Predict_Milk_Quality.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/andrewmuhoro/MilkGrade/blob/main/Predict_Milk_Quality.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Introduction**"
      ],
      "metadata": {
        "id": "2t9viYOjUr5L"
      },
      "id": "2t9viYOjUr5L"
    },
    {
      "cell_type": "markdown",
      "source": [
        "In this notebook, the goal is to predict the milk quality. The central goal is to design a model that makes proper classifications for new milk samples. In other words, one which exhibits good generalization. by seven observable milk variables\n",
        "\n",
        "We can classify milk into three groups: low, medium, and high quality. The milk dataset contains 429 instances of low quality, 374 instances of medium quality, and 256 instances of high quality.\n",
        "\n",
        "This dataset is manually collected from observations. It helps us to build machine learning models to predict the quality of milk. \n",
        "This dataset consists of 7 independent variables ie pH, Temperature, Taste, Odor, Fat, Turbidity, and Color:\n",
        "\n",
        "1. pH: This feature defines pH of the milk, which is in the range of 3 to 9.5. \n",
        "\n",
        "2. Temperature: This feature defines the temperature of the milk, and its range is from 34'C to 90'C.\n",
        "\n",
        "3. Taste: This feature defines the taste of the milk and takes the possibles values: 1 (good) or 0 (bad).\n",
        "\n",
        "4. Odor: This feature defines the odor of the milk and takes the possibles values: 1 (good) or 0 (bad).\n",
        "\n",
        "5. Fat: This feature defines fat of the milk and takes the possibles values: 1 (good) or 0 (bad).\n",
        "\n",
        "6. Turbidity: This feature defines the turbidity of the milk and takes the possibles values: 1 (good) or 0 (bad).\n",
        "\n",
        "7. Color: This feature defines the color of the milk, which is in the range of 240 to 255.\n",
        "\n",
        "8. Grade: This is the target and takes the values: low_quality,medium_quality or high_quality."
      ],
      "metadata": {
        "id": "oc4rzXhJ5EMi"
      },
      "id": "oc4rzXhJ5EMi"
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Data Gathering**"
      ],
      "metadata": {
        "id": "SbU6Nu55U2Bv"
      },
      "id": "SbU6Nu55U2Bv"
    },
    {
      "cell_type": "code",
      "execution_count": 41,
      "id": "57f96b97",
      "metadata": {
        "id": "57f96b97"
      },
      "outputs": [],
      "source": [
        "# Install kaggle\n",
        "# Kaggle will enable us to access and use data directly stored on the kaggle platform\n",
        "!pip install -q kaggle\n",
        "from google.colab import files\n",
        "from IPython.display import clear_output\n",
        "\n",
        "files.upload()\n",
        "clear_output()\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Upload your kaggle.json to colab environment\n",
        "! mkdir ~/.kaggle"
      ],
      "metadata": {
        "id": "RHSG-1aL-3X-"
      },
      "id": "RHSG-1aL-3X-",
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# remove any existing ~/.kaggle directory and \n",
        "# create a new one\n",
        "# move your kaggle.json to ~/.kaggle\n",
        "!rm -r ~/.kaggle\n",
        "!mkdir ~/.kaggle\n",
        "!mv ./kaggle.json ~/.kaggle/\n",
        "!chmod 600 ~/.kaggle/kaggle.json"
      ],
      "metadata": {
        "id": "pbsvq0Ge-_dD"
      },
      "id": "pbsvq0Ge-_dD",
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# List kaggle datasets\n",
        "! kaggle datasets list "
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WhrAf1_E_GTO",
        "outputId": "c1f6c141-49db-4814-9641-ceba5bd32142"
      },
      "id": "WhrAf1_E_GTO",
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "ref                                                           title                                             size  lastUpdated          downloadCount  voteCount  usabilityRating  \n",
            "------------------------------------------------------------  -----------------------------------------------  -----  -------------------  -------------  ---------  ---------------  \n",
            "salvatorerastelli/spotify-and-youtube                         Spotify and Youtube                                9MB  2023-03-20 15:43:25           4448        174  1.0              \n",
            "ashishraut64/global-methane-emissions                         Global Emissions.                                 31KB  2023-03-27 09:02:51           1407         40  1.0              \n",
            "muhammedzidan/car-prices-market                               Car Prices Market                                787KB  2023-04-01 10:07:17            768         27  1.0              \n",
            "thitiwat/top-100-most-expensive-football-transfers            ⚽ Top 100 most expensive football transfers        2KB  2023-03-27 04:06:28            742         29  1.0              \n",
            "davidms8/product-sales-detail                                 Product Sales - Detail 2023                        2MB  2023-03-15 14:50:30           1668         45  0.8235294        \n",
            "dgoenrique/netflix-movies-and-tv-shows                        Netflix Movies and TV Shows                        2MB  2023-03-13 18:49:00           2740         73  1.0              \n",
            "lamiatabassum/top-50-us-tech-companies-2022-2023-dataset      USA Tech Companies Stats                           2KB  2023-03-25 06:06:20           1059         37  1.0              \n",
            "arnabchaki/popular-video-games-1980-2023                      Popular Video Games 1980 - 2023 🎮                  1MB  2023-03-23 16:16:51           2320         74  1.0              \n",
            "rkiattisak/student-performance-in-mathematics                 Student performance prediction                     9KB  2023-03-12 04:32:56           5736        118  1.0              \n",
            "usamabuttar/world-happiness-report-2005-present               World Happiness Report, 2005-Present             123KB  2023-03-25 17:38:26           1381         38  1.0              \n",
            "datascientistanna/customers-dataset                           Shop Customer Data                                23KB  2023-02-07 18:42:21          18315        371  1.0              \n",
            "kylegraupe/ukrainerussia-conflict-dataset                     Ukraine/Russia Conflict Dataset                    4MB  2023-03-27 16:51:19            665         35  0.9411765        \n",
            "mehmettahiraslan/customer-shopping-dataset                    Customer Shopping Dataset - Retail Sales Data      2MB  2023-03-09 07:44:35           3653         76  1.0              \n",
            "ulrikthygepedersen/meat-consumption                           Meat Consumption                                 137KB  2023-03-02 09:59:13           1059         27  1.0              \n",
            "rajkumarpandey02/list-of-countries-by-gdp-sector-composition  List of Countries by GDP Sector Composition        8KB  2023-03-20 04:42:10           1727         60  1.0              \n",
            "ajaypalsinghlo/world-happiness-report-2023                    World Happiness Report 2023                        7KB  2023-03-22 07:29:42           1509         44  0.9411765        \n",
            "amaanansari09/most-streamed-songs-all-time                    Most Streamed Songs (All Time)                     8KB  2023-03-08 10:56:24           3865         85  1.0              \n",
            "mazlumi/mbti-personality-type-twitter-dataset                 MBTI Personality Type Twitter Dataset             33MB  2023-03-22 12:11:45            528         39  1.0              \n",
            "shreyajagani13/used-car-dataset                               Used Car Dataset                                  12KB  2023-03-19 06:38:54           1621         35  1.0              \n",
            "hrokrin/the-largest-diamond-dataset-currely-on-kaggle         The largest diamond dataset currently on Kaggle    4MB  2023-03-16 20:29:04            566         25  1.0              \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Download the dataset\n",
        "!kaggle datasets download -d harinuu/milk-quality-prediction"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PZWDdLyY7ZEA",
        "outputId": "f6e66edd-bfb1-4eb3-ef7f-bcf10beb6ffd"
      },
      "id": "PZWDdLyY7ZEA",
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading milk-quality-prediction.zip to /content\n",
            "\r  0% 0.00/1.24k [00:00<?, ?B/s]\n",
            "\r100% 1.24k/1.24k [00:00<00:00, 1.15MB/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Unzip the downloaded dataset\n",
        "!unzip milk-quality-prediction.zip"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Bw38Pteq_laB",
        "outputId": "0a0d2160-c463-4922-e655-c77c31c1ec1b"
      },
      "id": "Bw38Pteq_laB",
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  milk-quality-prediction.zip\n",
            "  inflating: milknew.csv             \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# List all files\n",
        "!ls"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2-X3PSuUQF1W",
        "outputId": "e822dc48-f2d8-43e5-dbeb-649df6407001"
      },
      "id": "2-X3PSuUQF1W",
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "milknew.csv  milk-quality-prediction.zip  sample_data\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Data Exploration**"
      ],
      "metadata": {
        "id": "0cYbUdj3AA4l"
      },
      "id": "0cYbUdj3AA4l"
    },
    {
      "cell_type": "code",
      "source": [
        "# Import necessary Libraries\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import seaborn as sb\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.style as style\n",
        "%matplotlib inline\n",
        "\n",
        "from sklearn.preprocessing import LabelEncoder, StandardScaler\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.metrics import classification_report"
      ],
      "metadata": {
        "id": "Pwi6xYJX-XXO"
      },
      "id": "Pwi6xYJX-XXO",
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Load data into a dataframe\n",
        "data = pd.read_csv('milknew.csv')"
      ],
      "metadata": {
        "id": "ULGD7hSa9DA4"
      },
      "id": "ULGD7hSa9DA4",
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "id": "b3fa0733",
      "metadata": {
        "id": "b3fa0733",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "1ff39c53-07b7-4d83-ce1d-587b2df17589"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "    pH  Temprature  Taste  Odor  Fat   Turbidity  Colour   Grade\n",
