Logistic Regression in Python- Predict the Probability of Default of an Individual.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Data Understanding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn import preprocessing\n",
    "import matplotlib.pyplot as plt \n",
    "plt.rc(\"font\", size=14)\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import train_test_split\n",
    "import seaborn as sns\n",
    "sns.set(style=\"white\")\n",
    "sns.set(style=\"whitegrid\", color_codes=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(41188, 10)\n",
      "['loan_applicant_id', 'age', 'education', 'years_with_current_employer', 'years_at_current_address', 'household_income', 'debt_to_income_ratio', 'credit_card_debt', 'other_debt', 'y']\n"
     ]
    }
   ],
   "source": [
    "data=pd.read_csv(\"bank.csv\")\n",
    "data = data.dropna()\n",
    "print(data.shape)\n",
    "print(list(data.columns))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>loan_applicant_id</th>\n",
       "      <th>age</th>\n",
       "      <th>education</th>\n",
       "      <th>years_with_current_employer</th>\n",
       "      <th>years_at_current_address</th>\n",
       "      <th>household_income</th>\n",
       "      <th>debt_to_income_ratio</th>\n",
       "      <th>credit_card_debt</th>\n",
       "      <th>other_debt</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>191</td>\n",
       "      <td>44</td>\n",
       "      <td>university.degree</td>\n",
       "      <td>10</td>\n",
       "      <td>20</td>\n",
       "      <td>192</td>\n",
       "      <td>12.116645</td>\n",
       "      <td>14.377313</td>\n",
       "      <td>8.886645</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>34318</td>\n",
       "      <td>34</td>\n",
       "      <td>high.school</td>\n",
       "      <td>3</td>\n",
       "      <td>18</td>\n",
       "      <td>57</td>\n",
       "      <td>14.264229</td>\n",
       "      <td>5.137880</td>\n",
       "      <td>2.992730</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>14932</td>\n",
       "      <td>45</td>\n",
       "      <td>university.degree</td>\n",
       "      <td>14</td>\n",
       "      <td>24</td>\n",
       "      <td>212</td>\n",
       "      <td>7.285681</td>\n",
       "      <td>10.460306</td>\n",
       "      <td>4.985339</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2776</td>\n",
       "      <td>33</td>\n",
       "      <td>illiterate</td>\n",
       "      <td>12</td>\n",
       "      <td>5</td>\n",
       "      <td>418</td>\n",
       "      <td>11.386272</td>\n",
       "      <td>3.040189</td>\n",
       "      <td>44.554429</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11915</td>\n",
       "      <td>20</td>\n",
       "      <td>basic</td>\n",
       "      <td>4</td>\n",
       "      <td>19</td>\n",
       "      <td>122</td>\n",
       "      <td>28.418494</td>\n",
       "      <td>14.560450</td>\n",
       "      <td>20.110112</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   loan_applicant_id  age          education  years_with_current_employer  \\\n",
       "0                191   44  university.degree                           10   \n",
       "1              34318   34        high.school                            3   \n",
       "2              14932   45  university.degree                           14   \n",
       "3               2776   33         illiterate                           12   \n",
       "4              11915   20              basic                            4   \n",
       "\n",
       "   years_at_current_address  household_income  debt_to_income_ratio  \\\n",
       "0                        20               192             12.116645   \n",
       "1                        18                57             14.264229   \n",
       "2                        24               212              7.285681   \n",
       "3                         5               418             11.386272   \n",
       "4                        19               122             28.418494   \n",
       "\n",
       "   credit_card_debt  other_debt  y  \n",
       "0         14.377313    8.886645  1  \n",
       "1          5.137880    2.992730  0  \n",
       "2         10.460306    4.985339  0  \n",
       "3          3.040189   44.554429  1  \n",
       "4         14.560450   20.110112  0  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['university.degree', 'high.school', 'illiterate', 'basic',\n",
       "       'professional.course'], dtype=object)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['education'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    36548\n",
       "1     4640\n",
       "Name: y, dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['y'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(x='y', data=data, palette='hls')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m percentage of no default is 88.73458288821988\n",
      "\u001b[1m percentage of default 11.265417111780131\n"
     ]
    }
   ],
   "source": [
    "count_no_default = len(data[data['y']==0])\n",
    "count_default = len(data[data['y']==1])\n",
    "pct_of_no_default = count_no_default/(count_no_default+count_default)\n",
    "print(\"\\033[1m percentage of no default is\", pct_of_no_default*100)\n",
    "pct_of_default = count_default/(count_no_default+count_default)\n",
    "print(\"\\033[1m percentage of default\", pct_of_default*100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>loan_applicant_id</th>\n",
       "      <th>age</th>\n",
       "      <th>years_with_current_employer</th>\n",
       "      <th>years_at_current_address</th>\n",
       "      <th>household_income</th>\n",
       "      <th>debt_to_income_ratio</th>\n",
       "      <th>credit_card_debt</th>\n",
       "      <th>other_debt</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>y</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></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>22914.5</td>\n",
       "      <td>37.975676</td>\n",
       "      <td>13.447521</td>\n",
       "      <td>15.497729</td>\n",
       "      <td>128.300646</td>\n",
       "      <td>15.514879</td>\n",
       "      <td>7.810996</td>\n",
       "      <td>12.127154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2320.5</td>\n",
       "      <td>38.261207</td>\n",
       "      <td>14.356034</td>\n",
       "      <td>14.500000</td>\n",
       "      <td>229.554957</td>\n",
       "      <td>21.810028</td>\n",
       "      <td>23.489629</td>\n",
       "      <td>26.605214</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   loan_applicant_id        age  years_with_current_employer  \\\n",
       "y                                                              \n",
       "0            22914.5  37.975676                    13.447521   \n",
       "1             2320.5  38.261207                    14.356034   \n",
       "\n",
       "   years_at_current_address  household_income  debt_to_income_ratio  \\\n",
       "y                                                                     \n",
       "0                 15.497729        128.300646             15.514879   \n",
       "1                 14.500000        229.554957             21.810028   \n",
       "\n",
       "   credit_card_debt  other_debt  \n",
       "y                                \n",
       "0          7.810996   12.127154  \n",
       "1         23.489629   26.605214  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.groupby('y').mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>loan_applicant_id</th>\n",
       "      <th>age</th>\n",
       "      <th>years_with_current_employer</th>\n",
       "      <th>years_at_current_address</th>\n",
       "      <th>household_income</th>\n",
       "      <th>debt_to_income_ratio</th>\n",
       "      <th>credit_card_debt</th>\n",
       "      <th>other_debt</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>basic</th>\n",
       "      <td>20715.466902</td>\n",
       "      <td>37.926138</td>\n",
       "      <td>13.452422</td>\n",
       "      <td>15.372353</td>\n",
       "      <td>139.645291</td>\n",
       "      <td>16.259856</td>\n",
       "      <td>9.623007</td>\n",
       "      <td>13.578631</td>\n",
       "      <td>0.112558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>high.school</th>\n",
       "      <td>20496.678068</td>\n",
       "      <td>38.027093</td>\n",
       "      <td>13.585877</td>\n",
       "      <td>15.314944</td>\n",
       "      <td>139.625598</td>\n",
       "      <td>16.192938</td>\n",
       "      <td>9.512649</td>\n",
       "      <td>13.766241</td>\n",
       "      <td>0.112296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>illiterate</th>\n",
       "      <td>20381.154156</td>\n",
       "      <td>38.010513</td>\n",
       "      <td>13.483130</td>\n",
       "      <td>15.424694</td>\n",
       "      <td>139.980562</td>\n",
       "      <td>16.270517</td>\n",
       "      <td>9.645607</td>\n",
       "      <td>13.962909</td>\n",
       "      <td>0.116137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>professional.course</th>\n",
       "      <td>20543.622328</td>\n",
       "      <td>38.135690</td>\n",
       "      <td>13.641278</td>\n",
       "      <td>15.310860</td>\n",
       "      <td>138.129981</td>\n",
       "      <td>16.138116</td>\n",
