Loan Eligibility Prediction.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "9a923872",
   "metadata": {},
   "source": [
    "### Our Goal\n",
    "\n",
    "Given historical data on loans given out with information on whether or not the borrower defaulted (charge-off), can we build a model thatcan predict wether or nor a borrower will pay back their loan? This way in the future when we get a new potential customer we can assess whether or not they are likely to pay back the loan. Keep in mind classification metrics when evaluating the performance of your model!\n",
    "\n",
    "The \"loan_status\" column contains our label.\n",
    "\n",
    "### Data Overview"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "132a568e",
   "metadata": {},
   "source": [
    "----\n",
    "-----\n",
    "There are many LendingClub data sets on Kaggle. Here is the information on this particular data set:\n",
    "\n",
    "<table border=\"1\" class=\"dataframe\">\n",
    "  <thead>\n",
    "    <tr style=\"text-align: right;\">\n",
    "      <th></th>\n",
    "      <th>LoanStatNew</th>\n",
    "      <th>Description</th>\n",
    "    </tr>\n",
    "  </thead>\n",
    "  <tbody>\n",
    "    <tr>\n",
    "      <th>0</th>\n",
    "      <td>loan_amnt</td>\n",
    "      <td>The listed amount of the loan applied for by the borrower. If at some point in time, the credit department reduces the loan amount, then it will be reflected in this value.</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>1</th>\n",
    "      <td>term</td>\n",
    "      <td>The number of payments on the loan. Values are in months and can be either 36 or 60.</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>2</th>\n",
    "      <td>int_rate</td>\n",
    "      <td>Interest Rate on the loan</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>3</th>\n",
    "      <td>installment</td>\n",
    "      <td>The monthly payment owed by the borrower if the loan originates.</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>4</th>\n",
    "      <td>grade</td>\n",
    "      <td>LC assigned loan grade</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>5</th>\n",
    "      <td>sub_grade</td>\n",
    "      <td>LC assigned loan subgrade</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>6</th>\n",
    "      <td>emp_title</td>\n",
    "      <td>The job title supplied by the Borrower when applying for the loan.*</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>7</th>\n",
    "      <td>emp_length</td>\n",
    "      <td>Employment length in years. Possible values are between 0 and 10 where 0 means less than one year and 10 means ten or more years.</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>8</th>\n",
    "      <td>home_ownership</td>\n",
    "      <td>The home ownership status provided by the borrower during registration or obtained from the credit report. Our values are: RENT, OWN, MORTGAGE, OTHER</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>9</th>\n",
    "      <td>annual_inc</td>\n",
    "      <td>The self-reported annual income provided by the borrower during registration.</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>10</th>\n",
    "      <td>verification_status</td>\n",
    "      <td>Indicates if income was verified by LC, not verified, or if the income source was verified</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>11</th>\n",
    "      <td>issue_d</td>\n",
    "      <td>The month which the loan was funded</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>12</th>\n",
    "      <td>loan_status</td>\n",
    "      <td>Current status of the loan</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>13</th>\n",
    "      <td>purpose</td>\n",
    "      <td>A category provided by the borrower for the loan request.</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>14</th>\n",
    "      <td>title</td>\n",
    "      <td>The loan title provided by the borrower</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>15</th>\n",
    "      <td>zip_code</td>\n",
    "      <td>The first 3 numbers of the zip code provided by the borrower in the loan application.</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>16</th>\n",
    "      <td>addr_state</td>\n",
    "      <td>The state provided by the borrower in the loan application</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>17</th>\n",
    "      <td>dti</td>\n",
    "      <td>A ratio calculated using the borrower’s total monthly debt payments on the total debt obligations, excluding mortgage and the requested LC loan, divided by the borrower’s self-reported monthly income.</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>18</th>\n",
    "      <td>earliest_cr_line</td>\n",
    "      <td>The month the borrower's earliest reported credit line was opened</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>19</th>\n",
    "      <td>open_acc</td>\n",
    "      <td>The number of open credit lines in the borrower's credit file.</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>20</th>\n",
    "      <td>pub_rec</td>\n",
    "      <td>Number of derogatory public records</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>21</th>\n",
    "      <td>revol_bal</td>\n",
    "      <td>Total credit revolving balance</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>22</th>\n",
    "      <td>revol_util</td>\n",
    "      <td>Revolving line utilization rate, or the amount of credit the borrower is using relative to all available revolving credit.</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>23</th>\n",
    "      <td>total_acc</td>\n",
    "      <td>The total number of credit lines currently in the borrower's credit file</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>24</th>\n",
    "      <td>initial_list_status</td>\n",
    "      <td>The initial listing status of the loan. Possible values are – W, F</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>25</th>\n",
    "      <td>application_type</td>\n",
    "      <td>Indicates whether the loan is an individual application or a joint application with two co-borrowers</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>26</th>\n",
    "      <td>mort_acc</td>\n",
    "      <td>Number of mortgage accounts.</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "      <th>27</th>\n",
    "      <td>pub_rec_bankruptcies</td>\n",
    "      <td>Number of public record bankruptcies</td>\n",
    "    </tr>\n",
    "  </tbody>\n",
    "</table>\n",
    "\n",
    "---\n",
    "----"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "97cebea3",
   "metadata": {},
   "source": [
    "## Starter Code\n",
    "\n",
    "#### Note: We also provide feature information on the data as a .csv file for easy lookup throughout the notebook:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "id": "128bdf1e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "id": "10e59793",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_info = pd.read_csv('./Data/lending_club_info.csv', index_col='LoanStatNew')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "id": "4b33bd87",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Revolving line utilization rate, or the amount of credit the borrower is using relative to all available revolving credit.\n"
     ]
    }
   ],
   "source": [
    "print(data_info.loc['revol_util']['Description'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "id": "16c113bb",
   "metadata": {},
   "outputs": [],
   "source": [
    "def feat_info(col_name):\n",
    "    print(data_info.loc[col_name]['Description'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "id": "0c9df4df",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of mortgage accounts.\n"
     ]
    }
   ],
   "source": [
    "feat_info('mort_acc')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "baf989e9",
   "metadata": {},
   "source": [
    "## Loading the data and other imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "id": "12f29651",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "id": "3c031521",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('./Data/lending_club_loan_two.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "id": "490b0870",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 396030 entries, 0 to 396029\n",
      "Data columns (total 27 columns):\n",
      " #   Column                Non-Null Count   Dtype  \n",
      "---  ------                --------------   -----  \n",
      " 0   loan_amnt             396030 non-null  float64\n",
      " 1   term                  396030 non-null  object \n",
      " 2   int_rate              396030 non-null  float64\n",
      " 3   installment           396030 non-null  float64\n",
      " 4   grade                 396030 non-null  object \n",
      " 5   sub_grade             396030 non-null  object \n",
      " 6   emp_title             373103 non-null  object \n",
      " 7   emp_length            377729 non-null  object \n",
      " 8   home_ownership        396030 non-null  object \n",
      " 9   annual_inc            396030 non-null  float64\n",
      " 10  verification_status   396030 non-null  object \n",
      " 11  issue_d               396030 non-null  object \n",
      " 12  loan_status           396030 non-null  object \n",
      " 13  purpose               396030 non-null  object \n",
      " 14  title                 394275 non-null  object \n",
      " 15  dti                   396030 non-null  float64\n",
      " 16  earliest_cr_line      396030 non-null  object \n",
      " 17  open_acc              396030 non-null  float64\n",
      " 18  pub_rec               396030 non-null  float64\n",
      " 19  revol_bal             396030 non-null  float64\n",
      " 20  revol_util            395754 non-null  float64\n",
      " 21  total_acc             396030 non-null  float64\n",
      " 22  initial_list_status   396030 non-null  object \n",
      " 23  application_type      396030 non-null  object \n",
      " 24  mort_acc              358235 non-null  float64\n",
      " 25  pub_rec_bankruptcies  395495 non-null  float64\n",
      " 26  address               396030 non-null  object \n",
      "dtypes: float64(12), object(15)\n",
      "memory usage: 81.6+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c120f5e1",
   "metadata": {},
   "source": [
    "------\n",
    "\n",
    "# Section 1: Exploratory Data Analysis\n",
    "\n",
    "**OVERALL GOAL: Get an understanding for which variables are important, view summary statistics, and visualize the data**\n",
    "\n",
    "------"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 183,
   "id": "4cfa9727",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='loan_status', ylabel='count'>"
      ]
     },
     "execution_count": 183,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(x='loan_status', data=df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "id": "f325e4bd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='loan_amnt', ylabel='Count'>"
      ]
     },
     "execution_count": 184,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "sns.histplot(data=df, x='loan_amnt', bins=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "id": "d7216f1c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>loan_amnt</th>\n",
       "      <th>int_rate</th>\n",
       "      <th>installment</th>\n",
       "      <th>annual_inc</th>\n",
       "      <th>dti</th>\n",
       "      <th>open_acc</th>\n",
       "      <th>pub_rec</th>\n",
       "      <th>revol_bal</th>\n",
       "      <th>revol_util</th>\n",
       "      <th>total_acc</th>\n",
       "      <th>mort_acc</th>\n",
       "      <th>pub_rec_bankruptcies</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>loan_amnt</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.168921</td>\n",
       "      <td>0.953929</td>\n",
       "      <td>0.336887</td>\n",
       "      <td>0.016636</td>\n",
       "      <td>0.198556</td>\n",
       "      <td>-0.077779</td>\n",
       "      <td>0.328320</td>\n",
       "      <td>0.099911</td>\n",
       "      <td>0.223886</td>\n",
       "      <td>0.222315</td>\n",
       "      <td>-0.106539</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>int_rate</th>\n",
       "      <td>0.168921</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.162758</td>\n",
       "      <td>-0.056771</td>\n",
       "      <td>0.079038</td>\n",
       "      <td>0.011649</td>\n",
       "      <td>0.060986</td>\n",
       "      <td>-0.011280</td>\n",
       "      <td>0.293659</td>\n",
       "      <td>-0.036404</td>\n",
       "      <td>-0.082583</td>\n",
       "      <td>0.057450</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>installment</th>\n",
       "      <td>0.953929</td>\n",
       "      <td>0.162758</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.330381</td>\n",
       "      <td>0.015786</td>\n",
       "      <td>0.188973</td>\n",
       "      <td>-0.067892</td>\n",
       "      <td>0.316455</td>\n",
       "      <td>0.123915</td>\n",
       "      <td>0.202430</td>\n",
       "      <td>0.193694</td>\n",
       "      <td>-0.098628</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>annual_inc</th>\n",
       "      <td>0.336887</td>\n",
       "      <td>-0.056771</td>\n",
       "      <td>0.330381</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.081685</td>\n",
       "      <td>0.136150</td>\n",
       "      <td>-0.013720</td>\n",
       "      <td>0.299773</td>\n",
       "      <td>0.027871</td>\n",
       "      <td>0.193023</td>\n",
       "      <td>0.236320</td>\n",
       "      <td>-0.050162</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dti</th>\n",
       "      <td>0.016636</td>\n",
       "      <td>0.079038</td>\n",
       "      <td>0.015786</td>\n",
       "      <td>-0.081685</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.136181</td>\n",
       "      <td>-0.017639</td>\n",
       "      <td>0.063571</td>\n",
       "      <td>0.088375</td>\n",
       "      <td>0.102128</td>\n",
       "      <td>-0.025439</td>\n",
       "      <td>-0.014558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>open_acc</th>\n",
       "      <td>0.198556</td>\n",
       "      <td>0.011649</td>\n",
       "      <td>0.188973</td>\n",
       "      <td>0.136150</td>\n",
       "      <td>0.136181</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.018392</td>\n",
       "      <td>0.221192</td>\n",
       "      <td>-0.131420</td>\n",
       "      <td>0.680728</td>\n",
       "      <td>0.109205</td>\n",
       "      <td>-0.027732</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pub_rec</th>\n",
       "      <td>-0.077779</td>\n",
       "      <td>0.060986</td>\n",
       "      <td>-0.067892</td>\n",
       "      <td>-0.013720</td>\n",
       "      <td>-0.017639</td>\n",
       "      <td>-0.018392</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.101664</td>\n",
       "      <td>-0.075910</td>\n",
       "      <td>0.019723</td>\n",
       "      <td>0.011552</td>\n",
       "      <td>0.699408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>revol_bal</th>\n",
       "      <td>0.328320</td>\n",
       "      <td>-0.011280</td>\n",
       "      <td>0.316455</td>\n",
       "      <td>0.299773</td>\n",
       "      <td>0.063571</td>\n",
       "      <td>0.221192</td>\n",
       "      <td>-0.101664</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.226346</td>\n",
       "      <td>0.191616</td>\n",
       "      <td>0.194925</td>\n",
       "      <td>-0.124532</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>revol_util</th>\n",
       "      <td>0.099911</td>\n",
       "      <td>0.293659</td>\n",
       "      <td>0.123915</td>\n",
       "      <td>0.027871</td>\n",
