Predict_dafault_for_bank.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# loading the libraries\n",
    "\n",
    "import csv\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from matplotlib import pyplot\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn import preprocessing\n",
    "from sklearn import metrics\n",
    "from sklearn.linear_model import LogisticRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# opening the data in pd dataframes\n",
    "\n",
    "credit_data = pd.read_csv('/Users/georgekalfas/Documents/Term 1/Ditigal Technologies and Value Creation/3rd Individual assignment/GitHub upload/Credit_data.csv')\n",
    "new_appl = pd.read_csv('/Users/georgekalfas/Documents/Term 1/Ditigal Technologies and Value Creation/3rd Individual assignment/GitHub upload/New_applicantions.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>ID</th>\n",
       "      <th>LIMIT_BAL</th>\n",
       "      <th>SEX</th>\n",
       "      <th>EDUCATION</th>\n",
       "      <th>MARRIAGE</th>\n",
       "      <th>AGE</th>\n",
       "      <th>PAY_1</th>\n",
       "      <th>PAY_2</th>\n",
       "      <th>PAY_3</th>\n",
       "      <th>PAY_4</th>\n",
       "      <th>...</th>\n",
       "      <th>BILL_AMT4</th>\n",
       "      <th>BILL_AMT5</th>\n",
       "      <th>BILL_AMT6</th>\n",
       "      <th>PAY_AMT1</th>\n",
       "      <th>PAY_AMT2</th>\n",
       "      <th>PAY_AMT3</th>\n",
       "      <th>PAY_AMT4</th>\n",
       "      <th>PAY_AMT5</th>\n",
       "      <th>PAY_AMT6</th>\n",
       "      <th>default_0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>20000</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>689.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>90000</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>34</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>14331.0</td>\n",
       "      <td>14948.0</td>\n",
       "      <td>15549.0</td>\n",
       "      <td>1518.0</td>\n",
       "      <td>1500.0</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>5000.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>50000</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>28314.0</td>\n",
       "      <td>28959.0</td>\n",
       "      <td>29547.0</td>\n",
       "      <td>2000.0</td>\n",
       "      <td>2019.0</td>\n",
       "      <td>1200.0</td>\n",
       "      <td>1100.0</td>\n",
       "      <td>1069.0</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>50000</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>57</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>20940.0</td>\n",
       "      <td>19146.0</td>\n",
       "      <td>19131.0</td>\n",
       "      <td>2000.0</td>\n",
       "      <td>36681.0</td>\n",
       "      <td>10000.0</td>\n",
       "      <td>9000.0</td>\n",
       "      <td>689.0</td>\n",
       "      <td>679.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>50000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>37</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>19394.0</td>\n",
       "      <td>19619.0</td>\n",
       "      <td>20024.0</td>\n",
       "      <td>2500.0</td>\n",
       "      <td>1815.0</td>\n",
       "      <td>657.0</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>800.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   ID  LIMIT_BAL  SEX  EDUCATION  MARRIAGE  AGE  PAY_1  PAY_2  PAY_3  PAY_4  \\\n",
       "0   1      20000    2          2         1   24      2      2     -1     -1   \n",
       "1   2      90000    2          2         2   34      0      0      0      0   \n",
       "2   3      50000    2          2         1   37      0      0      0      0   \n",
       "3   4      50000    1          2         1   57     -1      0     -1      0   \n",
       "4   5      50000    1          1         2   37      0      0      0      0   \n",
       "\n",
       "   ...  BILL_AMT4  BILL_AMT5  BILL_AMT6  PAY_AMT1  PAY_AMT2  PAY_AMT3  \\\n",
       "0  ...        0.0        0.0        0.0       0.0     689.0       0.0   \n",
       "1  ...    14331.0    14948.0    15549.0    1518.0    1500.0    1000.0   \n",
       "2  ...    28314.0    28959.0    29547.0    2000.0    2019.0    1200.0   \n",
       "3  ...    20940.0    19146.0    19131.0    2000.0   36681.0   10000.0   \n",
       "4  ...    19394.0    19619.0    20024.0    2500.0    1815.0     657.0   \n",
       "\n",
       "   PAY_AMT4  PAY_AMT5  PAY_AMT6  default_0  \n",
       "0       0.0       0.0       0.0          1  \n",
       "1    1000.0    1000.0    5000.0          0  \n",
       "2    1100.0    1069.0    1000.0          0  \n",
       "3    9000.0     689.0     679.0          0  \n",
       "4    1000.0    1000.0     800.0          0  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking the head of 24.000 customers \n",
    "\n",
    "credit_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.22108333333333333"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# By analysing our dataframe we can realize that default_0 = 1 are the ones who couldn't pay and the default_0 = 0 are the ones who paid\n",
    "# Here we are calculating the percentage of people that we should not be giving them loans and its 22%\n",
    "\n",
    "np.mean(credit_data.default_0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# dropping ID column (unnecessary column)\n",
    "\n",
    "credit_data = credit_data.drop(['ID'], axis = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will try to predict costumer default, by using only some of the key columns in the data set (plotting some of our variables to understand better our data):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 414.75x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# LIMIT_BAL: Total amount of credit line with the bank (including all individual and family/supplementary credit) \n",
    "# and PAY_1: Repayment status 1 month ago – in September ( the values for PAY_1 vary because we have costumers paying and other who don't)\n",
    "\n",
    "sns.relplot(x = \"LIMIT_BAL\", y = \"PAY_1\", hue = \"default_0\", data = credit_data)\n",
    "\n",
    "#plotting the figure\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 414.75x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# running again the same but for PAY_6: repayment status 6 months ago (just to get e better view of comparison between 1 month and 6 months)\n",
    "\n",
    "sns.relplot(x = \"LIMIT_BAL\", y = \"PAY_6\", hue = \"default_0\", data = credit_data)\n",
    "\n",
    "#plotting the figure\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 414.75x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting between bill amount and pay amount for both categories of default\n",
    "\n",
    "sns.relplot(x=\"BILL_AMT1\",y=\"PAY_AMT1\",hue=\"default_0\",data=credit_data)\n",
    "\n",
    "# plotting the figure\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's try to get a heatmap of our variables and see the correlations between them, if there are any."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# corr() method is used to find the pairwise correlation of all columns in the dataframe\n",
    "\n",
    "corr = credit_data.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/gq/bllqt9gj1bq5wzshmf8x9wcw0000gn/T/ipykernel_90298/1252692548.py:19: FutureWarning: this method is deprecated in favour of `Styler.format(precision=..)`\n",
      "  corr.style.background_gradient(cmap, axis=0)\\\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_e19b5 th {\n",
       "  font-size: 7pt;\n",
       "}\n",
       "#T_e19b5 td {\n",
       "  padding: 0em 0em;\n",
       "}\n",
       "#T_e19b5 th:hover {\n",
       "  font-size: 12pt;\n",
       "}\n",
       "#T_e19b5 tr:hover td:hover {\n",
       "  max-width: 200px;\n",
       "  font-size: 12pt;\n",
       "}\n",
       "#T_e19b5_row0_col0, #T_e19b5_row1_col1, #T_e19b5_row2_col2, #T_e19b5_row3_col3, #T_e19b5_row4_col4, #T_e19b5_row5_col5, #T_e19b5_row6_col6, #T_e19b5_row7_col7, #T_e19b5_row8_col8, #T_e19b5_row9_col9, #T_e19b5_row10_col10, #T_e19b5_row11_col11, #T_e19b5_row12_col12, #T_e19b5_row13_col13, #T_e19b5_row14_col14, #T_e19b5_row15_col15, #T_e19b5_row16_col16, #T_e19b5_row17_col17, #T_e19b5_row18_col18, #T_e19b5_row19_col19, #T_e19b5_row20_col20, #T_e19b5_row21_col21, #T_e19b5_row22_col22, #T_e19b5_row23_col23 {\n",
       "  background-color: #4479bb;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row0_col1, #T_e19b5_row23_col0 {\n",
       "  background-color: #dd637c;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row0_col2, #T_e19b5_row0_col5, #T_e19b5_row0_col6, #T_e19b5_row0_col7, #T_e19b5_row0_col8, #T_e19b5_row0_col9, #T_e19b5_row0_col10, #T_e19b5_row0_col23, #T_e19b5_row1_col11, #T_e19b5_row1_col12, #T_e19b5_row1_col13, #T_e19b5_row1_col14, #T_e19b5_row3_col4, #T_e19b5_row3_col13, #T_e19b5_row3_col14, #T_e19b5_row3_col15, #T_e19b5_row3_col16, #T_e19b5_row4_col1, #T_e19b5_row4_col3, #T_e19b5_row5_col17, #T_e19b5_row5_col18, #T_e19b5_row5_col19, #T_e19b5_row5_col20, #T_e19b5_row5_col21, #T_e19b5_row5_col22, #T_e19b5_row6_col0, #T_e19b5_row6_col17, #T_e19b5_row7_col18, #T_e19b5_row8_col19, #T_e19b5_row23_col21 {\n",
       "  background-color: #d73c5b;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row0_col3, #T_e19b5_row1_col4, #T_e19b5_row12_col22, #T_e19b5_row15_col18, #T_e19b5_row22_col19 {\n",
       "  background-color: #e38b9d;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row0_col4, #T_e19b5_row13_col7, #T_e19b5_row15_col7, #T_e19b5_row16_col6, #T_e19b5_row16_col7, #T_e19b5_row21_col0, #T_e19b5_row22_col0 {\n",
       "  background-color: #edccd2;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row0_col11, #T_e19b5_row10_col16, #T_e19b5_row23_col4 {\n",
       "  background-color: #e8aab6;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row0_col12, #T_e19b5_row0_col13, #T_e19b5_row2_col5, #T_e19b5_row2_col8, #T_e19b5_row10_col23, #T_e19b5_row18_col19, #T_e19b5_row19_col18, #T_e19b5_row21_col3 {\n",
       "  background-color: #e7a7b4;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row0_col14, #T_e19b5_row0_col15, #T_e19b5_row2_col7, #T_e19b5_row5_col3, #T_e19b5_row17_col19, #T_e19b5_row20_col4, #T_e19b5_row20_col15, #T_e19b5_row21_col4 {\n",
       "  background-color: #e8abb7;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row0_col16, #T_e19b5_row13_col17, #T_e19b5_row15_col19, #T_e19b5_row17_col12 {\n",
       "  background-color: #e7a8b5;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row0_col17, #T_e19b5_row2_col10, #T_e19b5_row10_col12 {\n",
       "  background-color: #e597a7;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row0_col18, #T_e19b5_row8_col11, #T_e19b5_row11_col22, #T_e19b5_row17_col7 {\n",
       "  background-color: #e38d9e;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row0_col19, #T_e19b5_row6_col11, #T_e19b5_row6_col12, #T_e19b5_row7_col4, #T_e19b5_row8_col16, #T_e19b5_row9_col2, #T_e19b5_row9_col4, #T_e19b5_row10_col4, #T_e19b5_row16_col17, #T_e19b5_row20_col16 {\n",
       "  background-color: #e599a8;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row0_col20, #T_e19b5_row3_col7, #T_e19b5_row8_col12, #T_e19b5_row9_col12, #T_e19b5_row14_col18, #T_e19b5_row17_col14, #T_e19b5_row19_col16, #T_e19b5_row22_col17 {\n",
       "  background-color: #e595a5;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row0_col21, #T_e19b5_row0_col22, #T_e19b5_row6_col4, #T_e19b5_row7_col12, #T_e19b5_row8_col4, #T_e19b5_row9_col13, #T_e19b5_row10_col13, #T_e19b5_row17_col13 {\n",
       "  background-color: #e59aa9;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col0, #T_e19b5_row3_col6, #T_e19b5_row7_col13, #T_e19b5_row7_col14, #T_e19b5_row7_col15, #T_e19b5_row10_col2, #T_e19b5_row18_col20, #T_e19b5_row20_col18, #T_e19b5_row21_col18 {\n",
       "  background-color: #e494a4;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col2, #T_e19b5_row4_col6, #T_e19b5_row13_col2, #T_e19b5_row13_col20, #T_e19b5_row17_col10, #T_e19b5_row18_col15, #T_e19b5_row19_col6, #T_e19b5_row20_col6, #T_e19b5_row22_col8 {\n",
       "  background-color: #e18093;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col3, #T_e19b5_row2_col9, #T_e19b5_row15_col3, #T_e19b5_row15_col17, #T_e19b5_row23_col3 {\n",
       "  background-color: #e6a0ae;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col5, #T_e19b5_row17_col6, #T_e19b5_row21_col5, #T_e19b5_row22_col5 {\n",
       "  background-color: #e0778c;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col6, #T_e19b5_row15_col2, #T_e19b5_row18_col7, #T_e19b5_row20_col8, #T_e19b5_row20_col11 {\n",
       "  background-color: #e17a8f;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col7, #T_e19b5_row4_col8, #T_e19b5_row16_col2, #T_e19b5_row21_col9, #T_e19b5_row22_col10 {\n",
       "  background-color: #e0798d;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col8, #T_e19b5_row16_col22, #T_e19b5_row17_col11, #T_e19b5_row18_col5, #T_e19b5_row19_col5, #T_e19b5_row20_col12, #T_e19b5_row21_col15 {\n",
       "  background-color: #e0768b;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col9, #T_e19b5_row1_col10, #T_e19b5_row2_col23, #T_e19b5_row4_col9, #T_e19b5_row12_col18, #T_e19b5_row18_col2, #T_e19b5_row19_col8, #T_e19b5_row20_col13 {\n",
       "  background-color: #e07389;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col15, #T_e19b5_row2_col16, #T_e19b5_row3_col12, #T_e19b5_row7_col0, #T_e19b5_row9_col20, #T_e19b5_row10_col21, #T_e19b5_row23_col13, #T_e19b5_row23_col17, #T_e19b5_row23_col18, #T_e19b5_row23_col20 {\n",
       "  background-color: #d83e5d;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col16, #T_e19b5_row3_col11, #T_e19b5_row6_col18, #T_e19b5_row23_col22 {\n",
       "  background-color: #d73d5c;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col17, #T_e19b5_row4_col14, #T_e19b5_row4_col15, #T_e19b5_row4_col21, #T_e19b5_row4_col22, #T_e19b5_row9_col19, #T_e19b5_row10_col17, #T_e19b5_row13_col1, #T_e19b5_row14_col1 {\n",
       "  background-color: #db5570;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col18, #T_e19b5_row2_col0, #T_e19b5_row3_col2, #T_e19b5_row3_col19, #T_e19b5_row8_col18 {\n",
       "  background-color: #da516c;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col19, #T_e19b5_row1_col21, #T_e19b5_row2_col11, #T_e19b5_row3_col18, #T_e19b5_row3_col20, #T_e19b5_row3_col21, #T_e19b5_row10_col1, #T_e19b5_row23_col1 {\n",
       "  background-color: #da4e6a;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col20, #T_e19b5_row4_col16, #T_e19b5_row11_col1, #T_e19b5_row12_col1 {\n",
       "  background-color: #db526e;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col22, #T_e19b5_row3_col22, #T_e19b5_row10_col0 {\n",
       "  background-color: #da4d69;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row1_col23, #T_e19b5_row4_col19, #T_e19b5_row17_col1, #T_e19b5_row18_col1, #T_e19b5_row20_col1, #T_e19b5_row21_col1 {\n",
       "  background-color: #dc5c76;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row2_col1 {\n",
       "  background-color: #dd6179;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row2_col3, #T_e19b5_row12_col2, #T_e19b5_row12_col20, #T_e19b5_row17_col9, #T_e19b5_row17_col21, #T_e19b5_row19_col10, #T_e19b5_row20_col21, #T_e19b5_row21_col7, #T_e19b5_row21_col11, #T_e19b5_row22_col7 {\n",
       "  background-color: #e28194;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row2_col4 {\n",
       "  background-color: #eed4d9;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row2_col6, #T_e19b5_row7_col3, #T_e19b5_row8_col3, #T_e19b5_row8_col23, #T_e19b5_row9_col3, #T_e19b5_row10_col3 {\n",
       "  background-color: #e9afbb;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row2_col12, #T_e19b5_row9_col1 {\n",
       "  background-color: #da4b68;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row2_col13, #T_e19b5_row2_col18, #T_e19b5_row9_col22 {\n",
       "  background-color: #d94764;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row2_col14, #T_e19b5_row5_col0, #T_e19b5_row6_col21, #T_e19b5_row6_col22, #T_e19b5_row7_col19, #T_e19b5_row7_col22, #T_e19b5_row8_col0, #T_e19b5_row8_col20, #T_e19b5_row8_col21, #T_e19b5_row9_col21, #T_e19b5_row23_col15 {\n",
       "  background-color: #d84361;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row2_col15, #T_e19b5_row2_col22, #T_e19b5_row6_col19, #T_e19b5_row6_col20, #T_e19b5_row7_col20, #T_e19b5_row7_col21, #T_e19b5_row23_col12, #T_e19b5_row23_col16 {\n",
       "  background-color: #d8415f;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row2_col17, #T_e19b5_row5_col1 {\n",
       "  background-color: #d94a67;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row2_col19, #T_e19b5_row2_col20 {\n",
       "  background-color: #d84462;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row2_col21, #T_e19b5_row23_col11, #T_e19b5_row23_col14, #T_e19b5_row23_col19 {\n",
       "  background-color: #d8405e;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row3_col0, #T_e19b5_row11_col18, #T_e19b5_row17_col2, #T_e19b5_row17_col5, #T_e19b5_row19_col13, #T_e19b5_row20_col9, #T_e19b5_row21_col10 {\n",
       "  background-color: #df7288;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row3_col1, #T_e19b5_row3_col17, #T_e19b5_row8_col17, #T_e19b5_row9_col17, #T_e19b5_row10_col19, #T_e19b5_row17_col23 {\n",
       "  background-color: #db546f;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row3_col5, #T_e19b5_row3_col9, #T_e19b5_row7_col11, #T_e19b5_row9_col11, #T_e19b5_row13_col21, #T_e19b5_row13_col22, #T_e19b5_row14_col22 {\n",
       "  background-color: #e48e9f;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row3_col8, #T_e19b5_row6_col16, #T_e19b5_row17_col15, #T_e19b5_row18_col21 {\n",
       "  background-color: #e491a1;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row3_col10, #T_e19b5_row5_col12, #T_e19b5_row16_col18 {\n",
       "  background-color: #e38a9c;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row3_col23 {\n",
       "  background-color: #dd627b;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row4_col0, #T_e19b5_row18_col13 {\n",
       "  background-color: #eab7c1;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row4_col2, #T_e19b5_row21_col16 {\n",
       "  background-color: #e9b1bc;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row4_col5, #T_e19b5_row14_col2, #T_e19b5_row19_col12, #T_e19b5_row21_col14 {\n",
       "  background-color: #e17c90;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row4_col7, #T_e19b5_row14_col20, #T_e19b5_row18_col6, #T_e19b5_row18_col16, #T_e19b5_row21_col12, #T_e19b5_row22_col9 {\n",
       "  background-color: #e17d91;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row4_col10, #T_e19b5_row19_col2, #T_e19b5_row20_col2, #T_e19b5_row21_col2, #T_e19b5_row22_col2 {\n",
       "  background-color: #df7086;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row4_col11, #T_e19b5_row4_col12, #T_e19b5_row4_col18, #T_e19b5_row10_col20, #T_e19b5_row18_col23, #T_e19b5_row19_col1, #T_e19b5_row19_col23 {\n",
       "  background-color: #dc5973;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row4_col13, #T_e19b5_row7_col17 {\n",
       "  background-color: #db5771;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row4_col17 {\n",
       "  background-color: #dd5f78;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row4_col20, #T_e19b5_row15_col1, #T_e19b5_row16_col1, #T_e19b5_row20_col23, #T_e19b5_row21_col23 {\n",
       "  background-color: #db5872;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row4_col23, #T_e19b5_row20_col14 {\n",
       "  background-color: #df6f85;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row5_col2, #T_e19b5_row5_col4, #T_e19b5_row8_col13, #T_e19b5_row8_col14, #T_e19b5_row8_col15, #T_e19b5_row17_col20 {\n",
       "  background-color: #e69caa;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row5_col6, #T_e19b5_row7_col9 {\n",
       "  background-color: #9bb6d7;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row5_col7 {\n",
       "  background-color: #b8c9e0;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row5_col8 {\n",
       "  background-color: #c3d1e4;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row5_col9 {\n",
       "  background-color: #ccd8e7;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row5_col10 {\n",
       "  background-color: #d8e0eb;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row5_col11, #T_e19b5_row11_col19, #T_e19b5_row11_col21, #T_e19b5_row17_col16, #T_e19b5_row18_col14, #T_e19b5_row19_col22, #T_e19b5_row21_col19 {\n",
       "  background-color: #e3889a;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row5_col13, #T_e19b5_row5_col14, #T_e19b5_row11_col20, #T_e19b5_row12_col21, #T_e19b5_row18_col22, #T_e19b5_row20_col22, #T_e19b5_row21_col6, #T_e19b5_row21_col13, #T_e19b5_row21_col20, #T_e19b5_row21_col22, #T_e19b5_row22_col6, #T_e19b5_row22_col12, #T_e19b5_row22_col14, #T_e19b5_row22_col21, #T_e19b5_row23_col2 {\n",
       "  background-color: #e28497;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row5_col15, #T_e19b5_row14_col21, #T_e19b5_row17_col8, #T_e19b5_row19_col9, #T_e19b5_row19_col21, #T_e19b5_row20_col10, #T_e19b5_row22_col11, #T_e19b5_row22_col13 {\n",
       "  background-color: #e28698;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row5_col16, #T_e19b5_row11_col2, #T_e19b5_row11_col17, #T_e19b5_row18_col9 {\n",
       "  background-color: #e28396;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row5_col23, #T_e19b5_row16_col9 {\n",
       "  background-color: #eed2d7;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row6_col1, #T_e19b5_row7_col1, #T_e19b5_row8_col22, #T_e19b5_row10_col22 {\n",
