From 6386fde2300d45c80a29e99651db8cbe1831bcc0 Mon Sep 17 00:00:00 2001
From: rakshakanshu <93030745+rakshakanshu@users.noreply.github.com>
Date: Tue, 12 Jul 2022 22:33:22 +0530
Subject: [PATCH 1/2] Reuploaded assignment 1

KMeans Clustering Using Sklearn
---
 Assignment 1/210156_Anshu_1.ipynb | 1833 +++++++++++++++++++++++++++++
 1 file changed, 1833 insertions(+)
 create mode 100644 Assignment 1/210156_Anshu_1.ipynb

diff --git a/Assignment 1/210156_Anshu_1.ipynb b/Assignment 1/210156_Anshu_1.ipynb
new file mode 100644
index 0000000..20a1e32
--- /dev/null
+++ b/Assignment 1/210156_Anshu_1.ipynb	
@@ -0,0 +1,1833 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 195,
+   "id": "f39b2a88",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import matplotlib as mpl\n",
+    "import matplotlib.pyplot as plt\n",
+    "import seaborn as sns\n",
+    "%matplotlib inline\n",
+    "from math import sqrt"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 196,
+   "id": "ecef9f64",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df = pd.read_csv(r'C:\\Users\\anshu\\Downloads\\Dataset.csv')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 197,
+   "id": "58a5f8f1",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "['CUST_ID', 'BALANCE', 'BALANCE_FREQUENCY', 'PURCHASES', 'ONEOFF_PURCHASES', 'INSTALLMENTS_PURCHASES', 'CASH_ADVANCE', 'PURCHASES_FREQUENCY', 'ONEOFF_PURCHASES_FREQUENCY', 'PURCHASES_INSTALLMENTS_FREQUENCY', 'CASH_ADVANCE_FREQUENCY', 'CASH_ADVANCE_TRX', 'PURCHASES_TRX', 'CREDIT_LIMIT', 'PAYMENTS', 'MINIMUM_PAYMENTS', 'PRC_FULL_PAYMENT', 'TENURE']\n"
+     ]
+    }
+   ],
+   "source": [
+    "columns_names=df.columns.tolist()\n",
+    "# tolist(), used to convert the data elements of an array into a list\n",
+    "print(columns_names)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 198,
+   "id": "4bb88690",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(8950, 18)"
+      ]
+     },
+     "execution_count": 198,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 199,
+   "id": "e3a1b590",
+   "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>CUST_ID</th>\n",
+       "      <th>BALANCE</th>\n",
+       "      <th>BALANCE_FREQUENCY</th>\n",
+       "      <th>PURCHASES</th>\n",
+       "      <th>ONEOFF_PURCHASES</th>\n",
+       "      <th>INSTALLMENTS_PURCHASES</th>\n",
+       "      <th>CASH_ADVANCE</th>\n",
+       "      <th>PURCHASES_FREQUENCY</th>\n",
+       "      <th>ONEOFF_PURCHASES_FREQUENCY</th>\n",
+       "      <th>PURCHASES_INSTALLMENTS_FREQUENCY</th>\n",
+       "      <th>CASH_ADVANCE_FREQUENCY</th>\n",
+       "      <th>CASH_ADVANCE_TRX</th>\n",
+       "      <th>PURCHASES_TRX</th>\n",
+       "      <th>CREDIT_LIMIT</th>\n",
+       "      <th>PAYMENTS</th>\n",
+       "      <th>MINIMUM_PAYMENTS</th>\n",
+       "      <th>PRC_FULL_PAYMENT</th>\n",
+       "      <th>TENURE</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>C10001</td>\n",
+       "      <td>40.900749</td>\n",
+       "      <td>0.818182</td>\n",
+       "      <td>95.40</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>95.4</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.166667</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.083333</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>2</td>\n",
+       "      <td>1000.0</td>\n",
+       "      <td>201.802084</td>\n",
+       "      <td>139.509787</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>12</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>C10002</td>\n",
+       "      <td>3202.467416</td>\n",
+       "      <td>0.909091</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>6442.945483</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.250000</td>\n",
+       "      <td>4</td>\n",
+       "      <td>0</td>\n",
+       "      <td>7000.0</td>\n",
+       "      <td>4103.032597</td>\n",
+       "      <td>1072.340217</td>\n",
+       "      <td>0.222222</td>\n",
+       "      <td>12</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>C10003</td>\n",
+       "      <td>2495.148862</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>773.17</td>\n",
+       "      <td>773.17</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>12</td>\n",
+       "      <td>7500.0</td>\n",
+       "      <td>622.066742</td>\n",
+       "      <td>627.284787</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>12</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>C10004</td>\n",
+       "      <td>1666.670542</td>\n",
+       "      <td>0.636364</td>\n",
+       "      <td>1499.00</td>\n",
+       "      <td>1499.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>205.788017</td>\n",
+       "      <td>0.083333</td>\n",
+       "      <td>0.083333</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.083333</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>7500.0</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>12</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>C10005</td>\n",
+       "      <td>817.714335</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>16.00</td>\n",
+       "      <td>16.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.083333</td>\n",
+       "      <td>0.083333</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1200.0</td>\n",
+       "      <td>678.334763</td>\n",
+       "      <td>244.791237</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>12</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "  CUST_ID      BALANCE  BALANCE_FREQUENCY  PURCHASES  ONEOFF_PURCHASES  \\\n",
+       "0  C10001    40.900749           0.818182      95.40              0.00   \n",
+       "1  C10002  3202.467416           0.909091       0.00              0.00   \n",
+       "2  C10003  2495.148862           1.000000     773.17            773.17   \n",
+       "3  C10004  1666.670542           0.636364    1499.00           1499.00   \n",
+       "4  C10005   817.714335           1.000000      16.00             16.00   \n",
+       "\n",
+       "   INSTALLMENTS_PURCHASES  CASH_ADVANCE  PURCHASES_FREQUENCY  \\\n",
+       "0                    95.4      0.000000             0.166667   \n",
+       "1                     0.0   6442.945483             0.000000   \n",
+       "2                     0.0      0.000000             1.000000   \n",
+       "3                     0.0    205.788017             0.083333   \n",
+       "4                     0.0      0.000000             0.083333   \n",
+       "\n",
+       "   ONEOFF_PURCHASES_FREQUENCY  PURCHASES_INSTALLMENTS_FREQUENCY  \\\n",
+       "0                    0.000000                          0.083333   \n",
+       "1                    0.000000                          0.000000   \n",
+       "2                    1.000000                          0.000000   \n",
+       "3                    0.083333                          0.000000   \n",
+       "4                    0.083333                          0.000000   \n",
+       "\n",
+       "   CASH_ADVANCE_FREQUENCY  CASH_ADVANCE_TRX  PURCHASES_TRX  CREDIT_LIMIT  \\\n",
+       "0                0.000000                 0              2        1000.0   \n",
+       "1                0.250000                 4              0        7000.0   \n",
+       "2                0.000000                 0             12        7500.0   \n",
+       "3                0.083333                 1              1        7500.0   \n",
+       "4                0.000000                 0              1        1200.0   \n",
+       "\n",
+       "      PAYMENTS  MINIMUM_PAYMENTS  PRC_FULL_PAYMENT  TENURE  \n",
+       "0   201.802084        139.509787          0.000000      12  \n",
+       "1  4103.032597       1072.340217          0.222222      12  \n",
+       "2   622.066742        627.284787          0.000000      12  \n",
+       "3     0.000000               NaN          0.000000      12  \n",
+       "4   678.334763        244.791237          0.000000      12  "
+      ]
+     },
+     "execution_count": 199,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.head()\n",
+    "# The head() function is used to get the first n rows. This function returns the first n rows for the object based on position. It is useful for quickly testing if your object has the right type of data in it"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 200,
+   "id": "1c893514",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "<class 'pandas.core.frame.DataFrame'>\n",
+      "RangeIndex: 8950 entries, 0 to 8949\n",
+      "Data columns (total 18 columns):\n",
+      " #   Column                            Non-Null Count  Dtype  \n",
+      "---  ------                            --------------  -----  \n",
+      " 0   CUST_ID                           8950 non-null   object \n",
+      " 1   BALANCE                           8950 non-null   float64\n",
+      " 2   BALANCE_FREQUENCY                 8950 non-null   float64\n",
+      " 3   PURCHASES                         8950 non-null   float64\n",
+      " 4   ONEOFF_PURCHASES                  8950 non-null   float64\n",
+      " 5   INSTALLMENTS_PURCHASES            8950 non-null   float64\n",
+      " 6   CASH_ADVANCE                      8950 non-null   float64\n",
+      " 7   PURCHASES_FREQUENCY               8950 non-null   float64\n",
+      " 8   ONEOFF_PURCHASES_FREQUENCY        8950 non-null   float64\n",
+      " 9   PURCHASES_INSTALLMENTS_FREQUENCY  8950 non-null   float64\n",
+      " 10  CASH_ADVANCE_FREQUENCY            8950 non-null   float64\n",
+      " 11  CASH_ADVANCE_TRX                  8950 non-null   int64  \n",
+      " 12  PURCHASES_TRX                     8950 non-null   int64  \n",
+      " 13  CREDIT_LIMIT                      8949 non-null   float64\n",
+      " 14  PAYMENTS                          8950 non-null   float64\n",
+      " 15  MINIMUM_PAYMENTS                  8637 non-null   float64\n",
+      " 16  PRC_FULL_PAYMENT                  8950 non-null   float64\n",
+      " 17  TENURE                            8950 non-null   int64  \n",
+      "dtypes: float64(14), int64(3), object(1)\n",
+      "memory usage: 1.2+ MB\n"
+     ]
+    }
+   ],
+   "source": [
+    "df.info()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 201,
+   "id": "7d77b352",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "CUST_ID                               0\n",
+       "BALANCE                               0\n",
+       "BALANCE_FREQUENCY                     0\n",
+       "PURCHASES                             0\n",
+       "ONEOFF_PURCHASES                      0\n",
+       "INSTALLMENTS_PURCHASES                0\n",
+       "CASH_ADVANCE                          0\n",
+       "PURCHASES_FREQUENCY                   0\n",
+       "ONEOFF_PURCHASES_FREQUENCY            0\n",
+       "PURCHASES_INSTALLMENTS_FREQUENCY      0\n",
+       "CASH_ADVANCE_FREQUENCY                0\n",
+       "CASH_ADVANCE_TRX                      0\n",
+       "PURCHASES_TRX                         0\n",
+       "CREDIT_LIMIT                          1\n",
+       "PAYMENTS                              0\n",
+       "MINIMUM_PAYMENTS                    313\n",
+       "PRC_FULL_PAYMENT                      0\n",
+       "TENURE                                0\n",
+       "dtype: int64"
+      ]
+     },
+     "execution_count": 201,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.isna().sum()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 202,
+   "id": "ab1eb418",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "CUST_ID                             0\n",
+       "BALANCE                             0\n",
+       "BALANCE_FREQUENCY                   0\n",
+       "PURCHASES                           0\n",
+       "ONEOFF_PURCHASES                    0\n",
+       "INSTALLMENTS_PURCHASES              0\n",
+       "CASH_ADVANCE                        0\n",
+       "PURCHASES_FREQUENCY                 0\n",
+       "ONEOFF_PURCHASES_FREQUENCY          0\n",
+       "PURCHASES_INSTALLMENTS_FREQUENCY    0\n",
+       "CASH_ADVANCE_FREQUENCY              0\n",
+       "CASH_ADVANCE_TRX                    0\n",
+       "PURCHASES_TRX                       0\n",
+       "CREDIT_LIMIT                        0\n",
+       "PAYMENTS                            0\n",
+       "MINIMUM_PAYMENTS                    0\n",
+       "PRC_FULL_PAYMENT                    0\n",
+       "TENURE                              0\n",
+       "dtype: int64"
+      ]
+     },
+     "execution_count": 202,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df['CREDIT_LIMIT'].fillna(value = df['CREDIT_LIMIT'].mean() , inplace = True)\n",
+    "df['MINIMUM_PAYMENTS'].fillna(value = df['MINIMUM_PAYMENTS'].mean() , inplace = True)\n",
+    "df.isna().sum()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 203,
+   "id": "34cec8c5",
+   "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>BALANCE</th>\n",
+       "      <th>BALANCE_FREQUENCY</th>\n",
+       "      <th>PURCHASES</th>\n",
+       "      <th>ONEOFF_PURCHASES</th>\n",
+       "      <th>INSTALLMENTS_PURCHASES</th>\n",
+       "      <th>CASH_ADVANCE</th>\n",
+       "      <th>PURCHASES_FREQUENCY</th>\n",
+       "      <th>ONEOFF_PURCHASES_FREQUENCY</th>\n",
+       "      <th>PURCHASES_INSTALLMENTS_FREQUENCY</th>\n",
+       "      <th>CASH_ADVANCE_FREQUENCY</th>\n",
+       "      <th>CASH_ADVANCE_TRX</th>\n",
+       "      <th>PURCHASES_TRX</th>\n",
+       "      <th>CREDIT_LIMIT</th>\n",
+       "      <th>PAYMENTS</th>\n",
+       "      <th>MINIMUM_PAYMENTS</th>\n",
+       "      <th>PRC_FULL_PAYMENT</th>\n",
+       "      <th>TENURE</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>40.900749</td>\n",
+       "      <td>0.818182</td>\n",
+       "      <td>95.40</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>95.4</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.166667</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.083333</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>2</td>\n",
+       "      <td>1000.0</td>\n",
+       "      <td>201.802084</td>\n",
+       "      <td>139.509787</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>12</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>3202.467416</td>\n",
+       "      <td>0.909091</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>6442.945483</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.250000</td>\n",
+       "      <td>4</td>\n",
+       "      <td>0</td>\n",
+       "      <td>7000.0</td>\n",
+       "      <td>4103.032597</td>\n",
+       "      <td>1072.340217</td>\n",
+       "      <td>0.222222</td>\n",
+       "      <td>12</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>2495.148862</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>773.17</td>\n",
