Project_ Classification_ Loan Eligibility Prediction - Learner_s Notebook-2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "OC4AgbDge2Co"
   },
   "source": [
    "# Project: Classification - Loan Eligibility Prediction\n",
    "**Marks: 30**\n",
    "\n",
    "Dear Learner,\n",
    "\n",
    "Welcome to project on Classification. We will use the credit risk data for the exercise.\n",
    "Do read the problem statement and the guidelines around the same.\n",
    "\n",
    "----\n",
    "## Context: \n",
    "-------\n",
    "\n",
    "Credit risk is nothing but the default in payment of any loan by the borrower. In Banking sector this is an important factor to be considered before approving the loan of an applicant.Dream Housing Finance company deals in all home loans. They have presence across all urban, semi urban and rural areas. Customer first apply for home loan after that company validates the customer eligibility for loan.\n",
    "\n",
    "---------\n",
    "## Objective:\n",
    "------------\n",
    "Company wants to automate the loan eligibility process (real time) based on customer detail provided while filling online application form. These details are Gender, Marital Status, Education, Number of Dependents, Income, Loan Amount, Credit History and others. To automate this process, they have given a problem to identify the customers segments, those are eligible for loan amount so that they can specifically target these customers. Here they have provided a partial data set.\n",
    "\n",
    "--------\n",
    "## Attributes Information:\n",
    "------------\n",
    "\n",
    "* Variable:\t------------------------------     Description\n",
    "* Loan_ID:\t-------------------------------          Unique Loan ID\n",
    "* Gender:\t  --------------------------------        Male/ Female\n",
    "* Married:\t-------------------------------          Applicant married (Y/N)\n",
    "* Dependents:--------------------------\t      Number of dependents\n",
    "* Education:----------------------------      Applicant Education (Graduate/ Under Graduate)\n",
    "* Self_Employed:--------------------\t  Self employed (Y/N)\n",
    "* ApplicantIncome:-----------------\t  Applicant income\n",
    "* CoapplicantIncome:------------- Coapplicant income\n",
    "* LoanAmount:------------------------\t      Loan amount in thousands\n",
    "* Loan_Amount_Term:------------  Term of loan in months\n",
    "* Credit_History:----------------------\t  credit history meets guidelines\n",
    "* Property_Area:----------------------\t  Urban/ Semi Urban/ Rural\n",
    "* Loan_Status:-------------------------\t      Loan approved (Y/N)\n",
    "\n",
    "---------------------------\n",
    "## Guidelines\n",
    "-----------------------------------------\n",
    "- The exercise consists of semi written code blocks. You need to fill the blocks as per the instructions to achieve the required results.\n",
    "- To be able to complete the assessment in the expected time, do not change the variable names. The codes might throw errors when the names are changed. \n",
    "- The marks of each requirement is mentioned in the question.\n",
    "- You can raise your issues on the discussion forum on the Olympus.\n",
    "- You can use Jupyter Notebook/ Google Colab(recommended) for this assessment.\n",
    "- Uncomment the code snippets and work on them\n",
    "--------------------------------------------\n",
    "Wishing you all the best!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 110998,
     "status": "ok",
     "timestamp": 1605037888979,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "Hfe1_fzF6QU4",
    "outputId": "ce05247c-48ef-4ad7-cbb2-e6c3a8e52ddb"
   },
   "outputs": [],
   "source": [
    "#from google.colab import drive\n",
    "#drive.mount('/content/drive')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Lbci6FiLe2Cq"
   },
   "source": [
    "### Import Libraries and Load Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "oL8YsUkIe2Cr"
   },
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "1rarAFQve2Cv"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "# importing ploting libraries\n",
    "import matplotlib.pyplot as plt\n",
    "# To enable plotting graphs in Jupyter notebook\n",
    "%matplotlib inline\n",
    "\n",
    "#importing seaborn for statistical plots\n",
    "import seaborn as sns\n",
    "\n",
    "#Let us break the X and y dataframes into training set and test set. For this we will use\n",
    "#Sklearn package's data splitting function which is based on random function\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "import numpy as np\n",
    "from scipy import stats\n",
    "\n",
    "# calculate accuracy measures and confusion matrix\n",
    "from sklearn import metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 202
    },
    "executionInfo": {
     "elapsed": 1021,
     "status": "ok",
     "timestamp": 1605039643575,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "ObV-NWPae2Cx",
    "outputId": "37198518-307c-40d8-a8f3-68fababac2dd"
   },
   "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": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('CreditRisk.csv')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 688,
     "status": "ok",
     "timestamp": 1605039644919,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "UKoRtVvge2C0",
    "outputId": "dfa911b4-8b94-4def-d607-459eef7ed37d"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(614, 13)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1SL8PmEne2C4"
   },
   "source": [
    "### Exploring data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 1248,
     "status": "ok",
     "timestamp": 1605039646884,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "R-kkimqee2C4",
    "outputId": "72b5991d-3c51-43cd-cd09-b78f1d138d24"
   },
   "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": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 233
    },
    "executionInfo": {
     "elapsed": 1063,
     "status": "ok",
     "timestamp": 1605039647703,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "nQEUfcrqe2C7",
    "outputId": "bb4331d9-6aa4-40ea-faf5-b17123b1a9a9"
   },
   "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",
       "    <tr>\n",
       "      <th>Credit_History</th>\n",
       "      <td>564.0</td>\n",
       "      <td>0.842199</td>\n",
       "      <td>0.364878</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Loan_Status</th>\n",
       "      <td>614.0</td>\n",
       "      <td>0.687296</td>\n",
       "      <td>0.463973</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.0</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",
       "Credit_History     564.0     0.842199     0.364878    0.0     1.0     1.0   \n",
       "Loan_Status        614.0     0.687296     0.463973    0.0     0.0     1.0   \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  \n",
       "Credit_History        1.00      1.0  \n",
       "Loan_Status           1.00      1.0  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Lets analysze the distribution of the various attribute\n",
    "\n",
    "df.describe().T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "iVUi2pH1e2C-"
   },
   "source": [
    "**Insights:** `CoapplicantIncome and LoanAmount has 0's that can't happen so no need to fill these values`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 1304,
     "status": "ok",
     "timestamp": 1605039648365,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "03n3XcrMe2C_",
    "outputId": "090b5569-5f5e-4f90-efd2-0173887cb4c3"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Loan_ID              614\n",
       "Gender                 2\n",
       "Married                2\n",
       "Dependents             4\n",
       "Education              2\n",
       "Self_Employed          2\n",
       "ApplicantIncome      505\n",
       "CoapplicantIncome    287\n",
       "LoanAmount           204\n",
       "Loan_Amount_Term      10\n",
       "Credit_History         2\n",
       "Property_Area          3\n",
       "Loan_Status            2\n",
       "dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.nunique() # Number of unique values in a column"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "-cJT4fE6e2DB"
   },
   "source": [
    "**Insights:**\n",
    "- `Loan_Amount_Term is categorical as it has only 10 unique values`\n",
    "- `Loan_ID column has all unique values and it not a continous column so it has 614 categories one for each row and that would not provide any info to the algorithm`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 202
    },
    "executionInfo": {
     "elapsed": 1096,
     "status": "ok",
     "timestamp": 1605039648366,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "1P-VrXXse2DC",
    "outputId": "cbe2fe63-462a-4d7c-ad3f-7c07cf461c36"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Gender</th>\n",
       "      <th>Married</th>\n",
       "      <th>Dependents</th>\n",
       "      <th>Education</th>\n",
       "      <th>Self_Employed</th>\n",
       "      <th>ApplicantIncome</th>\n",
       "      <th>CoapplicantIncome</th>\n",
       "      <th>LoanAmount</th>\n",
       "      <th>Loan_Amount_Term</th>\n",
       "      <th>Credit_History</th>\n",
       "      <th>Property_Area</th>\n",
       "      <th>Loan_Status</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\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>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>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>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>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": [
       "  Gender Married Dependents     Education Self_Employed  ApplicantIncome  \\\n",
       "0   Male      No          0      Graduate            No             5849   \n",
       "1   Male     Yes          1      Graduate            No             4583   \n",
       "2   Male     Yes          0      Graduate           Yes             3000   \n",
       "3   Male     Yes          0  Not Graduate            No             2583   \n",
       "4   Male      No          0      Graduate            No             6000   \n",
       "\n",
       "   CoapplicantIncome  LoanAmount  Loan_Amount_Term  Credit_History  \\\n",
       "0                0.0           0             360.0             1.0   \n",
       "1             1508.0         128             360.0             1.0   \n",
       "2                0.0          66             360.0             1.0   \n",
       "3             2358.0         120             360.0             1.0   \n",
       "4                0.0         141             360.0             1.0   \n",
       "\n",
       "  Property_Area  Loan_Status  \n",
       "0         Urban            1  \n",
       "1         Rural            0  \n",
       "2         Urban            1  \n",
       "3         Urban            1  \n",
       "4         Urban            1  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cr_df = df.drop('Loan_ID', axis =1 ) # dropping this column as it will be 1-1 mapping anyways\n",
    "cr_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 802
    },
    "executionInfo": {
     "elapsed": 2179,
     "status": "ok",
     "timestamp": 1605039649639,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "Q69J75RFe2DE",
    "outputId": "9a8e26be-5fef-4310-ca4c-950396feb127"
   },
   "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"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# distplot for continous columns\n",
    "for i in ['ApplicantIncome','CoapplicantIncome','LoanAmount']:\n",
    "    sns.distplot(df[i])\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 1548,
     "status": "ok",
     "timestamp": 1605039650293,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "I7UiIZym9ScJ",
    "outputId": "7e7eb7ab-79fa-4695-ab2c-d54af5c70c86"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "360.0    0.853333\n",
       "180.0    0.073333\n",
       "480.0    0.025000\n",
       "300.0    0.021667\n",
       "84.0     0.006667\n",
       "240.0    0.006667\n",
       "120.0    0.005000\n",
       "36.0     0.003333\n",
       "60.0     0.003333\n",
       "12.0     0.001667\n",
       "Name: Loan_Amount_Term, dtype: float64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cr_df['Loan_Amount_Term'].value_counts(normalize = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "G5rn8Y_Ue2DK"
   },
   "outputs": [],
   "source": [
    "#The Loan_Amount_Term is highly skewed - so we will delete this column\n",
    "cr_df.drop(['Loan_Amount_Term'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 1247,
     "status": "ok",
     "timestamp": 1605039653472,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "kPTyqzqMe2DP",
    "outputId": "ef168bbe-184b-41dc-af42-4e6b2463410f"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Gender                object\n",
       "Married               object\n",
       "Dependents            object\n",
       "Education             object\n",
       "Self_Employed         object\n",
       "ApplicantIncome        int64\n",
       "CoapplicantIncome    float64\n",
       "LoanAmount             int64\n",
       "Credit_History       float64\n",
       "Property_Area         object\n",
       "Loan_Status            int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cr_df.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "UWmFVTdH-hMO"
   },
   "source": [
    "### Complete the below snippets to generate the countplots for all the categorical columns (1 mark)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "NXmuhNGne2DS"
   },
   "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"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEHCAYAAABBW1qbAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAATOUlEQVR4nO3df9CdZX3n8ffH8MO6qIA8UEzShrHZrrhto30W6bp/KHRaoK3BLjg4bU0tu7EzuOrUdYvdnYq7y0w7rdJqu8ykCxKcVqBYF2rpdjFCHbcr9EEi8qOMWaTykCx5kB9CnWY3+N0/zpWbQ3KSHGjucx5y3q+ZM+e+r/u67/MNk+GT67p/paqQJAngJdMuQJK0fBgKkqSOoSBJ6hgKkqSOoSBJ6hwx7QL+IU444YRas2bNtMuQpBeVO+6449Gqmhu17UUdCmvWrGFhYWHaZUjSi0qSv93fNqePJEkdQ0GS1DEUJEkdQ0GS1DEUJEkdQ0GS1DEUJEkdQ0GS1DEUJEmdF/UdzdLh7Jv/8YemXYKWoe/79a/1enxHCpKkjqEgSeoYCpKkjqEgSeoYCpKkjqEgSer0HgpJViS5M8nn2vopSW5L8vUk1yY5qrUf3da3te1r+q5NkvRckxgpvA+4b2j9N4HLqmot8DhwYWu/EHi8qn4AuKz1kyRNUK+hkGQV8FPAf23rAc4Arm9dNgPntuX1bZ22/czWX5I0IX2PFH4H+HfAd9v6q4Anqmp3W18EVrbllcBDAG37k62/JGlCeguFJD8N7KyqO4abR3StMbYNH3djkoUkC0tLS4egUknSHn2OFN4EvDXJg8A1DKaNfgc4NsmeZy6tAra35UVgNUDb/krgsb0PWlWbqmq+qubn5uZ6LF+SZk9voVBVH6qqVVW1BrgA+EJV/RxwC3Be67YBuKEt39jWadu/UFX7jBQkSf2Zxn0Kvwr8SpJtDM4ZXNHarwBe1dp/Bbh4CrVJ0kybyKOzq+pW4Na2/ABw2og+fw+cP4l6JEmjeUezJKljKEiSOoaCJKljKEiSOoaCJKljKEiSOoaCJKljKEiSOoaCJKljKEiSOoaCJKljKEiSOoaCJKljKEiSOoaCJKljKEiSOr2FQpKXJrk9yVeT3JPkI639qiTfSLK1fda19iT5eJJtSe5K8oa+apMkjdbnm9d2AWdU1dNJjgS+lOTP27YPVtX1e/U/G1jbPm8ELm/fkqQJ6W2kUANPt9Uj26cOsMt64Oq235eBY5Oc3Fd9kqR99XpOIcmKJFuBncDNVXVb23RpmyK6LMnRrW0l8NDQ7outbe9jbkyykGRhaWmpz/Ilaeb0GgpV9UxVrQNWAacl+afAh4B/Avwz4HjgV1v3jDrEiGNuqqr5qpqfm5vrqXJJmk0Tufqoqp4AbgXOqqodbYpoF/BJ4LTWbRFYPbTbKmD7JOqTJA30efXRXJJj2/L3AD8O/M2e8wRJApwL3N12uRF4Z7sK6XTgyara0Vd9kqR99Xn10cnA5iQrGITPdVX1uSRfSDLHYLpoK/DLrf9NwDnANuA7wLt6rE2SNEJvoVBVdwGvH9F+xn76F3BRX/VIkg7OO5olSR1DQZLUMRQkSR1DQZLUMRQkSR1DQZLUMRQkSR1DQZLUMRQkSR1DQZLUMRQkSR1DQZLUMRQkSR1DQZLUMRQkSZ0+37z20iS3J/lqknuSfKS1n5LktiRfT3JtkqNa+9FtfVvbvqav2iRJo/U5UtgFnFFVPwKsA85qr9n8TeCyqloLPA5c2PpfCDxeVT8AXNb6SZImqLdQqIGn2+qR7VPAGcD1rX0zg/c0A6xv67TtZ7b3OEuSJqTXcwpJViTZCuwEbgb+N/BEVe1uXRaBlW15JfAQQNv+JPCqPuuTJD1Xr6FQVc9U1TpgFXAa8NpR3dr3qFFB7d2QZGOShSQLS0tLh65YSdJkrj6qqieAW4HTgWOTHNE2rQK2t+VFYDVA2/5K4LERx9pUVfNVNT83N9d36ZI0U/q8+mguybFt+XuAHwfuA24BzmvdNgA3tOUb2zpt+xeqap+RgiSpP0ccvMsLdjKwOckKBuFzXVV9Lsm9wDVJ/jNwJ3BF638F8Kkk2xiMEC7osTZJ0gi9hUJV3QW8fkT7AwzOL+zd/vfA+X3VI0k6OO9oliR1DAVJUsdQkCR1DAVJUsdQkCR1DAVJUsdQkCR1DAVJUsdQkCR1DAVJUsdQkCR1DAVJUsdQkCR1DAVJUsdQkCR1DAVJUqfP13GuTnJLkvuS3JPkfa39kiQPJ9naPucM7fOhJNuS3J/kJ/uqTZI0Wp+v49wNfKCqvpLk5cAdSW5u2y6rqt8e7pzkVAav4Hwd8Grg80n+cVU902ONkqQhvY0UqmpHVX2lLT8F3AesPMAu64FrqmpXVX0D2MaI13ZKkvozkXMKSdYweF/zba3pPUnuSnJlkuNa20rgoaHdFhkRIkk2JllIsrC0tNRj1ZI0e3oPhSTHAJ8B3l9V3wYuB14DrAN2AB/d03XE7rVPQ9Wmqpqvqvm5ubmeqpak2dRrKCQ5kkEg/GFV/QlAVT1SVc9U1XeBP+DZKaJFYPXQ7quA7X3WJ0l6rrFCIcmWcdr22h7gCuC+qvrYUPvJQ93eBtzdlm8ELkhydJJTgLXA7ePUJ0k6NA549VGSlwIvA05oc/97pnheweAKoQN5E/ALwNeSbG1tvwa8I8k6BlNDDwLvBqiqe5JcB9zL4Mqli7zySJIm62CXpL4beD+DALiDZ0Ph28DvH2jHqvoSo88T3HSAfS4FLj1ITZKknhwwFKrqd4HfTfJvquoTE6pJkjQlY928VlWfSPLPgTXD+1TV1T3VJUmagrFCIcmnGFxGuhXYM89fgKEgSYeRcR9zMQ+cWlX73DcgSTp8jHufwt3A9/ZZiCRp+sYdKZwA3JvkdmDXnsaqemsvVUmSpmLcULikzyIkScvDuFcf/WXfhUiSpm/cq4+e4tmH0x0FHAn8XVW9oq/CJEmTN+5I4eXD60nOxXcdSNJh5wU9JbWq/htwxiGuRZI0ZeNOH/3s0OpLGNy34D0LknSYGffqo58ZWt7N4Omm6w95NZKkqRr3nMK7+i5EkjR9475kZ1WSzybZmeSRJJ9Jsqrv4iRJkzXuieZPMngz2quBlcCftrb9SrI6yS1J7ktyT5L3tfbjk9yc5Ovt+7jWniQfT7ItyV1J3vDC/1iSpBdi3FCYq6pPVtXu9rkKmDvIPruBD1TVa4HTgYuSnApcDGypqrXAlrYOcDaDV3CuBTYClz+/P4ok6R9q3FB4NMnPJ1nRPj8PfOtAO1TVjqr6Slt+CriPwShjPbC5ddsMnNuW1wNX18CXgWP3ep+zJKln44bCLwFvB/4PsAM4Dxj75HOSNcDrgduAk6pqBwyCAzixdVsJPDS022JrkyRNyLih8J+ADVU1V1UnMgiJS8bZMckxwGeA91fVtw/UdUTbPvdCJNmYZCHJwtLS0jglSJLGNG4o/HBVPb5npaoeY/Av/wNKciSDQPjDqvqT1vzInmmh9r2ztS8Cq4d2XwVs3/uYVbWpquaran5u7mCnNSRJz8e4ofCSPVcJweAKIg5yj0OSAFcA91XVx4Y23QhsaMsbgBuG2t/ZrkI6HXhyzzSTJGkyxr2j+aPAXyW5nsGUztuBSw+yz5uAXwC+lmRra/s14DeA65JcCHwTOL9tuwk4B9gGfIfncc5CknRojHtH89VJFhg8BC/Az1bVvQfZ50uMPk8AcOaI/gVcNE49h9KPfvDqSf+kXgTu+K13TrsEaSrGHSnQQuCAQSBJenF7QY/OliQdngwFSVLHUJAkdQwFSVLHUJAkdQwFSVLHUJAkdQwFSVLHUJAkdQwFSVLHUJAkdQwFSVLHUJAkdQwFSVLHUJAkdXoLhSRXJtmZ5O6htkuSPJxka/ucM7TtQ0m2Jbk/yU/2VZckaf/6HClcBZw1ov2yqlrXPjcBJDkVuAB4XdvnvyRZ0WNtkqQReguFqvoi8NiY3dcD11TVrqr6BoP3NJ/WV22SpNGmcU7hPUnuatNLx7W2lcBDQ30WW9s+kmxMspBkYWlpqe9aJWmmTDoULgdeA6wDdgAfbe0Z0bdGHaCqNlXVfFXNz83N9VOlJM2oiYZCVT1SVc9U1XeBP+DZKaJFYPVQ11XA9knWJkmacCgkOXlo9W3AniuTbgQuSHJ0klOAtcDtk6xNkgRH9HXgJJ8G3gyckGQR+DDw5iTrGEwNPQi8G6Cq7klyHXAvsBu4qKqe6as2SdJovYVCVb1jRPMVB+h/KXBpX/VIkg7OO5olSR1DQZLUMRQkSR1DQZLUMRQkSR1DQZLUMRQkSR1DQZLUMRQkSR1DQZLUMRQkSR1DQZLUMRQkSR1DQZLUMRQkSR1DQZLU6S0UklyZZGeSu4fajk9yc5Kvt+/jWnuSfDzJtiR3JXlDX3VJkvavz5HCVcBZe7VdDGypqrXAlrYOcDaD9zKvBTYCl/dYlyRpP3oLhar6IvDYXs3rgc1teTNw7lD71TXwZeDYJCf3VZskabRJn1M4qap2ALTvE1v7SuChoX6LrW0fSTYmWUiysLS01GuxkjRrlsuJ5oxoq1Edq2pTVc1X1fzc3FzPZUnSbJl0KDyyZ1qofe9s7YvA6qF+q4DtE65NkmbepEPhRmBDW94A3DDU/s52FdLpwJN7ppkkSZNzRF8HTvJp4M3ACUkWgQ8DvwFcl+RC4JvA+a37TcA5wDbgO8C7+qpLkrR/vYVCVb1jP5vOHNG3gIv6qkWSNJ7lcqJZkrQMGAqSpI6hIEnqGAqSpI6hIEnqGAqSpI6hIEnqGAqSpI6hIEnqGAqSpI6hIEnqGAqSpI6hIEnqGAqSpI6hIEnq9PY+hQNJ8iDwFPAMsLuq5pMcD1wLrAEeBN5eVY9Poz5JmlXTHCm8parWVdV8W78Y2FJVa4EtbV2SNEHLafpoPbC5LW8Gzp1iLZI0k6YVCgX8jyR3JNnY2k6qqh0A7fvEUTsm2ZhkIcnC0tLShMqVpNkwlXMKwJuqanuSE4Gbk/zNuDtW1SZgE8D8/Hz1VaAkzaKpjBSqanv73gl8FjgNeCTJyQDte+c0apOkWTbxUEjyj5K8fM8y8BPA3cCNwIbWbQNww6Rrk6RZN43po5OAzybZ8/t/VFX/PclfA9cluRD4JnD+FGqTpJk28VCoqgeAHxnR/i3gzEnXI0l61nK6JFWSNGWGgiSpYyhIkjqGgiSpYyhIkjqGgiSpYyhIkjqGgiSpYyhIkjqGgiSpYyhIkjqGgiSpYyhIkjqGgiSpYyhIkjqGgiSps+xCIclZSe5Psi3JxdOuR5JmybIKhSQrgN8HzgZOBd6R5NTpViVJs2NZhQJwGrCtqh6oqv8LXAOsn3JNkjQzJv6O5oNYCTw0tL4IvHG4Q5KNwMa2+nSS+ydU2yw4AXh02kUsB/ntDdMuQc/l3809PpxDcZTv39+G5RYKo/609ZyVqk3ApsmUM1uSLFTV/LTrkPbm383JWW7TR4vA6qH1VcD2KdUiSTNnuYXCXwNrk5yS5CjgAuDGKdckSTNjWU0fVdXuJO8B/gJYAVxZVfdMuaxZ4rScliv/bk5IqurgvSRJM2G5TR9JkqbIUJAkdQwF+WgRLVtJrkyyM8nd065lVhgKM85Hi2iZuwo4a9pFzBJDQT5aRMtWVX0ReGzadcwSQ0GjHi2yckq1SJoyQ0EHfbSIpNlhKMhHi0jqGAry0SKSOobCjKuq3cCeR4vcB1zno0W0XCT5NPC/gB9MspjkwmnXdLjzMReSpI4jBUlSx1CQJHUMBUlSx1CQJHUMBUlSx1CQJHUMBR32kjw9hd/890nuSXJXkq1J3tja35/kZWPsP1Y/6VDzPgUd9pI8XVXHTPD3fgz4GPDmqtqV5ATgqKranuRBYL6qHj3IMcbqJx1qjhQ0k5J8f5It7V/yW5J8X2v/mSS3JbkzyeeTnNTaL2kvfLk1yQNJ3nuAw58MPFpVuwCq6tEWCO8FXg3ckuSWdtzLkyy0UcVHWtuoft1oJ8l5Sa5qy+cnuTvJV5N88RD/Z9IMcqSgw96okUKSPwWur6rNSX4JeGtVnZvkOOCJqqok/wp4bVV9IMklwE8AbwFeDtwPfG9V/b8Rv3cM8CXgZcDngWur6i/btgcZGgEkOb6qHmsvO9oCvLeq7hrRr/szJDkP+Omq+sUkXwPOqqqHkxxbVU8cyv92mj2OFDSrfgz4o7b8KeBftOVVwF+0/9l+EHjd0D5/VlW72v+odwInjTpwVT0N/CiwEVgCrk3yi/up4+1JvgLc2X7r+b717n8CVyX518CK57mvtA9DQRrYM2T+BPB7VfVDwLuBlw712TW0/AxwxH4PVvVMVd1aVR9m8MDBf7l3nySnAP8WOLOqfhj4s71+b1R9DPepql8G/gODx59vTfKq/dUkjcNQ0Kz6KwaPCQf4OQbTPQCvBB5uyxteyIGT/GCStUNN64C/bctPMZh+AngF8HfAk+3cxdlD+wz3A3gkyWuTvAR429BvvaaqbquqXwce5bnvxpCet/3+S0c6jLwsyeLQ+seA9wJXJvkggymed7VtlwB/nORh4MvAKS/g944BPpHkWGA3sI3BVBLAJuDPk+yoqrckuRO4B3iAwVQQo/oBFwOfY/Dq1LvbbwD8VgugMDgn8dUXUK/U8USzJKnj9JEkqeP0kfQCtZO6W0ZsOrOqvjXpeqRDwekjSVLH6SNJUsdQkCR1DAVJUsdQkCR1/j+3KVDgoC+u1QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "for i in ['Gender','Married','Education','Dependents','Self_Employed','Credit_History','Property_Area','Loan_Status']:\n",
    "    sns.countplot(cr_df[i])\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "8hWiWiyQB_B8"
   },
   "source": [
