project-basic-models-comparison.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Importing Libaries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "from sklearn.metrics import confusion_matrix,accuracy_score,classification_report,roc_auc_score,roc_curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "df=pd.read_csv(\"C:/Users/Nantha/Desktop/Nantha Kumar/Capstone/train.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>UniqueID</th>\n",
       "      <th>disbursed_amount</th>\n",
       "      <th>asset_cost</th>\n",
       "      <th>ltv</th>\n",
       "      <th>branch_id</th>\n",
       "      <th>supplier_id</th>\n",
       "      <th>manufacturer_id</th>\n",
       "      <th>Current_pincode_ID</th>\n",
       "      <th>Date.of.Birth</th>\n",
       "      <th>Employment.Type</th>\n",
       "      <th>...</th>\n",
       "      <th>SEC.SANCTIONED.AMOUNT</th>\n",
       "      <th>SEC.DISBURSED.AMOUNT</th>\n",
       "      <th>PRIMARY.INSTAL.AMT</th>\n",
       "      <th>SEC.INSTAL.AMT</th>\n",
       "      <th>NEW.ACCTS.IN.LAST.SIX.MONTHS</th>\n",
       "      <th>DELINQUENT.ACCTS.IN.LAST.SIX.MONTHS</th>\n",
       "      <th>AVERAGE.ACCT.AGE</th>\n",
       "      <th>CREDIT.HISTORY.LENGTH</th>\n",
       "      <th>NO.OF_INQUIRIES</th>\n",
       "      <th>loan_default</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>420825</td>\n",
       "      <td>50578</td>\n",
       "      <td>58400</td>\n",
       "      <td>89.55</td>\n",
       "      <td>67</td>\n",
       "      <td>22807</td>\n",
       "      <td>45</td>\n",
       "      <td>1441</td>\n",
       "      <td>01-01-84</td>\n",
       "      <td>Salaried</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0yrs 0mon</td>\n",
       "      <td>0yrs 0mon</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>537409</td>\n",
       "      <td>47145</td>\n",
       "      <td>65550</td>\n",
       "      <td>73.23</td>\n",
       "      <td>67</td>\n",
       "      <td>22807</td>\n",
       "      <td>45</td>\n",
       "      <td>1502</td>\n",
       "      <td>31-07-85</td>\n",
       "      <td>Self employed</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1991</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1yrs 11mon</td>\n",
       "      <td>1yrs 11mon</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>417566</td>\n",
       "      <td>53278</td>\n",
       "      <td>61360</td>\n",
       "      <td>89.63</td>\n",
       "      <td>67</td>\n",
       "      <td>22807</td>\n",
       "      <td>45</td>\n",
       "      <td>1497</td>\n",
       "      <td>24-08-85</td>\n",
       "      <td>Self employed</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0yrs 0mon</td>\n",
       "      <td>0yrs 0mon</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>624493</td>\n",
       "      <td>57513</td>\n",
       "      <td>66113</td>\n",
       "      <td>88.48</td>\n",
       "      <td>67</td>\n",
       "      <td>22807</td>\n",
       "      <td>45</td>\n",
       "      <td>1501</td>\n",
       "      <td>30-12-93</td>\n",
       "      <td>Self employed</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0yrs 8mon</td>\n",
       "      <td>1yrs 3mon</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>539055</td>\n",
       "      <td>52378</td>\n",
       "      <td>60300</td>\n",
       "      <td>88.39</td>\n",
       "      <td>67</td>\n",
       "      <td>22807</td>\n",
       "      <td>45</td>\n",
       "      <td>1495</td>\n",
       "      <td>09-12-77</td>\n",
       "      <td>Self employed</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0yrs 0mon</td>\n",
       "      <td>0yrs 0mon</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 41 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   UniqueID  disbursed_amount  asset_cost    ltv  branch_id  supplier_id  \\\n",
       "0    420825             50578       58400  89.55         67        22807   \n",
       "1    537409             47145       65550  73.23         67        22807   \n",
       "2    417566             53278       61360  89.63         67        22807   \n",
       "3    624493             57513       66113  88.48         67        22807   \n",
       "4    539055             52378       60300  88.39         67        22807   \n",
       "\n",
       "   manufacturer_id  Current_pincode_ID Date.of.Birth Employment.Type  ...  \\\n",
       "0               45                1441      01-01-84        Salaried  ...   \n",
       "1               45                1502      31-07-85   Self employed  ...   \n",
       "2               45                1497      24-08-85   Self employed  ...   \n",
       "3               45                1501      30-12-93   Self employed  ...   \n",
       "4               45                1495      09-12-77   Self employed  ...   \n",
       "\n",
       "  SEC.SANCTIONED.AMOUNT  SEC.DISBURSED.AMOUNT  PRIMARY.INSTAL.AMT  \\\n",
       "0                     0                     0                   0   \n",
       "1                     0                     0                1991   \n",
       "2                     0                     0                   0   \n",
       "3                     0                     0                  31   \n",
       "4                     0                     0                   0   \n",
       "\n",
       "   SEC.INSTAL.AMT  NEW.ACCTS.IN.LAST.SIX.MONTHS  \\\n",
       "0               0                             0   \n",
       "1               0                             0   \n",
