{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/hhboorstein/ds_portfolio/blob/main/DSBA_Ensemble.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Work Visa: Ensemble Methods"
      ],
      "metadata": {
        "id": "5S8NuBE3nvlD"
      },
      "id": "5S8NuBE3nvlD"
    },
    {
      "cell_type": "markdown",
      "source": [
        "The number of applications for US work visas is growing. To assist with the review process, we will develop an ML solution to filter out many candidates. This filtering will be trained on a data set of previous applications and the resulting case statuses.\n",
        "\n",
        "This data has features describing both the applicant and their sponsoring employer:\n",
        "\n",
        "* case_id: unique identifier for each application\n",
        "\n",
        "* continent: employee continent of origin\n",
        "\n",
        "* education_of_employee: level of education\n",
        "\n",
        "* has_job_experience: binary flag\n",
        "\n",
        "* requires_job_training: binary flag\n",
        "\n",
        "* no_of_employees: size of sponsoring employer's company\n",
        "\n",
        "* yr_of_estab: sponsoring company's year of establishment\n",
        "\n",
        "* region_of_employment: applicant's intended region of employment in the US\n",
        "\n",
        "* prevailing_wage: regional average wage for a given domain of labor; useful to keep job market competative without underpaying foreign workers\n",
        "\n",
        "* unit_of_wage: frequency of payment\n",
        "\n",
        "* full_time_position: binary flag; Y: full time\n",
        "\n",
        "* case_status: binary flag"
      ],
      "metadata": {
        "id": "sCbkoxqqnLPK"
      },
      "id": "sCbkoxqqnLPK"
    },
    {
      "cell_type": "markdown",
      "id": "dirty-island",
      "metadata": {
        "id": "dirty-island"
      },
      "source": [
        "## Importing necessary libraries and data"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip uninstall scikit-learn -y\n",
        "\n",
        "!pip install -U scikit-learn"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vSdfBXwUfVm_",
        "outputId": "d0d4ac89-81dc-40a4-e776-7d43aae22378"
      },
      "id": "vSdfBXwUfVm_",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found existing installation: scikit-learn 1.0.2\n",
            "Uninstalling scikit-learn-1.0.2:\n",
            "  Successfully uninstalled scikit-learn-1.0.2\n",
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting scikit-learn\n",
            "  Downloading scikit_learn-1.1.3-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (31.2 MB)\n",
            "\u001b[K     |████████████████████████████████| 31.2 MB 1.1 MB/s \n",
            "\u001b[?25hRequirement already satisfied: numpy>=1.17.3 in /usr/local/lib/python3.8/dist-packages (from scikit-learn) (1.21.6)\n",
            "Requirement already satisfied: joblib>=1.0.0 in /usr/local/lib/python3.8/dist-packages (from scikit-learn) (1.2.0)\n",
            "Requirement already satisfied: scipy>=1.3.2 in /usr/local/lib/python3.8/dist-packages (from scikit-learn) (1.7.3)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.8/dist-packages (from scikit-learn) (3.1.0)\n",
            "Installing collected packages: scikit-learn\n",
            "Successfully installed scikit-learn-1.1.3\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "statewide-still",
      "metadata": {
        "id": "statewide-still"
      },
      "outputs": [],
      "source": [
        "# math and data\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "# plotting\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "sns.set_theme()\n",
        "\n",
        "# model building\n",
        "from sklearn import metrics\n",
        "from sklearn.model_selection import train_test_split, GridSearchCV\n",
        "\n",
        "# models\n",
        "from sklearn import tree\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.ensemble import BaggingClassifier, RandomForestClassifier\n",
        "from sklearn.ensemble import AdaBoostClassifier, GradientBoostingClassifier\n",
        "from sklearn.ensemble import StackingClassifier\n",
        "from xgboost import XGBClassifier"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "visa=pd.read_csv('dataset.csv')"
      ],
      "metadata": {
        "id": "QTUEqr4ReaRG"
      },
      "id": "QTUEqr4ReaRG",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "visa_copy=visa.copy()"
      ],
      "metadata": {
        "id": "1etBGBuhffbJ"
      },
      "id": "1etBGBuhffbJ",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "id": "desperate-infection",
      "metadata": {
        "id": "desperate-infection"
      },
      "source": [
        "## Data Overview\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "visa.sample(10)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 479
        },
        "id": "_9Sk7d9TflmP",
        "outputId": "9bd17233-17a0-4c34-ca0a-03dd47d52c06"
      },
      "id": "_9Sk7d9TflmP",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "         case_id      continent education_of_employee has_job_experience  \\\n",
              "24759  EZYV24760         Europe            Bachelor's                  Y   \n",
              "17290  EZYV17291           Asia              Master's                  Y   \n",
              "11584  EZYV11585           Asia              Master's                  Y   \n",
              "5717    EZYV5718         Europe              Master's                  N   \n",
              "10098  EZYV10099  North America             Doctorate                  Y   \n",
              "18484  EZYV18485           Asia              Master's                  Y   \n",
              "8757    EZYV8758  North America            Bachelor's                  Y   \n",
              "6137    EZYV6138           Asia              Master's                  N   \n",
              "1178    EZYV1179  North America              Master's                  Y   \n",
              "12869  EZYV12870           Asia           High School                  Y   \n",
              "\n",
              "      requires_job_training  no_of_employees  yr_of_estab  \\\n",
              "24759                     N              955         2001   \n",
              "17290                     N             1305         2011   \n",
              "11584                     N              645         2007   \n",
              "5717                      Y             3343         1911   \n",
              "10098                     Y             1307         2001   \n",
              "18484                     N             1851         2007   \n",
              "8757                      N             7778         1968   \n",
              "6137                      N             3406         2010   \n",
              "1178                      N             1409         2001   \n",
              "12869                     N             2981         2006   \n",
              "\n",
              "      region_of_employment  prevailing_wage unit_of_wage full_time_position  \\\n",
              "24759            Northeast        136814.79         Year                  Y   \n",
              "17290              Midwest        221669.11         Year                  N   \n",
              "11584            Northeast         57470.10         Year                  N   \n",
              "5717                  West        147919.62         Year                  Y   \n",
              "10098                South         25838.77         Year                  Y   \n",
              "18484            Northeast         58934.94         Year                  Y   \n",
              "8757             Northeast        115183.30         Year                  Y   \n",
              "6137                 South        161136.84         Year                  N   \n",
              "1178                 South        124682.10         Year                  Y   \n",
              "12869            Northeast        110582.48         Year                  Y   \n",
              "\n",
              "      case_status  \n",
              "24759      Denied  \n",
              "17290      Denied  \n",
              "11584   Certified  \n",
              "5717    Certified  \n",
              "10098   Certified  \n",
              "18484   Certified  \n",
              "8757    Certified  \n",
              "6137    Certified  \n",
              "1178    Certified  \n",
              "12869      Denied  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-42042815-88b4-4777-b11b-fbe6d00a5813\">\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>case_id</th>\n",
              "      <th>continent</th>\n",
              "      <th>education_of_employee</th>\n",
              "      <th>has_job_experience</th>\n",
              "      <th>requires_job_training</th>\n",
              "      <th>no_of_employees</th>\n",
              "      <th>yr_of_estab</th>\n",
              "      <th>region_of_employment</th>\n",
              "      <th>prevailing_wage</th>\n",
              "      <th>unit_of_wage</th>\n",
              "      <th>full_time_position</th>\n",
              "      <th>case_status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>24759</th>\n",
              "      <td>EZYV24760</td>\n",
              "      <td>Europe</td>\n",
              "      <td>Bachelor's</td>\n",
              "      <td>Y</td>\n",
              "      <td>N</td>\n",
              "      <td>955</td>\n",
              "      <td>2001</td>\n",
              "      <td>Northeast</td>\n",
              "      <td>136814.79</td>\n",
              "      <td>Year</td>\n",
              "      <td>Y</td>\n",
              "      <td>Denied</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17290</th>\n",
              "      <td>EZYV17291</td>\n",
              "      <td>Asia</td>\n",
              "      <td>Master's</td>\n",
              "      <td>Y</td>\n",
              "      <td>N</td>\n",
              "      <td>1305</td>\n",
              "      <td>2011</td>\n",
              "      <td>Midwest</td>\n",
              "      <td>221669.11</td>\n",
              "      <td>Year</td>\n",
              "      <td>N</td>\n",
              "      <td>Denied</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11584</th>\n",
              "      <td>EZYV11585</td>\n",
              "      <td>Asia</td>\n",
              "      <td>Master's</td>\n",
              "      <td>Y</td>\n",
              "      <td>N</td>\n",
              "      <td>645</td>\n",
              "      <td>2007</td>\n",
              "      <td>Northeast</td>\n",
              "      <td>57470.10</td>\n",
              "      <td>Year</td>\n",
              "      <td>N</td>\n",
              "      <td>Certified</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5717</th>\n",
              "      <td>EZYV5718</td>\n",
              "      <td>Europe</td>\n",
              "      <td>Master's</td>\n",
              "      <td>N</td>\n",
              "      <td>Y</td>\n",
              "      <td>3343</td>\n",
              "      <td>1911</td>\n",
              "      <td>West</td>\n",
              "      <td>147919.62</td>\n",
              "      <td>Year</td>\n",
              "      <td>Y</td>\n",
              "      <td>Certified</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10098</th>\n",
              "      <td>EZYV10099</td>\n",
              "      <td>North America</td>\n",
              "      <td>Doctorate</td>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "      <td>1307</td>\n",
              "      <td>2001</td>\n",
              "      <td>South</td>\n",
              "      <td>25838.77</td>\n",
              "      <td>Year</td>\n",
              "      <td>Y</td>\n",
              "      <td>Certified</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18484</th>\n",
              "      <td>EZYV18485</td>\n",
              "      <td>Asia</td>\n",
              "      <td>Master's</td>\n",
              "      <td>Y</td>\n",
              "      <td>N</td>\n",
              "      <td>1851</td>\n",
              "      <td>2007</td>\n",
              "      <td>Northeast</td>\n",
              "      <td>58934.94</td>\n",
              "      <td>Year</td>\n",
              "      <td>Y</td>\n",
              "      <td>Certified</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8757</th>\n",
              "      <td>EZYV8758</td>\n",
              "      <td>North America</td>\n",
              "      <td>Bachelor's</td>\n",
              "      <td>Y</td>\n",
              "      <td>N</td>\n",
              "      <td>7778</td>\n",
              "      <td>1968</td>\n",
              "      <td>Northeast</td>\n",
              "      <td>115183.30</td>\n",
              "      <td>Year</td>\n",
              "      <td>Y</td>\n",
              "      <td>Certified</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6137</th>\n",
              "      <td>EZYV6138</td>\n",
              "      <td>Asia</td>\n",
              "      <td>Master's</td>\n",
              "      <td>N</td>\n",
              "      <td>N</td>\n",
              "      <td>3406</td>\n",
              "      <td>2010</td>\n",
              "      <td>South</td>\n",
              "      <td>161136.84</td>\n",
              "      <td>Year</td>\n",
              "      <td>N</td>\n",
              "      <td>Certified</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1178</th>\n",
              "      <td>EZYV1179</td>\n",
              "      <td>North America</td>\n",
              "      <td>Master's</td>\n",
              "      <td>Y</td>\n",
              "      <td>N</td>\n",
              "      <td>1409</td>\n",
              "      <td>2001</td>\n",
              "      <td>South</td>\n",
              "      <td>124682.10</td>\n",
              "      <td>Year</td>\n",
              "      <td>Y</td>\n",
              "      <td>Certified</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12869</th>\n",
              "      <td>EZYV12870</td>\n",
              "      <td>Asia</td>\n",
              "      <td>High School</td>\n",
              "      <td>Y</td>\n",
              "      <td>N</td>\n",
              "      <td>2981</td>\n",
              "      <td>2006</td>\n",
              "      <td>Northeast</td>\n",
              "      <td>110582.48</td>\n",
              "      <td>Year</td>\n",
              "      <td>Y</td>\n",
              "      <td>Denied</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-42042815-88b4-4777-b11b-fbe6d00a5813')\"\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-42042815-88b4-4777-b11b-fbe6d00a5813 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-42042815-88b4-4777-b11b-fbe6d00a5813');\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": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "visa.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "45jNV4adf-ho",
        "outputId": "e864df18-b27c-458c-fe84-bbf7490737d9"
      },
      "id": "45jNV4adf-ho",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(25480, 12)"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "persistent-juice",
      "metadata": {
        "id": "persistent-juice",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d67d2622-64f1-4dfa-91ef-1a35e35af3b1"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 25480 entries, 0 to 25479\n",
            "Data columns (total 12 columns):\n",
            " #   Column                 Non-Null Count  Dtype  \n",
            "---  ------                 --------------  -----  \n",
            " 0   case_id                25480 non-null  object \n",
            " 1   continent              25480 non-null  object \n",
            " 2   education_of_employee  25480 non-null  object \n",
            " 3   has_job_experience     25480 non-null  object \n",
            " 4   requires_job_training  25480 non-null  object \n",
            " 5   no_of_employees        25480 non-null  int64  \n",
            " 6   yr_of_estab            25480 non-null  int64  \n",
            " 7   region_of_employment   25480 non-null  object \n",
            " 8   prevailing_wage        25480 non-null  float64\n",
            " 9   unit_of_wage           25480 non-null  object \n",
            " 10  full_time_position     25480 non-null  object \n",
            " 11  case_status            25480 non-null  object \n",
            "dtypes: float64(1), int64(2), object(9)\n",
            "memory usage: 2.3+ MB\n"
          ]
        }
      ],
      "source": [
        "visa.info()"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "There are twelve features with 25480 records each. While we record no NaN values, there may very well be missing or erroneous data. Most of the columns are object type, which we can convert to categorical during our pre-processing later."
      ],
      "metadata": {
        "id": "8pWt6N2YxUbA"
      },
      "id": "8pWt6N2YxUbA"
    },
    {
      "cell_type": "markdown",
      "id": "seasonal-calibration",
      "metadata": {
        "id": "seasonal-calibration"
      },
      "source": [
        "## Exploratory Data Analysis (EDA)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "visa.describe().T"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        },
        "id": "eT5a4t1cgDME",
        "outputId": "205ff54e-0c61-4df6-dbe6-21c4a7bcfff4"
      },
      "id": "eT5a4t1cgDME",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                   count          mean           std        min       25%  \\\n",
              "no_of_employees  25480.0   5667.043210  22877.928848   -26.0000   1022.00   \n",
              "yr_of_estab      25480.0   1979.409929     42.366929  1800.0000   1976.00   \n",
              "prevailing_wage  25480.0  74455.814592  52815.942327     2.1367  34015.48   \n",
              "\n",
              "                      50%          75%        max  \n",
              "no_of_employees   2109.00    3504.0000  602069.00  \n",
              "yr_of_estab       1997.00    2005.0000    2016.00  \n",
              "prevailing_wage  70308.21  107735.5125  319210.27  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-2d115f60-b1c0-4350-802e-685b15b901b6\">\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>count</th>\n",
              "      <th>mean</th>\n",
              "      <th>std</th>\n",
              "      <th>min</th>\n",
              "      <th>25%</th>\n",
              "      <th>50%</th>\n",
              "      <th>75%</th>\n",
              "      <th>max</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>no_of_employees</th>\n",
              "      <td>25480.0</td>\n",
              "      <td>5667.043210</td>\n",
              "      <td>22877.928848</td>\n",
              "      <td>-26.0000</td>\n",
              "      <td>1022.00</td>\n",
              "      <td>2109.00</td>\n",
              "      <td>3504.0000</td>\n",
              "      <td>602069.00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>yr_of_estab</th>\n",
              "      <td>25480.0</td>\n",
              "      <td>1979.409929</td>\n",
              "      <td>42.366929</td>\n",
              "      <td>1800.0000</td>\n",
              "      <td>1976.00</td>\n",
              "      <td>1997.00</td>\n",
              "      <td>2005.0000</td>\n",
              "      <td>2016.00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>prevailing_wage</th>\n",
              "      <td>25480.0</td>\n",
              "      <td>74455.814592</td>\n",
              "      <td>52815.942327</td>\n",
              "      <td>2.1367</td>\n",
              "      <td>34015.48</td>\n",
              "      <td>70308.21</td>\n",
              "      <td>107735.5125</td>\n",
              "      <td>319210.27</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2d115f60-b1c0-4350-802e-685b15b901b6')\"\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-2d115f60-b1c0-4350-802e-685b15b901b6 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-2d115f60-b1c0-4350-802e-685b15b901b6');\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": 8
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* We see that the minimum number of employees is -26, an impossibility! This indicates that we already have at least one apparent error in our data.\n",
        "\n",
        "* The median number of employees is around 2100, so a midsize company. The largest in our records has over 600,000 employees.\n",
        "\n",
        "* The oldest company we have records for was founded in 1800, but the mean year is around 1979.\n",
        "\n",
        "* The median local wage is \\$70,308.21, and the mean is several thousand higher, indicating some skew toward higher costs of living. There is at least one shockingly low value, as the minimum is just \\$2.14. Perhaps this is hourly wage, though even then \\$2/hr is quite low. This hints that we must diligently explore outliers for potential errors."
      ],
      "metadata": {
        "id": "FVQNxg_GgZxg"
      },
      "id": "FVQNxg_GgZxg"
    },
    {
      "cell_type": "code",
      "source": [
        "visa.describe(include='object').T"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 332
        },
        "id": "_iQsP9-ggHe0",
        "outputId": "4f808fad-59b0-49ef-f149-6948496e5bfb"
      },
      "id": "_iQsP9-ggHe0",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                       count unique         top   freq\n",
              "case_id                25480  25480      EZYV01      1\n",
              "continent              25480      6        Asia  16861\n",
              "education_of_employee  25480      4  Bachelor's  10234\n",
              "has_job_experience     25480      2           Y  14802\n",
              "requires_job_training  25480      2           N  22525\n",
              "region_of_employment   25480      5   Northeast   7195\n",
              "unit_of_wage           25480      4        Year  22962\n",
              "full_time_position     25480      2           Y  22773\n",
              "case_status            25480      2   Certified  17018"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-2119e249-8a60-4df6-beb4-f103eceb2c06\">\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>count</th>\n",
              "      <th>unique</th>\n",
              "      <th>top</th>\n",
              "      <th>freq</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>case_id</th>\n",
              "      <td>25480</td>\n",
              "      <td>25480</td>\n",
              "      <td>EZYV01</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>continent</th>\n",
              "      <td>25480</td>\n",
              "      <td>6</td>\n",
              "      <td>Asia</td>\n",
              "      <td>16861</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>education_of_employee</th>\n",
              "      <td>25480</td>\n",
              "      <td>4</td>\n",
              "      <td>Bachelor's</td>\n",
              "      <td>10234</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>has_job_experience</th>\n",
              "      <td>25480</td>\n",
              "      <td>2</td>\n",
              "      <td>Y</td>\n",
              "      <td>14802</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>requires_job_training</th>\n",
              "      <td>25480</td>\n",
              "      <td>2</td>\n",
              "      <td>N</td>\n",
              "      <td>22525</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>region_of_employment</th>\n",
              "      <td>25480</td>\n",
              "      <td>5</td>\n",
              "      <td>Northeast</td>\n",
              "      <td>7195</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>unit_of_wage</th>\n",
              "      <td>25480</td>\n",
              "      <td>4</td>\n",
              "      <td>Year</td>\n",
              "      <td>22962</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>full_time_position</th>\n",
              "      <td>25480</td>\n",
              "      <td>2</td>\n",
              "      <td>Y</td>\n",
              "      <td>22773</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>case_status</th>\n",
              "      <td>25480</td>\n",
              "      <td>2</td>\n",
              "      <td>Certified</td>\n",
              "      <td>17018</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2119e249-8a60-4df6-beb4-f103eceb2c06')\"\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-2119e249-8a60-4df6-beb4-f103eceb2c06 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-2119e249-8a60-4df6-beb4-f103eceb2c06');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* With 25480 unique case IDs, we can be fairly confident that there are no duplicates in our records.\n",
        "\n",
        "* The data covers 6 continents of origin, with Asia being the most frequent.\n",
        "\n",
        "* The plurality of applicants have a Bachelor's level of education. We will explore the other levels on record shortly.\n",
        "\n",
        "* The majority of applicants do have some job experience, and around 88% of jobs do not require training.\n",
        "\n",
        "* Nearly all wages are recorded as yearly salaries, though the data dictionary indicates that hourly, weekly, and monthly are options too."
      ],
      "metadata": {
        "id": "Fng3k8BkzIhL"
      },
      "id": "Fng3k8BkzIhL"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "right-permit",
      "metadata": {
        "id": "right-permit",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c6c62a2b-f397-40ed-d203-32615509c19f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Values in continent feature\n",
            "Asia             16861\n",
            "Europe            3732\n",
            "North America     3292\n",
            "South America      852\n",
            "Africa             551\n",
            "Oceania            192\n",
            "Name: continent, dtype: int64\n",
            "---------------------------------------------\n",
            "Values in education_of_employee feature\n",
            "Bachelor's     10234\n",
            "Master's        9634\n",
            "High School     3420\n",
            "Doctorate       2192\n",
            "Name: education_of_employee, dtype: int64\n",
            "---------------------------------------------\n",
            "Values in has_job_experience feature\n",
            "Y    14802\n",
            "N    10678\n",
            "Name: has_job_experience, dtype: int64\n",
            "---------------------------------------------\n",
            "Values in requires_job_training feature\n",
            "N    22525\n",
            "Y     2955\n",
            "Name: requires_job_training, dtype: int64\n",
            "---------------------------------------------\n",
            "Values in region_of_employment feature\n",
            "Northeast    7195\n",
            "South        7017\n",
            "West         6586\n",
            "Midwest      4307\n",
            "Island        375\n",
            "Name: region_of_employment, dtype: int64\n",
            "---------------------------------------------\n",
            "Values in unit_of_wage feature\n",
            "Year     22962\n",
            "Hour      2157\n",
            "Week       272\n",
            "Month       89\n",
            "Name: unit_of_wage, dtype: int64\n",
            "---------------------------------------------\n",
            "Values in full_time_position feature\n",
            "Y    22773\n",
            "N     2707\n",
            "Name: full_time_position, dtype: int64\n",
            "---------------------------------------------\n",
            "Values in case_status feature\n",
            "Certified    17018\n",
            "Denied        8462\n",
            "Name: case_status, dtype: int64\n",
            "---------------------------------------------\n"
          ]
        }
      ],
      "source": [
        "for col in visa.drop('case_id',axis=1).select_dtypes('object').columns:\n",
        "  print('Values in',col,'feature')\n",
        "  print(visa[col].value_counts())\n",
        "  print('-'*45)"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* The only continent of origin not represented is (predictably) Antarctica. After Asia, the most common continents are Europe and North America (likely Canada and Mexico).\n",
        "\n",
        "* Other than a Bachelor's degree, many applicants have a Master's. Some also have a high school diploma or a doctorate, but these are less common.\n",
        "\n",
        "* There are five options for region of employment: Northeast, South, West, Midwest, and Island (in decreasing level of frequency). Northeast and South have nearly the same number of records.\n",
        "\n",
        "* The dependent variable in this study is ```case_status```, with outcomes Certified or Denied. About two-thirds of cases are Certified."
      ],
      "metadata": {
        "id": "EjVl26e93LEX"
      },
      "id": "EjVl26e93LEX"
    },
    {
      "cell_type": "code",
      "source": [
        "# quick plotting function\n",
        "def plott(col=None):\n",
        "  '''Quick plot a countplot for categorical\n",
        "  data and a histogram for numeric data.'''\n",
        "  plt.figure(figsize=(8,5))\n",
        "  if col==None:\n",
        "    return\n",
        "  elif visa[col].dtype=='object':\n",
        "    plt.title('Countplot of '+col,fontsize=14)\n",
        "    sns.countplot(data=visa,x=col,\n",
        "                  order=visa[col].value_counts().index.tolist());\n",
        "  else:\n",
        "    plt.title('Histogram of '+col,fontsize=14)\n",
        "    sns.histplot(data=visa,x=col);"
      ],
      "metadata": {
        "id": "OhuPoWJX4zrh"
      },
      "id": "OhuPoWJX4zrh",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plott('continent')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "Iv4GGconqc-R",
        "outputId": "10695d26-f495-4cfb-983d-0a6e0b1ab0da"
      },
      "id": "Iv4GGconqc-R",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We find that Asia is by far the most frequent continent of origin, while Oceania is the least."
      ],
      "metadata": {
        "id": "Q6Auwmhp8-Pf"
      },
      "id": "Q6Auwmhp8-Pf"
    },
    {
      "cell_type": "code",
      "source": [
        "plott('education_of_employee')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "Ii-IjHtVqc6M",
        "outputId": "040546c6-c8f3-4e3d-a2e2-ff66aaf3bb64"
      },
      "id": "Ii-IjHtVqc6M",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Bachelor's and Master's education are almost equally common, with both around 10000 records each."
      ],
      "metadata": {
        "id": "sL111zEg9FO8"
      },
      "id": "sL111zEg9FO8"
    },
    {
      "cell_type": "code",
      "source": [
        "plott('has_job_experience')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "I0XeSsPFqc29",
        "outputId": "ecac6e2d-1442-408a-bf4c-e9cfebfd484c"
      },
      "id": "I0XeSsPFqc29",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "visa['has_job_experience'].value_counts(normalize=True)"
      ],
      "metadata": {
        "id": "Jv_wkqWa92wL",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3c233b56-4963-4fbc-aa6c-0368b0b05b64"
      },
      "id": "Jv_wkqWa92wL",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Y    0.580926\n",
              "N    0.419074\n",
              "Name: has_job_experience, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The majority of applicants have some job experience, but over 10000 (about 40%) have none."
      ],
      "metadata": {
        "id": "rAcy6OtY9fuC"
      },
      "id": "rAcy6OtY9fuC"
    },
    {
      "cell_type": "code",
      "source": [
        "plott('requires_job_training')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "2xRPgoaBqcz0",
        "outputId": "e12d110d-01b3-4b7a-ae08-151a2efbc3c2"
      },
      "id": "2xRPgoaBqcz0",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Unlike the previous feature, which had fairly balanced classes, most applicants do NOT require job training."
      ],
      "metadata": {
        "id": "YTL-Inno9vo4"
      },
      "id": "YTL-Inno9vo4"
    },
    {
      "cell_type": "code",
      "source": [
        "plott('no_of_employees')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "nY8LC0PArFkp",
        "outputId": "7a79b2c5-2e33-4ca6-9d0c-b941d27e6a51"
      },
      "id": "nY8LC0PArFkp",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(16,5))\n",
        "plt.suptitle('Feature: no_of_employees',fontsize=18)\n",
        "\n",
        "plt.subplot(1,2,1)\n",
        "plt.title('Boxplot (with fliers)',fontsize=14)\n",
        "sns.boxplot(data=visa,x='no_of_employees')\n",
        "\n",
        "plt.subplot(1,2,2)\n",
        "plt.title('Boxplot (fliers removed)',fontsize=14)\n",
        "sns.boxplot(data=visa,x='no_of_employees',showfliers=False)\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 375
        },
        "id": "iE_XQm8DrRHY",
        "outputId": "1018bb78-e224-4d3b-9ca3-3ae81d45961e"
      },
      "id": "iE_XQm8DrRHY",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1152x360 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "With such skewed data, the histogram lends little insight. Boxplots give a better idea of data concentration and distribution. Most of our records are concentrated below around 7000, but there are numerous extreme values on the high end.\n",
        "\n",
        "We will not treat these extreme values though. It is entirely reasonable that many records reflect applicants to large companies."
      ],
      "metadata": {
        "id": "u28-xVje_08o"
      },
      "id": "u28-xVje_08o"
    },
    {
      "cell_type": "code",
      "source": [
        "plott('yr_of_estab')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "oBesojUXraRQ",
        "outputId": "f5667e22-af0a-4255-d2a0-0687624f4585"
      },
      "id": "oBesojUXraRQ",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "visa['yr_of_estab'].describe()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sOiEzS5uFAn1",
        "outputId": "dbdab4c0-23f6-49a3-ebd7-a79b06bb6515"
      },
      "id": "sOiEzS5uFAn1",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "count    25480.000000\n",
              "mean      1979.409929\n",
              "std         42.366929\n",
              "min       1800.000000\n",
              "25%       1976.000000\n",
              "50%       1997.000000\n",
              "75%       2005.000000\n",
              "max       2016.000000\n",
              "Name: yr_of_estab, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We find that 75% of companies in our records were founded in 1976 or later. Like the previous feature, we have many extreme values, this time on the lower end. As before, all these values are entirely sensible, so we will not alter them.\n",
        "\n",
        "However, we will convert this feature to 'years since founded', which will flip the distribution from left-skewed to right-skewed but otherwise leave the data unchanged."
      ],
      "metadata": {
        "id": "Lb65bfu8A2nt"
      },
      "id": "Lb65bfu8A2nt"
    },
    {
      "cell_type": "code",
      "source": [
        "plott('region_of_employment')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "8ji8Ea-IGFLJ",
        "outputId": "7f657e9e-67e1-416b-fa69-0c4ea4c07960"
      },
      "id": "8ji8Ea-IGFLJ",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Northeast, South, and West are all common regions. Island is decidedly uncommon, with fewer than 500 records."
      ],
      "metadata": {
        "id": "01h5otEPH_nT"
      },
      "id": "01h5otEPH_nT"
    },
    {
      "cell_type": "code",
      "source": [
        "plott('prevailing_wage')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "4ayouQkMpW7B",
        "outputId": "32cf16b3-91ea-4f4b-e79d-c54de42f776c"
      },
      "id": "4ayouQkMpW7B",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Aside from the spike around \\$0, prevailing wage follows a right-skewed normal distribution. It may well be that the spike is due to other wage units, such as hourly."
      ],
      "metadata": {
