Training_Testing.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.metrics import roc_curve, auc\n",
    "from sklearn.metrics import classification_report, confusion_matrix, accuracy_score\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1819, 25)\n"
     ]
    },
    {
     "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>url</th>\n",
       "      <th>label</th>\n",
       "      <th>url_length</th>\n",
       "      <th>special_character_count</th>\n",
       "      <th>having_ip_address</th>\n",
       "      <th>dns_response_time</th>\n",
       "      <th>handshake_time</th>\n",
       "      <th>data_time</th>\n",
       "      <th>data_length</th>\n",
       "      <th>redirect</th>\n",
       "      <th>...</th>\n",
       "      <th>external_link_count</th>\n",
       "      <th>total_strings_count</th>\n",
       "      <th>average_string_length</th>\n",
       "      <th>max_len_strings</th>\n",
       "      <th>strings_above_avglength</th>\n",
       "      <th>whois_creation_date</th>\n",
       "      <th>whois_expiry_date</th>\n",
       "      <th>whois_last_updated</th>\n",
       "      <th>whois_country</th>\n",
       "      <th>whois_city</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>twitter.com/aQuariusrecOrds</td>\n",
       "      <td>good</td>\n",
       "      <td>27</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1.866579</td>\n",
       "      <td>809.915781</td>\n",
       "      <td>185.235262</td>\n",
       "      <td>301912.0</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>73.0</td>\n",
       "      <td>26397.0</td>\n",
       "      <td>63.022881</td>\n",
       "      <td>46438.0</td>\n",
       "      <td>2436.0</td>\n",
       "      <td>2000-01-21T16:28:17Z</td>\n",
       "      <td>2020-01-21T16:28:17Z</td>\n",
       "      <td>2018-12-07T19:32:35Z</td>\n",
       "      <td>US</td>\n",
       "      <td>CA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>tong464.org/</td>\n",
       "      <td>good</td>\n",
       "      <td>12</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>654.555559</td>\n",
       "      <td>3379.443407</td>\n",
       "      <td>799.807549</td>\n",
       "      <td>162844.0</td>\n",
       "      <td>True</td>\n",
       "      <td>...</td>\n",
       "      <td>18.0</td>\n",
       "      <td>9289.0</td>\n",
       "      <td>11.865217</td>\n",
       "      <td>5491.0</td>\n",
       "      <td>2936.0</td>\n",
       "      <td>2006-04-11T19:49:29Z</td>\n",
       "      <td>2025-04-11T19:49:29Z</td>\n",
       "      <td>2016-04-07T14:34:10Z</td>\n",
       "      <td>country_not_available</td>\n",
       "      <td>state_not_available</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>legacy.com/obituaries/orlandosentinel/obituary...</td>\n",
       "      <td>good</td>\n",
       "      <td>92</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>57.567120</td>\n",
       "      <td>2735.617161</td>\n",
       "      <td>29.102325</td>\n",
       "      <td>129957.0</td>\n",
       "      <td>True</td>\n",
       "      <td>...</td>\n",
       "      <td>58.0</td>\n",
       "      <td>22093.0</td>\n",
       "      <td>44.528946</td>\n",
       "      <td>28934.0</td>\n",
       "      <td>4508.0</td>\n",
       "      <td>1994-02-01T05:00:00Z</td>\n",
       "      <td>2021-12-20T17:37:52Z</td>\n",
       "      <td>2019-04-01T20:15:54Z</td>\n",
       "      <td>US</td>\n",
       "      <td>FL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>natcath.org/ncr_onli.htm</td>\n",
       "      <td>good</td>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1.390219</td>\n",
       "      <td>1009.941101</td>\n",
       "      <td>0.138521</td>\n",
       "      <td>3453.0</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1998-06-18T04:00:00Z</td>\n",
       "      <td>2024-06-17T04:00:00Z</td>\n",
       "      <td>2019-04-19T06:57:43Z</td>\n",
       "      <td>country_not_available</td>\n",
       "      <td>state_not_available</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>wn.com/San_Diego_State_Aztecs</td>\n",
       "      <td>good</td>\n",
       "      <td>29</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1.077890</td>\n",
       "      <td>6096.054077</td>\n",
       "      <td>350.750208</td>\n",
       "      <td>204520.0</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>41.0</td>\n",
       "      <td>8248.0</td>\n",
       "      <td>12.376940</td>\n",
       "      <td>5491.0</td>\n",
       "      <td>2477.0</td>\n",
       "      <td>1995-03-23T05:00:00Z</td>\n",
       "      <td>2029-03-24T04:00:00Z</td>\n",
       "      <td>2019-04-12T19:51:39Z</td>\n",
       "      <td>REDACTED FOR PRIVACY</td>\n",
       "      <td>REDACTED FOR PRIVACY</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 url label  url_length  \\\n",
       "0                        twitter.com/aQuariusrecOrds  good          27   \n",
       "1                                       tong464.org/  good          12   \n",
       "2  legacy.com/obituaries/orlandosentinel/obituary...  good          92   \n",
       "3                           natcath.org/ncr_onli.htm  good          24   \n",
       "4                      wn.com/San_Diego_State_Aztecs  good          29   \n",
       "\n",
       "   special_character_count  having_ip_address  dns_response_time  \\\n",
       "0                        2                  0           1.866579   \n",
       "1                        2                  0         654.555559   \n",
       "2                       12                  0          57.567120   \n",
       "3                        3                  0           1.390219   \n",
       "4                        2                  0           1.077890   \n",
       "\n",
       "   handshake_time   data_time  data_length redirect          ...           \\\n",
       "0      809.915781  185.235262     301912.0    False          ...            \n",
       "1     3379.443407  799.807549     162844.0     True          ...            \n",
       "2     2735.617161   29.102325     129957.0     True          ...            \n",
       "3     1009.941101    0.138521       3453.0    False          ...            \n",
       "4     6096.054077  350.750208     204520.0    False          ...            \n",
       "\n",
       "  external_link_count  total_strings_count average_string_length  \\\n",
       "0                73.0              26397.0             63.022881   \n",
       "1                18.0               9289.0             11.865217   \n",
       "2                58.0              22093.0             44.528946   \n",
       "3                 1.0                  8.0              8.000000   \n",
       "4                41.0               8248.0             12.376940   \n",
       "\n",
       "   max_len_strings  strings_above_avglength   whois_creation_date  \\\n",
       "0          46438.0                   2436.0  2000-01-21T16:28:17Z   \n",
       "1           5491.0                   2936.0  2006-04-11T19:49:29Z   \n",
       "2          28934.0                   4508.0  1994-02-01T05:00:00Z   \n",
       "3              8.0                      0.0  1998-06-18T04:00:00Z   \n",
       "4           5491.0                   2477.0  1995-03-23T05:00:00Z   \n",
       "\n",
       "      whois_expiry_date    whois_last_updated          whois_country  \\\n",
       "0  2020-01-21T16:28:17Z  2018-12-07T19:32:35Z                     US   \n",
       "1  2025-04-11T19:49:29Z  2016-04-07T14:34:10Z  country_not_available   \n",
       "2  2021-12-20T17:37:52Z  2019-04-01T20:15:54Z                     US   \n",
       "3  2024-06-17T04:00:00Z  2019-04-19T06:57:43Z  country_not_available   \n",
       "4  2029-03-24T04:00:00Z  2019-04-12T19:51:39Z   REDACTED FOR PRIVACY   \n",
       "\n",
       "             whois_city  \n",
       "0                    CA  \n",
       "1   state_not_available  \n",
       "2                    FL  \n",
       "3   state_not_available  \n",
       "4  REDACTED FOR PRIVACY  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"final_data7.csv\")\n",
    "print(df.shape)\n",
    "df.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>url</th>\n",
       "      <th>label</th>\n",
       "      <th>url_length</th>\n",
       "      <th>special_character_count</th>\n",
       "      <th>having_ip_address</th>\n",
       "      <th>dns_response_time</th>\n",
       "      <th>handshake_time</th>\n",
       "      <th>data_time</th>\n",
       "      <th>data_length</th>\n",
       "      <th>redirect</th>\n",
       "      <th>...</th>\n",
       "      <th>external_link_count</th>\n",
       "      <th>total_strings_count</th>\n",
       "      <th>average_string_length</th>\n",
       "      <th>max_len_strings</th>\n",
       "      <th>strings_above_avglength</th>\n",
       "      <th>whois_creation_date</th>\n",
       "      <th>whois_expiry_date</th>\n",
       "      <th>whois_last_updated</th>\n",
       "      <th>whois_country</th>\n",
       "      <th>whois_city</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1819</td>\n",
       "      <td>1819</td>\n",
       "      <td>1819.000000</td>\n",
       "      <td>1819.000000</td>\n",
       "      <td>1819.000000</td>\n",
       "      <td>1819.000000</td>\n",
       "      <td>1791.000000</td>\n",
       "      <td>1791.000000</td>\n",
       "      <td>1.791000e+03</td>\n",
       "      <td>1791</td>\n",
       "      <td>...</td>\n",
       "      <td>1792.000000</td>\n",
       "      <td>1811.000000</td>\n",
       "      <td>1811.000000</td>\n",
       "      <td>1.811000e+03</td>\n",
       "      <td>1811.000000</td>\n",
       "      <td>1819</td>\n",
       "      <td>1819</td>\n",
       "      <td>1819</td>\n",
       "      <td>1817</td>\n",
       "      <td>1806</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>1819</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>901</td>\n",
       "      <td>885</td>\n",
       "      <td>911</td>\n",
       "      <td>43</td>\n",
       "      <td>92</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>ligotti.net/showthread.php?t=3887</td>\n",
       "      <td>good</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>cdate_not_available</td>\n",
       "      <td>rdate_not_available</td>\n",
       "      <td>udate_not_available</td>\n",
       "      <td>country_not_available</td>\n",
       "      <td>state_not_available</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>1</td>\n",
       "      <td>1598</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1168</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>544</td>\n",
       "      <td>564</td>\n",
       "      <td>555</td>\n",
       "      <td>1034</td>\n",
       "      <td>1027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>47.843321</td>\n",
       "      <td>6.155580</td>\n",
       "      <td>0.009346</td>\n",
       "      <td>128.337215</td>\n",
       "      <td>1560.233217</td>\n",
       "      <td>310.214560</td>\n",
       "      <td>8.489945e+04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>31.673549</td>\n",
       "      <td>11181.467145</td>\n",
       "      <td>22.892563</td>\n",
       "      <td>1.735832e+04</td>\n",
       "      <td>2830.214798</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>31.768850</td>\n",
       "      <td>4.106997</td>\n",
       "      <td>0.096247</td>\n",
       "      <td>216.588207</td>\n",
       "      <td>4235.769567</td>\n",
       "      <td>635.436642</td>\n",
       "      <td>2.003764e+05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>62.636892</td>\n",
       "      <td>27324.442122</td>\n",
       "      <td>23.622174</td>\n",
       "      <td>4.809249e+04</td>\n",
       "      <td>9010.493238</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.024557</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.267433</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>7.678465</td>\n",
       "      <td>5.600000e+01</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>43.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>23.092508</td>\n",
       "      <td>758.058310</td>\n",
       "      <td>2.724886</td>\n",
       "      <td>1.015800e+04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>3791.000000</td>\n",
       "      <td>16.809524</td>\n",
       "      <td>5.491000e+03</td>\n",
       "      <td>824.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>59.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>228.833914</td>\n",
       "      <td>2049.501777</td>\n",
       "      <td>441.988230</td>\n",
       "      <td>9.391350e+04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>41.000000</td>\n",
       "      <td>11957.000000</td>\n",
       "      <td>31.470965</td>\n",
       "      <td>2.795600e+04</td>\n",
       "      <td>2549.500000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>641.000000</td>\n",
       "      <td>68.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2596.153498</td>\n",
       "      <td>130470.530500</td>\n",
       "      <td>7074.467182</td>\n",
       "      <td>3.869084e+06</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>1018.000000</td>\n",
       "      <td>305185.000000</td>\n",
       "      <td>230.000000</td>\n",
       "      <td>1.091701e+06</td>\n",
       "      <td>106880.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>11 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      url label   url_length  \\\n",
       "count                                1819  1819  1819.000000   \n",
       "unique                               1819     2          NaN   \n",
       "top     ligotti.net/showthread.php?t=3887  good          NaN   \n",
       "freq                                    1  1598          NaN   \n",
       "mean                                  NaN   NaN    47.843321   \n",
       "std                                   NaN   NaN    31.768850   \n",
       "min                                   NaN   NaN     9.000000   \n",
       "25%                                   NaN   NaN    30.000000   \n",
       "50%                                   NaN   NaN    43.000000   \n",
       "75%                                   NaN   NaN    59.000000   \n",
       "max                                   NaN   NaN   641.000000   \n",
       "\n",
       "        special_character_count  having_ip_address  dns_response_time  \\\n",
       "count               1819.000000        1819.000000        1819.000000   \n",
       "unique                      NaN                NaN                NaN   \n",
       "top                         NaN                NaN                NaN   \n",
       "freq                        NaN                NaN                NaN   \n",
       "mean                   6.155580           0.009346         128.337215   \n",
       "std                    4.106997           0.096247         216.588207   \n",
       "min                    2.000000           0.000000           0.024557   \n",
       "25%                    4.000000           0.000000           1.267433   \n",
       "50%                    5.000000           0.000000          23.092508   \n",
       "75%                    8.000000           0.000000         228.833914   \n",
       "max                   68.000000           1.000000        2596.153498   \n",
       "\n",
       "        handshake_time    data_time   data_length redirect  \\\n",
       "count      1791.000000  1791.000000  1.791000e+03     1791   \n",
       "unique             NaN          NaN           NaN        2   \n",
       "top                NaN          NaN           NaN    False   \n",
       "freq               NaN          NaN           NaN     1168   \n",
       "mean       1560.233217   310.214560  8.489945e+04      NaN   \n",
       "std        4235.769567   635.436642  2.003764e+05      NaN   \n",
       "min           0.000000     0.000000  0.000000e+00      NaN   \n",
       "25%           0.000000     0.000000  0.000000e+00      NaN   \n",
       "50%         758.058310     2.724886  1.015800e+04      NaN   \n",
       "75%        2049.501777   441.988230  9.391350e+04      NaN   \n",
       "max      130470.530500  7074.467182  3.869084e+06      NaN   \n",
       "\n",
       "               ...          external_link_count  total_strings_count  \\\n",
       "count          ...                  1792.000000          1811.000000   \n",
       "unique         ...                          NaN                  NaN   \n",
       "top            ...                          NaN                  NaN   \n",
       "freq           ...                          NaN                  NaN   \n",
       "mean           ...                    31.673549         11181.467145   \n",
       "std            ...                    62.636892         27324.442122   \n",
       "min            ...                     0.000000             0.000000   \n",
       "25%            ...                     1.000000            18.000000   \n",
       "50%            ...                    17.000000          3791.000000   \n",
       "75%            ...                    41.000000         11957.000000   \n",
       "max            ...                  1018.000000        305185.000000   \n",
       "\n",
       "       average_string_length  max_len_strings  strings_above_avglength  \\\n",
       "count            1811.000000     1.811000e+03              1811.000000   \n",
       "unique                   NaN              NaN                      NaN   \n",
       "top                      NaN              NaN                      NaN   \n",
       "freq                     NaN              NaN                      NaN   \n",
       "mean               22.892563     1.735832e+04              2830.214798   \n",
       "std                23.622174     4.809249e+04              9010.493238   \n",
       "min                 0.000000     0.000000e+00                 0.000000   \n",
       "25%                 7.678465     5.600000e+01                 5.000000   \n",
       "50%                16.809524     5.491000e+03               824.000000   \n",
       "75%                31.470965     2.795600e+04              2549.500000   \n",
       "max               230.000000     1.091701e+06            106880.000000   \n",
       "\n",
       "        whois_creation_date    whois_expiry_date   whois_last_updated  \\\n",
       "count                  1819                 1819                 1819   \n",
       "unique                  901                  885                  911   \n",
       "top     cdate_not_available  rdate_not_available  udate_not_available   \n",
       "freq                    544                  564                  555   \n",
       "mean                    NaN                  NaN                  NaN   \n",
       "std                     NaN                  NaN                  NaN   \n",
       "min                     NaN                  NaN                  NaN   \n",
       "25%                     NaN                  NaN                  NaN   \n",
       "50%                     NaN                  NaN                  NaN   \n",
       "75%                     NaN                  NaN                  NaN   \n",
       "max                     NaN                  NaN                  NaN   \n",
       "\n",
       "                whois_country           whois_city  \n",
       "count                    1817                 1806  \n",
       "unique                     43                   92  \n",
       "top     country_not_available  state_not_available  \n",
       "freq                     1034                 1027  \n",
       "mean                      NaN                  NaN  \n",
       "std                       NaN                  NaN  \n",
       "min                       NaN                  NaN  \n",
       "25%                       NaN                  NaN  \n",
       "50%                       NaN                  NaN  \n",
