{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[Problem Statement](#ps)\n",
    "\n",
    "[Hypothesis Generation](#hg)\n",
    "\n",
    "[Getting the system ready and loading the data](#ld)\n",
    "\n",
    "[Understanding the data](#ud)\n",
    "\n",
    "[Exploratory Data Analysis (EDA)](#ed)\n",
    "\n",
    "[Univariate Analysis](#uv)\n",
    "\n",
    "[Bivariate Analysis](#bv)\n",
    "\n",
    "[Missing value and outlier treatment](#mv)\n",
    "\n",
    "[Evaluation Metrics for classification problems](#ev)\n",
    "\n",
    "[Model Building : Part I](#mb)\n",
    "\n",
    "[Logistic Regression using stratified k-folds cross validation](#lreg)\n",
    "\n",
    "[Feature Engineering](#fe)\n",
    "\n",
    "[Model Building : Part II](#mbb)\n",
    "\n",
    "[Logistic Regression](#lrr)\n",
    "\n",
    "[Decision tree](#dt)\n",
    "\n",
    "[Random Forest](#rf)\n",
    "\n",
    "[XGBoost](#xgb)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id=\"ps\"> </a>\n",
    "\n",
    "Understanding the problem statement is the first and foremost step. This would help you give an intuition of what you will face ahead of time. Let us see the problem statement -\n",
    "\n",
    "Dream Housing Finance company deals in all home loans. They have presence across all urban, semi urban and rural areas. Customer first apply for home loan after that company validates the customer eligibility for loan. Company wants to automate the loan eligibility process (real time) based on customer detail provided while filling online application form. These details are Gender, Marital Status, Education, Number of Dependents, Income, Loan Amount, Credit History and others. To automate this process, they have given a problem to identify the customers segments, those are eligible for loan amount so that they can specifically target these customers."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id=\"hg\"> </a>\n",
    "\n",
    "After looking at the problem statement, we will now move into hypothesis generation. It is the process of listing out all the possible factors that can affect the outcome.\n",
    "\n",
    "What is hypothesis generation?\n",
    "This is a very important stage in any data science/machine learning pipeline. It involves understanding the problem in detail by brainstorming as many factors as possible which can impact the outcome. It is done by understanding the problem statement thoroughly and before looking at the data.\n",
    "\n",
    "Below are some of the factors which I think can affect the Loan Approval (dependent variable for this loan prediction problem):\n",
    "\n",
    "Salary: Applicants with high income should have more chances of loan approval.\n",
    "Previous history: Applicants who have repayed their previous debts should have higher chances of loan approval.\n",
    "Loan amount: Loan approval should also depend on the loan amount. If the loan amount is less, chances of loan approval should be high.\n",
    "Loan term: Loan for less time period and less amount should have higher chances of approval.\n",
    "EMI: Lesser the amount to be paid monthly to repay the loan, higher the chances of loan approval.\n",
    "These are some of the factors which i think can affect the target variable, you can come up with many more factors.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id=\"ld\"> </a>\n",
    "\n",
    "### Loading data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd \n",
    "import numpy as np                     # For mathematical calculations \n",
    "import seaborn as sns                  # For data visualization \n",
    "import matplotlib.pyplot as plt        # For plotting graphs \n",
    "%matplotlib inline \n",
    "import warnings                        # To ignore any warnings warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "\n",
    "train = pd.read_csv(\"train.csv\") \n",
    "\n",
    "test = pd.read_csv(\"test.csv\")\n",
    "\n",
    "train_original = train.copy() \n",
    "\n",
    "test_original = test.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Loan_ID', 'Gender', 'Married', 'Dependents', 'Education',\n",
       "       'Self_Employed', 'ApplicantIncome', 'CoapplicantIncome', 'LoanAmount',\n",
       "       'Loan_Amount_Term', 'Credit_History', 'Property_Area', 'Loan_Status'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have 12 independent variables and 1 target variable, i.e. Loan_Status in the train dataset. Let’s also have a look at the columns of test dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Loan_ID', 'Gender', 'Married', 'Dependents', 'Education',\n",
       "       'Self_Employed', 'ApplicantIncome', 'CoapplicantIncome', 'LoanAmount',\n",
       "       'Loan_Amount_Term', 'Credit_History', 'Property_Area'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have similar features in the test dataset as the train dataset except the Loan_Status. We will predict the Loan_Status using the model built using the train data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Loan_ID               object\n",
       "Gender                object\n",
       "Married               object\n",
       "Dependents            object\n",
       "Education             object\n",
       "Self_Employed         object\n",
       "ApplicantIncome        int64\n",
       "CoapplicantIncome    float64\n",
       "LoanAmount           float64\n",
       "Loan_Amount_Term     float64\n",
       "Credit_History       float64\n",
       "Property_Area         object\n",
       "Loan_Status           object\n",
       "dtype: object"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see there are three format of data types:\n",
    "\n",
    "object: Object format means variables are categorical. Categorical variables in our dataset are: Loan_ID, Gender, Married, Dependents, Education, Self_Employed, Property_Area, Loan_Status\n",
    "\n",
    "int64: It represents the integer variables. ApplicantIncome is of this format.\n",
    "\n",
    "float64: It represents the variable which have some decimal values involved. They are also numerical variables. Numerical variables in our dataset are: CoapplicantIncome, LoanAmount, Loan_Amount_Term, and Credit_History"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((614, 13), (367, 12))"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.shape, test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this section, we will do univariate analysis. It is the simplest form of analyzing data where we examine each variable individually. For categorical features we can use frequency table or bar plots which will calculate the number of each category in a particular variable. For numerical features, probability density plots can be used to look at the distribution of the variable."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Target Variable\n",
    "\n",
    "We will first look at the target variable, i.e., Loan_Status. As it is a categorical variable, let us look at its frequency table, percentage distribution and bar plot.\n",
    "\n",
    "Frequency table of a variable will give us the count of each category in that variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Y    422\n",
       "N    192\n",
       "Name: Loan_Status, dtype: int64"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train['Loan_Status'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Y    0.687296\n",
       "N    0.312704\n",
       "Name: Loan_Status, dtype: float64"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Normalize can be set to True to print proportions instead of number\n",
    "\n",
    "train['Loan_Status'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f1a3e6fa400>"
      ]
     },
     "execution_count": 81,
     "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": [
    "train['Loan_Status'].value_counts().plot.bar()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id=\"uv\"> </a>\n",
    "\n",
    "Now lets visualize each variable separately. Different types of variables are Categorical, ordinal and numerical.\n",
    "\n",
    "Categorical features: These features have categories (Gender, Married, Self_Employed, Credit_History, Loan_Status)\n",
    "\n",
    "Ordinal features: Variables in categorical features having some order involved (Dependents, Education, Property_Area)\n",
    "\n",
    "Numerical features: These features have numerical values (ApplicantIncome, CoapplicantIncome, LoanAmount, Loan_Amount_Term)\n",
    "\n",
    "Let’s visualize the categorical and ordinal features first."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Independent Variable (Categorical)\n",
    "\n",
    "plt.figure(1)\n",
    "plt.subplot(221)\n",
    "train['Gender'].value_counts(normalize=True).plot.bar(figsize=(20,10), title= 'Gender') \n",
    "plt.subplot(222)\n",
    "train['Married'].value_counts(normalize=True).plot.bar(title= 'Married')\n",
    "plt.subplot(223)\n",
    "train['Self_Employed'].value_counts(normalize=True).plot.bar(title= 'Self_Employed')\n",
    "plt.subplot(224)\n",
    "train['Credit_History'].value_counts(normalize=True).plot.bar(title= 'Credit_History')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It can be inferred from the above bar plots that:\n",
    "\n",
    "80% applicants in the dataset are male.\n",
    "Around 65% of the applicants in the dataset are married.\n",
    "Around 15% applicants in the dataset are self employed.\n",
    "Around 85% applicants have repaid their debts."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1728x432 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Independent Variable (Ordinal)\n",
    "\n",
    "plt.figure(1)\n",
    "plt.subplot(131)\n",
    "train['Dependents'].value_counts(normalize=True).plot.bar(figsize=(24,6), title= 'Dependents') \n",
    "plt.subplot(132)\n",
    "train['Education'].value_counts(normalize=True).plot.bar(title= 'Education')\n",
    "plt.subplot(133)\n",
    "train['Property_Area'].value_counts(normalize=True).plot.bar(title= 'Property_Area') \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Independent Variable (Numerical)\n",
    "\n",
    "Till now we have seen the categorical and ordinal variables and now lets visualize the numerical variables. Lets look at the distribution of Applicant income first."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(1)\n",
    "plt.subplot(121)\n",
    "sns.distplot(train['ApplicantIncome'])\n",
    "plt.subplot(122)\n",
    "train['ApplicantIncome'].plot.box(figsize=(16,5))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It can be inferred that most of the data in the distribution of applicant income is towards left which means it is not normally distributed. We will try to make it normal in later sections as algorithms works better if the data is normally distributed.\n",
    "\n",
    "The boxplot confirms the presence of a lot of outliers/extreme values. This can be attributed to the income disparity in the society. Part of this can be driven by the fact that we are looking at people with different education levels. Let us segregate them by Education:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,0.98,'')"
      ]
     },
     "execution_count": 85,
     "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": [
    "train.boxplot(column='ApplicantIncome', by = 'Education')\n",
    "plt.suptitle(\"\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that there are a higher number of graduates with very high incomes, which are appearing to be the outliers.\n",
    "\n",
    "Let’s look at the Coapplicant income distribution."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(1)\n",
    "plt.subplot(121)\n",
    "sns.distplot(train['CoapplicantIncome'])\n",
    "plt.subplot(122)\n",
