{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "\n",
        "from sklearn.linear_model import LinearRegression\n",
        "from sklearn.ensemble import RandomForestRegressor\n",
        "\n",
        "from sklearn.metrics import (\n",
        "    mean_absolute_error,\n",
        "    mean_squared_error,\n",
        "    r2_score\n",
        ")\n",
        "\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')"
      ],
      "metadata": {
        "id": "PyazhfeIZvdW"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv(\"crop.csv\")"
      ],
      "metadata": {
        "id": "XBexo3iZZ_9W"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(df.shape)\n",
        "\n",
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 296
        },
        "id": "suy5IrmlaDBm",
        "outputId": "2c8c57f8-ac16-4cb7-8a79-5ed40db650c4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(19689, 13)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "           Crop  Crop_Year       Season  State     Area  Production  \\\n",
              "0      Arecanut       1997  Whole Year   Assam  73814.0       56708   \n",
              "1     Arhar/Tur       1997  Kharif       Assam   6637.0        4685   \n",
              "2   Castor seed       1997  Kharif       Assam    796.0          22   \n",
              "3      Coconut        1997  Whole Year   Assam  19656.0   126905000   \n",
              "4  Cotton(lint)       1997  Kharif       Assam   1739.0         794   \n",
              "\n",
              "   Annual_Rainfall  Fertilizer  Pesticide     Yield  Avg_Temperature  \\\n",
              "0           2051.4  7024878.38   22882.34     0.796           23.692   \n",
              "1           2051.4   631643.29    2057.47     0.710           23.692   \n",
              "2           2051.4    75755.32     246.76     0.238           23.692   \n",
              "3           2051.4  1870661.52    6093.36  5238.052           23.692   \n",
              "4           2051.4   165500.63     539.09     0.421           23.692   \n",
              "\n",
              "   Max_Temperature  Min_Temperature  \n",
              "0           33.435           14.779  \n",
              "1           33.435           14.779  \n",
              "2           33.435           14.779  \n",
              "3           33.435           14.779  \n",
              "4           33.435           14.779  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-2a889b82-ddd4-457a-9e6b-b859e06748a6\" 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>Crop</th>\n",
              "      <th>Crop_Year</th>\n",
              "      <th>Season</th>\n",
              "      <th>State</th>\n",
              "      <th>Area</th>\n",
              "      <th>Production</th>\n",
              "      <th>Annual_Rainfall</th>\n",
              "      <th>Fertilizer</th>\n",
              "      <th>Pesticide</th>\n",
              "      <th>Yield</th>\n",
              "      <th>Avg_Temperature</th>\n",
              "      <th>Max_Temperature</th>\n",
              "      <th>Min_Temperature</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Arecanut</td>\n",
              "      <td>1997</td>\n",
              "      <td>Whole Year</td>\n",
              "      <td>Assam</td>\n",
              "      <td>73814.0</td>\n",
              "      <td>56708</td>\n",
              "      <td>2051.4</td>\n",
              "      <td>7024878.38</td>\n",
              "      <td>22882.34</td>\n",
              "      <td>0.796</td>\n",
              "      <td>23.692</td>\n",
              "      <td>33.435</td>\n",
              "      <td>14.779</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Arhar/Tur</td>\n",
              "      <td>1997</td>\n",
              "      <td>Kharif</td>\n",
              "      <td>Assam</td>\n",
              "      <td>6637.0</td>\n",
              "      <td>4685</td>\n",
              "      <td>2051.4</td>\n",
              "      <td>631643.29</td>\n",
              "      <td>2057.47</td>\n",
              "      <td>0.710</td>\n",
              "      <td>23.692</td>\n",
              "      <td>33.435</td>\n",
              "      <td>14.779</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Castor seed</td>\n",
              "      <td>1997</td>\n",
              "      <td>Kharif</td>\n",
              "      <td>Assam</td>\n",
              "      <td>796.0</td>\n",
              "      <td>22</td>\n",
              "      <td>2051.4</td>\n",
              "      <td>75755.32</td>\n",
              "      <td>246.76</td>\n",
              "      <td>0.238</td>\n",
              "      <td>23.692</td>\n",
              "      <td>33.435</td>\n",
              "      <td>14.779</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Coconut</td>\n",
              "      <td>1997</td>\n",