        "id": "E9uKFaQSIJKB"
      },
      "id": "E9uKFaQSIJKB"
    },
    {
      "cell_type": "code",
      "source": [
        "plott()\n",
        "plt.title('Prevailing Yearly Wage',fontsize=14)\n",
        "sns.histplot(data=visa.loc[visa['unit_of_wage']=='Year'],\n",
        "             x='prevailing_wage');"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "MIe8av9nR-S3",
        "outputId": "77d7f8d4-295c-46e8-ad0b-ec057d800437"
      },
      "id": "MIe8av9nR-S3",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfwAAAFTCAYAAADcGL4oAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de1hU1cI/8O/MMFwMEUHAwTyZniTCGwJS9pqFeDDjollppF3U9FVLqteOqAWa5Qn1lJmSeCqro2V5VFD0iJra0XpDrRQRrVRSDOQyiHGHmdm/P/yxX8ZhhhmdG+zv53l8npm19uy91szgd/ZtLZkgCAKIiIioU5M7ugFERERkewx8IiIiCWDgExERSQADn4iISAIY+ERERBLAwCciIpIABj5RB7Rt2zaEhoaa/byzef/99xEbG+voZhB1KAx8IiOSk5MRFBSEoKAghISEYNSoUUhLS0NdXZ2jm4axY8di//79N11vDb/++isGDhyI3bt365ULgoApU6Zg+vTpNt3+rfj73/+O0aNH65VduXIFQUFBmDlzpl75d999h6CgIFy6dMmeTSSyOgY+kQnDhw/HkSNHsH//frz00kv4/PPPkZaW1uayGo0G9hrHyt3dHb6+vjddbw133XUXXn75ZSxZsgTl5eVi+aeffopffvkFy5Yts/o2dTodtFrtLa8nMjISly5dwpUrV8Sy77//HiqVCsePH9fbRm5uLgIDA/GnP/3plrdL5EgMfCITXF1d4efnB5VKhbi4OMTFxeHrr78G8H+Hlbdt24bo6GgMHDgQdXV1qK6uxuuvv4777rsPoaGhmDx5Mk6dOgUAqKmpwaBBg3DgwAG97Rw5cgQhISFQq9UAgJUrVyImJgaDBg1CVFQUli9fjsbGRnH59g7Z31jf0tZdu3YhOjoaoaGhmD17NiorK8VlNBoNli1bhoiICERERGDZsmVITU3FlClTjG7n2WefxV133YXXX38dAFBYWIh3330Xqamp8Pf3x9atWzF27FgMHDgQMTEx+OSTT6DT6cTXb9iwAXFxcRgyZAhGjBiBRYsW4Y8//jDoxzfffIPY2FgMGDAA58+f12vDsWPHEBISovejAwDeffddxMXFtdnuoUOHQqlU4vvvvxfLcnNzkZCQgNtuuw2nT5/WK7/33nsBAFlZWZgwYQJCQ0Nx3333Ye7cuSgtLdVb96FDhxATE4OBAwfiqaeewq5duxAUFITLly+Ly/z444+YPHkyBg8ejBEjRiA1NRU1NTVG32cia2DgE1nA3d0dzc3N4vPLly8jOzsb7733HrKysuDq6ooZM2agtLQUGRkZyMzMRHh4OJ555hmUlZXB09MTDz30EHbu3Km33p07d2L48OHiXrmHhweWLVuG3bt3IzU1Fbt378YHH3xwS23//fffsXv3bqxZswYff/wxzpw5g1WrVon1H3/8MbZv344333wTX375JXQ6HbKzs02uUy6X4+2338bRo0exZcsWJCcnIzo6GmPHjsVXX32Fd999F3PnzsXu3bsxf/58/OMf/8Dnn38uvl4mk2HhwoXIzs7G3//+d+Tl5WHp0qV622hsbER6ejqWLFmCXbt2ITAwUK8+IiICvXv3RmZmplim0+mQmZmJxx57rM12d+nSBQMHDkRubq5Ylpubi8jISERERIjltbW1OHXqFCIjIwEAzc3NmDt3Lnbs2IGMjAxcvXoVr7zyiriO4uJivPDCC3jwwQeRlZWFKVOmYMWKFXrb/vnnnzFt2jRERUUhKysLa9aswdmzZ7Fw4UKT7zXRLROIqE3z588XZsyYIT4/efKkMGzYMCEpKUkQBEFYvXq1cM899wjl5eXiMt99950wZMgQob6+Xm9d8fHxwvr16wVBEIT9+/cLgwYNEqqrqwVBEIT6+nohNDRU2LFjh9G2fP7550J0dLT4fOvWrcKQIUPMfr569WphwIABwh9//CGWpaen663z/vvvFzIyMsTnOp1O+Mtf/iJMnjzZaLtabNmyRbj77ruFESNGCNeuXRMEQRBGjhwpbN++XW+5DRs2CA8//LDR9XzzzTdCSEiIoNVqxX70799fOHXqlN5yq1evFh555BHx+YcffiiMGTNGfH7o0CEhJCREqKysNLqtd999V3jooYcEQRCEoqIiISQkRKirqxM2b94sTJs2TWxP//79hZKSkjbXce7cOb36lStX6rVDEAThgw8+EPr37y8UFRUJgiAIr776qrBgwQK9ZQoKCoT+/fsLFRUVRttLdKtcHP2Dg8iZHT58GKGhodBoNNBoNBg1apR4+BoAAgIC0KNHD/H56dOnUV9fj/vuu09vPY2NjSgqKgIAPPDAA3B3d8f+/fsxbtw4HDhwAIIgIDo6Wlx+z549+PTTT3Hp0iXU1dVBq9XqHQq/GYGBgejatav43N/fXzyFUF1djfLycgwcOFCsl8lkGDRokN55bmMee+wxrF69GpMnT4aXlxcqKytRUlKC1NRULFmyRFzuxusc/vd//xfr16/H+fPnUV1dDZ1Oh+bmZpSXlyMgIAAA4OLiguDgYJPbHz9+PN599138+OOPGDp0KLZu3Yro6Gh0797d6GvuvfdefPDBB7h8+TJyc3MxcOBAeHh4YNiwYXj77beh0WiQm5uLO+64Az179gRw/fNt2SOvqqoS11VcXIyePXviwoULeu8hAAwePFjv+enTp3Hx4kX8+9//Fsta3pNLly7Z/NoLki4GPpEJ4eHhWLp0KVxcXODv7w+lUqlX36VLF73nOp0OPXr0wKZNmwzW5enpCQBQKpV4+OGHsXPnTowbNw47duzA6NGj4eHhAQA4ceIEXnnlFcyZMwcjRoyAl5cXDhw4YPRiQXPd2HaZTGbViwxdXFygUCgAQPxxsmTJEqPXGvz++++YOXMmnnjiCcydOxfe3t4oKCjAK6+8onfaxNXVVVyvMT4+PoiKisLWrVtx55134sCBA1i3bp3J14SGhsLV1RVHjx7F0aNHMWzYMADAnXfeidtuuw35+fk4evSoeP6+rq4O06ZNw/Dhw7F8+XL4+Pjg6tWreOqpp/Ta2x6dTofHH38czz77rEFdy48cIltg4BOZ4OHhgTvuuMPs5UNCQlBRUQG5XI7evXsbXS4+Ph6TJ0/GuXPncOTIEb1w+vHHHxEQEIA5c+aIZcXFxTfXATN17doVfn5+OHXqlHh0QhAEnDp1Cn5+fhavr0ePHvD398elS5cwbty4NpfJz89Hc3MzFixYIAb6oUOHbroPLT8cevfuDT8/PwwfPtzk8m5ubhgyZAhyc3ORm5urd1fBsGHD8PXXX+P06dN47rnnAAAXLlzA1atX8fLLL4uf7d69e/XW2bdvX/GizhZ5eXl6z++55x6cO3fOou8VkTXwoj0iKxo+fDiGDh2K2bNn45tvvkFRURF++uknrF69GsePHxeXGzp0KAIDA/E///M/8Pb21jsF0KdPH5SWlmLHjh0oKirC559/3u7Fc9bw9NNP46OPPsK+fftw4cIFvP322wZXvlti7ty5+PDDD/HJJ5/gwoUL+OWXX5CZmYmMjAwAwB133AGdTodPP/0URUVFyM7OxqeffnrT27v//vvh7e2NNWvWYPz48ZDL2//vLTIyEvv27UNFRYXekYiIiAhs2rQJWq1WvGAvMDAQrq6u2LRpE4qKinDo0CG89957euubNGkSLl26hLS0NFy4cAF79+7Fl19+CeD6ERUAeP7555GXl4eUlBQUFBTg4sWLOHjwIFJSUm6670TmYOATWZFMJsP69esRGRmJ119/HQ8//DBeeuklFBYWwt/fX2/ZuLg4nD17Fo888ojeIeuoqChMmzYNy5YtQ3x8PL777jvMnTvX5m2fOnUq4uPjsWDBAkycOBEAMHr0aLi5ud3U+h5//HEsW7YMWVlZSEhIwFNPPYUvv/wSt99+OwDg7rvvxqJFi7BhwwY88sgj2LJlC/7617/edPtlMhkeffRRaDQaPProo2a95t5770Vtba14/r7FsGHDUFtbi7vuuks8p+7j44O0tDTs378fY8eOxZo1a5CcnKy3vl69euH999/HgQMHkJCQgE8++UQ8UtPyPt59993YuHEjfv/9d0yePBkJCQl45513eO6ebE4mWPMkHhF1KuPGjUNYWJjehYrOLDU1FZcuXcKGDRsc3RTRp59+Kh7hadnLJ3IEnsMnIgDXL6I7cuQIIiIioNFo8NVXX+Hnn382uC/eGVVXV+PcuXPIysrSG1vAETZt2oSBAweie/fuOHnyJNLT0zF+/HiGPTkcA5+IAFwfRCczMxPLly+HTqfDn//8Z/zjH/8wuM3MGc2ePRt5eXl47LHH8OCDDzq0LRcvXsS6detQVVWFnj17YtKkSXoXYBI5Cg/pExERSQAv2iMiIpIABj4REZEEMPCJiIgkoNNftHf1ai10OutcpuDr6wm1WppTWEq57wD7L+X+S7nvgLT739H6LpfL0L37bUbrO33g63SC1QK/ZX1SJeW+A+y/lPsv5b4D0u5/Z+o7D+kTERFJAAOfiIhIAhj4REREEsDAJyIikgAGPhERkQQw8ImIiCSAgU9ERCQBDHwiIiIJYOATERFJQKcfaY+cl04mQ12DxqC8i7sL5Jy1mYjIqhj45DB1DRqkbz1pUD57wmB4uikc0CIios6Lh/SJiIgkgIFPREQkATykT1Y7l26t9SiVCtQ0am9pHUREpI+BT1Y7l26t9TQ0aZCxLe+W1kFERPp4SJ+IiEgCuIdPHZqx0wge7i6o5y1/REQiBj7ZXFvn5AFAkN36uo2dRpj56CCD0wIATw0QkXQx8Mnm2jonD1wPZSIisg+ewyciIpIABj4REZEEMPCJiIgkgIFPREQkAXYL/IMHD2LcuHFISEhAfHw89u7dCwAoLCzExIkTERMTg4kTJ+K3334TX2Oqjmyv5er6mkYtLpdVi491MitcXk9ERHZll6v0BUHAX//6V2zatAn9+/fH2bNn8eSTTyI6OhqpqalITExEQkICsrKykJKSgs8++wwATNaR7bW+ul6pdEFz8/X72pMmDUVzs21usyMiItuw2215crkc1dXVAIDq6mr4+/vj6tWrKCgowIYNGwAAsbGxWLp0KSorKyEIgtE6Hx8fezWb2sDb7IiIOh67BL5MJsOqVaswe/ZsdOnSBbW1tVi/fj1KSkoQEBAAheL6QCgKhQL+/v4oKSmBIAhG6xj40mPLwXuIiKTALoGv0WiQkZGB9PR0hIWF4YcffsBLL72E5cuX23zbvr6eVl2fn19Xq67PGTSWVUOpNPwqyGQyvfKWxzeWG1vemuWNzVp8tOO0wbLT4kMsWrfSVXHTn2Fn/OwtIeX+S7nvgLT735n6bpfAP3PmDMrKyhAWFgYACAsLg4eHB9zc3FBaWgqtVguFQgGtVouysjKoVCoIgmC0zhJqdQ10OuuMne7n1xXl5dVWWZczaW7SiufnWxMEQSxvfQ6/dbmx5a1dbq11Nzdpb+oz7Kyfvbmk3H8p9x2Qdv87Wt/lcpnJnVy7XKXfs2dPXLlyBRcuXAAAnD9/Hmq1GnfccQeCg4ORnZ0NAMjOzkZwcDB8fHzg6+trtI6IiIgsY5c9fD8/PyxevBhJSUmQ/f9bupYtWwZvb28sXrwYycnJSE9Ph5eXF9LS0sTXmaojIiIi89ntKv34+HjEx8cblPfr1w9btmxp8zWm6oiIiMh8nC1PQozNHc8r3YmIOj8GvoSYmjueiIg6N46lT0REJAEMfCIiIglg4BMREUkAA5+IiEgCeNEeSYqxMfm7uLtALlhnREYiImfEwCdJMTbT3+wJg+HppnBAi4iI7IOH9ImIiCSAgU9ERCQBPKRPZELL6ISNZdVobvq/c/88509EHQ0Dn8iEltEJW08PDPCcPxF1PDykT0REJAEMfCIiIgngIX0iGL8/nzMJElFnwcAngvH78zmTIBF1FjykT0REJAEMfCIiIglg4BMREUkAA5+IiEgCeNFeB9YyCtyNOAocERHdiIHfgbWMAncjjgJne5xml4g6GrsE/uXLlzFnzhzxeXV1NWpqanD06FEUFhYiOTkZVVVV8Pb2RlpaGvr06QMAJuuIHMnYbXxJk4aiuZk/BIjI+dgl8G+//XZkZWWJz9966y1otdf/U0xNTUViYiISEhKQlZWFlJQUfPbZZ+3WETkjYz8EeNSFiBzN7hftNTU1YefOnZgwYQLUajUKCgoQGxsLAIiNjUVBQQEqKytN1hEREZFl7H4O/8CBAwgICEBISAjy8/MREBAAheL6no9CoYC/vz9KSkogCILROh8fH7O35+vradX2+/l1ter6bkVjWTWUSsOPUOmqaLOdxpaXyWRmlbc8Nnd5a5Y7Yps3lreut3Q9xj6TjqSjt/9WSLnvgLT735n6bvfA37p1KyZMmGC37anVNdDprHPu1M+vK8rLq62yLmtobtLqTdnaurytdhpbXhCEdstbTw9rzvLWLnfENluX3zg9rqXrMfaZdBTO9t23Jyn3HZB2/zta3+VymcmdXLse0i8tLcWxY8cQFxcHAFCpVCgtLRXP52u1WpSVlUGlUpmsIyIiIsvYNfC3b9+OkSNHonv37gAAX19fBAcHIzs7GwCQnZ2N4OBg+Pj4mKwj01puGbvxH2d+IyKSLrse0t++fTsWLVqkV7Z48WIkJycjPT0dXl5eSEtLM6uOjOPMb0REdCO7Bn5OTo5BWb9+/bBly5Y2lzdVR0RERObjWPpEREQSwMAnIiKSAAY+ERGRBDDwiYiIJICBT0REJAEMfCIiIglg4BMREUkAA5+IiEgCGPhEREQSYPfZ8shyOpkMdQ1tzMzGsfGJiMhMDPwOoK5Bg/StJw3KOTY+ERGZi4f0iYiIJICBT0REJAEMfCIiIgngOXwiO1AqFahp1BqUd3F3gVwQHNAiIpIaBj6RHTQ0aZCxLc+gfPaEwfB0UzigRUQkNTykT0REJAEMfCIiIglg4BMREUkAA5+IiEgCGPhEREQSYLfAb2xsRGpqKv7yl78gLi4Or7/+OgCgsLAQEydORExMDCZOnIjffvtNfI2pOiIiIjKf3W7LW7FiBdzc3JCTkwOZTIaKigoAQGpqKhITE5GQkICsrCykpKTgs88+a7euM+IkOUREZCt2Cfza2lpkZmbim2++gUx2Pb169OgBtVqNgoICbNiwAQAQGxuLpUuXorKyEoIgGK3z8fGxR7PtjpPkEBGRrdgl8IuKiuDt7Y01a9YgNzcXt912G5KSkuDu7o6AgAAoFNcHHlEoFPD390dJSQkEQTBaZ0ng+/p6WrUvfn5drbq+1hrLqqFUGn4kMpnMKcpbHjuiPc7wHrSut9b6la4Km36nrKmjtNMWpNx3QNr970x9t0vga7VaFBUV4Z577sH8+fNx8uRJ/Pd//zfee+89m29bra6BTmedoUv9/LqivLzaKutqS3OTFs3NbRzSFwSHlyuVLuJjR7TH0e9B6/5bc/3NTVqbfqesxdbffWcm5b4D0u5/R+u7XC4zuZNrl8BXqVRwcXFBbGwsAGDw4MHo3r073N3dUVpaCq1WC4VCAa1Wi7KyMqhUKgiCYLSOiIiILGOXq/R9fHwQGRmJb7/9FsD1q+/VajX69OmD4OBgZGdnAwCys7MRHBwMHx8f+Pr6Gq0jIiIiy9jtKv0lS5Zg4cKFSEtLg4uLC5YvXw4vLy8sXrwYycnJSE9Ph5eXF9LS0sTXmKojIiIi89kt8Hv37o1//vOfBuX9+vXDli1b2nyNqToiIiIyH0faIyIikgAGPhERkQQw8ImIiCSAgU9ERCQBDHwiIiIJsNtV+kRkSKlUoKZRa1Dexd0FcsE6I0QSEQEMfCKHamjSIGNbnkF50qShaG7mDwEish4GPpETMvZDYPaEwfB0UzigRUTU0fEcPhERkQQw8ImIiCSAgU9ERCQBDHwiIiIJYOATERFJAAOfiIhIAhj4REREEsDAJyIikgAOvOMAOpkMdQ0ag3JB5oDGEBGRJDDwHaCuQYP0rScNymc+OsgBrSEiIilg4NsQ9+SJiMhZMPBtiHvyZG2cXY+IbhYDn6gD4aQ6RHSz7Bb4UVFRcHV1hZubGwBg3rx5GDFiBE6cOIGUlBQ0NjaiV69eWLFiBXx9fQHAZB0RERGZz6635a1evRpZWVnIysrCiBEjoNPp8OqrryIlJQU5OTkIDw/HypUrAcBkHREREVnGoffh5+fnw83NDeHh4QCASZMmYc+ePe3WERERkWXMPqT/73//Gw8//LBB+Z49ezBmzBiz1jFv3jwIgoCwsDC88sorKCkpQWBgoFjv4+MDnU6Hqqoqk3Xe3t7mNhu+vp5mL2sOP7+uZi/bWFYNpdLwLZbJZB2yvOWxI9rjDO9B63pnaE9rSleFRd/Nm2Hr9TszKfcdkHb/O1PfzQ78RYsWtRn4KSkpZgX+pk2boFKp0NTUhLfeegtvvPEGRo8ebVlrb4JaXQOdzjpXL/v5dUV5ebXZyzc3adHc3MZteYLQ4cqVShfxsSPa4+j3oHX/naE9N2pu0lr03bSUpd/9zkTKfQek3f+O1ne5XGZyJ7fdwC8qKgJw/T+alset61xdXc1qiEqlAgC4uroiMTERs2bNwtNPP43i4mJxmcrKSsjlcnh7e0OlUhmtIyIiIsu0G/ijR4+GTCaDIAgGe+Q9evTAiy++2O5G6urqoNVq0bVrVwiCgN27dyM4OBgDBgxAQ0MDjh8/jvDwcGzevFk8WmCqzlGqqht5DzQREXVI7Qb+2bNnAQCTJ0/Gxo0bb2ojarUaL774IrRaLXQ6Hfr164fU1FTI5XIsX74cqamperfeATBZ5yg19U1tDqTDe6CJiMjZmX0O/2bDHgB69+6NzMzMNuuGDh2KnTt3WlznTIyNfsYhdImIyFmYHfhFRUVYtWoVzpw5g7q6Or26Q4cOWbtdHYqx0c84hC4RETkLswN/3rx56N27N+bPnw8PDw9btomIiIiszOzA//XXX/HFF19ALnfoWD1ERER0E8xO74iICBQUFNiyLURERGQjZu/h9+rVC9OnT8fo0aPRo0cPvbqkpCSrN4yIiIisx+zAr6+vx0MPPQSNRoMrV67Ysk1ERERkZWYH/t/+9jdbtoOIiIhsyKLb8ozp3bu3VRpDREREtmF24LceYreFTHZ9ZJkzZ85Yv2VERERkNWYHfssQuy3Ky8uxZs0acb56IiIicl43fVO9n58fFi1ahHfeecea7SEiIiIbuKVRdC5cuID6+nprtYWIiIhsxOxD+omJieI5e+D6bXrnzp3DnDlzbNIwIiIish6zA//xxx/Xe+7h4YG7774bffr0sXabiIiIyMrMDvzx48fbsh1ERERkQ2afw29ubsbq1asxatQoDBw4EKNGjcLq1avR1NRky/YRERGRFZi9h79ixQrk5eVhyZIlCAwMRHFxMdLT01FTU4OFCxfaso1E1A6lUoGaRq1BeRd3F8hbjZ1BRNJlduDv2bMHWVlZ6N69OwCgb9++uOeee5CQkMDAJ3KwhiYNMrblGZTPnjAYnm4KB7SIiJyN2Yf0BSN7CcbKiYiIyHmYHfhjxozBrFmzcPjwYZw/fx7/+c9/MGfOHIwZM8aW7SMiIiIrMPuQ/quvvooPPvgAb7zxBsrKyhAQEIBHHnkEs2bNsmX7iIiIyAra3cP/4YcfsGLFCri6uiIpKQn79u3DyZMnsXfvXjQ1NaGgoMCiDa5ZswZBQUH45ZdfAAAnTpxAfHw8YmJiMHXqVKjVanFZU3VERERkvnYDPyMjAxEREW3WRUZGYt26dWZv7PTp0zhx4gR69eoFANDpdHj11VeRkpKCnJwchIeHY+XKle3WERERkWXaDfwzZ85gxIgRbdYNHz4c+fn5Zm2oqakJb7zxBhYvXiyW5efnw83NTZxxb9KkSdizZ0+7dURERGSZds/h19TUoLm5GQqF4a09Go0GtbW1Zm3ovffeQ3x8PG6//XaxrKSkBIGBgeJzHx8f6HQ6VFVVmazz9vY2a5sA4Ovrafay7blcVg2l0vAtk8lkkihveeyI9jjDe9C63hnaY0650lUBP7+uBuU3w1rr6Yik3HdA2v3vTH1vN/D79u2LI0eOIDo62qDuyJEj6Nu3b7sb+emnn5Cfn4958+bdXCtvgVpdA53OercONjdrDMoEQej05Uqli/jYEe1x9HvQuv/O0B5zy5ubtCgvrzYot5SfX1errKcjknLfAWn3v6P1XS6XmdzJbfeQ/rPPPovU1FTs3bsXOp0OwPXz63v37sXixYvx3HPPtduIY8eO4fz58xg1ahSioqJw5coVTJs2DRcvXkRxcbG4XGVlJeRyOby9vaFSqYzWERERkWXa3cOPi4tDRUUF5s+fj+bmZnh7e6OqqgpKpRJz585FbGxsuxuZMWMGZsyYIT6PiorCunXr8Oc//xlfffUVjh8/jvDwcGzevFm8r3/AgAFoaGhos46IiIgsY9Z9+M899xwef/xx/PTTT+I59NDQUHh63tr5cblcjuXLlyM1NRWNjY3o1asXVqxY0W4dEZmHY+wTUQuzB97x9PQ0erW+pQ4cOCA+Hjp0KHbu3NnmcqbqiKh9HGOfiFqYPbQuERERdVwMfCIiIglg4BMREUkAA5+IiEgCGPhEREQSwMAnIiKSAAY+ERGRBDDwiYiIJICBT0REJAEMfCIiIglg4BMREUmA2WPpE1HnwUl1iKSHgU8kQZxUh0h6eEifiIhIAhj4REREEsDAJyIikgAGPhERkQQw8ImIiCSAgU9ERCQBDHwiIiIJYOATERFJgN0G3pk9ezYuX74MuVyOLl264PXXX0dwcDAKCwuRnJyMqqoqeHt7Iy0tDX369AEAk3VERERkPrvt4aelpWHHjh3IzMzE1KlTsXDhQgBAamoqEhMTkZOTg8TERKSkpIivMVVHRERE5rNb4Hft2lV8XFNTA5lMBrVajYKCAsTGxgIAYmNjUVBQgMrKSpN1REREZBm7jqW/aNEifPvttxAEAR9++CFKSkoQEBAAheL62N0KhQL+/v4oKSmBIAhG63x8fOzZbCIiog7ProH/1ltvAQAyMzOxfPlyJCUl2Xybvr6eVlvX5bJqKJWGb5lMJpNEectjR7THGdDqB2UAABRnSURBVN6D1vXO0B5blCtdFfDz62pQDsBouRRIue+AtPvfmfrukNnyxo0bh5SUFPTs2ROlpaXQarVQKBTQarUoKyuDSqWCIAhG6yyhVtdAp7PedJ/NzRqDMkEQOn25UukiPnZEexz9HrTuvzO0x1blzU1alJdXG5T7+XVts1wKpNx3QNr972h9l8tlJndy7XIOv7a2FiUlJeLzAwcOoFu3bvD19UVwcDCys7MBANnZ2QgODoaPj4/JOiIiIrKMXfbw6+vrkZSUhPr6esjlcnTr1g3r1q2DTCbD4sWLkZycjPT0dHh5eSEtLU18nak6IiIiMp9dAr9Hjx746quv2qzr168ftmzZYnEdERERmc8h5/CJyDkplQrUNGoNy6sbHdAaIrImBj4RiRqaNMjYlmdQnvTkULjJHNAgIrIajqVPREQkAQx8IiIiCWDgExERSQADn4iISAIY+ERERBLAwCciIpIABj4REZEEMPCJiIgkgIFPREQkARxpj4jaJZfJUNNoOJ1uF3cXyAXrTT9NRLbDwCeidtU3arBu60mD8tkTBsPTTeGAFhGRpXhIn4iISAIY+ERERBLAwCciIpIABj4REZEEMPCJiIgkgIFPREQkAQx8IiIiCWDgExERSQADn4iISALsEvhXr17F888/j5iYGMTFxeGFF15AZWUlAODEiROIj49HTEwMpk6dCrVaLb7OVB0RERGZzy6BL5PJMH36dOTk5GDnzp3o3bs3Vq5cCZ1Oh1dffRUpKSnIyclBeHg4Vq5cCQAm64iIiMgydgl8b29vREZGis+HDBmC4uJi5Ofnw83NDeHh4QCASZMmYc+ePQBgso6IiIgsY/fJc3Q6Hb744gtERUWhpKQEgYGBYp2Pjw90Oh2qqqpM1nl7e5u9PV9fT6u1/XJZNZRKw7dMJpNJorzlsSPa4wzvQet6Z2iPPcsBtFmudFXAz69rm8t3JlLooylS7n9n6rvdA3/p0qXo0qULJk+ejH379tl8e2p1DXQ6603f2dxsOEWoIAidvlypdBEfO6I9jn4PWvffGdpj73Kg7e8+BKDwcpVBcWeaNtfPryvKy6sd3QyHkXL/O1rf5XKZyZ1cuwZ+WloaLl68iHXr1kEul0OlUqG4uFisr6yshFwuh7e3t8k6InIODU0aZGzLMyjntLlEzsdut+W98847yM/Px9q1a+Hq6goAGDBgABoaGnD8+HEAwObNmzFmzJh264iIiMgydtnD//XXX5GRkYE+ffpg0qRJAIDbb78da9euxfLly5GamorGxkb06tULK1asAADI5XKjdURERGQZuwT+XXfdhZ9//rnNuqFDh2Lnzp0W1xEREZH5ONIeERGRBDDwiYiIJICBT0REJAEMfCIiIglg4BMREUkAA5+IiEgCGPhEREQSwMAnIiKSAAY+ERGRBDDwiYiIJMDu0+MSUeenVCpQ06g1KO9M0+YSdTQMfCKyOk6bS+R8eEifiIhIAhj4REREEsDAJyIikgCewyciu+HFfESOw8AnIrvhxXxEjsND+kRERBLAwCciIpIABj4REZEEMPCJiIgkwC6Bn5aWhqioKAQFBeGXX34RywsLCzFx4kTExMRg4sSJ+O2338yqIyIiIsvYJfBHjRqFTZs2oVevXnrlqampSExMRE5ODhITE5GSkmJWHREREVnGLoEfHh4OlUqlV6ZWq1FQUIDY2FgAQGxsLAoKClBZWWmyjoiIiCznsPvwS0pKEBAQAIXi+r23CoUC/v7+KCkpgSAIRut8fHws2o6vr6fV2ny5rBpKpeFbJpPJJFHe8tgR7XGG96B1vTO0x57lAGy6fqWrAn5+XdvcrjNw5rbZg5T735n63ukH3lGra6DTWW8Er+ZmjUGZIAidvlypdBEfO6I9jn4PWvffGdpj73LAtt/95iYtysur29yuo/n5dXXattmDlPvf0foul8tM7uQ6LPBVKhVKS0uh1WqhUCig1WpRVlYGlUoFQRCM1hEREZHlHHZbnq+vL4KDg5GdnQ0AyM7ORnBwMHx8fEzWERERkeXssof/5ptvYu/evaioqMBzzz0Hb29v7Nq1C4sXL0ZycjLS09Ph5eWFtLQ08TWm6oiIiMgydgn81157Da+99ppBeb9+/bBly5Y2X2OqjoiIiCzT6S/aIyLnx2lziWyPgU9EDmds2tykSUPR3MwfAkTWwMAnIqdl7IfA7AmD4emmcECLiDouBj4RdRo6mQx1DYb3+fOIABEDn4g6kboGDdK3njQot/SIQOsfDo1l1Whuun5agT8cqCNj4BMR3aD1D4fWoyzyVAJ1ZA4beIeIiIjsh3v4RNThGLuNT5BZtrylh+iNrcfD3QX1vHaAnBwDn4g6HGNX7898dJBFyxu77c/YDwdT2+XdBOTsGPhEJFmW/nAg6sgY+ERENmLJqQTeUki2xsAnIrIRSwYOstYthUTGMPCJiOysrT1/Y9cNGMMjAmQpBj4RkZ21tedv7LoBU3ckfGDlQYZa4w+HzoeBT0TkxKx1YaGxYLfWDwdyfgx8IqJOxNIjArwjQToY+EREnYitbzXkKYCOi4FPREQGWo4UtJ48COApgI6MgU9ERAZajhS0njwI4CmAjoyBT0REt8zSeQasVc5TCeZj4BMR0S2zdJ4Ba5XzVIL5nH563MLCQkycOBExMTGYOHEifvvtN0c3iYiIqMNx+sBPTU1FYmIicnJykJiYiJSUFEc3iYiInETLqYQb/+lkFg5dKAFOfUhfrVajoKAAGzZsAADExsZi6dKlqKyshI+Pj1nrkMut96ErBKB7VzeDcheFvNOXuyhdoGlWOKw9jn4PWvffGdrD8s733XeGvrZV7uzffY1Why/2njUofy5uAJo0huf23d0UaGjjWgN3N0Wb1wIYyxCdTGbReixd/ma0l3cyQXDeqx3y8/Mxf/587Nq1SywbO3YsVqxYgZCQEAe2jIiIqGNx+kP6REREdOucOvBVKhVKS0uh1V4/DKLValFWVgaVSuXglhEREXUsTh34vr6+CA4ORnZ2NgAgOzsbwcHBZp+/JyIiouuc+hw+AJw/fx7Jycn4448/4OXlhbS0NPTt29fRzSIiIupQnD7wiYiI6NY59SF9IiIisg4GPhERkQQw8ImIiCSAgU9ERCQBDHwzdJYJfKKiojBmzBgkJCQgISEBhw8fBgCcOHEC8fHxiImJwdSpU6FWq8XX2KLOHtLS0hAVFYWgoCD88ssvYrmpz9LedbZkrP/GvgNA5/keXL16Fc8//zxiYmIQFxeHF154AZWVlQ7po7P1PygoCHFxceLn//PPP4uvO3DgAMaMGYPRo0fjpZdeQn19vU3rbGX27NmIj4/HuHHjkJiYiDNnzgCQzt++SQK1a8qUKUJmZqYgCIKQmZkpTJkyxcEtujkPPfSQ8PPPP+uVabVaITo6Wjh27JggCIKwdu1aITk52WZ19nLs2DGhuLjYoM+mPkt719mSsf639R0QhM71Pbh69arw/fffi8/ffvttYcGCBXbvo7P1XxAEoX///kJNTY3Ba2pqaoThw4cLhYWFgiAIwsKFC4X333/fZnW29Mcff4iP9+3bJ4wbN04QBOn87ZvCwG9HRUWFEBYWJmg0GkEQBEGj0QhhYWGCWq12cMss19Z/9idPnhQeeeQR8blarRaGDBliszp7a91nU5+lvesc0f+2nrfozN+DPXv2CM8884zd++hs/RcE44G/e/duYcaMGeLzvLw8YezYsTars5ft27cL48ePl+TfflucerY8Z1BSUoKAgAAoFNdni1IoFPD390dJSUmHHPFv3rx5EAQBYWFheOWVV1BSUoLAwECx3sfHBzqdDlVVVTap8/b2tk9H22DqsxQEwa51jvzu3Pgd8PLy6rTfA51Ohy+++AJRUVF276Oz9b/FlClToNVq8cADD+DFF1+Eq6urQVsDAwNRUlICADaps7VFixbh22+/hSAI+PDDD/m3///xHL6EbNq0CTt27MDWrVshCALeeOMNRzeJ7Exq34GlS5eiS5cumDx5sqOb4hA39v/QoUPYtm0bNm3ahHPnzmHt2rUObqFtvPXWWzh06BBefvllLF++3NHNcRoM/HZ0pgl8Wtrs6uqKxMRE/Pjjj1CpVCguLhaXqayshFwuh7e3t03qHMnUZ2nvOkdp6zvQUt7ZvgdpaWm4ePEiVq1aBblcbvc+Olv/gf/7/D09PfH4448b/fyLi4vFZW1RZy/jxo1Dbm4uevbsKfm/fYCB367OMoFPXV0dqqurAQCCIGD37t0IDg7GgAED0NDQgOPHjwMANm/ejDFjxgCATeocydRnae86RzD2HQBs81k78nvwzjvvID8/H2vXroWrq6tD+uhs/b927RoaGhoAABqNBjk5OeLnP2LECJw6dUq8knzz5s14+OGHbVZnK7W1tXqnDQ4cOIBu3bpJ/m+/BcfSN0NnmMCnqKgIL774IrRaLXQ6Hfr164fXXnsN/v7++PHHH5GamorGxkb06tULK1asQI8ePQDAJnX28Oabb2Lv3r2oqKhA9+7d4e3tjV27dpn8LO1dZ+/+r1u3zuh3ALDNZ+2I78Gvv/6K2NhY9OnTB+7u7gCA22+/HWvXrrV7H52p/9OnT0dKSgpkMhk0Gg1CQ0OxcOFC3HbbbQCA/fv3Y8WKFdDpdAgODsbbb7+NLl262KzOFioqKjB79mzU19dDLpejW7dumD9/PkJCQiTzt28KA5+IiEgCeEifiIhIAhj4REREEsDAJyIikgAGPhERkQQw8ImIiCSAgU9EbVq3bh0WLVoEALh8+TKCgoKg0WgAANOnT8f27dsd2TwishBvyyOidl2+fBmjRo3C6dOn4eLCKTiIOiLu4RN1YC173ERE7WHgEzmhqKgoZGRkYOzYsYiIiMCCBQvQ2NiI3NxcPPDAA1i/fj3uv/9+LFiwADqdDuvXr0d0dDQiIyORlJSEqqoqANcPvW/cuFFv3fHx8di7dy+A6yPyjRw5EkOHDsWjjz4qDgMLAO+//z7mzZvXZvumTJmCLVu2AAC2bduGJ598EmlpaYiIiEBUVBS++eYbcdmioiI89dRTCA0NxbPPPoslS5YYXW+L+fPn4+OPPwYAlJaWIigoCJs2bQIAXLp0CcOGDYNOp8O1a9cwc+ZM3HvvvYiIiMDMmTNx5coVs7d94sQJTJo0CeHh4YiPj0dubq7pD4aoA2PgEzmpnTt34qOPPsK+fftQWFiI9PR0ANeHD7127RoOHjyIpUuX4p///Cf279+PjRs34vDhw+jWrZs4C15sbKw4njcAnDt3DsXFxXjwwQcBAAMHDkRmZiaOHj2K2NhYJCUlobGx0eK25uXl4c4778T333+P6dOnY9GiRWg5Wzhv3jwMGjQIubm5eOGFF5CVldXu+iIiInD06FEAwNGjR9G7d28cO3ZMfB4WFga5XA6dTodHH30UBw8exMGDB+Hm5qY3A6CpbZeWlmLmzJmYNWsWjh49ivnz52Pu3LmorKy0uP9EHQEDn8hJPfXUU1CpVPD29sasWbOwa9cuAIBcLsfcuXPh6uoKd3d3bN68GS+//DJ69uwJV1dXvPDCC8jJyYFGo0F0dDTOnj2L33//HcD1HxGjR48WJ1RJSEhA9+7d4eLigqlTp6KpqQmFhYUWtzUwMBBPPPEEFAoFxo8fj/LyclRUVKC4uBinTp0S2xseHq43N7sxw4YNww8//ACdTodjx45h+vTp4sxux44dw7BhwwAA3bt3R0xMDDw8PODp6YlZs2aJPwza23ZWVhYeeOABjBw5EnK5HPfffz8GDBigd3SCqDPh1TdETqr1VJqBgYEoKysDcD3k3NzcxLri4mLMmTNHnAIVuP6jQK1WIyAgACNHjsSuXbswY8YMZGdn48033xSX++ijj/Cvf/0LZWVlkMlkqKmpwdWrVy1ua+sJYTw8PABcn53v6tWr6Natm1jW0q/WM5q15U9/+hM8PDxw5swZ/PDDD5gzZw7+9a9/4cKFCzh27BimTJkCAKivr8ff/vY3HD58GNeuXQNwfca0lulITW27uLgYe/bswcGDB8V6jUaDyMhIi/tP1BEw8ImcVOtQLC4uFme1k8lkesv17NkTy5YtQ1hYWJvriY2NxZo1axAREYHGxkYx0I4fP44PP/wQn3zyCe666y7I5XJERETAmjfu+Pn54dq1a6ivrxeDt72wbxEREYGcnBw0NzcjICAAERERyMzMxLVr18RpXT/++GMUFhbiq6++gp+fH86cOYNx48ZBEIR2t61SqZCQkKD3A4ioM+MhfSIn9fnnn+PKlSuoqqrCunXrMHbs2DaXe/LJJ7Fq1SrxsH1lZSX2798v1o8cORLFxcVYvXo1xo4dKx4JqK2thUKhgI+PDzQaDdasWYOamhqr9qFXr14YMGAA3n//fTQ1NeGnn37S26M2ZdiwYdi4cSPCw8MBAJGRkdi4cSPCwsKgUCjEPri5ucHLywtVVVVYs2aN2duOj4/HwYMHcfjwYWi1WvGiyNYX/RF1Jgx8IicVGxuLqVOnIjo6Gn/6058wa9asNpd7+umnERUVhalTpyI0NBRPPPEE8vLyxHpXV1eMHj0a3333HWJjY8Xy//qv/8KIESMQExODqKgouLm56Z1GsJaVK1fixIkTiIyMxKpVqzB27FjxGgJTIiIiUFtbi4iICABAWFgYGhoaxB8AAPDMM8+gsbER9957LyZOnIgRI0aYvW2VSoX09HRkZGTgvvvuw8iRI/HRRx9Bp9NZsfdEzoMD7xA5oaioKLz55psYPny4o5tidS+99BL69u2LuXPnSmrbRI7GPXwisqm8vDxcunQJOp0O//nPf/D1118jOjq602+byNnwoj0isqmKigq8+OKLqKqqQs+ePbF48WLcc8892LFjB1JTUw2WDwwMFG9BtNW2iaSIh/SJiIgkgIf0iYiIJICBT0REJAEMfCIiIglg4BMREUkAA5+IiEgCGPhEREQS8P8A3g60G2WOEsEAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Indeed, the distribution looks far less erratic when just focused on yearly prevailing wage."
      ],
      "metadata": {
        "id": "NDlJOKkgSjhy"
      },
      "id": "NDlJOKkgSjhy"
    },
    {
      "cell_type": "code",
      "source": [
        "plott('unit_of_wage')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "fX4y3E7WGI-l",
        "outputId": "cff43504-1993-48a8-acf0-40032afa0153"
      },
      "id": "fX4y3E7WGI-l",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Most of our records consider cases with yearly wages. Weekly and montly are vastly less common, as was shown earlier in the value counts table."
      ],
      "metadata": {
        "id": "CKdkEZRiU4qO"
      },
      "id": "CKdkEZRiU4qO"
    },
    {
      "cell_type": "code",
      "source": [
        "plott('full_time_position')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "5bFcm8W7GLSA",
        "outputId": "f0cb169b-0fd5-4d07-bc55-6bfe47b69a2b"
      },
      "id": "5bFcm8W7GLSA",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Most positions are full time."
      ],
      "metadata": {
        "id": "TdOu9YRmVcuV"
      },
      "id": "TdOu9YRmVcuV"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "mechanical-interference",
      "metadata": {
        "id": "mechanical-interference",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "outputId": "0c51cac3-1f70-40ff-b8db-26e8547f3e93"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "plott('case_status')"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Only one-thirds of cases are Denied."
      ],
      "metadata": {
        "id": "sHqQ22OjWR_n"
      },
      "id": "sHqQ22OjWR_n"
    },
    {
      "cell_type": "code",
      "source": [
        "visa['case_status'].value_counts(normalize=True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zWoyNkQnVm6t",
        "outputId": "fe9bb334-0813-4676-9445-4bb2fe1ba886"
      },
      "id": "zWoyNkQnVm6t",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Certified    0.667896\n",
              "Denied       0.332104\n",
              "Name: case_status, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let's dig deeper into prevailing wage."
      ],
      "metadata": {
        "id": "RRx3brEwVsQm"
      },
      "id": "RRx3brEwVsQm"
    },
    {
      "cell_type": "code",
      "source": [
        "visa.groupby(by='unit_of_wage')['prevailing_wage'].describe().T"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "89uvnFlrQ735",
        "outputId": "3c7aadd0-455f-4e2c-dc37-1543c3454b4f"
      },
      "id": "89uvnFlrQ735",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "unit_of_wage         Hour          Month           Week           Year\n",
              "count         2157.000000      89.000000     272.000000   22962.000000\n",
              "mean           414.570513   87592.864045   85606.820515   81228.077133\n",
              "std            275.015000   59525.124924   44802.704810   49951.473223\n",
              "min              2.136700    1599.280000    2183.230000     100.000000\n",
              "25%            152.700300   44986.240000   51408.277500   43715.955000\n",
              "50%            372.652300   81826.010000   85075.820000   76174.500000\n",
              "75%            637.311100  121629.600000  111331.910000  111341.960000\n",
              "max            999.919500  264362.950000  280175.950000  319210.270000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-3e5e6e2e-cd1f-4d12-8016-af6be8d989df\">\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>unit_of_wage</th>\n",
              "      <th>Hour</th>\n",
              "      <th>Month</th>\n",
              "      <th>Week</th>\n",
              "      <th>Year</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>2157.000000</td>\n",
              "      <td>89.000000</td>\n",
              "      <td>272.000000</td>\n",
              "      <td>22962.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>414.570513</td>\n",
              "      <td>87592.864045</td>\n",
              "      <td>85606.820515</td>\n",
              "      <td>81228.077133</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>275.015000</td>\n",
              "      <td>59525.124924</td>\n",
              "      <td>44802.704810</td>\n",
              "      <td>49951.473223</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>2.136700</td>\n",
              "      <td>1599.280000</td>\n",
              "      <td>2183.230000</td>\n",
              "      <td>100.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>152.700300</td>\n",
              "      <td>44986.240000</td>\n",
              "      <td>51408.277500</td>\n",
              "      <td>43715.955000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>372.652300</td>\n",
              "      <td>81826.010000</td>\n",
              "      <td>85075.820000</td>\n",
              "      <td>76174.500000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>637.311100</td>\n",
              "      <td>121629.600000</td>\n",
              "      <td>111331.910000</td>\n",
              "      <td>111341.960000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>999.919500</td>\n",
              "      <td>264362.950000</td>\n",
              "      <td>280175.950000</td>\n",
              "      <td>319210.270000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-3e5e6e2e-cd1f-4d12-8016-af6be8d989df')\"\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-3e5e6e2e-cd1f-4d12-8016-af6be8d989df 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-3e5e6e2e-cd1f-4d12-8016-af6be8d989df');\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": 28
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "I am skeptical about the records tagged with 'Week' for ```unit_of_wage```. According to our records, 75% of jobs with a weekly wage pay at least \\$51,000 per WEEK! As we have limited visibility into the source of these data, we will preserve these records, but ideally I would like to do further research to justify these entries."
      ],
      "metadata": {
        "id": "4L2xsRJsX6hf"
      },
      "id": "4L2xsRJsX6hf"
    },
    {
      "cell_type": "code",
      "source": [
        "plott()\n",
        "plt.title('Prevailing Wage by Unit',fontsize=14)\n",
        "sns.boxplot(data=visa,\n",
        "             x='prevailing_wage',\n",
        "             y='unit_of_wage');"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "9NV89sp8S46m",
        "outputId": "a08cb3fc-1045-41a2-9605-fb56cf871bf6"
      },
      "id": "9NV89sp8S46m",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The boxplot above is further evidence that weekly&mdash;and even monthly&mdash;data looks erroneous. Many of the weekly wages would result in a yearly earning well over a million USD! Perhaps this is personal bias, but it seems far fetched to have this many exceptionally high-paying jobs in this data set.\n",
        "\n",
        "Without any evidence to the contrary, however, we will leave these records intact."
      ],
      "metadata": {
        "id": "qzBqzFuYb08e"
      },
      "id": "qzBqzFuYb08e"
    },
    {
      "cell_type": "code",
      "source": [
        "plott()\n",
        "plt.title('Prevailing Hourly Wage',fontsize=14)\n",
        "sns.boxplot(data=visa.loc[visa['unit_of_wage']=='Hour'],\n",
        "            x='prevailing_wage');"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "HA_jRjSyTXN3",
        "outputId": "f27bc85c-6482-454f-ccf6-f3523a9d9507"
      },
      "id": "HA_jRjSyTXN3",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcwAAAFTCAYAAAC06zwQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAeZElEQVR4nO3de3RNZ8LH8V9OLhLjEpeIuJQyRItUxElc2qaNaAYR2tUy1apZqNSlqNGF6WtpUW2M6Whpiw4z01E1Leraq7JURiWUUF1q6lKJBrmQlITIydnvH5YzTkvyRG4S389a/sjeJ/t5zhPH195nJ/GwLMsSAAAolq2qJwAAQHVAMAEAMEAwAQAwQDABADBAMAEAMEAwAQAwQDBx21m7dq1CQ0ONP67pbrfnC9wsgokKMW3aNAUHBys4OFgdO3ZU7969lZCQoPz8/Kqemvr166ctW7bc9P7ycvLkSQUHB+vbb7/91b74+HhNmzatwudQHlatWqUuXbro8uXLrm2XL1/WPffco9jYWLfHnjhxQsHBwfr6668re5pAmRFMVJiePXsqMTFRW7Zs0aRJk7Ry5UolJCRc97EOh0OV9TM0fH191ahRo5veX5MUFhaW+RgRERG6ePGiDhw44Np24MAB1a1bVz/++KPOnj3r2p6UlCQfHx917dq1zOMClY1gosL4+PgoICBAQUFBGjBggAYMGKAvv/xSkrRw4ULFxsZq7dq1io6OVufOnZWfn6/z589rxowZ6tGjh0JDQ/Xkk0+6zsAuXLigkJAQbd261W2cxMREdezYUdnZ2ZKk+fPnKyYmRiEhIYqKitK8efNUUFDgenxJlyB/uf/qXDdv3qzo6GiFhoZq7NixbiFwOByaO3eu7Ha77Ha75s6dq5kzZ2rYsGFlX0hdOWN7+eWX1bNnT3Xu3FmDBw/Wnj17XPuTkpIUHBzsNqdfnsFefcz27dv16KOPqlOnTkpMTHQb5+TJk+rQocOvzno/+OADRUREuJ1FXnXnnXeqSZMmSkpKcm3btWuXunfvrk6dOrltT0pKUpcuXVSrVi199dVXGjp0qOx2u8LDwzVy5EgdPXrU7dj79+/Xww8/rM6dO2vQoEHavn27goOD3Y555MgRjR49WqGhoerRo4cmT56szMzM0iwvYIRgotL4+vq6ndGcPHlSmzZt0uuvv67169fLx8dHo0eP1pkzZ7RkyRKtW7dO3bp10/Dhw5WRkaE6derowQcf1MaNG92Ou3HjRvXs2dN1Vujn56e5c+fq448/1syZM/Xxxx/r7bffLtPcf/rpJ3388cdatGiRli9frkOHDmnBggWu/cuXL9dHH32kOXPm6N///recTqc2bdpUpjGvNW/ePH3yySeaO3eu1q1bp/bt2+vpp59WRkZGqY81f/58TZo0SZ988onuuecet30tWrRQr169tGbNGrfta9as0cCBA+Xj43PdY0ZERPwqjBERETfcLkkXL17U8OHD9eGHH+rdd99VnTp19Mwzz7iinJeXp/j4eLVp00Zr167V888/r3nz5rmNm5GRoSeeeELt2rXT6tWr9fe//135+fkaO3asnE5nqdcGKA7BRKU4cOCANm7cqB49eri2FRYWat68eerYsaPat2+vPXv26Pvvv9cbb7yhkJAQtWrVSpMmTVLLli21fv16SVJcXJy2bt2qCxcuSJIuXbqkL774QnFxca7jjhs3TmFhYWrRooUiIyMVHx+vzZs3l2n+DodDr776qjp06KDQ0FANHjzY7X24d999V08//bRiYmLUpk0bvfDCC2rcuLHRsZ988kmFhoa6/dmxY4drf35+vlatWqUpU6bogQceUNu2bfXSSy+pUaNGeu+990r9XMaPH697771XLVu2VMOGDX+1/7HHHtPmzZtdZ+VHjx5VSkqKHn300RseMyIiQikpKbp8+bIKCgqUkpKi8PBw2e12VzCPHj2qzMxMde/eXZIUExOjmJgYtW7dWh06dNArr7yikydPui7tbty4UU6nUy+//LLatWunXr166ZlnnnEb9/3331eHDh30/PPPq23bturQoYMSEhJ04MABHTx4sNRrAxTHq6ongJprx44dCg0NlcPhkMPhUO/evTVjxgzX/sDAQLeofPfdd7p48aJbVCWpoKBAaWlpkqT7779fvr6+2rJliwYNGqStW7fKsixFR0e7Hv/pp5/qn//8p1JTU5Wfn6+ioqIyn200a9ZMdevWdX3cpEkT1yXg8+fPKzMzU507d3bt9/DwUEhIiE6fPl3isf/yl7+oXbt2btuuXafU1FQVFha6ve/n6empLl26/OoSpolOnToVu793796aNWuWPv/8cw0YMECrV69WSEiI2rdvf8PP6d69uwoKCrRv3z5ZlqWGDRuqVatWCggIUFpamjIzM5WUlCQ/Pz/XWW1qaqpef/117d+/X2fPnpVlWXI6nTp16pQk6dixY2rXrp18fX1d4/zyjPi7777Tnj17rnuJPTU1VSEhIcbrApSEYKLCdOvWTbNnz5aXl5eaNGkib29vt/21a9d2+9jpdKpx48bXPWuqU6eOJMnb21t9+/bVxo0bNWjQIG3YsEF9+vSRn5+fJCklJUWTJ0/WuHHjdN9996levXraunXrDW82MvXLuXt4eJTbTUqBgYFq1aqV27arz6ckHh4ekiSb7dcXixwOx3U/p6Rje3t7a+DAgVqzZo369u2rDRs2aMKECcV+TsuWLdW8eXMlJyfLsizZ7XZJV77GHTt2VHJyspKTkxUWFuZay/j4eDVt2lSzZs1SYGCgPD091b9//1LdiOR0OhUZGampU6f+at/tcuMWKg+XZFFh/Pz81KpVKzVv3vxXwbmejh07KisrSzabTa1atXL7c+0/fnFxcfr666915MgRJSYmul2O3bt3rwIDAzVu3DiFhISodevWSk9Pr5Dnd1XdunUVEBDgdqOMZVnX/XaRm3HHHXfI29tbe/fudW0rKipSSkqK2rZtK0muS6vXvqd56NChmx7zscceU1JSklauXKm8vDz179+/xM+5+n7lte9TSlJ4eLh27dql5ORk1/Zz587p2LFjio+PV8+ePdW2bVvl5eW5Rb5Nmzb64YcfdOnSJde2a+/Ela78nTly5IiaNWv2q78zV/+TBZQXgolbRs+ePdW1a1eNHTtW27dvV1pamvbt26c33njD7Y7Qrl27qlmzZvrjH/8of39/t0u4rVu31pkzZ7RhwwalpaVp5cqV5XrzzY089dRTWrZsmb744gsdO3ZMr776arndqVm7dm09/vjjmj9/vrZv366jR4/qxRdfVHZ2toYOHSrpSlSDgoK0aNEiHT9+XImJiWW60alNmzYKCwvTvHnzFBMTYxSfq+9j7t+/X+Hh4a7tdrtdmzdvVnZ2tuv9y/r166tBgwb68MMPdeLECSUnJ2vmzJny8vrfRa/Y2FjZbDb93//9n44cOaKdO3dqyZIlkv53Zj106FCdP39ezz33nPbv36+0tDTt3LlTM2bMcL3PDZQXgolbhoeHh5YuXaqIiAjNmDFDffv21aRJk3T8+HE1adLE7bEDBgzQ999/r/79+8vT09O1PSoqSiNHjtTcuXMVFxennTt3lng5sTyMGDFCcXFxmj59uoYMGSJJ6tOnj2rVqlUux3/++efVt29fTZ8+XQMHDtThw4f1zjvvuNbF29tbr732mtLS0jRw4EAtXLhQkydPLtOYjz76qAoLC4u92edaERERKiwsVKNGjdwuMYeFhamgoEB16tRRx44dJV25hPzXv/5Vhw8fVmxsrGbNmqWJEye63YVbp04dLV68WEeOHNGgQYM0b948jR8/XpJc6xoYGKj3339fNptNo0aNUv/+/fXSSy/Jx8fnhnf0AjfLw6qs7xYHbjODBg1SWFiY2w081cnSpUu1Zs0affbZZ1U9FZctW7Zo/Pjx2rlz53Xv8AUqEjf9AOXgp59+UmJioux2uxwOhz744AMdPnxYs2fPruqplVpeXp7S09P17rvvasyYMVU6l48++kgtW7ZU06ZN9cMPP2ju3Ll68MEHiSWqBMEEyoHNZtO6des0b948OZ1O/fa3v9U777zj9q0m1cXs2bO1adMmRUVFuS4vV5WsrCwtXLhQGRkZCggIUGRkpKZMmVKlc8Lti0uyAAAY4KYfAAAMEEwAAAwQTAAADJR408+5c3lyOsvnbc5GjeooO5tvJi4L1rDsWMPywTqWHWtYduW9hjabhxo0+M1195UYTKfTKrdgXj0eyoY1LDvWsHywjmXHGpZdZa0hl2QBADBAMAEAMEAwAQAwQDABADBAMAEAMEAwAQAwQDABADBAMAEAMEAwAQAwQDABADBAMAEAMEAwAQAwQDABADBAMAEAMEAwAQAwQDABADBAMAEAMEAwAQAwQDABADBAMAEAMEAwAQAwQDABADBAMAEAMOBV1ROo6VaufFdpaSfK7Xje3p4qLCwqt+PVBLm5OZKk+vX9jR7PGpaPsqxjy5atNHToU+U8I6BiEcwKlpZ2Qod/OCJPX7N/zFF6RZeuBDPzZ0cVzwQmrn69gOqGYFYCT19/1W7Vu6qnUWPln/hSkljjauLq1wuobngPEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAANelTXQf/7zlerV81PnzvbKGhIAUINVdlcqLZiJidvl7e1JMAEA5aKyu8IlWQAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADHhV1kC5uTn6+edcJSTMrqwhbwmpqSfkLPKs6mkAtwyn45JSU0/cdv8WXI+3t6cKC4uqehrVVmrqCTVq1LDSxuMMEwAAA5V2hlm/vr8aN26kyZP/VFlD3hISEmbrSFpWVU8DuGXYvHx1R8vGmjp1RlVPpcoFBNRVZub5qp5GtZWQMFve3pV3BY8zTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMeFXWQPfeG6l69fwqazgAQA1X2V2ptGD26nW/AgLqKjPzfGUNCQCowSq7K1ySBQDAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwADBBADAAMEEAMAAwQQAwIBXVU/gdlB0KUf5J76s6mnUWEWXciSJNa4mrny9Glf1NIBSI5gVrGXLVuV6PG9vTxUWFpXrMau73Nwrf43r1/c3ejxrWD5ufh0bl/vrAqgMBLOCDR36VLkeLyCgrjIzz5frMW83rGH5YB1xu+E9TAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAwQTAAADBBMAAAMEEwAAAx4lfQAm82jXAcs7+PdjljDsmMNywfrWHasYdmV5xoWdywPy7KschsJAIAaikuyAAAYIJgAABggmAAAGCCYAAAYIJgAABggmAAAGCCYAAAYIJgAABggmAAAGKiUYB4/flxDhgxRTEyMhgwZoh9//LEyhq1Wzp07p6effloxMTEaMGCAxo8fr7Nnz0qSUlJSFBcXp5iYGI0YMULZ2dmuzytu3+1s0aJFCg4O1n//+19JrGFpFRQUaObMmXrooYc0YMAAzZgxQ1Lxr2Ve5+62bdumQYMGaeDAgYqLi9Pnn38uiTUsSUJCgqKiotxev9LNr1u5rqlVCYYNG2atW7fOsizLWrdunTVs2LDKGLZaOXfunLVr1y7Xx6+++qo1ffp0q6ioyIqOjrZ2795tWZZlvfnmm9a0adMsy7KK3Xc7O3jwoDVy5EjrwQcftA4fPswa3oTZs2dbL7/8suV0Oi3LsqzMzEzLsop/LfM6/x+n02l169bNOnz4sGVZlnXo0CGrS5cuVlFREWtYgt27d1vp6emu1+9VN7tu5bmmFR7MrKwsKywszHI4HJZlWZbD4bDCwsKs7Ozsih66Wvv000+t4cOHW/v377f69+/v2p6dnW116dLFsiyr2H23q4KCAmvw4MFWWlqa6wXHGpbOhQsXrLCwMOvChQtu24t7LfM6d+d0Oq3w8HBrz549lmVZVnJysvXQQw+xhqVwbTBvdt3Ke01L/G0lZXXq1CkFBgbK09NTkuTp6akmTZro1KlTatiwYUUPXy05nU69//77ioqK0qlTp9SsWTPXvoYNG8rpdConJ6fYff7+/lUx9Sr3+uuvKy4uTi1atHBtYw1LJy0tTf7+/lq0aJGSkpL0m9/8RhMnTpSvr+8NX8uWZfE6v4aHh4cWLFigsWPHqnbt2srLy9PSpUuL/feQNbyxm1238l5Tbvq5Bc2ePVu1a9fWk08+WdVTqVb27dungwcPaujQoVU9lWqtqKhIaWlpuvvuu7V27VpNmTJFzz77rPLz86t6atWGw+HQkiVL9NZbb2nbtm16++23NWnSJNawmqvwM8ygoCCdOXNGRUVF8vT0VFFRkTIyMhQUFFTRQ1dLCQkJOnHihBYvXiybzaagoCClp6e79p89e1Y2m03+/v7F7rsd7d69W0ePHlXv3r0lSadPn9bIkSM1bNgw1rAUgoKC5OXlpdjYWEnSPffcowYNGsjX1/eGr2XLsnidX+PQoUPKyMhQWFiYJCksLEx+fn6qVasWa3gTiutIcetW3mta4WeYjRo10l133aVNmzZJkjZt2qS77rrrtr/EcD2vvfaaDh48qDfffFM+Pj6SpE6dOunSpUvas2ePJGnVqlX63e9+V+K+29Ho0aOVmJiorVu3auvWrWratKmWLVumUaNGsYal0LBhQ0VEROg///mPpCt3GWZnZ6t169Y3fC3zOnfXtGlTnT59WseOHZMkHT16VNnZ2WrVqhVreBOKW5ub3XczKuUXSB89elTTpk3Tzz//rHr16ikhIUFt2rSp6GGrlR9++EGxsbFq3bq1fH19JUktWrTQm2++qb1792rmzJkqKChQ8+bN9ec//1mNGzeWpGL33e6ioqK0ePFitW/fnjUspbS0NP3pT39STk6OvLy8NGnSJEVGRhb7WuZ17m7Dhg1655135OHhIUmaMGGCoqOjWcMSzJkzR59//rmysrLUoEED+fv7a/PmzTe9buW5ppUSTAAAqjtu+gEAwADBBADAAMEEAMAAwQQAwADBBADAAMEEKsjixYv1wgsvSJJOnjyp4OBgORwOSdKoUaP00UcfVeX0AJQS31YCVIKTJ0+qd+/e+u677+TlVeE/YAtABeAME7e1q2d8AFASgokaKSoqSkuWLFG/fv1kt9s1ffp0FRQUKCkpSffff7+WLl2qXr16afr06XI6nVq6dKmio6MVERGhiRMnKicnR9KVS6crVqxwO/a1vwx4zpw5ioyMVNeuXfXII4+4fsSeJC1cuFBTpky57vyGDRumDz/8UJK0du1aPf7440pISJDdbldUVJS2b9/uemxaWpqeeOIJhYaG6g9/+INeeumlGx73qqlTp2r58uWSpDNnzig4OFjvvfeeJCk1NVXh4eFyOp3Kzc1VfHy8unfvLrvdrvj4eJ0+fdp47JSUFP3+979Xt27dFBcXp6SkpOK/MEA1RjBRY23cuFHLli3TF198oePHj+utt96SJGVlZSk3N1fbtm3T7Nmz9a9//UtbtmzRihUrtGPHDtWvX1+zZs2SJMXGxrp+DqUkHTlyROnp6XrggQckSZ07d9a6deuUnJys2NhYTZw4UQUFBaWe64EDB3TnnXdq165dGjVqlF544QVdfbdkypQpCgkJUVJSksaPH6/169eXeDy73a7k5GRJUnJyslq2bKndu3e7Pg4LC5PNZpPT6dQjjzyibdu2adu2bapVq5bruZc09pkzZxQfH68xY8YoOTlZU6dO1YQJE3T27NlSP3+gOiCYqLGeeOIJBQUFyd/fX2PGjNHmzZslSTabTRMmTJCPj498fX21atUqPffcc2ratKl8fHw0fvx4ffbZZ3I4HIqOjtb333+vn376SdKVCPfp08f1w/EHDhyoBg0ayMvLSyNGjNDly5d1/PjxUs+1WbNmGjx4sDw9PfXwww8rMzNTWVlZSk9P17fffuuab7du3RQVFVXi8cLDw/XNN9/I6XRq9+7dGjVqlPbu3Svpym91CQ8PlyQ1aNBAMTEx8vPzU506dTRmzBhXWEsae/369br//vsVGRkpm82mXr16qVOnTm5nx0BNwt0HqLGu/RU+zZo1U0ZGhqQrkahVq5ZrX3p6usaNGyeb7X//f7TZbMrOzlZgYKAiIyO1efNmjR49Wps2bdKcOXNcj1u2bJlWr16tjIwMeXh46MKFCzp37lyp53rtD3v38/OTJOXn5+vcuXOqX7++a9vV53Xq1Klij3fHHXfIz89Phw4d0jfffKNx48Zp9erVOnbsmHbv3q1hw4ZJki5evKhXXnlFO3bsUG5uriQpLy/P9WuQihs7PT1dn376qbZt2+ba73A4FBERUernD1QHBBM11rVRSU9PV5MmTSTJ9dsjrmratKnmzp3r+t2FvxQbG6tFixbJbreroKDAFYQ9e/bob3/7m/7xj3+oXbt2stlsstvtKs8bzwMCApSbm6uLFy+6wlVSLK+y2+367LPPVFhYqMDAQNntdq1bt065ubm66667JEnLly/X8ePH9cEHHyggIECHDh3SoEGDZFlWiWMHBQVp4MCBbv+BAGoyLsmixlq5cqVOnz6tnJwcLV68WP369bvu4x5//HEtWLDAddn17Nmz2rJli2t/ZGSk0tPT9cYbb6hfv36uM9G8vDx5enqqYcOGcjgcWrRokS5cuFCuz6F58+bq1KmTFi5cqMuXL2vfvn1uZ3TFCQ8P14oVK9StWzdJUkREhFasWKGwsDB5enq6nkOtWrVUr1495eTkaNGiRcZjx8XFadu2bdqxY4eKiopcN1Vde9MQUJMQTNRYsbGxGjFihKKjo3XHHXdozJgx133cU089paioKI0YMUKhoaEaPHiwDhw44Nrv4+OjPn36aOfOnYqNjXVtv/fee3XfffcpJiZGUVFRqlWr1k3/JvfizJ8/XykpKYqIiNCCBQvUr18/13uoxbHb7crLy5PdbpckhYWF6dKlS66AStLw4cNVUFCg7t27a8iQIbrvvvuMxw4KCtJbb72lJUuWqEePHoqMjNSyZcvkdDrL8dkDtw5+cAFqpKioKM2ZM0c9e/as6qmUu0mTJqlNmzaaMGHCbTU2UNU4wwRucQcOHFBqaqqcTqe++uorffnll4qOjq7xYwO3Gm76AW5xWVlZevbZZ5WTk6OmTZvqxRdf1N13360NGzZo5syZv3p8s2bNXN9CU1FjA7cjLskCAGCAS7IAABggmAAAGCCYAAAYIJgAABggmAAAGCCYAAAY+H+IgP2Bec9v9QAAAABJRU5ErkJggg==\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "b=visa.loc[visa['unit_of_wage']=='Hour']['prevailing_wage'].argmax()\n",
        "visa.iloc[b]['full_time_position']"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "-SCMc2pco29j",
        "outputId": "91af4e73-1761-45d4-9921-b69e022087b9"
      },
      "id": "-SCMc2pco29j",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'Y'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "a=visa.loc[visa['unit_of_wage']=='Hour']['prevailing_wage'].max()\n",
        "print('The maximum hourly wage is ${} for a full time position, equivalently ${} per year!'.format(a,2080*a))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MIlbtpGhoGHa",
        "outputId": "701a1f3c-dfc3-4730-faef-47f4d03236ad"
      },
      "id": "MIlbtpGhoGHa",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The maximum hourly wage is $999.9195 for a full time position, equivalently $2079832.56 per year!\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "For the record, some hourly data points seem unreasonable too: The maximum wage is about \\$1000 per hour, or around \\$2 million per year. Again, there is no evidence that this data is necessarily erroneous; it simply stands out."
