starter-notebook - Labs 1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "9d7f3214",
   "metadata": {
    "papermill": {
     "duration": 0.036252,
     "end_time": "2021-08-13T07:16:36.548737",
     "exception": false,
     "start_time": "2021-08-13T07:16:36.512485",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Telecom Churn Prediction - Starter Notebook\n",
    "\n",
    "**Author:** Akshay Sehgal (www.akshaysehgal.com)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fb5052e4",
   "metadata": {
    "papermill": {
     "duration": 0.035836,
     "end_time": "2021-08-13T07:16:36.620553",
     "exception": false,
     "start_time": "2021-08-13T07:16:36.584717",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "The goal of this notebook is to provide an overview of how write a notebook and create a submission file that successfully solves the churn prediction problem. Please download the datasets, unzip and place them in the same folder as this notebook.\n",
    "\n",
    "We are going to follow the process called CRISP-DM.\n",
    "\n",
    "<img src=\"https://upload.wikimedia.org/wikipedia/commons/thumb/b/b9/CRISP-DM_Process_Diagram.png/639px-CRISP-DM_Process_Diagram.png\" style=\"height: 400px; width:400px;\"/>\n",
    "\n",
    "After Business and Data Understanding via EDA, we want to prepare data for modelling. Then evaluate and submit our predictions."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9290ee78",
   "metadata": {
    "papermill": {
     "duration": 0.034552,
     "end_time": "2021-08-13T07:16:36.690028",
     "exception": false,
     "start_time": "2021-08-13T07:16:36.655476",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# 0. Problem statement\n",
    "\n",
    "In the telecom industry, customers are able to choose from multiple service providers and actively switch from one operator to another. In this highly competitive market, the telecommunications industry experiences an average of 15-25% annual churn rate. Given the fact that it costs 5-10 times more to acquire a new customer than to retain an existing one, customer retention has now become even more important than customer acquisition.\n",
    "\n",
    "For many incumbent operators, retaining high profitable customers is the number one business\n",
    "goal. To reduce customer churn, telecom companies need to predict which customers are at high risk of churn. In this project, you will analyze customer-level data of a leading telecom firm, build predictive models to identify customers at high risk of churn, and identify the main indicators of churn.\n",
    "\n",
    "In this competition, your goal is *to build a machine learning model that is able to predict churning customers based on the features provided for their usage.*\n",
    "\n",
    "**Customer behaviour during churn:**\n",
    "\n",
    "Customers usually do not decide to switch to another competitor instantly, but rather over a\n",
    "period of time (this is especially applicable to high-value customers). In churn prediction, we\n",
    "assume that there are three phases of customer lifecycle :\n",
    "\n",
    "1. <u>The ‘good’ phase:</u> In this phase, the customer is happy with the service and behaves as usual.\n",
    "\n",
    "2. <u>The ‘action’ phase:</u> The customer experience starts to sore in this phase, for e.g. he/she gets a compelling offer from a competitor, faces unjust charges, becomes unhappy with service quality etc. In this phase, the customer usually shows different behaviour than the ‘good’ months. It is crucial to identify high-churn-risk customers in this phase, since some corrective actions can be taken at this point (such as matching the competitor’s offer/improving the service quality etc.)\n",
    "\n",
    "3. <u>The ‘churn’ phase:</u> In this phase, the customer is said to have churned. In this case, since you are working over a four-month window, the first two months are the ‘good’ phase, the third month is the ‘action’ phase, while the fourth month (September) is the ‘churn’ phase."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3da713b1",
   "metadata": {
    "papermill": {
     "duration": 0.034501,
     "end_time": "2021-08-13T07:16:36.760335",
     "exception": false,
     "start_time": "2021-08-13T07:16:36.725834",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# 1. Loading dependencies & datasets\n",
    "\n",
    "Lets start by loading our dependencies. We can keep adding any imports to this cell block, as we write mode and mode code."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2e47254e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:36.838077Z",
     "iopub.status.busy": "2021-08-13T07:16:36.837038Z",
     "iopub.status.idle": "2021-08-13T07:16:38.158096Z",
     "shell.execute_reply": "2021-08-13T07:16:38.157302Z"
    },
    "papermill": {
     "duration": 1.362112,
     "end_time": "2021-08-13T07:16:38.158342",
     "exception": false,
     "start_time": "2021-08-13T07:16:36.796230",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "#Data Structures\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import re\n",
    "import os\n",
    "\n",
    "### For installing missingno library, type this command in terminal\n",
    "#pip install missingno\n",
    "\n",
    "import missingno as msno\n",
    "\n",
    "#Sklearn\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.manifold import TSNE\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.metrics import confusion_matrix, precision_score, recall_score\n",
    "\n",
    "#Plotting\n",
    "import matplotlib.pyplot as plt\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "import seaborn as sns\n",
    "\n",
    "#Others\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8d5046d0",
   "metadata": {
    "papermill": {
     "duration": 0.03468,
     "end_time": "2021-08-13T07:16:38.240579",
     "exception": false,
     "start_time": "2021-08-13T07:16:38.205899",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Next, we load our datasets and the data dictionary file.\n",
    "\n",
    "The **train.csv** file contains both dependent and independent features, while the **test.csv** contains only the independent variables. \n",
    "\n",
    "So, for model selection, I will create our own train/test dataset from the **train.csv** and use the model to predict the solution using the features in unseen test.csv data for submission."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e5474122",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:38.315005Z",
     "iopub.status.busy": "2021-08-13T07:16:38.314375Z",
     "iopub.status.idle": "2021-08-13T07:16:38.320099Z",
     "shell.execute_reply": "2021-08-13T07:16:38.319451Z"
    },
    "papermill": {
     "duration": 0.044801,
     "end_time": "2021-08-13T07:16:38.320264",
     "exception": false,
     "start_time": "2021-08-13T07:16:38.275463",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/kaggle/input/kagglelabstest2021/data_dictionary.csv\n",
      "/kaggle/input/kagglelabstest2021/sample.csv\n",
      "/kaggle/input/kagglelabstest2021/train.csv\n",
      "/kaggle/input/kagglelabstest2021/test.csv\n"
     ]
    }
   ],
   "source": [
    "#COMMENT THIS SECTION INCASE RUNNING THIS NOTEBOOK LOCALLY\n",
    "\n",
    "#Checking the kaggle paths for the uploaded datasets\n",
    "import os\n",
    "for dirname, _, filenames in os.walk('/kaggle/input'):\n",
    "    for filename in filenames:\n",
    "        print(os.path.join(dirname, filename))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3b94d1d5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:38.396407Z",
     "iopub.status.busy": "2021-08-13T07:16:38.395775Z",
     "iopub.status.idle": "2021-08-13T07:16:40.947829Z",
     "shell.execute_reply": "2021-08-13T07:16:40.948376Z"
    },
    "papermill": {
     "duration": 2.591587,
     "end_time": "2021-08-13T07:16:40.948543",
     "exception": false,
     "start_time": "2021-08-13T07:16:38.356956",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(69999, 172)\n",
      "(30000, 171)\n",
      "(30000, 2)\n",
      "(36, 2)\n"
     ]
    }
   ],
   "source": [
    "#INCASE RUNNING THIS LOCALLY, PASS THE RELATIVE PATH OF THE CSV FILES BELOW\n",
    "#(e.g. if files are in same folder as notebook, simple write \"train.csv\" as path)\n",
    "\n",
    "data = pd.read_csv(\"/kaggle/input/kagglelabstest2021/train.csv\")\n",
    "unseen = pd.read_csv(\"/kaggle/input/kagglelabstest2021/test.csv\")\n",
    "sample = pd.read_csv(\"/kaggle/input/kagglelabstest2021/sample.csv\")\n",
    "data_dict = pd.read_csv(\"/kaggle/input/kagglelabstest2021/data_dictionary.csv\")\n",
    "\n",
    "print(data.shape)\n",
    "print(unseen.shape)\n",
    "print(sample.shape)\n",
    "print(data_dict.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13614e3d",
   "metadata": {
    "papermill": {
     "duration": 0.035613,
     "end_time": "2021-08-13T07:16:41.020316",
     "exception": false,
     "start_time": "2021-08-13T07:16:40.984703",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "1. Lets analyze the data dictionary versus the churn dataset.\n",
    "2. The data dictonary contains a list of abbrevations which provide you all the information you need to understand what a specific feature/variable in the churn dataset represents\n",
    "3. Example: \n",
    "\n",
    "> \"arpu_7\" -> Average revenue per user + KPI for the month of July\n",
    ">\n",
    "> \"onnet_mou_6\" ->  All kind of calls within the same operator network + Minutes of usage voice calls + KPI for the month of June\n",
    ">\n",
    ">\"night_pck_user_8\" -> Scheme to use during specific night hours only + Prepaid service schemes called PACKS + KPI for the month of August\n",
    ">\n",
    ">\"max_rech_data_7\" -> Maximum + Recharge + Mobile internet + KPI for the month of July\n",
    "\n",
    "Its important to understand the definitions of each feature that you are working with, take notes on which feature you think might impact the churn rate of a user, and what sort of analysis could you do to understand the distribution of the feature better."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "adfb4dc5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:41.095411Z",
     "iopub.status.busy": "2021-08-13T07:16:41.094812Z",
     "iopub.status.idle": "2021-08-13T07:16:41.123189Z",
     "shell.execute_reply": "2021-08-13T07:16:41.122683Z"
    },
    "papermill": {
     "duration": 0.067353,
     "end_time": "2021-08-13T07:16:41.123346",
     "exception": false,
     "start_time": "2021-08-13T07:16:41.055993",
     "status": "completed"
    },
    "tags": []
   },
   "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>Acronyms</th>\n",
       "      <th>Description</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CIRCLE_ID</td>\n",
       "      <td>Telecom circle area to which the customer belo...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>LOC</td>\n",
       "      <td>Local calls  within same telecom circle</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>STD</td>\n",
       "      <td>STD calls  outside the calling circle</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>IC</td>\n",
       "      <td>Incoming calls</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>OG</td>\n",
       "      <td>Outgoing calls</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>T2T</td>\n",
       "      <td>Operator T to T ie within same operator mobile...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>T2M</td>\n",
       "      <td>Operator T to other operator mobile</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>T2O</td>\n",
       "      <td>Operator T to other operator fixed line</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>T2F</td>\n",
       "      <td>Operator T to fixed lines of T</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>T2C</td>\n",
       "      <td>Operator T to its own call center</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>ARPU</td>\n",
       "      <td>Average revenue per user</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>MOU</td>\n",
       "      <td>Minutes of usage  voice calls</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>AON</td>\n",
       "      <td>Age on network  number of days the customer is...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>ONNET</td>\n",
       "      <td>All kind of calls within the same operator net...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>OFFNET</td>\n",
       "      <td>All kind of calls outside the operator T network</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>ROAM</td>\n",
       "      <td>Indicates that customer is in roaming zone dur...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>SPL</td>\n",
       "      <td>Special calls</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>ISD</td>\n",
       "      <td>ISD calls</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>RECH</td>\n",
       "      <td>Recharge</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>NUM</td>\n",
       "      <td>Number</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>AMT</td>\n",
       "      <td>Amount in local currency</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>MAX</td>\n",
       "      <td>Maximum</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>DATA</td>\n",
       "      <td>Mobile internet</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>3G</td>\n",
       "      <td>G network</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>AV</td>\n",
       "      <td>Average</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>VOL</td>\n",
       "      <td>Mobile internet usage volume in MB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>2G</td>\n",
       "      <td>G network</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>PCK</td>\n",
       "      <td>Prepaid service schemes called  PACKS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>NIGHT</td>\n",
       "      <td>Scheme to use during specific night hours only</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>MONTHLY</td>\n",
       "      <td>Service schemes with validity equivalent to a ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>SACHET</td>\n",
       "      <td>Service schemes with validity smaller than a m...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>*.6</td>\n",
       "      <td>KPI for the month of June</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>*.7</td>\n",
       "      <td>KPI for the month of July</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>*.8</td>\n",
       "      <td>KPI for the month of August</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>FB_USER</td>\n",
       "      <td>Service scheme to avail services of Facebook a...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>VBC</td>\n",
       "      <td>Volume based cost  when no specific scheme is ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Acronyms                                        Description\n",
       "0     CIRCLE_ID  Telecom circle area to which the customer belo...\n",
       "1           LOC            Local calls  within same telecom circle\n",
       "2           STD              STD calls  outside the calling circle\n",
       "3            IC                                     Incoming calls\n",
       "4            OG                                     Outgoing calls\n",
       "5           T2T  Operator T to T ie within same operator mobile...\n",
       "6       T2M                    Operator T to other operator mobile\n",
       "7       T2O                Operator T to other operator fixed line\n",
       "8       T2F                         Operator T to fixed lines of T\n",
       "9       T2C                      Operator T to its own call center\n",
       "10     ARPU                               Average revenue per user\n",
       "11      MOU                          Minutes of usage  voice calls\n",
       "12      AON      Age on network  number of days the customer is...\n",
       "13     ONNET     All kind of calls within the same operator net...\n",
       "14   OFFNET       All kind of calls outside the operator T network\n",
       "15         ROAM  Indicates that customer is in roaming zone dur...\n",
       "16       SPL                                         Special calls\n",
       "17      ISD                                              ISD calls\n",
       "18     RECH                                               Recharge\n",
       "19      NUM                                                 Number\n",
       "20      AMT                               Amount in local currency\n",
       "21      MAX                                                Maximum\n",
       "22     DATA                                        Mobile internet\n",
       "23       3G                                              G network\n",
       "24       AV                                                Average\n",
       "25      VOL                     Mobile internet usage volume in MB\n",
       "26       2G                                              G network\n",
       "27      PCK                  Prepaid service schemes called  PACKS\n",
       "28    NIGHT         Scheme to use during specific night hours only\n",
       "29  MONTHLY      Service schemes with validity equivalent to a ...\n",
       "30    SACHET     Service schemes with validity smaller than a m...\n",
       "31      *.6                              KPI for the month of June\n",
       "32      *.7                              KPI for the month of July\n",
       "33      *.8                            KPI for the month of August\n",
       "34      FB_USER  Service scheme to avail services of Facebook a...\n",
       "35      VBC      Volume based cost  when no specific scheme is ..."
