MachineLearning_Classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "4GWfh0lCO7e7"
   },
   "source": [
    "# **Credit Risk- Loan Elligibility Prediction**\n",
    "\n",
    "Goal:\n",
    "------------------\n",
    "\n",
    "Build a model to predict whether the loan should be approved or not. This would help the company prevent potential losses, save time and focus more on eligible customers.\n",
    "\n",
    "--------------------------\n",
    "Data Attributes:\n",
    "--------------------------\n",
    "\n",
    "The data contains the following attributes:\n",
    "\n",
    "* **Loan_ID**: Unique Loan ID\n",
    "* **Gender**: Gender of the applicant - Male/Female\n",
    "* **Married**: Whether the applicant is married or not (Yes/No)\n",
    "* **Dependents**: Number of dependents of the applicant\n",
    "* **Education**: Applicant's education (Graduate/Not Graduate)\n",
    "* **Self_Employed**: Whether the applicant is self-employed (Yes/No)\n",
    "* **ApplicantIncome**: The income of the applicant (\\$)\n",
    "* **CoapplicantIncome**: The co-applicant's income in case of a joint loan and 0 otherwise (\\$)\n",
    "* **LoanAmount**: Loan amount (dollars in thousands) \n",
    "* **Loan_Amount_Term**: Term of loan in months\n",
    "* **Credit_History**: Whether the applicant's credit history meets required guidelines\n",
    "* **Property_Area**: The area the property pertaining to the loan belongs to - Urban/Semi-Urban/Rural\n",
    "* **Loan_Status**: Loan approved (1 - Yes, 0 - No)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "tCM0C1YSO7ff"
   },
   "source": [
    "### Importing the necessary libraries "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "HWgrS7UdO7fi"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "#To scale the data using z-score \n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "#Algorithms to use\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "from sklearn.discriminant_analysis import QuadraticDiscriminantAnalysis\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn import tree\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "#Metrics to evaluate the model\n",
    "from sklearn.metrics import confusion_matrix, classification_report, recall_score,  precision_recall_curve\n",
    "from sklearn import metrics\n",
    "\n",
    "#For hyperparameter tuning\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "UTBJn4meO7fq",
    "outputId": "087d9d1f-e924-4e37-97a8-872a338df5f6"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Loan_ID</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Married</th>\n",
       "      <th>Dependents</th>\n",
       "      <th>Education</th>\n",
       "      <th>Self_Employed</th>\n",
       "      <th>ApplicantIncome</th>\n",
       "      <th>CoapplicantIncome</th>\n",
       "      <th>LoanAmount</th>\n",
       "      <th>Loan_Amount_Term</th>\n",
       "      <th>Credit_History</th>\n",
       "      <th>Property_Area</th>\n",
       "      <th>Loan_Status</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LP001002</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>5849</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>LP001003</td>\n",
       "      <td>Male</td>\n",
       "      <td>Yes</td>\n",
       "      <td>1</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>4583</td>\n",
       "      <td>1508.0</td>\n",
       "      <td>128</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Rural</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LP001005</td>\n",
       "      <td>Male</td>\n",
       "      <td>Yes</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>Yes</td>\n",
       "      <td>3000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>66</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>LP001006</td>\n",
       "      <td>Male</td>\n",
       "      <td>Yes</td>\n",
       "      <td>0</td>\n",
       "      <td>Not Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>2583</td>\n",
       "      <td>2358.0</td>\n",
       "      <td>120</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LP001008</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>6000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>141</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Loan_ID Gender Married Dependents     Education Self_Employed  \\\n",
       "0  LP001002   Male      No          0      Graduate            No   \n",
       "1  LP001003   Male     Yes          1      Graduate            No   \n",
       "2  LP001005   Male     Yes          0      Graduate           Yes   \n",
       "3  LP001006   Male     Yes          0  Not Graduate            No   \n",
       "4  LP001008   Male      No          0      Graduate            No   \n",
       "\n",
       "   ApplicantIncome  CoapplicantIncome  LoanAmount  Loan_Amount_Term  \\\n",
       "0             5849                0.0           0             360.0   \n",
       "1             4583             1508.0         128             360.0   \n",
       "2             3000                0.0          66             360.0   \n",
       "3             2583             2358.0         120             360.0   \n",
       "4             6000                0.0         141             360.0   \n",
       "\n",
       "   Credit_History Property_Area  Loan_Status  \n",
       "0             1.0         Urban            1  \n",
       "1             1.0         Rural            0  \n",
       "2             1.0         Urban            1  \n",
       "3             1.0         Urban            1  \n",
       "4             1.0         Urban            1  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Loading the data\n",
    "data = pd.read_csv(\"CreditRisk.csv\")\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "id": "i1ki3y1vO7gI",
    "outputId": "99969b10-df9c-4360-ad91-8d2a2847fc57"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 614 entries, 0 to 613\n",
      "Data columns (total 13 columns):\n",
      " #   Column             Non-Null Count  Dtype  \n",
      "---  ------             --------------  -----  \n",
      " 0   Loan_ID            614 non-null    object \n",
      " 1   Gender             601 non-null    object \n",
      " 2   Married            611 non-null    object \n",
      " 3   Dependents         599 non-null    object \n",
      " 4   Education          614 non-null    object \n",
      " 5   Self_Employed      582 non-null    object \n",
      " 6   ApplicantIncome    614 non-null    int64  \n",
      " 7   CoapplicantIncome  614 non-null    float64\n",
      " 8   LoanAmount         614 non-null    int64  \n",
      " 9   Loan_Amount_Term   600 non-null    float64\n",
      " 10  Credit_History     564 non-null    float64\n",
      " 11  Property_Area      614 non-null    object \n",
      " 12  Loan_Status        614 non-null    int64  \n",
      "dtypes: float64(3), int64(3), object(7)\n",
      "memory usage: 62.5+ KB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ZkaIUWERO7gL"
   },
   "source": [
    "**Observations:**\n",
    "\n",
    "- There are **614 observations and 13 columns** in the data.\n",
    "- ApplicantIncome, CoapplicantIncome, LoanAmount, Loan_Amount_Term, Credit_History, and Loan_Status are numeric data types. **The rest of the variables are of the object data type.**\n",
    "- There are several columns with less than 614 non-null entries i.e. **these columns have missing values.**\n",
    "- **Loan_ID column is an identifier.** Let's check if each entry of the column is unique."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "id": "4gLzBxziO7gX",
    "outputId": "a55f342b-861d-43fd-9635-12e238824157"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "614"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.Loan_ID.nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "hFmPeGCjO7gb"
   },
   "source": [
    "**Observations:**\n",
    "- We can see that **all the entries of this column are unique.** Hence, this column would not add any value to our analysis. \n",
    "- Let's drop this column."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Ps3m9-FRO7gd"
   },
   "source": [
    "## **Exploratory Data Analysis and Data Preprocessing**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "fjtZPbQrO7gp"
   },
   "outputs": [],
   "source": [
    "#Dropping the Loan_ID column**\n",
    "data.drop(columns=['Loan_ID'], inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "y_2yAAi7O7gq"
   },
   "source": [
    "**Summary Statistics**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "psLyeifzO7gz",
    "outputId": "645a4565-e6ac-4380-ee81-7ea0f7975c3e"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ApplicantIncome</th>\n",
       "      <td>614.0</td>\n",
       "      <td>5403.459283</td>\n",
       "      <td>6109.041673</td>\n",
       "      <td>150.0</td>\n",
       "      <td>2877.5</td>\n",
       "      <td>3812.5</td>\n",
       "      <td>5795.00</td>\n",
       "      <td>81000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CoapplicantIncome</th>\n",
       "      <td>614.0</td>\n",
       "      <td>1621.245798</td>\n",
       "      <td>2926.248369</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1188.5</td>\n",
       "      <td>2297.25</td>\n",
       "      <td>41667.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LoanAmount</th>\n",
       "      <td>614.0</td>\n",
       "      <td>141.166124</td>\n",
       "      <td>88.340630</td>\n",
       "      <td>0.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>164.75</td>\n",
       "      <td>700.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Loan_Amount_Term</th>\n",
       "      <td>600.0</td>\n",
       "      <td>342.000000</td>\n",
       "      <td>65.120410</td>\n",
       "      <td>12.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>360.00</td>\n",
       "      <td>480.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   count         mean          std    min     25%     50%  \\\n",
       "ApplicantIncome    614.0  5403.459283  6109.041673  150.0  2877.5  3812.5   \n",
       "CoapplicantIncome  614.0  1621.245798  2926.248369    0.0     0.0  1188.5   \n",
       "LoanAmount         614.0   141.166124    88.340630    0.0    98.0   125.0   \n",
       "Loan_Amount_Term   600.0   342.000000    65.120410   12.0   360.0   360.0   \n",
       "\n",
       "                       75%      max  \n",
       "ApplicantIncome    5795.00  81000.0  \n",
       "CoapplicantIncome  2297.25  41667.0  \n",
       "LoanAmount          164.75    700.0  \n",
       "Loan_Amount_Term    360.00    480.0  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Creating a list of numerical columns\n",
    "num_cols = ['ApplicantIncome', 'CoapplicantIncome', 'LoanAmount', 'Loan_Amount_Term']\n",
    "\n",
    "data[num_cols].describe().T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "d8uDsaE3O7g1"
   },
   "source": [
    "**Observations:**\n",
    "\n",
    "- The average income of applicants is about 5,403 dollars. It has a large range of values from 150 to 81,000. It would be interesting to see if low applicant income implies a low loan amount.\n",
    "- The average co-applicant income is about 1,621 dollars which is much lower than the applicant's income. The 25th percentile value is 0 and the median value is 1,188 dollars which implies that the number of joint home loans is more than the number of non-joint loans.\n",
    "- The two columns `ApplicantIncome` and `CoapplicantIncome` give the same information i.e. income of applicants. It would be better to have that information in one column only. We can add these two columns to get the total income per application.\n",
    "- The 25th, 50th, and 75th percentile value of the loan term is 360 months i.e. 30 years. This implies that the majority of home loans in this dataset are for 30 years of term.\n",
    "- We can convert the scale of the loan term from months to years to make it easier to read.\n",
    "- The average loan amount is 141 dollars. It has a large range of values, which is to be expected. \n",
    "- The loan amount has zero values, which is absurd since loan amounts are meant to be non-zero. So we can treat 0's as missing values.\n",
    "\n",
    "Before imputing the missing values and 0's in `LoanAmount`, let's do some feature engineering and check the distributions, counts, and outliers for different variables."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Ad8XRsk2O7g7"
   },
   "source": [
    "## **Feature Engineering**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "Q3hc42zOO7g8"
   },
   "outputs": [],
   "source": [
    "#Converting the scale of loan term from months to years\n",
    "data['Loan_Amount_Term']=data['Loan_Amount_Term']/12"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "SB_h92fNO7g-"
   },
   "outputs": [],
   "source": [
    "#Adding the applicant and co-applicant income to get the total income per application\n",
    "data['total_income']=data['ApplicantIncome'] + data['CoapplicantIncome']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "id": "DHLiPJYzO7hD"
   },
   "outputs": [],
   "source": [
    "#Dropping the columns as we created a new column which captures the same information\n",
    "data.drop(columns=['ApplicantIncome', 'CoapplicantIncome'], inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "def histogram_boxplot(feature, figsize=(15,7), bins = None):\n",
    "    f, (ax_box, ax_hist) = plt.subplots(nrows = 2, # Number of rows of the subplot grid= 2\n",
    "                                           sharex = True, # x-axis will be shared among all subplots\n",
    "                                           gridspec_kw = {\"height_ratios\": (.25, .75)}, \n",
    "                                           figsize = figsize \n",
    "                                           ) # creating the 2 subplots\n",
    "    sns.boxplot(feature, ax=ax_box, showmeans=True, color='coral') # boxplot will be created and a star will indicate the mean value of the column\n",
    "    sns.distplot(feature, kde=F, ax=ax_hist, bins=bins) if bins else sns.distplot(feature, kde=False, ax=ax_hist) # For histogram\n",
    "    ax_hist.axvline(np.mean(feature), color='r', linestyle='--', label='Mean') # Add mean to the histogram\n",
    "    ax_hist.axvline(np.median(feature), color='k', linestyle='-', label='Median') # Add median to the histogram\n",
    "    ax_hist.legend()    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x504 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x504 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for col in ['LoanAmount', 'total_income']:\n",
    "    histogram_boxplot(data[col])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "8jJ8R6KkO7hF"
   },
   "source": [
    "## Univariate analysis "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "def perc_on_bar(z):\n",
    "    total = len(data[z]) # length of the column\n",
    "    plt.figure(figsize=(15,5))\n",
    "    ax = sns.countplot(data[z],palette='Paired',order = data[z].value_counts().index)\n",
    "    for p in ax.patches:\n",
    "        percentage = '{:.1f}%'.format(100 * p.get_height()/total) # percentage of each class of the category\n",
    "        x = p.get_x() + p.get_width() / 2 - 0.05 # width of the plot\n",
    "        y = p.get_y() + p.get_height()           # height of the plot\n",
    "        ax.annotate(percentage, (x, y), size = 12) \n",
    "    plt.show() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3sAAAE9CAYAAACyU3u7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3df5yWdZ3v8dcnQCXRAZoxEVAwfySg+WPwwaaLpu1K/tatFq2Nyo1KdLHUox5/4SbKnix3E83o9EM3T+iJVOpgRgaJbqlDjciPpRAURw1oEY1Ml8HP+WMuaJARR5l77pmL1/PxuB/3fX2v7/W9PxePx+34vn58r8hMJEmSJEnl8o5qFyBJkiRJ6niGPUmSJEkqIcOeJEmSJJWQYU+SJEmSSsiwJ0mSJEklZNiTJEmSpBLqWe0CtkdtbW0OGTKk2mVIkiRJUlXMnz//D5lZ19a6bh32hgwZQkNDQ7XLkCRJkqSqiIin32idl3FKkiRJUgkZ9iRJkiSphAx7etuOPfZYdtllF/r06UOfPn048MADN697+eWXOffcc6mtraWmpobRo0e3Ocarr77KOeecwz777MNuu+3GYYcdxn333bd5/TPPPMOoUaPo378/F1544Rbbjhkzxst4JUmSpDdQsbAXEbtExKMR8XhELIqIa4r2SRHxbEQ0Fq8TW21zWUQsi4ilEXFCpWpTx5k6dSrr169n/fr1LF26dHP7+PHjWbt2LUuWLGHt2rXceOONbW7f3NzM4MGD+cUvfsGLL77Il770JT760Y/y1FNPAXD99dczbtw4VqxYwT333LM53N15553su+++1NfXV3wfJUmSpO6okhO0vAocl5nrI6IX8FBEbDplc2Nm3tC6c0QMA8YCw4G9gJ9FxAGZubGCNaoCli5dysyZM2lqamL33XcH4Igjjmiz76677sqkSZM2L5988skMHTqU+fPnM2TIEFasWMHEiROpqalh5MiRLF++nAMOOIApU6YwZ86cztgdSZIkqVuq2Jm9bLG+WOxVvHIbm5wGTM/MVzNzBbAMOLJS9aljXHbZZdTW1nLUUUcxd+5cAB555BH22Wcfrr76amprazn44IOZMWNGu8ZbtWoVv/3tbxk+fDgAI0aMYPbs2axbt46GhgaGDRvGlVdeyQUXXEDfvn0rtVuSJElSt1fRe/YiokdENAKrgdmZ+Uix6ryIWBAR346IfkXbQOCZVps3FW3qov7lX/6F5cuX8+yzzzJ+/HhOOeUUnnzySZqamli4cCE1NTU899xzTJ06lXHjxrFkyZJtjrdhwwY+9rGPMW7cON773vcCLWFy3rx5HHPMMUyYMIENGzawYMECTjnlFM4++2xGjx7N1KlTO2N3JUmSpG4lMrd1sq2DviSiL3A3cD6wBvgDLWf5vgQMyMxPR8TNwC8z83vFNt8CZmXmjNeNNR4YD7D33nsf8fTTb/hYCXWyMWPGcNJJJ9Hc3Mwll1zCyy+/TM+eLVcKn3LKKXzwgx9k4sSJbW772muvcfbZZ/PSSy9x77330qtXrzb7jB49mltvvZXbb7+dvn37ctFFF3H44Ydz1113MWzYsIrunyRJktTVRMT8zGxzIotOmY0zM9cBc4ExmbkqMzdm5mvAN/nLpZpNwOBWmw0CnmtjrGmZWZ+Z9XV1bT4oXlUSEWQmhxxyyFvaLjM555xzWLVqFTNmzGgz6AFMmzaNUaNGMWLECJ544gnq6+vZaaedOPjgg1m4cGFH7IIkSZJUGpWcjbOuOKNHRPQGPgj8Z0QMaNXtDGDT/6XPBMZGxM4RMRTYH3i0UvVp+6xbt47777+fV155hebmZu644w4efPBBTjjhBEaPHs3ee+/N9ddfT3NzMw8//DBz587lhBPanmD185//PEuWLOFHP/oRvXv3brPP6tWrufnmmzdP5jJ06FDmzJnD+vXraWhoYN99963UrkqSJEndUiXP7A0A5kTEAuAxWu7Z+zHwvyLiiaL9A8AXADJzEXAXsBj4CTDBmTi7rg0bNnDFFVdQV1dHbW0tN910E/fccw8HHnggvXr14t5772XWrFnU1NTwmc98httvv33zfXjXXXcdH/rQhwB4+umn+cY3vkFjYyN77rnn5mf23XHHHVt830UXXcRVV11Fnz59gJZ7+X7+858zePBgTj31VB/BIEmSJL1Op9yzVyn19fXZ0Q/Vvnv+sg4dT+pIZxyxX7VLkCRJUhdS9Xv2JEmSJEmdy7AnSZIkSSVk2JMkSZKkEjLsSZIkSVIJGfYkSZIkqYQMe5IkSZJUQoY9SZIkSSohw54kSZIklZBhT5IkSZJKyLAnSZIkSSVk2JMkSZKkEjLsSZIkSVIJGfYkSZIkqYQMe5IkSZJUQoY9SZIkSSohw54kSZIklZBhT5IkSZJKyLAnSZIkSSVk2JMkSZKkEjLsSZIkSVIJGfYkSZIkqYQMe5IkSZJUQoY9SZIkSSohw54kSZIklZBhT5IkSZJKyLAnSZIkSSVk2JMkSZKkEjLsSZIkSVIJGfYkSZIkqYQqFvYiYpeIeDQiHo+IRRFxTdHePyJmR8Tvivd+rba5LCKWRcTSiDihUrVJkiRJUtlV8szeq8Bxmfk+4FBgTESMAi4FHsjM/YEHimUiYhgwFhgOjAFuiYgeFaxPkiRJkkqrYmEvW6wvFnsVrwROA24r2m8DTi8+nwZMz8xXM3MFsAw4slL1SZIkSVKZVfSevYjoERGNwGpgdmY+Arw7M58HKN73KLoPBJ5ptXlT0SZJkiRJeosqGvYyc2NmHgoMAo6MiBHb6B5tDbFVp4jxEdEQEQ1r1qzpqFIlSZIkqVQ6ZTbOzFwHzKXlXrxVETEAoHhfXXRrAga32mwQ8FwbY03LzPrMrK+rq6to3ZIkSZLUXVVyNs66iOhbfO4NfBD4T2AmMK7oNg64t/g8ExgbETtHxFBgf+DRStUnSZIkSWXWs4JjDwBuK2bUfAdwV2b+OCJ+CdwVEecAK4GPAGTmooi4C1gMNAMTMnNjBeuTJEmSpNKqWNjLzAXAYW20/xdw/BtsMxmYXKmaJEmSJGlH0Sn37EmSJEmSOpdhT5IkSZJKyLAnSZIkSSVk2JMkSZKkEjLsSZIkSVIJGfYkSZIkqYQMe5IkSZJUQoY9SZIkSSohw54kSZIklZBhT5IkSZJKyLAnSZIkSSVk2JMkSZKkEjLsSZIkSVIJGfYkSZIkqYQMe5IkSZJUQoY9SZIkSSohw54kSZIklZBhT5IkSZJKyLAnSZIkSSVk2JMkSZKkEjLsSZIkSVIJGfYkSZIkqYQMe5IkSZJUQoY9SZIkSSohw54kSZIklZBhT5IkSZJKyLAnSZIkSSVk2JMkSZKkEqpY2IuIwRExJyKWRMSiiJhYtE+KiGcjorF4ndhqm8siYllELI2IEypVmyRJkiSVXc8Kjt0MXJiZv46I3YD5ETG7WHdjZt7QunNEDAPGAsOBvYCfRcQBmbmxgjVKkiRJUilV7MxeZj6fmb8uPv8RWAIM3MYmpwHTM/PVzFwBLAOOrFR9kiRJklRmnXLPXkQMAQ4DHimazouIBRHx7YjoV7QNBJ5ptVkT2w6HkiRJkqQ3UPGwFxF9gBnABZn5EvB14D3AocDzwFc2dW1j82xjvPER0RARDWvWrKlQ1ZIkSZLUvVU07EVEL1qC3h2Z+UOAzFyVmRsz8zXgm/zlUs0mYHCrzQcBz71+zMyclpn1mVlfV1dXyfIlSZIkqduq5GycAXwLWJKZX23VPqBVtzOAhcXnmcDYiNg5IoYC+wOPVqo+SZIkSSqzSs7GeRTwD8ATEdFYtP1P4KyIOJSWSzSfAj4LkJmLIuIuYDEtM3lOcCZOSZIkSXp7Khb2MvMh2r4Pb9Y2tpkMTK5UTZIkSZK0o+iU2TglSZIkSZ3LsCdJkiRJJWTYkyRJkqQSMuxJkiRJUgkZ9iRJkiSphAx7kiRJklRChj1JkiRJKiHDniRJkiSVkGFPkiRJkkrIsCdJkiRJJWTYkyRJkqQSMuxJkiRJUgkZ9iRJkiSphAx7kiRJklRChj1JkiRJKiHDniRJkiSVkGFPkiRJkkrIsCdJkiRJJWTYkyRJkqQSMuxJkiRJUgkZ9iRJkiSphAx7kiRJklRChj1JkiRJKiHDniRJkiSVkGFPkiRJkkrIsCdJkiRJJWTYkyRJkqQSMuxJkiRJUgkZ9iRJkiSphCoW9iJicETMiYglEbEoIiYW7f0jYnZE/K5479dqm8siYllELI2IEypVmyRJkiSVXSXP7DUDF2bmQcAoYEJEDAMuBR7IzP2BB4plinVjgeHAGOCWiOhRwfokSZIkqbQqFvYy8/nM/HXx+Y/AEmAgcBpwW9HtNuD04vNpwPTMfDUzVwDLgCMrVZ8kSZIklVmn3LMXEUOAw4BHgHdn5vPQEgiBPYpuA4FnWm3WVLRJkiRJkt6iioe9iOgDzAAuyMyXttW1jbZsY7zxEdEQEQ1r1qzpqDIlSZIkqVQqGvYiohctQe+OzPxh0bwqIgYU6wcAq4v2JmBwq80HAc+9fszMnJaZ9ZlZX1dXV7niJUmSJKkba1fYi4gH2tP2uvUBfAtYkplfbbVqJjCu+DwOuLdV+9iI2DkihgL7A4+2pz5JkiRJ0pZ6bmtlROwCvBOoLR6RsOlSy92Bvd5k7KOAfwCeiIjGou1/AlOAuyLiHGAl8BGAzFwUEXcBi2mZyXNCZm5867skSZIkSdpm2AM+C1xAS7Cbz1/C3kvAzdvaMDMfou378ACOf4NtJgOT36QmSZIkSdKb2GbYy8x/A/4tIs7PzJs6qSZJkiRJ0nZ6szN7AGTmTRHxfmBI620y8/YK1SVJkiRJ2g7tCnsR8e/Ae4BGYNN9dAkY9iRJkiSpC2pX2APqgWGZudVz7yRJkiRJXU97n7O3ENizkoVIkiRJkjpOe8/s1QKLI+JR4NVNjZl5akWqkiRJkiRtl/aGvUmVLEKSJEmS1LHaOxvnLypdiCRJkiSp47R3Ns4/0jL7JsBOQC/gT5m5e6UKkyRJkiS9fe09s7db6+WIOB04siIVSZIkSZK2W3tn49xCZt4DHNfBtUiSJEmSOkh7L+M8s9XiO2h57p7P3JMkSZKkLqq9s3Ge0upzM/AUcFqHVyNJkiRJ6hDtvWfvU5UuRJIkSZLUcdp1z15EDIqIuyNidUSsiogZETGo0sVJkiRJkt6e9k7Q8h1gJrAXMBD4UdEmSZIkSeqC2hv26jLzO5nZXLy+C9RVsC5JkiRJ0nZob9j7Q0R8PCJ6FK+PA/9VycIkSZIkSW9fe8Pep4GPAr8Hngc+DDhpiyRJkiR1Ue199MKXgHGZ+QJARPQHbqAlBEqSJEmSupj2ntk7ZFPQA8jMtcBhlSlJkiRJkrS92hv23hER/TYtFGf22ntWUJIkSZLUydob2L4C/EdE/ABIWu7fm1yxqiRJkiRJ26VdYS8zb4+IBuA4IIAzM3NxRSuTJEmSJL1t7b4Uswh3BjxJkiRJ6gbae8+eJEmSJKkbMexJkiRJUgkZ9iRJkiSphCoW9iLi2xGxOiIWtmqbFBHPRkRj8Tqx1brLImJZRCyNiBMqVZckSZIk7QgqeWbvu8CYNtpvzMxDi9csgIgYBowFhhfb3BIRPSpYmyRJkiSVWsXCXmY+CKxtZ/fTgOmZ+WpmrgCWAUdWqjZJkiRJKrtq3LN3XkQsKC7z7Fe0DQSeadWnqWiTJEmSJL0NnR32vg68BzgUeB74StEebfTNtgaIiPER0RARDWvWrKlMlZIkSZLUzXVq2MvMVZm5MTNfA77JXy7VbAIGt+o6CHjuDcaYlpn1mVlfV1dX2YIlSZIkqZvq1LAXEQNaLZ4BbJqpcyYwNiJ2joihwP7Ao51ZmyRJkiSVSc9KDRwR3weOBWojogm4Gjg2Ig6l5RLNp4DPAmTmooi4C1gMNAMTMnNjpWqTJEmSpLKrWNjLzLPaaP7WNvpPBiZXqh5JkiRJ2pFUYzZOSZIkSVKFGfYkSZIkqYQMe5IkSZJUQoY9SZIkSSohw54kSZIklZBhT5IkSZJKyLAnSZIkSSVk2JMkSZKkEjLsSZIkSVIJGfYkSZIkqYQMe5IkSZJUQoY9SZIkSSohw54kSZIklZBhT5IkSZJKyLAnSZIkSSVk2JMkSZKkEjLsSZIkSVIJGfYkSZIkqYQMe5IkSZJUQoY9SZIkSSohw54kSZIklZBhT5IkSZJKyLAnSVUydepU6uvr2XnnnfnkJz+5xbqXX36Zc889l9raWmpqahg9evQ2x5o+fToHHXQQu+66K+95z3uYN28eAM888wyjRo2if//+XHjhhVtsM2bMGBoaGjp0nyRJUtdh2JOkKtlrr7244oor+PSnP73VuvHjx7N27VqWLFnC2rVrufHGG99wnNmzZ3PJJZfwne98hz/+8Y88+OCD7LvvvgBcf/31jBs3jhUrVnDPPfdsDnd33nkn++67L/X19ZXZOakTdOQBE4Df/e537LLLLnz84x/f3OYBE0ndWc9qFyBJO6ozzzwTgIaGBpqamja3L126lJkzZ9LU1MTuu+8OwBFHHPGG41x99dVcddVVjBo1CoCBAwduXrdixQomTpxITU0NI0eOZPny5RxwwAFMmTKFOXPmVGK3pE6z6YDJ/fffz5///Oct1o0fP57m5maWLFlC//79aWxsfNPxJkyYwMiRI7do23TA5Oyzz+bwww/nrLPOor6+3gMmkroFz+xJUhfzyCOPsM8++3D11VdTW1vLwQcfzIwZM9rsu3HjRhoaGlizZg377bcfgwYN4rzzztv8P74jRoxg9uzZrFu3joaGBoYNG8aVV17JBRdcQN++fTtzt6QOd+aZZ3L66afzrne9a4v2TQdMpk2bRl1dHT169NjmARNouRS6b9++HH/88Vu0r1ixguOOO26LAyYvvfQSU6ZM4brrruvwfZKkjmTYk6QupqmpiYULF1JTU8Nzzz3H1KlTGTduHEuWLNmq76pVq9iwYQM/+MEPmDdvHo2NjfzmN7/h2muvBeCyyy5j3rx5HHPMMUyYMIENGzawYMECTjnlFM4++2xGjx7N1KlTO3sXpYp6KwdMAF566SWuuuoqvvKVr2y1zgMmkrozw54kdTG9e/emV69eXHHFFey0004cc8wxfOADH+CnP/1pm30Bzj//fAYMGEBtbS1f/OIXmTVrFgD9+/fnzjvv5PHHH2fixImcf/753HTTTUyZMoURI0bws5/9jFtvvZXFixd36j5KlfRWDpgAXHnllZxzzjkMHjx4q3UeMJHUnXnPniR1MYcccki7+/br149BgwYREW/ad9q0aYwaNYoRI0bwxBNP8IUvfIGddtqJgw8+mIULFzJs2LDtKVvqMlofMOnZs+cWB0wOOuigLfo2Njbys5/9jN/85jdtjrXpgAnAa6+9xujRo7n11ls3HzD57ne/y+GHH85xxx3nb0hSl1OxM3sR8e2IWB0RC1u19Y+I2RHxu+K9X6t1l0XEsohYGhEnVKouSeoqmpubeeWVV9i4cSMbN27klVdeobm5mdGjR7P33ntz/fXX09zczMMPP8zcuXM54YS2/9P4qU99iptuuonVq1fzwgsv8K//+q+cfPLJW/RZvXo1N998M5MmTQJg6NChzJkzh/Xr19PQ0LB59k6pDN7KAZO5c+fy1FNPsffee7Pnnntyww03MGPGDA4//PCt+r7+gEl9ff0WB0wkqaup5GWc3wXGvK7tUuCBzNwfeKBYJiKGAWOB4cU2t0REjwrWJklVd+2119K7d2+mTJnC9773PXr37s21115Lr169uPfee5k1axY1NTV85jOf4fbbb+e9730vANdddx0f+tCHNo9z5ZVXMnLkSA444AAOOuggDjvsMC6//PItvuuiiy7iqquuok+fPkDLpWk///nPGTx4MKeeeqozCqpb6ogDJuPHj+fJJ5+ksbGRxsZGPve5z3HSSSdx//33b9HPAyaSuqPIzMoNHjEE+HFmjiiWlwLHZubzETEAmJuZB0bEZQCZeX3R735gUmb+clvj19fXZ0c/3+bu+cs6dDypI51xxH7VLqFdjrj49mqXILVp/pc/Ue0S1IEmTZrENddcs0Xb1VdfzaRJk1i0aBH/+I//yIIFC9hnn32YPHkyZ5xxBtBywGTevHncd999bY65bNkyvve9723R/olPfIJTTjmFj3zkI0DL8/c+/OEP89vf/pZPf/rTbU7uIkmdISLmZ2abR207O+yty8y+rda/kJn9ImIq8KvM/F7R/i3gvsz8wbbGN+xpR2PYk7ZPdwl7P150S7VLkNp08vBzq12CpNfZVtjrKrNxtjWzQJspNCLGR0RDRDSsWbOmwmVJkiRJUvfU2WFvVXH5JsX76qK9CWg93/Eg4Lm2BsjMaZlZn5n1dXV1FS1WkiRJkrqrzg57M4FxxedxwL2t2sdGxM4RMRTYH3i0k2uTJEmSpNKo2HP2IuL7wLFAbUQ0AVcDU4C7IuIcYCXwEYDMXBQRdwGLgWZgQmZurFRtkiRJklR2FQt7mXnWG6w6/g36TwYmV6oeSZIkSdqRdJUJWiRJkiRJHciwJ0mSJEklZNiTJEmSpBIy7EmSJElSCRn2JEmSJKmEDHuSJEmSVEKGPUmSJEkqIcOeJEmSJJWQYU+SJEmSSsiwJ0mSJEklZNiTJEmSpBIy7EmSJElSCRn2JEmSJKmEDHuSJEmSVEKGPUmSJEkqIcOeJEmSJJWQYU+SJEmSSsiwJ0mSJEklZNiTJEmSpBIy7EmSJElSCRn2JEmSJKmEDHuSJEmSVEKGPUmSJEkqIcOeJEmStIN66qmnOPHEE+nXrx977rkn5513Hs3NzVv1mz59OgceeCA1NTXssccejBs3jpdeemnz+gsuuIB+/frxV3/1Vzz77LOb2++44w4mTpzYKfuirRn2JEmSpB3Uueeeyx577MHzzz9PY2Mjv/jFL7jlllu26nfUUUfx8MMP8+KLL7J8+XKam5u54oorAHj00UeZP38+v//97zn66KO5/vrrAXjxxRe54YYb+Od//udO3Sf9hWFPkiRJ2kGtWLGCj370o+yyyy7sueeejBkzhkWLFm3Vb/DgwdTW1m5e7tGjB8uWLds8xtFHH83OO+/M8ccfz/LlywG4/PLLufjii6mpqemcndFWDHuSJEnSDmrixIlMnz6dl19+mWeffZb77ruPMWPGtNn3oYceoqamht12240ZM2ZwwQUXADB8+HDmzZvHn//8Zx544AGGDx9OQ0MDS5cu5eyzz+7M3dHrGPYkSZKkHdQxxxzDokWL2H333Rk0aBD19fWcfvrpbfY9+uijefHFF2lqauLiiy9myJAhAIwYMYK/+7u/Y9SoUaxcuZJLLrmEiRMn8rWvfY2vfe1rjB49mo997GOsW7euE/dMUKWwFxFPRcQTEdEYEQ1FW/+ImB0Rvyve+1WjNkmSJGlH8Nprr3HCCSdw5pln8qc//Yk//OEPvPDCC1xyySXb3G7gwIGMGTOGsWPHbm77whe+wOOPP86dd97JnXfeyV//9V/z2muvMW3aNB544AEOOuggpkyZUuld0utU88zeBzLz0MysL5YvBR7IzP2BB4plSZIkSRWwdu1annnmGc477zx23nln3vWud/GpT32KWbNmvem2zc3NPPnkk1u1r1q1im984xtcddVVLFy4kEMOOYRevXoxcuRIFixYUInd0DZ0pcs4TwNuKz7fBrR9/liSJEnSdqutrWXo0KF8/etfp7m5mXXr1nHbbbfxvve9b6u+d9xxBytXriQzefrpp7n88ss5/vjjt+r3xS9+kWuuuYZ3vvOdDB06lMcee4z169czd+5c9t13387YLbVSrbCXwE8jYn5EjC/a3p2ZzwMU73tUqTZJkiRph/DDH/6Qn/zkJ9TV1bHffvvRs2dPbrzxRlauXEmfPn1YuXIlAIsXL+b9738/ffr04aijjuLAAw/km9/85hZjzZkzh3Xr1nHGGWcAcOSRR3LSSScxePBg5syZw6WXeuFeZ4vM7PwvjdgrM5+LiD2A2cD5wMzM7NuqzwuZudV9e0U4HA+w9957H/H00093aG13z1/WoeNJHemMI/ardgntcsTFt1e7BKlN87/8iWqX0C4/XrT1M66kruDk4edWu4R2+cBNR1W7BKlNc85/uMPHjIj5rW6N20JVzuxl5nPF+2rgbuBIYFVEDAAo3le/wbbTMrM+M+vr6uo6q2RJkiRJ6lY6PexFxK4Rsdumz8DfAguBmcC4ots44N7Ork2SJEmSyqJnFb7z3cDdEbHp+/9PZv4kIh4D7oqIc4CVwEeqUJskSZIklUKnh73MXA5sNcVPZv4XsPWUPpIkSZKkt6wrPXpBkiRJktRBDHuSJEmSVEKGPUmSJEkqIcOeJEmSJJWQYU+SJEmSSsiwJ0mSJEklZNiTJEmSpBIy7EmSJElSCRn2JEmSJKmEDHuSJEmSVEKGPUmSJEkqIcOeJEmSJJWQYU+SJEmSSsiwJ0mSJEklZNiTJEmSpBIy7EmSJElSCRn2JEmSJKmEDHuSJEmSVEKGPUmSJEkqIcOeJEmSJJWQYU+SJEmSSsiwJ0mSJEklZNiTJEmSpBIy7EmSJElSCRn2JEmSJKmEDHuSJEmSVEKGPUmSJEkqIcOeJEmSJJWQYU+SJEmSSqjLhb2IGBMRSyNiWURcWu16JEmSJKk76lJhLyJ6ADcDHwKGAWdFxLDqViVJkiRJ3U+XCnvAkcCyzFyemf8NTAdOq3JNkiRJktTtdLWwNxB4ptVyU9EmSZIkSXoLela7gNeJNtpyiw4R44HxxeL6iFha8aq0PWqBP1S7CKkb8zfUgeKGcdUuQZ3P31CHmlDtAlQd/o46SPxTW3Fnu+3zRiu6WthrAga3Wh4EPNe6Q2ZOA6Z1ZlF6+yKiITPrq12H1F35G5K2j78hafv5O+q+utplnI8B+0fE0IjYCRgLzKxyTZIkSZLU7XSpM3uZ2RwR5wH3Az2Ab2fmoiqXJUmSJEndTpcKewCZOQuYVe061GG85FbaPv6GpO3jb0jafv6OuqnIzDfvJUmSJEnqVrraPXuSJEmSpA5g2FNFRMSYiFgaEcsi4tJq1yN1NxHx7YhYHRELq12L1B1FxPY9S9YAAAVBSURBVOCImBMRSyJiUURMrHZNUlcWEbtExKMR8Xjxm7mm2jVp+3kZpzpcRPQAfgv8DS2P03gMOCszF1e1MKkbiYjRwHrg9swcUe16pO4mIgYAAzLz1xGxGzAfON2/RVLbIiKAXTNzfUT0Ah4CJmbmr1r1eSozh1SrRr11ntlTJRwJLMvM5Zn538B04LQq1yR1K5n5ILC22nVI3VVmPp+Zvy4+/xFYAgysblVS15Ut1heLvYqXZ4W6OcOeKmEg8Eyr5Sb8AytJqpKIGAIcBjxS3Uqkri0iekREI7AamJ2Z/ma6uS736AWVQrTR5pEhSVKni4g+wAzggsx8qdr1SF1ZZm4EDo2IvsDdETGClquzPlJ02asIgwAPZ+aEatSp9jPsqRKagMGtlgcBz1WpFknSDqq472gGcEdm/rDa9UjdRWaui4i5wJjMnAxMhs337B1a1eL0lngZpyrhMWD/iBgaETsBY4GZVa5JkrQDKSab+BawJDO/Wu16pK4uIuqKM3pERG/gg8B/VrcqbS/DnjpcZjYD5wH303JD/F2Zuai6VUndS0R8H/glcGBENEXEOdWuSepmjgL+ATguIhqL14nVLkrqwgYAcyJiAS0H7mdn5o+rXJO2k49ekCRJkqQS8syeJEmSJJWQYU+SJEmSSsiwJ0mSJEklZNiTJEmSpBIy7EmSJElSCRn2JEmlEREbiyn2F0XE4xHxxYio2t+6iHgqImrf5ranR8Swjq5JkrTjMOxJksrkz5l5aGYOB/4GOBG4uso1vV2nA4Y9SdLbZtiTJJVSZq4GxgPnRYseEfHliHgsIhZExGcBIuLYiHgwIu6OiMURceums4ER8bcR8cuI+HVE/N+I6FO0PxUR1xTtT0TEe4v2d0XETyPiNxHxDSA21RMRH4+IR4szj9+IiB5F+/qImFycifxVRLw7It4PnAp8uej/noj4p6K+BRExvVP/MSVJ3ZJhT5JUWpm5nJa/dXsA5wAvZuZIYCTwmYgYWnQ9ErgQOBh4D3BmcfnlFcAHM/NwoAH4Yqvh/1C0fx24qGi7GngoMw8DZgJ7A0TEQcDfA0dl5qHARuBjxTa7Ar/KzPcBDwKfycz/KLa/uDhT+SRwKXBYZh4CfK7D/pEkSaXVs9oFSJJUYZvOrv0tcEhEfLhYrgH2B/4beLQIhkTE94GjgVdouYzy4YgA2An4Zatxf1i8zwfOLD6P3vQ5M/9fRLxQtB8PHAE8VozVG1hdrPtv4MetxvqbN9iPBcAdEXEPcE87912StAMz7EmSSisi9qXlLNpqWkLf+Zl5/+v6HAvk6zbNov/szDzrDYZ/tXjfyJZ/T18/FsVYt2XmZW2s25CZm7Z5/VitnURLmDwVuDIihmdm8xv0lSTJyzglSeUUEXXArcDUIkzdD3w+InoV6w+IiF2L7kdGxNDiXr2/Bx4CfgUcFRH7Ff3fGREHvMnXPkhxeWZEfAjoV7Q/AHw4IvYo1vWPiH3eZKw/ArsV/d8BDM7MOcD/APoCfdrz7yBJ2nF5Zk+SVCa9I6IR6AU0A/8OfLVY97+BIcCvo+VayjW0zHgJLZdnTqHlnr0Hgbsz87WI+CTw/YjYueh3BfDbbXz/NUX/XwO/AFYCZObiiLgC+GkR3DYAE4CntzHWdOCbEfFPwFjgWxFRQ8tZwhszc107/j0kSTuw+MuVI5Ik7XiKyzgvysyTq12LJEkdycs4JUmSJKmEPLMnSZIkSSXkmT1JkiRJKiHDniRJkiSVkGFPkiRJkkrIsCdJkiRJJWTYkyRJkqQSMuxJkiRJUgn9f+nfLM7ZY0OKAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3sAAAE+CAYAAAA0xwkVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAdP0lEQVR4nO3dfbReVX0n8O/PIASlkkQC4SWUIOiIoQreULQUpmBLlAqUUZta2qi0KdPggNVWmQpl7KCZVe0bL3VYvsVqG1gqiB0qDUis9g0vcgUSSokETQAJiIjREkiy54/7gDch4AXuk5t78vms9aznnH322c/v/HXzzd7nnGqtBQAAgG55zngXAAAAwNgT9gAAADpI2AMAAOggYQ8AAKCDhD0AAIAOEvYAAAA6aKfxLuDZ2GOPPdoBBxww3mUAAACMixtuuOH+1tr0rR2b0GHvgAMOyODg4HiXAQAAMC6q6ltPdswyTgAAgA4S9gAAADpI2IMJ6M4778zrXve6TJ06NTNmzMgZZ5yRDRs25JFHHskb3vCGHHDAAamqLFu2bFTj3X777Zk8eXJOPfXUx9tWr16dI488MtOmTcs73/nOzfrPnTvXEmoAgO2csAcT0O/+7u9mzz33zD333JOhoaF8+ctfzsUXX5wkOeqoo/KpT30qM2bMGPV4CxcuzJw5czZr+8AHPpD58+dn1apVueKKKx4Pd5deemkOPPDADAwMjN0FAQAw5oQ9mIBWrVqVN73pTZk8eXJmzJiRuXPnZvny5dl5551z1lln5aijjsqkSZNGNdaSJUsyZcqUHHfccU/4jWOPPTa777575syZkzvuuCMPPfRQFi1alPe///39uCwAAMaQsAcT0JlnnpklS5bkRz/6Ue666678/d//febOnfu0x3nooYdy7rnn5kMf+tATjs2ePTtLly7Ngw8+mMHBwRxyyCE555xzctZZZ2XKlCljcRkAAPSRsAcT0DHHHJPly5fnBS94Qfbbb78MDAzk5JNPftrjnHPOOTnttNMyc+bMJxw7++yz85WvfCXHHHNMFi5cmEcffTQ33XRTXv/61+fNb35zjj766Fx44YVjcTkAAPRBX8NeVd1ZVTdX1VBVDfbaplXV0qq6vfc9dUT/s6tqZVXdVlXH97M2mKg2bdqU448/Pqecckp++MMf5v7778/3vve9vPvd735a4wwNDeWaa67JO97xjq0enzZtWi699NJ84xvfyJlnnpm3v/3tueCCC7Jo0aLMnj0711xzTT784Q9nxYoVY3FZAACMsW0xs/cLrbVXtNYee5rDe5Jc21o7OMm1vf1U1SFJ5iV5WZK5SS6uqtHddAQ7kAceeCCrV6/OGWeckV122SUvfOEL89a3vjVXXXXV0xpn2bJlufPOO7P//vtnxowZ+eAHP5jPfvazOfzww5/Q95JLLsmRRx6Z2bNn5+abb87AwEB23nnnHHroobnlllvG6tIAABhD47GM86Qki3vbi5OcPKJ9SWttfWttVZKVSY4Yh/pgu7bHHntk1qxZ+au/+qts2LAhDz74YBYvXpyXv/zlSZL169fn4YcfTpI88sgjefjhh9Nae8I4CxYsyDe/+c0MDQ1laGgop59+ek444YRcffXVm/Vbu3ZtLrroopx33nlJklmzZuW6667LunXrMjg4mAMPPLC/FwwAwDPS77DXkvxDVd1QVQt6bXu11u5Jkt73nr32fZOsHnHuml4bsIXPfe5z+eIXv5jp06fnoIMOyk477ZQ/+7M/S5K85CUvya677pq77rorxx9/fHbdddd861vfSpK8//3vz2tf+9okyfOe97zMmDHj8c9uu+2WyZMnZ/r06Zv91rve9a6ce+652W233ZIM38v3pS99KTNnzsyJJ57oFQwAANup2tr/+I/Z4FX7tNburqo9kyxN8vYkV7bWpozo873W2tSquijJv7TWPtVr/2iSq1prn91izAVJFiTJ/vvv/8rH/hG7Pbv8hpXjXQLAhPMrrzxovEsAgO1eVd0w4pa5zfR1Zq+1dnfve22SyzO8LPPeqtq7V9jeSdb2uq9JMvKRgPsluXsrY17SWhtorQ1sOQMBAADAsL6Fvap6flX91GPbSX4pyS1Jrkwyv9dtfpLP97avTDKvqnapqllJDk5yfb/qAwAA6LKd+jj2Xkkur6rHfudvWmtfrKqvJbmsqk5L8u0kb0yS1tryqrosyYokG5IsbK1t7GN9AAAAndW3sNdauyPJy7fS/t0kxz3JOecnOb9fNQEAAOwoxuPVCwAAAPSZsAcAANBBwh4AAEAHCXsAAAAdJOwBAAB0kLAHAADQQcIeAABABwl7AAAAHSTsAQAAdJCwBwAA0EHCHgAAQAcJewAAAB0k7AEAAHSQsAcAANBBwh4AAEAHCXsAAAAdJOwBAAB0kLAHAADQQcIeAABABwl7AAAAHSTsAQAAdJCwBwAA0EHCHgAAQAcJewAAAB0k7AEAAHSQsAcAANBBwh4AAEAHCXsAAAAdJOwBAAB0kLAHAADQQcIeAABABwl7AAAAHSTsAQAAdJCwBwAA0EHCHgAAQAcJewAAAB0k7AEAAHSQsAcAANBBwh4AAEAHCXsAAAAdJOwBAAB0kLAHAADQQcIeAABAB/U97FXVpKq6sar+rrc/raqWVtXtve+pI/qeXVUrq+q2qjq+37UBAAB01baY2Tszya0j9t+T5NrW2sFJru3tp6oOSTIvycuSzE1ycVVN2gb1AQAAdE5fw15V7ZfkhCQfGdF8UpLFve3FSU4e0b6ktba+tbYqycokR/SzPgAAgK7q98zenyf5gySbRrTt1Vq7J0l633v22vdNsnpEvzW9NgAAAJ6mvoW9qvrlJGtbazeM9pSttLWtjLugqgaravC+++57VjUCAAB0VT9n9n4uyYlVdWeSJUmOrapPJbm3qvZOkt732l7/NUlmjjh/vyR3bzloa+2S1tpAa21g+vTpfSwfAABg4upb2Gutnd1a26+1dkCGH7zypdbaqUmuTDK/121+ks/3tq9MMq+qdqmqWUkOTnJ9v+oDAADosp3G4TcXJbmsqk5L8u0kb0yS1tryqrosyYokG5IsbK1tHIf6AAAAJrxtEvZaa8uSLOttfzfJcU/S7/wk52+LmgAAALpsW7xnDwAAgG1M2AMAAOggYQ8AAKCDhD0AAIAOEvYAAAA6SNgDAADoIGEPAACgg4Q9AACADhL2AAAAOkjYAwAA6CBhDwAAoIOEPQAAgA4S9gAAADpI2AMAAOggYQ8AAKCDhD0AAIAOEvYAAAA6SNgDAADoIGEPAACgg4Q9AACADhL2AAAAOkjYAwAA6CBhDwAAoIOEPQAAgA4S9gAAADpI2AMAAOggYQ8AAKCDhD0AAIAOEvYAAAA6SNgDAADoIGEPAACgg4Q9AACADhL2AAAAOkjYAwAA6CBhDwAAoIOEPQAAgA4S9gAAADpI2AMAAOggYQ8AAKCDhD0AAIAOEvYAAAA6SNgDAADoIGEPAACgg/oW9qpqclVdX1XfqKrlVfW/eu3TqmppVd3e+5464pyzq2plVd1WVcf3qzYAAICu6+fM3vokx7bWXp7kFUnmVtWRSd6T5NrW2sFJru3tp6oOSTIvycuSzE1ycVVN6mN9AAAAndW3sNeGrevtPrf3aUlOSrK41744ycm97ZOSLGmtrW+trUqyMskR/aoPAACgy/p6z15VTaqqoSRrkyxtrf1bkr1aa/ckSe97z173fZOsHnH6ml4bAAAAT1Nfw15rbWNr7RVJ9ktyRFXNforutbUhntCpakFVDVbV4H333TdWpQIAAHTKNnkaZ2vtwSTLMnwv3r1VtXeS9L7X9rqtSTJzxGn7Jbl7K2Nd0lobaK0NTJ8+va91AwAATFT9fBrn9Kqa0tveNclrkvx7kiuTzO91m5/k873tK5PMq6pdqmpWkoOTXN+v+gAAALpspz6OvXeSxb0naj4nyWWttb+rqn9JcllVnZbk20nemCStteVVdVmSFUk2JFnYWtvYx/oAAAA6q29hr7V2U5LDttL+3STHPck55yc5v181AQAA7Ci2yT17AAAAbFvCHgAAQAeNKuxV1bWjaQMAAGD78JT37FXV5CTPS7JHVU3Nj9+F94Ik+/S5NgAAAJ6hn/SAlt9JclaGg90N+XHYeyjJRX2sCwAAgGfhKcNea+0vkvxFVb29tXbBNqoJAACAZ2lUr15orV1QVa9OcsDIc1prn+xTXQAAADwLowp7VfXXSV6UZCjJYy86b0mEPQAAgO3QaF+qPpDkkNZa62cxAAAAjI3RvmfvliQz+lkIAAAAY2e0M3t7JFlRVdcnWf9YY2vtxL5UBQAAwLMy2rB3Xj+LAAAAYGyN9mmcX+53IQAAAIyd0T6N8wcZfvpmkuyc5LlJfthae0G/CgMAAOCZG+3M3k+N3K+qk5Mc0ZeKAAAAeNZG+zTOzbTWrkhy7BjXAgAAwBgZ7TLOU0bsPifD793zzj0AAIDt1Gifxvn6EdsbktyZ5KQxrwYAAIAxMdp79t7a70IAAAAYO6O6Z6+q9quqy6tqbVXdW1Wfrar9+l0cAAAAz8xoH9Dy8SRXJtknyb5JvtBrAwAAYDs02rA3vbX28dbaht7nE0mm97EuAAAAnoXRhr37q+rUqprU+5ya5Lv9LAwAAIBnbrRh721J3pTkO0nuSfKGJB7aAgAAsJ0a7asX/jjJ/Nba95KkqqYl+WCGQyAAAADbmdHO7P3MY0EvSVprDyQ5rD8lAQAA8GyNNuw9p6qmPrbTm9kb7awgAAAA29hoA9uHkvxzVX0mScvw/Xvn960qAAAAnpVRhb3W2ierajDJsUkqySmttRV9rQwAAIBnbNRLMXvhTsADAACYAEZ7zx4AAAATiLAHAADQQcIeAABABwl7AAAAHSTsAQAAdJCwBwAA0EHCHgAAQAcJewAAAB0k7AEAAHSQsAcAANBBwh4AAEAHCXsAAAAd1LewV1Uzq+q6qrq1qpZX1Zm99mlVtbSqbu99Tx1xztlVtbKqbquq4/tVGwAAQNf1c2ZvQ5J3ttZemuTIJAur6pAk70lybWvt4CTX9vbTOzYvycuSzE1ycVVN6mN9AAAAndW3sNdau6e19vXe9g+S3Jpk3yQnJVnc67Y4ycm97ZOSLGmtrW+trUqyMskR/aoPAACgy7bJPXtVdUCSw5L8W5K9Wmv3JMOBMMmevW77Jlk94rQ1vTYAAACepr6HvaraLclnk5zVWnvoqbpupa1tZbwFVTVYVYP33XffWJUJAADQKX0Ne1X13AwHvU+31j7Xa763qvbuHd87ydpe+5okM0ecvl+Su7ccs7V2SWttoLU2MH369P4VDwAAMIH182mcleSjSW5trf3piENXJpnf256f5PMj2udV1S5VNSvJwUmu71d9AAAAXbZTH8f+uSS/keTmqhrqtf3PJIuSXFZVpyX5dpI3JklrbXlVXZZkRYaf5Lmwtbaxj/UBAAB0Vt/CXmvtq9n6fXhJctyTnHN+kvP7VRMAAMCOYps8jRMAAIBtS9gDAADoIGEPAACgg4Q9AACADhL2AAAAOkjYAwAA6CBhDwAAoIOEPQAAgA4S9gAAADpI2AMAAOggYQ8AAKCDhD0AAIAOEvYAAAA6SNgDAADoIGEPAACgg4Q9AACADhL2AAAAOkjYAwAA6CBhDwAAoIOEPQAAgA4S9gAAADpI2AMAAOggYQ8AAKCDhD0AAIAOEvYAAAA6SNgDAADoIGEPAACgg4Q9AACADhL2AAAAOkjYAwAA6CBhDwAAoIOEPQAAgA4S9gAAADpI2AMAAOggYQ8AAKCDhD0AAIAOEvYAAAA6SNgDAADoIGEPAACgg4Q9AACADhL2AAAAOkjYAwAA6CBhDwAAoIP6Fvaq6mNVtbaqbhnRNq2qllbV7b3vqSOOnV1VK6vqtqo6vl91AQAA7Aj6ObP3iSRzt2h7T5JrW2sHJ7m2t5+qOiTJvCQv651zcVVN6mNtAAAAnda3sNda+8ckD2zRfFKSxb3txUlOHtG+pLW2vrW2KsnKJEf0qzYAAICu29b37O3VWrsnSXrfe/ba902yekS/Nb02AAAAnoHt5QEttZW2ttWOVQuqarCqBu+7774+lwUAADAxbeuwd29V7Z0kve+1vfY1SWaO6Ldfkru3NkBr7ZLW2kBrbWD69Ol9LRYAgM1deOGFGRgYyC677JK3vOUtj7evWLEiAwMDmTp1aqZOnZrXvOY1WbFixU8c7/bbb8/kyZNz6qmnPt62evXqHHnkkZk2bVre+c53btZ/7ty5GRwcHLPrgS7b1mHvyiTze9vzk3x+RPu8qtqlqmYlOTjJ9du4NgAAfoJ99tkn733ve/O2t73tCe2f+cxn8sADD+T+++/PiSeemHnz5v3E8RYuXJg5c+Zs1vaBD3wg8+fPz6pVq3LFFVc8Hu4uvfTSHHjggRkYGBi7C4IO26lfA1fV3yb5r0n2qKo1Sf4oyaIkl1XVaUm+neSNSdJaW15VlyVZkWRDkoWttY39qg0AgGfmlFNOSZIMDg5mzZo1j7dPmTIlU6ZMSZK01jJp0qSsXLnyKcdasmRJpkyZkle/+tWb9V21alXOPPPM7L777pkzZ07uuOOOvPjFL86iRYty3XXX9eGqoJv6FvZaa7/2JIeOe5L+5yc5v1/1AADQf1OmTMm6deuyadOmvO9973vSfg899FDOPffcXHvttfnoRz+62bHZs2dn6dKl2WuvvTI4OJj3vve9Oeecc3LWWWc9HiiBn2x7eUALAAAd8OCDD+b73/9+Lrzwwhx22GFP2u+cc87JaaedlpkzZz7h2Nlnn52vfOUrOeaYY7Jw4cI8+uijuemmm/L6178+b37zm3P00Ufnwgsv7OdlQCf0bWYPAIAd0/Of//ycfvrpmT59em699dbsueeemx0fGhrKNddckxtvvHGr50+bNi2XXnppkmTTpk05+uij8+EPfziLFi3K7Nmz84lPfCKHH354jj322BxyyCF9vx6YqIQ9AADG3KZNm/KjH/0od9111xPC3rJly3LnnXdm//33T5KsW7cuGzduzIoVK/L1r399s76XXHJJjjzyyMyePTs333xz3vGOd2TnnXfOoYcemltuuUXYg6dgGScAAKO2YcOGPPzww9m4cWM2btyYhx9+OBs2bMjSpUtz4403ZuPGjXnooYfye7/3e5k6dWpe+tKXPmGMBQsW5Jvf/GaGhoYyNDSU008/PSeccEKuvvrqzfqtXbs2F110Uc4777wkyaxZs3Lddddl3bp1GRwczIEHHrgtLhkmrGptq+8unxAGBgbaRHjPyuU3PPWTqAB4ol955UHjXUKnvPL3PzneJdARd//T5fnOv1yxWduMV52cXffYN3f/0+fy6A8eyHN22jnPmzEr+xz9xjxv+vDs3Xf+9QtZt+a2HPSGd211zPUP3ptZJ5y+WfudV/3f7P6iwzL1JUckSR556Lu548oLs/5738kLZx+d/X7hyZ4HCM/eDX/ym+NdwqhU1Q2tta2+j0TY2waEPYCnT9gbW8IewNPThbBnGScAAEAHCXsAAAAdJOwBAAB0kLAHAADQQcIeAABABwl7AAAAHSTsAQAAdJCwBwAA0EHCHgAAQAcJewAAAB0k7AEAAHSQsAcAANBBwh4AAEAHCXsAAAAdJOwBAAB0kLAHAADQQcIeAABABwl7AAAAHSTsAQAAdJCwBwAA0EHCHgAAQAcJewAAAB0k7AEAAHSQsAcAANBBwh4AAEAHCXsAAAAdJOwBAAB0kLAHAADQQcIeAABABwl7AAAAHSTsAQAAdJCwBwAA0EHCHgAAQAcJewAAAB0k7AEAAHSQsAcAANBBwh4AAEAHbXdhr6rmVtVtVbWyqt4z3vUAAABMRNtV2KuqSUkuSvLaJIck+bWqOmR8qwIAAJh4tquwl+SIJCtba3e01h5JsiTJSeNcEwAAwISzvYW9fZOsHrG/ptcGAADA07DTeBewhdpKW9usQ9WCJAt6u+uq6ra+VwXdtkeS+8e7CAB2WP4OsV2qD84f7xJG66ef7MD2FvbWJJk5Yn+/JHeP7NBauyTJJduyKOiyqhpsrQ2Mdx0A7Jj8HYL+2d6WcX4tycFVNauqdk4yL8mV41wTAADAhLNdzey11jZU1RlJrk4yKcnHWmvLx7ksAACACWe7CntJ0lq7KslV410H7EAsiwZgPPk7BH1SrbWf3AsAAIAJZXu7Zw8AAIAxIOzBDqKqWlV9aMT+u6rqvHEsCYAdQA37alW9dkTbm6rqi+NZF+wIhD3YcaxPckpV7THehQCw42jD9wydnuRPq2pyVT0/yflJFo5vZdB9wh7sODZk+Cb4d2x5oKp+uqquraqbet/7b/vyAOiq1totSb6Q5N1J/ijJp5L8YVV9rapurKqTkqSqXlZV11fVUO9v0sHjWDZMeB7QAjuIqlqXZJ8kNyV5eZLfTrJba+28qvpCks+01hZX1duSnNhaO3kcywWgY3ozel9P8kiSv0uyvLX2qaqakuT6JIclWZTkX1trn+69c3lSa+0/x61omOCEPdhBVNW61tpuVfW+JI8m+c/8OOzdn2Tv1tqjVfXcJPe01iz3BGBM9f4GrUvypiSTM7zqJEmmJTk+w4HvD5N8MsnnWmu3j0ed0BXb3Xv2gL778wz/z+rHn6KP/wUCoB829T6V5L+11m7b4vitVfVvSU5IcnVV/VZr7UvbukjoCvfswQ6mtfZAksuSnDai+Z+TzOtt/3qSr27rugDYoVyd5O1VVUlSVYf1vg9Mckdr7S+TXJnkZ8avRJj4hD3YMX0oychlmv8jyVur6qYkv5HkzHGpCoAdxR8neW6Sm6rqlt5+kvxqkluqaijJf8nwck7gGXLPHgAAQAeZ2QMAAOggYQ8AAKCDhD0AAIAOEvYAAAA6SNgDAADoIGEPAACgg4Q9ACacqvrDqlpeVTdV1VBV/exT9P1EVb2ht/3zvfOGqmrXrfQ9oKr+s3f8sc9vjlHN68ZinKcY//HrBIAk2Wm8CwCAp6OqXpXkl5Mc3lpbX1V7JNl5lKf/epIPttY+/hR9vtlae8WzrRMAxpuZPQAmmr2T3N9aW58krbX7W2t3V9Urq+rLVXVDVV1dVXuPPKmqfivJm5KcW1Wffro/WlXrqur/9Ma/pqqOqKplVXVHVZ3Y6/OWqvp8VX2xqm6rqj/ayjhVVX9SVbdU1c1V9au99r+uqpNG9Pt0VZ1YVZN6/b/Wm8n8nRHjXFhVK6rq/yXZ8+leEwDdJuwBMNH8Q5KZVfUfVXVxVR1TVc9NckGSN7TWXpnkY0nOH3lSa+0jSa5M8vuttV9/ivFftMUyzp/vtT8/ybLe+D9I8r+T/GKSX0nyvhHnH5HhGcRXJHljVQ1sMf4pvWMvT/KaJH/SC6YfSfLWJKmq3ZO8OslVSU5L8v3W2pwkc5L8dlXN6v3uS5IcmuS3e/0B4HGWcQIwobTW1lXVK5P8fJJfSHJphoPX7CRLqypJJiW55xn+xJMt43wkyRd72zcnWd9ae7Sqbk5ywIh+S1tr302SqvpckqOSDI44flSSv22tbUxyb1V9Ocmc1tqVVXVRVe2Z4UD42dbahqr6pSQ/M+J+vN2THJzk6BHj3F1VX3qG1wtARwl7AEw4vYCzLMmyXthamGR5a+1VffzZR1trrbe9Kcljy0g3VdXIv6dti/O23K+n+I2/zvCs4LwkbxvR/+2ttas3G6TqdVsZGwAeZxknABNKVb2kqg4e0fSKJLcmmd57eEuq6rlV9bJxKTD5xaqa1nva58lJ/mmL4/+Y5Fd79+JNz/AM3fW9Y59IclaStNaW99quTvLfe0tVU1Uvrqrn98aZ1xtn7wzPcgLA48zsATDR7JbkgqqakmRDkpVJFiS5JMlf9u532ynJnydZ/qSjPLkXVdXQiP2Ptdb+8mmc/9UMz9AdlORvWmuDWxy/PMmrknwjwzNzf9Ba+06StNburapbk1wxov9HMrxM9Os1vEb1vgyHyMuTHJvhJaX/keTLT6NGAHYA9eMVKQDAs1FVb0ky0Fo74xme/7wMh7fDW2vfH8vaANjxWMYJANuBqnpNkn9PcoGgB8BYMLMHwA6nqg7N8FLLkda31n52POoBgH4Q9gAAADrIMk4AAIAOEvYAAAA6SNgDAADoIGEPAACgg4Q9AACADvr/7WTuj/7cf20AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cat_col= ['Gender', 'Married', 'Dependents', 'Education', 'Self_Employed', 'Credit_History','Property_Area', 'Loan_Status']\n",