       "75%                       NaN                  NaN  \n",
       "max                       NaN                  NaN  \n",
       "\n",
       "[11 rows x 25 columns]"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Quick statistical summary of data\n",
    "df.describe(include='all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "url                          0\n",
      "label                        0\n",
      "url_length                   0\n",
      "special_character_count      0\n",
      "having_ip_address            0\n",
      "dns_response_time            0\n",
      "handshake_time              28\n",
      "data_time                   28\n",
      "data_length                 28\n",
      "redirect                    28\n",
      "protocol                    28\n",
      "http_status_code            28\n",
      "server_name                901\n",
      "scripts_count               27\n",
      "iframe_count                27\n",
      "external_link_count         27\n",
      "total_strings_count          8\n",
      "average_string_length        8\n",
      "max_len_strings              8\n",
      "strings_above_avglength      8\n",
      "whois_creation_date          0\n",
      "whois_expiry_date            0\n",
      "whois_last_updated           0\n",
      "whois_country                2\n",
      "whois_city                  13\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "#check the number of null values at each feature\n",
    "print(df.isnull().sum())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see a huge amount of null values at server, but other null values are less in number so we can remove those points which have null values.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "url                          0\n",
      "label                        0\n",
      "url_length                   0\n",
      "special_character_count      0\n",
      "having_ip_address            0\n",
      "dns_response_time            0\n",
      "handshake_time               0\n",
      "data_time                    0\n",
      "data_length                  0\n",
      "redirect                     0\n",
      "protocol                     0\n",
      "http_status_code             0\n",
      "server_name                848\n",
      "scripts_count                0\n",
      "iframe_count                 0\n",
      "external_link_count          0\n",
      "total_strings_count          0\n",
      "average_string_length        0\n",
      "max_len_strings              0\n",
      "strings_above_avglength      0\n",
      "whois_creation_date          0\n",
      "whois_expiry_date            0\n",
      "whois_last_updated           0\n",
      "whois_country                0\n",
      "whois_city                   0\n",
      "dtype: int64\n",
      "shape: (1757, 25)\n"
     ]
    }
   ],
   "source": [
    "df = df.dropna(axis=0, subset=['strings_above_avglength' , 'scripts_count' , 'data_time','whois_city'])\n",
    "df= pd.DataFrame(df)\n",
    "print(df.isnull().sum())\n",
    "print(\"shape:\", df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "# we interpolate those datapoints where server column is null\n",
    "df = df.interpolate()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = df.iloc[:,2:] #remove the unique url feature as that doesn't signify anything\n",
    "label = df[\"label\"]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1757, 10)\n"
     ]
    },
    {
     "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>having_ip_address</th>\n",
       "      <th>redirect</th>\n",
       "      <th>protocol</th>\n",
       "      <th>http_status_code</th>\n",
       "      <th>server_name</th>\n",
       "      <th>whois_creation_date</th>\n",
       "      <th>whois_expiry_date</th>\n",
       "      <th>whois_last_updated</th>\n",
       "      <th>whois_country</th>\n",
       "      <th>whois_city</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "      <td>https</td>\n",
       "      <td>200.0</td>\n",
       "      <td>tsa_k</td>\n",
       "      <td>2000-01-21T16:28:17Z</td>\n",
       "      <td>2020-01-21T16:28:17Z</td>\n",
       "      <td>2018-12-07T19:32:35Z</td>\n",
       "      <td>US</td>\n",
       "      <td>CA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "      <td>http</td>\n",
       "      <td>200.0</td>\n",
       "      <td>Apache</td>\n",
       "      <td>2006-04-11T19:49:29Z</td>\n",
       "      <td>2025-04-11T19:49:29Z</td>\n",
       "      <td>2016-04-07T14:34:10Z</td>\n",
       "      <td>country_not_available</td>\n",
       "      <td>state_not_available</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "      <td>https</td>\n",
       "      <td>200.0</td>\n",
       "      <td>cloudflare</td>\n",
       "      <td>1994-02-01T05:00:00Z</td>\n",
       "      <td>2021-12-20T17:37:52Z</td>\n",
       "      <td>2019-04-01T20:15:54Z</td>\n",
       "      <td>US</td>\n",
       "      <td>FL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "      <td>http</td>\n",
       "      <td>200.0</td>\n",
       "      <td>nginx/1.14.1</td>\n",
       "      <td>1998-06-18T04:00:00Z</td>\n",
       "      <td>2024-06-17T04:00:00Z</td>\n",
       "      <td>2019-04-19T06:57:43Z</td>\n",
       "      <td>country_not_available</td>\n",
       "      <td>state_not_available</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "      <td>https</td>\n",
       "      <td>200.0</td>\n",
       "      <td>nginx</td>\n",
       "      <td>1995-03-23T05:00:00Z</td>\n",
       "      <td>2029-03-24T04:00:00Z</td>\n",
       "      <td>2019-04-12T19:51:39Z</td>\n",
       "      <td>REDACTED FOR PRIVACY</td>\n",
       "      <td>REDACTED FOR PRIVACY</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   having_ip_address redirect protocol  http_status_code   server_name  \\\n",
       "0                  0    False    https             200.0         tsa_k   \n",
       "1                  0     True     http             200.0        Apache   \n",
       "2                  0     True    https             200.0    cloudflare   \n",
       "3                  0    False     http             200.0  nginx/1.14.1   \n",
       "4                  0    False    https             200.0         nginx   \n",
       "\n",
       "    whois_creation_date     whois_expiry_date    whois_last_updated  \\\n",
       "0  2000-01-21T16:28:17Z  2020-01-21T16:28:17Z  2018-12-07T19:32:35Z   \n",
       "1  2006-04-11T19:49:29Z  2025-04-11T19:49:29Z  2016-04-07T14:34:10Z   \n",
       "2  1994-02-01T05:00:00Z  2021-12-20T17:37:52Z  2019-04-01T20:15:54Z   \n",
       "3  1998-06-18T04:00:00Z  2024-06-17T04:00:00Z  2019-04-19T06:57:43Z   \n",
       "4  1995-03-23T05:00:00Z  2029-03-24T04:00:00Z  2019-04-12T19:51:39Z   \n",
       "\n",
       "           whois_country            whois_city  \n",
       "0                     US                    CA  \n",
       "1  country_not_available   state_not_available  \n",
       "2                     US                    FL  \n",
       "3  country_not_available   state_not_available  \n",
       "4   REDACTED FOR PRIVACY  REDACTED FOR PRIVACY  "
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "list_categorical_features = [\"having_ip_address\", \"redirect\", \"protocol\", \"http_status_code\",\n",
    "                             \"server_name\",\"whois_creation_date\",\"whois_expiry_date\",\"whois_last_updated\",\\\n",
    "                             \"whois_country\",\"whois_city\"]\n",
    "# list_categorical_features = [\"having_ip_address\", \"redirect\", \"protocol\", \"http_status_code\", \"whois_country\",\"whois_city\"]\n",
    "data_cat = data[list_categorical_features]\n",
    "print(data_cat.shape)\n",
    "data_cat.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "nan"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_cat[\"server_name\"][8]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\USER\\Anaconda3\\lib\\site-packages\\pandas\\core\\indexing.py:543: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self.obj[item] = s\n"
     ]
    }
   ],
   "source": [
    "# for i  in range(len(data_cat)):\n",
    "#     try:\n",
    "#         val1,val2,val3 = data_cat.iloc[i , [5,6,7]]\n",
    "#         if val1 == \"cdate_not_available\":\n",
    "#             data_cat.iloc[i, 5] = np.nan\n",
    "#         if val2 == \"rdate_not_available\":\n",
    "#             data_cat.iloc[i, 6] = np.nan\n",
    "#         if val3 == \"udate_not_available\":\n",
    "#             data_cat.iloc[i, 7] = np.nan\n",
    "        \n",
    "#         else:\n",
    "#             print(val1, val2, val3)\n",
    "#             if type(val1) != int and type(val1) != float:\n",
    "#                 try:\n",
    "#                     data_cat.iloc[i, 5] = int(val1.split(\"-\")[0])\n",
    "#                 except:\n",
    "#                     val1 = val1.split(\".\")\n",
    "# #                     print(val1)\n",
    "#                     try:\n",
    "#                         val1 = val1[2].split(\" \")\n",
    "                    \n",
    "                        \n",
    "#                         data_cat.iloc[i, 5] = int(val1[0])\n",
    "#                     except:\n",
    "# #                         data_cat.drop([i], axis=0)\n",
    "#                         data_cat.iloc[i, 5] = val1\n",
    "#             if type(val2) != int and type(val2) != float:\n",
    "#                 data_cat.iloc[i, 6] = int(val2.split(\"-\")[0])\n",
    "#             if type(val3) != int and type(val3) != float:\n",
    "#                 data_cat.iloc[i, 7] = int(val3.split(\"-\")[0])\n",
    "            \n",
    "#     except:\n",
    "#         print(data_cat.loc[i , 5])\n",
    "\n",
    "for i  in range(len(data_cat)):\n",
    "\n",
    "    try:\n",
    "        val1,val2,val3 = data_cat.iloc[i , [5,6,7]]\n",
    "#         print(val1, val2, val3)\n",
    "    #             if type(val1) != int and type(val1) != float:\n",
    "\n",
    "        data_cat.iloc[i, 5] = (val1.split(\"-\")[0])\n",
    "\n",
    "    #             if type(val2) != int and type(val2) != float:\n",
    "        data_cat.iloc[i, 6] = (val2.split(\"-\")[0])\n",
    "    #             if type(val3) != int and type(val3) != float:\n",
    "        data_cat.iloc[i, 7] = (val3.split(\"-\")[0])\n",
    "            \n",
    "    except:\n",
    "        print(data_cat.loc[i , 5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(170,)"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_cat[\"server_name\"].value_counts().shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1757, 376)"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoded_cat_data = pd.get_dummies(data_cat)\n",
    "encoded_cat_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "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>having_ip_address</th>\n",
       "      <th>http_status_code</th>\n",
       "      <th>redirect_False</th>\n",
       "      <th>redirect_True</th>\n",
       "      <th>protocol_http</th>\n",
       "      <th>protocol_https</th>\n",
       "      <th>server_name_516</th>\n",
       "      <th>server_name_ATS</th>\n",
       "      <th>server_name_ATS/7.1.2</th>\n",
       "      <th>server_name_AmazonS3</th>\n",
       "      <th>...</th>\n",
       "      <th>whois_city_Victoria</th>\n",
       "      <th>whois_city_WA</th>\n",
       "      <th>whois_city_WI</th>\n",
       "      <th>whois_city_Wilayah Persekutuan</th>\n",
       "      <th>whois_city_ca</th>\n",
       "      <th>whois_city_ilocos norte r3</th>\n",
       "      <th>whois_city_jilin</th>\n",
       "      <th>whois_city_ma</th>\n",
       "      <th>whois_city_nj</th>\n",
       "      <th>whois_city_state_not_available</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>200.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>200.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>200.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>200.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>200.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 376 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   having_ip_address  http_status_code  redirect_False  redirect_True  \\\n",
       "0                  0             200.0               1              0   \n",
       "1                  0             200.0               0              1   \n",
       "2                  0             200.0               0              1   \n",
       "3                  0             200.0               1              0   \n",
       "4                  0             200.0               1              0   \n",
       "\n",
       "   protocol_http  protocol_https  server_name_516  server_name_ATS  \\\n",
       "0              0               1                0                0   \n",
       "1              1               0                0                0   \n",
       "2              0               1                0                0   \n",
       "3              1               0                0                0   \n",
       "4              0               1                0                0   \n",
       "\n",
       "   server_name_ATS/7.1.2  server_name_AmazonS3  \\\n",
       "0                      0                     0   \n",
       "1                      0                     0   \n",
       "2                      0                     0   \n",
       "3                      0                     0   \n",
       "4                      0                     0   \n",
       "\n",
       "                ...                whois_city_Victoria  whois_city_WA  \\\n",
       "0               ...                                  0              0   \n",
       "1               ...                                  0              0   \n",
       "2               ...                                  0              0   \n",
       "3               ...                                  0              0   \n",
       "4               ...                                  0              0   \n",
       "\n",
       "   whois_city_WI  whois_city_Wilayah Persekutuan  whois_city_ca  \\\n",
       "0              0                               0              0   \n",
       "1              0                               0              0   \n",
       "2              0                               0              0   \n",
       "3              0                               0              0   \n",
       "4              0                               0              0   \n",
       "\n",
       "   whois_city_ilocos norte r3  whois_city_jilin  whois_city_ma  whois_city_nj  \\\n",
       "0                           0                 0              0              0   \n",
       "1                           0                 0              0              0   \n",
       "2                           0                 0              0              0   \n",
       "3                           0                 0              0              0   \n",
       "4                           0                 0              0              0   \n",
       "\n",
       "   whois_city_state_not_available  \n",
       "0                               0  \n",
       "1                               1  \n",
       "2                               0  \n",
       "3                               1  \n",
       "4                               0  \n",
       "\n",
       "[5 rows x 376 columns]"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoded_cat_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "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>url_length</th>\n",
       "      <th>special_character_count</th>\n",
       "      <th>dns_response_time</th>\n",
       "      <th>handshake_time</th>\n",
       "      <th>data_time</th>\n",
       "      <th>data_length</th>\n",
       "      <th>scripts_count</th>\n",
       "      <th>iframe_count</th>\n",
       "      <th>external_link_count</th>\n",
       "      <th>total_strings_count</th>\n",
       "      <th>average_string_length</th>\n",
       "      <th>max_len_strings</th>\n",
       "      <th>strings_above_avglength</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>27</td>\n",
       "      <td>2</td>\n",
       "      <td>1.866579</td>\n",
       "      <td>809.915781</td>\n",
       "      <td>185.235262</td>\n",
       "      <td>301912.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>73.0</td>\n",
       "      <td>26397.0</td>\n",
       "      <td>63.022881</td>\n",
       "      <td>46438.0</td>\n",
       "      <td>2436.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>12</td>\n",
       "      <td>2</td>\n",
       "      <td>654.555559</td>\n",
       "      <td>3379.443407</td>\n",
       "      <td>799.807549</td>\n",
       "      <td>162844.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>9289.0</td>\n",
       "      <td>11.865217</td>\n",
       "      <td>5491.0</td>\n",
       "      <td>2936.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>92</td>\n",
       "      <td>12</td>\n",
       "      <td>57.567120</td>\n",
       "      <td>2735.617161</td>\n",
       "      <td>29.102325</td>\n",
       "      <td>129957.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>22093.0</td>\n",
       "      <td>44.528946</td>\n",
       "      <td>28934.0</td>\n",
       "      <td>4508.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "      <td>1.390219</td>\n",
       "      <td>1009.941101</td>\n",
       "      <td>0.138521</td>\n",
       "      <td>3453.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>29</td>\n",
       "      <td>2</td>\n",
       "      <td>1.077890</td>\n",
       "      <td>6096.054077</td>\n",
       "      <td>350.750208</td>\n",
       "      <td>204520.0</td>\n",
       "      <td>85.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>41.0</td>\n",
       "      <td>8248.0</td>\n",
       "      <td>12.376940</td>\n",
       "      <td>5491.0</td>\n",
       "      <td>2477.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   url_length  special_character_count  dns_response_time  handshake_time  \\\n",
       "0          27                        2           1.866579      809.915781   \n",
       "1          12                        2         654.555559     3379.443407   \n",
       "2          92                       12          57.567120     2735.617161   \n",
       "3          24                        3           1.390219     1009.941101   \n",
       "4          29                        2           1.077890     6096.054077   \n",
       "\n",
       "    data_time  data_length  scripts_count  iframe_count  external_link_count  \\\n",
       "0  185.235262     301912.0            8.0           6.0                 73.0   \n",
       "1  799.807549     162844.0           33.0           0.0                 18.0   \n",
       "2   29.102325     129957.0           55.0           2.0                 58.0   \n",
       "3    0.138521       3453.0            1.0           0.0                  1.0   \n",
       "4  350.750208     204520.0           85.0           4.0                 41.0   \n",
       "\n",
       "   total_strings_count  average_string_length  max_len_strings  \\\n",
       "0              26397.0              63.022881          46438.0   \n",
       "1               9289.0              11.865217           5491.0   \n",
       "2              22093.0              44.528946          28934.0   \n",
       "3                  8.0               8.000000              8.0   \n",
       "4               8248.0              12.376940           5491.0   \n",
       "\n",
       "   strings_above_avglength  \n",
       "0                   2436.0  \n",
       "1                   2936.0  \n",
       "2                   4508.0  \n",
       "3                      0.0  \n",
       "4                   2477.0  "
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#joining the encoded categorical data with numerical data\n",