              "0  6.6          35      1     0     1          0     254    high\n",
              "1  6.6          36      0     1     0          1     253    high\n",
              "2  8.5          70      1     1     1          1     246     low\n",
              "3  9.5          34      1     1     0          1     255     low\n",
              "4  6.6          37      0     0     0          0     255  medium"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-e5423155-ab13-4b09-bfac-13ae09c6cae6\">\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>pH</th>\n",
              "      <th>Temprature</th>\n",
              "      <th>Taste</th>\n",
              "      <th>Odor</th>\n",
              "      <th>Fat</th>\n",
              "      <th>Turbidity</th>\n",
              "      <th>Colour</th>\n",
              "      <th>Grade</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>6.6</td>\n",
              "      <td>35</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>254</td>\n",
              "      <td>high</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>6.6</td>\n",
              "      <td>36</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>253</td>\n",
              "      <td>high</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>8.5</td>\n",
              "      <td>70</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>246</td>\n",
              "      <td>low</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>9.5</td>\n",
              "      <td>34</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>255</td>\n",
              "      <td>low</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>6.6</td>\n",
              "      <td>37</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>255</td>\n",
              "      <td>medium</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e5423155-ab13-4b09-bfac-13ae09c6cae6')\"\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-e5423155-ab13-4b09-bfac-13ae09c6cae6 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-e5423155-ab13-4b09-bfac-13ae09c6cae6');\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": 11
        }
      ],
      "source": [
        "# Check the top 5 rows\n",
        "data.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "id": "bd6f3305",
      "metadata": {
        "id": "bd6f3305",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "b458bfdd-4aaf-401a-ab6d-4c0039de787e"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       pH  Temprature  Taste  Odor  Fat   Turbidity  Colour   Grade\n",
              "1054  6.7          45      1     1     0          0     247  medium\n",
              "1055  6.7          38      1     0     1          0     255    high\n",
              "1056  3.0          40      1     1     1          1     255     low\n",
              "1057  6.8          43      1     0     1          0     250    high\n",
              "1058  8.6          55      0     1     1          1     255     low"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-30b31afa-271b-462f-8092-4a3e9b6bdbbc\">\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>pH</th>\n",
              "      <th>Temprature</th>\n",
              "      <th>Taste</th>\n",
              "      <th>Odor</th>\n",
              "      <th>Fat</th>\n",
              "      <th>Turbidity</th>\n",
              "      <th>Colour</th>\n",
              "      <th>Grade</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1054</th>\n",
              "      <td>6.7</td>\n",
              "      <td>45</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>247</td>\n",
              "      <td>medium</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1055</th>\n",
              "      <td>6.7</td>\n",
              "      <td>38</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>255</td>\n",
              "      <td>high</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1056</th>\n",
              "      <td>3.0</td>\n",
              "      <td>40</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>255</td>\n",
              "      <td>low</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1057</th>\n",
              "      <td>6.8</td>\n",
              "      <td>43</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>250</td>\n",
              "      <td>high</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1058</th>\n",
              "      <td>8.6</td>\n",
              "      <td>55</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>255</td>\n",
              "      <td>low</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-30b31afa-271b-462f-8092-4a3e9b6bdbbc')\"\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-30b31afa-271b-462f-8092-4a3e9b6bdbbc 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-30b31afa-271b-462f-8092-4a3e9b6bdbbc');\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": [
        "# Check the last 5 rows\n",
        "data.tail()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "id": "6d45063b",
      "metadata": {
        "id": "6d45063b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "190ee4ae-d198-4b5a-d114-b0afd652b256"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1059, 8)"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ],
      "source": [
        "# Check rows, columns count\n",
        "data.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "id": "d0bed35a",
      "metadata": {
        "id": "d0bed35a",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4c5fdead-fdf9-43ce-d491-1bd739c43f02"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['pH', 'Temprature', 'Taste', 'Odor', 'Fat ', 'Turbidity', 'Colour',\n",
              "       'Grade'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ],
      "source": [
        "# Check column headers\n",
        "data.keys()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "id": "5e0a2513",
      "metadata": {
        "id": "5e0a2513",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f104b46b-f920-431c-b142-7c031a556111"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 1059 entries, 0 to 1058\n",
            "Data columns (total 8 columns):\n",
            " #   Column      Non-Null Count  Dtype  \n",
            "---  ------      --------------  -----  \n",
            " 0   pH          1059 non-null   float64\n",
            " 1   Temprature  1059 non-null   int64  \n",
            " 2   Taste       1059 non-null   int64  \n",
            " 3   Odor        1059 non-null   int64  \n",
            " 4   Fat         1059 non-null   int64  \n",
            " 5   Turbidity   1059 non-null   int64  \n",
            " 6   Colour      1059 non-null   int64  \n",
            " 7   Grade       1059 non-null   object \n",
            "dtypes: float64(1), int64(6), object(1)\n",
            "memory usage: 66.3+ KB\n"
          ]
        }
      ],
      "source": [
        "data.info()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "id": "9a8f9572",
      "metadata": {
        "id": "9a8f9572",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "outputId": "549684b0-3491-45c6-b4f9-432b99927352"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                pH   Temprature        Taste\n",
              "count  1059.000000  1059.000000  1059.000000\n",
              "mean      6.630123    44.226629     0.546742\n",
              "std       1.399679    10.098364     0.498046\n",
              "min       3.000000    34.000000     0.000000\n",
              "25%       6.500000    38.000000     0.000000\n",
              "50%       6.700000    41.000000     1.000000\n",
              "75%       6.800000    45.000000     1.000000\n",
              "max       9.500000    90.000000     1.000000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-be7d20aa-c14b-4078-bce5-1b58f4885164\">\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>pH</th>\n",
              "      <th>Temprature</th>\n",
              "      <th>Taste</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>1059.000000</td>\n",
              "      <td>1059.000000</td>\n",
              "      <td>1059.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>6.630123</td>\n",
              "      <td>44.226629</td>\n",
              "      <td>0.546742</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>1.399679</td>\n",
              "      <td>10.098364</td>\n",
              "      <td>0.498046</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>3.000000</td>\n",
              "      <td>34.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>6.500000</td>\n",
              "      <td>38.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>6.700000</td>\n",
              "      <td>41.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>6.800000</td>\n",
              "      <td>45.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>9.500000</td>\n",
              "      <td>90.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-be7d20aa-c14b-4078-bce5-1b58f4885164')\"\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-be7d20aa-c14b-4078-bce5-1b58f4885164 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-be7d20aa-c14b-4078-bce5-1b58f4885164');\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": 16
        }
      ],
      "source": [
        "# Get a summary of listed cols\n",
        "data[['pH','Temprature','Taste']].describe()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "id": "afa65fac",
      "metadata": {
        "id": "afa65fac",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3f766102-bb58-4bc8-ed68-258a50f850e4"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "pH            0\n",
              "Temprature    0\n",
              "Taste         0\n",
              "Odor          0\n",
              "Fat           0\n",
              "Turbidity     0\n",
              "Colour        0\n",
              "Grade         0\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ],
      "source": [
        "# Check for null values\n",
        "data.isnull().sum()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "id": "eb000f81",
      "metadata": {
        "id": "eb000f81",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "61a5a9e8-2f60-4e47-98ea-f78efb6db77d"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "976"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ],
      "source": [
        "# Check for duplicates\n",
        "data.duplicated().sum()"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "> The dataset has no key, this maybe the reason why we are picking up so many duplicate records."