       "      <td>9.343046</td>\n",
       "      <td>13.543704</td>\n",
       "      <td>0.109208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>university.degree</th>\n",
       "      <td>20828.736579</td>\n",
       "      <td>37.941197</td>\n",
       "      <td>13.585645</td>\n",
       "      <td>15.501012</td>\n",
       "      <td>141.127128</td>\n",
       "      <td>16.258214</td>\n",
       "      <td>9.758195</td>\n",
       "      <td>13.936749</td>\n",
       "      <td>0.113082</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     loan_applicant_id        age  \\\n",
       "education                                           \n",
       "basic                     20715.466902  37.926138   \n",
       "high.school               20496.678068  38.027093   \n",
       "illiterate                20381.154156  38.010513   \n",
       "professional.course       20543.622328  38.135690   \n",
       "university.degree         20828.736579  37.941197   \n",
       "\n",
       "                     years_with_current_employer  years_at_current_address  \\\n",
       "education                                                                    \n",
       "basic                                  13.452422                 15.372353   \n",
       "high.school                            13.585877                 15.314944   \n",
       "illiterate                             13.483130                 15.424694   \n",
       "professional.course                    13.641278                 15.310860   \n",
       "university.degree                      13.585645                 15.501012   \n",
       "\n",
       "                     household_income  debt_to_income_ratio  credit_card_debt  \\\n",
       "education                                                                       \n",
       "basic                      139.645291             16.259856          9.623007   \n",
       "high.school                139.625598             16.192938          9.512649   \n",
       "illiterate                 139.980562             16.270517          9.645607   \n",
       "professional.course        138.129981             16.138116          9.343046   \n",
       "university.degree          141.127128             16.258214          9.758195   \n",
       "\n",
       "                     other_debt         y  \n",
       "education                                  \n",
       "basic                 13.578631  0.112558  \n",
       "high.school           13.766241  0.112296  \n",
       "illiterate            13.962909  0.116137  \n",
       "professional.course   13.543704  0.109208  \n",
       "university.degree     13.936749  0.113082  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.groupby('education').mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='age', ylabel='Density'>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAm8AAAF2CAYAAAA4B39GAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAABlzklEQVR4nO3deZxU5Z0v/s85p/Z975UdBNlV4h6IWUDFDoYxE5cr+cURx9yMermJuUYdMkl0nHiZi9k0GZNJxgRHHZPAYBREDXGBiOICyNpA79219lJV3bWdc35/FLa29FLdXdXV1f15v168pOqcOv0tq6E/nOd5vo+gqqoKIiIiIioJYrELICIiIqLcMbwRERERlRCGNyIiIqISwvBGREREVEIY3oiIiIhKiKbYBYyFRCKBQ4cOwev1QpKkYpdDRERENCBZlhEMBrFw4UIYDIazjk+K8Hbo0CHcdNNNxS6DiIiIKGdbtmzBsmXLznp+UoQ3r9cLIPs/oby8vMjVEBEREQ2sra0NN910U29++aRJEd4+HCotLy9HdXV1kashIiIiGtpAU724YIGIiIiohDC8EREREZUQhjciIiKiEjIp5rwRERHRxJdOp9HU1IREIlHsUnJmMBhQXV0NrVab82sY3oiIiGhCaGpqgtVqxfTp0yEIQrHLGZKqqgiHw2hqasKMGTNyfh2HTYmIiGhCSCQScLvdJRHcAEAQBLjd7mHfKWR4IyIiogmjVILbh0ZSL8MbERERUQlheCMiIiIqIQxvRERERP24//77sXnz5t7H27Ztwze+8Y0iVpTF8EZERETUj5tuugm///3vkclkAADPPPMMrr/++iJXxVYhRERERP0699xzUV1djd27d2PGjBkIBAK4/PLLi10WwxsREdFYyMTaIXeFISdiUNMpSBYHNFY3JKsTgsCBsPHqw7tv06dPx9/+7d+Oi9WsDG9EREQFoKST6D7xNuLH3kSi6RjUZDckqwuizghB0kBOxCDHOwEAppnnwTzvIpjmLIMgSkWunD5u1apV+Nd//VccO3YMzz77bLHLAcDwRkRElFepcAs6/7oV8SN7oXVXwTBlHlwrrodkc/d7hy0TjSDZcgLtrz6N8MtPwHn5l2FZ+GmGuHFCp9Nh1apVCIVCcLlcxS4HAMMbERFRXqSCjYi8+hQSdYdgOudT8Kz+OiSTbcjXaawuaOZeBNM5FyLlr0PnvufQuW87fF/639C5q8agchpMd3c33nrrLWzcuLHYpfQq6CD79u3bcfXVV2PlypXYsmXLWcePHDmCtWvXYtWqVbjvvvt6V3O8/fbbWLt2LWpqanD77bejszN7W7mrqwu33XYbrrrqKtx0000IBoOFLJ+IiGhIcrwTwed/jpYn7odktMG75k5YF38mp+D2cYIgQF8+A67PfxWGaYvQ8pt70fnOi1BVtUCV01Bee+01fOYzn8GnP/1pLF26tNjl9CpYePP7/di8eTOefPJJbN26FU8//TRqa2v7nHP33Xdj48aN2LlzJ1RVxTPPPAMA+M53voOHH34Y27dvx+zZs/GrX/0KAPDII49g2bJleOGFF/DlL38ZDz74YKHKJyIiGpSqKujcvxONP78TSk8M3ppvwLLgMoha/aiuKwgCzOcsg/sL/x86925F+KXfMMAVyac//Wns27cPGzZsKHYpfRQsvO3ZswcXX3wxHA4HTCYTVq1ahR07dvQeb25uRiKR6E2ya9eu7T3+/PPPY/bs2Uin0/D7/bDZsv962b17N2pqagAA11xzDV599VWk0+lCvQUiIqJ+pUJNaPmP+9C1fwdcn1sH27IrIepNef0aGrsX7i/8f+g5fQDB534GVZHzen0qXQULb4FAAF6vt/exz+eD3+8f8LjX6+09rtVqcezYMaxYsQJvvvkmVq9efdZrNBoNLBYLIpFIod4CERFRH6oio33PH9HyH/dBXzkH7pVfg9ZZVrCvJ+pNcH32fyAVbERw+095B44AFDC8KYrSpxeKqqp9Hg91fO7cudizZw/+5//8nwPerlRVFaLI3jhERFR4qXAzmn/zHcSP7oX7ylthnnvhmPRnE7V6uFZ8Bcm202h/fXy0qqDiKth3XXl5eZ8FBcFgED6fb8DjoVAIPp8PyWQSL730Uu/zX/ziF3Hs2DEA2bt3oVAIAJDJZBCPx+FwOAr1FoiIiKCqCjrefA4tv/kODNXz4Prs/4DG4hzTGgSNDs7lX0H0nZ2IHX5jTL82jT8FC2+XXnop9u7di0gkgp6eHrz44otYvnx57/Gqqiro9Xrs378fQHaz1+XLl0Oj0eB73/seDh06BAB44YUXcP755wMAVqxYga1btwLIzotbtmwZtFptod4CERFNcumOAFp/uxGxA6/AvfLvxuxuW38kkxXOFdcj9MK/IemvK0oNND4UrM9bWVkZNmzYgHXr1iGdTuO6667D4sWLsX79etx5551YtGgRNm3ahPvvvx+xWAwLFizAunXrIEkSNm/ejI0bN0KWZZSVlfWuKr3rrrtwzz33YPXq1bBardi0aVOhyicioklMVVV0vbsL7X/eAvP8S2GedwmEcTBNR+ssh/X8lQhsfQRVf/cwRI2u2CWNe1/7wYsIdfTk/boehxG//seVQ563fft2PPbYY8hkMvjqV7+Km266adRfu6BNemtqanpXh37o8ccf7/39vHnz+t1qYtmyZfjDH/5w1vMOhwM///nP818oERHRGen2NgSfexRydydcn/8qtA7f0C8aQ8YZi5FsPo7In7fA84WvFbuccS/U0YN//vpleb/uvY8NPXz9Ydu0P/zhD9DpdLj++utx0UUXYfbs2aP62txhgWgcU1IJpCOtSLe3ItMVhhyLQI51QkknoKYSgKoCoghB0kA0WCAaLdBY3dDYvdDavdC6qyDqjcV+G0QlQZUz6Nz3HDr2/AHm+ZfBMe9vxuUWVYIgwP6pqxF8/hcwz1kG4/RFxS6JBvDxtmkAetum/cM//MOorsvwRjROKOkkkq0nkWg6imRLLVJtpyHH2yFZPdBYXZBMNohGMySLHRqtD4KkAwQAqgpVzkBNJaCkepBsOYHu2v2QY+3IdAYhGizQ+aZCXzkH+srZMFSdM+zO70QTXU/DBwg9/wuIehPcq/4OGqu72CUNSjSYYb/wGgSfexRTbv8xBA3nf49H/bVNO3DgwKivy/BGVCSqIiPZfALdp99Hz6n3kPLXQeMsg85dDZ2nGuZzPgXJ6h7VPBtVVSDHOpBub0M60pr9OqFGSCY7DFPOhXHaAhimzofGUdanVQ/RZJGOtCL8ym+RbDoG6/krYZg6v2T+LBiqz0HPyXfQ8eZ/w3nZ3xS7HOrHUG3RRorhjWgMyT0xdNfuR/z4W0icfh+S2QFd+QyY514Ex+XXjXpbnU8SBDG76bXVBePU+QAAVVGQ6QwgFWxA7IPXEH7ltxBECYap82GcvgiGqQugdVWUzA8wopHIdIXRsecPiH3wGszzLoa35hsQSnDyv/X8lQjv/CWsi6+Axuoqdjn0CeXl5Xj77bd7H3+ybdpIMbwRFVgm2o74sTcRP7IHybaT0JfPhL5iNiyrv16U4UtBFKF1lkPrLAfOuRCqqkKORpDyn0b86F/R/upTUBUFhuq5MEydD0PVXOjKZ3BVG00IqXALOv+6FfEje2GcdR481/xPSAZLscsaMY3VBePsCxB++T9Qdu342n+Tsm3TfvKTnyASicBoNOLFF1/ED37wg1Ffl+GNqADkeCfiR/cieuhVpAINMFTPhXHGIjguvXbc/eteEARobG5obG6Y5iwDAGRiHUgHG5BoPIrouy8h0xmA1lUJXflM6MtnQlc2DTp3NUSTjXfoaNxT5TS6a99F59vPI+Wvg2n2+fDW/ANEg7nYpeWFZcHlCD33MySaT8BQNafY5dDHDNQ2bbQY3ojyREkn0X3ibUTffwWJpmMwVM2Badb5cH76byFIpfVHTWNxQGNxwDgj+5eMmkln5821t6Kn7gCi772EdEcAgihC4yyH1lEGjbMCWrsHGpsbktUNjcUJ0WQtWkNTmtxUOYNE4xHEDr+B+JG90Ng9MM46H45L1kCQJtbkflGrh3nBcrS/+p+ouGFjscsZdzwOY05tPUZy3Vz01zZttErrJwrROJRsO42ud3chfvh1aF2VMExfBNunrs77/LViEjRa6LxToPNO6X1OVVUoiRjkaDsysXbIsQjSwXrI3V2Qu6NQerqgpHogGqyQTFZIJhskkx2S2Q7J4sw+PvN7jcUJyeIsuZBL44eqqkhHWpCo/wDdp99H4vQBSFY39FXnwH3lrWO+ndVYM81aiuAHr/HuWz9yaaRbavg3JdEIqHIasSN70bnvOcjRMIwzz4PnqtsgmR3FLm3MCIIAyWiFZLRC55va7zmqnIGS7M7+SsR7f6XCzVBbTkBJxCH3xKD0dEHuiUEyWqGxeaBxlkHrrIDWXQmdpxpadyVEHfvVUZaqKsh0BJAK1CPZdhrJ5uNItp6EoNFA55sBnW9q0eaUFosgaWCefznaX30aFTfcX+xyqMAY3oiGQUnE0fnOTnTtew4amxfmcy6EvuqccbFtzngkSJozd9yG/iGqKjKUnijkeCcy0XZkuoJINB+H3BVCpisIyeKEzjcd+opZ2Z51FbMgGUt3ojkNTVXk7HB9qAnpcDNSwUakgo1IR1og6gzQOCugcXhhmHIurOd9flKFtf6YZp+H4OHXkWiphaFydB38aXxjeCPKgdwTQ8dftyH6zk7oK+fA+Zkbs6s1KW8EUYJkdmTbp/im9TmmKjIyXWFk2v1IBRsQP/pXpCMt0Fhd0FfPg3HqfParK2HZkOZHKlCf/RWsRzrYiHRnEJLZDo3Nm11UY3XDWnUONHYP78T2I3v37bLs3bfr7yt2OVRADG9Eg1DSSXS++d/ofPM5GKbMhfvK9RN+7sx4JIgStA4ftA4fjDOyWwGpitwb5qKHPtavbtoCmGYsgWHawnG3JyVlhzzT4RYkW04g2VKb3U0k1AjRYIbWUQ6N3QOdZypMsy+AxurmzgHDZJp1HgIH/4J0pAVaV2Wxy6ECYXgj6oeqqogf2YPwS/8BrbsS7pVfg8bmKXZZ9DGCKEHrroTWXQkz0KdfXfTQawi//AREnQGG6YtgmrEYhmkLGbyLQEn1INF8HInGo0g0HEay9SREvQk6TxU0zgqYF1wOh6sCos5Q7FInBEGjhXH2+eh86wV4Vv1dscuhAmF4I/qEdHsbgs89CjnWDvvFX4S+bHqxS6IcfLJfnaqqyHQGkWo7ha53XkTw+V9AY3HCMH1RdieJKedCY3EUu+wJR07EkWg8gkT9IfTUf4B0qBladwW0nmoYz6zEliZIf7XxyjxnGUIv/AKuz9wIUc/h5fqf/D3krlDeryvZPJh2xy9yOjcWi+H666/Hz3/+c1RXV4/6azO8EZ2hqgo69z2PjtefgXn+ZXBc/jcQRKnYZdEICYLQO9RqnncxVEVBur0VKX8dOvc9h+BzP4Nktme3BZtyLvRV50DrrmRfumGSe6JINBxBT/0hJOoPId3eBp13KrTeKbAsWgGdp2rC9VUb7ySzHbqyGYge+DPsn7q62OUUndwVQsX/+F7er9v6u+/mdN7777+P+++/H3V1dXn72gxvRMjuiODfuhlKdxfcK/8OGpt7xNdKpDJoCcbRHIwh3NmDcGcCsZ40kikZqbQMSRQgiiL0WgkWowYWkw5uuwEehxFepxFlLjO0GgaIfBNEETp3FXTuKmD+Zdk9Xjuyc+Zih19H5C//CSWVgL5sBvRVc7LbmJXPgMZZzkD3MZmuEBKNR9HT8AESDYeR6QxB55sKnXcqrOd9HlpXJfv1jQPmuRei860/wbbsSn7/FtkzzzyD7373u/j2t7+dt2vyTxhNej31hxDY+ggM0xfBcemXhn23TVVVNPpjOFoXwdH6CMKdCXjsBrjsRtjNOsyqcsBk0ECrEaGVRCgAVEVFKi0jkcognsigI5ZEkz+G9lgCHdEknFYDppRbMb3chumVNrjtBq6izDNBFKF1VUDrquh9Tk7EkA61IB1pQdf+HUhHWqEku7Nbg51pUqx1VUHrqoDGWTahGjH3R0n2INl2CsnWWiSajiLZfAJqJg2dbxq0nmrYll0FrauCd6jHIa13KgRRRM+p92GadV6xy5nUHnzwwbxfk+GNJrXO/TvR/pf/hOOSa6EfZl+kUEcP9h8J4L0TQYgiMK3MhmXzyuBzmiBKIw9asqyiPZpAoL0bB0+GsGtfPRRVxcxKO+ZMcWLWFAec1okdGopFMlggVZ8DQ/U5vc8pqQQynUFkOgJI+uvQXftOtm1JLALJYM62sbB7oXH6oLF6oLG6IVk/3DXCURJDhmomjXSHH6lQto9ayl+HlL8OciwCjbMCOlcltO5qmOddkt0Jg/+QGPcEQYBx1nnoeu9lhrcJiOGNJiVVkRHe9Wt0n9ifXUlqzW2YVFFUHK2PYM+BVrSF45hVZcdnL6iG224A8vQDTZIEeBzG7L55MwCoKqLdabSE4jhQG8ILe+tgMmgwd5oTc6e6MKPKBo3EYZFCEXWGs7YGAwBVUbI7Q8Q7kYl1QI53IB1uzTYa7t01IgpBa+htVCwZrRDN9ux2YUYbRIMFosEMUW/K/jKYIOqMEHUGCDrDqIe7VEXO7m5xpiY53gk51o5MNIJMRwCZzgDSnUHI8Q5ozA5o7D5INje07iqY5iyDxubmXbUSZpy2CIFtPzqzewkbWk8kDG806ahyGv4/bkamKwz3qltyavaZzih454gfr73fDI0kYf4MF1acVw1pFHfYciYIsJp1mGvWYe40J1RVRaijB02BGF7YexrtXUnMrLJj/gw35k13wmwc/3d6JgJBFAdsKvwhVVWgphLZMPfxbcJ6Ysh0hqCmk1BSCaipBJT0mf9mUlDTSajpFARJA0GjhaDRAZIm+1iUAFE6c/dL6P06UGSoSgZqJp39lU5CldMQtIZsKNSbIBrMkM4Exuyk9mmQLNn3wJA28Yh6I/SVsxE//AZsF6wqdjmURwxvNKko6ST8z/4QajoN12duGHJidSKZwV8PteGN95vhcRpx6aJKlLtNebvLNhKCIMDrNMHrNOG8uT70JDNo8sfw7vEAtr9+EmUuMxbNcmPBTA+cNg6vFpMgiBDO3FUbLlVVATkDVT4TxhQZqiIDigKoSjawAYCaDZIQREAUswFP0kDQ6LK/OMQ5qRlnLEbX+y8zvE0wDG80aSjpJNqeehCCpIHj09cNeqch1p3GG++3YN/hNlT7LFh18XS47OOziahRr8GcqQ7MmeqALKtoDsZwqrkTu99pgt2sw6LZXiya7ckO7VLJEAQB0GizOwwwgxdVRlbgD3cj0N6N9mgSHV0J9KRkZGQFiqJCp5Wg10qwmnRwWHVwWA3wOU2wW3QQxeKGZ33FLHS+uR2pUBN0ntH3FytFks2Tc1uP4V53OF555ZW8fW2GN5oUVDkD/x82AaII+yXXDriRfKQzgdfea8Z7J4KYVWXHFz89E1azboyrHTlJEjC13Iqp5VYoioq2cBx1rV14/f1m2C16LJ3jwZI5XtgsTANEA0lnFNS1dKG2qQO1je0IdvTAbtHDZTPAbNDCdKbFj1YjQgCQlhVkMgpiPSkE2rsR7U6hI5pEMi2jzGVCldeCap8F1T4rvE7jmAY6QZRgnL4Y0fdfgftz68bs644nuTbSLSUMbzThqaqCwPafQOmJwbn8K2cFN1VVcbqlC3sOtOBUSxfmTXXibz4zByZjaf/xEEUBlV4LKr0WXLywAq3hOE43d+KV/U2ocJtx/lwfFs52w6Ar7fc5lGRKRrQ7hZ5EBj2pDNIZBbKiAqoKSRQhSQJ0WglGvQYGvQZWk3bSLgBRFBWRrgTawt0Id/agI5pEVzyFZFpGJqNAEACdVoJBJ8FhM8Bl1cPrNKHSY4ZBX9rfR+mMgqN1ERyoDaG2sQMumx4VXgsuOLccXodxRPNbUykZka4Egh09OFAbwstvNyKeSKPCZUZ1mQVVXgsqPJYRXz9XxplL0L77Sbg++z/Y822CKO0/bUQ5iPx5C9LBJrg+e1OfOW6x7jTePxHEWx+0Ia0oOHe6C1/53BxotRNv4rYoCqjyZn9YXLKoEg1tXXj3eAB/euM05k5z4IJ55ZhVbS/6EM9oxHvSaArE0BKMoS0SR7C9B+3RJGRZgcWog04nQaeRoJUECKIAQRCgKtkgl84oSGVkJJIyupMZ6DQibGY97BYdXDYD3HYjXHZ99r82w4RpotydyKC+tRN1rVHUt3WhNRSHUa+By2aAzayD2aBFpSfbNFqSxOyWY3K2R2GsO4VjkR68+UEbwp0JWE1azKy0Y2a1A7OrHbCYxv/CGUVRcbq5E/uPBnCkPgKvw4gZFTZc99k5MBpG/+NRp5NQ7jGj3PPRdmCplIxgZw+C7T1451gA4X0NiHWnYLfq4bEb4bYbYLfoYTfrYNRrYdRL0GolaCQRkihABQAVUFQVspz9/pUVJTsVUs1+DUkE9DoJBp0GZqMWWocPglaPZNNxGKbMG/X7ouJjeKMJLXrwL4gdehWeVbdC0OjQFU/haF0EH5wMo8EfxdRyK5bNL0Olx1zURQhjSZIEzKiyY0aVHYlkBiebO/GnN06hJyXj/HN8uOBcX7ZNyTjXFU/hZFMHTjZ14lRLJ3oSGXid2XDltBows8IOq1kHg04a1merqioSKRndPWnEetKIdqdR39aFD06l0BlPIRpPwWLUwm03wH2mpcuHAc9h1Y3rO5nR7hTqW6M41dyJU82d6Igl4HOa4XMZsWCGG585rxo63fD/8aIoKjqiSbSG4njrcBu2/eUk3HYD5k134twZblR6zONq4YQ/0o13jwXw7vEgDFoJM6vtY3a3XaeTev8h9SFZVtAZS6EzlkSsJ41GfxTHk5neXVnSsgJZVqEoau+3sigK2V+CAEk88w+SM9dTFCCVyb42lZbhtBpwocGL0J93YErNNHid4//P92ioqjquvt+Gon6Yuodh/P4tQzRKsYajCO74FUJzarDnzTacbu5EV3cKU3xWTCmz4NIllRPmDspIGfQaLJjpxoKZbkQ6EzjR2I6f/+EA3DYjls33YdEsz7gZDlMUFY3+KA7XRXCsrh2d8SSqvBaUu0347AVT4LTp8/IXtiAIMOo1MOo1cPcTYhVFRaw7ha54Nsw1tkVx+HQY0XgKXd0paCUJNrMONosOdnP2DorFrIPFqIXFqIXJoIHJoIVRpxlVM+fBqKqK7kQG/kg3WoJxNAWiaPBH0ZPIoNxtQpnbhIsXlsNjN+alBlEU4LIb4LIbsGCWG7Kswh+Jo9EfxTtHj0JWVZw73YUFM9xF60vYEU3iYG0Q7x0Poqs7hVnVDnzhU1PHxUIkSRJ7///lWyajoCueQjQgQNu0C3duehmzprjw5c/OweI5npIKObkwGAwIh8Nwu90l8d5UVUU4HIbBMLzPXlBHEvlKTFNTEz73uc/h5ZdfRnX15FxtU0rSGQUtoRiC7T0IdfQgkcr+C1RW+n6rCgBkVYUsq0hlZMR70oj3pNEeTaK7I4L1wrN4V1yIbtt0eOxGlLtNcNvHdrJwKVJkFU2BKGqbOtAcjGP2FAfOn+vFnKkuaMair93HpFIyTjR14PCpMI41tMNo0GCqz4rqMiu8jnH4WaoqepIy4ons92I8kc7OtUtmkEhlt0PL/ldGMiVDpxF7h7cMOin7e332sf7MCkadVoRG89Gw2Yc/jxRFhfzxbdZ6MuiMJdHVnUKkMwFVBZy2j4Z6fU4THNYitA5RVbRHU2jwd6HRH0V7VxIzKm2YN92F2dWOgoUnVVURaO/BkdMRHD4dRqijB9MrbJhRZUel2wxhvH3vjAH9vicQ/9TX8E7Eij0HW+Cw6LHhxvNR6Zk4DXzT6TSampqQSCSKXUrODAYDqqurodV+NNVgqNzC8EbjQqw7hRffrMf+owEca2iHw6KH06qHzayD7sx8jz4/qFVAhdo7ZCBJ4pkffhpYjBrMPvobSEYz5NnLi/emJoBkKtM7xNYeS+HcaU4smu3BrGpHwe5atnclcbwhgsOnI6hv64LPacKUMiumlVthMZXOyt+hqKqKdFpBKi0jmZGROvP7D+ffZTIK0pnsfCZZyc5vUlS1d16TIAIiRGg0ArQaCQatBJMhO8fJatLBqB/ecPFY6Ulm0ByIoTmY/aXVSJhRYcPUciuqyywoc5lH9L2VkVUEInE0BWI43ZL9noUATC2zYWqZBeVuy9g01R7HNKf3ADozUuf9LRRVxV8PtWL3O024+cpzcdWl00viTtVkMVRuGR/jITRpdcaS+K+Xj+Oltxoxb6oTS2Z7sGb5LBhHMVSnOfYStKlOJBeszGOlk5Nep8G5M9w4d4YbsZ406lq6sGtfA57adQzTym2YO82JaRU2VLjMIxp+U1UV0XgKda1RnGrJzl/rSWZQ5bVgSpkFlyysGNEcrFIgCEJ2EYVOwsS57zE0o16D2VMcmD3FAagqOmIptIXjON7QjjcOtKAjmoTVnF0o8tHEfQlajQRJErJ3HGUV8UR2TmJHNIlIZwKd8STsFj08DiO8DiOuumQ6bGbduAywxSJ750B3aDuw9MsQBQGXLqrE7GoHnn7pOFpDcdzyxQUMcCWC4Y2K5oNTYTz827cxd5oT37huCRx56D0mtjdAd/h5JM//CsDtfvLKYtRi4Sw3Fs5yI5WS0RSMobaxE3sOtCDWk4bPaYLPZUKZ0wirSQeLKXvXVBQBqNmWHT0pGZ2x7A/bQEc3WkNxKCpQ7jKhzGXCp5dWwW038AfIZCEIcFj1cFg/+rOvyCqiZ+YUdifS2WHgLqW3Ia54ZqVwdrhZwvQKGxbP9sJm1kKapC1ecqWaPYAoQYzUQXHPAAD4nCb8Xc0C/PpPh6GoKm5ds5B//koAwxsVxba/nMTTLx3D2s/MxtxprvxcVJGh++uvkZ71aahGR36uSf3S6STMrLJjZpUdwEf9rNqjCTQFYkgk5WxPNVnBh7s46bQitBoRJkN24v6sSjs+Nb8cFoOGd0eolygJsFv1sFvZSDrvBAGyZzakpnd7wxsAmAxafG31fPzmT4fx2xeOYt3V5xaxSMoFwxuNuR1767Dt1ZO4fe1iOK35m6ysPbID0Bogl/EvnrHWXz8rIhp/FPcMaGv/gvSStX2eNxm0WHf1fDz67PtYNMuN8+b6ilQh5YL3mGlM7T/qx29fOIKbrzo3r8FN6GqF9vhLSJ/zWd7FISIagGIrh5DohBCPnHXMYtRi7RWzsfk/30FHNFmE6ihXDG80ZhrauvCvW/bjhpVz89sEVlWgf/M/kJ52MVSDLX/XJSKaaAQRsms6pNaD/R6eXe3Akjle/OuT+0fUPJbGBsMbjQlFUfHjZ97DFcumYFp5fgOWpu5NCJkE5Koleb0uEdFEpLimQ2p+f8Djn1s2BcH2buw50DqGVdFwMLzRmHj5rQb0JDO48Nzy/F44nYD2wB+Qnr2Cw6VERDmQXdMhhWqBTKrf45IkYuVF0/CbP32AjKyMcXWUC4Y3KriueAq/+dNh1Fw+M+8d8bVHXoDiqIZir8zrdYmIJiytAYrFBylwbMBT5kxxwmbWYefeurGri3LG8EYF9x9/OowFM919NmLOByEeyq6amnFZXq9LRDTRKa7pkFoODHrOygun4T93HUN3Ij1GVVGuGN6ooMKdPXj9/WZ8btmUvF9bd2ArMlVLAYM179cmIprIZPcMSC0HgUEWJVR6LZhZZcd/v3pqDCujXDC8UUFtf+0Uls7xwmTQDn3yMAhdrZDaDiMz5YK8XpeIaDJQzW5AlSHEAoOe9+klVXh+z2nOfRtnChretm/fjquvvhorV67Eli1bzjp+5MgRrF27FqtWrcJ9992HTCYDANi/fz+uu+46rFmzBl/96lfR3NwMANi3bx8uuugirFmzBmvWrMF3vvOdQpZPo9STzGDnX+tx6eL8z0fTHfxvZKrPBzQTZ6NyIqIxIwhQnFMhtR0Z9LRytxlOmwF/PcSVp+NJwcKb3+/H5s2b8eSTT2Lr1q14+umnUVtb2+ecu+++Gxs3bsTOnTuhqiqeeeaZ3ucfeOABbNu2DTU1NXjggQcAAIcOHcItt9yCbdu2Ydu2bXjooYcKVT7lwUv76jGj0gaXLX/NeAFA6GiCFDiOTPXSvF6XiGgyURzVkNoOD3nehfPLsf01Dp2OJwULb3v27MHFF18Mh8MBk8mEVatWYceOHb3Hm5ubkUgksHTpUgDA2rVrsWPHDqRSKdx1112YN28eAGDu3Llobc0m/oMHD+L1119HTU0Nbr/99t7nafyRFRVb/3KyQHfdtmWHS6X8DsUSEU0msnMqpOBxQBl8SHTBDBeaAzHUt3aNUWU0lIKFt0AgAK/X2/vY5/PB7/cPeNzr9cLv90On02HNmjUAAEVR8NOf/hSf//znAQBWqxU333wztm/fjhUrVmDDhg2FKp9G6Z2jfhj1mrw35BU6miGFTyFTtTiv1yUimnT0Fqg6C8SOhkFPkyQRy84tw/bXefdtvChYeFMUBcLHmqaqqtrn8VDHU6kUvvWtbyGTyeDv//7vAQDf//73sXLlSgDADTfcgNraWkSj0UK9BRqF3e80YfEc79AnDpP26E5kKpfyrhsRUR4ozilDznsDgE/NL8Nr7zUjkcyMQVU0lIKFt/LycgSDwd7HwWAQPp9vwOOhUKj3eDwex6233opMJoPHHnsMWq0WiqLgsccegyzLfb6OJEmFegs0Qsm0jLeP+LFwpjuv1xW626FpOYBM1aK8XpeIaLKSnVMg+Yee92Yz61Hts2L/0cFXp9LYKFh4u/TSS7F3715EIhH09PTgxRdfxPLly3uPV1VVQa/XY//+/QCAbdu29R6/++67MW3aNDzyyCPQ6bKrCUVRxK5du7Bz504AwNatW7FkyRKYTKZCvQUaobeP+FHltcBqyu9KUM3xVyCXnQto87ipPRHRJKY4qiGG6wB56Ea882e48Np7zYUvioZUsPBWVlaGDRs2YN26dbj22mtxzTXXYPHixVi/fj0OHjwIANi0aRMeeughXHnlleju7sa6detw+PBhvPzyy3jnnXfwpS99CWvWrMH69esBAD/84Q/xxBNPYPXq1fj973/fuwqVxpfd7zTl/a4b0gloT72OTPV5+b0uEdFkptFDtXghhk4Oeer8GW68cyyAZFoe8lwqLE0hL15TU4Oampo+zz3++OO9v583bx6effbZPsfnz5+PY8f6329tzpw5eOqpp/JfKOVNdyKN948H8fkbz8/rdTWnXofsnArVaM/rdYmIJjvZUQ3JfwRK2bxBz7MYtaj2WfDOUT8uWcT9pIuJOyxQXu077Mf0Clt+d1RQFWhP/Bky+7oREeWd4qgedJP6j5s/w4VX3+XQabExvFFevfpu/odMxcAxQBCh2Cryel0iIgIUWwXEjuYc5725sf8oh06LjeGN8iadUXDoZBjnTHXm9braE7uRqVwEfKyVDBER5YlGB9XsgRipG/JUq0mHKq8Z73DVaVExvFHeHKuPwOMwwGzM35Cp0NMJyX8U8hBzMYiIaORkewWkwPGczp071YW3DrcVuCIaDMMb5c27x4KYVeXI6zU1p16H7DsH0Ojzel0iIvqIYq+CmGN4mz3FgXePB6GqaoGrooEwvFHe7D/mx6zqPK4GVRRoTr6WHTIlIqKCURxVkCKnAWXouWw+pxHpjIzWcHwMKqP+MLxRXkS7U2gOxPK6l6kYOJrtQWQty9s1iYioH1ojVL0NYkfTkKcKgoDZ1Q68fzw45LlUGAxvlBcHToQwvdIGjZS/bynN6b3IlJ2bt+sREdHAFEclxOCJnM6dWWnnVllFxPBGebH/qB+zKh35u2AmCU3LAchlc/N3TSIiGpBir8x50cKsagcOnQxBVjjvrRgY3mjUVFXFu8eDmD3FkbdrSk3vQrFXATruXUtENBYUezWkUC2Qw0IEm1kHm1mPk00dhS+MzsLwRqPWFu5GOiPD58zfhvGa03t5142IaAypBitUUQMhlttw6MxqO949xqHTYmB4o1E7fDqMGZV2CHlqoiv0dEKK1EH2zMrL9YiIKDeKvRJS6FRO586u4ry3YmF4o1E7fDqCaq8lb9eT6vdB9s4GpDzuj0pERENSrGUQwydzOnd6pR0nmzuQ4lZZY47hjUbtaF0EU8utebuepv5NyD4OmRIRjTXFVgExlFt402sl+JwmnGruLHBV9EkMbzQq3Yk0/JFulLvNebmeEA9B7I5AcUzJy/WIiCh3qtULMRYEMsmczp9SZsXR+kiBq6JPYnijUTne0I4qnyVv/d2kxneyc91EfmsSEY05UQPFUgYxUp/T6dU+Cz44FS5wUfRJ/AlJo3KkLoJqX/7mu2ka90P2zM7b9YiIaHhUWxmkHIdOp5ZZcay+nfucjjGGNxqVw6cjmOLLz3w3obsdYtQPxckhUyKiYlFs5TnPe3PZDMjICoIdPQWuij6O4Y1GTFFUnGhoz9tiBanp3TNDplJerkdERMOn2Cqym9TncDdNEARMLbfhWF37GFRGH2J4oxFrDsZg0GtgNenycj1Nw1scMiUiKjJVb4UKAUI8t7ls1T4LDtdx3ttYYnijETtWH8HUsjwNmfZ0QuxsgeKampfrERHRCAkCVFsFxHBuzXqnlllxpI4rTscSwxuN2OHTEVTlqTmv1HIQsnsGIGrycj0iIho5ZRiLFqq8FjT6o2zWO4YY3mjEaps68rbSVGp+D4prel6uRUREo6NYyyGGT+d0rk4rocxlQi03qR8zDG80IhlZQXMglp/mvHIaUvA4ZPf00V+LiIhGTbGWQexqAZTc7qZV+6w43sBFC2OF4Y1GpCkQg9NmgE47+pWhUuAYFIsP0BrzUBkREY2aRgfVYIfY2ZLT6RVuM040dhS2JurF8EYjcqq5AxVuU16uJTUf4JApEdE4o1jLILbnttNCpcfMPU7HEMMbjcjJpk6U5WPIVFUhtRzILlYgIqJxQ7H6cp735nOZEIh0I8lFC2OC4Y1G5GRTJyryEN6EzhZAVaCa3XmoioiI8kW1lkGM1OV0rkYS4XWaUN/aVdiiCADDG42Aqqqoa+1EhWf04U3TcgCKeyYgCHmojIiI8kWxeCF2+QE5ndP5FRw6HTMMbzRswfYeaDRiXnZWyA6ZTh99UURElF+SFqrJBbGjOafTy9kuZMwwvNGwnWzuRKUnD/3d0j0QO5qgOKpHfy0iIso7xebLedFChceMkwxvY4LhjYbtVHMHylyjX2kqBY5BsVcCkjYPVRERUb4pltwXLVR4zGjwRyErQ29oT6PD8EbDdrIpP/PdpNbDUBxT8lAREREVgmItg5TjogWDTgObWY/mQLSwRRHDGw3fqZb8rDSV2g5Ddk3LQ0VERFQIqsUDIRYCMqmczq/0mHGqhStOC43hjYYl2p1CrCcNl90wqusI8TCEdDdUizdPlRERUd6JGigWD8SOxpxOL3ebOO9tDDC80bDUt3ah3G2COMrWHlLbEcjOqWwRQkQ0zqkWL8T2hpzOrXBbuOJ0DDC80bA0+qPwOfOwWKHtEBQn57sREY13isULMZJjePOY2ah3DDC80bDUt0XhdYxyA3lFgeQ/BtnJ+W5EROOdMow7b1aTFrKioiOaLHBVkxvDGw1LfWvXqO+8iR0NUHUmwGDNU1VERFQoqsULMeoH5MyQ5wqCgHK3GQ1+3n0rJIY3GpbGwOiHTSX/UQ6ZEhGVCkkL1WiH2NWa0+lehxENbWwXUkgMb5SzaHcKyZQMu2V022KJ/qOQuasCEVHJUCy+nIdOvU4j6jjvraAY3ihnjf4oylwmCKNZIarIkMKnoNgZ3oiISoVq8eQc3nxOExctFFhBw9v27dtx9dVXY+XKldiyZctZx48cOYK1a9di1apVuO+++5DJZMfT9+/fj+uuuw5r1qzBV7/6VTQ3ZzfF7erqwm233YarrroKN910E4LBYCHLp09o9EfhHe18t/YGqAY7oBvlogciIhozitUHMZLbHqdlLhMaAzGoKrfJKpSChTe/34/NmzfjySefxNatW/H000+jtra2zzl33303