    "### Calculate proportions of 1s and 0s in loan status and comment on the error rate if our model predicted everyone as eligible for loan (1 mark)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "ubYinGBTe2DW"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    68.729642\n",
       "0    31.270358\n",
       "Name: Loan_Status, dtype: float64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculate baseline proportion. Comment on the error rate if our model predicted everyone as eligible for a loan.\n",
    "\n",
    "(cr_df['Loan_Status'].value_counts()/cr_df['Loan_Status'].count())*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    422\n",
       "0    192\n",
       "Name: Loan_Status, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cr_df['Loan_Status'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Td42Bodde2DY"
   },
   "source": [
    "**Insights:**\n",
    "Here we have 192 case that are not eligible which is %31.27 are not eligible. If the model predict everyone is eligible %31.27 is wrong\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "HLCxLTnDe2DZ"
   },
   "source": [
    "### Null Values Treatment (4 marks)\n",
    "\n",
    "- Fill Credit history and Dependents null values with 0\n",
    "- Fill self employed null values with 'No'\n",
    "- Fill null values in gender where CoapplicantIncome is 0 with female and all other missing values as males. Comment on why this makes sense."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3140,
     "status": "ok",
     "timestamp": 1605032448098,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "iVw8oN80e2DZ",
    "outputId": "86862558-dd7b-4e66-ef7c-6324cc9b6001"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Gender               13\n",
       "Married               3\n",
       "Dependents           15\n",
       "Education             0\n",
       "Self_Employed        32\n",
       "ApplicantIncome       0\n",
       "CoapplicantIncome     0\n",
       "LoanAmount            0\n",
       "Credit_History       50\n",
       "Property_Area         0\n",
       "Loan_Status           0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cr_df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "id": "1Wyv8j-UHQrx"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Gender               13\n",
       "Married               3\n",
       "Dependents            0\n",
       "Education             0\n",
       "Self_Employed        32\n",
       "ApplicantIncome       0\n",
       "CoapplicantIncome     0\n",
       "LoanAmount            0\n",
       "Credit_History        0\n",
       "Property_Area         0\n",
       "Loan_Status           0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Fill Credit history and Dependents null values with 0\n",
    "\n",
    "cr_df['Credit_History'] = cr_df['Credit_History'].fillna(0)\n",
    "\n",
    "cr_df['Dependents'] = cr_df['Dependents'].fillna(0)\n",
    "\n",
    "cr_df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Gender               13\n",
       "Married               3\n",
       "Dependents            0\n",
       "Education             0\n",
       "Self_Employed         0\n",
       "ApplicantIncome       0\n",
       "CoapplicantIncome     0\n",
       "LoanAmount            0\n",
       "Credit_History        0\n",
       "Property_Area         0\n",
       "Loan_Status           0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fill self employed null values with 'No'\n",
    "cr_df['Self_Employed'].fillna('No', inplace = True) \n",
    "cr_df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Gender               13\n",
       "Married               3\n",
       "Dependents            0\n",
       "Education             0\n",
       "Self_Employed         0\n",
       "ApplicantIncome       0\n",
       "CoapplicantIncome     0\n",
       "LoanAmount            0\n",
       "Credit_History        0\n",
       "Property_Area         0\n",
       "Loan_Status           0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fill null values in gender where CoapplicantIncome is 0 with female and all other missing values as males. Comment on why this makes sense.\n",
    "\n",
    "cr_df[cr_df['CoapplicantIncome'] == 0 ]['Gender'].fillna('Female', inplace = True)\n",
    "cr_df[cr_df['CoapplicantIncome'] != 0 ]['Gender'].fillna('Male', inplace = True)\n",
    "cr_df.isnull().sum()\n",
    "\n",
    "#Its not working and I don't know what is the problem, \n",
    "#Can you please provide a feedback on this?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Gender</th>\n",
       "      <th>Married</th>\n",
       "      <th>Dependents</th>\n",
       "      <th>Education</th>\n",
       "      <th>Self_Employed</th>\n",
       "      <th>ApplicantIncome</th>\n",
       "      <th>CoapplicantIncome</th>\n",
       "      <th>LoanAmount</th>\n",
       "      <th>Credit_History</th>\n",
       "      <th>Property_Area</th>\n",
       "      <th>Loan_Status</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Yes</td>\n",
       "      <td>2</td>\n",
       "      <td>Not Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>3365</td>\n",
       "      <td>1917.0</td>\n",
       "      <td>112</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Rural</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>126</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Yes</td>\n",
       "      <td>3+</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>23803</td>\n",
       "      <td>0.0</td>\n",
       "      <td>370</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Rural</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>171</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Yes</td>\n",
       "      <td>3+</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>51763</td>\n",
       "      <td>0.0</td>\n",
       "      <td>700</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>188</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Yes</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>Yes</td>\n",
       "      <td>674</td>\n",
       "      <td>5296.0</td>\n",
       "      <td>168</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Rural</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>314</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Yes</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>2473</td>\n",
       "      <td>1843.0</td>\n",
       "      <td>159</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Rural</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>334</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Yes</td>\n",
       "      <td>1</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>Yes</td>\n",
       "      <td>9833</td>\n",
       "      <td>1833.0</td>\n",
       "      <td>182</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>460</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Yes</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>Yes</td>\n",
       "      <td>2083</td>\n",
       "      <td>4083.0</td>\n",
       "      <td>160</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Semiurban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>467</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Yes</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>16692</td>\n",
       "      <td>0.0</td>\n",
       "      <td>110</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Semiurban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>477</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Yes</td>\n",
       "      <td>2</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>2873</td>\n",
       "      <td>1872.0</td>\n",
       "      <td>132</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Semiurban</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>507</th>\n",
       "      <td>NaN</td>\n",
       "      <td>No</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>3583</td>\n",
       "      <td>0.0</td>\n",
       "      <td>96</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>576</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Yes</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>3087</td>\n",
       "      <td>2210.0</td>\n",
       "      <td>136</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Semiurban</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>588</th>\n",
       "      <td>NaN</td>\n",
       "      <td>No</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>4750</td>\n",
       "      <td>0.0</td>\n",
       "      <td>94</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Semiurban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>592</th>\n",
       "      <td>NaN</td>\n",
       "      <td>No</td>\n",
       "      <td>3+</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>Yes</td>\n",
       "      <td>9357</td>\n",
       "      <td>0.0</td>\n",
       "      <td>292</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Semiurban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Gender Married Dependents     Education Self_Employed  ApplicantIncome  \\\n",
       "23     NaN     Yes          2  Not Graduate            No             3365   \n",
       "126    NaN     Yes         3+      Graduate            No            23803   \n",
       "171    NaN     Yes         3+      Graduate            No            51763   \n",
       "188    NaN     Yes          0      Graduate           Yes              674   \n",
       "314    NaN     Yes          0      Graduate            No             2473   \n",
       "334    NaN     Yes          1      Graduate           Yes             9833   \n",
       "460    NaN     Yes          0      Graduate           Yes             2083   \n",
       "467    NaN     Yes          0      Graduate            No            16692   \n",
       "477    NaN     Yes          2      Graduate            No             2873   \n",
       "507    NaN      No          0      Graduate            No             3583   \n",
       "576    NaN     Yes          0      Graduate            No             3087   \n",
       "588    NaN      No          0      Graduate            No             4750   \n",
       "592    NaN      No         3+      Graduate           Yes             9357   \n",
       "\n",
       "     CoapplicantIncome  LoanAmount  Credit_History Property_Area  Loan_Status  \n",
       "23              1917.0         112             0.0         Rural            0  \n",
       "126                0.0         370             1.0         Rural            1  \n",
       "171                0.0         700             1.0         Urban            1  \n",
       "188             5296.0         168             1.0         Rural            1  \n",
       "314             1843.0         159             1.0         Rural            0  \n",
       "334             1833.0         182             1.0         Urban            1  \n",
       "460             4083.0         160             0.0     Semiurban            1  \n",
       "467                0.0         110             1.0     Semiurban            1  \n",
       "477             1872.0         132             0.0     Semiurban            0  \n",
       "507                0.0          96             1.0         Urban            0  \n",
       "576             2210.0         136             0.0     Semiurban            0  \n",
       "588                0.0          94             1.0     Semiurban            1  \n",
       "592                0.0         292             1.0     Semiurban            1  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cr_df[cr_df['Gender'].isnull()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Gender               0\n",
       "Married              3\n",
       "Dependents           0\n",
       "Education            0\n",
       "Self_Employed        0\n",
       "ApplicantIncome      0\n",
       "CoapplicantIncome    0\n",
       "LoanAmount           0\n",
       "Credit_History       0\n",
       "Property_Area        0\n",
       "Loan_Status          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#the above code didn't work and no one answered my question in slack, discusion form or in the mentored session!!!!!\n",
    "\n",
    "cr_df.loc[126,'Gender'] = 'Female'\n",
    "cr_df.loc[171,'Gender'] = 'Female'\n",
    "cr_df.loc[467,'Gender'] = 'Female'\n",
    "cr_df.loc[507,'Gender'] = 'Female'\n",
    "cr_df.loc[588,'Gender'] = 'Female'\n",
    "cr_df.loc[592,'Gender'] = 'Female'\n",
    "\n",
    "\n",
    "cr_df.loc[23,'Gender'] = 'Male'\n",
    "cr_df.loc[188,'Gender'] = 'Male'\n",
    "cr_df.loc[314,'Gender'] = 'Male'\n",
    "cr_df.loc[334,'Gender'] = 'Male'\n",
    "cr_df.loc[460,'Gender'] = 'Male'\n",
    "cr_df.loc[477,'Gender'] = 'Male'\n",
    "cr_df.loc[576,'Gender'] = 'Male'\n",
    "\n",
    "cr_df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-here it consider that if the co applivant income is zero the head of house hold is women \n",
    "and if its not 0 the head of house hold is male so the co applicant is a female\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3107,
     "status": "ok",
     "timestamp": 1605032448101,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "SNAGJwKge2Dl",
    "outputId": "7bab5d7c-7d55-4e8c-948b-93e44d4e2806",
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Loan_Status  Gender  Property_Area  Married\n",
       "0            Female  Rural          No           7\n",
       "                                    Yes          3\n",
       "                     Semiurban      No          12\n",
       "                                    Yes          1\n",
       "                     Urban          No          10\n",
       "                                    Yes          4\n",
       "             Male    Rural          Yes         39\n",
       "                                    No          18\n",
       "                     Semiurban      Yes         27\n",
       "                                    No          12\n",
       "                     Urban          Yes         35\n",
       "                                    No          19\n",
       "1            Female  Rural          No          12\n",
       "                                    Yes          2\n",
       "                     Semiurban      No          22\n",
       "                                    Yes         19\n",
       "                     Urban          No          17\n",
       "                                    Yes          2\n",
       "             Male    Rural          Yes         68\n",
       "                                    No          26\n",
       "                     Semiurban      Yes        100\n",
       "                                    No          32\n",
       "                     Urban          Yes         88\n",
       "                                    No          23\n",
       "Name: Married, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Drawing relationship between Loan_Status, Gender, Property_Area, Married\n",
    "\n",
    "df.groupby(['Loan_Status','Gender','Property_Area'])['Married'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3101,
     "status": "ok",
     "timestamp": 1605032448102,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "WdYcFGiSe2Dn",
    "outputId": "d00b23b5-11a0-4e55-ab5b-d1f7e2cdff16"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Gender</th>\n",
       "      <th>Married</th>\n",
       "      <th>Dependents</th>\n",
       "      <th>Education</th>\n",
       "      <th>Self_Employed</th>\n",
       "      <th>ApplicantIncome</th>\n",
       "      <th>CoapplicantIncome</th>\n",
       "      <th>LoanAmount</th>\n",
       "      <th>Credit_History</th>\n",
       "      <th>Property_Area</th>\n",
       "      <th>Loan_Status</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>104</th>\n",
       "      <td>Male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>3816</td>\n",
       "      <td>754.0</td>\n",
       "      <td>160</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>228</th>\n",
       "      <td>Male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>4758</td>\n",
       "      <td>0.0</td>\n",
       "      <td>158</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Semiurban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>435</th>\n",
       "      <td>Female</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>Graduate</td>\n",
       "      <td>No</td>\n",
       "      <td>10047</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Semiurban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Gender Married Dependents Education Self_Employed  ApplicantIncome  \\\n",
       "104    Male     NaN          0  Graduate            No             3816   \n",
       "228    Male     NaN          0  Graduate            No             4758   \n",
       "435  Female     NaN          0  Graduate            No            10047   \n",
       "\n",
       "     CoapplicantIncome  LoanAmount  Credit_History Property_Area  Loan_Status  \n",
       "104              754.0         160             1.0         Urban            1  \n",
       "228                0.0         158             1.0     Semiurban            1  \n",
       "435                0.0           0             1.0     Semiurban            1  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cr_df[cr_df['Married'].isnull()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "pTa2mSFve2Dq"
   },
   "source": [
    "**Insights:**\n",
    "- `When Gender is Male Loan Status is 1 and Propert_Area is Urban then mostly they are married.(mode)`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "id": "Nf2HrgBWe2Dq"
   },
   "outputs": [],
   "source": [
    "## Replacing with mode observed in above mentioned relationship:\n",
    "\n",
    "cr_df.loc[104,'Married'] = 'Yes'\n",
    "cr_df.loc[228,'Married'] = 'Yes'\n",
    "cr_df.loc[435,'Married'] = 'No'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3087,
     "status": "ok",
     "timestamp": 1605032448103,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "J8MtBZ2Ae2Ds",
    "outputId": "491866e2-0196-4251-fb84-552bfb4364b1"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Gender               0\n",
       "Married              0\n",
       "Dependents           0\n",
       "Education            0\n",
       "Self_Employed        0\n",
       "ApplicantIncome      0\n",
       "CoapplicantIncome    0\n",
       "LoanAmount           0\n",
       "Credit_History       0\n",
       "Property_Area        0\n",
       "Loan_Status          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cr_df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Njqqbxmre2Du"
   },
   "source": [
    "### Bivariate Analysis (3 marks)\n",
    "\n",
    "- Draw a pairplot with Loan_Status hue. Comment on the visualisation\n",
    "- Use group by on loan status and figure out the mean and median values of other numerical features per loan status category.\n",
    "- Create a cross tab for dependents and loan status and state on the findings\n",
    "- Create a cross tab for the property type and loan status and state the findings\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>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>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>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>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\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>1.0</td>\n",
       "      <td>Rural</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\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>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\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>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\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>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Gender Married Dependents     Education Self_Employed  ApplicantIncome  \\\n",
       "0   Male      No          0      Graduate            No             5849   \n",
       "1   Male     Yes          1      Graduate            No             4583   \n",
       "2   Male     Yes          0      Graduate           Yes             3000   \n",
       "3   Male     Yes          0  Not Graduate            No             2583   \n",
       "4   Male      No          0      Graduate            No             6000   \n",
       "\n",
       "   CoapplicantIncome  LoanAmount  Credit_History Property_Area  Loan_Status  \n",
       "0                0.0           0             1.0         Urban            1  \n",
       "1             1508.0         128             1.0         Rural            0  \n",
       "2                0.0          66             1.0         Urban            1  \n",