      ],
      "metadata": {
        "id": "drRJhsU5B1DN"
      },
      "id": "drRJhsU5B1DN"
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Data Cleaning**"
      ],
      "metadata": {
        "id": "6iWWfYGhUivX"
      },
      "id": "6iWWfYGhUivX"
    },
    {
      "cell_type": "code",
      "source": [
        "# Create a copy of the original dataset\n",
        "data_cp = data.copy()"
      ],
      "metadata": {
        "id": "GDnJqSlnE7S2"
      },
      "id": "GDnJqSlnE7S2",
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Rename columns\n",
        "data_cp = data_cp.rename(columns={'Temprature':'Temperature'})\n",
        "\n",
        "# Remove any empty spaces in header col\n",
        "data_cp.columns = data_cp.columns.str.replace(' ', '')"
      ],
      "metadata": {
        "id": "de_0KLiIVDmN"
      },
      "id": "de_0KLiIVDmN",
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Data Visualization**"
      ],
      "metadata": {
        "id": "_nwil54rCSgM"
      },
      "id": "_nwil54rCSgM"
    },
    {
      "cell_type": "code",
      "source": [
        "# plot a scatter plot to visualize any correlation between fat and quaity grade \n",
        "plt.scatter(data_cp['Fat'], data_cp['Grade'])\n",
        "plt.xlabel('Fat Content')\n",
        "plt.ylabel('Quality Grade ')\n",
        "plt.title('Correlation between Fat and Quality Grade')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "vCWmsD74TmZ8",
        "outputId": "1b82d5d5-b18e-44df-c9c8-3a52f518cff9"
      },
      "id": "vCWmsD74TmZ8",
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAloAAAHHCAYAAABnS/bqAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABDuElEQVR4nO3deXyM5/7/8fckssuGIHaC2sVybLG11aOooqeldU7squrUUUX5aoVqayvlaKuqbWgp1U0XWrvWVkuD1r6mtastQUhIrt8ffpljJGGG3IlJX8/HYx4Pc93X3POZa2Yyb/d9zTU2Y4wRAAAAsp1HbhcAAACQVxG0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELSAbDZz5kzZbDbFx8dn2z7j4+Nls9k0c+bMbNuns5o3b65q1arl+P3i3pWbr8c70a1bN5UpU8ahzWazaeTIkblST276qz7u3ETQgls4cOCA+vTpo3LlysnX11dBQUGKiorSlClTdPny5dwuL9t88sknmjx5cm6XYbnXX39dCxYsyO0yLDNy5EjZbLZML++++67T+1m3bp1Gjhyp8+fPW1dsDvjjjz/0zDPPqEyZMvLx8VHhwoXVoUMHrVu3LrdLs7N6rL/99lu1bdtWRYoUkbe3twoUKKCmTZtq4sSJSkxMtOQ+cW/Il9sFALezcOFCPfHEE/Lx8VGXLl1UrVo1paSkaM2aNRo8eLB27Nih9957L7fLzBaffPKJtm/frgEDBji0ly5dWpcvX5aXl1fuFJbNXn/9dT3++ONq3759bpdiqWnTpil//vwObfXr13f69uvWrdOoUaPUrVs3hYSEZHN1OWPt2rVq3bq1JKlXr16qUqWKTpw4oZkzZ6px48Z6++231bdv3xyv6/Lly8qX738fgVaNdVpamnr27KmZM2eqevXqevbZZ1WyZElduHBB69ev10svvaRFixZp+fLl2XafuLcQtHBPO3TokJ588kmVLl1aK1asUHh4uH1bv379tH//fi1cuPCu78cYoytXrsjPzy/DtitXrsjb21seHrl3ANhms8nX1zfX7h935vHHH1ehQoVyu4xcc+7cOT3++OPy8/PT2rVrFRERYd82cOBAtWzZUs8995xq1aqlBg0a5GhtOfV+Gj9+vGbOnKnnn39eEydOlM1ms2/7z3/+o+PHj+ujjz665T7S0tKUkpLC3wA3xalD3NPGjx+vixcv6oMPPnAIWenKly+v//znP/br165d0+jRoxURESEfHx+VKVNG//d//6fk5GSH25UpU0aPPPKIFi9erLp168rPz0/Tp0/XqlWrZLPZNG/ePL300ksqXry4/P397Yf2N2zYoIcffljBwcHy9/dXs2bNtHbt2ts+jq+//lpt2rRRsWLF5OPjo4iICI0ePVqpqan2Ps2bN9fChQv1+++/208zpc8ryWpOzIoVK9SkSRMFBAQoJCRE7dq1065duxz6pJ/G2r9/v/1/68HBwerevbuSkpJuW3u6X375RY0aNZKfn5/Kli2b6Smw5ORkxcTEqHz58vLx8VHJkiU1ZMgQh/G32Wy6dOmSZs2aZX+c3bp106+//iqbzaZvvvnG4T5tNptq167tcD+tWrXKcGTo+++/t49FYGCg2rRpox07dmSocffu3Xr88cdVoEAB+fr6qm7dug73Kf1vnt3atWs1cOBAhYWFKSAgQB06dNCff/7p9Jhl5ddff1W3bt3sp8KLFi2qHj166MyZM/Y+I0eO1ODBgyVJZcuWtY/Vreb+rV69Wk888YRKlSplH//nn38+w+n1bt26KX/+/Dp69Kjat2+v/PnzKywsTIMGDXJ4TUrS+fPn1a1bNwUHByskJERdu3Z1+vTa9OnTdeLECU2YMMEhZEmSn5+fZs2aJUl65ZVXHB73jWEkXWZzH515X2XlxrlKtxrrZs2aqWbNmpnu47777lPLli2zvI+kpCSNGzdOVatW1YQJEzJ9XOHh4XrxxRcz1Pbvf/9bc+bMUdWqVeXj46MffvhBkvTGG2+oUaNGKliwoPz8/FSnTh19/vnnGfabnJys559/XmFhYQoMDNSjjz6qI0eOZFrn0aNH1aNHDxUpUkQ+Pj6qWrWqPvzwwywfF1zDES3c07799luVK1dOjRo1cqp/r169NGvWLD3++ON64YUXtGHDBo0ZM0a7du3SV1995dB3z549euqpp9SnTx/17t1b9913n33b6NGj5e3trUGDBik5OVne3t5asWKFWrVqpTp16igmJkYeHh6KjY3VAw88oNWrV6tevXpZ1jVz5kzlz59fAwcOVP78+bVixQqNGDFCiYmJmjBhgiRp+PDhSkhI0JEjR/Tmm29KUobTTjdatmyZWrVqpXLlymnkyJG6fPmypk6dqqioKMXFxWWY/NuxY0eVLVtWY8aMUVxcnN5//30VLlxY48aNu+24njt3Tq1bt1bHjh311FNPaf78+erbt6+8vb3Vo0cPSdf/1/3oo49qzZo1evrpp1W5cmX99ttvevPNN7V37177nKyPP/5YvXr1Ur169fT0009LkiIiIlStWjWFhITop59+0qOPPirpenDw8PDQtm3blJiYqKCgIKWlpWndunX226bvs2vXrmrZsqXGjRunpKQkTZs2TY0bN9aWLVvsY7Fjxw5FRUWpePHiGjp0qAICAjR//ny1b99eX3zxhTp06ODwuJ977jmFhoYqJiZG8fHxmjx5sv7973/r008/ve2YSdLZs2cdrnt6eio0NFRLly7VwYMH1b17dxUtWtR++nvHjh36+eefZbPZ9Nhjj2nv3r2aO3eu3nzzTfuRsbCwsCzv77PPPlNSUpL69u2rggULauPGjZo6daqOHDmizz77zKFvamqqWrZsqfr16+uNN97QsmXLNHHiREVERNhP5Rlj1K5dO61Zs0bPPPOMKleurK+++kpdu3Z16vF/++238vX1VceOHTPdXrZsWTVu3FjLli3TlStXXD5i48z7yhm3Guvo6Gj17t1b27dvd/hSyKZNm7R371699NJLWe53zZo1On/+vAYNGiRPT0+XHtuKFSs0f/58/fvf/1ahQoXsr+EpU6bo0Ucf1T//+U+lpKRo3rx5euKJJ/Tdd9+pTZs29tv36tVLs2fPVufOndWoUSOtWLHCYXu6kydPqkGDBvZwFxYWpu+//149e/ZUYmJihmkMuAMGuEclJCQYSaZdu3ZO9d+6dauRZHr16uXQPmjQICPJrFixwt5WunRpI8n88MMPDn1XrlxpJJly5cqZpKQke3taWpqpUKGCadmypUlLS7O3JyUlmbJly5qHHnrI3hYbG2skmUOHDjn0u1mfPn2Mv7+/uXLlir2tTZs2pnTp0hn6Hjp0yEgysbGx9rbIyEhTuHBhc+bMGXvbtm3bjIeHh+nSpYu9LSYmxkgyPXr0cNhnhw4dTMGCBTPc182aNWtmJJmJEyfa25KTk+33n5KSYowx5uOPPzYeHh5