2nd_Place_Beginners_UmojaHack_Africa_2022_Faulty_Air_Quality_Sensor_Challenge.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BpeiZTU7y188"
      },
      "source": [
        "Mount your Google Drive"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oZhekY1O_z0K",
        "outputId": "43f13d3b-edd4-4961-df8e-42fc2d45fd8c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0so031txy19J"
      },
      "source": [
        "Import libraries"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "NuGKrcam_4Ci"
      },
      "outputs": [],
      "source": [
        "# Import libraries\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import accuracy_score, confusion_matrix, classification_report, ConfusionMatrixDisplay\n",
        "pd.options.display.float_format = '{:.5f}'.format\n",
        "\n",
        "%matplotlib inline\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ggsO5rGpy19M"
      },
      "source": [
        "Bring in the CSV files"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "MPFCEbH0__jZ",
        "outputId": "5b1fc676-4b7a-422b-fe6a-815929612ef3"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                ID            Datetime  Sensor1_PM2.5  Sensor2_PM2.5  \\\n",
              "0  ID_QF0ZTQJ2SF5Q 2021-11-03 04:06:31       52.58000       49.52000   \n",
              "1  ID_4GTK689CNX5S 2021-11-08 18:43:23       35.25000       33.40000   \n",
              "2  ID_DL7VVKW9U7XQ 2021-11-07 09:50:33       19.18000       23.50000   \n",
              "3  ID_6XQOMBXM2DG3 2022-01-01 18:55:15       19.40000       15.48000   \n",
              "4  ID_UQZW9ISJY9QE 2021-11-05 22:23:48       38.30000       34.77000   \n",
              "\n",
              "   Temperature  Relative_Humidity  Offset_fault  \n",
              "0     17.40000           96.00000             0  \n",
              "1     25.00000           75.00000             0  \n",
              "2     24.90000           75.00000             0  \n",
              "3     24.90000           70.00000             0  \n",
              "4     20.90000           89.00000             0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-1f247eac-e5d4-466f-93f2-3f71e8806013\">\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>ID</th>\n",
              "      <th>Datetime</th>\n",
              "      <th>Sensor1_PM2.5</th>\n",
              "      <th>Sensor2_PM2.5</th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Relative_Humidity</th>\n",
              "      <th>Offset_fault</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ID_QF0ZTQJ2SF5Q</td>\n",
              "      <td>2021-11-03 04:06:31</td>\n",
              "      <td>52.58000</td>\n",
              "      <td>49.52000</td>\n",
              "      <td>17.40000</td>\n",
              "      <td>96.00000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ID_4GTK689CNX5S</td>\n",
              "      <td>2021-11-08 18:43:23</td>\n",
              "      <td>35.25000</td>\n",
              "      <td>33.40000</td>\n",
              "      <td>25.00000</td>\n",
              "      <td>75.00000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ID_DL7VVKW9U7XQ</td>\n",
              "      <td>2021-11-07 09:50:33</td>\n",
              "      <td>19.18000</td>\n",
              "      <td>23.50000</td>\n",
              "      <td>24.90000</td>\n",
              "      <td>75.00000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ID_6XQOMBXM2DG3</td>\n",
              "      <td>2022-01-01 18:55:15</td>\n",
              "      <td>19.40000</td>\n",
              "      <td>15.48000</td>\n",
              "      <td>24.90000</td>\n",
              "      <td>70.00000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ID_UQZW9ISJY9QE</td>\n",
              "      <td>2021-11-05 22:23:48</td>\n",
              "      <td>38.30000</td>\n",
              "      <td>34.77000</td>\n",
              "      <td>20.90000</td>\n",
              "      <td>89.00000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1f247eac-e5d4-466f-93f2-3f71e8806013')\"\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-1f247eac-e5d4-466f-93f2-3f71e8806013 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-1f247eac-e5d4-466f-93f2-3f71e8806013');\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": 3
        }
      ],
      "source": [
        "# Load files\n",
        "train = pd.read_csv('/content/drive/MyDrive/machine_learning_stories/Umoja2022/Beg/train.csv', parse_dates = ['Datetime'])\n",
        "test = pd.read_csv('/content/drive/MyDrive/machine_learning_stories/Umoja2022/Beg/test.csv', parse_dates = ['Datetime'])\n",
        "samplesubmission = pd.read_csv('/content/drive/MyDrive/machine_learning_stories/Umoja2022/Beg/SampleSubmission.csv')\n",
        "\n",
        "# Preview train dataset\n",
        "train.head()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5cN_jWray19O"
      },
      "source": [
        "Sort the train and test set by date, gives the model a sense of seeing the data as a time series"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "IsKJFnXRshqi"
      },
      "outputs": [],
      "source": [
        "train = train.sort_values(by=['Datetime']).reset_index(drop=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "NnDQ38z_sxMq"
      },
      "outputs": [],
      "source": [
        "test = test.sort_values(by=['Datetime']).reset_index(drop=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "roGBXKZxM242",
        "outputId": "55a20fd4-7e32-4c3b-9d6b-99ef22453bda"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       Sensor1_PM2.5  Sensor2_PM2.5  Temperature  Relative_Humidity  \\\n",
              "count   293563.00000   293563.00000 293628.00000       293628.00000   \n",
              "mean        40.75120       38.18302     23.31655           78.65362   \n",
              "std         28.52618       28.64776      3.83697           14.77272   \n",
              "min         -9.45000       -9.70000     16.70000           27.00000   \n",
              "25%         22.43000       19.52000     20.40000           68.00000   \n",
              "50%         36.47000       33.72000     21.90000           84.00000   \n",
              "75%         49.42000       47.57000     26.30000           91.00000   \n",
              "max        997.96000      997.26000     34.90000           99.00000   \n",
              "\n",
              "       Offset_fault  \n",
              "count  297177.00000  \n",
              "mean        0.37107  \n",
              "std         0.48309  \n",
              "min         0.00000  \n",
              "25%         0.00000  \n",
              "50%         0.00000  \n",
              "75%         1.00000  \n",
              "max         1.00000  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-02889c00-2e02-4c98-b1f7-95c8167132fa\">\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>Sensor1_PM2.5</th>\n",
              "      <th>Sensor2_PM2.5</th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Relative_Humidity</th>\n",
              "      <th>Offset_fault</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>293563.00000</td>\n",
              "      <td>293563.00000</td>\n",
              "      <td>293628.00000</td>\n",
              "      <td>293628.00000</td>\n",
              "      <td>297177.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>40.75120</td>\n",
              "      <td>38.18302</td>\n",
              "      <td>23.31655</td>\n",
              "      <td>78.65362</td>\n",
              "      <td>0.37107</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>28.52618</td>\n",
              "      <td>28.64776</td>\n",
              "      <td>3.83697</td>\n",
              "      <td>14.77272</td>\n",
              "      <td>0.48309</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>-9.45000</td>\n",
              "      <td>-9.70000</td>\n",
              "      <td>16.70000</td>\n",
              "      <td>27.00000</td>\n",
              "      <td>0.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>22.43000</td>\n",
              "      <td>19.52000</td>\n",
              "      <td>20.40000</td>\n",
              "      <td>68.00000</td>\n",
              "      <td>0.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>36.47000</td>\n",
              "      <td>33.72000</td>\n",
              "      <td>21.90000</td>\n",
              "      <td>84.00000</td>\n",
              "      <td>0.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>49.42000</td>\n",
              "      <td>47.57000</td>\n",
              "      <td>26.30000</td>\n",
              "      <td>91.00000</td>\n",
              "      <td>1.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>997.96000</td>\n",
              "      <td>997.26000</td>\n",
              "      <td>34.90000</td>\n",
              "      <td>99.00000</td>\n",
              "      <td>1.00000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-02889c00-2e02-4c98-b1f7-95c8167132fa')\"\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-02889c00-2e02-4c98-b1f7-95c8167132fa 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-02889c00-2e02-4c98-b1f7-95c8167132fa');\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": 6
        }
      ],
      "source": [
        "train.describe()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 250
        },
        "id": "-3rS3EpnNFPQ",
        "outputId": "51b9ecff-bbdd-4e01-9e14-c14aa7581187"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                     ID            Datetime  Sensor1_PM2.5  Sensor2_PM2.5  \\\n",
              "127356  ID_0Q6FJWMIY1UD 2022-03-14 14:45:42       16.58000       13.15000   \n",
              "127357  ID_J5QHC8KEKLOS 2022-03-14 14:47:04       14.33000       12.77000   \n",
              "127358  ID_27Q3I4ZW27OC 2022-03-14 14:48:07       15.60000       15.97000   \n",
              "127359  ID_17GIG572BMR3 2022-03-14 14:49:22       12.87000       13.08000   \n",
              "127360  ID_XZOXV4ACPR32 2022-03-14 14:50:35       10.57000       11.43000   \n",
              "\n",
              "        Temperature  Relative_Humidity  \n",
              "127356          NaN                NaN  \n",
              "127357          NaN                NaN  \n",
              "127358          NaN                NaN  \n",
              "127359          NaN                NaN  \n",
              "127360          NaN                NaN  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a035b063-6cd9-43cf-a62a-0080cb1940ae\">\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>ID</th>\n",
              "      <th>Datetime</th>\n",
              "      <th>Sensor1_PM2.5</th>\n",
              "      <th>Sensor2_PM2.5</th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Relative_Humidity</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>127356</th>\n",
              "      <td>ID_0Q6FJWMIY1UD</td>\n",
              "      <td>2022-03-14 14:45:42</td>\n",
              "      <td>16.58000</td>\n",
              "      <td>13.15000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>127357</th>\n",
              "      <td>ID_J5QHC8KEKLOS</td>\n",
              "      <td>2022-03-14 14:47:04</td>\n",
              "      <td>14.33000</td>\n",
              "      <td>12.77000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>127358</th>\n",
              "      <td>ID_27Q3I4ZW27OC</td>\n",
              "      <td>2022-03-14 14:48:07</td>\n",
              "      <td>15.60000</td>\n",
              "      <td>15.97000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>127359</th>\n",
              "      <td>ID_17GIG572BMR3</td>\n",
              "      <td>2022-03-14 14:49:22</td>\n",
              "      <td>12.87000</td>\n",
              "      <td>13.08000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>127360</th>\n",
              "      <td>ID_XZOXV4ACPR32</td>\n",
              "      <td>2022-03-14 14:50:35</td>\n",
              "      <td>10.57000</td>\n",
              "      <td>11.43000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a035b063-6cd9-43cf-a62a-0080cb1940ae')\"\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-a035b063-6cd9-43cf-a62a-0080cb1940ae 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-a035b063-6cd9-43cf-a62a-0080cb1940ae');\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": 7
        }
      ],
      "source": [
        "test.tail()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BLld9k1Xy19U"
      },
      "source": [
        "Check out Outliers in the train set for sensor1"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "F6JU4LbKOyiu",
        "outputId": "5e5e4a24-1598-414e-d27c-5b647769e6e4"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Outliers from Z-scores method:  [129.98, 127.18, 131.42, 133.37, 135.58, 136.85, 128.73, 135.03, 143.73, 133.38, 140.48, 133.77, 134.05, 130.82, 129.02, 129.12, 130.1, 133.45, 141.98, 141.82, 142.23, 150.68, 145.83, 146.58, 141.85, 155.7, 137.95, 144.77, 139.68, 147.47, 142.65, 147.77, 139.73, 142.23, 128.75, 132.57, 127.2, 127.68, 138.98, 127.45, 138.03, 129.25, 141.17, 135.67, 141.32, 132.03, 141.32, 137.53, 138.5, 138.6, 145.97, 136.73, 145.53, 136.95, 149.33, 133.75, 132.57, 147.12, 139.08, 142.7, 135.92, 143.85, 130.78, 143.02, 136.62, 140.12, 138.62, 143.48, 130.83, 142.65, 131.85, 134.17, 131.15, 133.08, 127.12, 127.07, 127.62, 465.45, 440.12, 431.46, 148.88, 164.38, 145.68, 127.48, 128.2, 126.67, 136.02, 130.77, 131.77, 495.64, 493.32, 472.56, 421.45, 479.0, 491.2, 444.61, 127.17, 284.0, 283.18, 405.97, 133.62, 345.0, 145.82, 144.7, 141.92, 151.13, 154.25, 152.27, 152.88, 148.07, 147.0, 146.0, 149.53, 138.8, 142.52, 127.68, 131.9, 128.0, 128.17, 128.18, 130.3, 139.58, 137.42, 146.42, 132.03, 138.35, 141.35, 152.07, 151.0, 166.4, 158.42, 191.32, 177.93, 207.55, 138.75, 197.93, 222.25, 133.5, 189.05, 185.25, 158.27, 202.1, 184.95, 133.63, 236.1, 207.7, 140.23, 200.85, 218.65, 152.85, 171.63, 150.72, 157.22, 148.93, 153.87, 135.47, 150.88, 140.75, 791.27, 789.85, 536.77, 505.33, 997.96, 878.23, 723.51, 568.81, 511.31, 690.51, 853.81, 733.91, 910.28, 962.35, 684.22, 788.51, 773.74, 946.34, 720.47, 866.74, 698.18, 686.77, 551.38, 703.53, 823.6, 584.03, 913.1, 581.13, 519.23, 706.42, 594.39, 937.76, 659.89, 680.09, 830.54, 768.2, 759.14, 506.31, 868.62, 774.55, 667.59, 709.68, 881.59, 891.81, 851.66, 536.35, 706.51, 723.85, 553.01, 671.57, 731.95, 827.31, 543.0, 827.75, 650.95, 873.65, 696.26, 699.62, 829.0, 649.62, 790.2, 673.81, 734.4, 746.86, 954.78, 171.18, 177.03, 183.83, 195.23, 176.92, 199.25, 181.97, 195.57, 176.25, 160.77, 183.17, 160.85, 168.12, 150.78, 163.98, 132.12, 137.68, 131.97, 138.65, 135.2, 135.82, 206.83, 212.33, 160.92, 244.77, 169.98, 249.62, 226.83, 153.5, 242.27, 196.32, 127.55, 196.57, 136.37, 176.65, 127.67, 150.3, 144.55, 145.18, 143.35, 157.37, 130.9, 141.47, 132.65, 136.77, 128.22, 133.15, 126.87, 127.22, 127.03, 127.3, 129.87, 126.37, 129.33, 128.4, 128.42, 134.47, 130.6, 134.58, 142.98, 139.48, 149.02, 142.13, 152.33, 155.33, 164.72, 152.07, 166.17, 144.22, 153.78, 150.92, 160.22, 151.05, 144.43, 157.05, 147.35, 159.13, 148.45, 155.97, 144.95, 151.48, 144.87, 154.17, 140.63, 149.95, 136.7, 145.57, 137.95, 137.18, 131.85, 138.65, 127.77, 131.23, 136.63, 126.8, 129.48, 129.72, 128.13, 130.17, 130.22, 127.0, 128.12, 127.17, 160.38, 145.18, 135.07, 157.58, 127.4, 172.17, 197.48, 128.22, 188.03, 202.83, 180.67, 201.67, 140.03, 159.42, 139.1, 139.82, 142.52, 133.58, 151.47, 128.63, 134.28, 136.72, 167.6, 183.93, 209.15, 170.68, 173.75, 147.7, 132.72, 128.05, 157.62, 187.6, 156.1, 184.65, 169.88, 172.75, 194.55, 140.7, 142.98, 170.37, 173.03, 159.33, 180.07, 132.48, 133.73, 149.95, 157.08, 146.68, 161.5, 156.75, 159.98, 145.82, 160.28, 152.88, 151.03, 152.43, 156.9, 142.3, 145.75, 130.6, 141.07, 135.8, 154.28, 149.6, 147.28, 147.75, 155.7, 147.45, 151.93, 146.38, 139.43, 145.33, 146.62, 149.07, 146.17, 146.93, 146.82, 153.1, 153.45, 153.4, 153.12, 162.38, 155.45, 155.33, 160.62, 175.52, 172.83, 131.27, 203.75, 204.73, 143.78, 203.25, 207.05, 138.55, 219.82, 209.68, 145.98, 224.5, 228.67, 151.68, 227.37, 235.45, 154.48, 237.93, 242.25, 157.67, 240.82, 245.5, 159.38, 236.67, 161.93, 256.58, 239.45, 153.48, 253.97, 251.5, 162.18, 257.02, 165.52, 239.82, 256.13, 162.78, 236.65, 262.63, 163.03, 236.05, 254.6, 173.42, 248.68, 157.97, 244.72, 233.48, 160.43, 250.97, 238.17, 156.28, 246.48, 232.15, 151.58, 228.28, 248.63, 149.83, 219.95, 239.32, 150.57, 203.68, 142.52, 228.18, 200.98, 140.03, 211.03, 184.97, 127.98, 199.13, 128.72, 187.15, 186.82, 183.83, 182.15, 157.73, 182.32, 171.83, 169.85, 163.27, 170.95, 156.63, 169.5, 