    "train['CoapplicantIncome'].plot.box(figsize=(16,5)) \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see a similar distribution as that of the applicant income. Majority of coapplicant’s income ranges from 0 to 5000. We also see a lot of outliers in the coapplicant income and it is not normally distributed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7IAAAFACAYAAACBRUDPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3Xt81uWd5//XlTNJOAXCKRwCghqM4oF6qExbah11etCdrTviztRWVtedNuMM7k9t89jZbX+NU3a77XRop6wWptoZoh23nTpWa23FTqmK4gmBiCJyCAiEAAGCCQlc+0duaDhJgCTf3Hdez8cjj9zf676+9/2+W/T2w3UKMUYkSZIkSUoXWUkHkCRJkiTpVFjISpIkSZLSioWsJEmSJCmtWMhKkiRJktKKhawkSZIkKa1YyEqSJEmS0oqFrCRJkiQprVjISpIkSZLSioWsJEmSJCmt5CQd4FQMHz48lpeXJx1DkpQBXn755e0xxtKkc6Q7v5slSd3lVL6b06qQLS8vZ9myZUnHkCRlgBDC+qQzZAK/myVJ3eVUvpudWixJkiRJSisWspIkSZKktGIhK0mSJElKKxaykiRJkqS0YiErSZIkSUorFrKSJEmSpLRiIStJkiRJSisWspIkZagQwpAQwqMhhDdDCHUhhCtCCCUhhKdDCG+nfg9N9Q0hhL8LIawJISwPIVycdH4p3VVVVVFQUEAIgYKCAqqqqpKOJGUMC1lJkjLXd4BfxBjPBaYBdcC9wK9jjFOAX6euAa4DpqR+bge+3/txpcxRVVXF/Pnzue+++2hubua+++5j/vz5FrNSN7GQlSQpA4UQBgEfARYAxBj3xxh3AdcDD6a6PQjckHp8PfBQ7PACMCSEMLqXY0sZ44EHHmDu3LnMmTOHwsJC5syZw9y5c3nggQeSjiZlBAtZSZIy0ySgAfiHEMKrIYQfhBCKgJExxvcAUr9HpPqXARs73V+fajtGCOH2EMKyEMKyhoaGnvsEUhprbW3ljjvuOKLtjjvuoLW1NaFEUmbJSTqA0suipRtO2ufmy8b3QhJJ0knkABcDVTHGpSGE7/D7acTHE47TFo/XMcZ4P3A/wPTp04/bR+rv8vPzmT9/PnPmzDncNn/+fPLz8xNMJWUOR2QlScpM9UB9jHFp6vpROgrbrYemDKd+b+vUf1yn+8cCm3spq5RxbrvtNu655x6+9a1vsW/fPr71rW9xzz33cNtttyUdTcoIjshKkpSBYoxbQggbQwjnxBhXA1cBq1I/twDfSP3+WeqWx4AvhRAeBi4Dmg5NQZZ06ubNmwfAV77yFe666y7y8/O54447DrdLOjMWspIkZa4q4J9CCHnAWuALdMzG+nEIYTawAbgx1fcJ4I+ANcC+VF9JZ2DevHkWrlIP6VIhG0K4lo4t/LOBH8QYv3HU8/nAQ8AlQCPwJzHGdSGEYXRMZfoQ8MMY45c63XMJ8ENgAB1fnnfGGF1nI0lSN4kxvgZMP85TVx2nbwS+2OOhJEnqBiddIxtCyAa+R8f5clOBWSGEqUd1mw3sjDFOBr4NzE21twD/Dfivx3np79NxTt2hM+uuPZ0PIEmSJEnqX7qy2dOlwJoY49oY437gYTrOmuus85l0jwJXhRBCjLE5xriEjoL2sNTmEoNijM+n/gb4IX5/jp0kSZIkSSfUlUK2K+fKHe4TY2wHmoBhJ3nN+pO8piRJkiRJx+hKIduVc+W6fPbcqfb30HVJkiRJUmddKWS7cq7c4T4hhBxgMLDjJK859iSvCXQcuh5jnB5jnF5aWtqFuJIkSVLyamtrqaysJDs7m8rKSmpra5OOJGWMrhSyLwFTQggTU9v330THWXOdPUbHWXQAnwWe+aAdiFPn0u0JIVweQgjA5/j9OXaSJElSWqutraW6upp58+bR0tLCvHnzqK6utpiVuslJC9nUmtcvAU8BdcCPY4wrQwhfCyF8JtVtATAshLAGmAPce+j+EMI64FvA50MI9Z12PP4vwA/oOK/uHeDJ7vlIkiRJUrJqampYsGABM2fOJDc3l5kzZ7JgwQJqamqSjiZlhC6dIxtjfIKOs147t/11p8ct/P5A9aPvLT9B+zKgsqtBJUmSpHRRV1fHjBkzjmibMWMGdXV1CSWSMktXphZLkiRJOgUVFRUsWbLkiLYlS5ZQUVGRUCIps1jISpIkSd2surqa2bNns3jxYtra2li8eDGzZ8+muro66WhSRujS1GJJkiRJXTdr1iwAqqqqqKuro6KigpqamsPtks6MhawkSZLUA2bNmmXhKvUQpxZLkiRJktKKhawkSZIkKa1YyEqSJEmS0oqFrCRJkiQprVjISpIkSZLSioWsJEmSJCmtWMhKkiRJktKKhawkSZIkKa1YyEqSJEmS0oqFrCRJkiQprVjISpIkSZLSioWsJEmSJCmtWMhKkiRJktKKhawkSZIkKa1YyEqSJEmS0oqFrCRJkiQprVjISpIkSZLSioWsJEmSJCmtWMhKkiRJktKKhawkSZIkKa1YyEqSJEmS0oqFrCRJGSqEsC6E8EYI4bUQwrJUW0kI4ekQwtup30NT7SGE8HchhDUhhOUhhIuTTS+lv9raWiorK8nOzqayspLa2tqkI0kZw0JWkqTMNjPGeGGMcXrq+l7g1zHGKcCvU9cA1wFTUj+3A9/v9aRSBqmtraW6upp58+bR0tLCvHnzqK6utpiVuomFrCRJ/cv1wIOpxw8CN3Rqfyh2eAEYEkIYnURAKRPU1NQwbdo0rrvuOvLy8rjuuuuYNm0aNTU1SUeTMoKFrCRJmSsCvwwhvBxCuD3VNjLG+B5A6veIVHsZsLHTvfWptmOEEG4PISwLISxraGjooehSelu5ciWPP/449913H83Nzdx33308/vjjrFy5MuloUkbISTqAMs+ipRtO2ufmy8b3QhJJ6veujDFuDiGMAJ4OIbz5AX3Dcdri8TrGGO8H7geYPn36cftI/V0Igdtuu405c+YAMGfOHNasWcP8+fMTTiZlBkdkJUnKUDHGzanf24CfApcCWw9NGU793pbqXg+M63T7WGBz76WVMkuMkSeffJLFixfT1tbG4sWLefLJJ4nRv/uRuoOFrCRJGSiEUBRCGHjoMfCHwArgMeCWVLdbgJ+lHj8GfC61e/HlQNOhKciSTl1+fj5XXnklVVVVFBQUUFVVxZVXXkl+fn7S0aSMYCErSVJmGgksCSG8DrwI/DzG+AvgG8DVIYS3gatT1wBPAGuBNcADwJ/3fmQpc9x222088sgj3HrrrezZs4dbb72VRx55hNtuuy3paFJGcI2sJEkZKMa4Fph2nPZG4KrjtEfgi70QTeoX5s2bB8BXvvIV7rrrLvLz87njjjsOt0s6M47ISpIkSZLSioWsJEmS1M2qqqqYP3/+EcfvzJ8/n6qqqqSjSRnBQlaSJEnqZg888ABz585lzpw5FBYWMmfOHObOncsDDzyQdDQpI1jISpIkSd2stbWVO+6444i2O+64g9bW1oQSSZnFQlaSJEnqZvn5+cyfP/+Itvnz53v8jtRN3LVYkiRJ6ma33XYb99xzD9AxEjt//nzuueeeY0ZpJZ0eC1lJkiSpm3n8jtSzLGQlSZKkHjBv3jwLV6mHdGmNbAjh2hDC6hDCmhDCvcd5Pj+E8Ejq+aUhhPJOz3051b46hHBNp/a/CiGsDCGsCCHUhhAKuuMDSZIkSZIy20kL2RBCNvA94DpgKjArhDD1qG6zgZ0xxsnAt4G5qXunAjcB5wHXAn8fQsgOIZQBfwFMjzFWAtmpfpIkSZIkfaCujMheCqyJMa6NMe4HHgauP6rP9cCDqcePAleFEEKq/eEYY2uM8V1gTer1oGNa84AQQg5QCGw+s48iSZIkSeoPulLIlgEbO13Xp9qO2yfG2A40AcNOdG+McRPwTWAD8B7QFGP85fHePIRwewhhWQhhWUNDQxfiSpIkSZIyWVcK2XCcttjFPsdtDyEMpWO0diIwBigKIfzp8d48xnh/jHF6jHF6aWlpF+JKkiRJkjJZVwrZemBcp+uxHDsN+HCf1FThwcCOD7j3E8C7McaGGGMb8BPgw6fzASRJkiRJ/UtXCtmXgCkhhIkhhDw6NmV67Kg+jwG3pB5/FngmxhhT7TeldjWeCEwBXqRjSvHlIYTC1Fraq4C6M/84kiRJkqRMd9JzZGOM7SGELwFP0bG78MIY48oQwteAZTHGx4AFwI9CCGvoGIm9KXXvyhDCj4FVQDvwxRjjAWBpCOFR4JVU+6vA/d3/8SRJkiRJmeakhSxAjPEJ4Imj2v660+MW4MYT3FsD1Byn/b8D//1UwkqSJEmS1JWpxZIkSZIk9RkWspIkSVIPqK2tpbKykuzsbCorK6mtrU06kpQxujS1WJIkSVLX1dbWUl1dzYIFC5gxYwZLlixh9uzZAMyaNSvhdFL6c0RW3a61/QBL1mxnxaYmGve2cjAefeywJElSZqupqWHBggXMnDmT3NxcZs6cyYIFC6ipOWbrGEmnwRFZdbsnV2zhxXd3HL7OzQ6cN2Ywn71kLFkhJJhMkiSpd9TV1TFjxowj2mbMmEFdnSdOSt3BEVl1q3Xbm3nx3R1cMWkYf/6xs/jji8qoHDOY1zbuYnn9rqTjSZIk9YqKigqWLFlyRNuSJUuoqKhIKJGUWSxk1W3aDxzkp69uYkhhLn943kjGDi1kenkJ//6SsYwZUsAvV26l7cDBpGNKkiT1uOrqambPns3ixYtpa2tj8eLFzJ49m+rq6qSjSRnBqcXqNr95q4GGva3cckU5+TnZh9uzQuC6ytEsWPIuz7/TyEfOLk0wpSRJUs87tKFTVVUVdXV1VFRUUFNT40ZPUjexkFW32La7hWffauCCsYM5Z9TAY54/q7SYc0YO5Nm3tjF9wtAEEkqSJPWuWbNmWbhKPcSpxTpjMUZ++tom8rKz+NQFY07Y79rKUbS2HWTx6m29mE6SJElSprGQ1RnbsruF9Y37uKpiBMX5Jx7kHzmogEsmDOWFtTtY39jciwklSZIkZRILWZ2xlZt3E4DzywaftO8npo4kKwvmPbOm54NJkiRJykgWsjpjqzbvZvywQgYW5J6076CCXC4oG8JTK7bQ2n6gF9JJkiRJyjQWsjojjXtb2bK7hfNGD+ryPeeVDWJPazvPrWnswWSSJEmSMpWFrM7Iqvd2AzB1zMmnFR8yubSYgfk5PLnivZ6KJUmSJCmDWcjqjKzcvJvRgwsoKcrr8j052Vl8vGIEv1y1lbYDB3swnSRJkqRMZCGr07anpY2NO/YxdUzXpxUfcl3laHbta2Pp2h09kEySdEgIITuE8GoI4fHU9cQQwtIQwtshhEdCCHmp9vzU9ZrU8+VJ5pYyQW1tLZWVlWRnZ1NZWUltbW3SkaSMYSGr07bqvd1E4LzRXZ9WfMhHzy5lQG6204slqefdCdR1up4LfDvGOAXYCcxOtc8GdsYYJwPfTvWTdJpqa2uprq5m3rx5tLS0MG/ePKqrqy1mpW5iIavTtmrzboYV5TFyUP4p3zsgL5uZ55by1MqtHDgYeyCdJCmEMBb4JPCD1HUAPg48muryIHBD6vH1qWtSz1+V6i/pNNTU1LBgwQJmzpxJbm4uM2fOZMGCBdTU1CQdTcoIFrI6Le/vP8A7DXuZOmYQp/vfOddWjmb73lZeXr+zm9NJklL+FrgbOLQhwTBgV4yxPXVdD5SlHpcBGwFSzzel+h8jhHB7CGFZCGFZQ0NDT2WX0lpdXR0zZsw4om3GjBnU1dWd4A5Jp8JCVqdl9dbdHIyc0rE7R/v4uSPIy8lyerEk9YAQwqeAbTHGlzs3H6dr7MJzRzbGeH+McXqMcXppaekZJpUyU0VFBUuWLDmibcmSJVRUVCSUSMosFrI6Las272ZgQQ5jSwpP+zWK83P4yJRSfrFiCwedXixJ3e1K4DMhhHXAw3RMKf5bYEgIISfVZyywOfW4HhgHkHp+MOCOfNJpqq6uZvbs2SxevJi2tjYWL17M7Nmzqa6uTjqalBFyTt5FOlKMkbXbmzln5ECyznD51HWVo/hV3VZer9/FReOHdlNCSVKM8cvAlwFCCB8D/muM8T+GEP4Z+Cwdxe0twM9StzyWun4+9fwzMUb/llE6TbNmzQKgqqqKuro6KioqqKmpOdwu6cxYyOqUbd+7n337D1A+rOiMX+sTFSPJCrB4dYOFrCT1jnuAh0MIXwdeBRak2hcAPwohrKFjJPamhPJJGWPWrFkWrlIPcWqxTtn6xmYAJgw7/WnFhwwuzOX8ssE8t2b7Gb+WJOn4YozPxhg/lXq8NsZ4aYxxcozxxhhja6q9JXU9OfX82mRTS+nPc2SlnmMhq1O2vnEfA3KzGT7w1I/dOZ4PTx7Oaxt30dzafvLOkiRJacBzZKWeZSGrU7Z+RzMThhWe8frYQz581jDaD0ZeXOeeIpIkKTPU1NRw8803U1VVRUFBAVVVVdx8882eIyt1E9fI6pTsbW1n+979XDKhpNtec/qEEvKys3huzXZmnjOi215XkiQpKatWraK5uZmFCxcyY8YMlixZwq233sr69euTjiZlBEdkdUo2pNbHlnfD+thDBuRlc/GEIfxuTWO3vaYkSVKS8vLyqKqqYubMmeTm5jJz5kyqqqrIy8tLOpqUERyR1SlZ37iP7KzAmCEDzuh1Fi3dcMR1cX4uL6zdwQ/+bS2F+R1/LG++bPwZvYckSVJS9u/fz3e/+10uuuiiwyOy3/3ud9m/f3/S0aSM4IisTsm6xmbGDhlAbnb3/tGZXNpxlM8725u79XUlSZKSMHXq1OOukZ06dWrS0aSMYCGrLmtpO8DmXS3dcuzO0cqGFpKXk8U7DXu7/bUlSZJ6W3V1NYsWLTpi1+JFixZRXV2ddDQpIzi1WF32+sZdHIiRCcOKuv21s7MCE4cV8c42C1lJkpT+Zs2aBUBVVRV1dXVUVFRQU1NzuF3SmXFEVl22bP1OACaUdP+ILMBZI4ppbN7Prn2uHZEkSZJ0Yo7IqsuWrdtB6cD8w5sxdbezUutk1zY0c/EEd/STJEnpq7a2lurqahYsWHB4s6fZs2cDOCordQNHZNUlBw9GXl6/s8dGYwFGDiqgKC/bdbKSJCnt1dTUsGDBgiOO31mwYAE1NTVJR5MygoWsumRNw152t7RT3gPrYw/JCoFJpcW807CXGGOPvY8kSVJPq6urY8aMGUe0zZgxg7q6uoQSSZnFQlZd8tK6HQA9smNxZ2eVFrO7pZ3te10nK0mS0ldFRQVf/epXqaysJDs7m8rKSr761a9SUVGRdDQpI1jIqkuWb2yipCiPkqKeXbs6cXjHiO86z5OVJElpbObMmcydO5dbb72VPXv2cOuttzJ37lxmzpyZdDQpI1jIqkuWb2ri/LLBhBB69H2GF+dRnJ/Du40WspIkKX0tXryYe+65h4ULFzJw4EAWLlzIPffcw+LFi5OOJmUEdy3WSbW0HeCtrXv4RMWIHn+vEAIThxfx7vZmYow9XjhLkiT1hLq6Ol599VW+/vWvH25ra2vjb/7mbxJMJWWOLo3IhhCuDSGsDiGsCSHce5zn80MIj6SeXxpCKO/03JdT7atDCNd0ah8SQng0hPBmCKEuhHBFd3wgdb9V7+3mwMHI+WWDe+X9yocX0fR+G/U73++V95MkSepuFRUVLFmy5Ii2JUuWuEZW6iYnLWRDCNnA94DrgKnArBDC1KO6zQZ2xhgnA98G5qbunQrcBJwHXAv8fer1AL4D/CLGeC4wDXALtz7qjfomAC4YO6RX3u/QOtml7+7olfeTJEnqbtXV1cyePZvFixfT1tbG4sWLmT17NtXV1UlHkzJCV6YWXwqsiTGuBQghPAxcD6zq1Od64H+kHj8KfDd0zAm9Hng4xtgKvBtCWANcGkJYCXwE+DxAjHE/4Da1fdTy+iZKB+YzclB+r7zfiIH5DMjNZunaRj57ydheeU9JkqTuNGvWLACqqqqoq6ujoqKCmpqaw+2SzkxXCtkyYGOn63rgshP1iTG2hxCagGGp9heOurcMeB9oAP4hhDANeBm4M8Z4zA4/IYTbgdsBxo8f34W46m5vbNrFBb2w0dMhWal1si+uc0RWkiSlr1mzZlm4Sj2kK2tkj1e9xC72OVF7DnAx8P0Y40VAM3DM2luAGOP9McbpMcbppaWlXYir7tTc2s6abXs5f2zvrI89pHx4Eesb97GlqaVX31eSJElS39eVQrYeGNfpeiyw+UR9Qgg5wGBgxwfcWw/UxxiXptofpaOwVR+z6r3dHIxwQS8Xsr9fJ9vYq+8rSZIkqe/rSiH7EjAlhDAxhJBHx+ZNjx3V5zHgltTjzwLPxBhjqv2m1K7GE4EpwIsxxi3AxhDCOal7ruLINbfqI5anNnqq7KUdiw8ZPbiAgfk5bvgkSZIk6RgnLWRjjO3Al4Cn6NhZ+McxxpUhhK+FED6T6rYAGJbazGkOqWnCMcaVwI/pKFJ/AXwxxnggdU8V8E8hhOXAhcB93fex1F2W1+9i9OACRgws6NX3zQqB6eVDWbrWEVlJkpSeamtrqaysJDs7m8rKSmpra5OOJGWMrmz2RIzxCeCJo9r+utPjFuDGE9xbA9Qcp/01YPqphFXve6O+qdfOjz3apROHsXh1A9v3tjK8uHd2TJYkSeoOtbW13HnnnRQVFRFjpLm5mTvvvBPADaCkbtCVqcXqp3a3tLF2e3Ovr4895LJJJQC86PRiSZKUZu6++26ys7NZuHAhra2tLFy4kOzsbO6+++6ko0kZwUJWJ7RiU8f62PPHDknk/c8vG8yA3GwLWUmSlHbq6+t56KGHmDlzJrm5ucycOZOHHnqI+vr6pKNJGcFCVif0Rmqjp6SmFudmZ3HJhKG84DpZSZKUhp555pkj1sg+88wzSUeSMkaX1siqf1q+qYmxQwdQUpSXWIbLJ5XwzV++xc7m/QxNMIckSdKpKCkpYe7cuWRlZXHw4EHefPNNVq1aRUlJSdLRpIzgiKxO6I36psTWxx5y+aRhAB7DI0mS0kpraysxRjpOpOTw49bW1oSTSZnBQlbHtWvffjbs2Mf5Zcmsjz3kgrFDKMjNcnqxJElKK83NzRQVFTF+/HhCCIwfP56ioiKam5uTjiZlBAtZHdeKTbuB5NbHHpKXk8X0CSUWspIkKe18+tOfpqioiBACRUVFfPrTn046kpQxLGR1XG+kdiyuLBuUcJKOdbJvbtnDzub9SUeRpLQSQigIIbwYQng9hLAyhPDVVPvEEMLSEMLbIYRHQgh5qfb81PWa1PPlSeaX0t0jjzzCrbfeyp49e7j11lt55JFHko4kZQwLWR3Xik1NjCsZwJDC5DdYcp2sJJ22VuDjMcZpwIXAtSGEy4G5wLdjjFOAncDsVP/ZwM4Y42Tg26l+kk5DVlYWMUbuvvtuioqKuPvuu4kxkpXlf35L3cF/knRcb2xqSnxa8SGuk5Wk0xM77E1d5qZ+IvBx4NFU+4PADanH16euST1/VQgh9FJcKaMcPHgQgAMHDhzx+1C7pDNjIatjNO1rY8OOfVT2kULWdbKSdPpCCNkhhNeAbcDTwDvArhhje6pLPVCWelwGbARIPd8EDOvdxFJmyMnJobi4mPLyckIIlJeXU1xcTE6Op19K3cFCVsdYuTm1PnZM3yhkwXWyknS6YowHYowXAmOBS4GK43VL/T7e6Gs8uiGEcHsIYVkIYVlDQ0P3hZUySHt7O0VFRSxcuJDW1lYWLlxIUVER7e3tJ79Z0kn5V0I6xqGNnpKeWrxo6YbDj/e0dPxL/5u/XM15qQL75svGJ5JLktJRjHFXCOFZ4HJgSAghJzXqOhbYnOpWD4wD6kMIOcBg4JgNCmKM9wP3A0yfPv2YQldSh8suu4zrrruO1tZW8vPzueaaa3jssceSjiVlBEdkdYw3NjVRNmQAQ4uS3+jpkLKhA8jNDqzd7tlrktRVIYTSEMKQ1OMBwCeAOmAx8NlUt1uAn6UeP5a6JvX8MzFGC1XpNJSUlPD4449z33330dzczH333cfjjz9OSUlJ0tGkjOCIrI6xog9t9HRITlYWE0qKeLfBQlaSTsFo4MEQQjYdf3n94xjj4yGEVcDDIYSvA68CC1L9FwA/CiGsoWMk9qYkQkuZoLCwkJaWFu69917uuusucnNzKSgooLCwMOloUkawkNURdre0sa5xH5+9ZGzSUY4xsbSIp1dtZV9rO4X5/tGVpJOJMS4HLjpO+1o61sse3d4C3NgL0aSMt2nTJoYNG0ZxcTEbNmygrKyMvXv3smnTpqSjSRnBqcU6wspNuwH6zI7FnU0aXgTAu42OykqSpL4tLy+Pa665hqKijv9+KSoq4pprriEvr+8s3ZLSmYWsjrCij2z0dDyH1sm+4/RiSZLUx7W2tlJbW0tjY8fxgY2NjdTW1tLa2ppwMikzWMjqCG9samLM4AKGFecnHeUYOVlZlA8r4p1te5OOIkmS9IFycnIoLCykoKCAGOPh9bGeIyt1DwtZHWHF5ibO64OjsYdMHlFMw95Wmt5vSzqKJEnSCXmOrNSzLGR12N7Wdt7d3twnpxUfMnlEMQBrHJWVJEl93Be+8AWqqqooKCigqqqKL3zhC0lHkjKGcxt02MpNTcTYN9fHHjJyUAFF+Tms2bYn6SiSJKkfCSGc8j3f+MY3Dj9euXIlK1euPO3X8khn6UiOyOqwN1IbPfXFHYsPyQqByaVFrGlo9l/okiSp18QYT+ln0aJFlJaWUl5eDgTKy8spLS1l0aJFp/xa/jePdCwLWR22cvNuRg7Kp3Rg39voqbPJI4ppbm3nzS2OykqSpL5p1qxZfOc73+k4ficEioqK+M53vsOsWbOSjiZlBAtZHfbGpqY+Pa34kMkjBgKw5O3tCSeRJEk6sVmzZrFixQom3P0YK1assIiVupGFrICOjZ7eadjbp6cVHzJ4QC6lxfksWWMhK0mSJPVHFrIC4I36jo2epo0bknSULpk8opil7zbS2n4g6SiSJEmSepmFrABYXr8LgGlj06eQbWk7yMvrdyYdRZIkSVIvs5AVAMvrmxg7dAAlRXlJR+mSicOLyM4K/M7pxZIkSVK/YyErAF6v35U2o7EABbnZXDRuiBs+SZKFURg8AAAgAElEQVQkSf2Qhaxo3NtK/c73mTau72/01NmVk4ezfFMTu/btTzqKJEmSpF5kISuW1zcBcEEajcgCfOTs4cSIuxdLkiRJ/YyFrHi9fhchkBZH73Q2bewQBg/I5dnVDUlHkSRJktSLLGTF8vomJpcWU5yfk3SUU5KTncVHzi7l2dUNHDwYk44jSZIkqZdYyPZzMUaW1+9Km/NjjzbznFK2721lxeampKNIkiRJ6iUWsv3c5qYWtu/dz7Sx6TWt+JCPnl1KCLD4TacXS5IkSf2FhWw/9/rGXUD6bfR0yLDifKaNHcLi1duSjiJJkiSpl1jI9nOv1+8iNztw7uiBSUc5bTPPGcHr9bto3NuadBRJkiRJvcBCtp9bvrGJitGDyM/JTjrKafv4uSOIEX7zltOLJUmSpP4gvbapVbc6eDCyYlMTN1xUBsCipRsSTnR6zhsziOHF+Sxe3cAfXzw26TiSJEmSepgjsv3Y2u3N7Glt54I03ejpkKyswMfOKeU3q7fRfuBg0nEkSZIk9bAuFbIhhGtDCKtDCGtCCPce5/n8EMIjqeeXhhDKOz335VT76hDCNUfdlx1CeDWE8PiZfhCdukMbPaXr0TudffzcEexuaefV1GeSJEmSlLlOWsiGELKB7wHXAVOBWSGEqUd1mw3sjDFOBr4NzE3dOxW4CTgPuBb4+9TrHXInUHemH0KnZ3n9LgrzsjmrtDjpKGdsxpThZGcFFr/p7sWSJElSpuvKiOylwJoY49oY437gYeD6o/pcDzyYevwocFUIIaTaH44xtsYY3wXWpF6PEMJY4JPAD878Y+h0vLpxFxeMHUx2Vkg6yhkbVJDL9AlDecZCVpIkScp4XSlky4CNna7rU23H7RNjbAeagGEnufdvgbuBD1zUGEK4PYSwLISwrKHBXWm7y7797azcvJvpE0qSjtJtPn7uCN7csodNu95POookSZKkHtSVQvZ4w3Wxi32O2x5C+BSwLcb48snePMZ4f4xxeoxxemlp6cnTqkte39jEgYORSyYMTTpKt/nE1JEA/GrV1oSTSJIkSepJXSlk64Fxna7HAptP1CeEkAMMBnZ8wL1XAp8JIayjY6ryx0MI/3ga+XWaXtmwE4CLx2dOIXtWaTGTSot42kJWkiRJymhdKWRfAqaEECaGEPLo2LzpsaP6PAbcknr8WeCZGGNMtd+U2tV4IjAFeDHG+OUY49gYY3nq9Z6JMf5pN3weddGydTuYMqKYwYW5SUfpVldPHckLaxtper8t6SiSJEmSekjOyTrEGNtDCF8CngKygYUxxpUhhK8By2KMjwELgB+FENbQMRJ7U+relSGEHwOrgHbgizHGAz30WdRFBw9GXtmwi+sqRyUdpdv94dSR/J/frOXZ1du4/sIyFi3dcNJ7br5sfC8kkyRJktRdTlrIAsQYnwCeOKrtrzs9bgFuPMG9NUDNB7z2s8CzXcmh7rF2+16a3m/j4gxaH3vIheOGMrw4n6dXbeX6C4/ek0yS+o8QwjjgIWAUHRsr3h9j/E4IoQR4BCgH1gH/Ica4M3XawHeAPwL2AZ+PMb6SRHZJkk6mK1OLlWFeXt+xPnZ6Bhay2VmBT1SM4DerG9jf/oEbYktSpmsH7ooxVgCXA19Mne9+L/DrGOMU4Nepa+g4L35K6ud24Pu9H1mSpK6xkO2Hlq3bydDCXCYOL0o6So+4eupI9rS288LaxqSjSFJiYozvHRpRjTHuAeroOAKv89nvDwI3pB5fDzwUO7wADAkhjO7l2JIkdYmFbD/08oadXDJhKB2zyDLPlZOHMyA3m1+u2pJ0FEnqE0II5cBFwFJgZIzxPegodoERqW5dOTdekqQ+wUK2n9nRvJ+1Dc1cMqEk6Sg9piA3m4+cPZxfrdpGx+bZktR/hRCKgf8L/GWMcfcHdT1O23H/JRpCuD2EsCyEsKyhoaE7YkqSdEosZPuZV1LrYy/JwPWxnV09dRRbdrewadf7SUeRpMSEEHLpKGL/Kcb4k1Tz1kNThlO/t6Xau3JuPAAxxvtjjNNjjNNLS0t7JrwkSR/AQrafeXnDTnKyAheMHZx0lB718XNHkBWg7r0PGnyQpMyV2oV4AVAXY/xWp6c6n/1+C/CzTu2fCx0uB5oOTUGWJKmvsZDtZ15ev5PzygZTkJuddJQeVVKUx/TyElZZyErqv64E/gz4eAjhtdTPHwHfAK4OIbwNXJ26ho5j9tYCa4AHgD9PILMkSV3SpXNklRn2tx/k9Y27+NPLJyQdpVdce94ovvb4KrbvaWX4wPyk40hSr4oxLuH4614BrjpO/wh8sUdDSZLUTSxk+5FvP/0Wre0H2bf/AIuWbkg6To+7trKjkF25uYmPnjPi5DdIkiRJSgtOLe5H1jU2AzChpDDhJL1jzJABjB06gBWbnV4sSZIkZRIL2X5kbUMzw4vzGDQgN+kovea8MYPZtOt9du3bn3QUSZIkSd3EqcX9RPuBg6xrbGba2CFJR+k2XZkeXTlmEE+t3MLKzbu5cvLwXkglSZIkqac5IttPrNi8m9b2g0wqLUo6Sq8aVpzPqEEFrNjclHQUSZIkSd3EQrafeP6dRgAmDu9fhSzAeWMGsaFxH3ta2pKOIkmSJKkbWMj2E8+9s50RA/MZWNB/1scecl7ZYCJ4pqwkSZKUIVwj2w/sbz/IsnU7mTYuc9bHnoqRA/MZXpzHys27uWzisKTjSJKkhEz76i9pej+ZGVrl9/48kfcdPCCX1//7Hyby3lJPspDtB5bX7+L9tgNM6ofTigFCCJw3ZjC/fbuBffvbKczzj70kSf1R0/ttrPvGJ5OO0auSKqClnubU4n7g+XcaCYF+W8hCxzrZgxHq3tuTdBRJkiRJZ8hCth94fm0j544aRGF+/x2JLBsygKGFuazY5O7FkiRJUrqzkM1wLW0HeHn9Tq6Y1L/XhoYQOL9sMG9v28O+/e1Jx5EkSZJ0BixkM9yrG3bR2n6QK87q34UswPllQzgYYdVmdy+WJEmS0pmFbIZ7fm0jWQEunViSdJTEjRlSQElRHm84vViSJElKaxayGe6FdxqpLBvM4AH97/zYox2aXvxOw172tjq9WJIkSUpXFrIZ7P39B3h1o+tjO7tg7GCnF0uSJElpzkI2g72wtpG2A5ErJw9POkqfMWpQAcOL81i+aVfSUSRJkiSdJgvZDPbs6m0MyM12fWwnHdOLh/BuQzN7WtqSjiNJkiTpNFjIZqgYI4tXN3DFWcMoyM1OOk6fcv7YwURgpdOLJUmSpLRkIZuh3t3ezIYd+/jYOaVJR+lzRg0qYMTAfJbXu3uxJEmSlI4sZDPUs6sbAPjY2SMSTtI3nV82mPWNzex+3+nFkiRJUrrJSTqAuseipRuOuK59cQPDi/NZsmY7rEkoVB82bewQfv3mNl6vd9MnSZIkKd04IpuB9rcf5N3tzZwzsjjpKH3W8IH5jBs6gFc3WMhKkiRJ6cZCNgOt3b6X9oORs0cNTDpKn3bh+KFs2d3imbKSJElSmrGQzUBvbd1DbnZg4rCipKP0aReUDSY7BH76an3SUSRJkiSdAtfIZpgYI29t3ctZpcXkZPv3FB+kKD+Hc0YN5F9e28w9157r/16SJGW4gRX3cv6D9yYdo1cNrAD4ZNIxpG5nIZthGvfuZ0fzfmZMHp50lLRw4bghLHpxA0vWbOdj57jDsyRJmWxP3TdY943+VdSV3/vzpCNIPcIhqAyzeuseAM4e6frYrjh31EAGD8jlp69uSjqKJEmSpC6ykM0wb23dQ2lxPiVFeUlHSQs52Vl8etponlq5hT0tnikrSZIkpQML2Qzy/v4DrG1o5hx3Kz4l/+6isbS0HeTJFVuSjiJJkiSpCyxkM8jqrbs5ECOVYwYlHSWtXDx+COXDCvnJK+5eLEmSJKUDC9kMsmLTbgYV5DC2pDDpKGklhMBnLxnLC2t3sLZhb9JxJEmSJJ2EhWyG2N9+kLe37WHqmMFkhZB0nLTzJx8aT2524EcvrE86iiRJkqST6FIhG0K4NoSwOoSwJoRwzOFbIYT8EMIjqeeXhhDKOz335VT76hDCNam2cSGExSGEuhDCyhDCnd31gfqrt7buoe1A5DynFZ+W0oH5XFc5mkdfrmff/vak40hStwghLAwhbAshrOjUVhJCeDqE8Hbq99BUewgh/F3qO3t5COHi5JJLkvTBTlrIhhCyge8B1wFTgVkhhKlHdZsN7IwxTga+DcxN3TsVuAk4D7gW+PvU67UDd8UYK4DLgS8e5zV1ClZsbqIwL5vyYUVJR0lbn7tiAnta2vmXVzcnHUWSussP6fj+7exe4NcxxinAr1PX0PE9PyX1czvw/V7KKEnSKevKiOylwJoY49oY437gYeD6o/pcDzyYevwocFUIIaTaH44xtsYY3wXWAJfGGN+LMb4CEGPcA9QBZWf+cfqn1vYDrN6yh6mjB5Gd5bTi03XJhKFUjB7EQ8+vI8aYdBxJOmMxxn8DdhzV3Pk7+0Hghk7tD8UOLwBDQgijeyepJEmnJqcLfcqAjZ2u64HLTtQnxtgeQmgChqXaXzjq3iMK1tQ05IuApaeQW538bs12WtsPct6YwUlHSWshBD53xQS+/JM3WLZ+Jx8qL0k6kiT1hJExxvcAYozvhRBGpNqP931fBrx39AuEEG6nY9SW8ePH92xaqZuV3/vzpCP0qsEDcpOOIPWIrhSyxxviO3q46kR9PvDeEEIx8H+Bv4wx7j7um/tleVJPvrGF/JwszhrhtOIzdf2FY/ibJ+p46Pn1FrKS+puufN93NMZ4P3A/wPTp053CorSx7hufTOR9y+/9eWLvLWWqrkwtrgfGdboeCxy9iPBwnxBCDjCYjqlMJ7w3hJBLRxH7TzHGn5zozWOM98cYp8cYp5eWlnYhbv/SfuAgT9dtpWL0IHKy3IT6TBXm5XDj9HE8+cZ7bNvdknQcSeoJWw9NGU793pZq78r3vSRJfUJXKp+XgCkhhIkhhDw6Nm967Kg+jwG3pB5/FngmdiwyfAy4KbWr8UQ6NpB4MbV+dgFQF2P8Vnd8kP5q6bs72LWvzd2Ku9GfXj6B9oORRS9uSDqKJPWEzt/ZtwA/69T+udTuxZcDTYemIEuS1NectJCNMbYDXwKeomNTph/HGFeGEL4WQvhMqtsCYFgIYQ0wh9QOiDHGlcCPgVXAL4AvxhgPAFcCfwZ8PITwWurnj7r5s/ULP3/jPQbkZjNlxMCko2SMicOLuOrcEfzD79axu6Ut6TiSdNpCCLXA88A5IYT6EMJs4BvA1SGEt4GrU9cATwBr6diY8QHgzxOILElSl3RljSwxxifo+ILr3PbXnR63ADee4N4aoOaotiUcfy2OTkFL2wEef30z15w3krwcpxV3p7+6+mw+NW8JC5e8y19+4uyk40jSaYkxzjrBU1cdp28EvtiziSRJ6h5WP2nsmTe3sbulnT++eGzSUTJOZdlgrj1vFAt++y47m/cnHUeSJElSJxayaewnr9QzclA+V04ennSUjPRXV5/N3v3t3P/btUlHkSRJktSJhWya2r63lWdXN3DDRWVkZzlLuyecM2ogn5k2hh/+bh0Ne1qTjiNJkiQpxUI2Tf3r65tpPxj544ucVtyT7rxqCq3tB5j/m3eSjiJJkiQpxUI2Tf3klU1Ulg3inFHuVtyTJpUW8+8vHsuPXljPe03vJx1HkiRJEhayaemtrXt4Y1OTo7G95C+umgLAf/uXlXRs6ilJkiQpSRayaegnr2wiOyvwmQvHJB2lXxhXUsjd15zDr+q28ujL9UnHkSRJkvo9C9k0c+Bg5F9e3cTHzi5leHF+0nH6jVuvnMhlE0v42r+uYtMupxhLkiRJSbKQTTNL1mxny+4W/t3FZUlH6VeysgLfvHEaB2Pk//vn1zl40CnGkiRJUlIsZNPMQ8+tY3hxHldPHZl0lH5nXEkh/+1TU3nunUYeen5d0nEkSZKkfstCNo2sb2zmmdXbuPnS8eTnZCcdp1/6kw+N4+PnjuBvnnyTl9btSDqOJEmS1C9ZyKaRh55fT3YI/MfLJyQdpd8KIfA/P3sBZUMG8IV/eInXNu5KOpIkSZLU7+QkHUBd09zazo+XbeS680czclBB0nEyyqKlG07a5+bLxh9+PLw4n0W3Xc5/+D/P87kFS1l02+VUlg3uyYiSJEmSOnFENk389NVN7Glp5/MfdjS2Lxg1uIBFt13GwIJc/nTBUt7csjvpSJIkSVK/YSGbBmKMPPjcOirLBnHx+KFJx1HK2KGFLLrtMvJzsrjx+8/zs9c2JR1JkiRJ6hecWpwGnnunkbe37eWbN04jhJB0HHUyYVgRj97xYf7qkde48+HXWPzmNr52QyWDCnKP6XuqU5glSZIkHZ8jsmngh8+to6Qoj09dMDrpKDqOcSWFPHz75cy5+mz+dfl7XPe3v+XZ1duI0bNmJUmSpJ5gIdvHrdm2h1/VbeXmS8dTkOuRO31VTnYWf3HVFP75jivIzQ58/h9e4qb7X+Dl9TuTjiZJkiRlHAvZPu5vf/U2hbnZ3DpjYtJR1AUXjx/KU3/1Eb76mfN4p6GZf//95/hPD77EqxssaCVJkqTu4hrZPuzNLbv5+Rvv8ecfO4uSoryk46iL8nOyueXD5dw4fSz/8Lt1/J/fvMOv6p6jfFgRH5kynLNHDSTLtc6SJEnSabOQ7cO+86u3Kc7L4bY/mJR0FJ2GwrwcvjhzMrd8uJxHXtrI3/36bR56YT0jBubzB1NKmTZuMDlZToqQJEmSTpWFbB+1cnMTT67Ywp1XTWFIoaOxfd3JdiQekJvNf/3Dc3hj0y7+7a3t/N9X6nl61RaunDycD5WXuP5ZkiRJOgUWsn3Ut59+m0EFOa6NzSDZWYELxw1l2tghvL1tL//2dgNPrtjC4tXb+PBZw/nwWcOSjihJkiSlBQvZPmh5/S5+VbeVu64+m8EDjj2PVOkthMDZIwdy9siB1O/cx7OrG3jmzW0sWbOdhr2t3PYHkxhenJ90TEmSJKnPspDtY2KM/M9frGZIYS6fv7I86ThKOdnU4dM1dmghf3r5BLbsbuHZ1dt44N/W8uBz6/iPl03gP39kEiMGFfTI+0qSJEnpzJ1m+pifvbaZJWu285dXTWFggaOx/cWoQQXc9KHx/GrOR/nk+WP44XPr+IP/uZj/8dhKNu96P+l4kiRJUp9iIduH7Nq3n///8VVMGzeEP7uiPOk4SsCk0mL+93+YxjN3fZQbLizjH19Yz0f/12LueXQ5725vTjqeJEmS1Cc4tbgPue+JOna938aP/t35ZGf9/pzRnprWqr6n8//X08YNYfywQn77dscuxz9etpHzygbz9RsquXDckARTSpIkScmykO0jXljbyI+X1fOfPzqJqWMGJR1HfcTQwjw+M20MM88p5XdrGln6biM3fO93fKh8KLNnTOLqqSOP+EsPSZIkqT+wkO0DWtsP8JWfvsG4kgH85VVnJx1HfdDAglyurRzFx84pJQILl7zLHf/4MuNLCvmzyydw4/SxnjcsSZKkfsNCNmGLlm7g58s3s7ahmc9/uJyfvrop6Ujqwwpys7n5svHccsUEnlq5lQefX0fNE3X876dXc8OFZfzZFRN4fWPTSV/n5svG93xYSZIkqYdYyCbs1Q07+d07jVw+aRhnjxyYdByliZzsLD55wWg+ecFoVm3ezY9eWMdPX93Ewy9tZHxJIZdNLKGybDC52e7nJkmSpMxjIZug1zfu4qevbmLS8CI+ef7opOMoTU0dM4i/+eMLuPfaCh59pZ6/X7yGf365np+/8R7TJ5Rw2cQShhY57ViSpDMRwpntSRHmntn7xxjP7AWkDGMhm5Bte1r4zz96meKCHGZdOt4Ne9RlH7SL9YDcbP7q6rNZ29DMC2sb+e3bDfz27QbOGTWQyycNY/KIYrLO8ItYUmYLIVwLfAfIBn4QY/xGwpGkPsFCUupbLGQT0NJ2gP/yj6/Q9H4b/+kPJlKU7/8N6j5ZITB5RDGTRxSza99+Xlq3g5fW7eSHz62jpCiPS8tLuLZyFCWO0ko6SgghG/gecDVQD7wUQngsxrgq2WSSJB3JCqqXNb3fxn968CVe2bCT7866mKb325KOpAw2pDCPq6eOYua5I1i1eTcvrN3BL1Zu4ZnV27iuchR/Mn0cl08aRpYzAiR1uBRYE2NcCxBCeBi4HrCQlST1KRayvWjb7hY+t/BF3mnYy9/ddBGfvGD0B04TlbpLTlYWF4wdwgVjh7Bldwu79u3np69u4mevbWZcyQBuvGQcn5k2hvLhRUlH7fO68s+su0IrjZUBGztd1wOXHd0phHA7cDvA+PH+eZck9T4L2V7y7vZm/mzBUnY072fh5z/EH0wpTTqS+qlRgwqYc/XZfOWPKnhq5RYeeWkj33r6Lb719FtUjB7EJ88fxTXnjWLyiOIz3thCUto53j/0xywMjDHeD9wPMH36dBcOSpJ6nYVsD4sx8ujL9Xz953VkZwVqb7ucaeOGJB1LoiA3m+svLOP6C8vYtOt9frFiC0+88R7f/OVbfPOXb1E6MJ/LJw3j8kklXDx+KJNKi8jPyU46tqSeVQ+M63Q9FticUBZJkk7IQrYHbWjcx5d/upzfrWnkQ+VD+V+fnebUTfVJZUMGMHvGRGbPmMiWphYWr97GC2sbef6dRv719Y7/hs3OCkwcXsTZI4sZN7SQkYMKGD24gBGDCigpymPIgFwGDcg9Zgfuk03FTddpuPvbD9Lc2s6Bg5EIRCJZBAbkZfOPL6w/6e7QfelzO11anbwETAkhTAQ2ATcBNycbSZKkY1nI9oCtu1v4xxfW88Bv15KTlcXXb6jk5kvHu6GO0sKowQXMunQ8sy4dT4yRd7c3s2Lzbt7asofVW/ewavNufrVqG/sPHDzm3hBgYH4OgwbkMqggl0EDctjT0s6A3GwK83IozMumOD+H4oKcw78PHIx98vippn1tvLN9L+u2N7O+cR8bdnT8vLNtL83722k7cOLZlAEozM9hUEEOw4rzKS3OY3hxPiMHFTBiYD452Vm990GkUxBjbA8hfAl4io7jdxbGGFcmHEuSpGN0qZA92ZlyIYR84CHgEqAR+JMY47rUc18GZgMHgL+IMT7VlddMNzFGXtmwkx8+t54n33iPAzFy7Xmj+OtPT2X04AFJx5NOSwiBSaXFTCothmm/b48xsnNfG+81vc+23a3sen8/u/a1sXNfG03/r717DZK0qu84/v1N99xndnaZWZaFXXa56bqKCBpE1xjAGE2MolUYWbeQsqgyL4xAJQUxeRFJKlapLzRBLCuKeKFQRNCSUgSViwYNLCCEVbm4IMa9sTt7mdvOdE/3/PPiObO0y1yadefyNL9PVVc/feZ5Tp9/T09P/885z3MOlhkaqzA4Ns7A6Dh7hkqMlqscLFepTrEG36fueILl3ZNJXhvHLmllRXcbK5a0sry7ld6uVno7s0SwveXoTW0eG6+ye7DE7qExtu0frUlWR3hmzwh7R8o1rwMc39PO6mPaOamvk87WYnZrKRxKSkU2KnuwXGWkVGG4VGVwdJwdB0b51faBQycZNgmO7W7jod/tY/3KJaw7bgmvWNlNb1frUYutXtWJYLhUYXB0nPHqBOPVYKLmdyRlFwprLogDB8t0tBRpKToJb3QRcTtw+0K3w8zMbCazJrJ1ril3KbA/Ik6VdBHwSeB9ktaTTUt6JXA88GNJL0vH5Hqduohgx8AY9z+9l59t7ee+rf3sHirR3Vrkkjeu5eJz1ngasS1ac3G17NZigeOWFDhuSduUP48ISmk67nCpwtDY5P04g2NZMvXL7QMMPT3OSLk6ZR0txSZ62pvpaW+muy0b4W1vLtDWXGD7/lEk0aQsAZ+IoDrx/G2sUqWzpcjQWEq4p1j66rglbZzY28Fb16/g5OWdnNzXxdq+TlYf037o/OAjee0q1Qn2jpR5bnCMnQNj7BwY5b7f9PPtX2w/tM/y7lZO6utkbW9H9pzLOujraqWvq4XerlY6Wwu0FJpecAGuiYmgXJ1gcGw8e03HKgyMjrP/YJn9I2V++pt+DpYrjJSqHCxXOFiuMlquMjpepVR54aj6dD5155MAdLUWWdbZzDEdWbuyDoes42FFdxvH9WSdEX2drZ6FYmZmZnOmnhHZetaUuwC4Om3fAlyr7NvWBcBNEVECfitpa6qPOuqcFxFBJX3RnRyRGB2vMpq+8A2PVegfKbNvuMTekTK/33eQp/eM8Mye4UNftns7W3jjqX386Wl9vOP0lXS2esa22eEk0ZaSztlGHysTEwynRHd4rMKrTuihf6TEwOg4g6NZEjo0VmG0XGX/yDij41UOHCwzEdnfdEQ2mlhoEoUmUWxqorXYRF9XC2v7Olna3syKJa0cm6b6rlrWzqplHbQ1z83FrIqFJlYsaWPFkjZevSore//rT2TvcIkndg3x+M5Bntw1xLN7R7j7iT30D2+btq6WQjZCWo2gUs0+v2bT3lygszWb3t3T3szKnjbamrNOgNbmAs0F0VxoornQRKEm95wIqEwE45UJXrWqh5FSJY28l9k3UmbXwBiPbRtg70iJwwfbi02irysbVV/enY2qL+tsoae9mWUdLXSn6eWdralDoqVAa7GJ1mKBlmITxSZRLGS/u8nOCTMzM7NJ9WRc9awpd2ifdH7NANCbyu8/7NgT0vas69TNhQ9+eTM/27qXamRT6KaY6TgtCVYuaeOUY7t47+tWc8ryTl675hjWHdftkQezo6jY1MTSjhaWdrQA8Dd/snqWI/J5waLerlY2nNrKhlP7/qB8uFRhx4FR+odL7B0u0z9c4mA5G0EtVyYYr05QaNKhBLSl2ER3a5Hutmykuqe9maUdLSzraOb2LbuOyjnIM71249UJ+odLPDdYYtfAGM8NjrFrcIz+oRJ7hks8NzjGr3cMcmC0zNh4/aPAtSRoSiPu5687lidsqI0AAAqQSURBVP+6+HVHGoqZmZk1gHoS2XrWlJtun+nKpzrJasqUsnbRdWBY0pPTtHNePAv8zx9XRR/QfxSastAaJQ5onFgaJQ44LJZNR6nSo1XPizDt72QB2vJH2bSI3l9bgS984IgPr41jzdFoz0vdww8/3C/pdwvdDrNFbtF8hpotcnX/b64nka1nTbnJfbZJKgI9wL5Zjq1rnbraRdcbgaSHIiL3QwmNEgc0TiyNEgc0TiyNEgc0TiyNEsdiEhHLF7oNZoudP3vMjr56Lj95aE05SS1kF2+67bB9bgMuSdsXAndHRKTyiyS1pjXpTgM211mnmZmZmZmZ2QvMOiI73Zpykv4NeCgibgO+BNyQLua0jywxJe13M9lFnCrAhyOiCuB16szMzMzMzOxI1HV53anWlIuIf6nZHgPeO82xHwc+Xk+dLxGNMk26UeKAxomlUeKAxomlUeKAxomlUeIws3zxZ4/ZUaZ4MZftNTMzMzMzM1tg9Zwja2ZmZmZmZrZoOJE1MzMzMzOzXHEiO08kvV3Sk5K2SvroQrdnNpKul7Rb0i9ryo6R9CNJv0n3y1K5JF2TYntM0lkL1/I/JGm1pHskPS7pV5IuT+V5jKVN0mZJ/5ti+ddUfpKkB1Is30xXAiddLfybKZYHJK1dyPYfTlJB0iOSvpce5y4OSc9K2iLpUUkPpbLcvbcAJC2VdIukJ9LfyxvyFoukl6ffxeRtUNIVeYvDzOaOpOF5eI73SApJ6+b6uWZpxxWSOhayDWZzyYnsPJBUAD4H/CWwHtgoaf3CtmpWXwHefljZR4G7IuI04K70GLK4Tku3DwGfn6c21qMC/ENEvAI4B/hweu3zGEsJOD8izgBeA7xd0jnAJ4HPpFj2A5em/S8F9kfEqcBn0n6LyeXA4zWP8xrHeRHxmpr1AfP43gL4T+COiFgHnEH2u8lVLBHxZPpdvAZ4LXAQ+A45i8PMcm8jcB9pFY8FdAXgRNYalhPZ+XE2sDUinomIMnATcMECt2lGEfFTsqWUal0AfDVtfxV4d0351yJzP7BU0sr5aenMImJnRPwibQ+RfTk/gXzGEhEx2ZPcnG4BnA/cksoPj2UyxluAt0jSPDV3RpJWAe8ArkuPRQ7jmEbu3luSlgBvJltKjYgoR8QBchhLjbcAT0fE78h3HGY2xyStkXRXmplxl6QTU/k700ygRyT9WNKKVH61splr90p6RtJlNXV1ARvIOmEvqik/V9JPJN0s6SlJn5C0SdlMqy2STpmlLV+RdGFNfcM19d5bM6PmxjTb5DLgeOAeSffM+YtotgCcyM6PE4Df1zzelsryZkVE7IQsQQSOTeW5iC9NST0TeICcxqJsOu6jwG7gR8DTwIGIqKRdatt7KJb08wGgd35bPK3/AK4CJtLjXvIZRwA/lPSwpA+lsjy+t04G9gBfTl/YrpPUST5jmXQR8I20nec4zGzuXUvWqfVq4EbgmlR+H3BORJxJNghxVc0x64C3kQ1WfExScyp/N9nslqeAfYedsnAG2Wyk04GLgZdFxNlknbofmaUtMzmTbPR1Pdnn+YaIuAbYQTZr6Lz6XgazfHEiOz+mGj1qpHWPFn18qYf0VuCKiBicadcpyhZNLBFRTdMmV5H983zFVLul+0UZi6S/BnZHxMO1xVPsuqjjSDZExFlkU1Q/LOnNM+y7mOMoAmcBn09f2EZ4fvrtVBZzLCg7v/pdwLdm23WKskUTh5nNmzcAX0/bNwBvSturgDslbQGuBF5Zc8z3I6IUEf1kncsrUvlGsqSXdL+x5pgH00yxEllH9A9T+RZg7SxtmcnmiNgWERPAozV1mTU0J7LzYxuwuubxKrJesrx5bnLaXbrfncoXdXypl/RW4MaI+HYqzmUsk9K0z3vJzvtdKqmYflTb3kOxpJ/38MLp4gthA/AuSc+S/ZM/n2yENm9xEBE70v1usnMxzyaf761twLaIeCA9voUssc1jLJB1LPwiIp5Lj/Mah5ktjMkOrc8C10bE6cDfAm01+5RqtqtAUVIv2f+069L/uCuB99WcDlN7zETN4wmyDsWZ2lIhfW9P9bXM1JaZgjNrFE5k58eDwGnKrsraQjbl7bYFbtORuA24JG1fAny3pvwD6ZyMc4CByWl8Cy192H8JeDwiPl3zozzGslzS0rTdDvw52Tm/9wCT580cHstkjBcCd0fEgo82RcQ/RcSqiFhL9rdwd0RsImdxSOqU1D25DfwF8Ety+N6KiF3A7yW9PBW9Bfg1OYwl2cjz04ohv3GY2fz4Oc+fz7qJbEoxZB2n29P2JYcfNIULyaYFr4mItRGxGvgt9Y2qztaWZ8kuYgfZ+f3NzG4I6H4Rz22WK+6xmQcRUZH0d8CdQAG4PiJ+tcDNmpGkbwDnAn2StgEfAz4B3CzpUuD/gPem3W8H/grYSnaV0A/Oe4Ont4HsPJQt6dxSgH8mn7GsBL6q7CrYTcDNEfE9Sb8GbpL078AjpAv2pPsbJG0lG8Fc6KsnzuYfyVccK4DvpI72IvD1iLhD0oPk770F2flZN6bOtmfI2tdEzmJRttTEW8lGTybl8e/dzOZGR/peM+nTwGXA9ZKuJLtewORnwdXAtyRtB+4HTpql7o1knze1bgXeD3yzzvZN15YvAt+VtJns6usjddT1BeAHknb6PFlrRFoEAxtmZmZmZmZmdfPUYjMzMzMzM8sVJ7JmZmZmZmaWK05kzczMzMzMLFecyJqZmZmZmVmuOJE1MzMzMzOzXHEiazZHJA3Pw3O8R1JIWjfXzzVLO65Iy56YmZmZmc05J7Jm+baRbLH0hV5b9QrAiayZmZmZzQsnsmbzSNIaSXdJeizdn5jK3ynpAUmPSPqxpBWp/GpJ10u6V9Izki6rqasL2ABcSk0iK+lcST+RdLOkpyR9QtImSZslbZF0yixt+YqkC2vqG66p915Jt0h6QtKNylwGHA/cI+meOX8RzczMzOwlz4ms2fy6FvhaRLwauBG4JpXfB5wTEWcCNwFX1RyzDngbcDbwMUnNqfzdwB0R8RSwT9JZNcecAVwOnA5cDLwsIs4GrgM+MktbZnIm2ejreuBkYENEXAPsAM6LiPPqexnMzMzMzI6cE1mz+fUG4Otp+wbgTWl7FXCnpC3AlcAra475fkSUIqIf2A2sSOUbyZJe0v3GmmMejIidEVECngZ+mMq3AGtnactMNkfEtoiYAB6tqcvMzMzMbN4UF7oBZi9xke4/C3w6Im6TdC5wdc0+pZrtKlCU1AucD7xKUgAFICRdNcUxEzWPJ5j+736yLRVSJ5ckAS0ztWWm4MzMzMzM5oJHZM3m1895/nzWTWRTigF6gO1p+5I66rmQbFrwmohYGxGrgd9S36jqbG15Fnht2r4AaGZ2Q0D3i3huMzMzM7Mj5kTWbO50SNpWc/t74DLgg5IeIzt39fK079XAtyT9N9BfR90bge8cVnYr8P4X0b7p2vJF4M8kbQZeD4zUUdcXgB/4Yk9mZmZmNh8UEbPvZWZmZmZmZrZIeETWzMzMzMzMcsWJrJmZmZmZmeWKE1kzMzMzMzPLFSeyZmZmZmZmlitOZM3MzMzMzCxXnMiamZmZmZlZrjiRNTMzMzMzs1z5f0+jVNIHsyz8AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1152x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#### Let’s look at the distribution of LoanAmount variable.\n",
    "\n",
    "plt.figure(1)\n",
    "plt.subplot(121)\n",
    "df=train.dropna()\n",
    "\n",
    "sns.distplot(df['LoanAmount']); \n",
    "plt.subplot(122)\n",
    "df['LoanAmount'].plot.box(figsize=(16,5)) \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see a lot of outliers in this variable and the distribution is fairly normal. We will treat the outliers in later sections.\n",
    "\n",
    "Now we would like to know how well each feature correlate with Loan Status. So, in the next section we will look at bivariate analysis."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id=\"bv\"> </a>\n",
    "\n",
    "### Bivariate\n",
    "\n",
    "\n",
    "Lets recall some of the hypotheses that we generated earlier:\n",
    "\n",
    "Applicants with high income should have more chances of loan approval.\n",
    "Applicants who have repaid their previous debts should have higher chances of loan approval.\n",
    "Loan approval should also depend on the loan amount. If the loan amount is less, chances of loan approval should be high.\n",
    "Lesser the amount to be paid monthly to repay the loan, higher the chances of loan approval.\n",
    "Lets try to test the above mentioned hypotheses using bivariate analysis\n",
    "\n",
    "After looking at every variable individually in univariate analysis, we will now explore them again with respect to the target variable.\n",
    "\n",
    "\n",
    "Categorical Independent Variable vs Target Variable\n",
    "\n",
    "\n",
    "\n",
    "First of all we will find the relation between target variable and categorical independent variables. Let us look at the stacked bar plot now which will give us the proportion of approved and unapproved loans."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f1a43d7ba20>"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Gender=pd.crosstab(train['Gender'],train['Loan_Status'])\n",
    "Gender.div(Gender.sum(1).astype(float), axis=0).plot(kind=\"bar\", stacked=True, figsize=(4,4))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It can be inferred that the proportion of male and female applicants is more or less same for both approved and unapproved loans.\n",
    "\n",
