Loan Approval ML Model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "1f27360b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd \n",
    "import numpy as np \n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "04def756",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import precision_score,confusion_matrix,classification_report,accuracy_score\n",
    "from sklearn.impute import KNNImputer\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier,VotingClassifier,GradientBoostingClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.feature_selection import SelectFromModel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "08e6c2ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "# loading the data\n",
    "data = pd.read_csv('Loan Approval.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "908e898a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# coping data\n",
    "df=data.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "51f6ac47",
   "metadata": {},
   "outputs": [],
   "source": [
    "# define features and target\n",
    "x=df.drop([\"Loan_ID\",\"Loan_Status\"],axis=1)\n",
    "y=df[\"Loan_Status\"]\n",
    "\n",
    "# split the data into train and test sets\n",
    "from sklearn.model_selection import train_test_split\n",
    "train_x,test_x,train_y,test_y = train_test_split(x,y,test_size=0.2,random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d4d6746",
   "metadata": {},
   "source": [
    "EDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "bec5c3fd",
   "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>Gender</th>\n",
       "      <th>Married</th>\n",
       "      <th>Dependents</th>\n",
       "      <th>Education</th>\n",
       "      <th>Self_Employed</th>\n",
       "      <th>ApplicantIncome</th>\n",
       "      <th>CoapplicantIncome</th>\n",
       "      <th>LoanAmount</th>\n",
       "      <th>Loan_Amount_Term</th>\n",
       "      <th>Credit_History</th>\n",
       "      <th>Property_Area</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>90</th>\n",
       "      <td>Male</td>\n",
       "      <td>Yes</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>2958</td>\n",
       "      <td>2900.0</td>\n",
       "      <td>131.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Semiurban</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>533</th>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>1</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>11250</td>\n",
       "      <td>0.0</td>\n",
       "      <td>196.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Semiurban</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>452</th>\n",
       "      <td>Male</td>\n",
       "      <td>Yes</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>3948</td>\n",
       "      <td>1733.0</td>\n",
       "      <td>149.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Rural</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>355</th>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>3813</td>\n",
       "      <td>0.0</td>\n",
       "      <td>116.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>266</th>\n",
       "      <td>Male</td>\n",
       "      <td>Yes</td>\n",
       "      <td>2</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>4708</td>\n",
       "      <td>1387.0</td>\n",
       "      <td>150.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Semiurban</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Gender Married Dependents Education Self_Employed  ApplicantIncome  \\\n",
       "90     Male     Yes          0  Graduate            No             2958   \n",
       "533    Male      No          1  Graduate            No            11250   \n",
       "452    Male     Yes          0  Graduate            No             3948   \n",
       "355  Female      No        NaN  Graduate            No             3813   \n",
       "266    Male     Yes          2  Graduate            No             4708   \n",
       "\n",
       "     CoapplicantIncome  LoanAmount  Loan_Amount_Term  Credit_History  \\\n",
       "90              2900.0       131.0             360.0             1.0   \n",
       "533                0.0       196.0             360.0             NaN   \n",
       "452             1733.0       149.0             360.0             0.0   \n",
       "355                0.0       116.0             180.0             1.0   \n",
       "266             1387.0       150.0             360.0             1.0   \n",
       "\n",
       "    Property_Area  \n",
       "90      Semiurban  \n",
       "533     Semiurban  \n",
       "452         Rural  \n",
       "355         Urban  \n",
       "266     Semiurban  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#check for first 5 rows of the train data\n",
    "train_x.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0d9321e6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 491 entries, 90 to 559\n",
      "Data columns (total 11 columns):\n",
      " #   Column             Non-Null Count  Dtype  \n",