       "      <td>0.088375</td>\n",
       "      <td>-0.131420</td>\n",
       "      <td>-0.075910</td>\n",
       "      <td>0.226346</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.104273</td>\n",
       "      <td>0.007514</td>\n",
       "      <td>-0.086751</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_acc</th>\n",
       "      <td>0.223886</td>\n",
       "      <td>-0.036404</td>\n",
       "      <td>0.202430</td>\n",
       "      <td>0.193023</td>\n",
       "      <td>0.102128</td>\n",
       "      <td>0.680728</td>\n",
       "      <td>0.019723</td>\n",
       "      <td>0.191616</td>\n",
       "      <td>-0.104273</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.381072</td>\n",
       "      <td>0.042035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mort_acc</th>\n",
       "      <td>0.222315</td>\n",
       "      <td>-0.082583</td>\n",
       "      <td>0.193694</td>\n",
       "      <td>0.236320</td>\n",
       "      <td>-0.025439</td>\n",
       "      <td>0.109205</td>\n",
       "      <td>0.011552</td>\n",
       "      <td>0.194925</td>\n",
       "      <td>0.007514</td>\n",
       "      <td>0.381072</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.027239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pub_rec_bankruptcies</th>\n",
       "      <td>-0.106539</td>\n",
       "      <td>0.057450</td>\n",
       "      <td>-0.098628</td>\n",
       "      <td>-0.050162</td>\n",
       "      <td>-0.014558</td>\n",
       "      <td>-0.027732</td>\n",
       "      <td>0.699408</td>\n",
       "      <td>-0.124532</td>\n",
       "      <td>-0.086751</td>\n",
       "      <td>0.042035</td>\n",
       "      <td>0.027239</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      loan_amnt  int_rate  installment  annual_inc       dti  \\\n",
       "loan_amnt              1.000000  0.168921     0.953929    0.336887  0.016636   \n",
       "int_rate               0.168921  1.000000     0.162758   -0.056771  0.079038   \n",
       "installment            0.953929  0.162758     1.000000    0.330381  0.015786   \n",
       "annual_inc             0.336887 -0.056771     0.330381    1.000000 -0.081685   \n",
       "dti                    0.016636  0.079038     0.015786   -0.081685  1.000000   \n",
       "open_acc               0.198556  0.011649     0.188973    0.136150  0.136181   \n",
       "pub_rec               -0.077779  0.060986    -0.067892   -0.013720 -0.017639   \n",
       "revol_bal              0.328320 -0.011280     0.316455    0.299773  0.063571   \n",
       "revol_util             0.099911  0.293659     0.123915    0.027871  0.088375   \n",
       "total_acc              0.223886 -0.036404     0.202430    0.193023  0.102128   \n",
       "mort_acc               0.222315 -0.082583     0.193694    0.236320 -0.025439   \n",
       "pub_rec_bankruptcies  -0.106539  0.057450    -0.098628   -0.050162 -0.014558   \n",
       "\n",
       "                      open_acc   pub_rec  revol_bal  revol_util  total_acc  \\\n",
       "loan_amnt             0.198556 -0.077779   0.328320    0.099911   0.223886   \n",
       "int_rate              0.011649  0.060986  -0.011280    0.293659  -0.036404   \n",
       "installment           0.188973 -0.067892   0.316455    0.123915   0.202430   \n",
       "annual_inc            0.136150 -0.013720   0.299773    0.027871   0.193023   \n",
       "dti                   0.136181 -0.017639   0.063571    0.088375   0.102128   \n",
       "open_acc              1.000000 -0.018392   0.221192   -0.131420   0.680728   \n",
       "pub_rec              -0.018392  1.000000  -0.101664   -0.075910   0.019723   \n",
       "revol_bal             0.221192 -0.101664   1.000000    0.226346   0.191616   \n",
       "revol_util           -0.131420 -0.075910   0.226346    1.000000  -0.104273   \n",
       "total_acc             0.680728  0.019723   0.191616   -0.104273   1.000000   \n",
       "mort_acc              0.109205  0.011552   0.194925    0.007514   0.381072   \n",
       "pub_rec_bankruptcies -0.027732  0.699408  -0.124532   -0.086751   0.042035   \n",
       "\n",
       "                      mort_acc  pub_rec_bankruptcies  \n",
       "loan_amnt             0.222315             -0.106539  \n",
       "int_rate             -0.082583              0.057450  \n",
       "installment           0.193694             -0.098628  \n",
       "annual_inc            0.236320             -0.050162  \n",
       "dti                  -0.025439             -0.014558  \n",
       "open_acc              0.109205             -0.027732  \n",
       "pub_rec               0.011552              0.699408  \n",
       "revol_bal             0.194925             -0.124532  \n",
       "revol_util            0.007514             -0.086751  \n",
       "total_acc             0.381072              0.042035  \n",
       "mort_acc              1.000000              0.027239  \n",
       "pub_rec_bankruptcies  0.027239              1.000000  "
      ]
     },
     "execution_count": 185,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "id": "d2193879",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 186,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "sns.heatmap(df.corr(), annot=True, cmap='YlGnBu')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e117a8e1",
   "metadata": {},
   "source": [
    "**Since there's a high correlation b/w installments and Loan Amount, let's explore them further!**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 187,
   "id": "af995d5b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The monthly payment owed by the borrower if the loan originates.\n",
      "The listed amount of the loan applied for by the borrower. If at some point in time, the credit department reduces the loan amount, then it will be reflected in this value.\n"
     ]
    }
   ],
   "source": [
    "feat_info('installment')\n",
    "feat_info('loan_amnt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 188,
   "id": "8dd55652",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='installment', ylabel='loan_amnt'>"
      ]
     },
     "execution_count": 188,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.scatterplot(x='installment', y='loan_amnt', data=df)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0405b851",
   "metadata": {},
   "source": [
    "**Let's analyze the relation of \"Loan Status\" with the loan amount.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "id": "84d2944f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='loan_status', ylabel='loan_amnt'>"
      ]
     },
     "execution_count": 189,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZIAAAEHCAYAAACEKcAKAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAgkElEQVR4nO3df5xVdb3v8debwRQzVHAk7oBBDZ1SMsrRS9k94Y+SYyp60hM9ugfscqPjNaJzu5XaOSfrPPCRj255hNRzMb2iWUqUiR6lVDQ7NwIHIxF/5BQovwJERfzFkfFz/1jfkTXDZmY2a/bsmdnv5+OxH3utz/5+1/4u3M5nf9d37e9XEYGZmdn+GlTtBpiZWf/mRGJmZoU4kZiZWSFOJGZmVogTiZmZFTK42g2ohiOOOCLGjBlT7WaYmfUrK1eufDYi6jvGazKRjBkzhubm5mo3w8ysX5H0dKm4L22ZmVkhTiRmZlaIE4mZmRXiRGJmZoU4kVghkyZNevNh1pfMnTuXSZMmcdVVV1W7KQNeryQSSXWSfifpzrQ/TNI9kp5Kz4fnyl4sqUXSk5JOy8WPk7Q6vTZXklL8QEm3pvhySWN645zMrG/72c9+BsBPfvKTKrdk4OutHsls4PHc/kXAfRExDrgv7SPpaGAqcAwwGbhaUl2qcw0wExiXHpNTfAbwfEQ0AlcAl1f2VKxNx16IeyXWV8ydO7fdvnsllVXxRCJpFPAJ4Ae58BRgQdpeAJydi98SEbsiYi3QApwgaSQwNCKWRTbv/Y0d6rQdaxFwSltvxcxqU1tvpI17JZXVGz2SfwG+CryRi42IiM0A6fnIFG8A1ufKbUixhrTdMd6uTkTsBnYAwzs2QtJMSc2Smrdt21bwlMzMrE1FE4mkM4CtEbGyu1VKxKKTeGd12gci5kdEU0Q01dfv9Qt/MzPbT5WeIuVE4CxJpwMHAUMl/RDYImlkRGxOl622pvIbgNG5+qOATSk+qkQ8X2eDpMHAocBzlTohM+v7Bg0axBtvvNFu3yqnov+6EXFxRIyKiDFkg+hLI+K/AouB6anYdOD2tL0YmJruxBpLNqi+Il3+2ilpYhr/mNahTtuxzk3v4fWDzWrYGWec0W7/zDPPrFJLakO10vS3gY9Jegr4WNonItYAC4HHgCXAhRHRmupcQDZg3wL8Ebg7xa8DhktqAf4n6Q4wq7yO9zT4HgfrK6ZPn95uf9q0aVVqSW3otdl/I+IB4IG0vR04ZR/l5gBzSsSbgfEl4q8B5/VgU62b6uvr2bp165v7Rx55ZCelzXrP8OHDOeuss7jjjjs466yzGD58r/tvrAfV5DTy1jO2b9/ebv/ZZ5+tUkvM9jZ9+nTWrVvn3kgvcCKx/ZYfzCy1b1ZNw4cP3+uHiVYZvpXB9lvHexp8j4NZbXIiMbMBafv27Xzxi1/c6xKs9TwnEjMbkBYsWMDq1au58cYbq92UAc+JxMwGnO3bt7NkyRIigiVLlrhXUmFOJLbf6urqOt03q5YFCxa8efNHa2ureyUV5kRi+621tbXTfbNquffee9m9ezcAu3fv5p577qlyiwY2JxIzG3BOPfVUBg/Oft0wePBgPvaxj1W5RQObf0fST82bN4+WlpZqN2Mvs2fPrsr7NjY2MmvWrKq8t/U906dPZ8mSJUB2ydU/Sqws90jMbMAZPnw4kydPRhKTJ0/2FCkVplr8EVlTU1M0NzdXuxn93tKlS/nWt7715v43vvENTjrppCq2yPqKvtBjXr9+PS+++CLvfve7OeCAA6raloHSY5a0MiKaOsbdI7H9dvLJJ7+5PXjwYCcR61N27drFgQceWPUkUgs8RmKFjB49mvXr1/P1r3+92k2xPqQvfPtuG6+78sorq9ySgc+JxAoZNmwYw4YNc2/ErIb50paZmRVS0UQi6SBJKyT9XtIaSd9M8UslbZS0Kj1Oz9W5WFKLpCclnZaLHydpdXptblpyl7Qs760pvlzSmEqek5mZtVfpHsku4OSIeD8wAZgsaWJ67YqImJAedwFIOppsbfdjgMnA1ZLa5t24BphJto77uPQ6wAzg+YhoBK4ALq/wOZmZWU5FE0lkXkq7B6RHZ/cbTwFuiYhdEbGWbH32EySNBIZGxLLI7le+ETg7V2dB2l4EnNLWWzEzs8qr+BiJpDpJq4CtwD0RsTy99AVJj0i6XtLhKdYArM9V35BiDWm7Y7xdnYjYDewA9vr1kaSZkpolNW/btq1nTs7MzCqfSCKiNSImAKPIehfjyS5TvYvsctdm4LupeKmeRHQS76xOx3bMj4imiGiqr68v6xzMzGzfeu2urYh4AXgAmBwRW1KCeQO4FjghFdsAjM5VGwVsSvFRJeLt6kgaDBwKPFeZszAzs44qfddWvaTD0vYQ4FTgiTTm0eYc4NG0vRiYmu7EGks2qL4iIjYDOyVNTOMf04Dbc3Wmp+1zgaVRi/O+mJlVSaV/kDgSWJDuvBoELIyIOyXdJGkC2SWodcDnASJijaSFwGPAbuDCiGhb5OIC4AZgCHB3egBcB9wkqYWsJzK1wudkZmY5FU0kEfEI8IES8b/tpM4cYE6JeDMwvkT8NeC8Yi01M7P95V+2m5lZIU4kZmZWiBOJmZkV4kRiZmaFOJGYmVkhTiRmZlaIE4mZmRXiRGJmZoU4kZiZWSFOJGZmVogTiZmZFeJEYmZmhTiRmJlZIU4kZmZWiBOJmZkV4kRiZmaFVHqp3YMkrZD0e0lrJH0zxYdJukfSU+n58FydiyW1SHpS0mm5+HGSVqfX5qYld0nL8t6a4ssljankOZmZWXuV7pHsAk6OiPcDE4DJkiYCFwH3RcQ44L60j6SjyZbKPQaYDFydlukFuAaYSbaO+7j0OsAM4PmIaASuAC6v8DmZmVlORRNJZF5KuwekRwBTgAUpvgA4O21PAW6JiF0RsRZoAU6QNBIYGhHLIiKAGzvUaTvWIuCUtt6KmZlVXsXHSCTVSVoFbAXuiYjlwIiI2AyQno9MxRuA9bnqG1KsIW13jLerExG7gR3A8BLtmCmpWVLztm3beujszMys4okkIlojYgIwiqx3Mb6T4qV6EtFJvLM6HdsxPyKaIqKpvr6+i1abmVl39dpdWxHxAvAA2djGlnS5ivS8NRXbAIzOVRsFbErxUSXi7epIGgwcCjxXiXMwM7O9VfqurXpJh6XtIcCpwBPAYmB6KjYduD1tLwampjuxxpINqq9Il792SpqYxj+mdajTdqxzgaVpHMXMzHrB4AoffySwIN15NQhYGBF3SloGLJQ0A3gGOA8gItZIWgg8BuwGLoyI1nSsC4AbgCHA3ekBcB1wk6QWsp7I1Aqfk5mZ5VQ0kUTEI8AHSsS3A6fso84cYE6JeDOw1/hKRLxGSkRmZtb7/Mt2MzMrxInEzMwKcSIxM7NCnEjMzKwQJxIzMyvEicTMzApxIjEzs0KcSMzMrBAnEjMzK8SJxMzMCnEiMTOzQpxIzMysECcSMzMrxInEzMwKcSIxM7NCnEjMzKyQSi+1O1rS/ZIel7RG0uwUv1TSRkmr0uP0XJ2LJbVIelLSabn4cZJWp9fmpiV3Scvy3priyyWNqeQ5mZlZe5XukewGvhwR7wUmAhdKOjq9dkVETEiPuwDSa1OBY4DJwNVpmV6Aa4CZZOu4j0uvA8wAno+IRuAK4PIKn5OZmeVUNJFExOaIeDht7wQeBxo6qTIFuCUidkXEWqAFOEHSSGBoRCyLiABuBM7O1VmQthcBp7T1VszMrPK6nUgkje1OrJP6Y8jWb1+eQl+Q9Iik6yUdnmINwPpctQ0p1pC2O8bb1YmI3cAOYHiJ958pqVlS87Zt27rbbDMz60I5PZKflogt6k5FSYek+l+KiBfJLlO9C5gAbAa+21a0RPXoJN5ZnfaBiPkR0RQRTfX19d1ptpmZdcPgrgpIeg/ZmMWhkv4699JQ4KBu1D+ALIncHBE/A4iILbnXrwXuTLsbgNG56qOATSk+qkQ8X2eDpMHAocBzXbXLzMx6Rnd6JH8BnAEcBpyZe3wQ+FxnFdNYxXXA4xHxvVx8ZK7YOcCjaXsxMDXdiTWWbFB9RURsBnZKmpiOOQ24PVdneto+F1iaxlHMzKwXdNkjiYjbgdslfSgilpV5/BOBvwVWS1qVYpcAn5Y0gewS1Drg8+m91khaCDxGdsfXhRHRmupdANwADAHuTg/IEtVNklrIeiJTy2yjmZkV0GUiyWmRdAkwJl8vIv7bvipExL9Tegzjrk7qzAHmlIg3A+NLxF8Dzuus4WZmVjnlJJLbgV8D9wKtXZQ1M7MaUU4iOTgivlaxlpiZWb9Uzu2/d+anMjEzM4PyEslssmTyqqQXJe2U9GKlGmZmZv1Dty9tRcTbKtkQMzPrn8oZI0HSsex919bPerhNZmbWj3Q7kUi6HjgWWAO8kcIBOJGYmdWwcnokEyPi6K6LmZlZLSlnsH1Zbi0RMzMzoLweyQKyZPJnYBfZL9YjIo6tSMvMzKxfKCeRXE+aN4s9YyRmZlbjykkkz0TE4oq1xMzM+qVyEskTkn4E3EF2aQvw7b9mZrWunEQyhCyBfDwX8+2/ZmY1rpxftn+2kg0xM7P+qZwfJB4EzCBbdvfNJXY7W4/EzMwGvnJ+R3IT8HbgNOBXZOum7+ysgqTRku6X9LikNZJmp/gwSfdIeio9H56rc7GkFklPSjotFz9O0ur02ty05C5pWd5bU3y5pDFlnJOZmRVUTiJpjIh/BF6OiAXAJ4D3dVFnN/DliHgvMBG4MP2o8SLgvogYB9yX9kmvTSXr9UwGrpZUl451DTCTbB33cel1yHpJz0dEI3AFcHkZ52RmZgWVM9j+enp+QdJ44M9kEzjuU0RsBjan7Z2SHgcagCnApFRsAfAA8LUUvyUidgFr0zrsJ0haBwxtWzNe0o3A2WTrtk8BLk3HWgR8X5IiIso4N7MBYd68ebS0tFS7GX1C27/D7Nmzq9ySvqGxsZFZs2ZV5NjlJJL56RLUPwCLgUOAf+xu5XTJ6QPAcmBESjJExGZJR6ZiDcBvc9U2pNjrabtjvK3O+nSs3ZJ2AMOBZzu8/0yyHg1HHXVUd5tt1q+0tLTw1JrfcdQhXg37La9nF1x2Pd1c5ZZU3zMv1XVdqIBy7tr6Qdp8EHhnx9clTU+XvPYi6RDgp8CXIuLFNLxRsmipt+4k3lmd9oGI+cB8gKamJvdWbMA66pBWLvmg15yzPS57eGhFj1/OGElXSvYfJR1AlkRuzv14cYukken1kcDWFN8AjM5VHwVsSvFRJeLt6kgaDBwKPFf0ZMzMrHvKWtiqC3v1DNKdVdcBj0fE93IvLQamA99Oz7fn4j+S9D3gP5ENqq+IiNa0tO9Esktj04B5HY61DDgXWFrp8RFfh97D16Hbq+R1aLO+qicTSak/3ieSJnqUtCrFLiFLIAslzQCeAc4DiIg1khYCj5Hd8XVhRLRd7L0AuIHsF/Z3pwdkieqmNDD/HNldXxXV0tLCqkcfp/XgYZV+qz5v0H9k/9lX/mlLlVtSfXWvuCNstamiPZKI+PdS8eSUUsGImAPMKRFvBsaXiL9GSkS9qfXgYbz6ntN7+22tDxvyxF3VboJZVfTkGMn/68FjmZlZP1HOFCkHAp8k++3Im/Ui4lvp+Qs93TgzM+v7yrm0dTuwA1hJbhp5MzOrbeUkklERMbnrYmZmVkvKGSP5jaSu5tYyM7MaU06P5CPA+ZLWkl3aEhARcWxFWmZmZv1COYnkryrWCjMz67fKmWvraYA0weJBXRQ3M7Ma0e0xEklnSXoKWEu2sNU69vy63MzMalQ5g+3/TLY41R8iYizZL9P9I0QzsxpXTiJ5PSK2A4MkDYqI+4EJlWmWmZn1F+UMtr+Q1hX5NXCzpK1kEyuamVkNK6dHMgV4FfgSsAT4I3BmBdpkZmb9SDl3bb0saQRwPLAduDtd6jIzsxpWzl1bfwOsIJuy/W+A5ZLOrVTDzMysfyhnjOTrwPERsRVAUj1wL7CoEg0zM7P+oZwxkkFtSSTZ3lV9SddL2irp0VzsUkkbJa1Kj9Nzr10sqUXSk5JOy8WPk7Q6vTY3LeGLpAMl3ZriyyWNKeN8zMysB5STSJZI+oWk8yWdD/wb0NWScDcApWYMviIiJqTHXQCSjiZbJveYVOdqSXWp/DXATLI13MfljjkDeD4iGoErgMvLOB8zM+sB3U4kEfEVYD5wLPB+YH5EfK2LOg+SraPeHVOAWyJiV0SsBVqAEySNBIZGxLKICOBG4OxcnQVpexFwSltvxczMekdZa7ZHxE+Bn/bA+35B0jSgGfhyRDwPNAC/zZXZkGKvp+2OcdLz+tS23ZJ2AMOBZzu+oaSZZL0ajjrqqB44BTMzg270SCTtlPRiicdOSS/ux3teA7yL7Ffxm4Hvtr1VibLRSbyzOnsHI+ZHRFNENNXX15fVYDMz27cueyQR8baefMOI2NK2Lela4M60uwEYnSs6CtiU4qNKxPN1NkgaDBxK9y+lmZlZDyhnsL1HpDGPNucAbXd0LQampjuxxpINqq+IiM3ATkkT0/jHNLL149vqTE/b5wJL0ziKmZn1krLGSMol6cfAJOAISRuAbwCTJE0guwS1Dvg8QESskbQQeIxsDq8LI6I1HeoCsjvAhpBNXd82ff11wE2SWsh6IlMreT5mfd3GjRt5eWcdlz08tNpNsT7k6Z11vHXjxoodv6KJJCI+XSJ8XSfl5wBzSsSbgfEl4q+R/dLezMyqpKKJxMx6V0NDA7t2b+aSD+7PfTA2UF328FAObGjouuB+6vUxEjMzG1icSMzMrBAnEjMzK8SJxMzMCnEiMTOzQpxIzMysECcSMzMrxL8j2Q8bN26k7pUdDHmiq+VYrJbUvbKdjRt3V7sZZr3OPRIzMyvEPZL90NDQwJ93DebV95zedWGrGUOeuIuGhhHVboZZr3OPxMzMCnEiMTOzQpxIzMysECcSMzMrxInEzMwKqWgikXS9pK2SHs3Fhkm6R9JT6fnw3GsXS2qR9KSk03Lx4yStTq/NTUvukpblvTXFl0saU8nzMTOzvVW6R3IDMLlD7CLgvogYB9yX9pF0NNlSucekOldLqkt1rgFmkq3jPi53zBnA8xHRCFwBXF6xMzEzs5Iqmkgi4kGytdTzpgAL0vYC4Oxc/JaI2BURa4EW4ARJI4GhEbEsIgK4sUOdtmMtAk5p662YmVnvqMYYyYiI2AyQno9M8QZgfa7chhRrSNsd4+3qRMRuYAcwvNSbSpopqVlS87Zt23roVMzMrC8NtpfqSUQn8c7q7B2MmB8RTRHRVF9fv59NNDOzjqqRSLaky1Wk560pvgEYnSs3CtiU4qNKxNvVkTQYOJS9L6WZmVkFVSORLAamp+3pwO25+NR0J9ZYskH1Feny105JE9P4x7QOddqOdS6wNI2jmJlZL6nopI2SfgxMAo6QtAH4BvBtYKGkGcAzwHkAEbFG0kLgMWA3cGFEtKZDXUB2B9gQ4O70ALgOuElSC1lPZGolz8fMzPZW0UQSEZ/ex0un7KP8HGBOiXgzML5E/DVSIjIzs+roS4PtZmbWD3k9ErMB5pmX6rjs4aHVbkbVbXkl+5484uA3qtyS6nvmpTrGVfD4TiRmA0hjY2O1m9Bn/EdLCwAHvsP/JuOo7GfDicRsAJk1a1a1m9BnzJ49G4Arr7yyyi0Z+DxGYmZmhTiRmJlZIb60tZ/qXnmOIU/cVe1mVN2g114E4I2DPLhb98pzwIhqN8Os1zmR7AcPaO7R0rITgMZ3+g8ojPBnw2qSE8l+8IDmHh7QNDOPkZiZWSFOJGZmVogTiZmZFeJEYmZmhTiRmJlZIU4kZmZWiBOJmZkVUrVEImmdpNWSVklqTrFhku6R9FR6PjxX/mJJLZKelHRaLn5cOk6LpLlpOV4zM+sl1e6RnBQREyKiKe1fBNwXEeOA+9I+ko4mW0b3GGAycLWkulTnGmAm2UzJ49LrZmbWS6qdSDqaAixI2wuAs3PxWyJiV0SsBVqAEySNBIZGxLKICODGXB0zM+sF1UwkAfxS0kpJM1NsRERsBkjPR6Z4A7A+V3dDijWk7Y7xvUiaKalZUvO2bdt68DTMzGpbNefaOjEiNkk6ErhH0hOdlC017hGdxPcORswH5gM0NTWVLGNmZuWrWo8kIjal563AbcAJwJZ0uYr0vDUV3wCMzlUfBWxK8VEl4mZm1kuqkkgkvVXS29q2gY8DjwKLgemp2HTg9rS9GJgq6UBJY8kG1Veky187JU1Md2tNy9UxM7NeUK1LWyOA29KduoOBH0XEEkkPAQslzQCeAc4DiIg1khYCjwG7gQsjojUd6wLgBmAIcHd6mJlZL6lKIomIPwHvLxHfDpyyjzpzgDkl4s3A+J5uo5mZdU9fu/3XzMz6GScSMzMrxInEzMwKcSIxM7NCnEjMzKwQJxIzMyvEicTMzApxIjEzs0KcSMzMrBAnEjMzK8SJxMzMCnEiMTOzQpxIzMysECcSMzMrpJpL7ZrZADVv3jxaWlqq2oa29589e3ZV2wHQ2NjIrFmzqt2MinEiMbMBaciQIdVuQs1QRFS7DYVJmgxcCdQBP4iIb3dWvqmpKZqbm3ulbZXSF77xwZ5vfY2NjVVtx0D/xmfWF0haGRFNHeP9foxEUh1wFfBXwNHApyUdXd1W1Y4hQ4b4m59ZjRsIl7ZOAFrS8r1IugWYQra++4Dlb99m1lf0+x4J0ACsz+1vSDEzM+sFAyGRqERsr4EfSTMlNUtq3rZtWy80y8ysNgyERLIBGJ3bHwVs6lgoIuZHRFNENNXX1/da48zMBrqBkEgeAsZJGivpLcBUYHGV22RmVjP6/WB7ROyW9AXgF2S3/14fEWuq3Cwzs5rR7xMJQETcBdxV7XaYmdWigXBpy8zMqsiJxMzMChkQU6SUS9I24Olqt2MAOQJ4ttqNMCvBn82e9Y6I2Ou215pMJNazJDWXmn/HrNr82ewdvrRlZmaFOJGYmVkhTiTWE+ZXuwFm++DPZi/wGImZmRXiHomZmRXiRGJmZoU4kdQISa2SVuUeYzope76k76ftSyX9rzLe5wZJa9N7PCzpQ12U/00nxzm3u+9rfZ+kt0u6RdIfJT0m6S5J75Y0SdKdVW5byc+bMv8g6SlJf5B0v6Rjcq+fJ+lxSfen/R9LekTS3/dm+6ttQMy1Zd3yakRM6KX3+kpELJL0ceD/AMfuq2BEfLiX2mRVJEnAbcCCiJiaYhOAET1w7MERsbvocfbhQuDDwPsj4pX0mV4s6ZiIeA2YAfyPiLhf0tuBD0fEOyrUlj7LPZIaJmmdpCPSdpOkBzop+y5JD+f2x0la2cVbPAg0SjpE0n2ph7Ja0pTccV5Kz5L0/fRN9d+AI4ucm/U5JwGvR8S/tgUiYlVE/DrtHiJpkaQnJN2cEg+S/knSQ5IelTQ/F39A0mWSfgXMlnR86gksk/QdSY+mcnVp/6H0+udTvLuft68BsyLildTmXwK/AT4j6Z+AjwD/Kuk7wC+BI1Nv/L/07D9f3+ZEUjuG5C5r3VZu5Yj4I7AjfYsE+CxwQxfVzgRWA68B50TEB8n+oHy37Q9CzjnAXwDvAz5H9i3QBo7xQGdfPD4AfAk4GngncGKKfz8ijo+I8cAQ4IxcncMi4qMR8V3g/wJ/FxEfAlpzZWYAOyLieOB44HOSxtKNz5ukocBb02c/rxk4JiK+lbY/ExFfAc4C/hgRE3IJsiY4kdSOV9MHfEJEnLOfx/gB8FlJdcCngB/to9x3JK0CZpL9jyzgMkmPAPcCDex9SeMvgR9HRGtEbAKW7mcbrX9aEREbIuINYBUwJsVPkrRc0mrgZOCYXJ1bASQdBrwtItrG2/Kfy48D09LncTkwHBhHsc+bKLGcdy3zGElt282eLxMHdaP8T4FvkP1PtzIitu+j3FciYlHbjqTzgXrguIh4XdK6fbyf/+ccuNYAnd08sSu33QoMlnQQcDXQFBHrJV1K+8/Ny+m5Y+82T2SXpn7RLiidTheft4h4UdLLkt4ZEX/KvfRB4Fed1a017pHUtnXAcWn7k10VToOLvwCuIbuU0F2HAltTEjkJKDUY+SAwNV3THkl2CcwGjqXAgZI+1xZI4xof7aROW9J4VtIh7CMRRcTzwE5JE1Noau7lXwAXSDogvee7Jb2V7n/evgPMlTQk1T+VbFxkX73xmuQeSW37JnCdpEvIuv3dcTPw12QDi911M3CHpGayyxZPlChzG9mli9XAH/A3vgElIkLSOcC/SLqIbNxsHdm4SMM+6rwg6Vqyz8Q64KFO3mIGcK2kl4EHgB0p/gOyy2QPp3G5bcDZdP/zNg84HFgtqRX4MzAlIl7t4pRriqdIsbIo+03JoRHxj9Vui1kbSYdERNsdgBcBIyNidpWbVTPcI7FuS3d7vYvsm5xZX/IJSReT/U17Gji/us2pLe6RmJlZIR5sNzOzQpxIzMysECcSMzMrxInEzMwKcSIx60LbxJJVeu9Jkrqcd6y75cwqwYnErG+bRPcmsOxuObMe50Ri1k1p6vHvpCnNV0v6VIqXnCZf0hhlix5dK2mNpF+2TbWxj+N/MU1r/oiyBaDGAH8H/H3b1OSSzkyTGP5O0r2SRuyjXLuFmrRnuv6Rkh5M5R6ttenOrTL8OxKzLkh6KSIOkfRJsj/Yk4EjyKbs+M9k024cnCb5OwL4LdkMs+8AWsgmHVwlaSGwOCJ+uI/32QSMjYhdkg5LU4RcCrwUEf87lTkceCFNOfLfgfdGxJdLlLsBuLNt8szcOXwZOCgi5qRZnA+OiJ09/69mtcS/bDfrvo+Qph4HtihbVOl44G6yafL/EniD9tPkr42IVWl7JXumRy/lEeBmST8Hfr6PMqOAW9NEg28B1pZ5Dg8B16dJDH+ea5vZfvOlLbPu29d05Z9hzzT5E4At7Jm5dq/p0Ts5/ieAq8hmZF4pqVTZeWSLPb0P+Dz7nv7/zSUC0mSFbwGIiAfJ1uLYCNwkaVon7THrFicSs+57EPhUmnq8nuwP8gq6N01+pyQNAkZHxP3AV4HDgEOAncDbckUPJUsCANNz8Y7l1rFniYApQNs06u9Ibb0WuI5sbQ2zQpxIzLrvNrLLT78nW1/jqxHxZ7Jp8pvSNPmfofQ0+V2pA36YVgL8HXBFRLwA3AGcoz3rgF8K/ETSr4Fnc/U7lrsW+KikFWTjOG2LQE0CVkn6HdkaNFfuR1vN2vFgu5mZFeIeiZmZFeK7tsx6maSrgBM7hK+MiHKWLzbrM3xpy8zMCvGlLTMzK8SJxMzMCnEiMTOzQpxIzMyskP8PPwIXu9UYQAkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(x='loan_status', y='loan_amnt', data=df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "id": "dfce34af",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>loan_status</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Charged Off</th>\n",