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_dict"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d91e8d29",
   "metadata": {
    "papermill": {
     "duration": 0.036844,
     "end_time": "2021-08-13T07:16:41.198447",
     "exception": false,
     "start_time": "2021-08-13T07:16:41.161603",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "For the purpose of this **starter notebook**, we I will restrict the dataset to only a small set of variables. \n",
    "\n",
    "The approach I use here is to understand each Acronym, figure our what variable might be important and filter out variable names based on the combinations of acrynoms using REGEX. So, if I want the total minutes a person has spent on outgoing calls, I need acronyms, TOTAL, OG and MOU. So corresponding regex is ```total.+og.+mou```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "78b62d9f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:41.286772Z",
     "iopub.status.busy": "2021-08-13T07:16:41.285532Z",
     "iopub.status.idle": "2021-08-13T07:16:41.298276Z",
     "shell.execute_reply": "2021-08-13T07:16:41.298717Z"
    },
    "papermill": {
     "duration": 0.06318,
     "end_time": "2021-08-13T07:16:41.298901",
     "exception": false,
     "start_time": "2021-08-13T07:16:41.235721",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "ids = ['id','circle_id']\n",
    "total_amounts = [i for i in list(data.columns) if re.search('total.+amt',i)]\n",
    "total_outgoing_minutes = [i for i in list(data.columns) if re.search('total.+og.+mou',i)]\n",
    "offnetwork_minutes = [i for i in list(data.columns) if re.search('offnet',i)]\n",
    "average_revenue_3g = [i for i in list(data.columns) if re.search('arpu.+3g',i)]\n",
    "average_revenue_2g = [i for i in list(data.columns) if re.search('arpu.+2g',i)]\n",
    "volume_3g = [i for i in list(data.columns) if re.search('vol.+3g',i)]\n",
    "volume_2g = [i for i in list(data.columns) if re.search('vol.+2g',i)]\n",
    "age_on_network = [i for i in list(data.columns) if re.search('aon',i)]\n",
    "\n",
    "#Storing them in a single flat list\n",
    "variables = [*ids, \n",
    "             *total_amounts, \n",
    "             *total_outgoing_minutes, \n",
    "             *offnetwork_minutes, \n",
    "             *average_revenue_3g, \n",
    "             *average_revenue_2g,\n",
    "             *volume_3g,\n",
    "             *volume_2g,\n",
    "             *age_on_network, \n",
    "             'churn_probability']\n",
    "\n",
    "data = data[variables].set_index('id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d40bdb11",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:41.375999Z",
     "iopub.status.busy": "2021-08-13T07:16:41.375128Z",
     "iopub.status.idle": "2021-08-13T07:16:41.401438Z",
     "shell.execute_reply": "2021-08-13T07:16:41.400962Z"
    },
    "papermill": {
     "duration": 0.066079,
     "end_time": "2021-08-13T07:16:41.401769",
     "exception": false,
     "start_time": "2021-08-13T07:16:41.335690",
     "status": "completed"
    },
    "tags": []
   },
   "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>circle_id</th>\n",
       "      <th>total_rech_amt_6</th>\n",
       "      <th>total_rech_amt_7</th>\n",
       "      <th>total_rech_amt_8</th>\n",
       "      <th>total_og_mou_6</th>\n",
       "      <th>total_og_mou_7</th>\n",
       "      <th>total_og_mou_8</th>\n",
       "      <th>offnet_mou_6</th>\n",
       "      <th>offnet_mou_7</th>\n",
       "      <th>offnet_mou_8</th>\n",
       "      <th>...</th>\n",
       "      <th>arpu_2g_7</th>\n",
       "      <th>arpu_2g_8</th>\n",
       "      <th>vol_3g_mb_6</th>\n",
       "      <th>vol_3g_mb_7</th>\n",
       "      <th>vol_3g_mb_8</th>\n",
       "      <th>vol_2g_mb_6</th>\n",
       "      <th>vol_2g_mb_7</th>\n",
       "      <th>vol_2g_mb_8</th>\n",
       "      <th>aon</th>\n",
       "      <th>churn_probability</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>109</td>\n",
       "      <td>77</td>\n",
       "      <td>65</td>\n",
       "      <td>10</td>\n",
       "      <td>81.21</td>\n",
       "      <td>221.68</td>\n",
       "      <td>3.63</td>\n",
       "      <td>32.24</td>\n",
       "      <td>96.68</td>\n",
       "      <td>2.33</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1958</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>109</td>\n",
       "      <td>0</td>\n",
       "      <td>145</td>\n",
       "      <td>50</td>\n",
       "      <td>0.00</td>\n",
       "      <td>30.73</td>\n",
       "      <td>31.66</td>\n",
       "      <td>0.00</td>\n",
       "      <td>25.99</td>\n",
       "      <td>30.89</td>\n",
       "      <td>...</td>\n",
       "      <td>122.08</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.96</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>352.91</td>\n",
       "      <td>0.00</td>\n",
       "      <td>710</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>109</td>\n",
       "      <td>70</td>\n",
       "      <td>120</td>\n",
       "      <td>0</td>\n",
       "      <td>56.49</td>\n",
       "      <td>99.36</td>\n",
       "      <td>0.00</td>\n",
       "      <td>53.99</td>\n",
       "      <td>82.05</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>882</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>109</td>\n",
       "      <td>160</td>\n",
       "      <td>240</td>\n",
       "      <td>130</td>\n",
       "      <td>76.03</td>\n",
       "      <td>95.98</td>\n",
       "      <td>53.84</td>\n",
       "      <td>68.76</td>\n",
       "      <td>78.48</td>\n",
       "      <td>50.23</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>982</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>109</td>\n",
       "      <td>290</td>\n",
       "      <td>136</td>\n",
       "      <td>122</td>\n",
       "      <td>63.26</td>\n",
       "      <td>42.94</td>\n",
       "      <td>15.76</td>\n",
       "      <td>56.99</td>\n",
       "      <td>38.11</td>\n",
       "      <td>9.63</td>\n",
       "      <td>...</td>\n",
       "      <td>35.12</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>390.8</td>\n",
       "      <td>308.89</td>\n",
       "      <td>213.47</td>\n",
       "      <td>647</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    circle_id  total_rech_amt_6  total_rech_amt_7  total_rech_amt_8  \\\n",
       "id                                                                    \n",
       "0         109                77                65                10   \n",
       "1         109                 0               145                50   \n",
       "2         109                70               120                 0   \n",
       "3         109               160               240               130   \n",
       "4         109               290               136               122   \n",
       "\n",
       "    total_og_mou_6  total_og_mou_7  total_og_mou_8  offnet_mou_6  \\\n",
       "id                                                                 \n",
       "0            81.21          221.68            3.63         32.24   \n",
       "1             0.00           30.73           31.66          0.00   \n",
       "2            56.49           99.36            0.00         53.99   \n",
       "3            76.03           95.98           53.84         68.76   \n",
       "4            63.26           42.94           15.76         56.99   \n",
       "\n",
       "    offnet_mou_7  offnet_mou_8  ...  arpu_2g_7  arpu_2g_8  vol_3g_mb_6  \\\n",
       "id                              ...                                      \n",
       "0          96.68          2.33  ...        NaN        NaN          0.0   \n",
       "1          25.99         30.89  ...     122.08        NaN          0.0   \n",
       "2          82.05          0.00  ...        NaN        NaN          0.0   \n",
       "3          78.48         50.23  ...        NaN        NaN          0.0   \n",
       "4          38.11          9.63  ...      35.12        0.0          0.0   \n",
       "\n",
       "    vol_3g_mb_7  vol_3g_mb_8  vol_2g_mb_6  vol_2g_mb_7  vol_2g_mb_8   aon  \\\n",
       "id                                                                          \n",
       "0          0.00          0.0          0.0         0.00         0.00  1958   \n",
       "1          3.96          0.0          0.0       352.91         0.00   710   \n",
       "2          0.00          0.0          0.0         0.00         0.00   882   \n",
       "3          0.00          0.0          0.0         0.00         0.00   982   \n",
       "4          0.00          0.0        390.8       308.89       213.47   647   \n",
       "\n",
       "    churn_probability  \n",
       "id                     \n",
       "0                   0  \n",
       "1                   0  \n",
       "2                   0  \n",
       "3                   0  \n",
       "4                   0  \n",
       "\n",
       "[5 rows x 24 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e3fb9add",
   "metadata": {
    "papermill": {
     "duration": 0.037694,
     "end_time": "2021-08-13T07:16:41.476864",
     "exception": false,