       "3             2358.0         120             1.0         Urban            1  \n",
       "4                0.0         141             1.0         Urban            1  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cr_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "id": "WKedEJqap85e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x7ff06b0b8b50>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnQAAAImCAYAAADaLxczAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nOzde3xcdZ3/8dfnzCW3XpK2aUHaCiIUq6uw1svKrosKtoiAgNzvKqwiq7i6K6g/L4iuN1ZxFRUQbbkVcEFQEAQUVFaBclmVQqWgtIHSppe0uc/lfH9/fM8kk2QmmTSZzCR5Px+PeSTznTMn3yTn8pnP92bOOURERERk8goqXQERERERGRsFdCIiIiKTnAI6ERERkUlOAZ2IiIjIJKeATkRERGSSU0AnIiIiMslNu4BuxYoVDtBDj/F8lIWOVT3K8CgbHa96lOEhozDtArqtW7dWugoiJdGxKpOJjleRypp2AZ2IiIjIVKOATkRERGSSU0AnIiIiMskpoBMRERGZ5BTQlWrH3yAMK10LERERkSEU0JWiaztc9jq48+OVronIEGHoaG3v5YUdXbS29xKGGu0vUu103sp4K2tAZ2YfM7MnzezPZnaDmdWa2T5m9pCZPWNmN5pZMtq2Jnq+Pnp977z9XBSVrzOz5XnlK6Ky9WZ2Ydl+kd5d/uuaq8v2I0R2Rxg61m1u55jLH+Tgr/6aYy5/kHWb23VzEKliOm+lHMoW0JnZXsBHgGXOudcAMeAk4KvAN51z+wE7gPdHb3k/sMM590rgm9F2mNnS6H2vBlYAl5tZzMxiwHeBw4GlwMnRtuMv05v3faosP0Jkd2zrTHHOqjW07OgGoGVHN+esWsO2Th2nItVK562UQ7mbXONAnZnFgXpgE/B24CfR6yuB90TfHx09J3r9HWZmUflq51yvc+6vwHrgjdFjvXPuOedcClgdbTv+0t3937/0x7L8CJHdkcpk+24KOS07ukllshWqkYiMROetlEPZAjrn3AvAN4AN+EBuJ/Ao0Oacy0SbtQB7Rd/vBWyM3puJtp+bXz7oPcXKx1+mp//7nray/AiR3ZGMx1jYVDegbGFTHcl4rEI1EpGR6LyVcihnk2sTPmO2D/AyoAHfPDpYrtOAFXlttOWF6nKuma0xszWtra0jVX2o/Axduqf4diJjNNpjdW5DkivPWNZ3c1jYVMeVZyxjbkOy3FUVGfu1dZrSeSvlEC/jvg8F/uqcawUws1uAtwCNZhaPsnALgRej7VuARUBL1EQ7G9ieV56T/55i5QM4564ArgBYtmzZ6Hud5vehS3cX305kjEZ7rAaBsWTBTG4972BSmSzJeIy5DUmCoNDnHZHxNeZr6zSl81bKoZx96DYAbzaz+qgv3DuAtcCvgfdG25wJ3BZ9f3v0nOj1XznnXFR+UjQKdh9gP+Bh4BFgv2jUbBI/cOL2svwmme7C34tUgSAwmmfWsFdTPc0za3RTEJkEdN7KeCtbhs4595CZ/QR4DMgAj+M/yd0BrDazS6KyH0Zv+SFwjZmtx2fmTor286SZ3YQPBjPAh51zWQAzOx+4Gz+C9mrn3JNl+WXym1nV5CoiIiJVppxNrjjnPgd8blDxc/gRqoO37QGOL7KfLwFfKlB+J3Dn2Gs6gvxBEcrQiYiISJXRShGlyChDJyIiItVLAV0pBoxy7apcPUREREQKUEBXitwo10T9wGydiIiISBVQQFeKTDfEkhCv1bQlIiIiUnUU0JUi3RMFdEll6ERERKTqKKArRZShc7GkMnQiIiJSdRTQlSLTS0t3nOfa4wroREREpOoooCtFupuuMMHWtJpcRUREpPoooCtFpocekvS6BE7TloiIiEiVUUBXikwPvSToIUm2V02uIiIiUl0U0JUgTHXT45L0kCTTqwydiIiIVBcFdCUIU930kKTHJQlTCuhERESkuiigK4GLmly7SWIaFCEiIiJVRgFdKdLd9OKbXGNhb6VrIyIiIjKAArpSZHr6+tAlXS84V+kaiYiIiPRRQFcCy6ZIEScMkr4goyydiIiIVA8FdKUI06SJE8SigE5z0YmIiEgVUUBXgiDMkCFGPKEMnYiIiFQfBXQjcY7AZUgTI5GI+7KsAjoRERGpHgroRhJmAci4GMlEwpdlUhWskIiIiMhACuhGEqYByBInEQV0mZSW/xIREZHqEa90Bape1gd0xGIEMf/n6u3t1h9OqkcYQlerzxzHk1DfDIE+q4lMS7oeTFuKS0YSZgCIBQGxmM/Q9fZ001DJOonkhCFsWQurT4a2DdC4GE66AeYv1UVcZLrR9WBa0394JFFAZ0GcIB5l6HrU5CpVoqu1/+IN/uvqk325iEwvuh5MawroRhI1uVoQIxb305akejUPnVSJTKr/4p3TtkEDd0SmI10PprURAzoz29/M7jOzP0fPX2tmnyl/1apENCgiiMWIR33o0r09layRSL940jer5Gtc7MtFZHrR9WBaKyVDdyVwEZAGcM79ETiplJ2bWaOZ/cTMnjazp8zsH8xsjpndY2bPRF+bom3NzL5tZuvN7I9m9vd5+zkz2v4ZMzszr/z1Zvan6D3fNjMbzS9fkqxvcg2CGPFolKsCOqka9c2+j0zuIp7rM1PfXNl6icjE0/VgWitlUES9c+7hQbFSpsT9Xwbc5Zx7r5klgXrgU8B9zrmvmNmFwIXAJ4HDgf2ix5uA7wFvMrM5wOeAZYADHjWz251zO6JtzgX+ANwJrAB+UWLdStOXoYsTj+emLVFAJ1UiCHyH5w/cq1FtItOdrgfTWikB3VYz2xcfTGFm7wU2jfQmM5sFvBU4C8A5lwJSZnY0cEi02UrgfnxAdzSwyjnngD9E2b09o23vcc5tj/Z7D7DCzO4HZjnnfh+VrwLew3gHdFEfulgsRiIK6LKah06qSRDAjAWVroWIVANdD6atUgK6DwNXAAeY2QvAX4HTSnjfK4BW4Edm9jrgUeCjwALn3CYA59wmM5sfbb8XsDHv/S1R2XDlLQXKx1cuQxcEJJL+z5VJa+kvERERqR4j5mGdc8855w4FmoEDnHP/6Jz7Wwn7jgN/D3zPOXcQ0IlvXi2mUP83txvlQ3dsdq6ZrTGzNa2toxy+HS39RRAnGWXoQjW5SpmM6VgVmWA6XkWqRymjXBvN7CPAF4EvRYMPvl3CvluAFufcQ9Hzn+ADvM1RUyrR1y152y/Ke/9C4MURyhcWKB/COXeFc26Zc25Zc/MoO4fmVooIYtQkfIYuzChDJ+UxpmNVZILpeBWpHqX0lLwT2Bv4E77ZNPcYlnPuJWCjmS2Jit4BrAVuB3IjVc8Ebou+vx04Ixrt+mZgZ9Q0ezfwTjNrikbEvhO4O3qt3czeHI1uPSNvX+MmG83fEwQxYrGAlIsRppWhExERkepRSh+6Wufcv+3m/v8VuC4a4foccDY+iLzJzN4PbACOj7a9E3gXsB7oirbFObfdzL4IPBJtd3FugATwIeDHQB1+MMT4DogA0ukUMfzEwgBpEqAMnYiIiFSRUgK6a8zsHODnQF8kkxdUFeWcewI/3chg7yiwrcMPwCi0n6uBqwuUrwFeM1I9xiKT9hk6C/yfKmUJnAI6ERERqSKlBHQp4OvAp+kfdODwo1invFyTq8XyMnRZLaMiIiIi1aOUgO7fgFc657aWuzLVKJehC6Im14zFCbLK0ImIiEj1KGVQxJP4Pm3TUjYdTSyc14fOFNCJiIhIFSklQ5cFnjCzXzOwD91HylarKtI3yjVqcs1aAlOTq4iIiFSRUgK6n0aPaSmb8Rk66wvo4sRCBXQiIiJSPUYM6JxzK6NpR/aPitY559LlrVb1CKMMXSwa5ZoNksQyCuhERESkeowY0JnZIcBK4G/45bYWmdmZzrnflLdq1SHX5BqLMnShxYm57kpWSWSgMISuVsikIJ6E+ma/QLeISCXp2jShSmlyvRR4p3NuHYCZ7Q/cALy+nBWrFmHU5NoX0AUJ4m5XJask0i8MYctaWH0ytG2AxsVw0g0wf6kunCJSObo2TbhS/qqJXDAH4Jz7C5AoX5Wqy+CAzgVxki5FNnTDvU1kYnS19l8wwX9dfbIvFxGpFF2bJlwpGbo1ZvZD4Jro+amUsJbrVBFmh2bokqTpTmeZUVPKn0+kjDKp/gtmTtsGXy4iUim6Nk24UjJ0H8LPRfcR4KPAWuCD5axUNXFRQBfPpYiDBEnL0NWbqWCtRCLxpG/KyNe42JeLiFSKrk0TrpSALg5c5pw71jl3DPBtIFbealWPMJsm5WIkYuYLggQ1pOlKZStbMRHwnYxPuqH/wpnrp1LfXNl6icj0pmvThCulzfA+4FCgI3peB/wSeEu5KlVNXDZNhjiJXAgbS1JLik0K6KQaBIHvZPyBezWSTESqh65NE66UgK7WOZcL5nDOdZhZfRnrVFVcJk2GGInoGHTxJHWWojs1babik2oXBDBjQaVrISIykK5NE6qUULnTzP4+98TMXg9Mn4nYwgxpYsSjFleL+fb/nu5pu7ytiIiIVJlSMnQXADeb2YvR8z2BE8tXpSqTjTJ0UZNrEHXo7O3uGOZNIiIiIhOnlKW/HjGzA4Al+JUinp5OS38R+j50uTERFq8BINWjDJ2IiIhUh1InUnsDsHe0/UFmhnNuVdlqVUVcNkM2r2U6FvdzKqd7OitVJREREZEBSlnL9RpgX+AJIDe00wHTIqCzMEM2b5aWIOEzdAroREREpFqUkqFbBix1zk3Pta6iJteceML3ocv2qslVREREqkMpo1z/DOxR7opUKwszZC0vQxcNisikps9AXxEREalupWTo5gFrzexhoDdX6Jw7qmy1qiIWpgc0uYaBD+hcShk6ERERqQ6lBHSfL3clqlngMqQt1pfKdDHfhy6rgE5ERESqRCnTljwwERWpVr7JNd4X0IWBH+WKAjoRERGpEkUDOjNrx49mtehr30uAc87NKnPdqkLgMoR5Ta65DJ1L91SqSiIiIiIDFA3onHMzJ7Ii1crCDKEl+57n+tBZRoMiREREpDqMOMo1moduxLJh3h8zs8fN7OfR833M7CEze8bMbjTz0ZKZ1UTP10ev7523j4ui8nVmtjyvfEVUtt7MLiy1TqMRuAyh9ce9YUwBnYiIiFSXUqYteXX+EzOLA68fxc/4KPBU3vOvAt90zu0H7ADeH5W/H9jhnHsl8M1oO8xsKXBSVI8VwOVRkBgDvgscDiwFTo62HVexQU2uBHGyBAQZNblKlQhD6NgMbRv91zCsdI1EZLLTdWXSKRrQRVmxduC1ZrYrerQDm4HbStm5mS0EjgCuip4b8HbgJ9EmK4H3RN8fHT0nev0d0fZHA6udc73Oub8C64E3Ro/1zrnnnHMpYHW07biKuSwubx46gJTVEGSVoZMqEIawZS1cdSh86zX+65a1pV98ddGWqUjH9diM9boiFVE0oHPO/WfUj+7rzrlZ0WOmc26uc+6iEvf/LeA/gNxRMBdoc85louctwF7R93sBG6OfnQF2Rtv3lQ96T7HycRWQIRwU0KUtSSzbW+QdIhOoqxVWnwxtG/zztg3+eVfryO/VRVumIh3XYzeW64pUzIhNrs65i8xsLzN7i5m9NfcY6X1m9m5gi3Pu0fziQj9ihNdGW16oLuea2RozW9PaOroDMu4yuGDg2JGMJYlne5iuq6FJ+Yz6WM2k+i+6OW0bfPlIdNGWMRrLtbVsdFyP3ViuK1IxI85DZ2ZfwfdhWwtko2IH/GaEtx4MHGVm7wJqgVn4jF2jmcWjLNxC4MVo+xZgEdAS9dObDWzPK8/Jf0+x8gGcc1cAVwAsW7ZsVFFYjKFNrpkgSZJeejMhtYlYkXeKjN6oj9V4EpYcAQeeDHVN0L0DnrjBl49EF20Zo7FcW8tmuOM6DH1gl0n5c6S+GYJSupJPM/EkNC4e+HdsXFzadUUqppSVIo4BljjnRtXGGDXLXgRgZocAn3DOnWpmNwPvxfd5O5P+/ni3R89/H73+K+ecM7PbgevN7L+AlwH7AQ/jM3T7mdk+wAv4oPOU0dSxFL4P3cATPhskqSNFR29GAZ1UVt1c+Of/gJtO9xffxsVwwjW+fCS6aMtUVOy4TtT5ptdc9q5xMZx0A8xfqqBusPpm/7cZ/Leqb650zWQYpRzFzwGJcfyZnwT+zczW4/vI/TAq/yEwNyr/N+BCAOfck8BN+AzhXcCHnXPZKMN3PnA3fhTtTdG24ypOFgY1uYa5gK4nU+RdIhOke1t/MAf+602n+/KR5C7ajYv9c120ZSoodlyHWTXFlioIfKD7gXvhgj/7rwp8q14pGbou4Akzuw/oy9I55z5S6g9xzt0P3B99/xx+hOrgbXqA44u8/0vAlwqU3wncWWo9Rss5R5zMkCbXMFZLvbXT0auATipsLM2m+RdtNUHJVFHsuN71groYjEYQwIwFla6FjEIpAd3t0WPaSWedz9ANCuiyiQZm8xIvKaCTShtrs6ku2jIVFTqu1cVAprgRAzrn3MqRtpmqUpksMyyEYFCGLtFAk3WyXk2uUmnq6yJSGp0rMsWVMsp1P+A/8asx1ObKnXOvKGO9qkI6FbUwD+pD5xL1zKKTjp50BWolkicICJtfRfbse7BsChdLEpvRTKBmU5GBxrGLQRg6tnWmSGWyJOMx5jYkCYJCM2mJTJxSmlx/BHwOvxzX24CzKTwH3JSTSucCuoEZOks2EDNHT9dO/GwpIpURho51Wzo5Z9VaWnZ0s7CpjivPWMaSBTN1gxEZbBy6GIShY93mds5ZtUbnnFSVUj6a1Dnn7gPMOfe8c+7z+OW7prx0KuosO+gTXJBsACDTuWOiqyQywLbOVN+NBaBlRzfnrFrDtk519BYpB51zUq1KydD1mFkAPGNm5+PnfJtf3mpVh1TKN6naoCbXXECXVUAnFZbKZPtuLDktO7pJZbJF3iEiY6FzTqpVKRm6C4B64CPA64HT8BMAT3nZqMnVBg+KiAK6sHvnhNdJJF8yHmNhU92AsoVNdSTjmvBapBx0zkm1KmUt10eccx3OuRbn3NnOueOcc3+YiMpVWjrtU+iDA7psvN5/09020VUSGWBuQ5Irz1jWd4PJ9eeZ26CpGETKQeecVKtSRrneAxzvnGuLnjcBq51zy8tduUrLFAvoEj5DZ70K6KSygsBYsmAmt553sEbciUwAnXNSrUrpQzcvF8wBOOd2mNm06EOXC+iCQX3ochm6eGrXhNdJZLAgMJpn1lS6GiLThs45qUal9KELzWxx7omZvRxw5atS9SiWoQvj9YSYAjoRERGpCqVk6D4N/M7MHoievxU4t3xVqh6ZaI2/IDaos6sFdFsdyYwCOhEREam8Upb+usvM/h54M35C4Y8557aWvWZVIJv205YEwdDRS13WQG2mY6KrJCIiIjJE0SZXMzsg+vr3wGLgRfwcdIujsikvWyxDB/QGDdSF7RNdJREREZEhhsvQfRw4B7i0wGuOabBaRDbjM3SxAhm63lg9DalOMtmQeEzrZoqIiEjlFA3onHPnRF/fNnHVqS7DZegy8QZms4ldPRnmaP4hERERqaCiAZ2ZHTvcG51zt4x/dapLGAV0sVh8yLDeTLyBOdbFzu60AjoRERGpqOGaXI8c5jUHTIOALmpyjcXIDHrNJeqZTSebu9MTXzERERGRPMM1uZ49kRWpRrmAzgo0ubpEA/XWS3tHJ9A4wTUTERER6Tdib34zm2tm3zazx8zsUTO7zMzmTkTlKi3MRhML29CALqjxy391tW+f0DqJiIiIDFbK8MzVQCtwHPDe6Psby1mpauGyUUNrgVGusRq//FePAjqptDCEjs3QttF/DcNK10hESqFzV8ZRKStFzHHOfTHv+SVm9p5yVaiahFnf5OoKZOhiUYYu3aGATiooDGHLWlh9MrRtgMbFcNINMH8pBJpOR6Rq6dyVcVbKUfNrMzvJzILocQJwR7krVg1cJhfQDY17g6QP6DJdbRNaJ5EBulr7bwjgv64+2ZeLSPXSuSvjrJSA7l+A64FU9FgN/JuZtZvZ1F7MNJehK9DkGiZ8QOe6d0xolUQGyKT6bwg5bRt8uYhUL527Ms5GDOicczOdc4FzLh49gqhspnNu1kRUslLCsHiTa5jwfejoVoZOKiie9E01+RoX+3IRqV46d2WcldRQb2bHmtl/mdml06X/HIANk6HLxn2GLuid2klKqXL1zbiTbui/MTQu9s/rm3d/n+qoLRMsDB2t7b28sKOL1vZewnDwVO5TUH2z7zOXd+4ylnNX5+20N+KgCDO7HHglcENU9EEzO8w59+ER3rcIWAXsAYTAFc65y8xsDn6U7N7A34ATnHM7zMyAy4B3AV3AWc65x6J9nQl8Jtr1Jc65lVH564EfA3XAncBHnXPjdyWIMnQU6EPnYklSJIilFNBJ5YQYW+v2Yc6ZdxK4DKHF2R7MYR5W2qe1ITtUR22ZWGHoWLe5nXNWraFlRzcLm+q48oxlLFkwkyCwSlevfILAn1cfuNc3s8aTPpjbnfMsDHFb1mJ556076QasnOdtGPr+fmOtu4ybUv76/wwsd879yDn3I3zAdUgJ78sAH3fOvQp4M/BhM1sKXAjc55zbD7gveg5wOLBf9DgX+B5AFAB+DngT8Ebgc2bWFL3ne9G2ufetKKFeJXPZrP9qhf9MXdZAMq2ATiqnvaeXOb0vEN/6FMEu/3VO7wu09/SWvI/87Ei6fYs6asuE2taZ6gvmAFp2dHPOqjVs65wGfcmCAGYsgMZF/utuBkSus7U/mANo24CtPhnXWdp5WzBDOlzGL/fB76pD4Vuv8V+3rFVWsMJKmbZkHbAYeD56vgj440hvcs5tAjZF37eb2VPAXsDR9AeEK4H7gU9G5auiDNsfzKzRzPaMtr3HObcdwMzuAVaY2f3ALOfc76PyVcB7gF+U8DuVJkyTxaBIQNcT1FOT7Ri3HycyWnWZXcQ7N8MdH+/7ZB4/+nLqahuB2hHfPzg78vsP7cee6qgtEyiVyfYFczktO7pJZbIVqtHkE2Z6iRU4b8NML0M7DA16b4EM6TXvewN7Z58fkPEbkKkvNkL3A/f6wFQqopSPA3OBp8zs/iiIWgs0m9ntZnZ7KT/EzPYGDgIeAhZEwV4u6JsfbbYXsDHvbS1R2XDlLQXKx082TXaYmLc3Vk992DE9+ntIVUqEPXDbeQMvrLed58tLMDg78mJHqI7aMqGS8RgLm+oGlC1sqiMZHykUkZyMJQqet2kSI/ZJLJQh7dj+0pCM34BMvUboVqVSArrP4ptDPwd8HvhPYE/g0ugxLDObAfwPcIFzbrj2yUKdJdxulBeqw7lmtsbM1rS2lt50FLgM2WE+36RiDcyik45UpuR9igxn1MdqGBa+sJbY9DE4O3LJ/a1sO3Ll+HXUliltd6+t+eY2JLnyjGV9QV2uD93chjF+iJhGgwS6Ek20HT3wvG07eiUnX7+eYy5/kHWb24sGdYUypI3JIteVXMCmEbpVacQmV+fcA2Z2IHAKcALwV+D7zrkHRnqvmSXwwdx1zrlbouLNZranc25T1KS6JSpvwTfn5iwEXozKDxlUfn9UvrDA9oV+hyuAKwCWLVtWejotzJAtMGVJTjbewCxeYmdXmlm1iZJ3K1LMaI/VbKyWeOPigRffxsW+vISfl8uO5C7oj2/cxacerOE7Z99DwqXV2VmGtdvX1jxBYCxZMJNbzzuYVCZLMh5jbkNybAMixmtwTyU6/u/Gz5xdV8PzM/Zj4+G3sO+cBM9uT/PZO1/i8Y0+h3LOqjXcet7BNM+sGfLewdcAgLZUwKIC15W+gC03Qnfw31cf/Cqq6DXfzPYHTgJOBrbhR6aac+5tpew4GrX6Q+Ap59x/5b10O3Am8JXo62155eeb2Wr8AIidUdB3N/DlvIEQ7wQucs5tjyY3fjO+KfcM4L9LqVupgnD4JtcwUc8s62JTd3pAJCoyYYIAjl8JXVshUQ/pLqif138DGOHmkMuO5PefueCwA4jNnAlTeYShVJUgsILBRkkGH+N1c6HjpbH38SolKBzvgG83A9EgMF4+dwbbahfTmspw5I/uH/D6cH0SC10DZszZw4+SLRawjecIXRk3w32Ifxr4LXCkc249gJl9bBT7Phg4HfiTmT0RlX0KH8jdZGbvBzYAx0ev3YkfQbseP23J2QBR4PZF4JFou4tzAySAD9E/bckvGM8BEYCFGcJhMnQu0cBsOlnXpX4DUhmxMDV00I4FvrxzK7S/NOzNoSzZEZGJUigAOuEacEWaDNPdvgl2UAASho5tnamB58BwHf/rm6GzFVIdsP1ZeOCr0LFl7FP8jGGwQS4obm1nSMZtuD6Jxa4BxggBW26ErlSN4QK64/AZul+b2V34Jb9Kvso75343zPbvKLC9AwrObeecuxq4ukD5GuA1pdZptJKul3QwTJ+AZD1xC+ns3Ako1SwVYDHobR8wypWjL8dmLIAXH+svh6I3hzFlR0QmSqFsWKEA6KbT4ZSb/bkwuMlw61/guuMHfLgJsYLz4B1Ql8KK9U8dHEQe9R341cVjH+k5DoMNCmXcRuqTWPgaYArYJpmiHyOcc7c6504EDsD3WfsYsMDMvmdm75yg+lVU3KX86KEiYkm/WkT3ru1FtxEpqzBdcJQrYdo3wWokmkwFheY92/xnSHUXOca7fZCVP7jn6Mt9Ji23TTRqs9g8eMVGjoZhdmgQefv5cPAF/efX7g7GGIfBBvkZtwc/+TZuPe/gqT9JswClreXa6Zy7zjn3bvzAgyfonwx4Sku6FJmgeEAXr/EBXW+HAjqpkDBbPGjr3qGRaDI1FMrE3XgqZHoKH+M7W3zGbPmX4X13w1l3wH2fh5Y1/dtF54kLs1z27pfx2395Jf9z+r4ctGgWLTu62cGsIUtztR29ktb2IkFkXZNfoSHMDAg83Wgm3B1uObBRjNrNZdz2aqqneWaNgrlpYlQN/c657c65Hzjn3l6uClWLMHQkXZqsFb/5xWrqAUh37pioaokMFCucRWDnRnjwWz4rkXdzyJxwPWHdvImvp8hYFG2K7PaDgvIDoONXwRM3+ODt7k9BcgbEanwft3yNiyGWZF7Xs7z+nuNZtPKNvP6e47lqRQPLlzZjQayv43/2o/Skk7EAACAASURBVH/iT4ffwtl3dtKyK1v4nEt34U68Drv707u9YsOAwQYX/Nl/nb/Uv6aVGWQEGpJSRCobUmspssM0uYbJmf6bTmXopIKO++HAG9pxV8GvL4GOLbQn5/H0u29l45kP8+hhN/Phe7vZ1qV5E6XKjLTMlJnPtJ14LSxc5stzmTgzn4k76w7/dc1VcODJ8JEn+gOihiKZryBGcOMpAwKwuT87k2++e6HvcxZ1/H+JeRz5o7/w+MZdhedqPPE6eNnfE9Y3w7o7Bv5u0YoNJSu0HFixwRJakk/ylDJV1bTUk85SQ5owmF10m0xyFgCxbp1UUiEWh3gNHHFp/7QlNbNxx68k5WKccd36vrmocj575PgvqVRwlKCaeaQUw03VAYUHIDz0A3jTv/hm1YMv8Jm4/AzeX38zdHBCoWk2dr1QMPNXF2SxvOM3f662xzfu4gN3wcWH38LS+TXEEjV9I0AzbZuIFRiMkbFE3xT1JZ0rgweAFJtAXP1hJY8CuiI6ejPUkCITK56hywV0iZ5tE1UtkYFcFm48behovrPvYldsDq0d6QGbL2yqw8wIQzduAVehtSCvPGOZOmJLaYabqgMKD0A46w74xYW+WfWJG/xUJTedPvwkt4Wm2cgNQhh0/tigfqaDR462dqSJz1qAzR44X2NXoonuo1fSeNuZfXVpO3olKZuNtffSVJfgmdaO4c+VQgHuidfBkiMGZv/UH1YGUUBXRFcqS72lycSGOWGCOLtsBrW9CuikMiybghnzfVNTXZMfCPHgt7Bsirkzh05f8NXjXsvnb/8zHztsybgFXMVGCRabmV5kgJGm6ij0GgZHfhMO/2r/ZMKjmOQ2lyVz4UzmnXg9wf3/6ZtpG5p90Fc3d8D2I83XmNtfGDo6G17JxsNvoTEZ0pYKID6Xz177OK0dvfzg9Ndz2b1/Gf5cKTYA5IzbYfOfSluZoRIrXEjFKaAroqM3wxxSdA4zyhWg3WbTkFYfOqkMF6/F3vH5/qlLoukZXLyWYFcLB8xIcuuH/oHnt3ezrTPFN+5ex+Mb21i7qX3cAq5Ca0EONzO9yABFsmR92adir81Y0N98uStFMt7I3FkjN1+GdfNYt6Wz70PIh/7p5fz7IRcS3Hhq3zkUnng9zF9KYNb33iCepLmhGYKaQbsfmKF+59L5fOaIpWDG1i0dfPv2p3h8YxsA/3LNo/y/dy/ll2v7B2gMOVeKBbhBvLSgdbyWPZNJR//dIrp6fR+6wan3wTpis5mV1ShXqRCXLTgPnYVp+NZrsB8eypzO9Xzpjif5l2se5fGNbRy0qJH/9+6ldKUytLb3Fl20u1S5/kX5hpuZXmSA4abqGOa1XCB1zOUPcvBXf114EfpC89dtWcu37nm670PIoXvH+4M5gLYNBDeeQmrXZj/lyM8+BpuegO1/hbbnIZsZMIgj2755wP5+uXYLp1z1EDUxmJXdzrffNW/AdCiDJ/gdcq4MNxfd4MEShWgAxbSlDF0RHb0Zakn5aSGG0R2fRWPqbzjn8MvXikygMFP407wL/WjAgy8glu7khhNfzsk3Pg8EfGL5Ej75P38ct/5uuzMzvUifIIDmA+DsX0A27a+5M/boD1iKrBm6rb135Kb+AsFNcOMpfOO0uzj29Yv59dqXeO2e9XDqTyCIwa4X4b4vQMsagkwP9usv+8EXt58/sD9bbSNsfRoS9STSXXztbYvY0t7bNwCpeUaCuV3PMv8eP4J2UeNirjpyJZ96sIb5M2v6BlgUPFfGuvD9OKw2IZOTAroiunt6SFiWWDxJepjtehOzmcsuutNZ6pP6c8oEC+KFm6SCBLz9s303otrGxaw6ZhVPZhbyiZ/8cXT93Uboj6P1YGVMwhBany7eRFhkzdDchMDz640tXY5L7m/l8Y27cGHWZ88yKZ/BLhDctLV3cMujO/nuobXEVx3R/3OPXwnHXgUtj+Es5vvV5YK56L3c/xV46ycGLLc3+4RV3HDa/vyttZ2GWJY5M2qJ3XvR0OlQzrib5Ky6oecKzs+TlzvHmg/Y/YXvR2rClilLEUgRPd1dAATx4TN06cRsZlkXG9t2UT9/zkRUTaSfxf3kwYP60GGxITeimbeewYFn/pKWHd0ctKiRDx6yL3vPqWNxTSeJ9Eukd9YQm9FMEMtr/imxP47Wg5XdtjsL0och8wpkwK56uol5Xc9Cbm65990Fp97sp/RxoV9ZJVFHU20NH3vLHOI3HTPw5958Jpx2K+7lbyKWaIBZew0NCA882W+X/74Hvk7tWz/BAT/vH93KUd+Bzs39q1PkpkOJBwPOlTCbJdyytn8+vLH2eRtrhk8mLQV0RfT2+IAuNsKnmqCuEYDtL21gkQI6mWiZbr+k0dHfhVkv8zesXS+CK9AUO2M+NZbmiY8ewIzaJB2ZLA2pv5FYdVrfhT9zwvWwYGl/ULc7N1uR0didJsKu1oITAn/irDsJfvwuX75wmf9AcsfH/Ujwd3wefvYRaNvAjMbFLDn9p4V/btdWrKGZuMX6lvPq227hMljwd0NGlRcM8m4/329342m+rMB0KGHo6Nm5mfpBv8uYzrH81SY0ynVaUUBXRKrUgG72y6AFul5YC689cCKqJtIviMGit8DMPaFzK3S2whM34A75JPYPH4HFb/Q3nmwaamdjPz6Cxih4azztFrj5tAE3kvhNp5A++x6C2Xv4m2EmBe/5Xv+Nq2WN+uPI+NqdJsIiQWCQ36f04Avg99/1QdW8/f2Ahhnz/ettG7Dtz/b/3Ki/KQ3Nvn+cBf6cufvTPtN2+/n+vYd/Hbq3909knMvE1TUVDg4bmvt+n/DE6wkGZcm2dabItHdQPx593jRVybSngK6IXEDnRhgUUTNnod9uy9qy10lkiEQdvPa9cO2xA24wdv9XYfklsOrogU2xeTc0urYVvAlZNlW4qfX4lX6Zpc5W/3NFxkPd3KETA59wzZC54AbIBYEz5vtArK4J0l2E8Vq/UsOM+bDgNfAPH4affnDgMbzmh/D4tbD2NjjhWnjgqwUGPlwL9XP9RL6dm31QOP9VsP25/r5z0J+JO/PnBYPSjto92HHmw2zpciyqX8z8QQFWKpPlpY6QPcfa501TlQiatqSodK8P6MJg+JMqWVvPJjeXura/TES1RAZK9wxcKSJ3gznwZOhuG7jG5R++529+OZ2tBadHcLFk4abWm8/0a2fe8XFof0kLg8v46N4GD3xt4LH6wNd8eTH1zXDarb4Z9e5PwY+PgDs+TmbXFrJn/NyXu2x/hi633998A976H3D6T/sDwXdePHTgw42n+e4Lp97sBygAOOf74hXKpnXv8B+Y8qZX2XbkSk6/eSP/9IP1fPTnL2LB0Gl8kvEYVzy6a8jasOGJ14+uz5umKhGUoSsqEwV0boSJhQE2xBayR8f6cldJZKgio/hoaIa6Rsj2+u8b5vuMXSblm5fe9hmYvdBnFtpfhF9+Bjq24E68jrg5XCaFFdpvrmlJ/ehknLhMClt3x5BF7d3hX6XoOOkggJqZcO3AQQ01PzmN7Jl3+EFCZ/4M3v4ZPxLczJ8D7/oadG3v60tH42I47ZbC59DOjf7Dy4nXwh9/4ssH96kD/zzXreHYK6F+Di5eS52r4dpTIOYyJAOHZTZBW8xnt+vmQBAwtyHJBYcdwKfueZpzD7uZPWcENM6cQe3sYeaZK0RTlQjK0BWVTflpHdxwS39FNiVezsvSz0NvR7mrJTJQkCg8CemMBT6bkKiHa46B777BN79mU3DUd/0N7TvLYOW7IV4Lx/8YTrwWq52N3fFxn50otN/uaBLtAjeLMHS0tvfywo6ucZmwWKaHDPGhx9qSI3AYtG30U5BkM30T+fY9z3QPDGIWLoPlXybmMj4bl6j3H2iuO84f69cd5/uZ/vHmgZmsjs2Fj/Vsuj9b9/ozYMtTUD9vSCaOY6+Euy7y9cmm4drjsG/9HfUrD6Wh6wVqOzYS/Phw7LLXwY/f5fez/TkIw74pfy455nXs8bLFJOa8nNrGPQeONC9FkcmI05bQeTiNKKArIpvqASAsJUNX/xoSZOBvvy13tUQGsmDoDeb4lZBsgJZHhjbH3niqf8+M+f1lN50BO573GY8wA+/4LPbI1bgTrx2436O+4wdG5J7n9fEpadZ+kQJCi8ExV/Qfa0uOgH/+d4IfH563usOTPghqex62P4fb8iRs/Uv/ew46DY75gc9Gb34SnrgBMj3+2M4//m86Hd74fh/85bhw6Dl09OV+wFHufV3b4FVHwfXH+1HluWbcIy71H5LW3YFr2nvoqi1dW+GWcwaW3XYe7Hiurzk0N+XPXk31NM+s2b35GwusqLHtyJWcf9sGnYfTiJpciwjTUYZuhD50AO2zl9CxvZbk03eRXHJ4uasm0i/TA39cDafc7G9AYRZ+/99wyEW+E3ehEard2/2kw7+6uL9sxoKBAyiO+g7MWEDHKT+jLuYwg+DuT/vtC8xrta0zNfKs/SIFxMhCLOmDo0S9bxq97rj+IGjG/CiLtjdsXQcN8/ygnzd/EE5YBQ98HZZ9wGei+wZVrPLnQqFmSAJ4z/ehp833BW1o7p9iJDcVyX2fh0O/4N/TuNj3N62f2z+gKDcVCfjArnExlu4Z+vOK9blL1I9vc2g0VUn67HtobdvFpo6QS+7yEy0/ualD5+E0oYCuCBcFdOEIo1wBmhsSPBC+luVP3gorLvGZDpGJEK+F157kMwe5m9kxV0CqE65778DRfevvhZcd6JuNulrhPT+Ajpcg3eWbbvNHwN5+Ppx1BzOuP9I/X3IE4fIvseuQL5KsqaN29gKCvD4+qUy2L5jLGbLouEgBgcvCkz+Fg071H0osNnDet7wVT/pGoP7Tx+EnZ/lj9ujvDQ0AO1t9P7XB/d2WHAE9O/oz17n9vfMS3480NwlwrntB7sPNQz+Aw75QuP9cusufX7teGPp6uqv4e8Z75YYgYEs4i4O/9+iAYp2H04eaXItJ+ybXUjJ0r50HV2aPJJ7aCb/5erlrJtLPhUObeW49t3/OrROv9Vm6rq1w0OkwZ1/o3QUO36E8Gh1I+yZY8bX+pqi2DVg2b06vdXcQrDqKukScZ1s72dkzcEG8ZDzGwqaBU5kMWXRcpIAwloTXHOs/lHxnGWxf3990+LbPFB6BOmO+z4wdexUkavu7EBx0ms++1czyTbTHDxw9yjsvLtAN4TQ/RcrhX/fHfy7Ia3y5z9o99AM45JN+cMXgptkTr4XaJj/o4uErSR8/qJtC/Tzfx25wc27TK8qycoPOw+lNGbpiMj0QjDxtCcDMJIRN+3JXzyGsePAyP//Ra0+YgErKtJfruJ2vbYNvthqc2cg1T731E/C//z00CDziUj+Vw42n+e1jcT9S0IK+Ztuajo383S8+RPbE66Hu1X0j8eY2JLnyjGV9za4FFx0fL5pAdUqJZXv9B+Fck6cLfZb5f//bj8QuOHqzZ+Dci0df7rseLHv/wPL3fB+OvcI3l+58of/9g/fXvd2fS+/9kV9Xtn4evPiYr8+ys6BmNqw60geOuXqmu/zDAQ3NZA+7mBe6k/S8+1aaahz1dfXUJBLEyBKcfRcWpn32MW+U63ib0PNQqo4CugLC0BHPdkNQ2ihXgDcugPPWfoCnm58j+fAVCuhkYgRB8WkUchO1Qv/gh+Vf9vPJLf/ywGkicv16EvX9wd8vLoQ3fwju/ZxfOPzoy/3ry79MLN3ppzuZ+TIIgr7RekMWHd+dDt7DmUQTqIahY1tnqrx/j6nAbOgEwCdcA4d/xQdXhY7vHX8dOtDg1P8Z2PTatsHvc/mXfX+6VIfP2hXaX2ern8/u9FvhuuPhI4/7fqcdW+CYK3Au9NP4FOo/99MPwRGXErvuePaOBiN84PZOWjvSE953bcLOQ6lK1XUFrBJd6SxzrIMMccJYbUnveetekIwF/Nze6kcXbnu2zLUUwWfPjvrO0NGoQVA4E5GbR65hUHNPrl/PzD18pq633Qd8t53ns3a5m2btLH/ju3o5/OhwH1xFEwyPy2i9kUySCVQ16ncUHP3BHPSPRg0zfhWHwcf3Cat8eb62DRBLFJ+TMd3lj/1C+8uN3m7b0D9dz64X4bgfwRGXkq2bQ2tvrPg0PrkPQ9HPm/uzM7n40D0qlhmbkPNQqpICugI6ezPMZSfd8Vn+02MJ5tbCca+E/2r1fZDC9b8qZxVFPIfv45M/G/5DP/Cz2he7AeX69uTf1I65wo+K7dkFTfvAH2/0r+WCwNz3nVsrGky5IhOouiqbQLXYqN9tndVVz6oQFuk2YDGfIfvVxf64/vDD/R82OrYM3L5xsR9QUWxOxsaX+6Aut7/TboH33e33mxvtncvUvef78H83gAGNe5ON15OqnUPb0SsLB4L58zNGdV86v4YlC2YqmJIJNekDOjNbYWbrzGy9mV04HvvcvKuHOdZOb3zWqN533L6wcH4zW1wjv7r35zz90q7xqI5IcbnmqtzyR3d/yj9P9xTORDxxQzTHVuD7C/3rY361iLpG+NG74Mq3+T5If3dCfwfx3M0qd8PLN8Gz0Wes8ETKGRt5NPpEKveo3yk1iXOxQCxI+GO1Y4tv5rz3C76/5B++X/jYjiULz8mYaPDzMtbM9E25HVt8RtA5f77kgrkTrvH99/7vej9y/MdHwHffQHLVu5ix8xm+sgaefvetZM9/zAeWv7q4vytCbn7G6OfGEsUzY1PqfydVZVL3oTOzGPBd4DCgBXjEzG53zq0dy35fbOtmD9tFmBzd9CM1MfjMG4yOh/Zj//anOPu6x7jjI/9EbUIjjKSMcutV5ubQ+v13YfmX+jN3s/aChnk++FvxFd805ULfpPXL/wfv+Cxcf8LQ9WCPuBTidX5OrsbFcOJ1cP9XBv7s0S4iPkY7mEXsyJXM/dmZff2tth25kiyzmD9htRhZbrRhflA3XqMNc825gzu+T9aMUDZWR/zoy/tHa+cGOex6AXD+A0fXNh+QPbrKH6/ZlJ97MdXu55J76Ad+BGuyoX8+u7RfvpGw1x+3rzvFrxF77JV+cEO8Ds76he9bF6+Buz/tuxmcevOQkeONt53J8YfdzIqr1nHQoll85pDFvO7YHxFP1AzMGBaYozHfVPvfSXWZ1AEd8EZgvXPuOQAzWw0cDYwpoGvZ0c2r2YXV7jHq95pB/R6vZI/2R2hrfZHvP/AsFxy6/1iqI1JUqmYuyUMuxG48NW9erev8ze9dX4NML2x/Fn7x7/6mc/xKn4377aXw+LV+J4ddXLDJK9X0SjZ0xFh8zNUkkrVY/Vx426dg858GDkgow/QLxVgQ41MPZjj3sJuZX29s6XJc8eAuLjmmuj40lXO04ZSbxLmuieyMPYjlB2INzXDfxT7AWrgMlv8n3PM5OPBkH8zhBs69eOK1fqqSMPRT9OQkG+APV/hjvWsHrPgyzjnMhbDtGZ/tO+STUNeIO/yr/oMQFFzHeH69D7ge37iLj/48b8BDQ7Nf17iEUddT7n8nVWWyB3R7ARvznrcAbxq8kZmdC5wLsHjx4sEvD/FCWzdzbBfdtaNrcs3panoVAKcueJ6rfzeXD/zTK5hRM9n/1DIRRnusJpI19DbtT81Zd/qMWxCH2tlYNhVNkVAPza/y83UFMZ/1uOWcgROoxhIFR/65eB2z584h3pDEctmD+UtLvnmVQ24x82qflqGcow2raRLn0R6vhcTjcTJN+5BKzCAI08TIYo9cDW/6F//hoWUN/O5bPuByWT8QKJaEs+70z4ME4LBv7Oen2UnU+yAriPk+oUuW497wPghipGMNbOtMsWf7n/x2B54Md/wbdGwhfdY9tGSb6Ny+ib8rcD60pfxxPuSYCwLfT68E1fS/k6lnskcZha6OQzokOOeuAK4AWLZs2YgdFlq376DBeulI7l5A1z3rFWTiDRw1cx3f3vxarv3D83zwn/fdrX3J9DLaYzUIjGSyhq3p+aRclmQsxtxEnGy6F5wjFmbADFt1lG9mevtnBzQPhSdej2HY4Cavk26gZtYCmgcHa6O4eZXDZJqWITfacLyVszl3tEZ7vBYTj8ehcU//JAzh4PMhDHFn/4IwDMlYgq5EE7PrivRNa9/sj9v7vuCP8bxVUsITruFT9+1g9aObAFi+tJnvHbYnwY2n9G3jTrqB5Kz57I2xs34R2ROvJzbo9YUzF/HgJ/ca0zFXTf87mXome0DXAizKe74QeHGsO+3YvhmATGL3AjosoGvOUhZt/wOvfdn7uOq3z3HWW/ZWXzopi4KBQ8O8vr46zTMSrDpmFTNvPcN35D7iUtycfbHkDILc9CXJ+v6MR7wuynBU55ipcgVKk8WUnzw270ODAbHoMex/vCFanH71yQOOceK1vJSdze+eexjwwdMFhx0A8xsGZJotyjQHQFNDrZ80e9DrTUFAU8PYfrUp/7+TiprsAd0jwH5mtg/wAnAScMpYd5re5TMY2d3M0AHsmv8GFm55hPOWbOSDv2/kxkc2cuZb9h64Uc8u+NtvYf8VvnlAZJwMzmRlkgHu/ff6pti8G1ifhnmVq6yMymTKUk6YaHH6QkHaHqEr/LcaLtNcpky0/ndSTpM6oHPOZczsfOBu/Ie4q51zT45ln12pDA29myEJmZrZu72fXQveTPqZ63nzpuvYf8EnufI3z3HqmxYTb30S/m+177/x+LV+tv2DL/ALP4uMo6GZrNImyZbqN92zlAUVCcKq7W9VbfWRqaM621RGwTl3p3Nuf+fcvs65L411fxu2d3Fg8CxZYvTM2L1OvgAuiLPt5UfQ+NKD/OvC9bS0dfPrhx6FlUfiHvoB/OZrvjP64rf4OYw2+zj0mc3tHPnfv+PC//kjre29Y/11REREZBqY1Bm6cnhgXSsH2Xo6Z7y85HVci9m2aDmNLz7A4c98nn+fcQyv/uWddFov70l9hZq6GXzxjfM4aE4KNj0Ov/k6f/qHyzj96ofIZB1PbdrF9s4UV5yxbJx+MxEREZmqJn2Gbrzd8+cXODD2LOmm/ca+syDOhgM/QTZWx4czKwF4X/ZTLF28J63hTI7/eS8/fqaG9JKj4MlbufzK71IfZLn8rWnOe1U3v1y7mTV/2z72eoiIiMiUpgxdnpYdXez5wt3UJXtpnbN0XPaZrpvP+oP/i3hqJ92x2XzSjJoYnLw/fONx+PyDPfwnh3Nb8n/5TvB1nEsS/20PbwWW1b6eL/7PJ7jpo8upCczPIVY/x0+WOQrdqSzfuvcv/P65bSyeU89/LD+AxXOjxaTDLGx8CFqfhk3/55t+5+wLb//00OV4Rqv9JT9fVP2cse2nt8Pva84ryj7ycv2Wdr5859Ns7ejl0Fct4IP/vC/JuD73iIhIdTPnptc6csuWLXNr1qwZUPbSzh7+9Nj/8sRDv+KM7mupb5jJhrd82U9gWUahgzVb4E/b4KDZHSzv+jmxMEVn06tIdr3E/GduoiWcw9rZ/8Q/Zh9mRlcL2aCGLYsPZ8f8N5NONuIs6J94z7m+7x1AGLK5vZdfPbWFHV29LGqqZ/OuXkIchx0wj1fXbGXx335