155.47, 152.42, 160.62, 148.55, 151.53, 138.15, 151.55, 135.15, 147.08, 130.28, 132.97, 128.08, 126.95, 127.53, 130.92, 229.27, 186.25, 146.87, 177.47, 194.7, 139.42, 174.8, 130.83, 126.65, 128.28, 134.58, 128.23, 127.73, 129.4, 126.58, 128.82, 128.42, 127.87, 126.93, 128.17, 137.38, 132.88, 141.98, 134.02, 132.78, 126.63, 131.13, 127.1, 128.82, 128.03, 128.95, 128.32, 132.2, 132.03, 131.07, 136.75, 130.38, 131.75, 128.87, 134.83, 130.98, 133.95, 133.75, 137.02, 137.38, 140.5, 129.7, 132.65, 132.43, 131.27, 135.2, 139.32, 135.22, 135.17, 129.48, 135.93, 136.33, 133.73, 138.62, 143.92, 135.17, 140.27, 139.03, 138.03, 142.38, 142.2, 142.48, 134.48, 136.93, 130.77, 132.13, 129.08, 132.67, 130.5, 129.17, 138.07, 144.1, 161.73, 127.12, 138.02, 159.13, 170.22, 136.72, 141.62, 152.3, 128.82, 143.63, 129.17, 128.2, 127.28, 129.33, 131.05, 132.95, 128.33, 132.75, 132.85, 135.02, 134.97, 135.05, 134.47, 133.98, 131.87, 137.08, 131.62, 140.25, 138.77, 143.38, 141.62, 145.43, 141.15, 152.05, 143.82, 155.75, 151.37, 150.88, 154.93, 155.15, 156.45, 149.55, 159.32, 153.77, 161.93, 158.62, 159.42, 154.05, 167.45, 149.73, 157.72, 149.0, 152.18, 151.45, 151.05, 146.08, 154.55, 146.3, 143.98, 150.97, 144.08, 141.2, 141.18, 145.95, 147.57, 152.25, 147.68, 147.07, 148.72, 157.52, 151.63, 155.95, 149.25, 147.68, 148.82, 147.68, 142.33, 154.57, 145.97, 158.47, 144.12, 151.97, 160.43, 153.25, 158.85, 142.38, 160.65, 131.45, 139.93, 131.27, 136.38, 129.85, 133.55, 134.63, 136.48, 139.22, 133.7, 138.65, 133.23, 134.07, 132.82, 133.85, 129.27, 131.67, 136.22, 133.87, 127.83, 142.13, 135.7, 137.98, 140.48, 151.27, 145.6, 149.87, 147.72, 145.55, 129.4, 128.92, 126.7, 128.1, 127.0, 133.57, 138.12, 144.42, 139.87, 146.45, 136.9, 126.8, 141.48, 132.33, 136.7, 133.45, 133.98, 126.88, 135.87, 134.93, 141.73, 140.62, 150.2, 157.95, 177.3, 126.85, 208.47, 140.15, 206.75, 218.8, 138.05, 214.03, 192.97, 141.2, 180.27, 136.97, 150.37, 127.13, 148.83, 127.68, 134.0, 135.53, 129.37, 130.65, 130.68, 128.58, 129.8, 132.35, 127.82, 128.85, 128.73, 127.0, 128.5, 126.65, 126.72, 130.17, 128.0, 131.65, 133.77, 127.12, 156.65, 190.72, 144.25, 212.08, 191.98, 178.7, 188.45, 146.98, 213.67, 205.52, 139.82, 210.3, 212.22, 165.6, 201.43, 139.12, 209.5, 218.5, 166.28, 239.48, 147.52, 221.35, 228.02, 148.13, 208.78, 215.9, 153.48, 192.42, 145.98, 222.15, 186.07, 133.68, 212.47, 194.32, 141.02, 204.25, 180.08, 133.55, 195.52, 170.45, 131.82, 166.43, 127.82, 192.83, 158.23, 184.27, 157.23, 171.7, 154.85, 173.13, 156.75, 166.0, 148.87, 153.65, 133.33, 145.38, 131.17, 147.05, 137.1, 142.58, 132.12, 144.1, 135.53, 133.37, 130.72, 136.9, 129.05, 138.97, 150.22, 160.08, 150.33, 129.9, 180.02, 179.25, 162.28, 172.18, 167.73, 178.17, 156.85, 173.43, 145.83, 165.3, 158.05, 163.28, 180.43, 162.57, 166.83, 128.38, 175.37, 185.55, 168.6, 187.0, 167.02, 179.88, 173.48, 187.5, 151.9, 166.28, 153.52, 177.93, 154.55, 174.82, 166.15, 164.82, 137.28, 128.12, 151.4, 134.2, 136.27, 196.12, 214.17, 133.07, 178.2, 201.43, 161.13, 178.57, 167.77, 174.68, 163.17, 167.0, 156.97, 155.43, 144.05, 138.9, 126.95, 136.28, 137.12, 138.47, 139.98, 141.22, 145.28, 149.88, 160.77, 147.05, 159.82, 152.35, 136.27, 158.75, 147.67, 157.75, 139.57, 165.9, 130.15, 154.25, 148.52, 156.98, 158.65, 171.83, 142.45, 161.27, 136.02, 139.17, 131.88, 139.35, 134.63, 142.35, 127.52, 137.43, 132.87, 131.9, 140.47, 131.47, 131.98, 144.45, 128.6, 162.97, 126.38, 128.82, 127.72, 155.48, 198.78, 219.73, 127.13, 129.35, 133.93, 136.87, 130.87, 138.97, 134.07, 134.97, 185.27, 127.38, 127.97, 138.77, 133.18, 155.53, 149.3, 129.35, 147.97, 132.38, 139.17, 128.52, 128.3, 134.33, 133.12, 131.83, 134.55, 129.12, 128.63, 131.25, 127.18, 126.97, 128.17, 128.52, 132.62, 135.1, 143.3, 137.23, 143.85, 143.55, 142.92, 127.65, 145.27, 139.97, 140.6, 127.65, 144.75, 142.97, 137.65, 140.92, 140.82, 143.88, 139.12, 137.05, 131.88, 131.97, 132.18, 126.8, 129.53, 140.27, 147.18, 152.95, 131.87, 158.15, 131.98, 158.47, 159.88, 133.88, 164.43, 137.85, 167.18, 132.88, 165.57, 140.53, 158.12, 136.47, 158.17, 139.28, 156.57, 128.82, 148.6, 128.22, 142.4, 147.62, 135.28, 131.93, 136.88, 139.17, 135.63, 126.63, 126.7, 136.53, 128.57, 129.4, 126.35, 128.2, 133.07, 131.83, 131.87, 134.7, 137.53, 136.92, 134.62, 139.43, 140.15, 151.7, 132.73, 152.67, 165.83, 139.4, 167.17, 135.92, 131.27, 140.8, 181.33, 147.85, 151.88, 163.9, 187.03, 156.88, 178.65, 144.25, 159.75, 136.77, 128.98, 129.27, 137.73, 128.08, 127.57, 131.87, 136.07, 138.38, 149.52, 143.5, 146.37, 146.62, 146.1, 126.48, 144.5, 130.98, 156.05, 132.38, 153.48, 135.45, 155.25, 131.87, 162.35, 130.97, 144.48, 130.37, 148.13, 138.12, 131.13, 130.78, 130.97, 127.52, 128.72, 131.12, 126.6, 127.88, 130.05, 127.52, 135.55, 135.88, 135.0, 128.67, 130.2, 130.15, 133.3, 131.27, 135.83, 137.72, 136.42, 137.85, 141.17, 127.27, 149.42, 131.8, 155.95, 139.58, 169.2, 147.97, 201.48, 165.87, 190.52, 127.0, 163.77, 194.75, 131.32, 166.78, 132.02, 194.85, 136.05, 175.88, 199.38, 132.97, 169.97, 190.27, 130.03, 160.92, 180.37, 166.28, 142.92, 168.95, 146.23, 160.08, 137.75, 137.32, 132.1, 128.78, 143.35, 137.25, 149.42, 138.88, 161.23, 167.95, 145.95, 166.3, 146.98, 163.32, 144.0, 158.65, 138.33, 162.58, 135.8, 160.0, 138.78, 130.13, 131.07, 135.85, 134.35, 136.32, 143.47, 147.8, 149.23, 127.1, 149.17, 144.77, 145.53, 145.93, 128.73, 149.62, 136.2, 136.0, 127.03, 129.35, 128.98, 130.75, 128.08, 130.0, 128.08, 126.82, 131.3, 139.3, 136.9, 127.97, 131.35, 128.22, 135.27, 136.38, 138.77, 133.53, 144.1, 137.98, 137.27, 135.73, 131.78, 126.92, 128.2, 136.23, 133.2, 136.3, 127.03, 127.93, 126.98, 136.37, 131.32, 139.42, 128.57, 129.17, 130.28, 142.38, 146.97, 131.47, 142.43, 128.1, 139.7, 134.48, 132.63, 142.4, 140.92, 130.18, 141.23, 138.85, 140.05, 142.25, 134.73, 147.0, 127.48, 166.38, 142.98, 172.0, 147.08, 162.83, 142.37, 154.88, 133.62, 154.77, 134.75, 148.77, 134.48, 147.48, 133.95, 134.67, 129.85, 130.37, 141.2, 142.78, 138.47, 143.13, 131.52, 132.23, 133.72, 143.73, 140.05, 130.55, 141.62, 129.08, 139.75, 129.53, 142.53, 141.5, 143.02, 146.82, 143.02, 127.87, 143.18, 126.82, 141.52, 127.42, 136.4, 139.0, 127.78, 132.28, 134.02, 138.0, 138.5, 136.98, 133.63, 129.32, 129.87, 134.65, 133.28, 140.5, 127.95, 145.82, 130.02, 150.2, 140.48, 157.55, 139.0, 156.0, 138.9, 158.5, 139.37, 155.02, 134.85, 152.3, 135.83, 158.18, 145.55, 130.43, 146.68, 152.73, 137.68, 147.45, 132.2, 138.97, 127.72, 141.12, 132.58, 143.25, 145.73, 128.48, 135.25, 137.53, 133.77, 128.88, 134.55, 134.38, 131.42, 130.97, 135.13, 132.38, 128.55, 128.87, 129.15, 129.45, 131.53, 131.0, 128.83, 128.77, 127.07, 165.48, 161.18, 166.78, 140.02, 136.87, 155.2, 130.68, 138.67, 157.9, 130.3, 161.62, 181.47, 225.03, 136.87, 175.33, 196.1, 237.77, 142.87, 155.15, 126.93, 152.1, 141.42, 161.83, 158.58, 178.45, 150.25, 184.77, 177.08, 150.47, 177.13, 143.6, 140.1, 151.28, 149.57, 127.68, 126.5, 144.45, 160.07, 131.6, 157.93, 132.83, 140.17, 140.62, 157.05, 135.2, 143.72, 132.58, 142.13, 127.2, 148.08, 135.4, 148.55, 130.58, 129.68, 135.07, 138.17, 135.25, 144.47, 130.08, 152.93, 134.05, 153.77, 132.48, 158.88, 159.18, 140.9, 156.97, 136.2, 151.17, 131.48, 129.45, 131.77, 130.27, 127.78, 158.07, 135.15, 170.73, 136.85, 142.53, 151.18, 171.37, 153.02, 186.15, 130.32, 163.58, 192.83, 136.48, 200.02, 178.25, 139.58, 235.25, 205.65, 161.33, 253.03, 168.78, 235.77, 185.33, 274.25, 246.23, 187.48, 284.33, 241.15, 175.18, 236.05, 175.78, 172.72, 214.32, 242.78, 160.28, 204.33, 159.02, 198.02, 164.15, 205.32, 237.87, 162.63, 199.4, 218.85, 155.12, 149.38, 223.62, 198.25, 160.55, 217.73, 188.48, 154.52, 212.52, 188.62, 152.75, 219.38, 190.97, 154.25, 147.18, 186.63, 151.58, 190.25, 203.93, 141.67, 184.07, 143.53, 201.95, 184.82, 141.15, 210.75, 183.33, 141.73, 207.42, 138.88, 171.37, 198.55, 144.68, 180.7, 202.25, 142.47, 197.72, 173.83, 137.53, 182.78, 135.53, 144.4, 189.45, 173.28, 159.58, 131.43, 192.45, 161.55, 133.55, 160.93, 132.25, 126.63, 156.35, 177.78, 153.98, 169.5, 137.15, 157.25, 139.65, 158.62, 141.45, 138.3, 136.13, 126.52, 129.25, 139.22, 129.58, 128.45, 130.68, 134.88, 130.42, 128.38, 135.33, 141.32, 168.85, 141.42, 159.77, 142.6, 147.0, 132.37, 138.38, 139.1, 137.0, 130.28, 129.17, 129.8, 127.77, 129.12, 129.97, 132.75, 136.98, 135.93, 132.13, 128.6, 130.52, 129.1, 126.78, 127.83, 132.15, 133.73, 146.48, 141.75, 136.27, 151.43, 148.02, 127.57, 129.4, 130.0, 150.58, 131.1, 158.25, 154.55, 132.13, 152.58, 133.95, 155.7, 135.72, 161.95, 141.02, 163.45, 138.92, 164.45, 133.58, 149.9, 135.37, 166.0, 134.4, 151.72, 156.45, 135.52, 135.25, 129.15, 179.52, 138.27, 129.37, 150.07, 154.28, 140.08, 130.82, 132.83, 139.08, 126.45, 133.82, 133.18, 136.12, 139.42, 141.1, 128.62, 142.75, 142.85, 126.78, 140.67, 138.27, 133.2, 129.52, 129.8, 136.47, 137.62, 141.65, 126.58, 128.57, 134.57, 303.53, 127.9, 132.13, 143.2, 131.4, 126.4, 128.92, 126.53, 127.43, 126.68, 142.32, 152.95, 149.25, 158.17, 154.27, 128.08, 150.3, 131.18, 151.13, 144.2, 142.33, 145.6, 147.6, 149.13, 127.33, 158.4, 132.48, 153.72, 151.53, 144.77, 141.2, 155.8, 127.1, 164.6, 133.57, 143.75, 130.78, 131.07, 128.03, 132.12, 137.33, 133.87, 134.1, 130.35, 128.03, 128.95, 129.27, 134.15, 128.8, 139.12, 144.43, 144.9, 151.4, 146.82, 154.38, 154.58, 131.03, 159.25, 130.4, 159.05, 131.87, 159.83, 158.62, 132.87, 168.45, 139.32, 173.83, 147.38, 174.87, 143.02, 173.03, 127.37, 153.2, 175.9, 148.25, 170.95, 143.48, 167.9, 144.7, 151.75, 127.72, 141.63, 129.05, 127.15, 133.03, 136.57, 135.38, 138.32, 134.3, 135.32, 138.33, 130.0, 136.48, 134.8, 135.45, 127.55, 128.15, 127.28, 128.98, 132.88, 128.72, 134.47, 131.35, 127.5, 126.7, 128.57, 140.33, 129.42, 142.18, 128.3, 130.83, 129.82, 132.03, 126.6, 154.25, 131.03, 151.15, 127.78, 144.82, 147.4, 128.95, 162.88, 127.33, 163.07, 139.38, 126.57, 177.65, 155.55, 136.03, 202.72, 139.67, 168.92, 192.63, 137.15, 163.48, 182.13, 135.58, 193.37, 171.92, 131.22, 196.38, 169.48, 145.4, 204.77, 147.97, 175.42, 197.0, 129.17, 162.77, 128.2, 191.23, 158.33, 169.82, 133.42, 163.27, 176.22, 145.92, 131.07, 173.78, 151.6, 134.25, 178.87, 156.17, 133.57, 191.55, 160.0, 131.07, 183.43, 142.05, 161.67, 204.88, 136.05, 162.58, 182.08, 167.82, 138.87, 155.03, 137.15, 158.95, 135.88, 155.38, 132.13, 150.05, 129.55, 153.37, 127.33, 156.45, 146.27, 139.4, 142.25, 145.55, 143.42, 142.15, 138.35, 131.0, 133.67, 140.68, 132.27, 130.18, 131.48, 131.3, 127.17, 133.68, 126.87, 129.8, 128.38, 129.1, 128.23, 127.53, 126.93, 127.5, 128.27, 133.0, 128.77, 218.1, 127.68, 131.87, 128.82, 132.47, 135.43, 135.72, 131.42, 136.5, 137.68, 140.43, 139.7, 145.88, 144.97, 139.78, 147.43, 142.9, 142.75, 139.55, 129.62, 133.92, 129.0, 126.73, 129.3, 126.42, 127.85, 131.47, 129.97, 136.55, 136.08, 139.12, 143.73, 140.43, 134.33, 139.03, 137.68, 133.43, 134.82, 134.07, 138.88, 143.5, 147.15, 136.97, 138.68, 135.78, 136.0, 137.98, 141.73, 131.27, 137.8, 131.32, 132.48, 130.97, 134.52, 130.82, 138.02, 139.17, 137.82, 142.65, 134.75, 133.83, 136.57, 134.85, 131.78, 133.35, 131.52, 135.77, 131.25, 130.2, 126.67, 133.38, 136.03, 132.9, 130.77, 134.35, 137.32, 131.85, 134.78, 136.25, 126.42, 133.02, 128.72, 128.22, 141.07, 130.17, 137.0, 135.87, 131.57, 133.22, 133.93, 132.73, 127.6, 129.65, 127.1, 128.25, 134.37, 130.7, 136.57, 130.87, 130.78, 129.13, 129.55, 129.1, 126.33, 131.1, 132.67, 131.53, 133.28, 128.9, 129.5, 133.13, 130.8, 132.03, 128.87, 130.55, 127.05, 132.08, 135.67, 132.9, 133.72, 135.43, 136.05, 134.07, 134.33, 134.38, 130.28, 130.72, 141.27, 134.22, 141.38, 137.28, 139.3, 139.8, 135.4, 140.67, 133.28, 135.73, 134.68, 134.93, 131.1, 132.22, 127.48, 128.42, 130.37, 130.12, 127.65, 130.98, 126.73, 132.47, 130.5, 130.43, 134.97, 130.52, 127.0, 131.33, 140.3, 142.02, 139.3, 147.42, 131.95, 134.73, 127.68, 136.7, 144.52, 133.75, 128.43, 144.73, 162.32, 138.13, 175.37, 143.65, 174.83, 150.18, 176.98, 141.12, 181.67, 174.48, 157.2, 169.27, 153.25, 183.08, 126.83, 152.55, 194.47, 158.03, 180.3, 175.05, 162.23, 155.15, 131.75, 151.43, 142.02, 145.35, 127.78, 148.37, 141.52, 127.3, 152.75, 127.87, 148.78, 127.35, 151.6, 128.33, 152.92, 128.72, 155.9, 153.1, 130.05, 150.28, 129.92, 126.43, 144.85, 139.93, 147.47, 140.18, 136.02, 139.8, 134.1, 129.43, 130.7, 130.6, 128.35, 133.82, 127.37, 129.25, 130.58, 129.25, 127.77, 150.48, 179.0, 148.7, 158.37, 157.32, 137.68, 148.78, 158.17, 135.22, 170.53, 165.42, 130.17, 151.6, 129.2, 146.08, 153.08, 151.77, 129.93, 148.12, 143.08, 127.0, 144.8, 128.07, 151.13, 126.4, 150.02, 129.48, 148.63, 133.02, 149.85, 129.5, 148.65, 148.48, 149.88, 151.47, 127.98, 149.95, 128.08, 150.4, 145.28, 128.2, 150.65, 127.07, 148.3, 149.42, 126.45, 142.02, 150.42, 128.48, 145.83, 141.52, 150.45, 143.13, 150.13, 128.43, 152.43, 139.75, 147.15, 129.83, 146.12, 130.87, 145.52, 145.37, 145.48, 129.08, 127.72, 127.87, 138.77, 139.03, 136.58, 138.13, 136.88, 144.27, 157.6, 129.98, 155.38, 135.25, 169.6, 142.43, 171.38, 151.6, 172.68, 167.83, 140.42, 171.53, 136.83, 172.4, 148.33, 176.73, 143.9, 169.35, 141.55, 169.18, 142.08, 155.68, 158.28, 138.23, 157.53, 141.6, 158.2, 134.52, 157.62, 134.15, 160.18, 138.75, 156.98, 133.45, 151.62, 151.27, 149.9, 126.53, 150.83, 129.53, 152.68, 133.87, 151.27, 152.0, 129.08, 157.78, 157.0, 128.82, 134.03, 151.5, 153.93, 129.73, 148.47, 147.77, 128.4, 148.5, 130.43, 143.67, 150.65, 144.73, 146.23, 126.92, 145.52, 146.98, 128.32, 144.8, 130.17, 143.72, 130.88, 150.92, 153.43, 128.53, 154.73, 161.47, 136.07, 158.05, 134.18, 159.42, 141.52, 165.62, 138.6, 160.85, 174.28, 143.93, 170.88, 147.82, 173.82, 147.95, 169.98, 146.97, 168.33, 150.02, 174.6, 151.48, 172.13, 173.57, 151.37, 180.73, 155.23, 128.42, 184.7, 164.88, 188.55, 127.02, 152.1, 183.3, 129.58, 160.82, 191.98, 126.78, 171.62, 205.68, 137.47, 192.52, 128.78, 175.13, 172.72, 152.0, 176.5, 149.98, 175.23, 163.38, 177.08, 149.78, 181.97, 157.75, 182.28, 187.97, 159.9, 132.58, 187.43, 160.5, 128.07, 189.08, 161.15, 186.0, 129.98, 158.62, 180.87, 126.58, 156.43, 176.55, 160.08, 181.75, 182.07, 153.48, 128.85, 180.07, 152.07, 180.52, 129.85, 155.48, 186.77, 152.33, 178.53, 127.93, 152.85, 186.33, 131.08, 155.37, 178.47, 174.4, 151.2, 177.78, 152.47, 172.97, 154.25, 177.03, 150.55, 170.7, 151.6, 173.6, 154.63, 187.57, 130.88, 187.58, 159.17, 189.27, 153.87, 133.2, 187.7, 160.75, 184.28, 159.12, 179.03, 146.75, 176.08, 148.65, 165.42, 173.33, 141.15, 164.32, 145.85, 163.82, 143.72, 161.48, 143.28, 158.38, 135.15, 164.55, 138.15, 155.25, 159.3, 138.82, 156.47, 133.0, 153.45, 136.63, 150.38, 159.43, 127.45, 156.78, 138.52, 160.23, 154.78, 134.55, 154.13, 134.82, 157.08, 130.75, 150.42, 139.58, 142.18, 138.68, 138.35, 134.33, 132.57, 130.92, 127.02, 127.02, 128.25, 138.92, 138.82, 148.68, 139.47, 139.78, 127.32, 143.55, 143.07, 144.63, 142.38, 146.1, 142.13, 139.38, 142.7, 144.72, 139.67, 126.53, 144.2, 135.98, 135.48, 134.45, 143.73, 146.0, 141.53, 146.23, 131.45, 153.08, 151.05, 130.78, 148.15, 128.38, 156.1, 127.4, 146.62, 132.82, 156.65, 132.23, 151.0, 128.23, 155.8, 131.05, 151.68, 150.88, 154.0, 135.82, 154.65, 135.65, 156.4, 138.38, 160.43, 137.07, 166.88, 163.48, 136.97, 159.27, 139.78, 163.25, 143.35, 162.13, 140.02, 163.5, 134.87, 166.65, 135.68, 169.07, 160.37, 140.38, 159.18, 137.3, 156.08, 141.32, 158.03, 139.22, 164.68, 127.72, 155.58, 133.07, 154.42, 156.53, 136.55, 152.93, 132.2, 159.08, 132.77, 158.82, 139.92, 165.4, 131.63, 166.37, 141.68, 171.05, 173.22, 141.77, 175.58, 155.12, 174.17, 148.83, 174.13, 147.6, 180.48, 150.68, 174.55, 147.73, 168.5, 150.2, 174.22, 163.22, 141.27, 159.98, 140.67, 159.38, 135.03, 159.93, 135.48, 158.05, 134.75, 155.55, 136.92, 152.08, 156.38, 133.97, 152.78, 130.75, 154.47, 130.78, 153.85, 126.67, 152.95, 128.08, 157.77, 154.8, 140.85, 157.32, 134.47, 155.33, 136.82, 146.8, 129.57, 152.67, 127.2, 147.18, 150.68, 140.6, 137.92, 137.02, 134.3, 132.57, 127.32, 132.48, 127.33, 129.22, 130.58, 128.88, 135.12, 145.42, 144.85, 138.43, 130.95, 132.18, 138.92, 131.9, 134.67, 144.33, 143.97, 150.77, 146.9, 153.42, 155.22, 132.12, 148.63, 126.98, 146.33, 128.35, 144.15, 143.63, 127.2, 145.72, 143.67, 144.47, 128.75, 143.82, 151.23, 147.02, 128.22, 153.9, 126.42, 145.97, 152.65, 143.47, 146.38, 145.65, 141.25, 146.95, 128.75, 142.75, 144.18, 129.5, 144.25, 128.08, 140.8, 140.27, 143.27, 136.72, 140.8, 138.87, 136.2, 136.98, 128.43, 135.8, 136.77, 142.4, 131.75, 146.03, 138.17, 150.8, 149.63, 143.72, 143.37, 145.53, 151.98, 153.43, 149.97, 128.93, 152.65, 126.88, 156.08, 131.47, 156.48, 156.53, 132.63, 152.57, 131.38, 153.88, 135.45, 152.02, 134.2, 154.37, 132.0, 156.12, 138.22, 155.83, 153.13, 138.9, 154.27, 136.98, 161.03, 136.95, 149.85, 132.87, 154.43, 132.38, 156.35, 135.02, 156.75, 156.72, 137.97, 163.5, 138.17, 158.88, 135.92, 159.97, 136.32, 159.45, 136.63, 155.75, 131.5, 151.2, 157.68, 131.18, 152.3, 130.32, 150.55, 