       "  background-color: #d94663;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row6_col2, #T_e19b5_row9_col16, #T_e19b5_row11_col3, #T_e19b5_row12_col3, #T_e19b5_row13_col3, #T_e19b5_row14_col3, #T_e19b5_row16_col3, #T_e19b5_row19_col20 {\n",
       "  background-color: #e6a1af;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row6_col3, #T_e19b5_row9_col23, #T_e19b5_row10_col15, #T_e19b5_row18_col4, #T_e19b5_row19_col14, #T_e19b5_row22_col4 {\n",
       "  background-color: #e8adb8;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row6_col5 {\n",
       "  background-color: #9eb8d8;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row6_col7, #T_e19b5_row7_col6 {\n",
       "  background-color: #82a4cf;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row6_col8 {\n",
       "  background-color: #a1b9d9;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row6_col9 {\n",
       "  background-color: #adc2dd;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row6_col10 {\n",
       "  background-color: #bacbe1;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row6_col13, #T_e19b5_row6_col14, #T_e19b5_row6_col15, #T_e19b5_row7_col16 {\n",
       "  background-color: #e492a3;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row6_col23, #T_e19b5_row11_col10, #T_e19b5_row13_col5, #T_e19b5_row14_col5, #T_e19b5_row15_col5 {\n",
       "  background-color: #ebbec6;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row7_col2, #T_e19b5_row20_col17 {\n",
       "  background-color: #e69ead;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row7_col5 {\n",
       "  background-color: #b9cae0;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row7_col8, #T_e19b5_row8_col7 {\n",
       "  background-color: #7fa2ce;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row7_col10 {\n",
       "  background-color: #aac0dc;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row7_col23, #T_e19b5_row11_col4, #T_e19b5_row12_col4, #T_e19b5_row13_col4, #T_e19b5_row15_col20 {\n",
       "  background-color: #e9b5bf;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row8_col1, #T_e19b5_row9_col0 {\n",
       "  background-color: #d94865;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row8_col2, #T_e19b5_row19_col15 {\n",
       "  background-color: #e69dab;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row8_col5 {\n",
       "  background-color: #c1d0e3;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row8_col6 {\n",
       "  background-color: #9fb8d8;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row8_col9, #T_e19b5_row10_col9, #T_e19b5_row13_col16, #T_e19b5_row16_col13 {\n",
       "  background-color: #769ccb;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row8_col10 {\n",
       "  background-color: #92b0d4;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row9_col5 {\n",
       "  background-color: #c9d6e6;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row9_col6 {\n",
       "  background-color: #a9bfdb;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row9_col7 {\n",
       "  background-color: #98b3d6;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row9_col8 {\n",
       "  background-color: #759bca;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row9_col10 {\n",
       "  background-color: #779dcb;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row9_col14, #T_e19b5_row14_col17, #T_e19b5_row16_col20, #T_e19b5_row20_col3 {\n",
       "  background-color: #e7a4b1;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row9_col15, #T_e19b5_row17_col3, #T_e19b5_row18_col3, #T_e19b5_row19_col3, #T_e19b5_row22_col3 {\n",
       "  background-color: #e7a5b2;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row9_col18, #T_e19b5_row10_col18 {\n",
       "  background-color: #da506b;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row10_col5 {\n",
       "  background-color: #d4dde9;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row10_col6 {\n",
       "  background-color: #b5c8df;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row10_col7 {\n",
       "  background-color: #a6bdda;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row10_col8 {\n",
       "  background-color: #91afd4;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row10_col11, #T_e19b5_row17_col22 {\n",
       "  background-color: #e48fa0;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row10_col14, #T_e19b5_row16_col19, #T_e19b5_row20_col19 {\n",
       "  background-color: #e7a3b0;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row11_col0, #T_e19b5_row13_col0 {\n",
       "  background-color: #efdde0;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row11_col5, #T_e19b5_row18_col0 {\n",
       "  background-color: #ebbfc8;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row11_col6, #T_e19b5_row12_col6, #T_e19b5_row14_col10 {\n",
       "  background-color: #edd0d6;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row11_col7, #T_e19b5_row13_col10, #T_e19b5_row19_col0 {\n",
       "  background-color: #ecc8cf;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row11_col8 {\n",
       "  background-color: #ebc2ca;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row11_col9, #T_e19b5_row12_col5, #T_e19b5_row13_col18, #T_e19b5_row23_col9 {\n",
       "  background-color: #ebc1c9;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row11_col12, #T_e19b5_row12_col11 {\n",
       "  background-color: #5484c0;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row11_col13, #T_e19b5_row13_col11 {\n",
       "  background-color: #6892c6;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row11_col14, #T_e19b5_row12_col15, #T_e19b5_row14_col11, #T_e19b5_row15_col12 {\n",
       "  background-color: #7299c9;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row11_col15, #T_e19b5_row12_col16, #T_e19b5_row15_col11 {\n",
       "  background-color: #7da0cd;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row11_col16, #T_e19b5_row16_col11 {\n",
       "  background-color: #86a7d0;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row11_col23 {\n",
       "  background-color: #dd657d;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row12_col0 {\n",
       "  background-color: #efdcdf;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row12_col7, #T_e19b5_row13_col8, #T_e19b5_row14_col8, #T_e19b5_row15_col8, #T_e19b5_row23_col7 {\n",
       "  background-color: #edcfd5;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row12_col8 {\n",
       "  background-color: #ecc9d0;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row12_col9, #T_e19b5_row20_col0, #T_e19b5_row23_col8 {\n",
       "  background-color: #ecc6ce;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row12_col10 {\n",
       "  background-color: #ecc3cb;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row12_col13, #T_e19b5_row13_col12 {\n",
       "  background-color: #5c8ac2;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row12_col14, #T_e19b5_row14_col12 {\n",
       "  background-color: #6791c6;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row12_col17, #T_e19b5_row14_col4, #T_e19b5_row15_col4 {\n",
       "  background-color: #e9b4be;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row12_col19, #T_e19b5_row15_col22, #T_e19b5_row18_col8, #T_e19b5_row21_col17, #T_e19b5_row22_col18, #T_e19b5_row22_col20 {\n",
       "  background-color: #e28799;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row12_col23, #T_e19b5_row13_col23, #T_e19b5_row18_col11 {\n",
       "  background-color: #de667e;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row13_col6, #T_e19b5_row14_col6, #T_e19b5_row14_col7, #T_e19b5_row15_col6, #T_e19b5_row16_col8 {\n",
       "  background-color: #edcdd3;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row13_col9 {\n",
       "  background-color: #edcad1;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row13_col14, #T_e19b5_row14_col13 {\n",
       "  background-color: #5e8bc3;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row13_col15, #T_e19b5_row15_col13 {\n",
       "  background-color: #6b94c7;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row13_col19, #T_e19b5_row15_col21, #T_e19b5_row18_col10, #T_e19b5_row19_col7, #T_e19b5_row19_col11, #T_e19b5_row20_col7, #T_e19b5_row21_col8, #T_e19b5_row22_col15 {\n",
       "  background-color: #e17e92;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row14_col0, #T_e19b5_row15_col0 {\n",
       "  background-color: #f0e0e3;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row14_col9, #T_e19b5_row15_col9 {\n",
       "  background-color: #eed3d8;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row14_col15, #T_e19b5_row15_col14 {\n",
       "  background-color: #5887c1;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row14_col16, #T_e19b5_row16_col14 {\n",
       "  background-color: #6690c6;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row14_col19, #T_e19b5_row17_col18, #T_e19b5_row23_col10 {\n",
       "  background-color: #eab8c2;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row14_col23 {\n",
       "  background-color: #de687f;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row15_col10, #T_e19b5_row23_col6 {\n",
       "  background-color: #efd7dc;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row15_col16, #T_e19b5_row16_col15 {\n",
       "  background-color: #5585c0;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row15_col23, #T_e19b5_row16_col23, #T_e19b5_row18_col12 {\n",
       "  background-color: #de6980;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row16_col0 {\n",
       "  background-color: #f0dee2;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row16_col4 {\n",
       "  background-color: #e9b2bd;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row16_col5 {\n",
       "  background-color: #eabcc5;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row16_col10 {\n",
       "  background-color: #eed6db;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row16_col12 {\n",
       "  background-color: #7b9fcd;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row16_col21 {\n",
       "  background-color: #eab9c3;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row17_col0 {\n",
       "  background-color: #ecc5cc;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row17_col4, #T_e19b5_row19_col4, #T_e19b5_row19_col17 {\n",
       "  background-color: #e8aeba;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row18_col17 {\n",
       "  background-color: #eabbc4;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row20_col5 {\n",
       "  background-color: #e0748a;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row22_col1, #T_e19b5_row22_col23 {\n",
       "  background-color: #dc5b75;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row22_col16 {\n",
       "  background-color: #de6c83;\n",
       "  color: #f1f1f1;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "#T_e19b5_row23_col5 {\n",
       "  background-color: #f1e5e7;\n",
       "  color: #000000;\n",
       "  max-width: 80px;\n",
       "  font-size: 10pt;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_e19b5\">\n",
       "  <caption>Correlations Heatmap</caption>\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_e19b5_level0_col0\" class=\"col_heading level0 col0\" >LIMIT_BAL</th>\n",
       "      <th id=\"T_e19b5_level0_col1\" class=\"col_heading level0 col1\" >SEX</th>\n",
       "      <th id=\"T_e19b5_level0_col2\" class=\"col_heading level0 col2\" >EDUCATION</th>\n",
       "      <th id=\"T_e19b5_level0_col3\" class=\"col_heading level0 col3\" >MARRIAGE</th>\n",
       "      <th id=\"T_e19b5_level0_col4\" class=\"col_heading level0 col4\" >AGE</th>\n",
       "      <th id=\"T_e19b5_level0_col5\" class=\"col_heading level0 col5\" >PAY_1</th>\n",
       "      <th id=\"T_e19b5_level0_col6\" class=\"col_heading level0 col6\" >PAY_2</th>\n",
       "      <th id=\"T_e19b5_level0_col7\" class=\"col_heading level0 col7\" >PAY_3</th>\n",
       "      <th id=\"T_e19b5_level0_col8\" class=\"col_heading level0 col8\" >PAY_4</th>\n",
       "      <th id=\"T_e19b5_level0_col9\" class=\"col_heading level0 col9\" >PAY_5</th>\n",
       "      <th id=\"T_e19b5_level0_col10\" class=\"col_heading level0 col10\" >PAY_6</th>\n",
       "      <th id=\"T_e19b5_level0_col11\" class=\"col_heading level0 col11\" >BILL_AMT1</th>\n",
       "      <th id=\"T_e19b5_level0_col12\" class=\"col_heading level0 col12\" >BILL_AMT2</th>\n",
       "      <th id=\"T_e19b5_level0_col13\" class=\"col_heading level0 col13\" >BILL_AMT3</th>\n",
       "      <th id=\"T_e19b5_level0_col14\" class=\"col_heading level0 col14\" >BILL_AMT4</th>\n",
       "      <th id=\"T_e19b5_level0_col15\" class=\"col_heading level0 col15\" >BILL_AMT5</th>\n",
       "      <th id=\"T_e19b5_level0_col16\" class=\"col_heading level0 col16\" >BILL_AMT6</th>\n",
       "      <th id=\"T_e19b5_level0_col17\" class=\"col_heading level0 col17\" >PAY_AMT1</th>\n",
       "      <th id=\"T_e19b5_level0_col18\" class=\"col_heading level0 col18\" >PAY_AMT2</th>\n",
       "      <th id=\"T_e19b5_level0_col19\" class=\"col_heading level0 col19\" >PAY_AMT3</th>\n",
       "      <th id=\"T_e19b5_level0_col20\" class=\"col_heading level0 col20\" >PAY_AMT4</th>\n",
       "      <th id=\"T_e19b5_level0_col21\" class=\"col_heading level0 col21\" >PAY_AMT5</th>\n",
       "      <th id=\"T_e19b5_level0_col22\" class=\"col_heading level0 col22\" >PAY_AMT6</th>\n",
       "      <th id=\"T_e19b5_level0_col23\" class=\"col_heading level0 col23\" >default_0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row0\" class=\"row_heading level0 row0\" >LIMIT_BAL</th>\n",
       "      <td id=\"T_e19b5_row0_col0\" class=\"data row0 col0\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row0_col1\" class=\"data row0 col1\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row0_col2\" class=\"data row0 col2\" >-0.22</td>\n",
       "      <td id=\"T_e19b5_row0_col3\" class=\"data row0 col3\" >-0.10</td>\n",
       "      <td id=\"T_e19b5_row0_col4\" class=\"data row0 col4\" >0.14</td>\n",
       "      <td id=\"T_e19b5_row0_col5\" class=\"data row0 col5\" >-0.27</td>\n",
       "      <td id=\"T_e19b5_row0_col6\" class=\"data row0 col6\" >-0.29</td>\n",
       "      <td id=\"T_e19b5_row0_col7\" class=\"data row0 col7\" >-0.28</td>\n",
       "      <td id=\"T_e19b5_row0_col8\" class=\"data row0 col8\" >-0.27</td>\n",
       "      <td id=\"T_e19b5_row0_col9\" class=\"data row0 col9\" >-0.25</td>\n",
       "      <td id=\"T_e19b5_row0_col10\" class=\"data row0 col10\" >-0.23</td>\n",
       "      <td id=\"T_e19b5_row0_col11\" class=\"data row0 col11\" >0.28</td>\n",
       "      <td id=\"T_e19b5_row0_col12\" class=\"data row0 col12\" >0.27</td>\n",
       "      <td id=\"T_e19b5_row0_col13\" class=\"data row0 col13\" >0.28</td>\n",
       "      <td id=\"T_e19b5_row0_col14\" class=\"data row0 col14\" >0.29</td>\n",
       "      <td id=\"T_e19b5_row0_col15\" class=\"data row0 col15\" >0.29</td>\n",
       "      <td id=\"T_e19b5_row0_col16\" class=\"data row0 col16\" >0.29</td>\n",
       "      <td id=\"T_e19b5_row0_col17\" class=\"data row0 col17\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row0_col18\" class=\"data row0 col18\" >0.17</td>\n",
       "      <td id=\"T_e19b5_row0_col19\" class=\"data row0 col19\" >0.21</td>\n",
       "      <td id=\"T_e19b5_row0_col20\" class=\"data row0 col20\" >0.20</td>\n",
       "      <td id=\"T_e19b5_row0_col21\" class=\"data row0 col21\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row0_col22\" class=\"data row0 col22\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row0_col23\" class=\"data row0 col23\" >-0.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row1\" class=\"row_heading level0 row1\" >SEX</th>\n",
       "      <td id=\"T_e19b5_row1_col0\" class=\"data row1 col0\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row1_col1\" class=\"data row1 col1\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row1_col2\" class=\"data row1 col2\" >0.01</td>\n",
       "      <td id=\"T_e19b5_row1_col3\" class=\"data row1 col3\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row1_col4\" class=\"data row1 col4\" >-0.10</td>\n",
       "      <td id=\"T_e19b5_row1_col5\" class=\"data row1 col5\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row1_col6\" class=\"data row1 col6\" >-0.07</td>\n",
       "      <td id=\"T_e19b5_row1_col7\" class=\"data row1 col7\" >-0.07</td>\n",
       "      <td id=\"T_e19b5_row1_col8\" class=\"data row1 col8\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row1_col9\" class=\"data row1 col9\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row1_col10\" class=\"data row1 col10\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row1_col11\" class=\"data row1 col11\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row1_col12\" class=\"data row1 col12\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row1_col13\" class=\"data row1 col13\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row1_col14\" class=\"data row1 col14\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row1_col15\" class=\"data row1 col15\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row1_col16\" class=\"data row1 col16\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row1_col17\" class=\"data row1 col17\" >0.00</td>\n",
       "      <td id=\"T_e19b5_row1_col18\" class=\"data row1 col18\" >0.00</td>\n",
       "      <td id=\"T_e19b5_row1_col19\" class=\"data row1 col19\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row1_col20\" class=\"data row1 col20\" >0.00</td>\n",
       "      <td id=\"T_e19b5_row1_col21\" class=\"data row1 col21\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row1_col22\" class=\"data row1 col22\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row1_col23\" class=\"data row1 col23\" >-0.04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row2\" class=\"row_heading level0 row2\" >EDUCATION</th>\n",
       "      <td id=\"T_e19b5_row2_col0\" class=\"data row2 col0\" >-0.22</td>\n",
       "      <td id=\"T_e19b5_row2_col1\" class=\"data row2 col1\" >0.01</td>\n",
       "      <td id=\"T_e19b5_row2_col2\" class=\"data row2 col2\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row2_col3\" class=\"data row2 col3\" >-0.15</td>\n",
       "      <td id=\"T_e19b5_row2_col4\" class=\"data row2 col4\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row2_col5\" class=\"data row2 col5\" >0.10</td>\n",
       "      <td id=\"T_e19b5_row2_col6\" class=\"data row2 col6\" >0.12</td>\n",
       "      <td id=\"T_e19b5_row2_col7\" class=\"data row2 col7\" >0.11</td>\n",
       "      <td id=\"T_e19b5_row2_col8\" class=\"data row2 col8\" >0.11</td>\n",
       "      <td id=\"T_e19b5_row2_col9\" class=\"data row2 col9\" >0.10</td>\n",
       "      <td id=\"T_e19b5_row2_col10\" class=\"data row2 col10\" >0.08</td>\n",
       "      <td id=\"T_e19b5_row2_col11\" class=\"data row2 col11\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row2_col12\" class=\"data row2 col12\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row2_col13\" class=\"data row2 col13\" >0.01</td>\n",
       "      <td id=\"T_e19b5_row2_col14\" class=\"data row2 col14\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row2_col15\" class=\"data row2 col15\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row2_col16\" class=\"data row2 col16\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row2_col17\" class=\"data row2 col17\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row2_col18\" class=\"data row2 col18\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row2_col19\" class=\"data row2 col19\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row2_col20\" class=\"data row2 col20\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row2_col21\" class=\"data row2 col21\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row2_col22\" class=\"data row2 col22\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row2_col23\" class=\"data row2 col23\" >0.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row3\" class=\"row_heading level0 row3\" >MARRIAGE</th>\n",
       "      <td id=\"T_e19b5_row3_col0\" class=\"data row3 col0\" >-0.10</td>\n",
       "      <td id=\"T_e19b5_row3_col1\" class=\"data row3 col1\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row3_col2\" class=\"data row3 col2\" >-0.15</td>\n",
       "      <td id=\"T_e19b5_row3_col3\" class=\"data row3 col3\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row3_col4\" class=\"data row3 col4\" >-0.42</td>\n",
       "      <td id=\"T_e19b5_row3_col5\" class=\"data row3 col5\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row3_col6\" class=\"data row3 col6\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row3_col7\" class=\"data row3 col7\" >0.03</td>\n",
       "      <td id=\"T_e19b5_row3_col8\" class=\"data row3 col8\" >0.03</td>\n",
       "      <td id=\"T_e19b5_row3_col9\" class=\"data row3 col9\" >0.04</td>\n",
       "      <td id=\"T_e19b5_row3_col10\" class=\"data row3 col10\" >0.04</td>\n",