      "---  ------             --------------  -----  \n",
      " 0   Gender             479 non-null    object \n",
      " 1   Married            488 non-null    object \n",
      " 2   Dependents         476 non-null    object \n",
      " 3   Education          491 non-null    object \n",
      " 4   Self_Employed      463 non-null    object \n",
      " 5   ApplicantIncome    491 non-null    int64  \n",
      " 6   CoapplicantIncome  491 non-null    float64\n",
      " 7   LoanAmount         471 non-null    float64\n",
      " 8   Loan_Amount_Term   480 non-null    float64\n",
      " 9   Credit_History     450 non-null    float64\n",
      " 10  Property_Area      491 non-null    object \n",
      "dtypes: float64(4), int64(1), object(6)\n",
      "memory usage: 46.0+ KB\n"
     ]
    }
   ],
   "source": [
    "#check columns informations\n",
    "train_x.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "dea348bd",
   "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>ApplicantIncome</th>\n",
       "      <th>CoapplicantIncome</th>\n",
       "      <th>LoanAmount</th>\n",
       "      <th>Loan_Amount_Term</th>\n",
       "      <th>Credit_History</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>491.000000</td>\n",
       "      <td>491.000000</td>\n",
       "      <td>471.000000</td>\n",
       "      <td>480.000000</td>\n",
       "      <td>450.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5619.018330</td>\n",
       "      <td>1525.863381</td>\n",
       "      <td>147.324841</td>\n",
       "      <td>342.550000</td>\n",
       "      <td>0.837778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>6659.021257</td>\n",
       "      <td>2787.398712</td>\n",
       "      <td>88.482349</td>\n",
       "      <td>63.953251</td>\n",
       "      <td>0.369065</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>150.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>36.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2932.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>360.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>3850.000000</td>\n",
       "      <td>997.000000</td>\n",
       "      <td>128.000000</td>\n",
       "      <td>360.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>5763.000000</td>\n",
       "      <td>2167.500000</td>\n",
       "      <td>165.500000</td>\n",
       "      <td>360.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>81000.000000</td>\n",
       "      <td>41667.000000</td>\n",
       "      <td>700.000000</td>\n",
       "      <td>480.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       ApplicantIncome  CoapplicantIncome  LoanAmount  Loan_Amount_Term  \\\n",
       "count       491.000000         491.000000  471.000000        480.000000   \n",
       "mean       5619.018330        1525.863381  147.324841        342.550000   \n",
       "std        6659.021257        2787.398712   88.482349         63.953251   \n",
       "min         150.000000           0.000000    9.000000         36.000000   \n",
       "25%        2932.000000           0.000000  100.000000        360.000000   \n",
       "50%        3850.000000         997.000000  128.000000        360.000000   \n",
       "75%        5763.000000        2167.500000  165.500000        360.000000   \n",
       "max       81000.000000       41667.000000  700.000000        480.000000   \n",
       "\n",
       "       Credit_History  \n",
       "count      450.000000  \n",
       "mean         0.837778  \n",
       "std          0.369065  \n",
       "min          0.000000  \n",
       "25%          1.000000  \n",
       "50%          1.000000  \n",
       "75%          1.000000  \n",
       "max          1.000000  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#check basic informations of numeric columns\n",
    "train_x.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "caa9abc6",
   "metadata": {},
   "source": [
    "we can identify some right skewnees in the continuous  columns "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "8af948ff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Gender                 0\n",
       "Married                0\n",
       "Dependents             0\n",
       "Education              0\n",
       "Self_Employed          0\n",
       "ApplicantIncome        0\n",
       "CoapplicantIncome    228\n",
       "LoanAmount             0\n",
       "Loan_Amount_Term       0\n",
       "Credit_History        73\n",
       "Property_Area          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#check for any zero values to handle it if it's not reasonable\n",
    "train_x[train_x==0].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "8dfc1212",
   "metadata": {},
   "outputs": [],
   "source": [
    "#check unique values of categorical columns and it's frequancy on our data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "d07afbd0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Male      389\n",
       "Female     90\n",