131.65, 157.97, 130.22, 150.35, 130.5, 149.8, 128.37, 149.6, 144.9, 147.83, 141.73, 138.15, 137.85, 133.55, 134.4, 130.25, 135.73, 135.08, 133.82, 129.82, 131.87, 128.82, 132.67, 128.25, 133.73, 134.05, 133.08, 136.5, 139.93, 147.85, 145.5, 135.5, 142.2, 142.8, 147.95, 132.07, 151.58, 151.68, 133.0, 156.03, 131.3, 159.65, 136.0, 158.75, 135.8, 153.1, 127.75, 156.32, 153.98, 149.38, 148.05, 128.75, 144.02, 148.9, 144.67, 138.3, 145.82, 145.2, 143.68, 145.85, 147.23, 127.05, 144.93, 128.7, 152.83, 126.53, 151.22, 151.72, 145.92, 127.47, 143.45, 145.32, 129.6, 150.63, 127.28, 141.42, 141.82, 139.57, 142.12, 157.28, 127.73, 141.53, 150.6, 129.9, 154.33, 154.58, 154.15, 129.32, 159.97, 130.27, 165.33, 136.37, 167.23, 140.8, 166.53, 135.97, 157.03, 131.38, 159.43, 126.55, 156.32, 151.57, 133.03, 158.3, 141.0, 164.43, 136.55, 161.33, 139.68, 168.22, 139.77, 170.48, 136.23, 168.45, 166.17, 137.57, 172.0, 144.07, 162.57, 134.35, 166.18, 135.53, 159.8, 135.72, 169.03, 173.42, 137.92, 169.1, 154.53, 182.87, 149.63, 178.37, 156.48, 174.02, 148.4, 172.02, 154.55, 185.82, 182.87, 156.65, 177.38, 127.6, 155.35, 180.33, 157.9, 189.4, 126.47, 166.33, 183.55, 164.4, 131.17, 186.58, 171.55, 186.13, 179.12, 151.48, 172.87, 151.53, 181.52, 152.5, 184.73, 162.95, 179.3, 131.73, 159.55, 188.27, 129.83, 161.73, 191.07, 187.08, 150.32, 178.97, 148.37, 177.58, 152.23, 187.88, 147.68, 184.62, 150.98, 176.5, 156.03, 174.57, 179.33, 148.93, 173.15, 151.35, 182.68, 147.47, 172.42, 148.62, 171.42, 149.68, 172.97, 138.85, 162.75, 159.82, 144.0, 159.6, 137.7, 157.75, 133.05, 150.68, 128.78, 162.6, 139.33, 159.8, 138.58, 176.0, 173.98, 158.63, 150.83, 179.55, 169.35, 149.38, 172.03, 150.6, 187.25, 137.53, 165.95, 204.2, 139.7, 176.27, 146.43, 209.82, 144.28, 215.7, 180.23, 149.62, 212.23, 188.18, 148.87, 211.32, 174.4, 140.63, 208.58, 168.68, 134.95, 193.02, 135.05, 161.77, 190.58, 132.03, 187.8, 165.4, 127.48, 179.6, 152.38, 131.87, 184.65, 148.55, 169.48, 141.38, 168.45, 138.78, 166.92, 138.33, 165.1, 146.33, 129.68, 129.45, 127.2, 126.87, 133.37, 134.35, 141.73, 135.98, 137.33, 137.17, 132.72, 140.62, 139.18, 138.67, 135.25, 133.65, 132.22, 132.18, 131.83, 129.9, 132.0, 130.93, 132.72, 131.93, 130.1, 131.05, 129.93, 136.1, 134.48, 139.78, 130.43, 133.98, 132.98, 136.27, 135.1, 135.02, 136.67, 128.45, 130.67, 131.27, 133.42, 131.57, 129.92, 130.55, 132.08, 130.68, 130.78, 126.78, 126.57, 130.17, 135.03, 132.98, 138.12, 137.32, 130.48, 127.78, 127.27, 134.75, 131.67, 132.77, 135.18, 130.78, 134.63, 128.0, 129.75, 132.38, 132.7, 129.25, 132.6, 131.85, 132.65, 137.05, 137.43, 139.08, 135.38, 138.93, 143.77, 137.83, 137.67, 141.03, 140.95, 142.62, 138.38, 144.62, 150.77, 148.47, 130.25, 154.97, 129.95, 152.23, 133.43, 148.35, 149.7, 132.2, 146.4, 133.3, 146.37, 151.45, 130.23, 151.87, 151.35, 127.82, 153.98, 148.03, 152.27, 150.58, 128.0, 146.75, 128.12, 145.92, 142.08, 135.63, 136.43, 132.65, 129.93, 127.03, 131.23, 145.9, 150.42, 127.3, 150.6, 128.3, 141.12, 145.85, 142.85, 136.3, 130.72, 127.22, 127.7, 126.45, 129.62, 130.6, 128.18, 127.18, 135.48, 131.02, 128.23, 127.25, 130.6, 127.93, 129.88, 126.92, 130.17, 128.83, 127.2, 130.8, 133.78, 139.7, 137.67, 137.83, 135.38, 142.4, 139.98, 142.37, 140.77, 137.87, 135.0, 126.62, 129.9, 128.02, 127.15, 127.88, 128.02, 142.65, 130.48, 128.87, 135.28, 140.8, 135.02, 138.32, 145.28, 146.55, 145.8, 156.57, 137.52, 172.97, 181.13, 144.37, 183.85, 146.97, 193.42, 129.53, 172.95, 196.65, 127.53, 161.35, 173.85, 146.6, 162.1, 134.32, 153.48, 145.88, 142.92, 139.88, 133.63, 132.0, 131.42, 135.92, 134.08, 132.08, 130.27, 134.62, 129.42, 131.77, 128.13, 127.65, 129.9, 128.32, 132.65, 136.47, 133.05, 126.58, 132.05, 129.23, 130.9, 129.77, 134.67, 130.23, 128.87, 126.53, 127.48, 127.33, 127.0, 127.42, 130.65, 129.15, 129.67, 132.08, 128.78, 131.57, 127.0, 134.77, 140.78, 141.63, 143.8, 149.45, 155.68, 159.42, 150.98, 176.18, 134.08, 185.42, 146.28, 189.37, 150.12, 182.4, 149.57, 170.87, 141.38, 157.43, 134.65, 142.72, 129.38, 145.65, 150.85, 129.87, 152.35, 126.33, 151.62, 132.48, 154.02, 130.75, 156.47, 138.5, 165.08, 178.0, 133.43, 178.13, 144.82, 180.3, 151.58, 128.98, 187.58, 155.7, 182.52, 155.27, 181.73, 148.23, 177.13, 169.37, 146.63, 166.7, 142.35, 165.93, 140.6, 157.53, 133.67, 153.67, 128.65, 143.52, 128.72, 152.15, 154.75, 130.83, 157.97, 128.4, 155.63, 131.33, 166.35, 127.18, 156.88, 129.25, 159.5, 162.58, 138.77, 150.62, 129.45, 154.33, 128.98, 154.05, 157.32, 153.83, 154.5, 128.37, 156.12, 127.38, 153.22, 128.9, 154.4, 132.08, 170.07, 143.82, 183.18, 139.93, 172.2, 178.28, 144.67, 174.32, 145.32, 172.7, 143.6, 167.6, 141.53, 167.18, 144.4, 169.68, 169.1, 139.67, 167.6, 137.65, 156.75, 130.07, 138.88, 128.3, 130.87, 126.77, 128.73, 127.33, 135.38, 138.13, 134.27, 132.02, 129.32, 138.13, 139.22, 137.62, 140.93, 139.15, 147.87, 127.4, 149.68, 147.53, 149.37, 149.87, 145.77, 144.33, 131.03, 154.57, 128.88, 148.42, 132.7, 148.43, 141.23, 145.3, 146.7, 147.22, 139.62, 141.12, 135.05, 138.03, 137.0, 137.87, 135.52, 136.45, 133.25, 133.22, 139.28, 141.08, 135.48, 134.23, 142.45, 139.45, 147.32, 131.48, 149.55, 154.33, 126.62, 151.77, 130.43, 158.17, 157.42, 129.85, 160.15, 131.9, 155.77, 132.28, 160.12, 133.3, 156.87, 129.97, 159.67, 135.48, 153.8, 147.47, 130.62, 150.77, 129.02, 145.63, 126.8, 152.97, 130.12, 150.7, 128.33, 150.97, 127.58, 152.92, 151.23, 147.6, 128.07, 150.38, 149.68, 148.33, 151.95, 150.72, 128.75, 148.35, 149.6, 131.28, 148.13, 126.58, 153.7, 128.3, 153.55, 127.18, 150.28, 153.12, 127.55, 152.58, 157.52, 129.07, 151.53, 126.85, 162.25, 137.92, 156.2, 133.35, 128.05, 153.48, 135.75, 154.57, 157.7, 136.38, 159.78, 134.85, 160.18, 135.63, 158.53, 140.45, 160.87, 131.58, 158.35, 138.18, 164.77, 163.05, 140.73, 162.33, 137.35, 163.48, 130.9, 164.53, 139.88, 161.82, 131.73, 153.77, 128.0, 160.42, 129.73, 160.58, 158.18, 132.03, 155.4, 131.33, 164.15, 138.93, 161.93, 139.42, 160.4, 142.88, 164.45, 130.83, 169.88, 140.57, 163.32, 159.67, 136.27, 168.18, 139.98, 165.13, 140.12, 162.15, 146.45, 163.1, 135.52, 169.75, 136.92, 168.17, 162.0, 143.22, 174.8, 151.47, 177.48, 150.17, 171.63, 150.18, 167.72, 147.23, 162.17, 131.82, 154.58, 151.85, 127.2, 148.6, 127.83, 148.55, 126.97, 140.95, 138.67, 138.63, 137.38, 132.62, 132.4, 132.42, 127.77, 127.43, 129.18, 126.48, 127.55, 127.73, 128.55, 129.07, 127.67, 133.93, 137.05, 133.58, 134.23, 132.5, 133.77, 138.42, 135.55, 143.6, 137.6, 136.28, 140.4, 140.13, 145.3, 147.17, 146.07, 149.35, 148.75, 143.27, 141.92, 141.47, 145.03, 141.73, 144.03, 144.63, 142.72, 149.4, 137.32, 138.0, 127.42, 140.88, 145.07, 145.77, 141.7, 145.88, 146.2, 139.32, 141.1, 146.73, 145.47, 142.22, 138.53, 144.4, 143.78, 141.63, 145.7, 145.15, 142.13, 141.83, 135.82, 141.67, 139.03, 135.85, 139.9, 132.72, 130.4, 128.85, 130.77, 132.33, 136.75, 132.4, 138.55, 136.12, 134.13, 133.47, 134.95, 135.05, 132.83, 129.32, 132.83, 143.38, 135.95, 141.28, 141.45, 141.2, 145.75, 139.43, 141.78, 140.5, 145.58, 144.77, 141.15, 138.27, 140.23, 132.05, 130.53, 134.72, 133.48, 143.83, 144.15, 126.75, 158.25, 136.72, 161.08, 136.53, 153.47, 141.02, 170.25, 159.25, 135.17, 156.63, 129.05, 164.12, 136.38, 157.83, 128.75, 159.17, 131.3, 160.7, 131.25, 163.23, 163.1, 151.18, 175.95, 151.45, 174.52, 147.77, 173.75, 153.55, 182.1, 146.38, 176.7, 146.1, 177.3, 180.97, 153.52, 177.1, 148.17, 176.02, 151.53, 173.8, 146.45, 176.98, 147.47, 175.68, 149.58, 181.57, 176.18, 153.95, 181.08, 160.57, 179.25, 163.83, 179.08, 157.45, 179.07, 151.58, 178.62, 186.15, 159.62, 191.85, 159.78, 177.12, 154.57, 182.58, 151.68, 170.33, 150.72, 178.32, 154.53, 171.88, 176.25, 154.57, 168.97, 158.67, 179.33, 147.08, 178.02, 148.28, 168.58, 143.3, 170.2, 167.52, 136.22, 169.93, 136.68, 160.85, 135.2, 161.32, 133.82, 156.43, 137.13, 160.57, 132.8, 165.47, 162.38, 136.7, 162.68, 134.55, 148.28, 145.03, 151.25, 150.68, 140.83, 147.97, 132.47, 133.12, 142.97, 137.83, 150.12, 132.58, 151.68, 146.35, 170.8, 133.82, 174.68, 127.28, 140.02, 178.17, 132.35, 170.67, 143.53, 133.55, 175.97, 145.43, 157.72, 142.55, 158.52, 138.22, 156.87, 128.68, 149.83, 140.45, 137.87, 127.78, 128.32, 131.02, 132.85, 127.48, 126.88, 129.78, 130.07, 127.68, 128.45, 129.62, 132.53, 129.2, 130.18, 133.58, 132.53, 129.63, 128.25, 126.6, 131.45, 129.47, 129.72, 138.85, 136.62, 149.22, 154.15, 153.67, 129.45, 152.02, 135.52, 164.8, 129.18, 170.65, 136.35, 167.52, 140.33, 170.78, 143.62, 171.8, 176.67, 144.55, 170.23, 142.83, 169.2, 139.17, 163.28, 140.23, 170.98, 134.48, 155.02, 136.05, 160.63, 152.03, 128.23, 148.67, 145.1, 140.83, 140.15, 127.72, 126.65, 126.73, 128.48, 127.17, 128.45, 127.37, 127.28, 127.57, 128.47, 128.78, 130.3, 131.02, 129.32, 130.93, 130.87, 131.85, 128.05, 128.58, 126.67, 128.73, 127.82, 132.53, 129.15, 129.03, 126.78, 127.02, 133.52, 126.6, 128.5, 127.55, 129.83, 127.27, 129.53, 133.9, 129.93, 134.48, 139.55, 130.4, 136.43, 139.98, 132.28, 131.58, 130.5, 134.08, 135.38, 132.37, 128.67, 131.7, 133.47, 135.65, 133.0, 133.22, 127.63, 126.52, 130.32, 130.38, 135.43, 137.1, 128.07, 133.15, 161.6, 155.42, 126.92, 153.65, 151.97, 140.47, 140.8, 138.4, 130.22, 130.38, 132.12, 134.95, 136.03, 143.3, 143.98, 134.85, 135.3, 135.38, 133.15, 137.0, 127.63, 142.8, 127.18, 154.85, 154.78, 132.68, 127.8, 151.28, 132.03, 143.18, 145.57, 140.08, 133.18, 135.03, 140.58, 134.8, 134.93, 136.7, 132.15, 127.07, 132.88, 134.23, 142.23, 139.97, 142.48, 135.85, 134.25, 132.35, 154.9, 128.3, 146.03, 126.98, 127.63, 128.53, 132.37, 127.85, 133.7, 128.75, 127.47, 156.68, 159.45, 161.65, 176.55, 144.85, 158.32, 135.13, 130.85, 127.82, 159.43, 127.08, 152.07, 159.53, 134.4, 146.22, 143.53, 146.9, 141.55, 146.4, 139.63, 149.67, 163.18, 131.05, 162.68, 134.98, 168.75, 136.92, 160.72, 129.45, 165.83, 133.25, 163.27, 153.97, 131.75, 155.25, 139.48, 159.58, 132.82, 173.02, 145.58, 154.22, 128.52, 161.57, 133.0, 156.63, 158.48, 130.83, 158.7, 133.6, 158.73, 132.67, 156.33, 132.53, 166.67, 140.97, 174.08, 145.08, 187.25, 179.0, 145.8, 186.12, 128.58, 151.23, 192.42, 159.28, 182.38, 149.87, 191.98, 132.32, 155.17, 201.9, 129.88, 132.15, 192.6, 157.32, 127.57, 187.23, 157.72, 127.18, 192.3, 150.08, 128.05, 187.73, 151.82, 186.0, 127.58, 151.78, 183.83, 129.18, 175.87, 145.7, 159.2, 135.55, 166.92, 135.33, 175.22, 141.13, 161.17, 127.53, 160.98, 144.37, 145.73, 140.98, 135.33, 140.45, 128.97, 133.13, 146.12, 128.98, 161.52, 136.5, 174.63, 171.13, 143.15, 167.67, 142.98, 162.18, 138.97, 154.93, 131.97, 160.97, 129.32, 161.2, 132.12, 162.23, 156.62, 136.1, 154.18, 129.23, 141.87, 146.62, 144.58, 135.48, 137.63, 146.55, 149.83, 130.65, 146.93, 144.5, 151.92, 146.92, 140.92, 142.3, 139.97, 149.32, 147.7, 141.18, 146.28, 147.32, 143.47, 141.93, 145.57, 151.87, 128.12, 162.58, 131.55, 157.35, 167.12, 155.52, 151.67, 130.92, 141.42, 143.67, 142.47, 143.45, 129.93, 139.12, 139.57, 137.88, 138.77, 151.78, 126.73, 149.27, 140.15, 145.57, 142.38, 140.73, 148.7, 140.28, 143.33, 138.62, 132.67, 133.15, 134.58, 136.97, 143.95, 143.4, 140.8, 139.15, 141.83, 130.15, 129.55, 134.0, 131.28, 132.47, 131.13, 133.67, 141.38, 137.02, 136.1, 131.4, 141.2, 146.65, 134.45, 171.73, 173.78, 156.75, 162.03, 144.42, 167.53, 145.33, 172.75, 137.82, 164.57, 130.3, 149.18, 137.58, 162.87, 150.08, 128.72, 151.12, 151.18, 149.13, 139.65, 143.98, 139.2, 138.27, 135.23, 135.08, 134.55, 148.75, 145.77, 144.82, 141.37, 141.88, 139.33, 136.22, 136.35, 134.23, 130.45, 137.57, 139.07, 131.98, 126.52, 128.32, 128.8, 132.32, 129.3, 127.07, 126.72, 129.13, 134.08, 128.05, 129.72, 133.87, 134.73, 130.85, 126.53, 127.05, 128.87, 127.88, 126.97, 126.7, 130.42, 129.85, 132.53, 144.13, 151.12, 128.13, 153.35, 132.5, 156.92, 151.82, 151.98, 147.72, 142.38, 137.72, 130.57, 127.93, 138.48, 140.17, 143.73, 137.03, 139.82, 138.87, 139.65, 141.05, 139.4, 140.6, 143.37, 137.13, 142.85, 139.42, 141.9, 135.45, 138.42, 140.08, 144.15, 138.45, 136.85, 132.92, 128.23, 132.53, 127.53, 127.42, 126.57, 127.47, 127.93, 126.57, 130.9, 129.23, 127.93, 135.37, 128.23, 130.53, 130.82, 135.48, 127.33, 129.43, 127.03, 129.57, 136.68, 127.5, 128.45, 128.55, 126.52, 127.2, 127.0, 127.92, 128.52, 129.78, 129.45, 132.05, 127.45, 127.97, 127.58, 127.83, 127.67, 126.97, 132.03, 126.33, 142.78, 151.35, 155.23, 160.57, 132.9, 159.1, 130.95, 157.82, 137.07, 155.08, 138.18, 170.85, 142.43, 165.98, 131.45, 157.13, 160.0, 131.87, 145.62, 128.87, 133.2, 137.08, 137.32, 128.5, 135.52, 139.53, 148.2, 149.65, 126.8, 153.55, 130.9, 170.72, 190.63, 151.88, 222.15, 142.13, 178.25, 182.65, 161.37, 149.58, 136.27, 133.02, 129.93, 156.65, 147.22, 144.43, 142.35, 135.23, 138.17, 135.82, 131.27]\n"
          ]
        }
      ],
      "source": [
        "import numpy as np\n",
        "outliers = []\n",
        "def detect_outliers_zscore(data):\n",
        "    thres = 3\n",
        "    mean = np.mean(data)\n",
        "    std = np.std(data)\n",
        "    # print(mean, std)\n",
        "    for i in data:\n",
        "        z_score = (i-mean)/std\n",
        "        if (np.abs(z_score) > thres):\n",
        "            outliers.append(i)\n",
        "    return outliers# Driver code\n",
        "sample_outliers = detect_outliers_zscore(train[\"Sensor1_PM2.5\"])\n",
        "print(\"Outliers from Z-scores method: \", sample_outliers)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8WRl4ZpnRzeL",
        "outputId": "96735042-b247-4097-fd84-d0ca5f4710d2"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ID                      0\n",
              "Datetime                0\n",
              "Sensor1_PM2.5        3614\n",
              "Sensor2_PM2.5        3614\n",
              "Temperature          3549\n",
              "Relative_Humidity    3549\n",
              "Offset_fault            0\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ],
      "source": [
        "train.isnull().sum()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Fhnjcefqy19Z"
      },
      "source": [
        "If sensor1 has outliers, drop the row"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "bUwPVfl0QIpP"
      },
      "outputs": [],
      "source": [
        "train = train[train[\"Sensor1_PM2.5\"]<min(sample_outliers)].reset_index(drop = True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ghETEp-pR6pW",
        "outputId": "8990a7dc-0dbf-4f00-cb3d-88cb156071cb"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ID                      0\n",
              "Datetime                0\n",
              "Sensor1_PM2.5           0\n",
              "Sensor2_PM2.5           0\n",
              "Temperature          3533\n",
              "Relative_Humidity    3533\n",
              "Offset_fault            0\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ],
      "source": [
        "train.isnull().sum()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5Eo9Qa-By19a"
      },
      "source": [
        "Look at the Temperature and Humidity field, it makes sense because we've sorted it by time. Let's take the advantage to Foward fill the Temperature and Relative Humidity, leveraging the timing."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 834
        },
        "id": "vAQ7sKJls15P",
        "outputId": "b26c38e7-82a2-427b-acf5-fd321bcc8b04"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                     ID            Datetime  Sensor1_PM2.5  Sensor2_PM2.5  \\\n",
              "0       ID_37STUCVVPZKP 2021-10-15 16:00:31        9.17000        9.28000   \n",
              "1       ID_YZZUOMF0TQIF 2021-10-15 16:00:34        9.93000       11.25000   \n",
              "2       ID_UF68QLIPG3LE 2021-10-15 16:01:01        6.53000        3.15000   \n",
              "3       ID_KKCO4VATOSWB 2021-10-15 16:01:54        8.30000        8.35000   \n",
              "4       ID_8YNKABZP11SS 2021-10-15 16:02:06        6.87000        9.57000   \n",
              "...                 ...                 ...            ...            ...   \n",
              "289323  ID_AATX8BDC1PK6 2022-01-21 07:33:21       49.82000       46.03000   \n",
              "289324  ID_1B5HQEHUQK3H 2022-01-21 07:33:40       40.25000       31.67000   \n",
              "289325  ID_5JPOTNIC5BNN 2022-01-21 07:34:18       44.30000       52.37000   \n",
              "289326  ID_RCP3FX3CB2CK 2022-01-21 07:34:46       48.83000       46.25000   \n",
              "289327  ID_FFX2NV9UA6PX 2022-01-21 07:34:57       40.43000       31.93000   \n",
              "\n",
              "        Temperature  Relative_Humidity  Offset_fault  \n",
              "0          29.90000           53.00000             0  \n",
              "1          29.90000           53.00000             0  \n",
              "2          29.90000           53.00000             1  \n",
              "3          29.90000           53.00000             0  \n",