     "start_time": "2021-08-13T07:16:41.439170",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Let's look at each variable's datatype:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "cbd5d726",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:41.557516Z",
     "iopub.status.busy": "2021-08-13T07:16:41.556428Z",
     "iopub.status.idle": "2021-08-13T07:16:41.578390Z",
     "shell.execute_reply": "2021-08-13T07:16:41.577859Z"
    },
    "papermill": {
     "duration": 0.064419,
     "end_time": "2021-08-13T07:16:41.578526",
     "exception": false,
     "start_time": "2021-08-13T07:16:41.514107",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 69999 entries, 0 to 69998\n",
      "Data columns (total 24 columns):\n",
      " #   Column             Non-Null Count  Dtype  \n",
      "---  ------             --------------  -----  \n",
      " 0   circle_id          69999 non-null  int64  \n",
      " 1   total_rech_amt_6   69999 non-null  int64  \n",
      " 2   total_rech_amt_7   69999 non-null  int64  \n",
      " 3   total_rech_amt_8   69999 non-null  int64  \n",
      " 4   total_og_mou_6     69999 non-null  float64\n",
      " 5   total_og_mou_7     69999 non-null  float64\n",
      " 6   total_og_mou_8     69999 non-null  float64\n",
      " 7   offnet_mou_6       67231 non-null  float64\n",
      " 8   offnet_mou_7       67312 non-null  float64\n",
      " 9   offnet_mou_8       66296 non-null  float64\n",
      " 10  arpu_3g_6          17568 non-null  float64\n",
      " 11  arpu_3g_7          17865 non-null  float64\n",
      " 12  arpu_3g_8          18417 non-null  float64\n",
      " 13  arpu_2g_6          17568 non-null  float64\n",
      " 14  arpu_2g_7          17865 non-null  float64\n",
      " 15  arpu_2g_8          18417 non-null  float64\n",
      " 16  vol_3g_mb_6        69999 non-null  float64\n",
      " 17  vol_3g_mb_7        69999 non-null  float64\n",
      " 18  vol_3g_mb_8        69999 non-null  float64\n",
      " 19  vol_2g_mb_6        69999 non-null  float64\n",
      " 20  vol_2g_mb_7        69999 non-null  float64\n",
      " 21  vol_2g_mb_8        69999 non-null  float64\n",
      " 22  aon                69999 non-null  int64  \n",
      " 23  churn_probability  69999 non-null  int64  \n",
      "dtypes: float64(18), int64(6)\n",
      "memory usage: 13.4 MB\n"
     ]
    }
   ],
   "source": [
    "data.info(verbose=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f71a0be",
   "metadata": {
    "papermill": {
     "duration": 0.037487,
     "end_time": "2021-08-13T07:16:41.653911",
     "exception": false,
     "start_time": "2021-08-13T07:16:41.616424",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Let's also summarize the features using the df.describe method:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "0ed1a2ee",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:41.734400Z",
     "iopub.status.busy": "2021-08-13T07:16:41.733760Z",
     "iopub.status.idle": "2021-08-13T07:16:41.851570Z",
     "shell.execute_reply": "2021-08-13T07:16:41.852036Z"
    },
    "papermill": {
     "duration": 0.160583,
     "end_time": "2021-08-13T07:16:41.852218",
     "exception": false,
     "start_time": "2021-08-13T07:16:41.691635",
     "status": "completed"
    },
    "tags": []
   },
   "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>circle_id</th>\n",
       "      <th>total_rech_amt_6</th>\n",
       "      <th>total_rech_amt_7</th>\n",
       "      <th>total_rech_amt_8</th>\n",
       "      <th>total_og_mou_6</th>\n",
       "      <th>total_og_mou_7</th>\n",
       "      <th>total_og_mou_8</th>\n",
       "      <th>offnet_mou_6</th>\n",
       "      <th>offnet_mou_7</th>\n",
       "      <th>offnet_mou_8</th>\n",
       "      <th>...</th>\n",
       "      <th>arpu_2g_7</th>\n",
       "      <th>arpu_2g_8</th>\n",
       "      <th>vol_3g_mb_6</th>\n",
       "      <th>vol_3g_mb_7</th>\n",
       "      <th>vol_3g_mb_8</th>\n",
       "      <th>vol_2g_mb_6</th>\n",
       "      <th>vol_2g_mb_7</th>\n",
       "      <th>vol_2g_mb_8</th>\n",
       "      <th>aon</th>\n",
       "      <th>churn_probability</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>69999.0</td>\n",
       "      <td>69999.000000</td>\n",
       "      <td>69999.000000</td>\n",
       "      <td>69999.000000</td>\n",
       "      <td>69999.000000</td>\n",
       "      <td>69999.000000</td>\n",
       "      <td>69999.000000</td>\n",
       "      <td>67231.000000</td>\n",
       "      <td>67312.000000</td>\n",
       "      <td>66296.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>17865.000000</td>\n",
       "      <td>18417.000000</td>\n",
       "      <td>69999.000000</td>\n",
       "      <td>69999.000000</td>\n",
       "      <td>69999.000000</td>\n",
       "      <td>69999.000000</td>\n",
       "      <td>69999.000000</td>\n",
       "      <td>69999.000000</td>\n",
       "      <td>69999.000000</td>\n",
       "      <td>69999.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>109.0</td>\n",
       "      <td>328.139788</td>\n",
       "      <td>322.376363</td>\n",
       "      <td>323.846355</td>\n",
       "      <td>306.451436</td>\n",
       "      <td>310.572674</td>\n",
       "      <td>304.513065</td>\n",
       "      <td>198.874771</td>\n",
       "      <td>197.153383</td>\n",
       "      <td>196.543577</td>\n",
       "      <td>...</td>\n",
       "      <td>85.846074</td>\n",
       "      <td>86.348404</td>\n",
       "      <td>122.171882</td>\n",
       "      <td>128.934444</td>\n",
       "      <td>135.486541</td>\n",
       "      <td>51.773924</td>\n",
       "      <td>51.240204</td>\n",
       "      <td>50.127506</td>\n",
       "      <td>1220.639709</td>\n",
       "      <td>0.101887</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.0</td>\n",
       "      <td>404.211068</td>\n",
       "      <td>411.070120</td>\n",
       "      <td>426.181405</td>\n",
       "      <td>465.502866</td>\n",
       "      <td>479.131770</td>\n",
       "      <td>477.936832</td>\n",
       "      <td>316.818355</td>\n",
       "      <td>322.482226</td>\n",
       "      <td>324.089234</td>\n",
       "      <td>...</td>\n",
       "      <td>178.067280</td>\n",
       "      <td>170.297094</td>\n",
       "      <td>554.869965</td>\n",
       "      <td>554.096072</td>\n",
       "      <td>568.310234</td>\n",
       "      <td>212.513909</td>\n",
       "      <td>211.114667</td>\n",
       "      <td>213.101403</td>\n",
       "      <td>952.426321</td>\n",
       "      <td>0.302502</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>109.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>-13.090000</td>\n",
       "      <td>-55.830000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>180.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>109.0</td>\n",
       "      <td>110.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>44.780000</td>\n",
       "      <td>42.910000</td>\n",
       "      <td>38.710000</td>\n",
       "      <td>34.860000</td>\n",
       "      <td>32.240000</td>\n",
       "      <td>31.575000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>468.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>109.0</td>\n",
       "      <td>229.000000</td>\n",
       "      <td>220.000000</td>\n",
       "      <td>225.000000</td>\n",
       "      <td>145.280000</td>\n",
       "      <td>141.230000</td>\n",
       "      <td>138.360000</td>\n",
       "      <td>96.480000</td>\n",
       "      <td>91.885000</td>\n",
       "      <td>91.800000</td>\n",
       "      <td>...</td>\n",
       "      <td>8.800000</td>\n",
       "      <td>9.090000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>868.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>109.0</td>\n",
       "      <td>438.000000</td>\n",
       "      <td>430.000000</td>\n",
       "      <td>436.000000</td>\n",
       "      <td>374.305000</td>\n",
       "      <td>380.045000</td>\n",
       "      <td>370.895000</td>\n",
       "      <td>232.990000</td>\n",
       "      <td>227.630000</td>\n",
       "      <td>229.345000</td>\n",
       "      <td>...</td>\n",
       "      <td>122.070000</td>\n",
       "      <td>122.070000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1813.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>109.0</td>\n",
       "      <td>35190.000000</td>\n",
       "      <td>40335.000000</td>\n",
       "      <td>45320.000000</td>\n",
       "      <td>10674.030000</td>\n",
       "      <td>8285.640000</td>\n",
       "      <td>14043.060000</td>\n",
       "      <td>8362.360000</td>\n",
       "      <td>7043.980000</td>\n",
       "      <td>14007.340000</td>\n",
       "      <td>...</td>\n",
       "      <td>4809.360000</td>\n",
       "      <td>3483.170000</td>\n",
       "      <td>45735.400000</td>\n",
       "      <td>28144.120000</td>\n",
       "      <td>30036.060000</td>\n",
       "      <td>10285.900000</td>\n",
       "      <td>7873.550000</td>\n",
       "      <td>11117.610000</td>\n",
       "      <td>4337.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       circle_id  total_rech_amt_6  total_rech_amt_7  total_rech_amt_8  \\\n",