       "Name: Gender, dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Gender'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "1a9c3f7e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Yes    323\n",
       "No     165\n",
       "Name: Married, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Married'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "40533e9f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "No     394\n",
       "Yes     69\n",
       "Name: Self_Employed, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Self_Employed'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a8025561",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Graduate        383\n",
       "Not Graduate    108\n",
       "Name: Education, dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Education'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "db646efa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     264\n",
       "1      86\n",
       "2      85\n",
       "3+     41\n",
       "Name: Dependents, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Dependents'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "d7f9659e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Semiurban    185\n",
       "Urban        163\n",
       "Rural        143\n",
       "Name: Property_Area, dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Property_Area'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "91f400a2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "360.0    410\n",
       "180.0     34\n",
       "480.0     12\n",
       "300.0     11\n",
       "240.0      4\n",
       "84.0       3\n",
       "60.0       2\n",
       "36.0       2\n",
       "120.0      2\n",
       "Name: Loan_Amount_Term, dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Loan_Amount_Term'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "abfe8b60",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0    377\n",
       "0.0     73\n",
       "Name: Credit_History, dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Credit_History'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "40f94496",
   "metadata": {},
   "source": [
    "from the previous EDA we confirm that our data have impalanced classes sow we will use models that can deal with this like random forest for example "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "6fb72a7a",
   "metadata": {},
   "outputs": [],
   "source": [
    "#check the distribution of the cont data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "8df55146",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='ApplicantIncome', ylabel='Count'>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig=plt.figure(figsize=(6,3))\n",
    "sns.histplot(train_x['ApplicantIncome'],kde=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "635c524f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='CoapplicantIncome', ylabel='Count'>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig=plt.figure(figsize=(6,3))\n",
    "sns.histplot(train_x['CoapplicantIncome'],kde=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "c74a6aaa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='LoanAmount', ylabel='Count'>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig=plt.figure(figsize=(6,3))\n",
    "sns.histplot(train_x['LoanAmount'],kde=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "ff1c9ab1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#identify relationship between con columns with each other \n",
    "sns.heatmap(train_x[train_x.columns[5:8]].corr(),annot=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3632869e",
   "metadata": {},
   "source": [
    "we can see that there a strong relationship between loan amount and applicantincome "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed990983",
   "metadata": {},
   "source": [
    "data preparation and cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "67487610",
   "metadata": {},
   "outputs": [],
   "source": [
    "# as we can see the con data is right skewed so we will apply a sqrt transformation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "45a3ec69",
   "metadata": {},
   "outputs": [],
   "source": [
    "# perform sqrt transformation on columns\n",
    "train_x[['LoanAmount','ApplicantIncome','CoapplicantIncome']]=np.sqrt(train_x[['LoanAmount','ApplicantIncome','CoapplicantIncome']])+1\n",
    "test_x[['LoanAmount','ApplicantIncome','CoapplicantIncome']]=np.sqrt(test_x[['LoanAmount','ApplicantIncome','CoapplicantIncome']])+1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "c6db74e2",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='LoanAmount', ylabel='Count'>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# constract a sample to see the diffrence \n",
    "fig=plt.figure(figsize=(6,3))\n",
    "sns.histplot(train_x['LoanAmount'],kde=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e847de88",
   "metadata": {},
   "source": [