              "4          29.90000           53.00000             0  \n",
              "...             ...                ...           ...  \n",
              "289323     16.90000           79.00000             0  \n",
              "289324     16.90000           79.00000             1  \n",
              "289325     16.90000           79.00000             0  \n",
              "289326     16.90000           79.00000             0  \n",
              "289327     16.90000           79.00000             1  \n",
              "\n",
              "[289328 rows x 7 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a12dbb7c-bfac-4131-9885-9b1abe1c20e9\">\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>ID</th>\n",
              "      <th>Datetime</th>\n",
              "      <th>Sensor1_PM2.5</th>\n",
              "      <th>Sensor2_PM2.5</th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Relative_Humidity</th>\n",
              "      <th>Offset_fault</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ID_37STUCVVPZKP</td>\n",
              "      <td>2021-10-15 16:00:31</td>\n",
              "      <td>9.17000</td>\n",
              "      <td>9.28000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ID_YZZUOMF0TQIF</td>\n",
              "      <td>2021-10-15 16:00:34</td>\n",
              "      <td>9.93000</td>\n",
              "      <td>11.25000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ID_UF68QLIPG3LE</td>\n",
              "      <td>2021-10-15 16:01:01</td>\n",
              "      <td>6.53000</td>\n",
              "      <td>3.15000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ID_KKCO4VATOSWB</td>\n",
              "      <td>2021-10-15 16:01:54</td>\n",
              "      <td>8.30000</td>\n",
              "      <td>8.35000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ID_8YNKABZP11SS</td>\n",
              "      <td>2021-10-15 16:02:06</td>\n",
              "      <td>6.87000</td>\n",
              "      <td>9.57000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</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",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>289323</th>\n",
              "      <td>ID_AATX8BDC1PK6</td>\n",
              "      <td>2022-01-21 07:33:21</td>\n",
              "      <td>49.82000</td>\n",
              "      <td>46.03000</td>\n",
              "      <td>16.90000</td>\n",
              "      <td>79.00000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>289324</th>\n",
              "      <td>ID_1B5HQEHUQK3H</td>\n",
              "      <td>2022-01-21 07:33:40</td>\n",
              "      <td>40.25000</td>\n",
              "      <td>31.67000</td>\n",
              "      <td>16.90000</td>\n",
              "      <td>79.00000</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>289325</th>\n",
              "      <td>ID_5JPOTNIC5BNN</td>\n",
              "      <td>2022-01-21 07:34:18</td>\n",
              "      <td>44.30000</td>\n",
              "      <td>52.37000</td>\n",
              "      <td>16.90000</td>\n",
              "      <td>79.00000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>289326</th>\n",
              "      <td>ID_RCP3FX3CB2CK</td>\n",
              "      <td>2022-01-21 07:34:46</td>\n",
              "      <td>48.83000</td>\n",
              "      <td>46.25000</td>\n",
              "      <td>16.90000</td>\n",
              "      <td>79.00000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>289327</th>\n",
              "      <td>ID_FFX2NV9UA6PX</td>\n",
              "      <td>2022-01-21 07:34:57</td>\n",
              "      <td>40.43000</td>\n",
              "      <td>31.93000</td>\n",
              "      <td>16.90000</td>\n",
              "      <td>79.00000</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>289328 rows × 7 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a12dbb7c-bfac-4131-9885-9b1abe1c20e9')\"\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-a12dbb7c-bfac-4131-9885-9b1abe1c20e9 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-a12dbb7c-bfac-4131-9885-9b1abe1c20e9');\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": 12
        }
      ],
      "source": [
        "\n",
        "# applying ffill() method to fill the missing values\n",
        "test.ffill(axis = 0)\n",
        "\n",
        "# applying ffill() method to fill the missing values\n",
        "train.ffill(axis = 0)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "id": "DB8dd5jtuG80"
      },
      "outputs": [],
      "source": [
        "test_orig = test.copy()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5ObzuksBADAE",
        "outputId": "6e2c6f43-dfe1-4bf2-c79f-53cab32b5c64"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((289328, 7), (127361, 6), (127361, 2))"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ],
      "source": [
        "# Check size and shape of datasets\n",
        "train.shape, test.shape, samplesubmission.shape"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "S99gPI3isQYd",
        "outputId": "ae67bfe3-e71b-46f7-9555-5236decb9f3f"
      },
      "source": [
        "Descriptions and Visualizations."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 625
        },
        "id": "u6JhNcyCAF_-",
        "outputId": "256c7707-105f-4732-d2a2-8b471b5c912f"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                     ID             Datetime  Sensor1_PM2.5  Sensor2_PM2.5  \\\n",
              "count            289328               289328   289328.00000   289328.00000   \n",
              "unique           289328               286059            NaN            NaN   \n",
              "top     ID_37STUCVVPZKP  2021-11-06 18:29:26            NaN            NaN   \n",
              "freq                  1                    3            NaN            NaN   \n",
              "first               NaN  2021-10-15 16:00:31            NaN            NaN   \n",
              "last                NaN  2022-01-21 07:34:57            NaN            NaN   \n",
              "mean                NaN                  NaN       39.04612       36.57356   \n",
              "std                 NaN                  NaN       23.10803       23.57386   \n",
              "min                 NaN                  NaN       -9.45000       -9.70000   \n",
              "25%                 NaN                  NaN       22.18000       19.32000   \n",
              "50%                 NaN                  NaN       36.12000       33.37000   \n",
              "75%                 NaN                  NaN       48.72000       46.88000   \n",
              "max                 NaN                  NaN      126.32000      208.87000   \n",
              "\n",
              "        Temperature  Relative_Humidity  Offset_fault  \n",
              "count  285795.00000       285795.00000  289328.00000  \n",
              "unique          NaN                NaN           NaN  \n",
              "top             NaN                NaN           NaN  \n",
              "freq            NaN                NaN           NaN  \n",
              "first           NaN                NaN           NaN  \n",
              "last            NaN                NaN           NaN  \n",
              "mean       23.36399           78.51096       0.37519  \n",
              "std         3.84427           14.83530       0.48417  \n",
              "min        16.70000           27.00000       0.00000  \n",
              "25%        20.40000           67.00000       0.00000  \n",
              "50%        22.00000           84.00000       0.00000  \n",
              "75%        26.40000           91.00000       1.00000  \n",
              "max        34.90000           99.00000       1.00000  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-749def24-a141-4a92-bfe4-55cf1fb00ac1\">\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>ID</th>\n",
              "      <th>Datetime</th>\n",
              "      <th>Sensor1_PM2.5</th>\n",
              "      <th>Sensor2_PM2.5</th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Relative_Humidity</th>\n",
              "      <th>Offset_fault</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>289328</td>\n",
              "      <td>289328</td>\n",
              "      <td>289328.00000</td>\n",
              "      <td>289328.00000</td>\n",
              "      <td>285795.00000</td>\n",
              "      <td>285795.00000</td>\n",
              "      <td>289328.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>unique</th>\n",
              "      <td>289328</td>\n",
              "      <td>286059</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>top</th>\n",
              "      <td>ID_37STUCVVPZKP</td>\n",
              "      <td>2021-11-06 18:29:26</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>freq</th>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>first</th>\n",
              "      <td>NaN</td>\n",
              "      <td>2021-10-15 16:00:31</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>last</th>\n",
              "      <td>NaN</td>\n",
              "      <td>2022-01-21 07:34:57</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>39.04612</td>\n",
              "      <td>36.57356</td>\n",
              "      <td>23.36399</td>\n",
              "      <td>78.51096</td>\n",
              "      <td>0.37519</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>23.10803</td>\n",
              "      <td>23.57386</td>\n",
              "      <td>3.84427</td>\n",
              "      <td>14.83530</td>\n",
              "      <td>0.48417</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>-9.45000</td>\n",
              "      <td>-9.70000</td>\n",
              "      <td>16.70000</td>\n",
              "      <td>27.00000</td>\n",
              "      <td>0.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>22.18000</td>\n",
              "      <td>19.32000</td>\n",
              "      <td>20.40000</td>\n",
              "      <td>67.00000</td>\n",
              "      <td>0.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>36.12000</td>\n",
              "      <td>33.37000</td>\n",
              "      <td>22.00000</td>\n",
              "      <td>84.00000</td>\n",
              "      <td>0.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>48.72000</td>\n",
              "      <td>46.88000</td>\n",
              "      <td>26.40000</td>\n",
              "      <td>91.00000</td>\n",
              "      <td>1.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>126.32000</td>\n",
              "      <td>208.87000</td>\n",
              "      <td>34.90000</td>\n",
              "      <td>99.00000</td>\n",
              "      <td>1.00000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-749def24-a141-4a92-bfe4-55cf1fb00ac1')\"\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-749def24-a141-4a92-bfe4-55cf1fb00ac1 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-749def24-a141-4a92-bfe4-55cf1fb00ac1');\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": 15
        }
      ],
      "source": [
        "# Train statistical summary\n",
        "train.describe(include = 'all')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 351
        },
        "id": "ilBcZjhWAH5G",
        "outputId": "58bb24fa-2f7d-4faa-b88a-117243344d9a"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# Target variable distribution\n",
        "sns.set_style('darkgrid')\n",
        "plt.figure(figsize=(8, 5))\n",
        "sns.countplot(train[\"Offset_fault\"])\n",
        "plt.title('Target variable distribution')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3rrfuR1zAK_S",
        "outputId": "66c550bc-47a5-4bfd-fd45-aa57aa112ff9"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(True, True)"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ],
      "source": [
        "# Check for missing values\n",
        "train.isnull().sum().any(), test.isnull().sum().any() "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 609
        },
        "id": "AMwsKL5OAMsk",
        "outputId": "e66bf8ff-d9a5-459d-b45c-d1df578bcbec"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 648x720 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# Plot missing values in train set\n",
        "ax = train.isna().sum().sort_values().plot(kind = 'barh', figsize = (9, 10))\n",
        "plt.title('Percentage of Missing Values Per Column in Train Set', fontdict={'size':15})\n",
        "for p in ax.patches:\n",
        "    percentage ='{:,.0f}%'.format((p.get_width()/train.shape[0])*100)\n",
        "    width, height =p.get_width(),p.get_height()\n",
        "    x=p.get_x()+width+0.02\n",
        "    y=p.get_y()+height/2\n",
        "    ax.annotate(percentage,(x,y))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ah4vOdG2y19g"
      },
      "source": [
        "Other feature engineering"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "id": "9PeUb5X3AOGm"
      },
      "outputs": [],
      "source": [
        "target = train.Offset_fault\n",
        "train=train.drop(\"Offset_fault\", axis =1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 112
        },
        "id": "J6fteJzIARyQ",
        "outputId": "5b524024-625c-4837-cbda-ae39a44f6945"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                ID            Datetime  Sensor1_PM2.5  Sensor2_PM2.5  \\\n",
              "0  ID_37STUCVVPZKP 2021-10-15 16:00:31        9.17000        9.28000   \n",
              "1  ID_YZZUOMF0TQIF 2021-10-15 16:00:34        9.93000       11.25000   \n",
              "\n",
              "   Temperature  Relative_Humidity  \n",
              "0     29.90000           53.00000  \n",
              "1     29.90000           53.00000  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-377c70cb-725c-4d33-9738-0b3f952d4098\">\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>ID</th>\n",
              "      <th>Datetime</th>\n",
              "      <th>Sensor1_PM2.5</th>\n",
              "      <th>Sensor2_PM2.5</th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Relative_Humidity</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ID_37STUCVVPZKP</td>\n",
              "      <td>2021-10-15 16:00:31</td>\n",
              "      <td>9.17000</td>\n",
              "      <td>9.28000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ID_YZZUOMF0TQIF</td>\n",
              "      <td>2021-10-15 16:00:34</td>\n",
              "      <td>9.93000</td>\n",
              "      <td>11.25000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-377c70cb-725c-4d33-9738-0b3f952d4098')\"\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-377c70cb-725c-4d33-9738-0b3f952d4098 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-377c70cb-725c-4d33-9738-0b3f952d4098');\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": 20
        }
      ],
      "source": [
        "train.head(2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "id": "t4_hnlATApwH"
      },
      "outputs": [],
      "source": [
        "data = train.append(test)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "id": "ur_hrTK1A8mT"
      },
      "outputs": [],
      "source": [
        "# Convert datetime variables to datetime objects\n",
        "data[\"Datetime\"] = pd.to_datetime(data.Datetime)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 112
        },
        "id": "d5BC8lwsCQUl",
        "outputId": "10fa098e-d644-4abb-d934-0d70262e92ab"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                ID            Datetime  Sensor1_PM2.5  Sensor2_PM2.5  \\\n",
              "0  ID_37STUCVVPZKP 2021-10-15 16:00:31        9.17000        9.28000   \n",
              "1  ID_YZZUOMF0TQIF 2021-10-15 16:00:34        9.93000       11.25000   \n",
              "\n",
              "   Temperature  Relative_Humidity  \n",
              "0     29.90000           53.00000  \n",
              "1     29.90000           53.00000  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-7f4a4c1d-c084-4b9c-8bce-2d9ff034d1cc\">\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>ID</th>\n",
              "      <th>Datetime</th>\n",
              "      <th>Sensor1_PM2.5</th>\n",
              "      <th>Sensor2_PM2.5</th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Relative_Humidity</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ID_37STUCVVPZKP</td>\n",
              "      <td>2021-10-15 16:00:31</td>\n",
              "      <td>9.17000</td>\n",
              "      <td>9.28000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ID_YZZUOMF0TQIF</td>\n",