       "count    69999.0      69999.000000      69999.000000      69999.000000   \n",
       "mean       109.0        328.139788        322.376363        323.846355   \n",
       "std          0.0        404.211068        411.070120        426.181405   \n",
       "min        109.0          0.000000          0.000000          0.000000   \n",
       "25%        109.0        110.000000        100.000000         90.000000   \n",
       "50%        109.0        229.000000        220.000000        225.000000   \n",
       "75%        109.0        438.000000        430.000000        436.000000   \n",
       "max        109.0      35190.000000      40335.000000      45320.000000   \n",
       "\n",
       "       total_og_mou_6  total_og_mou_7  total_og_mou_8  offnet_mou_6  \\\n",
       "count    69999.000000    69999.000000    69999.000000  67231.000000   \n",
       "mean       306.451436      310.572674      304.513065    198.874771   \n",
       "std        465.502866      479.131770      477.936832    316.818355   \n",
       "min          0.000000        0.000000        0.000000      0.000000   \n",
       "25%         44.780000       42.910000       38.710000     34.860000   \n",
       "50%        145.280000      141.230000      138.360000     96.480000   \n",
       "75%        374.305000      380.045000      370.895000    232.990000   \n",
       "max      10674.030000     8285.640000    14043.060000   8362.360000   \n",
       "\n",
       "       offnet_mou_7  offnet_mou_8  ...     arpu_2g_7     arpu_2g_8  \\\n",
       "count  67312.000000  66296.000000  ...  17865.000000  18417.000000   \n",
       "mean     197.153383    196.543577  ...     85.846074     86.348404   \n",
       "std      322.482226    324.089234  ...    178.067280    170.297094   \n",
       "min        0.000000      0.000000  ...    -13.090000    -55.830000   \n",
       "25%       32.240000     31.575000  ...      0.000000      0.000000   \n",
       "50%       91.885000     91.800000  ...      8.800000      9.090000   \n",
       "75%      227.630000    229.345000  ...    122.070000    122.070000   \n",
       "max     7043.980000  14007.340000  ...   4809.360000   3483.170000   \n",
       "\n",
       "        vol_3g_mb_6   vol_3g_mb_7   vol_3g_mb_8   vol_2g_mb_6   vol_2g_mb_7  \\\n",
       "count  69999.000000  69999.000000  69999.000000  69999.000000  69999.000000   \n",
       "mean     122.171882    128.934444    135.486541     51.773924     51.240204   \n",
       "std      554.869965    554.096072    568.310234    212.513909    211.114667   \n",
       "min        0.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "25%        0.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "50%        0.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "75%        0.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "max    45735.400000  28144.120000  30036.060000  10285.900000   7873.550000   \n",
       "\n",
       "        vol_2g_mb_8           aon  churn_probability  \n",
       "count  69999.000000  69999.000000       69999.000000  \n",
       "mean      50.127506   1220.639709           0.101887  \n",
       "std      213.101403    952.426321           0.302502  \n",
       "min        0.000000    180.000000           0.000000  \n",
       "25%        0.000000    468.000000           0.000000  \n",
       "50%        0.000000    868.000000           0.000000  \n",
       "75%        0.000000   1813.000000           0.000000  \n",
       "max    11117.610000   4337.000000           1.000000  \n",
       "\n",
       "[8 rows x 24 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe(include=\"all\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "717a47b1",
   "metadata": {
    "papermill": {
     "duration": 0.03869,
     "end_time": "2021-08-13T07:16:41.929599",
     "exception": false,
     "start_time": "2021-08-13T07:16:41.890909",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# 2. Create X, y and then Train test split\n",
    "\n",
    "Lets create X and y datasets and skip \"circle_id\" since it has only 1 unique value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "17d8d0f3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:42.012089Z",
     "iopub.status.busy": "2021-08-13T07:16:42.011484Z",
     "iopub.status.idle": "2021-08-13T07:16:42.016893Z",
     "shell.execute_reply": "2021-08-13T07:16:42.016407Z"
    },
    "papermill": {
     "duration": 0.04928,
     "end_time": "2021-08-13T07:16:42.017029",
     "exception": false,
     "start_time": "2021-08-13T07:16:41.967749",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([109])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['circle_id'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "bd2a70a2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:42.107906Z",
     "iopub.status.busy": "2021-08-13T07:16:42.107312Z",
     "iopub.status.idle": "2021-08-13T07:16:42.111099Z",
     "shell.execute_reply": "2021-08-13T07:16:42.110536Z"
    },
    "papermill": {
     "duration": 0.054463,
     "end_time": "2021-08-13T07:16:42.111249",
     "exception": false,
     "start_time": "2021-08-13T07:16:42.056786",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((69999, 22), (69999,))"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = data.drop(['circle_id'],1).iloc[:,:-1]\n",
    "y = data.iloc[:,-1]\n",
    "\n",
    "X.shape, y.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b1e9eca",
   "metadata": {
    "papermill": {
     "duration": 0.039731,
     "end_time": "2021-08-13T07:16:42.189842",
     "exception": false,
     "start_time": "2021-08-13T07:16:42.150111",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Splitting train and test data to avoid any contamination of the test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "19049f7a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:42.274868Z",
     "iopub.status.busy": "2021-08-13T07:16:42.274265Z",
     "iopub.status.idle": "2021-08-13T07:16:42.293801Z",
     "shell.execute_reply": "2021-08-13T07:16:42.294259Z"
    },
    "papermill": {
     "duration": 0.065525,
     "end_time": "2021-08-13T07:16:42.294433",
     "exception": false,
     "start_time": "2021-08-13T07:16:42.228908",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((55999, 22), (14000, 22), (55999,), (14000,))"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)\n",
    "\n",
    "X_train.shape, X_test.shape, y_train.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "383c1ddb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:42.376039Z",
     "iopub.status.busy": "2021-08-13T07:16:42.375133Z",
     "iopub.status.idle": "2021-08-13T07:16:42.401238Z",
     "shell.execute_reply": "2021-08-13T07:16:42.401707Z"
    },
    "papermill": {
     "duration": 0.068615,
     "end_time": "2021-08-13T07:16:42.402101",
     "exception": false,
     "start_time": "2021-08-13T07:16:42.333486",
     "status": "completed"
    },
    "tags": []
   },
   "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>total_rech_amt_6</th>\n",
       "      <th>total_rech_amt_7</th>\n",
       "      <th>total_rech_amt_8</th>\n",
       "      <th>total_og_mou_6</th>\n",
       "      <th>total_og_mou_7</th>\n",
       "      <th>total_og_mou_8</th>\n",
       "      <th>offnet_mou_6</th>\n",
       "      <th>offnet_mou_7</th>\n",
       "      <th>offnet_mou_8</th>\n",
       "      <th>arpu_3g_6</th>\n",
       "      <th>...</th>\n",
       "      <th>arpu_2g_6</th>\n",
       "      <th>arpu_2g_7</th>\n",
       "      <th>arpu_2g_8</th>\n",
       "      <th>vol_3g_mb_6</th>\n",
       "      <th>vol_3g_mb_7</th>\n",
       "      <th>vol_3g_mb_8</th>\n",
       "      <th>vol_2g_mb_6</th>\n",
       "      <th>vol_2g_mb_7</th>\n",
       "      <th>vol_2g_mb_8</th>\n",
       "      <th>aon</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>68230</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>212</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>413.36</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>356.11</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>479</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35918</th>\n",
       "      <td>90</td>\n",
       "      <td>100</td>\n",
       "      <td>47</td>\n",
       "      <td>1.16</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.80</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20.60</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>123.60</td>\n",
       "      <td>767</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33141</th>\n",
       "      <td>193</td>\n",
       "      <td>240</td>\n",
       "      <td>335</td>\n",
       "      <td>81.43</td>\n",
       "      <td>89.08</td>\n",
       "      <td>90.91</td>\n",
       "      <td>76.09</td>\n",
       "      <td>85.86</td>\n",
       "      <td>85.69</td>\n",
       "      <td>58.65</td>\n",
       "      <td>...</td>\n",
       "      <td>68.7</td>\n",
       "      <td>109.78</td>\n",
       "      <td>82.38</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>271.49</td>\n",
       "      <td>280.33</td>\n",
       "      <td>678.31</td>\n",
       "      <td>1248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020</th>\n",