       "      <td>77673.0</td>\n",
       "      <td>15126.300967</td>\n",
       "      <td>8505.090557</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>8525.0</td>\n",
       "      <td>14000.0</td>\n",
       "      <td>20000.0</td>\n",
       "      <td>40000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fully Paid</th>\n",
       "      <td>318357.0</td>\n",
       "      <td>13866.878771</td>\n",
       "      <td>8302.319699</td>\n",
       "      <td>500.0</td>\n",
       "      <td>7500.0</td>\n",
       "      <td>12000.0</td>\n",
       "      <td>19225.0</td>\n",
       "      <td>40000.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                count          mean          std     min     25%      50%  \\\n",
       "loan_status                                                                 \n",
       "Charged Off   77673.0  15126.300967  8505.090557  1000.0  8525.0  14000.0   \n",
       "Fully Paid   318357.0  13866.878771  8302.319699   500.0  7500.0  12000.0   \n",
       "\n",
       "                 75%      max  \n",
       "loan_status                    \n",
       "Charged Off  20000.0  40000.0  \n",
       "Fully Paid   19225.0  40000.0  "
      ]
     },
     "execution_count": 190,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('loan_status')['loan_amnt'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1326e9fe",
   "metadata": {},
   "source": [
    "**Let's explore the Grade and SubGrade columns that LendingClub attributes to the loans. What are the unique possible grades and subgrades?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "id": "a72bdd4c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['A', 'B', 'C', 'D', 'E', 'F', 'G']"
      ]
     },
     "execution_count": 191,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sorted(df['grade'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "id": "4ae27fc2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['A1',\n",
       " 'A2',\n",
       " 'A3',\n",
       " 'A4',\n",
       " 'A5',\n",
       " 'B1',\n",
       " 'B2',\n",
       " 'B3',\n",
       " 'B4',\n",
       " 'B5',\n",
       " 'C1',\n",
       " 'C2',\n",
       " 'C3',\n",
       " 'C4',\n",
       " 'C5',\n",
       " 'D1',\n",
       " 'D2',\n",
       " 'D3',\n",
       " 'D4',\n",
       " 'D5',\n",
       " 'E1',\n",
       " 'E2',\n",
       " 'E3',\n",
       " 'E4',\n",
       " 'E5',\n",
       " 'F1',\n",
       " 'F2',\n",
       " 'F3',\n",
       " 'F4',\n",
       " 'F5',\n",
       " 'G1',\n",
       " 'G2',\n",
       " 'G3',\n",
       " 'G4',\n",
       " 'G5']"
      ]
     },
     "execution_count": 192,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sorted(df['sub_grade'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "id": "ce9cf5c4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='grade', ylabel='count'>"
      ]
     },
     "execution_count": 193,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(data=df, x='grade', hue='loan_status')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 194,
   "id": "5473a634",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='sub_grade', ylabel='count'>"
      ]
     },
     "execution_count": 194,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12,4))\n",
    "sub_grade_ordered = sorted(df['sub_grade'].unique())\n",
    "sns.countplot(data=df, x='sub_grade', palette='coolwarm', order=sub_grade_ordered)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 195,
   "id": "955c3dd2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='sub_grade', ylabel='count'>"
      ]
     },
     "execution_count": 195,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12,4))\n",
    "sub_grade_ordered = sorted(df['sub_grade'].unique())\n",
    "sns.countplot(data=df, x='sub_grade', palette='coolwarm', order=sub_grade_ordered, hue='loan_status')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe6f9423",
   "metadata": {},
   "source": [
    "*Note: People from lower grades are finding it difficult to default the loan!*\n",
    "\n",
    "*Especially, F and G subgrades don't get paid back that often*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 196,
   "id": "fc605c2b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='sub_grade', ylabel='count'>"
      ]
     },
     "execution_count": 196,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fng = df[(df['grade']=='F') | (df['grade']=='G')]\n",
    "\n",
    "plt.figure(figsize=(12,4))\n",
    "sub_grade_ordered = sorted(fng['sub_grade'].unique())\n",
    "sns.countplot(data=fng, x='sub_grade', palette='coolwarm', order=sub_grade_ordered, hue='loan_status')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a0e3e538",
   "metadata": {},
   "source": [
    "**Let's create a new column called 'loan_repaid' to make our life easier. It will contain a 1 if the loan status was \"Fully Paid\" and a 0 if it was \"Charged Off\".**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "id": "725e58e5",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['loan_repaid'] = df['loan_status']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 198,
   "id": "4357fa13",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['loan_repaid'] = df['loan_repaid'].apply(lambda x: 1 if x.strip()=='Fully Paid' else 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 199,
   "id": "18a47ee7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>loan_status</th>\n",
       "      <th>loan_repaid</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Fully Paid</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Fully Paid</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Fully Paid</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Fully Paid</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Charged Off</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>396025</th>\n",
       "      <td>Fully Paid</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>396026</th>\n",
       "      <td>Fully Paid</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>396027</th>\n",
       "      <td>Fully Paid</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>396028</th>\n",
       "      <td>Fully Paid</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>396029</th>\n",
       "      <td>Fully Paid</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>396030 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        loan_status  loan_repaid\n",
       "0        Fully Paid            1\n",
       "1        Fully Paid            1\n",
       "2        Fully Paid            1\n",
       "3        Fully Paid            1\n",
       "4       Charged Off            0\n",
       "...             ...          ...\n",
       "396025   Fully Paid            1\n",
       "396026   Fully Paid            1\n",
       "396027   Fully Paid            1\n",
       "396028   Fully Paid            1\n",
       "396029   Fully Paid            1\n",
       "\n",
       "[396030 rows x 2 columns]"
      ]
     },
     "execution_count": 199,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[:, ['loan_status', 'loan_repaid']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 200,
   "id": "9c6bc346",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 200,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.corr()['loan_repaid'].sort_values().drop('loan_repaid').plot(kind = 'bar')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0673c884",
   "metadata": {},
   "source": [
    "---\n",
    "---\n",
    "# Section 2: Data PreProcessing\n",
    "\n",
    "**Section Goals: Remove or fill any missing data. Remove unnecessary or repetitive features. Convert categorical string features to dummy variables.**\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 201,
   "id": "b732297b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>loan_amnt</th>\n",
       "      <th>term</th>\n",
       "      <th>int_rate</th>\n",
       "      <th>installment</th>\n",
       "      <th>grade</th>\n",
       "      <th>sub_grade</th>\n",
       "      <th>emp_title</th>\n",
       "      <th>emp_length</th>\n",
       "      <th>home_ownership</th>\n",
       "      <th>annual_inc</th>\n",
       "      <th>...</th>\n",
       "      <th>pub_rec</th>\n",
       "      <th>revol_bal</th>\n",
       "      <th>revol_util</th>\n",
       "      <th>total_acc</th>\n",
       "      <th>initial_list_status</th>\n",
       "      <th>application_type</th>\n",
       "      <th>mort_acc</th>\n",
       "      <th>pub_rec_bankruptcies</th>\n",
       "      <th>address</th>\n",
       "      <th>loan_repaid</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>36 months</td>\n",
       "      <td>11.44</td>\n",
       "      <td>329.48</td>\n",
       "      <td>B</td>\n",
       "      <td>B4</td>\n",
       "      <td>Marketing</td>\n",
       "      <td>10+ years</td>\n",
       "      <td>RENT</td>\n",
       "      <td>117000.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>36369.0</td>\n",
       "      <td>41.8</td>\n",
       "      <td>25.0</td>\n",
       "      <td>w</td>\n",
       "      <td>INDIVIDUAL</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0174 Michelle Gateway\\nMendozaberg, OK 22690</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8000.0</td>\n",
       "      <td>36 months</td>\n",
       "      <td>11.99</td>\n",
       "      <td>265.68</td>\n",
       "      <td>B</td>\n",
       "      <td>B5</td>\n",
       "      <td>Credit analyst</td>\n",
       "      <td>4 years</td>\n",
       "      <td>MORTGAGE</td>\n",
       "      <td>65000.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>20131.0</td>\n",
       "      <td>53.3</td>\n",
       "      <td>27.0</td>\n",
       "      <td>f</td>\n",
       "      <td>INDIVIDUAL</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1076 Carney Fort Apt. 347\\nLoganmouth, SD 05113</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15600.0</td>\n",
       "      <td>36 months</td>\n",
       "      <td>10.49</td>\n",
       "      <td>506.97</td>\n",
       "      <td>B</td>\n",
       "      <td>B3</td>\n",
       "      <td>Statistician</td>\n",
       "      <td>&lt; 1 year</td>\n",
       "      <td>RENT</td>\n",
       "      <td>43057.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11987.0</td>\n",
       "      <td>92.2</td>\n",
       "      <td>26.0</td>\n",
       "      <td>f</td>\n",
       "      <td>INDIVIDUAL</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>87025 Mark Dale Apt. 269\\nNew Sabrina, WV 05113</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7200.0</td>\n",
       "      <td>36 months</td>\n",
       "      <td>6.49</td>\n",
       "      <td>220.65</td>\n",
       "      <td>A</td>\n",
       "      <td>A2</td>\n",
       "      <td>Client Advocate</td>\n",
       "      <td>6 years</td>\n",
       "      <td>RENT</td>\n",
       "      <td>54000.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5472.0</td>\n",
       "      <td>21.5</td>\n",
       "      <td>13.0</td>\n",
       "      <td>f</td>\n",
       "      <td>INDIVIDUAL</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>823 Reid Ford\\nDelacruzside, MA 00813</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24375.0</td>\n",
       "      <td>60 months</td>\n",
       "      <td>17.27</td>\n",
       "      <td>609.33</td>\n",
       "      <td>C</td>\n",
       "      <td>C5</td>\n",
       "      <td>Destiny Management Inc.</td>\n",
       "      <td>9 years</td>\n",
       "      <td>MORTGAGE</td>\n",
       "      <td>55000.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>24584.0</td>\n",
       "      <td>69.8</td>\n",
       "      <td>43.0</td>\n",
       "      <td>f</td>\n",
       "      <td>INDIVIDUAL</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>679 Luna Roads\\nGreggshire, VA 11650</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   loan_amnt        term  int_rate  installment grade sub_grade  \\\n",
       "0    10000.0   36 months     11.44       329.48     B        B4   \n",
       "1     8000.0   36 months     11.99       265.68     B        B5   \n",
       "2    15600.0   36 months     10.49       506.97     B        B3   \n",
       "3     7200.0   36 months      6.49       220.65     A        A2   \n",
       "4    24375.0   60 months     17.27       609.33     C        C5   \n",
       "\n",
       "                 emp_title emp_length home_ownership  annual_inc  ... pub_rec  \\\n",
       "0                Marketing  10+ years           RENT    117000.0  ...     0.0   \n",
       "1          Credit analyst     4 years       MORTGAGE     65000.0  ...     0.0   \n",
       "2             Statistician   < 1 year           RENT     43057.0  ...     0.0   \n",
       "3          Client Advocate    6 years           RENT     54000.0  ...     0.0   \n",
       "4  Destiny Management Inc.    9 years       MORTGAGE     55000.0  ...     0.0   \n",
       "\n",
       "  revol_bal revol_util total_acc initial_list_status  application_type  \\\n",
       "0   36369.0       41.8      25.0                   w        INDIVIDUAL   \n",
       "1   20131.0       53.3      27.0                   f        INDIVIDUAL   \n",
       "2   11987.0       92.2      26.0                   f        INDIVIDUAL   \n",
       "3    5472.0       21.5      13.0                   f        INDIVIDUAL   \n",
       "4   24584.0       69.8      43.0                   f        INDIVIDUAL   \n",
       "\n",
       "  mort_acc  pub_rec_bankruptcies  \\\n",
       "0      0.0                   0.0   \n",