    "\n",
    "for col in cat_col:\n",
    "    perc_on_bar(col)    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "id": "AA2J6KT6O7hG",
    "outputId": "5d5fa96a-ea9d-4765-daa1-35a468a35f3d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Male      0.813644\n",
      "Female    0.186356\n",
      "Name: Gender, dtype: float64\n",
      "****************************************\n",
      "Yes    0.651391\n",
      "No     0.348609\n",
      "Name: Married, dtype: float64\n",
      "****************************************\n",
      "0     0.575960\n",
      "1     0.170284\n",
      "2     0.168614\n",
      "3+    0.085142\n",
      "Name: Dependents, dtype: float64\n",
      "****************************************\n",
      "Graduate        0.781759\n",
      "Not Graduate    0.218241\n",
      "Name: Education, dtype: float64\n",
      "****************************************\n",
      "No     0.859107\n",
      "Yes    0.140893\n",
      "Name: Self_Employed, dtype: float64\n",
      "****************************************\n",
      "1.0    0.842199\n",
      "0.0    0.157801\n",
      "Name: Credit_History, dtype: float64\n",
      "****************************************\n",
      "Semiurban    0.379479\n",
      "Urban        0.328990\n",
      "Rural        0.291531\n",
      "Name: Property_Area, dtype: float64\n",
      "****************************************\n",
      "1    0.687296\n",
      "0    0.312704\n",
      "Name: Loan_Status, dtype: float64\n",
      "****************************************\n"
     ]
    }
   ],
   "source": [
    "#Checking the percentage of each category for categorical variables\n",
    "for col in cat_col:\n",
    "    print(data[col].value_counts(normalize=True))  #The parameter normalize=True gives the percentage of each category\n",
    "    print('*'*40)                                  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "UpuwOACuO7hI"
   },
   "source": [
    "**Observations:**\n",
    "\n",
    "- Percentage of male customers (approx 81%) is more than the percentage of female customers (approx 19%).\n",
    "- About 65% of customers are married. This indicates that married people apply more for the home loan as compared to non-married people.\n",
    "- The majority of customers have 0 or 1 dependents. Some of these entries are 3+ which means 3 or more dependents. Since we don't know the actual number, we can treat this as a separate category.\n",
    "- About 78% of customers are graduate. It looks like the majority of customers are educated.\n",
    "- Only ~14% of customers are self-employed. This indicates that salaried people apply more for the home loan.\n",
    "- The majority of customers, about 84%, have credit history as per the required guidelines.\n",
    "- The count of observations is approximately evenly distributed among Urban, Semi-Urban, and Rural areas. This implies that the company has a presence in all kinds of areas.\n",
    "- The number of home loans approved is comparatively higher than the number of those not approved. About 69% of applied loans get approved. Although we have limited data, the reason for this could either be a fairly lenient attitude from the company towards approving loans OR that many of the customers applying for these home loans are actually eligible for them."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "frx2ZYaeO7ha"
   },
   "source": [
    "#### Imputing missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "id": "8fW0Z4u9O7hc"
   },
   "outputs": [],
   "source": [
    "#Imputing missing values with mode for the categorical variables \n",
    "for col in ['Gender', 'Married', 'Dependents', 'Self_Employed', 'Loan_Amount_Term', 'Credit_History']:\n",
    "    data[col].fillna(value=data[col].mode()[0], inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "id": "vODNuyMLO7hd"
   },
   "outputs": [],
   "source": [
    "#Replacing 0's with null values in loan amount \n",
    "data.LoanAmount.replace(0, np.nan, inplace=True)\n",
    "\n",
    "#Imputing null values in loan amount with the median\n",
    "data.LoanAmount.fillna(value=data.LoanAmount.median(), inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Bivariate Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x900 with 20 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#### pairwise-plot\n",
    "sns.pairplot(data, diag_kind='kde', corner=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "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": [
    "ax=sns.regplot(x=\"LoanAmount\",y=\"total_income\",data=data)\n",
    "ax.set(xlabel=\"LoanAmount\", ylabel=\"total_income\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observations:**\n",
    "- The plot shows that the loan amount is positively correlated with total income. This implies that the loan amount for higher-income applicants is progressively higher.\n",
    "- There are some outliers visible, showing applicants with low income having been given loans of a higher amount."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Let's check the relationship of total income with loan status**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "def boxplot(z):\n",
    "    plt.figure(figsize=(15, 5)) \n",
    "    g=sns.boxplot(x=z, y=data['total_income']) \n",
    "    g.set(ylabel=\"Total Income\")\n",
    "    plt.show()\n",
    "    plt.figure(figsize=(15, 5))\n",
    "    plt.title('Without Outliers')\n",
    "    sns.boxplot(x=z, y=data['total_income'],showfliers=False) #turning off the outliers\n",
    "    plt.show()    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "boxplot(data['Loan_Status'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observations:**\n",
    "- Except for some outliers, there is no major difference between the income of those customers who are eligible versus those who are not. \n",
    "- This implies that income alone might not be a good indicator of a customer's loan eligibility."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "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 relationship between education and loan status**\n",
    "sns.countplot(x=data.Education, hue=data.Loan_Status)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observations:**\n",
    "- We can see that graduate customers are more likely to get loans.\n",
    "- This might be because, in general, graduate customers have higher income when compared to non-graduates. Let's check if this is true."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4kAAAFNCAYAAACtyCreAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de7xddX3n/9fboBCqqEikmIBBw9gCRYTIoNaWyrTS/lSohYqDgh1m0jJMmtqxCrVTtW0cL/UCseBQLwRFEVFH7EgVsSpWBAMCMVzKsQgkRAjiBeWiiZ/fH/t7lpvDSXJOcvbZycnr+Xjsx177s9Z3rc8OPNi8812XVBWSJEmSJAE8atgNSJIkSZK2HYZESZIkSVLHkChJkiRJ6hgSJUmSJEkdQ6IkSZIkqWNIlCRJkiR1DImSpBktyfOT3LyJ9fOTVJKdprOvQUpyRJLVfZ9XJTliiC1JkrYjhkRJ0nYlyelJPjumdstGasdX1eVV9Yy++neS/Kdp6vXcJH+3mW2S5C9avw8kuT3JW5LsPInjVJIFG1tfVQdU1Zcm0bokaQdmSJQkbW++AjwvySyAJL8MPBo4ZExtQdt2W3cmsAg4EXgc8LvAC4ALh9kUwEyaXZUkTZwhUZK0vfkGvVB4cPv8G8C/ADePqX27qu7sP/UyyYeAfYDPJPlxktf27feENot3T5LXjxaT7Jzk3UnubK93j87yJXlVkq/2Nzc6q5dkEXAC8Np2rM+M/SJJ9gP+O3BCVV1RVeurahXwB8BRSV7QtvtSkv/aN647bpLRIHxdO87LxjlON3ua5FFJTkvy7STfS3Jhkt3butFTb09OcjvwxSS7JPlw2/YHSb6RZM9N/hOSJG3XDImSpO1KVf0UuJJeEKS9Xw58dUztEbOIVfVK4HbgxVX12Kp6W9/qXweeARwJ/HWSX2311wOH0wugzwQOA/5qAn2eA5wPvK0d68XjbHYksLqqrhoz9g7g68BvT+A4o9/5me04H9vMkD8FjgF+E3gK8H3gH8Zs85vArwIvBE4CHg/sDTwJ+BPggc31JUnafhkSJUnboy/zi0D4fHoh8fIxtS9Pcp9vqqoHquo64Dp6gRB6s4F/U1V3V9U64E3AK7em+T57AGs3sm5tWz/V/hh4fVWtrqqHgDcCx445tfSNVfWTqnoA+Bm9cLigqjZU1dVV9aMB9CVJ2kYYEiVJ26OvAL+e5InAnKq6Bfga8NxWO5DJX4/43b7l+4HHtuWnALf1rbut1abCPcBeG1m3V1s/1Z4KfKqdOvoD4EZgA9B/CukdfcsfAj4HXNBOt31bkkcPoC9J0jbCkChJ2h5dQe8UyEXAvwK02a07W+3Oqrp1I2Nrkse6k16wGrVPqwH8BNh1dEW7Yc5kjvVFYO8kh/UXk+xN7xTXy8Y7DjD2OJNxB/C7VfWEvtcuVbVmvL6r6mdV9aaq2h94LvAiejfZkSTNUIZESdJ2p50GuQL4c3qnmY76aqttahbxLuBpkzjcR4G/SjInyR7AXwMfbuuuAw5IcnCSXeidujnhY1XVvwHvBc5PcniSWUkOAD4BfKGqvtA2vRZ4aZJd26MuTt6K7/ReYGmSpwK073X0xjZO8ltJfq3dOfZH9E4/3TDBY0mStkOGREnS9urLwJPpBcNRl7fapkLi/6YX+n6Q5DUTOM7f0Quk1wMrgWtabTTk/Q3wBeCWMb0AvB/Yvx3r/25k//8DeB+94Plj4J+BL9G7w+modwE/pRcGl9O7IU6/NwLL23H+cDPf5wzgYuDzSe6jd4Oc/7iJ7X8ZuIheQLyR3p/7hzexvSRpO5eqyZ51I0mSJEmaqZxJlCRJkiR1DImSJEmSpI4hUZIkSZLUMSRKkiRJkjqGREmSJElSZ6dhNzAMe+yxR82fP3/YbUiSJEnSUFx99dX3VNWc8dbtkCFx/vz5rFixYthtSJIkSdJQJLltY+s83VSSJEmS1DEkSpIkSZI6hkRJkiRJUseQKEmSJEnqGBIlSZIkSR1DoiRJkiSpY0iUJEmSJHUMiZIkSZKkjiFRkiRJktTZadgNSNNh2bJljIyMDLuNGWHNmjUAzJ07d8idzBwLFixg8eLFw25DkiQJMCRKmqQHHnhg2C1IkiRpgAyJ2iE4SzN1lixZAsAZZ5wx5E4kSZI0CF6TKEmSJEnqDDQkJvlAkruTfGucda9JUkn26KudnmQkyc1JXthXPzTJyrbuzCRp9Z2TfKzVr0wyf5DfR5IkSZJmukHPJJ4LHDW2mGRv4LeB2/tq+wPHAwe0MWclmdVWnw0sAvZrr9F9ngx8v6oWAO8C3jqQbyFJkiRJO4iBhsSq+gpw7zir3gW8Fqi+2tHABVX1UFXdCowAhyXZC9itqq6oqgLOA47pG7O8LV8EHDk6yyhJkiRJmrxpvyYxyUuANVV13ZhVc4E7+j6vbrW5bXls/WFjqmo98EPgSQNoW5IkSZJ2CNN6d9MkuwKvB35nvNXj1GoT9U2NGe/Yi+idsso+++yz2V4lSZIkaUc03TOJTwf2Ba5L8h1gHnBNkl+mN0O4d9+284A7W33eOHX6xyTZCXg845/eSlWdU1ULq2rhnDlzpuwLSZIkSdJMMq0hsapWVtWTq2p+Vc2nF/IOqarvAhcDx7c7lu5L7wY1V1XVWuC+JIe36w1PBD7ddnkxcFJbPhb4YrtuUZIkSZK0BQb9CIyPAlcAz0iyOsnJG9u2qlYBFwI3AP8MnFpVG9rqU4D30buZzbeBS1r9/cCTkowAfw6cNpAvIkmSJEk7iIFek1hVL9/M+vljPi8Flo6z3QrgwHHqDwLHbV2XkiRJkqRR0353U0mSJEnStsuQKEmSJEnqGBIlSZIkSR1DoiRJkiSpY0iUJEmSJHUMiZIkSZKkjiFRkiRJktQxJEqSJEmSOoZESZIkSVLHkChJkiRJ6hgSJUmSJEkdQ6IkSZIkqWNIlCRJkiR1DImSJEmSpI4hUZIkSZLUMSRKkiRJkjqGREmSJElSx5AoSZIkSeoYEiVJkiRJHUOiJEmSJKljSJQkSZIkdQyJkiRJkqSOIVGSJEmS1DEkSpIkSZI6hkRJkiRJUseQKEmSJEnqGBIlSZIkSR1DoiRJkiSpY0iUJEmSJHUMiZIkSZKkzkBDYpIPJLk7ybf6am9PclOS65N8KskT+tadnmQkyc1JXthXPzTJyrbuzCRp9Z2TfKzVr0wyf5DfR5IkSZJmukHPJJ4LHDWmdilwYFUdBPwbcDpAkv2B44ED2pizksxqY84GFgH7tdfoPk8Gvl9VC4B3AW8d2DeRJEmSpB3AQENiVX0FuHdM7fNVtb59/Dowry0fDVxQVQ9V1a3ACHBYkr2A3arqiqoq4DzgmL4xy9vyRcCRo7OMkiRJkqTJG/Y1if8FuKQtzwXu6Fu3utXmtuWx9YeNacHzh8CTBtivJEmSJM1oQwuJSV4PrAfOHy2Ns1ltor6pMeMdb1GSFUlWrFu3brLtSpIkSdIOYSghMclJwIuAE9oppNCbIdy7b7N5wJ2tPm+c+sPGJNkJeDxjTm8dVVXnVNXCqlo4Z86cqfoqkiRJkjSjTHtITHIU8DrgJVV1f9+qi4Hj2x1L96V3g5qrqmotcF+Sw9v1hicCn+4bc1JbPhb4Yl/olCRJkiRN0k6D3HmSjwJHAHskWQ28gd7dTHcGLm33mPl6Vf1JVa1KciFwA73TUE+tqg1tV6fQu1PqbHrXMI5ex/h+4ENJRujNIB4/yO8jSZIkSTPdQENiVb18nPL7N7H9UmDpOPUVwIHj1B8EjtuaHiVJkiRJvzDsu5tKkiRJkrYhhkRJkiRJUseQKEmSJEnqGBIlSZIkSR1DoiRJkiSpY0iUJEmSJHUMiZIkSZKkjiFRkiRJktQxJEqSJEmSOoZESZIkSVLHkChJkiRJ6hgSJUmSJEkdQ6IkSZIkqWNIlCRJkiR1DImSJEmSpI4hUZIkSZLUMSRKkiRJkjqGREmSJElSx5AoSZIkSeoYEiVJkiRJHUOiJEmSJKljSJQkSZIkdQyJkiRJkqSOIVGSJEmS1DEkSpIkSZI6hkRJkiRJUseQKEmSJEnqGBIlSZIkSR1DoiRJkiSpM9CQmOQDSe5O8q2+2u5JLk1yS3t/Yt+605OMJLk5yQv76ocmWdnWnZkkrb5zko+1+pVJ5g/y+0iSJEnSTDfomcRzgaPG1E4DLquq/YDL2meS7A8cDxzQxpyVZFYbczawCNivvUb3eTLw/apaALwLeOvAvokkSZIk7QAGGhKr6ivAvWPKRwPL2/Jy4Ji++gVV9VBV3QqMAIcl2QvYraquqKoCzhszZnRfFwFHjs4ySpIkSZImbxjXJO5ZVWsB2vuTW30ucEffdqtbbW5bHlt/2JiqWg/8EHjSwDqXJEmSpBluW7pxzXgzgLWJ+qbGPHLnyaIkK5KsWLdu3Ra2KEmSJEkz2zBC4l3tFFLa+92tvhrYu2+7ecCdrT5vnPrDxiTZCXg8jzy9FYCqOqeqFlbVwjlz5kzRV5EkSZKkmWUYIfFi4KS2fBLw6b768e2OpfvSu0HNVe2U1PuSHN6uNzxxzJjRfR0LfLFdtyhJkiRJ2gI7DXLnST4KHAHskWQ18AbgLcCFSU4GbgeOA6iqVUkuBG4A1gOnVtWGtqtT6N0pdTZwSXsBvB/4UJIRejOIxw/y+0iSJEnSTDfQkFhVL9/IqiM3sv1SYOk49RXAgePUH6SFTEmSJEnS1tuWblwjSZIkSRoyQ6IkSZIkqWNIlCRJkiR1DImSJEmSpI4hUZIkSZLUmVBITLJrkv+V5B/b5/2SvGiwrUmSJEmSpttEZxI/CDwEPKd9Xg383UA6kiRJkiQNzURD4tOr6m3AzwCq6gEgA+tKkiRJkjQUEw2JP00yGyiAJE+nN7MoSZIkSZpBdprgdm8A/hnYO8n5wPOAVw2qKUmSJEnScEwoJFbVpUmuAQ6nd5rpkqq6Z6CdSZIkSZKm3WQegTEXmAU8BviNJC8dTEuSJEmSpGGZ0Exikg8ABwGrgJ+3cgGfHFBfkiRJkqQhmOg1iYdX1f4D7USSJEmSNHQTPd30iiSGREmSJEma4SY6k7icXlD8Lr1HXwSoqjpoYJ1JkiRJkqbdREPiB4BXAiv5xTWJkiRJkqQZZqIh8faqunignUiSJEmShm6iIfGmJB8BPkPvdFMAqsq7m0qSJEnSDDLRkDibXjj8nb6aj8CQJEmSpBlmQiGxqv5o0I1IkiRJkoZvQo/ASDIvyaeS3J3kriSfSDJv0M1JkiRJkqbXRJ+T+EHgYuApwFx61yZ+cFBNSZIkSZKGY6IhcU5VfbCq1rfXucCcAfYlSZIkSRqCiYbEe5K8Isms9noF8L1BNiZJkiRJmn4TDYn/BfhD4LvAWuDYVpMkSZIkzSATvbvp7cBLBtyL+ixbtoyRkZFhtyE9wui/l0uWLBlyJ9L4FixYwOLFi4fdhiRJ260JhcQky4ElVfWD9vmJwDuqytnEARkZGeHab93Ihl13H3Yr0sM86qcFwNX/fteQO5Eeadb99w67BUmStnsTConAQaMBEaCqvp/kWQPqSc2GXXfngV/5vWG3IUnbjdk3fXbYLUiStN2b6DWJj2qzhwAk2Z2JB8xxJXl1klVJvpXko0l2SbJ7kkuT3NLe+495epKRJDcneWFf/dAkK9u6M5Nka/qSJEmSpB3ZREPiO4CvJfnbJH8LfA1425YeNMlc4E+BhVV1IDALOB44DbisqvYDLmufSbJ/W38AcBRwVpJZbXdnA4uA/drrqC3tS5IkSZJ2dBMKiVV1Hr07mt4F3A28tKo+tJXH3gmYnWQnYFfgTuBoYHlbvxw4pi0fDVxQVQ9V1a3ACHBYkr2A3arqiqoq4Ly+MZIkSZKkSZrMKaM3Ad8fHZNkn3bX00mrqjVJ/h64HXgA+HxVfT7JnlW1tm2zNsmT25C5wNf7drG61X7WlsfWJUmSJElbYKJ3N10MvIHeTOIGIEABB23JQdu1hkcD+wI/AD6e5BWbGjJOrTZRH++Yi+idlso+++wzqX4lSZIkaUcx0ZnEJcAzqup7U3Tc/wTcWlXrAJJ8EngucFeSvdos4l70Tm2F3gzh3n3j59E7PXV1Wx5bf4SqOgc4B2DhwoXjBklJkiRJ2tFNNCTeAfxwCo97O3B4kl3pnW56JLAC+AlwEvCW9v7ptv3FwEeSvBN4Cr0b1FxVVRuS3JfkcOBK4ERg2RT2KUmStNWWLVvGyMjIsNuYMdasWQPA3LleZTQVFixYwOLFi4fdhrYhEw2J/w58Kcn/Ax4aLVbVO7fkoFV1ZZKLgGuA9cA36c3yPRa4MMnJ9ILkcW37VUkuBG5o259aVRva7k4BzgVmA5e0lyRJkmaoBx54YNgtSDPaREPi7e31mPbaalX1BnrXOfZ7iN6s4njbLwWWjlNfARw4FT1JkiQNgrM0U2vJkiUAnHHGGUPuRJqZJhQSq+pNg25EkiRJkjR8mwyJSd5dVX+W5DOMc9fQqnrJwDqTJEmSJE27zc0kfqi9//2gG5EkSZIkDd8mQ2JVXd3ev7yp7ZJ8oqr+YCobkyRJkiRNv0dN0X6eNkX7kSRJkiQN0VSFRB9OL0mSJEkzwFSFREmSJEnSDDBVITFTtB9JkiRJ0hBNVUh83RTtR5IkSZI0RJt7TuJKxr/eMEBV1UH0Fj4/gN4kSZIkSdNsc89JfNG0dCFJkiRJ2iZs7jmJt01XI5IkSZKk4ZvQNYlJDk/yjSQ/TvLTJBuS/GjQzUmSJEmSptdEb1zzHuDlwC3AbOC/AssG1ZQkSZIkaTg2d01ip6pGksyqqg3AB5N8bYB9SZIkSZKGYKIh8f4kjwGuTfI2YC3wS4NrS5IkSZI0DBM93fSVbdv/AfwE2Bt46aCakiRJkiQNx0RD4jFV9WBV/aiq3lRVf46Px5AkSZKkGWeiIfGkcWqvmsI+JEmSJEnbgE1ek5jk5cB/BvZNcnHfqt2A7w2yMUmSJEnS9NvcjWu+Ru8mNXsA7+ir3wdcP6imJEmSJEnDscmQWFW3AbcBz0myJ/DsturGqlo/6OYkSZIkSdNrQtckJjkOuAo4DvhD4Mokxw6yMUmSJEnS9JvocxL/Cnh2Vd0NkGQO8AXgokE1JkmSJEmafhO9u+mjRgNi871JjJUkSZIkbScmOpN4SZLPAR9tn18GfHYwLUmSJEmShmWis4EF/B/gIOCZwDkD60iSJEmSNDQTnUn87ap6HfDJ0UKSNwGvG0hXkiRJkqSh2GRITHIK8N+BpyXpfy7i44B/HWRjkiRJkqTpt7nTTT8CvBi4uL2Pvg6tqldszYGTPCHJRUluSnJjkuck2T3JpUluae9P7Nv+9CQjSW5O8sK++qFJVrZ1ZybJ1vQlSZIkSTuyTYbEqvphVX2nql5eVbf1ve6dgmOfAfxzVf0KvescbwROAy6rqv2Ay9pnkuwPHA8cABwFnJVkVtvP2cAiYL/2OmoKepMkSZKkHdJQHmORZDfgN4D3A1TVT6vqB8DRwPK22XLgmLZ8NHBBVT1UVbcCI8BhSfYCdquqK6qqgPP6xkiSJEmSJmlYzzp8GrAO+GCSbyZ5X5JfAvasqrUA7f3Jbfu5wB1941e32ty2PLYuSZIkSdoCwwqJOwGHAGdX1bOAn9BOLd2I8a4zrE3UH7mDZFGSFUlWrFu3brL9SpIkSdIOYVghcTWwuqqubJ8vohca72qnkNLe7+7bfu++8fOAO1t93jj1R6iqc6pqYVUtnDNnzpR9EUmSJEmaSYYSEqvqu8AdSZ7RSkcCN9C7i+pJrXYS8Om2fDFwfJKdk+xL7wY1V7VTUu9Lcni7q+mJfWMkSZIkSZO0yeckDthi4PwkjwH+HfgjeqH1wiQnA7cDxwFU1aokF9ILkuuBU6tqQ9vPKcC5wGzgkvaSJEmSJG2BoYXEqroWWDjOqiM3sv1SYOk49RXAgVPb3fCtWbOGWff/kNk3fXbYrUjSdmPW/d9jzZr1w25jRli2bBkjIyPDbkMa1+i/m0uWLBlyJ9IjLViwgMWLFw+7ja0yzJlESZK0jRoZGeGWVd9kn8du2PzG0jR7zM96V0w9dNuKIXciPdztP561+Y22A4bEbdTcuXP57kM78cCv/N6wW5Gk7cbsmz7L3Ll7DruNGWOfx27gLw/50bDbkKTtxpuv2W3YLUyJYd3dVJIkSZK0DTIkSpIkSZI6hkRJkiRJUseQKEmSJEnqGBIlSZIkSR1DoiRJkiSpY0iUJEmSJHUMiZIkSZKkjiFRkiRJktQxJEqSJEmSOoZESZIkSVLHkChJkiRJ6hgSJUmSJEkdQ6IkSZIkqWNIlCRJkiR1DImSJEmSpM5Ow25AkiRte9asWcNP7pvFm6/ZbditSNJ247b7ZvFLa9YMu42t5kyiJEmSJKnjTKIkSXqEuXPn8tD6tfzlIT8adiuStN148zW7sfPcucNuY6s5kyhJkiRJ6hgSJUmSJEkdQ6IkSZIkqWNIlCRJkiR1DImSJEmSpI4hUZIkSZLU8REY27BZ99/L7Js+O+w2pId51IO92+H/fBcfsK1tz6z77wX2HHYbkiRt1wyJ26gFCxYMuwVpXCMj9wGw4Gn+j7i2RXv6309JkraSIXEbtXjx4mG3II1ryZIlAJxxxhlD7kSSJEmDMNRrEpPMSvLNJP/UPu+e5NIkt7T3J/Zte3qSkSQ3J3lhX/3QJCvbujOTZBjfRZIkSZJmgmHfuGYJcGPf59OAy6pqP+Cy9pkk+wPHAwcARwFnJZnVxpwNLAL2a6+jpqd1SZIkSZp5hhYSk8wD/j/gfX3lo4HlbXk5cExf/YKqeqiqbgVGgMOS7AXsVlVXVFUB5/WNkSRJkiRN0jBnEt8NvBb4eV9tz6paC9Den9zqc4E7+rZb3Wpz2/LYuiRJkiRpCwwlJCZ5EXB3VV090SHj1GoT9fGOuSjJiiQr1q1bN8HDSpIkSdKOZVh3N30e8JIkvwfsAuyW5MPAXUn2qqq17VTSu9v2q4G9+8bPA+5s9Xnj1B+hqs4BzgFYuHDhuEFSkiT9wu0/nsWbr/GZqNr23HV/b55jz11/vpktpel1+49nsd+wm5gCQwmJVXU6cDpAkiOA11TVK5K8HTgJeEt7/3QbcjHwkSTvBJ5C7wY1V1XVhiT3JTkcuBI4EVg2rV9GkqQZyOdNalv205ERAHZ+qv+eatuyHzPjv5/b2nMS3wJcmORk4HbgOICqWpXkQuAGYD1walVtaGNOAc4FZgOXtJckSdoKPq9X2zKf2SsN1tBDYlV9CfhSW/4ecORGtlsKLB2nvgI4cHAdSpIkSdKOY9jPSZQkSZIkbUMMiZIkSZKkjiFRkiRJktQxJEqSJEmSOoZESZIkSVLHkChJkiRJ6hgSJUmSJEkdQ6IkSZIkqWNIlCRJkiR1DImSJEmSpI4hUZIkSZLUMSRKkiRJkjqGREmSJElSx5AoSZIkSeoYEiVJkiRJHUOiJEmSJKljSJQkSZIkdQyJkiRJkqSOIVGSJEmS1DEkSpIkSZI6hkRJkiRJUseQKEmSJEnqGBIlSZIkSR1DoiRJkiSpY0iUJEmSJHUMiZIkSZKkjiFRkiRJktQxJEqSJEmSOoZESZIkSVJnKCExyd5J/iXJjUlWJVnS6rsnuTTJLe39iX1jTk8ykuTmJC/sqx+aZGVbd2aSDOM7SZIkSdJMMKyZxPXA/6yqXwUOB05Nsj9wGnBZVe0HXNY+09YdDxwAHAWclWRW29fZwCJgv/Y6ajq/iCRJkiTNJEMJiVW1tqquacv3ATcCc4GjgeVts+XAMW35aOCCqnqoqm4FRoDDkuwF7FZVV1RVAef1jZEkSZIkTdLQr0lMMh94FnAlsGdVrYVekASe3DabC9zRN2x1q81ty2PrkiRJkqQtMNSQmOSxwCeAP6uqH21q03FqtYn6eMdalGRFkhXr1q2bfLOSJEmStAPYaVgHTvJoegHx/Kr6ZCvflWSvqlrbTiW9u9VXA3v3DZ8H3Nnq88apP0JVnQOcA7Bw4cJxg6RmrmXLljEyMjLsNmaE0T/HJUuWDLmTmWPBggUsXrx42G1IkiQBw7u7aYD3AzdW1Tv7Vl0MnNSWTwI+3Vc/PsnOSfald4Oaq9opqfclObzt88S+MZIGYPbs2cyePXvYbUiSJGlAhjWT+DzglcDKJNe22l8CbwEuTHIycDtwHEBVrUpyIXADvTujnlpVG9q4U4BzgdnAJe0lPYyzNJIkSdLEDCUkVtVXGf96QoAjNzJmKbB0nPoK4MCp606SJEmSdlxDv7upJEmSJGnbYUiUJEmSJHUMiZIkSZKkjiFRkiRJktQZ2nMSJUmSdhQ+r3dq+czeqeXzejWWIVGSJEnbFZ/XKw2WIVGSJGnAnKWRtD3xmkRJkiRJUseQKEmSJEnqGBIlSZIkSR1DoiRJkiSpY0iUJEmSJHUMiZIkSZKkjiFRkiRJktQxJEqSJEmSOoZESZIkSVLHkChJkiRJ6hgSJUmSJEmdVNWwe5h2SdYBtw27D2k7tgdwz7CbkCTt0PwtkrbOU6tqzngrdsiQKGnrJFlRVQuH3Yckacflb5E0OJ5uKkmSJEnqGBIlSZIkSR1DoqQtcc6wG5Ak7fD8LZIGxGsSJUmSJEkdZxIlSZIkSR1DojSDJdkzyUeS/HuSq5NckeT3t2J/b0zymi0cOz/Jf97SY0uSth1JKsk7+j6/JskbNzPmmCT7b2L9K5Jcn2RVkuuSvC/JE7ayzx9vxdhXJXnK1hxf2l4ZEqUZKkmA/wt8paqeVlWHAscD88Zst9M0tTQfMCRK0szwEPDSJHtMYswxwLghMclRwKuB362qA4BDgK8Be46z7azJt7tFXgUYErVDMiRKM9cLgJ9W1XtHC1V1W1Uta387+vEknwE+n+SxSS5Lck2SlUmOHh2T5PVJbk7yBeAZffUvJVnYlvdI8p22PD/J5W1f1yR5bhvyFuD5Sa5N8uoks5K8Pck32t8c//Hg/0gkSVNkPb0bx7x67IokT22/Kde3933ab8FLgLe334Gnjxn2euA1VbUGoKo2VNUHqurmts/vJPnrJF8Fjkvy39rvx3VJPpFk17bdvu2smW8k+du+no5I8k99n9+T5FVt+a/b9t9Kck56jtufSd0AAAX/SURBVAUWAue3fmcnOTTJl9uZOZ9LstfU/XFK2xZDojRzHQBcs4n1zwFOqqoXAA8Cv19VhwC/Bbyj/UiOzj4+C3gp8OwJHPdu4Lfbvl4GnNnqpwGXV9XBVfUu4GTgh1X17Lbf/5Zk30l/S0nSsPwDcEKSx4+pvwc4r6oOAs4HzqyqrwEXA3/Rfge+PWbM5n6zAB6sql+vqguAT1bVs6vqmcCN9H5TAM4Azm6/Ld+d4Pd4T9vXgcBs4EVVdRGwAjihqg6mF4qXAce2M3M+ACyd4P6l7Y4hUdpBJPmH9jeu32ilS6vq3tHVwJuTXA98AZhL7xSf5wOfqqr7q+pH9H7gN+fRwD8mWQl8nI2cWgT8DnBikmuBK4EnAfttyXeTJE2/9rtwHvCnY1Y9B/hIW/4Q8OuT2W+SX2uzd99O8rK+VR/rWz6wnbWyEjiBXsgEeB7w0b5jT8RvJbmy7esFffvq9wzgQODS9rv1V4y5fEOaSabrWiRJ028V8AejH6rq1HbtyIpW+knfticAc4BDq+pn7dTRXUaHbmT/6/nFXzTt0ld/NXAX8My2/sGNjA+wuKo+N6FvI0naFr2b3gzgBzexzUSet7aK3nWI/1JVK4GDk7yH3szeqP7frXOBY6rqunba6BGbOV7/bxa0360kuwBnAQur6o52851dHjmcAKuq6jkT+C7Sds+ZRGnm+iKwS5JT+mq7bmTbxwN3t4D4W8BTW/0rwO+3azEeB7y4b8x3gEPb8rFj9rW2qn4OvBIYvcHAfcDj+rb7HHBKkkcDJPkPSX5pMl9QkjRc7YyUC/nF6Z7Qu+HM8W35BOCrbXns70C//w38fZL+2bnZG9mWtp+17TfkhL76v4459qjbgP2T7NxOjz2y1UcD4T1JHsvDf8/6+70ZmJPkOQBJHp1kvBlHaUYwJEozVFUVvTvJ/WaSW5NcBSwHXjfO5ucDC5OsoPejelPbxzX0Tu+5FvgEcHnfmL+nF/K+BvTf3e4s4KQkXwf+A7/4m9/rgfXtlNdXA+8DbgCuSfIt4P/g2Q2StD16Bw//HfhT4I/aJQyvBJa0+gXAXyT55tgb11TVZ+ldw35Jkhvab8sGen+hOJ7/Re9ShUtpv1nNEuDUdmlFd61kVd1BL8xeT+8375ut/gPgH4GV9O4I/o2+fZ0LvLedXjqLXoB8a5Lr6P0uPhdphkrv/yMlSZIkSXImUZIkSZLUx5AoSZIkSeoYEiVJkiRJHUOiJEmSJKljSJQkSZIkdQyJkiSNI8mGJNf2vU4bZ5sjkvzTFB/3iCTP7fv8J0lOnMpjSJK0KT6TTJKk8T1QVQcP4bhHAD+m90Byquq9Q+hBkrQDcyZRkqRJSHJUkpuSfBV4aV/9jUle0/f5W0nmt+UTk1yf5LokH2q1Fye5sj1Y/AtJ9mzb/wnw6jZ7+fz+/SY5OMnX274+leSJrf6lJG9NclWSf0vy/Gn645AkzUCGREmSxjd7zOmmL0uyC/CPwIuB5wO/vLmdJDkAeD3wgqp6JrCkrfoqcHhVPQu4AHhtVX0HeC/wrqo6uKouH7O784DXVdVBwErgDX3rdqqqw4A/G1OXJGlSPN1UkqTxPeJ00yQHA7dW1S3t84eBRZvZzwuAi6rqHoCqurfV5wEfS7IX8Bjg1k3tJMnjgSdU1ZdbaTnw8b5NPtnerwbmb6YnSZI2yplESZImpzZSX8/Df1d3ae/ZyJhlwHuq6teAP+7bfks91N434F8CS5K2giFRkqSJuwnYN8nT2+eX9637DnAIQJJDgH1b/TLgD5M8qa3bvdUfD6xpyyf17ec+4HFjD1xVPwS+33e94SuBL4/dTpKkrWVIlCRpfGOvSXxLVT1I7/TS/9duXHNb3/afAHZPci1wCvBvAFW1ClgKfDnJdcA72/ZvBD6e5HLgnr79fAb4/dEb14zp6STg7UmuBw4G/mYqv7AkSQCp2thZM5IkSZKkHY0ziZIkSZKkjiFRkiRJktQxJEqSJEmSOoZESZIkSVLHkChJkiRJ6hgSJUmSJEkdQ6IkSZIkqWNIlCRJkiR1/n/wG6Ea8i+XGQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "boxplot(data['Education'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observations:**\n",