    "impute missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "4dc84335",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Gender               12\n",
       "Married               3\n",
       "Dependents           15\n",
       "Education             0\n",
       "Self_Employed        28\n",
       "ApplicantIncome       0\n",
       "CoapplicantIncome     0\n",
       "LoanAmount           20\n",
       "Loan_Amount_Term     11\n",
       "Credit_History       41\n",
       "Property_Area         0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "123af4c0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# encode non numeric values \n",
    "train_x['Married'].replace(['Yes','No'],[1,0], inplace=True)\n",
    "test_x['Married'].replace(['Yes','No'],[1,0], inplace=True)\n",
    "\n",
    "train_x['Gender'].replace(['Male','Female'],[1,0], inplace=True)\n",
    "test_x['Gender'].replace(['Male','Female'],[1,0], inplace=True)\n",
    "\n",
    "train_x['Dependents'].replace(['3+','2','1','0'],[3,2,1,0], inplace=True)\n",
    "test_x['Dependents'].replace(['3+','2','1','0'],[3,2,1,0], inplace=True)\n",
    "\n",
    "train_x['Education'].replace(['Graduate','Not Graduate'],[1,0], inplace=True)\n",
    "test_x['Education'].replace(['Graduate','Not Graduate'],[1,0], inplace=True)\n",
    "\n",
    "train_x['Self_Employed'].replace(['Yes','No'],[1,0], inplace=True)\n",
    "test_x['Self_Employed'].replace(['Yes','No'],[1,0], inplace=True)\n",
    "\n",
    "train_x['Property_Area'].replace(['Urban','Semiurban','Rural'],[2,1,0], inplace=True)\n",
    "test_x['Property_Area'].replace(['Urban','Semiurban','Rural'],[2,1,0], inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "00f7df91",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# impute misssing data using knn imputer\n",
    "col=train_x.columns\n",
    "imputer = KNNImputer()\n",
    "train_x = imputer.fit_transform(train_x)\n",
    "test_x = imputer.fit_transform(test_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "43cca89f",
   "metadata": {},
   "outputs": [],
   "source": [
    "#convert the numpy array into dataframe\n",
    "train_x = pd.DataFrame(train_x,columns=col)\n",
    "test_x = pd.DataFrame(test_x,columns=col)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "f876cc15",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0    412\n",
       "1.0     69\n",
       "0.2      6\n",
       "0.4      3\n",
       "0.6      1\n",
       "Name: Self_Employed, dtype: int64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Self_Employed'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "3a9e350e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0    325\n",
       "0.0    165\n",
       "0.4      1\n",
       "Name: Married, dtype: int64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Married'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "1568bdf3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0    393\n",
       "0.0     90\n",
       "0.6      3\n",
       "0.4      3\n",
       "0.8      2\n",
       "Name: Gender, dtype: int64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Gender'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "103e4cb9",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0    264\n",
       "1.0     88\n",
       "2.0     85\n",
       "3.0     41\n",
       "1.2      4\n",
       "0.6      3\n",
       "1.4      2\n",
       "0.4      2\n",
       "1.6      1\n",
       "1.8      1\n",
       "Name: Dependents, dtype: int64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Dependents'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "774680e4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "360.0    414\n",
       "180.0     34\n",
       "480.0     12\n",
       "300.0     11\n",
       "240.0      4\n",
       "84.0       3\n",
       "60.0       2\n",
       "324.0      2\n",
       "36.0       2\n",
       "120.0      2\n",
       "372.0      2\n",
       "312.0      1\n",
       "288.0      1\n",
       "348.0      1\n",
       "Name: Loan_Amount_Term, dtype: int64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Loan_Amount_Term'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "d85491c5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0    390\n",
       "0.0     73\n",
       "0.8     17\n",
       "0.6     10\n",
       "0.4      1\n",
       "Name: Credit_History, dtype: int64"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Credit_History'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "ba1cf137",
   "metadata": {},
   "outputs": [],
   "source": [
    "# as we can see there are some values that should be handeled we will use a threshold of 0.5 in binary classes\n",
    "# and x.5 in more than 2 classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "5d3b1e41",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_x['Self_Employed'] = train_x['Self_Employed'].apply(lambda x: 1 if x > 0.5 else 0)\n",
    "test_x['Self_Employed'] = test_x['Self_Employed'].apply(lambda x: 1 if x > 0.5 else 0)\n",
    "\n",
    "train_x['Married'] = train_x['Married'].apply(lambda x: 1 if x > 0.5 else 0)\n",