       "2               0                             0   \n",
       "3               0                             0   \n",
       "4               0                             0   \n",
       "\n",
       "   DELINQUENT.ACCTS.IN.LAST.SIX.MONTHS  AVERAGE.ACCT.AGE  \\\n",
       "0                                    0         0yrs 0mon   \n",
       "1                                    1        1yrs 11mon   \n",
       "2                                    0         0yrs 0mon   \n",
       "3                                    0         0yrs 8mon   \n",
       "4                                    0         0yrs 0mon   \n",
       "\n",
       "   CREDIT.HISTORY.LENGTH  NO.OF_INQUIRIES  loan_default  \n",
       "0              0yrs 0mon                0             0  \n",
       "1             1yrs 11mon                0             1  \n",
       "2              0yrs 0mon                0             0  \n",
       "3              1yrs 3mon                1             1  \n",
       "4              0yrs 0mon                1             1  \n",
       "\n",
       "[5 rows x 41 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def cns_desc(x):\n",
    "    if x<300:\n",
    "        return 0\n",
    "    elif (x>=300) and (x<=350):\n",
    "        return 1\n",
    "    elif (x>350) and (x<=570):\n",
    "        return 2\n",
    "    elif (x>570) and (x<=630):\n",
    "        return 3\n",
    "    elif (x>630) and (x<=705):\n",
    "        return 4\n",
    "    else:\n",
    "        return 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"Date.of.Birth\"]=pd.to_datetime(df[\"Date.of.Birth\"])\n",
    "\n",
    "df[\"DisbursalDate\"]=pd.to_datetime(df[\"DisbursalDate\"])\n",
    "df[\"DisbursalDate_Month\"]=df[\"DisbursalDate\"].dt.month\n",
    "\n",
    "df[\"Age_at_disbursement\"]=((df['DisbursalDate']-df[\"Date.of.Birth\"])/365).apply(lambda x:x.days)\n",
    "\n",
    "df[\"Employment.Type\"]=df[\"Employment.Type\"].map({'Salaried':0,\"Self employed\":1,np.nan:2})\n",
    "\n",
    "df[\"PERFORM_CNS.SCORE.DESCRIPTION\"]=df[\"PERFORM_CNS.SCORE\"].apply(cns_desc)\n",
    "\n",
    "df[\"AVERAGE.ACCT.AGE\"]=df[\"AVERAGE.ACCT.AGE\"].apply(lambda x:(int((x.split(\"y\"))[0])*12)+(int((x.split())[1].split(\"m\")[0])))\n",
    "\n",
    "df[\"CREDIT.HISTORY.LENGTH\"]=df[\"CREDIT.HISTORY.LENGTH\"].apply(lambda x:(int((x.split(\"y\"))[0])*12)+(int((x.split())[1].split(\"m\")[0])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20e06697808>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(df['Age_at_disbursement'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['NO.OF.ACCTS']=df['PRI.NO.OF.ACCTS']+df['SEC.NO.OF.ACCTS']\n",
    "df['ACTIVE.ACCTS']=df['PRI.ACTIVE.ACCTS']+df['SEC.ACTIVE.ACCTS']\n",
    "df['OVERDUE.ACCTS']=df['PRI.OVERDUE.ACCTS']+df['SEC.OVERDUE.ACCTS']\n",
    "df['CURRENT.BALANCE']=df['PRI.CURRENT.BALANCE']+df['SEC.CURRENT.BALANCE']\n",
    "df['SANCTIONED.AMOUNT']=df['PRI.SANCTIONED.AMOUNT']+df['SEC.SANCTIONED.AMOUNT']\n",
    "df['DISBURSED.AMOUNT']=df['PRI.DISBURSED.AMOUNT']+df['SEC.DISBURSED.AMOUNT']\n",
    "df['INSTAL.AMT']=df['PRIMARY.INSTAL.AMT']+df['SEC.INSTAL.AMT']\n",
    "\n",
    "df=df.drop(['PRI.NO.OF.ACCTS','SEC.NO.OF.ACCTS','PRI.ACTIVE.ACCTS','SEC.ACTIVE.ACCTS','PRI.OVERDUE.ACCTS','SEC.OVERDUE.ACCTS','PRI.CURRENT.BALANCE','SEC.CURRENT.BALANCE','PRI.SANCTIONED.AMOUNT','SEC.SANCTIONED.AMOUNT','PRI.DISBURSED.AMOUNT','SEC.DISBURSED.AMOUNT','PRIMARY.INSTAL.AMT','SEC.INSTAL.AMT'],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "UniqueID                               233154\n",
       "disbursed_amount                        24565\n",
       "asset_cost                              46252\n",
       "ltv                                      6579\n",
       "branch_id                                  82\n",
       "supplier_id                              2953\n",
       "manufacturer_id                            11\n",
       "Current_pincode_ID                       6698\n",
       "Date.of.Birth                           15433\n",
       "Employment.Type                             3\n",
       "DisbursalDate                              84\n",
       "State_ID                                   22\n",
       "Employee_code_ID                         3270\n",
       "MobileNo_Avl_Flag                           1\n",
       "Aadhar_flag                                 2\n",
       "PAN_flag                                    2\n",
       "VoterID_flag                                2\n",
       "Driving_flag                                2\n",
       "Passport_flag                               2\n",
       "PERFORM_CNS.SCORE                         573\n",
       "PERFORM_CNS.SCORE.DESCRIPTION               6\n",
       "NEW.ACCTS.IN.LAST.SIX.MONTHS               26\n",
       "DELINQUENT.ACCTS.IN.LAST.SIX.MONTHS        14\n",
       "AVERAGE.ACCT.AGE                          192\n",
       "CREDIT.HISTORY.LENGTH                     294\n",
       "NO.OF_INQUIRIES                            25\n",
       "loan_default                                2\n",
       "DisbursalDate_Month                        12\n",
       "Age_at_disbursement                        50\n",
       "NO.OF.ACCTS                               108\n",
       "ACTIVE.ACCTS                               41\n",
       "OVERDUE.ACCTS                              22\n",
       "CURRENT.BALANCE                         72483\n",
       "SANCTIONED.AMOUNT                       45367\n",
       "DISBURSED.AMOUNT                        48958\n",
       "INSTAL.AMT                              28540\n",