      ],
      "metadata": {
        "id": "QxwGcOX1n4Xu"
      },
      "id": "QxwGcOX1n4Xu"
    },
    {
      "cell_type": "code",
      "source": [
        "plott()\n",
        "plt.title('Prevailing Wage by Region',fontsize=14)\n",
        "sns.boxplot(data=visa,\n",
        "               x='prevailing_wage',\n",
        "               y='region_of_employment');"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "DSlArTDVEgBb",
        "outputId": "b3a7779d-9b17-41d4-bd3e-cbd075e57939"
      },
      "id": "DSlArTDVEgBb",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The middle 50% of the data for prevailing wage is higher in the midwest and island regions. The northeast has the lowest first quartile. Every region has many extreme values on the high end."
      ],
      "metadata": {
        "id": "2T1jV3NL7oxk"
      },
      "id": "2T1jV3NL7oxk"
    },
    {
      "cell_type": "code",
      "source": [
        "sns.catplot(data=visa,\n",
        "            x='case_status',\n",
        "            col='education_of_employee',\n",
        "            kind='count',\n",
        "            col_wrap=2);"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 725
        },
        "id": "stcW_7X1FIig",
        "outputId": "fa9c9c00-f108-424a-8fb5-9eadff2aad51"
      },
      "id": "stcW_7X1FIig",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x720 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Looking toward our dependent variable, it appears that ```case_status``` is influenced by education. The trouble here is that it is difficult to compare the regions, as the scale is different for each. Is the _percentage_ Certified for Doctorate any greater or less than that for, say, Bachelor's?\n",
        "\n",
        "Let's instead examine the percentage Certified and Denied."
      ],
      "metadata": {
        "id": "YofS02QG8SIF"
      },
      "id": "YofS02QG8SIF"
    },
    {
      "cell_type": "code",
      "source": [
        "# dataframe of percentages\n",
        "a=visa.groupby('education_of_employee')['case_status'].value_counts(normalize=True)\n",
        "b=pd.DataFrame(index=['High School',\"Bachelor's\",\"Master's\",'Doctorate'],\n",
        "               columns=['Certified','Denied'])\n",
        "for (c,d) in a.index:\n",
        "  b.loc[c,d]=a[(c,d)]\n",
        "b"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "omF0GmgRF4MK",
        "outputId": "6cefc292-cfac-417a-f9a6-f90d9e455eb7"
      },
      "id": "omF0GmgRF4MK",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "            Certified    Denied\n",
              "High School  0.340351  0.659649\n",
              "Bachelor's   0.622142  0.377858\n",
              "Master's     0.786278  0.213722\n",
              "Doctorate    0.872263  0.127737"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-4e87a7cb-6f7e-4d9a-884b-41e373971971\">\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>Certified</th>\n",
              "      <th>Denied</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>High School</th>\n",
              "      <td>0.340351</td>\n",
              "      <td>0.659649</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bachelor's</th>\n",
              "      <td>0.622142</td>\n",
              "      <td>0.377858</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Master's</th>\n",
              "      <td>0.786278</td>\n",
              "      <td>0.213722</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Doctorate</th>\n",
              "      <td>0.872263</td>\n",
              "      <td>0.127737</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4e87a7cb-6f7e-4d9a-884b-41e373971971')\"\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-4e87a7cb-6f7e-4d9a-884b-41e373971971 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-4e87a7cb-6f7e-4d9a-884b-41e373971971');\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": 35
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# barplot of percentages\n",
        "plott()\n",
        "plt.title('Percent Certified by Education Level',fontsize=14)\n",
        "plt.bar(b.index,\n",
        "        b['Certified'],\n",
        "        label='Certified')\n",
        "plt.bar(b.index,\n",
        "        b['Denied'],\n",
        "        bottom=b['Certified'],\n",
        "        label='Denied')\n",
        "plt.legend(loc='lower right')\n",
        "plt.xlabel('Education Level')\n",
        "plt.ylabel('Percent Certified/Denied')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "8ySgUscIJ58y",
        "outputId": "fa94bd63-9e74-4898-8f0a-51f546501454"
      },
      "id": "8ySgUscIJ58y",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Indeed, here we can compare ```case_status``` on like scales. We find that applicants with a Doctorate are most often certified (over 87% of the time), while employees with only a High School education are certified around a third of the time. We find a clear trend: Visa status is directly connected with level of education. Higher levels of education lead to a greater percentage of visas certifed.\n",
        "\n",
        "To plot the rest of our features as percentages, we make the above code into a function."
      ],
      "metadata": {
        "id": "I3RqCDa89j9i"
      },
      "id": "I3RqCDa89j9i"
    },
    {
      "cell_type": "code",
      "source": [
        "def percent_status(col):\n",
        "  '''Plot percent Certified/Denied\n",
        "  for classes of a categorical variable.'''\n",
        "\n",
        "  # generate dataframe of percentages\n",
        "  a=visa.groupby(col)['case_status'].value_counts(normalize=True)\n",
        "  #compute ascending order of classes\n",
        "  ser=pd.Series(dtype='float')\n",
        "  for name in visa[col].unique():\n",
        "    ser[name]=a[(name,'Certified')]\n",
        "  # dataframe\n",
        "  b=pd.DataFrame(index=ser.sort_values().index.tolist(),\n",
        "                 columns=['Certified','Denied'])\n",
        "  for (c,d) in a.index:\n",
        "    b.loc[c,d]=a[(c,d)]\n",
        "  \n",
        "  # plot percentages\n",
        "  plott()\n",
        "  plt.title('Percent Certified by '+col,fontsize=14)\n",
        "  plt.bar(b.index,b['Certified'],label='Certified')\n",
        "  plt.bar(b.index,b['Denied'],bottom=b['Certified'],label='Denied')\n",
        "  plt.legend(loc='lower right')\n",
        "  plt.xlabel(col)\n",
        "  plt.ylabel('Percent Certified/Denied')\n",
        "  plt.show()"
      ],
      "metadata": {
        "id": "Mf3NfypwtYps"
      },
      "id": "Mf3NfypwtYps",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "percent_status('continent')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "ie73TsCbw4ky",
        "outputId": "508fa520-1786-4355-b453-5cd5dabb39ea"
      },
      "id": "ie73TsCbw4ky",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Europe, Africa, and Asia see the greatest percentage of certified visas.\n",
        "* South America sees the least.\n",
        "* Every region has over 50% certification rate."
      ],
      "metadata": {
        "id": "oLMPOdP9-bXd"
      },
      "id": "oLMPOdP9-bXd"
    },
    {
      "cell_type": "code",
      "source": [
        "percent_status('has_job_experience')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "15xyJn4S9Vcm",
        "outputId": "946b4c76-6581-4ace-af7a-6aca265f6555"
      },
      "id": "15xyJn4S9Vcm",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Having job experince certainly helps, as the proportion of Certified visas is higher for applicants with job experience. Let's test whether this difference is significant, assuming a level of siginificance of 5%. Please note that our populations are binomially distributed, independent, and simply randomly sampled; they furthermore satisfy the sample size inequalities. The assumptions for a two proportion z-test are therefore met. Our null hypothesis is that our sample proportions come from  populations with the same ratios of Certification to Denial; the alternative is that the ratios are truly different."
      ],
      "metadata": {
        "id": "32RM_732-x7-"
      },
      "id": "32RM_732-x7-"
    },
    {
      "cell_type": "code",
      "source": [
        "from statsmodels.stats.proportion import proportions_ztest\n",
        "\n",
        "def cert_ztest(col):\n",
        "  '''Run a two proportions independent\n",
        "  z-test for case_status.'''\n",
        "  # collect data\n",
        "  size=visa[col].value_counts()\n",
        "  a=visa.groupby(col)['case_status'].value_counts()\n",
        "  cert=[]\n",
        "  for idx in size.index.tolist():\n",
        "    cert.append(a[idx,'Certified'])\n",
        "  # run test\n",
        "  t,p_val=proportions_ztest(cert,size)\n",
        "  print('The p-value is',p_val)"
      ],
      "metadata": {
        "id": "KTyGNBDBCQPj"
      },
      "id": "KTyGNBDBCQPj",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "cert_ztest('has_job_experience')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "N8kIy3TIComH",
        "outputId": "6566e90d-b603-4e5f-e423-9b5a91c4508c"
      },
      "id": "N8kIy3TIComH",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The p-value is 1.2710489965841227e-206\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "With an astoundingly low p-value, we can confidently conclude that these sample proportions reflect a real-world difference for visa applications: We find that a greater proportion of applications are certified when the applicant has previous job experience."
      ],
      "metadata": {
        "id": "5zpQvyoil34u"
      },
      "id": "5zpQvyoil34u"
    },
    {
      "cell_type": "code",
      "source": [
        "percent_status('requires_job_training')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "5TBKUwD19a4A",
        "outputId": "a4e17603-7d53-4318-844b-deefcb64cee5"
      },
      "id": "5TBKUwD19a4A",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "visa.groupby('requires_job_training')['case_status'].value_counts(normalize=True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "odC2i4ZsDgVI",
        "outputId": "fe6d20bc-ebdb-4971-927d-1a5454ae2474"
      },
      "id": "odC2i4ZsDgVI",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "requires_job_training  case_status\n",
              "N                      Certified      0.666459\n",
              "                       Denied         0.333541\n",
              "Y                      Certified      0.678849\n",
              "                       Denied         0.321151\n",
              "Name: case_status, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Unlike the previous feature, the requirement of job training has very close proportions: 66.6% Certified (N) versus 67.9% (Y). There is apparently no difference in ```case_status``` based on the job training requirement. To confirm, we can run another two proportion z-test. (As before, the assumptions for the test are met.) We set the level of siginificance at 0.05. Our null hypothesis is that our sample proportions come from  populations with the same ratios of Certification to Denial; the alternative is that the ratios are truly different."
      ],
      "metadata": {
        "id": "VTl-G-WyZvG5"
      },
      "id": "VTl-G-WyZvG5"
    },
    {
      "cell_type": "code",
      "source": [
        "cert_ztest('requires_job_training')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GpPEclpDDRu6",
        "outputId": "98d68242-15c8-46ee-fd65-df8b24e7c51b"
      },
      "id": "GpPEclpDDRu6",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The p-value is 0.1787590242870024\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Our p-value exceeds 0.05, so we fail to reject the null hypothesis. We must conclude that there is no significant difference between the two proportions. Put another way, ```requires_job_training``` does not impact ```case_status``` on its own."
      ],
      "metadata": {
        "id": "UFd8cU8FarTH"
      },
      "id": "UFd8cU8FarTH"
    },
    {
      "cell_type": "code",
      "source": [
        "percent_status('region_of_employment')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "Pne0v1rv9egO",
        "outputId": "525506ee-cc9e-426a-bc70-f25bf72ad619"
      },
      "id": "Pne0v1rv9egO",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "There seems to be little difference in visa certification rate among island, western, and northeastern jobs. The south and midwest look to have appreciably higher rates."
      ],
      "metadata": {
        "id": "rRfhzqqXbwlQ"
      },
      "id": "rRfhzqqXbwlQ"
    },
    {
      "cell_type": "code",
      "source": [
        "percent_status('unit_of_wage')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "du7GY_gx9mdA",
        "outputId": "6cda8957-28f8-46fc-a5e0-e6c4685e8c6e"
      },
      "id": "du7GY_gx9mdA",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Applicants to hourly jobs see far less than 50% certification. This is well below the rest, with yearly salaried jobs being most likely to lead to visa certification."
      ],
      "metadata": {
        "id": "wsCZ8xqFcfVq"
      },
      "id": "wsCZ8xqFcfVq"
    },
    {
      "cell_type": "code",
      "source": [
        "percent_status('full_time_position')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "G-JVGarR9r0y",
        "outputId": "8cbc7f5b-5cf5-4dcd-bae3-79aaea5e0989"
      },
      "id": "G-JVGarR9r0y",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "visa.groupby('full_time_position')['case_status'].value_counts(normalize=True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0wr9nW0NDnP2",
        "outputId": "399cee15-5503-4286-cd4c-56900e625593"
      },
      "id": "0wr9nW0NDnP2",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "full_time_position  case_status\n",
              "N                   Certified      0.685260\n",
              "                    Denied         0.314740\n",
              "Y                   Certified      0.665832\n",
              "                    Denied         0.334168\n",
              "Name: case_status, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 48
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Similar to ```requires_job_training```, it appears here that there is not much difference in the resulting certification based on whether the position is full time. We find that 68.5% are Certified for part time work and 66.6% are Certified for full time work. Our assumptions being once more satisfied, we can run a two proportion z-test. We set the level of siginificance at 0.05. Our null hypothesis is that our sample proportions come from populations with the same ratios of Certification to Denial; the alternative is that the ratios are truly different."
      ],
      "metadata": {
        "id": "E0RvP4UAebK0"
      },
      "id": "E0RvP4UAebK0"
    },
    {
      "cell_type": "code",
      "source": [
        "cert_ztest('full_time_position')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "W5QUkuBsDXVG",
        "outputId": "c56c2f61-d214-4578-d116-ea0d262e2011"
      },
      "id": "W5QUkuBsDXVG",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The p-value is 0.042452929825717224\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "With a p-value under 0.05, we can reject the null hypothesis. In this case, there is a statistically significant difference between these two proportions. That means that an applicant is more likely to be certified for part time work than full time work."
      ],
      "metadata": {
        "id": "boxWk9Nze8PA"
      },
      "id": "boxWk9Nze8PA"
    },
    {
      "cell_type": "code",
      "source": [
        "plott()\n",
        "plt.title('Prevailing Wage and Case Status',fontsize=14)\n",
        "sns.boxplot(data=visa,x='prevailing_wage',\n",
        "            y='case_status',\n",
        "            order=['Certified','Denied']);"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "mJVJIYA5-jQV",
        "outputId": "4265f790-7481-4a84-f9f2-b1d9306474f8"
      },
      "id": "mJVJIYA5-jQV",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "visa.groupby('case_status')['prevailing_wage'].describe().T"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "GAA2oCQKi2ZO",
        "outputId": "88d87f35-e113-49e0-84d7-dbbe3ba5eb01"
      },
      "id": "GAA2oCQKi2ZO",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "case_status      Certified         Denied\n",
              "count         17018.000000    8462.000000\n",
              "mean          77293.619243   68748.681580\n",
              "std           52042.715576   53890.166031\n",
              "min               2.136700       2.956100\n",
              "25%           38375.330000   23497.295000\n",
              "50%           72486.270000   65431.460000\n",
              "75%          108879.107500  105097.640000\n",
              "max          318446.050000  319210.270000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-959d0b95-f8a3-4bf8-b94e-e364f8151f7e\">\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>case_status</th>\n",
              "      <th>Certified</th>\n",
              "      <th>Denied</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>17018.000000</td>\n",
              "      <td>8462.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>77293.619243</td>\n",
              "      <td>68748.681580</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>52042.715576</td>\n",
              "      <td>53890.166031</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>2.136700</td>\n",
              "      <td>2.956100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>38375.330000</td>\n",
              "      <td>23497.295000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>72486.270000</td>\n",
              "      <td>65431.460000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>108879.107500</td>\n",
              "      <td>105097.640000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>318446.050000</td>\n",
              "      <td>319210.270000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-959d0b95-f8a3-4bf8-b94e-e364f8151f7e')\"\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-959d0b95-f8a3-4bf8-b94e-e364f8151f7e 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-959d0b95-f8a3-4bf8-b94e-e364f8151f7e');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 51
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The mean prevailing wage for certified applications is higher than that for denied applications by about \\$8500. The middle 50% for certified is concentrated higher too. Interestingly, the whiskers extend further for denied applications. Taken with its wider IQR, this shows that denied applications have greater variance in prevailing wage.\n",
        "\n",
        "This is also true across various regions."
      ],
      "metadata": {
        "id": "jgq3D4hNiz4B"
      },
      "id": "jgq3D4hNiz4B"
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(15,15))\n",
        "for idx,region in enumerate(visa['region_of_employment'].unique()):\n",
        "  plt.subplot(3,2,idx+1)\n",
        "  plt.title('Prevailing wage in '+region)\n",
        "  sns.boxplot(data=visa.loc[visa['region_of_employment']==region],\n",
        "              x='prevailing_wage',\n",
        "              y='case_status',\n",
        "              order=['Certified','Denied'])\n",
        "plt.tight_layout()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 975
        },
        "id": "VHiACbj9j9_x",
        "outputId": "30e838dd-2f8b-4ab5-a339-d86fb3d5569e"
      },
      "id": "VHiACbj9j9_x",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1080x1080 with 5 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Denied applications in the northeast seem to have some of the lowest prevailing wages. The prevailing wages for the island region, on the other hand, appear generally higher than the rest, with certified cases having the greatest middle 50% among the regions. This is likely due to the higher cost of living on islands, as most goods must be imported."
      ],
      "metadata": {
        "id": "oqwzvPPsll7k"
      },
      "id": "oqwzvPPsll7k"
    },
    {
      "cell_type": "markdown",
      "id": "alleged-spirituality",
      "metadata": {
        "id": "alleged-spirituality"
      },
      "source": [
        "## Data Preprocessing"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Feature Engineering"
      ],
      "metadata": {
        "id": "D42w4a2XME_w"
      },
      "id": "D42w4a2XME_w"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "increasing-louisiana",
      "metadata": {
        "id": "increasing-louisiana"
      },
      "outputs": [],
      "source": [
        "visa['years_in_business']=2016-visa['yr_of_estab']\n",
        "visa.drop('yr_of_estab',axis=1,inplace=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Rather than recording the year businesses were established, we will instead consider the number of years in business."
      ],
      "metadata": {
        "id": "5jZn2oPQmVeI"
      },
      "id": "5jZn2oPQmVeI"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Outlier detection and treatment"
      ],
      "metadata": {
        "id": "p0zlVioEL_Ur"
      },
      "id": "p0zlVioEL_Ur"
    },
    {
      "cell_type": "code",
      "source": [
        "visa.loc[visa['no_of_employees']<1].shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OMci272TZbRL",
        "outputId": "ceb4919f-3308-4146-c9b7-8b755d2dd3e5"
      },
      "id": "OMci272TZbRL",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(33, 12)"
            ]
          },
          "metadata": {},
          "execution_count": 54
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "There are 33 records where the application lists a negative number of employees. These clearly constitute an error, and with no clear method to impute these values, our best recourse is to drop the rows."
      ],
      "metadata": {
        "id": "T1F5fZnSmg7T"
      },
      "id": "T1F5fZnSmg7T"
    },
    {
      "cell_type": "code",
      "source": [
        "idx=visa.loc[visa['no_of_employees']<1].index\n",
        "visa.drop(idx,axis=0,inplace=True)"
      ],
      "metadata": {
        "id": "wYh2e25MZ8gK"
      },
      "id": "wYh2e25MZ8gK",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Data Prep"
      ],
      "metadata": {
        "id": "98y_YHTcsxfY"
      },
      "id": "98y_YHTcsxfY"
    },
    {
      "cell_type": "code",
      "source": [
        "visa.drop('case_id',axis=1,inplace=True)"
      ],
      "metadata": {
        "id": "dMdzuF2Ss6Bx"
      },
      "id": "dMdzuF2Ss6Bx",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "We drop ```case_id```, as the unique identifiers will not lend any predictive power to our model."
      ],
      "metadata": {
        "id": "-dH6OfgEm8lX"
      },
      "id": "-dH6OfgEm8lX"
    },
    {
      "cell_type": "code",
      "source": [
        "# convert object dtype to category\n",
        "for col in visa.select_dtypes('object').columns:\n",
        "  visa[col]=pd.Categorical(visa[col])"
      ],
      "metadata": {
        "id": "QwEcXShij5xt"
      },
      "id": "QwEcXShij5xt",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "After making all object variables into category type, we encode them numerically."
      ],
      "metadata": {
        "id": "bIOODwvqnEqM"
      },
      "id": "bIOODwvqnEqM"
    },
    {
      "cell_type": "code",
      "source": [
        "# define categorical ordering\n",
        "order_struct={\n",
        "    'education_of_employee':{'High School':1,\n",
        "                             \"Bachelor's\":2,\n",
        "                             \"Master's\":3,\n",
        "                             'Doctorate':4},\n",
        "    'has_job_experience':{'Y':1,'N':0},\n",
        "    'requires_job_training':{'Y':1,'N':0},\n",
        "    'full_time_position':{'Y':1,'N':0},\n",
        "    'case_status':{'Certified':1,'Denied':0}\n",
        "              }\n",
        "no_order={'continent','region_of_employment','unit_of_wage'}"
      ],
      "metadata": {
        "id": "mNz4WKZokEJQ"
      },
      "id": "mNz4WKZokEJQ",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# convert data to numeric\n",
        "visa.replace(order_struct,inplace=True)\n",
        "visa=pd.get_dummies(visa,columns=no_order)"
      ],
      "metadata": {
        "id": "qpng9JadkEG9"
      },
      "id": "qpng9JadkEG9",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "We use one hot encoding for the categorical variables with no inherent order."
      ],
      "metadata": {
        "id": "BMrp-4oNnRWN"
      },
      "id": "BMrp-4oNnRWN"
    },
    {
      "cell_type": "code",
      "source": [
        "X=visa.drop('case_status',axis=1)\n",
        "y=visa['case_status']"
      ],
      "metadata": {
        "id": "27lzEJkwl3ks"
      },
      "id": "27lzEJkwl3ks",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# split\n",
        "X_train,X_test,y_train,y_test=train_test_split(X,y,\n",
        "                                               test_size=0.3,\n",
        "                                               stratify=y,\n",
        "                                               random_state=57)"
      ],
      "metadata": {
        "id": "SfnEoAyol8_C"
      },
      "id": "SfnEoAyol8_C",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now that we've split our data, it is ready for modeling."
      ],
      "metadata": {
        "id": "TvDh5DZJzqbx"
      },
      "id": "TvDh5DZJzqbx"
    },
    {
      "cell_type": "markdown",
      "id": "difficult-union",
      "metadata": {
        "id": "difficult-union"
      },
      "source": [
        "## Second EDA\n"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We only removed 33 records out of 25480. That's only 0.1% of the data. Accordingly, there is not enough change in our data to produce appreciably different visualizations. We only include plots for features that were directly altered/engineered. "
      ],
      "metadata": {
        "id": "7Vcxvad0iiBJ"
      },
      "id": "7Vcxvad0iiBJ"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "interested-talent",
      "metadata": {
        "id": "interested-talent",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "outputId": "403235c3-abaa-4675-f829-a33baf55be9d"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "plott('years_in_business')"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plott()\n",
        "plt.title('Boxplot of Years Since Founding',fontsize=14)\n",
        "sns.boxplot(data=visa,x='years_in_business');"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "CUQLZS-aCsAq",
        "outputId": "819b4097-3aad-4102-a915-c88aeb14ce0f"
      },
      "id": "CUQLZS-aCsAq",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* We find that most businesses in our records are less than 50 years old.\n",
        "* There are a steady supply of businesses greater than 75 years old, about the same amount for each year. It only starts to trail off after around 175 years."
      ],
      "metadata": {
        "id": "_ceJO-gBCl1z"
      },
      "id": "_ceJO-gBCl1z"
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(16,5))\n",
        "plt.suptitle('Feature: no_of_employees',fontsize=18)\n",
        "\n",
        "plt.subplot(1,2,1)\n",
        "plt.title('Boxplot (with fliers)',fontsize=14)\n",
        "sns.boxplot(data=visa,x='no_of_employees')\n",
        "\n",
        "plt.subplot(1,2,2)\n",
        "plt.title('Boxplot (fliers removed)',fontsize=14)\n",
        "sns.boxplot(data=visa,x='no_of_employees',showfliers=False)\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 375
        },
        "id": "q9XjYFQVigdt",
        "outputId": "fc605a9c-0c5e-4feb-b32f-69cec4a74467"
      },
      "id": "q9XjYFQVigdt",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1152x360 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "With so few records removed, there is little difference in the plots for number of employees. Put another way, the comments on this feature from the first EDA are still applicable. The middle 50% of observations is between about 1000 and 3500. There are also tiny companies, with not more than several employees, and massive companies, with over half a million employees, in our data set."
      ],
      "metadata": {
        "id": "QwZR8EcYDvtv"
      },
      "id": "QwZR8EcYDvtv"
    },
    {
      "cell_type": "markdown",
      "id": "domestic-iceland",
      "metadata": {
        "id": "domestic-iceland"
      },
      "source": [
        "## Building bagging and boosting models"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "When scoring our models, we must find a good balance between false positives and false negatives.\n",
        "* A false positive is a certified visa that should have been denied. We want to reduce false positives because these workers may be ineffective and not benefit the company.\n",
        "* On the other hand, a false negative constitutes a denied visa that should have been certified. This deprives the workforce of an asset, someone who could benefit US companies.\n",
        "\n",
        "With this in mind, we will score our models on F1, as it balances recall and precision, reducing both false positives and false negatives."
      ],
      "metadata": {
        "id": "dfMMD01bMkKI"
      },
      "id": "dfMMD01bMkKI"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Functions"
      ],
      "metadata": {
        "id": "9V9RR10TmfdZ"
      },
      "id": "9V9RR10TmfdZ"
    },
    {
      "cell_type": "code",
      "source": [
        "def scores(model):\n",
        "  '''Print training and testing\n",
        "  metrics for the specified model.'''\n",
        "\n",
        "  score_idx=['Accuracy','Precision','Recall','F1']\n",
        "  X_train_pred=model.predict(X_train)\n",
        "  score_list_train=[metrics.accuracy_score(y_train,X_train_pred),\n",
        "                    metrics.precision_score(y_train,X_train_pred),\n",
        "                    metrics.recall_score(y_train,X_train_pred),\n",
        "                    metrics.f1_score(y_train,X_train_pred)]\n",
        "\n",
        "  X_test_pred=model.predict(X_test)\n",
        "  score_list_test=[metrics.accuracy_score(y_test,X_test_pred),\n",
        "                   metrics.precision_score(y_test,X_test_pred),\n",
        "                   metrics.recall_score(y_test,X_test_pred),\n",
        "                   metrics.f1_score(y_test,X_test_pred)]\n",
        "\n",
        "  df=pd.DataFrame(index=score_idx)\n",
        "  df['Train']=score_list_train\n",
        "  df['Test']=score_list_test\n",
        "  return df"
      ],
      "metadata": {
        "id": "H92Kf99dmey8"
      },
      "id": "H92Kf99dmey8",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "This function gets the scores of a model on training data and testing data."
      ],
      "metadata": {
        "id": "5E894mMhyiGz"
      },
      "id": "5E894mMhyiGz"
    },
    {
      "cell_type": "code",
      "source": [
        "model_comp_table=pd.DataFrame(columns=['Train Acc','Test Acc','Train F1','Test F1','Status'])\n",
        "y_train_test_lens=[y_train.shape[0],y_test.shape[0]]\n",
        "\n",
        "def tabulate(model,name):\n",
        "  '''Compute train/test accuracy and\n",
        "  F1 for a given model. Add to table.'''\n",
        "\n",
        "  # run predictions with model\n",
        "  X_train_pred=model.predict(X_train)\n",
        "  X_test_pred=model.predict(X_test)\n",
        "\n",
        "  # run overfitting test\n",
        "  train_acc_count=np.logical_not(np.logical_xor(y_train,X_train_pred)).sum()\n",
        "  test_acc_count=np.logical_not(np.logical_xor(y_test,X_test_pred)).sum()\n",
        "  t,p_val=proportions_ztest([train_acc_count,test_acc_count],y_train_test_lens)\n",
        "  \n",
        "  # assign rating based on p-value\n",
        "  rating=str()\n",
        "  if p_val<0.05:\n",
        "    rating='Overfit'\n",
        "  else:\n",
        "    rating='General'\n",
        "\n",
        "  # collect data for new table row\n",
        "  model_comp_table.loc[name]=[metrics.accuracy_score(y_train,X_train_pred),\n",
        "                              metrics.accuracy_score(y_test,X_test_pred),\n",
        "                              metrics.f1_score(y_train,X_train_pred),\n",
        "                              metrics.f1_score(y_test,X_test_pred),\n",
        "                              rating]\n",
        "\n",
        "  return model_comp_table"
      ],
      "metadata": {
        "id": "w1AaK_l2MpHa"
      },
      "id": "w1AaK_l2MpHa",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "This function adds accuracy and F1 scores to a DataFrame that we can use to compare models. It also runs a two proportion z-test to determine whether the training accuracy and testing accuracy are different enough for the model to be considered overfit. We assume a level of significance of 5% in the interpretation of the p-value. All conditions are necessarily met to justify the test."
      ],
      "metadata": {
        "id": "z-SFO41TyoAE"
      },
      "id": "z-SFO41TyoAE"
    },
    {
      "cell_type": "code",
      "source": [
        "def confusion_heatmap(model,show_scores=False):\n",
        "  '''Heatmap of confusion matrix of\n",
        "  model predictions on test data.'''\n",
        "\n",
        "  actual=y_test\n",
        "  predicted=model.predict(X_test)\n",
        "  # generate confusion matrix\n",
        "  cm=metrics.confusion_matrix(actual,predicted)\n",
        "  cm=np.flip(cm).T\n",
        "\n",
        "  # heatmap labels\n",
        "  labels=['TP','FP','FN','TN']\n",
        "  cm_labels=np.array(cm).flatten()\n",
        "  cm_percents=np.round((cm_labels/np.sum(cm))*100,3)\n",
        "  annot_labels=[]\n",
        "  for i in range(4):\n",
        "    annot_labels.append(str(labels[i])+'\\nCount:'+str(cm_labels[i])+'\\n'+str(cm_percents[i])+'%')\n",
        "  annot_labels=np.array(annot_labels).reshape(2,2)\n",
        "\n",
        "  # print figure\n",
        "  plt.figure(figsize=(8,5))\n",
        "  plt.title('Confusion Matrix',fontsize=20)\n",
        "  sns.heatmap(data=cm,\n",
        "              annot=annot_labels,\n",
        "              annot_kws={'fontsize':'x-large'},\n",
        "              xticklabels=[1,0],\n",
        "              yticklabels=[1,0],\n",
        "              cmap='Greens',\n",
        "              fmt='s')\n",
        "  plt.xlabel('Actual',fontsize=14)\n",
        "  plt.ylabel('Predicted',fontsize=14)\n",
        "  plt.tight_layout();\n",
        "  return"
      ],
      "metadata": {
        "id": "2EmydL_wKjLV"
      },
      "id": "2EmydL_wKjLV",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "The function prints a heatmap with the confusion matrix for a given model."
      ],
      "metadata": {
        "id": "NyHKiMj7yuEU"
      },
      "id": "NyHKiMj7yuEU"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Bagging"
      ],
      "metadata": {
        "id": "rq-dyuyXmnsG"
      },
      "id": "rq-dyuyXmnsG"
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Baseline: Decision Tree"
      ],
      "metadata": {
        "id": "ZxNLV1DPMKB6"
      },
      "id": "ZxNLV1DPMKB6"
    },
    {
      "cell_type": "code",
      "source": [
        "dtree=tree.DecisionTreeClassifier(random_state=1)\n",
        "dtree.fit(X_train,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 75
        },
        "id": "GRaA0HEoMO31",
        "outputId": "a2028ded-ecac-4181-f7a9-e94003b8825e"
      },
      "id": "GRaA0HEoMO31",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "DecisionTreeClassifier(random_state=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>DecisionTreeClassifier(random_state=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier(random_state=1)</pre></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 68
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "As a baseline comparison, we train a single decision tree."
      ],
      "metadata": {
        "id": "t4-JxFmwowRs"
      },
      "id": "t4-JxFmwowRs"
    },
    {
      "cell_type": "code",
      "source": [
        "scores(dtree)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "oqeHvfkfMfuN",
        "outputId": "875c6c2b-7d4b-49d2-96c2-c97c5591b89d"
      },
      "id": "oqeHvfkfMfuN",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "           Train      Test\n",
              "Accuracy     1.0  0.649902\n",
              "Precision    1.0  0.741256\n",
              "Recall       1.0  0.731229\n",
              "F1           1.0  0.736208"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-8d3c316e-8ade-4ae0-a8dc-1bbd28481501\">\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>Train</th>\n",
              "      <th>Test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Accuracy</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.649902</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Precision</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.741256</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Recall</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.731229</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>F1</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.736208</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8d3c316e-8ade-4ae0-a8dc-1bbd28481501')\"\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-8d3c316e-8ade-4ae0-a8dc-1bbd28481501 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-8d3c316e-8ade-4ae0-a8dc-1bbd28481501');\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": 69
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Predictably, this model is comically overfit."
      ],
      "metadata": {
        "id": "KyPG350poz1_"
      },
      "id": "KyPG350poz1_"
    },
    {
      "cell_type": "code",
      "source": [
        "tabulate(dtree,'dTree (baseline)')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 81
        },
        "id": "J0gMS3utNk7J",
        "outputId": "d61fa348-ecc0-4f41-9d71-42f751ef746b"
      },
      "id": "J0gMS3utNk7J",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                  Train Acc  Test Acc  Train F1   Test F1   Status\n",
              "dTree (baseline)        1.0  0.649902       1.0  0.736208  Overfit"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-c97e2614-d332-4806-9173-4239908beb27\">\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>Train Acc</th>\n",
              "      <th>Test Acc</th>\n",
              "      <th>Train F1</th>\n",
              "      <th>Test F1</th>\n",
              "      <th>Status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>dTree (baseline)</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.649902</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.736208</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c97e2614-d332-4806-9173-4239908beb27')\"\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-c97e2614-d332-4806-9173-4239908beb27 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-c97e2614-d332-4806-9173-4239908beb27');\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": 70
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We add this model to a DataFrame that we will use to compare the different models."
      ],
      "metadata": {
        "id": "RFRVW1HapqYG"
      },
      "id": "RFRVW1HapqYG"
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Bagging Classifier"
      ],
      "metadata": {
        "id": "7kEKiBjBmzTl"
      },
      "id": "7kEKiBjBmzTl"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "unknown-institution",
      "metadata": {
        "id": "unknown-institution",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 75
        },
        "outputId": "0d948351-e024-49ca-bdee-7e2c75807cec"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "BaggingClassifier(random_state=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-2 {color: black;background-color: white;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>BaggingClassifier(random_state=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" checked><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">BaggingClassifier</label><div class=\"sk-toggleable__content\"><pre>BaggingClassifier(random_state=1)</pre></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 71
        }
      ],
      "source": [
        "bag=BaggingClassifier(random_state=1)\n",
        "bag.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The default estimator on a bagging classifier is a decision tree classifier, so this model consists of many parallel decision trees trained on samples taken with replacement (bootstrap samples)."
      ],
      "metadata": {
        "id": "AJyhm9Nmp8x3"
      },
      "id": "AJyhm9Nmp8x3"
    },
    {
      "cell_type": "code",
      "source": [
        "scores(bag)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "z9-gRQ_gm-Fd",
        "outputId": "cb265cfb-5771-4cf5-ac77-e8a5b67d0d20"
      },
      "id": "z9-gRQ_gm-Fd",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              Train      Test\n",
              "Accuracy   0.984168  0.698232\n",
              "Precision  0.990625  0.770660\n",
              "Recall     0.985630  0.780631\n",
              "F1         0.988121  0.775614"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-d662615b-c379-4c5b-bcac-af12845640fc\">\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>Train</th>\n",
              "      <th>Test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Accuracy</th>\n",
              "      <td>0.984168</td>\n",
              "      <td>0.698232</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Precision</th>\n",
              "      <td>0.990625</td>\n",
              "      <td>0.770660</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Recall</th>\n",
              "      <td>0.985630</td>\n",
              "      <td>0.780631</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>F1</th>\n",
              "      <td>0.988121</td>\n",
              "      <td>0.775614</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-d662615b-c379-4c5b-bcac-af12845640fc')\"\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-d662615b-c379-4c5b-bcac-af12845640fc 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-d662615b-c379-4c5b-bcac-af12845640fc');\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": 72
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tabulate(bag,'Bagging Classifier')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 112
        },
        "id": "OMnQipoMN-kE",
        "outputId": "5cf498ee-c005-466e-8d21-93615ba7125f"
      },
      "id": "OMnQipoMN-kE",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                    Train Acc  Test Acc  Train F1   Test F1   Status\n",
              "dTree (baseline)     1.000000  0.649902  1.000000  0.736208  Overfit\n",
              "Bagging Classifier   0.984168  0.698232  0.988121  0.775614  Overfit"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-186ea120-891c-4b9d-a6ac-d29fc7a68087\">\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>Train Acc</th>\n",
              "      <th>Test Acc</th>\n",
              "      <th>Train F1</th>\n",
              "      <th>Test F1</th>\n",
              "      <th>Status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>dTree (baseline)</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.649902</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.736208</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging Classifier</th>\n",
              "      <td>0.984168</td>\n",
              "      <td>0.698232</td>\n",
              "      <td>0.988121</td>\n",
              "      <td>0.775614</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-186ea120-891c-4b9d-a6ac-d29fc7a68087')\"\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-186ea120-891c-4b9d-a6ac-d29fc7a68087 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-186ea120-891c-4b9d-a6ac-d29fc7a68087');\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": 73
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "This model is also overfit. The test performance is nonetheless slightly better than the single decision tree."
      ],
      "metadata": {
        "id": "clAMCwDfqa7x"
      },
      "id": "clAMCwDfqa7x"
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Random Forest Classifier"
      ],
      "metadata": {
        "id": "ijwUZPIonA_E"
      },
      "id": "ijwUZPIonA_E"
    },
    {
      "cell_type": "code",
      "source": [
        "rf=RandomForestClassifier(random_state=1)\n",
        "rf.fit(X_train,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 75
        },
        "id": "EADQ2fMgm-Cw",
        "outputId": "aa89cb64-e0e3-4c2b-a6dd-f581752545c1"
      },
      "id": "EADQ2fMgm-Cw",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(random_state=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-3 {color: black;background-color: white;}#sk-container-id-3 pre{padding: 0;}#sk-container-id-3 div.sk-toggleable {background-color: white;}#sk-container-id-3 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-3 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-3 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-3 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-3 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-3 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-3 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-3 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-3 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-3 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-3 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-3 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-3 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-3 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-3 div.sk-item {position: relative;z-index: 1;}#sk-container-id-3 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-3 div.sk-item::before, #sk-container-id-3 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-3 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-3 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-3 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-3 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-3 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-3 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-3 div.sk-label-container {text-align: center;}#sk-container-id-3 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-3 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-3\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomForestClassifier(random_state=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" checked><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier(random_state=1)</pre></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 74
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "scores(rf)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "wIZL3NH5m9_0",
        "outputId": "6cc771d2-2af2-4b79-fdd6-62dad7e16e1e"
      },
      "id": "wIZL3NH5m9_0",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "           Train      Test\n",
              "Accuracy     1.0  0.725082\n",
              "Precision    1.0  0.771920\n",
              "Recall       1.0  0.835326\n",
              "F1           1.0  0.802373"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-ba718bf3-bf71-4d28-91c3-f656d580b525\">\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>Train</th>\n",
              "      <th>Test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Accuracy</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.725082</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Precision</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.771920</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Recall</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.835326</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>F1</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.802373</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ba718bf3-bf71-4d28-91c3-f656d580b525')\"\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-ba718bf3-bf71-4d28-91c3-f656d580b525 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-ba718bf3-bf71-4d28-91c3-f656d580b525');\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": 75
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "As with the others, the random forest classifier overfits on the training data. Test performance is better than before, though this model is not generalized."
      ],
      "metadata": {
        "id": "EOoBbE-jzVSR"
      },
      "id": "EOoBbE-jzVSR"
    },
    {
      "cell_type": "code",
      "source": [
        "tabulate(rf,'Random Forest')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        },
        "id": "MKve2C7POJhY",
        "outputId": "0c4a7864-811f-4234-f173-e8caaa21aef8"
      },
      "id": "MKve2C7POJhY",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                    Train Acc  Test Acc  Train F1   Test F1   Status\n",
              "dTree (baseline)     1.000000  0.649902  1.000000  0.736208  Overfit\n",
              "Bagging Classifier   0.984168  0.698232  0.988121  0.775614  Overfit\n",
              "Random Forest        1.000000  0.725082  1.000000  0.802373  Overfit"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-7468a813-bfe2-425a-bba3-765f5599aa36\">\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>Train Acc</th>\n",
              "      <th>Test Acc</th>\n",
              "      <th>Train F1</th>\n",
              "      <th>Test F1</th>\n",
              "      <th>Status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>dTree (baseline)</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.649902</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.736208</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging Classifier</th>\n",
              "      <td>0.984168</td>\n",
              "      <td>0.698232</td>\n",
              "      <td>0.988121</td>\n",
              "      <td>0.775614</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.725082</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.802373</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-7468a813-bfe2-425a-bba3-765f5599aa36')\"\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-7468a813-bfe2-425a-bba3-765f5599aa36 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-7468a813-bfe2-425a-bba3-765f5599aa36');\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": 76
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Boosting"
      ],
      "metadata": {
        "id": "a4dgTVuqn6xH"
      },
      "id": "a4dgTVuqn6xH"
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### AdaBoost"
      ],
      "metadata": {
        "id": "K3q-FvDXn4R_"
      },
      "id": "K3q-FvDXn4R_"
    },
    {
      "cell_type": "code",
      "source": [
        "abc=AdaBoostClassifier(random_state=1)\n",
        "abc.fit(X_train,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 75
        },
        "id": "F8XdF1ODm98q",
        "outputId": "26bac9d2-65ba-47d5-f7d2-0d30d8494333"
      },
      "id": "F8XdF1ODm98q",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "AdaBoostClassifier(random_state=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-4 {color: black;background-color: white;}#sk-container-id-4 pre{padding: 0;}#sk-container-id-4 div.sk-toggleable {background-color: white;}#sk-container-id-4 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-4 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-4 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-4 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-4 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-4 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-4 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-4 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-4 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-4 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-4 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-4 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-4 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-4 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-4 div.sk-item {position: relative;z-index: 1;}#sk-container-id-4 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-4 div.sk-item::before, #sk-container-id-4 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-4 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-4 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-4 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-4 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-4 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-4 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-4 div.sk-label-container {text-align: center;}#sk-container-id-4 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-4 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-4\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>AdaBoostClassifier(random_state=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" checked><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">AdaBoostClassifier</label><div class=\"sk-toggleable__content\"><pre>AdaBoostClassifier(random_state=1)</pre></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 77
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "scores(abc)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "dSgxaywDm958",
        "outputId": "4afda85b-2411-42d3-f950-40a64f81a63f"
      },
      "id": "dSgxaywDm958",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              Train      Test\n",
              "Accuracy   0.734786  0.733857\n",
              "Precision  0.756469  0.755537\n",
              "Recall     0.889328  0.889433\n",
              "F1         0.817536  0.817036"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-d1b8a71e-c7e6-4550-817b-093a701c4a0f\">\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>Train</th>\n",
              "      <th>Test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Accuracy</th>\n",
              "      <td>0.734786</td>\n",
              "      <td>0.733857</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Precision</th>\n",
              "      <td>0.756469</td>\n",
              "      <td>0.755537</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Recall</th>\n",
              "      <td>0.889328</td>\n",
              "      <td>0.889433</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>F1</th>\n",
              "      <td>0.817536</td>\n",
              "      <td>0.817036</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-d1b8a71e-c7e6-4550-817b-093a701c4a0f')\"\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-d1b8a71e-c7e6-4550-817b-093a701c4a0f 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-d1b8a71e-c7e6-4550-817b-093a701c4a0f');\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": 78
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The basic AdaBoost Classifier has good performance and does not suffer from overfitting. Accuracy on training and testing is around 73% and F1 is about 82%. This is a much more promising baseline."
      ],
      "metadata": {
        "id": "eXyKTQ9G0DRq"
      },
      "id": "eXyKTQ9G0DRq"
    },
    {
      "cell_type": "code",
      "source": [
        "tabulate(abc,'AdaBoost')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "lZkWK7QfOOUU",
        "outputId": "fb443ed2-73c0-425f-c261-658dd10f0d32"
      },
      "id": "lZkWK7QfOOUU",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                    Train Acc  Test Acc  Train F1   Test F1   Status\n",
              "dTree (baseline)     1.000000  0.649902  1.000000  0.736208  Overfit\n",
              "Bagging Classifier   0.984168  0.698232  0.988121  0.775614  Overfit\n",
              "Random Forest        1.000000  0.725082  1.000000  0.802373  Overfit\n",
              "AdaBoost             0.734786  0.733857  0.817536  0.817036  General"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-826da3dc-e093-4ee8-91ab-e4f63dacfccc\">\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>Train Acc</th>\n",
              "      <th>Test Acc</th>\n",
              "      <th>Train F1</th>\n",
              "      <th>Test F1</th>\n",
              "      <th>Status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>dTree (baseline)</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.649902</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.736208</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging Classifier</th>\n",
              "      <td>0.984168</td>\n",
              "      <td>0.698232</td>\n",
              "      <td>0.988121</td>\n",
              "      <td>0.775614</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.725082</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.802373</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost</th>\n",
              "      <td>0.734786</td>\n",
              "      <td>0.733857</td>\n",
              "      <td>0.817536</td>\n",
              "      <td>0.817036</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-826da3dc-e093-4ee8-91ab-e4f63dacfccc')\"\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-826da3dc-e093-4ee8-91ab-e4f63dacfccc 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-826da3dc-e093-4ee8-91ab-e4f63dacfccc');\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": 79
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Gradient Boosting"
      ],
      "metadata": {
        "id": "jtmIXFVioEce"
      },
      "id": "jtmIXFVioEce"
    },
    {
      "cell_type": "code",
      "source": [
        "gbc=GradientBoostingClassifier(random_state=1)\n",
        "gbc.fit(X_train,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 75
        },
        "id": "6TnlfLAaoIz8",
        "outputId": "d7c03baa-0172-44b5-dcf1-9ea0a8c28721"
      },
      "id": "6TnlfLAaoIz8",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GradientBoostingClassifier(random_state=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-5 {color: black;background-color: white;}#sk-container-id-5 pre{padding: 0;}#sk-container-id-5 div.sk-toggleable {background-color: white;}#sk-container-id-5 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-5 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-5 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-5 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-5 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-5 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-5 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-5 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-5 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-5 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-5 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-5 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-5 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-5 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-5 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-5 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-5 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-5 div.sk-item {position: relative;z-index: 1;}#sk-container-id-5 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-5 div.sk-item::before, #sk-container-id-5 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-5 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-5 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-5 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-5 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-5 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-5 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-5 div.sk-label-container {text-align: center;}#sk-container-id-5 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-5 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-5\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GradientBoostingClassifier(random_state=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-5\" type=\"checkbox\" checked><label for=\"sk-estimator-id-5\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GradientBoostingClassifier</label><div class=\"sk-toggleable__content\"><pre>GradientBoostingClassifier(random_state=1)</pre></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 80
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "scores(gbc)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "73U_JdKdoIxT",
        "outputId": "49230ecd-1c7d-4dc3-ba1d-fb4a3c7c4043"
      },
      "id": "73U_JdKdoIxT",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              Train      Test\n",
              "Accuracy   0.754997  0.749312\n",
              "Precision  0.782289  0.780200\n",
              "Recall     0.877479  0.869829\n",
              "F1         0.827155  0.822581"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-d10982fe-4faf-4e6d-bee8-f9f8555d1884\">\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>Train</th>\n",
              "      <th>Test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Accuracy</th>\n",
              "      <td>0.754997</td>\n",
              "      <td>0.749312</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Precision</th>\n",
              "      <td>0.782289</td>\n",
              "      <td>0.780200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Recall</th>\n",
              "      <td>0.877479</td>\n",
              "      <td>0.869829</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>F1</th>\n",
              "      <td>0.827155</td>\n",
              "      <td>0.822581</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-d10982fe-4faf-4e6d-bee8-f9f8555d1884')\"\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-d10982fe-4faf-4e6d-bee8-f9f8555d1884 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-d10982fe-4faf-4e6d-bee8-f9f8555d1884');\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": 81
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Gradient Boosting also does not appear overfit, with the added bonus of slightly better performance over AdaBoost.\n",
        "* Train and test accuracy is around 75%.\n",
        "* Train and test F1 is around 82%."