       "1      3.0                   0.0   \n",
       "2      0.0                   0.0   \n",
       "3      0.0                   0.0   \n",
       "4      1.0                   0.0   \n",
       "\n",
       "                                           address  loan_repaid  \n",
       "0     0174 Michelle Gateway\\nMendozaberg, OK 22690            1  \n",
       "1  1076 Carney Fort Apt. 347\\nLoganmouth, SD 05113            1  \n",
       "2  87025 Mark Dale Apt. 269\\nNew Sabrina, WV 05113            1  \n",
       "3            823 Reid Ford\\nDelacruzside, MA 00813            1  \n",
       "4             679 Luna Roads\\nGreggshire, VA 11650            0  \n",
       "\n",
       "[5 rows x 28 columns]"
      ]
     },
     "execution_count": 201,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4bd63a00",
   "metadata": {},
   "source": [
    "# Missing Data\n",
    "\n",
    "**Let's explore this missing data columns. We will use a variety of factors to decide whether or not they would be useful, to see if we should keep, discard, or fill in the missing data.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 202,
   "id": "c87bb94f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "396030"
      ]
     },
     "execution_count": 202,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 203,
   "id": "12f4a4dc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "loan_amnt                   0\n",
       "term                        0\n",
       "int_rate                    0\n",
       "installment                 0\n",
       "grade                       0\n",
       "sub_grade                   0\n",
       "emp_title               22927\n",
       "emp_length              18301\n",
       "home_ownership              0\n",
       "annual_inc                  0\n",
       "verification_status         0\n",
       "issue_d                     0\n",
       "loan_status                 0\n",
       "purpose                     0\n",
       "title                    1755\n",
       "dti                         0\n",
       "earliest_cr_line            0\n",
       "open_acc                    0\n",
       "pub_rec                     0\n",
       "revol_bal                   0\n",
       "revol_util                276\n",
       "total_acc                   0\n",
       "initial_list_status         0\n",
       "application_type            0\n",
       "mort_acc                37795\n",
       "pub_rec_bankruptcies      535\n",
       "address                     0\n",
       "loan_repaid                 0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 203,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 204,
   "id": "e96fbc2a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "loan_amnt               0.000000\n",
       "term                    0.000000\n",
       "int_rate                0.000000\n",
       "installment             0.000000\n",
       "grade                   0.000000\n",
       "sub_grade               0.000000\n",
       "emp_title               5.789208\n",
       "emp_length              4.621115\n",
       "home_ownership          0.000000\n",
       "annual_inc              0.000000\n",
       "verification_status     0.000000\n",
       "issue_d                 0.000000\n",
       "loan_status             0.000000\n",
       "purpose                 0.000000\n",
       "title                   0.443148\n",
       "dti                     0.000000\n",
       "earliest_cr_line        0.000000\n",
       "open_acc                0.000000\n",
       "pub_rec                 0.000000\n",
       "revol_bal               0.000000\n",
       "revol_util              0.069692\n",
       "total_acc               0.000000\n",
       "initial_list_status     0.000000\n",
       "application_type        0.000000\n",
       "mort_acc                9.543469\n",
       "pub_rec_bankruptcies    0.135091\n",
       "address                 0.000000\n",
       "loan_repaid             0.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 204,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(df.isnull().sum()/len(df))*100 #in percentage"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "id": "3ecda4a6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The job title supplied by the Borrower when applying for the loan.*\n",
      "Employment length in years. Possible values are between 0 and 10 where 0 means less than one year and 10 means ten or more years. \n"
     ]
    }
   ],
   "source": [
    "feat_info('emp_title')\n",
    "feat_info('emp_length')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "id": "9f5a7889",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "173105"
      ]
     },
     "execution_count": 206,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['emp_title'].nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 207,
   "id": "1917a729",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Teacher                    4389\n",
       "Manager                    4250\n",
       "Registered Nurse           1856\n",
       "RN                         1846\n",
       "Supervisor                 1830\n",
       "                           ... \n",
       "Postman                       1\n",
       "McCarthy & Holthus, LLC       1\n",
       "jp flooring                   1\n",
       "Histology Technologist        1\n",
       "Gracon Services, Inc          1\n",
       "Name: emp_title, Length: 173105, dtype: int64"
      ]
     },
     "execution_count": 207,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['emp_title'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f92decc6",
   "metadata": {},
   "source": [
    "**Realistically there are too many unique job titles to try to convert this to a dummy variable feature. Let's remove that emp_title column.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 208,
   "id": "2e816683",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop('emp_title', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 209,
   "id": "dce3d63d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['1 year',\n",
       " '10+ years',\n",
       " '2 years',\n",
       " '3 years',\n",
       " '4 years',\n",
       " '5 years',\n",
       " '6 years',\n",
       " '7 years',\n",
       " '8 years',\n",
       " '9 years',\n",
       " '< 1 year']"
      ]
     },
     "execution_count": 209,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sorted(df['emp_length'].dropna().unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "id": "98db1d95",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='emp_length', ylabel='count'>"
      ]
     },
     "execution_count": 210,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "sorted_emp_length = sorted(df['emp_length'].dropna().unique())\n",
    "sns.countplot(x='emp_length', data=df, order=sorted_emp_length)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 211,
   "id": "00c290a9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='emp_length', ylabel='count'>"
      ]
     },
     "execution_count": 211,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAncAAAF0CAYAAABBth9RAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAs2ElEQVR4nO3de5yXZZ3/8ddHIFEQj3iCtmENUxQbZTysrXig1E6ou5q4mlqE5arZyV1t92d2cPPQ2bWDRh6KPIQRmFtpIpJnBsUAySBBxRDRlKREBT+/P+4bGmhmGGBmvjM3r+fj8X3M/b3u677u65ovzLznuk+RmUiSJKkaNqt1ByRJktR+DHeSJEkVYriTJEmqEMOdJElShRjuJEmSKsRwJ0mSVCE9a92BrmKHHXbIurq6WndDkiRpnaZPn/58ZvZvbp3hrlRXV0djY2OtuyFJkrROEfFkS+s8LCtJklQhhjtJkqQKMdxJkiRViOfcSZKkVr3++ussXLiQ5cuX17orm5zevXszcOBAevXq1eZtDHeSJKlVCxcuZKuttqKuro6IqHV3NhmZyQsvvMDChQsZNGhQm7fzsKwkSWrV8uXL2X777Q12nSwi2H777dd7xtRwJ0mS1slgVxsb8n033EmSJFWI4U6SJHW6vn371mzfU6ZM4b777mu3el2N4U6SJG1SDHeSJEkdJDM577zz2HvvvRk6dCg33XQTAMuWLWPEiBHst99+DB06lIkTJwKwYMEC9txzT8aMGcNee+3FkUceySuvvNJi+9/61rcYMmQI++yzD6NGjWLBggV897vf5etf/zr19fX85je/4dZbb+XAAw9k33335Z3vfCeLFy9utt7pp5/O+PHjV7e9avZx0aJFDB8+nPr6evbee29+85vfdOB3bN28FYokSaqZn/70p8yYMYNHH32U559/nv3335/hw4fTv39/JkyYQL9+/Xj++ec56KCDGDlyJABz587lhhtu4Oqrr+YDH/gAt9xyC6ecckqz7V9yySXMnz+fzTffnJdeeoltttmGj33sY/Tt25fPfOYzALz44os88MADRATf//73ueyyy/jqV7/6d/XGjh3b7D5+/OMfc9RRR/Ff//VfrFy5kr/+9a8d8J1qO8OdJEmqmXvuuYeTTjqJHj16sNNOO3HooYcybdo03v3ud/PZz36WqVOnstlmm/HMM8+wePFiAAYNGkR9fT0Aw4YNY8GCBS22v88++3DyySdz7LHHcuyxxzZbZ+HChZx44oksWrSI1157bb3uKQew//778+EPf5jXX3+dY489dnXfasVwt4kYdt717d7m9MtPbfc2JUmblsxstnzcuHEsWbKE6dOn06tXL+rq6lbf723zzTdfXa9Hjx6tHpa97bbbmDp1KpMmTeKLX/wis2fP/rs655xzDp/61KcYOXIkU6ZM4aKLLmq2rZ49e/LGG2+s7vdrr70GwPDhw5k6dSq33XYbH/zgBznvvPM49dTa/Y70nDtJklQzw4cP56abbmLlypUsWbKEqVOncsABB7B06VJ23HFHevXqxV133cWTTz653m2/8cYbPP300xx++OFcdtllvPTSSyxbtoytttqKl19+eXW9pUuXMmDAAACuu+661eVr16urq2P69OkATJw4kddffx2AJ598kh133JExY8YwevRoHn744Q36XrQXw50kSaqZ4447jn322Ye3v/3tHHHEEVx22WXsvPPOnHzyyTQ2NtLQ0MC4cePYY4891rvtlStXcsoppzB06FD23XdfPvnJT7LNNtvw/ve/nwkTJqy+UOKiiy7ihBNO4JBDDmGHHXZYvf3a9caMGcPdd9/NAQccwIMPPkifPn2A4qra+vp69t13X2655RbOPffcdvv+bIhoaTp0U9PQ0JCNjY217kaH8bCsJGlDzZkzhz333LPW3dhkNff9j4jpmdnQXH1n7iRJkiqkw8JdRPwgIp6LiFlNyraLiDsiYm75ddsm6y6IiHkR8XhEHNWkfFhEzCzXfSvKh6xFxOYRcVNZ/mBE1DXZ5rRyH3Mj4rSOGqMkSeoazjrrLOrr69d4XXPNNbXuVk105NWy1wL/CzQ9Hng+cGdmXhIR55fv/zMihgCjgL2AXYFfR8TumbkS+A5wBvAA8H/A0cAvgNHAi5n51ogYBVwKnBgR2wGfAxqABKZHxKTMfLEDxypJkmroyiuvrHUXuowOm7nLzKnAn9YqPgZYdRnKdcCxTcpvzMxXM3M+MA84ICJ2Afpl5v1ZnBx4/VrbrGprPDCinNU7CrgjM/9UBro7KAKhJElS5XX2OXc7ZeYigPLrjmX5AODpJvUWlmUDyuW1y9fYJjNXAEuB7VtpS5IkqfK6ygUV0UxZtlK+odusudOIMyKiMSIalyxZ0qaOSpIkdWWdHe4Wl4daKb8+V5YvBN7cpN5A4I9l+cBmytfYJiJ6AltTHAZuqa2/k5lXZWZDZjb0799/I4YlSZI6Uo8ePda4WKK1R45de+21nH322QBcdNFFfOUrX2nzfk4//fTVjzfbb7/9uP/++1utf/DBB7fYzvjx49u83/bU2Y8fmwScBlxSfp3YpPzHEfE1igsqBgMPZebKiHg5Ig4CHgROBa5Yq637geOByZmZEfEr4H+aXIl7JHBBxw9NkqRNQ3vfO7Ut903dYostmDFjRrvutyWXX345xx9/PLfffjsf/ehH+e1vf9ti3fvuu69T+rQ+OvJWKDdQBK+3RcTCiBhNEereFRFzgXeV78nM2cDNwGPAL4GzyitlAc4Evk9xkcUfKK6UBRgLbB8R84BPUVx5S2b+CfgiMK18faEskyRJFVJXV8fzzz8PQGNjI4cddliLdf/whz+w3377rX4/d+5chg0b1mr7w4cPZ968eSxbtowRI0aw3377MXToUCZOnLi6Tt++fYHiWbNnn302Q4YM4b3vfS/PPfdcS812uA6bucvMk1pYNaKF+hcDFzdT3gjs3Uz5cuCEFtr6AfCDNndWkiR1aa+88gr19fUADBo0iAkTJqzX9rvtthtbb701M2bMWH0PvNNPP73VbW699VaGDh1K7969mTBhAv369eP555/noIMOYuTIkZS33gVgwoQJPP7448ycOZPFixczZMgQPvzhD6/vMNtFZx+WlSRJWm/tcVj2Ix/5CNdccw1f+9rXuOmmm3jooYearXfeeefxpS99if79+zN27Fgyk89+9rNMnTqVzTbbjGeeeYbFixez8847r95m6tSpnHTSSfTo0YNdd92VI444YqP6ujEMd5IkqVvq2bMnb7zxBgDLly9fZ/1//dd/5fOf/zxHHHEEw4YNY/vtt2+23qpz7la59tprWbJkCdOnT6dXr17U1dU1u7+mM3m11FVuhSJJkrRe6urqmD59OgC33HLLOuv37t2bo446ijPPPJMPfehDbd7P0qVL2XHHHenVqxd33XUXTz755N/VGT58ODfeeCMrV65k0aJF3HXXXW0fSDsz3EmSpG7pc5/7HOeeey6HHHIIPXr0aNM2J598MhHBkUce2eb9nHzyyTQ2NtLQ0MC4cePYY489/q7Occcdx+DBgxk6dChnnnkmhx56aJvbb29RPNVLDQ0N2djYWOtudJj2vmwd2nbpuiSp+5szZw577rlnrbvRLr7yla+wdOlSvvjFL9a6K23W3Pc/IqZnZkNz9T3nTpIkbRKOO+44/vCHPzB58uRad6VDGe4kSdImYX1vn9Jdec6dJElShRjuJEmSKsRwJ0mSVCGGO0mSpAox3EmSpC7v2WefZdSoUey2224MGTKE97znPfz+979nypQpvO9976tp304//XTGjx//d+WZyZe+9CUGDx7M7rvvzuGHH87s2bNXr//JT37CnnvuyeGHHw7ASSedxD777MPXv/71jeqPV8tKkqT18tQXhrZre/9w4cxW12cmxx13HKeddho33ngjADNmzGDx4sUbve8VK1bQs2fHxKErr7yS++67j0cffZQtt9yS22+/nZEjRzJ79mx69+7N2LFj+fa3v83hhx/Os88+y3333dfs0y/Wl+FOkiR1aXfddRe9evXiYx/72Oqy+vp6AKZMmcKyZcs4/vjjmTVrFsOGDeNHP/oREcEXvvAFbr31Vl555RUOPvhgvve97xERHHbYYRx88MHce++9jBw5kuHDhzN69Gj69OnDP//zP/OLX/yCWbNmsXLlSs4//3ymTJnCq6++yllnncVHP/pRMpNzzjmHyZMnM2jQIFp6IMSll17KlClT2HLLLQE48sgjOfjggxk3bhzPPPMM99xzD/Pnz2fkyJH86le/4rnnnqO+vp4rrriCQw45ZIO/Xx6WlSRJXdqq0NaSRx55hG984xs89thjPPHEE9x7770AnH322UybNo1Zs2bxyiuv8POf/3z1Ni+99BJ33303n/70p/nQhz7Ed7/7Xe6///41HmM2duxYtt56a6ZNm8a0adO4+uqrmT9/PhMmTODxxx9n5syZXH311dx3331/16c///nP/OUvf2G33XZbo7yhoYHZs2dz4YUXrn6c2eWXX86kSZPYbbfdmDFjxkYFOzDcSZKkbu6AAw5g4MCBbLbZZtTX17NgwQKgmPE78MADGTp0KJMnT17jfLcTTzwRKELeyy+/zMEHHwzAv/3bv62uc/vtt3P99ddTX1/PgQceyAsvvMDcuXOZOnUqJ510Ej169GDXXXfliCOOaHNfM5OIaIdRt8xwJ0mSurS99tqL6dOnt7h+8803X73co0cPVqxYwfLly/n3f/93xo8fz8yZMxkzZgzLly9fXa9Pnz4ALR5SXbXuiiuuYMaMGcyYMYP58+dz5JFHAqwzoPXr148+ffrwxBNPrFH+8MMPM2TIkFa33ViGO0mS1KUdccQRvPrqq1x99dWry6ZNm8bdd9/d4jargtwOO+zAsmXLmr2aFWDbbbdlq6224oEHHgBYfcEGwFFHHcV3vvMdXn/9dQB+//vf85e//IXhw4dz4403snLlShYtWsRdd93VbNvnnXceH//4x3nllVcA+PWvf80999yzxuxgR/CCCkmS1KVFBBMmTOATn/gEl1xyCb1796auro5vfOMbPPPMM81us8022zBmzBiGDh1KXV0d+++/f4vtjx07ljFjxtCnTx8OO+wwtt56awA+8pGPsGDBAvbbbz8yk/79+/Ozn/2M4447jsmTJzN06FB23313Dj300GbbPeecc3jxxRcZOnQoPXr0YOedd2bixIlsscUWG/9NaUW0Nh25KWloaMjGxsZad6PDDDvv+nZvc/rlp7Z7m5KkrmfOnDnsueeete5Gh1m2bBl9+/YF4JJLLmHRokV885vfrHGv/qa5739ETM/MhubqO3MnSZI2abfddhtf/vKXWbFiBW95y1u49tpra92ljWK4kyRJm7QTTzxx9dWzVeAFFZIkSRViuJMkSevkOfq1sSHfd8OdJElqVe/evXnhhRcMeJ0sM3nhhRfo3bv3em3nOXeSJKlVAwcOZOHChSxZsqTWXdnk9O7dm4EDB67XNoY7SZLUql69ejFo0KBad0Nt5GFZSZKkCjHcSZIkVYjhTpIkqUIMd5IkSRViuJMkSaoQw50kSVKFGO4kSZIqxHAnSZJUIYY7SZKkCjHcSZIkVYjhTpIkqUIMd5IkSRViuJMkSaoQw50kSVKFGO4kSZIqxHAnSZJUIYY7SZKkCjHcSZIkVYjhTpIkqUIMd5IkSRViuJMkSaoQw50kSVKFGO4kSZIqxHAnSZJUIYY7SZKkCqlJuIuIT0bE7IiYFRE3RETviNguIu6IiLnl122b1L8gIuZFxOMRcVST8mERMbNc962IiLJ884i4qSx/MCLqajBMSZKkTtfp4S4iBgAfBxoyc2+gBzAKOB+4MzMHA3eW74mIIeX6vYCjgW9HRI+yue8AZwCDy9fRZflo4MXMfCvwdeDSThiaJElSzdXqsGxPYIuI6AlsCfwROAa4rlx/HXBsuXwMcGNmvpqZ84F5wAERsQvQLzPvz8wErl9rm1VtjQdGrJrVkyRJqrJOD3eZ+QzwFeApYBGwNDNvB3bKzEVlnUXAjuUmA4CnmzSxsCwbUC6vXb7GNpm5AlgKbL92XyLijIhojIjGJUuWtM8AJUmSaqgWh2W3pZhZGwTsCvSJiFNa26SZsmylvLVt1izIvCozGzKzoX///q13XJIkqRuoxWHZdwLzM3NJZr4O/BQ4GFhcHmql/PpcWX8h8OYm2w+kOIy7sFxeu3yNbcpDv1sDf+qQ0UiSJHUhtQh3TwEHRcSW5XlwI4A5wCTgtLLOacDEcnkSMKq8AnYQxYUTD5WHbl+OiIPKdk5da5tVbR0PTC7Py5MkSaq0np29w8x8MCLGAw8DK4BHgKuAvsDNETGaIgCeUNafHRE3A4+V9c/KzJVlc2cC1wJbAL8oXwBjgR9GxDyKGbtRnTA0SZKkmuv0cAeQmZ8DPrdW8asUs3jN1b8YuLiZ8kZg72bKl1OGQ0mSpE2JT6iQJEmqEMOdJElShRjuJEmSKsRwJ0mSVCGGO0mSpAox3EmSJFWI4U6SJKlCDHeSJEkVYriTJEmqEMOdJElShRjuJEmSKsRwJ0mSVCGGO0mSpAox3EmSJFWI4U6SJKlCDHeSJEkVYriTJEmqEMOdJElShRjuJEmSKsRwJ0mSVCGGO0mSpAox3EmSJFWI4U6SJKlCDHeSJEkVYriTJEmqEMOdJElShRjuJEmSKsRwJ0mSVCGGO0mSpAox3EmSJFWI4U6SJKlCDHeSJEkVYriTJEmqEMOdJElShRjuJEmSKsRwJ0mSVCGGO0mSpAox3EmSJFWI4U6SJKlCDHeSJEkVYriTJEmqEMOdJElShRjuJEmSKsRwJ0mSVCGGO0mSpAox3EmSJFWI4U6SJKlCDHeSJEkVYriTJEmqEMOdJElShRjuJEmSKqQm4S4itomI8RHxu4iYExH/FBHbRcQdETG3/Lptk/oXRMS8iHg8Io5qUj4sImaW674VEVGWbx4RN5XlD0ZEXQ2GKUmS1OlqNXP3TeCXmbkH8HZgDnA+cGdmDgbuLN8TEUOAUcBewNHAtyOiR9nOd4AzgMHl6+iyfDTwYma+Ffg6cGlnDEqSJKnWOj3cRUQ/YDgwFiAzX8vMl4BjgOvKatcBx5bLxwA3ZuarmTkfmAccEBG7AP0y8/7MTOD6tbZZ1dZ4YMSqWT1JkqQqq8XM3T8CS4BrIuKRiPh+RPQBdsrMRQDl1x3L+gOAp5tsv7AsG1Aur12+xjaZuQJYCmzfMcORJEnqOmoR7noC+wHfycx9gb9QHoJtQXMzbtlKeWvbrNlwxBkR0RgRjUuWLGm915IkSd1ALcLdQmBhZj5Yvh9PEfYWl4daKb8+16T+m5tsPxD4Y1k+sJnyNbaJiJ7A1sCf1u5IZl6VmQ2Z2dC/f/92GJokSVJtdXq4y8xngacj4m1l0QjgMWAScFpZdhowsVyeBIwqr4AdRHHhxEPloduXI+Kg8ny6U9faZlVbxwOTy/PyJEmSKq1njfZ7DjAuIt4EPAF8iCJo3hwRo4GngBMAMnN2RNxMEQBXAGdl5sqynTOBa4EtgF+ULygu1vhhRMyjmLEb1RmDkiRJqrWahLvMnAE0NLNqRAv1LwYubqa8Edi7mfLllOFQkiRpU+ITKiRJkiqkTeEuIu5sS5kkSZJqq9XDshHRG9gS2KF8HNiqW4z0A3bt4L5JkiRpPa3rnLuPAp+gCHLT+Vu4+zNwZcd1S5IkSRui1XCXmd8EvhkR52TmFZ3UJ0mSJG2gNl0tm5lXRMTBQF3TbTLz+g7qlyRJkjZAm8JdRPwQ2A2YAay6x1wChjtJkqQupK33uWsAhviUB0mSpK6trfe5mwXs3JEdkSRJ0sZr68zdDsBjEfEQ8Oqqwswc2SG9kiRJ0gZpa7i7qCM7IUmSpPbR1qtl7+7ojkiSJGnjtfVq2Zcpro4FeBPQC/hLZvbrqI5JkiRp/bV15m6rpu8j4ljggI7okCRJkjZcW6+WXUNm/gw4on27IkmSpI3V1sOy/9Lk7WYU973znneSJEldTFuvln1/k+UVwALgmHbvjSRJkjZKW8+5+1BHd0SSJEkbr03n3EXEwIiYEBHPRcTiiLglIgZ2dOckSZK0ftp6QcU1wCRgV2AAcGtZJkmSpC6kreGuf2Zek5kryte1QP8O7JckSZI2QFvD3fMRcUpE9ChfpwAvdGTHJEmStP7aGu4+DHwAeBZYBBwPeJGFJElSF9PWW6F8ETgtM18EiIjtgK9QhD5JkiR1EW2dudtnVbADyMw/Aft2TJckSZK0odoa7jaLiG1XvSln7to66ydJkqRO0taA9lXgvogYT/HYsQ8AF3dYryRJkrRB2vqEiusjohE4AgjgXzLzsQ7tmSRJktZbmw+tlmHOQCdJktSFtfWcO0mSJHUDhjtJkqQKMdxJkiRViOFOkiSpQgx3kiRJFWK4kyRJqhDDnSRJUoUY7iRJkirEcCdJklQhhjtJkqQKMdxJkiRViOFOkiSpQgx3kiRJFWK4kyRJqhDDnSRJUoUY7iRJkirEcCdJklQhhjtJkqQKMdxJkiRViOFOkiSpQgx3kiRJFWK4kyRJqhDDnSRJUoXULNxFRI+IeCQifl6+3y4i7oiIueXXbZvUvSAi5kXE4xFxVJPyYRExs1z3rYiIsnzziLipLH8wIuo6fYCSJEk1UMuZu3OBOU3enw/cmZmDgTvL90TEEGAUsBdwNPDtiOhRbvMd4AxgcPk6uiwfDbyYmW8Fvg5c2rFDkSRJ6hpqEu4iYiDwXuD7TYqPAa4rl68Djm1SfmNmvpqZ84F5wAERsQvQLzPvz8wErl9rm1VtjQdGrJrVkyRJqrJazdx9A/gP4I0mZTtl5iKA8uuOZfkA