    "Now let us visualize the remaining categorical variables vs target variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEOCAYAAABy7Vf3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAFhRJREFUeJzt3XuUVeWZ5/HvYyFgxHiB0kQKA4k48dLeuqJptW0d1EjHaMYhqDHt0O2I2qLdg7GDExOR0WmjSdvtCh0kaqIZIqKjkaQxGI2J3fES8IYjtBGvVHAU0aiICOjTf5wDlkVBnSoOHHj5ftZiefbez9nnqb3kx3ve2pfITCRJZdmq0Q1IkurPcJekAhnuklQgw12SCmS4S1KBDHdJKpDhLkkFMtwlqUCGuyQVqFejPnjAgAE5ePDgRn28JG2WHn744Vczs7mruoaF++DBg5k9e3ajPl6SNksR8UItdU7LSFKBDHdJKpDhLkkFaticuyS1t2LFCtra2li2bFmjW9kk9O3bl5aWFrbeeusevd9wl7RJaGtrY7vttmPw4MFERKPbaajMZPHixbS1tTFkyJAe7aPLaZmIuD4iXomI/7eW7RERV0fE/IiYExEH9qgTSVu0ZcuW0b9//y0+2AEigv79+6/Xt5ha5tx/CBy7ju3DgaHVP6OB7/W4G0lbNIP9A+t7LLoM98y8D3htHSUnADdmxYPADhHx8fXqSpK0Xuox5z4QWNBuua267qWOhRExmsront12260OH93B+O3rv88NYfwbje6gNh7P+vFY1tfCRxvdQW12PaBhH12PUyE7++7Q6VO3M3NyZrZmZmtzc5dXz0rawvXr12+jf+Zl/3Qtex85gn2PGsn+R5/MQ488AcA/fn8KS995p8v311q3odUj3NuAQe2WW4CFddivJG1UD8x+nJ/d/a888vMfM+fuadx98/cYtOvHAPjHa3/M0ne6/gVnrXUbWj3CfTpwWvWsmc8Cb2TmGlMyklQPL7zwAsNGnsm+R41k2MgzefH3lbj56V2/5uDjTuOAY07hqJPO4uVFiwEY/51J/NXY8Rwx4gw++Sdf4Orrblrrvl965VUG7LQDffr0BmDATjuy68eaufq6m1j48iKO/NKZHDliNABnj/vftA4/lb2PHMHF366cR9JZXftvH7feeiujRo0C4JZbbmGfffZhv/324/DDD6/vQaK2UyFvAh4A/lNEtEXE6RFxVkScVS2ZATwLzAe+D/x13buUpKoxY8Zw2ojjmHP3NE49cTjnfeNKAA476AAe/OkNPHrXTZx8wue44p9vWP2ef5//PDOnTOS3/3Ijl/zDZFasWNHpvo/5sz9hwcKX2eOwL/LXF/49v37gYQDOO/0Udt2lmXtvuYZ7b50MwGVfO4fZd05hzt038+sHH2HO3N91Wrc2EyZMYObMmTz++ONMnz69HofmQ7r8hWpmntLF9gTOqVtHkrQODzzwALd99xsA/MV//Tx/d+nVALS99DInnT2Ol155leXLVzBkt11Xv+fzww6jT5/e9OnTm50H7MjLi16jZddd1th3v20/wsM/n8K/PvQo994/i5POHsflF57LqJOOX6N22k9/weQpt7HyvZW89PKrzH36Ofbda4+af45DDz2UUaNGMXLkSE488cTuHoYueW8ZSZu1VaeDn/uNKxjzlyfxxD3TuOZbX2fZu8tX16yaZgFoampi5XvvrXV/TU1NHHFIK5d89Wy+e+nX+L8z7lmj5rkXf8+3r7mRe26exJy7p/H5YX/KsmXvrqW/D845aX9R0qRJk7j00ktZsGAB+++/P4sXL675Z66F4S5ps3LIIYcw9Y6ZAEy57U4OO6hyuuEbby5h4McqZ+HdcMtPe7Tvp+Y/z9PPvrh6+bEnn+ITLZXLdrbrty1vLVkKwJtvvc2222zD9h/tx8uLFnPnvb9Z/Z72dQC77LIL8+bN4/333+f2229fvf6ZZ57h4IMPZsKECQwYMIAFC9qfUb7+vLeMpE3W0qVLaWlpWb08duxYrr76av7q1C9x5aQbad5pR35w1XgAxp9/Jl8682sM/Fgznz3wj3huQfdP2luydCnnXnQFf3jzLXr1amL3wYOYfMVFAIw+9USGf+VcPr7zAO69dTIH7PNp9j5yBJ/crYVDP7Pf6n18qO7+WVx++eUcd9xxDBo0iH322YclS5YAcMEFF/D000+TmQwbNoz99tuv0556KipT5htfa2tr1v1JTF4oUl8ez/rxWHZp3rx57LnnnrUVbyEXMXV2TCLi4cxs7eq9TstIUoGclpG0xVn82h8YdtJZa6y/5+ZJ9N9phwZ0VH+Gu6QtTv+dduCxX0xtdBsblNMyklQgw12SCmS4S1KBnHOXVIzBV9f3hrTPn7drlzUx8EDGjv4K37l4LADfnnQjS95eyvjz1/yF7cbkyF2S1kOfPr257c5f8uprrze6lQ8x3CVpPfRqamL0qSdy1eQpjW7lQwx3SVpP54wayZTb7+SNN99qdCurGe6StJ4+ul0/ThtxHFdft+mcO2+4S1Id/O1//zLXTf0Jby9t/PNTwXCXpLrYacftGfmFo7nupjsa3QrgqZCSClLLqYsb0vln/gXf/cG0hvawiuEuSethydMfPKhjl+b+LH3m/gZ28wGnZSSpQIa7JBXIcJekAhnuklQgw12SCmS4S1KBPBVSUjkmH1Hf/Y3+1To3ZyZ/+l9O5+vnnc7w/3woANOm38X1N9/Bz6dMrG8v3eTIXZJ6KCKYdPn/ZOwl32HZsnd5e+k7fP1bE5l42bhGt+bIXZLWxz6f3p0vHH043/rnH/L20nc4bcRxfGrwoEa3ZbhL0vq6eOyZHPi5L9O799bMnvF/Gt0OYLhL0nrb9iPbcNLxx9Bv223o06d3o9sBnHOXpLrYaqtgq602nUjddDqRJNVNTdMyEXEs8E9AE3BtZl7eYftuwA3ADtWacZk5o869StK6jf5VozvYZHQZ7hHRBEwEjgbagFkRMT0z57YruwiYlpnfi4i9gBnA4A3QryRtksaff1ajW/iQWqZlDgLmZ+azmbkcmAqc0KEmgY9WX28PLKxfi5Kk7qplWmYgsKDdchtwcIea8cBdEXEusC1wVF26kyT1SC0j9+hkXXZYPgX4YWa2AH8O/Cgi1th3RIyOiNkRMXvRokXd71ZS0TI7RsuWa32PRS3h3ga0v9yqhTWnXU4HplUbegDoCwzouKPMnJyZrZnZ2tzc3LOOJRWpb9++LF682ICnEuyLFy+mb9++Pd5HLdMys4ChETEE+D1wMvDlDjUvAsOAH0bEnlTC3aG5pJq1tLTQ1tZGTd/q//DKhm+oHt6Y1+O39u3bl5aWlh6/v8twz8yVETEGmEnlNMfrM/PJiJgAzM7M6cD5wPcj4n9QmbIZlf7zK6kbtt56a4YMGVJb8fjPbthm6mX8Gw376JrOc6+esz6jw7pvtns9Fzi0vq1JknrKK1QlqUCGuyQVyHCXpAIZ7pJUIMNdkgpkuEtSgQx3SSqQ4S5JBTLcJalAhrskFchwl6QCGe6SVCDDXZIKZLhLUoEMd0kqkOEuSQUy3CWpQIa7JBXIcJekAhnuklQgw12SCmS4S1KBDHdJKpDhLkkFMtwlqUCGuyQVyHCXpAIZ7pJUIMNdkgpkuEtSgQx3SSqQ4S5JBTLcJalANYV7RBwbEU9FxPyIGLeWmpERMTcinoyIH9e3TUlSd/TqqiAimoCJwNFAGzArIqZn5tx2NUOBC4FDM/P1iNh5QzUsSepaLSP3g4D5mflsZi4HpgIndKg5A5iYma8DZOYr9W1TktQdtYT7QGBBu+W26rr29gD2iIjfRMSDEXFsZzuKiNERMTsiZi9atKhnHUuSulRLuEcn67LDci9gKHAEcApwbUTssMabMidnZmtmtjY3N3e3V0lSjWoJ9zZgULvlFmBhJzV3ZOaKzHwOeIpK2EuSGqCWcJ8FDI2IIRHRGzgZmN6h5ifAkQARMYDKNM2z9WxUklS7LsM9M1cCY4CZwDxgWmY+GRETIuL4atlMYHFEzAXuBS7IzMUbqmlJ0rp1eSokQGbOAGZ0WPfNdq8TGFv9I0lqMK9QlaQCGe6SVCDDXZIKZLhLUoEMd0kqkOEuSQUy3CWpQIa7JBXIcJekAhnuklQgw12SCmS4S1KBDHdJKpDhLkkFMtwlqUCGuyQVyHCXpAIZ7pJUIMNdkgpkuEtSgQx3SSqQ4S5JBTLcJalAhrskFchwl6QCGe6SVCDDXZIKZLhLUoEMd0kqkOEuSQUy3CWpQIa7JBWopnCPiGMj4qmImB8R49ZRNyIiMiJa69eiJKm7ugz3iGgCJgLDgb2AUyJir07qtgPOAx6qd5OSpO6pZeR+EDA/M5/NzOXAVOCETur+F3AFsKyO/UmSeqCWcB8ILGi33FZdt1pEHAAMysyfrWtHETE6ImZHxOxFixZ1u1lJUm1qCffoZF2u3hixFXAVcH5XO8rMyZnZmpmtzc3NtXcpSeqWWsK9DRjUbrkFWNhueTtgH+BXEfE88Flgur9UlaTGqSXcZwFDI2JIRPQGTgamr9qYmW9k5oDMHJyZg4EHgeMzc/YG6ViS1KUuwz0zVwJjgJnAPGBaZj4ZERMi4vgN3aAkqft61VKUmTOAGR3WfXMttUesf1uSpPXhFaqSVCDDXZIKZLhLUoEMd0kqUE2/UNWWafCyHze6hZo83+gGtNH5/2bXHLlLUoEMd0kqkOEuSQUy3CWpQIa7JBXIcJekAhnuklQgw12SClTURUxe2CBJFY7cJalAhrskFchwl6QCGe6SVCDDXZIKZLhLUoEMd0kqkOEuSQUy3CWpQIa7JBXIcJekAhnuklSgom4cJm2qvKmdNjZH7pJUIMNdkgpkuEtSgQx3SSqQ4S5JBTLcJalANYV7RBwbEU9FxPyIGNfJ9rERMTci5kTEPRHxifq3KkmqVZfhHhFNwERgOLAXcEpE7NWh7FGgNTP3BW4Frqh3o5Kk2tUycj8ImJ+Zz2bmcmAqcEL7gsy8NzOXVhcfBFrq26YkqTtqCfeBwIJ2y23VdWtzOnBnZxsiYnREzI6I2YsWLaq9S0lSt9QS7tHJuuy0MOIrQCtwZWfbM3NyZrZmZmtzc3PtXUqSuqWWe8u0AYPaLbcACzsWRcRRwNeBP8vMd+vTniSpJ2oZuc8ChkbEkIjoDZwMTG9fEBEHANcAx2fmK/VvU5LUHV2Ge2auBMYAM4F5wLTMfDIiJkTE8dWyK4F+wC0R8VhETF/L7iRJG0FNt/zNzBnAjA7rvtnu9VF17kuStB68QlWSCmS4S1KBDHdJKpDhLkkFMtwlqUCGuyQVyHCXpAIZ7pJUIMNdkgpkuEtSgQx3SSqQ4S5JBTLcJalAhrskFchwl6QCGe6SVCDDXZIKZLhLUoEMd0kqkOEuSQUy3CWpQIa7JBXIcJekAhnuklQgw12SCmS4S1KBDHdJKpDhLkkFMtwlqUCGuyQVyHCXpAIZ7pJUoJrCPSKOjYinImJ+RIzrZHufiLi5uv2hiBhc70YlSbXrMtwjogmYCAwH9gJOiYi9OpSdDryembsDVwHfqnejkqTa1TJyPwiYn5nPZuZyYCpwQoeaE4Abqq9vBYZFRNSvTUlSd/SqoWYgsKDdchtw8NpqMnNlRLwB9AdebV8UEaOB0dXFJRHxVE+a3sgG0OHnWF+xZX+v8XjWj8eyvjaX4/mJWopqCffORuDZgxoyczIwuYbP3GRExOzMbG10H6XweNaPx7K+SjuetUzLtAGD2i23AAvXVhMRvYDtgdfq0aAkqftqCfdZwNCIGBIRvYGTgekdaqYD/636egTwy8xcY+QuSdo4upyWqc6hjwFmAk3A9Zn5ZERMAGZn5nTgOuBHETGfyoj95A3Z9Ea2WU0jbQY8nvXjsayvoo5nOMCWpPJ4haokFchwl6QCGe6SVKBaznPfokTEp6lccTuQyrn6C4HpmTmvoY1JUjc4cm8nIr5G5fYKAfyWymmgAdzU2Q3TpI0pIj4dEcMiol+H9cc2qidtujxbpp2I+B2wd2au6LC+N/BkZg5tTGfliYi/zMwfNLqPzUVEnAecA8wD9gf+JjPvqG57JDMPbGR/m7uIuDAz/77RfdSTI/cPex/YtZP1H69uU/1c0ugGNjNnAH+cmV8EjgC+ERF/U93mTfrW35ca3UC9Oef+YX8L3BMRT/PBzdJ2A3YHxjSsq81URMxZ2yZgl43ZSwGaMnMJQGY+HxFHALdGxCcw3NUJp2U6iIitqNzmeCCVvzRtwKzMfK+hjW2GIuJl4HPA6x03AfdnZmffktSJiPglMDYzH2u3rhdwPXBqZjY1rLnNVEQ8R+WkiaDy7Xxh9XVm5icb2Vs9OHLvIDPfBx5sdB+F+BnQr30grRIRv9r47WzWTgNWtl+RmSuB0yLimsa0tHnLzCGrXkfEo5l5QCP7qTdH7pK2eCWGu79QlST4TaMbqDdH7pK2SBFxEJX59VkRsTeV3w/9e2bOaHBrdWG4S9riRMTFwHAqv3f8BZWTKH4NHAXMzMzLGtheXRjukrY4EfEElYvB+gD/H2jJzDcjYhvgoczct6EN1oFz7pK2RCsz873MXAo8k5lvAmTmOxRywaLhLmlLtDwiPlJ9/cerVkbE9hQS7k7LSNriRESfzHy3k/UDgI9n5hMNaKuuDHdJKpDTMpJUIMNdkgpkuGuTFhHvRcRjEfFkRDweEWOrN3drVD/PV+dle/LeL0bEXvXuSeqM4a5N3TuZuX9m7g0cDfw5cHGDe+qpLwKGuzYKw12bjcx8BRgNjImKpoi4MiJmRcSciDgTICKOiIj7IuL2iJgbEZNWjfYj4piIeCAiHomIW1Y9sq46Ir+kuv6J6rN0iYj+EXFXRDxavfvi6nunR8RXIuK31W8W10REU3X9koi4rPpN48GI2CUiDgGOB66s1n8qIs6r9jcnIqZu1IOp4hnu2qxk5rNU/r/dGTgdeCMzPwN8BjgjIlbdxvUg4Hzgj4BPASdWp1MuAo6qPpZuNjC23e5fra7/HvDV6rqLgX+r3jFwOpWHtxARewInAYdm5v7Ae8Cp1fdsCzyYmfsB9wFnZOb91fdfUP0m8gwwDjigejXkWXU7SBLez12bp1Wj52OAfSNiRHV5e2AosBz4bfUfAiLiJuAwYBmVaZHfRARAb+CBdvu9rfrfh4ETq68PX/U6M/8lIlY9eGQYlYtfZlX3tQ3wSnXbcir3sl+1r6PX8nPMAaZExE+An9T4s0s1Mdy1WYmIT1IZJb9CJeTPzcyZHWqOoPKEnfZWPXHnF5l5ylp2v+qilvf48N+Nzi4GCeCGzLywk20r8oMLSDruq73PU/nH43gqz0Tdu/oADmm9OS2jzUZENAOTgO9Ww3MmcHZEbF3dvkdEbFstPygihlTn2k8C/o3KE7YOjYjdq/UfiYg9uvjY+6hOt0TEcGDH6vp7gBERsXN1207V55muy1vAdtX6rYBBmXkv8HfADkC/Wo6DVAtH7trUbRMRjwFbU3nM3I+Af6huuxYYDDwSlbmRRVTOSIHKdMvlVObc7wNuz8z3I2IUcFNE9KnWXQT8bh2ff0m1/hEqt4R9ESAz50bERcBd1aBeAZwDvLCOfU0Fvh8R5wEnA9dV72USwFWZ+YcajodUE28/oOJUp2W+mpnHNboXqVGclpGkAjlyl6QCOXKXpAIZ7pJUIMNdkgpkuEtSgQx3SSrQfwAuTO22flkfxwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Married=pd.crosstab(train['Married'],train['Loan_Status'])\n",
    "\n",
    "Dependents=pd.crosstab(train['Dependents'],train['Loan_Status'])\n",
    "\n",
    "Education=pd.crosstab(train['Education'],train['Loan_Status'])\n",
    "\n",
    "Self_Employed=pd.crosstab(train['Self_Employed'],train['Loan_Status']) \n",
    "\n",
    "Married.div(Married.sum(1).astype(float), axis=0).plot(kind=\"bar\", stacked=True, figsize=(4,4))\n",
    "plt.show() \n",
    "\n",
    "Dependents.div(Dependents.sum(1).astype(float), axis=0).plot(kind=\"bar\", stacked=True)\n",
    "plt.show() \n",
    "\n",
    "Education.div(Education.sum(1).astype(float), axis=0).plot(kind=\"bar\", stacked=True, figsize=(4,4))\n",
    "plt.show() \n",
    "\n",
    "Self_Employed.div(Self_Employed.sum(1).astype(float), axis=0).plot(kind=\"bar\", stacked=True, figsize=(4,4))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "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>Loan_Status</th>\n",
       "      <th>N</th>\n",
       "      <th>Y</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Married</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>No</th>\n",
       "      <td>79</td>\n",
       "      <td>134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Yes</th>\n",
       "      <td>113</td>\n",
       "      <td>285</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Loan_Status    N    Y\n",
       "Married              \n",
       "No            79  134\n",
       "Yes          113  285"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Married"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "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>Loan_Status</th>\n",
       "      <th>N</th>\n",
       "      <th>Y</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Married</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>No</th>\n",
       "      <td>0.370892</td>\n",
       "      <td>0.629108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Yes</th>\n",
       "      <td>0.283920</td>\n",
       "      <td>0.716080</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Loan_Status         N         Y\n",
       "Married                        \n",
       "No           0.370892  0.629108\n",
       "Yes          0.283920  0.716080"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Married.div(Married.sum(1).astype(float), axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Proportion of married applicants is higher for the approved loans.\n",
    "\n",
    "Distribution of applicants with 1 or 3+ dependents is similar across both the categories of Loan_Status.\n",
    "\n",
    "There is nothing significant we can infer from Self_Employed vs Loan_Status plot.\n",
    "\n",
    "Now we will look at the relationship between remaining categorical independent variables and Loan_Status."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQcAAAERCAYAAAB2JGIhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAFVBJREFUeJzt3X+UVWW9x/H3h0FAgTRg/AEDDhqVSv5qUgtLWVhJmtzrVdA0LuVttPyxDCtxUUlk94ppElcLKcwfCzUkrdGLYtJoLRNj8AcGRCKajLiQMEkgAux7/zgHPBweZg7Dmdkz8nmtNWud/eznPPs7Z2Y+s5+999lHEYGZWbFOWRdgZu2Tw8HMkhwOZpbkcDCzJIeDmSU5HMwsyeFgZkkOBzNLcjiYWZLDwcySOme14T59+kR1dXVWmzfbYy1YsOCvEVHZXL/MwqG6upqGhoasNm+2x5L0l1L6eVphZkkOBzNLcjiYWVJmxxzMWsPmzZtpbGxk48aNWZeSuW7dulFVVcVee+3Vouc3Gw6SbgVOB16PiMGJ9QJ+CHwG2ACMiYinW1SN2W5qbGykZ8+eVFdXk/vV3DNFBGvWrKGxsZGBAwe2aIxSphW3Aac2sX44MCj/VQv8uEWVmJXBxo0b6d279x4dDACS6N27927tQTUbDhHxW+CNJrqMAO6InHnAfpIOanFFZrtpTw+GrXb3dSjHAcl+wIqC5cZ8m5l1YOU4IJmKp+RdayXVkpt6MGDAgDJsehdM2Ldtt9fWJqzNuoLWVerP79MzYeX2u9I9Bg1h3QtPtEJRO/e9H/6Uu375MBUVneikTtwyaTzHH/shJv9kBrXnn8k+e+/9Tue+x+zw/MmTJ1NbW8s+++zThlVvrxx7Do1A/4LlKmBlqmNETIuImoioqaxs9upNsw7pyYbnePDR3/H0w3ex8NGZPPrzH9O/74EATP7pXWz4R/PHASZPnsyGDRtau9QmlSMc6oDRyjkBWBsRr5VhXLNW85fGlQwbeSFHnjKSYSMv5JVXc7+yDzzyOMefPppjPnUup4y6iFWr1wAw4YapfHHsBE4+60sc8tHPMmX63Tsd+7XX/0qfXvvRtWsXAPr0ei99D6xkyvS7WblqNUPPvpChZ9UC8OVx/01NTQ1HHHEEV199NQBTpkxh5cqVDB06lKFDhwLQo0ePbePPmjWLMWPGAHDvvfcyePBgjjrqKD7xiU+U9TVqNhwk3Q08CXxAUqOkCyRdJOmifJfZwHJgGfAT4CtlrdCsFVwyfhKjzzqdhY/O5Lwzh3PZt74PwInHHcO8B27nmUfu5pwRn+a6H92+7Tl/WvYyc2bczB/+7w6+84NpbN68OTn2p076KCtWruL9J/4bX7nqf3j8yQUAXHbBufQ9oJL6e2+hftY0AL535cU0NDSwcOFCHn/8cRYuXMhll11G3759qa+vp76+vsnvY+LEicyZM4fnnnuOurq6crw02zR7zCEizm1mfQAXl60iszbw5ILnue+n1wPw+f84jW9cMwWAxtdWMerL43jt9b+yadNmBg7ou+05pw07ka5du9C1axf27/NeVq1+g6q+B+wwdo/u+7Dg4Rn87qlnqP/9fEZ9eRzXXnUpY0adsUPfmQ/8mmmfH8uWLVt47bXXWLx4MUceeWTJ38eQIUMYM2YMI0eO5Mwzz9zVl6FJvnzaDNh61u/Sb13HJV8YxfNzZ3LLpPFs/OembX22ThMAKioq2PL22zsdr6KigpM/VsN3vvZlbrrmSn4xe+4OfV565VWuv+UO5s6dy8KFCznttNN2el1C4WnJwj5Tp07lmmuuYcWKFRx99NGsWbOm5O+5OQ4H2yN9rOZI7vnVHABm3PcQJx6XO2Ow9u/r6Hdg7mD57fc+0KKxly57mReWv7Jt+dlFSzm4KnfpT88e3XlrXe5A49/fWk/3vfdm3333ZdWqVTz00EPbntOzZ0/eeuutbcsHHHAAS5Ys4V//+hf333//tvYXX3yR448/nokTJ9KnTx9WrCi8qmD3+L0V9q634R8bqfrwOxf5jq09nynf/QZfHDuB70+9g8pe7+VnN04AYMIVF3L2hVfS78BKTjj2Q7y0InnirUnrNmzg0m9ex5t/f4vOnSt4X3V/pl33TQBqzzuT4edfykH796F+1jSOGfxBjjjiCA455BCGDBmybYza2lqGDx/OQQcdRH19Pddeey2nn346/fv3Z/Dgwaxbtw6Ar3/967zwwgtEBMOGDeOoo47ajVdqe8rqg3RramqiTW/24uscOrYSf35LPj2Tww7ev5WLKbPEdQ7lsmTJEg477LDt2iQtiIia5p7raYWZJXlaYdZCa954k2GjLtqhfe7Pp9K7134ZVFReDgezFurdaz+e/fU9WZfRajytMLMkh4OZJTkczCzJxxzMClRP2fXrGpry8mV9m+0jibFjx3LDDTcAcP3117Nu3TomTJhQ1lp21R4TDtUb78q6hFb1ctYFWIt17dqV++67j6uuuoo+ffpkXc42nlaYZaxz587U1tZy4403Zl3KdhwOZu3AxRdfzIwZM1i7tv1c6epwMGsH3vOe9zB69GimTJmSdSnbOBzM2onLL7+c6dOns379+qxLARwOZu1Gr169GDlyJNOnT8+6FGAPOlthVopSTj22piuuuIKbbrop0xq2cjiYZWzrvRkgd1OXrO86vZWnFWaW5HAwsySHg5klORzMLMnhYGZJDgczS/KpTLNC004u73i1jzW5OiL4+IknMn78eIYPHw7AzJkzufXWW3n44YfLW8sucjiYZUgSU6dO5eyzz2bo0KG8/fbbjB8/PvNgAIeDWeYGDx7MZz/7WSZNmsT69esZPXo0hx56aNZlORzM2oOrr76aY489li5dutCmH/bUBIeDWTvQvXt3Ro0aRY8ePejatWvW5QA+W2HWbnTq1IlOndrPn2T7qcTM2pWSphWSTgV+CFQAP42Ia4vWDwBuB/bL9xkXEbPLXKtZ66t9LOsK2o1mw0FSBXAz8EmgEZgvqS4iFhd0+yYwMyJ+LOlwYDZQ3Qr1mr1rZX0r+mKlTCuOA5ZFxPKI2ATcA4wo6hPAe/KP9wXKe/N/M2tzpUwr+gErCpYbgeOL+kwAHpF0KdAdOKUs1ZlZZkrZc1CiLYqWzwVui4gq4DPAnZJ2GFtSraQGSQ2rV6/e9WrNmhVEFP967pl293UoJRwagf4Fy1XsOG24AJiZL+hJoBuww0f3RMS0iKiJiJrKysqWVWzWhG5rl7Nm/ZY9PiAigjVr1tCtW7cWj1HKtGI+MEjSQOBV4Bzgc0V9XgGGAbdJOoxcOHjXwNpc1dOTaORKVu97COmd3nZo7ZJWGbZbt25UVVW1+PnNhkNEbJF0CTCH3GnKWyNikaSJQENE1AFXAD+R9FVyU44xsadHt2Vir01vMnDeVVmXsWsmtJ9PuSpU0nUO+WsWZhe1fbvg8WJgSHlLM7Ms+QpJM0tyOJhZksPBzJIcDmaW5HAwsySHg5klORzMLMnhYGZJDgczS3I4mFmSw8HMkhwOZpbkcDCzJIeDmSU5HMwsyeFgZkkOBzNLcjiYWZLDwcySHA5mluRwMLMkh4OZJTkczCzJ4WBmSQ4HM0tyOJhZksPBzJIcDmaW5HAwsySHg5klORzMLMnhYGZJDgczSyopHCSdKmmppGWSxu2kz0hJiyUtknRXecs0s7bWubkOkiqAm4FPAo3AfEl1EbG4oM8g4CpgSET8TdL+rVWwmbWNUvYcjgOWRcTyiNgE3AOMKOrzJeDmiPgbQES8Xt4yzaytlRIO/YAVBcuN+bZC7wfeL+kJSfMknVquAs0sG81OKwAl2iIxziDgZKAK+J2kwRHx5nYDSbVALcCAAQN2uVgzazul7Dk0Av0LlquAlYk+v4qIzRHxErCUXFhsJyKmRURNRNRUVla2tGYzawOlhMN8YJCkgZK6AOcAdUV9fgkMBZDUh9w0Y3k5CzWzttVsOETEFuASYA6wBJgZEYskTZR0Rr7bHGCNpMVAPfD1iFjTWkWbWesr5ZgDETEbmF3U9u2CxwGMzX+Z2buAr5A0sySHg5klORzMLMnhYGZJDgczS3I4mFmSw8HMkhwOZpbkcDCzJIeDmSU5HMwsyeFgZkkOBzNLcjiYWZLDwcySHA5mluRwMLMkh4OZJTkczCzJ4WBmSQ4HM0tyOJhZksPBzJIcDmaW5HAwsySHg5klORzMLMnhYGZJDgczS3I4mFmSw8HMkhwOZpZUUjhIOlXSUknLJI1rot9ZkkJSTflKNLMsNBsOkiqAm4HhwOHAuZIOT/TrCVwGPFXuIs2s7ZWy53AcsCwilkfEJuAeYESi33eB64CNZazPzDJSSjj0A1YULDfm27aRdAzQPyIeLGNtZpahUsJBibbYtlLqBNwIXNHsQFKtpAZJDatXry69SjNrc6WEQyPQv2C5ClhZsNwTGAw8Jull4ASgLnVQMiKmRURNRNRUVla2vGoza3WlhMN8YJCkgZK6AOcAdVtXRsTaiOgTEdURUQ3MA86IiIZWqdjM2kSz4RARW4BLgDnAEmBmRCySNFHSGa1doJllo3MpnSJiNjC7qO3bO+l78u6XZWZZ8xWSZpbkcDCzJIeDmSU5HMwsyeFgZkkOBzNLcjiYWZLDwcySHA5mluRwMLMkh4OZJTkczCzJ4WBmSQ4HM0tyOJhZksPBzJIcDmaW5HAwsySHg5klORzMLMnhYGZJDgczS3I4mFmSw8HMkhwOZpbkcDCzJIeDmSU5HMwsyeFgZkkOBzNLcjiYWZLDwcySHA5mllRSOEg6VdJSScskjUusHytpsaSFkuZKOrj8pZpZW2o2HCRVADcDw4HDgXMlHV7U7RmgJiKOBGYB15W7UDNrW6XsORwHLIuI5RGxCbgHGFHYISLqI2JDfnEeUFXeMs2srXUuoU8/YEXBciNwfBP9LwAe2p2izIpVb7wr6xJazctZF7ATpYSDEm2R7CidD9QAJ+1kfS1QCzBgwIASSzSzLJQyrWgE+hcsVwEriztJOgUYD5wREf9MDRQR0yKiJiJqKisrW1KvmbWRUsJhPjBI0kBJXYBzgLrCDpKOAW4hFwyvl79MM2trzYZDRGwBLgHmAEuAmRGxSNJESWfku30f6AHcK+lZSXU7Gc7MOohSjjkQEbOB2UVt3y54fEqZ6zKzjPkKSTNLcjiYWZLDwcySHA5mluRwMLMkh4OZJTkczCzJ4WBmSQ4HM0tyOJhZksPBzJIcDmaW5HAwsySHg5klORzMLMnhYGZJDgczS3I4mFmSw8HMkhwOZpbkcDCzJIeDmSU5HMwsyeFgZkkOBzNLcjiYWZLDwcySHA5mluRwMLMkh4OZJTkczCzJ4WBmSSWFg6RTJS2VtEzSuMT6rpJ+nl//lKTqchdqZm2r2XCQVAHcDAwHDgfOlXR4UbcLgL9FxPuAG4FJ5S7UzNpWKXsOxwHLImJ5RGwC7gFGFPUZAdyefzwLGCZJ5SvTzNpaKeHQD1hRsNyYb0v2iYgtwFqgdzkKNLNsdC6hT2oPIFrQB0m1QG1+cZ2kpSVsv6PqA/y1rTYmT+TK6d3+szu4lE6lhEMj0L9guQpYuZM+jZI6A/sCbxQPFBHTgGmlFNbRSWqIiJqs67Bd559dTinTivnAIEkDJXUBzgHqivrUAf+Zf3wW8JuI2GHPwcw6jmb3HCJii6RLgDlABXBrRCySNBFoiIg6YDpwp6Rl5PYYzmnNos2s9cn/4FuHpNr8NMo6GP/schwOZpbky6fNLMnhYGZJDgczS3I4lJmkXpLem3UdZrvL4VAGkgZIukfSauApYL6k1/Nt1dlWZ6WSdICkYyUdI+mArOvJms9WlIGkJ4HJwKyIeDvfVgGcDVweESdkWZ81TdLRwFRyV/a+mm+uAt4EvhIRT2dVW5YcDmUg6YWIGLSr66x9kPQscGFEPFXUfgJwS0QclU1l2SrlvRXWvAWSfkTubetb38Han9wl5c9kVpWVqntxMABExDxJ3bMoqD3wnkMZ5N9zcgG5+1r0I/cu1RXAA8D0iPhnhuVZMyRNAQ4F7mD7cB8NvBQRl2RVW5YcDmaApOFsH+6NQF1EzM60sAw5HFqZpNMj4sGs6zDbVT6V2fo+knUB1nL5GxTtkXxAskwkfZB3dkuD3A1x6iLi6kwLs921x94L1XsOZSDpSnI33hXwB3I3yBFwd+pW/tahbMq6gKz4mEMZSPozcEREbC5q7wIs8nUOHZekVyJiQNZ1ZMHTivL4F9AX+EtR+0H5ddaOSVq4s1XAHnsZtcOhPC4H5kp6gXfOkw8A3gfskefIO5gDgE8DfytqF/D7ti+nfXA4lEFEPCzp/eQ+AKjwPPn8re+1sHbtQaBHRDxbvELSY21fTvvgYw5mluSzFWaW5HAwsySHg5klORw6GEkH5u8w9aKkxZJm5w+GtmSsMZJuyj++SNLogva+zTz3MUk1BcvVkv6Yf1yTf6fjzp5bLelzLanZ2o7PVnQgkgTcD9weEefk244mdyruz/nlipacIYmIqQWLY4A/suNnopY6VgPQ0ESXauBzwF2ljimpc/4T3K2NeM+hYxkKbC78Q86ffquQVC/pLuB5AEnnS/qDpGcl3ZK/bR2SviDpz5IeB4ZsHUfSBElfk3QWUAPMyD93710tUtLJkh7MPz4pP86zkp6R1BO4Fvh4vu2rkrpJ+pmk5/N9huafO0bSvZIeAB6RdKekEQXbmSHpjF1+Fa0k3nPoWAYDC3ay7jhgcES8JOkwYBQwJCI25+9SdZ6kXwPfAT4MrAXqKbpTVUTMyn826tfyewBNmSHpH/nHXUhfDfo14OKIeEJSD2AjMC4//ukAkq7Ib/tD+TewPVIwVfoocGREvCHpJOCrwK8k7Qt8jHc+wNnKzHsO7x5/iIiX8o+HkQuA+fn7Iw4DDgGOBx6LiNURsQn4+W5u87yIODoijgY+s5M+TwA/kHQZsN9OpgYnAncCRMSfyF2GvjUcfh0Rb+TXPQ68T9L+wLnALzzVaD0Oh45lEbk/+pT1BY9F7rjE0fmvD0TEhPy6Nr3qLSKuBf4L2BuYl98zKNbU26LXFy3fCZwHfAH4WVmKtCSHQ8fyG6CrpC9tbZD0EeCkon5zgbPy/2G3ftDOweQ+U+NkSb0l7UXu1vkpbwE9y1GwpEMj4vmImETuIOUHE+P/ltwfPPnpxABg6U6GvI3ce1mIiEXlqNHSfMyhA4mIkPTvwOT8fSI2Ai8Dvyzqt1jSN8nN3TsBm8nN++dJmgA8CbwGPA1UJDZ1GzA1fzzhoxHxj0SfUl2eP8D4NrAYeIjcsYktkp7Lb+tH+e09D2wBxkTEP3MnZ3Z4DVZJWlL8PVv5+b0V1qFI2ofcGZljI2Jt1vW8m3laYR2GpFOAPwH/62Bofd5zsCZJuh8YWNR8ZUTMyaIeazsOBzNL8rTCzJIcDmaW5HAwsySHg5klORzMLOn/AfN+7T0BeOp/AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "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": [
    "Credit_History = pd.crosstab(train['Credit_History'],train['Loan_Status'])\n",
    "\n",
    "Property_Area=pd.crosstab(train['Property_Area'],train['Loan_Status'])\n",
    "\n",
    "Credit_History.div(Credit_History.sum(1).astype(float), axis=0).plot(kind=\"bar\", stacked=True, figsize=(4,4))\n",
    "plt.show() \n",
    "\n",
    "Property_Area.div(Property_Area.sum(1).astype(float), axis=0).plot(kind=\"bar\", stacked=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It seems people with credit history as 1 are more likely to get their loans approved.\n",
    "Proportion of loans getting approved in semiurban area is higher as compared to that in rural or urban areas.\n",
    "Now let’s visualize numerical independent variables with respect to target variable."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Numerical Independent Variable vs Target Variable\n",
    "\n",
    "We will try to find the mean income of people for which the loan has been approved vs the mean income of people for which the loan has not been approved."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f1a3e4fee48>"
      ]
     },
     "execution_count": 93,
     "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": [
    "train.groupby('Loan_Status')['ApplicantIncome'].mean().plot.bar()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here the y-axis represents the mean applicant income. We don’t see any change in the mean income. So, let’s make bins for the applicant income variable based on the values in it and analyze the corresponding loan status for each bin."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "bins=[0,2500,4000,6000,81000]\n",
    "group=['Low','Average','High', 'Very high']\n",
    "train['Income_bin']=pd.cut(train['ApplicantIncome'],bins,labels=group)\n",
    "\n",
    "Income_bin=pd.crosstab(train['Income_bin'],train['Loan_Status'])\n",
    "Income_bin.div(Income_bin.sum(1).astype(float), axis=0).plot(kind=\"bar\", stacked=True)\n",
    "plt.xlabel('ApplicantIncome')\n",
    "P = plt.ylabel('Percentage')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It can be inferred that Applicant income does not affect the chances of loan approval which contradicts our hypothesis in which we assumed that if the applicant income is high the chances of loan approval will also be high.\n",
    "\n",
    "We will analyze the coapplicant income and loan amount variable in similar manner."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "bins=[0,1000,3000,42000]\n",
    "\n",
    "group=['Low','Average','High']\n",
    "\n",
    "train['Coapplicant_Income_bin']=pd.cut(train['CoapplicantIncome'],bins,labels=group)\n",
    "\n",
    "Coapplicant_Income_bin=pd.crosstab(train['Coapplicant_Income_bin'],train['Loan_Status'])\n",
    "\n",
    "Coapplicant_Income_bin.div(Coapplicant_Income_bin.sum(1).astype(float), axis=0).plot(kind=\"bar\", stacked=True)\n",
    "\n",
    "plt.xlabel('CoapplicantIncome')\n",
    "\n",
    "P = plt.ylabel('Percentage')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It shows that if coapplicant’s income is less the chances of loan approval are high. But this does not look right. The possible reason behind this may be that most of the applicants don’t have any coapplicant so the coapplicant income for such applicants is 0 and hence the loan approval is not dependent on it. So we can make a new variable in which we will combine the applicant’s and coapplicant’s income to visualize the combined effect of income on loan approval.\n",
    "\n",
    "Let us combine the Applicant Income and Coapplicant Income and see the combined effect of Total Income on the Loan_Status."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "train['Total_Income']=train['ApplicantIncome']+train['CoapplicantIncome']\n",
    "\n",
    "bins=[0,2500,4000,6000,81000]\n",
    "\n",
    "group=['Low','Average','High', 'Very high']\n",
    "\n",
    "train['Total_Income_bin'] = pd.cut(train['Total_Income'],bins,labels=group)\n",
    "\n",
    "Total_Income_bin=pd.crosstab(train['Total_Income_bin'],train['Loan_Status'])\n",
    "\n",
    "Total_Income_bin.div(Total_Income_bin.sum(1).astype(float), axis=0).plot(kind=\"bar\", stacked=True)\n",
    "\n",
    "plt.xlabel('Total_Income')\n",
    "\n",
    "P = plt.ylabel('Percentage')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that Proportion of loans getting approved for applicants having low Total_Income is very less as compared to that of applicants with Average, High and Very High Income."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Let’s visualize the Loan amount variable.\n",
    "\n",
    "bins=[0,100,200,700]\n",
    "\n",
    "group=['Low','Average','High']\n",
    "\n",
    "train['LoanAmount_bin']=pd.cut(train['LoanAmount'],bins,labels=group)\n",
    "\n",
    "LoanAmount_bin=pd.crosstab(train['LoanAmount_bin'],train['Loan_Status'])\n",
    "\n",
    "LoanAmount_bin.div(LoanAmount_bin.sum(1).astype(float), axis=0).plot(kind=\"bar\", stacked=True)\n",
    "\n",
    "plt.xlabel('LoanAmount')\n",
    "\n",
    "P = plt.ylabel('Percentage')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It can be seen that the proportion of approved loans is higher for Low and Average Loan Amount as compared to that of High Loan Amount which supports our hypothesis in which we considered that the chances of loan approval will be high when the loan amount is less.\n",
    "\n",
    "Let’s drop the bins which we created for the exploration part. We will change the 3+ in dependents variable to 3 to make it a numerical variable.We will also convert the target variable’s categories into 0 and 1 so that we can find its correlation with numerical variables. One more reason to do so is few models like logistic regression takes only numeric values as input. We will replace N with 0 and Y with 1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "train=train.drop(['Income_bin', 'Coapplicant_Income_bin', 'LoanAmount_bin', 'Total_Income_bin', 'Total_Income'], axis=1)\n",
    "\n",
    "train['Dependents'].replace('3+', 3,inplace=True)\n",
    "\n",
    "test['Dependents'].replace('3+', 3,inplace=True)\n",
    "\n",
    "train['Loan_Status'].replace('N', 0,inplace=True)\n",
    "\n",
    "train['Loan_Status'].replace('Y', 1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Now lets look at the correlation between all the numerical variables. We will use the heat map to visualize the correlation. Heatmaps visualize data through variations in coloring. The variables with darker color means their correlation is more.\n",
    "\n",
    "matrix = train.corr()\n",
    "\n",
    "f, ax = plt.subplots(figsize=(9, 6))\n",
    "\n",
    "sns.heatmap(matrix, vmax=.8, square=True, cmap=\"BuPu\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see that the most correlated variables are (ApplicantIncome - LoanAmount) and (Credit_History - Loan_Status). LoanAmount is also correlated with CoapplicantIncome."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id=\"mv\"> </a>\n",
    "\n",
    "### Missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Loan_ID               0\n",
       "Gender               13\n",
       "Married               3\n",
       "Dependents           15\n",
       "Education             0\n",
       "Self_Employed        32\n",
       "ApplicantIncome       0\n",
       "CoapplicantIncome     0\n",
       "LoanAmount           22\n",
       "Loan_Amount_Term     14\n",
       "Credit_History       50\n",
       "Property_Area         0\n",
       "Loan_Status           0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# After exploring all the variables in our data, we can now impute the missing values and treat the outliers because missing data and outliers can have adverse effect on the model performance.\n",
    "\n",
    "#Missing value imputation\n",
    "#Let’s list out feature-wise count of missing values.\n",
    "\n",
    "train.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [],
   "source": [
    "#There are missing values in Gender, Married, Dependents, Self_Employed, LoanAmount, Loan_Amount_Term and Credit_History features.\n",
    "\n",
    "#We will treat the missing values in all the features one by one.\n",
    "#We can consider these methods to fill the missing values:\n",
    "\n",
    "#For numerical variables: imputation using mean or median\n",
    "#For categorical variables: imputation using mode\n",
    "\n",
    "#There are very less missing values in Gender, Married, Dependents, Credit_History and Self_Employed features so we can fill them using the mode of the features. \n",
    "\n",
    "train['Gender'].fillna(train['Gender'].mode()[0], inplace=True)\n",
    "train['Married'].fillna(train['Married'].mode()[0], inplace=True)\n",
    "train['Dependents'].fillna(train['Dependents'].mode()[0], inplace=True)\n",
    "train['Self_Employed'].fillna(train['Self_Employed'].mode()[0], inplace=True)\n",
    "train['Credit_History'].fillna(train['Credit_History'].mode()[0], inplace=True)\n",
    "\n",
    "\n",
    "#Now let’s try to find a way to fill the missing values in Loan_Amount_Term. We will look at the value count of the Loan amount term variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "360.0    512\n",
       "180.0     44\n",
       "480.0     15\n",