Nm7ciJ07d0JVVTzzzDO9zz/wwAPYtm0bampq8MADDwAAHnnkESxbtgwvvPACvvzlL+PBBx8sVPnUj7rW0a80lQLHuBE9EVGJUSxeiJ3NgKoMea7FqAVUrjgtpIKFtz179uDiiy+Gw+GAyWTCqlWrsGPHjt7jzc3NSCQSWLp0KQBg7dq12LFjB1KpFO666y7MmzcPADB37ly0tmYnSe7evRs1NTUAgGuuuQavvvoq0ul0od4CfUJDWxd8ztGFt+x8t6o8VURERGNCa4SqM0KIDT3i1bvilIsWCqZg4S0QCMDr/WjrI5/PB7/fP+Bxr9cLv98PnU6HNWvWAAAURcFPf/pTfP7znz/rNRqNBhaLBZFIpFBvgT5h1A16Od+NiKhkqZayYS1aaPAzvBVKwcKboih9Jrarqtrn8VDHU6kUvvWtbyGTyeDv//7v+/0aqqpCFLnmYizEetLoTmRgt+pHfA3OdyMiKl2KOfdFC16HiStOC6hgyae8vLzPgoJgMAifzzfg8VAo1Hs8Ho/j1ltvRSaTwWOPPQatVgsge/cuFAoBADKZDOLxOBwOR6HeAn1M05mVpqPZ05Tz3YiISpdi9UHKcdGCz2VEfRvDW6EULLxdeuml2Lt3LyKRCHp6evDiiy9i+fLlvcerqqqg1+uxf/9+AMC2bdt6j999992YNm0aHnnkEeh0H/UUW7FiBbZu3QoAeP7557Fs2bLeYEeF1eCPwsv5bkREk5Zq8ULsaM7p3DKnCU3+KFecFoimUBcuKyvDhg0bsG7dOqTTaVx33XVYvHgx1q9fjzvvvBOLFi3Cpk2bcP/99yMWi2HBggVYt24dDh8+jJdffhmzZ8/Gl770JQDZO26PP/447rrrLtxzzz1YvXo1rFYrNm3aVKjy6RPq27rgGc1KU0WBFD6F9OwV+SuKiIjGjKq3AKoMoacTqtE+6LlmoxYQBLRHk3DZDGNU4eRRsPAGADU1Nb2rQz/0+OOP9/5+3rx5ePbZZ/scnz9/Po4dO9bv9RwOB37+85/nv1AaUmNbFAtnukf8erGzGareCuhG1yeOiIiKRBCyOy10NkMeIrwJgoBylwkNbV0MbwXA2f6Uk5ZQHO5R3HkTQ7VQ7BV5rIiIiMaaavZA7GjM6Vyv04hGf6zAFU1ODG80pHRGQbgzMap/PUmB41BslXmsioiIxppicUNszy28uW1GNAXYLqQQGN5oSP5IHE6rHhpphN8uqgoxdBKKneGNiKiUKRZvznfePA4DmgK881YIDG80pJZgfFSLFYTuCAQlA9XoyF9RREQ05lSTG0IsBMhD727kcRjRHGR4KwSGNxpSUyA2qiHT7F23KmAUPeKIiGgckDRQTU6IXa1Dnuq0GtAVTyGRyoxBYZMLwxsNqSkQhds+ivluweNQbFysQEQ0EShmL8SOpiHPE0UBbrsBraH4GFQ1uTC80ZCaArFRDZtKwVrIDs53IyKaCFSzK+dFCxw6LQyGNxpSa2gUc95S3RDiYagW39DnEhHRuKdYvMNYccpFC4XA8EaD6k6kEU+kYTPrhj65H1L4VHbIVJTyXBkRERWDYvFC7GwGctj6yuMwotHPdiH5xvBGg2oJxeFzGke8Ib0YrIViK89zVUREVDQ6MyAIEHo6hjyVw6aFwfBGg2oJxuC2j2K+W/gkFysQEU0kH26TlcOiBa/DiJZgnBvU5xnDGw2qeTRtQhQFYqSe4Y2IaIJRzK7s0OkQTAYtRAHoiCXHoKrJg+GNBtUQiMIzwjtvQlcrVJ0Z0I38zh0REY0/qtmT86IFr9OEZi5ayCuGNxpUdneFkd15k8KnedeNiGgCUiyenO68AR/Oe2Ovt3xieKMBqao6qjYhYugkFFtZnqsiIqJiU00uCLEgIA+9e0K2XQhXnOYTwxsNqCOWhCBk5yyMBBcrEBFNUJIWqtEOIeof8lS2C8m/nMLbHXfcgT179hS6Fhpn2kLd8I6mOW93O1SzJ79FERHRuKCacxs6ZbuQ/MspvH3hC1/Ao48+ilWrVuFXv/oVOjo6ClwWjQet4ZGvNBUjddn+bmzOS0Q0ISkmd07tQlw2A0IdCcgK24XkS07h7Ytf/CJ+97vf4dFHH0U4HMZ1112Hu+++GwcOHCh0fVREraE4HDb9iF4rhU5BsXK+GxHRRKVYPBA7hl5xqtWIsJq0CHX0jEFVk0POc94URUF9fT3q6uogyzLcbjf+6Z/+CT/+8Y8LWR8VUVMwBrd1pIsVuLMCEdFEplo8EDtbcjrX4zCiNcSh03zR5HLS5s2b8Yc//AFTpkzBjTfeiB/96EfQarXo7u7GFVdcgTvvvLPQdVIRtIXiWDDdPfwXqiqkSD3Ssz6d/6KIiGhcUA12CKluINUN6EyDnuu0GtAaimPpOWNU3ASXU3iLRCJ4/PHHMW/evD7Pm0wm/Ou//mtBCqPiawt3w2Uf/pw3IRaEKmkBvaUAVRER0bggCGf6vbVA8c4e9FSXTc9eb3mU07CpLMtnBbcP77Zdfvnl+a+Kii7ek0Y6o8BiHH6bEDFSB5VDpkREE16uK05ddq44zadB77x997vfhd/vx/79+xGJRHqfz2QyaGzMbVsMKk2t4ezOCoIgDPu1UvgUFIuvAFUREdF4ophdOS1acNsNaAvzzlu+DBrerrvuOpw4cQLHjh3DqlWrep+XJAlLly4tdG1URK2h+MjbhIRPIzN1WZ4rIiKi8UY1eyA17B/yPLfNgECkG4qiQhSHf1OA+ho0vC1atAiLFi3CZZddhrIytn2YTNrCcTitIwhvigyxsxmKdXX+iyIionFFMXshdrUCqgoMMlKj00ow6DWIdCVGvOUifWTQ8HbXXXfhRz/6EW699dZ+j2/fvr0gRVHxNQdicI7gzpvY2QzV6AA0I+sPR0REJURnBCQNhJ52qCbXoKdm24WMfL9s+sig4W39+vUAgH/8x38ck2Jo/GgJxXHJItuwXydG6ticl4hoElHMXoidLZCHCG8umwEtoTgWzea2iaM16GrThQsXAgAuvPBCVFRU4MILL0R3dzfeeustnHvuuWNSIBVHWzgO90juvIVOQ7FysQIR0WShml05rTh1WvVoYaPevMipVcjGjRvx+OOP4+TJk7j//vvR1NSEe++9t9C1UZEk0zKi3WnYLcMf+pQip6FY2SaEiGiyUMxuCO1D73HqthvQHGB4y4ecwtuhQ4fwT//0T9i1axe+9KUv4aGHHkJz89Apm0pTWzgOl00//BVBmSSEeAiqxVuYwoiIaNxRzO6c7ry57Ua0sl1IXuQU3lRVhSiKeOONN3DxxRcDABKJREELo+JpC8Xhtg9/QqnY3gDF4gVEqQBVERHReKSaPRCjAUCRBz3PZTPAH+6GqqpjVNnElVN4mzp1KtavX4+mpiZceOGF+OY3v4m5c+cWujYqktZwN5zW4Q+ZiuE6qFysQEQ0uUhaqHoLhFhw0NOMeg20GhEd0eQYFTZx5bS36UMPPYRdu3bhggsugFarxbJly3DttdcWuDQqlpbQyNqEcGcFIqLJST2zx6k8xNaIHocRLaH4iH7G0EdyuvNmMpmwbNkydHV14YMPPsDixYtx6tSpQtdGRdIaGuFK0/YGKNzTlIho0lFMLogdQy9acNr0aA1x3tto5XTn7Uc/+hH+/d//HW63u/c5QRDw8ssvF6wwKh5/uBvO84YZ3lJxCIkuqCZnYYoiIqJxSzW7IXbmEN6s3OM0H3IKb9u2bcOLL77ILbImAUVREezoGfacN7G9IducV8jpZi4REU0gisUDTf1bQ57nshrQzF5vo5bTT9qKigoGt0ki0pWA2aiBTju8FaNSpB4qm/MSEU1KqtEJoacdyKQGPc9lz644pdHJ6c7bJZdcgocffhif+9znYDB8NJy2YMGCghVGxZHt8TaC+W7h01BsFQWoiIiIxj1RgmpyQYy2QXFOHfA0l00Pf4ThbbRyCm9/+MMfAAA7duzofY5z3iamtnA3nNaRLVbIVC/Nf0FERFQSFLMHYkfzoOHNatKhJ5lBdyINk0E7htVNLDmFt1deeaXQddA40RaOwzHcHm/JGIRUHKqRixWIiCYr1eSCMMROC4IgwG03wB/pxoxK+xhVNvHkNOctHo/j+9//Pr761a+io6MDGzduRDw+9GqR7du34+qrr8bKlSuxZcuWs44fOXIEa9euxapVq3Dfffchk8n0Of7II4/gJz/5Se/jffv24aKLLsKaNWuwZs0afOc738mlfBqGllAcrmHeeZPaG7L7mQrD3E6LiIgmDMXizqldiNvOFaejlVN4e+CBB2C1WhEOh6HX6xGLxbBx48ZBX+P3+7F582Y8+eST2Lp1K55++mnU1tb2Oefuu+/Gxo0bsXPnTqiqimeeeQYAEI1Gce+99+LXv/51n/MPHTqEW265Bdu2bcO2bdvw0EMPDee9Ug7awnE4bcNcaRqug2LlfqZERJOZanZD7God8jyHxYA2LloYlZzC25EjR7BhwwZoNBoYjUZs2rQJR44cGfQ1e/bswcUXXwyHwwGTyYRVq1b1mTPX3NyMRCKBpUuXAgDWrl3be/zll1/G9OnT8bWvfa3PNQ8ePIjXX38dNTU1uP3229HaOvQ3CQ2PP9I97AULYuQ0VO6sQEQ0qakGO4RUN5DqGfQ8p02PFrYLGZWcwpso9j1NluWznvukQCAAr/ejuzE+nw9+v3/A416vt/f4tddei9tuuw2S1LddhdVqxc0334zt27djxYoV2LBhQy7lU44SyQx6khlYzbphva63xxsREU1egpBdtNDVMuhpLpuBuyyMUk7h7VOf+hT+7//9v0gkEnjttdfwD//wD7jooosGfY2iKBA+NgdKVdU+j4c63p/vf//7WLlyJQDghhtuQG1tLaLRaC5vgXLw4V03cThz1xJdENIJqEZHweoiIqLSkN1pYfBFCy4bh01HK6fw9q1vfQsmkwlWqxWPPPII5s2bh29/+9uDvqa8vBzBYLD3cTAYhM/nG/B4KBTqc/yTFEXBY489BlmW+zz/ybtzNHIj6fEmfbifKRcrEBFNeop56D1OnVYDwp09kBV1jKqaeIYMb7t27cLNN9+MX/7yl2hqaoLVasX5558PvX7wSe2XXnop9u7di0gkgp6eHrz44otYvnx57/Gqqiro9Xrs378fQHYLro8fP6tQUcSuXbuwc+dOAMDWrVuxZMkSmEymnN4oDa0t0j3sNiFipB6K2VOgioiIqJSoZ3q9DUarEWEx6RDqGHxuHA1s0D5vL7zwAjZv3ow777wT8+bNgyAIOHjwIB588EEkk8neIcz+lJWVYcOGDVi3bh3S6TSuu+46LF68GOvXr8edd96JRYsWYdOmTbj//vsRi8WwYMECrFu3btBif/jDH+If//Ef8bOf/QwulwsPP/zwyN419astFB/+nqaROiiOKQWqiIiISolidmfnvKnqoCMyblu2XUiZizdgRmLQ8PbEE0/gN7/5DSorK3ufmzVrFpYsWYJ777130PAGADU1Naipqenz3OOPP977+3nz5uHZZ58d8PV33HFHn8dz5szBU089NejXpJFrCcUxf4ZrWK8R2xuQmXJBgSoiIqKSojNng1syChhsA57mPBPelsxhm6mRGHTYNB6P9wluH5oxYwaSyWTBiqLiGHabkGSMixWIiOgjggDF4h1y0YLDqueK01EYNLwNthhAVTnRcCJRVRXB9m44hxHepEh9tkUIFysQEdEZqskFsXOIdiFWA1oY3kYsp9WmNPG1R5PQ6yTotbmv3hUj9VAsvOVNREQfUcxDb5PFLbJGZ9A5b8eOHcP5559/1vOqqiKVShWsKBp7beE4XPbh7qxQB8VRVaCKiIioFKlmN6S6vw56jtNmQCDCXm8jNWh427Vr11jVQUXmj3TDOcwN6bOLFc4O90RENHllV5y2AaoCCP0P8JkNGmQUFbGeNCxG7RhXWPoGDW9VVbyrMlm0heJwWIbRJiQZg5Du5mIFIiLqS2uEqtFD6I5AHaAPqCAI8NgN8IfjsFQ7xra+CYBz3ggA0BoeXo836cP9TLlYgYiIPkG1eIds1uu0GdDGodMRYXgjAEBbeHhtQrhYgYiIBqLksOLUYdHDz0ULI8LwRgDOzHkbVnirg2oZeC9aIiKavFSzG2JH46DnOG1sFzJSDG+EdEZBVzwJ+zDmvIkfDpsSERF9gmLx5NTrjY16R4bhjRBs74bdoock5jh/LRWHkIxBNTkLWxgREZUk1eSCEAsCijzgOS6bHn7OeRsRhjdCW6Qb7uEMmXKxAhERDUbSQjXYIET9A57isBoQ7uyBrHDHpuFieCP4w3E4htHjTWpv4Hw3IiIalGoefI9TrUaExahDuKNnDKuaGBjeCG3hbjiG0SZEDNdBsXKlKRERDUwxO3PbJivCeW/DxfBGaAnFhtXjLTtsyjtvREQ0MMXsGbLXm8OqR1uY896Gi+GN4I8Mo8dbqgdCoguq0VXYooiIqKSpZg/EriF6vVn1XHE6AgxvhEB7T8493sSORigWHyDyW4eIiAamGh0QerqATHLAc1w2A1rZqHfY+BN4kov3pJGRFZgNg25z20tsr4fK+W5ERDQUUcw26+1sHfAUl82ANoa3YWN4m+T8Z9qECDm2/ZDCddwWi4iIcqKY3YOuOHXZDAiw19uwMbxNcm3h+PD2NOXOCkRElCPV5Bp0xanFqEUyLaM7kR7Dqkofw9sk5490w57rStN0AkJPO1QTFysQEdHQFItn0PAmCALcdiN3WhgmhrdJriUUz7lNiNjRBMXsBUSpwFUREdFEkN3jdPB2IS6bnvPehonhbZJrC+U+bMrFCkRENCw6C6AqQKJrwFMcFgN7vQ0Tw9sk1xYZRngL12XbhBAREeVCEKBYvIMOnTpterSEYmNYVOljeJvEZEVFqCOR87Cp1F7PxQpERDQsqnnwoVOXzcBGvcPE8DaJRToTMBs10GpymMOWSUKIR6Ca3YUvjIiIJgzF7ILY3jjgcZfNAD+HTYeF4W0Sa4vE4bYbczpXbG+EYvFwsQIREQ2LYvYOusep02pAqDMBWVHHsKrSxvA2ifnDw1hp2t4AlfPdiIhomFSzG2LUDyhyv8e1GhEWoxbhjp4xrqx0MbxNYq3hbjhyDW+ROihcaUpERMOl0UE1WCDEggOe4rYb0BbhvLdcMbxNYi2hGFzW3FaacrECERGNlGoeasWpAa0hznvLFcPbJNYW7s6tTUgmBSEW5mIFIiIaEcU8+DZZDisb9Q4Hw9skFoh0w5lDeBM7GqFa3ICoGYOqiIhoolHMHogdg6w4tRrY620YGN4mqe5EGj3JDKwm7ZDniu0NbM5LREQjplq8EDtbBjzushnQxmHTnDG8TVL+SDfcdgMEQRjyXDF8muGNiIhGTDXaISSjQLr/FaUuuwH+doa3XDG8TVI5z3cDILU3QLEyvBER0QgJ4qD93swGDTKyglhPeowLK00Mb5OUPxLPrU1IJgUhFoJq8RS+KCIimrBUy8Dz3gRBgMdu5KKFHDG8TVItoTicObQJETs+3FmBixWIiGjkFLNniG2y9NwmK0cMb5NUayie07CpGKnnzgpERDRqqsU76IpTh9WAVt55ywnD2yTlz3HOmxg+zfluREQ0aorFA7GrDVCUfo87rXq0BNkuJBcMb5OQoqgIdfbktK+pFKmDYisfg6qIiGhC0+ih6kwQYoF+D7tsBrSGeOctFwxvk1CoswcmgwY6rTT4iekEhJ52qCbX2BRGREQTmmrxDTh0mt3flHPecsHwNgm1huLw2I1Dnie2N2Tnu4lDhDwiIqIcKGb3gIsWHFYDOqJJpDPyGFdVegoa3rZv346rr74aK1euxJYtW846fuTIEaxduxarVq3Cfffdh0wm0+f4I488gp/85Ce9j7u6unDbbbfhqquuwk033YRgMFjI8iestnCOixXa6znfjYiI8kaxeAcMb5IowGnVw8+7b0MqWHjz+/3YvHkznnzySWzduhVPP/00amtr+5xz9913Y+PGjdi5cydUVcUzzzwDAIhGo7j33nvx61//us/5jzzyCJYtW4YXXngBX/7yl/Hggw8WqvwJrSUYz22+G3dWICKiPMpuk9V/o14gO3TKeW9DK1h427NnDy6++GI4HA6YTCasWrUKO3bs6D3e3NyMRCKBpUuXAgDWrl3be/zll1/G9OnT8bWvfa3PNXfv3o2amhoAwDXXXINXX30V6TS7MQ9XczAGlz23NiGKrWwMKiIioslANdggZBJAsv9VpU4uWshJwcJbIBCA1+vtfezz+eD3+wc87vV6e49fe+21uO222yBJfedaffw1Go0GFosFkUikUG9hwmoNx+Eeas5bqhtCoouLFYiIKH8EAcogixZcNgOa2S5kSAULb4qi9Nn0XFXVPo+HOp4LVVUhilxzMRyqqubU4613vpvA/79ERJQ/qtkDsaOp32NuuwEtQd55G0rBfjKXl5f3WVAQDAbh8/kGPB4Khfoc74/P50MoFAIAZDIZxONxOByO/BY+wXVEk9BoRBj1g293JYXroFo5ZEpERPmlWDwQIw39HnPbuMtCLgoW3i699FLs3bsXkUgEPT09ePHFF7F8+fLe41VVVdDr9di/fz8AYNu2bX2O92fFihXYunUrAOD555/HsmXLoNVqC/UWJqTWcBweRy47K5yCwvBGRER5plh9kDr6D29OmwHhzgQycv+7MFBWwcJbWVkZNmzYgHXr1uHaa6/FNddcg8WLF2P9+vU4ePAgAGDTpk146KGHcOWVV6K7uxvr1q0b9Jp33XUX3nvvPaxevRpPPvkkNm7cWKjyJ6zWUBzuHPc05c4KRESUb6rZDSEWBjKps45pJBF2iw7B9p4iVFY6Bh87G6Wampre1aEfevzxx3t/P2/ePDz77LMDvv6OO+7o89jhcODnP/95foucZFpCcTisg4c3obsdgpyGarCPUVVERDRpiBqoZjfEzmYo7hlnHf6wXUiFx1yE4koDZ6NPMi3B2JB33rJ33SqAYS4gISIiyoVi9UJs73/oNLvHKVecDobhbZJpCcXhHqLHW3a+G5vzEhFRYShmL8RIXb/HnFYDmtnrbVAMb5OMP4etsaTwaTbnJSKiglGtvgHvvLnt7PU2FIa3SSTanYKsqDAbB1mhqyoQ2xugWLlYgYiICkOxeCF2+QE5c9Yxl82ANt55GxTD2yTSGorD4zAO2gxZiAagag2AzjSGlRER0aQiaaGaHBC7Ws865LIZEGjvgayoRSisNDC8TSKtoVyHTCvGqCIiIpqsFEv/Q6c6rQSLUYtwB9uFDIThbRJpCcXgtOoHPUcMn4Zq4WIFIiIqLNXsgdhe3+8xj8OAFq44HRDD2yTS6I/B4xh8Q3oxfIqLFYiIqOAUqw9ipP/w5rYb0RxgeBsIw9sk0hwcIrzJaYjRNm6LRUREBadYvBA7WwDl7K2w3HYDGhneBsTwNkmoqorWUBzeQcKbGKmHYvYAEveLJSKiAtMaoOrNEKJtZx3yOIxoCkSLUFRpYHibJDpiSYgCYDIMHMzE8Cmo3M+UiIjGiGIth9TP0KnXYURzkO1CBsLwNkm0BOPwOgdv/yGFTnIzeiIiGjOqxQcxfPqs5x1WAzqiSSTTchGqGv8Y3iaJpkBs8G2xVPXMYoXKsSuKiIgmNcVWBjF86qznJVGA285mvQNheJskmgJRuO0Dz3cTutshKDJUg20MqyIioslMsZZBjLb1u9OC12FEE7fJ6hfD2yTRFIjBM8idNzF8Eoq9Ehhk9wUiIqK8krRQjU6InU1nHXLZDWwXMgCGt0miZYg2IVLoJPczJSKiMadYy/rt9+axG9Ho54rT/jC8TQKyrCDQ3jPonDcxdBKKndtiERHR2FKsZRBDJ8963uMwopnDpv1ieJsE/O3dsJl10Gqk/k+Q0xC7Wtmcl4iIxpxiK+u3XYjHbkBLMAZV5Qb1n8TwNglk24QM0py3vQGq2c3mvERENOZUswdCdxhIJ/o8bzZqoapAVzxVpMrGL4a3SaApEIPLNsSQqZVDpkREVASilO331t7Q52lBEOB1cui0Pwxvk0BTIArPIG1CpMCx7EpTIiKiIsguWqg763m3gxvU94fhbRJo8sfgcQxw501VIIVPQXZUjW1RREREZyhWH6TQ2c163TYDe731g+FtEmgOxQa88yZ0tUGV9IDeMsZVERERZSm2CoiRs7fJ8jiMaGhju5BPYnib4GI9afQkMrBb9f0el4K1HDIlIqKiUo0OCHIaQnekz/M+p4m93vrB8DbBNfmjKHOZIA6wc4IUPM7wRkRExSUIUOyVED8xdOqxGxDpSnCD+k9geJvg6tuig7cJCdZC4Xw3IiIqMsVaDilU2+c5SRLhsRvRxLtvfTC8TXD1rZ3wDrAtltAdgSCnoBqdY1wVERFRX4q9EmKw9qznfS4OnX4Sw9sEV98Whc9l6veYGKyFYq/iZvRERFR0itUHMdoGZPo25fU4DKjnooU+GN4muAZ/FGXO/sObFDwOmfuZEhHReCBpoZq9Z/V7K3OaUNfaVZyaximGtwks1pNGT3Kolaac70ZEROODYq+A9IlN6n0uE9uFfALD2wTW2DbIStNkDEJ3BKrFO/aFERER9UO2VZw1781tM6A9mkAilSlSVeMPw9sE1uDvgm+gIdPAMSiOakCUxrgqIiKi/im2CkiR04Cq9j4nSSI8DiOauE1WL4a3CayutWvAlaaS/wgUe/UYV0RERDQIgxWqqIEQC/R5usxp5IrTj2F4m8DqW7vgG6DHm+Q/Btk5ZYwrIiIiGpziqIIUONHnOY/ThHouWujF8DaBNfpj/bYJEXo6ICSjnO9GRETjjmKvghQ42uc5H1ec9sHwNkHFulPoSWXgsJy90lT0H4PimML+bkRENO4ojmqIgeN95r35OGzaB8PbBNXgj6LcZYLQT0CT/EcgOzjfjYiIxh/V6ABUFUI81Puc225EezTJFadnMLxNUPVt0UFWmh6FwvluREQ0HgkCFGc1pMCx3qckUYDPyX5vH2J4m6BqGztQ1t98t1gIQiYN1eQqQlVERERDU+yVkPx9571VeMw41dxZpIrGF4a3CepUcwcqPOaznpcCZ1aZcr4bERGNU4pjylnz3spdJtQ2dRSvqHGE4W0CkhUVjf5Y/+Gt9SCHTImIaFzLzntT+sx7q/CYcZLhDQDD24TUHIjCZtbBoNP0PaDIkPxHIbumF6UuIiKinAgCFMeUPvPeKjxmNPijkBV1kBdODgUNb9u3b8fVV1+NlStXYsuWLWcdP3LkCNauXYtVq1bhvvvuQyaTXUXS0tKCm266CVdeeSW+/vWvIx6PAwD27duHiy66CGvWrMGaNWvwne98p5Dll6xTzZ2o9J59102M1EHV2wC9pQhVERER5U5x9J33ZtBpYDPr0RzgooWChTe/34/NmzfjySefxNatW/H000+jtrbvZrN33303Nm7ciJ07d0JVVTzzzDMAgO9973u48cYbsWPHDixcuBCPPvooAODQoUO45ZZbsG3bNmzbtg0PPfRQocovabVNnShz9Tdkegiya1oRKiIiIhoexTkVov9on3lvlVy0AKCA4W3Pnj24+OKL4XA4YDKZsGrVKuzYsaP3eHNzMxKJBJYuXQoAWLt2LXbs2IF0Oo233noLq1at6vM8ABw8eBCvv/46ampqcPvtt6O1tbVQ5Ze02qYOVLj7CW8tB6EwvBERUQlQjQ5A0kDobO59rtxtQm0Tw1vBwlsgEIDX+9H2Sz6fD36/f8DjXq8Xfr8f7e3tsFgs0Gg0fZ4HAKvViptvvhnbt2/HihUrsGHDhkKVX7JUVUVdSz/DpokuiLEAFHtlcQojIiIaJsU5HZrWQ72PK9wWrjhFAcOboih9uvurqtrn8UDHP3kegN7H3//+97Fy5UoAwA033IDa2lpEoxz7/rhgew8kSYTVpOvzvNR2BIpzKiBKRaqMiIhoeGTXVEgfC2+VXjPqWjqhqpN70ULBwlt5eTmCwWDv42AwCJ/PN+DxUCgEn88Hl8uFaDQKWZb7vE5RFDz22GO9z39IkhhGPu5kcyeqvGcvSNC0HoTs5JApERGVDsUxBWKkHkgnAABWkw6SJCLY3lPkyoqrYOHt0ksvxd69exGJRNDT04MXX3wRy5cv7z1eVVUFvV6P/fv3AwC2bduG5cuXQ6vVYtmyZXj++ecBAFu3bsXy5cshiiJ27dqFnTt39j6/ZMkSmEz9bwE1WZ1q7mdnBVWB1HYYint6UWoiIiIaEY0Oiq0CUvB471NVXgtOtUzueW8FC29lZWXYsGED1q1bh2uvvRbXXHMNFi9ejPXr1+PgwYMAgE2bNuGhhx7ClVdeie7ubqxbtw4A8N3vfhfPPPMMrr76arz99tv4X//rfwEAfvjDH+KJJ57A6tWr8fvf/x4PPPBAocovWScaz95ZQQydgqozQzXYilQVERHRyCjOqZBaPho6LXOZJn2zXs3Qp4xcTU0Nampq+jz3+OOP9/5+3rx5ePbZZ896XVVVFX7729+e9fycOXPw1FNP5b/QCUJVVZxq7sQVF/TdQUHT9C5k98wiVUVERDRysns6dIdf6H1c5bXgg9PhIlZUfNxhYQIJdvQgIytwWvUfPamqkJregeydXbzCiIiIRkg1eyBkEhCiAQDA1HIrTjS0Q5nEOy0wvE0gR+simFZh67NaV+hsBhQZqsU7yCuJiIjGKUGA7JoBqTU75cpq0sGg16A5GCtyYcXD8DaBHDkdQbWv70pTTdM7ULyzgU+0XyEiIioVsmcmNI37ex9PLbfiSF2kiBUVF8PbBHK4LoKpZX0XJUiN70L2zCpSRURERKOnOKdC7GgCEtnertU+Kw6fZnijEpdIZdAUiKLqYzsrCLEQxEQnd1UgIqLSJmmhuKZD0/I+AGBamRVH6ibvogWGtwmi9kyLEK3mo6bFUtO7kD0zAYEfMxERlTbZMwtS4zsAgDK3GeHOBGLdqSJXVRz8qT5BHKmLoNpn7fOcpmEfZA9XmRIRUemT3TMghWqBdAKSKGBKmRVH69uLXVZRMLxNEB+cCmNq2UfhTYj6IcYjULglFhERTQQaPRR7Ve+q02qvBUcm6bw3hrcJQFVVHG9o7xPeNHV/heybC4j8iImIaGLIrjrNDp1OLbfi8CSd98af7BNAaygOrUaE3XKmOa+qQlP3JjJlc4tbGBERUR7JntmQ/IeBTBJTy6yobeyALCvFLmvMMbxNAB+cCmNa+UctQsTwKUAAVGtZEasiIiLKM50pO3Ta9C5MBi2cVgNONk++TeoZ3iaAt48GMLPK3vtYU/cmZN88NuYlIqIJR/bNg/b0HgDArGo73jkWKHJFY4/hrcTJiooDJ4KYXe3IPqHI0DS+DZlDpkRENAHJnpkQ2xsgdLdjdrUD+4/4i13SmGN4K3GnmjtgMWl757tJze9BMbmhGh3FLYyIiKgQJC1k7xxo6vdheoUNp1u70J1IF7uqMcXwVuLeORbArA/vugHQnvgz5MpFxSuIiIiowOSyedCc3gOdRsTUMisOnZxcq04Z3krc/iMBzK5yAACErlaIna2QvWzMS0REE5dirwIyCYgdjZhVbcf+o5Nr6JThrYR1J9I41dKJGZXZlabaE7uRqVgAiNIQryQiIiphggC57Fxoav+C2dWOSbdogeGthB06md1VQaeVgEwSmvp9HDIlIqJJIVO5CJrGt1FuBWI9afgj3cUuacwwvJWw/Uf9mFWdbRGiqd8HxV4J1WAb4lVEREQTgN4C2T0TutN7MGeKA+9OortvDG8lSlVVvHXYjzlTnICqQHtsFzJVS4pdFhER0ZjJVC2B9sQrmFNlx96DrcUuZ8wwvJWoE40dEASg3GWC1PweIIhQnFOLXRYREdGYUW0VgNaIxfomHKkLI9qdKnZJY4LhrUT95Z0mLJzlgQBAe+g5pKd+ijsqEBHRpJOpWgrTqT9jdrVj0tx9Y3grQYqi4rX3mrFolgdS6yEIchqKZ1axyyIiIhpzsncOhKgfF5d14y/vNBW7nDHB8FaCjtRFYDRoUOY0QvvBc8hMXca7bkRENDmJEjLTLsTc0J9xorEDnbFksSsqOIa3ErT7nSYsnOmG1HYYQiIK2XdOsUsiIiIqGrl8AaR4AMsr49hzoKXY5RQcw1uJkWUFew60YNEMJ3TvPo30zMsAgR8jERFNYqKEzLSL8Gn5r9i9v7HY1RQcf+qXmHePB+Gw6FEWfAuqxsC5bkRERADksnNhUqLQBI+hLRwvdjkFxfBWYrb+5SQuOccK3QfPIT17Oee6ERERAYAoIjPjUnzZ/Baee/VEsaspKIa3ElLf1oXTLZ24ILEXsncWVIu32CURERGNG4p3DvQWK1Lvv4DuRLrY5RQMw1sJ2br7JK6ckYS26R2kp19a7HKIiIjGF0GAOu+zuEJ3ELtffa/Y1RQMw1uJ6Igmsf9gHS6OPIf0OZ8DdMZil0RERDTuqEYHusuWQLNvCzKyUuxyCoLhrUQ8v+c0bna9B9U5BYpnZrHLISIiGrcMcy+GW+jCoed/X+xSCoLhrQR0xpI4veclTFMbs4sUiIiIaGCiBh0zvgDtgT+iu+VksavJO4a3ErDt97uwVr8XmUU1gEZX7HKIiIjGvfJpU3BIuxj1//kvUJLdxS4nrxjexrnaIyexpPEpZOZ+AarVV+xyiIiISoMgoGrxhaiNW9H87CaocqbYFeUNw9s4lo52oOOP/4Iu90KIFXOKXQ4REVFJcdr0CPougr8tjMB//xiqIhe7pLxgeBunMl1h1P7bt9ECH5yLLit2OURERCXpvHkVeC62EB0tDQi98AuoqlrskkaN4W0cSkdaUPfL/4P3Yl6UXfgFCCI/JiIiopHQ6SSsuGA6/jNwLmINxxH44/+Dkk4Wu6xRYSoYZ2JH9qLx3+/BG9FqeM9bAYtJW+ySiIiISprPZcLCuZV4un0J0t1RtDxxPzLR9mKXNWIMb+OEkuxB8IV/Q2Dnr/B88nyYZp+PCo+52GURERFNCOdOc8LltOAZ/xwI7qlo+uU3ETuyt9hljQjDW5Gpqorowb+g8bF/QGdbM37XsQxl02di/gx3sUsjIiKaOAQBly6ugMOqx2+O2KG5YA0iLz+Btv/6IdKdgWJXNywFDW/bt2/H1VdfjZUrV2LLli1nHT9y5AjWrl2LVatW4b777kMmk13G29LSgptuuglXXnklvv71ryMejwMAurq6cNttt+Gqq67CTTfdhGAwWMjyC0qVM4geehXNv/wmOt74PRrKVuDfT0/FpxZNY3AjIiIqAEEQcNHCCsyutuMXr4TROvdvIOpNaH78Wwj+6VGkO0ojxBUsvPn9fmzevBlPPvkktm7diqeffhq1tbV9zrn77ruxceNG7Ny5E6qq4plnngEAfO9738ONN96IHTt2YOHChXj00UcBAI888giWLVuGF154AV/+8pfx4IMPFqr8glBVFcm2Uwi//AQafno7Ot/cji7fUmzpWIY3myWsvmwGplbYil0mERHRhLZwlgdXXDAFu/Y145l6H5RLb4GSSaP5l99Ey5Z/QuyD16GkEsUuc0AFC2979uzBxRdfDIfDAZPJhFWrVmHHjh29x5ubm5FIJLB06VIAwNq1a7Fjxw6k02m89dZbWLVqVZ/nAWD37t2oqakBAFxzzTV49dVXkU6nC/UWRk1VZKTCzYge/AsC23+Chp/ejranH0I0FEBd2WfxG/9C/P6Qgvkz3bjqkmlwWPXFLpmIiGhSKHObsGb5LHjsBvx61yn8rtaD5vk3QfXMROe+51D/yC1o+d130b53K3oaPoCS6il2yb00hbpwIBCA1+vtfezz+XDgwIEBj3u9Xvj9frS3t8NisUCj0fR5/pOv0Wg0sFgsiEQiKCsrG7QWWc425Wtra8vPmxuAkuhG9L2XkO7wI9MZRLrDD1FvhtbhhdZdhUbLMrxyNIZMg4pyTxfOqbTD6zQCQgaBcLigtREREdHZyhyAd4kLraEYdu8/jbb2blgNlai57CLY0kEEDr2D9OsvINMZhGS2Q2P3Qe+bBttFNRAEoSA1fZhXPswvn1Sw8KYoSp83papqn8cDHf/keQAG/J+jqirEHHqgfTg37qabbhrWeyAiIqLJ6fe/H+qMHxW8hmAwiGnTpp31fMHCW3l5Od5+++0+Bfh8vj7HP77gIBQKwefzweVyIRqNQpZlSJLU53U+nw+hUAjl5eXIZDKIx+NwOBxD1rJw4UJs2bIFXq8XkiTl700SERER5ZksywgGg1i4cGG/xwsW3i699FL85Cc/QSQSgdFoxIsvvogf/OAHvcerqqqg1+uxf/9+XHDBBdi2bRuWL18OrVaLZcuW4fnnn0dNTQ22bt2K5cuXAwBWrFiBrVu34vbbb8fzzz+PZcuWQasduomtwWDAsmXLCvVWiYiIiPKqvztuHxLUAm7ytX37dvziF79AOp3Gddddh/Xr12P9+vW48847sWjRIhw9ehT3338/YrEYFixYgIceegg6nQ7Nzc245557EA6HUVFRgf/3//4f7HY7Ojo6cM8996CxsRFWqxWbNm1CdXV1oconIiIiGncKGt6IiIiIKL+4wwIRERFRCWF4IyIiIiohDG9EREREJYThjYiIiKiEMLwRERERlRCGN+r105/+FKtXr8bq1avx8MMPA8juUVtTU4OVK1di8+bNRa6QBvOjH/0IV199NVavXo1f//rXAPj5laIf/vCHuOeeewDw8yslN998M1avXo01a9ZgzZo1eP/99/n5lZBXXnkFa9euxVVXXYUHHngAwDj/86cSqar6xhtvqF/5ylfUZDKpplIpdd26der27dvVFStWqA0NDWo6nVZvueUWdffu3cUulfrx5ptvqtdff72aTqfVnp4e9YorrlCPHDnCz6/E7NmzR73ooovU//N//o/a09PDz69EKIqiXn755Wo6ne59jp9f6WhoaFAvv/xytbW1VU2lUuoNN9yg7t69e1x/frzzRgAAr9eLe+65BzqdDlqtFrNmzUJdXR2mTZuGKVOmQKPRoKamBjt27Ch2qdSPCy+8EE888QQ0Gg3C4TBkWUZXVxc/vxLS0dGBzZs34/bbbwcAHDhwgJ9fiTh16hQA4JZbbsEXv/hF/O53v+PnV0J27dqFq6++GuXl5dBqtdi8eTOMRuO4/vwY3ggAMGfOHCxduhQAUFdXhxdeeAGCIMDr9fae4/P54Pf7i1QhDUWr1eLHP/4xVq9ejUsuuQSBQICfXwnZuHEjNmzYAJvNBgD8/EpIV1cXLrnkEvzsZz/Db37zGzz11FNoaWnh51ci6uvrIcsybr/9dqxZswZPPvnkuP/zx/BGfZw4cQK33HILvv3tb2PKlCkQBKH3mKqqfR7T+HPnnXdi7969aG1tRV1dHT+/EvFf//VfqKiowCWXXNL7nKIo/PxKxHnnnYeHH34YVqsVLpcL1113HX784x/z8ysRsixj7969+Od//mc8/fTTOHDgABobG8f151ewjemp9Ozfvx933nkn7r33XqxevRr79u1DMBjsPR4MBuHz+YpYIQ3k5MmTSKVSOPfcc2E0GrFy5Urs2LEDkiT1nsPPb/x6/vnnEQwGsWbNGnR2dqK7uxvNzc38/ErE22+/jXQ63Ru+VVVFVVUV//4sER6PB5dccglcLhcA4POf//y4//uTd94IANDa2opvfOMb2LRpE1avXg0AWLJkCU6fPt17S/m5557D8uXLi1wp9aepqQn3338/UqkUUqkUXn75ZVx//fX8/ErEr3/9azz33HPYtm0b7rzzTnz2s5/FL3/5S35+JSIajeLhhx9GMplELBbDH//4R/zv//2/+fmViCuuuAKvv/46urq6IMsyXnvtNVx55ZXj+vPjnTcCAPzqV79CMpnEv/zLv/Q+d/311+Nf/uVfcMcddyCZTGLFihW48sori1glDWTFihU4cOAArr32WkiShJUrV2L16tVwuVz8/EqUXq/nn78SccUVV+D999/HtddeC0VRcOONN+K8887j51cilixZgltvvRU33ngj0uk0LrvsMtxwww2YOXPmuP38BFVV1WIXQURERES54bApERERUQlheCMiIiIqIQxvRERERCWE4Y2IiIiohDC8EREREZUQhjciIiKiEsLwRkRERFRC2KSXiGgAiqLgn//5n/H+++8jHo9DVVU88MADmDFjBr7zne+goaEBDocDXq8Xc+bMwR133IGTJ0/iwQcfREdHB2RZxs0334zrrruu2G+FiCYQhjciogG8//77CAQCePrppyGKIv7t3/4Njz/+OEwmE2bPno1f/OIXCAQCWLt2LebMmYNMJoM777wTDz/8MBYsWIBoNIqvfOUrmD17NpYuXVrst0NEEwTDGxHRAM477zzY7XY89dRTaGxsxJtvvgmz2Yy33noLf/zjHwEAPp+vd9ucuro6NDQ04N577+29RiKRwOHDhxneiChvGN6IiAawe/duPPjgg/ja176Gz33uc5g5cyb++7//GxqNBh/fWVAUs9OHZVmG1WrFtm3beo+FQiFYrdYxr52IJi4uWCAiGsAbb7yBK664AjfeeCMWLlyIl156CbIsY8WKFXj22WcBAO3t7XjppZcgCAJmzJgBg8HQG95aW1txzTXX4NChQ8V8G0Q0wXBjeiKiAZw8eRLf/OY3IcsyMpkMLrvsMrz44ovYtm0b7r///t4FC6qq4jOf+QxuvfVWHD16tHfBQiaTwbp163DDDTcU+60Q0QTC8EZENExbtmzB/Pnzcd555yGVSuHGG2/EHXfcgRUrVhS7NCKaBDjnjYhomGbPno0f/OAHUBQF6XQaV155JYMbEY0Z3nkjIiIiKiFcsEBERERUQhjeiIiIiEoIwxsRERFRCWF4IyIiIiohDG9EREREJYThjYiIiKiE/P8mxVvWxJwoMAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "sns.set(style=\"white\")\n",