              "      <td>Whole Year</td>\n",
              "      <td>Assam</td>\n",
              "      <td>19656.0</td>\n",
              "      <td>126905000</td>\n",
              "      <td>2051.4</td>\n",
              "      <td>1870661.52</td>\n",
              "      <td>6093.36</td>\n",
              "      <td>5238.052</td>\n",
              "      <td>23.692</td>\n",
              "      <td>33.435</td>\n",
              "      <td>14.779</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Cotton(lint)</td>\n",
              "      <td>1997</td>\n",
              "      <td>Kharif</td>\n",
              "      <td>Assam</td>\n",
              "      <td>1739.0</td>\n",
              "      <td>794</td>\n",
              "      <td>2051.4</td>\n",
              "      <td>165500.63</td>\n",
              "      <td>539.09</td>\n",
              "      <td>0.421</td>\n",
              "      <td>23.692</td>\n",
              "      <td>33.435</td>\n",
              "      <td>14.779</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2a889b82-ddd4-457a-9e6b-b859e06748a6')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-2a889b82-ddd4-457a-9e6b-b859e06748a6 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-2a889b82-ddd4-457a-9e6b-b859e06748a6');\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",
              "\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "df",
              "summary": "{\n  \"name\": \"df\",\n  \"rows\": 19689,\n  \"fields\": [\n    {\n      \"column\": \"Crop\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 55,\n        \"samples\": [\n          \"Groundnut\",\n          \"Dry chillies\",\n          \"Horse-gram\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Crop_Year\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 6,\n        \"min\": 1997,\n        \"max\": 2020,\n        \"num_unique_values\": 24,\n        \"samples\": [\n          2005,\n          2013,\n          1997\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Season\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 6,\n        \"samples\": [\n          \"Whole Year \",\n          \"Kharif     \",\n          \"Winter     \"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"State\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 30,\n        \"samples\": [\n          \"Telangana\",\n          \"Punjab\",\n          \"Jharkhand\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Area\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 732828.6758877125,\n        \"min\": 0.5,\n        \"max\": 50808100.0,\n        \"num_unique_values\": 13644,\n        \"samples\": [\n          636951.0,\n          2283.0,\n          11488.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Production\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 263056839,\n        \"min\": 0,\n        \"max\": 6326000000,\n        \"num_unique_values\": 14016,\n        \"samples\": [\n          263316,\n          142,\n          284572\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Annual_Rainfall\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 816.9095916652778,\n        \"min\": 301.3,\n        \"max\": 6552.7,\n        \"num_unique_values\": 634,\n        \"samples\": [\n          1019.9,\n          1711.5,\n          800.8\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Fertilizer\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 94946004.48252656,\n        \"min\": 54.17,\n        \"max\": 4835406877.0,\n        \"num_unique_values\": 18598,\n        \"samples\": [\n          45900.48,\n          199161.6,\n          879144.96\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Pesticide\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 213287.3548595091,\n        \"min\": 0.09,\n        \"max\": 15750511.0,\n        \"num_unique_values\": 17405,\n        \"samples\": [\n          4031.61,\n          1069.2,\n          4430.14\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Yield\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 878.3061935577682,\n        \"min\": 0.0,\n        \"max\": 21105.0,\n        \"num_unique_values\": 6174,\n        \"samples\": [\n          2.715,\n          2.174,\n          6.932\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Avg_Temperature\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 5.847671914595723,\n        \"min\": -5.115,\n        \"max\": 28.652,\n        \"num_unique_values\": 622,\n        \"samples\": [\n          26.237,\n          26.286,\n          26.193\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Max_Temperature\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 5.732338741447859,\n        \"min\": 7.701,\n        \"max\": 40.632,\n        \"num_unique_values\": 639,\n        \"samples\": [\n          26.985,\n          30.858,\n          17.074\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Min_Temperature\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 6.5717127573816745,\n        \"min\": -19.359,\n        \"max\": 24.972,\n        \"num_unique_values\": 633,\n        \"samples\": [\n          14.442,\n          15.154,\n          12.594\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.isnull().sum()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 492