CQ8dfAUjHG+iYsTcz258Fi7HhgA/QNXtfQovjgrjfq+vbO4bDOQfOYbi+8kTvTuZvvIu5m38HwNY9/pHWvd5JqmYOYSzp11CMamku9z6Hc36fRGVBmKFx22Ps+befksh00NWwiI2vOJHOWa9g55zX0Vsz3MjMwsd1scM9HToee34H1/7heWoSAYua6nn6pXb2bW7glDe9nFfOn8Fb95uHWcHRaGUZolboWBUZo7INp9TxKmWg4b+jMO0COjNrBZ4vYdN5wNYRt6qcaq5fNdcNxr9+W51zK8Zxf8CojtWcav+7jzf9vqNXlmMVdut4LWSy/E9Vz/FVrJ5lO16nomkX0JXKzNY456p2REI116+a6wbVX7/dNVV/r2L0+049k+V3VD3H12SpZ7VT5yARERGRSU4BnYiIiMgkp4CuuCsqXYERVHP9qrluUP31211T9fcqRr/v1DNZfkfVc3xNlnpWNfWhExEREZnklKETERERmeQU0ImIiIhMcgroRERERCY5BXQiIiIik9y0C+hWrFjRv9aUHnqMz6MsdKzqUYZH2eh41aMMDxmFaRfQbd06GVZBEdGxKpOLjleRypp2AZ2IiIjIVKOATkRERGSSU0AnIiIiMskpoBMRERGZ5BTQiYiIiExyCuhEpoJsGn73TUh3V7omIiJSAfFKV6AahaFjW2eKVCZLMh5jbkOSILBKV0ukuD/eCPd+HlKd8PbPVLo2IlKE7i9SLgroBglDx7rN7Zyzag0tO7pZ2FTHlWcsY8mCmTrppHqFGf9116bK1kNEitL9RcpJTa6DbOtM9Z1sAC07ujln1Rq2daYqXDORYcSS/mu6s7L1EJGidH+RclJAN0gqk+072XJadnSTymQrVCORkWW7d/lvUgroRKqV7i9STgroBknGYyxsqhtQtrCpjmQ8VqEaiYzszjXrAHAK6ESqlu4vUk5lD+jMLGZmj5vZz6Pn+5jZQ2b2jJndaGbJqLwmer4+en3vvH1cFJWvM7PleeUrorL1ZnbheNR3bkOSK89Y1nfS5fo4zG1IjsfuRcrixc2bAch2bKtwTUSkGN1fpJwmYlDER4GngFnR868C33TOrTaz7wPvB74Xfd3hnHulmZ0UbXeimS0FTgJeDbwMuNfM9o/29V3gMKAFeMTMbnfOrR1LZYPAWLJgJreed7BGIcmksSDRCw4yvV0a6SRSpXR/kXIqa4bOzBYCRwBXRc8NeDvwk2iTlcB7ou+Pjp4Tvf6OaPujgdXOuV7n3F+B9cAbo8d659xzzrkUsDradsyCwGieWcNeTfU0z6zRySZVrznZA0CoeehEqpruL1Iu5W5y/RbwH0AYPZ8LtDnnojkWaAH2ir7fC9gIEL2+M9q+r3zQe4qVD2Fm55rZGjNb09raOtbfSaRsdvdYnW1RIJfpbuGO6AAAIABJREFULVPNRIbStVWkepQtoDOzdwNbnHOP5hcX2NSN8Npoy4cWOneFc26Zc25Zc3PzMLUWqazdPVZrwi4AYqECOpk4uraKVI9ydrc5GDjKzN4F1OL70H0LaDSzeJSFWwi8GG3fAiwCWswsDswGtueV5+S/p1i5yLQSC9MA1LhecA5MzTgiItNJ2TJ0zrmLnHMLnXN74wc1/Mo5dyrwa+C90WZnArdF398ePSd6/VfOOReVnxSNgt0H2A94GHgE2C8aNZuMfsbt5fp9RKpZ4NJ934dpZelERKabSgyI+ySw2swuAR4HfhiV/xC4xszW4zNzJwE45540s5uAtUAG+LBzLgtgZucDdwMx4Grn3JMT+puIVIl4XkCX6u2iNllbwdqIiMhEm5CAzjl3P3B/9P1z+BGqg7fpAY4v8v4vAV8qUH4ncOc4VlVkUorlBXS93Z3UzpxTwdqIiMhE00oRIlNA3GVIOT/bfKq3q8K1ERGRiaaATmQKSJCmg3oA0j0K6EREphsFdCJTQMJl6DQf0GV6NbmwiMh0o4BOZJJzzhEnQ7f59SEzanIVEZl2FNCJTHKZbEiNZeiOMnRpZehERKYdBXQik1xPr1/HtTfwAV02pYBORGS6UUAnMsn1RgFdOpYL6NTkKiIy3SigE5nk0lFAl4n5PnShMnQiItOOAjqRSS6XoQvjUUCnpb9ERKYdBXQik1y61wdwri+g66lkdUREpAIU0IlMcumoidUSPqBzGWXoRESmGwV0IpNcKsrQWTIX0ClDJyIy3cQrXYFqFoaObZ0pUpksyXiMuQ1JgsAqXS2RATJRn7l4ogYAl05VsjoiE0rXaRFPAV0RYehYt7mdc1atoWVHNwub6rjyjGUsWTBTFwupKpmoz1wsnqDHJSCrDJ1MD7pOi/QbVZOrmTWUqyLVZltnqu8iAdCyo5tzVq1hW6eyH1JdMimfoYvF4qRIQEbHqEwPuk6L9CspoDOzt5jZWuCp6PnrzOzystaswlKZbN9FIqdlRzepTLZCNRIpLJtrco1HAV1WgyJketB1WqRfqRm6bwLLgW0Azrn/A95arkpVg2Q8xsKmugFlC5vqSMZjFaqRSGHZlG9iDeJx0pYgUEAn04Su0yL9Sm5ydc5tHFQ0pT8CzW1IcuUZy/ouFrm+GXMbkhWumchAYdTEGg9ipIljWTU3yfSg67RIv1IHRWw0s7cAzsySwEeIml+nqiAwliyYya3nHazRU1LVspk0ALFchi5UQCfTg67TIv1KDeg+CFwG7AW0AL8EPlyuSlWLIDCaZ9ZUuhoiw8pGGbpYLEaGBDE1uco0ouu0iFdSQOec2wqcWua6iMhuyGXoCGJkLE4QpitbIRERmXAlBXRmtg/wr8De+e9xzh1VnmqJSKlcFNA5i5GxBDVOGToRkemm1CbXnwI/BH4GhOWrjoiMVrYvoIuTtQSxsKPCNRIRkYlWakDX45z7dllrIiK7J8xvck0QU5OriMi0U2pAd5mZfQ4/GKKvPcc591hZaiUiJXPZjP9qAWGQIKGVIkREpp1S56H7O+Ac4CvApdHjG8O9wcxqzexhM/s/M3vSzL4Qle9jZg+Z2TNmdmM0DQpmVhM9Xx+9vnfevi6KyteZ2fK88hVR2Xozu3A0v7jIlBH2N7mGFifhlKETEZluSs3QHcP/Z+/+4+Oq6sT/v953JpNMJkmTpukPmgYoItBVVqDyQ1RQQFCUAkoLIi2sgi6wLq674kf9iuLqqruuogguoEL50VJUpArYBQRUrEhZflpEfrfpr6RpfmeS+XHP949zJzOTTJKbZCaZTN7PR+cxmTt37pykd8687znnfQ4sNcaM59J/AHivMaZHRMqAP4jI/cC/AN81xqwXkR8BHweu9+7bjTFvEpFzgW8Bq0RkGXAu8HfAfsCDIvJm7z1+CJyCnUrlCRHZaIzZOo4yKjXjiZtqoQuQdMooQwM6pZSabfy20D0D1I7nwMZKjc4u824GeC/wM2/7LcCZ3s8rvMd4z58kIuJtX2+MGTDGvAa8DBzt3V42xrzqBZrrvX2Vml1SY+bEwUgZoXFddymllCoFflvoFgB/FZEnyB5DN+q0JSISAJ4E3oRtTXsF6DDGJLxdmrGTFePdb/eOmxCRTqDe2/6njMNmvmb7kO3HjFCOS4BLAJqamkYrslLTaiLnqrgJEgRABDdQRogExhjs9ZBShaN1q1LFw29Ad9VEDm6MSQJvE5Fa4G7gsFy7efe5vn3MKNtztS6aHNswxtwA3ACwfPnynPsoVQwmcq6KmyCJtxi5lFEucQYSScrL/H68lZoYrVuVKh6+ulyNMY8CfwWqvdsL3jZfjDEdwCPAsUCtiKS+aRqBnd7PzcASAO/5OcC+zO1DXjPSdqVmlcEWOsA49qPV398/nUVSSik1xXwFdCKyEvgzcA6wEnhcRD4yxmsavJY5RCQMnAy8ADwMpF67BrjH+3mj9xjv+d8aY4y3/VwvC/ZA4GCvLE8AB3tZsyFs4sRGP7+PUqVETJxk6hopUAZArD86jSVSSik11fz2yXwReLsxpgVssAY8SDq5IZdFwC3eODoH2GCM+bWIbAXWi8i/A09hV6DAu79VRF7GtsydC2CM+YuIbAC2AgngMq8rFxG5HNgEBICfGGP+4vP3UapkiJvE9a7NxAvo+gc0oFNKqdnEb0DnpII5TxtjtO4ZY54Fjsix/VVshurQ7f3YFsBcx/o68PUc2+8D7hu15EqVOMckSIo3hk5b6JRSalbyG9D9RkQ2Aeu8x6uA+wtTJKXUeDhuAtf7KDteQDegLXRKKTWr+ArojDH/JiJnA+/EZp3eYIy5u6AlU0r54pgESce20EnAfqRj0b7pLJJSSqkp5iug85IR7jPG/MJ7HBaRA4wxrxeycEqpsTkmiSt2BEQgaFvo4jHNclVKqdnE70oRdwFuxuOkt00pNc0CpLtcA4FUQKddrkopNZv4DeiCmeu4ej+HClMkpdR4BEwC10uKSLXQJXUMnVJKzSp+A7pWERlc5ktEVgB7C1MkpdR4OBkBXdAL6BJx7XJVSqnZxG+W66eA20XkWmxSxHZgdcFKpZTyxRhDwLiYwRY6+5FOxAZGe5lSSqkS4zfL9RXgWBGpAsQY013YYiml/Ei4hqAkcMW2zAXL7L2rY+iUUmpW8ZvlWg58GDgACIoIAMaYqwtWMqXUmOJJlzKSGKkA0vPQJeNeC92DX4VkDE4dNi+3UkqpEuK3y/UeoBN4EtC+HKWKRDxpCJIcHEOH47XQJQbAGPjDf9vtx10GNftNUymVUkoVmt+ArtEYc1pBS6KUGrd40iVAEryAzngtdCbeD+2vp3fcthne8uFpKKFSSqmp4DfL9Y8i8taClkQpNW6JpPG6XG1A5zreNVqiH/b8Jb1j165pKJ1SSqmp4reF7p3AhSLyGrbLVQBjjDm8YCVTSo0pnnQJkiThLf1lnBAughOPQndGENetAZ1SSpUyvwHd+wtaCqXUhMSTLiFJYsT7KItDlAoCiV4bxEkAquZD187pLahSSqmCGjWgE5G53o86TYlSRSjhGiIkBsfQAUSlglCyB7r3QLgOKudpC51SSpW4sVrongQMtot1KAMszXuJlFK+xRK2y9U46YCuX8KEEn02iKuca4O6zu3TWEqllFKFNmpAZ4w5cKoKopQav4RrkyIyW+gGpIKQ2wfdbRCeC6EI9HdOYymVUkoVmq8sVxF5yM82pdTUSiVF4KSvzQakgnLXa6EL19mAbqDLzkunlFKqJI01hq4CiADzRKSOdNdrDaCzlCo1zexKEYmsLteYE2ZOYjdE222Xq1MGbgLiUQhVTmNplVJKFcpYY+g+CVyBDd6eJB3QdQE/LGC5lFI+xBMuZZLdQhcPhGkyO+2ntbIe3KR9or9TAzqllCpRY42huwa4RkT+yRjzgykqk1LKp2QiBoA4AVIdqnGngoB4j8JzId5nf+7vhJpFU19IpZRSBedrHjpjzA9E5B3AAZmvMcasLVC5lFI+JOJx+0NGUkQyEE7vUDkXol7DuiZGKKVUyfKbFHEr8F/YFSPe7t2WF7BcSikfBlvoAhlj6IKR9A6RBpsUARrQKaVKioj0TMN7flFE/iIiz4rI0yJyjLf9ChEZc0yL3/0mwu9KEcuBZcZompxSxcT1Ajok/VFur1iS3qG8Oh3QDXRNYcmUUqq0iMhxwAeBI40xAyIyDwh5T18B3Ab0jXEYv/uNm68WOuB5YOF4DiwiS0TkYRF5wYtm/9nbPldEHhCRl7z7Om+7iMj3ReRlL/I9MuNYa7z9XxKRNRnbjxKR57zXfF9Eck2ArFTJSnpdrk5Glmt75QHZO4Wq7H1/xxSVSimlpoeI7C8iD3lxxEMi0uRt/5CIPC4iT4nIgyKywNv+FRH5iYg8IiKvisinRzn8ImCvMWYAwBiz1xiz03vNfsDDIvKwd9zrRWSLF/981duWa7/BVkYR+YiI3Oz9fI6IPC8iz4jI7/z87n4DunnAVhHZJCIbU7cxXpMAPmuMOQw4FrhMRJYBnwceMsYcDDzkPQa7XuzB3u0S4Hrvl5oLXAUcAxwNXJUKAr19Lsl43Wk+fx9fXNfQ3tvPQMcuku3bMN17wHXz+RZKTUoyObzLNVxVx+vuAra9abXdoF2uajZwXejZAx3b7X0e62rXNbR2D7CjvY/W7gFcVzuriti1wFpjzOHA7cD3ve1/AI41xhwBrAc+l/GaQ4FTSccYZSMc+3+BJSLyNxG5TkROADDGfB/YCbzHGPMeb98vGmOWA4cDJ4jI4SPsN5IvA6caY/4eOMPPL+63y/UrPvcbZIzZBezyfu4WkReAxcAK4ERvt1uAR4Arve1rvW7dP4lIrYgs8vZ9wBizD0BEHgBOE5FHgBpjzGZv+1rgTOD+8ZY1F9c1vNHWQ13PS5TfswY6tkFtE+bcdcj8ZeD4jYWVKhw3PgCAZExbsigCJ8a+y7fqK2gCCITsXHQa0KlS5brQshXWnzdYV3PuOshDXe26hhf3dHPx2i00t0dprAtz4+rlHLKgGsfRTqEidBxwtvfzrcC3vZ8bgTu9uCIEvJbxmnu9VrcBEWkBFgDNQw9sjOkRkaOAdwHv8Y73eWPMzTnKsVJELsHGWYuAZcCz4/g9HgNuFpENwC/8vMDXmW6MeTTXzW+pROQA4AjgcWCBF+ylgr753m6LgcwFJ5u9baNtb86xPS/aemP07NtNbSqYA+jYhqw/D/pa8/U2Sk2KSdouV8nocm2ogKDA650ZLRS6/JcqZX2t6WAO7H2e6uq23thgMAfQ3B7l4rVbaOuNTfrYakqkmlN/AFxrjHkrdo7diox9BjJ+TjJKY5cxJmmMecQYcxVwOfDhofuIyIHAvwIneS2F9w55v1zlI3MfY8yngC8BS4CnRaR+pDKl+M1yPVZEnhCRHhGJiUhSRHyNsBaRKuDnwBXGmNFek+tSx0xge64yXOL1ZW9pbfX3AY8lktSG3HQFkdKxDRL6QVaFMd5zNZnwxtAF0vVPwIGFERvQPbk7wUDSaECnCmIidWtBJGIFq6tjieRgMJfS3B4llkhO+tiqIP4InOv9fD62qxVgDrDD+3nN0Bf5ISKHiMjBGZveBrzh/dwNVHs/1wC9QKc3Vu/9Ga/J3A9gj4gcJiIOcFbGex1kjHncGPNlYC82sBuV37boa4HzgJeAMPAJb9uovH7onwO3G2NSTYZ7vCZPvPsWb3vzkAI3YvuaR9vemGP7MMaYG4wxy40xyxsaGsYqNgChYICOmGOb7jPVNkEwlPtFSk3SeM9VNzE8KQJgYSU89EaCD9/Tx6kbeogFItCvWa4qvyZStxZEMFSwujoUDNBYF87a1lgXJhQMjPAKNYUqRaQ54/YvwKeBi0TkWeAC4J+9fb8C3CUiv8cGSBNRBdwiIlu94y8jPSTtBuB+EXnYGPMM8BTwF+An2O5Thu7nPf488Gvgt3jD1Dz/6SV9Pg/8DnhmrML5HUOHMeZlEQkYY5LAT0Xkj6Pt72Wc/hh4wRjz3xlPbcRGx9/07u/J2H65iKzHJkB0GmN2icgm4BsZiRDvA/6fMWafiHSLyLHYrtzV2CbVvKiPhOiZu5COFbeku11TY+gqp7HiUiqTlxTBkIBuUQS2tEDIgde7DHsrKtlPs1xVqapssGPmho6hy0NdXR8JcePq5cPG0NVH9MJ+uhljRmqUem+Ofe8hHW9kbv/KkMdvGeX9ngTeMcJzPyAjBjHGXOhzv58BP8ux39lDt43Fb0DXJyIhbD/ut7FRZGSM1xyPjY6fE5GnvW1fwAZyG0Tk48A24BzvufuADwAvY+dnuQjAC9y+Bjzh7Xd1KkEC+EfgZmyr4f3kKSECwHGE/eur6KxcxsCF/0vQxHGC5UikQRMiVNFIdbkODej286atfE8j/LYZOk2Y/fp17KcqUY5jEyA+8aDtZg2GbDCXh7racYRDFlRz96XHE0skCQUD1EdCmhChio7fgO4CbPfs5cBnsF2gwwYCZjLG/IHc49wATsqxvwEuG+FYP8E2Ww7dvgUYMZqeLMcR6iIV2AQVpYqQlxRhJPujvHSO/fCdsgSea4O2ZCVEtYVOlTDHgaoFBTq00FBdXpBjq+LjJSA8lOOpk4wxbVNdHr/8BnR7gZgxph/4qogEAD27lZpmbiqgG9JC95Z6WHsKzK2AhjC0xsLgdk9HEZVSakbxgra3TXc5xstve/RDQObaY2HgwfwXRyk1Lt4YuqEtdGCDOYAFlbA7FobEgGZoK6VUifIb0FUYYwaXp/B+LsjiskqpcRihhS7TgjDsjnsfV13PVSmlSpLfgK53yNqqRwHRUfZXSk0B4ybsDzJyQDc/DD3Gm3ZBAzqllCpJfsfQXYGdvyU1z9siYFVhiqSU8m0wKWKUgK4SnsML6HQuOqWUmlIichpwDRAAbjLGfLMQ7+MroDPGPCEihwKHYJPn/mqMiReiQEqpcRjsch35ozw/DN1ol6tSSk01L4n0h8Ap2AURnhCRjcaYrfl+r1EDOhF5rzHmtyIydIK7g0WEjNUflFLTwR27hW5uBXQPdrlqpqtSSuUykEget7c79p2E6y4KOs6uedWhz5YHA5snedijgZeNMa8CeIsnrACmNqADTsAuR/GhHM8ZQAM6paZT0o6hGy0pIuiAW+a10GmXq1JKDTOQSB73tz09G//xtifneSuCHHD9x47a+OYFVWdMMqhbDGzPeNyMXQ0r70YN6IwxV3n3FxXizZVSk2N8tNABBEKVkEC7XJVSKoe93bHvpII5gOb2KP9425Pz7rzkuO8srgvnXO7Lp1wLLJhJHG9EY3W5/stozw9Zo1UpNcUkNQ/dKGPoAMpCYQ3olFJqBAnXXZQK5lKa26MkXHeyS0U1Y1fXSmkEdo6w76SMNW1J9Rg3pdR08pIiyDGxcKaqijIGKNMuV6WUyiHoOLsa68JZ2xrrwgQdZ9ckD/0ENu/gQBEJAecCGyd5zJzG6nL9aiHeVCmVJ948dGN1udaV28SIcm2hU0qpYeZVhz57/ceOyhxDx/UfO2rvvOrQZydzXGNMQkQuBzZhpy35iTHmL3kp9BC+pi0RkaXYOVSOxfb9bgY+k8raUEpND3HjJHFAcg3TSKv1ArraaJfvySeVUmq2KA8GNr95QdUZd15yXL6zXDHG3Afcl4dijspv3X4Hdh6Vs7zH5wLrKFCmhlLKH3ETJBm9dQ68gI5K4r0dGtAppVQO5cHA5kkmQEwrv0t/iTHmVmNMwrvdRoGyNJRS/ombIDnG+DlItdBVkox2TkGplFJKTTW/F+sPi8jngfXYQG4VcK+IzAUwxuwrUPmUUqNwTAJ3lDnoUurKoYcwpl8DOqWUKkV+A7rUuq2fHLL9H7AB3tK8lUgp5Ztj4r66XOvKoZVKnFjzFJRKKaXUVPO7luuBhS6IUmp8jDEE3ASujy7XOeXwqgkTjOvSX0opVYr8ZrlWAJcC78S2yP0e+JExpr+AZVNKjSKWdAlK0ldAV+ZAzAlTluwD1wXH7/BZpZRSM4HfWn0t8HfAD4BrgWXArYUqlFJqbLGESxlJ3DHmoEuJByM4GF0tQimlppCI/EREWkTk+UK+j98xdIcYY/4+4/HDIvJMIQqklPInlnAJjiOgS5RVwwAQ3Qfh2sIWTimlVMrN2MawtYV8E78tdE+JyLGpByJyDPBYYYqklPIjlnQJkhhzlYgUt6zK/tDXXsBSKaXUDJUYOI6O7X9k32uv0bH9jyQGjsvHYY0xvwMKPhuI3xa6Y4DVIrLNe9wEvCAizwHGGHN4QUpXpFzX0NYbI5ZIEgoGqI+EcJzRZ+pXKt9SXa5+AzrKq6AH20KnlCpJ+v00QYmB42h5YSMbLphHxzaobTqAlbduZP5hZxAsn/RqEVPBb0B3WkFLMYO4ruHFPd1cvHYLqfXebly9nEMWVOf80OiHSxVKqsvVOP4+xoGKavu67lZChSyYUmpEhfxOGO/3k8rQ0/KdwWAOoGMbbLhgHhfe9x1ql8yI1SN8dbkaY94wxrwBRLFZrsZuHtw+a7T1xgY/LADN7VEuXruFtt7YsH1TH66zrnuM47/1MGdd9xgv7unGdXWRDTV5AwmXMvHf5VoWtl2ufR0thSyWUmoEhf5OGM/3kxrCTSwaDOZSOrbZ7TOEr4BORM4QkZeA14BHgdeB+8d4zbCsDhGZKyIPiMhL3n2dt11E5Psi8rKIPCsiR2a8Zo23/0sisiZj+1Ei8pz3mu+LjLE6eZ7EEsnBD0tKc3uUWCI5bF/9cKlCiiXH1+VaGY6QNEJ/194Cl0wplUuhvxPG8/2khnCCu6htyt5W22S3zxB+kyK+BhwL/M2bZPgkxk6KuJnhXbWfBx4yxhwMPOQ9Bng/cLB3uwS4HmwACFyFHcN3NHBVKgj09rkk43VT0i0cCgZorAtnbWusCxMKDv9S1Q+XKiTb5ZoAn12utRUOnUSId7cVuGRKqVwK/Z0wnu8nNUTV/M+y8ta9g0FdbROsvHUvVfM/O9lDi8g6YDNwiIg0i8jHJ3vMXPwGdHFjTBvgiIhjjHkYeNtoLxghq2MFcIv38y3AmRnb1xrrT0CtiCwCTgUeMMbsM8a0Aw8Ap3nP1RhjNhtjDDYV+EymQH0kxI2rlw9+aFJjFOojw0cl6YdLFVJqDB0+1nIFqC2HdlON26cBnVLTodDfCeP5flJDBMs3M/+wM7jwvs18+unXufC+zflKiDDGnGeMWWSMKTPGNBpjfpyPIg/lNymiQ0SqsCtE3C4iLUBiAu+3wBizC8AYs0tE5nvbFwPbM/Zr9raNtr05x/aCcxzhkAXV3H3p8WMOak19uIYOUNUPl8qHVJYrPrtca0PQQRXzozptiVLTodDfCeP5flI5BMs3z5QEiFz8BnQrgH7gCuB8YA5wdR7LketsMxPYnvvgIpdgu2dpamoaabfhXBf6WiERg2AIKhvAcXAcoaG6fMyX64dLjdd4zlU7D53/FrqyAHRIFYv6NaBT+THhunUqjVCPT4ep+E7w+/2kSo/fLNdeYB52nFobsN7rgh2vPV53Kd59Kt2uGViSsV8jsHOM7Y05to9U/huMMcuNMcsbGhr8ldR1oWUr3HQyfO8t9r5lq90+DqkP1+K6ShqqyzWYU6Maz7kaS9i1XP2OoQPoD1RTHu+YbDGVAiZYt06lPNXj+aTfCapQ/Ga5rgT+DJwDrAQeF5GPTOD9NgKpTNU1wD0Z21d72a7HAp1e1+wm4H0iUuclQ7wP2OQ91y0ix3rZraszjpUffa2w/jyy5qRZf57drlQRsF2uCcRnCx1AsqyayqSu5apmCa3H1Szi99L+i8DbjTEtACLSADwI/GykF3hZHScC80SkGZut+k1gg5fhsQ0bIALcB3wAeBnoAy4CMMbsE5GvAU94+11tjEklWvwjNpM2jJ1CZdRpVMYtESPnnDTxqL26SzXZF1FzvppdBpI2oEuOI6CjvIrwwADuQB9OeWXhCqdUMRipHk9404QMrb/D9RBt0/pczUh+AzonFcx52hijdc8Yc94IT52UY18DXDbCcX4C/CTH9i3AW0Yrw6QEQzZtObMyqG2CvX+DgR6Yv8xua9mavgKsbYJVt0PEqwS0MlAFlMpylXF0uQYrqqAL9rbuYn7jQQUsnVJFYKR6PBhKd8dm1t8rb4VHvw0v3msfn7vO1vWFrse1YUDlgd8z5jcisklELhSRC4F7sa1qpauywX6YM+ekOeNaePRbsP48TG9r7ub8O8+HHVuKYqyGKm2pLNfxdLkGI3Yax727ZtUCL2q2ylWPn7vObs9Vf2+4AN52XvrxOLpnXdfQ2j3AjvY+WrsH/K/+UITj/NTMNOqlvYi8CTvVyL+JyNnAO7EZppuB26egfNPHceyV2YX3QccbEG2H5zbA8VdAuA6THMA1EMjVnB+uS1cGn3gQqhZMz++gSlosniREHAmW+X5NZfVcADr3aECnZoFUPf6JB4e3fo3UHVu9MPtxYuxVHCa1hupI4/z0u0ON01gtdN8DugGMMb8wxvyLMeYz2Na57xW6cNPOcaAsDPE++yE//gp4eh3cfDrOzadjkgl7xde4HFbdBhfeC+ffBca7svJZGSg1EYlEjIAYcPzPYVVVYwO6gX3bx9hTqRLhODYwql1i772uTBMIkXOpp4paW6enHgfH+Hy5LsnuPdTGdnPNB/fjiCU141vSa6xxfn65LvTsgY7t9l5b+GadsQbfHGCMeXboRmPMFhE5oCAlKiJuMgldu3Du/Wx6jMUZ10LvHmjeQvCBL5H46M8I9rbAPZem9znzR7ZC6GkZuzLI+cY6nkKNzY3bJYSM47+Fzqmopp8y3I4dhSqWUgXnuoa23tik5nLrcOZQfc5tBO/6WHb9/sBV9uJ90xfS3bMjFwRatlK2/jwWdWxjUW0TN33oFj7xG3hqe5e/Jb1GG+fnV67xgFM1/k8VjbH+pytGeS7xLKY4AAAgAElEQVQ8ynMznusa+jv34Nz50eym8I2X2w87wIv34gbD6WAutc8vPwUnXDl2ZZD7jXU8hfLFjfUDYAL+AzpE2OfU43SPOG2jUkUt1b151nWPcfy3Huas6x7jxT3d/sesefpiLm3UwKnfsL0rp34Dfnu1TYhY8Bbb5TlWQJSju7T+V2v40okN/pf0Gm2cn1+9I3Tb9ur0LLPJWC10T4jIxcaYGzM3etOOPFm4Yk2/tt4Yie4eKqvm2w96uM6Oo3vse/ZngNom2+2aq7l83pthTtP4r450PIXyySRsQOeOo4UOoK9sLpHoHuJJl7KAXr2rmaWtNzY4Vg0Y7N68+9Ljx7VCQigYYHdPggWbvgBV8+2F+slftUNsyqshMm/sg4zQXbqoyvG/pNdo4/z8SkRH6LaN+j+GmvHGCuiuAO4WkfNJB3DLgRBwViELNt1iiSTdiTIWnfSV7O7UFddBoBxqm+hYcQu7O10OzdVcXhaeWFN3vsZTqJLnxgeA8XW5AiTK57Iw+iKv7e3lzQuqC1E0pQomlkgOBnMpze1Rf92bGcNZ5gVCxBsW0X3OBqpje7Pr+XPXQXju2HX4CN2lDbU1LKz2kRCRkhrnN1ESyN1t63OdZ1UaxppLbo8x5h3AV4HXvdtXjTHHGWN2F75406cy5HDAnAAEymwLXeNy+2H50/VQvQCz+lfMqZvHIQtrMKtun1xzeabgCAN1JzIWT5W2wRa68Z0bwchcFsg+XtipS4CpmScUDNBYlz3ix1f35pDhLPLjk1kUe51wVV32sJmq+RDrhq5maH8duockGGQmH0ggZ3dpWfV8HMzUJSmUhW1jQ2Y5Vlxnt6tZw9eMpMaYh4GHC1yW4pFMUNv1N+TO87MHyz63Ad66Em4+HclssYvMgzW/BnEmlcTguoZOmUPNqjsIpMbuTTZAVCXLxL0xdONsoauonktIkmzfvg2OWDL2C5QqIvWREDeuXj5sipCRujdTCRQ1yX2UDxnOIuvPI7h6Y3pb43I47du22/Xm04cnGMDw5IOP3Q0ffxCSGd2lufYrZJJCeC5UL4LTvwNlld7MDIvsdh/ykWSipp//KeZnC9eFzu3pYA7SyRBrfgW3fCh7+z2XwpnXg9OBW7MEN9aPSbYQqGrACfhv7s6cx6ihqoyr3/8LDpkXoixUgUQ0y1UNJ8mJdbm64XoA9u54BTg+38VSquAW1JRz5yXHkjRQUeYwLzJkkXuva9W4LsZNID0DUFFmW98yuyU7ttlppg453U4oPP8w2PcqpGY2SO2TGscMw8c433aWfa424+KoZ4/vsdB5CaYcB+YuhYrqcY/Dm9QceqqoaEA3VF+r10yeYxzbSAkQNY3Q34lz8/txvKuxxMo7YMEy30Fd5kDf5vYoH/ppF411YTvQdyqCOZ0qZeZJpFroxvcxHogsAiC+50Vc12ilrWaMkYKPeZHyzJ1s69jD30CO+SSBjZczL7NH5dn18ObT0sltoQic8Dm7SsSZ19sWrtHGMXdssy153iTzRNsxrkvWp8jnWOi8BlMTHIeXryQTNf30G3uoRMymeg8dx3bI6TbQufxJuPRxOOJjdnttE+DaJcEyrsaCGz5Kssd/yvikBvpOlk6VMiOlWujGO4YuFl5AQoIsSbzOa229hSiaUgWRCj4aqsr4+QUHsW7lYhJde+iMDqTHtnXthIe/Aad8FYLlw8dAL/+4nWPuwats12Ssx96q5tuZDOJ9uccxB0IgAhc/bOca9SaZZ9MXML2tdt7SVBlM0k4yn5qguHG5N+l8Mms83UjBlK8JifNkWr97VF5pC91QwZD9oJ5xre1m7dhmg7kTPpc9puKctVAxFw4+GTZ9CY6+2M5flNKxDUn6/1CmBvpmfrB8z2M0WTpVyozkxicwDx2AE6A3vJg3J5r5vzfaOaihqgClUyr/YokkDVVl3HRahPpfnQMd21hS24S76g6Ihm335+qNcNxlcNvZ2WOgf3u17Va9a40N3t775XQdnzlO+qh/sC15GVmv5tw7kIFue/wck8w7d55P4uO/xendkz1uLtUiePi5w7No5y8rimBqWr97VF5pC91QlQ1wytVQUQNr7oXLn4DT/sM2x2cGPHethmMuhlClrSTmHph9nNomXKfM9yLNqYG+qeytsQb65pVOlTIjmdjEkiIATM1iDnWaeWq7Zrqq6ed3YftQMMDVJy+k/ldrsupj586PQvurdlugHOK9tvt01W02eEtNCF+z2O5z/BXpYM47Bhsvt12xv/kcBIK4a+5l4LKneGXFL0lGFqaDucz9U5PMd2wjkIgOvzC+51LMe75k54PLLM/686CvdeIZu3k0rd89Kq+0hS6XRD9sWJ2+msrMgkrp2Gab9n9yqt1n1W22Je/Fe6G2ie6z1vJqVzmXrXvM15gIxxEOWVDN3ZceP/WZRvlYekZNOXeCEwsDDFQtYT/5I399fQfw1jyXTCn/XNfwRlsPPft2Uxty2R1z6Jm7kP3rq4bVf/WVQernl9vgKDXRe/MWW3eVVdquzb7WdFJDZutc7f6YcC3yD5vs2LdcdXq4DnpacANhvvDgXtY/uYsjltTwi3MZeX+A2iYk0T98n6r50NuSuzyJGPU148vYLYRp/e5ReaUB3VBDux8PfLe9zxXwpJZV6dgGd37MLh9z8lXQ28qORC39YsY1wNRxZHoGoaaWnhmaYq9TpRQtY4ydtiQ4sRa6gUgjANL6Aj0DJ1NVrlWBmh6d0QHqel7iwPvXDHahdqy4hc7KZdRFMlafdF2c1hey66lUcNTTYqfpOOOHcMc5w1vS/uF/IdGPdO2EeNRONZWrTp+zBC74JY4E+MbJDhceWUeZ4yL7Xsm9f7Q9XY6uHcP3OeFK5M6PDS/P6d+BYKhogqlp++5ReaVdrkMN7X5857/Ai5tsC1zmpI0rb4WWF9L7dWyzt9s/Am6S+kqhIxoH7JiIaCzhu/t1ymUuPXPF8/7WMFTTaiDhUmZsl7gZZ1IEQLTmAAAOl1d4Vrtd1TSqjLdTe092F2rtPWuojLdn75hrrO/Gy+262R/dYLcFQyO3kq09A/73i+DGbYLD0Dp91e2AwL5X4Bcfx7nldA51mllaIzbp7Yxrh+x/G8xpTK8B++cb7TEy9knWLc3dsjf3oMEL5lQwtbiukobqcm0ZUxOml+WZXBfcRPoqq3E5BCvgoBOgrw3O/xk4QXv7801w1Gp4aZNt8k9drXmVTN2a+/jRI68AdkzCC7u7+dqvtxbv/D6TXXpGTamegQTl2AuGiXS5Jirq6a+Yz7HJF3hqewfveJOPdSuVKoCAG88Z9ATcePa2kcb61r8JBrrhjpXwoe+n55QL10EyDnX7w9oVwxMhjvu0nRDeJG2d/viNsPn72S1/91yKnP9z2wL426vT63rH+6CyHn76/nRr4QevgYZDiV30AHs7utjV4xLpIPfSkKEqvWBWeadnVKa+Vtj0xfSV2Pv/067+0LsXfvmP8MOj4dYz7YfzzafYpvvjr0hXAI99zx6nYxs9/TGe2t5BY12Yb334cH70yCvZKemZy8cUelkYVXK6+xOEJUaSADgTG0AdnXsYxwX+ylOvt+W5dEr55zplNghbdZsdtuKNR866UMm82M5U22SDsVS3ZvVCOPXfIdJgL7D/fCO4yeGJEI3L7QwFt3wQvn+EncHgrWfbyeMzkyg6tkG8l+TK22xQd+fH7HdB9X64kYXEL3qAxKefI37RA7gNh0EgSLB6AZ2hhfzzr3fy/zbtomPFLVmtdu6qO2z5lMozbaHLlIjZpIbePTbDtb/DXuH96fr0lVm03T5+39dsRbLobXDRJnjk67alDqC2iUhlhN/924m8sLub/9r0Ik9t7+CIJbV86sSDSCQTuHv+ZjOzpmJZGFVyevoTRIgSD0x8rcbeusNo3PkoXduewZijESmyVmM1K/SH6ig74XNIaiaB2ibMylvpD9UxOJgg82I7c6qRVbeDMekgTRxoe9kmSATL4V2ftRfMtU3ZiRC5slw3rLZj2977ZTt9yfzD4B82QVmEREU9zscfRIwLJokxLv1dLXzm181s2tqaNSEw2JUs1l9yLP3xJG/0x9n9wbuJBJK09BmWVDYxX+t5VQAa0GVKZXs2b4HkANx5Plx4HxzzyeHzFQXL7XxGPS12rqHjPg3H/RN0boM5SwiSZGGgizte3zcYzP3rqYdw5c+f5ZoP7seiBz6q876pCeseiFMl/SQDFWPvPIK+Ors25eGxp3mltZc3zdf56NTUq052pIM5sHN4briA6o8/ACy021wXjv+0nXbkwnttq1vPHtsiF+2wgVfVAuhszs4oXXGdXQli5dr0hPGp7NRc3be1+8P/3WYnH741PedcaNXtuFULcXpbkTvPRzq2UVnbxDc+dAst3TU8tb2Li9duYePlx7OnayAra/VbHz6cr2zaM9hjc/elQ6a4UipP9DIhgxueZ5fsqm0CCdgPsxvPPV+RMel0+Xsuhc43bNfXlpsh2oHctZrQze/jX49w+dS7DuBTJx7ElT9/lub2KPMrRed9U5PS3Z8gwuQCunh4Ht2VSzjJeYotr+/LY+mUGodc0310bIPEgLfywy4bjP3iYruSzc2n296Tuv2hexfc/mE7fZRx05P3po5xz6VQMccuUj//sPT0UpXzbBC46rb0ag61TdDxBrzlLPjdf2UHmHeeTyC6b9ga3/W/WsOXTrTdpzb5LTls5Ycrf/4snzrxIJ3fTRWcBnQZ2voSXPZglO6P/QYTCNoPvnFHWNc1lq4IUld8fW12MO7dlwyOvwhu+CifO34Ohy+uHvyQt/SZ3GNBdN435VOqy9UNTjygA4jOP5K3B/7K86+8np+CKTVe4uRealEE9r0G/Z3wyDeHd4/2d6fHzsHIdbU49rmffgD+tglOvNJObfKTU+0SYO/9sn2/M6612ax3rbH1+NDjOIGcx59faYcqNNaFSRqTc+WHwxbaqUmKMiFOlQwN6DLEEkk2bW0l4AgiDpz6ddj3au7ga98r6VnCU+MzelvTTfmpCSc7tuF0NbMg+iqnLrNXcndt7cM959bh06CE66foN1UzXc9AgirpxwQnPoYOoGf+UQRxKX/twTyVTKlxCoTsUoqp+tBbalFuPt0GXXecY4e9pC6gwQvUcvR05KqrMTao69hmW+lyzQt30pdtFmuq12Vo0kJtk+3mzXH8lj4z2PpWUZZ75YdwKKhTkqiC04AuQ2oZlnIzYK/o8OYfWrk2O/hKXcmF69LjNIKVdg3Y1EST0fb0/r2tOHd+lO9+sJHGujDnLKvE+d23baLFhffa+0e/DVHNNlT+dPfHiRBFJhnQRWuW0hms513RR9jT1Z+n0ik1Dsk4/O4/0/XhSV8evtRi5jJbkDvAivUMnyvujGvt8JnuncMTI1I6tkHf3qykNqoWDD/O5h8MO7676g6WNDYNtr7Ni5TrMlpq2sz4pAgROQ24BggANxljvjnRY6XWtHOlnUBZpR3b0dNi5zg6/Ts2cyranjEz+UJbCT27Ht79OTjlq3bMxzm3wP3/lj2fUcc2wk6Suy89nrr4brjrXptRm+n935rU30LNHt2pFrqyyXW5Ig47G97Nu3f+koe3bmXBsUfmp4BK+WZsXZiqDy+8N3fQlWo1ywywVlyXHjcX64XH/yd7RoLH/8fWqxKED99ku29zzQsX78s+NsYep2YxhGth05ds+fraMRfdjxgDJokTDDM/UpE1O0ExrPygZicxpghXLvBJRALA34BTgGbgCeA8Y8zWkV6zfPlys2XLlhGP6boGt3cvQXfATio80G2vIANBGOix3QPxPlthlIVtU36s1wZyva3w9DrMiVcilfW2EjFJm1ghDkYcQEAcZPN1dmJiJ2CnPwlUYNwYGMAkQBziTgXRYA3VFeU4GJu6n4jZsXaVDeOa4sR1DW29seGVjOumjxsI2fLEoxN6jyFvOKny5v04vt9uhL/T6ApSW492rn7x7ue48plTiS95J7sPXTOp93F6Wzjsj1fw8KKP855P/vekjqWKXsEii7Hq1hH17IGbTk4HWatus2PbUlORnHSVDazEAafM1qnJOATKbE+KE7SZrLEem0jRtxfi/bZ+rt7P1hfJuN1PgtDXku52Ta34ULXQzmzgJgAwoSqI9SJO0L5vMmbrxrIIpmtHOjmitgmz6nY7d50xYFyMm0QcB5GALasEbBmG1F2ua+iMDlAZbydo4jjBciQVtPa2QiJqX1sWtkkdMGpdmKq7jJtkrtNLIBkF1yURqCBQ1YATGOd8lckE9OxO/62rFtrvwfQbjl03T77+1kh4HGZ6C93RwMvGmFcBRGQ9sAIYMaAbi+MIVM7FDHQhndvtYNzjLoPNPxw+fclHbrYn54bVWVOayCPfgqMvthVI6uqxtgk541p7xfi+q+GtH4bbzk6/buVaJFAO61alU+VXXIcTWcDe+GIaoq8hQ9da9TlvnesaXtzTPWwB6EPmR4avjbjiOnjoK7YFcqJz47kutGwdvjbseI+Vr+P4frsR/k5FOJC5pz9OJVHaJ5kUAeBG5vPnwBEctXsD9H8FKmomX0ClfEoE5xBYeWt66pKn19kxxY9+284jF+9LT+he2wQf+aldaeG2szLqT2//3j1w0lfgV5+2QdZJX8mqg+1Y5Trb4xKug4o6eOlBaDoG7krX47JyLTxxE7z2u+w6cfU9wzJd5c7z4YK7ob8L7lqDZE5v9fj/2O+Nx/8H3vOFwbrLdQ1vtPVQ1/MS5allz2qbMB+7G0lEYf1Hs+vk6kX2/TJ/54y6MFV3fe+Bv/Kdk2sIDrQO/t5ltU0kV96Bu2CZ/6AumYA9z6e7vlN/uwVvsUGdn7p5iutvNfPH0C0Gtmc8bva2TUpbXwIT67Xz0L3tPPjlp+z90OlLom3pYC61bePldt+a/Yan0Keec93hY0Q2rLYB4JCU+2Dn68x196WDudRz68+zVz5+fp/e2LBU+ovXbiHZk2NtxHsuTc+QPo73yJJrzcWJHCtfx/FppL9TW2/xTSfTH+0jiIs7iYmFMz0x72xqTDf9f/xRXo6nlF+B/lbk0YwxxW87D577OZz2H7aOHVqP/uwiO99nVv15gX3d8Vek98/8OXO/xADcfg5077YB0kEnpoO5wf1W23lFh9aJPS25u4N799qWwVz1feo+o+5q643Rs2/3sDVspf3VdDCXOs49l0L7q/Y2Ql2YqrsuOaqGqt5tw37vwIaP2vrer57dOb6jLrDbwV/dPMX1t5r5LXS5mk2G9SGLyCXAJQBNTU3DXjBULJFETDx7Aspcg2nLKkce6zFCijvhutzZWamsraHbyipxTGJS89bFEsmcqfSSHBi5jON8jywjrbk43mPl6zg+jfR3iiWSBXm/XPyeq+Il0CTL8jMZ8LzFB/HgziN49+YfwDs+pa10ypfx1q05uYnsMXQpR60euY4tqxy+LbPegpETIFL1bOr5kerq1JJ6mcfOnJw4xUt8G9xnaJky7726K5ZIUhvKMc2K3983tT3jeINznEruY0hyHPVmMvf6uiS99XX91M1TXH+rmd9C1wwsyXjcCOwcupMx5gZjzHJjzPKGhrHX0AsFAxinLJ2tmnmfKd6XO02+sh66duZ+Ltpux1rkem7oeEZvsK4rwUnNW5fK3s3UWBfGBMpHLuM43yNLasWNCZY378fxaaS/Uyg4sbVSJ8LvuVoetVe58fI5eXnfQ+vgR+bDhOJdtntIKR/GW7fm5IxQvxkzch2bSmLI3BZtz66nc9XZmfVs6vkRpiPBTWYfG2x38KrbhmfAPr1u5DJl3nt1VygYoCPm+P9OifflPn7G8RrrwnaO0xGOYQLjqDcDZbnLEfDW1/VTN09x/a1mfkD3BHCwiBwoIiHgXGDjZA9aHwnRVz7PDnZ9eh2c+SN7PzQlPlwPZ92QvW3VbfDkWnjoq3bsQ64Pvogdj5A1D93a7MWnvbETiTkHsM+Zizl3XfZz566zA0x9/j65UukDVQ32OEPek8e+N+73yFKZ47gTOVa+juPTSH+nYptywBiD6WkBIBGqzcsxgw449Ut5TI6CzdfabEClpkCyogEzpD40K2+zCQVVC4bXox/5KcxpGj6P59PrbN2V2j/z58z9MNnPP3V79jx4qfp48w+G14kn/Bu0vgyr77ErTZz6DXsBdNxldvWJXPV96j6j7qqPhKiau5COFbdk/951S+HcO4bXyXVL7W2EujBVd93wZBc9kaZhv3dy5R22vveramGO76hb7XbwVzdPcf2tZniWK4CIfAD4Hnbakp8YY74+2v5+M7Fc1zAQG6Aits/LXjK2g9e43s+ODcwkYLe5XhZVsNxmibpJCIazslwRr6VHAhAI2EwsN5F+XTLuvZdmueb9OL7frvizXFu6+/nvb36Rb5bdxIvv+gGJivxMSH3v6/C751/l1+Vfgvd8EU74XF6Oq4pK8WW5Aon+fgL9rYP1YayinrJYF5IaQePGbZ3qlNl6N1hhW6JS9WdFrV0OzE3Y54xrM1ODFfZ1qfo5FLFj6FLHK6v0jhGw+7vJdH0c67X34tg60QlggmE7