       "dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['UniqueID', 'disbursed_amount', 'asset_cost', 'ltv', 'branch_id',\n",
       "       'supplier_id', 'manufacturer_id', 'Current_pincode_ID', 'Date.of.Birth',\n",
       "       'Employment.Type', 'DisbursalDate', 'State_ID', 'Employee_code_ID',\n",
       "       'MobileNo_Avl_Flag', 'Aadhar_flag', 'PAN_flag', 'VoterID_flag',\n",
       "       'Driving_flag', 'Passport_flag', 'PERFORM_CNS.SCORE',\n",
       "       'PERFORM_CNS.SCORE.DESCRIPTION', 'NEW.ACCTS.IN.LAST.SIX.MONTHS',\n",
       "       'DELINQUENT.ACCTS.IN.LAST.SIX.MONTHS', 'AVERAGE.ACCT.AGE',\n",
       "       'CREDIT.HISTORY.LENGTH', 'NO.OF_INQUIRIES', 'loan_default',\n",
       "       'DisbursalDate_Month', 'Age_at_disbursement', 'NO.OF.ACCTS',\n",
       "       'ACTIVE.ACCTS', 'OVERDUE.ACCTS', 'CURRENT.BALANCE', 'SANCTIONED.AMOUNT',\n",
       "       'DISBURSED.AMOUNT', 'INSTAL.AMT'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "df=df.drop(columns=['Date.of.Birth',\"DisbursalDate\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "cols=df.columns.to_list()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "cat_cols=['UniqueID','branch_id','supplier_id', 'manufacturer_id', 'Current_pincode_ID','Employment.Type', 'State_ID',\n",
    "          'Employee_code_ID','MobileNo_Avl_Flag', 'Aadhar_flag', 'PAN_flag', 'VoterID_flag','Driving_flag', 'Passport_flag',\n",
    "       'PERFORM_CNS.SCORE.DESCRIPTION','DisbursalDate_Month']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "target_col=\"loan_default\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_cols=['disbursed_amount','asset_cost','ltv','PERFORM_CNS.SCORE','NEW.ACCTS.IN.LAST.SIX.MONTHS',\n",
    "          'DELINQUENT.ACCTS.IN.LAST.SIX.MONTHS','AVERAGE.ACCT.AGE','CREDIT.HISTORY.LENGTH','NO.OF_INQUIRIES',\n",
    "          'Age_at_disbursement','NO.OF.ACCTS','ACTIVE.ACCTS','OVERDUE.ACCTS','CURRENT.BALANCE','SANCTIONED.AMOUNT',\n",
    "             'DISBURSED.AMOUNT','INSTAL.AMT']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "UniqueID                         233154\n",
       "branch_id                            82\n",
       "supplier_id                        2953\n",
       "manufacturer_id                      11\n",
       "Current_pincode_ID                 6698\n",
       "Employment.Type                       3\n",
       "State_ID                             22\n",
       "Employee_code_ID                   3270\n",
       "MobileNo_Avl_Flag                     1\n",
       "Aadhar_flag                           2\n",
       "PAN_flag                              2\n",
       "VoterID_flag                          2\n",
       "Driving_flag                          2\n",
       "Passport_flag                         2\n",
       "PERFORM_CNS.SCORE.DESCRIPTION         6\n",
       "DisbursalDate_Month                  12\n",
       "dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[cat_cols].nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we see no of unique elements in some categorical columns is higher,we couldn't take dummies as it will cause curse of \n",
    "dimensionality.Let's drop them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "linear_models_df=df.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "linear_models_df=linear_models_df.drop(columns=['UniqueID','supplier_id','Current_pincode_ID','Employee_code_ID',\"MobileNo_Avl_Flag\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "cols_to_be_dummied=[\"branch_id\",\"State_ID\",\"manufacturer_id\",'Employment.Type','DisbursalDate_Month','PERFORM_CNS.SCORE.DESCRIPTION']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "linear_models_final_df=pd.get_dummies(data=linear_models_df,columns=cols_to_be_dummied,drop_first=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "disbursed_amount                       0.077675\n",
       "asset_cost                             0.014261\n",
       "ltv                                    0.098208\n",
       "branch_id                              0.030193\n",
       "manufacturer_id                       -0.025039\n",
       "Employment.Type                        0.024823\n",
       "State_ID                               0.048075\n",
       "Aadhar_flag                           -0.041593\n",
       "PAN_flag                               0.002046\n",
       "VoterID_flag                           0.043747\n",
       "Driving_flag                          -0.005821\n",
       "Passport_flag                         -0.007602\n",
       "PERFORM_CNS.SCORE                     -0.057929\n",
       "PERFORM_CNS.SCORE.DESCRIPTION         -0.067798\n",
       "NEW.ACCTS.IN.LAST.SIX.MONTHS          -0.029400\n",
       "DELINQUENT.ACCTS.IN.LAST.SIX.MONTHS    0.034462\n",
       "AVERAGE.ACCT.AGE                      -0.024781\n",
       "CREDIT.HISTORY.LENGTH                 -0.042126\n",
       "NO.OF_INQUIRIES                        0.043678\n",
       "loan_default                           1.000000\n",
       "DisbursalDate_Month                    0.011411\n",
       "Age_at_disbursement                    0.008926\n",
       "NO.OF.ACCTS                           -0.035963\n",
       "ACTIVE.ACCTS                          -0.041511\n",
       "OVERDUE.ACCTS                          0.039469\n",
       "CURRENT.BALANCE                       -0.027839\n",
       "SANCTIONED.AMOUNT                     -0.011749\n",
       "DISBURSED.AMOUNT                      -0.011591\n",
       "INSTAL.AMT                            -0.010707\n",