    "- We can see that this does appear reasonable; graduate customers seem to earn more on average."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "** Let's check how credit history is related to loan status **"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x248fd897d00>"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(x=data.Credit_History, hue=data.Loan_Status)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observations:**\n",
    "- The plot shows that credit history is an important factor while checking loan eligibility. There are very few customers whose loan was approved even when their credit history did not meet required guidelines."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Let's check how property area is related with loan status**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x248fe16deb0>"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(x=data.Property_Area, hue=data.Loan_Status)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observations:**\n",
    "\n",
    "- The plot shows that more loans are approved for properties in semi-urban areas. \n",
    "- This could be due to several reasons. The bank might be charging higher interest rates for semi-urban areas or the current customer base of the company from semi-urban areas might actually be more eligible for home loans based on loan applicant features. We cannot be certain as we don't have the data to support this claim.\n",
    "\n",
    "**Now that we have processed and explored our data, let's prepare it for modeling.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "zHsHD8S3O7hf"
   },
   "source": [
    "#### Preparing data for modeling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "id": "ia9I9xy7O7hg"
   },
   "outputs": [],
   "source": [
    "#Separating target variable and other variables\n",
    "X=data.drop(columns='Loan_Status')\n",
    "Y=data['Loan_Status']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "id": "eHbwSygpO7hh"
   },
   "outputs": [],
   "source": [
    "#Creating dummy variables \n",
    "#drop_first=True is used to avoid redundant variables\n",
    "X = pd.get_dummies(X, drop_first=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "id": "jzevkL5mO7hi"
   },
   "outputs": [],
   "source": [
    "#Splitting the data into train and test sets\n",
    "X_train,X_test,y_train,y_test=train_test_split(X, Y, test_size=0.30, random_state=7)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "HVZae82KO7hj"
   },
   "source": [
    "### **Model Building**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "2-sTuiJdO7hj"
   },
   "source": [
    "### **Building Classification Models**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "jhGfGwRCO7hm"
   },
   "source": [
    "**Before training the model, let's choose the appropriate model evaluation criterion as per the problem on hand.**\n",
    "\n",
    "### Model evaluation criterion:\n",
    "\n",
    "#### Since this is a binary classification problem, the model can make wrong predictions in one of two ways:\n",
    "1. Predicting a customer is not eligible for a loan when the customer actually is eligible - **Loss of opportunity**\n",
    "2. Predicting a customer is eligible for a loan when the customer is actually not - **Financial Loss**\n",
    "\n",
    "#### Which case is more important? \n",
    "* **Predicting that the customer is eligible when he is not**, because the person might default or not be able to return the loan, which would result in financial loss to the company.\n",
    "\n",
    "#### How to reduce this loss i.e need to reduce False Negatives?\n",
    "* In classification, the class of interest is considered the positive class. Here, the class of interest is 0 i.e. identifying non-eligible customers. So, the company wants to minimize the number of false negatives, in other words **Recall** is the evaluation criterion to be maximized."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "dEsDqZayO7hv"
   },
   "source": [
    "#### let's create a function to calculate and print the classification report and confusion matrix so that we don't have to rewrite the same code repeatedly for each model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "id": "lknK3VvDO7hx"
   },
   "outputs": [],
   "source": [
    "#function to print classification report and get confusion matrix in a proper format\n",
    "\n",
    "def metrics_score(actual, predicted):\n",
    "    print(classification_report(actual, predicted))\n",
    "    cm = confusion_matrix(actual, predicted)\n",
    "    plt.figure(figsize=(8,5))\n",
    "    sns.heatmap(cm, annot=True,  fmt='.2f', xticklabels=['Not Eligible', 'Eligible'], yticklabels=['Not Eligible', 'Eligible'])\n",
    "    plt.ylabel('Actual')\n",
    "    plt.xlabel('Predicted')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " **Linear Discriminant Analysis (LDA)**\n",
    " Linear discriminant analysis (LDA) is generally used to classify patterns between two classes; however, it can be extended to classify multiple patterns. LDA assumes that all classes are linearly separable and according to this, multiple linear discrimination functions representing several hyperplanes in the feature space are created to distinguish between the classes. If there are two classes, then the LDA draws one hyperplane and projects the data onto this hyperplane in such a way as to maximize the separation of the two categories. This hyperplane is created according to two criteria considered simultaneously:\n",
    "\n",
    "- Maximizing the distance between the means of two classes;\n",
    "- Minimizing the variation between each category."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearDiscriminantAnalysis()"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#fitting lda model\n",
    "lda = LinearDiscriminantAnalysis()\n",
    "lda.fit(x_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.92      0.41      0.57       131\n",
      "           1       0.79      0.98      0.88       298\n",
      "\n",
      "    accuracy                           0.81       429\n",
      "   macro avg       0.85      0.70      0.72       429\n",
      "weighted avg       0.83      0.81      0.78       429\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#checking model performance of lda\n",
    "y_pred_train_lda = lda.predict(x_train)\n",
    "metrics_score(y_train, y_pred_train_lda)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Credit_History</th>\n",
       "      <td>4.823615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Married_Yes</th>\n",
       "      <td>0.751970</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Property_Area_Semiurban</th>\n",
       "      <td>0.742081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dependents_2</th>\n",
       "      <td>0.452345</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dependents_3+</th>\n",
       "      <td>0.427453</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Self_Employed_Yes</th>\n",
       "      <td>0.191595</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Property_Area_Urban</th>\n",
       "      <td>0.146022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_income</th>\n",
       "      <td>-0.000031</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LoanAmount</th>\n",
       "      <td>-0.000989</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Loan_Amount_Term</th>\n",
       "      <td>-0.007488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dependents_1</th>\n",
       "      <td>-0.233224</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Gender_Male</th>\n",
       "      <td>-0.337370</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Education_Not Graduate</th>\n",
       "      <td>-0.460198</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                0\n",
       "Credit_History           4.823615\n",
       "Married_Yes              0.751970\n",
       "Property_Area_Semiurban  0.742081\n",
       "Dependents_2             0.452345\n",
       "Dependents_3+            0.427453\n",
       "Self_Employed_Yes        0.191595\n",
       "Property_Area_Urban      0.146022\n",
       "total_income            -0.000031\n",
       "LoanAmount              -0.000989\n",
       "Loan_Amount_Term        -0.007488\n",
       "Dependents_1            -0.233224\n",
       "Gender_Male             -0.337370\n",
       "Education_Not Graduate  -0.460198"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#creating list of column names\n",
    "cols=X.columns\n",
    "\n",
    "#saving coefficients of lda model\n",
    "coef_lda=lda.coef_\n",
    "\n",
    "#printing the cofficients of lda\n",
    "pd.DataFrame(coef_lda,columns=cols).T.sort_values(by=0,ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Precision-Recall Curve for LDA\n",
    "\n",
    "**Precision-Recall curves summarize the trade-off between the true positive rate and the positive predictive value for a predictive model using different probability thresholds.**\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_scores_lda=lda.predict_proba(x_train) #predict_proba gives the probability of each observation belonging to each class\n",
    "\n",
    "\n",
    "precisions_lda, recalls_lda, thresholds_lda = precision_recall_curve(y_train, y_scores_lda[:,1])\n",
    "\n",
    "#Plot values of precisions, recalls, and thresholds\n",
    "plt.figure(figsize=(10,7))\n",
    "plt.plot(thresholds_lda, precisions_lda[:-1], 'b--', label='precision')\n",
    "plt.plot(thresholds_lda, recalls_lda[:-1], 'g--', label = 'recall')\n",
    "plt.xlabel('Threshold')\n",
    "plt.legend(loc='upper left')\n",
    "plt.ylim([0,1])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.92      0.41      0.57       131\n",
      "           1       0.79      0.98      0.88       298\n",
      "\n",
      "    accuracy                           0.81       429\n",
      "   macro avg       0.85      0.70      0.72       429\n",
      "weighted avg       0.83      0.81      0.78       429\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "optimal_threshold1=.35\n",
    "y_pred_train_lda = lda.predict_proba(x_train)\n",
    "metrics_score(y_train, y_pred_train_lda[:,1]>optimal_threshold1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.93      0.46      0.62        61\n",
      "           1       0.79      0.98      0.87       124\n",
      "\n",
      "    accuracy                           0.81       185\n",
      "   macro avg       0.86      0.72      0.74       185\n",
      "weighted avg       0.84      0.81      0.79       185\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "optimal_threshold1=.35\n",
    "y_pred_test_lda = lda.predict_proba(x_test)\n",
    "metrics_score(y_test, y_pred_test_lda[:,1]>optimal_threshold1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Quadratic Discriminant Analysis\n",
    "Quadratic discriminant analysis (QDA) is a probabilistic parametric classification technique which represents an evolution of LDA for nonlinear class separations. QDA, like LDA, is based on the hypothesis that the probability density distributions are multivariate normal but, in this case, the dispersion is not the same for all of the categories.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "QuadraticDiscriminantAnalysis()"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#fitting qda model\n",
    "qda = QuadraticDiscriminantAnalysis()\n",
    "qda.fit(x_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.84      0.45      0.59       131\n",
      "           1       0.80      0.96      0.87       298\n",
      "\n",
      "    accuracy                           0.81       429\n",
      "   macro avg       0.82      0.71      0.73       429\n",
      "weighted avg       0.81      0.81      0.79       429\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#checking model performance on the training data\n",
    "y_pred_train_qda = qda.predict(x_train)\n",
    "metrics_score(y_train, y_pred_train_qda)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Let's move on to another model - Logistic Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(random_state=1)"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "log_reg= LogisticRegression(random_state=1)\n",
    "\n",
    "log_reg.fit(x_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Credit_History             3.006569\n",
       "Property_Area_Semiurban    0.725591\n",
       "Married_Yes                0.425012\n",
       "Dependents_2               0.394610\n",
       "Self_Employed_Yes          0.084766\n",
       "total_income              -0.000035\n",
       "LoanAmount                -0.000569\n",
       "Property_Area_Urban       -0.008981\n",
       "Dependents_3+             -0.033317\n",
       "Loan_Amount_Term          -0.061075\n",
       "Gender_Male               -0.216884\n",
       "Dependents_1              -0.246658\n",
       "Education_Not Graduate    -0.537382\n",
       "dtype: float64"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series(log_reg.coef_[0], index=x_train.columns).sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observations:**\n",
    "\n",
    "- The attribute Credit history stands out among all the features which is fairly reasonable to imply that loan applicants with good credit history are likely get approved for loan.\n",
    "- The features Property Area_Semi Urban, Married, Dependents_2, Dependents_3 also seem to positively affect the Loan approval rate.It might be that applicants situated in Semiurban areas satisfy the credit history and other loan requirements and are more likely to get approved.\n",
    "- Education_Not Graduate seem to be negatively affecting the Loan approval rate which indicates that applicants who are not graduates are less likely to get approved.\n",
    "- The remaining features have very low coefficients and does not seem to drive the loan approval as much."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>odds</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Credit_History</th>\n",
       "      <td>20.217922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Property_Area_Semiurban</th>\n",
       "      <td>2.065952</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Married_Yes</th>\n",
       "      <td>1.529608</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dependents_2</th>\n",
       "      <td>1.483805</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Self_Employed_Yes</th>\n",
       "      <td>1.088462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_income</th>\n",
       "      <td>0.999965</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LoanAmount</th>\n",
       "      <td>0.999432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Property_Area_Urban</th>\n",
       "      <td>0.991059</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dependents_3+</th>\n",
       "      <td>0.967232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Loan_Amount_Term</th>\n",
       "      <td>0.940753</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Gender_Male</th>\n",
       "      <td>0.805023</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dependents_1</th>\n",
       "      <td>0.781408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Education_Not Graduate</th>\n",
       "      <td>0.584276</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                              odds\n",
       "Credit_History           20.217922\n",
       "Property_Area_Semiurban   2.065952\n",
       "Married_Yes               1.529608\n",
       "Dependents_2              1.483805\n",
       "Self_Employed_Yes         1.088462\n",
       "total_income              0.999965\n",
       "LoanAmount                0.999432\n",
       "Property_Area_Urban       0.991059\n",
       "Dependents_3+             0.967232\n",
       "Loan_Amount_Term          0.940753\n",
       "Gender_Male               0.805023\n",
       "Dependents_1              0.781408\n",
       "Education_Not Graduate    0.584276"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "odds = np.exp(log_reg.coef_[0]) #finding the odds\n",
    "\n",
    "#adding the odds to a dataframe and sorting the values\n",
    "pd.DataFrame(odds, x_train.columns, columns=['odds']).sort_values(by='odds', ascending=False) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observations:\n",
    "\n",
    "- The odds of getting a loan approval is much higher if the applicant has a good Credit history.This suggests that Credit history is a very good predictor of whether the loan gets approved or not.\n",
    "- The odds of applicant with 2 or more dependents getting loan approval is almost double than those with only one dependent.\n",
    "- For applicants owning properties in Semiurban areas, the odds of getting approved for loan is much higher than those in urban and rural areas.\n",