m9erVDrd/9913jSSzdu1ae1tAQIDp2rVrhvtq06aNqVevnv36Y489Zh577DHj6elpvv/+e2OMMXFxcUaS+frrr40xxly4cMGEhISY3r17O+zrxIkTJjg42KH9wQcfNNWrV3cY87S0NNOoUSNToUIFe1v6c9iiRQuH5/v55583np6e5vz587ccs/Qxv/mS/txm9nqYO3eukWR++ukne9uECRMyvJZuJbP9jhkzxthsNvP777/b27p27WokmVdeecWhb61atUydOnXs1xcsWGAkmfHjx9vbrl27Zpo0aZLh9ZiZkJAQU7NmzVv26d+/v5Fkfv31V2PM/8buZnfzvuratWuG95UkExMTY7+e1VifP3/e+Pr6mhdffDFD3QEBAebixYtZPrYpU6YYSWbBggUO7deuXTN//vmnw+XG15kk4+HhYXbs2JFhnzc/5pSUFFOtWjXzwAMP2NvS/xY+++yzDn07d+6c4XH37NnThIeHm9OnTzv0ffLJJ01wcHCmYwzXcOoQ96z003WBgYFO9V+0aJGk63M/bvTCCy9IUoa5XGXLls3ysH/Xrl0d5mtt3bpV+/btU+fOnXXmzBmdPn1ap0+f1qVLl/Tggw/qp59+UlpaWpa13bivCxcu6PTp02rSpImSkpK0e/dupx7fjY4fP66tW7eqW7duKlCggL29Ro0aeuihh+xjcaNnnnnG4XqTJk105swZp77xlC9fPvXp08d+3dvbW3369NGpU6f0yy+/SLp+NKVy5cqqVKmSfXxOnz6tBx54QJK0cuXK295PkyZNFBcXp0uXLkm6fkSgdevWioyM1OrVqyVdP8pls9nUuHFjSdLSpUt1/vx5PfXUUw736+npqfr169vv9+zZs1qxYoU6duxofw5Onz6tM2fOqGXLltq3b5+OHj3qUM/TTz/tcLqnSZMmSk1N1e+//37bxyJJX3zxhZYuXWq/zJkzR5Lj6+HKlSs6ffq0fY5SXFycU/vOzI37vXTpkk6fPq1GjRrJGKMtW7Zk6J/Za+LgwYP264sWLVK+fPkcJqt7enrqueeec6qeCxcu3Pb9m779woULTu3zRtn9vspMcHCw2rVrp7lz58oYI+n60cBPP/1U7du3V0BAQJa3TX9v3Xxk+rffflNYWJjD5cbTxpLUrFkzValSJcM+b3zM586dU0JCgv19ky79/d+/f3+H2958dMoYoy+++EJt27aVMcbh/dOyZUslJCTc1esR13HqEPesoKAgSc7/Af7999/l4eGh8uXLO7QXLVpUISEhGT4cy5Ytm+W+bt62b98+SbrlKZOEhASFhoZmum3Hjh166aWXtGLFigzBJiEhIct9ZiX9sdx4ujNd5cqVtXjxYl26dMnhQ6BUqVIO/dJrPXfunH2ss1KsWLEMHygVK1aUdH3+WIMGDbRv3z7t2rUry1Nbp06dus2juv5Bf+3aNa1fv14lS5bUqVOn1KRJE+3YscMhaFWpUsUeMNOfm/RAd7P0x7Z//34ZY/Tyyy/r5ZdfzrLG4sWL26/fasyc0bRp00wnw589e1ajRo3SvHnzMozLnbwe0v3xxx8aMWKEvvnmmww13rxfX1/fDM9VaGiow+1+//13hYeHZwgKmb3uMhMYGHjb92/69sKFCzu1zxtl9/sqK126dNGnn36q1atXq2nTplq2bJlOnjyp6OjoW94uPURevHjRob18+fJaunSpJOmjjz7Sxx9/nOG2Wf19+u677/Tqq69q69atGeY+pkv/W3jzvLibn7c///xT58+f13vvvZflN7eded/i1ghauGcFBQWpWLFi2r59u0u3y2zCaWYy+4ZhVtvSj1ZNmDBBkZGRmd4mq/lU58+fV7NmzRQUFKRXXnlFERER8vX1VVxcnF588cVbHgnLTlnNEUn/X/rdSktLU/Xq1TVp0qRMt5csWfK2+6hbt658fX31008/qVSpUipcuLAqVqyoJk2a6J133lFycrJWr17tMJcqffw+/vhjFS1aNMM+07/Cn95v0KBBWR7JvDmkWzVmHTt21Lp16zR48GBFRkYqf/78SktL08MPP3zHr4fU1FQ99NBDOnv2rF588UVVqlRJAQEBOnr0qLp165Zhv67OGboTVapUUVxcnJKTk+Xj45Npn19//VXe3t72gJvV+zezSfo59b5q2bKlihQpotmzZ6tp06aaPXu2ihYtqhYtWtzydpUqVZIkbd++Xe3atbO358+f337bNWvWZHrbzP4+rV69Wo8++qiaNm2qd955R+Hh4fLy8lJsbKw++eQTlx9X+hj961//yvI/kTVq1HB5v3BE0MI97ZFHHtF7772n9evXq2HDhrfsW7p0aaWlpWnfvn2qXLmyvf3kyZM6f/68Spcufcd1pP/PMCgo6LZ/XG+2atUqnTlzRl9++aWaNm1qbz906FCGvs6GxPTHsmfPngzbdu/erUKFCt3ylIarjh07luEI2d69eyXJPkk3IiJC27Zt04MPPnjbx5HVdm9vb9WrV0+rV69WqVKl1KRJE0nXj3QlJydrzpw5OnnypMM4pj83hQsXvuVzU65cOUmSl5eXy89hdjp37pyWL1+uUaNGacSIEfb29CNzN3L29SBdPx21d+9ezZo1S126dLG3px85uROlS5fW8uXLdfHiRYf/SGT2ustM27ZttW7dOn322Wf617/+lWF7fHy8Vq9erXbt2tmDRfpRw/PnzzusZ3XzEWlX3lfOuNVYe3p6qnPnzpo5c6bGjRunBQsWqHfv3rcNq02aNFFwcLDmzZunYcOG3fUSMV988YV8fX21ePFih+AaGxvr0C/9b+GBAwccjmLd/LylfyMxNTU1V98TeR1ztHBPGzJkiAICAtSrVy+dPHkyw/YDBw5oypQpkmRfFPHmldXTj7Bk9o0bZ9WpU0cRERF64403MpwGkHTLr/yn/zG+8ShISkqK3nnnnQx9AwICnDrlER4ersjISM2aNcvhq/bbt2/XkiVL7GORXa5du6bp06fbr6ekpGj69OkKCwtTnTp1JF0/SnP06FHNmDEjw+0vX75sn3clXX+cWS0R0KRJE23YsEErV660B61ChQqpcuXK9m9IprdL1482BAUF6fXXX9fVq1cz7C/9uSlcuLCaN2+u6dOn6/jx41n2s1pmrwcp4+tWkj3YOrOcQmb7NcbY3x93onXr1rp27ZqmTZtmb0tNTdXUqVOdun2fPn1UtGhRDR482GHul3R9blr37t1ls9k0ZMgQe3t6cP7pp5/sbenLgdzIlfeVM2431tHR0Tp37pz69OmjixcvZhocb+bv768hQ4Zo+/btGjp0aKZHQl05Ourp6SmbzeZwdC8+Pj7Dryy0atVKkvTf//7Xof3m15inp6f+8Y9/6Isvvsj0zEFOvSfyOo5o4Z4WERGhTz75RJ06dVLlypUdVoZP/59yt27dJEk1a9ZU165d9d5779lPK2zcuFGzZs1S+/btdf/9999xHR4eHnr//ffVqlUrVa1aVd27d1fx4sV19OhRrVy5UkFBQfr2228zvW2jRo0UGhqqrl27qn///rLZbPr4448z/QNbp04dffrppxo4cKD+9re/KX/+/Grbtm2m+50wYYJatWqlhg0bqmfPnvblHYKDg7P9t8yKFSumcePGKT4+XhUrVtSnn36qrVu36r333rOvVh8dHa358+frmWee0cqVKxUVFaXU1FTt3r1b8+fPt69Zlv44ly1bpkmTJqlYsWIqW7asfV2sJk2a6LXXXtPhw4cdAlXTpk01ffp0lSlTRiVKlLC3BwUFadq0aYqOjlbt2rX15JNPKiwsTH/88YcWLlyoqKgovfXWW5Kkt99+W40bN1b16tXVu3dvlStXTidPntT69et15MgRbdu2LVvHLTNBQUFq2rSpxo8fr6tXr6p48eJasmRJpkdi0kPs8OHD9eSTT8rLy0tt27bN9GhlpUqVFBERoUGDBuno0aMKCgrSF1984fR8ssy0bdtWUVFRGjp0qOLj41WlShV9+eWXTs9/Cg0N1eeff67WrVurdu3aGVaGP3jwoN566y2HNdH+/ve/q1SpUurZs6cGDx4sT09Pffjhh/bnNJ0r7ytn3G6sa9WqpWrVqtm/9HHz2m5ZGTp0qHbt2qUJEyZoyZIl+sc//qESJUro3LlziouL02effabChQs7tbRFmzZtNGnSJD388MPq3LmzTp06pbffflvly5fXr7/+au8XGRmpp556Su+8844SEhLUqFEjLV++XPv378+wz7Fjx2rlypWqX7++evfurSpVqujs2bOKi4vTsmXLMixRgjuQ8190BFy3d+9e07t3b1OmTBnj7e1tAgMDTVRUlJk6darD17ivXr1qRo0aZcqWLWu8vLxMyZIlzbBhwxz6GHN9eYc2bdpkuJ/05R0+++yzTOvYsmWLeeyxx0zBggWNj4+PKV26tOnYsaNZvny5vU9mX0Nfu3atadCggfHz8zPFihUzQ4YMMYsXLzaSzMqVK+39Ll68aDp37mxCQkIclgPIbHkHY4xZtmyZiYqKMn5+fiYoKMi0bdvW7Ny506FP+tfl//zzT4f2zOrMTLNmzUzVqlXN5s2bTcOGDY2vr68pXbq0eeuttzL0TUlJMePGjTNVq1Y1Pj4+JjQ01NSpU8eMGjXKJCQk2Pvt3r3bNG3a1Pj5+RlJDks9JCYmGk9PTxMYGGiuXbtmb589e7aRZKKjozOtc+XKlaZly5YmODjY+Pr6moiICNOtWzezefNmh34HDhwwXbp0MUWLFjVeXl6mePHi5pFHHjGff/55hrHZtGlThvu4+TnLTFZjnu7IkSOmQ4cOJiQkxAQHB5snnnjCHDt2LMNX740xZvTo0aZ48eLGw8Pjts/Xzp07TYsWLUz+/PlNoUKFTO/evc22bdsyvHa6du1qAgICsqz7RmfOnDHR0dEmKCjIBAcHm+joaLNlyxanlndIFx8fb55++mlTqlQpky9fPvtyF8uWLcu0/y+//GLq169vvL29TalSpcykSZPu6n3lzPIOxtx+rMePH28kmddff