+       "      <td>773.17</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>12</td>\n",
+       "      <td>7500.0</td>\n",
+       "      <td>622.066742</td>\n",
+       "      <td>627.284787</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>12</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>1666.670542</td>\n",
+       "      <td>0.636364</td>\n",
+       "      <td>1499.00</td>\n",
+       "      <td>1499.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>205.788017</td>\n",
+       "      <td>0.083333</td>\n",
+       "      <td>0.083333</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.083333</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>7500.0</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>864.206542</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>12</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>817.714335</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>16.00</td>\n",
+       "      <td>16.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.083333</td>\n",
+       "      <td>0.083333</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1200.0</td>\n",
+       "      <td>678.334763</td>\n",
+       "      <td>244.791237</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>12</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       BALANCE  BALANCE_FREQUENCY  PURCHASES  ONEOFF_PURCHASES  \\\n",
+       "0    40.900749           0.818182      95.40              0.00   \n",
+       "1  3202.467416           0.909091       0.00              0.00   \n",
+       "2  2495.148862           1.000000     773.17            773.17   \n",
+       "3  1666.670542           0.636364    1499.00           1499.00   \n",
+       "4   817.714335           1.000000      16.00             16.00   \n",
+       "\n",
+       "   INSTALLMENTS_PURCHASES  CASH_ADVANCE  PURCHASES_FREQUENCY  \\\n",
+       "0                    95.4      0.000000             0.166667   \n",
+       "1                     0.0   6442.945483             0.000000   \n",
+       "2                     0.0      0.000000             1.000000   \n",
+       "3                     0.0    205.788017             0.083333   \n",
+       "4                     0.0      0.000000             0.083333   \n",
+       "\n",
+       "   ONEOFF_PURCHASES_FREQUENCY  PURCHASES_INSTALLMENTS_FREQUENCY  \\\n",
+       "0                    0.000000                          0.083333   \n",
+       "1                    0.000000                          0.000000   \n",
+       "2                    1.000000                          0.000000   \n",
+       "3                    0.083333                          0.000000   \n",
+       "4                    0.083333                          0.000000   \n",
+       "\n",
+       "   CASH_ADVANCE_FREQUENCY  CASH_ADVANCE_TRX  PURCHASES_TRX  CREDIT_LIMIT  \\\n",
+       "0                0.000000                 0              2        1000.0   \n",
+       "1                0.250000                 4              0        7000.0   \n",
+       "2                0.000000                 0             12        7500.0   \n",
+       "3                0.083333                 1              1        7500.0   \n",
+       "4                0.000000                 0              1        1200.0   \n",
+       "\n",
+       "      PAYMENTS  MINIMUM_PAYMENTS  PRC_FULL_PAYMENT  TENURE  \n",
+       "0   201.802084        139.509787          0.000000      12  \n",
+       "1  4103.032597       1072.340217          0.222222      12  \n",
+       "2   622.066742        627.284787          0.000000      12  \n",
+       "3     0.000000        864.206542          0.000000      12  \n",
+       "4   678.334763        244.791237          0.000000      12  "
+      ]
+     },
+     "execution_count": 203,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df_drop = df.drop(labels=['CUST_ID'],axis = 1)\n",
+    "df_drop.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 204,
+   "id": "259bbb7b",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5, 1.0, 'Correlation between different fearures')"
+      ]
+     },
+     "execution_count": 204,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 720x720 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "correlation = df_drop.corr()\n",
+    "plt.figure(figsize=(10,10))\n",
+    "sns.heatmap(correlation, vmax=1, square=True,annot=True,cmap='cubehelix')\n",
+    "# annot: If True, write the data value in each cell\n",
+    "# You can customize the colors in your heatmap with the cmap parameter of the heatmap() function in seaborn\n",
+    "\n",
+    "plt.title('Correlation between different fearures')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 205,
+   "id": "a580d1e8",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "BALANCE                              2.393386\n",
+       "BALANCE_FREQUENCY                   -2.023266\n",
+       "PURCHASES                            8.144269\n",
+       "ONEOFF_PURCHASES                    10.045083\n",
+       "INSTALLMENTS_PURCHASES               7.299120\n",
+       "CASH_ADVANCE                         5.166609\n",
+       "PURCHASES_FREQUENCY                  0.060164\n",
+       "ONEOFF_PURCHASES_FREQUENCY           1.535613\n",
+       "PURCHASES_INSTALLMENTS_FREQUENCY     0.509201\n",
+       "CASH_ADVANCE_FREQUENCY               1.828686\n",
+       "CASH_ADVANCE_TRX                     5.721298\n",
+       "PURCHASES_TRX                        4.630655\n",
+       "CREDIT_LIMIT                         1.522549\n",
+       "PAYMENTS                             5.907620\n",
+       "MINIMUM_PAYMENTS                    13.867357\n",
+       "PRC_FULL_PAYMENT                     1.942820\n",
+       "TENURE                              -2.943017\n",
+       "dtype: float64"
+      ]
+     },
+     "execution_count": 205,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df_drop.skew()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 206,
+   "id": "7fc764ad",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from scipy.stats import skew"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 207,
+   "id": "d8d6de1b",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "BALANCE\n",
+      "2.392984897743557\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "BALANCE_FREQUENCY\n",
+      "-2.022926407947498\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "PURCHASES\n",
+      "8.14290403970826\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "ONEOFF_PURCHASES\n",
+      "10.043399272872994\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "INSTALLMENTS_PURCHASES\n",
+      "7.297896535603957\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "CASH_ADVANCE\n",
+      "5.165743121623765\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "PURCHASES_FREQUENCY\n",
+      "0.06015415199708297\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "ONEOFF_PURCHASES_FREQUENCY\n",
+      "1.535355406271309\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "PURCHASES_INSTALLMENTS_FREQUENCY\n",
+      "0.5091158200492595\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "CASH_ADVANCE_FREQUENCY\n",
+      "1.828379768476593\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "CASH_ADVANCE_TRX\n",
+      "5.720339281754528\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWAAAAFgCAYAAACFYaNMAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAX00lEQVR4nO3de7BlZXnn8e8PkIs4RJi0BLuZQEyPCnhBWodIxlLRok0s0RnRZjS0NUyYEpJ4KxV0akanQo2JVrxMBhxKHZpoQYiXSExAGQQvFUZsFUFu0opCK4HGlBFPRxR45o/1HlwcTncfmrPPe06f76dq117r2Wvt/Zymz68X717rXakqJEkLb7feDUjScmUAS1InBrAkdWIAS1InBrAkdbJH7wYmZe3atXXJJZf0bkOSADJbcZc9Ar7rrrt6tyBJ27XLBrAkLXYGsCR1YgBLUicGsCR1YgBLUicGsCR1YgBLUicGsCR1YgBLUicGsCR1YgBLUicGsCR1YgA3VcXU1BTeI0/SQjGAm61bt/LK91/C1q1be7ciaZkwgEf22HPv3i1IWkYMYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4MYEnqxACWpE4mHsBJdk/yjSSfaesHJLk0yc3tef/Rtmck2ZTkpiTHjepHJbm2vfaBJJl035I0aQtxBPw64IbR+unAZVW1GrisrZPkMGAdcDiwFjgrye5tn7OBU4DV7bF2AfqWpImaaAAnWQX8LvChUfl4YENb3gC8dFS/oKruqapbgE3As5IcBOxXVVdWVQHnjfaRpCVr0kfA7wPeAtw/qh1YVbcDtOfHtfpK4LbRdptbbWVbnlmXpCVtYgGc5MXAnVX1tbnuMkuttlOf7TNPSbIxycYtW7bM8WMlqY9JHgEfA7wkyfeAC4DnJ/kocEcbVqA939m23wwcPNp/FfDDVl81S/0hquqcqlpTVWtWrFgxnz+LJM27iQVwVZ1RVauq6hCGL9c+X1WvBi4C1rfN1gOfbssXAeuS7JXkUIYv265qwxR3Jzm6nf1w0mgfSVqy9ujwme8CLkxyMnArcAJAVV2X5ELgeuBe4LSquq/t81rgXGAf4OL2kKQlbUECuKquAK5oyz8Cjt3GdmcCZ85S3wgcMbkOJWnheSWcJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJwawJHViAEtSJxML4CR7J7kqyTeTXJfkna1+QJJLk9zcnvcf7XNGkk1Jbkpy3Kh+VJJr22sfSJJJ9S1JC2WSR8D3AM+vqqcBTwfWJjkaOB24rKpWA5e1dZIcBqwDDgfWAmcl2b2919nAKcDq9lg7wb4laUFMLIBr8NO2+qj2KOB4YEOrbwBe2paPBy6oqnuq6hZgE/CsJAcB+1XVlVVVwHmjfSRpyZroGHCS3ZNcDdwJXFpVXwEOrKrbAdrz49rmK4HbRrtvbrWVbXlmXZKWtIkGcFXdV1VPB1YxHM0esZ3NZxvXre3UH/oGySlJNibZuGXLlofdryQtpAU5C6KqfgxcwTB2e0cbVqA939k22wwcPNptFfDDVl81S322zzmnqtZU1ZoVK1bM548gSfNukmdBrEjy2La8D/AC4EbgImB922w98Om2fBGwLsleSQ5l+LLtqjZMcXeSo9vZDyeN9pGkJWuPCb73QcCGdibDbsCFVfWZJFcCFyY5GbgVOAGgqq5LciFwPXAvcFpV3dfe67XAucA+wMXtIUlL2sQCuKquAY6cpf4j4Nht7HMmcOYs9Y3A9saPJWnJ8Uo4SerEAJakTgxgSerEAJakTgxgSerEAJakTgxgSerEAJakTgxgSerEAJakTgxgSerEAJakTgxgSerEAJakTgxgSerEAJakTgxgSerEAJakTuYUwEmOmUtNkjR3cz0C/p9zrEmS5mi7N+VM8lvAs4EVSd44emk/YPdJNiZJu7od3RV5T+Axbbt/Mar/BHj5pJqSpOVguwFcVV8AvpDk3Kr6/gL1JEnLwo6OgKftleQc4JDxPlX1/Ek0JUnLwVwD+K+ADwIfAu6bXDuStHzMNYDvraqzJ9qJJC0zcz0N7W+SnJrkoCQHTD8m2pkk7eLmegS8vj2/eVQr4Dfmtx1JWj7mFMBVdeikG5Gk5WZOAZzkpNnqVXXe/LYjScvHXIcgnjla3hs4Fvg6YABL0k6a6xDEH47Xk/wK8BcT6UiSlomdnY5yK7B6PhuRpOVmrmPAf8Nw1gMMk/A8GbhwUk1J0nIw1zHg94yW7wW+X1WbJ9CPJC0bcxqCaJPy3MgwI9r+wM8n2ZQkLQdzvSPGK4CrgBOAVwBfSeJ0lJL0CMx1COLtwDOr6k6AJCuA/wt8fFKNSdKubq5nQew2Hb7Njx7GvpKkWcz1CPiSJJ8Fzm/rrwT+bjItSdLysKN7wv0mcGBVvTnJvwN+GwhwJfCxBehPknZZOxpGeB9wN0BVfbKq3lhVb2A4+n3fZFuTpF3bjgL4kKq6ZmaxqjYy3J5IkrSTdhTAe2/ntX3msxFJWm52FMBfTfL7M4tJTga+NpmWJGl52NFZEK8HPpXkVfwycNcAewIvm2BfkrTL224AV9UdwLOTPA84opX/tqo+P/HOJGkXN9f5gC8HLp9wL5K0rHg1myR1YgBLUicGsCR1YgBLUicGsCR1YgBLUicGsCR1MrEATnJwksuT3JDkuiSva/UDklya5Ob2vP9onzOSbEpyU5LjRvWjklzbXvtAkkyqb0laKJM8Ar4XeFNVPRk4GjgtyWHA6cBlVbUauKyt015bBxwOrAXOSrJ7e6+zgVOA1e2xdoJ9S9KCmFgAV9XtVfX1tnw3cAOwEjge2NA22wC8tC0fD1xQVfdU1S3AJuBZSQ4C9quqK6uqgPNG+0jSkrUgY8BJDgGOBL7CcIeN22EIaeBxbbOVwG2j3Ta32sq2PLM+2+eckmRjko1btmyZ159BkubbxAM4yWOATwCvr6qfbG/TWWq1nfpDi1XnVNWaqlqzYsWKh9+sJC2giQZwkkcxhO/HquqTrXxHG1agPU/fbXkzcPBo91XAD1t91Sx1SVrSJnkWRIAPAzdU1Z+NXroIWN+W1wOfHtXXJdkryaEMX7Zd1YYp7k5ydHvPk0b7SNKSNdfb0u+MY4DfA65NcnWrvQ14F3Bhu6vGrcAJAFV1XZILgesZzqA4rarua/u9FjiX4TZIF7eHJC1pEwvgqvoys4/fAhy7jX3OBM6cpb6RX04IL0m7BK+Ek6RODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6sQAlqRODGBJ6mRiAZzkI0nuTPKtUe2AJJcmubk97z967Ywkm5LclOS4Uf2oJNe21z6QJJPqWZIW0iSPgM8F1s6onQ5cVlWrgcvaOkkOA9YBh7d9zkqye9vnbOAUYHV7zHxPSVqSJhbAVfVF4B9nlI8HNrTlDcBLR/ULquqeqroF2AQ8K8lBwH5VdWVVFXDeaB9JWtIWegz4wKq6HaA9P67VVwK3jbbb3Gor2/LM+qySnJJkY5KNW7ZsmdfGJWm+LZYv4WYb163t1GdVVedU1ZqqWrNixYp5a06SJmGhA/iONqxAe76z1TcDB4+2WwX8sNVXzVKXpCVvoQP4ImB9W14PfHpUX5dkrySHMnzZdlUbprg7ydHt7IeTRvtI0pK2x6TeOMn5wHOBX02yGfhvwLuAC5OcDNwKnABQVdcluRC4HrgXOK2q7mtv9VqGMyr2AS5uD0la8iYWwFV14jZeOnYb258JnDlLfSNwxDy2JkmLwmL5Ek6Slh0DeIapqSmmpqZ6tyFpGTCAJakTA1iSOjGAJakTA1iSOjGAJakTA1iSOjGAJakTA1iSOjGAJakTA1iSOjGAJakTA1iSOjGAJakTA1iSOjGAJakTA3gbqoqpqSmqtnkTZkl6RAzgbdi6dSuvfP8lbN26tXcrknZRBvB27LHn3r1bkLQLM4AlqRMDWJI6MYAlqRMDWJI6MYAlqRMDWJI6MYAlqRMDWJI6MYB3wEuSJU2KAbwDXpIsaVIM4DnwkmRJk2AAS1InBvAcORYsab4ZwHPkWLCk+WYAPwyOBUuaTwawJHViAO8Ex4MlzQcDeCc4HixpPhjAO8nxYEmPlAEsSZ0YwDvJcWBJj5QBvJPu+8U9vOZ/f8FxYEk7zQB+BHbfa5/eLUhawgzgeeBwhKSdYQDPA09Lk7QzDOB54mlpkh4uA3geTU1NMTU11bsNSUuEATxhhrKkbTGAJakTA3gCZp4V4VkSkmZjAE/AzLMixuuGsaRpBvCEzDwrYnp969atvOJ9F7NlyxZDWFrmDOAOkjxwGbNHxNLyZQB3Mn0Z8/TwxPTZEvfff/+Dng1made1ZAI4ydokNyXZlOT03v3Mpz323PuBIL7rrrse9DwdzOMgnj5qnhnSHk1LS8uSCOAkuwP/C3gRcBhwYpLD+nY1/6bHicfP4y/wpsN4ZlhP16emph4S2ts6I2NbR9qGuLRw9ujdwBw9C9hUVd8FSHIBcDxw/Xx+yL0//9mDLpqYXp/5PP3abvfd/6D12fYF5rzv9j5vtos5pmtbtmzh1A1/z1nrn/2g9Q2nvhCA9Wddylnrn82+++77oPXpfcbbTk1NPbA+vb2kwXz/TmQpHOkkeTmwtqr+U1v/PeDfVNUfzNjuFOCUtvpE4KaH+VG/Ctz1CNtdKPY6GUupV1ha/S7nXu+qqrUzi0vlCDiz1B7yL0dVnQOcs9MfkmysqjU7u/9CstfJWEq9wtLq114fakmMAQObgYNH66uAH3bqRZLmxVIJ4K8Cq5McmmRPYB1wUeeeJOkRWRJDEFV1b5I/AD4L7A58pKqum8BH7fTwRQf2OhlLqVdYWv3a6wxL4ks4SdoVLZUhCEna5RjAktSJAczivsw5ycFJLk9yQ5Lrkryu1Q9IcmmSm9vz/r17nZZk9yTfSPKZtr6Ye31sko8nubH9Gf/WYu03yRva34FvJTk/yd6LpdckH0lyZ5JvjWrb7C3JGe337aYkxy2CXt/d/g5ck+RTSR67EL0u+wBeApc53wu8qaqeDBwNnNb6Ox24rKpWA5e19cXidcANo/XF3Ov7gUuq6knA0xj6XnT9JlkJ/BGwpqqOYPgyeh2Lp9dzgZkXGszaW/v7uw44vO1zVvs9XCjn8tBeLwWOqKqnAt8GzoDJ97rsA5jRZc5V9XNg+jLnRaGqbq+qr7fluxkCYiVDjxvaZhuAl3ZpcIYkq4DfBT40Ki/WXvcDngN8GKCqfl5VP2aR9stw1tI+SfYAHs1wLvyi6LWqvgj844zytno7Hrigqu6pqluATQy/hwtitl6r6nNVdW9b/X8M1xpMvFcDeAiz20brm1tt0UlyCHAk8BXgwKq6HYaQBh7XsbWx9wFvAe4f1RZrr78BbAH+Txsy+VCSfVmE/VbVD4D3ALcCtwP/VFWfYxH2OrKt3hb779x/BC5uyxPt1QCe42XOvSV5DPAJ4PVV9ZPe/cwmyYuBO6vqa717maM9gGcAZ1fVkcAUi2C4YTZt/PR44FDg8cC+SV7dt6udtmh/55K8nWHY72PTpVk2m7deDeAlcJlzkkcxhO/HquqTrXxHkoPa6wcBd/bqb+QY4CVJvscwlPP8JB9lcfYKw3/7zVX1lbb+cYZAXoz9vgC4paq2VNUvgE8Cz2Zx9jptW70tyt+5JOuBFwOvql9eIDHRXg3gRX6Zc5IwjFHeUFV/NnrpImB9W14PfHqhe5upqs6oqlVVdQjDn+Pnq+rVLMJeAarqH4DbkjyxlY5lmOJ0MfZ7K3B0kke3vxPHMnwfsBh7nbat3i4C1iXZK8mhwGrgqg79PSDJWuCtwEuqauvopcn2Oj1p93J+AL/D8M3nd4C39+5nRm+/zfC/PNcAV7fH7wD/kuGb5Zvb8wG9e53R93OBz7TlRdsr8HRgY/vz/Wtg/8XaL/BO4EbgW8BfAHstll6B8xnGpn/BcNR48vZ6A97eft9uAl60CHrdxDDWO/079sGF6NVLkSWpE4cgJKkTA1iSOjGAJakTA1iSOjGAJakTA1iSOjGANa+S/FqSC5J8J8n1Sf4uyb9ur70hyc+S/Mpo+0cn+ViSa9s0i19ul12T5Kcz3vs1Sf58Dj18M8n5M2rnJrmlvfbtJOe1GcZIcsXMaQaTvD7JWW15RZJfJPnPM7b5XpJPjNZfnuTc0fqLkmzMMM3ljUne0+rvSPKDJFePHo+d5ec4bvT6T9t0iFe33p+b5J/aHBYPvHfb741JPjxaf1WSv93Rn5sWngGsedOu0PoUcEVVPaGqDgPeBhzYNjmR4crDl412ex1wR1U9pYZpFk9mOEF+Z3t4MsPf6+e0iXXG3lxVTwOeCHwDuLxd/Xg+w5V7Y+taHeAEhhmyTpzlI9ckOXyWPo4A/hx4dQ1TiR4BfHe0yXur6umjx49nvkdVfXb6dYaLRV7V1k9qm3yphjksjgRenOSYVv8AcFSSY1qw/zHwh7P0rs4MYM2n5wG/qKoPTheq6uqq+lKSJwCPAf4LDw6yg4AfjLa/qarueQQ9/AeGq8Q+B7xktg1q8F7gHxjmgf44Q4DtBQ/MOvd44MttlxOBNwGrpo+aR97D8I/MTG8BzqyqG9tn3ltVZz2Cn2ubquqfGa7eWjn9WcCpDPNc/ynDTWy/u803UDcGsObTEcC2ZkI7keGI8kvAE5NMT034EeCtSa5M8sdJVo/22Wf8v+nAf59DD68E/rJ91mxHrGNfB55UVT9iuL5/epLudcBfVlUlORj4taq6Criwvf/YhcAzkvzmjPr2/iwA3jD62S7f4U+1HW2mtNXAF6drVfX3DHNFvIAhhLUIGcBaKOsYJra+n2EmrxNgOEJmmJf33cABwFfbMALAP4//Nx34r9v7gCTPBLZU1fcZ5h54RrZ/i57xVIPjYYjx8MM6hpCFYYa3maF+X+v9jO31NovxEMTzHua+0/5tkmsYjuQ/U8PkQsAD05euAR4FrNjJ99eEGcCaT9cBR80sJnkqwxHapW2qynWMgqyqflpVn6yqU4GPMkw2tDNOBJ7UPuM7wH7Av9/O9kfyy1sn/TVwbJJnAPtUuwtJe8/XtPe8CHjajKN0GIY8ngP8q1Ft1j+LefalGm6h8xTgtUmePnrtnQx/lmcC751wH9pJBrDm0+eBvZL8/nShHZW+H3hHVR3SHo8HVib59fZF0f5t2z0Z7sv3/Yf7wUl2Yziqfur05zBMYP6QYYgM/ohh/PkSGP4RAK5gGBI5v233RGDfqlo5es//wYwv7GqYn/e9wOtH5XcDbxudAbJbkjc+3J9rLqrq262vt7bPegrDbaH+BDgH+PUkL5zEZ+uRMYA1b2qYWu9lwAvbaWjXAe9gmJryUzM2/xRDkD0B+EKSaxnOTNjIMPn8w/Uc4Ac13Lpn2heBw9ImBQfeneSbDFOPPhN4Xg33AZx2PsONOS9o6yfO0vcnmH1s+cMMd9gAoKquYQjk85PcwDCF5EGj7cdjwFe3L/4eiQ8ynPlxKHA28Iaq+lkb8jkVeH/7B06LiNNRSlInHgFLUid77HgTaXHJcOPEE2aU/6qqzuzRz3xoV+L9yYzyLVX1stm2167BIQhJ6sQhCEnqxACWpE4MYEnqxACWpE7+P6tjZyOeu2EVAAAAAElFTkSuQmCC\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "PURCHASES_TRX\n",
