project_notebook.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "week1_assignment.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "authorship_tag": "ABX9TyNuQKJaSFopFQ5Ma2Qnxc4n",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/samarakoon-ryan/NASANearestEarthObjectsModeling/blob/cleanup-notebook/project_notebook.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "# Predicting Nearest Earth Objects using NASA Dataset\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "c070mxOJqQHQ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Background\n",
        "\n",
        "Outer Space is vast and the number of foreign objects in space is infinite. The distances in space are numbers we cannot fathom, but NASA considers objects within 70,000 Km of Earth as Nearest Earth Objects. These objects or asteroids can be dangerous to Earth and NASA constantly monitors various parameters of these asteroids. Considering the large amount of data available on the current nearest earth objects, it would be useful to use the various parameters of these data to predict whether a nearest earth object is of any concern to Earth."
      ],
      "metadata": {
        "id": "elqVCrOqqhtY"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Project Description\n",
        "\n",
        "The NASA Nearest Earth Objects dataset includes over 90,000 datapoints on asteroids. These datapoints include distance, velocity, magnitude, and diameter among other things. All datapoints are also labeled with a boolean value for whether the asteroid is considered hazardous to Earth or not. After some cleaning and feature engineering, I plan to use 10,000 datapoints as a training dataset and another 10,000 datapoints as the test dataset. The test dataset will not include the \"hazardous\" column as this is what I'll be planning to predict based on the other columns.\n",
        "\n",
        "Dataset Source - https://www.kaggle.com/datasets/sameepvani/nasa-nearest-earth-objects\n"
      ],
      "metadata": {
        "id": "ATjsOuzgqyCa"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Basic Exploratory Data Analysis\n",
        "\n",
        "1. What is the relationship between hazardousness and relative velocity?\n",
        "2. What is the relationship between hazardousness and absolute magnitude?\n",
        "3. What is the distribution of the size of nearest Earth objects?\n",
        "    - Are there any significant outliers?\n",
        "4. What is the relationship between relative velocity and absolute magnitude?"
      ],
      "metadata": {
        "id": "egPSP2PoImLL"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Import dependencies\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "\n",
        "from sklearn.model_selection import train_test_split, GridSearchCV\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.pipeline import Pipeline \n",
        "from sklearn.compose import ColumnTransformer, make_column_selector\n",
        "from sklearn.impute import SimpleImputer\n",
        "from sklearn.preprocessing import StandardScaler, FunctionTransformer\n",
        "from sklearn import config_context\n",
        "from sklearn.metrics import classification_report, confusion_matrix, ConfusionMatrixDisplay"
      ],
      "metadata": {
        "id": "hkSVyFrQK8N1"
      },
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# load dataset\n",
        "neos = pd.read_csv(\"neo.csv\")\n",
        "\n",
        "# display head\n",
        "neos.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "4rdCQZsYLhS6",
        "outputId": "3c80630e-f6fe-436b-ff27-e66fb44e4657"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        id                 name  est_diameter_min  est_diameter_max  \\\n",
              "0  2162635  162635 (2000 SS164)          1.198271          2.679415   \n",
              "1  2277475    277475 (2005 WK4)          0.265800          0.594347   \n",
              "2  2512244   512244 (2015 YE18)          0.722030          1.614507   \n",
              "3  3596030          (2012 BV13)          0.096506          0.215794   \n",
              "4  3667127          (2014 GE35)          0.255009          0.570217   \n",
              "\n",
              "   relative_velocity  miss_distance orbiting_body  sentry_object  \\\n",
              "0       13569.249224   5.483974e+07         Earth          False   \n",
              "1       73588.726663   6.143813e+07         Earth          False   \n",
              "2      114258.692129   4.979872e+07         Earth          False   \n",
              "3       24764.303138   2.543497e+07         Earth          False   \n",
              "4       42737.733765   4.627557e+07         Earth          False   \n",
              "\n",
              "   absolute_magnitude  hazardous  \n",
              "0               16.73      False  \n",
              "1               20.00       True  \n",
              "2               17.83      False  \n",
              "3               22.20      False  \n",
              "4               20.09       True  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-8f59e68d-e361-4a5a-b2d8-82e25dc69afb\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>name</th>\n",
              "      <th>est_diameter_min</th>\n",
              "      <th>est_diameter_max</th>\n",
              "      <th>relative_velocity</th>\n",
              "      <th>miss_distance</th>\n",
              "      <th>orbiting_body</th>\n",
              "      <th>sentry_object</th>\n",
              "      <th>absolute_magnitude</th>\n",
              "      <th>hazardous</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2162635</td>\n",
              "      <td>162635 (2000 SS164)</td>\n",
              "      <td>1.198271</td>\n",
              "      <td>2.679415</td>\n",
              "      <td>13569.249224</td>\n",
              "      <td>5.483974e+07</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>16.73</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2277475</td>\n",
              "      <td>277475 (2005 WK4)</td>\n",
              "      <td>0.265800</td>\n",
              "      <td>0.594347</td>\n",
              "      <td>73588.726663</td>\n",
              "      <td>6.143813e+07</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>20.00</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2512244</td>\n",
              "      <td>512244 (2015 YE18)</td>\n",
              "      <td>0.722030</td>\n",
              "      <td>1.614507</td>\n",
              "      <td>114258.692129</td>\n",
              "      <td>4.979872e+07</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>17.83</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3596030</td>\n",
              "      <td>(2012 BV13)</td>\n",
              "      <td>0.096506</td>\n",
              "      <td>0.215794</td>\n",
              "      <td>24764.303138</td>\n",
              "      <td>2.543497e+07</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>22.20</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>3667127</td>\n",
              "      <td>(2014 GE35)</td>\n",
              "      <td>0.255009</td>\n",
              "      <td>0.570217</td>\n",
              "      <td>42737.733765</td>\n",
              "      <td>4.627557e+07</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>20.09</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8f59e68d-e361-4a5a-b2d8-82e25dc69afb')\"\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-8f59e68d-e361-4a5a-b2d8-82e25dc69afb 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-8f59e68d-e361-4a5a-b2d8-82e25dc69afb');\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": 2
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# display descriptive statistics for valid columns\n",
        "neos.describe()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "LfrVmzcsP2VH",
        "outputId": "f9b19d48-0cb4-4578-c4e5-3c94d205dd4e"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                 id  est_diameter_min  est_diameter_max  relative_velocity  \\\n",
              "count  9.083600e+04      90836.000000      90836.000000       90836.000000   \n",
              "mean   1.438288e+07          0.127432          0.284947       48066.918918   \n",
              "std    2.087202e+07          0.298511          0.667491       25293.296961   \n",
              "min    2.000433e+06          0.000609          0.001362         203.346433   \n",
              "25%    3.448110e+06          0.019256          0.043057       28619.020645   \n",
              "50%    3.748362e+06          0.048368          0.108153       44190.117890   \n",
              "75%    3.884023e+06          0.143402          0.320656       62923.604633   \n",
              "max    5.427591e+07         37.892650         84.730541      236990.128088   \n",
              "\n",
              "       miss_distance  absolute_magnitude  \n",
              "count   9.083600e+04        90836.000000  \n",
              "mean    3.706655e+07           23.527103  \n",
              "std     2.235204e+07            2.894086  \n",
              "min     6.745533e+03            9.230000  \n",
              "25%     1.721082e+07           21.340000  \n",
              "50%     3.784658e+07           23.700000  \n",
              "75%     5.654900e+07           25.700000  \n",
              "max     7.479865e+07           33.200000  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-29ef2377-c7c8-41af-bbd3-286f830fd885\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>est_diameter_min</th>\n",
              "      <th>est_diameter_max</th>\n",
              "      <th>relative_velocity</th>\n",
              "      <th>miss_distance</th>\n",
              "      <th>absolute_magnitude</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>9.083600e+04</td>\n",
              "      <td>90836.000000</td>\n",
              "      <td>90836.000000</td>\n",
              "      <td>90836.000000</td>\n",
              "      <td>9.083600e+04</td>\n",
              "      <td>90836.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>1.438288e+07</td>\n",
              "      <td>0.127432</td>\n",
              "      <td>0.284947</td>\n",
              "      <td>48066.918918</td>\n",
              "      <td>3.706655e+07</td>\n",
              "      <td>23.527103</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>2.087202e+07</td>\n",
              "      <td>0.298511</td>\n",
              "      <td>0.667491</td>\n",
              "      <td>25293.296961</td>\n",
              "      <td>2.235204e+07</td>\n",
              "      <td>2.894086</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>2.000433e+06</td>\n",
              "      <td>0.000609</td>\n",
              "      <td>0.001362</td>\n",
              "      <td>203.346433</td>\n",
              "      <td>6.745533e+03</td>\n",
              "      <td>9.230000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>3.448110e+06</td>\n",
              "      <td>0.019256</td>\n",
              "      <td>0.043057</td>\n",
              "      <td>28619.020645</td>\n",
              "      <td>1.721082e+07</td>\n",
              "      <td>21.340000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>3.748362e+06</td>\n",
              "      <td>0.048368</td>\n",
              "      <td>0.108153</td>\n",
              "      <td>44190.117890</td>\n",
              "      <td>3.784658e+07</td>\n",
              "      <td>23.700000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>3.884023e+06</td>\n",
              "      <td>0.143402</td>\n",
              "      <td>0.320656</td>\n",
              "      <td>62923.604633</td>\n",