    "test_x['Married'] = test_x['Married'].apply(lambda x: 1 if x > 0.5 else 0)\n",
    "\n",
    "train_x['Gender'] = train_x['Gender'].apply(lambda x: 1 if x > 0.5 else 0)\n",
    "test_x['Gender'] = test_x['Gender'].apply(lambda x: 1 if x > 0.5 else 0)\n",
    "\n",
    "train_x['Dependents'] = train_x['Dependents'].apply(lambda x: 0 if x < 0.5 else (1 if x <= 1.5 else 2))\n",
    "test_x['Dependents'] = test_x['Dependents'].apply(lambda x: 0 if x < 0.5 else (1 if x <= 1.5 else 2))\n",
    "\n",
    "train_x['Credit_History'] = train_x['Credit_History'].apply(lambda x: 1 if x > 0.5 else 0)\n",
    "test_x['Credit_History'] = test_x['Credit_History'].apply(lambda x: 1 if x > 0.5 else 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "c3623292",
   "metadata": {},
   "outputs": [],
   "source": [
    "# decode non numeric data again\n",
    "train_x['Married'].replace([1,0],['Yes','No'], inplace=True)\n",
    "test_x['Married'].replace([1,0],['Yes','No'], inplace=True)\n",
    "\n",
    "train_x['Gender'].replace([1,0],['Male','Female'], inplace=True)\n",
    "test_x['Gender'].replace([1,0],['Male','Female'], inplace=True)\n",
    "\n",
    "train_x['Dependents'].replace([3,2,1,0],['3+','2','1','0'], inplace=True)\n",
    "test_x['Dependents'].replace([3,2,1,0],['3+','2','1','0'], inplace=True)\n",
    "\n",
    "train_x['Education'].replace([1,0],['Graduate','Not Graduate'], inplace=True)\n",
    "test_x['Education'].replace([1,0],['Graduate','Not Graduate'], inplace=True)\n",
    "\n",
    "train_x['Self_Employed'].replace([1,0],['Yes','No'], inplace=True)\n",
    "test_x['Self_Employed'].replace([1,0],['Yes','No'], inplace=True)\n",
    "\n",
    "train_x['Property_Area'].replace([2,1,0],['Urban','Semiurban','Rural'], inplace=True)\n",
    "test_x['Property_Area'].replace([2,1,0],['Urban','Semiurban','Rural'], inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "ced830a6",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Gender               0\n",
       "Married              0\n",
       "Dependents           0\n",
       "Education            0\n",
       "Self_Employed        0\n",
       "ApplicantIncome      0\n",
       "CoapplicantIncome    0\n",
       "LoanAmount           0\n",
       "Loan_Amount_Term     0\n",
       "Credit_History       0\n",
       "Property_Area        0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2101501d",
   "metadata": {},
   "source": [
    "loan amount term column has many values which will lead for more complexity so we will reduce the values by adding new category called other "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "2548d3b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_x['Loan_Amount_Term'] = np.where((train_x['Loan_Amount_Term'] >= 0) & (train_x['Loan_Amount_Term'] <= 250),\n",
    "    180,train_x['Loan_Amount_Term'])\n",
    "test_x['Loan_Amount_Term'] = np.where((test_x['Loan_Amount_Term'] >= 0) & (test_x['Loan_Amount_Term'] <= 250),\n",
    "    180,test_x['Loan_Amount_Term'])\n",
    "\n",
    "train_x['Loan_Amount_Term'] = np.where((train_x['Loan_Amount_Term'] >= 251) & (train_x['Loan_Amount_Term'] <= 310),\n",
    "    300,train_x['Loan_Amount_Term'])\n",
    "test_x['Loan_Amount_Term'] = np.where((test_x['Loan_Amount_Term'] >= 251) & (test_x['Loan_Amount_Term'] <= 310),\n",
    "    300,test_x['Loan_Amount_Term'])\n",
    "\n",
    "train_x['Loan_Amount_Term'] = np.where((train_x['Loan_Amount_Term'] >= 311) & (train_x['Loan_Amount_Term'] <= 400),\n",
    "    360,train_x['Loan_Amount_Term'])\n",
    "test_x['Loan_Amount_Term'] = np.where((test_x['Loan_Amount_Term'] >= 311) & (test_x['Loan_Amount_Term'] <= 400),\n",
    "    360,test_x['Loan_Amount_Term'])\n",
    "\n",
    "train_x['Loan_Amount_Term']=np.where(train_x['Loan_Amount_Term'] > 400,480,train_x['Loan_Amount_Term'])\n",
    "test_x['Loan_Amount_Term']=np.where(test_x['Loan_Amount_Term'] > 400,480,test_x['Loan_Amount_Term'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "63975b06",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "360.0    420\n",
       "180.0     47\n",
       "480.0     12\n",
       "300.0     12\n",
       "Name: Loan_Amount_Term, dtype: int64"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x['Loan_Amount_Term'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "1e9b5a8f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# apply one hot encoding on categorical data\n",
    "train_x=pd.get_dummies(train_x,columns=['Gender','Married','Dependents','Loan_Amount_Term','Credit_History',\n",
    "                                        'Education','Self_Employed','Property_Area'])\n",
    "test_x=pd.get_dummies(test_x,columns=['Gender','Married','Dependents','Loan_Amount_Term','Credit_History',\n",
    "                                        'Education','Self_Employed','Property_Area'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91f952af",
   "metadata": {},
   "source": [
    "applying lable encoding on the target "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "73999e37",