92px32jr776yrRu3dqEhYWZfPnymZCQENO4cWMzYcKEDMuFSDL9+vXLdD8ffPCBqVChgvHx8TGVKlUysbGxmT5vly9fNv379zcFCxY0AQEBpm3btubw4cOZPu6TJ0+afv36mZIlSxovLy9TtGhR8+CDD5r33nvP5ceJjGzGZNNMWAAAnLB8+XK1bt1ajRs31vfffy9vb+/cLskpU6ZM0fPPP6/4+PgM30gFskLQAgDkuHnz5qlz58566qmnNHv2bJcm/ucGY4xq1qypggULOrUmHJCOoAUAQBYuXbqkb775RitXrtSMGTP09ddf238iCnAGQQsAgCzEx8erbNmyCgkJ0bPPPqvXXnstt0uCmyFoAQAAWIR1tAAAACxC0AIAALAIC5bmorS0NB07dkyBgYH3/DduAADAdcYYXbhwQcWKFbvtTysRtHLRsWPHnPqhXQAAcO85fPiwwy9VZIaglYsCAwMlXX+igoKCcrkaAADgjMTERJUsWdL+OX4rBK1clH66MCgoiKAFAICbcWbaD5PhAQAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAs8pdYGb558+aKjIzU5MmTJUllypTRgAEDNGDAgFytyyplhi7M0BY/tk0uVAIAQO4oO3ShzA3XbZIO5cJn4V/yiNamTZv09NNP53YZlsgsZN2qHQCAvKbMTSFLkoxy57PwLxm0wsLC5O/vn9tlZLvbvYAIWwCAvO5e+yzM1aDVvHlzPffccxowYIBCQ0NVpEgRzZgxQ5cuXVL37t0VGBio8uXL6/vvv7ffZvv27WrVqpXy58+vIkWKKDo6WqdPn7Zvv3Tpkrp06aL8+fMrPDxcEydOzHC/ZcqUsZ9GjI+Pl81m09atW+3bz58/L5vNplWrVkmSVq1aJZvNpsWLF6tWrVry8/PTAw88oFOnTun7779X5cqVFRQUpM6dOyspKcmSsbodZ184hC0AQF5V1snPOGf7ZYdcP6I1a9YsFSpUSBs3btRzzz2nvn376oknnlCjRo0UFxenv//974qOjlZSUpLOnz+vBx54QLVq1dLmzZv1ww8/6OTJk+rYsaN9f4MHD9aPP/6or7/+WkuWLNGqVasUFxeXLbWOHDlSb731ltatW6fDhw+rY8eOmjx5sj755BMtXLhQS5Ys0dSpU7O8fXJyshITEx0uAAAge9x8uvBu+2WHXJ8MX7NmTb300kuSpGHDhmns2LEqVKiQevfuLUkaMWKEpk2bpl9//VXLli1TrVq19Prrr9tv/+GHH6pkyZLau3evihUrpg8++ECzZ8/Wgw8+KOl6kCtRokS21Prqq68qKipKktSzZ08NGzZMBw4cULly5SRJjz/+uFauXKkXX3wx09uPGTNGo0aNypZaAADAvS/Xj2jVqFHD/m9PT08VLFhQ1atXt7cVKVJEknTq1Clt27ZNK1euVP78+e2XSpUqSZIOHDigAwcOKCUlRfXr17ffvkCBArrvvvuyvdYiRYrI39/fHrLS206dOpXl7YcNG6aEhAT75fDhw9lSFwAAuDfl+hEtLy8vh+s2m82hzWazSZLS0tJ08eJFtW3bVuPGjcuwn/DwcO3fv9/l+/fwuJ41jfnfgcSrV6/ettab60xvS0tLy/K+fHx85OPj43KNAADg9mxy7rSgzepCbpDrR7RcUbt2be3YsUNlypRR+fLlHS4BAQGKiIiQl5eXNmzYYL/NuXPntHfv3iz3GRYWJkk6fvy4ve3GifHuwtl1slhPCwCQVzm7TlZOrqflVkGrX79+Onv2rJ566ilt2rRJBw4c0OLFi9W9e3elpqYqf/786tmzpwYPHqwVK1Zo+/bt6tatm/2oVWb8/PzUoEEDjR07Vrt27dKPP/5onzPmbm4XoghZAIC87l77LHSroFWsWDGtXbtWqamp+vvf/67q1atrwIABCgkJsYepCRMmqEmTJmrbtq1atGihxo0bq06dOrfc74cffqhr166pTp06GjBggF599dWceDiWyOoFRMgCAPxVxI9tk+H0oE2581loMzdOTkKOSkxMVHBwsBISEhQUFJTb5QAAACe48vntVke0AAAA3AlBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwyB0FrY8//lhRUVEqVqyYfv/9d0nS5MmT9fXXX2drcQAAAO7M5aA1bdo0DRw4UK1bt9b58+eVmpoqSQoJCdHkyZOzuz4AAAC35XLQmjp1qmbMmKHhw4fL09PT3l63bl399ttv2VocAACAO3M5aB06dEi1atXK0O7j46NLly5lS1EAAAB5gctBq2zZstq6dWuG9h9++EGVK1fOjpoAAADyhHyu3mDgwIHq16+frly5ImOMNm7cqLlz52rMmDF6//33ragRAADALbkctHr16iU/Pz+99NJLSkpKUufOnVWsWDFNmTJFTz75pBU1AgAAuCWbMcbc6Y2TkpJ08eJFFS5cODtr+stITExUcHCwEhISFBQUlNvlAAAAJ7jy+e3yEa0b+fv7y9/f/252AQAAkGc5FbRq1aolm83m1A7j4uLuqiAAAIC8wqmg1b59e/u/r1y5onfeeUdVqlRRw4YNJUk///yzduzYoWeffdaSIgEAANyRU0ErJibG/u9evXqpf//+Gj16dIY+hw8fzt7qAAAA3JjLk+GDg4O1efNmVahQwaF93759qlu3rhISErK1wLyMyfAAALgfVz6/XV6w1M/PT2vXrs3QvnbtWvn6+rq6OwAAgDzL5W8dDhgwQH379lVcXJzq1asnSdqwYYM+/PBDvfzyy9leIAAAgLtyOWgNHTpU5cqV05QpUzR79mxJUuXKlRUbG6uOHTtme4EAAADu6q4WLMXdYY4WAADux9I5WgAAAHCOy6cOU1NT9eabb2r+/Pn6448/lJKS4rD97Nmz2VYcAACAO3P5iNaoUaM0adIkderUSQkJCRo4cKAee+wxeXh4aOTIkRaUCAAA4J5cDlpz5szRjBkz9MILLyhfvnx66qmn9P7772vEiBH6+eefragRAADALbkctE6cOKHq1atLkvLnz29foPSRRx7RwoULs7c6AAAAN+Zy0CpRooSOHz8uSYqIiNCSJUskSZs2bZKPj0/2VgcAAODGXA5aHTp00PLlyyVJzz33nF5++WVVqFBBXbp0UY8ePbK9QAAAAHd11+to/fzzz1q3bp0qVKigtm3bZlddfwmsowUAgPtx5fPbpeUdrl69qj59+ujll19W2bJlJUkNGjRQgwYN7rxaAACAPMqlU4deXl764osvrKoFAAAgT3F5jlb79u21YMECC0oBAADIW1xeGb5ChQp65ZVXtHbtWtWpU0cBAQEO2/v3759txQEAALgzlyfDp8/NynRnNpsOHjx410X9VTAZHgAA92PZZHhJOnTo0B0XBgAA8Ffi8hwtAAAAOMfpI1rnz5/X3Llz1bdvX0nSP//5T12+fNm+3dPTUzNmzFBISEi2FwkAAOCOnD6iNWPGDK1Zs8Z+/ZtvvpGHh4eCg4MVHBys3377TZMnT7aiRgAAALfkdND6/PPP1b17d4e28ePHKzY2VrGxsRozZoy+/vrrbC8QAADAXTkdtA4ePKj77rvPfv2+++6Tt7e3/XrNmjW1b9++7K0OAADAjTkdtC5duqSEhAT79c2bN6tEiRIO29PS0rK3OgAAADfmdNAqV66c4uListy+efPmW66xBQAA8FfjdNDq0KGDXnrpJZ08eTLDthMnTigmJkYdOnTI1uIAAADcmdMrw1+4cEH169fXkSNHFB0drYoVK0qS9uzZo9mzZ6t48eLauHGjAgMDLS04uzVv3lyRkZG58o1Jq1aGbzx0oY7ccL2EpDVj22Tb/gEAuNeVGbowQ1t8Nn0WWrIyfGBgoNauXathw4Zp7ty5On/+vCQpJCREnTt31uuvv+52ISsvyuyFdeT/t2fXCwwAgHtZZp+F6e05/Vno0srwoaGhevfdd3XmzBmdOHFCJ06c0JkzZ/Tuu++qQIECVtUIJ2X1wnJ2OwAA7u5e+yy8o5/gsdlsKly4sAoXLiybzZbdNeWac+fOqUuXLgoNDZW/v79atWplX7LCGKOwsDB9/vnn9v6RkZEKDw+3X1+zZo18fHyUlJSU47U3dvKF42w/AADcjbMhKifDFr91eINu3bpp8+bN+uabb7R+/XoZY9S6dWtdvXpVNptNTZs21apVqyRdD2W7du3S5cuXtXv3bknSjz/+qL/97W/y9/fPdP/JyclKTEx0uGSXI7fv4lI/AABw9wha/9++ffv0zTff6P3331eTJk1Us2ZNzZkzR0ePHtWCBQskXZ84nx60fvrpJ9WqVcuhbdWqVWrWrFmW9zFmzBj7TxYFBwerZMmSFj8qAACQmwha/9+uXbuUL18+1a9f395WsGBB3Xfffdq1a5ckqVmzZtq5c6f+/PNP/fjjj2revLk9aF29elXr1q1T8+bNs7yPYcOGKSEhwX45fPiw1Q8LAADkIpeD1sGDB62owy1Ur15dBQoU0I8//ugQtH788Udt2rRJV69eVaNGjbK8vY+Pj4KCghwu2aXE7bu41A8AANw9l4NW+fLldf/992v27Nm6cuWKFTXlisqVK+vatWvasGGDve3MmTPas2ePqlSpIun6lwCaNGmir7/+Wjt27FDjxo1Vo0YNJScna/r06apbt64CAgJypX5n18liPS0AQF7l7NINObnEg8tBKy4uTjVq1NDAgQNVtGhR9enTRxs3brSithxVoUIFtWvXTr1799aaNWu0bds2/etf/1Lx4sXVrl07e7/mzZtr7ty5ioyMVP78+eXh4aGmTZtqzpw5t5yflRNu98JhHS0AQF53r30Wuhy0IiMjNWXKFB07dkwffvihjh8/rsaNG6tatWqaNGmS/vzzTyvqzBGxsbGqU6eOHnnkETVs2FDGGC1atEheXl72Ps2aNVNqaqrDXKzmzZtnaMst8WPbZDg9WEKELADAX0dWn3m58Vno9E