              "      <td>5.654900e+07</td>\n",
              "      <td>25.700000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>5.427591e+07</td>\n",
              "      <td>37.892650</td>\n",
              "      <td>84.730541</td>\n",
              "      <td>236990.128088</td>\n",
              "      <td>7.479865e+07</td>\n",
              "      <td>33.200000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-29ef2377-c7c8-41af-bbd3-286f830fd885')\"\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-29ef2377-c7c8-41af-bbd3-286f830fd885 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-29ef2377-c7c8-41af-bbd3-286f830fd885');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# display column info\n",
        "neos.info()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lfc2M328Pm3A",
        "outputId": "92b26a51-f84d-4486-9ea4-7c0040487b4d"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 90836 entries, 0 to 90835\n",
            "Data columns (total 10 columns):\n",
            " #   Column              Non-Null Count  Dtype  \n",
            "---  ------              --------------  -----  \n",
            " 0   id                  90836 non-null  int64  \n",
            " 1   name                90836 non-null  object \n",
            " 2   est_diameter_min    90836 non-null  float64\n",
            " 3   est_diameter_max    90836 non-null  float64\n",
            " 4   relative_velocity   90836 non-null  float64\n",
            " 5   miss_distance       90836 non-null  float64\n",
            " 6   orbiting_body       90836 non-null  object \n",
            " 7   sentry_object       90836 non-null  bool   \n",
            " 8   absolute_magnitude  90836 non-null  float64\n",
            " 9   hazardous           90836 non-null  bool   \n",
            "dtypes: bool(2), float64(5), int64(1), object(2)\n",
            "memory usage: 5.7+ MB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Violin plots for both hazardous types looking at relative velocity\n",
        "sns.violinplot(data=neos, y='relative_velocity', x='hazardous').set(title='Relative Velocity for Hazardousness')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        },
        "id": "agzl8Go_bSiF",
        "outputId": "885138cd-8585-4887-bc7b-c630e287d4aa"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[Text(0.5, 1.0, 'Relative Velocity for Hazardousness')]"
            ]
          },
          "metadata": {},
          "execution_count": 5
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# plot mean relative velocity per hazardousness\n",
        "neos.groupby('hazardous')['relative_velocity'].mean().plot.bar(title='Average Relative Velocity for Hazardousness')\n",
        "plt.show()\n",
        "neos.groupby('hazardous')['relative_velocity'].mean().round(2)\n",
        "rv_haz = neos.groupby('hazardous')['relative_velocity'].mean()[1].round(2)\n",
        "rv_non = neos.groupby('hazardous')['relative_velocity'].mean()[0].round(2)\n",
        "\n",
        "percentage_increase = ((rv_haz - rv_non) / rv_non * 100).round()\n",
        "print(f'Hazardous NEOs on average had a {percentage_increase}% higher relative velocity.')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 328
        },
        "id": "P-8Bqu_Ysf1H",
        "outputId": "bf89874a-bcba-495a-f856-82201d2dc773"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Hazardous NEOs on average had a 35.0% higher relative velocity.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# calculate range for relatively velocity\n",
        "rv_maximum = neos.groupby('hazardous')['relative_velocity'].max()\n",
        "rv_minimum = neos.groupby('hazardous')['relative_velocity'].min()\n",
        "rv_range = rv_maximum - rv_minimum\n",
        "rv_range.plot.bar(title='Relative Velocity Range between Hazardous Types')\n",
        "rv_range.round(2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 380
        },
        "id": "_GKNs_VCtRuW",
        "outputId": "1d7293ce-05f7-4f8f-97d7-786b8b1b00a1"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "hazardous\n",
              "False    236786.78\n",
              "True     187478.68\n",
              "Name: relative_velocity, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 7
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Violin plots for both hazardous types looking at absolute magnitude\n",
        "sns.violinplot(data=neos, y='absolute_magnitude', x='hazardous').set(title='Absolute Magnitude for Hazardousness')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        },
        "id": "8Hy1F4XmTlIY",
        "outputId": "40b7e8d4-cd41-4cdb-e02d-1aa607b7e884"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[Text(0.5, 1.0, 'Absolute Magnitude for Hazardousness')]"
            ]
          },
          "metadata": {},
          "execution_count": 8
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# plot mean absolute magnitude per hazardousness\n",
        "neos.groupby('hazardous')['absolute_magnitude'].mean().plot.bar(title='Average Absolute Magnitude betweeen Hazardousness')\n",
        "plt.show()\n",
        "neos.groupby('hazardous')['absolute_magnitude'].mean().round(2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 380
        },
        "id": "A7KTl_ZYuDiK",
        "outputId": "56154bc6-b71e-41ab-94ca-22fa069eef8d"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "hazardous\n",
              "False    23.87\n",
              "True     20.31\n",
              "Name: absolute_magnitude, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# calculate range for absolute magnitude\n",
        "absmag_maximum = neos.groupby('hazardous')['absolute_magnitude'].max()\n",
        "abgmag_minimum = neos.groupby('hazardous')['absolute_magnitude'].min()\n",
        "absmag_range = absmag_maximum - abgmag_minimum\n",
        "absmag_range.plot.bar(title='Absolute Magnitude Range between Hazardousness')\n",
        "absmag_range.round(2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 380
        },
        "id": "qKKF20J8uQvg",
        "outputId": "93ce313a-de2b-4448-9f7a-eaca45373cb2"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "hazardous\n",
              "False    23.97\n",
              "True      8.36\n",
              "Name: absolute_magnitude, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 10
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Visualize distribution of NEO min size\n",
        "sns.violinplot(data=neos, y='est_diameter_min', x='hazardous').set(title='Minimum Estimated Diameter per Hazardous Type')\n",
        "\n",
        "haz_min_diam_mean = neos.groupby('hazardous')['est_diameter_min'].mean()[1].round(2)\n",
        "non_min_diam_mean = neos.groupby('hazardous')['est_diameter_min'].mean()[0].round(2)\n",
        "percent_change_min_diam_mean = ((haz_min_diam_mean - non_min_diam_mean) / non_min_diam_mean * 100).round()\n",
        "\n",
        "haz_min_diam_max = neos.groupby('hazardous')['est_diameter_min'].max()[1].round(2)\n",
        "non_min_diam_max = neos.groupby('hazardous')['est_diameter_min'].max()[0].round(2)\n",
        "\n",
        "print(f'The estimated minimum diameter on average is {percent_change_min_diam_mean}% larger for hazardous NEOs.\\n')\n",
        "print(f'The average estimated minimum diameter for hazardous NEOs is {haz_min_diam_mean}km.')\n",
        "print(f'The average estimated minimum diameter for non-hazardous NEOs is {non_min_diam_mean}km.\\n')\n",
        "\n",
        "print(f'The maximum estimated minimum diameter for hazardous NEOs is {haz_min_diam_max}km.')\n",
        "print(f'The maximum estimated minimum diameter for non-hazardous NEOs is {non_min_diam_max}km.')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 417
        },
        "id": "vE9LkfLXT5OW",
        "outputId": "039b0f1d-6e57-42e0-8cc4-ce3368cb28a0"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The estimated minimum diameter on average is 164.0% larger for hazardous NEOs.\n",
            "\n",
            "The average estimated minimum diameter for hazardous NEOs is 0.29km.\n",
            "The average estimated minimum diameter for non-hazardous NEOs is 0.11km.\n",
            "\n",
            "The maximum estimated minimum diameter for hazardous NEOs is 4.14km.\n",
            "The maximum estimated minimum diameter for non-hazardous NEOs is 37.89km.\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Visualize distribution of NEO max size\n",
        "sns.violinplot(data=neos, y='est_diameter_max', x='hazardous').set(title='Maximum Estimated Diameter per Hazardous Type')\n",
        "\n",
        "haz_max_diam_mean = neos.groupby('hazardous')['est_diameter_max'].mean()[1].round(2)\n",
        "non_max_diam_mean = neos.groupby('hazardous')['est_diameter_max'].mean()[0].round(2)\n",
        "percent_change_max_diam_mean = ((haz_max_diam_mean - non_max_diam_mean) / non_max_diam_mean * 100).round()\n",
        "\n",
        "haz_max_diam_max = neos.groupby('hazardous')['est_diameter_max'].max()[1].round(2)\n",
        "non_max_diam_max = neos.groupby('hazardous')['est_diameter_max'].max()[0].round(2)\n",
        "\n",
        "print(f'The estimated maximum diameter on average is {percent_change_max_diam_mean}% larger for hazardous NEOs.\\n')\n",
        "\n",
        "\n",
        "print(f'The maximum estimated maximum diameter for hazardous NEOs is {haz_max_diam_max}km.')\n",
        "print(f'The maximum estimated maximum diameter for non-hazardous NEOs is {non_max_diam_max}km.')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 364
        },
        "id": "_SRC7GR_W8gj",
        "outputId": "84e0b8b9-03ef-4bd6-8adb-5957724a44f8"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The estimated maximum diameter on average is 175.0% larger for hazardous NEOs.\n",
            "\n",
            "The maximum estimated maximum diameter for hazardous NEOs is 9.25km.\n",
            "The maximum estimated maximum diameter for non-hazardous NEOs is 84.73km.