        },
        "id": "09sOunLwaFfV",
        "outputId": "7a880994-3569-4e8f-f295-6a82f68a9ff7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Crop               0\n",
              "Crop_Year          0\n",
              "Season             0\n",
              "State              0\n",
              "Area               0\n",
              "Production         0\n",
              "Annual_Rainfall    0\n",
              "Fertilizer         0\n",
              "Pesticide          0\n",
              "Yield              0\n",
              "Avg_Temperature    0\n",
              "Max_Temperature    0\n",
              "Min_Temperature    0\n",
              "dtype: int64"
            ],
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>0</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Crop</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Crop_Year</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Season</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>State</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Area</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Production</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Annual_Rainfall</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Fertilizer</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Pesticide</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Yield</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Avg_Temperature</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Max_Temperature</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Min_Temperature</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div><br><label><b>dtype:</b> int64</label>"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.duplicated().sum()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bvcSWxXBaIKW",
        "outputId": "59a8345d-cd88-4b90-a497-6cd6d8eb3da1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "np.int64(0)"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.info()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "msFurYqgaJll",
        "outputId": "aae67041-10ad-4439-9dee-7ee453ddaa9b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 19689 entries, 0 to 19688\n",
            "Data columns (total 13 columns):\n",
            " #   Column           Non-Null Count  Dtype  \n",
            "---  ------           --------------  -----  \n",
            " 0   Crop             19689 non-null  object \n",
            " 1   Crop_Year        19689 non-null  int64  \n",
            " 2   Season           19689 non-null  object \n",
            " 3   State            19689 non-null  object \n",
            " 4   Area             19689 non-null  float64\n",
            " 5   Production       19689 non-null  int64  \n",
            " 6   Annual_Rainfall  19689 non-null  float64\n",
            " 7   Fertilizer       19689 non-null  float64\n",
            " 8   Pesticide        19689 non-null  float64\n",
            " 9   Yield            19689 non-null  float64\n",
            " 10  Avg_Temperature  19689 non-null  float64\n",
            " 11  Max_Temperature  19689 non-null  float64\n",
            " 12  Min_Temperature  19689 non-null  float64\n",
            "dtypes: float64(8), int64(2), object(3)\n",
            "memory usage: 2.0+ MB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.columns = (\n",
        "    df.columns\n",
        "    .str.lower()\n",
        "    .str.strip()\n",
        ")"
      ],
      "metadata": {
        "id": "eh20c7sGaLYd"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df['crop'] = (\n",
        "    df['crop']\n",
        "    .str.upper()\n",
        "    .str.strip()\n",
        ")\n",
        "\n",
        "df['season'] = (\n",
        "    df['season']\n",
        "    .str.upper()\n",
        "    .str.strip()\n",
        ")\n",
        "\n",
        "df['state'] = (\n",
        "    df['state']\n",
        "    .str.upper()\n",
        "    .str.strip()\n",
        ")"
      ],
      "metadata": {
        "id": "pb4X87pTaNJu"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df['yield'].describe()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 335
        },
        "id": "Vx2U01afaO-m",
        "outputId": "44bda5d0-26c1-445d-8e09-6853df6d55dc"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "count    19689.000000\n",
              "mean        79.954011\n",