       "      <td id=\"T_e19b5_row3_col11\" class=\"data row3 col11\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row3_col12\" class=\"data row3 col12\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row3_col13\" class=\"data row3 col13\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row3_col14\" class=\"data row3 col14\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row3_col15\" class=\"data row3 col15\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row3_col16\" class=\"data row3 col16\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row3_col17\" class=\"data row3 col17\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row3_col18\" class=\"data row3 col18\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row3_col19\" class=\"data row3 col19\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row3_col20\" class=\"data row3 col20\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row3_col21\" class=\"data row3 col21\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row3_col22\" class=\"data row3 col22\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row3_col23\" class=\"data row3 col23\" >-0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row4\" class=\"row_heading level0 row4\" >AGE</th>\n",
       "      <td id=\"T_e19b5_row4_col0\" class=\"data row4 col0\" >0.14</td>\n",
       "      <td id=\"T_e19b5_row4_col1\" class=\"data row4 col1\" >-0.10</td>\n",
       "      <td id=\"T_e19b5_row4_col2\" class=\"data row4 col2\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row4_col3\" class=\"data row4 col3\" >-0.42</td>\n",
       "      <td id=\"T_e19b5_row4_col4\" class=\"data row4 col4\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row4_col5\" class=\"data row4 col5\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row4_col6\" class=\"data row4 col6\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row4_col7\" class=\"data row4 col7\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row4_col8\" class=\"data row4 col8\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row4_col9\" class=\"data row4 col9\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row4_col10\" class=\"data row4 col10\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row4_col11\" class=\"data row4 col11\" >0.06</td>\n",
       "      <td id=\"T_e19b5_row4_col12\" class=\"data row4 col12\" >0.06</td>\n",
       "      <td id=\"T_e19b5_row4_col13\" class=\"data row4 col13\" >0.05</td>\n",
       "      <td id=\"T_e19b5_row4_col14\" class=\"data row4 col14\" >0.05</td>\n",
       "      <td id=\"T_e19b5_row4_col15\" class=\"data row4 col15\" >0.05</td>\n",
       "      <td id=\"T_e19b5_row4_col16\" class=\"data row4 col16\" >0.05</td>\n",
       "      <td id=\"T_e19b5_row4_col17\" class=\"data row4 col17\" >0.03</td>\n",
       "      <td id=\"T_e19b5_row4_col18\" class=\"data row4 col18\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row4_col19\" class=\"data row4 col19\" >0.03</td>\n",
       "      <td id=\"T_e19b5_row4_col20\" class=\"data row4 col20\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row4_col21\" class=\"data row4 col21\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row4_col22\" class=\"data row4 col22\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row4_col23\" class=\"data row4 col23\" >0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row5\" class=\"row_heading level0 row5\" >PAY_1</th>\n",
       "      <td id=\"T_e19b5_row5_col0\" class=\"data row5 col0\" >-0.27</td>\n",
       "      <td id=\"T_e19b5_row5_col1\" class=\"data row5 col1\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row5_col2\" class=\"data row5 col2\" >0.10</td>\n",
       "      <td id=\"T_e19b5_row5_col3\" class=\"data row5 col3\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row5_col4\" class=\"data row5 col4\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row5_col5\" class=\"data row5 col5\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row5_col6\" class=\"data row5 col6\" >0.67</td>\n",
       "      <td id=\"T_e19b5_row5_col7\" class=\"data row5 col7\" >0.57</td>\n",
       "      <td id=\"T_e19b5_row5_col8\" class=\"data row5 col8\" >0.54</td>\n",
       "      <td id=\"T_e19b5_row5_col9\" class=\"data row5 col9\" >0.51</td>\n",
       "      <td id=\"T_e19b5_row5_col10\" class=\"data row5 col10\" >0.47</td>\n",
       "      <td id=\"T_e19b5_row5_col11\" class=\"data row5 col11\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row5_col12\" class=\"data row5 col12\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row5_col13\" class=\"data row5 col13\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row5_col14\" class=\"data row5 col14\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row5_col15\" class=\"data row5 col15\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row5_col16\" class=\"data row5 col16\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row5_col17\" class=\"data row5 col17\" >-0.08</td>\n",
       "      <td id=\"T_e19b5_row5_col18\" class=\"data row5 col18\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row5_col19\" class=\"data row5 col19\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row5_col20\" class=\"data row5 col20\" >-0.07</td>\n",
       "      <td id=\"T_e19b5_row5_col21\" class=\"data row5 col21\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row5_col22\" class=\"data row5 col22\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row5_col23\" class=\"data row5 col23\" >0.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row6\" class=\"row_heading level0 row6\" >PAY_2</th>\n",
       "      <td id=\"T_e19b5_row6_col0\" class=\"data row6 col0\" >-0.29</td>\n",
       "      <td id=\"T_e19b5_row6_col1\" class=\"data row6 col1\" >-0.07</td>\n",
       "      <td id=\"T_e19b5_row6_col2\" class=\"data row6 col2\" >0.12</td>\n",
       "      <td id=\"T_e19b5_row6_col3\" class=\"data row6 col3\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row6_col4\" class=\"data row6 col4\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row6_col5\" class=\"data row6 col5\" >0.67</td>\n",
       "      <td id=\"T_e19b5_row6_col6\" class=\"data row6 col6\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row6_col7\" class=\"data row6 col7\" >0.77</td>\n",
       "      <td id=\"T_e19b5_row6_col8\" class=\"data row6 col8\" >0.66</td>\n",
       "      <td id=\"T_e19b5_row6_col9\" class=\"data row6 col9\" >0.62</td>\n",
       "      <td id=\"T_e19b5_row6_col10\" class=\"data row6 col10\" >0.58</td>\n",
       "      <td id=\"T_e19b5_row6_col11\" class=\"data row6 col11\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row6_col12\" class=\"data row6 col12\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row6_col13\" class=\"data row6 col13\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row6_col14\" class=\"data row6 col14\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row6_col15\" class=\"data row6 col15\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row6_col16\" class=\"data row6 col16\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row6_col17\" class=\"data row6 col17\" >-0.08</td>\n",
       "      <td id=\"T_e19b5_row6_col18\" class=\"data row6 col18\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row6_col19\" class=\"data row6 col19\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row6_col20\" class=\"data row6 col20\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row6_col21\" class=\"data row6 col21\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row6_col22\" class=\"data row6 col22\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row6_col23\" class=\"data row6 col23\" >0.26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row7\" class=\"row_heading level0 row7\" >PAY_3</th>\n",
       "      <td id=\"T_e19b5_row7_col0\" class=\"data row7 col0\" >-0.28</td>\n",
       "      <td id=\"T_e19b5_row7_col1\" class=\"data row7 col1\" >-0.07</td>\n",
       "      <td id=\"T_e19b5_row7_col2\" class=\"data row7 col2\" >0.11</td>\n",
       "      <td id=\"T_e19b5_row7_col3\" class=\"data row7 col3\" >0.03</td>\n",
       "      <td id=\"T_e19b5_row7_col4\" class=\"data row7 col4\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row7_col5\" class=\"data row7 col5\" >0.57</td>\n",
       "      <td id=\"T_e19b5_row7_col6\" class=\"data row7 col6\" >0.77</td>\n",
       "      <td id=\"T_e19b5_row7_col7\" class=\"data row7 col7\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row7_col8\" class=\"data row7 col8\" >0.78</td>\n",
       "      <td id=\"T_e19b5_row7_col9\" class=\"data row7 col9\" >0.69</td>\n",
       "      <td id=\"T_e19b5_row7_col10\" class=\"data row7 col10\" >0.63</td>\n",
       "      <td id=\"T_e19b5_row7_col11\" class=\"data row7 col11\" >0.21</td>\n",
       "      <td id=\"T_e19b5_row7_col12\" class=\"data row7 col12\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row7_col13\" class=\"data row7 col13\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row7_col14\" class=\"data row7 col14\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row7_col15\" class=\"data row7 col15\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row7_col16\" class=\"data row7 col16\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row7_col17\" class=\"data row7 col17\" >0.00</td>\n",
       "      <td id=\"T_e19b5_row7_col18\" class=\"data row7 col18\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row7_col19\" class=\"data row7 col19\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row7_col20\" class=\"data row7 col20\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row7_col21\" class=\"data row7 col21\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row7_col22\" class=\"data row7 col22\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row7_col23\" class=\"data row7 col23\" >0.23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row8\" class=\"row_heading level0 row8\" >PAY_4</th>\n",
       "      <td id=\"T_e19b5_row8_col0\" class=\"data row8 col0\" >-0.27</td>\n",
       "      <td id=\"T_e19b5_row8_col1\" class=\"data row8 col1\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row8_col2\" class=\"data row8 col2\" >0.11</td>\n",
       "      <td id=\"T_e19b5_row8_col3\" class=\"data row8 col3\" >0.03</td>\n",
       "      <td id=\"T_e19b5_row8_col4\" class=\"data row8 col4\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row8_col5\" class=\"data row8 col5\" >0.54</td>\n",
       "      <td id=\"T_e19b5_row8_col6\" class=\"data row8 col6\" >0.66</td>\n",
       "      <td id=\"T_e19b5_row8_col7\" class=\"data row8 col7\" >0.78</td>\n",
       "      <td id=\"T_e19b5_row8_col8\" class=\"data row8 col8\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row8_col9\" class=\"data row8 col9\" >0.82</td>\n",
       "      <td id=\"T_e19b5_row8_col10\" class=\"data row8 col10\" >0.72</td>\n",
       "      <td id=\"T_e19b5_row8_col11\" class=\"data row8 col11\" >0.20</td>\n",
       "      <td id=\"T_e19b5_row8_col12\" class=\"data row8 col12\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row8_col13\" class=\"data row8 col13\" >0.25</td>\n",
       "      <td id=\"T_e19b5_row8_col14\" class=\"data row8 col14\" >0.25</td>\n",
       "      <td id=\"T_e19b5_row8_col15\" class=\"data row8 col15\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row8_col16\" class=\"data row8 col16\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row8_col17\" class=\"data row8 col17\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row8_col18\" class=\"data row8 col18\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row8_col19\" class=\"data row8 col19\" >-0.07</td>\n",
       "      <td id=\"T_e19b5_row8_col20\" class=\"data row8 col20\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row8_col21\" class=\"data row8 col21\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row8_col22\" class=\"data row8 col22\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row8_col23\" class=\"data row8 col23\" >0.22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row9\" class=\"row_heading level0 row9\" >PAY_5</th>\n",
       "      <td id=\"T_e19b5_row9_col0\" class=\"data row9 col0\" >-0.25</td>\n",
       "      <td id=\"T_e19b5_row9_col1\" class=\"data row9 col1\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row9_col2\" class=\"data row9 col2\" >0.10</td>\n",
       "      <td id=\"T_e19b5_row9_col3\" class=\"data row9 col3\" >0.04</td>\n",
       "      <td id=\"T_e19b5_row9_col4\" class=\"data row9 col4\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row9_col5\" class=\"data row9 col5\" >0.51</td>\n",
       "      <td id=\"T_e19b5_row9_col6\" class=\"data row9 col6\" >0.62</td>\n",
       "      <td id=\"T_e19b5_row9_col7\" class=\"data row9 col7\" >0.69</td>\n",
       "      <td id=\"T_e19b5_row9_col8\" class=\"data row9 col8\" >0.82</td>\n",
       "      <td id=\"T_e19b5_row9_col9\" class=\"data row9 col9\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row9_col10\" class=\"data row9 col10\" >0.82</td>\n",
       "      <td id=\"T_e19b5_row9_col11\" class=\"data row9 col11\" >0.21</td>\n",
       "      <td id=\"T_e19b5_row9_col12\" class=\"data row9 col12\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row9_col13\" class=\"data row9 col13\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row9_col14\" class=\"data row9 col14\" >0.27</td>\n",
       "      <td id=\"T_e19b5_row9_col15\" class=\"data row9 col15\" >0.27</td>\n",
       "      <td id=\"T_e19b5_row9_col16\" class=\"data row9 col16\" >0.26</td>\n",
       "      <td id=\"T_e19b5_row9_col17\" class=\"data row9 col17\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row9_col18\" class=\"data row9 col18\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row9_col19\" class=\"data row9 col19\" >0.01</td>\n",
       "      <td id=\"T_e19b5_row9_col20\" class=\"data row9 col20\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row9_col21\" class=\"data row9 col21\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row9_col22\" class=\"data row9 col22\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row9_col23\" class=\"data row9 col23\" >0.21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row10\" class=\"row_heading level0 row10\" >PAY_6</th>\n",
       "      <td id=\"T_e19b5_row10_col0\" class=\"data row10 col0\" >-0.23</td>\n",
       "      <td id=\"T_e19b5_row10_col1\" class=\"data row10 col1\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row10_col2\" class=\"data row10 col2\" >0.08</td>\n",
       "      <td id=\"T_e19b5_row10_col3\" class=\"data row10 col3\" >0.04</td>\n",
       "      <td id=\"T_e19b5_row10_col4\" class=\"data row10 col4\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row10_col5\" class=\"data row10 col5\" >0.47</td>\n",
       "      <td id=\"T_e19b5_row10_col6\" class=\"data row10 col6\" >0.58</td>\n",
       "      <td id=\"T_e19b5_row10_col7\" class=\"data row10 col7\" >0.63</td>\n",
       "      <td id=\"T_e19b5_row10_col8\" class=\"data row10 col8\" >0.72</td>\n",
       "      <td id=\"T_e19b5_row10_col9\" class=\"data row10 col9\" >0.82</td>\n",
       "      <td id=\"T_e19b5_row10_col10\" class=\"data row10 col10\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row10_col11\" class=\"data row10 col11\" >0.21</td>\n",
       "      <td id=\"T_e19b5_row10_col12\" class=\"data row10 col12\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row10_col13\" class=\"data row10 col13\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row10_col14\" class=\"data row10 col14\" >0.27</td>\n",
       "      <td id=\"T_e19b5_row10_col15\" class=\"data row10 col15\" >0.29</td>\n",
       "      <td id=\"T_e19b5_row10_col16\" class=\"data row10 col16\" >0.29</td>\n",
       "      <td id=\"T_e19b5_row10_col17\" class=\"data row10 col17\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row10_col18\" class=\"data row10 col18\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row10_col19\" class=\"data row10 col19\" >0.01</td>\n",
       "      <td id=\"T_e19b5_row10_col20\" class=\"data row10 col20\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row10_col21\" class=\"data row10 col21\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row10_col22\" class=\"data row10 col22\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row10_col23\" class=\"data row10 col23\" >0.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row11\" class=\"row_heading level0 row11\" >BILL_AMT1</th>\n",
       "      <td id=\"T_e19b5_row11_col0\" class=\"data row11 col0\" >0.28</td>\n",
       "      <td id=\"T_e19b5_row11_col1\" class=\"data row11 col1\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row11_col2\" class=\"data row11 col2\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row11_col3\" class=\"data row11 col3\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row11_col4\" class=\"data row11 col4\" >0.06</td>\n",
       "      <td id=\"T_e19b5_row11_col5\" class=\"data row11 col5\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row11_col6\" class=\"data row11 col6\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row11_col7\" class=\"data row11 col7\" >0.21</td>\n",
       "      <td id=\"T_e19b5_row11_col8\" class=\"data row11 col8\" >0.20</td>\n",
       "      <td id=\"T_e19b5_row11_col9\" class=\"data row11 col9\" >0.21</td>\n",
       "      <td id=\"T_e19b5_row11_col10\" class=\"data row11 col10\" >0.21</td>\n",
       "      <td id=\"T_e19b5_row11_col11\" class=\"data row11 col11\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row11_col12\" class=\"data row11 col12\" >0.95</td>\n",
       "      <td id=\"T_e19b5_row11_col13\" class=\"data row11 col13\" >0.89</td>\n",
       "      <td id=\"T_e19b5_row11_col14\" class=\"data row11 col14\" >0.86</td>\n",
       "      <td id=\"T_e19b5_row11_col15\" class=\"data row11 col15\" >0.83</td>\n",
       "      <td id=\"T_e19b5_row11_col16\" class=\"data row11 col16\" >0.80</td>\n",
       "      <td id=\"T_e19b5_row11_col17\" class=\"data row11 col17\" >0.14</td>\n",
       "      <td id=\"T_e19b5_row11_col18\" class=\"data row11 col18\" >0.09</td>\n",
       "      <td id=\"T_e19b5_row11_col19\" class=\"data row11 col19\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row11_col20\" class=\"data row11 col20\" >0.15</td>\n",
       "      <td id=\"T_e19b5_row11_col21\" class=\"data row11 col21\" >0.17</td>\n",
       "      <td id=\"T_e19b5_row11_col22\" class=\"data row11 col22\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row11_col23\" class=\"data row11 col23\" >-0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row12\" class=\"row_heading level0 row12\" >BILL_AMT2</th>\n",
       "      <td id=\"T_e19b5_row12_col0\" class=\"data row12 col0\" >0.27</td>\n",
       "      <td id=\"T_e19b5_row12_col1\" class=\"data row12 col1\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row12_col2\" class=\"data row12 col2\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row12_col3\" class=\"data row12 col3\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row12_col4\" class=\"data row12 col4\" >0.06</td>\n",
       "      <td id=\"T_e19b5_row12_col5\" class=\"data row12 col5\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row12_col6\" class=\"data row12 col6\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row12_col7\" class=\"data row12 col7\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row12_col8\" class=\"data row12 col8\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row12_col9\" class=\"data row12 col9\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row12_col10\" class=\"data row12 col10\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row12_col11\" class=\"data row12 col11\" >0.95</td>\n",
       "      <td id=\"T_e19b5_row12_col12\" class=\"data row12 col12\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row12_col13\" class=\"data row12 col13\" >0.92</td>\n",
       "      <td id=\"T_e19b5_row12_col14\" class=\"data row12 col14\" >0.89</td>\n",
       "      <td id=\"T_e19b5_row12_col15\" class=\"data row12 col15\" >0.86</td>\n",
       "      <td id=\"T_e19b5_row12_col16\" class=\"data row12 col16\" >0.83</td>\n",
       "      <td id=\"T_e19b5_row12_col17\" class=\"data row12 col17\" >0.28</td>\n",
       "      <td id=\"T_e19b5_row12_col18\" class=\"data row12 col18\" >0.10</td>\n",
       "      <td id=\"T_e19b5_row12_col19\" class=\"data row12 col19\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row12_col20\" class=\"data row12 col20\" >0.14</td>\n",
       "      <td id=\"T_e19b5_row12_col21\" class=\"data row12 col21\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row12_col22\" class=\"data row12 col22\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row12_col23\" class=\"data row12 col23\" >-0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row13\" class=\"row_heading level0 row13\" >BILL_AMT3</th>\n",
       "      <td id=\"T_e19b5_row13_col0\" class=\"data row13 col0\" >0.28</td>\n",
       "      <td id=\"T_e19b5_row13_col1\" class=\"data row13 col1\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row13_col2\" class=\"data row13 col2\" >0.01</td>\n",
       "      <td id=\"T_e19b5_row13_col3\" class=\"data row13 col3\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row13_col4\" class=\"data row13 col4\" >0.05</td>\n",
       "      <td id=\"T_e19b5_row13_col5\" class=\"data row13 col5\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row13_col6\" class=\"data row13 col6\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row13_col7\" class=\"data row13 col7\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row13_col8\" class=\"data row13 col8\" >0.25</td>\n",
       "      <td id=\"T_e19b5_row13_col9\" class=\"data row13 col9\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row13_col10\" class=\"data row13 col10\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row13_col11\" class=\"data row13 col11\" >0.89</td>\n",
       "      <td id=\"T_e19b5_row13_col12\" class=\"data row13 col12\" >0.92</td>\n",
       "      <td id=\"T_e19b5_row13_col13\" class=\"data row13 col13\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row13_col14\" class=\"data row13 col14\" >0.92</td>\n",
       "      <td id=\"T_e19b5_row13_col15\" class=\"data row13 col15\" >0.88</td>\n",
       "      <td id=\"T_e19b5_row13_col16\" class=\"data row13 col16\" >0.85</td>\n",