              "      <td>2021-10-15 16:00:34</td>\n",
              "      <td>9.93000</td>\n",
              "      <td>11.25000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-7f4a4c1d-c084-4b9c-8bce-2d9ff034d1cc')\"\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-7f4a4c1d-c084-4b9c-8bce-2d9ff034d1cc 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-7f4a4c1d-c084-4b9c-8bce-2d9ff034d1cc');\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": 23
        }
      ],
      "source": [
        "data.head(2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "metadata": {
        "id": "APpICs6swM2V"
      },
      "outputs": [],
      "source": [
        "def time_of_day(number):\n",
        "  if(number>5 and number<12):\n",
        "    return(1)\n",
        "  elif(number>12 and number<17):\n",
        "    return(2)\n",
        "  else:\n",
        "    return(3)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 112
        },
        "id": "27jXRsVKwkSx",
        "outputId": "904b5a5c-a1d4-4eae-ab76-57cd075bd35d"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                ID            Datetime  Sensor1_PM2.5  Sensor2_PM2.5  \\\n",
              "0  ID_37STUCVVPZKP 2021-10-15 16:00:31        9.17000        9.28000   \n",
              "1  ID_YZZUOMF0TQIF 2021-10-15 16:00:34        9.93000       11.25000   \n",
              "\n",
              "   Temperature  Relative_Humidity  \n",
              "0     29.90000           53.00000  \n",
              "1     29.90000           53.00000  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-b9cf2057-2427-42c3-988f-6f8696694c16\">\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>ID</th>\n",
              "      <th>Datetime</th>\n",
              "      <th>Sensor1_PM2.5</th>\n",
              "      <th>Sensor2_PM2.5</th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Relative_Humidity</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ID_37STUCVVPZKP</td>\n",
              "      <td>2021-10-15 16:00:31</td>\n",
              "      <td>9.17000</td>\n",
              "      <td>9.28000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ID_YZZUOMF0TQIF</td>\n",
              "      <td>2021-10-15 16:00:34</td>\n",
              "      <td>9.93000</td>\n",
              "      <td>11.25000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-b9cf2057-2427-42c3-988f-6f8696694c16')\"\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-b9cf2057-2427-42c3-988f-6f8696694c16 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-b9cf2057-2427-42c3-988f-6f8696694c16');\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
        }
      ],
      "source": [
        "data.head(2)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PzlS9Lxcy19k"
      },
      "source": [
        "More Feature Engineering"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "id": "RJ0VTNJStt2s"
      },
      "outputs": [],
      "source": [
        "data[\"Sensor difference\"] = (data[\"Sensor1_PM2.5\"] - data[\"Sensor2_PM2.5\"]).abs()\n",
        "data[\"Sensor div\"] = (data[\"Sensor1_PM2.5\"] / data[\"Sensor2_PM2.5\"]).abs()\n",
        "data[\"Sensor R/T\"] = (data[\"Relative_Humidity\"] / data[\"Temperature\"]).abs()\n",
        "data[\"SD\"] = data[[\"Sensor1_PM2.5\",\"Sensor2_PM2.5\"]].std(axis=1)\n",
        "data[\"Mean\"] = data[[\"Sensor1_PM2.5\",\"Sensor2_PM2.5\"]].mean(axis=1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {
        "id": "BSLc2wcX9Kz8"
      },
      "outputs": [],
      "source": [
        "data['Rolling_3'] = data['Sensor1_PM2.5'].rolling(3).mean()\n",
        "data['Rolling_5'] = data['Sensor1_PM2.5'].rolling(5).mean()\n",
        "\n",
        "data['Rolling_2_3'] = data['Sensor2_PM2.5'].rolling(3).mean()\n",
        "data['Rolling_2_5'] = data['Sensor2_PM2.5'].rolling(5).mean()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "id": "b7mztWcf9S1l"
      },
      "outputs": [],
      "source": [
        "data[\"Sensor1_PM2_rolling_mean_60\"] = data.rolling(60).mean()[\"Sensor1_PM2.5\"]\n",
        "data[\"Sensor1_PM2_rolling_max_60\"] = data.rolling(60).max()[\"Sensor1_PM2.5\"]\n",
        "data[\"Sensor1_PM2_rolling_min_60\"] = data.rolling(60).min()[\"Sensor1_PM2.5\"]\n",
        "\n",
        "data[\"Sensor1_PM2_rolling_mean_30\"] = data.rolling(30).mean()[\"Sensor1_PM2.5\"]\n",
        "data[\"Sensor1_PM2_rolling_max_30\"] = data.rolling(30).max()[\"Sensor1_PM2.5\"]\n",
        "data[\"Sensor1_PM2_rolling_min_30\"] = data.rolling(30).min()[\"Sensor1_PM2.5\"]\n",
        "\n",
        "data[\"Sensor1_PM2_rolling_mean_10\"] = data.rolling(10).mean()[\"Sensor1_PM2.5\"]\n",
        "data[\"Sensor1_PM2_rolling_max_10\"] = data.rolling(10).max()[\"Sensor1_PM2.5\"]\n",
        "data[\"Sensor1_PM2_rolling_min_10\"] = data.rolling(10).min()[\"Sensor1_PM2.5\"]\n",
        "\n",
        "\n",
        "data[\"Sensor2_PM2_rolling_mean_60\"] = data.rolling(60).mean()[\"Sensor2_PM2.5\"]\n",
        "data[\"Sensor2_PM2_rolling_max_60\"] = data.rolling(60).max()[\"Sensor2_PM2.5\"]\n",
        "data[\"Sensor2_PM2_rolling_min_60\"] = data.rolling(60).min()[\"Sensor2_PM2.5\"]\n",
        "\n",
        "data[\"Sensor2_PM2_rolling_mean_30\"] = data.rolling(30).mean()[\"Sensor2_PM2.5\"]\n",
        "data[\"Sensor2_PM2_rolling_max_30\"] = data.rolling(30).max()[\"Sensor2_PM2.5\"]\n",
        "data[\"Sensor2_PM2_rolling_min_30\"] = data.rolling(30).min()[\"Sensor2_PM2.5\"]\n",
        "\n",
        "data[\"Sensor2_PM2_rolling_mean_10\"] = data.rolling(10).mean()[\"Sensor2_PM2.5\"]\n",
        "data[\"Sensor2_PM2_rolling_max_10\"] = data.rolling(10).max()[\"Sensor2_PM2.5\"]\n",
        "data[\"Sensor2_PM2_rolling_min_10\"] = data.rolling(10).min()[\"Sensor2_PM2.5\"]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ooXH-rZX7eGe",
        "outputId": "0ad1af46-e5fe-4d00-a2c9-23e8d7d7ddc7"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "34.9"
            ]
          },
          "metadata": {},
          "execution_count": 29
        }
      ],
      "source": [
        "data.Relative_Humidity.min()\n",
        "data.Temperature.max()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "metadata": {
        "id": "JcFs8947Axm6"
      },
      "outputs": [],
      "source": [
        "# Extract day, month year and hour from the Datetime column\n",
        "# day\n",
        "data['Datetime_day'] = data.Datetime.dt.day\n",
        "\n",
        "# month\n",
        "data['Datetime_month'] = data.Datetime.dt.month\n",
        " \n",
        "# year\n",
        "#data['Datetime_year'] = data.Datetime.dt.year\n",
        "\n",
        "# hour\n",
        "data['Datetime_hour'] = data.Datetime.dt.hour\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 292
        },
        "id": "v54iJdRjIM_L",
        "outputId": "5042f0bc-0467-4c81-d8a5-18a88102a4a5"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                ID            Datetime  Sensor1_PM2.5  Sensor2_PM2.5  \\\n",
              "0  ID_37STUCVVPZKP 2021-10-15 16:00:31        9.17000        9.28000   \n",
              "1  ID_YZZUOMF0TQIF 2021-10-15 16:00:34        9.93000       11.25000   \n",
              "\n",
              "   Temperature  Relative_Humidity  Sensor difference  Sensor div  Sensor R/T  \\\n",
              "0     29.90000           53.00000            0.11000     0.98815     1.77258   \n",
              "1     29.90000           53.00000            1.32000     0.88267     1.77258   \n",
              "\n",
              "       SD  ...  Sensor2_PM2_rolling_min_60  Sensor2_PM2_rolling_mean_30  \\\n",
              "0 0.07778  ...                         NaN                          NaN   \n",
              "1 0.93338  ...                         NaN                          NaN   \n",
              "\n",
              "   Sensor2_PM2_rolling_max_30  Sensor2_PM2_rolling_min_30  \\\n",
              "0                         NaN                         NaN   \n",
              "1                         NaN                         NaN   \n",
              "\n",
              "   Sensor2_PM2_rolling_mean_10  Sensor2_PM2_rolling_max_10  \\\n",
              "0                          NaN                         NaN   \n",
              "1                          NaN                         NaN   \n",
              "\n",
              "   Sensor2_PM2_rolling_min_10  Datetime_day  Datetime_month  Datetime_hour  \n",
              "0                         NaN            15              10             16  \n",
              "1                         NaN            15              10             16  \n",
              "\n",
              "[2 rows x 36 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-d7b8bd60-5902-4080-8e36-c5d5720e52fe\">\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>ID</th>\n",
              "      <th>Datetime</th>\n",
              "      <th>Sensor1_PM2.5</th>\n",
              "      <th>Sensor2_PM2.5</th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Relative_Humidity</th>\n",
              "      <th>Sensor difference</th>\n",
              "      <th>Sensor div</th>\n",
              "      <th>Sensor R/T</th>\n",
              "      <th>SD</th>\n",
              "      <th>...</th>\n",
              "      <th>Sensor2_PM2_rolling_min_60</th>\n",
              "      <th>Sensor2_PM2_rolling_mean_30</th>\n",
              "      <th>Sensor2_PM2_rolling_max_30</th>\n",
              "      <th>Sensor2_PM2_rolling_min_30</th>\n",
              "      <th>Sensor2_PM2_rolling_mean_10</th>\n",
              "      <th>Sensor2_PM2_rolling_max_10</th>\n",
              "      <th>Sensor2_PM2_rolling_min_10</th>\n",
              "      <th>Datetime_day</th>\n",
              "      <th>Datetime_month</th>\n",
              "      <th>Datetime_hour</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ID_37STUCVVPZKP</td>\n",
              "      <td>2021-10-15 16:00:31</td>\n",
              "      <td>9.17000</td>\n",
              "      <td>9.28000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>0.11000</td>\n",
              "      <td>0.98815</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>0.07778</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ID_YZZUOMF0TQIF</td>\n",
              "      <td>2021-10-15 16:00:34</td>\n",
              "      <td>9.93000</td>\n",
              "      <td>11.25000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>1.32000</td>\n",
              "      <td>0.88267</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>0.93338</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>2 rows × 36 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-d7b8bd60-5902-4080-8e36-c5d5720e52fe')\"\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-d7b8bd60-5902-4080-8e36-c5d5720e52fe 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-d7b8bd60-5902-4080-8e36-c5d5720e52fe');\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": 31
        }
      ],
      "source": [
        "data.head(2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "metadata": {
        "id": "XfYqcutOIUFA"
      },
      "outputs": [],
      "source": [
        "data.drop(\"Datetime\", axis = 1, inplace = True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 33,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 474
        },
        "id": "_Jt1DGm_Ia5l",
        "outputId": "2a0a7be9-12e9-4d4f-f3ab-1eb5e3a71885"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                ID  Sensor1_PM2.5  Sensor2_PM2.5  Temperature  \\\n",
              "0  ID_37STUCVVPZKP        9.17000        9.28000     29.90000   \n",
              "1  ID_YZZUOMF0TQIF        9.93000       11.25000     29.90000   \n",
              "2  ID_UF68QLIPG3LE        6.53000        3.15000     29.90000   \n",
              "3  ID_KKCO4VATOSWB        8.30000        8.35000     29.90000   \n",
              "4  ID_8YNKABZP11SS        6.87000        9.57000     29.90000   \n",
              "5  ID_4M1B7B99G22Z        4.73000        2.37000     29.90000   \n",
              "6  ID_0NN3PW2X3D2J        8.47000        9.73000     29.90000   \n",
              "7  ID_EKSBUQK9HYXN        5.88000        3.62000     29.90000   \n",
              "8  ID_Z5SXLLBTR2DG        9.15000       10.85000     29.90000   \n",
              "9  ID_04QJPRKRN029        9.28000        8.67000     29.90000   \n",
              "\n",
              "   Relative_Humidity  Sensor difference  Sensor div  Sensor R/T      SD  \\\n",
              "0           53.00000            0.11000     0.98815     1.77258 0.07778   \n",
              "1           53.00000            1.32000     0.88267     1.77258 0.93338   \n",
              "2           53.00000            3.38000     2.07302     1.77258 2.39002   \n",
              "3           53.00000            0.05000     0.99401     1.77258 0.03536   \n",
              "4           53.00000            2.70000     0.71787     1.77258 1.90919   \n",
              "5           53.00000            2.36000     1.99578     1.77258 1.66877   \n",
              "6           53.00000            1.26000     0.87050     1.77258 0.89095   \n",
              "7           53.00000            2.26000     1.62431     1.77258 1.59806   \n",
              "8           53.00000            1.70000     0.84332     1.77258 1.20208   \n",
              "9           53.00000            0.61000     1.07036     1.77258 0.43134   \n",
              "\n",
              "      Mean  ...  Sensor2_PM2_rolling_min_60  Sensor2_PM2_rolling_mean_30  \\\n",
              "0  9.22500  ...                         NaN                          NaN   \n",
              "1 10.59000  ...                         NaN                          NaN   \n",
              "2  4.84000  ...                         NaN                          NaN   \n",
              "3  8.32500  ...                         NaN                          NaN   \n",
              "4  8.22000  ...                         NaN                          NaN   \n",
              "5  3.55000  ...                         NaN                          NaN   \n",
              "6  9.10000  ...                         NaN                          NaN   \n",
              "7  4.75000  ...                         NaN                          NaN   \n",
              "8 10.00000  ...                         NaN                          NaN   \n",
              "9  8.97500  ...                         NaN                          NaN   \n",
              "\n",
              "   Sensor2_PM2_rolling_max_30  Sensor2_PM2_rolling_min_30  \\\n",
              "0                         NaN                         NaN   \n",
              "1                         NaN                         NaN   \n",
              "2                         NaN                         NaN   \n",
              "3                         NaN                         NaN   \n",
              "4                         NaN                         NaN   \n",
              "5                         NaN                         NaN   \n",
              "6                         NaN                         NaN   \n",
              "7                         NaN                         NaN   \n",
              "8                         NaN                         NaN   \n",
              "9                         NaN                         NaN   \n",
              "\n",
              "   Sensor2_PM2_rolling_mean_10  Sensor2_PM2_rolling_max_10  \\\n",
              "0                          NaN                         NaN   \n",
              "1                          NaN                         NaN   \n",
              "2                          NaN                         NaN   \n",
              "3                          NaN                         NaN   \n",
              "4                          NaN                         NaN   \n",
              "5                          NaN                         NaN   \n",
              "6                          NaN                         NaN   \n",
              "7                          NaN                         NaN   \n",
              "8                          NaN                         NaN   \n",
              "9                      7.68400                    11.25000   \n",
              "\n",
              "   Sensor2_PM2_rolling_min_10  Datetime_day  Datetime_month  Datetime_hour  \n",
              "0                         NaN            15              10             16  \n",
              "1                         NaN            15              10             16  \n",
              "2                         NaN            15              10             16  \n",