       "Name: loan_default, dtype: float64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "linear_models_df.corr()['loan_default']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Since Current balance has negative values we will take scaled between 0 and 1**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "scalar=MinMaxScaler()\n",
    "linear_models_final_df[\"CURRENT.BALANCE\"]=scalar.fit_transform(linear_models_final_df[\"CURRENT.BALANCE\"].values.reshape(-1,1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>Scores</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>disbursed_amount</td>\n",
       "      <td>4.492240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>asset_cost</td>\n",
       "      <td>6.133485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>ltv</td>\n",
       "      <td>-1.075766</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>PERFORM_CNS.SCORE</td>\n",
       "      <td>0.445150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>NEW.ACCTS.IN.LAST.SIX.MONTHS</td>\n",
       "      <td>4.839326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>DELINQUENT.ACCTS.IN.LAST.SIX.MONTHS</td>\n",
       "      <td>6.641996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>AVERAGE.ACCT.AGE</td>\n",
       "      <td>3.285142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>CREDIT.HISTORY.LENGTH</td>\n",
       "      <td>2.969155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>NO.OF_INQUIRIES</td>\n",
       "      <td>7.870683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Age_at_disbursement</td>\n",
       "      <td>-2.250138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>NO.OF.ACCTS</td>\n",
       "      <td>9.474425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>ACTIVE.ACCTS</td>\n",
       "      <td>5.278660</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>OVERDUE.ACCTS</td>\n",
       "      <td>7.312614</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>CURRENT.BALANCE</td>\n",
       "      <td>28.589453</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>SANCTIONED.AMOUNT</td>\n",
       "      <td>320.010642</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>DISBURSED.AMOUNT</td>\n",
       "      <td>318.903561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>INSTAL.AMT</td>\n",
       "      <td>68.804991</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         Scores\n",
       "disbursed_amount                       4.492240\n",
       "asset_cost                             6.133485\n",
       "ltv                                   -1.075766\n",
       "PERFORM_CNS.SCORE                      0.445150\n",
       "NEW.ACCTS.IN.LAST.SIX.MONTHS           4.839326\n",
       "DELINQUENT.ACCTS.IN.LAST.SIX.MONTHS    6.641996\n",
       "AVERAGE.ACCT.AGE                       3.285142\n",
       "CREDIT.HISTORY.LENGTH                  2.969155\n",
       "NO.OF_INQUIRIES                        7.870683\n",
       "Age_at_disbursement                   -2.250138\n",
       "NO.OF.ACCTS                            9.474425\n",
       "ACTIVE.ACCTS                           5.278660\n",
       "OVERDUE.ACCTS                          7.312614\n",
       "CURRENT.BALANCE                       28.589453\n",
       "SANCTIONED.AMOUNT                    320.010642\n",
       "DISBURSED.AMOUNT                     318.903561\n",
       "INSTAL.AMT                            68.804991"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "val=[]\n",
    "for i in num_cols:\n",
    "    val.append(linear_models_df[i].skew())\n",
    "skew_df=pd.DataFrame(index=num_cols,data=val,columns=[\"Scores\"])\n",
    "skew_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "cols_to_taken_log=skew_df[skew_df[\"Scores\"]>2].index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in cols_to_taken_log:\n",
    "    linear_models_final_df[i]=np.log(linear_models_final_df[i]+1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "X=linear_models_final_df.drop('loan_default',axis=1)\n",
    "y=linear_models_final_df['loan_default']\n",
    "model_scores={}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>disbursed_amount</th>\n",
       "      <th>asset_cost</th>\n",
       "      <th>ltv</th>\n",
       "      <th>Aadhar_flag</th>\n",
       "      <th>PAN_flag</th>\n",
       "      <th>VoterID_flag</th>\n",
       "      <th>Driving_flag</th>\n",
       "      <th>Passport_flag</th>\n",
       "      <th>PERFORM_CNS.SCORE</th>\n",
       "      <th>NEW.ACCTS.IN.LAST.SIX.MONTHS</th>\n",
       "      <th>...</th>\n",
       "      <th>DisbursalDate_Month_8</th>\n",
       "      <th>DisbursalDate_Month_9</th>\n",
       "      <th>DisbursalDate_Month_10</th>\n",
       "      <th>DisbursalDate_Month_11</th>\n",
       "      <th>DisbursalDate_Month_12</th>\n",
       "      <th>PERFORM_CNS.SCORE.DESCRIPTION_1</th>\n",
       "      <th>PERFORM_CNS.SCORE.DESCRIPTION_2</th>\n",
       "      <th>PERFORM_CNS.SCORE.DESCRIPTION_3</th>\n",
       "      <th>PERFORM_CNS.SCORE.DESCRIPTION_4</th>\n",
       "      <th>PERFORM_CNS.SCORE.DESCRIPTION_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>10.831292</td>\n",
       "      <td>10.975088</td>\n",
       "      <td>89.55</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>10.761004</td>\n",
       "      <td>11.090584</td>\n",
       "      <td>73.23</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>598</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>10.883298</td>\n",
       "      <td>11.024530</td>\n",