/wZCU5OVnvvPOOhg0bppSUFHl7e6tjx44aN26cwxpTyMiqn+ABAADWceXz+46/dbh582Y9++yzCg8P16RJkzRo0CAdOHBAS5cu1bFjxxxOtwEAAPwVOf1bh+kmTZqk2NhY7dmzR61bt9ZHH32k1q1by8PjemYrW7asZs6cqTJlymR3rQAAAG7F5aA1bdo09ejRQ926dcvy1GDhwoX1wQcf3HVxAAAA7szlOVrx8fEqVaqU/QhWOmOMDh8+rFKlSmVrgXkZc7QAAHA/ls7RioiI0OnTpzO0nz17VmXLlnV1dwAAAHmWy0ErqwNgFy9elK+v710XBAAAkFc4PUdr4MCBkq6vjj5ixAj5+/vbt6WmpmrDhg2KjIzM9gIBAADcldNBa8uWLZKuH9H67bff5O3tbd/m7e2tmjVratCgQdlfIQAAgJtyOmitXLlSktS9e3dNmTKFydsAAAC34fLyDrGxsVbUAQAAkOc4FbQee+wxzZw5U0FBQXrsscdu2ffLL7/MlsIAAADcnVNBKzg4WDabzf5vAAAA3N5d/6g07hwLlgIA4H5y5EelAQAAcGtOnTqsVauW/dTh7cTFxd1VQQAAAHmFU0Grffv2FpcBAACQ9zBHKxcxRwsAAPfDHC0AAIB7gMsLlqampurNN9/U/Pnz9ccffyglJcVh+9mzZ7OtOAAAAHfm8hGtUaNGadKkSerUqZMSEhI0cOBAPfbYY/Lw8NDIkSMtKBEAAMA9uRy05syZoxkzZuiFF15Qvnz59NRTT+n999/XiBEj9PPPP1tRIwAAgFtyOWidOHFC1atXlyTlz59fCQkJkqRHHnlECxcuzN7qAAAA3JjLQatEiRI6fvy4JCkiIkJLliyRJG3atEk+Pj7ZWx0AAIAbczlodejQQcuXL5ckPffcc3r55ZdVoUIFdenSRT169Mj2AgEAANzVXa+jtX79eq1fv14VKlRQ27Zts6uuvwTW0QIAwP248vnt8vION2vYsKEaNmx4t7sBAADIc1wOWh999NEtt3fp0uWOiwEAAMhLXD51GBoa6nD96tWrSkpKkre3t/z9/Vmw1AWcOgQAwP1Y+hM8586dc7hcvHhRe/bsUePGjTV37tw7LhoAACCvyZbfOqxQoYLGjh2r//znP9mxOwAAgDwh235UOl++fDp27Fh27Q4AAMDtuTwZ/ptvvnG4bozR8ePH9dZbbykqKirbCgMAAHB3Lget9u3bO1y32WwKCwvTAw88oIkTJ2ZXXQAAAG7P5aCVlpZmRR0AAAB5zh3P0Tp9+rQSExOzsxYAAIA8xaWgdf78efXr10+FChVSkSJFFBoaqqJFi2rYsGFKSkqyqkYAAAC35PSpw7Nnz6phw4Y6evSo/vnPf6py5cqSpJ07d2rq1KlaunSp1qxZo19//VU///yz+vfvb1nRAAAA7sDpoPXKK6/I29tbBw4cUJEiRTJs+/vf/67o6GgtWbJE//3vf7O9UAAAAHfjdNBasGCBpk+fniFkSVLRokU1fvx4tW7dWjExMeratWu2FgkAAOCOnJ6jdfz4cVWtWjXL7dWqVZOHh4diYmKypTAAAAB353TQKlSokOLj47PcfujQIRUuXDg7agIAAMgTnA5aLVu21PDhw5WSkpJhW3Jysl5++WU9/PDD2VocAACAO7MZY4wzHY8cOaK6devKx8dH/fr1U6VKlWSM0a5du/TOO+8oOTlZmzZtUqlSpayuOc9ITExUcHCwEhISFBQUlNvlAAAAJ7jy+e30ZPgSJUpo/fr1evbZZzVs2DCl5zObzaaHHnpIb731FiELAADgBi79BE/ZsmX1/fff69y5c9q3b58kqXz58ipQoIAlxQEAALgzl3/rUJJCQ0NVr1697K4FAAAgT7nj3zoEAADArRG0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwCEELAADAIgQtAAAAixC0AAAALJIvtwvIKc2bN1dkZKQmT56c6XabzaavvvpK7du3d2p/q1at0v33369z584pJCQk2+rMDqlpRhsPndWpC1dUONBX9coWkKeHLbfLAgAgx5QZujBDW/zYNjlex18maN3O8ePHFRoamttl3LUfth/XqG936njCFXtbeLCvYtpW0cPVwnOxMgAAckZmISu9PafDFqcO/7+iRYvKx8cnt8u4Kz9sP66+s+McQpYknUi4or6z4/TD9uO5VBkAADkjq5Dl7Pbs9pcKWmlpaRoyZIgKFCigokWLauTIkfZtNptNCxYssF9ft26dIiMj5evrq7p162rBggWy2WzaunWrwz5/+eUX1a1bV/7+/mrUqJH27NmTMw/mJqlpRqO+3SmTybb0tlHf7lRqWmY9AABwf86GqJwMW3+poDVr1iwFBARow4YNGj9+vF555RUtXbo0Q7/ExES1bdtW1atXV1xcnEaPHq0XX3wx030OHz5cEydO1ObNm5UvXz716NEjy/tPTk5WYmKiwyW7bDx0NsORrBsZSccTrmjjobPZdp8AAODW/lJBq0aNGoqJiVGFChXUpUsX1a1bV8uXL8/Q75NPPpHNZtOMGTNUpUoVtWrVSoMHD850n6+99pqaNWumKlWqaOjQoVq3bp2uXMk88IwZM0bBwcH2S8mSJbPtsZ26kHXIupN+AADg7v3lgtaNwsPDderUqQz99uzZoxo1asjX19feVq9evdvuMzz8+mTzzPYpScOGDVNCQoL9cvjwYZcfQ1YKB/revpML/QAAwN37S33r0MvLy+G6zWZTWlpatu3TZru+hEJW+/Tx8bFswn29sgUUHuyrEwlXMp2nZZNUNPj6Ug8AACBn/KWOaDnrvvvu02+//abk5GR726ZNm3Kxotvz9LAppm0VSddD1Y3Sr8e0rcJ6WgCAPMvZpRtycokHglYmOnfurLS0ND399NPatWuXFi9erDfeeEPS/45a3Yserhauaf+qraLBjqcHiwb7atq/arOOFgAgz7tdiMrpdbT+UqcOnRUUFKRvv/1Wffv2VWRkpKpXr64RI0aoc+fODvO27kUPVwvXQ1WKsjI8AOAvK35sm3tmZXibMYaFlZwwZ84cde/eXQkJCfLz88uWfSYmJio4OFgJCQkKCgrKln0CAABrufL5zRGtLHz00UcqV66cihcvrm3btunFF19Ux44dsy1kAQCAvI+glYUTJ05oxIgROnHihMLDw/XEE0/otddey+2yAACAG+HUYS7i1CEAAO7Hlc9vvnUIAABgEYIWAACARQhaAAAAFiFoAQAAWISgBQAAYBGCFgAAgEVYRysXpa+skZiYmMuVAAAAZ6V/bjuzQhZBKxdduHBBklSyZMlcrgQAALjqwoULCg4OvmUfFizNRWlpaTp27JgCAwNls2Xvjz4nJiaqZMmSOnz4MIuhWohxzhmMc85gnHMOY50zrBpnY4wuXLigYsWKycPj1rOwOKKVizw8PFSiRAlL7yMoKIg3cQ5gnHMG45wzGOecw1jnDCvG+XZHstIxGR4AAMAiBC0AAACLELTyKB8fH8XExMjHxye3S8nTGOecwTjnDMY55zDWOeNeGGcmwwMAAFiEI1oAAAAWIWgBAABYhKAFAABgEYIWAACARQhabuztt99WmTJl5Ovrq/r162vjxo237P/ZZ5+pUqVK8vX1VfXq1bVo0aIcqtS9uTLOM2bMUJMmTRQaGqrQ0FC1aNHits8LrnP19Zxu3rx5stlsat++vbUF5hGujvP58+fVr18/hYeHy8fHRxUrVuRvhxNcHefJkyfrvvvuk5+fn0qWLKnnn39eV65cyaFq3dNPP/2ktm3bqlixYrLZbFqwYMFtb7Nq1SrVrl1bPj4+Kl++vGbOnGl5nTJwS/PmzTPe3t7mww8/NDt27DC9e/c2ISEh5uTJk5n2X7t2rfH09DTjx483O3fuNC+99JLx8vIyv/32Ww5X7l5cHefOnTubt99+22zZssXs2rXLdOvWzQQHB5sjR47kcOXuxdVxTnfo0CFTvHhx06RJE9OuXbucKdaNuTrOycnJpm7duqZ169ZmzZo15tChQ2bVqlVm69atOVy5e3F1nOfMmWN8fHzMnDlzzKFDh8zixYtNeHi4ef7553O4cveyaNEiM3z4cPPll18aSearr766Zf+DBw8af39/M3DgQLNz504zdepU4+npaX744QdL6yRoual69eqZfv362a+npqaaYsWKmTFjxmTav2PHjqZNmzYObfXr1zd9+vSxtE535+o43+zatWsmMDDQzJo1y6oS84Q7Gedr166ZRo0amffff9907dqVoOUEV8d52rRpply5ciYlJSWnSswTXB3nfv36mQceeMChbeDAgSYqKsrSOvMSZ4LWkCFDTNWqVR3aOnXqZFq2bGlhZcZw6tANpaSk6JdfflGLFi3sbR4eHmrRooXWr1+f6W3Wr1/v0F+SWrZsmWV/3Nk43ywpKUlXr15VgQIFrCrT7d3pOL/yyisqXLiwevbsmRNlur07GedvvvlGDRs2VL9+/VSkSBFVq1ZNr7/+ulJTU3OqbLdzJ+PcqFEj/fLLL/bTiwcPHtSiRYvUunXrHKn5ryK3Pgf5UWk3dPr0aaWmpqpIkSIO7UWKFNHu3bszvc2JEycy7X/ixAnL6nR3dzLON3vxxRdVrFixDG9u/M+djPOaNWv0wQcfaOvWrTlQYd5wJ+N88OBBrVixQv/85z+1aNEi7d+/X88++6yuXr2qmJiYnCjb7dzJOHfu3FmnT59W48aNZYzRtWvX9Mwzz+j//u//cqLkv4ysPgcTExN1+fJl+fn5WXK/HNECLDJ27FjNmzdPX331lXx9fXO7nDzjwoULio6O1owZM1SoUKHcLidPS0tLU+HChfXee++pTp066tSpk4YPH6533303t0vLU1atWqXXX39d77zzjuLi4vTll19q4cKFGj16dG6XhmzAES03VKhQIXl6eurkyZMO7SdPnlTRokUzvU3RokVd6o87G+d0b7zxhsaOHatly5apRo0aVpbp9lwd5wMHDig