finalNotebook.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Dataset Information\n",
    "Our training dataset contains information from 12,165 hours over 731 days with 14 features (information categories) and 1 target for each hour. The file named train.csv contains the training set of the problem, while the file test.csv contains the test set of the problem. Each line of these two files corresponds to a bike sharing record characterized by the following variables:\n",
    "\n",
    "Variable | Description --- | --- | --- season | season (1:winter, 2:spring, 3:summer, 4:fall) yr | year (0: 2011, 1:2012) mnth | month (1 to 12) hr | hour (0 to 23) holiday | weather day is holiday or not weekday|day of the week workingday | if day is neither weekend nor holiday is 1, otherwise is 0. weathersit | 1: Clear, Few clouds, Partly cloudy, Partly cloudy | 2: Mist + Cloudy, Mist + Broken clouds, Mist + Few clouds, Mist | 3: Light Snow, Light Rain + Thunderstorm + Scattered clouds, Light Rain + Scattered clouds | 4: Heavy Rain + Ice Pallets + Thunderstorm + Mist, Snow + Fog temp | Normalized temperature in Celsius. The values are derived via (t-t_min)/(t_max-t_min), t_min=-8, t_max=+39 (only in hourly scale) atemp | Normalized feeling temperature in Celsius. The values are derived via (t-t_min)/(t_max-t_min), t_min=-16, t_max=+50 (only in hourly scale) hum | Normalized humidity. The values are divided to 100 (max) windspeed | Normalized wind speed. The values are divided to 67 (max) casual | count of casual users registered | count of registered users cnt | count of total rental bikes including both casual and registered\n",
    "\n",
    "The training set contains the variable cnt which is related to the number of bicycles rented in any given 1 hour time-period. The cnt variable is missing from the test set as it is the variable to be predicted.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Importing the libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing the libraries\n",
    "\n",
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import seaborn as sn\n",
    "from scipy import stats\n",
    "from numpy import median\n",
    "from sklearn import metrics\n",
    "from sklearn import model_selection\n",
    "from sklearn.metrics import mean_squared_log_error, r2_score\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.linear_model import RidgeCV\n",
    "from sklearn.svm import LinearSVR\n",
    "import xgboost as xgb\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "The code that loads the training data from file train.csv into a DataFrame of the Pandas library and prints the first 5 lines, is given below. So you can see the 15 variables of the problem."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>season</th>\n",
       "      <th>yr</th>\n",
       "      <th>mnth</th>\n",
       "      <th>hr</th>\n",
       "      <th>holiday</th>\n",
       "      <th>weekday</th>\n",
       "      <th>workingday</th>\n",
       "      <th>weathersit</th>\n",
       "      <th>temp</th>\n",
       "      <th>atemp</th>\n",
       "      <th>hum</th>\n",
       "      <th>windspeed</th>\n",
       "      <th>casual</th>\n",
       "      <th>registered</th>\n",
       "      <th>cnt</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.56</td>\n",
       "      <td>0.5303</td>\n",
       "      <td>0.64</td>\n",
       "      <td>0.3582</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.2727</td>\n",
       "      <td>0.65</td>\n",
       "      <td>0.2537</td>\n",
       "      <td>9</td>\n",
       "      <td>59</td>\n",
       "      <td>68</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.44</td>\n",
       "      <td>0.4394</td>\n",
       "      <td>0.41</td>\n",
       "      <td>0.1940</td>\n",
       "      <td>74</td>\n",
       "      <td>270</td>\n",
       "      <td>344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.1970</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.1642</td>\n",
       "      <td>3</td>\n",
       "      <td>219</td>\n",
       "      <td>222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.2727</td>\n",
       "      <td>0.45</td>\n",
       "      <td>0.2985</td>\n",
       "      <td>30</td>\n",
       "      <td>180</td>\n",
       "      <td>210</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0.20</td>\n",
       "      <td>0.2273</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.0896</td>\n",
       "      <td>1</td>\n",
       "      <td>19</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.66</td>\n",
       "      <td>0.6212</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.2239</td>\n",
       "      <td>203</td>\n",
       "      <td>260</td>\n",
       "      <td>463</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.5455</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.1642</td>\n",
       "      <td>1</td>\n",
       "      <td>55</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.48</td>\n",
       "      <td>0.2836</td>\n",
       "      <td>46</td>\n",
       "      <td>120</td>\n",
       "      <td>166</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>14</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.40</td>\n",
       "      <td>0.4091</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.1642</td>\n",
       "      <td>14</td>\n",
       "      <td>126</td>\n",
       "      <td>140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.33</td>\n",
       "      <td>0.2537</td>\n",
       "      <td>109</td>\n",
       "      <td>324</td>\n",
       "      <td>433</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>14</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.64</td>\n",
       "      <td>0.6212</td>\n",
       "      <td>0.27</td>\n",
       "      <td>0.2836</td>\n",
       "      <td>59</td>\n",
       "      <td>198</td>\n",
       "      <td>257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>22</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.3485</td>\n",
       "      <td>0.40</td>\n",
       "      <td>0.2239</td>\n",
       "      <td>15</td>\n",
       "      <td>92</td>\n",
       "      <td>107</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.80</td>\n",
       "      <td>0.7273</td>\n",
       "      <td>0.43</td>\n",
       "      <td>0.2239</td>\n",
       "      <td>89</td>\n",
       "      <td>346</td>\n",
       "      <td>435</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.2424</td>\n",
       "      <td>0.55</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>3</td>\n",
       "      <td>13</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    season  yr  mnth  hr  holiday  weekday  workingday  weathersit  temp  \\\n",
       "0        4   0    11   0        0        2           1           1  0.56   \n",
       "1        1   0     2  21        0        6           0           1  0.28   \n",
       "2        4   1    11  12        0        3           1           1  0.44   \n",
       "3        1   1     1   7        0        2           1           2  0.18   \n",
       "4        1   1     3  16        0        1           1           1  0.30   \n",
       "5        1   0     2   2        0        0           0           2  0.20   \n",
       "6        2   1     6  15        0        0           0           1  0.66   \n",
       "7        3   0     9   6        0        4           1           3  0.62   \n",
       "8        4   0    10  11        0        2           1           1  0.52   \n",
       "9        1   1     2  14        0        1           1           1  0.40   \n",
       "10       3   1     7  16        0        4           1           1  0.92   \n",
       "11       2   1     4  14        0        2           1           1  0.64   \n",
       "12       1   1     2  22        0        5           1           1  0.36   \n",
       "13       3   1     7  16        0        2           1           1  0.80   \n",
       "14       1   0     1   2        0        6           0           2  0.18   \n",
       "\n",
       "     atemp   hum  windspeed  casual  registered  cnt  \n",
       "0   0.5303  0.64     0.3582       7          15   22  \n",
       "1   0.2727  0.65     0.2537       9          59   68  \n",
       "2   0.4394  0.41     0.1940      74         270  344  \n",
       "3   0.1970  0.93     0.1642       3         219  222  \n",
       "4   0.2727  0.45     0.2985      30         180  210  \n",
       "5   0.2273  0.75     0.0896       1          19   20  \n",
       "6   0.6212  0.50     0.2239     203         260  463  \n",
       "7   0.5455  0.94     0.1642       1          55   56  \n",
       "8   0.5000  0.48     0.2836      46         120  166  \n",
       "9   0.4091  0.35     0.1642      14         126  140  \n",
       "10  0.8333  0.33     0.2537     109         324  433  \n",
       "11  0.6212  0.27     0.2836      59         198  257  \n",
       "12  0.3485  0.40     0.2239      15          92  107  \n",
       "13  0.7273  0.43     0.2239      89         346  435  \n",
       "14  0.2424  0.55     0.0000       3          13   16  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Reading training data from the local storage\n",
    "filename = 'dataset/train.csv'\n",
    "df_train = pd.read_csv(filename)\n",
    "df_train.head(15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Also, as for the dimensions of the training set, it consists of 12165 lines and 15 columns. That is, you have the data from 12165 bike sharing records for which you know the number of bicycles rented cnt."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(12165, 15)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Instead, the file named test.csv contains the test set of the problem. As with the training set, you are now given code that loads the test set into a DataFrame of the Pandas library and prints its first 5 lines."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>season</th>\n",
       "      <th>yr</th>\n",
       "      <th>mnth</th>\n",
       "      <th>hr</th>\n",
       "      <th>holiday</th>\n",
       "      <th>weekday</th>\n",
       "      <th>workingday</th>\n",
       "      <th>weathersit</th>\n",
       "      <th>temp</th>\n",
       "      <th>atemp</th>\n",
       "      <th>hum</th>\n",
       "      <th>windspeed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.26</td>\n",
       "      <td>0.2121</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.4478</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.42</td>\n",
       "      <td>0.4478</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.2121</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.2836</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.74</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.2239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.2424</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   season  yr  mnth  hr  holiday  weekday  workingday  weathersit  temp  \\\n",
       "0       1   1     1  20        0        4           1           1  0.26   \n",
       "1       4   1    10  15        0        5           1           1  0.52   \n",
       "2       1   0     1  10        0        0           0           1  0.22   \n",
       "3       3   1     8  18        0        3           1           1  0.74   \n",
       "4       1   1     2   8        0        1           1           1  0.24   \n",
       "\n",
       "    atemp   hum  windspeed  \n",
       "0  0.2121  0.52     0.4478  \n",
       "1  0.5000  0.42     0.4478  \n",
       "2  0.2121  0.51     0.2836  \n",
       "3  0.6667  0.51     0.2239  \n",
       "4  0.2424  0.60     0.0000  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test = pd.read_csv('dataset/test.csv')\n",
    "df_test.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As for the dimensions of the test set, as you can see below, it consists of 5214 lines and 12 columns. The missing columns are about the cnt variable that we are aiming to predict,the casual and registered variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5214, 12)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see the distribution of continuous variables so we have an idea of some basic statistical details."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>temp</th>\n",
       "      <th>atemp</th>\n",
       "      <th>hum</th>\n",
       "      <th>windspeed</th>\n",
       "      <th>casual</th>\n",
       "      <th>registered</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>12165.00000</td>\n",
       "      <td>12165.000000</td>\n",
       "      <td>12165.000000</td>\n",
       "      <td>12165.000000</td>\n",
       "      <td>12165.000000</td>\n",
       "      <td>12165.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.49612</td>\n",
       "      <td>0.474749</td>\n",
       "      <td>0.627587</td>\n",
       "      <td>0.190263</td>\n",
       "      <td>35.563337</td>\n",
       "      <td>152.837731</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.19240</td>\n",
       "      <td>0.171742</td>\n",
       "      <td>0.193743</td>\n",
       "      <td>0.122868</td>\n",
       "      <td>49.246702</td>\n",
       "      <td>150.233116</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.02000</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>0.34000</td>\n",
       "      <td>0.333300</td>\n",
       "      <td>0.480000</td>\n",
       "      <td>0.104500</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>34.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.50000</td>\n",
       "      <td>0.484800</td>\n",
       "      <td>0.630000</td>\n",
       "      <td>0.194000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>116.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.64000</td>\n",
       "      <td>0.621200</td>\n",
       "      <td>0.790000</td>\n",
       "      <td>0.253700</td>\n",
       "      <td>48.000000</td>\n",
       "      <td>218.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.850700</td>\n",
       "      <td>362.000000</td>\n",
       "      <td>886.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              temp         atemp           hum     windspeed        casual  \\\n",
       "count  12165.00000  12165.000000  12165.000000  12165.000000  12165.000000   \n",
       "mean       0.49612      0.474749      0.627587      0.190263     35.563337   \n",
       "std        0.19240      0.171742      0.193743      0.122868     49.246702   \n",
       "min        0.02000      0.000000      0.000000      0.000000      0.000000   \n",
       "25%        0.34000      0.333300      0.480000      0.104500      4.000000   \n",
       "50%        0.50000      0.484800      0.630000      0.194000     17.000000   \n",
       "75%        0.64000      0.621200      0.790000      0.253700     48.000000   \n",
       "max        1.00000      1.000000      1.000000      0.850700    362.000000   \n",
       "\n",
       "         registered  \n",
       "count  12165.000000  \n",
       "mean     152.837731  \n",
       "std      150.233116  \n",
       "min        0.000000  \n",
       "25%       34.000000  \n",
       "50%      116.000000  \n",
       "75%      218.000000  \n",
       "max      886.000000  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train[['temp','atemp','hum','windspeed','casual','registered']].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluation - RMSLE Score"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One common way to evaluate a regression model is through calculating Mean Squared Error (MSE) or Root Mean Squared Error (RMSE). \n",
    "\n",
    "$$ RMSE = \\sqrt{\\frac{1}{N} \\sum_{i=1}^N (\\hat{y_i} - y_i)^2} $$\n",
    "\n",
    "In this particular competition, the metric to evaluate your model is **Root Mean Square Logarithmic Error** (RMSLE). It is particularly helpful when we want to penalize an under-predicted estimate greater than an over-predicted estimate. If $\\hat{y_i}$ is the predicted value of the $I$-th sample, and ${y_i}$ is the corresponding true value, then the RMSLE estimated over $N$ samples is defined as:\n",
    "\n",
    "$$ RMSLE = \\sqrt{\\frac{1}{N} \\sum_{i=1}^N (\\log(1+\\hat{y_i}) - \\log(1+y_i))^2 } $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Goal"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As mentioned above, your goal is to predict the number of bicycles (variable `cnt`) rented for each 1 hour period-time of the test set. This is a supervised learning problem. You will need to preprocess the training set data, and then use it to train a regressor, which you will use to make predictions for the test set data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Variable Engineering\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check existence of null values in the dataset:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "season        0\n",
       "yr            0\n",
       "mnth          0\n",
       "hr            0\n",
       "holiday       0\n",
       "weekday       0\n",
       "workingday    0\n",
       "weathersit    0\n",
       "temp          0\n",
       "atemp         0\n",
       "hum           0\n",
       "windspeed     0\n",
       "casual        0\n",
       "registered    0\n",
       "cnt           0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It seems that no null value is present in the dataset.\n",
    "\n",
    "Lets change column names for a nice precise reading:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train.rename(columns={'weathersit':'weather',\n",
    "                     'mnth':'month',\n",
    "                     'hr':'hour',\n",
    "                     'yr':'year',\n",
    "                     'hum': 'humidity',\n",
    "                     'cnt':'count'},inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>season</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>hour</th>\n",
       "      <th>holiday</th>\n",
       "      <th>weekday</th>\n",
       "      <th>workingday</th>\n",
       "      <th>weather</th>\n",
       "      <th>temp</th>\n",
       "      <th>atemp</th>\n",
       "      <th>humidity</th>\n",
       "      <th>windspeed</th>\n",
       "      <th>casual</th>\n",
       "      <th>registered</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.56</td>\n",
       "      <td>0.5303</td>\n",
       "      <td>0.64</td>\n",
       "      <td>0.3582</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.2727</td>\n",
       "      <td>0.65</td>\n",
       "      <td>0.2537</td>\n",
       "      <td>9</td>\n",
       "      <td>59</td>\n",
       "      <td>68</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.44</td>\n",
       "      <td>0.4394</td>\n",
       "      <td>0.41</td>\n",
       "      <td>0.1940</td>\n",
       "      <td>74</td>\n",
       "      <td>270</td>\n",
       "      <td>344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.1970</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.1642</td>\n",
       "      <td>3</td>\n",
       "      <td>219</td>\n",
       "      <td>222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.2727</td>\n",
       "      <td>0.45</td>\n",
       "      <td>0.2985</td>\n",
       "      <td>30</td>\n",
       "      <td>180</td>\n",
       "      <td>210</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   season  year  month  hour  holiday  weekday  workingday  weather  temp  \\\n",
       "0       4     0     11     0        0        2           1        1  0.56   \n",
       "1       1     0      2    21        0        6           0        1  0.28   \n",
       "2       4     1     11    12        0        3           1        1  0.44   \n",
       "3       1     1      1     7        0        2           1        2  0.18   \n",
       "4       1     1      3    16        0        1           1        1  0.30   \n",
       "\n",
       "    atemp  humidity  windspeed  casual  registered  count  \n",
       "0  0.5303      0.64     0.3582       7          15     22  \n",
       "1  0.2727      0.65     0.2537       9          59     68  \n",
       "2  0.4394      0.41     0.1940      74         270    344  \n",
       "3  0.1970      0.93     0.1642       3         219    222  \n",
       "4  0.2727      0.45     0.2985      30         180    210  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_test.rename(columns={'weathersit':'weather',\n",
    "                     'mnth':'month',\n",
    "                     'hr':'hour',\n",
    "                     'yr':'year',\n",
    "                     'hum': 'humidity',\n",
    "                     'cnt':'count'},inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check data type for each column:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>season</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>hour</th>\n",
       "      <th>holiday</th>\n",
       "      <th>weekday</th>\n",
       "      <th>workingday</th>\n",
       "      <th>weather</th>\n",
       "      <th>temp</th>\n",
       "      <th>atemp</th>\n",
       "      <th>humidity</th>\n",
       "      <th>windspeed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.26</td>\n",
       "      <td>0.2121</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.4478</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.42</td>\n",
       "      <td>0.4478</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.2121</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.2836</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.74</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.2239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.2424</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   season  year  month  hour  holiday  weekday  workingday  weather  temp  \\\n",
       "0       1     1      1    20        0        4           1        1  0.26   \n",
       "1       4     1     10    15        0        5           1        1  0.52   \n",
       "2       1     0      1    10        0        0           0        1  0.22   \n",
       "3       3     1      8    18        0        3           1        1  0.74   \n",
       "4       1     1      2     8        0        1           1        1  0.24   \n",
       "\n",
       "    atemp  humidity  windspeed  \n",
       "0  0.2121      0.52     0.4478  \n",
       "1  0.5000      0.42     0.4478  \n",
       "2  0.2121      0.51     0.2836  \n",
       "3  0.6667      0.51     0.2239  \n",
       "4  0.2424      0.60     0.0000  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "season          int64\n",
       "year            int64\n",
       "month           int64\n",
       "hour            int64\n",
       "holiday         int64\n",
       "weekday         int64\n",
       "workingday      int64\n",
       "weather         int64\n",
       "temp          float64\n",
       "atemp         float64\n",
       "humidity      float64\n",
       "windspeed     float64\n",
       "casual          int64\n",
       "registered      int64\n",
       "count           int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Outliers"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We tried to cut the outliers  from df_train according to one function that calculates the difference between the absolute value of variable 'count' and the average of value and then checks if this result is less than 3*(standard deviaton of 'count'). We run the following 3 lines of code and data was decreased from 12165 to 11989 so we applied our model on this data.\n",
    "- df_trainWithoutOutliers = df_train[np.abs(df_train[\"count\"]-df_train[\"count\"].mean())<=(3*df_train[\"count\"].std())]\n",
    "- print (\"Shape Of The Before Ouliers: \",df_train.shape)\n",
    "- print (\"Shape Of The After Ouliers: \",df_trainWithoutOutliers.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data Exploration "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Categorical Features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Season "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3    3130\n",
       "2    3090\n",
       "4    2988\n",
       "1    2957\n",
       "Name: season, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.season.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can infer that people rent more bikes in Summer and Spring than in Winter and Autumn from the above function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Anaconda\\lib\\site-packages\\seaborn\\categorical.py:3666: UserWarning: The `factorplot` function has been renamed to `catplot`. The original name will be removed in a future release. Please update your code. Note that the default `kind` in `factorplot` (`'point'`) has changed `'strip'` in `catplot`.\n",
      "  warnings.warn(msg)\n",
      "D:\\Anaconda\\lib\\site-packages\\seaborn\\categorical.py:3672: UserWarning: The `size` paramter has been renamed to `height`; please update your code.\n",
      "  warnings.warn(msg, UserWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x251c1605ac8>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAFgCAYAAACbqJP/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAUm0lEQVR4nO3df7Bc5X3f8ffHgOMkJgEH4WJJKYwrp8ZOLWyNICXTODgFQdqIpKaB1rZMaOR2IGPPeNrBmalx7DBN4l+NGYeOEmQgcUyZYtcqpRCVEDNxakA44oeQKbeYGFkqulgYO/WUVOTbP/a5kzW6ulrJ2rv7cN+vmZ0953ues/u9O6PPHj179myqCklSP14y6QYkSYfH4JakzhjcktQZg1uSOmNwS1Jnjp10A+Owbt26uv322yfdhiR9rzJf8UV5xP30009PugVJGpsXZXBL0ouZwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzrwoL+sqTdrZ15w96RYW3Rd/5YuTbmHJ8IhbkjpjcEtSZwxuSeqMc9waydc++OOTbmHR/ej7H5p0C9K8POKWpM4Y3JLUGYNbkjpjcEtSZ8YW3EleluTeJA8k2ZHk11r9tCT3JHksyX9M8tJW/762PtO2nzr0WO9r9UeTnDeuniWpB+M84n4OOKeq3gCsBtYlOQv4TeDjVbUKeAa4rI2/DHimqv4O8PE2jiSnAxcDrwPWAb+T5Jgx9i1JU21swV0Df9lWj2u3As4B/lOr3wBc2JbXt3Xa9rckSavfVFXPVdVXgRlg7bj6lqRpN9Y57iTHJNkO7AW2Av8L+GZV7W9DdgHL2/Jy4EmAtv1Z4EeG6/PsM/xcG5NsS7JtdnZ2HH+OJE2FsX4Bp6qeB1YnOQH4HPDa+Ya1+xxk28HqL3yuTcAmgDVr1hywXdJ0+8I/+KlJt7DofuruLxzRfotyVklVfRP4E+As4IQkc28YK4DdbXkXsBKgbf9hYN9wfZ59JGnJGedZJcvakTZJvh/4GWAncBfw1jZsA/D5trylrdO2/3FVVatf3M46OQ1YBdw7rr4ladqNc6rkFOCGdgbIS4Cbq+rWJI8ANyX5deDPgeva+OuA308yw+BI+2KAqtqR5GbgEWA/cHmbgjlib/rXN34vu3fp/g+/Y9ItSDpKxhbcVfUgcMY89ceZ56yQqvq/wEUHeayrgauPdo+S1CO/OSlJnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOjO24E6yMsldSXYm2ZHk3a3+gSRfT7K93S4Y2ud9SWaSPJrkvKH6ulabSXLluHqWpB4cO8bH3g+8t6q+nOR44P4kW9u2j1fVR4YHJzkduBh4HfAq4L8neU3b/EngHwK7gPuSbKmqR8bYuyRNrbEFd1XtAfa05W8n2QksX2CX9cBNVfUc8NUkM8Datm2mqh4HSHJTG2twS1qSFmWOO8mpwBnAPa10RZIHk2xOcmKrLQeeHNptV6sdrP7C59iYZFuSbbOzs0f5L5Ck6TH24E7ycuAW4D1V9S3gWuDVwGoGR+QfnRs6z+61QP27C1WbqmpNVa1ZtmzZUeldkqbROOe4SXIcg9D+dFV9FqCqnhra/rvArW11F7ByaPcVwO62fLC6JC054zyrJMB1wM6q+thQ/ZShYT8PPNyWtwAXJ/m+JKcBq4B7gfuAVUlOS/JSBh9gbhlX35I07cZ5xH028HbgoSTbW+1XgUuSrGYw3fEE8C6AqtqR5GYGHzruBy6vqucBklwB3AEcA2yuqh1j7FuSpto4zyr5U+afn75tgX2uBq6ep37bQvtJ0lLiNyclqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmfGFtxJVia5K8nOJDuSvLvVX5Fka5LH2v2JrZ4kn0gyk+TBJG8ceqwNbfxjSTaMq2dJ6sE4j7j3A++tqtcCZwGXJzkduBK4s6pWAXe2dYDzgVXtthG4FgZBD1wFnAmsBa6aC3tJWorGFtxVtaeqvtyWvw3sBJYD64Eb2rAbgAvb8nrgxhr4EnBCklOA84CtVbWvqp4BtgLrxtW3JE27RZnjTnIqcAZwD/DKqtoDg3AHTm7DlgNPDu22q9UOVn/hc2xMsi3JttnZ2aP9J0jS1Bh7cCd5OXAL8J6q+tZCQ+ep1QL17y5UbaqqNVW1ZtmyZUfWrCR1YKzBneQ4BqH96ar6bCs/1aZAaPd7W30XsHJo9xXA7gXqkrQkjfOskgDXATur6mNDm7YAc2eGbAA+P1R/Rzu75Czg2TaVcgdwbpIT24eS57aaJC1Jx47xsc8G3g48lGR7q/0q8BvAzUkuA74GXNS23QZcAMwA3wEuBaiqfUk+BNzXxn2wqvaNsW9JmmpjC+6q+lPmn58GeMs84wu4/CCPtRnYfPS6k6R++c1JSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnRgruJHeOUpMkjd+Cv4CT5GXADwAntd97nPtFmx8CXjXm3iRJ8zjUT5e9C3gPg5C+n78J7m8BnxxjX5Kkg1gwuKvqt4HfTvIrVXXNIvUkSVrASD8WXFXXJPn7wKnD+1TVjWPqS5J0ECMFd5LfB14NbAeeb+UCDG5JWmQjBTewBji9qmqczUiSDm3U87gfBv7WOBuRJI1m1CPuk4BHktwLPDdXrKqfG0tXkqSDGjW4PzDOJiRJoxv1rJIvjLsRSdJoRj2r5NsMziIBeClwHPB/quqHxtWYJGl+ox5xHz+8nuRCYO1YOpIkLeiIrg5YVf8ZOOco9yJJGsGoUyW/MLT6EgbndXtOtyRNwKhnlfzjoeX9wBPA+qPejSTpkEad47503I1IkkYz6g8prEjyuSR7kzyV5JYkK8bdnCTpQKN+OPkpYAuD63IvB/5Lq0mSFtmowb2sqj5VVfvb7Xpg2Rj7kiQdxKjB/XSStyU5pt3eBnxjnI1JkuY3anD/EvBPgf8N7AHeCiz4gWWSzW1O/OGh2geSfD3J9na7YGjb+5LMJHk0yXlD9XWtNpPkysP54yTpxWjU4P4QsKGqllXVyQyC/AOH2Od6YN089Y9X1ep2uw0gyenAxcDr2j6/M3d0z+C3Lc8HTgcuaWMlackaNbj/XlU9M7dSVfuAMxbaoaruBvaN+PjrgZuq6rmq+ioww+Ar9WuBmap6vKr+CrgJzx+XtMSNGtwvSXLi3EqSVzD6l3de6IokD7aplLnHXA48OTRmV6sdrH6AJBuTbEuybXZ29ghbk6TpN2pwfxT4syQfSvJB4M+A3zqC57uWwW9XrmYwV/7RVs88Y2uB+oHFqk1Vtaaq1ixb5gkvkl68Rv3m5I1JtjG4sFSAX6iqRw73yarqqbnlJL8L3NpWdwErh4auAHa35YPVJWlJGnm6owX1YYf1sCSnVNWetvrzDH7LEgZf7vnDJB9j8CWfVcC9DN4kViU5Dfg6gw8w/9n30oMk9e5I56kPKclngDcDJyXZBVwFvDnJagbTHU8A7wKoqh1JbmbwxrAfuLyqnm+PcwVwB3AMsLmqdoyrZ0nqwdiCu6oumad83QLjrwaunqd+G3DbUWxNkrp2RD+kIEmaHINbkjpjcEtSZwxuSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpM2ML7iSbk+xN8vBQ7RVJtiZ5rN2f2OpJ8okkM0keTPLGoX02tPGPJdkwrn4lqRfjPOK+Hlj3gtqVwJ1VtQq4s60DnA+sareNwLUwCHrgKuBMYC1w1VzYS9JSNbbgrqq7gX0vKK8HbmjLNwAXDtVvrIEvASckOQU4D9haVfuq6hlgKwe+GUjSkrLYc9yvrKo9AO3+5FZfDjw5NG5Xqx2sfoAkG5NsS7Jtdnb2qDcuSdNiWj6czDy1WqB+YLFqU1Wtqao1y5YtO6rNSdI0WezgfqpNgdDu97b6LmDl0LgVwO4F6pK0ZC12cG8B5s4M2QB8fqj+jnZ2yVnAs20q5Q7g3CQntg8lz201SVqyjh3XAyf5DPBm4KQkuxicHfIbwM1JLgO+BlzUht8GXADMAN8BLgWoqn1JPgTc18Z9sKpe+IGnJC0pYwvuqrrkIJveMs/YAi4/yONsBjYfxdYkqWvT8uGkJGlEBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnJhLcSZ5I8lCS7Um2tdorkmxN8li7P7HVk+QTSWaSPJjkjZPoWZKmxSSPuH+6qlZX1Zq2fiVwZ1WtAu5s6wDnA6vabSNw7aJ3KklTZJqmStYDN7TlG4ALh+o31sCXgBOSnDKJBiVpGkwquAv4oyT3J9nYaq+sqj0A7f7kVl8OPDm0765W+y5JNibZlmTb7OzsGFuXpMk6dkLPe3ZV7U5yMrA1yVcWGJt5anVAoWoTsAlgzZo1B2yXpBeLiRxxV9Xudr8X+BywFnhqbgqk3e9tw3cBK4d2XwHsXrxuJWm6LHpwJ/nBJMfPLQPnAg8DW4ANbdgG4PNteQvwjnZ2yVnAs3NTKpK0FE1iquSVwOeSzD3/H1bV7UnuA25OchnwNeCiNv424AJgBvgOcOnityxJ02PRg7uqHgfeME/9G8Bb5qkXcPkitCZJXZim0wElSSMwuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjrTTXAnWZfk0SQzSa6cdD+SNCldBHeSY4BPAucDpwOXJDl9sl1J0mR0EdzAWmCmqh6vqr8CbgLWT7gnSZqIVNWkezikJG8F1lXVv2jrbwfOrKorhsZsBDa21R8DHl30RkdzEvD0pJvoiK/X4fM1OzzT/Ho9XVXrXlg8dhKdHIHMU/uud5yq2gRsWpx2jlySbVW1ZtJ99MLX6/D5mh2eHl+vXqZKdgErh9ZXALsn1IskTVQvwX0fsCrJaUleClwMbJlwT5I0EV1MlVTV/iRXAHcAxwCbq2rHhNs6UlM/nTNlfL0On6/Z4enu9eriw0lJ0t/oZapEktQY3JLUGYN7kSTZnGRvkocn3UsPkqxMcleSnUl2JHn3pHuaZkleluTeJA+01+vXJt1TL5Ick+TPk9w66V5GZXAvnuuBA06k10HtB95bVa8FzgIu9zIHC3oOOKeq3gCsBtYlOWvCPfXi3cDOSTdxOAzuRVJVdwP7Jt1HL6pqT1V9uS1/m8E/rOWT7Wp61cBfttXj2s0zDw4hyQrgZ4Hfm3Qvh8Pg1tRLcipwBnDPZDuZbu2//NuBvcDWqvL1OrR/D/wb4K8n3cjhMLg11ZK8HLgFeE9VfWvS/Uyzqnq+qlYz+Gbx2iSvn3RP0yzJPwL2VtX9k+7lcBncmlpJjmMQ2p+uqs9Oup9eVNU3gT/Bz1QO5Wzg55I8weCKo+ck+YPJtjQag1tTKUmA64CdVfWxSfcz7ZIsS3JCW/5+4GeAr0y2q+lWVe+rqhVVdSqDy2j8cVW9bcJtjcTgXiRJPgP8D+DHkuxKctmke5pyZwNvZ3AUtL3dLph0U1PsFOCuJA8yuLbP1qrq5vQ2HR6/8i5JnfGIW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcWpKS/GCS/9quX/1wkl9M8qYkX0hyf5I7kpzSxv5ykvva2FuS/ECrX9T2fSDJ3a32siSfSvJQu8bzT7f6O5N8NsntSR5L8luT++vVO7+AoyUpyT8B1lXVL7f1Hwb+G7C+qmaT/CJwXlX9UpIfqapvtHG/DjxVVdckeag9xteTnFBV30zyXuD1VXVpkr8L/BHwGgZfqX4/g6scPgc8CvxkVT25yH+6XgS6+JV3aQweAj6S5DeBW4FngNcDWweXSeEYYE8b+/oW2CcALwfuaPUvAtcnuRmYuwjWTwLXAFTVV5L8BYPgBrizqp4FSPII8LcBg1uHzeDWklRV/zPJm4ALgH8HbAV2VNVPzDP8euDCqnogyTuBN7fH+JdJzmRwIf7tSVYDWeBpnxtafh7//ekIOcetJSnJq4DvVNUfAB8BzgSWJfmJtv24JK9rw48H9rTLzP7zocd4dVXdU1XvB54GVgJ3z41J8hrgRxlMi0hHje/4Wqp+HPhwkr8G/h/wrxj8zuUn2nz3sQx+HWUH8G8Z/PrOXzCYYjm+PcaHk6xicJR9J/AAg0up/oc2/70feGdVPdemX6Sjwg8nJakzTpVIUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktSZ/w+8QQ8x4nquLwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sn.factorplot(x='season',data=df_train,kind='count',size=5,aspect=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Year"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    6104\n",
       "0    6061\n",
       "Name: year, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.year.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "More rentals at second year (2012)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x251c168a390>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAFgCAYAAACbqJP/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAATPUlEQVR4nO3df6zd9V3H8ecLOpxubpTtgtjWgNro8Md+XQFdYnSYUvBH0ciCcXJlTaoJ6kyMyoyxCiNq/DG3RYnN6FYWlZHNSTUL2HSbi1EYZSDjh6QVF7gp0sta0Em2BXz7x/nUHcq9t2fY77330/t8JDfnfD/fzzn3c5Lm2W++53vOTVUhSerHKcu9AEnSV8dwS1JnDLckdcZwS1JnDLckdWbNci9gCJs3b67bbrttuZchSf9fmW/wpDzifvLJJ5d7CZI0mJMy3JJ0MjPcktQZwy1JnTHcktQZwy1JnTHcktQZwy1JnTHcktQZwy1JnTHcktQZwy1JnTHcktQZwy1JnTkpv9ZVWukevfa7lnsJWgLf9FufHeR5PeKWpM4MesSd5HTgfcB3AgW8DXgY+BBwDvA54C1VdSRJgHcDlwLPAD9bVZ9pzzMD/GZ72ndW1a4h1/3GX71pyKfXCnH3H1y53EuQXpShj7jfDdxWVd8OvBZ4CLgG2FtVG4G9bRvgEmBj+9kG3ACQ5AxgO3ABcD6wPcnagdctSSvWYOFO8grg+4EbAarqy1X1FLAFOHrEvAu4rN3fAtxUI3cApyc5G7gY2FNVh6vqCLAH2DzUuiVppRvyiPubgTng/UnuSfK+JC8DzqqqxwHa7Zlt/jrgsbHHz7axhcafJ8m2JPuS7Jubmzvxr0aSVoghw70GeANwQ1W9HvhvvnJaZD7z/VHMWmT8+QNVO6pquqqmp6amXsx6JakLQ4Z7Fpitqjvb9ocZhfyJdgqEdntobP6GscevBw4uMi5Jq9Jg4a6q/wAeS/Jtbegi4EFgNzDTxmaAW9v93cCVGbkQeLqdSrkd2JRkbXtTclMbk6RVaegP4Pwi8BdJTgMeAa5i9J/FLUm2Ao8Cl7e5H2N0KeABRpcDXgVQVYeTXAfc1eZdW1WHB163JK1Yg4a7qu4FpufZddE8cwu4eoHn2QnsPLGrk6Q++clJSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSerMoOFO8rkkn01yb5J9beyMJHuS7G+3a9t4krwnyYEk9yV5w9jzzLT5+5PMDLlmSVrpluKI+wer6nVVNd22rwH2VtVGYG/bBrgE2Nh+tgE3wCj0wHbgAuB8YPvR2EvSarQcp0q2ALva/V3AZWPjN9XIHcDpSc4GLgb2VNXhqjoC7AE2L/WiJWmlGDrcBfx9kruTbGtjZ1XV4wDt9sw2vg54bOyxs21sofHnSbItyb4k++bm5k7wy5CklWPNwM//pqo6mORMYE+Sf11kbuYZq0XGnz9QtQPYATA9Pf2C/ZJ0shj0iLuqDrbbQ8BHGZ2jfqKdAqHdHmrTZ4ENYw9fDxxcZFySVqXBwp3kZUm+/uh9YBNwP7AbOHplyAxwa7u/G7iyXV1yIfB0O5VyO7Apydr2puSmNiZJq9KQp0rOAj6a5Ojv+cuqui3JXcAtSbYCjwKXt/kfAy4FDgDPAFcBVNXhJNcBd7V511bV4QHXLUkr2mDhrqpHgNfOM/554KJ5xgu4eoHn2gnsPNFrlKQe+clJSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSeqM4ZakzhhuSerM4OFOcmqSe5L8Xds+N8mdSfYn+VCS09r417TtA23/OWPP8Y42/nCSi4desyStZEtxxP124KGx7d8H3lVVG4EjwNY2vhU4UlXfCryrzSPJecAVwHcAm4E/S3LqEqxbklakQcOdZD3ww8D72naANwMfblN2AZe1+1vaNm3/RW3+FuDmqvpSVf07cAA4f8h1S9JKNvQR958Avwb8T9t+FfBUVT3btmeBde3+OuAxgLb/6Tb//8bnecz/SbItyb4k++bm5k7065CkFWOwcCf5EeBQVd09PjzP1DrOvsUe85WBqh1VNV1V01NTU1/1eiWpF2sGfO43AT+W5FLgpcArGB2Bn55kTTuqXg8cbPNngQ3AbJI1wCuBw2PjR40/RpJWncGOuKvqHVW1vqrOYfTm4ser6qeBTwA/2abNALe2+7vbNm3/x6uq2vgV7aqTc4GNwKeHWrckrXRDHnEv5NeBm5O8E7gHuLGN3wh8MMkBRkfaVwBU1QNJbgEeBJ4Frq6q55Z+2ZK0MixJuKvqk8An2/1HmOeqkKr6InD5Ao+/Hrh+uBVKUj/85KQkdcZwS1JnDLckdcZwS1JnDLckdcZwS1JnDLckdcZwS1JnDLckdcZwS1JnDLckdcZwS1JnDLckdcZwS1JnDLckdcZwS1JnDLckdcZwS1JnDLckdcZwS1JnDLckdWaicCfZO8mYJGl4axbbmeSlwNcBr06yFkjb9QrgGwdemyRpHouGG/g54JcZRfpuvhLu/wT+dMB1SZIWsGi4q+rdwLuT/GJVvXeJ1iRJWsTxjrgBqKr3Jvk+4Jzxx1TVTQOtS5K0gInCneSDwLcA9wLPteECDLckLbGJwg1MA+dVVQ25GEnS8U16Hff9wDcMuRBJ0mQmPeJ+NfBgkk8DXzo6WFU/NsiqJEkLmjTcvz3kIiRJk5v0qpJ/GHohkqTJTHpVyX8xuooE4DTgJcB/V9UrhlqYJGl+kx5xf/34dpLLgPMHWZEkaVEv6tsBq+pvgDef4LVIkiYw6amSnxjbPIXRdd1e0y1Jy2DSq0p+dOz+s8DngC0nfDWSpOOa9Bz3VUMvRJI0mUn/kML6JB9NcijJE0k+kmT90IuTJL3QpG9Ovh/Yzeh7udcBf9vGFpTkpUk+neRfkjyQ5Hfa+LlJ7kyyP8mHkpzWxr+mbR9o+88Ze653tPGHk1z81b9MSTp5TBruqap6f1U9234+AEwd5zFfAt5cVa8FXgdsTnIh8PvAu6pqI3AE2NrmbwWOVNW3Au9q80hyHnAF8B3AZuDPkpw68SuUpJPMpOF+Mslbk5zaft4KfH6xB9TIF9rmS9pPMbqM8MNtfBdwWbu/pW3T9l+UJG385qr6UlX9O3AAryGXtIpNGu63AW8B/gN4HPhJ4LhvWLbI3wscAvYA/wY8VVXPtimzjE690G4fA2j7nwZeNT4+z2PGf9e2JPuS7Jubm5vwZUlSfyYN93XATFVNVdWZjEL+28d7UFU9V1WvA9YzOkp+zXzT2m0W2LfQ+LG/a0dVTVfV9NTU8c7iSFK/Jg33d1fVkaMbVXUYeP2kv6SqngI+CVwInJ7k6GWI64GD7f4ssAGg7X8lcHh8fJ7HSNKqM2m4T0my9uhGkjM4zjXgSaaSnN7ufy3wQ8BDwCcYnWoBmAFubfd3t23a/o+3v7izG7iiXXVyLrAR+PSE65akk86kn5z8I+CfknyY0WmKtwDXH+cxZwO72hUgpwC3VNXfJXkQuDnJO4F7gBvb/BuBDyY5wOhI+wqAqnogyS3Ag4w+tXl1VT2HJK1Sk35y8qYk+xhdERLgJ6rqweM85j7mOZ1SVY8wz1UhVfVF4PIFnut6jv8fhSStCpMecdNCvWisJUnDe1Ff6ypJWj6GW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6Y7glqTOGW5I6M1i4k2xI8okkDyV5IMnb2/gZSfYk2d9u17bxJHlPkgNJ7kvyhrHnmmnz9yeZGWrNktSDIY+4nwV+papeA1wIXJ3kPOAaYG9VbQT2tm2AS4CN7WcbcAOMQg9sBy4Azge2H429JK1Gg4W7qh6vqs+0+/8FPASsA7YAu9q0XcBl7f4W4KYauQM4PcnZwMXAnqo6XFVHgD3A5qHWLUkr3ZKc405yDvB64E7grKp6HEZxB85s09YBj409bLaNLTR+7O/YlmRfkn1zc3Mn+iVI0ooxeLiTvBz4CPDLVfWfi02dZ6wWGX/+QNWOqpququmpqakXt1hJ6sCg4U7yEkbR/ouq+us2/EQ7BUK7PdTGZ4ENYw9fDxxcZFySVqUhryoJcCPwUFX98diu3cDRK0NmgFvHxq9sV5dcCDzdTqXcDmxKsra9KbmpjUnSqrRmwOd+E/AzwGeT3NvGfgP4PeCWJFuBR4HL276PAZcCB4BngKsAqupwkuuAu9q8a6vq8IDrlqQVbbBwV9U/Mv/5aYCL5plfwNULPNdOYOeJW50k9ctPTkpSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwy3JHXGcEtSZwYLd5KdSQ4luX9s7Iwke5Lsb7dr23iSvCfJgST3JXnD2GNm2vz9SWaGWq8k9WLII+4PAJuPGbsG2FtVG4G9bRvgEmBj+9kG3ACj0APbgQuA84HtR2MvSavVYOGuqk8Bh48Z3gLsavd3AZeNjd9UI3cApyc5G7gY2FNVh6vqCLCHF/5nIEmrylKf4z6rqh4HaLdntvF1wGNj82bb2ELjL5BkW5J9SfbNzc2d8IVL0kqxUt6czDxjtcj4CwerdlTVdFVNT01NndDFSdJKstThfqKdAqHdHmrjs8CGsXnrgYOLjEvSqrXU4d4NHL0yZAa4dWz8ynZ1yYXA0+1Uyu3ApiRr25uSm9qYJK1aa4Z64iR/BfwA8Ooks4yuDvk94JYkW4FHgcvb9I8BlwIHgGeAqwCq6nCS64C72rxrq+rYNzwlaVUZLNxV9VML7LponrkFXL3A8+wEdp7ApUlS11bKm5OSpAkZbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM4YbknqjOGWpM50E+4km5M8nORAkmuWez2StFy6CHeSU4E/BS4BzgN+Ksl5y7sqSVoeXYQbOB84UFWPVNWXgZuBLcu8JklaFmuWewETWgc8NrY9C1wwPiHJNmBb2/xCkoeXaG0ni1cDTy73IpZS/nBmuZew2qy6f2Nsz//3GW6rqs3HDvYS7vlefT1vo2oHsGNplnPySbKvqqaXex06eflv7MTp5VTJLLBhbHs9cHCZ1iJJy6qXcN8FbExybpLTgCuA3cu8JklaFl2cKqmqZ5P8AnA7cCqws6oeWOZlnWw8zaSh+W/sBElVHX+WJGnF6OVUiSSpMdyS1BnDvcr5VQIaWpKdSQ4luX+513KyMNyrmF8loCXyAeAFHyLRi2e4Vze/SkCDq6pPAYeXex0nE8O9us33VQLrlmktkiZkuFe3436VgKSVx3Cvbn6VgNQhw726+VUCUocM9ypWVc8CR79K4CHgFr9KQCdakr8C/hn4tiSzSbYu95p650feJakzHnFLUmcMtyR1xnBLUmcMtyR1xnBLUmcMtyR1xnBLA2vfwiidMIZbGpPkuiRvH9u+PskvJfnVJHcluS/J74zt/5skdyd5IMm2sfEvJLk2yZ3A9y7xy9BJznBLz3cjMAOQ5BRGXwPwBLCR0dfgvg54Y5Lvb/PfVlVvBKaBX0ryqjb+MuD+qrqgqv5xKV+ATn5d/JV3aalU1eeSfD7J64GzgHuA7wE2tfsAL2cU8k8xivWPt/ENbfzzwHPAR5Zy7Vo9DLf0Qu8Dfhb4BmAncBHwu1X15+OTkvwA8EPA91bVM0k+Cby07f5iVT23VAvW6uKpEumFPsroT219D6Mv4LodeFuSlwMkWZfkTOCVwJEW7W8HLlyuBWt18YhbOkZVfTnJJ4Cn2lHz3yd5DfDPSQC+ALwVuA34+ST3AQ8DdyzXmrW6+O2A0jHam5KfAS6vqv3LvR7pWJ4qkca0v3J/ANhrtLVSecQtSZ3xiFuSOmO4JakzhluSOmO4JakzhluSOvO/PljANi1GJoEAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sn.factorplot(x='year',data=df_train,kind='count',size=5,aspect=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "12    1048\n",