\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX8AAAEWCAYAAACOv5f1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3dd5xcdb3/8dd7d9MrhEggARIJIEUBjSj649IUGwgiSpMmXvR6seG9iogCXkVsWMBykRYFpCtdROoVQkkQhBCUFghJCAEJ6Vs/vz/OdzeTdctMMmdns+f9fDzmsafP58ye854z33PmjCICMzMrlrpaF2BmZn3P4W9mVkAOfzOzAnL4m5kVkMPfzKyAHP5mZgXk8K8CScslvbHWdeSpP6+jpJA0dT3mP0XS+dWsyfqepGMl/aXWdWwoBnz4S5orqUnSJp2G/zWFxuT1fY6IGBkRz67vcqpN0sVp3ZeXPB4tY767JH2qdFhe65j3DpvWZbWkZZKWSpol6WRJQ9qniYgzI+JTPS0np9pOl3RJXz/v+kjb1Lc7DZuc9qWGWtVVC532qzZJq0r6j6x1fb0Z8OGfPAcc3t4j6c3A8NqV06e+n4K7/bFzrQuqgRMjYhSwGfBl4DDgZkmqbVnrJ++w7c9h3h9qK92vgBeAA0qGXVrr+npTlPD/LXB0Sf8xwG9KJ5D0ofRpYKmkeZJOLxl3qKTnJI1O/R+Q9JKk8am/o9khHRn9QtIt6QjgXkkTJP1E0muSnpS0a8my12qyKD2ykrSXpBclfUXSy5IWSjpI0gcl/UPSPyWdsi4viKShki6R9KqkJZIekrSppO8AewDnpvrPrcI6nizpmXT0/YSkj6Th2wO/AnZPy1mShg+R9ENJL0haJOlXkoaVLO+/02uxQNIny13niFgREXcBHwZ2Bz6UlrfWEbikq9L/93VJ90jasWRcpeu+uaRrJC1O29Dn0/D3A6cAh6rkE5mkMZIuSOs3X9K3JdWnccem5/uxpFeB07v4v54u6WpJV6TX+2FJO/dWT6d5L5G0FDi23Ne2Uw097Uvt21X7o6V9fHfbSXfrLmmcpOvT8zwIbN2pjnel7fr19PddJePmSnpPp3W/JHV3uW+Uue6Dle2Xby4Z9gZJKyWN15p9+hRJr6Q6jiyZtsdtv5qKEv73A6MlbZ92pMOAzh+3V5C9QYwlC4X/kHQQQERcAdwH/EzSOOAC4FMRsbib5/s4cCqwCdAIzAAeTv1XA2dXUPsEYCgwEfgm8GvgE8DbyEL6G5KmVLC8dscAY4AtgHHAZ4BVEfF14P/IjpZHRsSJ3cxfyTo+k2odA5wBXCJps4iYk553RnqusWn6s4BtgV2AqSXr3h6a/wW8F9gGeA8ViogXgJmppq7ckpb9hrROnY/iylp3SXXADcCjaR32Bb4o6X0R8UfgTOCKTp/ILgZa0nrvCuwHlDZJvQN4FtgU+E439R8IXAVsDFwG/EHSoJ7q6TTv1WT7wboevfa0L51YcrT8/4DXgOvSfF1uJz2s+8+B1WSf6D6ZHgBI2hi4CfgZ2fZ9NnBT2n970+W+Uc6KR0QTcDnZPtrucOD2kryYQLatTEzPdZ6k7dK4brf9qouIAf0A5pIFxKnAd4H3A7cBDUAAk7uZ7yfAj0v6x5J9tHsM+N9O0wYwNXVfDPy6ZNzngDkl/W8GlnQ1b8n8307de5FtdPWpf1Sa/h0l088CDupmHS4m2zmWlDymp3GfJHtDe0sX891F9uZWlXXsYvmPAAem7mOBv5SME1l4bF0ybHfgudR9IXBWybhtO7+Gva1LGn55+zqQHUFf0s38Y9Pyx1S67mRh9UKn5X0NuKir5yULtUZgWMmww4E7S16rF7qqs2T604H7S/rrgIVkoVpOPff0svyutqml6TVqKGdfSsPGk+2bh1WwnbxQMq4eaAbeVDLszPZtCTgKeLDT8mYAx6buucB7Or1ul/S2b/RQa8fy2l9nQKl/JvDxWLNPtwAjSua9EvgGvWz71X7UvN2sD/0WuAeYQqcmHwBJ7yB7190JGAwMITt6AiAilki6CjgJ+Ggvz7WopHtVF/0jK6j71YhoLZm3q+X3tLwfRsSpXQz/LdmRzeWSxpJ9Evp6RDSXWVfZ6yjpaLLXbXIaNJLsyKcr48nOx8zSmiZ5ke3sAJuTveG1e77MejubSLaDryV9MvwO8LFUS1satQnweuoud923AjZXas5K6sk+WXVlK2AQsLBk3euAeSXTzOs8Uxc6pomINkkvkr1uUUY95Sx/rW1K2UUTz5X097gvSRpE9unisoi4vGR4b9tJaW3jyQ7gSoeVbgub86/bxvNk//ferNe+EREPSFoJ7CVpIdkR/PUlk7wWESs61bU5vW/7VVWUZh8i4nmyDfSDwLVdTHIZ2T9oi4gYQ9YWveY/IO1CdkTwO7KPktWykrVPPk+o4rK7FRHNEXFGROwAvAvYnzXnRap2q1dJW5E1VZ0IjIusaedx1ry2nZ/rFbIA3TEixqbHmMiaCSA7it2iZPot16GmLciazboK4SPImj7eQ/bRf3L7bJU+D1kwPVeyHmMjYlREfDCN77zu88iO/DcpmX50ROxYMk05/5uO1yc19UwCFpRRT7nL702P+xJwDtmnhdI3kN62k861LSY7gu5uW1hA9mZKp/HzU/cKutnvetk3yjWdrOnnKODqiFhdMm4jSSM61bWA3rf9qipM+CfHA/t0etdtNwr4Z0SslrQbWQgA2Qkgsnf/U4DjgImSPlulmh4BjpBUn9qz96zScnskaW9Jb05HukvJPkK3H+UuAqp1Tf8Isp12cXre48iOCNstAiZJGgzZkSpZCPxY0hvSPBNL2qWvBI6VtIOk4cBp5RYiabikPcnamB8Ebu5islFkAfwqWTicWe7yu/AgsEzSVyUNS//jnSS9PY1fBExOAU1ELAT+BPxI0mhJdZK2TjVX4m2SDlZ2RcwX0/rcX0Y91dLTvvRpsm38yPS/btfbdrKW9Gn4WrITv8Ml7UDWft7uZmBbSUdIapB0KLADcGMa/whwWDoXMg04pKTGnvaNcl0CfITsDeBfWhqAM9LJ4T3I3lyuKmPbr6pChX9EPBMRM7sZ/VngW5KWkZ1gubJk3HeBeRHxy4hoJPuHflvSNlUo6wvAAWRtp0cCf6jCMkt9RWtfXfFKGj6B7KP3UmAOcDfZx12AnwKHKLt6Zb0+5UTEE8CPyNpbF5G1id9bMskdwGzgpZLavgo8Ddyv7KqTPwPbpeXdQtaGfEea5o4yyjg3/V8XpXmvAd7fKXza/YbsY/h84Amy0FwnKaD2Jzt59xzZkd35ZJ8oYE1TyKuSHk7dR5M1lTxBdjL0arITmpW4Djg0zX8UcHA6mu2tnmrpaV86nOzAYkHJNnlKGdtJV04kaxp6iexcxEXtIyLiVbJ1/TLZG/lXgP0jon0b+wbZ1UGvkZ1cvqxkuT3tG2WJiHlkFwEE//oJ86X0vAvITqp/JiKeTOO63farrf2EhJkNAMoum5waEZ/obVrLl6QLgQWdzo/sRXZieVLNCkuKdMLXzKxPpJPgB5NdrtsvFarZx8wsb5L+h+xk9Q8i4rnepq8VN/uYmRWQj/zNzApog2nz32STTWLy5Mm1LsPMbIMya9asVyJifOfhG0z4T548mZkzu7tK08zMuiKpy2/Bu9nHzKyAHP5mZgXk8DczKyCHv5lZATn8zcwKyOFvZlZADn8zswJy+BeQb+lhZg7/Avr617/Oz3/+81qXYWY15PAvoPvuu4+rrrqq9wnNbMBy+JuZFZDD38ysgBz+ZmYF5PA3Mysgh7+ZWQE5/M3MCij38Jf0JUmzJT0u6XeShkqaIukBSU9LukLS4LzrMDOzNXINf0kTgc8D0yJiJ6AeOAz4HvDjiJgKvAYcn2cdZma2tr5o9mkAhklqAIYDC4F9gKvT+OnAQX1Qh5mZJbmGf0TMB34IvEAW+q8Ds4AlEdGSJnsRmJhnHWZmtra8m302Ag4EpgCbAyOA91cw/wmSZkqauXjx4pyqNDMrnrybfd4DPBcRiyOiGbgWeDcwNjUDAUwC5nc1c0ScFxHTImLa+PHjcy7VzKw48g7/F4B3ShouScC+wBPAncAhaZpjgOtyrsPMzErk3eb/ANmJ3YeBx9LznQd8FThJ0tPAOOCCPOswM7O1NfQ+yfqJiNOA0zoNfhbYLe/nNjOzrvkbvmZmBeTwNzMrIIe/mVkBOfzNzArI4W9mVkAOfzOzAnL4m5kVkMPfzKyAHP5mZgXk8DczKyCHv5lZATn8zcwKyOFvZlZADn8zswJy+JuZFZDD38ysgBz+ZmYF5PA3Mysgh7+ZWQE5/M3MCsjhb2ZWQA5/M7MCcvibmRWQw9/MrIAc/mZmBeTwNzMrIIe/mVkBOfzNzArI4W9mVkAOfzOzAnL4m5kVkMPfzKyAHP5mZgXk8DczKyCHv5lZATn8zcwKyOFvZlZADn8zswLKPfwljZV0taQnJc2RtLukjSXdJump9HejvOswM7M1yg5/Scd36q+XdFoZs/4U+GNEvAnYGZgDnAzcHhHbALenfjMz6yOVHPnvK+lmSZtJ2hG4HxjV0wySxgD/BlwAEBFNEbEEOBCYniabDhxUceVmZrbOGsqdMCKOkHQo8BiwAjgiIu7tZbYpwGLgIkk7A7OALwCbRsTCNM1LwKZdzSzpBOAEgC233LLcUs3MrBeVNPtsQxbc1wDPA0dJGt7LbA3AW4FfRsSuZG8aazXxREQA0dXMEXFeREyLiGnjx48vt1QzM+tFJc0+NwDfiIhPA3sCTwEP9TLPi8CLEfFA6r+a7M1gkaTNANLflyuq2szM1ksl4b9bRNwO2dF6RPwI+EhPM0TES8A8SdulQfsCTwDXA8ekYccA11VUtZmZrZdK2vyXStoJ2AEYWjLqH73M+jngUkmDgWeB48jedK5MVxA9D3y8oqrNzGy9lB3+6bLOvcjC/2bgA8BfgN/0NF9EPAJM62LUvmVXaWZmVVVJs88hZIH9UkQcR3bN/phcqjIzs1xVEv6rIqINaJE0muwk7Rb5lGVmZnkqu9kHmClpLPBrsuv1lwMzcqnKzMxyVckJ38+mzl9J+iMwOiL+lk9ZZmaWp0qO/JH0FmBy+3ySpkbEtTnUZWZmOarkap8LgbcAs4G2NDgAh7+Z2QamkiP/d0bEDrlVYmZmfaaSq31mSHL4m5kNAJUc+f+G7A3gJaARENmdHt6SS2VmZpabSsL/AuAosls6t/UyrZmZ9WOVhP/iiLg+t0rMzKzPVBL+f5V0GdmtnRvbB/pSTzOzDU8l4T+MLPT3KxnmSz3NzDZAlXzD97iexkv6WkR8d/1LMjOzvFVyqWdvPlbFZZmZWY6qGf6q4rLMzCxH1Qz/Ln+E3czM+h8f+ZuZFVBZ4S+pXtKXepnsqirUY2ZmfaCs8I+IVuDwXqY5syoVmZlZ7iq5zv9eSecCVwAr2gdGxMNVr8rMzHJVSfjvkv5+q2RYAPtUrxwzM+sLlXzJa+88CzEzs75T9tU+kjaVdIGkW1L/DpKOz680MzPLSyWXel4M3Apsnvr/AXyx2gWZmVn+Kgn/TSLiStK9/COiBWjNpSozM8tVJeG/QtI40jd5Jb0TeD2XqszMLFeVXO1zEnA9sLWke4Hx+GZuZmYbpErCfzawJ7Ad2a0c/k51bw9hZmZ9pJLwnhERLRExOyIej4hmYEZehZmZWX56PfKXNAGYCAyTtCtrbuA2GhieY21mZpaTcpp93gccC0wCzi4ZvhQ4JYeazMwsZ72Gf0RMB6ZL+mhEXNMHNZmZWc4qafO/19/wNTMbGCoJ/4vwN3zNzAYEf8PXzKyA/A1fM7MCWt9v+B5SzoyS6oGZwPyI2F/SFOByYBwwCzgqIpoqqtzMzNZZ2Uf+6Re79gTeBXwa2DEi/lbm7F8A5pT0fw/4cURMBV4DfOLYzKwPVXI//3rgg8C+wH7A5ySdVMZ8k4APAeenfpH9+tfVaZLpwEGVlW1mZuujkmafG4DVwGOkk75l+gnwFWBU6h8HLEknjAFeJPsGsZmZ9ZFKwn9SRLylkoVL2h94OSJmSdqrosqy+U8ATgDYcsstK53dzMy6UcnVPrdI2q/C5b8b+LCkuWQnePcBfgqMldT+xjMJmN/VzBFxXkRMi4hp48ePr/CpzcysO5WE//3A7yWtkrRU0jJJS3uaISK+FhGTImIycBhwR0QcCdzJmiuFjgGuW4fazcxsHVUS/mcDuwPDI2J0RIyKiNHr+LxfBU6S9DTZOYAL1nE5Zma2Dipp858HPB4RsS5PFBF3AXel7meB3dZlOWZmtv4qCf9ngbvSjd0a2wdGxNndz2JmZv1RJeH/XHoMTg8zM9tAlR3+EXFGnoWYmVnfKTv8JY0n+7LWjsDQ9uERsU8OdZmZWY4qudrnUuBJYApwBjAXeCiHmszMLGeVhP+4iLgAaI6IuyPik2Rf2jIzsw1MJSd8m9PfhZI+BCwANq5+SWZmlrdKwv/bksYAXwbOAUYDX8qlKjMzy1UlV/vcmDpfB/bOpxwzM+sLvYa/pK9ExPclnUP6CcdSEfH5XCozM7PclHPk3/4LXDPzLMTMzPpOr+EfETekv9PzL8fMzPpCOc0+N9BFc0+7iPhwVSsyM7PcldPs88P092BgAnBJ6j8cWJRHUWZmlq9ymn3uBpD0o4iYVjLqBkk+D2BmtgGq5Bu+IyS9sb1H0hRgRPVLMjOzvFXyJa8vkd3P/1lAwFakH1c3M7MNSyVf8vqjpG2AN6VBT0ZEx4+6SHpvRNxW7QLNzKz6Kmn2ISIaI+LR9GjsNPp7VazLzMxyVFH490JVXJaZmeWomuG/Tj/sbmZmfa+a4W9mZhuIssNf0pBehs2tRkFmZpa/So78Z/Q0LCIOXv9yzMysL5Rzb58JwERgmKRdWXNidzQwPMfazMwsJ+Vc5/8+4FhgEvAj1oT/MuCUfMoyM7M8lXNvn+nAdEkfjYhr+qAmMzPLWSVt/pMkjVbmfEkPS9ovt8rMzCw3lYT/JyNiKbAfMA44Cjgrl6rMzCxXlYR/e1v/h4DfRMRs/K1eM7MNUiXhP0vSrcAHgFsljQLa8inLzMzyVEn4Hw/cC9wYESuBjYAv5lKVmZnlqpLw/zmwKfD+1L8MOLvqFZmZWe4q+TGXd0TEWyX9FSAiXpM0OKe6zMwsR5Uc+TdLqifdvVPSeNzmb2a2Qaok/H8G/B54g6TvAH8BzsylKjMzy1UlP+N4qaRZwL5kl3geFBFzcqvMzMxyU0mbPxHxJPBkTrWYmVkfyfXHXCRtIelOSU9Imi3pC2n4xpJuk/RU+rtRnnWYmdna8v4lrxbgyxGxA/BO4D8l7QCcDNweEdsAt6d+MzPrI7mGf0QsjIiHU/cyYA7ZbwMcCExPk00HDsqzDjMzW1uf/YavpMnArsADwKYRsTCNeonsy2NdzXOCpJmSZi5evLhP6jQzK4I+CX9JI4FrgC+mO4N2iIggfXegs4g4LyKmRcS08ePH90GlZmbFkHv4SxpEFvyXRsS1afAiSZul8ZsBL+ddh5mZrZH31T4CLgDmRETpfYCuB45J3ccA1+VZh5mZra2i6/zXwbvJfvTlMUmPpGGnkP0IzJWSjgeeBz6ecx1mZlYi1/CPiL/Q/Q++7Jvnc5uZWff67GofMzPrPxz+ZmYF5PA3Mysgh7+ZWQE5/M3MCsjhb2ZWQA5/M7MCcvibmRWQw9/MrIAc/mZmBeTwNzMrIIe/mVkBOfzNzArI4W9mVkAOfzOzAnL4m5kVkMPfzKyAHP5mZgXk8DezfuPFF1/kvPPOIyJqXcqA5/A3s37j3HPP5bLLLmPevHm1LmXAc/ibWb8xf/58ANra2mpcycDn8DezfkNSrUsoDIe/mfUb7W39bvPPn8PfzKyAHP5m1m+0N/u4+Sd/Dn8z6zcaGxsBWL16dY0rGfgc/mbWbzQ0NAAwaNCgGlcy8Dn8zazfqKurW+uv5cevsJn1G+3X97e0tNS4koHP4W9m/Ub7JZ5NTU01rmTgc/ibWb9RX18PwIgRI2pcycDn8DezfmPVqlUArFixosaVDHwOfzPrN5YuXQrAgw8+WONKBj6Hv5n1C6tXr+5o6/ddPfPn8C8YX0Vh/dVTTz3V0f3wrJk1rKQYHP4Fs2DBgo7ulStX1rASs7XNnj27o3vJ60t5/vnna1jNwOfwL5g//elPHd133313DSsxyzQ2NnLLLbdw0YUXsNNGTZy52xKGNMDpp32TJ5980vf2z4lqdetUSe8HfgrUA+dHxFk9TT9t2rSYOdMfBdfFypUr+fOf/8z119/A008/ReuYSdQ3LYNVr7PLLrtwwAEHsMceezB48OBal2oDWESwYsUK5s6dy5w5c3jqqad4+ql/8PwLL9Da2saYwVnI1xGMGBS8tKqBljYYMXwYW289la2nTmX77bdn2223ZeLEib4FRJkkzYqIaf8yvBbhL6ke+AfwXuBF4CHg8Ih4ort5HP6VmzdvHldccQW33fZnGhtXEw1DoK2FQMTgUUT9IBpaV8PqpYwcNYoPfuADHHrooYwbN67WpVvO2traaG1tpaWlhZaWFlpbWzv6S4c3Nzf3+GhsbOx4tJ+wXb16NStXrmTVqlWsXLGClSuXs3TpUpYsWUpTc3NHDRsNhS1GNLHVyBa2HdvCjXOH8ffX1wT6NqOb2WtiI88sbeCF5Q28uGIQq1vW5NXIEcMZM2YMI0eOZPiIkQwfPpzhw4czbNgwhgwZ0vEYOnQogwcP7ngMGjSIhoaGju7SR0NDA/X19dTX13d0NzQ0rDV8Q7vjaHfh31CLYoDdgKcj4lkASZcDBwLdhr9V7he/+AUzZsygeZNtaHrjdgyZex8jYyX7778/N954I8tDLNvpo9QvXUDLy09w5ZVX0trayuc+97lal14oy5cv5+STT2b27Nkd33DtLmD684+cNNTBkHoxqB6G1gfD6lsZWtfGqIZg04Y2Rm8WjB7UxuYjWpk8qoWNhqy9Llc/M5yhQ4d2bJ9NbS3ssVkje2yW3emzLWDBinqeX17P4lX1LGtexdKm11j1qlixSLzSVs/q1joaW0VzK6xuDfrry9XTG0jpNnDcccdx9NFH51JDrcJ/IlB6LdeLwDtqVMuAtffeezNjxgzahoyirmkFdc0r2f+g/TnxxBOJCK667iYaXpsLQNvgkQDss88+Nay4mObPn8/jjz++1rD+HPJdGdog6uugThABq1tFU1sDyxSoKRteB0hBvYJBCgbVtTGoLhhUB4PrgpdW1nHAwWu2zxt/fxXnPDaS5jbR3AZNbaIl6mhuq6MNaAsRZG8KbUHqzkK1vg6GSzS1Bc2tNXxhulHO/zciuPDCCwdc+JdF0gnACQBbbrlljavZ8Oyzzz5ccOGFLJr/cMewG2+8kYjgpptuoq5lNcOeubNj3C677sqOO+5Yi1ILbbvttuOOO+5Yq6mlc3fpsIggImhra+v4W9q9Ln9bW1t7XGY1Hy0tLTQ1NWWPxkaWNq6mqamRFi1aa/ts1SBeHrYVgwcPZsjQoYwaPKSj6aa9+aWuro76+nrq6uqQtN7d7cvs7W850/T2t6GhoaOpqf1v52Htt7vIQ63a/HcHTo+I96X+rwFExHe7m8dt/utm2bJlvPLKKwCcddZZ/P3vf+8Yt91223HyySd39E+YMIFhw4b1eY1mAOeccw6PPvooy5cvZ+TIkey8885ugqyC/tbm/xCwjaQpwHzgMOCIGtUyoI0aNYpRo0YBsNNOOzF06NCOcVOnTmXKlCm1Ks1sLQ76vlWT8I+IFkknAreSXep5YUTM7mU2W0/eucysXc3a/CPiZuDmWj2/mVmR+Ru+ZmYF5PA3Mysgh7+ZWQE5/M3MCsjhb2ZWQDW7q2elJC0GfIPv6tgEeKXWRZh1w9tndW0VEeM7D9xgwt+qR9LMrr7xZ9YfePvsG272MTMrIIe/mVkBOfyL6bxaF2DWA2+ffcBt/mZmBeQjfzOzAnL4m5kVUL/+JS8rn6RW4LGSQQdFxNxupl0eESP7pDAzQNI44PbUOwFoBRan/t0ioqkmhRWY2/wHiEoC3eFvtSTpdGB5RPywZFhDRLTUrqricbPPACVppKTbJT0s6TFJB3YxzWaS7pH0iKTHJe2Rhu8naUaa9ypJfqOwqpN0saRfSXoA+L6k0yX9V8n4xyVNTt2fkPRg2lb/V1J+P25bEA7/gWNY2jEekfR7YDXwkYh4K7A38CNJ6jTPEcCtEbELsDPwiKRNgFOB96R5ZwIn9d1qWMFMAt4VEd1uY5K2Bw4F3p221VbgyD6qb8Bym//AsSrtGABIGgScKenfgDZgIrAp8FLJPA8BF6Zp/xARj0jaE9gBuDe9VwwGZvTROljxXBURrb1Msy/wNuChtE0OA17Ou7CBzuE/cB0JjAfeFhHNkuYCQ0sniIh70pvDh4CLJZ0NvAbcFhGH93XBVkgrSrpbWLs1on17FTA9Ir7WZ1UVgJt9Bq4xwMsp+PcGtuo8gaStgEUR8WvgfOCtwP3AuyVNTdOMkLRtH9ZtxTWXbBtE0luBKWn47cAhkt6Qxm2ctl1bDz7yH7guBW6Q9BhZu/2TXUyzF/DfkpqB5cDREbFY0rHA7yQNSdOdCvwj/5Kt4K4BjpY0G3iAtM1FxBOSTgX+JKkOaAb+E9/ifb34Uk8zswJys4+ZWQE5/M3MCsjhb2ZWQA5/M7MCcvibmRWQw98GNEmTJT1eo+e+S5J/iNz6JYe/WRVI8ndmbIPi8LciqJf0a0mzJf1J0jBJ/y7pIUmPSrpG0nCAkpvjPSJplaQ9Je2W7nL6V0n3SdouTXuspOsl3QHcnpZ7uaQ56eZ6w9oLkHR4urvq45K+VzJ8eUn3IZIuTt0fS9M+KumePnqdrEAc/lYE2wA/j4gdgSXAR4FrI+LtEbEzMAc4HiAidkk3yPsG2Tej7yP7dvQeEbEr8E3gzJJlvxU4JCL2BP4DWBkR2wOnkd2MDEmbA98D9gF2Ad4u6aBeav4m8L5U34fX9wUw68wfVa0InouIR1L3LGAysJOkbwNjgZHAre0TS9oG+AGwd7o30gRgehoewKCSZd8WEf9M3f8G/AwgIv4m6W9p+NuBu6EFOh4AAAE5SURBVCJicVr+pWnaP/RQ871kN9u7Erh23VbbrHs+8rciaCzpbiU76LkYODEi3gycQbqDZPrhmiuBf4+IhWme/wHujIidgANY++6opXelXBel91fpWG5EfIbsnkpbALPSzyCaVY3D34pqFLAw/ZZB6Q+DXAhcFBH/VzJsDDA/dR/bwzLvIfuBHCTtBLwlDX8Q2FPSJukXqA4H7k7jFknaPt2w7CPtC5K0dUQ8EBHfJPut2y3WYR3NuuXwt6L6BtmdI+8l3fE03Sb4EOCTJSd9pwHfB74r6a/03FT6S2CkpDnAt8iamEifIE4G7gQeBWZFxHVpnpOBG8nOLSwsWdYP2k8Qp3GPVmGdzTr4rp5mZgXkI38zswJy+JuZFZDD38ysgBz+ZmYF5PA3Mysgh7+ZWQE5/M3MCuj/A1Au58g4C2pWAAAAAElFTkSuQmCC\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# plot relative velocity and absolute magnitude\n",
        "sns.scatterplot(x=neos['absolute_magnitude'], y=neos['relative_velocity']).set(title='Absolute Magnitude vs. Relative Velocity')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 313
        },
        "id": "WJb7G1y4X4xv",
        "outputId": "7c5731e7-569c-4c6b-ca48-433661fc5caa"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[Text(0.5, 1.0, 'Absolute Magnitude vs. Relative Velocity')]"
            ]
          },
          "metadata": {},
          "execution_count": 13
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Exploratory Analysis Findings\n",
        "1. It's possible that a higher relative velocity may be an indication of a more hazardous NEO. Although the NEOs not labeled as hazardous had a larger range than the hazardous NEOs, the hazardous NEOs were on average 35% faster in relative velocity than non-hazardous NEOs. \n",
        "\n",
        "2. Pertaining to absolute magnitude, the hazardous labeled NEO had a much tighter distribution with most absolute magnitudes lying very close to the mean. The non-hazardous NEOs had a higher absolute magnitude on average but also had a much larger range which could explain the higher average. \n",
        "\n",
        "3. Using the estimated minimum diameter, NEOs that are hazardous are on average 164% larger than non-hazardous NEOs. Using the estimated maximum diameter, NEOs that are hazardous are similarly on average 175% larger than non-hazardous NEOs. For both parameters, non-hazardous NEOs had much larger ranges and extremely large outliers. Even though the distribution for hazardous NEOs was much tighter, the average remained higher which is much more compelling.\n",
        "\n",
        "4. There certainly appears to be a moderate negative correlation between absolute magnitude and relative velocity. As absolute magnitude increases, relative velocity appears to decrease slightly."