+Pl5t27a1t6WlpUmS8uXLpz179igiIsLaot3Qnbyew8PD5eXlJU9PT3tb5cqVdeLECaWkpMjb29vSmt3RnYzzyy+/rOjoaPXq1UuSVL16dV26dElPP/20hg8fLg8Pjolkh6w+B4OCgiw7miVxRMsteXt7q06dOlq+fLm9LS0tTcuXL1fDhg0zvU3Dhg0d+kvS0qVLs+yPOxtnSRo/frxGjx6tH374QXXr1s2JUt2aq+NcqVIl/fbbb9q6dav98uijj+r+++/X1q1bVbJkyZws323cyes5KipK+/fvtwdZSdq7d6/Cw8MJWVm4k3FOSkrKEKbSw63h54izTa59Dlo61R6WmTdvnvHx8TEzZ840O3fuNE8//bQJCQkxJ06cMMYYEx0dbYYOHWrvv3btWpMvXz7zxhtvmF27dpmYmBiWd3CCq+M8duxY4+3tbT7//HNz/Phx++XChQu59RDcgqvjfDO+degcV8f5jz/+MIGBgebf//632bNnj/nuu+9M4cKFzauvvppbD8EtuDrOMTExJjAw0MydO9ccPHjQLFmyxERERJiOHTvm1kNwCxcuXDBbtmwxW7ZsMZLMpEmTzJYtW8zvv/9ujDFm6NChJjo62t4/fXmHwYMHm127dpm3336b5R1wa1OnTjWlSpUy3t7epl69eubnn3+2b2vWrJnp2rWrQ//58+ebihUrGm9vb1O1alWzcOHCHK7YPbkyzqVLlzaSMlxiYmJyvnA34+rr+UYELee5Os7r1q0z9evXNz4+PqZcuXLmtddeM9euXcvhqt2PK+N89epVM3LkSBMREWF8fX1NyZIlzbPPPmvOnTuX84W7kZUrV2b69zZ9bLt27WqaNWuW4TaRkZHG29vblCtXzsTGxlpep80YjksCAABYgTlaAAAAFiFoAQAAWISgBQAAYBGCFgAAgEUIWgAAABYhaAEAAFiEoAUAAGARghYAAIBFCFoA3Fq3bt1ks9kyXPbv33/b286cOVMhISFO3U9KSorGjx+vmjVryt/fX4UKFVJUVJRiY2N19erVu3wU/zNy5EhFRkZm2/5u1Lx5cw0YMMCSfQPIXL7cLgAA7tbDDz+s2NhYh7awsLBs239KSopatmypbdu2afTo0YqKilJQUJB+/vlnvfHGG6pVq5Zl4QiAm7P8R34AwEK3+p3DiRMnmmrVqhl/f39TokQJ07dvX/sPfGf2O2lZ/SbluHHjjIeHh4mLi8uwLSUlxVy8eNEYY8yVK1fMc889Z8LCwoyPj4+JiooyGzdutPdNv89ly5aZOnXqGD8/P9OwYUOze/duY4wxsbGxGWpK/y22c+fOmZ49e5pChQqZwMBAc//995utW7fa9x0TE2Nq1qxpPvroI1O6dGkTFBRkOnXqZBITE+3jdPO+Dx065MpQA7gDnDoEkGd5eHjov//9r3bs2KFZs2ZpxYoVGjJkiCSpUaNGmjx5soKCgnT8+HEdP35cgwYNynQ/c+bMUYsWLVSrVq0M27y8vBQQECBJGjJkiL744gvNmjVLcXFxKl++vFq2bKmzZ8863Gb48OGaOHGiNm/erHz58qlHjx6SpE6dOumFF15Q1apV7TV16tRJkvTEE0/o1KlT+v777/XLL7+odu3aevDBBx32feDAAS1YsEDfffedvvvuO/34448aO3asJGnKlClq2LChevfubd93yZIl73KEAdxWbic9ALgbXbt2NZ6eniYgIMB+efzxxzPt+9lnn5mCBQvar8fGxprg4ODb3oefn5/p37//LftcvHjReHl5mTlz5tjbUlJSTLFixcz48eONMY5HtNItXLjQSDKXL182xvzvyNSNVq9ebYKCgsyVK1cc2iMiIsz06dPtt/P397cfwTLGmMGDB5v69evbrzdr1sz85z//ue3jBZB9mKMFwO3df//9mjZtmv16+hGmZcuWacyYMdq9e7cSExN17do1XblyRUlJSfL393d6/8aY2/Y5cOCArl69qqioKHubl5eX6tWrp127djn0rVGjhv3f4eHhkqRTp06pVKlSme5727ZtunjxogoWLOjQfvnyZR04cMB+vUyZMgoMDHTY96lTp25bOwDrELQAuL2AgACVL1/eoS0+Pl6PPPKI+vbtq9dee00FChTQmjVr1LNnT6WkpLgUtCpWrKjdu3dnW71eXl72f9tsNklSWlpalv0vXryo8PBwrVq1KsO2G781eeN+0/d9q/0CsB5ztADkSb/88ovS0tI0ceJENWjQQBUrVtSxY8cc+nh7eys1NfW2++rcubOWLVumLVu2ZNh29epVXbp0SREREfL29tbatWsdtm3atElVqlRxuu7Maqpdu7ZOnDihfPnyqXz58g6XQoUK3dW+AViLoAUgTypfvryuXr2qqVOn6uDBg/r444/17rvvOvQpU6aMLl68qOXLl+v06dNKSkrKdF8DBgxQVFSUHnzwQb399tvatm2bDh48qPnz56tBgwbat2+fAgIC1LdvXw0ePFg//PCDdu7cqd69eyspKUk9e/Z0uu4yZcro0KFD2rp1q06fPq3k5GS1aNFCDRs2VPv27bVkyRLFx8dr3bp1Gj58uDZv3uzSvjds2KD4+HidPn2ao11ADiBoAciTatasqUmTJmncuHGqVq2a5syZozFjxjj0adSokZ555hl16tRJYWFhGj9+fKb78vHx0dKlSzVkyBBNnz5dDRo00N/+9jf997//Vf/+/VWtWjVJ0tixY/WPf/xD0dHRql27tvbv36/FixcrNDTU6br/8Y9/6OGHH9b999+vsLAwzZ07VzabTYsWLVLTpk3VvXt3VaxYUU8++aR+//13FSlSxOl9Dxo0SJ6enqpSpYrCwsL0xx9/OH1bAHfGZpyZ5QkAAACXcUQLAADAIgQtAAAAixC0AAAALELQAgAAsAhBCwAAwCIELQAAAIsQtAAAACxC0AIAALAIQQsAAMAiBC0AAACLELQAAAAsQtACAACwyP8DQYWsKbnL2oMAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "id": "85fc338a",
      "metadata": {
        "id": "85fc338a"
      },
      "outputs": [],
      "source": [
        "counts = data['Grade'].value_counts()\n",
        "grade_percentages = [(count/len(data_cp))*100 for count in counts]\n",
        "grade_label = counts.index.tolist()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "id": "3712f67c",
      "metadata": {
        "id": "3712f67c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 444
        },
        "outputId": "faf8382f-dd0a-4a3d-945e-2e98fd1e008e"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 500x500 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "sb.set_style('dark')\n",
        "plt.figure(figsize=(5,5))\n",
        "plt.title('Milk Quality Grade %')\n",
        "plt.pie(grade_percentages, labels=grade_label, autopct='%1.1f%%',shadow=True);"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "b44ea7be",
      "metadata": {
        "id": "b44ea7be"
      },
      "source": [
        "# **Train & Evaluate the Model**\n",
        "Metrics such as accuracy, precision, and recall can be used measure the performance of the model."
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Split the dataset into features(to use to predict) and target variable(to predict)\n",
        "X = data_cp.drop(['Grade'], axis=1)\n",
        "y = data_cp['Grade']"
      ],
      "metadata": {
        "id": "jQi_xT7KoOpO"
      },
      "id": "jQi_xT7KoOpO",
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "'''\n",
        "# Preprocess the data\n",
        "# Encode categorical variables\n",
        "# Not necessary in this case as the feature-values are already in the numerical category format\n",
        "Le = LabelEncoder()\n",
        "X['Taste'] = Le.fit_transform(X['Taste'])\n",
        "X['Odor'] = Le.fit_transform(X['Odor'])\n",
        "X['Fat'] = Le.fit_transform(X['Fat'])\n",
        "X['Turbidity'] = Le.fit_transform(X['Turbidity'])\n",
        "'''"
      ],