       "5     1046\n",
       "3     1042\n",
       "7     1038\n",
       "10    1030\n",
       "6     1027\n",
       "8     1023\n",
       "11    1011\n",
       "9     1002\n",
       "4     1001\n",
       "1      968\n",
       "2      929\n",
       "Name: month, dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.month.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the function above we can infer that more rentals take place in December which is unexpected because we saw earlier that Winter has not as many bike rentals as Summer and Spring."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x251c1b85cf8>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 540x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sn.factorplot(x='month',data=df_train,kind='count',size=5,aspect=1.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Hour"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9     518\n",
       "20    517\n",
       "10    515\n",
       "12    514\n",
       "21    513\n",
       "19    512\n",
       "11    512\n",
       "13    511\n",
       "17    510\n",
       "23    510\n",
       "0     510\n",
       "5     509\n",
       "15    509\n",
       "18    505\n",
       "2     504\n",
       "22    504\n",
       "1     504\n",
       "6     503\n",
       "14    503\n",
       "8     503\n",
       "16    502\n",
       "7     499\n",
       "3     490\n",
       "4     488\n",
       "Name: hour, dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.hour.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the fuction above we can see the distribution of bike rentals throughout the day."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x251c1bdccc0>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 540x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sn.factorplot(x='hour',data=df_train,kind='count',size=5,aspect=1.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Holiday"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    11837\n",
       "1      328\n",
       "Name: holiday, dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.holiday.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can infer from the function above that people rent bikes at non-vacation days at higher rate than at vacation ones."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Anaconda\\lib\\site-packages\\seaborn\\categorical.py:3666: UserWarning: The `factorplot` function has been renamed to `catplot`. The original name will be removed in a future release. Please update your code. Note that the default `kind` in `factorplot` (`'point'`) has changed `'strip'` in `catplot`.\n",
      "  warnings.warn(msg)\n",
      "D:\\Anaconda\\lib\\site-packages\\seaborn\\categorical.py:3672: UserWarning: The `size` paramter has been renamed to `height`; please update your code.\n",
      "  warnings.warn(msg, UserWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x1382aff0048>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 540x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sn.factorplot(x='holiday',data=df_train,kind='count',size=5,aspect=1.5)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Weekday"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6    1776\n",
       "4    1763\n",
       "0    1762\n",
       "5    1737\n",
       "3    1730\n",
       "2    1704\n",
       "1    1693\n",
       "Name: weekday, dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.weekday.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The number of rentals is distributed at weekdays."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x251c32aa7f0>"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 540x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sn.factorplot(x='weekday',data=df_train,kind='count',size=5,aspect=1.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Workingday"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    8299\n",
       "0    3866\n",
       "Name: workingday, dtype: int64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.workingday.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have more bike rentals at working days."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x251c3235588>"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 540x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sn.factorplot(x='workingday',data=df_train,kind='count',size=5,aspect=1.5)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Weather"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    8004\n",
       "2    3127\n",
       "3    1031\n",
       "4       3\n",
       "Name: weather, dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.weather.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1: Clear, Few clouds, Partly cloudy, Partly cloudy | 2: Mist + Cloudy, Mist + Broken clouds, Mist + Few clouds, Mist | 3: Light Snow, Light Rain + Thunderstorm + Scattered clouds, Light Rain + Scattered clouds | 4: Heavy Rain + Ice Pallets + Thunderstorm + Mist, Snow + Fog temp\n",
    "\n",
    "- At days with heavy rain,ice pallets,thunderstorms and mists or foggy days with or without snow, we have only 3 rentals at train set. \n",
    "- At days with clear weather we have the most rentals(8004).\n",
    "- At days with cloudy weather the number is diminishing .\n",
    "- At days with light snow we have a few rentals but this number continues to be smaller than the corresponding one with clear weather.\n",
    "\n",
    "So the weather feature is very important for our model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x1fb0ad1e4e0>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 540x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sn.factorplot(x='weather',data=df_train,kind='count',size=5,aspect=1.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Numerical Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Relation between windspeed and count')]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Regression plot is used to verify if a pattern can be observed between `count` and numerical variables\n",
    "\n",
    "fig,[ax1,ax2,ax3] = plt.subplots(ncols = 3, figsize = (20,8))\n",
    "plt.rc('xtick', labelsize=10) \n",
    "plt.rc('ytick', labelsize=10) \n",
    "\n",
    "sn.regplot(x = 'temp', y = 'count',data = df_train, ax = ax1)\n",
    "ax1.set(title=\"Relation between temperature and count\")\n",
    "sn.regplot(x = 'humidity', y = 'count',data = df_train, ax = ax2)\n",
    "ax2.set(title=\"Relation between humidity and count\")\n",
    "sn.regplot(x = 'windspeed', y = 'count',data = df_train, ax = ax3)\n",
    "ax3.set(title=\"Relation between windspeed and count\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we see ,we can't infer something useful from this plots for our model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Visualisation of variation for all the continuous variables using histograms."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig,axes = plt.subplots(2,2)\n",
    "axes[0,0].hist(x=\"temp\",data=df_train,edgecolor=\"black\",linewidth=2,color='#ff4125')\n",
    "axes[0,0].set_title(\"Variation of temp\")\n",
    "axes[0,1].hist(x=\"atemp\",data=df_train,edgecolor=\"black\",linewidth=2,color='#ff4125')\n",
    "axes[0,1].set_title(\"Variation of atemp\")\n",
    "axes[1,0].hist(x=\"windspeed\",data=df_train,edgecolor=\"black\",linewidth=2,color='#ff4125')\n",
    "axes[1,0].set_title(\"Variation of windspeed\")\n",
    "axes[1,1].hist(x=\"humidity\",data=df_train,edgecolor=\"black\",linewidth=2,color='#ff4125')\n",
    "axes[1,1].set_title(\"Variation of humidity\")\n",
    "fig.set_size_inches(10,10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we make a heatmap to show us the correlation between the features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1fb0c762588>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_corr = df_train[['temp', 'atemp', 'humidity', 'windspeed','season' ,'workingday', 'weekday' , 'holiday'  ,'weather' , 'casual', 'registered','hour','year','month', 'count']].corr()\n",
    "mask = np.array(data_corr)\n",
    "mask[np.tril_indices_from(mask)] = False\n",
    "fig = plt.subplots(figsize=(15, 10))\n",
    "sn.heatmap(data_corr, mask=mask, vmax=1, square=True, annot=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- We drop atemp feature,on the grounds that is strongly correlated with temp feature. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Casual and registered contain direct information about the bike sharing count which we want to predict (data leakage). Therefore they are not considered in the feature set."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Windspeed has little correlation with count."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train = df_train.drop(['atemp', 'casual', 'registered', 'windspeed'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>season</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>hour</th>\n",
       "      <th>holiday</th>\n",
       "      <th>weekday</th>\n",
       "      <th>workingday</th>\n",
       "      <th>weather</th>\n",
       "      <th>temp</th>\n",
       "      <th>humidity</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.56</td>\n",
       "      <td>0.64</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.65</td>\n",
       "      <td>68</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.44</td>\n",
       "      <td>0.41</td>\n",
       "      <td>344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.93</td>\n",
       "      <td>222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.45</td>\n",
       "      <td>210</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   season  year  month  hour  holiday  weekday  workingday  weather  temp  \\\n",
       "0       4     0     11     0        0        2           1        1  0.56   \n",
       "1       1     0      2    21        0        6           0        1  0.28   \n",
       "2       4     1     11    12        0        3           1        1  0.44   \n",
       "3       1     1      1     7        0        2           1        2  0.18   \n",
       "4       1     1      3    16        0        1           1        1  0.30   \n",
       "\n",
       "   humidity  count  \n",
       "0      0.64     22  \n",
       "1      0.65     68  \n",
       "2      0.41    344  \n",
       "3      0.93    222  \n",
       "4      0.45    210  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(12165, 11)"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We did 1-hot encoding to variables 'month', 'year' in following lines, first at train set and then at test set. We can apply this procedure only to categorical variables. We ended up to do this to these 2 variables after several tests and combinations with other variables.\n",
    "- We tried 1-hot encoding to 'weather' but we observed that number 4 (Heavy Rain + Ice Pallets + Thunderstorm + Mist, Snow + Fog temp) did not exist in test set so we could not create 2 tables (train + test) with the same number of columns.\n",
    "We saw from the heatmap that 'hour' and 'season' are very important for our model so we tried to apply 1-hot encoding.\n",
    "- As for the 'hour', they had been created so many feature columns and this led the model to perform low despite the fact that we tried to apply PCA method to decrease the dimensionality.\n",
    "- As for the 'season', results were not good neither by itself nor when we tried to combine it with 'month' and 'year'.\n",
    "\n",
    "We could have combined 'season' and 'year' but we observed from the heatmap that 'season' and 'month' are strongly correlated so we decided to use only the one variable from these two."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>season</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>hour</th>\n",
       "      <th>holiday</th>\n",
       "      <th>weekday</th>\n",
       "      <th>workingday</th>\n",
       "      <th>weather</th>\n",
       "      <th>temp</th>\n",
       "      <th>humidity</th>\n",
       "      <th>count</th>\n",
       "      <th>year_0</th>\n",
       "      <th>year_1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.56</td>\n",
       "      <td>0.64</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.65</td>\n",
       "      <td>68</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.44</td>\n",
       "      <td>0.41</td>\n",
       "      <td>344</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.93</td>\n",
       "      <td>222</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.45</td>\n",
       "      <td>210</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   season  year  month  hour  holiday  weekday  workingday  weather  temp  \\\n",
       "0       4     0     11     0        0        2           1        1  0.56   \n",
       "1       1     0      2    21        0        6           0        1  0.28   \n",
       "2       4     1     11    12        0        3           1        1  0.44   \n",
       "3       1     1      1     7        0        2           1        2  0.18   \n",
       "4       1     1      3    16        0        1           1        1  0.30   \n",
       "\n",
       "   humidity  count  year_0  year_1  \n",
       "0      0.64     22       1       0  \n",
       "1      0.65     68       1       0  \n",
       "2      0.41    344       0       1  \n",
       "3      0.93    222       0       1  \n",
       "4      0.45    210       0       1  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hour=pd.get_dummies(df_train['year'],prefix='year')\n",
    "df_train=pd.concat([df_train,hour],axis=1)\n",
    "df_train.head()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1-hot encoding for month feature."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>season</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>hour</th>\n",
       "      <th>holiday</th>\n",
       "      <th>weekday</th>\n",
       "      <th>workingday</th>\n",
       "      <th>weather</th>\n",
       "      <th>temp</th>\n",
       "      <th>humidity</th>\n",
       "      <th>...</th>\n",
       "      <th>month_3</th>\n",
       "      <th>month_4</th>\n",
       "      <th>month_5</th>\n",
       "      <th>month_6</th>\n",
       "      <th>month_7</th>\n",
       "      <th>month_8</th>\n",
       "      <th>month_9</th>\n",
       "      <th>month_10</th>\n",
       "      <th>month_11</th>\n",
       "      <th>month_12</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.56</td>\n",
       "      <td>0.64</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>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.65</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>2</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.44</td>\n",
       "      <td>0.41</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>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.93</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>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.45</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   season  year  month  hour  holiday  weekday  workingday  weather  temp  \\\n",
       "0       4     0     11     0        0        2           1        1  0.56   \n",
       "1       1     0      2    21        0        6           0        1  0.28   \n",
       "2       4     1     11    12        0        3           1        1  0.44   \n",
       "3       1     1      1     7        0        2           1        2  0.18   \n",
       "4       1     1      3    16        0        1           1        1  0.30   \n",
       "\n",
       "   humidity  ...  month_3  month_4  month_5  month_6  month_7  month_8  \\\n",
       "0      0.64  ...        0        0        0        0        0        0   \n",
       "1      0.65  ...        0        0        0        0        0        0   \n",
       "2      0.41  ...        0        0        0        0        0        0   \n",
       "3      0.93  ...        0        0        0        0        0        0   \n",
       "4      0.45  ...        1        0        0        0        0        0   \n",
       "\n",
       "   month_9  month_10  month_11  month_12  \n",
       "0        0         0         1         0  \n",
       "1        0         0         0         0  \n",
       "2        0         0         1         0  \n",
       "3        0         0         0         0  \n",
       "4        0         0         0         0  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hour=pd.get_dummies(df_train['month'],prefix='month')\n",
    "df_train=pd.concat([df_train,hour],axis=1)\n",
    "df_train.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>season</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>hour</th>\n",
       "      <th>holiday</th>\n",
       "      <th>weekday</th>\n",
       "      <th>workingday</th>\n",
       "      <th>weather</th>\n",
       "      <th>temp</th>\n",
       "      <th>atemp</th>\n",
       "      <th>humidity</th>\n",
       "      <th>windspeed</th>\n",
       "      <th>year_0</th>\n",
       "      <th>year_1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.26</td>\n",
       "      <td>0.2121</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.4478</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.42</td>\n",
       "      <td>0.4478</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.2121</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.2836</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.74</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.2239</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.2424</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   season  year  month  hour  holiday  weekday  workingday  weather  temp  \\\n",
       "0       1     1      1    20        0        4           1        1  0.26   \n",
       "1       4     1     10    15        0        5           1        1  0.52   \n",
       "2       1     0      1    10        0        0           0        1  0.22   \n",
       "3       3     1      8    18        0        3           1        1  0.74   \n",
       "4       1     1      2     8        0        1           1        1  0.24   \n",
       "\n",
       "    atemp  humidity  windspeed  year_0  year_1  \n",
       "0  0.2121      0.52     0.4478       0       1  \n",
       "1  0.5000      0.42     0.4478       0       1  \n",
       "2  0.2121      0.51     0.2836       1       0  \n",
       "3  0.6667      0.51     0.2239       0       1  \n",
       "4  0.2424      0.60     0.0000       0       1  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hour=pd.get_dummies(df_test['year'],prefix='year')\n",
    "df_test=pd.concat([df_test,hour],axis=1)\n",
    "df_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>season</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>hour</th>\n",
       "      <th>holiday</th>\n",
       "      <th>weekday</th>\n",
       "      <th>workingday</th>\n",
       "      <th>weather</th>\n",
       "      <th>temp</th>\n",
       "      <th>atemp</th>\n",
       "      <th>...</th>\n",
       "      <th>month_3</th>\n",
       "      <th>month_4</th>\n",
       "      <th>month_5</th>\n",
       "      <th>month_6</th>\n",
       "      <th>month_7</th>\n",
       "      <th>month_8</th>\n",
       "      <th>month_9</th>\n",
       "      <th>month_10</th>\n",
       "      <th>month_11</th>\n",
       "      <th>month_12</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.26</td>\n",
       "      <td>0.2121</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>4</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.5000</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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.2121</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>3</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.74</td>\n",
       "      <td>0.6667</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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.2424</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 × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   season  year  month  hour  holiday  weekday  workingday  weather  temp  \\\n",
       "0       1     1      1    20        0        4           1        1  0.26   \n",
       "1       4     1     10    15        0        5           1        1  0.52   \n",
       "2       1     0      1    10        0        0           0        1  0.22   \n",
       "3       3     1      8    18        0        3           1        1  0.74   \n",
       "4       1     1      2     8        0        1           1        1  0.24   \n",
       "\n",
       "    atemp  ...  month_3  month_4  month_5  month_6  month_7  month_8  month_9  \\\n",
       "0  0.2121  ...        0        0        0        0        0        0        0   \n",
       "1  0.5000  ...        0        0        0        0        0        0        0   \n",
       "2  0.2121  ...        0        0        0        0        0        0        0   \n",
       "3  0.6667  ...        0        0        0        0        0        1        0   \n",
       "4  0.2424  ...        0        0        0        0        0        0        0   \n",
       "\n",
       "   month_10  month_11  month_12  \n",
       "0         0         0         0  \n",
       "1         1         0         0  \n",
       "2         0         0         0  \n",
       "3         0         0         0  \n",
       "4         0         0         0  \n",
       "\n",
       "[5 rows x 26 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hour=pd.get_dummies(df_test['month'],prefix='month')\n",
    "df_test=pd.concat([df_test,hour],axis=1)\n",
    "df_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train = df_train.drop(['year','month'], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Some data types need to be changed from numerical to categorical in order for the model to interpret better these variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train['weekday'] = df_train.weekday.astype('category')\n",
    "df_train['hour'] = df_train.hour.astype('category')\n",
    "df_train['holiday'] = df_train.holiday.astype('category')\n",
    "df_train['weather'] = df_train.weather.astype('category')\n",
    "df_train['workingday'] = df_train.workingday.astype('category')\n",
    "df_train['season'] = df_train.season.astype('category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "season        category\n",
       "hour          category\n",
       "holiday       category\n",
       "weekday       category\n",
       "workingday    category\n",
       "weather       category\n",
       "temp           float64\n",
       "humidity       float64\n",
       "count            int64\n",
       "year_0           uint8\n",
       "year_1           uint8\n",
       "month_1          uint8\n",
       "month_2          uint8\n",
       "month_3          uint8\n",
       "month_4          uint8\n",
       "month_5          uint8\n",
       "month_6          uint8\n",
       "month_7          uint8\n",
       "month_8          uint8\n",
       "month_9          uint8\n",
       "month_10         uint8\n",
       "month_11         uint8\n",
       "month_12         uint8\n",
       "dtype: object"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(12165, 23)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>season</th>\n",
       "      <th>hour</th>\n",
       "      <th>holiday</th>\n",
       "      <th>weekday</th>\n",
       "      <th>workingday</th>\n",
       "      <th>weather</th>\n",
       "      <th>temp</th>\n",
       "      <th>humidity</th>\n",
       "      <th>count</th>\n",
       "      <th>year_0</th>\n",
       "      <th>...</th>\n",
       "      <th>month_3</th>\n",
       "      <th>month_4</th>\n",
       "      <th>month_5</th>\n",
       "      <th>month_6</th>\n",
       "      <th>month_7</th>\n",
       "      <th>month_8</th>\n",
       "      <th>month_9</th>\n",
       "      <th>month_10</th>\n",
       "      <th>month_11</th>\n",
       "      <th>month_12</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.56</td>\n",
       "      <td>0.64</td>\n",
       "      <td>22</td>\n",
       "      <td>1</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>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.65</td>\n",
       "      <td>68</td>\n",
       "      <td>1</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>2</th>\n",
       "      <td>4</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.44</td>\n",
       "      <td>0.41</td>\n",
       "      <td>344</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>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  season hour holiday weekday workingday weather  temp  humidity  count  \\\n",
       "0      4    0       0       2          1       1  0.56      0.64     22   \n",
       "1      1   21       0       6          0       1  0.28      0.65     68   \n",
       "2      4   12       0       3          1       1  0.44      0.41    344   \n",
       "\n",
       "   year_0  ...  month_3  month_4  month_5  month_6  month_7  month_8  month_9  \\\n",
       "0       1  ...        0        0        0        0        0        0        0   \n",
       "1       1  ...        0        0        0        0        0        0        0   \n",
       "2       0  ...        0        0        0        0        0        0        0   \n",
       "\n",
       "   month_10  month_11  month_12  \n",
       "0         0         1         0  \n",
       "1         0         0         0  \n",
       "2         0         1         0  \n",
       "\n",
       "[3 rows x 23 columns]"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Building"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Because outlier value will bring some big error,we used log to avoid outliers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = np.array(df_train.drop(df_train[[ 'count']], 1))\n",
    "\n",
    "y = df_train['count']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=42)\n",
    "\n",
    "y_trainlog = np.log1p(y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### KNN"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The k-nearest neighbors (KNN) algorithm is a supervised machine learning algorithm that can be used to solve both classification and regression problems. It assumes that similar things exist in close proximity. In other words, similar things are near to each other so it calculates the straight-line distance  between these points on a graph. We present below its implementation: \n",
    "1. Load the data.\n",
    "2. Initialize K to your chosen number of neighbors.\n",
    "3. For each example in the data\n",
    "    - 3.1 Calculate the distance between the query example and the current example from the data.\n",
    "    - 3.2 Add the distance and the index of the example to an ordered collection.\n",
    "4. Sort the ordered collection of distances and indices from smallest to largest (in ascending order) by the distances.\n",
    "5. Pick the first K entries from the sorted collection.\n",
    "6. Get the labels of the selected K entries.\n",
    "7. If regression (a real number as output), return the mean of the K labels.\n",
    "8. If classification (a discrete value as output), return the mode of the K labels.\n",
    "\n",
    "We tried to apply KNN to our model but as you can see below, it did not improve our RMSLE score."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "knn = KNeighborsRegressor(n_jobs=-1)\n",
    "knn.fit(X_train, y_trainlog)\n",
    "y_pred= knn.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- n_jobs : The number of parallel jobs to run for neighbors search,-1 means using all processors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "True in (y_pred < 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i, y  in enumerate(y_pred):\n",
    "    if y_pred[i] < 0:\n",
    "        y_pred[i] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSLE: 0.39962138873333625\n",
      "R2: 0.8801970604914774\n"
     ]
    }
   ],
   "source": [
    "print('RMSLE:', np.sqrt(mean_squared_log_error(y_test,np.exp( y_pred))))\n",
    "print('R2:', r2_score(y_test, np.exp( y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Comparison between the actual vs predicted values from KNN')]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEWCAYAAACXGLsWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nO3dd3wc5Z348c9XZdVsWZZ7lxuEEghgsAnFFOcIhB9w6SEBDCQmCRxJyF1IuLuEJFzKXS5ADi6EXk05kguEhCMUU4MrvRrjKluyZXVptbta7fP743l2PVpvk7Srtt/366WXdmdmZ2ZnZuc7TxdjDEoppVQqBUO9A0oppYY/DRZKKaXS0mChlFIqLQ0WSiml0tJgoZRSKi0NFkoppdLKu2AhIl8Wkb8O8ja3isiywdzmUBKRk0Skdqj3I5tE5GoRuXeo92MgROROEbnGvT5BRN4fpO0aEVmQ5XXWuPUWZXO9abYpInKHiDSLyNrB2u5w0e9gISLnish6EekQkToReVxEjs/mzuWCMeY+Y8zfDfV+ZGokBJpc3AwGSkSWi8iLQ70fw5Ux5gVjzIHpltPj2MvxwCeAmcaYYwZ74/HnQkQqReQlEfm9iBS7hwEjIsd4llkgIsbz/lkRCYjILM+0ZSKyNd32+xUsROQK4DrgZ8AUYDbw38DZ/VnfYBnMpxClckmv5SExB9hqjOlMNHOQUznjgaeAbcAXjDHdblYTcE2aj3cC/9rnjRpj+vQHjAM6gM+lWKYEG0x2ub/rgBI37ySgFvgesAeoA84BzgA2ui97lWddVwMPAw8C7cArwOGe+d8HPnTz3gH+3jNvOfAScK3nIC4HXnTzxc3bA7QCbwCHer7n3UAD9oT8C1DgWe+LwK+AZmALcHqK47EV+IHbv2bgDqDUM/9M4DWgBfgbcJibfg8QAbrcMf8ecBfwXTd/BmCAb7r3C9z3lFTrdfOmA793328LcHncMX/Iff924G1gUZLv9rzbh063j1/wnOPves7xhXHXx6+A7cBu4CagLMn65wPPAI3AXuA+oMozfxbwB/c9GoEbgIOAANDj9qnFLfss8NW46+NFz/vrgR1AG7ABOCHumNybZB/fBc70vC9y+3okUArc6/atBVgHTOnrdeI5plcC9cA9GZzjI7C/l3bs7+cB4Brv+vp5HFOeP+Cf3DnfBVzkro8FCb7vF4H1cdO+AzzqXn8KeNWdjx3A1Z7latx6izzHblmy8wUsccenBXgdOCnuOtjsjtMW4MsJ9vXiuGPx4xTn5GvAJuxv8VFgumc9Bvgm8IHb3k+x1/jL7ns+BPiSXB/Lsfedie643IG7J7n5dwK/dvuy1HNPMJ5lngV+5La9wE1bhg2Cqe/96RZIsMOfBMLRk5RkmZ8Aq4HJwCR3kn7quUjDwA+BYndgG4CVwFjgEHdS5nlOejfwWbf8P7oTWuzmfw574yvA3qg6gWmegxsG/gH7Ay6jd7A4DXtTqMIGjoM8n70beMTtUw02kF3sWW+32/dC4BvYH4akuAm8hf1BVmMDWPRHeyT2hrrYresCt3yJ57PeH8FFwJ/c63OxgfJBz7xH0q3XHasN7hz4gHnYH8tpnmMewAbwQuDnwOoU57vXzcBzjn/iztkZgB8Y7+Zfh/0RVbvj+yfg50nWvQCb9C/BXkvPA9e5eYXYH/61QAX2xnx8okDg+aGkChZfASZgr5XvYn90pZ5jkixY/BC4z/P+U8B77vUl7vuVu/09Cqjsx3USPaa/dMeiLM059mEfcr7jzsFnsdfsfsGiH8cx6fnD3h92A4e6da2Mvz486ynH3rQWeqatA77o2cePYq/Xw9x6z+lrsMA+VDVir8MC7PXUiL2eKrA36QPdstOAQ1LdrBNc595zcgr7HhRKgP8Cno/7rTwKVGLvdUHgaexvcBz2QeGCFNt/B/vwdhNx9xtssLgGuJx997hEweKr2KASPT45CxZfBurTLPMhcIbn/WnRnXEHuAsodO/HugO42LP8Bs9FcTWeG5U72XV4nvritv0acLbn4G5PdsLdid2IferwRuhCdxIP9ky7BHjWs45NcRe9AaamuAl83fP+DOBD9/q3uEDqmf8++54MttL7RzAf+3RU4C6YS9j3o78LuCLderE3l/jj8gPgDs8xf8oz72CgK8X5ThQsuvA8UGBvakuwQbkTmO+ZdyywJcPr7xzgVc/nGkjw4EI/gkWCdTTjUrGkDhYLsDe9cvf+PuCH7vVFxD3xp9hequvkJCBE7xRpqnN8InEPMG4/EgWLjI9juvMH3A78wjPvgPjrI27993qO1ULvcUyw7HXAte51DZkHiytxT/2e+U9gg2sF9vf0GZKkblMci0Tn5Dbg3z3vx2CDdI3nt3KcZ/4G4ErP+//EPQwl2X67W9/iBPPvxAaLEmyq73SSB4tJ2NyUQ8gwWPSnzKIRmJgmf2469qkmapubFluHMabHve5y/3d75ndhD3LUjugLY0wEm/SbDiAi54vIayLSIiIt2CeaiYk+G88Y8ww2uX0jsFtEbhaRSvf56JOZ9zvM8Lyv96zH71569zmedz+8x2MO8N3o/rvvMIvex8u7zx9ik8EfA04AHgN2iciB2JvEcxmsdw4wPW7eVdjyp/2+HzZVUNrHPNlGY0w4bh1jsBdpObDBs+3/c9P3IyKTReQBEdkpIm3Ym0v0/M4CtsVtp99E5Lsi8q6ItLr9GkfvaykhY8wmbFbU/xORcuAs7BM12KzEJ4AHRGSXiPy7iBSnWF2y6wSgwRgT8LxPdY6nAzuNu0N41pdIX45juvM3PcF3SGUl8CX3+lzgj9Hfk4gsFpFVItIgIq3A18ngfCQwB/hc3HE6HpuL0InNkfg6UCcifxaRj/Rh3fHnpNe9zxjTgb1neu8d8fe6VPe+eK9jc1ceF5EjEi1gjAlis7d+ig3uiZZpwN77fpJiW730J1i8jM2iOCfFMruwJyhqtpvWX96S+wJgJvYGOQe4BbgMmGCMqcIm470HyPtj2Y8x5jfGmKOwEfYAbH7rXmz0jv8OO7PxHeh9PHYA/2aMqfL8lRtj7k+x/89hsxV8xpid7v35wHhsyirdendgnwS988YaY84YwPfL1F7sD+IQz7bHGWOS/UB+jj0GhxljKrFZRdHzuwOYnSSIJTpundgbXdTU6AsROQH7BPp5bHZZFfbJK+GPLYH7sTe9s4F3XADBGNNtjPmxMeZg4OPYMobzU6wn2XWS6DulOsd1wAwR8e7/7CTb7MtxTHf+6hJ8h1T+in34/Bj2+K30zFuJzbKZZYwZh8t6SbKepOcW+/3uiTtOFcaYXwAYY54wxnwCmwX1Hvaekqn449Pr3iciFdiszYHcO3pv0JjrgV8AT4rIoUkWuwP7sPP3KVb1H8DJ2KzRtPocLIwxrdg82htF5BwRKXfVtk4XkX93i90P/IuITBKRiW75gdRRP0pEPu0u5m9js4hWY5OQBpuERkQuxKYsMiIiR7unl2LsxRYAelyq5yHg30RkrAtKVwzwO1wqIjNFpBr7FP+gm34L8HW3HyIiFSLyKREZ6+bvxuZnej2HDZDPu/fPYstlXvSk2FKtdy3QJiJXikiZiBSKyKEicnQ/v1uifUzIpQxvAa4VkckAIjJDRE5L8pGxuMJVEZmBDeZRa7E3p1+471cqIsd59mmmiPg8y78GfNpdswuwhZbe7YRx2TEi8kNsvnKmHgD+Dlt+FbvhicjJIvJRESnE5o13YwtJk0l2nSSS6hy/7L7P5SJSJCKfBpJV98z4OGZw/h4ClovIwS6V9aMU+49LzTyMvXFVA096Zo8FmowxAbHVQc9NsarXgC+6e9Ei7MNU1L3YVN9p7lovFdsWaKaITBGRs9xNPYi91lKdn3RWAheKyMdEpARbY3SNMWbrANa5H2PMv2MrZDwlNlchfn4YmxV3ZYp1tGCzvb6XyTb7VXXWGPNr7M3zX7A/rh3Ym9cf3SLXAOuxtYvexNbISFedK5VHsEnFZuA84NPuie0d7Jd9GXtRfxRbKJipSuyF34xNOjZia3mAvfl2Ygt+X8ReBLcP4DusxD5FbXZ/1wAYY9ZjC8pvcPuxCZs3GfVzbOBtEZF/dNOew/6QosHiRexTVfR9yvW6gPL/sFlZW7BPi7din0T642rgLrePn89g+Svd/qx2WUtPAcnq/P8YW1jYCvwZW2MH6PU9FmDzaGux1wnYGlRvA/UistdNuxabx7wbW75zn2c7TwCPY8uwtmEfHJJmYcYzxtRhr8OP0/sGPxV7M2zDZlU9R+qHjoTXSZJtpjrHIeDT7n0z9rj8Icl6+nock54/Y8zj2LKFZ9wyz6T4rt7vvAz4n7issG8CPxGRduwD50Mp1vGv2PK8Zuw1EwvYxpgd2BTfVey7X/0T9v5XgK3MsAtbe2mp226/GGOedvvye2wAno+t9ZV1xpifYn+3T4vI/ASLRFOYqVxPhsExWsVy2BKRq7GFY18Z6n1RKpfENoz6qjHmqaHeF6Xi5V13H0oppfpOg4VSSqm0hn02lFJKqaGnKQullFJpDYvOyCZOnGhqamqGejeUUmpE2bBhw15jTMIGrdk2LIJFTU0N69evH+rdUEqpEUVE0rWQzxrNhlJKKZWWBgullFJpabBQSimVlgYLpZRSaWmwUEoplZYGC6WUUmllFCxE5Dsi8raIvCUi97sufueKyBoR+UBEHox2YSwiJe79Jje/JpdfQCmlVO6lDRZuDIHLgUXGmEOxQ45+ETvu7LXGmIXYboGjYwNcDDQbYxZgu4T+ZS52XCml1ODJNBuqCChzgw+VY/tIPwXbTz/YsQGiI+ed7d7j5p8aN1qXUkqpAYhEDK3+7kHdZtpg4Ybt/BV2UJQ67CA0G4AWz0AltewbY3YGbtAYN78VO6xgLyKyQkTWi8j6hoaGgX4PpZTKC+2Bbmqbu2jsDA7qdjPJhhqPTS3MxQ5GXgGcnmDRaPe1iVIR+3Vta4y52RizyBizaNKkQenaRCmlRqxguIddLV3saQvw4qYGzr997aBuP5O+oZYBW4wx0XGu/4AdOrJKRIpc6mEm+waWr8UO2F7rsq3GYYcrVEop1Uc9EUOzP0RbVzfbG/3897ObWLu1edD3I5NgsR1Y4gZf7wJOxY6vvQo7KPoDwAXYcbIBHnXvX3bznzE6aIZSSvVZW6Cb5s4Qrf5u7l69lf99dRc9EXs7PWrOeAatF0EyCBbGmDUi8jDwChAGXgVuBv4MPCAi17hpt7mP3AbcIyKbsCmKnAxWrpRSo1Wgu4e9HUG6Qj08/lY9t7+4hZYuW6A9bVwp3zxpPksPmMQfvjl4+5RRF+XGmB8BP4qbvBk4JsGyAeBzA981pZQaXS5d+QqbGzqZN6mCG889cr/54Z4ITf4QHYEwb9S2cMOqD9m0pwOA0uICvrJ4Dp89aiYVJUVMGOMb1H0fFuNZKKVUPtjc0Mm7dW37TTfG0NYVptkfoq61i5uf38yq9/fVEj3tkClcfPxcJo8tZXy5j8qyIga7RYIGC6WUGkL+UJjGjhDtgW4eXLeDB9btIBiOAPCRqWP5h1MWcNC0SirLihlf7qOwYGiarWmwUEqpIdDdE6GpM0RHoJvnNjZw03Ob2dNu205UV/hYccJclh08hYqSIqorfJQUFQ7p/mqwUEqpQdARDNPiDwHQ1Blk4+52djZ3ccOqTbxR2wpAcaHwmSNn8pUlsxlX5mPCGB/lvuFxmx4ee6GUUqPYuq1NLL9jLf5gDwC724KcdcOLRCL7Wiwft2ACX186n9nV5VSVDU25RCoaLJRSKoc6gmEuuH0N/lCk1/Qe93ZOdTmXnbKARTXVjC0tGtJyiVQ0WCilVI5EIoYH1m4nkqRZclGB8JkjZ3DCwklUV/jwFQ3fIYY0WCilVA60B7pp7uzm1e1NBLojCZcJRwxtgTBTx5UO8t71nQYLpZTKokB3D42dIZo6gty7ehuPv7U76bJlxQUsnDJmEPeu/zRYKKVUFvREDE2dIVq7QjzxVj23vriFZjfmhJCg622goEA487Dpg7qf/aXBQimlBqi1q5sWf4jXd7Rww6pNbNztuugoKuDcxbM5aFolP3z0LQKhCAYQgXJfIXdeeAwVJSPjNjwy9lIppYahaId/O5ttFx1Pv7cnNu/Uj0xmxYnzmDG+jAkVJZzx0Wks+/Vz1LUGmFpZylNXLB0xgQI0WCilVJ+FXevrxo4gD22o5f412wm4LjoOnDKWS0+ez+Gzqqgq9zGurDj2uapyH3WtAarKfSMqUIAGC6WUypgxhtYuO8ZEtIuO+rYAAOPLi/naCfM47dCpVLl+nAqGYXuJ/tJgoZRSGYh2+PdeXRs3PruJ13bYLjqibSW+smQOkytLqa7wUVyYuL3EvEkVvf6PJBoslFIqhWiHf7uau7j9b1v48xt1sUZ2S+ZV882T5jNv0hgmVJRQ5kvd2V+iMSxGCg0WSimVgDGGZn83jR1BHnltJ3f+bRsdwTAAs6vL+eZJ8zl2/gTGV/ioLC1Os7aRT4OFUkrF6QiGaeoIsXrzXm589kO2NfoBqCgpZPnHazjnYzOYMKaEqrLiUVUukYoGC6WUckLhCI2dQTbt6eC3z37I3z5sBGyjuk8dNo2LjqthxvjylOUSo5UGC6VU3otEDM3+EPVtAe5dvY2HN9TS3WMLJg6bOY7LTl7AwdMrmTimhNLioR2EaKhosFBK5bX2QDd7O4I88VY9t7ywhcZOO0DR5LElfH3pfE79yGSqx/gYmwflEqlosFBK5aVoh3+vbW/mv57ZxHv17QCUFBXwpWNm8cWjZzO5sjSvyiVS0WChlMor0Q7/tuzt4JYXtvDkO/t6hT35wEmsOHEe8yeNYXwelkukosFCKZU3Wru6qW/t4n/W13Lvmm2xcSYWTB7DZSfP5+i5E5hQ4cvbcolUNFgopUa9QHcPDe0BnnmvgZue+5C6VttFR1VZMRcfP5czD5vOxLFaLpGKBgul1KgV7fDvzZ2t3LhqE69sbwGgsED49BEzqGvt4k9v7OK9+nZu/PLIbV09GDRYKKVGnWiHf9sb/dz+0hYefX1XrIuOY+baLjoOnl7JBbetZePuDgoLtGwiHQ0WSqlRxR8Ks7stwB9f3ckdL22lLWC76Jg5voxvnjSfpQdO3lcuIVrLKVMaLJRSo0J3T4TGjhAvbWrgxlUfsnlvJwAVvkLOO3YOnztqFlPGlTJmhI0jMVzoUVNKjWiRiKGlq5t369q46bkPeeGDvYDtouP0j07lq8fPY96kCsaVFSOakug3DRZKqRGrIximtsnPPau38dD6HbEuOg6dXsllpyzgyDnjqS73UaTtJQZMg4VSasQJhu3Y1396vY5bnt8c66Jj0pgSVpw4jzM+OpUJafpx6giGafHbz7X4Q3QEw5pFlYIeGaXUiBGJGJr8IdZuaeSGZzbxTp3tosNXVMAXFs3kvCU1TB9flvamv25rE8vvWIs/1ANAfVuAxT97ijsvPIaja6pz/j1GIg0WSqkRoS3Qzabd7fzu+c088fa+LjqWHjCJry+dx0HTKjMql+gIhll+x1o6gz2xacZAZ7CH5XesZe1Vy6jQFMZ+9IgopYa1QHcPda0B7l+zjXtWb6er297k502q4LKTF3DCwklUV/gozLCzv8de34UxiecZA4+9sYsvHD07W7s/amiwUEoNSz0RQ2NHkL++U89vn93MzpYuACpLi7j4+Ll85qiZTBpbQklR3/px2trYGct+iucP9bB1r3/A+z4aabBQSg0rxhjausK8tqOZG1ZtYt3WZgAKBM45YgZfPX4ucyZU9DurqGZCBeW+woQBo9xXSM3E8gHt/2ilwUIpNWx0hXrY0tjBrS9s4Y+v7ox10bFoznguO3kBR8weT2VZ0YDaS5x5+HR++ud3Es4TgTMPm97vdY9mGQULEakCbgUOBQxwEfA+8CBQA2wFPm+MaRZ7Fq8HzgD8wHJjzCtZ33Ol1KgR7onQ0B7kofU7uP2lrbR2dQMwo6qMb5w0j7872FaFzbRcIpUxJUXceeExsdpQxtggUe4r5M4Lj9HC7SQyPSrXA/9njPmsiPiAcuAq4GljzC9E5PvA94ErgdOBhe5vMfBb918ppXqJdvi36v0GbnjmAz5ssF10lBUXct6S2Xx5yRymjivtc7lEOkfXVLP2qmUs+/Vz1LUGmFpZylNXLNVAkULaZo0iUgmcCNwGYIwJGWNagLOBu9xidwHnuNdnA3cbazVQJSLTsr7nSqkRzR8Ks35rM9964DW+8+BrsUDxyUOmcv/XFrO9qYuv3/sKVzz0ek62X1FSRFW5D4Cqcp8GijQyOTrzgAbgDhE5HNgAfAuYYoypAzDG1InIZLf8DGCH5/O1blpd1vZaKTWsXLryFTY3dDJvUgU3npt6XIhQOMLO5i5uf2kzD66vJRS2o9UdPG0sl5+ykGPnT6SyrIhtTX7erWsbjN1XGcgkWBQBRwL/YIxZIyLXY7OckkmUqbhfrWYRWQGsAJg9W+s0KzWSbW7oTHtjj0QMzf4Q//vqTn733GYaOoIATBjj45IT5/H3R8yguiI75RIq+zIJFrVArTFmjXv/MDZY7BaRaS5VMQ3Y41l+lufzM4Fd8Ss1xtwM3AywaNGiJE1klFKjQUcwzMubGrn+6Y28tcsGleJC4fOLZnHR8TXMHF+e9XIJlV1pg4Uxpl5EdojIgcaY94FTgXfc3wXAL9z/R9xHHgUuE5EHsAXbrdHsKqVUfgmGe/hgdwf//ewmHn+zPpbFcMLCiVx28gI+OnMc5T4tKxgJMj1L/wDc52pCbQYuxBaOPyQiFwPbgc+5Zf+CrTa7CVt19sKs7rFSatjriRj2tAe4629bueflbXS6BnBzJ1Zw+SkLePyteq7+09vMmzQmbRmHGh4yChbGmNeARQlmnZpgWQNcOsD9UkqNUG2Bbh5/s47/emYTtc37uuhYftxczl08i0ljSrn5hS28W9dO4iLOwTNvUkWv/yo5Tf8ppbKmMxjmq3etZ+2WJsB20XHW4dP5xkkLmDuxAl9RZoMQDdZYE5qqyZwGC6XyVF+qu6b6TE/EEHQ9wW5v8rO9yXbEd+TsKr516kKOnlvdp3IJHWtieNJgoVSe8lZ3zTRweD9jjKG5s5uVa7extbEztsy0caVcevJ8zjxsep/HvdaxJoYvPepKqYzaSXgZY/jzG3Vc9/QHbNrTEZteXVHMw984lqmVZUnbS6TKYtKxJoYvHcVcKZUx4+7ktc1+Lrv/1VigiN7sJ48tZUZVedJAsW5rE4t/9hT1bQFgXxbTuq22jEPHmhi+NFgopdIyxrCrpYvd7ibf4bKJPjJ1LDd95UhmjC8DSJnl5M1iiqYevFlMncFwbKyJRHSsiaGlwUKpPBcxZr9sIa+OQDd3v7yNs298iWa/7Tq8QOCKTxzA/SuWcNohUynIoFwikyymMw+fTrJV6VgTQ0uDhVJ57oPd7QmzhULhCM9t3MO5t67hR4++TUN7MPaZiIGbntvEpj0ddIZ6UgabqEyymKJjTVSUFMaChghUlOhYE0NNj7xSo1AmtZsi7jE/4nnaj2YLnX/bak46cAr/91b9/r2AOv5QhPNuW0OhCP7u9NVcMx3OVMeaGJ40ZaHUCHfpylc4/foXuHTlvgEpo7WbNjd0Jv1cmxuNLpGubsPjLlBMqPBRXJg4byjQHaEzlLwMwqsvWUw61sTwo8FCqREuk8AQryMYpi2QPFiA7RX28lMXcOZh0+ju6VvH0NEyCC/NYhrZNFgolWei1Vf9wcTlBwCFBXD5qQv49qkHcNC0yqQ1lJJJVs01msU0tbIUgKmVpay9apm2zB4BNFgolUd6VV9NsVxJUQEXHTePggJJmX2UTKpqrprFNDJpsFAqjzz2+i56eiIpl6koKeSuixbHbuLJso/KfQWUFie+hWg119FHQ7pSo0x8dxqX3LOB7U1+5k4spydiCISTpykqSgoT9r+UrIbSO3VtsU7/jIkGES2DGI30bCo1iiTqsbX+nXqMgR1NnbGW18lUlhYnvclHs4/qWgOx7KNcVnPVsSaGFw0WSo0SEWMS9tgaFQ0UIiRtSV1ZVtzn7SYKIulkEgh0rInhRYOFUqNEW1d30iAANkh8/5Mf4aDpY/nGva/0yjoSbOO8TLrtyAYNBCOPFnArNQwlamiXTigcSdqdBkBVWTGXLJ3PiQsn71d9deGUsQPeZzW6acpCqWEo2fgS8d14eAuzg+EeCqR39x1epcX72krEZx3Nm1RBgYiWD6ikNFgoNYJ4g0i0MDtaRpG28DpFeUSm2UJa6Jy/NFgoNUx4Uw3pRIzhgtvX4A+lbjOR7fIILWvIXxoslBom+jK06d6OIIHuxIGirLgAX1EhrV3dTBlbAgj1bYH9hjBVqi+0gFupDPWn0Dnbwq71dWNHKGnZRFd3hCI3rOme9iC72xMPYapUX2iwUCpD/endNVsBpjsc4Y6XtrB5b/ptl/sKY12KRwwZdR+uVDqaHlWjTiYD/wyWvmQtRUWMiY010eIP8dd36rn2rxt5t749tsz48mIC3WG6uvdPXtiiieTlE4m6D1cqHU1ZqFGnPymA4cQ7zGlda4AVd2+IBYpoecPUcWXcffGSpGNDdKfoLDBZ9+FKpaLBQqkhEp9F5R3mNL4ltgj8+nOHM7bUBosWf4iDplUmHRvCV5T8p52q+3ClktFsKKUGkTeLzJtF1REMU9ec/Gm/SOAH//smIZdi8I51nahfpsNmVrG7rT5hIXi0+/DnP9gLaJsJlRkNFiovDGY5Rqr2EonKMPyhMIt++teUXYd3R4DIvqwlb2H1zPH7pxJ+d95RvXqgTdR9+FCX56iRRYOFygt9LWgeSHDp67a2Nfa//MAYYoXh8XLZfXg8bdk9+mmwUCqB/tRi6gtjIjR1Jr7J94U/1ENZcfLxsfvTfXh/aCpl9NNgodQQ2Nzgj5U/pOMrEgQhGN5/+XJfYcrCbKWyRa8ypQbJpj3tbG+y1XkzDRTjyop58XunUFSYuN2ESP8GLFKqrzRYKJWB+HGtO4LhtK2zI8bEPrOtsZPTr38h1kNsaXEBcybYgumCFP37TR1XyuTKUlconbhNxWANWKTym2ZDKZVCbbOfpf+xikoDpTwAABqHSURBVNrmrlg7iGi11YljSlIWTm+sbydavyk6KFF0vImaCeWI2Ge1WdXl7O0Iphy5bjALq5VKRFMWSsXxpiK6Qj1sa/TTEzH79bG0oylxoIgGlUQVYaPtHlq7wrHlyn1FGY1cFy2sBnJaWK1UInq1KeXhbZsAEE7WtSuJg0FdSxfbMujsr741QH2r7dKjxR/CQK9aS0oNN5qyUHmlttmftJyhIxiOjTwX391GIt5lAt1hrntqI6f++jkCCWot7fdZ9gWb+labreUPaU+wavjKOGUhIoXAemCnMeZMEZkLPABUA68A5xljQiJSAtwNHAU0Al8wxmzN+p4rlUCigujoa7DZSsnaTzz2+q6MgkSUyL5GcUf85Cm6um1qREic6kjGYLO1ukK56dxPG8ypbOhLNtS3gHeBSvf+l8C1xpgHROQm4GLgt+5/szFmgYh80S33hSzus1IJxWch1bcFWHTNkwCxNgrRbCV/KLxfK+1bXtgc+2wmooFlZ0tX7+n93H/v57J5g9cGcyobMgoWIjIT+BTwb8AVIiLAKcC5bpG7gKuxweJs9xrgYeAGERFj+vLMplTfXHLPBp58p3fHecaQdOjRHU1+SosLed8zRkS6QOFNMRQVgDe3qay4MJay6C/vL6QvN3hNOajBkGnK4jrge0C0isYEoMUYE81krQVmuNczgB0AxpiwiLS65fd6VygiK4AVALNnz+7v/qs8EJ8CSNRv0xu1LUmHGU3EsH+fSpVlxdS5Qud4IjB5TAm724OI9A4UVWVF+IoGHiyi2Vp9pSkHNRjSBgsRORPYY4zZICInRScnWNRkMG/fBGNuBm4GWLRokaY6VFLx/TQl6rcplEGhspcx7Nd9RrRNQ4G4AmjPVTl5TAnNLrjE39BbA2H2PTclVllaRI8x+IM9SbOpCkRYMGVMyhSCpiLUUMkkZXEccJaInAGUYsssrgOqRKTIpS5mAtFxGmuBWUCtiBQB4wAdIV7lVH/6R2ruDCWcvnDKWNq6uqlrDcSynna3B5OuJ5PUQEVJEU9dsTTWqC76RGXo3XX40TXVKdejqQg1VNL+wowxPzDGzDTG1ABfBJ4xxnwZWAV81i12AfCIe/2oe4+b/4yWV6hc60//SNGLsrlzX60psE/40b6YEl24/eldo7KsuFejugOmjmXquP1HuFNquBpIO4srsYXdm7BlEre56bcBE9z0K4DvD2wXlUqtIxiOlT8I+/JBo91mpLO7rXc7h2Z/kB1NXUmXz/TRR9jX71N8/00FItoaW40ofbpCjTHPAs+615uBYxIsEwA+l4V9Uyqt+OqyXlMrS6ksK+5V4ymRaDuHzqBt51DfmjzLCTJvR1FZVszUcaVpt6/USKCPM2pE8Ta6a+4Msvz2tXR6AoX3Jl5ZVpyTHlkzzVMtKhDtEVaNGhos1LCUaBxrfyjssotscNjdFkx5427r6s5JP0sTKnwEwj0pazZB/wrdlRqu9GpWw1K0euzmhn2d8u1o8vfqtyndE34wHImlQgpTDRrRRxPHlrD2qmVMqSxJuZwOSqRGEw0WaljzDiDU1zp1zZ0h6ttsI7se12Kvvxe8SO/C6oqSIsZXlLj39BqUKFmhtlIjmWZDqZxJ1NI608/sbLaFzR/s3jeAUF/rX5sEH+pL072y4gJ8RYW0dnUztbKUw2dWsa3Jv1+DOG+7jPhC9fhGdN7/3lSTUsOdBguVFYkCQ6KW1ulEP1Pi8vsz7cIjWkNJ3It0H5s2rhRjDPVtyWs+FRQIU8eV0urKPm4676jEy7lqsInGoogPkt73p1//Qpq9VGr40GwolRWJyhgGoqcPHT2JEAsuUytLGV+RvqygqtxHSXFh4vWh41srFU9TFmpYyiRUeMeqji5fVe6jsSN1OwmwZSHJhkUFWPXdk5jshjlVSmnKQg0jNiurA8is5XWisaob2gPszSBYtHV1Jw9IAqve35PBHiQ2b1IFB02rTNvZX6bLKTUcaMpCDRubGzr36wk2lcqy4lg3H9099nN7O0JuXhGBUA+hnn0hoUBgfLmPyZWlhMLJh041Brbu7duodd6C60wL87VTQDWSaLBQObe5oYNLV76S8ubobZkN+0a0S6SwQOiJGD7Y3R4rAPcu/rUT5vKtZQcgEOvlddq4Up66YmmsD6YH1m7nn//4VsKyERGomVjep++oN3412mmwUDkXDEd4fmMDp1//Ai3+EFXlvl5P4LH+nYKZDR4UvcEniicCfHvZAbGg4K2l5O2s78zDp/PTP79DZ4JtCnDmYdPT7oeOLaHyiQYLNShC4Qjv1rUh0Gs0uo5gmOV3rE140+6PggLhsTd28YWjU4++OKakiDsvPCbWCaE3S2pWdXlGvcBqakLlEy3gVgPmzUJq8fceGyJefGLgsdd39Wso0WR6Iibj8oaja6pZe9UyprpaT6XFBRw4dSyHzhiXvR1SapTQlIUakPguwutaAxx29RMsnjcho8/f8sLmhN2L91e5r7BP5Q3RAYnqWgPMnTiGx791Qq/5mtWklKXBQvVbsiykiIE1mxt7VWuNL0iubfZz+vUvsLMl+SBD/SGSWXlDpjSrSSlLs6FUv6XKQooY269TVHztJn+oh3fr2gacBeUdFS/a6lpHnVMq+zRYqH7b2tiZMgspVY8d0ZRGqiqyyUQDw90XH6PjWCs1SPQRTPVbzYSKWJuH/upTH1DYAvKplfvaTCSrGhuVSZmDlksolZ4GC9VvZx4+nav+982cb0fEFlxPHFPCtkZ/0sCQSCZlDlouoVR6mg2lerl05Sucfv0LXLrylbTzxpQUMau6by2dM1XkGdkumsVU7tNnG6WGiv76VC+pxqCIn3fJPRvY2WxrMxVI5mNPZKKwQGLlGX1JSSilckNTFiqh9+raOPbnTyedv25rE0++Ux+7oWczUCSjvbQqNXT0cU0lZICmzn0d+3lbaTd3Brng9rU5DRCRBHVqE5UtaOG0UoNDg4VKK76VdqqhSLOluyezSKSF00oNDg0WKsY7+BDYaq272wJZ7ehPKTUyabBQMW/WtvYafCgcMZzwy2eQLIxD7SsUIqZvjfB8hQWUFhdoFpNSw4AGizxmUxKdzJtUQUN7kO0JxqS2I80NvHBi/mTbT1SymlbxROCn5xyStqtxpdTg0GCRB7xBwZvHH60KGzGG9+vbU6yh7woLoFCk17CmUQdOHUtbV3evcS3iFYhktUNApdTAaNXZPBANCpsbOhPO39aYePpAHDClMpaaiFcgQlW5D4A5E8qpKCkkmtMV7ffpgRVLtG2FUsOIBgtFyFNOkS2Jqr4mUu4r6jUAkXYIqNTwpMEiT3nbTeSivcSHezpo8YcoKUp/iUU7BARtra3UcKW/yjz0pVtWs2ZzY04b1YUjhqpyH0fOGR8rL0mWDaaUGv40WOSZjmA454EiqsUf4pefOZYxLqUQ7YBQA4dSI48Gi1EqWgOqxR+Kddvxfn0bi655clACBUB9W4DFP3uKOy88hqNrqnvVxPIGDqXU8KfBYhRIVDU2WgOqpKgg1tAuYiDQnf3C7GSMgc5gD8vvWMvaq5b1KouI76ZD+3hSanjTYDEKpOpWPJiDmk7xoiPYRf/HMwYee2NXygZ22seTUsOb1oYaBXY2+3v97wiG2d3aNWjbN3H/4/lDPWzdu3/rcKXUyKEpi1EgmnroCIZZ+h+r2N7oz0IHHclTCvstJzb1EP0fr9xXSM3E3Iyop5QaHGlTFiIyS0RWici7IvK2iHzLTa8WkSdF5AP3f7ybLiLyGxHZJCJviIjmL3ikGrZ0oCIGtmUrUAhMcQ3loPcwp/stG/c/0bq06w6lRrZMsqHCwHeNMQcBS4BLReRg4PvA08aYhcDT7j3A6cBC97cC+G3W93oES9f1Rl/tbgvQ3ZP9conZ1eWMr7AN5UqKCij3FQJ2+NT4rjkWz5vAQdMqWTxvQsKuO+688BhtaKfUCJf2F2yMqQPq3Ot2EXkXmAGcDZzkFrsLeBa40k2/2xhjgNUiUiUi09x61AB5az4tnlvNDx95O+vbKBA4dMa4WECbN2lMrG3EnOpyXq9toa41wNTKUp66YmmvQNAZDLPs188lna+UGpn69CsWkRrgCGANMCUaAIwxdSIy2S02A9jh+Vitm9YrWIjICmzKg9mzR1c31Ml6eR3oejqCYV7Z1kxda4DGjiB/fiM38ffgaZXceO6RvdpCeL/H6de/QF1rIGHXHNGuO5LNV0qNTBn/kkVkDPB74NvGmLYUA+IkmrFfNrox5mbgZoBFixYNUjOxwZGqKmsmokFiZ7OftkCYzQ0dfOmW1bxR2xIb2nRPe26GNhXgwUuOBZJXZ9U2EUrln4yChYgUYwPFfcaYP7jJu6PZSyIyDdjjptcCszwfnwnsytYO5wNvgzqwtZ1e/rBxULY9e0J52tRAutSSBhOlRp+0wUJsEuI24F1jzK89sx4FLgB+4f4/4pl+mYg8ACwGWrW8InMdwTBb9tpxsHPRdXg65b6BZxtpAzulRp9M7gzHAecBb4rIa27aVdgg8ZCIXAxsBz7n5v0FOAPYBPiBC7O6x6PYuq1NLL9jbaxLjsHOm5s2rlRTA0qphDKpDfUiyavQn5pgeQNcOsD9yjsdwTBfvHk1PTns5a/cV0iguydhR4IVJYVac0kplZR29zFMPLyhlkiOu4OdVV3Og5ccq20hlFJ9psFiEHlHp2vxh+gIhgGb/XTNY2/nNNtJxDauO7qmWocxVUr1mQaLQbJuaxOLf/YU9W0BYN9YD89vbGD5HWvJdVm2MXDc/ImADmOqlOo7vUsMgkTlEd6xHhJ1vpdt2pmfUmogNGWRZYmymh57fReRJBEhYrJX66lAtDM/pVRuaLDIomhWU12rzWqqa7VZTS9vbsx56sFXKDx4ybHMntA79aAF2EqpbNC7R5Z0BMMsv2MtncGeXtM7gz08/lbu2yS+eOUpTK4s5dAZ4ygtLmRbYyeB7oh25qeUygpNWWTJY6/vSpp66A7nNllR7itksqvddOO5R/LEt09k7sQxgBZgK6WyQ+8iWbK1sTPWyV+8bIeKArFlHUUFQjhimDOh762utf8mpVRfaLDoh0RdkNdMqMBXWEAoBwMRRfmKCpg7sYK6li7aAmEKXbDoD+2/SSnVFxosMuQNEPFdkF+68hVe39GS00BRWACv/usnqCgpiu1Liz9EVblPUwdKqZzTYJGhfcOhdsS6Do/atKeD2uaunG27QOCBFcfGyh40VaCUGmxawN1HwXCEoKe5dUcwTL2rKpstN59/FNPGlcbeL5wyts/dccybVMFB0yo11aGUygpNWWTI26guOs7EtsZODv/xX7PWU6wAa646lcmVpVz75Aex9hoFyUclTEpTH0qpbNKURQbWbW3ig93tsffR0OAP9WS1S/G7LjomVgV23qSK/bK7lFJqqOjdKI1ov0457j2cRXPGc+IBk2Lvbzz3SOZNGpPbjSqlVIY0WKRxwzMf5HRAIoAzPzqVh7/x8ZxuQymlBkLLLOJ4q8j+8jOHcesLm3O6PQFO8KQovLThnFJquNBgEef5jQ20B8K8V9fGa9tbbE98ORyWqKBAkvYGq4XUSqnhQoOFRzRQgA0PO1ty13ZCxPbppL3BKqVGAr1LOc9vbOD829fmdBvlvgIqS33UtwW0N1il1IiiBdzYGk8X3bUup9uoKCnkrosWc1TNeA6aVsmRc8ZroFBKjRh6twIe3lBLuCd35RIzq8p44jsnUlFS1OeW2EopNRxoygJY9d7unK27oqQwFiiUUmqk0mABvL6jdcDrOGjaWCpKCikssF1zFBaIDmeqlBo18voudunKV3ivro2Wru5+r8Pbn1NnMMxjb+xi614/NRPLOfOw6RoolFKjQt7eyTqC4V5VZfvrkqXzYv05VZQU8YWjZ2dj95RSaljJi2ARP7Lduq1NnHvzy3QPcKyicl8B/3DKwuzspFJKDWN5ESyiAxfVNvs57brn+WB3+4A7BiwsEO66aLFmMyml8sKovtNduvIVNu3pYFtjJwDtgTDv17en+VRyBQJV5T5mji/j/q8t0UChlMobo/pu99bOVrY1+ge8Hm/XHNpOQimVj0ZtsOgIhtnRNLBAMXdiBW1d3czQlIRSKs+NyrtfRzDMVb9/Y0DlEndffAwnLkzcdbhSSuWbURcs1m1t4iu3riYY7n+kWHbQZA0USinlMaqCRXugmy/fuppQPwNFUYFw/4olWi6hlFJxRk13Hx/sbufT//23fgcKgOUfr9FAoZRSCYz4lEWrv5trn9rI3S9vHVAZRbmvkIVTxmRtv5RSajQZscGiJ2JYuXY7//nX92nx979vpygRkg5vqpRS+W5EBouXP2zkx396m/cG0MAuSoc3VUqp9HJydxSRTwLXA4XArcaYX2RjvTua/PzsL+/y+Fv1djvAzPFl7Gju21jZBQJzJmgbCqWUylTW75AiUgjcCHwCqAXWicijxph3+rtOfyjMTc9+yO+e30wwbHv/O2J2FT8+6xAuve+VPq2rqED42/dPifUUq5RSKr1cPE4fA2wyxmwGEJEHgLOBPgcLYwyPvr6Ln//lPerbAgBMqSzhqjMO4qzDpyMilPkKM15fSZFw71eXaKBQSqk+ykWwmAHs8LyvBRbHLyQiK4AVALNn7z8GxJu1rVz9p7fZsK0ZAF9hAZcsncfXl87vlWX02SNn8LPH30+7UxcfP5crPnGAZjcppVQ/5OLOKQmm7Vep1RhzM3AzwKJFi2LzG9qD/McT7/E/62tjH/rkIVP5508dxKzq8v1WfO6SGn71142EehLXmz1p4SRu/MqRGiSUUmoAcnEHrQVmed7PBHal+1AoHOHOv23h+qc+oDPUA8ABU8Zw9VmH8PH5E5N+bkxJEfd9bQnn37aaQLeJBZhCgd+edxR/d/DU/n8TpZRSAIgxAxwFKH6FIkXARuBUYCewDjjXGPN2ss8ceOjhZuoF18W6Ex9XVsw/nnYgXzp6FkWFmTUy1/GvlVL5RkQ2GGMWDca2sn43NcaEReQy4Als1dnbUwUKgK2NfoKNfgpF+MqS2XznEwdQVe7r03Z1/GullMqdnDx6G2P+AvylL585bv4EfnTWIRwwZWwudkkppdQADIt8mjnV5dz71cWIJCobV0opNdSGRa+zlWXFGiiUUmoYGxbBQiml1PCmwUIppVRaGiyUUkqlpcFCKaVUWhoslFJKpaXBQimlVFoaLJRSSqWlwUIppVRaWe9IsF87IdIAbMvCqiYCe7OwntFIj01yemwS0+OS3HA5NnOMMZMGY0PDIlhki4isH6weGEcaPTbJ6bFJTI9Lcvl4bDQbSimlVFoaLJRSSqU12oLFzUO9A8OYHpvk9Ngkpsclubw7NqOqzEIppVRujLaUhVJKqRzQYKGUUiqtURMsROSTIvK+iGwSke8P9f4MJhGZJSKrRORdEXlbRL7lpleLyJMi8oH7P95NFxH5jTtWb4jIkUP7DXJPRApF5FURecy9nysia9yxeVBEfG56iXu/yc2vGcr9zjURqRKRh0XkPXf9HKvXDYjId9xv6S0RuV9ESvP9mhkVwUJECoEbgdOBg4EvicjBQ7tXgyoMfNcYcxCwBLjUff/vA08bYxYCT7v3YI/TQve3Avjt4O/yoPsW8K7n/S+Ba92xaQYudtMvBpqNMQuAa91yo9n1wP8ZYz4CHI49Rnl93YjIDOByYJEx5lCgEPgi+X7NGGNG/B9wLPCE5/0PgB8M9X4N4fF4BPgE8D4wzU2bBrzvXv8O+JJn+dhyo/EPmIm96Z0CPAYItvVtUfz1AzwBHOteF7nlZKi/Q46OSyWwJf775ft1A8wAdgDV7hp4DDgt36+ZUZGyYN/Jjap10/KOSwIfAawBphhj6gDc/8lusXw7XtcB3wMi7v0EoMUYE3bvvd8/dmzc/Fa3/Gg0D2gA7nBZdLeKSAV5ft0YY3YCvwK2A3XYa2ADeX7NjJZgIQmm5V2dYBEZA/we+LYxpi3VogmmjcrjJSJnAnuMMRu8kxMsajKYN9oUAUcCvzXGHAF0si/LKZG8ODaujOZsYC4wHajAZsHFy6trZrQEi1pgluf9TGDXEO3LkBCRYmyguM8Y8wc3ebeITHPzpwF73PR8Ol7HAWeJyFbgAWxW1HVAlYgUuWW83z92bNz8cUDTYO7wIKoFao0xa9z7h7HBI9+vm2XAFmNMgzGmG/gD8HHy/JoZLcFiHbDQ1VbwYQujHh3ifRo0IiLAbcC7xphfe2Y9ClzgXl+ALcuITj/f1W5ZArRGsx1GG2PMD4wxM40xNdjr4hljzJeBVcBn3WLxxyZ6zD7rlh91T4kAxph6YIeIHOgmnQq8g14324ElIlLuflvR45Lf18xQF5pk6w84A9gIfAj881DvzyB/9+Oxyd43gNfc3xnYfNOngQ/c/2q3vGBrj30IvImt9THk32MQjtNJwGPu9TxgLbAJ+B+gxE0vde83ufnzhnq/c3xMPgasd9fOH4Hxet0YgB8D7wFvAfcAJfl+zWh3H0oppdIaLdlQSimlckiDhVJKqbQ0WCillEpLg4VSSqm0NFgopZRKS4OFUlkiIt8WkfKh3g+lckGrziqVJa6V+CJjzN6h3helsk1TFiqviMj5biyG10XkHhGZIyJPu2lPi8hst9ydIvJZz+c63P+TRORZzxgQ97kWzZdj+xFaJSKrhubbKZU7RekXUWp0EJFDgH8GjjPG7BWRauAu4G5jzF0ichHwG+CcNKs6AjgE2zfQS259vxGRK4CTNWWhRiNNWah8cgrwcPRmboxpwo5LsNLNvwfbdUo6a40xtcaYCLZrlZoc7KtSw4oGC5VPhPRdR0fnh3G/D9eZnM+zTNDzugdNoas8oMFC5ZOngc+LyASwY5QDf8P2RgvwZeBF93orcJR7fTZQnMH624Gx2dpZpYYTfSJSecMY87aI/BvwnIj0AK9ix1q+XUT+CTtq3IVu8VuAR0RkLTbIdGawiZuBx0Wkzhhzcva/gVJDR6vOKqWUSkuzoZRSSqWlwUIppVRaGiyUUkqlpcFCKaVUWhoslFJKpaXBQimlVFoaLJRSSqX1/wF/2jcTAQfz3AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sn.regplot(y_test,np.exp( y_pred), x_bins = 200)\n",
    "ax.set(title = \"Comparison between the actual vs predicted values from KNN\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### XGBoost Regressor"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "XGBoost is a decision-tree-based ensemble Machine Learning algorithm that uses a gradient boosting framework. Gradient boosting is an approach where new models are created that predict the residuals or errors of prior models and then added together to make the final prediction. It is called gradient boosting because it uses a gradient descent algorithm to minimize the loss when adding new models. The key features of this algorithm are :\n",
    "1. Sparse Aware implementation with automatic handling of missing data values.\n",
    "2. Block Structure to support the parallelization of tree construction.\n",
    "3. Continued Training so that you can further boost an already fitted model on new data.\n",
    "\n",
    "Our RMSLE score was not improved with the implementation of this algorithm so we did not apply it to our model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Anaconda\\lib\\site-packages\\xgboost\\core.py:587: FutureWarning: Series.base is deprecated and will be removed in a future version\n",
      "  if getattr(data, 'base', None) is not None and \\\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[13:47:52] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n"
     ]
    }
   ],
   "source": [
    "xg_reg = xgb.XGBRegressor()\n",
    "xg_reg.fit(X_train, y_trainlog)\n",
    "y_pred= xg_reg.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "True in (y_pred < 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i, y  in enumerate(y_pred):\n",
    "    if y_pred[i] < 0:\n",
    "        y_pred[i] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSLE: 0.3890022022240694\n",
      "R2: 0.8388085886587142\n"
     ]
    }
   ],
   "source": [
    "print('RMSLE:', np.sqrt(mean_squared_log_error(y_test,np.exp( y_pred))))\n",
    "print('R2:', r2_score(y_test, np.exp( y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Comparison between the actual vs predicted values from XGB')]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sn.regplot(y_test,np.exp( y_pred), x_bins = 200)\n",
    "ax.set(title = \"Comparison between the actual vs predicted values from XGB\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forest Regressor"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A random forest is a meta estimator that fits a number of classifying decision trees on various sub-samples \n",
    "of the dataset and uses averaging to improve the predictive accuracy and control over-fitting\n",
    "The default values for the parameters controlling the size of the trees (e.g. max_depth, min_samples_leaf, etc.) \n",
    "lead to fully grown and unpruned trees which can potentially be very large on some data sets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Anaconda\\lib\\site-packages\\sklearn\\ensemble\\forest.py:245: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "clf = RandomForestRegressor(n_jobs=-1, random_state=0)\n",
    "clf.fit(X_train, y_trainlog)\n",
    "y_pred = clf.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- n_jobs : The number of parallel jobs to run for neighbors search,-1 means using all processors.\n",
    "- Controls both the randomness of the bootstrapping of the samples used when building trees and the sampling of the features to consider when looking for the best split at each node, 0 is a popular integer random seed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "True in (y_pred < 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i, y  in enumerate(y_pred):\n",
    "    if y_pred[i] < 0:\n",
    "        y_pred[i] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSLE: 0.34509249950869675\n",
      "R2: 0.9297074367985756\n"
     ]
    }
   ],
   "source": [
    "print('RMSLE:', np.sqrt(mean_squared_log_error(y_test,np.exp( y_pred))))\n",
    "print('R2:', r2_score(y_test, np.exp( y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Comparison between the actual vs predicted values from Random Forest Regressor')]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sn.regplot(y_test,np.exp( y_pred), x_bins = 200)\n",
    "ax.set(title = \"Comparison between the actual vs predicted values from Random Forest Regressor\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forest Regressor with GridSearch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters = {'n_estimators': [200, 400], 'max_depth': [15, 25]}\n",
    "clf_cv = GridSearchCV(clf, parameters, cv=5, n_jobs=-1)\n",
    "clf_cv.fit(X_train, y_trainlog)\n",
    "y_pred = clf_cv.best_estimator_.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- So we use \"n_estimators\" for the number of trees in the forest and max_depth for the max depth of these trees.\n",
    "We put these parameters in arrays and our algorithm(in get score method) will choose the best values for our estimator.\n",
    "- cv=5 : Determines the cross-validation splitting strategy. 5 is the number of folds in a (Stratified)KFold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "True in (y_pred < 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i, y  in enumerate(y_pred):\n",
    "    if y_pred[i] < 0:\n",
    "        y_pred[i] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSLE: 0.32624763419688274\n",
      "R2: 0.9375337258790005\n"
     ]
    }
   ],
   "source": [
    "print('RMSLE:', np.sqrt(mean_squared_log_error(y_test,np.exp( y_pred))))\n",
    "print('R2:', r2_score(y_test, np.exp( y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Comparison between the actual vs predicted values from Random Forest Regressor')]"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sn.regplot(y_test,np.exp( y_pred), x_bins = 200)\n",
    "ax.set(title = \"Comparison between the actual vs predicted values from Random Forest Regressor\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  Gradient Boosting Regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "gbm = GradientBoostingRegressor(n_estimators=4000,alpha=0.01)\n",
    "gbm.fit(X_train,y_trainlog)\n",
    "y_pred = gbm.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- alpha : These regularization parameter limits how extreme the weights (or influence) of the leaves in a tree can become."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "True in (y_pred < 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i, y  in enumerate(y_pred):\n",
    "    if y_pred[i] < 0:\n",
    "        y_pred[i] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSLE: 0.2959370381825778\n",
      "R2: 0.9457606742413951\n"
     ]
    }
   ],
   "source": [
    "print('RMSLE:', np.sqrt(mean_squared_log_error(y_test,np.exp( y_pred))))\n",
    "print('R2:', r2_score(y_test, np.exp( y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Comparison between the actual vs predicted values from Gradient Boosting Regressor')]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sn.regplot(y_test,np.exp( y_pred), x_bins = 200)\n",
    "ax.set(title = \"Comparison between the actual vs predicted values from Gradient Boosting Regressor\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We fit the 2 regressors (train our model)\n",
    "gbm as GradientBoostingRegressor\n",
    "and \n",
    "clf_cv as RandomForestRegressor with GridSearchCV.\n",
    "GridSearch let us combine an estimator with a grid search preamble to tune hyper-parameters.\n",
    "This method picks the optimal parameter from  grid search and uses it with our estimator(RandomForestRegressor)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_score(X_train, X_test, y_train, y_test):\n",
    "    y_trainlog = np.log1p(y_train)\n",
    "    clf_cv.fit(X_train, y_trainlog)\n",
    "    gbm.fit(X_train, y_trainlog)\n",
    "    y_pred1 = clf_cv.best_estimator_.predict(X_test)\n",
    "    y_pred = gbm.predict(X_test)\n",
    "    print('RMSLE:', np.sqrt(mean_squared_log_error(y_test,np.exp( y_pred))))\n",
    "    print('R2:', r2_score(y_test, np.exp( y_pred)))\n",
    "    print('RMSLE:', np.sqrt(mean_squared_log_error(y_test,np.exp( y_pred1))))\n",
    "    print('R2:', r2_score(y_test, np.exp( y_pred1)))\n",
    "    #return np.sqrt(mean_squared_log_error(y_test,np.exp( y_pred)))\n",
    "    return y_pred"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Cross Validation\n",
    "Cross Validation improved our score a lot.\n",
    "Its a very usefull tool to avoid overfitting and also gave us an idea of how our code will perform on other data.\n",
    "Splits' performance in cross validation is related with the amount of data used in the model.\n",
    "We tried many combinations about cross validation but the 10-split gave us the best results,\n",
    "more or less splits will end up scoring less points."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "kf = StratifiedKFold(n_splits = 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We observed that there is a relation between 'count' ,  'registered'  and 'casual' (registered + casual = count) so we tried to predict 'registered' and 'casual' separately  and then we  added the results of these 2 predictions. Nevertheless the results were worse than before (prediction of 'count' without the mixture of 'registered' and 'casual') so we did not apply this to our model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Anaconda\\lib\\site-packages\\sklearn\\model_selection\\_split.py:657: Warning: The least populated class in y has only 1 members, which is too few. The minimum number of members in any class cannot be less than n_splits=10.\n",
      "  % (min_groups, self.n_splits)), Warning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSLE: 0.2715540058182328\n",
      "R2: 0.9617599993848172\n",
      "RMSLE: 0.2902213921319653\n",
      "R2: 0.9441878029742011\n",
      "RMSLE: 0.27987900005224275\n",
      "R2: 0.9542517335403785\n",
      "RMSLE: 0.3078838792966988\n",
      "R2: 0.9416330459308445\n",
      "RMSLE: 0.284590531031506\n",
      "R2: 0.9403240565644932\n",
      "RMSLE: 0.3124018559314789\n",
      "R2: 0.937229033138751\n",
      "RMSLE: 0.29983443027317014\n",
      "R2: 0.934504982151679\n",
      "RMSLE: 0.31692625585603335\n",
      "R2: 0.9273686475412997\n",
      "RMSLE: 0.3091153373829097\n",
      "R2: 0.9396214585057495\n",
      "RMSLE: 0.3325847789005261\n",
      "R2: 0.9334417985897975\n",
      "RMSLE: 0.31504412331684223\n",
      "R2: 0.9447004136446796\n",
      "RMSLE: 0.331744621251857\n",
      "R2: 0.9302901618097535\n",
      "RMSLE: 0.3088888835495268\n",
      "R2: 0.9322636721527194\n",
      "RMSLE: 0.33848161258009357\n",
      "R2: 0.9218556643415223\n",
      "RMSLE: 0.31668820320065644\n",
      "R2: 0.9279689118225144\n",
      "RMSLE: 0.34385492271479606\n",
      "R2: 0.9298580603807783\n",
      "RMSLE: 0.31802210599515107\n",
      "R2: 0.9260928858404053\n",
      "RMSLE: 0.3387523284743355\n",
      "R2: 0.9244622133820288\n",
      "RMSLE: 0.3190798207314306\n",
      "R2: 0.9005578021079977\n",
      "RMSLE: 0.3615811165500343\n",
      "R2: 0.8887062582116336\n"
     ]
    }
   ],
   "source": [
    "X = np.array(df_train.drop(df_train[[ 'count']], 1))\n",
    "\n",
    "y = df_train['count']\n",
    "\n",
    "\n",
    "\n",
    "gbm = GradientBoostingRegressor(n_estimators=4000,alpha=0.01)\n",
    "\n",
    "sc =[]\n",
    "y_pred=[]\n",
    "y_pred1=[]\n",
    "y_t=[]\n",
    "\n",
    "clf = RandomForestRegressor(n_jobs=-1, random_state=0)\n",
    "parameters = {'n_estimators': [200, 400], 'max_depth': [15, 25]}\n",
    "clf_cv = GridSearchCV(clf, parameters, cv=5, n_jobs=-1)\n",
    "\n",
    "for train_index,test_index in kf.split(X,y):\n",
    "    X_train, X_test, y_train, y_test = X[train_index],X[test_index],y[train_index],y[test_index]\n",
    "    y_t.append(y_test)\n",
    "    y_pred.append(get_score(X_train, X_test, y_train, y_test))\n",
    "    \n",
    "    \n",
    "    \n",
    "    \n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(11257, 22)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(11257,)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Importance from Gradient Boosting Regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2.67656532e-02 7.63018661e-01 1.53748251e-03 1.08289457e-02\n",
      " 6.17872643e-02 1.47211345e-02 6.96985775e-02 1.60761878e-02\n",
      " 1.56799773e-02 1.32956583e-02 1.12655126e-03 6.11208622e-04\n",
      " 3.32888700e-04 8.01344916e-04 5.39975246e-04 2.12670754e-04\n",
      " 9.22856655e-05 2.89319542e-04 1.35153184e-04 3.61349847e-04\n",
      " 5.53295860e-04 1.53441422e-03]\n"
     ]
    }
   ],
   "source": [
    "###importance of features for Gradient Boosting algorithm \n",
    "importances = gbm.feature_importances_\n",
    "print(importances)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Feature ranking:\n",
      "1. feature hour (0.763019)\n",
      "2. feature temp (0.069699)\n",
      "3. feature workingday (0.061787)\n",
      "4. feature season (0.026766)\n",
      "5. feature humidity (0.016076)\n",
      "6. feature year_0 (0.015680)\n",
      "7. feature weather (0.014721)\n",
      "8. feature year_1 (0.013296)\n",
      "9. feature weekday (0.010829)\n",
      "10. feature holiday (0.001537)\n",
      "11. feature month_12 (0.001534)\n",
      "12. feature month_1 (0.001127)\n",
      "13. feature month_4 (0.000801)\n",
      "14. feature month_2 (0.000611)\n",
      "15. feature month_11 (0.000553)\n",
      "16. feature month_5 (0.000540)\n",
      "17. feature month_10 (0.000361)\n",
      "18. feature month_3 (0.000333)\n",
      "19. feature month_8 (0.000289)\n",
      "20. feature month_6 (0.000213)\n",
      "21. feature month_9 (0.000135)\n",
      "22. feature month_7 (0.000092)\n"
     ]
    }
   ],
   "source": [
    "indices = np.argsort(importances)[::-1]\n",
    "print(\"Feature ranking:\")\n",
    "features=['season','hour','holiday','weekday','workingday','weather','temp','humidity','year_0','year_1','month_1','month_2','month_3','month_4','month_5','month_6','month_7','month_8','month_9','month_10','month_11','month_12']\n",
    "for f in range(X.shape[1]):\n",
    "    print(\"%d. feature %s (%f)\" % (f + 1, features[indices[f]], importances[indices[f]]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x23109ca7da0>"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "(pd.Series(gbm.feature_importances_, index=features)\n",
    "   .plot(kind='barh')) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Comparison between the actual vs predicted values from Gradient Boosting Regressor for all cross-validation splits."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x23109890908>"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA20AAAV8CAYAAACoqHfOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nOzde3jcZZ3//+c9Mzk056ZpS5mcmrYUW6SVLSiKrhxUqEoXrETgu3SlyrriAWF/ntb9rXutsoKK8hVRoawWgSVYURCqoICHggptoba0lLY5Nz2QNIfJaTKH+/vHHDpJZiYzyaRJ2tfDq1cy8znMPWVM++r9vt+3sdYiIiIiIiIi05NjqgcgIiIiIiIiiSm0iYiIiIiITGMKbSIiIiIiItOYQpuIiIiIiMg0ptAmIiIiIiIyjSm0iYiIiIiITGOuqR4AQFlZma2urp7qYYiIiIiIiEyJbdu2tVtr58Y7Ni1CW3V1NVu3bp3qYYiIiIiIiEwJY0xTomMqjxQREREREZnGFNpERERERESmMYU2ERERERGRKdQ31Jf0+LRY0yYiIiIiInKy6vH20NDZQH1nPfWd9TR0NdDc3UxTdxOtPa0cGziW9HqFNhERERERkXGy1tI12EVjVyMHOg9Q31lPY1cjTV1NNHc309zTTI+3Z0KvodAmIiIiIiInJY/Xw0X3X0RzVzOVJZU8e92zFOYUpnUPay0dAx3HZ8q6hoey1p5WPEOelO6V48yhvKicyuJKqkqqqC6uZuHshdTMruGdX31nwusU2kRERERE5KSzpXkLqx9cTd9QH0GCtPe3477DzeZrN3NB5QXR86y1HO07OmymrKmricbuxmgo6/f1p/SaeVl5x0NZcRXVJdXUzK6hpqSGhbMXMi9/HsaYtN+LQpuIiIiIiEyZ2k217G3fy9KypdStrcvIPT1eD6sfXD1sBixIEM+Qh4s3XsxVy6+izdNGc08olA36B1O6b0F2ARVFFVQWV1JdUk1VSRU1JTXUzA6Fsjmz5owrlI1FoU1ERERERKaEx+vhhZYXaO1ppWOgA4/Xk3b5YiAYoM3TFipf7KqnobOB39b/lj5f/I6MQ8EhHtj5QNxjxTnFoVBWEpopW1iykIWzF7Jo9iKqS6opyS2ZlFA2FoU2ERERERE54WLLFwHaetrili/6g35ae1pp7AyVLzZ0NYTWlHWH1pS1edrwB/1pvfbi2Yu5dMmlLCxZGJopK62hqriK4tzijL7HTFFoExERERGRjDTtSOe1EpUvXrTxItYsXcNBz0Faelo45DlEwAZSum9ZXhmVRZUECbLryC78dnSYy8/K54sXfJH156zP2PuZbAptIiIiIiLTTKbWeXm8HuperWNfxz6WzFlC7fLauEEs1aYd4zHoH6S5u5nGzsZo98VnG55NuKG0L+hj055NcY+dVnAaFUUVVBVXhbovllSzaPYiFs5eSGVxJXlZedH37b7DHbero8M4qD2rdkLv6URTaBMRERERmUYysc4Ljgexfl8/ARvAaZzc/NTNo4JYslmv1Q+upu2WNgqyCxK+zoce+RCvHn2VefnzqF1eGy1fbO5uprm7mSN9R9Ie+5lzzmTNmWtC5Yuza6guqaaiuIJcV25K1xfmFLL52s2j3n9eVh6br92c9P1MR8ZaO9VjYNWqVXbr1q1TPQwRERERkSk1csbLgYP87Py0Z7ySzTQVZhcOC2Ibtm/gpt/cFLdxR35WPrddchvvrn43DZ0NHOg6QGNXI42dx9eUdQx0pDQmp3GyoHABVcVV+II+trdtT1i+eOeld2akfLF3qJe6XXXsP7afxaWLqT2rdtoGNmPMNmvtqnjHNNMmIiIiIjINTHTGK1bdq3UEbTDusaANUrerLhqKdh3ZlbDTYp+vj0/9+lMpvweD4ez5Z3P2/LNDe5SF9yerLqnGXeTG5XBF3+uJKF8syC6YUWvXElFoExERERGZBtIJWmPZ17EvaRC7dcutfPev36Wlu4Vub3dK98x2Zoc2ji6q5PTC09m0exNDwaFh51gs9Z31bLl+S9KAGVu+GG9WcbrOhk0VhTYRERERkWlgrKC1/9j+Uc9ba2nvbw+VLHaFWuI3djWypXkLBoMl/lKo+s76lMaU48zhyWueZNncZcwvmI/DOIBQSeVjex8bFdog9YB5QeUFtN3SxrLvL6OlpwV3kZvdN+5WYItDoU1EREREZBpYMmcJ+Vn5cYNbrjOXYwPHuPVPt0YDWkNnA/Wd9QSJPzuXiNM4uW7FdZwx54zo5tFHeo9wzc+vod/Xn9JauvEEzHgKsgsonVVKS08LpbNKFdgSUGgTERERkRkpU23xp3IsQRvkkOcQTd1NGAy+gC/ueYOBQe7Zfs+Y91tYspAzy86kuqQagB+/8mO8fi8Wi8FQkF2QMIgd+tdDKc96JQuY+Vn5LC5dPOZYI5aWLR32VUYbM7QZY5YCsZ+8GuD/B0qAjwNvhJ//srV2c/iaLwHrgQDwGWvtU5kctIiIiIicnFLd4DlTbfEzIdlYAsEAbZ620MxYVwO3P387bZ42XA4XBdkFHPQcZCgwusQwnpLcEiqKKnAXunmm4Rl8wdEBr72/nUc+/Eg0bN3+ntu5cOOFNHc3U1lcyXPrnksYxAqyCzi/4nxK20tZWrY06axX7fJabn7q5rjH0m0kMtWBeyZIq+W/McYJHATeCnwU6LXWfmvEOcuA/wXOA04HfgecYW3ibczV8l9EREREUm13P562+KluMp2u3zf+ng889AH6ff3R9WMu42LZvGV0D3Zz0HMQf3B0W/t45syaQ0VxKJT9ofEP9Pp6KZtVxhPXPMGb5r6JopwiYOwW/Zlqlz+WTG1PICGZbPl/MXDAWttkjEl0zhrgYWutF2gwxuwnFOD+nOZriYiIiMgpItV29+Npi5/qJtPxxvTuje+msbORsvwyPnPeZ2jztNHQ1UBTdxONXY20edpGXee3fv525G+jni/LK6N7sBtf0EdBdgH/+e7/jJYyVhVXkZ+dHz135Q9XsuPIDtxFbt5a/tZh98nUerKJUiOREyfd0PYRQrNoEZ8yxlwHbAVusdZ2Am7gLzHntIafG8YYcwNwA0BlZWWawxARERGRmShR+WOq7e7TbYsfL+QFbCAa8uo/W8+xgWPRxh4NXQ00dDaw641d7HljT3T27NjgsbT2K3MZF1efdTXXnn0t1SXVtPa0ckXdFQSCoeKz/qF+vvr7r7L52s0sm7ss5ftCZteTTVQ6JZUyfimHNmNMNnA58KXwUz8A/guw4a/fBq4H4k3BjarBtNbeA9wDofLItEYtIiIiIjPOyHK69v523He42Xzt5pRnj9KdZbp/x/0JyxM9Qx7mfnNuWu/BYRxc+aYrOaP0DLYf2s5vDvwm7nl+68dd5OZ9i9+Hx+vh3HvPTWt2MFlzjkyuJ8sErUmbfOnMtF0GbLfWHgGIfAUwxtwLPBF+2ApUxFxXDoyeNxYRERGRjEm1gcdUGaus8daLb01p9ijZLFOWI4tth7ZxZd2VNHU30dzdTHt/e0rjcxon7iI3lUWV+IN+trZtxW9Hh71ZrllcuuhS1p+zng3bN/Cn5j+NOebxbJqdLAhpY+pTjyONc68mpjTSGLMg5tgVwK7w948DHzHG5BhjFgJLgBcnOlARERGRU11kxmb+N+eHZm68oQC0pXkL7jvcbG/bztH+o2xv2477DjdbmrdM8YiPGyu4GEx04+aRrLXkZ+fz3b98l21t2xj0D8Y9zxf08dv63/KL137B9kPbxwxs2c5sPv/2z9P42UYGvzJI001N/On6P/GuqnfFDWwwfDavdnltwjHHznhNxhq0yHoyd1FoFZK7yE3bLW1qAHKSSmmmzRiTB7wH+OeYp283xqwkVPrYGDlmrX3VGPMIsBvwAzcm6xwpIiIiImNLVFq46apNrH1kbVqld+Mx0T3RxgouzzQ8w7oV6/jRth+NamXf7+/n6p9fPeZrnF54OmfOCTX2qJldQ3VJNfPy53FF3RVxXzvHmcO///2/j/r9SXXNWKozXpO1Bk3ryU4dabX8nyxq+S8iIiLT1XTYwNnj9eC+wz0smEXkOHNwOVyT2v7d4/Ww7O5ltPa0Ul5Uzu5P7o5bejny98payxv9b9DU1cR9L9/H/7z8P3H3FkvFLNcsKosrqSyupLqkGnehm++9+D06Bjo4veB09nxqT7Ql/kjxukfmZeUl7B6Z7Pe7MLtwVBDuHeqNdlCsKKoY1UEx3fvJqSmTLf9FRERETil72/ey48iOKR1DstLCQDCAN+CNe2xk6d14AmjsDB9AW09btHnIOyrewZG+IzR2NdLU1cSfmv7Eod5DHOg8wJl3nUlzdzMD/oGUXicvK4+q4iqqSqrY17GPfl8/1SXV3HnpnVSVVDE3by4jt5z6xWu/oGOgg7n5cxMGNjheSli3q479x/azuHQxtWfVJgxK6a4ZG2vGS2vQZKIU2kRERESmuWSlhX7rJ8uRFXcGa2Tp3VgBdGSo6x7s5tIHLh322pHSy3f9+F1kO7PjBsbeoV72duwd9lxhdiFleWU0dzcTCK+cMRhmuWbxyIcfYfWS1aNC2ViSdVgcqSC7IK0Zx0jQu3DjhTR3N1NZXMlz655LGLDGCsHa00wmQqFNREREZBxOZLfGZGui8lx5BAnGDW1jtX8PBAMc9ByMzpT9sfGPHO47zP5j+6m5s2ZYwBrJYocFttm5sxnwDzDoH6Qsr4yvvPMroU2jS6qoLqmmJLcECAW6VIPQWCa7XLUgu4CXPv5SRu9XOquUlp4WSmeVKrBJyhTaRERERNKUbL+xyBqpZKEu3TLFZPtyOR1OfnHVL1j7yNpRpXePX/04b/S9wUsHX6Kpu4nDvYcB2H9sP1XfraLN0xZ3D7M+Xx8NXQ1jjmvtm9byH+/+D6qKqwCi695yXblc/5br44bYTAehmSad2UGRCDUiEREREUkgXgMOYMymEq8cfiXh+qUV81ek1NRjpJFB0YGDvKw87r38XubmzeW1jtf4+h+/zrGBY2Q5sijKLeJw7+GEa+FGMhgsx/9emOPM4QNnfIDN+zbHXZcW2+Qk3tgi71ct6EVSk6wRiUKbiIiInPQm2oAjNoj8y7n/wvdf/H7Cbo23XXIbX3rmS3FD3SzXLJwOJ/1D/SmFmwHfAM3dzTR1N9HY1cjejr385OWf0DPUg8HgD/qHBa1EDIYFBQvo8nbR7+tnXv48/uvC/6KquIq5+XP5+x//Pb2+3lHXFWQVgAmVNI4UCajWWnVGFMkAdY8UERGRk16yYJasAUe86zze0B5n8fY++86fv5OwbX2fr48nX38y4ezWyBmryD3f+9P3cuvFt9LmaaOxqzG0xqy7iaN9R1N67w7jwF3ojrbDj/xaWLKQqpIqKooqGAoMsezuZfT7+sl2ZnP1WVdTmFPIhu0bEgY/i+VTqz7F3VvvTtj1cMP2DUk3za7bVTfhLQdETnUKbSIiIjJjJApmHq+HF1peoLWnlY6BDjxeT7Tk0OP10DHQATDqGMQPdMla7BtjyHHmxO2amJ+Vj8Um7PSYyIB/gM899bmEx53GSUVRRSiUza6mpqQm2uCjqriK8qJyspxZCa9P1rZ/rE2vjTFJux6OdX3slgMiMj4KbSIiIjIjJApmyQIJkPDYBZUXJAx0yYLIUGCIbGd2wmMHPQdHrQ9LRUFWAQP+gWHt8HNdufz0ip+y5sw1uBzj+2tbslnD1Q+u5taLb03YmTKyZUCyrofJOluO3HJARMbHMdUDEBERERmpdlMtK3+4ktpNoXb1W5q34L7DTVtPG3A8fD194OloIAkSmhmLBJLLHriMyx68LO6x1Q+u5ukDT8e955bmLSyZs4Q8V17csTmNM+FGzr6gj51Hd6Yd2GY5Z+G3/mHt9S2WAf8AH33sowz6B9O6X6xks4ZBG8RgcJj4fyWM3TJgadlSVsxfMarrYe3y2pSuF5Hx00ybiIiITDuxJYvJZoou/9/LE85ADQWHMMTfrDkQDHD5/14+rMQxcs93/+Td1Myuod/fH/9aG6C9v33YcwbDkjlLWFiykOqSaowx/OTln+ANeLFYDIb8rHyCBOn3jb5vkCAuE/99THRd2Fjli609rWy+dnPC7o+RWbVEDVwKcwpTul5Exk+hTURERE64ZHuYjSxZ3LhjY8KZokAwEHdtGYRKFRNJFMggFMr2HdsX95jTOLl86eW8veLtVJdU86NtP+KQ5xDL5i7jkQ8/Muzcb77nm9F1YOVF5ey+cXfCrQA+uPSDPLTzobivOdF1YamUL15QeUHSdWtjmej1IpKcQpuIiIicUMk2pobRa9BufurmhN0a/daP0ziHlRVmwkXVF/Gp8z7FvPx5XLXpKto8bVQUVYwKImuXrU14j4LsAs6vOJ/S9lKWli2lILsgGm4u3Hghzd3NVBZX8ty653h418M89tpjk7IuLNnG3LHli8nWraVioteLSGIKbSIiInLCJCt1vOyBy0btCRYkSDCYfHPo8QQ2p3HicrgSdoC85s3XcMWbrgBgbt5c2jxt4woi8UoKC7ILeOnjLw17LtVgNR4qXxSZ+dSIRERERCbFyGYikLwpRr+/n/6hxGWLyYxcu5bjzOG2S25j4z9spCCrAEf4rzwOHBRmh0JMog6QU9E8IxKsCrML4451osEqMsPnLnID4C5y03ZL26gNvRM1GxGRqaWZNhEREUkq2abVyfZNe775eQ56DnLQc5C7XryLw72H+cWeXyRsipEozEXEttGPtMO/54P3cOWbriRog9H1VCPLGK9805Vx11qlOvsUCTCTHWQSlU5maiYslfLFRM1GRGRqKbSJiIhIQsk2rT7ae5Q/NP6BI31HaOpu4panbuGg5yA7j+5kzxt7ogGrvb+dT//602O+lgn/L9KeP1Z+Vj63XXIbtz1/27DGHrHBY+T6sYh4a8sg9eYZJzLIxCudFBFRaBMREZlmks1spXJ8pGSdGpN5+sDTXPHwFQz4BwBo7Wml9LZSFs9ZzOHew3QNdkXP7Rrs4o6/3JH0fg7j4Iozr+BXe3/FUHB0Z8f8rPxRa9pir123ch1/bP5j3GAGycNVomNqnpFZJ2pWUuRUk1JoM8Y0Ah4gAPittauMMaVAHVANNAJXWWs7jTEGuBNYDfQD/2St3Z75oYuIiExv6YYrSD6zlcrxkZJ1ajx7/tk0djVy4+Ybqe+spyC7gLPmnUVjZyON3Y0cGzg26n5+6+e19tcSvp4DB06HM263x1muWVy2+DJuettNCcsSgaQli5Mx66WgkTkqrxSZHOnMtF1orY3dSfKLwDPW2m8YY74YfvwF4DJgSfjXW4EfhL+KiIicUmI3iE5FbMCCULv7SMC6oPKCpMdXzF9B3at17OvYx5I5S7hq2VV0ebt430/fN2xPskinxnf++J1xx/B6x+tjjtNpnBgMfusfdcxhHAnb80f2G1t/zvqka7dO9H5fChoiMt1NpDxyDfDu8Pcbgd8TCm1rgPuttRb4izGmxBizwFp7aCIDFRERmUzjmRXLpGSt8Fc/uJq9n9qb8PjFGy8GE2rk4Q/6MRg+/quPpz2GHGcO17z5GpaULuGFlhd4Yt8Tcc9L1mLfb/1kObLiBrfY/caSrd1KtAZNRORUlWpos8DTxhgL/Mhaew8wPxLErLWHjDHzwue6gZaYa1vDzw0LbcaYG4AbACorK8f/DkRERDIg0ayYx+sZNoNVu7w2pfVgHq+HjoEOgJTKGJO1wh8KDPGBhz4QXVs26viI9WGRBiBjed+i97GleQv9vn4sFl/Ax6bdm9h87Wbm5s/lucbn4nZ6zHZmYzBx9zjLc+URJBg3tKXTSl+zX5NDpaAiM1Oqoe0d1tq2cDD7rTEmcTE7IzZKCRn1p0c4+N0DsGrVqtT+dBERkZPWVM90xRMpR+z39ROwAZzGyc1P3RwtVxzrukRljgBX/ewqXj36KqcVnsbH3vIxfrrjpwlb4XsDXrYfTn95uMvhwoEjbtOPPFcef2j6A4P+wehzI2f2Em32nO3IBkPc0OZ0OPnFVb9g7SNrtZHzNDRd/r8lIulJKbRZa9vCX48aY34BnAcciZQ9GmMWAEfDp7cCFTGXlwNtGRyziIichNJd/5VJ8WbFgFHliAEbiIaatlva4gaQZGWOF228iHdUvoOmriYauhoA2N2+m2cbnh1zjLNzZ9Pj7UlamjiSP+hPuIF0wAZwOeL/NSBog2zetznpPmaQuGFIqq30RUQkNY6xTjDG5BtjCiPfA+8FdgGPA+vCp60DHgt//zhwnQl5G9Ct9WwiIqcWj9fDhu0b+MJvv8CG7RuiISjZ+fFCU+2mWlb+cCW1m1IrqRvPa29p3oL7DjdtPaF/X4zMin3tT19LWK4YtEEe+ttD7DqyizPvOpOi/y7CfYebf3z0Hzn33nPjtqwH8AV9/L7x99HAFmtx6WJcJn6IKswu5NVPvkpeVl7S9zJSflY+n3vr5yjMLsQR/iPfgYPC7EI+tOxDCWf2Ig1DIuHLXeQGwF3kpu2WNi6ovCDpMTi+Lm3F/BWcX3G+ApuIyASYUL+QJCcYUwP8IvzQBTxkrf26MWYO8AhQCTQDH7bWHgu3/L8LuJRQy/+PWmu3JnuNVatW2a1bk54iIiIzRLySwrysvIQlhSNb0kdmbDZdtYn1j6+ntac1tJHyJ3ePuZYscq+gDdLn6yM/Kx+HcSR8bY/Xg/sO97BZsQiXccXtjhhhMCmvHYvlMI64YbAwu5BNV21KWFY4sntk5HheVmgNWb+vP+49224JhdGRnRof3vUwN/3mprjBLT8rnzsvvZP156wHYOUPV7LjyA5WzF/BK594Zdi507GsVURkJjLGbLPWrop7bKzQdiIotImITC/j/Yt4shAUCRCxMy7JzofQrFC88DLe117/+Pro+7r/H+7nWy98i6/98WsMBgbj3HF8nMaJ0zjjriNL1sAjEpRqz6qNlhVWFFWMKivsHeoddfyVw6+kFVYhvf9WCmYiIpMvWWibSMt/ERE5SY13fVmyDohBG6RuVx1P1z8dDQDvqXlPwvMhtBYs8nWstWTJXtvr93Jl3ZVsad7CgH+AnUd38sirj6T9/iJcxoXT4YwbvnKcOaFxxxmKtTZumIPjJYkF2QWUziqlpaeF0lmlo95rvHb4kVLFul117D+2n8Wli6k9qzZpSWJhTmHSNWux1yqoiYhMLYU2ERHJmH0d+5Kuk3r16Ku80PICrT2tdAx0cHrB6QnPjycS/Nafs57eoV6auppo7GqksauR+3fcn/BeQ8Ehflv/22H3GUuuM5d1K9bx4M4HGfAPDCv1/ODSD/LQzofiXtfv7+faN1/L43sfHxWG/uXcf+H7L34/YUliZA+zsdqyxwtRBdkF0XLGVKlhiIjIzKDySBERGSXZGqZ4IuVzWc4s9ryxJ24oyXXmgoEh/1A0yGQ5szDGDGs7P5b5+fPxBX0cGziW1nuKJ8+VhzEm7nhj14ONnMFKZT1YvDJHa21a5aMnSrr/vUVEJPOSlUeO2T1SREROLYk6OUL8bo4er4cXWl5gx5EdHPIcItSParTBwCCD/sFhJY/egDetwAZwpO/IqMBmMLzltLeQ5chK617GGD517qfidleMlAhGZrD++5L/Zv056ynILqB2eS0OE/+P0MgG0pEyRyBa5hgpSUz2eiIiIiMptImITKHxtLSfyL08Xg/n3nsu8785n3PvPXdUO/xE7e+3NG8Bjq9129u+N+75hzyHCAQDzHLNGhZKIs05JsrlcMXdd8xi2X9sP09c8wSF2YXR13IaZ9Ig1+frwxiTtHV9PBMJX2O1yp8KS8uWsmL+ioTlmCIiMrW0pk1EZAplakPpyGxXZK2Yx+sZ1R5/ZLv49v523He42XztZr734vfY88Ye9rTvwR883uZ+ZAOQCGst+zv2876fvo9+f/+w8wf8A8NeN0iQsTrjz5k1hz5fH4P+QYpzirnxvBv57p+/y4B/AIsdtSZsKDC6mUfQBmnpbhnVkGPAP8AXf/fFpOvIxmr+Ec9E1oON5/UmkxqNiIhMb5ppExGZ4caaHYNQqFv94Go8Q55RHRkve+Aynm9+np1HdxIIBuK+xqB/kEsfuJTX2l8D4G9H/8aSu5YMC2yJFGYXUl5YnnCmLc+Vx22X3MbSOaFZnuqSar5+0dc58