    "import matplotlib\n",
    "matplotlib.rcParams['figure.figsize'] = (10.0 , 6.0)\n",
    "sns.kdeplot( data['age'].loc[data['y'] == 0], hue=data['y'], shade=True)\n",
    "sns.kdeplot( data['age'].loc[data['y'] == 1], hue=data['y'], shade=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    36548.000000\n",
       "mean        37.975676\n",
       "std         10.686491\n",
       "min         20.000000\n",
       "25%         29.000000\n",
       "50%         38.000000\n",
       "75%         47.000000\n",
       "max         56.000000\n",
       "Name: age, dtype: float64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['age'].loc[data['y'] == 0].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    4640.000000\n",
       "mean       38.261207\n",
       "std        10.112340\n",
       "min        21.000000\n",
       "25%        29.000000\n",
       "50%        38.000000\n",
       "75%        47.000000\n",
       "max        55.000000\n",
       "Name: age, dtype: float64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['age'].loc[data['y'] == 1].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "table=pd.crosstab(data.education,data.y)\n",
    "table.div(table.sum(1).astype(float), axis=0).plot(kind='bar', stacked=True)\n",
    "plt.title('Stacked Bar Chart of Education vs Default')\n",
    "plt.xlabel('Education')\n",
    "plt.ylabel('Proportion of Applicants')\n",
    "plt.savefig('education_vs_def_stack')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='years_with_current_employer', ylabel='Density'>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.kdeplot( data['years_with_current_employer'].loc[data['y'] == 0], hue=data['y'], shade=True)\n",
    "sns.kdeplot( data['years_with_current_employer'].loc[data['y'] == 1], hue=data['y'], shade=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    36548.000000\n",
       "mean        13.447521\n",
       "std          8.063183\n",
       "min          0.000000\n",
       "25%          6.000000\n",
       "50%         13.000000\n",
       "75%         20.000000\n",
       "max         27.000000\n",
       "Name: years_with_current_employer, dtype: float64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['years_with_current_employer'].loc[data['y'] == 0].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    4640.000000\n",
       "mean       14.356034\n",
       "std         8.722690\n",
       "min         0.000000\n",
       "25%         7.000000\n",
       "50%        14.000000\n",
       "75%        22.000000\n",
       "max        29.000000\n",
       "Name: years_with_current_employer, dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['years_with_current_employer'].loc[data['y'] == 1].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='years_at_current_address', ylabel='Density'>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.kdeplot( data['years_at_current_address'].loc[data['y'] == 0], hue=data['y'], shade=True)\n",
    "sns.kdeplot( data['years_at_current_address'].loc[data['y'] == 1], hue=data['y'], shade=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    36548.000000\n",
       "mean        15.497729\n",
       "std          9.245698\n",
       "min          0.000000\n",
       "25%          7.000000\n",
       "50%         16.000000\n",
       "75%         24.000000\n",
       "max         31.000000\n",
       "Name: years_at_current_address, dtype: float64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['years_at_current_address'].loc[data['y'] == 0].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    4640.000000\n",
       "mean       14.500000\n",
       "std         8.637087\n",
       "min         0.000000\n",
       "25%         7.000000\n",
       "50%        15.000000\n",
       "75%        22.000000\n",
       "max        29.000000\n",
       "Name: years_at_current_address, dtype: float64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['years_at_current_address'].loc[data['y'] == 1].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='household_income', ylabel='Density'>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.kdeplot( data['household_income'].loc[data['y'] == 0], hue=data['y'], shade=True)\n",
    "sns.kdeplot( data['household_income'].loc[data['y'] == 1], hue=data['y'], shade=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    36548.000000\n",
       "mean       128.300646\n",
       "std         66.317774\n",
       "min         14.000000\n",
       "25%         71.000000\n",
       "50%        128.000000\n",
       "75%        186.000000\n",
       "max        242.000000\n",
       "Name: household_income, dtype: float64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['household_income'].loc[data['y'] == 0].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    4640.000000\n",
       "mean      229.554957\n",
       "std       124.486932\n",
       "min        14.000000\n",
       "25%       122.000000\n",
       "50%       231.000000\n",
       "75%       336.000000\n",
       "max       446.000000\n",
       "Name: household_income, dtype: float64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['household_income'].loc[data['y'] == 1].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='debt_to_income_ratio', ylabel='Density'>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.kdeplot( data['debt_to_income_ratio'].loc[data['y'] == 0], hue=data['y'], shade=True)\n",
    "sns.kdeplot( data['debt_to_income_ratio'].loc[data['y'] == 1], hue=data['y'], shade=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    36548.000000\n",
       "mean        15.514879\n",
       "std          8.664839\n",
       "min          0.400386\n",
       "25%          8.040746\n",
       "50%         15.530198\n",
       "75%         22.934795\n",
       "max         30.599571\n",
       "Name: debt_to_income_ratio, dtype: float64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['debt_to_income_ratio'].loc[data['y'] == 0].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    4640.000000\n",
       "mean       21.810028\n",
       "std        11.108836\n",
       "min         2.407115\n",
       "25%        12.293277\n",
       "50%        21.830901\n",
       "75%        31.243570\n",
       "max        41.293984\n",
       "Name: debt_to_income_ratio, dtype: float64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['debt_to_income_ratio'].loc[data['y'] == 1].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='credit_card_debt', ylabel='Density'>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.kdeplot( data['credit_card_debt'].loc[data['y'] == 0], hue=data['y'], shade=True)\n",
    "sns.kdeplot( data['credit_card_debt'].loc[data['y'] == 1], hue=data['y'], shade=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    36548.000000\n",
       "mean         7.810996\n",
       "std          8.427304\n",
       "min          0.005668\n",
       "25%          1.651448\n",
       "50%          4.724167\n",
       "75%         11.125912\n",
       "max         55.343899\n",
       "Name: credit_card_debt, dtype: float64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['credit_card_debt'].loc[data['y'] == 0].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    4640.000000\n",
       "mean       23.489629\n",
       "std        24.276876\n",
       "min         0.056798\n",
       "25%         5.807341\n",
       "50%        14.527972\n",
       "75%        33.632070\n",
       "max       149.015859\n",
       "Name: credit_card_debt, dtype: float64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['credit_card_debt'].loc[data['y'] == 1].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='other_debt', ylabel='Density'>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.kdeplot( data['other_debt'].loc[data['y'] == 0], hue=data['y'], shade=True)\n",
    "sns.kdeplot( data['other_debt'].loc[data['y'] == 1], hue=data['y'], shade=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    36548.000000\n",
       "mean        12.127154\n",
       "std         11.343904\n",
       "min          0.021759\n",
       "25%          3.551734\n",
       "50%          8.479551\n",
       "75%         17.334192\n",
       "max         68.666080\n",
       "Name: other_debt, dtype: float64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['other_debt'].loc[data['y'] == 0].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    4640.000000\n",
       "mean       26.605214\n",
       "std        26.301693\n",
       "min         0.222414\n",
       "25%         7.536917\n",
       "50%        17.818418\n",
       "75%        36.710018\n",
       "max       159.198221\n",
       "Name: other_debt, dtype: float64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['other_debt'].loc[data['y'] == 1].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3. Data Preparation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "cat_vars=['education']\n",
    "for var in cat_vars:\n",
    "    cat_list='var'+'_'+var\n",
    "    cat_list = pd.get_dummies(data[var], prefix=var)\n",
    "    data1=data.join(cat_list)\n",
    "    data=data1\n",
    "cat_vars=['education']\n",
    "data_vars=data.columns.values.tolist()\n",