VMUJZu+f8V2gWa6Mb5/RFVc2WpGb6WPoMMbcB9yX7+M6jhCuqICK/fJ9aN9C3s0Se7U6QY4jNFSX53piUscd5Q3zc9xClW/Etxvh71REXtrTQwMdACRD+elyBThhMfx461K2Vr6dZZuvtdl5Ffk7vlIjCVZUQEV69Ew5QEWOcWNZhlzIVFTnu1jD+Iku/EYgjiPURSqARcOfrB6hzhulLsyuu9J/uzKf5RkmEIQ5jSM/76dunuL6e7ab6V2uSs06L+7uZqG0Ey+rxjj5uyarKoN37wf/X9eZ0N+Ju/ZMdv366+zYvAG3Y0fe3kcppVT+aUCn1AzzUks3bwlsIxaZ9Aw9w3zkTfCScyBXupeyfccOFm35Nos3XYzzvWVErzsRnrkzPVh8iBd3d/PdB/7GnU9sI5Zw8142pZRSI5vxXa5KzTYv7+rkUHmD7pqT837sxir41jvgF6+8k+8E38nfz4lS1ruTjm3Pc/qeP3DQ3ZfYweKnfxeWvH3wdbf+6Q2+uvEvJFw75mndn7fzw/OPZHFtfubJGyoaS9I9EKehqhwZOt1PkXBdw66ufuojISrK8p8p3TuQoD+epL6quIcIKKWmhgZ0Ss0gxhiSLS9QLjFaapYW5D2aquGKt6UehYGD2N10EGs2f4jlscf5xr47qPzxyXDkaqLv+iLf/kMbP33sdY5squWT7z6Iv+zs4sbfv8rp1/ye/17197z30PyMoTHG8PT2Du7741Ps3foowWQf7ZVLOeb497Lq2KXUJDuh/Q1oPCov7zdRrd0D3Lr5dW57fBv7emOUBx1OWbaAT7xrKW9bMrms5KRreOiFPfz4D6/x+Gv7AFi2qIaPHtPE2UcupjJkq3RjTNEGukqpwpjxWa7jNZlMLKVGMGVZrsYYXtnTxY7nf8/8YB9uWaQQb51T5wB87QnY1h7l8xV3cx73Yww8a5ZiqvejqWGOzfhDiMbivLi7i56BBLXhIJVlAcoCqTEexstetNnjAoOPZbA+svuIt0/SdekbiLNfopkmJ3um+R5TwYumiYMDu4kFwvzHm+6w2ZA++Y17UkUzGLx/dgoZ77mE67KjPcpfdnaRdA1H7l/H4Y1z2NEe5bGX99IbS7J0XoSlDRHCoSChgDN4LHvc9PFS75e5rSsa57kdnXT0xZlXFeKENzcQCjj86bV9vLa3l0h5gIPnV7Onq58rTzuUM4/I2SVflFmuSo1Ar0rGYdYFdCLSCrzhY9d5wN4CF2cyirl8xVw2yH/59hpjTsvj8YBxnaspxf53zzf9fcevIOcqTOh8zWWm/J9qOfNrpHIW7HwtRbMuoPNLRLYYY5ZPdzlGUszlK+ayQfGXb6JK9fcaif6+pWem/I5azvyaKeUsdprlqpRSSik1w2lAp5RSSik1w2lAN7IbprsAYyjm8hVz2aD4yzdRpfp7jUR/39IzU35HLWd+zZRyFjUdQ6eUUkopNcNpC51SSiml1AynAZ1SSiml1AynAZ1SSiml1AynAZ1SSiml1Aw36wK60047zVtzSG96y9utIPRc1VsBbgWj56veCnBT4zDrArq9e2fCKihK6bmqZhY9X5WaXrMuoFNKKaWUKjUa0CmllFJKzXAa0CmllFJKzXAa0CmllFJKzXAa0CmllFJKzXDB6S6AKlGuC32tkIhBMASVDeDo9UNB6N9aqeHuvxJCETjpy9NdEqWmhAZ0Kv9cF1q2wvrzoGMb1DbBuetg/jINNPJN/9ZKDZeIwf/dAsbAOz8D5dXTXSKlCk5rfJV/fa3pAAPs/frz7HaVX/q3Vmq4HU9CPAqJfti6cbpLo9SU0IBO5V8ilg4wUjq22e0qv/RvrdRwr//B3lfOg7/cPb1lUWqKaECn8i8Ysl1/mWqb7HaVX/q3Vmq4138Pc5fC/MOg7eXpLo1SU0IDOpV/lQ12HFcq0EiN66psmN5ylSL9Wys13N6XoO4A20LXvdOOpVOqxGlShMo/x7GD8j/xoGZeFpr+rZXK5rrQ2wKVx0NFLSQGoG8fROqnu2RKFZQGdKowHAeqFkx3KWYH/VsrldbXBm4CwnOhcq7d1tWsAZ0qeXoZr5RSqnR077L3lXNtlytA187pK49SU0QDOqWUUqWjZ4+9D8+FiBfQdTZPX3mUmiIa0CmllCodgy109XYMnRPQFjo1K+gYumKjyzip8dJzRqm07t32Plxng7nKedC1Y3rLpNQU0ICumOgyTmq89JxRKlv3biivgUCZfVxZr12ualbQGr+Y6DJOarz0nFEqW/fudHYrQLgWelqmrzxKTREN6IqJLuOkxkvPGaWyde+y3a0p5TUQ3Td95VFqimhAV0x0GSc1XnrOKJWtpyVHQNehq0WokqcBXTHRZZzUeOk5o1S2/nYor04/Lq8Gk2G5v10AACAASURBVIT+zukrk1JTQJMiioku46TGS88ZpdKScYj1Qqgqva28xt5H99nxdEqVKA3oio0u46TGS88Zpaxoh70f2kIHdj3XuUunvkxKTRG9jFdKKVUaou32PrOFrsJroevTxAhV2ooqoBORWhH5mYj8VUReEJHjRGSuiDwgIi9593XeviIi3xeRl0XkWRE5crrLr5RSahqlArpcLXSa6apKXFEFdMA1wG+MMYcCfw+8AHweeMgYczDwkPcY4P3Awd7tEuD6qS9unrmuXYewY7u9d93Z9f5qYpIJO3HqvtfsfTIx3SVSanr0e12uucbQaQudKnFFM4ZORGqAdwMXAhhjYkBMRFYAJ3q73QI8AlwJrADWGmMM8CevdW+RMWbXFBc9P6Z7xv/pfn81MckE7HkeNlyQ/n9beSsseAsEiubjrdTUGGyhywjoQhEQR1voVMkrpm/qpUAr8FMReUpEbhKRCLAgFaR59/O9/RcD2zNe3+xtm5mme8b/6X5/NTE9u9PBHNj7DRfY7UrNNqmkiFBGl6s4tsVOW+hUiSumgC4IHAlcb4w5Augl3b2ai+TYlnPmSBG5RES2iMiW1tYiDVCme8b/6X5/BUzgXE3Gc/+/JeOFKaBSGYqubh1Miohkb6+ogb62qS+PUlOomAK6ZqDZGPO49/hn2ABvj4gsAvDuWzL2X5Lx+kZgZ64DG2NuMMYsN8Ysb2go0glXp3vG/+l+fwVM4FwNlOX+f0stTK5UARVd3Rptt8GcE8jeXl6tXa6q5BVNQGeM2Q1sF5FDvE0nAVuBjcAab9sa4B7v543Aai/b9Vigc8aOn4Ppn/F/ut9fTUzVQjtmLvP/beWtdrtSs01/R3Z3a0qoWrtcVckrtlHT/wTcLiIh4FXgImzQuUFEPg5sA87x9r0P+ADwMtDn7TtzOQ5uw2EkL3oAScYwgRCBqgacqUpI0BUHZqZA0CZAXHS/7WYNlNlgbhwJEa5raOuNEUskCQUD1EdCOE6uEQ1KFbloe3ZCREp5DXTtmPryKDWFiiqgM8Y8DSzP8dRJOfY1wGUFL9QUcV3Diy29XLx2K83tURrrwty4ejmHLKieui9XXXFgZgoEYU7jhF7quoYX93Rz8dot03feKZUv0Y7h4+cAyiPpKU2UKlHa/FIk2npjg1+qAM3tUS5eu4W2Xk1KUIWj550qKdF9I3e5xno1WUiVNA3oikQskRz8Uk1pbo8SSySnqURqNtDzTpWUaMcIXa5V6eeVKlEa0BWJUDBAY104a1tjXZhQMDDCK5SaPD3vVEkZ6IKyyuHbU612qWlNlCpBGtAVifpIiBtXLx/8ck2NZaqP6LQhqnD0vFMlI94PydgIY+hSLXQa0KnSVVRJEbOZ4wiHLKjm7kuP12xDNWX0vFMlY6DL3udsofMCOk2MUCVMA7oi4jhCQ3X5dBdDzTJ63qmSMNBt77WFTs1S2uWqlFJq5uvvtPdlOQK6wTF02kKnSpcGdEoppWa+VJdrKFeXqxfkaQudKmEa0CmllJr5+lMBXY4WOidgt+sYOlXCNKBTSik1842WFAF2+S9toVMlTAM6pZRSM99oLXSp7TqGTpUwDeiUUkrNfBktdP0Jw/N7k/TETPr5UJW20KmSptOWFAHXNbT1xnQeMDUhev4ohW2hC1bgisOqjb080+py8v5BbjrN64INVUHP7ukto1IFpAHdBPj+AnVd6GuFRAyCIahsAMcZdqwX93QPLpCemqn/kAXV+qWsxuS6hjfaeujZt5vakMvumEPP3IXsX1815eePBpZqWg10QijCo9sTPNPqcmgdPPhGgj/vSnD0oqCdi26vdrmq0qVdruOUCsDOuu4xjv/Ww5x13WO8uKcb1zVDd4SWrXDTyfC9t9j7lq12e4a23thgMAd2YfSL126hrTc2Vb9ScXNd6NkDHdvt/ZC/32zXGR1gXt/LvPX+s1lyy9G89f6zmdf3Mp3RgSkth+/PhVKF0t8FoQi3/iVGfQVcfQxUh2D9C3H7fKjKjqEzek6q0qQB3Tj5DsD6WmH9edCxzT7u2GYf97Vm7RZLJAePldLcHiWWSBbsd5gxfAbFs1kk3k713auzzrPqu1cTiU/tWCG9MFHTbqALEwzzp51JjlsIkTI4qgEe2Z7ANQbKq8FNQKx3ukuqVEEULKATkYf8bJtpcgVgDVVl1CT3ZbciJWLpL9mUjm12e4ZQMDC4MHrKqcsamCedJNu3YbpncauUz6B4NitzB3KeZ2XuJFroUq2iXbugs9lX66hemKhp199FD5X0JWDZXLvpqAbY1294fq+bXs9VEyNUicp7QCciFSIyF5gnInUiMte7HQDsl+/3m2pDA7AjltTw0w9EKL/5fdmtSIEQ1DZlv7i2yY6ly1AfCXHj6uWDxzx1WQPXnlRB+c3vI3DNW5Efn4yZra1SPoPiWc0J5D7PnMDEjpdqFf3VZ2Dv3+Cn7/fVOprrwqSxLkwoOMFyKDVe/Z20Juw5+HdeQHfkfHv/h+aEbaEDnVxYlaxCtNB9EngSONS7T93uAX5YgPebUkMDsKtPXkjtPWuGtyI5ATh3XfrLtrbJPq5syDqe4wiHLKjm7kuP53efew/ffv9+lN11ftbxZLa2SgX9BcWzWTJQASuuyz7PVlxnt09EqlX0befBxst9t44O/VykknvqI/p/pabIQBc7Y2EWVMI879qithwWR+CZlqS20KmSl/csV2PMNcA1IvJPxpgf5Pv40y0zAIslkixkb+5WpHgU5i+DTzw4apZr6pgN1eXsaO+jq6eXOdoqZVU22CA41e06QlA8m+1zI5QH5zHn9O/YGfLjfXQG5zHgRpg/kQOmWkXDdeNqHR36udAsVzXlBrrYYcIsrcnefHAtPN2StFmuoJMLq5JVsGlLjDE/EJF3AAdkvo8xZm2h3nOqpAIwAHrKbaCR+eWXakVyHKha4Pu4oWCA3TGHJSMdb7ZxHN9B8WwlToDPPdzNJUc1MT8otMQMN2zu4t/PmmBXZ6pVNNo+8nk9gqzPhVJTKRGDxADNiTD7DxnY8+ZaeGSHoTUZoQG0hU6VrEImRdwK/BfwTuDt3m15od5v2qRakcboWvWjPhKiau5COlbcknU8M5tbpVJBce0Se6/BXJb6SIgrTjmUf/71Tt71Py/zz7/eyRWnHDrxrs7U+fz0Ojjj2ryc10oV3EA3AF2mkv2rs596c629f6bT64fVgE6VqEJOLLwcWGZMiU/6k8dWJMcR9q+vorNyGQMX/i9BE8cJliMRbZVSueW9qzN1Pn/ouzYB4qL77bxd2jqqitlAJwDdJszRQwK6A2tsy8Xz7SFOdoKaFKFKViEDuueBhcCuAr5HcRhn1+rohxLqIhXAorwcT5W+vHd15vF8VmpK9Nt1XHulksVV2U9VBGG/Kvhru2szXbWFTpWoQgZ084CtIvJnYHBSLGPMGQV8T6WUUrPNgA3oQuWVlOVoRD6gGra2JSFcpUkRqmQVMqD7SgGPPTP5WNtVqXHT80rNdl4LXXlFZc6nD6yBP+wyJGsiBLSFTpWoQma5PlqoY89IqQlbh07BMX+ZfvmqidPzSqnBFrpIZTjn0wd4U5n0SBVzdAydKlGFzHLtFpEu79YvIkkR6SrU+xU9XcZKFYKeV0oR7batblWR3AFdam66fW4E+rSFTpWmQrbQZeUaiciZwNGFer+ip8tYqULQ80opujraCAO1Vbm7XBvCUBmE3YkIB0b3TW3hlJoiU9YnY4z5JfDeqXq/oqPLWKlC0PNKKXq79hE1IRZEcrdRiNhxdK8NVEOsB+L9U1xCpQqvkF2uZ2fcPiIi3wRKe0660eRxAmKlBul5pRQDPR10U8nC3A108P+zd+ZxTlVn4/8+2WbfmRmWYQAVoUIVFbWWvpWqFBUVbVVcQVu1rbWt3dRaf9b27Wtdamtba92qgqICtRQU9w1b6oaIgmwi4MzAMDMMs69J7vn9cW4mmZnMnmQSON/PJ5/cnNyce5I8uffJs6Lj6LY2246j5n2xWZjBEEOimeV6Vsi2D9gFzO3rRSLiBNYCu5VSZ4rIBOBpIBdYB1ymlGoXkSRgEXAsUA3MU0rtiug7iAShGYhp+XDVG7rPq8lGNEQChwPyJ+sCwH4vON2QPtLIleGgwtdcSyMppLl73mdCBmz1ZYIHaKqCrKKYrc9giAXRjKG7YpAv/RGwGQi0WL4D+KNS6mkRuR/4NvA3+75GKXWYiFxo7zdviMuOLP3JQDQlJwxDwbKgZhfU7AB3KnibtTsp9xAjR4aDh7Z62hy9mOeACVnwX2VfVpqMhc5w4BFNl2uRiCwXkUoRqRCRZ0Sk179E9vNzgIftx4KOu/uHvctC4Bx7e679GPv5U+z944e+MhADCt/Dp8I9U/V95SY9bjD0h5b90FAOq34Kj83R9w3letxgOEhwtDfgc/au0I3LgH1k6QdGoTMcgETzL/yjwEpgNDAGeNYe6417gOuBgEaTB9QqpXz24zJ7Luz7UgD7+Tp7//ihrwxEU3LCMFS8LbDims4ytOIaPW4wHAT4LUWyvxHLFb5kSYAUF3hSAhY6c441HHhEU6HLV0o9qpTy2bfHgB4jtUXkTKBSKfVB6HCYXVU/nus699UislZE1lZVxfCH3FcGoik5YejCgGVV+cPLkPJHZ4EGQwjDdm4NoaK+lXSaEU/vFjqA/Ixk2nEbhc5wQBJNhW6fiFwqIk77dik6eaEnZgBni8gudBLEyWiLXbaIBGL9ioA99nYZMBbAfj4LCOtnUko9qJSarpSanp8fxew/y4LGCqgt1fcpeb1nIJqSE4YuDFhWXSk9yFDv1ooB0VWuTUiAwSZm59ZeKN3fTAYtuJL6VugmZAn7VCa+BqPQGQ48oqnQfQu4ANgLlAPn2WNhUUr9QilVpJQaD1wIvK6UugR4w34twAJghb290n6M/fzrSqlhK4ti+f1YFZ90joer2qIzEK98Fa7bqO9DEyJMyQnDUEnrQYbS+idDlqWoamhjd00zVQ1tWJbquoOJ8zTENaX76kiVNjw99HENZUIm7FOZNNfujcHKDIbYEs0s1xLg7AhMdQPwtIj8FvgQ+Ls9/nfgcRHZjrbMXRiBYw0Ky1K01lWQuuTi7vFwV74K6YXhX+hwaAXvylf7znI12bCGcAyhbIllKbZWNHDVorWU1bRQlJPCQ/OnM6kwA4fDjmjoKc6zN7k2GGJIdbW2tiWnpNHWx74TMqFaZVJUXxn9hRkMMSZqCp1dP+4HwPjQ4yil+lTylFJvAm/a2zsI0zJMKdUKnB+RxQ6R6qZ2fA2NpA4mHs7h6PvCaBqwG3rCsrQleBCyUd3U3qHMAZTVtHDVorUsv2YG+RlJeicT52mIcxpqbPepJ63PfQtTYbtk4WjeGuVVGQyxJ5rawL/QxYT/AtwdcjvgaPf52dNoRS8ezmTDxow+XZDxxhBko93n71DmApTVtNDuC0moMHGehjinqU6HZvtdfSt0DgGfJ5NU734YvggdgyEqRFOha1VK/Vkp9YZSanXgFsXjDRsel5MHP6in+qyFnWKZrHlPRiYezlhJYkLABXnufWuYcccbnHvfGrZWNMS1Uqd6kA3VD9nwuJwU5XROnijKScHjcgYHTJynIc5pa7AVOnffCh2ASs7BgxfVbGo1Gg4sotn6608i8ivgZQiGNiil1kXxmBHBshTVTe20+/x4XE7y0jzBmKIw5KV5uG7WZG56ZQtXz1rGqHQH2RnpJGcVYiFUN7T1e66wBKwkoRduYyWJOP1yQcYZPnHjDiMbPnHTSxckQMvtQ/Ond4uhy0sLkat+xnkO9DdjGBrm8w7ib67R9/2w0AG40/OgEfbt2UH+xPgqXWowDIVoKnRfBC5Dlx8JpMQp+3Hc0q9A8S44HMKkwgx+e+5RtPv8uF1Oku2L4kDnCkvAStI1TspYSSJKv1yQcUYNmTjPWkjesws6ZKP6rIX4yaSgj9cG5Hb5NTN6Vwz6iPMczG/GMHjM5x2kud2H29sAbrD6aaHLyMqFvVBe8hn5E4+L8goNhtgRTZfrucAhSqmTlFJfs29xrcxBz1aa6qbeXVgOh5CfkcSYnFTyM5JwOGTQc4WZPGglCVf+xBAR+uWCjDPE4eSmNT4+mLWM0gXv8cGsZdy0xoc4+rfmcHI7UCIm54Z+YT7vIHvrWsmkCei/yzUvV1vlavbuitayDIZhIZoWuo+AbCCh8sMjaaWJqMWnP9mwhiHRLxdknBFw9w/nmhPRspnImM87yN66VrKkCZ/Dg3L0FWSg8aRl4cNBa3VplFdnMMSWaCp0hcAWEXmfzjF0kahNFzUCVprQE2ZYK00/6sL1ey5DXNBvF2QcMeg1R7CuoZHz2GI+7yDlda1k0YTPld7/F4mDWslBGnZHb2EGwzAQTZ/dr9Bu19tIoLIlAStNwPUW1uJhWbB/B5R/BLWf6/v9O7pVz+/XXIa4IhIuyFgz4DX7fVpu9++E8vXw7I+H1P3ByHlsMZ93kL312kKn3H13iQilyZ1Lelslrd6Dz6ppOHCJZqeITiVKRGQGcDEQ16VL+mXxaNkPDeWw6qfBJIW590FKNqSN6DTXxPx0ln7nRLx+C7fTQUF6YigJhgOLQFaksvzkN29HllwSlN2z74U3boOz/jgot34iWjYTGfN5Bymva2G6s7nfCREBfMm5FLaWsL2ykaljsqK0OoMhtkQ1ql5EponInSKyC/gtsDmax4sUfVo8vC2w4prOxVxXXKPHQ7AsxadVjVzwwNucdNebXPDA23xa1RjXdc0MBx6h9fVKy0qCyhzo+5XXwrSLtPzWlkJjxYCtdYlo2UxkzOet2VvXSq40Y7kGZqFzpuUxWqrZUl4fpZUZDLEn4gqdiBwuIreIyGbgXqAUEDvL9d5IHy/mWBbK8oUv5mr52V3TTE1TK6qhAlVXiq++gvx0Hax7MGejGaKIZWklrAdlLDQrsijTGb5IdVo+7NsG90yFh08dkgvWYIgVe2q1y7W/Ga4B3Bn5pEg7paW7orMwg2EYiIbLdQvwb+AspdR2ABH5cRSOMyyopiqkfnfYQr876vz87PkPePSMNGTFApy1JXwxu5iHz1rIlS8COPjuzENpbvdR1cBB6yYxRJA++vxalqLd5+fu84/CUorUlLbwRapTR8DLv9SPA+3DrnzVZFYb4pq99a2kq0Za+llUOIA3bSQAjXu2AP8ThZUZDLEnGi7XbwJ7gTdE5CEROQU4YLQWy9cGr/4Kzrm/UzskNW8xP3t+DzfPzCd7xYJOLq28Zxfwu9mj+NnsSfzzgxL27S3Ft/9zWmvLsfwmKNcwBJqrdPzb7Nvg8lX6/o3boLmqw9U678F3mPfgO7R6LX72/O5uLerUvCdgzR+hbG1wXtNazhDntHr91Da1kqqaB2yha08bBYC17zOU6elqOECIuIVOKbUcWC4iacA5wI+BQhH5G7BcKfVypI8ZS3zixplWCMoPly0Hyw9NVTQnF/JhaSkFqQVhXVrFWS7++MIObpvhIu/Z8zusKda8J6FwiikSbBgclgUnfEfHwYUmOVhWtwK0qR4nL22qorIhk9+duZwJWQ4QJ063B5fdPqkD01rOEOdU1reRTSMAfnfGgF7rTR6BT1wUesuoamyjICM5Gks0GGJKNLNcm4DFwGIRyQXOB25E93ZNOAJZgk5nFu6Trsex9LKgUnbB45Q2ezh6bCY5aUnwrZegqQrW3KOtHtnFOFxJXH1sZlCZA6gtwbHkYuPaMgwe5Yd3H9CWuZQcaKnRj8+4k3Z/5wK0tS3ejlIX+VJP0uIFnWTYAbB1VbfWcqZvqCEeKa9rIU90UoPPkzmwF4uDxqSRTPCVs6W8wSh0hgOCaBYWRkSc6ALDLuAF+5ZwhPZOfPibxeQ+d1lnpWzpZUy44jUePSON9CfP6mwpefcBrJm/oNmTw+j0hvAB6ca1ZRgs4gxvoRNntwK097/5GXeddyTpvhrynv1GNxlWV7yAnH5Hp0LDpm+oIV7ZW986eIUOsNJHMaF5D2/ureerh5u+2IbEJ2p+PhH5AVABvAKssm/PRet4UcOy8DdUkN2+lz+dOZqcJBVWKfOo1m6xc6y8lsZTbuN7r7RQ2+wjKyM9GLsUwLi2DENAKX9QmYMOuVPK360AbVVjG4WZyRxRkBRWhkUpyB6rrcUORzfZP3pspsnUNsQN5XWt5KEVOv8gFDp/+kjGOyrYsqc20kszGIaFaFrofgRMUkpVR/EY0cXOIHQ/fRGjaksYlV2M99IVYbMExdsa9iJZ09TGS5uq+KT8PVb94MtY857Ubtb0AjjpBsg9FJR9LBNHZxggyrKQsCV0rJ4L0DY3hc90dXXphtJF9gPZ2h+W1h+UfUMN8cXeulZGuRqAwVno2tKK8OCjYfcW4NgIr85giD3R1CBKgboozh99mquC5SAAaktwv3Iz3vOf6JQlyNz7dPeIMNa3Jr/ur1hW00Jjm4WjcApc9QbM+aPuNPGXY+Dvpu6XYXCIwxlW7sSh5S5sAdrUfB0jFyrDITFzQFjZz3t2ATfPzD9o+4Ya4ovyuhZGe5pQyICTIgBaMycAkFmzkXafOfcaEp9oWuh2AG+KyCqgLTColPpDFI8ZWXzt3a1uW1fRdPLtZFz+PA7LBw4n8o8r4JRfacUu0EEioOhZ+sLXcRF0OHQge9dq/abul2EwiCO83In+rxY+ocEB+ZPhihfA7wWnG9JHdrYQh5P92hJGpTsO2r6hhviivLaVUc5G/JLRIe8DoS11FF7x8AV2smNfI5NHDtzKZzDEE9FU6Ersm8e+JR4uT1jXVJbbj7zwS90uacQk7Tp1p8CLN8JZf4bM0eBwQv0eCtIc3Ztn+9q1yzU0M3HNPVFNjjCZigcovlb4+Gm4eJmWOcsPb/8F/udnPSc0FKThqNqi69VNuwjS8lF+LypzLA6XfUroQfbzszMZmWESIgzDT3l9K/muenyuQSpiDieNaeOYWr+LLeUNRqEzJDzRLFvy62jNHTMCrqnQC196IYLAV38Gy4JlH7h0OaQV6n+Ki8/rGM+Z9wTP/+DLpCeH9Ft0p8Apt3a3qrhTovI2TKbiAYw7FY68EJ48v4sspXarQxdIaHjju0fgeOO2Ttmxkl0M8xaj0vIRhwNS8rTsd+lA4c4oACMzhmHG67fY19BGTk79oBIiAvizxjO14T+8saeWc44eE8EVGgyxJ2oKnYjkA9cDU4COIj9KqZOjdcyIE3BNzbyxw0Uq2cVw6T/hrd8HLWzKgtZa1Oz/Q6o/1da32hK9/5JLybziBUgtCs5r+YPKHOj7FdfAt1+Nytvo6cK+/JoZ5GckReWYhhhhebXshFp8vc2AQll+/nTmaApShcpmxW/frOLD0nrE36b/oHTJjpUll+g5XrpJK3P5k3UYgK+9UykTg2G4qWpoQwFZVi0+z6hBz9OWNZ4Ru1+hunQr+lJlMCQu0XS5LgaWAGcC3wUWAFVRPF50aKnuHu/mbQlaN9ILtLXtH1fobMNAHbDXf6OLCteW6DilUPzh45PwR8fl2u7rXGAWtFJnMhUPAPxeLYMn39K5Ft28JxjhaqDgFW25G2tnqd60JgnlTIK0/PAymJJjYjoNcU95XSsA6f5amjyTBj1PIDEiqWoD8I1ILM1gGDai+Xc7Tyn1d8CrlFqtlPoW8KUoHi/iWJbC723TF7ii6TDvCd0vMzkzePGccV13a9vKa/U46Iur091pTq+4Y1qPLlBgNhSTqXiA4HDpGM5u1rZLcdTu7Jal+sczi3Cm69ABJs0JyvS8J/TjlpqO/eO14LVlKaoa2thd00xVQxuWZXpxHmzsrWvFhY9kXz1+9+Bdrm1pRfhwMbbtU+qavX2/wGCIY6Kp0AV+HeUiMkdEjgaKentBPBGIO9tU2aYvdCffol1Rj82Bhr3BC2XAohFKwNIRaHyeVthpzmtXlHRrkN6tbEQE6VpgtluShiFxcSVD7iHhZdCd2m0sxeHH0bIPXMmomTcEZfqlm+CrP4dtL+p947TgdeA3dO59a5hxxxuce98atlY0GKXuIKO8rqWjqLDPkzXoeZTDRW3KWKbKDrZVNkRqeQbDsBBNl+tvRSQL+CnwFyAT+HEUjxdRAnFn+eluls77Le4n5gYvmk1VwQzAlprwRVqzxsLs25A374Cv3QQFR1Dd5O2IZatsyOTmWcsYle4gPzvTDjaPjn7dY4FZE9ye+HiboGlfeBn0NnfeN7sYsXzw8BlIIFYu1LK8bL6Oodv5VlT/YAwFEw9qAG2hG+/aD4A3ZcSQ5mrPHM/U5vd4rrye48bnRmJ5BsOwEDULnVLqOaVUnVJqo1Lqa0qpY5VSK6N1vEgTiDv7sLSe/S3+zhfLNffoOLnsYr09977uhYZrP4cll+pm509fBE1VnWLZPiyt57dvVlHeaOkg9eaqqBYWDltg1pD4iAM2r4LzO1t81bzFqKziznJ5wePw/iOdY+VCqS2Bwqk6dq7giLhMgDDxoAbQJUsOT9Ytu7zJQ1PoVM4hZEsTlaWfRmJpBsOwEc0s1yK0Ze4rgAX8B/iRUqosWseMJKGNzcvq/RSGWkDK1sK7D8Alz0BbHaQVwDl/0895m7Wr68Xrg5PVlkB7I6nJWR1zHj02k4dPSyPv2ZByExc+FbcXUkOc4kyCqd+At+7S1rW0fEgvQJLSYeV1nWsdrr5TZ7e+Tc+WZU9qXCdChP4uA5h40IOPvbWtHOOuAR94k4dmVWvNHA+A2rMemDX0xRkMw0Q0NYdHgZXAKGAM8Kw9lhDkpbp4+arJvP29iaQluWg4f2lna8dJP4cV34OHT4VnvhUMJh8xCf59t1b6AmQXw/7PyLbqOmLZbp6ZT96zC7p3i2hOvERgwzDib9eu0q2rtEX4kdmwaC54W4Njj80JWovTbDdqqJUZdJzo/JU6EaKxIm7b0Jl4UAPoGLqxzv34XGlYrtS+X9ALbelj8eMgy49l0gAAIABJREFUq/YTlDKxmIbEJZoxdPlKqVAF7jERuS6Kx4scloWjajOpT19Eqt2Y3D/vSSovfJE8VyvO5krwZMDX/0/H0625R18ws4th7l91SZOKDUHLm13GRM57jEmFBSy/ZgY53r3hXV5xmlloiFMsf3g5svxa9tILdMa1XZ9OZRXpWoq2lVnNX4nlTMbRXIUsOjvurcUmHtTgtxSVDW2Myt2HNzlvyPMpp4d9niIOa9lBVWMbBRnJfb/IYIhDoqnQ7RORS4Gn7McXAdVRPF7kCNOY3LnkYlIvfpZGRxqZ7jTklVs6ukdwzv3wwSKYeCq8ZjfImHM3ZI+Dfdt0TbrGSnB5OmLZaEwO7/KKw8xCQ/yiHC5k0hwtiwHX6vqndLHri5fpPxwhHUnUBYvxLngJ/K14JYmfPLebq4/N5NhXEqe3cMdvyHBQUt3Yhs9SjPBX4ksbukIH0Jg+nqlt69m6t8EodIaEJZp/v78FXADsBcqB84AretpZRMaKyBsisllEPhGRH9njuSLyioh8at/n2OMiIn8Wke0i8rGIHBOxlffQmDzdu58s1Yi8ebu2wr10k3ZxPfENOPI8+PRVbflorNSK3qu/1pa7xsruWYOBtmIxKl1iODCx3Glw0vWdy4+cdD1YPmje161GomPpJXisZjw127Fa66lsaKMo09m3tdiytCu2tjSuXbKGA589dlHhHG/FkBMiOsiZwAipp/TzzyIzn8EwDESzl2sJcHbomO1yvaeHl/iAnyql1olIBvCBiLwCXA68ppS6XURuBG4EbgBOBybatxOAv9n3Q6eHxuQkZegSEWHaJrHkUl2gdeq5er9PluM/7XZ8s36HuDy4MgpwhLqvHA7t0jKtlQxDwOlrhqWXdZbFpZehrngBaaoKr6g17IVVPyVj7n387rQi0pIkvLyL7ca0LKjc1K2vazy6ZA0HPuW1LaTSSrKvnroIuFwBHNm6RGrj7k0kWP17g6GDWJ+Nf9LTE0qpcqXUOnu7AdiMTqaYCyy0d1sInGNvzwUWKc07QLaIDL6pXyip+VjznuxsPTv7Xl2Vv6lKu1Jn3xassF80XV/o6vdA/W5tsRs9jV21Xib9fiMnP7CZrZVNnYqfWpaiqsnLbm8GVc4CrNTo1aEzHMD423WcXGjHh/QC3RLM2xy+I0lLTUf/4AlZDpq8qnOCREDexc4cDROCYBJ4DMPF7toWiqUSgPaUyHg02lNHAuDbtz0i8xkMw0E0Y+jC0a/IZREZDxwNvAsUKqXKQSt9IlJg7zYGKA15WZk9Vh5mvquBqwGKi4u7Pt0dh4N9qYey/8zlTMp1Ia4k7cKyfLoqv+ULFmWdNEfH0LU1QGqeVvbmrwCHi7HONF68chJpTj+19RU0ZnnItOpQvnZaLSc3P1fGS5uqOjL1JhVmmODug5wBy6orGU6/S7tXAVxJ+rErGQqmallsrNR/RNY/pbOz2xr0n5CytbgdFq0+pcvwhJY4efcBOONObZFub9bPrbknmL1tEngMDEJeI0B5XStHuPcA0JY2JiJz+pJyaMdNUv3nKKUQMedhQ+IRa4Wuz5xwEUkHngGuU0rV9/LDCvdE2PmVUg8CDwJMnz69X3np4nBCSh74KqCuNBiLdMkyWPXTYH/XE76jLXIBV9QFi8DvA+XHk17IZM9+eGQWY7OLURc8DqvvRLauIjW7mNvOWkhlQyYfltabavcGYBCyKg5obwrKpF3YWlr2a2UsEBqQXQwXPAGfvgQfPq4tcO8+gKO1luKUFPjS9zolT3DO/fD89Xr8tVu1Umhna1O21iTwGIDBnVuHyp7aFo7z7EX5HB2WtSEjDuo8IylqKWd3bQtFOUMrhWIwDAcR9/GJSIOI1Ie5NQCj+3itG63MLVZK/dMergi4Uu37Snu8DBgb8vIiYE+k3kdemofD05qQupLOgeXu1OD2jOu6x9ItnQ+tNXY9sLPBnQRHX6obpi+9TMff2fvmPbuAm2dql4Gpdm8YFP62bokPrLgGPOlhZPNSmHKudsmuvBa+/r/gyUCePF8rbYEwgjl3g/LrunUrrtFynl6gjzX3Pv2n5tLlJoHHMCzsrm1hkmM37akFKGfk/lS0pxYyTvayrcL0dDUkJhFX6JRSGUqpzDC3DKVUjxZB0aa4vwOblVJ/CHlqJbDA3l4ArAgZn29nu34JqAu4ZiOFWF6twKUXwGX/gu+/p3u0futFbZ3rqX1SoCl6IFnixB8EH6fkdNq3IFUbGk21e8Og6KkOnbKC40XTdWzdOX/TFr1TfqVl2uHWz8++Td8HihAvPl/vF5grqwhOvkVbAf96vL73tcbm/RkMXSivbWWcKqMttVf7wICRjJGMk0q2ltdHdF6DIVbE2uXaGzOAy4ANIrLeHrsJuB1YKiLfBkqA8+3nngfOALYDzfRSEmXAWBa+hkrcDhc4nDDrt7D86k4uLU6/C1rrwmcHBrpGgH7O4Qz/XHYxlc3KVLs3DB6HK7wMOpzBwsIn39LF9fq4lt+Fc7oVv+5wpwbkNLsYXCmwcn7C1KkzHLi0+fzsb2xmZPJuatKPjOzk6YUkiZeq3TvQxRMMhsQibtIqlVL/UUqJUupIpdQ0+/a8UqpaKXWKUmqifb/f3l8ppb6vlDpUKfVFpdTavo7RLywLVbkJz6OzEKcbMkYHlTkIurSa90HeYdryEZodOPc+HTweILu4o2q/uuBxHZhuj1vznmRsUTHLr5lhEiIMg8PhgnMf7CyD5z6oEx8ueAJOuiGM6/UyLb+hYyuvha/d3FmGA9vtDaariSEuqKhrY7zsxYUvYgkRAQLxeC0VJtPVkJjEk4UuPmiuQp6+SFs2Git08/NAAkSghVJLjS4c7G+H1AIdd9S8X4970nQAOXQocfWOTLbPWsarH/r4yRl/wH36HbprRGo+BaZUiWEoOJyQmqvj3typulRJah44PWDV6z8d4TJU3V2CvgOWum+9DL421Df/Thseyuq8jE2GJNPVxBAH7K5tYYrsAqA1Y3xE524PFCmu/Ry/pXCaP9iGBMNoE10JdImYcR28eYdW0CbN0W6r0Gr8SoHTDX+crMeUBe89pPefvwL1w/U0XfIc2x3juXzJTr75+Gf87d+fU2llQvZYHVDeXDX8lfdNB4DERhxaFkNxuqGpApZcAn8+WsvrybfoPyWglTFvc+fXZBfD/s+gagssOgtprceTWUhW3kjq3GHqMpquJoZhoLyuhamOXfjFTVtaZGPovMl5WDgYpSr4vLoponMbDLHAKHRdCXSJSMnRGanihK//prvbatkCXY8u9PGpt2p37KK5qLYGLl1awhUL1/HdmYcCIYkPgcr7D58K90zV95WbYq9Mxcs6DIPH2wJPnq8TGQIJDdWfwjNXdnepzrguWJIkdUT3QsKr79DKYG0JLLkER8s+8jOSKMhKxVE4RcfMXbdR35suEYZhYE9tC1+UnbRmFOtwg0jicNHiyaNI9plMV0NCYs7IXUnNR134lF1lf5wu1RCorB9KbYmOjQt93FSp3Vq1JTiWXMLNM/Mpq2khO8XdOfEhXirvx8s6DIPH8nWXzdDSOgFqSyB/sna/vnoLvPBzWPAcXPsBXLxMJ0Q0VgaTIbrGyDkcOgEie6y+N8qcYRgor21mqnMnbZnjozK/lTqCsVLJ1r2NUZnfYIgm5qzcBQuhNuMw2gunoVKydT2ulprwLZQc7s5urNB/jLUlTC3w8Pb3JjItt52V1345mPgQcOuGMhxB5vGyDsPgCWS5htJTy6+qLbo0SdlarbxZXmjYoxMkGiu15S6Q0NNTjJxx0RuGEe++HWTQQkvGIVGZ35eaz3iHsdAZEhOj0IVgWYqtFQ2cde9/ueWlMh0X958/akvd3Pu6ZxI6nHD2X3WM3fkLIau4k4KXVLudUY8eR9JjXye3YTuOQCOLgFs3lOEIMo+XdRgGjztZW9guWaaTcy5ZBnkTdYZrqLx2ybDm7Hvh5Vt0DF52Mery54NWt+xibaXuGiNnXPSGYSazZhMArVGy0HmTC8hnPzv27o/K/AZDNDEKXQjVTe1ctWgtZTUt3HrqSEQp2PmWHSun4LLlurjwNx7S1g1fK9SVwCn/DzYu167K0+8KtgDbZNdA7urKTM3XQeWBC+6kOTB/pbaMxdLq0XUdJtg98bAsunXB87WDO0Vnvl6+SrtZXck6JjTw+PXf6E4QGaPghRuRe6bCP6/Gf8Yf2H7Oc9RmHNbdrWpc9IZhZmTzVnw4aUsf2/fOg6A9RZ/72qs/p8107jEkGKZsSQjtPj9lNS0AJIkftrykLRv1uzv3b+1aqHXufXD4LO3Omr8S5v4VVt+lL6AfPqEnD3VlOhw6qPzKV/UFualKtwkLzHfhU7EJOg9dh68dRHQSSHOVVupMnFT8oxS07O/cy/XcByF3AiRnBZMjLlmms127FSB2aTltqoCytTiXXkrdrGWkZFnkpHU5lnHRG4aRumYvk6zPqEwpRkU6IcLGayt0Y6hg574mJo/MjMpxDIZoYK7YIXhcTopyUvQDyweHzoTVd2rrxgWP6wtguP6tK66BzNF6u7FCd5DYuqpzm6+ursxAkLnDoctLDJfVw+HQyltrHTx6OvxhsnGlJRKWt3vh6+VXg98Lvja45Bn41ktahs+5v3vYQO3nncua1JZwaK47fBs646I3DCO79jUy1bGL+rTxUTtGwEJXJPvYutfE0RkSC6PQhZCX5uGh+dOZfUS+jo9zebTlAnTHiAWroHBKzxmv2cXa2paS07nWVzhXZiC4vL15+K0expWWuITLcq0t0eNZY7XCHiiv4/QEwwYuf14rg0kZ3cqaZKV4wrehMy56wzBSWbadHGnEmzUhasfwJeVgiYtiR5VJjDAkHEahC8HhECYVpPG3WSnICzeCO13Hy+UeBt5GnRHoa9cxb6EEFLmz79WB595mbf0oOCJ83a7Q4PKKjcNv9TCutMQlXJZrIAMbBQ17ddeT7HGgfLD8O7D4PC2v79yvXba2ZY60fB37KRK+DV2oi97UozPEGG/pOgBcedFT6BAH3pQRTPRUm9IlhoTDnIm74GjZh2PJxfCl72oF7v1H7Ri3ufDIbF3E9as/Dyp12cU6o1BZ8O4DMPMG7d565WbtwnR5utftCrWIrblHK4LDafUwrrTExeHqnoE99z5ob9T9XP/1Pbj3OFh4Jvh9cNqdWsFbtkDHzi1bECw4nJILvtauKRZdjhcH9ehM6ZSDEk/VBnw4UVnRSYgI4E3OZ5yx0BkSEJMU0ZWAtSpztC7rMOvX8MQ3unSJmK9LRcz+re7h6k6GlCz9+N93BxMhAvOFYlm6un9gvrK1OuNw9m1QOBU8qbFPSAi40gJKpnGlJQ6+NnjtVi0/gT7Dr90K33xE//noGus5526twC25VO8fsMwFlMDkHEgfOZzvqHcC1u2usmoshQc8efWbKXEUoZzR/aPZnpLPyLp1lOxvprndR6rHXCYNiYGR1FAsS8cbZRfrmLhpF0FzdXh3ZGstZI7R7tV/XAELnoWXbtbJEAG6WrkCF6OGPfq5UKXupZu0Cyu9MPrvsytds11dHpPlmig4nLoo8JJLg2PZxVrRCye37lR9yy4OFszOHA3iAn87JGeCM45PCz3Few7Xb6cXLEtR3dROu8+Px+UkL80T3pVt6BulGNe+jU3JR5PT995DwpuST66/llRa+bSikaPGZkf5iAZDZIjjM/cw0FwFL/1Su0CbqrTloqmqs/IF+nFqHrxwA5z0c+3Cat6PmnkjUrGhZytX4GKUXgDnPQYt1fri6m2GnEOG1yIWcKUZEgtXks7AXnpZUO4ueFyXMgknt4FEnbPv1SEC5y/Uf0jK1urx6zZCam7s30d/SZB4z0CR8kBdy0Drv45uMYYBUVf5ObnUU582IeoKXTDTtYqNe+qMQmdIGIwJJhRfu7awvf4bHROXXqiTHLrGuJ2/UMe+bV0FS+fDSTeg0kbwxv5cNpz+T/w/2oD3ileoTDmEqiYvlqWC8wcuRr5WXTvssTn63tc6PO/ZkNj4WnVpndm3BYsGr74TUOHlNnscFEyBtBFw2u+0QhhQ5hIhbjJB4j1Di5QDlNW0cNWitVQ3xZfimShUbH5Hb+SMj/qxvCkFAExOquaDXTVRP57BECmMhS6UwMWibC0sPAt+/qlOcnjzDn2hzP8CVG3Wjc0DF8HaEsiZgLxwI8d9+ed4ssfg97WwtbKBW179lKpGb/CfeWD+GdfBv76bEG4jQ5xj+fUfi1BXP8DX/xde/n9abtPyg3L1zyu17GYX6xp1DrceT5S4yQSJ9wwtUh6grKaFdtN9YFC0lK7Dr4SM/HFRP1Zb6igAjkvfx4O7TAswQ+JgFLpQul4sfD6dKXjabbrnpeXXFo1Qsouhrgy2riKjYoMOOl98Pl/MLubhsxZy5YtNXLVoLcuvmUF+mj1/e2NCuI0MCYDTozOup10UTIpY/xQ0lAfHmqpgzZ9h+uVBZe6Cx7UVOikTfrgenG6dDOFwxHfsV4LEewaKlIcqdUU5KeELNhv6JKlyA59RRE5aUt87DxHLnYY3KZsj3Hsp3ddCZUMrBRnJUT+uwTBUjEIXSsjFQvl9SHMVvH0fnPCd7q2+XrtVB6Offa920UIw6Nzeznt2ATfPWsY3H//M/mfu1iVNnJ7w8U1x5jYyJACeNDjp+u4xdHs3aJkN5dRf6a4RmaO1MtRWH8yEzS7GP+9JGrMmsqeuPb5jvxIg3jNQpLzr5xi2YLOhTwqatrDONYVRMRLBttTRFFtlALz9WTVzp42JzYENhiEQX39r4wH7YqEUOnNw2kXhW32d96i2xr3+m84xSC0hMRe1JRSkSvCfeXMVPHEurPje8NeeMxwYtDUElTnQ90svgwlf7bxfdrFu8+VOgY+ehpodugZdyOucSy6Gpn0m9isCOBzCpMIMll8zgzU3fI3l18yIL6U4gWiv2UOutZ/atCgWFO56zLQx5DTvJCPZyeqtpmOOITEwFroeEMtOYAjU6gol0OrLmaStdNDZchcgu5jadkfwn3m9PWd6AYjoNkyWH9xp2moSZ24jQwLQU+svZQWtwNnFMO8JSCvUrtWjLtRyN/s2ndwTEg+aJCb2K1I4HEJ+RvRdhAc6n29cw0TAnTc+ZsdsSxuNy9vAV0f6eXNbFZaljDJuiHuMQtcTgZZKfm9496jDpcs+XLxM16TLKgJvaycFzz/vSYoyx5KVkqRPBi473umE78CK74dcbBdrhc5gGCgBOe0mn87OxYbfvAO+8mOdFbvimqDsBUIG7Ni6FsvEfhnii9pP38anHBSOHh+zY7al6fPxV3NrWLUrh3UlNUwfH8flfAwGjMu1R5Q44fxFugtEuNZK4tCK2Wu/QXnSIWM05B7Sqc+ls3AKOWnJwX92qfkw+/+6u3CXXKLdsQbDQHGn6pi5UPm84HFob9LWt5YardRNuwgyCoPKHOj7ldd2tP6qPmshd6zex13nHUlRTgqAif0yDDvJez/gM0cxGamxS0xozdDZtMe6PyfZ7eAfH5TF7NgGw2AxFroeEBFdYDVg0Zh9m7bCuVKgvQGUH/IOR83+PySzKOguDResbVlaYfO1a4tKekFni4rJcDUMlrZ6nQBx+SrtfnW4YMdqKJwCJ9/SOZnnsuVh3bOqcCobT/8nt7y4l6pGL1fPPIx/XvNlvD4r/rJcDQcsjW0+ln+4G5dD+MYxY0hyOdlZWc+Eti1szPgfMmK4Fr8nk7bUkeTuX88JE2ax8qM9/HLOF8hIdsdwFQbDwDAKXU9YPti3VW83Vmprx8m3wMr5IdmEi5DVd2mrW/a48DFw4XpPBmLtEqmgqyE+cadC3mG6QHVAvs59EDJGdUt6YP+O8O5ZdyojR2dz78VjjAJnGBaa233M//u7rCupBeCR/+zkt+dM5fW33uQX0krO6MPwxXpNWYeTUfkBs2YWsHpbFQ/9eyc/mXV4jFdhMPQf43INg2UprdCtvgMyi+CCRXDSDd1dpUvna1dWY0XPLtNwvSdXXKPnA5PhahgalheWX91ZvpZfrS3IXa1xq+/QyRFdsqslLZ/8jCTG5KSSn5FklDlDzHnkPztZV1LLD0+eyM9nT6K6qY15D75Dy6erAZD82CtSLdkTcbdVMyW5mhMm5PLQW5+xZW99zNdhMPQXY6ELQ11LG9kOl7bMNVbAew/pGl7hsgnzJ0HD3p5dpj31nhxxuI61i9PCqIYEwe8NL19+r64511QVzGRtrNTFg7/9Kvjjtyiv4eCisc3HQ//eyTHF2Zx4aB4Ak887ig8+r+GiLZtp847Gmxr7uoPN2ZMAyNz7NvNP/CY3/2sDlz/yPg/OP5Yji0x/V0P8Yc7kYUj11iAv3ADf/Luu27V1FezbFr6HZM0u3T7JnRJ+sp56T7pTIHusjrkzF1TDYHE4w8uXsuCR2fDSTTpUYNIcnU2dlq+TI4zsGeKE5z8up67F26l4b6rHxcxxSYypW0dD/rHDsq62tDG0po0mf8cz5KZ5uP60yfgsi3P+uob/96+N7De1GQ1xhjmbh8GlvFqJe+kXOhYpu1hbOboWAz77Xu3GWn61rusVjkA7MVNE2BANHO7wWdj1u/XjQCbrKbdoZc4ocIY4Y/mHuxmVlczEgvRO43mfv4BD+ajPP2Z4FiZC7eiZZFatI6VuO+Pz0vjduUcy64iRLH73c0666w0e+c9OvH5reNZnMHTBuFzD4HAl6Qtj2VpdXX/ufTru7fXf6O4QORN0/9bQLhH+Hv6tJUjvSUMCk5yp5dKdCt5mff/i9cHna0t0rcTUnOFbo8EQhor6Vt7ZUc03jilCRPA0lZNcv5P2tFGMXX83zZmH0JI1cdjWVzv6fyjY8QyH/vcGNp/8CKPqt3FL8mv84pByVtZO4LfPtbNi/W7+eskxFOWkDts6DQYwCl1YxOEMKnGv/RpOvyt4wUwrgFd+pS14AfrKUk2A3pOGxEREOluHs8fDq7cG/2iAls/0QmMVNsQdL2+qQAEnHpJH4dYnmPD+rxGl5VmJg9Kjfqxrfg4Tfk8WZVO+R/HH93D80qMBsMSF35PBD9ue5cxRMzi74vuc+ef/cM+F05g5qWDY1mowJLxCJyKnAX8CnMDDSqnbhzypt0WXFZl9G4w8SrfpSsnWNb6cSTDzBqjY0FEmQs1bjN9v4WyoQFLzoKW6uzUutBZdH1Y6y1JUN7XT7vObMhKGXrFSRiDeFsSTruXU4YaZN3aST+YtRrlTobYEy5WMhQPxteJ3uGl25wQ7mRgMMealjXsZnZVMcXIz49bdQVP2JKrHnY67tYbmnMm0pRcN9xJpKDyeHcfdStr+TbSnFtA4YhqWM4Xc0pc4ZOsi/nXYBL5deR5XPPo+PzhlIj86ZSJO83syDAMJrdCJiBP4KzALKAPeF5GVSqlNQ5rY5dEZgS210LI/2Pw8u1h3jyh5F+av0PsqC3n5FlxbV8GkOaiTrkdC97/wKcifDFVbOteiu/Ap7YrtotRZlmJrRUNHg/RApX7T2NsQjnafj6TWelh6qZatSXN0vFzAogzg9CB/PxVqS3BmF+O06yC6GitpmbuQz9MnMi4v3ciXIabUNXt5Z0c1p00dyZjND+PwtVA++Qra08f0/eIY05J9OC3ZnUun7C8+DU9LJYftfIK/nDSTu3eO48+vfcoHn+/nzvOOYkx2D4lyBkOUSPRAruOB7UqpHUqpduBpYO6QZ03Nxz/vSTjkpKAyB/p+2Xw4dCYsmgvihMfPDbpfp10UVOYC+z99ETTu7V6L7umLwtauq25q71DmQDdGv2rRWqpNRpUhDEmtVUhAmQNdF/HJ82Hx+brYcEuNfty1DuKM66C2hOwVC2jcv9fIlyHmvLq5Ap+lOGF8LiN2PUvjiKPiUpnrjYrDLqQ1vZjJ7/6Ca0/I5ur/OYT3d9Vw6t2reWD1ZyZhwhBTEl2hGwOUhjwus8c6ISJXi8haEVlbVdWPnqkOB5Uph6CsMMVZa0t0qYjaku7FW1Nyeq4JFm48TO26dp+/U2N00Epdu6+HLFrDAcWAZdXy9S6DPclkSk7HdrbHMvJlGBQDltcQnt9Qzoh0D0c6d5LUtIf6guOitMrooZweyqZeg6u9hkPf+SVfm5TP7887iimjM/ndC1s440//5t0d1cO9TMNBQqIrdOF8RKrbgFIPKqWmK6Wm5+f3LzBc4dAxc+FqfFl+fS9daoC11ITf3+kOPx4mkcLjcnY0Rg9QlJOCx+Xs17oNic2AZbWrjHaVwZ5ksqWmY7u23WHkyzAoBnNuBahv9fLWp1UcPz6XvNKXUeIYtnpzQ6Uto5jKQy8gt/RlCrYvJT8jiZ9+fRI/+/ok6lq8zHvwHX6ydD37GtuGe6mGA5xEV+jKgLEhj4uAPZGYuCA9iXp3LuqCxzvX+Dp/EXy4WLcDs3y6b2bg+fVPdd//wqd0df5+1qLLS/Pw0PzpHUpdIIYuL830ejV0x5+c31nmusrg+qe0rHatU7fmHq3MzV1Ieu5II1+GmPL65kq8fsUJh+SRV/IiTTlH4PdkDPeyBk31uDNozJ3ChPduIbvsNQCOHZfDnecdyTnTxrBy/R5O/v2b3PPqNj7ZU4dS3ewOBsOQkUQWLBFxAduAU4DdwPvAxUqpT3p6zfTp09XatWt7eroTPp9Fu7edlLYqrbw5XNra5m9HuVLA78MvTpTlx2F5cbqTTJbrwUlUvpz+yqqvtRVna1BG/cn5OL21Olvb4QRXsq6T6PeZLFdD1L7ogZxbr160lrWf1/DwGRkc8+xs9ky+gpqxs6K1tJjg8DYyft3tJNfvYs+UK9kz5Tv4knRow+7aFp54ZxcfldahgJxUN8eNz+X0L45kzhdH43Elum0lapgT0wBI6CxXpZRPRK4FXkKXLXmkN2VuoLhcDlyuZEgZ2+25gJSF/RmGqzk3gFp0DoeQn5HU73UaDm5cycmQHJRRF0DyyLBMP8UXAAASrUlEQVT7Ou1bYD8jZYZYU9XQxhtbKznlC4WMKNUJZQ3504d5VUPHcqez69ibGbl1IWM+eZCRWx6n8rDzKf/CFYzJHscNp32B2uZ21pfWsrm8nnUlNby8qYK7X97GXecd1dHH1mAYLAmt0AEopZ4Hnh/udRgMBoOhb55+rwSvXzFrcgEj3voXTdmH40s+MLqYWK5k9kz5DtXj5pC36zkKtz1J4bYn2fuFyymZ9lOyU5OYOamAmZMKsJTi47JaFr39ORc/9E7EathtLq/no9JaXE4HJ0zIZWyu6WBxsJDwCp3BYDAYEoOGVi+P/XcXRxZlMcm7idT6z9h9xNXDvayI05ZexJ6p36XysHnk7/gHozc9TObet9n21b/SlqHjWR0iTBubw+SRmTy6Zid/fu1TNu6u408XTiMj2T3gY27fs4//PnU70+peZTIOlvhncr3/a5x5VBG/OGMyo7JMXbwDHaPQGQwGgyEm/Pm1T6luaufHsw6ncMvN+F0p1Bd+abiXFTV8yTmUH3EVjSOOYfQn93PkqrPY/uW7qCn+esc+yS4HP5vaxAL/W1R9tpXNv6/niyNTSMkZBaOPgYlfh7xDdSeYMLT5/Dz+wn84/v3rmO/YQVnqRDKdXn7X9HfmZ33M/I1XcfKmCn506kS+/ZUJuJ0DiNdr3g++VsgY1ePxDfGDUegMBoPBEHVWrN/NQ//eySmTCzjav5H8XSupGn8Wlit5uJcWdRoKjmVHxv9R9PFfmLz6u5RPvpyaopNJ27+J/M/+QWrddgBakrLZ4c1hS6mLwyq3krFhGbz0C8gZrxW7sSdA5hjIHI3KGMnqrZW8/ezDfLflYZKcfrZ84Tr8Y46nTilydr/OpK0LeTPr1/wq+UZuf8HPsrWlfPsrh3DioXkkux1UN7ZTVtPMjn1NNLb68CtFUVYyx/vWMnHL/Tj22Eku2ePhxO/D0ZeCJ8SFqxS0N4I7rcckP3ztuh+v06gb0Sahs1wHw0AysQyGfjKsWa4GwwCIaZZrq9fPupIalq/bzbIPyjiywM3tU8qY+MGvUc4ktp94O8p58KTmiOWlcNti8kpf7hhrzppIzeiZNBQci9+TSUUz3L8R3quAaSlVXD1iA9P9H5FXtwGnv7XTfD7lwCUW+1ImUHP0tbSnjer0fErddoo+/hMubyMfjr6Q+6q+yNraTOpJ1bVWA/vRymhHLcc4tjHf8SJfdOyiTI1gdfIpFGSlMb31bXLqN2MlZWGNm4FyuHHU7MJZuxPa6sHpgcIvQtGxkFUE7U1QtRUqPoH9n4GyIGcCFB1n36ZD3mHgTtHVI3q2/hmz4AAwCp3BMHSMQmdIFGKq0N37+qf8/uVteFwOTj+ikLvKLsLTvJe2zPGUH/8LvAnW6itSuJr24m6uwJs2Gl+YeqQA7+3x8vxnXt4r9+G1IIl2xkolo2Q/o2U/U1JrOCJHUTBhCm0jj9VWsDA422rJ//gB0vesQZRuRWbhwBIngkIAh/J17N+SOpoPC87lWf6Hj/YJW6v9+BUcJ1u40PUG02Q7Diz86SM5bPwESC+A1lqo2gb7tupySQhkjtaWxdwJugj//h1QuRma93VeYNZY+PHGnj4qo9ANgINOoRORKuDzfuw6AtjX517DRzyvL57XBpFf3z6l1GkRnA8YkKwGiPfPPdKY9ztwoiKrMCh5DUeifKdmnZGlp3VGTV4PRA46ha6/iMhapVTcFkeK5/XF89og/tc3WA7U99UT5v0eeCTKezTrjCyJss54x5SnNhgMBoPBYEhwjEJnMBgMBoPBkOAYha5nHhzuBfRBPK8vntcG8b++wXKgvq+eMO/3wCNR3qNZZ2RJlHXGNSaGzmAwGAwGgyHBMRY6g8FgMBgMhgTHKHRhEJHTRGSriGwXkRujeJyxIvKGiGwWkU9E5Ef2+K0isltE1tu3M0Je8wt7XVtFZHZfaxaRCSLyroh8KiJLRMQzgPXtEpEN9hrW2mO5IvKKPd8rIpJjj4uI/Nk+/scickzIPAvs/T8VkQUh48fa82+3X9vvmkMiMink81kvIvUicl28fHaxJlYyG2l6+Q3EhZxFCxFxisiHIvKc/TisrIlIkv14u/38+JA5BiTP8U68rLsXmRzwuSVG643IeTrKa4zY+drQC0opcwu5AU7gM+AQwAN8BBwRpWONAo6xtzOAbcARwK3Az8Lsf4S9niRggr1OZ29rBpYCF9rb9wPfG8D6dgEjuozdCdxob98I3GFvnwG8gC4E+SXgXXs8F9hh3+fY2zn2c+8BJ9qveQE4fQjf2V5gXLx8dgeqzMbwNxB3chbh9/0T4Engud5kDbgGuN/evhBYMlh5judbPK27F5kc0LklhuvdxRDP08PwXQ/qfD2cMpoIN2Oh687xwHal1A6lVDvwNDA3GgdSSpUrpdbZ2w3AZqC30ulzgaeVUm1KqZ3Adnu9YddsWyJOBv5hv34hcM4Qlz3XnqfrfHOBRUrzDpAtIqOA2cArSqn9Sqka4BXgNPu5TKXU20r/ghcNYW2nAJ8ppXorahoPn120iJnMRppefgPxKGcRQUSKgDnAw/bj3mQt9HP4B3CKvf+A5Dn672rIxM26I3heHk4G+vuJJUM5Xxt6wSh03RkDlIY8LqP3H3NEsF0pRwPv2kPX2ibxRwLm8l7W1tN4HlCrVEdfl4G+FwW8LCIfiMjV9lihUqoc9IkPKBjk2sbY213HB8OFwFMhj+Phs4slwyKzkabLbyAe5SxS3ANcD1j2495kreN92c/X2fsP9HOId+Jy3UM8L8eKSJynY8lQzteGXjAKXXfCxddENRVYRNKBZ4DrlFL1wN+AQ4FpQDlwdx9rG+h4f5mhlDoGOB34voh8tZd9Y702fVAda3Q2sMweipfPLpYk0lrDEuY30OOuYcYS5nsUkTOBSqXUB6HDYXZVfTyXEO93AMTduiNwXo4VkThPx4QInK8NvWAUuu6UAWNDHhcBe6J1MBFxo08ai5VS/wRQSlUopfxKKQt4iKCpuae19TS+D21Sdw3mvSil9tj3lcByex0VARO9fV85yLWV2dtdxwfK6cA6pVSFvda4+OxiTExlNtKE+w0Qf3IWKWYAZ4vILrRb8WS0xa4nWet4X/bzWcB+Bv45xDtxte4InZdjQoTO07FiqOdrQ28MdxBfvN0AFzqgegLB4NwpUTqWoGN67ukyPipk+8foWAKAKXQOFN2BDjDtcc3of0KhwdbX9HNtaUBGyPZ/gdOAu+gcbHunvT2HzsG279njucBOdKB6jr2daz/3vr1vIFj9jEF8hk8DV8TTZ3cgy2wMfwNxJWdReu8zCSZFhJU14Pt0TopYOlh5judbPK27F5kc0LklRmuNyHk6hp/tkM7Xwy2n8X4b9gXE4w2dCbQNnVnzyyge5ytoM/LHwHr7dgbwOLDBHl/ZReh/aa9rKyHZej2tGZ019h46qHQZkNTPtR1i/6A+Aj4JzImO33kN+NS+D1w0BfirffwNwPSQub5lH397lx/zdGCj/Zp7sQtdD+DzSwWqgayQsWH/7A5kmY3hbyBu5CyK730mQYUurKwByfbj7fbzhwxWnuP9Fi/r7kUmB3xuicFaI3aejsFaI3K+Nreeb6ZThMFgMBgMBkOCY2LoDAaDwWAwGBIco9AZDAaDwWAwJDhGoTMYDAaDwWBIcIxCZzAYDAaDwZDgGIXOYDAYDAaDIcExCt0wIiLniogSkclDmOMxETnP3n5YRI6I3ApBRG7q8rgxkvMb4gcRGSkiT4vIZyKySUSeF5HDY3DcW0XkZ/b2b0Tk1AjPf52IpIY83iUiIyJ5DEPiEYtzWSTO8RFaR6ffgOHAxCh0w8tFwH/QBUOHjFLqSqXUpkjMFcJNfe9iSHTshu/LgTeVUocqpY5Af/eFsVyHUuoWpdSrEZ72OnQNLIMh1kT0HD8EzG/gIMAodMOE3SdwBvBt7B+7iMwUkbdEZLltIblfRBz2c40icreIrBOR10QkP8ycb4rIdHv7NHvfj0TkNXvseBH5r4h8aN9PsscvF5F/isiLIvKpiNxpj98OpIjIehFZ3OVYM+3j/UNEtojIYlspQESOs+f/SETeE5EMEUkWkUdFZIN9/K+FHPtfIvKsiOwUkWtF5Cf2Pu+ISK6936H2+j4QkX8P9z/eA5CvAV6l1P2BAaXUeuA/InKXiGy0v7t5oOXXlsN19vhce3y8LQ8LRTfc/kfAMmBbxu6wZeI9ETms6yK6WJzDydF4+/tfZ9++bO8bVh5F5IfAaOANEXmjy7HGi8hmEXlIRD4RkZdFJMV+7jARedU+9jpb/qSHz2KmiKwWkaUisk1EbheRS+w1bxCRQ+398kXkGRF5377NiPSXaBgaIjLOluuP7ftie/wsEXnXPi+9KiKF9vitopvKvykiO2x5C8zV7Rxvj/dXXnpaS8dvxH7cGDLvgH4DhgOM4a5sfLDegEuBv9vb/wWOQVeOb0VX/3YCrwDn2fso4BJ7+xbgXnv7sZB93kRXxc8HSoEJ9nigSngm4LK3TwWesbcvR7dWyUJXpv8cGGs/19hl3Y32/UygDt1jzwG8ja6w7rHnOi70mMBPgUftsclAiX2sy9FV8DPsddcB37X3+yO6MTboaucT7e0TgNeH+zs8kG7AD4E/hhn/pi2HTrS1rgQYZX+nmfY+I+zvUIDxtqzOsJ97BPiZvb2LYCX7+QS7JNwass9jwHm9yFEqkGyPTQTW9iaPIccdEfKedtlrHg/4gGn2+FLgUnv7XeBcezvZPm5Pn8VMoNbeTgJ2A7+2X/sj7BZSwJMhayoGNg/3934w3+hybrPHngUW2NvfAv5lb+dARyH+K4G7Q2T3v/b3PgLdCcFtP9ftHB8iq/2Rl57W8hj2OT/0fQzkN2BuB+bNWOiGj4vQfe2w7y+yt99TSu1QSvmBp9BKEoAFLLG3nwgZD8eXgLeUUjsBlFL77fEsYJmIbEQrS1NCXvOaUqpOKdUKbALG9eM9vKeUKlO6sfJ69AVyElCulHrfPna9Uspnr/dxe2wLWmkMxGe9oZRqUEpVoU9Iz9rjG4Dx9j/dL9trXw88gD4ZGqLPV4CnlG6gXQGsBo5DK2+3icjHwKvAGILu2VKl1Bp7u6usPhVyf2Ivx+1JjtzAQyKyAd0SKzRmNJw89sVOpS2RAB+g5S0DGKOUWm4fu1Up1dzLZwHwvlKqXCnVhm5X9LI9viFkHacC99oyvBLItI9liB9ORCveoM9XAdktAl6y5e7ndD53rlJKtSml9gGVBH8HPZ3joX/y0tNaemMwvwHDAYJruBdwMCIiecDJwFQRUeh//Ap43r4PpafebL31bJMenv9ftPJ0roiMR1v0ArSFbPvpn2yEe01Px5Z+zmOFPLbsOR1ArVJqWj/WZBgcn6AtY13p6Xu7BG1RPVYp5RWRXWhLFvQuwz1thztuuOd/DFQAR6HlojXkuUjIcAo9v+ehyDDo9Z6olGrpx7oM8UFABv8C/EEptVJEZqItcwG6yV1P53gRuT7Ma3qSl57W4sMOlxIRQVuze1xLb2/OcGBhLHTDw3nAIqXUOKXUeKXUWGAn+h/Y8SIyQXTs3Dx0QC3o7ypwwb04ZDwcbwMnicgEALHj0NAWut329uX9XKtXRNz93BdgCzBaRI6zj50hIi7gLbQSgOjMyWJ00+U+UUrVAztF5Hz79SIiRw1gTYa+eR1IEpGrAgP2d1gDzBMRp+i4za+im8RnAZW2Mvc1Olt0i0UkYH0LBIUHmBdy/3Yv6+lJjrLQljsLuAx9oeyLBrRLv1/Y8lYmIufYx04SHQf4FuE/i/7yMnBt4IGImD8o8cd/Cca7XUJQdkPPnQv6MU9v5/ihrmUXcKy9PRdtte6LAf0GDImJUeiGh4vQGYWhPINW1N4Gbgc2ok8Agf2agCki8gH6n99veprcdl1eDfxTRD4i6Kq9E/idiKyhfxdCgAeBj6VLUkQvx25HX6z/Yh/7FbTl5j7AabsslgCX2+6G/nIJ8G17zk/QJzJDhFBKKeBcYJbosiWfoK0QTwIfAx+hlb7rlVJ7gcXAdBFZi/5utoRMtxlYYLtjc4G/hTyXJCLvomOFftzLenqTowUi8g7aZd/Uj7f3IPDCAAPCLwN+aL+H/wIj0b/FcJ9Ff/kh+jP7WEQ2Ad8dwGsNkSdVRMpCbj9Bf0dX2N/7/2/Xjm0QDIEwgB69o1g6hiu4hds4gq0b2DmFA1janwXY/RoKNV7yXktCKIB8HLeLvk8j+lk4ttbOEXGbmPvdHT/r1VoO0R/sl+j9xN86AxTzbPLkD4xS/j4ztwtj98xc/X5VMG985Z8yc70wdo2Izeg1AuCDVOgAAIpToQMAKE6FDgCgOIEOAKA4gQ4AoDiBDgCgOIEOAKA4gQ4AoLgHUoEmQiVoBIUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 608.75x540 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Your Code here for the pairplot\n",
    "#numerical features\n",
    "sns.pairplot(cr_df, vars = ['ApplicantIncome','CoapplicantIncome','LoanAmount'], hue = 'Loan_Status')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "the plot does not show any diffrence between the two groups with diffrent loan statuse for this numeric variables they have same distributions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Gender</th>\n",
       "      <th>Married</th>\n",
       "      <th>Dependents</th>\n",
       "      <th>Education</th>\n",
       "      <th>Self_Employed</th>\n",
       "      <th>ApplicantIncome</th>\n",
       "      <th>CoapplicantIncome</th>\n",
       "      <th>LoanAmount</th>\n",
       "      <th>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>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>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\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>1.0</td>\n",
       "      <td>Rural</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\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>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\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>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\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>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Gender Married Dependents     Education Self_Employed  ApplicantIncome  \\\n",
       "0   Male      No          0      Graduate            No             5849   \n",
       "1   Male     Yes          1      Graduate            No             4583   \n",
       "2   Male     Yes          0      Graduate           Yes             3000   \n",
       "3   Male     Yes          0  Not Graduate            No             2583   \n",
       "4   Male      No          0      Graduate            No             6000   \n",
       "\n",
       "   CoapplicantIncome  LoanAmount  Credit_History Property_Area  Loan_Status  \n",
       "0                0.0           0             1.0         Urban            1  \n",
       "1             1508.0         128             1.0         Rural            0  \n",
       "2                0.0          66             1.0         Urban            1  \n",
       "3             2358.0         120             1.0         Urban            1  \n",
       "4                0.0         141             1.0         Urban            1  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cr_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>ApplicantIncome</th>\n",
       "      <th>CoapplicantIncome</th>\n",
       "      <th>LoanAmount</th>\n",
       "      <th>Credit_History</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Loan_Status</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3833.5</td>\n",
       "      <td>268.0</td>\n",
       "      <td>126.5</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3812.5</td>\n",
       "      <td>1239.5</td>\n",
       "      <td>125.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             ApplicantIncome  CoapplicantIncome  LoanAmount  Credit_History\n",
       "Loan_Status                                                                \n",
       "0                     3833.5              268.0       126.5             1.0\n",
       "1                     3812.5             1239.5       125.0             1.0"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Uncomment below and fill in the blank\n",
    "cr_df.groupby('Loan_Status').median()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "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>ApplicantIncome</th>\n",
       "      <th>CoapplicantIncome</th>\n",
       "      <th>LoanAmount</th>\n",
       "      <th>Credit_History</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Loan_Status</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5446.078125</td>\n",
       "      <td>1877.807292</td>\n",
       "      <td>142.557292</td>\n",
       "      <td>0.505208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5384.068720</td>\n",
       "      <td>1504.516398</td>\n",
       "      <td>140.533175</td>\n",
       "      <td>0.895735</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             ApplicantIncome  CoapplicantIncome  LoanAmount  Credit_History\n",
       "Loan_Status                                                                \n",
       "0                5446.078125        1877.807292  142.557292        0.505208\n",
       "1                5384.068720        1504.516398  140.533175        0.895735"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cr_df.groupby('Loan_Status').mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Loan_Status</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Gender</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Female</th>\n",
       "      <td>0.322034</td>\n",
       "      <td>0.677966</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Male</th>\n",
       "      <td>0.310484</td>\n",
       "      <td>0.689516</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Loan_Status         0         1\n",
       "Gender                         \n",
       "Female       0.322034  0.677966\n",
       "Male         0.310484  0.689516"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(cr_df['Gender'],cr_df['Loan_Status'],normalize='index')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Loan_Status</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Married</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>No</th>\n",
       "      <td>0.369159</td>\n",
       "      <td>0.630841</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Yes</th>\n",
       "      <td>0.282500</td>\n",
       "      <td>0.717500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Loan_Status         0         1\n",
       "Married                        \n",
       "No           0.369159  0.630841\n",
       "Yes          0.282500  0.717500"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(cr_df['Married'],cr_df['Loan_Status'],normalize='index')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Loan_Status</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Education</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Graduate</th>\n",
       "      <td>0.291667</td>\n",
       "      <td>0.708333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Not Graduate</th>\n",
       "      <td>0.388060</td>\n",
       "      <td>0.611940</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Loan_Status          0         1\n",
       "Education                       \n",
       "Graduate      0.291667  0.708333\n",
       "Not Graduate  0.388060  0.611940"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(cr_df['Education'],cr_df['Loan_Status'],normalize='index')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Loan_Status</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Self_Employed</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>No</th>\n",
       "      <td>0.864583</td>\n",
       "      <td>0.867299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Yes</th>\n",
       "      <td>0.135417</td>\n",
       "      <td>0.132701</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Loan_Status           0         1\n",
       "Self_Employed                    \n",
       "No             0.864583  0.867299\n",
       "Yes            0.135417  0.132701"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(cr_df['Self_Employed'],cr_df['Loan_Status'],normalize='columns')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "3l2vbivye2D-"
   },
   "source": [
    "**Insights:**\n",
    "- `Self Employed people have same distribution, We should check how this feature influences final model using coefficients`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Loan_Status</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dependents</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.600000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.310145</td>\n",
       "      <td>0.689855</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.352941</td>\n",
       "      <td>0.647059</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.247525</td>\n",
       "      <td>0.752475</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3+</th>\n",
       "      <td>0.352941</td>\n",
       "      <td>0.647059</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Loan_Status         0         1\n",
       "Dependents                     \n",
       "0            0.400000  0.600000\n",
       "0            0.310145  0.689855\n",
       "1            0.352941  0.647059\n",
       "2            0.247525  0.752475\n",
       "3+           0.352941  0.647059"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Your code here for cross tabs\n",
    "#Create a cross tab for dependents and loan status and state on the findings\n",
    "pd.crosstab(cr_df['Dependents'],cr_df['Loan_Status'],normalize='index')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Loan_Status</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Property_Area</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Rural</th>\n",
       "      <td>0.385475</td>\n",
       "      <td>0.614525</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Semiurban</th>\n",
       "      <td>0.231760</td>\n",
       "      <td>0.768240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Urban</th>\n",
       "      <td>0.341584</td>\n",
       "      <td>0.658416</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Loan_Status           0         1\n",
       "Property_Area                    \n",
       "Rural          0.385475  0.614525\n",
       "Semiurban      0.231760  0.768240\n",
       "Urban          0.341584  0.658416"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Create a cross tab for the property type and loan status and state the findings\n",
    "pd.crosstab(cr_df['Property_Area'],cr_df['Loan_Status'],normalize='index')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "iyASSsoIopYv"
   },
   "source": [
    "### Data Preparation (2 marks)\n",
    "- Split the data into independent and dependent variables \n",
    "- create dummy variables wherever necessary and add in the independent features set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define X and Y variables\n",
    "\n",
    "# Uncomment below to answer\n",
    "X = cr_df.drop(columns = 'Loan_Status')\n",
    "Y = cr_df['Loan_Status'].astype('category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 614 entries, 0 to 613\n",
      "Data columns (total 10 columns):\n",
      " #   Column             Non-Null Count  Dtype  \n",
      "---  ------             --------------  -----  \n",
      " 0   Gender             614 non-null    object \n",
      " 1   Married            614 non-null    object \n",
      " 2   Dependents         614 non-null    object \n",
      " 3   Education          614 non-null    object \n",