+      "4.629879142324137\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "CREDIT_LIMIT\n",
+      "1.5222938500528531\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "PAYMENTS\n",
+      "5.906629644423578\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "MINIMUM_PAYMENTS\n",
+      "13.86503301346577\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "PRC_FULL_PAYMENT\n",
+      "1.942494313531204\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "TENURE\n",
+      "-2.942524021534386\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 360x360 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "for col in df_drop:\n",
+    "    print(col)\n",
+    "    print(skew(df_drop[col]))\n",
+    "    \n",
+    "    plt.figure()\n",
+    "    sns.displot(df_drop[col])\n",
+    "    plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 208,
+   "id": "e1c515ea",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df = np.sqrt(df_drop)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 209,
+   "id": "6e3dea08",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "BALANCE                             0.829498\n",
+       "BALANCE_FREQUENCY                  -2.819495\n",
+       "PURCHASES                           1.730752\n",
+       "ONEOFF_PURCHASES                    2.129460\n",
+       "INSTALLMENTS_PURCHASES              1.546939\n",
+       "CASH_ADVANCE                        1.486159\n",
+       "PURCHASES_FREQUENCY                -0.421872\n",
+       "ONEOFF_PURCHASES_FREQUENCY          0.724607\n",
+       "PURCHASES_INSTALLMENTS_FREQUENCY    0.130409\n",
+       "CASH_ADVANCE_FREQUENCY              0.706976\n",
+       "CASH_ADVANCE_TRX                    1.417779\n",
+       "PURCHASES_TRX                       1.185757\n",
+       "CREDIT_LIMIT                        0.680933\n",
+       "PAYMENTS                            1.951535\n",
+       "MINIMUM_PAYMENTS                    3.805340\n",
+       "PRC_FULL_PAYMENT                    1.297280\n",
+       "TENURE                             -3.064332\n",
+       "dtype: float64"
+      ]
+     },
+     "execution_count": 209,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.skew()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 211,
+   "id": "4ee0c061",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.preprocessing import StandardScaler\n",
+    "X_std = StandardScaler().fit_transform(df)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 212,
+   "id": "595850c8",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0, 0.5, 'Cumulative explained variance')"
+      ]
+     },
+     "execution_count": 212,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from sklearn.decomposition import PCA\n",
+    "pca = PCA().fit(X_std)\n",
+    "plt.plot(np.cumsum(pca.explained_variance_ratio_))\n",
+    "plt.xlim(0,7,1)\n",
+    "plt.xlabel('Number of components')\n",
+    "plt.ylabel('Cumulative explained variance')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 213,
+   "id": "88ef6afd",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.decomposition import PCA \n",
+    "sklearn_pca = PCA(n_components=7)\n",
+    "Y_sklearn = sklearn_pca.fit_transform(X_std)\n",
+    "#Y_sklearn = pd.DataFrame(Y_sklearn)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 214,
+   "id": "b16a8dcd",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[[-0.88512191 -2.48301693  0.23090242 ...  0.04336989 -0.38202562\n",
+      "  -0.35720617]\n",
+      " [-3.00034342  2.01508943 -0.16533426 ...  1.67093843 -0.28801526\n",
+      "   0.94274938]\n",
+      " [ 1.19172624  0.38517395 -1.92678896 ... -0.55010278 -0.23006842\n",
+      "   0.52287556]\n",
+      " ...\n",
+      " [ 0.10596162 -3.06675754  1.18931984 ... -2.96585047  1.26333337\n",
+      "   1.97973232]\n",
+      " [-2.84716017 -2.51797947 -0.29519488 ... -2.99036079  0.69668999\n",
+      "   1.77427724]\n",
+      " [-0.16460436 -0.5243077  -1.64424995 ... -4.69253162  1.53231934\n",
+      "   0.09281539]]\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(Y_sklearn)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 215,
+   "id": "6d138bc4",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(8950, 7)"
+      ]
+     },
+     "execution_count": 215,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "Y_sklearn.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 216,
+   "id": "72c6f986",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([[-0.88512191, -2.48301693,  0.23090242, ...,  0.04336989,\n",
+       "        -0.38202562, -0.35720617],\n",
+       "       [-3.00034342,  2.01508943, -0.16533426, ...,  1.67093843,\n",
+       "        -0.28801526,  0.94274938],\n",
+       "       [ 1.19172624,  0.38517395, -1.92678896, ..., -0.55010278,\n",
+       "        -0.23006842,  0.52287556],\n",
+       "       ...,\n",
+       "       [ 0.10596162, -3.06675754,  1.18931984, ..., -2.96585047,\n",
+       "         1.26333337,  1.97973232],\n",
+       "       [-2.84716017, -2.51797947, -0.29519488, ..., -2.99036079,\n",
+       "         0.69668999,  1.77427724],\n",
+       "       [-0.16460436, -0.5243077 , -1.64424995, ..., -4.69253162,\n",
+       "         1.53231934,  0.09281539]])"
+      ]
+     },
+     "execution_count": 216,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "Y_sklearn"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 217,
+   "id": "a7a6d91a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.cluster import KMeans\n",
+    "from sklearn.decomposition import PCA\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 218,
+   "id": "9f2b89fd",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "PCA(n_components=7)"
+      ]
+     },
+     "execution_count": 218,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "sklearn_pca"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 219,
+   "id": "151eb549",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data = pd.DataFrame(Y_sklearn , columns = ['PC1','PC2','PC3','PC4','PC5','PC6','PC7'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 220,
+   "id": "c2b410ee",
+   "metadata": {
+    "scrolled": true
+   },
+   "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>PC1</th>\n",
+       "      <th>PC2</th>\n",
+       "      <th>PC3</th>\n",
+       "      <th>PC4</th>\n",
+       "      <th>PC5</th>\n",
+       "      <th>PC6</th>\n",
+       "      <th>PC7</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>-0.885122</td>\n",
+       "      <td>-2.483017</td>\n",
+       "      <td>0.230902</td>\n",
+       "      <td>0.807129</td>\n",
+       "      <td>0.043370</td>\n",
+       "      <td>-0.382026</td>\n",
+       "      <td>-0.357206</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>-3.000343</td>\n",
+       "      <td>2.015089</td>\n",
+       "      <td>-0.165334</td>\n",
+       "      <td>-1.087171</td>\n",
+       "      <td>1.670938</td>\n",
+       "      <td>-0.288015</td>\n",
+       "      <td>0.942749</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>1.191726</td>\n",
+       "      <td>0.385174</td>\n",
+       "      <td>-1.926789</td>\n",
+       "      <td>1.859338</td>\n",
+       "      <td>-0.550103</td>\n",
+       "      <td>-0.230068</td>\n",
+       "      <td>0.522876</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>-0.794805</td>\n",
+       "      <td>0.218433</td>\n",
+       "      <td>-1.661542</td>\n",
+       "      <td>1.195618</td>\n",
+       "      <td>0.058950</td>\n",
+       "      <td>0.798510</td>\n",
+       "      <td>-0.086756</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>-1.265058</td>\n",
+       "      <td>-1.593317</td>\n",
+       "      <td>-0.689436</td>\n",
+       "      <td>1.339644</td>\n",
+       "      <td>-0.114019</td>\n",
+       "      <td>-0.837737</td>\n",
+       "      <td>0.231600</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",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8945</th>\n",
+       "      <td>0.668484</td>\n",
+       "      <td>-2.871696</td>\n",
+       "      <td>1.452469</td>\n",
+       "      <td>-2.236975</td>\n",
+       "      <td>-2.854943</td>\n",
+       "      <td>0.766981</td>\n",
+       "      <td>2.343375</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8946</th>\n",
+       "      <td>0.262604</td>\n",
+       "      <td>-2.240280</td>\n",
+       "      <td>1.844972</td>\n",
+       "      <td>-0.568118</td>\n",
+       "      <td>-3.339266</td>\n",
+       "      <td>1.706628</td>\n",
+       "      <td>1.774529</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8947</th>\n",
+       "      <td>0.105962</td>\n",
+       "      <td>-3.066758</td>\n",
+       "      <td>1.189320</td>\n",
+       "      <td>-1.775107</td>\n",
+       "      <td>-2.965850</td>\n",
+       "      <td>1.263333</td>\n",
+       "      <td>1.979732</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8948</th>\n",
+       "      <td>-2.847160</td>\n",
+       "      <td>-2.517979</td>\n",
+       "      <td>-0.295195</td>\n",
+       "      <td>-2.148352</td>\n",
+       "      <td>-2.990361</td>\n",
+       "      <td>0.696690</td>\n",
+       "      <td>1.774277</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8949</th>\n",
+       "      <td>-0.164604</td>\n",
+       "      <td>-0.524308</td>\n",
+       "      <td>-1.644250</td>\n",
+       "      <td>-1.315554</td>\n",
+       "      <td>-4.692532</td>\n",
+       "      <td>1.532319</td>\n",
+       "      <td>0.092815</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>8950 rows × 7 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "           PC1       PC2       PC3       PC4       PC5       PC6       PC7\n",
+       "0    -0.885122 -2.483017  0.230902  0.807129  0.043370 -0.382026 -0.357206\n",
+       "1    -3.000343  2.015089 -0.165334 -1.087171  1.670938 -0.288015  0.942749\n",
+       "2     1.191726  0.385174 -1.926789  1.859338 -0.550103 -0.230068  0.522876\n",
+       "3    -0.794805  0.218433 -1.661542  1.195618  0.058950  0.798510 -0.086756\n",
+       "4    -1.265058 -1.593317 -0.689436  1.339644 -0.114019 -0.837737  0.231600\n",
+       "...        ...       ...       ...       ...       ...       ...       ...\n",
+       "8945  0.668484 -2.871696  1.452469 -2.236975 -2.854943  0.766981  2.343375\n",
+       "8946  0.262604 -2.240280  1.844972 -0.568118 -3.339266  1.706628  1.774529\n",
+       "8947  0.105962 -3.066758  1.189320 -1.775107 -2.965850  1.263333  1.979732\n",
+       "8948 -2.847160 -2.517979 -0.295195 -2.148352 -2.990361  0.696690  1.774277\n",
+       "8949 -0.164604 -0.524308 -1.644250 -1.315554 -4.692532  1.532319  0.092815\n",
+       "\n",
+       "[8950 rows x 7 columns]"
+      ]
+     },
+     "execution_count": 220,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 221,
+   "id": "63ea21d6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "pca = PCA(n_components = 2)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 222,
+   "id": "0adfdb90",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "PCA(n_components=2)"
+      ]
+     },
+     "execution_count": 222,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "pca.fit(data)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 223,
+   "id": "c5922337",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "pca =pca.transform(data)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 224,
+   "id": "6d65e836",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(8950, 2)"
+      ]
+     },
+     "execution_count": 224,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "pca.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 225,
+   "id": "252df0a8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.cluster import KMeans\n",
+    "wcss = [] \n",
+    "r = range(1,11)\n",
+    "for i in range(1,11): \n",
+    "    kmeans = KMeans(n_clusters = i, init = 'k-means++', random_state = 42)\n",
+    "    kmeans.fit(Y_sklearn) \n",
+    "    wcss.append(kmeans.inertia_)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 226,
+   "id": "28fd2e64",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "elbow=pd.DataFrame({'Clusters':i,'WSS':wcss})"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 227,
+   "id": "46ef1e28",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.plot(range(1,11),wcss,'bx-')\n",
+    "plt.plot(range(1,11), wcss)\n",
+    "plt.xlabel('PC')\n",
+    "plt.ylabel('Inertia')\n",
+    "plt.title('Elbow Method')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 228,
+   "id": "a5ba0ab6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "kmeans_pca = KMeans(n_clusters = 3 , init = 'k-means++')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 229,
+   "id": "2e63ebd0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "label = kmeans_pca.fit_predict(pca)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 230,
+   "id": "01bb5b66",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "X = data[[\"PC1\",\"PC2\",\"PC3\",\"PC4\",\"PC5\",\"PC6\",\"PC7\"]]\n",
+    "plt.scatter(X[\"PC1\"], X[\"PC2\"], c=\"r\")\n",
+    "plt.xlabel(\"PC1\")\n",
+    "plt.ylabel(\"PC2\")\n",
+    "plt.show()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 231,
+   "id": "6048af80",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "filtered_label0 = pca[label==0]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 232,
+   "id": "24adb1ae",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "show = np.unique(label)\n",
+    "for i in show:\n",
+    "    plt.scatter(pca[label == i , 0] ,pca[label == i , 1] , label = i)\n",
+    "plt.xlabel(\"PC1\")\n",
+    "plt.ylabel(\"PC2\")\n",
+    "plt.legend()\n",
+    "plt.show()"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

From e0ae2431960896895aff4bdc590dcc4933df3be1 Mon Sep 17 00:00:00 2001
From: rakshakanshu <93030745+rakshakanshu@users.noreply.github.com>
Date: Fri, 15 Jul 2022 00:06:51 +0530
Subject: [PATCH 2/2] Assingment 2 (code till cell 22)

---
 Assignment 2/210156_Anshu_2 (1).ipynb | 2692 +++++++++++++++++++++++++
 1 file changed, 2692 insertions(+)
 create mode 100644 Assignment 2/210156_Anshu_2 (1).ipynb

diff --git a/Assignment 2/210156_Anshu_2 (1).ipynb b/Assignment 2/210156_Anshu_2 (1).ipynb
new file mode 100644
index 0000000..e97f47f
--- /dev/null
+++ b/Assignment 2/210156_Anshu_2 (1).ipynb	
@@ -0,0 +1,2692 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "0d5f58b8",
+   "metadata": {},
+   "source": [
+    "## Environment Setup"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 154,
+   "id": "c0d00887",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# import pandas, numpy, seaborn, pyplot\n",