       "      <td id=\"T_e19b5_row13_col17\" class=\"data row13 col17\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row13_col18\" class=\"data row13 col18\" >0.32</td>\n",
       "      <td id=\"T_e19b5_row13_col19\" class=\"data row13 col19\" >0.13</td>\n",
       "      <td id=\"T_e19b5_row13_col20\" class=\"data row13 col20\" >0.14</td>\n",
       "      <td id=\"T_e19b5_row13_col21\" class=\"data row13 col21\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row13_col22\" class=\"data row13 col22\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row13_col23\" class=\"data row13 col23\" >-0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row14\" class=\"row_heading level0 row14\" >BILL_AMT4</th>\n",
       "      <td id=\"T_e19b5_row14_col0\" class=\"data row14 col0\" >0.29</td>\n",
       "      <td id=\"T_e19b5_row14_col1\" class=\"data row14 col1\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row14_col2\" class=\"data row14 col2\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row14_col3\" class=\"data row14 col3\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row14_col4\" class=\"data row14 col4\" >0.05</td>\n",
       "      <td id=\"T_e19b5_row14_col5\" class=\"data row14 col5\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row14_col6\" class=\"data row14 col6\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row14_col7\" class=\"data row14 col7\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row14_col8\" class=\"data row14 col8\" >0.25</td>\n",
       "      <td id=\"T_e19b5_row14_col9\" class=\"data row14 col9\" >0.27</td>\n",
       "      <td id=\"T_e19b5_row14_col10\" class=\"data row14 col10\" >0.27</td>\n",
       "      <td id=\"T_e19b5_row14_col11\" class=\"data row14 col11\" >0.86</td>\n",
       "      <td id=\"T_e19b5_row14_col12\" class=\"data row14 col12\" >0.89</td>\n",
       "      <td id=\"T_e19b5_row14_col13\" class=\"data row14 col13\" >0.92</td>\n",
       "      <td id=\"T_e19b5_row14_col14\" class=\"data row14 col14\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row14_col15\" class=\"data row14 col15\" >0.94</td>\n",
       "      <td id=\"T_e19b5_row14_col16\" class=\"data row14 col16\" >0.90</td>\n",
       "      <td id=\"T_e19b5_row14_col17\" class=\"data row14 col17\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row14_col18\" class=\"data row14 col18\" >0.20</td>\n",
       "      <td id=\"T_e19b5_row14_col19\" class=\"data row14 col19\" >0.30</td>\n",
       "      <td id=\"T_e19b5_row14_col20\" class=\"data row14 col20\" >0.13</td>\n",
       "      <td id=\"T_e19b5_row14_col21\" class=\"data row14 col21\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row14_col22\" class=\"data row14 col22\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row14_col23\" class=\"data row14 col23\" >-0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row15\" class=\"row_heading level0 row15\" >BILL_AMT5</th>\n",
       "      <td id=\"T_e19b5_row15_col0\" class=\"data row15 col0\" >0.29</td>\n",
       "      <td id=\"T_e19b5_row15_col1\" class=\"data row15 col1\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row15_col2\" class=\"data row15 col2\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row15_col3\" class=\"data row15 col3\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row15_col4\" class=\"data row15 col4\" >0.05</td>\n",
       "      <td id=\"T_e19b5_row15_col5\" class=\"data row15 col5\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row15_col6\" class=\"data row15 col6\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row15_col7\" class=\"data row15 col7\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row15_col8\" class=\"data row15 col8\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row15_col9\" class=\"data row15 col9\" >0.27</td>\n",
       "      <td id=\"T_e19b5_row15_col10\" class=\"data row15 col10\" >0.29</td>\n",
       "      <td id=\"T_e19b5_row15_col11\" class=\"data row15 col11\" >0.83</td>\n",
       "      <td id=\"T_e19b5_row15_col12\" class=\"data row15 col12\" >0.86</td>\n",
       "      <td id=\"T_e19b5_row15_col13\" class=\"data row15 col13\" >0.88</td>\n",
       "      <td id=\"T_e19b5_row15_col14\" class=\"data row15 col14\" >0.94</td>\n",
       "      <td id=\"T_e19b5_row15_col15\" class=\"data row15 col15\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row15_col16\" class=\"data row15 col16\" >0.95</td>\n",
       "      <td id=\"T_e19b5_row15_col17\" class=\"data row15 col17\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row15_col18\" class=\"data row15 col18\" >0.17</td>\n",
       "      <td id=\"T_e19b5_row15_col19\" class=\"data row15 col19\" >0.25</td>\n",
       "      <td id=\"T_e19b5_row15_col20\" class=\"data row15 col20\" >0.29</td>\n",
       "      <td id=\"T_e19b5_row15_col21\" class=\"data row15 col21\" >0.14</td>\n",
       "      <td id=\"T_e19b5_row15_col22\" class=\"data row15 col22\" >0.17</td>\n",
       "      <td id=\"T_e19b5_row15_col23\" class=\"data row15 col23\" >-0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row16\" class=\"row_heading level0 row16\" >BILL_AMT6</th>\n",
       "      <td id=\"T_e19b5_row16_col0\" class=\"data row16 col0\" >0.29</td>\n",
       "      <td id=\"T_e19b5_row16_col1\" class=\"data row16 col1\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row16_col2\" class=\"data row16 col2\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row16_col3\" class=\"data row16 col3\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row16_col4\" class=\"data row16 col4\" >0.05</td>\n",
       "      <td id=\"T_e19b5_row16_col5\" class=\"data row16 col5\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row16_col6\" class=\"data row16 col6\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row16_col7\" class=\"data row16 col7\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row16_col8\" class=\"data row16 col8\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row16_col9\" class=\"data row16 col9\" >0.26</td>\n",
       "      <td id=\"T_e19b5_row16_col10\" class=\"data row16 col10\" >0.29</td>\n",
       "      <td id=\"T_e19b5_row16_col11\" class=\"data row16 col11\" >0.80</td>\n",
       "      <td id=\"T_e19b5_row16_col12\" class=\"data row16 col12\" >0.83</td>\n",
       "      <td id=\"T_e19b5_row16_col13\" class=\"data row16 col13\" >0.85</td>\n",
       "      <td id=\"T_e19b5_row16_col14\" class=\"data row16 col14\" >0.90</td>\n",
       "      <td id=\"T_e19b5_row16_col15\" class=\"data row16 col15\" >0.95</td>\n",
       "      <td id=\"T_e19b5_row16_col16\" class=\"data row16 col16\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row16_col17\" class=\"data row16 col17\" >0.20</td>\n",
       "      <td id=\"T_e19b5_row16_col18\" class=\"data row16 col18\" >0.17</td>\n",
       "      <td id=\"T_e19b5_row16_col19\" class=\"data row16 col19\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row16_col20\" class=\"data row16 col20\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row16_col21\" class=\"data row16 col21\" >0.31</td>\n",
       "      <td id=\"T_e19b5_row16_col22\" class=\"data row16 col22\" >0.12</td>\n",
       "      <td id=\"T_e19b5_row16_col23\" class=\"data row16 col23\" >-0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row17\" class=\"row_heading level0 row17\" >PAY_AMT1</th>\n",
       "      <td id=\"T_e19b5_row17_col0\" class=\"data row17 col0\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row17_col1\" class=\"data row17 col1\" >0.00</td>\n",
       "      <td id=\"T_e19b5_row17_col2\" class=\"data row17 col2\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row17_col3\" class=\"data row17 col3\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row17_col4\" class=\"data row17 col4\" >0.03</td>\n",
       "      <td id=\"T_e19b5_row17_col5\" class=\"data row17 col5\" >-0.08</td>\n",
       "      <td id=\"T_e19b5_row17_col6\" class=\"data row17 col6\" >-0.08</td>\n",
       "      <td id=\"T_e19b5_row17_col7\" class=\"data row17 col7\" >0.00</td>\n",
       "      <td id=\"T_e19b5_row17_col8\" class=\"data row17 col8\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row17_col9\" class=\"data row17 col9\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row17_col10\" class=\"data row17 col10\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row17_col11\" class=\"data row17 col11\" >0.14</td>\n",
       "      <td id=\"T_e19b5_row17_col12\" class=\"data row17 col12\" >0.28</td>\n",
       "      <td id=\"T_e19b5_row17_col13\" class=\"data row17 col13\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row17_col14\" class=\"data row17 col14\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row17_col15\" class=\"data row17 col15\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row17_col16\" class=\"data row17 col16\" >0.20</td>\n",
       "      <td id=\"T_e19b5_row17_col17\" class=\"data row17 col17\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row17_col18\" class=\"data row17 col18\" >0.30</td>\n",
       "      <td id=\"T_e19b5_row17_col19\" class=\"data row17 col19\" >0.26</td>\n",
       "      <td id=\"T_e19b5_row17_col20\" class=\"data row17 col20\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row17_col21\" class=\"data row17 col21\" >0.15</td>\n",
       "      <td id=\"T_e19b5_row17_col22\" class=\"data row17 col22\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row17_col23\" class=\"data row17 col23\" >-0.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row18\" class=\"row_heading level0 row18\" >PAY_AMT2</th>\n",
       "      <td id=\"T_e19b5_row18_col0\" class=\"data row18 col0\" >0.17</td>\n",
       "      <td id=\"T_e19b5_row18_col1\" class=\"data row18 col1\" >0.00</td>\n",
       "      <td id=\"T_e19b5_row18_col2\" class=\"data row18 col2\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row18_col3\" class=\"data row18 col3\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row18_col4\" class=\"data row18 col4\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row18_col5\" class=\"data row18 col5\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row18_col6\" class=\"data row18 col6\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row18_col7\" class=\"data row18 col7\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row18_col8\" class=\"data row18 col8\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row18_col9\" class=\"data row18 col9\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row18_col10\" class=\"data row18 col10\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row18_col11\" class=\"data row18 col11\" >0.09</td>\n",
       "      <td id=\"T_e19b5_row18_col12\" class=\"data row18 col12\" >0.10</td>\n",
       "      <td id=\"T_e19b5_row18_col13\" class=\"data row18 col13\" >0.32</td>\n",
       "      <td id=\"T_e19b5_row18_col14\" class=\"data row18 col14\" >0.20</td>\n",
       "      <td id=\"T_e19b5_row18_col15\" class=\"data row18 col15\" >0.17</td>\n",
       "      <td id=\"T_e19b5_row18_col16\" class=\"data row18 col16\" >0.17</td>\n",
       "      <td id=\"T_e19b5_row18_col17\" class=\"data row18 col17\" >0.30</td>\n",
       "      <td id=\"T_e19b5_row18_col18\" class=\"data row18 col18\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row18_col19\" class=\"data row18 col19\" >0.25</td>\n",
       "      <td id=\"T_e19b5_row18_col20\" class=\"data row18 col20\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row18_col21\" class=\"data row18 col21\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row18_col22\" class=\"data row18 col22\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row18_col23\" class=\"data row18 col23\" >-0.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row19\" class=\"row_heading level0 row19\" >PAY_AMT3</th>\n",
       "      <td id=\"T_e19b5_row19_col0\" class=\"data row19 col0\" >0.21</td>\n",
       "      <td id=\"T_e19b5_row19_col1\" class=\"data row19 col1\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row19_col2\" class=\"data row19 col2\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row19_col3\" class=\"data row19 col3\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row19_col4\" class=\"data row19 col4\" >0.03</td>\n",
       "      <td id=\"T_e19b5_row19_col5\" class=\"data row19 col5\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row19_col6\" class=\"data row19 col6\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row19_col7\" class=\"data row19 col7\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row19_col8\" class=\"data row19 col8\" >-0.07</td>\n",
       "      <td id=\"T_e19b5_row19_col9\" class=\"data row19 col9\" >0.01</td>\n",
       "      <td id=\"T_e19b5_row19_col10\" class=\"data row19 col10\" >0.01</td>\n",
       "      <td id=\"T_e19b5_row19_col11\" class=\"data row19 col11\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row19_col12\" class=\"data row19 col12\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row19_col13\" class=\"data row19 col13\" >0.13</td>\n",
       "      <td id=\"T_e19b5_row19_col14\" class=\"data row19 col14\" >0.30</td>\n",
       "      <td id=\"T_e19b5_row19_col15\" class=\"data row19 col15\" >0.25</td>\n",
       "      <td id=\"T_e19b5_row19_col16\" class=\"data row19 col16\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row19_col17\" class=\"data row19 col17\" >0.26</td>\n",
       "      <td id=\"T_e19b5_row19_col18\" class=\"data row19 col18\" >0.25</td>\n",
       "      <td id=\"T_e19b5_row19_col19\" class=\"data row19 col19\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row19_col20\" class=\"data row19 col20\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row19_col21\" class=\"data row19 col21\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row19_col22\" class=\"data row19 col22\" >0.17</td>\n",
       "      <td id=\"T_e19b5_row19_col23\" class=\"data row19 col23\" >-0.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row20\" class=\"row_heading level0 row20\" >PAY_AMT4</th>\n",
       "      <td id=\"T_e19b5_row20_col0\" class=\"data row20 col0\" >0.20</td>\n",
       "      <td id=\"T_e19b5_row20_col1\" class=\"data row20 col1\" >0.00</td>\n",
       "      <td id=\"T_e19b5_row20_col2\" class=\"data row20 col2\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row20_col3\" class=\"data row20 col3\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row20_col4\" class=\"data row20 col4\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row20_col5\" class=\"data row20 col5\" >-0.07</td>\n",
       "      <td id=\"T_e19b5_row20_col6\" class=\"data row20 col6\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row20_col7\" class=\"data row20 col7\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row20_col8\" class=\"data row20 col8\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row20_col9\" class=\"data row20 col9\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row20_col10\" class=\"data row20 col10\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row20_col11\" class=\"data row20 col11\" >0.15</td>\n",
       "      <td id=\"T_e19b5_row20_col12\" class=\"data row20 col12\" >0.14</td>\n",
       "      <td id=\"T_e19b5_row20_col13\" class=\"data row20 col13\" >0.14</td>\n",
       "      <td id=\"T_e19b5_row20_col14\" class=\"data row20 col14\" >0.13</td>\n",
       "      <td id=\"T_e19b5_row20_col15\" class=\"data row20 col15\" >0.29</td>\n",
       "      <td id=\"T_e19b5_row20_col16\" class=\"data row20 col16\" >0.24</td>\n",
       "      <td id=\"T_e19b5_row20_col17\" class=\"data row20 col17\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row20_col18\" class=\"data row20 col18\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row20_col19\" class=\"data row20 col19\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row20_col20\" class=\"data row20 col20\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row20_col21\" class=\"data row20 col21\" >0.15</td>\n",
       "      <td id=\"T_e19b5_row20_col22\" class=\"data row20 col22\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row20_col23\" class=\"data row20 col23\" >-0.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row21\" class=\"row_heading level0 row21\" >PAY_AMT5</th>\n",
       "      <td id=\"T_e19b5_row21_col0\" class=\"data row21 col0\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row21_col1\" class=\"data row21 col1\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row21_col2\" class=\"data row21 col2\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row21_col3\" class=\"data row21 col3\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row21_col4\" class=\"data row21 col4\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row21_col5\" class=\"data row21 col5\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row21_col6\" class=\"data row21 col6\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row21_col7\" class=\"data row21 col7\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row21_col8\" class=\"data row21 col8\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row21_col9\" class=\"data row21 col9\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row21_col10\" class=\"data row21 col10\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row21_col11\" class=\"data row21 col11\" >0.17</td>\n",
       "      <td id=\"T_e19b5_row21_col12\" class=\"data row21 col12\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row21_col13\" class=\"data row21 col13\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row21_col14\" class=\"data row21 col14\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row21_col15\" class=\"data row21 col15\" >0.14</td>\n",
       "      <td id=\"T_e19b5_row21_col16\" class=\"data row21 col16\" >0.31</td>\n",
       "      <td id=\"T_e19b5_row21_col17\" class=\"data row21 col17\" >0.15</td>\n",
       "      <td id=\"T_e19b5_row21_col18\" class=\"data row21 col18\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row21_col19\" class=\"data row21 col19\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row21_col20\" class=\"data row21 col20\" >0.15</td>\n",
       "      <td id=\"T_e19b5_row21_col21\" class=\"data row21 col21\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row21_col22\" class=\"data row21 col22\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row21_col23\" class=\"data row21 col23\" >-0.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row22\" class=\"row_heading level0 row22\" >PAY_AMT6</th>\n",
       "      <td id=\"T_e19b5_row22_col0\" class=\"data row22 col0\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row22_col1\" class=\"data row22 col1\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row22_col2\" class=\"data row22 col2\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row22_col3\" class=\"data row22 col3\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row22_col4\" class=\"data row22 col4\" >0.02</td>\n",
       "      <td id=\"T_e19b5_row22_col5\" class=\"data row22 col5\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row22_col6\" class=\"data row22 col6\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row22_col7\" class=\"data row22 col7\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row22_col8\" class=\"data row22 col8\" >-0.03</td>\n",
       "      <td id=\"T_e19b5_row22_col9\" class=\"data row22 col9\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row22_col10\" class=\"data row22 col10\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row22_col11\" class=\"data row22 col11\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row22_col12\" class=\"data row22 col12\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row22_col13\" class=\"data row22 col13\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row22_col14\" class=\"data row22 col14\" >0.18</td>\n",
       "      <td id=\"T_e19b5_row22_col15\" class=\"data row22 col15\" >0.17</td>\n",
       "      <td id=\"T_e19b5_row22_col16\" class=\"data row22 col16\" >0.12</td>\n",
       "      <td id=\"T_e19b5_row22_col17\" class=\"data row22 col17\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row22_col18\" class=\"data row22 col18\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row22_col19\" class=\"data row22 col19\" >0.17</td>\n",
       "      <td id=\"T_e19b5_row22_col20\" class=\"data row22 col20\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row22_col21\" class=\"data row22 col21\" >0.16</td>\n",
       "      <td id=\"T_e19b5_row22_col22\" class=\"data row22 col22\" >1.00</td>\n",
       "      <td id=\"T_e19b5_row22_col23\" class=\"data row22 col23\" >-0.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_e19b5_level0_row23\" class=\"row_heading level0 row23\" >default_0</th>\n",
       "      <td id=\"T_e19b5_row23_col0\" class=\"data row23 col0\" >-0.15</td>\n",
       "      <td id=\"T_e19b5_row23_col1\" class=\"data row23 col1\" >-0.04</td>\n",
       "      <td id=\"T_e19b5_row23_col2\" class=\"data row23 col2\" >0.03</td>\n",
       "      <td id=\"T_e19b5_row23_col3\" class=\"data row23 col3\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row23_col4\" class=\"data row23 col4\" >0.01</td>\n",
       "      <td id=\"T_e19b5_row23_col5\" class=\"data row23 col5\" >0.32</td>\n",
       "      <td id=\"T_e19b5_row23_col6\" class=\"data row23 col6\" >0.26</td>\n",
       "      <td id=\"T_e19b5_row23_col7\" class=\"data row23 col7\" >0.23</td>\n",
       "      <td id=\"T_e19b5_row23_col8\" class=\"data row23 col8\" >0.22</td>\n",
       "      <td id=\"T_e19b5_row23_col9\" class=\"data row23 col9\" >0.21</td>\n",
       "      <td id=\"T_e19b5_row23_col10\" class=\"data row23 col10\" >0.19</td>\n",
       "      <td id=\"T_e19b5_row23_col11\" class=\"data row23 col11\" >-0.02</td>\n",