    "data_numeric = data.drop(list_categorical_features, axis = 1)\n",
    "data_numeric.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1757, 389)\n"
     ]
    },
    {
     "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>url_length</th>\n",
       "      <th>special_character_count</th>\n",
       "      <th>dns_response_time</th>\n",
       "      <th>handshake_time</th>\n",
       "      <th>data_time</th>\n",
       "      <th>data_length</th>\n",
       "      <th>scripts_count</th>\n",
       "      <th>iframe_count</th>\n",
       "      <th>external_link_count</th>\n",
       "      <th>total_strings_count</th>\n",
       "      <th>...</th>\n",
       "      <th>whois_city_Victoria</th>\n",
       "      <th>whois_city_WA</th>\n",
       "      <th>whois_city_WI</th>\n",
       "      <th>whois_city_Wilayah Persekutuan</th>\n",
       "      <th>whois_city_ca</th>\n",
       "      <th>whois_city_ilocos norte r3</th>\n",
       "      <th>whois_city_jilin</th>\n",
       "      <th>whois_city_ma</th>\n",
       "      <th>whois_city_nj</th>\n",
       "      <th>whois_city_state_not_available</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>27</td>\n",
       "      <td>2</td>\n",
       "      <td>1.866579</td>\n",
       "      <td>809.915781</td>\n",
       "      <td>185.235262</td>\n",
       "      <td>301912.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>73.0</td>\n",
       "      <td>26397.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>12</td>\n",
       "      <td>2</td>\n",
       "      <td>654.555559</td>\n",
       "      <td>3379.443407</td>\n",
       "      <td>799.807549</td>\n",
       "      <td>162844.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>9289.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>92</td>\n",
       "      <td>12</td>\n",
       "      <td>57.567120</td>\n",
       "      <td>2735.617161</td>\n",
       "      <td>29.102325</td>\n",
       "      <td>129957.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>22093.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "      <td>1.390219</td>\n",
       "      <td>1009.941101</td>\n",
       "      <td>0.138521</td>\n",
       "      <td>3453.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>29</td>\n",
       "      <td>2</td>\n",
       "      <td>1.077890</td>\n",
       "      <td>6096.054077</td>\n",
       "      <td>350.750208</td>\n",
       "      <td>204520.0</td>\n",
       "      <td>85.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>41.0</td>\n",
       "      <td>8248.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 389 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   url_length  special_character_count  dns_response_time  handshake_time  \\\n",
       "0          27                        2           1.866579      809.915781   \n",
       "1          12                        2         654.555559     3379.443407   \n",
       "2          92                       12          57.567120     2735.617161   \n",
       "3          24                        3           1.390219     1009.941101   \n",
       "4          29                        2           1.077890     6096.054077   \n",
       "\n",
       "    data_time  data_length  scripts_count  iframe_count  external_link_count  \\\n",
       "0  185.235262     301912.0            8.0           6.0                 73.0   \n",
       "1  799.807549     162844.0           33.0           0.0                 18.0   \n",
       "2   29.102325     129957.0           55.0           2.0                 58.0   \n",
       "3    0.138521       3453.0            1.0           0.0                  1.0   \n",
       "4  350.750208     204520.0           85.0           4.0                 41.0   \n",
       "\n",
       "   total_strings_count               ...                whois_city_Victoria  \\\n",
       "0              26397.0               ...                                  0   \n",
       "1               9289.0               ...                                  0   \n",
       "2              22093.0               ...                                  0   \n",
       "3                  8.0               ...                                  0   \n",
       "4               8248.0               ...                                  0   \n",
       "\n",
       "   whois_city_WA  whois_city_WI  whois_city_Wilayah Persekutuan  \\\n",
       "0              0              0                               0   \n",
       "1              0              0                               0   \n",
       "2              0              0                               0   \n",
       "3              0              0                               0   \n",
       "4              0              0                               0   \n",
       "\n",
       "   whois_city_ca  whois_city_ilocos norte r3  whois_city_jilin  whois_city_ma  \\\n",
       "0              0                           0                 0              0   \n",
       "1              0                           0                 0              0   \n",
       "2              0                           0                 0              0   \n",
       "3              0                           0                 0              0   \n",
       "4              0                           0                 0              0   \n",
       "\n",
       "   whois_city_nj  whois_city_state_not_available  \n",
       "0              0                               0  \n",
       "1              0                               1  \n",
       "2              0                               0  \n",
       "3              0                               1  \n",
       "4              0                               0  \n",
       "\n",
       "[5 rows x 389 columns]"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_final = pd.concat([data_numeric, encoded_cat_data], axis=1, sort = False)\n",
    "print(data_final.shape)\n",
    "data_final.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "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>url_length</th>\n",
       "      <th>special_character_count</th>\n",
       "      <th>dns_response_time</th>\n",
       "      <th>handshake_time</th>\n",
       "      <th>data_time</th>\n",
       "      <th>data_length</th>\n",
       "      <th>scripts_count</th>\n",
       "      <th>iframe_count</th>\n",
       "      <th>external_link_count</th>\n",
       "      <th>total_strings_count</th>\n",
       "      <th>...</th>\n",
       "      <th>whois_city_Victoria</th>\n",
       "      <th>whois_city_WA</th>\n",
       "      <th>whois_city_WI</th>\n",
       "      <th>whois_city_Wilayah Persekutuan</th>\n",
       "      <th>whois_city_ca</th>\n",
       "      <th>whois_city_ilocos norte r3</th>\n",
       "      <th>whois_city_jilin</th>\n",
       "      <th>whois_city_ma</th>\n",
       "      <th>whois_city_nj</th>\n",
       "      <th>whois_city_state_not_available</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1.757000e+03</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "      <td>1757.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>47.905521</td>\n",
       "      <td>6.149687</td>\n",
       "      <td>123.738056</td>\n",
       "      <td>1574.808989</td>\n",
       "      <td>314.575958</td>\n",
       "      <td>8.608122e+04</td>\n",
       "      <td>14.711440</td>\n",
       "      <td>0.372795</td>\n",
       "      <td>31.854297</td>\n",
       "      <td>11340.444508</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000569</td>\n",
       "      <td>0.024474</td>\n",
       "      <td>0.000569</td>\n",
       "      <td>0.001138</td>\n",
       "      <td>0.001138</td>\n",
       "      <td>0.000569</td>\n",
       "      <td>0.000569</td>\n",
       "      <td>0.000569</td>\n",
       "      <td>0.001707</td>\n",
       "      <td>0.574274</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>31.972585</td>\n",
       "      <td>4.105363</td>\n",
       "      <td>193.332407</td>\n",
       "      <td>4270.575643</td>\n",
       "      <td>640.166030</td>\n",
       "      <td>2.020096e+05</td>\n",
       "      <td>22.005367</td>\n",
       "      <td>0.865308</td>\n",
       "      <td>63.097092</td>\n",
       "      <td>27603.311401</td>\n",
       "      <td>...</td>\n",
       "      <td>0.023857</td>\n",
       "      <td>0.154558</td>\n",
       "      <td>0.023857</td>\n",
       "      <td>0.033729</td>\n",
       "      <td>0.033729</td>\n",
       "      <td>0.023857</td>\n",
       "      <td>0.023857</td>\n",
       "      <td>0.023857</td>\n",
       "      <td>0.041298</td>\n",
       "      <td>0.494593</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>9.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.024557</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>30.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1.266956</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>43.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>22.622108</td>\n",
       "      <td>778.821230</td>\n",
       "      <td>3.920317</td>\n",
       "      <td>1.137700e+04</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>3910.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>59.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>228.809118</td>\n",
       "      <td>2051.874638</td>\n",
       "      <td>446.742773</td>\n",
       "      <td>9.517400e+04</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>41.000000</td>\n",
       "      <td>12257.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>641.000000</td>\n",
       "      <td>68.000000</td>\n",
       "      <td>1375.710964</td>\n",
       "      <td>130470.530500</td>\n",
       "      <td>7074.467182</td>\n",
       "      <td>3.869084e+06</td>\n",
       "      <td>443.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>1018.000000</td>\n",
       "      <td>305185.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 389 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        url_length  special_character_count  dns_response_time  \\\n",
       "count  1757.000000              1757.000000        1757.000000   \n",
       "mean     47.905521                 6.149687         123.738056   \n",
       "std      31.972585                 4.105363         193.332407   \n",
       "min       9.000000                 2.000000           0.024557   \n",
       "25%      30.000000                 4.000000           1.266956   \n",
       "50%      43.000000                 5.000000          22.622108   \n",
       "75%      59.000000                 8.000000         228.809118   \n",
       "max     641.000000                68.000000        1375.710964   \n",
       "\n",
       "       handshake_time    data_time   data_length  scripts_count  iframe_count  \\\n",
       "count     1757.000000  1757.000000  1.757000e+03    1757.000000   1757.000000   \n",
       "mean      1574.808989   314.575958  8.608122e+04      14.711440      0.372795   \n",
       "std       4270.575643   640.166030  2.020096e+05      22.005367      0.865308   \n",
       "min          0.000000     0.000000  0.000000e+00       0.000000      0.000000   \n",
       "25%          0.000000     0.000000  0.000000e+00       1.000000      0.000000   \n",
       "50%        778.821230     3.920317  1.137700e+04       8.000000      0.000000   \n",
       "75%       2051.874638   446.742773  9.517400e+04      20.000000      0.000000   \n",
       "max     130470.530500  7074.467182  3.869084e+06     443.000000      6.000000   \n",
       "\n",
       "       external_link_count  total_strings_count  \\\n",
       "count          1757.000000          1757.000000   \n",
       "mean             31.854297         11340.444508   \n",
       "std              63.097092         27603.311401   \n",
       "min               0.000000             0.000000   \n",
       "25%               1.000000            20.000000   \n",
       "50%              17.000000          3910.000000   \n",
       "75%              41.000000         12257.000000   \n",
       "max            1018.000000        305185.000000   \n",
       "\n",
       "                    ...                whois_city_Victoria  whois_city_WA  \\\n",
       "count               ...                        1757.000000    1757.000000   \n",
       "mean                ...                           0.000569       0.024474   \n",
       "std                 ...                           0.023857       0.154558   \n",
       "min                 ...                           0.000000       0.000000   \n",
       "25%                 ...                           0.000000       0.000000   \n",
       "50%                 ...                           0.000000       0.000000   \n",
       "75%                 ...                           0.000000       0.000000   \n",
       "max                 ...                           1.000000       1.000000   \n",
       "\n",
       "       whois_city_WI  whois_city_Wilayah Persekutuan  whois_city_ca  \\\n",
       "count    1757.000000                     1757.000000    1757.000000   \n",
       "mean        0.000569                        0.001138       0.001138   \n",
       "std         0.023857                        0.033729       0.033729   \n",
       "min         0.000000                        0.000000       0.000000   \n",
       "25%         0.000000                        0.000000       0.000000   \n",
       "50%         0.000000                        0.000000       0.000000   \n",
       "75%         0.000000                        0.000000       0.000000   \n",
       "max         1.000000                        1.000000       1.000000   \n",
       "\n",
       "       whois_city_ilocos norte r3  whois_city_jilin  whois_city_ma  \\\n",
       "count                 1757.000000       1757.000000    1757.000000   \n",
       "mean                     0.000569          0.000569       0.000569   \n",
       "std                      0.023857          0.023857       0.023857   \n",
       "min                      0.000000          0.000000       0.000000   \n",
       "25%                      0.000000          0.000000       0.000000   \n",
       "50%                      0.000000          0.000000       0.000000   \n",
       "75%                      0.000000          0.000000       0.000000   \n",
       "max                      1.000000          1.000000       1.000000   \n",
       "\n",
       "       whois_city_nj  whois_city_state_not_available  \n",
       "count    1757.000000                     1757.000000  \n",
       "mean        0.001707                        0.574274  \n",
       "std         0.041298                        0.494593  \n",
       "min         0.000000                        0.000000  \n",
       "25%         0.000000                        0.000000  \n",
       "50%         0.000000                        1.000000  \n",
       "75%         0.000000                        1.000000  \n",
       "max         1.000000                        1.000000  \n",
       "\n",
       "[8 rows x 389 columns]"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_final.describe(include='all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 1] [ 208 1549]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.preprocessing import LabelEncoder\n",
    "label = LabelEncoder().fit_transform(label)\n",
    "value, counts = np.unique(label, return_counts=True)\n",
    "print(value, counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1229, 389)\n",
      "(528, 389)\n",
      "(1229,)\n",
      "(528,)\n"
     ]
    }
   ],
   "source": [
    "# Get a training and test dataset\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(data_final, label, test_size=0.3)\n",
    "print(X_train.shape)\n",
    "print(X_test.shape)\n",
    "print(Y_train.shape)\n",
    "print(Y_test.shape)\n",
    "prediction = dict()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight={0: 7, 1: 1},\n",
       "            criterion='gini', max_depth=100, max_features='auto',\n",
       "            max_leaf_nodes=None, min_impurity_decrease=0.0,\n",
       "            min_impurity_split=None, min_samples_leaf=1,\n",
       "            min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "            n_estimators=100, n_jobs=-1, oob_score=False,\n",
       "            random_state=None, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "rf = RandomForestClassifier(n_estimators=100, n_jobs=-1, max_depth=100, criterion = 'gini', class_weight={0:7,1:1})\n",
    "rf.fit(X_train, Y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {},
   "outputs": [],
   "source": [
    "def eval_model(Y_test, Pred):\n",
    "    print(classification_report(Y_test, Pred))\n",
    "    print(confusion_matrix(Y_test, Pred))\n",
    "    print(\"accuracy_score : \", accuracy_score(Y_test, Pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.67      0.26      0.38        68\n",
      "           1       0.90      0.98      0.94       460\n",
      "\n",
      "   micro avg       0.89      0.89      0.89       528\n",
      "   macro avg       0.78      0.62      0.66       528\n",
      "weighted avg       0.87      0.89      0.87       528\n",
      "\n",
      "[[ 18  50]\n",
      " [  9 451]]\n",
      "accuracy_score :  0.8882575757575758\n"
     ]
    }
   ],
   "source": [
    "pred = rf.predict(X_test)\n",
    "eval_model(Y_test, pred)\n",
    "prediction[\"Random Forest\"] = pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.48      0.29      0.36        68\n",
      "           1       0.90      0.95      0.93       460\n",
      "\n",
      "   micro avg       0.87      0.87      0.87       528\n",
      "   macro avg       0.69      0.62      0.64       528\n",
      "weighted avg       0.85      0.87      0.85       528\n",
      "\n",
      "[[ 20  48]\n",
      " [ 22 438]]\n",
      "accuracy_score :  0.8674242424242424\n"
     ]
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "clf = KNeighborsClassifier(n_neighbors = 3)\n",
    "clf.fit(X_train, Y_train)\n",
    "pred = clf.predict(X_test)\n",
    "prediction[\"KNN\"] = pred\n",
    "eval_model(Y_test, pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [],
   "source": [
    "def formatt(x):\n",
    "    if x == 'negative':\n",
    "        return 0\n",
    "    return 1\n",
    "def AuROC(prediction, y_test=None):\n",
    "\n",
    "    \n",
    "    vfunc = np.vectorize(formatt)\n",
    "\n",
    "    cmp = 0\n",
    "    colors = ['b', 'g', 'y', 'm', 'k','c']\n",
    "    for model, predicted in prediction.items():\n",
    "        if model == \"Logistic\" or model == \"SVM\" or model == \"SVM_Linear\":\n",
    "            false_positive_rate, true_positive_rate, thresholds = roc_curve(y_test, predicted)\n",
    "        else:\n",
    "            false_positive_rate, true_positive_rate, thresholds = roc_curve(Y_test, predicted)\n",
    "        roc_auc = auc(false_positive_rate, true_positive_rate)\n",
    "        plt.plot(false_positive_rate, true_positive_rate, colors[cmp], label='%s: AUC %0.2f'% (model,roc_auc))\n",
    "        cmp += 1\n",
    "\n",
    "    plt.title('Classifiers comparaison with ROC')\n",
    "    plt.legend(loc='lower right')\n",
    "    plt.plot([0,1],[0,1],'r--')\n",
    "    plt.xlim([-0.1,1.2])\n",
    "    plt.ylim([-0.1,1.2])\n",
    "    plt.ylabel('True Positive Rate')\n",
    "    plt.xlabel('False Positive Rate')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [],
   "source": [
    "features = list(data_final)\n",
    "feature_importances= dict()\n",
    "imps = []\n",
    "from sklearn.feature_selection import SelectFromModel\n",
    "for feature in zip(features, rf.feature_importances_):\n",
    "    feature_importances[feature[0]] = feature[1]\n",
    "for key in feature_importances.keys():\n",
    "#     print(key)\n",
    "    imps.append(feature_importances[key])\n",
    "imps = np.array(imps)\n",