       "      <td>89.63</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>10.959784</td>\n",
       "      <td>11.099136</td>\n",
       "      <td>88.48</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>305</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>10.866261</td>\n",
       "      <td>11.007104</td>\n",
       "      <td>88.39</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 153 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   disbursed_amount  asset_cost    ltv  Aadhar_flag  PAN_flag  VoterID_flag  \\\n",
       "0         10.831292   10.975088  89.55            1         0             0   \n",
       "1         10.761004   11.090584  73.23            1         0             0   \n",
       "2         10.883298   11.024530  89.63            1         0             0   \n",
       "3         10.959784   11.099136  88.48            1         0             0   \n",
       "4         10.866261   11.007104  88.39            1         0             0   \n",
       "\n",
       "   Driving_flag  Passport_flag  PERFORM_CNS.SCORE  \\\n",
       "0             0              0                  0   \n",
       "1             0              0                598   \n",
       "2             0              0                  0   \n",
       "3             0              0                305   \n",
       "4             0              0                  0   \n",
       "\n",
       "   NEW.ACCTS.IN.LAST.SIX.MONTHS  ...  DisbursalDate_Month_8  \\\n",
       "0                           0.0  ...                      0   \n",
       "1                           0.0  ...                      0   \n",
       "2                           0.0  ...                      0   \n",
       "3                           0.0  ...                      0   \n",
       "4                           0.0  ...                      0   \n",
       "\n",
       "   DisbursalDate_Month_9  DisbursalDate_Month_10  DisbursalDate_Month_11  \\\n",
       "0                      0                       0                       0   \n",
       "1                      1                       0                       0   \n",
       "2                      0                       0                       0   \n",
       "3                      0                       1                       0   \n",
       "4                      1                       0                       0   \n",
       "\n",
       "   DisbursalDate_Month_12  PERFORM_CNS.SCORE.DESCRIPTION_1  \\\n",
       "0                       0                                0   \n",
       "1                       0                                0   \n",
       "2                       0                                0   \n",
       "3                       0                                1   \n",
       "4                       0                                0   \n",
       "\n",
       "   PERFORM_CNS.SCORE.DESCRIPTION_2  PERFORM_CNS.SCORE.DESCRIPTION_3  \\\n",
       "0                                0                                0   \n",
       "1                                0                                1   \n",
       "2                                0                                0   \n",
       "3                                0                                0   \n",
       "4                                0                                0   \n",
       "\n",
       "   PERFORM_CNS.SCORE.DESCRIPTION_4  PERFORM_CNS.SCORE.DESCRIPTION_5  \n",
       "0                                0                                0  \n",
       "1                                0                                0  \n",
       "2                                0                                0  \n",
       "3                                0                                0  \n",
       "4                                0                                0  \n",
       "\n",
       "[5 rows x 153 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "linear_models_final_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model Building"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "#model selection from sklearn import model_selection\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "#ensemble\n",
    "from sklearn.ensemble import RandomForestClassifier,BaggingClassifier,AdaBoostClassifier,GradientBoostingClassifier\n",
    "from xgboost import XGBClassifier\n",
    "#metrics\n",
    "from sklearn.metrics import precision_score, recall_score, f1_score, roc_auc_score, accuracy_score, classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split,KFold,cross_val_score\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(163207,)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((69947, 152), (69947,))"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape,y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#dict to store various model results for comparison\n",
    "results = {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "#model result is a function for model building and calculating scores\n",
    "def modelresult(model_name,i):\n",
    "    print(i,\"\\n\")\n",
    "    classifier = i\n",
    "    classifier.fit(X_train, y_train)\n",
    "    y_predt=classifier.predict(X_train)\n",
    "    y_pred=classifier.predict(X_test)\n",
    "    y_pred_prob=classifier.predict_proba(X_test)[:,1]\n",
    "    results[model_name] = {'Accuracy' : accuracy_score(y_test,y_pred), 'roc_auc_score' : roc_auc_score(y_test,y_pred_prob)}\n",
    "    print(\"\\n Accuracy for the Train:\",accuracy_score(y_train,y_predt))\n",