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "train_y.replace(['Y','N'],[1,0],inplace=True)\n",
    "test_y.replace(['Y','N'],[1,0],inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "5511be87",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Loan_Status\n",
       "1              332\n",
       "0              159\n",
       "dtype: int64"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_y.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0b14859e",
   "metadata": {},
   "source": [
    "model selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "ccc183bf",
   "metadata": {},
   "outputs": [],
   "source": [
    "log_clf=LogisticRegression()\n",
    "gb_clf = GradientBoostingClassifier()\n",
    "rf = RandomForestClassifier()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f2eacd1a",
   "metadata": {},
   "source": [
    "logistic regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "d002a109",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "precision score on train data : 78.17745803357315\n",
      "precision score on test data : 83.01886792452831\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\maher\\anaconda3\\Lib\\site-packages\\sklearn\\utils\\validation.py:1184: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "C:\\Users\\maher\\anaconda3\\Lib\\site-packages\\sklearn\\linear_model\\_logistic.py:460: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n"
     ]
    }
   ],
   "source": [
    " log_clf.fit(train_x,train_y)\n",
    "y_train_pred = log_clf.predict(train_x)\n",
    "train_precision_score = precision_score(train_y, y_train_pred)\n",
    "\n",
    "y_test_pred = log_clf.predict(test_x)\n",
    "test_precision_score = precision_score(test_y, y_test_pred)\n",
    "\n",
    "print(f'precision score on train data : {train_precision_score * 100}')\n",
    "print(f'precision score on test data : {test_precision_score * 100}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "58d302c8",
   "metadata": {},
   "source": [
    "random forrest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "61dec97f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "precision score on train data : 100.0\n",
      "precision score on test data : 81.63265306122449\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\maher\\anaconda3\\Lib\\site-packages\\sklearn\\base.py:1151: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  return fit_method(estimator, *args, **kwargs)\n"
     ]
    }
   ],
   "source": [
    " rf.fit(train_x,train_y)\n",
    "y_train_pred = rf.predict(train_x)\n",
    "train_precision_score = precision_score(train_y, y_train_pred)\n",
    "\n",
    "y_test_pred = rf.predict(test_x)\n",
    "test_precision_score = precision_score(test_y, y_test_pred)\n",
    "\n",
    "print(f'precision score on train data : {train_precision_score * 100}')\n",
    "print(f'precision score on test data : {test_precision_score * 100}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6133da99",
   "metadata": {},
   "source": [
    "gradient boosting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "b0917c8b",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "precision score on train data : 86.42297650130548\n",
      "precision score on test data : 84.84848484848484\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\maher\\anaconda3\\Lib\\site-packages\\sklearn\\ensemble\\_gb.py:424: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n"
     ]
    }
   ],
   "source": [
    "gb_clf.fit(train_x,train_y)\n",
    "y_train_pred = gb_clf.predict(train_x)\n",
    "train_precision_score = precision_score(train_y, y_train_pred)\n",
    "\n",
    "y_test_pred = gb_clf.predict(test_x)\n",
    "test_precision_score = precision_score(test_y, y_test_pred)\n",
    "\n",
    "print(f'precision score on train data : {train_precision_score * 100}')\n",
    "print(f'precision score on test data : {test_precision_score * 100}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "29f61b6f",
   "metadata": {},
   "outputs": [],
   "source": [
    "#seems gradient boosting is most suitable , so let's see his confusion matrix "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "562a38cc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[18 15]\n",
      " [ 6 84]]\n"
     ]
    }
   ],
   "source": [
    "cm = confusion_matrix(test_y, y_test_pred)\n",
    "print(cm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "e57d29a7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.75      0.55      0.63        33\n",
      "           1       0.85      0.93      0.89        90\n",
      "\n",
      "    accuracy                           0.83       123\n",
      "   macro avg       0.80      0.74      0.76       123\n",
      "weighted avg       0.82      0.83      0.82       123\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(test_y, y_test_pred))"
   ]
  }
 ],
 "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": "3.11.4"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}