      "metadata": {
        "id": "4ZMQWFmsodm1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 70
        },
        "outputId": "2cc0b508-d1b5-4549-a8b9-7d5a55b594c7"
      },
      "id": "4ZMQWFmsodm1",
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "\"\\n# Preprocess the data\\n# Encode categorical variables\\n# Not necessary in this case as the feature-values are already in the numerical category format\\nLe = LabelEncoder()\\nX['Taste'] = Le.fit_transform(X['Taste'])\\nX['Odor'] = Le.fit_transform(X['Odor'])\\nX['Fat'] = Le.fit_transform(X['Fat'])\\nX['Turbidity'] = Le.fit_transform(X['Turbidity'])\\n\""
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(X.columns)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6aVeMv2GpW2S",
        "outputId": "37f51138-aba8-4c56-a68e-d2314ab5449e"
      },
      "id": "6aVeMv2GpW2S",
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Index(['pH', 'Temperature', 'Taste', 'Odor', 'Fat', 'Turbidity', 'Colour'], dtype='object')\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Support Vector Machines (SVM) is sensitive to differences in scale and distribution, so using a scaler-object we can: standardize and normalize the features."
      ],
      "metadata": {
        "id": "S1_MVSp6NhhA"
      },
      "id": "S1_MVSp6NhhA"
    },
    {
      "cell_type": "code",
      "source": [
        "# Scale numerical variables\n",
        "scaler = StandardScaler()\n",
        "\n",
        "# Transform col-values to have zero mean and variance\n",
        "X[['pH', 'Temperature', 'Fat', 'Turbidity']] = scaler.fit_transform(X[['pH', 'Temperature', 'Fat', 'Turbidity']])"
      ],
      "metadata": {
        "id": "avYaCxVKohfG"
      },
      "id": "avYaCxVKohfG",
      "execution_count": 39,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Split the data into training and testing sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)"
      ],
      "metadata": {
        "id": "mNb5vpwpoj76"
      },
      "id": "mNb5vpwpoj76",
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Train and evaluate a support vector machine classifier\n",
        "svm = SVC(kernel='linear')\n",
        "svm.fit(X_train, y_train)\n",
        "\n",
        "y_pred_svm = svm.predict(X_test)\n",
        "accuracy_svm = accuracy_score(y_test, y_pred_svm)\n",
        "\n",
        "print(\"Accuracy of SVM classifier: {:.2f}%\".format(accuracy_svm * 100))\n",
        "print('------------------------------------------------------------------')\n",
        "print(classification_report(y_test, y_pred_svm))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9_K4wpK0omSf",
        "outputId": "0b706cd6-7cb7-4a42-b49b-3927f51aba27"
      },
      "id": "9_K4wpK0omSf",
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy of SVM classifier: 87.42%\n",
            "------------------------------------------------------------------\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "        high       0.74      0.92      0.82        76\n",
            "         low       0.90      0.85      0.88       115\n",
            "      medium       0.96      0.87      0.91       127\n",
            "\n",
            "    accuracy                           0.87       318\n",
            "   macro avg       0.87      0.88      0.87       318\n",
            "weighted avg       0.89      0.87      0.88       318\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Perform random classifier classification\n",
        "rf = RandomForestClassifier(random_state=42)\n",
        "rf.fit(X_train, y_train)\n",
        "\n",
        "y_pred_rf = rf.predict(X_test)\n",
        "accuracy_rf = accuracy_score(y_test, y_pred_rf)\n",
        "\n",
        "print(\"Accuracy of Random Forest classifier: {:.2f}%\".format(accuracy_rf * 100))\n",
        "print('------------------------------------------------------------------')\n",
        "print(classification_report(y_test, y_pred_rf))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9XPZRdHEop4f",
        "outputId": "6786dd79-0dc5-4514-9cc1-73d0498a0aa5"
      },
      "id": "9XPZRdHEop4f",
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy of Random Forest classifier: 99.69%\n",
            "------------------------------------------------------------------\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "        high       0.99      1.00      0.99        76\n",
            "         low       1.00      0.99      1.00       115\n",
            "      medium       1.00      1.00      1.00       127\n",
            "\n",
            "    accuracy                           1.00       318\n",
            "   macro avg       1.00      1.00      1.00       318\n",
            "weighted avg       1.00      1.00      1.00       318\n",
            "\n"
          ]
        }
      ]
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.8"
    },
    "colab": {
      "provenance": [],
      "include_colab_link": true
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}