    "    print(\"\\n Accuracy for the Test\",accuracy_score(y_test,y_pred),\"\\n\")\n",
    "    print(\"\\n ROC AUC score\",roc_auc_score(y_test,y_pred_prob),\"\\n\")\n",
    "    from sklearn.metrics import classification_report\n",
    "    print(classification_report(y_test,y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
      "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
      "                   multi_class='warn', n_jobs=None, penalty='l2',\n",
      "                   random_state=1, solver='warn', tol=0.0001, verbose=0,\n",
      "                   warm_start=False) \n",
      "\n",
      "\n",
      " Accuracy for the Train: 0.7828218152407679\n",
      "\n",
      " Accuracy for the Test 0.7829928374340572 \n",
      "\n",
      "\n",
      " ROC AUC score 0.6475496967596035 \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.78      1.00      0.88     54776\n",
      "           1       0.48      0.01      0.01     15171\n",
      "\n",
      "    accuracy                           0.78     69947\n",
      "   macro avg       0.63      0.50      0.44     69947\n",
      "weighted avg       0.72      0.78      0.69     69947\n",
      "\n"
     ]
    }
   ],
   "source": [
    "lr = LogisticRegression(random_state=1)\n",
    "modelresult(\"Linear Reg\",lr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "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>Linear Reg</th>\n",
       "      <th>DT-Entropy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>Accuracy</td>\n",
       "      <td>0.782993</td>\n",
       "      <td>0.676098</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>roc_auc_score</td>\n",
       "      <td>0.647550</td>\n",
       "      <td>0.529636</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Linear Reg  DT-Entropy\n",
       "Accuracy         0.782993    0.676098\n",
       "roc_auc_score    0.647550    0.529636"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(results)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Decision Tree classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DecisionTreeClassifier(class_weight=None, criterion='entropy', max_depth=None,\n",
      "                       max_features=None, max_leaf_nodes=None,\n",
      "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "                       min_samples_leaf=1, min_samples_split=2,\n",
      "                       min_weight_fraction_leaf=0.0, presort=False,\n",
      "                       random_state=1, splitter='best') \n",
      "\n",
      "\n",
      " Accuracy for the Train: 0.9997365309086007\n",
      "\n",
      " Accuracy for the Test 0.6760976167669808 \n",
      "\n",
      "\n",
      " ROC AUC score 0.529635960358134 \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.80      0.79      0.79     54776\n",
      "           1       0.26      0.27      0.27     15171\n",
      "\n",
      "    accuracy                           0.68     69947\n",
      "   macro avg       0.53      0.53      0.53     69947\n",
      "weighted avg       0.68      0.68      0.68     69947\n",
      "\n"
     ]
    }
   ],
   "source": [
    "dcg = DecisionTreeClassifier(criterion='entropy',random_state=1)\n",
    "modelresult(\"DT-Entropy\" ,dcg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Decision Tree classifier with gini"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=None,\n",
      "                       max_features=None, max_leaf_nodes=None,\n",
      "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "                       min_samples_leaf=1, min_samples_split=2,\n",
      "                       min_weight_fraction_leaf=0.0, presort=False,\n",
      "                       random_state=1, splitter='best') \n",
      "\n",
      "\n",
      " Accuracy for the Train: 0.9997365309086007\n",
      "\n",
      " Accuracy for the Test 0.6747823351966489 \n",
      "\n",
      "\n",
      " ROC AUC score 0.5323006885855466 \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.80      0.78      0.79     54776\n",
      "           1       0.26      0.28      0.27     15171\n",
      "\n",
      "    accuracy                           0.67     69947\n",
      "   macro avg       0.53      0.53      0.53     69947\n",
      "weighted avg       0.68      0.67      0.68     69947\n",
      "\n"
     ]
    }
   ],
   "source": [
    "dcg = DecisionTreeClassifier(criterion='gini',random_state=1)\n",
    "modelresult(\"DT-Gini\",dcg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## RandomForest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
      "                       max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
      "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "                       min_samples_leaf=1, min_samples_split=2,\n",
      "                       min_weight_fraction_leaf=0.0, n_estimators=100,\n",
      "                       n_jobs=None, oob_score=False, random_state=1, verbose=0,\n",
      "                       warm_start=False) \n",
      "\n",
      "\n",
      " Accuracy for the Train: 0.9997120221559125\n",
      "\n",
      " Accuracy for the Test 0.7771598496004117 \n",
      "\n",
      "\n",
      " ROC AUC score 0.6301906615443204 \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.79      0.98      0.87     54776\n",
      "           1       0.39      0.05      0.08     15171\n",
      "\n",
      "    accuracy                           0.78     69947\n",
      "   macro avg       0.59      0.51      0.48     69947\n",
      "weighted avg       0.70      0.78      0.70     69947\n",