    "indexes = np.argsort(imps, axis=-1, kind='quicksort', order=None)\n",
    "sorted_imps = sorted(imps)\n",
    "#pick the top 10 features\n",
    "VALUES = sorted_imps[-30:]\n",
    "feature_indexes = indexes[-30:]\n",
    "imp_features=[]\n",
    "for i in feature_indexes:\n",
    "    imp_features.append(features[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cd93951fd0>"
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.Series(index = imp_features, data = VALUES).plot(kind='barh', title='Variable Importances', figsize = (10,10))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\USER\\Anaconda3\\lib\\site-packages\\sklearn\\preprocessing\\data.py:625: DataConversionWarning: Data with input dtype uint8, int64, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.partial_fit(X, y)\n",
      "C:\\Users\\USER\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:4: DataConversionWarning: Data with input dtype uint8, int64, float64 were all converted to float64 by StandardScaler.\n",
      "  after removing the cwd from sys.path.\n",
      "C:\\Users\\USER\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:5: DataConversionWarning: Data with input dtype uint8, int64, float64 were all converted to float64 by StandardScaler.\n",
      "  \"\"\"\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(1229, 389)"
      ]
     },
     "execution_count": 179,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "SS = StandardScaler()\n",
    "SS.fit(X_train)\n",
    "scaled_X_train = SS.transform(X_train)\n",
    "scaled_X_test = SS.transform(X_test)\n",
    "scaled_X_train.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "clf = LogisticRegression(solver=\"liblinear\", class_weight=\"balanced\").fit(scaled_X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.43      0.78      0.55        68\n",
      "           1       0.96      0.85      0.90       460\n",
      "\n",
      "   micro avg       0.84      0.84      0.84       528\n",
      "   macro avg       0.70      0.81      0.73       528\n",
      "weighted avg       0.89      0.84      0.86       528\n",
      "\n",
      "[[ 53  15]\n",
      " [ 71 389]]\n",
      "accuracy_score :  0.8371212121212122\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAEWCAYAAAB42tAoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzsnXd4VEX3xz8nvRNCrxKpUiRSRVFUXgsoIjZERFEQO4piLy/601dFEBvSBaUEpKOgCEox9I6hl1BCCqmbnm3z++NuMIQkJJBlU+bzPPtk771z55692T3fO3NmzohSCo1Go9FoisLN1QZoNBqNpnyjhUKj0Wg0xaKFQqPRaDTFooVCo9FoNMWihUKj0Wg0xaKFQqPRaDTFooWiEiMio0RklhPr3ycitzjei4hMF5EUEdkqIjeJyCFnXbsyIyKNRSRDRNxdbUtBROQ3EXmimOMzROTjK2mTxvlooajgiMijIrLd4VhiHT/k7lfi2kqpNkqptY7N7sDtQEOlVBel1N9KqZZXwo7KhlLqlFIqQCllc7UtBVFK9VJK/QggIoNFJOJS6xKRJiKiHN/dDBE5ISJvFVJusIj8IyJZIhInIhNEJLhAmRYiMl9EEkXEJCJ7ReTV8ii2FREtFBUYEXkV+Ar4H1AHaAx8D/R1gTlXASeUUpmXW5GIeJSBPeUW7bwuIFgpFQA8CLwvIrfnHRCR14DPgdeBasD1GN+1VSLi5SjTFNgCnAbaKaWqAQ8BnYDAK/lBKi1KKf2qgC+MH00G8FAxZUYBs/JtzwfiABOwHmiT71hvYD+QDpwBRjr21wR+BVKBZOBvwM1x7ATwH2AIkAPYHDZ9CNwCROervz6wEEgAooDhBexcAMwC0oChQBdgu2M7HviymM/ZF9jtKHsMuCvfNZc57D4KPF3gmvMd10wH/gFaAG8DZzGczh35yq8FPgW2Ou7fUiCkhPd2BjABWAFkOu7Z3cAuh82ngVH5yjcBFODh2B4MHHfYGQUMdOx3A94DTjps/gmoVqCOJ4BTQCLwbhH3L9Tx/837v04FzuY7Pgt4Jd99GApcU+B/nprvs44Hljvs3QI0LeK6531Ox76twOuO90GOuh8ucF6A4/M+lc++5a7+TVbml8sN0K9L/MfBXYA1/4+skDKjOF8onsJ4wvLGaInszncsFrjJ8b460MHx/lNgIuDpeN0EiOPYCeA/jveDgYh89d2CQygcDm0H8AHgBVztcHx35rPTAtznKOsLbAIGOY4HANcX8Rm7YDjn2x3nNgBaOY6tw2hh+QBhGCLVM981c4A7AQ8MJxsFvOv4nE8DUfmusxZDQNsC/hiiV9J7O8Nh440OG30c96edY/taDDG8z1G+CQ4H6rhWGtDScaweDhFyXPOo434GAIuAmQXqmOK4n+2BXOCaIu7jKaCj4/0hx//nmnzHrst3H4YW9j/P91mTHf8XD2A2MLeIa577nI7t64EsoN/FvuPAj0C4430c8KSrf5OV+aW7niouNYBEpZS1pCcopX5QSqUrpXIxHGV7EanmOGwBWotIkFIqRSm1M9/+esBVSimLMmIPpU0Q1hmopZT6SCllVkodx3Bgj+Qrs0kptUQpZVdKZTuu20xEaiqlMpRSm4uoewjwg1JqlePcM0qpgyLSCCNu8qZSKkcptRvjSXlQvnP/VkqtdNzD+UAt4DOllAWYCzQp0Bc+UykVqYzutfeBh/O6kS5ybwGWKqU2OGzMUUqtVUr949jeC4QDPYr4jHagrYj4KqVilVL7HPsHYrS0jiulMjBaQ48U6Lr7UCmVrZTaA+zBEIzCWAf0EJG6ju0Fju1QjCf7PUWcVxiLlFJbHfd1NoZIF0eiiGRjPBx8Dyxx7K9J0d/xWMdxMH4LsaWwT1NKtFBUXJKAmiXtzxcRdxH5TESOiUgaRmsA/v2xPYDR/XRSRNaJSDfH/i8wnlr/EJHjhQUbS8BVQH0RSc17Ae9gxFXyOF3gnCEYXUEHRWSbiNxTRN2NMLqbClIfSFZKpefbdxKjxZFHfL732RhOyZZvG4wn9cJsPInR8qhZgntb8FxEpKuIrBGRBBExAc8WKA+AQ5T6O47HishyEWmV7zOeLGCTB+ff17h877MKfJ78rMNo5dyM0XW2FkO4emAIqr2I8wqjpNfMo6ajzEiHDZ6O/YkU/R2v5zgOxm+hXins05QSLRQVl00YXSf3lbD8oxh9+f/BiG80cewXAKXUNqVUX6A2xhPdz4796Uqp15RSVwN9gFdFpGcpbT2N0Y0TnO8VqJTqna/Mea0UpdQRpdQAhz2fAwtExL+IupsWsj8GCBGR/MHMxhjdR5dKowJ1WTCcVbH31kHBVtgcjPhJI2UEXycWKP/viUar53YMZ3gQozUGxme8qoBNVs4XwJKyDqNb8RbH+wiMrrIeju1CTbuE6xRekVI2pdRYjO/0847dmzC6y+7PX9bxPegF/OnYtRrjQUfjJLRQVFCUUiaMPv/xInKfiPiJiKeI9BKR0YWcEojxo0sC/DBGSgEgIl4iMlBEqjm6XdIwgpSIyD0i0kxEJN/+0g7b3AqkicibIuLreAJvKyKdizpBRB4TkVqOJ9lUx+7CrjsNeFJEeoqIm4g0EJFWSqnTwEbgUxHxEZFrMVops0tpe34eE5HWIuIHfAQscLRAiry3xRCI0eLJEZEuGGJzASJSR0TudTjHXIzgbt59CAdGiEioiAQ4rjuvNN2ReSiljmC0oh4D1iul8gYRPEDRQhEPNMwbfVRGfAa8ISI+ju/4h8C3InKX4/vdBKObMBqY6Tjnv8ANIvJFXteZ4zs7q+AwWs2loYWiAqOU+hJ4FWPkSwLG0/WL/NvHm5+fMLomzmCMbirY5z8IOOHoOnkWw2EANMd4YsvA0Yes/p07UVI7bRitkTCMgHEiRrygWjGn3QXsE5EM4GvgEaVUTiF1bwWeBMZhBIzX8e9T9gCMp/sYYDHwX6XUqtLYXoCZGMHaOIyA9HDH/ovd28J4HvhIRNIxBP/nIsq5Aa9hfIZkjCf8vCfuHxw2rce4rznAS6X5QAVYByQppU7l2xaM0VmF8RewD4gTkcQiypSW5UAKxmAClFKjMbopx2A8qOQNg+3piAehlDoGdMP4X+9zdOUtxBg1l47msskbvaLRaIpBRNZijHKa6mpbNJorjW5RaDQajaZYnCYUIvKDiJwVkcgijg90TLPfKyIbRaSoYXsajUajcSFO63oSkZsx+rV/Ukq1LeT4DcABpVSKiPTCmJna1SnGaDQajeaScVpOHaXUescIhaKOb8y3uRlo6CxbNBqNRnPplJfka0OA34o6KCLDgGEA/v7+HVu1alVUUY1Go9EUwo4dOxKVUrUu5VyXC4WI3IohFEWmxlZKTQYmA3Tq1Elt3779Clmn0Wg0lQMROXnxUoXjUqFwTIKaCvRSSiW50haNRqPRFI7LhseKSGOMbJeDlFKHXWWHRqPRaIrHaS0KEQnHyBtTU0SiMabZewIopSZizEatAXxvZIfAqpTq5Cx7NBqNRnNpOHPU04CLHB+KsQCKRqPRaMoxema2RqPRaIpFC4VGo9FoikULhUaj0WiKRQuFRqPRaIpFC4VGo9FoikULhUaj0WiKRQuFRqPRaIpFC4VGo9FoikULhUaj0WiKRQuFRqPRaIpFC4VGo9FoikULhUaj0WiKRQuFRqPRaIpFC4VGo9FoikULhUaj0WiKRQuFRqPRaIpFC4VGo9FoikULhUaj0WiKRQuFRqPRaIpFC4VGo9FoikULhUaj0WiKRQuFRqPRaIpFC4VGo9FoikULhUaj0WiKRQuFRqPRaIrFaUIhIj+IyFkRiSziuIjINyJyVET2ikgHZ9mi0Wg0mkvHmS2KGcBdxRzvBTR3vIYBE5xoi0aj0ZSKxFOJ2Gw2V5tRLvBwVsVKqfUi0qSYIn2Bn5RSCtgsIsEiUk8pFessmzQaTeVFKTCbIT0dMjKMV/73edspmTZiLTmctsWR6HGabP/TWAJSsPmbsfq6YXX3of/8LewObcpVHoFMfHeEqz+ay3GaUJSABsDpfNvRjn0XCIWIDMNoddC4ceMrYpxGo3EuFsvFnXpJj6WbbWR4mrEFWaBWGtRLgDopuNVKx7tOJu7BdlQNL8z1A7C4+xWwpBHQiEDS6HjgEKM/H0Pnk1H82K4DSW++4YpbU+5wpVBIIftUYQWVUpOByQCdOnUqtIxGo3EeVuv5jro4R15Sh282X+SinjYIsUB1M971zHjVzcWtXgrUMWGvnoUt0ILVX2Hz9cDm4V1IBW74KTdCxEJ1UoyXSsbPnkWQWKnu7kEdnyCuCmhIM3tLUvt8S+cTqzEBv/QcwON/zELc9HgfcK1QRGNIeR4NgRgX2aLRVGk2boSvvoLU1MKdfE5Oyevy9YWAAAgMNP4GBEC1atCwIfgG2XGrYUaFmFHVzFgCzeT6mcnwySLFI50U92xMYiHDTciVf91TruMF4G1PJ0ilUINk6rilECIphJB8TggCxUx1dy9qeYdQM6ApNYLaEhzQER+fULy96yPyr/NXSnFo8iFeHP4CI8x/sdonmGbLf6XPbTeWzY2tJLhSKJYBL4rIXKArYNLxCY3myjNvHjz+OAQHw9VXG469bt1/nXx+h19wOzAQvPztZPuYyfQyk+ZhJsFqJt5iId5sJtacS3R2FofMOSRabGSqwjoSwM2agbctGX9LCtVVCi0kmfruKTT0SKGOW9K/LQJS8PDww8OrEf6+TakW0BI/37b4+IQ6Xo1xc/Mq0efO2pPIwTsGMvTsNvZKGt2feYv3vvs/PDxc6RbLJ067IyISDtwC1BSRaOC/gCeAUmoisALoDRwFsoAnnWWLRqO5EKXg88/h7behe3dYsgRq1DCO5drtnDWbiTObiTcbjj/ObCbKsR1nNhNnziUuK5e09MJ7g8WWhcpNQizJ+NpSCLQn05gUarsl08Qrk6s9TTTxPEtdtwS8PCznvJG4BeDjE4q/X1N8fFqcEwFf31B8fJrg7u5/WZ/bnmvn9JCZeM1+gQ5kcl9gLSav3kynLp0uq97KjBiDjioOnTp1Utu3b3e1GRpNhcNstxtO32zmTLaFsdPN/B1ppkVXM+16mEm0/SsIqVZroXV42HNxs5iw5yZizT0L5hTEkkyQPYUakkJjz3Ra+php6ZtJfe9cgt2z8JbM8+pwc/PBx6dJvlaAIQCGEITi4VEdkcJbHpdLyrKTxD/6BK0y13EU+O2efgz9eTa+vr5OuV55QkR2KKUuSQ11G0ujqcCYHU/+eQ4+Pt8Tf3yB/SkFnf/1xuuEshCTkImbORVrbgI5WbGQmwiWFDAngzmF6iTT2l9oHeRDkwBv6vkKIUG5+Esa7ioFwZ6vYnd8fBrh49O8QGvAeHl51TkvTnAlMJ81c3DEQU7PeZS72MdE3wBaLZjHS717X1E7KipaKDSacobFbudsCRx/vNlMchFP/oHu7tTx8qKulxdt/PwI87bz56HFJKQeBnOeACSDJQUPd3eaBdfjmurBNA3wo0Edd2p5WQl0z8RbJYM1HqVyz6vfy6uew/HfcJ4IGAHjhri5lQ/XouyK+K8i2frBCl7P+oE4DvPKXb0YET6H4OBgV5tXYSgf/02NppKT5/wv5vjjinH+Ae7u1PXyoo6nJ9f4+XFrcDB1vLzOCUIdT89z237u7ufOWxO1hgHz76Ohtyftjt1JTc9A+vasTZM6Jtxs8VhyT2GzHT/vWh4Sgo9XE3x8Ol8gBD4+V+HuXv67ajL+SefsA19T+8gnNCSXxGrB/DBtAQ888ICrTatwaKHQaC4Ri91OQmFP/g5ByL8/qRjnn+fgW/n50cPh/PM7/rqFOP+C2O1mcnJOkpMdRWpKFLHZUeTkRHEmeTtZOceZ28VRsP0cANzc/PCwObqEqt92QcDYw6NaWd+uK4Yt08bp1zfhNeFlrmYnW4Cfbr6ZffPmUbduXVebVyHRQqHR5MNa2JN/IY4/rhjn7+/mds65t8zn/Ot4ep7bn/fyL8b550cpGzk5p8jJiSIn5wTZDiHIe+XmniH/fFURT7JUEJHJSWSb63Ho96F4u7Xm3XdDadQoFE/PWk4LGLuSpOVJRA/9hZZxz2DHzJteXjT/5hu+GzasUn7eK4UWCk2lx+p48r+Y44+3WEiyWApND+CXz/m38PPjpuDgCxx/3VI6//wopbBYEsjJibpABLKzo8jNPYVSlnxnCN7eDfDxCSU4+LbzAsaeXo147c//MWnnVFpZBnDws+ncfZc3c+cacx8qI7lncjny0n6OLT7F+IAf6YuZtR068NnPP9O0aVNXm1fh0UKhqZBY7XYS87p9inH88WYzicU4/zwH39zPj+5FOP46np4ElMEkLKvVVEAETpwnDHZ71nnlPT1r4eMTSmBgJ2rXfui8oaTGxLIL01ZkmDN4eEF/VhxZQcuEtzj4/Sc8/5wbX38NlXEembIpznx9EvPbo2lgnkc/f+Fkbhq3ffopc19/HfdLEG3NhVTCr46momJTioQSOP64Ypy/b74n/6Y+PtwYFFSo46/r5VUmzv88+20555x//tZA3nurNeW88u7ugY6WQDNCQm6/YF6Bh0fpHv/jMuK4e87d7I7bTei+iRxe8Axjx8CIEVAZe13Stqdx+vEVNDjwX4I5zFKgfqNWzA9fTVhYmKvNq1RoodA4FZtS/z75X2TET6LFct5o/Dx88z35X+3jQ7egoPP7+vO1BALc3Z3WF223W8nNPV2oCOTknMBsPj8DjYj3uYlkQUFdL5hTUJYTyw4kHKDX7F6czUig9uplxG67mwUL4P77y6T6coU1zUrUO0fxHP8p1zCLZIH+Chq/9horP/4YHx8fV5tY6dBCoSk1ec6/JEM9i3L+PnlP/p6ehDqcf2FDPes62fnnRymF2RxXhBBEkZNzGsi/kE3exLJQQkJ6nZttnCcEXl51r8jEsvUn19N3bl/c7N54zFqHLbkTa9bA9dc7/dJXFKUUCQsSOPryUTJjM6gbtJY5aXbGNGjI17Nm0aNHD1ebWGnRQqEBDOefVLDLp4iWQEIxzj/PwTfx8aGrw/kXNtQz8Ao5//wopbBaU4oJGJ/Ebj8/TaqXV118fEIJCrqB2rXPn2FsTCzzvKKfoSDh/4QzeOlgarpdTcJXK7g6JJQVm43kfpWJ7Khsjj6zh2qrviI59EbebTmVg4dOMfDJJ1n/1VcEBQW52sRKjRaKSozd4fwv5vjjLRbOms2FOn9vkXPO/SofH7rkdfsUcPx1vLwIcoHzL4jNlukQgcJjBTZb2nnlPTyqGwno/NtSs2affMHivAR05XNimVKK0RtG89afb3EVN3Hyf0vo0SWERYsgJMTV1pUddoud02NPY/rvIppbvsCXWMacXkhs9er8vGQJffv2dbWJVQItFBUMe74n/4vl90kwmylsxV9vkXPOvbGPD52Dgooc6lkenH9+jIllp4oMGFssCeeVd3PzO+f4g4NvOi9g7OsbWiEnllntVob/NpwJ2ycQmvkIUeOmM7C/D9OmgXdh6/dUUFIjUjn29A7qHRzHtSzntI8vA3Ogeu/eRE6ZQu3atV1tYpVBC0U5ZlVyMrPj488ThLNFOH8vh/Ov6+VFQ29vOgYEFDnUs5qHR7ly/vlRykZubkyhw0f/nVj2b9tHxANv76vw9Q2lZs37LggYV7aJZZnmTB5Z+Ai/Hv6Vq06/SdQP/+P999z48MPKM7LJkmTh2JvHiJsWR/PA6dSV3xjn6cX/3N0Y/cMPDB48uFL9TysCWijKMW8cP87R7Gxa+fnRwMuLjgEBRQ71LM/OPz/GxLLEYgLGJ4uZWHbLBULg7d0AkaoxVj4uI44+4X3YGbuTejvHc2bF8/wwDZ6sJCu5KKWInxnPqRHbITUJvxeu4fVjqRz63Y5/165s/fFHQkNDXW1mlUQLRTklzWplb0YG7191FaMq2I/Dak0rdPho3j67/fz1CTw9a+LjE0pAQAdq1nygQBK6wieWVTUOJh6k1+xexKWfJWjFEjIP9OH336FnT1dbVjZkHszkyLOH8V63gA4e40mvH0KrnzNINZn45IsvGDFihJ4850K0UJRTNqelYQdurFb++tBtthxyc08WOnLImFiWfF75vIllPj5NqV79PwUCxqGlnlhW1fj75N/0ndsXZfWEGWsJsnRmxUZo08bVll0+thwbp/53ivjPttFCjSOEzRwNqU2f6CjqtW/P6j//pF27dq42s8qjhaKcssFkwg243gXD/oyJZdFFBIxPYDbHnFdexOuc4w8K6lxIwDikQnSLlUfmRc7j8SWPU51Qzo5bQYfQq/nlF6hXz9WWXT7Jq5I58vwR3I/uobPHCHC382FQCP939iwj33yTDz/8EO/KFJ2vwGihKKdEmEy0Dwgg0EkJeqzWNDIz9xUaK8jNPY1S+TOjuuHt3Qhf31BCQu68IE5wpSaWVSWUUozdNJbXV71OA2t3zoxdQp//1CA8HPwvb8lol5Mbl8uxV4+REH4Gn2ZBNFh6F5s/nMSgnTuRBvVZu3Qp3bt3d7WZmnxooSiHWOx2tqSl8aSTHhut1nS2br3mvJbBvxPLrsfHZ0CBgHEjl08sq0rY7DZe/v1lxm8bT4PUhznz3Y+89JwP48ZBRe6mV3ZFzOQYjr95mAaZ87gheAUHJs7krlf7sHfvXp5++mnGjh1LYGCgq03VFEALRTlkT0YGmXY73Z0Unzh7NhyzOYYWLSZTrVp3x4plfk65lqZ0ZJozGbBwAL8c/oW6x0dyZubnfDXOjZdfdrVll0fGngwOP3sY2+addAz4Ej/bAQ43aEPvXrdjDQnhl19+4Z577nG1mZoi0EJRDtmQZswevtFJ8YmYmEn4+7ejXr2hOnZQjojPiKdPeB92xO6gxpZvMa19kcWL4L77XG3ZpWPNsHJi1Amix52kqfdPNHSfg827Gu83bMXH+/Zx//33M3HiRGrVquVqUzXFoIWiHLLBZKKxtzcNnZAFMy1tOxkZO2ne/DstEuWIQ4mH6DW7FzFpcfguXYR7dF/WroUuXS56arklcVkiR148Qu7pXOoNaUCD04kcyejCf/bswRQTw48//sigQYP097ACoCOQ5QylFBEmk9O6nWJjJ+Hm5kvt2gOdUr+m9Gw4tYEbfriBxLQMrNPW0Di7L5s3V1yRyInOIbJfJAf6biM06zs6zKtO8CfVedBTaLlxI826dGHv3r08/vjjWiQqCE4VChG5S0QOichREXmrkOONRWSNiOwSkb0i0tuZ9lQETuTkEGs2O2X+hNWaRnx8OLVrP4KnZ3CZ168pPfP3zafnTz2R7Bqkj9vETaFd2bABKtgcS8B4yImZEsO2Ntuwr1hJt+rDqJs0h+N/TaBdu3asWL2aL7/8ktWrV3PVVVe52lxNKSiRUIiIl4g0K03FYuRVGA/0AloDA0SkdYFi7wE/K6WuAx4Bvi/NNSojESYTgFNaFPHxc7DbM6lf/5kyr1tTOpRSjN04locXPEy1zI4kjd7IoHuasnIlVK/uautKT3ZUNntu38PxYTto7TeWa80jkZq+/K9XL66bNImGDRuyY8cORowYgZub7sioaFz0PyYidwP/AKsc22EisrgEdXcBjiqljiulzMBcoGBOYAXkRWyrATFUcTaYTAS5u9OmjAfLK6WIjZ2Ev397AgMraJ9GJSFv+OvIVSOplfAAZ8esZtQbNfnxR/DycrV1pUPZFdHfRrOt7TbSt6bT/o7fCUn4lVMDB9I6N5f3V67k3XffZfPmzbSpDFPJqyglCWZ/BHQF1gAopXaXsHXRADidbzvaUU9+RgF/iMhLgD/wn8IqEpFhwDCAxo0bl+DSFZcNaWl0CwrCvYz7btPTt5GRsZvmzb/X/cIuJMuSxcBFA1lycAkhh14lZcEXzJjqxhNPuNqy0pN1OItDQw6RFRFF7RuhyZze4NOSca+lMXL2bJo2bUpERATdunVztamay6QkQmFRSqUWcC6FrWtfkMK8UcHzBgAzlFJjRaQbMFNE2iqlzltDRyk1GZgM0KlTp5Jcu0KSYrEQmZlJfycMFYyJmYSbmx916uggtqtIyEygT3gftp7ZSmDE19i2DeePlXDrra62rHTYrXaix0Vz4v0o6rr/QXv/75Hsq9md2IRBjz/Ovn37ePbZZxkzZgz+FX0auQYomVAcEJGHATcRCQVeBjaX4LxooFG+7YZc2LU0BLgLQCm1SUR8gJrA2RLUX+nYlDd/oozjE1aribNn51K79gA8PPSSka7gcNJhes/uzenUM3guXkhIaj9WbITWBaN25ZyMyAwOPXUIy7YjXFfrWwITNqJuuIHJXbrw0vXXU7NmTVasWEGvXr1cbaqmDClJVOlFoCPGajGLgBwMsbgY24DmIhIqIl4YweplBcqcAnoCiMg1gA+QQBUlwmTCQ4QuZTzRLj5+FnZ7lg5iu4iNpzdyw7QbiEs1YZ6yhvZe/di8uWKJhN1i58T/nWBHhx24H9lNF+8hBGTvJeG//+UmpXj2q6+47777+Oeff7RIVEJK0qK4Uyn1JvBm3g4RuR9DNIpEKWUVkReBlYA78INSap+IfARsV0otA14DpojICIxuqcFKqUrbtXQxNphMXBcQgH8ZJvRRShETM4mAgOsIDOxUZvVqSsbC/Qt5bPFjeOc2JHP8b/S9qRlz5oBfBcqYkr4rnYNPHiRrTwo1+zeg+bjOyGd7mVOnDs/87394eHgwa9YsHn30UR3/qqwopYp9ATsL2bfjYuc569WxY0dVGcm12ZTPunXqlSNHyrTe1NRNas0a1JkzE8u0Xs3FGbdpnJJRoqqP7KbwS1CvvKKU1epqq0qOLcemjr17TK11W6VOBAxT1lqNlEpKUqdOnVK9e/dWgOrZs6c6deqUq03VlACMB/RL8rtFtihE5E6M+EEDEfky36Eg8i9arCkTdqank+OERIDGTGx/atd+tEzr1RSNzW7jtT9e4+stXxMc24/U6bP5ZowvL73kastKTtqWNA4+eRC3A3voGvIlPskHUXc+wNTJkxnxySfYbDa+/vprXnzxRT0vogpQXNfTWSASIyaxL9/+dOCCWdaay8MZiQAtllTOnp1HnTqP4eGhUzdfCbIt2QxcNJDFBxflDSe5AAAgAElEQVQTuO9lcleMZcl8d+6919WWlQxblo2oD6I48+VJmgb8RAO32YhXLU6NG0f/efPYvHAhd9xxBxMnTtTrV1chihQKpdQuYJeIzFZK5VxBm6okESYTTX18qFuGK3rFx8/Ebs/WQewrREJmAvfOvZct0VvwWTsOvwOv8Ota6FRBQkOp61M5NOQQ2UezqTesAfVPJ2OrNZDRtWsz6o03CAoKYubMmQwcOFDHIqoYJQlmNxCRTzDScJxLZ6qUauE0q6oYSik2mEz0Cgkp0zpjYycTENCRwMCOZVavpnCOJh+l1+xenEyJxm3BfJqqB1i+BSpCSiNrupXjbx8nfvxRmgX9hN+sN6k2sBURa15l6HPPcejQIR577DG+/PJLnQ68ilKSzsUZwHSMCXS9gJ8x0nFoyoij2dkkWCxlOn8iLW0TmZmRujVxBdgcvZlu07pxJikFy9Q/6VH7ASIiKoZIJK9KZlu7bWSPX8L1gU9TN/1nPM+u5ZlnnuGm224jNzeX33//nZkzZ2qRqMKUpEXhp5RaKSJjlFLHgPdE5G9nG1aVcEYiwJiYSbi7B1K79oAyq1NzIYsPLObRRY/imd2A7Am/Mfje5kyaVP5zNllSLRwbeYzEaYdoGTSZWqyABq1YO+gFHv3iC+Lj43n11Vf56KOP9OxqTYmEIleMDsljIvIscAao7VyzqhYbTCaqe3jQqowG11ssKSQk/EzduoPx8Agokzo1F/LNlm945fdXCEjrQvqkZXz0Zm3eew/Ke/d94q+JHH7mMOY4M2FdV1Btxx+kDR/O0Kgo5r/7LmFhYSxbtoxOFSW4onE6JRGKEUAAMBz4BCPL61PONKqqEWEycWO1ariVkYeJj/8Juz2HevWGlUl9mvOxKzsj/xjJuM3jCIi+j+zZs5k52Y/HHnO1ZcVjSbJw5OUjpM4+QFBzC4239Cag+XUs+KIWQ7/9FrPZzGeffcarr76Kp6enq83VlCMuKhRKqS2Ot+nAIAARaehMo6oSiWYzh7KzGVy3bpnUpxwzsQMDOxMYeF2Z1Kn5l2xLNoMWD2LhgYX47n0Jj7XjWP2bOz16uNqy4jm74CxHnj9MjaRf6eozAbfA5hz0u5Zh9zxDREQEt912G5MmTaJZs1ItO6OpIhQrFCLSGSNdeIRSKlFE2mCk8rgNI8mf5jLZWMaJAE2mDWRlHaBly6llUp/mXxKzEuk7ty8bT2/E46+x1Dk9gt82CK1audqyojHHmzn8wmEyFv5Du8CvCLJvxd6pO9+FhTHyuuvw9/dn+vTpPPHEE3rIq6ZIihz1JCKfArOBgcDvIvIuxpoUewA9NLaMiDCZ8BKhc2DZTIiLjZ2Eu3sQtWs/Uib1aQyOJR/jhmk3sPX0Dpj/Mx1zX2XL5vIrEkop4mbFsbX1VnKXbaSL5xACOcCxkSO5NimJ4d99x/3338+BAwcYPHiwFglNsRTXougLtFdKZYtICEaK8PZKqUNXxrSqwQaTiY6BgfiUQSJAiyWJs2fnU6/eENzd9UiVsmJL9Bb6hPchPcOGdfqf9Ot0I7Nmld/EfjnRORx+9jCpy+MI6FaTlpP6Y5m0l4/S0vh07FgaNmzI8uXL6d27yi9RrykhxQlFjlIqG0AplSwiB7VIlC05Nhvb09N5uWHZ9OLFxf2EUrnUr6+D2GXF0oNLGbBwAJJZj5zJv/HqEy0YPRrKMMFvmaGUInZaLMdfPUTD7Nm0rv477kv3sGzjWl5YsoSYmBheeuklPv74YwLLqAWrqRoUJxRXi0heKnEBmuTbRil1v1MtqwJsT0/HrFSZxCf+DWJ3JSCgfRlYp/lu63cM/204vimdyJ72K999VpsXXnC1VYWTfSKbw08fxrp6Mx39x+JrPUJ29z4MG/Ik03/5hXbt2rFw4UK6di24GrFGc3GKE4oHCmx/50xDqiJ5iQBvKINEgCbT32RnH6Jlyx8uu66qjl3ZeWPVG4zdNBbfk/eiFs1hWbg/99zjassuRNkVZ74/w/E3DxNqmUpDmQdBdfjj8ed4eM4ccnJy+OSTT3j99df1kFfNJVNcUsA/r6QhVZEIk4mWvr7UKoNpvMZM7GrUrt2/DCyruuRYc3h88ePM3z8fz90vUG3r1yz/y50OHVxt2YVkHc7i0NBDmP42Uf2OEOqrdEzBDzDwzBlWTJhAjx49mDx5Mi1a6LEnmstDJ5J3EXal2OiYaHe5mM2JJCQsoG7dQbi7l9MIawUgKSuJ//z0H+bvn4/b6i9ocfRbtmwqfyKhbIpTY06x69p11NnyCa0/86LVsmsY3b0zdZctY+P+/UyZMoW//vpLi4SmTCjJzGyNEziYlUWy1Vom+Z3i439EKbOeiX0ZHE85Tu/ZvTmaGAUL5tGz/sPM/xXKeB2pyyZzXyYHnzqI59ZVdPb5Ck9LAlGpzenZ6RkiIyN56KGH+Oabb6hbRhM4NRoohVCIiLdSKteZxlQlNjgSAV5ui8IIYk8mKKgbAQHtysK0Kse2M9u4Z849pKZbsP24miG338SECVCeuvTtFjunPj/FmQ/30NxtPLVZha1JK75sdzOvf/459evXZ+nSpdxbUVZI0lQoLtr1JCJdROQf4Ihju72IfOt0yyo5G0wmanl60tzX97LqSU1dS3b2YZ1O/BJZdmgZt8y4BVOSH+YJG/nkmZuYMqV8iUT6rnR2dtnJifdP0LLZMmrZ13B4wABaZmTw+oIFPP/88+zfv1+LhMZplCRG8Q1wD5AEoJTaA9zqTKOqAnmJAC93Rmxs7GQ8PIKpVevhMrKs6vD9tu/pN68f9vjW2CdvZs43rXjnnfKT/dWea+f4e8f5p9NK3E/tp82iNth/fYc3evakZXg43kFBRERE8N133xFUhkvoajQFKUnXk5tS6mQBh2Zzkj1VgrjcXI7l5PBs/fqXVY/ZnEBCwkLq138Od/fLa5lUJezKztur32b0xtF4Rd2D729zWbrUn5tucrVl/5K2JY2DTx6g2oEFdPWchDRsxo8pVl7rPJLMzEw+/PBD3nrrLbzK+8IXmkpBSYTitIh0AZSIuAMvAYeda1blJm/+xOUGsuPiZqCURc/ELgU51hwGLxnMvH3zcN/5HA32fcNv6z1o2dLVlhnYsmxEfRBF4pdbucbrS6qxg+zrujDMzY1ZQ4bQvXt3Jk+ezDXXXONqUzVViJIIxXMY3U+NgXhgtWOf5hLZYDLh4+ZGh8tIo6CUndjYyQQF3Yi/f5sytK7ykpydTL+5/Vh/aj2s+pxOttf5ZZNQXlb4TF2fyqEhh/A4upsuHq8gnp781rsfD6xYgae3NxMnTuTpp5/GzU2PatdcWUoiFFallE5FWoZsMJnoHBiI12X84FNT15CdfZSrrvqgDC2rvESlRNF7dm8OJx6HhXN4sNUAfvoJLnMsQZlgzbBy/K3jxI2Pwiu0GlevvJ+zP/3N4zt3smrxYvr168e3335LgwYNXG2qpopSEk+1TURWiMgTIlKqR2ARuUtEDonIURF5q4gyD4vIfhHZJyJzSlN/RSTLZmNnRsZldzvFxEzGw6M6tWo9WEaWVV62x2zn+qndOBofh33GKl7vNYB588qHSCSvTmZ7m414jv+UbgGDab28Dv/3+6c0CA8nMjWVRYsWsWjRIi0SGpdSkhXumorIDcAjwIcishuYq5SaW9x5jnjGeOB2IBpDcJYppfbnK9MceBu4USmVIiKVfi3urWlpWC8zEaDZfJbExMU0aPCCDmJfhF8P/8rD8/tjS6uFbcYaJnx4Dc8+62qrwGqycmzkMTKm/sW1XmPx4xgxnW6h9x03syc6mmeeeYbPPvuM4OBgV5uq0ZQshYdSaqNSajjQAUjDWNDoYnQBjiqljiulzMBcjDUu8vM0MF4pleK4ztkSW15BiXBMtLucRIBxcdNRyqJnYl+ESdsn0Te8L9a4VrjP2Myv08uHSCT+msjW1hvxm/oBHeRFvEOyGNOjBw3WriU3IID169czceJELRKackNJJtwFiMhAEfkF2AokADeUoO4GwOl829GOfflpAbQQkQ0isllE7irChmEisl1EtickJJTg0uWXDWlptPHzo/olzuhSyk5MzGSqVbsZf3898qUw8oa/Prv8WeT4XdRYto4Nv9fF1ev0WJIsHBh0gMg+kXgEe1P35gyO3NKDZrk5vLNxIx988AG7d+/mpvI0TlejoWQtikjgemC0UqqZUuo1pdSWEpxX2LQlVWDbA2gO3AIMAKaKyAWPUUqpyUqpTkqpTrXKyxCVS8CmFJsuMxFgSsqf5OQc10NiiyDXmstjix7jsw2fITuG0XrPUrZtCOC661xrV8LCBHa0+oug2e/Q9EVFyIIa9PHMpeWaNdRr1Ypdu3bx4Ycf4u3t7VpDNZpCKMmop6uVUvZLqDsaaJRvuyHGcqoFy2xWSlmAKBE5hCEc2y7heuWefZmZmGy2ywpkx8RMwsOjBjVrFlwuRJOSnUK/ef1Yd3IdrP6U2/3fZP56wZWTls3xZo68eAT7gmV08PwKT0nkjzgP+nVcjoeHB9999x3PPfecHvKqKdcUKRQiMlYp9RqwUEQKtgRKssLdNqC5iIQCZzCC4Y8WKLMEoyUxQ0RqYnRFHS+F/RWKy00EmJsbR1LSUho0GI67u09ZmlbhOZl6krtm9eJw4lFYNJunr3+U8eNdl7NJKcXZOWeJenEroWlfU4c/ybqqGUPcqzNrwQL69OnD999/T8MyWgZXo3EmxbUo5jn+XtLKdkopq4i8CKwE3IEflFL7ROQjYLtSapnj2B0ish8jLcjrSqmkS7leRSDCZKKelxehPpfm5I0gtlV3OxVgZ+xOes26myRTNvbZf/DZs7fwxhuuy9mUeyaXw88eJunXJFrVW0jtjPX80e1G+m7aRLVatfj555958MEHLzvPl0ZzpShuhbutjrfXKKXOEwuHAFx0BTyl1ApgRYF9H+R7r4BXHa9Kz4bLSARozMSeQnDwLfj5lZN8E+WAFUdW8NDPD2M21cBtzmpmj21Dfxct8qeUIu6HOE69shEPSxpNx/2H/U3uZ/BLf7JywwaGDh3K6NGjqV69umsM1GgukZJ0jD5VyL4hZW1IZSc6J4eTubnceIkd5ikpq8jJidJDYvMxecdk7g2/l9yYFgTO28Saea4TiewT2ey9fTdpQ7+gU/YTtAv9kvd3vUfPfn057uvLmjVrmDJlihYJTYWkuBhFf4y4QqiILMp3KBBIdbZhlY3LTQQYEzMJT8+a1Kp1sdBQ5UcpxXt/vcf/Iv6HHLuL0G0/8/vqQJo3d4EtdkXMhBjOvL6OFrlfEMxuYlu14d7YGHbPOcg777zDe++9h295mAau0VwixcUotmKsQdEQY4Z1HunALmcaVRnZYDLh5+ZG+4CAUp+bmxtLYuIyGjUagZtb1R4+abaZeWrpU8z+ZzbsGMr1yRNY9rcHNWteeVuyjmRxaMgh7H9vpZPbyyg/L8Y1acurkZF06dKFHVOmcO211155wzSaMqa4GEUUEIWRLVZzmUSYTFwfFITnJQyDjIv7AbBV+W6n1JxU+s29n7Un18Cfn/BQvbf5abVwiWMDLhllU0R/Fc3Jdw+ifHwInXQnO+bewIDNm4mLiuKrr77ixRdfxN3d/coaptE4iSK9loisc/xNEZHkfK8UEUm+ciZWfNKtVvZkZFzSsFilbMTETCE4+Db8/FzQt1JOOGU6xQ1Tu7MuKgIWzeTNG95hbviVF4nM/Zns6rYZ28j36CqPETA1h/unPkyXNWtoeeut7Nu3j5dfflmLhKZSUVzXU95ypy5o1FcutqSlYefS5k8kJ/9Bbu5JmjYdXfaGVRB2xe6i16y7SUjNQub9zqQ3b+Ppp6+sDXaLndOjT5M0ajkt1Rf4c4KdTdtw1yO9ICSEOXPm8Mgjj+ghr5pKSZEtinyzsRsB7kopG9ANeAbwvwK2VRoiTCbcgG6XMOLJCGLXombN+8resArA70d/p/sPN5N41gPvORGsGH/lRSJ9dzo7O2/F7b2RXGd9EffgDIbWq0fHffu4e9AgDhw4wIABA7RIaCotJekwX4KxDGpT4CfgGqDSrxtRlmxIS6Odvz9BHiXJmPIvublnSEr6lbp1n8TNreqtjTxt5zTunn0POWeaUWvpZjYtacudd16569tz7UR9EMXOzjvJjbUS3NHE2lYtqZ2UyFo/P1avXs306dOpUaPGlTNKo3EBJREKuyMX0/3AV0qpl7gwC6ymCKx2O5tMpksaFhsbmxfEvsKP0C5GKcUHaz5g6C9DsR/9D623rmf7mvq0b3/lbEjbmsausDV4/9/LNLg7i5OfnqLFqd3cfuQIz7/5Jnv37qVnz55XziCNxoWUaClUEXkIGATk9X+4KINOxWNvZiaZdnup4xNK2YiNnUL16v/Bz6+Zk6wrf5htZoYsHcqsf2bCzqe4wzKRBWs8uYzlxUuFLdvGif+eIGtMOO3kazzdUvg26iQvD9lLx44dWbFyJde5OhWtRnOFKenM7Fsx0owfdyT5C3euWZWHS00EmJz8O7m5p6lX7xlnmFUuMeWYuHNmL0Mk/vqIZ+pPZfmyKycSqX+nsrvNSgK/GEo79T6mOl7c7O3F20ePMnbsWDZv3qxFQlMlKclSqJEiMhxoJiKtMFat+8T5plUOIkwmGnl707iU4ziNIHZtata810mWlS9Om05z50+9OZh4EJb+yOiBjzNy5JVJ7GfNsBL1dhRnvjtDi6BwanpsYGK9Rrx0+jS33XEHkRMnEhoaWmwdFouF6OhocnJynG+wRlMMPj4+NGzYEM8yTJ18UaEQkZuAmRipwgWoKyKDlFIbysyKSopSig0mEzeVcknLnJzTJCUtp3HjN6pEEHtP3B7u+Kk3iaYMPBb8zuyPe/LQQ1fm2smrkznx5N/YohOo+fz1TPGqy4Jv7cRmZTF95kwGDhxYotFM0dHRBAYG0qRJEz36SeMylFIkJSURHR190Yeb0lCSGMU4oLdSaj+AiFyDIRydysyKSsqp3FzOmM2lDmQbM7HtVSKI/cexP+gX/iA5piCCfolg+fR23FCShXYvE6vJyrHXjiDTJtFeppDeqD7Xr3bj0OHDDBo0iDVffknNUuQFycnJ0SKhcTkiQo0aNSjrJaNLIhReeSIBoJQ6ICKV/zG3DIjIi0+UYv6E3W4lNnYq1avfga/v1c4yrVwwfdd0nl42DHt8a67asJzVyxvStKnzr5u0PImTT63i6rOfEsxe9tdrQO/TR5EmTVi5ciV33HHHJdWrRUJTHnDG97AkweydIjJJRLo7XhPQSQFLxAaTiUB3d9qVIhFgcvJv5OZGU79+5Q1iG8Nf/8tTy57CduxWuu7/m+1/OV8kLEkWDgw6wIl75hOW8AS+PkcZUa0a7WJjePC114iMjLxkkdBoKjMlEYpngWPAG8CbGEuVVl4vVoZEmEx0CwrCvRQKHxMzCS+vutSo0ceJlrkOs83ME4uf5P/WfwS7nuRhy3LW/B6Es+esJSxMYPs16zg79yzm4W1Z1LQxTXKyWBsaytZt2xgzZgz+/hU74YC7uzthYWG0bduWPn36kJpaNqsBnDhxgrZt25ZJXfkZNWoUDRo0ICwsjLCwMN56660yv0Yeu3fvZsWKFRcv6GDx4sWICAcPHjy3b+3atdxzzz3nlRs8eDALFiwAjAENb731Fs2bN6dt27Z06dKF33777YK6o6Ki6Nq1K82bN6d///6YzeZCbdi7dy/dunWjTZs2tGvXjpycHLKysrj77rtp1aoVbdq0ceo9y0+xQiEi7YC7gMVKqXuVUn2UUl8opfTQjouQarEQmZlZqmGxOTmnSE7+jbp1n8LNrfJNVUnLTePOn+5m5j8/wppRvN1mGuGzPZ2a2M8cb2b//TvJenAEHVMfZf8L67lrxt0Mio5mxOefs3XrVjp27Og8A64gvr6+7N69m8jISEJCQhg/fvzFT3IxI0aMYPfu3ezevZvPPvusxOfZbLZSXae0QhEeHk737t2ZO3duic95//33iY2NJTIyksjISH755RfS09MvKPfmm28yYsQIjhw5QvXq1Zk2bdoFZaxWK4899hgTJ05k3759rF279twoppEjR3Lw4EF27drFhg0bChWjsqa4hYvewVjJbifQWUQ+Ukr94HSLKgmb09JQlG6hotjYqYCqlEHs6LRo7phxNweS9iO/TGfKC4MZ4sR1EpVSnA0/S9xzi2iW9hn+nGJFcG3e+foDOvfsyaRJk2jqpL6uV16B3bvLts6wMPjqq5KX79atG3v37gUgIyODvn37kpKSgsVi4eOPP6Zv376cOHGCXr160b17dzZu3EiDBg1YunQpvr6+7Nixg6eeego/Pz+6d+9+rt6cnByee+45tm/fjoeHB19++SW33norM2bMYMmSJdhsNiIjI3nttdcwm83MnDkTb29vVqxYQUhISIls//PPPxk5ciRWq5XOnTszYcIEvL29adKkCU899RR//PEHL774Ip07d+aFF14gISEBPz8/pkyZQqtWrZg/fz4ffvgh7u7uVKtWjdWrV/PBBx+QnZ1NREQEb7/9Nv2LWQoxIyODDRs2sGbNGu69915GjRp1UZuzsrKYMmUKUVFReHsba8bUqVOHhx9++LxySin++usv5swxsiA98cQTjBo1iueee+68cn/88QfXXnst7R3pCPLSxPj5+XHrrUa+Vi8vLzp06EB0dHSJ7uvlUFyLYiBwrVLqIaAz8FwxZTUFiDCZcAe6ljCQbQSxpxEScie+vk2catuVZm/8XjpOuJ6D8VH4LlrB7585VyRyz+QS2Wc3loHPc23aC5gDUrnb3Z1BNivjpk9n1apVThOJ8oDNZuPPP//k3nuNOTg+Pj4sXryYnTt3smbNGl577TWM5erhyJEjvPDCC+zbt4/g4GAWLlwIwJNPPsk333zDpk2bzqs7r5Xyzz//EB4ezhNPPHFu7khkZCRz5sxh69atvPvuu/j5+bFr1y66devGTz/9VKit48aNO9f1tHLlSnJychg8eDDz5s3jn3/+wWq1MmHChHPlfXx8iIiI4JFHHmHYsGF8++237NixgzFjxvD8888D8NFHH7Fy5Ur27NnDsmXL8PLy4qOPPqJ///7s3r2b/v37s337doYOHVqoTUuWLOGuu+6iRYsWhISEsHPnzove86NHj9K4cWOCLvJ7T0pKIjg4GA9H3reGDRty5syZC8odPnwYEeHOO++kQ4cOjB59Yfbo1NRUfvnllyuSSqa4UU+5SqlMAKVUgoiUfsWdKsyGtDTCAgLwL+G6BMnJyzGbY6hXr/x3F5SGVcdW0XfOA+SYgqi9KoJVc66lXTvnXEspRdz0OI6+ehSVYyekaQxzkoN5NiWFPgMGcOCrr6hdu7ZzLp6P0jz5lyXZ2dmEhYVx4sQJOnbsyO233w4Y9+Wdd95h/fr1uLm5cebMGeLj4wEIDQ0lLCwMgI4dO3LixAlMJhOpqan06NEDgEGDBp3r3oiIiOCll14CoFWrVlx11VUcPnwYgFtvvZXAwEACAwOpVq0affoYcbZ27dqda90UZMSIEYwcOfLc9p49ewgNDaVFixaA8cQ9fvx4XnnlFYBzLYGMjAw2btzIQ/km3OTm5gJw4403MnjwYB5++GHuv7/wpYM7derE1KlTCz0WHh5+7nqPPPII4eHhdOjQocjRRKUZZZQn0Bc732q1EhERwbZt2/Dz86Nnz5507NjxnChYrVYGDBjA8OHDufpq54+OLE4ors63VrYATfOvna2U0os3F4HFbmdLWhrD6tUr8TlGELseNWrcc/HCFYQZu2cwZOnT2OOvoc3uFfzxW0Pq13fOtXJO5nB08DZC1n6Od4dHmdpqLdPmbKZh48bMmzWL3r17O+fC5Yi8GIXJZOKee+5h/PjxDB8+nNmzZ5OQkMCOHTvw9PSkSZMm51oBed0kYATDs7OzUUoV6fwKc3R55K/Lzc3t3LabmxtWq7VEn6G4+oFzAw7sdjvBwcHsLqSPb+LEiWzZsoXly5cTFhZWaJmiSEpK4q+//iIyMhIRwWazISKMHj2aGjVqkJKScl755ORkatasSbNmzTh16hTp6ekEFpNzpmbNmqSmpmK1WvHw8CA6Opr6hfwoGjZsSI8ePc7N5enduzc7d+48JxTDhg2jefPm5wTN2RTXSngAY63s8cB3BbYr12NvGbMrI4PsUiQCzM4+QXLy79SrNwQ3t9KlIi+PKKX4cO1HPLn0SezHe3DHmb/Z/IdzRELZFWe+P8PxVmNose4+6spvhB9/ganhUxn+yivs27evSohEfqpVq8Y333zDmDFjsFgsmEwmateujaenJ2vWrOHkyZPFnh8cHEy1atWIiIgAYPbs2eeO3Xzzzee2Dx8+zKlTp2jZsmWZ2d6qVStOnDjB0aNHAZg5c+a5lk1+goKCCA0NZf78+YDxnduzZw8Ax44do2vXrnz00UfUrFmT06dPExgYWGhguSALFizg8ccf5+TJk5w4cYLTp08TGhpKREQEzZs3JyYmhgMHDgBw8uRJ9uzZQ1hYGH5+fgwZMoThw4efG8UUGxvLrFmzzqtfRLj11lvPjZT68ccf6du37wV23Hnnnezdu5esrCysVivr1q2jdevWALz33nuYTCa+uoJN1+IWLvqzuNcVs7ACUtpEgEYQG+rVK7zPtCJhsVl4YvEQRq37L+x+nGcCVrB8YTVKMZWkxGQdzSLyxtV4vfAorXPeJzrARkdl589Gjdi8eTPjxo0jwBkXrgBcd911tG/fnrlz5zJw4EC2b99Op06dmD17Nq1atbro+dOnT+eFF16gW7du+Pr6ntv//PPPY7PZaNeuHf3792fGjBnntSQuFx8fH6ZPn85DDz1Eu3btcHNz49lnny207OzZs5k2bRrt27enTZs2LF26FIDXX3+ddu3a0bZtW26++Wbat2/Prbfeyv79+wkLC2PevHlFxijCw8Pp172IS8kAACAASURBVK/fefseeOAB5syZg7e3N7NmzeLJJ58kLCyMBx98kKlTp1LN8Tv/+OOPqVWrFq1bt6Zt27bcd9991KpV64JrfP7553z55Zc0a9aMpKQkhjgCdsuWLeODDz4AoHr16rz66qt07tyZsLAwOnTowN133010dDSffPIJ+/fvp0OHDoSFhRXZhVamKKWc9sIYWnsIOAq8VUy5BwEFdLpYnR07dlTlnfv/+UeFbtpUorI2m1lt2FBP7dnT28lWOR9Tjkn1mHq7YhSKWz9QY8bYld1e9texW+3q1NhTap3vOnXSa4CyuHmoUd7eyt/L6//bO/Pwmq71j39WEjMx0xBDJKGRkNRMtQ1qvC3XPLWooa0aSqv81K26vTXV3JaitFQrMZOqmmmJeYghlCBRQ0gQGUTm9/fHOTlNZDqJjKzP8+znOXvvtdd61zn77Hev6fvK1KlTJSYmJvsLzYALFy7kepkaTVqkdj8CJySLz/Ic6+dQSlli6KJqC9wEjiulvCSJHIgxXSlgNHA0p2zJTcQoBNjWzKmA9+9vJSYmkCpVFuewZTnLrbBbvP7jv/jrwXmsti/Hc/xgunfP/nIeXXjEtf57ifa5w8NWlRkSEcDfx+Oo0rwlp5YsMQ2CajSa7MNsR6GUKiIi0ZnIuwkGSfJrxus9gS7AhSfS/Q/4ChjHM8DVx4+5Gxtr9voJwyB2VcqVK7j96OfunqPNj50IDn9I6e2/sX1he5o1y94yEmITuDHzOjFT5uMUv5Q7ZcvQ0vsBxYoXZ/ayZQwePFhrLWk0OUSGU16VUk2UUucAP+O+q1LqGzPyrgrcSLJ/kydCqCqlXgKqicjWDGx4Vyl1Qil1IrtVEbMb77AwwDwhwMeP/QkJ2YmNzdACO4i9138vTZe2JPhePDX2HODUuux3EuE+4Vxw3UCZz97EMf5rDpew4uWQO3Tu0oWLFy8yZMgQ7SQ0mhzEnLURXwNvAPcBROQMhoh3GZHaP9c09824LmMe8HFGGYnIUhFpJCKNUhscyk94h4ZSxsqKumboBgUGfg8obGxycPVZDvLTmVW0+6kDj+9Uo8nZI5za5kZ2TulOiE7Af7I/fo1+oe7F/lgWusIAoF+Z0nzr5cXatWt54YUXsq9AjUaTKuY4CgsReXI+nTlCKzeBakn2bYHbSfZLAS7AfqVUANAM8FJKFeg4FwdDQ2lhbY1FBm+4CQmxBAb+QPnynShatFq6afMbIsJ/933JwM0DiPdvSc/wg/z5a3XMHJYxi7BjYZxy+5Pr/7vOsVfi+bZUMRxioyk9ciS+Fy6YFnNpNJqcx5z+jhtKqSaAGAeoRwGXzbjuOOBojLF9C+gD9Es8KSKhgCkyjFJqPzBORE6Yb37+4n5sLBcjI3mrcuWM0973Ijb2boGLiR0bH8vQzR/w0/llcOZtPnVZxpf/LZxtIUvjH8dzfdIlLOdNxcXid95r2oyf9v+Os7Mzm77/nubNm2dPQRqNxmzMaVEMBz4CqgN3Mbz5Z6j7JCJxwEhgB3ARWCsivkqpL5RSz2Qg6EPG9RPmDGTfvr2EIkWqUb58x5w2K9sIjw6n7Y+d+en8MtSB/7D8jZVM/SL7nMTDgw/5q/YyXpjXnhqsZr3VI34/tZsvvviCU6dOaSeRAUnXjGzbtg1HR0f+/vtvpkyZQvHixQkKCko1rVKKjz/+pwd49uzZZgnhgUGrqWjRooQa732AFStWMHLkyGTp3N3dOXHC8A4YERHBe++9h729Pc7Ozrz66qscPZpy0uPJkyepV68eDg4OjB49Os1V2/v378fNzQ1nZ2fT4rwbN27QqlUrnJyccHZ2ZsGCBWbVR5M6GToKEQkSkT4iUsG49RGRe+ZkLiLbRKS2iNiLyFTjscki4pVKWveC3JoAw/hEIaVonM4SfoDHj68SErILG5uhGBpp+Z/b4bdptOhV/rixi6K7lrJr4v8YPDh7PERcRBx+Iy/y6JV3cL75Pg8LB/M6sLRpU/48e5bPPvuMwoV1UEVz2bNnD6NGjWL79u1Ur14dMEhHzJkzJ9X0RYoUYePGjdy7Z9bfOhkeHh40btyYTZs2mX3N0KFDKVeuHH5+fvj6+rJixYpUyx4+fDhLly7Fz88PPz8/tm/fniLNw4cP+eCDD/Dy8sLX19e0UtvKyoo5c+Zw8eJFjhw5wsKFC7lw4ckJlxpzybDrSSn1PUkGoRMRkXdzxKICjHdYGA1LlaJYBkKAt29/D1jwwguDc8ewp8Q3yBf3ZR259yiEivt/Ze+SjmRXHJuQPSFcGnqJiIAIKpa7wKZQS6YVKcznXy9g2LBhWFgUPC3KMdvH4HMne3XG3V5wY36HjCUbDhw4wLBhw9i2bVsyhdzBgwezYsUKJkyYkELu28rKinfffZd58+YxdepUs226evUqERERzJo1i2nTpjFo0CCzrjl69Ci//PKL6betVatWCmG7wMBAwsLCTK3IAQMGsHnzZjp2TN4CX716Nd26dTM5xETRRxsbG2yMWmulSpXCycmJW7dumWQwNJnDnH/hbmCPcfMGKgGZWU/xXBCdkMDxsLAMp8UmJMRw586PlC//BkWL2uaSdVlnn/8+Gi9+mXsPYnnx0J+cWZ89TiIuNA6/gUeIer0/t6L2MtZhLC89uMy+Ll048ddfvPfeewXSSeQl0dHRdOnShc2bN6eQ6ShZsiSDBw9OswtmxIgR/PLLL8m6kCC5rMSTeHh40LdvX1555RUuXbqUrGsrLXx9fXFzc8Myg5epW7duYWv7z/8jPTnukJAQ3N3dadiwYapy5gEBAZw+fZqmTZtmaJ8mdTJsUYjImqT7SqlVwK4cs6iAcjI8nGiRDPWd7t3bQmxsUIGIib3K5xcGbX6HhGBH2tzdxuZfa2SLZtP93+5xb8ASaj6YjSWh/HZ3O/fVC2zYuDGFzk5BxJw3/5ygUKFCtGjRguXLl6fqEEaPHo2bm1uy8YhErK2tGTBgAF9//XUybafOnTub4lo8iaenJ5s2bcLCwoJu3bqxbt06RowYkety3CdPnmTPnj08fvyY5s2b06xZM9MK/YiICLp37878+fMzjBWhSZusvLLZATWy25CCzkEzhQANg9jVKVeufW6YlSVEhP/uncaALW+REPAyw6wOsmPN0zuJ2Aex+PX8k4Q3ulLnwX+4VCiSBghF3nuPixcvPhNOIi+xsLBg7dq1HD9+nGnTpqU4X6ZMGfr168eiRYtSvX7MmDEsX76cR48eZVjW2bNn8fPzo23bttSsWRNPT088PDwA0pXjdnZ25syZMyQkJKSbv62tbbLIbenJcXfo0IESJUpQoUIFXn31VZOKbGxsLN27d6d///5pxqXQmIc5K7NDlFIPjNtDDK2JT3PetIKFd2gojsWKUSmdQdfIyCs8fLgHG5th+XYQOy4hjgHr3mPKgUlwth+zXLezdEFZzIy/lCbBm4I5VvcYbJiPtTrEJ0C/WnYsOnCA7777zqTAqXk6ihcvztatW03Kqk/y0UcfsWTJklTjQ5QrV45evXqlet2TeHh4MGXKFAICAggICOD27dvcunWL69ev07hxY7y9vblz5w4AJ06cIDo6mmrVqmFvb0+jRo34/PPPk0XZS1R+TcTGxoZSpUpx5MgRRISffvopVTnuLl26cODAAeLi4oiMjOTo0aM4OTkhIgwZMgQnJyc++ugjs747Tdqk6yiUoa3nClQ0bmVFpJaIrM0N4woKiUKAGU2LDQxcClhiY5M/B7EjYiJovawzP1/8HqtDn7K+/yrGjX06CemYoBgu/2sHAd02sqvILlqV2sdLlhYUnzyZk2fOJIvHrMkeypUrx/bt2/nyyy9TPIArVKhA165dTdHgnuTjjz9ONgMprTEKT0/PFC3Arl274unpSeXKlVmwYAGdOnXCzc2NMWPG4OHhYRpzWrZsGXfu3MHBwYF69eoxbNiwVFsL3333HUOHDsXBwQF7e3vTQPbixYtZvNggounk5ESHDh2oX78+TZo0YejQobi4uODt7c2qVavYu3evKdTqtm3bMvEtapKRkbwscDKr0rQ5seVHmfGLERHCvn2y7PbtNNPEx0fJwYMV5Ny5f+eiZeZzO+y21J7dQJhsKaXcl8jRo0+XX0JCgtxZdUuuFh8tsRSRi0VKCiDNmzeX8+fPZ4/R+QgtM67JT+SFzPgxpVQDEck4wvhzijlCgMHBm4iNvZcvV2JfCL7Aa0s7cS/yHraHvfjzh07Y2WU9v+hb0Vx/63cq7/+Uylxkh4UFY6ysWDh3Ie+//76ezaTRFDDSdBRKKSsxrK5uCQxTSl0FHmEQ+xMRaZBLNuZ7DoaGUt7KijrFi6eZJjBwKUWL1qRcuXa5aFnG7Pf/g44//ZuoiCI0/OsPdm1oSNmyWctLRLjz4x3ujvai3qORhFkI/RMg4l//YteiRcmmO2o0moJDei2KY0AD4N+5ZEuBxTs0lJdLl05z+l9k5GUePtyHnd1UDKK5+YOfTq82xLW+Z0/XyG14bqhJVhdAR12Pwu+dU9zaF4ZH1QM4RsbjWb4CUxYtonv37loGXKMpwKTnKBSAiFzNJVsKJEExMfg9fsww4yrQ1Lh9eylKWeWbldgiwpTdM/ni0ES4/hrja2xixudls6TZJAlC4NdXiR8/mVqxWxlQ1hqfW7cYNmwYB2bOpGxWmycajSbfkJ6jqKiUSnNemYjMzQF7ChzeGayfiI+P4s6dFZQv34UiRfI+dkJcQhyD1ozil8uLUb59WNppBUMHZW1mU+SVSG71WI3NmcmUJJClgEW5SuzftN8kzqbRaAo+6TkKS6AkqQcg0hjxDg2liFI0TEMI8N69jcTF3adKlbyXxoqIiaDj8j4cDPqNIscnsHXsNF5vk/muMIkXbs4LwPL/xuAY74W/sqCrsqDpxIl4/+c/FC1aNAes12g0eUV6T4lAEflCRP6b2pZrFuZzDoaG0tjamiJpzOS5fXspRYvWomzZ13PZsuTcibiD23x3Dt75nXKHFnHqqxlZchKPLj7idMvTeH9ylAtWh5gDDGrYgLk+Pnz55ZfaSeQhJbNBX+X27dv06NEjzfMPHz5MtrI7o/Tm0KVLlxQS8oMGDWL9+vXJjiWt3+XLl+nUqRMODg44OTnRq1cv7t69myLvlStX4ujoiKOjIytXrky1fB8fH5o1a4abmxuNGjXi2LFjAPz11180b96cIkWKMHv27KeqY4EnrXmzwOmszrnNyS0/raOIjIuTQvv3y4QrV1I9HxFxUfbtQwICpueyZcm5EHRByn1RU/i0uDh0+lUCAzOfR3xMvPw96ZTcVu3ki6L9pViRYlKyeHFZsGCBxMXFZb/RBYz8sI6iRIkSOV6Gv7+/ODs7Z1t+ISEhYmtrKy+++KJcu3bNdHzgwIGybt26ZGkT6/f48WNxcHAQLy8v07m9e/fKuXPnkqW/f/++2NnZyf379+XBgwdiZ2cnDx48SGFD27ZtZdu2bSIi8ttvv8lrr70mIiJ3796VY8eOyaeffiqzZs3KlvrmFrm5jqJN7riqgsvx8HBiRdJckR0YaBjEtrF5J5ct+4d91/6k48p/Ex1ZiFdv/sG2tY0wI5x3MiJ8wrnb9RuqBEzHigiuREGrTp347rvvTPLOmn/w8xtDRET2yoyXLOmGo2PmxQavX7/O4MGDCQ4OpmLFivz4449Ur16dq1ev0r9/f+Lj4+nYsSNz584lIiKCgIAA3njjDc6fP4+vry/vvPMOMTExJCQksGHDBj777DOuXr2Km5sbbdu2ZcSIEab08fHxTJgwgR07dqCUYtiwYYwaNSpd+zZs2MCbb75J5cqV8fT0ZOLEiRnWafXq1TRv3jxZONxWrVqlSLdjxw7atm1rklVv27Yt27dvp2/fvsnSKaUIM66FCg0NNa0Sr1SpEpUqVeK3337L0KZnnTT7HkTkQW4aUhBJFAJskYqjMAxir6RCha4ULpxxaNScYMVJT15f2ZboB5V4J+EIe3/OnJNIiEng77GHiXypHfYBkzhPBG3LlaOjhwdbt27VTqIAMHLkSAYMGMDZs2fp378/o0ePBuDDDz/kww8/5Pjx46nKZ4BBKuPDDz/Ex8eHEydOYGtry4wZM7C3t8fHx4dZs2YlS7906VL8/f05ffq0qTyAyZMn4+WVIlYZ8I9Ued++fU2ighlx/vx5GjZsmGG6W7duUa3aP/Ho05Iqnz9/Pp988gnVqlVj3LhxTJ8+3Sw7nifMWZmtSQPv0FCcihenXKFCKc4FB68nLu4BNja5P4gtIny+Yzb/OzoebrRkWv0tTBxTLuMLkxB2PIxLgy8h5ydRj6N8DDwcOJCNc+emCHyjSU5W3vxzisOHD7Nx40YA3n77bcaPH286vnnzZgD69evHuHHjUlzbvHlzpk6dys2bN+nWrRuOjo7plrV7927ef/99rKwMj5XE++SLL75INf3du3e5cuUKLVu2RCmFlZUV58+fx8XFJdV1N5ldiyNmSpV/9913zJs3j+7du7N27VqGDBnC7t27M1XWs07+Wf1VwEgQ4VBYWDrdTksoWtSesmVb56pd8Qnx9PtlJP87Oh7Lv3qxrvOuTDmJ+MfxXH93L6ebLOO/1/5LK/bxRvVqdNq9m+UrVmgnUcDJzMO2X79+eHl5UaxYMdq3b8/evXvTTS8imcp/zZo1hISEYGdnR82aNQkICMDT0xNIKVWeKFMO4OzszMmTJzPM39bWlhs3bpj205IqX7lypUmGvGfPnqbBbM0/aEeRRS48esTDuLhU1088enSB0NCDVKnybq6uxH4U84hXFnbF8+oiip/+hEMfedDj3+bPQgr98wE3qn+EzfftKa3Gsz3qd0ZNmMDWv/6iTRs9ZFUQadGihenh+8svv5jUeps1a8aGDRsATOef5Nq1a9SqVYvRo0fTuXNnzp49S6lSpQgPD081fbt27Vi8eLFJwvzBg/R7rz08PNi+fbtJqvzkyZMmW9zd3VmzZg0xMTEArFixwjQO0a9fPw4dOpRs7GD79u2cO3cuWf7t27dn586dhISEEBISws6dO2nfPmUcmCpVqvDHH38AsHfv3gxbTs8lWR0Fz6stv8x6WnzrlrBvn1yJjExx7vLlD2X//kISHX031+y5E35H7KY1FiZbyAudvxV/f/OvjYuIk4D+W+Ue9iIgW0A61q8vp0+fzjF7nzXyw6wnpZRUrVrVtM2ZM0f8/f2lVatWUq9ePWndurVcv35dREQuX74sTZo0kcaNG8uUKVOkSpUqIpJ8VtO0adOkbt264urqKu3bt5f79++LiEjfvn3F2dlZxo0blyx9bGysjB07VpycnKR+/fryzTffiIjIZ599Jlu2bElmq7+/v1SpUkUSEhKSHX/ppZfkyJEjIiIyZcoUcXFxEVdXV+nWrZsEBQWZ0l28eFHat28vDg4O4uTkJL1795Y7d+6k+E6WL18u9vb2Ym9vLz/88IPp+JAhQ+T48eMiInLgwAFp0KCB1K9fX5o0aSInTpwQEZHAwECpWrWqlCpVSkqXLi1Vq1aV0NDQrPw0uU52z3rK8wd/Zrf84ijeunBBKh88mOJGj4uLlAMHysj5871zzZaLQX9Jmc/thEnFpF6PLRISYv61D/Y8EB+bJRKLhQSDDChcWObOmSOxsbE5Z/AzSH5wFJnh0aNHpnvXw8NDOnfunMcWabKTvJAZ16RCWkKAwcHriIt7mGsxsfddPUiHlV2IibLkX6H72fhLE7OE/eJC4/Afc4YDK84wv/ACupLAxVatmLl8OXZPozGuKRCcPHmSkSNHIiKUKVOGH374Ia9N0uRjtKPIArejo/GPimJU1aopz91eQrFijpQp457jdvx4bB1DfnsbCanBmIq/M3dmLbOE/e5vukHkgHFUivDiPyqWWOuy1J7/M1P69dMqr88Jr7zyiim2tEaTETk60qqU6qCUuqSUuqKU+r9Uzn+klLqglDqrlNqjlKqRk/ZkF2kJAUZEnCcs7BA2Nu/m6ANXRPh062wG/94LbjViUaNDzJucsZOIfRDL9fYrsOz2EtUi1uJBFG/06sHFixfp37+/dhIajSZVcsxRKKUsgYVAR6Au0FcpVfeJZKeBRiJSH1gPfJVT9mQn3qGhFLOw4KUntHUMK7EL88ILg3Ks7PiEeHqvHM30k59Q6HJPtvfbzfCB5TO8Lnh9IHeqdqHGzncI5j59K79ArR07WOrpaZp2qNFoNKmRk11PTYArInINQCnlCXQBLiQmEJF9SdIfAd7KQXuyjYOhoTS1tqZQEiHA+PhI7tz5iYoVu1O4cM48eCNjI2n1bT+OhW3B2vdjDk35Cue66fv6mKAYLo+8zIZ1G6iv9uAJhHz4IcumTqVEZrU8NBrNc0lOOoqqwI0k+zeBpumkHwL8noP2ZAsRcXH4RETwf0/IVwQFrSU+PjTHBrHvRgTRZP6b/B13nOoXvub4t6OoVCnt9CJC8FJfokaPYlpMKB6cpkF9N5YuX2aW/IFGo9EkkpNjFKl1eKdcUw8opd4CGgGz0jj/rlLqhFLqRHBwcDaamHmOhocTDylWZAcGLqFYsTqULv1qtpd5Megytb9qzt9R52gasJGLK9J3EtG3ovi70XSKv9+YF2L2U6zQeb766iuOnjiuncQzzNSpU3F2dqZ+/fq4ublx9OhRpkyZkkJoz8fHBycnJwBq1qzJK6+8kuy8m5sbLi4uZpV5+vRplFLs2LHDdCwgICDF9VOmTEkm1T179mxefPFFXFxccHV15aeffkqR94MHD2jbti2Ojo60bds22UrtpFhaWuLm5oabmxudO3c2Hff396dp06Y4OjrSu3dv0+I9TebJSUdxE6iWZN8WuP1kIqXU68AkoLOIRKeWkYgsFZFGItKoYsWKOWKsuXiHhqKA5kkcRUTEWcLCjhhXYmfvgPDuy964ftOcsKhw+sbsw3v5vylePPW0IsLdOScIrtGMGqcmcY4ohjdtxqcXL/LJJ5+YNHg0zx6HDx9m69atnDp1irNnz7J7926qVatG3759WbNmTbK0np6e9OvXz7QfHh5ukrq4ePFipsr18PCgZcuWZgv6gUFscNeuXRw7dozz58/z559/pqrLNGPGDNq0aYOfnx9t2rRhxowZqeZXrFgxfHx88PHxSSY+OGHCBMaOHYufnx9ly5Zl+fLlmaqb5h9y8slxHHBUStkBt4A+QL+kCZRSLwFLgA4iEpSDtmQb3qGh1CtRgtJJHrqGmNhFeOGFgdla1jLvDby7oz8SWp3PHX5nymj7NNNG/R3FuaHnCNo1GHfOM7FYcZwWfsuyQYP0bKZcxm+MHxE+EdmaZ0m3kjjOT1taIjAwkAoVKlCkiCGsbdIJCmXKlOHo0aM0bWro+V27dm2yFkCvXr1Ys2YN48aNM6m5rlq1KkObRIT169eza9cuXnnlFaKioswKXDVt2jT27duHtbU1AKVLl2bgwJT/nS1btrB//34ABg4ciLu7OzNnzsww/0Tb9u7dy+rVq03XT5kyheHDh5t1vSY5OdaiEJE4YCSwA7gIrBURX6XUF0qpxPbhLAzhVtcppXyUUqlrEecT4kU4HBaWbFpsfPwj7t5dRcWKPShUKOPZR+YyftM8hu3qibrTgJ9bH0rTSUiCcOeLA/xa+3N67u5JT84z6c3OjA3wZ8A772gn8ZzQrl07bty4Qe3atfnggw9M2kUAffv2NWkoHTlyhPLlyyfTM+rRo4dJYfbXX39NFufh9u3bdOrUKdUyvb29sbOzw97eHnd3d7Zt25ahneHh4YSHh2Nvn/ZLTyJ3797FxsYGABsbG4KCUn+XjIqKolGjRjRr1sykiHv//n3KlCljakWnJTGuMY8c7YsQkW3AtieOTU7yOW/jg2aScxERhMfHJ3MUQUFriI8Py7ZB7PiEeHou+5hNgQso4t+NPSN+5uUmxVJN+/hSOIEdJlA1YAk1SSDBxpYNy7fRsWPHbLFFkzXSe/PPKUqWLMnJkyc5cOAA+/bto3fv3syYMYNBgwbRp08fWrRowZw5c/D09EwRuKdcuXKULVsWT09PnJycKJ6kb7NKlSppOgAPDw/69OkDQJ8+fVi1ahXdunVL8+VEKZVphVlz+Pvvv6lSpQrXrl2jdevW1KtXz9RaebJ8TdbQndaZIDFQUdKB7Nu3l1C8uBOlS7d86vwjYyN5Zd5bnHq8ifKXx3J86izsalqmSCfxwt2Ju0iYPZhacostwMkhQ7gwf362xE3WFEwsLS1xd3fH3d2devXqsXLlSgYNGkS1atWoWbMmf/zxBxs2bODw4cMpru3duzcjRoxgxYoVZpUVHx/Phg0b8PLyYurUqYgI9+/fJzw8PIVEOBgGpu3s7LC2tqZEiRImZdr0qFy5MoGBgdjY2BAYGEilNGZwJEqH16pVC3d3d06fPk337t15+PAhcXFxWFlZpSkxrjEPLTOeCbzDwqhauDDVjf3A4eE+hIcfy5aV2EERwdSe2oZTkZupfW0+VxfNTdVJPPrrEUddvqbCrA5Yyi0+rlqdKkeP8sWyZdpJPMdcunQJPz8/076Pjw81avwjdNC3b1/Gjh2Lvb09tra2Ka7v2rUr48ePT1WGOzV2796Nq6srN27cICAggOvXr9O9e3c2b95MyZIlsbGxYc+ePYDBSWzfvt0kcT5x4kRGjBhhCj8aFhbG0qVLU5TRuXNnVq5cCRhiRnTp0iVFmpCQEKKjDXNg7t27h7e3N3Xr1kUpRatWrVi/fn2612vMJKtqgnm15aV6bLVDh6T3+fOm/UuXhsv+/UUkJub+U+V7/vZlKfmpvTCpqLQesUFiYlKmiY+NF//PfORjq4/FmhLyXwtLmfef/0hMaok1uU5eq8eeOHFCmjdvLk5OTlKvXj3p2rWrBAcHm84HBQWJlZWVfPfdd8muq1GjRrJ0Ismlxm/duiUdO3ZMUd7AgQNT5LVlyxbp0KGDiIj4+vqKu7u7uLq6iqurq/z8PrTBIwAAG31JREFU88+mdAkJCTJz5kypXbu2ODs7i5ubm6xatSpFGffu3ZPWrVuLg4ODtG7d2iRzfvz4cRkyZIiIiHh7e4uLi4vUr19fXFxcZNmyZabrr169Ko0bNxZ7e3vp0aOHREVFZfxFPiNkt3qsklSmpeVnGjVqJCdOnMj1cv+OiqLGkSN87eDAKFtb4uIiOHy4ChUq/Bsnp5RzwM1lh+9h3lzdmdg44d1Sv7J4UvMUmk0RR+4S+K/BlH+wgxeJx/lld5b+uFQHWMlHXLx40bQ2QaPJa1K7H5VSJ0WkUVby011PZvKkEGBQkCfx8eFPNYj93f6NdPRsTWxEGebUPcyS/yR3EgkxCdzo/yOPmzvg+GAb660smf3tt+w9oKNwaTSa3EMPZpvJwdBQSlpaUt+ojxQYuITixZ2xtm6RpfzGrlnA/ItjsQxuyqaeXrzZJvlCwrAjDwhs14U64Qe5BHzd8jVGrPPkhRdeeNqqaDQaTabQLQoz8Q4NpZm1NVYWFoSHnyI8/ESWVmInSAJvfDuW+X+Nofjf/+bEqD3JnET843jOjD3DoOZDORF+kG+KleTq+vX878B+7SQ0Gk2eoFsUZhAaF8e5R4+YbJQPuX17CRYWRalc+e1M5RMZ85hmX73NufgNvBAwmtMz5vJC5X9mNoX96sfN3r354HEAh3lIlWEfMH3ODEqVKpWt9dFoNJrMoB2FGRwJCyMBeNnamri4cIKCVlOxYm8KFSprdh53wu7x0swu3Cl0mPqBczmyaCzFjOvo4iPiuPbGF1T6Yzr2xPFyeRvmbD1Es2bNcqZCGo1Gkwm0ozAD79BQLIGm1tYEBf1AfHxEpgaxz968SvOFHYm0+pvO0WvZ9F0PEkNZPFx9lruDulIn9hqHUZx+fyRfLphDYXMCX2s0Gk0uoB2FGRwMDcW1ZElKWVlx6fYSSpSoh7W1eW/7W32O0nXtG8QlCB9X3Mvs0YbB77iwOPa8t4dbngPoRRBzq9Sk086tfODsnJNV0Wg0mkyjB7MzIDYhgaNGIcCwsBNERJyiSpX3zBrEXrBjM503tCL+sTVLmx4yOYng7w4yu0ov3vR8k4mFI9ky5UvG3LjKi9pJaJ6C3IpHsX//ft54440Ux4cOHcqFCxdSuSL3CA4OplChQixZsiTZ8SdVC1asWMHIkSNN+z/99BMuLi44OztTt27dZLEzEomOjqZ37944ODjQtGlTAgIC0rQjPj6el156Kdn3tHfvXho0aICLiwsDBw4kLi4ui7XMfXSLIgPOREQQmZBAy9KlCQz8CguLYlSq1D/D60as/IZF/h9S6EFjfn/7V9o0q0Ts3UguvDaUFy950gbh0GtvsHj1Eq1B84wxZswYfHx8sjVPNzc35s+fn+b5pPEoihQpwr1794iJiaFv37507NiR6dOnm9KmFY+iWrVqmY5HkZRly5Zl+VpzSNRtSo9169bRrFkzPDw8eO8987qHf//9d+bPn8/OnTupUqUKUVFRqcqsL1++nLJly3LlyhU8PT2ZMGFCilgfiSxYsAAnJyeTTElCQgIDBw5kz5491K5dm8mTJ7Ny5UqGDBlilo15jW5RZECiEGDTEhbcvetBpUp9KFSoTJrpEySB9nPGsShgNNaBnTk/bh9tmlXi1vSt3LSpiuslD3ZZFSZ4yTK89v+qnYQmW0gtHkWVKlWoU6eOKR5FImvXrjWpvsI/8SgAUzyKrODu7k6iakLJkiWZNGkSrq6uNGvWjLt37wKGN/7u3bvTuHFjGjdujLe3NwDHjh2jRYsWvPTSS7Ro0YJLly4Bhjf/nj178uabb9KuXbsMbfDw8GDOnDncvHnTbFnx6dOnM3v2bNN/sWjRogwbNixFui1btpjiZvTo0YM9e/akGnDp5s2b/PbbbwwdOtR07P79+xQpUoTatWsD0LZtWzZs2GCWffmCrGp/5NWW21pPPc6fl5qHD8vNm9/Jvn1IaOiRNNM+in4sL37WU5iCVH9vpDwIiZPooGjZ6PaJxIIEgixs3loePnyYizXQ5AZ5rfUUHh4urq6u4ujoKMOHD5f9+/ebzn311VcyZswYERE5fPiwNGrUyHSuRo0acunSJWnevLmIiLi5uYmvr69J6yk19u3bJ//6179SHH/ttdfk+PHjIiICiJeXl4iIfPLJJ/K///1PRET69u0rBw4cEBGR69evy4svvigiIqGhoRIbGysiIrt27ZJu3bqJiMiPP/4oVatWNek8iYi4urqmatfff/8tDg4OIiIyceJEmTNnjulciRIlkqX98ccfZcSIESIiUrZsWbP+k87OznLjxg3Tfq1atVLoZImIdO/eXU6cOJHse0pISJDq1aubvp/Ro0eLi4tLhmVmlezWetItinQQEQ6GhvKytTWBgUsoUcKVUqWapJr2Vsh9anz2On9ZrqPJw9lcXrCAwB+P8ka1N+juM4sFxctxbcuvfHBoD6WfiLet0TwtifEoli5dSsWKFendu7dJMrxPnz6sX7+ehISETMejyCqFCxc29c83bNjQ1J+/e/duRo4caYpvHRYWRnh4OKGhofTs2RMXFxfGjh2Lr6+vKa+2bdtSrlw5035a3Xqenp706tXLVOeMwrNmdrGspNJ6eDKPrVu3UqlSpRSx6ZVSeHp6MnbsWJo0aUKpUqUKVGjigmNpHuAfFcWdmBgaFAkhIsgHR8dFqd5cJ69d45XFHXlc5Dr9rNaybHgrzlR7mVrBR/HFgv8bPpGR8z43dQtoNDlBbsajyIhChQqZ/iuWlpamgduEhAQOHz5MsWLJg3GNGjWKVq1asWnTJgICAnB3dzedK2GUzckIDw8P7t69yy+//AIYovP5+fnh6OhIsWLFiImJMU07f/DggSlcrLOzMydPnqR169bp5m9ra8uNGzewtbUlLi6O0NDQZA4MDFH/vLy82LZtG1FRUYSFhfHWW2/x888/07x5cw4cOADAzp07uXz5sln1yg/oFkU6JAoBOkRuxMKiOJUr90uRZsORYzT9vhmP1T2m2O1ikv8tQupUpVHwEXaXqsCuw4eYtmiadhKaHCW341FklXbt2vHtt98msxMgNDSUqlWrAmTJWV26dIlHjx5x69YtAgICCAgIYOLEiaYQsK+99ho///wzAI8fP2bt2rW0atUKMMTHGD9+PHfu3AEMs5u+/vrrFGUkjY+xfv16WrduneLFcfr06dy8eZOAgAA8PT1p3bq1qdzEUK7R0dHMnDmT999/P9P1zCu0o0iHg6GhlLa0xPrBEipV6ouVVfIuo5mbveix1R2JLslqx510GPsedZePJYJY1g0aRa+Hd6jbLPWuKo0mO4mIiGDgwIHUrVuX+vXrc+HCBaZMmWI637NnT3x9fZMNYielVKlSTJgwweyFnnv27MHW1ta0pdZKSY2vv/6aEydOUL9+ferWrcvixYsBGD9+PBMnTuTll18mPj4+3Tzc3NxSHPPw8KBr167JjnXv3t3U/bRgwQI2btyIm5sbzZo1o2fPnrz66qsAdOrUiREjRvD666/j7OxMw4YNU526OmTIEO7fv4+DgwNz585lxowZQPpxxZMya9YsnJycqF+/Pm+++WaGLZj8hI5HkQ4ux45RibtMjuxMgwbHsLZubDo3ZMlCfrg9mqL3G/Ht4/HMXDGRzxP8UBXteOXgDqrV1jLgzxM6HoUmP5Hd8Sj0GEUaPIiNxTcyklcK/UHJki9RqpTh+41PSKDN9P/jj7hZtPDpwuydPgyL6klI4fJYzP2FPh/01UHcNRrNM4V2FGlw2LhQpk7sbqrYDUcpRURUFG5TBuFfeC1zf3DnnVteFEUYXqcF/Q5vpWxZ80UCNZr8zo4dO5gwYUKyY3Z2dmzatCmPLNLkFdpRpIF3aChWJFBX3aBSpX5cD3rASzP/jeONS6zeUIYmCfs5ZlGE2LmLGPHh4Lw2V6PJdtq3b5/jg9uagoF2FGlw4OF9HPGjxgvdOeZ3jzZLO+G2szw9LoTiRDTr67fiX4d+pZiZU/c0Go2moKJnPaVCTEICx8PDceEMZwNf4T9j2tPl22iOXPBmTRlH/D220uPMXu0kNBrNc0GOtiiUUh2ABYAlsExEZjxxvgjwE9AQuA/0FpGAnLTJHE6GhxMtFtiH3SNo0Kf8HnKTcyjqv/MlE77/PywtLTPORKPRaJ4RcsxRKKUsgYVAW+AmcFwp5SUiSXWIhwAhIuKglOoDzAR655RN5rLn7lleunyZt0b8inNcFHuKVKDWb7/yaRsdcU6j0Tx/5GTXUxPgiohcE5EYwBPo8kSaLsBK4+f1QBuVx3NLo6NiObJwM8eGD6d8XDS/dnqX1o+DsNNOQpPPsbS0xM3NzbQFBAQkix2xYsUKLCwsOHv2rOkaFxeXZHEVTp8+jVKKHTt2JMv7yXgOaTFlyhSqVq2Km5sbdevWTaa3NGjQIOzs7Ez2pbb6OTVcXV1T6FMlVaoFCAgISBZD49ixY7z66qvUqVOHF198kaFDhxIZGZki7+nTp+Pg4ECdOnVS1DmRPXv20KBBA9zc3GjZsiVXrlwBYOzYsaa61K5dmzJl0laVLujkZNdTVeBGkv2bQNO00ohInFIqFCgP3EuaSCn1LvAuQPXq1XPKXgAsrCw4WsGGr15pw9tfzefNJnVztDzNs8cYPz98IiKyNU+3kiWZ75j+Is5ixYqlEMx7MriOra0tU6dOTTOOgoeHBy1btsTDwyPLM57Gjh3LuHHj8PPzo2HDhvTo0YNChQoBhtXJPXr0MDuvixcvkpCQwJ9//smjR4/M0n26e/cuPXv2xNPTk+bNmyMibNiwgfDw8GSChxcuXMDT0xNfX19u377N66+/zuXLl1N0LQ8fPpwtW7bg5OTEokWL+PLLL1mxYgXz5s0zpfnmm284ffq02fUqaORkiyK1lsGTy8DNSYOILBWRRiLSqGLFitliXFoUsrLkyqgPmLBnO9W0k9A8Y7zxxhv4+vqa4j0kRURYv349K1asYOfOnURFRT1VWY6OjhQvXpyQkJAs57F69Wrefvtt2rVrh5eXl1nXLFy4kIEDB9K8eXPAoNzao0cPKleunCzdli1b6NOnD0WKFMHOzg4HBweOHTuWIj+llCkAUWhoaKoxZJ4mjkdBICdbFDeBakn2bYHbaaS5qZSyAkoDD3LQJrMoXebpZZY1zy8ZvfnnFI8fPzbpIKW1MM7CwoLx48czbdo0k8