       "      <td>10</td>\n",
       "      <td>120</td>\n",
       "      <td>120</td>\n",
       "      <td>54.63</td>\n",
       "      <td>125.98</td>\n",
       "      <td>80.78</td>\n",
       "      <td>42.46</td>\n",
       "      <td>105.98</td>\n",
       "      <td>56.06</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>853</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56082</th>\n",
       "      <td>177</td>\n",
       "      <td>230</td>\n",
       "      <td>160</td>\n",
       "      <td>145.88</td>\n",
       "      <td>129.19</td>\n",
       "      <td>109.04</td>\n",
       "      <td>127.88</td>\n",
       "      <td>122.01</td>\n",
       "      <td>106.08</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1905</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       total_rech_amt_6  total_rech_amt_7  total_rech_amt_8  total_og_mou_6  \\\n",
       "id                                                                            \n",
       "68230                 0                 0               212            0.00   \n",
       "35918                90               100                47            1.16   \n",
       "33141               193               240               335           81.43   \n",
       "2020                 10               120               120           54.63   \n",
       "56082               177               230               160          145.88   \n",
       "\n",
       "       total_og_mou_7  total_og_mou_8  offnet_mou_6  offnet_mou_7  \\\n",
       "id                                                                  \n",
       "68230            0.00          413.36          0.00          0.00   \n",
       "35918            0.00            0.00          0.80          0.00   \n",
       "33141           89.08           90.91         76.09         85.86   \n",
       "2020           125.98           80.78         42.46        105.98   \n",
       "56082          129.19          109.04        127.88        122.01   \n",
       "\n",
       "       offnet_mou_8  arpu_3g_6  ...  arpu_2g_6  arpu_2g_7  arpu_2g_8  \\\n",
       "id                              ...                                    \n",
       "68230        356.11        NaN  ...        NaN        NaN        NaN   \n",
       "35918          0.00        NaN  ...        NaN        NaN      20.60   \n",
       "33141         85.69      58.65  ...       68.7     109.78      82.38   \n",
       "2020          56.06        NaN  ...        NaN        NaN        NaN   \n",
       "56082        106.08       0.00  ...        0.0        NaN        NaN   \n",
       "\n",
       "       vol_3g_mb_6  vol_3g_mb_7  vol_3g_mb_8  vol_2g_mb_6  vol_2g_mb_7  \\\n",
       "id                                                                       \n",
       "68230          0.0          0.0          0.0         0.00         0.00   \n",
       "35918          0.0          0.0          0.0         0.00         0.00   \n",
       "33141          0.0          0.0          0.0       271.49       280.33   \n",
       "2020           0.0          0.0          0.0         0.00         0.00   \n",
       "56082          0.0          0.0          0.0         0.00         0.00   \n",
       "\n",
       "       vol_2g_mb_8   aon  \n",
       "id                        \n",
       "68230         0.00   479  \n",
       "35918       123.60   767  \n",
       "33141       678.31  1248  \n",
       "2020          0.00   853  \n",
       "56082         0.00  1905  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6a29fbf5",
   "metadata": {
    "papermill": {
     "duration": 0.039575,
     "end_time": "2021-08-13T07:16:42.482838",
     "exception": false,
     "start_time": "2021-08-13T07:16:42.443263",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# 3. Handling Missing data\n",
    "\n",
    "First lets analyse the missing data. We can use missingno library for quick visualizations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6360a919",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:42.565967Z",
     "iopub.status.busy": "2021-08-13T07:16:42.565075Z",
     "iopub.status.idle": "2021-08-13T07:16:44.221811Z",
     "shell.execute_reply": "2021-08-13T07:16:44.222300Z"
    },
    "papermill": {
     "duration": 1.699725,
     "end_time": "2021-08-13T07:16:44.222465",
     "exception": false,
     "start_time": "2021-08-13T07:16:42.522740",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1728x720 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "msno.bar(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "1217cc5e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:44.310021Z",
     "iopub.status.busy": "2021-08-13T07:16:44.309094Z",
     "iopub.status.idle": "2021-08-13T07:16:45.259337Z",
     "shell.execute_reply": "2021-08-13T07:16:45.259749Z"
    },
    "papermill": {
     "duration": 0.995529,
     "end_time": "2021-08-13T07:16:45.259912",
     "exception": false,
     "start_time": "2021-08-13T07:16:44.264383",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "msno.matrix(X_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d509008",
   "metadata": {
    "papermill": {
     "duration": 0.043599,
     "end_time": "2021-08-13T07:16:45.347986",
     "exception": false,
     "start_time": "2021-08-13T07:16:45.304387",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Lets also calculate the % missing data for each column:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "910cbe80",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:45.443329Z",
     "iopub.status.busy": "2021-08-13T07:16:45.442761Z",
     "iopub.status.idle": "2021-08-13T07:16:45.453451Z",
     "shell.execute_reply": "2021-08-13T07:16:45.453853Z"
    },
    "papermill": {
     "duration": 0.06183,
     "end_time": "2021-08-13T07:16:45.454014",
     "exception": false,
     "start_time": "2021-08-13T07:16:45.392184",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "total_rech_amt_6     0.000000\n",
       "total_rech_amt_7     0.000000\n",
       "total_rech_amt_8     0.000000\n",
       "total_og_mou_6       0.000000\n",
       "total_og_mou_7       0.000000\n",
       "total_og_mou_8       0.000000\n",
       "offnet_mou_6         3.955428\n",
       "offnet_mou_7         3.898284\n",
       "offnet_mou_8         5.328667\n",
       "arpu_3g_6           74.899552\n",
       "arpu_3g_7           74.458472\n",
       "arpu_3g_8           73.792389\n",
       "arpu_2g_6           74.899552\n",
       "arpu_2g_7           74.458472\n",
       "arpu_2g_8           73.792389\n",
       "vol_3g_mb_6          0.000000\n",
       "vol_3g_mb_7          0.000000\n",
       "vol_3g_mb_8          0.000000\n",
       "vol_2g_mb_6          0.000000\n",
       "vol_2g_mb_7          0.000000\n",
       "vol_2g_mb_8          0.000000\n",
       "aon                  0.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "missing_data_percent = 100*X_train.isnull().sum()/len(y_train)\n",
    "missing_data_percent"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "30dbf492",
   "metadata": {
    "papermill": {
     "duration": 0.043963,
     "end_time": "2021-08-13T07:16:45.542693",
     "exception": false,
     "start_time": "2021-08-13T07:16:45.498730",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Since too much missing information would make a column not really a great predictor for churn, we drop these columns and keep only the ones which have less than 40% missing data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "e94c195c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:45.634867Z",
     "iopub.status.busy": "2021-08-13T07:16:45.634230Z",
     "iopub.status.idle": "2021-08-13T07:16:45.641492Z",
     "shell.execute_reply": "2021-08-13T07:16:45.640994Z"
    },
    "papermill": {
     "duration": 0.054309,
     "end_time": "2021-08-13T07:16:45.641625",
     "exception": false,
     "start_time": "2021-08-13T07:16:45.587316",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['total_rech_amt_6', 'total_rech_amt_7', 'total_rech_amt_8',\n",
       "       'total_og_mou_6', 'total_og_mou_7', 'total_og_mou_8', 'offnet_mou_6',\n",
       "       'offnet_mou_7', 'offnet_mou_8', 'vol_3g_mb_6', 'vol_3g_mb_7',\n",
       "       'vol_3g_mb_8', 'vol_2g_mb_6', 'vol_2g_mb_7', 'vol_2g_mb_8', 'aon'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_vars = missing_data_percent[missing_data_percent.le(40)].index\n",
    "new_vars"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "1935c440",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:45.734737Z",
     "iopub.status.busy": "2021-08-13T07:16:45.734145Z",
     "iopub.status.idle": "2021-08-13T07:16:45.742290Z",
     "shell.execute_reply": "2021-08-13T07:16:45.741799Z"
    },
    "papermill": {
     "duration": 0.05574,
     "end_time": "2021-08-13T07:16:45.742430",
     "exception": false,
     "start_time": "2021-08-13T07:16:45.686690",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(55999, 16)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_filtered = X_train[new_vars]\n",
    "X_train_filtered.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a60c8a2f",
   "metadata": {
    "papermill": {