      ],
      "metadata": {
        "id": "ot4jmucd0aL4"
      },
      "id": "ot4jmucd0aL4"
    },
    {
      "cell_type": "code",
      "source": [
        "tabulate(gbc,'Gradient Boosting')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "TOwnGrNNOS23",
        "outputId": "1f49e7bb-ae14-4f9d-a342-2e9539d0e482"
      },
      "id": "TOwnGrNNOS23",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                    Train Acc  Test Acc  Train F1   Test F1   Status\n",
              "dTree (baseline)     1.000000  0.649902  1.000000  0.736208  Overfit\n",
              "Bagging Classifier   0.984168  0.698232  0.988121  0.775614  Overfit\n",
              "Random Forest        1.000000  0.725082  1.000000  0.802373  Overfit\n",
              "AdaBoost             0.734786  0.733857  0.817536  0.817036  General\n",
              "Gradient Boosting    0.754997  0.749312  0.827155  0.822581  General"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-00772681-ee58-422a-aa95-12cee78948fc\">\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>Train Acc</th>\n",
              "      <th>Test Acc</th>\n",
              "      <th>Train F1</th>\n",
              "      <th>Test F1</th>\n",
              "      <th>Status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>dTree (baseline)</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.649902</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.736208</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging Classifier</th>\n",
              "      <td>0.984168</td>\n",
              "      <td>0.698232</td>\n",
              "      <td>0.988121</td>\n",
              "      <td>0.775614</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.725082</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.802373</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost</th>\n",
              "      <td>0.734786</td>\n",
              "      <td>0.733857</td>\n",
              "      <td>0.817536</td>\n",
              "      <td>0.817036</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Gradient Boosting</th>\n",
              "      <td>0.754997</td>\n",
              "      <td>0.749312</td>\n",
              "      <td>0.827155</td>\n",
              "      <td>0.822581</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-00772681-ee58-422a-aa95-12cee78948fc')\"\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-00772681-ee58-422a-aa95-12cee78948fc 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-00772681-ee58-422a-aa95-12cee78948fc');\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": 82
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### XGBoost"
      ],
      "metadata": {
        "id": "NOVeGYmToWe1"
      },
      "id": "NOVeGYmToWe1"
    },
    {
      "cell_type": "code",
      "source": [
        "xgbc=XGBClassifier(random_state=1)\n",
        "xgbc.fit(X_train,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 75
        },
        "id": "jLxVwUOWoIuA",
        "outputId": "8984d247-6e24-459c-ea40-2f993ce433f4"
      },
      "id": "jLxVwUOWoIuA",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "XGBClassifier(random_state=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-6 {color: black;background-color: white;}#sk-container-id-6 pre{padding: 0;}#sk-container-id-6 div.sk-toggleable {background-color: white;}#sk-container-id-6 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-6 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-6 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-6 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-6 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-6 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-6 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-6 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-6 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-6 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-6 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-6 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-6 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-6 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-6 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-6 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-6 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-6 div.sk-item {position: relative;z-index: 1;}#sk-container-id-6 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-6 div.sk-item::before, #sk-container-id-6 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-6 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-6 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-6 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-6 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-6 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-6 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-6 div.sk-label-container {text-align: center;}#sk-container-id-6 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-6 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-6\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>XGBClassifier(random_state=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-6\" type=\"checkbox\" checked><label for=\"sk-estimator-id-6\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(random_state=1)</pre></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 83
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "scores(xgbc)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "HtXSPfq-ofCD",
        "outputId": "f990e132-ed89-4293-d556-5b9649b7db9a"
      },
      "id": "HtXSPfq-ofCD",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              Train      Test\n",
              "Accuracy   0.753762  0.750098\n",
              "Precision  0.780624  0.779646\n",
              "Recall     0.878235  0.872574\n",
              "F1         0.826558  0.823497"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-18dfe198-4d2f-4371-9864-adfd84f15715\">\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>Train</th>\n",
              "      <th>Test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Accuracy</th>\n",
              "      <td>0.753762</td>\n",
              "      <td>0.750098</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Precision</th>\n",
              "      <td>0.780624</td>\n",
              "      <td>0.779646</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Recall</th>\n",
              "      <td>0.878235</td>\n",
              "      <td>0.872574</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>F1</th>\n",
              "      <td>0.826558</td>\n",
              "      <td>0.823497</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-18dfe198-4d2f-4371-9864-adfd84f15715')\"\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-18dfe198-4d2f-4371-9864-adfd84f15715 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-18dfe198-4d2f-4371-9864-adfd84f15715');\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": 84
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "XGBoost offers near identical performance to the gradient boosting above."
      ],
      "metadata": {
        "id": "NY0R6k53mzP8"
      },
      "id": "NY0R6k53mzP8"
    },
    {
      "cell_type": "code",
      "source": [
        "tabulate(xgbc,'XGBoost')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238
        },
        "id": "-FG0Bp1bOYSP",
        "outputId": "7f60bb93-d39b-4226-99ed-867904455ed9"
      },
      "id": "-FG0Bp1bOYSP",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                    Train Acc  Test Acc  Train F1   Test F1   Status\n",
              "dTree (baseline)     1.000000  0.649902  1.000000  0.736208  Overfit\n",
              "Bagging Classifier   0.984168  0.698232  0.988121  0.775614  Overfit\n",
              "Random Forest        1.000000  0.725082  1.000000  0.802373  Overfit\n",
              "AdaBoost             0.734786  0.733857  0.817536  0.817036  General\n",
              "Gradient Boosting    0.754997  0.749312  0.827155  0.822581  General\n",
              "XGBoost              0.753762  0.750098  0.826558  0.823497  General"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a920b2d6-3587-4567-ba29-76cc47bc46cb\">\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>Train Acc</th>\n",
              "      <th>Test Acc</th>\n",
              "      <th>Train F1</th>\n",
              "      <th>Test F1</th>\n",
              "      <th>Status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>dTree (baseline)</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.649902</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.736208</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging Classifier</th>\n",
              "      <td>0.984168</td>\n",
              "      <td>0.698232</td>\n",
              "      <td>0.988121</td>\n",
              "      <td>0.775614</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.725082</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.802373</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost</th>\n",
              "      <td>0.734786</td>\n",
              "      <td>0.733857</td>\n",
              "      <td>0.817536</td>\n",
              "      <td>0.817036</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Gradient Boosting</th>\n",
              "      <td>0.754997</td>\n",
              "      <td>0.749312</td>\n",
              "      <td>0.827155</td>\n",
              "      <td>0.822581</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBoost</th>\n",
              "      <td>0.753762</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.826558</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a920b2d6-3587-4567-ba29-76cc47bc46cb')\"\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-a920b2d6-3587-4567-ba29-76cc47bc46cb 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-a920b2d6-3587-4567-ba29-76cc47bc46cb');\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": 85
        }
      ]
    },
    {
      "cell_type": "markdown",
      "id": "prime-athletics",
      "metadata": {
        "id": "prime-athletics"
      },
      "source": [
        "##  Will tuning the hyperparameters improve the model performance?"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Bagging"
      ],
      "metadata": {
        "id": "7wYCv0lbFh3g"
      },
      "id": "7wYCv0lbFh3g"
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Tuned Decision Tree"
      ],
      "metadata": {
        "id": "Yhhn7CVBFslx"
      },
      "id": "Yhhn7CVBFslx"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "banned-difficulty",
      "metadata": {
        "id": "banned-difficulty"
      },
      "outputs": [],
      "source": [
        "dtree_tuned=tree.DecisionTreeClassifier(random_state=1)\n",
        "\n",
        "params={'max_depth':np.arange(3,10),\n",
        "        'min_samples_leaf':np.arange(5,10),\n",
        "        'max_features':[None,'sqrt'],\n",
        "        'max_leaf_nodes':np.arange(5,25,5),\n",
        "        'min_impurity_decrease':[0.0,0.0005,0.001],\n",
        "        'class_weight':[None,'balanced']}"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* We will test values of tree depth from 3 to 9.\n",
        "* We will consider limiting the maximum number of features to consider when calculating the best split.\n",
        "* We will try balancing the class weights, though I don't expect this will improve performance."
      ],
      "metadata": {
        "id": "iKD6FUP5kPxs"
      },
      "id": "iKD6FUP5kPxs"
    },
    {
      "cell_type": "code",
      "source": [
        "go=GridSearchCV(estimator=dtree_tuned,\n",
        "                param_grid=params,\n",
        "                scoring='f1',\n",
        "                n_jobs=-1,\n",
        "                cv=5,\n",
        "                verbose=1)\n",
        "go.fit(X_train,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 135
        },
        "id": "UOf6QwdoGDqv",
        "outputId": "b54a3b37-2395-46be-f2da-2c7b98a7d94e"
      },
      "id": "UOf6QwdoGDqv",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Fitting 5 folds for each of 1680 candidates, totalling 8400 fits\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSearchCV(cv=5, estimator=DecisionTreeClassifier(random_state=1), n_jobs=-1,\n",
              "             param_grid={'class_weight': [None, 'balanced'],\n",
              "                         'max_depth': array([3, 4, 5, 6, 7, 8, 9]),\n",
              "                         'max_features': [None, 'sqrt'],\n",
              "                         'max_leaf_nodes': array([ 5, 10, 15, 20]),\n",
              "                         'min_impurity_decrease': [0.0, 0.0005, 0.001],\n",
              "                         'min_samples_leaf': array([5, 6, 7, 8, 9])},\n",
              "             scoring='f1', verbose=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-7 {color: black;background-color: white;}#sk-container-id-7 pre{padding: 0;}#sk-container-id-7 div.sk-toggleable {background-color: white;}#sk-container-id-7 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-7 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-7 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-7 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-7 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-7 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-7 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-7 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-7 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-7 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-7 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-7 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-7 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-7 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-7 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-7 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-7 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-7 div.sk-item {position: relative;z-index: 1;}#sk-container-id-7 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-7 div.sk-item::before, #sk-container-id-7 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-7 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-7 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-7 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-7 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-7 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-7 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-7 div.sk-label-container {text-align: center;}#sk-container-id-7 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-7 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-7\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(cv=5, estimator=DecisionTreeClassifier(random_state=1), n_jobs=-1,\n",
              "             param_grid={&#x27;class_weight&#x27;: [None, &#x27;balanced&#x27;],\n",
              "                         &#x27;max_depth&#x27;: array([3, 4, 5, 6, 7, 8, 9]),\n",
              "                         &#x27;max_features&#x27;: [None, &#x27;sqrt&#x27;],\n",
              "                         &#x27;max_leaf_nodes&#x27;: array([ 5, 10, 15, 20]),\n",
              "                         &#x27;min_impurity_decrease&#x27;: [0.0, 0.0005, 0.001],\n",
              "                         &#x27;min_samples_leaf&#x27;: array([5, 6, 7, 8, 9])},\n",
              "             scoring=&#x27;f1&#x27;, verbose=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-7\" type=\"checkbox\" ><label for=\"sk-estimator-id-7\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GridSearchCV</label><div class=\"sk-toggleable__content\"><pre>GridSearchCV(cv=5, estimator=DecisionTreeClassifier(random_state=1), n_jobs=-1,\n",
              "             param_grid={&#x27;class_weight&#x27;: [None, &#x27;balanced&#x27;],\n",
              "                         &#x27;max_depth&#x27;: array([3, 4, 5, 6, 7, 8, 9]),\n",
              "                         &#x27;max_features&#x27;: [None, &#x27;sqrt&#x27;],\n",
              "                         &#x27;max_leaf_nodes&#x27;: array([ 5, 10, 15, 20]),\n",
              "                         &#x27;min_impurity_decrease&#x27;: [0.0, 0.0005, 0.001],\n",
              "                         &#x27;min_samples_leaf&#x27;: array([5, 6, 7, 8, 9])},\n",
              "             scoring=&#x27;f1&#x27;, verbose=1)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-8\" type=\"checkbox\" ><label for=\"sk-estimator-id-8\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier(random_state=1)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-9\" type=\"checkbox\" ><label for=\"sk-estimator-id-9\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier(random_state=1)</pre></div></div></div></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 87
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "go.best_params_"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nwa8W4TNH-Im",
        "outputId": "2aa1119a-6fcf-499d-a047-5b94eaa3a322"
      },
      "id": "nwa8W4TNH-Im",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'class_weight': None,\n",
              " 'max_depth': 7,\n",
              " 'max_features': None,\n",
              " 'max_leaf_nodes': 20,\n",
              " 'min_impurity_decrease': 0.0,\n",
              " 'min_samples_leaf': 5}"
            ]
          },
          "metadata": {},
          "execution_count": 88
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Grid Search Findings\n",
        "* Our response variable is not so unbalanced that a rebalancing of class weights is required.\n",
        "* A maximum tree depth of 7 seems to be ideal, with 20 as the upper bound for leaf nodes.\n",
        "* The model requires at least 5 samples for each leaf. Along with limiting the depth and maximum number of leaves, this reduces overfitting.\n",
        "* We get best performance when considering all features to determine the best split."
      ],
      "metadata": {
        "id": "JsZXlrv32-c8"
      },
      "id": "JsZXlrv32-c8"
    },
    {
      "cell_type": "code",
      "source": [
        "dtree_tuned=go.best_estimator_\n",
        "\n",
        "dtree_tuned.fit(X_train,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 92
        },
        "id": "7PBKlY9g0baz",
        "outputId": "81437f1e-157b-45b4-ea76-599ec7b46217"
      },
      "id": "7PBKlY9g0baz",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "DecisionTreeClassifier(max_depth=7, max_leaf_nodes=20, min_samples_leaf=5,\n",
              "                       random_state=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-8 {color: black;background-color: white;}#sk-container-id-8 pre{padding: 0;}#sk-container-id-8 div.sk-toggleable {background-color: white;}#sk-container-id-8 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-8 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-8 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-8 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-8 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-8 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-8 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-8 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-8 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-8 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-8 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-8 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-8 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-8 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-8 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-8 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-8 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-8 div.sk-item {position: relative;z-index: 1;}#sk-container-id-8 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-8 div.sk-item::before, #sk-container-id-8 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-8 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-8 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-8 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-8 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-8 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-8 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-8 div.sk-label-container {text-align: center;}#sk-container-id-8 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-8 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-8\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>DecisionTreeClassifier(max_depth=7, max_leaf_nodes=20, min_samples_leaf=5,\n",
              "                       random_state=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-10\" type=\"checkbox\" checked><label for=\"sk-estimator-id-10\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier(max_depth=7, max_leaf_nodes=20, min_samples_leaf=5,\n",
              "                       random_state=1)</pre></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 89
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We refit the best estimator on the full training set. (In the grid search, we employ cross validation, so there is a holdout set used to test each model. The upside is a more reliable search, but the cost is that we have yet to train the best estimator on the full training set.)"
      ],
      "metadata": {
        "id": "-t8REJ-elCjQ"
      },
      "id": "-t8REJ-elCjQ"
    },
    {
      "cell_type": "code",
      "source": [
        "scores(dtree_tuned)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "SAYQ5m16II0W",
        "outputId": "a2b59de8-892a-48ca-bebc-f775c6a313f6"
      },
      "id": "SAYQ5m16II0W",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              Train      Test\n",
              "Accuracy   0.749382  0.750098\n",
              "Precision  0.777421  0.779646\n",
              "Recall     0.875546  0.872574\n",
              "F1         0.823571  0.823497"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-90349d09-5423-4e23-8a34-33f0f5a805e3\">\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>Train</th>\n",
              "      <th>Test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Accuracy</th>\n",
              "      <td>0.749382</td>\n",
              "      <td>0.750098</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Precision</th>\n",
              "      <td>0.777421</td>\n",
              "      <td>0.779646</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Recall</th>\n",
              "      <td>0.875546</td>\n",
              "      <td>0.872574</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>F1</th>\n",
              "      <td>0.823571</td>\n",
              "      <td>0.823497</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-90349d09-5423-4e23-8a34-33f0f5a805e3')\"\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-90349d09-5423-4e23-8a34-33f0f5a805e3 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-90349d09-5423-4e23-8a34-33f0f5a805e3');\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": 90
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tabulate(dtree_tuned,'dTree (tuned)')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "id": "8cXyWbtO0Zrb",
        "outputId": "f0cd5279-e127-4537-e796-3be5c600fa3b"
      },
      "id": "8cXyWbtO0Zrb",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                    Train Acc  Test Acc  Train F1   Test F1   Status\n",
              "dTree (baseline)     1.000000  0.649902  1.000000  0.736208  Overfit\n",
              "Bagging Classifier   0.984168  0.698232  0.988121  0.775614  Overfit\n",
              "Random Forest        1.000000  0.725082  1.000000  0.802373  Overfit\n",
              "AdaBoost             0.734786  0.733857  0.817536  0.817036  General\n",
              "Gradient Boosting    0.754997  0.749312  0.827155  0.822581  General\n",
              "XGBoost              0.753762  0.750098  0.826558  0.823497  General\n",
              "dTree (tuned)        0.749382  0.750098  0.823571  0.823497  General"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-8747f689-e89a-47f6-b84b-4c4f767def29\">\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>Train Acc</th>\n",
              "      <th>Test Acc</th>\n",
              "      <th>Train F1</th>\n",
              "      <th>Test F1</th>\n",
              "      <th>Status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>dTree (baseline)</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.649902</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.736208</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging Classifier</th>\n",
              "      <td>0.984168</td>\n",
              "      <td>0.698232</td>\n",
              "      <td>0.988121</td>\n",
              "      <td>0.775614</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.725082</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.802373</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost</th>\n",
              "      <td>0.734786</td>\n",
              "      <td>0.733857</td>\n",
              "      <td>0.817536</td>\n",
              "      <td>0.817036</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Gradient Boosting</th>\n",
              "      <td>0.754997</td>\n",
              "      <td>0.749312</td>\n",
              "      <td>0.827155</td>\n",
              "      <td>0.822581</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBoost</th>\n",
              "      <td>0.753762</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.826558</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>dTree (tuned)</th>\n",
              "      <td>0.749382</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.823571</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8747f689-e89a-47f6-b84b-4c4f767def29')\"\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-8747f689-e89a-47f6-b84b-4c4f767def29 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-8747f689-e89a-47f6-b84b-4c4f767def29');\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": 91
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "With around 75% accuracy, the tuned decision tree has similar performance to untuned boosting models. F1 score is around 82%. Importantly, tuning this tree has eliminated overfitting."
      ],
      "metadata": {
        "id": "srhmxkXM3qjx"
      },
      "id": "srhmxkXM3qjx"
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Tuned Bagging Classifier"
      ],
      "metadata": {
        "id": "BfZwNqvgsK4I"
      },
      "id": "BfZwNqvgsK4I"
    },
    {
      "cell_type": "code",
      "source": [
        "params={'base_estimator':[tree.DecisionTreeClassifier(random_state=2)],\n",
        "        'base_estimator__max_depth':[None,1,2,3,4,5],\n",
        "        'n_estimators':[10,20,30,40,50],\n",
        "        'max_samples':[0.7,0.8,0.9,1.0],\n",
        "        'max_features':[0.7,0.8,0.9,1.0],}"
      ],
      "metadata": {
        "id": "kvtyHI7fszix"
      },
      "id": "kvtyHI7fszix",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "* We will test different depths for the base estimator (Decision Tree Classifier).\n",
        "* We will consider between 10 and 50 estimators.\n",
        "* Max samples and max features will also be adjusted, testing different fractions to include."
      ],
      "metadata": {
        "id": "7P6GifmERS4w"
      },
      "id": "7P6GifmERS4w"
    },
    {
      "cell_type": "code",
      "source": [
        "bag_tuned=BaggingClassifier(random_state=1)\n",
        "\n",
        "go=GridSearchCV(estimator=bag_tuned,\n",
        "                param_grid=params,\n",
        "                scoring='f1',\n",
        "                n_jobs=-1,\n",
        "                cv=5,\n",
        "                verbose=1)\n",
        "\n",
        "go.fit(X_train,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 135
        },
        "id": "AoI3vy0gtn4V",
        "outputId": "26ecdf7f-c169-4e73-cbe4-a30c6dc6c42d"
      },
      "id": "AoI3vy0gtn4V",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Fitting 5 folds for each of 480 candidates, totalling 2400 fits\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSearchCV(cv=5, estimator=BaggingClassifier(random_state=1), n_jobs=-1,\n",
              "             param_grid={'base_estimator': [DecisionTreeClassifier(max_depth=5,\n",
              "                                                                   random_state=2)],\n",
              "                         'base_estimator__max_depth': [None, 1, 2, 3, 4, 5],\n",
              "                         'max_features': [0.7, 0.8, 0.9, 1.0],\n",
              "                         'max_samples': [0.7, 0.8, 0.9, 1.0],\n",
              "                         'n_estimators': [10, 20, 30, 40, 50]},\n",
              "             scoring='f1', verbose=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-9 {color: black;background-color: white;}#sk-container-id-9 pre{padding: 0;}#sk-container-id-9 div.sk-toggleable {background-color: white;}#sk-container-id-9 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-9 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-9 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-9 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-9 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-9 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-9 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-9 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-9 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-9 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-9 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-9 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-9 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-9 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-9 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-9 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-9 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-9 div.sk-item {position: relative;z-index: 1;}#sk-container-id-9 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-9 div.sk-item::before, #sk-container-id-9 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-9 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-9 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-9 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-9 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-9 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-9 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-9 div.sk-label-container {text-align: center;}#sk-container-id-9 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-9 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-9\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(cv=5, estimator=BaggingClassifier(random_state=1), n_jobs=-1,\n",
              "             param_grid={&#x27;base_estimator&#x27;: [DecisionTreeClassifier(max_depth=5,\n",
              "                                                                   random_state=2)],\n",
              "                         &#x27;base_estimator__max_depth&#x27;: [None, 1, 2, 3, 4, 5],\n",
              "                         &#x27;max_features&#x27;: [0.7, 0.8, 0.9, 1.0],\n",
              "                         &#x27;max_samples&#x27;: [0.7, 0.8, 0.9, 1.0],\n",
              "                         &#x27;n_estimators&#x27;: [10, 20, 30, 40, 50]},\n",
              "             scoring=&#x27;f1&#x27;, verbose=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-11\" type=\"checkbox\" ><label for=\"sk-estimator-id-11\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GridSearchCV</label><div class=\"sk-toggleable__content\"><pre>GridSearchCV(cv=5, estimator=BaggingClassifier(random_state=1), n_jobs=-1,\n",
              "             param_grid={&#x27;base_estimator&#x27;: [DecisionTreeClassifier(max_depth=5,\n",
              "                                                                   random_state=2)],\n",
              "                         &#x27;base_estimator__max_depth&#x27;: [None, 1, 2, 3, 4, 5],\n",
              "                         &#x27;max_features&#x27;: [0.7, 0.8, 0.9, 1.0],\n",
              "                         &#x27;max_samples&#x27;: [0.7, 0.8, 0.9, 1.0],\n",
              "                         &#x27;n_estimators&#x27;: [10, 20, 30, 40, 50]},\n",
              "             scoring=&#x27;f1&#x27;, verbose=1)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-12\" type=\"checkbox\" ><label for=\"sk-estimator-id-12\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: BaggingClassifier</label><div class=\"sk-toggleable__content\"><pre>BaggingClassifier(random_state=1)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-13\" type=\"checkbox\" ><label for=\"sk-estimator-id-13\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">BaggingClassifier</label><div class=\"sk-toggleable__content\"><pre>BaggingClassifier(random_state=1)</pre></div></div></div></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 93
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "go.best_params_"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZnXGPydTvMfu",
        "outputId": "607a03bd-d7df-44f0-e266-3fc4ffb919c1"
      },
      "id": "ZnXGPydTvMfu",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'base_estimator': DecisionTreeClassifier(max_depth=5, random_state=2),\n",
              " 'base_estimator__max_depth': 5,\n",
              " 'max_features': 0.9,\n",
              " 'max_samples': 0.8,\n",
              " 'n_estimators': 10}"
            ]
          },
          "metadata": {},
          "execution_count": 94
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Trees with a depth of 5 score best on F1.\n",
        "* Our model performs better with fewer estimators.\n",
        "* Taking less than 100% of the samples yields a higher-scoring model."
      ],
      "metadata": {
        "id": "Gf12bCFhRoYt"
      },
      "id": "Gf12bCFhRoYt"
    },
    {
      "cell_type": "code",
      "source": [
        "bag_tuned=go.best_estimator_\n",
        "\n",
        "bag_tuned.fit(X_train,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 118
        },
        "id": "oGHuXI1-0udc",
        "outputId": "02f080bf-6817-4dfa-b2b7-b2c11cac117c"
      },
      "id": "oGHuXI1-0udc",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "BaggingClassifier(base_estimator=DecisionTreeClassifier(max_depth=5,\n",
              "                                                        random_state=2),\n",
              "                  max_features=0.9, max_samples=0.8, random_state=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-10 {color: black;background-color: white;}#sk-container-id-10 pre{padding: 0;}#sk-container-id-10 div.sk-toggleable {background-color: white;}#sk-container-id-10 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-10 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-10 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-10 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-10 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-10 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-10 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-10 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-10 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-10 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-10 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-10 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-10 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-10 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-10 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-10 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-10 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-10 div.sk-item {position: relative;z-index: 1;}#sk-container-id-10 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-10 div.sk-item::before, #sk-container-id-10 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-10 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-10 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-10 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-10 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-10 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-10 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-10 div.sk-label-container {text-align: center;}#sk-container-id-10 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-10 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-10\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>BaggingClassifier(base_estimator=DecisionTreeClassifier(max_depth=5,\n",
              "                                                        random_state=2),\n",
              "                  max_features=0.9, max_samples=0.8, random_state=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-14\" type=\"checkbox\" ><label for=\"sk-estimator-id-14\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">BaggingClassifier</label><div class=\"sk-toggleable__content\"><pre>BaggingClassifier(base_estimator=DecisionTreeClassifier(max_depth=5,\n",
              "                                                        random_state=2),\n",
              "                  max_features=0.9, max_samples=0.8, random_state=1)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-15\" type=\"checkbox\" ><label for=\"sk-estimator-id-15\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">base_estimator: DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier(max_depth=5, random_state=2)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-16\" type=\"checkbox\" ><label for=\"sk-estimator-id-16\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier(max_depth=5, random_state=2)</pre></div></div></div></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 95
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "scores(bag_tuned)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "WdpMRiVrvPb0",
        "outputId": "d4b5984b-ef9f-46e4-ec35-ac911646257f"
      },
      "id": "WdpMRiVrvPb0",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              Train      Test\n",
              "Accuracy   0.738098  0.735298\n",
              "Precision  0.747520  0.746795\n",
              "Recall     0.918067  0.913546\n",
              "F1         0.824062  0.821797"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-12093c28-9ca5-4bfb-b5e0-23f196d7e1be\">\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>Train</th>\n",
              "      <th>Test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Accuracy</th>\n",
              "      <td>0.738098</td>\n",
              "      <td>0.735298</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Precision</th>\n",
              "      <td>0.747520</td>\n",
              "      <td>0.746795</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Recall</th>\n",
              "      <td>0.918067</td>\n",
              "      <td>0.913546</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>F1</th>\n",
              "      <td>0.824062</td>\n",
              "      <td>0.821797</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-12093c28-9ca5-4bfb-b5e0-23f196d7e1be')\"\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-12093c28-9ca5-4bfb-b5e0-23f196d7e1be 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-12093c28-9ca5-4bfb-b5e0-23f196d7e1be');\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": 96
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Note that the tuned bagging classifier has great recall. This model does a great job at reducing false negatives, meaning the US job market is not losing out on promising talent. On the other hand, precision is lower, which leaves us with an F1 score comparable to other generalized models."
      ],
      "metadata": {
        "id": "al5mYAiP1-Qz"
      },
      "id": "al5mYAiP1-Qz"
    },
    {
      "cell_type": "code",
      "source": [
        "tabulate(bag_tuned,'Bagging clfr (tuned)')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "G7K98RXD1DVl",
        "outputId": "06a36460-6970-4b6d-a0f5-f6bc7d37b4a7"
      },
      "id": "G7K98RXD1DVl",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                      Train Acc  Test Acc  Train F1   Test F1   Status\n",
              "dTree (baseline)       1.000000  0.649902  1.000000  0.736208  Overfit\n",
              "Bagging Classifier     0.984168  0.698232  0.988121  0.775614  Overfit\n",
              "Random Forest          1.000000  0.725082  1.000000  0.802373  Overfit\n",
              "AdaBoost               0.734786  0.733857  0.817536  0.817036  General\n",
              "Gradient Boosting      0.754997  0.749312  0.827155  0.822581  General\n",
              "XGBoost                0.753762  0.750098  0.826558  0.823497  General\n",
              "dTree (tuned)          0.749382  0.750098  0.823571  0.823497  General\n",
              "Bagging clfr (tuned)   0.738098  0.735298  0.824062  0.821797  General"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-6b1ffdc3-98d5-4afd-88f2-e042dde789c9\">\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>Train Acc</th>\n",
              "      <th>Test Acc</th>\n",
              "      <th>Train F1</th>\n",
              "      <th>Test F1</th>\n",
              "      <th>Status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>dTree (baseline)</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.649902</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.736208</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging Classifier</th>\n",
              "      <td>0.984168</td>\n",
              "      <td>0.698232</td>\n",
              "      <td>0.988121</td>\n",
              "      <td>0.775614</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.725082</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.802373</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost</th>\n",
              "      <td>0.734786</td>\n",
              "      <td>0.733857</td>\n",
              "      <td>0.817536</td>\n",
              "      <td>0.817036</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Gradient Boosting</th>\n",
              "      <td>0.754997</td>\n",
              "      <td>0.749312</td>\n",
              "      <td>0.827155</td>\n",
              "      <td>0.822581</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBoost</th>\n",
              "      <td>0.753762</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.826558</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>dTree (tuned)</th>\n",
              "      <td>0.749382</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.823571</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging clfr (tuned)</th>\n",
              "      <td>0.738098</td>\n",
              "      <td>0.735298</td>\n",
              "      <td>0.824062</td>\n",
              "      <td>0.821797</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-6b1ffdc3-98d5-4afd-88f2-e042dde789c9')\"\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-6b1ffdc3-98d5-4afd-88f2-e042dde789c9 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-6b1ffdc3-98d5-4afd-88f2-e042dde789c9');\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": 97
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "As an experiment, we try another grid search with a second option for the base estimator: Logistic Regression."
      ],
      "metadata": {
        "id": "k3SKSEPJ1fJ0"
      },
      "id": "k3SKSEPJ1fJ0"
    },
    {
      "cell_type": "code",
      "source": [
        "params={'base_estimator':[tree.DecisionTreeClassifier(random_state=2),\n",
        "                          LogisticRegression(random_state=2,max_iter=1000)],\n",
        "        'n_estimators':[10,20,30,40,50],\n",
        "        'max_samples':[0.7,0.8,0.9,1.0],\n",
        "        'max_features':[0.7,0.8,0.9,1.0],}"
      ],
      "metadata": {
        "id": "2-QkrVVYvY6c"
      },
      "id": "2-QkrVVYvY6c",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "bag_tuned=BaggingClassifier(random_state=1)\n",
        "\n",
        "go=GridSearchCV(estimator=bag_tuned,\n",
        "                param_grid=params,\n",
        "                scoring='f1',\n",
        "                n_jobs=-1,\n",
        "                cv=5,\n",
        "                verbose=1)\n",
        "\n",
        "go.fit(X_train,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 135
        },
        "id": "_8wlJa1Tv9Uj",
        "outputId": "585dd445-9a78-4e3c-87fc-2f89a03dbc01"
      },
      "id": "_8wlJa1Tv9Uj",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Fitting 5 folds for each of 160 candidates, totalling 800 fits\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSearchCV(cv=5, estimator=BaggingClassifier(random_state=1), n_jobs=-1,\n",
              "             param_grid={'base_estimator': [DecisionTreeClassifier(random_state=2),\n",
              "                                            LogisticRegression(max_iter=1000,\n",
              "                                                               random_state=2)],\n",
              "                         'max_features': [0.7, 0.8, 0.9, 1.0],\n",
              "                         'max_samples': [0.7, 0.8, 0.9, 1.0],\n",
              "                         'n_estimators': [10, 20, 30, 40, 50]},\n",
              "             scoring='f1', verbose=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-11 {color: black;background-color: white;}#sk-container-id-11 pre{padding: 0;}#sk-container-id-11 div.sk-toggleable {background-color: white;}#sk-container-id-11 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-11 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-11 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-11 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-11 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-11 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-11 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-11 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-11 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-11 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-11 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-11 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-11 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-11 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-11 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-11 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-11 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-11 div.sk-item {position: relative;z-index: 1;}#sk-container-id-11 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-11 div.sk-item::before, #sk-container-id-11 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-11 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-11 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-11 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-11 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-11 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-11 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-11 div.sk-label-container {text-align: center;}#sk-container-id-11 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-11 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-11\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(cv=5, estimator=BaggingClassifier(random_state=1), n_jobs=-1,\n",
              "             param_grid={&#x27;base_estimator&#x27;: [DecisionTreeClassifier(random_state=2),\n",
              "                                            LogisticRegression(max_iter=1000,\n",
              "                                                               random_state=2)],\n",
              "                         &#x27;max_features&#x27;: [0.7, 0.8, 0.9, 1.0],\n",
              "                         &#x27;max_samples&#x27;: [0.7, 0.8, 0.9, 1.0],\n",
              "                         &#x27;n_estimators&#x27;: [10, 20, 30, 40, 50]},\n",
              "             scoring=&#x27;f1&#x27;, verbose=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-17\" type=\"checkbox\" ><label for=\"sk-estimator-id-17\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GridSearchCV</label><div class=\"sk-toggleable__content\"><pre>GridSearchCV(cv=5, estimator=BaggingClassifier(random_state=1), n_jobs=-1,\n",
              "             param_grid={&#x27;base_estimator&#x27;: [DecisionTreeClassifier(random_state=2),\n",
              "                                            LogisticRegression(max_iter=1000,\n",
              "                                                               random_state=2)],\n",
              "                         &#x27;max_features&#x27;: [0.7, 0.8, 0.9, 1.0],\n",
              "                         &#x27;max_samples&#x27;: [0.7, 0.8, 0.9, 1.0],\n",
              "                         &#x27;n_estimators&#x27;: [10, 20, 30, 40, 50]},\n",
              "             scoring=&#x27;f1&#x27;, verbose=1)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-18\" type=\"checkbox\" ><label for=\"sk-estimator-id-18\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: BaggingClassifier</label><div class=\"sk-toggleable__content\"><pre>BaggingClassifier(random_state=1)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-19\" type=\"checkbox\" ><label for=\"sk-estimator-id-19\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">BaggingClassifier</label><div class=\"sk-toggleable__content\"><pre>BaggingClassifier(random_state=1)</pre></div></div></div></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 99
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "go.best_params_"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5-4pEIVJwlEq",
        "outputId": "59ccc460-6f25-4021-f65f-aa6d8c91b9bb"
      },
      "id": "5-4pEIVJwlEq",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'base_estimator': DecisionTreeClassifier(random_state=2),\n",
              " 'max_features': 0.7,\n",
              " 'max_samples': 0.8,\n",
              " 'n_estimators': 50}"
            ]
          },
          "metadata": {},
          "execution_count": 100
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "After fitting the grid search, we find that the decision tree classifier was still the strongest base estimator for our metric. However, this time more estimators were required."
      ],
      "metadata": {
        "id": "kNmsKPH51nDz"
      },
      "id": "kNmsKPH51nDz"
    },
    {
      "cell_type": "code",
      "source": [
        "scores(go)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "SkP7EhTHwn9K",
        "outputId": "23f5d355-d13e-4d10-9841-26ace6bede53"
      },
      "id": "SkP7EhTHwn9K",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              Train      Test\n",
              "Accuracy   0.996351  0.722200\n",
              "Precision  0.995313  0.749916\n",
              "Recall     0.999244  0.876495\n",
              "F1         0.997274  0.808280"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-49cdfac5-2a28-4f86-804f-95d5bbbd4df1\">\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>Train</th>\n",
              "      <th>Test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Accuracy</th>\n",
              "      <td>0.996351</td>\n",
              "      <td>0.722200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Precision</th>\n",
              "      <td>0.995313</td>\n",
              "      <td>0.749916</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Recall</th>\n",
              "      <td>0.999244</td>\n",
              "      <td>0.876495</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>F1</th>\n",
              "      <td>0.997274</td>\n",
              "      <td>0.808280</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-49cdfac5-2a28-4f86-804f-95d5bbbd4df1')\"\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-49cdfac5-2a28-4f86-804f-95d5bbbd4df1 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-49cdfac5-2a28-4f86-804f-95d5bbbd4df1');\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": 101
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "It appears our model now overfits, as we have not limited the tree depth. We will not consider this model."
      ],
      "metadata": {
        "id": "iBRJKc-21vH9"
      },
      "id": "iBRJKc-21vH9"
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Tuned Random Forest"
      ],
      "metadata": {
        "id": "tF2Sz8y2wJVH"
      },
      "id": "tF2Sz8y2wJVH"
    },
    {
      "cell_type": "code",
      "source": [
        "params={'n_estimators':np.append(np.arange(150,350,50),1000),\n",
        "        'max_depth':np.arange(1,6),\n",
        "        'min_samples_split':np.arange(4,9),\n",
        "        'max_features':['sqrt','log2'],\n",
        "        'max_samples':[0.25,0.5,0.75,1.0]}"
      ],
      "metadata": {
        "id": "z96RaViKwJC1"
      },
      "id": "z96RaViKwJC1",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "* We consider a number of estimators between 150 and 300. We also consider a huge model: 1000 trees.\n",
        "* Various depths will be studied.\n",
        "* As with the previous bagging model, we will test various values for the maximum number of features and samples."
      ],
      "metadata": {
        "id": "C1JhNGBtUb7Y"
      },
      "id": "C1JhNGBtUb7Y"
    },
    {
      "cell_type": "code",
      "source": [
        "rf_tuned=RandomForestClassifier(random_state=1,warm_start=True)\n",
        "\n",
        "go=GridSearchCV(estimator=rf_tuned,\n",
        "                param_grid=params,\n",
        "                scoring='f1',\n",
        "                n_jobs=-1,\n",
        "                cv=5,\n",
        "                verbose=1)\n",
        "\n",
        "go.fit(X_train,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 135
        },
        "id": "2RLLslbkyNyM",
        "outputId": "c3abbc60-7752-4a3c-8a16-ecd0854ba090"
      },
      "id": "2RLLslbkyNyM",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Fitting 5 folds for each of 1000 candidates, totalling 5000 fits\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSearchCV(cv=5,\n",
              "             estimator=RandomForestClassifier(random_state=1, warm_start=True),\n",
              "             n_jobs=-1,\n",
              "             param_grid={'max_depth': array([1, 2, 3, 4, 5]),\n",
              "                         'max_features': ['sqrt', 'log2'],\n",
              "                         'max_samples': [0.25, 0.5, 0.75, 1.0],\n",
              "                         'min_samples_split': array([4, 5, 6, 7, 8]),\n",
              "                         'n_estimators': array([ 150,  200,  250,  300, 1000])},\n",
              "             scoring='f1', verbose=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-12 {color: black;background-color: white;}#sk-container-id-12 pre{padding: 0;}#sk-container-id-12 div.sk-toggleable {background-color: white;}#sk-container-id-12 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-12 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-12 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-12 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-12 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-12 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-12 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-12 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-12 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-12 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-12 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-12 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-12 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-12 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-12 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-12 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-12 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-12 div.sk-item {position: relative;z-index: 1;}#sk-container-id-12 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-12 div.sk-item::before, #sk-container-id-12 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-12 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-12 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-12 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-12 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-12 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-12 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-12 div.sk-label-container {text-align: center;}#sk-container-id-12 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-12 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-12\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(cv=5,\n",
              "             estimator=RandomForestClassifier(random_state=1, warm_start=True),\n",
              "             n_jobs=-1,\n",
              "             param_grid={&#x27;max_depth&#x27;: array([1, 2, 3, 4, 5]),\n",
              "                         &#x27;max_features&#x27;: [&#x27;sqrt&#x27;, &#x27;log2&#x27;],\n",
              "                         &#x27;max_samples&#x27;: [0.25, 0.5, 0.75, 1.0],\n",
              "                         &#x27;min_samples_split&#x27;: array([4, 5, 6, 7, 8]),\n",
              "                         &#x27;n_estimators&#x27;: array([ 150,  200,  250,  300, 1000])},\n",
              "             scoring=&#x27;f1&#x27;, verbose=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-20\" type=\"checkbox\" ><label for=\"sk-estimator-id-20\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GridSearchCV</label><div class=\"sk-toggleable__content\"><pre>GridSearchCV(cv=5,\n",
              "             estimator=RandomForestClassifier(random_state=1, warm_start=True),\n",
              "             n_jobs=-1,\n",
              "             param_grid={&#x27;max_depth&#x27;: array([1, 2, 3, 4, 5]),\n",
              "                         &#x27;max_features&#x27;: [&#x27;sqrt&#x27;, &#x27;log2&#x27;],\n",
              "                         &#x27;max_samples&#x27;: [0.25, 0.5, 0.75, 1.0],\n",
              "                         &#x27;min_samples_split&#x27;: array([4, 5, 6, 7, 8]),\n",
              "                         &#x27;n_estimators&#x27;: array([ 150,  200,  250,  300, 1000])},\n",
              "             scoring=&#x27;f1&#x27;, verbose=1)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-21\" type=\"checkbox\" ><label for=\"sk-estimator-id-21\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier(random_state=1, warm_start=True)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-22\" type=\"checkbox\" ><label for=\"sk-estimator-id-22\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier(random_state=1, warm_start=True)</pre></div></div></div></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 103
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "go.best_params_"
      ],
      "metadata": {
        "id": "yUrmvgxG0-of",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "b7c0276d-3ec6-4134-8469-2249eca03a71"
      },
      "id": "yUrmvgxG0-of",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'max_depth': 5,\n",
              " 'max_features': 'sqrt',\n",
              " 'max_samples': 0.75,\n",
              " 'min_samples_split': 7,\n",
              " 'n_estimators': 300}"
            ]
          },
          "metadata": {},
          "execution_count": 104
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Limiting the number of features at each split by the square root of the total number of features yielded better results than by using log base 2.\n",
        "* A maximum tree depth of 5 worked best.\n",
        "* We required 250 estimators in the highest scoring model."
      ],
      "metadata": {
        "id": "U6niOgWXU_Af"
      },
      "id": "U6niOgWXU_Af"
    },
    {
      "cell_type": "code",
      "source": [
        "# train best estimator omitting warm_start parameter\n",
        "rf_tuned=RandomForestClassifier(max_depth=5,\n",
        "                                max_features='sqrt',\n",
        "                                max_samples=0.75,\n",
        "                                min_samples_split=8,\n",
        "                                n_estimators=250,\n",
        "                                random_state=1)\n",
        "\n",
        "rf_tuned.fit(X_train,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 92
        },
        "id": "z-KnX-k9bURO",
        "outputId": "36064664-0dfd-477e-eaf2-a3a117dd169a"
      },
      "id": "z-KnX-k9bURO",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(max_depth=5, max_samples=0.75, min_samples_split=8,\n",
              "                       n_estimators=250, random_state=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-13 {color: black;background-color: white;}#sk-container-id-13 pre{padding: 0;}#sk-container-id-13 div.sk-toggleable {background-color: white;}#sk-container-id-13 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-13 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-13 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-13 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-13 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-13 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-13 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-13 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-13 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-13 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-13 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-13 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-13 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-13 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-13 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-13 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-13 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-13 div.sk-item {position: relative;z-index: 1;}#sk-container-id-13 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-13 div.sk-item::before, #sk-container-id-13 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-13 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-13 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-13 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-13 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-13 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-13 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-13 div.sk-label-container {text-align: center;}#sk-container-id-13 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-13 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-13\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomForestClassifier(max_depth=5, max_samples=0.75, min_samples_split=8,\n",
              "                       n_estimators=250, random_state=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-23\" type=\"checkbox\" checked><label for=\"sk-estimator-id-23\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier(max_depth=5, max_samples=0.75, min_samples_split=8,\n",
              "                       n_estimators=250, random_state=1)</pre></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 105
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "scores(rf_tuned)"
      ],
      "metadata": {
        "id": "f3wdjOpl08Iq",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "outputId": "cf489f28-dcf1-405e-89ab-b0f4a8445100"
      },
      "id": "f3wdjOpl08Iq",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              Train      Test\n",
              "Accuracy   0.725803  0.723772\n",
              "Precision  0.729101  0.727011\n",
              "Recall     0.938151  0.939228\n",
              "F1         0.820520  0.819605"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-b3919964-4d6b-4ab0-ac2a-d4fd25bbbd8b\">\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>Train</th>\n",
              "      <th>Test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Accuracy</th>\n",
              "      <td>0.725803</td>\n",
              "      <td>0.723772</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Precision</th>\n",
              "      <td>0.729101</td>\n",
              "      <td>0.727011</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Recall</th>\n",
              "      <td>0.938151</td>\n",
              "      <td>0.939228</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>F1</th>\n",
              "      <td>0.820520</td>\n",
              "      <td>0.819605</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-b3919964-4d6b-4ab0-ac2a-d4fd25bbbd8b')\"\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-b3919964-4d6b-4ab0-ac2a-d4fd25bbbd8b 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-b3919964-4d6b-4ab0-ac2a-d4fd25bbbd8b');\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": 106
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Similar to the tuned bagging classifier above, we're getting over 90% recall with this model. Unfortunately, accuracy is still below 75% and F1 is no higher than before."
      ],
      "metadata": {
        "id": "0OsQqVeFWF1V"
      },
      "id": "0OsQqVeFWF1V"
    },
    {
      "cell_type": "code",
      "source": [
        "tabulate(rf_tuned,'Random Forest (tuned)')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 332
        },
        "id": "VhjFfUp8bb3A",
        "outputId": "7dc85199-135f-4238-cb4f-4f073d147084"
      },
      "id": "VhjFfUp8bb3A",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                       Train Acc  Test Acc  Train F1   Test F1   Status\n",
              "dTree (baseline)        1.000000  0.649902  1.000000  0.736208  Overfit\n",
              "Bagging Classifier      0.984168  0.698232  0.988121  0.775614  Overfit\n",
              "Random Forest           1.000000  0.725082  1.000000  0.802373  Overfit\n",
              "AdaBoost                0.734786  0.733857  0.817536  0.817036  General\n",
              "Gradient Boosting       0.754997  0.749312  0.827155  0.822581  General\n",
              "XGBoost                 0.753762  0.750098  0.826558  0.823497  General\n",
              "dTree (tuned)           0.749382  0.750098  0.823571  0.823497  General\n",
              "Bagging clfr (tuned)    0.738098  0.735298  0.824062  0.821797  General\n",
              "Random Forest (tuned)   0.725803  0.723772  0.820520  0.819605  General"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-88b7cbdf-fe99-44f6-8d4a-575db526e077\">\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>Train Acc</th>\n",
              "      <th>Test Acc</th>\n",
              "      <th>Train F1</th>\n",
              "      <th>Test F1</th>\n",
              "      <th>Status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>dTree (baseline)</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.649902</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.736208</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging Classifier</th>\n",
              "      <td>0.984168</td>\n",
              "      <td>0.698232</td>\n",
              "      <td>0.988121</td>\n",
              "      <td>0.775614</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.725082</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.802373</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost</th>\n",
              "      <td>0.734786</td>\n",
              "      <td>0.733857</td>\n",
              "      <td>0.817536</td>\n",
              "      <td>0.817036</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Gradient Boosting</th>\n",
              "      <td>0.754997</td>\n",
              "      <td>0.749312</td>\n",
              "      <td>0.827155</td>\n",
              "      <td>0.822581</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBoost</th>\n",
              "      <td>0.753762</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.826558</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>dTree (tuned)</th>\n",
              "      <td>0.749382</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.823571</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging clfr (tuned)</th>\n",
              "      <td>0.738098</td>\n",
              "      <td>0.735298</td>\n",
              "      <td>0.824062</td>\n",
              "      <td>0.821797</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest (tuned)</th>\n",
              "      <td>0.725803</td>\n",
              "      <td>0.723772</td>\n",
              "      <td>0.820520</td>\n",
              "      <td>0.819605</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-88b7cbdf-fe99-44f6-8d4a-575db526e077')\"\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-88b7cbdf-fe99-44f6-8d4a-575db526e077 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-88b7cbdf-fe99-44f6-8d4a-575db526e077');\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": 107
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Boosting"
      ],
      "metadata": {
        "id": "cwGoAvY3lZ9c"
      },
      "id": "cwGoAvY3lZ9c"
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Tuned AdaBoost"
      ],
      "metadata": {
        "id": "D_WCWGywlfOp"
      },
      "id": "D_WCWGywlfOp"
    },
    {
      "cell_type": "code",
      "source": [
        "params={'base_estimator':[tree.DecisionTreeClassifier(random_state=2)],\n",
        "        'base_estimator__max_depth':[1,2,3],\n",
        "        'n_estimators':np.arange(10,80,10),\n",
        "        'learning_rate':np.linspace(0.1,1,10)}"
      ],
      "metadata": {
        "id": "blU6ewjOl_rS"
      },
      "id": "blU6ewjOl_rS",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Next we'll train an AdaBoost classifier, using a decision tree as the base estimator. We will consider between 10 and 70 estimators within the ensemble, various depths of the trees, and several different learning rates."