BdIt7e3tjZ2WFvb4+7uzvbtm2jW7duWbbn1KlTODo6UqlSJdOxTz75hC+//BKAVatWUa9ePYYOHcr7779Po0Yp1SXWrFnDrl27uHTpEt9++61ZD+Pz58+bAgylx61bt2jW7J8uZVtb21SDHS1btoxOnTpRrFgxrK2tOXLkSLLz169fx9/fv0BpN2WWnGxRHAcclVJ2SqnCQB/gyVcCLyDxF+0B7JWCJj6l0eQTEruefHx80l093a9fP44cOYK/v3+y4x4eHibRQHPiOaTFvHnzqFOnDk2bNk0mTAiGrqdEG+vVqwcYHsSpOYnjx49TsWJFatSoQZs2bTh16pSpdZLaUGZOxJdIrM+2bdu4efMm77zzDh999FGy856envTo0eOZng2ZY45CROKAkcAO4CKwVkR8lVJfKKU6G5MtB8orpa4AHwH/l1P2aDQaA1ZWVnz88cfMnDnTdCw+Pp4NGzbwxRdfULNmTUaNGsXvv/9OeHh4pvMfO3Ysly5dYs2aNQwYMCDLXVgeHh789ddf1KxZE3t7e8LCwkzhQ8uXL5+sSyu1+BIZkRhfIpGbN2+m6FYKDg7mzJkzNG1qGF7t3bs3hw4dSpYmtWBQzxo5uuBORLaJSG0RsReRqcZjk0XEy/g5SkR6ioiDiDQRkWs5aY9GozEwaNAgdu/eTXBwMGCIPOfq6sqNGzcICAjg+vXrdO/enc2bN2e5jG7dutGoUaMUXVzmkJCQwLp16zh79qwpvsSWLVtMrRx3d3d+/vlnU6tg5cqVpvgSI0eOZOXKlcnihP/888+meBOJdO7cGU9PT6Kjo/H398fPz48mTZKHBShbtiyhoaGmIEO7du1Kpsp66dIlQkJCTOMhzyp6ZbZG8xxSuHBhRo8ebQqmk1Y8h9WrVwMQGRmZLP7E3LlzzSpn8uTJzJ07l4SEhDTTDB06lCdDB/z5559UrVrVFMwI4NVXX+XChQsEBgby7rvvUqpUKVxdXXF1dSUiIoJx48YBULlyZTw9PRk3bhx16tTBycmJAwcOYG1tnawMZ2dnevXqRd26denQoQMLFy40dR916tSJ27dvY2Vlxffff0/37t1xdXVl1apVzJo1y5RHYnfds64YreNRaDTZgI5HoclPZHc8Ct2i0Gg0Gk26aPVYjUaTaaZOncq6deuSHevZsyeTJk3KI4s0OYl2FBpNNiEiz3xfdSKTJk3STiGfkhPDCbrrSaPJBooWLcr9+/dz5E+q0ZiLiHD//n2KFi2arfnqFoVGkw3Y2tpy8+ZN03RTjSavKFq0KLa2ttmap3YUGk02UKhQIezs7PLaDI0mR9BdTxqNRqNJF+0oNBqNRpMu2lFoNBqNJl0K3MpspVQwcD0XiqrAEwGUCjDPUl3g2arPs1QXeLbq8yzVBaCOiJTKyoUFbjBbRHI2cpERpdSJrC53z288S3WBZ6s+z1Jd4Nmqz7NUFzDUJ6vX6q4njUaj0aSLdhQajUajSRftKNJmaV4bkI08S3WBZ6s+z1Jd4Nmqz7NUF3iK+hS4wWyNRqPR5C66RaHRaDSadNGOQqPRaDTp8tw7CqVUB6XUJaXUFaXU/6VyvohSao3x/FGlVM3ct9I8zKjLR0qpC0qps0qpPUqpGnlhp7lkVJ8k6XoopUQplW+nMppTF6VUL+Pv46uUWp3bNmYGM+616kqpfUqp08b7rVNe2GkOSqkflFJBSqnzaZxXSqmvjXU9q5RqkNs2mosZdelvrMNZpdQhpZSrWRmLyHO7AZbAVaAWUBg4A9R9Is0HwGLj5z7Amry2+ynq0goobvw8PL/Wxdz6GNOVAv4EjgCN8trup/htHIHTQFnjfqW8tvsp67MUGG78XBcIyGu706nPq0AD4Hwa5zsBvwMKaAYczWubn6IuLZLcYx3Nrcvz3qJoAlwRkWsiEgN4Al2eSNMFWGn8vB5oo/JndJoM6yIi+0Qk0rh7BMheLeLsxZzfBuB/wFdAVG4al0nMqcswYKGIhACISFAu25gZzKmPANbGz6WB27loX6YQkT+BB+kk6QL8JAaOAGWUUja5Y13myKguInIo8R4jE8+A591RVAVuJNm/aTyWahoRiQNCgfK5Yl3mMKcuSRmC4S0pv5JhfZRSLwHVRGRrbhqWBcz5bWoDtZVS3kqpI0qpDrlmXeYxpz5TgLeUUjeBbcCo3DEtR8jsf6ugYPYzoMBJeGQzqbUMnpwvbE6a/IDZdiql3gIaAa/lqEVPR7r1UUpZAPOAQbll0FNgzm9jhaH7yR3DW94BpZSLiDzMYduygjn16QusEJE5SqnmwCpjfRJy3rxsp6A8A8xGKdUKg6NoaU76571FcROolmTflpRNZFMapZQVhmZ0es3UvMKcuqCUeh2YBHQWkehcsi0rZFSfUoALsF8pFYCh79grnw5om3ufbRGRWBHxBy5hcBz5EXPqMwRYCyAih4GiGET2CiJm/bcKCkqp+sAyoIuI3DfnmufdURwHHJVSdkqpwhgGq72eSOMFDDR+7gHsFeNIUD4jw7oYu2qWYHAS+bkPHDKoj4iEikgFEakpIjUx9Ld2FpEsC5/lIObcZ5sxTDZAKVUBQ1fUtVy10nzMqc/fQBsApZQTBkdRUOPEegEDjLOfmgGhIhKY10ZlBaVUdWAj8LaIXDb7wrwepc/rDcOMhssYZnFMMh77AsNDBww3+DrgCnAMqJXXNj9FXXYDdwEf4+aV1zY/TX2eSLuffDrryczfRgFzgQvAOaBPXtv8lPWpC3hjmBHlA7TLa5vTqYsHEAjEYmg9DAHeB95P8tssNNb1XD6/zzKqyzIgJMkz4IQ5+WoJD41Go9Gky/Pe9aTRaDSaDNCOQqPRaDTpoh2FRqPRaNJFOwqNRqPRpIt2FBqNRqNJF+0oNPkOpVS8UsonyVYznbQ101LKzGSZ+41qqGeMMhp1spDH+0qpAcbPg5RSVZKcW6aUqpvNdh5XSrmZcc0YpVTxpy1b8/yiHYUmP/JYRNySbAG5VG5/EXHFIAI5K7MXi8hiEfnJuDsIqJLk3FARuZAtVv5j5yLMs3MMoB2FJstoR6EpEBhbDgeUUqeMW4tU0jgrpY4ZWyFnlVKOxuNvJTm+RCllmUFxfwIOxmvbGGMqnDNq/RcxHp+h/ontMdt4bIpSapxSqgcGLa1fjGUWM7YEGimlhiulvkpi8yCl1DdZtPMwScTplFLfKaVOKEM8i/8aj43G4LD2KaX2GY+1U0odNn6P65RSJTMoR/Ocox2FJj9SLEm30ybjsSCgrYg0AHoDX6dy3fvAAhFxw/CgvmmUj+gNvGw8Hg/0z6D8N4FzSqmiwAqgt4jUwyDcN1wpVQ7oCjiLSH3gy6QXi8h64ASGN383EXmc5PR6oFuS/d7Amiza2QGD9Ecik0SkEVAfeE0pVV9EvsagS9RKRFoZ5UH+A7xu/C5PAB9lUI7mOed5V4/V5E8eGx+WSSkEfGvsk4/HoIX0JIeBSUopW2CjiPgppdoADYHjyhBGpBgGp5MavyilHgMBGGSx6wD+8o8mzkpgBPAthvgXy5RSvwFmy5yLSLBS6ppRM8jPWIa3Md/M2FkCQwChpNHWeiml3sXwv7bBIKNx9olrmxmPexvLKYzhe9No0kQ7Ck1BYSwGnSpXDC3hFIGKRGS1Uuoo8C9gh1JqKAadnpUiMtGMMvpLElFBpVSqcUdEJE4p1QSD6F0fYCTQOhN1WQP0Av4CNomIKMNT22w7MWgozcCgQdRNKWUHjAMai0iIUmoFBp2yJ1HALhHpmwl7Nc85uutJU1AoDQSKIZ7B2xjeppOhlKoFXDN2t3hh6ILZA/RQSlUypimnzI8V/hdQUynlYNx/G/jD2KdfWkS2YRgoTm3mUTgGKfTU2Aj8G0PMhjXGY5myU0RiMXQhNTN2W1kDj4BQpVRlDGEuU7PlCPByYp2UUsWVUqm1zjQaE9pRaAoKi4CBSqkjGLqdHqWSpjdwXinlA7yIIXzlBQwP1J1KqbPALgzdMhkiIlHAO8A6pdQ5IAFYjOGhu9WY3x8YWjtPsgJYnDiY/US+IRhUYmuIyDHjsUzbaRz7mAOME5EzGGJu+wI/YOjOSmQp8LtSap+IBGOYkeVhLOcIhu9Ko0kTrR6r0Wg0mnTRLQqNRqPRpIt2FBqNRqNJF+0oNBqNRpMu2lFoNBqNJl20o9BoNBpNumhHodFoNJp00Y5Co9FoNOny/xmDIezXjwRqAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pred = clf.predict(scaled_X_test)\n",
    "prediction[\"Logistic\"] = pred\n",
    "eval_model(Y_test, pred)\n",
    "AuROC(prediction, Y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\USER\\Anaconda3\\lib\\site-packages\\sklearn\\svm\\base.py:196: FutureWarning: The default value of gamma will change from 'auto' to 'scale' in version 0.22 to account better for unscaled features. Set gamma explicitly to 'auto' or 'scale' to avoid this warning.\n",
      "  \"avoid this warning.\", FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.40      0.75      0.52        68\n",
      "           1       0.96      0.83      0.89       460\n",
      "\n",
      "   micro avg       0.82      0.82      0.82       528\n",
      "   macro avg       0.68      0.79      0.71       528\n",
      "weighted avg       0.89      0.82      0.84       528\n",
      "\n",
      "[[ 51  17]\n",
      " [ 76 384]]\n",
      "accuracy_score :  0.8238636363636364\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "from sklearn.svm import SVC\n",
    "clf = SVC(class_weight = {0:7, 1:1}, kernel = \"rbf\")\n",
    "clf.fit(scaled_X_train,Y_train)\n",
    "pred = clf.predict(scaled_X_test)\n",
    "eval_model(Y_test, pred)\n",
    "prediction[\"SVM\"] = pred\n",
    "AuROC(prediction, Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.42      0.68      0.52        68\n",
      "           1       0.95      0.86      0.90       460\n",
      "\n",
      "   micro avg       0.84      0.84      0.84       528\n",
      "   macro avg       0.68      0.77      0.71       528\n",
      "weighted avg       0.88      0.84      0.85       528\n",
      "\n",
      "[[ 46  22]\n",
      " [ 64 396]]\n",
      "accuracy_score :  0.8371212121212122\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\USER\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\stochastic_gradient.py:166: FutureWarning: max_iter and tol parameters have been added in SGDClassifier in 0.19. If both are left unset, they default to max_iter=5 and tol=None. If tol is not None, max_iter defaults to max_iter=1000. From 0.21, default max_iter will be 1000, and default tol will be 1e-3.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.linear_model import SGDClassifier\n",
    "clf = SGDClassifier(class_weight = {0:7, 1:1})\n",
    "clf.fit(scaled_X_train,Y_train)\n",
    "pred = clf.predict(scaled_X_test)\n",
    "eval_model(Y_test, pred)\n",
    "prediction[\"SVM_Linear\"] = pred\n",
    "AuROC(prediction, Y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Trying for any improvement possible with this data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'bootstrap': [True, False],\n",
      " 'max_depth': [10, 25, 41, 56, 72, 87, 103, 118, 134, 150, None],\n",
      " 'max_features': ['auto', 'sqrt'],\n",
      " 'min_samples_leaf': [3, 6, 9],\n",
      " 'min_samples_split': [3, 8, 13, 20],\n",
      " 'n_estimators': [50, 90, 131, 172, 213, 254, 295, 336, 377, 418, 459, 500]}\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "import pprint\n",
    "# Number of trees in random forest\n",
    "n_estimators = [int(x) for x in np.linspace(start = 50, stop = 500, num = 12)]\n",
    "# Number of features to consider at every split\n",
    "max_features = ['auto', 'sqrt']\n",
    "# Maximum number of levels in tree\n",
    "max_depth = [int(x) for x in np.linspace(10, 150, num = 10)]\n",
    "max_depth.append(None)\n",
    "# Minimum number of samples required to split a node\n",
    "min_samples_split = [3, 8, 13, 20]\n",
    "# Minimum number of samples required at each leaf node\n",
    "min_samples_leaf = [3, 6, 9]\n",
    "# Method of selecting samples for training each tree\n",
    "bootstrap = [True, False]\n",
    "# Create the random grid\n",
    "random_grid = {'n_estimators': n_estimators,\n",
    "               'max_features': max_features,\n",
    "               'max_depth': max_depth,\n",
    "               'min_samples_split': min_samples_split,\n",
    "               'min_samples_leaf': min_samples_leaf,\n",
    "               'bootstrap': bootstrap}\n",
    "pprint.pprint(random_grid)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Number of settings : 2*11*2*3*4*12 = 6336 combinations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 100 candidates, totalling 500 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  25 tasks      | elapsed:    4.5s\n",
      "[Parallel(n_jobs=-1)]: Done 146 tasks      | elapsed:   18.2s\n",
      "[Parallel(n_jobs=-1)]: Done 349 tasks      | elapsed:   43.1s\n",
      "[Parallel(n_jobs=-1)]: Done 500 out of 500 | elapsed:  1.0min finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5, error_score='raise-deprecating',\n",
       "          estimator=RandomForestClassifier(bootstrap=True, class_weight={0: 7, 1: 1},\n",
       "            criterion='gini', max_depth=None, max_features='auto',\n",
       "            max_leaf_nodes=None, min_impurity_decrease=0.0,\n",
       "            min_impurity_split=None, min_samples_leaf=1,\n",
       "            min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "            n_estimators='warn', n_jobs=None, oob_score=False,\n",
       "            random_state=None, verbose=0, warm_start=False),\n",
       "          fit_params=None, iid='warn', n_iter=100, n_jobs=-1,\n",
       "          param_distributions={'n_estimators': [50, 90, 131, 172, 213, 254, 295, 336, 377, 418, 459, 500], 'max_features': ['auto', 'sqrt'], 'max_depth': [10, 25, 41, 56, 72, 87, 103, 118, 134, 150, None], 'min_samples_split': [3, 8, 13, 20], 'min_samples_leaf': [3, 6, 9], 'bootstrap': [True, False]},\n",
       "          pre_dispatch='2*n_jobs', random_state=42, refit=True,\n",
       "          return_train_score='warn', scoring=None, verbose=2)"
      ]
     },
     "execution_count": 155,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Use the random grid to search for best hyperparameters\n",
    "# First create the base model to tune\n",
    "rf = RandomForestClassifier(class_weight={0:7,1:1})\n",
    "# Random search of parameters, using 3 fold cross validation, \n",
    "# search across 100 different combinations, and use all available cores\n",
    "rf_random = RandomizedSearchCV(estimator = rf, param_distributions = random_grid, n_iter = 100, cv = 5, \\\n",
    "                               verbose=2, random_state=42, n_jobs = -1,)\n",
    "# Fit the random search model\n",
    "rf_random.fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'n_estimators': 254,\n",
       " 'min_samples_split': 3,\n",
       " 'min_samples_leaf': 3,\n",
       " 'max_features': 'sqrt',\n",
       " 'max_depth': 134,\n",
       " 'bootstrap': False}"
      ]
     },
     "execution_count": 156,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf_random.best_params_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">We have narrowed down the combination, before proceeding with more detailed combination lets just check the performance once"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.45      0.77      0.57        65\n",
      "           1       0.96      0.87      0.91       463\n",
      "\n",
      "   micro avg       0.86      0.86      0.86       528\n",
      "   macro avg       0.71      0.82      0.74       528\n",
      "weighted avg       0.90      0.86      0.87       528\n",
      "\n",
      "[[ 50  15]\n",
      " [ 61 402]]\n",
      "accuracy_score :  0.8560606060606061\n"
     ]
    }
   ],
   "source": [
    "rf = rf_random.best_estimator_\n",
    "pred = rf.predict(X_test)\n",
    "eval_model(Y_test, pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is significant improvement in recall. So we can keep this model for further testing purposes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "pickle.dump(rf_random,open(\"final_random_forest.sav\", \"wb\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle.dump(clf, open(\"SVM_LINEAR_MODEL.sav\",\"wb\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.36      0.75      0.48        69\n",
      "           1       0.96      0.80      0.87       459\n",
      "\n",
      "   micro avg       0.79      0.79      0.79       528\n",
      "   macro avg       0.66      0.77      0.68       528\n",
      "weighted avg       0.88      0.79      0.82       528\n",
      "\n",
      "[[ 52  17]\n",
      " [ 94 365]]\n",
      "accuracy_score :  0.7897727272727273\n"
     ]
    }
   ],
   "source": [
    "model = pickle.load(open(\"SVM_LINEAR_MODEL.sav\",\"rb\"))\n",
    "result = model.predict(x_test)\n",
    "eval_model(y_test, result)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle.dump(clf, open(\"Logistic_final.sav\", \"wb\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.51      0.87      0.64        68\n",
      "           1       0.98      0.88      0.93       460\n",
      "\n",
      "   micro avg       0.88      0.88      0.88       528\n",
      "   macro avg       0.75      0.87      0.79       528\n",
      "weighted avg       0.92      0.88      0.89       528\n",
      "\n",
      "[[ 59   9]\n",
      " [ 56 404]]\n",
      "accuracy_score :  0.8768939393939394\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAEWCAYAAAB42tAoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzsnXd4VFX6xz8nk96pofdACESQooJYWHUFC1h+Ai66gIVVUBTBgqigu2LHgiyugoWOSFVRUQQ1FDHUJBAIIQECZNIndTLt/P64kxBCMpmUSeN8nmeezL333HvfuZl5v6e85z1CSolCoVAoFBXhVt8GKBQKhaJho4RCoVAoFA5RQqFQKBQKhyihUCgUCoVDlFAoFAqFwiFKKBQKhULhECUUTRghxFwhxHIXXj9WCHGj/b0QQnwhhMgSQuwVQlwnhDjmqns3ZYQQnYQQeUIIXX3bUhYhxA9CiAkOjn8phPhPXdqkcD1KKBo5Qoh/CCGi7I7lvP2HPKwu7i2l7COl3GHfHAbcAnSQUl4lpfxDStmrLuxoakgpT0sp/aWU1vq2pSxSypFSyq8AhBAThRCR1b2WEKKLEELav7t5QogkIcQL5ZSbKISIFkIUCCFShBCLhBDBZcr0FEKsFUKkCyEMQojDQohnGqLYNkaUUDRihBDPAB8A84AQoBPwX2B0PZjTGUiSUubX9EJCCPdasKfBopzXJQRLKf2B/wNeFkLcUnxACDEDeAt4FggCrkH7rv0shPC0l+kO/AmcASKklEHAfcAgIKAuP0iTRUqpXo3whfajyQPuc1BmLrC81PZaIAUwAL8DfUoduw04AuQCZ4GZ9v0tge+AbCAT+ANwsx9LAm4GHgaMgNVu06vAjUByqeu3A9YBaUAiMK2Mnd8Ay4Ec4BHgKiDKvq0H5jv4nKOBg/ayCcCIUvfcbLf7BPBomXuutd8zF4gGegKzgFQ0p/P3UuV3AG8Ae+3PbxPQ3Mln+yWwCNgC5Nuf2e3AAbvNZ4C5pcp3ASTgbt+eCJy025kIjLfvdwNeAk7ZbV4KBJW5xgTgNJAOzK7g+XW1/3+L/6+LgdRSx5cDT5d6Do8Avcv8z7NLfdaFwPd2e/8Euldw34s+p33fXuBZ+/tA+7XHlDnP3/55Hypl3/f1/Ztsyq96N0C9qvmPgxGApfSPrJwyc7lYKB5Cq2F5obVEDpY6dh64zv6+GTDA/v4N4BPAw/66DhD2Y0nAzfb3E4HIUte7EbtQ2B3aPuAVwBPoZnd8t5ay0wzcZS/rA+wGHrQf9weuqeAzXoXmnG+xn9seCLMf+w2theUN9EcTqZtK3dMI3Aq4oznZRGC2/XM+CiSWus8ONAHtC/ihiZ6zz/ZLu43X2m30tj+fCPv2FWhieJe9fBfsDtR+rxygl/1YW+wiZL/nCfvz9AfWA8vKXOMz+/PsBxQBvSt4jqeBgfb3x+z/n96ljl1Z6jk8Ut7/vNRnzbT/X9yBFcDqCu5Z8jnt29cABcDdlX3Hga+AVfb3KcCk+v5NNuWX6npqvLQA0qWUFmdPkFJ+LqXMlVIWoTnKfkKIIPthMxAuhAiUUmZJKfeX2t8W6CylNEtt7KGqCcIGA62klK9JKU1SypNoDmxcqTK7pZQbpZQ2KWWh/b49hBAtpZR5Uso9FVz7YeBzKeXP9nPPSinjhBAd0cZNnpdSGqWUB9Fqyg+WOvcPKeVP9me4FmgFvCmlNAOrgS5l+sKXSSljpNa99jIwprgbqZJnC7BJSrnTbqNRSrlDShlt3z4MrAJuqOAz2oC+QggfKeV5KWWsff94tJbWSSllHlpraFyZrrtXpZSFUspDwCE0wSiP34AbhBBt7Nvf2Le7otXsD1VwXnmsl1LutT/XFWgi7Yh0IUQhWuXgv8BG+/6WVPwdP28/Dtpv4XwV7FNUESUUjZcMoKWz/flCCJ0Q4k0hRIIQIgetNQAXfmz3onU/nRJC/CaEGGLf/w5arXWrEOJkeYONTtAZaCeEyC5+AS+ijasUc6bMOQ+jdQXFCSH+EkLcUcG1O6J1N5WlHZAppcwtte8UWoujGH2p94VoTslaahu0mnp5Np5Ca3m0dOLZlj0XIcTVQojtQog0IYQBeKxMeQDsojTWfvy8EOJ7IURYqc94qoxN7lz8XFNKvS8o83lK8xtaK+d6tK6zHWjCdQOaoNoqOK88nL1nMS3tZWbabfCw70+n4u94W/tx0H4Lbatgn6KKKKFovOxG6zq5y8ny/0Dry78ZbXyji32/AJBS/iWlHA20RqvRfW3fnyulnCGl7AbcCTwjhLipiraeQevGCS71CpBS3laqzEWtFCllvJTyfrs9bwHfCCH8Krh293L2nwOaCyFKD2Z2Qus+qi4dy1zLjOasHD5bO2VbYSvRxk86Sm3w9ZMy5S+cqLV6bkFzhnForTHQPmPnMjZZuFgAneU3tG7FG+3vI9G6ym6wb5drWjXuU/6FpLRKKd9D+05Pse/ejdZddk/psvbvwUhgm33XL2gVHYWLUELRSJFSGtD6/BcKIe4SQvgKITyEECOFEG+Xc0oA2o8uA/BFi5QCQAjhKYQYL4QIsne75KANUiKEuEMI0UMIIUrtr2rY5l4gRwjxvBDCx14D7yuEGFzRCUKIB4QQrew12Wz77vLuuwSYJIS4SQjhJoRoL4QIk1KeAXYBbwghvIUQV6C1UlZU0fbSPCCECBdC+AKvAd/YWyAVPlsHBKC1eIxCiKvQxOYShBAhQohRdudYhDa4W/wcVgHThRBdhRD+9vuuqUp3ZDFSyni0VtQDwO9SyuIggnupWCj0QIfi6KNa4k3gOSGEt/07/iqwQAgxwv797oLWTZgMLLOfMwcYKoR4p7jrzP6dXV42jFZRPZRQNGKklPOBZ9AiX9LQatdPcKGPtzRL0bomzqJFN5Xt838QSLJ3nTyG5jAAQtFqbHnY+5DlhbkTztppRWuN9EcbME5HGy8IcnDaCCBWCJEHfAiMk1Iay7n2XmAS8D7agPFvXKhl349Wuz8HbADmSCl/rortZViGNlibgjYgPc2+v7JnWx5TgNeEELlogv91BeXcgBlonyETrYZfXOP+3G7T72jP1Qg8WZUPVIbfgAwp5elS2wItOqs8fgVigRQhRHoFZarK90AWWjABUsq30bop30WrqBSHwd5kHw9CSpkADEH7X8fau/LWoUXN5aKoMcXRKwqFwgFCiB1oUU6L69sWhaKuUS0KhUKhUDjEZUIhhPhcCJEqhIip4Ph4+zT7w0KIXUKIisL2FAqFQlGPuKzrSQhxPVq/9lIpZd9yjg8Fjkops4QQI9Fmpl7tEmMUCoVCUW1cllNHSvm7PUKhouO7Sm3uATq4yhaFQqFQVJ+GknztYeCHig4KISYDkwH8/PwGhoWFVVRUoVAoFOWwb9++dCllq+qcW+9CIYQYjiYUFabGllJ+CnwKMGjQIBkVFVVH1ikUCkXTQAhxqvJS5VOvQmGfBLUYGCmlzKhPWxQKhUJRPvUWHiuE6ISW7fJBKeXx+rJDoVAoFI5xWYtCCLEKLW9MSyFEMto0ew8AKeUnaLNRWwD/1bJDYJFSDnKVPQqFQqGoHq6Merq/kuOPoC2AolAoFIoGjJqZrVAoFAqHKKFQKBQKhUOUUCgUCoXCIUooFAqFQuEQJRQKhUKhcIgSCoVCoVA4RAmFQqFQKByihEKhUCgUDlFCoVAoFAqHKKFQKBQKhUOUUCgUCoXCIUooFAqFQuEQJRQKhUKhcIgSCoVCoVA4RAmFQqFQKByihEKhUCgUDlFCoVAoFAqHKKFQKBQKhUOUUCgUCoXCIUooFAqFQuEQJRQKhUKhcIgSCoVCoVA4RAmFQqFQKByihEKhUCgUDlFCoVAoFAqHuEwohBCfCyFShRAxFRwXQoiPhBAnhBCHhRADXGWLQqFQKKqPK1sUXwIjHBwfCYTaX5OBRS60RaFQKKqExWLBVFRU32Y0CNxddWEp5e9CiC4OiowGlkopJbBHCBEshGgrpTzvKpsUCkXTRUowmSA3F/LytFfp98XbWflWzpuNnLGmkCJSyXFPQ4g0zKKQQncdBd6ePLTrN2JDe9DZPYhPZj9V3x+t3nGZUDhBe+BMqe1k+75LhEIIMRmt1UGnTp3qxDiFQuFazObKnbqzx3JNVvI8TFgDzdAqB9qmQUgWtMiHFiYIlHgEuCNDfLB4epWxJFB75eQQHhPDug8/5erTp1jRfxCn/jWtPh5Ng6M+hUKUs0+WV1BK+SnwKcCgQYPKLaNQKFyHxXKxo3bkyJ11+CZTJTf1sEJzMzQz4dXWhGebItzaZkGIAVuzAqwBZsx+Equ3BzYPz3Iu4IZXgQdemSZERg6mE6kUZp6FrCzIysI900C7Ih3dvL3o164VD+dl0ffXXyAoCFauZPy4cSDKc1OXH/UpFMlAx1LbHYBz9WSLQnFZs2sXfPABZGeX7+SNRuev5eMD/v4QEKD99ffXfG+HDuATaMOthQnZ3IQMMmEOMFHkayLPu4As91yydIUYhJk8N0GRuOCeiuwvAMy5uBcZCCow0inVSssMiedZI9bkHHLO6UnPPE5K1lHM2akUmc2YEXQIaE2vjp0I79uLAdcN5aqbhxMa2hOdTqddU0q44w647z7tQbRqVVuPtklQn0KxGXhCCLEauBowqPEJhaLuWbMG/vlPCA6Gbt00x96mzQUnX9rhl90OCABPPxuF3ibyPU3kuJtIs5jQm83oTSbOm4pILizgmMlIutlKvqyghm7JA1MmFGaBKQtMmfhYCgkr8qNPth+hqb60SnCnIDaXc6ezSTSd4yQnSSSROHJKLtPSL4iwTl24bfj1DLzhWgZcfQ3h4eH4+Phces+CAnjtNXjsMejSBdavB6+y3VIKcKFQCCFWATcCLYUQycAcwANASvkJsAW4DTgBFACTXGWLQqG4FCnhrbdg1iwYNgw2boQWLbRjRTYbqSYTKSYTepPm+FNMJhLt2ykmEymmIlIKisjJLb83WFgLkEUZmgCYssCs/RXmbJrpJCEeXnT09qOvJZjwjGBCzvvhd9oHW2IIpxP8SUjNJFGeIoG9/MxJ9OhLru3n5U2vzl25u/8IBgwdwhX9+hMREUGL4g9QGTt2wCOPQEKC1tR54gklEg5wZdTT/ZUcl8BUV91foVBcjMlm05y+ycTZQjPvfWHijxgTPT8yEXKDiXtPm9AnaIKQbbGUew13WxFuZgO2onQsRakltX9NCLLwkYW09fSmo48/nQLa0D6gPe1btqe9Z19C0j3xSzTBSSPGY2aKTug4l2jmZL6ePSRqLQRxkjPyDBas2v3cdPTo3JnrrxxK/4GDiIiIICIigs6dOyOqM35gMMBzz8Gnn0L37vDrrzB8eE0e62VBfXY9KRSKGmKy1/yLa/z6UjV+fZn9WWWd/zXaK0maOZeWj5spG0tRGsaC81CUDuZSImDKIsTTgw7+rWkf2F4TgJD2tA+8kvYB7ekQ2IF2Ae3wzLKRHX0cQ8x5CiINGI9LzCd9yT7rxl6b1lV0kpMkuh8n0XaaAtuFwY+O7doR0b8fY68YVyIIvXr1wtOzvIHqajJvHixeDDNnwquvgq9v7V27CSO0in3jYdCgQTIqKqq+zVAoXIbZZiPVCcevN5nIrKDmH6DTEeLpSRtPT0I8PPCw5rHt2FrSso+XagVkgjkLX5275viLBaD0e/vfNv5t8NB5AGA1Wck5mkR2dBJ5R9IoOFaI6YQH1qQginI8SSKJRBJJdDtBotcxTtrOkFFkKLGtWXAwEVdcUSIGERER9O3bl8DAQNc80LQ0SE+H3r21FsXx4zB4sGvu1YARQuyTUg6qzrmqRaFQ1AHFzr8yx5/iwPn763Qljr+3ry/Dg4MJ8fS8SBCKt32Lo3mA7YnbuXvN3XjgQ0DkFMyZHXhlentG3aAJQZBXULndOMbUHLIOxZMTe4DoowaM8VbMCb7Is82xWgXn0AaUk3yOkugdT6JMJlmkYbNXPr08vAgPC+f2iNEXiULbtm2r121UVaSEVatg2jTo3BmiorTwq8tQJGqKEgqFopqYbTbSyqv52wWh9P4MB86/2MGH+fpyg935l3b8bcpx/s6yMnolEzdOpI1nDzI++oEWus58/z1ERGjHrSYrhpjTZEefJDc2jcLjhRSdcMOaGAyGACSSTMyc1CWTFBRDknciic3PkpidgtGsTYQQRkH39t0ZGHEtE0sJQvfu3XF3rycXk5wMjz8O330HV10FS5aoORE1QAmFQlEKS3k1/3Icf4oD5+/n5lbi3HuVcv4hHh4l+4tfftVw/s4gpeTNyDd58dcXCfW4gczXV3B/eBpTR0VjWWhgd7wFc4IPtuTmYHUHBAX4kxhwkqRmRzjV6gyJwedJyDhHVl4uWIFMCAkJIaJfBCMi7i0RhPDwcHwbUl//gQNwww3aLMH587UWhYue8+WCGqNQNHks9pp/ZY5fbzaTYTaXmx7At5TzL3H2ZRx/Gxc7/8qwmW3kHj9H5uET/L7tJ6wnLITou+N9tj263ICSchb3As6EHOJ08AlOeZ4l0ZRKQvp5zuhTSsr4+fnRt2/fS8YRWjXkiWhmM3h4aH+ffhpmzNAmhiiAmo1RKKFQNEosNhvpxd0+Dhy/3mQi3YHzd8bxh3h44F9fXSjlUJRRiCH6BIboZPLjsig8ZsF80hvbmWZg8SgpZwvK5HTzeE76JZHdUk9SURoJ6SnEJ53CbDYDoNPp6NWr1yWC0KVLF9zcGslyNRaLNpt60SJtHKJZs/q2qEGiBrMVTQKrlKQ54