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import seaborn as sns\n",
+    "import matplotlib.pyplot as plt\n",
+    "import scipy.stats as stats"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 155,
+   "id": "8718c756",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# import train_test_split, RepeatedStratifiedKFold, cross_val_score, LogisticRegression, roc_curve, roc_auc_score,\n",
+    "# confusion_matrix, precision_recall_curve, auc, f_classif, Pipeline, BaseEstimator, TransformerMixin, chi2_contingency\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from sklearn.model_selection import RepeatedStratifiedKFold\n",
+    "from sklearn import datasets, linear_model \n",
+    "from sklearn.model_selection import cross_val_score\n",
+    "from sklearn.datasets import load_iris\n",
+    "from sklearn.linear_model import LogisticRegression\n",
+    "from sklearn import metrics\n",
+    "from sklearn.metrics import roc_auc_score\n",
+    "from sklearn.metrics import confusion_matrix\n",
+    "from sklearn.metrics import precision_recall_curve\n",
+    "from sklearn.metrics import auc\n",
+    "from sklearn.feature_selection import SelectFpr , f_classif\n",
+    "from sklearn.pipeline import Pipeline\n",
+    "from sklearn.base import BaseEstimator , TransformerMixin\n",
+    "from scipy.stats import chi2_contingency\n",
+    "from datetime import date\n",
+    "from sklearn.feature_selection import SelectKBest\n",
+    "from sklearn.feature_selection import chi2\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "452fe749",
+   "metadata": {},
+   "source": [
+    "## Import Data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 156,
+   "id": "8dfd3c5f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# load the dataset\n",
+    "df = pd.read_csv(r'C:\\Users\\anshu\\Downloads\\credit_risk_dataset.csv',low_memory=False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 157,
+   "id": "5db1ea7b",
+   "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>member_id</th>\n",
+       "      <th>loan_amnt</th>\n",
+       "      <th>funded_amnt</th>\n",
+       "      <th>funded_amnt_inv</th>\n",
+       "      <th>term</th>\n",
+       "      <th>int_rate</th>\n",
+       "      <th>installment</th>\n",
+       "      <th>grade</th>\n",
+       "      <th>sub_grade</th>\n",
+       "      <th>...</th>\n",
+       "      <th>total_bal_il</th>\n",
+       "      <th>il_util</th>\n",
+       "      <th>open_rv_12m</th>\n",
+       "      <th>open_rv_24m</th>\n",
+       "      <th>max_bal_bc</th>\n",
+       "      <th>all_util</th>\n",
+       "      <th>total_rev_hi_lim</th>\n",
+       "      <th>inq_fi</th>\n",
+       "      <th>total_cu_tl</th>\n",
+       "      <th>inq_last_12m</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1077501</td>\n",
+       "      <td>1296599</td>\n",
+       "      <td>5000</td>\n",
+       "      <td>5000</td>\n",
+       "      <td>4975.0</td>\n",
+       "      <td>36 months</td>\n",
+       "      <td>10.65</td>\n",
+       "      <td>162.87</td>\n",
+       "      <td>B</td>\n",
+       "      <td>B2</td>\n",
+       "      <td>...</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>1077430</td>\n",
+       "      <td>1314167</td>\n",
+       "      <td>2500</td>\n",
+       "      <td>2500</td>\n",
+       "      <td>2500.0</td>\n",
+       "      <td>60 months</td>\n",
+       "      <td>15.27</td>\n",
+       "      <td>59.83</td>\n",
+       "      <td>C</td>\n",
+       "      <td>C4</td>\n",
+       "      <td>...</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>1077175</td>\n",
+       "      <td>1313524</td>\n",
+       "      <td>2400</td>\n",
+       "      <td>2400</td>\n",
+       "      <td>2400.0</td>\n",
+       "      <td>36 months</td>\n",
+       "      <td>15.96</td>\n",
+       "      <td>84.33</td>\n",
+       "      <td>C</td>\n",
+       "      <td>C5</td>\n",
+       "      <td>...</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>1076863</td>\n",
+       "      <td>1277178</td>\n",
+       "      <td>10000</td>\n",
+       "      <td>10000</td>\n",
+       "      <td>10000.0</td>\n",
+       "      <td>36 months</td>\n",
+       "      <td>13.49</td>\n",
+       "      <td>339.31</td>\n",
+       "      <td>C</td>\n",
+       "      <td>C1</td>\n",
+       "      <td>...</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>1075358</td>\n",
+       "      <td>1311748</td>\n",
+       "      <td>3000</td>\n",
+       "      <td>3000</td>\n",
+       "      <td>3000.0</td>\n",
+       "      <td>60 months</td>\n",
+       "      <td>12.69</td>\n",
+       "      <td>67.79</td>\n",
+       "      <td>B</td>\n",
+       "      <td>B5</td>\n",
+       "      <td>...</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 74 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "        id  member_id  loan_amnt  funded_amnt  funded_amnt_inv        term  \\\n",
+       "0  1077501    1296599       5000         5000           4975.0   36 months   \n",
+       "1  1077430    1314167       2500         2500           2500.0   60 months   \n",
+       "2  1077175    1313524       2400         2400           2400.0   36 months   \n",
+       "3  1076863    1277178      10000        10000          10000.0   36 months   \n",
+       "4  1075358    1311748       3000         3000           3000.0   60 months   \n",
+       "\n",
+       "   int_rate  installment grade sub_grade  ... total_bal_il il_util  \\\n",
+       "0     10.65       162.87     B        B2  ...          NaN     NaN   \n",
+       "1     15.27        59.83     C        C4  ...          NaN     NaN   \n",
+       "2     15.96        84.33     C        C5  ...          NaN     NaN   \n",
+       "3     13.49       339.31     C        C1  ...          NaN     NaN   \n",
+       "4     12.69        67.79     B        B5  ...          NaN     NaN   \n",
+       "\n",
+       "  open_rv_12m  open_rv_24m max_bal_bc all_util total_rev_hi_lim inq_fi  \\\n",
+       "0         NaN          NaN        NaN      NaN              NaN    NaN   \n",
+       "1         NaN          NaN        NaN      NaN              NaN    NaN   \n",
+       "2         NaN          NaN        NaN      NaN              NaN    NaN   \n",
+       "3         NaN          NaN        NaN      NaN              NaN    NaN   \n",
+       "4         NaN          NaN        NaN      NaN              NaN    NaN   \n",
+       "\n",
+       "  total_cu_tl inq_last_12m  \n",
+       "0         NaN          NaN  \n",
+       "1         NaN          NaN  \n",
+       "2         NaN          NaN  \n",
+       "3         NaN          NaN  \n",
+       "4         NaN          NaN  \n",
+       "\n",
+       "[5 rows x 74 columns]"
+      ]
+     },
+     "execution_count": 157,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.head()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "88db398b",
+   "metadata": {},
+   "source": [
+    "## Data Exploration"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 158,
+   "id": "0a185963",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Index(['id', 'member_id', 'loan_amnt', 'funded_amnt', 'funded_amnt_inv',\n",
+       "       'term', 'int_rate', 'installment', 'grade', 'sub_grade', 'emp_title',\n",
+       "       'emp_length', 'home_ownership', 'annual_inc', 'verification_status',\n",
+       "       'issue_d', 'loan_status', 'pymnt_plan', 'url', 'desc', 'purpose',\n",
+       "       'title', 'zip_code', 'addr_state', 'dti', 'delinq_2yrs',\n",
+       "       'earliest_cr_line', 'inq_last_6mths', 'mths_since_last_delinq',\n",
+       "       'mths_since_last_record', 'open_acc', 'pub_rec', 'revol_bal',\n",
+       "       'revol_util', 'total_acc', 'initial_list_status', 'out_prncp',\n",
+       "       'out_prncp_inv', 'total_pymnt', 'total_pymnt_inv', 'total_rec_prncp',\n",
+       "       'total_rec_int', 'total_rec_late_fee', 'recoveries',\n",
+       "       'collection_recovery_fee', 'last_pymnt_d', 'last_pymnt_amnt',\n",
+       "       'next_pymnt_d', 'last_credit_pull_d', 'collections_12_mths_ex_med',\n",
+       "       'mths_since_last_major_derog', 'policy_code', 'application_type',\n",
+       "       'annual_inc_joint', 'dti_joint', 'verification_status_joint',\n",
+       "       'acc_now_delinq', 'tot_coll_amt', 'tot_cur_bal', 'open_acc_6m',\n",
+       "       'open_il_6m', 'open_il_12m', 'open_il_24m', 'mths_since_rcnt_il',\n",
+       "       'total_bal_il', 'il_util', 'open_rv_12m', 'open_rv_24m', 'max_bal_bc',\n",
+       "       'all_util', 'total_rev_hi_lim', 'inq_fi', 'total_cu_tl',\n",
+       "       'inq_last_12m'],\n",
+       "      dtype='object')"
+      ]
+     },
+     "execution_count": 158,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# explore the dataset(use the data dictionary)\n",
+    "# look at the columns, which data type is present in a column, range of values in each column, their mean, etc.\n",
+    "df.columns"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 159,
+   "id": "336cf009",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "<class 'pandas.core.frame.DataFrame'>\n",
+      "RangeIndex: 466285 entries, 0 to 466284\n",
+      "Data columns (total 74 columns):\n",
+      " #   Column                       Non-Null Count   Dtype  \n",
+      "---  ------                       --------------   -----  \n",
+      " 0   id                           466285 non-null  int64  \n",
+      " 1   member_id                    466285 non-null  int64  \n",
+      " 2   loan_amnt                    466285 non-null  int64  \n",
+      " 3   funded_amnt                  466285 non-null  int64  \n",
+      " 4   funded_amnt_inv              466285 non-null  float64\n",
+      " 5   term                         466285 non-null  object \n",
+      " 6   int_rate                     466285 non-null  float64\n",
+      " 7   installment                  466285 non-null  float64\n",
+      " 8   grade                        466285 non-null  object \n",
+      " 9   sub_grade                    466285 non-null  object \n",
+      " 10  emp_title                    438697 non-null  object \n",
+      " 11  emp_length                   445277 non-null  object \n",
+      " 12  home_ownership               466285 non-null  object \n",
+      " 13  annual_inc                   466281 non-null  float64\n",
+      " 14  verification_status          466285 non-null  object \n",
+      " 15  issue_d                      466285 non-null  object \n",
+      " 16  loan_status                  466285 non-null  object \n",
+      " 17  pymnt_plan                   466285 non-null  object \n",
+      " 18  url                          466285 non-null  object \n",
+      " 19  desc                         125983 non-null  object \n",
+      " 20  purpose                      466285 non-null  object \n",
+      " 21  title                        466265 non-null  object \n",
+      " 22  zip_code                     466285 non-null  object \n",
+      " 23  addr_state                   466285 non-null  object \n",
+      " 24  dti                          466285 non-null  float64\n",
+      " 25  delinq_2yrs                  466256 non-null  float64\n",
+      " 26  earliest_cr_line             466256 non-null  object \n",
+      " 27  inq_last_6mths               466256 non-null  float64\n",
+      " 28  mths_since_last_delinq       215934 non-null  float64\n",
+      " 29  mths_since_last_record       62638 non-null   float64\n",
+      " 30  open_acc                     466256 non-null  float64\n",
+      " 31  pub_rec                      466256 non-null  float64\n",
+      " 32  revol_bal                    466285 non-null  int64  \n",
+      " 33  revol_util                   465945 non-null  float64\n",
+      " 34  total_acc                    466256 non-null  float64\n",
+      " 35  initial_list_status          466285 non-null  object \n",
+      " 36  out_prncp                    466285 non-null  float64\n",
+      " 37  out_prncp_inv                466285 non-null  float64\n",
+      " 38  total_pymnt                  466285 non-null  float64\n",
+      " 39  total_pymnt_inv              466285 non-null  float64\n",
+      " 40  total_rec_prncp              466285 non-null  float64\n",
+      " 41  total_rec_int                466285 non-null  float64\n",
+      " 42  total_rec_late_fee           466285 non-null  float64\n",
+      " 43  recoveries                   466285 non-null  float64\n",
+      " 44  collection_recovery_fee      466285 non-null  float64\n",
+      " 45  last_pymnt_d                 465909 non-null  object \n",
+      " 46  last_pymnt_amnt              466285 non-null  float64\n",
+      " 47  next_pymnt_d                 239071 non-null  object \n",
+      " 48  last_credit_pull_d           466243 non-null  object \n",
+      " 49  collections_12_mths_ex_med   466140 non-null  float64\n",
+      " 50  mths_since_last_major_derog  98974 non-null   float64\n",
+      " 51  policy_code                  466285 non-null  int64  \n",
+      " 52  application_type             466285 non-null  object \n",
+      " 53  annual_inc_joint             0 non-null       float64\n",
+      " 54  dti_joint                    0 non-null       float64\n",
+      " 55  verification_status_joint    0 non-null       float64\n",
+      " 56  acc_now_delinq               466256 non-null  float64\n",
+      " 57  tot_coll_amt                 396009 non-null  float64\n",
+      " 58  tot_cur_bal                  396009 non-null  float64\n",
+      " 59  open_acc_6m                  0 non-null       float64\n",
+      " 60  open_il_6m                   0 non-null       float64\n",
+      " 61  open_il_12m                  0 non-null       float64\n",
+      " 62  open_il_24m                  0 non-null       float64\n",
+      " 63  mths_since_rcnt_il           0 non-null       float64\n",
+      " 64  total_bal_il                 0 non-null       float64\n",
+      " 65  il_util                      0 non-null       float64\n",
+      " 66  open_rv_12m                  0 non-null       float64\n",
+      " 67  open_rv_24m                  0 non-null       float64\n",
+      " 68  max_bal_bc                   0 non-null       float64\n",
+      " 69  all_util                     0 non-null       float64\n",
+      " 70  total_rev_hi_lim             396009 non-null  float64\n",
+      " 71  inq_fi                       0 non-null       float64\n",
+      " 72  total_cu_tl                  0 non-null       float64\n",
+      " 73  inq_last_12m                 0 non-null       float64\n",
+      "dtypes: float64(46), int64(6), object(22)\n",
+      "memory usage: 263.3+ MB\n"
+     ]
+    }
+   ],
+   "source": [
+    "df.info()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 160,