v8dobyoHDg+G4VlzA2hR5YzrluxbsxSRhjfbNNYG0gnu6dmt0REJFWaaRMRmaFqN9UmnR277IHL+O9L/puW7hYO9x4mYOMHsl5fL72+XoCEm0X7gj6eb3k+rfF9dMVH+fb7vk1Jbgm9Q70JG3A4HU5qz6rl6fqngeMdE+PNRi2Zs4T8rPwxuy/GSrW1/Xhnm5JdN95jIiIisRTaRESSmMxNheM1/BhZ0pjM3va97Dy6M7qmaqReXy+f+fVnEgaxyZSflc87Kt/B7FmzgdSCU7zf35Gt72uX13LzUzfHfc3YWbOR1NpeRERmMoU2EZEEkq0T83g9XHT/RTR3NVNZUsljtY+xpm5N9PGz1z2bNIDFri+D4yWNm6/dPKwhhcfroe7VOvZ17GPJnCXULq+lMKeQjv4OjvQdAY5vQB3PVAQ2iB+gxhOcRga52PDX7+sftnfaWA1A4m1KLSIiMhNonzYRkThGNu0wGLKd2Xxo2YcoLyzn7q130z/UHz1msdGvsTNI8ToCeryeMffqcjlcPLrnUT72+McI2ABDgaHomrDC7EK6vF2T/nsA4c6PDie+oC/hOdnObALBwKgAlagbYiZmL3uHekftnaYQJiIiM1myfdoU2kQkrsksCzzREs1WJTs/UahKx8jNkmM3oN59dHfcJh4OHGS7stPeu2w8IiEz2bGKogo+/47PJ+2+eNslt5HrylWAEhERmYBkoU3lkSISV6Za0U+mkSWK8UoSIzNmsaV0Nz91c9KZoLpX6wjaxCWHqeod6mXeN+cxN38uqxev5ld7f8WAfyBpWAoSnHBgS3b/PFce8/Ln0e/vp7ywnNePvU7vUO/oexiDtZbSWaWsW7GOLz/z5bj3cxgH61auU0gTERGZRGr5LyIzUqTN/fa27RztP8q2tm2U3FbC7NtmRzeNjm1zH+mcGLCB6L5jI8OKx9jeVyAAACAASURBVOthw/YN3Lvt3oRt5dNhsQz4B2jubuaH234Y3b9sMteZ5WXl8ZbT3pIwRDkdTnZ+cidH/vUI2/55G7++9tdxN4iumV0TvWYiG0mLiIjIxGmmTUSmpdhZNHeRG4CDPQejTT8iYSzCYrHW0jXYxfa27bjvcPMv5/5LwhmzoA1St6uOq5ZfRd2rdTzX8ByPvvYovoAvYWv86S4/K587L72T9eesj84wBm2QPl8f+Vn5OIxjVMhK1Bxk/ePryc/Kj3ZtVPdFERGRqaM1bSIyisfrYdndy2jtaaW8qJzdn9ydViv6iRrZBCSWAwdZziyMMRMuIyzKLmIoOITX752yLovJShnTNXINXTrNOlJdw7jyhyvZcWQHK+av4JVPvJKRcYuIiIjWtIlIGlJtRZ8psU1CKoorGPQP8m/P/BtDwaG45wcJ4g14M/LaPUM9GbmPwzgI2mD0azoKsgvwDHmYnTubocBQ0rLMSGfGb1zyjVBjkCQbRUfuvf6c9SmNY6Y3mxERETmZKbSJnETG2/Exct2i2Yv4bf1vh5UdBglG14DFzuJkwsgmIdOZweAwjmgzk1muWdx43o0YDItLF/P+M97Pk68/yc93/5xnG59NKVjmZ+VTkluCZ8hDeVE5jV2Ncc/LceZw43k3sqxsWXS27LoV101JqeLIza5FRERk8o0Z2owxFcD9wGlAELjHWnunMearwMeBN8Knftlauzl8zZeA9UAA+Iy19qlJGLuIjDCejo+xG0g3dzcnDE+BYICz7j6LAd9Awk6NydahRc6PzKztOrqLH2370QlpbT8RDhwECXLWvLP47Fs/m7TUcP0567lq+VW473CnFNocxkFJbgktPS3R9WaxZaHJ9nsryC6gdFYpLT0tlM4qPWFryzQjJyIicuKlMtPmB26x1m43xhQC24wxvw0f+4619luxJxtjlgEfAZYDpwO/M8acYe00/2d0mXZOpn3CpquRpZCdg50Jz+3399PU3QTAG/1vMPebc/nQsg9xfvn5APy55c88+tqj0fVhR/uPRq+NnP/OqneypXkLQ/6hUWvVpiuHw8HysuW8ae6bUio1jHRajJ1BjJRNxtt8+3svfg+HcbC0bGm02ceFGy+kubuZyuJKnlv3nJp9iIiInOLGDG3W2kPAofD3HmPMHsCd5JI1wMPWWi/QYIzZD5wH/DkD45VTyEzYJ2yypbIPWbr3i8xyHe09yqY9m/AFfWnfx2LxBrw8tPMhHtr5UMrn/67+d+MZdsoy2dQjsn5sPGv5IuErtgnI+894P+fde96ocsZ4M2gvffyllF5HpYoiIiKnhrTWtBljqoG3AH8F3gF8yhhzHbCV0GxcJ6FA95eYy1qJE/KMMTcANwBUVlaOY+giJ7eRHRTb+9uTNgTxeD10DHQA0DHQgcfrGRbwovfz9WVk4+jpJs+VR44rh87BznGHN4Mh25nN2mVrubD6wqTdFscSrwnI+RXnU9peytKypRmZPdMMtIiIyKkh5c21jTEFwM+Bm6y1PcAPgEXASkIzcd+OnBrn8lF/e7LW3mOtXWWtXTV37ty0By5yMovdFDpSRhhpCHLRxou468W78HiPNwuJbDTd1tMGhNaRzbl9Dl/43RfweD209bTx9z/5+9D9ZmBgWzF/Be7CZBP8ofLN+fnzWTF/BRdWXxh3I+jvr/4+hdmFOI0TCM2mFWQVcOacM5mXP4+/O/3vaP98Ow9c+QDrz1mf8bLEurV1vPKJVxS2REREJC0pzbQZY7IIBbYHrbWPAlhrj8Qcvxd4IvywFaiIubwcaMvIaEVOcpFyyNfaX4uuMxvJF/Tx2d98ls899TmqS6o5e97Zozo+Wiy+oI/bn7+d7/31ewRsYMaFtdjNoC+ovIDaTbW80PIChzyH4jZLyc/K51/f/q/R2a3eod64a8OuW3FdynuXiYiIiEwHqXSPNMB9wB5r7R0xzy8Ir3cDuALYFf7+ceAhY8wdhBqRLAFezOioRU4CI9erff3Cr7P2Z2vpHeods7QvaIMEbZD9x/bT0NmQtF3+gH8g00PPOIdxMGfWHCqKKrj+nOtp7W4dFajq1tbh8Xpw3+EeFlBj71F7Vm30caK1YensXSYiIiIyHaQy0/YO4B+BncaYV8LPfRm42hizklDpYyPwzwDW2leNMY8Auwl1nrxRnSPlVDFyo2iAlu4WlsxZQu3y2mjL+3PvPZfXO16PhrM3+t/gfQ++b1yvOd33N0smWUv7eGI7MwZtkD5f37AZOc2YiYiIyMkole6RW4i/Tm1zkmu+Dnx9AuMSmXGSbRSdn5XP537zOS4/83J+tutn+Ozwjo2Z6nh4IuVl5dHv6x/2nNM4sViKc4pZWLKQ68+5nj+3/Jmf7/k5voAvujF15JxFpYvSbmkfrzOjShxFRETkZGasnfq/LK5atcpu3bp1qoch04jH62HZ3cto7WmlvKic3Z/cPWar+9hZrtiZrfG89sYdG3ni9dAyzUtqLuGhnQ/R0tNCflY+V5x5BYtKFwGw/9h+Ogc6yXZm85MdP8Ef9Kf/ZqeYw4SadVhrE4bHHGcO3oA3Oqu16apNtHS3sP/YfsqLysEQt6QxoneoVyFLREREJAljzDZr7aq4xxTaZLoZ2eo+lRK6yDUjS+Y2XbWJf3v236Lrxh6rfYzN+zez6+guOgc6KZ1VyvJ5y6MBb0vzFt770/fOiHVgE+UyLj5x7idYOX8lq5esZuldS+OuFSvIKuAb7/lG0lAmIiIiIhOj0CYzRrJGE4XZhbTd0kZBdsGotWNfeuZL9A71xr2nAwdBgtG9uxzGMayTosM4yHJkcdniy3hs72PTulQxE5tHO4yD/KzRIXhkeedENpcWERERkfQkC21pba4tMtnqXq1L2Jp+wDfAjU/eyLVvvpa1P1sbDRdjBZnIPmeRc0beP2iDeANefrn3lxl6F5mV68oly5HFpqs28aXffYmXD788ZnBL9HuS5cjiO5d+h3Ur1o2aLdNaMREREZHpSTNtMi1EZs7u3XYvL7Yl3iHCZVz47cxbNzYeBsON593Iyvkrh4WnyIzYyK0BImvTgjbI4tLFHOk9klaJqYiIiIhMHc20yQlTu6mWve17WVq2lLq1dQnPi4S0XUd38Vr7azzb8GxKG0CfrIHNYRxgSSlgRWbELtx4IU1dTeRn5XPlsitZVraMzfs3c+DYAZaWLeW+y++Lu7m0iIiIiMwsmmmTlMUGsg0f3BANXZ0DnRRkF9A50Mmv9v2K3qFeSnJL+KcV/xRt8gFE16AB3L31bvwBP4OBwal8S5Mqy2ThcDjwBrwJzzlzzpm8dENoA2gFLBEREZFTlxqRyLh5vB4uuv8imrua8Qx5GPAPMC9vHl2DXfiCvjHXVuW6cjEYjDEYDH2+vhM08qkTWYO2+drQVoaxnTAjImFNwUxEREREQOWRMk4jW+9HHO0/mvI9Bv0zfybt1otuZdOe0L5kkVLEmpIaMFB/rJ5H9zxKr6+XgqyCaJli7Bq0SCmjZtFEREREZDw003aKip1Bcxe5uebN1/C7+t8Boc2kH/jbA/ztyN+mdfv7iXIZF5efeTmb920Gy7BSTYOhILtAjTtERERE5ITQTNspJrbJR+wG0qsXr2bz/s081/Acj772KF6/F4vlaP9RXj78cvT6pw48NYWjHz+ncZLtzMYYg9fvje41Fu+52P3Heod6qdtVx+43dvPjV35M52An5UXl7L5xt2bERERERGTKKbTNYLGzZZUllTxW+xgX3X8Rr3e8PmqGzGmcocCCkwCBKRpx5uW58gjYAGuXreXC6gupPSvc9GTEXmPxnosEsoLsAtafsx6AVk9rtNmKApuIiIiITAcqj5yhRq43G2uD6Zkg8h4cxsEs1yz6fH04cGCMIceVM2ymzOVwDQtqClgiIiIiMpOpPHIaG6uUMbalfu9Qb7S1/qY9m/AFfdH7zOTA5jIuPnHuJ6KbSEN6M2UiIiIiIiczzbSdYB6vh407NvLE60/Q3t/OjiM7sEE7rGTRgWNYt8bpzGVcXLLoEhaXLuZY/zGeaXiGPl8fJbklHOk9MixYjjRybZmIiIiIyKlKM20nUOzM2dHeoxzqPcTh3sMEbZAB3wCtntYxZ8Wma2CLlC8aDC6Hi5vPv5mvvOsrCWe8IiWcQRukz9dHflY+BsON592IwWjGTEREREQkBQpt4xSZMfvla7/kYM/BaDBp87TN6FLFRH5Z+0u+9qevpbXX2AWVF9B2S5vKGkVEREREJkDlkSnweD38aNuPuG/7fRztO0qQIF2DXVM9rIxx4GD5vOUc6DzAgG8Al8OFL+jDZVw4HU4ev/px3rvovVM9TBERERGRk9aUlEcaYy4F7gScwAZr7Tcm67Um05bmLVx8/8UMBYameijjsur0Vew4vAOHceANeIcdy8/Kx2Ecw9aURfYs08yYiIiIiMj0MCmhzRjjBL4PvAdoBV4yxjxurd09Ga+XKdZauga7aOxqpKm7idfaX+Pfn/13/NY/1UNLy8gZstggVl5UDgZau1vjhrLYPctERERERGTqTdZM23nAfmttPYAx5mFgDTCloc1aS8dARyiUdTXR2NUYDWiRrz3enqkc4rjkunJxGVfCBh8KYiIiIiIiM9dkhTY30BLzuBV4a+wJxpgbgBsAKisrM/Ki1lqO9B2JBrLYMBYJaP2+/pTulZ+VzyzXLNoH2jMytolwGidrlq6hMKeQwpxCPF5P9OucvDksK1umMkYRERERkZPUZIU2E+e5YR1PrLX3APdAqBFJKjcN2iCHew8fnyGLzJZ1h75v6m5i0D+Y0gALswtZOHshVcVVVJdUU11SHf2+qqSKObPmcN/L9/GJJz5BwAbGvuE4GAyX1FzC0rKl0QBWU1KDN+Dld/W/AwvvX/p+1q1Yp0AmIiIiInKKmqzQ1gpUxDwuB9rGuigQDNDmaRs1Oxb5vrm7OeWGILNzZ1NVUhUNY1XFVcNCWkluCcbEy5bH1S6v5aZf30Sfvy+l14znjNIzcDqcRLp0OowDd5GbNWeuSRrGbj7/5nG/poiIiIiInDwmpeW/McYFvA5cDBwEXgKusda+Gu/8ooVFtuymMlp6WvAHU2v6UZZXFneWLDJTVpRTlJH3kk73yPLCcgI2QI4zh7dXvp1vv/fbnFZwWkbGISIiIiIiJ69kLf8nbZ82Y8xq4LuEWv7/j7X26wnPPd1Y/nn4c/Py542aJYs+Lqk6oeWCvUO93LP1Hja8vIGjvUdxOV0U5xTjcrhSmjUTERERERFJZkpCWzoWnLHAfvWhr0bLGSuLK8nLypvqYYmIiIiIiJwQU7K5djrcRW7+edU/j32iiIiIiIjIKcYx1QMQERERERGRxBTaREREREREpjGFNhERERERkWlsWjQiMca8ATRN9TjGUAa0T/UgRML0eZTpRJ9HmU70eZTpRJ9HSUeVtXZuvAPTIrTNBMaYrYm6uYicaPo8ynSiz6NMJ/o8ynSiz6NkisojRUREREREpjGFNhERERERkWlMoS1190z1AERi6PMo04k+jzKd6PMo04k+j5IRWtMmIiIiIiIyjWmmTUREREREZBpTaBuDMeZSY8xeY8x+Y8wXp3o8cvIzxlQYY54zxuwxxrxqjPls+PlSY8xvjTH7wl9nh583xpj/G/6M/s0Yc87UvgM5GRljnMaYl40xT4QfLzTG/DX8eawzxmSHn88JP94fPl49leOWk5MxpsQYs8kY81r4Z+X5+hkpU8UY87nwn9e7jDH/a4zJ1c9IyTSFtiSMMU7g+8BlwDLgamPMsqkdlZwC/MAt1to3AW8Dbgx/7r4IPGOtXQI8E34Moc/nkvCvG4AfnPghyyngs8CemMe3Ad8Jfx47gfXh59cDndbaxcB3wueJZNqdwG+stWcCKwh9NvUzUk44Y4wb+Aywylp7FuAEPoJ+RkqGKbQldx6w31pbb60dAh4G1kzxmOQkZ609ZK3dHv7eQ+gvI25Cn72N4dM2Av8Q/n4NcL8N+QtQYoxZcIKHLScxY0w58H5gQ/ixAS4CNoVPGfl5jHxONwEXh88XyQhjTBHwLuA+AGvtkLW2C/2MlKnjAmYZY1xAHnAI/YyUDFNoS84NtMQ8bg0/J3JChMsm3gL8FZhvrT0EoWAHzAufps+pTLbvAp8HguHHc4Aua60//Dj2Mxf9PIaPd4fPF8mUGuAN4Mfhkt0Nxph89DNSpoC19iDwLaCZUFjrBrahn5GSYQptycX7lw+125QTwhhTAPwcuMla25Ps1DjP6XMqGWGM+QBw1Fq7LfbpOKfaFI6JZIILOAf4gbX2LUAfx0sh49FnUiZNeO3kGmAhcDqQT6gkdyT9jJQJUWhLrhWoiHlcDrRN0VjkFGKMySIU2B601j4afvpIpKQn/PVo+Hl9TmUyvQO43BjTSKhE/CJCM28l4VIgGP6Zi34ew8eLgWMncsBy0msFWq21fw0/3kQoxOlnpEyFS4AGa+0b1lof8CjwdvQzUjJMoS25l4Al4Q5A2YQWlj4+xWOSk1y4tv0+YI+19o6YQ48D68LfrwMei3n+unCHtLcB3ZESIZGJstZ+yVpbbq2tJvQz8Flr7bXAc8Da8GkjP4+Rz+na8Pn6V2TJGGvtYaDFGLM0/NTFwG70M1KmRjPwNmNMXvjP78jnUT8jJaO0ufYYjDGrCf2rshP4H2vt16d4SHKSM8ZcAPwJ2MnxNURfJrSu7RGgktAfEh+21h4L/yFxF3Ap0A981Fq79YQPXE56xph3A/9qrf2AMaaG0MxbKfAy8H+stV5jTC7wU0JrMY8BH7HW1k/VmOXkZIxZSagxTjZQD3yU0D9E62eknHDGmP8Eagl1f34Z+BihtWv6GSkZo9AmIiIiIiIyjak8UkREREREZBpTaBMREREREZnGFNpERERERESmMYU2ERERERGRaUyhTUREREREZBpTaBMREUmDMeYmY0zeVI9DREROHWr5LyIikgZjTCOwylrbPtVjERGRU8O0CG1lZWW2urp6qochIiIiIiIyJbZt29ZurZ0b75jrRA8mnurqarZu3TrVwxAREREREZkSxpimRMe0pk1ERERERGQaU2gTERERERGZxhTaREREREREprFpsaZNREREREQk0zxeD3Wv1rGvYx9L5iyhdnkthTmFUzqm3qFeDhw7QENnA/Vd9TR1NdHUnXA5G6DQJiIiIiIiJ6EtzVtY/eBq+n39BGwAp3Fy81M3s/nazVxQecGkvW73YDcHjh2gvrOehu4GGjsbae5ppqW7hdaeVjoGOtK+p0KbiIiIiIhMudpNtext38vSsqXUra2b0L08Xg+rH1yNZ8gTfS5gA3iGQs+33dJGQXbBuO59bODY8VDW1UBjVyPN3c209IRCWddgV0r3yXJkcVrBaVQUVVBeXM4jPJLw3JRCmzHmc8DHAAvsBD4KLAAeBkqB7cA/WmuHjDE5wP3A3wEdQK21tjGlkYuIiIiIyClpb/tedhzZkZF71b1aR9AG4x4L2iB1u+pYf876UcestbT3t7P/2H4auhpo6GygqTtUvtjc3czBnoPDgmAyWY4s3EXuUCgrKqeqpIqFxQupLqmmprSGyqJKsl3Z0fMnFNqMMW7gM8Aya+2AMeYR4CPAauA71tqHjTE/BNYDPwh/7bTWLjbGfAS4DahN6Z2JiIiIiMi4ZXK26kTyeD3RssGOgQ48Xs+E1p7t69hHn68v7rE+Xx/PNTxHljOLxq7GYTNlB3sOJrxupFxXLu5CN+VF5VQWV1JVXEV1STXVJdUsmr2I8uJyXI7MFDamehcXMMsY4wPygEPARcA14eMbga8SCm1rwt8DbALuMsYYa63NyIhFRERERCSuTM5WnSiRtWd9Q6Gw1NbThvsO97jWngVtkIM9BwnYANmObIaCQ3HPe3DXgzy468Gk98rPyqe8qDw0S1ZcRWVxJQtLQjNli0oXsaBgAQ7HiWnGP2Zos9YeNMZ8C2gGBoCngW1Al7XWHz6tFXCHv3cDLeFr/caYbmAO0J7hsYuIiIiIyAwWb+1ZkGDCtWf+oJ/Wnlb2H9tPY2cjDV3Hyxdbulto87ThC/pSeu3inOLQTFlxKJRFZsoWzl7I4tmLKcsrO2GhbCyplEfOJjR7thDoAn4GXBbn1MhMmklyLPa+NwA3AFRWVqY4XBERERERiSfTJYYnQrK1Z16/lw8/8mGKc4uj5YuHPIcI2EBK9y7MLqTP1zfs/rnOXP7vZf+XK990JXPy5mTkPZwIqZRHXgI0WGvfADDGPAq8HSgxxrjCs23lQFv4/FagAmg1xriAYuDYyJtaa+8B7gFYtWqVSidFRERERMYpkyWGiWRivZzX76WxqzE0U9bVyMYdGxOuIRsKDvGbA79JeK+yvDLche6468kWzl5IcW4xvUO9LPv+Mlp6WqgoqmD3jbvH3TVyKqUS2pqBtxlj8giVR14MbAWeA9YS6iC5DngsfP7j4cd/Dh9/VuvZRERERGQ6makNO+JJt8RwvFJZL9fv66exqzG0eXRXQ6h8saspOlN2tO9oWq9ZVVzFOQvOobK4MtR1saSGhaULqSmpIT87f8zrC7ILOL/ifErbS1latnRGBjZIbU3bX40xmwi19fcDLxOaIXsSeNgY87Xwc/eFL7kP+KkxZj+hGbaPTMbARURERETGw+P18ELLC9GNjmdCGWEy421vP14D/gF+ueeX1HfV09jVGA1lrZ5W2vtTa2PhMA7m58/n9MLT2XF4B/5oq4zjCrML2fXJXRMOWjM9lEOK3SOttf8B/MeIp+uB8+KcOwh8eOJDExERERHJrBNRRniijdXefv+x/Snfq2uwi/rOeg50HqChM7RxdFN3E01dTex+YzcAr3e8zhWPXJH0PlmOLBYULqCiqIKKogqqS6qpKqmipqSGmtIaqoqryHJmAcP/mwQJ4sBBfnY+m6/dPGNnxjItMxsHiIiIiIhMcyeqjPBEWzJnCflZ+XGDW35WPotLFwOhjaM7Bjpo6GzgQOcBGrsah5Uvtva0prxxNMCCggUsKV1CZcnxNWU1s2tCe5QVleN0OFO6zwWVF9B2S1t07Zm7yD1j155NFoU2ERERETklTLSMcCLr4CZzDV3t8lpufurmuMeGAkM8sPMBbn/hdlp7Wun39ad0z1xXLuVF5ZxeeDovtLyAPzi6fLF3qJcnr30yI+GqILuA0lmltPS0UDqrVIFtBIU2ERERETklTKSMcKLr4DKx6XXQBjncezhUvnjs+ExZc3czRTlFcWfJfEEfv2/8/ajnIxtHVxZXhpp8FIf2J4vsUTY3fy7GGDZs38C2tm1xQ1um18stLVs67Kscp9AmIiIiIqeEVMsIRzpR6+ACwQAHPQdp6GygvrOe+s56mrqbaOxqpKUntHH0UGAopXsZDGeWnRnaNLoktHH0wtkLo+WLpbNKMSbe9srDZXK93FhOhoYhk0WhTUREREROCcnKCB3GQe1ZtaOez/Q6OG/AyzP1z1DfWR9dT9bUHVpTdqj3UNwZrXhKZ5VSXhgzU1YSmin78jNfZt+xfZw9/2xe+cQrKY8rkfEGXckshTYREREROWkkWztWmFPI5ms3p9WpMJ11cF6/l+bu5ugsWUNXA03dTdR31rPz6E4AXmt/jUt+esmY72Nu3txo+WJVSRXVxaEmHwtLFlJTWpMwKP5s98/Iy8rLWInheIKuZJ5Cm4iIiIicNMZaO5Zup8KxygNvf+F2frD1B7T2tHKk70jK4yydVcqi2Yuis2TVJdUsLFnIotmLqJ5dTa4rN+V7xcp0ieF4gq5knkKbiIiIyAw2mV0Jp+PrZkKiToUeryc6M9bQ2UBDVwN/aPwDDhwEiT/b9nrH66OecxgHpxeezoKCBbx86OW4G0f7Aj6eXffsjAg9ask/9RTaRERERGawdLsSphO2kp2biW6ImebxeugY6ACI2+Gxa7Ar2nHxjb43AGjoamDFD1fQ3N1M12BXWq/nMA4+9KYPsaR0SajzYslCFpUuwl3oJsuZxYbtG7jpNzfh901+58XJppb8U0uhTUREROQUkk7YymQwm+w9zv7U9KdQCV+4lLG1p5U5t8/h7xb8HT1DPbT2tNLj7Rl1XY+3h78d+duw57Kd2bgL3dEmHw7j4OFdD+MNeIFQZ8aC7IIxu0eeyM6LcnJTaBMRERGRUcaatUrXRAJg5FpfwMeLB1+MNvpo7GqksauRpq4mXj82ukzRF/Txl4N/GfV8riuXiqIKKoorqC6upqqkKtTgY3YNNbNrmF8wH4dxDLvmrtV3ceHGC2nubqayuJLn1j035myTOi9Kpii0iYiIiMxQmQ5WEWPtSzYZrxvZODpSvlh/rJ7G7lAg29O+B4Dd7bt564a3pnVfl3HxwTM+yIeXfzja5GNu3tyU9iiLVZBdwEsffymta062zova/HrqKLSJiIiIzEDj2fA5lbA11r5km67axNpH1qa90bS1FoDeoV427tgYbfTR3NVMU3cTBz0HU944ujin+PhMWUk1u4/u5g/Nf4h7rt/6WVq2lKvffHVK986kk63z4kxrOHMyMZH/A02lVatW2a1bt071MEREREQmRaY7LXq8Htx3uIcFq4jC7MK4Gz7HhryR4SE2bEWaZ8Qr6ctz5REkyKB/cNSxgqwCXlj/Au397dR31Q+bKWvqbqK1pzXl91c6q5TKokrys/N58eCL+II+ILSWLD8rn1//n1+nPOb8rHzuvPTOKW340TvUG+28WFFUoc6LEpcxZpu1dlW8Y5ppExEREZlkE23oMTL0pbPhM4w9exYb8pI1z+j39+M0zrjHen29nP3Ds1N6PwbDkjlLWD53eXR/soWzQ2vKqoqryM/OjwbTSGADsFh6fb2jxjzdyxDVeVEmKqXQZowpATYAZwEWuB7YC9QB1UAjcJW1ttOECoTvBFYD/cA/WWu3Z3zkIiIiIjNAJtZ/jQx96XYlTCXkXf3mGNT40wAAIABJREFUq2nqasIb8JLlyBoWlmIFbCDpWA2G0wpOo7yoHHehmyf3PTnqXhbLIc8htt2wLWGASSeYnmxliCIjOcY+BQiFsN9Ya88EVgB7gC8Cz1hrlwDPhB8DXAYsCf+6AfhBRkcsIiIiMkNsad6C+w43bT1twPH1X1uat0TP8Xg9nHvvucz/5nzOvfdcPN7RJY8jRboSxhOvK+FYIe/Tv/40+bfms+zuZdz51zsTBrZkchw53HrRrQx+ZZC2W9p48eMv8v4z3k+2Mzvu+ZHglUi6wTSyAbS7yA2Au8hN2y1tSdfZicwUY4Y2Y0wR8C7gPgBr7ZC1tgtYA2wMn7YR+Ifw92uA+23IX4ASY8yCjI9cREREZBqp3VTLyh+upHZTqBQvtiQxSGjGKLYksXeoNxrqtrdt52j/Uba3bY8b6kbO1NUurx3Vkj4iYAN0DHTw2V9/ljUPr2HlD1dy10t3JR37gH8g+r3L4eK0/NNGlUHOcs1i4z9spCAr/qxVtiubT7/108NC2kT2KUs3mMLxMkRAZYhyUkmlPLIGeAP4sTFmBbAN+Cww31p7CMBae8gYMy98vhtoibm+NfzcodibGmNuIDQTR2Vl5UTeg4iIiMikS9ZMxOP18ELLC7T2tEaD1VjlfRtf2ciXnvlS0nVmrxx+JRrwAA72HGTet+Zx/crruaTmEh7f+/iocsVB/yBf+N0XUn5f2Y5svrf6eyybu4yq4ipOLzwdp8OZsHlGzeyalMsQJ7JP2XRfp5YutcuXiUgltLmAc4BPW2v/aoy5k+OlkPHE2/RiVItKa+09wD0Q6h6ZwjhEREREpkyiZiKJWu9/cOkHk84yPfn6kwlD3aB/kEt/eil/OfiXYaHMYhn0D3L31rsTjnOWaxYVxRVUFlVSVVJFzewaFpYspMfbwy1P30K/rx+LTdg9MiJR84xIGWIk0LmL3Am7IU4keI13ndp0DUdqly8TkUpoawVarbV/DT/eRCi0HTHGLAjPsi0AjsacXxFzfTnQlqkBi4iIiEwXyboy/nz3zxPOMmU5sth5dGfCUOcL+ni+9fmEr2swLJ+7nAsqL+CxvY9xqPcQS0qX8Pz1z1OWV5Zw4+hrz742pbA1llS7IU60QUg6ATFC4UhORmOuabPWHgZajDGRf664GNgNPA6sCz+3Dngs/P3jwHUm5G1Ad6SMUkRERGSqjafxR7x1ZZC8w6E/6Kff1x/3mC/oo9WTeN8yg2Fu3tyExy2WD5zxAX7wgR8wLz+0QiUvK4+5+XMTBjZIf83X0rKlrJi/YkKzVhNtEKJ1aiKp79P2aeBBY0w2UA98lFDge8QYsx5oBj4cPnczoXb/+wm1/P9oRkcsIiIiMk4jN5hu72/H/f/Yu/P4uqs6/+Ovc5ek2dN0T5o0pS3FwkxrbZlBUGQRoaiMPyphhhmrU8Wl4FLmoeDPcZzf6IgomwIyCI4FGSmWYZMCgi0zFhRpoZWydKFtlqZt2qZJbnKz3OX8/rgL9ybfu2Rf+n764JHce7/3e783XEM//Zzz/txSEV8i6Ov2cf7951PXUkdVaRWbPrWJHUd2cMkvL4l3xQ62HWTKTVM4b+55vNn0ZspuWbpo/Pll8zm17FSe2/ecY1JjYU4h3z7321z//PUD2g82VIaqa6U5ZSKDk1XRZq3dDjhN577A4VgLrBnkdYmIiIgMmFPxBaRcynjJLy/hx5f8mDUb19AV7MJiafI3UXpjKTb6vxiLJRAO8Nt3fpv2GtzGzQVzL+Ci+Rdx4+9v5FjnMSqKKnj7mrfjRUvvIjJx6eDiGYv55u++6Xju8RjEISIDl22nTURERGRA0qUuDgenblr5zeXUnFFDT6jH8TntgXb+8Yl/7HN/LKrfidu4Oa/6PH5f93u6Q919Hs/35vNIzSMU5hTywI4HONZ5jKn5U5O6TLGlg+etO4+61jqqSqrYvGpz/Jhs9oP1N3hjrAZ1pDMer1lkKKloExERkWGVKnXRqRtWlFs0oNfwdfuoba3lzaNv8qlHP5VURIUJ0x5o577X7hvwe3ASsiGWlS/jXz70L4MqrApzCnnlc684vkY2QRz9LYSHqnAeyWJR4SJyslPRJiIiIiMu096y3lq6WjjQcoDalloOtBxgX8u++Pd1rXWc6DoxqOtx4yZE6j1oTmL7yoajsEo0VveDjVaxKHIyUtEmIiIiIypdTP6HH/gw3z732/z8tZ/T1N6EMYawDScdm47buNMGgHhdXsfgD4/bwyTXpJShIk4S95UV5hRyVuVZlB0rY+HUhUNeWGl5oMjJTUWbiIjIKBnK5YFjla/bxzH/MQAOtR9i3fZ1PLn7SToDnY7HdwW7UoZvQGRwdFVJFQd9B2nvaWdW4Sxu+cgtVJdWM6dkDk/ufpK1z65Nmbh4zfJruGvrXfgDfkI2hNu4yffms+GKDax8eGXK13S73Ph7/GnnjA1nJ0ldKpGTm4mEPY6uZcuW2a1bt472ZYiIiIyYdKmB2c6vGmqpAkMyFZdX/PoK3mh6g5lFM/nsez/LgZYD7G/Zz/bD29nauDUpeTEbZZPK8Af8dIW6KMkt4SeX/ITTpp7GnNI5TMuPzCFbcvcSdhzZweIZi9n+he1J11pxS4VjZ64op4jG6xoBWL9zPXub9zK/bD41Z9RQmFOY9t/JkplL4ssfK4srBzyUWkQkFWPMNmutU2K/Om0iIiIjLd3ywBUPrqDxusaMBUG6RMaBpjU6BYZsqdsSn1FmsRz1H2XqD6dy4SkX0hnojBdoAG8ee5NN+zdlfB2DwWVcjssYc125+IP+eMqjr9vHmo1r2HjVxvgQaadB17Eisii3KKvExdVLV/d57UxJjmNxX5mInBxUtImIiIyw9W+sJ2ydo+TDNsz6nesdi4pE6RIZX6p/iYa2hj4FTezx3l2zSZ5JNLQ10N7TDsDh9sN8+rFPs7d5Ly/Vv9RnRllPqIeNezamvLYFZQsozCnk9SOvE7TBPo/nefIIEyYU7Fu0dYe7SUzZ713Mbj+8PV6QATS2NfYJMMlUfKWTLslRRGS0qGgTEREZYXuO70kZeNER6GBv8960z0/VaUpc3gfvFjSPXfkYlcWV/Gb3b7jhdzfE4/Cb/E2U3FgSD/uIOdJxhHU71qW9BrdxAzh2yw63H+azSz/La4dfc3yuP+jnqr+4iid2PZHUDfO6vRhj6Ap29XlO2IZZt30dN/zuhqw6lMNRfCkMRERGi2u0L0BERGSo1GyoYcndS6jZUDOmztXbgikLKPAWOD4Wi5FPZUvdFipuqaCxLbI3q7GtkfKby7n31Xu56IGL8PX44gOhYwXNBfdfwKl3nMra367tMwTaYh27fh7jYdG0RSmvI2RDuIzzHyPCNsxx//G07/G86vNovK6RiuIKACqKK/jS8i85FmwQKWaf2v1Uxg7lcFq/cj3bv7BdoSAiMuLUaRMRkQkj1ZLBkTpXtvvM7v3Yvax9dq3jORJj5GOyGRz9uSc/169rzSRog7zT/A4F3gLHrqDHeByj8yFSYE3Jm5KyqIu9x97zxxZNW5Ty9Qq8BVjsoDqUIiLjlYo2ERGZENKFU6Qy0MAOp+dl2kuWWAQ6hWUYDLmeXD6z5DN8e/O32X9iP7WtkeHRgx0cDbBkxhK2H9me+cAE6WaeuV1ucl25KQus06efnlUgSOKSw5rTa9IWs5eeeim/r/t9ytdM16EUERnPtDxSRETGPaclgxW3VLClbkva58UKqV3HdiXd71QAJj72Uv1L7Diyg5fqX8LX7Uv7+tZajvuP4w/4ATjacZQvP/1lbnrxJqpKqjDGAJFlil3BLn78px9z6x9v5bFdj/Ha4deSCrbYPrJUPMb572ILvAX85Yy/TLlcMRV/0M/l77mcopwiXNE/MrhwUZRTxBN/+0TGTlosECRxCWTjdY1JIw0SlxzGilmn19t41UZWLV6V8TVFRCYiFW0iIjJmZbOvLDE+v/derhUPrognIjo9z6kwS1eAOT0260ezUu4lO/c/z6Xw+4VM/eFU9jTviTynvZGf/OknPLn7Sd44+kZSJyvPk8epU07lonkX8Y9L/pHi3GIApuVPY/c1u7nr0rtSFl75nnw8bueizWVc3HjhjSkLnlQS954tLV/K9ILpLC1fSuN1jVw076K0BVZiIEhZXhlAVlH56Qq9TEWdYvhFZKLKenmkMcYNbAUOWms/aoyZCzwElAGvAv9gre0xxuQC9wPvA44DNdbaA0N+5SIiMuFls69sIPH5qVIWN1yxgZUPr3RMJ7zo/osIE+6zn6wj6LzHKvZ4rMOWyG3cfHzhxzlr9llUl1ZTXVqdNDi69/Ud9x/nffe8jw1XbEhZeLldbh694lFWPrzScTnirKJZ8eWK/oCfkA3hNu54/L7TdSbuPXNKYowVWLGh0xXFFUMydLowp5CzKs+i7FgZC6cuTDrfcL2miMhY1p89bV8B3gKKo7d/ANxqrX3IGHM3sBr4afTrCWvtfGPMldHjtF5BRGQCGehesP7Ido9aNvH5vUNAUg22vvTBS+PLFXvrDHUO6H14XB6C4eRZZSEb4vl9z3P/J+7vU2ykG7y98uGV8cLSqTDLNJ8s9vj6nevZ27yX+WXzqTmjJmn2Waq9Z6mkK7AGI93nqneAiQo2EZnosirajDGzgUuB7wFrTeS/aOcDfxc9ZB3wHSJF22XR7wE2AHcYY4y11iIicpIYiaJmNA1lSqOTVJ2wjVdtZPGMxUnDoVctXpU2cXBW4Sx+seMXHG4/TF1rHTUbaugMOBdgQRuEIfyvVY47B4MhSN8B06m6gJk6h/Wt9WkLs0zzyQpzCvu85mC7V5k+45pvJiIyONl22m4Dvg7E/opzCtBirY39V6gBqIh+XwHUA1hrg8aY1ujxx4bkikVExoHhLmoGaziKyoGe09ftSyrCHq95PGWn6aIHLsLtcuPv8RMmzDH/Md5qeosQzgmHHYEOvvLsV+K3T3Sd4Om9Tw/o/XldXlzG1WfOWTrWWnrCPSmvzSmiPpvO4XAMjh7O7tVE/IsLEZGRlLFoM8Z8FGiy1m4zxnwodrfDoTaLxxLPezVwNUBVVVVWFysiIkMjVVGZTeGVKsAjXdx9KokdtVgRdsqPT0m9RDGY3CHLtKesv/I9+YQJOw54znXngsGxaMvz5CUVk7HlhV9c/kXu/NOd/Yqojw3eVqx9eureicjJJJsYqbOBjxtjDhAJHjmfSOet1Jh4tvBsoDH6fQNQCRB9vARo7n1Sa+091tpl1tpl06ZNG9SbEBGR7KWLs08VgR/jlJ4440czmHnzzKzj9mOJkJevv9wx9bE71O1YNGViMPFI/HxvPhfPu5gcd06/zuF2uXn8yscd0wmf/vunefqqpx0f++0//JZD1x3qk3j4rQ98q98R9TWn1yjWPguJowJERCa6jEWbtfYGa+1sa201cCWwyVp7FbAZWBk9bBXwePT7J6K3iT6+SfvZRORkkq4oGm0DnWcGqaP1O4Od+AP+lHH7n1j/CRb+ZCHn/PwcbnrxJjbu2ciOIzt4fNfjSUsgB+ufzvonzph+BgALyhbwlzP+kp6Q89JEiOw3c4qNv2jeRY4R9+dUnZM2jj4WyLF4xmLOqjyLwpzCAUXUj2as/cKpC1k8Y7G6VyIiY0x/0iN7+wbwkDHmu8BrwH3R++8DHjDG7CXSYbtycJcoIjJ+pAvQSBwoHDPcgSWJ+8UqiivY07wnaW5ZYoG165pdjmmNsXPsOrYr5V4rJ+097Uz74bR412x3825erH8x/njifLLBKvAWsHDqQn6777fx+zItM/zBhT/gFzt+0e8wj3RpiU7/DgcS8jFasfbqWomIjE39KtqstS8AL0S/3wec6XBMF/DJIbg2EZFxJV1U+4oHV9B4XWOfP3QPZ2BJ7/1iR/1HsSmiEQOhAKf8+BR6gpHOVKzY/P4F3+cbz38Df8Cf8rmpWOyAljkCTHJPAgM9wZ74HrF8b37GeWKxom3h1IXUnF7D2mfXOp7fZVysWrKKNWeuGdD19be4GUjIx3BF6YuIyPgzmE6biIjwbrfM6/b2a8hztnPIspWpq5au6OoKJRdXsWLzmqevGfD1DIbX7WX3tbv52K8+ltQJyzRPrHcxFRsmPZD5Y0NtIMEZ6nyJiAiAGQvbzZYtW2a3bt062pchIieB4ViOuOTuJew4soPp+dNp8jelPO76s6/n+xd+H+jbBYsVE8vLl3O883j8+noPhU6Mxt/0qU189snPsuvYLqbkTeGVxlfi5zOYfnfGxoLeg6KdtPe0x5cNVhZXZlw22N/jRURERoMxZpu1dpnTY+q0iciwGmtDpge6HNHX7WP9G+vZc3wPC6YsoOb0mvh+r1i3LBAOkO/Nd1y+B/Crnb/imx/4JkDKZZQv1L4Q79b5un3xGP1DvkOU31KeNJ+s/OZy8rx5HPUf7fNaI1WwuXCR68nFGENXoCtegLpdbgLhQFbnyPfkM71gOv6gv8/eMif9XWqoZYYiIjLeqWgTkWE1loZMD3Q5Yqwr5g/4CdkQbuNm7bNrufHCG7n++evjoSMtXS1pi6X61nrKbiqjLK8sZWEXW/0QK9Ji5+7dwQsTpj3QTnugvc85+sNg8Lg8/R4aHevkxQIygKTQjK+f/fXIzyaL4BK3y83rX3q9X8VUf5cajoW/MBARERmobOa0iYiMewONuk8MF4mlHYZsCF+PjzUb1yTF3ycWbIa+w6HDhAmGgzR1NKVMToydo8nfRHtP+7B0zAq9hVQURSLrZxfPpvartf2eZxYbfh3rdPWOu1+1eFXKWWMxg4mx14wuERE5mahoExkHYsOIazZoqO5ApJovljhLLJX1b6xPGS6SSr4nnzxv3qCueSjFCsjEIdFnV50dL7BmFc1KORfszhV3Ot5/yuRT+rxOYiGVatZYobeQ06ac1mf+mYiIiKSm5ZEi48BYWmLYH0OdjjhQ6QqvUDjEGXedQWegk6rSKh6veZyNezfG96690fRGv2aTAfiDfsdO22go9BZSMqmEg76DSbO+ehdK6eaCfWrxp/rcf87PMxdaozVrTEREZKJR0SZyEhqqcJB053EaMl1+czkVxRWc6DwRTz/sXcQlxtanOqb38U4BIYn2HN+TsvDyB/3UttYCcNR/lIpbK3AZF2EbCdSIdeb6a7STG93GTb43n41XbeQnf/pJ/N/TQAI+nII8st1TphAQERGRwVPRJnISGqrOXarzpBoy3R5oZ9fxXQAc8x+j4paKpGj33jH4TsfE1Gyo4dVDr3Kg5QDW2qSAkMTjfd0+DrcfxuvyZkwzjBVasa7cQAs2r8vL5LzJNHWkjv/vbagi+nPduaw5cw2Lpi6i5owax65aOqmKsd5FeX+Kfe07ExERGRwVbSIyIOmWPmazDyy2p+zcX5zL35z2N/zisl+kjMGPHfPIFY/EX/vFuhc56DuYdM5YQMiF91/Il5Z/CY/xcOfWOwmFQ1nHzw+GwVCYUxjvbsWKylA4lLIgcxs3MwtncqTjCMFwMOW5XbjI9+aTn5NPU0cT0/MjEfmxEQDZzDfLhgosERGRsUdFm4j0m9PSx8SOWLrliL2FbZjN+zezbse6lIVe2IbZtH8Td/zpDv5Q/wf+++3/pjuYOp6+O9TNrX+8tf9vbJAKcwppvK4xqbvV3tPOuf95Lq8dfs2xcMv35vP2NW+z/fD2PsO2E4u02H6w1U+sji91vO/j93HeuvOoa63Lar6ZiIiIjE8q2kROMgMJB0ncZ1ZRXMGe5j1JiYuxjtj5687nlo/cQmVJJQXegqwLtxNdJ1j77Nq03bCWrhauffrarM6XjaFajhjjwuW4Z6swp5Btn9/WZ+lnYmcsVuQ1XtfYpwhLLNIKcwr7dMJe+dwrQ/YeREREZGwysUGuo2nZsmV269ato30ZImOSr9vHorsW0dDWwOzi2bz5pTezTmBMDAq592P38t3//S63/fE2AuEAFuu4pK53EMj3zvseK3+9Ml5sZCp2YvHuA90PNl4V5RTFu2yptPe0x5MUK4srlaQoIiIiccaYbdbaZU6PqdMmMoY5LUMsu6mM6tJqls5amnb/ka/bx0v1L9HQ1sAh3yFm3TyrT+crcVZZ43WNfZboHfUf5SMPfiTpOZm6U+O5WIsVpLH0SCd5njzcLjedgc54+EkspTFTAZYqnVFEREQkHRVtImNUqgTGcDjM3ua95HlSD2/uXew1+dOnGIZtmHXb13HD725Ier3Rjq0faXNK5vCHz/6Bp3Y/xeYDm3nkrUfoCfb0Wc64ZOYS1u9cz97mvcwvmx9PacxGtlH5IiIiIjEZizZjTCVwPzATCAP3WGtvN8aUAeuBauAAcIW19oQxxgC3AysAP/Bpa+2rw3P5IqNrqOadOcmUwFjfVs8df7qD+tZ6FkxZwIr5K7hs/WUcaDlAc2dzxvTGRB2BDh57+zF6Qj1DcemjJl2HDDLvY7vyjCuZWTiT1UtXs3rpau7uuTvlYOjVS1cP6BqVzigiIiL9lU2nLQhcZ6191RhTBGwzxjwHfBr4nbX2RmPM9cD1wDeAS4AF0X/+Cvhp9KvIhNPfeWepBkE7FX+ZEhgTgzk8Lk/auPhMXLh4fv/zA37+aErsgAGseHAFYRumI9BBgbcAg2HNmWswGBp9jTzy1iOOP9cCbwHzy+Yn3afB0CIiIjIWZCzarLWHgEPR733GmLeACuAy4EPRw9YBLxAp2i4D7reRhJM/GmNKjTGzoucRGdd6B3vEUhjrWuv42jNf4/Tpp1Nzeg1An+Jsx5EdfQqKtc+uZcMVG+J7zxLTHBdMWZB1AuNgCjYYmX1o8f1iuMj15GKMoTPQmbLzNck9CQzx5YmJ8j35dAYjz+3dAWu8rjHl0kVft49H337U8fVcxkXNGTV97ldnTEREREZbv9IjjTHVwP8CZwB11trShMdOWGsnG2N+A9xord0Svf93wDestVt7netq4GqAqqqq99XW1g7yrYgMjXRLHpfcvYQdR3Ywv2w+R9qP0N7TnlR0uIyLXHekIOkOdseDKvI8eYQJ4w/4HV+z97K9BWULyPfm8+cjfx43+8rcxk3IhhwfK/QWcurUU2loa4hH2QOcdsdpfQZkxxTlFLH72t2c+bMzI8sTiyq44QM30NDawPyy+Wzcu5F3mt/p99LUdNH7gxlKLSIiIjIYQ5IeaYwpBB4BvmqtbYtsXXM+1OG+Pn/qtNbeA9wDkcj/bK9DTh7DuV8slcTExd4zzBLnm73T/I5jMRW2YTqDnUn3hWyI9kB7n2MT9T7XnuY9g3kbw8aNG+My5Lpz491Cl3Gx4YoN1LfWx8M7AqFAn2RFp4Lo7KqzefXQqxxoOUA4HO5TRM0snBlPW5yaP5U1y9fEnzvQPWWp5qFp6aOIiIiMVVkVbcYYL5GC7UFr7X9H7z4SW/ZojJkFxOLpGoDKhKfPBhqH6oLl5JCueIrJpqhzOiY2h+xAywEKvAV84rRPcPr006kqrorPI4NIvH7FLRVJe6Vij42X7lcir8vL2rPWYjDc+cqdSSmR2crPyWf3tbt5avdTjssPY+Ed2SYrxv6dJM4v673ccTjSFgtzCjWUWkRERMaNjMsjo2mQ64Bma+1XE+7/IXA8IYikzFr7dWPMpcA1RNIj/wr4sbX2zHSvoeHakijb5WuxpYqLZyxm+xe2O54rdszs4tlc/p7LWb9zPYc7Dvc5zoUr5b6uHFcOFksgHBiaNzgKXMbFjy/+MWvOjHSqKm+tpKGtIfvnj8ASwtHorIqIiIiMFYNdHnk28A/A68aY2J+MvwncCDxsjFkN1AGfjD62kUjBtpdI5P9nBnHtcpJJNZsscQB0YU5h0lLFVJ04X7ePQ75I/k1DWwO3v3x7ytdNF8TREx7fMfgQSUZctWRV/PaZ5WfS6GtMGY8f24c3u3g2zV3NI7KEUIWaiIiIiLNs0iO34LxPDeACh+MtsMbhWBnn+tsJGUjnJN1ssrANs37nehZOXZi0VDFxGWOsC7SlbguX/PKSjHvJxqt8bz6nTTmNc6vP5c5X7iQYDjr+3GJ7zjZetTGp4Hqk5pE+HU2DIcedw8pFKzmv+rx+DYwWERERkeGTdRCJSH9mkmWzJy3x2PPvP5+6ljpcxpUy4r4j0MEbTW/wtWe/5tiJO+8X5/GZJZ+hvaedDW9tGNfLGdMp8BZw+8W3x4M4/t95/y++h+z5/c/TGegkz5vHhXMvTLunTIEcIiIiIuNDvyL/h4v2tI0P2ewhq9lQkzYNsPd+qN7dnnRcuHjvzPfy+tHX6QmN/yWLALnuXLpD3f16TlFOUXyZqIiIiIhMDEMS+S8nt2z3kL1Y92KfuVu996RZa1n/xnp2Nu3kP7b9B13BrqyuIUyYbYe3Dc0bGgMKvYXc+OEbaWhtiHfIGn2NeN1eZhfN7rP0MTE+XwWbiIiIyMlDnTbJyCnNMd+bz5fO/BJYWDBlQTwuv/ew6UT5nnw+MOcDbN6/OTJ8up8dprEsz5OH2+WmM9CZNJ9swxUb2LRvE7e+fKtj8ZUpibG9pz3r+HwRERERGb/SddpUtAkQ6ZKtf2M9e47vYcGUBayYv4KNezdm1Q0zmHE5twxgZuFMzp1zLo/vejw+EDodl3FhrcVlXIRsKCnoY8nMJSkLLBVfIiIiIpKOijZJydft47v/+11u++Nt8e7XJPckukJduIwrZZLjeOQxHqYWTCUUDjGndE5S6EasqHqj6Q1+uu2njkVqrjuXA189QGFOoQowERERERlSKtpOUr27ZzWn1wDE7wO485U7U6Y1TiQVRRW8fc3bWRVXseWg/oA/aanjcA6WFhEREZGTm4JITjJO3TOvy8vnn/x8fOLeROqgpTLJM4lgOEh1aTVLZy3NuhsWi8JXN01ERERExgIVbeNEYtessqQSgL3Ne2lqb+Ko/yhul5uPnvpRTi07lcsfvrzPUOn4zLLRb6xmpcBbQNiGCdtwysABhaiFAAAgAElEQVSSAm8BBsPs4tkc7zxOgbeAjy78KL5uH1Pyp7Bo6qIBF1uFOYXxOWgiIiIiIqNJRdsoyLRsMfG+dTvW8cCOB9h2aBsWm7ZD9uw7z47I9Q+nXHcua85cEy+4ANZtX8dTe54CCxfOu5BcTy4NrQ3qgImIiIjISUF72kZY7/1SMbFUwsQURheujAOnx6ocd05WA7C9Li+BcCAphVH7xkRERETkZKM9bSMs1knb2bSTE50nKMwp5ETnCerb6nmp/iXHQsypgzbeCrbE2WT1rfUZkxjf+NIbvHDgBe0bExERERFJQ0XbAPTeX9YV7OKZvc9wsO0gvh4fjb7GcTu3LJ3YHrI1Z67BYJhfNp9LT72Up3Y/lbLw+j+L/g8rHlxB2IbpCHQkddTmlc1jXtm8UXxHIiIiIiJjn5ZHZsHX7WPdjnVseHMDrx95neau5tG+pGHhcXlwGzfBcDAedZ/nyUsq0gbSDdNgaRERERGR9EZleaQx5mLgdsAN3GutvXG4Xms4banbwkUPXERnsHO0L2VA3LgJEcp8IJDnyWP3tbvTds4GQkmMIiIiIiIDNyxFmzHGDdwJfBhoAF4xxjxhrX1zOF5vqHUHu6lrrePNo29Ss6EmZeT8WOd1ebn14lu5/D2X88ibj8QTGE+bdho/e/VnWGv7LFmcWThTBZaIiIiIyBgyLMsjjTFnAd+x1n4kevsGAGvt952OH+nlkf6An7rWOg60HOBAywFqW2o50Br92nKAQ+2HRuxaBsuNm8WzFrPj8A7cLjc9oZ6skhi1ZFFEREREZOwYjeWRFUB9wu0G4K+G6bX68HX7qG2tjRdhta217xZorbU0dTRldR6DGROBIgYDRPac5bhz+nTHzqk6p99FmJYsioiIiIiMD8NVtBmH+5KqH2PM1cDVAFVVVf06eUtXS5+CLLEwa+7MLijE4/JQWVxJdWk11aXVzCmZE/laGvn6zN5n+MozX8lq3thAGQxzS+cyr2weF8+/GIDn9z1PMBRkWuE0ZhbOTBo0naowUxEmIiIiIjIxjbnlkdZamjubk7tjvZYvtna3ZnUdOe4c5pTMYU7pnHhBlliglReV43a5Uz7f1+2j/OZy2gPt/fsB9GIwuHCxeOZiOoOdGAwVxRVcdtplrFq8SssSRUREREROcumWRw5X0eYBdgMXAAeBV4C/s9a+4XT89AXT7Zn/ema8Y9bek12RlOfJi3fFEouy2PczCmfgMq5BvZf+pkfOLppNyIbwurzMKprF8orlLJmxRHvGREREREQkpREv2qIvugK4jUjk/8+ttd9LeWy5sXy+7/2FOYXMLZ3bp1MW+35q/lSMcVqJObTae9pZt30dj7z1CG8dfYtAOMC0gmm8Z+p7aO9px2M8XLrwUnXNRERERERkQEalaOuP0lNK7afv+nRyYVY6h8mTJo9IUSYiIiIiIjKaRmW4dn/ML5vPbRffNtqXISIiIiIiMuYMbsOXiIiIiIiIDCsVbSIiIiIiImOYijYREREREZExbEwEkRhjjgK1o30dGUwFjo32RYgMEX2eZaLRZ1omGn2mZSLR5zk7c6y105weGBNF23hgjNmaKs1FZLzR51kmGn2mZaLRZ1omEn2eB0/LI0VERERERMYwFW0iIiIiIiJjmIq27N0z2hcgMoT0eZaJRp9pmWj0mZaJRJ/nQdKeNhERERERkTFMnTYREREREZExTEWbiIiIiIjIGKaiLQNjzMXGmF3GmL3GmOtH+3pEsmGMqTTGbDbGvGWMecMY85Xo/WXGmOeMMXuiXydH7zfGmB9HP+d/NsYsHd13INKXMcZtjHnNGPOb6O25xpiXo5/n9caYnOj9udHbe6OPV4/mdYs4McaUGmM2GGPejv6uPku/o2U8M8Z8Lfpnjp3GmF8ZYybp9/TQUdGWhjHGDdwJXAIsAv7WGLNodK9KJCtB4Dpr7XuAvwbWRD+71wO/s9YuAH4XvQ2Rz/iC6D9XAz8d+UsWyegrwFsJt38A3Br9PJ8AVkfvXw2csNbOB26NHicy1twOPGOtPQ1YTOSzrd/RMi4ZYyqALwPLrLVnAG7gSvR7esioaEvvTGCvtXaftbYHeAi4bJSvSSQja+0ha+2r0e99RP4wUEHk87suetg64G+i318G3G8j/giUGmNmjfBli6RkjJkNXArcG71tgPOBDdFDen+eY5/zDcAF0eNFxgRjTDHwQeA+AGttj7W2Bf2OlvHNA+QZYzxAPnAI/Z4eMira0qsA6hNuN0TvExk3oksO3gu8DMyw1h6CSGEHTI8eps+6jHW3AV8HwtHbU4AWa20wejvxMxv/PEcfb40eLzJWnAIcBf4zuuT3XmNMAfodLeOUtfYg8COgjkix1gpsQ7+nh4yKtvScKn7NSJBxwxhTCDwCfNVa25buUIf79FmXMcEY81GgyVq7LfFuh0NtFo+JjAUeYCnwU2vte4EO3l0K6USfaRnTovsvLwPmAuVAAZFlvb3p9/QAqWhLrwGoTLg9G2gcpWsR6RdjjJdIwfagtfa/o3cfiS2piX5tit6vz7qMZWcDHzfGHCCyTP18Ip230ugyHEj+zMY/z9HHS4DmkbxgkQwagAZr7cvR2xuIFHH6HS3j1YXAfmvtUWttAPhv4P3o9/SQUdGW3ivAgmjyTQ6RDZVPjPI1iWQUXRd+H/CWtfaWhIeeAFZFv18FPJ5w/6eiCWV/DbTGluiIjDZr7Q3W2tnW2moiv4c3WWuvAjYDK6OH9f48xz7nK6PH629wZcyw1h4G6o0xC6N3XQC8iX5Hy/hVB/y1MSY/+meQ2Gdav6eHiNHPJz1jzAoif6PrBn5urf3eKF+SSEbGmHOA3wOv8+4eoG8S2df2MFBF5BfsJ621zdFfsHcAFwN+4DPW2q0jfuEiGRhjPgT8k7X2o8aYU4h03sqA14C/t9Z2G2MmAQ8Q2cvZDFxprd03Wtcs4sQYs4RIsE4OsA/4DJG/TNfvaBmXjDH/CtQQSbB+Dfgskb1r+j09BFS0iYiIiIiIjGFaHikiIiIiIjKGqWgTEREREREZw1S0iYiIiIiIjGEq2kRERERERMYwFW0iIiIiIiJjmIo2ERGRfjDGfNUYkz/a1yEiIiePMRH5P3XqVFtdXT3alyEiIiIiIjIqtm3bdsxaO83pMc9IX4yT6upqtm7VjEgRERERETk5GWNqUz2m5ZEiIiIiIiJjmIo2ERERERGRMWxMLI8UEREREREZDjUbath1bBcLpy5k/cr1o305A6KiTUREREREJqxdx3ax48iO0b6MQclqeaQx5mvGmDeMMTuNMb8yxkwyxsw1xrxsjNljjFlvjMmJHpsbvb03+nj1cL4BERERERERJ75uH8c7jwNwvPM4vm7fKF9RsrAN0x3spqOnI+1xGTttxpgK4MvAImttpzHmYeBKYAVwq7X2IWPM3cBq4KfRryestfONMVcCPwBqBvd2REREREREsrelbgsrHlwRL4ga2xqpuKWCjVdt5Jyqc0bkGsI2TCAUoCfUw+H2w+w/sZ8DrQeoa62jvrWeBl8Djb5GGn2Nac+T7fJID5BnjAkA+cAh4Hzg76KPrwO+Q6Rouyz6PcAG4A5jjLFjYSCciIiIiIhMeL5uHyseXIGv593OWpgwvp7I/Y3XNVKYUzjo1wmFQwTDQbqCXdS11nGg5QC1rbXUtdRR11bHQd/BeFHWFewa8OtkLNqstQeNMT8C6oBO4LfANqDFWhuMHtYAVES/rwDqo88NGmNagSnAsQFfpYiIiIjISWoiBGmMtPVvrCdsw46PhW2Y9TvXs3rp6oznCYVDBMIB2rvbqW2tZX/LfmpbaiOdsrZ6DvoOcrDtIEc6jhAMBzOeD6DAW0BFUQXlReWUF5czu2g2VSVVXPOda1I+J5vlkZOJdM/mAi3Ar4FLHA6NddJMmscSz3s1cDVAVVVVpssQERERETkpTYQgjZG25/geOgLO+8Q6Ah3sbd4LQDAcJBgOcqLzBPtb9rP/xH7qWusi/7TV0djWyEHfQY75j2H7ljSOyvLKKC8qjxdmlcWVVJVUMad0DtWl1UzNm0qOJwePy4PX5cWYSPl0DYMo2oALgf3W2qMAxpj/Bt4PlBpjPNFu22wgthCzAagEGowxHqAEaO59UmvtPcA9AMuWLdPSSRERERERGRILpiygwFvgWLh5XV5+X/d7PvzAhyNLF9saaeluyeq8BsOMwhmRLllRpEtWWRIpyqpLq5lTMofSSaWRgsztxeMamrD+bM5SB/y1MSafyPLIC4CtwGZgJfAQsAp4PHr8E9Hbf4g+vkn72UREREREZKjFQj7q2+ojIR/RPWX7TuxLuYcsEA7wYv2Ljo95Xd54QTaraBaVxZVUFlcyp2QOcyfPpaqkinxvfrwoc5mswvgHLZs9bS8bYzYArwJB4DUiHbKngIeMMd+N3ndf9Cn3AQ8YY/YS6bBdORwXLiIiIiIiE5e1lmA4SEegI1KINe+jtrWW2pbayH6ytoMc9B3kcPthAuFA1uedXTybeZPnUVFUQWVJJZUllcwtmcuc0jlUFFeQ487B64p0yWJLF0dbVv06a+2/AP/S6+59wJkOx3YBnxz8pYmIiIiInNyc5owV5RaN8lUNDWstgXCA1q5W9p3YFwn5iCYv1rfV09AWicNv6mjKej9Z6aTS+F6yiuIKZhXM4u5td3Oi6wSzCmex7eptTM6bjNflxe1yD/M7HDpDs8hSRERERESG1FiYMzYY1lp6Qj00dTSxvyW6dLGlltrWWhraGuJF2YmuE1mdz2CYVjCtT8BHVWkV1SXVVJdWMzlvcp+Aj417N3Ki6wTTC6Yzq2jWcL7lYaOiTURERERkjBmpOWODEbZhuoPdNPoa2Xdi37szylrraGhriC9fTJXi2JvH5WFm4UwqiiqoKK5ITl2MFmUFOQV4XJ4hC/gYL06udysiIiIiMg4M1ZyxwQjbMP4ePwdaD3Cg5UBSHH5DW0N8cHS2+8nyPHnvpi4Wz44XZdWlkYKssriSXE/ukAd8LJy6MOnreKSiTURERERkjMl2zthghMIh2rrb2NeyL568WNdaR31rfXxwdFNHU8risbfS3FLKi8vjSxcrSyKpi3NK5nDK5FOYXjCdHHfOiAd8TISB5CraRERERGTM8XX7OP/+86lrqaOqtIpNn9o0YQI4spFuzliBt4D5ZfMzniMUDnHUf5R9zZGliwdaD/TplDV39hmnnNK0/GnxpYuzi2dHli6WRAZGnzL5lHEZ8DFemLEwQm3ZsmV269ato30ZIiIiIjIGJAZwhAnjwkVBTkG/AjhqNtSw69guFk5dOC47Lb5uHxW3VCTtaYspyimi8bpGct25SfvJYnvKEuPw23vas3o9t3HH95PNLn53YHSsSzanZA6FuYV4XJ4Rm012sjHGbLPWLnN6TJ02ERERERkzhiqAY9exXew4smM4L3VYFeUW8WjNo3z8Vx/HH/TH7/cYD3NL5/KXP/1LGn2NdIe6szpfrjs3HoMfHxhdGumSzS2dS2VxJZO8k066gI/xQv9WRERERGTMGAsBHCPF1+1j/4n9vNPyDgdOvLt0Mbaf7Ej7EUI2lPScoA3y56Y/9zlXUU5RPOAjtnSxuqSa6smRomxW4SxyPDnqko1TKtpEREREZMwYiQCOkWCt5bj/eLwg6xOH7zvIMf+xfp1zkmcS5889/90o/JI5zJ08l7mlc5maP3XEAz5k5KhoExEREZExYygCOHzdPo53HgfgeOdxfN2+IQ8xCYfDNLY3sv/E/kjyokPIR1t3W1bnchs3MwpnxPeTVZVUJUXhr3p0FTuP7mThlIU89XdPDen7kPFBRZuIiIiIDNhQB37UnF7D2mfXOj7mMi5qzqhJ+/zEEBOAxrZGKm6p6FeICUAwFKS2tfbdkI/WA9S1RJYuNrQ1sL9lf9ZR+DnuHMoLo0sXS5JTF+eWzmVO6RwmeSalXLqoNEZR0SYiIiIiAzbUgR9FuUVsvGpjyvTIdCEk/Qkx6Qx0RuaTNe9nf0tkRll9Wz31rfU0+Bo43H6YYDiY1TUXeAuYXTw7HvKR2CWbWzqXiuIKctw5g/vByElNRZuIiIjIGDXeY+sH6pyqc2i8rpFFdy6ivq2eiuIK3lzzZsbUyHQhJp2BTj7w8w9gjKGhrYGj/qNZX09ZXll86aLX5eWZvc/QE+559wAL93783n518vpj4dSFSV/l5KOiTURERGSMGu+x9YNRmFNIWV4Z9W31lOWVJRVs1lqaOprYd2If+1r2xZMXf7vvtylDTII2yPYj2/vc7zIuZhTMeDcKv6QysnSxpJq5k+dyyuRTKM4txhgTn52WVLABHcGOfo0j6K+TqWAXZyraRERERCTO1+3j/PvPp66ljqrSKjZ9atOQh3hkErZhGlobmDxpMpXFlQRCAVY9turdkI+2g3QGO/t1Tpdx8b5Z7+O86vMi88lKqplXNo+qkiryvHlZneNkGkcgY0tWRZsxphS4FzgDsMA/AruA9UA1cAC4wlp7wkRyRm8HVgB+4NPW2leH/MpFRERExrDxuLQxMcQjTJhj/mMDCvHIpCfUE0ldbIkmL7ZEkxfbIkXZId8hAuFA0nPePPam47nyvflUFEW6ZDMLZ/LrN3/d57kQ2Xe2adWmQXXCJso4Ahl/su203Q48Y61daYzJAfKBbwK/s9beaIy5Hrge+AZwCbAg+s9fAT+NfhURERE5aYyVpY3ZFo/9CfFIfI5TtH5HTwfvnHiHd5rfcUxebOpowmKzuv7JkyZTURyNwi+OBHzMKZ3D3NLI0sVp+dNwud5NXfzi8i8OKMQkG0MxjkBkIDIWbcaYYuCDwKcBrLU9QI8x5jLgQ9HD1gEvECnaLgPut9Za4I/GmFJjzCxr7aEhv3oRERGRCWqoZo1lWzxmu/TPWktzZzOPvPUIX3n6K3SFugBoaGug9MZSCnIKkgq/dAyG6QXTmV08O2lgdPXkak4pPYW5k+dSMqkk+zfLwENMsjHYcQQiA5VNp+0U4Cjwn8aYxcA24CvAjFghZq09ZIyZHj2+AqhPeH5D9D4VbSIiIjKhpOpiDbbgynbW2FAuwcy09O/f/vffuOmlmzjYdjDlcbHOXIzX5WVW4azIbLLiyMDoWJds7uS5zC2dS64nd1DX7SRdiMlgDGYcgchgZFO0eYClwLXW2peNMbcTWQqZinG4r0//2xhzNXA1QFVVVRaXISIiIjK2OHWx0hVci2csZv0b69lzfA8Lpiyg5vSaPsVcf5YpDmQJZiAUoL61/t3kxZYD1LbW8srBVzCYlMsWa1trszp/jiuHL//Vl7nmzGuoLKlMOTB6vBrOTp5IKtkUbQ1Ag7X25ejtDUSKtiOxZY/GmFlAU8LxlQnPnw009j6ptfYe4B6AZcuWZbeoWURERKSfRjIQJF3BddEDF+F2ubHW0hHooMBbwNpn1/bpnmW7TDFVN88f8FPbUss7J95h/4n9NPoifwzbfXw35TeXc6TjSMrzp+I2bj61+FPML5tPdWk1G3dv5MGdDzoe2xPuwePyMKd0Tr9eYzwZrk6eSCoZizZr7WFjTL0xZqG1dhdwAfBm9J9VwI3Rr49Hn/IEcI0x5iEiASSt2s8mIiIiwyndMsWX6l+ioa1hUPvCnDgVTWmHO/eKqI8tMezdPcu0THFn007ue/U+rtl4TfJ+sh+UUphTSFt3W8rX72xPvobe+8nCNsx9r94XP2/i0r/EwtIf8PPYrsfGdCCHBlLLRJJteuS1wIPR5Mh9wGcAF/CwMWY1UAd8MnrsRiJx/3uJRP5/ZkivWERERKSX/i5THGx8fapzf2zhx1IWXKn0DvmYXjCdXHcu3aFux+Nve/m2lOdJLNg8Lg+zCmdxvPM4/oCfafnT+M6HvhNPXawurXbcT3bjhTdy3rrzqGuto6qkis2rNvfpJI2HQI7h7qqqKJSRZCIhj6Nr2bJlduvWraN9GSIiIjJOLbl7CTuO7GDxjMVs/8J2fN0+Km6pcEwxLMopcoyvz1a6c+e6c/G4PP0u3CqLK/G4PBz0HaQn1DOg64J395N9+a++THlROf6An0V3LaKhrYHZxbN580tvDlmnsfdMt1RdORHJjjFmm7V2mdNjE2tnqIiIiAjZ7QvLxNft495X7+Ubz32De1+9F1+3L+O5XbgcBztnUt9Wz/6W/WkLthxXDt/+4Lf5+7/4+5THxPaTVZZU8oeGP1BxSwWNbZE9bbFu4Ja6Lf2+PiexQI6K4goAKooraLyuUQWbyDDIdnmkiIiIyJjktLcs076wvc17gdR74WJdJH/AT8iGcBs3a59dy8OffJgX615Mee7OUKfj/em4jZvPLv0sp045Nb50ce7kubiMK55QWFlcGU8ovPfVe3n07UfT7icbyKDsgVAgh8jIUNEmIiIiIypVoTSQlMdUe8u+uPyLFHgLMgZlJO6Fs9bS1NHEm0ffZMWDK+JhHAAhG8LX4+OSBy/J6rrcxh1/Xp4nj6vfdzVu4+auV+6iJ9RDmDBu4ybfm592OaFTQZTNfrKHdj6UVQKliIwPKtpERERkRKWaLdbfmWPpukl3/ekujHEaHRspWoI2yHde+A51rXUAvHXsLfL/PZ+uYJfjc7KV585j6+e3cuqUU1l2zzJ2HNnBqVNO5baLI+Eh/3rev7J+53r2Nu9lftl8as6o6Xd3KpsBz9l2GkVkfFDRJiIiIgPW3+5Yqtli2Tyv91DqdHvLekI9vHfWe9nWuI2gDSY91hns5Au/+UKf47N17fJr+eTpn+TS/7rUsWhaNG1RyucW5hT2q8OVKqEw04DnBVMWZNVpFJHxQUWbiIiIDFh/umMDjeB3Sim85qlrqJ5cnbKb1BPu4eWDL6c859T8qcwuns2+E/to625jVuEs7lxxJ6dMPoX/qf0fvvm7b6YseBbPXMwH5nwgbdE0VNIVwun2k41kJL+i70WGn4o2ERERGXK9O3CZgjHeXvM2TR1NQGRQ9Hf/97scaj/E3uN7eX7f84QJJz2vO9zNruO7Ur6+wVBdWs2y8mXMKZ1DdUl1POCjqqSKfG8+8O6ogOkF0/nEez4BwCmTT+Fbm77leN7EgidTCMdIFDOpXiObJZRDZbjnoYmIijYRERHJINUSyHRLHXt34H6181cEw8nLFGPae9qpuLUifvt453H+efM/Z7wuFy4wOC6RLMwp5M9f/HPa4iTV9ScWPGEbpiPQQYG3AJdx9avgGYliJt1rZFpCKSLjh4o2ERGRLAwk2XA8vV46Tksgey91PNh2kFk3z+J753+PgpwCDvkOAbC3eS9zbptDQ1tDyv1nFut4v9u4ec+097Czaafj42HCXHXGVTyx64l+d5MyLdWMFTyDDQ0ZbYrkF5kYVLSJiIhk4Ov28VL9SzS0NfQrPCNmIAVYf5MUB8op4CPVe+sKdvH0nqd569hbXP/89UlDpC2WjkAHX332q0nP6Qh00NHqvO8sk0meSSyduZT9J/an3F92XvV53P3Ru/vVTcp2hlk2oSHazyUiI0FFm4iISBoDDc9INJAo+4EkLKY7n1Nh5hTwce3T17L6vasxxlDbUktda12807Xr+C5W/NeKfr9+riuXf/7QP/Pvv/93/AF/1s/rCHQwJW8KLuNyfDy2v6y/3aR0qZP9nWE22l1QETk5qGgTERFJIduOTKZz9KcAG2yR2Lurt6VuC5f88hL8AT9hwhgMa55awwWnXMBv3/ktIRtKem9dwS7ufOXOlOf3urzkefNo627LeC0x3eFufrDlBzxa8ygrH16ZVCR63V6MMY7z0Qq8BZw+/fQhD9TQDDMRGW+c/+pKREREsurI+Lp9LP/Zcmb8cAbLf7YcX/e7Bd6Wui1U3FJBY1sj8G4BtqVui+M5E4vEWFpiYpHY3tMORAqzJXcvoWZDJMUwEAqw78Q+ntr9FM+98xw7juzgN7t/w/vvez8f/M8P0h5oj5/PYukJ9/D03qeTCrZEBsOiqYv46KkfJceVk3R/rjuXLyz7AgXegv78KAnbMPWt9TRe10hFcSR0pKK4gv1f2Y/X5XV8TqyTFttftrR8KdMLprO0fCmN1zUmFbELpy5k8YzFWS1TjM0wc6IZZiIyFqnTJiIikkKmjszmA5v52rNfi3eAjvmPxbtii2csztilW/3E6qSuWLoiMRgO8s+b/pnTpp7G5v2bOeo/yp7mPZTfXM7h9sN9wjz8AT9/aPjDgN63xfKReR/h3tfupSfck3R/e6Cdu/50F8aYfp0z1sEqzCnkrMqzKDtWxsKpC5lVNCurTlphTiGvfO6VlOfvzzLFkZxhJiIyFFS0iYjIhDaYFMZYR8apcMv35PPIW48kLetLLMr+/YJ/T9ulW7d9XTzc5Kj/KC/WvcjGPRtTFomdwU5ue/m2pPv8AX+/9oj15nV5k8JEYgq8BRzvPJ427fGaZddw19a7+rXUMdbB6v3vYaSj6UdyhpmIyFDIenmkMcZtjHnNGPOb6O25xpiXjTF7jDHrjTE50ftzo7f3Rh+vHp5LFxERySwWArLrWOpBzKnUnF6TMgQjZEO4jdvxsbAN89Tup9J26a59+loa2hoAaPQ1cs5/nsOjbz+a9npcxsWckjnxpX3TC6Zz38fvY+1fryXPk5ft2wIiRafb5Xz9LuNiSv6UtNdvjBnQUsdUYmEiwIhE08cKxcTr773kUkRkrOjPnravAG8l3P4BcKu1dgFwAojFLK0GTlhr5wO3Ro8TERl1vfcBycTnFAKSTu/PSKwjU5RTFBnkTGSgc1FOEZcvujxtUfOng3/CkHoJYarZZKmeU+AtoPnrzbz+xdeZnDcZgBx3Dp9c9Ek8Lg+dwc607603t8vN41c+7vjeNl61kUXTFmXc99W70IotdUx1zrHWwRrpQlFEZKCyKtqMMbOBS4F7o7cNcD6wIXrIOuBvot9fFr1N9PELTH8XvouIDIPBdFxk5KUL+MhGf0NAwKTNLM4AACAASURBVPkz8r5Z7+OFT7/AlPwpABTkFHDx/It5ueHltEVZc1dzysIslXxPPl9//9cdi55n/v4ZXm963fE9YUhZYE1yT2KSZ5JjEXXRvItSBnyk6zImds16B4CogyUiMvSy3dN2G/B1IJZRPAVosdYGo7cbgIro9xVAPYC1NmiMaY0ef2xIrlhERCa83vPDegd8nH//+dS11FFVWsWmT23is09+NmnfWn+i+lu7WqltreWto29R21oLwO7ju3nv3e/loO8gR/1Hk67N1+Pj12/+Ou31e4yHLy7/Im7j5j+2/QddwS4sFhcu3C634z4yAH/QH1922Ht/l7WWilsqHN9TumAQr9vL7mt387FffYy61jqqSqrYvGpzxoCPbPd9Oe0T7B02km0HS4OqRUScZSzajDEfBZqstduMMR+K3e1wqM3iscTzXg1cDVBVVZXVxYqIyPiXatBz4uOpCq6LHrgIt8uNv8efVMzNKJyRNFsrXQpjZ6CT99/3fiyWutY6x3ljncFOth/ZnnSf27gpLyqnqqSK6tJqqkurCYaD/PjlHycVZbGiJtZZ+rfz/y2pAPv62V/n+uevd1xa2XvZYeKw6HtfvbffwSCxa5lZODNt8mIqgwkIGcjQaQ2qHh4qhkXGv2w6bWcDHzfGrAAmAcVEOm+lxhhPtNs2G2iMHt8AVAINxhgPUAI09z6ptfYe4B6AZcuW9W/9iIjIBJMp4XAwCYgjKdN1xjpo/oA/HuSx9tm1SUVO2oKr176tWDHna44UePtb9vPpRz/NC7UvpNxvFrRBXm96Pav3M8k9icf+9jHeM/U9lBeV43H1/c/mtz74rbRFTe8CbNXiVXzzd990fL3eyw4Tv2YaP5CqQzfYfVpOBaSML2P5d4aIZCdj0WatvQG4ASDaafsna+1VxphfAyuBh4BVwOPRpzwRvf2H6OObrLUqykRE0ojtpRro42OBr9sXj7CPhX4U5RbFC7l5k+fx3L7nkjpoIRvqM7fshf2pC65M2rrbWPfndWmPcRs3H5zzQVYsWMGckjlsP7yd2/54G/5g3+h8t8tNQ2sDH5n3kZTn629RM9Blh+nGD6Tr0ImIyPg3mDlt3wAeMsZ8F3gNuC96/33AA8aYvUQ6bFcO7hJFZCwaL50fGT6Jn4Frz7w2XoTAuwEZG6/aGC8408396gn18PknPx8fGj1YHuPB5XLRE+rp81i+N58n/vYJVj+xml/++Zd0BbscCzZ4dyB0JpmWn/V+PLbs8Lx15znuM3OigdAiIievfhVt1toXgBei3+8DznQ4pgv45BBcm4iMYeOh8zNQI12QOsXSJ+7xGklO793pvsSu2jH/MTbu2Uh7T3v8PLFlixc/cDH5OfkAHO04Sneo2/F1u0Pd/NfO/xqy95HryeWa5an3eFlr49c/edJk8r35jkOqEwdCp5Ppc5IqrKM/+8yy7dBp/5KIyMQzmE6biMiENJIFaWJKIiR3qGJ7vFIVdUNdXDotbwT63LfjyA5WPLgiXqQd9B1Mec6OYAcdwch7S1WwxRhM2tllBd4CwoQdi6s+r5tmj9f2w9upuKUi/jNv6WpJ+bpjrYOVTTDIcPxFw0QuBCfyexORiUNFm4icFMZSNyvxmjLF0m8/vD3jssNU5+4di+/0fmOF35S8KbzS+ErS60z/4XQw0B2MFFsNbQ2U/aCMMOGUyxwHw6lwchkXU/KmMKd0DptXbU76eYRJfQ2Je7wSo+ettX1+5omv68KVsoM1VozGvrWJvAR6Ir83EZk4VLSJyISXTTdrNKRLSfQH/Jx2x2kc6ThCMByM3x8r6i755SWUTCoB4Jj/GHf86Q7qW+tZMGUBVcVVrPz1yqQZZ+U3l/OlM78ElnjMfntPO/9z4H840nGkz+uHCdMV6upzf9AG+9w3EC7jImzDuI0bj8uDMYauYN/Xy/Pk8f0Lvs/qpauBvp2mVN25xA5Z4h/K08Xm53vyyfXkcqLrxJAlL4qIiAwFFW0iMqH1Z8hy7PiR6sili3AP2RAHfQcxjqMvoT3QHn/uQd9Brn36WiCSjBiyoaRjw4RpD7Rz04s3xe/73JOfG4q3MGAfmfcRFs9YzPyy+bxx9A1u/eOtjsc5BYEkdprmlc3jSPuRtHu8EqX7mfuDfgpzI89R8qKIiIwlrtG+ABGR4ZSumxW2YdbvfLcLs6VuCxW3VNDYFhk7GevIbanbktVr+bp9LP/Zcmb8cAbLf7Y8vicslViEezqp9lqleqx3wTbSYkVmqmITIksXL3/P5Xz/wkgHbdG0RSl/DqmCQBZOXcjiGYtZOmspjdc1srR8KdMLprO0PHI7VQc13c+8wFtAVUkVi2cs1v4mEREZU9RpE5EJpXc4R6aBxLEuTn87cr0lLsGMLUlMtwTT1+3jrlfuyipUY7wo9BZSMqmEg76DlBeV09rdmpQoGdM73GMgUfa99yFlm8KY6bUyxe6LiIiMBnXaRKTfnJYQjgWx9MMdR3bwUv1L+Lp9GTsrsS5OfzpyTq8bK/hi4RiJBV/vwmVL3RbKby5n++HtaTtpY1WeJ4/CnEJc0f+EuHBRlFPE03//NGdXnc3iGYs5u+psnr7qaYpyivoc13vpYizKPptjB2skX2s4xTqN6giKiJwcjLWj/weGZcuW2a1bt472ZYhIFnp3lBL3EI1mqEeq69pwxQZWPrwyqYMWU5RTFO+gfeO5b3DTSzc5nDniw3M/zMdP+3g87KPm9Jr4Xrd7X72Xrz7z1ZQdvcmTJnPVX1zFY7se45j/mGPgxljjwkWuJzcSEBLo6vPvesnMJVkNhm7vaY+HhlQWV6YN9+jPsYM1kq8lIiKSDWPMNmvtMqfHtDxSRLI22CWEo3FdKx9eGS/c0oVVxDpyqQqv5/Y/x3P7nwMiBc1Xn/kqN190M6WTSvnln3+Z8nkAJ7pOcMcrdwzqPcZSEiuKKlIuO+wtFl/fH70j9oGUM8GyWZLYn3j6kYyyH43YfBERkYFS0SYiWctmCWEsmn0kZbqu+tb6tAOJfd0+uoJdBEKBrF4vTJiOQAdfeOoLQ/YeYnLcOfSEevrcb4zBWsvU/Kk8tPKhlLPK3MZNvjefDVds4P9u+r/UtdZRNqmMBl8DnYFOQjaE27jJceek7KL17pgmzjkbSHHTn+HFGnQsIiLSl4o2kUHqHXwxkWUb6jFcUv2sf/TSjzJeV2JnpXRSKQfbDlLbWsvz+57n9j/eTsiGRj15scBbwDXLr+GurXf16QouL1/O8c7jLJy6sM+ssoqiCm74wA00tDYwv2w+NWfUUJhTyEXzLoqfu72nnfU717O3eW/8GEjdRUs02M91f54/0f8/JCIiMhAq2kQGadexXew4smO0L2PI+bp9nH//+dS11FFVWsWmT21Ku4TQKZo98RwVxRUAHGw7GD8f0Oc1YvvEfN0+1r8RSX+M7SFL9bNOtzQR4K5X7uLtY2+z+/huAF5vep3T7jyt/z+UYeYyLr517rf41rnfylhMFeYUJnXA1ixfk/bchTmFjl3Qk3mJoLp6IiIyXqhoE5E+UsXXb7hiAy7jHDrbO5q99zma/E3xx475jzHjRzOSlucd9R+l7AdlXHDKBRTnFvPk7icJhALx5Xxfe+Zr5OfkA9DU0cTm/Zs55j9GbWstnYHOtO+nraeNx3Y9NgQ/meERW9KYuMcum2JqKLpSJ3Phoq6eiIiMFyraZMw5mZYbjgWxn/e8yfOoa6vjQMsBmjubk/aIZRPqseGKDTy08yH2HN9DZUklN/zuhpRhGWHCdAaTCy2LJWiDPPvOs32OD9kQ/5+9O4+Pq77u//86MxqtI3lftdhgG4NNa8dhCW1KADsUTILb4NhNSEMSJ6RpSKDQNDTN99H098hC0gJxCgklJqmdUhAYUrKYENYkTgIBDAQvGBuwtXq3pZFka5n5/P6YhRnpzmgkS5qR/H7y8GM0996585l7kazj8znn09bdRlt39HzNbc1csuGSAX/WeDOPfFHoL2Tp7KVcteCqxJTGuJEKpvQ9JiIikv8UtEneGavTDfNV/Hq/euBVcGTsNujV1GNc8Tjee/p7WXH/ikRmzGe+tI1BhlNFUQWFvkIOHT/kud/MwJFV4GZY4viBdmBMPkem9yryF/HAqgeGpY5MRERExo5+gzYzqwY2ANOBCHC3c26tmU0EaoHZwB5glXPuqJkZsBZYDnQAH3PObRme4YtIb161YPE6seRjLtlwCXuO7aHlRAtAVkFWe3c739/yfRyOgD8ARNvZP7D9gZTjchGwAbR1tuH3+dPuH8i4goVBXv/c67z/vvdT11JHVXkV2NvNWPo7VzAQ5IzJZ9DQ2pDo3tjR1ZF2yQERERGRdLLJtPUANznntphZOfCimT0OfAx40jl3i5ndDNwMfBG4HJgX+3M+8L3Yo8iYE+oMcfj4YQAOHz9MqDPUJ0Aa7vdPbuTxtYu/xsoHVxJx0Zb0ZYEybnzsxpQ27vFas7autkFNFXyu8Tmeb3x+0Nmn4RQhQiSSflyF/kIMozPcmfaY5IBqenB6n7XIQp0hKm+r9Fysu/frk1vnJy/mnKlTo4iIiEhv/QZtzrlmoDn2dcjMdgCVwArgothh64FniAZtK4ANzjkHPGtm481sRuw8InkpUx1dun3JjTYAmlqbqLyt0nOdq8HIlDELdYb46q+/yref/TbdkW4cjoMdB/nLe/8y5RzxroqX/8/lfGPZN9h+cDvf3/J9eiI9JzW2fAzYslHoKwTDM2jrvah0uoCqvKicTVdvSmmyYhgQnXaZbbdHBWwiIiKSrQHVtJnZbOAdwHPAtHgg5pxrNrOpscMqgfqklzXEtilok7yVqY7Oa1+oM8Tye5enZFvizTqW37ucppuaMv5SHg/Ith7YytHjR5lYMpGFUxcmArN4QOiVMYNoEBZvyhGXKWvW1t3G5x/9fF414Rhuhf5CwpFwovtkvDsjkBJwZVpUOp3e66RVVVQxvng8rx54ddi7PYqIiMipJ+ugzcyCwEPADc651mjpmvehHtv6/KZoZtcC1wLU1NRkOwyREZVu+uMlGy5JZNh6i7gItVtrPdfEgrczdN3hbk6ETyS2+8zHjY/dyIa/3sBVD1yVUjMVz5hd9qPL6HE9Gaf3pZPrgK28sDztlEIvRf4iAFYuWElVeRV3vnAn7V3tWX2OskAZ31z2TYoLilMWk44HUykLUw9yqmLvzNnOQzsH9HoRERGRbGUVtJlZgGjAdq9z7uHY5v3xaY9mNgOIL8LUAFQnvbwKaOp9Tufc3cDdAOecc86p88//o8yp0H4/XWDWe/pjQ2sDFbdUMG/iPJrbmtNOEWzvbmf3kd0p54/Xnc0on8Frh17zDLoiLpqp++vav0471vaezItID6WAL0B3pHvIzvfu6nfz67pfey6EbRizxs3iA2d9gMPHDzOpdBILJi9ICbSSF5w2DJ/5CLuw53v5zMc1i69JG4gN1VTF5O+J1Ruja9SdiuudiYiIyPDKpnukAfcAO5xztyXt+glwDXBL7PGRpO3Xmdn9RBuQtKiebfQa6fb7I93YI11dWnwtMq/M0K4ju/o9731b7+Nz532Otc+tTak7S15gOp8tmbGEjy/+ODc/cbNnkJXMb/5om398aQPZskAZV5xxBZvrN3vuDxYGefXvX80YPAULg4kFp8+eejbXn389T+95mod2PERXT9eAuzIO9T9CjNV/1BAREZHcyybT9ufA3wKvmtnLsW1fIhqsPWBma4A64IOxfZuItvvfTbTl/8eHdMQyZg13Y4+4eD3ZluYtfZpyxOvSrrzvypPKMu1t2Uvl7ZVDMdwRYxjBwmDieoc6Q3zpyS+lPT5YGOShVQ9R31LP7iO7qaqo4uYnb/ZcUDue+Vo0fVGfWj2f+Qbc+t5nPtYsWcOaJWu4q+sudWUUERGRMS2b7pGb8a5TA1jqcbwDPnuS45JTzMk29oifI7n9/VMffSolS5fccdHhMgZlg6kZG80ml05m9vjZKV0Tvbokxp056Uyev/b5PvckHpR1dHf0aQASLAwmGnjUbq31rDXrT3zqYfIUxOQMXKYmICIiIiKj1YC6R8qpZSSnKtZuq027WHF/jT0gNUsXIcLBjoNM+tYk/uGCf+DLf/FlXtn/imfHxVNZ78yal3iQdfH6i6lrqaNmXE3GdvjZBGXBwmDGe5mJpiCKiIjIqUhBm3gaqamKcbsO70pbO9W7sUdvXlm6eCbtW7/9Frf+9lYiRHLePTEX4nVmhlHoL2TF/BX8ofEPtHe397seWVywMNhngen+jh9sUCYiIiIifSlokz6GYqpiNu+RvE5ZU6iJIn+R57TEAiugKdREU2sTm3ZvYuuBrRxoO8DBjoP4fX7KC8szNssI491hMF/48ePz+QZcQxevB7tl2S3RhiG91h3buGoj//LUv2SVIRvtvKZNioiIiIwVFi1By61zzjnHvfDCC7kehsSs27KOG35xg2cgVBYoY+1laz0zKfGasj3H9tDR3UFpoJTZ42en1JYl15XFOypmw2c+Ii5Cob+QrnDXyX3APJHchANSF3w2jAJfAasWruKRnY/Q0dWRkjFbuWAlF8++ODH1sK2rLespjCIiIiKSf8zsRefcOV77lGmTPvqbqrjtwLY+2+PTKdu62hKBWEd3B4c7DjPl36dw5fwrae1s5ck3n6TH9fR5fX/i9W65DNh85mNSySSqK6r5xJJP8OaRN3l4x8M0t0VXtOiJ9KRdNyzujIln8Pl3fZ6GlgbPBZ+9Aq9sArKBTmEUERERkdFDmTZJTFXcdXgX8ybN49jxY3z56S+n7aDow8d757yXsyafxcKpC1k+dzln3nmm55pmY0E2DTsA2rra+Oqvvsrtz91OT6SHiIvgMx9+83PjBTfy5Qu/rOyXiIiIiHjKlGlT0HaKSW6LX1lRyXtmvYfvPv9dzIzOcCfF/mJOhE9kfT7D8JufAn8BJ3qyf10+8OHj0rmXEo6E+fXeX6etp7v6T6/mjuV3ZB1wtXW1DbqlvYiIiIicmhS0CdC3Lf6pzGc+Gm9sZHpwOqHOEJW3VXpmCssLy4ek8YqIiIiISCaqaRvDendhjNdATSyZmJi6uGn3JrY0b+H7W75PT2Tg9WSjUcAXAKJ1ZsnNTnzmoyxQxqarNzE9OB1IXUQ63aLQIiIiIiK5oqBtFAp1hlj/ynp+9MqPeLH5RczMMxgzDIfDb/5+G2SMFcnB1uLpi6ndWsv2g9s5fPwwk0onsWDyAs/pitksCi0iIiIikguaHpkD8aDrZ6//jJ5ID1PLpjKtbBpzJs7hRM8JnnjzCQCWnb4MgCfefIKeSA8Tiiew59getjRvGTPTG/34wfAMKov8RTicZ8dIHz7OnHwms8bNYtmcZRQVFHl2ZBQRERERGQ00PTIPxKcxPv3W02zcsTGr1vWPvfHYCIxs+Jw56Uz2tOyhK9yVaNkPJNYgi3dUfHnfyyy/dzkRF6G9uz2xftnGVRtZ+cBKz2tVVljGc596TsGZiIiIiIx5CtpGwC93/5IVtSvo7ukmzOibpljkL2LlgpU8uP1BIpFIVuuslQXK+Mc/+0dWn7263ymKmaYmxmvNegd0qjUTERERkVOFpkeepPhUx/977f/Y17aPKaVTmFw6mUMdh9jXto9QZ4jGtsZcD3PQivxF7LlhD9OD0xOt7B9/43Eeeu0hwpFwSpOPZEPZdVEt9EVERERkrFPL/5OUHJg1tjbSE+mhtauV493Hx8yC0gFfIFFD1juj5bWgdDyQenrP0zy04yH85s/qdSIiIiIi0ldOgjYzuwxYC/iBdc65W9Idm89B2+a6zVz6o0s53nM810MZFslBVrzb4kAzWsqEiYiIiIicnBEP2szMD7wOvBdoAJ4HPuSc2+51fL4GbftC+5jznTl09HTkeij98uPnvXPfy9yJc9l9eDdPvvUkPvPRGe6k0F9IJBJh2ZxlLJiygNPHnw6Gui2KiIiIiOSJXHSPPA/Y7Zx7MzaA+4EVgGfQlgvOOY6eOMreY3vZ27L37cekrw91HMr1MDPy4eO8yvP4yKKPcM2ia1ICL2W/RERERETGhuEK2iqB+qTnDcD5w/ReniIuwr62fRmDsrautpEcUtbii2IXWAE9rgcfPhyOi2ZfRIGvgAIr4Ir5V/QJ1JIFC4OsWbJmhEcuIiIiIiJDbbiCNvPYljIP08yuBa4FqKmpGfAbdIe7aWhtSA3IYo97ju2hvrU+q7XQAEoDpcwaN4tZ42dFH2Nf/3H/H7nt97fRHeke8PgGqrSglLALs3LBSi6efTFXnHEFP3/958qUiYiIiIic4oarpu0C4CvOub+MPf9nAOfcN7yO96pp6+juoK6lLiUg29OyJ/r82F6a2ppSFmzOZELxhD4B2axxs5g9fjazxs9iUskkzPrGmaHOEDNvnUlb99Bl5Hz4eOfMdzK5ZDLL5iyjqKBItWUiIiIiIqe4XNS0PQ/MM7PTgEbgb4APpzv4QPsBbnrspkSWrK6ljoMdB7N+sxnBGamZsl6P5UXlg/oQ5UXlPPqRRwfUPTI+tTH+WFNRQ7AwSGVFJSvOXJFxSqOIiIiIiEhvw9nyfznwbaIt/3/gnPta2mNnmuPT3vsKfAVUVVT1zZDFnldXVFNUUDQsnyGurauN9S+v55Gdj9DQ2kA4Eqals4XucDdFBUWcNfksPrDgA1x11lWa0igiIiIiIgOW94trl9SUuPd8/T2eWbKZ5TPx+/y5HqKIiIiIiMiwycX0yAFZOHUhv/jIL3I9DBERERERkbzjy/UAREREREREJD0FbSIiIiIiInlMQZuIiIiIiEgey4tGJGZ2ENib63H0YzJwKNeDOAXpuueGrntu6Lrnjq59bui654aue27ouufGaLrus5xzU7x25EXQNhqY2QvpurnI8NF1zw1d99zQdc8dXfvc0HXPDV333NB1z42xct01PVJERERERCSPKWgTERERERHJYwrasnd3rgdwitJ1zw1d99zQdc8dXfvc0HXPDV333NB1z40xcd1V0yYiIiIiIpLHlGkTERERERHJYwraRERERERE8piCtn6Y2WVmttPMdpvZzbkez1hjZj8wswNmtjVp20Qze9zMdsUeJ8S2m5l9J3Yv/mhmS3I38tHNzKrN7Gkz22Fm28zs+th2XfthZGbFZvYHM3sldt3/Lbb9NDN7Lnbda82sMLa9KPZ8d2z/7FyOf7QzM7+ZvWRmP4s913UfZma2x8xeNbOXzeyF2Db9nBlmZjbezDaa2Wuxn/MX6LoPPzObH/t/Pf6n1cxu0LUffmb2D7G/V7ea2X2xv2/H1M94BW0ZmJkfuBO4HFgAfMjMFuR2VGPOfwOX9dp2M/Ckc24e8GTsOUTvw7zYn2uB743QGMeiHuAm59xZwLuAz8b+39a1H16dwCXOuUXAYuAyM3sX8E3g9th1PwqsiR2/BjjqnJsL3B47TgbvemBH0nNd95FxsXNucdI6Sfo5M/zWAr9wzp0JLCL6/72u+zBzzu2M/b++GHgn0AH8GF37YWVmlcDngXOcc2cDfuBvGGM/4xW0ZXYesNs596Zzrgu4H1iR4zGNKc65XwNHem1eAayPfb0e+Kuk7Rtc1LPAeDObMTIjHVucc83OuS2xr0NE/0KvRNd+WMWuX1vsaSD2xwGXABtj23tf9/j92AgsNTMboeGOKWZWBVwBrIs9N3Tdc0U/Z4aRmVUAFwL3ADjnupxzx9B1H2lLgTecc3vRtR8JBUCJmRUApUAzY+xnvIK2zCqB+qTnDbFtMrymOeeaIRpcAFNj23U/hkFsWsA7gOfQtR92sSl6LwMHgMeBN4Bjzrme2CHJ1zZx3WP7W4BJIzviMePbwD8BkdjzSei6jwQH/NLMXjSza2Pb9HNmeJ0OHAR+GJsOvM7MytB1H2l/A9wX+1rXfhg55xqB/wDqiAZrLcCLjLGf8QraMvOKurVGQu7ofgwxMwsCDwE3OOdaMx3qsU3XfhCcc+HY1Jkqotn8s7wOiz3qug8BM3sfcMA592LyZo9Ddd2H3p8755YQnQb2WTO7MMOxuu5DowBYAnzPOfcOoJ23p+N50XUfYrHaqSuBB/s71GObrv0AxWoEVwCnATOBMqI/c3ob1T/jFbRl1gBUJz2vAppyNJZTyf749IDY44HYdt2PIWRmAaIB273OuYdjm3XtR0hsutIzRGsKx8emdEDqtU1c99j+cfSdTiz9+3PgSjPbQ3Sa+yVEM2+67sPMOdcUezxAtLbnPPRzZrg1AA3OuedizzcSDeJ03UfO5cAW59z+2HNd++G1DHjLOXfQOdcNPAz8GWPsZ7yCtsyeB+bFus8UEk11/yTHYzoV/AS4Jvb1NcAjSds/Guu29C6gJT7dQAYmNnf7HmCHc+62pF269sPIzKaY2fjY1yVE/6LZATwNrIwd1vu6x+/HSuAp51ze/2tgvnHO/bNzrso5N5voz/GnnHNXo+s+rMyszMzK418DlwJb0c+ZYeWc2wfUm9n82KalwHZ03UfSh3h7aiTo2g+3OuBdZlYa+/0m/v/8mPoZb6NgjDllZsuJ/ousH/iBc+5rOR7SmGJm9wEXAZOB/cC/Av8HPADUEP1G/KBz7kjsG/EOot0mO4CPO+deyMW4RzszezfwG+BV3q7x+RLRujZd+2FiZn9KtPjZT/QfzR5wzv1/ZnY60QzQROAl4CPOuU4zKwZ+RLTm8AjwN865N3Mz+rHBzC4C/tE59z5d9+EVu74/jj0tAP7XOfc1M5uEfs4MKzNbTLTpTiHwJvBxYj9z0HUfVmZWSrRe6nTnXEtsm/6fH2YWXUJnNdHu2C8BnyRauzZmfsYraBMREREREcljmh4pIiIiIiKSxxS0iYiIiIiI5DEFbSIiIiIiInlMQZuIiIiIiEgeU9AmIiIiIiKSxxS0iYiIDICZ3RBr6y0iIjIi8qLl/+TJk93s2bNzPQwREREREZGcePHF3E+J9gAAIABJREFUFw8556Z47SsY6cF4mT17Ni+8oLUERURERETk1GRme9Pt0/RIERERERGRPKagTUREREREJI8paBMREREREcljeVHTJiIiIiIiMhxWb1zNzkM7mT95PrUra3M9nEFR0CYiIiIiImPWzkM7eWX/K7kexknR9EgREREREZE8pqBNREREREQkjyloExERERGRMSnUGeLw8cMAHD5+mFBnKMcjGhwFbSIiIiIiMuZsrttM5W2VNLU2AdDU2kTlbZVsrtuc45ENnII2EREREREZU0KdIZbfu5xQV4gIEQAiRAh1Rbe3dbXleIQDo6BNRERERETGlNpttURcxHNfxEWo3Zpfrf+dcxn3q+W/iIiIiIiMKbsO76K9u91zX3t3O7uP7B7R8fREeuiJ9NAd7qY73M2+tn3sbdlLfWs9Da0NNIYaM75eQZuIiIiIiIwp8ybNoyxQ5hm4lQXKmDtx7pC9V8RFUgKyw8cPU9dSR11LHQ2tDTS0NtDc1kxzWzP72vaxr20fXeGuAb2HgjYRERERERlTVi9czY2P3ei5z2c+Vp+9OqvzOOcIuzDd4W56Ij20drYmArL61nrqW+tpDr0dkDWHmtNm+LyUBkqZHpzOjOAMfstv0x6noE1ERERERE7K6o2r2XloJ/Mnz6d2Ze7rxcqLytl09SaW37uc9q52IkTw4aOssIxNV28iWBgEIBwJR7NkkW6Odx+PBmMt9SnTFpvbmhOB2bETx7IeQ8AXSARk08ujj5XllVSNq6K6oppZ42YxqWQSAX+AAl8BpWtK055LQZuIiIiIiJyUnYd28sr+V3I9jBR/Xv3n7L1hL3/yvT+hMdTIlLIpfPeK77LtwDYe2/1YYtrivrZ9NLc1c7D9II7MDUHifOZjSukUZpbPTARmMytmUlXxdkA2PTidQn8hBb4CAv4AfvNjZoP6LAraRERERERk1Ik39+jq6eJgx0H2tuxNTFtsaGmgqa2JfW37ONB+AID97fu56oGrsjr3xJKJ0QxZLCCrrKhMBGTV46qprqimuKCYAl9B4s9gA7JsKGgTERERkZzJt2l1Y9Vou87JzT2OHj8aDchao1MXG1obaAo10RRqormtmf1t+zneczzrcwcLg4mAbGb5TGaWRzNkVRVVzBo3i5pxNQQLgyMWkGVDQZuIiIiI5Ew+Tqsbi/LpOjvnElmy9q72RIYs0W0x1JDS3KO1szXrcxf6C5kRnMGM8hnRKYvlM3l4x8M0tzUzZ8IcnvjoE0wqmZRXAVk2FLSJiIiIiIyQ0ZbxGox4c48TPSdoaG1ImbbY2NqYyJLta9vH4eOHsz6v3/xMC057e8pieSWVFZWJKYs142qYXjY90dgjHpBtrttMc1szwcIgs8fPHr4PPoyyCtrMbDywDjgbcMAngJ1ALTAb2AOscs4dtWiouhZYDnQAH3PObRnykYuIiIwRp8IvcSISlU8Zr8FwztEdia5Htr99P28dfYv6lnr2t+8H4M2jb3Lu989N1JKFXTjrc08pnfJ2U4/ymYk6sppxNdSMq6GyvDKljizbDNn8yfNTHkejbDNta4FfOOdWmlkhUAp8CXjSOXeLmd0M3Ax8EbgcmBf7cz7wvdijiIiIeBjtv8SJSHZCnaFEZunw8cOEOkOUF5XneFSp4tMWD7UfenvaYmt02mI8SxZfILoz3Nnn9aGuEC/te6nP9nFF4xI1ZDPKY63vkwKy6orqlDqyoZyyOBb+MazfoM3MKoALgY8BOOe6gC4zWwFcFDtsPfAM0aBtBbDBOeeAZ81svJnNcM41D/noRURERERGgc11mxNrhgE0tTZReVslm67exLtr3j1i43C4REC2t2Vvyppk8cYeA10gureAL8C1S67lPbPfk5i2OL54PAFfYNTUkOWbbDJtpwMHgR+a2SLgReB6YFo8EHPONZvZ1NjxlUB90usbYtsUtImIiIhIwmjIPPU2mOnMoc4Qy+9dTqgrlNgWIUKoK7q96aamxGLPJ6Mn0kNHVwf1rfVv15HFui2+cfQNALYd2MaU/5iS9TkDvgDTgtMSzT2qyqsSnRYnl07mgw98kPae1ACvO9LNhj9u4Jb33jIkn0uyC9oKgCXA55xzz5nZWqJTIdPxCp37rFJnZtcC1wLU1NRkMQwRERERGSvyIfMU6gxRu62WXYd3MW/SPFYvXN1v0DiY6cy122qJuIjnvoiLULu1ljVL1mQ8RzgSpivcRVOoiT3H9ry9HlnStMWmtiYOdRxK+15ASo1ZfIHoeEAWn7JYVVHFrPGzmDVuFjPKZ1DoLyTgC/TJkK3bss77N/8BfC7JTjZBWwPQ4Jx7LvZ8I9GgbX982qOZzQAOJB1fnfT6KqCp90mdc3cDdwOcc8452S09LiIiIiKj3khlnjKJB40d3R2EXRi/+bnxsRuHJWjcdXhX2umG7d3t7Dq8i66eLg50HGDvsbfb3/eetri/bT/dke6s33dC8QTGFY2jvrU+JVgr8Zdw1/vuYsWZKygJlFDgK8BnviH/XLuP7B7wOcVbv0Gbc26fmdWb2Xzn3E5gKbA99uca4JbY4yOxl/wEuM7M7ifagKRF9WwiIiIiY9dApwwORebpZHgFjWEX7jdoHOx0zjkT51DiL+F4uO8C0D7zcc/L9/CdP3xnUAtEzwjOYGbFzMS0xZpxNYksGcDsb8/u08HxePg41z16HR9Y8AEK/YVZv2dv8ybNoyxQ5hm4lQXKmDtx7qDPLamy7R75OeDeWOfIN4GPAz7gATNbA9QBH4wdu4lou//dRFv+f3xIRywiIiIieWWgUwZznaEZTNCYbjrnzz78M94x/R3UtdQlpi327rbYFGryDNji73eo41DKtuQFolPWIquINvU4bcJpTCyZ2G+GbN2Wdbi+VUoZP+dArF64mhsfu9Fzn898rD579aDPLamyCtqccy8D53jsWupxrAM+e5LjEhEREZExargyNNlm/AYSNHb2dLLj4A4u3XBpSuAVn8550X9flDYw6o/f/LzvjPdx7sxzE63vZ42fxYzgDAL+wKCmLCYb7uC4vKicTVdvSgSzESL48FFWWMamqzepCckQyjbTJiIiIiLSx2CmDA5XhibbjF+moLHACnh096M8/ubjiTqyTAtE9w7YJpdOfntx6F5rkU0uncz7//f9NLY1Ul1RzfbPbh/WwGYkpi++u+bdNN3UxII7F1DfWk9lReWwf65TkYI2ERERERmUwXaAHEyGZqB1c845jhw/wlvH3ko094g39qhrqaOju8PzdT2uZ8DdIa866yq+8p6vUDOuhrLCMvw+f9pjJ5dNprGtkYklE4c9sBmp6YvBwiATSyZS31o/Ip/rVKSgTURERGQQBrNe11hysh0gB5qh6Z1FC3WG2Nuylz1H91DXWscbR95g1+FdAGw7uI3yb5QPeIFowzh76tksmLLg7db342axpXkLt/7uVs+6tLJAGZfPvZyzp52d1XvMnzw/5XE4afri2KGgTURERGQQBrNe11gyFB0g02VoOns6Uxp77DqyK1F/te3gNiZ8cwLHThxLe96eSA89kZ7E84AvwLSyacysmJnS2GNa2TRu2XwLBzsOMnvcbJ75+DOegcyy05ex9rm14DFLcqAZq5EO8DV9cWxQ0CYiIiIyimWT8RuOrOBgm1yEI+HEAtF7j+1lf/t+AN469hZL/msJTaEmDnYcTBsQ9kR6UgI2i63u7NUMpMRfwvPXPs8Zk84g4A94nu+axdek/5Axoz1jpemLo5+CNhERETmljYZpjpnGmE3GbziygpmaXBT7izl64ii3/v5W6lvqE+3vG0ON7Gvb57lAdGtnKy/teynr9y/xl/DbNb/l2YZn+cLjX/Ach8/n49mGZ1k4deHAPpwHZawklxS0iYiIyCltNExzPJkxDnZB6N5Wb1zN9oPbmVk+k+vOu45jJ47RHe4bfAGcCJ/gv178r37PWRYow8xwzjGpZBIf+pMPUV1Rzazxs6LTFfc+w81P3Jw2INvSvIU9x/aM2JpvylhJrihoExEREUkjH7JwmYKu/gKygXR37Ap3JaYs7jm2h/rW+miWLNRAfUs9rx9+HYdj64Gt/PKNX/Y77vgC0fEasqqKKqrHVTNr3CxmjZvFaeNPY0LJBMws7Tl+9Mcf9RuQjURbe5FcU9AmIiIiksZQZeEGG/xlCrqAjAFZpu6OS9cv5eo/vZr97fsT0xYPdRwa0GeaVDKJ2eNnMz04nWf2PEN7dzsTiyfywAcf4KwpZzEjOCNjQJaNbAKyVQtXjUhb+7iR7P4oEqegTURERMTDUE0rDHWG+F3972hobRjQeTIFXZf/z+Vg0NbV1mff0vVL+cQ7PsGzjc8mArreuiJd/PDlH6Z974klE6mqqGJa2TSe3vN0SifGxDnCXTzzsWi3xcV3LeaV/a9QPa6apacv7fezZSubdcaChcERbRKSr3WPMrb5cj0AERERkXyzuW4zlbdV0tTaBLydxdpct3nIzxPqDHHu989l2r9P49zvn0uoMxqkZWqp397dnjEgu+vFu3h538tE8H49wKxxs/jooo/y0T/9KEX+osR2w+gOd3Pn8jtZtXBVyr5k8bb+wynetbG8sBxf7NdWHz7KC8tTArJ4k5DKikoAKisqabqpKeMC3yKjiTJtIiIikreGu6bMK5sGnNSi0cnn7u88zzU8x5X3XUlHTwcABzoOMPGbE/nT6X/KW0ffSlvP5dXePllVeRUzy2eypXkLPa5vlqwsUMb/u/D/sWrhKipvq6Qz3Jly7vgYP7nkk1k1+RjOKYPZdm1UkxAZyxS0iYiISN4azs6O6erFPnPuZ/pdNHrVwlX9Tp3831f/13NaIUSnNU77j2l0dHf02dfjetjSvCXj2C32n1cmrSxQxlcu+koiIEsOGuPiUwvv33p/xs96uONwVk0+hnvKYLAwyAXVFzDx0ETmT56vgGwQVIs3uiloExERyaGhqpsai4by2vTO2GXKgt3++9s91xGDaHbp6T1Pc8MvbkgEMg2tDUz+1mTeN/99hCPhlPXI0mXEHM4zYIsrsAKWz1vOL9/4JSfCJ/rsLwuU9alpixtIrVd/C2RPKpmEz7yraYajyUcmqiU7OcN9/RQUDi/VtImIiOTIUNVNjVbparlg6K9NPGO389BOIHO9GEChr9Bzuw8f9756L23dbSkBWVeki4d3PMwjOx/hheYXaG5rzjiFMeALcPaUs9Pu73E9LJiygMc/+rhnPdejH3mUR69+9KRrveLdGb2UBcpYOHVhVjVlIrUra3n5715WcD1MFLSJiIjkQHKmJz7FLbneySuD0p/VG1ez+K7FrN44ctmPwYoHZVuatnCg4wBbmrYkgrKhvja9M3b72/bzbMOzaTNM3ZFuuiJdnvsyNfbwmY93Vb6LL737S9x1xV089MGHKC0o9Ty2uKCYT5/z6YwB09yJczMGXdk234hPLVw0bREXVF+QEmitXri630zaaGryMX/yfBZNW6Rsj4w5mh4pIiKSA5kyPfG6qTVL1gzonMNZ/zWU+mvQ8fWlXx/QtUme+vijv/4Rja2N1LXUsefYHn5T9xs2vLIhMd2xobWB6bdOH9S4C6yAuZPm8tqh19KO7aLZF/G1pV9LbJsanJp2euKiaYv40pNf8jxX8tTDTA02sm2+kS77Ee/O2F+7/NFSU6Ysj4xVCtpERERyoL9aonhXvqEyHF0YB3vO/gLWn7/+84zX5td7f02wMMielj28ceQNfrLzJ5zoOcGrB17lwW0P9ttZsT/F/mIe+9vHqK6o5sIfXkhDqIHqimq2f3Y792+9P6WeLVlyY464eJbq4vUXU9dSR824Gp6+5ulE0DOS64ul098Y4xQQieRO1kGbmfmBF4BG59z7zOw04H5gIrAF+FvnXJeZFQEbgHcCh4HVzrk9Qz5yERGRUSxeS5TtL/8nq78s3GACsMFm9voLWPe37SfgC6RtBrLhjxvY8McNfbb3DgSDhUE6ujo8pzSWFJRw3bnXcdeLd3kGTPGpf5NKJ9EQakhksbJZ7Lm3YGGQ5z/1vOdrsm1nn6nJw1A0gMg0RhHJvYFk2q4HdgAVseffBG53zt1vZncBa4DvxR6POufmmtnfxI7L/8n1IiIiI2gwv/wPl1BniN/V/46G1oasuzT219kx1Bmidlstuw7vYt6keSyfu5yjJ45S31pPY6iRAivwXD8M4KX9L/U75oqiCkKdIc+sWlmgjD037OHff/vvfOt33/J8/fGe4/h9/qwCpmTZTicciGymHmYKpJUBExn7sgrazKwKuAL4GnCjmRlwCfDh2CHrga8QDdpWxL4G2AjcYWbmnDu5uQoiIiJjyFD/8j/Y9vjp1ipLzjb195rG1kZm3DqDry/9OhWFFdE6sj9uSLtGWX+KC4qZWDKR/W37Cbvw29v9xdx95d381fy/onZbbdppigCPvPZIVtnM/mrCvLJY2WbHBkKBl4hkkm2m7dvAPwHxn/6TgGPOJf6JrAGojH1dCdQDOOd6zKwldvyhIRmxiIjIGDHYX/57T2UcTOAF/TcEabqpieKCYppCTdS31FPXUsfrh1/na7/5WsrURYejvbud639xfb+fucBXwPTgdCqKKnj90Osp2bbSglJqP1jLFfOuwMxo62pLXJt4TVn82mRTE/ilv/jSSWcz0wVT2TYAEREZCv0GbWb2PuCAc+5FM7sovtnjUJfFvuTzXgtcC1BTU5PVYEVERPJd72mBqxeuzpjxSvfLf6Yas+RasmwCr3QBxX1b7yMcCXvua+9qp/q2ao51HhvQ5y8uKKarpyttHdnay9byqXd+CiBjUAaZA6NssmjZZjO1KLCI5LtsMm1/DlxpZsuBYqI1bd8GxptZQSzbVgU0xY5vAKqBBjMrAMYBR3qf1Dl3N3A3wDnnnKOpkyIiw2A4OgZKquRr/LnzPsfye5fT0d1B2IXxm58bH7sxkfEayP1I1+Sj9zTI9a+sT9uJMRwJc+vvbuXcynM51BGd8LK3ZS9/8cO/oKG1gbqWuvRdHIn0CdgmFE/Ab34OHU8/eWbRtEU81/ic577jPcd58+ibiecn00Y+25rAbLKZ+t4QkXzXb9DmnPtn4J8BYpm2f3TOXW1mDwIriXaQvAZ4JPaSn8Se/z62/ynVs4mI5MZoWbcrW/kWhCY38DjUcYhNuzalLPwcduGUjFe29yNdfZpXLdkNv7ghpe4rWUdPB1/51VdSth07cYzNdZv7HUPAF+DDf/JhfrP3N7x57E3Onno2r37mVdZtWZex5f38SfPZemBr1l0x+7uP6bJgA6kJ1FRGERntTmadti8C95vZV4GXgHti2+8BfmRmu4lm2P7m5IYoIiIyuA6H6Qw0+PM6vk8dWagp7fpg8QWhs7G5bjOX/8/liaCnsbWRyd+azLI5y3hs92MpAZrDpQ3YkgV8ASIuQtiFKSko4dPv/DTzJs1jSukUPvZ/H6Ojp6PPa4oLirlj+R28+wfRmji/+aPXop8M1y3LbuHHr/047f6BdsXMdH8GUhOoKZAiMpoNKGhzzj0DPBP7+k3gPI9jTgAfHIKxiYiIAIPrcJhOpuDPKzjzOh7oU0eWaUHn9u52th3YlsieHWg/wFNvPcWR40eoa6mjMdQIwM7DO7nwhxemnMvh6Ip0sWnXpgF9Tog29vjBih/wyZ98ko7uaGDW2dPJPS/dk7h2M8pnDKiDZX8ZrhnlMxL7k6eJlgZKh2XB6GynWA5HZlaBoIiMFMuHmYvnnHOOe+GFF3I9DBGRMWfxXYt5Zf8rLJq2iJf/7uVcD2dQQp0hKm+rTAmQ4soLyzM22ugtOfjzWky59/VKd/xnzv0Md/7hzrTdC3vz4cPF/hsuhf5CesI9KePcuGojKx9Y2e+1S9cQJNQZYsF3F9DQ2kBVRRXb/357IsDtr4lIW1cbtVtr2X1kN3MnzmX12as1LVFEJAMze9E5d47XPt9ID0ZERGQgarfVpm+WMYBph8ldFuOdDZO7LDaHmlPqyJpam9Ief+vvbs06YIu/Ll3ANn/S29mh0kBpxvMEfAHP7WWBMm679DYqK6Kr71RWVNJ0U1PmRiNJ1y5e8wUkar42122m8rZKmlqjfcbi2c14PZzXa5IFC4OsWbKGbyz7BmuWrFHAJiJyEhS0iYjISVu9cTWL71rM6o0Dq1fK5vXZrMeVzXkzBX/d4W5OW3saja3RaYqNrY3UfLsm7ftmU0cGYLH/fOb9121ZoIwv/NkXmDNhDhANfsoCZZ7HlhaU4vf5Pff5zMc1i6/pE0QN9tr1F+DGm63MnzyfRdMWaXqgiMgwO5lGJCIiIkDm9vT9rVnWX43Z7+p/h9/8noGSVzdCr/Me6jhEeaA8bQBzInwi5Xm2DT68lBSU8K8X/StHO44yb9I8th3cxu3P3u55bO/AaULxBFpOtHge6/f5+fGqH7PygZVZ159ls5ZZXHJ9VjbZzTVL1uRFB08RkVOBgjYRkVPUQBeBznSe3u3pP/nTT7KleQt7ju3BOZd2zbL4MZFINEBIbjCyaNqiRNBlmOd7+8zHqoWrWHH/CnYc3MHUsqmsWriK39f/no3bN9LjegBoDDVyz8v3eJ5jMEoLSvnceZ/juy98N219XNy6LeuyDpx85svY5KO/bom9G2Nku5YZpDbq+OLjXxx0dlNERIaeGpGIiIxRmZpIxBtsRFyE9u52ygJliYBhIN0Y0zXqmFI2JWUR5WTlheXsvG4n537/3ETXxN5KCkrw+/y0d7WnrQXzm5+pZVM5cvwIneHOrMc8FOJNPICMzTggu0Yq7/7Bu1MaoPTX5GMgDWb6a77ipb+12NZetpY1S9b0e51ERCR7akQiInKKydRE4qraq3jPf7+HUFco8Ut5e3d7n3ql/mSqe0oXsEG0fuz075xOU6gp7THHe47T1tWWsdti2IVpbmv2DNjSZeaKfEUUFxTji/3158NHkT+6zUuxv7jP8eWF5YnpiPF284umLeKC6gsytsgvLyxPe57e+mvyMRDx7FzvJiWZgvPVC1enrcMbzFprIiJychS0iYiMMf01kXiu8bm09UrtXe1cvP7irN4nU91TJifCJzjRc+Kk2t+PLx7Pje+6kSJ/kef+dOfujHTy9+f8fUoA89b1b6XtyhjwB3jr+rdYMnMJU8umsmTmkj4BT+3KWl7+u5ezWgQ603kGYqANQAYaBA4m0BQRkeGjmjYRkVEkmzq0/ppItHa2pj1/hAg7D+1M1KX1Xmg6+TwvNb80oLb3Q2nWuFmcNeUsCnwFA5oaWRYoY8GUBSmLMScvBu01hXB6cDrPf+r5kx5zsDA4JOeB4Vkourf+6udERGTkKGgTEcljqzeuTgROnzvvcyy/dzkd3R2ejT3i+mvzXuz3ngoYF+oKMfPWmUwomUB9az2HOg7xz0/8M/Wt9ew9tpf61nqaQk10R7qH9LNmw29+ZpbPZP7k+Rk/ZzrxqX2967HyMUDp3VQkF+IZuvrW+pOepikiIoOnoE1EpJdQZ4hLNlxC3bE6asbX8NRHnxpUV8WhGEdyy/pNuzal1JuFXZhQV4hlG5bx9+f+PQumLGD1wtU8tecpfPgSUyN7693e3ktbdxtt3dH3agw1cstvbznpzxPwBfCZDzPjRI/3GAyjLFBGhAgd3R199pcGShPBVKaujMX+YjDo6unKqjU+5F+Aonb6IiISp5o2EZEk8QYeW5q2cKDjAFuatiQaeAxGqDPEud8/l2n/Po1zv38uTa1NrNuyji8+/kXWbVlHqLNvR8HkcSQaiYSa0jYI6Qx3cvuzt3PDL25g5q0z2d+2P23ANlh+87Ps9GWUF0aD18mlk3luzXP84upfpNQ9pVNgBXz4Tz6csX4MYGb5TJr/sZnHPvJYv/VUmZplxGvRBtJ8I5e0SLWIiGSiTJvIKSZ5up3+JT9VcgOPuOQGHk03NWXMvvSuN6upqGHlg28vhHyw4yCVt1dSXFDMiZ4TlAXKPKc3eo0jm6Yd8YxTPEM2lMIuzBNvPpEIoo50HGHZj5ax6epNKdMKDfMca0mghDuW30GwMOhZP+bz+Zg9fjZLZiwhWBhMTFe8eP3F1LXUUTOuhqeveTrl+sebZWSqRUuuXct15iwTfS+KiEgmCtpETjE7D+3klf2v5HoYQ24oForur4FH7dbatGtTxdfCSq43C7twyjHxYCY+NTAeZC2/dzmNNzZyqOMQV/zvFbx17C06e4Z23bGALzAkNWhe3SibbmpKTCucM3EO+9v2ewZR8aAp2/qxbBp39HeugQZD+VBHJiIi0puCNpFRbDRmzYZjzF4B03WbruPK+Vfyh8Y/0N7dzuzxs3lk9SOsqF2RtlatvwYetVtrufP5O2lsbUxMu4t/vevIrj71ZtkKdYUY/83xg2qfn62pZVM5duLYkHd7jAezycHOPVfek1VANlT1Y/G10oYiozZavo9EROTUoqBNZBQbaNYs1Bni8PHDABw+fphQZ2jEG2xkM+aBBHZeUwnDLkw4HObB7Q8mth3uOEzl7ZWJ6XuHOg5ReVtlytTE6nHV+MyXNnh6/K3HE18f6Djg+fVgDSRgSzcFsT/p6r9KCkrw+/x0dHUMuBauvbud3Ud297lP2QRRQ5nVUrDVP2URRURGLwVtIqeIeDaqvSuaaWlqbeoTtOSL5MCuvwDukg2XJD5TJvEgJ/4Yn9538fqLOWvyWUwpncIfGv8wrNmuoRAMBLnlvbfwxuE3+N6L30vbhbG37kh3xvqvxdMX91uX5qUsUMbciXP7bM8miFKgNbJ0vUVERi8FbSKngJNtsNGfbDNj2WT6eh+z4+AOXj3watrz7Ty086Q6JfZEetKeP58kB1jxIPsDCz7A8nuXE3GRfqc9FvmL+q3/SleXlimIi697JiIiIsOn35b/ZlZtZk+b2Q4z22Zm18e2TzSzx81sV+xxQmy7mdl3zGy3mf3RzJYM94cQkcyyabBxMuKZsZ0Rso0SAAAgAElEQVSHdqY9pk8L+1imb3PdZkKdIdZtWcfVD13N1P+YSmNrIxCtF9t6YCvwdpDX+3zp2uDnk3jHRcOAaPv8bMVf49WyPh6Erb1sLf9w/j9QXOC9aLbPfJxbeS7wdi0Z0KeWLN52fsmMJTTd1JSo26uqqMqqBb+IiIgMj2wybT3ATc65LWZWDrxoZo8DHwOedM7dYmY3AzcDXwQuB+bF/pwPfC/2KCI50l+Djd1HdvfZnm03xv6yZ6HOEOtfWc9Nj91EV6QrsT2e6fuLH/5FdMFlLG23RUidzrlo2iIu/5/Lh6W1/cmK18T5zU9poJSNqzbyL0/9S6Jt/U8/9FO++PgXuW/rfVl1c/T7/Jw1+SzOmnKWZ3AULAwmOlrGM29e0x+zmQLbO0ua3Cjk0jmXZtXxUcYe1cKJiORev0Gbc64ZaI59HTKzHUAlsAK4KHbYeuAZokHbCmCDc84Bz5rZeDObETuPiAyRgTQVmTdpHmWBMs/AzasmqXc3RoBP/fRTzJ80P9GCvXZbLU+/9TQPv/Zwoj197zq5RB1dd3vGWrFs6sjiQd6FP7yQooKirGu5RlK83qyhpYG5E+ey+uzVBAuDXDrn0pTj7lh+Bz9+7cd0d6UP2gYacEH2rfSz1fuX9aHs+Cijh2rhRERyb0A1bWY2G3gH8BwwLR6IOeeazWxq7LBKoD7pZQ2xbSlBm5ldC1wLUFNTM4ihi5y6BtpUZPXC1dz42I2e5+pdk+RV/xa38/BOJnxzAn6fH7/5Od5zPGV/PLC6ZP0lfH3p1/m3X/3bkE9fdLisArbe65INtuMi4LnmWrKBBljpFoUuDZRSWljKgfYDgw64sml/n23mxOuXdWVdRERERp5FE2JZHGgWBH4FfM0597CZHXPOjU/af9Q5N8HMfg58wzm3Obb9SeCfnHMvpjv3Oeec41544YWT+iAiY1XvJh9NrU3M+c85noFLeWF5n6Yi8WmOyVkxh0sJNBZNW8QlGy6h7lgdbd1tHO8+PugAJ9fKAmV8c9k3KS4oZvvB7fzw5R9y9MRRqiuquf786/mnJ/4pq8xe/PpsXLWR+pZ6dh/ZTVVFFT94+Qc0tDZQVV4FBg2tDdSMq+Hpa54ecIDV1tWWyIpVV1Sz/bPbWfOTNaNu7T0RERE5eWb2onPuHK99WWXazCwAPATc65x7OLZ5f3zao5nNAOILFTUA1UkvrwKaBjd0EUluf7+5bjPLNiyjM9zpeWy8qUi8xslr0em4GeUzmFo2lfff935aO1txzo3aQC2Zz3xcs/iaRADVEGpIBEHXvvNa/u1X/+aZRUx+/aSSScwaP8szEPvseZ8dsrF6ZcUUqImIiEhv/WbazMyI1qwdcc7dkLT934HDSY1IJjrn/snMrgCuA5YTbUDyHefceZneQ5k2EW+hzhALvruAhtYGKssr2d++n55IT8bXVBRV8InFn6CyopIvP/VlzwDPMPw+P+FIeEwEahANtsoC/U9RTFdnZxjBwmBerlsnIiIiY1+mTFs2Qdu7gd8Ar0JiMaQvEa1rewCoAeqADzrnjsSCvDuAy4AO4OPOuYwRmYI2yWfZdlEc7Hm3HtjK0eNHCRYG+fmun9PW1UawMMh5lefx09d/mpjOeDI1WaNJaUEpx3sGPj3z8rmX88AHH8hqimJbVxu1W2vZfnA7D+94mPbu9rSZNREREZGRcFJB20hQ0Db6ZLuY8mji9Zm8pheWBkoHnY3pXV/WFe7Kqr5qLIkHn8lt/v3mp8BXwMoFK7l49sVs2r2JN468Qc24Gn72+s/6DeDKAmWsvWxtYlqoiIiIyGhz0jVtIr0l11mNFb0/k1cXxbALE+qKbu/d8CP5dV6ZOa82+mPBQDJjhf5CppROoTHUyMIpC7n+/OvZfWR3Snt8ICX4Wrp+Kc/sfSZjzV3vDpgiIiIiY4mCNjllJWfW1r1/XZ81z2q31abNgsUbfqxauColQKupqGHlgysTgZlhfPqnn2bOxDnsPbY3ZXHp0aK0oJSKogr2t+9PBE1embE/7v8je47twTmXNigt8hdxftX5vHHkDeZPnp9VZuzJa55MTGd8es/TPLTjIbrD3X0yn5rWKCIiImOVpkfKoCy+azGv7H+FRdMW8fLfvZzr4QB9pzd6ZbyARB3Zhlc2cPTEUaaWTqWjp4OOrg4iRDCMQn8hp084nR2HdqR9v/Nnns9L+16ix/UQcZFRU3NW5C/C7/PT0d2R1fHxZQQAarfWembG4noHV109XYk1yAa6UHQ68ffINA4RERGR0UY1bTLk8i1oS+6yOKF4AstOX8ZPX/8pPvPR0d1BWaAsGliZEYlEOBHuf3HmsajIX8SeG/aw+8juPrV6PvPhcImOkicbaLV1tXHx+oupa6kb9DpmIiIiIqcK1bSNIWOxAUg62XZtjNeKtXW1AXD0xFEe3P5gyjHt3e0jMuaRVuAroGZcDXUtdURcJO10zrJAGT7zsenqTUwPTmd6cDpNNzX1yVgBicWeKysq2f7Z7YMOtIKFQZ7/1POD/mwiIiIiEqWgbZTJhwYgoc5Qn/qvk2mBnxycVY+Lrsv++/rf8/BrDycyZYX+Qj7zs8+w9PSlzJkwJ9EW/+jxo2zcsZHuSPeQfLZ8U+QvoivclWj5b5jndMPkFvaHjx9mUukkTh9/Ohg0tDR4TiMMFgY9a8omlkykvrWeiSUTlRkTERERyQMK2mRAEgsTd0UzV02tTVTeVjnoKXTx80VcJGM2rCscbeDx2BuPDW7geciHj4A/wIr5KygJlFBeVM7Pdv6Mtu42goEgH1jwARZMXsDa59by6oFXOXvq2Rw7ccwzC5YuABuM+ZPnpzyKiIiISG4paJOsebXAjxAh1BVi2YZlvHX9WwQLg1kvRO11vlNFwBfg9stu55pF16Rks/7z8v/sc+x//iG6zWc+Lqi+gImHJjJ/8vxhy4KN9Wm3IiIiIqONgjbpV3z64sbtGxMZr946w51U316daPQR78L4mZ99hvef8X627NtCe3c7s8fP5qmPPgXAdZuu40TPqdcQJFgY5NGrH806M5mc+VJAJSIiInLqUdA2igx1LVm690iuL3vjyBt89/nvJtraZxJ2YZI73jscPa6HH+/8cWLboY5DjP/m+IwLJY9WAV+A7kh3SqfKzp5Owi6Mz3z4zc+NF9zIly/88oCyZArURERERE5tCtpGiWxqyUKdIS7ZcAl7ju2hLFDGX5/51yycupDlc5ezafcmzymL8SBt64GtvHboNZ566ynCLtxvgHYyhvPcyXz4iOD9Xn78RIh4Bo7BQJBPvfNT3Pn8nfREosGq3/wU+gvTBmI3vOsGfv76z/t0YtR6YiIiIiJysrRO2ygQ6gxReVulZ+1XwBfgtr+8jaryKj708IfSTjeML/xcWlBK2IW5asFVVJVX8d0XvktPuGdMrVtWM66GlQtWsmDyAqrHVbPygZUp65GVBkrZdPUmgD5rlcX3JXdkVCAmIiIiIsNNi2uPcnf84Q5u+uVNaevJxiqvBZ/T8ZmPsoD3QtBewVc80Mq0T0RERERkpChoGwVCnSHWv7Ke/3vt/2hsbcTMmFY2jQJfAU+89USuhzciCqyAVQtX8Wc1f5aythiQsgZZeVF5op4v1BliUukkFkxeoIBLREREREYtBW05lNzYY96keYn6si3NW3i5+WW6Il1MKJrAr+p+NWYXiM5WeWE5TTc1KfASERERkVNOpqBNjUiGWHLGbNfhXTSEGhKdEuN1ZZkaZIw2hrFq4Soe2fkIfvNnXCA7k2BhkE1Xb1LAJiIiIiLSi4K2LPWevpjcAdHv8zOtbBoAv97767QBWbwmazQGbPGAM/nRzLho1kXcv/L+lNqwqooqMHjzyJs8vONh2rrbCAaCnF91Po/sfITucPdJt8EXERERETlVDNv0SDO7DFgL+IF1zrlb0h2b79Mjf7n7l7z//vefco1Azph4Bp9/1+cT9WVXnHFFn7b2Aw201PhDRERERKSvEa9pMzM/8DrwXqABeB74kHNuu9fx+Ry0/fKNX/KX//OXuR7GgPjxA2A+w29+OsOdFPoLCUei2a10tXMFVsDfnft3LJ62WMGUiIiIiMgIykVN23nAbufcm7EB3A+sADyDtnzQ0d1BfUs99a31icc3j77Jhlc25HpoWSkpKCHiIqxcsJKLZ1+cdk0xgK/+6qvc/tztGEZnuJOyQBk+83m2yxcRERERkdwarkzbSuAy59wnY8//FjjfOXed1/HDnWnrCnfR2NqYEpD1fjx8/PCwvf9Q8Jsf5xwYFPgK6Ap3JRbKTg7Uss2OaZqiiIiIiEj+yEWmzTy2pUSHZnYtcC1ATU3NoN8o4iLsa9uXEoTVtdSlBGT72vZlXJg5mc98zAzOpHpcNUeOH2Hn4Z2DHtvJqKmoYc7EOcwon8H04PTEOmTQN3s2mGArWBhkzZI1Qz1sEREREREZYsMVtDUA1UnPq4Cm5AOcc3cDd0M00+Z1EuccR44f6RuMJT1vDDXSE+nJemBTSqdQPa6a6opqasbVUF1RnXhePa6ameUzKfBFL8u6Leu4/tHr6ejpGNinz4IPH7PGz2Jvy17MGWHCFFgBfp+fn3zoJ1w659K0r1WwJSIiIiJy6hiu6ZEFRBuRLAUaiTYi+bBzbpvX8Wf8yRnuC//9hT4BWUNrA8d7jmf9vuOKxlFVUeUZjMUfiwuKsz5fqDNE5W2VhLpCWb+mt9njZnPahNOYXDqZYyeOUWAFXDH/Cq5ZdA3BwqCmKYqIiIiIyMh3j4y96XLg20Rb/v/AOfe1tMfONMenM5+vuKCYqooqZo2b9XYQFgvEasbVUFVRRUVRxZB+BoDNdZtZfu9ywpEwHT0diXXKaipqKA2UAtEpldOC09IGZiIiIiIiIpnkJGgbCKs0V/OFmj5TFpO/nlQyCTOvUrnhp2yYiIiIiIgMp7wP2vJ5nTYREREREZHhlilo8430YERERERERCR7CtpERERERETymII2ERERERGRPKagTUREREREJI/lRSMSMzsI7M31OPoxGTiU60HIgOm+jU66b6OX7t3opPs2eunejU66b6PTcN+3Wc65KV478iJoGw3M7IV03Vwkf+m+jU66b6OX7t3opPs2eunejU66b6NTLu+bpkeKiIiIiIjkMQVtIiIiIiIieUxBW/buzvUAZFB030Yn3bfRS/dudNJ9G71070Yn3bfRKWf3TTVtIiIiIiIieUyZNhERERERkTymoK0fZnaZme00s91mdnOuxyOpzOwHZnbAzLYmbZtoZo+b2a7Y44TYdjOz78Tu5R/NbEnuRn5qM7NqM3vazHaY2TYzuz62Xfcuj5lZsZn9wcxeid23f4ttP83Mnovdt1ozK4xtL4o93x3bPzuX4z/VmZnfzF4ys5/Fnuu+jQJmtsfMXjWzl83shdg2/azMc2Y23sw2mtlrsb/rLtB9y39mNj/2vRb/02pmN+TDvVPQloGZ+YE7gcuBBcCHzGxBbkclvfw3cFmvbTcDTzrn5gFPxp5D9D7Oi/25FvjeCI1R+uoBbnLOnQW8C/hs7HtL9y6/dQKXOOcWAYuBy8zsXcA3gdtj9+0osCZ2/BrgqHNuLnB77DjJneuBHUnPdd9Gj4udc4uTWo3rZ2X+Wwv8wjl3JrCI6Pee7luec87tjH2vLQbeCXQAPyYP7p2CtszOA3Y75950znUB9wMrcjwmSeKc+zVwpNfmFcD62Nfrgb9K2r7BRT0LjDezGSMzUknmnGt2zm2JfR0i+pdZJbp3eS12/dtiTwOxPw64BNgY2977vsXv50ZgqZnZCA1XkphZFXAFsC723NB9G830szKPmVkFcCFwD4Bzrss5dwzdt9FmKfCGc24veXDvFLRlVgnUJz1viG2T/DbNOff/s3fv8XGXZd7HP/dMZtJkcmqaHpJJ0rTQLRRWai0oKw+HcqhUF9zdYlSeFXlw0QVZEHRB19c+7r7cB1fWAj54wrIKglKsojxSTkJ31y6KtIVqoZSWNp0kkx7SHGaSNKeZ+/ljDp0kM5PJcSbp982rr2R+v9/85p6529Cr13VfdwtEggNgQfS45jMHRUuv3g28guYu50VL7F4HjgIvAO8AHdbawegliXMTn7fo+U5g3vSOWKLuA/4eCEcfz0PzNlNY4HljzA5jzI3RY/pZmduWAseAH0RLkjcaYzxo3maajwI/iX6f9blT0JZesn9ZVLvNmUvzmWOMMUXAz4DbrLWBdJcmOaa5ywJrbShaNlJNpBrhzGSXRb9q3nKAMeZDwFFr7Y7Ew0ku1bzlpvdba1cRKcO62RhzYZprNXe5IQ9YBXzHWvtuoJuT5XTJaN5yTHSN71XAT0e7NMmxKZk7BW3pNQE1CY+rAX+WxiKZOxJLTUe/Ho0e13zmEGOMi0jA9pi19ufRw5q7GSJa6vMfRNYklhlj8qKnEucmPm/R86WMLGeWqfd+4CpjTAORMv81RDJvmrcZwFrrj349SmRtzXnoZ2WuawKarLWvRB9vJhLEad5mjiuBndbaI9HHWZ87BW3pvQosi3bYchNJkz6V5THJ6J4Crot+fx3wy4Tjn4h2+nkf0BlLdcv0iq6PeQjYY63dkHBKc5fDjDHzjTFl0e8LgMuIrEfcCqyPXjZ83mLzuR54yWpz0Glnrf2itbbaWltH5P9jL1lrr0XzlvOMMR5jTHHse+AKYDf6WZnTrLWHgUZjzPLooUuBN9G8zSQf42RpJOTA3Glz7VEYY9YR+RdJJ/Dv1tp/yfKQJIEx5ifAxUAFcAT438AvgCeAWsAHXGOtbYsGCg8Q6TbZA1xvrd2ejXGf6owxFwC/Af7IyTU2XyKyrk1zl6OMMe8isgDbSeQf/Z6w1v6zMWYpkQxOOfAa8D+ttX3GmDnAj4isWWwDPmqtPZCd0QuAMeZi4PPW2g9p3nJfdI6ejD7MA35srf0XY8w89LMypxljVhJp/OMGDgDXE/25ieYtpxljComsU1tqre2MHsv6nzkFbSIiIiIiIjlM5ZEiIiIiIiI5TEGbiIiIiIhIDlPQJiIiIiIiksMUtImIiIiIiOQwBW0iIiIiIiI5TEGbiIjIGBhjbou2hBYREZkWavkvIiIyBsaYBmC1tbY122MREZFTQ04EbRUVFbauri7bwxAREREREcmKHTt2tFpr5yc7lzfdg0mmrq6O7du18buIiIiIiJyajDGHUp3TmjYREREREZEcpqBNREREREQkh+VEeaSIiIiIiMhUqN9cz97WvSyvWM6m9ZuyPZxxUdAmIiIiIiKz1t7Wvew6sivbw5gQlUeKiIiIiIjkMAVtIiIiIiIyqeo317Pyuyup31yf1XEE+4IcP3EcgOMnjhPsC2Z1POOloE1ERERERCZVrCRxb+verI1hm28b3g1e/AE/AP6AH+8GL9t827I2pvFS0CYiIiIiIrNKsC/IusfWEewPEiYMQJgwwf7I8a7+riyPcGwUtImIiIiIyKyy6Y1NhG046bmwDbNpd/a7SFprGQgN0DPQQ9uJtrTXqnukiIiIiIjMKvuO76N7oDvpue6Bbva37Z/yMYTCIQbDg/SH+jnafZSGjgZ8nT58AR9NgSb8QT8twRZaulo42n007b0UtImIiIiIyKRJ1vyjOL94WsewbN4yPC5P0sDN4/JwevnpE7q/tZbB8CCD4UE6+zo51HEoEpB1+mgMNNIcbMYf9OMP+jncdZjewd4JvZ6x1k7oBpNh9erVdvv27dkehoiIiIiITMA23zbWPbaO7v5uwoRx4MDj9rDl2i1cUHvBtI0j2BfEu8FLsH9kt8hidzH+O/wUuYtSPj+WJesd7KUp0MShzpNBWXOwmeZAMy1dLbQEW2jvbc94XAV5BVQWV1JZVElVcRXVJdXUlNRQU1LDNWdfs8NauzrZ85RpExERERGRCUts/hGT2PxjtEBpMhXnF7Pl2i1JA8inP/40+c58evp7aD3RyqGOQzR0NNAUaKIx0EhToCkekB3pPsJgeDCj13QYBws8C6gsqqSyuHJIQFZTWsOSsiXML5yPy+nC5XThMJm3F1HQJiIiIiIiE5ZJ848bVt0w5eMI2zADoQFWVKxgy7Vb+PDjH+b4ieMUugu54rQr+IeX/oGWrhb8QT89Az0Z37c0vzSeJfMWe6kuqaa6pJrFZYupLa2ltrSWgrwC8hx5OB3OSX1PGQVtxpgyYCNwNmCB/wXsBTYBdUAD8BFrbbsxxgD3A+uAHuCT1tqdkzpqERERERHJKdPV/GMwPEjfYB8twRYaOhviZYtNgaZ4g4/DXYc51nNsyPO6+rv42Z6fJb2ny+E6GZCVePEWe6kpqYkHY3VldcwtmEueI488x/TnvTJ9xfuBZ621640xbqAQ+BLworX2a8aYu4C7gDuBK4Fl0V/vBb4T/SoiIiIiIgnqN9ezt3UvyyuWs2l99tvQT8RkNP+IZcnaT7TT0NkQWUvW4YuXLjYHmzncdZiWYAsD4YExjW9O3hwurrt4SEBWU1rD4tLFVBVX4Xa6yXPkEclB5ZZRgzZjTAlwIfBJAGttP9BvjLkauDh62cPAfxAJ2q4GHrGRDie/M8aUGWMqrbUtkz56ERERkVlgNv3FXcZmb+tedh3ZNan3zNbvp/qz6rn9uduTnnMYB/Vn1zMYHqSnv4fGQGOkBX7AFw/KYh0XW7paCPQFMn7dQlch3mJvPEsWW0tWW1pL3dw6vrL1KxzoOMAZ885g0zUz889XJpm2pcAx4AfGmHOAHcCtwMJYIGatbTHGLIhe7wUaE57fFD2moE1EREQkian4i7ucurL1+8nj9vCL+l9w1U+uonvwZLbN5XBx9oKzufAHF+IP+jnWcyzl2rfh8hx5LPQspLL45DqyWGOPurI6FpcuZl7BPNx57pSNPZ786JOT8v6yKZOgLQ9YBdxirX3FGHM/kVLIVJLlE0fsK2CMuRG4EaC2tjaDYYiIiIiISLYMhgcJ9gU52HGQQx2HONR5KFK22Nk4JEs2fE+ygfAAv236bdJ7zp0zl6riKqqKq/CWRMsWS2qpLYusI/MWe5mTN2fSG3vMNJkEbU1Ak7X2lejjzUSCtiOxskdjTCVwNOH6moTnVwP+4Te11j4IPAiRfdrGOX4RERERmeFO1fLQqdiEerz3tNbSO9hLc7CZg+0Hh2wUHWvu4Q/6x7QnGYDb6ea8qvOoKqmKB2Q1pTUsLltMXVkdJfklWWnsMdOM+glZaw8bYxqNMcuttXuBS4E3o7+uA74W/frL6FOeAj5rjHmcSAOSTq1nExEREZFUTsXy0MRNqAH8AT/eDd4JbUKd7p7v9b6X4yeOxwOyQ52HaOxsHLKWbLx7knlLvFQXV1NTGllHtrh0MZ/+1afZ07qHMyvO5Df/6zfjej9yUqZh7S3AY9HOkQeA6wEH8IQx5gbAB1wTvXYLkXb/+4m0/L9+UkcsIiIiIjKDTfYm1NZaWoItrP3RWnoGT+47FrvnhT+4kEJXYcp2/MmU5JdQVVRFVUlVZB1Zcc2QDFlNSQ0FroKU68jcTnfGryWjyyhos9a+DqxOcurSJNda4OYJjktERERETgFTUSKY68a6CfVAaAB/0E9DRwMNHQ00Bhrj+5LFyhaH70mWyGKHBGyxPcm8xZH9yKpLqyPryEprI0FZaR3lheUpAzKZfiogFREREZGsmIoSwZlgtE2ov7f9ezz7zrM0B5rj+5L1h/on9JofXPZBvvBnX6CurI6q4ipcTteE7ifTS0GbiIiIiEy7yS4RzFV9g334On2RjovRjaJ/4/sNDuNImW17teVVXm15Nem5QldhpNNi8dD9yN5qfYsHdzxIb6h3xHM8Lg9/ccZfcFHdRZP63mT6KGgTERERkWk31hLBiQr2BVnzyBp8HT5qy2p56RMvTbgMMxwOc7T7KAc6DnCo41B8o+jGQCP+oJ/mYDNHu49mvCcZgMGwqnIVNaU18YCstjTS/n5J2RLmFczD4RhZthjsC/KD13+QNGiLbWw9nZZXLB/yVSZGQZuIiIhIFp2Ka7pg9BLB/W37k54bT/CVWIYZJkxrT2tGZZhdfV00dDRwsGNox8XmYHO84+LwPcnSie1J5i3x4nK4eOGdF+gPnyx79OR5ePavnx1XaWhxfjFbrt0y5H06cOBxe9hy7ZZpz1qeSls3TAcFbSIiIiJZkstruia6d9poz182bxkelydp4OZxeTi9/PQRx8cTfKUrw1z7o7Vs/shmjnQfiexJ1tlIY+DkRtFtJ9oyfr9z8uakbH9fV1ZHXVndiOCyq7+LFd9aQWOgkZqSGt68+c0JBVcX1F6A/w5//J7eEu+E7ym5QUGbiIiISBZkc01XJgHZRPdOG+359WfVc/tztyc9l6ycb6yfV1tPGwc7DrJx58aU2bCewR7W/XjdqO/FYFjoWUhlcSU1pTVUF1fHA7LFZYtZMncJCz0LMcaMeq9ERe4iygvKaQw0Ul5QPinzPRX3lOxT0CYiIiKSBdO9pitRLmxmPdZyvnSfV+9gL5f/6HIK8goipYuB5nHtSeYtiTT3iK8jK61jw+820Bxo5oz5Z8yYkj+tJ5t9FLSJiIiIZMF413RN1GSuoZtoCWWycr43bnqDrv4uXva9TENnA4c6DtEYaOT5d55P+XkNhAf4XdPvkp5zGidhG8ZiR5yb45zD1y//Ore895aUY/z8C5/nD0f/MOYsWqamIsCaKcGlZE5Bm4iIiEgWjGdN10RN5hq6YF+QlxtfpinQNKbgr6u/i4PtByMt8DsO4ev0EeyLlDwe7jpMxT0VY96TzGD40wV/yvk1558sWyyNlC0WuYuoubdmSFlljMvp4vp3Xz+m15psCrAkEwraRERERLJgrGu6Mr5viuzXZK6hSxX8PfWxp6grq+ONo2/g6/QBsK9tH5c+fCmHuw7THGyms68z5X0HwgNDHntcnvieZIuKFvHzPT8f0m0xpshdxH/f8N8pxz/erv3fSi4AACAASURBVIqnamdPyT0K2kREROSUMtGSvskyVS3aU61Xm+gaOmstbSfaeOPoG6x9dO2Q/cBiwd8lD18y4nk9Az281PDSiON5jjwWFS3CW+ylKdBEX6iP6uJqvnzhl6krq2Pp3KWUzSkbUpY4vHtkpp9XrAzzkocvwdfpo7a0lq3XbU37nFzu7CmnHgVtIiIickrJhSYcMdPZon20NXR7ju1hb+teDnYcpKGjgZauFgD2t+1n2TeX0Rxs5sTgiXG/vtvh5oF1D7C8YjlLy5ZSWVyJ0+Ec0z0m8nkVuYt49W9ezeh1stnZUyQZBW0iIiIiWZRpi/ZMMoSpyvnCNsx8z3zynfn0hfqSPvcbv/sG3/jdN0Yc7x7oZn975k1RnDgJERpx3OV04TAOLlx8Ycb3SqbIXcT5NedT3lrO8orlUxI8ZbOzp0gyCtpEREREctxoTT86ezt58q0n+dtf/W28bLEp0ETZv5ZRUVhB+4n2EevFUjFEyhETuy26HW6+fOGXufL0K9nWuI1/ePEf6BnsGfHcPJPHoB1Met/J7Ig51WWt2ersKZKKgjYRERE5ZUx2Y4npWB+39eBW/vwnfx4PIpoCTZR/vZxzFp5DV38X/qA/aWdEiGSFjnYfTXt/t8PNHX92B1ecdgXzC+fzvo3vo2uga8g1/eF+7nn5Hj53/udYXrGcf9z6j0nv5XQ4yXfkT2tHzKmQjc6eIuk4sj0AERERkUT1m+tZ+d2V1G8eX/fEVLb5tuHd4MUf8AMnG0ts820b1/1i2a9dR3bxcuPL8bb1Y2Gt5Wj3UXoGIlmro91HuWXLLVz9k6t5z4PvYcE9C1jzyJoRwcNgeJAdLTvYe3xvyoAtxu1wc8u5t7D7b3cTuCtA8ItBakpqAKgpqeH4ncf5P5f+Hy6uu5jfNv026X5mcLIsMNZApdhdjCP6V0kHDordxTz1sadwmOR/vZxIR8zpVn9W/ax4HzJ7KNMmIiIiOWUqGoVMdmOJTDsLtgRb2LhzI3848gfy8/KZXzifI91HaAo00RhopCXYMmSNWUtXCw+8+kBGY8gzeVy9/GquOesafvX2r3j0j48mva4/3I/H7eGsBWfFj6VaE5ZpWWC6boxT0RFzuk1VZ0+R8VLQJiIiMstMVslerrTGnwyT2VgiXQC45uE1XLT4Ilq6WjjUeYiu/q40dxop35nP+2vfT01JDbWltbza/CrPvvNs0msH7SDL5i2j/ux6gv1BnnzryYzL+VLN51jKAlN1Y5zOjphTaba8D5kdFLSJiIjMUKmCqsnKVGWjNf5krDlL9rmMtbGEtZb23nZ8nT4a2hs41HmIhs4GfJ0+Xmt5LWUwNhAe4NcHf512fHkmj8+s/gzL5i2jrqyOurI6rv3Ztew+tpszKs7gxU+8GL92486N/Mb3m1GDqMnaqHuy7pNpR8xcN1veh8x8CtpERERmqGRB1WQ32phOk7GZcaoui+kySG6Hm91Hd/OJJz9BU6CJpkATB9oPELIj29ZnYuXClSyZu4Sn336a/nD/iPP5efmsXLRySGYv1X5lmQZRk1XOp7LAkZZXLB/yVSQbMg7ajDFOYDvQbK39kDFmCfA4UA7sBP7aWttvjMkHHgHeAxwH6q21DZM+chERERliMoKebMl0zVm6ks3Y+49lwZoDzSy4ZwHXrbyOUDhE72Bv0tfuD/fzq32/Sjm2OXlz8BZ7qS6ppqa0huM9x3nxwItJAzKPy8Nnz/ssbx9/O+l5GFvL+LEEUZNVzqeywKFmemmwzA5jybTdCuwBSqKP/xW411r7uDHmu8ANwHeiX9uttacbYz4avU4tdkRERKbQZDfamIzxrHlkDb4OH7Vltbz0iZdGZPwSA7DLl16e0ZqzWHbxxOAJtuzbQkNHAw0dDbzT9g6/2PuLIfewWHpDvXxvx/fSjnVewTyWzl1KTUkNDuPgqb1PxQMugyHP5PHDD/8wHvgG+4J4N3jp7x8ZlMWyX4/vfnzSWsaPJYiarHI+lQWK5JaMgjZjTDXwQeBfgNuNMQZYA3w8esnDwFeIBG1XR78H2Aw8YIwx1trk/WNFRERkwjJptPH8gedHZKmmotlIYsYvTJjWntYRGb/hZYxVRVVp15x99Tdf5Ru//QZ7j+8F4O3jb/PBH38wo/EYDCvmr+D86vNZVLSIb2//Nm0n2qgsquSNm95gbsHc+Ji8G7xDMmQWS9dA15DAN5Ps11jXho1WgjcTgyiVFYpMnkwzbfcBfw/E/olsHtBhbXzL+ybAG/3eCzQCWGsHjTGd0etbE29ojLkRuBGgtrZ2vOMXEZEcMJu6DMLMfD+jNdp44+gbI9Z6AUnXf2Uq2eeULuN35aNX8oe//QMvHXyJzz7z2Xi5YlOgiftfuT/tazV0NIw45nK4qCquiuw1duI4e1r3JH2uxfLnf/Ln3H3Z3QD8v7f/H20n2ljgWRAP2GBsHSZHy34lBnY9Az2EbAincVLoKky6NiyT32czLQiaKX92RGaCUYM2Y8yHgKPW2h3GmItjh5NcajM4d/KAtQ8CDwKsXr1aWTgRkRksG10Gp9JMfD/pGm3Mcc7hOzu+Q/9gJIPkD/hZ+G8LMcbQO9AbP5aYDcukocnwz6m7v5v7fncf/aHka7m6BrpY+s2lSc+l2tAZwGmcrFu2juf2PzekbDHfmc+jf/koF9RewMadG7nt2dsyKkdMFfyMtcPkaNmvWGC3afcm9rft5/Ty06k/u37cWbJMg6CZFtyJyOgyybS9H7jKGLMOmENkTdt9QJkxJi+abasG/NHrm4AaoMkYkweUAm2TPnIREZFT2PCgat3p67jdJC/H6w0NbcARJsyJwRMjjsXWv23+yGbWP7F+SEOTqm9U8cMP/5BFRYtoDDTS0NFAU6AJgL3H91L+r+W097ZP6D3lmTyMMQyGB7HYeMlhbDzpyhbHUo44GXuUxWRS1pjp/m+TRRkukdnHMdoF1tovWmurrbV1wEeBl6y11wJbgfXRy64Dfhn9/qnoY6LnX9J6NhERkZPqN9ez8rsrqd+cWZ+uYF+Qc79/LgvvWci53z+X5/c/j3eDF38g8u+l/oCfM751Bl+77GsUu4txRP/37sBBvjOfOXlzMh5b32AfH3zsgwT7g4SJlAqGCdM10MX6n67ngh9cwMd+9jG++OIX40Fj72BvRgFbvjOf91e/P+X5QTvILefdQnVJNQDeEi/+O/z4On2jli3GyhGHv/9id3HGrerrz6rHYZL/1SjVHmWb1m/i9c+8rkBJRKbUqEFbGncSaUqyn8iatYeixx8C5kWP3w7cNbEhioiIzB6xBhy7juzi5caX42vLUtnm24Z3g5ed/p0c7TnKDv8O1j62dkRQFewPctev7+LtW95mVdUqFngWsKpqFTede1PKVvfJ9If7GYwvWU8uMTCKmeOcw3c++B323LSHIlfyAMntdPPRP/0oHpcn6XmPy8OK+SsoLygHiJccZlq2GCtHTHz//jv8GW93MBmBn4jIVBhT0Gat/Q9r7Yei3x+w1p5nrT3dWnuNtbYverw3+vj06PkDUzFwERGR6TTW7FgysQAsMUPm3eBlm29b0usTm3rEArR0a7/6Q/383TN/x/uq38f51ecTCof4/s7vj3u8yXx29WcB4uOJ6Q318vcv/D3VpdU88z+fodhdjNNENox2Gmc88LnunOvGnM2KlS0mM7xsschdxKt/8ypHPn+EV//m1XHvUeYtifRXi2X7cn2fOxGZ3cayT5uIiMi0ykYXx1QNOFI1J8l0jJnso3bDUzcMudemNzYRCocyHntfqI+fvvnTMbzbkdwONy6nK+W6rkB/IKMOi+kacGS6WXTMWNvnT1SRu4jza86nvLWc5RXLlWETkaxT0CYiIjlrurs4Ju4vBiczYZs/sjllJ8VMx5iunfxAaIDbnr2N/zr0XxzuOsyB9gOc/e2z2de2L2UnxlQqiyp518J3UVtay+LSxSwuW0x7bztf/PUX6RnoiTf4yM/Lj3ePTAycYk0/knEYB/MK52VUqpiuAcdo7fKHN/fIZF+0yaY1aupCKZJLFLSJiIiQPhO29tG18TVOw1vjp2Kt5cObPsyeY3tY4FnAnLw5KYOd3lAvD732UPxxsD/IG8feGPN7KHYX8/YtbycNYq5fef2IIAlIGjilC5Dean1rzB0Wk0nXLj9ZwDRaoCeTT4GrSO5Q0CYikmXTWQI4Fa+VyX5euXTf4a+x5pE1+Dp8FLgKCNnUpYjDm35c+eiVPH3t0zQHmwE40H6Aa392LS1dLfg6fTQHm+MNQPa17Rvz2PKd+Xzhz77Aht9uoGewJ+k1DhwZZ51SBUnJygDTBUjnLDxn0koVx5rJGW1fNBGR2UpBm4hIlk1nCeBkv1aqcsLELNR4AsVM7juZYx/eVCMTXQNdXPTDi+KPg/1Bfrz7x0mvNRiWzl1KQ0dD2sAwUV+oj/tfuZ8nP/pkfM+04WWMn3rqUxPOOqWak1QB0mSWKiqTIyKSmYm0/BcRkVNYss6GiY01uvq7xtzePtP7jiax0+PwPc4CvQF8nT7W/mho2/zJ4nK4uPncmynIK8Bg4sePdh/lm1d+c0z7qIVtmMbOxqRt7K847YoRrfFHs7xiOecsPGfCa5TUYVFEZHop0yYiMg7Z6GqYa9I11gjbMF/9z6/y7e3fHnO2bLT7xroTptI32McfDv+Bt46/RUtXCxX3VMSbeRztOUrpv5Zm+hbHxeVw8dBrDw3ZG81ih+yjdt73z4tnyNavWM+9v7s36b1ijT1ibewnajJ/r2arVFHNMUTkVKSgTURkHKa7q+FUSlzXVVtWy0ufeCmjtWOjbXh87yv3Dul8OFp7+1gg3DvYm/a+rx9+nd83/x5fp499x/fx4sEX8XX66A/10zPQw7GeY/Hrj3YfHeOnMXE9gz24HK6k58I2zNNvPz0k2Fkxf8W4G3tMdQCTiwHSqfqPJCJyalPQJiJyikjW2GPXkV1D1ia19rRmvHYstuFxsmDDYBgMDyZ9Xixblhj4xsoomwJNlLhLyHfm0xfqS/r8B159gAdefWAsb32EPEceDhz0h0e20y/MKyQ/L5/23nZqSmrYeNXGEWvKXE5XpF1+QjYtfm+Tx0B4IOnrxjJnicHQRPYgm+oARgGSiEhuUNAmIjIDjTU7lqyxR9U3qggTpmfgZGfCWDbskocv4cyKM1lWvowrl13JvuP7WDZvGfVn1cdfJ12wYbFYa5Oe6x7oZvObmznYcRCAPcf2UPq1UiyR6wP9gbF/IGM0GB7E7XQnPed0OKkuqaa9t53ygnKuOO2KEZ0UX/nUKyx/YDm9jAzanA4n+Y78tJmzuy+7e8jx6d6DTEREZhY1IhERmWG2+bbh3eBlp38nR3uOstO/E+8GL9t825Jen6qxR9dA15CALVE4HOaPR//IL/b+gtuevY2vv/x1bnv2NrwbvDz/zvNs3LmRr/7XV/nb1X9LkbtoSMONTDz7zrME+iLBWX+4Px6wpZPvzOeey+9h7WlrqSquwmmcY3rNRB6Xh8+993MjmoIUuyOdER1m6P8eY+u3INL0o7K4ki3Xbkn6/Kc+9tSI58ekypzN1MYek9XYRERE0lOmTURkBkm3AfS6x9Zx+WmX807bO0MapKRr7JFKPLiz4XjGKPZ17aNrcTlcDIQHcBgH1tqMgq6xGl6m+ObNb2Kt5f5X7scf9E/o3g7j4MsXfZkvX/TljDdrHr6+K91eZuPJnM3EPchUPikiMj0UtImIzACxcsi9rXtTNukIhUNsPbiV9t52jp84jj/gZ8v+LXx/x/dTPme8Ymu2xhIMOnCQ58jDGJNyvVqinsEeaktrqS2tZXnFcl4//PqQEs/xcBonha7CIcFTss2lkzXgSBagpAq00gV06eRi4w8REck+BW0iMqtNRWv+ZA09Mum2mPj8TW9sYt/xfdSU1tDa05r2XpluAt0z2EPPYKTcsTnQjPdeLwYzJVmwsXA73Vxadyl/teKveOPYGynb2w/ncXn4/J99nhtW3UCwL4h3g3dIhnEs8p353HzezayoWEH92fVDgqdkvy8m4/fKeDJnylyJiEgyCtpEZFbLtDV/psFdsoYe6botDr9v7Pk9Az2EbGjItcnulawcMhOxQC3bARtAf6ifcxadww2rbmDjzo0pO04Ol7j+K5MST4/LQ9iGI10dB3pHlCVO1fqwdNkxZc5ERGQyKGgTkSmRC5tPjyUjliq4S+zS6C3xsq9tH139XfHzievJ9n52L1c9ftWQjo7D29qnC8Bi97ry0Su5+7K78XX4aOhoGLLX2WRyO9zcdv5t/Fn1n/GZX32Gw92Hx32vdGvbEvcaS9dxMn6vJOu/0u0JB/A+7/v41KpPxYO8sZYlTkS639/KnImIyGRQ0CYiUyLbm0+PlhFLLFFcNm9Z0izO8LLEYz3HUmauBkIDLP3mUvoH++P7nVV9o4pCdyEQCRof3vVwRmvAuga6uOWZWybw7kcXa+wRC2Z+vPvHvNz4Mi3BlhEZwBi3w510XzOINA3BMCSgjUnMmBXnFydt0lHoKqS6pJq23jZqS2vZet3WIYFWuj3hPC4Pn1r1KW5YdUP82Exr6CEiIpKOgjYRyVnDs3XDA63EPcMSjdZhcfNHNrP+ifXxEkWnccaDqVhGDhhxj3Slhr2hoft1xVrqdw1EgpjmQDO3PnNr2jVpY+U0Tt5T+R68JV6e2/9cfD1bOg4chAmPCGZin2+qdWMuh4sNH9jAsvJlIz67WGMPIKOOieNp0jHWDahVligiIrOJgjYRmXQTbdQRk5iti2W9Yi3oPS4Ptz93e9K1SunWP4XCIa5+/Gp6B08GWYmZpeZAM/Pvmc85i86Z1LJEy+S3xS90FfLidS9ircW7wZvZkwykGkaqLNjwNWH+O/xs2r2J/W37Ob389CGNPTINxorcRUm7NqYy2tiGP19liSIiMpsYa9P/JcIYUwM8AiwCwsCD1tr7jTHlwCagDmgAPmKtbTfGGOB+YB3QA3zSWrsz3WusXr3abt++fYJvReSkXFhPdaoaXlI4kUYQK7+7kl1HdlGaXxrZGDpJIFbsLsZ/h3/IX9rvfOFOvv7y1yf8XnJVss90+Oce6xrpMA7CNhzPiJ1bdS7HTxxP+2ejq78rHngNL6PMRGzezll4Dq9/5vVJec+TNTYREZFcZYzZYa1dnexcJpm2QeAOa+1OY0wxsMMY8wLwSeBFa+3XjDF3AXcBdwJXAsuiv94LfCf6VWTaZHs91WyX2Jwj1nCjOL941LLExOBqtFLHxGxdV39XysxZ90A3Vd+oYnnF8vg40q1/yhXJWvGna8+fuNF0sgzW8JLD6pJqfv83v+fpt59OmhFLJ5c3eR5rhk5ERGQ2GDVos9a2AC3R74PGmD2AF7gauDh62cPAfxAJ2q4GHrGRFN7vjDFlxpjK6H1EZixl7yKGZ3Rae1rjDT7ean0rZXAV7A9Sd18df/2uv8btdPPt7d/mxMCJ+Lqo25+7nc0f2Yyv08fWg1v5+Vs/p28wsgFzqsYYENncOdgfZKd/J1XfqOKJa56grrSOnoHR13dlkzGGioIKakpqONJ9hOZgM1XFVXT2dSZt5uF0OLlkySW80/ZOymBleECzqGjRkOYcY5HLa8JO5T9/IiJyahrTmjZjTB3wbuAVYGEsELPWthhjFkQv8wKNCU9rih5T0CYz2qmYvUvWCCRdJu1Tqz6VNrt1/MRx7nvlvhHHQzZEsD/I2kfX4jTOtEFaKrHGH+t+vG7Mz82GgrwC7r70bm5YdQMrv7uS5mAzFYUVPL7+8SH7uCU2+sikvHSyApqJ3CeXAz4REZGZKOOgzRhTBPwMuM1aG4gsXUt+aZJjI+p9jDE3AjcC1NbWZjoMEZkmwb4gLze+TFOgKd5MJF2Dj7ANc7zneLw74XiNJ2DLRX9S/icc7DjIQHgg6fnugW72t+0fcTxW5piq0cdMoEyYiIjI5MooaDPGuIgEbI9Za38ePXwkVvZojKkEjkaPNwE1CU+vBvzD72mtfRB4ECKNSMY5fjnFqERx8qT7LIfvcdYUaKLsX8uoLqlOmUnrHuim2F086R0Sp4OJ/lvTeMZuMDiMI2lW7IHfP8DfPfN3o244PVyRu2jcZY0iIiIy+4watEW7QT4E7LHWbkg49RRwHfC16NdfJhz/rDHmcSINSDq1nk0my6lYojhRyRp+ACOyaLEmIMlKICGSSfN1+tI2y/jW9m9N7ZtJMN4yyuGK3cW8fcvbnPf982gMNGKi/yVu+hwmnHSNnMM4+NCyD3HV8quSZsWuO+c6Pvfc5xgMDyZ97vC9xURERESSySTT9n7gr4E/GmNivZu/RCRYe8IYcwPgA66JnttCpN3/fiIt/6+f1BGLnCIy3Ug6neF7m7mdbj7zq8/gdDgZCEXK9vwBf7yRyAW1F6QtgYTxZaMmW2FeJJAKDWYetM1xzgEDA6GBEVmxRUWL4t0Sz15wNh29HUP2GXv98Ovj2sagOL+YrddtzXhvMREREZFkMukeuY3k69QALk1yvQVunuC4RHJKus2iU7W/n4hYsJXYjCLVRtKxMSTLpg3PmMU2iw6FTgY7sUYi/+MH/4PyOeXk5+XndKt8iHRSfPIjT7L+ifXxYGg0LqeLt295e9QW+A7jGNFSPrbO7JKHL8HX6aO2tJat123NKOiayHNFREREYIzdI0Vmu2TrvIav70rMTAEp29/HgquxrsNLVp4Y66645uE1nL3gbAbDg5w5/0w2rd+UNJt209M3sWbJmqRleem09baN6frxchgHBc4CToROpMzqJSt/HJ7higVDe1v30j3QnfJec/LmxDNqmawVSzZPRe4iXv2bVzN4dyON9lx1WxQREZF0FLTJjJEu2zWW65IFUfWb69lzbA/72/ZzYvAEvk4fn3v2c5xWfhpffPGLQ/bNimWmrnz0SjAkPRfbSNpaG187lnhPgMbOxqRljw/vejieERtuIDzAa4dfA+Bw12H2HN3D2h+tpWfw5Hqr2HOfe+e5MXy6U6cwr5AFngV0D3TjcXn4yxV/yYqKFZwYPMFdv74raVYvz+Rx7buuZWfLTv549I+4HC4GwgMjNpWOBUN3vnAnX3/56ynHcNPqm0Ztl5/NwElNdURERCQdBW0yI6TLdiX+ZTyT65I1M9nZsnNI+/X23nbue+U+8hx5KbNV/aH+tO3c//Glf2TjaxvjQV3snok8Ls+Qssdtvm3c/tztKe+b6FjPMVZ8Z8Wo12Wb0+Hkjzf9cUQ54J0v3JmyDHPQDlJZVMmZ88/EYRy0nWijvKA85abSy+Ytw+PyJL2f0zhZMX/0z0mBk4iIiOQqBW2S80bb0Pny0y7nnbZ3OG3uabxw4IWU1/nv8Cf9C78/4OedtneSvna68sL+cPJsGEQ6Ld77yr2jvrdYkHHRDy/iE+/6BD/c9cNRnzOTFLmLUjbcSBdoxdrh333Z3Rm9Tv1Z9dz+3O1JzxW6CtWlUURERGY0R7YHIDKa0TZ0frX5VXYd2cXv/b9Pe92m3ZtGlE4+v/95ln5zadY7IoZtOCcDNodxxPcwG3EOB8XuYr617lsUu4txGueQ8ws8C2i5oyVlWWL9WfU4TPIfQWNth1+cX8yWa7dQ7C7GEf2xFhufujSKiIjITKegTXLevuP70m7o3NHbAUBnb2fa67Y2bMW7wYs/ENnrvSnQxNrH1tIX6puageewwrzCeHCTjsflwX+Hn5qSGgC8xV68xd7I9yVe/Hf4uencm/Df4ed7H/oed73/rvi1lUWVaYOlxEArFvA5jXPcgVasMYm3ZOj4RlvLJiIiIpLrVB4pWTfafmTpyujgZCOQxIYgw7kcLja/ufmUDNASuZ1uLq27lL9a8VfUn10f33+sq79rSLZxTt4cXA7XiD3MKgorAGgONlNeUD6kIUisK+Mz+5+hMdCY0Xhigdam3ZvStuHPVJG7aES7fhEREZGZTkGbjLkl/UQMD9BqS2pZ/9P18Xb1scYcmz+yGV+nj33H9zHfMz/t2rJYsJGuxDGTxh6ngnxnPk985Il4MJPYNv9Qx6Eh3R1TBU+jdVkcaxfGxIBvMqihiIiIiMw2CtpOccG+YLwlfao2+qNlwpJdu/vobtpPtFNeUM5ZC85i3enruP+V+7nvd/cxaAdTrj2LZdPWPro2bXbtVDInbw6//OgvufWZW9l7fO+o6+/ynfl8+IwP88u9v6R/sJ8w4SH7mw0PxDLZfywxEBstKFLQJCIiIjK5jLXZbcAAsHr1art9+/ZsD+OUk9gef/hf7GPrgGLX9Az0ELIhnMZJQV4BN513E32DfUMCs1jWbCA0QG+oN/46sT22BCoKKyjMK+Rw9+GUe7ElchgHzbc3s6hoERApAV3xrRU0BhpZULiAnsEeevp7ks5f4rU1JTVD9jcTERERkdxijNlhrV2d7JwybbNEsC/ImkfW4OvwUVtWy0ufeClpxix2jbfEy762fSk3jb77srvZ37af7+34Hr2DJwOwkA3RNdDF1/976EbGDhyESZ49m0kBm8HgcXkIE6ZnoGf0J0StqlzF0x9/mqfffpqtDVv52Z6fJc1yXVB7AcG+IN4N3rRBW+JzYgEbjFyz9dBVD3HJw5fg6/RRW1rL1uu2DllnFluLlrj+TERERERmFmXaZoGxZMxi1xhM2jI7h3GkLGGczVZVruI/P/mf8QYdiWvtwjaMMYbegd6Un3NMV39XymAKks9ZoauQ6pJq2nrbkj5nPKZzvaKIiIiIjF+6TJuCthyW7i/ciWvHhmfDYordxfjv8GOtxbvBO2TTaTnJaZwUugpHBF9d/V0juhoCaYOxsVD5ooiIiIjEqDxyhkhs+FFTWsN/+/6b5mDziAYhz+9/nqs3XU0oHEpbehgKh7j56ZtpDjZzYuDEdL2NGWNR0SI+ec4nU7aZT9XVcLSmHZlSe3oRERERyYQybTlieMOPRIlleK3dxE3VcQAAIABJREFUrfzFE3+R8X3TrTWbiZzGidvpxhhD32AfIRuKl3LGvjqNE2MMZXPKaDvRlrTM0+PycP8H7p/UVvMiIiIiIuOlTFuOyqTEEU42CLn4BxcTIpT0mlRmQsCWbn2dy+Hia5d9jfy8fJo6m4aUKSaWLn7wTz7I028/PaSUMV1ZqMM44vcREREREcllyrRNseH7lhW5i2g/0c6+tn28dvg1nA5nRq3fZ6I8k8dVZ1zFln1bcOCgZ7CHPJOH0+Fk0/pN3PLMLfH1XBuv2sj6J4Zusu0wjqRNPsYikyYtIiIiIiLZpkzbNEoM0t448gZbD20lbMMpM0mh0NgyZ7nC7XQTDoe57LTLKM0v5Zd7f8lAaCC+l1tiY49kDT2K3EX8ePeP4+u5rjjtCvx3+JNeNxEX1F6A/w5/vOGHt8Srhh8iIiIiMqMo05ahYF+Qh3c9zC/e+gXNgWaMMSz0LKSisILWnlYOdx2mu7+b5q5mrLVp2+nnusK8QgbCA1gsTuOkL9SHx+XBYLj5vJsxmBFBVarALFeo9b2IiIiI5LKstPw3xnwAuB9wAhuttV9LdW2uB23P73+eP3/8z2ddGaPH5YmXIiYLyICcDsRERERERGaLaS+PNMY4gW8BlwNNwKvGmKestW9OxetNpeffeZ61j63N9jAmzZy8ObgcLjZ/ZDONnY2jBmTqrigiIiIikl1TtabtPGC/tfYAgDHmceBqIKeDtsHwIC3BFnydPnydPva37ecr//mVbA9rVG6Hm/5w6ixgnsnjstMuY8X8FayoWKGMmYiIiIjIDDJVQZsXaEx43AS8d4peKyPWWtp72/F1+mjsbIwHZo2Bk9/7g/4Re6TlqsQujK09rWxt2MrP9vwMp3HSPdCN2+nGWsvt59/Oly/8soI0EREREZEZaqqCNpPk2JDFc8aYG4EbAWprayf8gr2DvfFgLBaINXY24gv48HVEjnUPdGd8v4rCCpzGyZHuIxMe23icW3UuZflllBWU0dHbEc+WDd+vLBaM3bDqBr7b/12tQRMRERERmWWmKmhrAmoSHlcD/sQLrLUPAg9CpBFJupuFbZgjXUfiGbHhGTJfp49jPccyHlxBXgE1JTXUltVSW1JLbWktNaU11JTUsLhsMdUl1RS6Ctm4cyO3PnMrPYM9Gd97NA4cGAyrqlbhwMHrR17HaZxD9jB76mNPccVpV4z53kXuIq1BExERERGZZaYqaHsVWGaMWQI0Ax8FPp7q4pANsfvo7qGli4GT3zcFmhgID2T0wg7joLKoktrSaDBWUhMPymLH5hXMw5hkycCh6s+q5/bnbs/sHScbCw7eU/UeyvLLmF80n0VFi0asKcv1VvkiIiIiIpJdU9nyfx1wH5GW//9urf2XlNdWGcunM7tv2ZyyeCCW+Ct2rKq4CpfTNSnvAWCbbxvrHltHKByiZ7AHg8FiqSutY8ncJUP2abPW4jAOvCVerj7jaq475zoFYCIiIiIiMqqs7NM2FrGgze104y32srh0MbVlCVmyhGxZSX7JtI9P2TAREREREZlKOR+0nXnOmXbrf29lgWcBDuPI9nBERERERESm1bRvrj1WHpeHRUWLsj0MERERERGRnKO0loiIiIiISA5T0CYiIiIiIpLDFLSJiIiIiIjksJxoRGKMOQYcyvY4RlEBtGZ7EDKpNKezj+Z0dtK8zj6a09lJ8zr7aE6n12Jr7fxkJ3IiaJsJjDHbU3VzkZlJczr7aE5nJ83r7KM5nZ00r7OP5jR3qDxSREREREQkhyloExERERERyWEK2jL3YLYHIJNOczr7aE5nJ83r7KM5nZ00r7OP5jRHaE2biIiIiIhIDlOmTUREREREJIcpaBuFMeYDxpi9xpj9xpi7sj0eyZwx5t+NMUeNMbsTjpUbY14wxuyLfp0bPW6MMd+MzvMfjDGrsjdyScUYU2OM2WqM2WOMecMYc2v0uOZ1hjLGzDHG/N4Ysys6p/8UPb7EGPNKdE43GWPc0eP50cf7o+frsjl+Sc0Y4zTGvGaM+VX0seZ0hjPGNBhj/miMed0Ysz16TD9/ZzhjTJkxZrMx5q3o/1/P17zmHgVtaRhjnMC3gCuBFcDHjDErsjsqGYMfAh8Yduwu4EVr7TLgxehjiMzxsuivG4HvTNMYZWwGgTustWcC7wNujv6Z1LzOXH3AGmvtOcBK4APGmPcB/wrcG53TduCG6PU3AO3W2tOBe6PXSW66FdiT8FhzOjtcYq1dmdAGXj9/Z777gWettWcA5xD5c6t5zTEK2tI7D9hvrT1gre0HHgeuzvKYJEPW2v8C2oYdvhp4OPr9w8CHE44/YiN+B5QZYyqnZ6SSKWtti7V2Z/T7IJH/sXjRvM5Y0bnpij50RX9ZYA2wOXp8+JzG5nozcKkxxkzTcCVDxphq4IPAxuhjg+Z0ttLP3xnMGFMCXAg8BGCt7bfWdqB5zTkK2tLzAo0Jj5uix2TmWmitbYFIAAAsiB7XXM8w0RKqdwOvoHmd0aJldK8DR4EXgHeADmvtYPSSxHmLz2n0fCcwb3pHLBm4D/h7IBx9PA/N6WxggeeNMTuMMTdGj+nn78y2FDgG/CBazrzRGONB85pzFLSll+xf+tRuc3bSXM8gxpgi4GfAbdbaQLpLkxzTvOYYa23IWrsSqCZS4XBmssuiXzWnOc4Y8yHgqLV2R+LhJJdqTmee91trVxEpkbvZGHNhmms1rzNDHrAK+I619t1ANydLIZPRvGaJgrb0moCahMfVgD9LY5HJcSSWxo9+PRo9rrmeIYwxLiIB22PW2p9HD2teZ4FoSc5/EFmvWGaMyYueSpy3+JxGz5cysgxasuv9wFXGmAYiywrWEMm8aU5nOGutP/r1KPAkkX9k0c/fma0JaLLWvhJ9vJlIEKd5zTEK2tJ7FVgW7XjlBj4KPJXlMcnEPAVcF/3+OuCXCcc/Ee2K9D6gM1YWILkjus7lIWCPtXZDwinN6wxljJlvjCmLfl8AXEZkreJWYH30suFzGpvr9cBLVhuO5hRr7RettdXW2joi/998yVp7LZrTGc0Y4zHGFMe+B64AdqOfvzOatfYw0GiMWR49dCnwJprXnKPNtUdhjFlH5F8IncC/W2v/JctDkgwZY34CXAxUAEeA/w38AngCqAV8wDXW2rZoMPAAkW6TPcD11trt2Ri3pGaMuQD4DfBHTq6V+RKRdW2a1xnIGPMuIovcnUT+IfEJa+0/G2OWEsnSlAOvAf/TWttnjJkD/IjIesY24KPW2gPZGb2MxhhzMfB5a+2HNKczW3T+now+zAN+bK39F2PMPPTzd0Yzxqwk0jTIDRwArif68xjNa85Q0CYiIiIiIpLDVB4pIiIiIiKSwxS0iYiIiIiI5DAFbSIiIiIiIjlMQZuIiIiIiEgOU9AmIiIiIiKSwxS0iYiIjIEx5jZjTGG2xyEiIqcOtfwXEREZA2NMA7DaWtua7bGIiMipISeCtoqKCltXV5ftYYiIiIiIiGTFjh07Wq2185Ody5vuwSRTV1fH9u3aTF1ERERERE5NxphDqc5pTZuIiIiIiEgOU9AmIiIiIiKSw3KiPFJERERERGQq1G+uZ2/rXpZXLGfT+k3ZHs64KGgTEREREZFZa2/rXnYd2ZXtYUyIyiNFRERERGRWCvYFOX7iOADHTxwn2BfM8ojGR0GbiIiIiIjMOtt82/Bu8OIP+AHwB/x4N3jZ5tuW5ZGNXUZBmzGmzBiz2RjzljFmjzHmfGNMuTHmBWPMvujXudFrjTHmm8aY/caYPxhjVk3tWxARERERETkp2Bdk3WPrCPYHCRMGIEyYYH/keFd/V5ZHODaZZtruB5611p4BnAPsAe4CXrTWLgNejD4GuBJYFv11I/CdSR2xiIiIiIhIGpve2ETYhpOeC9swm3bnVkOSVGONGbURiTGmBLgQ+CSAtbYf6DfGXA1cHL3sYeA/gDuBq4FHrLUW+F00S1dprW0Z31sQERERERHJ3L7j++ge6E56rnugm/1t+6d1PKFwiO6Bbg51HIr86jyEL+CjsbOR5kAz/i5/2udn0j1yKXAM+IEx5hxgB3ArsDAWiFlrW4wxC6LXe4HGhOc3RY8NCdqMMTcSycRRW1ubwTBERERERERGt2zeMjwuT9LAzePycHr56ZP2WtZaBkIDHOk+QkNHAw0dDfg6fTQGGmkKNMWDsmPdx7DYcb1GJkFbHrAKuMVa+4ox5n5OlkImY5IcGzE6a+2DwIMAq1evHt/oRUREREREhqk/q57bn7s96TmHcVB/dn3G9wqFQ/QM9NDQ0cChzkMcbD9IY6CRxkAkS9YcbKYl2MKJwRMZ33N+4XwqiyupKqrCW+KluqSaf+KfUl6fSdDWBDRZa1+JPt5MJGg7Eit7NMZUAkcTrq9JeH41kD7fJyIiIiIiMkmK84vZcu0W1j22ju7+bsKEceDA4/aw5dotFLmL4tcOhAY43HU4kiXrbMDX4RsSlPmDflp7WjPOkhXkFcQDsqqSKqpLqqkpqWFx6eLIr7LFeNweXA4XTocz/rx0QZuJLD1LzxjzG+BT1tq9xpivAJ7oqePW2q8ZY+4Cyq21f2+M+SDwWWAd8F7gm9ba89Ldf/Xq1Xb79u2jjkNERERERCRTwb4gZ37rTJqDzVQUVHDnBXdytPsojZ3R0sVgMy1dLfQO9mZ0P4Nhvmd+JCArjgRksaCsbm4ddWV1LChcgMvpIs+RhzHJihBT3NuYHdba1cnOZZJpA7gFeMwY4wYOANcT6Tz5hDHmBsAHXBO9dguRgG0/0BO9VkREREREZFINhgbxd/nja8kOdUQafDR1RgIyf9Af31y79UQrX3jhC2nvV+gqpLKokqriKrzFXqpLT2bJ6srqqC2txeP2kOfIw2Gmb8vrjII2a+3rQLKo79Ik11rg5gmOS0REREREZpFgX5A1j6zB1+GjtqyWlz7xEsX5xSmvt9YS7AtysOMgDZ3RgCza4KM50Bxv8NEf6s94DIuKFsWDsliGrKa0hrrSOurm1jG/cD5up3tI2WIuyDTTJiIiIiIiM0z95nr2tu5lecVyNq3P3t5k23zbhqwva+1ppfLfKvnhX/yQ+YXz4x0X40FZNEvWdqIt49fwuDx4i71UFlcOCchqS2sjWbKSWgpcBWMuW8wFCtpERERERGapva172XVkV1Zeu7O3k4PtB9nTuofrf3k9faG++LkwYboHu7nmp9ekucNJDuNgoWchlcWVeIu98YCspqSGurLIWrL5nvnTXrY4XRS0iYiIiIgMkysZqlwVCodoDDRG1pK1R1rhNwYaI5tFByNt8Dt6OzK+X7G7OB6QeUsiQVltaS21JbUsmbuEmtIaCvIKcq5scbooaBMRERGRWWOygq1sZqhyQWdvJwc7DnKw/SANHQ00BhrxdfpoCjThD/o53HWYgfBARvcymLTt8j+7+rNs+MCGGVm2OF0UtImIiIjIrHGqB1uZGAwN0hhojDT4iHZcTNyXrCXYQkdf5lmy0vzSeJYstpastrSW2tJalpQt4dcHf80Xnv8CPYM9I57rcXlYWbkSl9M1mW9x1lHQJiIiIiKSINgXjLeJP37iOMG+YNouh+lko8yyo7eDA20HaOho4FjPMQAaOhp478b30hxo5nDXYUI2lNG98hx5LCpaNKRssaakhsVlkY2il5QtYW7B3LRli5XFldz167uSnnMYB/Vn14/9TZ5iFLSJiIiIyKwwGcFWYpdDAH/Aj3eDly3XbuGC2gvGPKbJzvwNhgdp7ByaJRvSBj/YTLA/OOJ5nX2d/L759yOOl+WXRbJkwwKyWJasuqSa/Lz8CY25OL+YLdduGdI90oEDj9vDlmu3UOQumtD9TwUK2kRERERkxhtLsJUq+xXsC7LusXVDgp4wYYL9keP+O/xTHmB0nOjgQMeB+GbRsTb4TYHIZtFHuo5knCUbzuVw8en3fJpLllwS77g4d87caVlHdkHtBfjv8LPiWytoDDTiLfHy5s1vKmDLkII2EREREZnRxhpspcp+bXpjE2EbTvoaYRtm0+5N3LDqhjGNKzHz19bTRntvOwc7DnKo4xCHOg/FA7JYUNbV35Xx/csLyiObRBdXU11SHV9HNr9wPtf89JoRa8gGwgM8vOth7r7s7qwES0XuIs6vOZ/y1nKWVyxXwDYGCtpEREREZEabrGBr3/F9dA90Jz3XPdDN/rb9KZ9rraX9RHs8S7b14Fa+v/P78Q6LTYEm5t0zL4N3E+FyuKgqroo09ygd2dxjydwllOSXJH3uxp0bU2bPxhN8TiZtnzA+CtpEREREZEabSLCVaNm8ZXhcnqT3KswrpCS/hBfeeYFDHYdo6GygsTPScTHWBj/VGJKZVzBvaHOP0hoWly5mcVmkuUdVcdW49ySbrM9DcoeCNhERERGZ0dIFWx6Xh9PLT48/TtaspMhdRNuJNpaVLyMUTr5erGewhy+99KWMxpPnyCMUDiXdm2yOcw7/dsW/cfN5N2d0r/EYy+chM4OCNhERERGZ0erPquf2525Pes5hHHz4jA+z//h+nnr7Kb704pfoC/UBkZLF0q+V4na648cyMb9w/pAsWW1pbTxLtnTuUu797b3c89t7kj63N9RLU6Bp7G9yDEb7PNRif+ZR0CYiIiIySbKxJ1cuGO19T+XnYq2ld7CXf7vi37j1mVvpDfXGzzmMA6fDScU9Famfjx0SsM3Jm0NlUSXNwWb6Q/0Uu4v550v+meXzllNXVseSsiXMcc1JO6Y/qfiTrGa61GJ/9lHQJiIiIjIJgn1BXm58maZA04Q3ZJ5p0u1FNtHPpXcwkpk62H6QA+0H4h0XGzsbaQo20RxoTpklC9swHb0do76G2+HmjvPv4HPnf46KwgqMMaz87kp2HdnF0rlLue19t2U8XsiNTJda7M8uCtpEREREJmiyN2SebFOZ6Uq3ofVon4u1lmM9x+J7kjW0N0SCskB0X7JAM8d6jmU8lkJXIaFwiL5QH3PnzOWmc2+ipqSGJXOX8NM3fsrG1zYmfV5/uB+LZb5n/gQ/jYhcyXQVuYsoLyinMdBIeUG5ArYZTEGbiIiIyATkwobMo0mXCZuIdEHZOQvP4cpHr6Rr4OS+Y7HP5aIfXkRtSS3+Lj/9of6MX8/lcHH2grNZXLp4SLfFxaWLOW3uacwtmMu7v/dudh3ZRW1pLV9d89X4c32dPn6y+yfTVrKoTJdMJgVtIiIiIhMw2RsyzxSdvZ184NEPDAmC4kHZDy5ijmsOPQM9SZ8btmEaOhuGHPO4PHhLvFSXnNyTrG+wj//7yv/lROgEAKFwiP1t+/nmld9MmcFcXrF8yNeYsZYsprrPWCjTJZNFQZuIiIjIBEzXnljjLXFMV76YTnd/N42BRho6GjjYfnDoWrJAE42BRkI2eXv8MOGUAVvMmro13HzezZEsWflplOaXDtkQOtgXxLvBGw/YYvcdLYOZ6rMZa8niqdRIRnKfgjYRERGZtTIJdCa63muie2Jl+vrjKXFMVb74q4//itPLT8fX6eNg+0EaOiJryQ60H2Cbbxu9g71J9xgbiz+d/6e83fZ20iYhHpeHj//px/nLM/8y5fOnIoOpkkWZqRS0iYiIyKyVSaAz0fVeE+kUmGlnxbFmy4J9Qd5qfYvLH7l8SAv8xDVlmXIYB1XFVXiLT+5JVltay762fTy086Eh94/xuDx8evWn+eKLX0watGXSQXGqMpjTXbI4GWWWIgraRERERCZgvJ0CM+04Ofy65kAzld+o5J7L76E0v5SDnQfxdfgipYuBRpoDzf+fvTsPk6us8/7/vqt67+olnb3XBBIDyJCICT4RBkhgHIkLzmOkdXCMM0EQAZfER5aH53J+DmrAGRAFYRDUoCBhGkdBAsiqBhy2GEggBEJIek06nfRSvXdV3b8/aklV96mlu6vT2+fllau7qs45dU6fIvYn931/v7T1JS9zH1aUXcT8gvm8c+Qdx+mOeRl57L5i95Dr8PZ5ufe1ex1Dm8u4WLdsHUvnLR1xBcXRjmBOFOM9zVKhcWpIObQZY9zAK0CDtfbjxpiFwANACbAd+Cdrbb8xJhu4F/ggcASottbuT/uZi4iIyLiaro2knQx32l2iipMX/OoCnl73NIe7DvNWy1tc+/S1DAQGIttZLF0DXXxl61dGdc7rl63nlo/eQkF2AXdvv5uvP/51x4BksY5TEVMJq6OZjjgRep1NBdP9v82pYjgjbV8DdgOFocc3ArdYax8wxtwJrAfuCH1ttdYuMsZ8NrSd/qsSERGZYuJNK0wU5kb62kikMqVwpEU6nHiyPKysWElJSwlLZi1xDCa+gI8mbxO3v3w7fT7nhtCdA5186O4PpfSeeZl5nDDjBCoKKyIl8BcUL2BB8QKer32ef33uX+n2DS0Ikp+Zz8qKlZFrHelUxFRC2UinI45lrzONPslkk1JoM8aUAx8DvgtsMMHSPquBfwxtshn4V4Kh7cLQ9wA1wG3GGGOtHd1qVhEREZkUEq0RG+lrw5XK1MN0N8S21vKfH/9PatuDUxV//NKPOdB2gAPtB6jvqKe+vZ6mzqa4FRcHy3JnkZORQ0dfR9xtvnrGV/n++d93fO1v5vwN//anf3N8bfBI1WimIqYSVkcaksaqcIhGn2SySXWk7YfAt4DwPz3NBNqstb7Q43qgLPR9GVAHYK31GWPaQ9u3pOWMRUREZFJK56hWmNPoXCrNrq21w26IPeAfoNHbSG17LQfaD8SWwe8IlsHv7O9ktHLcOdyw+ga+sfIb/OyvP4s7bTFZmBrOSNVopyImC0GjCUmphEKRqS5paDPGfBxotta+aow5N/y0w6Y2hdeij3spcClAZWVlSicrIiIi6XM816QlG9UaaaBzGp1LpVS8xcbdpt/fz6WPXEppQSkH2g5Q21FLfUc9BzsPxt1nsGx3drBRdEF5pNpiVXEVC4oWUFVcRXFOMYt/vDgmNIZlujO5bPllwbA0yjCV6kjVWE5FTAeNjMl0l8pI25nAJ40xa4AcgmvafggUG2MyQqNt5UBjaPt6oAKoN8ZkAEXA0cEHtdbeBdwFsHz5ck2dFBEROc7SOR0xzCl8AQlHtWouqmHtg2uHPU0xXtBLtj7rv3f/N219bXG36fP38etdv056rS7jYmX5Sj4w7wORQLZgxgIqCiuYkz8nplG0k1RCUjrCVKprysIBb9XmVdS211JZVMmz654d98AmIimENmvttcC1AKGRtm9aay82xvwXsJZgBcl1wO9CuzwcevyX0OvPaD2biIjIxDIWUxXjjaZdvuLyuCNU/oCfCx+4kF7f0F5i4WmK6x9eP2RE0KkM/rx/n8f1Z1/PgfYDZJgMfJFVHLEe3fto0muZnTebU2afQmVRJfM88/jRiz8a0m8sYAO8fuh1Hv/84yMKNqmGpOPZENqT5eHlL72c9uOKyOiMpk/b1cADxpgbgL8C94Sevwf4pTFmL8ERts+O7hRFREQknYZTgCPVKZSJ1pHd8pdbYkrWR+v2dZPpynR8LWADbN6xmedrn6fB20CDt4E7X7mT2vZafvD8D2JCmcXS7evmumeuS3r9s/NmU1pQyq7mXY5FQTyZHvZ9bV8kFN29/W4yXBmOTaLD0y0Hl8NPVaohabQNoVUtUWRyG1Zos9Y+BzwX+n4fcIbDNr3AZ9JwbiIiIpJmqRTpiA4EqU6hTLSOzBhDtjvbMfS4cccNdF0DXVz52JWRxy3dLVz+6OVJz2VO/hzKC8p5/dDrkWBnMORl5vHw5x5m9cLVQGx4TTTtcKTl8NNtNMFLa8JEJrfRjLSJiEx7ai4sx1M6Pm+pFOkYyahRomDT7+8nw+X8K4ef1MrfD8e/LPsXvn/+9+ns74xMKSwvLB8ypTDVaYejKYefTvo7RmT6co33CYiITGbhUYg9LXvG+1RkGkjH5204o0bxioqEX2vpDnbzaeps4kjPkbjTHCHYVDpdstxZZLuzHV+LDlHhKYVA3CmF4XLyS+cuZWXFSsdtqt9fjcs4/8qUSgVHEZHR0kibiIhMeuM54jnZRltTHTXaVruNC351QWS7ho4GZt40k9ULV3Oo6xCvHXwNG+ro09zVzD1/vWfI8aIZDHM9cznafZT+QD+eTA/fPvfbLJm5hCM9R7hq61V0DXRhsbhw4XbFnzbZ7+8ny53l+NpIQlSy+zbRy+GLyNSnkTYREZnQqmuqWXbnMqpr4v8iPp4jnul471SuMRFvn5cVP13B3B/MZcVPV8SMiA15rwSjRv3+fv77rf9m6R1LOfvnZ9M50BkJZhbLQGCAJ959gh0Hd0Sej2YwuI075nFuRi4Prn2Qvuv7aNrYxMmzTwbgxJIT+eaHv8knlnyCLy77Ik3fbKK8sByAssIybv77m8nPzHc8z/zMfL7xoW9QkFWAK/SrjAsXBVkFQ0LUkllLWDp36agLcISnUpYVlkXOsXFjY8KWBCIi6aKRNhERmdDGopfYYCMdLUtH2Xxvn5cX6l6gvqN+RMcYXEyjpbuFspvLeOiih6goquCtlrf4+uNf53DXYXIzczlp1kl4sjyOTZ0HAgM8+k7icvgm9L8AQ9fF5WXmceP5N3Lj8zfGXUcWr5jG4OqI65au47qnnStBuoyL68+5nuvPuT7perR0jn6Gp1KWtJSwZNYSjbCJyHGj0CYiIpPaaIPTSENTOsrmD+cY0dfZ0t3C2y1vU9dRx8fv/zi9/qE9zj7yq48MOeduXzfP1z0/5HmD4aRZJ7GgeAGVRZW8efhN/lz7Z8frtqH/Oeka6KK+oz5hafpUQ1SqUxJHUwZ/JCbDFFgRmXoU2kREJKGJvGZrOKEnnfsPp2x+vFCY6Bjn/uJcTphxAvM88/j8aZ9nW+02fr3z15Hy9Q3eBpbcPvLpftnubO773/dx3TPX8faRtzlt7mns+PKOyOt3b7+b7U3bHde9ZbmzMBjH8v3prqSYSnVH9R8TkelAoU1RaMfdAAAgAElEQVRERBI6HtMTR2K4/caGs/8Fv7qA75//fera61g8czHV76+OGX1LtWy+Uygs/Y9SNv/DZl5qeIl+f7/jMfzWzztH3+Gdo+/EHfFKRYbJiGlAHXnelUFbbxvL5i0jNyN3SOCpfn81G57Y4HjMLFcWGBxDW7gIyI9f+vGIztcpgCVrKj3R/iFBRGQsKLSJiMiklCw4rdq8igH/QNwRwkT7dw508rXHv0bABnAbNxue2BAz+pasbP5f6v5CcU4xF//m4phwEyBA50Ann37w0ylf58zcmbT2tDquIct0ZeIyLscAFS+whc9x79G9cQNPsqmJwJhUUox3PhpNE5HpTqFNREQmpWTBqba9luau5hHtD0QCnd/6I6N3td+opaOvAwhOMXQKSwD37LiHe3YkLoGfiMFgsZw25zQ+uuij3PTCTY7bDQQG4pa+d7vcZLuyR9wQOtnUxESvpTtkaTRNRKY7lfwXEZFx41SqPtXy9+F+Y07yMvIY8Ad7fA1uCp3K/k46+zspubGEqh9WcdMLN8UNbKm6cvmVFGQ5FzwxxkS+JjrPRKXvH/7cw6NuCJ2oOXWi17as3cKOL+9Q2BIRSROFNhGREXKqWiip21a7jbKby9jeuJ3m7ma2N24PPm7anlLfs0T9xrp93bT3tgPHios8t/85DrQd4M8H/sx9r99HfUf9sILX4KqJhdmFMe9vMORk5PB/Pvx/kobB/Mx8ls1fxtaLt1KQVRDpbeY2bjyZHmblzQKCn6s1i9YkDF/Xn3M9jRsbOb30dObkz+H00tNp3NjIR078SOT4yXqZiYjIxKbpkSIiIzDaqoUjlaiS40Su8hituqaa3Yd3s7tlN77AsTVX4SIgnUc7gWBgaexojFvO/5JHLmGuZy49bT34A/4hZejDa8DCx121eVVazt9gyM/M59F/fJRl85ZFpgiWF5bz4iUvsuS2JQmnXcKxkS5PlofGjY1s2bWFvUf3Yq3l9ldup6WrBQh+rk66/SQ2nb+Ja566JuEaspe/9PKQ9wlPcVy1eRW17bVUFlXy7LpnFdhERCYZhTYRkWEabdXC0UhUyXEiVXlMFCD3tOxhZ/POyOjPYOHw1dDRQNktZRhM5PHcf5/LhpUbyM3I5al9T3G05yjZ7mz8+Id9ji7j4syKM/mbOX/D/Tvvp62vjdl5s+ka6KJ7oDvufhZL50Bn5F5HN1t+9J1H4xY3gWDg82R5YsKWJ8vD+tPX4+3zUnZzGZ39nZHtw5+ra566hrevepszfnpGwkbSTjxZHsdAlyoVARERGX8KbSIiw5RqufexdjxG1lJpXO3t87L63tXUttVSWVzJ76p/l1KzaqdqiNHC4S36a4+vh+/++bsx2410bVnABthxcAdbL95KS09L5Gd51RlXsea+NXQPdOO38cNg+F5H/+yvfvLqhKNs4ZE1p7CV7HP16NuPHvdG0qAiICIiE4FCm4jIMCWrWrj36N5hHzOVADY4QO0+vJudzTsdX4sXlIZzHuHwMrjHWFlhGa09rVQWV/LdVd9l7X+tjUzbO9x9OGZ0LLzP3Z+8mzn5c9hzZA/vtb037PMarKqoiuauZnp8PRRlF/GTNT9hyawl/OnAn7j+mevp9sUfKYvmFLyAyJTFn27/KS82vOi4r9O9DhcNiff5KM4pjhu2xuJzJSIiU4NCm4jIMCX6xTyVUupOkk1tdFpD1+htBIK/0JfdXDbq9XXePm9khKylu4Wt72wdMlWvc6CTPUeCBUIOdx/m7+/7+5hjDB4dC+/z2Yc+m/J5pKKpswmf3xc57y8/+mW2XryVS06/hG8/9+2UjxMvDIWnLFosu5p3pXyvEzWldhkXK8pWxD2XsfhciYjI1KDqkSIiw5SoamGqpdSHI3oNXXRxjfBUun2t+4a85u33cs4vzuGTv/7kkJL6TsKVHBs7gkGw0dsYE9icDC78cTz1+/uHXO+a+9ZgjBlSMTE86uckWRga7r0ON6V2qtj4xy/+kYcueiht7yUiItOHQpuISAJOPcMS/WKeqJS6U0+yVCRa6wRgrXN4CtgAj7z9CK82vhpTUn9b7bbIfoc6D/Hc/uf4u1/+XUzwG4tAlmGcJ3eEQ1WicJWK8FTHcMXEssIyAEoLSuPek2RhaCT3evD7lxWW0bixMemo50g/V+NpyawlLJ27VEVKRETGmKZHiogQDFRb3tjCO0feYfHMxVS/v5qC7IK40xbDv5iHy72Hq/mtf3g9V269csjatOjpjQECtHS3RKYwLp27NGY9WmNHI1v3bo2cyxvNbyQsbpEsYEVPVQyPwJXkltDa05qw0EY6ZZgMfNbn+JrFMid/DqfOPpWXG1+O/IwMBouNfHXhwu1yMxAYcDxO9FTHcOPnuo46ZuXN4oG1D8T8/J1K5sczkrL5nixPTFXJVANXvM/VRAxsoCIlIiLHi0KbiAzLZOkFNhzhQBWuFug2bjY8sYGtF2+N2W7wtUcHg3A1v8EhL1lPso/88iO4XW66+4OFM8Jl7nMycuj19ZKfmU+PrycSXNIhYIOh8Xhyu9xku7Ljrtf63urvsf709XT2d8b0PXvpSy/FlLn/1pnfCvYrG+a6r9H2KxtJ2fyR/vfh9LkSEZHpTaFNRIZlIvUCSwennmt+68fb7+WCX11AUU4RAC3dLbTUttDgbUhY+n5wBcdkPcl6fD0xj8PBrNfXC5C0SfPxkEpgdOGKOzqWn5VPzUU1rH1wrfO+UVMUBweWeZ55MY/XLV3HdU9fl/Q4MLS/2Gj7lU0E6pkmIjI9KbSJyLga75G7ROvFOgc6I6GpwdsQeT66OmOYUwXH0v8oJS8rD0jek2yi8mR6KMoposHbwJy8OXT7uukZ6ImMSOZl5lFzUQ2XPHxJ3NGx8PS+rRdvTWmKYqJgEl73lcpxJvNIcLyfwWS+JhERGbmkoc0YUwHcC8wDAsBd1tpbjTElwBZgAbAfuMha22qMMcCtwBqgG/iitXb72Jy+iEx2qYzcjWWwS9QbC5zXi4WnNq7evJo5+XOAYAXH6PAXLnXfOZC4AuNEkuXOwuf3DQlCV269kgZvA/ML5rPtX7axZdcW9h7dy6KSRVSfWj1k/dbg0bFwkEp1vVayezzZ1n2NhMKZiIhES2WkzQdstNZuN8YUAK8aY54Evgg8ba3dZIy5BrgGuBq4AFgc+vMh4I7QVxGRYYvuHTbcptHePi+r711NbVstlcWVPPOFZ4bsm6wZciIDgYHICFyi6o7HU6KpjIley8/M58bzb+QXr/0i4ZqvcP+ywQaHjHgjRelar6V1XyIiMp0kDW3W2iagKfS91xizGygDLgTODW22GXiOYGi7ELjXBmtQ/48xptgYMz90HBGRSJja37af9t52AMdANrihdENHA8U3FlOYXciikkWOIcxp3wABmrubKdxUSEFWAQuKF3Dy7JPZsnZLwmbIk01+Zj5XrriSn7zyk5gKjBAcMSwtKKW9r92x/5rLuFi3bB1XnHFFWs5FI0UiIiLpM6w1bcaYBcAHgBeBueEgZq1tMsbMCW1WBtRF7VYfei4mtBljLgUuBaisrBzBqct0MN7rnST9wmGqs78zZtSnoaMhsk7srMqzHAuEWCzWWtp623i18VVm/2A2nz7l06wsXwlAXXsdi2cuZs2iNUP2DfP2e9nZvJPmrmY2bdtEbXstp845lRfrX5xQ684yXBkx1SZT4TIurj/neq4/5/qYCowrylbw7tF3WTJrCVedcdWQSpl5mXnHtQ/YcItpxNteRTlERGS6SDm0GWM8wEPA1621HcGla86bOjw3ZD6OtfYu4C6A5cuXp7+Lq0wJU61S4WTnVB1xOFMVz/nFOew4uMNxip7F4u0PBrXGjY3JG0pj6fP3cf/O+7l/5/2R5124cLmcKzVGO9R1iGufvjalc0+nLHcWBkOfv8/x9Rk5M3jzijd56M2H2PCHDfT7+4dsk5uRG2kTEK8QR/Qas8H/4NG4sdFxXVo86Q5Hw/0HmHjb6x9yRERkukgptBljMgkGtvustb8JPX0oPO3RGDMfaA49Xw9URO1eDjSm64RFprrhjC6msmYr2f5ODaWdDJ6qGK6guKJ0BUd6jkTO1+mYrx16zXF0zUnABtiyawv//sK/j2idWYAAgUBqI2bRa7wMBmMMpQWlfGDeB/jmh7/Jx+//eGSaYTI57hww0O/rT7h9lisLDHFDW3lhOfM887jijCtYOm9p3CqJy+YtS1iII9FnJ966tHgUjkRERMZXKtUjDXAPsNtae3PUSw8D64BNoa+/i3r+SmPMAwQLkLRrPZtI6lIdXRy8ZquluyVmemGq+wdsgK6BLvIz8yMNpQfv7+3zcsF9F8SshQpXUHx2/7NYLC3dLbzvx+9j79G9kSDkNm6uePQKAgRSnurXNdDFN574huMIU7pFB8jw1Mv23nbu//T9eLI8sRUKC8qwWBq9zv8GlenO5O2r3o6Uug+Xx3caDQOGhDFMMLC6zLFRwmQNoVWIQ0REZHpIZaTtTOCfgJ3GmB2h564jGNYeNMasB2qBz4Re20qw3P9egiX//zmtZywijuu9wiEqPL1w8C/x0SNgT+57ktcOvRYz/TA8quW0/w1/vsGxeAUcCz7RfczC/NaP3/qHf30Oa9GOl/BI3/rT18dUKJyVN4sls5awvWk7+9v2Y60dsiYsutR9uDx+vNGwwSXri3OK2dm8c8j5TIWG0CIiIjI6qVSP3IbzOjWA8xy2t0B6yo+JiKNE6726BrrYvGNzTBXA8KhauPhEIgEbYPOOzWRnZPPOkXeoKKrglr/cktbzn8i6BrrYe3Sv42vhaYKd/Z1x14RFr/8a3L8sOggPfg2ChURUVENEREQGG1b1SJl+pkP1xol0jakW+kjUEDpgA2z4wwaWzlvK0rlL2fzaZjY+sZH+QGrTDbsGutjwhw1kujLpGugi05XJQGBg5BeVZol6jaVjn/zMfBaVLIo8dirCkWhN2ODPUKLP1Hh/3kRERGRyUGiThKZD9caxusbhFPmA+IU+nNaYJWsI3e/vZ/Xm1WS7s+n2dQ+78XO/vz+ypmysA5vLuFI+v/eVvI/CnELePfou7X3tYElaJCQ/M5+V5St56r2nUj6f6lOrI48VrERERGS8Ja+LLTJOnEadJotttdsou7mMrz/+dW564SYu+/1lFG4q5KTbTnK8jug1auEQEr1GrcnbxIqfrmDuD+ay4qcrWLNoTUzBCicDgQE6BzqHHdiOB4OhIKuAP//znyktKE1pn/zMfL515rd4+Usvc/Tqo7Rf005ZYRkApZ7SuD8Pl3Fx4UkXkp+Zn/D4LlwUZBUc135lIiIiIqnQSJtMSMMZdRqN0fYdGzySdskjl7D78G52t+yOqZYYDk57juyh5KYSFhQv4PT5p0dGcRKtUevs76Ts5rLI9L7D3Yep+GEFuRm5I77u8VZaUMpbV76FJ8vDjJwZ1HfUJ91n8AjY4DVh4abRTuXxl85dynVPX+d43PDUSaey+SIiIiITgUKbTDgjqYwYLdU1aqMJhoPL5buNm8seuYxMdyZ9/j5M3No94Av42Ht0Lz0DPZFRt5o3a+JOdRy8Fitcmn4kPczSyWDIzcyld6B3yBRFg8FlXLhdbnx+35Cy9rPyZkXu4cmzT6bH18P+tv0EAoEhx3JqHB3m1DQ6Xnn8rRdvdQx10X3mFNhERERkIlJokxEbqwIeiUadosuxO/H2eXmh7gXqO+pjplQOHhEDRhQMvX1ex8Ie4YqM4YbJqRS9aPA2ULypGJfLhT8w/LL4421W3iz2fW1fJCSVF5SDgfqO+khgAmJCVJ+vb0hZ++iKjNE90SD4MxrOCFii8vjhnmeJGlKLiIiITEQKbTJiY1XAI1FlxETl2J1Gzub++1yMMfT5+iI9tTY8sYHLV1weNxj6A36uePQK5nnmxYS8FT9dwdtH3h525cJEAgQIBMZvzZkLF5nuTIwx9Pp6h7VvTkZOSj3Eol+vrqmOW9Z+8HTHPS17aPA2pLVxdHTfNTWkFhERkclCoU0mnIqiCrLd2ZFRq2hu4+a+nfexr20fd3/i7sgIWkVRBdc+fW1MA+gAAXp8PTH7+60fb7+XW/5yS9yqiN2+bn75+i+xWNzGzVcf+yrAkGNNNG7jTtiDLT8zH5dxUXNRDZc8fElktOnFS15kyW1L6GV4oa04p3jY55hsRDb69eqaYFhW3zIRERGZ7hTapqB0TVscTZGOkdpWu41rn7rWMbBBcHpkXUcdfe/1UXpzKT0DPfitHxeupKXfo/mtP24whGPTG/3WP+HDGgQD2ZUrruQnr/wk0kDbbdzkZuRyxRlXYDAxTaCjR7TmF8yPrPeK3jcvM49N52/imqeuiVkH5nK5WFC8gJNnnzym16RS+8k59ZATERGRqUehbQpKx7TFZEU60h3o4q0Vi6e5uznm8XACGwTDX3R1x7HiNm6AhCNg6eAyLq4/53quP+d6tuzawt6je2NC2mCDA1F4vZfTvl9Y+oW4xT1kfCnYioiITA8KbTKEt8/Lqs2rYkvWRxXpqLmohrUPrk1bOf5wQOzz9yUNbOlcTzbWQQogLzOPt696m+V3LafB25C244YbUodHxKIrK8Yr0pKMJ8vjuG8q69YmE41OiYiIyGSj0CZDbH5tc9xqhuHgFh14ElVdDI+g/f7t3wNw/gnnk5ORQ117HYtnLubsyrP5u1/+3bCLYEwGniwPWy/eyjzPPM6sPJPtTdvZ37ofn40/wuc2biw2YUNsT6aHTX+3ifr2+oSjaeJMo1MiIiIy2Si0TTGjnba4rXYbG57YkHBEK94IVa+vl8seuYzywnIWz1xMZWEln9ryqZg1YU+8+0Tk+3BT48ksNyM3WHkx1KvMYMhwZbBh5QauP/v6SJiKLmt/4q0nxkzvdBs3Ga4M1p6yllULVnHri7eys3knniwP3QPdkQAX3a8snQ3GRURERGRiU2ibQuKtQ6u5qIba9tqYPmXhIOft87LljS3sat5Fc2czNbtr4lZVTGYgMMD9u+5PefvJHNjcxs03P/xNrj/7eoCU13x5sjycu/Bcdh/eTXZGNucvPH/IaNkf9v0hUhb/nk/eo75iIiIiItOcsXb8f3Fevny5feWVV8b7NCY1b5+XspvLYppFRwuXgw+P6nz6lE9TXlAeqTaYaDqexHIZFw0bGpjnmXdc3m+smpiLiIiIyMRhjHnVWrvc6TWNtE0RW97YkjB4hac0+q0fv9/P/TtTHxEbL8mmT2a6MtmwcgP1HfU8tPshBvwDkWCa5c7CGIPBxG3UPVi4BUC4n5lTufvw9MTjFdhAa7BEREREpjuFtgkqPG0x3DgaiBTviJ7eGPZG8xsph5PJICcjh69+6Kvc/tLtjtdlMNzy97dwxRlXAHBn/51DytUDkeestdz28m1xf0bxinuo3L2IiIiIjDdNj5xgvH1ebvjTDfzwf36IL+Bz7D9mMJy38DzeN/N9tPa08vi7j9Pa2zoOZzs2st3Z7P/6fvIz8+NO+cxwZdB6deuwAlRnfycr7lrBniN7MMYQsIHIqJqKe4iIiIjIeEo0PVKhbRxEl8H3BXzMyJlBW28bHX0d7Di4I6Xm0uMt3Ccs3TxZHh67+LFIgAoXV+ke6I5MfQz3JRtpyOrs70ypAbWIiIiIyPGi0HacDJ7S2Ovr5fG9j9PQ0YAxhrn5cwH404E/OY6gTUS5Gbn0+HqoKKzgM6d8hiM9R5iZN5NTZp3Cva/fy7babRhM0kbVbuNm9cLVfHTRR6nvqOf2l2/HYOjz95HlzsJaO6RMfphCloiIiIhMdQptaTa4TP7h7sO09rby2qHXwJKwefJkYDCRxtA/funHCSsXRgeq8sJy3m19l9tfvj04tdMG4o6MKYiJiIiIiByj0DZK4emMv33rt7zb+i517XVA/CbTk1GGyaC0sJSegR6qiqtGVXBDgUxEREREZHjGpeS/MeajwK2AG7jbWrtprN5rLG2r3cZHfvkRenw9430qaRUup5+ONWKDebI8rD99fVqOJSIiIiIy3Y1JaDPGuIHbgb8D6oGXjTEPW2vfHIv3S7eu/i5q22t5q+UtPvfQ5+jz9433KaVVvPL2IiIiIiIy8YzVSNsZwF5r7T4AY8wDwIXAuIc2f8DPwc6D1LbXxv7pOPb90Z6j432ao+bUYFrl7UVEREREJp+xCm1lQF3U43rgQ2P0XjG8fV7qOuqGhLID7Qeoba+lvqMeXyC1QiFjVdZ+OMKVFa8840rKC8t5at9T+Pw+ZntmU5JbwtHuoxzuPkyGyeD8E88nOyM7ZgQN0PoyEREREZFJbKxCm3F4LqbiiTHmUuBSgMrKypQO6g/4aepsGjpKFvVnOE2mS3JLqCyqDP4prKSiqIKqoqrIc4++8yhXPXYV/f6x65tmMCybt4zXD70OBIub5GXk4bd+1p6yllULVsUErQ0rNwz7PbS+TERERERk8hqr0FYPVEQ9Lgcaozew1t4F3AXB6pEAHX0dCQNZfUd9yhUbM12ZVBRVRAJYRWFsIKsoqkg64vS5Uz/Hxic2jji0uY0bay2VRZXkZeYx1zOXWXmzaOttI8Nk8LElH2Pd0nV4sjyquCgiIiIiIo7GpOS/MSYDeBs4D2gAXgb+0Vr7htP2eVV5NuvyLNr72lN+j5m5M4+NkoVDWXEwlFUVVTHXMxeXcY36WlKpHunChcVy/gnns2TWErx93kgDaoUvERERERFJZlz6tBlj1gA/JFjy/2fW2u/G3bbUWC479jjLnUVFYYVjIAs/zs/KH5PzdtLZ38nmHZv53Z7f0eRtYn7BfD666KND1o8pnImIiIiIyEhM+ObaFSdX2FseuiUSyubkz0nLKJmIiIiIiMhkMC7NtYdjbv5c1p6ydrxPQ0REREREZMLRcJaIiIiIiMgEptAmIiIiIiIygSm0iYiIiIiITGATohCJMeYwcGC8zyOBWUDLeJ+EjBvd/+lN93960/2f3nT/pzfd/+ltPO5/lbV2ttMLEyK0TXTGmFfiVXKRqU/3f3rT/Z/edP+nN93/6U33f3qbaPdf0yNFREREREQmMIU2ERERERGRCUyhLTV3jfcJyLjS/Z/edP+nN93/6U33f3rT/Z/eJtT915o2ERERERGRCUwjbSIiIiIiIhOYQlsSxpiPGmP2GGP2GmOuGe/zkfQzxvzMGNNsjNkV9VyJMeZJY8w7oa8zQs8bY8yPQp+H140xp4/fmUs6GGMqjDHPGmN2G2PeMMZ8LfS8PgPTgDEmxxjzkjHmtdD9//9Czy80xrwYuv9bjDFZoeezQ4/3hl5fMJ7nL6NnjHEbY/5qjPl96LHu/TRhjNlvjNlpjNlhjHkl9Jz+7p8mjDHFxpgaY8xbod8BVk7k+6/QloAxxg3cDlwAnAJ8zhhzyvielYyBXwAfHfTcNcDT1trFwNOhxxD8LCwO/bkUuOM4naOMHR+w0Vp7MvC/gCtC/53rMzA99AGrrbVLgWXAR40x/wu4EbgldP9bgfWh7dcDrdbaRcAtoe1kcvsasDvqse799LLKWrssqrS7/u6fPm4FHrfWngQsJfj3wIS9/wptiZ0B7LXW7rPW9gMPABeO8zlJmllr/wQcHfT0hcDm0PebgU9FPX+vDfofoNgYM//4nKmMBWttk7V2e+h7L8G/tMvQZ2BaCN3HztDDzNAfC6wGakLPD77/4c9FDXCeMcYcp9OVNDPGlAMfA+4OPTbo3k93+rt/GjDGFAJnA/cAWGv7rbVtTOD7r9CWWBlQF/W4PvScTH1zrbVNEPylHpgTel6fiSksNN3pA8CL6DMwbYSmx+0AmoEngXeBNmutL7RJ9D2O3P/Q6+3AzON7xpJGPwS+BQRCj2eiez+dWOAPxphXjTGXhp7T3/3TwwnAYeDnoenRdxtj8pnA91+hLTGnf0FTuc3pTZ+JKcoY4wEeAr5ure1ItKnDc/oMTGLWWr+1dhlQTnCGxclOm4W+6v5PEcaYjwPN1tpXo5922FT3fuo601p7OsGpb1cYY85OsK3u/9SSAZwO3GGt/QDQxbGpkE7G/f4rtCVWD1REPS4HGsfpXOT4OhQe9g59bQ49r8/EFGSMySQY2O6z1v4m9LQ+A9NMaGrMcwTXNhYbYzJCL0Xf48j9D71exNDp1TI5nAl80hizn+Dyh9UER95076cJa21j6Gsz8N8E/9FGf/dPD/VAvbX2xdDjGoIhbsLef4W2xF4GFocqSWUBnwUeHudzkuPjYWBd6Pt1wO+inv9CqIrQ/wLaw8PoMjmF1qTcA+y21t4c9ZI+A9OAMWa2MaY49H0ucD7BdY3PAmtDmw2+/+HPxVrgGauGp5OStfZaa225tXYBwf9/f8ZaezG699OCMSbfGFMQ/h74CLAL/d0/LVhrDwJ1xpgloafOA95kAt9/NddOwhizhuC/vLmBn1lrvzvOpyRpZoz5NXAuMAs4BHwb+C3wIFAJ1AKfsdYeDf2CfxvBapPdwD9ba18Zj/OW9DDGnAX8GdjJsXUt1xFc16bPwBRnjDmN4GJzN8F/yHzQWvsdY8wJBEdfSoC/Ap+31vYZY3KAXxJc+3gU+Ky1dt/4nL2kizHmXOCb1tqP695PD6H7/N+hhxnA/dba7xpjZqK/+6cFY8wygkWIsoB9wD8T+v8BJuD9V2gTERERERGZwDQ9UkREREREZAJTaBMREREREZnAFNpEREREREQmMIU2ERERERGRCUyhTUREREREZAJTaBMRERkGY8zXjTF5430eIiIyfajkv4iIyDAYY/YDy621LeN9LiIiMj1MiNA2a9Ysu2DBgvE+DRERERERkXHx6quvtlhrZzu9lnG8T8bJggULeOUVNZUXEREREZHpyRhzIN5rWtMmIiIiIiIygSm0iYiIiIiITGATYnqkiIiIiIjIWKiuqWZPyx6WzFrClrVbxvt0RiSlkTZjTLExpsYY85YxZrcxZqUxpsQY8woupDkAACAASURBVKQx5p3Q1xmhbY0x5kfGmL3GmNeNMaeP7SWIiIiIiIg429Oyh9cOvcaelj3jfSojlur0yFuBx621JwFLgd3ANcDT1trFwNOhxwAXAItDfy4F7kjrGYuIiIiIiEwjSUObMaYQOBu4B8Ba22+tbQMuBDaHNtsMfCr0/YXAvTbof4BiY8z8tJ+5iIiIiIjIJBawAfp8fXT1dyXcLpU1bScAh4GfG2OWAq8CXwPmWmubAKy1TcaYOaHty4C6qP3rQ881RR/UGHMpwZE4KisrUzgNERERERGRycMf8DMQGMAX8DHgH2AgMEB7bzv72/ZzoP0ADd4GGjsaaexsTHicVEJbBnA6cJW19kVjzK0cmwrpxDg8N6SDt7X2LuAugOXLl49/h28REREREZFh8AV8MYFswD9Ak7eJ/e37qWuvC4Yyb2PMn/a+9mG/TyqhrR6ot9a+GHpcQzC0HTLGzA+Nss0HmqO2r4javxxIHB1FREREREQmEGvtsVAWCmTdA90caD9AbVstdR3HQllTZ1NkxKzf35/S8TNdmcwvmM98z3zKCsr4Db+Ju23S0GatPWiMqTPGLLHW7gHOA94M/VkHbAp9/V1ol4eBK40xDwAfAtrD0yhFREREREQmAmvtkKmLR3uOsr81OHWxvqP+WCjzNtHgbaCluwU7dBKho4KsAuYXBAPZ/IL5lBaUUlFQQUVRBVXFVZQWlJLtzibDlUGmO5NccuMeK9U+bVcB9xljsoB9wD8TLGLyoDFmPVALfCa07VZgDbAX6A5tKyIiIiIi42Aq9CkbiYANcLT7KC3dLQA0dzWz7cA2Dncfpra9lnpvfUwga/Q20tnfmdKxDYbZ+bMjgSw8WlZRVEFVURVVxVXMzJ0ZCWQZrgwyXZkY47SSLLmUQpu1dgew3OGl8xy2tcAVIzobEREREZFJZqKHonCfsqlmcJEPb7+X2vZaDrQdoLa9lpcbX+aRPY/gsz4Amjqb+Ntf/G1Kx852Z0dGx+Z7QqNkhRWRUFZZVEleZl5MKMtwpToeNnxjd2QRERERkWlgqoai8RYOY76Aj35/P4e7DnOg/QAH2g9Q115Ho7eRBm9DcD2ZtzEyopaKouwiSgtKY/5UFAanLVYVVTHfM59Md2ZaRsnSQaFNRERERESOq3CRj/BIWc9AD3UddZGRspj1ZKFQ1j3QndKxXcZFQVYB3j4vAQJDXs915/Kdc7/Dv5z+L5FA5na5032JaaXQJiIiIiIiaRUu8hEeKWvvC/Ymq22r5UBHMJSFS+A3eZs42HkQv/WndOycjJxjI2SeUsoKyygvLKeyqJKqoioqCivYtG0Tt750q+P+Pf4emjqb8GR50nnJY0qhTUREREREhsUf8EdGyvp9/RzqOhQMZR211LbVDulP1trbmvKxS3JLIoFsfsH8mLVkVUVVzPXMjZm26DRKdurcU8nPzKdroGvIa/mZ+SwqWTSq6z/eFNpERERERCRGuMjHgH+AnoGeyFqy2vZaattrYwJZU2cTvb7elI6b4cpgbv7cSNXF8ChZOJBVFVdRmF0YCWQZrowRrSWrfn81G57Y4Piay7ioPrV62MccTwptIiIiIiIOJnpVyFR4+7wc6TkCwJGeI3j7vBRkF0SmLQ4EBmjtaeW9tvcigWxwkY/mrmYCdujaMCd5mXmUFZQFqy6GRsnCoWxB8QLKC8vJzsiOBLKxWktWkF3A1ou3sua+NXT1dxEggAsX+Vn5bL1466SaGgkKbSIiIiIyTSULZalUhYwXisZTeD3Znw78iU/9+lN0+YJTBOs76pl10yxWLVzFQGAgsp6sva895WPPzpsdCWRlBbGjZAtmLGBm7kyy3FmREvjjWXHxrMqzaNzYyCm3n0JdRx1lhWW8ecWbky6wgUKbiIiIiExToy3Vv612W2QkB6Cxo5Gym8vYevFWzqo8K12nOUTABiIjZZ39ndS21wbXk7XXUtdRR31HPXUddbzc8DIWG7Nvf6CfJ959wvG4ma7MSG+yUk9ppLhHRVEFC4oXUFVUhSfLE+lN5jKuMbvGdPFkeSjJLaGuo46S3JJJGdhAoU1EREREJpF0TVlMNkKWyutr7luDt98beS5AAG9/8PnGjY0jDgjRTaNbulp4r+29yHqyuva6mCIfLd0tQ4JZMi7jYkXpClYvXE1lYSWVxZUsKF5AWUEZWe6sSJEPmTh0N0RERERk0khHI+tkI2SpjKBteWNL3HVeARtgy64trD99vePr4abR/f5+6jrq2N8aqrrYXhvsT9YRCmWdjXT2d6Z0TQbDnPw5kVL49R31cX9OARtg1YJVfO+876V0bBl/Cm0iIiIiMm0kGyHbc+WelEbQ3jnyjmM5eYCugS52H97Noc5gGfwDbQdiQ1lopOxQ5yEGAgMpnXe2OzsydbGsoIyKwgoqiyupLKxkYfFCKosqycvKi6wlu3v73Xz98a9PmZL3051Cm4iIiIhMG8lGyK556pqErz+w8wG+sOwLVBRWkOvOpcffM2Q7g+HOV+/kP/7nP1I+r+LsYkoLS5nvOdaXLNwsemHxQuZ55pGVESzwkcpasqlW8n66U2gTERERmQAme3n5VM9/vK8z2QjZWy1vJXz9O3/6Dpue30Sjt9ExsAFYbMwxXMbFPM885nuOVVysLK48VnWxeAEzcmdERsnSYaqVvJ/uFNpEREREJoB0rNUaT6me/3hf54klJ5KXkUe3r3vIa27j5t3WdzGYuMU96jrqEh4/w5XBRadcxIcrPxwJZJVFleRk5JDpyjyuJfCnUsn76U6hTURERKa08R7Zmawm4s8tlZ5o4SIfTd6mmKqL4fVkdR11joENwG/9keM7cRs3nz/t85ww44TIKNkczxw+cu9HaOhsoKKwYsKFoqlS8n66U2gTERGRKW28R3bG2liFq+H83FJtMD2aRtThio7haor1HfXM+cEcLlt+GRmujEgp/AZvA03eJvr8fSkdN8xlXJxVeRbL5i3DH/Bzz/Z76PX3AsE1avmZ+Tz2+ccc+6/Nyp9FQ2fDhA1FS2Ytifk63UyF61doExERERlnIw0z3j4vL9S9QH1H/bBDULrOJ9UG06lsZ63FF/BxpOcI77W+F2kYvffoXn6+4+f4rT/mvXv9vdz64q0JryU/M5+ygjLKCoNrySoKK6gqrmJO/hwu//3lHOw6SHlhObuv2B0TuDadvykyrbC8sHzCjaANx0QZKR0vU+H6FdpERERExlGqoSdd+6XzfFJtMJ1ou/M2n8fa96/lUOehSCn8jr6OYZ1zRWEFp809LVJ1cUHxgsif2XmzyXBlOK4l+9fn/pWDXQeZmTtzSCDTtEKZSBTaRERERMZJqqEnXful+3wSlc/v9/XzpYe/RGlBKX888Me4FRn7A/3cv/N+x9cyXZmUFpQyEBig0dsY97wv/puL+f7530/lEkUmJYU2ERERkXGSrGfYll1bWH/6+rTtN9rzuf/1+/nEkk9Epi7eu+PeuGGsL9DHA288kNL7VhVVsWrhKqqKqqgsqmRB8QIWFi+krLCMLHfWmDWKTtdap6mwZkomNoU2ERERkXGSrGfY3qN707pfMrsP70543Csfu5LLHr0s5eMVZRexqGQR/oCfXc278FnfkG3yM/P5f2f/v4Qhc6waRadrrdNUWDMlE5tCm4iIiExZo6lWeDwsnrmY/Mz8YY8gjWQ/ay3ePi/72vaxv21/pMhHXXsddR11kfVkiQwEBiLfZ7uzmeeZR11HnePoXH5GPvUb6iNr2spuLouZdhmWSugar0bRGkGTicJY69w48Hhavny5feWVV8b7NERERGQKiS6oMfiX/NEU6kinRGGmIKsg4Zq2ePvlZuTyowt+xMHOg9R11FHXXkd9Rz0N3gaO9hwd8blmubL48Zofc8rsU1hYvJD5nvm4XK6Uf87puB+d/Z2Rio4TsSeayGgYY1611i53fC3V0GaMcQOvAA3W2o8bYxYCDwAlwHbgn6y1/caYbOBe4IPAEaDaWrs/0bEV2kRERCSdRhqGxkOqYabf3x8ZIXuv7T221W7jgV0P4AsMnXKYjNu4meeZFymDX1VURUVhsOpia28rX936VXp8PSmHq1TDVDpC17I7l/HaoddYOncpO768Y9jXLjJRJQptw5ke+TVgN1AYenwjcIu19gFjzJ3AeuCO0NdWa+0iY8xnQ9uNbKKxiIiIyAiMVaGOsXBW5Vk0bmxkyW1LaPQ2MiN3BhtXbqTmzRpu+cst1HUER8oOdR2Ke02D5WXmUVZwrC9ZZVElVcVVkQIflUWVZLoz4+5/0fsvioSrssKypOEq1fL4KqMvMjIphTZjTDnwMeC7wAYTbHSxGvjH0CabgX8lGNouDH0PUAPcZowxdiLMwxQREZEpqbqmmj0te1gyawlb1m4Zs0Idyd43Hmstjd5G9rVGrSfrqI2ZutjW2wYE195d98x1Cd93Zu5MygrLgn3JQs2iq4qrWFi8kBNnnMisvFmOfclSNZJwpfVfImMn1ZG2HwLfAsIrd2cCbdZGSgDVA2Wh78uAOgBrrc8Y0x7aviUtZywiIiKTXrKwk2oYguBUyBfqXqC+oz5SbGSkBT6Ga0/LHl479BoBG+DtI2+z7+g+9rfv50DbAQ60H6C+o566jjoavY30+/tTPm5FYQXlheWUF5ZHSuBXFVdxQvEJLCxeSF5WXlrOP5HhhrBUKygq3IkMX9LQZoz5ONBsrX3VGHNu+GmHTW0Kr0Uf91LgUoDKysqUTlZERESOv+EEqFT3DYedeJxedzpW9HowgMaORspuLqPmohpcxuV4bKdqhcmusb23nXdb342sJzvQdoC69jrePvI2ADubd7LkttRCiCfLEwlkkVGyoipu+NMNvNv6LqfNOY3XLo//szlexqqMvcrjiwxfKiNtZwKfNMasAXIIrmn7IVBsjMkIjbaVA+EasfVABVBvjMkAioAhpYqstXcBd0GwEMloL0RERETGRrKANVb7JjuWt8/LmvvWxBQbCRDA2+9l7YNrqbmohrUPrk1aIj5gA+w6tIs3W97kaM9RvvPH73Cg/QC17bU0dDRQ31HvWNAknjn5cyKB7I3mN+ge6GZB8QJ+dMGPOHHGiRTnFjvu99jex/BkeTQCJSJDJA1t1tprgWsBQiNt37TWXmyM+S9gLcEKkuuA34V2eTj0+C+h15/RejYREZHJaSL3OUtWbKSuvY7GjY2cfNvJ1HvrKckt4bq/vY5H9jzCbS/dFinw0eRtivQfq+uo49vPfTvue2a5sigtLKUwu5A3m9+MaRadl5HHI//4CKsXro48t+zOZext3cvs/Nl8sPSDia9HI1Ap0fRKmY5G01z7auABY8wNwF+Be0LP3wP80hizl+AI22dHd4oiIiJyPAyeIhhv6uFY9zlLNSgmKzZy3TPXcc3T19DSHVxW39LTwoY/bEj43gbDSbNOCpbAL6qgquhYxcWFMxYy1zOXrv4uym4uiwlsAN2+bj71wKcmVDuBqUjhVqajYYU2a+1zwHOh7/cBZzhs0wt8Jg3nJiIiIiM0knVo0dMPE009XHPfmjELJoODYkNHA/P/Yz7fP//7kcbQdR11nH/v+exq3pXwWM1dzUOem++ZHynuUVFUQSAQ4K5X76LX3wsEQ1t9Rz13feKuuMF0MrUTEJGpYTQjbSIiImNmNMUvZPRryVINJt4+L6vvXU1tWy2VxZU884VnhjV9MmAD7D68m32t+3ir5S2uffrayFRFAIula6CLrz721chzR3uO8vR7Tyc8bobJYMPKDbxv5vv43rbvsa9135ACH+EG3OHABqkF0+PVTkBEJEyhTUREJqR0FrCQ5AZPSXyj+Y2kwSR6VCxAgJbulpjpk9ZafIHgFMK23jZuev4m9rftZ1/rPt48/CYQrLp4yk9OSekcDQaLJScjh4v/5mIWlyym19fLjc/fSK+vF4uNKTYSHin78Us/Du4/qG/ZSEfMUm0nMJHXA4rI5KLQJiIiMg4m0kii09q1O165g5yMHHp9vUO2z8/Mp7ywnAt+dQGdA52R58OjVOf+4lyqiqto8jbR4+sB4ED7Aa5+6uq45+A2bvIy8xJWabShDkL9vn4efOPBSDDb+OGNnHL7KdR11FFWWMabV7wZM0IWr3DFSEfMqt9fzYYnnNfGhdsJjNd6QBGZmpwbmIiIiIxAdU01y+5cRnVNdfKNpxhvn5e7t9/N1U9ezd3b78bbl7hEfHgkcU/LnlG972h/5tFr1wIER50CBOjz9zkGNoAeXw/XPXNdTGCL5rd+9rXuiwQ2CIaZxSWLcRu34z65Gbl877zvkZ+Zn/Sco6cwdvZ34snysLJiJUvnLmVlxcohUxq3rN3Cji/vGBKOwyNmThI14C7ILmDrxVspyCrAFfpVyoWLgqzg89Zax59p9DmLiAyHQpuIiKRNuoLIZLOtdhtlN5fx5d9/mZteuIkv//7LlN1cxrbabWP+3iP9mVtrOdh5kE3bNtHn6xvWvgEboKOvI+E2qxes5r5/uI85+XOAYAGQr37oq+Rk5DifDxaDidsQO955bNkVDGLxglki1e+vHlYD7mhnVZ5F48ZGygrLACgrLKNxYyNnVZ6V0rRLEZHh0PRIERGRUXCqsui3/oTFLNK11inRccKhobO/k3v+eg/7W/ezv30/T+x9gva+dnwBX9xgEU+WO4tPnfQpTpxxIvUd9Ty460H6AkMDX35mPitKV/DlR78cmR7Y5G1iwxMbYoqMROsa6KK+o56tF2+NWSeXyGiLfoRHzKLfL14DbieeLA8luSXUddRRklsS2V6FSkQk3RTaREREQkayzmy4xSxSWeuUynk8vvdxPr3l03T7ugGo76in5MYSTiw5kbbeNg51HQLg3dZ3ueThS1L7AQyS487hhtU3cO9r9/J68+ucPOvkyPl4+7z89q3f0tc/NLQZDLe/cnvMNMAAAQKB+CEsPB0xPIIVXqM2I2cG/f7+pEU/Rmrw+zmtiRuuVAuViIikSqFNRESmpcHByNvn5YW6F6jvqB/W6NdwRlVS6X1mreX52udp8DbQ6G3kZ9t/xsGug+xv209tey117XXUdtQ6rovyWR97jgydJmkwVBZV0tDRENMQOi8jj1/971+x7rfrHAuAZLozuWz5ZbzU+BLGmJhCHtGjVN0D3fitP1JM5PIVl3P7S7cn/dlFi56OGD2CVV5Yzv62/Un3GY14I2apcCpykkqhEhGR4VBoExGRCed4lEqPbikwmkp/wxlVSTQq1z3Qzcm3nUyDtyFSJfFw92HWPzL6Js0Wy4H2A0Pf09fNut+uo+aiGtY+uDbuFMF4o33hUaotu7aw9+heFpUsovrUav7tj/8WN8hCcJqlz+9LaTqiy7hGNYVxrDn9bEY77VJEZDCFNhERmVDiBagVpSs40nNkWFMX4zV+jg6FLd0tXHDfBUOm8g0e/dryxhbeOfIOi2cupvr91ZEQmWhUxW/9/OjFH3HNU9eQnZHNQGAgbpjxWz/13vqE15Ofmc/VZ17NK42v8PDbD6f0M0gmYAPUtdeNeIqgJ8szpJdZsiB74/k3cuPzNyZ8r+gRrLGYwjjWJuM5i8jEpdAmIiITRqLpg88deG5YhTPiNX7edP4mrnnqmmOh0NsYGdkaLGAD3PDHG/jJKz+JTAF04eJrj32NmXkzae9tx5Pt4cyKM3ly35P4rT9m/15fL683v57S+ZrQ/xIV3+gZ6OEHL/yAy1dcztPvPZ1wNCtV4Smco5kiOFiy6YHrlq3jp9t/mvC9BgfzdJ7f8TIZz1lEJiaFNhERGXfePi9b3thCzZs19Pv7HbexNhiskk2XrK6pZvfh3exu2Y0vcGz9Vjj8XbH1itjjxglsEAw0P3jhBzFBKkCAbl833R3BAiAd/R00ehtTus5MVybGGMdrzHBlxK2sOPgafvLSTzDGpPSeyYxFYQxNDxQRSS/1aRMRkbRwWoeWiugeZ0+8+wR9fueeYeFwFZ4uGa8H2p6WPexs3pmwUuFwJCs7H5bpyuT7532f31z0G/74xT86Nm0eCAzgNm48WZ4hTZm/sfIbKTWWhuDP4orlVwxp7uzJ9JCXmZfilQVFF8ZYMmsJS+cujSmqMVKJ+phNROm8dhGRdFNoExGRUQsHr8aO4IhTdLCqrqlm2Z3LqK4ZWjEvejrk4KmF8YRHm1ZtXsVpd5wWc9yj3Ucjpe5TDVvpkuXOYnbebP7h5H/g7SNvx93OZVxsOm/TkDBz/d9en3Jj6a6BLowxQ0JR0zebeOLzTwwJcwVZBdy+5nbH56NHvkbSoDqR8PRAYMj0wIkWktJ97SIi6aTpkSIyKiPpayVTS6J1aBf86gKKcopo8DbEjL6Fi3oc7DyYclgbzBfwsbN5J3uP7mXFXSvY376flu6WtFyTwZCfmc9XzvgKt790e0prx6LL+ydrA1DfUe+41inVxtLhKY1Oa6YSFcD4wtIvTJjCGNPl7wunlgAiIsOl0CaCgsdoRJdNl+kpURn7zoHOSHhp7Ghk7r/PxZhgwY2ugS4yTEZM37CR6PH18ErTK6M6BoAn0xMJmOWF5bx5xZtYa7nj5TtS2j96bdhImysPDlwG47jmbvCUxuivEL8AhifLw8qKlZS0lLBk1hKtLTsO9P8pIpIOmh4pwrHgsadlaFNakdFINDVwLI8Vb1un50d7jolGleDYWrQAAXp8PXQPdEe2H21gG45k0wQf+/xjzMqbBRybyhcuqBG9fdzjRwWp6vdXx53qmKy5cjhYLZ27lFULVqV9SuNkngY40aZUiogcLxppExEZQ+kYiQyPBB9oP0Bbb9uQ18OVF8M9xNYsWsMLdS9Q31EfU2nR2+cd8jwQ81xjRyMXbrlwSF+zRBbPXExeRh7dvu5RXedYS2WaoNOoVXj0a9XmVdS211KSU0K9t56egR781o/buMnLzIsJUqOtnhgdqDr7OyfMlMbxNhmDpohIOii0iYhMAImm6IaDX6YrE4gteR/uRRbpIWZcBGwAQ7AcfLggyJDeZFFTFXsHegFo6Gig7JayyJS8cF+zrRdv5dTZp3LO5nOoa6ujOLeYLy77Ik3eJva37Wd/2/60BjaDwZPlYa5nLnuP7mVRySIOdR5Kus4rHrdxU1pQysqKlTHTBJ2mD8YLBZ4sDy9/6eXI487+Trbs2hI5v+pTq4cEqWTNlVNd66ReXyIiotAmIjJGnErgO41aOY2ARW8XnkoY7uEVDmI1F9Ww9sG1sQVAQmvLoqckOvUmC09VjBbeZ/C+Z//87Jh1Va19rXz7uW+P4CcSX15GHnPy59Dt66ayqJJn1z3LWT8LlobPz8xPaZ1X3GNn5jmOTo2mQIQny8P609entF28NWTDGTVSMQsRkelNoU1EZAyER8CiR7bCo1ZnVZ4VmdL47HvP8pu3fkOfr89xu8aORt49+m7MscNh6pO//iQZrrH/azxeQHIbN5d+8FIWlSyiqqiK2fmz+VzN52jsbKSsoIz2vnY6+ztTeg+3y83Or+yMO4oUPdpkjIk02o6Wm5GL2+Wmu787pemIx2uqXTreR9MCRUSmN4U2EZE0S1QC//x7z+fCJRfy8NsPM+AfGFLuProHWWVRJfvb9scNTf6AP24j6uMhJyOHD87/YMyI0+z82TR2NjIrbxYPrH1gyJqu7IxsjDH0+frirgcLize6dMKME2KmS0aHs2Xzlmn91wSjUUIRkdFTaBOREUt1+t9IHK82DNHvAyR8T2+fl9X3rk5apCNRCfw+fx8Pvvlg0vPyB/zsa92XcJvjWXnRSXRfMifx1nQBSdeDwdDRpehf/u/55D1xw5nWf00sGiUUERk9hTYRGZFk0/9Gw2mN1yWPXJJyiEsl8A2uyBiwAVp7W+OuK4u+3gCBSJGOmotq+L/P/N+YIJesBH4qhrNma7zkZeTx612/5md//Vnk2geLt6YrlfVggw2+l+o3JiIi00XS0GaMqQDuBeYBAeAua+2txpgSYAuwANgPXGStbTXGGOBWYA3QDXzRWrt9bE5fRMZDoul/a+5bQ+PGxmH9Eh0dsq464yrHMBiuJJiKeGX2o98nvE1ORg4Abxx+g3BOCr/nitIVHOk5wokzTuTJfU86Xu/f/+rvceGKCXKXr7g8bmPldMpyZfGVFV/hP1/9T/p8fcGpgqHqkckYDDkZOZHqkQECkQIf4WO4jZvcjFwCBOgeGFodstvXTV17Xcy1z/XMHbLdWI20xDuupuOJiMhUk8pImw/YaK3dbowpAF41xjwJfBF42lq7yRhzDXANcDVwAbA49OdDwB2hryIyRSSa/hewAbbs2jKskZRwgArYQNww2NU6sgAULvixq3kXT777JK29rRzpOcKMnBlAcBpi+LwHv+dzB54jYAMc6TmSMAiFy9CH9/vB8z8Y85GynIwcPlz+YZ7d/ywXLL6ANYvWRKYbfux9H+Nj932Mvx78a9zz+MD8D/DHL/4RIDLNsLywnJe+9BKPvv1ozNTFHQd3xLQVcBt3ZC3e4Gsf6X1KJ03HExGRqSZpaLPWNgFNoe+9xpjdQBlwIXBuaLPNwHMEQ9v/z96dh7dVnnnj/z5HixfJS5w4sS3bSUnSQAKTEAIlMxRIChRCeeni4rZ0mulkoC1LCwnXQFvm/TG/gQLtNEBbWsoESmihGAJlKaFhSZhOhlKSQNLsxFksy3LiXZItW+vz/iEdRZKPNluyZfv74eoVWzqSjpTTJF/f93M/1wJ4WoZGe70vhCgXQlSHn4eIJoFk7X+p1jnFi14XZ3Pahg3mUKnTAuM3i062xmzANwDLegsGfAMxoavN2Qab0wbg9Bj9ZK854BnIqGqW68AmIHD8e8dx5e+ujFQU40Pyrm/tGtbSGT2wI7qFNbrNsMpcNey51LVp6jo0u8uOFw++qP2Z5H9X55TBiiMR0eSR0Zo2IcQcAOcC+CuAWWoQk1K2CyFmhg+zAGiNepgtfFtMaBNC3Ajg/BvYJQAAIABJREFURgCor68fwakT5YexGpiRT+ZPn5+w/c9kMGFexTzNx6lVryPdRzB/+nzUl9aj4YWGSCtk71BvwtdUg5DWZtFBBNHh7kD5g+W4dPalkccc6z2mWSFLN1Spx/V6Ep9XNhTrizEUGIJBMaScBqkIBZfOvhQmgynlEBg1bK3YuAJWhzWy/1l862o61230vmR3vnVnwhAbRBAzTTMZFPLAVPnziIhoKkg7tAkhzABeBHCblNIZWrqmfajGbcP+hSSlfBzA4wCwbNky/myWJqxE66cms8ZFjVi7Za3mfYpQ0Hh247Awq1Z9tFrsMpFos2gg1OK47cQ2nFV5FgBo7uU1Xs6YdgY6BzojIVNAwKgzomFhA1bMWYHGsxvx3L7ncNufbksYiKYVToP1dit2n9wdqiCmMQTGbDRjxw07svpeUoX2H6380YgGjRAREZE2JZ2DhBAGhALbM1LKl8I3nxJCVIfvrwbQEb7dBqAu6uG1AOzZOV0iygclBSXYfP1mlBhLoIT/GFGgoMQYul1Kifda38OeU3vwXut7sDvtkbVqalAbSWBLh4TEgc4Dka+zTWj+XCo1X8AH+zo7ltYsxUzTTJxXcx66/rULv/vi77Bm6RqYjWY0LmqEIhL/sVxbWgspZeSzjF9PtuqZVWlvZj0ayc5TDe1ERESUPSlDW3ga5BMADkop10fd9SqA1eGvVwN4Jer2b4iQCwE4uJ6N8pnWXmMTicvjwoYPN+DOt+7Ehg83jMn5uzwu3L7ldhToC6BTdAAAS6kF9nWhn89Y1ltgd4a+tjvtOONnZyRdOzZRFOuLR7wPnclgilS9Tt1xCjtu2DGsTTFRGFYDkiKUtIbA5Fqq0M7x+0RERNmVTnvkPwD4RwB7hRC7w7f9AMADAJ4XQqwBYAXw5fB9mxEa99+M0Mj/b2b1jImyKJd7jWUq3bVxWuPxo1sO125ZO+rzjx7wYSm1AAgN77CUWnDJ7Evwyx2/hC/oi6lkGXQGXPybi7H75O6Y24MIplynNVG4/W5cNfsq/Nn654wGk5gMJtzx93ekdazWhtTlheXY27EXQHaHwIxGuuvliIiIaPTSmR65Hdrr1ADgMxrHSwDDF5sQ5Zls7zU2WumsjYvedLrL3YXNRzbHtMMFZCDh+ScLYv987j+j1dE6bDiIOuBD1eHuwEcnP9I8t2O9xyL7fE1U6v5k6r5r8UwGE67+5NXY3ro94+fNpGXQbDSjoqgCrc5WVBSFpjoqQsGCGQtGPAQmF3KxXo6IiIiGy2h6JNFkko29xrRaK0fSPpfO8wyrCrrsCQOSev7XLboOTfubsO34Nrx06CV4/B5IyGFB7NY3bgWAEQ8HUeUisBUoBRCKgNfvjQzwGOlrqe9PhP+LDmYzTTNxUf1FONJ9BAe7DiIYHH5tKELB6iWrsbhqseYo/fiJltEj9kfzA4DoyqvL40o5BIaIiIgml7QGkRBNRqNtM9tu3