4Okm9RaWZQPK5bXL19gmM1cAS4Ht1+5ERJwREY0R0bhkyZKNHJIkSVLtdXq4i4j3Ac9l5vS2btJMWbZS3to2axZkXpWZDZnZ0L9//zZ2R5IkqevqWYN9vgMYGRHvAXoD/SLiR8DiiNglMxeVh1yfK+svBN7cZPuBwB/L8oHNlDfdZmFE9AS2Bv7UUQOSJEnqKjp95i4zL8jMgZlZR3GhxOTMPAWYBJxWVjsNmFguTwJGlVfADqK4cOKh8tDtyxFxUHk+3alrbbOqrePLffzdzJ0kSVLV1GLmriWXADdHxGjgKeAEgMycHRE3A48BK4CzMnNluc2ZwLXAFsAvyhfAWOCHETGPYsZuVGcNQpIkqZZqGu4ycwowpVx+ARjRQr2LgYubKW8E9m6mfDllOJQkSdqU+IQKSZKkCjHcSZIkVYjhTpIkqUIMd5IkSRViuJMkSaoQw50kSVKFGO4kSZIqxHAnSZJUIYY7SZKkCjHcSZIkVYjhTpIkqUIMd5IkSRViuJMkSaoQw50kSVKFGO4kSZIqxHAnSZJUIYY7SZKkCjHcSZIkVYjhTpIkqUIMd5IkSRViuJMkSaoQw50kSVKFGO4kSZIqxHAnSZJUIYY7SZKkCjHcSZIkVYjhTpIkqUIMd5IkSRXSs9YdkNbXsPOub9f2pl9+aru2J0lSLTlzJ0mSVCGGO0mSpAox3EmSJFWI4U6SJKlCDHeSJEkVYriTJEmqEMOdJElShRjuJEmSKsRwJ0mSVCE+oUKqIZ+2IUlqb4Y7SZK0SavaH9oelpUkSaoQw50kSVKFGO4kSZIqxHAnSZJUIYY7SZKkCvFqWW2wp74wtF3b+4cLZ7Zre5IkbYqcuZMkSaoQw50kSVKFGO4kSZIqxHAnSZJUIV5QIUlNVO0xRJI2PZ0+cxcRb46IuyJiTkTMjohzy/LtIuKOiJhbft22yTYXRMS8iHg8Io5qUj4sImaW674VEVGWbx4RN5XlD0ZEXWePU5IkqRZqMXO3Avh0Zj4cEVsB0yPiDuB04M7MvCQizgfOB/4zIoYAo4C9gF2BX0fE7pm5EvgOcAbwAPB/wNHAL4DRwIuZ+daIGAVcCpy4IZ31r3hJktSddPrMXWYuysyHy+WXgTnAAOAY4Lqy2nXAseXyMcCNmflqZs4H5gEHRMQuQL/MvD8zE7h+rW1WtTUeGLFqVk+SJKnKanpBRXm4dF/gQWCnzFwERQAEdiyrDQCebrLZwrJsQLm8dvka22TmCmApsH0z+z8jIhojonHJkiXtNCpJkqTaqVm4i4i+wC3AJzLzz61VbaYsWylvbZs1CzKvysyGzGzo37//urosSZLU5dXkatmI6EUR7MZl5k/L4sURsUtmLioPuT5Xli8E3txk84HAH8vygc2UN91mYUT0BLYG/tQhg5HU7uemguenStKGqsXVsgGMBeZk5tearJoEnFYunwZMbFI+qrwCdhAwGHioPHT7ckQcVLZ56lrbrGrreGByeV6eJElSpdVi5u4dwAeBmRExoyz7LHAJcHNEjAaeAk4AyMzZEXEz8BjFlbZnlVfKApwJXAtsQXGV7C/K8rHADyNiHsWM3agOHpMkSVKX0OnhLjPvoflz4gBGtLDNxcDFzZQ3Ans3U76cMhxKkiRtSnz8mCRJUoUY7iRJkirEcCdJklQhhjtJkqQKqcl97iRJHa9Kz8au0likjubMnSRJUoUY7iRJkirEcCdJklQhhjtJkqQK8YIKSZI6SXtfGAJeHKK/58ydJElShRjuJEmSKsRwJ0mSVCGGO0mSpAox3EmSJFWI4U6SJKlCDHeSJEkVYriTJEmqEG9iLEmS1lt735DZmzG3H2fuJEmSKsRwJ0mSVCGGO0mSpAox3EmSJFWI4U6SJKlCDHeSJEkVYriTJEmqEMOdJElShRjuJEmSKsRwJ0mSVCGGO0mSpAox3EmSJFWI4U6SJKlCDHeSJEkVYriTJEmqkJ617oBUa099YWi7t/kPF85s9zYlSWoLw50kSVI7au9Jg/WdMDDcSRVS6x8okqTa85w7SZKkCjHcSZIkVYiHZTuZh80kSVJHMtxJUgfyamxJnc1wJ6lLcpZbkjaM59xJkiRViDN3kqQ2qdJsqmNpmbPc3Z8zd5IkSRViuJMkSaoQD8tKkqSa88ry9uPMnSRJUoUY7iRJkirEcCdJklQhhjtJkqQKqXS4i4ijI+LxiJgXEefXuj+SJEkdrbLhLiJ6AFcC7waGACdFxJDa9kqSJKljVTbcAQcA8zLzicx8DbgROKbGfZIkSepQVQ53A4Cnm7xfWJZJkiRVVmRmrfvQISLiBOCozPxI+f6DwAGZeU6TOmcAZ5Rv3wY83gld2wF4vhP20xmqMpaqjAMcS1dUlXGAY+mqqjKWqowDOmcsb8nM/s2tqPITKhYCb27yfiDwx6YVMvMq4KrO7FRENGZmQ2fus6NUZSxVGQc4lq6oKuMAx9JVVWUsVRkH1H4sVT4sOw0YHBGDIuJNwChgUo37JEmS1KEqO3OXmSsi4mzgV0AP4AeZObvG3ZIkSepQlQ13AJn5f8D/1bofa+nUw8AdrCpjqco4wLF0RVUZBziWrqoqY6nKOKDGY6nsBRWSJEmboiqfcydJkrTJMdxtgIj4QUQ8FxGzat2XlrTUx4jYLiLuiIi55ddta9XHjRURb46IuyJiTkTMjohza92nDRERvSPioYh4tBzH52vdp40VET0i4pGI+Hmt+7IxImJBRMyMiBkR0Vjr/myMiNgmIsZHxO/K/zP/VOs+ra+IeFv5Wax6/TkiPlHrfm2oiPhk+X9+VkTcEBG9a92nDRUR55bjmN2dP5O2iIgTynG+ERFd8upew92GuRY4urN3GhHrc47ktTTfx/OBOzNzMHBn+X7t/VwUEadvSB83xnqOD2AF8OnM3BM4CDirsx4xtwF9bc2rwBGZ+XagHjg6Ig5qx/Zb1M7jaOpcYE4Htd2sDhzL4ZlZ35m3NeigsXwT+GVm7gG8nU76fNpzLJn5ePlZ1APDgL8CE9qr/da092cSEQOAjwMNmbk3xYV/o9pzH63su73HsjcwhuLJUG8H3hcRg9tzH63su8OuHYiIN0VEn2ZWzQL+BZjaUftuoT8REW3KbYa7DZCZU4E/tbQ+IraKiPkR0at836+cAegVEbtFxC8jYnpE/CYi9ijrvD8iHixnO34dETuV5RdFxFURcTtwfTv08RjgunL5OuDYtrbZZHwjImJCk/fvioiflstHRsT9EfFwRPwkIvqW5RdGxLTyL7urIiLK8ikR8T8RcTdwbvkX0axyFqvV/ziZuSgzHy6XX6b4ZbXGU0g68rOIiL3KGbcZEfHbDf1hloVl5dte5WuNk2G7wzia9HUg8F7g+y2s7zZjacNYu8VYIqIfMBwYC5CZr2XmS91xLE2MAP6QmU9243H0BLaIIqBsyVr3Yu1GY9kTeCAz/5qZK4C7geO66ViIiD0j4qsUDzbYfe31mTknM1t96EFE/DAijmnyflxEjIziqMblUfw+/G1EfLRc3zci7ozid+fMVdtGRF0UM+3fBh5mzfv3tiwzfW3AC6gDZrWy/hrg2HL5DOCr5fKdwOBy+UBgcrm8LX+7wOUjTepfBEwHtmiPPgIvrfX+xWa2uwg4vZV2A/gd0L98/2Pg/RR35J4K9CnL/xO4sFzersn2PwTeXy5PAb7dZN1MYEC5vM16jvUpoF9nfRbAFcDJ5fKbNuQzatLHHsAMYBlwaWf+m2rPcZRtjKeYVTkM+Hk3H8t8ih+o04EzuutYKGaEH6KY0X+EInj36Y5jadLXHwBnd9fPpGzjXIr/80uAcd11LBTh7vfA9hQh9X7giu40FqAP8CHgHuDecn9brWObKRQzr82tOxT4Wbm8NcXPkp7luP+7LN8caAQGlev6leU7APMoftfWAW8AB63PeCp9K5Qa+z7wH8DPKP7BjIliFutg4CdRTFxB8eFC8QSNmyJiF4p/nPObtDUpM1/pyM5GxFCK0AWwM/Ba/O28iRGZ+cKqupmZEfFD4JSIuAb4J+BUisPAQ4B7y/G9ieI/OcDhEfEfFP/xtwNmA7eW625q0pV7gWsj4mbgp23se1/gFuATmfnnZqp01GdxP/BfUcxU/TQz57alv83JzJVAfURsA0yIiL0zc+1zOrv8OCLifcBzmTk9Ig5rpWqXH0vpHZn5x4jYEbgjIn6Xxax4dxtLT2A/4JzMfDAivklxSsb/64ZjIYob048ELmihSpcfRxTnOx9D8Yv9pbJfp2Tmj7rbWDJzTkRcCtxBEVYfpThtZm1deSyLgN8CH8nM323A9mvIzLsj4sryZ8e/ALdkcf/dI4F9IuL4surWwGCKp2r9T0QMpwhzA4CdyjpPZuYD69sBXxv2l0odrczclXUepUjvD5Xv+wGLWqg7BRhZLh8GTCmXLwI+0159pJhm3qVc3gV4vJntLqKVmbuyzq4UfzmdCVxWlr0fuKGZur2BxcCbm7R/UZNxN6xV/0DgC8DTwPbr6EcvihtVf6oWnwWwG8V5M09QnDfXHv+2PtfSZ97VxwF8meKH1ALgWYpzon7UHcfSwv+L7vq57AwsaPL+EOC27jiWsq1jgNvXUadLjwM4ARjb5P2pNDmK0Z3G0kwf/gf49+40FuBIiomGOcCFFM9tXdc2U2hh5q5c/5/AJ4EHgb3Kslsonnu/dt3Ty/33Kt8voPgdXsc6skZzL8+561jXAzdQTEWTxazS/Ig4AVafHPn2su7WwDPl8mkd2KdJTdo/DZi4IY1k5h8pzg/5b4pDPQAPAO+IiLcCRMSWEbE7RbgDeL78K+14WhARu2Xmg5l5IcVDl1s8vyCKP/PGAnMy82vr6HK7fxYR8Y/AE5n5LYrv6z7r6ENL7fQvZ+yIiC2Ad1Ic9u5W4yj7c0FmDszMOoqTwydn5indcSwR0Scitlq1TPHDv6Ur5Lv0WDLzWeDpiHhbWTQCeKw7jqV0UtnH1nT1cTwFHFT+nAyKz6Sli1y6+lgoZ6iIiH+gmKlq6fPpkmPJzNsz80Tgn4GlwMQozvOrW9+2mrgW+ETZ/qonZP0KODP+du7h7uXPl60pjnq8HhGHA2/ZiP0a7jZERNxAMQ38tohYGBGjW6g6juKcgab/yE8GRkfEoxSHJo8pyy+imJb+DUWo6ag+XgK8KyLmAu8q32+occDTmfkYQGYuofjr44aI+C1F2NsjixO3r6Y4n+5nFM/9bcnlUZxMOovi/L1HW6n7DuCDwBHxt1sjvKeVvrb3Z3EiMCsiZgB7sB4XvKxlF+Cu8ns2DbgjM1u6hUhXHsf66upj2Qm4p+zLQxQzXb9soW5XHwvAOcC48t9ZPcXsSnO69FgiYkuKn13rOm2jS48jMx+kOD/1YYqfjZvR8lMNuvRYSrdExGMUp9uclZkvtlCvS48lM1/IzG9mcUX2Z4GVa9eJiOMiYiHFKUm3RcSvWmhrMUVgv6ZJ8fcp/rB6uPw99z2K0ybGAQ1R3HLpZFr+A79NfEJFB4rimPoxmfnBWvelI0TE/wKPZObYWvdlXaryWVRlHOBYuqqqjKUq4wDH0l2Vf4jMBPbLzKWduW8vqOggEXEF8G6gpZmkbi0ipgN/AT5d676sS1U+i6qMAxxLV1WVsVRlHOBYuquIeCfFFd1f6+xgB87cSZIkVYrn3EmSJFWI4U6SJKlCDHeSJEkVYriTJEmqEMOdJHWAiFjWAW3WN72XYxQPUP9Me+9HUvdmuJOk7qOeTeA2EpI2juFO0iYlIk6JiIfKJ5p8LyJ6RMSyiLg0IqaXjxw6ICKmRMQTETGy3O70iJgYEb+MiMcj4nPrsc/zImJaRPw2Ij5fltVFxJyIuDoiZkfE7VE8fo6I2L+se39EXB4RsyLiTRTPXD6x7PuJZfNDmvT14+387ZLUDRnuJG0yImJPikcVvaN8vNBKikf99KF4KPkw4GXgSxSPuDqOIlCtckBZvx44ISIa2rDPI4HB5bb1wLCIGF6uHgxcmZl7AS8B/1qWXwN8LDP/qewjmfkaxQPNb8rM+sy8qay7B3BU2f7nVj2zUtKmyydUSNqUjACGAdMiAmAL4DngNWDVc2NnAq+WD/CeCdQ12f6OzHwBICJ+SvGQ8cZ17PPI8vVI+b4vRah7CpifmTPK8ulAXURsA2yVmfeV5T8G3tdK+7dl5qvAqxHxHMUzcReuo0+SKsxwJ2lTEsB1mXnBGoURn8m/Pa7nDeBVgMx8IyKa/pxc+5E+bXnETwBfzszvrbXPulX7Ka2kCJvRhjabWrsNf65LmzgPy0ralNwJHB8ROwJExHYR8Zb12P5d5TZbAMcC97Zhm18BH46IvuU+B6zaf3My80Xg5Yg4qCwa1WT1y8BW69FfSZsgw52kTUZmPgb8N3B7RPwWuAPYZT2auAf4ITADuCUz13VIlsy8neLQ6v3lYd7xrDugjQauioj7KWbyVj14/C6KCyiaXlAhSWuIvx2JkCS1JCJOBxoy8+xO2FffzFxWLp8P7JKZ53b0fiVVg+dmSFLX896IuIDiZ/STwOm17Y6k7sSZO0naQBGxPcV5fGsbseqqWknqbIY7SZKkCvGCCkmSpAox3EmSJFWI4U6SJKlCDHeSJEkVYriTJEmqkP8PPZf0prDoJm4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "sorted_emp_length = sorted(df['emp_length'].dropna().unique())\n",
    "sns.countplot(x='emp_length', data=df, order=sorted_emp_length, hue='loan_status')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c6063ab6",
   "metadata": {},
   "source": [
    "**This still doesn't really inform us if there is a strong relationship between employment length and being charged off, let's analyse percentage of charge offs per category. Essentially informing us what percent of people per employment category didn't pay back their loan.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 212,
   "id": "de7e26fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "emp_co = df[df['loan_status']==\"Charged Off\"].groupby(\"emp_length\").count()['loan_status']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 213,
   "id": "fdd45b23",
   "metadata": {},
   "outputs": [],
   "source": [
    "emp_fp = df[df['loan_status']==\"Fully Paid\"].groupby(\"emp_length\").count()['loan_status']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 214,
   "id": "27c273ad",
   "metadata": {},
   "outputs": [],
   "source": [
    "emp_len = emp_co/emp_fp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 215,
   "id": "60050a55",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "emp_length\n",
       "1 year       0.248649\n",
       "10+ years    0.225770\n",
       "2 years      0.239560\n",
       "3 years      0.242593\n",
       "4 years      0.238213\n",
       "5 years      0.237911\n",
       "6 years      0.233341\n",
       "7 years      0.241887\n",
       "8 years      0.249625\n",
       "9 years      0.250735\n",
       "< 1 year     0.260830\n",
       "Name: loan_status, dtype: float64"
      ]
     },
     "execution_count": 215,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "emp_len"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 216,
   "id": "4e1b700f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='emp_length'>"
      ]
     },
     "execution_count": 216,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "emp_len.plot(kind='bar')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12abfadf",
   "metadata": {},
   "source": [
    "**Charge off rates are extremely similar across all employment lengths. We will drop the emp_length column here.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 217,
   "id": "70d17e40",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop('emp_length', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 218,
   "id": "9626190e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>loan_amnt</th>\n",
       "      <th>term</th>\n",
       "      <th>int_rate</th>\n",
       "      <th>installment</th>\n",
       "      <th>grade</th>\n",
       "      <th>sub_grade</th>\n",
       "      <th>home_ownership</th>\n",
       "      <th>annual_inc</th>\n",
       "      <th>verification_status</th>\n",
       "      <th>issue_d</th>\n",
       "      <th>...</th>\n",
       "      <th>pub_rec</th>\n",
       "      <th>revol_bal</th>\n",
       "      <th>revol_util</th>\n",
       "      <th>total_acc</th>\n",
       "      <th>initial_list_status</th>\n",
       "      <th>application_type</th>\n",
       "      <th>mort_acc</th>\n",
       "      <th>pub_rec_bankruptcies</th>\n",
       "      <th>address</th>\n",
       "      <th>loan_repaid</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>36 months</td>\n",
       "      <td>11.44</td>\n",
       "      <td>329.48</td>\n",
       "      <td>B</td>\n",
       "      <td>B4</td>\n",
       "      <td>RENT</td>\n",
       "      <td>117000.0</td>\n",
       "      <td>Not Verified</td>\n",
       "      <td>Jan-2015</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>36369.0</td>\n",
       "      <td>41.8</td>\n",
       "      <td>25.0</td>\n",
       "      <td>w</td>\n",
       "      <td>INDIVIDUAL</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0174 Michelle Gateway\\nMendozaberg, OK 22690</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8000.0</td>\n",
       "      <td>36 months</td>\n",
       "      <td>11.99</td>\n",
       "      <td>265.68</td>\n",
       "      <td>B</td>\n",
       "      <td>B5</td>\n",
       "      <td>MORTGAGE</td>\n",
       "      <td>65000.0</td>\n",
       "      <td>Not Verified</td>\n",
       "      <td>Jan-2015</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>20131.0</td>\n",
       "      <td>53.3</td>\n",
       "      <td>27.0</td>\n",
       "      <td>f</td>\n",
       "      <td>INDIVIDUAL</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1076 Carney Fort Apt. 347\\nLoganmouth, SD 05113</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15600.0</td>\n",
       "      <td>36 months</td>\n",
       "      <td>10.49</td>\n",
       "      <td>506.97</td>\n",
       "      <td>B</td>\n",
       "      <td>B3</td>\n",
       "      <td>RENT</td>\n",
       "      <td>43057.0</td>\n",
       "      <td>Source Verified</td>\n",
       "      <td>Jan-2015</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11987.0</td>\n",
       "      <td>92.2</td>\n",
       "      <td>26.0</td>\n",
       "      <td>f</td>\n",
       "      <td>INDIVIDUAL</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>87025 Mark Dale Apt. 269\\nNew Sabrina, WV 05113</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7200.0</td>\n",
       "      <td>36 months</td>\n",
       "      <td>6.49</td>\n",
       "      <td>220.65</td>\n",
       "      <td>A</td>\n",
       "      <td>A2</td>\n",
       "      <td>RENT</td>\n",
       "      <td>54000.0</td>\n",
       "      <td>Not Verified</td>\n",
       "      <td>Nov-2014</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5472.0</td>\n",
       "      <td>21.5</td>\n",
       "      <td>13.0</td>\n",
       "      <td>f</td>\n",
       "      <td>INDIVIDUAL</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>823 Reid Ford\\nDelacruzside, MA 00813</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24375.0</td>\n",
       "      <td>60 months</td>\n",
       "      <td>17.27</td>\n",
       "      <td>609.33</td>\n",
       "      <td>C</td>\n",
       "      <td>C5</td>\n",
       "      <td>MORTGAGE</td>\n",
       "      <td>55000.0</td>\n",
       "      <td>Verified</td>\n",
       "      <td>Apr-2013</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>24584.0</td>\n",
       "      <td>69.8</td>\n",
       "      <td>43.0</td>\n",
       "      <td>f</td>\n",
       "      <td>INDIVIDUAL</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>679 Luna Roads\\nGreggshire, VA 11650</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   loan_amnt        term  int_rate  installment grade sub_grade  \\\n",
       "0    10000.0   36 months     11.44       329.48     B        B4   \n",
       "1     8000.0   36 months     11.99       265.68     B        B5   \n",
       "2    15600.0   36 months     10.49       506.97     B        B3   \n",
       "3     7200.0   36 months      6.49       220.65     A        A2   \n",
       "4    24375.0   60 months     17.27       609.33     C        C5   \n",
       "\n",
       "  home_ownership  annual_inc verification_status   issue_d  ... pub_rec  \\\n",
       "0           RENT    117000.0        Not Verified  Jan-2015  ...     0.0   \n",
       "1       MORTGAGE     65000.0        Not Verified  Jan-2015  ...     0.0   \n",
       "2           RENT     43057.0     Source Verified  Jan-2015  ...     0.0   \n",
       "3           RENT     54000.0        Not Verified  Nov-2014  ...     0.0   \n",
       "4       MORTGAGE     55000.0            Verified  Apr-2013  ...     0.0   \n",
       "\n",
       "  revol_bal revol_util  total_acc initial_list_status  application_type  \\\n",
       "0   36369.0       41.8       25.0                   w        INDIVIDUAL   \n",
       "1   20131.0       53.3       27.0                   f        INDIVIDUAL   \n",
       "2   11987.0       92.2       26.0                   f        INDIVIDUAL   \n",
       "3    5472.0       21.5       13.0                   f        INDIVIDUAL   \n",
       "4   24584.0       69.8       43.0                   f        INDIVIDUAL   \n",
       "\n",
       "   mort_acc  pub_rec_bankruptcies  \\\n",
       "0       0.0                   0.0   \n",
       "1       3.0                   0.0   \n",
       "2       0.0                   0.0   \n",
       "3       0.0                   0.0   \n",
       "4       1.0                   0.0   \n",
       "\n",
       "                                           address  loan_repaid  \n",
       "0     0174 Michelle Gateway\\nMendozaberg, OK 22690            1  \n",
       "1  1076 Carney Fort Apt. 347\\nLoganmouth, SD 05113            1  \n",
       "2  87025 Mark Dale Apt. 269\\nNew Sabrina, WV 05113            1  \n",
       "3            823 Reid Ford\\nDelacruzside, MA 00813            1  \n",
       "4             679 Luna Roads\\nGreggshire, VA 11650            0  \n",
       "\n",
       "[5 rows x 26 columns]"
      ]
     },
     "execution_count": 218,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 219,
   "id": "68d79a89",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "loan_amnt                   0\n",