+   "id": "cc753927",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "id                       0\n",
+       "member_id                0\n",
+       "loan_amnt                0\n",
+       "funded_amnt              0\n",
+       "funded_amnt_inv          0\n",
+       "                     ...  \n",
+       "all_util            466285\n",
+       "total_rev_hi_lim     70276\n",
+       "inq_fi              466285\n",
+       "total_cu_tl         466285\n",
+       "inq_last_12m        466285\n",
+       "Length: 74, dtype: int64"
+      ]
+     },
+     "execution_count": 160,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# get a list of columns that have more than 80% null values\n",
+    "df.isna().sum()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 161,
+   "id": "e8794aeb",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Index(['emp_title', 'emp_length', 'annual_inc', 'desc', 'title', 'delinq_2yrs',\n",
+       "       'earliest_cr_line', 'inq_last_6mths', 'mths_since_last_delinq',\n",
+       "       'mths_since_last_record', 'open_acc', 'pub_rec', 'revol_util',\n",
+       "       'total_acc', 'last_pymnt_d', 'next_pymnt_d', 'last_credit_pull_d',\n",
+       "       'collections_12_mths_ex_med', 'mths_since_last_major_derog',\n",
+       "       'annual_inc_joint', 'dti_joint', 'verification_status_joint',\n",
+       "       'acc_now_delinq', 'tot_coll_amt', 'tot_cur_bal', 'open_acc_6m',\n",
+       "       'open_il_6m', 'open_il_12m', 'open_il_24m', 'mths_since_rcnt_il',\n",
+       "       'total_bal_il', 'il_util', 'open_rv_12m', 'open_rv_24m', 'max_bal_bc',\n",
+       "       'all_util', 'total_rev_hi_lim', 'inq_fi', 'total_cu_tl',\n",
+       "       'inq_last_12m'],\n",
+       "      dtype='object')"
+      ]
+     },
+     "execution_count": 161,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.columns[df.isnull().sum() > 0.8]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 162,
+   "id": "987bca7b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#Delete columns containing 80% or more than 80% NAN values\n",
+    "perc = 80"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 163,
+   "id": "161a7f1c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "min_count =  int(((100 - perc)/100)*df.shape[0] + 1)\n",
+    "df = df.dropna( axis=1, thresh=min_count)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 164,
+   "id": "2ff9f66a",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "<class 'pandas.core.frame.DataFrame'>\n",
+      "RangeIndex: 466285 entries, 0 to 466284\n",
+      "Data columns (total 56 columns):\n",
+      " #   Column                       Non-Null Count   Dtype  \n",
+      "---  ------                       --------------   -----  \n",
+      " 0   id                           466285 non-null  int64  \n",
+      " 1   member_id                    466285 non-null  int64  \n",
+      " 2   loan_amnt                    466285 non-null  int64  \n",
+      " 3   funded_amnt                  466285 non-null  int64  \n",
+      " 4   funded_amnt_inv              466285 non-null  float64\n",
+      " 5   term                         466285 non-null  object \n",
+      " 6   int_rate                     466285 non-null  float64\n",
+      " 7   installment                  466285 non-null  float64\n",
+      " 8   grade                        466285 non-null  object \n",
+      " 9   sub_grade                    466285 non-null  object \n",
+      " 10  emp_title                    438697 non-null  object \n",
+      " 11  emp_length                   445277 non-null  object \n",
+      " 12  home_ownership               466285 non-null  object \n",
+      " 13  annual_inc                   466281 non-null  float64\n",
+      " 14  verification_status          466285 non-null  object \n",
+      " 15  issue_d                      466285 non-null  object \n",
+      " 16  loan_status                  466285 non-null  object \n",
+      " 17  pymnt_plan                   466285 non-null  object \n",
+      " 18  url                          466285 non-null  object \n",
+      " 19  desc                         125983 non-null  object \n",
+      " 20  purpose                      466285 non-null  object \n",
+      " 21  title                        466265 non-null  object \n",
+      " 22  zip_code                     466285 non-null  object \n",
+      " 23  addr_state                   466285 non-null  object \n",
+      " 24  dti                          466285 non-null  float64\n",
+      " 25  delinq_2yrs                  466256 non-null  float64\n",
+      " 26  earliest_cr_line             466256 non-null  object \n",
+      " 27  inq_last_6mths               466256 non-null  float64\n",
+      " 28  mths_since_last_delinq       215934 non-null  float64\n",
+      " 29  open_acc                     466256 non-null  float64\n",
+      " 30  pub_rec                      466256 non-null  float64\n",
+      " 31  revol_bal                    466285 non-null  int64  \n",
+      " 32  revol_util                   465945 non-null  float64\n",
+      " 33  total_acc                    466256 non-null  float64\n",
+      " 34  initial_list_status          466285 non-null  object \n",
+      " 35  out_prncp                    466285 non-null  float64\n",
+      " 36  out_prncp_inv                466285 non-null  float64\n",
+      " 37  total_pymnt                  466285 non-null  float64\n",
+      " 38  total_pymnt_inv              466285 non-null  float64\n",
+      " 39  total_rec_prncp              466285 non-null  float64\n",
+      " 40  total_rec_int                466285 non-null  float64\n",
+      " 41  total_rec_late_fee           466285 non-null  float64\n",
+      " 42  recoveries                   466285 non-null  float64\n",
+      " 43  collection_recovery_fee      466285 non-null  float64\n",
+      " 44  last_pymnt_d                 465909 non-null  object \n",
+      " 45  last_pymnt_amnt              466285 non-null  float64\n",
+      " 46  next_pymnt_d                 239071 non-null  object \n",
+      " 47  last_credit_pull_d           466243 non-null  object \n",
+      " 48  collections_12_mths_ex_med   466140 non-null  float64\n",
+      " 49  mths_since_last_major_derog  98974 non-null   float64\n",
+      " 50  policy_code                  466285 non-null  int64  \n",
+      " 51  application_type             466285 non-null  object \n",
+      " 52  acc_now_delinq               466256 non-null  float64\n",
+      " 53  tot_coll_amt                 396009 non-null  float64\n",
+      " 54  tot_cur_bal                  396009 non-null  float64\n",
+      " 55  total_rev_hi_lim             396009 non-null  float64\n",
+      "dtypes: float64(28), int64(6), object(22)\n",
+      "memory usage: 199.2+ MB\n"
+     ]
+    }
+   ],
+   "source": [
+    "df.info()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 165,
+   "id": "21713972",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# remove duplicate columns\n",
+    "def DuplicateColumns(df):\n",
+    "    duplicatenames = set()\n",
+    "    for x in range(df.shape[1]):\n",
+    "        col = df.iloc[:, x]\n",
+    "        for y in range(x + 1, df.shape[1]):\n",
+    "            othercol = df.iloc[:, y]\n",
+    "            if col.equals(othercol):\n",
+    "                duplicatenames.add(df.colums.values[y])\n",
+    "    return list(duplicatenames)\n",
+    "df = df.drop(columns = DuplicateColumns(df))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 166,
+   "id": "a3bbe42b",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['id',\n",
+       " 'member_id',\n",
+       " 'loan_amnt',\n",
+       " 'funded_amnt',\n",
+       " 'funded_amnt_inv',\n",
+       " 'term',\n",
+       " 'int_rate',\n",
+       " 'installment',\n",
+       " 'grade',\n",
+       " 'sub_grade',\n",
+       " 'emp_title',\n",
+       " 'emp_length',\n",
+       " 'home_ownership',\n",
+       " 'annual_inc',\n",
+       " 'verification_status',\n",
+       " 'issue_d',\n",
+       " 'loan_status',\n",
+       " 'pymnt_plan',\n",
+       " 'url',\n",
+       " 'desc',\n",
+       " 'purpose',\n",
+       " 'title',\n",
+       " 'zip_code',\n",
+       " 'addr_state',\n",
+       " 'dti',\n",
+       " 'delinq_2yrs',\n",
+       " 'earliest_cr_line',\n",
+       " 'inq_last_6mths',\n",
+       " 'mths_since_last_delinq',\n",
+       " 'open_acc',\n",
+       " 'pub_rec',\n",
+       " 'revol_bal',\n",
+       " 'revol_util',\n",
+       " 'total_acc',\n",
+       " 'initial_list_status',\n",
+       " 'out_prncp',\n",
+       " 'out_prncp_inv',\n",
+       " 'total_pymnt',\n",
+       " 'total_pymnt_inv',\n",
+       " 'total_rec_prncp',\n",
+       " 'total_rec_int',\n",
+       " 'total_rec_late_fee',\n",
+       " 'recoveries',\n",
+       " 'collection_recovery_fee',\n",
+       " 'last_pymnt_d',\n",
+       " 'last_pymnt_amnt',\n",
+       " 'next_pymnt_d',\n",
+       " 'last_credit_pull_d',\n",
+       " 'collections_12_mths_ex_med',\n",
+       " 'mths_since_last_major_derog',\n",
+       " 'policy_code',\n",
+       " 'application_type',\n",
+       " 'acc_now_delinq',\n",
+       " 'tot_coll_amt',\n",
+       " 'tot_cur_bal',\n",
+       " 'total_rev_hi_lim']"
+      ]
+     },
+     "execution_count": 166,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "list(df)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 167,
+   "id": "a1534071",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# drop redundant(no longer useful) and forward-looking columns\n",
+    "df_drop = df.drop(labels = ['id','member_id','sub_grade','emp_title','url','desc','title','zip_code','total_rec_prncp','total_rec_late_fee','recoveries','collection_recovery_fee','next_pymnt_d'],axis = 1,inplace = True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 168,
+   "id": "f3a847b0",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>loan_amnt</th>\n",
+       "      <th>funded_amnt</th>\n",
+       "      <th>funded_amnt_inv</th>\n",
+       "      <th>term</th>\n",
+       "      <th>int_rate</th>\n",
+       "      <th>installment</th>\n",
+       "      <th>grade</th>\n",
+       "      <th>emp_length</th>\n",
+       "      <th>home_ownership</th>\n",
+       "      <th>annual_inc</th>\n",
+       "      <th>...</th>\n",
+       "      <th>last_pymnt_amnt</th>\n",
+       "      <th>last_credit_pull_d</th>\n",
+       "      <th>collections_12_mths_ex_med</th>\n",
+       "      <th>mths_since_last_major_derog</th>\n",
+       "      <th>policy_code</th>\n",
+       "      <th>application_type</th>\n",
+       "      <th>acc_now_delinq</th>\n",
+       "      <th>tot_coll_amt</th>\n",
+       "      <th>tot_cur_bal</th>\n",
+       "      <th>total_rev_hi_lim</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>5000</td>\n",
+       "      <td>5000</td>\n",
+       "      <td>4975.0</td>\n",
+       "      <td>36 months</td>\n",
+       "      <td>10.65</td>\n",
+       "      <td>162.87</td>\n",
+       "      <td>B</td>\n",
+       "      <td>10+ years</td>\n",
+       "      <td>RENT</td>\n",
+       "      <td>24000.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>171.62</td>\n",
+       "      <td>Jan-16</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>1</td>\n",
+       "      <td>INDIVIDUAL</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>2500</td>\n",
+       "      <td>2500</td>\n",
+       "      <td>2500.0</td>\n",
+       "      <td>60 months</td>\n",
+       "      <td>15.27</td>\n",
+       "      <td>59.83</td>\n",
+       "      <td>C</td>\n",
+       "      <td>&lt; 1 year</td>\n",
+       "      <td>RENT</td>\n",
+       "      <td>30000.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>119.66</td>\n",
+       "      <td>Sep-13</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>1</td>\n",
+       "      <td>INDIVIDUAL</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>2400</td>\n",
+       "      <td>2400</td>\n",
+       "      <td>2400.0</td>\n",
+       "      <td>36 months</td>\n",
+       "      <td>15.96</td>\n",
+       "      <td>84.33</td>\n",
+       "      <td>C</td>\n",
+       "      <td>10+ years</td>\n",
+       "      <td>RENT</td>\n",
+       "      <td>12252.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>649.91</td>\n",
+       "      <td>Jan-16</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>1</td>\n",
+       "      <td>INDIVIDUAL</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>10000</td>\n",
+       "      <td>10000</td>\n",
+       "      <td>10000.0</td>\n",
+       "      <td>36 months</td>\n",
+       "      <td>13.49</td>\n",
+       "      <td>339.31</td>\n",
+       "      <td>C</td>\n",
+       "      <td>10+ years</td>\n",
+       "      <td>RENT</td>\n",
+       "      <td>49200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>357.48</td>\n",
+       "      <td>Jan-15</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>1</td>\n",
+       "      <td>INDIVIDUAL</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>3000</td>\n",
+       "      <td>3000</td>\n",
+       "      <td>3000.0</td>\n",
+       "      <td>60 months</td>\n",
+       "      <td>12.69</td>\n",
+       "      <td>67.79</td>\n",
+       "      <td>B</td>\n",
+       "      <td>1 year</td>\n",
+       "      <td>RENT</td>\n",
+       "      <td>80000.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>67.79</td>\n",
+       "      <td>Jan-16</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>1</td>\n",
+       "      <td>INDIVIDUAL</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 43 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   loan_amnt  funded_amnt  funded_amnt_inv        term  int_rate  installment  \\\n",
+       "0       5000         5000           4975.0   36 months     10.65       162.87   \n",
+       "1       2500         2500           2500.0   60 months     15.27        59.83   \n",
+       "2       2400         2400           2400.0   36 months     15.96        84.33   \n",
+       "3      10000        10000          10000.0   36 months     13.49       339.31   \n",
+       "4       3000         3000           3000.0   60 months     12.69        67.79   \n",
+       "\n",
+       "  grade emp_length home_ownership  annual_inc  ... last_pymnt_amnt  \\\n",
+       "0     B  10+ years           RENT     24000.0  ...          171.62   \n",
+       "1     C   < 1 year           RENT     30000.0  ...          119.66   \n",
+       "2     C  10+ years           RENT     12252.0  ...          649.91   \n",
+       "3     C  10+ years           RENT     49200.0  ...          357.48   \n",
+       "4     B     1 year           RENT     80000.0  ...           67.79   \n",
+       "\n",
+       "  last_credit_pull_d collections_12_mths_ex_med mths_since_last_major_derog  \\\n",