              "3                         NaN            15              10             16  \n",
              "4                         NaN            15              10             16  \n",
              "5                         NaN            15              10             16  \n",
              "6                         NaN            15              10             16  \n",
              "7                         NaN            15              10             16  \n",
              "8                         NaN            15              10             16  \n",
              "9                     2.37000            15              10             16  \n",
              "\n",
              "[10 rows x 35 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-b487abe5-6dc1-4981-a054-4e3bac0cce6d\">\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>ID</th>\n",
              "      <th>Sensor1_PM2.5</th>\n",
              "      <th>Sensor2_PM2.5</th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Relative_Humidity</th>\n",
              "      <th>Sensor difference</th>\n",
              "      <th>Sensor div</th>\n",
              "      <th>Sensor R/T</th>\n",
              "      <th>SD</th>\n",
              "      <th>Mean</th>\n",
              "      <th>...</th>\n",
              "      <th>Sensor2_PM2_rolling_min_60</th>\n",
              "      <th>Sensor2_PM2_rolling_mean_30</th>\n",
              "      <th>Sensor2_PM2_rolling_max_30</th>\n",
              "      <th>Sensor2_PM2_rolling_min_30</th>\n",
              "      <th>Sensor2_PM2_rolling_mean_10</th>\n",
              "      <th>Sensor2_PM2_rolling_max_10</th>\n",
              "      <th>Sensor2_PM2_rolling_min_10</th>\n",
              "      <th>Datetime_day</th>\n",
              "      <th>Datetime_month</th>\n",
              "      <th>Datetime_hour</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ID_37STUCVVPZKP</td>\n",
              "      <td>9.17000</td>\n",
              "      <td>9.28000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>0.11000</td>\n",
              "      <td>0.98815</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>0.07778</td>\n",
              "      <td>9.22500</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ID_YZZUOMF0TQIF</td>\n",
              "      <td>9.93000</td>\n",
              "      <td>11.25000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>1.32000</td>\n",
              "      <td>0.88267</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>0.93338</td>\n",
              "      <td>10.59000</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ID_UF68QLIPG3LE</td>\n",
              "      <td>6.53000</td>\n",
              "      <td>3.15000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>3.38000</td>\n",
              "      <td>2.07302</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>2.39002</td>\n",
              "      <td>4.84000</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ID_KKCO4VATOSWB</td>\n",
              "      <td>8.30000</td>\n",
              "      <td>8.35000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>0.05000</td>\n",
              "      <td>0.99401</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>0.03536</td>\n",
              "      <td>8.32500</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ID_8YNKABZP11SS</td>\n",
              "      <td>6.87000</td>\n",
              "      <td>9.57000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>2.70000</td>\n",
              "      <td>0.71787</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>1.90919</td>\n",
              "      <td>8.22000</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>ID_4M1B7B99G22Z</td>\n",
              "      <td>4.73000</td>\n",
              "      <td>2.37000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>2.36000</td>\n",
              "      <td>1.99578</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>1.66877</td>\n",
              "      <td>3.55000</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>ID_0NN3PW2X3D2J</td>\n",
              "      <td>8.47000</td>\n",
              "      <td>9.73000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>1.26000</td>\n",
              "      <td>0.87050</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>0.89095</td>\n",
              "      <td>9.10000</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>ID_EKSBUQK9HYXN</td>\n",
              "      <td>5.88000</td>\n",
              "      <td>3.62000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>2.26000</td>\n",
              "      <td>1.62431</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>1.59806</td>\n",
              "      <td>4.75000</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>ID_Z5SXLLBTR2DG</td>\n",
              "      <td>9.15000</td>\n",
              "      <td>10.85000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>1.70000</td>\n",
              "      <td>0.84332</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>1.20208</td>\n",
              "      <td>10.00000</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>ID_04QJPRKRN029</td>\n",
              "      <td>9.28000</td>\n",
              "      <td>8.67000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>0.61000</td>\n",
              "      <td>1.07036</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>0.43134</td>\n",
              "      <td>8.97500</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>7.68400</td>\n",
              "      <td>11.25000</td>\n",
              "      <td>2.37000</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>10 rows × 35 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-b487abe5-6dc1-4981-a054-4e3bac0cce6d')\"\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-b487abe5-6dc1-4981-a054-4e3bac0cce6d 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-b487abe5-6dc1-4981-a054-4e3bac0cce6d');\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": 33
        }
      ],
      "source": [
        "data.head(10)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "J5MtyiHEy19s"
      },
      "source": [
        "More Visualizations"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 34,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 453
        },
        "id": "wHK2LhB-A30Y",
        "outputId": "6821c848-73c3-4422-9e2a-5fef50d63139"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1008x504 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# Datetime month countplot\n",
        "plt.figure(figsize = (14, 7))\n",
        "sns.countplot(x = 'Datetime_month', data = data)\n",
        "plt.title('Datetime month count plot')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 35,
      "metadata": {
        "id": "MHdtcwhhMNgB"
      },
      "outputs": [],
      "source": [
        "data[\"time_of_day\"] = [time_of_day(i) for i in data.Datetime_hour]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 36,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 504
        },
        "id": "dhxsTOc3Mb22",
        "outputId": "81b8dada-6785-4102-9c9a-2c9016f6acc9"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                     ID  Sensor1_PM2.5  Sensor2_PM2.5  Temperature  \\\n",
              "0       ID_37STUCVVPZKP        9.17000        9.28000     29.90000   \n",
              "1       ID_YZZUOMF0TQIF        9.93000       11.25000     29.90000   \n",
              "2       ID_UF68QLIPG3LE        6.53000        3.15000     29.90000   \n",
              "3       ID_KKCO4VATOSWB        8.30000        8.35000     29.90000   \n",
              "4       ID_8YNKABZP11SS        6.87000        9.57000     29.90000   \n",
              "...                 ...            ...            ...          ...   \n",
              "127356  ID_0Q6FJWMIY1UD       16.58000       13.15000          NaN   \n",
              "127357  ID_J5QHC8KEKLOS       14.33000       12.77000          NaN   \n",
              "127358  ID_27Q3I4ZW27OC       15.60000       15.97000          NaN   \n",
              "127359  ID_17GIG572BMR3       12.87000       13.08000          NaN   \n",
              "127360  ID_XZOXV4ACPR32       10.57000       11.43000          NaN   \n",
              "\n",
              "        Relative_Humidity  Sensor difference  Sensor div  Sensor R/T      SD  \\\n",
              "0                53.00000            0.11000     0.98815     1.77258 0.07778   \n",
              "1                53.00000            1.32000     0.88267     1.77258 0.93338   \n",
              "2                53.00000            3.38000     2.07302     1.77258 2.39002   \n",
              "3                53.00000            0.05000     0.99401     1.77258 0.03536   \n",
              "4                53.00000            2.70000     0.71787     1.77258 1.90919   \n",
              "...                   ...                ...         ...         ...     ...   \n",
              "127356                NaN            3.43000     1.26084         NaN 2.42538   \n",
              "127357                NaN            1.56000     1.12216         NaN 1.10309   \n",
              "127358                NaN            0.37000     0.97683         NaN 0.26163   \n",
              "127359                NaN            0.21000     0.98394         NaN 0.14849   \n",
              "127360                NaN            0.86000     0.92476         NaN 0.60811   \n",
              "\n",
              "           Mean  ...  Sensor2_PM2_rolling_mean_30  Sensor2_PM2_rolling_max_30  \\\n",
              "0       9.22500  ...                          NaN                         NaN   \n",
              "1      10.59000  ...                          NaN                         NaN   \n",
              "2       4.84000  ...                          NaN                         NaN   \n",
              "3       8.32500  ...                          NaN                         NaN   \n",
              "4       8.22000  ...                          NaN                         NaN   \n",
              "...         ...  ...                          ...                         ...   \n",
              "127356 14.86500  ...                     17.26633                    24.78000   \n",
              "127357 13.55000  ...                     17.00433                    24.78000   \n",
              "127358 15.78500  ...                     16.71067                    23.03000   \n",
              "127359 12.97500  ...                     16.50167                    23.03000   \n",
              "127360 11.00000  ...                     16.19100                    23.03000   \n",
              "\n",
              "        Sensor2_PM2_rolling_min_30  Sensor2_PM2_rolling_mean_10  \\\n",
              "0                              NaN                          NaN   \n",
              "1                              NaN                          NaN   \n",
              "2                              NaN                          NaN   \n",
              "3                              NaN                          NaN   \n",
              "4                              NaN                          NaN   \n",
              "...                            ...                          ...   \n",
              "127356                    12.58000                     14.85900   \n",
              "127357                    12.58000                     14.20300   \n",
              "127358                    12.58000                     14.52200   \n",
              "127359                    12.58000                     14.07500   \n",
              "127360                    11.43000                     13.96000   \n",
              "\n",
              "        Sensor2_PM2_rolling_max_10  Sensor2_PM2_rolling_min_10  Datetime_day  \\\n",
              "0                              NaN                         NaN            15   \n",
              "1                              NaN                         NaN            15   \n",
              "2                              NaN                         NaN            15   \n",
              "3                              NaN                         NaN            15   \n",
              "4                              NaN                         NaN            15   \n",
              "...                            ...                         ...           ...   \n",
              "127356                    19.33000                    12.58000            14   \n",
              "127357                    17.55000                    12.58000            14   \n",
              "127358                    17.55000                    12.58000            14   \n",
              "127359                    16.75000                    12.58000            14   \n",
              "127360                    16.75000                    11.43000            14   \n",
              "\n",
              "        Datetime_month  Datetime_hour  time_of_day  \n",
              "0                   10             16            2  \n",
              "1                   10             16            2  \n",
              "2                   10             16            2  \n",
              "3                   10             16            2  \n",
              "4                   10             16            2  \n",
              "...                ...            ...          ...  \n",
              "127356               3             14            2  \n",
              "127357               3             14            2  \n",
              "127358               3             14            2  \n",
              "127359               3             14            2  \n",
              "127360               3             14            2  \n",
              "\n",
              "[416689 rows x 36 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-cd8595c9-1da9-4a3b-b47f-be7ea2ef7c75\">\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>ID</th>\n",
              "      <th>Sensor1_PM2.5</th>\n",
              "      <th>Sensor2_PM2.5</th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Relative_Humidity</th>\n",
              "      <th>Sensor difference</th>\n",
              "      <th>Sensor div</th>\n",
              "      <th>Sensor R/T</th>\n",
              "      <th>SD</th>\n",
              "      <th>Mean</th>\n",
              "      <th>...</th>\n",
              "      <th>Sensor2_PM2_rolling_mean_30</th>\n",
              "      <th>Sensor2_PM2_rolling_max_30</th>\n",
              "      <th>Sensor2_PM2_rolling_min_30</th>\n",
              "      <th>Sensor2_PM2_rolling_mean_10</th>\n",
              "      <th>Sensor2_PM2_rolling_max_10</th>\n",
              "      <th>Sensor2_PM2_rolling_min_10</th>\n",
              "      <th>Datetime_day</th>\n",
              "      <th>Datetime_month</th>\n",
              "      <th>Datetime_hour</th>\n",
              "      <th>time_of_day</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ID_37STUCVVPZKP</td>\n",
              "      <td>9.17000</td>\n",
              "      <td>9.28000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>0.11000</td>\n",
              "      <td>0.98815</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>0.07778</td>\n",
              "      <td>9.22500</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ID_YZZUOMF0TQIF</td>\n",
              "      <td>9.93000</td>\n",
              "      <td>11.25000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>1.32000</td>\n",
              "      <td>0.88267</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>0.93338</td>\n",
              "      <td>10.59000</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ID_UF68QLIPG3LE</td>\n",
              "      <td>6.53000</td>\n",
              "      <td>3.15000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>3.38000</td>\n",
              "      <td>2.07302</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>2.39002</td>\n",
              "      <td>4.84000</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ID_KKCO4VATOSWB</td>\n",
              "      <td>8.30000</td>\n",
              "      <td>8.35000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>0.05000</td>\n",
              "      <td>0.99401</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>0.03536</td>\n",
              "      <td>8.32500</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ID_8YNKABZP11SS</td>\n",
              "      <td>6.87000</td>\n",
              "      <td>9.57000</td>\n",
              "      <td>29.90000</td>\n",
              "      <td>53.00000</td>\n",
              "      <td>2.70000</td>\n",