     "duration": 0.045444,
     "end_time": "2021-08-13T07:16:45.832903",
     "exception": false,
     "start_time": "2021-08-13T07:16:45.787459",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Next, we try imputation on variables with any amount of missing data still left. There are multiple ways of imputing data, and each will require a good business understanding of what the missing data is and how you may handle it.\n",
    "\n",
    "Some tips while working with missing data - \n",
    "\n",
    "1. Can simply replace missing values directly with a constant value such as 0\n",
    "2. In certain cases you may want to replace it with the average value for each column respectively\n",
    "3. For timeseries data, you may consider using linear or spline interplolation between a set of points, if you have data available for some of the months, and missing for the others.\n",
    "4. You can consider more advance methods for imputation such as MICE.\n",
    "\n",
    "In our case, I will just demostrate a simple imputation with constant values as zeros."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "7ece2cff",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:45.927493Z",
     "iopub.status.busy": "2021-08-13T07:16:45.926849Z",
     "iopub.status.idle": "2021-08-13T07:16:45.934908Z",
     "shell.execute_reply": "2021-08-13T07:16:45.934456Z"
    },
    "papermill": {
     "duration": 0.056957,
     "end_time": "2021-08-13T07:16:45.935041",
     "exception": false,
     "start_time": "2021-08-13T07:16:45.878084",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['offnet_mou_6', 'offnet_mou_7', 'offnet_mou_8'], dtype='object')"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "missing_data_percent = X_train_filtered.isnull().any()\n",
    "impute_cols = missing_data_percent[missing_data_percent.gt(0)].index\n",
    "impute_cols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "cb1574ed",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:46.034023Z",
     "iopub.status.busy": "2021-08-13T07:16:46.032946Z",
     "iopub.status.idle": "2021-08-13T07:16:46.065689Z",
     "shell.execute_reply": "2021-08-13T07:16:46.065145Z"
    },
    "papermill": {
     "duration": 0.085462,
     "end_time": "2021-08-13T07:16:46.065826",
     "exception": false,
     "start_time": "2021-08-13T07:16:45.980364",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "imp = SimpleImputer(strategy='constant', fill_value=0)\n",
    "X_train_filtered[impute_cols] = imp.fit_transform(X_train_filtered[impute_cols])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "fb20d440",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:46.161326Z",
     "iopub.status.busy": "2021-08-13T07:16:46.160412Z",
     "iopub.status.idle": "2021-08-13T07:16:47.024882Z",
     "shell.execute_reply": "2021-08-13T07:16:47.024341Z"
    },
    "papermill": {
     "duration": 0.914023,
     "end_time": "2021-08-13T07:16:47.025018",
     "exception": false,
     "start_time": "2021-08-13T07:16:46.110995",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1728x720 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "msno.bar(X_train_filtered)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "fdac87cc",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:47.125828Z",
     "iopub.status.busy": "2021-08-13T07:16:47.124911Z",
     "iopub.status.idle": "2021-08-13T07:16:47.205022Z",
     "shell.execute_reply": "2021-08-13T07:16:47.205459Z"
    },
    "papermill": {
     "duration": 0.132076,
     "end_time": "2021-08-13T07:16:47.205627",
     "exception": false,
     "start_time": "2021-08-13T07:16:47.073551",
     "status": "completed"
    },
    "tags": []
   },
   "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>total_rech_amt_6</th>\n",
       "      <th>total_rech_amt_7</th>\n",
       "      <th>total_rech_amt_8</th>\n",
       "      <th>total_og_mou_6</th>\n",
       "      <th>total_og_mou_7</th>\n",
       "      <th>total_og_mou_8</th>\n",
       "      <th>offnet_mou_6</th>\n",
       "      <th>offnet_mou_7</th>\n",
       "      <th>offnet_mou_8</th>\n",
       "      <th>vol_3g_mb_6</th>\n",
       "      <th>vol_3g_mb_7</th>\n",
       "      <th>vol_3g_mb_8</th>\n",
       "      <th>vol_2g_mb_6</th>\n",
       "      <th>vol_2g_mb_7</th>\n",
       "      <th>vol_2g_mb_8</th>\n",
       "      <th>aon</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.000000</td>\n",
       "      <td>55999.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>328.443079</td>\n",
       "      <td>323.267148</td>\n",
       "      <td>324.287059</td>\n",
       "      <td>306.923158</td>\n",
       "      <td>311.161165</td>\n",
       "      <td>305.737210</td>\n",
       "      <td>191.002213</td>\n",
       "      <td>189.402289</td>\n",
       "      <td>186.110931</td>\n",
       "      <td>121.980578</td>\n",
       "      <td>129.689519</td>\n",
       "      <td>136.102482</td>\n",
       "      <td>51.888564</td>\n",
       "      <td>51.034042</td>\n",
       "      <td>49.858109</td>\n",
       "      <td>1220.13177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>406.974780</td>\n",
       "      <td>415.563194</td>\n",
       "      <td>428.009628</td>\n",
       "      <td>468.546044</td>\n",
       "      <td>481.746916</td>\n",
       "      <td>479.937889</td>\n",
       "      <td>312.965428</td>\n",
       "      <td>317.878777</td>\n",
       "      <td>313.269475</td>\n",
       "      <td>568.335811</td>\n",
       "      <td>564.476438</td>\n",
       "      <td>575.708914</td>\n",
       "      <td>210.934853</td>\n",
       "      <td>206.699198</td>\n",
       "      <td>208.060250</td>\n",
       "      <td>952.09862</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>180.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>110.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>45.190000</td>\n",
       "      <td>43.085000</td>\n",
       "      <td>38.710000</td>\n",
       "      <td>29.110000</td>\n",
       "      <td>26.610000</td>\n",
       "      <td>23.790000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>468.50000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>230.000000</td>\n",
       "      <td>220.000000</td>\n",
       "      <td>225.000000</td>\n",
       "      <td>145.640000</td>\n",
       "      <td>141.830000</td>\n",
       "      <td>138.140000</td>\n",
       "      <td>89.940000</td>\n",
       "      <td>85.560000</td>\n",
       "      <td>82.960000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>868.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>437.000000</td>\n",
       "      <td>430.000000</td>\n",
       "      <td>436.000000</td>\n",
       "      <td>373.985000</td>\n",
       "      <td>379.240000</td>\n",
       "      <td>371.905000</td>\n",
       "      <td>223.580000</td>\n",
       "      <td>218.070000</td>\n",
       "      <td>216.135000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1809.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>35190.000000</td>\n",
       "      <td>40335.000000</td>\n",
       "      <td>45320.000000</td>\n",
       "      <td>10674.030000</td>\n",
       "      <td>8285.640000</td>\n",
       "      <td>11046.740000</td>\n",
       "      <td>8362.360000</td>\n",
       "      <td>7043.980000</td>\n",
       "      <td>7460.210000</td>\n",
       "      <td>45735.400000</td>\n",
       "      <td>28144.120000</td>\n",
       "      <td>30036.060000</td>\n",
       "      <td>10285.900000</td>\n",
       "      <td>7873.550000</td>\n",
       "      <td>8255.670000</td>\n",
       "      <td>4321.00000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       total_rech_amt_6  total_rech_amt_7  total_rech_amt_8  total_og_mou_6  \\\n",
       "count      55999.000000      55999.000000      55999.000000    55999.000000   \n",
       "mean         328.443079        323.267148        324.287059      306.923158   \n",
       "std          406.974780        415.563194        428.009628      468.546044   \n",
       "min            0.000000          0.000000          0.000000        0.000000   \n",
       "25%          110.000000        100.000000         90.000000       45.190000   \n",
       "50%          230.000000        220.000000        225.000000      145.640000   \n",
       "75%          437.000000        430.000000        436.000000      373.985000   \n",
       "max        35190.000000      40335.000000      45320.000000    10674.030000   \n",
       "\n",
       "       total_og_mou_7  total_og_mou_8  offnet_mou_6  offnet_mou_7  \\\n",
       "count    55999.000000    55999.000000  55999.000000  55999.000000   \n",
       "mean       311.161165      305.737210    191.002213    189.402289   \n",
       "std        481.746916      479.937889    312.965428    317.878777   \n",
       "min          0.000000        0.000000      0.000000      0.000000   \n",
       "25%         43.085000       38.710000     29.110000     26.610000   \n",
       "50%        141.830000      138.140000     89.940000     85.560000   \n",
       "75%        379.240000      371.905000    223.580000    218.070000   \n",
       "max       8285.640000    11046.740000   8362.360000   7043.980000   \n",