       "      <td id=\"T_e19b5_row23_col12\" class=\"data row23 col12\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row23_col13\" class=\"data row23 col13\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row23_col14\" class=\"data row23 col14\" >-0.01</td>\n",
       "      <td id=\"T_e19b5_row23_col15\" class=\"data row23 col15\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row23_col16\" class=\"data row23 col16\" >-0.00</td>\n",
       "      <td id=\"T_e19b5_row23_col17\" class=\"data row23 col17\" >-0.07</td>\n",
       "      <td id=\"T_e19b5_row23_col18\" class=\"data row23 col18\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row23_col19\" class=\"data row23 col19\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row23_col20\" class=\"data row23 col20\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row23_col21\" class=\"data row23 col21\" >-0.06</td>\n",
       "      <td id=\"T_e19b5_row23_col22\" class=\"data row23 col22\" >-0.05</td>\n",
       "      <td id=\"T_e19b5_row23_col23\" class=\"data row23 col23\" >1.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7fb07162d0d0>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# creating our heatmap \n",
    "\n",
    "cmap = cmap=sns.diverging_palette(5, 250, as_cmap=True)\n",
    "\n",
    "# creating a function to pass in our figure\n",
    "\n",
    "def cor_heatmap():\n",
    "    return [dict(selector=\"th\",\n",
    "                 props=[(\"font-size\", \"7pt\")]),\n",
    "            dict(selector=\"td\",\n",
    "                 props=[('padding', \"0em 0em\")]),\n",
    "            dict(selector=\"th:hover\",\n",
    "                 props=[(\"font-size\", \"12pt\")]),\n",
    "            dict(selector=\"tr:hover td:hover\",\n",
    "                 props=[('max-width', '200px'),\n",
    "                        ('font-size', '12pt')])]\n",
    "\n",
    "\n",
    "corr.style.background_gradient(cmap, axis=0)\\\n",
    "    .set_properties(**{'max-width': '80px', 'font-size': '10pt'})\\\n",
    "    .set_caption(\"Correlations Heatmap\")\\\n",
    "    .set_precision(2)\\\n",
    "    .set_table_styles(cor_heatmap())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A correlation heatmap is graphical representation of correlation matrix representing correlation between different variables. The value of correlation can take any values from -1 to 1. Identifing highly correlated variables will allow us to streamline the feature selection process. Understanding a dataset is very important when creating a model, and a heatmap is one of the many tools that can be used to achieve that.\n",
    "\n",
    "By checking on our heatmap, we can realise that red means positive correlation and blue means negative one. The dark red are meant to have strong positive correaltion between them, while the blue ones mean they have a negative correlation. Also, we have some variables that are close to 0 and that means there is no correlation between these two variables. The 1 are the same variables. \n",
    "\n",
    "As the color becomes darker in either direction (red or blue) that means that those variables are more highly correlated and should not be paired together in the same model.\n",
    "\n",
    "In order now to continue with our prediction we are going to apply some methods (like scaling/normalising and transforming some of our variables into dummy variables) to get better results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before proceeding with our data cleaning and feature engineering, we are going to run a linear regression to our data by using scikit and see what we get."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = credit_data[[\"default_0\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = credit_data.drop(columns=[\"default_0\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_other, y_train, y_other = train_test_split(X, y, test_size=0.4, random_state = 657)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test, X_validation, y_test, y_validation = train_test_split(X_other, y_other, test_size=0.5, random_state = 343)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(max_iter=2000)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classifier_LR = LogisticRegression(max_iter=2000)\n",
    "classifier_LR.fit(X_train, y_train.values.ravel())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>Variable</th>\n",
       "      <th>Coefficient</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>PAY_1</td>\n",
       "      <td>0.001147</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>PAY_2</td>\n",
       "      <td>0.000903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>PAY_3</td>\n",
       "      <td>0.000821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>PAY_4</td>\n",
       "      <td>0.000783</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>PAY_5</td>\n",
       "      <td>0.000729</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Variable Coefficient\n",
       "5    PAY_1    0.001147\n",
       "6    PAY_2    0.000903\n",
       "7    PAY_3    0.000821\n",
       "8    PAY_4    0.000783\n",
       "9    PAY_5    0.000729"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# getting the coefficients of our dataset, he higher the coefficient of variation, the greater the level of dispersion around the mean. \n",
    "# it is generally expressed as a percentage. The lower the value of the coefficient of variation, the more precise the estimate.\n",
    "\n",
    "summary = pd.DataFrame([X_train.columns,classifier_LR.coef_[0]]).T.sort_values(by = 1, ascending = False)\n",
    "summary.columns = ['Variable','Coefficient']\n",
    "\n",
    "#highest coef\n",
    "\n",
    "summary.head(n = 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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>Variable</th>\n",
       "      <th>Coefficient</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>PAY_AMT1</td>\n",
       "      <td>-0.000024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>SEX</td>\n",
       "      <td>-0.000925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>MARRIAGE</td>\n",
       "      <td>-0.000946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>-0.001152</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AGE</td>\n",
       "      <td>-0.016516</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Variable Coefficient\n",
       "17   PAY_AMT1   -0.000024\n",
       "1         SEX   -0.000925\n",
       "3    MARRIAGE   -0.000946\n",
       "2   EDUCATION   -0.001152\n",
       "4         AGE   -0.016516"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# lowest coef\n",
    "\n",
    "summary.tail(n = 5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we see above there are also negative coefficients, because as the independent variable increases, the dependent variable tends to decrease. The coefficient value signifies how much the mean of the dependent variable changes given a one-unit shift in the independent variable, while holding other variables in the model constant."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A receiver operating characteristic curve (or known as ROC curve) is a graphical plot that illustrates the diagnostic ability of a binary classifier system as its discrimination threshold is varied. Let's check it out."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting the data in order to check the grey area below our curve (known as AUC) measuring how good our model is before picking a threshold\n",
    "\n",
    "y_pred_prob = classifier_LR.predict_proba(X_validation)[:,1]\n",
    "fpr, tpr, thresholds = metrics.roc_curve(y_validation,y_pred_prob)\n",
    "pyplot.plot(fpr, tpr, linewidth=4)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6347302427428545"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# measuring the grey area under the curve \n",
    "\n",
    "metrics.roc_auc_score(y_validation, y_pred_prob)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Based on our data, we can directly compute the costs and profits of our predictions. \n",
    "\n",
    "1. True negative: We predicted no (they will not default), and they didn't default. The net profit then is £1500.\n",
    "2. False negative: We predicted no, but they actually did default (also known as a \"Type II classification error\"). The net profit then is  -5000£ (we lose money).\n",
    "3. False positive: We predicted yes, but they didn't default (known as a \"Type I classification error\"). The net profit then is £0.\n",
    "4. True positive: We predicted yes (they will default), and the actually default. The net profit then is £0.\n",
    "\n",
    "Given this, we can now calculate the profits we would get from the validation customers, using the confusion matrix (we start again with a threshold of 0.7):\n",
    "\n",
    "Given this, we can now calculate the profits we would get from the validation customers, using the confusion matrix (to find the threshold we want to start with we run a loop to find it).\n",
    "\n",
    "The above steps can also illustrate on the so called confusion matrix."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The profit at threshold 0.0 is 0\n",
      "The profit at threshold 0.01 is 45000\n",
      "The profit at threshold 0.02 is 92000\n",
      "The profit at threshold 0.03 is 111000\n",
      "The profit at threshold 0.04 is 144500\n",
      "The profit at threshold 0.05 is 206500\n",
      "The profit at threshold 0.06 is 259500\n",
      "The profit at threshold 0.07 is 264000\n",
      "The profit at threshold 0.08 is 327500\n",
      "The profit at threshold 0.09 is 375500\n",
      "The profit at threshold 0.1 is 440500\n",
      "The profit at threshold 0.11 is 512000\n",
      "The profit at threshold 0.12 is 560000\n",
      "The profit at threshold 0.13 is 651500\n",
      "The profit at threshold 0.14 is 711000\n",
      "The profit at threshold 0.15 is 771000\n",
      "The profit at threshold 0.16 is 843000\n",
      "The profit at threshold 0.17 is 897000\n",
      "The profit at threshold 0.18 is 921500\n",
      "The profit at threshold 0.19 is 988500\n",
      "The profit at threshold 0.2 is 1051500\n",
      "The profit at threshold 0.21 is 1087000\n",
      "The profit at threshold 0.22 is 1150500\n",
      "The profit at threshold 0.23 is 1192500\n",
      "The profit at threshold 0.24 is 1177500\n",
      "The profit at threshold 0.25 is 1226500\n",
      "The profit at threshold 0.26 is 1085500\n",
      "The profit at threshold 0.27 is 1130000\n",
      "The profit at threshold 0.28 is 1084000\n",
      "The profit at threshold 0.29 is 1009000\n",
      "The profit at threshold 0.3 is 995000\n",
      "The profit at threshold 0.31 is 880500\n",
      "The profit at threshold 0.32 is 817500\n",
      "The profit at threshold 0.33 is 741000\n",
      "The profit at threshold 0.34 is 632500\n",
      "The profit at threshold 0.35000000000000003 is 514500\n",
      "The profit at threshold 0.36 is 429500\n",
      "The profit at threshold 0.37 is 289500\n",
      "The profit at threshold 0.38 is 274500\n",
      "The profit at threshold 0.39 is 220000\n",
      "The profit at threshold 0.4 is 191500\n",
      "The profit at threshold 0.41000000000000003 is 169500\n",
      "The profit at threshold 0.42 is 144500\n",
      "The profit at threshold 0.43 is 144500\n",
      "The profit at threshold 0.44 is 139500\n",
      "The profit at threshold 0.45 is 139500\n",
      "The profit at threshold 0.46 is 139500\n",
      "The profit at threshold 0.47000000000000003 is 139500\n",
      "The profit at threshold 0.48 is 139500\n",
      "The profit at threshold 0.49 is 139500\n",
      "The profit at threshold 0.5 is 139500\n",
      "The profit at threshold 0.51 is 139500\n",
      "The profit at threshold 0.52 is 139500\n",
      "The profit at threshold 0.53 is 141000\n",
      "The profit at threshold 0.54 is 141000\n",
      "The profit at threshold 0.55 is 141000\n",
      "The profit at threshold 0.56 is 141000\n",
      "The profit at threshold 0.5700000000000001 is 141000\n",
      "The profit at threshold 0.58 is 141000\n",
      "The profit at threshold 0.59 is 141000\n",
      "The profit at threshold 0.6 is 141000\n",
      "The profit at threshold 0.61 is 141000\n",
      "The profit at threshold 0.62 is 141000\n",
      "The profit at threshold 0.63 is 141000\n",
      "The profit at threshold 0.64 is 141000\n",
      "The profit at threshold 0.65 is 141000\n",
      "The profit at threshold 0.66 is 141000\n",
      "The profit at threshold 0.67 is 141000\n",
      "The profit at threshold 0.68 is 141000\n",
      "The profit at threshold 0.6900000000000001 is 141000\n",
      "The profit at threshold 0.7000000000000001 is 141000\n",
      "The profit at threshold 0.71 is 141000\n",
      "The profit at threshold 0.72 is 141000\n",
      "The profit at threshold 0.73 is 141000\n",
      "The profit at threshold 0.74 is 141000\n",
      "The profit at threshold 0.75 is 141000\n",
      "The profit at threshold 0.76 is 141000\n",
      "The profit at threshold 0.77 is 141000\n",
      "The profit at threshold 0.78 is 141000\n",
      "The profit at threshold 0.79 is 141000\n",
      "The profit at threshold 0.8 is 141000\n",
      "The profit at threshold 0.81 is 141000\n",
      "The profit at threshold 0.8200000000000001 is 141000\n",
      "The profit at threshold 0.8300000000000001 is 141000\n",
      "The profit at threshold 0.84 is 141000\n",
      "The profit at threshold 0.85 is 141000\n",
      "The profit at threshold 0.86 is 141000\n",
      "The profit at threshold 0.87 is 141000\n",
      "The profit at threshold 0.88 is 141000\n",
      "The profit at threshold 0.89 is 141000\n",
      "The profit at threshold 0.9 is 141000\n",
      "The profit at threshold 0.91 is 141000\n",
      "The profit at threshold 0.92 is 141000\n",
      "The profit at threshold 0.93 is 141000\n",
      "The profit at threshold 0.9400000000000001 is 141000\n",
      "The profit at threshold 0.9500000000000001 is 141000\n",
      "The profit at threshold 0.96 is 141000\n",
      "The profit at threshold 0.97 is 141000\n",
      "The profit at threshold 0.98 is 141000\n",
      "The profit at threshold 0.99 is 141000\n"
     ]
    }
   ],
   "source": [
    "thres = []\n",
    "profits = []\n",
    "threshold = np.arange(0,1, 0.01)\n",
    "for i in threshold:\n",
    "    y_validation_pred = np.where(y_pred_prob < i, 0, 1)\n",
    "    cm = metrics.confusion_matrix(y_validation, y_validation_pred)\n",
    "    profit = cm[0][0] * 1500 + cm[1][0] * (-5000) + cm[0][1] * 0 + cm[1][1] * 0\n",
    "    print(\"The profit at threshold \" + str(i) + \" is \" + str(profit))\n",
    "    profits.append(profit)\n",
    "    thres.append(i)\n",
    "\n",
    "# creating a dataframe with the threshold and profits\n",
    "\n",
    "df = pd.DataFrame({'Threshold' : thres, 'Profits' : profits})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Threshold    99\n",
       "Profits      25\n",
       "dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# going to our dataframe and find the max profit in which line is it\n",
    "\n",
    "df.idxmax(axis = 0, skipna = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Threshold  Profits\n",
      "25       0.25  1226500\n"
     ]
    }
   ],
   "source": [
    "# printing out the max profit with a threshold of 0.25 and a profit up to 1225000\n",
    "\n",
    "print(df.loc[[25]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[2121 1593]\n",
      " [ 391  695]]\n",
      "0.5866666666666667\n"
     ]
    }
   ],
   "source": [
    "# setting the correct threshold (0.25) in order to get our confusion matrix and accuracy\n",
    "\n",
    "y_validation_pred = np.where(y_pred_prob < 0.25, 0, 1)\n",
    "\n",
    "# confusion matrix \n",
    "\n",
    "print(metrics.confusion_matrix(y_validation, y_validation_pred))\n",
    "\n",
    "# getting the accuracy\n",
    "\n",
    "print(metrics.accuracy_score(y_validation,y_validation_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before continuing to predict our new applicants, we will pre-process/use of feature engineering in the data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So from our data it is obvious that we will have to transform some of the them (dummy variables, use of scaling/normalizing method) in order to\n",
    "be able to get better results we want (improvement of the ROC curve)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Changing some columns into dummy variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "#creating the dummy variables of \"Sex, Education, Marriage, and Pay\" (we know that for dummy variables n-1)\n",
    "\n",
    "credit_data = pd.get_dummies(credit_data, columns = ['SEX'])\n",
    "credit_data = credit_data.drop([\"SEX_2\"], axis = 1)\n",
    "credit_data = credit_data.rename({'SEX_1': 'GENDER'}, axis=1) # here we will notice that instead of 1 = male and 2 = female, we have 1 = male and 0 = female\n",
    "\n",
    "# continuing with education\n",
    "\n",
    "credit_data = pd.get_dummies(credit_data, columns = ['EDUCATION']) \n",
    "credit_data = credit_data.drop([\"EDUCATION_0\", \"EDUCATION_5\", \"EDUCATION_6\"], axis = 1) # so at this part, and based on our details of the data set we can notice that for education that we have 4 = other and 0, 5, 6 = unknown what we will do is to combine these two options because in case we have all 0 in our dataframe means that our costumer belongs in one of these two categories\n",
    "credit_data = credit_data.rename({'EDUCATION_1' : 'GRADUATE', 'EDUCATION_2' : 'UNDERGRADUATE', 'EDUCATION_3' : 'HIGH SCHOOL', 'EDUCATION_4' : 'OTHER_UNKNOWN'}, axis = 1) # so in every column based on the name of the column if we find 1 means that this guys has this kind of education, otherwise its 0 means he belongs to a different category\n",
    "\n",
    "# continuing with marriage\n",
    "credit_data = pd.get_dummies(credit_data, columns = ['MARRIAGE'])\n",
    "credit_data['MARRIAGE_0_3'] = (credit_data.MARRIAGE_0 | credit_data.MARRIAGE_3).astype(int) # same as education we combine other, unknown in one dummy variable because as far as we get in both married, single 0 we know he belongs to the other options\n",
    "credit_data = credit_data.drop([\"MARRIAGE_0\", \"MARRIAGE_3\"], axis = 1)\n",
    "credit_data = credit_data.rename({'MARRIAGE_1' : 'MARRIED', 'MARRIAGE_2' : 'SINGLE', 'MARRIAGE_0_3' : 'NOT MARRIED-NOT SINGLE-OTHER'}, axis = 1) # same as education about 1 and 0 (the explanation), so if for both cases is 0 we know that our costumer belongs to other or unknown\n",
    "\n",
    "# last creating the pay into dummies too\n",
    "\n",
    "credit_data[\"PAY_1\"] = credit_data[\"PAY_1\"].replace(-2, 0).replace( -1, 0)\n",
    "credit_data[\"PAY_1\"] = credit_data[\"PAY_1\"].replace(2, 1).replace(3, 1).replace(4, 1).replace(5, 1).replace(6, 1).replace(7, 1).replace(8, 1).replace(9, 1)\n",
    "\n",
    "credit_data[\"PAY_2\"] = credit_data[\"PAY_2\"].replace(-2, 0).replace(-1, 0)\n",
    "credit_data[\"PAY_2\"] = credit_data[\"PAY_2\"].replace(2, 1).replace(3, 1).replace(4, 1).replace(5, 1).replace(6, 1).replace(7, 1).replace(8, 1).replace(9, 1)\n",
    "\n",
    "credit_data[\"PAY_3\"] = credit_data[\"PAY_3\"].replace(-2, 0).replace(-1, 0)\n",
    "credit_data[\"PAY_3\"] = credit_data[\"PAY_3\"].replace(2, 1).replace(3, 1).replace(4, 1).replace(5, 1).replace(6, 1).replace(7, 1).replace(8, 1).replace(9, 1)\n",
    "\n",
    "credit_data[\"PAY_4\"] = credit_data[\"PAY_4\"].replace(-2, 0).replace(-1, 0)\n",
    "credit_data[\"PAY_4\"] = credit_data[\"PAY_4\"].replace(2, 1).replace(3, 1).replace(4, 1).replace(5, 1).replace(6, 1).replace(7, 1).replace(8, 1).replace(9, 1)\n",
    "\n",
    "credit_data[\"PAY_5\"] = credit_data[\"PAY_5\"].replace(-2, 0).replace(-1, 0)\n",
    "credit_data[\"PAY_5\"] = credit_data[\"PAY_5\"].replace(2, 1).replace(3, 1).replace(4, 1).replace(5, 1).replace(6, 1).replace(7, 1).replace(8, 1).replace(9, 1)\n",
    "\n",
    "credit_data[\"PAY_6\"] = credit_data[\"PAY_6\"].replace(-2, 0).replace(-1, 0)\n",
    "credit_data[\"PAY_6\"] = credit_data[\"PAY_6\"].replace(2, 1).replace(3, 1).replace(4, 1).replace(5, 1).replace(6, 1).replace(7, 1).replace(8, 1).replace(9, 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Scaling/normalizing limit balance, bill amount and pay amount "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "#scaling limit bal, bill amt and pay amt \n",
    "\n",
    "credit_data[['LIMIT_BAL']] = preprocessing.scale(credit_data[['LIMIT_BAL']])\n",
    "\n",
    "credit_data[[\n",
    "    'BILL_AMT1', 'BILL_AMT2', 'BILL_AMT3', 'BILL_AMT4', 'BILL_AMT5',\n",
    "    'BILL_AMT6'\n",
    "]] = preprocessing.scale(credit_data[[\n",
    "    'BILL_AMT1', 'BILL_AMT2', 'BILL_AMT3', 'BILL_AMT4', 'BILL_AMT5',\n",
    "    'BILL_AMT6'\n",
    "]])\n",
    "\n",
    "credit_data[[\n",
    "    'PAY_AMT1', 'PAY_AMT2', 'PAY_AMT3', 'PAY_AMT4', 'PAY_AMT5',\n",
    "    'PAY_AMT6'\n",
    "]] = preprocessing.scale(credit_data[[\n",
    "    'PAY_AMT1', 'PAY_AMT2', 'PAY_AMT3', 'PAY_AMT4', 'PAY_AMT5',\n",
    "    'PAY_AMT6'\n",
    "]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>LIMIT_BAL</th>\n",
       "      <th>AGE</th>\n",
       "      <th>PAY_1</th>\n",
       "      <th>PAY_2</th>\n",
       "      <th>PAY_3</th>\n",
       "      <th>PAY_4</th>\n",
       "      <th>PAY_5</th>\n",
       "      <th>PAY_6</th>\n",
       "      <th>BILL_AMT1</th>\n",
       "      <th>BILL_AMT2</th>\n",
       "      <th>...</th>\n",
       "      <th>PAY_AMT6</th>\n",
       "      <th>default_0</th>\n",