    "- The marital status- Married is also a good predictor of whether the loan gets accepted or rejected. We can see that for married applicants, the odds of getting approved is 1.2 times higher than for those with other marital status.   \n",
    "\n",
    "**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Now, let's check the performance of the model on the training set**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.89      0.42      0.57       131\n",
      "           1       0.79      0.98      0.88       298\n",
      "\n",
      "    accuracy                           0.81       429\n",
      "   macro avg       0.84      0.70      0.72       429\n",
      "weighted avg       0.82      0.81      0.78       429\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#predict on the training data \n",
    "y_pred_train = log_reg.predict(x_train)\n",
    "\n",
    "# Checking performance on the training data\n",
    "metrics_score(y_train, y_pred_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observations:\n",
    "\n",
    "- We can see in the classification report that there is 81% accuracy on the training dataset.\n",
    "- However the recall is only about 42% for class of interest which indicates that the model performance would not be as good in differentiating those applicants who are more likely to default.\n",
    "- We can see from the confusion matrix that the values of False negatives(73) is too high which means this model fails to identify the applicants who are more likely to default or at risk. \n",
    "\n",
    "**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Precision-Recall Curve for Logistic Regression**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_scores=log_reg.predict_proba(x_train) #predict_proba gives the probability of each observation belonging to each class\n",
    "\n",
    "\n",
    "precisions, recalls, thresholds = precision_recall_curve(y_train, y_scores[:,1])\n",
    "\n",
    "#Plot values of precisions, recalls, and thresholds\n",
    "plt.figure(figsize=(10,7))\n",
    "plt.plot(thresholds, precisions[:-1], 'b--', label='precision')\n",
    "plt.plot(thresholds, recalls[:-1], 'g--', label = 'recall')\n",
    "plt.xlabel('Threshold')\n",
    "plt.legend(loc='upper left')\n",
    "plt.ylim([0,1])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observations:**\n",
    "- We can see that precision and recall are balanced for threshold of about 0.7.\n",
    "- Let's try to calculate the exact threshold where precision and recall are equal."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.6883198225148315\n"
     ]
    }
   ],
   "source": [
    "#calculating the exact threshold where precision and recall are equal.\n",
    "for i in np.arange(len(thresholds)):\n",
    "    if precisions[i]==recalls[i]:\n",
    "        print(thresholds[i])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- The threshold of 0.75 would give a balanced precision and recall.\n",
    "- We can choose the threshold to be a bit higher to make the recall higher but then the precision would drop.\n",
    "- **This decision depends on the stakeholders and other business driven factors.** For example, if the company can bear the cost of some false positives and be more conservative while approving loans then we can choose for the recall to be higher.\n",
    "- Let's choose the recall to be 0.75 for the current scenario, since we have mentioned that the company wants to optimize recall.\n",
    "- **Note:** We also don't want the precision to be so poor that it outweighs the benefits of a higher recall. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.48      0.75      0.58       131\n",
      "           1       0.85      0.64      0.73       298\n",
      "\n",
      "    accuracy                           0.67       429\n",
      "   macro avg       0.67      0.69      0.66       429\n",
      "weighted avg       0.74      0.67      0.69       429\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "optimal_threshold1 = 0.75\n",
    "metrics_score(y_train, y_scores[:,1]>optimal_threshold1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.50      0.72      0.59        61\n",
      "           1       0.82      0.65      0.72       124\n",
      "\n",
      "    accuracy                           0.67       185\n",
      "   macro avg       0.66      0.68      0.66       185\n",
      "weighted avg       0.72      0.67      0.68       185\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Checking performance on the testing data\n",
    "y_pred_test = log_reg.predict_proba(x_test)\n",
    "metrics_score(y_test, y_pred_test[:,1]>optimal_threshold1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observations:**\n",
    "\n",
    "- After adjusting for the threshold, the accuracy of the model goes down, but the recall improves significantly.There is definitely a trade-off between precision and recall, but since we are more interested in optimizing the recall, we can ignore the trade-off here.\n",
    "- This model also minimizes the False negatives i.e predicting the applicant is eligible when he is actually not which in turn helps minimize the Financial losses.\n",
    "- Additionally, the f1-score of 0.58 for train and the test set favors the Logistic Regression after tuning in the threshhold.\n",
    "- The overall model performance is better than before adjusting for the threshhold.\n",
    "- Let us check out the model performance with other algorithms.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### K - Nearest Neighbors (KNN)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- KNN is a distance based algorithm and all distance based algorithms are affected by the scale of the data.\n",
    "- We will scale the attributes (dataframe X defined above) before building the KNN model.\n",
    "\n",
    "**But before actually building the model, we need to identify the value of K to be used in KNN. We will perform the following steps for the same.**\n",
    "\n",
    "- For every value of K (from 1 to 20), split the train set into new train and validation set (30 times)\n",
    "- Scale the training data and validation data\n",
    "- Take the average of the error on these train and the validation sets for each K\n",
    "- Plot the average train vs the validation set error for all Ks \n",
    "- Choose a suitable K from the plot where the two errors are comparable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{1: [0.0, 0.3135658914728682],\n",
       " 2: [0.13323615160349853, 0.36007751937984483],\n",
       " 3: [0.16462585034013605, 0.26860465116279075],\n",
       " 4: [0.17716229348882406, 0.29728682170542636],\n",
       " 5: [0.1946550048590865, 0.2453488372093023],\n",
       " 6: [0.1904761904761905, 0.2585271317829456],\n",
       " 7: [0.20524781341107878, 0.237984496124031],\n",
       " 8: [0.1970845481049563, 0.23488372093023263],\n",
       " 9: [0.21049562682215742, 0.23992248062015503],\n",
       " 10: [0.19825072886297374, 0.23255813953488372],\n",
       " 11: [0.21341107871720116, 0.23410852713178293],\n",
       " 12: [0.20641399416909623, 0.22131782945736442],\n",
       " 13: [0.21525753158406222, 0.24108527131782947],\n",
       " 14: [0.21273080660835766, 0.20930232558139536],\n",
       " 15: [0.22011661807580177, 0.2306201550387597],\n",
       " 16: [0.2076773566569485, 0.22403100775193802],\n",
       " 17: [0.2240038872691934, 0.21705426356589153],\n",
       " 18: [0.2118561710398445, 0.2224806201550388],\n",
       " 19: [0.21875607385811474, 0.24457364341085278],\n",
       " 20: [0.21778425655976674, 0.21744186046511632]}"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn = KNeighborsClassifier()\n",
    "ss = StandardScaler()\n",
    "\n",
    "# We select the best value of k for which the error rate is the least in the validation data\n",
    "# Let us loop over a few values of the k to determine the best k\n",
    "\n",
    "train_error = []\n",
    "test_error = []\n",
    "knn_many_split = {}\n",
    "\n",
    "error_df_knn = pd.DataFrame()\n",
    "features = X.columns\n",
    "\n",
    "for k in range(1,21):\n",
    "    train_error = []\n",
    "    test_error = []\n",
    "    lista = []\n",
    "    knn = KNeighborsClassifier(n_neighbors=k)\n",
    "    for i in range(30):\n",
    "        x_train_new, x_val, y_train_new, y_val = train_test_split(x_train, y_train, test_size = 0.20)\n",
    "    \n",
    "        #let us scale the data\n",
    "        x_train_new = ss.fit_transform(x_train_new)  #fit_transform the training data\n",
    "        x_val = ss.transform(x_val) #transform the validation set\n",
    "        #Fitting knn on training data\n",
    "        knn.fit(x_train_new, y_train_new)\n",
    "        #Calculating error on training and validation data\n",
    "        train_error.append(1 - knn.score(x_train_new, y_train_new)) \n",
    "        test_error.append(1 - knn.score(x_val, y_val))\n",
    "    lista.append(sum(train_error)/len(train_error))\n",
    "    lista.append(sum(test_error)/len(test_error))\n",
    "    knn_many_split[k] = lista\n",
    "\n",
    "knn_many_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlkAAAFlCAYAAADYqP0MAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3yV5f3/8deVPQgJhISQQdg7JEAYCg6coAi4t21ta23r11p/VbG1WqtV22+/HX6rdX+tdS+GinsUBRECBMKeCRkQQkhC9jrX7487QIAAJ+PkZLyfj0ceZ9z3fc7nQOS8va7r/tzGWouIiIiItC0fbxcgIiIi0hUpZImIiIh4gEKWiIiIiAcoZImIiIh4gEKWiIiIiAcoZImIiIh4gJ+3C2hKnz597IABA7xdhoiIiMgprVq1ar+1NurY5ztkyBowYABpaWneLkNERETklIwxWU09r+lCEREREQ9QyBIRERHxAIUsEREREQ/okGuyREREpHOora0lJyeHqqoqb5ficUFBQcTHx+Pv7+/W/gpZIiIi0mI5OTmEhYUxYMAAjDHeLsdjrLUUFhaSk5PDwIED3TpG04UiIiLSYlVVVURGRnbpgAVgjCEyMrJZI3YKWSIiItIqXT1gHdLcz6mQJSIiIp1WcXExTz75ZIuO/dvf/kZFRUUbV3SEQpaIiIh0Wh05ZGnhu4iIiHRa8+bNY8eOHaSkpHD++ecTHR3Nm2++SXV1NZdeeikPPvgg5eXlXHXVVeTk5FBfX89vf/tb8vPzycvLY/r06fTp04cvv/yyzWtTyBIREZE28eB7G9iYd7BNX3NUbE8euGT0Cbc/9thjrF+/nvT0dD755BPefvttVqxYgbWW2bNns2TJEgoKCoiNjeWDDz4AoKSkhPDwcP7yl7/w5Zdf0qdPnzat+RBNF3rY5r0Hqat3ebsMERGRLu+TTz7hk08+Ydy4cYwfP57Nmzezbds2kpKS+Oyzz7jnnnv4+uuvCQ8Pb5d6NJLlQZ9uzOfHL6Vx/6xR3DzNvZ4aIiIindXJRpzag7WWe++9l5/85CfHbVu1ahWLFy/m3nvv5YILLuD+++/3eD0ayfKQgtJq5r2zDoD5a3K9XI2IiEjXFBYWRmlpKQAXXnghL7zwAmVlZQDk5uayb98+8vLyCAkJ4YYbbuBXv/oVq1evPu5YT9BIlgdYa5n3zjpKq+u4dlICr63IZkdBGYOjeni7NBERkS4lMjKSqVOnMmbMGGbOnMl1113HaaedBkCPHj14+eWX2b59O3fddRc+Pj74+/vzz3/+E4BbbrmFmTNn0q9fP48sfDfW2jZ/0dZKTU21aWlp3i6jxV5bsZt7383g/lmjmDW2H1Me/Zzbpg/hzguGe7s0ERGRNrVp0yZGjhzp7TLaTVOf1xizylqbeuy+mi5sY5n7y3no/Y1MHRLJ908fQHTPIE4f3IcF6Xl0xEArIiIinuFWyDLGzDDGbDHGbDfGzGti+xxjzDpjTLoxJs0YM63RtkxjTMahbW1ZfEdTV+/il2+m4+dj+POVyfj4OO3356TEsvtABenZxV6uUERERNrLKUOWMcYXeAKYCYwCrjXGjDpmt8+BZGttCnAz8Nwx26dba1OaGkrrSv751Q7W7C7m4UuT6BcefPj5C8fEEODnw8L0PC9WJyIiIu3JnZGsScB2a+1Oa20N8Dowp/EO1toye2QuLBTodvNi63KK+fvn25iTEsvs5NijtvUM8ue8kdG8vy5PPbNERES6CXdCVhyQ3ehxTsNzRzHGXGqM2Qx8gDOadYgFPjHGrDLG3NKaYjuqypp67ngjnaiwQH4/e0yT+8xJiWN/WQ3fbN/fztWJiIiIN7gTskwTzx03UmWtnW+tHQHMBR5qtGmqtXY8znTjz40xZzb5Jsbc0rCeK62goMCNsjqORz/cxM6Ccv7nymTCQ/yb3Ofs4VH0DPJjkaYMRUREugV3QlYOkNDocTxwwqRgrV0CDDbG9Gl4nNdwuw+YjzP92NRxz1hrU621qVFRUW6W731fbdnHS99m8cNpAzl9yImvfRTo58tFSf34eMNeKmvq27FCERGRrqu4uJgnn3yy2cdddNFFFBd79oQ0d0LWSmCoMWagMSYAuAZY1HgHY8wQY4xpuD8eCAAKjTGhxpiwhudDgQuA9W35AbypqLyGu99ex7C+PbjrwlP3wJqTEkd5TT2fbspvh+pERES6vhOFrPr6kw9oLF68mIiICE+VBbjR8d1aW2eMuQ34GPAFXrDWbjDG3Nqw/SngcuAmY0wtUAlcba21xpi+wPyG/OUHvGqt/chDn6VdWWv59fwMiipq+L8fTCTI3/eUx0we2JuYnkEsXJN73OJ4ERERab558+axY8cOUlJS8Pf3p0ePHvTr14/09HQ2btzI3Llzyc7Opqqqil/84hfccouzPHzAgAGkpaVRVlbGzJkzmTZtGsuWLSMuLo6FCxcSHBx8inc+Nbcuq2OtXQwsPua5pxrd/yPwxyaO2wkkt7LGDund1bl8uH4v98wYwehY967m7eNjmJ0Sywvf7KKovIZeoQEerlJERKQdfTgP9ma07WvGJMHMx064+bHHHmP9+vWkp6fz1VdfcfHFF7N+/XoGDhwIwAsvvEDv3r2prKxk4sSJXH755URGRh71Gtu2beO1117j2Wef5aqrruKdd97hhhtuaHXp6vjeAtkHKnhg0QYmDejNLWcOataxc1JiqXNZPsjY46HqREREuq9JkyYdDlgAjz/+OMnJyUyZMoXs7Gy2bdt23DEDBw4kJSUFgAkTJpCZmdkmtegC0c1U77L8v7fWAvA/VyXj69PUyZcnNqpfT4ZG92Bhei43TEn0RIkiIiLecZIRp/YSGhp6+P5XX33FZ599xrfffktISAhnn302VVVVxx0TGBh4+L6vry+VlZVtUotGsprpua93smLXAX43ezQJvUOafbwxhrnj4liZWUROUYUHKhQREek+wsLCKC0tbXJbSUkJvXr1IiQkhM2bN7N8+fJ2rU0hqxk25h3kz59sYcboGC4ff1w/VrcdWvS+aK16ZomIiLRGZGQkU6dOZcyYMdx1111HbZsxYwZ1dXWMHTuW3/72t0yZMqVdazNHrobTcaSmptq0tI51Lemq2nrm/GMpBypq+PiOM+ndykXrl/9zGWVVdXz8yyZ7s4qIiHQKmzZtYuTIkd4uo9009XmNMauauj6zRrLc9D+fbGFLfil/umJsqwMWwNyUWLbkl7Jpz8E2qE5EREQ6GoUsNyzbsZ/nvtnFjVMSmT48uk1e8+Kxsfj5GBak57bJ64mIiEjHopB1CiWVtfzqzbUMjAzl1xe13XBo79AAzhjah/fS83C5Ot6UrYiIiLSOQtYpPLBwPfml1fz16hSCA07d1b055o6LI6+kipWZB9r0dUVERNpTR1zf7QnN/ZwKWSfx3to8FqTncfs5Q0lOaPvrG50/qi8hAb4sSNdZhiIi0jkFBQVRWFjY5YOWtZbCwkKCgoLcPkbNSE9gb0kV9y1YT0pCBD+fPtgj7xES4McFo/qyOGMPD84eTYCfMq+IiHQu8fHx5OTkUFBQ4O1SPC4oKIj4+Hi391fIaoLLZbnr7bXU1Ln469Up+Pl6LvzMSYljQXoe/9lawPmj+nrsfURERDzB39//qMvYyBEaOmnCS99m8vW2/dw3ayQD+4Secv/WmDa0D71DA3SWoYiISBejkHWMbfmlPPrhZs4ZEc11k/p7/P38fX2YNbYfn23Mp7Sq1uPvJyIiIu1DIauRmjoXv3wzndBAPx67PAljmnfx55aakxJHdZ2LTzbkt8v7iYiIiOcpZDXy98+3sj73II9elkR0mPtnD7TW+P4RJPQO1pShiIhIF6KQ1SAt8wD//GoHV6XGc+HomHZ9b2MMc5LjWLp9P/tKq9r1vUVERMQzFLKAsuo67nxzLXG9grn/ktFeqWHuuFhcFt5fu8cr7y8iIiJtSyELeOi9jeQUVfDXq1LoEeidrhZDosMY1a8nC9eqMamIiEhX0O1D1scb9vJGWja3njWY1AG9vVrL3HGxrM0uZtf+cq/WISIiIq3XrUNWQWk1976bwejYntxx3jBvl8Ps5DiMgYVaAC8iItLpdduQZa3lnnfWUV5dx9+uTukQl7SJCQ9iysBIFqbndflrQImIiHR13k8WXvLaimy+2LyPeTNHMLRvmLfLOWxOSiy79peTkVvi7VJERESkFbplyMrcX85D729k2pA+fO+0Ad4u5ygzk/oR4OvDgjVaAC8iItKZdbuQVe+y/PLNdAL8fPjzlcn4+LRPV3d3hQf7M31EFO+ty6PepSlDERGRzso7/Qq8yMfA9ZMTCQ3wJSa8/bq6N8fclDg+3pDPsh37OWNolLfLERERkRbodiHLGMMVE+K9XcZJTR8RTVigHwvT8xSyREREOqluN13YGQT5+zJjTAwfrd9LVW29t8sRERGRFlDI6qDmjoujrLqOzzft83YpIiIi0gIKWR3UlEGRRIcFskCNSUVERDolhawOytfHcElyLF9t2UdJRa23yxEREZFmUsjqwOamxFFbb1m8fo+3SxEREZFmcitkGWNmGGO2GGO2G2PmNbF9jjFmnTEm3RiTZoyZ5u6xcmJj4noyKCqUBWs0ZSgiItLZnDJkGWN8gSeAmcAo4FpjzKhjdvscSLbWpgA3A88141g5AWMMc1Pi+G7XAfKKK71djoiIiDSDOyNZk4Dt1tqd1toa4HVgTuMdrLVl9sgVjUMB6+6xcnKzk2MBeG+tLrMjIiLSmbgTsuKA7EaPcxqeO4ox5lJjzGbgA5zRLLePlRMb0CeUlIQIFqQrZImIiHQm7oSspi7ud9xF9ay18621I4C5wEPNORbAGHNLw3qutIKCAjfK6j7mpsSyac9BtuaXersUERERcZM7ISsHSGj0OB444bCKtXYJMNgY06c5x1prn7HWplprU6OidCmZxi4eG4uvj+kwC+A/25jPJxv2ersMERGRDs2dkLUSGGqMGWiMCQCuARY13sEYM8QYYxrujwcCgEJ3jpVTiwoLZOqQPixMz+PI0rf2Z63liS+386OX0rjt1TVs31fmtVpEREQ6ulOGLGttHXAb8DGwCXjTWrvBGHOrMebWht0uB9YbY9Jxzia82jqaPNYTH6Srm5sSS25xJauyirzy/nX1Lu5bsJ7//ngLFyf1I8jfh9/Mz/Bq6BMREenI/NzZyVq7GFh8zHNPNbr/R+CP7h4rzXfB6BiC/DNYkJ5L6oDe7freFTV13P7aGj7btI+fnj2Yuy4Yzusrs/n1/AzeWZ3LFRPi27UeERGRzkAd3zuJHoF+nDeyLx+s20Ntvavd3rewrJprn/2OLzbv46E5o7lnxgh8fAzXTExgQmIv/vDBRg6U17RbPSIiIp2FQlYnMjcljqKKWr7e1j5nX2YVlnP5P5exec9BnrphAjeeNuDwNh8fwyOXJlFaVccjize1Sz0iIiKdiUJWJ3LmsCgiQvxZsMbzPbPSs4u57MlllFTW8uqPp3DB6Jjj9hkeE8aPzxzE26ty+HZHocdrEhER6UwUsjqRAD8fLk7qx6cb8ymvrvPY+3y+KZ9rnvmWkEBf3vnp6UxI7HXCfW8/ZygJvYP5zYIMquvqPVaTiIhIZ6OQ1cnMSYmjsraeTzfme+T1X/kuix+/lMawvmG8+9OpDIrqcdL9gwN8eWjOGHYWlPPUVzs9UpOIiEhnpJDVyaQm9iIuIpgF6W3bmNRay58/3sJv5q/nrGFRvH7LFKLCAt069uzh0VySHMsTX25nZ4F6Z4mIiIBCVqfj42OYnRLL19v2s7+suk1es7bexa/eWsc/vtzONRMTePamVEIC3OrucdhvZ40k0N+H38xfr95ZIiIiKGR1SnNT4qh3WT5Yt6fVr1VaVcvNL67kndU53Hn+MB69LAk/3+b/WkSHBTFv5gi+3VnIu6s7xuV/REREvEkhqxMaHhPGiJgwFrZyyjD/YBVXP72cZTsK+dMVY7n93KE0XB2pRa6d2J/x/SP4w+JNFKl3loiIdHMKWZ3UnJQ4Vu8uZndhRYuO376vlMueXEZmYTkvfH8iV6UmnPqgU/DxMTxyWRIHK2t59EP1zhIRke5NIauTmp0SC9Ci0awVuw5w2ZPLqK5z8eZPTuOsYVFtVteImJ786IxBvJmWw/Kd6p0lIiLdl0JWJxUXEcykgb1ZkJ7brIXmizP2cMPz39EnLJD5PzudMXHhbV7bL84dSnyvYH4zX72zRESk+1LI6sTmpMSyo6CcDXkH3dr/+W928fNXVzM2Lpx3bj2dhN4hHqkrOMCXh+aOYUdBOU//R72zRESke1LI6sQuTuqHv6855ZShy2V56P2NPPT+Ri4cFcPLP5pMr9AAj9Y2fXg0F4/txz++3M6u/eUefS8REZGOSCGrE4sICeCsYdEsWptHvavpKcOq2nr+6/U1PP/NLr5/+gCeuH48Qf6+7VLfA7NGEejrw30LMtQ7S0REuh2FrE5u7rhY8g9W810Ti8xLKmq56YUVfLBuD7+5aCQPXDIKX5+Wt2horuieQdw9cwRLtxe2eYd6ERGRjk4hq5M7d0RfQgN8WZied9TzucWVXP7UMtJ3F/P4teP48ZmDWtUDq6Wun9SflIQIHnpfvbNERKR7Ucjq5IIDfLlwTAyL1++hqtY5k29DXgmXPrGU/INV/OvmScxOjvVafT4+hkcvS6KkspbHPtzstTpERETam0JWFzA3JY7Sqjq+2rKPr7cVcPXTy/H1Mbzz09M5bXCkt8tjZL+e/OiMgbyRls2KXQe8XY6IiEi7UMjqAk4fHEmfHgH86eMt/OD/VhLfK5j5P5vKsL5h3i7tsF+cO5S4iGB+PT+DmjqXt8sRERHxOIWsLsDP14dZY2PZWVDOpIG9efPW04gJD/J2WUcJCfDj4blj2L6vjGeW7PB2OSIiIh7n5+0CpG3cfu5QBkf34OrUBAL8OmZ2nj4imouT+vH4F9uZNTaWAX1CvV2SiIiIx3TMb2Nptt6hAdw4JbHDBqxD7r/kUO+s9eqdJSIiXVrH/kaWLqdvzyDunjGcb7bvZ9HavFMfICIi0kkpZEm7u25yIskJETz0/kaKK9Q7S0REuiaFLGl3vj6GRy4dQ1FFLX/8SL2zRESka1LIEq8YHRvOD6cN5LUV2azMVO8sERHpehSyxGvuOK+hd9a76p0lIiJdj0KWeE1IgB+/nzOabfvKePbrnV6ro7qunkVr83jwvQ1s31fmtTpERKRrUZ8s8apzR/bloqQYHv98G7PG9iMxsv16Z+0urODVFbt5Ky2bwvIajIFXlu/mtnOGcOtZgzt8OwwREenY9C0iXvfAJaPxb6feWXX1Lj7esJebXljBmf/9Jc9+vZPUAb146eZJfHfvuVwwui9/+XQrFz/+NWlaKyYiIq1gOmJDyNTUVJuWlubtMqQd/WtZJg8s2sDfr0lhTkpcm7/+npJKXl+Rzesrd5N/sJqYnkFcO6k/V09MOO4SRF9szue3CzaQW1zJ9ZP7c/eMEYQH+7d5TSIi0jUYY1ZZa1OPe14hSzqCepflsieXkltcyed3nk14SOtDjctlWbKtgFe+283nm/KxwFnDorh+ciLTh0fh53vigdzy6jr+8ulW/m/pLvr0COTB2aOZMSYGY0yr6xIRka6lVSHLGDMD+DvgCzxnrX3smO3XA/c0PCwDfmqtXduwLRMoBeqBuqaKOJZCVve0Ia+E2f9YytUTE3jk0qQWv87+smreTMvmtRW7yT5QSZ8eAVyVmsC1k/qT0DukWa+1LqeYee9ksHHPQc4b2ZffzxlNbERwi2sTEZGup8UhyxjjC2wFzgdygJXAtdbajY32OR3YZK0tMsbMBH5nrZ3csC0TSLXW7ne3WIWs7usPH2zk2a938c5PT2NCYm+3j7PW8t2uA7y8PIuPN+yltt5y2qBIrp/SnwtGxbRqEXttvYsXvtnFXz/biq8x3D1jBDdMScTXR6NaIiLSupB1Gk5ourDh8b0A1tpHT7B/L2C9tTau4XEmClnipvLqOi746xJ6BPrx/u3T8D/JlB5ASUUt76zO4ZXvsthRUE7PID+umJDAdZP7MyS6R5vWln2ggl/Pz+DrbftJSYjgscuTGBHTs03fQ0REOp8ThSx3/vc+Dshu9Din4bkT+SHwYaPHFvjEGLPKGHPLSQq8xRiTZoxJKygocKMs6YpCA/14cPZotuSXnrB3lrWWNbuL+NVba5n0yGf8/v2N9Az2589XJrPiN+dx/yWj2jxgAST0DuGlmyfxt6tT2H2gglmPf8OfPtpMVW19m7+XiIh0fu70yWpqTqTJ4S9jzHSckDWt0dNTrbV5xpho4FNjzGZr7ZLjXtDaZ4BnwBnJcqMu6aLOG9WXGaNj+Ptn25iVFEv/SGcdVVl1HQvTc3ll+W427jlIaIAvV0yI57rJ/RkdG94utRljmDsujrOGRfGHxZt48qsdLM7YwyOXJnH6kD7tUoOIiHQObTZdaIwZC8wHZlprt57gtX4HlFlr/3yy99R0oewtqeK8v/yH8Ym9uGfGcF79bjcL1uRSXlPPyH49uWFKf+akxNEj0Lv9dJdu38+v52eQVVjBFRPi+c1FI+kVGuDVmkREpH21Zk2WH87C93OBXJyF79dZazc02qc/8AVwk7V2WaPnQwEfa21pw/1Pgd9baz862XsqZAnAi0t38bv3nPMrAv18uCQ5lusn9yclIaJDtVKoqq3n8c+38cySnfQM9uf+WaOYkxLboWoUEekoCkqriQoL9HYZbepEIeuUwwDW2jpjzG3AxzgtHF6w1m4wxtzasP0p4H4gEniy4YvlUKuGvsD8huf8gFdPFbBEDrnxtAHkFlcSEx7MFePj26R3licE+fty94wRXJIcy7x3M7jjjXTeXZPLH+aOaXbLCBGRrqq4ooYHFm1gYXoe3zstkd/OGnXSfoVdgZqRirShepfl5eVZ/OmjzdRby53nD+PmqQO7/D8kIiIn8+WWfcx7Zx2FZTVMG9qHr7YUcMbQPvzjuvFd4ooarTm7UETc5Otj+N7pA/j0zrOYNiSKRxZvZs4TS8nIKfF2aSKA0yZlfa5+H6V9lFXXce+76/jB/60kPNifBT+fyos/mMSfLh/L8p2FXPbkUjL3l3u7TI/RSJaIh1hr+Wj9Xh5YtIH9ZdXcPHUgvzx/GKFeXqzf1ipq6thZUE5ecSVD+4YxIDJE69E6oEO/j79/fyN7Sqr4zUUj+fGZg7xdlnRhy3cW8qu31pJbXMktZw7izvOHEejne9T2W19eBcBTN0xgyqBIb5Xaarp2oYiXlFTW8qePNvPKd7uJiwjm4UvHMH14tLfLahZrLfkHq9lRUMaOgjJ2FpQ79/eVkVdSddS+kaEBjE/sxYSGn6S4cIL8fU/wytIesgrLuX/hBv6ztYARMWHERQTz+eZ93D1jOD87e4i3y5Mupqq2nv/+eAsvLN1F/94h/OWq5BNewSOrsJybX1xJVmEFf7h0DFdP7N/O1bYNhSwRL1uZeYB7381g+74yJg3oTf/IEKLCAokOC2y4DTr82FujXVW19WQWljshal9ZQ6gqZ2dBGeU1R5qu9gj0Y1BUKIOjejC44bZveBCb95SyKquI1buL2NUwBeDvaxgdG05qo+AV3TPIK5+vu6mqrefp/+zkia+24+9juPOC4XzvtEQA/t9ba1mYnsed5w/j9nOHerlS6SrSs4u58810dhaUc9NpicybOYKQgJP/e3awqpbbXl3Dkq0F/GjaQO69aGSnu2yZQpZIB1BdV8+zS3byycZ8CkqrKSitps51/H+DIQG+x