              "      <td>0.71787</td>\n",
              "      <td>1.77258</td>\n",
              "      <td>1.90919</td>\n",
              "      <td>8.22000</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "      <td>2</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",
              "      <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>127356</th>\n",
              "      <td>ID_0Q6FJWMIY1UD</td>\n",
              "      <td>16.58000</td>\n",
              "      <td>13.15000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3.43000</td>\n",
              "      <td>1.26084</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2.42538</td>\n",
              "      <td>14.86500</td>\n",
              "      <td>...</td>\n",
              "      <td>17.26633</td>\n",
              "      <td>24.78000</td>\n",
              "      <td>12.58000</td>\n",
              "      <td>14.85900</td>\n",
              "      <td>19.33000</td>\n",
              "      <td>12.58000</td>\n",
              "      <td>14</td>\n",
              "      <td>3</td>\n",
              "      <td>14</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>127357</th>\n",
              "      <td>ID_J5QHC8KEKLOS</td>\n",
              "      <td>14.33000</td>\n",
              "      <td>12.77000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.56000</td>\n",
              "      <td>1.12216</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.10309</td>\n",
              "      <td>13.55000</td>\n",
              "      <td>...</td>\n",
              "      <td>17.00433</td>\n",
              "      <td>24.78000</td>\n",
              "      <td>12.58000</td>\n",
              "      <td>14.20300</td>\n",
              "      <td>17.55000</td>\n",
              "      <td>12.58000</td>\n",
              "      <td>14</td>\n",
              "      <td>3</td>\n",
              "      <td>14</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>127358</th>\n",
              "      <td>ID_27Q3I4ZW27OC</td>\n",
              "      <td>15.60000</td>\n",
              "      <td>15.97000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.37000</td>\n",
              "      <td>0.97683</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.26163</td>\n",
              "      <td>15.78500</td>\n",
              "      <td>...</td>\n",
              "      <td>16.71067</td>\n",
              "      <td>23.03000</td>\n",
              "      <td>12.58000</td>\n",
              "      <td>14.52200</td>\n",
              "      <td>17.55000</td>\n",
              "      <td>12.58000</td>\n",
              "      <td>14</td>\n",
              "      <td>3</td>\n",
              "      <td>14</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>127359</th>\n",
              "      <td>ID_17GIG572BMR3</td>\n",
              "      <td>12.87000</td>\n",
              "      <td>13.08000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.21000</td>\n",
              "      <td>0.98394</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.14849</td>\n",
              "      <td>12.97500</td>\n",
              "      <td>...</td>\n",
              "      <td>16.50167</td>\n",
              "      <td>23.03000</td>\n",
              "      <td>12.58000</td>\n",
              "      <td>14.07500</td>\n",
              "      <td>16.75000</td>\n",
              "      <td>12.58000</td>\n",
              "      <td>14</td>\n",
              "      <td>3</td>\n",
              "      <td>14</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>127360</th>\n",
              "      <td>ID_XZOXV4ACPR32</td>\n",
              "      <td>10.57000</td>\n",
              "      <td>11.43000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.86000</td>\n",
              "      <td>0.92476</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.60811</td>\n",
              "      <td>11.00000</td>\n",
              "      <td>...</td>\n",
              "      <td>16.19100</td>\n",
              "      <td>23.03000</td>\n",
              "      <td>11.43000</td>\n",
              "      <td>13.96000</td>\n",
              "      <td>16.75000</td>\n",
              "      <td>11.43000</td>\n",
              "      <td>14</td>\n",
              "      <td>3</td>\n",
              "      <td>14</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>416689 rows × 36 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-cd8595c9-1da9-4a3b-b47f-be7ea2ef7c75')\"\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-cd8595c9-1da9-4a3b-b47f-be7ea2ef7c75 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-cd8595c9-1da9-4a3b-b47f-be7ea2ef7c75');\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": 36
        }
      ],
      "source": [
        "data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 37,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2sUBet4lD_G0",
        "outputId": "1b191d28-e434-4f13-c45a-3583f26d2726"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((289328, 35), (127361, 35))"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ],
      "source": [
        "data = data.drop(\"ID\", axis = 1)\n",
        "# Separate train and test data from the combined dataframe\n",
        "df_train = data[:train.shape[0]]\n",
        "df_test = data[train.shape[0]:]\n",
        "\n",
        "X = df_train\n",
        "y = target\n",
        "\n",
        "# Check the shapes of the split dataset\n",
        "df_train.shape, df_test.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 38,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oYH9KQ0tFJdw",
        "outputId": "f3d6a9c8-7187-40dd-e32a-4be686fdb01a"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 289328 entries, 0 to 289327\n",
            "Data columns (total 35 columns):\n",
            " #   Column                       Non-Null Count   Dtype  \n",
            "---  ------                       --------------   -----  \n",
            " 0   Sensor1_PM2.5                289328 non-null  float64\n",
            " 1   Sensor2_PM2.5                289328 non-null  float64\n",
            " 2   Temperature                  285795 non-null  float64\n",
            " 3   Relative_Humidity            285795 non-null  float64\n",
            " 4   Sensor difference            289328 non-null  float64\n",
            " 5   Sensor div                   289328 non-null  float64\n",
            " 6   Sensor R/T                   285795 non-null  float64\n",
            " 7   SD                           289328 non-null  float64\n",
            " 8   Mean                         289328 non-null  float64\n",
            " 9   Rolling_3                    289326 non-null  float64\n",
            " 10  Rolling_5                    289324 non-null  float64\n",
            " 11  Rolling_2_3                  289326 non-null  float64\n",
            " 12  Rolling_2_5                  289324 non-null  float64\n",
            " 13  Sensor1_PM2_rolling_mean_60  289269 non-null  float64\n",
            " 14  Sensor1_PM2_rolling_max_60   289269 non-null  float64\n",
            " 15  Sensor1_PM2_rolling_min_60   289269 non-null  float64\n",
            " 16  Sensor1_PM2_rolling_mean_30  289299 non-null  float64\n",
            " 17  Sensor1_PM2_rolling_max_30   289299 non-null  float64\n",
            " 18  Sensor1_PM2_rolling_min_30   289299 non-null  float64\n",
            " 19  Sensor1_PM2_rolling_mean_10  289319 non-null  float64\n",
            " 20  Sensor1_PM2_rolling_max_10   289319 non-null  float64\n",
            " 21  Sensor1_PM2_rolling_min_10   289319 non-null  float64\n",
            " 22  Sensor2_PM2_rolling_mean_60  289269 non-null  float64\n",
            " 23  Sensor2_PM2_rolling_max_60   289269 non-null  float64\n",
            " 24  Sensor2_PM2_rolling_min_60   289269 non-null  float64\n",
            " 25  Sensor2_PM2_rolling_mean_30  289299 non-null  float64\n",
            " 26  Sensor2_PM2_rolling_max_30   289299 non-null  float64\n",
            " 27  Sensor2_PM2_rolling_min_30   289299 non-null  float64\n",
            " 28  Sensor2_PM2_rolling_mean_10  289319 non-null  float64\n",
            " 29  Sensor2_PM2_rolling_max_10   289319 non-null  float64\n",
            " 30  Sensor2_PM2_rolling_min_10   289319 non-null  float64\n",
            " 31  Datetime_day                 289328 non-null  int64  \n",
            " 32  Datetime_month               289328 non-null  int64  \n",
            " 33  Datetime_hour                289328 non-null  int64  \n",
            " 34  time_of_day                  289328 non-null  int64  \n",
            "dtypes: float64(31), int64(4)\n",
            "memory usage: 79.5 MB\n"
          ]
        }
      ],
      "source": [
        "X.info()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eitbu60Ry19u"
      },
      "source": [
        "Modelling"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 39,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zNuuCwvwwsZ1",
        "outputId": "b3d942c6-121f-4f12-c314-bfdd211bd018"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting catboost\n",
            "  Downloading catboost-1.0.4-cp37-none-manylinux1_x86_64.whl (76.1 MB)\n",
            "\u001b[K     |████████████████████████████████| 76.1 MB 1.1 MB/s \n",
            "\u001b[?25hRequirement already satisfied: scipy in /usr/local/lib/python3.7/dist-packages (from catboost) (1.4.1)\n",
            "Requirement already satisfied: plotly in /usr/local/lib/python3.7/dist-packages (from catboost) (5.5.0)\n",
            "Requirement already satisfied: graphviz in /usr/local/lib/python3.7/dist-packages (from catboost) (0.10.1)\n",
            "Requirement already satisfied: pandas>=0.24.0 in /usr/local/lib/python3.7/dist-packages (from catboost) (1.3.5)\n",
            "Requirement already satisfied: numpy>=1.16.0 in /usr/local/lib/python3.7/dist-packages (from catboost) (1.21.5)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from catboost) (1.15.0)\n",
            "Requirement already satisfied: matplotlib in /usr/local/lib/python3.7/dist-packages (from catboost) (3.2.2)\n",
            "Requirement already satisfied: pytz>=2017.3 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.24.0->catboost) (2018.9)\n",
            "Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.24.0->catboost) (2.8.2)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.7/dist-packages (from matplotlib->catboost) (0.11.0)\n",
            "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib->catboost) (3.0.7)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib->catboost) (1.3.2)\n",
            "Requirement already satisfied: tenacity>=6.2.0 in /usr/local/lib/python3.7/dist-packages (from plotly->catboost) (8.0.1)\n",
            "Installing collected packages: catboost\n",
            "Successfully installed catboost-1.0.4\n"
          ]
        }
      ],
      "source": [
        "!pip install catboost"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 40,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RNL045ZLYLiT",
        "outputId": "cd2dea46-d4bd-48e8-8a5b-8db81915bd0e"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Warning: Overfitting detector is active, thus evaluation metric is calculated on every iteration. 'metric_period' is ignored for evaluation metric.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0:\tlearn: 0.9337660\ttest: 0.9337660\ttest1: 0.9796772\tbest: 0.9796772 (0)\ttotal: 210ms\tremaining: 2m 48s\n",
            "100:\tlearn: 0.9928301\ttest: 0.9928301\ttest1: 0.9854146\tbest: 0.9945391 (16)\ttotal: 10.5s\tremaining: 1m 12s\n",
            "200:\tlearn: 0.9953379\ttest: 0.9953379\ttest1: 0.9875229\tbest: 0.9945391 (16)\ttotal: 33s\tremaining: 1m 38s\n",
            "Stopped by overfitting detector  (200 iterations wait)\n",
            "\n",
            "bestTest = 0.9945391076\n",
            "bestIteration = 16\n",
            "\n",
            "Shrink model to first 17 iterations.\n",
            "Accuracy:  0.9945391075934055\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Warning: Overfitting detector is active, thus evaluation metric is calculated on every iteration. 'metric_period' is ignored for evaluation metric.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0:\tlearn: 0.9345648\ttest: 0.9345648\ttest1: 0.9680296\tbest: 0.9680296 (0)\ttotal: 356ms\tremaining: 4m 44s\n",
            "100:\tlearn: 0.9925191\ttest: 0.9925191\ttest1: 0.9946774\tbest: 0.9946774 (95)\ttotal: 10.2s\tremaining: 1m 10s\n",
            "200:\tlearn: 0.9952764\ttest: 0.9952764\ttest1: 0.9950921\tbest: 0.9951267 (159)\ttotal: 19.3s\tremaining: 57.5s\n",
            "300:\tlearn: 0.9968548\ttest: 0.9968548\ttest1: 0.9952995\tbest: 0.9954377 (282)\ttotal: 28.6s\tremaining: 47.4s\n",
            "400:\tlearn: 0.9977765\ttest: 0.9977765\ttest1: 0.9956451\tbest: 0.9956797 (391)\ttotal: 45.8s\tremaining: 45.6s\n",
            "500:\tlearn: 0.9983871\ttest: 0.9983871\ttest1: 0.9956105\tbest: 0.9956797 (391)\ttotal: 55s\tremaining: 32.8s\n",
            "Stopped by overfitting detector  (200 iterations wait)\n",
            "\n",
            "bestTest = 0.9956796737\n",
            "bestIteration = 391\n",
            "\n",
            "Shrink model to first 392 iterations.\n",
            "Accuracy:  0.99567967372896\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Warning: Overfitting detector is active, thus evaluation metric is calculated on every iteration. 'metric_period' is ignored for evaluation metric.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0:\tlearn: 0.9348605\ttest: 0.9348605\ttest1: 0.9607714\tbest: 0.9607714 (0)\ttotal: 138ms\tremaining: 1m 50s\n",
            "100:\tlearn: 0.9927072\ttest: 0.9927072\ttest1: 0.9941244\tbest: 0.9941589 (96)\ttotal: 9.59s\tremaining: 1m 6s\n",
            "200:\tlearn: 0.9953724\ttest: 0.9953724\ttest1: 0.9946774\tbest: 0.9947810 (192)\ttotal: 18.8s\tremaining: 55.9s\n",
            "300:\tlearn: 0.9968817\ttest: 0.9968817\ttest1: 0.9952304\tbest: 0.9953340 (244)\ttotal: 27.9s\tremaining: 46.3s\n",
            "400:\tlearn: 0.9979070\ttest: 0.9979070\ttest1: 0.9953340\tbest: 0.9954032 (360)\ttotal: 39.1s\tremaining: 38.9s\n",
            "500:\tlearn: 0.9984562\ttest: 0.9984562\ttest1: 0.9953686\tbest: 0.9954377 (433)\ttotal: 48.3s\tremaining: 28.8s\n",
            "600:\tlearn: 0.9989670\ttest: 0.9989670\ttest1: 0.9954377\tbest: 0.9954723 (530)\ttotal: 57.5s\tremaining: 19s\n",
            "700:\tlearn: 0.9993126\ttest: 0.9993126\ttest1: 0.9955069\tbest: 0.9955069 (697)\ttotal: 1m 6s\tremaining: 9.4s\n",
            "799:\tlearn: 0.9994624\ttest: 0.9994624\ttest1: 0.9955414\tbest: 0.9955760 (760)\ttotal: 1m 16s\tremaining: 0us\n",
            "\n",
            "bestTest = 0.9955759859\n",
            "bestIteration = 760\n",
            "\n",
            "Shrink model to first 761 iterations.\n",
            "Accuracy:  0.9955759858984551\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Warning: Overfitting detector is active, thus evaluation metric is calculated on every iteration. 'metric_period' is ignored for evaluation metric.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0:\tlearn: 0.9343305\ttest: 0.9343305\ttest1: 0.9677531\tbest: 0.9677531 (0)\ttotal: 141ms\tremaining: 1m 52s\n",
            "100:\tlearn: 0.9928608\ttest: 0.9928608\ttest1: 0.9951612\tbest: 0.9951958 (96)\ttotal: 9.64s\tremaining: 1m 6s\n",
            "200:\tlearn: 0.9954915\ttest: 0.9954915\ttest1: 0.9956451\tbest: 0.9958525 (175)\ttotal: 18.9s\tremaining: 56.2s\n",
            "300:\tlearn: 0.9968471\ttest: 0.9968471\ttest1: 0.9958870\tbest: 0.9958870 (300)\ttotal: 28.1s\tremaining: 46.6s\n",
            "400:\tlearn: 0.9976881\ttest: 0.9976881\ttest1: 0.9962327\tbest: 0.9963709 (364)\ttotal: 37.1s\tremaining: 36.9s\n",
            "500:\tlearn: 0.9984255\ttest: 0.9984255\ttest1: 0.9959907\tbest: 0.9963709 (364)\ttotal: 46.3s\tremaining: 27.6s\n",
            "Stopped by overfitting detector  (200 iterations wait)\n",
            "\n",
            "bestTest = 0.9963709259\n",
            "bestIteration = 364\n",
            "\n",
            "Shrink model to first 365 iterations.\n",
            "Accuracy:  0.9963709259323265\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Warning: Overfitting detector is active, thus evaluation metric is calculated on every iteration. 'metric_period' is ignored for evaluation metric.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0:\tlearn: 0.9366117\ttest: 0.9366117\ttest1: 0.9508865\tbest: 0.9508865 (0)\ttotal: 148ms\tremaining: 1m 58s\n",
            "100:\tlearn: 0.9930337\ttest: 0.9930337\ttest1: 0.9928801\tbest: 0.9929147 (97)\ttotal: 9.55s\tremaining: 1m 6s\n",