      "\n"
     ]
    }
   ],
   "source": [
    "rfc = RandomForestClassifier(n_estimators=100,random_state=1)\n",
    "modelresult(\"RandomForest\",rfc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Bagging Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BaggingClassifier(base_estimator=None, bootstrap=True, bootstrap_features=False,\n",
      "                  max_features=1.0, max_samples=1.0, n_estimators=10,\n",
      "                  n_jobs=None, oob_score=False, random_state=1, verbose=0,\n",
      "                  warm_start=False) \n",
      "\n",
      "\n",
      " Accuracy for the Train: 0.9750868528923392\n",
      "\n",
      " Accuracy for the Test 0.7650506812300742 \n",
      "\n",
      "\n",
      " ROC AUC score 0.587350816605213 \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.79      0.95      0.86     54776\n",
      "           1       0.34      0.09      0.14     15171\n",
      "\n",
      "    accuracy                           0.77     69947\n",
      "   macro avg       0.56      0.52      0.50     69947\n",
      "weighted avg       0.69      0.77      0.71     69947\n",
      "\n"
     ]
    }
   ],
   "source": [
    "bgc = BaggingClassifier(random_state=1)\n",
    "modelresult(\"Bagging\",bgc)  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## AdaBoost Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AdaBoostClassifier(algorithm='SAMME.R', base_estimator=None, learning_rate=1.0,\n",
      "                   n_estimators=50, random_state=1) \n",
      "\n",
      "\n",
      " Accuracy for the Train: 0.7827115258536705\n",
      "\n",
      " Accuracy for the Test 0.7826783135802822 \n",
      "\n",
      "\n",
      " ROC AUC score 0.6535504293938926 \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.79      0.99      0.88     54776\n",
      "           1       0.47      0.02      0.03     15171\n",
      "\n",
      "    accuracy                           0.78     69947\n",
      "   macro avg       0.63      0.51      0.45     69947\n",
      "weighted avg       0.72      0.78      0.69     69947\n",
      "\n"
     ]
    }
   ],
   "source": [
    "ada = AdaBoostClassifier(random_state=1)\n",
    "modelresult(\"Adaboost\", ada)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## XGBoost Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
      "              colsample_bynode=1, colsample_bytree=1, gamma=0,\n",
      "              learning_rate=0.2, max_delta_step=0, max_depth=3,\n",
      "              min_child_weight=1, missing=None, n_estimators=100, n_jobs=-1,\n",
      "              nthread=None, objective='binary:logistic', random_state=1,\n",
      "              reg_alpha=0, reg_lambda=1, scale_pos_weight=1, seed=None,\n",
      "              silent=None, subsample=1, verbosity=1) \n",
      "\n",
      "\n",
      " Accuracy for the Train: 0.7838328012891604\n",
      "\n",
      " Accuracy for the Test 0.783693367835647 \n",
      "\n",
      "\n",
      " ROC AUC score 0.6646233684499697 \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.78      1.00      0.88     54776\n",
      "           1       0.57      0.01      0.02     15171\n",
      "\n",
      "    accuracy                           0.78     69947\n",
      "   macro avg       0.68      0.50      0.45     69947\n",
      "weighted avg       0.74      0.78      0.69     69947\n",
      "\n"
     ]
    }
   ],
   "source": [
    "xgb =  XGBClassifier(random_state=1, n_jobs=-1, learning_rate=0.2,n_estimators=100, max_depth=3)\n",
    "modelresult(\"XGboost\",xgb)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "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>Accuracy</th>\n",
       "      <th>roc_auc_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>Linear Reg</td>\n",
       "      <td>0.782993</td>\n",
       "      <td>0.647550</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>DT-Entropy</td>\n",
       "      <td>0.676098</td>\n",
       "      <td>0.529636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>DT-Gini</td>\n",
       "      <td>0.674782</td>\n",
       "      <td>0.532301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>RandomForest</td>\n",
       "      <td>0.777160</td>\n",
       "      <td>0.630191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Bagging</td>\n",
       "      <td>0.765051</td>\n",
       "      <td>0.587351</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Adaboost</td>\n",
       "      <td>0.782678</td>\n",
       "      <td>0.653550</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>XGboost</td>\n",
       "      <td>0.783693</td>\n",
       "      <td>0.664623</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Accuracy  roc_auc_score\n",
       "Linear Reg    0.782993       0.647550\n",
       "DT-Entropy    0.676098       0.529636\n",
       "DT-Gini       0.674782       0.532301\n",
       "RandomForest  0.777160       0.630191\n",
       "Bagging       0.765051       0.587351\n",
       "Adaboost      0.782678       0.653550\n",
       "XGboost       0.783693       0.664623"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(results).T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can observe that Xgboost classifier has highest accuracy score of 78.3% and roc_auc_score of 66% of all models built."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# --------------------------------------old code------------------------"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "# def validation(X,y,model,cv=230,scoring=\"roc_auc\"):\n",