      " 4   Self_Employed      614 non-null    object \n",
      " 5   ApplicantIncome    614 non-null    int64  \n",
      " 6   CoapplicantIncome  614 non-null    float64\n",
      " 7   LoanAmount         614 non-null    int64  \n",
      " 8   Credit_History     614 non-null    float64\n",
      " 9   Property_Area      614 non-null    object \n",
      "dtypes: float64(2), int64(2), object(6)\n",
      "memory usage: 48.1+ KB\n"
     ]
    }
   ],
   "source": [
    "X.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "id": "jxRMhHU-e2EE"
   },
   "outputs": [],
   "source": [
    "# Convert categorical vriables to dummy variables\n",
    "\n",
    "#Uncomment below to answer\n",
    "X = pd.get_dummies(X,columns=['Gender', 'Married', 'Dependents', 'Education', 'Self_Employed', 'Property_Area'], drop_first=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "id": "xEfIZNcle2EH"
   },
   "outputs": [],
   "source": [
    "## Split into training and test set\n",
    "from sklearn.model_selection import train_test_split\n",
    "\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": "5411AIKsoflB"
   },
   "source": [
    "### Model building using Logistic Regression from Sklearn (10 marks)\n",
    "\n",
    "- Fit a Logistic Regression model on all of the train data (2 marks)\n",
    "- Predict on the test data and append the predictions as a column in the test set (2 marks)\n",
    "- Complete the function to draw a confusion matrix and pass on the test set predictions along with actual values to generate the consfusion matrix as a heatmap (3 marks)\n",
    "- Generate the classification report using the classification_report funciont(1 mark)\n",
    "- Store the model coeffs in a dataframe (2 marks)\n",
    "-"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "executionInfo": {
     "elapsed": 1735,
     "status": "ok",
     "timestamp": 1605043769676,
     "user": {
      "displayName": "Shrish Chandra Pandey",
      "photoUrl": "",
      "userId": "03777729159992968707"
     },
     "user_tz": -330
    },
    "id": "FWQWq97Re2EL"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=False,\n",
       "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
       "                   multi_class='auto', n_jobs=None, penalty='l2',\n",
       "                   random_state=42, solver='lbfgs', tol=0.0001, verbose=0,\n",
       "                   warm_start=False)"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report, classification_report\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "# Uncomment and answer below\n",
    "logreg = LogisticRegression(random_state=42,fit_intercept=False)\n",
    "logreg.fit(X_train, Y_train)              # fit the model on train data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "id": "egK-n26Ve2EN"
   },
   "outputs": [],
   "source": [
    "#Uncomment to answer\n",
    "y_predict = logreg.predict(X_test)             # Predicting the target variable on test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "id": "_trOGkFfe2EQ"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ApplicantIncome</th>\n",
       "      <th>CoapplicantIncome</th>\n",
       "      <th>LoanAmount</th>\n",
       "      <th>Credit_History</th>\n",
       "      <th>Gender_Male</th>\n",
       "      <th>Married_Yes</th>\n",
       "      <th>Dependents_0</th>\n",
       "      <th>Dependents_1</th>\n",
       "      <th>Dependents_2</th>\n",
       "      <th>Dependents_3+</th>\n",
       "      <th>Education_Not Graduate</th>\n",
       "      <th>Self_Employed_Yes</th>\n",
       "      <th>Property_Area_Semiurban</th>\n",
       "      <th>Property_Area_Urban</th>\n",
       "      <th>Observed Loan Status</th>\n",
       "      <th>Predicted Loan Status</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>220</th>\n",
       "      <td>2221</td>\n",
       "      <td>0.0</td>\n",
       "      <td>60</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>2071</td>\n",
       "      <td>754.0</td>\n",
       "      <td>94</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>569</th>\n",
       "      <td>3166</td>\n",
       "      <td>2064.0</td>\n",
       "      <td>104</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>330</th>\n",
       "      <td>4384</td>\n",
       "      <td>1793.0</td>\n",
       "      <td>117</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>407</th>\n",
       "      <td>2213</td>\n",
       "      <td>0.0</td>\n",
       "      <td>66</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     ApplicantIncome  CoapplicantIncome  LoanAmount  Credit_History  \\\n",
       "220             2221                0.0          60             0.0   \n",
       "109             2071              754.0          94             1.0   \n",
       "569             3166             2064.0         104             0.0   \n",
       "330             4384             1793.0         117             1.0   \n",
       "407             2213                0.0          66             1.0   \n",
       "\n",
       "     Gender_Male  Married_Yes  Dependents_0  Dependents_1  Dependents_2  \\\n",
       "220            1            1             1             0             0   \n",
       "109            1            1             0             0             0   \n",
       "569            1            1             1             0             0   \n",
       "330            1            0             0             1             0   \n",
       "407            0            0             1             0             0   \n",
       "\n",
       "     Dependents_3+  Education_Not Graduate  Self_Employed_Yes  \\\n",
       "220              0                       0                  0   \n",
       "109              1                       1                  0   \n",
       "569              0                       0                  0   \n",
       "330              0                       0                  0   \n",
       "407              0                       1                  0   \n",
       "\n",
       "     Property_Area_Semiurban  Property_Area_Urban Observed Loan Status  \\\n",
       "220                        0                    1                    0   \n",
       "109                        1                    0                    1   \n",
       "569                        0                    1                    0   \n",
       "330                        0                    1                    1   \n",
       "407                        0                    0                    1   \n",
       "\n",
       "     Predicted Loan Status  \n",
       "220                      0  \n",
       "109                      1  \n",
       "569                      0  \n",
       "330                      1  \n",
       "407                      1  "
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Observe the predicted and observed classes in a dataframe.\n",
    "\n",
    "z = X_test.copy()\n",
    "\n",
    "# Uncomment below to answer\n",
    "z['Observed Loan Status'] = Y_test\n",
    "z['Predicted Loan Status'] = y_predict\n",
    "z.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "#import required libraries\n",
    "from sklearn.metrics import recall_score, confusion_matrix, classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "id": "dzEvwZVie2ES"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW8AAAEHCAYAAABhm8cjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAaVElEQVR4nO3deXxU9dXH8c9JAm4ISdgSA09FpVpF64IUFS0uKGgt9KlYrUXaYrGu1FqXShG1rl2wWlEbpUrVKrg9WBUF4wYVEUFQEG1wYwtrCKhQJZnz/DE3MUiWyZDJzA++b1/3lZnf3Ln3XI2Hw7m/e6+5OyIiEpasdAcgIiJNp+QtIhIgJW8RkQApeYuIBEjJW0QkQDnpDqAhx3bpp6kwspWi7N3THYJkoAc/ecK2dRub13yYcM5p1WGvbd7ftlDlLSISoIyuvEVEWlSsKt0RJEzJW0SkWlVluiNImJK3iEjEPZbuEBKm5C0iUi2m5C0iEh5V3iIiAdIJSxGRAKnyFhEJj2u2iYhIgHTCUkQkQGqbiIgESCcsRUQCpMpbRCRA6nmLiARIs01ERMLjrp63iEh41PMWEQmQet4iIgFS5S0iEqCqzemOIGFK3iIi1dQ2EREJkNomIiIBUuUtIhIgJW8RkfDoIh0RkRDp8ngRkQAF1DbJSncAIiIZw2OJL40ws7+b2Sozm19rLN/MpppZafQzLxo3M7vdzBaZ2dtmdmhj21fyFhGpFoslvjTufqD/18auBErcvTtQEr0HGAB0j5bhwF2NbVzJW0SkWjNW3u7+KlD+teGBwPjo9XhgUK3xf3jc60CumRU2tH0lbxGRak2ovM1suJm9WWsZnsAeOrt7GUD0s1M0XgQsqbXe0misXjphKSJSrQmzTdy9GChupj1bXbto6AtK3iIi1VI/22SlmRW6e1nUFlkVjS8FutZarwuwvKENqW0iIlKtGXve9XgKGBq9HgpMqjV+djTrpDewvrq9Uh9V3iIi1Zqx8jazh4G+QAczWwqMBm4GJprZMGAxMDha/VngZGARsBH4WWPbV/IWEanWjHcVdPcz6/no+DrWdeCCpmxfyVtEpFqlLo8XEQmPNzjBI6MoeYuIVAvo3iZK3iIi1ZS8RUQCpMegiYgESJW3iEiAqvQkHRGR8KjyFhEJkHreIiLh8ZjmeYuIhEdtExGRAKltIiISoErNNhERCY/aJtKQjoUd+e1tl5PfMR+PxXj6n8/y+LgnufrOkXTdO/4wjTZtd+OzDZ/zi5N+udX3D+/bkwuvPZ/s7CyeeXgyD4+dAEBB1wKuvvMqds9tS+k7pdw44hYqN4dzl7QdXaudWvG7ideT07oV2TlZvPHsDJ64dULN52dfew7HDD6Wc/Y/q87vn3r+/9L3R8cTq4rxj2vG8c6rcwE46LuHMGT0z8nKzuLlR17gX3c92SLHEyTdmEoaUlVVxV3X/Y3S+YvYZbdd+NvkO3nz1dlcd/4NNeucN+pcPv/0862+m5WVxYjrL+KyH1/B6rI13P3MHbw2ZQaflC7m3KvO4dF7nuClp17mkptGcPIZ/Xnqgadb8tBkG2z+YjM3njmaLzb+l+ycbEY9dgPzXn6LD976D90O3Jtd2+5a73f36N6F3qf24Yp+I8jrnM+VD13Db/peCMDQ3/+Cm8+6lvIVa7nuqT8w+4VZLC9d2lKHFZaAKu+UPgbNzPYzsyvM7HYzuy16/a1U7jME5avKKZ2/CIBNn29iceliOhR02GKdvqceQ8mkl7b67n4H78vyj5dTtngFlZsreXHSyxx14pEAHHLUwbzyzKsAPP/oFPqcdFSKj0Sa2xcb/wtAdk42Oa1ywB3LyuLMkWfzyE0P1Pu9w/r14vV/Tafyy0pWL1nFyo/L2Pvgfdj74H1Y+XEZq5espGpzJa//azqH9evVUocTnpgnvqRZypK3mV0BPEL8qchvALOi1w+b2ZWp2m9oOnfpzD499mHhW+/VjB30nQNZt7qCZR8t22r9DoUdWFW2uub96hVr6FDYgbZ5bflsw2fEquKVw+qyNXQoaJ/6A5BmZVlZ3PDsn7lzzn28M20eH8wt5cShA5gzdRYVq9bV+728gnzKy9bUvC9fsZa8gvbkFbSnvGztV+Nla8kryE/pMQStqirxJc1S2TYZBhzg7ptrD5rZGGAB8We5bcXMhgPDAb6Zux977NYlhSGm18677sx1xVcz9pq72PjZxprx4wYeW2fVDWDYVmPujlkd480XqrQQj8UYefKl7Np2V35VfAX79tqfXqccyQ0/GtXg9+r67089vxf6xaifq20CQAzYo47xwuizOrl7sbv3dPee23Pizs7J5rri0bzw5ItMmzy9ZjwrO4ujB/ThpX+9XOf3VpetplNhx5r3HQs6sHbFWtaXr6dN2zZkZcf/k3YsjI9LmDZu2MjCGQvY/4gedP5GAX9+5U5unX43rXfZiT+/Mnar9cvL1pJf+FXrLb+gPetWllO+Yi35hV/9DSy/MD4u9VDbBIBfASVmNtnMiqPlOaAEGJHC/Qbh8j9dyieLFvPoPY9vMX7Y0Yey5IMlrKn1V+Da3pv3PkXdiijoWkBOqxyOG9iX16bOAOCt1+bx3VOOAeCkwSfy7ymvpfYgpFntnt+25qRkq51a06PPQXz0zgdcePgwLunzSy7p80u+3PQFl3536+fUzpk6i96n9iGndQ4du3aioFshH8xdxIfzFlHQrZCOXTuR3SqH3qf2Yc7UWS19aOHwWOJLmqWsbeLuz5nZN4FeQBHxfvdSYJa7p79hlEY9Dj+AE0/rxwcLP+Se5+8G4N5b/s7MF9/guO8fS8n/bdkyad+5Pb/546/57dkjiVXFuH3UHfzhoZvIyspi8oTn+fg/nwBQfOM9jLpzJMMu/yml8z/g2Ueea/Fjk+Tldsrj3DEXkZWVhWVlMfPpfzP3xdn1rn/oCYfT7aC9eXzMIywrXcLMZ/7NLS/cTqyyivtH3YPHYjgw/up7ufwfV5OVncUrE0tYVrqk5Q4qNBlQUSfKPIPnNR7bpV/mBidpU5S9e7pDkAz04CdP1NHgb5rPrzkz4Zyz2zUPb/P+toXmeYuIVMuAWSSJUvIWEakWUNtEyVtEJBLSVEElbxGRaqq8RUQCpOQtIhKgDJi/nSglbxGRiFcqeYuIhEdtExGRAAU02ySl9/MWEQlKM96YyswuMbMFZjbfzB42s53NrJuZzTSzUjObYGatkw1VyVtEpFozJW8zKwIuBnq6ew8gGzgDuAW41d27A+uI3zo7KUreIiIRr4olvCQgB9jFzHKAXYEy4Djgsejz8cCgZGNV8hYRqdaEytvMhpvZm7WW4dWbcfdlwJ+AxcST9npgNlDh7tVPBV9K/I6rSdEJSxGRiDdhtom7FwPFdX1mZnnAQKAbUAE8CgyoazNNjzJOyVtEpFrzTRU8AfjI3VcDmNkTwJFArpnlRNV3F2B5sjtQ20REpFqsCUvDFgO9zWxXiz9I9HjgXeAl4LRonaHApGRDVfIWEYl4zBNeGtyO+0ziJybnAO8Qz7XFwBXAr81sEdAeGJdsrGqbiIhUq2y+KyzdfTQw+mvDHxJ/NOQ2U/IWEYk05YRluil5i4hUC+fqeCVvEZFqqrxFREKkyltEJDw11z4GQMlbRCQS0IN0lLxFRGooeYuIhEeVt4hIgJS8RUQCpOQtIhIgr7J0h5AwJW8RkYjHlLxFRIKjtomISIDcVXmLiARHlbeISIDU8xYRCVBMs01ERMITUuWd0DMszawkkTERkZC5J76kW4OVt5ntDOwKdDCzPKD6j6W2wB4pjk1EpEWFVHk31jY5F/gV8UQ9m6+S9wZgbArjEhFpcdvNVEF3vw24zcwucve/tlBMIiJpURXQCcuEet7ACjPbHcDMfmdmT5jZoSmMS0Skxblbwku6JZq8R7n7p2bWBzgJGA/clbqwRERanscs4SXdEk3eVdHPU4C73H0S0Do1IYmIpMd2M9uklmVm9jfgBOAWM9uJxBO/iEgQMqGiTlSiCfh04Hmgv7tXAPnAZSmLSkQkDWJuCS/pllDl7e4bzWwV0AcoBSqjnyIi241YQJV3QsnbzEYDPYF9gfuAVsCDwFGpC01EpGVlQkWdqER73j8ADgHmALj78uqpgyIi24tMmAKYqEST95fu7mbmAGa2WwpjEhFJi0yYRZKoRJP3xGi2Sa6Z/QL4OXBv6sKKm7bq3VTvQgK0afm0dIcg26ntrm3i7n8ys37E72myL3C1u09NaWQiIi2sOdsmZpZLvMjtATjxovd9YAKwJ/AxcLq7r0tm+4neEvYWd5/q7pe5+2/cfaqZ3ZLMDkVEMlWVW8JLAm4DnnP3/YBvAwuBK4ESd+8OlETvk5LoPO9+dYwNSHanIiKZqLnmeZtZW+AYYByAu38ZXSMzkPjtRYh+Dko21gaTt5mdZ2bvAPua2du1lo+At5PdqYhIJmrKjanMbLiZvVlrGV5rU3sBq4H7zOwtM7s3mujR2d3L4vvyMqBTsrE21vP+JzAZuIkty/tP3b28+o2Z5SXbtxERyRRNeXi8uxcDxfV8nAMcClzk7jPN7Da2oUVS3w4aCm49sB44s5HtlBAPVEQkWE6znbBcCix195nR+8eIJ++VZlbo7mVmVgisSnYHzXVzqXDm14iI1KPSLeGlIe6+AlhiZvtGQ8cD7wJPAUOjsaHApGRjba6nxwc0tV1EpG7NWHkDXAQ8ZGatgQ+BnxEvmCea2TBgMTA42Y03V/IWEQleU3rejXH3ucTvCfV1xzfH9hubbfKsme2ZwHbUNhGR4DmW8JJujfW87wemmNlIM2vVwHrN8ieJiEg6xZqwpFtjs00mmtkzwNXAm2b2ALXidvcx0c/yejYhIhKMTEjKiUqk570Z+BzYCdidsI5PRCRhVZb+dkiiGkzeZtYfGEN8esuh7r6xRaISEUmDWAb0shPVWOU9Ehjs7gtaIhgRkXQKac5zYz3vo1sqEBGRdAupJ6x53iIikdj20vMWEdmRbDdtExGRHUllOIW3kreISLXtabaJiMgOQ20TEZEAxcIpvJW8RUSqaaqgiEiAqlR5i4iER5W3iEiAlLxFRALUyKMpM4qSt4hIRJW3iEiAlLxFRAKk2SYiIgFS5S0iEiAlbxGRAOneJiIiAdK9TUREAlSV7gCaQMlbRCQSC6hxouQtIhLRCUsRkQCFU3creYuI1FDlLSISIM02EREJUFVAjZOsdAcgIpIpYk1YEmFm2Wb2lpk9Hb3vZmYzzazUzCaYWetkY1XyFhGJxPCElwSNABbWen8LcKu7dwfWAcOSjVXJW0Qk4k1YGmNmXYBTgHuj9wYcBzwWrTIeGJRsrEreIiKRprRNzGy4mb1Zaxn+tc39Bbicr7os7YEKd6+M3i8FipKNVScsRUQiTbnC0t2LgeK6PjOz7wGr3H22mfWtHq5rM02NsZqSt4hIpBnvbXIU8H0zOxnYGWhLvBLPNbOcqPruAixPdgdqm4iIRLwJ/zS4HfffunsXd98TOAN40d3PAl4CTotWGwpMSjZWJW8RkUhzTxWswxXAr81sEfEe+LhkN6TknQEuunAYc98qYd7cF7n4onPqXOfWMdfx3rvTmTN7Kocc3KNmfMiQwSxcMJ2FC6YzZMjglgpZmsnvbhzDMaecwaCf/LJm7PkXpzHwrHM5sM/JzF/4n5rxZWUrOezYgfxw6AX8cOgFXPuHv9a5zfUbPuWcEVdx8o+Gcc6Iq1i/4VMA3J0bb72LAaf/nB+cfR7vvr8otQcXoBRMFcTdX3b370WvP3T3Xu6+j7sPdvcvko1VyTvNDjhgX4YN+zFHHHkKhx7Wj1NOPoF99um2xToD+h9H9326sd/+fTjvvCsYe8dNAOTl5TJq5CUc2ed7HHHUKYwaeQm5ue3ScRiSpEEn9+PuMddvMbbPXt/gLzeO4rBaf0hX61pUyOPjx/L4+LGMvvyiOrd57wMT6d3zYJ6dMI7ePQ9m3IMTAZg2YxaLly7n2QnjuObyi/n9n+5o/gMKXHNOFUw1Je8022+/7sycOYdNm/5LVVUVr057nUED+2+xzqmnnsQDD8Wnhs58Yw7tcttRUNCJE0/8Li+UTGPdugoqKtbzQsk0TjqpbxqOQpLV8+ADadd29y3G9t7zf+j2jS5Jb/OlaTMYOOAEAAYOOIEXX50RH5/+Ot/vfzxmxrd7fItPP/2M1WvKkw9+O1SJJ7ykm5J3mi1Y8B5HH92b/Pw8dtllZwb0P44uXfbYYp2iPQpYuuSrk9LLlpZRtEdBfHxprfFl8XHZfi0rW8FpP72An15wGbPnzq9znbXrKujYIR+Ajh3yKa9YD8DK1Wsp6NShZr3OnTqwcvWa1AcdkOY6YdkS0jJV0Mx+5u73pWPfmea99xbxxz+O5bnJD/P5Z58z7+13qarccsJS/MKsLbl7PeMpC1XSrGP7PKY+8Q9y27VlwXulXPzb65j04N202W23hL7vdfxy1PU7tCML6Zaw6aq8r63vg9pXLcVin7dkTGlz3/2P0Os7/Tn2+B+ybl0FpYs+2uLzpcvK6NL1q2q8qEshy8tWxsdrVelFRYUsL1vRYnFLy2rdujW57doCcMB+3elaVMjHi5dttV77vNyadsjqNeXkR+dBCjp1YMWqryrtlavW0KlD+xaIPBwhVd4pS95m9nY9yztA5/q+5+7F7t7T3XtmZSVWUYSuY8f4/0Bdu+7BoEEDeGTC/23x+dNPT2HIWfGpod/pdSgb1m9gxYpVTJnyCv1OOIbc3Hbk5raj3wnHMGXKKy0ev7SM8nUVVFXF/1a2ZFkZi5csp2tR4Vbr9e3Tm0mTXwBg0uQXOPboI2rGn3quBHdn3vyFtGmzW017ReJaYKpgs0ll26QzcBLxO2fVZsBrKdxvcB6dcA/57fPYvLmSiy8eSUXFeob/YggAxfc8wLOTS+jf/zjeX/hvNm7axDnn/BqAdesquOHGv/D6a88AcP0Nt7JuXUXajkOa7rLRNzPrrbepqNjA8YN+wvnDhtCubRtuuvUuyivWc/5lo9mv+14U33oDs+fO5457HyA7J5vsrCyuvuzCmpOdV9/0F04fdDI9vvVNzhlyOpeOupEnnn6ews4dGXP9SACOOeJwps2YxYDTf84uO+/M76+6JJ2HnpFiAfUdra4+WLNs2GwccJ+7T6/js3+6+48b20ZO66Jw/k1Ki9m0fFq6Q5AM1KrDXtvcwP/xN36QcM755ydPpvWEQcoqb3ev9z61iSRuEZGWlgm97ETpxlQiIpFM6GUnSslbRCTSlMve003JW0QkoraJiEiA1DYREQlQlYeTvpW8RUQi4aRuJW8RkRrqeYuIBEizTUREApSqK85TQclbRCRSpcpbRCQ8apuIiARIbRMRkQCp8hYRCZCmCoqIBCikhzEoeYuIRDTbREQkQOp5i4gESLNNREQCpMpbRCRAmm0iIhIgtU1ERAKkhzGIiAQopJ53VroDEBHJFN6EfxpiZl3N7CUzW2hmC8xsRDSeb2ZTzaw0+pmXbKxK3iIikZh7wksjKoFL3f1bQG/gAjPbH7gSKHH37kBJ9D4pSt4iIpHmqrzdvczd50SvPwUWAkXAQGB8tNp4YFCysarnLSISacoJSzMbDgyvNVTs7sV1rLcncAgwE+js7mUQT/Bm1inZWJW8RUQiTbkxVZSot0rWtZlZG+Bx4FfuvsHMti3AWtQ2ERGJNFfbBMDMWhFP3A+5+xPR8EozK4w+LwRWJRurkreISKS5TlhavMQeByx09zG1PnoKGBq9HgpMSjZWtU1ERCLNeHn8UcAQ4B0zmxuNXQXcDEw0s2HAYmBwsjtQ8hYRiXgzXWHp7tOB+hrcxzfHPpS8RUQiujxeRCRAIV0er+QtIhLRXQVFRAKkBxCLiARID2MQEQmQ2iYiIgHSbBMRkQCp5y0iEiC1TUREAqR53iIiAVLlLSISIJ2wFBEJkE5YiogESG0TEZEA6QpLEZEAqfIWEQlQSMnbQgp2R2Zmw6OnVYvU0O/FjksPIA7H8HQHIBlJvxc7KCVvEZEAKXmLiARIyTsc6mtKXfR7sYPSCUsRkQCp8hYRCZCSt4hIgJS8A2Bm/c3sfTNbZGZXpjseST8z+7uZrTKz+emORdJDyTvDmVk2MBYYAOwPnGlm+6c3KskA9wP90x2EpI+Sd+brBSxy9w/d/UvgEWBgmmOSNHP3V4HydMch6aPknfmKgCW13i+NxkRkB6bknfmsjjHN7xTZwSl5Z76lQNda77sAy9MUi4hkCCXvzDcL6G5m3cysNXAG8FSaYxKRNFPyznDuXglcCDwPLAQmuvuC9EYl6WZmDwMzgH3NbKmZDUt3TNKydHm8iEiAVHmLiARIyVtEJEBK3iIiAVLyFhEJkJK3iEiAlLxFRAKk5C3bFTPbs/o2qWbW08xub2T9q1omMpHmpXneEgQzy3b3qgTW2xN42t17JLjdz9y9zTaGJ9LiVHlLSpjZ781sRK33N5jZxXWs19fMXjWzJ83sXTO728yyos8+M7PrzGwmcISZHWZmr5jZbDN73swKo/UOM7N5ZjYDuOBr2346et3GzO4zs3fM7G0z+6GZ3QzsYmZzzeyhFP8rEWlWSt6SKuOAoQBRMj4DqC9B9gIuBQ4E9gb+NxrfDZjv7t8BZgJ/BU5z98OAvwM3ROvdB1zs7kc0EM8oYL27H+juBwEvuvuVwCZ3P9jdz0ryOEXSIifdAcj2yd0/NrO1ZnYI0Bl4y93X1rP6G+7+IdTcs6MP8BhQBTwerbMv0AOYamYA2UCZmbUDct39lWi9B4g/dejrTiD+B0h1fOu25fhE0k3JW1LpXuCnQAHxSrk+Xz/xUv3+v7X63AYs+Hp1bWa5dXy/LpbgeiJBUNtEUulJ4s9ZPJz4XRHr0yu65W0W8CNgeh3rvA90NLMjAMyslZkd4O4VwHoz6xOtV1/7YwrxuzMSfT8vernZzFolfEQiGULJW1ImeubmS8RvY9vQTJEZwM3AfOAj4km/rm2dBtxiZvOAucCR0cc/A8ZGJyw31bOP64E8M5sfff/YaLwYeFsnLCU0miooKRNV0nOAwe5eWs86fYHfuPv3WjI2kdCp8paUMLP9gUVASX2JW0SSp8pbWoSZHUh8JkhtX0TTAEWkiZS8RUQCpLaJiEiAlLxFRAKk5C0iEiAlbxGRAP0/qyxwTJDH4d0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "## function to get confusion matrix in a proper format\n",
    "\n",
    "# Uncommment below to answer\n",
    "\n",
    "def draw_cm( actual, predicted ):\n",
    "     cm = confusion_matrix(Y_test,y_predict)\n",
    "     sns.heatmap(cm, annot=True,  fmt='.2f', xticklabels = [0,1] , yticklabels = [0,1] )\n",
    "     plt.ylabel('Y_test')\n",
    "     plt.xlabel('y_predict')\n",
    "     plt.show()\n",
    "\n",
    "#Call function here - \n",
    "draw_cm(Y_test,y_predict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "id": "d5AgHy3cZVdq"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.75      0.44      0.56        61\n",
      "           1       0.77      0.93      0.84       124\n",
      "\n",
      "    accuracy                           0.77       185\n",
      "   macro avg       0.76      0.69      0.70       185\n",
      "weighted avg       0.76      0.77      0.75       185\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(Y_test,y_predict))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "id": "rZ1ng7Cwe2EZ"
   },
   "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>Col</th>\n",
       "      <th>Coeff</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Credit_History</td>\n",
       "      <td>1.66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Property_Area_Semiurban</td>\n",
       "      <td>0.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Education_Not Graduate</td>\n",
       "      <td>0.46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Married_Yes</td>\n",
       "      <td>0.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Dependents_2</td>\n",
       "      <td>0.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Gender_Male</td>\n",
       "      <td>0.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Dependents_0</td>\n",
       "      <td>0.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Dependents_1</td>\n",
       "      <td>0.12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Property_Area_Urban</td>\n",
       "      <td>0.12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Self_Employed_Yes</td>\n",
       "      <td>0.04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Dependents_3+</td>\n",
       "      <td>0.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ApplicantIncome</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CoapplicantIncome</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LoanAmount</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                        Col  Coeff\n",
       "3            Credit_History   1.66\n",
       "12  Property_Area_Semiurban   0.59\n",
       "10   Education_Not Graduate   0.46\n",
       "5               Married_Yes   0.33\n",
       "8              Dependents_2   0.32\n",
       "4               Gender_Male   0.25\n",
       "6              Dependents_0   0.19\n",
       "7              Dependents_1   0.12\n",
       "13      Property_Area_Urban   0.12\n",
       "11        Self_Employed_Yes   0.04\n",
       "9             Dependents_3+   0.03\n",
       "0           ApplicantIncome   0.00\n",
       "1         CoapplicantIncome   0.00\n",
       "2                LoanAmount   0.00"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Coefficients \n",
    "# Uncomment below to answer\n",
    "fi = pd.DataFrame()\n",
    "fi['Col'] = X.columns\n",
    "fi['Coeff'] = np.round(abs(logreg.coef_[0]),2)\n",
    "fi.sort_values(by='Coeff',ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "8geTPobgFc8k"
   },
   "source": [
    "### kNN model Building (10 marks)\n",
    "\n",
    "- For every value of K, split the train set into new train and validation set 30 times and take the average of the error on these train and the validation sets per k. Create a dictionary that stores the average train and validation error for all Ks from 1 to 14.  (2 marks)\n",
    "\n",
    "- Use the dictionary to plot the average train vs the validation set error for all ks. (2 mark) \n",
    "\n",
    "- Choose the apt k from the plot where the two errors are comparable and fit the entire training set and predict on the test set. (1 mark)\n",
    "\n",
    "- Store the results of the test set in the dataframe created above. (1 mark)\n",
    "\n",
    "- Draw confusion matrix calling the function created above.(1 marks)\n",
    "\n",