      ],
      "metadata": {
        "id": "z7WWoW2zW-ud"
      },
      "id": "z7WWoW2zW-ud"
    },
    {
      "cell_type": "code",
      "source": [
        "abc_tuned=AdaBoostClassifier(random_state=1)\n",
        "\n",
        "go=GridSearchCV(estimator=abc_tuned,\n",
        "                param_grid=params,\n",
        "                scoring='f1',\n",
        "                n_jobs=-1,\n",
        "                cv=5,\n",
        "                verbose=1)\n",
        "\n",
        "go.fit(X_train,y_train)"
      ],
      "metadata": {
        "id": "X-VgWmCnmaOX",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 135
        },
        "outputId": "70af74f0-05b0-4a09-cee8-bdd57f6ca537"
      },
      "id": "X-VgWmCnmaOX",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Fitting 5 folds for each of 210 candidates, totalling 1050 fits\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSearchCV(cv=5, estimator=AdaBoostClassifier(random_state=1), n_jobs=-1,\n",
              "             param_grid={'base_estimator': [DecisionTreeClassifier(max_depth=3,\n",
              "                                                                   random_state=2)],\n",
              "                         'base_estimator__max_depth': [1, 2, 3],\n",
              "                         'learning_rate': array([0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1. ]),\n",
              "                         'n_estimators': array([10, 20, 30, 40, 50, 60, 70])},\n",
              "             scoring='f1', verbose=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-14 {color: black;background-color: white;}#sk-container-id-14 pre{padding: 0;}#sk-container-id-14 div.sk-toggleable {background-color: white;}#sk-container-id-14 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-14 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-14 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-14 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-14 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-14 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-14 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-14 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-14 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-14 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-14 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-14 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-14 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-14 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-14 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-14 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-14 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-14 div.sk-item {position: relative;z-index: 1;}#sk-container-id-14 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-14 div.sk-item::before, #sk-container-id-14 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-14 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-14 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-14 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-14 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-14 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-14 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-14 div.sk-label-container {text-align: center;}#sk-container-id-14 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-14 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-14\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(cv=5, estimator=AdaBoostClassifier(random_state=1), n_jobs=-1,\n",
              "             param_grid={&#x27;base_estimator&#x27;: [DecisionTreeClassifier(max_depth=3,\n",
              "                                                                   random_state=2)],\n",
              "                         &#x27;base_estimator__max_depth&#x27;: [1, 2, 3],\n",
              "                         &#x27;learning_rate&#x27;: array([0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1. ]),\n",
              "                         &#x27;n_estimators&#x27;: array([10, 20, 30, 40, 50, 60, 70])},\n",
              "             scoring=&#x27;f1&#x27;, verbose=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-24\" type=\"checkbox\" ><label for=\"sk-estimator-id-24\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GridSearchCV</label><div class=\"sk-toggleable__content\"><pre>GridSearchCV(cv=5, estimator=AdaBoostClassifier(random_state=1), n_jobs=-1,\n",
              "             param_grid={&#x27;base_estimator&#x27;: [DecisionTreeClassifier(max_depth=3,\n",
              "                                                                   random_state=2)],\n",
              "                         &#x27;base_estimator__max_depth&#x27;: [1, 2, 3],\n",
              "                         &#x27;learning_rate&#x27;: array([0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1. ]),\n",
              "                         &#x27;n_estimators&#x27;: array([10, 20, 30, 40, 50, 60, 70])},\n",
              "             scoring=&#x27;f1&#x27;, verbose=1)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-25\" type=\"checkbox\" ><label for=\"sk-estimator-id-25\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: AdaBoostClassifier</label><div class=\"sk-toggleable__content\"><pre>AdaBoostClassifier(random_state=1)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-26\" type=\"checkbox\" ><label for=\"sk-estimator-id-26\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">AdaBoostClassifier</label><div class=\"sk-toggleable__content\"><pre>AdaBoostClassifier(random_state=1)</pre></div></div></div></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 109
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "go.best_params_"
      ],
      "metadata": {
        "id": "5CkHBoR7mnyW",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6fb2d370-0cc6-4fbe-bd3b-82718ea741ae"
      },
      "id": "5CkHBoR7mnyW",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'base_estimator': DecisionTreeClassifier(max_depth=3, random_state=2),\n",
              " 'base_estimator__max_depth': 3,\n",
              " 'learning_rate': 0.1,\n",
              " 'n_estimators': 50}"
            ]
          },
          "metadata": {},
          "execution_count": 110
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "With a depth of 3 and a slower learning rate, our best model required 50 estimators."
      ],
      "metadata": {
        "id": "g5GxTXqkXVqO"
      },
      "id": "g5GxTXqkXVqO"
    },
    {
      "cell_type": "code",
      "source": [
        "abc_tuned=go.best_estimator_\n",
        "\n",
        "abc_tuned.fit(X_train,y_train)"
      ],
      "metadata": {
        "id": "KZIRckXI4cvD",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 118
        },
        "outputId": "faf026d4-e8d8-4a30-810d-b7cf551747ef"
      },
      "id": "KZIRckXI4cvD",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "AdaBoostClassifier(base_estimator=DecisionTreeClassifier(max_depth=3,\n",
              "                                                         random_state=2),\n",
              "                   learning_rate=0.1, random_state=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-15 {color: black;background-color: white;}#sk-container-id-15 pre{padding: 0;}#sk-container-id-15 div.sk-toggleable {background-color: white;}#sk-container-id-15 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-15 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-15 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-15 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-15 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-15 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-15 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-15 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-15 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-15 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-15 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-15 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-15 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-15 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-15 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-15 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-15 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-15 div.sk-item {position: relative;z-index: 1;}#sk-container-id-15 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-15 div.sk-item::before, #sk-container-id-15 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-15 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-15 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-15 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-15 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-15 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-15 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-15 div.sk-label-container {text-align: center;}#sk-container-id-15 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-15 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-15\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>AdaBoostClassifier(base_estimator=DecisionTreeClassifier(max_depth=3,\n",
              "                                                         random_state=2),\n",
              "                   learning_rate=0.1, random_state=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-27\" type=\"checkbox\" ><label for=\"sk-estimator-id-27\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">AdaBoostClassifier</label><div class=\"sk-toggleable__content\"><pre>AdaBoostClassifier(base_estimator=DecisionTreeClassifier(max_depth=3,\n",
              "                                                         random_state=2),\n",
              "                   learning_rate=0.1, random_state=1)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-28\" type=\"checkbox\" ><label for=\"sk-estimator-id-28\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">base_estimator: DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier(max_depth=3, random_state=2)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-29\" type=\"checkbox\" ><label for=\"sk-estimator-id-29\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier(max_depth=3, random_state=2)</pre></div></div></div></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 111
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "scores(abc_tuned)"
      ],
      "metadata": {
        "id": "68hdXW9hmpPL",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "outputId": "d1684263-14e9-49d5-f2ad-875c8cb288ed"
      },
      "id": "68hdXW9hmpPL",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              Train      Test\n",
              "Accuracy   0.750561  0.749050\n",
              "Precision  0.778933  0.778166\n",
              "Recall     0.874958  0.873358\n",
              "F1         0.824158  0.823019"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-792871fc-94c6-4ed5-80e5-e6672c384e10\">\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>Train</th>\n",
              "      <th>Test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Accuracy</th>\n",
              "      <td>0.750561</td>\n",
              "      <td>0.749050</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Precision</th>\n",
              "      <td>0.778933</td>\n",
              "      <td>0.778166</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Recall</th>\n",
              "      <td>0.874958</td>\n",
              "      <td>0.873358</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>F1</th>\n",
              "      <td>0.824158</td>\n",
              "      <td>0.823019</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-792871fc-94c6-4ed5-80e5-e6672c384e10')\"\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-792871fc-94c6-4ed5-80e5-e6672c384e10 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-792871fc-94c6-4ed5-80e5-e6672c384e10');\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": 112
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We get similar F1 performance to the tuned random forest, but better accuracy. Recall is lower than some other models."
      ],
      "metadata": {
        "id": "lC1GTEDiXffM"
      },
      "id": "lC1GTEDiXffM"
    },
    {
      "cell_type": "code",
      "source": [
        "tabulate(abc_tuned,'AdaBoost (tuned)')"
      ],
      "metadata": {
        "id": "aaR_kAlk4cKs",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "outputId": "24a11f68-fecc-4bcd-8d99-ab8347dc7a52"
      },
      "id": "aaR_kAlk4cKs",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                       Train Acc  Test Acc  Train F1   Test F1   Status\n",
              "dTree (baseline)        1.000000  0.649902  1.000000  0.736208  Overfit\n",
              "Bagging Classifier      0.984168  0.698232  0.988121  0.775614  Overfit\n",
              "Random Forest           1.000000  0.725082  1.000000  0.802373  Overfit\n",
              "AdaBoost                0.734786  0.733857  0.817536  0.817036  General\n",
              "Gradient Boosting       0.754997  0.749312  0.827155  0.822581  General\n",
              "XGBoost                 0.753762  0.750098  0.826558  0.823497  General\n",
              "dTree (tuned)           0.749382  0.750098  0.823571  0.823497  General\n",
              "Bagging clfr (tuned)    0.738098  0.735298  0.824062  0.821797  General\n",
              "Random Forest (tuned)   0.725803  0.723772  0.820520  0.819605  General\n",
              "AdaBoost (tuned)        0.750561  0.749050  0.824158  0.823019  General"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-f21daacc-878c-4846-a7ba-0a1303e60083\">\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>Train Acc</th>\n",
              "      <th>Test Acc</th>\n",
              "      <th>Train F1</th>\n",
              "      <th>Test F1</th>\n",
              "      <th>Status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>dTree (baseline)</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.649902</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.736208</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging Classifier</th>\n",
              "      <td>0.984168</td>\n",
              "      <td>0.698232</td>\n",
              "      <td>0.988121</td>\n",
              "      <td>0.775614</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.725082</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.802373</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost</th>\n",
              "      <td>0.734786</td>\n",
              "      <td>0.733857</td>\n",
              "      <td>0.817536</td>\n",
              "      <td>0.817036</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Gradient Boosting</th>\n",
              "      <td>0.754997</td>\n",
              "      <td>0.749312</td>\n",
              "      <td>0.827155</td>\n",
              "      <td>0.822581</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBoost</th>\n",
              "      <td>0.753762</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.826558</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>dTree (tuned)</th>\n",
              "      <td>0.749382</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.823571</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging clfr (tuned)</th>\n",
              "      <td>0.738098</td>\n",
              "      <td>0.735298</td>\n",
              "      <td>0.824062</td>\n",
              "      <td>0.821797</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest (tuned)</th>\n",
              "      <td>0.725803</td>\n",
              "      <td>0.723772</td>\n",
              "      <td>0.820520</td>\n",
              "      <td>0.819605</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost (tuned)</th>\n",
              "      <td>0.750561</td>\n",
              "      <td>0.749050</td>\n",
              "      <td>0.824158</td>\n",
              "      <td>0.823019</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f21daacc-878c-4846-a7ba-0a1303e60083')\"\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-f21daacc-878c-4846-a7ba-0a1303e60083 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-f21daacc-878c-4846-a7ba-0a1303e60083');\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": 113
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Tuned Gradient Boosting"
      ],
      "metadata": {
        "id": "s2c61dRfnsud"
      },
      "id": "s2c61dRfnsud"
    },
    {
      "cell_type": "code",
      "source": [
        "params={'init':[None,'zero',AdaBoostClassifier(random_state=2)],\n",
        "        'n_estimators':[50,75,100,150],\n",
        "        'max_features':[None,'sqrt']+np.linspace(0.7,1.0,4).tolist()}"
      ],
      "metadata": {
        "id": "AKOorlEHnxCZ"
      },
      "id": "AKOorlEHnxCZ",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "We will next tune the gradient booster with several different initialization conditions. We will also check the number of estimators and the maximum number of features when looking for the best split."
      ],
      "metadata": {
        "id": "HgFe3YXZXptg"
      },
      "id": "HgFe3YXZXptg"
    },
    {
      "cell_type": "code",
      "source": [
        "gbc_tuned=GradientBoostingClassifier(random_state=1,\n",
        "                                     warm_start=True)\n",
        "\n",
        "go=GridSearchCV(estimator=gbc_tuned,\n",
        "                param_grid=params,\n",
        "                scoring='f1',\n",
        "                n_jobs=-1,\n",
        "                cv=5,\n",
        "                verbose=1)\n",
        "\n",
        "go.fit(X_train,y_train)"
      ],
      "metadata": {
        "id": "WwkTVB2LoN2Q",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 135
        },
        "outputId": "24a7822a-6189-471c-fd9b-3d396eb291f6"
      },
      "id": "WwkTVB2LoN2Q",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Fitting 5 folds for each of 72 candidates, totalling 360 fits\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSearchCV(cv=5,\n",
              "             estimator=GradientBoostingClassifier(random_state=1,\n",
              "                                                  warm_start=True),\n",
              "             n_jobs=-1,\n",
              "             param_grid={'init': [None, 'zero',\n",
              "                                  AdaBoostClassifier(random_state=2)],\n",
              "                         'max_features': [None, 'sqrt', 0.7, 0.7999999999999999,\n",
              "                                          0.9, 1.0],\n",
              "                         'n_estimators': [50, 75, 100, 150]},\n",
              "             scoring='f1', verbose=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-16 {color: black;background-color: white;}#sk-container-id-16 pre{padding: 0;}#sk-container-id-16 div.sk-toggleable {background-color: white;}#sk-container-id-16 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-16 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-16 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-16 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-16 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-16 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-16 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-16 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-16 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-16 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-16 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-16 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-16 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-16 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-16 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-16 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-16 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-16 div.sk-item {position: relative;z-index: 1;}#sk-container-id-16 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-16 div.sk-item::before, #sk-container-id-16 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-16 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-16 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-16 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-16 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-16 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-16 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-16 div.sk-label-container {text-align: center;}#sk-container-id-16 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-16 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-16\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(cv=5,\n",
              "             estimator=GradientBoostingClassifier(random_state=1,\n",
              "                                                  warm_start=True),\n",
              "             n_jobs=-1,\n",
              "             param_grid={&#x27;init&#x27;: [None, &#x27;zero&#x27;,\n",
              "                                  AdaBoostClassifier(random_state=2)],\n",
              "                         &#x27;max_features&#x27;: [None, &#x27;sqrt&#x27;, 0.7, 0.7999999999999999,\n",
              "                                          0.9, 1.0],\n",
              "                         &#x27;n_estimators&#x27;: [50, 75, 100, 150]},\n",
              "             scoring=&#x27;f1&#x27;, verbose=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-30\" type=\"checkbox\" ><label for=\"sk-estimator-id-30\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GridSearchCV</label><div class=\"sk-toggleable__content\"><pre>GridSearchCV(cv=5,\n",
              "             estimator=GradientBoostingClassifier(random_state=1,\n",
              "                                                  warm_start=True),\n",
              "             n_jobs=-1,\n",
              "             param_grid={&#x27;init&#x27;: [None, &#x27;zero&#x27;,\n",
              "                                  AdaBoostClassifier(random_state=2)],\n",
              "                         &#x27;max_features&#x27;: [None, &#x27;sqrt&#x27;, 0.7, 0.7999999999999999,\n",
              "                                          0.9, 1.0],\n",
              "                         &#x27;n_estimators&#x27;: [50, 75, 100, 150]},\n",
              "             scoring=&#x27;f1&#x27;, verbose=1)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-31\" type=\"checkbox\" ><label for=\"sk-estimator-id-31\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: GradientBoostingClassifier</label><div class=\"sk-toggleable__content\"><pre>GradientBoostingClassifier(random_state=1, warm_start=True)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-32\" type=\"checkbox\" ><label for=\"sk-estimator-id-32\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GradientBoostingClassifier</label><div class=\"sk-toggleable__content\"><pre>GradientBoostingClassifier(random_state=1, warm_start=True)</pre></div></div></div></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 115
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "go.best_params_"
      ],
      "metadata": {
        "id": "7rZupNsYoWRR",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "65e593ca-1a7a-489d-8194-e5b22deffc65"
      },
      "id": "7rZupNsYoWRR",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'init': 'zero', 'max_features': 0.7, 'n_estimators': 50}"
            ]
          },
          "metadata": {},
          "execution_count": 116
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# train best estimator omitting warm_start parameter\n",
        "gbc_tuned=GradientBoostingClassifier(random_state=1,\n",
        "                                     init=AdaBoostClassifier(random_state=2),\n",
        "                                     max_features=0.7,\n",
        "                                     n_estimators=100)\n",
        "\n",
        "gbc_tuned.fit(X_train,y_train)"
      ],
      "metadata": {
        "id": "mRqe4OIO4q-h",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 118
        },
        "outputId": "365bd4f6-f45c-4f3b-f036-bb94b166319a"
      },
      "id": "mRqe4OIO4q-h",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GradientBoostingClassifier(init=AdaBoostClassifier(random_state=2),\n",
              "                           max_features=0.7, random_state=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-17 {color: black;background-color: white;}#sk-container-id-17 pre{padding: 0;}#sk-container-id-17 div.sk-toggleable {background-color: white;}#sk-container-id-17 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-17 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-17 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-17 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-17 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-17 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-17 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-17 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-17 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-17 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-17 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-17 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-17 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-17 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-17 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-17 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-17 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-17 div.sk-item {position: relative;z-index: 1;}#sk-container-id-17 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-17 div.sk-item::before, #sk-container-id-17 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-17 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-17 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-17 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-17 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-17 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-17 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-17 div.sk-label-container {text-align: center;}#sk-container-id-17 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-17 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-17\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GradientBoostingClassifier(init=AdaBoostClassifier(random_state=2),\n",
              "                           max_features=0.7, random_state=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-33\" type=\"checkbox\" ><label for=\"sk-estimator-id-33\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GradientBoostingClassifier</label><div class=\"sk-toggleable__content\"><pre>GradientBoostingClassifier(init=AdaBoostClassifier(random_state=2),\n",
              "                           max_features=0.7, random_state=1)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-34\" type=\"checkbox\" ><label for=\"sk-estimator-id-34\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">init: AdaBoostClassifier</label><div class=\"sk-toggleable__content\"><pre>AdaBoostClassifier(random_state=2)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-35\" type=\"checkbox\" ><label for=\"sk-estimator-id-35\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">AdaBoostClassifier</label><div class=\"sk-toggleable__content\"><pre>AdaBoostClassifier(random_state=2)</pre></div></div></div></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 117
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "scores(gbc_tuned)"
      ],
      "metadata": {
        "id": "iNYNLWrzoYGm",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "outputId": "d99f5cc0-0c7a-4d49-9134-57d0a6dcf20d"
      },
      "id": "iNYNLWrzoYGm",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              Train      Test\n",
              "Accuracy   0.754379  0.751801\n",
              "Precision  0.782449  0.782020\n",
              "Recall     0.875882  0.871398\n",
              "F1         0.826533  0.824293"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a67fe10e-1fce-49a4-8fee-01b537354c3d\">\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>Train</th>\n",
              "      <th>Test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Accuracy</th>\n",
              "      <td>0.754379</td>\n",
              "      <td>0.751801</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Precision</th>\n",
              "      <td>0.782449</td>\n",
              "      <td>0.782020</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Recall</th>\n",
              "      <td>0.875882</td>\n",
              "      <td>0.871398</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>F1</th>\n",
              "      <td>0.826533</td>\n",
              "      <td>0.824293</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a67fe10e-1fce-49a4-8fee-01b537354c3d')\"\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-a67fe10e-1fce-49a4-8fee-01b537354c3d 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-a67fe10e-1fce-49a4-8fee-01b537354c3d');\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": 118
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "As with AdaBoost, our accuracy is over 75%. The F1 score does not seem to be able to improve above 82%."
      ],
      "metadata": {
        "id": "N5RntnrUYmm7"
      },
      "id": "N5RntnrUYmm7"
    },
    {
      "cell_type": "code",
      "source": [
        "tabulate(gbc_tuned,'Grad Boost (tuned)')"
      ],
      "metadata": {
        "id": "4oKx6N6L4xq8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 394
        },
        "outputId": "9ce68fac-255a-4e90-de2a-dcdc6b0bb59b"
      },
      "id": "4oKx6N6L4xq8",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                       Train Acc  Test Acc  Train F1   Test F1   Status\n",
              "dTree (baseline)        1.000000  0.649902  1.000000  0.736208  Overfit\n",
              "Bagging Classifier      0.984168  0.698232  0.988121  0.775614  Overfit\n",
              "Random Forest           1.000000  0.725082  1.000000  0.802373  Overfit\n",
              "AdaBoost                0.734786  0.733857  0.817536  0.817036  General\n",
              "Gradient Boosting       0.754997  0.749312  0.827155  0.822581  General\n",
              "XGBoost                 0.753762  0.750098  0.826558  0.823497  General\n",
              "dTree (tuned)           0.749382  0.750098  0.823571  0.823497  General\n",
              "Bagging clfr (tuned)    0.738098  0.735298  0.824062  0.821797  General\n",
              "Random Forest (tuned)   0.725803  0.723772  0.820520  0.819605  General\n",
              "AdaBoost (tuned)        0.750561  0.749050  0.824158  0.823019  General\n",
              "Grad Boost (tuned)      0.754379  0.751801  0.826533  0.824293  General"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-1c88c852-bcce-48bb-88e6-85c756046ba7\">\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>Train Acc</th>\n",
              "      <th>Test Acc</th>\n",
              "      <th>Train F1</th>\n",
              "      <th>Test F1</th>\n",
              "      <th>Status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>dTree (baseline)</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.649902</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.736208</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging Classifier</th>\n",
              "      <td>0.984168</td>\n",
              "      <td>0.698232</td>\n",
              "      <td>0.988121</td>\n",
              "      <td>0.775614</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.725082</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.802373</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost</th>\n",
              "      <td>0.734786</td>\n",
              "      <td>0.733857</td>\n",
              "      <td>0.817536</td>\n",
              "      <td>0.817036</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Gradient Boosting</th>\n",
              "      <td>0.754997</td>\n",
              "      <td>0.749312</td>\n",
              "      <td>0.827155</td>\n",
              "      <td>0.822581</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBoost</th>\n",
              "      <td>0.753762</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.826558</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>dTree (tuned)</th>\n",
              "      <td>0.749382</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.823571</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging clfr (tuned)</th>\n",
              "      <td>0.738098</td>\n",
              "      <td>0.735298</td>\n",
              "      <td>0.824062</td>\n",
              "      <td>0.821797</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest (tuned)</th>\n",
              "      <td>0.725803</td>\n",
              "      <td>0.723772</td>\n",
              "      <td>0.820520</td>\n",
              "      <td>0.819605</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost (tuned)</th>\n",
              "      <td>0.750561</td>\n",
              "      <td>0.749050</td>\n",
              "      <td>0.824158</td>\n",
              "      <td>0.823019</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Grad Boost (tuned)</th>\n",
              "      <td>0.754379</td>\n",
              "      <td>0.751801</td>\n",
              "      <td>0.826533</td>\n",
              "      <td>0.824293</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1c88c852-bcce-48bb-88e6-85c756046ba7')\"\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-1c88c852-bcce-48bb-88e6-85c756046ba7 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-1c88c852-bcce-48bb-88e6-85c756046ba7');\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": 119
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Tuned XGBoost"
      ],
      "metadata": {
        "id": "WxsPtYuTp9S4"
      },
      "id": "WxsPtYuTp9S4"
    },
    {
      "cell_type": "code",
      "source": [
        "params={'eta':[0.1,0.2,0.3],\n",
        "        'gamma':[0,2],\n",
        "        'subsample':[0.5,0.75,1.0],\n",
        "        'colsample_by_tree':[0.5,0.75,1.0],\n",
        "        'colsample_bylevel':[0.5,0.75,1.0],\n",
        "        'scale_pos_weight':[0.5,1.0]}"
      ],
      "metadata": {
        "id": "bW4SuH0As4sw"
      },
      "id": "bW4SuH0As4sw",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "We will try many parameters for XGBoost.\n",
        "* The parameter eta is the learning rate for the model.\n",
        "* The parameter gamma concern loss reduction when partitioning a leaf node on a tree.\n",
        "* Several sampling parameters control how the model samples the training data during fitting.\n",
        "* The last parameter (```scale_pos_weight```) can help with biased classes in the response variable."
      ],
      "metadata": {
        "id": "Z79wJl01YwoV"
      },
      "id": "Z79wJl01YwoV"
    },
    {
      "cell_type": "code",
      "source": [
        "xgbc_tuned=XGBClassifier(random_state=1)\n",
        "\n",
        "go=GridSearchCV(estimator=xgbc_tuned,\n",
        "                param_grid=params,\n",
        "                scoring='f1',\n",
        "                cv=5,\n",
        "                verbose=1)\n",
        "\n",
        "go.fit(X_train,y_train)"
      ],
      "metadata": {
        "id": "LNP6jSd-qAB6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 135
        },
        "outputId": "d4c8aa95-2bf2-4a7a-ca19-7b00393c714d"
      },
      "id": "LNP6jSd-qAB6",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Fitting 5 folds for each of 324 candidates, totalling 1620 fits\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSearchCV(cv=5, estimator=XGBClassifier(random_state=1),\n",
              "             param_grid={'colsample_by_tree': [0.5, 0.75, 1.0],\n",
              "                         'colsample_bylevel': [0.5, 0.75, 1.0],\n",
              "                         'eta': [0.1, 0.2, 0.3], 'gamma': [0, 2],\n",
              "                         'scale_pos_weight': [0.5, 1.0],\n",
              "                         'subsample': [0.5, 0.75, 1.0]},\n",
              "             scoring='f1', verbose=1)"
            ],
            "text/html": [
              "<style>#sk-container-id-18 {color: black;background-color: white;}#sk-container-id-18 pre{padding: 0;}#sk-container-id-18 div.sk-toggleable {background-color: white;}#sk-container-id-18 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-18 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-18 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-18 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-18 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-18 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-18 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-18 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-18 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-18 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-18 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-18 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-18 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-18 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-18 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-18 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-18 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-18 div.sk-item {position: relative;z-index: 1;}#sk-container-id-18 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-18 div.sk-item::before, #sk-container-id-18 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-18 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-18 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-18 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-18 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-18 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-18 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-18 div.sk-label-container {text-align: center;}#sk-container-id-18 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-18 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-18\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(cv=5, estimator=XGBClassifier(random_state=1),\n",
              "             param_grid={&#x27;colsample_by_tree&#x27;: [0.5, 0.75, 1.0],\n",
              "                         &#x27;colsample_bylevel&#x27;: [0.5, 0.75, 1.0],\n",
              "                         &#x27;eta&#x27;: [0.1, 0.2, 0.3], &#x27;gamma&#x27;: [0, 2],\n",
              "                         &#x27;scale_pos_weight&#x27;: [0.5, 1.0],\n",
              "                         &#x27;subsample&#x27;: [0.5, 0.75, 1.0]},\n",
              "             scoring=&#x27;f1&#x27;, verbose=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-36\" type=\"checkbox\" ><label for=\"sk-estimator-id-36\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GridSearchCV</label><div class=\"sk-toggleable__content\"><pre>GridSearchCV(cv=5, estimator=XGBClassifier(random_state=1),\n",
              "             param_grid={&#x27;colsample_by_tree&#x27;: [0.5, 0.75, 1.0],\n",
              "                         &#x27;colsample_bylevel&#x27;: [0.5, 0.75, 1.0],\n",
              "                         &#x27;eta&#x27;: [0.1, 0.2, 0.3], &#x27;gamma&#x27;: [0, 2],\n",
              "                         &#x27;scale_pos_weight&#x27;: [0.5, 1.0],\n",
              "                         &#x27;subsample&#x27;: [0.5, 0.75, 1.0]},\n",
              "             scoring=&#x27;f1&#x27;, verbose=1)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-37\" type=\"checkbox\" ><label for=\"sk-estimator-id-37\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(random_state=1)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-38\" type=\"checkbox\" ><label for=\"sk-estimator-id-38\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(random_state=1)</pre></div></div></div></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 121
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "go.best_params_"
      ],
      "metadata": {
        "id": "ALq-QBkB1Q-m",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "29dc0d09-0cd8-4a65-aed6-4e58d932a800"
      },
      "id": "ALq-QBkB1Q-m",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'colsample_by_tree': 0.5,\n",
              " 'colsample_bylevel': 1.0,\n",
              " 'eta': 0.1,\n",
              " 'gamma': 0,\n",
              " 'scale_pos_weight': 1.0,\n",
              " 'subsample': 1.0}"
            ]
          },
          "metadata": {},
          "execution_count": 122
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* A lower learning rate ended up being better for F1.\n",
        "* The classifier did not need to adjust class weights for better performance.\n",
        "* While ```colsample_by_tree``` is less than 100%, both ```colsample_by_level``` and ```subsample``` are maxed out at 1."
      ],
      "metadata": {
        "id": "mPeh54OAZmYX"
      },
      "id": "mPeh54OAZmYX"
    },
    {
      "cell_type": "code",
      "source": [
        "xgbc_tuned=go.best_estimator_\n",
        "\n",
        "xgbc_tuned.fit(X_train,y_train)"
      ],
      "metadata": {
        "id": "lDFYa2pA44n_",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 92
        },
        "outputId": "6256ea64-4be5-4e1e-dc55-145a6b181377"
      },
      "id": "lDFYa2pA44n_",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "XGBClassifier(colsample_by_tree=0.5, colsample_bylevel=1.0, eta=0.1,\n",
              "              random_state=1, scale_pos_weight=1.0, subsample=1.0)"
            ],
            "text/html": [
              "<style>#sk-container-id-19 {color: black;background-color: white;}#sk-container-id-19 pre{padding: 0;}#sk-container-id-19 div.sk-toggleable {background-color: white;}#sk-container-id-19 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-19 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-19 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-19 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-19 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-19 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-19 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-19 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-19 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-19 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-19 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-19 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-19 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-19 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-19 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-19 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-19 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-19 div.sk-item {position: relative;z-index: 1;}#sk-container-id-19 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-19 div.sk-item::before, #sk-container-id-19 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-19 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-19 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-19 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-19 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-19 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-19 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-19 div.sk-label-container {text-align: center;}#sk-container-id-19 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-19 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-19\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>XGBClassifier(colsample_by_tree=0.5, colsample_bylevel=1.0, eta=0.1,\n",
              "              random_state=1, scale_pos_weight=1.0, subsample=1.0)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-39\" type=\"checkbox\" checked><label for=\"sk-estimator-id-39\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(colsample_by_tree=0.5, colsample_bylevel=1.0, eta=0.1,\n",
              "              random_state=1, scale_pos_weight=1.0, subsample=1.0)</pre></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 123
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "scores(xgbc_tuned)"
      ],
      "metadata": {
        "id": "abyxG0VW1Y8e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "outputId": "dc6ea90f-ce4a-4982-b98d-e42f087147bb"
      },
      "id": "abyxG0VW1Y8e",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              Train      Test\n",
              "Accuracy   0.753762  0.750098\n",
              "Precision  0.780624  0.779646\n",
              "Recall     0.878235  0.872574\n",
              "F1         0.826558  0.823497"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-eb3646a1-771e-4750-b4b3-c6dc1ab59f42\">\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>Train</th>\n",
              "      <th>Test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Accuracy</th>\n",
              "      <td>0.753762</td>\n",
              "      <td>0.750098</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Precision</th>\n",
              "      <td>0.780624</td>\n",
              "      <td>0.779646</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Recall</th>\n",
              "      <td>0.878235</td>\n",
              "      <td>0.872574</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>F1</th>\n",
              "      <td>0.826558</td>\n",
              "      <td>0.823497</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-eb3646a1-771e-4750-b4b3-c6dc1ab59f42')\"\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-eb3646a1-771e-4750-b4b3-c6dc1ab59f42 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-eb3646a1-771e-4750-b4b3-c6dc1ab59f42');\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": 124
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "After training the best estimator on the full training set, we find that performance is comparable&mdash;in fact, nearly identical&mdash;to the other tuned boosting models."
      ],
      "metadata": {
        "id": "P-FQrBmCaIEP"
      },
      "id": "P-FQrBmCaIEP"
    },
    {
      "cell_type": "code",
      "source": [
        "tabulate(xgbc_tuned,'XGBoost (tuned)')"
      ],
      "metadata": {
        "id": "MusEIuCk5ArY",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 426
        },
        "outputId": "c46dd4dd-d6e8-4fd7-cba9-23ba0740a076"
      },
      "id": "MusEIuCk5ArY",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                       Train Acc  Test Acc  Train F1   Test F1   Status\n",
              "dTree (baseline)        1.000000  0.649902  1.000000  0.736208  Overfit\n",
              "Bagging Classifier      0.984168  0.698232  0.988121  0.775614  Overfit\n",
              "Random Forest           1.000000  0.725082  1.000000  0.802373  Overfit\n",
              "AdaBoost                0.734786  0.733857  0.817536  0.817036  General\n",
              "Gradient Boosting       0.754997  0.749312  0.827155  0.822581  General\n",
              "XGBoost                 0.753762  0.750098  0.826558  0.823497  General\n",
              "dTree (tuned)           0.749382  0.750098  0.823571  0.823497  General\n",
              "Bagging clfr (tuned)    0.738098  0.735298  0.824062  0.821797  General\n",
              "Random Forest (tuned)   0.725803  0.723772  0.820520  0.819605  General\n",
              "AdaBoost (tuned)        0.750561  0.749050  0.824158  0.823019  General\n",
              "Grad Boost (tuned)      0.754379  0.751801  0.826533  0.824293  General\n",
              "XGBoost (tuned)         0.753762  0.750098  0.826558  0.823497  General"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-ab8a28fb-08b4-4bde-9b67-4b4594d44e73\">\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>Train Acc</th>\n",
              "      <th>Test Acc</th>\n",
              "      <th>Train F1</th>\n",
              "      <th>Test F1</th>\n",
              "      <th>Status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>dTree (baseline)</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.649902</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.736208</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging Classifier</th>\n",
              "      <td>0.984168</td>\n",
              "      <td>0.698232</td>\n",
              "      <td>0.988121</td>\n",
              "      <td>0.775614</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.725082</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.802373</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost</th>\n",
              "      <td>0.734786</td>\n",
              "      <td>0.733857</td>\n",
              "      <td>0.817536</td>\n",
              "      <td>0.817036</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Gradient Boosting</th>\n",
              "      <td>0.754997</td>\n",
              "      <td>0.749312</td>\n",
              "      <td>0.827155</td>\n",
              "      <td>0.822581</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBoost</th>\n",
              "      <td>0.753762</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.826558</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>dTree (tuned)</th>\n",
              "      <td>0.749382</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.823571</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging clfr (tuned)</th>\n",
              "      <td>0.738098</td>\n",
              "      <td>0.735298</td>\n",
              "      <td>0.824062</td>\n",
              "      <td>0.821797</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest (tuned)</th>\n",
              "      <td>0.725803</td>\n",
              "      <td>0.723772</td>\n",
              "      <td>0.820520</td>\n",
              "      <td>0.819605</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost (tuned)</th>\n",
              "      <td>0.750561</td>\n",
              "      <td>0.749050</td>\n",
              "      <td>0.824158</td>\n",
              "      <td>0.823019</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Grad Boost (tuned)</th>\n",
              "      <td>0.754379</td>\n",
              "      <td>0.751801</td>\n",
              "      <td>0.826533</td>\n",
              "      <td>0.824293</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBoost (tuned)</th>\n",
              "      <td>0.753762</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.826558</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ab8a28fb-08b4-4bde-9b67-4b4594d44e73')\"\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-ab8a28fb-08b4-4bde-9b67-4b4594d44e73 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-ab8a28fb-08b4-4bde-9b67-4b4594d44e73');\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": 125
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Stacking Classifier"
      ],
      "metadata": {
        "id": "n2H7XaAw5LO7"
      },
      "id": "n2H7XaAw5LO7"
    },
    {
      "cell_type": "code",
      "source": [
        "ests=[('Decision Tree',dtree_tuned),\n",
        "      ('Bagging Classifier',bag_tuned),\n",
        "      ('Random Forest',rf_tuned),\n",
        "      ('AdaBoost',abc_tuned),\n",
        "      ('Gradient Boosting',gbc_tuned),\n",
        "      ('XGBoost',xgbc_tuned)]"
      ],
      "metadata": {
        "id": "I1I3qC6RcwOW"
      },
      "id": "I1I3qC6RcwOW",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "With all six ensemble models tuned, we will now use a stacking classifier to hopefully gain additional predictive power. "
      ],
      "metadata": {
        "id": "OzbwAZDMaY7W"
      },
      "id": "OzbwAZDMaY7W"
    },
    {
      "cell_type": "code",
      "source": [
        "stack=StackingClassifier(estimators=ests,\n",
        "                         final_estimator=XGBClassifier(random_state=1),\n",
        "                         cv=5)\n",
        "stack.fit(X_train,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 258
        },
        "id": "W-BXivhnchcf",
        "outputId": "7c4dd587-b183-441c-9528-111d5d7c4595"
      },
      "id": "W-BXivhnchcf",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "StackingClassifier(cv=5,\n",
              "                   estimators=[('Decision Tree',\n",
              "                                DecisionTreeClassifier(max_depth=7,\n",
              "                                                       max_leaf_nodes=20,\n",
              "                                                       min_samples_leaf=5,\n",
              "                                                       random_state=1)),\n",
              "                               ('Bagging Classifier',\n",
              "                                BaggingClassifier(random_state=1)),\n",
              "                               ('Random Forest',\n",
              "                                RandomForestClassifier(max_depth=5,\n",
              "                                                       max_samples=0.75,\n",
              "                                                       min_samples_split=8,\n",
              "                                                       n_estimators=250,\n",
              "                                                       random_state=1)),\n",
              "                               ('AdaBoost',\n",
              "                                AdaBoostClassifier(...3,\n",
              "                                                                                         random_state=2),\n",
              "                                                   learning_rate=0.1,\n",
              "                                                   random_state=1)),\n",
              "                               ('Gradient Boosting',\n",
              "                                GradientBoostingClassifier(init=AdaBoostClassifier(random_state=2),\n",
              "                                                           max_features=0.7,\n",
              "                                                           random_state=1)),\n",
              "                               ('XGBoost',\n",
              "                                XGBClassifier(colsample_by_tree=0.5,\n",
              "                                              colsample_bylevel=1.0, eta=0.1,\n",
              "                                              random_state=1,\n",
              "                                              scale_pos_weight=1.0,\n",
              "                                              subsample=1.0))],\n",
              "                   final_estimator=XGBClassifier(random_state=1))"
            ],
            "text/html": [
              "<style>#sk-container-id-20 {color: black;background-color: white;}#sk-container-id-20 pre{padding: 0;}#sk-container-id-20 div.sk-toggleable {background-color: white;}#sk-container-id-20 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-20 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-20 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-20 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-20 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-20 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-20 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-20 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-20 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-20 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-20 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-20 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-20 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-20 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-20 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-20 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-20 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-20 div.sk-item {position: relative;z-index: 1;}#sk-container-id-20 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-20 div.sk-item::before, #sk-container-id-20 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-20 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-20 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-20 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-20 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-20 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-20 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-20 div.sk-label-container {text-align: center;}#sk-container-id-20 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-20 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-20\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>StackingClassifier(cv=5,\n",
              "                   estimators=[(&#x27;Decision Tree&#x27;,\n",
              "                                DecisionTreeClassifier(max_depth=7,\n",
              "                                                       max_leaf_nodes=20,\n",
              "                                                       min_samples_leaf=5,\n",
              "                                                       random_state=1)),\n",
              "                               (&#x27;Bagging Classifier&#x27;,\n",
              "                                BaggingClassifier(random_state=1)),\n",
              "                               (&#x27;Random Forest&#x27;,\n",
              "                                RandomForestClassifier(max_depth=5,\n",
              "                                                       max_samples=0.75,\n",
              "                                                       min_samples_split=8,\n",
              "                                                       n_estimators=250,\n",
              "                                                       random_state=1)),\n",
              "                               (&#x27;AdaBoost&#x27;,\n",
              "                                AdaBoostClassifier(...3,\n",
              "                                                                                         random_state=2),\n",
              "                                                   learning_rate=0.1,\n",
              "                                                   random_state=1)),\n",
              "                               (&#x27;Gradient Boosting&#x27;,\n",
              "                                GradientBoostingClassifier(init=AdaBoostClassifier(random_state=2),\n",
              "                                                           max_features=0.7,\n",
              "                                                           random_state=1)),\n",
              "                               (&#x27;XGBoost&#x27;,\n",
              "                                XGBClassifier(colsample_by_tree=0.5,\n",
              "                                              colsample_bylevel=1.0, eta=0.1,\n",
              "                                              random_state=1,\n",
              "                                              scale_pos_weight=1.0,\n",
              "                                              subsample=1.0))],\n",
              "                   final_estimator=XGBClassifier(random_state=1))</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-40\" type=\"checkbox\" ><label for=\"sk-estimator-id-40\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StackingClassifier</label><div class=\"sk-toggleable__content\"><pre>StackingClassifier(cv=5,\n",
              "                   estimators=[(&#x27;Decision Tree&#x27;,\n",
              "                                DecisionTreeClassifier(max_depth=7,\n",
              "                                                       max_leaf_nodes=20,\n",
              "                                                       min_samples_leaf=5,\n",
              "                                                       random_state=1)),\n",
              "                               (&#x27;Bagging Classifier&#x27;,\n",
              "                                BaggingClassifier(random_state=1)),\n",
              "                               (&#x27;Random Forest&#x27;,\n",
              "                                RandomForestClassifier(max_depth=5,\n",
              "                                                       max_samples=0.75,\n",
              "                                                       min_samples_split=8,\n",
              "                                                       n_estimators=250,\n",
              "                                                       random_state=1)),\n",
              "                               (&#x27;AdaBoost&#x27;,\n",
              "                                AdaBoostClassifier(...3,\n",
              "                                                                                         random_state=2),\n",
              "                                                   learning_rate=0.1,\n",
              "                                                   random_state=1)),\n",
              "                               (&#x27;Gradient Boosting&#x27;,\n",
              "                                GradientBoostingClassifier(init=AdaBoostClassifier(random_state=2),\n",
              "                                                           max_features=0.7,\n",
              "                                                           random_state=1)),\n",
              "                               (&#x27;XGBoost&#x27;,\n",
              "                                XGBClassifier(colsample_by_tree=0.5,\n",
              "                                              colsample_bylevel=1.0, eta=0.1,\n",
              "                                              random_state=1,\n",
              "                                              scale_pos_weight=1.0,\n",
              "                                              subsample=1.0))],\n",
              "                   final_estimator=XGBClassifier(random_state=1))</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>Decision Tree</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-41\" type=\"checkbox\" ><label for=\"sk-estimator-id-41\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier(max_depth=7, max_leaf_nodes=20, min_samples_leaf=5,\n",
              "                       random_state=1)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>Bagging Classifier</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-42\" type=\"checkbox\" ><label for=\"sk-estimator-id-42\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">BaggingClassifier</label><div class=\"sk-toggleable__content\"><pre>BaggingClassifier(random_state=1)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>Random Forest</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-43\" type=\"checkbox\" ><label for=\"sk-estimator-id-43\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier(max_depth=5, max_samples=0.75, min_samples_split=8,\n",
              "                       n_estimators=250, random_state=1)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>AdaBoost</label></div></div><div class=\"sk-serial\"><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-44\" type=\"checkbox\" ><label for=\"sk-estimator-id-44\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">base_estimator: DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier(max_depth=3, random_state=2)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-45\" type=\"checkbox\" ><label for=\"sk-estimator-id-45\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier(max_depth=3, random_state=2)</pre></div></div></div></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>Gradient Boosting</label></div></div><div class=\"sk-serial\"><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-46\" type=\"checkbox\" ><label for=\"sk-estimator-id-46\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">init: AdaBoostClassifier</label><div class=\"sk-toggleable__content\"><pre>AdaBoostClassifier(random_state=2)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-47\" type=\"checkbox\" ><label for=\"sk-estimator-id-47\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">AdaBoostClassifier</label><div class=\"sk-toggleable__content\"><pre>AdaBoostClassifier(random_state=2)</pre></div></div></div></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>XGBoost</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-48\" type=\"checkbox\" ><label for=\"sk-estimator-id-48\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(colsample_by_tree=0.5, colsample_bylevel=1.0, eta=0.1,\n",
              "              random_state=1, scale_pos_weight=1.0, subsample=1.0)</pre></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>final_estimator</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-49\" type=\"checkbox\" ><label for=\"sk-estimator-id-49\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(random_state=1)</pre></div></div></div></div></div></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 128
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "scores(stack)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "OOTOzgUyddm_",
        "outputId": "270ffaf6-a89d-4601-b73c-7b2bac5ae91d"
      },
      "id": "OOTOzgUyddm_",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              Train      Test\n",
              "Accuracy   0.754772  0.749836\n",
              "Precision  0.783841  0.782640\n",
              "Recall     0.873950  0.866105\n",
              "F1         0.826446  0.822259"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-509b03a2-b5b6-40ae-b2f9-549a684e8561\">\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>Train</th>\n",
              "      <th>Test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Accuracy</th>\n",
              "      <td>0.754772</td>\n",
              "      <td>0.749836</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Precision</th>\n",
              "      <td>0.783841</td>\n",
              "      <td>0.782640</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Recall</th>\n",
              "      <td>0.873950</td>\n",
              "      <td>0.866105</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>F1</th>\n",
              "      <td>0.826446</td>\n",
              "      <td>0.822259</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-509b03a2-b5b6-40ae-b2f9-549a684e8561')\"\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-509b03a2-b5b6-40ae-b2f9-549a684e8561 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-509b03a2-b5b6-40ae-b2f9-549a684e8561');\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": 129
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The F1 score is at best several thousandths higher than other well-performing models. The improvement is not as drastic as could be hoped."
      ],
      "metadata": {
        "id": "w0RuVkiLar3h"
      },
      "id": "w0RuVkiLar3h"
    },
    {
      "cell_type": "code",
      "source": [
        "tabulate(stack,'Stacking')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 457
        },
        "id": "eQ2G9SUUddkS",
        "outputId": "ca813758-6727-4b95-c160-295b26e4542b"
      },
      "id": "eQ2G9SUUddkS",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                       Train Acc  Test Acc  Train F1   Test F1   Status\n",
              "dTree (baseline)        1.000000  0.649902  1.000000  0.736208  Overfit\n",
              "Bagging Classifier      0.984168  0.698232  0.988121  0.775614  Overfit\n",
              "Random Forest           1.000000  0.725082  1.000000  0.802373  Overfit\n",
              "AdaBoost                0.734786  0.733857  0.817536  0.817036  General\n",
              "Gradient Boosting       0.754997  0.749312  0.827155  0.822581  General\n",
              "XGBoost                 0.753762  0.750098  0.826558  0.823497  General\n",
              "dTree (tuned)           0.749382  0.750098  0.823571  0.823497  General\n",
              "Bagging clfr (tuned)    0.738098  0.735298  0.824062  0.821797  General\n",
              "Random Forest (tuned)   0.725803  0.723772  0.820520  0.819605  General\n",
              "AdaBoost (tuned)        0.750561  0.749050  0.824158  0.823019  General\n",
              "Grad Boost (tuned)      0.754379  0.751801  0.826533  0.824293  General\n",
              "XGBoost (tuned)         0.753762  0.750098  0.826558  0.823497  General\n",
              "Stacking                0.754772  0.749836  0.826446  0.822259  General"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a24400f5-e5a1-4a01-b988-af42deec1be7\">\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>Train Acc</th>\n",
              "      <th>Test Acc</th>\n",
              "      <th>Train F1</th>\n",
              "      <th>Test F1</th>\n",
              "      <th>Status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>dTree (baseline)</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.649902</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.736208</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging Classifier</th>\n",
              "      <td>0.984168</td>\n",
              "      <td>0.698232</td>\n",
              "      <td>0.988121</td>\n",
              "      <td>0.775614</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.725082</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.802373</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost</th>\n",
              "      <td>0.734786</td>\n",
              "      <td>0.733857</td>\n",
              "      <td>0.817536</td>\n",
              "      <td>0.817036</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Gradient Boosting</th>\n",
              "      <td>0.754997</td>\n",
              "      <td>0.749312</td>\n",
              "      <td>0.827155</td>\n",
              "      <td>0.822581</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBoost</th>\n",
              "      <td>0.753762</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.826558</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>dTree (tuned)</th>\n",
              "      <td>0.749382</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.823571</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging clfr (tuned)</th>\n",
              "      <td>0.738098</td>\n",
              "      <td>0.735298</td>\n",
              "      <td>0.824062</td>\n",
              "      <td>0.821797</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest (tuned)</th>\n",
              "      <td>0.725803</td>\n",
              "      <td>0.723772</td>\n",
              "      <td>0.820520</td>\n",
              "      <td>0.819605</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost (tuned)</th>\n",
              "      <td>0.750561</td>\n",
              "      <td>0.749050</td>\n",
              "      <td>0.824158</td>\n",
              "      <td>0.823019</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Grad Boost (tuned)</th>\n",
              "      <td>0.754379</td>\n",
              "      <td>0.751801</td>\n",
              "      <td>0.826533</td>\n",
              "      <td>0.824293</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBoost (tuned)</th>\n",
              "      <td>0.753762</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.826558</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Stacking</th>\n",
              "      <td>0.754772</td>\n",
              "      <td>0.749836</td>\n",
              "      <td>0.826446</td>\n",
              "      <td>0.822259</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a24400f5-e5a1-4a01-b988-af42deec1be7')\"\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-a24400f5-e5a1-4a01-b988-af42deec1be7 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-a24400f5-e5a1-4a01-b988-af42deec1be7');\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": 130
        }
      ]
    },
    {
      "cell_type": "markdown",
      "id": "obvious-maine",
      "metadata": {
        "id": "obvious-maine"
      },
      "source": [
        "## Model Performance Comparison and Conclusions"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Best Model"
      ],
      "metadata": {
        "id": "in-ISvTOv8BX"
      },
      "id": "in-ISvTOv8BX"
    },
    {
      "cell_type": "markdown",
      "source": [
        "We will examine the best model and compare it with some of the other high performing models."