      ],
      "metadata": {
        "id": "fhaq22R8ropW"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Train/Test Split\n",
        "\n",
        "The dataset contains just over 90,000 datapoints. I will randomly select 40,000 (44%) for training and 40,000 (44%)for testing. I'll then reserve the remaining 10,000+ (12%) for my golden holdout."
      ],
      "metadata": {
        "id": "nawrgM2LCG0y"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Perform Train/Test split\n",
        "class_col = 'hazardous'\n",
        "random_seed = 1234\n",
        "\n",
        "X_train, X_test, y_train, y_test = train_test_split(neos.drop(columns=class_col), neos[class_col],\n",
        "                                                   test_size=0.44, train_size= 0.44, random_state=random_seed, stratify=neos[class_col])"
      ],
      "metadata": {
        "id": "nueRrpy0eaVc"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Check dimensions of X\n",
        "# X Train\n",
        "print('On X train: ')\n",
        "print('X train dimensions: ', X_train.shape)\n",
        "display(X_train.head())\n",
        "\n",
        "# X test\n",
        "print('\\nOn X test: ')\n",
        "print('X test dimensions: ', X_test.shape)\n",
        "display(X_test.head())"
      ],
      "metadata": {
        "id": "5TfMYfhcg-ax",
        "outputId": "9068b797-f70b-4fb2-9079-1007717ff8f9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 482
        }
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "On X train: \n",
            "X train dimensions:  (39967, 9)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "            id                name  est_diameter_min  est_diameter_max  \\\n",
              "48136  2446833  446833 (2001 RB12)          0.167708          0.375008   \n",
              "58768  3740667           (2016 BU)          0.012149          0.027167   \n",
              "1576   2249886  249886 (2001 RY11)          0.817627          1.828269   \n",
              "16737  2380929  380929 (2006 HU30)          0.321037          0.717860   \n",
              "42924  3644815         (2013 NH10)          0.063761          0.142574   \n",
              "\n",
              "       relative_velocity  miss_distance orbiting_body  sentry_object  \\\n",
              "48136       35449.205475   6.530534e+07         Earth          False   \n",
              "58768       30285.492225   5.044515e+07         Earth          False   \n",
              "1576        41502.282921   4.218968e+07         Earth          False   \n",
              "16737       44153.448899   6.596702e+07         Earth          False   \n",
              "42924       44034.483020   5.991513e+07         Earth          False   \n",
              "\n",
              "       absolute_magnitude  \n",
              "48136               21.00  \n",
              "58768               26.70  \n",
              "1576                17.56  \n",
              "16737               19.59  \n",
              "42924               23.10  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-c5373ba1-4b99-47bd-9755-d53e2ea1fe56\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>name</th>\n",
              "      <th>est_diameter_min</th>\n",
              "      <th>est_diameter_max</th>\n",
              "      <th>relative_velocity</th>\n",
              "      <th>miss_distance</th>\n",
              "      <th>orbiting_body</th>\n",
              "      <th>sentry_object</th>\n",
              "      <th>absolute_magnitude</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>48136</th>\n",
              "      <td>2446833</td>\n",
              "      <td>446833 (2001 RB12)</td>\n",
              "      <td>0.167708</td>\n",
              "      <td>0.375008</td>\n",
              "      <td>35449.205475</td>\n",
              "      <td>6.530534e+07</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>21.00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>58768</th>\n",
              "      <td>3740667</td>\n",
              "      <td>(2016 BU)</td>\n",
              "      <td>0.012149</td>\n",
              "      <td>0.027167</td>\n",
              "      <td>30285.492225</td>\n",
              "      <td>5.044515e+07</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>26.70</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1576</th>\n",
              "      <td>2249886</td>\n",
              "      <td>249886 (2001 RY11)</td>\n",
              "      <td>0.817627</td>\n",
              "      <td>1.828269</td>\n",
              "      <td>41502.282921</td>\n",
              "      <td>4.218968e+07</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>17.56</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16737</th>\n",
              "      <td>2380929</td>\n",
              "      <td>380929 (2006 HU30)</td>\n",
              "      <td>0.321037</td>\n",
              "      <td>0.717860</td>\n",
              "      <td>44153.448899</td>\n",
              "      <td>6.596702e+07</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>19.59</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>42924</th>\n",
              "      <td>3644815</td>\n",
              "      <td>(2013 NH10)</td>\n",
              "      <td>0.063761</td>\n",
              "      <td>0.142574</td>\n",
              "      <td>44034.483020</td>\n",
              "      <td>5.991513e+07</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>23.10</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c5373ba1-4b99-47bd-9755-d53e2ea1fe56')\"\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-c5373ba1-4b99-47bd-9755-d53e2ea1fe56 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-c5373ba1-4b99-47bd-9755-d53e2ea1fe56');\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": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "On X test: \n",
            "X test dimensions:  (39968, 9)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "             id         name  est_diameter_min  est_diameter_max  \\\n",
              "50018   3799259  (2017 UR52)          0.152952          0.342011   \n",
              "83347   3536591   (2010 NG1)          0.232571          0.520044   \n",
              "44090   3759719   (2016 SY1)          0.011603          0.025944   \n",
              "74407  54016809   (2020 HW7)          0.010105          0.022596   \n",
              "80218  54050980   (2020 PB1)          0.167708          0.375008   \n",
              "\n",
              "       relative_velocity  miss_distance orbiting_body  sentry_object  \\\n",
              "50018      194056.414989   4.888464e+07         Earth          False   \n",
              "83347       63647.376068   4.830891e+07         Earth          False   \n",
              "44090       55686.275337   1.277171e+06         Earth          False   \n",
              "74407       27892.028691   7.104528e+07         Earth          False   \n",
              "80218       58749.031927   4.584174e+07         Earth          False   \n",
              "\n",
              "       absolute_magnitude  \n",
              "50018               21.20  \n",
              "83347               20.29  \n",
              "44090               26.80  \n",
              "74407               27.10  \n",
              "80218               21.00  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-2a567c06-ea9d-4d70-ba69-a729d459e2d8\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>name</th>\n",
              "      <th>est_diameter_min</th>\n",
              "      <th>est_diameter_max</th>\n",
              "      <th>relative_velocity</th>\n",
              "      <th>miss_distance</th>\n",
              "      <th>orbiting_body</th>\n",
              "      <th>sentry_object</th>\n",
              "      <th>absolute_magnitude</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>50018</th>\n",
              "      <td>3799259</td>\n",
              "      <td>(2017 UR52)</td>\n",
              "      <td>0.152952</td>\n",
              "      <td>0.342011</td>\n",
              "      <td>194056.414989</td>\n",
              "      <td>4.888464e+07</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>21.20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>83347</th>\n",
              "      <td>3536591</td>\n",
              "      <td>(2010 NG1)</td>\n",
              "      <td>0.232571</td>\n",
              "      <td>0.520044</td>\n",
              "      <td>63647.376068</td>\n",
              "      <td>4.830891e+07</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>20.29</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>44090</th>\n",
              "      <td>3759719</td>\n",
              "      <td>(2016 SY1)</td>\n",
              "      <td>0.011603</td>\n",
              "      <td>0.025944</td>\n",
              "      <td>55686.275337</td>\n",
              "      <td>1.277171e+06</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>26.80</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>74407</th>\n",
              "      <td>54016809</td>\n",
              "      <td>(2020 HW7)</td>\n",
              "      <td>0.010105</td>\n",
              "      <td>0.022596</td>\n",
              "      <td>27892.028691</td>\n",
              "      <td>7.104528e+07</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>27.10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>80218</th>\n",
              "      <td>54050980</td>\n",
              "      <td>(2020 PB1)</td>\n",
              "      <td>0.167708</td>\n",
              "      <td>0.375008</td>\n",
              "      <td>58749.031927</td>\n",
              "      <td>4.584174e+07</td>\n",
              "      <td>Earth</td>\n",
              "      <td>False</td>\n",
              "      <td>21.00</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2a567c06-ea9d-4d70-ba69-a729d459e2d8')\"\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-2a567c06-ea9d-4d70-ba69-a729d459e2d8 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-2a567c06-ea9d-4d70-ba69-a729d459e2d8');\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": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Check dimensions of Y\n",
        "# Y Train\n",
        "print('On y train: ')\n",
        "print('y train dimensions: ', y_train.shape)\n",
        "display(y_train.head())\n",
        "\n",
        "# Y test\n",
        "print('\\nOn y test: ')\n",
        "print('y test dimensions: ', y_test.shape)\n",
        "display(y_test.head())"
      ],
      "metadata": {
        "id": "0gvRpANmhkDd",
        "outputId": "fc26d0c4-26bd-49f1-c435-bc8c2e00964d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        }
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "On y train: \n",
            "y train dimensions:  (39967,)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "48136    False\n",
              "58768    False\n",
              "1576     False\n",
              "16737    False\n",
              "42924    False\n",
              "Name: hazardous, dtype: bool"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "On y test: \n",
            "y test dimensions:  (39968,)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "50018    False\n",
              "83347    False\n",
              "44090    False\n",
              "74407    False\n",
              "80218     True\n",
              "Name: hazardous, dtype: bool"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Feature Engineering\n",
        "\n",
        "- I plan to drop the `orbiting_body` and `sentry_object` columns as they have the same value for each row in both columns respectively. I'll also drop `id` as it is a numerical value that the model may think is of relevance, when indeed it is not. All the drops will be performed in the pipeline using \"drop\" in a `FunctionsTransformer` using `dropper`.\n",
        "- I'd like to take the average of `est_diameter_min` and `est_diameter_max` and put those values into an `est_diameter` column to see whether an average of these columns could provide better modeling. I'll do this by creating a function and then passing it through `FunctionTransformer`. \n",
        "- I'd like to try to create a function that extracts the year out of `name` since it's possible that year could have an affect on where NEOs and other planets may be in relation to Earth. I'll do this by creating a couple of functions and then passing it through `FunctionTransformer`. \n",
        "- Test values will be applied to the functions doing transformations to confirm it's working properly."