              "std        878.306194\n",
              "min          0.000000\n",
              "25%          0.600000\n",
              "50%          1.030000\n",
              "75%          2.389000\n",
              "max      21105.000000\n",
              "Name: yield, dtype: float64"
            ],
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>yield</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>19689.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>79.954011</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>878.306194</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>0.600000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>1.030000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>2.389000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>21105.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div><br><label><b>dtype:</b> float64</label>"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(10,5))\n",
        "\n",
        "sns.histplot(df['yield'], bins=100)\n",
        "\n",
        "plt.title(\"Yield Distribution\")\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 476
        },
        "id": "Sj0Lcd3RaRm-",
        "outputId": "6e9c5862-1d89-4de7-e756-6f9708de6fa6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x500 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "upper_limit = df['yield'].quantile(0.99)\n",
        "\n",
        "upper_limit"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oXEPl5P_aTgW",
        "outputId": "9531437b-fc0e-4c0d-ec15-a8e36342354f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "np.float64(104.27367999999956)"
            ]
          },
          "metadata": {},
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df = df[\n",
        "    df['yield'] > 0\n",
        "]\n",
        "\n",
        "df = df[\n",
        "    df['yield'] <= upper_limit\n",
        "]"
      ],
      "metadata": {
        "id": "Aw578rMEaaXV"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df['yield_log'] = np.log1p(\n",
        "    df['yield']\n",
        ")"
      ],
      "metadata": {
        "id": "3BLP6tRqadKF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(10,5))\n",
        "\n",
        "sns.histplot(df['yield_log'], bins=100)\n",
        "\n",
        "plt.title(\"Log Transformed Yield\")\n",
        "\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 480
        },
        "id": "YQvS9hdHaezl",
        "outputId": "3f3cca62-8b00-4489-95d0-98adf2d2fa45"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x500 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df['temp_range'] = (\n",
        "    df['max_temperature']\n",
        "    -\n",
        "    df['min_temperature']\n",
        ")"
      ],
      "metadata": {
        "id": "m7k_RuPVagbe"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df['rainfall_intensity'] = (\n",
        "    df['annual_rainfall'] / 12\n",
        ")"
      ],
      "metadata": {
        "id": "QZ8tbeChaiS-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df['fertilizer_per_area'] = (\n",
        "    df['fertilizer']\n",
        "    /\n",
        "    (df['area'] + 1)\n",
        ")"
      ],
      "metadata": {
        "id": "3BE8JHiFajdu"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df['pesticide_per_area'] = (\n",
        "    df['pesticide']\n",
        "    /\n",
        "    (df['area'] + 1)\n",
        ")"
      ],
      "metadata": {
        "id": "cYbdc-2LamOu"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df['area_log'] = np.log1p(\n",
        "    df['area']\n",
        ")"
      ],
      "metadata": {
        "id": "3apv-ygmankm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df['years_from_2000'] = (\n",
        "    df['crop_year'] - 2000\n",
        ")"
      ],
      "metadata": {
        "id": "PsDCFZMNao_O"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "cat_cols = [\n",
        "    'crop',\n",
        "    'season',\n",
        "    'state'\n",
        "]\n",
        "\n",
        "label_encoders = {}\n",
        "\n",
        "for col in cat_cols:\n",
        "\n",
        "    le = LabelEncoder()\n",
        "\n",
        "    df[col] = le.fit_transform(\n",
        "        df[col]\n",
        "    )\n",
        "\n",
        "    label_encoders[col] = le"
      ],
      "metadata": {
        "id": "7Zie1zYLaqXm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X = df.drop(\n",
        "    columns=[\n",
        "        'yield',\n",
        "        'yield_log',\n",
        "        'production'\n",
        "    ]\n",
        ")\n",
        "\n",
        "y = df['yield_log']"
      ],
      "metadata": {
        "id": "ke-hrF6CarkN"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(\n",
        "    X,\n",
        "    y,\n",
        "    test_size=0.2,\n",
        "    random_state=42\n",