       "      <th>GENDER</th>\n",
       "      <th>GRADUATE</th>\n",
       "      <th>UNDERGRADUATE</th>\n",
       "      <th>HIGH SCHOOL</th>\n",
       "      <th>OTHER_UNKNOWN</th>\n",
       "      <th>MARRIED</th>\n",
       "      <th>SINGLE</th>\n",
       "      <th>NOT MARRIED-NOT SINGLE-OTHER</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.138345</td>\n",
       "      <td>24</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.640803</td>\n",
       "      <td>-0.646348</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.289940</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.599489</td>\n",
       "      <td>34</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.296915</td>\n",
       "      <td>-0.492875</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.015424</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.907406</td>\n",
       "      <td>37</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.055884</td>\n",
       "      <td>-0.012355</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.235037</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.907406</td>\n",
       "      <td>57</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.576930</td>\n",
       "      <td>-0.610273</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.252661</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.907406</td>\n",
       "      <td>37</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.180517</td>\n",
       "      <td>0.111772</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.246018</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23995</th>\n",
       "      <td>-0.676468</td>\n",
       "      <td>34</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.291277</td>\n",
       "      <td>0.401706</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.070327</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23996</th>\n",
       "      <td>-0.137612</td>\n",
       "      <td>43</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.671083</td>\n",
       "      <td>-0.664245</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.289940</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23997</th>\n",
       "      <td>-1.061365</td>\n",
       "      <td>37</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.645529</td>\n",
       "      <td>-0.642780</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.119740</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23998</th>\n",
       "      <td>-0.676468</td>\n",
       "      <td>41</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.716272</td>\n",
       "      <td>0.411132</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.190895</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23999</th>\n",
       "      <td>-0.907406</td>\n",
       "      <td>46</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.043133</td>\n",
       "      <td>-0.002914</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.235037</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>24000 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       LIMIT_BAL  AGE  PAY_1  PAY_2  PAY_3  PAY_4  PAY_5  PAY_6  BILL_AMT1  \\\n",
       "0      -1.138345   24      1      1      0      0      0      0  -0.640803   \n",
       "1      -0.599489   34      0      0      0      0      0      0  -0.296915   \n",
       "2      -0.907406   37      0      0      0      0      0      0  -0.055884   \n",
       "3      -0.907406   57      0      0      0      0      0      0  -0.576930   \n",
       "4      -0.907406   37      0      0      0      0      0      0   0.180517   \n",
       "...          ...  ...    ...    ...    ...    ...    ...    ...        ...   \n",
       "23995  -0.676468   34      1      1      1      1      1      1   0.291277   \n",
       "23996  -0.137612   43      0      0      0      0      0      0  -0.671083   \n",
       "23997  -1.061365   37      1      1      1      0      0      0  -0.645529   \n",
       "23998  -0.676468   41      1      0      0      0      0      0  -0.716272   \n",
       "23999  -0.907406   46      0      0      0      0      0      0  -0.043133   \n",
       "\n",
       "       BILL_AMT2  ...  PAY_AMT6  default_0  GENDER  GRADUATE  UNDERGRADUATE  \\\n",
       "0      -0.646348  ... -0.289940          1       0         0              1   \n",
       "1      -0.492875  ... -0.015424          0       0         0              1   \n",
       "2      -0.012355  ... -0.235037          0       0         0              1   \n",
       "3      -0.610273  ... -0.252661          0       1         0              1   \n",
       "4       0.111772  ... -0.246018          0       1         1              0   \n",
       "...          ...  ...       ...        ...     ...       ...            ...   \n",
       "23995   0.401706  ... -0.070327          1       1         0              1   \n",
       "23996  -0.664245  ... -0.289940          0       1         0              0   \n",
       "23997  -0.642780  ... -0.119740          1       1         0              1   \n",
       "23998   0.411132  ... -0.190895          1       1         0              0   \n",
       "23999  -0.002914  ... -0.235037          1       1         0              1   \n",
       "\n",
       "       HIGH SCHOOL  OTHER_UNKNOWN  MARRIED  SINGLE  \\\n",
       "0                0              0        1       0   \n",
       "1                0              0        0       1   \n",
       "2                0              0        1       0   \n",
       "3                0              0        1       0   \n",
       "4                0              0        0       1   \n",
       "...            ...            ...      ...     ...   \n",
       "23995            0              0        0       1   \n",
       "23996            1              0        0       1   \n",
       "23997            0              0        0       1   \n",
       "23998            1              0        1       0   \n",
       "23999            0              0        1       0   \n",
       "\n",
       "       NOT MARRIED-NOT SINGLE-OTHER  \n",
       "0                                 0  \n",
       "1                                 0  \n",
       "2                                 0  \n",
       "3                                 0  \n",
       "4                                 0  \n",
       "...                             ...  \n",
       "23995                             0  \n",
       "23996                             0  \n",
       "23997                             0  \n",
       "23998                             0  \n",
       "23999                             0  \n",
       "\n",
       "[24000 rows x 29 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# checking our new dataframe with the new columns \n",
    "\n",
    "credit_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We are going now to rerun our data with the transformations in order to see how much our profits have been improved."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = credit_data[[\"default_0\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = credit_data.drop(columns=[\"default_0\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_other, y_train, y_other = train_test_split(X, y, test_size=0.4, random_state = 657)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test, X_validation, y_test, y_validation = train_test_split(X_other, y_other, test_size=0.5, random_state = 343)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(max_iter=2000)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classifier_LR = LogisticRegression(max_iter=2000)\n",
    "classifier_LR.fit(X_train, y_train.values.ravel())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>Variable</th>\n",
       "      <th>Coefficient</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>PAY_1</td>\n",
       "      <td>1.285347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>GRADUATE</td>\n",
       "      <td>0.820563</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>UNDERGRADUATE</td>\n",
       "      <td>0.806026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>HIGH SCHOOL</td>\n",
       "      <td>0.739316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>PAY_4</td>\n",
       "      <td>0.403128</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Variable Coefficient\n",
       "2           PAY_1    1.285347\n",
       "21       GRADUATE    0.820563\n",
       "22  UNDERGRADUATE    0.806026\n",
       "23    HIGH SCHOOL    0.739316\n",
       "5           PAY_4    0.403128"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# getting the coefficients  \n",
    "\n",
    "summary = pd.DataFrame([X_train.columns,classifier_LR.coef_[0]]).T.sort_values(by = 1, ascending = False)\n",
    "summary.columns = ['Variable','Coefficient']\n",
    "summary.head(n=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "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>Variable</th>\n",
       "      <th>Coefficient</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>PAY_AMT1</td>\n",
       "      <td>-0.131624</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LIMIT_BAL</td>\n",
       "      <td>-0.184001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>PAY_AMT2</td>\n",
       "      <td>-0.252733</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>SINGLE</td>\n",
       "      <td>-0.256039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>OTHER_UNKNOWN</td>\n",
       "      <td>-0.604961</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Variable Coefficient\n",
       "14       PAY_AMT1   -0.131624\n",
       "0       LIMIT_BAL   -0.184001\n",
       "15       PAY_AMT2   -0.252733\n",
       "26         SINGLE   -0.256039\n",
       "24  OTHER_UNKNOWN   -0.604961"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "summary.tail(n=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting the data in order to check if there was an imporvement in the grey area below our curve measuring how good our model is before picking a threshold\n",
    "\n",
    "y_pred_prob = classifier_LR.predict_proba(X_validation)[:,1]\n",
    "fpr, tpr, thresholds = metrics.roc_curve(y_validation,y_pred_prob)\n",
    "pyplot.plot(fpr, tpr, linewidth=4)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.763274023628677\n"
     ]
    }
   ],
   "source": [
    "print(metrics.roc_auc_score(y_validation, y_pred_prob))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finding the best threshold in our final model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The profit at threshold 0.0 is 0\n",
      "The profit at threshold 0.01 is 10500\n",
      "The profit at threshold 0.02 is 17500\n",
      "The profit at threshold 0.03 is 24500\n",
      "The profit at threshold 0.04 is 61000\n",
      "The profit at threshold 0.05 is 94500\n",
      "The profit at threshold 0.06 is 158500\n",
      "The profit at threshold 0.07 is 253000\n",
      "The profit at threshold 0.08 is 358000\n",
      "The profit at threshold 0.09 is 488000\n",
      "The profit at threshold 0.1 is 590500\n",
      "The profit at threshold 0.11 is 740500\n",
      "The profit at threshold 0.12 is 1055500\n",
      "The profit at threshold 0.13 is 1411000\n",
      "The profit at threshold 0.14 is 1761000\n",
      "The profit at threshold 0.15 is 2028000\n",
      "The profit at threshold 0.16 is 2239000\n",
      "The profit at threshold 0.17 is 2300500\n",
      "The profit at threshold 0.18 is 2363000\n",
      "The profit at threshold 0.19 is 2366500\n",
      "The profit at threshold 0.2 is 2343000\n",
      "The profit at threshold 0.21 is 2344500\n",
      "The profit at threshold 0.22 is 2330000\n",
      "The profit at threshold 0.23 is 2343000\n",
      "The profit at threshold 0.24 is 2337000\n",
      "The profit at threshold 0.25 is 2338000\n",
      "The profit at threshold 0.26 is 2329000\n",
      "The profit at threshold 0.27 is 2342000\n",
      "The profit at threshold 0.28 is 2324000\n",
      "The profit at threshold 0.29 is 2307000\n",
      "The profit at threshold 0.3 is 2294000\n",
      "The profit at threshold 0.31 is 2291500\n",
      "The profit at threshold 0.32 is 2305000\n",
      "The profit at threshold 0.33 is 2271500\n",
      "The profit at threshold 0.34 is 2238500\n",
      "The profit at threshold 0.35000000000000003 is 2216500\n",
      "The profit at threshold 0.36 is 2156500\n",
      "The profit at threshold 0.37 is 2121000\n",
      "The profit at threshold 0.38 is 2043500\n",
      "The profit at threshold 0.39 is 1985000\n",
      "The profit at threshold 0.4 is 1972000\n",
      "The profit at threshold 0.41000000000000003 is 1897000\n",
      "The profit at threshold 0.42 is 1854500\n",
      "The profit at threshold 0.43 is 1807500\n",
      "The profit at threshold 0.44 is 1749000\n",
      "The profit at threshold 0.45 is 1707500\n",
      "The profit at threshold 0.46 is 1674500\n",
      "The profit at threshold 0.47000000000000003 is 1627000\n",
      "The profit at threshold 0.48 is 1592000\n",
      "The profit at threshold 0.49 is 1574500\n",
      "The profit at threshold 0.5 is 1542000\n",
      "The profit at threshold 0.51 is 1498500\n",
      "The profit at threshold 0.52 is 1440500\n",
      "The profit at threshold 0.53 is 1381000\n",
      "The profit at threshold 0.54 is 1353500\n",
      "The profit at threshold 0.55 is 1307500\n",
      "The profit at threshold 0.56 is 1298000\n",
      "The profit at threshold 0.5700000000000001 is 1263500\n",
      "The profit at threshold 0.58 is 1213000\n",
      "The profit at threshold 0.59 is 1175500\n",
      "The profit at threshold 0.6 is 1164500\n",
      "The profit at threshold 0.61 is 1107000\n",
      "The profit at threshold 0.62 is 1042500\n",
      "The profit at threshold 0.63 is 1009000\n",
      "The profit at threshold 0.64 is 994000\n",
      "The profit at threshold 0.65 is 969500\n",
      "The profit at threshold 0.66 is 934000\n",
      "The profit at threshold 0.67 is 913000\n",
      "The profit at threshold 0.68 is 899000\n",
      "The profit at threshold 0.6900000000000001 is 863500\n",
      "The profit at threshold 0.7000000000000001 is 816500\n",
      "The profit at threshold 0.71 is 781000\n",
      "The profit at threshold 0.72 is 763500\n",
      "The profit at threshold 0.73 is 737500\n",
      "The profit at threshold 0.74 is 667500\n",
      "The profit at threshold 0.75 is 599000\n",
      "The profit at threshold 0.76 is 526500\n",
      "The profit at threshold 0.77 is 422000\n",
      "The profit at threshold 0.78 is 305000\n",
      "The profit at threshold 0.79 is 240500\n",
      "The profit at threshold 0.8 is 193000\n",
      "The profit at threshold 0.81 is 159500\n",
      "The profit at threshold 0.8200000000000001 is 146000\n",
      "The profit at threshold 0.8300000000000001 is 141000\n",
      "The profit at threshold 0.84 is 141000\n",
      "The profit at threshold 0.85 is 141000\n",
      "The profit at threshold 0.86 is 141000\n",
      "The profit at threshold 0.87 is 141000\n",
      "The profit at threshold 0.88 is 141000\n",
      "The profit at threshold 0.89 is 141000\n",
      "The profit at threshold 0.9 is 141000\n",
      "The profit at threshold 0.91 is 141000\n",
      "The profit at threshold 0.92 is 141000\n",
      "The profit at threshold 0.93 is 141000\n",
      "The profit at threshold 0.9400000000000001 is 141000\n",
      "The profit at threshold 0.9500000000000001 is 141000\n",
      "The profit at threshold 0.96 is 141000\n",
      "The profit at threshold 0.97 is 141000\n",
      "The profit at threshold 0.98 is 141000\n",
      "The profit at threshold 0.99 is 141000\n"
     ]
    }
   ],
   "source": [
    "thres = []\n",
    "profits = []\n",
    "threshold = np.arange(0,1, 0.01)\n",
    "\n",
    "for i in threshold:\n",
    "    y_validation_pred = np.where(y_pred_prob < i, 0, 1)\n",
    "    cm = metrics.confusion_matrix(y_validation, y_validation_pred)\n",
    "    profit = cm[0][0] * 1500 + cm[1][0] * (-5000) + cm[0][1] * 0 + cm[1][1] * 0\n",
    "    print(\"The profit at threshold \" + str(i) + \" is \" + str(profit))\n",
    "    profits.append(profit)\n",
    "    thres.append(i)\n",
    "\n",
    "# creating a dataframe for threshold and profit \n",
    "\n",
    "df = pd.DataFrame({'Threshold' : thres, 'Profits' : profits})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Threshold    99\n",
       "Profits      19\n",
       "dtype: int64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.idxmax(axis = 0, skipna = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Threshold  Profits\n",
      "19       0.19  2366500\n"
     ]
    }
   ],
   "source": [
    "print(df.loc[[19]])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The max profit at a threshold of 0.19 is up to 2371500."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_validation_pred = np.where(y_pred_prob < 0.19, 0, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[2901  813]\n",
      " [ 397  689]]\n",
      "0.7479166666666667\n"
     ]
    }
   ],
   "source": [
    "# getting the confusion matrix and the accuracy of our model after the transformations\n",
    "\n",
    "print(metrics.confusion_matrix(y_validation, y_validation_pred))\n",
    "print(metrics.accuracy_score(y_validation,y_validation_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So based on the confusion metrics what we have above is:\n",
    "\n",
    "True Negatives(TN): 2901\n",
    "\n",
    "False Positives(FP): 813\n",
    "\n",
    "False Negatives(FN): 396\n",
    "\n",
    "True Positives(TP): 690\n",
    "\n",
    "And the Accuracy classification score is: 0748125."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If we compare the two dataframes (before and after the transformations) we will notice that our accuracy has been improved up to close 0.2."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, it's time to predict our new applicants based on the results of our model above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "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>ID</th>\n",
       "      <th>LIMIT_BAL</th>\n",
       "      <th>SEX</th>\n",
       "      <th>EDUCATION</th>\n",
       "      <th>MARRIAGE</th>\n",
       "      <th>AGE</th>\n",
       "      <th>PAY_1</th>\n",
       "      <th>PAY_2</th>\n",
       "      <th>PAY_3</th>\n",
       "      <th>PAY_4</th>\n",
       "      <th>...</th>\n",
       "      <th>BILL_AMT4</th>\n",
       "      <th>BILL_AMT5</th>\n",
       "      <th>BILL_AMT6</th>\n",
       "      <th>PAY_AMT1</th>\n",
       "      <th>PAY_AMT2</th>\n",
       "      <th>PAY_AMT3</th>\n",
       "      <th>PAY_AMT4</th>\n",
       "      <th>PAY_AMT5</th>\n",
       "      <th>PAY_AMT6</th>\n",
       "      <th>Unnamed: 24</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>n1000-1</td>\n",
       "      <td>500000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>542653</td>\n",
       "      <td>483003</td>\n",
       "      <td>473944</td>\n",
       "      <td>55000.0</td>\n",
       "      <td>40000.0</td>\n",
       "      <td>38000.0</td>\n",
       "      <td>20239</td>\n",
       "      <td>13750.0</td>\n",
       "      <td>13770</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>n1000-2</td>\n",
       "      <td>210000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>29</td>\n",
       "      <td>-2</td>\n",
       "      <td>-2</td>\n",
       "      <td>-2</td>\n",
       "      <td>-2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>n1000-3</td>\n",
       "      <td>150000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>27</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>89775</td>\n",
       "      <td>87725</td>\n",
       "      <td>40788</td>\n",
       "      <td>4031.0</td>\n",
       "      <td>10006.0</td>\n",
       "      <td>3266.0</td>\n",
       "      <td>4040</td>\n",
       "      <td>1698.0</td>\n",
       "      <td>800</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>n1000-4</td>\n",
       "      <td>20000</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>38</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>18240</td>\n",
       "      <td>17928</td>\n",
       "      <td>150</td>\n",
       "      <td>1699.0</td>\n",
       "      <td>1460.0</td>\n",
       "      <td>626.0</td>\n",
       "      <td>1750</td>\n",
       "      <td>150.0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>n1000-5</td>\n",
       "      <td>400000</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>...</td>\n",
       "      <td>7839</td>\n",
       "      <td>14837</td>\n",
       "      <td>7959</td>\n",
       "      <td>9677.0</td>\n",
       "      <td>11867.0</td>\n",
       "      <td>7839.0</td>\n",
       "      <td>14837</td>\n",
       "      <td>7959.0</td>\n",
       "      <td>5712</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        ID  LIMIT_BAL  SEX  EDUCATION  MARRIAGE  AGE  PAY_1  PAY_2  PAY_3  \\\n",
       "0  n1000-1     500000    1          1         2   29      0      0      0   \n",
       "1  n1000-2     210000    1          1         2   29     -2     -2     -2   \n",
       "2  n1000-3     150000    1          1         2   27      0      0      0   \n",
       "3  n1000-4      20000    1          2         1   38      0      0      0   \n",
       "4  n1000-5     400000    1          2         1   34     -1     -1     -1   \n",
       "\n",
       "   PAY_4  ...  BILL_AMT4  BILL_AMT5  BILL_AMT6  PAY_AMT1  PAY_AMT2  PAY_AMT3  \\\n",
       "0      0  ...     542653     483003     473944   55000.0   40000.0   38000.0   \n",