       "term                        0\n",
       "int_rate                    0\n",
       "installment                 0\n",
       "grade                       0\n",
       "sub_grade                   0\n",
       "home_ownership              0\n",
       "annual_inc                  0\n",
       "verification_status         0\n",
       "issue_d                     0\n",
       "loan_status                 0\n",
       "purpose                     0\n",
       "title                    1755\n",
       "dti                         0\n",
       "earliest_cr_line            0\n",
       "open_acc                    0\n",
       "pub_rec                     0\n",
       "revol_bal                   0\n",
       "revol_util                276\n",
       "total_acc                   0\n",
       "initial_list_status         0\n",
       "application_type            0\n",
       "mort_acc                37795\n",
       "pub_rec_bankruptcies      535\n",
       "address                     0\n",
       "loan_repaid                 0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 219,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "id": "46974849",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0                   Vacation\n",
       "1         Debt consolidation\n",
       "2    Credit card refinancing\n",
       "3    Credit card refinancing\n",
       "4      Credit Card Refinance\n",
       "5         Debt consolidation\n",
       "6           Home improvement\n",
       "7       No More Credit Cards\n",
       "8         Debt consolidation\n",
       "9         Debt Consolidation\n",
       "Name: title, dtype: object"
      ]
     },
     "execution_count": 220,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['title'].head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 221,
   "id": "80d6859c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0              vacation\n",
       "1    debt_consolidation\n",
       "2           credit_card\n",
       "3           credit_card\n",
       "4           credit_card\n",
       "5    debt_consolidation\n",
       "6      home_improvement\n",
       "7           credit_card\n",
       "8    debt_consolidation\n",
       "9    debt_consolidation\n",
       "Name: purpose, dtype: object"
      ]
     },
     "execution_count": 221,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['purpose'].head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c994bb5",
   "metadata": {},
   "source": [
    "**The title column is simply a string subcategory/description of the purpose column. We will drop the title column.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 222,
   "id": "6411c027",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop('title', axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9494c465",
   "metadata": {},
   "source": [
    "---\n",
    "**Let's explore what the mort_acc feature represents**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "id": "d1c903ad",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of mortgage accounts.\n"
     ]
    }
   ],
   "source": [
    "feat_info('mort_acc')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "id": "66279a7a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0     139777\n",
       "1.0      60416\n",
       "2.0      49948\n",
       "3.0      38049\n",
       "4.0      27887\n",
       "5.0      18194\n",
       "6.0      11069\n",
       "7.0       6052\n",
       "8.0       3121\n",
       "9.0       1656\n",
       "10.0       865\n",
       "11.0       479\n",
       "12.0       264\n",
       "13.0       146\n",
       "14.0       107\n",
       "15.0        61\n",
       "16.0        37\n",
       "17.0        22\n",
       "18.0        18\n",
       "19.0        15\n",
       "20.0        13\n",
       "24.0        10\n",
       "22.0         7\n",
       "21.0         4\n",
       "25.0         4\n",
       "27.0         3\n",
       "32.0         2\n",
       "31.0         2\n",
       "23.0         2\n",
       "26.0         2\n",
       "28.0         1\n",
       "30.0         1\n",
       "34.0         1\n",
       "Name: mort_acc, dtype: int64"
      ]
     },
     "execution_count": 224,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['mort_acc'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "id": "76da935b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "loan_amnt               0.222315\n",
       "int_rate               -0.082583\n",
       "installment             0.193694\n",
       "annual_inc              0.236320\n",
       "dti                    -0.025439\n",
       "open_acc                0.109205\n",
       "pub_rec                 0.011552\n",
       "revol_bal               0.194925\n",
       "revol_util              0.007514\n",
       "total_acc               0.381072\n",
       "mort_acc                1.000000\n",
       "pub_rec_bankruptcies    0.027239\n",
       "loan_repaid             0.073111\n",
       "Name: mort_acc, dtype: float64"
      ]
     },
     "execution_count": 225,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.corr()['mort_acc']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d2899d38",
   "metadata": {},
   "source": [
    "**Looks like the total_acc feature correlates with the mort_acc , this makes sense! Let's try this fillna() approach. We will group the dataframe by the total_acc and calculate the mean value for the mort_acc per total_acc entry.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 226,
   "id": "9ea3b7c2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "total_acc\n",
       "2.0      0.000000\n",
       "3.0      0.052023\n",
       "4.0      0.066743\n",
       "5.0      0.103289\n",
       "6.0      0.151293\n",
       "           ...   \n",
       "124.0    1.000000\n",
       "129.0    1.000000\n",
       "135.0    3.000000\n",
       "150.0    2.000000\n",
       "151.0    0.000000\n",
       "Name: mort_acc, Length: 118, dtype: float64"
      ]
     },
     "execution_count": 226,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Mean of mort_acc column per total_acc\n",
    "df.groupby('total_acc')['mort_acc'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "id": "07a83909",
   "metadata": {},
   "outputs": [],
   "source": [
    "tot_acc_mean = df.groupby('total_acc')['mort_acc'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 228,
   "id": "e90ca476",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.10328947368421053"
      ]
     },
     "execution_count": 228,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tot_acc_mean[5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "id": "c76b1afe",
   "metadata": {},
   "outputs": [],
   "source": [
    "def fill_mort(total_acc, mort_acc):\n",
    "    if np.isnan(mort_acc):\n",
    "        return tot_acc_mean[total_acc]\n",
    "    else:\n",
    "        return mort_acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 230,
   "id": "a827d92e",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['mort_acc'] = df.apply(lambda x: fill_mort(x['total_acc'], x['mort_acc']), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 231,
   "id": "2682d2f6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "loan_amnt                 0\n",
       "term                      0\n",
       "int_rate                  0\n",
       "installment               0\n",
       "grade                     0\n",
       "sub_grade                 0\n",
       "home_ownership            0\n",
       "annual_inc                0\n",
       "verification_status       0\n",
       "issue_d                   0\n",
       "loan_status               0\n",
       "purpose                   0\n",
       "dti                       0\n",
       "earliest_cr_line          0\n",
       "open_acc                  0\n",
       "pub_rec                   0\n",
       "revol_bal                 0\n",
       "revol_util              276\n",
       "total_acc                 0\n",
       "initial_list_status       0\n",
       "application_type          0\n",
       "mort_acc                  0\n",
       "pub_rec_bankruptcies    535\n",
       "address                   0\n",
       "loan_repaid               0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 231,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "435af80f",
   "metadata": {},
   "source": [
    "**revol_util and the pub_rec_bankruptcies have missing data points, but they account for less than 0.5% of the total data. We will remove the rows that are missing those values in those columns with dropna().**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 232,
   "id": "d8e13ec2",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 233,
   "id": "3866a4de",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "loan_amnt               0\n",
       "term                    0\n",
       "int_rate                0\n",
       "installment             0\n",
       "grade                   0\n",
       "sub_grade               0\n",
       "home_ownership          0\n",
       "annual_inc              0\n",
       "verification_status     0\n",
       "issue_d                 0\n",
       "loan_status             0\n",
       "purpose                 0\n",
       "dti                     0\n",
       "earliest_cr_line        0\n",
       "open_acc                0\n",
       "pub_rec                 0\n",
       "revol_bal               0\n",
       "revol_util              0\n",
       "total_acc               0\n",
       "initial_list_status     0\n",
       "application_type        0\n",
       "mort_acc                0\n",
       "pub_rec_bankruptcies    0\n",
       "address                 0\n",
       "loan_repaid             0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 233,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "712a3bac",
   "metadata": {},
   "source": [
    "## Categorical Variables and Dummy Variables\n",
    "\n",
    "**We're done working with the missing data! Now we just need to deal with the string values due to the categorical columns.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 234,
   "id": "f0b7b52c",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['term', 'grade', 'sub_grade', 'home_ownership', 'verification_status',\n",
       "       'issue_d', 'loan_status', 'purpose', 'earliest_cr_line',\n",
       "       'initial_list_status', 'application_type', 'address'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 234,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.select_dtypes(['object']).columns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "72a70dd0",
   "metadata": {},
   "source": [
    "---\n",
    "**Let's now go through all the string features to see what we should do with them.**\n",
    "\n",
    "---\n",
    "\n",
    "\n",
    "### term feature\n",
    "\n",
    "**Let's convert the term feature into either a 36 or 60 integer numeric data type using .apply() or .map().**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 235,
   "id": "d30c3548",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 36 months    301247\n",
       " 60 months     93972\n",
       "Name: term, dtype: int64"
      ]
     },
     "execution_count": 235,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['term'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "id": "19d7a301",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['term'] = df['term'].apply(lambda term: int(term[:3]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3a5cc6b0",
   "metadata": {},
   "source": [
    "### grade feature\n",
    "\n",
    "**We already know grade is part of sub_grade, so we'll just drop the grade feature.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "id": "0ea96aba",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop('grade', axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "580caf73",
   "metadata": {},
   "source": [
    "**Let's Convert the subgrade into dummy variables. Then concatenate these new columns to the original dataframe.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 238,
   "id": "97c3a89c",
   "metadata": {},
   "outputs": [],
   "source": [
    "subgrade_dummies = pd.get_dummies(df['sub_grade'],drop_first=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 239,
   "id": "b8a18364",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.concat([df.drop('sub_grade',axis=1),subgrade_dummies],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 240,
   "id": "ac51c290",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['loan_amnt', 'term', 'int_rate', 'installment', 'home_ownership',\n",
       "       'annual_inc', 'verification_status', 'issue_d', 'loan_status',\n",
       "       'purpose', 'dti', 'earliest_cr_line', 'open_acc', 'pub_rec',\n",
       "       'revol_bal', 'revol_util', 'total_acc', 'initial_list_status',\n",
       "       'application_type', 'mort_acc', 'pub_rec_bankruptcies', 'address',\n",
       "       'loan_repaid', 'A2', 'A3', 'A4', 'A5', 'B1', 'B2', 'B3', 'B4', 'B5',\n",
       "       'C1', 'C2', 'C3', 'C4', 'C5', 'D1', 'D2', 'D3', 'D4', 'D5', 'E1', 'E2',\n",
       "       'E3', 'E4', 'E5', 'F1', 'F2', 'F3', 'F4', 'F5', 'G1', 'G2', 'G3', 'G4',\n",
       "       'G5'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 240,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 241,
   "id": "ccdedd23",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['home_ownership', 'verification_status', 'issue_d', 'loan_status',\n",
       "       'purpose', 'earliest_cr_line', 'initial_list_status',\n",
       "       'application_type', 'address'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 241,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.select_dtypes(['object']).columns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b65bb49d",
   "metadata": {},
   "source": [
    "### verification_status, application_type,initial_list_status,purpose \n",
    "**Let's convert these columns: ['verification_status', 'application_type','initial_list_status','purpose'] into dummy variables and concatenate them with the original dataframe.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 250,
   "id": "e9581ad1",
   "metadata": {},
   "outputs": [],
   "source": [
    "verf_dummies = pd.get_dummies(df['verification_status'], drop_first=True)\n",
    "df = pd.concat([df.drop('verification_status', axis=1), verf_dummies], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 338,
   "id": "a0d5ee0a",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "\"None of [Index(['application_type', 'initial_list_status', 'purpose'], dtype='object')] are in the [columns]\"",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "Input \u001b[1;32mIn [338]\u001b[0m, in \u001b[0;36m<cell line: 1>\u001b[1;34m()\u001b[0m\n\u001b[1;32m----> 1\u001b[0m dummies \u001b[38;5;241m=\u001b[39m pd\u001b[38;5;241m.\u001b[39mget_dummies(\u001b[43mdf\u001b[49m\u001b[43m[\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mapplication_type\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43minitial_list_status\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mpurpose\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m]\u001b[49m, drop_first\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m)\n\u001b[0;32m      2\u001b[0m df \u001b[38;5;241m=\u001b[39m df\u001b[38;5;241m.\u001b[39mdrop([\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mapplication_type\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124minitial_list_status\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mpurpose\u001b[39m\u001b[38;5;124m'\u001b[39m], axis\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m1\u001b[39m)\n",