        ")"
      ],
      "metadata": {
        "id": "khwki5VaatG1"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "lr = LinearRegression()\n",
        "\n",
        "lr.fit(X_train, y_train)\n",
        "\n",
        "lr_preds = lr.predict(X_test)"
      ],
      "metadata": {
        "id": "WZyNNLoAaukF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "lr_mae = mean_absolute_error(\n",
        "    y_test,\n",
        "    lr_preds\n",
        ")\n",
        "\n",
        "lr_rmse = np.sqrt(\n",
        "    mean_squared_error(\n",
        "        y_test,\n",
        "        lr_preds\n",
        "    )\n",
        ")\n",
        "\n",
        "lr_r2 = r2_score(\n",
        "    y_test,\n",
        "    lr_preds\n",
        ")\n",
        "\n",
        "print(\"\\n===== Linear Regression =====\")\n",
        "\n",
        "print(\"MAE :\", lr_mae)\n",
        "print(\"RMSE:\", lr_rmse)\n",
        "print(\"R2  :\", lr_r2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JckWpr-jazr2",
        "outputId": "3ad42879-5f9b-413f-bcbe-c6e3bec69c3f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "===== Linear Regression =====\n",
            "MAE : 0.5672166832870797\n",
            "RMSE: 0.7926024339360608\n",
            "R2  : 0.157989005731081\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "rf = RandomForestRegressor(\n",
        "    n_estimators=200,\n",
        "    max_depth=20,\n",
        "    random_state=42,\n",
        "    n_jobs=-1\n",
        ")\n",
        "\n",
        "rf.fit(X_train, y_train)\n",
        "\n",
        "rf_preds = rf.predict(X_test)\n",
        "\n",
        "rf_mae = mean_absolute_error(\n",
        "    y_test,\n",
        "    rf_preds\n",
        ")\n",
        "\n",
        "rf_rmse = np.sqrt(\n",
        "    mean_squared_error(\n",
        "        y_test,\n",
        "        rf_preds\n",
        "    )\n",
        ")\n",
        "\n",
        "rf_r2 = r2_score(\n",
        "    y_test,\n",
        "    rf_preds\n",
        ")\n",
        "\n",
        "print(\"\\n===== Random Forest =====\")\n",
        "\n",
        "print(\"MAE :\", rf_mae)\n",
        "print(\"RMSE:\", rf_rmse)\n",
        "print(\"R2  :\", rf_r2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SEMMgUFma1uN",
        "outputId": "8df52a6c-40d0-4b32-9788-c40c4224c17f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "===== Random Forest =====\n",
            "MAE : 0.14796353524305828\n",
            "RMSE: 0.2538107575013197\n",
            "R2  : 0.9136570248061089\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from xgboost import XGBRegressor\n",
        "\n",
        "xgb = XGBRegressor(\n",
        "    n_estimators=500,\n",
        "    learning_rate=0.05,\n",
        "    max_depth=10,\n",
        "    subsample=0.8,\n",
        "    colsample_bytree=0.8,\n",
        "    random_state=42,\n",
        "    n_jobs=-1\n",
        ")\n",
        "\n",
        "xgb.fit(X_train, y_train)\n",
        "\n",
        "xgb_preds = xgb.predict(X_test)"
      ],
      "metadata": {
        "id": "jxkj0mtqbI4e"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "xgb_mae = mean_absolute_error(\n",
        "    y_test,\n",
        "    xgb_preds\n",
        ")\n",
        "\n",
        "xgb_rmse = np.sqrt(\n",
        "    mean_squared_error(\n",
        "        y_test,\n",
        "        xgb_preds\n",
        "    )\n",
        ")\n",
        "\n",
        "xgb_r2 = r2_score(\n",
        "    y_test,\n",
        "    xgb_preds\n",
        ")\n",
        "\n",
        "print(\"\\n===== XGBoost =====\")\n",
        "\n",
        "print(\"MAE :\", xgb_mae)\n",
        "print(\"RMSE:\", xgb_rmse)\n",
        "print(\"R2  :\", xgb_r2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SwSWF77SbwFN",
        "outputId": "ce0a0fd3-7953-456a-e64b-0269a03b81d3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "===== XGBoost =====\n",
            "MAE : 0.12744668015546914\n",
            "RMSE: 0.21186923381829612\n",
            "R2  : 0.9398351620882156\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "importance_df = pd.DataFrame({\n",
        "\n",
        "    'Feature': X.columns,\n",
        "\n",
        "    'Importance': xgb.feature_importances_\n",
        "})\n",
        "\n",
        "importance_df = importance_df.sort_values(\n",
        "    by='Importance',\n",
        "    ascending=False\n",
        ")\n",
        "\n",
        "importance_df.head(10)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "id": "_1aGGypWb0Ut",
        "outputId": "6140c364-bb49-410b-b6d4-881117a15302"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "               Feature  Importance\n",
              "0                 crop    0.276683\n",
              "2               season    0.243610\n",
              "15            area_log    0.065771\n",