fhTHDh/n1I1/+7e3lwbGFiu42/j6dmgnH9ZbBYbefGpZEcnkBurp+BYPkXxAktiIGQVRwf5gLs7dNBT2PE0ib3/4Jfc3RzKPkn6+UBy0rPAcGHWcseOHYmIiODOe+69KPzUqzFPNIuOhocfhr/+glGjtNaEotZpuL8URZPAKuWFmn8lET/pZjO2cq7hU6rm383bmyGBgRf39ZdqCfjrdHUTUVNLmLOKyD6cgCH2DHlHsjAeN2FKsLcOzMWtg+YQ7IZb53S4IZ3zwWkke2aTVJjNiRQ9MbFHOffHhTRpwt0PrJ0YMeJ2Ro/WBKFPnz4EBwfXz4d0BVarln5j3jytBbFmjZanqRH97xsTSigUVabY+TsT6lmR8/curvl7eNDV7vzLC/Vs0widf1mkVZKfkEHW4RPkxqZQEJeH8YTAejIAmVncOvACXUvokIJ71ww8b0whs2UByZ55nCrM4VjSOWJiYojfHI/Npj1RT09PwsPD6XtNXzILM3EP8Ub+vhyfnNv49lvBNdfU32d2OW5uWjfTuHFat5OzqTsU1UIJhQLQnH9G2S6fCloCaQ6cf7GD7+LtzdV2519eqGdAI3f+5WE2mDBEJ2GIPk3e0QwKj5swJXhhOx0MpuLZxcEQKHDrkobnDVl4hbphbOfGWQ8jSbl5HD12kujoaGJXxVJYWAiAEIJu3boRERHBmDFjSgaZQ0NDWXt0LRM3TaSlWzfSPthCt+Zd2fJnEx3Dzc/XZlM//jh07aqS+NUhSiiaMDa786/M8evNZlJNpnKdv5cQJc69s7c3VxV3+5Rx/CGengQ2QedfFmmVFCRm21sH58mPy6UoHiyJ/sj0IHspd3BrCe1TcO+Wjvf1enzDfHHrEsA5DysnUw3Exh4lOjqa6M+jSU9PL7l+69atiYiI4F//+lfJwHKfPn3w8/O72A4peXvn27yw7QU6cx2n5m3khquas349NG9ehw+krti2DR59FBITtUyvU6YokahDlFA0Mmylav6V5fdJM5nsqdUuxkuIEufeydubwYGBFUb8XA7OvzzMOWZyYk5jiDlF3pEMCo8XYTrhifVU6dZBIASA6JyKx7UZeIe64x8ehE+vNqR7unPshI6YmHOaICyKJjExseT6vr6+9O3bl9GjR18Uhtq6detKbbPYLEz7YRqLohbRNX8cie9/wfix3ixZ0gR9Z3Y2PPuslp8pNBR++w2uv76+rbrsUELRgPk5M5MVev1FgpBagfP3tDv/Np6edPDyYqC/f4WhnkHu7pel8y+LtEkKknLIPpxATuw5CuJyKYqXmE/6IdOKWwdu4NYc2qWg65qO77AUfHr6ENCnFYF9umAQ7Tl61F0Tg+hoordGExcXh8m+fJtOp6Nnz54MHjyYSZMmlQhC165dqxV+mm/KZ9y6cXx3/Ds6n3mexM/n8fJLbrz6ahMdx33jDfjiC3j+eZgzR1sVSVHnqHkUDZgro6I4UVhImK+vwxj/Np6eyvk7wJJrIedIMobDp8g9mk7hMSOmBC2JHUWlquD+uYjOejy6FeDd0w2/sCAC+7YnuE9PCs2+xMbGXhCE6GhiYmIwGC4ku+vQoUNJd1GxIISFheHt7V0rnyMlL4U7V93J/vP7Cdm3gLQtU/j0U5jU1FZySU2FjAwtiV9ODsTHw8CB9W1Vo0fNo2iC5FgsHM7L4+XOnZnbtWt9m9PgkTZJ4ek8sqMTyIk5R35cDkXHbZgTfZH6UmGhbsHQVo+uaxo+15zHp5c3Ab1bEnRFFwI7DcJicScuLu6CIKxbSXR0NGfPni25RFBQEBEREdx///0XTVJr5sKJXnHpcYxcMZKU3FQCt2wk/+id/Pgj3HSTy25Z90gJK1bAU09p4xBRURAYqESiAaCEooGyJycHG3BtUFClZS83zOYMCgqOkZ8fR8obVvJ/CdJaB8bi1oEv+NkQnVNwH5yGd+gZ/HoHENinHcF9euITOBwpITEx8YIgrFxDdHQ08fHxWK1a556npye9e/fmxhtvvGjmcocOHeq09fbHqT8YvXo00uIBX+4g0DyYLbugT586M8H1nD6tLUn6ww8wZIhK4tfAUELRQNlpMOAGXOOqHP0NHJvNgtGYREFBXMmrsPAYBQVxmM32KKFNo+Dj6bhdGY/3fefw6elFQHgLgiI6E9jlSjw8NJFNTU0lOjqaH/dGE71EE4TY2FgKCgpK7tetWzf69u3LvfdemLUcGhqKh4dHeebVGWti1vDPjf+kGV1JfX8LA7p249tvoW3bejWrdtm/X0viZ7PBhx/C1KkqiV8DQwlFAyXSYKCfvz8BDTjNRG1gsRgoKDhWShCO2UUhHikvpGPw8GiNr28vWra8G1/fMERCbxIW+dDstmZEfPsIwk2Qn59PbGws23+PJnrhqpLWQlpaWsl1WrZsSUREBI888kiJIPTp0wd/f//6+PgVIqXkvd3v8ezPz9LeMoyz723kzptbsGoVlImUbbyYTODpqeU8nzgRnnlGmx+haHA0bS/USDHbbPyZk8OkJlJtlNKG0Xi63NaByXQhY6kQ7vj49MDHpxctWtyJr28Yvr698PXthYfHhckBxiwjGyduJN43isKwQl65Zw7R0Vr4aXFwho+PD3369OGOO+64qNsoJCSkzj9/VbHarDz141Ms/Gsh7bPHcPbjr3jycW/ef7+JVLSL039/8gns26el4FiwoL6tUjhACUUD5FBeHvk2G8Ma2fiExZJHYeHxcloHx7GVWhvZ3b0Zvr5hNG8+spQYhOHt3Q03twtdPVJKzp49y65df14UbXQk+ghmm9bacPvAjdDQUAYMGMCECRMuCj/VNUKvmm/K5/519/Pt8W9pc3ImZ5e9xQfvu/HUU/VtWS1x6BA89JDW3XTXXSqJXyNBCUUDZGeOlvHz2gY4PiGlpKjo7CUtg4KCOIqKkkuVdMPbuyu+vmE0a3bLRYLg4dHyksHg7Oxs9u//85Lw0+zs7JIy7dq1o2fzntxju4fB9wzmby/9jd69e9da+Gl9o8/Tc+eqO9l3fh8t/lyAYccTbFiv+dNGj9WqzYN46y1t6vjatXDvvWrAupGghKIBstNgoJOXFx3q0QFarYUUFsZf0jooKDiGzZZfUk6nC8TXtxfBwcMvEgMfnx64uV06TbioqIjDhw9fJAjR0dEkJ18QmcDAQPr27cvYsWMvCj/1Ou/FvsH7CLwpkH5f90Pomo6TOZZ+jJErRnIuJwWfTevRJY9mxw5tFc8mgZub1poYP17rdmqSeUaaLkooGhhSSiINBobXQUpoKSUmk77c1oHReApKVnwQeHl1wtc3jLZtr7tIEDw92zgMFd22bRu7d+8uEYTjx4+XhJ96eHgQFhbG9ddff9E4QseOHS+5prXAyr4x+9AF6Oi9vHeTEomdp3cyavUozEU6LEu20833ar7f0wTGdfPytFbE1KlalsJ167TBa0Wjw6VCIYQYAXwI6IDFUso3yxzvBHwFBNvLvCCl3OJKmxo6SUYj502mWp0/YbMVUViYUE7rIA6r9cLCNm5uvvj69iIwcAht2kwq1ToIRaer+prIK1euZPz48QB06dKFiIgI7r777pIWQq9evZwOP42fFk/B0QKu2HoFXm2aTkKjtbFreXDDg/hbO5H74Q/c2K8769c3gUXafv4ZJk+GpCTo0UPL+KpEotHilFAIITyBTlLKE85eWAihAxYCtwDJwF9CiM1SyiOlir0EfC2lXCSECAe2AF2cvUdTJNKeEqImA9k2m5nTp98iN/dPe0vhJJTKDevp2R5f3zBCQh4sEQNf3zC8vNojRO0sf2kwGJgxYwaDBg1i27ZtBNZgvEW/Qk/KkhQ6ze5E85ubRpeFlJL5u+cz8+eZtDYOJfWjTTx4b0sWL27k/jQrS1ur+osvoFcv+OMPGDasvq1S1JBKhUIIcTswH/AEugoh+gNzpJR3V3LqVcAJKeVJ+3VWA6OB0kIhgWIPEgSc4zJnp8FAoE5HnxoEy6emriEp6WV8fcPx97+S1q3vL9U66Im7e0AtWlw+c+bMQa/Xs3nz5hqJRMHxAo4/dpygYUF0mdul9gysR6w2K9N/ms6CvQtolXYvqf9bxtyXfHjllSYwtvvmm7B0KcyaBa+8Ak0k0OByx5kWxWvA1cB2ACnlQSFEDyfOaw+cKbWdbL9OaeYCW4UQTwJ+wM3lXUgIMRmYDNCpUycnbt142ZmTw5DAQHQ18Bh6/XK8vDozeHB0rbUQqsKhQ4dYsGABkydPZvDgwdW+jtVo5cjYIwgvQe9VvXFzr/vPUtsUmAsYv348G+M20vzYM2R98w5fLnZjwoT6tqwG6PVaEr/wcJg9W1t17sqqUxIWAAAgAElEQVQr69sqRS3izC/PLKXMLrPPmZSz5Xm6sufdD3wppewA3AYsE+V4Ninlp1LKQVLKQa1atXLi1o2TLLOZmPz8GnU7FRWdJyvrZ0JCHqgXkbDZbEydOpXmzZszb968Gl0rYWYCeQfz6P1Vb7w7NP6aaVp+Gn/76m9sittEQOSHWLe8x9afGrFISAlffaVleX3wQW07MFCJRBPEmRbFUSHEGMBNCNEVeArY48R5yUDHUtsduLRr6WFgBICUcrcQwhtoCaQ6cf0mx+7i+RM1EIrU1NWAjZCQ8bVkVdVYunQpO3fuZMmSJTSvQQhk2ro0zi08R4cZHWhxe+NfD/l4xnFuW3EbZ7LP4rFhHc2z72bLLq0S3ihJSoJ//Qu2boVrr9UWFmr0/WaKinCmyvkEMBBtNHQ9YEQTi8r4CwgVQnS1D4aPAzaXKXMauAlACNEb8AbSuEyJNBhwF4KrajLwq1+Ov/9A/Px616JlzpGVlcVzzz3HkCFDmDhxYrWvU5hYSNzDcQRcFUC3eY1/8eddZ3YxdMlQUrINmD7bTj/Pu9mzpxGLxL590Lcv7NoFH38Mv/8OYWH1bZXChTjTorhVSvk88HzxDiHEPWiiUSFSSosQ4gngJ7TQ18+llLFCiNeAKCnlZmAG8JkQYjpat9RE2dhWUqpFdhoMXOnvj181U0/k5x8hL28/3bu/X8uWOcfs2bPJyMhg69at1Vq9DcBmsnFknBbvEL46HDfPxj0use7IOh7Y8ABeRR3IX/gDo6/rwcqV4Fv1aOP6p6hIW2u1Xz945BGYPh06d65vqxR1gZTS4QvYX86+fZWd56rXwIEDZVOkyGqV3r/9Jp+Oj6/2NRISZsnt292k0Xi+Fi1zjqioKCmEkE8++WSNrhM/I15uZ7tM/Sa1liyrP97f/b4Uc4VsNnOIxDdNPv20lBZLfVtVDUwmKV9/XcrOnaXMyKhvaxTVBK2CXi2/W2GLQghxK9r4QXshxPxShwIpHZSvqBX25+ZirEEiQClt6PUraN7873h5tall6xxjs9mYMmUKrVu35t///ne1r5P+XTrJ7yXTbmo7Wt3beIMWrDYrM7bO4MM/PyT4/N1kf7GCj9714ckn69uyanDggJbE7+BB+L//09aMUFx2OOp6SgVi0MYkYkvtzwVecKVRlyM1TQRoMERSVHSabt1qFmlUHRYvXszevXtZtmwZQdUUOmOykbgJcfj396f7u91r2cK6o9BcyPj149kQt4GA2Kco2vIeG9fqGDWqvi2rIhaLNg/i7behVSst/cY999S3VYp6okKhkFIeAA4IIVZIKY0VlVPUDpEGA929vWnjVb30FHr9ctzc/GjZsm5TjaanpzNr1iyuv/76knQdVcVmsXH0/qNIkyR8TTg678aXHhy08NdRq0fxZ/KfeO94H9+jT/PdDhhUreXs6xmdDmJi4J//hPfeawI5RRQ1wZnB7PZCiNeBcLSoJACklD1dZtVlhpSSnQYDI6sZTmq1GklN/ZpWre5Gp6vb5c9eeOEFDAYDCxcurPY60klzkzBEGui9oje+PRvjKC+cyDzByBUjOZWVjNs3a+ku7+X7PxvZWG9urtaKePLJC0n86nkpWEXDwJmQki+BL9Am0I0EvgZWu9Cmy44ThYWkmc3Vnj+Rmfk9VquBkJAHa9kyx+zZs4clS5bw9NNP07dv32pdI/PnTE7PO02bh9sQ8o+Gv/pceexJ3sOQJUM4m5GFefE2bmh9L5GRjUwkfvpJC3n98EMtoR8okVCU4IxQ+EopfwKQUiZIKV8ChrvWrMuLmiYC1OuX4+nZhuDgv9WmWQ6xWq1MmTKFdu3aMWfOnGpdoyiliKMPHMW3ty+hH4XWsoV1w4ajGxj+1XCKDEEULtjNxJuG8sMPUAdZ4muHjAyYMAFGjNBidiMjtYl0CkUpnBGKIqH1KSQIIR4TQtwJtHaxXZcVOw0Gmrm7E1aN4HqzOZOMjO9p3fp+3NzqbnmRRYsWceDAAd5//30CAqqeZFBaJUfHH8Waa6XP133Q+Ta+cYmP/vyIe7++F4/MfuR+sIvXng7l888bWfbXt9+GlSu1HE0HDsDQofVtkaIB4oxnmQ74A9OA19GyvD7kSqMuNyINBq4NCsKtGn38aWlrkdJMSMgDLrCsfPR6PS+99BI333wz9913X7WuceqNU2T/mk2vJb3w61O34yo1xSZtzNw6k/f3vI9/8l0UrljBsk99eaDu/gU14/x5rSXRty+89BL84x/aJDqFogIqFQop5Z/2t7nAgwBCiA6uNOpyIt1k4lhhIRPbVG/uQ0rKspJ04nXFc889R0FBAR9//HG1BrCzf88maU4Srce3ps2kup3zUVMKzYU8uOFB1h1dh8/hJ3Hf8T6//KDjhhvq2zInkBK+/BKeeQa6d4e//oKAACUSikpx2PUkhBgshLhLCNHSvt1HCLEU55ICKpxgVw0SARYWniQnZ6c9U2zdJGT7448/WLp0KTNnzqRXr15VPt+UZuLI/Ufw6e5Dz0U968zu2iC9IJ2bl93MuqPrcP/1PUIOfMjunY1EJBIT4e9/1ybPXXGF1t3UiJ69on6pUCiEEG8AK4DxwI9CiNloa1IcAlRobC0RaTDgKQSDq9HPr9evBCAk5B+1bVa5mM1mpkyZQqdOnZg9e3aVz5c2SdyEOMwZZsK/Dsc9oPEs2Z6QmcDQJUPZe2YfrP2agUXP8Oce0Thy4RUn8fvzT1i0CLZvh57qJ6xwHke/1NFAPylloRCiOVqK8H5SymN1Y9rlwU6DgYEBAXhXMRGglBK9fjlBQdfj7V03cZgLFiwgJiaG9evX41eNFfjOzD9D5g+ZhC4MJaC/61fZqy3+TP6TO1fdSW6eFcsX27h70LUsX94IEvsZjdoKc/36aZFM06dDx46Vn6dQlMFR15NRSlkIIKXMBOKUSNQuRquVqNzcaoXF5uZGUVh4rM7mTpw7d445c+YwcuRI7rqr6rO/DXsMJM5KpOW9LWn3eDsXWOgaNsVtYvhXw8nPCsC4cDfP3Hcta9c2cJEwm+E//9HWrM7MBHd3mD9fiYSi2jhqUXQTQhSnEhdAl1LbSClV4pcaEpWbi0nKao1P6PXLEcKTVq3+zwWWXcqMGTMwm80sWLCgyuMK5iwzR8YdwauDF70W92o04xIf7/2YaT9MwydrEIVLvuPjN1szdWp9W1UJUVHw8MNw+DCMGaOS+ClqBUdCcW+Z7Y9dacjlSHEiwKFVTARos5lJTV1FixZ34uHh+pld27ZtY/Xq1cyZM4fu3auWsE9KybGHj2E6a+LKnVfiEdzwZ/vapI3nfn6O93a/h8+pUcj1K9m8yo877qhvyxxgscCLL2p5mUJCYMMGqEbLT6EoD0dJAbfVpSGXI5EGA718fGhVxRlaWVm/YDan1cncCZPJxBNPPEG3bt14/vnnKz+hDGcXniV9Qzrd3+tO4FXVX7mvrjBajPxzwz9Ze2QtHgenErT3Q77/VceAAfVtWSXodHDsmBbV9M47jWhquKIx0HjCTpoYNinZZTBwV8uWVT5Xr1+Gu3tzWrS4zQWWXcz8+fOJi4vju+++w8fHp0rn5u7PJWFGAi3uaEGH6Q1/6k1GQQajV49m55mduP3yDj2zZ7Blt6BTp/q2rAJycrQJc9OmQY8e8M03Kj+TwiUooagn4goKyLRYqjyQbbHkkp6+kTZtJuDm5tpcEadPn+bf//43o0eP5vbbb6/SuZYcC0fGHsGztSdhX4Y1+HGJk1knuW3FbZxIT4Rv1nBTuzGs/Q6qmX7L9WzZokUynTunhb726KFEQuEynF6QWAhRvYUSFOWy054IsKoD2enpG7DZCuuk22n69OlIKfnwww+rdJ6UkuP/Ok5hYiG9V/XGo0XDdmB/nf2LIYuHkJiaivWLX3j4mjF8/30DFYn0dHjgAbj9dggMhF27YPLk+rZK0cSpVCiEEFcJIaKBePt2PyHEApdb1sTZaTDQysOD0Cp25+j1y/H27kJgoGuTt/3444+sX7+el156ic5VzJd9fsl5Ulen0vW1rgQPa9h95ZuPbebGL2/EkOGLadEuXv/XdXz2WQOunL/zDqxZA3PmwP79cPXV9W2R4jLAmRbFR8AdQAaAlPIQKs14jSlOBFiVLpmionNkZW1zecoOo9HIE088Qc+ePZkxY0aVzs2LyePEkydodkszOr3QUDv3Nf7713+5e83d2PTh2D7dw8qPwnjxxQaY2eLcOYiO1t6/9JImEHPnQjVXQ1QoqoozQuEmpTxVZp/VFcZcLqQUFZFgNFZ5fezU1FWAzeXdTu+88w4JCQl8/PHHeFXBGVnzrRwZcwRdkI7ey3oj3Bqax9WwSRvP//w8U7dMxf3kbfis2cG2TSHcf399W1YGKWHxYggPh4kTte2AAIiIqG/LFJcZzgxmnxFCXAVIIYQOeBI47lqzmjbF8yeqOpCt1y8nIGAwvr5VT8bnLImJicybN4/77ruPW265pUrnxj8ZT0FcAf1+7odnSMNclMFoMTJx40TWxK5Bt/9x2sd+xA+/u1ON/Iau5eRJePRR+PVXuOEGTTAaXFNHcbngjFA8jtb91AnQA7/Y9ymqyU6DAW83NwZUIRFgXl4MeXkH6dGjagPLVWXatGnodDrmz59fpfNSlqWQ8kUKnV/uTLObmrnIupqRWZjJ3avv5vfTv8PPbzHI+izf7ha0alXflpUhKgquv15LvfG//8Ejj4Cb03EnCkWt44xQWKSU41xuyWXEToOBwQEBeFbhx6/XLwd0tG7tun/F5s2b+e6773jnnXfo0MH5eQ8Fxwo4/vhxgq4PovMrDXOh6MSsRG5bcRvH00/CupX8X9j9LF0KVYwlcC2FhZpB/fvDlCnw9NNQhf+DQuEqnPFUfwkhtgghJgghqpTyUwgxQghxTAhxQgjxQgVlxgghjgghYoUQK6ty/cZIgdXK/ry8KnU7SWkjNXUFzZvfiqena1ahLSgoYNq0aYSHh/PUU085fZ610ErsmFjcvN0IXxmOm3vDq/lGnYvimsVDOKFPwfblzzw78n7WrGlAImEywauvaqm/MzK0lsS77yqRUDQYnFnhrrsQYigwDnhVCHEQWC2lXO3oPPt4xkLgFiAZTXA2SymPlCoTCswCrpVSZgkhmvxa3HtzcrBUMRFgdvbvFBUl063b2y6z64033uDUqVNs374djyrEhibMSCD/cD4R30fg1b7hReF8d/w7xqwdizWnFdYvt7Po1d489lh9W1WKvXu1JH4xMdqSpApFA8Sp6p+UcpeUchowAMhBW9CoMq4CTkgpT0opTcBqtDUuSvMosFBKmWW/T6rTljdSIu0T7aqSCFCvX45O50/LlmUfX+0QHx/P22+/zfjx47nxxhudPi91bSrnFp2j47MdaXFbC5fYVhP+F/U/Rq8ajSUlDN2Xe/juiwYkEhYLzJwJQ4ZAVhZ8+y2sWAEtGt5zVCicmXDnL4QYL4T4FtgLpAHOzPZqD5wptZ1s31eankBPIcROIcQeIcSICmyYLISIEkJEpaWlOXHrhsvOnBz6+PrSzMlau9VqJC1tLS1b3oNOV/uLIEgpeeKJJ/D29ubdd991+rzCk4Uce+QYgdcE0vX1rrVuV02wSRuzfpnFY98/hjg5ghabf2Pnj224zfWpsZxHp4MTJ7TIpthYGnZqWsXljjOD2THAt8DbUso/qnDt8mL5ZDn3DwVuBDoAfwgh+kopsy86ScpPgU8BBg0aVPYajQarlOw2GBjb2vketoyMb7Fac1y2QNG6devYunUrH374IW3atHHqHJvJxpGxRxBugt6reuPm0XDGJYosRUzaNIlVMasQ+yYTfnohW3a6N4zufoMBZs/WBqmLk/i5q3RrioaPM9/SblLK6qx+kgyUXlKrA9pyqmXL7JFSmoFEIcQxNOH4qxr3a/DE5udjsFqrNJCt1y/H07MtzZrV/mT4vLw8pk+fTr9+/ZgyZYrT55184SS5Ubn0Wd8Hny4NZUQYsgqzuHvN3fx26jf45Q1u8Xuetb8Lqjiv0TV89x089hicP69FNfXooURC0Wio8JsqhHhPSjkDWCeEuKQW78QKd38BoUKIrsBZtMHwsqN1G4H7gS+FEC3RuqJOVsH+RkVVEwGaTOlkZm6hffun0GIDapd///vfJCcns2bNGtyddFrpm9NJfj+Z9k+2p9XdDWcCwqnsU4xYPpLj6Sdg/QoeveYfLFzYAHI2paXBU0/BqlXajOoNG2Dw4Ho2SqGoGo68wxr732qtbCeltAghngB+AnTA51LKWCHEa0CUlHKz/djfhRBH0NKCPCulzKjO/RoDkQYDbT096ert7VT5tLS1SGlxScqOI0eOMH/+fCZNmsTQoc4lGDSeNhI3MQ7/Af50f6dqK925kv3n9zNy+e1kGAqxrdjKm4/dyHPPNZCJzO++q3UxvfoqvPACVHGRKoWiISCkdNzlL4R4Qkr5cWX76opBgwbJqKio+rh1jemyezeDAwNZ26ePU+X37x+K1ZrLoEGHazUJoJSSv/3tbxw6dIhjx47RyompyTazjYM3HiQ/Op+B+wfi26P2B9arw5b4Ldz39RhMhhaIlVtY9l4fxo6tZ6OSkyEzE664AvLy4NQpcPJ/rlC4CiHEPinloOqc68wo5EPl7Hu4Oje7nEk2GjlVVOR0IsDCwgRycna7JFPsqlWr2LFjB/PmzXNKJACS5iSRsyuHnp/2bDAi8em+Txm1ahRF53oSsGY329fUs0jYbFrKjfBwmDRJS+Ln769EQtHocTRGMRZtXKGrEGJ9qUMBQHb5ZykqoqqJAPV6bapK69a1OwkrJyeHGTNmMGjQIB599FGnzsn8KZPTb5ym7aNtCRkXUqv2VAcpJS/9+hLzIuchEkbQ9a+v+fGXAEJD69Go+Hgt1PW33+Cmm+DTTxtI35dCUXMcjVHsRVuDogPaDOticoEDrjSqKbLTYMDXzY1+/v6VlpVSotcvJzj4Rry9O1ZavirMmTMHvV7P5s2b0ekqHyAvOl/E0QeP4tfXjx4f9KhVW6qDyWrioU0PsSJ6Bex7hGsyF7H5D3eqsfR47REVBdddp60PsXgxPPSQEglFk6JCoZBSJgKJaNliFTUk0mDgmsBAPJxIBJibu5fCwng6dXq+Vm04fPgwCxYsYPLkyQx2IvJGWiVHxx/Fmm8lfE04Ot/aj7yqCtnGbO5efQ87Tm2Hba9zX9tZLP1F4GRsQO1TOonftGladFO7dvVkjELhOir0WkKI3+x/s4QQmaVeWUKIzLozsfGTa7FwKC/P6bBYvX45QnjRqtX/1ZoNNpuNKVOmEBwczLx585w659Trp8jenk3owlD8wv1qzZbqcNpwmqGLh/FbYiSsX8bzQ19k9ap6EomiIm0p0tBQbQ1rd3d46y0lEoomi6Oup+IZXvXZqG8S/JmTgw3n5k/YbGZSU1fTsuUo3N2rtrCRI5YtW8bOnTtZvHgxzZs3r7R81o4skl5NIuTBENpMcG7Gtqs4cP4AI5ffTlp2AWLNj/zv+b/h5PBK7bNnj5bE78gReOABtU6E4rKgwm95qdnYHQGdlNIKDAH+BdRv9bKREWkw4AYMcSLiKStrK2Zzeq3OncjKyuLZZ59lyJAhTJo0qdLypjQTR/9xFJ9QH0L/G+rS9bkr48cTPzLs8+tJT3XHa2UkWxbWk0hYLPDMMzB0KOTkwPffw7Jl4IToKhSNHWeqQxvRlkHtDiwFegNNft2I2mRnTg4Rfn4EOjH7OSVlGe7uLWjevNz8iNXipZdeIiMjg//+97+4VVIDljZJ3D/jMGea6fN1H9z96y/NxJL9S7h9xR0Yz/ag1aY97N7Yl1tvrSdjdDpIStLScMTG0rAyDCoUrsUZobDZczHdA3wgpXySS7PAKirAYrOx22BwKizWYskhI2MTrVuPxc2tdmbw7tu3j0WLFjF16lT69+9fafkz754h88dMQj8Mxf+KyiO0XIGUkle2v8Ij3z6C7cTNhO/9najt7ejXr44Nyc7WhCE+XotiWrsW/vtfGkbyKIWi7nBqKVQhxH3Ag8Bd9n31nUGn0XA4P598m82p8Ym0tPXYbMZa63YqHsBu3bo1r732WqXlDbsMnHzxJK3GtKLt5La1YkNVMVlNPLzpEZZHL4P9D/F38yd8s92DKiwvXjts2gSPPw6pqVpuptBQrVWhUFyGODszezhamvGT9iR/q1xrVtOhKokA9frleHt3IzDwmlq595IlS9i7dy/vvPMOwcHBDsuaM80cuf8I3p296fVpr3oZlzAYDdy6bKQmEr++xr/aLeb7zXUsEno9jB0Ld90FrVvDn39qg9cKxWWMM0uhxgghpgE9hBBhaKvWve5605oGkQYDHb286FRJHKfRmEx29q907vxyrTjp9PR0XnjhBa6//noeeMBxC0VKSdxDcZjOm7hy15W4B9X9uMQZwxluXXobcelxsOkr3h7/T2bOrId5a/Pnw8aN8Prr8OyzTqefNZvNJCcnYzQaXWygQuEYb29vOnToUKUljSujUo8ghLgOWIaWKlwAbYQQD0opd9aaFU0UKSU7DQauq6Q2D5CaugqQtdbtNGvWLAwGAwsXLqxUeM4uOEvGpgx6fNCDwEF13/9+KOUQf196G+mGPNy/+ZEV/7mJ++6rQwPOnNGS+PXrBy+/DBMnQu/eVbpEcnIyAQEBdOnSpV6jxBSXN1JKMjIySE5OpmvX2lt50pmup/eB26SU10ophwK3Ax/WmgVNmNNFRZw1mZwayNbrlxMQcDW+vjVPWLRnzx4WL17M008/Td++fR2WzYnKIWFmAi1GtaD9tLqPUdiasJWhi68jPU0QuC6SHV/UoUjYbNrgdHi41r1UnMSviiIBYDQaadGihRIJRb0ihKBFixa13rJ1Rig8pZRHijeklEcBlVTfCSKLxycqiZLJyztMfv7hWmlNWK1WpkyZQrt27ZgzZ47DshaDhSNjj+DZxpOwL8Lq3Ml9ceALblt+O4XnutJp6x6ivo/AyaUxas7x43DjjTB1KgwZoq0ZUcPPr0RC0RBwxffQmc7o/UKI/6F1PwGMRyUFdIqdBgMBOh0RlSQC1FJ2uNO6dc1zZH/yySccOHCA1atXE+BgFFhKybHJxzCeMnLl71fi0bzuAtmklMzZMZd///4aJNzCNae/4btfA2nRoo4M+OsvLYmfjw98/rnW1aScvEJRIc60KB4DEoDngOfRlir9lyuNaipEGgwMCQxE58AJSWlFr19J8+Yj8PSs2dKier2e2bNnc/PNNzNmzBiHZc9/ep60r9Po9no3gobWXqqQyjBZTUzYMEkTiQOTGGP+nu0/1pFI5OdrfwcMgOnTtTQckyY1GZHQ6XT079+fvn37cuedd5KdXTurASQlJVXahVkd5s6dS/v27enfvz/9+/fnhRdeqPV7FHPw4EG2bNnidPkNGzYghCAuLq5k344dO7jjjjsuKjdx4kS++eYbQAtoeOGFFwgNDaVv375cddVV/PDDD5dcOzExkauvvprQ0FDGjh2LyWQq14bDhw8zZMgQ+vTpQ0REBEajkYKCAm6//XbCwsLo06ePS59ZaRwKhRAiAhgBbJBSjpJS3imlfEdKqUI7KiHbbCYmP7/SsNjs7N8wmc7WSrfTc889R0FBAR9//LHD5mfe4Tzin4qn2a3N6Phs7aYxd0ROUQ63Lr2dZdFfwfa5zOqzhFUrPFyf2M9ohNmzoWdPLYmfTgdvvAFt62euiKvw8fHh4MGDxMTE0Lx5cxYuXFj5SfXM9OnTOXjwIAcPHuTNN990+jyr1Vql+1RVKFatWsWwYcNYvXq10+e8/PLLnD9/npiYGGJiYvj222/Jzc29pNzzzz/P9OnTiY+Pp1mzZixZsuSSMhaLhQceeIBPPvmE2NhYduzYURLFNHPmTOLi4jhw4AA7d+4sV4xqG0cLF72ItpLdfmCwEOI1KeXnLreoibAnJwdJ5QsV6fXL0ekCaNHizhrd748//mDp0qXMmjWLXr16VVjOkmchdkwsHs096L20N8KtbmrTyTnJ/P3L2zmacQTx7Rd8NnVi3UxP2LVLG6iOi4MJE+pk0tzTT8PBg7V7zf794YMPnC8/ZMgQDh8+DEBeXh6jR48mKysLs9nMf/7zH0aPHk1SUhIjR45k2LBh7Nq1i/bt27Np0yZ8fHzYt28fDz30EL6+vgwbNqzkukajkccff5yoqCjc3d2ZP38+w4cP58svv2Tjxo1YrVZiYmKYMWMGJpOJZcuW4eXlxZYtW5xKRgmwbds2Zs6cicViYfDgwSxatAgvLy+6dOnCQw89xNatW3niiScYPHgwU6dOJS0tDV9fXz777DPCwsJYu3Ytr776KjqdjqCgIH755RdeeeUVCgsLiYyMZNasWYx1sBRiXl4eO3fuZPv27YwaNYq5c+dWanNBQQGfffYZiYmJeHl5ARASEnJJy15Kya+//srKlVoWpAkTJjB37lwef/zxi8pt3bqVK664gn72dAQt7E1uX19fhg/X8rV6enoyYMAAkpOTnXquNcFRi2I8cIWU8j5gMPC4g7KKMkQaDOiAqx0MZFuthaSlfUOrVvei01V/eVGz2cyUKVPo1KkTs2fPdlg2fmo8hfGF9F7ZG8/WdROTcFh/mIGLriFOn4jP+i38+GYdiITFoq0PMWwYFBTAjz/Cl19Cs2YuvnH9Y7Va2bZtG6NGjQK0uPoNGzawf/9+tm/fzowZM5BSAhAfH8/UqVOJjY0lODiYdevWATBp0iQ++ugjdu/efdG1i1sp0dHRrFq1igkTJpRE2MTExLBy5Ur27t3L7Nmz8fX15cCBAwwZMoSlS5eWa+v7779f0vX0008/YTQamThxImvWrCE6OhqLxcKiRYtKynt7exMZGcm4ceOYPHkyCxYsYN++fbz77rtMmTIFgNdee42ffvqJQ4cOsXnzZjw9PXnttdcYO3YsB7wpYJEAACAASURBVA8eZOzYsURFRfHII4+Ua9PGjRsZMWIEPXv2pHnz5uzfv7/SZ37ixAk6depEYCWBKxkZGQQHB+Nuz/vWoUMHzp49e0m548ePI4Tg1ltvZcCAAbz99tuXlMnOzubbb7/lpptuqtS+muJoMLtISpkPIKVME0KofMpVYGdODv39/fFzUIPNyNiM1ZpLSMiDNbrXxx9/TExMDOvXr8fPr+LEvilfpaBfqqfL3C40u7FuHObPCT8zeuW9GA2BtP45kp9XXkFERB3cWKeDs2e1qKZ586jL6d1VqfnXJoWFhfTv35+kpCQGDhzILbfcAmi12BdffJHff/8dNzc3zp49i16vB6Br164lOcAGDhxIUlISBoOB7OxsbrjhBgAefPDBku6NyMhInnzySQDCwsLo3Lkzx48fB2D48OEEBAQQEBBAUFAQd96ptZIjIiJKWjdlmT59OjNnzizZPnToEF27dqVnz56AVuNeuHAhTz/9NEBJSyAvL49du3ZxX6lY6qKiIgCuvfZaJk6cyJgxY7jnnnvKve+gQYNYvHhxucdWrVpVcr9x48axatUqBgwYUGF3blWijIoFurLzLRYLkZGR/PXXX/j6+nLTTTcxcODAElH4//bOPD6m6/3j70PsxE4RJCJIgqR2qpa2qrSltS9tUa22gqJKfbWoltop1aqftvbE1qKqVUvs+xJb7BLEkgQxSSRkmef3x0ymiUySSSQSnPfrNa/MPffcc56TmbnPPdvniYuLo3v37gwaNIgqVarYXH9GSc1RVEkUK1sBzoljZ4uI9U9AQ6zRyP7wcPqlMQYeHLyEvHkrUKxY8wzXdf36dcaMGUObNm146623Usx37/Q9zvU/R7EWxaj8ZeUM15ceFvgtoO/aDzEGu+Lut4F//3bI2tg+YWEwYoRpR7WLCyxf/kzpMyXMURgMBt544w3mzJnDoEGDWLp0KaGhoRw+fJg8efLg6Oho6QUkDJOAaTI8OjoaEUnx5mftRpdA4rJy5cplOc6VKxdxcXE2tSG18gHLg5DRaKRYsWL4WRnjmzt3Lvv37+evv/7C09PTap6UuH37Nlu3buXkyZMopYiPj0cpxeTJkylZsiRhYWFJ8t+5c4dSpUpRtWpVrly5QkRERKqrDUuVKsXdu3eJi4vDzs6OoKAgylv5UTg4ONC8eXNKmWP8tm3bliNHjlgcRb9+/XBxcbE4tKwmtV5CR0yxsucAPzx0nPNnybKRo5GRRKchBBgTE8qdO/9QtmwPlMr4zSxhLHj27Nkp/rjjo+Px7+JP7kK5cV3qisqdtfMSIsLX28bRZ20fjJea8+q1nez7N4udxO+/mzbO/for7NplSnuGnERiihYtyqxZs5g6dSqxsbEYDAbKlClDnjx58PX15fLly6leX6xYMYoWLcou8/9x6dKllnPNmjWzHJ87d44rV66kOieWXmrUqEFgYCAXLlwATAG3Eno2ibG3t8fJyYmVK1cCpu/csWPHALh48SINGzZk3LhxlCpViqtXr1KkSBGrE8sPs2rVKt577z0uX75MYGAgV69excnJiV27duHi4sL169c5ffo0AJcvX+bYsWN4enpSsGBB+vbty6BBgyyrmG7cuMGSJUuSlK+UomXLlpaVUgsXLqR9+/bJ7GjdujXHjx8nKiqKuLg4tm/fjpubG2AKG2AwGJj5GLuuqQUu2pLa67FZ+ARiixBgaOgKROIeabXT1q1b8fHx4YsvvsDZ2TnFfBcGX+DeyXu4LnYlX/l8KebLDGLjY+n1R1/Gbh8Dfu/xUeEN/LW6KGlsJck4N29Cp07QsSM895xpj4QNwZmedp5//nk8PDzw8fGhZ8+eHDp0iHr16rF06VJq1KiR5vW//fYbXl5eNG7cmAIFCljS+/fvT3x8PLVq1aJr164sWLAgSU/iUcmfPz+//fYbnTt3platWuTKlYuPP/7Yat6lS5fyyy+/4OHhgbu7O2vXrgXg888/p1atWtSsWZNmzZrh4eFBy5Yt8ff3x9PTk+XLl6c4R+Ht7c3bb7+dJK1jx44sW7aMfPnysWTJEvr06YOnpyedOnVi/vz5FDX/zr/99ltKly6Nm5sbNWvW5K233qJ06eRL3idNmsT06dOpWrUqt2/fpq95wm7dunWMHj0agOLFizN06FDq16+Pp6cnderU4fXXXycoKIjx48fj7+9PnTp18PT0THEILVMRkSx7YVpaexa4AHyRSr5OgAD10iqzbt26ktPpcOKEOO3dm2qeQ4cayoEDtTJcx4MHD6RGjRri5OQkUVFRKea76X1TfPGVi19czHBdtmK4b5Dm81sJYxFajpapU41iNGZxpcOHi+TLJzJhgkhMTBZXljL+/v7ZVrdG8zDWvo/AIcngvTzLZEKVaTxlDtAKCAIOKqXWSSI5EHO+IsAgYH9W2fI4EbMQYKtUlgJGRZ0nImI/VaokX8lgKzNmzODMmTOsX78+yRNfknouRHGu3znsm9jjOM4xw3XZwrXwa7zy2+ucuXMSu39+wWf4+3TsmEWVBQaa5iOefx5Gj4b334dMHP7QaDRJsXklk1Iqvf3LBpgkyS+JSAzgAyQfjINvgMnAU7GJ72J0NMGxsanunwgOXgooypTpnqE6rly5wrhx42jfvj2vv/661TzGB0b8u/qj7BRu3m7kypN1i9ZOBJ/g+R8bcSbkIkXX/8XO77PISRiNMHs21KwJH35oEvErVEg7CY0mi0nz7qGUaqCUOgGcNx97KKVm21B2BeBqouMgHgqhqpR6HqgoIuvTsKGfUuqQUupQaGioDVVnH7vDw4GUhQBFhODgJRQr1pL8+R0yVMeQIUMQEb7/PmUR34vDLxJ5JJIaC2qQv1LWbX3eGrCVhvOaEnornspbdnJkZWsaZU7cpaScPm3SZxo0yPR39eqnRnpDo8np2PKYOQt4A7gNICLHMEW8Swtrv2LL2jfzvowZwGdpFSQi80SknojUszY5lJPYbTBQzM4OtxT2M4SH7+P+/YsZnsT+559/+P333/nyyy+pXNn6MtfQNaFcm3UNh8EOlGpXKkP12MKiY4t5ddFrRN+sSIPj+ziywZMsWdJ94IBpa/KZM7BoEWzYACm0XaPRZD62OIpcIvLwejpbhFaCgMRCQg7A9UTHRYCawDalVCDQCFinlKpnQ9k5ll0GA03s7cmVwtNucPAScuXKT+nS6R+buX//PgMHDqRatWp89pl1/3r/8n3O9jlLkXpFqDIpazbiiAhf+35LrzXvER/QlM4Ru9jxZyVsVGiwnchI09+6dU17I/z94d13dU9Co3nM2OIoriqlGgCilMqtlBoMnLPhuoOAi1LKSSmVF+gGrEs4KSIGESklIo4i4gjsA9qJyKH0NyNncDs2ltNRUSkuizUaYwgJWU7Jku2xs0t/JLkpU6Zw4cIFfvjhB6tLEo2xRvy7+SNGwc3HjVx5M39eIjY+lt6/92Psjq/g2Lv8r9I/LF9YjExcIWkS8Rs50rRpLjTUtB/i22+hbNlMrESj0diKLXeST4ChQCUgGNOTf5q6TyISBwwANgKngRUickopNU4p1S7jJudc9pj3T6Q0kX3nzkbi4m5naNgpICCACRMm0LlzZ4s0Q7I8XwYQvi+c6v9XnQLO1ldCPQoRDyJo9Vs7Fp2cj9r5Jb+8sZDx4/Jm7gP+rl2mkKQTJ0LbtjbHrNZA4USbVTZs2ICLiwtXrlxh7NixFCxYkJCQEKt5lVJJeqhTp061SQgPTKvv8ufPj8H83QdYsGABAwYMSJKvRYsWHDpkegaMjIzko48+wtnZGXd3d5o1a8b+/ckXPR4+fJhatWpRtWpVBg0alOKu7W3btuHp6Ym7u7tlc97Vq1dp2bIlrq6uuLu7pzqfp7GBjK6rza5XTt5HMeLCBcmzbZtExcVZPX/yZGfZtauUxMenf73/m2++KYUKFZKrV69aPX9rwy3xxVfOfnw23WXbwrXwa1JtuqcwOrfkf2GebN6cyRXExop4eYmAiKOjyKZNmVxB1pIT9lEUKlRIREQ2b94sVapUkQsXLoiIyJgxY6RixYoyfPjwZHlFRPLlyyeOjo4SGhoqIiJTpkyRMWPG2FRn/fr1pWnTpvLbb79Z0n777Tfx8vJKkq958+Zy8OBBERHp2rWrfPHFFxIfHy8iIhcvXpT169dbLXvPnj1iNBrltddekw0bNiTLExYWJq6urnL58mUREQkODhYRkevXr8vhw4dFRCQ8PFxcXFzk1KlTNrXpaeCx76NQSv0fiSahEzmYflnhuJ5kdoeHU7dIEQpYkY6IizNw69Y6ypf/kFy50veU/Oeff/Lnn38yefJkHBySr5R6cO0BZ947Q6HahXCenvIO7YxyKuQULea34da9MEpv+5OtP7ch0+PY2NlBcLBJ8fXbb8m6rdxZz+B/BuN3M3N1xj2f82Tma2lLNuzcuZMPP/yQDRs2JNmt//7777NgwQJGjBiRTO7bzs6Ofv36MWPGDMaPH2+zTRcvXiQyMpIpU6YwYcIEevfubdM1+/fvZ+nSpeTKZRrQqFKlSjJhuxs3bhAeHk7jxo0BeO+991izZg1t2rRJkm/ZsmV06NCBSpUqAVCmTBkAypUrRzmz1lqRIkVwdXXl2rVrFhkMTfqwZehpM7DF/NoNlAEeZKVRTyIPjEYOhoenuCw2NHQ1Ig/SPewUFRXFoEGDcHNzsyoAZowz4t/Dn/joeNyWu5G7QObqG/kG+FJ/7gvcuhNLjT07OLYqE53E7dumWBFnz5qOly83Sa8+wU4iO3nw4AHt27dnzZo1yWQ6ChcuzPvvv5/iEIyXlxdLly5NMoQESWUlHsbb25vu3bvz4osvcvbs2SRDWylx6tQpPD09yZ2GDte1a9eSPBSlJscdFhZGixYtqFu3rlU588DAQI4ePUrDhg3TtE9jnTR7FCKyPPGxUmoxsCnLLHpCORwRwQORFCeyg4OXUKBAVYoUaZCucr/77jsCAwPx9fW1RLhKzOVvLmPYYaDGohoUqpGyxHhGWOy3lN5r+mAMdeHl4A2s+bNy5tzDRWDVKhgwAO7cMe2LqF4dcj0dSva2PPlnBXny5KFJkyb88ssvVh3CoEGD8PT0tLpizt7envfee49Zs2Yl2enfrl07S1yLh/Hx8eGPP/4gV65cdOjQgZUrV+Ll5fXY5bgPHz7Mli1biI6OpnHjxjRq1MgiUx4ZGUnHjh2ZOXNmmrEiNCmTkV+mE6AXsT/ErlSEAO/fv8rdu9soW/addP1Yzp8/z+TJk+nZsyctWrRIdj5saxiXv7nMc72f47l3n8uw7Q8jIny9dQLvrX0HY+ALfGi3i43LM8lJ3LgBHTpAly5QsSIcPgw2DFlo0iZXrlysWLGCgwcPMmHChGTnixUrRo8ePfjxxx+tXj948GB++eUX7iXEFk+F48ePc/78eVq1aoWjoyM+Pj54e3sDpCrH7e7uzrFjxzAajamW7+DgkCRyW2py3K+99hqFChWiVKlSNGvWzKIiGxsbS8eOHenZs2eKcSk0tmHLzuwwpdQd8+supt7E/7LetCeL3QYDLgUKUCZv8qhxISHLAEnXsJOIMHDgQPLnz8/UqVOTnY8JjuF0z9MUrF4Qlx9cHsX0JMQZ43hv5UeM3TkKjvdgisc/zPu+eOYpds+caYo2N3ky7NsHtWtnUsEaMIXKXL9+vUVZ9WGGDh3Kzz//bDU+RIkSJejSpYvV6x7G29ubsWPHEhgYSGBgINevX+fatWtcvnyZ+vXrs3v3bm7evAnAoUOHePDgARUrVsTZ2Zl69eoxZswYS6/h/PnzFuXXBMqVK0eRIkXYt28fIsKiRYusynG3b9+enTt3EhcXR1RUFPv378fV1RURoW/fvri6ujJ06FCb/neaVEhtphvT7upKQG7zS2V01jyzXjlx1ZPRaJSSO3dKn9OnrZ7bv99dDh9unK4yV61aJYDMnDkzeZnxRvFr5Sfb82+XiOMRGbb7YSIeRMiLP7cRxiJ2r/5PVq2Oz5yCL10SMa9AkchIkXPnMqfcHEROWvUkInLlyhVxdHSUNWvWyJgxY2TKlCmWc0OGDBHTTz/5dTdv3pQCBQpYVj2tXbtWvvrqq2R1OTo6yumHvu9DhgyRiRMniojImjVr5PnnnxcPDw954YUXLCuQREQMBoN88MEHUqVKFalZs6Y0b95cDhw4kKyOgwcPiru7u1SpUkW8vLzEaJYi/umnn+Snn36y5Js8ebK4urqKu7u7zJgxQ0REdu7cKYDUqlVLPDw8xMPDQ/7666+0/4lPCZm96intDHA4o4VnxSsnOorTkZGCr6/Mv3492bmICD/x9UWCgubYXF5ERIQ4ODiIh4eHxMbGJjsfOCFQfPGVa/OuPZLdibkefl2qTa0jjM4tRVr8LPv3Z0KhcXEiM2eKFCwoUreuZL3mePaRExyFRpNAdsiMH1BK1RGRtCOMP6OkJgR48+ZilLKjdOkuNpf3zTffEBQUxPLlyy1B2BO4u+suAV8FUKZbGcp9kHqoVVvxD/Wn+by23Iq6hcPedez4tS1OTo9aqL9pRdO+fdCmDfz8s5be0GieUFJ0FEopOzHtrm4KfKiUugjcwzQcJSJS5zHZmOPZZTBQ0s6O6gULJkkXiSckZBklSrQhb17bxPlOnz7N9OnT6dOnD02aNElyLvZ2LKe7nya/Y36q/VwtXRPjKbEtYDttFr3F/ch81D2znU2r61K8+CMWun8/NGsGRYrAkiXQo4d2EhrNE0xqPYoDQB3grcdkyxPLboOBF4oWTXbjDgvzJSbmBmXLvmtTOSKCl5cXhQsXZtKkScnOnel9hpiQGOrsrYOd/aPHnFp0dJkprvUtZ96O2oDPakeszMXbTkSEyTnUqwcjRpiWv5o3QGk0mieX1FY9KQARuWjt9Zjsy/GExMRwPjraqr5TcPAScue2p2TJN2wqy8fHB19fXyZMmJAs1m7QzCBur7+N81RnitQp8kg2iwhjNk2k17qeGC83Znip3az+5RGcRFQUDB+eVMRv3DjtJDSap4TUHktLK6VSXFcmItOzwJ4njt0p7J+Ij4/i1q3VlC7dhdy50xboCw8P57PPPqNevXr065dUHSX8QDiXRlyi1FulqDCgQgol2EacMY7eywey9Nxc1KluzGu7gA96P4L06/bt8MEHcOGCKercI3VJNBpNTiQ1R5EbKIz1AEQaM7sNBvIpRd0iSZ/yb91aS3x8pM17J8aOHcvNmzdZu3ZtEnmD2Lux+HfzJ2/5vFT/tfojzUtExkTS5pdu7Ar5i3wHR7B+yAReeTmDu6Hj4mDgQJg7F6pUgS1b4KWXMmybRqPJuaR2l7ghIuNE5Gtrr8dmYQ5nl8FAfXt78j0kPxEcvIR8+RwoVqx5mmWcOHGCWbNm0a9fP+rXr29JFxHOfXiOB1cf4ObjRp7iGZfcvhl5E8+ZLdh1829K7PmRI5MnZtxJgEnELywMhg6FEye0k8gBFM6ErfPXr1+nU6dOKZ6/e/dukp3daeW3hfbt21vE/xLo3bs3q1atSpKWuH3nzp2jbdu2VK1aFVdXV7p06UJwcHCyshcuXIiLiwsuLi4sXLjQav1+fn40atQIT09P6tWrx4EDBwA4c+YMjRs3Jl++fFY3vT5TpLRuFjia0TW3WfnKSfsoouLiJM+2bTLCLOecwIMHweLrm1suXBiRZhlGo1GaNm0qJUuWlFu3biU5F/RjkPjiK5cnX34kO/1D/KXEOEfhfwWlats/5caNDBYUGirSq5fImTOm4/hM2pD3FJAT9lEk3jiXVQQEBIi7u3umlRcWFiYODg5So0YNuXTpkiW9V69esnLlyiR5E9oXHR0tVatWlXXr1lnObd26VU6cOJEk/+3bt8XJyUlu374td+7cEScnJ7lz504yG1q1amWRMP/rr7+kefPmImKSLD9w4ID873//S7Jh8Ungce6jePnxuKonl4MREcSKJJvIDglZDsTbNOy0aNEidu3axfz58ylZsqQlPcIvggtDLlCiTQkqflYxlRJSx/fSDtosfIsHUXloFrSdDSvqkUI475QRMSm7DhwId+9Cy5ZPlYhfZnP+/GAiIzNXZrxwYU9cXNIvNnj58mXef/99QkNDKV26NL/99huVKlXi4sWL9OzZk/j4eNq0acP06dOJjIwkMDCQN954g5MnT3Lq1Cn69OlDTEwMRqOR1atX89VXX3Hx4kU8PT1p1aoVXl5elvzx8fGMGDGCjRs3opTiww8/ZODAganat3r1at58803Kli2Lj48PI0eOTLNNy5Yto3Hjxrz55puWtJYtWybLt3HjRlq1amWRVW/VqhX//PMP3bt3T5JPKUW4eS+UwWCwaEqVKVOGMmXK8Ndff6Vp09NOir90EbnzOA15EkkQAmzykKMIDl5MoUIeFC6cuh53WFgYn3/+OY0bN6ZPnz6W9LiIOPy7+pOnZB5qLKyBypWxeYkFh314ZWErHtwpQx/jPrYuyYCTuHYN3noLuncHJyc4cgR69cqQPZrHz4ABA3jvvfc4fvw4PXv2ZNCgQQB8+umnfPrppxw8eNCq2B7A3Llz+fTTT/Hz8+PQoUM4ODgwceJEnJ2d8fPzY8qUKUnyz5s3j4CAAI4ePWqpD2D06NGsW7fOWhUWqfLu3btbRAXT4uTJk9StWzfNfNeuXaNixf8eslKSKp85cyaff/45FStWZNiwYXz33Xc22fEs8eiL8Z9hdhsMuBYsSIlE8t9RUWeJiDiIs3PaY5pfffUVt2/f5t9//7UEcRERzn1yjugL0Xhu9SRv6fSvIhIRxmycyjf7h8PVpkyovZaRg0ukfaE1Zs+GTZtg6lQYPJjMUwd8esnIk39WsXfvXn7//XcA3n33XYYPH25JX7NmDQA9evRg2LBhya5t3Lgx48ePJygoiA4dOuDikrr45ObNm/n4448tagIJT/Ljxo2zmj84OJgLFy7QtGlTlFLY2dlx8uRJatasaXXRRnoXcphGW9Iu46effmLGjBl07NiRFStW0LdvXzZv3pyuup529NhBBjGKsCc8PNmwU3DwUkBRpkx36xeaOXLkCD/99BNeXl54enpa0m8uuEnI0hAcxzpSrHmxdNsVb4ynx9IBfLN/OLnPdGFlu03pdxIXL5rkvwG++gqOH4fPPtNO4ikgPTfbHj16sG7dOgoUKEDr1q3ZunVrqvlFJF3lL1++nLCwMJycnHB0dCQwMBAfHx8guVR5gkw5gLu7O4cTvp+p4ODgwNWrVy3HKUmVL1y40CJD3rlzZ8tktuY/tKPIIP737nE3Li7J/gkRITh4CcWLv0y+fNa78wBGo5H+/ftTunTpJE9b9/zvcd7rPMVeKkbl/6U/5Me9mHu8OOdtfC7+SMGjn7NnqDed3spvewHx8TB9OtSqBR99ZJqbKFQIqlZNty2anEGTJk0sN9+lS5fStGlTABo1asTq1asBLOcf5tKlS1SpUoVBgwbRrl07jh8/TpEiRYiIiLCa/9VXX2Xu3LkWCfM7d1Ifvfb29uaff/6xSJUfPnzYYkuLFi1Yvnw5MTExACxYsMAyD9GjRw/27NmTZO7gn3/+4cSJE0nKb926Nf/++y9hYWGEhYXx77//0rp162R2lC9fnu3btwOwdevWNHtOzyQZnQXPrldOWfU099o1wddXLkRFWdLu3t0lvr7IjRsLUr123rx5AsiiRYssaXH34mS/+37ZVWaX3L9+P9323Iy4KU4T6gujc8lz7X6QgIB0FnDihEiDBiIg8uabIkFB6bbhWSYnrHpSSkmFChUsr2nTpklAQIC0bNlSatWqJS+99JJcvmxaQXfu3Dlp0KCB1K9fX8aOHSvly5cXkaSrmiZMmCBubm7i4eEhrVu3ltu3b4uISPfu3cXd3V2GDRuWJH9sbKwMGTJEXF1dpXbt2jJ79mwREfnqq69k7dq1SWwNCAiQ8uXLW6TDE3j++edl3759IiIyduxYqVmzpnh4eEiHDh0kJCTEku/06dPSunVrqVq1qri6ukrXrl3l5s2byf4nv/zyizg7O4uzs7P8+uuvlvS+ffvKwYMHRcQkSV6nTh2pXbu2NGjQQA4dOiQiIjdu3JAKFSpIkSJFpGjRolKhQgUxGAwZ+WgeO49dZjynvXKKo3jH31/K7tqV5It+9uzHsn17AYmNDU/xulu3bkmJEiWkWbNmSa4988EZ8VW+cvvf2+m25XTIGSk2xkkYVUBqdVorYWHpLGDfPpE8eURKlRLx9n6q5cCzipzgKNLDvXv3LN8/b29vadeuXTZbpMlMskNmXGOFh4UAjcYYQkKWU6rUW9jZpazFNHLkSAwGA3PmzLFcG7wsmBvzb1Dpf5Uo0Sp98wm+F3fx2sL2xNzPzeuGbfy+tIHtKhrh4WBvbxLxGzUKvLyglG0qt5onm8OHDzNgwABEhGLFivHrr79mt0maHIx2FBng+oMHBNy/z8AK/+ku3bnzN3FxYanundi/fz/z589n6NCh1KxpWjobdT6Kcx+dw/4Fexy/dkyXHb8dWEnfv95FwiozuPTfTJ9UxTY176goGD0aFi827aouUwbGjElX3ZonmxdffNESW1qjSYssncxWSr2mlDqrlLqglPrCyvmhSil/pdRxpdQWpVT6Z3CzAWtCgDdvLiZPntIUL97K6jXx8fH079+fcuXKMcZ8U46/H49/F39UXoWbtxu57Gz7OESE/62fyvt/d4Fr9fix3h5mjLbRSfj6miarp02Dt9+G/OmY7NZoNM8kWdajUErlBuYArYAg4KBSap2I+CfKdhSoJyJRSqlPgMlA16yyKbPYbTBQIFcunjdrz8TG3uX27T8pX/5jcuWyrsc0d+5cjhw5go+PD0XMAoKXPr9EpF8kNdfVJH9F227Y8cZ4ui8azMrLP5DnXGfW913Eqy/ZcG1cnGload480yomX19o0cKmOjUazbNNVvYoGgAXROSSiMQAPkD7xBlExFdEjXvAWwAAIABJREFUosyH+wCHLLQn09hlMNDQ3p485k1yoaGrEIlJcdgpJCSEUaNG8corr9CliykkaujvoVz74RoOQx0o9aZt8wJRsVE0+b4jKy//gP2pzzg6ysc2JwEmET+DAT7/HI4d005Co9HYTFY6igrA1UTHQea0lOgL/J2F9mQKkXFx+EVGJomPHRy8hAIFqlGkSD2r1wwfPpyoqChmz56NUorogGjOvH+GIvWLUOW7KjbVGxwZguuklhwwrKPSqVmc/2Eq7m5pfHwhIfDee3DmjOl42TKYPBkeCtmq0Wg0qZGVjsLaiHnyPfWAUuodoB4wJYXz/ZRSh5RSh0JDQzPRxPSzPyKCeLDsyL5//zIGw3bKln3H6q7UXbt2sXDhQj777DNq1KiBMcaIfzfT6Jvbcjdy5U37Izgdco5qkxtz5f4JGgb+zukFA1MPHicCS5eCmxv4+MDBg6Z0LeL3VDN+/Hjc3d2pXbs2np6e7N+/n7FjxyYT2vPz88PV1RUAR0dHXnzxxSTnPT09LYst0uLo0aMopdi4caMlLTAwMNn1Y8eOTSLVPXXqVGrUqEHNmjXx8PBg0aJFycq+c+cOrVq1wsXFhVatWiXZqZ2Ar68vnp6ellf+/Pkt0iQ9e/akevXq1KxZk/fff5/Y2Fib2qRJTlbeOYKAxLKnDsD1hzMppV4BRgHtROSBtYJEZJ6I1BOReg+HCH3c7DYYUEBjs6MIDl4GQNmyPZPljYuLo3///lSqVIkvv/wSgIBRAUQciKD6/OoUcEo78t3mc7vxmN2Y8PsRdI/xZfcvb6XeIbh6Fd58E955xxSa1M8P3rUtZrfmyWXv3r2sX7+eI0eOcPz4cTZv3kzFihXp3r07y5cvT5LXx8eHHj16WI4jIiIsUhenT59OV73e3t40bdrUZkE/MM3Xbdq0iQMHDnDy5El27NhhVZdp4sSJvPzyy5w/f56XX36ZiRMnJsvTsmVL/Pz88PPzY+vWrRQsWJBXX30VMDmKM2fOcOLECaKjo5k/f3662qb5j6xcHnsQcFFKOQHXgG5Aj8QZlFLPAz8Dr4lISBbakmnsNhioVagQRe3szJIdi7G3f4ECBZIPIf3www+cOHGC33//nUKFCnH7r9tcnXqV8v3LU6ZT2vGk5+9eTb+NPRFDJcZU/Zuxg5zTNnDOHNNE9cyZMGCA1mfKBs4PPk+kX2SmllnYszAuM1OWlrhx4walSpUiXz5TWNtSifbDFCtWjP3799OwYUMAVqxYkaQH0KVLF5YvX86wYcMsaq6LFy9O0yYRYdWqVWzatIkXX3yR+/fvk9+GVXQTJkzA19cXe/PwbdGiRellRZF47dq1bNu2DYBevXrRokULJk2alGK5q1atok2bNhQ0P0m1bdvWcq5BgwYEBQWlaZvGOlnWoxCROGAAsBE4DawQkVNKqXFKqXbmbFMwhVtdqZTyU0pZ1yLOIcSLsDc83LIsNjLSj6io01Ynsa9fv87o0aNp06YNb731FveD7nO612kKeRTCeVraN/zhf8zgw02dUTfrsOSlPak7ifPn4dAh0/vRo+HkSfj0U+0kniFeffVVrl69SrVq1ejfv79Fuwige/fuFg2lffv2UbJkySR6Rp06dbIozP75559J4jxcv349yQ03Mbt378bJyQlnZ2datGjBhg0b0rQzIiKCiIgInJ3T/g0EBwdTrlw5AMqVK0dISOrPkj4+PsliTQDExsayePFiXnvttTTr1FgnSzfcicgGYMNDaaMTvX8lK+vPbE5ERhIRH29xFMHBS1AqD2XKdE6Wd9iwYcTExDB79mwkXjjd4zTG+0bcV7iTO3/KN/B4Yzyd53/GHze+J19AB7Z4LeGFBikMUcXFwYwZJudQsyYcOGCaqHZyypT2ajJGak/+WUXhwoU5fPgwO3fuxNfXl65duzJx4kR69+5Nt27daNKkCdOmTbN6My1RogTFixfHx8cHV1dXyxM5mATzUnIA3t7edOvWDYBu3bqxePFiOnTokKKCrFIKkfQpzNrKjRs3OHHihFXRv/79+9OsWbNkczEa29E7s9NBQqCipkWLYjTGERKyjBIl2pInT8kk+bZu3Yq3tzejR4/G2dmZgK8CMOw04LrElYLVUp5giIqN4sUZ73Ak+g9KnhvCwfFTcHJMwakcPw59+5p6Eu3bw48/YtuOO83TSu7cuWnRogUtWrSgVq1aLFy4kN69e1OxYkUcHR3Zvn07q1evZu/evcmu7dq1K15eXixYsMCmuuLj41m9ejXr1q1j/PjxiAi3b98mIiIimUQ4mCamnZycsLe3p1ChQhZl2tQoW7YsN27coFy5cty4cYMyqazgWLFiBW+//TZ58iTdx/T1118TGhrKzz//bFO7NNbRy2DSwe7wcCrkzUulfPm4e3crMTE3ee65pBPFMTExeHl54eTkxBdffMGdzXe4PP4yz73/HGV7lk2x7JDIUKqNf5kjUWuodmkmF3+cnrKT2L8f6taFK1dgxQr44w9IIUqZ5tng7NmznD9/3nLs5+dH5cr/CR10796dIUOG4OzsjIND8u1Kb7/9NsOHD7f6RG6NzZs34+HhwdWrVwkMDOTy5ct07NiRNWvWULhwYcqVK8eWLVsAk5P4559/LBLnI0eOxMvLyxJ+NDw8nHnz5iWro127dixcuBAwxYxo3759sjwJJMytJGb+/Pls3LgRb29vS2AwTQbJqJpgdr2yUz224p490vXkSRER8fd/V3bsKCpxcdFJ8kyaNEkAWb9+vdy/cV92ld0l+932S1xkXIrlnrx+Tgr/z1kYlV9e8lotMTEpZLx71/Q3Lk7k669Fbt3KjGZpMoHsVo89dOiQNG7cWFxdXaVWrVry9ttvS2hoqOV8SEiI2NnZyU8//ZTkusqVKyfJJ5JUavzatWvSpk2bZPX16tUrWVlr166V1157TURETp06JS1atBAPDw/x8PCQJUuWWPIZjUaZNGmSVKtWTdzd3cXT01MWL16crI5bt27JSy+9JFWrVpWXXnrJInN+8OBB6du3bxJ7y5cvL/Hx8Umuz507t1SpUsViw9dff53yP/ApQ8uMZ5OjuBwdLfj6yqyrVyUuLlK2by8kZ858kCTPlStXpGDBgtK+fXsxxhnl6MtHZXuB7RJ5MjLFcv85uUfy/K+UMLyk9Ptmj3WF78hIkU8/FSldWiQ4OJNbpskMsttRaDSJyWxHoftjNpJYCPDWrTUYjfeSrXYaMmQIIsLMmTO5MvEKd7fcxWW2C4XcC1kt86dtv9PG5yViI4sxzW0vP3/ZOPk0w+bNponq77+HLl2gQNp7LzQajSYz0Y7CRnYZDBTOnZvahQoRHLyEfPkqUbTof6soNm7cyOrVqxk1ahTFrhYjYHQAZXqU4bn3n7Na3pDl39N/WydyhXqyrt0ehvZ6aKVMXJxpsrpVK8ibF3bsgB9+gCIpx7rQaDSarECverKR3QYDjeztMcaFcufOv1SqNAKlTH72wYMHDBgwgGrVqjGo9yBONDxBgSoFqDa3WrKlgEYx0m7OZ/x1eyYFr7zN7s+W4OluZSWUnR3cvw9ffGFa/qp7EhqNJpvQjsIGDHFxnLh3j9GlSxMS4gMYkww7TZkyhQsXLrDxn40EfBRAbGgstfbVwq5I0n9vVEw0jSa/y4n41TwXOIijE6fzXNlEK5uCg2HoUPjyS3B1hSVL9JJXjUaT7eihJxvYFx6OEXjB3p6bNxdTuPDzFCrkBkBAQADjx4+nc+fOuJ505c5fd6g6vSpFnk86RHQz/BbOX7/CibjfqX19Opd+/P4/JyFiijbn5garVsHhw6Z07SQ0Gk0OQDsKG9htMJAbqGl3g8jIw5Qt+9/eiU8//ZTcuXPz9btfc+mLS5TqUIry/ZPuaTgedBHn75pwM9dh2j1YwdGfhvw3knTlCrz+ukkOvHp1k4jfOymHU9VoNJrHjXYUNrDLYMCjcGGibnkDuShTxiRb8Oeff/Lnn3/y5edfcnfgXfI55KP6L9WTzEus99tP3R8bEWW8w2elt7L2u05J1b5/+sk0UT1rFuzcaRpy0mg0mhyEdhRpEGs0sj88nCZF7QkJWUrx4q+QL185oqOjGTRoEG5ubrT2a03MtRjcfNzIU+w/CYHvN66h3eqWxEfbM6/hHqYOamI6cfasSZcJ4KuvTCJ+AwdqET/NI/G44lFs27aNN954I1n6Bx98gL+/v5UrHh+hoaHkyZMnmWRHYXPY4gQWLFjAgAEDLMeLFi2iZs2auLu74+bmliR2RgIPHjyga9euVK1alYYNGxIYGJgsz9mzZ5PEx7C3t2fmzJmAKSZHhQoVLOdsEVHMKejJ7DQ4FhlJlNFIHbsb3L8fiKPjOAC+++47AgMDWTFoBYZZBpynOWPf8L+od14LZ/NjwKfkuVOfv9/9k5cblYHYWJg2DcaOhVq1/hPxc3TMnsZpsoTBgwfj5+eXqWV6enpabjjWSByPIl++fNy6dYuYmBi6d+9OmzZt+O677yx5U4pHUbFixXTHo0hMVsd7iIuLw84u9VvWypUradSoEd7e3nz00Uc2lfv3338zc+ZM/v33X8qXL8/9+/etyqz/8ssvFC9enAsXLuDj48OIESOSxfqoXr265bOPj4+nQoUKvP3225bzQ4YMYdiwYTbZlZPQPYo0SBACdI7+g1y5ClKq1NucP3+eSZMm0aVNF8rMLUPJN0riMMSkn2MUI62nDePHwEHY32jHyWG+Jidx9Cg0bAgjR5rmJNat05PVmkzDWjyK8uXLU716dUs8igRWrFhhUX2F/+JRgHXNJFtp0aIFh8xy94ULF2bUqFF4eHjQqFEjgoODAdMTf8eOHalfvz7169dn9+7dABw4cIAmTZrw/PPP06RJE86ePQuYnvw7d+7Mm2++aQlIlBre3t5MmzaNoKAgrl27ZpPd3333HVOnTqW8WS8tf/78fPjhh8nyrV271hI3o1OnTmzZssVqwKUEtmzZgrOzcxLNrSeWjG7pzq7X45bw6HTypDju2SM7dxaTU6d6itFolNatW4u9vb2sd1wvexz2SMwtkzjTvQfRUuOrzsJYpNJHA+ROmFnfac8ekdy5RcqWFVm16rHar3k8ZLeER0REhHh4eIiLi4t88sknsm3bNsu5yZMny+DBg0VEZO/evVKvXj3LucqVK8vZs2elcePGIiLi6ekpp06dsmg9WcPX11def/31ZOnNmzeXgwcPiogIIOvWrRMRkc8//1y++eYbERHp3r277Ny5U0RELl++LDVq1BAREYPBILGxsSIismnTJunQoYOIiPz2229SoUIFi86TiIiHh4dVu65cuSJVq1YVEZGRI0fKtGnTLOcKFSqUJO9vv/0mXl5eIiJSvHhxuZugo5YK7u7ucvXqVctxlSpVkulkJaZPnz4ye/Zsy/GYMWOkcuXKUqtWLenTp4/cuXMnzTozipbweIyICLsMBurmDycu7i5ly77DH3/8wcaNG+nv3J9CVwvh6u1KnpJ5uBZ2m8pfvcKZ3CtpcHcq52fNojgRpoIaNoSvvwZ/f+jYMXsbpXkqSYhHMW/ePEqXLk3Xrl0tkuHdunVj1apVGI3GdMejyCh58+a1zGPUrVvXMp6/efNmBgwYgKenJ+3atSM8PJyIiAgMBgOdO3emZs2aDBkyhFOnTlnKatWqFSVKlLAcpzSs5+PjQ5cuXSxtTis8a3rjYoiV3kNKZcTExLBu3To6d/4vVs0nn3zCxYsX8fPzo1y5cnz22Wfpqj870XMUqRBw/z43Y2JwzbOHPHnKkDdvYwYProWbgxuvHH0Fp/FOFGtajMOXLvHi3DZE57tMD7sVLBnXBjXsU/D2Nk1Uly0Lo0Zld3M0TzmPMx5FWuTJk8dyE82dOzdxcXEAGI1G9u7dS4GHlAYGDhxIy5Yt+eOPPwgMDKRFixaWc4UKWddKexhvb2+Cg4NZunQpYIrOd/78eVxcXChQoAAxMTHkzZsXMEmfJ4SLdXd35/Dhw7z00kuplu/g4MDVq1dxcHAgLi4Og8GQxIEl5u+//6ZOnTqULftfaIHE7z/88EOrCwJyKrpHkQoJQoBV7vlQpkx3xo//jqtXr+IV4kWpVqWo9EUlVu87QMP/a0S0usXYKptZWr8oqlZNky5T9+5g45dco3kUHnc8iozy6quv8sMPPySxE8BgMFChQgWADDmrs2fPcu/ePa5du0ZgYCCBgYGMHDnSEgK2efPmLFmyBIDo6GhWrFhBy5YtAVN8jOHDh3Pz5k3AtLpp1qxZyepIHB9j1apVvPTSSyn2KKzN9dy4ccPy/o8//kh1ZVlOQzuKVNhlMGCfK57KnMdgaMq0adN4o+gbeBT3wHWxK5PX/Umn9S2QB4VZ1mw7Y3b8Aq1bQ/78pj0Rs2bBQ8vyNJqsIDIykl69euHm5kbt2rXx9/dn7NixlvOdO3fm1KlTSSaxE1OkSBFGjBhheeJOiy1btuDg4GB5WeulWGPWrFkcOnSI2rVr4+bmxty5cwEYPnw4I0eO5IUXXiA+Pj7VMjw9PZOleXt7J1ldBNCxY0fL8NP333/P77//jqenJ40aNaJz5840a9YMgLZt2+Ll5cUrr7yCu7s7devWtfSAEtO3b19u375N1apVmT59OhMnTgSSxxWPiopi06ZNdOjQIcn1w4cPp1atWtSuXRtfX19mzJiR1r8rx6CsjbvlZOrVqycJKyuympoHDlAsxo9peaYzcmR5Du8+zIKYBbTY3IJhF5bx6/VB5LtTF98P/6Rx7bLQs6cpXvWXX5qcheaZ4fTp05a9CRpNdmPt+6iUOiwi9TJSnp6jSIE7sbGcioqiLzs4cNQDX98VDGYwtb/04O1937E9bgrVr7Ti8B17CtndBspqET+NRvNUoh1FCuw1x/N1jj7J+G9uUD1Xdbo17k6bB6M4n9uHz3e/xKRDB1HR0dC1g0nQTzsJzVPExo0bGTFiRJI0Jycn/vjjj2yySJNdaEeRArvu3iU38RxbGkFIaChf239DT7eviYvZwIFVLtQP2ApNm8L8+SYxP43mKaN169ZZPrmteTLQk9kpsCPsOpXun2PVskBe53WWvPwvF8tu5sdT7akfegPmzIHt27WT0Gg0Tz26R2GFGKORw5ExFPE9RyEpTE3XQmwpvIEptTbT9n/1IDQUKlXKbjM1Go3msZClPQql1GtKqbNKqQtKqS+snM+nlFpuPr9fKeWYlfbYyqHwuzwgN3d3H2RefifGnlvOP4ccGda5qSkkqXYSGo3mGSLLHIVSKjcwB2gDuAHdlVJuD2XrC4SJSFVgBjApq+xJD+su7uX5c+c4tu8oXe8fI+rVdpTw3aQnqzUazTNJVvYoGgAXROSSiMQAPkD7h/K0Bxaa368CXlbpFWDJZOLijZxYsJUDn3xC2dxG7ixaRvENa00yHBpNDiZ37txJYiEEBgYmiR2xYMECcuXKxfHjxy3X1KxZM0lchaNHj6KUYuPGjUnKfjieQ0okjrng5uaWRG+pd+/eODk5WeyztvvZGh4eHsl2OSdWqgUIDAxMstP5wIEDNGvWjOrVq1OjRg0++OADoqKikpX93XffUbVqVapXr56szQls2bKFOnXq4OnpSdOmTblw4QJgkgxPaEu1atUoVqyYTe15EsnKOYoKwNVEx0FAw5TyiEicUsoAlARuJc6klOoH9AOolMXDPnExcRwoU44pzV9hiPcS8pctnaX1aZ4+Bp8/j19kZKaW6Vm4MDNdXFLNU6BAgWSCeQ8H13FwcGD8+PHJ4igk4O3tTdOmTfH29s7wiqeEmAvnz5+nbt26dOrUiTx5TAG9pkyZQqdOnWwu6/Tp0xiNRnbs2MG9e/ds0n0KDg6mc+fO+Pj40LhxY0SE1atXExERkUTw0N/fHx8fH06dOsX169d55ZVXOHfuHLkfCiD2ySefsHbtWlxdXfnxxx/59ttvWbBgQZKd1bNnz+bo0aM2t+tJIyt7FNZ6Bg9vA7clDyIyT0TqiUi90qWz9sadv0BeAocNYsif67ST0Dx1vPHGG5w6dcoS7yExIsKqVatYsGAB//77L/fv33+kulxcXChYsCBhYWEZLmPZsmW8++67vPrqq6xbt86ma+bMmUOvXr1o3LgxYFJ47dSpUxJRPjDFl+jWrRv58uXDycmJqlWrciAh8mQilFKEm/dVGQwGS9yKxDxKHI8ngazsUQQBFRMdOwDXU8gTpJSyA4oCd7LQJpsoVMA2vRuNxhppPflnFdHR0RYdpJQ2xuXKlYvhw4czYcIEi8BdArt378bJyQlnZ2datGjBhg0bkukVpYcjR47g4uJCmTJlLGmff/453377LQCLFy+mVq1afPDBB3z88cfUq5dcXWL58uVs2rSJs2fP8sMPP9h0Mz558qQlwFBqXLt2jUaNGlmOHRwcrAY7mj9/Pm3btqVAgQLY29uzb9++JOcvX75MQEBAmuqzTzJZ2aM4CLgopZyUUnmBbsDDjwTrgIRPtBOwVZ408SmNJoeQMPTk5+eX6u7pHj16sG/fPgICApKke3t7W0QDbYnnkBIzZsygevXqNGzYMIkwIZiGnhJsrFWrFmC6EVtzEgcPHqR06dJUrlyZl19+mSNHjlh6J9amMrMqvsSMGTPYsGEDQUFB9OnTh6FDhyY57+PjQ6dOnZINWT1NZJmjEJE4YACwETgNrBCRU0qpcUqpduZsvwAllVIXgKFAsiW0Go0mc7Gzs+Ozzz5j0qT/FhnGx8ezevVqxo0bh6OjIwMHDuTvv/8mIiIi3eUPGTKEs2fPsnz5ct57770MD2F5e3tz5swZHB0dcXZ2Jjw8nNWrVwNQsmTJJENa1uJLpEVCfIkEgoKCkg0rhYaGcuzYMRo2NE2vdu3alT179iTJYy0Y1NNGlu6jEJENIlJNRJxFZLw5bbSIrDO/vy8inUWkqog0EJFLWWmPRqMx0bt3bzZv3kxoaChgijzn4eHB1atXCQwM5PLly3Ts2JE1a9ZkuI4OHTpQr169ZENctmA0Glm5ciXHjx+3xJdYu3atpZfTokULlixZYukVLFy40BJfYsCAASxcuDBJnPAlS5ZY4k0k0K5dO3x8fHjw4AEBAQGcP3+eBg0aJMlTvHhxDAYD586dA2DTpk1JVFnPnj1LWFiYZT7kaUVLeGg0zyB58+Zl0KBBhISEACnHc1i2bBlgirGQOP7E9OnTbapn9OjRTJ8+HaPRmGKeDz74gIdDB+zYsYMKFSpYghkBNGvWDH9/f27cuEG/fv0oUqQIHh4eeHh4EBkZybBhwwBTJDkfHx+GDRtG9erVcXV1ZefOndjb2yepw93dnS5duuDm5sZrr73GnDlzLMNHbdu25fr169jZ2fF///d/dOzYEQ8PDxYvXsyUKVMsZSQM12Xzqv4sR8ej0GgyAR2PQpOTyOx4FLpHodFoNJpU0aKAGo0m3YwfP56VK1cmSevcuTOjRo3KJos0WYl2FBpNJiEiT/1YdQKjRo3STiGHkhXTCXroSaPJBPLnz8/t27ez5Eeq0diKiHD79m3y58+fqeXqHoVGkwk4ODgQFBRkWW6q0WQX+fPnx8HBIVPL1I5Co8kE8uTJg5OTU3abodFkCXroSaPRaDSpoh2FRqPRaFJFOwqNRqPRpMoTtzNbKRUKXH4MVZXioQBKTzBPU1vg6WrP09QWeLra8zS1BaC6iBTJyIVP3GS2iDyWaEJKqUMZ3e6e03ia2gJPV3ueprbA09Wep6ktYGpPRq/VQ08ajUajSRXtKDQajUaTKtpRpMy87DYgE3ma2gJPV3ueprbA09Wep6kt8AjteeImszUajUbzeNE9Co1Go9GkinYUGo1Go0mVZ95RKKVeU0qdVUpdUEp9YeV8PqXUcvP5/Uopx8dvpW3Y0JahSil/pdRxpdQWpVTl7LDTVtJqT6J8nZRSopTKsUsZbWmLUqqL+fM5pZRa9rhtTA82fNcqKaV8lVJHzd+3ttlhpy0opX5VSoUopU6mcF4ppWaZ23pcKVXncdtoKza0pae5DceVUnuUUh42FSwiz+wLyA1cBKoAeYFjgNtDefoDc83vuwHLs9vuR2hLS6Cg+f0nObUttrbHnK8IsAPYB9TLbrsf4bNxAY4Cxc3HZbLb7kdszzzgE/N7NyAwu+1OpT3NgDrAyRTOtwX+BhTQCNif3TY/QluaJPqOtbG1Lc96j6IBcEFELolIDOADtH8oT3tgofn9KuBllTOj06TZFhHxFZEo8+E+IHO1iDMXWz4bgG+AycD9x2lcOrGlLR8Cc0QkDEBEQh6zjenBlvYIYG9+XxS4/hjtSxcisgO4k0qW9sAiMbEPKKaUKvd4rEsfabVFRPYkfMdIxz3gWXcUFYCriY6DzGlW84hIHGAASj4W69KHLW1JTF9MT0k5lTTbo5R6HqgoIusfp2EZwJbPphpQTSm1Wym1Tyn12mOzLv3Y0p6xwDtKqSBgAzDw8ZiWJaT3t/WkYPM94ImT8MhkrPUMHl4vbEuenIDNdiql3gHqAc2z1KJHI9X2KKVyATOA3o/LoEfAls/GDtPwUwtMT3k7lVI1ReRuFtuWEWxpT3dggYhMU0o1Bhab22PMevMynSflHmAzSqmWmBxFU1vyP+s9iiCgYqJjB5J3kS15lFJ2mLrRqXVTswtb2oJS6hVgFNBORB48JtsyQlrtKQLUBLYppQIxjR2vy6ET2rZ+z9aKSKyIBABnMTmOnIgt7ekLrAAQkb1Afkwie08iNv22nhSUUrWB+UB7EbltyzXPuqM4CLgopZyUUnkxTVaveyjPOqCX+X0nYKuYZ4JyGGm2xTxU8zMmJ5GTx8AhjfaIiEFESomIo4g4YhpvbSciGRY+y0Js+Z6twbTYAKVUKUxDUZceq5W2Y0t7rgAvAyilXDE5iic1Tuw64D3z6qdGgEFEbmS3URlBKVUJ+B14V0TO2Xxhds/SZ/cL04qGc5hWcYwyp43DdNMB0xd8JXDuoD4tAAAEcklEQVQBOABUyW6bH6Etm4FgwM/8WpfdNj9Kex7Ku40cuurJxs9GAdMBf+AE0C27bX7E9rgBuzGtiPIDXs1um1NpizdwA4jF1HvoC3wMfJzos5ljbuuJHP49S6st84GwRPeAQ7aUqyU8NBqNRpMqz/rQk0aj0WjSQDsKjUaj0aSKdhQajUajSRXtKDQajUaTKtpRaDQajSZVtKPQ5DiUUvFKKb9EL8dU8jqmpJSZzjq3mdVQj5llNKpnoIyPlVLvmd/3VkqVT3RuvlLKLZPtPKiU8rThmsFKqYKPWrfm2UU7Ck1OJFpEPBO9Ah9TvT1FxAOTCOSU9F4sInNFZJH5sDdQPtG5D0TEP1Os/M/OH7HNzsGAdhSaDKMdheaJwNxz2KmUOmJ+NbGSx10pdcDcCzmulHIxp7+TKP1npVTuNKrbAVQ1X/uyOabCCbPWfz5z+kT1X2yPqea0sUqpYUqpTpi0tJaa6yxg7gnUU0p9opSanMjm3kqp2Rm0cy+JxOmUUj8ppQ4pUzyLr81pgzA5LF+llK857VWl1F7z/3GlUqpwGvVonnG0o9DkRAokGnb6w5wWArQSkTpAV2CWles+Br4XEU9MN+ogs3xEV+AFc3o80DON+t8ETiil8gMLgK4iUguTcN8nSqkSwNuAu4jUBr5NfLGIrAIOYXry9xSR6ESnVwEdEh13BZZn0M7XMEl/JDBKROoBtYHmSqnaIjILky5RSxFpaZYH+RJ4xfy/PAQMTaMezTPOs64eq8mZRJtvlonJA/xgHpOPx6SF9DB7gVFKKQfgdxE5r5R6GagLHFSmMCIFMDkdayxVSkUDgZhksasDAfKfJs5CwAv4AVP8i/lKqb8Am2XORSRUKXXJrBl03lzHbnO56bGzEKYAQomjrXVRSvXD9Lsuh0lG4/hD1zYyp+8215MX0/9No0kR7Sg0TwpDMOlUeWDqCScLVCQiy5RS+4HXgY1KqQ8w6fQsFJGRNtTRUxKJCiqlrMYdEZE4pVQDTKJ33YABwEvpaMtyoAtwBvhDRESZ7to224lJQ2kiJg2iDkopJ2AYUF9EwpRSCzDplD2MAjaJSPd02Kt5xtFDT5onhaLADTHFM3gX09N0EpRSVYBL5uGWdZiGYLYAnZRSZcx5SijbY4WfARyVUlXNx+8C281j+kVFZAOmiWJrK48iMEmhW+N34C1MMRuWm9PSZaeIxGIaQmpkHrayB+4BBqVUWUxhLq3Zsg94IaFNSqmCSilrvTONxoJ2FJonhR+BXkqpfZiGne5ZydMVOKmU8gNqYApf6Y/phvqvUuo4sAnTsEyaiMh9oA+wUil1AjACczHddNeby9uOqbfzMAuAuQmT2Q+VG4ZJJbayiBwwp6XbTvPcxzRgmIgcwxRz+xTwK6bhrATmAX8rpXxFJBTTiixvcz37MP2vNJoU0eqxGo1Go0kV3aPQaDQaTapoR6HRaDSaVNGOQqPRaDSpoh2FRqPRaFJFOwqNRqPRpIp2FBqNRqNJFe0oNBqNRpMq/w+vn0dfw/56wQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = pickle.load(open(\"final_random_forest.sav\",\"rb\"))\n",
    "result = model.predict(X_test)\n",
    "eval_model(Y_test, result)\n",
    "prediction[\"FINAL_RF\"] = result\n",
    "AuROC(prediction, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 194,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = rf_random.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "metadata": {},
   "outputs": [],
   "source": [
    "features = list(data_final)\n",
    "feature_importances= dict()\n",
    "imps = []\n",
    "from sklearn.feature_selection import SelectFromModel\n",
    "for feature in zip(features, model.feature_importances_):\n",
    "    feature_importances[feature[0]] = feature[1]\n",
    "for key in feature_importances.keys():\n",
    "#     print(key)\n",
    "    imps.append(feature_importances[key])\n",
    "imps = np.array(imps)\n",
    "indexes = np.argsort(imps, axis=-1, kind='quicksort', order=None)\n",
    "sorted_imps = sorted(imps)\n",
    "#pick the top 10 features\n",
    "VALUES = sorted_imps[-15:]\n",
    "feature_indexes = indexes[-15:]\n",
    "imp_features=[]\n",
    "for i in feature_indexes:\n",
    "    imp_features.append(features[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 199,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cd91352940>"
      ]
     },
     "execution_count": 199,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.Series(index = imp_features, data = VALUES).plot(kind='barh', title='Variable Importances', figsize = (10,5))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}