      ],
      "metadata": {
        "id": "w5dI0OEKAJC3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# This is the function transformer for estimated diameter average\n",
        "\n",
        "def combine_est_diameter(data_frame):\n",
        "    \"\"\"This function finds the average between `est_diameter_min` and `est_diameter_max` to \n",
        "    create a new column `est_diameter`. The former two columns are then dropped and the new\n",
        "    data_frame is returned.\"\"\"\n",
        "\n",
        "    data_frame['est_diameter'] = (data_frame['est_diameter_min'] + data_frame['est_diameter_max']) / 2\n",
        "    data_frame = data_frame.drop(columns=['est_diameter_min', 'est_diameter_max'])\n",
        "    return data_frame\n",
        "\n",
        "# test\n",
        "df = pd.DataFrame(columns=['est_diameter_min', 'est_diameter_max'], data= [[2,4], [1,3]])\n",
        "combine_est_diameter(df) # should return [[3.0], [2.0]]"
      ],
      "metadata": {
        "id": "0HGxyNtp5wJr",
        "outputId": "5aea3492-8632-4bde-b6f0-e531a890aad1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 112
        }
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   est_diameter\n",
              "0           3.0\n",
              "1           2.0"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-01192098-191c-473c-8cf6-985c60a1eb66\">\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>est_diameter</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>3.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-01192098-191c-473c-8cf6-985c60a1eb66')\"\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-01192098-191c-473c-8cf6-985c60a1eb66 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-01192098-191c-473c-8cf6-985c60a1eb66');\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": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Function takes in a name_value and finds the year.\n",
        "# Converts year into integer and returns in.\n",
        "def find_year(name_value):\n",
        "    try:\n",
        "        new_value = \"\"\n",
        "        start_idx = name_value.find(\"(\") + 1 \n",
        "        end_idx = start_idx + 4\n",
        "        new_value = name_value[start_idx : end_idx]\n",
        "        new_value = int(new_value)\n",
        "        return new_value\n",
        "    except:\n",
        "        return None\n",
        "\n",
        "# This is the transformer that converts name to year using the `find_year` function\n",
        "def year_transformer(data_frame):\n",
        "    \"\"\"This transformer applies the `find_year` function to the `name` column\n",
        "    and then creates a new column `year` with these values. `name` is then dropped\n",
        "    as well as null values.\"\"\"\n",
        "    data_frame['year'] = data_frame['name'].apply(find_year)\n",
        "    data_frame = data_frame.drop(columns=['name'])\n",
        "\n",
        "    return data_frame\n",
        "    \n",
        "# Example to test find_year()\n",
        "print(find_year(\"277475 (2015 WK4)\"))  # should return -> 2015\n",
        "print(type(find_year(\"277475 (2015 WK4)\")))     # should return -> <class 'int'>\n",
        "\n",
        "# Example to test year_transformer()\n",
        "df = pd.DataFrame(columns=['name'], data=[[\"277475 (2015 WK4)\"], [\"277475 (2000 WK4)\"]])\n",
        "year_transformer(df) # should return [[2015],[2000]]"
      ],
      "metadata": {
        "id": "1VvPXeU9-G6m",
        "outputId": "d44bfcff-87a0-47d9-902f-22db32c9b6d7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 147
        }
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2015\n",
            "<class 'int'>\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   year\n",
              "0  2015\n",
              "1  2000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-8f887bc2-d40a-4b94-9b95-0a617f3b9094\">\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>year</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2015</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8f887bc2-d40a-4b94-9b95-0a617f3b9094')\"\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-8f887bc2-d40a-4b94-9b95-0a617f3b9094 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-8f887bc2-d40a-4b94-9b95-0a617f3b9094');\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": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# This transformer drops `orbiting_body`, `sentry_object`, `id` for the standard pipeline\n",
        "def dropper(data_frame):\n",
        "    \"\"\"This transformer drops `orbiting_body`, `sentry_object`, `id`\"\"\"\n",
        "    data_frame = data_frame.drop(columns=[\"orbiting_body\", \"sentry_object\", \"id\"])\n",
        "\n",
        "    return data_frame"
      ],
      "metadata": {
        "id": "5WlqnDKfbv-b"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Feature Engineering Pipeline\n",
        "I will construct a pipeline for my feature engineering. This will belong in the beginning of the entire preprocessing pipeline."
      ],
      "metadata": {
        "id": "5iy4d3n_HXJt"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# All of the FunctionTransformers are setup here as well as the feature engineering pipeline\n",
        "\n",
        "est_diameter_transform = FunctionTransformer(combine_est_diameter)\n",
        "year_pipe_transform = FunctionTransformer(year_transformer)\n",
        "drop_columns_transform = FunctionTransformer(dropper)\n",
        "\n",
        "feat_eng = Pipeline([(\"diameter\", est_diameter_transform),\n",
        "                     (\"year\", year_pipe_transform),\n",
        "                     (\"dropper\", drop_columns_transform)])"
      ],
      "metadata": {
        "id": "UobZbz5tWpzR"
      },
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Pipeline\n",
        "\n",
        "Considering my data have no missing values, I won't need any imputation pipelines. Since my only non-numerical column will be a binary column value there will be no use for one hot encoding. However, a few features had some extreme outliers so I will consider using a scaler such as `RobustScaler` or `StandardScaler` to normalize the data. The problem I'm solving is a binary classification problem, so I will first implement with a Logistic Regression classification model."
      ],
      "metadata": {
        "id": "71Jo7hMGFx_t"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Build pipe for handling numerical and categorical data\n",
        "\n",
        "num_pipeline = Pipeline(steps=[('scale_num', StandardScaler()),\n",
        "                               ('num_impute', SimpleImputer(missing_values=np.nan, strategy='mean'))])\n",
        "\n",
        "cat_pipeline = Pipeline(steps=[('cat_impute', SimpleImputer(missing_values=np.nan, strategy='most_frequent'))])"
      ],
      "metadata": {
        "id": "hnW2sXi5mXFB"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Establish preprocessing pipe via column transformer\n",
        "\n",
        "preproc = ColumnTransformer([('num_pipe', num_pipeline, make_column_selector(dtype_include=np.number)),\n",
        "                             ('cat_pipe', cat_pipeline, make_column_selector(dtype_include=object))],\n",
        "                             remainder='passthrough')"
      ],
      "metadata": {
        "id": "DMt64hfmnPoT"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Logistic Regression Pipeline"
      ],
      "metadata": {
        "id": "NLe6P9OgS2Xq"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Generate pipeline for logistic regression model\n",
        "logistic_regression_pipe = Pipeline(steps=[\n",
        "                       ('FeatureEngineering', feat_eng),\n",
        "                       ('PreProcessing', preproc),\n",
        "                       ('mdl', LogisticRegression(penalty='elasticnet', solver='saga', tol=0.01))])\n",
        "\n",
        "# visualization of the pipeline\n",
        "with config_context(display='diagram'):\n",
        "    display(logistic_regression_pipe)"
      ],
      "metadata": {
        "id": "WF7WyoR5oDfT",
        "outputId": "9bcbaba4-5e08-492b-b618-d5041ae86c47",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        }
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Pipeline(steps=[('FeatureEngineering',\n",
              "                 Pipeline(steps=[('diameter',\n",
              "                                  FunctionTransformer(func=<function combine_est_diameter at 0x7f0546797b00>)),\n",
              "                                 ('year',\n",
              "                                  FunctionTransformer(func=<function year_transformer at 0x7f0546795560>)),\n",
              "                                 ('dropper',\n",
              "                                  FunctionTransformer(func=<function dropper at 0x7f0546795dd0>))])),\n",
              "                ('PreProcessing',\n",
              "                 ColumnTransformer(remainder='passthrough...\n",
              "                                                                  ('num_impute',\n",
              "                                                                   SimpleImputer())]),\n",
              "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50>),\n",
              "                                                 ('cat_pipe',\n",
              "                                                  Pipeline(steps=[('cat_impute',\n",
              "                                                                   SimpleImputer(strategy='most_frequent'))]),\n",
              "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950>)])),\n",
              "                ('mdl',\n",
              "                 LogisticRegression(penalty='elasticnet', solver='saga',\n",
              "                                    tol=0.01))])"
            ],
            "text/html": [
              "<style>#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f {color: black;background-color: white;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f pre{padding: 0;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-toggleable {background-color: white;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f 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-a86dbe95-ee6e-473b-8dbc-6debea4ff71f 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-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-estimator:hover {background-color: #d4ebff;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 2em;bottom: 0;left: 50%;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-item {z-index: 1;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-parallel::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 2em;bottom: 0;left: 50%;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-parallel-item {display: flex;flex-direction: column;position: relative;background-color: white;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-parallel-item:only-child::after {width: 0;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f 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;position: relative;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-label label {font-family: monospace;font-weight: bold;background-color: white;display: inline-block;line-height: 1.2em;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-label-container {position: relative;z-index: 2;text-align: center;}#sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f 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-a86dbe95-ee6e-473b-8dbc-6debea4ff71f div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-a86dbe95-ee6e-473b-8dbc-6debea4ff71f\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;FeatureEngineering&#x27;,\n",
              "                 Pipeline(steps=[(&#x27;diameter&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function combine_est_diameter at 0x7f0546797b00&gt;)),\n",
              "                                 (&#x27;year&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function year_transformer at 0x7f0546795560&gt;)),\n",
              "                                 (&#x27;dropper&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function dropper at 0x7f0546795dd0&gt;))])),\n",
              "                (&#x27;PreProcessing&#x27;,\n",
              "                 ColumnTransformer(remainder=&#x27;passthrough...\n",
              "                                                                  (&#x27;num_impute&#x27;,\n",
              "                                                                   SimpleImputer())]),\n",
              "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50&gt;),\n",
              "                                                 (&#x27;cat_pipe&#x27;,\n",
              "                                                  Pipeline(steps=[(&#x27;cat_impute&#x27;,\n",
              "                                                                   SimpleImputer(strategy=&#x27;most_frequent&#x27;))]),\n",
              "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950&gt;)])),\n",
              "                (&#x27;mdl&#x27;,\n",
              "                 LogisticRegression(penalty=&#x27;elasticnet&#x27;, solver=&#x27;saga&#x27;,\n",
              "                                    tol=0.01))])</pre><b>Please rerun this cell to show the HTML repr or trust the notebook.</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=\"0137abbf-6485-4de9-b46b-09717148f101\" type=\"checkbox\" ><label for=\"0137abbf-6485-4de9-b46b-09717148f101\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;FeatureEngineering&#x27;,\n",
              "                 Pipeline(steps=[(&#x27;diameter&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function combine_est_diameter at 0x7f0546797b00&gt;)),\n",
              "                                 (&#x27;year&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function year_transformer at 0x7f0546795560&gt;)),\n",
              "                                 (&#x27;dropper&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function dropper at 0x7f0546795dd0&gt;))])),\n",
              "                (&#x27;PreProcessing&#x27;,\n",
              "                 ColumnTransformer(remainder=&#x27;passthrough...\n",
              "                                                                  (&#x27;num_impute&#x27;,\n",
              "                                                                   SimpleImputer())]),\n",
              "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50&gt;),\n",
              "                                                 (&#x27;cat_pipe&#x27;,\n",
              "                                                  Pipeline(steps=[(&#x27;cat_impute&#x27;,\n",
              "                                                                   SimpleImputer(strategy=&#x27;most_frequent&#x27;))]),\n",
              "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950&gt;)])),\n",
              "                (&#x27;mdl&#x27;,\n",
              "                 LogisticRegression(penalty=&#x27;elasticnet&#x27;, solver=&#x27;saga&#x27;,\n",
              "                                    tol=0.01))])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"103b7751-e2c5-40c1-a5d4-fa73b4fbf35a\" type=\"checkbox\" ><label for=\"103b7751-e2c5-40c1-a5d4-fa73b4fbf35a\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FeatureEngineering: Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;diameter&#x27;,\n",
              "                 FunctionTransformer(func=&lt;function combine_est_diameter at 0x7f0546797b00&gt;)),\n",
              "                (&#x27;year&#x27;,\n",
              "                 FunctionTransformer(func=&lt;function year_transformer at 0x7f0546795560&gt;)),\n",
              "                (&#x27;dropper&#x27;,\n",
              "                 FunctionTransformer(func=&lt;function dropper at 0x7f0546795dd0&gt;))])</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=\"e78d5be0-e65a-4cb0-8d0c-0034668c0c79\" type=\"checkbox\" ><label for=\"e78d5be0-e65a-4cb0-8d0c-0034668c0c79\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FunctionTransformer</label><div class=\"sk-toggleable__content\"><pre>FunctionTransformer(func=&lt;function combine_est_diameter at 0x7f0546797b00&gt;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"71089c89-0cf4-444f-8c72-492cd7acaef5\" type=\"checkbox\" ><label for=\"71089c89-0cf4-444f-8c72-492cd7acaef5\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FunctionTransformer</label><div class=\"sk-toggleable__content\"><pre>FunctionTransformer(func=&lt;function year_transformer at 0x7f0546795560&gt;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"d3d75b83-fd6a-4be7-906d-cd6d4224423d\" type=\"checkbox\" ><label for=\"d3d75b83-fd6a-4be7-906d-cd6d4224423d\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FunctionTransformer</label><div class=\"sk-toggleable__content\"><pre>FunctionTransformer(func=&lt;function dropper at 0x7f0546795dd0&gt;)</pre></div></div></div></div></div><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=\"58f8945d-df06-4f92-bff3-8788eecaba56\" type=\"checkbox\" ><label for=\"58f8945d-df06-4f92-bff3-8788eecaba56\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">PreProcessing: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
              "                  transformers=[(&#x27;num_pipe&#x27;,\n",
              "                                 Pipeline(steps=[(&#x27;scale_num&#x27;,\n",
              "                                                  StandardScaler()),\n",
              "                                                 (&#x27;num_impute&#x27;,\n",
              "                                                  SimpleImputer())]),\n",
              "                                 &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50&gt;),\n",
              "                                (&#x27;cat_pipe&#x27;,\n",
              "                                 Pipeline(steps=[(&#x27;cat_impute&#x27;,\n",
              "                                                  SimpleImputer(strategy=&#x27;most_frequent&#x27;))]),\n",