4WvqMOPj9yPDA1s9j9G1loKy2saQlR5w8iUcz+7qILG/yTERQQfF6YGR/cgOizwlFOChWXVrN5dTFrWAVZnFbE2p4SaOhcA8b2CDweu8f17MSImTCcHtLElWwu4f+F6MgsrmDW2H/ddPIqY8CPhtt5luevttby7Opfbzx3KL88bqmleabGaOhf/+8U2nvxqB9Fhgfz3FclMG+p+g+a6ehcPf7CJF5dlMn14FI9fO46woM6zIF4hS6QDcrksxZW17CutoqC0mn0Hqykoa3xbRUFZNQUHqymtrjvueB8DkT0aj4YdH8yKymsOh6lDU30llbWHXyPI34eBfY4OUYP6hDIoKvSU/wfaHDV1LjbklRwe6UrLLGJfaTUAoQG+pPSPYEL/XoxP7MW4/r26xBlH3rC3pIqH3t/IBxl7GNgnlN/PGc0ZQ6Oa3LfeZbn33XW8mZbDz84ezF0XDlfQkmbbvPcgd76xlo17DnL5+Hjuv2RUi//7fXl5Fg8s2sDgqFCe/97ETtMGRyFLpJOrrKl3Rr/Kqo4OY6XVTkhreLy/rJomBseIDgtsCFGhDOrjhKnBUaHEhgfj44WheWstucWVrMoqOvyzac9BXBaMgaHRPZiQ2PvwiFdbLaivq3dRUVtPeXUd5dX1VNQ4t+XVdZTX1FFRc/S2wdE9uDipX4c/YaG23sW/lmXy10+3Uuey3DZ9CLecNeiohcZNcbks9y1cz6vf7ebHZwzk1xeNVNASt9S7LE8v2cFfP91KeLA/j1yaxAWjY1r9uku37+enL6/Cz9eHp2+cwMQBTa/n6kgUskS6iXqXpaii5nAQCw/2Z1BUKD07wdB7eXUda7OLndC1u4jVWUUcrHJG8HqHBjC+vxO4hkT3oOpQUKqpp6Lh9nBQqq6nvKaO8uqG0NQoSFU3TFm6w9/XUFtv6RHox+yUWK6d2J+k+Pa5TmZzpGUe4L4F69m8t5Tpw6N4cPaYw9f8dIe1lgcWbeClb7P4wdQB3D9rlIKWnNSu/eXc+WY6a3YXc1FSDA/PTaJ3G15SbGdBGT/8Vxq5RZU8clkSV0yIb7PX9gSFLBHpdFwuy/aCssMjXauzith5gp46AX4+hAb4EhLgR2igL6GBfoQG+BES0HA/0LfhsXP/8H4BfoQ03B7aL6ThOD8fw6qsIl5bkc0HGXlU1boYHduTayYmMGdcnNeD64HyGh5dvIm3VuUQGx7EA7NHc8Govi0KSNZaHnp/Ey8s3cWNUxJ5cPZor4xwSsfmcln+vTyLRz/cRICvDw/NHcPsZM9cRqykorxcnzgAACAASURBVJafvbqKpdsLufWswdx94fAO+zupkCUiXUJhWTU5RZVHglKAH8EBvgT4eXbhfEllLYvSc3ltRTYb9xwkyN+Hi5NiuXZSAhMSe7XryI/LZXkjLZs/frSZsqo6fnTGIG4/d0ir19BZa3nsw808vWQn105K4A9zkzrsl1pj+QerePLL7QT6+zI2PpyxcREk9A7WaFwbyymq4O6317FsRyFnD4/ij5ePpa+HzxSurXfxu0UbeOW73Zw/qi9/uzqlQ07dK2SJiLQBay0ZuSW8tiKbRem5lNfUMzS6B1dPTODy8fH0asMpk6aszy3hvgXrSc8uZvLA3jw8dwxD+4a12etba/nzJ1t44ssdXDkhnscuH9thT6e31vL2qhween8jVbXONHBNvXMbHuzP2PhwkuLCndv4CGLDgxS8WsBay1urcvj9exux1nLfrFFcMzGh3f4srbW89G0WD763geExPXnue6nERQS3y3u7SyFLRKSNlVfX8f66PF5bkU16djEBvj5cOCaGaycmMGVQZJuOAh2squUvn2zlpW8z6R0awK8vGsml4+I88kVnreXvn2/jb59t47Jxcfz3lckdLmjlFldy77sZLNlawKQBvfnjFWOJiwhma34pGbklrMspYV1OMVv2lh5ukxIZGkBSfDhj45zQNTY+3OMjMZ3dvtIqfv1uBp9t2sfkgb3585XJXjvj7z9bC7jtldUE+vvyzE0TGN+/l1fqaIpCloiIB23ee5DXV2Tz7uocDlbVkRgZwtUTE7hiQjzRYS3/IrfWsmhtHg9/sIn9ZdXcMDmRX10wnPAQz68H+8cX2/jzJ1u5JDmWv16V3CF6mblcltdW7ubRxZtxWcs9M0Zw45TEEwbaqtp6Nu8tJSOnmHU5JWTklrA1v/TwGbjRYYENI14RDSNe4fTpEdiOn6jjen9dHvctWE9FTT13Xzicm6cO9Pr08fZ9pdz8Yhp7D1bx31eMZU5KnFfrOUQhS0SkHVTV1vPh+j28tiKbFbsO4OdjOHdkNNdM6s+ZQ6OaNSK0o6CM+xeuZ+n2QsbGh/Pw3DGMjY/wYPXHe+o/O3jsw81clBTD368Zh78Xg9buwgrueWcd3+4sZOqQSB67bGyLRlUqa+rZuMcZ7crIKWFdbgk7CsoON+KNiwgmKc4JXIemHCNCPDsN3JEUlddw/6INvLc2j+T4cP7nqmSGRLfdlHRrHSiv4daXV7Fi1wH+65wh/PK8YV4PfwpZIiLtbEdBGW+szOadVTkUltcQFxHMlanxXJWaQOxJ1pRU1tTzxJfbeXrJDoL8fbl7xgium9Tfa1N2z329k4c/2MQFo/ryj+vGe/wkg2O5XJZ/fZvJnz7agp+P4TcXj+TqNl4TVFZdx4bcksNTjRm5JYcvDQXQv3dIo6nGcMbGR9DD0wuwK4uhuhQiEjz7Po18sTmfe97JoKi8hl+cO5Sfnj24Q4xgHqumzsV9CzJ4My2Hi5Ji+J8rUwgO8N41UhWyRES8pKbOxacb83l95W6+3rYfHwNnDYvimkn9OWdE9FGjQ59vyueBRRvIKarksvFx3DtzJFFh3p++enHpLn733kbOHRHNkzeMP2WT07ayo6CMe95eR1pWEWcPj+KRS5NOGlDbUkllLRtynZEuZ8SrmOwDlcCRhrnjEnqR0j+ClIQIhvUNa5sgXHEAvn0CvnsaakohYTKkXAejL4Ugz/RpK62q5eH3N/FGWjbD+4bxl6uTGR3b8XrCNWat5flvdvGHxZsYExvOszelHnXpqPakkCUi0gFkH6jgjZXZvLUqm/yD1USFBXLlhHjOGRHN00t28unGfIZG9+ChuWOYMijS2+Ue5eXlWdy3YD1nDYvi6RsnEOTvuaBVV+/i+W928ZdPtxLo58MDl4zmsvGeWejfHEXlNazLLSF9dzHp2UWkZxdTVOFcpiokwJekuHBS+kcwLiGClIRezfvSryyCb5+E756C6oMwag7EjIV1b8D+reAXBCMvcQLXwLPAp/l//tZa8g9Ws21fKVvzy9iWX8rWfOd+RU0dPzlrMHecN7TdQnRb+HxTPre/tobQQD+e+15qu0+pg0KWiEiHUlfv4sstBby+YjdfbtmHy0Kwvy93nDeUm6cN9Orap5N5fcVu7p2fwdTBfXj2plSPTNFszS/lrrfWsjanhAtG9eXhuWOIPtVZgLbhekztzFpLVmEF6dnFpGcXsya7mE15Bw+3kojpGURKQsTh0a6x8eHH9zOrLIbl/3R+qktg5Gw46x6IGXPoTSB3NaS/AuvfhqoS6BkHyddA8nXQZ0iTde0rrWZbfhlb80uPClWHrqIAzpUUhkb3YGjfHlw2Pr5DnbHXHJv3HuSHL6ZRWF7N/1yZwsVj+7Xr+ytkiYh0UHtLqliytYCpQ/t0uP4/TXl7VQ53vb2WKQMjef77qW12IfHaehdPfbWDx7/YRliQPw/OHs2ssf2aHr0q3QvZ30H2Cud2z1roOwbG3QBjLofg9h/NOKS6rp6NeQcPB6/07GKyCisA56Luw/qGMa5/LybF+HJG0dtEZjyHqSqBEbPg7HkQk3TiF6+tgi2LIf1V2PE5WBc1sRPZnTCX5cFnsuGAYVt+Kdv2lR11IfiIEH+GRYcxtG8PhvU9ctuVzqTcX1bNT/69ilVZRdx5/jD+65wh7TbyqZAlIiJtZmF6Lr98I53UxN688IOJrV4Evj63hLvfXsfGPQeZNbYfD84eTeShAOCqh30bnTC1+zvntjjL2eYbCLHjoN9YyPzG2c8vyJlqG3cDJE4DH++PChaWVbM2p5j03cVsysplbO7r3GTfI9xU8LlN5bPoH9BrcOrhUa9j235Ya9lfVnNkem9fGfvzMhm1/yMurv+SoT65VFl/PjeTWd1rJlUJ0xjaN7whUIXRp0eA96ZaywshdxXkpjm3+RvB+IBfgPN35Rfo/D36BTY8Pvb5hvuHfo7at9F93wBqjD//uySHjzYXMW1EHPfMSiYo0vMnDihkiYhIm3p/XR6/eD2dlIQIXvzBRMJacC3H6rp6/vHFdv751Q4iQgJ4eO4YZgwJgZyVR0apctKcBeAAodHQfzIkTHEWhPcb63zJgjOtlrcG1rwMGW87U28RiU7YSr62Xc/Sa1J1qbPeatk/oKqY8gHnszzhFr4qjSU9u5hNew4ebpwaFxFMSkIE4SH+bM8vY9u+0sNrvwB6BvkdDlDDokMZ75fJ0D2LCN7yrjMqFhbrTCemXAd9hrbfZ6ythD3rjg5VRZnONuMDUSOdvzPjC/XVUFcFddWNfqqgvqbp5121J33rplQTQNXdeR7vK6eQJSIiDld9ixZNN+XDjD3812trGB0Xzks3TyI82P0vs/TsYu5+K53Kgp3cOrCQK6JzCNyTBvkbAAsYZwowYZITqPpPdkKTOyMyNRWw+X1Y82/YtcR5rcHnOIFrxMVHgll7qC6FFc/Asv91FrcPm+GsuYobf9RuVbX1bMgrYc1uZ21X+u5iSqtqnSDVtwdDo8MY1nA/Kiyw6ZGp2irY+qEznbj9M7AuiJ/khK0xl7Xt2YkuFxRuc0LwoVCVvwFcDWu+esZD/ASIa/jplwKBPVr3fqcMZlVQ59zPyNrH2uwDXHvLvR5vf6KQJSLSXdWUQ+ZS2PGF87N/CwT2hJDeEBJ59E9wr+OfO/S8b9NTgp9s2MvPX13N8JgwXv7h5JM37qyrpjp7Nd988QG1mcuZ6LuNSIqdbQFhEJ8K/ac4wSouFYJ6tv7zF2U6oWPNK3Awx/ksSVfB+BtPvv6ptarLYOWzsPRxqDwAQy9w1lzFTfDcezZWutc5MzH9VSjY7EypjZjlBK5BZzc/aJfudcLUoVCVt8Y5CxKc36fYcc7f36FQFRbT1p+ow1LIEhFpjrpqWP8u1FU6X/bRo04YMjoclwv2rm0IVV/C7uXOVItfECROdUZQqsugotD58q8obPg5ADVlJ37doIgmApgT1DYd9Ofvyw4Q2iua3151BhGRMc7+FYWQs8KpIXsFrtzV+LhqACgMiCVs6FQCBpzmjFRFj2yzEbam/1zqYedXznTi5ved0Y9+yTDuRki6wglfbaGmHFY+B0v/7nz+IefB2fc6AcQbrIW81U7YyngbqopPPZ1YXQZ70huNUq2Cg7nONh8/Z4QxbsKRUBU5tEOsffMWhSwREXdUHYS0F5zT6cv2HnneP8SZ7oifAPETneAV3jGumwZASY4TqHZ+6dxWHnCej0lypskGTYf+p4H/KVoh1FYdE7wawtdRjxs9X77fmcJpivFxpqsA6+NPTvBwPipJZFfIaObMuozJY0e24R9AM1UccALHmpdgb4azmHrkJc504sCzWhYYaioahav9MPhcZ+QqYVLb199SddWw5dB04qdHphOTr3ECbk6a0y6iYNPhvzt6DXB+3w+Fqpgk8O/4Z8G2J4UsEfGsmnIICPV2FS1Xts8JViufdxZMDzobpt7hfMEcmiLJWQl71zkjIABh/Y588cRPbP2ak+ZoagoQoEdfJ1QNPsf5DD2iPVuHtVBbcTh4bdi+i//7bBUDg6v5/rgehIb1IsMM546vYUdRPd87LZG7Z4wg1NOXpGmOPWud0a11bzqjPOH9nRGecddDRP9TH19T4QTzpX+D8gIn0J59r7OGrCMr3et85vRXnOlEcEbz4iYcCVVxEyC0YzXF7YgUskSk7dXVONMuK5+HrG+ckJFyvTP1EtLb29W558BOZ0Hymlec8DRqthOujlmUfFhdNexd7yzyzVnphK+iXc424+NMK8anOl9S8anQZ3jbTKOcagpw8DkweLrz/l7uir58ZyE3v7iSmJ5BTBrYm9dXZjMgMoQ/XZHMpIEd+PeitqphsfzLzrQiwKCznOnEERcfP3pTWwlp/+eEq7J8ZwTs7Hsh8bR2L71VrD3S+qL3IK///nRGClki0nZKcmHVi7D6X86XS0SiM9Wy6z8NUy8BMPwiJ3ANPqdjrmXas875ctww31ljknwtnH57k92zT+lQH6CclUdOW68qcbYFhEHcuCNTjPGp7o8unWgKsG+SE6gGn+PeFKAXrMw8wPdfWEFlbT0/nDaQO88f7tUL+DZb8W5Ifw3SX3buB4VD0pXOdGLUSOf3/5u/OlPKA86A6b+GxNO9XbV4iUKWiLSOywW7vnJGrbYsdv7vd9iFMPFHztqTQ6M1e9Y56z3WveGEgrB+MPZqJ3BFDfPqR8Bap2HlN391umUHhEHqD2DKz6BnG16Gw+WCAzuOTDEee2p7eP+GKcaGEa9+yU5Q6ihTgG1kR0EZNXUuRvZrgzMEvcXlgsyvnVYQGxc568/8Q5wp0sRpMP1eGDDN21WKlylkiUjLVBY5oWnl805wCImE8TfBhB9Ar8QTH1dXA1s/co7d9gnYemc0J+X6tu/XcyouF2z5AL75mxN4QqNgyk8h9Yftd/mV2kpn7c+hKcbcVVCS7Wzz8YPIIVC445gpwIbRqg4wBSg4/y2sfwdyVkHKtTDwTG9XJB2EQpZIR3Bgp7O4uijL6QU04AyITQFfz3YjbpG8Nc6ZUhnvOG0MEiY7o1aj5jS/kWNpfkO/nleO9OsZeYkTuFp6Jpc76mog403nbK/9W51F7Kff7ixq7ghnRx3uO7TSGemKGtGhpwBFpGkKWSLelJMGyx6HTe85l5PoNcDplAwQ0ONI4BpwhjN15K01TLWVTm+olc85fXX8Q2Hslc6IT7+xrX/9Q/161rwC69921i31jHdGBVKucxbdtoXqUlj1L/j2CSjNc045n3oHjJrbMdeHiUin1qqQZYyZAfwd8AWes9Y+dsz264F7Gh6WAT+11q5159imKGRJl+BywbaPnW7Pu5dBYDhMvBkm/cRZ/1NW4JyRl/kN7Pr6yPqbgDDn7KQB046ELk82aARnmirtBeesqqpi54y4iT+C5Ks9N61XW+VM4a15xVl/hHWmyFKud0bLWtIKoXy/c224Fc86n2PAGTDtDmfNmKbbRMRDWhyyjDG+wFbgfCAHWAlca63d2Gif04FN1toiY8xM4HfW2snuHNsUhSzp1GqrnKmxb//hTFGFJzgLq8ffCIFhJz6ubJ+zwDazIXjt3+o8H9jTOWvpUOiKSWqb0OWqh60fO6NWOz531gWNmOWEqwHT2jeUlOTCutedwHVghzOCNnquE7gSTz91LUVZzp/36n871y4bcTFM+6X3OmyLSLfSmpB1Gk5ourDh8b0A1tpHT7B/L2C9tTauucceopAlnVLFAUh7Hr57Bsr3OWHo9F84YaEla65K9zYErobgVbjdeT4o3BnxGTDN+emb1Lw1TWX7nNYLq/7lLLwO6+csYh9/U9ueYdcS1kL2d86I2ob5ziVeeg1wwlbytRCRcPT++Ruc9VYZbzs9qpKvdv7MvX0Wo4h0KycKWe4sTogDshs9zgFO1sb2h8CHzT3WGHMLcAtA//5udNgV6SiKMp3F7Kv/DbXlznXKTv8vZ0F3a0aDwmKcpp5JVziPD+Y5p/dnLnFC15bFzvNBEU7oGniGE7qiRx8fuqyF3d86o1YbFzlnsA06G2Y8CsNmdpx1SsY469P6T4GZf3TWsK15Gb78A3z5iHM217gbnGC47H+d6Vj/UOdMwSk/61iXuRGRbs+df1mb+pZocvjLGDMdJ2Qdahri9rHW2meAZ8AZyXKjLhHvylvjrLfauMAZRUm60glXfUd75v16xjqL0Mde6TwuyW000vW1s74JnMtiJE51AknCZKdlwcrnnY7OgeEw6ceQenPTF4XtSAJCneupJV/jTAeufc05O/HdHzvbQyJh+n0w8Yedp7u8iHQr7oSsHKDxGH08kHfsTsaYscBzwExrbWFzjhXpNFwu2P6Zc6Zg5tfOeqnTboPJt7b/KEp4nDM9lny187g4+8h6rswlzuVBDokZC7P/F8Zc3jmvL9gr0bnQ7pl3OycLHNzjtIAICPF2ZSIiJ+TOmiw/nMXr5wK5OIvXr7PWbmi0T3/gC+Ama+2y5hzbFK3JksPqa52WB57qo+SuumrIeMuZoirYDGGxzhTVhO+1b1PN5ijKctY39R7kXORVZ9eJiHhEi9dkWWvrjDG3AR/jtGF4wVq7wRhza8P2p4D7gUjgSeP8Q15nrU090bFt9qmk66o66ASab59wLkUSkeBcH69X4jG3A5zpMU8FiMpip7XBd0871yjrOwYufRpGXwZ+AZ55z7bSK/HkHdlFRMSj1IxUOpa6GifULPkTVBQ6zSMj+kNxljMyU5zlXNqisYCwJsJXo9uWTI8VZzcsZv+Xc4bboLOdTuGDz9GIkIiIHKU1ZxeKeJ7LBRvehc9/7wSpgWfCeQ9C3Pjj9606eHToOnR7YCfs/NK5cGtjIX1OHMLCE44ekdqz1hlBW/+u83jM5c5i9rbodi4iIt2KQpZ4344v4NMHYO86p+fTDe+cvEN3UE+nB1VM0vHbrHW6fhdnOa0VGoewvDVOSwBX7ZH9jY+zvqpXIliX0+YgoIez3mryrcf3ZRIREXGTQpZ4T94a+Ox3sPMrZ0rwsmdhzBWtW+RuDPSIcn6a6vbtqofSPcePghVlOSNk5z0IE74PwREtr0FERASFLPGGAzvhi4dh/TsQ3BtmPOb0bfIL9Px7+/hCeLzzw1TPv5+IiHRbClnSfsoKnAXtaS+Ajz+c8SuYenvHbYEgIiLSCgpZ4nnVZU4rhmWPQ22lc428s+c5l40RERHpohSyxHPqa2HVi/CfP0J5AYycDefe3/Ev5yIiItIGFLKk7VkLG+bDFw85668Sp8I1r0HCRG9XJiIi0m4UsqRt7VoCn97vnDkYPQquexOGXqAGniIi0u0oZEnb2JvhtGPY/hn0jIe5/4SxVztn84mIiHRDClnSOkVZ8OUfYN2bzlmCFzwME38M/kHerkxERMSrFLKkZcoL4es/w8rnnK7pU38B036pJp4iIiINFLKkeQ7sgjX/hhXPOhdOTrkezr4XwuO8XZmIiEiHopAlp1Zb6Vzzb/VLkPk1YGDExXDObyF6hLerExER6ZAUsuTE8tKdUauMt6CqxLm+4PTfQMp1DZelERERkRNRyJKjVRbBurdgzUvOGYO+gTBqNoy7EQac0bqLN4uIiHQjClkCLhdkLoHV/3amBeurIWYsXPRnSLoCgnt5u0IREZFORyGrOyvJgfRXYc3LUJzltGAYfxOMvxH6JXu7OhERkU5NIau7qauBLYudtVY7vgDrgoFnOovYR84C/2BvVygiItIlKGR1F/s2OdOB616HikIIi4Uz/p/TgqH3QG9XJyIi0uUoZHVlVQdhw7tOuMpNAx9/GD7TmRIcfI4ueSMiIuJBClldjbWwe7kzHbhhPtRWQNQIuOAPkHwNhPbxdoUiIiLdgkJWV1Fd5lziZs3LULgNAno4ZwaOuwniU8EYb1coIiLSrShkdQWVxfDKFZCzEhKmwLQ7YNRcCOzh7cpERES6LYWszq7iAPx7LuRvhKtfhpGXeLsiERERQSGrcysrgJfmQOF2uOZVGHaBtysSERGRBgpZndXBPfDSbCjOhuvegMHTvV2RiIiINKKQ1RkVZzsBq2wf3PAODJjq7YpERETkGApZnU1RJvzrEqgsgRsXQMJEb1ckIiIiTVDI6kz2b3dGsGrK4XsLIXactysSERGRE1DI6iz2bXYClqsevv8BxIzxdkUiIiJyEj7eLkDcsDcDXrwIMPCDxQpYIiIinYBbIcsYM8MYs8UYs90YM6+J7SOMMd8aY6qNMb86ZlumMSbDGJNujElrq8K7jdxV8OIs8At2AlbUcG9XJCIiIm445XShMcYXeAI4H8gBVhpjFllrNzba7QBwOzD3BC8z3Vq7v7XFdju7v3M6uQf3gu+9B70SvV2RiIiIuMmdkaxJwHZr7U5rbQ3wOjCn8Q7W2n3W2pVArQdq7J4yv4F/XwqhUc4IlgKWiIhIp+JOyIoDshs9zml4zl0W+MQYs8oYc8uJdjLG3GKMSTPGpBUUFDTj5bug7Z/Dy1dARIITsMLjvV2RiIiINJM7Ics08ZxtxntMtdaOB2YCPzfGnNnUTtbaZ6y1qdba1KioqGa8fBez5SN47RqIHOKcRRgW4+2KREREpAXcCVk5QEKjx/FAnrtvYK3Na7jdB8zHmX6UpmxcBG/cAH1Hw/cWQWgfb1ckIiIiLeROyFoJDDXGDDTGBADXAIvceXFjTKgxJuzQfeACYH1Li+3SMt6Gt77vNBi9aSGE9PZ2RSIiItIKpzy70FpbZ4y5DfgY8AVesNZuMMbc2rD9KWNMDJAG9ARcxpg7gFFAH2C+MebQe71qrf3IMx+lE1vzCiz8OSROheteh8Awb1ckIiIireRWx3dr7WJg8THPPdXo/l6cacRjHQSSW1Ngl5f2Arz/Sxg0Ha55FQJCvF2RiIiItAF1fPem5f90AtbQC+Ha1xWwREREuhCFLG/55m/w0TwYeQlc/TL4B3m7IhEREWlDukB0e7MW/vMn+OoRGHMFXPo0+OqvQUREpKvRt3t7shY+/z188xdIuR5m/y/4+Hq7KhEREfEAhaz2Yi18/GtY/iRM+AFc/Bfw0WytiIhIV6WQ1R5cLlj8K0h7Hib/FGY8CqapRvoiIiLSVShkeZqrHt67Hda8DFPvgPN+p4AlIiLSDShkeZK1sOCnsO4NOGsenD1PAUtERKSbUMjypD1rnYB1xv+D6fd6uxoRERFpR1p57UlZy5zb1B96tw4RERFpdwpZnpS1FHoNgPA4b1ciIiIi7Uwhy1NcLmckK3GqtysRERERL1DI8pT9W6DyACSe7u1KRERExAsUsjwla6lzq5EsERGRbkkhy1Myl0JYrLMmS0RERLodhSxPsLZhPdbp6oslIiLSTSlkecKBnVC2FwZoqlBERKS7UsjyhEP9sbQeS0REpNtSyPKErKUQEgl9hnm7EhEREfEShSxPyFqq9VgiIiLdnEJWWyvOhuLdkDjN25WIiIiIFylktbXD67HUhFRERKQ7U8hqa1lLITAc+o72diUiIiLiRQpZbS1rGSSeBj6+3q5EREREvEghqy2V5kPhNk0VioiIiEJWm9qt/lgiIiLiUMhqS1nLwD8U+iV7uxIRERHxMoWstpS1DBImga+/tysRERERL1PIaisVByB/g6YKRUREBFDIaju7lwNWF4UWERERQCGr7WQtBd9AiB3v7UpERESkA1DIaitZSyE+FfyDvF2JiIiIdABuhSxjzAxjzBZjzHZjzLwmto8wxnxrjKk2xvyqOcd2CdWlsGet+mOJiIjIYacMWcYYX+AJYCYwCrjWGDPqmN0OALcDf27BsZ1f9ndgXVr0LiIiIoe5M5I1Cdhurd1pra0BXgfmNN7BWrvPWrsSqG3usV1C5lLw8XPaN4iIiIjgXsiKA7IbPc5peM4drTm288haBv1SICDU25WIiIhIB+FOyDJNPGfdfH23jzXG3GKMSTPGpBUUFLj58h1AbSXkrlLrBhERETmKOyErB0ho9DgeyHPz9d0+1lr7jLU21VqbGhUV5ebLdwA5aeCq1XosEREROYo7IWslMNQYM9AYEwBcAyxy8/Vbc2znkLUUMJAw2duViIiISAfid6odrLV1xpjbgI8BX+AFa+0GY8ytDdufMsbEAGlAT8BljLkDGGWtPdjUsZ76MF6RtRRikiA4wtuViIiISAdyypAFYK1dDCw+5rmnGt3fizMV6NaxXUZdDWSvhAnf93YlIiIi0sGo43tr5K2Buko1IRUREZHjKGS1RtZS51YhS0RERI6hkNUaWcsgagSE9vF2JSIiItLBKGS1VH0d7F6uUSwRERFpkkJWS+VnQE2p+mOJiIhIkxSyWiprmXOrkSwRERFpgkJWS2Utg14DoWestysRERGRDkghqyVcLufMQk0VioiIyAkoZLVEwWaoLNJFoUVEROSEFLJaQv2xRERE5BQUsloiayn0jIOIRG9XIiIiIh2UQlZzWessznYjlgAADBpJREFUek+cCsZ4uxoRERHpoBSymuvATijL11ShiIiInJRCVnNlfuPc6sxCEREROQmFrObKWgahUdBnqLcrERERkQ5MIau5spY5U4VajyUiIiInoZDVHMW7oWS3pgpFRETklBSymkPXKxQRERE3KWQ1R9ZSCAqH6NHerkREREQ6OIWs5shaBv1PBx/9sYmIiMjJKS24q3QvFG7XVKGIiIi4RSHLXYfWY+mi0CIiIuIGhSx3ZS0D/1CISfZ2JSIiItIJKGS5K2sp9J8Mvn7erkREREQ6AYUsd1QcgH0b1R9LRERE3KaQ5Y7d3zq3ClkiIiLiJoUsd2QuBd9AiBvv7UpERESkk1DIckfWUoifCH6B3q5EREREOgmFrFOpOgh716l1g4iIiDSLQtapZK8A61ITUhEREWkWhaxTyfoGfPyc6UIRERERNylknUrWMogdDwGh3q5EREREOhG3QpYxZoYxZosxZrsxZl4T240x5vGG7euMMeMbbcs0xmQYY9KNMWltWbzH1VRA7mpNFYqIiEiznbJ9uTHGF/j/7d1/jBxlGcDx70NLJVSg/ChQ2vJLG0ONIM1ZQJBoVEIbQtU/tMQEgiYNCSTyh4kkJIR/0agJhkAwEsGgEKNoY4pgjInxakmPCoVakNJ0y9FCD9AiEChHH/+Yudxm3e0t7c3O3e73k0xmd+Z922efvjv7dN7Z2buALwOjwOaIWJ+Z/2xqtgpYVi4XAXeX6wlfyMzXpi3qXhndDAff9/5YkiTpQ+vmTNZKYEdm7szMA8BDwJqWNmuAB7KwCVgQEYumOdbea2yEOKr4OR1JkqQPoZsiazHwUtPz0XJbt20SeDwinoyIdYcbaC0aw3D6p+CYE+qORJIkzTLdFFnRZlt+iDaXZuYKiinFGyPi8rZ/ScS6iBiJiJGxsbEuwqrY+HvFdKFThZIk6TB0U2SNAkubni8B9nTbJjMn1vuARyimH/9PZt6bmUOZObRw4cLuoq/Snn/A+LsWWZIk6bB0U2RtBpZFxDkRMQ9YC6xvabMeuLb8luHFwP7M3BsR8yPiOICImA9cATw7jfFXpzFcrM+8pN44JEnSrDTltwszczwibgIeA+YA92Xmtoi4odx/D7ABWA3sAN4Bri+7nwY8EhETf9cvM/OP0/4qqtDYCAvPg/kn1x2JJEmahaYssgAycwNFIdW87Z6mxwnc2KbfTuCCI4yx9z4Yh92b4Pxv1B2JJEmapbzjezuvbIUDb/mj0JIk6bBZZLXT2Fisz/RO75Ik6fBYZLXTGIaTzoXjZ//9VCVJUj0sslodPFicyfLWDZIk6QhYZLUa2w7v/sciS5IkHRGLrFa7yvtjneX1WJIk6fBZZLVqDMMJS+HEs+qORJIkzWIWWc0yy+uxPIslSZKOjEVWs9dfhLf3WWRJkqQjZpHVrPG3Yu1F75Ik6QhZZDVrbIT5p8LJH687EkmSNMtZZDWbuB6r+EFrSZKkw2aRNeHfDdj/klOFkiRpWlhkTZj4vUJ/FFqSJE0Di6wJjWE4ZgEsPK/uSCRJUh+wyJrQGC6uxzrKlEiSpCNnRQHw5l54Y6fXY0mSpGljkQWwu7wey5uQSpKkaWKRBcWPQs87Dk4/v+5IJElSn7DIguKbhWdeBHPm1h2JJEnqExZZb78OY9udKpQkSdPKImv334u1F71LkqRpZJHVGIa5x8AZF9YdiSRJ6iMWWY1hWPIZmPuRuiORJEl9ZLCLrHf3wyvPOFUoSZKm3WAXWbufgDzoRe+SJGnaDXaR1RiGo44upgslSZKm0YAXWRth8QqYd2zdkUiSpD4zuEXWgbdhzxanCiVJUiUGt8ga3QwHx+Gsy+qORJIk9aHBLbIaGyGOgqUr645EkiT1ocEusk4/H445vu5IJElSHxrMImv8vWK68GynCiVJUjW6KrIi4sqIeD4idkTELW32R0TcWe7fGhEruu1bi5e3wPi7XvQuSZIqM2WRFRFzgLuAVcBy4JqIWN7SbBWwrFzWAXd/iL691xgu1mdeUm8ckiSpb3VzJmslsCMzd2bmAeAhYE1LmzXAA1nYBCyIiEVd9u29xjCcuhyOPanuSCRJUp/qpshaDLzU9Hy03NZNm276AhAR6yJiJCJGxsbGugjrCJx4Niyvv9aTJEn9a24XbaLNtuyyTTd9i42Z9wL3AgwNDbVtM22u+nGlf7wkSVI3RdYosLTp+RJgT5dt5nXRV5Ikqe90M124GVgWEedExDxgLbC+pc164NryW4YXA/szc2+XfSVJkvrOlGeyMnM8Im4CHgPmAPdl5raIuKHcfw+wAVgN7ADeAa4/VN9KXokkSdIMEpnVXv50OIaGhnJkZKTuMCRJkqYUEU9m5lDr9sG847skSVLFLLIkSZIqYJElSZJUAYssSZKkClhkSZIkVcAiS5IkqQIWWZIkSRWwyJIkSaqARZYkSVIFZuQd3yNiDGjUHUfNTgFeqzuIGcJcFMzDJHMxyVxMMhcF8zCpV7k4KzMXtm6ckUWWICJG2t2ifxCZi4J5mGQuJpmLSeaiYB4m1Z0LpwslSZIqYJElSZJUAYusmeveugOYQcxFwTxMMheTzMUkc1EwD5NqzYXXZEmSJFXAM1mSJEkVsMiqUUQsjYi/RMT2iNgWEd9p0+bzEbE/Ip4ql9vqiLUXImJXRDxTvs6RNvsjIu6MiB0RsTUiVtQRZ5Ui4hNN/9ZPRcSbEXFzS5u+HRMRcV9E7IuIZ5u2nRQRf4qIF8r1iR36XhkRz5fj45beRV2NDrn4QUQ8V47/RyJiQYe+h3wvzTYdcnF7RLzc9D5Y3aFv34yLDnl4uCkHuyLiqQ59+21MtP38nHHHi8x0qWkBFgErysfHAf8Clre0+Tzwh7pj7VE+dgGnHGL/auBRIICLgSfqjrnifMwBXqG4/8pAjAngcmAF8GzTtu8Dt5SPbwHu6JCrF4FzgXnA063vpdm2dMjFFcDc8vEd7XJR7jvke2m2LR1ycTvw3Sn69dW4aJeHlv0/BG4bkDHR9vNzph0vPJNVo8zcm5lbysf/BbYDi+uNakZbAzyQhU3AgohYVHdQFfoi8GJmDsyNeTPzr8AbLZvXAPeXj+8HvtKm60pgR2buzMwDwENlv1mrXS4y8/HMHC+fbgKW9DywGnQYF93oq3FxqDxERABfB37V06BqcojPzxl1vLDImiEi4mzgQuCJNrsviYinI+LRiPhkTwPrrQQej4gnI2Jdm/2LgZeano/S30XpWjofMAdlTACclpl7oTiwAqe2aTNoYwPgWxRndtuZ6r3UL24qp07v6zAtNEjj4nPAq5n5Qof9fTsmWj4/Z9TxwiJrBoiIjwK/AW7OzDdbdm+hmC66APgJ8Ltex9dDl2bmCmAVcGNEXN6yP9r06cuvx0bEPOBq4Ndtdg/SmOjWwIwNgIi4FRgHHuzQZKr3Uj+4G/gY8GlgL8VUWatBGhfXcOizWH05Jqb4/OzYrc22SsaFRVbNIuJoigHyYGb+tnV/Zr6ZmW+VjzcAR0fEKT0Osycyc0+53gc8QnFKt9kosLTp+RJgT2+i67lVwJbMfLV1xyCNidKrE9PC5XpfmzYDMzYi4jrgKuCbWV5g0qqL99Ksl5mvZuYHmXkQ+CntX+NAjIuImAt8DXi4U5t+HBMdPj9n1PHCIqtG5Rz6z4DtmfmjDm1OL9sRESsp/s1e712UvRER8yPiuInHFBf4PtvSbD1wbfktw4uB/ROnhftQx/+VDsqYaLIeuK58fB3w+zZtNgPLIuKc8izg2rJfX4mIK4HvAVdn5jsd2nTzXpr1Wq7H/CrtX+NAjAvgS8BzmTnabmc/jolDfH7OrONF3d8QGOQFuIziFOVW4KlyWQ3cANxQtrkJ2Ebx7YdNwGfrjruiXJxbvsany9d7a7m9ORcB3EXxrZBngKG6464oF8dSFE0nNG0biDFBUVjuBd6n+N/mt4GTgT8DL5Trk8q2ZwAbmvqupviG0YsT42c2Lx1ysYPiWpKJ48U9rbno9F6azUuHXPyiPA5spfiAXNTv46JdHsrtP584PjS17fcx0enzc0YdL7zjuyRJUgWcLpQkSaqARZYkSVIFLLIkSZIqYJElSZJUAYssSZKkClhkSZIkVcAiS5IkqQIWWZIkSRX4Hw4lJXJTSCsSAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Extracting the train and the test error for each k in a list for easy plotting\n",
    "\n",
    "kltest = []\n",
    "vltest = []\n",
    "for k, v in knn_many_split.items():\n",
    "    kltest.append(k)\n",
    "    vltest.append(knn_many_split[k][1])\n",
    "\n",
    "kltrain = []\n",
    "vltrain = []\n",
    "\n",
    "for k, v in knn_many_split.items():\n",
    "    kltrain.append(k)\n",
    "    vltrain.append(knn_many_split[k][0])\n",
    "\n",
    "# Plotting k vs error \n",
    "plt.figure(figsize=(10,6))\n",
    "plt.plot(kltest,vltest, label = 'test' )\n",
    "plt.plot(kltrain,vltrain, label = 'train')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observations:**\n",
    "- We can see that the train error and the test error are more or less similar for K more than or equal to 8.\n",
    "- This implies that we would get generalized results if we choose K=8.\n",
    "- So, let's scale the full data and fit the model on the entire training set. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "# scaling the data\n",
    "scaler=StandardScaler()\n",
    "x_train_scaled=pd.DataFrame(scaler.fit_transform(x_train), columns=x_train.columns)  #fit_transform the training data\n",
    "x_test_scaled=pd.DataFrame(scaler.transform(x_test), columns=x_test.columns) #transform the testing data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(n_neighbors=8, weights='distance')"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fitting the model on the scaled data\n",
    "knn = KNeighborsClassifier(n_neighbors = 8, weights='distance')\n",
    "knn.fit(x_train_scaled, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Now, let's check the performance of the model on the training and testing data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00       131\n",
      "           1       1.00      1.00      1.00       298\n",
      "\n",
      "    accuracy                           1.00       429\n",
      "   macro avg       1.00      1.00      1.00       429\n",
      "weighted avg       1.00      1.00      1.00       429\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#predict on train data\n",
    "y_pred_train_knn =knn.predict(x_train_scaled)\n",
    "\n",
    "#checking performance of the model\n",
    "metrics_score(y_train, y_pred_train_knn)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observations:\n",
    "\n",
    "- KNN model seems to be overfitting on the training dataset. Let us check the performance on the test set.\n",