      "File \u001b[1;32m~\\anaconda3\\lib\\site-packages\\pandas\\core\\frame.py:3511\u001b[0m, in \u001b[0;36mDataFrame.__getitem__\u001b[1;34m(self, key)\u001b[0m\n\u001b[0;32m   3509\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m is_iterator(key):\n\u001b[0;32m   3510\u001b[0m         key \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mlist\u001b[39m(key)\n\u001b[1;32m-> 3511\u001b[0m     indexer \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcolumns\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_get_indexer_strict\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkey\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mcolumns\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m)\u001b[49m[\u001b[38;5;241m1\u001b[39m]\n\u001b[0;32m   3513\u001b[0m \u001b[38;5;66;03m# take() does not accept boolean indexers\u001b[39;00m\n\u001b[0;32m   3514\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mgetattr\u001b[39m(indexer, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mdtype\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;241m==\u001b[39m \u001b[38;5;28mbool\u001b[39m:\n",
      "File \u001b[1;32m~\\anaconda3\\lib\\site-packages\\pandas\\core\\indexes\\base.py:5782\u001b[0m, in \u001b[0;36mIndex._get_indexer_strict\u001b[1;34m(self, key, axis_name)\u001b[0m\n\u001b[0;32m   5779\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m   5780\u001b[0m     keyarr, indexer, new_indexer \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_reindex_non_unique(keyarr)\n\u001b[1;32m-> 5782\u001b[0m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_raise_if_missing\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkeyarr\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mindexer\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis_name\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m   5784\u001b[0m keyarr \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtake(indexer)\n\u001b[0;32m   5785\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(key, Index):\n\u001b[0;32m   5786\u001b[0m     \u001b[38;5;66;03m# GH 42790 - Preserve name from an Index\u001b[39;00m\n",
      "File \u001b[1;32m~\\anaconda3\\lib\\site-packages\\pandas\\core\\indexes\\base.py:5842\u001b[0m, in \u001b[0;36mIndex._raise_if_missing\u001b[1;34m(self, key, indexer, axis_name)\u001b[0m\n\u001b[0;32m   5840\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m use_interval_msg:\n\u001b[0;32m   5841\u001b[0m         key \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mlist\u001b[39m(key)\n\u001b[1;32m-> 5842\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mNone of [\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mkey\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m] are in the [\u001b[39m\u001b[38;5;132;01m{\u001b[39;00maxis_name\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m]\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m   5844\u001b[0m not_found \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mlist\u001b[39m(ensure_index(key)[missing_mask\u001b[38;5;241m.\u001b[39mnonzero()[\u001b[38;5;241m0\u001b[39m]]\u001b[38;5;241m.\u001b[39munique())\n\u001b[0;32m   5845\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mnot_found\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m not in index\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
      "\u001b[1;31mKeyError\u001b[0m: \"None of [Index(['application_type', 'initial_list_status', 'purpose'], dtype='object')] are in the [columns]\""
     ]
    }
   ],
   "source": [
    "dummies = pd.get_dummies(df[['application_type', 'initial_list_status', 'purpose']], drop_first=True)\n",
    "df = df.drop(['application_type', 'initial_list_status', 'purpose'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 339,
   "id": "cb7bc89f",
   "metadata": {},
   "outputs": [],
   "source": [
    "#actually we run the above cell twice! No need to worry about the Error.... :)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 256,
   "id": "ce9b62bd",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.concat([df, dummies], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "64884415",
   "metadata": {},
   "source": [
    "### home_ownership"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 259,
   "id": "1e33f2ed",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MORTGAGE    198022\n",
       "RENT        159395\n",
       "OWN          37660\n",
       "OTHER          110\n",
       "NONE            29\n",
       "ANY              3\n",
       "Name: home_ownership, dtype: int64"
      ]
     },
     "execution_count": 259,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['home_ownership'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b502760d",
   "metadata": {},
   "source": [
    "**Convert these to dummy variables, but replace NONE and ANY with OTHER, so that we end up with just 4 categories, MORTGAGE, RENT, OWN, OTHER. Then concatenate them with the original dataframe.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 260,
   "id": "90ffa8c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['home_ownership'] = df['home_ownership'].replace(['NONE', 'ANY'], 'OTHER')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 261,
   "id": "2087d2a0",
   "metadata": {},
   "outputs": [],
   "source": [
    "home_dummies = pd.get_dummies(df['home_ownership'], drop_first=True)\n",
    "df = pd.concat([df.drop('home_ownership', axis=1), home_dummies], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f8066f49",
   "metadata": {},
   "source": [
    "### address\n",
    "**Let's feature engineer a zip code column from the address in the data set. Create a column called 'zip_code' that extracts the zip code from the address column.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 291,
   "id": "c4295ca1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'22690'"
      ]
     },
     "execution_count": 291,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['address'][0][-5:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "id": "a4eea4a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['zip_code'] = df['address'].apply(lambda x:x[-5:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 283,
   "id": "1b195d96",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10"
      ]
     },
     "execution_count": 283,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['zip_code'].nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 285,
   "id": "52fac69b",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Dummies for zipcode\n",
    "zip_dummies = pd.get_dummies(df['zip_code'], drop_first=True)\n",
    "df = pd.concat([df.drop('zip_code', axis=1), zip_dummies], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c459e1e",
   "metadata": {},
   "source": [
    "### issue_d \n",
    "\n",
    "**This would be data leakage, we wouldn't know beforehand whether or not a loan would be issued when using our model, so in theory we wouldn't have an issue_date, going to drop this feature.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 278,
   "id": "40a9e4b6",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop('issue_d', axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fdbb890f",
   "metadata": {},
   "source": [
    "### earliest_cr_line\n",
    "**This appears to be a historical time stamp feature. We'll extract the year from this feature then convert it to a numeric feature. Set this new data to a feature column called 'earliest_cr_year'.Then drop the earliest_cr_line feature.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 288,
   "id": "b8fdb001",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'1990'"
      ]
     },
     "execution_count": 288,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['earliest_cr_line'][0][-4:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 289,
   "id": "08c2a5db",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['earliest_cr_year'] = df['earliest_cr_line'].apply(lambda cr:int(cr[-4:]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 290,
   "id": "09e993bf",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop('earliest_cr_line', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 293,
   "id": "c70ad3dd",
   "metadata": {},
   "outputs": [],
   "source": [
    "#We forgot to drop address...\n",
    "df = df.drop('address', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 294,
   "id": "00892326",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['loan_status'], dtype='object')"
      ]
     },
     "execution_count": 294,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.select_dtypes(['object']).columns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b0716c2",
   "metadata": {},
   "source": [
    "**Let's drop the load_status column since its a duplicate of the loan_repaid column which we created earlier. We'll use the loan_repaid column since its already in 0s and 1s.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 295,
   "id": "4a32248b",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop('loan_status', axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cdf626d3",
   "metadata": {},
   "source": [
    "# Section 3: Model Building & Training.\n",
    "## Train Test Split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 297,
   "id": "34647fcb",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 303,
   "id": "d7dab941",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df.drop('loan_repaid', axis=1).values\n",
    "y = df['loan_repaid'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 305,
   "id": "c7d4c8fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=101)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b3af927",
   "metadata": {},
   "source": [
    "## Normalizing the Data\n",
    "\n",
    "**We'll use a MinMaxScaler to normalize the feature data X_train and X_test. We don't want data leakge from the test set so we only fit on the X_train data.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 306,
   "id": "704e5185",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 308,
   "id": "89262dad",
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = MinMaxScaler()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 309,
   "id": "6d6773e6",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = scaler.fit_transform(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 310,
   "id": "86a1c839",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test = scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0b9d632b",
   "metadata": {},
   "source": [
    "# Creating the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 311,
   "id": "6e2e3214",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense,Dropout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 313,
   "id": "69182664",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "\n",
    "#input layer\n",
    "model.add(Dense(78, activation='relu'))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "#hidden layer\n",
    "model.add(Dense(39, activation='relu'))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "#hidden layer\n",
    "model.add(Dense(19, activation='relu'))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "#output layer\n",
    "model.add(Dense(1, activation='sigmoid'))\n",
    "\n",
    "#model compilation\n",
    "model.compile(loss='binary_crossentropy', optimizer='adam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 314,
   "id": "0f8a953f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/25\n",
      "1236/1236 [==============================] - 11s 8ms/step - loss: 0.2997 - val_loss: 0.2672\n",
      "Epoch 2/25\n",
      "1236/1236 [==============================] - 9s 8ms/step - loss: 0.2660 - val_loss: 0.2631\n",
      "Epoch 3/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2630 - val_loss: 0.2632\n",
      "Epoch 4/25\n",
      "1236/1236 [==============================] - 9s 8ms/step - loss: 0.2619 - val_loss: 0.2621\n",
      "Epoch 5/25\n",
      "1236/1236 [==============================] - 9s 8ms/step - loss: 0.2612 - val_loss: 0.2620\n",
      "Epoch 6/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2608 - val_loss: 0.2620\n",
      "Epoch 7/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2604 - val_loss: 0.2618\n",
      "Epoch 8/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2598 - val_loss: 0.2616\n",
      "Epoch 9/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2595 - val_loss: 0.2613\n",
      "Epoch 10/25\n",
      "1236/1236 [==============================] - 9s 8ms/step - loss: 0.2592 - val_loss: 0.2613\n",
      "Epoch 11/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2590 - val_loss: 0.2612\n",
      "Epoch 12/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2587 - val_loss: 0.2617\n",
      "Epoch 13/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2585 - val_loss: 0.2613\n",
      "Epoch 14/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2583 - val_loss: 0.2612\n",
      "Epoch 15/25\n",
      "1236/1236 [==============================] - 9s 8ms/step - loss: 0.2579 - val_loss: 0.2613\n",
      "Epoch 16/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2580 - val_loss: 0.2609\n",
      "Epoch 17/25\n",
      "1236/1236 [==============================] - 9s 8ms/step - loss: 0.2577 - val_loss: 0.2609\n",
      "Epoch 18/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2577 - val_loss: 0.2608\n",
      "Epoch 19/25\n",
      "1236/1236 [==============================] - 9s 8ms/step - loss: 0.2572 - val_loss: 0.2612\n",
      "Epoch 20/25\n",
      "1236/1236 [==============================] - 9s 8ms/step - loss: 0.2571 - val_loss: 0.2610\n",
      "Epoch 21/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2571 - val_loss: 0.2608\n",
      "Epoch 22/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2569 - val_loss: 0.2610\n",
      "Epoch 23/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2567 - val_loss: 0.2609\n",
      "Epoch 24/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2567 - val_loss: 0.2610\n",
      "Epoch 25/25\n",
      "1236/1236 [==============================] - 9s 7ms/step - loss: 0.2564 - val_loss: 0.2607\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1d883693ca0>"
      ]
     },
     "execution_count": 314,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(x=X_train, y=y_train, epochs=25, batch_size=256, validation_data=(X_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 315,
   "id": "cdda69fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Save the model\n",
    "# from tensorflow.keras.models import load_model\n",
    "# model.save('full_data_project_model.h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "00849024",
   "metadata": {},
   "source": [
    "# Section 4: Evaluating Model Performance.\n",
    "\n",
    "**Validation loss versus the training loss plots.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 317,
   "id": "07047e8d",
   "metadata": {},
   "outputs": [],
   "source": [
    "loses = pd.DataFrame(model.history.history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 319,