    "to_keep=[i for i in data_vars if i not in cat_vars]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\pandas\\core\\frame.py:4315: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  errors=errors,\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array(['age', 'years_with_current_employer', 'years_at_current_address',\n",
       "       'household_income', 'debt_to_income_ratio', 'credit_card_debt',\n",
       "       'other_debt', 'y', 'education_basic', 'education_high.school',\n",
       "       'education_illiterate', 'education_professional.course',\n",
       "       'education_university.degree'], dtype=object)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_final=data[to_keep]\n",
    "data_final.drop(['loan_applicant_id'], axis=1, inplace=True)\n",
    "data_final.columns.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m length of oversampled data is  48912\n",
      "\u001b[1m Number of no default in oversampled data 24456\n",
      "\u001b[1m Number of default 24456\n",
      "\u001b[1m Proportion of no default data in oversampled data is  0.5\n",
      "\u001b[1m Proportion of default data in oversampled data is  0.5\n"
     ]
    }
   ],
   "source": [
    "X = data_final.loc[:, data_final.columns != 'y']\n",
    "y = data_final.loc[:, data_final.columns == 'y']\n",
    "\n",
    "from imblearn.over_sampling import SMOTE\n",
    "\n",
    "os = SMOTE(random_state=42)\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)\n",
    "columns = X_train.columns\n",
    "\n",
    "os_data_X,os_data_y = os.fit_sample(X_train, y_train)\n",
    "os_data_X = pd.DataFrame(data=os_data_X,columns=columns )\n",
    "os_data_y = pd.DataFrame(data=os_data_y,columns=['y'])\n",
    "# we can Check the numbers of our data\n",
    "print(\"\\033[1m length of oversampled data is \",len(os_data_X))\n",
    "print(\"\\033[1m Number of no default in oversampled data\",len(os_data_y[os_data_y['y']==0]))\n",
    "print(\"\\033[1m Number of default\",len(os_data_y[os_data_y['y']==1]))\n",
    "print(\"\\033[1m Proportion of no default data in oversampled data is \",len(os_data_y[os_data_y['y']==0])/len(os_data_X))\n",
    "print(\"\\033[1m Proportion of default data in oversampled data is \",len(os_data_y[os_data_y['y']==1])/len(os_data_X))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:70: FutureWarning: Pass n_features_to_select=9 as keyword args. From version 0.25 passing these as positional arguments will result in an error\n",
      "  FutureWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py:764: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py:764: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py:764: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[False  True False  True  True False  True  True  True  True  True  True]\n",
      "[2 1 3 1 1 4 1 1 1 1 1 1]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py:764: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    }
   ],
   "source": [
    "data_final_vars=data_final.columns.values.tolist()\n",
    "y=['y']\n",
    "X=[i for i in data_final_vars if i not in y]\n",
    "\n",
    "from sklearn.feature_selection import RFE\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "logreg = LogisticRegression()\n",
    "\n",
    "rfe = RFE(logreg, 9)\n",
    "rfe = rfe.fit(os_data_X, os_data_y.values.ravel())\n",
    "print(rfe.support_)\n",
    "print(rfe.ranking_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['age', 'years_with_current_employer', 'years_at_current_address',\n",
       "       'household_income', 'debt_to_income_ratio', 'credit_card_debt',\n",
       "       'other_debt', 'education_basic', 'education_high.school',\n",
       "       'education_illiterate', 'education_professional.course',\n",
       "       'education_university.degree'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "os_data_X.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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>Feature</th>\n",
       "      <th>Importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>years_with_current_employer</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>household_income</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>debt_to_income_ratio</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>other_debt</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>education_basic</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>education_high.school</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>education_illiterate</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>education_professional.course</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>education_university.degree</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>age</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>years_at_current_address</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>credit_card_debt</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          Feature  Importance\n",
       "1     years_with_current_employer           1\n",
       "3                household_income           1\n",
       "4            debt_to_income_ratio           1\n",
       "6                      other_debt           1\n",
       "7                 education_basic           1\n",
       "8           education_high.school           1\n",
       "9            education_illiterate           1\n",
       "10  education_professional.course           1\n",
       "11    education_university.degree           1\n",
       "0                             age           2\n",
       "2        years_at_current_address           3\n",
       "5                credit_card_debt           4"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_X1 = pd.DataFrame({\n",
    "    'Feature': os_data_X.columns,\n",
    "    'Importance': rfe.ranking_},)\n",
    "data_X1.sort_values(by=['Importance'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['years_with_current_employer', 'household_income', 'debt_to_income_ratio', 'other_debt', 'education_basic', 'education_high.school', 'education_illiterate', 'education_professional.course', 'education_university.degree']\n",
      "9\n"
     ]
    }
   ],
   "source": [
    "cols=[]\n",
    "for i in range (0, len(data_X1[\"Importance\"])):\n",
    "    if data_X1[\"Importance\"][i] == 1:\n",
    "        cols.append(data_X1[\"Feature\"][i])\n",
    "print(cols)\n",
    "print(len(cols))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "X=os_data_X[cols]\n",
    "y=os_data_y['y']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4. Modeling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization terminated successfully.\n",
      "         Current function value: 0.269908\n",
      "         Iterations 8\n",
      "                                Results: Logit\n",
      "==============================================================================\n",
      "Model:                   Logit                Pseudo R-squared:     0.611     \n",
      "Dependent Variable:      y                    AIC:                  26421.4504\n",
      "Date:                    2021-10-12 01:51     BIC:                  26500.6304\n",
      "No. Observations:        48912                Log-Likelihood:       -13202.   \n",
      "Df Model:                8                    LL-Null:              -33903.   \n",
      "Df Residuals:            48903                LLR p-value:          0.0000    \n",
      "Converged:               1.0000               Scale:                1.0000    \n",
      "No. Iterations:          8.0000                                               \n",
      "------------------------------------------------------------------------------\n",
      "                               Coef.  Std.Err.    z     P>|z|   [0.025  0.975]\n",
      "------------------------------------------------------------------------------\n",
      "years_with_current_employer    0.0189   0.0018  10.3326 0.0000  0.0153  0.0225\n",
      "household_income               0.0173   0.0002  71.5619 0.0000  0.0168  0.0178\n",
      "debt_to_income_ratio           0.1207   0.0022  56.0241 0.0000  0.1165  0.1250\n",
      "other_debt                    -0.0362   0.0013 -27.5726 0.0000 -0.0388 -0.0336\n",
      "education_basic               -6.0912   0.0696 -87.5289 0.0000 -6.2276 -5.9548\n",
      "education_high.school         -6.1076   0.0697 -87.6225 0.0000 -6.2442 -5.9710\n",
      "education_illiterate          -6.0203   0.0690 -87.3070 0.0000 -6.1554 -5.8851\n",
      "education_professional.course -6.0154   0.0687 -87.4977 0.0000 -6.1501 -5.8806\n",
      "education_university.degree   -6.0792   0.0691 -88.0078 0.0000 -6.2146 -5.9438\n",
      "==============================================================================\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import statsmodels.api as sm\n",
    "logit_model=sm.Logit(y,X)\n",
    "result=logit_model.fit()\n",
    "print(result.summary2())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "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>p_value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>years_with_current_employer</th>\n",
       "      <td>5.019950e-25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>household_income</th>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>debt_to_income_ratio</th>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>other_debt</th>\n",