    "\n",
    "**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.78      0.48      0.59        61\n",
      "           1       0.78      0.94      0.85       124\n",
      "\n",
      "    accuracy                           0.78       185\n",
      "   macro avg       0.78      0.71      0.72       185\n",
      "weighted avg       0.78      0.78      0.77       185\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#predict on test data\n",
    "y_pred_test_knn = knn.predict(x_test_scaled)\n",
    "\n",
    "#checking performance of the model\n",
    "metrics_score(y_test, y_pred_test_knn)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observation:\n",
    "\n",
    "- Here the accuracy of the model is about 78% but the recall is only 0.48% for the class of interest.This will not help the bank in identifying the applicants at risk in an efficient manner.\n",
    "- We can see from the confusion matrix that the model fails to identify a large number of applicants who might default and lead to financial losses.\n",
    "- Examining the Logistic regression and KNN algorithms closely, the Logistic regression after adjusting the threshold seems to perform really well with a better f1-score and better recall.\n",
    "- Logistic Regression is able to minimize the Credit risk better than the KNN. \n",
    "\n",
    "**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "EWmSTtQLO7h1"
   },
   "source": [
    "## **Decision Tree**\n",
    "\n",
    "In this section, we will implement a decision tree classifier on the data.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {
    "id": "7KXjQaoQO7h4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(random_state=7)"
      ]
     },
     "execution_count": 129,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fitting the decision tree classifier on the training data\n",
    "d_tree = DecisionTreeClassifier(random_state=7) \n",
    "\n",
    "d_tree.fit(x_train,y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "rSEU_nFTO7h6"
   },
   "source": [
    "**Let's check the performance on the training data:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {
    "id": "wSStFV3LO7h-"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00       131\n",
      "           1       1.00      1.00      1.00       298\n",
      "\n",
      "    accuracy                           1.00       429\n",
      "   macro avg       1.00      1.00      1.00       429\n",
      "weighted avg       1.00      1.00      1.00       429\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Checking performance on the training data\n",
    "y_pred_train1 = d_tree.predict(x_train)\n",
    "metrics_score(y_train,y_pred_train1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "rQBoankb1AOS"
   },
   "source": [
    "**Observations:\n",
    "- The Decision Tree is giving us a 100% result on the training dataset.\n",
    "\n",
    "**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "JDoLC78mO7iD"
   },
   "source": [
    "**Let's check the performance of the model on the testing data:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {
    "id": "vDy9NKysO7iT"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.65      0.54      0.59        61\n",
      "           1       0.79      0.85      0.82       124\n",
      "\n",
      "    accuracy                           0.75       185\n",
      "   macro avg       0.72      0.70      0.71       185\n",
      "weighted avg       0.74      0.75      0.75       185\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Checking performance on the testing data\n",
    "y_pred_test1 = d_tree.predict(x_test)\n",
    "metrics_score(y_test,y_pred_test1 )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "RlT6eH8M1AOT"
   },
   "source": [
    "**Observations:\n",
    "\n",
    "- The Decision tree performs really well on the training set where as it does not perform well on the test set since the recall for class 0 on the test set is 0.54 as compared to 1 on the training set.   The metric scores indicate that the model overfits on the training data.\n",
    "- The precision of 0.65 estimates that the bank is able to correctly identify the non eligible customers only 65% of the time which is a huge loss in terms of financial losses since there will be large number of customers defaulting because this model will predict eligibility for those customers when they were actually not.  \n",
    "\n",
    "**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "JDLi2ITSO7iV"
   },
   "source": [
    "**Let's try hyperparameter tuning using GridSearchCV to find the optimal max_depth** in order to reduce overfitting of the model. We can tune some other hyperparameters as well."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "jVLhHs8WO7iW"
   },
   "source": [
    "## **Decision Tree - Hyperparameter Tuning**\n",
    "\n",
    "We will use the class_weight hyperparameter with value equal to {0:0.7, 1:0.3} which is approximately the opposite of the imbalance in the original data. \n",
    "\n",
    "**This would tell the model that 0 is the important class here.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {
    "id": "PVE4G3ncO7iX",
    "outputId": "0e9a4fc7-a289-42d4-a780-48f6d8162ebc"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(class_weight={0: 0.7, 1: 0.3}, criterion='entropy',\n",
       "                       max_depth=7, min_samples_leaf=20, random_state=7)"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Choose the type of classifier \n",
    "d_tree_tuned = DecisionTreeClassifier(random_state=7, class_weight={0:0.7, 1:0.3})\n",
    "\n",
    "# Grid of parameters to choose from\n",
    "parameters = {'max_depth': np.arange(2,10), \n",
    "              'criterion': ['gini', 'entropy'],\n",
    "              'min_samples_leaf': [5, 10, 20, 25]\n",
    "             }\n",
    "\n",
    "# Type of scoring used to compare parameter combinations - recall score for class 0\n",
    "scorer = metrics.make_scorer(recall_score, pos_label=0)\n",
    "\n",
    "# Run the grid search\n",
    "grid_obj = GridSearchCV(d_tree_tuned, parameters, scoring=scorer, cv=5)\n",
    "grid_obj = grid_obj.fit(x_train, y_train)\n",
    "\n",
    "# Set the classifier to the best combination of parameters\n",
    "d_tree_tuned = grid_obj.best_estimator_\n",
    "\n",
    "# Fit the best algorithm to the data \n",
    "d_tree_tuned.fit(x_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LsZNfwvgO7ic"
   },
   "source": [
    "We have tuned the model and fit the tuned model on the training data. Now, **let's check the model performance on the training and testing data.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {
    "id": "_4pznu8MO7if",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.57      0.79      0.66       131\n",
      "           1       0.89      0.74      0.81       298\n",
      "\n",
      "    accuracy                           0.75       429\n",
      "   macro avg       0.73      0.76      0.73       429\n",
      "weighted avg       0.79      0.75      0.76       429\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Checking performance on the training data\n",
    "y_pred_train2 = d_tree_tuned.predict(x_train)\n",
    "metrics_score(y_train,y_pred_train2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1PU_CuqNO7ih"
   },
   "source": [
    "**Let's check the model performance on the testing data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {
    "id": "T9c64oijO7ii"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.51      0.61      0.56        61\n",
      "           1       0.79      0.72      0.75       124\n",
      "\n",
      "    accuracy                           0.68       185\n",
      "   macro avg       0.65      0.66      0.65       185\n",
      "weighted avg       0.70      0.68      0.69       185\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Checking performance on the testing data\n",
    "y_pred_test2 = d_tree_tuned.predict(x_test)\n",
    "metrics_score(y_test,y_pred_test2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "FsFecGze1AOV"
   },
   "source": [
    "**Observations:**\n",
    "\n",
    "- After tuning in, there is a significant improvement in recall from our previous model performance on the testing set.The tuned model gives a better recall (0.61).\n",
    "- The precision and accuracy of the tuned model is lower than our base model.Only 51% of the times, the bank will be able to corectly classify the customers as non elligible which is not as efficient since this model will predict more number of false positives."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "QVN1JIBmO7iw"
   },
   "source": [
    "**Let's visualize the tuned decision tree** and observe the decision rules:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {
    "id": "m2DxT1hJO7iy",
    "outputId": "6aa3b5fa-292e-4d81-ddaf-1307a0dcf9a2"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "features = list(X.columns)\n",
    "\n",
    "plt.figure(figsize=(20,20))\n",
    "\n",
    "tree.plot_tree(d_tree_tuned,feature_names=features,filled=True,fontsize=9,node_ids=True,class_names=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "F6lb3jjRO7jS"
   },
   "source": [
    "**Let's look at the feature importance** of the tuned decision tree model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {
    "id": "0TaJ2BK5O7jT",
    "outputId": "d91c5833-bb2f-4f1a-e724-1b38210d12fa"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                              Imp\n",
      "Credit_History           0.574778\n",
      "total_income             0.206104\n",
      "LoanAmount               0.122203\n",
      "Married_Yes              0.054235\n",
      "Property_Area_Semiurban  0.042681\n",
      "Loan_Amount_Term         0.000000\n",
      "Gender_Male              0.000000\n",
      "Dependents_1             0.000000\n",
      "Dependents_2             0.000000\n",
      "Dependents_3+            0.000000\n",
      "Education_Not Graduate   0.000000\n",
      "Self_Employed_Yes        0.000000\n",
      "Property_Area_Urban      0.000000\n"
     ]
    }
   ],
   "source": [
    "# Importance of features in the tree building\n",
    "\n",
    "print (pd.DataFrame(d_tree_tuned.feature_importances_, columns = [\"Imp\"], index = x_train.columns).sort_values(by = 'Imp', ascending = False))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {
    "id": "uvCo_-YCO7jd",
    "outputId": "0d7133ca-3177-4f61-b309-ebe1b9bfcd6b"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plotting the feature importance\n",
    "importances = d_tree_tuned.feature_importances_\n",
    "indices = np.argsort(importances)\n",
    "\n",
    "plt.figure(figsize=(10,10))\n",
    "plt.title('Feature Importances')\n",
    "plt.barh(range(len(indices)), importances[indices], color='grey', align='center')\n",
    "plt.yticks(range(len(indices)), [features[i] for i in indices])\n",
    "plt.xlabel('Relative Importance')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "rG3nAwLzO7jo"
   },
   "source": [
    "**Observations:**\n",
    "- **Credit history is the most important feature** **followed by total income and loan amount** which makes sense.\n",
    "- **Married_Yes and Property_Area_Semiurban have some importance** and the rest of the variables have no impact in this model, while deciding loan eligibility.\n",
    "\n",
    "Now let's build another model - **a random forest classifier**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "OO2u6BuIO7js"
   },
   "source": [
    "## **Random Forest Classifier**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {
    "id": "9ZZBHm5wO7ju"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(random_state=7)"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fitting the decision tree classifier on the training data\n",
    "rf_estimator = RandomForestClassifier(random_state=7)\n",
    "rf_estimator.fit(x_train,y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "EqP5pHP9O7jv"
   },
   "source": [
    "**Let's check the performance of the model on the training data:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {
    "id": "DbN7ygMQO7jw"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00       131\n",
      "           1       1.00      1.00      1.00       298\n",
      "\n",
      "    accuracy                           1.00       429\n",
      "   macro avg       1.00      1.00      1.00       429\n",
      "weighted avg       1.00      1.00      1.00       429\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdAAAAE9CAYAAAC7hzNcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deZgdVbWw8XclhMEwQyCjDAYUnEAjF5SLDMqkkUENIHK5IgS5qAgqCPqJE8JVBgcUDSoighAFLoOoYEQDCjJEFAigQBCSNAlzQAMZen1/nEo4aXo6RVdOn+73x1PPqdq1q/Y+eZpevYfaFZmJJElqzJBmV0CSpFZkAJUkqQQDqCRJJRhAJUkqwQAqSVIJBlBJkkpYpdkV6MpVIw/y+Rq1vP2enN7sKkh9YsmiOVHVvRc//mDDv++Hbbh5ZfXpLVugkiSV0G9boJKkQaJ9abNrUIoBVJLUXNne7BqUYgCVJDVXuwFUkqSGpS1QSZJKsAUqSVIJtkAlSSrBWbiSJJVgC1SSpBIcA5UkqXHOwpUkqQxboJIklWALVJKkEpyFK0lSCbZAJUkqwTFQSZJKaNEWqC/UliSpBFugkqTmsgtXkqTGZToLV5KkxrXoGKgBVJLUXHbhSpJUgi1QSZJKcCUiSZJKsAUqSVIJjoFKklSCLVBJkkqwBSpJUgkGUEmSGudKRJIklWELVJKkEpxEJElSCbZAJUkqoUVboL5QW5KkEmyBSpKayy5cSZJKaNEuXAOoJKm5bIFKklRCiwZQJxFJkpor2xvfehAR4yLi+oi4JyLujohjivQvRMSciLij2Pauu+bEiLg/Iu6LiD16KqPSFmhE7AhskZnnRcQIYM3MnFVlmZKkFlNNC3QJ8MnMnBERawG3R8R1xbmzMvP0+swRsTVwIPBaYDTw24jYMrtZZ7CyFmhEnAycAJxYJA0DflpVeZKkFlVBCzQz2zJzRrH/LHAPMKabS/YBLs7MF4qG3v3Adt2VUWUX7n7Ae4B/AWTmXGCtCsuTJLWi9vbGtwZExKbAtsCfi6SPRsTfIuJHEbFekTYGeKTustl0H3ArDaCLMjOBBIiI4RWWJUlqVSVaoBExOSJuq9smd3briFgTuBT4RGYuAM4BXgVsA7QBZyzL2lnNuqt2lWOgUyPi+8C6EXEEcBhwboXlSZJaUYkx0MycAkzpLk9EDKMWPC/MzMuK6+bVnT8XuLo4nA2Mq7t8LDC3u/tXFkAz8/SIeCewAHg18PnMvK6HyyRJg00Fk4giIoAfAvdk5pl16aMys6043A+4q9i/ErgoIs6kNoloC+CW7sqodBZuETANmpKkrmW3PaVlvQ04BLgzIu4o0k4CDoqIbah1zz4EHFmrQt4dEVOBmdRm8B7d3QxcqCCARsSzdN5vHLU65tp9XaYkqYVV0ALNzBvpfFzzmm6uOQU4pbdl9HkAzUxn2kqSeq9FVyKqeiGFNwE7UmuR3piZf6myPElSC2rRxeSrXEjh88D5wAbAhsCPI+JzVZUnSWpRFT8HWpUqW6AHAdtm5vMAEXEaMAP4SoVlSpK0UlQZQB8CVgeeL45XAx6osDxJUiuqZhZu5aqYhfttamOeLwB3F4v3JvBO4Ma+Lk+S1OL6SZdso6pogd5WfN4OXF6X/vsKypIktToDaE1mnt/X95QkDWAtOgu3ii7cqZk5KSLupJMFFTLzDX1dpiSpdWW7Y6DLHFN8vruCe0uSBhq7cGuWLdKbmf/s63tLkgYgu3BX1MWauM9Qm2T0ycx8sKqyJUktxC7clziT2rvULqK2oO+BwEjgPuBHwM4Vli1JahUt2oVb2VJ+wJ6Z+f3MfDYzFxQvP907My8B1quwXElSK2nRpfyqDKDtETEpIoYU26S6c63ZXu9n3njWkex+1/d4+++/tjzt1ce/n7f/7n/Z6bensv3FJ7LaxrW/VYattyY7XPo59nrgPF731f/u8p7D1h3O9pecxC5/OpPtLzmJYesMX35u/Mf2YdebzmKXG89gxM5Oplb19th9Z+6+azr3zryR4z99dKd5zjrzS9w780Zm3H4d227zuoauVT+R2fjWD1QZQA+m9jLT+cC8Yv+DEbEG8NEKyx00HrnkD/z5oNNWSHvgu1fzh11PYPo7TmTedTPY8rj9AWh/YTH3/e/PmfnFC7u95/iP7cPjN9zF9W89jsdvuIvxH3sPAGtuOYbR++7A79/+aW7+wGm8/rTDYEhnr9qT+saQIUP41jdP4d0TP8jr37gLBxywL1tttcUKefbac1e2GL8Zr9l6R4466gS+c/apvb5W/Ygt0BVl5oOZOTEzN8zMEcX+/Zm5sHjRqV6mJ2++l0VPP7dC2pLnFi7fH/qK1VnW2F/67xd48pb7WPrCom7vOXKPN/PI1OkAPDJ1OiP3nFCkT2Du/91E+6IlLHz4Mf4161HW23Z8H34baUXbvWVbHnjgIWbNepjFixczdeoVvGfiHivkmThxDy648BcA/PmWGayz7jqMHLlRr65VP9KejW/9QBULKRyfmV+rWxN3BZn58b4uUyt6zWcmMfb9O7H42X9z03u/3NC1q41YhxfmPw3AC/OfZtUN1wZg9VHr8dTt9y/P93zbk6w+yqFsVWf0mJE8Mnvu8uPZc9rY7i3brpBnzOiRzH7kxTxzZrcxZvTIXl2rfqRFH2OpogV6T/F5G7X1cDtuqti9p03lt2/+KHMu/SObHtZHf3VHJ921/eOPQA1Q0cnPXHYY++oqT2+uVT9iC7QmM68qPhteEzciJgOTAf5nrQns+Qq7CF+OOZf/ke1+ejx///oven3NC489w2obrcsL859mtY3WZdHjCwB4fu6TrDF6g+X5Vh+1Ps8/+lSf11laZs7sNsaNHb38eOyYUbS1zVshz+w5bYwd92KeMWNHMbdtHquuumqP16r/yH4yptmoPm+BRsRVEXFlV1t312bmlMyckJkTDJ7lDN9s5PL9jfd4M8/dP7eb3C/16LW3M27STgCMm7QTj/7m9uXpo/fdgSGrrsIarxzB8M1H8tRf7u/uVtLLcuttdzB+/GZsuuk4hg0bxqRJ+3DV1deukOfqq6/lkIPfB8B/bPcmFjyzgEcfnd+ra6WXq4qFFE6v4J7qxJvO+RgbvHUrVl1/Ld4x42zu+/ov2Hi3bRg+fjS0J/+e/Rh3Hv/D5fl3u/VbrLLmGgxZdRVG7jmBmw88lef+Poc3nHEE//zJNJ7564Pc/+0refOUYxj3gZ1ZOOcJbj/iGwA8d99s2q68mZ2nn04uWcpdJ57Xb7pRNDAtXbqUYz7xOa755UUMHTKEH59/CTNn/p3JRxwCwJRzL+CaX01jzz135b57/si/Fy7k8MOP6/Za9VMt+rsk+npcICLWzswFXZx7ZWY+3Jv7XDXyoNb8F5Xq7Pfk9GZXQeoTSxbNqey5tX995YMN/74f/rmfNv05uiomEf1+2U5ETOtw7v8qKE+S1MqcRLRc/V8F63dzTpKkfrMwQqOqCKDZxX5nx5Kkwa6ftCgbVUUA3SgijqPW2ly2T3E8ooLyJEmtrEUXUqgigJ4LrNXJPsAPKihPktTKbIHWZOYX+/qekqSBq1UXUqjyhdqSJPXMFqgkSSW0aACt7HVmEbFZb9IkSYNctje+9QNVvlD70k7Ser+quSRpcHAhhZqIeA3wWmCdiNi/7tTawOp9XZ4kqbVlPwmIjapiDPTVwLuBdYGJdenPAkdUUJ4kqZUZQGsy8wrgiojYITNv6uv7S5IGmBZ9jKXKMdBHIuLyiJgfEfMi4tKIGFtheZKkVtSiY6BVBtDzgCuB0cAY4KoiTZKkF1UQQCNiXERcHxH3RMTdEXFMkb5+RFwXEf8oPteru+bEiLg/Iu6LiD16KqPKALpRZp6XmUuK7ce4Fq4kaeVYAnwyM7cCtgeOjoitgc8A0zJzC2BacUxx7kBqk2D3BL4bEUO7K6DKAPpYRHwwIoYW2weBJyosT5LUgjKz4a0X92zLzBnF/rPAPdR6Q/cBzi+ynQ/sW+zvA1ycmS9k5izgfmC77sqoMoAeBkwCHgXagPcVaZIkvahEF25ETI6I2+q2yV3dPiI2BbYF/gxsnJltUAuywEZFtjHAI3WXzS7SulTZUn6Z+TDwnqruL0kaIEpMCsrMKcCUnvJFxJrUFvb5RGYuiIgus3ZWTHf3rmIhhc93czoz88t9XaYkqXVVtZBCRAyjFjwvzMzLiuR5ETEqM9siYhQwv0ifDYyru3wsMLe7+1fRhfuvTjaADwMnVFCeJKmVVTMLN4AfAvdk5pl1p64EDi32DwWuqEs/MCJWK9Zt3wK4pbsyqlhI4Yxl+xGxFnAM8CHgYuCMrq6TJA1S1ayj8DbgEODOiLijSDsJOA2YGhEfBh4G3g+QmXdHxFRgJrUZvEdn5tLuCqhkDDQi1geOAw6mNsvpTZn5VBVlSZJaWxVduJl5I52PawLs1sU1pwCn9LaMKsZAvw7sT21w9/WZ+VxflyFJGkD6ycpCjapiDPST1FYf+hwwNyIWFNuzEbGggvIkSa2svcTWD1QxBlrls6WSpAHG15lJklRGP2lRNsoAKklqKlugkiSVYQtUkqTGpQFUkqQSDKCSJDWuVVugPnIiSVIJtkAlSc3Voi1QA6gkqalatQvXACpJaioDqCRJJRhAJUkqI7t661j/ZgCVJDWVLVBJkkrIdlugkiQ1zBaoJEklpGOgkiQ1zhaoJEklOAYqSVIJ2Zrv0zaASpKayxaoJEklGEAlSSrBLlxJkkpo1RaoL9SWJKkEW6CSpKYacAspRMS3gS57pjPz45XUSJI0qAzEhRRuW2m1kCQNWu0DrQWameevzIpIkganAdeFu0xEjABOALYGVl+Wnpm7VlgvSdIgMZBn4V4I3ANsBnwReAi4tcI6SZIGkczGt/6gNwF0g8z8IbA4M/+QmYcB21dcL0nSIJHt0fDWH/TmMZbFxWdbRLwLmAuMra5KkqTBZMBNIqrzlYhYB/gk8G1gbeDYSmslSRo0WnUSUY9duJl5dWY+k5l3ZeYumfnmzLxyZVROkjTwVTEGGhE/ioj5EXFXXdoXImJORNxRbHvXnTsxIu6PiPsiYo/e1Ls3s3DPo5MFFYqxUEmSXpaKunB/DJwN/KRD+lmZeXp9QkRsDRwIvBYYDfw2IrbMzKXdFdCbLtyr6/ZXB/ajNg4qSdLLVkUXbmZOj4hNe5l9H+DizHwBmBUR9wPbATd1d1GPATQzL60/joifAb/tZaUkSerWSn4s5aMR8V/UVtv7ZGY+BYwBbq7LM7tI61aZxeS3AF5Z4rqG7Pfk9KqLkCq3cO4Nza6C1O+V6cKNiMnA5LqkKZk5pYfLzgG+TG1Y8svAGcBhQGcV6DGs92YM9NkON3qU2spEkiS9bGW6cItg2VPA7HjNvGX7EXEuLw5RzgbG1WUdSy+GKnvThbtWIxWUJKkRK+s50IgYlZltxeF+wLIZulcCF0XEmdQmEW0B3NLT/XrTAp2Wmbv1lCZJUn9RzNfZGdgwImYDJwM7R8Q21HpVHwKOBMjMuyNiKjATWAIc3dMMXOj+faCrA68oCl+PF/uI16YWoSVJetmqmEOUmQd1kvzDbvKfApzSSBndtUCPBD5BLVjezosBdAHwnUYKkSSpKwNuKb/M/CbwzYj4WGZ+eyXWSZI0iAzYpfyA9ohYd9lBRKwXEf9TYZ0kSYNIe4mtP+hNAD0iM59edlA8dHpEdVWSJA0mSTS89Qe9WUhhSEREZm2tiIgYCqxabbUkSYNFez95QXajehNAfwNMjYjvUZss9RHgV5XWSpI0aLT3kxZlo3oTQE+gtlzSUdRm4v4FGFVlpSRJg0d/6ZJtVG/eB9pObZHdB4EJwG7APRXXS5I0SLTqJKLuFlLYktr70Q4CngAuAcjMXVZO1SRJg0GrtkC768K9F7gBmJiZ9wNExLErpVaSpEGjv7QoG9VdF+57qb155fqIODcidqPzV75IklRaq3bhdhlAM/PyzDwAeA3we+BYYOOIOCcidl9J9ZMkDXCt+hxobyYR/SszL8zMd1N7R9odwGcqr5kkaVBoj8a3/qA3j7Esl5lPAt8vNkmSXraB/ByoJEmVadGFiHq1Fq4kSerAFqgkqan6y6zaRhlAJUlN1R6OgUqS1LBWHQM1gEqSmsouXEmSSugvz3U2ygAqSWoqnwOVJKkEx0AlSSrBLlxJkkpwEpEkSSXYhStJUgl24UqSVIJduJIklWAAlSSphLQLV5KkxtkClSSpBAOoJEkltOpjLEOaXQFJklqRLVBJUlP5HKgkSSU4BipJUgmtGkAdA5UkNVWW2HoSET+KiPkRcVdd2voRcV1E/KP4XK/u3IkRcX9E3BcRe/Sm3gZQSVJTtUfjWy/8GNizQ9pngGmZuQUwrTgmIrYGDgReW1zz3YgY2lMBBlBJUlO1l9h6kpnTgSc7JO8DnF/snw/sW5d+cWa+kJmzgPuB7XoqwwAqSWqqKrpwu7BxZrYBFJ8bFeljgEfq8s0u0rplAJUkNVU72fAWEZMj4ra6bfLLqEJnncI9xmln4UqSmqrMLNzMnAJMafCyeRExKjPbImIUML9Inw2Mq8s3Fpjb081sgUqSmmolduFeCRxa7B8KXFGXfmBErBYRmwFbALf0dDNboJKkpqriOdCI+BmwM7BhRMwGTgZOA6ZGxIeBh4H3A2Tm3RExFZgJLAGOzsylPZVhAJUkNVUVS/ll5kFdnNqti/ynAKc0UoYBVJLUVO0t+j4WA6gkqalaM3xWPIkoIraMiGnLllKKiDdExOeqLFOS1FqqWEhhZah6Fu65wInAYoDM/Bu15ZIkSQLKPQfaH1QdQF+RmR2nAi+puExJkipX9Rjo4xHxKoou7oh4H9BWcZmSpBbSP9qTjas6gB5NbaWI10TEHGAW8MGKy5QktZD+MqbZqEoDaGY+CLwjIoYDQzLz2SrLkyS1nv4yptmoSgJoRBzXRToAmXlmFeVKklpPa4bP6lqga1V0X0nSAGMXbp3M/GIV95UkDTzZom3QqhdS2DwiroqIxyJifkRcERGbV1mmJKm1uJBC5y4CpgKjgNHAz4GfVVymJKmFuJBC5yIzL8jMJcX2U1p3vLil7LH7ztx913TunXkjx3/66E7znHXml7h35o3MuP06tt3mdQ1dK/WVtnmP8aGPnsDED0xmn4OP5IKp/wfAvf94kIMnH8t+hxzF0cefzHP/+hcAi5cs4aQvn85+hxzFxA9M5tyfXNLpfZ9Z8CyHH3MSex/wYQ4/5iSeWfDiQwDn/uQS9pp0GO8+8HD++Ofbq/+S6tZKfB9on6okgEbE+hGxPnB9RHwmIjaNiE0i4njgl1WUqRcNGTKEb33zFN498YO8/o27cMAB+7LVVluskGevPXdli/Gb8Zqtd+Soo07gO2ef2utrpb60ytChfPpjR3DVRVO4aMpZXHzZ1Tww65+cfNo3+MRRH+LyC85ht53eynkXXgrAtb+7gUWLF3P5Becw9Uff4udXXMOctnkvue8PLpjK9hO24ZpLfsj2E7bhhz+dCsADs/7Jr6b9gSt++j2+d+ZX+PLpZ7N0aY+vflSFbIGu6HbgNuAA4EjgeuD3wFHAhyoqU4Xt3rItDzzwELNmPczixYuZOvUK3jNxjxXyTJy4Bxdc+AsA/nzLDNZZdx1GjtyoV9dKfWnEhuuz9avHAzB8+CvYfJNxzHvsCR56eDYTtnk9ADu85U1c94cbgdrjcAuff54lS5bywguLGDZsGGsOf8VL7nv9DTexz17vAGCfvd7B76bfBMDvbriZvXZ7O6uuuipjR4/klWNHc+c9f18ZX1VdcAy0TmZulpmbF58dNycRVWz0mJE8Mnvu8uPZc9oYPXrkCnnGjB7J7EdezDNndhtjRo/s1bVSVea0zeOefzzAG177asZvvinX33gzANdefwOPznscgHfusiNrrL46u+zzAd65/3/x3wftzzprv/TJuSeeepoRG64P1IL0k08/A8D8x55g5MYjlufbeKMNmf/Y41V/NXUjS/zXH1S1kMKumfm7iNi/s/OZeVkV5apm2YIV9TKzV3l6c61UhX//eyHHfvYrnPDxI1lz+HC+fNKxnHrWOXzvvIvYecftGTas9uvqzpn3MXTIEH53xYUsePY5Dj3qU2w/YVvGjRnVq3I6++UbvPTnXitPf2lRNqqqhRTeDvwOmNjJuQQ6DaARMRmYDBBD12HIkOEVVW9gmzO7jXFjRy8/HjtmFG0dxohmz2lj7LgX84wZO4q5bfNYddVVe7xW6muLlyzhE5/9Cu/afRfeufPbANh8k3Gc+42vAvDQw7OZ/qfai52uue73vG37CQxbZRU2WG9dtnnD1tx97z9eEkA3WG9dHnv8SUZsuD6PPf4k66+7DgAbj9iQR+c9tjzfvPmPM2LEBivja6oL/aVF2aiqunBPLj4/1Ml2WDfXTcnMCZk5weBZ3q233cH48Zux6abjGDZsGJMm7cNVV1+7Qp6rr76WQw5+HwD/sd2bWPDMAh59dH6vrpX6Umby+VO/weabjOPQA1/stHriqacBaG9v5/vnX8ykffcGYNTGI7jl9r+Smfx74fP87e572WyTcS+57847bs8Vv/otAFf86rfs8p87ALDLjtvzq2l/YNGiRcye+ygPz57L67fasuqvqW606hhopYvJd7Em7jPA7Zl5R5VlD2ZLly7lmE98jmt+eRFDhwzhx+dfwsyZf2fyEYcAMOXcC7jmV9PYc89due+eP/LvhQs5/PDjur1Wqspf/nY3V/16Glu8alPee2jtsaljjjyUf86ey8WXXQ3AO97+VvZ71+4AHLT/RD731TPZ94MfIUn23Xt3Xj1+MwA+f+o3mLTv3rxuqy05/JBJfPL/fZXLrv4NozYewZlf+SwA4zffhD12/U/ec/CRrDJ0KJ897n8YOnRoE765lmlv0WGiqHJ8KyIuAiYAVxVJ7wJuBV4D/Dwzv9bVtausOqY1/0WlOgvn3tDsKkh9YtiGm1c2UHzIJvs3/Pv+gn9e1vSB66rfB7oB8KbMfA4gIk4GfgHsRO1Rly4DqCRpcGjV1lLVAfSVwKK648XAJpm5MCJeqLhsSVIL6C8LIzSq6gB6EXBzRFxRHE8Efla8YHtmxWVLklpAq87CrTSAZuaXI+IaYEcggI9k5m3F6YOrLFuS1Br6y6zaRlW1kMLambmgWA93VrEtO7d+Zj5ZRbmSpNZjF+6KLgLeTW2iUP2/TBTHLucnSQLswl1BZr67+NysivtLkgYOu3DrRMSbujufmTOqKFeS1Hpadb3tqrpwz+jmXAK7VlSuJKnFOAZaJzN3qeK+kqSBp1W7cCtZTD4ijq/bf3+Hc1+tokxJUmtq1feBVhJAgQPr9k/scG7PisqUJLWgdrLhrT+oagw0utjv7FiSNIg5iWhF2cV+Z8eSpEGsVcdAqwqgb4yIBdRam2sU+xTHq1dUpiSpBVU1phkRDwHPAkuBJZk5oVgh7xJgU+AhYFJmPlXm/pWMgWbm0MxcOzPXysxViv1lx8OqKFOS1JoqHgPdJTO3ycwJxfFngGmZuQUwrTgupapJRJIk9Uf7AOcX++cD+5a9kQFUktRUmdnw1ttbA9dGxO0RMblI2zgz24py24CNyta76veBSpLUrTKPpRQBcXJd0pTMnNIh29syc25EbARcFxH3voxqvoQBVJLUVGUmERXBsmPA7JhnbvE5PyIuB7YD5kXEqMxsi4hRwPwSVQbswpUkNVl7ZsNbTyJieESstWwf2B24C7gSOLTIdihwRdl62wKVJDVVRYsDbAxcHhFQi3UXZeavI+JWYGpEfBh4GHh/N/folgFUktRUVSzNl5kPAm/sJP0JYLe+KMMAKklqqv6ytm2jDKCSpKZyLVxJkkqwBSpJUgn95f2ejTKASpKayi5cSZJKsAtXkqQSbIFKklSCLVBJkkpwEpEkSSX0Zm3b/sjF5CVJKsEWqCSpqezClSSphFbtwjWASpKayhaoJEkl2AKVJKkEW6CSJJVgC1SSpBJsgUqSVEJme7OrUIoBVJLUVK6FK0lSCb6NRZKkEmyBSpJUgi1QSZJK8DEWSZJK8DEWSZJKsAtXkqQSnEQkSVIJrdoCHdLsCkiS1IpsgUqSmspZuJIkldCqXbgGUElSUzmJSJKkEmyBSpJUgmOgkiSV4EpEkiSVYAtUkqQSWnUM1IUUJElNlSX+642I2DMi7ouI+yPiM31db1ugkqSmqqIFGhFDge8A7wRmA7dGxJWZObOvyrAFKklqqsxseOuF7YD7M/PBzFwEXAzs05f1NoBKkpoqS2y9MAZ4pO54dpHWZ/ptF+6SRXOi2XUY6CJicmZOaXY9pJfLn+XWVub3fURMBibXJU3p8DPQ2T37tK/YFujgNrnnLFJL8Gd5kMnMKZk5oW7r+AfUbGBc3fFYYG5f1sEAKkkaiG4FtoiIzSJiVeBA4Mq+LKDfduFKklRWZi6JiI8CvwGGAj/KzLv7sgwD6ODmmJEGCn+W9RKZeQ1wTVX3j1ZdAUKSpGZyDFSSpBIMoP1MRGREnFF3/KmI+EIP1+wbEVt3ce4LETEnIu6o29aNiJ0j4uoiz3t6WuYqIkZHxC+K/f+OiLO7yPdcD19RWkFELO3w8/mZIv33ETGh2L8mItbt4T5fioh3FPsPRcSGneT5QkR8qorvocHHMdD+5wVg/4g4NTMf7+U1+wJXA10tUXVWZp5enxDx4iNSmXklPcxOy8y5wPt6WR+pEQszc5vuMmTm3j3dJDM/33dVknpmC7T/WUJtQsSxHU9ExCYRMS0i/lZ8vjIi3gq8B/h68df7qxotsL5FGRGvioibI+LW4i/654r0TSPirrrLxkXEr4uFmk/u4r6fLu7zt4j4YqP1kpapb1FGxP+LiHsj4rqI+NmyFmVE/Dgi6v/I+3RE3FJs4zu556uKn+HbI+KGiHjNSvo6GiAMoP3Td4CDI2KdDulnAz/JzDcAFwLfysw/UWs9fjozt8nMBzq537F13WPX91D2N4FvZuZb6P6h4+2Ag4FtgPcv62pbJiJ2B7Yo8m0DvDkiduqhbA1Oa3Towj2gq4zFz9l7gW2B/YEJXeUFFmTmdtT+v/lGJ+enAB/LzDcDnwK+W1u2w9wAAAQ9SURBVPobaFCyC7cfyswFEfET4OPAwrpTO1D7pQFwAfC1Xt7yJV243diBWpcwwEVAV9ddl5lPAETEZcCOwG1153cvtr8Ux2tSC6jTe1kPDR49duHW2RG4IjMXAkTEVd3k/Vnd51n1JyJiTeCtwM/rhjNW63WNJQyg/dk3gBnAed3kaeYzSB3L7ngcwKmZ+f2VVB8NDo2smZpd7EOt9+3pBgK39BJ24fZTmfkkMBX4cF3yn6gtRwW17tMbi/1ngbX6qOibqXWRUVdWZ94ZEetHxBrUWqx/7HD+N8BhxV/6RMSYiNioj+qowetGYGJErF78bL2rm7wH1H3eVH8iMxcAsyLi/QBR88YqKqyBywDav50B1E/F/zjwoYj4G3AIcEyRfjG1CRN/6WIS0bEdxpg27abMTwDHRcQtwCjgmS7y3UitG/kO4NLMrO++JTOvpdYFfFNE3An8gr4L8hpYOo6BntZVxsy8ldqY/1+By6gNG3T1M7paRPyZ2v8nL5mUR+2P0A9HxF+Bu+njd0Vq4HMlIq0gIl5BbUwqI+JA4KDM9BeL+o2IWDMznyt+VqcDkzNzRrPrpcHHMVB19Gbg7KjNrHgaOKzJ9ZE6mlIsHLI6cL7BU81iC1SSpBIcA5UkqQQDqCRJJRhAJUkqwQAqscIbQe6KiJ8XMzzL3mv5mqwR8YPo4k05xfmdi/WMGy2j07eNSFp5DKBSzcJiLeHXAYuAj9SfjIihZW6amYdnZldvyQHYmdqScpJajAFUeqkbgPFF6/D6iLgIuDMihkbE1+veMHMkLF/F5uyImBkRvwSWr7gUK77Tcs+ImBERfy3eprMptUC9bKGL/4yIERFxaVHGrRHxtuLaDSLi2mKxjO/T2JJ2kirgc6BSnYhYBdgL+HWRtB3wusycFRGTgWcy8y0RsRrwx4i4ltqbQV4NvB7YmNp7WX/U4b4jgHOBnYp7rZ+ZT0bE94Dnli32XwTrszLzxoh4JbUlEbcCTgZuzMwvRcS7gMmV/kNI6pEBVKpZIyLuKPZvAH5IrWv1lsycVaTvDryh7p2T61B7w8xOwM8ycykwNyJ+18n9twemL7tXsdZxZ94BbF33hpC1I2Ktooz9i2t/GRFPlfyekvqIAVSqeckrtYog9q/6JGrvj/xNh3x70/ObcaIXeaA2rLLDstd1daiLq55I/YhjoFLv/QY4KiKGAUTElhExnNp6rAcWY6SjgF06ufYm4O0RsVlx7fpFesc36VwLfHTZQUQsC+rTqS1+TkTsBazXZ99KUikGUKn3fkBtfHNGRNwFfJ9aL87lwD+AO4FzgD90vDAzH6M2bnlZ8faPS4pTVwH7LZtERO2NOxOKSUozeXE28BeBnSJiBrWu5Icr+o6Sesm1cCVJKsEWqCRJJRhAJUkqwQAqSVIJBlBJkkowgEqSVIIBVJKkEgygkiSVYACVJKmE/w/UpQGwEfvz7gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Checking performance on the training data\n",
    "y_pred_train3 = rf_estimator.predict(x_train)\n",
    "metrics_score(y_train, y_pred_train3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "-qNhC43Y1AOY"
   },
   "source": [
    "**Observations:\n",
    "\n",
    "-  The Random Forest is giving a 100% result on the training dataset which means the model overfits the training set.\n",
    "\n",
    "**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Qq34aT4aO7jy"
   },
   "source": [
    "**Let's confirm this by checking its performance on the testing data:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "id": "7eoRv4jcO7j1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.81      0.49      0.61        61\n",
      "           1       0.79      0.94      0.86       124\n",
      "\n",
      "    accuracy                           0.79       185\n",
      "   macro avg       0.80      0.72      0.74       185\n",
      "weighted avg       0.80      0.79      0.78       185\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Checking performance on the testing data\n",
    "y_pred_test3 = rf_estimator.predict(X_test)\n",
    "metrics_score(y_test, y_pred_test3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ZtvgL8yS1AOZ"
   },
   "source": [
    "**Observations:\n",
    "\n",
    "- The recall for RF classifier on the testing dataset is only 0.49 which is not very efficient, hence this model does not perform well on the testing set where as the model overfits the training set.\n",
    "- Precision on the testing set using Random Forest is very good.\n",
    "\n",
    "**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "pDVdnqzjO7j2"
   },
   "source": [
    " **Let's see if we can get a better model by tuning the random forest classifier:**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "9iADMfV-O7j3"
   },
   "source": [
    "## **Random Forest Classifier - Hyperparameter Tuning**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "w41zX02GO7j4"
   },
   "source": [
    "Let's try **tuning some of the important hyperparameters of the Random Forest Classifier**. \n",
    "\n",
    "We will **not** tune the `criterion` hyperparameter as we know from hyperparameter tuning for decision trees that `entropy` is a better splitting criterion for this data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {
    "id": "-3uYE75-O7j4"
   },
   "outputs": [],
   "source": [
    "# Choose the type of classifier\n",
    "rf_estimator_tuned = RandomForestClassifier(criterion=\"entropy\", random_state=7)\n",
    "\n",
    "# Grid of parameters to choose from\n",
    "parameters = {\"n_estimators\": [100, 110, 120],\n",
    "    \"max_depth\": [5, 6, 7],\n",
    "    \"max_features\": [0.8, 0.9, 1]\n",
    "             }\n",
    "\n",
    "# Type of scoring used to compare parameter combinations - recall score for class 0\n",
    "scorer = metrics.make_scorer(recall_score, pos_label=0)\n",
    "\n",
    "# Run the grid search\n",
    "grid_obj = GridSearchCV(rf_estimator_tuned, parameters, scoring=scorer, cv=5)\n",
    "grid_obj = grid_obj.fit(x_train, y_train)\n",
    "\n",
    "# Set the classifier to the best combination of parameters\n",
    "rf_estimator_tuned = grid_obj.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {
    "id": "o6DLcoGzO7j5",
    "outputId": "67c08c0f-f913-42b5-92f9-d38484c57030"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(criterion='entropy', max_depth=5, max_features=0.9,\n",
       "                       n_estimators=110, random_state=7)"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fitting the best algorithm to the training data\n",
    "rf_estimator_tuned.fit(x_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {
    "id": "OVXmmXdVO7j7",
    "outputId": "4fe6efba-e829-4c7a-8dcb-23ea829d0c3e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.98      0.48      0.65       131\n",
      "           1       0.81      1.00      0.90       298\n",
      "\n",
      "    accuracy                           0.84       429\n",
      "   macro avg       0.90      0.74      0.77       429\n",
      "weighted avg       0.87      0.84      0.82       429\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Checking performance on the training data\n",
    "y_pred_train4 = rf_estimator_tuned.predict(x_train)\n",
    "metrics_score(y_train, y_pred_train4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "tZ7AEkUmO7j8"
   },
   "source": [
    "**Observations:**\n",
    "- We can see that after hyperparameter tuning, the model is performing poorly on the train data as well.\n",
    "- We can try adding some other hyperparameters and/or changing values of some hyperparameters to tune the model and see if we can get a better performance.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {
    "id": "9XUQucgwO7kG"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(class_weight={0: 0.4, 1: 0.1}, criterion='entropy',\n",
       "                       max_depth=6, max_features=0.8, max_samples=0.9,\n",
       "                       min_samples_leaf=25, n_estimators=120, random_state=7)"
      ]
     },
     "execution_count": 145,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Choose the type of classifier\n",
    "rf_estimator_tuned = RandomForestClassifier(criterion=\"entropy\", random_state=7)\n",
    "\n",
    "# Grid of parameters to choose from\n",
    "parameters = {\"n_estimators\": [110, 120],\n",
    "    \"max_depth\": [6, 7],\n",
    "    \"min_samples_leaf\": [20, 25],\n",
    "    \"max_features\": [0.8, 0.9],\n",
    "    \"max_samples\": [0.9, 1],\n",
    "    \"class_weight\": [{0:0.7, 1:0.3}, \"balanced\", {0:0.4, 1:0.1}]\n",
    "             }\n",
    "\n",
    "# Type of scoring used to compare parameter combinations - recall score for class 0\n",
    "scorer = metrics.make_scorer(recall_score, pos_label=0)\n",
    "\n",
    "# Run the grid search on the training data using scorer=scorer and cv=5\n",
    "grid_obj = GridSearchCV(rf_estimator_tuned, parameters, scoring=scorer, cv=5)\n",
    "grid_obj = grid_obj.fit(x_train, y_train)\n",
    "# Save the best estimator to variable rf_estimator_tuned\n",
    "rf_estimator_tuned = grid_obj.best_estimator_\n",
    "\n",
    "#Fit the best estimator to the training data\n",
    "rf_estimator_tuned.fit(x_train,y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "9Id5gfK2O7kK"
   },
   "source": [
    "**Let's check the performance of the tuned model:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {
    "id": "-KmNFiv9O7kL"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.52      0.85      0.65       131\n",
      "           1       0.91      0.65      0.76       298\n",
      "\n",
      "    accuracy                           0.71       429\n",
      "   macro avg       0.71      0.75      0.70       429\n",
      "weighted avg       0.79      0.71      0.72       429\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Checking performance on the training data\n",
    "y_pred_train5 = rf_estimator_tuned.predict(x_train)\n",
    "metrics_score(y_train, y_pred_train5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_sL9NN0y1AOb"
   },
   "source": [
    "**Observations:\n",
    "\n",
    "- The recall on the training data after tuning the model is quite good(0.85). \n",
    "- The precision is only 0.52 on the training data.\n",
    "\n",
    "**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Nwv3LM3cO7kP"
   },
   "source": [
    "**Let's check the model performance on the test data:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {
    "id": "3Loo6OLKO7kY"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.50      0.70      0.59        61\n",
      "           1       0.82      0.65      0.73       124\n",
      "\n",
      "    accuracy                           0.67       185\n",
      "   macro avg       0.66      0.68      0.66       185\n",
      "weighted avg       0.71      0.67      0.68       185\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Checking performance on the testing data\n",
    "y_pred_test5 =rf_estimator_tuned.predict(x_test)\n",
    "metrics_score(y_test, y_pred_test5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "pYhZ3-ioO7ka"
   },
   "source": [
    "**Observations:\n",
    "\n",
    "- The tuned RF model performs well on the testing data.\n",
    "- Although, the precision is quite low, here we are more interested in minimizing the False negatives i.e maximizing the recall. The recall significantly improved from 0.49 to 0.70 on the testing data.\n",
    "- There is definitely a trade-off between precision and recall here which the bank might need to play close attention to if any predictions are made based on this model.\n",
    "\n",
    "**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "DCg7SaRTO7kb"
   },
   "source": [
    "**One of the drawbacks of ensemble models is that we lose the ability to obtain an interpretation of the model. We cannot observe the decision rules for random forests the way we did for decision trees. So, let's just check the feature importances of the model.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {
    "id": "i_E_LpzLO7kl",
    "outputId": "8dabfa51-ea81-4da8-8cb9-b75fce7d8ae3"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "importances = rf_estimator_tuned.feature_importances_\n",
    "indices = np.argsort(importances)\n",
    "feature_names = list(X.columns)\n",
    "\n",
    "plt.figure(figsize=(12,12))\n",
    "plt.title('Feature Importances')\n",
    "plt.barh(range(len(indices)), importances[indices], color='grey', align='center')\n",
    "plt.yticks(range(len(indices)), [feature_names[i] for i in indices])\n",
    "plt.xlabel('Relative Importance')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "D-4c6HACO7ko"
   },
   "source": [
    "**Observations:\n",
    "\n",
    "- The feature importance plot for RF Classifier gives a better overview of the features that are more influential on the loan eligibility. \n",
    "- The most important features that drives the eligibility decision are Credit History followed by total income, Loan amount, Property Area-Semiurban and Married-yes.\n",
    "- Other features like Education-Not Graduate, Property Area-Urban, Dependent-2 ,Gender-Male and Dependent-1 have very little impact on the decision.\n",
    "\n",
    "**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "dZ-0Sv5TO7kp"
   },
   "source": [
    "## **Conclusion and Recommendations**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vo8q0lEHO7kr"
   },
   "source": [
    "**Conclusions:\n",
    "\n",
    "- We have seen that **tuned Random Forest model does better in terms of recall on training and testing datasets.** The recall on training is ~0.85 and on testing is ~0.70 which is much better than the Decision tree and   base RF model..\n",
    "- Although the precision is quite low in both training(~ 0.52) and testing sets(~ 0.50), the bank needs to make a decision if they are willing to slip of the False positives inorder to maximize the recall.\n",
    "- Throughout our analysis, we have seen that **Credit history,total amount, loan amount, Property Area_Semiurban and Married-Yes are the most influential attributes in loan eligibility decision.** \n",
    "\n",
    "**\n",
    "\n",
    "**Recommendations:\n",
    "\n",
    "- Since **Credit History is the most important feature of Loan eligibility,** the bank should focus on the customer base with good credit history inorder to minimize the default loans.\n",
    "- We also saw that total income, loan amount,Property area-Semiurban and Married-Yes are also some of the very important attributes in the loan approval process which does make sense since customers who are married,   have higher income and own properties in Semi urban areas mostly come with a good credit history. This is the type of customer base that the bank should focus on in-oder to have most loan amount payed on a timely basis.\n",
    "- The bank should offer some incentives for customers falling into this category which could lead the bank to have a strong relationship and overall good customer base.The small incentives and good customer service provided to this group might help the bank make more business with great returns and minimize credit risk.\n",
    "\n",
    "**\n"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Learner_Notebook_Project_Classification.ipynb",
   "provenance": []
  },
  "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}