              "3                state    0.048912\n",
              "16     years_from_2000    0.047408\n",
              "9      max_temperature    0.039885\n",
              "10     min_temperature    0.039761\n",
              "11          temp_range    0.038381\n",
              "6           fertilizer    0.037074\n",
              "12  rainfall_intensity    0.032723"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-f92670f6-19fb-493d-8572-0136915ffb2b\" 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>Feature</th>\n",
              "      <th>Importance</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>crop</td>\n",
              "      <td>0.276683</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>season</td>\n",
              "      <td>0.243610</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>area_log</td>\n",
              "      <td>0.065771</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>state</td>\n",
              "      <td>0.048912</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>years_from_2000</td>\n",
              "      <td>0.047408</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>max_temperature</td>\n",
              "      <td>0.039885</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>min_temperature</td>\n",
              "      <td>0.039761</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>temp_range</td>\n",
              "      <td>0.038381</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>fertilizer</td>\n",
              "      <td>0.037074</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>rainfall_intensity</td>\n",
              "      <td>0.032723</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f92670f6-19fb-493d-8572-0136915ffb2b')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-f92670f6-19fb-493d-8572-0136915ffb2b 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-f92670f6-19fb-493d-8572-0136915ffb2b');\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",
              "\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "importance_df",
              "summary": "{\n  \"name\": \"importance_df\",\n  \"rows\": 17,\n  \"fields\": [\n    {\n      \"column\": \"Feature\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 17,\n        \"samples\": [\n          \"crop\",\n          \"season\",\n          \"max_temperature\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Importance\",\n      \"properties\": {\n        \"dtype\": \"float32\",\n        \"num_unique_values\": 17,\n        \"samples\": [\n          0.27668341994285583,\n          0.24360980093479156,\n          0.03988451138138771\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 50
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "top_features = importance_df.head(10)\n",
        "\n",
        "plt.figure(figsize=(10,6))\n",
        "\n",
        "sns.barplot(\n",
        "    x='Importance',\n",
        "    y='Feature',\n",
        "    data=top_features\n",
        ")\n",
        "\n",
        "plt.title(\"Top 10 Feature Importances\")\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 505
        },
        "id": "y-rdScdcb9tW",
        "outputId": "187ac73d-5c1d-4913-99ab-230ed2cb35e2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import shap\n",
        "\n",
        "explainer = shap.TreeExplainer(xgb)\n",
        "\n",
        "shap_values = explainer.shap_values(X_test)"
      ],
      "metadata": {
        "id": "lBiHdKBwcAyW"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "shap.summary_plot(\n",
        "    shap_values,\n",
        "    X_test\n",
        ")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 837
        },
        "id": "kE5jdb7ycT5G",
        "outputId": "0d71938a-85eb-4f61-8baa-6355eb46f68a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x830 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import joblib\n",
        "\n",
        "joblib.dump(\n",
        "    xgb,\n",
        "    'xgboost_model.pkl'\n",
        ")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "et-alffPdw7e",
        "outputId": "ab5464ba-b072-4339-e39c-180857596e2f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['xgboost_model.pkl']"
            ]
          },
          "metadata": {},
          "execution_count": 54
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "joblib.dump(\n",
        "    label_encoders,\n",
        "    'label_encoders.pkl'\n",
        ")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LWjO72TGd-Pu",
        "outputId": "6a03e15b-56c6-4f84-d2cb-70664c911928"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['label_encoders.pkl']"
            ]
          },
          "metadata": {},
          "execution_count": 55
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "020jtgcReFq9"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}