       "300.0     13\n",
       "84.0       4\n",
       "240.0      4\n",
       "120.0      3\n",
       "36.0       2\n",
       "60.0       2\n",
       "12.0       1\n",
       "Name: Loan_Amount_Term, dtype: int64"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train['Loan_Amount_Term'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It can be seen that in loan amount term variable, the value of 360 is repeating the most. So we will replace the missing values in this variable using the mode of this variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "train['Loan_Amount_Term'].fillna(train['Loan_Amount_Term'].mode()[0], inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [],
   "source": [
    " # Now we will see the LoanAmount variable. As it is a numerical variable, we can use mean or median to impute the missing values. We will use median to fill the null values as earlier we saw that loan amount have outliers so the mean will not be the proper approach as it is highly affected by the presence of outliers.\n",
    "\n",
    "train['LoanAmount'].fillna(train['LoanAmount'].median(), inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Loan_ID              0\n",
       "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",
       "Loan_Status          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Now lets check whether all the missing values are filled in the dataset.\n",
    "\n",
    "train.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "# As we can see that all the missing values have been filled in the test dataset. Let’s fill all the missing values in the test dataset too with the same approach.\n",
    "\n",
    "test['Gender'].fillna(train['Gender'].mode()[0], inplace=True)\n",
    "\n",
    "test['Dependents'].fillna(train['Dependents'].mode()[0], inplace=True)\n",
    "\n",
    "test['Self_Employed'].fillna(train['Self_Employed'].mode()[0], inplace=True)\n",
    "\n",
    "test['Credit_History'].fillna(train['Credit_History'].mode()[0], inplace=True)\n",
    "\n",
    "test['Loan_Amount_Term'].fillna(train['Loan_Amount_Term'].mode()[0], inplace=True)\n",
    "\n",
    "test['LoanAmount'].fillna(train['LoanAmount'].median(), inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Outlier\n",
    "\n",
    "As we saw earlier in univariate analysis, LoanAmount contains outliers so we have to treat them as the presence of outliers affects the distribution of the data. Let's examine what can happen to a data set with outliers. For the sample data set:\n",
    "\n",
    "1, 1, 2, 2, 2, 2, 3, 3, 3, 4, 4\n",
    "\n",
    "We find the following: mean, median, mode, and standard deviation\n",
    "\n",
    "Mean = 2.58\n",
    "\n",
    "Median = 2.5\n",
    "\n",
    "Mode = 2\n",
    "\n",
    "Standard Deviation = 1.08\n",
    "\n",
    "If we add an outlier to the data set:\n",
    "\n",
    "1, 1, 2, 2, 2, 2, 3, 3, 3, 4, 4, 400\n",
    "\n",
    "The new values of our statistics are:\n",
    "\n",
    "Mean = 35.38\n",
    "\n",
    "Median = 2.5\n",
    "\n",
    "Mode = 2\n",
    "\n",
    "Standard Deviation = 114.74\n",
    "\n",
    "It can be seen that having outliers often has a significant effect on the mean and standard deviation and hence affecting the distribution. We must take steps to remove outliers from our data sets.\n",
    "\n",
    "Due to these outliers bulk of the data in the loan amount is at the left and the right tail is longer. This is called right skewness. One way to remove the skewness is by doing the log transformation. As we take the log transformation, it does not affect the smaller values much, but reduces the larger values. So, we get a distribution similar to normal distribution.\n",
    "\n",
    "Let’s visualize the effect of log transformation. We will do the similar changes to the test file simultaneously."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "train['LoanAmount_log'] = np.log(train['LoanAmount'])\n",
    "\n",
    "train['LoanAmount_log'].hist(bins=20)\n",
    "\n",
    "test['LoanAmount_log'] = np.log(test['LoanAmount'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now the distribution looks much closer to normal and effect of extreme values has been significantly subsided. Let’s build a logistic regression model and make predictions for the test dataset."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id=\"ev\"> </a>\n",
    "\n",
    "### Evaluation metrics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The process of model building is not complete without evaluation of model’s performance. Suppose we have the predictions from the model, how can we decide whether the predictions are accurate? We can plot the results and compare them with the actual values, i.e. calculate the distance between the predictions and actual values. Lesser this distance more accurate will be the predictions. Since this is a classification problem, we can evaluate our models using any one of the following evaluation metrics:\n",
    "\n",
    "Accuracy: Let us understand it using the confusion matrix which is a tabular representation of Actual vs Predicted values. This is how a confusion matrix looks like:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "True Positive - Targets which are actually true(Y) and we have predicted them true(Y)\n",
    "True Negative - Targets which are actually false(N) and we have predicted them false(N)\n",
    "False Positive - Targets which are actually false(N) but we have predicted them true(T)\n",
    "False Negative - Targets which are actually true(T) but we have predicted them false(N)\n",
    "Using these values, we can calculate the accuracy of the model. The accuracy is given by\n",
    "\n",
    "Tp + TN / ALL\n",
    "\n",
    "\n",
    "Precision: It is a measure of correctness achieved in true prediction i.e. of observations labeled as true, how many are actually labeled true.\n",
    "Precision = TP / (TP + FP)\n",
    "\n",
    "Recall(Sensitivity) - It is a measure of actual observations which are predicted correctly i.e. how many observations of true class are labeled correctly. It is also known as ‘Sensitivity’.\n",
    "Recall = TP / (TP + FN)\n",
    "\n",
    "Specificity - It is a measure of how many observations of false class are labeled correctly.\n",
    "Specificity = TN / (TN + FP)\n",
    "\n",
    "Specificity and Sensitivity plays a crucial role in deriving ROC curve.\n",
    "\n",
    "ROC curve\n",
    "Receiver Operating Characteristic(ROC) summarizes the model’s performance by evaluating the trade offs between true positive rate (sensitivity) and false positive rate(1- specificity).\n",
    "The area under curve (AUC), referred to as index of accuracy(A) or concordance index, is a perfect performance metric for ROC curve. Higher the area under curve, better the prediction power of the model.\n",
    "\n",
    "\n",
    "![](https://s3.amazonaws.com/thinkific/file_uploads/118220/images/681/073/341/1549257180687.jpg)\n",
    "\n",
    "\n",
    "The area of this curve measures the ability of the model to correctly classify true positives and true negatives. We want our model to predict the true classes as true and false classes as false.\n",
    "So it can be said that we want the true positive rate to be 1. But we are not concerned with the true positive rate only but the false positive rate too. For example in our problem, we are not only concerned about predicting the Y classes as Y but we also want N classes to be predicted as N.\n",
    "We want to increase the area of the curve which will be maximum for class 2,3,4 and 5 in the above example.\n",
    "For class 1 when the false positive rate is 0.2, the true positive rate is around 0.6. But for class 2 the true positive rate is 1 at the same false positive rate. So, the AUC for class 2 will be much more as compared to the AUC for class 1. So, the model for class 2 will be better.\n",
    "The class 2,3,4 and 5 model will predict more accurately as compared to the class 0 and 1 model as the AUC is more for those classes."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id=\"mb\"> </a>\n",
    "\n",
    "### Model Building\n",
    "\n",
    "\n",
    "Let us make our first model to predict the target variable. We will start with Logistic Regression which is used for predicting binary outcome.\n",
    "\n",
    "Logistic Regression is a classification algorithm. It is used to predict a binary outcome (1 / 0, Yes / No, True / False) given a set of independent variables.\n",
    "Logistic regression is an estimation of Logit function. Logit function is simply a log of odds in favor of the event.\n",
    "This function creates a s-shaped curve with the probability estimate, which is very similar to the required step wise function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Lets drop the Loan_ID variable as it do not have any effect on the loan status. We will do the same changes to the test dataset which we did for the training dataset.\n",
    "\n",
    "train=train.drop('Loan_ID',axis=1) \n",
    "test=test.drop('Loan_ID',axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "#We will use scikit-learn (sklearn) for making different models which is an open source library for Python. It is one of the most efficient tool which contains many inbuilt functions that can be used for modeling in Python.\n",
    "#To learn further about sklearn, refer here: http://scikit-learn.org/stable/tutorial/index.html\n",
    "#Sklearn requires the target variable in a separate dataset. So, we will drop our target variable from the train dataset and save it in another dataset.\n",
    "\n",
    "X = train.drop('Loan_Status',1) \n",
    "y = train.Loan_Status\n",
    "\n",
    "\n",
    "# Now we will make dummy variables for the categorical variables. Dummy variable turns categorical variables into a series of 0 and 1, making them lot easier to quantify and compare. Let us understand the process of dummies first:\n",
    "#Consider the “Gender” variable. It has two classes, Male and Female.\n",
    "\n",
    "#As logistic regression takes only the numerical values as input, we have to change male and female into numerical value.\n",
    "#Once we apply dummies to this variable, it will convert the “Gender” variable into two variables(Gender_Male and Gender_Female), one for each class, i.e. Male and Female.\n",
    "#Gender_Male will have a value of 0 if the gender is Female and a value of 1 if the gender is Male."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [],
   "source": [
    "X=pd.get_dummies(X) \n",
    "train=pd.get_dummies(train) \n",
    "test=pd.get_dummies(test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we will train the model on training dataset and make predictions for the test dataset. But can we validate these predictions? One way of doing this is we can divide our train dataset into two parts: train and validation. We can train the model on this train part and using that make predictions for the validation part. In this way we can validate our predictions as we have the true predictions for the validation part (which we do not have for the test dataset).\n",
    "\n",
    "We will use the train_test_split function from sklearn to divide our train dataset. So, first let us import train_test_split."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "x_train, x_cv, y_train, y_cv = train_test_split(X,y, test_size =0.3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "          intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
       "          penalty='l2', random_state=1, solver='liblinear', tol=0.0001,\n",
       "          verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#The dataset has been divided into training and validation part. Let us import LogisticRegression and accuracy_score from sklearn and fit the logistic regression model.\n",
    "\n",
    "from sklearn.linear_model import LogisticRegression \n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "model = LogisticRegression() \n",
    "model.fit(x_train, y_train)\n",
    "\n",
    "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,          intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,          penalty='l2', random_state=1, solver='liblinear', tol=0.0001,          verbose=0, warm_start=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7891891891891892"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Here the C parameter represents inverse of regularization strength. Regularization is applying a penalty to increasing the magnitude of parameter values in order to reduce overfitting. Smaller values of C specify stronger regularization. To learn about other parameters, refer here: http://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html\n",
    "#Let’s predict the Loan_Status for validation set and calculate its accuracy.\n",
    "\n",
    "pred_cv = model.predict(x_cv)\n",
    "\n",
    "#Let us calculate how accurate our predictions are by calculating the accuracy.\n",
    "\n",
    "accuracy_score(y_cv,pred_cv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [],
   "source": [
    "#So our predictions are almost 80% accurate, i.e. we have identified 80% of the loan status correctly.\n",
    "\n",
    "#Let’s make predictions for the test dataset.\n",
    "\n",
    "pred_test = model.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Lets import the submission file which we have to submit on the solution checker.\n",