            "200:\tlearn: 0.9957718\ttest: 0.9957718\ttest1: 0.9922925\tbest: 0.9931220 (154)\ttotal: 18.8s\tremaining: 56.1s\n",
            "300:\tlearn: 0.9970737\ttest: 0.9970737\ttest1: 0.9929492\tbest: 0.9931220 (154)\ttotal: 28.1s\tremaining: 46.6s\n",
            "Stopped by overfitting detector  (200 iterations wait)\n",
            "\n",
            "bestTest = 0.9931220406\n",
            "bestIteration = 154\n",
            "\n",
            "Shrink model to first 155 iterations.\n",
            "Accuracy:  0.9931220405765043\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Warning: Overfitting detector is active, thus evaluation metric is calculated on every iteration. 'metric_period' is ignored for evaluation metric.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0:\tlearn: 0.9380710\ttest: 0.9380710\ttest1: 0.9382712\tbest: 0.9382712 (0)\ttotal: 142ms\tremaining: 1m 53s\n",
            "100:\tlearn: 0.9930375\ttest: 0.9930375\ttest1: 0.9917395\tbest: 0.9917395 (95)\ttotal: 9.54s\tremaining: 1m 6s\n",
            "200:\tlearn: 0.9957449\ttest: 0.9957449\ttest1: 0.9930184\tbest: 0.9930184 (200)\ttotal: 18.8s\tremaining: 56s\n",
            "300:\tlearn: 0.9971850\ttest: 0.9971850\ttest1: 0.9933294\tbest: 0.9934331 (264)\ttotal: 28.6s\tremaining: 47.5s\n",
            "400:\tlearn: 0.9980107\ttest: 0.9980107\ttest1: 0.9933294\tbest: 0.9936059 (326)\ttotal: 37.8s\tremaining: 37.6s\n",
            "500:\tlearn: 0.9986021\ttest: 0.9986021\ttest1: 0.9930184\tbest: 0.9936059 (326)\ttotal: 46.9s\tremaining: 28s\n",
            "Stopped by overfitting detector  (200 iterations wait)\n",
            "\n",
            "bestTest = 0.9936059171\n",
            "bestIteration = 326\n",
            "\n",
            "Shrink model to first 327 iterations.\n",
            "Accuracy:  0.9936059171188608\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Warning: Overfitting detector is active, thus evaluation metric is calculated on every iteration. 'metric_period' is ignored for evaluation metric.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0:\tlearn: 0.9380172\ttest: 0.9380172\ttest1: 0.9368541\tbest: 0.9368541 (0)\ttotal: 140ms\tremaining: 1m 51s\n",
            "100:\tlearn: 0.9940705\ttest: 0.9940705\ttest1: 0.9803339\tbest: 0.9804030 (93)\ttotal: 9.58s\tremaining: 1m 6s\n",
            "200:\tlearn: 0.9963018\ttest: 0.9963018\ttest1: 0.9815090\tbest: 0.9816473 (183)\ttotal: 18.8s\tremaining: 56s\n",
            "300:\tlearn: 0.9975000\ttest: 0.9975000\ttest1: 0.9817164\tbest: 0.9818546 (208)\ttotal: 27.9s\tremaining: 46.2s\n",
            "400:\tlearn: 0.9984485\ttest: 0.9984485\ttest1: 0.9820620\tbest: 0.9821311 (386)\ttotal: 37.2s\tremaining: 37s\n",
            "600:\tlearn: 0.9991628\ttest: 0.9991628\ttest1: 0.9820620\tbest: 0.9822348 (439)\ttotal: 55.1s\tremaining: 18.3s\n",
            "Stopped by overfitting detector  (200 iterations wait)\n",
            "\n",
            "bestTest = 0.9822348184\n",
            "bestIteration = 439\n",
            "\n",
            "Shrink model to first 440 iterations.\n",
            "Accuracy:  0.9822348183734836\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Warning: Overfitting detector is active, thus evaluation metric is calculated on every iteration. 'metric_period' is ignored for evaluation metric.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0:\tlearn: 0.9437816\ttest: 0.9437816\ttest1: 0.8642035\tbest: 0.8642035 (0)\ttotal: 142ms\tremaining: 1m 53s\n",
            "100:\tlearn: 0.9941666\ttest: 0.9941666\ttest1: 0.9599419\tbest: 0.9599419 (97)\ttotal: 9.63s\tremaining: 1m 6s\n",
            "200:\tlearn: 0.9964784\ttest: 0.9964784\ttest1: 0.9644351\tbest: 0.9646770 (199)\ttotal: 18.9s\tremaining: 56.3s\n",
            "300:\tlearn: 0.9976228\ttest: 0.9976228\ttest1: 0.9651609\tbest: 0.9652300 (298)\ttotal: 28.1s\tremaining: 46.6s\n",
            "400:\tlearn: 0.9984408\ttest: 0.9984408\ttest1: 0.9659904\tbest: 0.9660941 (393)\ttotal: 37.4s\tremaining: 37.2s\n",
            "500:\tlearn: 0.9989938\ttest: 0.9989938\ttest1: 0.9663360\tbest: 0.9665780 (492)\ttotal: 46.5s\tremaining: 27.7s\n",
            "600:\tlearn: 0.9993395\ttest: 0.9993395\ttest1: 0.9655065\tbest: 0.9665780 (492)\ttotal: 55.4s\tremaining: 18.3s\n",
            "Stopped by overfitting detector  (200 iterations wait)\n",
            "\n",
            "bestTest = 0.966577956\n",
            "bestIteration = 492\n",
            "\n",
            "Shrink model to first 493 iterations.\n",
            "Accuracy:  0.9665779559672346\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Warning: Overfitting detector is active, thus evaluation metric is calculated on every iteration. 'metric_period' is ignored for evaluation metric.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0:\tlearn: 0.9363585\ttest: 0.9363585\ttest1: 0.9040509\tbest: 0.9040509 (0)\ttotal: 142ms\tremaining: 1m 53s\n",
            "100:\tlearn: 0.9936251\ttest: 0.9936251\ttest1: 0.9767731\tbest: 0.9767731 (100)\ttotal: 9.65s\tremaining: 1m 6s\n",
            "200:\tlearn: 0.9962557\ttest: 0.9962557\ttest1: 0.9790889\tbest: 0.9790889 (200)\ttotal: 18.9s\tremaining: 56.3s\n",
            "300:\tlearn: 0.9974731\ttest: 0.9974731\ttest1: 0.9792272\tbest: 0.9798493 (254)\ttotal: 28.1s\tremaining: 46.6s\n",
            "400:\tlearn: 0.9982373\ttest: 0.9982373\ttest1: 0.9798147\tbest: 0.9799530 (394)\ttotal: 37.2s\tremaining: 37s\n",
            "500:\tlearn: 0.9986828\ttest: 0.9986828\ttest1: 0.9798839\tbest: 0.9801258 (452)\ttotal: 46.2s\tremaining: 27.6s\n",
            "600:\tlearn: 0.9989900\ttest: 0.9989900\ttest1: 0.9800221\tbest: 0.9802641 (549)\ttotal: 55.2s\tremaining: 18.3s\n",
            "700:\tlearn: 0.9992281\ttest: 0.9992281\ttest1: 0.9806788\tbest: 0.9807825 (683)\ttotal: 1m 4s\tremaining: 9.1s\n",
            "799:\tlearn: 0.9994931\ttest: 0.9994931\ttest1: 0.9805060\tbest: 0.9808517 (748)\ttotal: 1m 13s\tremaining: 0us\n",
            "\n",
            "bestTest = 0.9808516521\n",
            "bestIteration = 748\n",
            "\n",
            "Shrink model to first 749 iterations.\n",
            "Accuracy:  0.9808516521498687\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Warning: Overfitting detector is active, thus evaluation metric is calculated on every iteration. 'metric_period' is ignored for evaluation metric.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0:\tlearn: 0.9428870\ttest: 0.9428870\ttest1: 0.8874257\tbest: 0.8874257 (0)\ttotal: 139ms\tremaining: 1m 51s\n",
            "100:\tlearn: 0.9932295\ttest: 0.9932295\ttest1: 0.9765657\tbest: 0.9783631 (79)\ttotal: 9.56s\tremaining: 1m 6s\n",
            "200:\tlearn: 0.9958179\ttest: 0.9958179\ttest1: 0.9815084\tbest: 0.9819923 (158)\ttotal: 18.8s\tremaining: 56s\n",
            "300:\tlearn: 0.9970852\ttest: 0.9970852\ttest1: 0.9830637\tbest: 0.9830637 (300)\ttotal: 28s\tremaining: 46.4s\n",
            "400:\tlearn: 0.9979378\ttest: 0.9979378\ttest1: 0.9846191\tbest: 0.9847228 (388)\ttotal: 37.3s\tremaining: 37.1s\n",
            "500:\tlearn: 0.9985944\ttest: 0.9985944\ttest1: 0.9852067\tbest: 0.9852758 (499)\ttotal: 46.7s\tremaining: 27.9s\n",
            "600:\tlearn: 0.9990130\ttest: 0.9990130\ttest1: 0.9853449\tbest: 0.9856560 (547)\ttotal: 56s\tremaining: 18.5s\n",
            "700:\tlearn: 0.9992857\ttest: 0.9992857\ttest1: 0.9856215\tbest: 0.9856560 (547)\ttotal: 1m 5s\tremaining: 9.19s\n",
            "799:\tlearn: 0.9994470\ttest: 0.9994470\ttest1: 0.9858980\tbest: 0.9859325 (768)\ttotal: 1m 17s\tremaining: 0us\n",
            "\n",
            "bestTest = 0.9859325315\n",
            "bestIteration = 768\n",
            "\n",
            "Shrink model to first 769 iterations.\n",
            "Accuracy:  0.9859325314530624\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9884490608792162"
            ]
          },
          "metadata": {},
          "execution_count": 40
        }
      ],
      "source": [
        "from catboost import CatBoostClassifier\n",
        "errcb2=[]\n",
        "y_pred_totcb2=[]\n",
        "from sklearn.model_selection import KFold,StratifiedKFold, TimeSeriesSplit\n",
        "from sklearn.metrics import accuracy_score\n",
        "fold=StratifiedKFold(n_splits=10)#15#5#10\n",
        "i=1\n",
        "for train_index, test_index in fold.split(X,y):\n",
        "    X_train, X_test = X.iloc[train_index], X.iloc[test_index]\n",
        "    y_train, y_test = y.iloc[train_index], y.iloc[test_index]\n",
        "    m2  = CatBoostClassifier(n_estimators=800,eval_metric='Accuracy',od_wait=200, learning_rate=0.2,\n",
        "                              bootstrap_type='Bernoulli',metric_period=100,max_depth = 8,\n",
        "                     use_best_model=True)\n",
        "    m2.fit(X_train,y_train,eval_set=[(X_train,y_train),(X_test, y_test)], early_stopping_rounds=200,verbose=100,)\n",
        "    preds=m2.predict(X_test)\n",
        "    print(\"Accuracy: \",(accuracy_score(y_test,preds)))\n",
        "    errcb2.append((accuracy_score(y_test,preds)))\n",
        "    p2 = m2.predict(df_test)\n",
        "    y_pred_totcb2.append(p2)\n",
        "np.mean(errcb2)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3puJYnNey19w"
      },
      "source": [
        "See Feature importance"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 41,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 473
        },
        "id": "fJ13Sz7mnDhZ",
        "outputId": "6933242f-a6a5-4412-f404-a87d12ba1833"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1440x864 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "#0.984\n",
        "feature_importance_df = pd.DataFrame(m2.feature_importances_, columns=['importance'])\n",
        "feature_importance_df['feature'] = X_train.columns\n",
        "\n",
        "plt.figure(figsize=(20, 12));\n",
        "sns.barplot(x=\"importance\", y=\"feature\", data=feature_importance_df.sort_values(by = ['importance'], ascending = False).head(50))\n",
        "plt.title('CatboostClassifier features importance (top 50):');"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "T1jWFtH4y19y"
      },
      "source": [
        "Make submissions"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 42,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "NYLzem3BE9_B",
        "outputId": "2bd4ede2-afee-4bf6-c19c-b4d967edeac3"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                ID  Offset_fault\n",
              "0  ID_STL3V9B2OU12             0\n",
              "1  ID_U2EKAY4SGA94             0\n",
              "2  ID_UJFLEWQR0U4N             0\n",
              "3  ID_S2C2GPGA4R5I             0\n",
              "4  ID_5V2C2ZL221H8             0"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-35e351fc-1932-4f10-88de-03ce6e064930\">\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>ID</th>\n",
              "      <th>Offset_fault</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ID_STL3V9B2OU12</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ID_U2EKAY4SGA94</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ID_UJFLEWQR0U4N</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ID_S2C2GPGA4R5I</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ID_5V2C2ZL221H8</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-35e351fc-1932-4f10-88de-03ce6e064930')\"\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-35e351fc-1932-4f10-88de-03ce6e064930 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-35e351fc-1932-4f10-88de-03ce6e064930');\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": 42
        }
      ],
      "source": [
        "samplesubmission.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 43,
      "metadata": {
        "id": "iKeAgehadUom"
      },
      "outputs": [],
      "source": [
        "predictions = pd.DataFrame(y_pred_totcb2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 44,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4JbTmxVtmfgt",
        "outputId": "0a651c0b-8fcd-4761-e0ad-fdebe032c9c7"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "10"
            ]
          },
          "metadata": {},
          "execution_count": 44
        }
      ],
      "source": [
        "len(y_pred_totcb2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 45,
      "metadata": {
        "id": "3nGK1PFkdXE1"
      },
      "outputs": [],
      "source": [
        "#predictions = predictions.mode()\n",
        "predictions = np.where(np.mean(predictions,axis=0)> 0.45, 1, 0)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 46,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RjpuyxlTdc_K",
        "outputId": "e362f624-e85a-4a0a-e6f9-2ec6a77a7f08"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 0, 1, ..., 0, 1, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 46
        }
      ],
      "source": [
        "predictions = predictions.T\n",
        "predictions"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 47,
      "metadata": {
        "id": "NM_yqsj0dh6_"
      },
      "outputs": [],
      "source": [
        "sub = predictions"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 48,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7tga8TSBdk6x",
        "outputId": "5ad28b61-c8d2-405f-f8d1-e5c633494bfe"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "41771"
            ]
          },
          "metadata": {},
          "execution_count": 48
        }
      ],
      "source": [
        "sum(sub)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 49,
      "metadata": {
        "id": "a0wSU51MdmvJ"
      },
      "outputs": [],
      "source": [
        "samplesubmission.Offset_fault  = sub\n",
        "samplesubmission.ID = test_orig.ID"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 50,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "t3mMwUsgdot2",
        "outputId": "44899613-ba44-47b3-e0f3-2cbe49f87051"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    85590\n",
              "1    41771\n",
              "Name: Offset_fault, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 50
        }
      ],
      "source": [
        "samplesubmission.Offset_fault.value_counts()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 51,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "FIypzotmuV5F",
        "outputId": "643214a5-cc49-44a3-f986-33ea9cb50b97"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                ID  Offset_fault\n",
              "0  ID_H9H3ENR9M13N             0\n",
              "1  ID_JLD6XFEW7EAM             0\n",
              "2  ID_OQ01HEDYRZEU             1\n",
              "3  ID_NHK3H46SLYA2             1\n",
              "4  ID_NGRS57K2NFUF             0"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-11afdfbd-94a4-430f-86ee-a39db9a763af\">\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>ID</th>\n",
              "      <th>Offset_fault</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ID_H9H3ENR9M13N</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ID_JLD6XFEW7EAM</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ID_OQ01HEDYRZEU</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ID_NHK3H46SLYA2</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ID_NGRS57K2NFUF</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-11afdfbd-94a4-430f-86ee-a39db9a763af')\"\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-11afdfbd-94a4-430f-86ee-a39db9a763af 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-11afdfbd-94a4-430f-86ee-a39db9a763af');\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": 51
        }
      ],
      "source": [
        "samplesubmission.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 52,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "id": "C0K2fhEvdtkZ",
        "outputId": "006c48ab-6c3e-40a2-f124-35348b02b536"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_4a4f6fd1-6216-4f73-aa83-1e2a8af57314\", \"Reproducibility.csv\", 2292514)"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "# Write to csv file to submit to Zindi\n",
        "name = \"Reproducibility.csv\"\n",
        "samplesubmission.to_csv(name, index = False)\n",
        "from google.colab import files\n",
        "files.download(name)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "oCLXYvWOd0eg"
      },
      "outputs": [],
      "source": [
        ""
      ]
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [],
      "name": "2nd Place UmojaHack Africa 2022 Faulty Air Quality Sensor Challenge.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3 (ipykernel)",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.4"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}