+       "0             Jan-16                        0.0                         NaN   \n",
+       "1             Sep-13                        0.0                         NaN   \n",
+       "2             Jan-16                        0.0                         NaN   \n",
+       "3             Jan-15                        0.0                         NaN   \n",
+       "4             Jan-16                        0.0                         NaN   \n",
+       "\n",
+       "  policy_code application_type  acc_now_delinq  tot_coll_amt tot_cur_bal  \\\n",
+       "0           1       INDIVIDUAL             0.0           NaN         NaN   \n",
+       "1           1       INDIVIDUAL             0.0           NaN         NaN   \n",
+       "2           1       INDIVIDUAL             0.0           NaN         NaN   \n",
+       "3           1       INDIVIDUAL             0.0           NaN         NaN   \n",
+       "4           1       INDIVIDUAL             0.0           NaN         NaN   \n",
+       "\n",
+       "   total_rev_hi_lim  \n",
+       "0               NaN  \n",
+       "1               NaN  \n",
+       "2               NaN  \n",
+       "3               NaN  \n",
+       "4               NaN  \n",
+       "\n",
+       "[5 rows x 43 columns]"
+      ]
+     },
+     "execution_count": 168,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# re-explore the dataset\n",
+    "df.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 169,
+   "id": "78116f09",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "<class 'pandas.core.frame.DataFrame'>\n",
+      "RangeIndex: 466285 entries, 0 to 466284\n",
+      "Data columns (total 43 columns):\n",
+      " #   Column                       Non-Null Count   Dtype  \n",
+      "---  ------                       --------------   -----  \n",
+      " 0   loan_amnt                    466285 non-null  int64  \n",
+      " 1   funded_amnt                  466285 non-null  int64  \n",
+      " 2   funded_amnt_inv              466285 non-null  float64\n",
+      " 3   term                         466285 non-null  object \n",
+      " 4   int_rate                     466285 non-null  float64\n",
+      " 5   installment                  466285 non-null  float64\n",
+      " 6   grade                        466285 non-null  object \n",
+      " 7   emp_length                   445277 non-null  object \n",
+      " 8   home_ownership               466285 non-null  object \n",
+      " 9   annual_inc                   466281 non-null  float64\n",
+      " 10  verification_status          466285 non-null  object \n",
+      " 11  issue_d                      466285 non-null  object \n",
+      " 12  loan_status                  466285 non-null  object \n",
+      " 13  pymnt_plan                   466285 non-null  object \n",
+      " 14  purpose                      466285 non-null  object \n",
+      " 15  addr_state                   466285 non-null  object \n",
+      " 16  dti                          466285 non-null  float64\n",
+      " 17  delinq_2yrs                  466256 non-null  float64\n",
+      " 18  earliest_cr_line             466256 non-null  object \n",
+      " 19  inq_last_6mths               466256 non-null  float64\n",
+      " 20  mths_since_last_delinq       215934 non-null  float64\n",
+      " 21  open_acc                     466256 non-null  float64\n",
+      " 22  pub_rec                      466256 non-null  float64\n",
+      " 23  revol_bal                    466285 non-null  int64  \n",
+      " 24  revol_util                   465945 non-null  float64\n",
+      " 25  total_acc                    466256 non-null  float64\n",
+      " 26  initial_list_status          466285 non-null  object \n",
+      " 27  out_prncp                    466285 non-null  float64\n",
+      " 28  out_prncp_inv                466285 non-null  float64\n",
+      " 29  total_pymnt                  466285 non-null  float64\n",
+      " 30  total_pymnt_inv              466285 non-null  float64\n",
+      " 31  total_rec_int                466285 non-null  float64\n",
+      " 32  last_pymnt_d                 465909 non-null  object \n",
+      " 33  last_pymnt_amnt              466285 non-null  float64\n",
+      " 34  last_credit_pull_d           466243 non-null  object \n",
+      " 35  collections_12_mths_ex_med   466140 non-null  float64\n",
+      " 36  mths_since_last_major_derog  98974 non-null   float64\n",
+      " 37  policy_code                  466285 non-null  int64  \n",
+      " 38  application_type             466285 non-null  object \n",
+      " 39  acc_now_delinq               466256 non-null  float64\n",
+      " 40  tot_coll_amt                 396009 non-null  float64\n",
+      " 41  tot_cur_bal                  396009 non-null  float64\n",
+      " 42  total_rev_hi_lim             396009 non-null  float64\n",
+      "dtypes: float64(24), int64(4), object(15)\n",
+      "memory usage: 153.0+ MB\n"
+     ]
+    }
+   ],
+   "source": [
+    "df.info()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "e7300df0",
+   "metadata": {},
+   "source": [
+    "## Identify the target variable"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 170,
+   "id": "6ac92171",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# identify the target variable(target column) no need to code this just identify\n",
+    "#loan_status"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 171,
+   "id": "5f0d11eb",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array(['Fully Paid', 'Charged Off', 'Current', 'Default',\n",
+       "       'Late (31-120 days)', 'In Grace Period', 'Late (16-30 days)',\n",
+       "       'Does not meet the credit policy. Status:Fully Paid',\n",
+       "       'Does not meet the credit policy. Status:Charged Off'],\n",
+       "      dtype=object)"
+      ]
+     },
+     "execution_count": 171,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# explore the unique values in the target column\n",
+    "df.loan_status.unique()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 172,
+   "id": "37f20874",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create a new column based on the target column that will be our target variable\n",
+    "df['good_bad'] = np.where(df.loc[:, 'loan_status'].isin(['Charged Off', 'Default', 'Late (31-120 days)','Does not meet the credit policy. Status:Charged Off']), 0, 1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 173,
+   "id": "d7a50128",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Drop the original target column\n",
+    "df.drop(columns = ['loan_status'],inplace = True)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d62da664",
+   "metadata": {},
+   "source": [
+    "## Split Data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 174,
+   "id": "0af569dd",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# split data into 80/20 while keeping the distribution of bad loans in test set same as that in the pre-split dataset(X_train, y_train, etc)\n",
+    "X = df.drop('good_bad', axis = 1)\n",
+    "y = df['good_bad']\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 175,
+   "id": "a6a0e7da",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0         1\n",
+       "1         0\n",
+       "2         1\n",
+       "3         1\n",
+       "4         1\n",
+       "         ..\n",
+       "466280    1\n",
+       "466281    0\n",
+       "466282    1\n",
+       "466283    1\n",
+       "466284    1\n",
+       "Name: good_bad, Length: 466285, dtype: int32"
+      ]
+     },
+     "execution_count": 175,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "Y"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 176,
+   "id": "7dc94e2f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "X_train, X_test, y_train , y_test = train_test_split(X,y,test_size = 0.2,random_state = 42,stratify = y)\n",
+    "X_train, X_test = X_train.copy(), X_test.copy()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 177,
+   "id": "d4ba9e6b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# specifically hard copying the training sets to avoid Pandas' SetttingWithCopyWarning when we play around with this data later on\n",
+    "# you can refer to this link https://github.com/scikit-learn/scikit-learn/issues/8723\n",
+    "# this is currently an open issue between Pandas and Scikit-Learn teams"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 178,
+   "id": "45df823e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create a helper function clean up a column which has values given along with years, assign 0 to NANs and convert to numeric\n",
+    "def emp_length_converter(df, column):\n",
+    "    df[column] = df[column].str.replace('\\+ years', '')\n",
+    "    df[column] = df[column].str.replace('< 1 year', str(0))\n",
+    "    df[column] = df[column].str.replace(' years', '')\n",
+    "    df[column] = df[column].str.replace(' year', '')\n",
+    "    df[column] = pd.to_numeric(df[column])\n",
+    "    df[column].fillna(value = 0, inplace = True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 179,
+   "id": "4cf50e36",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "C:\\Users\\anshu\\AppData\\Local\\Temp\\ipykernel_2160\\2621041749.py:3: FutureWarning: The default value of regex will change from True to False in a future version.\n",
+      "  df[column] = df[column].str.replace('\\+ years', '')\n"
+     ]
+    }
+   ],
+   "source": [
+    "# apply to X_train\n",
+    "emp_length_converter(X_train, 'emp_length')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 180,
+   "id": "6f21a969",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([ 7., 10.,  3.,  4.,  2.,  0.,  1.,  6.,  5.,  8.,  9.])"
+      ]
+     },
+     "execution_count": 180,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# confirm our transformation by looking at unique values of this column\n",
+    "X_train['emp_length'].unique()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 181,
+   "id": "9818c5e9",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array(['Jan-15', 'Apr-13', 'Jun-14', 'Jan-16', 'Apr-12', 'Nov-12',\n",
+       "       'Jun-13', 'Sep-13', 'Jul-12', 'Oct-13', 'May-13', 'Feb-15',\n",
+       "       'Aug-15', 'Oct-12', 'Sep-12', nan, 'Dec-12', 'Dec-14', 'Aug-13',\n",
+       "       'Nov-13', 'Jan-14', 'Apr-14', 'Aug-14', 'Oct-14', 'Aug-12',\n",
+       "       'Jul-14', 'Jul-13', 'Apr-15', 'Feb-14', 'Sep-14', 'Jun-12',\n",
+       "       'Feb-13', 'Mar-13', 'May-14', 'Mar-15', 'Jan-13', 'Dec-13',\n",
+       "       'Feb-12', 'Mar-14', 'Sep-15', 'Nov-15', 'Dec-15', 'Jan-12',\n",
+       "       'Oct-15', 'Nov-14', 'Mar-12', 'May-12', 'Jun-15', 'May-15',\n",
+       "       'Jul-15', 'Dec-11', 'Nov-11', 'Oct-11', 'Sep-11', 'Aug-11',\n",
+       "       'Jul-11', 'Jun-11', 'May-11', 'Apr-11', 'Mar-11', 'Feb-11',\n",
+       "       'Jan-11', 'Dec-10', 'Nov-10', 'Oct-10', 'Sep-10', 'Aug-10',\n",
+       "       'Jul-10', 'Jun-10', 'May-10', 'Apr-10', 'Mar-10', 'Feb-10',\n",
+       "       'Jan-10', 'Dec-09', 'Nov-09', 'Oct-09', 'Sep-09', 'Aug-09',\n",
+       "       'Jul-09', 'Jun-09', 'May-09', 'Apr-09', 'Mar-09', 'Feb-09',\n",
+       "       'Jan-09', 'Dec-08', 'Oct-08', 'Aug-08', 'Jul-08', 'Sep-08',\n",
+       "       'Jun-08', 'May-08', 'Nov-08', 'Apr-08', 'Mar-08', 'Feb-08',\n",
+       "       'Jan-08', 'Dec-07'], dtype=object)"
+      ]
+     },
+     "execution_count": 181,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.last_pymnt_d.unique()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d86f2544",
+   "metadata": {},
+   "source": [
+    "### Date columns"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 182,
+   "id": "ba10320e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create a function to convert date columns to datetime format and create a new column as a difference between today and the respective date\n",
+    "# details of the function\n",
+    "    # store current month\n",
+    "    # convert to datetime format\n",
+    "    # calculate the difference in months and add to a new column\n",
+    "    # make any resulting -ve values to be equal to the max date\n",
+    "    # drop the original date column\n",
+    "# Note : In current date use the same data to maintain uniformilty across everyone's code. Use : 2020-08-01\n",
+    "\n",
+    "def convert_date_columns(df,column):\n",
+    "    month=df[column].str.split(pat='-',expand = True)[0]\n",
+    "    year='20'+df[column].str.split(pat='-',expand = True)[1]\n",
+    "    date=year+'-'+month+'-'+'01'\n",
+    "    current_date = pd.to_datetime('2020-08-01')\n",
+    "    df['mths_since_'+column]=round((current_date-pd.to_datetime(date))/np.timedelta64(1,'M'))\n",
+    "    df.drop(columns=[column],inplace = True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 183,
+   "id": "99a24bb3",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# apply to X_train\n",
+    "convert_date_columns(X_train,'last_pymnt_d')\n",
+    "convert_date_columns(X_train,'last_credit_pull_d')\n",
+    "convert_date_columns(X_train,'issue_d')\n",
+    "convert_date_columns(X_train,'earliest_cr_line')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 184,
+   "id": "4acfad7d",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "count    373003.000000\n",
+      "mean       -398.087517\n",
+      "std         536.013355\n",
+      "min        -952.000000\n",
+      "25%        -900.000000\n",
+      "50%        -771.000000\n",
+      "75%         203.000000\n",
+      "max         247.000000\n",
+      "Name: mths_since_earliest_cr_line, dtype: float64\n",
+      "count    373028.000000\n",
+      "mean         83.252485\n",
+      "std          14.339074\n",
+      "min          68.000000\n",
+      "25%          73.000000\n",
+      "50%          79.000000\n",
+      "75%          89.000000\n",
+      "max         158.000000\n",
+      "Name: mths_since_issue_d, dtype: float64\n",
+      "count    372717.000000\n",
+      "mean         63.289989\n",
+      "std          12.803859\n",
+      "min          55.000000\n",
+      "25%          55.000000\n",
+      "50%          56.000000\n",
+      "75%          67.000000\n",
+      "max         152.000000\n",
+      "Name: mths_since_last_pymnt_d, dtype: float64\n",
+      "count    372998.000000\n",
+      "mean         59.041810\n",
+      "std           9.630887\n",
+      "min          55.000000\n",
+      "25%          55.000000\n",
+      "50%          55.000000\n",
+      "75%          57.000000\n",
+      "max         159.000000\n",
+      "Name: mths_since_last_credit_pull_d, dtype: float64\n"
+     ]
+    }
+   ],
+   "source": [
+    "# check these new columns\n",
+    "print(X_train['mths_since_earliest_cr_line'].describe())\n",