   "id": "8215f70f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 319,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAD4CAYAAADrRI2NAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAmKklEQVR4nO3de5gcdZ3v8fe3rzPTPZlkOjO5TUISQAIkgBhQ3DVedldEREQ9GERABFl0ReE8y4Mej5dd1scLrpezD8rhIAc8osIqKh6j7K4XAsvlJMGQEAIBEkJmEpKZyUwmc+3b7/xRNTM9k55JTzIznXR9Xs/TT1VXV3X/ajr5/Kp+9atfm3MOEREJjlC5CyAiItNLwS8iEjAKfhGRgFHwi4gEjIJfRCRgIuUuQDGzZ892ixcvLncxRESOGxs2bGhzzjWUsu4xGfyLFy9m/fr15S6GiMhxw8x2lrqumnpERAJGwS8iEjAKfhGRgDkm2/hFJHgymQzNzc309/eXuyjHtKqqKpqamohGo0f8HiUFv5m9C/guEAbucs59bdTrFwO3AnkgC9zonHuslG1FRACam5upra1l8eLFmFm5i3NMcs7R3t5Oc3MzS5YsOeL3OWxTj5mFgduBC4DTgMvM7LRRq/0eONM5dxbwMeCuCWwrIkJ/fz+pVEqhPw4zI5VKHfVZUSlt/OcCLznntjvn0sBPgYsLV3DOdbvhYT4TgCt1WxGRQQr9w5uMv1Epwb8A2FXwvNlfNrowl5jZ88Bv8I76S97W3/46M1tvZutbW1tLKfsIzjn+5fcv8si2iW8rIhIkpQR/serlkEH8nXO/cM4tA96H195f8rb+9nc651Y651Y2NJR089nIQppx59rt/PH5fRPeVkQEIJlMlrsI06KU4G8GFhY8bwJ2j7Wyc24tcKKZzZ7otkerPhmjvSc9VW8vIlIRSgn+dcDJZrbEzGLAauChwhXM7CTzG57M7GwgBrSXsu1kqk/E2N8zMFVvLyIB4Zzj5ptvZvny5axYsYL7778fgD179rBq1SrOOussli9fzqOPPkoul+OjH/3o0Lrf/va3y1z6wztsd07nXNbMPgU8jNcl827n3BYzu95//Q7gA8CVZpYB+oAP+Rd7i247RftCKhGjuaNvqt5eRKbJP/x6C8/t7prU9zxt/gy+dNHpJa374IMPsnHjRp555hna2to455xzWLVqFT/+8Y85//zz+fznP08ul6O3t5eNGzfS0tLCs88+C0BnZ+eklnsqlNSP3zm3BlgzatkdBfNfB75e6rZTpT4RY1Pzgen4KBGpYI899hiXXXYZ4XCYOXPm8Na3vpV169Zxzjnn8LGPfYxMJsP73vc+zjrrLJYuXcr27du54YYbuPDCC3nnO99Z7uIfVkXduVufiNPRm8Y5p25hIsexUo/Mp8pw7/SRVq1axdq1a/nNb37DFVdcwc0338yVV17JM888w8MPP8ztt9/OAw88wN133z3NJZ6YihqrJ5WIkck5uvqz5S6KiBzHVq1axf33308ul6O1tZW1a9dy7rnnsnPnThobG/n4xz/ONddcw9NPP01bWxv5fJ4PfOAD3HrrrTz99NPlLv5hVdQRfyoZA2B/T5q66iMfx0JEgu2SSy7hiSee4Mwzz8TM+MY3vsHcuXO59957ue2224hGoySTSX74wx/S0tLC1VdfTT6fB+CrX/1qmUt/eBUV/PWJweAfYMnsRJlLIyLHm+7ubsC7L+i2227jtttuG/H6VVddxVVXXXXIdsfDUX6hCmvqiQPQ3q2+/CIiY6mo4K8vaOoREZHiKir4U35Tj+7eFREZW0UFf1U0TE0srKYeEZFxVFTwg4ZtEBE5nIoL/lRCA7WJiIyn4oLfO+JX8IuIjKUCgz+u4BeRKTfe2P2vvPIKy5cvn8bSTEzFBf9sf0z+scbaEBEJuoq6cxe8pp50Nk9POkcyXnG7JxIMv/0svLZ5ct9z7gq44GtjvnzLLbdwwgkn8MlPfhKAL3/5y5gZa9eupaOjg0wmwz/90z9x8cUT+9nw/v5+PvGJT7B+/XoikQjf+ta3ePvb386WLVu4+uqrSafT5PN5fv7znzN//nwuvfRSmpubyeVyfOELX+BDH/rQUe12MRWXjEPDNnSnFfwiUrLVq1dz4403DgX/Aw88wO9+9ztuuukmZsyYQVtbG29605t473vfO6HRf2+//XYANm/ezPPPP8873/lOtm3bxh133MFnPvMZLr/8ctLpNLlcjjVr1jB//nx+85vfAHDgwNQMM19xyTg4UFt7zwCLUjVlLo2IHJFxjsynyutf/3r27dvH7t27aW1tZdasWcybN4+bbrqJtWvXEgqFaGlpYe/evcydO7fk933ssce44YYbAFi2bBknnHAC27Zt47zzzuMrX/kKzc3NvP/97+fkk09mxYoV/P3f/z233HIL73nPe3jLW94yJftacW389f54PbrAKyIT9cEPfpCf/exn3H///axevZr77ruP1tZWNmzYwMaNG5kzZw79/f0Tes+xrjd++MMf5qGHHqK6uprzzz+fP/zhD7zuda9jw4YNrFixgs997nP84z/+42Ts1iEq74hfwzaIyBFavXo1H//4x2lra+ORRx7hgQceoLGxkWg0yh//+Ed27tw54fdctWoV9913H+94xzvYtm0br776Kqeccgrbt29n6dKlfPrTn2b79u1s2rSJZcuWUV9fz0c+8hGSyST33HPP5O8kFRj8g238GrZBRCbq9NNP5+DBgyxYsIB58+Zx+eWXc9FFF7Fy5UrOOussli1bNuH3/OQnP8n111/PihUriEQi3HPPPcTjce6//35+9KMfEY1GmTt3Ll/84hdZt24dN998M6FQiGg0yve///0p2EuwY7Hb48qVK9369euPaFvnHMu+8DuuPO8EPn/haZNcMhGZKlu3buXUU08tdzGOC8X+Vma2wTm3spTtK66N38w0bIOIyDgqrqkHvHH5dXFXRKba5s2bueKKK0Ysi8fjPPXUU2UqUWkqMvhTGrZB5LjknJtQH/lyW7FiBRs3bpzWz5yM5vmKa+oBf4ROXdwVOa5UVVXR3t6u4VbG4Zyjvb2dqqqqo3qfijzi1widIsefpqYmmpubaW1tLXdRjmlVVVU0NTUd1XtUZvAnY/RlcvSlc1THwuUujoiUIBqNsmTJknIXIxAqtqkHvGEbRERkpIoMfg3bICIytgoNft29KyIylooMfo3XIyIytooM/np/aOb9auMXETlERQZ/bTxCNGw64hcRKaIig9/MvL78auMXETlERQY/aNgGEZGxVG7wJzVCp4hIMRUb/Bq2QUSkuJKC38zeZWYvmNlLZvbZIq9fbmab/MfjZnZmwWufMbNnzWyLmd04iWUfl4JfRKS4wwa/mYWB24ELgNOAy8xs9E9b7QDe6pw7A7gVuNPfdjnwceBc4EzgPWZ28uQVf2ypRIzugSwD2dx0fJyIyHGjlCP+c4GXnHPbnXNp4KfAxYUrOOced851+E+fBAaHjjsVeNI51+ucywKPAJdMTtHHp2EbRESKKyX4FwC7Cp43+8vGcg3wW3/+WWCVmaXMrAZ4N7DwSAo6URq2QUSkuFKGZS72czhFfynBzN6OF/x/CeCc22pmXwf+HegGngGyY2x7HXAdwKJFi0oo1vhSSQ3bICJSTClH/M2MPEpvAnaPXsnMzgDuAi52zrUPLnfO/cA5d7ZzbhWwH3ix2Ic45+50zq10zq1saGiYyD4UNXjEr2EbRERGKiX41wEnm9kSM4sBq4GHClcws0XAg8AVzrlto15rLFjn/cBPJqPgh5NSU4+ISFGHbepxzmXN7FPAw0AYuNs5t8XMrvdfvwP4IpACvuf/UHLWObfSf4ufm1kKyAB/V3AReErNqIoSCZku7oqIjFLSTy8659YAa0Ytu6Ng/lrg2jG2fcvRFPBIhULGLPXlFxE5RMXeuQtec48u7oqIjFTRwa+7d0VEDqXgFxEJmIoO/lQiRnu3unOKiBSq6OCvT8Tp6s+SzubLXRQRkWNGZQe/f/duR6+ae0REBlV08OsmLhGRQ1V08A8P26DgFxEZVNHBP3TEr/F6RESGVHbwJzUmv4jIaBUd/DOro4RMwS8iUqiigz8UMmbVaNgGEZFCFR384N+9q149IiJDghH8OuIXERlS8cGfSsbUq0dEpEDFB7+O+EVERgpA8Mfp6M2QzWm8HhERCEDwD97E1dGbKXNJRESODRUf/Bq2QURkpIoP/lRSwzaIiBSq/OBPaNgGEZFCFR/8auoRERmp4oN/Vk0U0Jj8IiKDKj74I+EQM2uiOuIXEfFVfPCDbuISESkUiOBPJTRsg4jIoEAEf30ipjZ+ERFfQII/rqYeERFfIII/lYjR0Zsmn3flLoqISNkFI/iTMfIOOvs0Xo+ISCCCf/gmLl3gFREJRPAPDtugC7wiIgEJfg3bICIyLBDBPzxCp4JfRCQQwT+rRkf8IiKDAhH8sUiI2qqIgl9EhIAEPwwO26DgFxEpKfjN7F1m9oKZvWRmny3y+uVmtsl/PG5mZxa8dpOZbTGzZ83sJ2ZWNZk7UCpv2AZ15xQROWzwm1kYuB24ADgNuMzMThu12g7grc65M4BbgTv9bRcAnwZWOueWA2Fg9eQVv3QatkFExFPKEf+5wEvOue3OuTTwU+DiwhWcc4875zr8p08CTQUvR4BqM4sANcDuoy/2xKmpR0TEU0rwLwB2FTxv9peN5RrgtwDOuRbgm8CrwB7ggHPu34ptZGbXmdl6M1vf2tpaStknJJWM0dGTxjmN1yMiwVZK8FuRZUXT08zejhf8t/jPZ+GdHSwB5gMJM/tIsW2dc3c651Y651Y2NDSUUvYJqU/EyOYdXX3ZSX9vEZHjSSnB3wwsLHjeRJHmGjM7A7gLuNg51+4v/mtgh3Ou1TmXAR4E3nx0RT4ywzdx6QKviARbKcG/DjjZzJaYWQzv4uxDhSuY2SK8UL/CObet4KVXgTeZWY2ZGfBXwNbJKfrE1Pvj9egCr4gEXeRwKzjnsmb2KeBhvF45dzvntpjZ9f7rdwBfBFLA97x8J+s32zxlZj8DngaywJ/xe/xMt1RCwzaIiEAJwQ/gnFsDrBm17I6C+WuBa8fY9kvAl46ijJNCA7WJiHgCc+eugl9ExBOY4K+KhknEwhqTX0QCLzDBD1CfjKlXj4gEXrCCX8M2iIgEK/hnJ2Jq6hGRwAtU8NcnYjriF5HAC1bwJ73g13g9IhJkgQr+VCJGOpene0Dj9YhIcAUq+DVsg4hIwIJfwzaIiAQs+Ifu3lXPHhEJsGAGv474RSTAAhX8g2Pyt+nuXREJsEAFf00sQlU0pKYeEQm0QAU/QErDNohIwAUv+JMx9eoRkUALXPBr2AYRCToFv4hIwAQu+FMJjckvIsEWuOCvT8Tpz+TpTWu8HhEJpsAF/9CwDerSKSIBFbjg1927IhJ0wQv+pIJfRIItcME/2NTT1q0LvCISTIELfjX1iEjQBS74k/EIsUhIwS8igRW44Dczvy+/gl9EgilwwQ+6e1dEgi2wwa8jfhEJqkAGfyoRY7+GbRCRgApk8Ncn4voxFhEJrEAGfyoZoyedoz+TK3dRRESmXSCDX335RSTIFPwiIgETyODXsA0iEmTBDP5kHNARv4gEUyCDX009IhJkJQW/mb3LzF4ws5fM7LNFXr/czDb5j8fN7Ex/+SlmtrHg0WVmN07yPkzYjKoI0bDpJi4RCaTI4VYwszBwO/A3QDOwzswecs49V7DaDuCtzrkOM7sAuBN4o3PuBeCsgvdpAX4xubswcWbGrJqY+vKLSCCVcsR/LvCSc267cy4N/BS4uHAF59zjzrkO/+mTQFOR9/kr4GXn3M6jKfBk0bANIhJUpQT/AmBXwfNmf9lYrgF+W2T5auAnY21kZteZ2XozW9/a2lpCsY5OKqlhG0QkmEoJfiuyzBVd0ezteMF/y6jlMeC9wL+O9SHOuTudcyudcysbGhpKKNbRqU/EdXFXRALpsG38eEf4CwueNwG7R69kZmcAdwEXOOfaR718AfC0c27vkRZ0smlMfhEJqlKO+NcBJ5vZEv/IfTXwUOEKZrYIeBC4wjm3rch7XMY4zTzlUJ+IcbA/SzqbL3dRRESm1WGP+J1zWTP7FPAwEAbuds5tMbPr/dfvAL4IpIDvmRlA1jm3EsDMavB6BP3t1OzCkRnsy9/Rm2bOjKoyl0ZEZPqU0tSDc24NsGbUsjsK5q8Frh1j2168SuGYUjhsg4JfRIIkkHfugoZtEJHgCmzwa9gGEQmqwAb/YFNPu+7eFZGACWzw11VHCYdMR/wiEjiBDf5QyJhVE1VffhEJnMoK/ufXwP7tJa9en9CwDSISPJUT/H0d8Ivr4VefgnxpN2V5wa8jfhEJlsoJ/upZ8K6vws7/hP93Z0mbpBJxNfWISOBUTvADnPVhOPl8+I8vQ/vLh11dR/wiEkSVFfxmcNF3IRKDX34S8rlxV69PxOjszZDJabweEQmOygp+gBnz4ILbYNeT8OT3x111dnJ4vB4RkaCovOAHOONSOOVC+MOt0FpssFBPfcIbtmFfl3r2iEhwVGbwm8F7vg3RavjlJ8Zs8lmxoI6qaIhbfr6Jrv7MNBdSRKQ8KjP4AWrnwLu/CS3r4fF/KbrKolQNd3zkDWzbe5Br7llHX3r8awIiIpWgcoMfYPkH4NT3wh+/AvueL7rK205p5Dsfej0bdnZw/Y826IdZRKTiVXbwm8GF34J4Lfzyeshli6524Rnz+Nr7z+CRba3ceP+fyeWL/qSwiEhFqOzgB0g2wIX/DLv/DP/5nTFXu/Schfz3C09lzebX+NyDm8gr/EWkQpX0C1zHvdMvgecegj99DU65AOacXnS1a9+ylK7+LP/j9y+SjEf5wntOxf8pSRGRilH5R/yD3v1NqJ7pjeeTG7sHz01/fTJX/8Vi7v7PHXz39y9OX/lERKZJcII/kfK6eL62CR791pirmRlfuPA0PviGJr7zHy/yg8d2TGMhRUSmXnCCH+DUi2DFpbD2G7Bn05irhULG196/gguWz+XW//scD6zfNY2FFBGZWsEKfoALvg41Ke/GruzYQzVEwiG+s/os3nLybD77802s2bxnGgspIjJ1ghf8NfXeQG57n4W1t427ajwS5n9e8QbOXjSLz/z0zzyyrXWaCikiMnWCF/zg9ew588PwqN/Ncxw1sQg/+Og5nNxYy9/+n/Wse2X/NBVSRGRqmHPHXn/1lStXuvXr10/th/R1wvfeBLEkrPgg5NJeb59cxpvPF8znMgykB9j4Sisum2bZSUuZ+aYr4cR3QCg8teUUESmBmW1wzq0sad3ABj/AS7+H+6+ATA+EIhCKQjgG4WjBI+Yvj5Imwta9fTTlW0hZFwM184itvAJ7/Udg1glTX14RkTEo+CcinwMMQqW1eu3u7OMHf3qB/Rt/xcW537MqvAkDsovfRnTllbDsQojEp7TIIiKjKfinQV86x6837ebhx9ezfO+vuTTyCAusjWx8FuHXX4adfSU0nlruYopIQCj4p9mzLQf4yVM7aN34O97r/sD54Q1EyZJbcA7hN1zlDRkRT5a7mCJSwRT8ZdI9kOVXG1v49eObOb1tDR+O/IkTrYVcNEF49kn+dYQIWNi7KByKFEwjYKHh+VC4YL3B+cHloeLrRKqgehZU13vdVgfnq2fqIrRIhVPwl5lzjo27Ovnxkztp3vwnLnKPsCTeRao6xKzqMHVVIWKW964v5LPew+X9+dzw1OWGn7sc5PMF8/5ySvn+DKrqRlYGNfXD05p676a2wcfgcl2rEDluKPiPIQd6M/zqmRYeeaGVp3bsp3vA+02A181Jct7SFOedmOKNS1LMSsSO7AOcG64Esn3Q1wG9HdC335/f780PTkcs64D0wbHfO1ZbUCkUVA6J2ZBogESjN+z14Hy06sj2YarlMrB/O+x7zvtBnn3PQevz0L0PZsyHmYu8R91Cf34hzDzB21eNzirHCQX/MSqby/Ps7i6eeLmdJ7a3s27HfvoyOczg1LkzOO/EFG8+McU5S+qZURWdpkKl/Uqg3asQettHzvcVLmuHnnav+2sxsVq/Imj0Kodko1cpVNdDODLULdabRgq60BZ5LRz3zjgi8YL5Km+dscI4n4OOV2DfVu/R6k/bXvTuywDAoH6pd+E9OQcO7oHOV73HQNfI94vW+JXBwuGKIdnoLY8lIVYDsQREE940VuPNR8apxHNZ7++X7oV0D6S7/WmPv7zHO/tLnQQNy7wKV6QECv7jRDqbZ1NzJ0+83M7jL7ez4dUO0tk8IYPlC+p43ZxaljYkWDo7yYkNCRalaohHjoG2+nQv9LQOP7r3jf28dz+lNUdNQKRqZGUQiXnXOjp3QrZ/eL2Zi6DxNC9AG0/1HrNfB9Hq4u/b1wkHdvkVgT898Orw874S79oORf1KIel9VjY9HOqF5StFohEaTvH2YWi6zKtgjyWZfmh/yTuTGnzsex4yfd7vXyfnQq3/SM6B2nne8tp5UDO75O7UMjYF/3GqP5Pjz692Dp0NvNzazb6DA0OvhwwW1tewZLZXGSxtSLC0IcGJDUkaa+PH5o/G5LLekXQuM3w3dD5b8Ny/xjH6tVwasgOQG/Cm2QEvNA9Z5j/Ppb3mmcZToeFULyQnuyfVwEGvqSw9eMTeDZne4SP2wqP2wSP6TK9XQcUSo84OEv5ZQ+LQh8t7ZylDIfqC9yg8I6lJjawMko1esx/OmxbOF5uCV2lGa4bPVArPWmI13muj/01l+g4t276t0LHDKzd4lXD9Uv87mAHdr8HBvd7ZVbHK08J+ZeBXENWzvO8uXuv9jeK1Ix+xpPe+g+tEqrxKu6/IGevQWeuos9pc2quUozUFf4PBR3XBsmrv7xGt9v5uQ3fzp70KfXC+cPngw+W9Sq2wkius9OIzJrUpUcFfQQ72Z9jR1sP21h62t3az3Z/f0dZDXyY3tF4iFmZJQ4LFqQRLZnvTxbNrWJxKUJ+IHZuVgpTOOejaXVARDE63Qv+Bqfvc6GBzVg3gvDOfwYojFIH6E72Abzx1uBJKnTR2x4DsAHTvHa4IuvfCwde8R7c/7T/gVbIDB72ODEcjHC+4RjXYoSHllS/T5z96vGm616uoM73+8l5vWbZv1Jua3wQZG767f2gaH54H/wx4r/dexf62o89+6prgvL87ol2d9OA3s3cB3wXCwF3Oua+Nev1y4Bb/aTfwCefcM/5rM4G7gOV4/2I+5px7YrzPU/AfXj7veK2r368UunnZrwxeae+huaNvxA/G11ZFCiqDBEtm13BCKsGSVOLILyrLscE5L1h693vdfM0AGzXl0OXghfDg9YaMfwYzNN8zatrrhXD9idDoNzfVnzj+9YzJ2Lds/3AlMPhId498nu2Hqpl+wM8a2Tstljj6o+p83gt/C/lDuEywudU5r5yFldvB1/xKb8/ISrCqDv7rc0dUzEkNfjMLA9uAvwGagXXAZc655wrWeTOw1TnXYWYXAF92zr3Rf+1e4FHn3F1mFgNqnHOd432mgv/opLN5mjt6eaW9hx1tvbziVwg72nrY3dlH4e/IN9TGWT5/BisW1HH6gjpWLKhjXl2VzhBEyiHTN/Y1qMOYSPCX8mPr5wIvOee2+2/+U+BiYCj4nXOPF6z/JNDkrzsDWAV81F8vDYz96ycyKWKREEsbkixtOLSNeyCbY9f+Xl5p62VHWw9bX+tiS0sXj2xrHaoQ6hMxTvcrg+UL6lg+v46F9dWqDESm2hGG/kSVEvwLgMLfHmwG3jjO+tcAv/XnlwKtwP82szOBDcBnnHOH9Ac0s+uA6wAWLVpUQrHkSMQjYU5qrOWkxtoRy/vSOb8SOMDmlgM829LFnWu3k/VrgxlVEZYvqOP0+TNomlXDnBlxGmdUMWdGFQ3JOLGIemWIHC9KCf5ih3lF24fM7O14wf+XBe9/NnCDc+4pM/su8FngC4e8oXN3AneC19RTQrlkElXHwpy9aBZnL5o1tGwgm2Pba91eRbD7AFtaDnDvEztJZ/OHbJ9KxPyKIM6c2qoRFcPcGVXMn1mli8wix4hSgr8ZWFjwvAnYPXolMzsD7yLuBc659oJtm51zT/nPf4YX/HIciEfCrGiqY0VT3dCyfN6xvzfN3q5+9nUNsLern71dA+w92M8+f/653V20dQ+MuJbgvV+IBTOrmT+zmvkzq5g/s5oF/mP+zGrm1lVRFT0G7lMQqXClBP864GQzWwK0AKuBDxeuYGaLgAeBK5xz2waXO+deM7NdZnaKc+4F4K8ouDYgx59QyJidjDM7Gef0+WOvl83lae/xKojdnf3sOdDH7s4+dnf209LZx59eaB1xj8Kg2ck4C2Z6ZwqNM+I0JKtoqI0PPRprvc9W05LIkTts8Dvnsmb2KeBhvO6cdzvntpjZ9f7rdwBfBFLA9/xT+WzB1eUbgPv8Hj3bgasnfzfkWBMJh5jjN/Wc0VR8nYFsjr0HBmju7GV3Z79fMfTR0tnHzvZe1r2yn47eTNFtZ9ZEaUjG/crBqxRSyTipRIxUMkYqEac+EWN2Mk51TGcRIoV0A5cc09LZPO09A+zrGqD14ACt3d5038F+77m/bF/XAANFrj0A1MTC1CdipJJxZidiw/PJmHcmkRw+o6irjuo6hByXJrs7p0jZxCIh5tVVM69u/G5uzjl60znau9O09QywvztNe88A7T1p2rvTtHd783sO9LNldxftPQNkcoce9MTCIRpq48weVSEMNjPVJ2LMqIpSWxVhRnWURCysikKOOwp+qQhmRiIeIRGPsChVc9j1nXN09WWHziCGpgVnES2dfWzc1Ul7zwBjnRiHDGZU+xVBVdR7VEeoLZifX1dNU301C2fVMK+uikhY1yekvBT8EkhmRl1NlLqaKCc1jj+YWzaXZ39vmn1dA3T2Zujqz3CwP0NXX5au/gxdfRm6+rNDy3a299LVl+Fgf5aD/u8vDAqHjPkzq1g4q8Z71FezsL6GJn++ITn2YHu5vCOTyzOQzZPO5snkhqexiHemUhPTf2k5PP0rETmMSDhEY20VjbUT/6GZTC7Pawf62bW/l10dveza3+dPe/nDC/toHdWzqSrqBXg+jx/wOTI5RzqXHzH+0lgSsbDXVFXYTOXPFy5Tz6hgU/CLTKFoOMTC+hoW1hdvfupL52jpHFkhtB4cIBIOEQ2HiEdCxCIhomEjFg4PzQ8v96b9mTxto5qrXtzXzeMvt3Ogr3jPKDPv7syQGSEz77mNeo7XhTdkRlUkxClza4eG8ljRVMfcGRrX6Xik4Bcpo+pY8SE0JtNA1rvoXXj9ou3gAJlcnrwDhyPvIO8c+NO8P3R/3jmccziguz/Llt0jx3WanYyzYsHwuE5nNM1kzoxj9LchZIiCX6TCxSNh/27pyRkArC+d47k9B9jcfIDNLV0823JgzMqgqb6GsBmhEP5ZhBHyzyoMRjwPhbznkZARDhnRcMibhvxp2IiEQ0RCRiRsQ69FIyFqomFCIVU2pVLwi8iEVMfCvOGEet5wwvDvAfems2zd0zVmZTDVzCAZi1Bb5fWoSlYNz9dWRaiNF7wW97rizqyJMrM6Sl21d5H/mPhZ02mi4BeRo1YTixStDNq700NNRsNNSMNNS15T0nCzUs458nlHJue8Xkz5PLmcI5vPk807sjnnT/ND00zOcXDA61V1sD9Ld3+WgwMZ9vek2dneO7R8rBv8BlVFQ8ysjjGzJupVDH6lMLPGm8Yj4YJrIMPXPkKFZzEFZy/RcMgfoLCaOTOqCB9DZyQKfhGZEjWxCDX1x07EpLP5oUqgqz/Dgb4Mnb3e1JtPDy3r7Mvw6v7eodcLf+b0SERCxtw6rxJo8pvdFsyqHjFQ4XQOLXLsfCsiIlMoFgl54zklx/g94HEM+N1q887h8nhnJgVnLINnM/m89zznHOls3h+csJ8Wfzyqlo4+ntqxn9e6+g/pnlufiHFiQ4J/vf7Nk7XLY1Lwi4gcRjwSJn4EaXnK3OK9tbK5PPsOeneHt3R4AxO2dPaRn6aLIgp+EZFpFgmHhnpanbN4+j9ft+6JiASMgl9EJGAU/CIiAaPgFxEJGAW/iEjAKPhFRAJGwS8iEjAKfhGRgDE31o+JlpGZtQI7j3Dz2UDbJBbneBLkfYdg77/2PbgG9/8E51xDKRsck8F/NMxsvXNuZbnLUQ5B3ncI9v5r34O573Bk+6+mHhGRgFHwi4gETCUG/53lLkAZBXnfIdj7r30Prgnvf8W18YuIyPgq8YhfRETGoeAXEQmYigl+M3uXmb1gZi+Z2WfLXZ7pZmavmNlmM9toZuvLXZ6pZGZ3m9k+M3u2YFm9mf27mb3oT2eVs4xTaYz9/7KZtfjf/0Yze3c5yzhVzGyhmf3RzLaa2RYz+4y/vOK//3H2fcLffUW08ZtZGNgG/A3QDKwDLnPOPVfWgk0jM3sFWOmcq/gbWcxsFdAN/NA5t9xf9g1gv3Pua37FP8s5d0s5yzlVxtj/LwPdzrlvlrNsU83M5gHznHNPm1ktsAF4H/BRKvz7H2ffL2WC332lHPGfC7zknNvunEsDPwUuLnOZZIo459YC+0ctvhi415+/F+8/REUaY/8DwTm3xzn3tD9/ENgKLCAA3/84+z5hlRL8C4BdBc+bOcI/yHHMAf9mZhvM7LpyF6YM5jjn9oD3HwRoLHN5yuFTZrbJbwqquKaO0cxsMfB64CkC9v2P2neY4HdfKcFvRZYd/21YE/MXzrmzgQuAv/ObAyQ4vg+cCJwF7AH+uaylmWJmlgR+DtzonOsqd3mmU5F9n/B3XynB3wwsLHjeBOwuU1nKwjm325/uA36B1/wVJHv9NtDBttB9ZS7PtHLO7XXO5ZxzeeB/UcHfv5lF8YLvPufcg/7iQHz/xfb9SL77Sgn+dcDJZrbEzGLAauChMpdp2phZwr/Yg5klgHcCz46/VcV5CLjKn78K+FUZyzLtBkPPdwkV+v2bmQE/ALY6575V8FLFf/9j7fuRfPcV0asHwO/C9B0gDNztnPtKeUs0fcxsKd5RPkAE+HEl77+Z/QR4G95wtHuBLwG/BB4AFgGvAv/FOVeRF0DH2P+34Z3qO+AV4G8H27wriZn9JfAosBnI+4v/G15bd0V//+Ps+2VM8LuvmOAXEZHSVEpTj4iIlEjBLyISMAp+EZGAUfCLiASMgl9EJGAU/CIiAaPgFxEJmP8PbUaMlGC54O8AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "loses[['loss', 'val_loss']].plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 325,
   "id": "8ba09560",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2471/2471 [==============================] - 6s 2ms/step\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.98      0.44      0.61     15658\n",
      "           1       0.88      1.00      0.93     63386\n",
      "\n",
      "    accuracy                           0.89     79044\n",
      "   macro avg       0.93      0.72      0.77     79044\n",
      "weighted avg       0.90      0.89      0.87     79044\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Classification Report\n",
    "from sklearn.metrics import classification_report\n",
    "predictions = (model.predict(X_test) > 0.5).astype(\"int32\")\n",
    "print(classification_report(y_test, predictions))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 327,
   "id": "1874d227",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 6890  8768]\n",
      " [  135 63251]]\n"
     ]
    }
   ],
   "source": [
    "#Confusion Matrix\n",
    "from sklearn.metrics import confusion_matrix\n",
    "print(confusion_matrix(y_test, predictions))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6ca766b",
   "metadata": {},
   "source": [
    "**Try This: Given the customer below, would you offer this person a loan?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 328,
   "id": "cb07296a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "loan_amnt           25000.00\n",
       "term                   60.00\n",
       "int_rate               18.24\n",
       "installment           638.11\n",
       "annual_inc          61665.00\n",
       "                      ...   \n",
       "48052                   0.00\n",
       "70466                   0.00\n",
       "86630                   0.00\n",
       "93700                   0.00\n",
       "earliest_cr_year     1996.00\n",
       "Name: 305323, Length: 78, dtype: float64"
      ]
     },
     "execution_count": 328,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import random\n",
    "random.seed(101)\n",
    "random_ind = random.randint(0,len(df))\n",
    "\n",
    "new_customer = df.drop('loan_repaid',axis=1).iloc[random_ind]\n",
    "new_customer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 334,
   "id": "8f082502",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 31ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[1]])"
      ]
     },
     "execution_count": 334,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(model.predict(new_customer.values.reshape(1,78)) > 0.5).astype(\"int32\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 335,
   "id": "49788a6e",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Yes, he can pay back!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "98125302",
   "metadata": {},
   "source": [
    "**Now check, did this person actually end up paying back their loan?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 336,
   "id": "046c432f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 336,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.iloc[random_ind]['loan_repaid']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 337,
   "id": "2f424724",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Yes!!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f1091231",
   "metadata": {},
   "source": [
    "## Yay! Finally Done."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}