      ],
      "metadata": {
        "id": "QeYSt8eSa5Fb"
      },
      "id": "QeYSt8eSa5Fb"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "everyday-kinase",
      "metadata": {
        "id": "everyday-kinase",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 457
        },
        "outputId": "823aef45-accc-4519-cdb2-796577ec3cdd"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                       Train Acc  Test Acc  Train F1   Test F1   Status\n",
              "dTree (baseline)        1.000000  0.649902  1.000000  0.736208  Overfit\n",
              "Bagging Classifier      0.984168  0.698232  0.988121  0.775614  Overfit\n",
              "Random Forest           1.000000  0.725082  1.000000  0.802373  Overfit\n",
              "AdaBoost                0.734786  0.733857  0.817536  0.817036  General\n",
              "Gradient Boosting       0.754997  0.749312  0.827155  0.822581  General\n",
              "XGBoost                 0.753762  0.750098  0.826558  0.823497  General\n",
              "dTree (tuned)           0.749382  0.750098  0.823571  0.823497  General\n",
              "Bagging clfr (tuned)    0.738098  0.735298  0.824062  0.821797  General\n",
              "Random Forest (tuned)   0.725803  0.723772  0.820520  0.819605  General\n",
              "AdaBoost (tuned)        0.750561  0.749050  0.824158  0.823019  General\n",
              "Grad Boost (tuned)      0.754379  0.751801  0.826533  0.824293  General\n",
              "XGBoost (tuned)         0.753762  0.750098  0.826558  0.823497  General\n",
              "Stacking                0.754772  0.749836  0.826446  0.822259  General"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-8cfe6ade-53c9-46f0-b993-608ccad78ba2\">\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>Train Acc</th>\n",
              "      <th>Test Acc</th>\n",
              "      <th>Train F1</th>\n",
              "      <th>Test F1</th>\n",
              "      <th>Status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>dTree (baseline)</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.649902</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.736208</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging Classifier</th>\n",
              "      <td>0.984168</td>\n",
              "      <td>0.698232</td>\n",
              "      <td>0.988121</td>\n",
              "      <td>0.775614</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.725082</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.802373</td>\n",
              "      <td>Overfit</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost</th>\n",
              "      <td>0.734786</td>\n",
              "      <td>0.733857</td>\n",
              "      <td>0.817536</td>\n",
              "      <td>0.817036</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Gradient Boosting</th>\n",
              "      <td>0.754997</td>\n",
              "      <td>0.749312</td>\n",
              "      <td>0.827155</td>\n",
              "      <td>0.822581</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBoost</th>\n",
              "      <td>0.753762</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.826558</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>dTree (tuned)</th>\n",
              "      <td>0.749382</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.823571</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Bagging clfr (tuned)</th>\n",
              "      <td>0.738098</td>\n",
              "      <td>0.735298</td>\n",
              "      <td>0.824062</td>\n",
              "      <td>0.821797</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Random Forest (tuned)</th>\n",
              "      <td>0.725803</td>\n",
              "      <td>0.723772</td>\n",
              "      <td>0.820520</td>\n",
              "      <td>0.819605</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoost (tuned)</th>\n",
              "      <td>0.750561</td>\n",
              "      <td>0.749050</td>\n",
              "      <td>0.824158</td>\n",
              "      <td>0.823019</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Grad Boost (tuned)</th>\n",
              "      <td>0.754379</td>\n",
              "      <td>0.751801</td>\n",
              "      <td>0.826533</td>\n",
              "      <td>0.824293</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBoost (tuned)</th>\n",
              "      <td>0.753762</td>\n",
              "      <td>0.750098</td>\n",
              "      <td>0.826558</td>\n",
              "      <td>0.823497</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Stacking</th>\n",
              "      <td>0.754772</td>\n",
              "      <td>0.749836</td>\n",
              "      <td>0.826446</td>\n",
              "      <td>0.822259</td>\n",
              "      <td>General</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8cfe6ade-53c9-46f0-b993-608ccad78ba2')\"\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-8cfe6ade-53c9-46f0-b993-608ccad78ba2 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-8cfe6ade-53c9-46f0-b993-608ccad78ba2');\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": 154
        }
      ],
      "source": [
        "mct=model_comp_table\n",
        "mct"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We will define the 'best model' as that which has the highest F1 score on testing data."
      ],
      "metadata": {
        "id": "B7rf3R3vrXLz"
      },
      "id": "B7rf3R3vrXLz"
    },
    {
      "cell_type": "code",
      "source": [
        "mct.loc[mct['Status']=='General']['Test F1'].idxmax()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "GEb9M8JcrBmi",
        "outputId": "d83c22c4-8c3b-4bf7-85ee-3c309abb70bb"
      },
      "id": "GEb9M8JcrBmi",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'Grad Boost (tuned)'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 155
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The best generalized model is the tuned gradient boosting classifier."
      ],
      "metadata": {
        "id": "SNVNHrqCrRqC"
      },
      "id": "SNVNHrqCrRqC"
    },
    {
      "cell_type": "code",
      "source": [
        "mct.loc[mct['Status']=='General']['Test Acc'].idxmax()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "hhuDYfT7r9DH",
        "outputId": "0a08879b-3c71-4395-d32c-6c1af7433917"
      },
      "id": "hhuDYfT7r9DH",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'Grad Boost (tuned)'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 156
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "mct.loc['Grad Boost (tuned)']"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BB1Swd69rzd8",
        "outputId": "18d2114e-5a17-4ce3-86fe-654b258a511d"
      },
      "id": "BB1Swd69rzd8",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Train Acc    0.754379\n",
              "Test Acc     0.751801\n",
              "Train F1     0.826533\n",
              "Test F1      0.824293\n",
              "Status        General\n",
              "Name: Grad Boost (tuned), dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 157
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Incidentally, the tuned gradient boosting classifier is also the most accurate on test data, with over 75% accuracy. The F1 score of this model is a bit over 82%."
      ],
      "metadata": {
        "id": "HXull7Zmr2YW"
      },
      "id": "HXull7Zmr2YW"
    },
    {
      "cell_type": "code",
      "source": [
        "confusion_heatmap(gbc_tuned)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 365
        },
        "id": "kiFYuB5Arhkk",
        "outputId": "b162dc8e-38a5-4a00-b9ae-4563e8edfd41"
      },
      "id": "kiFYuB5Arhkk",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The tuned gradient boosting classifier allows few false negatives, but comparatively more false positives. This explains the higher recall score, as compared to precision.\n",
        "\n",
        "The percentage of true negatives is nearly identical to the percentage of false positives, around 16%. This means our model does a bad job at distinguishing Denied cases, as it's prediction is no better than random guessing. Put another way, there is only about a 50% chance that our model will correctly identify when a case is Denied."
      ],
      "metadata": {
        "id": "BLkr2LdqrsUe"
      },
      "id": "BLkr2LdqrsUe"
    },
    {
      "cell_type": "code",
      "source": [
        "# z-test\n",
        "num_obs=1239+1295\n",
        "t,p_val=proportions_ztest([1239,1295],[num_obs,num_obs])\n",
        "print('The p-value is',p_val)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eN7AScs70aA_",
        "outputId": "4d0be5f6-c4df-4100-a07f-c65ca259e575"
      },
      "id": "eN7AScs70aA_",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The p-value is 0.11565928243244895\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We can run a two proportions z-test to check our model's ability to predict Denied cases. With a p-value greater than 0.05, our model is only as good as random guessing."
      ],
      "metadata": {
        "id": "Xc1wMwrg0g-D"
      },
      "id": "Xc1wMwrg0g-D"
    },
    {
      "cell_type": "code",
      "source": [
        "# feature importance\n",
        "ser=pd.Series(gbc_tuned.feature_importances_,index=X_train.columns)\n",
        "ser=ser.sort_values(ascending=False)\n",
        "print('Feature importance')\n",
        "print('-'*10)\n",
        "ser"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CijVFR1EtzgG",
        "outputId": "0cabc501-fbb4-4c06-f250-d16f4fa65a77"
      },
      "id": "CijVFR1EtzgG",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Feature importance\n",
            "----------\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "education_of_employee             0.443459\n",
              "has_job_experience                0.150752\n",
              "unit_of_wage_Hour                 0.100343\n",
              "continent_Europe                  0.061272\n",
              "prevailing_wage                   0.054921\n",
              "unit_of_wage_Year                 0.030446\n",
              "region_of_employment_Midwest      0.029068\n",
              "continent_North America           0.022887\n",
              "region_of_employment_South        0.021740\n",
              "no_of_employees                   0.018451\n",
              "region_of_employment_West         0.017778\n",
              "years_in_business                 0.012811\n",
              "full_time_position                0.008663\n",
              "requires_job_training             0.008135\n",
              "continent_South America           0.006894\n",
              "region_of_employment_Northeast    0.005634\n",
              "continent_Asia                    0.003787\n",
              "region_of_employment_Island       0.000889\n",
              "unit_of_wage_Week                 0.000589\n",
              "continent_Oceania                 0.000537\n",
              "unit_of_wage_Month                0.000521\n",
              "continent_Africa                  0.000423\n",
              "dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 171
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Education of the applicant is most important, i.e., has the greatest predictive impact on ```case_status```.\n",
        "* Job experience is the next most influential feature.\n",
        "* Continent of origin has less importance, as do the years in business and full time status of the role."
      ],
      "metadata": {
        "id": "6aBP-XfjuNO_"
      },
      "id": "6aBP-XfjuNO_"
    },
    {
      "cell_type": "code",
      "source": [
        "plott()\n",
        "plt.title('Feature Importances in Tuned Gradient Boosting Classifier',fontsize=14)\n",
        "sns.barplot(x=ser,y=ser.index)\n",
        "plt.xlabel('Importance');"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "DJk2tTTVu02Z",
        "outputId": "12e0f203-da03-4d9a-ff26-34bd3aa3ad3a"
      },
      "id": "DJk2tTTVu02Z",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Relative importance can be gleaned with the plot above. Education is massively more important than anything else."
      ],
      "metadata": {
        "id": "9_Mmh35qvGhk"
      },
      "id": "9_Mmh35qvGhk"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Other Good Models"
      ],
      "metadata": {
        "id": "UZkFTwn7wC1w"
      },
      "id": "UZkFTwn7wC1w"
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Tuned Decision Tree"
      ],
      "metadata": {
        "id": "KvhRCNQAx8wM"
      },
      "id": "KvhRCNQAx8wM"
    },
    {
      "cell_type": "code",
      "source": [
        "mct.loc['dTree (tuned)']"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2oRsYgmIwFon",
        "outputId": "9adf54dd-b6d6-4bd9-9fc2-1c9d49ab0106"
      },
      "id": "2oRsYgmIwFon",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Train Acc    0.749382\n",
              "Test Acc     0.750098\n",
              "Train F1     0.823571\n",
              "Test F1      0.823497\n",
              "Status        General\n",
              "Name: dTree (tuned), dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 173
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The tuned decision tree also performed well on test data, with around 75% accuracy and 82% F1 score. This seems like the upper limit for performance."
      ],
      "metadata": {
        "id": "JaT9XRmgwdPN"
      },
      "id": "JaT9XRmgwdPN"
    },
    {
      "cell_type": "code",
      "source": [
        "confusion_heatmap(dtree_tuned)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 365
        },
        "id": "eIusXiLjw-bN",
        "outputId": "cb52664d-dd7d-4b0c-aaff-36f980641db2"
      },
      "id": "eIusXiLjw-bN",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Our confusion matrix for the tuned decision tree looks quite similar to that for the tuned gradient boosting classifier. It seems there is a trend in our models being unable to predict Denied applications."
      ],
      "metadata": {
        "id": "NghPUQCGxDGE"
      },
      "id": "NghPUQCGxDGE"
    },
    {
      "cell_type": "code",
      "source": [
        "# feature importance\n",
        "ser=pd.Series(gbc_tuned.feature_importances_,index=X_train.columns)\n",
        "ser=ser.sort_values(ascending=False)\n",
        "\n",
        "plott()\n",
        "plt.title('Feature Importances in Tuned Decision Tree',fontsize=14)\n",
        "sns.barplot(x=ser,y=ser.index)\n",
        "plt.xlabel('Importance');"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "id": "09icVx6hxVBA",
        "outputId": "ae1f5640-71ad-4650-dbba-047cb690dff1"
      },
      "id": "09icVx6hxVBA",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAApEAAAFTCAYAAABh41fpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeVhV1frA8e9hcgAynCi7WGYCDimaxpCGkl3NIUYVDdScyjRuEjdATRxBBDHFSFHTMg0HEERzqFS0QMxUMGdBBbs54QgIHOD8/uDHTmRWFI338zw9cfbZe621Fxx5WWuv9ao0Go0GIYQQQgghqkGrthsghBBCCCGePhJECiGEEEKIapMgUgghhBBCVJsEkUIIIYQQotokiBRCCCGEENUmQaQQQgghhKg2CSKFEEI8kczMzNi+fXttN+OR+OCDD/Dx8antZgBw8eJFzMzMOHr0aJXOd3d3Z+bMmY+4VeJpoFPbDRBCPP18fHzYtGlTqePR0dG0bdv2octPTExk+PDhJCQk0Lhx44cu70GZmZmxcOFC+vbtW2ttqMiT0k815ZdffqFRo0YPfL2ZmVmF7zs6OjJ37twHLv9RKv5eAqhUKho2bMgLL7yApaUlI0aMwMTEpMbqev755/nll18wMjKq0vmhoaHo6Dy68OHeey9PQEAATk5Oj6wNomokiBRC1AgbGxvmzZtX4lhVfyk9Tnl5eejp6dV2M2pcXl5ebTehxjVr1uyhrv/ll1+Ur/fs2cPUqVNLHKtfv/5Dlf84bN26lUaNGpGdnc3Jkyf55ptvePfdd1m6dCmvv/56jdShra1drb5+9tlna6Te8nTu3LnE9ykkJIRz584RGhqqHDM0NFS+LiwsRKPRoK2t/UjbJUqT6WwhRI3Q09OjWbNmJf4rHq3YtWsXTk5OvPrqq9jZ2bFgwYISQU9MTAzOzs507twZa2trPDw8uHz5MlA01VY8KmFtbY2ZmZkyDVjWtJqPjw8ffPCB8trd3R0/Pz8CAwOxsrJi6NChAJw9e5Zx48YpdXp6enL16tVq3bOZmRlr165l/PjxdOrUiT59+rB//34uXbrE6NGjsbCwwN7enmPHjinXREVF0blzZ3bt2kWfPn149dVXcXd3Jz09vUTZERERvP3223To0IG3336b9evXl6p7zZo1TJw4EQsLC7y8vMrtp7179zJs2DC6devG66+/zujRo0lJSVHKKp7O3LFjB++//z6dOnWiX79+/PrrryXqTElJ4cMPP+S1116jc+fODBkyhFOnTinvR0ZG0q9fP1599VX69OnDqlWrKCwsLHFPxfdsaWnJ6NGjyc/Pr7B/i6ezq9rGe937s1gcdBS/Tk1NpWvXrly/fr1UPxRP6yYmJmJmZkZCQgKDBg2iU6dOODk5lfh+Ahw6dAg3Nzc6depEjx498PPzIzMzU3n/7t27+Pj40LlzZ2xsbFiyZEm5bb5f48aNadasGS+++CJ9+vRh9erVtG3blsmTJ1NQUKCcV9lnLC8vj5CQEHr16kWHDh146623+Pbbb8u8b7VazezZs+nevTsdOnTA1taW4OBgpaz7P3e3bt3C29ubbt260bFjR0aOHMmZM2eU94t/5hMSEhgwYAAWFhZl/swXu//fkgYNGqCrq6u83rdvH9bW1sTFxTFgwAA6dOhASkoKeXl5BAUF8eabb9KpUyecnZ3Zt29fibJr4nMv/iZBpBDikdq3bx9eXl689957bN26FX9/f7Zv386CBQuUc9RqNR4eHmzevJmlS5dy48YNPD09gaKptuIRiK1bt/LLL78wZcqUarVh8+bNaDQa1qxZw7x587hy5Qrvvfcebdq0YePGjaxcuZLs7Gw++uijEkFPVXz11Vf079+fmJgYOnTogKenJ1OmTGHo0KFs2rSJ5s2bl3r2LS8vj8WLF+Pv78+6desoLCxk4sSJFGeh/fHHH5k1axYjRowgNjaW4cOHM2PGDHbt2lWinMWLF2Nra0tsbCxeXl7l9tPdu3cZMWIEGzZs4Ntvv8XAwIAPP/yw1OjlggULcHd3JyYmhldffRVPT0+ysrIAuHz5MsOGDUOlUrFy5UqioqIYNmyY0l/r169nwYIFeHh48MMPP+Dt7c2yZctYu3YtAEePHmXmzJlMmDCB7du3880339CjR49q9XVlbXxU5s+fz6effkpUVBRGRkZ4eXkp36tTp04xevRo7OzsiImJYfHixZw8eZLJkycr1wcGBvLrr7+yaNEiVq1axfHjx/ntt98eqC3a2tqMHDmS9PR0jh8/DlTtM+bj40N0dDQ+Pj5s27aNOXPm8Mwzz5RZx+rVq/nxxx9ZsGABO3fuZMGCBbRq1arcNvn4+JCUlERYWBgbNmygfv36jBkzhpycHOWcvLw8li5dir+/PxEREdy5c4fp06c/UB8A5ObmEhYWxowZM9i6dSstWrTA19eX3377jfnz57NlyxYcHR0ZP348J0+eBKjRz734fxohhHhI3t7emrZt22osLCyU/0aPHq3RaDSaYcOGaRYvXlzi/B9//FFjYWGhKSwsLLO8s2fPakxNTTV//fWXRqPRaPbv368xNTXVZGRklDjPzc1NM2PGjFJtGTduXIlzBgwYUOKcL774QjN8+PASx27evKkxNTXVJCUllXufpqammm3btpV4HRwcrLw+deqUxtTUVPP1118rx+5ve2RkpMbU1FRz8OBB5ZyLFy9qzM3NNb/++qtGo9FohgwZovHx8Sl1X66uriXqnjlzZolzyuun+2VlZWnMzc01v/32m0aj0WjS09M1pqammu+//14559KlSxpTU1PlnJCQEE3Pnj01ubm5ZZZpa2ur2bRpU4ljK1eu1LzzzjsajUaj2bFjh6ZLly6aO3fuVNi2e93b31VpY0W2bdumMTU1VV6X1VfFdSQnJ5c4Z+/evco5Bw8eLPGz+d///lfj6+tboq7jx49rTE1NNdeuXdNkZmZq2rdvr4mJiVHez8zM1Lz22msab2/vcttb0fey+POxdetWjUZT+Wfs3LlzGlNTU01cXFyZdd1/37NmzdIMHz683M/nvZ+74rIPHDigvH/79m1Nly5dNOvXr9doNH//zKekpCjnxMTEaNq3b19uHfeaMWOGxs3NTXldXN7Ro0eVYxcuXNCYmZlp/vzzzxLXjh8/XuPn56fRaB78cy/KJ89ECiFqRNeuXZk1a5byuvh5s2PHjpGcnMzy5cuV9woLC8nJyeHq1as0b96cY8eOKSM4N2/eVM773//+x3PPPffQbevQoUOJ18eOHePgwYN07ty51LlpaWl07NixymXfu3ijadOmAJiampY6dv36dWWxi5aWVok6XnjhBZo3b87Zs2exsbEhNTUVZ2fnEvW89tprpUYi77+v8qSlpbFw4UKSkpK4fv06Go2GwsJC/vrrr3LvpXnz5kq7AY4fP06XLl3KfJ70+vXr/PXXX/j5+TFjxgzleH5+vjJiZ2NjQ4sWLXjrrbfo3r073bt35+2338bAwKBK91CVNj4qZdWZkZHBc889x7Fjx7hw4QLbtm1Tzim+57S0NBo0aIBarS7xs6avr1/iZ+RBqVQqoPLP2PHjx9HS0sLKyqpK5To6OjJq1Cj69OnDG2+8ga2tLW+++SZaWqUnL1NSUtDS0sLCwkI5ZmhoiKmpKWfPnlWO6enp8fLLLyuvmzdvjlqt5tatWw/0jKWOjk6JRXvHjh1Do9HQv3//Eufl5eUp912Tn3tRRIJIIUSNaNCgAS+++GKp48VTtWWtaG7cuDHZ2dmMHj1aWZjTuHFjbty4wXvvvYdara6wTpVKpfzCLlbWNQ0aNCjVJltbW7y9vUud26RJkwrrvF9Zq1TLOnb/dFlxAFAd919z/32V54MPPuC5555j5syZGBsbo62tTf/+/Uv11b3tLq6rKtN8xefMmDGjzF/QAAYGBmzatInffvuN+Ph4li5dSkhICBs3bsTY2LhK9/EwbbxfWQFRec9nVlRnYWEhgwYNYuTIkaWuMzY25vz589VuW2WKg7PiFdqVfcaqq3379vz888/88ssvJCQk4O3tjbm5OStXriyz38pz78/r/Z+Jh/neQVFQeu9CGo1Gg0qlYuPGjaXqKv6DtiY/96KIBJFCiEeqXbt2pKamlhlgApw8eZIbN24wadIk5Zfizp07S5yjq6sLlP6F07hx41IPxZ86dYoXXnihwja1b9+ebdu20aJFC6Xsx6mwsJDk5GS6dOkCFI24XrlyhdatWwPw8ssvc+jQIQYNGqRc8/vvvyvvl6esfrpx4wapqan4+fmVGJGpaEFLWdq1a8fmzZvLXN3etGlTmjdvTlpaGg4ODuWWoaOjg7W1NdbW1nz88cfY2NiwZ88ehgwZUq221ITi4OrKlSvK1ydOnKh2Oe3atePs2bPl/nybmJigq6vLkSNHlJ/v7Oxszpw5Q8uWLatdX0FBAd988w0tW7ZURuIq+4y1bduWwsJC9u/fz5tvvlmlegwMDOjbty99+/bFycmJwYMHc+HChVLPRrZu3ZrCwkKOHDlCt27dAMjMzOT06dOPdQuetm3botFouHr1arkjrrX9uf8nkoU1QohHasKECWzZsoWFCxdy+vRpUlJS2L59u7IdUIsWLdDT02PNmjWkp6ezZ88eFi5cWKKMF154AZVKxZ49e7h+/bqykMLKyoq9e/fy888/k5qaSkBAQKkp2rIMGzaMO3fuMGnSJJKSkkhPTyc+Pp7PP/+8xKraR0VHRwd/f38OHz7MiRMn8Pb25pVXXsHGxgaAMWPGsHnzZtasWcP58+dZvXo1sbGxjBkzpsJyy+qnRo0aYWRkxIYNG7hw4QIHDhzAz8+v2vv8DRs2jOzsbD755BOSk5O5cOECW7ZsUQIvDw8Pli9fzqpVq0hNTeX06dNER0ezdOlSAHbv3s0333zD8ePH+fPPP9myZQtZWVmVBsaPSsuWLXn++edZvHgx586d45dffuGrr76qdjljx44lOTmZadOmcfz4cS5cuMDu3buZNm0aUDR17ezsTHBwML/++itnzpwptbK6ItevX+fq1atcuHCBnTt34u7uzokTJ/D391dG4ir7jLVq1Yp33nmHqVOnsmPHDtLT0zl48CDR0dFl1rly5Uq2bNlCSkoKFy5cIDY2FgMDgzIfLXnppZd46623mDZtGgcPHuTUqVN4eXlhYGDAwIEDq92fD6pVq1YMHDgQX19ftm/fTnp6OkePHmXFihXKH6W1/bn/J5KRSCHEI9WjRw+WLl1KWFgYX3/9Ndra2rz00kvKKEXjxo0JDAwkJCSENWvWKFvT3BswGRsb8/HHH/PFF18wdepUHBwcmDt3Ls7Ozpw6dUpZCfvee+/x9ttvc+PGjQrbZGxszPfff09ISAhjxowhNzeX559/nu7duz+WPST19PT48MMP8fb25n//+x8WFhYsXrxYmeLr3bs3U6dO5euvv8bf358WLVrg5+eHnZ1dpfdVVj8tWLCAOXPmMGDAAF588UW8vb3x8PCoVpuNjY357rvvmDdvHiNGjACKnhUs3upl0KBBNGjQgBUrVjB//nzq16/PK6+8gpubG1D0nNxPP/1EWFgYd+/epWXLlsyePZuuXbtWt/tqhK6uLiEhIcyYMQN7e3vatm2Lp6dnie2hqsLc3JzvvvuOL774Ajc3NwoLCzExMaF3797KOd7e3ty9e5eJEydSv3593NzcuHv3bpXKL37Gr3izcSsrKwIDA0tsNl7ZZwxg3rx5fPHFF8yePZsbN27w3HPPlTkFD0WB74oVKzh//jwqlYp27dqxbNmych+fCAgIwN/fn/Hjx5Obm0uXLl1Yvnz5Y9+HMyAggCVLlhAUFMTly5dp1KiRsp0U1P7n/p9Ipbn/gSIhhBCPTFRUFLNmzeLw4cO13RQhhHgoMp0thBBCCCGqTYJIIYQQQghRbTKdLYQQQgghqk1GIoUQQgghRLVJECmEEEIIIapNgkghhBBCCFFtsk+kEEIIIcQ/QEGempt3cmusPC0tFUZG+uW+L0GkENWU8d0mCu9k1XYzhBBCiBKajXejsDDnsdUn09lVdPHiRWXX+5p24sQJfvjhhxLH7O3tycl5fD8IAIcOHWLAgAE4ODiwf//+x1r3vdzd3dm9e3et1S+EEEKIykkQ+QQ4ceIE27dvL3EsJibmsaeMiomJwcHBgejo6HIT2AshhBBCgExnA5CUlERwcDBZWUVTlB4eHvTs2ZM1a9awatUqDAwMsLW1Vc5PTEwkMDCQqKioMl9v3LiRb7/9FijKz7p06VKeffZZPvjgA27cuEFubi4dO3ZkxowZZGVlsWjRIjIzM7G3t6dbt25MnToVMzMzDh06hL6+PsnJycyZM4fs7GwaNmzIlClT6NixIxcvXsTZ2RlXV1fi4uK4e/cuc+bMqTAX7bVr1/Dz8yMtLQ2A0aNH4+DgwPLly9m2bRv169cnNjaWdevWlRvEhoeHs3PnTgoKCjA2NmbWrFk0a9aM0NBQUlNTyczM5Pz587Rv355x48Yxd+5c/ve///H222/j7e0NFI02mpubc/jwYW7dusU777yDp6dnldu7bds2Nm3aRHh4OAB5eXnY2dmxfv16WrRoUW4b8/LyWLBgAb/99ht5eXmYmZkxffp09PXLf+ZDCCGEEKXV+SDy9u3b+Pn5ER4eTvPmzbly5QouLi7Mnz+fr776iujoaJo2bcr06dOrVF5iYiJLly5l7dq1NGvWjKysLHR0dNDW1iY4OBgjIyM0Gg3e3t5ERkYydOhQPDw82LNnD4sWLSpVXl5eHh4eHgQEBGBtbU18fDweHh7s3LkTgJs3b2JhYcGkSZPYvHkzwcHBRERElNu+2bNn06ZNG7788kuuXLmCk5MT7dq1Y8yYMZw9e5YOHTrg5uZW7vUxMTGkp6ezfv16tLS0WLt2LXPnzmX+/PkAHDt2jMjISBo2bIijoyPz589n+fLl5Ofn89ZbbzFkyBBeeuklAFJSUoiIiCA3NxdXV1c6d+5Mr169qtTet99+m3nz5pGeno6JiQk//PADnTp1okWLFhW2cfny5RgaGrJx40YAgoKCCA8PZ9KkSVX6/gohhBCiSJ0PIg8fPszFixcZO3asckylUpGYmEjPnj1p2rQpAEOGDGHbtm2Vlrdnzx7s7e1p1qwZgDLCVVBQwNdff83evXspLCzk1q1bVZquPnfuHLq6ulhbWwNgY2ODrq4u586dQ19fn4YNGyqBl4WFBYGBgRWWl5CQgI+PDwDNmzfH1taWxMRETE1NK20LwK5du/jjjz9wdHRU7svAwEB5v3v37hgaGgJgZmaGubk5enp66Onp0apVK9LS0pQg0sHBAR0dHXR0dOjXrx/79+8vFURW1N4hQ4YQERHBf//7X9auXcsnn3xSaRt37dpFZmYmO3bsAIqCdHNz8yrduxBCCCH+VueDSI1Gg5mZGWvWrClx/Ntvv+XSpUtlXqOtrc292SJzcytfTh8bG8vvv//OmjVrMDAwYMmSJZw/f/6h2g6gp6enfK2lpUV+fv5Dl1kRjUbD+PHjcXFxKfP9evXqKV9ra2uXel1QUFBjbRk8eDCOjo7Y2dlx+/ZtJdCuqI0ajQY/Pz/lXCGEEEI8mDq/sKZz585cuHChxGrk5ORkXn/9deLi4sjIyABQpj8BTExMSE9P59atW2g0GrZu3aq817NnT2JiYrh27RoAWVlZ5ObmcufOHYyMjDAwMODOnTts2bJFuab4WFlatWqFWq1W2peQkEB+fj6tWrV6oPu1trZm/fr1AFy9epW4uLhqLaKxs7Nj7dq13Lp1CygayTt58uQDtWXz5s3k5+eTnZ3Ntm3bymxHRe1t3LgxNjY2eHp6MmzYMFQqVaVttLOzY9WqVcrK98zMTFJSUh6o/UIIIURdVudHIhs1akRYWBhBQUH4+/ujVqsxMTFhyZIlfPjhhwwdOhQDAwPefPNN5RpjY2Pef/99nJycaNq0Kd26dePMmTMAWFpaMm7cON5//31UKhV6enosWbIEBwcHfv75Z/r27UuTJk147bXXlBFMa2trvv76a959911ef/11pk6dqtSlp6fHokWLSiysWbhwYYkRyOqYOnUq06ZNY+DAgQB4eXnRpk2bKl/v4ODAzZs3lecmNRoNQ4cOfaAp4ZdffhlXV1dlYc39U9lVaa+Liwvbt29Xpq4ra+O4ceNYvHgxLi4uqFQqVCoVEydOpHXr1tVuvxBCCFGXqTT3zssK8Zi4u7szatSoMgPH6ggLC+Pq1av4+fnVUMuEEEKIp1NBnprrt2puj2ktLRVNmhiU+36dH4kUT6/+/fujra3NihUrHmu9GRmZFBbK3161oVkzQ65eLfvRD/HoSf/XHun72iX9XzYZifwHOnHihLKi+V5ubm4MGjSoSmU4OTmVWgTTqVMnZs6cWSNtFKIuqulRgsdNfpHWHun72lVX+7+ykUgJIoWopqvffUXhnVu13QzxFDIe7/NU/yKqq79InwTS97WrrvZ/ZUFknV+dLYQQQgghqk+CyEfEzMxMSaP4qNjb2ytb1ZTn4sWLWFpaPtJ21LSq3JcQQgghapcsrHmKxcTE1HYTalR+fj46Ojr/uPsSQggh/okkiHyEVq9ezY8//sjNmzf57LPP6NOnDwCffvop586dQ61W07JlS/z9/WnUqBGpqan4+vpy9+5dCgsLcXR0ZPTo0eWWb2ZmxqFDh9DX1yc5ObnEXpJTpkyhY8eOyrlz587l119/BcDPz4+uXbuWW25mZiYBAQGcOnWK3NxcLC0t8fX15fz584waNYq1a9fywgsvsHjxYlJSUliwYAGhoaGcPXuWGzducOXKFdq0aYO/vz+Ghobk5eWxYMECfvvtN/Ly8jAzM2P69Ono6+vj4+ODtrY2586dIysri5iYmBL3lZqair+/Pzdu3ECtVjNixAicnZ2V+580aVKZfXz48GHmzZunjAZ/9tlndO/evcLyhBBCCFF1EkQ+QgYGBkRGRvL777/zySefKAHOlClTaNy4MQALFixg2bJleHl5sXbtWuzs7Pjggw8AlIwrlcnLy8PDw4OAgACsra2Jj4/Hw8ODnTt3AnDz5k3Mzc3x8fEhMTERT09Pfvrpp3I3LA8ICKBbt27MmTOHwsJCvLy8iIyMZPDgwUyaNIlJkybh4eFBbGwskZGRynW///470dHRNG3aFF9fX8LCwvD29mb58uUYGhoqWX+CgoIIDw9n0qRJQNFq8u+++46GDRuWaEd+fj5eXl4EBQXRunVrMjMzcXZ2xsLCQtkcvKw+vnnzJhMnTiQ0NJQuXbpQUFBAZmZmlcoTQgghRNVIEPkI9evXDwALCwuuXLlCbm4u9erVIyYmhtjYWNRqNdnZ2bz00ksAdOvWjaCgIO7evYulpWWV0xGeO3cOXV1dJR+0jY0Nurq6nDt3Dn19fXR1dXn33XeBoow69evXJzU1tdwsM7t27SI5OZmVK1cCkJOTg7GxMVCUDWb//v1MmDBByQNerGfPnjRt2hQoyiQze/ZspbzMzEx27NgBFAW999bdt2/fUgEkwPnz50lJScHT01M5plarSU1NVYK+svr4yJEjtG7dmi5dugBFObsbNWrE2bNnKy1PCCGEEFUjQeQjVK9ePaAoiIGikbWjR4/y/fffExERQePGjYmNjVVyQ/fp0wcLCwt+/fVXli1bRmRkJMHBwY+93RqNhrCwMExMTEq9l5eXx5kzZzA0NFTyilelPD8/PyXIvV9ZAWTxdUZGRhU+I1lWH1fUjsrKE0IIIUTVyOrsx+z27dsYGBjw7LPPkpeXV2I6+MKFCzRr1gwnJycmTJjA0aNHq1Rmq1atUKvV7N+/H4CEhATy8/Np1aoVUDTaFhsbC8DBgwfJycnh5ZdfLrc8Ozs7wsPDlc3Gr1+/Tnp6OgDz5s2jffv2rFy5Ej8/Py5duqRct2fPHq5fvw5AVFSUMpJqZ2fHqlWrlBXXmZmZpKSkVOm+6tevT3R0tHIsJSWFzMzMCq+zsLAgJSWFw4cPA1BQUMCtW7ceuDwhhBBClCYjkY9Zjx492Lx5M3369MHIyIiuXbsqweK2bduIjY1FV1cXlUrF5MmTKy1PpVKhp6fHokWLSiysWbhwofLM47PPPsvJkydZvnw5ACEhIeU+DwkwefJkgoKCsLe3R6VSoaury+TJkzl16hQHDhxgw4YN1KtXjwkTJuDp6cm3334LQNeuXZk0aRKXL1/mlVdeUbLmjBs3jsWLF+Pi4oJKpUKlUjFx4sRKp5B1dHRYsmQJ/v7+rFixgsLCQpo0acIXX3xR4XXPPvssoaGhzJ07l+zsbLS0tPD29sbGxuaByrtfM7fx1TpfiGIFeeraboIQQtQYyVjzlMrIyKBXr14kJSWhUqlquzmEhoaSnZ2Nt7d3bTflkZPc2bWnrmaNeFJI/9ce6fvaVVf7v7KMNTIS+RQ6fvw4H3/8MRMmTHgiAsi6pqIPlHi0CvPzarsJQggh/p8EkU+4xYsX8+OPP5Y6vn79epo0afLA5Z44cUKZbr6Xm5sbgwYNqnZ5H3/88QO35WlzYdVY8u9cqe1m1EmtP44Bcmu7GUIIIZDpbCGqTYLI2tP645g6OaX0pKirU3pPAun72lVX+7+y6WxZnf0Eu3z5Mu7u7srr0NBQ8vIebjrP19eX/v3788knnzxs8x4LHx8fvvvuuxLHAgMDCQ0NraUWCSGEEAJkOvuJZmxszOrVq5XXixcvZtSoURWurK7ItWvX2LFjBwcPHkRLS/5+KFacs1sIIYQQVSeRxGNy8eJFLC0tS70u/v+CBQtwcHCgT58+HDx4sNQ1M2bMAMDV1RV7e3tu375dbl3R0dEMHDiQgQMHMmHCBDIyMsjMzGT48OHk5OTg6OjIqlWryrx23759jBs3DihaAW5ubs62bdsAWLZsGSEhIUDRaKCzszPvvvsuI0aM4M8//1TK+O677/j3v/+Ns7MzixYtKnHfcXFxuLq64uTkxJAhQzhy5Eh1u7KEgoICAgMDGTBgAAMGDCAwMFDZ39Ld3Z3du3cr59772t3dnTlz5jB48GDGj5cte4QQQojqkiDyCXDz5k0sLCyIjo5mwoQJZWap8fPzAyAiIoKYmBieeeaZMpSBx5AAACAASURBVMs6ffo0wcHBrFixgtjYWNq0acOsWbMwMDAgPDwcQ0NDYmJiGDlyZJnXd+3alaSkJNRqNQkJCVhYWJCQkADA/v37lawzY8eOJTIyks2bNzNgwAClzSdPnmTp0qVEREQQGRlZIthNS0sjLCyM5cuXExUVxezZs6s0rR4eHo69vb3y372bha9bt44TJ04QFRVFVFQUx48fZ926dZWWCZCens7atWtZtmxZlc4XQgghxN9kDu8J0LBhQ3r16gUUZVsJDAx84LISExOxtbWlefPmwN8jl1XVoEED2rRpQ1JSEvHx8Xz00UcEBQWRl5fH0aNHlXzUe/fuZe3atWRnZ5dINXjgwAFsbW1p3LgxUJRDuzhbzr59+0hLS+O9995Tzs/Pz+fatWtKzu2yjBs3Djc3N+X1vf2TkJCAo6OjMsXv5OTETz/9xLBhwyq914EDB8o0thBCCPGA5DfoY6Kjo8O9C+Fzc//epuTeZxy1tLQqzP/8OFhZWbF//36SkpKYPn06TZo0YevWrZibm1OvXj3+/PNPAgIC2LhxIyYmJhw6dAgvL68qld2jRw/mzZv3iO+giLa2NoWFhcrre/scys/ZLYQQQojKyXT2Y9K0aVPUajUXLlwAYMuWLdUuQ19fv9I8z5aWlsTFxXH16lWgaD9JGxubatVjZWVFVFQUzz33HHp6elhbWxMaGqpMZWdmZqKrq0uzZs0oLCwkIiJCufb1119n7969Sg7tTZs2Ke+98cYb7Nu3jzNnzijHkpOTq9W2+1lbWxMdHY1arUatVhMdHa3cb8uWLZWUkmfPnuXEiRMPVZcQQggh/iYjkY+Jjo4OU6ZM4f3336dx48b07Nmz2mWMGjWK4cOHU79+fVavXl3mc5GmpqZ4eXkxatQoAExMTJg5c2a16unUqRM3btxQpoStra0JCQnBysoKADMzM/r27Uu/fv0wMjLC1tZWWQxkbm7OmDFjcHV1xcDAACsrKwwNDQF46aWXCAoKYsqUKeTk5KBWq+nSpQsdO3asdl8UGzJkCGlpaTg6OgLQvXt3Bg8eDBQ9t/mf//yHn3/+mXbt2tGuXbsHrudeL46UZyhri2SsEUKIJ4dsNi5qXGZmJgYGRZuThoaGcuHChTIXCz2tJHd27amrG/4+KaT/a4/0fe2qq/0vubPFYzd//nwOHTqEWq1+oJFQIYQQQjz5JIh8Sk2bNo2kpKQSx7S1tYmKiqrS9Rs2bCiVCQZg7ty5tG3b9qHaVrwdUVXFxcUp+0/ey9PTE1tb24dqy6NQ0V9l/yT5ebncuCXTx0IIIcom09lCVNPva0eRm/nPz51tM27LEzd9U1enlJ4U0v+1R/q+dtXV/pfc2XXQ7du3S22gPWXKFGXxy6MQFRXFuXPnKj3P3d2dt956q8Tm4YcOHXpk7RJCCCHEoyHT2f9At2/fZvny5YwdO1Y5NmfOnEda56ZNmzAyMqJVq1aVnjt16lRlc/UHIbmuhRBCiNonv4mfIIcPH2bevHlkZWUB8Nlnn/HMM88wZ84csrOzadiwIVOmTKFjx45cvHgRZ2dnXF1diYuL4+7du8yZM4euXbsyc+ZM7ty5g729PQ0aNCAiIgJ3d3dGjRpFr1698PHxQU9Pj/Pnz3Pp0iUlS45KpSIzM5OAgABOnTpFbm4ulpaW+Pr6oq2tjbu7Ox06dODIkSNcuXKFd955By8vLyIjI/njjz+YPXs2X3zxBd7e3tXemxKKtg46dOgQ+vr6pV6bmZkxceJE9uzZQ48ePXBzc8PPz4+0tDQARo8ejYODAwB2dnb069eP+Ph47ty5w4gRI5SMN6mpqfj7+3Pjxg3UajUjRozA2dm5Jr59QgghRJ0iQeQT4ubNm0ycOJHQ0FC6dOlCQUEBN27cwMXFhYCAAKytrYmPj8fDw4OdO3cq11hYWDBp0iQ2b95McHAwERERTJs2DWdnZ2JiYsqt78yZM6xatQqVSoWjoyPx8fG88cYbBAQE0K1bN+bMmUNhYaESJBbvvfjXX3+xZs0asrKy6N27Ny4uLjg7OxMdHa0EqZUpDjaLff311zRp0qTS6+rVq0dkZCQAn3zyCW3atOHLL7/kypUrODk50a5dO0xNTQHIyMggKiqKa9eu4eDgQNeuXXnllVfw8vIiKCiI1q1bk5mZibOzMxYWFrRu3brS+oUQQgjxNwkinxBHjhyhdevWSm5qbW1tMjIy0NXVVTLF2NjYoKury7lz59DX13+onNu9e/emXr16ALRr1460tDTeeOMNdu3aRXJyMitXrgQgJycHY2Nj5bq+ffuipaWFoaEhrVu3Ji0tjZdeeqla9/qg09nFG4pDUc5sHx8fAJo3b46trS2JiYlKEOni4gIUZQrq2bMnBw4cQEdHh5SUFDw9PZVy1Go1qampEkQKIYQQ1SRB5FPsYXJuFweQUBSwFhQUAKDRaAgLC8PExKRa19UEbW1tJb/4/Xmu4eFzXWs0GoyMjCocoRVCCCFE1cjq7CeEhYUFKSkpHD58GICCggKaNGmCWq1m//79QNHoW35+fqWLVwwMDMjJyalWUFnMzs6O8PBwJTi8fv066enplV6nr6/PnTsPt/3BvbmuY2NjKzzX2tqa9evXA3D16lXi4uKUtIzwd87u69evExcXh6WlJa1ataJ+/fpER0cr56WkpFSaj1wIIYQQpclI5BPi2WefJTQ0lLlz55KdnY2Wlhbe3t4sWrSoxMKahQsXlhiBLK+sgQMHMnDgQBo1akRERESV2zF58mSCgoKwt7dHpVKhq6vL5MmTyx2ZLDZkyBDmzp3LihUrKl1Yc/8zkR4eHrz11lv4+voybdo0DA0N6du3b4X1TZ06lWnTpjFw4EAAvLy8aNOmjfK+kZERTk5O3Llzhw8++AAzMzMAlixZgr+/PytWrKCwsJAmTZqUaIsQQgghqkY2Gxf/OHZ2dixZskR5PlI8mCcxY01d3fD3SSH9X3uk72tXXe1/yZ0tRA3LyMiksFD+9hJCCFG3SRApatyjzMtdFbt27Xqk5f+Tcmer83K5+YSNNgohhHg6SBApatygQYMYNGhQbTfjkfl5/Qju/kNyZw8YtQ2QIFIIIUT1yepsIYQQQghRbRJEigp9//33rFq1CoCoqCg8PDwAOHr0KJ9++mkttkwIIYQQtUmms/9B8vPz0dGp2W/p0KFDyzz+6quvMn/+/BqtSwghhBBPDxmJfAqYmZmxaNEi7O3t6dOnDzt27CjxXmhoKM7OzixevJjMzEymTJmCi4sLAwcOZPbs2RQUFHDw4EEcHBxKlOvk5MSBAwe4evUq7u7uODk50b9/f+bNm6ecExoaWmY6xcTERJycnAC4ePEilpaWLFiwAAcHB/r06cPBgweVc7/77jv+/e9/4+zszKJFi7C0tCz3XrOysrC0tFQ2O+/Xrx8zZswAIDk5GVdXV6BoM/JBgwbh4OCAg4MDCQkJShkHDx5U9smcPXs2vXr14vTp0wCkpqYyZswYnJ2deffdd5Vc3EIIIYSoHhmJfEpoaWkRExNDamoqQ4cOpWvXrjRp0gQoSkVYHAxNmTKFbt26MWfOHAoLC/Hy8iIyMpLBgweTnZ3NyZMnMTc359SpU9y+fZtu3bqRl5fHkiVL0NfXR61WM3r0aPbu3cubb75Z5fbdvHkTCwsLJk2axObNmwkODiYiIoKTJ0+ydOlSYmJiaNy4MbNnz66wHH19fV5++WWOHj1KixYtqF+/Pr///jtQlLGnOCtN9+7dGTBgACqVitTUVEaOHMnevXvJy8vD09OTkJAQunbtyo8//sjq1auBopFaLy8vgoKCaN26NZmZmTg7O2NhYSG5s4UQQohqkiDyKVG82vnll1+mXbt2HDlyhLfeegsAR0dH5bxdu3aRnJzMypUrAcjJycHY2BgABwcHNm3ahK+vL5s2bcLBwQGVSkVBQQHz5s3j8OHDaDQarl27xsmTJ6sVRDZs2JBevXoBRSkci0cvDxw4gK2tLY0bNwbAxcWlSikN4+PjadGiBXZ2diQmJnLp0iXi4+MZP348AOnp6Xz66adcvnwZHR0drl27xtWrV8nIyKB+/fp07doVgLfffptnnnkGgPPnz5OSkoKnp6dSl1qtJjU1VYJIIYQQopokiPwHaNiwofK1RqMhLCyszDSFDg4ODB48GE9PT7Zs2cK6desAWLlyJbdv32bDhg3Uq1ePzz//nNzc3Gq14d5UjFpaWg+Ut7uYlZUVoaGhvPDCC7i4uKBSqdi9ezcnTpygS5cuAHh6euLj40Pv3r0pLCykU6dOlbZZo9FgZGRETEzMA7dNCCGEEEXkmcinRPF09fnz5zl+/DgWFhZlnmdnZ0d4eLjyTOH169dJT08HoEWLFrzyyivMnj2bV155hRdeeAGAO3fu0KxZM+rVq8fly5f5+eefa6zdr7/+Onv37uX69esAbNq0qdJrLCwsOHXqFIcPH6ZTp07Y2NiwbNky2rdvrwSrd+7c4V//+hdQ1Dd5eUV7Hb788svcvXtXmQL/6aefuH37NgCtWrWifv36REdHK3WlpKSQmZlZY/crhBBC1BUyEvmUKCgowMHBgbt37zJz5kzlecj7TZ48maCgIOzt7VGpVOjq6jJ58mRlZNLR0ZHPPvusxOIZd3d3/vOf/zBgwACMjY2xtrausXabm5szZswYXF1dMTAwwMrKCkNDwwqv0dPT49VXX0VbWxtdXV1effVVbt26pTwPCeDr68tHH31Eo0aN6NGjB88++6xy7fz585k+fTpQFMQ2adIEQ0NDdHR0WLJkCf7+/qxYsYLCwkKaNGnCF198Ua17emvwN9XrhCeYOq96I85CCCFEMZVGo5EkwE84MzMzDh06hL6+fm035YFkZmZiYFCUKjA0NJQLFy4QHBz8WOrbv38/vr6+/Pzzz2hp1czAu+TOrj3Nmhly9eqd2m5GnSX9X3uk72tXXe1/LS1Vhal+ZSRSPHLz58/n0KFDqNVqTExMmDlz5iOtb+fOnaxatQqNRoOenh7BwcE1FkAKIYQQooiMRIpaceLECXx8fEodd3Nz+0fn3X4QanUON2+qa7sZT4S6OhrwpJD+rz3S97Wrrva/jESKJ1Lbtm2f2lXSGze6k5V5+bHVN2LkTkCCSCGEEE8WmeN7Sly+fBl3d3fldWhoqLIi+UH5+vrSv39/Pvnkk4dt3iMXHR2Nq6srhYWFyrEvv/zyqWi7EEII8U8kQeRTwtjYWMm8ArB48WLU6gcfnbp27Ro7duwgNja22quTa4ODgwNGRkbKJuqnTp1iw4YN+Pn5PXTZD7OnpRBCCFFXSRBZC4pzTd//uqIc1PdeU5xL2tXVFXt7e2UfxLJER0creaQnTJhARkYGmZmZDB8+nJycHBwdHVm1alWZ1+7bt49x48YBkJGRgbm5Odu2bQNg2bJlhISEABAYGKjkoh4xYgR//vmnUkZFebPj4uJwdXXFycmJIUOGcOTIkQr7bdasWXzzzTecPn0aX19fpk6dir6+PoGBgbi4uPDuu+/y3//+l6ysLKDi/Np2dnYEBwfj4uLCtGnTKqxXCCGEEKVJEPmEKc5BHR0dzYQJE8rcCqd49C0iIoKYmBglrd/9Tp8+TXBwMCtWrCA2NpY2bdowa9YsDAwMCA8Px9DQkJiYGEaOHFnm9V27diUpKQm1Wk1CQgIWFhZKILZ//35lP8mxY8cSGRnJ5s2bGTBggNLm4rzZERERREZGlgh209LSCAsLY/ny5URFRTF79uxKp6abNm2Kj48PQ4cO5ZVXXqF3794sX74cQ0NDNm7cyObNm2nevDnh4eFAUX7t9evXEx0dTUhICN7e3iXKy8zMZOPGjfj7+1dYrxBCCCFKk4U1T5jyclA/iMTERGxtbWnevDnw98hlVTVo0IA2bdqQlJREfHw8H330EUFBQeTl5XH06FElBeHevXtZu3Yt2dnZJaaGK8qbvW/fPtLS0njvvfeU8/Pz87l27RpNmzYtt039+vUjMDCQUaNGAUW5wjMzM9mxYwcAeXl5mJubA+Xn127WrBlQNEUuhBBCiAcjQWQt0NHR4d6dle7N+VyTOahrgpWVFfv37ycpKYnp06fTpEkTtm7dirm5OfXq1ePPP/8kICCAjRs3YmJiwqFDh/Dy8qpS2T169CiROaeqtLW1lX0fNRoNfn5+ZWbZqSy/9r05x4UQQghRPTKdXQuaNm2KWq3mwoULAGzZsqXaZejr61ea89nS0pK4uDiuXr0KwPr167GxsalWPVZWVkRFRfHcc8+hp6eHtbU1oaGhStCWmZmJrq4uzZo1o7CwkIiICOXaivJmv/HGG+zbt48zZ84ox5KTk6vVNih6tnHVqlXk5OQo7UlJSQHKz68thBBCiIcnI5G1QEdHhylTpvD+++/TuHFjevbsWe0yRo0axfDhw6lfvz6rV68u87lIU1NTvLy8lKnfB8kW06lTJ27cuMGwYcMAsLa2JiQkRMljbWZmRt++fenXrx9GRkbY2toqi4Eqypv90ksvERQUxJQpU8jJyUGtVtOlSxc6duxYrfaNGzeOxYsX4+LigkqlQqVSMXHiRFq3bl1ufu2H5eKyuvKTapBanfNY6xNCCCGqQjLWiEfqcefNfhwkd3btqatZI54U0v+1R/q+dtXV/peMNaJWPe682UIIIYR4PCSI/AeYNm0aSUlJJY5pa2sTFRVVpes3bNjAd999V+r43Llzadu27UO1rbqbgcfFxSn7T97L09MTW1vbh2pLTanor7KalqfO4ZbkzRZCCPEEkulsIappRcxwbmc9ntzZk4btqJNTKOWpq1NKTwrp/9ojfV+76mr/Vzad/UStzp4yZYqyKONxuXHjhrJ/4vLlyx9r3fcKDQ19qD0hH4fExER++eWXSs8LDQ3FzMyMPXv2KMeysrLo3LkzTk5OQOlc4PczMzNTMs88CidOnOCHH354ZOULIYQQ/3Q1HkQ+zL6Gc+bMoWvXrjXYmsolJCTwzDPPEBMTw5gxYx5r3U+bAwcO8Ouvv1bp3Pbt25fY0mf79u20atVKeX1/LvDH7cSJE2zfvr3W6hdCCCGedjXyTKSZmRkTJ05kz5499OjRgzFjxhAQEMCpU6fIzc3F0tISX19ftLW1OXv2LL6+vty9exdzc3PS0tIYP348vXr1wt3dnVGjRtGrVy+uXbuGn58faWlpAIwePVrJMGJnZ4e9vT3x8fFcvXqVUaNG4ebmVm77CgoKCA4OZt++fUDRJtdeXl789ttvzJs3j8zMTOzt7fn888/LDWLj4uL46quvyMvLQ1dXF19fXywsLEhMTGTOnDl07NiRpKQkdHR0mDdvHosXL+bMmTM8//zzhIaG0rBhQ0JDQzl79iw3btzgypUrtGnTBn9/f2Xbm8rae+3aNZydnfn555+pV68eAB9++CH9+/enc+fOODs7M3jwYPbt20dOTg7BwcFERESQlJRE/fr1CQsLU7K1hIeHs3PnTgoKCjA2NmbWrFk0a9aM0NBQzp07x507d0hPT6dly5YsXLiQtLQ0IiIiKCwsJD4+nv79+yt5tcvy+uuvs2fPHm7dukWjRo2Ijo7G0dFRCSwvXryIs7MziYmJAOzcuZOQkBDq1avHv//9b6WciIgITp06hZ+fH8nJyQwaNIgNGzbQsWNHpk+fTtu2bRkyZAhJSUkEBwcro5ceHh707NmTjIwMPv30UzIyMoCiLYrGjx/PokWLlO97t27dmDp1arn3IoQQQojSamwksl69ekRGRvLJJ58QEBBAt27d2LhxIzExMVy/fp3IyEgAPvvsM9zc3NiyZQsjRozg6NGjZZY3e/Zs2rRpQ2xsLCtWrCA4OJjTp08r7+fk5LBu3Tq+/fZb5s+fX+HU57p16zhx4gRRUVFERUVx/Phx1q1bh5WVFR4eHtjY2BATE1NuAFlZnueUlBTee+89YmNjsbCwYPTo0fj6+vLDDz+gpaXF1q1blXN///13QkJC2L59OwYGBoSFhVW5vcbGxnTr1k2Zhr148SJ//PEHffr0AYrybr/22mtER0fj4uLCyJEjlXa1b99eWTwTExNDeno669evZ9OmTbz55pvMnTtXqf+PP/5g/vz5bNu2jfz8fGJjYzEzM8PV1RUHBwdiYmIqDCABVCoV/fr1Y+vWraSnp5OdnY2pqWmZ5167do3PP/+csLAwYmJiSmTtsba2VvJ1JyQk0LlzZ/bv36+8tra25vbt2/j5+TF//nyioqJYsmQJ06ZN4/bt28TGxtKyZUtiY2OJjY1lwoQJGBkZlfi+SwAphBBCVF+NBZGOjo7K17t27WLFihXY29vj6OjIsWPHOHfuHJmZmZw+fZqBAwcC8Oqrr2JmZlZmeQkJCbi6ugLQvHlzbG1tlVErKMqhDPCvf/2LZ555hkuXLpXbtoSEBBwdHdHT00NPTw8nJyclMKmKe/M829vb4+XlpeR5BmjVqpWyirldu3a0bduW5557Diia1i3OTAPQs2dPJTe0i4uLEhBVtb3u7u6sXbsWKBqlc3Z2VoKuhg0bKhuXt2/fnueee05pV/v27ZVR3V27dhEfH4+joyP29vasXbuWP//8U6m/e/fuPPPMM6hUKjp27KhcV12Ojo7ExMSwadOmCvNUJyUl0a5dO15++WUAhgwZorz34osvkpuby6VLl0hISGDSpEkkJCTw119/oVaradmyJYcPH+bixYuMHTsWe3t7xo4di0ql4sKFC3Tq1Im9e/cSGBjI7t27JdWhEEIIUUNqbIufe385azQawsLCMDExKXFOcZo+lUr10PUVT+dC0XY2BQUFD11mRcrL85ySklJi5ExbW7tU2+7N1/ywunTpQkFBAb///jubNm1i48aNynv3592+v13FfaTRaBg/fjwuLi5l1lFT7TcxMUFPT4/169cTGxtbYiS5OqysrNi9ezcZGRlYWloya9Ys9uzZg6WlJVB0P2ZmZqxZs6bM6zdt2kR8fDwxMTGEh4fz/fffP1A7hBBCCPG3R7I6287OjvDwcCVouX79Ounp6RgYGNCmTRslV/SxY8fKDSysra1Zv349AFevXiUuLk5JtVdd1tbWREdHo1arUavVREdHVyuHdE3leQbYs2ePkks6KiqqzHuqrL3u7u54enrSuXNnnn/++Wq3wc7OjrVr13Lr1i0A8vLyOHnyZKXXGRgYcOdO9bY48PT0xMvLCyMjo3LPsbCw4Pjx45w/fx4o2rfyXlZWVixbtozOnTsDRYH0smXLlPzdnTt35sKFCyVGdZOTk9FoNMrPXf/+/fH19eXYsWMUFhY+0L0IIYQQ4m+PZLPxyZMnExQUhL29PSqVCl1dXSZPnoyJiQmBgYFMnjyZ8PBwTE1NMTU1LbWwBGDq1KlMmzZNmfr28vKiTZs2D9SeIUOGkJaWpky5d+/encGDB1f5+prK8wzQtWtXJk2axOXLl3nllVfw8fGpdnv79+/PzJkzlXzW1eXg4MDNmzeVxUgajYahQ4dibm5e4XW9e/cmOjoae3v7ShfWFOvcubMS/JWnSZMmzJo1iw8//JD69euXWFgDRUHkZ599pgSNVlZWyjOtAI0aNSIsLIygoCD8/f2V7DhLlizhwIEDrFq1Ci0tLQoLC5kxYwZaWlpYW1vz9ddf8+677/L666/Lc5FCCCFENT32zcazsrJo2LAhKpWKs2fP4u7uzvbt22nUqNHjbEatCA0NJTs7G29v74cq5+DBg0yfPp3Y2NgaeTRAPLkkY01JdXXD3yeF9H/tkb6vXXW1/5+43NmHDx9m3rx5FMeus2bNqhMBZE2ZPHky8fHxBAYGSgBZSzIyMikslERPQggh6rZ/VNpDJyenUgtsOnXqxMyZM6t0/cPmoK5LTpw4UeZUvJubG4MGDaqFFj3dctW53L6ZV9vNeOLV1dGAJ4X0f+2Rvq9ddbX/KxuJ/EcFkUI8DrO2jeBGds3mzg5x3l4n/4Gqrrr6D/mTQvq/9kjf16662v9PVe5sIYQQQgjxdJAgsppu377NsmXLShybMmUKBw8efGR1RkVFce7cuUrPc3d3x9raukT2Hnd3d3bv3l3tOi9evMi6detKHLOzs6vWXo9r167FzMyM48ePV7v+B7Fw4UIlm48QQgghHi0JIqvp9u3bLF++vMSxOXPmlJsysSZs2rRJ2UOxMg0aNGDlypUPVV9+fj5//vlnqSCyuiIjI7GyslJSXj5KBQUF/Oc//1EyGQkhhBDi0Xrsq7NrQ/GK8OIRus8++4zu3buTnJzMnDlzyM7OpmHDhkyZMoWOHTty8eJFnJ2dcXV1JS4ujrt37yqB4syZM7lz5w729vY0aNCAiIgI3N3dGTVqFL169cLHxwc9PT3Onz/PpUuXsLCwUFZSZ2ZmEhAQwKlTp8jNzcXS0hJfX1+0tbVxd3enQ4cOHDlyhCtXrvDOO+/g5eVFZGQkf/zxB7Nnz+aLL77A29u7wo3Sx40bx6JFixg2bBiNGzcu8d61a9fw8/NT0hiOHj1aSUdoZ2dHv3792L9/P6ampiQlJXHx4kXs7e158cUXWbRoEQDbtm3j888/5+rVq4waNUrZa/J+p0+f5vr16yxcuBAXFxe8vb2VDDp2dnYMHDiQ/fv3c/nyZT799FMyMjLYsmULt27dwt/fn27dugEQFxfHV199RV5eHrq6uvj6+mJhYUFiYiKzZ8+mQ4cOHD9+nE8++YQdO3bQoUMH3NzcyMvLY8GCBezbtw8tLS1MTEz48ssvOXXqFDNmzODu3bvk5uYyePBgRo4c+eA/XEIIIUQd9Y8PIm/evMnEiRMJDQ1VUgZmZmaSl5eHh4cHAQEBWFtbEx8fj4eHBzt37lSus7CwYNKkSWzevJng4GAiIiKYNm0azs7OxMTElFvnmTNnWLVqFSqVCkdHR+Lj43njjTcICAigW7duzJkzh8LCQiVILN5I/K+//mLNmjVkZWXRu3dvXFxccHZ2Jjo6UyYbpAAAIABJREFUWglSK2NsbIy9vT1Llixh8uTJJd6bPXs2bdq04csvv+TKlSs4OTnRrl07TE1NgaK0lMVpFBMTEwkMDCy1Mj0nJ4d169Zx8eJFBg4ciKOjI/r6+qXasXHjRhwcHPjXv/5F27Zt+emnn0qMEubl5bFu3TqSk5MZPnw4//3vf9m4cSM//PADISEhfP/996SlpREWFsaKFSswMDDgzJkzjB07lj179gBw9uxZZs6cqWxmvmPHDqX88PBw0tPTiYqKQk9PT8kS9MILL7Bq1Sr09PTIyspi0KBB9OjRg9atW1fat0IIIYT42z9+OvvIkSO0bt2aLl26AEVb9jRq1Ihz586hq6urZEGxsbFBV1dXefawYcOGStBmYWFBenp6levs3bs39erVQ09Pj3bt2ikjf7t27WLFihXY29vj6OjIsWPHSjzr2LdvX7S0tDA0NKR169bKddU1btw4tmzZwl9//VXieEJCAq6urgA0b94cW1tbEhMTlfeLRyUrUhwI/utf/+KZZ57h0qVLpc5Rq9Vs2bJFybjj6OhYakq7uJz27dtz9+5d3nnnHQA6dOig3Pe+fftIS0vjvffew97eHi8vL/Lz87l27RoAL774YrnZcHbv3s2IESOU0c/iUdmcnBwmT57MwIEDGTp0KFeuXKlSykchhBBClPSPH4l8UMXBB4CWlhb5+flVvrZevXrK19ra2srelRqNhrCwMExMTKp1XXUZGRnh5uamTEFXVcOGDSs9pypt3LVrF3fu3FGmif+PvTsPqLLM////ZFO/iGgloTNpogEqZUhuuMuY5oKHTcEUNTUnZ/zgMowKGBqCsqmjONa4FOM0DpggRxDNKdNwEMwlTXMBMg82uRCiHBBZf3/w4x5RloNiB+X9+AvOue/rft8XZBfXfa7rVV5eTk5ODj///LOS9V3VjpGRUbXvH+zrIUOGEB4e/tA1srKydKr3QWvXrsXCwoLQ0FCMjY2ZOXMm9+7da3A7QgghRHP3zM9E2tvbk5WVxalTp4DKBRi3b9/GysqKkpIS0tLSgMpZutLSUqysrOpsz8zMjKKiogYNKqs4OTmxefNmZeCVm5ur0wxn69atyc9v2P5UM2bM4MiRI9Xad3R0ZOfOnQDcvHmTw4cPK/nTDzIzM0Or1TbomlXi4uIIDAzk4MGDHDx4kEOHDuHm5tbgTdsHDRpESkoKGRkZymtnzpzR6dwRI0bw97//neLiyg28qx5n5+fn06FDB4yNjbl06dITXVUvhBBCPMue+ZnIdu3aERUVRWhoKIWFhRgaGiqLUzZs2FBtYc369eurzUDW1p6zszPOzs60bduWmJgYnWvx9/cnIiIClUqFgYEBJiYm+Pv71zozWcXT05PQ0FC2bdtW78KaKqampvz+979n5cqVymvLli0jMDAQZ2dnAHx9fbG2tq7xfFtbW6ysrBg/fjxdu3bVeVbz+vXrHDt2jMjIyGqvOzs74+fnxx/+8Aed2gHo0qULERERBAQEUFRURElJCQ4ODvTq1avec+fMmcOaNWtwcXHBxMREWRw0d+5cFi9ezK5du7CyslIW8DTE+2P+3uBz6nOvRGZDhRBCPF0ksUaIBpLsbP1prqkRTYX0v/5I3+tXc+1/SawRQgghhBCN7pl/nP2s+eyzz/j0008fej00NJQePXrooaLmp66/yh5FUUkx+XnyOFsIIcTTRQaRT5mJEycyceJEfZfRrE0/EMCNu780Wnv7VB+RjwwihRBCPF2eicfZTzq7uia3bt3Cy8sLlUr1UAzirykqKoqwsDC9XV8X6enpHDlypN7jfvrpJ+bMmaMsXHJzc2tQVndNoqKilBXaAEuXLq1xJlcIIYQQDdNkZiJLS0sxNn60ckJCQhq5mvodPXoUc3PzBq3Obq6OHTtGYWEhgwcPrvO4Dz74gKFDhypRitevX3/k34kqGzduZObMmfWuuhdCCCFEw+h1EGlra8u8efM4dOgQQ4YMYfbs2bVmS2dmZuLn58fdu3fp3r07Go2GuXPnMmLEiGrZ1fXlQ6tUKlJTU+vNfobKPSUjIyNJSUkBKje+9vX15ZtvviE8PBytVotKpeL999+nT58+NbZRV/ZzSEgIvXr14vTp0xgbGxMeHs7GjRvJyMigY8eOREVFYWpqSlRUFJmZmdy6dYsbN25gbW3NqlWraNOmjU715uTk4O7uzpdffqls6v3ee+8xbtw4evfujbu7O5MmTSIlJYWioiIl4vH06dO0atWKTZs2YWFhAVTGCR44cICysjIsLS1ZuXIlFhYWREVFcfnyZfLz88nOzqZz586sX78ejUZDTEwM5eXlpKamMm7cOObMmVNjX127dg1LS0vl+/u/ruvnamtry8mTJ5X4xarvq7YZ8vLywtDQkH/84x9AZa73tGnTHso2F0IIIYTu9P44u2XLlsTFxbFgwQIlW3rXrl2o1Wpyc3OVuLzFixczdepUkpKSmD59Ot99912N7VXlQycmJrJt2zYiIyOrPRKtyn7evn07a9asoaCgoNbaYmNjOX/+PPHx8cTHx/P9998TGxvLgAED8PHxYeDAgajV6loHkFXZz1u3biU+Pp7g4GAWLFigvJ+VlcWUKVNITEzE3t6eWbNm4efnR3JyMoaGhuzdu1c59sSJE6xdu5b9+/djZmbGpk2bdK7X0tKSvn37kpycDMDVq1c5e/Yso0ePBipzwt944w0SEhLw8PBgxowZSl12dnbK41+1Wk12djY7d+5k9+7dDB06lNDQUOX6Z8+eZc2aNezbt4/S0lISExOxtbXFy8sLFxcX1Gp1rQNIgNmzZys/54iIiGobi9f3c63J8uXLAYiJiUGtVmNubg5UZptv2bKFpKQkzp07R2pqap3tCCGEEOJheh9EVuUrQ+3Z0lqtlkuXLimbZL/22mvY2trW2F59+dC6ZD/f35arqystWrSgRYsWuLm5cfToUZ3vrb7sZysrK2VFdc+ePenRowcdOnQAKjOlr1y5orQ1fPhw2rdvD4CHh4eStKNrvd7e3uzYsQOoHFS5u7srj3hNTU0ZPny4ct0OHTooddnZ2VXL/k5NTcXV1RWVSsWOHTv46aeflOsPHjwYc3NzDAwM6NWrV4OzvydMmMDBgwd5++23uXv3LtOnTycpKUm5t7p+rg1RW7a5EEIIIXSn989E3p9/XFu2dFX8XmM8cmysfGpd1ZX9fP/n9IyMjB6qrTEznR0cHCgrK+PEiRPs3r2bXbt2Ke89mBP+YF33Z3/PnTsXDw+PGq/RGPU/99xzjB07lrFjx9KhQweSkpIYP358necYGRlRtWe+Ltf8tX8HhBBCiGeR3mci71dbtrSZmRnW1tbKrNS5c+dqfZTZkHzo+jg6OpKQkEBJSQklJSUkJCToFDlY5XGynx906NAhJf85Pj6+xnuqr15vb28WLVpE79696dixY4NrcHJyYseOHdy+fRuA4uJiLly4UO95ZmZmOmV/Hzp0SBkElpWVcfHiRV566SXl3mr7uXbu3Fn5eENiYmK1Nlu3bv3IGeBCCCGEqJ3eZyLvV1e2dFhYGP7+/mzevBkbGxtsbGweWlgCDcuHro+npycajUZ55D548GAmTZqk8/mPk/38oD59+rBw4UKuX7/OK6+8wtKlSxtc77hx4wgKCuLtt99u8PUBXFxcyMvLUxYjVVRUMHnyZLp3717neSNHjiQhIQGVSlXnwpr09HTCwsIwNjamrKyMV199lfnz5wN1/1z9/PwIDAykTZs2vPXWW9XanDlzJtOmTaNVq1bKwprH9fdRjbsbQFFJcf0HCSGEEE3MU5OdXVBQgKmpKQYGBmRmZuLt7c3+/ftp27atvkt74qKioigsLGTJkiWP1c7x48dZsWIFiYmJshr5MUh2tv401/zapkL6X3+k7/WrufZ/fdnZTWomsi6nTp0iPDxc+ezbypUrm8UAsrH4+/uTmpoq29kIIYQQolE8NTORT5Kbm9tDiytef/11goKCdDo/MDCQ06dPV3vNyMiI+Pj4RqvxWXH+/PkaH8VPnTq1WcU5FpWUkJ9XpO8ynjrNdTagqZD+1x/pe/1qrv1f30ykDCKFaKAZn3/IjcLbj9VGsuvSZvkP0uNqrv+QNxXS//ojfa9fzbX/6xtENqnV2eLX8cUXXzBmzBhcXFz44Ycf9FaHk5PTY2djCyGEEEI/nprPRIrGExMTg4+PD2PGjNF3KUIIIYR4SskgsgmxtbVl4cKF/Pvf/yYvL4/Fixcr0YRff/01a9eupaysjOeff56goCBefvnlWtu6cuUKgYGB5ObmYmxszMKFCxk6dCirVq3ixIkTXL58mR07dtS67U1xcTHr1q3jm2++obi4GFtbW1asWEHr1q1ZunQpLVq04McffyQ7O5s333yTESNGEBUVxbVr15g+fTrTp08HKmcbx44dS2pqKvn5+UyfPr3GvPLa6t26dSs//fSTEmGYk5PDhAkT+PLLLzEyMqq1Rq1WW2sO+8aNG0lKSqJly5YYGBiwfft2JRJRCCGEELqRx9lNjJmZGXFxcYSHhxMcHAzAL7/8wuLFi4mMjCQxMZHx48fj6+tbZzu+vr6MHz+exMREIiIi+POf/0xubi7+/v68+uqrLFu2rM59E7du3UqbNm3YtWsXe/bs4cUXX2Tz5s3K+xkZGWzdupXk5GQSExPZs2cPn376Kf/617/4y1/+Ui2T/JdffiE+Pp5//etffPTRRzVuUF5bvR4eHhw4cEBpLzY2lvHjx/P//t//q7PG2nLY8/LyiI6OJiEhAbVazaefflotNUkIIYQQupGZyCamKtvb3t6eGzducO/ePU6fPk337t155ZVXAHB3d+eDDz5Aq9ViZvbwB161Wi3nz5/H3d0dgFdeeYUePXrw7bff4uTkpFMdBw8eRKvV8vnnnwOVM5P3byo+cuRIJR7RysqKYcOGYWhoiKWlpZJJ3q1bNwAlJrF9+/YMHz6cY8eOVWurvnqdnJxQq9VMmjSJzz77jOjo6HprPHjwIGfOnOGTTz4BoKioCEtLS9q0aUPnzp1ZvHgxgwcPZvjw4TX2oRBCCCHqJoPIJqYq19nIyAiA0tJSvdRRUVHB8uXLcXR0rPH9B/Onn2Qe9dSpU/H19eWFF16gW7dudOnSpd4aa8thB9i5cycnT54kLS0NNzc3tm7dWm/qjhBCCCGqk8fZTwF7e3suXLhAVlYWALt376Znz561zqCZmZnRo0cPdu/eDUBWVhYXLlzA3t5e52s6OTkRHR1NUVHlXoZarVa5fkNV1ZGbm8vhw4fp379/g+q1tbWlXbt2rFq1qlpkY1011pbDrtVqyc3NpV+/fvj4+GBjY1Mt21wIIYQQupGZyKfA888/T3h4OL6+vpSWlvL8888TERFR5zmRkZEEBgYSHR2NsbEx4eHhPP/88zpfc86cOWzcuBEPDw8MDAwwMDBg3rx5yiPqhnjuuedwc3MjPz+f3//+99ja2ja43okTJ7Ju3TpGjBihU4215bCbmJjwf//3fxQVFVFRUUHPnj0ZNWpUg+9JCCGEaO5ks3HxRDk5OfHRRx9hY2PzWO0EBARgZWXF7NmzG6ky/ZLEmkfTXDf8bSqk//VH+l6/mmv/PzPZ2aJ5un79OtOmTcPCwoJly5bpuxwAfvlFS3m5/O0lhBCieZOZyKfY4cOHWbt27UOvL1q0iGHDhtV7/i+//MLMmTMfev3NN99k3rx5jVKjeJjMQj665job0FRI/+uP9L1+Ndf+l+xsIRrZO/v+yY1C7SOfv9f9983yH6PG0Fz/IW8qpP/1R/pev5pr/0t2thBCCCGEaHTP7CAyICCA48eP/6rXvHXrFl5eXqhUKrZu3fqrXvt+UVFRhIWF6e36ukhPT+fIkSN1HnPlyhV69+6t7JVZUVGBo6NjtXvbv39/tW1/GiI+Pp7Lly8/0rlCCCFEc9ekF9aUlpZibPxoJYaEhDRyNfU7evQo5ubmxMTE/OrXftocO3aMwsJCBg8eXOsxL7/8Mubm5pw7d47XX3+dzMxMfvvb31b74+DYsWP069fvkWrYvXs3zz33HFZWVo90vhBCCNGcNblBpK2tLfPmzePQoUMMGTKE2bNns3r1ai5evMi9e/fo378/fn5+GBkZkZmZiZ+fH3fv3qV79+5oNBrmzp3LiBEj8Pb2ZubMmYwYMYKcnByWL1+ORqMBYNasWbi4uACVW9CoVCpSU1O5efMmM2fOZOrUqbXWV1ZWRmRkJCkpKQAMGTIEX19fvvnmG8LDw9FqtahUKt5//3369OlTYxuHDx/mww8/pLi4GBMTE/z8/LC3tyc9PZ2QkBB69erF6dOnlf0SN27cSEZGBh07diQqKgpTU1OioqLIzMzk1q1b3LhxA2tra1atWkWbNm10qjcnJwd3d3e+/PJLJW3mvffeY9y4cfTu3Rt3d3cmTZpESkoKRUVFREZGEhMTw+nTp2nVqhWbNm3CwsICgM2bN3PgwAHKysqwtLRk5cqVWFhYEBUVxeXLl8nPzyc7O5vOnTuzfv16NBoNMTExlJeXk5qayrhx45gzZ06NfdWvXz/S09N5/fXXOXbsGKNGjSI5OVmJfDx27BgBAQF11vHFF1+wfv16DA0NKSsr4/333+fq1aucPXuW4OBg/vKXv7BkyRIGDhyo0++oEEIIIZro4+yWLVsSFxfHggULWL16NX379mXXrl2o1Wpyc3OJi4sDYPHixUydOpWkpCSmT5/Od999V2N7wcHBWFtbk5iYyLZt24iMjOTSpUvK+0VFRcTGxrJ9+3bWrFlDQUFBrbXFxsZy/vx54uPjiY+P5/vvvyc2NpYBAwbg4+PDwIEDUavVtQ4gNRoNmzZtYuvWrcTHxxMcHMyCBQuU97OyspgyZQqJiYnY29sza9Ys/Pz8SE5OxtDQkL179yrHnjhxgrVr17J//37MzMzYtGmTzvVaWlrSt29fkpOTAZRB1ejRowHIy8vjjTfeICEhAQ8PD2bMmKHUZWdnx6effgqAWq0mOzubnTt3snv3boYOHUpoaKhy/bNnz7JmzRr27dtHaWkpiYmJ2Nra4uXlhYuLC2q1utYBJED//v05duwYUDnr2LdvX3r37s3x48fJzc1Fo9HQu3fvOuvYsGEDQUFBqNVq1Go1dnZ2uLu78+qrr7Js2TLUarUMIIUQQogGanIzkQCurq7K1wcPHuTMmTN88sknQOWAz9LSEq1Wy6VLl3B2dgbgtddeqzEJBSofMy9duhSAF198kWHDhpGenq5sgD127FgAXnrpJczNzbl27VqtySxHjx7F1dWVFi1aAODm5sYXX3yh8+fyUlJS0Gg0TJkyRXmttLSUnJwcAKysrOjRowcAPXv25L///S8dOnQAwM7OjitXrijnDR8+nPbt2wPg4eFBcHBwg+r19vZm9erVuLq6EhMTg7u7u3Kcqakpw4cPV67boUMHpS47OztSU1OByp/P2bNnlZ9ZWVlZtTjGwYMHY25uDkCvXr2U2WBd9e/fn1WrVlFaWsq5c+d47bXX+Pnnn0lPT6eoqIhevXrRqlWrOusYMGAAq1evZtSoUQwdOvSxNz4XQgghRBMdRJqamipfV1RUsGnTJjp16lTtGK22cosVAwODx75e1eNcACMjIyVv+UkZMmQI4eHhD72elZWlDOKqanmwtnv37jVaHQ4ODpSVlXHixAl2797Nrl27lPfur8PQ0PChuqr6qKKigrlz5+Lh4VHjNR63/k6dOtG2bVsSExPp3LkzxsbG9O3bl48//ph79+4pn4esqw5/f38uXrxIWloa8+fP55133mHSpEkNqkMIIYQQ1TXJx9n3c3JyYvPmzcqgJTc3l+zsbMzMzLC2tiYpKQmAc+fOVXtEfT9HR0d27twJwM2bNzl8+DADBgx4pHocHR1JSEigpKSEkpISEhISGvQodNCgQaSkpJCRkaG8dubMmUeq5dChQ+Tm5gKVK41ruqf66vX29mbRokX07t2bjh07NrgGJycnduzYwe3btwEoLi7mwoUL9Z5nZmZGfr5ue27169ePjz76SBkwWlhYUFhYyOHDh+nfv3+9dfzwww/Y2toyffp0JkyYoHzsoXXr1jrXIIQQQojqmuRM5P38/f2JiIhApVJhYGCAiYkJ/v7+dOrUibCwMPz9/dm8eTM2NjbY2Ng8tLAEYNmyZQQGBiqPvn19fbG2tn6kejw9PdFoNMpj08GDBzdoVqtLly5EREQQEBBAUVERJSUlODg40KtXrwbX0qdPHxYuXMj169d55ZVXlEf2Dal33LhxBAUFPfI2OS4uLuTl5SmLkSoqKpg8eTLdu3ev87yRI0eSkJCASqWqc2ENVD7STkhIoG/fvsprDg4OqNVq7O3t661jzZo1XLlyBSMjI8zNzZWV+56enoSGhrJt27YGLaz5ZMyU+g+qQ1FJyWOdL4QQQjQFT3ViTUFBAaamphgYGJCZmYm3tzf79++nbdu2+i7tiYuKiqKwsJAlS5Y8VjvHjx9nxYoVJCYmNspHA5oDyc7Wn+aaGtFUSP/rj/S9fjXX/q8vsabJz0TW5dSpU4SHh1M1Dl65cmWzGEA2Fn9/f1JTUwkLC5MBpBBCCCEa5KmeiXyS3NzcHlpg8/rrrxMUFKTT+YGBgZw+fbraa0ZGRsTHxzdajc+K8+fP1/gofurUqUycOFEPFTWeopIS8vOK9F3GM6O5zgY0FdL/+iN9r1/Ntf/rm4mUQaQQDTQzOYEbhbXvJXq/JI8pzfIfnieluf5D3lRI/+uP9L1+Ndf+r28Q2eRXZ4tfj0qloqjo0WbN4uPj8fHxabRa9JF9LoQQQgjdPdWfiRS1e5TccbVa/YSqaTh9ZJ8LIYQQQncyiNSjrVu38tNPP7F8+XIAcnJymDBhAgcOHOCvf/0r33zzDcXFxdja2rJixQpat25NYmIi27dvp+T/3yZmyZIlODo6ApV7JY4dO5a0tDRsbGyYPXu2ki1eXl6Oq6srs2bNqrUeW1tbTp48SevWrRucKQ6Qn5/Pe++9h0ajoX379kRERGBpafnQSvL7v68p17p///7Vss+XLl1KixYt+PHHH7l27Rr29vbKYiCtVltrtvrGjRtJSkqiZcuWGBgYsH37dkxMTFiyZAmZmZkYGxtjZWXF+vXrG+PHKYQQQjQrMojUIw8PD8aNG4evry+tW7cmNjaW8ePHs337dtq0aaMkyERERLB582YWLlzI4MGDGT9+PAYGBvzwww/MmDGDr7/+WmlTq9Uq5wUHB+Pk5MTvf/97AGUjbl1VZYpfvXoVZ2dnXF1dad26da3HnzhxgoSEBLp27crGjRsJCQlhw4YNdV6jKte6d+/elJWVcffu3RqPy8jIIDo6GgMDA1xdXUlNTWXQoEFKtnpISAjl5eX4+voSFxfHqFGjiI6O5siRI7Rq1QqtVkurVq346quvKCgoUDLDG9onQgghhKgkg0g9ateuHU5OTqjVaiZNmsRnn31GdHQ0vr6+aLVaPv/8c6AyfaVq8+7s7Gz+9Kc/cf36dYyNjcnJyeHmzZtYWFgAlZtuV+nbty8RERHcvXuX/v37NzilpyGZ4gBvvPEGXbt2BWDixInK5u510TXXeuTIkUqEYs+ePdFoNAwaNKjWbPU2bdrQuXNnFi9ezODBgxk+fDhmZmZ0796drKwsPvjgA/r166fkgwshhBCiYWQQqWdTp07F19eXF154gW7dutGlSxcqKipYvny58pj6fosWLWLp0qWMHDmS8vJyXn/99Wp51Pfnjo8ePRp7e3v+85//sGXLFuLi4oiMjNS5tsbKFDcyMqK8vFz5/v56dc21rq2W2rLVAXbu3MnJkydJS0vDzc2NrVu30r17d5KSkkhLS+Prr79m3bp1JCYmVmtfCCGEEPWT1dl6ZmtrS7t27Vi1apUSPejk5ER0dLSyUlqr1ZKVlQVUfu7wpZdeAiAuLo7i4uJa275y5QoWFha4ubnxxz/+UcmMflJOnjzJjz/+qNRWNfP58ssvc+7cOcrLy9FqtRw6dEg5p7Zca13Vlq2u1WrJzc2lX79++Pj4YGNjQ0ZGBteuXcPIyIiRI0fi5+dHbm4ueXl5jXL/QgghRHMiM5FNwMSJE1m3bh0jRowAYM6cOWzcuBEPDw8MDAwwMDBg3rx5dOvWDT8/P/7whz/Qtm1bhgwZQrt27Wptd9++fSQmJmJiYoKBgQH+/v5P9D4cHBwICwvjypUrysIagDfffJPk5GTGjBnDb37zG+zs7JRzasu11lVt2eomJib83//9H0VFRVRUVNCzZ09GjRpFWloaa9asAaC8vJw5c+ZgaWnZeJ0ghBBCNBOy2XgTEBAQgJWVFbNnz9Z3KaKRSWJN42quG/42FdL/+iN9r1/Ntf+f6ezsp93169eZNm0aFhYWLFu2TN/lCB398ouW8nL520sIIUTzJoNIPbK0tFRWYP9aNm7cyL///e+HXv/444954YUX6j3/cTPFnwV1/VVWVFJKfl7N2xQJIYQQzxJ5nC1EA81OPsCNwpoHins8VM3ykcevpbk+UmoqpP/1R/pev5pr/0t2thBCCCGEaHQyiGxCvvjiC8aMGYOLiws//PBDjcdcvXqV/v37K9/b2tpSUFBQa5vnz59X0lmqqFQqZfugp8V3333Hn/70JwDu3LnDli1bqr0fEBDA8ePH9VGaEEII0SzJILIJiYmJwcfHR4kObAznz59n//791V5Tq9W0atWqUdr/tbz22mvK1jx37txh69at1d4PCQmhT58++ihNCCGEaJZkENlErFq1ihMnThAZGYm3t3e12cYHZx91devWLTZs2EBqaioqlYrg4GCg+uylk5MT69atw9PTk+HDh5OYmEh0dDQeHh68+eabfPPNN0p7hw8fxsvLCzc3Nzw9Pfn222/rvL63tzchISFKW2vXrlXeu3LlCtOnT1cyuavyv+/evYuPjw9jx45lwoQJzJ8/H4DfScIsAAAgAElEQVT09HTc3NwACAoKIj8/H5VKhZeXl3Ktr776CoCcnBz++Mc/4uzsjLOzMwkJCcp1nZycWL9+PZ6enjg5OfHpp582uF+FEEIIIauzmwx/f3/Onz/PzJkzsba2xt3d/bHbfO655/Dx8eHQoUNs2LCh1uOKi4uJjY3lzJkzTJs2jT//+c/s2rWL5ORk1q5dy7/+9S80Gg2bNm1i27ZtmJmZkZGRwbvvvlstfaYmWVlZxMTEcO/ePby8vOjduzcjRozA19eXSZMmMXHiRDIzM5kyZQr79u3jxIkTFBQUKI/gb9++/VCbgYGBuLu7o1ara7xmcHAw1tbW/PWvf+XGjRu4ubnRs2dPJZe7qKiI2NhYrl69qgxiW7durWOvCiGEEAJkJlIAY8eOBcDOzo67d+8yZswYAF599VU0Gg0AKSkpaDQapkyZgkqlwtfXl9LSUnJycups28XFBWNjY1q3bs3YsWNJS0tDq9Vy/vx5ZaD8yiuv0KNHD7799lu6d+9OVlYWH3zwAfv27aNFixYNvp+jR48qM5Qvvvgiw4YNIz09/aH7femllzA3N+fatWsNvoYQQgjR3MlMZBNkbGzM/Tsv3bt374ler2XLlgAYGRlV+97Q0JDS0lLluCFDhhAeHv5Ea+nUqRNJSUmkpaXx9ddfs27dOhITExv1GlX3B5X3/OC+l0IIIYSon8xENkHt27enpKSEK1euAJCUlPTIbZmZmZGf//h7Ww0aNIiUlBQyMjKU186cOVPveXv27KG0tJTCwkL27dvHgAEDMDMzo0ePHuzevRuofOR94cIF7O3tuXbtGkZGRowcORI/Pz9yc3PJy8t76J6KioqqDXDv5+joyM6dOwG4efMmhw8fZsCAAY9660IIIYSogcxENkHGxsYEBATwzjvv8PzzzzN8+PBHbsvR0ZGPP/6YCRMm0K9fv0eOV+zSpQsREREEBARQVFRESUkJDg4O9OrVq87zunbtipeXF7dv32bMmDGMGDECgMjISAIDA4mOjsbY2Jjw8HCef/55Dh8+rKzCLi8vZ86cOVhaWvLjjz8qbbZr105ZNNO2bVtiYmKqXXPZsmUEBgbi7OwMgK+vL9bW1o903zXZOnZUre8VldQ8sBVCCCGeNZJYI54Yb29vZs6cqQwcnxWSna0/zTU1oqmQ/tcf6Xv9aq79X19ijcxECtFAkp0thBBCyCDymeHm5vbQApHXX3+doKCgJ3rdw4cPV9v/scqiRYv4xz/+8USvrS9z9qVxs7DmxJ/d7sNpfn+rCiGEaI5kEPmMiI+P18t1hw0bxrBhw/RybSGEEELoj6zObuLeffddZa/GJ+X+XOq6xMfH4+Pj06C2o6Oj+eWXXx6pruvXr+Pt7a3TsU9jHrgQQgjxNJNBZCOpbbuZx7VlyxY6d+78RK93fy51Y9u+fXutg8jy8nLqWtdlaWmp8yPxpzEPXAghhHiaySDyMdja2hIVFYW7uzsbN25Eq9USEBCAh4cHzs7OBAcHK59TzMzMZOLEiYwbN46FCxcyadIkJevZycmJS5cuKe3e//39X1dlUU+aNIm5c+cCsHnzZjw8PHB1deW9997j5s2bAHzxxRc4OzujUqkYP358tcSWB92fSw2QkJCgbKHzxz/+sdogMD8/n/fee4+xY8cybdo0rl+/Xmu7H374ITdu3MDHxweVSkVmZiZRUVH4+Pgwc+ZMxo4dy507dwgLC8Pd3Z0JEyYwffp0fvrpJ+DhzHBbW1s++ugj3N3d+d3vfsfnn39e7b3788Bry8c+fvy4cm/BwcGMGDGiWt8LIYQQQjcyiHxMLVu2JC4ujgULFrB69Wr69u3Lrl27UKvV5ObmEhcXB8DixYt5++232bt3L9OnT+e77757pOtlZ2ezY8cOtmzZglqtJjs7m507d7J7926GDh1KaGgoABs2bCAoKAi1Wo1arcbOzk6n9i9dukRkZCTbtm0jMTERa2trVq5cqbx/4sQJFi9eTHJyMv369SMkJKTWtubOncuLL77Ihg0bUKvVvPLKK0DlJuWRkZHs37+ftm3b8u677xIXF8eePXsYP348kZGRtbZpZmZGXFwc4eHhBAcH13pcVT729u3bWbNmDQUFBRQXF7No0SKWL19OYmIi/fv357///a9O/SKEEEKI6mRhzWNydXVVvj548CBnzpzhk08+ASoHMpaWlmi1Wi5duoRKpQLA3t4eGxubR7qes7MzxsbGyvXOnj2r1FBWVoaZWeX2MwMGDGD16tWMGjWKoUOH6ny99PR0hg0bxosvvgiAl5eXUjfAG2+8QdeuXQGYOHGisqF3QwwdOpTnn39e+f7rr79mx44dFBYW1vuYvir32t7enhs3bnDv3r1qMYYPHnd/PnZJSQmtWrWiT58+ALz55puYm5s3uH4hhBBCyCDysZmamipfV1RUsGnTJjp16lTtGK1WW2cbRkZGlJeXK9/XlZX94PXmzp2Lh4fHQ8f5+/tz8eJF0tLSmD9/Pu+88w6TJk2q935+Da1bt1a+/umnn1i9ejW7du2iU6dOnDx5El9f31rPfTDnu7S0tMZBpORjCyGEEE+WPM5uRE5OTmzevFkZsOTm5pKdnY2ZmRk2NjYkJiYClY9z7/8cXufOnZXH20ePHiUnJ0fn6+3YsYPbt28DUFxczIULFwD44YcfsLW1Zfr06UyYMEHnx+f9+/fn8OHDymcrd+7cycCBA5X3T548qUQQxsXF1ZtJ3bp16zqzu7VaLSYmJlhYWFBeXv5QhGFj6tq1K3fv3uXEiRNA5edG79y588SuJ4QQQjzLZCayEfn7+xMREYFKpcLAwAATExP8/f3p1KkT4eHh+Pn5sWXLFmxsbHjttdeU8+bPn8/SpUv59NNPGTBgAL/5zW90up6Liwt5eXlMnToVqJyZnDx5Mt27d2fNmjVcuXIFIyMjzM3N6/zsIoCBgQEANjY2+Pr6MnPmTAA6depUbcNyBwcHwsLCuHLlCu3btyciIqLOdqdNm4a/vz+tWrWqcQW4ra0tb731FmPHjuW5555j2LBhHD9+XKf7b6gWLVqwZs0aVqxYAUC/fv144YUXaNOmTYPa2Tym9oGzZGcLIYRoLiQ7W0+aUq50cnIycXFxbNu2Td+lPHFarVb53GhaWhp+fn58+eWXGBrqPikv2dn601zza5sK6X/9kb7Xr+ba/5KdLeoUExPDxx9/TGBgoL5L+VUcOHCA6OhoKioqaNGiBZGRkQ0aQAohhBCiksxENiPvvfceP//8c7XXOnbsyEcfffRY7X722WfV9mKsEhoaSo8ePR6r7adFUUkZ+XmF+i7jmddcZwOaCul//ZG+16/m2v/1zUTKIFKIBvrDvovcLCyp9tpn7q82y39gfm3N9R/ypkL6X3+k7/WrufZ/fYNIeY7XAHfu3GHLli3VXgsICHhiC0GgMq/68uXL9R6XlpbGxIkTUalUjBkzhmnTplXbNqihrl69SmxsbLXXHkzWqc+OHTuwtbXl+++/f+Q6GmL9+vUkJyf/KtcSQgghmjv5TGQD3Llzh61bt/Luu+8qr9W36vlx7d69m+eeew4rK6tajyktLcXHx4ft27fTvXt3AL7//ntlxfWj+Omnn4iNjcXT0/OR26jaAiguLo6ePXs+cju6KCsrY/78+U/0GkIIIYT4n2d+JvLUqVNMnjyZCRMmMGHCBI4cOQJU7tXo6emJs7Mznp6enDlzBvhfXvO6detwcXFh9OjRykxjUFAQ+fn5qFQqvLy8gMpV1lUZ2EuXLiUwMJBp06YxatQoFi9eTNWnBerK1fb29iYsLIzJkyfzu9/9Ton9i4uL4+zZswQHB6NSqUhNTa3xHgsKCigsLKR9+/bKaz179lQGkbXd64OZ2fd/HxQURFZWFiqVCh8fH+WYffv21ZhJ/aBLly6Rm5tLSEgIe/fupbi4WHnPycmJdevW4enpyfDhw0lMTCQ6OhoPDw/