       "1     -2  ...          0          0          0       0.0       0.0       0.0   \n",
       "2      0  ...      89775      87725      40788    4031.0   10006.0    3266.0   \n",
       "3      0  ...      18240      17928        150    1699.0    1460.0     626.0   \n",
       "4     -1  ...       7839      14837       7959    9677.0   11867.0    7839.0   \n",
       "\n",
       "   PAY_AMT4  PAY_AMT5  PAY_AMT6  Unnamed: 24  \n",
       "0     20239   13750.0     13770          NaN  \n",
       "1         0       0.0         0          NaN  \n",
       "2      4040    1698.0       800          NaN  \n",
       "3      1750     150.0         0          NaN  \n",
       "4     14837    7959.0      5712          NaN  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_appl.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Starting with some transformations like getting dummy variables and scaling/normalising in our new dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "#creating the dummy variables of \"Sex, Education, Marriage, and Pay\"\n",
    "\n",
    "new_appl = pd.get_dummies(new_appl, columns = ['SEX'])\n",
    "new_appl = new_appl.drop([\"SEX_2\"], axis = 1)\n",
    "new_appl = new_appl.rename({'SEX_1': 'GENDER'}, axis=1) # here we will notice that instead of 1 = male and 2 = female, we have 1 = male and 0 = female\n",
    "\n",
    "# continuing with education\n",
    "\n",
    "new_appl = pd.get_dummies(new_appl, columns = ['EDUCATION']) \n",
    "new_appl = new_appl.drop([\"EDUCATION_0\", \"EDUCATION_5\", \"EDUCATION_6\"], axis = 1) # so at this part, and based on our details of the data set we can notice that for education that we have 4 = other and 0, 5, 6 = unknown what we will do is to combine these two options because in case we have all 0 in our dataframe means that our costumer belongs in one of these two categories\n",
    "new_appl = new_appl.rename({'EDUCATION_1' : 'GRADUATE', 'EDUCATION_2' : 'UNDERGRADUATE', 'EDUCATION_3' : 'HIGH SCHOOL', 'EDUCATION_4' : 'OTHER_UNKNOWN'}, axis = 1) # so in every column based on the name of the column if we find 1 means that this guys has this kind of education, otherwise its 0 means he belongs to a different category\n",
    "\n",
    "# continuing with marriage\n",
    "new_appl = pd.get_dummies(new_appl, columns = ['MARRIAGE'])\n",
    "new_appl['MARRIAGE_0_3'] = (new_appl.MARRIAGE_0 | new_appl.MARRIAGE_3).astype(int) # same as education we combine other, unknown in one dummy variable because as far as we get in both married, single 0 we know he belongs to the other options\n",
    "new_appl = new_appl.drop([\"MARRIAGE_0\", \"MARRIAGE_3\"], axis = 1)\n",
    "new_appl = new_appl.rename({'MARRIAGE_1' : 'MARRIED', 'MARRIAGE_2' : 'SINGLE', 'MARRIAGE_0_3' : 'NOT MARRIED-NOT SINGLE-OTHER'}, axis = 1) # same as education about 1 and 0 (the explanation), so if for both cases is 0 we know that our costumer belongs to other or unknown\n",
    "\n",
    "# last creating the pay into dummies too\n",
    "\n",
    "new_appl[\"PAY_1\"] = new_appl[\"PAY_1\"].replace(-2, 0).replace( -1, 0)\n",
    "new_appl[\"PAY_1\"] = new_appl[\"PAY_1\"].replace(2, 1).replace(3, 1).replace(4, 1).replace(5, 1).replace(6, 1).replace(7, 1).replace(8, 1).replace(9, 1)\n",
    "\n",
    "new_appl[\"PAY_2\"] = new_appl[\"PAY_2\"].replace(-2, 0).replace(-1, 0)\n",
    "new_appl[\"PAY_2\"] = new_appl[\"PAY_2\"].replace(2, 1).replace(3, 1).replace(4, 1).replace(5, 1).replace(6, 1).replace(7, 1).replace(8, 1).replace(9, 1)\n",
    "\n",
    "new_appl[\"PAY_3\"] = new_appl[\"PAY_3\"].replace(-2, 0).replace(-1, 0)\n",
    "new_appl[\"PAY_3\"] = new_appl[\"PAY_3\"].replace(2, 1).replace(3, 1).replace(4, 1).replace(5, 1).replace(6, 1).replace(7, 1).replace(8, 1).replace(9, 1)\n",
    "\n",
    "new_appl[\"PAY_4\"] = new_appl[\"PAY_4\"].replace(-2, 0).replace(-1, 0)\n",
    "new_appl[\"PAY_4\"] = new_appl[\"PAY_4\"].replace(2, 1).replace(3, 1).replace(4, 1).replace(5, 1).replace(6, 1).replace(7, 1).replace(8, 1).replace(9, 1)\n",
    "\n",
    "new_appl[\"PAY_5\"] = new_appl[\"PAY_5\"].replace(-2, 0).replace(-1, 0)\n",
    "new_appl[\"PAY_5\"] = new_appl[\"PAY_5\"].replace(2, 1).replace(3, 1).replace(4, 1).replace(5, 1).replace(6, 1).replace(7, 1).replace(8, 1).replace(9, 1)\n",
    "\n",
    "new_appl[\"PAY_6\"] = new_appl[\"PAY_6\"].replace(-2, 0).replace(-1, 0)\n",
    "new_appl[\"PAY_6\"] = new_appl[\"PAY_6\"].replace(2, 1).replace(3, 1).replace(4, 1).replace(5, 1).replace(6, 1).replace(7, 1).replace(8, 1).replace(9, 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Scaling/normalizing limit balance, bill amount and pay amount."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "#scaling limit bal, bill amt and pay amt \n",
    "\n",
    "new_appl[['LIMIT_BAL']] = preprocessing.scale(new_appl[['LIMIT_BAL']])\n",
    "\n",
    "new_appl[[\n",
    "    'BILL_AMT1', 'BILL_AMT2', 'BILL_AMT3', 'BILL_AMT4', 'BILL_AMT5',\n",
    "    'BILL_AMT6'\n",
    "]] = preprocessing.scale(new_appl[[\n",
    "    'BILL_AMT1', 'BILL_AMT2', 'BILL_AMT3', 'BILL_AMT4', 'BILL_AMT5',\n",
    "    'BILL_AMT6'\n",
    "]])\n",
    "\n",
    "new_appl[[\n",
    "    'PAY_AMT1', 'PAY_AMT2', 'PAY_AMT3', 'PAY_AMT4', 'PAY_AMT5',\n",
    "    'PAY_AMT6'\n",
    "]] = preprocessing.scale(new_appl[[\n",
    "    'PAY_AMT1', 'PAY_AMT2', 'PAY_AMT3', 'PAY_AMT4', 'PAY_AMT5',\n",
    "    'PAY_AMT6'\n",
    "]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "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>ID</th>\n",
       "      <th>LIMIT_BAL</th>\n",
       "      <th>AGE</th>\n",
       "      <th>PAY_1</th>\n",
       "      <th>PAY_2</th>\n",
       "      <th>PAY_3</th>\n",
       "      <th>PAY_4</th>\n",
       "      <th>PAY_5</th>\n",
       "      <th>PAY_6</th>\n",
       "      <th>BILL_AMT1</th>\n",
       "      <th>...</th>\n",
       "      <th>PAY_AMT6</th>\n",
       "      <th>Unnamed: 24</th>\n",
       "      <th>GENDER</th>\n",
       "      <th>GRADUATE</th>\n",
       "      <th>UNDERGRADUATE</th>\n",
       "      <th>HIGH SCHOOL</th>\n",
       "      <th>OTHER_UNKNOWN</th>\n",
       "      <th>MARRIED</th>\n",
       "      <th>SINGLE</th>\n",
       "      <th>NOT MARRIED-NOT SINGLE-OTHER</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>n1000-1</td>\n",
       "      <td>2.632642</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4.532902</td>\n",
       "      <td>...</td>\n",
       "      <td>0.734057</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>n1000-2</td>\n",
       "      <td>0.362718</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.716121</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.341001</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>n1000-3</td>\n",
       "      <td>-0.106921</td>\n",
       "      <td>27</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.510798</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.278543</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>n1000-4</td>\n",
       "      <td>-1.124473</td>\n",
       "      <td>38</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.459736</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.341001</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>n1000-5</td>\n",
       "      <td>1.849910</td>\n",
       "      <td>34</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.435671</td>\n",
       "      <td>...</td>\n",
       "      <td>0.104949</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 30 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        ID  LIMIT_BAL  AGE  PAY_1  PAY_2  PAY_3  PAY_4  PAY_5  PAY_6  \\\n",
       "0  n1000-1   2.632642   29      0      0      0      0      0      0   \n",
       "1  n1000-2   0.362718   29      0      0      0      0      0      0   \n",
       "2  n1000-3  -0.106921   27      0      0      0      0      0      0   \n",
       "3  n1000-4  -1.124473   38      0      0      0      0      0      0   \n",
       "4  n1000-5   1.849910   34      0      0      0      0      0      0   \n",
       "\n",
       "   BILL_AMT1  ...  PAY_AMT6  Unnamed: 24  GENDER  GRADUATE  UNDERGRADUATE  \\\n",
       "0   4.532902  ...  0.734057          NaN       1         1              0   \n",
       "1  -0.716121  ... -0.341001          NaN       1         1              0   \n",
       "2   0.510798  ... -0.278543          NaN       1         1              0   \n",
       "3  -0.459736  ... -0.341001          NaN       1         0              1   \n",
       "4  -0.435671  ...  0.104949          NaN       1         0              1   \n",
       "\n",
       "   HIGH SCHOOL  OTHER_UNKNOWN  MARRIED  SINGLE  NOT MARRIED-NOT SINGLE-OTHER  \n",
       "0            0              0        0       1                             0  \n",
       "1            0              0        0       1                             0  \n",
       "2            0              0        0       1                             0  \n",
       "3            0              0        1       0                             0  \n",
       "4            0              0        1       0                             0  \n",
       "\n",
       "[5 rows x 30 columns]"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_appl.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "# drop unnecessary columns\n",
    "\n",
    "new_appl = new_appl.drop(['Unnamed: 24'], axis = 1)\n",
    "new_appl1 = new_appl.drop(['ID'], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "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>LIMIT_BAL</th>\n",
       "      <th>AGE</th>\n",
       "      <th>PAY_1</th>\n",
       "      <th>PAY_2</th>\n",
       "      <th>PAY_3</th>\n",
       "      <th>PAY_4</th>\n",
       "      <th>PAY_5</th>\n",
       "      <th>PAY_6</th>\n",
       "      <th>BILL_AMT1</th>\n",
       "      <th>BILL_AMT2</th>\n",
       "      <th>...</th>\n",
       "      <th>PAY_AMT5</th>\n",
       "      <th>PAY_AMT6</th>\n",
       "      <th>GENDER</th>\n",
       "      <th>GRADUATE</th>\n",
       "      <th>UNDERGRADUATE</th>\n",
       "      <th>HIGH SCHOOL</th>\n",
       "      <th>OTHER_UNKNOWN</th>\n",
       "      <th>MARRIED</th>\n",
       "      <th>SINGLE</th>\n",
       "      <th>NOT MARRIED-NOT SINGLE-OTHER</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2.632642</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4.532902</td>\n",
       "      <td>5.458975</td>\n",
       "      <td>...</td>\n",
       "      <td>0.443790</td>\n",
       "      <td>0.734057</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.362718</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.716121</td>\n",
       "      <td>-0.717552</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.285717</td>\n",
       "      <td>-0.341001</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.106921</td>\n",
       "      <td>27</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.510798</td>\n",
       "      <td>0.573270</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.195630</td>\n",
       "      <td>-0.278543</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-1.124473</td>\n",
       "      <td>38</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.459736</td>\n",
       "      <td>-0.427226</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.277759</td>\n",
       "      <td>-0.341001</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.849910</td>\n",
       "      <td>34</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.435671</td>\n",
       "      <td>-0.572652</td>\n",
       "      <td>...</td>\n",
       "      <td>0.136548</td>\n",
       "      <td>0.104949</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   LIMIT_BAL  AGE  PAY_1  PAY_2  PAY_3  PAY_4  PAY_5  PAY_6  BILL_AMT1  \\\n",
       "0   2.632642   29      0      0      0      0      0      0   4.532902   \n",
       "1   0.362718   29      0      0      0      0      0      0  -0.716121   \n",
       "2  -0.106921   27      0      0      0      0      0      0   0.510798   \n",
       "3  -1.124473   38      0      0      0      0      0      0  -0.459736   \n",
       "4   1.849910   34      0      0      0      0      0      0  -0.435671   \n",
       "\n",
       "   BILL_AMT2  ...  PAY_AMT5  PAY_AMT6  GENDER  GRADUATE  UNDERGRADUATE  \\\n",
       "0   5.458975  ...  0.443790  0.734057       1         1              0   \n",
       "1  -0.717552  ... -0.285717 -0.341001       1         1              0   \n",
       "2   0.573270  ... -0.195630 -0.278543       1         1              0   \n",
       "3  -0.427226  ... -0.277759 -0.341001       1         0              1   \n",
       "4  -0.572652  ...  0.136548  0.104949       1         0              1   \n",
       "\n",
       "   HIGH SCHOOL  OTHER_UNKNOWN  MARRIED  SINGLE  NOT MARRIED-NOT SINGLE-OTHER  \n",
       "0            0              0        0       1                             0  \n",
       "1            0              0        0       1                             0  \n",
       "2            0              0        0       1                             0  \n",
       "3            0              0        1       0                             0  \n",
       "4            0              0        1       0                             0  \n",
       "\n",
       "[5 rows x 28 columns]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# checking the columns of our dataframe \n",
    "\n",
    "new_appl1.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we will follow the same steps by applying our predictions from before to the new applicants."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_prediction = new_appl1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([3.34020381e-02, 1.38227933e-01, 1.28201589e-01, 1.97066418e-01,\n",
       "       8.67576409e-02, 3.72842676e-01, 2.46955204e-01, 1.33720195e-01,\n",
       "       5.37817929e-01, 1.60785094e-01, 8.16329921e-02, 1.17530557e-01,\n",
       "       1.05330145e-01, 1.19789815e-01, 2.72265391e-01, 1.49788362e-01,\n",
       "       1.63732985e-01, 1.56499422e-01, 6.45401550e-01, 2.85663261e-01,\n",
       "       1.55355729e-01, 3.80947293e-01, 1.56649870e-01, 4.56316239e-01,\n",
       "       1.42779231e-01, 1.19891968e-01, 1.63286640e-01, 1.04539757e-01,\n",
       "       1.83321257e-01, 1.69274374e-01, 1.42244328e-01, 7.23867767e-01,\n",
       "       4.00360833e-01, 2.02379969e-01, 4.16298748e-01, 3.18162068e-01,\n",
       "       1.43047694e-01, 3.40516716e-01, 1.72892051e-01, 4.31071444e-01,\n",
       "       7.17605298e-01, 4.30981959e-01, 1.34016834e-01, 1.96321758e-01,\n",
       "       1.17765015e-01, 1.49702201e-01, 1.62607566e-01, 1.12955345e-01,\n",
       "       1.53383377e-01, 6.56327275e-04, 1.39389205e-01, 1.75610739e-01,\n",
       "       1.75506408e-01, 1.19608300e-01, 1.48031585e-01, 1.31616722e-01,\n",
       "       1.28443693e-01, 1.24551463e-01, 1.16849196e-01, 1.45435999e-01,\n",
       "       1.84178888e-01, 1.22173920e-01, 1.33492811e-01, 2.80106988e-01,\n",
       "       4.53887396e-02, 3.74464130e-01, 1.48141510e-01, 7.64907474e-01,\n",
       "       1.33382298e-01, 4.50310607e-01, 4.41136795e-01, 1.66451572e-01,\n",
       "       1.57691344e-01, 1.38351856e-01, 5.04833733e-01, 9.37197194e-02,\n",
       "       4.74004431e-01, 7.58681450e-01, 1.35405197e-01, 5.80814486e-01,\n",
       "       3.05752963e-01, 4.03781222e-01, 1.85433773e-01, 6.56908511e-01,\n",
       "       1.45184881e-01, 1.38862091e-01, 3.47223585e-01, 1.35417120e-01,\n",
       "       1.23264411e-01, 1.15494340e-01, 1.57454025e-01, 4.16631988e-01,\n",
       "       2.33226650e-01, 7.47823673e-01, 1.38390236e-01, 1.64986549e-01,\n",
       "       1.81286102e-01, 1.63060285e-01, 2.28016053e-01, 6.11699271e-01,\n",
       "       2.13408784e-01, 1.52750763e-01, 4.09316999e-01, 1.06365506e-01,\n",
       "       1.30901967e-01, 1.00392310e-01, 1.84724795e-01, 1.61701512e-01,\n",
       "       7.83668128e-01, 1.04918519e-01, 6.97658039e-01, 1.91083685e-01,\n",
       "       2.74356385e-01, 2.07038347e-01, 2.28969582e-02, 1.85765016e-01,\n",
       "       8.09986396e-02, 9.05543460e-02, 1.06189348e-01, 1.70817489e-01,\n",
       "       1.32973328e-01, 7.52034766e-01, 3.63597452e-01, 5.18098280e-01,\n",
       "       6.05463690e-01, 1.54428625e-01, 1.80528264e-01, 6.63990795e-02,\n",
       "       1.83366447e-01, 1.41745890e-01, 1.23319404e-01, 1.66396975e-01,\n",
       "       3.93437328e-01, 1.34460819e-01, 7.27088055e-01, 9.91722680e-02,\n",
       "       1.37489262e-01, 1.36773944e-01, 7.78549082e-01, 3.40090518e-01,\n",
       "       1.57336716e-01, 1.32090658e-01, 1.97460125e-01, 1.39479241e-01,\n",
       "       1.08964831e-01, 7.51851865e-02, 1.61258061e-01, 1.66665597e-01,\n",
       "       3.84760346e-01, 8.19168004e-01, 9.92196170e-02, 1.69935036e-01,\n",
       "       2.10660891e-02, 2.02094992e-01, 1.26388848e-01, 4.23082720e-01,\n",
       "       7.77364728e-01, 1.10658276e-01, 2.40209454e-01, 2.07987296e-01,\n",
       "       1.74555006e-01, 1.50654583e-01, 1.16781433e-01, 1.74792144e-01,\n",
       "       4.59059118e-01, 7.52382195e-02, 6.53797039e-01, 6.95648104e-01,\n",
       "       1.50618406e-01, 1.31707032e-01, 1.96872728e-01, 4.14335616e-01,\n",
       "       1.65625447e-01, 1.51136681e-01, 1.11784774e-01, 1.39195728e-01,\n",
       "       1.47780669e-01, 2.67470402e-01, 1.50669991e-01, 1.76864968e-01,\n",
       "       1.47123735e-01, 1.49104770e-01, 4.36489589e-01, 1.11906933e-01,\n",
       "       1.61680495e-01, 5.53709715e-01, 1.85861731e-01, 5.95637191e-01,\n",
       "       2.20570305e-01, 3.30603243e-01, 4.35034851e-01, 8.71759008e-02,\n",
       "       1.82002926e-01, 4.65358785e-02, 1.21807273e-01, 1.29793302e-01,\n",
       "       2.52419529e-01, 1.14166821e-01, 1.16317522e-01, 7.55257648e-01,\n",
       "       6.12059777e-01, 4.79447650e-01, 1.48983862e-01, 1.31909610e-01,\n",
       "       1.62625365e-01, 1.49812397e-01, 5.25568339e-02, 1.51343957e-01,\n",
       "       4.32220751e-01, 1.60727555e-01, 1.78708354e-01, 5.36747121e-01,\n",
       "       1.12139940e-01, 1.10751469e-01, 1.14563438e-01, 1.37073019e-01,\n",
       "       8.90776655e-02, 9.12445946e-02, 1.66849272e-01, 1.58446404e-01,\n",
       "       3.94070757e-01, 1.54997369e-01, 3.80088334e-01, 1.62349655e-01,\n",
       "       1.46094399e-01, 8.13431259e-02, 5.67105591e-01, 3.70552963e-01,\n",
       "       1.51140654e-01, 9.05481688e-02, 1.78231588e-02, 7.80892957e-01,\n",
       "       3.76494816e-01, 1.37462706e-01, 1.35740874e-01, 1.49398313e-01,\n",
       "       1.32449133e-01, 2.15506048e-02, 3.08271056e-01, 2.22154303e-01,\n",
       "       6.48610837e-02, 1.21155774e-01, 5.84576385e-01, 3.94440856e-01,\n",
       "       7.12884386e-01, 1.34382084e-01, 6.54117958e-01, 1.17239124e-01,\n",
       "       1.36999373e-01, 1.07763543e-01, 7.45363641e-01, 6.50106786e-01,\n",
       "       1.29582289e-01, 2.89192590e-01, 5.90421792e-02, 1.94935337e-01,\n",
       "       2.96431590e-01, 3.71138474e-01, 3.46296560e-01, 1.18834227e-01,\n",
       "       1.15826376e-01, 8.78441529e-02, 1.51119726e-01, 9.67928272e-02,\n",
       "       1.26481308e-01, 4.77315692e-01, 1.90643107e-01, 1.61896233e-01,\n",
       "       2.49222922e-01, 1.61127749e-01, 1.05207654e-01, 1.25729002e-01,\n",
       "       1.22263570e-01, 1.47469746e-01, 1.30789368e-01, 4.96108271e-01,\n",
       "       1.03044011e-01, 1.32485322e-01, 8.49231032e-02, 4.37089641e-01,\n",
       "       4.01001408e-01, 1.30265235e-01, 1.57736500e-01, 1.35181722e-01,\n",
       "       1.60066722e-01, 3.37521162e-01, 1.20470669e-01, 3.82281862e-01,\n",
       "       5.23954771e-01, 1.33698463e-01, 4.69068415e-01, 9.69032897e-02,\n",
       "       1.54958107e-01, 1.47514165e-01, 9.70101552e-02, 1.95905340e-01,\n",
       "       1.57587033e-01, 1.38999236e-01, 1.05858931e-01, 1.46304807e-01,\n",
       "       7.47207749e-02, 1.29242929e-01, 9.10982737e-02, 4.59755101e-01,\n",
       "       1.38916097e-01, 9.33705213e-02, 1.67712062e-01, 1.95587239e-01,\n",
       "       1.64459106e-01, 1.77783344e-01, 1.36898099e-01, 4.14065889e-01,\n",
       "       7.52664999e-01, 1.59316272e-01, 5.56452410e-01, 1.66036924e-01,\n",
       "       6.43221773e-01, 1.32030486e-01, 1.56226373e-01, 1.47549176e-01,\n",
       "       1.39492877e-01, 1.29209575e-01, 1.32418851e-01, 1.59249216e-01,\n",
       "       1.47289969e-01, 1.20055076e-01, 9.58257006e-02, 1.53221304e-01,\n",
       "       1.10166679e-01, 1.46770463e-01, 1.50139516e-01, 1.39059645e-01,\n",
       "       1.77324242e-01, 1.33518496e-01, 1.19102078e-01, 1.60364011e-01,\n",
       "       1.98563160e-01, 2.14456909e-01, 5.47008781e-02, 6.13084996e-01,\n",
       "       1.57994428e-01, 9.59311308e-02, 1.36821650e-01, 1.53610195e-01,\n",
       "       1.45543943e-01, 3.88182810e-01, 3.30666468e-01, 1.11721910e-01,\n",
       "       4.43009975e-01, 1.45559378e-01, 3.16307280e-01, 8.09868181e-01,\n",
       "       2.16189547e-01, 3.30381998e-02, 2.41757422e-01, 7.74890963e-01,\n",
       "       1.64638251e-01, 5.36340045e-01, 1.98276534e-01, 1.41529770e-01,\n",
       "       7.85052824e-01, 1.58194994e-01, 1.33964808e-01, 1.41708758e-01,\n",
       "       4.14814201e-01, 4.58844550e-01, 1.27931984e-01, 1.48972331e-01,\n",
       "       1.65205306e-01, 1.21862372e-01, 1.86771981e-01, 5.51500855e-01,\n",
       "       1.97370319e-01, 1.08461710e-01, 1.88595537e-01, 1.26280298e-01,\n",
       "       1.01730705e-01, 1.50866960e-01, 1.46250306e-01, 1.16275988e-01,\n",
       "       3.93793885e-01, 1.37096494e-01, 5.05304317e-01, 4.95608721e-01,\n",
       "       9.51170955e-02, 8.45890974e-02, 1.25207009e-01, 7.53075563e-02,\n",
       "       1.14841058e-01, 1.68730963e-01, 4.69364691e-01, 4.09594330e-01,\n",