              "                                 &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950&gt;)])</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=\"caea6046-8bf7-44c2-935d-9ee0f69e5c97\" type=\"checkbox\" ><label for=\"caea6046-8bf7-44c2-935d-9ee0f69e5c97\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">num_pipe</label><div class=\"sk-toggleable__content\"><pre>&lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50&gt;</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"8a92dde2-7c56-45ff-8d15-c111e373d832\" type=\"checkbox\" ><label for=\"8a92dde2-7c56-45ff-8d15-c111e373d832\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StandardScaler</label><div class=\"sk-toggleable__content\"><pre>StandardScaler()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"1ed57d01-b664-4aa1-b9c5-45fd0b7adccb\" type=\"checkbox\" ><label for=\"1ed57d01-b664-4aa1-b9c5-45fd0b7adccb\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer()</pre></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\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"123a894d-da18-4888-8459-c49b47d49674\" type=\"checkbox\" ><label for=\"123a894d-da18-4888-8459-c49b47d49674\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">cat_pipe</label><div class=\"sk-toggleable__content\"><pre>&lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950&gt;</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"083392ef-fd4e-45d4-9b4f-d72fe423016a\" type=\"checkbox\" ><label for=\"083392ef-fd4e-45d4-9b4f-d72fe423016a\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(strategy=&#x27;most_frequent&#x27;)</pre></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\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"80f97b75-29e8-4592-a73d-388d30ca0441\" type=\"checkbox\" ><label for=\"80f97b75-29e8-4592-a73d-388d30ca0441\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">remainder</label><div class=\"sk-toggleable__content\"><pre></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=\"4800bc93-81ae-4e1d-97f8-c8a3e5dd4591\" type=\"checkbox\" ><label for=\"4800bc93-81ae-4e1d-97f8-c8a3e5dd4591\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">passthrough</label><div class=\"sk-toggleable__content\"><pre>passthrough</pre></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"6d2571f8-429e-44de-81a0-d37c231196c2\" type=\"checkbox\" ><label for=\"6d2571f8-429e-44de-81a0-d37c231196c2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LogisticRegression</label><div class=\"sk-toggleable__content\"><pre>LogisticRegression(penalty=&#x27;elasticnet&#x27;, solver=&#x27;saga&#x27;, tol=0.01)</pre></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Support Vector Machines Pipeline"
      ],
      "metadata": {
        "id": "XqyUz9bGTAYF"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Generate pipeline for supper vector model\n",
        "support_vector_pipe = Pipeline(steps=[\n",
        "                       ('FeatureEngineering', feat_eng),\n",
        "                       ('preproc', preproc),\n",
        "                       ('mdl', SVC(gamma='auto'))])\n",
        "\n",
        "# visualization of the pipeline\n",
        "with config_context(display='diagram'):\n",
        "    display(support_vector_pipe)"
      ],
      "metadata": {
        "id": "JEaw1e-2TJoz",
        "outputId": "a27cd753-bc67-4dae-8951-ce2c807b76eb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        }
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Pipeline(steps=[('FeatureEngineering',\n",
              "                 Pipeline(steps=[('diameter',\n",
              "                                  FunctionTransformer(func=<function combine_est_diameter at 0x7f0546797b00>)),\n",
              "                                 ('year',\n",
              "                                  FunctionTransformer(func=<function year_transformer at 0x7f0546795560>)),\n",
              "                                 ('dropper',\n",
              "                                  FunctionTransformer(func=<function dropper at 0x7f0546795dd0>))])),\n",
              "                ('preproc',\n",
              "                 ColumnTransformer(remainder='passthrough',\n",
              "                                   tran...\n",
              "                                                  Pipeline(steps=[('scale_num',\n",
              "                                                                   StandardScaler()),\n",
              "                                                                  ('num_impute',\n",
              "                                                                   SimpleImputer())]),\n",
              "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50>),\n",
              "                                                 ('cat_pipe',\n",
              "                                                  Pipeline(steps=[('cat_impute',\n",
              "                                                                   SimpleImputer(strategy='most_frequent'))]),\n",
              "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950>)])),\n",
              "                ('mdl', SVC(gamma='auto'))])"
            ],
            "text/html": [
              "<style>#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 {color: black;background-color: white;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 pre{padding: 0;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-toggleable {background-color: white;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 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-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 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-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 2em;bottom: 0;left: 50%;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-item {z-index: 1;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-parallel::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 2em;bottom: 0;left: 50%;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-parallel-item {display: flex;flex-direction: column;position: relative;background-color: white;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-parallel-item:only-child::after {width: 0;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 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;position: relative;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-label label {font-family: monospace;font-weight: bold;background-color: white;display: inline-block;line-height: 1.2em;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-label-container {position: relative;z-index: 2;text-align: center;}#sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 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-305cfbd3-36a0-4454-a25e-1c2afb0a89a2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-305cfbd3-36a0-4454-a25e-1c2afb0a89a2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;FeatureEngineering&#x27;,\n",
              "                 Pipeline(steps=[(&#x27;diameter&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function combine_est_diameter at 0x7f0546797b00&gt;)),\n",
              "                                 (&#x27;year&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function year_transformer at 0x7f0546795560&gt;)),\n",
              "                                 (&#x27;dropper&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function dropper at 0x7f0546795dd0&gt;))])),\n",
              "                (&#x27;preproc&#x27;,\n",
              "                 ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
              "                                   tran...\n",
              "                                                  Pipeline(steps=[(&#x27;scale_num&#x27;,\n",
              "                                                                   StandardScaler()),\n",
              "                                                                  (&#x27;num_impute&#x27;,\n",
              "                                                                   SimpleImputer())]),\n",
              "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50&gt;),\n",
              "                                                 (&#x27;cat_pipe&#x27;,\n",
              "                                                  Pipeline(steps=[(&#x27;cat_impute&#x27;,\n",
              "                                                                   SimpleImputer(strategy=&#x27;most_frequent&#x27;))]),\n",
              "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950&gt;)])),\n",
              "                (&#x27;mdl&#x27;, SVC(gamma=&#x27;auto&#x27;))])</pre><b>Please rerun this cell to show the HTML repr or trust the notebook.</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=\"bf59df35-aef8-4a04-9d30-8d46b2440721\" type=\"checkbox\" ><label for=\"bf59df35-aef8-4a04-9d30-8d46b2440721\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;FeatureEngineering&#x27;,\n",
              "                 Pipeline(steps=[(&#x27;diameter&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function combine_est_diameter at 0x7f0546797b00&gt;)),\n",
              "                                 (&#x27;year&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function year_transformer at 0x7f0546795560&gt;)),\n",
              "                                 (&#x27;dropper&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function dropper at 0x7f0546795dd0&gt;))])),\n",
              "                (&#x27;preproc&#x27;,\n",
              "                 ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
              "                                   tran...\n",
              "                                                  Pipeline(steps=[(&#x27;scale_num&#x27;,\n",
              "                                                                   StandardScaler()),\n",
              "                                                                  (&#x27;num_impute&#x27;,\n",
              "                                                                   SimpleImputer())]),\n",
              "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50&gt;),\n",
              "                                                 (&#x27;cat_pipe&#x27;,\n",
              "                                                  Pipeline(steps=[(&#x27;cat_impute&#x27;,\n",
              "                                                                   SimpleImputer(strategy=&#x27;most_frequent&#x27;))]),\n",
              "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950&gt;)])),\n",
              "                (&#x27;mdl&#x27;, SVC(gamma=&#x27;auto&#x27;))])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"468f67a5-7008-4702-98b4-a0ba55c3d586\" type=\"checkbox\" ><label for=\"468f67a5-7008-4702-98b4-a0ba55c3d586\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FeatureEngineering: Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;diameter&#x27;,\n",
              "                 FunctionTransformer(func=&lt;function combine_est_diameter at 0x7f0546797b00&gt;)),\n",
              "                (&#x27;year&#x27;,\n",
              "                 FunctionTransformer(func=&lt;function year_transformer at 0x7f0546795560&gt;)),\n",
              "                (&#x27;dropper&#x27;,\n",
              "                 FunctionTransformer(func=&lt;function dropper at 0x7f0546795dd0&gt;))])</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=\"946f0268-8089-468e-96cd-9a6495e79a1c\" type=\"checkbox\" ><label for=\"946f0268-8089-468e-96cd-9a6495e79a1c\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FunctionTransformer</label><div class=\"sk-toggleable__content\"><pre>FunctionTransformer(func=&lt;function combine_est_diameter at 0x7f0546797b00&gt;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"e550460a-fb26-4395-9651-62212195ef13\" type=\"checkbox\" ><label for=\"e550460a-fb26-4395-9651-62212195ef13\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FunctionTransformer</label><div class=\"sk-toggleable__content\"><pre>FunctionTransformer(func=&lt;function year_transformer at 0x7f0546795560&gt;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"36266a51-f65a-43c3-ad56-f31085361eb3\" type=\"checkbox\" ><label for=\"36266a51-f65a-43c3-ad56-f31085361eb3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FunctionTransformer</label><div class=\"sk-toggleable__content\"><pre>FunctionTransformer(func=&lt;function dropper at 0x7f0546795dd0&gt;)</pre></div></div></div></div></div><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=\"c882de38-e7d4-4bc7-a4cc-74d736364a1b\" type=\"checkbox\" ><label for=\"c882de38-e7d4-4bc7-a4cc-74d736364a1b\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">preproc: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
              "                  transformers=[(&#x27;num_pipe&#x27;,\n",
              "                                 Pipeline(steps=[(&#x27;scale_num&#x27;,\n",
              "                                                  StandardScaler()),\n",
              "                                                 (&#x27;num_impute&#x27;,\n",
              "                                                  SimpleImputer())]),\n",
              "                                 &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50&gt;),\n",
              "                                (&#x27;cat_pipe&#x27;,\n",
              "                                 Pipeline(steps=[(&#x27;cat_impute&#x27;,\n",
              "                                                  SimpleImputer(strategy=&#x27;most_frequent&#x27;))]),\n",
              "                                 &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950&gt;)])</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=\"edeec1b2-de0c-4137-b5c3-52a988e76211\" type=\"checkbox\" ><label for=\"edeec1b2-de0c-4137-b5c3-52a988e76211\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">num_pipe</label><div class=\"sk-toggleable__content\"><pre>&lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50&gt;</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"8a608c7b-bdcb-4c6e-a75b-06094e3b5968\" type=\"checkbox\" ><label for=\"8a608c7b-bdcb-4c6e-a75b-06094e3b5968\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StandardScaler</label><div class=\"sk-toggleable__content\"><pre>StandardScaler()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"251b35f4-2d3e-4fe6-86c5-cec7ff539068\" type=\"checkbox\" ><label for=\"251b35f4-2d3e-4fe6-86c5-cec7ff539068\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer()</pre></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\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"a152e287-2af1-4760-b057-f2080b9c58b3\" type=\"checkbox\" ><label for=\"a152e287-2af1-4760-b057-f2080b9c58b3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">cat_pipe</label><div class=\"sk-toggleable__content\"><pre>&lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950&gt;</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"24afdef9-88f7-4be3-aeab-6bdffe4ce82f\" type=\"checkbox\" ><label for=\"24afdef9-88f7-4be3-aeab-6bdffe4ce82f\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(strategy=&#x27;most_frequent&#x27;)</pre></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\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"b6bab3d9-b82f-41dd-b627-f1f5f81d2af0\" type=\"checkbox\" ><label for=\"b6bab3d9-b82f-41dd-b627-f1f5f81d2af0\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">remainder</label><div class=\"sk-toggleable__content\"><pre></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=\"15f6a7a9-d8b9-4ccb-8c34-2e8a00d685c9\" type=\"checkbox\" ><label for=\"15f6a7a9-d8b9-4ccb-8c34-2e8a00d685c9\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">passthrough</label><div class=\"sk-toggleable__content\"><pre>passthrough</pre></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"7919c3b6-fe14-4ec4-8149-699037e1f4e8\" type=\"checkbox\" ><label for=\"7919c3b6-fe14-4ec4-8149-699037e1f4e8\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SVC</label><div class=\"sk-toggleable__content\"><pre>SVC(gamma=&#x27;auto&#x27;)</pre></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### K-Nearest Neighbors Pipeline"
      ],
      "metadata": {
        "id": "Ko5yyNI_xRDt"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Generate pipeline for supper vector model\n",
        "knn_pipe = Pipeline(steps=[\n",
        "                       ('FeatureEngineering', feat_eng),\n",
        "                       ('preproc', preproc),\n",
        "                       ('mdl', KNeighborsClassifier())])\n",
        "\n",
        "# visualization of the pipeline\n",
        "with config_context(display='diagram'):\n",
        "    display(knn_pipe)"
      ],
      "metadata": {
        "id": "Mx3bWnJLxUcu",
        "outputId": "611deac4-dac1-4fb8-e2c8-f9a36cee2da9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        }
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Pipeline(steps=[('FeatureEngineering',\n",
              "                 Pipeline(steps=[('diameter',\n",
              "                                  FunctionTransformer(func=<function combine_est_diameter at 0x7f0546797b00>)),\n",
              "                                 ('year',\n",
              "                                  FunctionTransformer(func=<function year_transformer at 0x7f0546795560>)),\n",
              "                                 ('dropper',\n",
              "                                  FunctionTransformer(func=<function dropper at 0x7f0546795dd0>))])),\n",
              "                ('preproc',\n",
              "                 ColumnTransformer(remainder='passthrough',\n",
              "                                   tran...\n",
              "                                                  Pipeline(steps=[('scale_num',\n",
              "                                                                   StandardScaler()),\n",
              "                                                                  ('num_impute',\n",
              "                                                                   SimpleImputer())]),\n",
              "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50>),\n",
              "                                                 ('cat_pipe',\n",
              "                                                  Pipeline(steps=[('cat_impute',\n",
              "                                                                   SimpleImputer(strategy='most_frequent'))]),\n",
              "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950>)])),\n",
              "                ('mdl', KNeighborsClassifier())])"
            ],
            "text/html": [
              "<style>#sk-57639786-4e6b-48e0-911d-52f88acb66d4 {color: black;background-color: white;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 pre{padding: 0;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-toggleable {background-color: white;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 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-57639786-4e6b-48e0-911d-52f88acb66d4 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-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-estimator:hover {background-color: #d4ebff;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 2em;bottom: 0;left: 50%;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-item {z-index: 1;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-parallel::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 2em;bottom: 0;left: 50%;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-parallel-item {display: flex;flex-direction: column;position: relative;background-color: white;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-parallel-item:only-child::after {width: 0;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 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;position: relative;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-label label {font-family: monospace;font-weight: bold;background-color: white;display: inline-block;line-height: 1.2em;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-label-container {position: relative;z-index: 2;text-align: center;}#sk-57639786-4e6b-48e0-911d-52f88acb66d4 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-57639786-4e6b-48e0-911d-52f88acb66d4 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-57639786-4e6b-48e0-911d-52f88acb66d4\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;FeatureEngineering&#x27;,\n",