       "      <td>2.369519e-167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_basic</th>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_high.school</th>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_illiterate</th>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_professional.course</th>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_university.degree</th>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                     p_value\n",
       "years_with_current_employer     5.019950e-25\n",
       "household_income                0.000000e+00\n",
       "debt_to_income_ratio            0.000000e+00\n",
       "other_debt                     2.369519e-167\n",
       "education_basic                 0.000000e+00\n",
       "education_high.school           0.000000e+00\n",
       "education_illiterate            0.000000e+00\n",
       "education_professional.course   0.000000e+00\n",
       "education_university.degree     0.000000e+00"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pvalue = pd.DataFrame(result.pvalues,columns={'p_value'},)\n",
    "pvalue"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['years_with_current_employer', 'household_income', 'debt_to_income_ratio', 'other_debt', 'education_basic', 'education_high.school', 'education_illiterate', 'education_professional.course', 'education_university.degree']\n",
      "9\n"
     ]
    }
   ],
   "source": [
    "pvs=[]\n",
    "for i in range (0, len(pvalue[\"p_value\"])):\n",
    "    if pvalue[\"p_value\"][i] < 0.05:\n",
    "        pvs.append(pvalue.index[i])\n",
    "\n",
    "if 'const' in pvs:\n",
    "    pvs.remove('const')\n",
    "else:\n",
    "    pvs \n",
    "print(pvs)\n",
    "print(len(pvs))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization terminated successfully.\n",
      "         Current function value: 0.269908\n",
      "         Iterations 8\n",
      "                           Logit Regression Results                           \n",
      "==============================================================================\n",
      "Dep. Variable:                      y   No. Observations:                48912\n",
      "Model:                          Logit   Df Residuals:                    48903\n",
      "Method:                           MLE   Df Model:                            8\n",
      "Date:                Tue, 12 Oct 2021   Pseudo R-squ.:                  0.6106\n",
      "Time:                        01:51:29   Log-Likelihood:                -13202.\n",
      "converged:                       True   LL-Null:                       -33903.\n",
      "Covariance Type:            nonrobust   LLR p-value:                     0.000\n",
      "=================================================================================================\n",
      "                                    coef    std err          z      P>|z|      [0.025      0.975]\n",
      "-------------------------------------------------------------------------------------------------\n",
      "years_with_current_employer       0.0189      0.002     10.333      0.000       0.015       0.023\n",
      "household_income                  0.0173      0.000     71.562      0.000       0.017       0.018\n",
      "debt_to_income_ratio              0.1207      0.002     56.024      0.000       0.117       0.125\n",
      "other_debt                       -0.0362      0.001    -27.573      0.000      -0.039      -0.034\n",
      "education_basic                  -6.0912      0.070    -87.529      0.000      -6.228      -5.955\n",
      "education_high.school            -6.1076      0.070    -87.623      0.000      -6.244      -5.971\n",
      "education_illiterate             -6.0203      0.069    -87.307      0.000      -6.155      -5.885\n",
      "education_professional.course    -6.0154      0.069    -87.498      0.000      -6.150      -5.881\n",
      "education_university.degree      -6.0792      0.069    -88.008      0.000      -6.215      -5.944\n",
      "=================================================================================================\n"
     ]
    }
   ],
   "source": [
    "X=os_data_X[pvs]\n",
    "y=os_data_y['y']\n",
    "\n",
    "logit_model=sm.Logit(y,X)\n",
    "result=logit_model.fit()\n",
    "print(result.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py:764: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LogisticRegression()"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn import metrics\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)\n",
    "logreg = LogisticRegression()\n",
    "logreg.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 5. Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of logistic regression classifier on test set: 0.91\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "y_pred = logreg.predict(X_test)\n",
    "print('Accuracy of logistic regression classifier on test set: {:.2f}'.format(accuracy_score(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[7860  169]\n",
      " [1350 6762]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "confusion_matrix = confusion_matrix(y_test, y_pred)\n",
    "print(confusion_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m The result is telling us that we have:  14622 correct predictions\u001b[1m\n",
      "\u001b[1m The result is telling us that we have:  1519 incorrect predictions\u001b[1m\n",
      "\u001b[1m We have a total predictions of:  16141\n"
     ]
    }
   ],
   "source": [
    "print(\"\\033[1m The result is telling us that we have: \",(confusion_matrix[0,0]+confusion_matrix[1,1]),\"correct predictions\\033[1m\")\n",
    "print(\"\\033[1m The result is telling us that we have: \",(confusion_matrix[0,1]+confusion_matrix[1,0]),\"incorrect predictions\\033[1m\")\n",
    "print(\"\\033[1m We have a total predictions of: \",(confusion_matrix.sum()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.85      0.98      0.91      8029\n",
      "           1       0.98      0.83      0.90      8112\n",
      "\n",
      "    accuracy                           0.91     16141\n",
      "   macro avg       0.91      0.91      0.91     16141\n",
      "weighted avg       0.91      0.91      0.91     16141\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "from sklearn.metrics import roc_curve\n",
    "sns.set(style=\"whitegrid\", color_codes=True)\n",
    "logit_roc_auc = roc_auc_score(y_test, logreg.predict(X_test))\n",
    "fpr, tpr, thresholds = roc_curve(y_test, logreg.predict_proba(X_test)[:,1])\n",
    "plt.figure()\n",
    "plt.plot(fpr, tpr, label='Logistic Regression (area = %0.2f)' % logit_roc_auc)\n",
    "plt.plot([0, 1], [0, 1],'r--')\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver operating characteristic')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.savefig('Log_ROC')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 6. Deployment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "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>loan_applicant_id</th>\n",
       "      <th>PD</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>191</td>\n",
       "      <td>0.206329</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>34318</td>\n",
       "      <td>0.041881</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>14932</td>\n",
       "      <td>0.201293</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2776</td>\n",
       "      <td>0.749639</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11915</td>\n",
       "      <td>0.239431</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>21925</td>\n",
       "      <td>0.439802</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>15110</td>\n",
       "      <td>0.034946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>19944</td>\n",
       "      <td>0.061453</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>22850</td>\n",
       "      <td>0.422680</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>37927</td>\n",
       "      <td>0.008191</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   loan_applicant_id        PD\n",
       "0                191  0.206329\n",
       "1              34318  0.041881\n",
       "2              14932  0.201293\n",
       "3               2776  0.749639\n",
       "4              11915  0.239431\n",
       "5              21925  0.439802\n",
       "6              15110  0.034946\n",
       "7              19944  0.061453\n",
       "8              22850  0.422680\n",
       "9              37927  0.008191"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['PD'] = logreg.predict_proba(data[X_train.columns])[:,1]\n",
    "data[['loan_applicant_id', 'PD']].head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['years_with_current_employer', 'household_income',\n",
       "       'debt_to_income_ratio', 'other_debt', 'education_basic',\n",
       "       'education_high.school', 'education_illiterate',\n",
       "       'education_professional.course', 'education_university.degree'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m This new loan applicant has a 4.19% chance of defaulting on a new debt\n"
     ]
    }
   ],
   "source": [
    "new_data = np.array([3,57,14.26,2.993,0,1,0,0,0]).reshape(1, -1)\n",
    "new_pred=logreg.predict_proba(new_data)[:,1][0]\n",
    "print(\"\\033[1m This new loan applicant has a {:.2%}\".format(new_pred), \"chance of defaulting on a new debt\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}