       "       7.63889938e-01, 7.48383190e-01, 1.38730418e-01, 1.01230223e-01,\n",
       "       1.54483731e-01, 2.06945059e-01, 5.17924384e-02, 1.08772438e-01,\n",
       "       7.31183885e-02, 1.10171518e-01, 1.92235717e-01, 1.29780724e-01,\n",
       "       5.72656755e-02, 1.10842063e-01, 6.32034168e-01, 1.74794744e-01,\n",
       "       1.06887784e-01, 7.73836116e-01, 1.24618949e-01, 5.99899171e-01,\n",
       "       1.66970515e-01, 9.39239455e-02, 2.11029975e-01, 1.39612528e-01,\n",
       "       7.48738042e-02, 1.30176222e-01, 1.29245862e-01, 1.41062411e-01,\n",
       "       4.47572022e-01, 5.26879121e-01, 9.81358096e-02, 4.70543696e-01,\n",
       "       8.69393574e-02, 1.33591767e-01, 1.15948889e-01, 4.74435298e-02,\n",
       "       1.10700345e-01, 1.23815645e-01, 1.10068393e-01, 4.71290926e-01,\n",
       "       1.71211103e-01, 3.80806961e-02, 6.36881472e-01, 6.37728753e-01,\n",
       "       1.44112322e-01, 1.40480721e-01, 1.76256377e-01, 2.52270593e-01,\n",
       "       1.49183301e-01, 1.19284571e-01, 2.45761349e-01, 1.03677553e-01,\n",
       "       7.17111484e-03, 8.77942784e-02, 1.64627122e-01, 1.58068494e-01,\n",
       "       1.51857454e-01, 1.32348023e-01, 1.26552226e-01, 2.02476044e-01,\n",
       "       8.09336893e-01, 7.34211132e-01, 4.61888587e-01, 1.08279263e-01,\n",
       "       1.01886572e-01, 1.72822297e-01, 7.67064137e-02, 1.46041845e-01,\n",
       "       1.10386928e-01, 6.66970962e-02, 1.49219452e-01, 7.78781671e-01,\n",
       "       1.22957711e-01, 7.71922274e-01, 1.91188005e-01, 2.57421423e-01,\n",
       "       1.37720378e-01, 3.56738299e-01, 7.75149825e-02, 1.56741571e-01,\n",
       "       1.51443695e-01, 4.01962656e-02, 1.02737994e-01, 1.88548502e-01,\n",
       "       1.19998732e-01, 3.27807403e-01, 1.59030785e-01, 9.56218408e-02,\n",
       "       9.29380686e-02, 1.03745162e-01, 3.63161191e-01, 6.48479939e-01,\n",
       "       1.18755786e-01, 7.77790260e-01, 9.25641462e-02, 1.09201087e-01,\n",
       "       1.39733459e-01, 1.49835007e-01, 1.36762217e-01, 1.17096275e-01,\n",
       "       1.38886159e-01, 4.04692256e-01, 1.39233387e-01, 1.80285581e-01,\n",
       "       9.48230295e-02, 1.18952974e-01, 4.46827769e-01, 8.06870641e-02,\n",
       "       1.53013659e-01, 8.05307354e-01, 4.00523267e-01, 1.51737876e-01,\n",
       "       6.23771297e-02, 1.62601179e-01, 1.60392595e-01, 9.04133368e-02,\n",
       "       7.19569315e-01, 7.88582220e-02, 9.08893943e-02, 1.85200423e-01,\n",
       "       1.72718253e-01, 1.17428324e-01, 1.67861124e-01, 1.77137809e-01,\n",
       "       6.76227906e-01, 4.39517618e-01, 4.40115552e-01, 9.89046696e-02,\n",
       "       2.42354707e-01, 1.03723902e-01, 1.22036141e-01, 7.04510035e-02,\n",
       "       1.37838107e-01, 8.71346479e-02, 1.34595271e-01, 1.36237600e-01,\n",
       "       1.96240458e-01, 3.46116247e-01, 1.46197498e-01, 9.36901783e-02,\n",
       "       4.36345812e-01, 1.18611393e-01, 1.66703919e-01, 1.48966213e-01,\n",
       "       1.39435919e-01, 4.05836067e-02, 1.36474079e-01, 1.29016444e-01,\n",
       "       6.44677487e-01, 3.85009993e-01, 7.42454250e-02, 5.17838713e-01,\n",
       "       3.48322885e-01, 1.46772726e-01, 2.21780458e-01, 1.31041757e-01,\n",
       "       1.69665697e-01, 6.05770053e-01, 7.85563085e-01, 1.67336577e-01,\n",
       "       1.54166741e-01, 8.08824183e-01, 4.19783896e-01, 1.56160431e-01,\n",
       "       5.64342591e-01, 3.12726912e-01, 4.48945802e-01, 6.76781416e-01,\n",
       "       1.02975192e-01, 1.27897825e-01, 1.12492875e-01, 7.95603920e-01,\n",
       "       1.38974783e-01, 3.84831877e-01, 1.49057580e-01, 3.55822554e-01,\n",
       "       4.07244646e-01, 1.43272910e-01, 3.48001166e-01, 1.27640695e-01,\n",
       "       1.27710748e-01, 1.15002814e-01, 4.83052020e-01, 4.36377926e-01,\n",
       "       1.16603793e-01, 3.12491925e-01, 8.79630870e-02, 8.69875646e-02,\n",
       "       3.76868328e-01, 7.72602741e-01, 2.07535178e-01, 4.05433129e-01,\n",
       "       4.80926347e-01, 2.06384712e-01, 1.79488218e-01, 1.54730851e-01,\n",
       "       4.48927781e-01, 1.00164842e-01, 2.46399485e-01, 1.17276746e-01,\n",
       "       1.35723774e-01, 1.00911664e-01, 1.30875185e-01, 1.71400553e-01,\n",
       "       7.92050169e-01, 9.92135702e-02, 1.63214116e-01, 6.39832293e-01,\n",
       "       5.64931768e-01, 1.17248083e-01, 1.69663291e-01, 4.15748235e-01,\n",
       "       8.10288822e-01, 1.84935584e-01, 2.06243835e-01, 1.05982174e-01,\n",
       "       1.51800156e-01, 1.68287318e-01, 1.60145948e-01, 1.48331278e-01,\n",
       "       3.11552172e-01, 1.39404716e-01, 1.43316888e-01, 1.00229599e-01,\n",
       "       1.30987995e-01, 2.61392098e-01, 9.85901252e-02, 1.22073056e-01,\n",
       "       1.78339085e-01, 1.11819860e-01, 9.87396746e-02, 5.72236455e-02,\n",
       "       1.29338799e-01, 1.08196439e-01, 2.10103004e-01, 6.34295425e-02,\n",
       "       1.78313476e-01, 1.08054126e-01, 9.04685243e-02, 1.24360387e-01,\n",
       "       3.39160758e-02, 3.65004230e-01, 1.80901674e-01, 3.79738503e-01,\n",
       "       3.46605957e-01, 1.43762066e-01, 1.22137841e-01, 7.96302385e-02,\n",
       "       1.24304658e-01, 1.58059400e-01, 7.96392937e-02, 3.28318663e-01,\n",
       "       1.60788389e-01, 1.05729485e-01, 7.97495659e-02, 5.41900955e-01,\n",
       "       1.78793848e-01, 1.21852823e-01, 1.23554766e-01, 4.39745233e-01,\n",
       "       1.60789478e-01, 2.73230305e-01, 1.31897560e-01, 1.67497445e-01,\n",
       "       3.11753599e-01, 9.37469262e-02, 5.40689754e-02, 5.03622994e-01,\n",
       "       9.66391505e-02, 4.73405390e-01, 1.44226910e-01, 9.54536142e-02,\n",
       "       1.70289751e-01, 1.63745910e-01, 1.28087114e-01, 1.19620884e-01,\n",
       "       1.15491835e-01, 1.78300467e-01, 2.64569475e-01, 1.53193721e-01,\n",
       "       5.44030618e-01, 1.32346862e-03, 1.70875715e-01, 1.83234166e-01,\n",
       "       1.74967955e-01, 1.60853464e-01, 1.34761488e-01, 1.54324199e-01,\n",
       "       1.27803120e-01, 2.02351631e-01, 2.52771320e-01, 1.49517719e-01,\n",
       "       1.17006651e-01, 5.52909682e-01, 1.19094899e-01, 2.02147152e-01,\n",
       "       1.76928575e-01, 1.00473718e-01, 1.94812863e-01, 7.49488579e-01,\n",
       "       1.09611159e-01, 1.46055055e-01, 1.71869369e-01, 2.61084124e-01,\n",
       "       1.63633175e-01, 1.38192528e-01, 1.89153820e-01, 1.92760147e-01,\n",
       "       1.13832068e-01, 7.41325544e-01, 1.96050427e-01, 1.13555699e-01,\n",
       "       1.24771094e-01, 1.03471589e-01, 3.80444627e-01, 3.01943652e-01,\n",
       "       2.31015778e-01, 3.52258265e-01, 7.97998045e-02, 1.36490145e-01,\n",
       "       1.21751345e-01, 1.06679893e-01, 2.50051794e-02, 3.14701777e-01,\n",
       "       9.21118040e-02, 1.33078097e-01, 1.39357922e-01, 6.86158867e-02,\n",
       "       1.63522165e-01, 1.78910923e-01, 3.92542955e-01, 7.71562266e-02,\n",
       "       1.53140679e-01, 2.04555904e-01, 7.93172809e-02, 5.65967054e-02,\n",
       "       2.36445251e-01, 1.36425279e-01, 8.88001240e-02, 1.10952346e-01,\n",
       "       9.77722565e-02, 1.54232896e-01, 1.28926509e-01, 1.38024130e-01,\n",
       "       1.13165472e-01, 5.75085169e-02, 6.68270184e-01, 1.42387308e-01,\n",
       "       1.69109394e-01, 3.13214605e-01, 4.70287277e-02, 1.47338785e-01,\n",
       "       1.50906956e-01, 4.11550594e-01, 1.19990280e-01, 1.35396958e-01,\n",
       "       9.68475488e-02, 1.57072295e-01, 1.39363854e-01, 5.25141172e-02,\n",
       "       6.58727315e-02, 1.16943611e-01, 1.35081448e-01, 8.54104995e-02,\n",
       "       4.10315387e-01, 1.63152405e-01, 6.46495833e-02, 1.29318949e-01,\n",
       "       2.67954187e-02, 7.24106511e-01, 1.50824905e-01, 1.09772852e-01,\n",
       "       1.49565314e-01, 1.37984144e-01, 9.39952032e-02, 1.65157208e-01,\n",
       "       6.51504538e-01, 1.40268097e-01, 1.44564776e-01, 7.02754803e-02,\n",
       "       7.30994632e-01, 1.26514146e-01, 8.01010876e-02, 3.92057846e-01,\n",
       "       4.36071099e-02, 8.47921696e-02, 1.90121188e-01, 1.14559497e-01,\n",
       "       1.46515272e-01, 1.44039443e-01, 9.38703928e-02, 1.19613248e-01,\n",
       "       5.56333320e-02, 1.82873847e-01, 5.61632092e-01, 1.75775982e-01,\n",
       "       1.35383709e-01, 4.04129453e-01, 4.94971330e-01, 7.70771788e-01,\n",
       "       4.15970295e-01, 4.54109337e-01, 4.16355905e-01, 1.15173167e-01,\n",
       "       9.02359555e-03, 1.08607085e-01, 4.59228540e-01, 1.75027429e-01,\n",
       "       6.53355688e-02, 1.65478266e-01, 1.46555253e-01, 1.44446545e-01,\n",
       "       1.58911136e-01, 1.37199806e-01, 3.58484877e-01, 4.62752051e-01,\n",
       "       1.33979784e-01, 1.85798455e-01, 1.63446604e-01, 1.73919521e-01,\n",
       "       6.69776794e-02, 6.81747172e-01, 5.07821456e-01, 1.35985924e-01,\n",
       "       1.10196391e-01, 1.04172232e-01, 3.20077541e-01, 1.27388404e-01,\n",
       "       1.73059841e-01, 1.30048070e-01, 1.35938351e-01, 1.89185509e-01,\n",
       "       1.05613208e-01, 1.64275674e-01, 1.72062702e-01, 5.55149990e-01,\n",
       "       5.33985629e-02, 1.85395858e-01, 1.16085243e-01, 6.61534060e-01,\n",
       "       1.69866114e-01, 4.64083894e-01, 1.52632816e-01, 5.55057513e-01,\n",
       "       1.76538946e-01, 4.58383579e-01, 1.68285048e-01, 2.93108746e-01,\n",
       "       2.85623938e-02, 1.66148970e-01, 1.49950543e-01, 3.89440787e-01,\n",
       "       1.62520644e-01, 1.39810544e-01, 4.25600286e-01, 1.39275792e-01,\n",
       "       1.45779661e-01, 9.46839589e-02, 1.02738452e-01, 1.61171961e-01,\n",
       "       9.76294097e-03, 2.36043107e-01, 4.90779005e-02, 1.07279774e-01,\n",
       "       1.55028988e-01, 1.61047655e-01, 8.80267167e-02, 1.68230411e-01,\n",
       "       5.82763933e-02, 1.32582953e-01, 1.23586187e-01, 7.22640799e-02,\n",
       "       1.56796988e-01, 8.99293021e-02, 1.46252835e-01, 1.35261997e-01,\n",
       "       1.64532402e-01, 1.53621753e-01, 5.90346973e-02, 4.32725365e-01,\n",
       "       1.41702741e-01, 4.24662015e-01, 1.27207415e-01, 3.02130061e-01,\n",
       "       1.28439480e-01, 1.59928006e-01, 6.73734628e-02, 4.82902315e-01,\n",
       "       1.75140625e-01, 2.00837575e-01, 1.29380380e-01, 1.46851398e-01,\n",
       "       1.44366756e-01, 1.77586380e-01, 3.69300385e-01, 1.91060304e-01,\n",
       "       1.64218342e-01, 1.63605001e-01, 1.73117358e-01, 6.16088233e-01,\n",
       "       7.22943617e-01, 3.83970841e-01, 1.88399250e-01, 1.21957846e-01,\n",
       "       1.51066859e-01, 5.09348721e-01, 1.07006296e-01, 1.97157043e-01,\n",
       "       1.09170660e-01, 9.54931435e-02, 1.28247555e-01, 1.27526707e-01,\n",
       "       1.47101478e-01, 1.24547733e-01, 7.63427576e-01, 1.55983592e-01,\n",
       "       1.10118550e-01, 3.82575887e-01, 3.94424533e-01, 1.58646446e-01,\n",
       "       1.32385895e-01, 3.49146589e-02, 4.96422675e-01, 1.75563263e-01,\n",
       "       5.01740865e-01, 2.43244500e-01, 1.48120050e-01, 3.63663155e-01,\n",
       "       1.47167714e-01, 1.33618437e-01, 1.50926829e-01, 1.51789350e-01,\n",
       "       1.30438526e-01, 1.33939552e-01, 1.33263608e-01, 9.02615705e-02,\n",
       "       1.39279425e-01, 1.22009967e-01, 3.59563670e-01, 3.22796444e-01,\n",
       "       1.90202940e-01, 1.37282806e-01, 1.39300183e-01, 1.30372209e-01,\n",
       "       8.58459582e-02, 1.11994763e-01, 1.84069146e-01, 1.07400667e-01,\n",
       "       7.90629924e-01, 2.06710847e-01, 1.41467297e-01, 1.77288852e-01,\n",
       "       1.49641554e-01, 1.61367402e-01, 1.43256680e-01, 3.28682680e-01,\n",
       "       1.45418454e-01, 1.41267589e-01, 5.64185320e-01, 1.51976675e-01,\n",
       "       1.16769710e-01, 1.12152697e-01, 1.48312471e-01, 6.16126100e-01,\n",
       "       1.55826470e-01, 3.83916537e-02, 1.17542965e-01, 1.02527588e-01,\n",
       "       1.30599237e-01, 9.56015319e-02, 1.25422711e-01, 1.39742779e-01,\n",
       "       1.11164915e-01, 2.75669542e-01, 1.60283221e-01, 7.41190699e-01,\n",
       "       1.99982622e-03, 4.22849311e-01, 1.63384699e-01, 5.71351691e-01,\n",
       "       1.71462173e-01, 1.50313623e-01, 1.77854378e-01, 7.84277913e-01,\n",
       "       1.75147894e-01, 1.59839387e-01, 1.68094505e-01, 4.51829960e-02,\n",
       "       3.03649661e-01, 1.25759118e-01, 2.57229197e-01, 8.62351246e-02,\n",
       "       1.95039705e-01, 1.45219510e-01, 6.50839096e-01, 1.32660970e-01,\n",
       "       4.32514371e-01, 7.90867515e-01, 1.30310112e-01, 5.88354021e-01,\n",
       "       9.91982668e-02, 1.50595643e-01, 1.44253243e-01, 6.41458163e-02,\n",
       "       5.16348432e-01, 4.42018302e-01, 1.42903737e-01, 1.57076780e-01])"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# checking all the probabilities before creating a dataframe with only probabilities and default costumers only\n",
    "\n",
    "y_pred_prob = classifier_LR.predict_proba(x_prediction)[:,1]\n",
    "y_pred_prob"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "# here we using the opposite of 0, 1(from our first dataset) because based on our data 1 = issue credit and 0 = do not issue\n",
    "# use of the previous threshold at 0.19\n",
    "\n",
    "y_validation_pred = np.where(y_pred_prob < 0.19, 1, 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 1 1 0 1 0 0 1 0 1 1 1 1 1 0 1 1 1 0 0 1 0 1 0 1 1 1 1 1 1 1 0 0 0 0 0 1 0 1 0 0 0 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 0 1 0 1 0 0 1 1 1 0 1 0 0 1 0 0 0 1 0 1 1 0 1 1 1 1 0 0 0 1 1 1 1 0 0 0 1 0 1 1 1 1 1 0 1 0 0 0 0 1 1 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 1 0 1 0 1 1 1 0 0 1 1 0 1 1 1 1 1 0 0 1 1 1 0 1 0 0 1 0 0 1 1 1 1 0 1 0 0 1 1 0 0 1 1 1 1 1 0 1 1 1 1 0 1 1 0 1 0 0 0 0 1 1 1 1 1 0 1 1 0 0 0 1 1 1 1 1 1 0 1 1 0 1 1 1 1 1 1 1 1 0 1 0 1 1 1 0 0 1 1 1 0 0 1 1 1 1 1 0 0 1 1 0 0 0 1 0 1 1 1 0 0 1 0 1 0 0 0 0 1 1 1 1 1 1 0 0 1 0 1 1 1 1 1 1 0 1 1 1 0 0 1 1 1 1 0 1 0 0 1 0 1 1 1 1 0 1 1 1 1 1 1 1 0 1 1 1 0 1 1 1 0 0 1 0 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 1 0 1 1 1 1 1 0 0 1 0 1 0 0 0 1 0 0 1 0 0 1 0 1 1 1 0 0 1 1 1 1 1 0 0 1 1 1 1 1 1 1 0 1 0 0 1 1 1 1 1 1 0 0 0 0 1 1 1 0 1 1 1 1 0 1 1 1 0 1 1 0 1 0 1 1 0 1 1 1 1 1 0 0 1 0 1 1 1 1 1 1 1 0 1 1 0 0 1 1 1 0 1 1 0 1 1 1 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 1 1 0 1 0 0 0 1 0 1 1 1 1 1 1 1 0 1 1 1 1 0 0 1 0 1 1 1 1 1 1 1 0 1 1 1 1 0 1 1 0 0 1 1 1 1 1 0 1 1 1 1 1 1 1 0 0 0 1 0 1 1 1 1 1 1 1 0 0 1 1 0 1 1 1 1 1 1 1 0 0 1 0 0 1 0 1 1 0 0 1 1 0 0 1 0 0 0 0 1 1 1 0 1 0 1 0 0 1 0 1 1 1 0 0 1 0 1 1 0 0 0 0 0 0 1 1 0 1 0 1 1 1 1 1 0 1 1 0 0 1 1 0 0 1 0 1 1 1 1 1 0 1 1 1 1 0 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 0 1 0 0 1 1 1 1 1 1 0 1 1 1 0 1 1 1 0 1 0 1 1 0 1 1 0 1 0 1 1 1 1 1 1 1 1 0 1 0 1 1 1 1 1 1 1 1 0 0 1 1 0 1 0 1 1 0 0 1 1 1 0 1 1 1 0 1 0 0 1 1 1 0 0 0 0 1 1 1 1 1 0 1 1 1 1 1 1 0 1 1 0 1 1 0 1 1 1 1 1 1 1 1 1 0 1 1 0 1 1 1 0 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 0 1 1 1 1 1 1 0 1 1 1 0 1 1 0 1 1 0 1 1 1 1 1 1 1 0 1 1 0 0 0 0 0 0 1 1 1 0 1 1 1 1 1 1 1 0 0 1 1 1 1 1 0 0 1 1 1 0 1 1 1 1 1 1 1 1 0 1 1 1 0 1 0 1 0 1 0 1 0 1 1 1 0 1 1 0 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 0 1 0 1 1 1 0 1 0 1 1 1 1 0 0 1 1 1 0 0 0 1 1 1 0 1 0 1 1 1 1 1 1 0 1 1 0 0 1 1 1 0 1 0 0 1 0 1 1 1 1 1 1 1 1 1 1 0 0 0 1 1 1 1 1 1 1 0 0 1 1 1 1 1 0 1 1 0 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0 1 0 1 0 1 0 1 1 1 0 1 1 1 1 0 1 0 1 0 1 0 1 0 0 1 0 1 1 1 1 0 0 1 1\n"
     ]
    }
   ],
   "source": [
    "print(*y_validation_pred, sep = \" \")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "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>ID</th>\n",
       "      <th>Issued</th>\n",
       "      <th>Probabilities</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>n1000-1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.033402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>n1000-2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.138228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>n1000-3</td>\n",
       "      <td>1</td>\n",
       "      <td>0.128202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>n1000-4</td>\n",
       "      <td>0</td>\n",
       "      <td>0.197066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>n1000-5</td>\n",
       "      <td>1</td>\n",
       "      <td>0.086758</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>995</th>\n",
       "      <td>n1000-996</td>\n",
       "      <td>1</td>\n",
       "      <td>0.064146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>n1000-997</td>\n",
       "      <td>0</td>\n",
       "      <td>0.516348</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>n1000-998</td>\n",
       "      <td>0</td>\n",
       "      <td>0.442018</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>n1000-999</td>\n",
       "      <td>1</td>\n",
       "      <td>0.142904</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>999</th>\n",
       "      <td>n1000-1000</td>\n",
       "      <td>1</td>\n",
       "      <td>0.157077</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             ID  Issued  Probabilities\n",
       "0       n1000-1       1       0.033402\n",
       "1       n1000-2       1       0.138228\n",
       "2       n1000-3       1       0.128202\n",
       "3       n1000-4       0       0.197066\n",
       "4       n1000-5       1       0.086758\n",
       "..          ...     ...            ...\n",
       "995   n1000-996       1       0.064146\n",
       "996   n1000-997       0       0.516348\n",
       "997   n1000-998       0       0.442018\n",
       "998   n1000-999       1       0.142904\n",
       "999  n1000-1000       1       0.157077\n",
       "\n",
       "[1000 rows x 3 columns]"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# now we are going to add the y_validation_pred in a dataframe with the id and the probabilities \n",
    "\n",
    "final_df = pd.DataFrame({'ID' : new_appl['ID'], 'Issued' : y_validation_pred, 'Probabilities': y_pred_prob})\n",
    "final_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.682"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# getting the percentage of people that are issued the loan\n",
    "\n",
    "np.mean(final_df.Issued)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looks like our percentage of people who should be issued credit is up to 68%."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "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>ID</th>\n",
       "      <th>Issued</th>\n",
       "      <th>Probabilities</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>n1000-50</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000656</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>673</th>\n",
       "      <td>n1000-674</td>\n",
       "      <td>1</td>\n",
       "      <td>0.001323</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>968</th>\n",
       "      <td>n1000-969</td>\n",
       "      <td>1</td>\n",
       "      <td>0.002000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>444</th>\n",
       "      <td>n1000-445</td>\n",
       "      <td>1</td>\n",
       "      <td>0.007171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>796</th>\n",
       "      <td>n1000-797</td>\n",
       "      <td>1</td>\n",
       "      <td>0.009024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>894</th>\n",
       "      <td>n1000-895</td>\n",
       "      <td>1</td>\n",
       "      <td>0.188399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>475</th>\n",
       "      <td>n1000-476</td>\n",
       "      <td>1</td>\n",
       "      <td>0.188549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>374</th>\n",
       "      <td>n1000-375</td>\n",
       "      <td>1</td>\n",
       "      <td>0.188596</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>698</th>\n",
       "      <td>n1000-699</td>\n",
       "      <td>1</td>\n",
       "      <td>0.189154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>823</th>\n",
       "      <td>n1000-824</td>\n",
       "      <td>1</td>\n",
       "      <td>0.189186</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>682 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            ID  Issued  Probabilities\n",
       "49    n1000-50       1       0.000656\n",
       "673  n1000-674       1       0.001323\n",
       "968  n1000-969       1       0.002000\n",
       "444  n1000-445       1       0.007171\n",
       "796  n1000-797       1       0.009024\n",
       "..         ...     ...            ...\n",
       "894  n1000-895       1       0.188399\n",
       "475  n1000-476       1       0.188549\n",
       "374  n1000-375       1       0.188596\n",
       "698  n1000-699       1       0.189154\n",
       "823  n1000-824       1       0.189186\n",
       "\n",
       "[682 rows x 3 columns]"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# now we are going to sort our dataframe based on the probabilities in order to find the \"best\" and \"worst\" clients (of the ones who are issued credit)\n",
    "\n",
    "final_df = final_df[final_df['Issued'] == 1].sort_values('Probabilities', ascending = True)\n",
    "final_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Based on the results above we can find the 3 \"best\" clients and the 3 \"worst\" ones. The ones with the lowest probability are the ones we suggest that they should be issued with a loan (n1000-50, n1000-674, n1000-969), and the ones with the highest probability are the ones that we should try avoid giving them the loan (n1000-699, n1000-779, n1000-933)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "14ca62385630e80b5c4ce7904863b62bddb6543d945de0683be07a870abfe95e"
  },
  "kernelspec": {
   "display_name": "Python 3.8.12 64-bit ('base': conda)",
   "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"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}