T5s7ZZlvQXbrZlVYdJ5ni81fQmXPHVJzPCJZKFlwDeAbSe2wbLegm//8dt4dt+zGPIPpQw6uRoOMhpGvRHHv3c8Uh0ETg8DSXcoiE7oUFdah5qSGiyetRhfOOsLkecr0hfhnJnn4NI5l+LF617E377zN2xbvS3pei21NVB9DnU9303n3xQzbGRpzVLY19mz2mrL9WRERERTDyttNKGNZo+0+dPnJ6wsJWsza9zUiIOdB3Gw6yD8QX/k9nRaK+P3KVs1b1XKFk0pJbad2JawKqilSFeEZ/c+m9etijqhg5QyptoVXQGLrlJVmatwfs35aHO1ISiDkfcV/f6SVQmLDcU4cPOBmN+Txk2NkQ2l468drXVl8Y83G80xm1Kr92WrZTDZxsjpnB8RERFNHgxtNKGNZo+0VfNWJQxCidrMoteUJaryJGqt1NqnTK/okWjPw0AwgIbnG7Dt+DZ4g96M3ttgYDCj48eayWDCg5c9iKf2PAWrwwpfwIfa0lrMnz4fVodVc7DFVfOvwpajWzSro8X6Ynx50ZfhD/rx4sEXI62UydoTU4X8+HVlWoEol5sXZ+P8iIiIaHJgaKMpSQ1Q8QQEzEYzNl23KTQVr8+K+vJ6bP3GVuw5tSdmTVmiKtaAbwC3vnErHtv1GOpL63G09yjqy+rx+pHXY0JiQAYQCCRuR3T73dhydMso32l+UteG3XxB+jOLkrWzuv1uVJurcf9l9+Mx72NZq0Alq3YRERERjRWGNpqwRjoE5EtNX8LLh19OWGX77Rd+i4bnT09P7HJ3oeLBCgRkIO12w0H/IHbZd2GXfRck5IirgflMQECv6LF2+Vqs/MTKmM8sfjBHUAYx4BuAyWCCIpQRrb1Kd2piNitQuaykEREREaWLoY0mpNHsr/aB/YOk68O+8uJXMOQfinwfRDCj9WSqfF5PlolZplmwlFpgc9pQW1KLUwOn0OZqQ21pbUwVK9Eaq28s/gaa9jWhuacZ8yrmofHsxhEFqcZFjZyaSERERFMSQxuNi+gBIgCwpXkLhBC4Yu4Vw6obIxneoYYCrcc6hhwJz0tCIhDMvwmK2aJX9Ghc1Ijltcvx5O4nYXPaUKgrxMmBk/AGhq+bMxlMuG/lfTHr86J/7+IHc2hVuMxGc8qtE9KhTk3UGrU/Vacmsn2TiIhoamBoo3ERP0DE4XFEbo+mVtSi2+tuCd6ScHjHgG8A1T+tRlAGYVAMcHldEBCRvbnSqX75gr5RvLP8VaArwInbTqDKXAUAkfVkLo8LlvUWzdCmVcFK1jKY6xDBqYmx2L5JREQ0NTC0UU5pjeSPX4tWWlAaOT56bZrL4xpWUUs0iEIVlEH0e/s175ss7Yp6RY9Zplk4NXAKBboCDPgGUKwvhtvvTvgYs9GMN65/IxLYomWzgjUWIYJTE4mIiGiqYWib4kazz1kq0ePxu9xdWPrrpTjedxxOjxNShgJUm7MNNtgij2lztqH8wXKUGEsw6B/UrP5MZTNNM3H0u0dhNprR7+2PWStWV1aHhucbIlsKqM6cfiZ23LgjabiZbBUstg0SERHRZMLQNsWNZp+zZOIHhbS52tDmaht2XHz1S0JCShlpl5yMdEIHo84IIQQ8fk9kzzb1NgERqZ4FZAANCxtwuPswPH4Pzqo8K+laMfs6+4iHfiTaLHoiYtsgERERTSYMbZRVLo8Llzx1CXaf3D1p2hHTUawvxkzTzMi6u0/VfgqvHH4lssmzgIBRZ0TDwgasmLMisk4sPmBp3ZZJeBrt0A+GHSIiIqL8w9A2hY1kn7PodsoN12xA0/4m7OvYh97BXgz6B/Hax6/B4/dMqMBWoCuAP+iPaSlUCYjQ0BMZmlCZiE7RYe9Ne2MCVr+3P7RBt8OK+rJ6bFu9bVgA0wpY2Zi0SERERESTB0PbJBQ/5r5xUeOwMDaSfc5cHhf+1/q/aHO14Uj3EVQeqoQv6BvRHmb5wmw048XrXkTD8w0xA0+i7//41o9xze+vgdVhRUVhBWwuG6SUKTeLNhvN2HHDjrF6K0REREQ0STG0TTJqGFOHUeiEDmu3rI0JY1pTGeP3OZNSRqpoLx96GX1DfXB5XZGAlmxS4XjQCR2KDEUY8A7EVPkEBPSKHhISQRm7SXb0gA51emL051ZsKMbm6zejylwVE77iB4CMdLNoGjkOGiEiIqKpRKhT/MbTsmXL5M6dO8f7NCY8db8trYpRibEksun0Lz74Be548w54Ap5hx+mEDudWnYt9nfvgDXjzvooWHa6WVC2JTD80KAaUF5ZjdvlsbFu9DUDqtWIMY0REREQ0XoQQu6SUy7TuY6VtAotvgxzyD2muywIAt8+NLzz3BSytXoqH3n8o4QbSARnAzvaxC9ACoU2y46tj8d+bDCbUltaiZ6gHtSW1+Oel/wybwzYsXEVPP4wfqpFqrdhoh3gQEREREeUCK23jxOVxYeXTK2Hts6K+vB5bv7E1sqF09HCPiqIKLJq5KGZdmsvjwr1/vhcPv/8w/NKPoAwOCzr5ToGCIII4Z+Y56Bvqi1THrpp3Fez9drT0tUTWjKnVMla9iIiIiGiySlZpY2gbY9GByxf0RYKWTugwr2IejvYeRSAYiAlgOqGDlBL15fUoLyjH3o69CStq+USBAiFEpMVSQkbaGc+vOR/dg92RNUm52uCbiIiIiGgiYHvkOIhvXVw1bxV+8t5P8LO//kxzdHxABnC4+7Dmc6kB7UTfiVyeclYU6ArgC/pQVlCGuRVz015PRkRERERE2hjassDlcWHjno14+dDLaHO2YdA/CJvTBkUo8AV9E651MRNGnRGfX/B5FBmKML14OhbOWKgZyrhWjIiIiIhoZBja0hAdyk72n0RlcSVmFM9Al7sLJxwnYO2zJqyeAZg0ga1AVwBfwIdCQyFKjaW4+5K7sXrxalbNiIiIiIhyKGehTQhxJYBHAOgAbJBSPpCr18ql7dbtuOK3V2DQPzjep5JTilBixvsrQoFBMeDaBdemrKIREREREVHu5CS0CSF0AB4FcDkAG4AdQohXpZQHcvF62ebxe9DqbMWhzkNoeKFBcz+zyaJAV4D9N+3HuyfexYHOA+ge7GZAIyIiIiLKI7mqtF0AoFlKeQwAhBDPAbgWwLiHNiklHB4HWvpaYHVY0eII/9rXEmp1dFhxqv/UhG9p1As9bv3UrZhbMRfHeo7hQOcBvHP8HShCgSfggclggiIUbL5+M+ZWzMXcirnjfcpERERERKQhV6HNAqA16nsbgE/l6LViBIIBtPe3R4KYGspO9IUCmdVhhcvrGotTyTmd0EGv6NHU0ASb04bXj7wOSODqBVdrrjXr9/ZziiMRERER0QSTq9AmNG6LKV0JIW4EcCMA1NfXp/3Eg77BSPhqcbSEKmbO06HM5rTBH/Sn9VyF+kLUldZhTvkc1JfVY3bZbMwunx35esvRLfjen74Hb8Cb9vllqkBXACklzq0+F6XGUlSaK1FRVAGXx4XpxdNxRvkZ8AQ8+FPzn9DuakelqRLVJdWoMlcNa2G8+YKbk76W2WjmFEciIiIiogkmJ5trCyGWA7hHSvnZ8PffBwAp5f1ax6uba0sp0TPYE9Oy2OJoOR3OHFZ0ujvTPo/pRdNRV1aHOWXhUFY+G7PLZke+riyuhBBa+TLE5XGh5qc16Pf1Z/T+k1Gg4Lzq8zB/xnzN4EVERERERFNPss21cxXa9AA+BvAZAG0AdgD4mpRyv9bxZWeUyZq1NWh1tmLAN5DWa+iEDjUlNZEgFl8lqyury0oQymR6pIDA7LLZKDYUw1JqwZXzrgQAvH3s7aRti0RERERENLWNeWgLv+gqAA8jNPL/SSnlfQmPrRES34q9rdhQfLplMRzIoqtkNSU10Ctjs81cv7cfG3dvxCuHX4m0KKr7tHUOdKK6pBrXnnktAxkREREREY3IuIS2TMz65Cz5g6d/EFMpqyiqSNq6SERERERENFkkC21jU6pKoa60Dt+78HvjfRpERERERER5RxnvEyAiIiIiIqLEGNqIiIiIiIjyGEMbERERERFRHsuLQSRCiE4ALeN9HknMANA13idB447XAQG8DojXAIXwOiCA1wFl9xqYLaWs1LojL0JbvhNC7Ew0yYWmDl4HBPA6IF4DFMLrgABeBzR21wDbI4mIiIiIiPIYQxsREREREVEeY2hLz+PjfQKUF3gdEMDrgHgNUAivAwJ4HdAYXQNc00ZERERERJTHWGkjIiIiIiLKYwxtREREREREeYyhLQUhxJVCiMNCiGYhxF3jfT6UO0KIJ4UQHUKIfVG3VQgh3hJCHAn/Oi18uxBC/Cx8XfxNCLF0/M6cskUIUSeE2CaEOCiE2C+E+F74dl4HU4gQolAI8YEQYk/4Ovj38O2fEEL8NXwdNAkhjOHbC8LfN4fvnzOe50/ZI4TQCSE+EkL8Mfw9r4EpRghxQgixVwixWwixM3wb/06YYoQQ5UKITUKIQ+F/Iywf6+uAoS0JIYQOwKMArgKwEMBXhRALx/esKIeeAnBl3G13AXhHSjkfwDvh74HQNTE//L8bAfxqjM6RcssPYJ2U8iwAFwK4Ofz/eV4HU4sHwEop5WIASwBcKYS4EMCDAB4KXwe9ANaEj18DoFdKOQ/AQ+HjaHL4HoCDUd/zGpiaVkgpl0TtxcW/E6aeRwD8SUp5JoDFCP25MKbXAUNbchcAaJZSHpNSegE8B+DacT4nyhEp5Z8B9MTdfC2AjeGvNwL4fNTtT8uQ9wGUCyGqx+ZMKVeklO1Syg/DX7sQ+kPZAl4HU0r497M//K0h/D8JYCWATeHb468D9frYBOAzQggxRqdLOSKEqAVwNYAN4e8FeA1QCP9OmEKEEKUALgbwBABIKb1Syj6M8XXA0JacBUBr1Pe28G00dcySUrYDoX/QA5gZvp3XxiQXbm86F8Bfwetgygm3xe0G0AHgLQBHAfRJKf3hQ6J/ryPXQfh+B4DpY3vGlAMPA/hXAMHw99PBa2AqkgDeFELsEkLcGL6NfydMLWcA6ATwm3C79AYhhAljfB0wtCWn9VMy7pFAAK+NSU0IYQbwIoDbpJTOZIdq3MbrYBKQUgaklEsA1CLUdXGW1mHhX3kdTDJCiM8B6JBS7oq+WeNQXgOT3z9IKZci1PJ2sxDi4iTH8jqYnPQAlgL4lZTyXAADON0KqSUn1wFDW3I2AHVR39cCsI/TudD4OKWWtMO/doRv57UxSQkhDAgFtmeklC+Fb+Z1MEWFW2DeRWiNY7kQQh++K/r3OnIdhO8vw/BWa5pY/gHA/xFCnEBoacRKhCpvvAamGCmlPfxrB4A/IPRDHP6dMLXYANiklH8Nf78JoRA3ptcBQ1tyOwDMD0+LMgL4CoBXx/mcaGy9CmB1+OvVAF6Juv0b4QlBFwJwqCVymrjCa1CeAHBQSrk+6i5eB1OIEKJSCFEe/roIwGUIrW/cBqAhfFj8daBeHw0Atkop+dP1CUxK+X0pZa2Ucg5Cf/dvlVJeD14DU4oQwiSEKFG/BnAFgH3g3wlTipTyJIBWIcSC8E2fAXAAY3wdCP6ZkpwQYhVCP13TAXhSSnnfOJ8S5YgQ4vcALgUwA8ApAP8fgJcBPA+gHoAVwJellD3hf9z/AqFpk24A35RS7hyP86bsEUJcBOB/AOzF6XUsP0BoXRuvgylCCPF3CC0q1yH0w83npZT/vxDiDISqLhUAPgLwdSmlRwhRCOC3CK2B7AHwFSnlsfE5e8o2IcSlAO6QUn6O18DUEv79/kP4Wz2AZ6WU9wkhpoN/J0wpQoglCA0lMgI4BuCbCP/9gDG6DhjaiIiIiIiI8hjbI4mIiIiIiPIYQxsREREREVEeY2gjIiIiIiLKYwxtREREREREeYyhjYiIiIiIKI8xtBEREWVACHGbEKJ4vM+DiIimjrwY+T9jxgw5Z86c8T4NIiIiIiKicbFr164uKWWl1n36sT4ZLXPmzMHOndx7kIiIiIiIpiYhREui+9geSURERERElMcY2oiIiIiIiPJYXrRHEhERERER5ULjpkYc7jqMBTMWoKmhabxPZ0QY2oiIiIiIaNI63HUYe07tGe/TGBW2RxIREREREeUxhjYiIiIiIqI8xvZIIiIiIiLKqWytK5sM69NGgqGNiIiIiIhyKlvryjJ9HpfHhe7BbgBA92A3XB4XSgpKRn0eY43tkURERERENOlst26HZb0FdqcdAGB32mFZb8F26/ZxPrPMMbQREREREVHe06qaJTt21TOr4PK6EEQQABBEEC5v6PZ+b/+YnHO2MLQREREREVFey7Rq1rS/CUEZ1LwvKINo2pc/6+GklPAFfEmP4Zo2IiIiIiLKW9FVM1V01cy+zg6z0RzzmCPdRzDgG9B8vgHfAJp7mnN6ztH8QT/8QT98AR/6vf2wOqywOqxodbbC5rTB7rKjvb896XMwtBERERERUc6MdhhIOlWzNUvXxNw+f/p8mAwmzeBmMpgwr2JeBu8gsaAMwhfwwR/0Y8g/hFZnayiQOUKBrM3Vhvb+drS72tHe346ewZ4RvQ5DGxERERER5cR263asemYVBryh8KS2NW6+fjMuqr8orecYSdWscVEj1m5Zq/kYRShoPLsx5etKKSNVMk/Ag5P9J9HS1wKrwxoJZHaXPRLIOt2dCcNlPAGBGcUzUF1SjWpzNarMVdiIjQmPZ2gjIiIiIqKsG0lbo5aRVM1KCkqw+frNkcAYRBAKFJiMJmy+fjPMRjMCwQD8QT+8AS+6B7txou8EWh2tkbbFNlcbTrpOor2/HSf7T8IXTL7uLFp5QTmqS0JhLDqYWUosqCurQ01JDUwGE/SKPvK/ZKFNSCnTfvFcWbZsmdy5c+d4nwYREREREWXJhg834LY/3ZYwbD1y5SPD2hq1uDwuWNZbYsKfqsRYohn+1CpZ71Avljy2BO397agoqsC6C9eh0915ukoWDmRunzvt91VsKEa1ORzESqpiAlltWS3qSutQVlAWE8gMOgP0ih6KSDwHUgixS0q5TOs+VtqIiIiIiCjrsjUMRKtqJiBgMpjw+y/9Hke6j0SGe9icNthctkjL4knXSfR5+gAAPYM9+OG2HyZ9LaPOiCpzVahCFhXMLCUW1JaGAtn0oumREBYdypIFstFiaCMiIiIioqwb6TCQoAyGBnv4hmBz2tDiaEGrsxXfWfYd/GLHL+D2uaFX9CjUF+Ka318DifQ7B9WqmNq6WGOuQU1JTSiQldWhylQVE8jSqZCNBYY2IiIiIiLKumTDQCCBT1k+hW3Ht8HqjJq26GyDvd+Ok/0ncar/FAIyoPlwX9CHrsGumNumF02PWT9Wba5GTUkNHnr/IbQ4WrBwxkL8+Zt/HtMKWbYwtBERERERUVaowz263d044TiBtcvX4v7t98Mb8EaOERDwBr0457Fz0n7e0oJSVJmr0DvYC1/Qh5nFM3HDeTegtrQ20rZoNpo1A9lvdv8GLY4WGHQGTC+enou3nXMMbUREREQ0pTRuasThrsNYMGMBmhqaxvt0xlUmn4WUEr6gDy6PC1aHFSf6TkQ2iW5ztcHuDA32sLvsCdeyAYCEjJnEWKQvimlZVNeSqUM9ZpfNRnlh+YgrZAtmLIj5dSJiaCMiIiKiKeVw12HsObVnvE8jL0R/Fv6gH4O+wchQjxZHS6ht0RVqW1Q3ie4d6k37+Q2KATNNM9Ez2INB/yAqiipw6wW3Rqpj9WX1mFE8A0adMSaU6RRd1t7jZAjmDG1ERERElBKrU7Em2ucRlEF4/B60u9pxou8EWhwtaO5pjkxwPNB5ADU/rUH3YHdGG0RXmipj1pBVl1RHhnrUl9ajuqQaBboCfPo3n8bejr2oK63DPZfek8N3OjkxtBERERFRSplUpyZaoBmJsa7WJftMpZTwBXzocHdEWhYj4++dtsh+ZKf6TyXcINoX9KHT3RlzW3lhOWpKamJaFi2lFtSX1aOurA61JbUoMhTBoBhSVsgmwrCPfMbQRkRERERJuTwudA92AwC6B7vh8rhQUlCS8Hi2H2aXP+jH/lP7sb9rP/qG+vDw+w/HtC3aXaFpi4P+wVG/VqGuEK999TV8csYnUVpQmlYgo9xjaCMiIiKihLZbt0c2NQYAu9MOy3oLNl+/GRfVXzSu5zZeFb1MQ2wyQRlEv6cfLY4WtDhaIoM9bA5baLiHyw67yw6X1wUAaHG04PYttyd9TqPOqN2yWFqHurI6vGd9D/f9z31w+93DHqtTdGhxtOCyuZeN6P1QbjC0EREREY2Bidgy6PK4sOqZVZHAAABBBOHyhm63r7PDbDSP2/mNR0UvkxArpYTH70GrszUSyKyO0J5kaiBr729Hl7tL66USEhCYP30+5lfMj7QsxqwjM1fHbBAdXyF748gbmoENAAZ8A5F1btk0GSY4jieGNiIiIqIxMBFbBpv2NyUcShGUQTTta8KapWvG+KxCRlrtGk2VLFmIvfzpy3HnRXfiVP8ptLna0OZqQ7urHR0DHQk3iNZSWVwZqZBNL56OFw+8CG/QG3OMhITdaccH//IBygrL0n5u1fzp82EymDTH8psMJsyrmJfxc6YyUX5Qka8Y2oiIiIhGYSJW0NJ1pPtIwv22clWRScdIWzYzeVwwGETXYFdo0mJfqEr2RvMbcPu0K1RDgSH8+3//e9LzLisoiwSympKaSIWstrQWs8tmo76sHsWG4kiF7De7f4NXD786LLQBoeC26cCmEYXmxkWNWLtlreZ9ilDQeHZjxs9JuZV2aBNC6ADsBNAmpfycEOITAJ4DUAHgQwD/KKX0CiEKADwN4DwA3QAapZQnsn7mRERERHkg3ytoowmV41GRSWWkLZvJHnfZ05fhO8u+g46BDthctsg6skQBLZHygnIsmrkoEsrqyuoie5HVldVhWuG0jIZ65Co0lxSUYPP1myMBNoggFCgwGU3YfP3mcW15JW2ZVNq+B+AggNLw9w8CeEhK+ZwQ4jEAawD8Kvxrr5RynhDiK+HjGNeJiIiIxsFoQuVIKjIjaT/MJFim07L59b/7emQvspa+FlidVmw9tjVhAPIEPHj4rw8nfE2DYsAs8yzohR6tzlbNdsciXRHu/8z9uOG8G7I2ZTGXofmi+otgX2fHwkcXotXZCkupBQduPsDAlqfSCm1CiFoAVwO4D8BaIYQAsBLA18KHbARwD0Kh7drw1wCwCcAvhBBCSimzd9pEREREE0c2pw2OZTtmphWZkbYtZhIsD3ceTlp9+u6fvosbXrsBEpn903OmaSbOqz4PlhILLKWWyFCP+rLTG0QP+gZR/3B9TLVOpdfp8fXFX8/qWPxctzGajWZUFFWg1dmKiqIKBrY8lm6l7WEA/wpA/dNlOoA+KaU//L0NgCX8tQVAKwBIKf1CCEf4+MzG4hARERFNApkGmVShLJOAk42wmG5FJhuTJqWU6OjvwPG+4zjhOBGZtGhzhsbftznb0N7fnvQ54lsapxVOQ01JDYIyiI8jYWMGAAAgAElEQVS7P9askhXri3Hvintxw3k3JH3uAn3BmLYVso2RVClDmxDicwA6pJS7hBCXqjdrHCrTuC/6eW8EcCMA1NfXp3WyRERERBPJSIJMttbIZXN/tXQqMum0LX554ZdjApm1L7Qnmbo2a2/HXsz66awM3+lpBsWAe1fei/kV81FfFqqSlRSUwKAY4Pa5YVlv0ayS6RQdvnrOV9N6DTXErti4AlaHFfVl9di2elvOAlSu2xg5in9iSKfS9g8A/o8QYhWAQoTWtD0MoFwIoQ9X22oB2MPH2wDUAbAJIfQAygD0xD+plPJxAI8DwLJly9g6SURERJNOtkfmp1s5y8X+aqn+cX+w82DStsWbNt+Ef3ntX5K+RnRLY4GuILRBdElVqGWxxBLaHLq8Dr3uXtzx5h1w+92QkDHVp0SBNJtVK7PRjB037Ej7+NHKZRvjZJt4OlmlDG1Syu8D+D4AhCttd0gprxdCvACgAaEJkqsBvBJ+yKvh7/8Svn8r17MRERHRVJTN6X+ZVM4yCYvprJHzB/348WU/DlXJ+k7gP/77P2B1WkNti87QJtFqmEzEGzg9tl6v6DHLNAtV5irUlNTgvdb30D3YjbrSOvzui79DfVk9ZplmRTaI1vL1xV/PuPo0kYdvsCI2tY1mn7Y7ATwnhLgXwEcAngjf/gSA3wohmhGqsH1ldKdIRERENDFla/pfppWzTMLioc5D+FvH3zDgG8Bz+54LtS06Qm2LaiA7NXAqYQhMh1ExYv2V6/HJik9idvlsWEosKNAXQK/o4fK4sPCXC4HBUKXt3Kpz01p3ZzaasbxuOSq6KrBgxoK0g9dEHb7BitjUllFok1K+C+Dd8NfHAFygccwQgC9n4dyIiIiIsmo0kxe1HpuqXTHT6X+Jni/TNstkYVEv9Hj72Nv48xN/RpurDVaHFQDQ3NOMr76Yel1XeUE5qkqqUGM+vTl0XWkd6svr0e3uxk1/vCnttsXRrrtjkKGpYjSVNiIiIqIJZTRDPuIfm07gyGQdVbLnS1U529K8Bd6gF62OVlgdVrQ4WhJuDO2Xfuxs36l5X7GhOLIxtKXUgtqS04Gsvqwes8tmo6ywLGHLIgB88awvptV+mIt1d0STFUMbERERUYYyCRzprKNK9nxXPH0FvrjwizAoBviCPs3zeeHgC3jh4Aspz1tA4KzKs3DmjDMhIPDa4dfgDZ5eayakwFOffyrj6ZLR0m0/zPaQFqLJjKGNiIiIJp1cb0CdaeCIDzJF+iK0OdvQ4mjBib4T2HRgEwZ9g5rPNxgYxDN7n0l6PopQMNM0E5YSS6hKVhLaHLrSVIkfvvNDdLo7YSmx4NAth2A2muHyuGBZb4kJbAAw4B8YsypXNoe0EE12DG1EREQ06Wi1QY5mo+n4x+7v2J80cPzt1N+wy74LJ/pOoMXRAqvDihN9JwAABzoPoPC+QviD/rTfT5G+CNUl1Wjpa4nZHLpIV4Qnrn0CXzjrCyjUF2o+9u1jb0cCrBrE8qHKla0hLURTAUMbERERtCszua7W0NgZzcCL+Me2Odvw6I5HoVf0CYPXzz74GX72wc8074tvcSwxlsBkMKHD3aEZpIp0Rfj5VT/HmqVr0O/tj7RZ1pXWpTWuXuvazYcqV6ZDWoimMoY2IiIiaFdmRjO0YqJLFFgnYpDNZP3ZkH8IrY5WtDha0NLXgo+7P8b6v6yHX54OZxIy4dqyeIX6wmGDPerL6jG7fDZml83GnPI5mFY0LdKuGH2OKr1OHwkw2RpXnw9Vrmxudp0p7nlGEw1DGxEREQ2TKLDmQ5DNNDgmawUc9A1i+RPLoQgFNqcNPYM9ozq3AqUAP7j4B/js3M9iTvkczDTNhBAi5eMyCTDZCBz5UuUar82uJ8oPHIhUDG1ERESUltGsCcsmreAYlEF0DHSEqmR9Legc6AQAHO87jru33p2wFdAv/djXsU/zPgGB8sJy9A71JjyXtReuxQsHXsioXTGRdANMNgLHeFa54k3Uza6JxhJDGxER0SQ10lZGrXC259SeUW2CPBpSSvQO9Ub2IOtydwEAWvpa8PdP/D3aXG042X8S3oB32GOdHiecHmfC51aEggtqLsCH7R8Om6QoIeH2uZO2ES6sXJjVwDGWASaXVS62HxJlF0MbERHRJDWSVkatgR01P61BEMGYzZqzuQnygHcArc5WtDpacbzvOKwOK6wOK1qdrWhztsHmtGHQP3wcfp+nD3+x/WXY7SaDCZ6AB/6gH8WGYnxzyTex4cMN8AQ8msde/3fXY2/H3mGhDQB0QhczrTGa2kb48w9+PoJ3ndhYBp5chUS2HxJlF0MbERERAUg8sKPf15/wMUEZxIqNK+AL+DQreg3PN2B/x35UlVThn5b8E6x9VjT3NOPlwy9HgmGiUKTFqDMiEAwgIAMo0hdhzdI1mDdtHurL6jGnfA5O9p9E46ZGBIOhNWxDviE8vedprP/setz19l2arYCvHX4tYfuk2+/G9edcj1cPvzpmbYQMPEQUj6GNiIhoEsjGVMdkAzsSGfAN4GjPUfQO9aJjoAO3bL4lpkrW6Q6tLTvUfQjvnng36XPphA7TCqehZ6gn5jwKdYV45KpHUG2uxtde/BrcgVDFz+P3YOPujZEWTZfHhUueukRzSuRdb9+Fj2/9GBf81wXDWgEPdR1K2gK5Ys4KPPa5x8Z8WAYRkYqhjYiIpjytNVzq19G3JRu6Mdaj8ONfLxtTHZPt3ZWMOqijvb8dj+54NOFx04umo3eoVzMYFuuLceiWQ1j0y0XD7h8KDGHdlnWAQEzVL75FM9WG0a9//LpmK2A6kxQn87AMrj8jyn8MbURENKVpreGa9Z+zIITAkG8ocluqoRtjPQo/+vVGEjodQ45IRczqsKKlrwXvtrwLRSgZV9tUBboCfHr2p1FbWov60nrMMs3CPf99DzrdnbCUWHDXRXeFWhQ1gqEQAndvvTvha3uDXghoj84PyiCa9jWNeMPofJqkOB7yoR2TwZEoOYY2IiKaFLQqXak2iJ47bS7eOvbWsHa6+KEXox26kU4VbqSVugHfACzrLTGhc+ZPZgIi1D4IADanDRUPVmBx1WL0e/vR5mpDvzfxOjUtAgJCiGFti49f8zh+/L8/xr7OfThzxpl46x/fAjA8DLe72rF2y9qEm1IP+AZwuPtwwtClNRky+rHNPc2j2jB6vPYLo5B8CI5E+YyhjYiIsmasWwSjaVW6Um0Q3T3YnVFVSa3orFm6Ju3HuDwuvNf6HmxOW8I2y3SOUfkCvpjQdbTnKCTk6XNEEEOBoWGP80s/drXvGna7QTGguqQalhILLKUWKFDwyqFX4AmGAp+AgMlgwhtffwNLqpZEQk30nmQ//ctPh70frYEm6nAQLSaDCQumL8C+jn2aocuoM0JAJJwAOa9iHq5bdF3KNsdkkx7NRjOW1y1HRVcFFsxYMCywJaoGsUpERLnG0EZERFmTyxbBXARCr9+b0RquZC12WrRaL+PbLOOPaXO2oeo/q/D9T38fJoMJLY6WSBujzWnDqf5TMSEt+utU9EKPaxdciy8v+jLmlM/Bff9zH1r6WnBm5Zkxn2m/tz8SzmpLa2MqTvHrurRaM0cy0EQRCh647AH84dAfNO83KsZQ9VAjtEWvO0vV5pgqYCW7thLdxyoREeUaQxsRjYvxrMjQxBS/hqtpf2gN0/zp87Fq3qq0h4ZEhwxf0IdiQ3HM/mPJJGqxS7SmTKva5PK6cPnTl2OGaQa63d3DWjElJNx+N/5t27+ldU6Z8Es/5k+fj8azG+HyuPDRyY9gc9rQM9QT85klG7oRHXoShdJrFlyTNAwbdUb4A/5hoaq6pDpp6AKQct1ZqjZH/nlDRBMRQxsRjYuxHtow2U2GEJzsPUSHonZXO2rW12DQN4iADEQGZ6hDKqKrWfHiQ0bfUF9GlSq1oqP1nGq7YpuzDZU/qcTS6qUJw+BQYAg2py3l61WZqrBo5iLUltZidtnsyF5kFUUVuOb316DN1QYBkfZ7UENnOhXARBUp9ffG5XHBst6iGUpfPPBi0rVlD172IB783wc1Q1Wq0JXOurPJPOmRiKYmhjYioklgMoTgRO8hPmB0uDti7lfb8NTgogaHq353FcoKywCEql92p31Y5Ss67ChQItWbAn1BZHpkEKFAWGwoxn0r78NLB1+C1WGF1WHF8d7j2Hp8K4IIxjynJ+DBX2x/GfVn8k9L/gn3X3b/sM/jkqcuiXwemYbOVfNXYcEvFmiGrehBK6nCf7IWSJ3QJdwwWxEKVi9Zjf/68L8Shqpka8sYyIhoKmJoIyKagCZiZc3lcWHl0yth7bOivrweW7+xNem+Z+pj4oNWuvp9/ZFKj91pxxk/OwNCaI+ML9QVwqAzwOUNtQh+5eyvwOa04c2jbyIYDEJCYsA3gO/+6btpv74I/xcd6DKh1YqZzuehFTqj2whfP/J60r3M0h20kmy8vtvvxvXnXI9XD7+ak7VlRERTDUMbEdEENNEqa9HVsiCC6HJ3pdz3DEhezUlHdPVNa4CFaigwFJm46PA48Otdv9Y8TkCg0lSJ2tJa1JbWotXRio9OfpTwtQ06Q9JR9clotWIm+zzUNkm1ZRCAZhvha4dfG9FeZvFSjddfMWcFHvvcY1xbRkSUBcp4nwAREeVe46ZGLHlsCRo3NaY+OMuiq0Nq1Sm6HS/RfmGNmxrxw3d+mNF0x2wo1BfiyrlX4oalN6DKXAUAmDttLo599xiG7h5C863NUISC91vfR89gD4oNxZrPYzKYcPunbkeJsQRK+K/bRJtDR1OgoMQY2uxZSokNH27AnW/diQ0fbsD+jv0JPw8JiZmmmVhetxxmoznSRgggpo1QDVuJzjnZXmbRGhc1QhHa/4yInuaodQ5ERJQZVtqIaMKaiC2CWkb7PrQmF0a3HabaAyzXn2Oy6pDajnfdouti3oPdacd26/Zh69dypdhQjE+UfwL7O/djwfQFeOPrbwAAljy2BCf7T8JsNOMT0z4xrGKYbAiIIhTcfcnduPuSuyPVppqSGjg8Ds2gGl8p231yNyzrLXD73AjIAHRCB72iR6G+EEP+4fuwmQwm/Gjlj1K2NjYuaky5l1k6SgpKUo7XB8ZvDzPunUZEkwlDGxFNWBOtRTCR0byPAd8ALOstCacApjMlMFufY6LwmGztk7pO7MY/3hgJdjanDZaHLKM+Hy2FukJAhPZniw8ZP//g59Ar+oT/yNdaT5ZokEl8cFEHZ8wonoHnGp7TDDqzzLPQ3NOMiqIKSCmHvVZABhAIaA/3ANIPXOmGrXSkmvQIjF8b5ET+QQ4RUTyGNiKiMZCratax3mMxVazotsPDtxxOuE9Y9JTAaPH7nzUuakw6LEQ9ftvxbXjp0Evw+EPrxtQNou+66C4c7zsOvaKHP+jXfI5090iLZlSMuPqTV6PUWIq/q/o7fOXsr+DCDReGgkOJJWE1y6Az4ONbP8YF/3XBsJChtbYuulqTrGJYrC9Ggb4AvUO9msEl+nnUoLNi4wpYHVbUl9Vj2+ptWPPqGpgMppSvlSx4phu40glb6RqPaY6sohHRVMPQRkQ0BrJZFYyuaEmp3ZoXlEEsf2J5pMKmdX/8lEC1ahfdjrd2y9qYqpyUEt2D3Wh1tGLL0S2459174A14h7UIqhtE/993/2823jKA0+2DdaV1mgEjnWrW5us3o8pclXbIiA7Yd751Z9JpieaC01W1+OeMD+pmoxk7btgxotcaCgxh7YVr8cKBF0YVuJKN1c93rKIR0VSTMrQJIeoAPA2gCkAQwONSykeEEBUAmgDMAXACwHVSyl4Rmqf8CIBVANwA/klK+WFuTp+IJqJUa7Amm2y+3/h2x0TrqQZ8A5BumXDcvDolMPrcjvYcjXm+gAzA5XXh07/5NCqLK1FWWIY2ZxsG/YMZnXNZQRmKDEXoGOjIyiTIdKo5qSpJI6nUpJqWWKArSPu5RvtaCysXZqW6xfBDRDQxpDM90g9gnZTyLAAXArhZCLEQwF0A3pFSzgfwTvh7ALgKwPzw/24E8KusnzXRJDSe0/3G0nbrdljWW2B32gGcXmO13bo9J683lp+ry+OKmfTn8rhSvt/o0NTl7sIvPvhFzOPjnz9+CmMixXrtiYYqAYENH27A9B9Ph81pA5B8k+ZOdyeae5ozDmwA4PQ4cbL/JAQEFKGg2FCccOpgMkqKv7IWzFiAxbMWR4JYssmFTQ1N2P3t3RmFllTTEs+3nB/z+qORzmTGdMR/JkRENDGlrLRJKdsBtIe/dgkhDgKwALgWwKXhwzYCeBfAneHbn5ahnp33hRDlQojq8PMQUQKTZahGMlqDHFKtsRrtWrBEn2u215ipFbDo1sLb/3Q7ggjGrNmKfr+brtuEhucbIlWzNlcbbn3jVgDQbE3MZM8ytz/5OjEJia7BrrTfn1Ex4oalN+DJ3U9iyD+UNOBpvRYQqtwBgC/gG1HFTVEULJqxaNw2Y041wCPZfnPZfq10K2uspBERTQ4ZrWkTQswBcC6AvwKYpQYxKWW7EGJm+DALgNaoh9nCtzG0EU1x6Yx+jx9Xnih0jbblMFVIjh7IUVdWBwBodbRqDufQCqMBGUC/T3v/MQAIBAO49rlrNUe3q49XWxOXzFqCTddtwp9P/HnM9yxT+YI+PLrz0aw8lxACBbqCpJtdA6fXsOUqGI1ENgd4pPta8QNLJtLaMyIiyo60Q5sQwgzgRQC3SSmdoaVr2odq3DbsR7JCiBsRap9EfX19uqdBRBNYqtHvzT3NaT1PojH259ecj+7B7kj1LFGwi7/9S01fwtHeo5HHxVfNosVXwBo3NeK91vcynoDo9rtTtvupdp/ajXk/T2/D41RKjaWoK6vD/s79GT0uk8paKt6AF0adMeH9lhILdt64U3PCY6ZyMWVwLKclag0sUXGCIhHR1JFWaBNCGBAKbM9IKV8K33xKbXsUQlQDUHdAtQGoi3p4LQB7/HNKKR8H8DgALFu2LHv/GiCirMjFiPpUwxXmVcQGE63QBSBhi+W7Le9GKnmJgt0Dlz2Au96+K+b2l10vIyiDCMoglv56KXaf3J0wpKgVMHWkvrpp9UikWpeWCbUqVVlUic7BzoTHVZmr4PA4Yh6TzXMAQgFvWuE0eAPehL/Xt5x/C36585cxrX8QoYrrjOIZGU14TGYytwdO5vdGRESxUv6YNzwN8gkAB6WU66PuehXA6vDXqwG8EnX7N0TIhQAcXM9GNPGo7YOHuw5n7TkzGa6QaIDHvf9zb8IWS3X8fZe7C1c9c1XMwA412N28+eZht6vPt69jX9LAFs3ldaFmfc2IA1u2KFBQYizBijkrsHjWYsyeNhsmg0nz2EJdIaxOa+QzzVZgU0ToHM6rOQ9nzzwbAFBbWpv09/ruS+6GfZ0dltLQJtqWUgsWVS6KOS6fh2jk87mNB34eRES5lU6l7R8A/COAvUKI3eHbfgDgAQDPCyHWALAC+HL4vs0IjftvRmjk/zezesZENGGlO1wh2cCSh/7yEHxBn+bzqyHE7rKPKJBks+qUa2qVLL510OVxwbLeovmYoYD2+jmV+nvxnfO/g0c/eDTt9XMmgykyRGbJY0tCzyWUtH6vo/cJi/8BQT5XkvL53MYDPw8iotxKZ3rkdmivUwOAz2gcLwHcPMrzIhoTuWgBnGxO9J3AkseWaH5GLo8LK59eCWufFfXl9dj6ja1Jh4Gon/flcy/HjrYdCdcrJRtY4gv6Urb15UP4EhDQCR1WfGIFPl3/aTy28zHY+4d1io9Yoj3LEgVjg84AIYTm4JP4ACilxK92JN6tRYGS1jTDdAZpRF9TauAjIiKiWBlNjySabKbCmP2RiF5L1u/tx55Te4ZNaIxeMxZEEF3uLljWW7Dpuk2wOqyRqYtPfvQk2pxtqC+vh8fvwd6OvQAwbL1SdIDec3JP0ipPPoSyZBbOWIhjfcfgC/jw1rG3sPX41sg5q4FHS3lhOe6++G784J0fwBvwJn2NZM+jNeGwYWEDHnr/Ic3jtQJgoirZpus24V9e/Ze0B4QkG6QRj4M1iIiItDG0EeWB0Y6vzyY1jPV7Q+Pq1emJbc42WNZbsPn6zVg8a3HC9sXP/u6z0AndsKmLXe6uhDV7l8cVGejRPdgdWZs2ERXri3Gs71hMRUv9LBQoUBQFwaB22KorrcO65evwKcunIr8HCQOqgMZc3tPiJxwurFyYcAiMVgBMNto+V5MTWe0mIiLSlt68aSLKmUQDN7Zbtw87tnFTI5Y8tgSNmxqH3ZcpreeKXksWHxYkJFxeF1ZsXIGzHj0r6Yj7+MAGxA786HJ3Rb4e8A3EvP82Zxvsruy1EWbD4lmL8bVzvgaDYkh5rLqxtpYiQxHWXbgOJcaSyDE6oYsM7FB/VQPTeTXnYVrhNChCiWwPoA4euXT2pRkNfkg6BEZRcM7Mc4Y9l9loxvK65Vg8azGW1y2PBDQOnSAiIhpbrLTRlJUP1a1kAzdWPbMqMtxBPTa6GpXp+UavP7OUWtAx0IE2V1vMcyVbS6YKBANoc7WN7A2HtblOB7P4vdmy2fpo0ptQXlSONlcbRPi/IIIxXycSv6Gzy+PCa4dfg8+rPQRFPf6aBdfg2b3Pah4z4BuAEAL2dXY07WtCc08z5lXMw+bmzTjaczQmBEW3FfZ7+0e0wXJ0u2GqITCJNq3Wqn6xIkZERDS2GNpoSkq0h1eyf7wmMpphJslCUlAG0bSvCWuWrkl5vokGgrg8LjTtb8K249vw0qGX4PF7ICHR4e6IvE70cyXb/FqVrVCVzvPohA5CCPiD/oyfv8QYCik3vX4T2lxtkOH/1NeWkCjSF0Gn6OD2uiNhTr0/fr2WVujROv65fc/hlUOvJN2Lzmw0Y83SNZHbo7/Wksm6sGjx12OylkciIiLKX2yPpCknuroVv4dX9FqudFsRR7OfWbKQNOAbQHNPc8rzffPom7Cst+BD+4focHdgl30XKn9Sict/ezlm/udMfPuP38az+57FkH9IMyhFP1ddWV3CPb5yQSQcTBsSkIGE67+SmVY4DfZ1dlxUfxEK9AWR1sJ4ilDwwGceiOwVVltaiy+c9YVh7YAqNfREH6/uSxYZqJLBXnTjQV2TBgyfPElERET5iZW2SYzj7LWlW91KZ7LkaFss50+fn3A4hFqVSXa+Lq8LV/7uypgwJiHhCXjw9rG30z4PAPD6vbhn2z1p782VDRISBsWQcN81AJGgqggFkIhUuJJV6mpLayNhZOWcldhp36l53IBvADanLWawxovXvZj0nNV1XureYuq5qV+nuxfdZMUJkERERNnH0DaJjcc4+3wLilrnk051Kx3JWhZ//sHPNT8HtV3xSPcRzJ8+H6vmrcJasVbz+dWqzH/893+Myfh7T9ADz5AnK8+VrmJ9MYIIJg1tQGi92JJZS6AoSmRd130r70PD8w0IyiAGfAMwGUzwBDyYUz4HZ1WeFXlsOsE4U6mu7XT2J5us8uH/90RERJMNQxtlVT7texY9uKPL3YWlv16KNmcbigxFKDYUa04/VP8Rn6qClmqAyJzyOZH9yFRqyHP73JEJg8WGYjxw2QO46+27ElZltp7YmnRPrlxLVdVK5vIzLsdn53024b5jOkWHP1z3BzQ83xDzWcYLIogr5l6B+y+7P+b2+IEejWc3DgtGjYsasXZL8mD85rE3AWS3OjTSdWhERERE8RjaaFKKr4K1udoiEw+ThRBPwIMfbf8Rvv36tyNrqexOOyp+XIE55XOwtHopmhqakrYsBoIB2Jw2AKHAZ3fa8bnffw67T+6Oed2ADMDldeGut+/Cx7d+jAv+64LIcIj9N+1HQAbwfuv7ON57fNwCGwAU6gsxr2IeDnYdzGggiMlgQuOiRqxZuiay71iiqYX2dXac/cuz0eJoSfhcWhWx+IEeWtJpV5xq1SG2MBIREU0sDG00IUW3GW49sRUevwdnVZ6FpoYmzSpYtOjgpFawFCgoNhSj0FCIY73HYo4PIohgMIjmnmYYdUZs+HADHtv5WMKWRbffDbc/VMVrc7bB8pAl6XvxBDz41mvfigTMjoEOVP6kEp7A2LYqJjK7bDb+9p2/xQTh6M8siKBm1TJ66EaqqYVmoxl7v7MXlvUWzd+30Q7w4NTEWFMtpBIREU10DG2UNclaCrO51k0ND+paJrVydmrgVCTMpdprDAitpyrQF6B3qBczimfA7Xej292d9DEHOg/gW3/8VlrPD6S33swb8OLVj1+NfJ+rsKYTOgRlMDSevsSCAd8A+ob6kj7GZDDhjr+/A0Di4LP75G7Nts/4oRvq1EJ14Ed8YMr1AI9Ur09ERESUrxjapohcDwhJtY9Ytta6aVXR1GDUMdCB6T+ejnOrz01rAqLb74ZO0QEAuga70g5i6R6XCUUoEBAIyMCIn0MNr/GbSBt1RjQsbMCKOStiNnF+9/i7aZ1XdIUrfnKi2WiOhLlUa8vSkeuKGNsCiYiIaCJiaJsicjkg5EtNX8LLh1+OCTPRQzns6+xZe61UVTRf0IcP2j5I+/nU8CdldiYwjlSmQdCoM8If8MdUo86vOR/dg92YWzEXVodVc2ph9Pqvuofqkr5Gob5Qs8KlFfrTWVsGpBeatIJhtrAtkIiIiCYihjYCkF4lzuVxYeXTK2Hts6K+vB5bv7EVJQUl+MD+QdJ9zzbu3jiivcy0xuP/29Z/y8k+Ytkamz8WTAYTHrzsQTy156lRjZOfVjgtMjAlniIUHP/ecVSZq7JxyhHphiaGKyIiIqLTGNomqfj1ZdMKpyU9Vh2NnyhUxQ+h6HJ3wbLegk3XbYJjyJHwuQd8A1j75lr4A6Gpg/Ftk1rBbHPzZmw7vg0vHXoJilDg9rlRqCvEDYEbsvDJ5A9FKCNqtVSEgtVLVuPmC24e1eufVXkWBv2DONF3AsFgcNgasmwHNiIiIiIaGTHebWEAsGzZMle5QTMAACAASURBVLlz587xPo0JSatCpjXlDyJU9Vo8azF2f3t35PFax6r/aF88azGa9jdhX8c+/HrXrzHkH9I8h5Hs41WgK8C1C67Fqx+/GhPMhgJDKNAV5M3kxGxRA5o6pOOaBdfg2b3PZvYccWPys6Xf2x9ZQ1ZXWjelpyoSERERjRchxC4p5TKt+1hpm+Di16ol2vRZzVQtjhbc/qfbsWjmIlxcfzEu/+3lMWFMXYt2xW+vgE7RYdA3mHI4xkhaCz0BD54/8HzMbUOBoch92fb5BZ/HmZVn4pH3H8Ggf3DY/YW6QkAAXr83El4VoSAgA6Nunbzl/FuwpGpJzJCO5/Y9h1cOvZKw1TM65EUmPuZoTH0u15ARERER0egxtE1AanVt7rS5w9aKpRrU0TfUh4f/+nDK1jytYDNRmQwmfO6Tn8N1i67Dox88qnmMQWfAx7d+jGt+f03MOjF1nL1aiRQQEEKg1FgKb9ALk8GEYkMxutxdmgHMZDBhSdWSYUM6Ghc1Yu2WtZrnYjaY8cDlD8DmsGFexbyYiY+5ClRcQ0ZERESUv9geOcG4PC4s/OVC2Jw2CAgAoUqX2jo3kra7iaxYX4yLZ1+Md46/A1/Qp3lMibEE9nV2mI3mpO2giVoO+739WLFxRcKhHy6PK+Gm0NGvHU89F639zbLZ/khERERE+S9ZeyRD2wSi/iO/39ufsGVPrQTlYi+xfKSGIgC497/vxUN/fQi+gC8myMaHoFQhbCRGGsD6vf1Z2d+MiIiIiCY2hrZJIFk1ZypShAKTYWwCWboYwIiIiIhopDiIZAKIHn1fVxba9Li5pxm9g70wG83Yad+Zk/3Jsk2v6KFX9AknTSajrrNLtd7OoBjw0JUPYfXi1cNCkdloxo4bdmT82tmQ7gbTRERERESZYGgbQ/F7kjUuakRJQQnebH4T1zZdi0AwkHBd1kTx7fO+jY17NmII6Ye2Al0BGhY2YMWcFbj6k1fj9Y9fR3NPM6SUeHTno3B73WmvPyMiIiIimmzYHpkjakDb17EPHf0dONJzBB+d/AiKUOAL+qCDDgBwRsUZONJzZJzPNjtMBhMeufIRLJixYNiwDxn+T6UIBTqhw9rla3H3xXcnbCMcz3ZHIiIiIqKxwjVtWRAfwjrdnZCQmFY4DX1DfTFf9w71Ys+pPZBSptzjLB/ooEMQwZT7kZkMpqQtmtGTEuPDVn1ZPY50H0GBvgCXfeIyrvkiIiIiIorCNW2jpLYv+gK+CRHCMlVsLMam6zah4fkGDPgGYtaTFeoLYVAM2HTdJrQ6WmPaFqWUGPANwGQwQREKNl+/ORLCxnNtGRERERHRZJKz0CaEuBLAIwB0ADZIKR/I1Wvl0pbmLbjymSvH+zSySid0mmPp7evsaNrXhAOdB9A92I3pxdOxcMZCzYrY3ZfczUmJRERERERjICftkUIIHYCPAVwOwAZgB4CvSikPaB0/nu2R/d5+tDpa0epsjfnV6rSipa9lQq43ExC4dM6l2G7dDkUo8AQ8kWpYdMWMYYuIiIiIKD+MR3vkBQCapZTHwifwHIBrAWiGtlzx+D2wOW2agazV0Qqb04beod6xPKWcim5TvKj+Iu4bRkREREQ0CeQqtFkAtEZ9bwPwqegDhBA3ArgRAOrr6zN+gUAwgPb+9tgw5rBGAlmrsxUdAx1pP59O6FBdUh0ZmlFXWocP2z/EO8ffyfjcskGv6OEP+qETOggpcOkZl0JAQC/0uGzuZSjQF8DmsKG2tBYQgM1hGxbMuG8YEREREdHEl6vQJjRui+nDlFI+DuBxINQeGXcfOt2dw9oWrQ4rrA4rWp2taHe1ZzQUZJZpFmpLa2NCWV1ZXeTXanM1dIou5jEbPtyAv7T+BW6/O+3XGQ290GPd36/DbRfeFtmrjBUyIiIiIqKpLVdr2pYDuEdK+dnw998HACnl/VrH1yyokZc9cBmsDitsThvaXG0Y8qe/OfO0wmmRQFZXWhf6NSqQWUosKNAXZPw+XB4XLOstcHldGT9WpUCBgMDSmqWYUTQjUiU71nMMJ/tPotPdCb3Q4+oFV2P14tUMZ0REREREU9CY79MmhNAjNIjkMwDaEBpE8jUp5X7N42uExLe0n8tkMKG2tBZ1ZXWoLw2FsfhKmcloyvp7UG23bseqZ1YhEAzA7XdHNsVeWrMU5QXlKC8qR99QHyAR83WluRJV5qqE0xeJiIiIiIhUYz6IRErpF0LcAmALQiP/n0wU2ADAXGDG58/5fEx1TK2YlReWQwitbsuxET0Kn+2KREREREQ01nJSacvUeI78JyIiIiIiGm/JKm3KWJ8MERERERERpY+hjYiIiIiIKI8xtBEREREREeWxvFjTJoToBNAy3ueRxAwAXeN9EjRp8fqiXOG1RbnCa4tyhdcW5VK+X1+zpZSVWnfkRWjLd0KInYkWBRKNFq8vyhVeW5QrvLYoV3htUS5N5OuL7ZFERERERER5jKGNiIiIiIgojzG0pefx8T4BmtR4fVGu8NqiXOG1RbnCa4tyacJeX1zTRkRERERElMdYaSMiIiIiIspjDG1ERERERER5jKEtBSHElUKIw0KIZiHEXeN9PjSxCCGeFEJ0CCH2/T/27j0+7rrO9/jrO5PJ/dY0vaS5NKXUQmGlWwFl4ayAeAG8HqrxgMLuoiwCu0LRlT3Hc/bykCO4CngWlEVkAcG1WkRRqoCCy1ZUoKVcS2ltm0wubZo2l8l9Lt/zx8xvOpPMTGaSSWaSvp8++kjym9/8fr+Z/DB55/P9fr4x22qMMU8ZY/ZEPi6KbDfGmP8XuddeMcZsyN2VS74zxjQaY54xxuwyxrxujPl8ZLvuL5kRY0yxMeZ5Y8zLkXvrnyLbVxlj/hC5tzYbYwoj24siX++NPN6cy+uX/GeMcRtjXjLG/Dzyte4tyQpjzAFjzKvGmJ3GmBcj2xbEz0WFthSMMW7gLuBCYB3wP4wx63J7VTLP3A98YMK2m4BfW2vXAL+OfA3h+2xN5N9VwLfn6BplfgoAN1prTwbeBVwb+f8n3V8yU2PA+dba04D1wAeMMe8CbgVuj9xbvcCVkf2vBHqttScCt0f2E0nl88CumK91b0k2nWetXR+zHtuC+Lmo0JbamcBea+0+a+048APgIzm+JplHrLXPAkcnbP4I8EDk8weAj8Zsf9CG/R6oNsbUzc2Vynxjre2y1u6IfO4j/AtQPbq/ZIYi98hg5EtP5J8Fzge2RLZPvLece24L8B5jjJmjy5V5xhjTAFwM3Bv52qB7S2bXgvi5qNCWWj3gjfm6PbJNZCaWWWu7IPyLN7A0sl33m0xLZMjQnwJ/QPeXZEFk+NpOoBt4Cvgj0GetDUR2ib1/ovdW5PF+YPHcXrHMI3cAfweEIl8vRveWZI8FnjTGbDfGXBXZtiB+Lhbk+gLyXKK/5miNBJktut8kY8aYcuAR4Hpr7UCKP0Lr/pK0WWuDwHpjTDXwKHByot0iH3VvSVqMMR8Euq21240x5zqbE+yqe0um62xrbacxZinwlDHmzRT7zqv7S5W21NqBxpivG4DOHF2LLByHnPJ75GN3ZLvuN8mIMcZDOLA9bK39cWSz7i/JGmttH/AbwvMmq40xzh97Y++f6L0VebyKycPCRQDOBj5sjDlAeMrJ+YQrb7q3JCustZ2Rj92E/+B0Jgvk56JCW2ovAGsiXY0KgU8Cj+X4mmT+ewy4IvL5FcBPY7ZfHulm9C6g3ynni0wUmdfxXWCXtfa2mId0f8mMGGOWRCpsGGNKgAsIz5l8BtgY2W3iveXccxuBp621efvXaskda+3fW2sbrLXNhH+netpaexm6tyQLjDFlxpgK53PgfcBrLJCfi0b3fmrGmIsI/xXIDdxnrb05x5ck84gx5j+Ac4Fa4BDwD8BPgB8CTUAb8HFr7dHIL+F3Eu42OQz8pbX2xVxct+Q/Y8w5wH8Br3Jsbsj/JDyvTfeXTJsx5u2EJ+u7Cf9x94fW2n82xpxAuDpSA7wEfMpaO2aMKQa+R3he5VHgk9bafbm5epkvIsMjv2Ct/aDuLcmGyH30aOTLAuD71tqbjTGLWQA/FxXaRERERERE8piGR4qIiIiIiOQxhTYREREREZE8ptAmIiIiIiKSxxTaRERERERE8phCm4iIiIiISB5TaBMREcmAMeZ6Y0xprq9DRESOH3nR8r+2ttY2Nzfn+jJERERERERyYvv27T3W2iWJHiuY64tJpLm5mRdfzNu17ERERERERGaVMaY12WMaHikiIiIiIpLHFNpERERERETymEKbiIiIiIhIHsuLOW0iIiIiIiKzoWVLC7t7drO2di2bN27O9eVMi0KbiIiIiIgsWLt7dvPyoZdzfRkzouGRIiIiIiIieUyVNhERERERWZB8Yz6OjBwB4MjIEXxjPiqKKnJ2PcFQEH/Ijz/oJxAK4A/5GRofomOgI+XzFNpERERERGTB2da2jYsevoih8SEAOgc6qb+tnq2XbeWcpnNm5ZyBUAB/0I8/5I9+fnTkKG39bbT2t9Lh66DL10WnrzP6r3uoG4tNeVyFNhERERERWVB8Yz4uevgifOO+6LYQIXzj4e2dN3ZSXlie0TGttdHqmFMpGwuO0THQgbffS9tAW1wY6/R10jXYxcDYwIxfj0KbiIiIiIgsKJtf30zIhhI+FrIhNr+2mSs3XDlp+8RKmW/MR1t/G20DbXj7veEg5usKV8wGu+jydeEP+dO6Jo/Lw4qKFayoWEFdRR31FfXUV9bTVNlEU1UT5/xj8uqfQpuIiIiIiCwoe47sYcg/lPCxIf8Qr3W/RvdgN4eHD9Pa30pbfxsdAx10+jqjgazT10nPcE/a56wqqooLZCvKV9BQ1RANZSsqV1DoLsTj8lDgKsDj9uAy6fWFVGgTEREREZFpyac10AKhQLRStqJiBSXuEkaCI5P2cxkX33/1+9yz4x6G/cNpHdtlXCwtWxqulJWHq2WxVbLmRc1UF1fHBbICV/ailkKbiIiIiIhMy1ytgebMJ4udU9Y/1k9rX7hK1tbfRudgZ3T4YvtAe8LABuFhkN3D3XHbiguK4wLZiooVNFY20ljVyMrqlTRWNlLqKY0LZOlWybJBoU1EREREJItyWX2ay3Nns52+tTau4+J4cJxDQ4c40HcAb78X74A3rvNih6+DvtG+jM/jNm7eu/q9vKPuHTRWhgNZU1UTy8uW43F7ooEsm1WybMivqxERERERmefmqvqUy3Nn2k4/ZENxa5MNjw/jHfCG55P1teH1eeNa4XcNdjEaGE3rWgpcBSwrW0Z9RX10PllDZQNN1U0sKVnCFT+9goODB6mvqOeVz71CdXH1nFbJskGhTURERETmpXyaT5UP5moh6VTt9C986ELeuPYN/CE/B/oOhIcvDoSbfDhdF9Ndm8xR6ikNN/aINPloqGgID1usWsnK6pU0VDRQVFCUtEq2rGwZBwcPUltaS01JTVbfi7mi0CYiIiIi81IuK1r5ZrYXknaGLQZCAe7dcS+BYCDhfkP+Id5259vSrpIBLCldEtd1saGygaaqcIOPVdWrWFyymMKCwvB8MpcHY8yMX898o9AmIiIiIpIlc1XtmnjOmSwkHbtodCAUYHBskNb+1mgrfGd9MqfRR/tAe9I10Cw2LrB5XB7qKuqizT3qK+qjgWxl1UpWVq2kvKg8GsjcLnf23piItbVr4z7ORwptIiIiIiJZMNvVrmSmWkj6B6/+gMvXXx5eNDrop2ekh/29+6OLRrcPtEe7Lnb6Ojk8fHja11LgKqDllBYu/ZNLWVm1khUVKygqKMpplWwhDJ1VaBMRERERmaGZVrtmYnfP7pQLSd/2+9v43qvfi84py3RtsrryuuicsqaqJpaWLuXzT3yekcDklvolBSXc/cG7Z+21Hq8U2kRERCQr1BRCjmdTVbs2v7aZKzdcOa1jB0PB6NpkA2MDHOg7EG6FP+Clrb+N57zP4cJFiMTn39Wzi109uyZtLy4ojgayuoo6Gqsaw23wI8MWV1avpMRTEl0wOrZKdvLSk6NVxRAhXLgoKyxj62VbFdhmgUKbiIiIZIWaQixsCuWp7TmyJ2W1a+/RvUmfG7s2WfdQN/v79tPa14p3wIu33xvtuNjp66R3tDej63Lh4t3N76a5upmGygZWVoXXJVu1aBXLypZR6A43+Mh0Ltk5TefQeWMn6+5ah3fAS31lPW9c+4YC2yxRaBMRERGZA7kOPTM9f7qhPNevM1fWLF5DmacsYXArLSilsbKRnqEevANe9vftjzb4cOaTdQ6G55QlGnKYiLM2mdPcw2Vc/HzPzxkPjkf3KfeU87NLf8a7V757VuaSlReWU1NSg3fAS01JjQLbLFJoExEREZkDua5EztX5c/06syGT4OksGv3BNR/k+l9cn3Cf0eAo//zsP/M3v/ybpEMoJ3LWJqurqKOhsoHGyvDQRadK1ljZSHFBcVyVbHB8MFr5aqxsnJPK10LozDgfpB3ajDFu4EWgw1r7QWPMKuAHQA2wA/i0tXbcGFMEPAi8AzgCtFhrD2T9ykVEREQkr+Si3f1siA2eznyy8eA4HQMdtPa3hueT9XvxDhwbutjl62IokHh4ZMiGODR0KG7bktIl0XXJ6ivraaqMtMGvXklzdTNLS5dGF4tOt0pWXljOWY1nUdNTw9ratXNS+Tqeqqm5lEml7fPALqAy8vWtwO3W2h8YY+4GrgS+HfnYa6090Rjzych+LVm8ZhERERHJMzNtd5+rYZWBUCC6NtmB/gO09rdGW94f6DvAOf9+TrQVvj/kz+jYhe5CLjzxQlZVr6KxqpHm6mZWVoVDWXlhOR63B5dxZfX1KEQtTGmFNmNMA3AxcDOwyYTj/vnApZFdHgD+kXBo+0jkc4AtwJ3GGGOttdm7bBERERHJF9lod5/psMp0q3pOk4/Dw4c50HeA1r7wotFOkw+nwUeitcn6x/r5ffvvJ22vLKoMd12srKehooHGqsZolWxV9Sou2XwJrx1+jZNrT+Ynn/xJ2q9JJJl0K213AH8HOP8lLAb6rLWByNftQH3k83rAC2CtDRhj+iP792TlikVEREQkr8xmu/tEnKre4PggAO0D7Sz/+nL+6dx/oqywLNzkY8BLx0BHtMmHs+9UDAa3y00gFKCysJLL3n4ZTVXHhi6eUH0CNSU1KatkmXZiFJnKlKHNGPNBoNtau90Yc66zOcGuNo3HYo97FXAVQFNTU1oXKyIiIiL5Zybt7lOx1uIP+RkYjaxN1n+A3T27+af//KdJQxWHA8N88VdfnPKYRe4iVlSsYEXFChorG8Nt8KsjbfCrV1FTXMO77nsX7QPtVBZXcusFt87LeXmysKRTaTsb+LAx5iKgmPCctjuAamNMQaTa1gB0RvZvBxqBdmNMAVAFHJ14UGvtPcA9AKeffrqGToqIiIjkqamGIqZqd1/mKePEmhMTHjdkQ4wHxjk0dChaCTs4eJDPPPYZ2gfa6fB10OXrip47HaWeUlZVr4oOXWyqbqKpMhzImhc1U1del7RKtq1tG2ffd/a05+WJzJYpQ5u19u+BvweIVNq+YK29zBjzI2Aj4Q6SVwA/jTzlscjXv4s8/rTms4mIiIjMT+k0GGk5pYVNT2xKfAALb1/+dh5/63Fa+yPzySLrkzmhLHZtskNDh/juS99NeCi3cVPiKUk51PG6M67j1vfemvHrzMa8PJHZMpN12r4E/MAY8xXgJcD5r+u7wPeMMXsJV9g+ObNLFBEREZGZmG4r/qmCjPcGL/6Qn31H9/GFP/sCNz97M+OhY4s7GwzDgWHO/M6ZGV2v27h5z6r38I4V72Bl1crwXLJFJ9BQ2cDDrzzMDU/ckLSq97bFb8voXI5szsvT2mWSbRmFNmvtb4DfRD7fB0z6L9BaOwp8PAvXJiIiIiIzNJNW/A+/8jCBYCDhY4Pjg9R9oy6uSjaRndDWYHHJYuor6mmoaqChsoHlZcv52m+/xmhwNG6/oA3yu/bf8UjLI5OqW5889ZPc+OSNCc/nMi5aTp3eSlPZnJentvuSbTOptImIiIhIjqVa3yxVpezChy5k79/u5fDwYfb37qe1v5W2/jba+tvCQxcHOvAOeAnaYMLzWmxcYPO4PCwvX86wf5hAKMCysmVcc8Y10Tb4zdXNVBZVxi0Ufe+Oe8OdFhOcIll1q6Kogq2XbY0G0RAhXLgoKyxj62Vbpz2Ecbrz8kTmgkKbiIiIzNh0h9/lylwv5Jzu+5PpdfnGfDznfY72gfZJxw3ZEPfuuBd/MPGC0IP+cKVsYjUsXR6Xh0tPvZRPn/ZpVi1aRWNlIx63J6NjTLe6dU7TOXTe2Mm6u9bhHfBSX1nPG9e+MaM5Z6nm5c2kgieSDQptIiIiMiMzGX6XK5ku5DwTmbw/mVzX0/ue5kP/8SGGA8NAeK2ymq/V8PZlb2dofIgOX8eUa5M5gc1gWFK2hIbKBuor6sPrklU2sbRsKddsvSbhEMjigmLuvPjOGQWlmVS3ygvLqSmpwTvgpaakZsZNQmargieSDQptIiIix6lsVJvmY8e96VYFp/N+zeT9GRwbZH/ffvb17uNA34FjQxd97Xj7vXT4OiY9JxAKsKNrR1rXVugq5OrTr+bq06+mubqZEk9Jwv1OXHzirAWZfKtuzUYFTyQbFNpERETmgdkYzpeNalM2O+7NhZlUBafzfqV6fwLBADc/ezN/suxPaO0Lt8Lf17sPgNe6X6PilukNLy0wBVz8tov57yf/d5aWLeWSzZdEq3GxigqKuPk9N08ZSGYzyORjdSvbFTyRbFBoExERmQfmcjhfJrLZcW+2zWVV0B/009bfxpN7n0z6/owER7jlt7ckfCy2+YfbuFlWvoz6inoaKxtpqmri5UMv88yBZxI+N2ADnFx7MpefdjkAT3z6iRmHotkMMqpuiUxNoU1EREQyElv1e+8J7827jnvJqpLZrAoOjA2wvzc8dLG1v5UDfQfw9nvxDoSHLR4aPJS06+JEJQUl1FfWU1dex/au7Qz7h1lUvIj7P3I/pyw9haaqpkkNPu7dcS/Pdzyf1vs+H0KRqlsiqSm0iYiIHIdm0u0xtup374fuzas5SZC8KplOVbBlSwtvHn6TldUr+dLZX2J/33729+7HO+CNG7pYdUvVjK+zpKCEN659g5VVK/mt97dc9PBFjPrD65X1j/bzqUc/xdbLtrK6ZvWk52Y6F2yhhiItYi3HC4U2ERGR40w2uz1ONSfpyseunFT1mut2+45UnQrdxs0jux6htb+V8eA4r3S/ws/e+lnC48RW0ApcBdSV19FQ2RAduriyeiXN1c2csOgEVlWvYnvX9qTvT3N187SGbebjXLBcmK37R2FQ8o1Cm4iISI7NRohJdsx0A0Kq50+s0KUafnfd1usmVb1me35eMBQOVX2jfXzjuW9EF40+0HeAYf/khhwQDmJ7ju6ZtN2pUHUMdMSFtRJ3Cd/779/joyd9NLw4NOH3/Kl9T7G2di3XnXlddN+phidOd9imc9zzHjiPtv42mqqaeOaKZ46bwDab5vKPCSLpUGgTERHJsalCzHSGMiY7ZroBIdHzp6rQTRx+l+i6nc8zeS0Tr7HL18W+3n3R+WStfa14ByLzyQY6ooG0tb+VLzz1hSmP6TIuzm48m/XL17PljS10DXaxpmYNz3/2edzGTf1t9ZPmp40ER/jLn/4l7z/x/dHXmmyRa0g9PHEmzVzKC8t54bMvTPkaRWR+U2gTERHJY9MZypgq5KUTEJKFrakqdFNd97KvL8MYE523lei1jAZGOdB3gP29+9nftz+6Ppm330u7r50uXxf+kD/t989gePuyt3Pq0lNZWbWSpuomlpct5+rHr+bg4EEaKxvjql7Ptj5L12AXpZ5SqouruXfHvVOG3LW1a2c03HQmC0yLyPFBoU1ERBacXM2Zmo5EAekzP/sMu3t2s3rRap7a91RGc52mCnlTBQRrLfW31U96/ufO+NyU4SX260QBbyQwEv+8yGs59/5zWbdkHV2+LnpGetJ+76qLq8Nt8KsaWV6+nO+/8n3GQ+Nx+1gs+3r3se2vtsW9V//wm3/g4ODBuKpXou/FVCH39e7XueGJG2a0jEC+LTCdzGzP89I8MpHkXLm+ABERkWxzhvbt7tk9ree3bGlh/d3radky9S/Lmew70ba2bdTfVk/nQLhK5QSkHV07ePnQyzzf+XzaQQni56uFCD8vNkAMjg/SckoLLpP4x7/BcNeLdyV8/u2/u33KCp21FoCDgwcZC4yl/T4EbZBXu19NGNhcxsW5K89l01mb+OYHvsljn3yM1z73Gh876WOsrFrJKUtP4ReX/YKzG8+e1Bbfkei9Wlu7ltOWnRYNCMm+F5hwmE2kzFPGkZEjGX2PEnGailQUVuCK/GrmwkVFYUVeNRXZvHEzO6/eOWt/CJnt44vMZ6q0iYiITJBuo4yp5jFN9dxkww2HesPhaDwwntFcp6nmq533wHn4g37OWHEGL3S+MKnr4OfO+Bx3PX9XwucbYyh0FU6qZEE4YNy9/W76RvsAODx8eOo3YIJ3rngnOw/tZCwYH/ZCNsT2ru387NKfxYWXfb374r5Hmc4LS7c5y7ee/xbGmITHdRkXi0sXZ2Vx8fmwlpqI5I4qbSIiItOQrDKzrW1bWs9PFbAIF6woLChMWeWZONdpquDS1t/Gy4de5sjIETpv7KS+sh6A+sp6OjZ1MDQ2lPT548HxhIENwgHHCWzTUeYpY23tWgpcif+WPLFilWgYozPsM9nxU80LS/W9sFiuPf3apFWwdUvWTfu8EznNSoAFtZaaiMycQpuIiEiG0hqGOMWwyVQByznmouJFSYcyuoyLi9ZcxBnfOYNl/7KMM75zBo1VjUkDRGlBKW4Tbk1/ZOQIt267lcHxQQAODR1iyb8s4a4XE1fZUikwBVx66qXcddFdnFB9AgCnLDmFisL0O0JmUrFK1b6O0wAAIABJREFUFpabqppSvlep5oVNFXaNMZNCbueNnZzTdE7K4ab5NB9NROY3hTYREZEMpdM2f6p5dakqQ4bwcLze0V62fHxLwirPLRfcwto717Kjcwfdw91s79zO3z31d/iDiTsrDgeG6RrsAqB9oJ2v/NdX6B3tBcJVtInDEicqchfx1KeeYvd1u2moaACgsbKR3pt6efiSh7nmjGuiQ0MLXAUJ52iVFJRQ6imddsUqVVje+MONbPlE4vdqqnlh6VTpklXB5st8NBGZ3xTaRERkQZiqspVuw5Bk7e5jpdNRcKpjpKrQ2Mj4yM6BTjb+aCM//PgPWV6+HIDqkmquPeNarv/l9XHhxWIZCYwkHcIYq8BVwOl1p/Oxkz7G9e+8ntvfdzuPtjzKzr/eydZLEweQX13+Ky5YfQFvW/w2FpcuBlIP4XPmaG1YsYGlZUvZsGID3V/s5tAXDk27YjVVWPb2e5NWxFKZabUs0WtN57wiIulSIxIREVkQpmoekujxiUsDpLsmWqq2+cXuYr69/duMB8ajx1jxjRVcc+Y1YMPPbTmlJVqhcYZTOkEtllNFuvDhC6Pbjo4c5Zbf3pLyvVhcshjfuI/x4DiVhZV89YKvclLtSVz7+LW8eeRNTllyCi9clXhB5tOWnzbthhgTW7YnW/j5rMazqOmpYW3t2kkVK+f9j22Q4lSs0mk2kmoR62TSOfdUtMi1iMwmhTYREZlX65plU2yQS9VBcOJ6W6nW1RoNjsZ9HSLEoH+Qr/32a0B46OM1j1/DOxveybB/mKKCorhzTsVt3CwqWUTPcPL1zD674bP8Yu8vePnQy6xatIprzrgGgKKCorTOMZ3gA6R97yTbb6oOirO5CPVcdG9MZx0yrVUmIokotImISNot7ufSVEEy2eMhG4rO1cqkDX+67fKd802skhlMdIhd0AaTnsdi8Yf8aXeZdNSU1HB05CinLj2V6868jut/eX3K8DJff/kvLyxPWImD2V+EeqqwOtP3NJ1Qezz90URE0qfQJiIicfKl6jad4Y4Arx9+PdoyP3Z440QT56693v36lPPURgIj7Ovdx8YfbqTL10VlUWW0SmaxKcPaRG7j5uI1F3PJuku4f+f97OrZRfdgd3SOWqwyTxllnjKOjhwF0gsvV264Mu1rSSSXoS/ZfZfuMMbZuvb5Hqjma5AXEYU2ERGZYKqwNBuhbrpVtVhOlSy2WhadE/bQhVQVVwHhgPbk3ifZ+KON0Zb3HQMd3PnCnbiNO2nwGgmMAOAb9/HIrkcS7lPuKWfYP5wweE0UtEHWLVnH5addzuWnXY5vzEf9bfUJh0q6jIvq4mq8A14gO3OwppLqe5vLX/7TGcY438PVbNH7IjJ/KbSJiCxQvjEfm1/fzJ4je+KaX6QycWhh50AnW/dujR7johMv4jnvc7QPtCccephJoHP2Xb1oNc93Ph93TOf8zsddh3fxaverKV9r+0B70scH/YMM+sMBrX2gnfc//P64xy2WQCiQ8non8rg8fON932Bt7VpWVq2ksaqRYCiYNHhNNHEO1lRB7Lqt18U9fy7mYCWT61/+Uw2hFBFZiBTaREQWIKcL4rB/mKAN4jZuNj2xaVIXxIlihxZ2DHRQf3s9LuMiZEPRj84aYok6K6Y7N8435ouGv1cOvRLd3jnQybKvL8MYw6h/NLqt09eZ9FhHR46y7OvLopWwmTCR/6VTKSt0F1LqKeV9q98Xt31i8Eom0RysVEEsUXUrk4YhiZYySGeuX76areCoIYQiko+0TpuIyAIT2wXRGeoXtMFoF0RnSGDs/k4nwpANxa375WyL/RjdHhl6+O77380lP7wk7evb1raN+tvq6RjoiB4v9pgjgZG4IYYhQtFz7z26l/d/7/2cctcp0bDnHfBmJbA513L9u66nsbIRgIrC5KHGaTE/kRO8nLXClpYupbywHLdxA+G5bKkWXk62iPPmjZvZefXOaYUV5z3vHAiHXydwZ9oM5Xgwk/dZRGS2qNImIjLHZqvRh2/Mx/kPns/unt1JG2qEbIjNr22ONqlwKnITg1wmQjbECx3prU+VqK1+Job8Qzy578lpPTcdZZ4y1i1ZF61eVRVXEbKhjFvMTxy+990Pf5fNr21m79G9nFhzIi2ntqSsimVS7Zlq30yWMhARkfyk0CYiMsey3V6/ZUsLO7p2cKDvAKFQKOWQvNjq0EwDVKyx4Fjc8Lue4R7a+ts4OnKUtv42WvtaeevIW9z30n0MB4ZnfL7Z4gxZ/Nfn/xWARcWL6B/tT7lvMhMDeSbdHDMJ81PtO9VSBrEhXkRE8tOUoc0Y0wg8CCwHQsA91tpvGmNqgM1AM3AA+IS1ttcYY4BvAhcBw8BfWGt3zM7li4jMrmRVMWf70ZGj1JTU5LQ9/q7DuxIO00uktKCU/3jtP7jvpfso8ZRk1KI+laHxIWq+VhNt5tHh62DlHSuzcuyZqC6uJhAKJK0kunAl7LwYW7361sXfmtUujbNtz5E9KZcySPfeERGR3Emn0hYAbrTW7jDGVADbjTFPAX8B/Npae4sx5ibgJuBLwIXAmsi/dwLfjnwUEZlz0x2K6Dyvtb+VvtG+SY871bLiguJoG/Z0zEYzCKfbYzqGA8N4+71pNdrIRLJQkGuNlY1JQ9eWT2zhM499JmHnxYn3Sq66NGbDmsVrKPOUZTzEU0RE8seUoc1a2wV0RT73GWN2AfXAR4BzI7s9APyGcGj7CPCgtdYCvzfGVBtj6iLHERGZNYkC2nSHIjrP87g8QHzAig1ewVD6lSpn/tjQePiX50TdFye+hti2+G0DbbT1teEP+WmobODkJSezeeNmxgPjU57bYOKafeQrFy6KCoowxjDiH4lec6bHwBDteJmqI2O6nRfnc4v5dBbiFhGR/JbRnDZjTDPwp8AfgGVOELPWdhljlkZ2qwdi/+zcHtmm0CYis2piQJtJVcupSvhDfuBYwLrlglu46Vc3RYOX8/hUlaZ0mkFYa+PWQOsc6JzUFt8JMb2jvYwERvCN+aLXkIjB4HF7KHQVMhwYTjq3KRmn42HQBin1lPLZDZ/lwZcfjFb3YsPgTLiMi6qiKlbXrOaZK54B4KQ7T6LD15H2MZxrqa+sp7q4Om5dt2QBLZOGH/O1m+BcLMQtIiKzK+3QZowpBx4BrrfWDoSnriXeNcG2ST/RjTFXAVcBNDU1pXsZIiJpSaeqlYxvzMe+3n1x25yAde3WaxM+Z1/vPgbHB6O/ADudHNv62miqbiIYCkavZaKh8SHOuOcMOnwd0X2cNdKcNdESBaO9R/ey5F+WMBYcS/paLJbx4DjjwamrcUDc+U6uPZlXP/cql/740mj1744P3IG338tPdv+EkA1lJbAtKl5E2w1tk8JDbWktHb4OXMYFlmjYcCpxo/7RuACyrHwZe4/ujbbLnyhRQJuvQSxTuVyIW0REZi6t0GaM8RAObA9ba38c2XzIGfZojKkDuiPb24HGmKc3AJNWRbXW3gPcA3D66afP/Ke+iEhEyIZm1OL8gZcfIDzCO33WWh7Y+QDXnnltXGAMEaJnuCflkMQQIXYf2R0XgJzPpwpFqQIbQE1xDSfUnEAoFOKV7leijUJilRaUUlRQRO9oLw2VDdEqVaG7ELfLPSnYXLjmQp744xNZm8fWUNmQ8PvhhKvVNat5oeOFuLABTAog59yXOowfLwEtmUyGg4qISH5Jp3ukAb4L7LLW3hbz0GPAFcAtkY8/jdl+nTHmB4QbkPRrPpuIzLbYoZDtA+1JuyJO1eJ8W9s2Nj2xKeMKksWy6clNrFm8ho0/3DgpMKbz/GyqL69n5+d2UltaC4Tfn/rb6hO293e73DRUNtA72hvthOkyrqRDBlN1I0yX27ix2Oi8s0RiQ9b6u9dPChsKICIicrxI/JMy3tnAp4HzjTE7I/8uIhzW3muM2QO8N/I1wFZgH7AX+A5wTfYvW0TkmG1t26i/rZ7OgXBRv3e0l2F/4rXAhvxDfP25ryd8zJl3lmqOWCrjwXE+9P0PTfv52fTp0z4dDWxwbF5TRWEFZZ4yINw5sKIwvD02OG3euJmdV+9MWplyuhEm4gyvNAlHykORu4hNZ23i3z74b5yy5JS0X8/a2rWctuy0tOaeTWd/ERGRfJZO98htJJ6nBvCeBPtbIPGkDxEBpt+GfqHI5uufzgLRh4YO4Rvz8ZmffSbuOlItQpyu8dA4uW7OmKyNuzOvafNrm9l7dC8n1pxIy6ktk9Ylm0qqboTGGKy1rK5ZzaHBQwz7hwnaIG7jptRTGjen8Ml9T6as6MXK9D45Hv+7EhGRhSuj7pEikh3TbUM/UbrhJ59Com/MF9chMVlHx3SuuWVLC895n0taVUumf7Sf+tvqo40rnOv68q+/nLfrjWUiVRv38sLyhENDM7kvUnUjdN7TMk9Z0oA4nXMmMjFoZhI8RURE5hOFNpF5LFn4mxh4shUSZyqTjo7pXPPunt20D7RnfB1OUxLf0XB17vXDr1N5S2XGx5lNTvv6TFrqz2Ub92TdCGObgSQLiNkyMfTl+g8SIiIisyWdOW0iMs84gWd3z+5cX0pU7DBGpzFHbEfHwfHBuH0nrq+W6HjOPsnmTwFJm1zEStRRcbpi53Slc+6JXMZFRWEF71jxDrpu7KKhsgEIt8Wfah5ZfWU9nTd2TrmkQbY43QgBNQMRERGZRaq0icyxmSz4PJ3jJNtv4jpiT1/+dPT5Mx1O6Rvzsfn1zew5soc1i9fQckpLyvliY8GxhO3y4Vg1bssntlBdVM3Drz7M79p/x46uHdEOkakqUTOdo5ZMsbsYDIwHxgkRSlwRm0ZDSGdY4cQOiQ2VDRzoO5DwOc48snwIThqiKCIikn0KbSJzaLoLPk8MUekeZ8g/RP1t9XELNlffWk2pp5Rh/zDWWiyWw8OHWfIvS7hk3SWc13weuw7v4tXuV2f0GkM2xJB/iDJPGZ97/HOUF5YnnS82Hhznb3/xtzz65qM82/psXPdFpxr3/ofeP63rmS0et4e3/uYtzvzOmXgHvNHtseurpQqTE0PfVEMbXcY15TyyfKAhiiIiItmn4ZEicyST4YETn/ec9zlePvQyz3mfo3OgM+3j7OvdF7efsy7W4PggIRuKCxhjwTG+/+r3ufrnV/N69+sAdPm6uHfHvQmHJzrXdu+Oe/nSU1/i3h33xl2bE9CG/EMEQgH6RvtSvj8hQvx6/6/zol2+w4WLkxafREVhBW7jBsLrizlt8peXL48OD0w1RDMRj9vD/s/vp76yHkhvaKMzj2zic5INm5wLaq0vIiIy+1RpE5kjqYYHxi74HFtV+5sz/2ZSRa3xjkbCK2skPs4DOx+IDodMtl8qsYtSdw93c/0vr2fTE5s4Y8UZHB4+TFFBEec3nw/At178FtZahvxDFBcU87nQ52ZtOOJMFboLCQQDaS107SgrLOOFq14ASNkFEVIvoB177tiKmhP6Mlkg2plHFvucXA5JVGVNRERk9im0icyRPUf2JB0eOOQfig5vc5qIhGxo0vpjIUIp50kN+YfY9OQmAsFwY410uw6m4lzz0weeBsIVpRc7X5y032hgdMbnmi1lnjJuveBWbv3trXgHvFN2ZEw0VHGqLoguXAmD28RzO10WneNOFbjSCWQKTiIiIgubQpvMmbleKywb50t2jOkce83iNZR5yhIGt0SLIfeO9k6rajUeHM/4OZnIRhCcay7j4or1V/CdHd/BO+CNLvwcOzes1FNKaWEp3UPdk4JVWgwJA/XEc0+sqE11/yiQiYiIiOa0yZyZ6zb02ThfsmNM59gtp7QkbQHvLIYc2+mxf7R/QSz0PBdchNvk33XRXVQUVuCK/F+bs92pmDnzrzbUbZg0N6zrC12c23wupy07jbMaz0o7sDnHPHfluWmdW3O/REREJFOqtEla5rpKlotrmNgC/6ctP03aKj/Tlv1OC/wPve1D/PjNHzMWGMNi44bh7Ty4M27+WuywyIXG6ZzoD/rj5tBlwmn8YbFxlbHLT7t80oLPTgCbeN9MnBs2nfsq9jmD44Npn3sm1FZfRETk+KLQJmlxKkvz5RoyDVaxLfRDhDg8fJj62+ujwcBpqX/LBbdw069uyqhl/8QW+KUFpdHH6irquOmcm9jyxhb+bfu/5fW8sGxy2uU//tbjPHPgGR7Z9Qhu444uEeAyLko8JXQPdSd8foEp4LK3X8aOrh282v1q3JDDRI065spcnVtDJmW6FPhFROYnhTZJKrayla8SVd8yXQstthW/I7YVPhxrqX/t1mvjnhvbaj92QeRUxx4ODEc/7/R18sUnv8hocHbDmtN440+W/gl9o314B7w0VjZyRv0Z/PHoH1lds5rtndtp7W+dlfM7TTomdk68csOVXLnhSu4ev3tSd8Y/++6fJQ1tJZ4S7rzoTs65L3l7/HToF1g53ijwi4jMTwptktRMq2vZHM6YrHI28RoThSQnWF340IV89YKv4u330ljVCIC338vBwYPTHqLnCIaCXPv4tSwvXx499u6e3Tz0ykMphzla7KwGNoOhvLA8uviyy7g4q/EsanpqJn1fvvTUl/jac1/L6PhF7iKuf9f14VBoLXe9eBcj/hGCNojbuCn1lLLlE1v4X0//L9r622iqauKZK56ZFG7LC8sndWd05v+5jAssCRegnmnomo1fYBUERUREJNsU2mTWZGtIZbLK2ZZPbIkLcp0DnfzZfX+WNCQN+ge54Zc3ELCBuO1TtX9Px3BgmIdeeSijNcAy5cwDGw+MRwOMy7gI2mDS619RsYI3r3szriKVLKik6m6ZSJG7iAPXH2B5+fLoti+/+8sJ1zN73+r3ZfBKw5zQs7pmNS90vDDr88SyJR+vSUREROY3hTaZUsiG6B3tBdJvvDFRpnPMnCrd6kWreWrfUwkrZ+9/6P3RTn0dAx3U314/5XVMDGyQvRb2sxnYAEaDo2x61yaebXs2rmrlNDAZ9g9HK1wWS8iGqC2tTXteVcspLWx6YlPSx5MNcYyVqGI2XbHhZ/3d6zOaJ6Zql4iIiCwkCm0ypdcPvx5dfyp2fti/Pv+v7O7ZzdGRo9SUxA+3iw1pXb4uVty2guHx4UnHSDbH7Dnvc7QPtLPn6B78QX/Sa3OC0nxcOyxTZZ4y1i1Zxzfe/4247ec0nUPnjZ1xFa5v/uGbvNr9anSfdEJMRVG4PX1sQxYnoG35xBY+89hnEla75kKmIUzVLhEREVlIFNokId+Yj57hHoC4BZ5jG280VzfzaverFBcU4x3wRveZOJyxezi+mUSq5h3OcwfHBwEY9g+zEJV7ygkRyuj1OWu5JTzehArXvz7/r3GPpxtinACYqG19rjoygkKYiIiIHN+0uLZMsq1tG/W31dPp60y6TzAUpH2gHSBaCXPmlTmNQKYaLhiyITa/duyX8dgmIrmsnLlw8Z5V7+E3f/Ebyj3ZDyfNVc10faGLJz71RMLFmDfUbYhbhHviIs3pmMlCzk5AA+ICmhaHFhEREckNY23uh5Wdfvrp9sUXX8z1ZQjh4FR/W/20FnY2GIwJdxFMN3QtK1uGteHFkU+qPYktb2zBH0o+HHKu3HT2TXz1gq/SeHtjNJxmQ0VhRVx1cXB8kPMeOG9SZ8Vk2+fK+rvX8/Khlzlt2WnsvHrnnJ1XRERE5HhljNlurT090WMaHnkc8o352Pz6ZvYc2cOaxWtoOaUl2hRk8+ubp93+3mLJ9I8Ah4YOAeEhlC8dfGla5822Mk8ZJ9acCMCZK86k09cZN0R0OpzW+xOrZeWF5bzw2Rcm7Z9su4iIiIgcfxTajhNOUHtm/zP8+M0f4w/6CdogLuPimsev4YazbuDL/+3LPHPgmQU1j+y9J7yX37f/npANMeQfotBdiLWWq95xFd964VsJK4Kxc8ceaXkkOs8utjtjqmB72rLTcLvcePu9DPmHKPOUsbJ65ZxXy0RERERkYVBoWwBSVc7gWHOPYCjIcCA+kIVsiJAN8bXffo07fnfHgurCWFpQSsspLfy45ccJ1w574o9PsPfo3uj+zmLQE6thibozNlY1svGHGyd1WUzWEXO+Uct8ERERkfyhOW3znBPInEpSaUEpgVCA9XXrqSisYFHxIn66+6d5MU9sJlzGRWlBaUYdFyfOH5uoZUsLuw7voqigiAtWXRAX6NIxOD4Y7bLYWNk4523wRURERGTh0Jy2POMb83H+g+fT1tdGfWU9f/Wnf4W335uwSjbxee++/93sPbqXEk8Jf97055MCmVNJe77j+Tl5LbOt1FPK1adfzbradbSc2hJdSNqpcEF4vpjLuLDWTqp6pQpRM20jX15YzlmNZ1HTE16jToFNRERERGaDKm1zLHYdsolDEd3GTYGrgA+v/TAFpoDDw4exWBYVL+JA3wF2dO2Yso3+QuIyLj560kd55BOPxG1PVOECctptUURERERkJlRpywO+MR8PvPwANz5xI+Oh8YT7BG2QYDDIj9740Rxf3cwUugqTvqZEXMaVshtj7NyyRPPDklW41G1RRERERBYihbYsc8LZz9/6OYFQIFolc9rZT7edfi65cGGMSXjtzrwxZ9ii02ExmXJPObe89xba+9tpqGzgvp330T7QTkNFA3+14a9o729Pa27ZTIc2ioiIiIjMFxoemYZEQaxvtC86dNH53GB45sAz8zKYJWIwVBdXc96q87jhXTfENTwp85ThMq64atjg+GC0w6K1lrtevAtrbdL9RUREREQkLNXwyFkLbcaYDwDfBNzAvdbaW5Ltm8+hbVvbNt73vfcxEhjJ9aVkVWlBKX/9jr/mWy9+i7HgWMJ9JnZfjA1l6VTDMt1fREREROR4NeehzRjjBt4C3gu0Ay8A/8Na+0ai/fM1tPnGfKz4xgoG/YO5vpSscwIZwFf+8yvc/ofbMRjGgmOqiomIiIiIzLFcNCI5E9hrrd0XuYAfAB8BEoa2XArZEIcGD9Ha30pbfxutfZGP/a28dPCleRnYPC4PIRtiw/INVJVU8Z8H/hOXcU0KZE7V65b33sKX3/1lVcVERERERPLQbIW2esAb83U78M5ZOldKo4FR2gfaae1rPRbM+ltp6wt/9A54GQ+m3/kwH7hw8bbFb2Nv714MBn/IT2lBKUEbZOO6jZzXfF5c6EpnmGJ5YTlXbrgyFy9HRERERERSmK3QZhJsixuHaYy5CrgKoKmpaVonsdbSO9o7qULW1t/Ggb4DePu9HBw6mPbxqourWVm1kpXVK2mqbOLg0EEe3fXorDYWcdZmK3AVMOQfotBdSCgU4txV52IwYKG6pJq+0T4KTAEXr72YK067gvLC8rTnjCmQiYiIiIjMX7MV2tqBxpivG4DO2B2stfcA90B4TluigwRDQTp9nXFhzKmYOV8Pjqc3fNFlXKwoX0FTVRMrq1eysmrlpM8riirinuMb8/HLPb+clSGSscMU1y9fP62hiQpjIiIiIiIL32w1Iikg3IjkPUAH4UYkl1prX0+0/9v+5G32i/d/MRrODvQdoK2/jQ5fB4FQIK1zlhSUhEOYUymLfO4Es/qKejxuT8avZTrdI93GDcCG5RtoXtQcrZJdsPoCigqK0l6LTEREREREjg+5avl/EXAH4Zb/91lrb0667wpj+evUx6stqaWpOhLKEgSz2tJajEk0KnPmBscHeWDnAzy+53ECwUB0uGLs0EUsLClfwvLy5ayrXadAJiIiIiIiactJaMuEqTe26YvHqmROMHOqZI2VjZQVluX6MkVERERERGZFLlr+Z2RD3Qa2X78915chIiIiIiKSd1y5vgAg3CVRREREREREJsmL0CYiIiIiIiKJKbSJiIiIiIjkMYU2ERERERGRPJYf3SONOQy05vo6UqgFenJ9EbLg6T6TuaJ7TeaC7jOZC7rPZC7M1X220lq7JNEDeRHa8p0x5sVk7TdFskX3mcwV3WsyF3SfyVzQfSZzIR/uMw2PFBERERERyWMKbSIiIiIiInlMoS099+T6AuS4oPtM5oruNZkLus9kLug+k7mQ8/tMc9pERERERETymCptIiIiIiIieUyhbQrGmA8YY3YbY/YaY27K9fXI/GWMuc8Y022MeS1mW40x5iljzJ7Ix0WR7cYY8/8i990rxpgNubtymU+MMY3GmGeMMbuMMa8bYz4f2a57TbLGGFNsjHneGPNy5D77p8j2VcaYP0Tus83GmMLI9qLI13sjjzfn8vplfjHGuI0xLxljfh75WveZZJ0x5oAx5lVjzE5jzIuRbXnzs1OhLQVjjBu4C7gQWAf8D2PMutxelcxj9wMfmLDtJuDX1to1wK8jX0P4nlsT+XcV8O05ukaZ/wLAjdbak4F3AddG/n9L95pk0xhwvrX2NGA98AFjzLuAW4HbI/dZL3BlZP8rgV5r7YnA7ZH9RNL1eWBXzNe6z2S2nGetXR/T3j9vfnYqtKV2JrDXWrvPWjsO/AD4SI6vSeYpa+2zwNEJmz8CPBD5/AHgozHbH7RhvweqjTF1c3OlMp9Za7ustTsin/sI/6JTj+41yaLI/TIY+dIT+WeB84Etke0T7zPn/tsCvMcYY+bocmUeM8Y0ABcD90a+Nug+k7mTNz87FdpSqwe8MV+3R7aJZMsya20XhH/ZBpZGtuvekxmLDA36U+AP6F6TLIsMWdsJdANPAX8E+qy1gcgusfdS9D6LPN4PLJ7bK5Z56g7g74BQ5OvF6D6T2WGBJ40x240xV0W25c3PzoLZPPgCkOivM2q3KXNB957MiDGmHHgEuN5aO5Dij82612RarLVBYL0xphp4FDg50W6Rj7rPJGPGmA8C3dba7caYc53NCXbVfSbZcLa1ttMYsxR4yhjzZop95/xeU6UttXagMebrBqAzR9ciC9Mhp5we+dgd2a57T6bNGOMhHNgettb+OLJZ95rMCmttH/AbwnMoq40xzh+EY++l6H0WebyKycPFRSY6G/iwMeYA4SlggbRVAAAgAElEQVQq5xOuvOk+k6yz1nZGPnYT/kPUmeTRz06FttReANZEuhQVAp8EHsvxNcnC8hhwReTzK4Cfxmy/PNKd6F1Av1OeF0klMn/ju8Aua+1tMQ/pXpOsMcYsiVTYMMaUABcQnj/5DLAxstvE+8y5/zYCT1stFCtTsNb+vbW2wVrbTPh3sKettZeh+0yyzBhTZoypcD4H3ge8Rh797NTi2lMwxlxE+K86buA+a+3NOb4kmaeMMf8BnAvUAoeAfwB+AvwQaALagI9ba49GfvG+k3C3yWHgL621L+biumV+McacA/wX8CrH5oD8T8Lz2nSvSVYYY95OeFK+m/AfgH9orf1nY8wJhCsiNcBLwKestWPGmGLge4TnWB4FPmmt3Zebq5f5KDI88gvW2g/qPpNsi9xTj0a+LAC+b6292RizmDz52anQJiIiIiIiksc0PFJERERERCSPKbSJiIiIiIjkMYU2ERERERGRPKbQJiIiIiIikscU2kRERERERPKYQpuIiEgGjDHXG2NKc30dIiJy/FDLfxERkQwYYw4Ap1tre3J9LSIicnzIi9BWW1trm5ubc30ZIiIiIiIiObF9+/Yea+2SRI8VzPXFJNLc3MyLL87qIuIiIiIiIiJ5yxjTmuwxzWkTERERERHJYwptIiIiIiIieSwvhkeKiIiIiEj2tWxpYXfPbtbWrmXzxs25vhyZJoU2EREREZEFanfPbl4+9HKuL0NmSMMjRURERERE8pgqbSIiInlEQ5lERGQihTYREZE8oqFMIiIykYZHioiIiIiI5DFV2kREREREJOs03Dt7FNpERERERCSrfGM+nvM+R/tAO0dGjuAb81FRVJGTa1kI4VHDI0VEREREJGu2tW2j/rZ6Ogc6Aegc6KT+tnq2tW3LyfU4c4V39+zOyflTCdkQY4ExhsaHUu6nSpuIiIiIiGSFb8zHRQ9fhG/cF90WIoRvPLy988ZOygvLc3iFc8daSyAUwB/yEwgFwp8H/dFtg+ODdAx0cHDwIAcHD6Y8lkKbiIiIiIgklOnQws2vbyZkQwkfC9kQm1/bzJUbrsz2ZebExCAWG9AGxwfp8nXRNRj+d2jwUPhzXxcHBw/SNdhF32hf2udSaBMRERERWYB8Yz6OjBwBmPa8skyXIdlzZA9D/sRD/Yb8Q+w9ujej8+dKMBScFMScgDbiH6HT10nnYGc0gB30HeTg0MFoKHPe93QYDEvKltBNd9J9FNpERERERBaYbW3buOjhi6JzpZx5ZVsv28o5TefM2nnXLF5DmacsYXAr85RxYs2Js3budFlrkw5ZHAuM0TXYRaevMxrGugaPVce6fF30DPdgsWmfr7a0lmVly6irqKOuvI7l5cupK69jWfky6srrWFq2FI/bw9ovrk16DIU2EREREZEFJJfzylpOaWHTE5sSPuYyLlpObZmV88ZKNGwRYDwwzu+9v6fd135suKITyCIVsu6hboI2mPa5qouro0HMCWPLK8IfnWBW6C7EbdwUuAri/nncnrivU1FoExERERFZQHI5r6yiqIKtl22NVvlChHDhoqywjK2XbZ1xWAzZUMI5ZOOBcbqHuunwdYSHLfoORqtj+3r3AfDmkTc5676z0j5XZVHlsSCWIJQtK1tGiackLpBNDGIFrgI8Lg/GmBm97rRDmzHGDbwIdFhrP2iMWQX8AKgBdgCfttaOG2OKgAeBdwBHgBZr7YEZXaWIiIiIyHEsk4YguZ5Xdk7TOXTe2Mm6u9bhHfBSX1nPG9e+MWVgS9Zt0R/0c2TkCN4Bb9ywxeh8skj3xbHgWOrjxwxpLPWURqthTgiLC2blyykrLJsUwJwQFvv1TANZOjKptH0e2AVURr6+FbjdWvsDY8zdwJXAtyMfe621JxpjPhnZb/broCIiIiKSFQthMeKFJpOGINmaVzaTRiblheXUlNTgHfBSU1JDeWE5wVAwGshiq2VHR47SPtBO+0D7sbljMcMWDw0dYtg/nNZ5AYrcRXFh7NnWZzkycoSGyga+ffG3WV6+nMrCSgrciUOYUzFzG/ecBLJ0pBXajDENwMXAzcAmE77684FLI7s8APwj4dD2kcjnAFuAO40xxlqb/mw9EREREcmZTDsGSn7JxryyTBuZhGxoUhhz5pL5xnw8sPOBY2FsQnOPwfHBtF+bx+WJG6a4vGx53Byy5eXLWVS8KDpMcTQwyrkPnAuEK3nvrH8ni0oWTTmHLN+ke7V3AH8HONF6MdBnrQ1Evm4H6iOf1wNeAGttwBjTH9m/J/aAxpirgKsAmpqapnv9IiIiIiILWqYVr5nOK0vVyOTChy7k9Wtfp7igGN+YL1oh6/AdWyTaCWNvHXkLgH19+/iLn/7FlK+zwFXAsrJlk7or1lUc67hYU1JDobsw4dwxp2rmdrmBycGzy9fF6v+3etY7aM6GKUObMeaDQLe1drsx5lxnc4JdbRqPHdtg7T3APQCnn366qnAiIiIiIhNMt3V/JvPKnDXJnKGL//7Sv+MP+hMedzgwzFn3nsVocJSjI0fTfh0u42JJ6ZKUc8iWli2lqKAo5fwxJ5BNJZcdNGdDOpW2s4EPG2MuAooJz2m7A6g2xhREqm0NQGdk/3agEWg3xhQAVUD631EREREREZlx8IidV1ZdXE0wFOTI8BHGAmN0+DrwDnjx9nvjhi0eHDrI3iN7GQ2OJjxmyIboHOyctL22tDYugNWV1/HQqw/R6etk9aLVPH7p4xQXFKdse+8yrpm/aRG57KA5G6YMbdbavwf+HiBSafuCtfYyY8yPgI2EO0heAfw08pTHIl//LvL405rPJiIisvCoWYXI7EoneFyx/oq4uWRjwTG6fF14+720D7TTNdgFwIG+A5z3wHl0DXZxeOhwRmuRxXIbN+evOp/3rX5ftGq2onwFpZ7SSUHsj71/ZM/RPZy0+CTW1iZfOHo25LqDZrbNZAbel4AfGGO+ArwEfDey/bvA94wxewlX2D45s0sUERGRfKRmFakp1MpM7e7ZnTJ43PfSffy+4/fRClmXr4vuoW78oclDG33jPl46+NKk7VVFVfGNPSKNPL76X19NWG0rKSjh/o/ez6LiRVO2vN/yiS0ZvuLsyVYHzXyRUWiz1v4G+E3k833AmQn2GQU+noVrExEREZm3FGqPWSgBNpuvw1obtx7ZeGCcnuEe2gbawlUyXzvPdzyP27iTVsWea3+O59qfS3qOMk8Z/pCf8eA4VUVVXH7a5cfmklUsp7GykcqiyoQt7/985Z/zoe9/iGH/8KRGJisqVszotc+FbHTQzCfzq9eliIiIiMw7CyXAZvI6Yocs+kN+ekd6aetvo62/jfaBdjoHO+Na3x8cPMhIYCSj61lZtZIVFSsmNfeoK6+jobKBmpIaPvDwB3jj8Bs0VjXyf9/zf+OCWSrnNp9L1xe6Ml4gO1/MtINmvlFoExEREZknFkrFKp+k+55ObLvfM9RDsaeYQChA/2g/rf2tx9rfD3RMWiA62TDHRArdhdHOioXuQv7Q/oe4IY+lBaU8+LEHObvp7ISdFmM7LHpcHiA8Fy3ToJJogez5JJMOmvlOoU1ERERknpiLilWma4LNdxPf04nDFgdGB/j5np9zwy9viM7xah9oZ/k3llNfWU//aD/9Y/1pn8/j8rC0bGl07bHYjov1FfU0VDawtGwphe7CaBgbDYxy+j2n0+5rp7Gycd4Gj1yY78HTodAmIiIiIsD01wSbT5ww5g/6GfYPM+wfBqBnqIcvPvnF8LBFZ4FoXxe9o70JjxO0Qdr62+K2uYwrHMhihikur1jO8rJwIGusaoxWzya2u0/V8r6KKhaXLqbd1z6vg4dMn0KbiIiIyAIxk+GTuVyMOFvDPkM2FJ1HNhoYpX2gPTqPrMPXER226ISynuEeLOGVqToGO/j6776e0fkKXAVccvIlbFy3kYbKBlZUrIhbi8zj8nD5Ty7n8bce15BWmRGFNhEREZF5IJ1hizMZPpnLxYjTuW5rbbRKdukjl7KrZxdLS5fyqdM+FZ1HFtvc4/Dw4aSvJxG3cXNS7UlxFbK6sjp+0/obfvbWzxI+JxAK0FzVzMZ1G5Med8+RPVkZ0uqsczbX653l6rwST6FNREREjhvztZHHXAxbnMlixKne16nCZtzjw0foGOhgyD9Ea18r3n4vXp83HMh8neHmHr6u6ILRu4/s5r+8/zXla6suqo52V1xWvoy68rpoh8X/88z/YW/vXk6qPYltf7VtUuv7yuJKnt7/dNL1vtYsXpP0vNmcHzjd+3WmoWs+/XeykCm0iYiIyHFjPraen6thizNZjDjZ++qEzcHxQSDcwKPu63V89YKvUlVcxba2bdy/8/5oZ8R2XzsNtzdkfO0NlQ2sql4VVyFrqGqgsbKRxqrwWmSx3RVjOyze+ttbgfBQx+ri6knHnu56X/kyP1Cha2FQaBMRERHJY3M1bHG64cRaG50X5g/62da2DW+/lz1H9/CVZ78S16oeYCgwxN/+8m/TuqaSgpLwcMWKOmpLavnF3l9MOh5A30gfP/rsj6gurp7U8n6mprPeVy7nB8rCpNAmIiKSJ463VusLWTa/lzMZtpiJZOGk1FPKoy2P4hvz8Vr3a7T1t+Ht94bnkfnCwxbf7HkTgDd63uC//ft/S+t8LuOKC3yxitxFfPnPv8zV77iawoJwp8UHX36QX+37VcLQZrH8/K2fZxxe0/0+ZbreVy7nB8rCpNAmIiKSB/JlKJXMXLa/lzMZtjjRxLlnIRuKrkXW2t9K91A3//vP/zf//Ow/Mzg+SIG7gOXly/nY5o/FVY2m4nF5KC4oTvmc9cvWs+PgjoSPjQXHGBoforasNrptf+9+hgPDCfefTnjN9PtUXljOWY1nUdNTw9ratSkrZXMVtOX4odAmIiKSYxpKNTfmopI5G9/LmQxbDIQCDI0P0TrQSltfG79t+y0dvg7a+tt4z4PviTb16BvtS3iM8eA4e3vjA4bbuFlatpTl5cspKihie+f2uOpXSUEJP/r4j9jft58vPfWlhEGrzFPGqUtPZfeR3WmH0WyG1+l+n9KdH5bNaxUBSLyCn4iIiMyZdIZSycxsa9tG/W31dA50AseqKtvattGypYX1d6+nZUvi8JOJ2fheOsMWKworcEV+dXPhotxTzo8+/qP/3969R0dV3nsD/z57LrlNLgTCJZNErFAM8UIRqBw9R0F6FHrRs/SQ9tVqK2JVpFpqK8d2rfecruOSVsWqgJ5KPcWqhRbbVytgFVHfF+nhIoLcBAJCJhdzgZCZZJLJXJ73j5k9zkz23rMnmSST8P10uUhm9szsSTY03/x+z+9BbXstthzfgmd2PoOTbScBAJ+2fopLnrsEpStLMepXozDt+Wn41vpvod5TDwBo627Dts+24UjrkbjAJiAwNm8ssq3ZAIBR2aPwyNWPYM2CNXjj229g3w/2oemhJhxbegx/u+1vONh0sFe7YlegC9957Tu4ufJm3bVlilCwYt4K3c2ktcJodVV1SscbGei/c+k8VyKAlTaijDdcx1MTkXnDsZVqOK2/S1ZVmVg0EQeaD6Tltcx+L5P92x6SIfgCPrjaXeHNod21uH/W/Xhm5zPo9HfColiQY8vB/Ffma64J8wV9ONxy2PBcrYoV3676Nq4svxJl+WWoKKxAWUEZcm25+Iff/gM+af4EFYUVePS6RzUfv/7ges3XVs9/8/HNhgM8JuRPSGnAR18GgugZ6L9z6TxXIoChjSjjDcfx1ESUmkxqpdILE7G3L521dFitvzOqqgRDQdS56wCkFj49Pg/mvjQXtedqUVFUgW23b0N+Vr6p76XH54m2KTZ4GvBfe/4LrV2tcLW7orc1ehrR3NmMoAxqvr4/5EeLtyXpeWZZsrBmwRos3bK0V5tiIBTA60dfx3PfeK5XiBBCJH1uM8Fn0fRFhgM8Uh3wkerxegbj71y6zpUIYGgjIiIacn1dszQQ9H5RpN4ekqFht/7OKFx4A95omDEbPmMHWIQQQqu3Nfq46qpqPLjlQc3HdQe68fiOx7H4r4ujFaoWbwvu2XRP0veQb8+P7kFW6ihFaX4pygrC+5B9/PnHeOrvT6Er2NXrcVbFig9Of6AbwvozydBs8HHYHSjOKYbL7UJxTnGv6yPZ/YlSPV7LYP2dS8e5EgEMbURERENuOLVStXW3DbtR5kbhIpaZ8Nne1Y4bfn8DOgOdvR53ze+uQVlBGfyy90h6AAjKII6eOap5n4DAV51fRXlhOZz5TjgLnNGNoS8ovADFOcXRTaETA9juht2agQ0IV7z0hn2o9/e1FTCdwWfKmClxf6b7+ETD6e8c9V9/r5dMwNBGRGnFNXhEfTMQrVQD8fexJ9CTUevvzLxHo3ChxR/0Y9lby+DucePE2RMQQsCqWNHU0YQ6Tx16gj2ajwvJEGrba3vdbhEWTJ8wHcFQEJ80fYKADPQ6Jseag0VfWYS7rrjL9HmqklW8poyegoPNB1NqBTTzQ246g0+q12c6rme2L54/RsLPIwxtRJRWXINH1HfpbqUaiL+Pdqu932uB9NaDqcyGTY/Pgx2uHahz1xmuR1PDRWKFTE93sBsvfPxC0uO0XHPBNbhr+l2oKKjAPZvuwZHWI6gaW4Vdi3fh4Xce1t2XzBvw4kTbiT69ZrKK14p5K/CXT/+ie79WRczsD7nDPfgMRvviSKjy0NBjaCMiIiIA+hMhY2+XUuqujzLTEme0HkxdR2YmbCZujFznrsP4J8bjF3N+gfysfLjcrvBgD3c96j31aOxoNBXYjNgVOxZWLcSfDv8JvqCv1/15tjx897Lv4rbLbgsfb7EDCLc+AsYVMUUoWL17NbZ+trVXiE0mWcUr1SmNqTITfM7n4DISqjw09BjaiIhoRGBrbv8khiB1KMeKeSuwfOvy6O2NnkZkWbOQa8tFt787pQBgZkNjKWWv4Jhtzcbp9tM4de4UTp87jRNtJ/D4jscRCMW3GXoDXjz0zkOm3q9VsSIkQ7rr87TYLDZcWXYlXj/6umZoSwytiUHFqCIWkuGvw96GvX2axJms4jXUFTH+nSTqH4Y2IiIaEdiaGy+VfdSMwtSSzUvijg0hhK5AFxw2BybkT0C9p14zAHh8Hmw4tAHHzxzH5NGTUV1VbTh6vyfYg+o/VWPrya3oCYXXjNW561C4ohAAdPcD0+KwO3Bh0YUozS+FM98ZnrJYWI6Kggr88K0f4uiZo6gqqcKqBat6VZ8siqXXZtGqTn8n6tx10aqV1+9FUAZhERbk2nJ7hdbEoKJVEUvUn0mcZqY0zi6fjeLWYkwZM2XYtDASEUMbERHRiKNXNdOr3hiFKT0SEopQAKBXQFBfPxgKwhvwIsuShaWbl+LiMRfrDjHxBX3YXLNZ83ViWYQFRdlF0UCqZcmMJVjxtRWa910+/nJkW7MxZcwUzerTT6/6abiyaLBmT33choMbUHO2BpOKJ6H6kmpTISjxNQWEZiAdqEmcrHgRDU8MbURERBkq2cAOvcekuo+a0T5mejr9ncix5QAAznWfw5M7noTL7cJn5z7Dm8fejAuBaivhvqZ9Kb2GKkvJQlFOEZo6m3DJ2Etw/6z78eBbD+oGq8mjJ+u2yyaGlsTq0x2X34FH3n1E8zxi2x8ddkefA1VsRUyvgjgUkziJKHMxtBEREaVJOtfVdfo74Vzp1BzY8eyuZ6Ovs/aba+PaELsD3Snvo2Z2H7NErd5WAMDp9tOm1pKpAzm0gooFFgQR1HycL+SD1x/eAPtM1xksmLQAy4T+tMQFkxfg3z/496RTJVWJ3yuz7Y/poEDRbJM0O4mTiM4PDG1ERER9oLVmK53r6k62nYwLX2q1bP7L81GYXRieiOhpROnKUnT5u6LhQhGK4ZqsmrM1cPvcWPjHhTh29hhGZY/CP17wj/AFeg/WMEtAYGLRRJTml+KM9ww+PfOp5nESErdeeiveOPpGrymG9868F6t3rdYNjh09HQDCrZ4Xr744bkBK7POsmLcCU1ZNMd0aqqU/7Y9mTRkzBSEZwpHWIwiFeoe2VDenHmrn83RIosHA0EZERMNeKkM3+vMaakgDgDV71sSFpR+99SPk2nPTdw46czc6/B3RYNPsbY67LyiDCErtahUQDldP/c9TWPHhF+u9PsNnunuHAeE1ZCEZiquOZSlZeOSfHsHL+1/G8bbjuHTspdh/bzisrt271rB1cc7EOXj+G8/3mmIopcRzu5/TPQ/19dXwunzrchxbegyzXpgVfZ6dd+3ElFVTUmoN1dOf9kcz1Ope4hYI6RzFP5i4Vo5oYDG0ERHRsKY1dKP0yVLcN+s+QCJaBdMKUGbbGdXXUNvlEgVlEB3+DnT4v6gG9WVseyytljlVKpMUEx+XOKreIiyYNn4aKgorUJJXgg0HN6Dd146xuWPx4Z0f4qLii9Dp74yGrPKC8uikyD8f+TMAxO3blmyjZ7VipTXFMHGyot6QDiDc6rnp2Ka4aYmbjm9KuTV0qKV7FD8rXkQjE0MbERFlNKNgpTd0o8PfgV99+CsA4VCy7G/LNAOUmXZGrddIxkx1J7G98l8u/he4fe5oG2C62RQbvv+V72Nc3jg88eET6Ap2AQhvln3szDH8+oZf4+qKq7Gzbif2N+3HhPwJmDQ6vKbKzObJqmQbPauP1QrJiQHGYXfoft21BnUYDVTJ5MEeqXx9k2HFi2hkYmgjIqKMZhSszIyqD8qgZoAy21LZl3H4KrW6c8e0O9DoaYTL7cLpc6fx/qn38d/7/jtu7dnivy7u02toybZk45fzfokn/v5EXHVMSgnnSmc0sAG9A2Y6qOFrzro5qG2vRUVhBd674z1TYSQ2wBRmFyIkQ4bj92MZDVThYA8iGs4Y2oiIKK1TD9P5OsmCVSqj6mPb41LZx6wv4/BVnf5OPPDWA7j7zbtTDn52ix3BULDXejIzbBYb7px+J17c92Jc9Wbt3rVJ2wdTaa8zOtZhd2D34t0pnXeiUdmj0N7drnmf2mr59sm3o+dgpjWTiGg4YmgjIjrPeXwe7HDtMD0evT9Sma5oJlilMqpebY9LZR+zkAxhnGMcsq3Z6A50p/p2o68bK9eWiy5/l2YQy7HmYHTOaNR56lA5phLb79wenWBYVlCG5e8uN2yfzLPlQRGK7hALM+2DqYT2gW7FU9+LUavlUI7rTyeuRRs6/NrTcJA0tAkhygG8BGA8gBCA30gpnxZCFAPYAGAigFMAFkop20R4NfLTABYA8AL4npRSfywVERENmVQqTkbSXakzClbzX56Px+Y9Ble7C+WF5XFDMIyo7XFG7Y4dPR0Y+/hY5NnykGfPQ4OnQXd8fqoEBPJsefjWxd/Cqwde1TymK9CFnlBP9PPECYaXj7+8V4DJteViyawlEBBJR9P3p31wMH+wjX2tVAd1DMa4/oHAtWhDh197Gg7MVNoCAH4spdwrhMgH8JEQ4h0A3wPwrpRyhRBiOYDlAB4GMB/A5Mh/XwXwXORPIiLKIKlUnJI9T7ordYbByt+BB956ACEZgkVYYLfYkWvLRbe/23DiotfvxUNvPwRf0IeuQJfmMRISXYEudAW60NrVqvtcAgJWYcW1F16LwqxCFGQVYMOhDejyd+lOPZSQ6PB34LXDrxkGpyxLlu7r9nfSYH/aBwfzB9vE10p1UMdAj+snIhpsSrIDpJSNaqVMSukBcASAE8CNANZFDlsH4KbIxzcCeEmG/Q+AIiHEhLSfORER9cvcl+ZGK2yJ1PVNyWyv3Q7nSica3OEBFmqlbnvt9l7Haq1P0zpm7d61eOGjFwxbHtVAF5RBdAW64A/4kW3LBhCeFqlFQuKc75xuYIvlsDtgV+ya99kUG55d8CzOLj+Lt7/7Nv608E/47Y2/xecPfQ5ngRMAUJRdhDxbnubje4I9uuegCAUznTNx+bjLdStaaoABkDTATBkzJe651MmO+fZ8KJEfARQoyLfnZ3z7IBHR+SylNW1CiIkAvgJgJ4BxUspGIBzshBBjI4c5AbhiHlYXua2xvydLlC6DNXSBMk8mfu+H4pw8Pg+Oth7VrUx1+jvxxI4nsGj6orjzAxD9eO031+pW6q753TUYnTMaFxRdgG23b8P+pv1J2zATNxlOhV/64feH2xi19lErzS9Fq7cVPcEeOOwO9AR64toQe73/nk7d4R92ix3ZluxeASe2GmRTbGjrbtN8vIREUVYRAqFAdDJi7Fo0M22pZlsVzYzV7+++YIOF646I6HxmOrQJIRwAXgPwoJTSbbCGQOuOXv/PJ4S4G8DdAFBRUWH2NIjSIpVhCOmUiYHhfDNU33sjZs5JvXbOdp1FcU5xNDTF7vO15fgWnGg7YXqj6GT7gXX6O3u1Po7KHoUDzQcAGLcwhmQILd4WnPGeQemTpQghBK/f+8X9kXB33brrcNtlt6HOXYetJ7emHNZiZVmy4Av6UJpfimdueAbPf/Q8GtwNGJc3Dnsa9yAQDAAAvD1eZFmzkGvRb6k0mtZotN+XGihsFhs6/Z261UKH3YHDSw73ed1Vf//9SOe+YIOF/2YS0fnMVGgTQtgQDmyvSCn/HLm5SQgxIVJlmwCgOXJ7HYDymIeXAei18YuU8jcAfgMAM2bMSG2WMdEwlYmBgYYH9drJtmbD5Xah098J50pn3IQ8dTS80bqyVDaK7gp0oXRlKbw94bDV4G5Ag+eLf87NjMJXN7rW0xPqwYv7Xkx6Lsnk2fKiIaQktwQ3T70ZN0+9GR6fB86Vzl7VwK5AFxw2BybkT0C9p15zDZrRa+kN7FCDhfq6ekIyBCnlkK67YuWKiGj4SLqmLTIN8rcAjkgpV8bc9QaAOyIf3wHg9ZjbbxdhVwJoV9soiYgonpl1XlpOtp2Ep8cTbQUMymA0dBitK0tlo+hWbys6ejqilagQQtHHnjp3Cjvrd+quH0vFRaMuwrTx05Ie58x36laEFKGgKLuo1+1G71dCQn7UFFcAABwsSURBVBFK9GOzzOz3pbV2LFajp1H3ezRYNtyyAfvu2ccKFhHRMJA0tAG4CsB3AcwVQuyL/LcAwAoAXxNCHAfwtcjnALAZwEkANQBeAHBf+k+b6PzQ1x/oh8pwO18j1RurMe35aajeqP3DebL7zTA7xCP26+oPhtdtSakfMmInQCa2QPZno+hY7b52fHD6A831Y6lQhILmzma0dbUh15areYxFWFBeUI6rKq7Cllu36A7RqCyp7DXAI9m+ZHm2PJQVlBmGTxHp+k91YIe6dmxCfu9ZXEbfIyIiokRJ2yOllNuhvU4NAK7TOF4CWNLP8yI676Vr/6zBUL2xGnsb9+LUuVMIhcJVjUw8X61QqTeaPlkrq9lWV621aBtu2WA4bn/OujmoHFOJypJKLJ21NO4HezUkmakMhWQI6w+sx02VN8HV7oLL7UKduw5WYUVABpI+3gyLsEARCgKhQErVqthz9PR40NHTofv4XFtu3KAMvSEaWtdZsn3JHvqHh7CwamGvFkqVw+ZAYXYh6j31fRrY4bCHWzAbPY2aa+fUKZ0cT09EREbMVNqIaJDF/kAf25qWqb+ZP9JyBDVna8LT8DL0fNM5mj6ViqIa7j7v+Bz7m/Zjh2sHGtwNuH/z/egOdGs+JhgK4kDzAXxY+yHmvzIfnh5PnwJRp78Ti99cjJLHSzD9N9Nx4/ob8erBV9MW2AAg25qNp65/CleUXoGS3JJou2GqYt+fOio/z5anWdly2B2YXT4bl4+7HLPLZxuGqOqqat1zUtscjcbgb7ltC8bkjgGQfLy+nrkT5xpO6dQbakJERKRiaCPKQMmm8pnZP2sw6Y02BzLjfM2EYHV/sFtfuxVjnxiLenc9gN7hLtXw1+gJL+n1h8JtjfXuejifcuKVA69Eb0ukBph6T/2ABF4FSlw7oICATbEZbuqsp9PfiTp3HXYv3o3mnzTjg+99gHx7frSlMFW51lzcMvUWLL9qOZ6+4Wk0/LhBs4Jmdj2W2X3J1FZGdZ81Z4FT97VTpVb7tBgNNSEiIlIxtBFloGTrcDLtN/M9AYP9rjLgfJNtIv2fH/wnnCudeGDLA3j14KvoDnRHg1NsuGv0NJqugG6v3Y7SJ0vR7G2Oez31efu7Fqw/Qgghx5qD6ROmY2zeWFxRegVOP3gadov2ZtJGEkOHGn6+e9l3YVNsKT+fN+DFBMcEPDbvMSyavigto+jNBjK9TasTN6hOlZlqHxERkRGGNqIMNNx+M2+36v+wnwnnW3uu1rA97amdT8HT44E34NU8BgC8fi+q1lTF7TUWK7ai2NrZihtevsFw1H26CAjkWHIgIDSnFOqRkLhvxn1oeqgJuxfvxoT8CZoVKYfNEW0P1KIVOhx2B1YtWIVsa7b+43TOdaCuF71AZkZ/pyyarfYRERHpYWgjyjDVG6vxzM5n4Av6NO/PxN/Mj8oepXtfX85XbVV8+J2HsXbvWjS4GzDzhZkY9/g4zHxhZkpTKT0+j24bIhAOPT1B/UqhKiiDaOtu062Qdfo78bNtP0PJ4yUoeaIkLRMazXDYHWj+aTPc/+aOVpLM0KqAalWkGh9qxNwL52JS8SRYFavp0BEbVNRWTIuwwKpY8aVRX4KiDH7lyUzFrL9VNT0D2X5JREQjn6nNtYlo8BxtPYoDzQcwqXgSmjqa0NnTiRBCUKAgz56Xkb+ZV1u/FKEAEv06X3VqprpptCIUhGQouvlxq7fV9FRK9bmM1oX1ZcCHnqbOpn49PpUNngUEHHZH3Nd3dvlsSJdEo6cxafulXkVLHfJR3BqedOmwO6IVpo6eDs2pjXrUoLLh4AbUnK3BpOJJqL6kGg67o9f32SIsyLXlDuj1baZSNpB7lqnVPpfb1eehJkREdH5iaCPKUHm2PN3R5kNNHWOvjq9XVZVU4Vz3uT6fr+YY/MhAFq01Zg0/bog+f+I5aT3XQLIICyyKBf6gv09B0GFzAAKaATN27LxFWBCUQZQVlPX6+qrvW298fSyjipZecOlL6HDYHZrj7I0C3UimVvDSXckjIqKRjaGNaJCkskeYKhN/M+/xebDDtQN17jqc6TqDBndD9H21dbdhVPaopOfr8Xkw96W5qD1XC2eBE3d+5U642l3YcGiD6WmJXf4uLNm0BKsWrILD7sCh5kM41HII7d3tWLVrFTYd36S7/iydROR/QRlEMGh+uEi2NRvdgW7k2fKgCAWbb90MANG9+RKrlfdvvh/1nnrYLDYEA0Hdr6/alhj7PLEGo6Jlll6gG8kGspJHREQjF0Mb0SDIlI2y9SpkZiW+D3V8vTrevcHdgAZPQ6/Xu2jURZg/eT6OnzkOAFizZw28PV6EEEKztxlLtyxN+VwCMoCXPnkJv//k97Aq1ui6tVPtp/r0fIlSaVVMpbI2Nm8s9t+zH5uObdKsMKWjuqpWseasm4Pa9lqU5Zfhzul3oq697rypaBEREY0kDG103ulLxau/r9er5U+nxS8dr6VWsCqKKrDt9m3R95ZYIUv1fWu9DzWsxLYuqvnFH/JHX++Tpk/w1om3BqTyJSHjBo0ICFiEJekG0nqhzCqsuPWyW7G3cS8ONB+AAkVz8qRFWBCSIdOBzabY4A/5McExAeMd43UrTOmqrjrsDuxevLtPjyUiIqLMwumRdF5JZWPkdEllo2ytQGmW+t72NuxFs7cZexv2Rt9bX9937BTH+zffn9LeYkdajqDOXQcgHKwGslUxS8lCvj0/+lrJzlOBAoti0bwvx5aDVQtWobKkEpeOvVR3yqFVsSLXlpv03NRpi7m2XNNTCY0mGGZZsgZkuiERERFlLlba6LwxmBWvWGY3ytZroRznGJf0NYze2/yX5/cacKHeN2fdHFSOqURlSSXWfnNtXJXu0TmP4pY/3RKd7pdKuyCQ3qmMyfhCPvh6vtgiQe+1Y9eIATCcXqi2j8Z+X2LXmX1zyjfx6oFXk56boiioGlOFypJK0y2pRsdNLJqIfffsM/U8A4GDNIiIiAYfQxudN8xUvAZiKIK6UbZWcFPHrhuFrs623o9LXJtm9N56Qj3RNWeJgqEgDjQfQFNHE0pXlsatM7v+levjjh3MEBZr/qT5ePHGF/HwOw/jDwf/YLjnmh41cCauETMzvTBxfVhFYQXeu+M9rD+4Hq9/+rpuII8dyZ+OdYuZEpY4SIOIiGjwMbRRRunvoAwjZite6VZdVY1lf1umeZ86dv13+36nvxmzRlY62noU+5v2Rz83em9GG0erQazZ26x7zFDKs+Xh5sqbMd4xHqsWrMJfPv0L/D2phzb1fSauETM7vVBrfZjR9xUASvNL8en9n6atesuwREREdP7imjbKKGoYOdp6NO3PrVa8tOhtNJwOd/31LoxzjINVsUbXNylQ4LA5cO/Me/GDv/4AD771oH4VMDIEIyRDmPnCTJQ8XoLDLYcBfLHuzei9DQa1kqdA0a3q9UXsXmLqKPt8e37062iWRVhQXlCe1iqV1vkoUGBVrJhUPAlXVVzFCY1ERESUFkLKoWl5ijVjxgy5Z8+eoT4NygDTnp+G/U37cfm4y+PW7aSjAme06XC+PV93TZvH58GGQxtw/MxxlBeWAwBc7S5MHj0Z1VXVyM/Kjztm26lt8AV80XViU9dMRZ27Ds58JwCg3lOPsblj4Q14EQqF4A0YD+hQpxfqTjtUrJgyegoOtxwelBbGXGsusqxZaOtuQ3lBOXYt3oVZL8yCy+1CeUE58rPyo6FSi0VYkGPNQa49F82dzXDmO7GwaiFW714d3Zg6du1ZYmthR09HtFUxx5qDVm+rfpUywuj7218dPR3REf3lBeUZswE6ERERDS9CiI+klDO07mN7JKVkINsXjSS2A6ZKHYVvUSxQhAIpZa9woP6gHfsel85aGjesIpZFWLDsb8uwYt4KLN+6vNcxZ7vOwrnSGR0s0uhphFqEau1q1a2sJVIrbXqBLBAK4FDLoZS+Hv1hUSwoKyhDW3cbinOKMd4xPm5EvUoRCiARDZx2ix23TL0FcybOQfUl1bj6xavR3NmMMbljsPL6lfjFnF+Y2p8stlVRDeJ6tL6/6eawOzC7fDaKW4sxZcwUBjYiIiJKO4Y2Skl/w5MRvf3T+ruv2nXrrsP7p9+PhqTY9r3EcBC7l1mrtxWbj2+Om7oYKyiD8PR4sGTzEs376z31cZ/H7mFmNrCZJSDwVedXsXTWUvzwrR9Gv15Gx/elKpdtzcbmWzfj2V3PQhGKYbthVUkVznWfg8vtQllBWa8QljhYoy/7k6ktirHTHdXvr9bgkYHC9WZEREQ0kBjaKCPojbtXq1iJt2u1zWlVAT0+T1xgA+IrVvlZ+Vh/cD2OnzkOAFizZw28PeF2xQZPw5BNTFTNKp2FmrYanO06a3ichMRHjR/hnk334N6Z92L1rtWaLYPRVkshkGprdJYlC5898BnGO8YbTkNMDGN6FSitoNOXCYnqdEe1SldWUIai7CIcaD7Qr82piYiIiDIFQxsNutj1X5NHT8aCSQt0x90nVrH09lWLrZCd6TqDBncDNtdsxsbDGw3DyZGWI3jwrQc1A85QBzZFKHj3jnex/uB63XOM5Q/54e/xY82uNRBCexiIumdYSW4Jdjfsjtt7LNeWi7KCMjR1NqHd1x5tbYxtMRzvGJ/0vPtTderrYxNbFAEkrQQSERERDRcMbTSo1Ipa7IbGVsWqGzL0xO6rllilq3fXw/mUE9nWbHQHug2fR0ImDUMDwUx74pdGfQkOuyPpaPlEEhL3z7gfa/as6bUhdGyFMnagh7r3mBqCje7LVGxRJCIiopGKoY0GjdYG0kEZRDAYNHiUNnVfNa3nVMNQssA2VBSh4OqKq/Fx48dx4TUkQ5CQsCk2+EP+6Aj/uHVb/s6k6+E6/Z0QQmhuCJ24R1ni3mNm7iMiIiKiwcXQRqYnQvZ3IMi6/evSVtXKseZACIGfvPMT+AK+tDznQEusdnX0dGDDwQ2oOVuDScWT8PTOp3Gg+QBybbmYWDQxrrVPXbe14eAGHG45jOc/eh5ev/ZWAeqecwxeRERERCMDQ1uGG8gR+9Ubq3Gk5QhqztagK9BlGMT0BoVsXLgRP9v2M9Seq0VFUQW23b4NAOLWrFVXVWN/034s+9uytE1N7Ap04bHtj6XlucyyCisWVi1Es7cZW09uNf04RSgYnTMaFxRdEFftctgdWDR9UfS4Z3c9CwCYWDQxbo86Vezx75x8BweaD+i+nroh9WDpywARIiIiIjKHoS3DDdSIfY/Pg/c/ex/N3ubobfXueoz+1Wj8aPaP8PN//Hk0vGm1IKoDQa5/+froRMIWbwtG/XIULIoFwVAwumfZ4r8uhiKUtI+57yu7xY7KMZU40HwgpXMKyAAqCisw58I5+Lvr77pVQ3W9mtEG0VpSCT6VJZXoCnTh1LlTyLJkodPfiTxbHhShDOieZHq4noyIiIho4IhUx34PhBkzZsg9e/YM9WlkHI/Pg6lrpqLOXRfe5+q+wym1I+rZXrsd81+ejw6/9v5jAGBTbLhv5n2YVDwJbx57E++feh++4PBoQzRj+VXL8fUvfx0LXlmAkAxFQ08gFEBPsEdzSEieLQ9P3/A0FlYthHOlMy7Eqhw2BwqzC1HvqUd5QfmA7xGW2GJZfUl1xg8MISIiIqLehBAfSSlnaN7H0JaZYtsRtab/aY3N31yzGcfPHEd5YTkAwNXuirYnxlbNSleW6m4YfT5Qw9ei6Yt6hZ4FkxdgyqopmoEs354f3WZA/f7EBj61yvXsrmcHrKWViIiIiEYmhrZhxuPz6FZysixZuHHKjXjj2BvwB/0IymC09VCrBTHPlgcBgftm3QdfwIetJ7fiUMuhwXorfaYIBQICQRk0NR4/Vr49H3t/sBeXPnep5gTJ2PClJVlgVrHKRURERETpwtA2zKzdu9bUZsojhbomToECRVEwsWgiLht3GRZMWoCaszVo8DTgtSOvmf56lBeUo/ZHtabDl5aOng5MXT0VLrdrUNociYiIiOj8ZhTaOIgkQ6jtjgebD+LtmrdHfGBTA1R0+qTBJs4enwd/+fQv5p5XKJjpnAngizH5avhyFjhNhy+H3YHZ5bNR3FqMKWOmMLARERER0ZBhaBtAWuvO/vzpn/HmsTcRCAUwNm8sRmWPwom2E9j2WXhUvj/kH+KzHlha4+//+aJ/NnxM7ObS6hqybEs2uoPdyLZmozvQHbemLLaK5rA7UJxTDJfbheKc4pTCF9ejEREREVEmYGjrp9gKWXNHM1q8LZCQEBD44PQHUIQCX9AHq2JFIBQY6tMdUOras1xrLroCXXHr0CzCglxbrunx94liN5dW15B9/ctfx6Zjm7imjIiIiIhGNIY2E5IFMyB5hSzTA1ueNQ/OAieOnz0OIYTp/cvybHnwBX2oKKxAcU4x5l04LxqgAKR1UEfiZtQAen2uhRs/ExEREdFwNmCDSIQQNwB4GoAFwFop5Qq9YzN5EIk6zMIf9KM72HsS4UjgsDuw5dYtuLri6riJiGUFZTjRdgKrd6+GgIAv6INFWCClxJdHfxk3XXwTK1xERERERGkw6NMjhRAWAMcAfA1AHYDdAL4jpTysdXymhDa3zw1Xuwt17jq43C7UnK3BkzueREBmdpWsL6yKFQICy2Yvw8//6eeGoYuj7YmIiIiIBtZQTI+cBaBGSnkycgLrAdwIQDO0DQav3wtXuwsutysazE63n47eVueu09wXbbhSB3WorIoVkMC8i+ZhaslUTB0z1XT40mpLJCIiIiKiwTFQoc0JwBXzeR2Arw7Qa6E70B2ujsWEslp3LWrba1HXXod6Tz3auttMP59VsaI0vxTBUBD1nvqBOu0+y7JkYcmsJbAJG1bvWQ0pJTr9ndEJihsXboSr3YXDLYdxpusMRueOTimkERERERFR5hio0CY0bovrwxRC3A3gbgCoqKjQfaKeYA/q3fXRlsXa9trof3XucCBr9baaPjFFKJjgmICygjKUF5ajoqAC5YXlKC8oj942Lm8cLIpl0De5Vqth1154LQQE3N1u7GvaB6tijQtlsRMYf37Nz9m6SEREREQ0gg3UmrbZAP5dSnl95PN/AwAp5WNax1deVin/4w//gdr22riWxXp3PZo7m+NGxxu+LgTG5o1FWUEZKgorUFFYEQ5iBeXRYDYhf0I4HJng8XngXOnsV9ukRVgQlEHYFBv8IT9sig0hGcL08dMxecxkFOcUw+Pz6FbDuJ6MiIiIiGjkG4pBJFaEB5FcB6Ae4UEk/0tKeUjz+FIh8YPkzzs6Z/QXFbLCil5VMmeBE3aLPY3vxNz0yMQKGSRQ4ijBeMd4TB0zlfuJERERERGRoUEPbZEXXQDg1wiP/H9RSvmo7rGlQhY9UBStilUUVsRVx8oKylBWUIYcW86AnGsyarXrcMthfN7xOVq8Lb2CGYMYERERERH11ZCEtlRMv2K63PvR3qE+DSIiIiIioiFhFNqUwT4ZLYrIiNMgIiIiIiLKOExLREREREREGYyhjYiIiIiIKIMxtBEREREREWWwjBhEIoRoAXB6qM/DwBgA5nfwJspcvJZpJOB1TCMFr2UaKXgtp8cFUsoSrTsyIrRlOiHEHr1JLkTDCa9lGgl4HdNIwWuZRgpeywOP7ZFEREREREQZjKGNiIiIiIgogzG0mfOboT4BojThtUwjAa9jGil4LdNIwWt5gHFNGxERERERUQZjpY2IiIiIiCiDMbQZEELcIIQ4KoSoEUIsH+rzITIihHhRCNEshDgYc1uxEOIdIcTxyJ+jIrcLIcQzkWv7EyHE9KE7c6J4QohyIcR7QogjQohDQogHIrfzeqZhQwiRLYTYJYTYH7mO/yNy+4VCiJ2R63iDEMIeuT0r8nlN5P6JQ3n+RImEEBYhxMdCiDcjn/NaHkQMbTqEEBYAqwHMBzAVwHeEEFOH9qyIDP0OwA0Jty0H8K6UcjKAdyOfA+HrenLkv7sBPDdI50hkRgDAj6WUlQCuBLAk8u8vr2caTnwA5kopLwcwDcANQogrAfwSwFOR67gNwKLI8YsAtEkpJwF4KnIcUSZ5AMCRmM95LQ8ihjZ9swDUSClPSil7AKwHcOMQnxORLinl/wVwNuHmGwGsi3y8DsBNMbe/JMP+B0CREGLC4JwpkTEpZaOUcm/kYw/CPyQ4weuZhpHI9dgR+dQW+U8CmAtgY+T2xOtYvb43ArhOCCEG6XSJDAkhygB8HcDayOcCvJYHFUObPicAV8zndZHbiIaTcVLKRiD8gzCAsZHbeX3TsBBpq/kKgJ3g9UzDTKSdbB+AZgDvADgB4JyUMhA5JPZajV7HkfvbAYwe3DMm0vVrAD8FEIp8Phq8lgcVQ5s+rd8IcNQmjRS8vinjCSEcAF4D8KCU0m10qMZtvJ5pyEkpg1LKaQDKEO7gqdQ6LPInr2PKSEKIbwBollJ+FHuzxqG8lgcQQ5u+OgDlMZ+XAWgYonMh6qsmtU0s8mdz5HZe35TRhBA2hAPbK1LKP0du5vVMw5KU8hyA9xFeo1kkhLBG7oq9VqPXceT+QvRueScaClcB+JYQ4hTCy4XmIlx547U8iBja9O0GMDkyGccO4NsA3hjicyJK1RsA7oh8fAeA12Nuvz0yde9KAO1q2xnRUIusffgtgCNSypUxd/F6pmFDCFEihCiKfJwDYB7C6zPfA3BL5LDE61i9vm8BsE1yM13KAFLKf5NSlkkpJyL88/A2KeWt4LU8qLi5tgEhxAKEf5NgAfCilPLRIT4lIl1CiD8AuBbAGABNAP43gP8D4I8AKgDUAvhXKeXZyA/FqxCeNukF8H0p5Z6hOG+iREKIqwH8PwAH8MX6iUcQXtfG65mGBSHEZQgPY7Ag/EvyP0opfyGE+BLC1YpiAB8DuE1K6RNCZAP4PcJrOM8C+LaU8uTQnD2RNiHEtQAeklJ+g9fy4GJoIyIiIiIiymBsjyQiIiIiIspgDG1EREREREQZjKGNiIiIiIgogzG0ERERERERZTCGNiIiIiIiogzG0EZERJQCIcSDQojcoT4PIiI6f3DkPxERUQqEEKcAzJBStg71uRAR0fmBlTYiIhpxhBC3CyE+EULsF0L8XghxgRDi3cht7wohKiLH/U4IcUvM4zoif14rhHhfCLFRCPGpEOIVEfZDAKUA3hNCvDc0746IiM431qE+ASIionQSQlQB+BmAq6SUrUKIYgDrALwkpVwnhLgTwDMAbkryVF8BUAWgAcCHked7RgixDMAcVtqIiGiwsNJGREQjzVwAG9VQJaU8C2A2gFcj9/8ewNUmnmeXlLJOShkCsA/AxAE4VyIioqQY2oiIaKQRAJIt2FbvDyDy/4VCCAHAHnOML+bjINidQkREQ4ShjYiIRpp3ASwUQowGgEh75A4A347cfyuA7ZGPTwG4IvLxjQBsJp7fAyA/XSdLRESUDH9rSEREI4qU8pAQ4lEAHwghggA+BvBDAC8KIX4CoAXA9yOHvwDgdSHELoTDXqeJl/gNgC1CiEYp5Zz0vwMiIqJ4HPlPRERERESUwdgeSURERERElMEY2oiIiIiIiDIYQxsREREREVEGY2gjIiIiIiLKYAxtREREREREGYyhjYiIiIiIKIMxtBEREREREWUwhjYiIiIiIqIM9v8BoxmVYfXfK8AAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1080x1800 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig,[ax1,ax2,ax3,ax4,ax5,ax6,ax7,ax8,ax9,ax10] = plt.subplots(nrows=10, figsize=(15,25))\n",
    "sn.regplot(y_t[0],np.exp( y_pred[0]), x_bins = 200,color='g',ax=ax1)\n",
    "sn.regplot(y_t[1],np.exp( y_pred[1]), x_bins = 200,color='g',ax=ax2)\n",
    "sn.regplot(y_t[2],np.exp( y_pred[2]), x_bins = 200,color='g',ax=ax3)\n",
    "sn.regplot(y_t[3],np.exp( y_pred[3]), x_bins = 200,color='g',ax=ax4)\n",
    "sn.regplot(y_t[4],np.exp( y_pred[4]), x_bins = 200,color='g',ax=ax5)\n",
    "sn.regplot(y_t[5],np.exp( y_pred[5]), x_bins = 200,color='g',ax=ax6)\n",
    "sn.regplot(y_t[6],np.exp( y_pred[6]), x_bins = 200,color='g',ax=ax7)\n",
    "sn.regplot(y_t[7],np.exp( y_pred[7]), x_bins = 200,color='g',ax=ax8)\n",
    "sn.regplot(y_t[8],np.exp( y_pred[8]), x_bins = 200,color='g',ax=ax9)\n",
    "sn.regplot(y_t[9],np.exp( y_pred[9]), x_bins = 200,color='g',ax=ax10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Kaggle Submission"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we will use the fitted model to make predictions for the given test set. First we load the test set and then follow the same preprocessing steps as for train set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>season</th>\n",
       "      <th>yr</th>\n",
       "      <th>mnth</th>\n",
       "      <th>hr</th>\n",
       "      <th>holiday</th>\n",
       "      <th>weekday</th>\n",
       "      <th>workingday</th>\n",
       "      <th>weathersit</th>\n",
       "      <th>temp</th>\n",
       "      <th>atemp</th>\n",
       "      <th>hum</th>\n",
       "      <th>windspeed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.26</td>\n",
       "      <td>0.2121</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.4478</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.42</td>\n",
       "      <td>0.4478</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.2121</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.2836</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.74</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.2239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.2424</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   season  yr  mnth  hr  holiday  weekday  workingday  weathersit  temp  \\\n",
       "0       1   1     1  20        0        4           1           1  0.26   \n",
       "1       4   1    10  15        0        5           1           1  0.52   \n",
       "2       1   0     1  10        0        0           0           1  0.22   \n",
       "3       3   1     8  18        0        3           1           1  0.74   \n",
       "4       1   1     2   8        0        1           1           1  0.24   \n",
       "\n",
       "    atemp   hum  windspeed  \n",
       "0  0.2121  0.52     0.4478  \n",
       "1  0.5000  0.42     0.4478  \n",
       "2  0.2121  0.51     0.2836  \n",
       "3  0.6667  0.51     0.2239  \n",
       "4  0.2424  0.60     0.0000  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filename = 'dataset/test.csv'\n",
    "\n",
    "df_test = pd.read_csv(filename)\n",
    "df_test.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5214, 12)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_test.rename(columns={'weathersit':'weather',\n",
    "                     'mnth':'month',\n",
    "                     'hr':'hour',\n",
    "                     'yr':'year',\n",
    "                     'hum': 'humidity',\n",
    "                     'cnt':'count'},inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "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>season</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>hour</th>\n",
       "      <th>holiday</th>\n",
       "      <th>weekday</th>\n",
       "      <th>workingday</th>\n",
       "      <th>weather</th>\n",
       "      <th>temp</th>\n",
       "      <th>atemp</th>\n",
       "      <th>humidity</th>\n",
       "      <th>windspeed</th>\n",
       "      <th>year_0</th>\n",
       "      <th>year_1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.26</td>\n",
       "      <td>0.2121</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.4478</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.42</td>\n",
       "      <td>0.4478</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.2121</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.2836</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.74</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.2239</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.2424</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   season  year  month  hour  holiday  weekday  workingday  weather  temp  \\\n",
       "0       1     1      1    20        0        4           1        1  0.26   \n",
       "1       4     1     10    15        0        5           1        1  0.52   \n",
       "2       1     0      1    10        0        0           0        1  0.22   \n",
       "3       3     1      8    18        0        3           1        1  0.74   \n",
       "4       1     1      2     8        0        1           1        1  0.24   \n",
       "\n",
       "    atemp  humidity  windspeed  year_0  year_1  \n",
       "0  0.2121      0.52     0.4478       0       1  \n",
       "1  0.5000      0.42     0.4478       0       1  \n",
       "2  0.2121      0.51     0.2836       1       0  \n",
       "3  0.6667      0.51     0.2239       0       1  \n",
       "4  0.2424      0.60     0.0000       0       1  "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "season=pd.get_dummies(df_test['year'],prefix='year')\n",
    "df_test=pd.concat([df_test,season],axis=1)\n",
    "df_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>season</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>hour</th>\n",
       "      <th>holiday</th>\n",
       "      <th>weekday</th>\n",
       "      <th>workingday</th>\n",
       "      <th>weather</th>\n",
       "      <th>temp</th>\n",
       "      <th>atemp</th>\n",
       "      <th>...</th>\n",
       "      <th>month_3</th>\n",
       "      <th>month_4</th>\n",
       "      <th>month_5</th>\n",
       "      <th>month_6</th>\n",
       "      <th>month_7</th>\n",
       "      <th>month_8</th>\n",
       "      <th>month_9</th>\n",
       "      <th>month_10</th>\n",
       "      <th>month_11</th>\n",
       "      <th>month_12</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.26</td>\n",
       "      <td>0.2121</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>4</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.5000</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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.2121</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>3</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.74</td>\n",
       "      <td>0.6667</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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.2424</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 × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   season  year  month  hour  holiday  weekday  workingday  weather  temp  \\\n",
       "0       1     1      1    20        0        4           1        1  0.26   \n",
       "1       4     1     10    15        0        5           1        1  0.52   \n",
       "2       1     0      1    10        0        0           0        1  0.22   \n",
       "3       3     1      8    18        0        3           1        1  0.74   \n",
       "4       1     1      2     8        0        1           1        1  0.24   \n",
       "\n",
       "    atemp  ...  month_3  month_4  month_5  month_6  month_7  month_8  month_9  \\\n",
       "0  0.2121  ...        0        0        0        0        0        0        0   \n",
       "1  0.5000  ...        0        0        0        0        0        0        0   \n",
       "2  0.2121  ...        0        0        0        0        0        0        0   \n",
       "3  0.6667  ...        0        0        0        0        0        1        0   \n",
       "4  0.2424  ...        0        0        0        0        0        0        0   \n",
       "\n",
       "   month_10  month_11  month_12  \n",
       "0         0         0         0  \n",
       "1         1         0         0  \n",
       "2         0         0         0  \n",
       "3         0         0         0  \n",
       "4         0         0         0  \n",
       "\n",
       "[5 rows x 26 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "season=pd.get_dummies(df_test['month'],prefix='month')\n",
    "df_test=pd.concat([df_test,season],axis=1)\n",
    "df_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_test = df_test.drop(['atemp', 'windspeed'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_test = df_test.drop(['year', 'month'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5214, 22)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test.shape\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_test['weekday'] = df_test.weekday.astype('category')\n",
    "df_test['hour'] = df_test.hour.astype('category')\n",
    "df_test['holiday'] = df_test.holiday.astype('category')\n",
    "df_test['weather'] = df_test.weather.astype('category')\n",
    "df_test['workingday'] = df_test.workingday.astype('category')\n",
    "df_test['season']  = df_test.season.astype('category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5214, 22)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "season        category\n",
       "hour          category\n",
       "holiday       category\n",
       "weekday       category\n",
       "workingday    category\n",
       "weather       category\n",
       "temp           float64\n",
       "humidity       float64\n",
       "year_0           uint8\n",
       "year_1           uint8\n",
       "month_1          uint8\n",
       "month_2          uint8\n",
       "month_3          uint8\n",
       "month_4          uint8\n",
       "month_5          uint8\n",
       "month_6          uint8\n",
       "month_7          uint8\n",
       "month_8          uint8\n",
       "month_9          uint8\n",
       "month_10         uint8\n",
       "month_11         uint8\n",
       "month_12         uint8\n",
       "dtype: object"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = gbm.predict(df_test)\n",
    "y_pred1 = clf_cv.predict(df_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We applied the combination of two models (Random Forest Regressor, Gradient Boosting Regressor) and we observed that Gradient Boosting gave us the best score so it was more influencial on our predictions than Random Forest Regressor. We tried the following combinations: 0.6/0.4 , 0.8/0.2 , 0.75/0.25 but the most effective result was this one: 0.7/0.3 se we applied this. We also tried to keep the average between the predictions of these two methods but this did not work."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = y_pred*0.7 + y_pred1*0.3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "True in (y_pred < 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i, y  in enumerate(y_pred):\n",
    "    if y_pred[i] < 0:\n",
    "        y_pred[i] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = pd.DataFrame()\n",
    "submission['Id'] = range(y_pred.shape[0])\n",
    "submission['Predicted'] = np.exp(y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission.to_csv(\"submission78.csv\", index=False)"
   ]
  },
  {
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}