+    "print(X_train['mths_since_issue_d'].describe())\n",
+    "print(X_train['mths_since_last_pymnt_d'].describe())\n",
+    "print(X_train['mths_since_last_credit_pull_d'].describe())"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "baf48e3e",
+   "metadata": {},
+   "source": [
+    "### Term column"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 185,
+   "id": "7832bb43",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# write a function to remove 'months' string from the 'term' column and convert it to numeric \n",
+    "# use the function on 'term' column of X_Train\n",
+    "def loan_term_converter(df, column):\n",
+    "    df[column] = pd.to_numeric(df[column].str.replace(' months', ''))\n",
+    "\n",
+    "loan_term_converter(X_train, 'term')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "168c666d",
+   "metadata": {},
+   "source": [
+    "## Feature Selection"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 186,
+   "id": "726b6bf8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# first divide training data into categorical and numerical subsets\n",
+    "X_train_cat = X_train.select_dtypes(include = 'object').copy()\n",
+    "X_train_num = X_train.select_dtypes(include = 'number').copy()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "08784212",
+   "metadata": {},
+   "source": [
+    "## Chi-squared statistic for categorical features"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 187,
+   "id": "1224a634",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# define an empty dictionary to store chi-squared test results\n",
+    "chi_sq = {}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 193,
+   "id": "12f30784",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "['grade', 'home_ownership', 'verification_status', 'pymnt_plan', 'purpose', 'addr_state', 'initial_list_status', 'application_type']\n"
+     ]
+    }
+   ],
+   "source": [
+    "Xt = X_train_cat.columns.tolist()\n",
+    "print(Xt)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "2f029372",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# loop over each column in the training set to calculate chi-statistic with the target variable\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 191,
+   "id": "79a44211",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# convert the dictionary to a DF"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "d11c692d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# keep only the top four categorical features"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "8949f3fa",
+   "metadata": {},
+   "source": [
+    "## ANOVA F-Statistic for numerical feature"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "f7c8d70f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# since f_class_if does not accept missing values, we will do a very crude imputation of missing values"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "815594f8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Calculate F Statistic and corresponding p values"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "66d26e6f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# convert to a DF"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "a0e14369",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# keep only the top 20 features and calculate pair-wise correlations between them\n",
+    "# save the top 20 numerical features in a list"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5a831c81",
+   "metadata": {},
+   "source": [
+    "## Pair wise correlations to detect multicollinearity"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "fb639daf",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# calculate pair-wise correlations between them"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "59b405e0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# drop 2 features based on their multicollinearity with other features"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "315f9251",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Define a helper function to drop the 4 categorical features with least p-values for chi squared test, 14 numerical features with least F-Statistic\n",
+    "# and 2 numerical features with high multicollinearity"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 34,
+   "id": "a483be96",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# apply to X_train"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "7b40d71b",
+   "metadata": {},
+   "source": [
+    "## creating dummy variables\n",
+    "### convert discrete variables to dummy variables"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 35,
+   "id": "cbb5d957",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# write a function to create dummy variables"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 36,
+   "id": "65ad5814",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# apply to our final four categorical variables"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "87b2556e",
+   "metadata": {},
+   "source": [
+    "## Update the test data set with all data cleaning procedures performed so far"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "id": "f2ce52f0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# also reindex the dummied test set variables to make sure all the feature columns in the train set are also available in the test set"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d96c459f",
+   "metadata": {},
+   "source": [
+    "## WoE Binning/Feature Engineering"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 38,
+   "id": "d91849df",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# we will analyze both categorical and numerical features based on their categorical/binned WoEs and IVs and then combine some of these binned categories together through a custom python class with fit_transform method"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 39,
+   "id": "9098de60",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create copies of the 4 training sets(by the names X_train_prepr, y_train_prepr, etc) to be preprocessed using WoE"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "556ef2ee",
+   "metadata": {},
+   "source": [
+    "## analyze WoEs and IVs of discrete features"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 40,
+   "id": "dd203042",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# write a function that takes 3 arguments: a dataframe (X_train_prepr), a string (column name), and a dataframe (y_train_prepr)\n",
+    "# the function should returns a dataframe as a result"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 41,
+   "id": "136285cd",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# set the default style of the graphs to the seaborn style. "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 42,
+   "id": "8cc8ddac",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# define a function for plotting WoE across categories that takes 2 arguments: a dataframe and a number"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 43,
+   "id": "ff26c024",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# apply these on all four categorical columns"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 44,
+   "id": "09c44927",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# observe graphs of WOE. If there is a continuous increase in WoE across the different categories then we do not need to combine any features together and should leave all these categories as they are"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 45,
+   "id": "c8b8839b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# if there are no missing values in the grade column leave it as it is, otherwise createa a separate and independent category for all Missing values that would never be combined with any other category\n",
+    "# you will come across this scenario when working through features"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 46,
+   "id": "c524dd5d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# define a function to calculate WoE of continuous variables\n",
+    "# this is same as the function we defined earlier for discrete variables\n",
+    "# the only difference are the 2 lines of code that need to be commented in the function that results in the df being sorted by continuous variable values"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 47,
+   "id": "c176da96",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# apply this on continuous variables"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 48,
+   "id": "82436e3d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# fine classing using the cut method if there are a large number of unique values"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 49,
+   "id": "6e66c07f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# don't use the features which have\n",
+    "    # very low IV\n",
+    "    # which have unusually high IV\n",
+    "    # WoE ranges between a very small range, implying low power of differentiating between good and bad loans"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 50,
+   "id": "d9c14866",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# if there is a feature for which most of the values are inside a particular range and very few outside then \n",
+    "    # create one category for values outside that range\n",
+    "    # apply your approach to all other values(which are inside the range)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 51,
+   "id": "7052db33",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# for some of the columns values would feel out of place like utilization being greater than 1 in some values which is very rare so filter those out"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 52,
+   "id": "b055d6fe",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# while plotting WOE if you have some doubt in curve you can also zoom on some portion to understand the nature of graph in that area"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 53,
+   "id": "9aac8722",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# if the IV is borderline close to the minimum or maximum ideal threshold, you can proceed without ignoring that feature"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "9865d6b8",
+   "metadata": {},
+   "source": [
+    "## Define Custom Class for WoE Binning/Reengineering"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 54,
+   "id": "3ecec1c7",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# here we will create a custom scikit-learn class to take care of all binning transformations on any given data set\n",
+    "# this custom class will help us in performing k fold cross validation"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 55,
+   "id": "bfeb5b21",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create a list of all the reference categories, i.e. one category from each of the global features"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 56,
+   "id": "b9d62257",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# this custom class will create new categorical dummy features based on the cut-off points that we manually identified based on the WoE plots and IV above"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 57,
+   "id": "260e7f4b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# structure this class so that it also allows a fit_transform method to be implemented on it, thereby allowing you to use it as part of a scikit-learn Pipeline "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "8594f479",
+   "metadata": {},
+   "source": [
+    "## PD Model"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 58,
+   "id": "1c007251",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# reconfirm shape of the 4 datasets"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 59,
+   "id": "400573de",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# define modeling pipeline"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 60,
+   "id": "46bbec7d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# define cross-validation criteria\n",
+    "# RepeatedStratifiedKFold automatially takes care of the class imbalance while splitting"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 61,
+   "id": "e76c496a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# fit and evaluate the logistic regression pipeline with cross-validation as defined in cv"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 62,
+   "id": "81862473",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# print the mean AUROC score and Gini"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 63,
+   "id": "ab3c480b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# fit the pipeline on the whole training set"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 64,
+   "id": "9426b9ed",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create a transformed training set through our WoE_Binning custom class"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 65,
+   "id": "a0a4d556",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# store the column names in X_train as a list"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 66,
+   "id": "96f2bfd1",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create a summary table of our logistic regression model(name it summary_table)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 67,
+   "id": "0b02d02b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create a new column in the dataframe, called 'Coefficients', with row values the transposed coefficients from the 'LogisticRegression' model"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 68,
+   "id": "bbefe8f3",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# increase the index of every row of the dataframe with 1 to store our model intercept in 1st row"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 69,
+   "id": "35d11b18",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# assign our model intercept to this new row"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 70,
+   "id": "1bdc664a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# sort the dataframe by index"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "2e39044f",
+   "metadata": {},
+   "source": [
+    "## Prediction Time!"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 71,
+   "id": "0fb857da",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# make preditions on our test set"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 72,
+   "id": "d751fae8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# get the predicted probabilities(name it y_hat_test_proba)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 73,