efPNNvvnmG+XYw4cP4+XlhZubG56ennz77bdKrc7Ozvj5+aFSqfj666+VbZKgcs/MsLAwxo8fz4QJE/jjH/8IwMWLF3n77bdxdXVl7NixREdH13oPQgghhKjdMz0TmZeXx7x584iKisLBwYGysjK0Wi3FxcX4+PiwevVqHB0dSU1NxcfHhwMHDijn2dvbs3DhQvbs2UNkZCQxMTEEBgbi7u6OWq2u9ZoZGRlER0djYGCAq6srqampDBo0SMnVDgkJoby8HF9fX+Li4pQUmZ9//pl//vOfFBQUMHLkSDw8PHB3dychIaHerYDatm3LpEmTGDVqFP369cPBwQFnZ2c6duxY773WJjAwkLCwMOLj46u9XpVJffXqVZydnXF1da2WQFNl165duLi48NJLL9GjRw+++OILJYoQKgd5sbGxnDlzhmnTpvHnP/+ZXbt2kZyczNq1a/nXv/6FRqNh06ZNbNu2DTMzMzIyMnj33Xc5dOgQAJmZmQQFBdG7d28APv/8c6X9zZs3k52dTXx8PC1atCA3NxeA3/72t0RHR9OiRQsKCgqYOHEiQ4YMoVu3bnX2hxBCCCGqe6YHkd9++y3dunXDwcEBqIy/a9u2LRcvXsTExARHR0cABg4ciImJCZcvX6Z169aYmpoqgzZ7e3vCwsJ0vubIkSOVyL2ePXui0WgYNGhQrbnaVd566y0MDQ1p06YN3bp1Q6PR0KVLF52vGxgYyDvvvENaWhpff/01f/vb34iLi+PevXu13uujqCmT+sEBWElJCUlJSUr6jKurK3FxcdUGkVVf29nZcffuXcaMGQPAq6++ikajASAlJQWNRsOUKVOU80pLS5VEn5dfflkZQD7oq6++YunSpbRo0QJAyeouKipixYoVXLx4EQMDA27cuMGFCxdkECmEEEI00DM9iHxUVQMPAENDQ0pLdU8hqS2zubZc7frOa4hOnTrRqVMnJk6cyOzZs/nqq6+qRRY+yMjIiPsX59eV2d2QOg8ePEh+fj4zZswAoLy8nJycHH7++Wc6duxYrZ2qDOyq7x/s7yFDhhAeHv7QNbKysqrliOtq7dq1WFhYEBoairGxMTNnztTpvoUQQghR3TP9mUh7e3uysrI4deoUULn44vbt21hZWVFSUkJaWhpQmVddWlpa5+IVADMzM4qKiho0qKxSW652ferLnobKz0QeOXJEGRDeuXOHq1ev8tJLL9V5r506dSI7O5vbt29TUVHB3r17q92rVqtt8H1C5Wc5AwMDOXjwIAcPHuTQoUO4ubk99Gi8PoMGDSIlJYWMjAzltarPc9ZnxIgR/P3vf1c+i1n1ODs/P58OHTpgbGzMpUuXnujKeiGEEOJZ9kzPRLZr146oqChCQ0MpLCzE0NCQJUuWMHDgQDZs2EBISAiFhYWYmpqyfv36ajOQtbXn7OyMs7Mzbdu2VR7X6qKuXO26eHp6EhoayrZt25TaH1RRUcE///lPVq5cScuWLSkrK8PZ2Zk333wToNZ7tbS05J133sHNzY327dvTt29fZcBma2uLlZUV48ePp2vXrmzYsEGn+7x+/TrHjh1TFgdVqVoE84c//EGndgC6dOlCREQEAQEBFBUVUVJSgoODA7169ar33Dlz5rBmzRpcXFwwMTHh5ZdfZsOGDcydO5fFixeza9curKys6Nu3r871CCGEEOJ/ZLNxIRqBJNb8Oprrhr9NhfS//kjf61dz7X/Jzhaikf3yi5bycvnbSwghRPMmg8iniGRUNw0P/lVWXFLGbZmFFEII0czIIPIpMnHiRCZOnKjvMpq9DZ9f53bh/1alv+/6Gz1WI4QQQujHM706WwghhBBCPBm/2iAyICDgV99O5datW3h5eaFSqdi6deuveu37RUVFNWjDcn1IT09XIiHrEhUVha2tLadPn6722qPeX1RUVLVIxPujC5+kq1evEhsb+8SvI4QQQjyrGjSIfJT9EauEhITQp0+fRz7/URw9ehRzc3PUajWzZ8/+Va/9tDl27Bj/+c9/dDr2t7/9LWvWrHms61X9Lm3cuJGSkpLHautR/PTTTzKIFEIIIR5DvZ+JtLW1Zd68eRw6dIghQ4Ywe/ZsVq9ezcWLF7l37x79+/fHz88PIyMjMjMz8fPz4+7du3Tv3h2NRsPcuXMZMWIE3t7eSgZ0Tk4Oy5cvV+LtZs2ahYuLC1C5KbdKpSI1NZWbN28yc+ZMpk6dWmt9ZWVlREZGkpKSAlQmnPj6+vLNN98QHh6OVqtFpVLx/vvv1zqIPXz4MB9++CHFxcWYmJjg5+eHvb096enphISE0KtXL06fPo2xsTHh4eFs3LiRjIwMOnbsSFRUFKampkRFRZGZmcmtW7e4ceMG1tbWrFq1ijZt2uhUb05ODu7u7nz55ZdKest7773HuHHj6N27N+7u7kyaNImUlBSKioqUPO/Tp0/TqlUrNm3ahIWFBVCZG33gwAHKysqwtLRk5cqVWFhYEBUVxeXLl8nPzyc7O5vOnTuzfv16NBoNMTExlJeXk5qayrhx45gzZ06tfT5q1ChSU1NJSUlhyJAhOt2fkZERS5cuxcjIiMuXL1NQUKDEUXp5eWFoaMg//vEPAC5dusS0adO4du2aEjtpYGCAVqut9Xfv448/Zu/evZSVldGyZUtWrFhBjx49uHv3LkuWLCEzMxNjY2OsrKxYv349QUFBXL16FZVKpewhKYQQQgjd6TQT2bJlS+Li4liwYAGrV6+mb9++7Nq1C7VaTW5uLnFxcQAsXryYqVOnkpSUxPTp0/nuu+9qbC84OBhra2sSExPZtm0bkZGRXLp0SXm/qKiI2NhYtm/fzpo1aygoKKi1ttjYWM6fP098fDzx8fF8//33xMbGMmDAAHx8fBg4cCBqtbrWAaRGo2HTpk1s3bqV+Ph4goODWbBggfJ+VlYWU6ZMITExEXt7e2bNmoWfnx/JyckYGhpWS3k5ceIEa9euZf/+/ZiZmbFp0yad67W0tKRv374kJycDlY9bz549y+jRowHIy8vjjTfeICEhAQ8PD2bMmKHUZWdnpzwCVqvVZGdns3PnTnbv3s3QoUMJDQ1Vrn/27FnWrFnDvn37KC0tJTExEVtbW7y8vHBxcUGtVtc5gAQwMDBg0aJFrFu3jge3Ga3t/qqcP3+erVu3olarWb58OQAxMTGo1WrMzc0ByMjIYMuWLSQlJXHu3DlSU1MB6vzdc3FxIS4ujoSEBObPn6+0feTIEQoKCkhOTmbPnj0EBQUBlVnj3bp1Q61WywBSCCGEeAQ6rc52dXVVvj548CBnzpzhk08+ASoHfJaWlmi1Wi5duoSzszMAr732Gra2tjW2d/ToUZYuXQrAiy++yLBhw0hPT8fGxgaAsWPHAvDSSy9hbm7OtWvX6NatW61tubq6Kmkzbm5ufPHFF7z99tu63BopKSloNBqmTJmivFZaWkpOTg4AVlZWyvY5PXv25L///S8dOnQAwM7OjitXrijnDR8+nPbt2wPg4eFBcHBwg+r19vZm9erVuLq6EhMTg7u7u3Kcqakpw4cPV67boUMHpS47OztloHXw4EHOnj2r/MzKysowM/vfljSDBw9WBmu9evVSZoMbavjw4fztb39j3759Ot8fwFtvvVVv5vXIkSOV2diePXui0WgYNGhQrb97UDk4/tvf/sbt27cxMDDgxx9/BKB79+5kZWXxwQcf0K9fP6UPhRBCCPF4dBpE3v8//YqKCjZt2vRQXF9VzrKBgcFjF1U1gAAwMjJS8qaflCFDhhAeHv7Q61lZWdWiEI2MjB6q7d69e41Wh4ODA2VlZZw4cYLdu3eza9cu5b376zA0NHyorqo+qqioYO7cuXh4eNR4jcas/09/+hMBAQG89dZbOp9T3wASav/51/a7V1xczPz58/n000+xs7Pj+vXrDB06FIBOnTqRlJREWloaX3/9NevWrSMxMVHneoUQQghRswavznZycmLz5s3K/9hzc3PJzs7GzMwMa2trkpKSADh37ly1R9T3c3R0ZOfOnQDcvHmTw4cPM2DAgEe6AUdHRxISEigpKaGkpISEhIQa86VrM2jQIFJSUpTMaIAzZ848Ui2HDh0iNzcXgPj4+Brvqb56vb29WbRoEb1796Zjx44NrsHJyYkdO3Zw+/ZtoHKAdeHChXrPMzMzIz+/YZFOffr0oUuXLtUGZQ39ebRu3Vr5A6Q+tf3uFRcXU1paqvTXjh07lHOuXbuGkZERI0eOxM/Pj9zcXPLy8jAzM9P5ukIIIYR4WIM3G/f39yciIgKVSoWBgQEmJib4+/vTqVMnwsLC8Pf3Z/PmzdjY2GBjY/PQwhKAZcuWERgYqDz69vX1xdra+pFuwNPTE41Gozy+HTx4MJMmTdL5/C5duhAREUFAQABFRUWUlJTg4OBAr169GlxLnz59WLhwIdevX+eVV15RHtk3pN5x48YRFBSk8+P4B7m4uJCXl6csRqqoqGDy5Ml07969zvNGjhxJQkICKpWq3oU191u4cGG1jzs09Ocxc+ZMpk2bRqtWrZSFNbWp63fPx8cHDw8P2rVrp3yOFODixYvKSvLy8nLmzJmDpaUlL7zwAlZWVowfP56uXbs26HORPqMtq31fXPJkZ8qFEEKIpsig4sGVEY+hoKAAU1NTDAwMyMzMxNvbm/3799O2bdvGukSTFRUVRWFhIUuWLHmsdo4fP86KFStITExslI8GiMYn2dn6Y2HRhps3GzZjLhqP9L/+SN/rV3Ptf0NDg4eifu/XqLGHp06dIjw8XFmxu3LlymYxgGws/v7+pKamKlvaiKap6j+okpJy8vJq3zlACCGEeJY16kzkk+Tm5vbQApvXX39d2bKlPoGBgdVSVqBy0UZ8fHyj1fisOH/+fI2P4qdOnSrZ3UDSvlwKC8uZ5N6+Wf5lqk/NdTagqZD+1x/pe/1qrv1f30zkUzOIFKKpkEGk/jTXf8ibCul//ZG+16/m2v/1DSJ/texs8fS4c+cOW7Zsqfbak84+j4+P5/LlyzodW1ZWxpAhQ5g7d65Ox69fv17ZxF0IIYQQjUMGkeIhd+7cYevWrdVee9LZ57t371Y2CK9PSkoKL774IidPnlQ2ha/L/PnzlQ3shRBCCNE4ZBD5DDl16hSTJ09mwoQJTJgwgSNHjnDmzBk8PT1xdnbG09NT2QPz6tWr9O/fn3Xr1uHi4sLo0aOVmcagoCDy8/NRqVR4eXkBlftXfvXVVwAsXbqUwMBApk2bxqhRo1i8eLGymEqr1RIQEICHhwfOzs4EBwcrn2X19vYmLCyMyZMn87vf/Y7IyEgA4uLiOHv2LMHBwUpuel3i4uLw8vJStiWqcvLkSVxdXZVtiqr2LF26dKkSC3n06FE8PT1xcXHB2dm5WmylEEIIIXTXqKuzhf7k5eUxb948oqKilOSbW7du4eHhwerVq3F0dCQ1NRUfHx8OHDignGNvb8/ChQvZs2cPkZGRxMTEEBgYiLu7O2q1utbrZWRkEB0djYGBAa6urqSmpjJo0CAl3zokJITy8nJ8fX2Ji4tT9or8+eef+ec//0lBQQEjR47Ew8MDd3d3EhISmDlzJiNGjKjzPnNzc0lLS2P16tV07dqV999/n9mzZwOwZcsWZs2axfjx46moqKhx8/SePXuyY8cOjIyMyMnJwc3NjcGDB8suAkIIIUQDySDyGfHtt9/SrVs3HBwcgMqV57/88gsmJiY4OjoCMHDgQExMTLh8+TKtW7fG1NRUGbTZ29sTFham8/UeJd8aKrOzDQ0NadOmDd26dUOj0dClSxedr7tnzx5GjBiBmZkZb7zxBmVlZZw6dYrevXvTv39/PvzwQ6WW119//aHzc3Nz8ff358qVKxgZGXH79m0uX76Mvb29zjUIIYQQQgaRzdqDedylpaU6n9vQfOv6ztNVXFwcubm5ODk5AZCfn09cXBy9e/dmxowZODk5kZqaysqVKxk0aBALFy6sdv6KFStwcnJi48aNGBgYMHr06EbNPxdCCCGaC/lM5DPC3t6erKwsTp06BVSuYH7hhRcoKSkhLS0NqPw8YGlpKVZWVnW2ZWZmRlFRUYMGlVVqy7euT+vWrevN7j5z5gz5+fkcOXKEgwcPcvDgQZKSkti/fz93797l8uXLdO7cGS8vL6ZNm8Z33333UBv5+fn89re/xcDAgP/85z9cuXKlwfcohBBCCJmJfGa0a9eOqKgoQkNDKSwsxNDQkCVLlrBhwwZCQkIoLCzE1NSU9evXV5uBrK0tZ2dnnJ2dadu2LTExMTrXUVe+dV08PT0JDQ1l27ZtLFmyhIEDBz50TFxcHOPGjauW5mNpaUnPnj3Zv38/3333Henp6ZiYmNCiRQuWLVv2UBt/+tOf+OCDD4iKiuK1117D1tZW53urMn7M80BlYo0QQgjRXMlm40I0kGRn609z3fC3qZD+1x/pe/1qrv0vm40LIYQQQohGJ4+zRZPz2WefKfs63i80NJQePXrooaLq2pqbciuvQN9lCCGEEHolg0jR5EycOJGJEyfqu4xaGZvIBL4QQgjx1P/f8ElnOtfk1q1beHl5oVKpHooH/DVFRUU1aG9HfUhPT+fIkSP1HqfLvdyfPNNY7k/iEUIIIYTumsRMZGlpKcbGj1ZKSEhII1dTv6NHj2Jubt6gVcvN1bFjxygsLGTw4MH6LkUIIYQQjUhvg0hbW1vmzZvHoUOHGDJkCLNnz2b16tVcvHiRe/fu0b9/f/z8/DAyMiIzMxM/Pz/u3r1L9+7d0Wg0zJ07lxEjRuDt7a3E5eXk5LB8+XI0Gg0As2bNwsXFBajcv7Aql/nmzZvMnDmTqVOn1lpfWVkZkZGRpKSkADBkyBB8fX355ptvCA8PR6vVolKpeP/99+nTp0+NbRw+fJgPP/yQ4uJiTExM8PPzw97envT0dEJCQujVqxenT5/G2NiY8PBwNm7cSEZGBh07diQqKgpTU1OioqLIzMzk1q1b3LhxA2tra1atWkWbNo9TfuUAACAASURBVG10qjcnJwd3d3e+/PJLZaPv9957j3HjxtG7d2/c3d2ZNGkSKSkpFBUVKdGHp0+fplWrVmzatAkLCwsANm/ezIEDBygrK8PS0pKVK1diYWFBVFQUly9fJj8/n+zsbDp37sz69evRaDTExMRQXl5Oamoq48aNY86cOfX+bpw8eZKVK1dSXl5OaWkpc+fOZfz48dWOOXr0KH/5y1+4d+8eZWVlyj1B5eziq6++yrfffsuNGzcYM2YMvr6+AMrvUmFhITY2NrLRuBBCCPGI9Po4u2XLlsTFxbFgwQIlc3nXrl2o1Wpyc3OJi4sDYPHixUydOpWkpCSmT59e4ybSAMHBwVhbW5OYmMi2bduIjIzk0qVLyvtFRUXExsayfft21qxZQ0FB7YsjYmNjOX/+PPHx8cTHx/P9998TGxvLgAED8PHxYeDAgajV6loHkBqNhk2bNrF161bi4+MJDg5mwYIFyvtZWVlMmTKFxMRE7O3tmTVrFn5+fiQnJ2NoaMjevXuVY0+cOMHatWvZv38/ZmZmbNq0Sed6LS0t6du3L8nJyQBcvXqVs2fPMnr0aKAyP/uNN94gISEBDw8PZsyYodRlZ2enPD5Wq9VkZ2ezc+dOdu/ezdChQwkNDVWuf/bsWdasWcO+ffsoLS0lMTERW1tbvLy8cHFxQa1W6zSAhP9lYKvVapKSkhg6dOhDx1RlYCckJPDJJ58QFhbG7du3lferMroTEhL47LPP+PHHH4HK36W3336bvXv31vm7JIQQQoi66fVxtqurq/J1bZnLWq2WS5cu4ezsDFDnBtFHjx5l6dKlALz44osMGzaM9PR0bGxsABg7diwAL730Eubm5ly7do1u3brV2parq6uyMbebmxtffPEFb7/9tk73lpKSgkajYcqUKcprpaWl5OTkAGBlZaWsNO7Zsyf//e9/6dChAwB2dnbVklSGDx9O+/btAfDw8CA4OLhB9Xp7e7N69WpcXV2JiYnB3d1dOc7U1JThw4cr1+3QoYNSl52dHampqUDlz+fs2bPKz6ysrAwzs//tHTV48GDMzc0B6NWrlzIb/CgaIwO7pozu9u3bc+nSJVQqFVCZ8lP1uyGEEEKIhtHrINLU1FT5urbMZa1WC1AtpeRRPW5uc0MNGTKE8PDwh17PysqqlhpjZGT0UG2N+ZjVwcGBsrIyTpw4we7du9m1a5fy3oP52Q/WdX8m9ty5c/Hw8KjxGo1Zf2NkYP/aP2shhBCiuWkyq7Nry1w2MzPD2tqapKQkAM6dO1ftEfX9HB0d2blzJwA3b97k8OHDDBgw4JHqcXR0JCEhgZKSEkpKSkhISKgxiq82gwYNIiUlhYyMDOW1M2fOPFIthw4dIjc3F4D4+Pga76m+er29vVm0aBG9e/emY8eODa7BycmJHTt2KI+Mi4uLuXDhQr3nmZmZ1ZuJ/aAnlYFtZmaGjY0NiYmJQOXPo7bfJSGEEELUrUmszoa6M5fDwsLw9/dn8+bN2NjYYGNj89DCEoBly5YRGBioPPr29fXF2tr6kerx9PREo9Eoj28HDx7MpEmTdD6/S5cuREREEBAQQFFRESUlJTg4ONCrV68G19KnTx8WLlzI9evXeeWVV5RH9g2pd9y4cQQFBen8OP5BLi4u5OXlKYuRKioqmDx5Mt27d6/zvJEjR5KQkIBKpdJ5Yc0//vGPJ5aBHR4ejp+fH1u2bMHGxobXXntNp/OEEEIIUd1TkZ1dUFCAqakpBgYGZGZm4u3tzf79+2nbtq2+S3vioqKiKCwsZMmSJY/VzvHjx/n/2rv3qKjq9fHjb2CAVNBEBaUs06WiGRJewAuiHssbiAgqpxNWJFTa5ejXvLEybwiIWoqZop5cXTykgFw0swsFEgjHBMzjJTMRJLyBBgMCA8zvD3/uIykwo+IgPK+1WjJ79v7sZx7KnvX5zGc/S5cuJSEh4b58NaAlq9LUSMcaA2mp/WubCsm/4UjuDaul5r+h3tlNZiayPpmZmaxevZqb9e6KFStaRAF5vyxevJjU1FRCQ0OlgLwP/iwuM3QIQgghhME9FDORjWnKlCm3bbro378/y5cv1+n6JUuWkJ2dXeuYiYkJMTEx9y3G5uLEiRN3XIp/8cUXm3Sbw78qLFRTU9Oi/7MxmJY6G9BUSP4NR3JvWC01/w3NRLb4IlIIfUkRaTgt9S/ypkLybziSe8NqqflvqIhsMruzhRBCCCHEw0OKyCbm4sWL+Pr6Kq/Dw8OprKy8pzEXLVrExIkTa3XMaapSU1MZP3688rqkpIS+ffvyxRdfKMe2b9/Ou+++e9f38PX15YcffrinOIUQQoiWTorIJsbGxobPPvtMeb1x40Y0Gs1dj3flyhUOHDhAQkICH3744f0IsVE5Ojpy/vx5pbPPzz//TL9+/UhPT1fOycjIYPDgwYYKUQghhBBIEdmozp8/j5OT022vb/75wQcfMHnyZMaOHcvhw4dvu2bZsmUA+Pj44OHhQXFxcZ33io2Nxd3dHXd3d2bPnk1hYSFqtZoZM2ZQXl6Op6cnO3bsuOO1Bw8eVJ7fWFhYiJ2dHfv37wdu9LFet24dAKGhoXh5eTFp0iReeukl8vPzlTE+//xznn/+eby8vNiwYUOtz52UlISPjw9Tpkxh+vTpZGVl1fk5HnnkEZ555hkyMjKAGwXjiy++yKlTpwCUzjvOzs5UVlYSGhqKt7c3kyZN4t1331X6oavVagIDA/H29sbd3Z2VK1fesWvNvn378PLy4sKFC3XGJIQQQojbSRFpINeuXcPBwYHY2Fhmz57NmjVrbjvn/fffByAyMpK4uDilN/Vf/frrr6xZs4bt27eTkJBAz549WbFiBRYWFkRERGBpaUlcXBwvv/zyHa8fOHAg2dnZaDQa0tLScHBwIC0tDYBDhw4xZMgQAPz9/YmOjiY+Ph43Nzcl5pMnT7JlyxYiIyOJjo6uVezm5uayadMmtm3bRkxMDCtXrmxwWd3JyUmZefzPf/6Ds7MzTz75JKdPn+b48eNYWlrStWtXtm3bhqWlJVFRUcTHx2NtbU1ERAQAwcHBDBo0iKioKOLi4igqKiI6OrrWfbZu3cru3bvZsWOH0rdcCCGEELp5KJ4T2Ry1bt2aUaNGAeDg4EBoaOhdj5Weno6rqyvW1tbA/2YuddWqVSt69uxJdnY2qampzJo1i7CwMCorK/nll19wdHQEIDk5mZ07d1JWVkZVVZVyfUZGBq6urlhZWQHg7e2ttBY8ePAgubm5/OMf/1DOr6qq4sqVK3Ts2PGO8QwePJjly5ejVqspKyvD2tqaQYMGkZGRQXl5ubKUnZiYiFqt5sCBA8CNVow3O+gkJiZy9OhRPvnkEwDKy8uxsbFR7hEeHo6trS0RERG1+oULIYQQQjdSRDYilUrFrU9QqqioUH6+tXAxNjauVZQZgrOzM4cOHSI7O5ulS5fSoUMH9u3bh52dHebm5uTn5xMcHExUVBRdu3blyJEjzJs3T6exXVxcWL16tc6xODo6kp+fzzfffKMUsIMGDeJf//oXFRUVPP/888CN1ovvv/++MlN6K61Wy6ZNm+jatesd7+Hg4MBPP/3EH3/8Qbdu3XSOTQghhBA3yHJ2I+rYsSMajYZz584BsHfvXr3HaNOmDWq1ut5znJycSEpK4vLlywDs2rWLoUOH6nUfZ2dnYmJi6Ny5M2ZmZgwZMoTw8HClQFOr1ZiamtKpUydqamqIjIxUrh08eDDJyckUFRUBsGfPHuW9YcOGcfDgQU6fPq0cO3r0aL2xmJubY29vz5YtW5RZx6effppjx47x888/K9+3HD16NDt27KC8vFyJ8cyZM8p7ERERyvcgi4qKyMvLU+7h4uLC0qVLCQgIqBWbEEIIIXQjRWQjUqlUBAYG8sorr+Dt7Y2JiYneY/j5+TFjxox6N9b06tWLefPm4efnh7u7OydPniQwMFCv+/Tv35+rV68qReOQIUPIz8/H2dkZgN69ezNu3DgmTJjA1KlTefzxx5Vr7ezsmDlzprJ5xsTEBEtLSwC6detGWFgYgYGBTJo0ifHjx/Pll182GI+TkxM5OTlKEWlqasqTTz5J27ZtlXsHBARgZ2enbJ554YUXlCJy8eLFGBsb4+Hhgbu7OzNnzuTixYu17jFkyBCCg4N54403OH78uF75EkIIIVo66Vgj7gu1Wo2FxY2n2oeHh3Pu3Lk7bhZqDqRjjeG01K4RTYXk33Ak94bVUvPfUMca+U6kuC/Wrl3LkSNH0Gg0dO3aVefe40IIIYR4OEkR+RBZsmQJ2dnZtY6ZmJgQExOj0/W7d+/m888/v+14SEgIffr0uafYbj6OSFdJSUnK8ydvNXfuXFxdXe8pFiGEEEI0PlnOFkJPspxtOC11SampkPwbjuTesFpq/htazpaNNQ+x4uJitm7dWutYYGCg0v2mMcTExHD27NkGz6upqWHz5s2MHTuW8ePHM2HCBHbt2tVocdXn3//+d53deoQQQghxd2Q5+yFWXFzMtm3b8Pf3V44FBQU16j337NlD+/bteeqpp+o97+OPP+bQoUPs2rWLdu3aUVBQwMyZMzE3N9frQej3w9///vcHej8hhBCiJZDlbAPIzMxk9erVSp/n+fPn07ZtW4KCgigrK6N169YEBgZib2/P+fPn8fLywsfHh6SkJK5fv05QUBADBw4kICCAlJQUevbsSatWrYiMjMTX1xc/Pz9GjRrFwoULMTMzIycnhwsXLiidcYyMjFCr1QQHB3Pq1CkqKipwcnJi0aJFmJiY4OvrS79+/cjKyuLSpUuMHz+eefPmER0dzcqVK7GyssLCwoIFCxbc8XmUFRUVDB48mJiYGHr06KEcT0pKYsWKFXz33XcAREVF8emnnwI3HuGzZcsWOnbsSFJSEh9//DGVlZWYmpqyaNEiHBwcuHz5MnPnzqW0tJSKigpcXV2ZP38+cGNH+NmzZykpKSEvL48nnniC9evX06pVK8LDwykrK2PBggWcOnWKZcuWcf36dSoqKpg2bVqd7SDrIsvZhtNSl5SaCsm/4UjuDaul5l92Zzcx165d48033yQ8PBxHR0eqq6u5evUq3t7eBAcHM2TIEFJTU3n77bf55ptvlGscHByYM2cO8fHxrFmzhsjISJYsWYKXlxdxcXF13u/06dPs2LEDIyMjPD09SU1NZdiwYUpv6aCgIGpqapQicdq0aQAUFBTwxRdfUFpaypgxY/D29sbLy4vY2FilSK3LuXPnMDU1rVVAwo0uMXl5eZSWlnLs2DG2bNnCzp076dSpE6WlpahUKqXX9vbt27GwsOD06dP4+/vz448/0rZtWzZv3kybNm3QaDS8+uqrJCcnM2LECACOHTtGVFQUlpaWvPrqqyQkJCif56bHHnuMHTt2YGZmRmlpKVOnTsXFxeW2WIUQQghRPykiH7CsrCx69OihtPMzMTGhsLAQU1NT5UHfQ4cOxdTUlLNnz9KmTZt76rM9ZswYzM3NAejbty+5ubkMGzaswd7S48aNw9jYGEtLS3r06EFubq7O7QEbmtzWarX8+OOPeHh40KlTJ+BGZx6ov9d269atWb16NZmZmWi1Wq5cucLJkyeVInL48OG0bdsWAHt7e3Jzc2+7d3l5OUuXLuXUqVMYGRlx6dIlTp48KUWkEEIIoScpIh8C99Jn+2YBCTcK1pttABvqLV3Xdbro1q0bGo2GM2fO1CrOsrKyePzxx5WHktelrl7bH330EcXFxezevRtzc3Pee++9Wv3I/xrzre/dtG7dOjp16kRISAgqlQo/P787nieEEEKI+snu7AfMwcGBM2fOkJmZCUB1dTUdOnRAo9Fw6NAhANLS0qiqqmpw84qFhQXl5eV6FZU3NdRbui5t2rShpKT+74WYm5vj7+/P0qVL+fPPP4Eby+NhYWG89dZbAIwcOZK4uDiuXLkCoHzPsb5e2yUlJXTq1Alzc3MuXrzI999/r/fnLikpoXPnzqhUKn799ddG3ckuhBBCNGcyE/mAPfroo4SHhxMSEkJZWRnGxsYsWLCADRs21NpYs379+lozkHWN5e7ujru7O+3atSMyMlLnOBYvXkxYWBgeHh4YGRlhamrK4sWL65yZvGn69OmEhISwffv2OjfWAMyaNQtjY2OmTp2q9AyfMWMGkydPBm70xg4ICOCVV17ByMgIMzMzNm/eXKvXdnl5ORqNBkdHR+zt7fH19eWdd97Bzc0NGxsbZflfH2+88Qbz588nKiqKp556ikGDBuk9hhBCCCFkd7YQepPd2YbTUndINhWSf8OR3BtWS82/PGxcCCGEEELcd7KcLe5aY/biFkIIIUTTJkWkuGtTp05l6tSphg5DCCGEEAYgy9kGdPHiRXx9fZXX4eHhVFZW3tOYixYtYuLEifzzn/+81/AeiIULF9K7d+9au7Hz8vKws7Pj7bffvqex09PTSUlJUV6fP38eJyenexpTCCGEEDdIEWlANjY2fPbZZ8rrjRs3otFo7nq8K1eucODAARISEvjwww/vR4gPxNNPP82ePXuU13v27KFv3773PG5GRgY//fTTPY8jhBBCiNtJEXmf/HWW6+brm39+8MEHTJ48mbFjxyrPJrz1mmXLlgHg4+ODh4cHxcXFdd4rNjZWebTP7NmzKSwsRK1WM2PGDMrLy/H09GTHjh13vPbgwYMEBAQAUFhYiJ2dHfv37wdg69atrFu3DoDQ0FC8vLyYNGkSL730Evn5+coYn3/+Oc8//zxeXl5s2LCh1udOSkrCx8eHKVOmMH36dLKyshrM3bhx4/j++++prq5Gq9Wyb98+3Nzcap0TERGBm5sbbm5uLFq0SOk7Hh4ezty5c/H392fcuHEEBARw/fp1Tp06RWRkJLGxsXh4eBAREaGMdaffhRBCCCH0I0XkA3Cz93VsbCyzZ89mzZo1t53z/vvvAxAZGUlcXJzSvu+vfv31V9asWcP27dtJSEigZ8+erFixAgsLCyIiIrC0tCQuLo6XX375jtcPHDiQ7OxsNBoNaWlpODg4kJaWBsChQ4eUZy/6+/sTHR1NfHw8bm5uSswnT55ky5YtREZGEh0dXavYvdn3etu2bcTExLBy5UqdltVbt26Ng4MDKSkppKen06tXLx599FHl/aSkJOLj44mMjCQhIYHq6mo2bdqkvH/s2DHWrl3L/v37qaqqIiEhgd69e+Pj48PkyZOJi4tTCmddfhdCCCGEaJhsrHkA7qX39V+lp6fj6uqKtbU18L+ZS121atWKnj17kp2dTWpqKrNmzSIsLIzKykp++eUXpad3cnIyO3fupKysrFZHnIyMDFxdXbGysgLA29ubhIQEoP6+1x07dqw3Lk9PT7788kvMzMzw9PTk2rVryntpaWlMmDBBaZc4bdo0Vq1apbyvS8/sm+7n70IIIYRoyWQm8j5RqVTc+tz2W/sx30vv68bg7OzMoUOHyM7OxtnZmQ4dOrBv3z7s7OwwNzcnPz+f4OBg1q5dy969e1m1apXOG35cXFyIi4tT/klJSWmwgIQbHWxOnjzJzz//zIgRI/T6PPr0+W5qvwshhBDiYSVF5H3SsWNHNBoN586dA2Dv3r16j9GmTRvUanW95zg5OZGUlMTly5cB2LVrV52tB+vi7OxMTEwMnTt3xszMjCFDhhAeHq4sZavVakxNTenUqRM1NTW12ikOHjyY5ORkioqKAGptiKmv73VDjIyMWLRoEYsWLUKlqj1BPmTIEPbv349arUar1RIVFaXTZ7awsGiwz7cQQggh7o4sZ98nKpWKwMBAXnnlFaysrBg5cqTeY/j5+TFjxgweeeQRPvvsszt+L7JXr17MmzcPPz8/ALp27cry5cv1uk///v25evUqL7zwAnCjSFu3bh3Ozs4A9O7dm3HjxjFhwgTat2+Pq6ursgHFzs6OmTNn4uPjg4WFBc7OzlhaWgLU2/daF3XNQLq6unLq1Cl8fHwA6NevH2+88UaD440ZM0bZWDNx4kQmTJigUxxCCCGEaJj0zhZ6U6vVyvcTw8PDOXfuXIvaoCK9sw2npfavbSok/4YjuTeslpr/hnpny0yk0NvatWs5cuQIGo3mrmZChRBCCPHwkyKyiVqyZAnZ2dm1jpmYmBATE6PT9Y3Z1/rm44h0lZSUpDx/8lZz587F1dX1nmIRQgghhGHIcrYQepLlbMNpqUtKTYXk33Ak94bVUvPf0HK27M4WQgghhBB6kyKyhSkuLmbr1q21jgUGBjZq+7+YmBjOnj2r07nV1dW4uLjctvs6JyeHyZMnM3nyZOLj42+77uLFi/j6+t6XeIUQQgjRMCkiW5ji4mK2bdtW61hQUBADBw5stHvu2bOHnJwcnc49ePAg1tbWHDlyhCtXrijHv/nmG5599lliY2OZNGlSrWuqqqqwsbHhs88+u59hCyGEEKIe8p3Ih0RmZiarV6+mtLQUgPnz59O2bVuCgoIoKyujdevWBAYGYm9vz/nz5/Hy8sLHx4ekpCSuX7+uFIoBAQGkpKTQs2dPWrVqRWRkJL6+vvj5+TFq1CgWLlyImZkZOTk5XLhwQWkNaGRkhFqtJjg4mFOnTlFRUYGTkxOLFi3CxMQEX19f+vXrR1ZWFpcuXWL8+PHMmzeP6OhoVq5ciZWVFRYWFixYsKDeB4W/9dZbjBgxgqysLJ566ilmzpxJfHw8oaGh1NTUYG1tTXh4OIGBgdjZ2ZGdnU27du14//338fLyIj09vc58DR8+nNDQUDIyMtBoNLRv355Vq1bx2GOP6fW7kO9EGk5L/V5SUyH5NxzJvWG11PzLI36agWvXrvHmm28SHh6Oo6Mj1dXVXL16FW9vb4KDgxkyZAipqam8/fbbfPPNN8o1Dg4OzJkzh/j4eNasWUNkZCRLlizBy8uLuLi4Ou93+vRpduzYgZGREZ6enqSmpjJs2DCCg4MZNGgQQUFB1NTUKEXitGnTACgoKOCLL76gtLSUMWPG4O3tjZeXF7GxsUqRWp+ioiIOHTpEcHAw3bt357333mPmzJlMmjSJc+fOUVZWxoIFC5Tz8/Ly2LlzJyqVivPnz9ebr5udgPz9/ZUxdu/ezZo1a/jggw/u7hcjhBBCtGBSRD4EsrKy6NGjB46OjsCNR/0UFhZiamqqtCocOnQopqamnD17ljZt2tC6dWulaLs5m6irMWPGKP2o+/btS25uLsOGDSMxMZGjR4/yySefAFBeXo6NjY1y3bhx4zA2NsbS0pIePXqQm5tLt27ddL5vfHw8o0aNwsLCggEDBlBdXU1mZibPPvvsHc93d3e/rUUi3Dlf7dq1AyA5OZmdO3dSVlYmfbOFEEKIeyBFZDNlZmam/GxsbKxXwXSzgIQbBVh1dTUAWq2WTZs20bVrV72u01V0dDRFRUWMHj0agJKSEqKjo+ssIlu3bq3X+Pn5+QQHBxMVFUXXrl05cuQI8+bN02sMIYQQQtwgG2seAg4ODpw5c4bMzEzgxg7mDh06oNFoOHToEABpaWlUVVXx1FNP1TuWhYUF5eXldzULN3r0aCIiIpTisKioiLy8vAava9OmDSUl9X+X5OjRo5SUlJCSkkJiYiKJiYns3buXr7/+muvXr+sV553y9eeff6JWqzE1NaVTp07U1NQQGRmp17hCCCGE+B+ZiXwIPProo4SHhxMSEkJZWRnGxsYsWLCADRs21NpYs379+lozkHWN5e7ujru7O+3atdOrkFq8eDFhYWF4eHhgZGSEqakpixcvrnNm8qbp06cTEhLC9u3b69xYEx0dzcSJEzEyMlKO2djY0LdvX77++mudY4S68zV06FDGjRvHhAkTaN++Pa6urnf1aCNjY6OGTxKNRvJvWJJ/w5HcG1ZLzH9Dn1l2ZwshhBBCCL3JcrYQQgghhNCbLGeLB2r37t18/vnntx0PCQmhT58+BohICCGEEHdDlrOFEEIIIYTeZDlbCCGEEELoTYpIIYQQQgihNykihRBCCCGE3qSIFEIIIYQQepMiUgghhBBC6E2KSCGEEEIIoTcpIoX4i7NnzzJ9+nTGjh3L9OnTycnJue2c6upqli1bxpgxY3juuefYvXv3gw+0mdIl/ykpKUyZMoV+/foRGhr64INsxnTJ/0cffcTEiRNxd3dnypQpHDx48MEH2gzpkvvo6Gjc3d3x8PDA3d2dTz/99MEH2kzpkv+bfv/9d/r37y9//2iFELX4+vpqY2NjtVqtVhsbG6v19fW97Zw9e/Zo/fz8tNXV1drCwkKti4uLNi8v70GH2izpkv+cnBzt8ePHtevWrdOGhIQ86BCbNV3yn5ycrC0rK9NqtVrtiRMntAMGDNBev379gcbZHOmS+5KSEm1NTY3y88iRI7UnTpx4oHE2V7rkX6vVaquqqrQvvviidu7cuS3+7x+ZiRTiFoWFhRw/fhw3NzcA3NzcOH78OEVFRbXO++qrr5g6dSrGxsZYWVkxZswYvv76a0OE3Kzomv8nn3ySPn36oFJJ0637Sdf8u7i40KpVKwB69+6NVqvl2rVrDzze5kTX3FtYWGBkZARAeXk5Go1GeS3unq75B4iIiGDkyJF069btAUfZ9EgRKcQtCgoKsLGxwcTEBAATExOsra0pKCi47TxbW1vldZcuXbhw4cIDjbU50jX/onHcTf5jY2N54okn6Ny584MKs1nSJ/fff/89EydOZNSoUcycOZPevXs/6HCbHV3zf/LkSVJSUnj55ZcNEGXTI0WkEEKIu5KRkcH69etZu3atoUNpUf72t7+xb98+Dhw4QFxcHL///ruhQ2oRNBoN7733HsuWLVOKzZZO1oKEuEWXLl24ePEi1dXVmJiYUF1dzaVLl+jSpctt5/3xxx/Y29sDt89Miruja/5F49An/5mZmbz77rts2rSJ7t27GyDa5uVu/t23tbXlmWee4ccff5TfwT3SJf+XL18mNzeXgIAAAIqLi9FqtajValasWGGo0A1KZiKFQXFUMgAABmpJREFUuEWHDh3o06cPe/fuBWDv3r306dMHKyurWueNGzeO3bt3U1NTQ1FREd999x1jx441RMjNiq75F41D1/wfPXqUOXPmsGHDBp5++mlDhNrs6Jr7M2fOKD8XFRWRnp5Or169HmiszZEu+be1tSU9PZ3ExEQSExN56aWXmDZtWostIAGMtFqt1tBBCNGUnDlzhoULF1JcXEzbtm0JDQ2le/fu+Pv78/bbb/PMM89QXV3N8uXL+emnnwDw9/dn+vTpBo68edAl/4cPH2bu3Lmo1Wq0Wi2WlpYEBQXh4uJi6PAferrk38vLi/z8fGxsbJTrVq9eLd/Nu0e65H7VqlX89NNPqFQqtFotU6dOxdfX19ChNwu65P9W4eHhlJWVsWDBAgNFbHhSRAohhBBCCL3JcrYQQgghhNCbFJFCCCGEEEJvUkQKIYQQQgi9SREphBBCCCH0JkWkEEIIIYTQmxSRQgghhBBCb1JECiGEqNfo0aNJTU01dBj4+vqye/duQ4chhPj/pO2hEEKIJk2r1SKPNBai6ZGHjQshhKjX6NGjWblyJRcuXGDXrl3Y29sTExNDu3btCAsLIycnh/Xr11NZWcn8+fPx9PQEYOHChZiZmZGXl0dWVhZPP/00oaGhPPbYYwAcOXKEoKAgcnJy6NatG4GBgTg6OgI3Zh0dHR1JT0/n+PHjPPfcc3z11VeoVCpUKhWenp4sWbKElStX8u2331JSUkK3bt1YvHgxAwcOBG50FPntt98wNzfn22+/xdbWlpCQEKXzSEFBAUFBQRw+fBitVsvEiRNZsmQJAFFRUWzfvp0rV65gb2/P8uXLlbiFEDfIcrYQQgidHT16lN69e5Oeno6bmxtz587ll19+4dtvvyUsLIzly5dTWlqqnJ+QkMCsWbNIT0/Hzs6OefPmAXDt2jVee+01fH19SU9P55VXXuG1117j6tWryrVxcXGsWLGCI0eOEBISwsCBA1myZAmZmZlKsffMM88QGxtLRkYGbm5uvPPOO1RUVChjJCYmMnHiRA4fPszo0aOVPsfV1dW89tpr2NrakpiYSHJyMhMmTADgu+++Y8uWLWzcuJG0tDQGDBjA//3f/zV6boV42EgRKYQQQmePP/44Xl5emJiYMGHCBAoKCpg9ezZmZmYMHz4cMzMzcnNzlfNHjhzJoEGDMDMzY86cOWRlZVFQUMCPP/7Ik08+yeTJk1GpVLi5udG9e3d++OEH5VpPT0969uyJSqXC1NT0jvF4eHjQvn17VCoVfn5+VFZWcvbsWeX9AQMG4OrqiomJCR4eHpw8eRK4UQxfunSJ+fPn07p1a8zNzZUZzMjISAICAujRowcqlYrXX3+dEydOkJ+f3xgpFeKhJd+JFEIIobMOHTooPz/yyCMAdOzYUTlmbm5eayayc+fOys9t2rShXbt2XLp0iUuXLmFra1trbFtbWy5evKi87tKlS4PxbN++naioKC5duoSRkRFqtbrWbOatsT3yyCNUVFRQVVVFQUEBtra2qFS3/2/wjz/+YNWqVYSGhirHtFotFy9elCVtIW4hRaQQQohGc+HCBeXn0tJS/vzzT6ytrbG2tuaPP/6odW5BQQEuLi7KayMjo3rHPnz4MNu2bWPHjh307NkTY2NjBg0apNMmnC5dulBQUEBVVdVthWSXLl14/fXXmTRpki4fUYgWS5azhRBCNJqkpCQOHz5MZWUl69evp3///nTp0gVXV1dycnJISEigqqqKr776it9++42RI0fWOVbHjh3Jy8tTXpeWlmJiYoKVlRVVVVVs3LgRtVqtU1z29vZ06tSJtWvXUlZWRkVFBT///DMAPj4+REREcPr0aQBKSkrYv3//3SdBiGZKikghhBCNxs3NjY8++ggnJyf++9//EhYWBkD79u3ZvHkzn3zyCU5OTmzbto3NmzdjZWVV51gzZszgwIEDDBo0iJUrVzJ8+HBcXFwYO3Yso0ePxtzcXKclcAATExM2b97MuXPnGDVqFCNGjFAKxeeee46ZM2cyd+5cHB0dcXNzIzk5+d6TIUQzI4/4EUII0SgWLlyIjY0Nc+bMMXQoQohGIDORQgghhBBCb1JECiGEEEIIvclythBCCCGE0JvMRAohhBBCCL1JESmEEEIIIfQmRaQQQgghhNCbFJFCCCGEEEJvUkQKIYQQQgi9SREphBBCCCH09v8ASvVK+ZK4eA4AAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Once more, education and job experience are the most important features in predicting ```case_status```."
      ],
      "metadata": {
        "id": "0skR8ezqxlEv"
      },
      "id": "0skR8ezqxlEv"
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Stacking Classifier"
      ],
      "metadata": {
        "id": "3msMRJmxx_O2"
      },
      "id": "3msMRJmxx_O2"
    },
    {
      "cell_type": "code",
      "source": [
        "mct.loc['Stacking']"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Kb9XRo0LyBX-",
        "outputId": "05742599-980d-40b4-e7dd-75df92fb9c98"
      },
      "id": "Kb9XRo0LyBX-",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Train Acc    0.754772\n",
              "Test Acc     0.749836\n",
              "Train F1     0.826446\n",
              "Test F1      0.822259\n",
              "Status        General\n",
              "Name: Stacking, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 176
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Despite stacking multiple models together, nothing was gained by implementing the stacking classifier. Accuracy is still around 75% and F1 is at 82%.\n",
        "\n",
        "It is possible that most of the models misclassified the same records, i.e., had the same false positives and false negatives. If this is the case, then stacking them together would not add any predictive power.\n",
        "\n",
        "We can get an idea from the confusion matrix."
      ],
      "metadata": {
        "id": "bsJYDyfayJiB"
      },
      "id": "bsJYDyfayJiB"
    },
    {
      "cell_type": "code",
      "source": [
        "confusion_heatmap(stack)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 365
        },
        "id": "7u90xFquyxLm",
        "outputId": "5f127702-030c-4ab1-879a-355a33336379"
      },
      "id": "7u90xFquyxLm",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x360 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Here, it appears the stacking classifier is slightly better than earlier models at predicting Denied cases. We can check whether its performance is better than random guessing with a two proportions z-test (significance=0.05)."
      ],
      "metadata": {
        "id": "j7Tuq5fHzWF5"
      },
      "id": "j7Tuq5fHzWF5"
    },
    {
      "cell_type": "code",
      "source": [
        "# z-test\n",
        "num_obs=1227+1307\n",
        "t,p_val=proportions_ztest([1227,1307],[num_obs,num_obs])\n",
        "print('The p-value is',p_val)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QiFTWQXDy7Ur",
        "outputId": "ab372386-f243-4e2b-b757-8291a3ad2e79"
      },
      "id": "QiFTWQXDy7Ur",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The p-value is 0.024607436753421648\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "With a p-value less than 0.05, we find that there is in fact a significant difference between the number of true negatives and false negatives. Thus the stacking classifier is better at predicting Denied cases than randomly guessing."
      ],
      "metadata": {
        "id": "0FIZ43OVzxZ5"
      },
      "id": "0FIZ43OVzxZ5"
    },
    {
      "cell_type": "markdown",
      "id": "nasty-retailer",
      "metadata": {
        "id": "nasty-retailer"
      },
      "source": [
        "## Actionable Insights and Recommendations"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Our model can correctly predict 75% of case statuses. While this is fairly good&mdash;certainly better than guessing&mdash;there's much room for improvement.\n",
        "* Our chosen scoring metric was F1, as it provides a good balance of recall and precision. Models maxed out at 82% for the F1 score. Meta-ensemble methods like the stacking classifier could not further improve this score.\n",
        "* Level of education has the greatest influence on case status prediction. One way for OFLC to manage the number of applications is to process an initial short form application to immediately rule out candidates. If this approach were used, level of education would surely be one of the most helpful questions to ask.\n",
        "* Previous job experience also has a big impact on case status. This too would be good to check on an initial automated screening. If the candidate passes the screening, they would be permitted to submit a complete application to be reviewed by a case reviewer at OFLC.\n",
        "* As was seen in the EDA, a higher level of education leads to a greater chance of being Certified. The ideal candidate has a Doctorate and previous job experience, is from Europe, and is seeking part-time employment in the Midwest.\n",
        "* Using machine learning models like those developed here, OFLC could in fact automate the entire screening! We could, for example, use a stacking classifier with a number of the estimators found above and logistic regression as the final estimator. Then we define two cutoffs: high probability of Certification and high probability of Denied. We can then assign statuses to the most obvious cases, leaving the more complicated cases on the edge for the reviewers at OFLC to consider. In effect, we will cut down on the number of applications that need to be reviewed by a human, while still allowing growth to the US job market."
      ],
      "metadata": {
        "id": "ZXxLTxqT1GfD"
      },
      "id": "ZXxLTxqT1GfD"
    }
  ],
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": [
        "difficult-union",
        "obvious-maine",
        "nasty-retailer"
      ],
      "include_colab_link": true
    },
    "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"
    },
    "gpuClass": "premium"
  },
  "nbformat": 4,
  "nbformat_minor": 5
}