    "#     results=cross_val_score(model,X=X,y=y,cv=cv,scoring=scoring,n_jobs=-1)\n",
    "#     print(\"mean_auc-\",results.mean())\n",
    "#     print(\"std_auc-\",results.std())\n",
    "#     return results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(233154, 34)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### LogisticRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# from sklearn.linear_model import LogisticRegression\n",
    "# model=LogisticRegression()\n",
    "# results=validation(X,y,model)\n",
    "# model_scores.update({\"Logistic_Regression\":results})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# model_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# results=validation(X,y,model,cv=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# model_scores.update({\"Logistic_Regression_cv10\":results})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# model_scores"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### LinearDiscriminantAnalysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "# model=LinearDiscriminantAnalysis()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# results=validation(X,y,model)\n",
    "# model_scores.update({\"LinearDiscriminantAnalysis\":results})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Tree Based Models:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# df1=df.copy()\n",
    "# # df1=df1.drop(columns=[\"UniqueID\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# df1[['supplier_id','Current_pincode_ID','Employee_code_ID']].nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# from sklearn.preprocessing import OneHotEncoder\n",
    "# for i in ['supplier_id','Current_pincode_ID','Employee_code_ID']:\n",
    "#     one_hot=OneHotEncoder(handle_unknown=\"ignore\",sparse=False)\n",
    "#     dummy_index=df1[i].value_counts()[:100].index.to_list()\n",
    "#     one_hot.fit(np.array(dummy_index).reshape(-1, 1))\n",
    "#     data=one_hot.transform(df1[i].values.reshape(-1, 1))\n",
    "#     dummy=pd.DataFrame(data=data,columns=[str(j)+i for j in dummy_index])\n",
    "#     df1=pd.concat([df1,dummy],axis=1)\n",
    "\n",
    "# df1=pd.get_dummies(data=df1,columns=cols_to_be_dummied,drop_first=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Xd=df1.drop('loan_default',axis=1)\n",
    "# yd=df1['loan_default']\n",
    "# from sklearn.tree import DecisionTreeClassifier\n",
    "# dtc_model=DecisionTreeClassifier()\n",
    "# # results=validation(X,y,model,cv=10)\n",
    "# # model_scores.update({\"Decision_Tree\":results})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "# from sklearn.model_selection import train_test_split,KFold,cross_val_score\n",
    "# X_traind, X_testd, y_traind, y_testd = train_test_split(Xd, yd, test_size=0.3, random_state=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=None,\n",
      "                       max_features=None, max_leaf_nodes=None,\n",
      "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "                       min_samples_leaf=1, min_samples_split=2,\n",
      "                       min_weight_fraction_leaf=0.0, presort=False,\n",
      "                       random_state=None, splitter='best') \n",
      "\n",
      "\n",
      " Accuracy for the Train: 0.9997365309086007\n",
      "\n",
      " Accuracy for the Test 0.6738673567129398 \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.80      0.78      0.79     54776\n",
      "           1       0.26      0.28      0.27     15171\n",
      "\n",
      "    accuracy                           0.67     69947\n",
      "   macro avg       0.53      0.53      0.53     69947\n",
      "weighted avg       0.68      0.67      0.68     69947\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# modelresult(dtc_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "var=[]\n",
    "mean=[]\n",
    "# for i in np.arange(1,21):\n",
    "kf = KFold(n_splits=3, shuffle=True, random_state=0)\n",
    "random = RandomForestClassifier(criterion='entropy',n_estimators=i,random_state=0)\n",
    "scores=cross_val_score(random,X,y,cv=3,scoring='roc_auc_score')\n",
    "mean.append(1-np.mean(scores,ddof=1))\n",
    "var.append(np.var(scores,ddof=1))\n",
    "print(i, np.var(scores,ddof=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "models_cv = []\n",
    "models_cv.append(('LinearRegression', lr))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import KFold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "models_cv = []\n",
    "models_cv.append(('LinearRegression', lr))\n",
    "# models.append(('BaggedLR',LR_bag))\n",
    "# models.append(('Adaboost',LR_AB))\n",
    "# models.append(('GBoost',LR_GB))\n",
    "# models.append(('KNN',KNN))\n",
    "\n",
    "\n",
    "\n",
    "# evaluate each model in turn\n",
    "results = []\n",
    "names = []\n",
    "for name, model in models_cv:\n",
    "    kfold = model_selection.KFold(shuffle=True,n_splits=3,random_state=0)\n",
    "    cv_results = model_selection.cross_val_score(model, X_new, Y,cv=kfold, scoring='neg_mean_squared_error')\n",
    "    results.append(np.sqrt(np.abs(cv_results)))\n",
    "    names.append(name)\n",
    "    print(\"%s: %f (%f)\" % (name, np.mean(np.sqrt(np.abs(cv_results))),np.std(np.sqrt(np.abs(cv_results)),ddof=1)))\n",
    "   # boxplot algorithm comparison\n",
    "fig = plt.figure()\n",
    "fig.suptitle('Algorithm Comparison')\n",
    "ax = fig.add_subplot(111)\n",
    "plt.boxplot(results)\n",
    "ax.set_xticklabels(names)\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}