    "- print the classification report(1 mark) \n",
    "- Provide business insights for the entire modelling process (including logistic regression)(2 marks)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "id": "D60Li_GdFmey"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{1: [0.0, 0.3643410852713177],\n",
       " 2: [0.1757045675413022, 0.41201550387596897],\n",
       " 3: [0.19222546161321669, 0.2945736434108526],\n",
       " 4: [0.2010689990281827, 0.315891472868217],\n",
       " 5: [0.21438289601554913, 0.2837209302325582],\n",
       " 6: [0.22264334305150632, 0.3073643410852713],\n",
       " 7: [0.23488824101068997, 0.3093023255813953],\n",
       " 8: [0.23965014577259475, 0.29573643410852707],\n",
       " 9: [0.25364431486880473, 0.31279069767441847],\n",
       " 10: [0.2552964042759961, 0.2992248062015504],\n",
       " 11: [0.26433430515063167, 0.3255813953488371],\n",
       " 12: [0.262779397473275, 0.30813953488372087],\n",
       " 13: [0.27842565597667635, 0.30077519379844964],\n",
       " 14: [0.2666666666666667, 0.3081395348837209]}"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "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",
    "# Uncomment below and fill in the blanks\n",
    "\n",
    "for k in range(1,15):\n",
    "    train_error = []\n",
    "    test_error = []\n",
    "    lista = []\n",
    "    knn = KNeighborsClassifier(n_neighbors=k)\n",
    "    for i in range(30):\n",
    "        x_train, x_val, y_train, y_val = train_test_split(X_train, Y_train, test_size = 0.20)\n",
    "     #let us scale the data. Notice the scaling done in the videos.\n",
    "        x_train = ss.fit_transform(x_train)\n",
    "        x_val = ss.transform(x_val) # transform the validation set\n",
    "        knn.fit(x_train, y_train)\n",
    "        train_error.append(1 - knn.score(x_train, y_train)) \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": 52,
   "metadata": {
    "id": "JY1X9z5UMpE-"
   },
   "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": [
    "# Extracting k and the test error 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 error\n",
    "\n",
    "#Uncomment and answer\n",
    "plt.figure(figsize=(10,10))\n",
    "plt.title('k vs Error Rate')\n",
    "plt.plot(kltest,vltest, label = 'test' )\n",
    "plt.plot(kltrain,vltrain, label = 'train')\n",
    "plt.xlabel('K')\n",
    "plt.ylabel('Error Rate')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "id": "50RL5C2yM8WS"
   },
   "outputs": [],
   "source": [
    "#Uncomment and answer\n",
    "\n",
    "# here it was writen 5 in lower lines so I used k = 5 and will do another one with K= 9\n",
    "#K=9 will be a better fit model which can generalize better.\n",
    "\n",
    "knn = KNeighborsClassifier(n_neighbors =5)\n",
    "\n",
    "knn.fit(X_train, Y_train)\n",
    "\n",
    "Y_pred = knn.predict(X_test)\n",
    "z[\"5NN predictions\"]  = Y_pred\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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>ApplicantIncome</th>\n",
       "      <th>CoapplicantIncome</th>\n",
       "      <th>LoanAmount</th>\n",
       "      <th>Credit_History</th>\n",
       "      <th>Gender_Male</th>\n",
       "      <th>Married_Yes</th>\n",
       "      <th>Dependents_0</th>\n",
       "      <th>Dependents_1</th>\n",
       "      <th>Dependents_2</th>\n",
       "      <th>Dependents_3+</th>\n",
       "      <th>Education_Not Graduate</th>\n",
       "      <th>Self_Employed_Yes</th>\n",
       "      <th>Property_Area_Semiurban</th>\n",
       "      <th>Property_Area_Urban</th>\n",
       "      <th>Observed Loan Status</th>\n",
       "      <th>Predicted Loan Status</th>\n",
       "      <th>5NN predictions</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>220</th>\n",
       "      <td>2221</td>\n",
       "      <td>0.0</td>\n",
       "      <td>60</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>2071</td>\n",
       "      <td>754.0</td>\n",
       "      <td>94</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>569</th>\n",
       "      <td>3166</td>\n",
       "      <td>2064.0</td>\n",
       "      <td>104</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>330</th>\n",
       "      <td>4384</td>\n",
       "      <td>1793.0</td>\n",
       "      <td>117</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>407</th>\n",
       "      <td>2213</td>\n",
       "      <td>0.0</td>\n",
       "      <td>66</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>563</th>\n",
       "      <td>5800</td>\n",
       "      <td>0.0</td>\n",
       "      <td>132</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>3200</td>\n",
       "      <td>2254.0</td>\n",
       "      <td>126</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>319</th>\n",
       "      <td>10000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>155</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>612</th>\n",
       "      <td>7583</td>\n",
       "      <td>0.0</td>\n",
       "      <td>187</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>281</th>\n",
       "      <td>3927</td>\n",
       "      <td>800.0</td>\n",
       "      <td>112</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>185 rows × 17 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     ApplicantIncome  CoapplicantIncome  LoanAmount  Credit_History  \\\n",
       "220             2221                0.0          60             0.0   \n",
       "109             2071              754.0          94             1.0   \n",
       "569             3166             2064.0         104             0.0   \n",
       "330             4384             1793.0         117             1.0   \n",
       "407             2213                0.0          66             1.0   \n",
       "..               ...                ...         ...             ...   \n",
       "563             5800                0.0         132             1.0   \n",
       "66              3200             2254.0         126             0.0   \n",
       "319            10000                0.0         155             1.0   \n",
       "612             7583                0.0         187             1.0   \n",
       "281             3927              800.0         112             1.0   \n",
       "\n",
       "     Gender_Male  Married_Yes  Dependents_0  Dependents_1  Dependents_2  \\\n",
       "220            1            1             1             0             0   \n",
       "109            1            1             0             0             0   \n",
       "569            1            1             1             0             0   \n",
       "330            1            0             0             1             0   \n",
       "407            0            0             1             0             0   \n",
       "..           ...          ...           ...           ...           ...   \n",
       "563            1            0             1             0             0   \n",
       "66             1            0             1             0             0   \n",
       "319            1            1             0             1             0   \n",
       "612            1            1             0             0             1   \n",
       "281            1            1             1             0             0   \n",
       "\n",
       "     Dependents_3+  Education_Not Graduate  Self_Employed_Yes  \\\n",
       "220              0                       0                  0   \n",
       "109              1                       1                  0   \n",
       "569              0                       0                  0   \n",
       "330              0                       0                  0   \n",
       "407              0                       1                  0   \n",
       "..             ...                     ...                ...   \n",
       "563              0                       1                  1   \n",
       "66               0                       1                  0   \n",
       "319              0                       0                  1   \n",
       "612              0                       0                  0   \n",
       "281              0                       0                  0   \n",
       "\n",
       "     Property_Area_Semiurban  Property_Area_Urban Observed Loan Status  \\\n",
       "220                        0                    1                    0   \n",
       "109                        1                    0                    1   \n",
       "569                        0                    1                    0   \n",
       "330                        0                    1                    1   \n",
       "407                        0                    0                    1   \n",
       "..                       ...                  ...                  ...   \n",
       "563                        1                    0                    1   \n",
       "66                         0                    1                    0   \n",
       "319                        0                    0                    0   \n",
       "612                        0                    1                    1   \n",
       "281                        1                    0                    1   \n",
       "\n",
       "     Predicted Loan Status  5NN predictions  \n",
       "220                      0                0  \n",
       "109                      1                1  \n",
       "569                      0                1  \n",
       "330                      1                1  \n",
       "407                      1                0  \n",
       "..                     ...              ...  \n",
       "563                      1                1  \n",
       "66                       0                1  \n",
       "319                      1                1  \n",
       "612                      1                1  \n",
       "281                      1                1  \n",
       "\n",
       "[185 rows x 17 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "z"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "id": "YCTHmdx5d3E2"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7ff06b05ccd0>"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWAAAAD4CAYAAADSIzzWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAATyklEQVR4nO3deZxVdf3H8deHLbeURRQYXDApbFUjo3LBLcFSqDTNMi2MVtO00iz1l6lpi1q/rH6Ta0aC2UL1E5dQUysRRXMJS364DY5gIaCJAXe+vz/mioMzMHeGO/fLPbyePM5j5iz33A8yvvnyOd9zbqSUkCTVXq/cBUjSxsoAlqRMDGBJysQAlqRMDGBJyqRPj79BvwanWaidKYPG5i5BG6AjmqfE+p5j5T/nV5w5fbfeab3fb304ApakTHp8BCxJNdVSyl1BxQxgScVSWpW7gooZwJIKJaWW3CVUzACWVCwtBrAk5eEIWJIy8SKcJGXiCFiS8kjOgpCkTLwIJ0mZ2IKQpEy8CCdJmTgClqRMvAgnSZl4EU6S8kjJHrAk5WEPWJIysQUhSZk4ApakTEorc1dQMQNYUrHYgpCkTOqoBeGnIksqlpaWypdORMRlEbEoIh5ss21gRNwUEY+Uvw4ob4+I+H5EzIuI+yNi987ObwBLKpYqBjBwBTDuFdtOBWamlEYCM8vrAOOBkeVlMvCjzk5uAEsqlFRaWfHS6blSug1Y/IrNE4Ary99fCUxss/2nqdWdQP+IGLqu8xvAkooltVS8RMTkiLi7zTK5gnfYNqXUDFD+uk15ewPwZJvjmsrb1sqLcJKKpQuzIFJKjUBjld45OnqLdb3AAJZULD0/C2JhRAxNKTWXWwyLytubgO3aHDcceGpdJ7IFIalYqnsRriO/BY4pf38MML3N9o+WZ0OMAZa+1KpYG0fAkoqliiPgiLgaGAtsHRFNwJnAecA1ETEJeAI4vHz4dcDBwDzgBeBjnZ3fAJZULKuq90D2lNKH1rJr/w6OTcBnu3J+A1hSsdTRnXAGsKRi8VkQkpSJI2BJysQRsCRl4ghYkjKp4iyInmYASyqWtM67fzcoBrCkYrEHLEmZGMCSlIkX4SQpk1IpdwUVM4AlFYstCEnKxACWpEzsAUtSHqnFecCSlIctCEnKxFkQkpSJI2BJysQA3rgd/7lJTJp0FBHBpZf+nO//9yXtjrnwgrMYP24/Xli+nEmTvsC99z2YoVL1tPfedRErn3+RVGohlUrcNO503nL6hxj27t1pWbGK5x9fyF0nNrJy2QvtXjtk3zez21lHE717Mf/nt/LwD35X+99APfJhPBuvN7zhdUyadBTveOd7WLFiJdf9fgrXzZjJvHmPrj5m/Lj9GLnzCEa9fk/evsfuXPyDb/LOPQ/JWLV60i2Hnc2Kxc+vXn/6tge5/9xppFILb/7qkexy/KHcf87UNV4TvYK3nnsstx7xTZY3L+bAGd/gqRvnsOwfC2pdfv2poxFwr84OiIhREXFKRHw/Ir5X/n6XWhRXj0aNGsmsWXNYvvxFSqUSt91+JxMnjFvjmEMOOYirplwLwKy75rBV/60YMmSbHOUqg4V/fIBUag2Jf82Zx2bDBrY7ZuBur+G5xxby7yeeoWVliSem30nDQW+tdan1qSVVvmS2zgCOiFOAqUAAdwGzy99fHRGn9nx59eehhx5mr73GMHDgADbddBPGj9uP4cOHrXFMw7AhND351Or1BU3NNAwbUutSVQMpJcZOPZUDbzibnT6yb7v9I47ch+ab/9pu+6ZDBrJ8wb9Wr7/QvJhNhwzo0VoLo1SqfMmssxbEJOANKaWVbTdGxAXAQ8B5Hb0oIiYDkwGi91b06rV5FUqtDw8/PI9vf/tirp9xNf9+/t/89f6/UVq15h90RLR7XaqjvpUqN/PQr/PiwiW8atCWjJ12Ks/Na+aZOx8GYJcTJpBKJR7/5Z/av7D9j0hd9TZzSgVqQbQAwzrYPrS8r0MppcaU0uiU0uiNKXxfcvkVU9nj7ePYd/8P8OyzS3ikTf8XoGlBM8O3e/k/a8PwoTzVvLDWZaoGXly4BID//GsZTTPuZuCuOwGw4+F7MeyA3bjzsz/s8HXLmxezacOg1eubDR3I8vK51ImitCCAE4GZETEjIhrLy/XATOCEni+vPg0e3Po/znbbDWPixPFMnfabNfb//vc3cvSHDwPg7XvszrKly3j66UU1r1M9q/emr6LP5pus/n7IPm9i6d+bGLLvmxn1uUO449jvUlq+osPXLr5vPq8eMYTNtxtMr7692X7CGBbccE8ty69fqaXyJbN1tiBSStdHxGuBPYAGWv9h1ATMTinlb6BsoH4x7ScMHDSAlStX8fnPf5UlS5Yy+RNHA9D4k6u4bsZMxo3bj7/P/RMvLF/OccedlLli9YRNBm/Jnpd9AYDo05vHf/1nnr7lfg7+83fp3a8v+0z9CtB6Ie6eUy5jk23787bvfoLbP/JtUqmFOaddwT5Xn9I6DW3qH50BUakNYGRbqejp3mOffg31819DNTNl0NjcJWgDdETzlI66313y7zOOrDhzNj9r6nq/3/pwHrCkYtkAWguVMoAlFUsdtSAMYEmFUk/T0AxgScXiCFiSMqmjAO70WRCSVFeqeCtyRHwhIh6KiAcj4uqI2CQiRkTErIh4JCKmRUS/7pZqAEsqlNSSKl7WJSIagM8Do1NKbwR6A0cC5wMXppRGAs/S+siGbjGAJRVLdW9F7gNsGhF9gM2AZmA/4Nry/iuBid0t1QCWVCwtLRUvETE5Iu5us0x+6TQppQXAd4AnaA3epcA9wJKU0qryYU203iXcLV6Ek1QsXbgIl1JqBBo72hcRA4AJwAhgCfALYHxHp+l6ka0MYEnFUr1ZEAcAj6aUngGIiF8B7wT6R0Sf8ih4OPDUOs6xTrYgJBVK6+fvVbZ04glgTERsFq0P8d4f+BtwC3BY+ZhjgOndrdUAllQsVboIl1KaRevFtjnAA7TmZSNwCnBSRMwDBgGXdrdUWxCSCqWz6WVdOldKZwJnvmLzfFof0bveDGBJxVJHd8IZwJKKpX6exWMASyqWtKp+EtgAllQs9ZO/BrCkYqnmRbieZgBLKhZHwJKUhyNgScrFEbAk5bH6OWV1wACWVCh19Kn0BrCkgjGAJSkPR8CSlIkBLEmZpFLkLqFiBrCkQnEELEmZpBZHwJKUhSNgScokJUfAkpSFI2BJyqTFWRCSlIcX4SQpEwNYkjJJ9fM4YANYUrE4ApakTJyGJkmZlJwFIUl5OAKWpEzsAUtSJs6CkKRMHAFLUialll65S6iYASypUGxBSFImLXU0C6J+xuqSVIGUouKlMxHRPyKujYiHI2JuRLwjIgZGxE0R8Uj564Du1moASyqUlCpfKvA94PqU0ijgLcBc4FRgZkppJDCzvN4tPd6C2KRPv55+C9Wh9z/wjdwlqKCq1YKIiC2BvYFjAVJKK4AVETEBGFs+7ErgVuCU7ryHI2BJhVJq6VXxEhGTI+LuNsvkNqfaCXgGuDwi7o2ISyJic2DblFIzQPnrNt2t1YtwkgqlK5MgUkqNQONadvcBdgeOTynNiojvsR7tho44ApZUKC0pKl460QQ0pZRmldevpTWQF0bEUIDy10XdrdUAllQo1ZoFkVJ6GngyIl5X3rQ/8Dfgt8Ax5W3HANO7W6stCEmFUuUPRT4emBIR/YD5wMdoHbheExGTgCeAw7t7cgNYUqEkqncjRkrpPmB0B7v2r8b5DWBJhbKqju6EM4AlFUo1R8A9zQCWVChV7gH3KANYUqE4ApakTBwBS1ImJUfAkpRHHX0ikQEsqVhaHAFLUh519IlEBrCkYvEinCRl0hK2ICQpi1LuArrAAJZUKM6CkKRMnAUhSZk4C0KSMrEFIUmZOA1NkjIpOQKWpDwcAUtSJgawJGVSRx8JZwBLKhZHwJKUibciS1ImzgOWpExsQUhSJgawJGXisyAkKRN7wJKUibMgJCmTljpqQhjAkgrFi3CSlEn9jH8NYEkFU08j4F65C5CkaloVqeKlEhHROyLujYjfl9dHRMSsiHgkIqZFRL/u1moASyqU1IWlQicAc9usnw9cmFIaCTwLTOpurQawpEJp6cLSmYgYDrwHuKS8HsB+wLXlQ64EJna3VgNYUqG0kCpeImJyRNzdZpn8itNdBHyZl/N6ELAkpbSqvN4ENHS3Vi/CSSqUrsyCSCk1Ao0d7YuI9wKLUkr3RMTYlzav51uuwQCWVChVnAXxLuDQiDgY2ATYktYRcf+I6FMeBQ8HnuruG9iCkFQoJVLFy7qklL6SUhqeUtoROBK4OaX0YeAW4LDyYccA07tbqwEsqVCqeRFuLU4BToqIebT2hC/t7olsQUgqlNQD98KllG4Fbi1/Px/YoxrnNYAlFYp3wm1kfvjj83n0sdncNfv61dvOPucrzLn3D9w5awZXT/0xW2316g5fe8CBezPnvpn89YFbOOnkT9WqZPWQr517AXu/50gmfuTlP8sbbr6dCR/+JG/a82AenPuP1dv/fNccPvjx43nf0Z/mgx8/nln33NfhOZcue47jTjiNg4+YxHEnnMbSZc/1+O+jnnVlGlpuBnAVTLnql0yceOwa226++Q7eNvogxrx9PI888ignf/Ez7V7Xq1cvLrjwLN4/8VhG7/5uDj/8UEaN2rlGVasnTDz4QH58wdlrbNt5px246NzTeeuub1xj+4D+W/KD8/+LX1/1I8752sl85azvdHjOS666hjGjd+W6aZcyZvSuXPqza3qs/iLogTvheowBXAV/+tNdPLt4yRrbbp55O6VS66OhZ8++l4aGIe1eN3r0W5j/f4/z2GNPsnLlSq699ne8570H1qRm9YzRu76JrbZc8187r9lxe0bsMLzdsbu8dme2GTwIgJ1H7MB/VqxgxYoV7Y675fa/MGH8AQBMGH8AN9/2lx6ovDhWkSpecjOAa+Doj36QG2/8Y7vtw4YNoWlB8+r1BQueZtiw9kGt4rvp1jvY5bWvoV+/9s91+dezSxi89UAABm89kMVLlta6vLqSuvArt24HcER8bB37Vt/et3LVxt2v+tKXP0tp1SqmTf1Nu32tt5WvKaX8PxSqrXnzH+eCH17GGV86PncphVCDaWhVsz4j4K+vbUdKqTGlNDqlNLpvn44vPm0Mjvrw+xk3fj8+/rETO9y/YEEzwxuGrl5vaBhCc/PCWpWnDcDTi57hhNO+wbmnf5Hthw/r8JhBA/rzzD8XA/DMPxczsP9WtSyx7hRmBBwR969leQDYtkY11qUDDtybk076FEcc/gmWL3+xw2Puued+XrPzjuyww3D69u3LYYcdwnX/+4caV6pclj33PJ/50pmc+Mlj2f3Nb1jrcWP3HMP0Ga0/F9Nn/IF993pHrUqsS/U0Ao51/ZM3IhYCB9H6zMs1dgF/Til1/Fd2G1tsNiL/XzM97PIrvsdee49h0KABLFr0T845+yJO/uKnedWr+rG4fHFu9l33csLnv8aQodtw8Q/P4wPv+zgA7z5oLOd/6wx69+7FVT/9Bd/+1sU5fys18+wTM3OX0CO+dOZ5zL73fpYsWcaggf35zKSj2WrLLfjmhT9i8ZKlvHqLLRg1cicaLzyH/7niai65ahrbD3/5YVqNF53DoAH9OeObF/HBiQfzxl1ey5Klyzj59HNpXvgMQ7cdzAVnf7Xdhb6i6Lv1Tuv9ofIf2eH9FWfOzx7/VdYPse8sgC8FLk8p3dHBvp+nlI7q7A02hgBW1xU1gLV+qhHAR+3wvooz5+eP/zprAK/zTriU0lqf9F5J+EpSrW0Ivd1KeSuypELZEHq7lTKAJRXKhnCLcaUMYEmFYgtCkjIp1dHNTAawpEKxBSFJmXgRTpIysQcsSZnYgpCkTOrpiYIGsKRC6ezj5jckBrCkQrEFIUmZ2IKQpEwcAUtSJk5Dk6RMvBVZkjKxBSFJmRjAkpSJsyAkKRNHwJKUibMgJCmTUqqfB1IawJIKpZ56wL1yFyBJ1dRCqnhZl4jYLiJuiYi5EfFQRJxQ3j4wIm6KiEfKXwd0t1YDWFKhpC786sQq4OSU0i7AGOCzEfF64FRgZkppJDCzvN4tBrCkQmlJqeJlXVJKzSmlOeXvnwPmAg3ABODK8mFXAhO7W6s9YEmF0hOzICJiR2A3YBawbUqpGVpDOiK26e55DWBJhdKVWRARMRmY3GZTY0qp8RXHbAH8EjgxpbQsIqpSJxjAkgqms9ZCW+WwbVzb/ojoS2v4Tkkp/aq8eWFEDC2PfocCi7pbqz1gSYVSrYtw0TrUvRSYm1K6oM2u3wLHlL8/Bpje3VodAUsqlK6MgDvxLuBo4IGIuK+87TTgPOCaiJgEPAEc3t03MIAlFUq1LsKllO4A1tbw3b8a72EASyqUUirlLqFiBrCkQqmnW5ENYEmF4uMoJSkTR8CSlEkVZ0H0OANYUqH4QHZJysQHsktSJvaAJSkTe8CSlIkjYEnKxHnAkpSJI2BJysRZEJKUiRfhJCkTWxCSlIl3wklSJo6AJSmTeuoBRz39bVHvImLyKz/yWvLnYuPlpyLX1uTcBWiD5M/FRsoAlqRMDGBJysQAri37fOqIPxcbKS/CSVImjoAlKRMDWJIyMYBrJCLGRcTfI2JeRJyaux7lFxGXRcSiiHgwdy3KwwCugYjoDVwMjAdeD3woIl6ftyptAK4AxuUuQvkYwLWxBzAvpTQ/pbQCmApMyFyTMksp3QYszl2H8jGAa6MBeLLNelN5m6SNmAFcG9HBNuf/SRs5A7g2moDt2qwPB57KVIukDYQBXBuzgZERMSIi+gFHAr/NXJOkzAzgGkgprQI+B9wAzAWuSSk9lLcq5RYRVwN/AV4XEU0RMSl3Taotb0WWpEwcAUtSJgawJGViAEtSJgawJGViAEtSJgawJGViAEtSJv8PeqfUlHSNvaMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Your code here for drawing the confusion matrix\n",
    "sns.heatmap(confusion_matrix(Y_test,Y_pred),annot=True,fmt='0.1f')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "id": "mJgMgXJzbZtf"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.43      0.15      0.22        61\n",
      "           1       0.68      0.90      0.78       124\n",
      "\n",
      "    accuracy                           0.65       185\n",
      "   macro avg       0.56      0.53      0.50       185\n",
      "weighted avg       0.60      0.65      0.59       185\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Print the classification report here\n",
    "print(classification_report( Y_test,Y_pred))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "knn = KNeighborsClassifier(n_neighbors =9)\n",
    "\n",
    "knn.fit(X_train, Y_train)\n",
    "\n",
    "Y_pred9 = knn.predict(X_test)\n",
    "z[\"9NN predictions\"]  = Y_pred9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7ff06b51a050>"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.heatmap(confusion_matrix(Y_test,Y_pred9),annot=True,fmt='0.1f')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.64      0.11      0.19        61\n",
      "           1       0.69      0.97      0.81       124\n",
      "\n",
      "    accuracy                           0.69       185\n",
      "   macro avg       0.66      0.54      0.50       185\n",
      "weighted avg       0.67      0.69      0.60       185\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report( Y_test,Y_pred9))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "n-jzsgfue2Ea"
   },
   "source": [
    "## <a id = \"link8\"></a>Business Insights "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "fRmpC9rHe2Ea"
   },
   "source": [
    "*Add insights here and comment on the best model*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "pSGnAL-fFTio"
   },
   "source": [
    "### What Confusion matrix means**\n",
    "\n",
    "**True Positive (observed=1,predicted=1):**\n",
    "\n",
    "Employee left and the model predicted he/she will\n",
    "\n",
    "**False Positive (observed=0,predicted=1):**\n",
    "\n",
    "Employee didn't leave and the model predicted he/she will\n",
    "\n",
    "**True Negative (observed=0,predicted=0):**\n",
    "\n",
    "Employee didn't leave and the model predicted he/she won't\n",
    "\n",
    "**False Negative (observed=1,predicted=0):**\n",
    "\n",
    "Employee left and the model predicted he/she won't\n",
    "\n",
    "**Logistic Regression **\n",
    "\n",
    "TP = 115 , FN = 9 , TN = 27 , FP = 34\n",
    "recall = 0.93  , precision = 0.77\n",
    "\n",
    "**5NN prediction** \n",
    "\n",
    "TP = 112 , FN = 12 ,TN = 9 , FP = 52\n",
    "recall = 0.9 , precision = 0.68\n",
    "\n",
    "**9NN prediction**\n",
    "\n",
    "TP = 120 , FN = 4 ,TN = 7 , FP = 54\n",
    "recall = 0.97  , precision = 0.69\n",
    "\n",
    "\n",
    "*Logistic regression model has a higher TN rate than KNN, has higher rate of TP than 5NN  but lower than 9NN. \n",
    "\n",
    "*Logistic regression model has lower FN rate than 5NN but higher than 9NN, has lower rate of FP than both. \n",
    "\n",
    "**Overall I think Logistic regression model is a better choice for this dataset.less FP -> higher precision .  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [
    "8hWiWiyQB_B8",
    "HLCxLTnDe2DZ",
    "Njqqbxmre2Du"
   ],
   "name": "Project: Classification: Loan Eligibility Prediction - Learner's Notebook.ipynb",
   "provenance": [],
   "toc_visible": true
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}