              "                 Pipeline(steps=[(&#x27;diameter&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function combine_est_diameter at 0x7f0546797b00&gt;)),\n",
              "                                 (&#x27;year&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function year_transformer at 0x7f0546795560&gt;)),\n",
              "                                 (&#x27;dropper&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function dropper at 0x7f0546795dd0&gt;))])),\n",
              "                (&#x27;preproc&#x27;,\n",
              "                 ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
              "                                   tran...\n",
              "                                                  Pipeline(steps=[(&#x27;scale_num&#x27;,\n",
              "                                                                   StandardScaler()),\n",
              "                                                                  (&#x27;num_impute&#x27;,\n",
              "                                                                   SimpleImputer())]),\n",
              "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50&gt;),\n",
              "                                                 (&#x27;cat_pipe&#x27;,\n",
              "                                                  Pipeline(steps=[(&#x27;cat_impute&#x27;,\n",
              "                                                                   SimpleImputer(strategy=&#x27;most_frequent&#x27;))]),\n",
              "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950&gt;)])),\n",
              "                (&#x27;mdl&#x27;, KNeighborsClassifier())])</pre><b>Please rerun this cell to show the HTML repr or trust the notebook.</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=\"717423a9-6776-4dce-89e4-8ed6536b5458\" type=\"checkbox\" ><label for=\"717423a9-6776-4dce-89e4-8ed6536b5458\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;FeatureEngineering&#x27;,\n",
              "                 Pipeline(steps=[(&#x27;diameter&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function combine_est_diameter at 0x7f0546797b00&gt;)),\n",
              "                                 (&#x27;year&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function year_transformer at 0x7f0546795560&gt;)),\n",
              "                                 (&#x27;dropper&#x27;,\n",
              "                                  FunctionTransformer(func=&lt;function dropper at 0x7f0546795dd0&gt;))])),\n",
              "                (&#x27;preproc&#x27;,\n",
              "                 ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
              "                                   tran...\n",
              "                                                  Pipeline(steps=[(&#x27;scale_num&#x27;,\n",
              "                                                                   StandardScaler()),\n",
              "                                                                  (&#x27;num_impute&#x27;,\n",
              "                                                                   SimpleImputer())]),\n",
              "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50&gt;),\n",
              "                                                 (&#x27;cat_pipe&#x27;,\n",
              "                                                  Pipeline(steps=[(&#x27;cat_impute&#x27;,\n",
              "                                                                   SimpleImputer(strategy=&#x27;most_frequent&#x27;))]),\n",
              "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950&gt;)])),\n",
              "                (&#x27;mdl&#x27;, KNeighborsClassifier())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"4b27cc21-eef4-4d14-8580-e0134109b4f6\" type=\"checkbox\" ><label for=\"4b27cc21-eef4-4d14-8580-e0134109b4f6\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FeatureEngineering: Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;diameter&#x27;,\n",
              "                 FunctionTransformer(func=&lt;function combine_est_diameter at 0x7f0546797b00&gt;)),\n",
              "                (&#x27;year&#x27;,\n",
              "                 FunctionTransformer(func=&lt;function year_transformer at 0x7f0546795560&gt;)),\n",
              "                (&#x27;dropper&#x27;,\n",
              "                 FunctionTransformer(func=&lt;function dropper at 0x7f0546795dd0&gt;))])</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=\"1afac193-5af2-4820-8eb0-82e32b5690fd\" type=\"checkbox\" ><label for=\"1afac193-5af2-4820-8eb0-82e32b5690fd\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FunctionTransformer</label><div class=\"sk-toggleable__content\"><pre>FunctionTransformer(func=&lt;function combine_est_diameter at 0x7f0546797b00&gt;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"3b40c6c8-3333-4fc3-8892-6c4ea3d773a8\" type=\"checkbox\" ><label for=\"3b40c6c8-3333-4fc3-8892-6c4ea3d773a8\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FunctionTransformer</label><div class=\"sk-toggleable__content\"><pre>FunctionTransformer(func=&lt;function year_transformer at 0x7f0546795560&gt;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"c346dc7e-e319-4f0c-85b9-d972b32c73b0\" type=\"checkbox\" ><label for=\"c346dc7e-e319-4f0c-85b9-d972b32c73b0\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FunctionTransformer</label><div class=\"sk-toggleable__content\"><pre>FunctionTransformer(func=&lt;function dropper at 0x7f0546795dd0&gt;)</pre></div></div></div></div></div><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=\"7571e78a-d911-479b-85fa-bd2814e7a2bd\" type=\"checkbox\" ><label for=\"7571e78a-d911-479b-85fa-bd2814e7a2bd\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">preproc: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
              "                  transformers=[(&#x27;num_pipe&#x27;,\n",
              "                                 Pipeline(steps=[(&#x27;scale_num&#x27;,\n",
              "                                                  StandardScaler()),\n",
              "                                                 (&#x27;num_impute&#x27;,\n",
              "                                                  SimpleImputer())]),\n",
              "                                 &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50&gt;),\n",
              "                                (&#x27;cat_pipe&#x27;,\n",
              "                                 Pipeline(steps=[(&#x27;cat_impute&#x27;,\n",
              "                                                  SimpleImputer(strategy=&#x27;most_frequent&#x27;))]),\n",
              "                                 &lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950&gt;)])</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=\"4634c581-75ea-470a-b20d-6076398ad7ca\" type=\"checkbox\" ><label for=\"4634c581-75ea-470a-b20d-6076398ad7ca\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">num_pipe</label><div class=\"sk-toggleable__content\"><pre>&lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50&gt;</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"f1a68f09-ec31-4b7d-90e9-cc86ac302a60\" type=\"checkbox\" ><label for=\"f1a68f09-ec31-4b7d-90e9-cc86ac302a60\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StandardScaler</label><div class=\"sk-toggleable__content\"><pre>StandardScaler()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"e02a9106-c979-4a31-96e8-63f676c735f8\" type=\"checkbox\" ><label for=\"e02a9106-c979-4a31-96e8-63f676c735f8\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer()</pre></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\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"5df88491-6df4-4de4-965d-916b7232116b\" type=\"checkbox\" ><label for=\"5df88491-6df4-4de4-965d-916b7232116b\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">cat_pipe</label><div class=\"sk-toggleable__content\"><pre>&lt;sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950&gt;</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"1003d907-5610-4593-a7b2-a784814c1dda\" type=\"checkbox\" ><label for=\"1003d907-5610-4593-a7b2-a784814c1dda\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(strategy=&#x27;most_frequent&#x27;)</pre></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\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"17ac2359-4b7e-4bfa-ae62-3cc4c5e7199d\" type=\"checkbox\" ><label for=\"17ac2359-4b7e-4bfa-ae62-3cc4c5e7199d\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">remainder</label><div class=\"sk-toggleable__content\"><pre></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=\"4069b13b-91b7-42d4-ad4d-540bbf6e3ea3\" type=\"checkbox\" ><label for=\"4069b13b-91b7-42d4-ad4d-540bbf6e3ea3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">passthrough</label><div class=\"sk-toggleable__content\"><pre>passthrough</pre></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"d8f4e43e-2b9c-4233-8c99-0670d6b49734\" type=\"checkbox\" ><label for=\"d8f4e43e-2b9c-4233-8c99-0670d6b49734\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">KNeighborsClassifier</label><div class=\"sk-toggleable__content\"><pre>KNeighborsClassifier()</pre></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Model Fitting and Evaluation\n",
        "\n",
        "Since I'm looking to see if an NEO is hazardous, this will be a binary classification problem. I have one primary assumption for feature importance:\n",
        "- `miss_distance` will most likely be the most important feature considering that we can assume that the closer an NEO is to Earth, the more potential dangerous it is.\n",
        "\n",
        "I anticipate the model to perform well (90%+) since there are some general trends with all of the numerical features. "
      ],
      "metadata": {
        "id": "ox8ZmXW0K6bM"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Logistic Regression Fit & Evaluation"
      ],
      "metadata": {
        "id": "5K4w7KExSgaG"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Setup Tuning Grid & Grid Search for Logistic Regression\n",
        "\n",
        "logistic_regression_tuning_grid = {'mdl__l1_ratio': np.linspace(0,1,5),\n",
        "                                   'mdl__C': np.logspace(-1,6,3)}\n",
        "\n",
        "logistic_regression_grid_search = GridSearchCV(logistic_regression_pipe, param_grid=logistic_regression_tuning_grid, return_train_score=True, n_jobs=-1)"
      ],
      "metadata": {
        "id": "N3qlF56GkfCK"
      },
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Find best logistic regression model parameters by fitting grid search\n",
        "logistic_regression_grid_search.fit(X_train, y_train)"
      ],
      "metadata": {
        "id": "-J7kYP4Tlqjl",
        "outputId": "e6352cf3-950d-4182-9d8a-7fb2f5be94a9",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSearchCV(estimator=Pipeline(steps=[('FeatureEngineering',\n",
              "                                        Pipeline(steps=[('diameter',\n",
              "                                                         FunctionTransformer(func=<function combine_est_diameter at 0x7f0546797b00>)),\n",
              "                                                        ('year',\n",
              "                                                         FunctionTransformer(func=<function year_transformer at 0x7f0546795560>)),\n",
              "                                                        ('dropper',\n",
              "                                                         FunctionTransformer(func=<function dropper at 0x7f0546795dd0>))])),\n",
              "                                       ('PreProcessing',\n",
              "                                        ColumnTransformer...\n",
              "                                                                                          SimpleImputer(strategy='most_frequent'))]),\n",
              "                                                                         <sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950>)])),\n",
              "                                       ('mdl',\n",
              "                                        LogisticRegression(penalty='elasticnet',\n",
              "                                                           solver='saga',\n",
              "                                                           tol=0.01))]),\n",
              "             n_jobs=-1,\n",
              "             param_grid={'mdl__C': array([1.00000000e-01, 3.16227766e+02, 1.00000000e+06]),\n",
              "                         'mdl__l1_ratio': array([0.  , 0.25, 0.5 , 0.75, 1.  ])},\n",
              "             return_train_score=True)"
            ]
          },
          "metadata": {},
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# logistic regression best model parameteres and values\n",
        "print(logistic_regression_grid_search.best_score_)\n",
        "print(logistic_regression_grid_search.best_params_)"
      ],
      "metadata": {
        "id": "Y2Xn7R2Dpyvl",
        "outputId": "a8c51bf0-8a98-4d3a-fe47-5a35a8fe4d9a",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.901743973437353\n",
            "{'mdl__C': 1000000.0, 'mdl__l1_ratio': 0.75}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# logistic regression best estimator pipeline\n",
        "logistic_regression_grid_search.best_estimator_"
      ],
      "metadata": {
        "id": "gqbIOxCEqTyN",
        "outputId": "75bb1cbc-4903-458f-d864-77a2daecf05d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Pipeline(steps=[('FeatureEngineering',\n",
              "                 Pipeline(steps=[('diameter',\n",
              "                                  FunctionTransformer(func=<function combine_est_diameter at 0x7f0546797b00>)),\n",
              "                                 ('year',\n",
              "                                  FunctionTransformer(func=<function year_transformer at 0x7f0546795560>)),\n",
              "                                 ('dropper',\n",
              "                                  FunctionTransformer(func=<function dropper at 0x7f0546795dd0>))])),\n",
              "                ('PreProcessing',\n",
              "                 ColumnTransformer(remainder='passthrough...\n",
              "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x7f05446a2390>),\n",
              "                                                 ('cat_pipe',\n",
              "                                                  Pipeline(steps=[('cat_impute',\n",
              "                                                                   SimpleImputer(strategy='most_frequent'))]),\n",
              "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x7f05446a2cd0>)])),\n",
              "                ('mdl',\n",
              "                 LogisticRegression(C=1000000.0, l1_ratio=0.75,\n",
              "                                    penalty='elasticnet', solver='saga',\n",
              "                                    tol=0.01))])"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# logistic regression feature weights\n",
        "lr_feature_weight = abs(logistic_regression_grid_search.best_estimator_['mdl'].coef_[0])\n",
        "lr_feature_names = ['est_diameter_min', 'est_diameter_max', 'relative_velocity', 'miss_distance', 'absolute_magnitude']\n",
        "\n",
        "lr_feature_importance = pd.DataFrame({'feature labels': lr_feature_names, 'feature weights': lr_feature_weight})\n",
        "lr_feature_importance.plot.bar(x='feature labels', y='feature weights')\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "jnHWRdxdqhFl",
        "outputId": "535116b5-5e5c-4b6d-8b69-b958e0c44fb0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 371
        }
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Support Vector Machines Fit & Evaluation"
      ],
      "metadata": {
        "id": "MwR_-innVuQ7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Setup Tuning Grid & Grid Search for Support Vector Machines Model\n",
        "\n",
        "support_vector_tuning_grid = {'mdl__C': [0.1,1],\n",
        "                              'mdl__kernel': ['linear', 'poly', 'rbf']}\n",
        "\n",
        "support_vector_grid_search = GridSearchCV(support_vector_pipe, param_grid=support_vector_tuning_grid, return_train_score=True, n_jobs=-1)"
      ],
      "metadata": {
        "id": "q6sUfzzhVzCs"
      },
      "execution_count": 32,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Find best support vector model parameters by fitting grid search\n",
        "support_vector_grid_search.fit(X_train, y_train)"
      ],
      "metadata": {
        "id": "c92Mx-bUX-1w",
        "outputId": "f145bd71-d957-42dc-d0e4-cefac4593a54",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSearchCV(estimator=Pipeline(steps=[('FeatureEngineering',\n",
              "                                        Pipeline(steps=[('diameter',\n",
              "                                                         FunctionTransformer(func=<function combine_est_diameter at 0x7f0546797b00>)),\n",
              "                                                        ('year',\n",
              "                                                         FunctionTransformer(func=<function year_transformer at 0x7f0546795560>)),\n",
              "                                                        ('dropper',\n",
              "                                                         FunctionTransformer(func=<function dropper at 0x7f0546795dd0>))])),\n",
              "                                       ('preproc',\n",
              "                                        ColumnTransformer(remai...\n",
              "                                                                         <sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3f50>),\n",
              "                                                                        ('cat_pipe',\n",
              "                                                                         Pipeline(steps=[('cat_impute',\n",
              "                                                                                          SimpleImputer(strategy='most_frequent'))]),\n",
              "                                                                         <sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950>)])),\n",
              "                                       ('mdl', SVC(gamma='auto'))]),\n",