    "\n",
    "submission=pd.read_csv(\"sample_submission.csv\")\n",
    "\n",
    "#We only need the Loan_ID and the corresponding Loan_Status for the final submission. we will fill these columns with the Loan_ID of test dataset and the predictions that we made, i.e., pred_test respectively.\n",
    "\n",
    "submission['Loan_Status']=pred_test \n",
    "submission['Loan_ID']=test_original['Loan_ID']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Remember we need predictions in Y and N. So let’s convert 1 and 0 to Y and N.\n",
    "\n",
    "submission['Loan_Status'].replace(0, 'N',inplace=True) \n",
    "submission['Loan_Status'].replace(1, 'Y',inplace=True)\n",
    "\n",
    "#Finally we will convert the submission to .csv format and make submission to check the accuracy on the leaderboard.\n",
    "\n",
    "pd.DataFrame(submission, columns=['Loan_ID','Loan_Status']).to_csv('logistic.csv')\n",
    "\n",
    "#From this submission we got an accuracy of 0.7847 on the leaderboard.\n",
    "#Instead of creating validation set, we can also make use of cross validation to validate our predictions. We will learn about this technique in next section."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id=\"lreg\"> </a>\n",
    "\n",
    "\n",
    "### Logistic Regression using stratified k-folds cross validation\n",
    "\n",
    "\n",
    "To check how robust our model is to unseen data, we can use Validation. It is a technique which involves reserving a particular sample of a dataset on which you do not train the model. Later, you test your model on this sample before finalizing it. Some of the common methods for validation are listed below:\n",
    "\n",
    "The validation set approach:\n",
    "\n",
    "\n",
    "k-fold cross validation\n",
    "\n",
    "Leave one out cross validation (LOOCV)\n",
    "\n",
    "Stratified k-fold cross validation\n",
    "\n",
    "In this section we will learn about stratified k-fold cross validation. Let us understand how it works:\n",
    "\n",
    "Stratification is the process of rearranging the data so as to ensure that each fold is a good representative of the whole.\n",
    "For example, in a binary classification problem where each class comprises of 50% of the data, it is best to arrange the data such that in every fold, each class comprises of about half the instances.\n",
    "It is generally a better approach when dealing with both bias and variance.\n",
    "A randomly selected fold might not adequately represent the minor class, particularly in cases where there is a huge class imbalance.\n",
    "\n",
    "\n",
    "![](https://s3.amazonaws.com/thinkific/file_uploads/118220/images/acf/a00/64a/1549258185586.jpg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "1 of kfold 5\n",
      "accuracy_score 0.7983870967741935\n",
      "\n",
      "2 of kfold 5\n",
      "accuracy_score 0.8306451612903226\n",
      "\n",
      "3 of kfold 5\n",
      "accuracy_score 0.8114754098360656\n",
      "\n",
      "4 of kfold 5\n",
      "accuracy_score 0.7950819672131147\n",
      "\n",
      "5 of kfold 5\n",
      "accuracy_score 0.8278688524590164\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "#Let’s import StratifiedKFold from sklearn and fit the model.\n",
    "\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "\n",
    "#Now let’s make a cross validation logistic model with stratified 5 folds and make predictions for test dataset.\n",
    "\n",
    "i=1\n",
    "\n",
    "kf = StratifiedKFold(n_splits=5,random_state=1,shuffle=True)\n",
    "\n",
    "for train_index,test_index in kf.split(X,y): \n",
    "    print('\\n{} of kfold {}'.format(i,kf.n_splits)) \n",
    "    \n",
    "    xtr,xvl = X.loc[train_index],X.loc[test_index]\n",
    "    ytr,yvl = y[train_index],y[test_index]\n",
    "    \n",
    "    model = LogisticRegression(random_state=1)\n",
    "    model.fit(xtr, ytr)\n",
    "    \n",
    "    pred_test = model.predict(xvl)\n",
    "    score = accuracy_score(yvl,pred_test)\n",
    "    \n",
    "    print('accuracy_score',score)\n",
    "    i+=1\n",
    "    \n",
    "    pred_test = model.predict(test)\n",
    "    pred=model.predict_proba(xvl)[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#The mean validation accuracy for this model turns out to be 0.81. Let us visualize the roc curve.\n",
    "\n",
    "from sklearn import metrics\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(yvl,  pred)\n",
    "\n",
    "auc = metrics.roc_auc_score(yvl, pred)\n",
    "\n",
    "plt.figure(figsize=(12,8))\n",
    "\n",
    "plt.plot(fpr,tpr,label=\"validation, auc=\"+str(auc))\n",
    "\n",
    "plt.xlabel('False Positive Rate')\n",
    "\n",
    "plt.ylabel('True Positive Rate')\n",
    "\n",
    "plt.legend(loc=4)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [],
   "source": [
    "#We got an auc value of 0.77.\n",
    "\n",
    "submission['Loan_Status']=pred_test\n",
    "submission['Loan_ID']=test_original['Loan_ID']\n",
    "\n",
    "#Remember we need predictions in Y and N. So let’s convert 1 and 0 to Y and N.\n",
    "\n",
    "submission['Loan_Status'].replace(0, 'N',inplace=True)\n",
    "submission['Loan_Status'].replace(1, 'Y',inplace=True)\n",
    "\n",
    "#Lets convert the submission to .csv format and make submission to check the accuracy on the leaderboard.\n",
    "\n",
    "pd.DataFrame(submission, columns=['Loan_ID','Loan_Status']).to_csv('Logistic.csv')\n",
    "\n",
    "#From this submission we got an accuracy of 0.78472 on the leaderboard. Now we will try to improve this accuracy using different approaches."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id=\"fe\"> </a>\n",
    "\n",
    "### Feature Engineering"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Based on the domain knowledge, we can come up with new features that might affect the target variable. We will create the following three new features:\n",
    "\n",
    "Total Income - As discussed during bivariate analysis we will combine the Applicant Income and Coapplicant Income. If the total income is high, chances of loan approval might also be high.\n",
    "EMI - EMI is the monthly amount to be paid by the applicant to repay the loan. Idea behind making this variable is that people who have high EMI’s might find it difficult to pay back the loan. We can calculate the EMI by taking the ratio of loan amount with respect to loan amount term.\n",
    "Balance Income - This is the income left after the EMI has been paid. Idea behind creating this variable is that if this value is high, the chances are high that a person will repay the loan and hence increasing the chances of loan approval."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [],
   "source": [
    "train['Total_Income']=train['ApplicantIncome']+train['CoapplicantIncome']\n",
    "\n",
    "test['Total_Income']=test['ApplicantIncome']+test['CoapplicantIncome']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Let’s check the distribution of Total Income.\n",
    "\n",
    "sns.distplot(train['Total_Income']);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#We can see it is shifted towards left, i.e., the distribution is right skewed. So, let’s take the log transformation to make the distribution normal.\n",
    "\n",
    "train['Total_Income_log'] = np.log(train['Total_Income'])\n",
    "\n",
    "sns.distplot(train['Total_Income_log']); \n",
    "\n",
    "\n",
    "test['Total_Income_log'] = np.log(test['Total_Income'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Now the distribution looks much closer to normal and effect of extreme values has been significantly subsided. Let’s create the EMI feature now.\n",
    "\n",
    "train['EMI']=train['LoanAmount']/train['Loan_Amount_Term']\n",
    "\n",
    "test['EMI']=test['LoanAmount']/test['Loan_Amount_Term']\n",
    "\n",
    "#Let’s check the distribution of EMI variable.\n",
    "\n",
    "sns.distplot(train['EMI']);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Let us create Balance Income feature now and check its distribution.\n",
    "\n",
    "train['Balance Income'] = train['Total_Income']-(train['EMI']*1000) # Multiply with 1000 to make the units equal test['Balance Income']=test['Total_Income']-(test['EMI']*1000)\n",
    "\n",
    "test['Balance Income'] = test['Total_Income']-(test['EMI']*1000) # Multiply with 1000 to make the units equal test['Balance Income']=test['Total_Income']-(test['EMI']*1000)\n",
    "\n",
    "sns.distplot(train['Balance Income']);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let us now drop the variables which we used to create these new features. Reason for doing this is, the correlation between those old features and these new features will be very high and logistic regression assumes that the variables are not highly correlated. We also wants to remove the noise from the dataset, so removing correlated features will help in reducing the noise too.\n",
    "\n",
    "train=train.drop(['ApplicantIncome', 'CoapplicantIncome', 'LoanAmount', 'Loan_Amount_Term'], axis=1) \n",
    "test=test.drop(['ApplicantIncome', 'CoapplicantIncome', 'LoanAmount', 'Loan_Amount_Term'], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id=\"mbb\"> </a>\n",
    "\n",
    "### Model Building : Part II]\n",
    "\n",
    "After creating new features, we can continue the model building process. So we will start with logistic regression model and then move over to more complex models like RandomForest and XGBoost.\n",
    "\n",
    "We will build the following models in this section.\n",
    "\n",
    "Logistic Regression\n",
    "Decision Tree\n",
    "Random Forest\n",
    "XGBoost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let’s prepare the data for feeding into the models.\n",
    "\n",
    "X = train.drop('Loan_Status',1) \n",
    "y = train.Loan_Status                # Save target variable in separate dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "1 of kfold 5\n",
      "accuracy_score 0.8064516129032258\n",
      "\n",
      "2 of kfold 5\n",
      "accuracy_score 0.8306451612903226\n",
      "\n",
      "3 of kfold 5\n",
      "accuracy_score 0.7786885245901639\n",
      "\n",
      "4 of kfold 5\n",
      "accuracy_score 0.7868852459016393\n",
      "\n",
      "5 of kfold 5\n",
      "accuracy_score 0.819672131147541\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "#Logistic Regression\n",
    "\n",
    "i=1 \n",
    "\n",
    "kf = StratifiedKFold(n_splits=5,random_state=1,shuffle=True) \n",
    "\n",
    "for train_index,test_index in kf.split(X,y):\n",
    "    \n",
    "    print('\\n{} of kfold {}'.format(i,kf.n_splits))     \n",
    "    \n",
    "    xtr,xvl = X.loc[train_index],X.loc[test_index]     \n",
    "    ytr,yvl = y[train_index],y[test_index]         \n",
    "\n",
    "    model = LogisticRegression(random_state=1)     \n",
    "    model.fit(xtr, ytr)     \n",
    "    pred_test = model.predict(xvl)     \n",
    "    score = accuracy_score(yvl,pred_test)     \n",
    "    print('accuracy_score',score)     \n",
    "    i+=1 \n",
    "    \n",
    "    pred_test = model.predict(test) \n",
    "    pred=model.predict_proba(xvl)[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [],
   "source": [
    "#The mean validation accuracy for this model is 0.812\n",
    "\n",
    "submission['Loan_Status']=pred_test # filling Loan_Status with predictions submission['Loan_ID']=test_original['Loan_ID'] # filling Loan_ID with test Loan_ID\n",
    "\n",
    "# replacing 0 and 1 with N and Y \n",
    "submission['Loan_Status'].replace(0, 'N',inplace=True) \n",
    "submission['Loan_Status'].replace(1, 'Y',inplace=True)\n",
    "\n",
    "# Converting submission file to .csv format \n",
    "pd.DataFrame(submission, columns=['Loan_ID','Loan_Status']).to_csv('Log2.csv')\n",
    "\n",
    "#From this submission we got an accuracy of 0.7847 on the leaderboard. So we can infer feature engineering has not improved the model. Let us look at some other algorithms."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "1 of kfold 5\n",
      "accuracy_score 0.7258064516129032\n",
      "\n",
      "2 of kfold 5\n",
      "accuracy_score 0.7419354838709677\n",
      "\n",
      "3 of kfold 5\n",
      "accuracy_score 0.7049180327868853\n",
      "\n",
      "4 of kfold 5\n",
      "accuracy_score 0.680327868852459\n",
      "\n",
      "5 of kfold 5\n",
      "accuracy_score 0.7049180327868853\n"
     ]
    }
   ],
   "source": [
    "# Decision Tree\n",
    "\n",
    "#Decision tree is a type of supervised learning algorithm(having a pre-defined target variable) that is mostly used in classification problems. In this technique, we split the population or sample into two or more homogeneous sets(or sub-populations) based on most significant splitter / differentiator in input variables.\n",
    "#Decision trees use multiple algorithms to decide to split a node in two or more sub-nodes. The creation of sub-nodes increases the homogeneity of resultant sub-nodes. In other words, we can say that purity of the node increases with respect to the target variable.\n",
    "\n",
    "\n",
    "from sklearn import tree\n",
    "\n",
    "#Let’s fit the decision tree model with 5 folds of cross validation.\n",
    "\n",
    "i=1 \n",
    "kf = StratifiedKFold(n_splits=5,random_state=1,shuffle=True) \n",
    "\n",
    "for train_index,test_index in kf.split(X,y):\n",
    "    print('\\n{} of kfold {}'.format(i,kf.n_splits))     \n",