+   "id": "457c48b4",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# select the probabilities of only the positive class (class 1 - default) "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 74,
+   "id": "1221e651",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# we will now create a new DF with actual classes and the predicted probabilities\n",
+    "# create a temp y_test DF to reset its index to allow proper concaternation with y_hat_test_proba"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 75,
+   "id": "120e4119",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# check the shape to make sure the number of rows is same as that in y_test"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 76,
+   "id": "eff09b21",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# rename the columns"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 77,
+   "id": "4a340a2b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# makes the index of one dataframe equal to the index of another dataframe"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "67f3f5c7",
+   "metadata": {},
+   "source": [
+    "## Confusion Matrix and AUROC on Test Set"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 78,
+   "id": "bcedd377",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# assign a threshold value to differentiate good with bad"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 79,
+   "id": "42884466",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# crate a new column for the predicted class based on predicted probabilities and threshold\n",
+    "# we will determine this optimal threshold later in this project"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 80,
+   "id": "a1b467c7",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create the confusion matrix"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 81,
+   "id": "32ae63bd",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# get the values required to plot a ROC curve"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 82,
+   "id": "32a48f01",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# plot the ROC curve"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 83,
+   "id": "32926203",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# plot a secondary diagonal line, with dashed line style and black color to represent a no-skill classifier"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 84,
+   "id": "b029d7fe",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# calculate the Area Under the Receiver Operating Characteristic Curve (AUROC) on our test set"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 85,
+   "id": "e421c4b8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# calculate Gini from AUROC"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 86,
+   "id": "a67d6a4a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# draw a PR curve\n",
+    "# calculate the no skill line as the proportion of the positive class"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 87,
+   "id": "e4408762",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# plot the no skill precision-recall curve"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 88,
+   "id": "9419ed09",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# calculate inputs for the PR curve"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 89,
+   "id": "b5a5bae3",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# calculate inputs for the PR curve"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 90,
+   "id": "7d78d0fe",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# plot PR curve"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 91,
+   "id": "a78dd2ac",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# calculate PR AUC"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "9dd5c621",
+   "metadata": {},
+   "source": [
+    "## Applying the Model - Scorecard Creation"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 92,
+   "id": "37913e85",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# print the summary_table"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 93,
+   "id": "e56fea64",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create a new dataframe with one column\n",
+    "# its values are the values from the 'reference_categories' list\n",
+    "# name it 'Feature name'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 94,
+   "id": "a26ea5f4",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create a second column called 'Coefficients' which contains only 0 values"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 95,
+   "id": "8da85214",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# concatenates two dataframes"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 96,
+   "id": "1b6bceac",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# we reset the index of a dataframe"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 97,
+   "id": "8674e55b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create a new column called 'Original feature name' which contains the value of the 'Feature name' column up to the column symbol"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 98,
+   "id": "5c9e748e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# define the min and max threshholds for our scorecard"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 99,
+   "id": "86b900c9",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# calculate the sum of the minimum coefficients of each category within the original feature name"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 100,
+   "id": "a346dbdc",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# calculate the sum of the maximum coefficients of each category within the original feature name"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 101,
+   "id": "ac8691f9",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create a new columns that has the imputed calculated Score based on the multiplication of the coefficient by the ratio of the differences between\n",
+    "# maximum & minimum score and maximum & minimum sum of cefficients"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 102,
+   "id": "1bf24313",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# update the calculated score of the Intercept (i.e. the default score for each loan)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 103,
+   "id": "dc2657e3",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# round the values of the 'Score - Calculation' column and store them in a new column"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 104,
+   "id": "543f897d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# check the min and max possible scores of our scorecard"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 105,
+   "id": "88818024",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# both our min and max scores are out by +1\n",
+    "# we need to manually adjust this\n",
+    "# to decide which one we'll evaluate based on the rounding differences of the minimum category within each Original Feature Name"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 106,
+   "id": "e188b1a5",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# we can get by deducting 1 from the Intercept"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 107,
+   "id": "feb1f6f1",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# recheck min and max possible scores"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "cd043803",
+   "metadata": {},
+   "source": [
+    "## Calculating credit scores for all observations in the test data set"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 108,
+   "id": "7ff8f92a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create a transformed test set through our WoE_Binning custom class"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 109,
+   "id": "95d3ba61",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# insert an Intercept column in its beginning to align with the # of rows in scorecard"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 110,
+   "id": "e6229a6f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# get the list of our final scorecard scores"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 111,
+   "id": "bf4b59e4",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# check the shapes of test set and scorecard before doing matrix dot multiplication"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 112,
+   "id": "9ee9b571",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# we can see that the test set has a few less columns than the rows in scorecard due to the reference categories\n",
+    "# since the reference categories will always be scored as 0 based on the scorecard, it is safe to add these categories to the end of test set with 0 values"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 113,
+   "id": "9cb4b22e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# need to reshape scorecard_scores so that it is of proper shape to allow for matrix dot multiplication"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 114,
+   "id": "60a94f3a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# matrix dot multiplication of test set with scorecard scores"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "1d8c99b3",
+   "metadata": {},
+   "source": [
+    "## Setting loan approval cut-offs"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 115,
+   "id": "869dcdea",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# calculate Youden's J-Statistic to identify the best threshhold"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 116,
+   "id": "cc9bdb99",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# locate the index of the largest J"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 117,
+   "id": "0401c25d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# print the best threshold"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 118,
+   "id": "13cbdedb",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# this means that based on the Youden's J statistic, this is the ideal probability threshold which minimizes the FPR and maximimizes the TPR\n",
+    "# which means all samples with a predicted probability higher than this should be classified as in Default and vice versa\n",
+    "# this ideal threshold might appear to be counterintuitive compared to the default probability threshold of 0.5 but remember that we used the class_weight parameter when fitting our logistic regression model that would have helped us\n",
+    "\n",
+    "# we can confirm this by looking at our original confusion matrix with the updated threshold"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 119,
+   "id": "8577ee5f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# update the threshold value"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 120,
+   "id": "fae9f7ec",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# crate a new column for the predicted class based on predicted probabilities and threshold"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 121,
+   "id": "e9cf41a8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create the confusion matrix"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 122,
+   "id": "ddfcf9bc",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# the updated confusion matrix would show a marked improvement in the TPR at marginal cost of lower TNR but at the same time, FNR has improved drastically with a corresponding marginal increase in FPR\n",
+    "\n",
+    "# find the corresponding acceptance and rejection rates on the test set at this ideal threshold"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 123,
+   "id": "b0794619",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create a new DF comprising of the thresholds from the ROC output"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 124,
+   "id": "b8b07a3a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# calculate Score corresponding to each threshold"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 125,
+   "id": "a838b102",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# define a function called 'n_approved' which assigns a value of 1 if a predicted probability is greater than the parameter p which is a threshold and a value of 0 if it is not\n",
+    "# then it sums the column.\n",
+    "# for given any percentage values the function will return the number of rows wih estimated probabilites greater than the threshold"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 126,
+   "id": "00b90a17",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# assuming that all credit applications above a given probability of being 'good' will be approved\n",
+    "# when we apply the 'n_approved' function to a threshold it will return the number of approved applications\n",
+    "# here we calculate the number of approved appliations for all thresholds"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 127,
+   "id": "cb181639",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# then we calculate the number of rejected applications for each threshold\n",
+    "# it is the difference between the total number of applications and the approved applications for that threshold"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 128,
+   "id": "1630dc56",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# approval rate equals the ratio of the approved applications and all applications"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 129,
+   "id": "b5873b1c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# rejection rate equals one minus approval rate"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 130,
+   "id": "3d7a0362",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# look at the approval and rejection rates at our ideal threshold"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 131,
+   "id": "b63291ad",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# compare the above rates with the case of the default 0.5 threshold"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 132,
+   "id": "b26cc736",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# you would find that 0.5 threshold would result in a very high rejection rate with a corresponding loss of business\n",
+    "\n",
+    "# we will stick with our ideal threshold and the corresponding Creidt Score of (fill yourself) and can also monitor the model's performance in production if more data were available"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "bb92abd1",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}