              "             param_grid={'mdl__C': [0.1, 1],\n",
              "                         'mdl__kernel': ['linear', 'poly', 'rbf']},\n",
              "             return_train_score=True)"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Support Vector machines best model parameteres and values\n",
        "print(support_vector_grid_search.best_score_)\n",
        "support_vector_grid_search.best_params_"
      ],
      "metadata": {
        "id": "hJ_OVn5br1vv",
        "outputId": "58e063d3-54a4-49f5-8660-d5f59e489f3b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.9106762794477941\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'mdl__C': 1, 'mdl__kernel': 'rbf'}"
            ]
          },
          "metadata": {},
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Support Vector machines best estimator pipeline\n",
        "support_vector_grid_search.best_estimator_"
      ],
      "metadata": {
        "id": "Sh9FZf99sCYx",
        "outputId": "079255dc-a565-4ddb-dcc8-81d36aec0831",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Pipeline(steps=[('FeatureEngineering',\n",
              "                 Pipeline(steps=[('diameter',\n",
              "                                  FunctionTransformer(func=<function combine_est_diameter at 0x7f0546797b00>)),\n",
              "                                 ('year',\n",
              "                                  FunctionTransformer(func=<function year_transformer at 0x7f0546795560>)),\n",
              "                                 ('dropper',\n",
              "                                  FunctionTransformer(func=<function dropper at 0x7f0546795dd0>))])),\n",
              "                ('preproc',\n",
              "                 ColumnTransformer(remainder='passthrough',\n",
              "                                   tran...\n",
              "                                                  Pipeline(steps=[('scale_num',\n",
              "                                                                   StandardScaler()),\n",
              "                                                                  ('num_impute',\n",
              "                                                                   SimpleImputer())]),\n",
              "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x7f0546808990>),\n",
              "                                                 ('cat_pipe',\n",
              "                                                  Pipeline(steps=[('cat_impute',\n",
              "                                                                   SimpleImputer(strategy='most_frequent'))]),\n",
              "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x7f055c733a50>)])),\n",
              "                ('mdl', SVC(C=1, gamma='auto'))])"
            ]
          },
          "metadata": {},
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### K-Nearest Neighbors Fit & Evaluation"
      ],
      "metadata": {
        "id": "-RYrxFsjzSyd"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Setup Tuning Grid & Grid Search for K-Nearest Neighbors Model\n",
        "knn_tuning_grid = {'mdl__n_neighbors': [1,5, 10],\n",
        "                    'mdl__leaf_size': [10, 20, 30],\n",
        "                    'mdl__p': [1,2],\n",
        "                    'mdl__weights': ['uniform', 'distance']}\n",
        "\n",
        "knn_grid_search = GridSearchCV(knn_pipe, param_grid=knn_tuning_grid, return_train_score=True, n_jobs=-1)"
      ],
      "metadata": {
        "id": "rOskZhzozZPo"
      },
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Find best k-nearest neighbors model parameters by fitting grid search\n",
        "knn_grid_search.fit(X_train, y_train)"
      ],
      "metadata": {
        "id": "Tv57j1zXz3h6",
        "outputId": "a46aa3a4-b832-4adc-f0d8-d63e9d646a6d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSearchCV(estimator=Pipeline(steps=[('FeatureEngineering',\n",
              "                                        Pipeline(steps=[('diameter',\n",
              "                                                         FunctionTransformer(func=<function combine_est_diameter at 0x7f0546797b00>)),\n",
              "                                                        ('year',\n",
              "                                                         FunctionTransformer(func=<function year_transformer at 0x7f0546795560>)),\n",
              "                                                        ('dropper',\n",
              "                                                         FunctionTransformer(func=<function dropper at 0x7f0546795dd0>))])),\n",
              "                                       ('preproc',\n",
              "                                        ColumnTransformer(remai...\n",
              "                                                                        ('cat_pipe',\n",
              "                                                                         Pipeline(steps=[('cat_impute',\n",
              "                                                                                          SimpleImputer(strategy='most_frequent'))]),\n",
              "                                                                         <sklearn.compose._column_transformer.make_column_selector object at 0x7f05467f3950>)])),\n",
              "                                       ('mdl', KNeighborsClassifier())]),\n",
              "             n_jobs=-1,\n",
              "             param_grid={'mdl__leaf_size': [10, 20, 30],\n",
              "                         'mdl__n_neighbors': [1, 5, 10], 'mdl__p': [1, 2],\n",
              "                         'mdl__weights': ['uniform', 'distance']},\n",
              "             return_train_score=True)"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# K-Nearest Neighbors best model parameteres and values\n",
        "print(knn_grid_search.best_score_)\n",
        "knn_grid_search.best_params_"
      ],
      "metadata": {
        "id": "YQmX79BjwSvx",
        "outputId": "e428b446-e7bb-486e-88dc-b832cd6a13e1",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.9136036657794859\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'mdl__leaf_size': 10,\n",
              " 'mdl__n_neighbors': 10,\n",
              " 'mdl__p': 1,\n",
              " 'mdl__weights': 'distance'}"
            ]
          },
          "metadata": {},
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# K-Nearest Neighbors best estimator pipeline\n",
        "knn_grid_search.best_estimator_"
      ],
      "metadata": {
        "id": "lEwkI9NAwwav",
        "outputId": "7f8d0cf3-7ced-445e-aec8-28783bde80fd",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Pipeline(steps=[('FeatureEngineering',\n",
              "                 Pipeline(steps=[('diameter',\n",
              "                                  FunctionTransformer(func=<function combine_est_diameter at 0x7f0546797b00>)),\n",
              "                                 ('year',\n",
              "                                  FunctionTransformer(func=<function year_transformer at 0x7f0546795560>)),\n",
              "                                 ('dropper',\n",
              "                                  FunctionTransformer(func=<function dropper at 0x7f0546795dd0>))])),\n",
              "                ('preproc',\n",
              "                 ColumnTransformer(remainder='passthrough',\n",
              "                                   tran...\n",
              "                                                                   SimpleImputer())]),\n",
              "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x7f05446a5410>),\n",
              "                                                 ('cat_pipe',\n",
              "                                                  Pipeline(steps=[('cat_impute',\n",
              "                                                                   SimpleImputer(strategy='most_frequent'))]),\n",
              "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x7f05446a5b90>)])),\n",
              "                ('mdl',\n",
              "                 KNeighborsClassifier(leaf_size=10, n_neighbors=10, p=1,\n",
              "                                      weights='distance'))])"
            ]
          },
          "metadata": {},
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Performance Metrics\n",
        "\n",
        "Given I have an extremely unbalanced dataset, I will assume that it will be difficult to accurately predict `true` values. Recall will be an importance metric to look at as the false negative result is very costly in this scenario. Precision and f1-score will also be looked at but won't be as important. I expect logistic regression to perform the worst but have no insight yet on which model will perform better between support vector machines and k-nearest neighbors."
      ],
      "metadata": {
        "id": "26rS_B-zq5c9"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Logistic Regression Performance Metrics"
      ],
      "metadata": {
        "id": "8MczBXCatOgA"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Logistic Regression Classification Report\n",
        "print(classification_report(y_test, logistic_regression_grid_search.best_estimator_.predict(X_test)))"
      ],
      "metadata": {
        "id": "9GVsIkzJt7U_",
        "outputId": "b3873b94-8ce5-458b-e79c-605d28f1d364",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "       False       0.90      0.99      0.95     36079\n",
            "        True       0.39      0.03      0.06      3889\n",
            "\n",
            "    accuracy                           0.90     39968\n",
            "   macro avg       0.65      0.51      0.50     39968\n",
            "weighted avg       0.86      0.90      0.86     39968\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Logistic Regression Confusion Matrix\n",
        "lr_cm = confusion_matrix(y_test, logistic_regression_grid_search.best_estimator_.predict(X_test))\n",
        "lr_disp = ConfusionMatrixDisplay(confusion_matrix=lr_cm,\n",
        "                                    display_labels=logistic_regression_grid_search.classes_)\n",
        "lr_disp.plot()\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "FH-JWA9MuR4m",
        "outputId": "c5269842-6030-4a9f-86bf-a657677d24e5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        }
      },
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Support Vector Machines Performance Metrics"
      ],
      "metadata": {
        "id": "7evZIkIdtUKg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Support Vector Machines Classification Report\n",
        "print(classification_report(y_test, support_vector_grid_search.best_estimator_.predict(X_test)))"
      ],
      "metadata": {
        "id": "0ICSeO-ztaZA",
        "outputId": "e98bd556-d9d9-4e5c-bb27-9cf01118a423",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "       False       0.91      1.00      0.95     36079\n",
            "        True       0.76      0.11      0.19      3889\n",
            "\n",
            "    accuracy                           0.91     39968\n",
            "   macro avg       0.83      0.55      0.57     39968\n",
            "weighted avg       0.90      0.91      0.88     39968\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Support Vector Machines Confusion Matrix\n",
        "svm_cm = confusion_matrix(y_test, support_vector_grid_search.best_estimator_.predict(X_test))\n",
        "svm_disp = ConfusionMatrixDisplay(confusion_matrix=svm_cm,\n",
        "                                    display_labels=support_vector_grid_search.classes_)\n",
        "svm_disp.plot()\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "YbovceAhtoAG",
        "outputId": "eb347e3a-36dc-499c-8ac8-c70e3d5227f5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        }
      },
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### K-Nearest Neighbors Performance Metrics"
      ],
      "metadata": {
        "id": "KJ6CxgXLw3bD"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# K-Nearest Neighbors Classification Report\n",
        "print(classification_report(y_test, knn_grid_search.best_estimator_.predict(X_test)))"
      ],
      "metadata": {
        "id": "kF3cOj81w7m-",
        "outputId": "1e1574c5-ccdc-4f66-e49c-47e9449d6fc6",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "       False       0.93      0.98      0.95     36079\n",
            "        True       0.60      0.32      0.42      3889\n",
            "\n",
            "    accuracy                           0.91     39968\n",
            "   macro avg       0.77      0.65      0.69     39968\n",
            "weighted avg       0.90      0.91      0.90     39968\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# K-Nearest Neighbors Confusion Matrix\n",
        "knn_cm = confusion_matrix(y_test, knn_grid_search.best_estimator_.predict(X_test))\n",
        "knn_disp = ConfusionMatrixDisplay(confusion_matrix=knn_cm,\n",
        "                                    display_labels=knn_grid_search.classes_)\n",
        "knn_disp.plot()\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "m-rPz3KGxS4p",
        "outputId": "e94e2adc-a967-477a-d121-8eff629660b6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        }
      },
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Summary of Preliminary Modeling Results\n",
        "\n",
        "3 different models were fitted and tuned using Grid Search for hyperparameter tuning. It was very clear from the beginning that the data were extremely unbalanced and this was quickly proven by disproportionately lower performance metrics for the `true` labeled result.  All models performed well in predicting whether a datapoint was a non-hazardous asteroid (`false`) with precision, recall, and f1 scores being all 0.90 or higher. \n",
        "\n",
        "\n",
        "\n",
        "Therefore the best performing model should be the one which did the best with predicting hazardous asteroids (`true`). The support vector machines model provided the highest precision by far for `true` results at 0.78 but had a very poor recall at 0.11 giving it an f1 score of 0.19. The k-nearest neighbors model had a precision of 0.58 and a recall of 0.32 giving it an f1 score of 0.42. The logistic regression model had low performance metrics across the board and won't be considered going foward. \n",
        "\n",
        "\n",
        "\n",
        "Since logistic regression assumes some sort of linearity this may have been the downfall of the model as it's not entirely certain if there are any linear relationships in this dataset. Support vector machines perform well with highly dimensional datasets and can also deal with unbalanced datasets with the `weights` parameter which was tuned during the grid search. K-nearest neighbors0 classifiers can also deal with unbalanced datasets in a similar way which could be a reason why it also performed better than the logistic regression model among other reasons. \n",
        "\n",
        "\n",
        "\n",
        "In this scenario I believe that recall is the most important metric because the false negative instance (predicting an asteroid to be non-hazardous and its true value is hazardous) is the most costly by far so we would want recall to be as high as possible. Given this I predict that the K-Nearest Neighbors model is most likely the best performing model when it comes to business value.\n",
        "\n",
        "    "
      ],
      "metadata": {
        "id": "q-1EgfhAyH5G"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Comparing models using business metrics\n",
        "\n",
        "    "
      ],
      "metadata": {
        "id": "wsi-pcWSTxBe"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Logistic Regression Business Costs\n",
        "\n",
        "lr_true_non = 10\n",
        "lr_false_non = 10 + (5*(lr_cm[0][1]))\n",
        "lr_false_haz = 10 + (100*(lr_cm[1][0]))\n",
        "lr_true_haz = 10 + (5*(lr_cm[1][1]))\n",
        "\n",
        "lr_total_cost = (lr_true_non + lr_false_non + lr_false_haz + lr_true_haz) // 1000\n",
        "\n",
        "print(f'The total business cost using a logistic regression model is ${lr_total_cost} B')"
      ],
      "metadata": {
        "id": "AnpDt7pJT-Nh",
        "outputId": "f26a03e8-8cf1-434c-d340-a9fb9669d8c1",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The total business cost using a logistic regression model is $378 B\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Support Vector Machines Business Costs\n",
        "\n",
        "svm_true_non = 10\n",
        "svm_false_non = 10 + (5*(svm_cm[0][1]))\n",
        "svm_false_haz = 10 + (100*(svm_cm[1][0]))\n",
        "svm_true_haz = 10 + (5*(svm_cm[1][1]))\n",
        "\n",
        "svm_total_cost = (svm_true_non + svm_false_non + svm_false_haz + svm_true_haz) // 1000\n",
        "\n",
        "print(f'The total business cost using a support vector machines model is ${svm_total_cost} B')"
      ],
      "metadata": {
        "id": "cLTQguJrUCTd",
        "outputId": "9ea86778-9ac6-49d3-a0e4-5e78fd774861",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The total business cost using a support vector machines model is $349 B\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# K-Nearest Neighbors Business Costs\n",
        "\n",
        "knn_true_non = 10\n",
        "knn_false_non = 10 + (5*(knn_cm[0][1]))\n",
        "knn_false_haz = 10 + (100*(knn_cm[1][0]))\n",
        "knn_true_haz = 10 + (5*(knn_cm[1][1]))\n",
        "\n",
        "knn_total_cost = (knn_true_non + knn_false_non + knn_false_haz + knn_true_haz) // 1000\n",
        "\n",
        "print(f'The total business cost using a k-nearest neighbors model is ${knn_total_cost} B')"
      ],
      "metadata": {
        "id": "QWHbS6z3UEvY",
        "outputId": "592d3e2e-4a6f-4777-b156-c68e62667368",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The total business cost using a k-nearest neighbors model is $273 B\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_comp_names = ['Log Reg', 'SVM', 'KNN']\n",
        "model_costs = [lr_total_cost, svm_total_cost, knn_total_cost]\n",
        "\n",
        "model_comp = pd.DataFrame({'Model Name': model_comp_names, 'Cost in $B': model_costs})\n",
        "model_comp.plot.bar(x='Model Name', y='Cost in $B', title='Cost in Billions of Dollars per Model')\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "l-0eJ8vAc2e4",
        "outputId": "144f472e-c0b0-4839-e31f-9425198c372f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 325
        }
      },
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Using the formulas in the README.md file I calculated business costs for all scenarios in the confusion matrix for each model. As predicted the K-Nearest Neighbors model had the lowest business cost. It was \\$76 B cheaper than the Support Vector Machines model and it was over \\$100 B cheaper than the Logistic Regression model."
      ],
      "metadata": {
        "id": "iVhF9EC1anl0"
      }
    }
  ]
}