    "    xtr,xvl = X.loc[train_index],X.loc[test_index]     \n",
    "    ytr,yvl = y[train_index],y[test_index]         \n",
    "    \n",
    "    model = tree.DecisionTreeClassifier(random_state=1)     \n",
    "    model.fit(xtr, ytr)     \n",
    "    \n",
    "    pred_test = model.predict(xvl)     \n",
    "    score = accuracy_score(yvl,pred_test)     \n",
    "    print('accuracy_score',score)     \n",
    "    i+=1 \n",
    "    pred_test = model.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [],
   "source": [
    "# The mean validation accuracy for this model is 0.69\n",
    "\n",
    "submission['Loan_Status']=pred_test            # filling Loan_Status with predictions submission['Loan_ID']=test_original['Loan_ID'] # filling Loan_ID with test Loan_ID\n",
    "\n",
    "# replacing 0 and 1 with N and Y \n",
    "submission['Loan_Status'].replace(0, 'N',inplace=True) \n",
    "submission['Loan_Status'].replace(1, 'Y',inplace=True)\n",
    "\n",
    "# Converting submission file to .csv format \n",
    "pd.DataFrame(submission, columns=['Loan_ID','Loan_Status']).to_csv('Decision Tree.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We got an accuracy of 0.63 which is much lesser than the accuracy from logistic regression model. So let’s build another model, i.e. Random Forest, a tree based ensemble algorithm and try to improve our model by improving the accuracy."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Random Forest\n",
    "\n",
    "RandomForest is a tree based bootstrapping algorithm wherein a certain no. of weak learners (decision trees) are combined to make a powerful prediction model.\n",
    "For every individual learner, a random sample of rows and a few randomly chosen variables are used to build a decision tree model.\n",
    "Final prediction can be a function of all the predictions made by the individual learners.\n",
    "In case of regression problem, the final prediction can be mean of all the predictions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "1 of kfold 5\n",
      "accuracy_score 0.8225806451612904\n",
      "\n",
      "2 of kfold 5\n",
      "accuracy_score 0.8145161290322581\n",
      "\n",
      "3 of kfold 5\n",
      "accuracy_score 0.7377049180327869\n",
      "\n",
      "4 of kfold 5\n",
      "accuracy_score 0.7295081967213115\n",
      "\n",
      "5 of kfold 5\n",
      "accuracy_score 0.8114754098360656\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/ensemble/forest.py:246: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n",
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/ensemble/forest.py:246: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n",
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/ensemble/forest.py:246: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n",
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/ensemble/forest.py:246: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n",
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/ensemble/forest.py:246: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "i=1 \n",
    "kf = StratifiedKFold(n_splits=5,random_state=1,shuffle=True) \n",
    "\n",
    "for train_index,test_index in kf.split(X,y):     \n",
    "    print('\\n{} of kfold {}'.format(i,kf.n_splits))     \n",
    "    xtr,xvl = X.loc[train_index],X.loc[test_index]     \n",
    "    ytr,yvl = y[train_index],y[test_index]         \n",
    "    model = RandomForestClassifier(random_state=1, max_depth=10)     \n",
    "    model.fit(xtr, ytr)     \n",
    "    pred_test = model.predict(xvl)     \n",
    "    score = accuracy_score(yvl,pred_test)     \n",
    "    print('accuracy_score',score)     \n",
    "    i+=1 \n",
    "\n",
    "pred_test = model.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n",
      "/home/akash/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_search.py:841: DeprecationWarning: The default of the `iid` parameter will change from True to False in version 0.22 and will be removed in 0.24. This will change numeric results when test-set sizes are unequal.\n",
      "  DeprecationWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv='warn', error_score='raise-deprecating',\n",
       "       estimator=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='warn', n_jobs=None,\n",
       "            oob_score=False, random_state=1, verbose=0, warm_start=False),\n",
       "       fit_params=None, iid='warn', n_jobs=None,\n",
       "       param_grid={'max_depth': [1, 3, 5, 7, 9, 11, 13, 15, 17, 19], 'n_estimators': [1, 21, 41, 61, 81, 101, 121, 141, 161, 181]},\n",
       "       pre_dispatch='2*n_jobs', refit=True, return_train_score='warn',\n",
       "       scoring=None, verbose=0)"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#The mean validation accuracy for this model is 0.766\n",
    "#We will try to improve the accuracy by tuning the hyperparameters for this model. We will use grid search to get the optimized values of hyper parameters. Grid-search is a way to select the best of a family of hyper parameters, parametrized by a grid of parameters.\n",
    "#We will tune the max_depth and n_estimators parameters. max_depth decides the maximum depth of the tree and n_estimators decides the number of trees that will be used in random forest model.\n",
    "\n",
    "\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "# Provide range for max_depth from 1 to 20 with an interval of 2 and from 1 to 200 with an interval of 20 for n_estimators \n",
    "paramgrid = {'max_depth': list(range(1, 20, 2)), 'n_estimators': list(range(1, 200, 20))}\n",
    "grid_search=GridSearchCV(RandomForestClassifier(random_state=1),paramgrid)\n",
    "\n",
    "from sklearn.model_selection import train_test_split \n",
    "x_train, x_cv, y_train, y_cv = train_test_split(X,y, test_size =0.3, random_state=1)\n",
    "\n",
    "# Fit the grid search model \n",
    "grid_search.fit(x_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "            max_depth=3, 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=141, n_jobs=None,\n",
       "            oob_score=False, random_state=1, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 143,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Estimating the optimized value \n",
    "grid_search.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "1 of kfold 5\n",
      "accuracy_score 0.7983870967741935\n",
      "\n",
      "2 of kfold 5\n",
      "accuracy_score 0.8225806451612904\n",
      "\n",
      "3 of kfold 5\n",
      "accuracy_score 0.8032786885245902\n",
      "\n",
      "4 of kfold 5\n",
      "accuracy_score 0.7786885245901639\n",
      "\n",
      "5 of kfold 5\n",
      "accuracy_score 0.819672131147541\n"
     ]
    }
   ],
   "source": [
    "#So, the optimized value for the max_depth variable is 3 and for n_estimator is 41. Now let’s build the model using these optimized values.\n",
    "\n",
    "i=1 \n",
    "kf = StratifiedKFold(n_splits=5,random_state=1,shuffle=True) \n",
    "\n",
    "for train_index,test_index in kf.split(X,y):     \n",
    "    print('\\n{} of kfold {}'.format(i,kf.n_splits))     \n",
    "    xtr,xvl = X.loc[train_index],X.loc[test_index]     \n",
    "    ytr,yvl = y[train_index],y[test_index]         \n",
    "    model = RandomForestClassifier(random_state=1, max_depth=3, n_estimators=41)\n",
    "    model.fit(xtr, ytr)     \n",
    "    pred_test = model.predict(xvl)     \n",
    "    score = accuracy_score(yvl,pred_test)     \n",
    "    print('accuracy_score',score)     \n",
    "    i+=1 \n",
    "    pred_test = model.predict(test) \n",
    "    pred2=model.predict_proba(test)[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "1 of kfold 5\n",
      "accuracy_score 0.7983870967741935\n",
      "\n",
      "2 of kfold 5\n",
      "accuracy_score 0.8225806451612904\n",
      "\n",
      "3 of kfold 5\n",
      "accuracy_score 0.8032786885245902\n",
      "\n",
      "4 of kfold 5\n",
      "accuracy_score 0.7786885245901639\n",
      "\n",
      "5 of kfold 5\n",
      "accuracy_score 0.819672131147541\n"
     ]
    }
   ],
   "source": [
    "#So, the optimized value for the max_depth variable is 3 and for n_estimator is 41. Now let’s build the model using these optimized values.\n",
    "\n",
    "i=1 \n",
    "kf = StratifiedKFold(n_splits=5,random_state=1,shuffle=True) \n",
    "for train_index,test_index in kf.split(X,y):     \n",
    "    print('\\n{} of kfold {}'.format(i,kf.n_splits))     \n",
    "    xtr,xvl = X.loc[train_index],X.loc[test_index]     \n",
    "    ytr,yvl = y[train_index],y[test_index]         \n",
    "    model = RandomForestClassifier(random_state=1, max_depth=3, n_estimators=41)\n",
    "    model.fit(xtr, ytr)     \n",
    "    pred_test = model.predict(xvl)     \n",
    "    score = accuracy_score(yvl,pred_test)     \n",
    "    print('accuracy_score',score)     \n",
    "    i+=1 \n",
    "    pred_test = model.predict(test) \n",
    "    pred2=model.predict_proba(test)[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f1a3e13a390>"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "submission['Loan_Status']=pred_test            # filling Loan_Status with predictions submission['Loan_ID']=test_original['Loan_ID'] # filling Loan_ID with test Loan_ID\n",
    "\n",
    "# replacing 0 and 1 with N and Y \n",
    "submission['Loan_Status'].replace(0, 'N',inplace=True) \n",
    "submission['Loan_Status'].replace(1, 'Y',inplace=True)\n",
    "\n",
    "# Converting submission file to .csv format \n",
    "pd.DataFrame(submission, columns=['Loan_ID','Loan_Status']).to_csv('Random Forest.csv')\n",
    "\n",
    "#We got an accuracy of 0.7638 from the random forest model on leaderboard.\n",
    "#Let us find the feature importance now, i.e. which features are most important for this problem. We will use feature_importances_ attribute of sklearn to do so.\n",
    "\n",
    "importances=pd.Series(model.feature_importances_, index=X.columns)\n",
    "\n",
    "importances.plot(kind='barh', figsize=(12,8))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "1 of kfold 5\n",
      "accuracy_score 0.782258064516129\n",
      "\n",
      "2 of kfold 5\n",
      "accuracy_score 0.8225806451612904\n",
      "\n",
      "3 of kfold 5\n",
      "accuracy_score 0.7622950819672131\n",
      "\n",
      "4 of kfold 5\n",
      "accuracy_score 0.7459016393442623\n",
      "\n",
      "5 of kfold 5\n",
      "accuracy_score 0.7868852459016393\n"
     ]
    }
   ],
   "source": [
    "#We can see that Credit_History is the most important feature followed by Balance Income, Total Income, EMI. So, feature engineering helped us in predicting our target variable.\n",
    "# XGBOOST\n",
    "\n",
    "from xgboost import XGBClassifier\n",
    "i=1 \n",
    "\n",
    "kf = StratifiedKFold(n_splits=5,random_state=1,shuffle=True) \n",
    "for train_index,test_index in kf.split(X,y):     \n",
    "    print('\\n{} of kfold {}'.format(i,kf.n_splits))     \n",
    "    xtr,xvl = X.loc[train_index],X.loc[test_index]     \n",
    "    ytr,yvl = y[train_index],y[test_index]         \n",
    "    model = XGBClassifier(n_estimators=50, max_depth=4)     \n",
    "    model.fit(xtr, ytr)     \n",
    "    pred_test = model.predict(xvl)     \n",
    "    score = accuracy_score(yvl,pred_test)     \n",
    "    print('accuracy_score',score)     \n",
    "    i+=1 \n",
    "    pred_test = model.predict(test) \n",
    "    pred3=model.predict_proba(test)[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {},
   "outputs": [],
   "source": [
    "# The mean validation accuracy for this model is 0.79\n",
    "\n",
    "submission['Loan_Status']=pred_test \n",
    "submission['Loan_ID']=test_original['Loan_ID']\n",
    "\n",
    "submission['Loan_Status'].replace(0, 'N',inplace=True) \n",
    "submission['Loan_Status'].replace(1, 'Y',inplace=True)\n",
    "pd.DataFrame(submission, columns=['Loan_ID','Loan_Status']).to_csv('XGBoost.csv')\n",
    "\n",
    "#We got an accuracy of 0.73611 with this model.\n",
    "#After trying and testing 4 different algorithms, the best accuracy on the public leaderboard is achieved by Logistic Regression (0.7847), followed by RandomForest (0.7638)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Hope this tutorial was helpful for you in understanding how a machine learning competition is approached and what are the steps one should go through to build a robust model. So, do replicate this analysis and let us know if you face any issues.\n",
    "\n",
    "What more can be tried?\n",
    "\n",
    "\n",
    "There are still quite a many things that can be tried to improve our models’ predictions. We create and add more variables, try different models with different subset of features and/or rows, etc. Some of the ideas are listed below:\n",
    "\n",
    "We can train the XGBoost model using grid search to optimize its hyperparameters and improve the accuracy.\n",
    "\n",
    "We can combine the applicants with 1,2,3 or more dependents and make a new feature as discussed in the EDA part.\n",
    "\n",
    "We can also make independent vs independent variable visualizations to discover some more patterns.\n",
    "\n",
    "We can also arrive at the EMI using a better formula which may include interest rates as well.\n",
    "\n",
    "We can even try ensemble modeling (combination of different models)"
   ]
  }
 ],
 "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}