       "\n",
       "       offnet_mou_8   vol_3g_mb_6   vol_3g_mb_7   vol_3g_mb_8   vol_2g_mb_6  \\\n",
       "count  55999.000000  55999.000000  55999.000000  55999.000000  55999.000000   \n",
       "mean     186.110931    121.980578    129.689519    136.102482     51.888564   \n",
       "std      313.269475    568.335811    564.476438    575.708914    210.934853   \n",
       "min        0.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "25%       23.790000      0.000000      0.000000      0.000000      0.000000   \n",
       "50%       82.960000      0.000000      0.000000      0.000000      0.000000   \n",
       "75%      216.135000      0.000000      0.000000      0.000000      0.000000   \n",
       "max     7460.210000  45735.400000  28144.120000  30036.060000  10285.900000   \n",
       "\n",
       "        vol_2g_mb_7   vol_2g_mb_8          aon  \n",
       "count  55999.000000  55999.000000  55999.00000  \n",
       "mean      51.034042     49.858109   1220.13177  \n",
       "std      206.699198    208.060250    952.09862  \n",
       "min        0.000000      0.000000    180.00000  \n",
       "25%        0.000000      0.000000    468.50000  \n",
       "50%        0.000000      0.000000    868.00000  \n",
       "75%        0.000000      0.000000   1809.00000  \n",
       "max     7873.550000   8255.670000   4321.00000  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_filtered.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9857d0c5",
   "metadata": {
    "papermill": {
     "duration": 0.047951,
     "end_time": "2021-08-13T07:16:47.301731",
     "exception": false,
     "start_time": "2021-08-13T07:16:47.253780",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# 4. Exploratory Data Analysis & Preprocessing\n",
    "\n",
    "Lets start by analysing the univariate distributions of each feature."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "cea7d58e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:47.401613Z",
     "iopub.status.busy": "2021-08-13T07:16:47.400680Z",
     "iopub.status.idle": "2021-08-13T07:16:48.234665Z",
     "shell.execute_reply": "2021-08-13T07:16:48.234180Z"
    },
    "papermill": {
     "duration": 0.885464,
     "end_time": "2021-08-13T07:16:48.234798",
     "exception": false,
     "start_time": "2021-08-13T07:16:47.349334",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3kAAAIMCAYAAABIaGUTAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAABcv0lEQVR4nO3deZxeZXnw8d81GXZkyQBqQDahooIbcQNEHXQA0apd3tpJ64LVumBtXyNibetbW62tUavWUhfyFmum6quiVqQMZlAEqhi6QBQtITbKouBERRYhk7nfP+4z4ZkxM5lMnplzzj2/7+czn8x9npPJdWWe5Vzn3iKlhCRJkiSpDD11ByBJkiRJ6h6LPEmSJEkqiEWeJEmSJBXEIk+SJEmSCmKRJ0mSJEkFsciTJEmSpIL01h3AXB100EHpyCOPrDsMSZIkSarFtdde++OU0sFTj7e2yDvyyCNZt25d3WFIkiRJUi0iYtP2jjtcU5IkSZIKYpEnSZIkSQWxyJMkSZKkgljkSZIkSVJBLPIkSZIkqSAWeZIkSZJUEIs8SZIkSSqIRZ4kSZIkFcQiT5IkSZIKYpEnSZIkSQWxyJMkSZKkgljkSZIkSVJBLPIkSZIkqSAWeZIkacGMjo6ycuVKNm/eXHcoklQsizxJkrRghoaGWL9+PWvWrKk7FEkqlkWeJElaEKOjowwPD5NSYnh42N48SZonFnmSJGlBDA0NMT4+DsD4+Li9eZI0Tyzy1FrO65CkdhkZGWFsbAyAsbExRkZGao5IkspkkafWcl6HJLVLf38/vb29APT29tLf319zRJJUJos8tZLzOiSpfQYHB+npyZcePT09rFixouaIJKlMFnlqJed1SFL79PX1MTAwQEQwMDDA0qVL6w5JkopkkadWcl6HJLXT4OAgxx9/vL14kjSPLPLUSs7rkKR26uvrY9WqVfbiSdI8sshTKzmvQ5IkSdo+izy1kvM6JEmSpO3rrTsAaa4GBwfZtGmTvXiSJElSB4s8tdbEvA5JkiRJD3C4piRJkiQVxCJPkiRJkgpikSdJkiRJBbHIkyRJkqSCWORJkiRJUkEs8iRJkiSpIBZ5kiRJklQQizxJkiRJKohFniRJkiQVxCJPkiRJkgpikSdJkiRJBbHIkyRJkqSCWORJkiRJUkEs8iRJkiSpIBZ5kiRJklQQizxJkiRJKohFniRJkiQVxCJPkiRJkgpikSdJkiRJBbHIkyRJkqSCWORJkiRJUkEs8iRJkiSpIBZ5kiRJklQQizxJkiRJKohFniRJkiQVxCJPkiRJkgpikSdJkiRJBbHIkyRJkqSCWORJkiRJUkEs8iRJkiSpIBZ5kiRJklQQizxJkiRJKohFniRJkiQVxCJvERkdHWXlypVs3ry57lAkSZIkzROLvEVkaGiI9evXs2bNmrpDkSRJkjRPLPIWidHRUYaHh0kpMTw8bG+eJEmSVCiLvEViaGiI8fFxAMbHx+3NkyRJkgo16yIvIpZExH9ExBer9lER8Y2I2BARn4yI3avje1TtDdXjR3b8jDdXx78bEad3HD+jOrYhIs7rYn6qjIyMMDY2BsDY2BgjIyM1RyRJkiRpPuxMT97rgRs62n8NvDeldAzwE+Dl1fGXAz+pjr+3Oo+IeBTwIuDRwBnA31eF4xLgg8CZwKOA367OVRf19/fT29sLQG9vL/39/TVHJEmSJGk+zKrIi4jDgLOAj1btAPqBT1enXAi8oPr++VWb6vHTqvOfD3wipXRfSul7wAbgSdXXhpTSxpTS/cAnqnPVRYODg/T05F93T08PK1asqDkiSZIkSfNhtj15fwucC4xX7T7gpymlsap9M3Bo9f2hwA8Aqsd/Vp2/7fiUvzPd8V8SEa+MiHURse6OO+6YZegC6OvrY2BggIhgYGCApUuX1h2SJEmSpHmwwyIvIp4L3J5SunYB4plRSunDKaXlKaXlBx98cN3htM7g4CDHH3+8vXiSJElSwXpncc7JwK9GxHOAPYH9gPcBB0REb9VbdxhwS3X+LcDDgJsjohfYHxjtOD6h8+9Md1xd1NfXx6pVq+oOQ5IkSdI82mFPXkrpzSmlw1JKR5IXThlJKa0ALgd+ozrtJcDnq++/ULWpHh9JKaXq+Iuq1TePAo4FrgG+CRxbrda5e/VvfKEr2UmSJEnSIjObnrzpvAn4RET8JfAfwAXV8QuAf4qIDcBmctFGSulbEfEp4NvAGPDalNJWgIg4B7gUWAKsTil9axfikiRJkqRFK3InW/ssX748rVu3ru4wJEmSJKkWEXFtSmn51OM7s0+eJEmSJKnhLPIkSZIkqSAWeZIkSZJUEIs8SZIkSSqIRZ4kSZIkFcQiT5IkSZIKYpEnSZIkSQWxyJMkSZKkgljkSZIkSVJBLPIkSZIkqSAWeZIkSZJUEIs8SZIkSSqIRZ4kSZIkFcQiT5IkSZIKYpEnSZIkSQWxyJMkSZKkgljkSZIkSVJBLPIkSZIkqSAWeZIkSZJUEIs8SZIkSSqIRZ4kSZIkFcQiT5IkSZIKYpEnSZIkSQWxyJMkSZKkgljkSZIkSVJBLPIkSZIkqSAWeZIkSZJUEIs8SZIkSSqIRZ4kSZIkFcQiT5IkSZIKYpEnSZIkSQWxyJMkSZKkgljkzWB0dJSVK1eyefPmukORJEmSpFmxyJvB0NAQ69evZ82aNXWHIkmSJEmzYpE3jdHRUYaHh0kpMTw8bG+eJEmSpFawyJvG0NAQ4+PjAIyPj9ubJ0mSJKkVLPKmMTIywtjYGABjY2OMjIzUHJEkSZIk7ZhF3jT6+/vp7e0FoLe3l/7+/pojkiRJkqQds8ibxuDgID09+b+np6eHFStW1ByRJEmSJO2YRd40+vr6GBgYICIYGBhg6dKldYckSZIkSTvUW3cATTY4OMimTZvsxZMkSZLUGhZ5M+jr62PVqlV1hyFJkiRJs+ZwTUmSJEkqiEWeJEmSJBXEIk+SJEmSCmKRJ0mSJEkFsciTJEmSpIJY5EmSJElSQSzyJEmSJKkgFnmSJEmSVBCLPEmSJEkqiEWeJEmSJBXEIk+SJEmSCmKRJ0mSJEkFsciTJEmSpIJY5EmSJElSQSzyJEmSJKkgFnmSJEmSVBCLPEmSJEkqiEWeJEmSJBXEIk+SJEmSCmKRJ0mSJEkFsciTJEmSpIJY5EmSJElSQSzyJEmSJKkgFnmSJEmSVBCLPEmSJEkqiEWeJEmSJBXEIk+SJEmSCmKRJ0mSJEkFsciTJEmSpIJY5EmSJElSQSzyJEmSJKkgFnmSJEmSVBCLPEmSJEkqiEWeJEmSJBXEIk+SJEkAjI6OsnLlSjZv3lx3KJJ2wQ6LvIjYMyKuiYj/iohvRcSfV8ePiohvRMSGiPhkROxeHd+jam+oHj+y42e9uTr+3Yg4veP4GdWxDRFx3jzkKUmSpB0YGhpi/fr1rFmzpu5QJO2C2fTk3Qf0p5QeCzwOOCMingL8NfDelNIxwE+Al1fnvxz4SXX8vdV5RMSjgBcBjwbOAP4+IpZExBLgg8CZwKOA367OlSRJ0gIZHR1leHiYlBLDw8P25kkttsMiL2V3Vc3dqq8E9AOfro5fCLyg+v75VZvq8dMiIqrjn0gp3ZdS+h6wAXhS9bUhpbQxpXQ/8InqXEmSJC2QoaEhxsfHARgfH7c3T2qxWc3Jq3rc/hO4HbgMuAn4aUpprDrlZuDQ6vtDgR8AVI//DOjrPD7l70x3fHtxvDIi1kXEujvuuGM2oUuSJGkWRkZGGBvLl3ZjY2OMjIzUHJGkuZpVkZdS2ppSehxwGLnn7bj5DGqGOD6cUlqeUlp+8MEH1xGCJElSkU466aRJ7ZNPPrmmSCTtqp1aXTOl9FPgcuCpwAER0Vs9dBhwS/X9LcDDAKrH9wdGO49P+TvTHZckSVJNUkp1hyBpjmazuubBEXFA9f1ewLOBG8jF3m9Up70E+Hz1/ReqNtXjIym/S3wBeFG1+uZRwLHANcA3gWOr1Tp3Jy/O8oUu5CZJkqRZuvrqq2dsS2qP2fTkPRS4PCKuIxdkl6WUvgi8CfjfEbGBPOfugur8C4C+6vj/Bs4DSCl9C/gU8G3gX4HXVsNAx4BzgEvJxeOnqnMlSZK0QPr7+1myZAkAS5Ysob+/v+aIJM1VtLUrfvny5WndunV1hyFJklSE0dFRXvrSl3L//fez++67c+GFF7J06dK6w5I0g4i4NqW0fOrxnZqTJ0mSpDL19fVx6qmnAvD0pz/dAk9qMYs8SZIkTdLWkV6SMos8SZIkMTo6yhVXXAHAFVdcwebNm2uOSNJcWeRJkiSJoaEhxsfHARgfH2fNmjU1RyRprizyJEmSxMjICGNjYwCMjY0xMjJSc0SS5soiT5IkSfT399Pb2wtAb2+vWyhILWaRJ0mSNEejo6OsXLmyiPlrg4OD9PTkS8Oenh5WrFhRc0SS5soiT5IkaY6GhoZYv359EfPX+vr6GBgYICIYGBhwCwWpxSzyJEmS5mB0dJTh4WFSSgwPDxfTm3f88cfbiye1nEWeJEnSHJS4GmVfXx+rVq2yF09qOYs8SZKkOXA1SklNZZEnSZI0B/39/SxZsgSAJUuWuBqlpMawyJMkSZqDwcFBUkoApJScxyapMSzyJEmSJKkgFnmSJElzMDQ0NGlfuRIWXpFUBos8SZKkOXDhFUlNZZEnSZI0B/39/fT29gLQ29vrwiuSGsMiT5IkaQ4GBwcnDdd04RVJTWGRJ0mSNAd9fX0MDAwQEQwMDLiBuKTG6K07AEmSpLYaHBxk06ZN9uJJahSLPEmSpDnq6+tj1apVdYchSZM4XFOSJEmSCmKRJ0mSJEkFsciTJEmSpIJY5EmSJElSQSzyJEmSJKkgFnmSJEmSVBCLPEmSJEkqiEWeJElaMKOjo6xcuZLNmzfXHYokFcsiT5IkLZihoSHWr1/PmjVr6g5FkoplkSdJkhbE6Ogow8PDpJQYHh62N0+S5olFniRJWhBDQ0OMj48DMD4+bm+eJM0TizxJkrQgRkZGGBsbA2BsbIyRkZGaI5KkMlnkSZKkBdHf309vby8Avb299Pf31xyRJJXJIk+SJC2IwcFBenrypUdPTw8rVqyoOSJJKpNFniRJWhB9fX0MDAwQEQwMDLB06dK6Q5KkIvXWHYAkSVo8BgcH2bRpk714kjSPLPIkSdKC6evrY9WqVXWHIUlFc7imJEmSJBXEIk+SJEmSCmKRJ0mSJEkFsciTJEmSpIJY5EmSJElSQSzyJEmSJKkgFnmSJEmSVBCLPEmSJEkqiEWeJEmSJBXEIk+SJEmSCmKRJ0mSJEkFsciTJEmSpIJY5EmSJElSQSzyJEmSJKkgFnmSJEmSVBCLPEmSJEkqiEWeJEnSHI2OjrJy5Uo2b95cdyiStI1FniRJ0hwNDQ2xfv161qxZU3cokrSNRZ4kSdIcjI6OMjw8TEqJ4eFhe/MkNYZFniRJ0hwMDQ0xPj4OwPj4uL15khrDIk+SJGkORkZGGBsbA2BsbIyRkZGaI5KkzCJPkiRpDvr7++nt7QWgt7eX/v7+miOSpMwiT5IkaQ4GBwfp6cmXUj09PaxYsaLmiCQps8iTJEmag76+PgYGBogIBgYGWLp0ad0hSRIAvXUHIEmS1FaDg4Ns2rTJXjxJjWKRJ0mSNEd9fX2sWrWq7jAkaRKHa0qSpAUzOjrKypUr3VNOkuaRRZ4kSVowQ0NDrF+/vpg95SxaJTWRRZ4kSVoQo6OjDA8Pk1JieHi4iMKotKJVUhks8iRJ0oIYGhpifHwcgPHx8dYXRiUWrZLKYJEnSZIWxMjICGNjYwCMjY0xMjJSc0S7prSiVVI5LPIkSdKC6O/vp7c3L+zd29tLf39/zRHtmtKKVknlsMiTJEkLYnBwkJ6efOnR09PT+r3lSitaJZXDIk+SJC2Ivr4+BgYGiAgGBgZYunRp3SHtktKKVknlsMiTJEkLZnBwkOOPP76Igqi0olVSOXrrDkCSJC0efX19rFq1qu4wumZwcJBNmzYVUbRKKodFniRJ0hyVVrRKKoPDNSVJkiSpIBZ5kiRJklSQHRZ5EfGwiLg8Ir4dEd+KiNdXx5dGxGURcWP154HV8YiI90fEhoi4LiKe0PGzXlKdf2NEvKTj+IkRcX31d94fETEfyUqSJElS6WbTkzcGvCGl9CjgKcBrI+JRwHnA2pTSscDaqg1wJnBs9fVK4HzIRSHwVuDJwJOAt04UhtU5r+j4e2fsemqSJKlpRkdHWblyJZs3b647FEkq1g6LvJTSbSmlf6++/zlwA3Ao8Hzgwuq0C4EXVN8/H/hYyr4OHBARDwVOBy5LKW1OKf0EuAw4o3psv5TS11NKCfhYx8+SJEkFGRoaYv369axZs6buUCSpWDs1Jy8ijgQeD3wDeHBK6bbqoR8CD66+PxT4Qcdfu7k6NtPxm7dzXJIkFWR0dJTh4WFSSgwPD9ubJ0nzZNZFXkTsC3wG+MOU0p2dj1U9cKnLsW0vhldGxLqIWHfHHXfM9z8nSZK6aGhoiPHxcQDGx8ftzZOkeTKrIi8idiMXeGtSSp+tDv+oGmpJ9eft1fFbgId1/PXDqmMzHT9sO8d/SUrpwyml5Sml5QcffPBsQpckSQ0xMjLC2NgYAGNjY4yMjNQckSSVaTarawZwAXBDSuk9HQ99AZhYIfMlwOc7jr+4WmXzKcDPqmGdlwIDEXFgteDKAHBp9didEfGU6t96ccfPkiRJhejv76e3txeA3t5e+vv7a45Ikso0m568k4HfBfoj4j+rr+cA7wSeHRE3As+q2gBfAjYCG4CPAK8BSCltBv4C+Gb19bbqGNU5H63+zk3AJV3ITZIkNcjg4CATuyRFBCtWrKg5IkkqU++OTkgpXQlMt2/dads5PwGvneZnrQZWb+f4OuD4HcUiSZLaq6+vj2XLlrFp0yaWLVvG0qVL6w5Jkoq0U6trSk3iXkuS1C6jo6PceuutANx2222+f0vSPLHIU2u515IktUvn6ppbt271/VuS5olFnlrJvZYkqX1GRkbYunUrkIs8V9eUpPlhkadWcq8lSWqfk046aca2JKk7LPLUSu61JEntN7HSpiSpuyzy1ErutSRJ7XP11VdPal911VU1RdI9pS0CVlo+0mJlkadWGhwcpKcnP317enrca0mSWqDEG3SlLQJWWj7SYmWRp1bq6+tjYGCAiGBgYMC9liSpBUq7QVfaImCl5SMtZhZ5aq3BwUGOP/741l8kSNJiUdoNutIWASstH2kxs8hTa/X19bFq1arWXyRI0mJS0g260hYBKy0faTGzyJuBk48lSequkm7QlTbHsLR8pMXMIm8GTj6WJEnTKW2OYWn5SIuZRd40nHwsSZJmUtocw9LykRYzi7xpOPlYkiTtSElzDKG8fKTFKlJKdccwJ8uXL0/r1q2bt5//whe+kHvuuWdbe++99+aiiy6at39PkiRJknZGRFybUlo+9bg9edNw8rEkSd3nomaSNP8s8qbh5GNJkrrPRc0kaf5Z5E3DyceSJHWXi5pJ0sKwyJuBk48lSeoeFzWTpIVhkTeDkjZsBedBSJLqNTIywtjYGABjY2OMjIzUHNGu87NVUhNZ5C0izoOQJNWpxEXNVq9ezfXXX88FF1xQdyiStI1F3iLhPAhJUt1KW9RsdHR0W2/kyMiIn62SGsMib5FwHoQkqW6lLWq2evXqSZ+t9uZJagqLvEWixHkQkqT2KWlRs8svv3zGtiTVxSJvkShxHoQkqX1KWtQsImZsS1JdLPIWidLmQUiSVLdnPOMZk9rPfOYz6wlEkqawyFskSpsHIUlS3c4+++xJN1DPPvvsmiOSpMwibxEpaR6EJEl16+vr2zb94bTTTvMGqqTGsMhbREqaByFJaqfSNg8/++yzOeGEE+zFk9QoFnmSJGnBDA0NsX79+mK28vEGqqQmssiTJEkLYnR0lOHhYVJKDA8PF9ObJ0lNY5EnSZIWxNDQ0KTNw0vpzZOkprHIkyRJC2JkZISxsTEAxsbGGBkZqTkiSSqTRZ4kSVoQ/f392zYMj4htK1O2WWkLyUgqg0WeJElaEGeeeSYpJQBSSpx11lk1R7TrSltIRlIZLPIkSdKCuOSSSyb15F188cU1R7RrXEhGUlNZ5EmSpAUxMjIyqSev7XPyXEhGUlNZ5EmSpAXR399Pb28vAL29va2fk1fiQjLOMZTKYJEnSZIWxODgID09+dKjp6eHFStW1BzRrimtaAVYvXo1119/PRdccEHdoUjaBRZ5kiRpQfT19TEwMEBEMDAwwNKlS+sOaZcMDg5OGq7Z9qJ1dHR0W2/kyMiIvXlSi1nkSZKkBXPmmWey1157FbGyZmlWr149qWi1N09qL4s8SZK0YC655BLuvffe1q+sCXnhlc7VQtu+8Mrll18+Y1tSe1jkSZKkBVHalgMjIyNs3boVgK1bt7Z+4ZWJgnW6tqT2sMiTJEkLorQtB0pbeOWpT33qpPZJJ51UUySSdpVFniRJWhClbTlQ2mqhe+yxx6T27rvvXlMkknaVRZ4kSVoQpfV8lbZa6NVXXz1jW1J7WORJkqQFUVrPF+Scjj/++CJymTo80+GaUntZ5EmSpAVRWs8X5JxWrVpVRC5TufCK1F4WeZIkacG4T15zTR2eedVVV9UUiaRdZZEnSZIWzEUXXcQ999zDZz7zmbpD0RRT50i2fc6ktJhZ5EmSpAUxOjrK2rVrAVi7dm3r98mDnNPKlSuLyOXMM8+c1La3VWovizxJkrQgVq9eTUoJgJQSF1xwQc0R7bqhoSHWr1/f+j3/IPeydvrsZz9bUySSdpVFniRJWhBT98Vr+z55o6OjDA8Pk1JieHi49b15X/nKVya1L7/88noCkbTLLPJmUNIQDEmS6jbRizddu22GhoYYHx8HYHx8vPW9eaX9fqTFzCJvBiUNwZAkqW4Te+RN126bkZERxsbGABgbG2t9z+Qzn/nMGduS2qPd767zqLQhGJIk1W1q0dD21Rv7+/vp7e0FoLe3t/X5nH322ZM2q3/5y19ec0SS5soibxqlDcGQJKluZ5999oztthkcHNy2YXhEsGLFipoj2jV9fX2cfPLJAJxyyilFbvAuLRYWedMobQiGJElN0NlT1HZ9fX0sW7YMgGXLlhVVFDkfT2q39r/DzpPShmBIklS3oaGhST1fbR8lMzo6yq233grAbbfd1vqpHaOjo1x55ZUAXHnlla3PR1rMLPKmMTg4OOluY9uHYEiSVLeRkRG2bt0KwNatW1s/SmZoaGhbj1cJUztK3MdQWqws8qbR19fHwMAAEcHAwEBRQzBK4RYXktQuJ5100qT2xPyvtiptasfUffHcJ09qL4u8GZx55pnstddenHXWWXWHou1wiwtJare2z/sqbWrHxIJz07UltYdF3gwuueQS7r33Xi6++OK6Q9EUbnEhSe1z9dVXz9hum9KmdpS2j6G0mPnqnYZFRLO5xYUktc/Unq6293yVNrXDzdClcljkTcMiotlKmwchSYvBmWeeOaldwnSIwcFBjj/++Nb34sEv71voZuhSe1nkTcMiotlKmwchSYvBJZdcMqldwnSIvr4+Vq1a1fpevAmdW1xIai+LvGlYRDRbafMgJGkxWLt27Yxt1WtoaGjSZ6ujmKT2ssibhkVEs5U2D0KSFoNDDjlkxrbqVdo+htJiZpE3DYuI5itpHoTawb0ZpV1z++23z9hWvabuYzi1Lak9LPJmYBHRbKXNg1DzuTejtGtOO+20SXO+TjvttJoj0kyclye1l0XeDCwiJE0YHR3l0ksvdVsVaRcMDg5Omu/uTdRmmbpv4VVXXVVTJJJ2lUWeJM3C0NDQthV3t2zZYm+eNAd9fX2cfvrpRASnn356ETdRN2zYwAtf+EI2btxYdyi7zEXnpHJY5EkN4pyv5lq7di0pJQBSSq4KKM1RaVMh3vGOd3DPPffw9re/ve5QdpmLzknlsMiTGsQ5X83lqoBSd5Q0FWLDhg3ccsstANx8882t781z0TmpHBZ5UkOMjo4yPDzsnK+GclVAqTtKGrHwjne8Y1K7lN68knpapcXKIk9qiKGhIcbHxwEYHx+3N69hXBVQ6o6SRixM9OJNuPnmm2uKpHtK6mmFsm4qSDvDIk9qiJGRkW0Le4yNjbkJbcO4KqC06xyxoIVW0k0FaWdY5EkN4apmzVbiqoDSQutcpXZsbKz1F97HH3/8pPYJJ5xQUyTdU1LPlzcVtJjtsMiLiNURcXtErO84tjQiLouIG6s/D6yOR0S8PyI2RMR1EfGEjr/zkur8GyPiJR3HT4yI66u/8/5w500tUq5q1nzOVZF2zcjIyKRh6W0fsfC9731vUrvtC68ArF69muuvv57Vq1fXHcoucxqEFrPZ9OT9I3DGlGPnAWtTSscCa6s2wJnAsdXXK4HzIReFwFuBJwNPAt46URhW57yi4+9N/bekRcFVzZqvtLkq0kI78cQTZ2y3zd133z1ju21GR0f58pe/DMBll13W+p4vp0FoMdthkZdSugKY+ip/PnBh9f2FwAs6jn8sZV8HDoiIhwKnA5ellDanlH4CXAacUT22X0rp6ylvQPWxjp8lLTr2FEkq2dSerqk9YW2z7777zthum6m9d23vzXMahBazuc7Je3BK6bbq+x8CD66+PxT4Qcd5N1fHZjp+83aOb1dEvDIi1kXEujvuuGOOoUvNZU+RpJKVthrlS1/60kntl7/85fUE0iVTe7rWrl1bUyTd4TQILWa7vPBK1QOXuhDLbP6tD6eUlqeUlh988MEL8U9KkqQuOeKII2Zst81FF100qf3pT3+6pki6Y2L+2nTttnEahBazuRZ5P6qGWlL9ObEr8C3AwzrOO6w6NtPxw7ZzXJIkFebcc8+d1D7vvPOmObMdpvZMTm2rfk6D0GI11yLvC8DECpkvAT7fcfzF1SqbTwF+Vg3rvBQYiIgDqwVXBoBLq8fujIinVKtqvrjjZ0mSpIIcc8wx2+at7bvvvhx99NE1R7Rrpi4I7gLhzeM0CC1Ws9lC4Z+BfwMeERE3R8TLgXcCz46IG4FnVW2ALwEbgQ3AR4DXAKSUNgN/AXyz+npbdYzqnI9Wf+cm4JLupKapStr7RpLUPqOjo/ziF78A4Be/+EXrP48e97jHTWo//vGPryeQLtlzzz1nbEtqj94dnZBS+u1pHjptO+cm4LXT/JzVwC8t05RSWgcc/8t/Q902NDTE+vXrWbNmDa973evqDkeStMgMDQ2RLxUgpdT6z6Pvf//7k9qbNm2qKZLuuO+++2ZsS2qPXV54Re0wOjrK8PAwKSWGh4dbf/dUktQ+IyMjbN26FYCtW7e2ft+y0dHRGdttM1GAT9eW1B4WeYvE0NDQtlWyxsfHWbNmTc0RSZIWm5NOOmnGtiSpOyzyFomRkRHGxsYAGBsba/3dU0lS+/z85z+f1L7rrrtqikSSymaRt0hMvVt68skn1xSJZuLiOJJKds0110xqf+Mb36gpEm3P1D2I3ZNYai+LvEXKcfbN1Lk4jrQQvLGghVTanK+99tprxnbbHHfccZPaj3zkI2uKRNKusshbJK6++uoZ26pfiYvjlFZAlJYPeGNBC+uQQw6Zsd02e++996T2PvvsU1Mk3XHttddOaq9bt66mSLSYlPjZ2gQWeYtEf38/PT35193T00N/f3/NEe260t4USlwc5/zzz+f666/n/PPPrzuUrli9ejXXX389q1f/0m4wrVTijQU125IlS2Zst83U1TR//OMf1xRJdzi1Q3XwZuP8sMhbJAYHBye1V6xYUVMk3VPam0Jpi+OMjo7yta99DYArrrii9QXE6Ojott/J2rVrW58PlHljQc122223zdhum4iYsd12bR9Oq+bzZuP8schTK42OjnLppZeSUuLSSy8t4k2hv7+f3t5eAHp7e1vf2zq1967tvXmrV6+eVBCV0JtX2o0FNd9DHvKQSe2HPvShNUXSHY9+9KMntU844YSaIumOq666asa21G3ebJw/FnmLxNDQ0KThmm1/EQ0NDU26OG17PpB7Wzt/R23vbb3yyisntSd69drqK1/5yqT25ZdfXk8gXVTajQU135YtWya177///poi6Y6bbrppUvvGG2+sKZLueNCDHjRjW81Q0nQVbzbOH4u8RaK0F9HatWu3DSNJKbF27dqaI9p1fX19DAwMEBEMDAywdOnSukPaJaWtoldaPlDejQU139Q5bFPbbXPvvffO2G6b22+/fca2mqGk6SrebJw/FnkzKOlOydQXTdtfRKWt0DbhzDPPZK+99uKss86qO5RdNnWVubavOlfiggSl3ViQpNKVNofNm43zxyJvBiXdKZl6gXrKKafUFEl3lHq38ZJLLuHee+/l4osvrjuUXbZ169YZ220zdVjZfffdV1Mk3TU4OMjxxx/vB2tDlXSzEcrbV67UG45qrtLmsHmzcf5Y5E2jtDslH/rQhya1274IxmmnnbZtFbOI4LTTTqs5ol3XuZhMCc+5Zz3rWTO22+Yb3/jGjO226uvrY9WqVX6wNlRJNxvhl+fkTW23zc9+9rMZ21K3lTb9BrzZOF8s8qZR2p2STZs2zdhum8HBwW1juHfbbbci3hg6F5PZsmVL659zJW7bIS2k0m42Qnk9/LvtttuMbdWvtN7wEuewebNxfljkTaO0OyVHHHHEjO226evr4+lPfzoAp556ahFvDCUuJlOS0uYYqvlKu9kI5S1gdNddd83Ybpu+vr5J7YMOOqimSLqntN5w57BptizyplHaIgvnnnvupPZ5551XUyTdV8rms6XN7Zi6j1zb95UrrQeiVCXdtS/tZmOJvPnTbCX2hpc4h62k9+0mscibpbbfbTzmmGO29d4dccQRHH300TVHtGtGR0e54oorAPjqV79axBtDaYvJTL0gbXvPZGlzDEtV0l37Eodllaa0LRSmbmnx4x//uKZIuqPE3nAobw5bSe/bTWKRN42rr756xnYbnXvuuey9995F9OINDQ1t60nZunVrEW8MpS0mM/HBOl27bZxj2Hyl3bV3WFbzTR1JUsrIklKU2hte0hy20t63m8Qibxr9/f0sWbIEgCVLlhRxB/WYY47hoosuan0vHuQ37s4ir4Q3bouI5uu84C5FScNkSrtrX+KwrNIsX758UvuJT3xiTZFoe+wNb77S3rebpJwrlS7rvIO6ZMmSIi64S7qYO/HEEye1p37Qqn4PfehDZ2y3zdDQ0KSe1lI+iEoaJlPiXfvShmWV5vvf//6kdttXri6NveHNV+L7dlNY5E2jr6+PZcuWAfnitIQ7qKtXr+b6669v/QIYABs2bJjUvvHGG2uKpHuGhoYmra7Z9ovuY445ZlL72GOPrSmS7iix97i0YTKlLZgFZQ3Lgl/uBW97r/htt902Y1v1sje8+extnT/tfnedR6Ojo9xyyy0A3HLLLa2/+BkdHd12Ubp27drW51PiB+vUhUnavlDJN7/5zUnta665pqZIuqPE3uPSh8m0fcEsKGsEBpQ3V1fNZ294s9nbOn8s8qZR2sIeq1evnnQxV0JvXmlK20KhtAUJNm7cOGO7jUobJlPiglklDaeV6lBab3hp7G2dPxZ50yhtY+qvfOUrk9qXX355PYF0SWlDfgB++MMfzthum9KWFp/o2Z9w88031xRJ95Q2TKa0BbNKG04rSdtjb+v8aP+V8TwprVeltCEyUy/e2r7dALDtYnu6tupV4qbHpQ2TGRwcnFTktT2f0ofTShLY2zpfLPKmUdrG1KWtdHj22WfP2G6ju+++e8a26nXPPffM2G6jvr4+Tj31VACe/vSnt/4DtrR8ShtOK0laOBZ505jaM9T2nqKpw3xKGPbTuZy9NN9KHCLcqYRFSjqVkM/UEQttH34qSVo4ZV2ldNHUpbhPOeWUmiLpjqlLibc9n9K2GwDYfffdZ2yrXn19fTO222h0dJSvfvWrAHz1q19t/c2f0vI588wzJ7XPOuusmiKR1CSlrbqr+WGRN40PfvCDk9of+MAHaopkfrT9Lndp2w0A3H///TO2Va/ShnBDvlmyZcsWALZs2dL6myWl5XPRRRdNan/2s5+tKRJJTeKqu5oNi7xpTF1Jb2q7ba688soZ222z//77T2ofcMAB9QQitdiXv/zlGdttU9rNn9JWRZa061x1V7NlkTeN0vb4Km3lxqnbC5SwGbq00Ep7X5i60ErbF16ZOuKi7SMwJO06V93VbFnkTWPqnLWnPe1pNUXSHXfdddeMbUmLT2nvC6XtNTl1bvjUudWSFh9X3dVsWeRN49WvfvWM7bY54ogjZmy3zaGHHjqpfdhhh9UUidReU19HU9tts3Xr1hnbbbPHHntMarsYk6T+/v5toy56e3tddVfTsshbJM4999xJ7fPOO6+mSLrjta997aT2OeecU1MkUnsdffTRk9oPf/jDa4pE21PaXGqpDqWtRDk4ODhpC6kVK1bUHJGayiJvGqtXr56x3TbHHHPMpDs/Uy/u2mbq8IS2L7Ag1eHaa6+d1F63bl1NkWh7SpsbLtWhtJUo+/r6WLZsGQDLli1r/dxjKK8QbwqLvGlMXcWs7WOeN2zYMGkM98aNG2uOaNeU9vuR6nDiiSdOai9fvrymSLqjtA3r77333hnbkmZW4kqUo6Oj3HrrrUBedK6EnEorxJui3Z+A86i0uR3veMc7JrXf/va31xRJd5T2+5HqMPVmz0033VRTJN1R4ob1kuauxJUoh4aGtq20W0JOJRbiTWGRt0hM3efv5ptvrikSSU1R2n6gd9xxx4zttpm68MrUtqSZlbgSZWk5DQ0NbbtRv3Xr1tYXrU1ikSdJi9RDH/rQGduq13333TdjW9LM+vv7Jy1SUsJKlKXlNDIyMqnIa3vRCnmK1Atf+MLap0ZZ5E2jtAnve+2114zttnnEIx4xqX3cccfVFInUXlOLhvvvv7+mSCSp+84888xtQxtTSpx11lk1R7TrSstp6n6gU9ttdM4553DPPff80krwC80ibxoTL6Dp2m1T2gT+TZs2TWr/z//8Tz2BSC02de7D6OhoTZFIUvddcsklk9oXX3xxTZF0T2k5lXazccOGDZPmTNbZm2eRt0hMbJ8wXbttfvGLX8zYliRJi9uXv/zlGdttNHU4Y9uHN1511VWT2m3fD3Tqvs119uZZ5C0SE5N0p2tLWnxK23JAkjotWbJkxnYblTa8cWL10+nabTN15F+d+fiJLkmLVGkfrpLU6e67756xXYK2rxlR2hoYTWKRJ0mSJLXA1OGMX/va12qKpDtKWwOjSSzyJEmSVJwSe4lKW2NB88dnhiRJkorT09OzbQ+2iXbb3XXXXTO2NX/OP//8Oa2W+cY3vnHGx48++mhe/epXzzWsabX/2S5JkiRN0Vngba/dRocccsiMbWmCPXmSJElSC9x5550ztjV/ZtPbdvrpp//SsXe9613zEc4O2ZMnSZIktYD7BDdbX1/fpPbBBx9cUyQWeZIkSZJqUNriOENDQ5PaH//4x2uKxCJPkiRJUg1K3EJholCtsxcPnJMnSZIkSV1xwgknAPXNxZtgT54kSZIkFcQiT5IkSZIKYpEnSZIkSQWxyJMkSZKkgljkSZIkSVJBLPIkSZIkqSAWeZIkSZJUEIs8SZIkSSqIRZ4kSZIkFcQiT5IkSZIK0lt3AJIkSZK00M4//3w2btzY1Z950003AfDGN76xqz/36KOP5tWvfvWsz7fIkyRJkrTobNy4kQ3f/i6H73dI137m7mMBwP03/6RrP/P7d96+03/HIk+SJEnSonT4fofwlpNW1B3GjN5+9Zqd/jvOyZMkSZKkgljkSZIkSVJBLPIkSZIkqSAWeZIkSZJUEIs8SZIkSSrIoltdc1f2w5hpv4ud3buim+aa047276grJ/N5wGJ6zpWWD9SbkyRJmtmtt97K3Xf+fE6rVy6kTXfezj633rtTf8eePEmSJEkqyKLryZvtXfXTTz/9l469613v6nY4XTGbnMynPj7nHmA+kiSpKZYtW8b94z9pxT55uy87cKf+zqIr8lSGBz3oQfz85z/f1t5///1rjEZqltKGCJeWDzhEWJI0v4op8nblImB7HvOYx3DddddNau/ow3W2ZvMhbD4zO+qooyblc8QRRyxoPvPh0ksvndRbdOmlly7ov1/ac67b6v79SJKk7vv+nbd3dU7ej+7+CQAP3mfnet5m8v07b+cYWtqTFxFnAO8DlgAfTSm9c2f+/saNG9nw7Rs4fP+l8xLf/bf8qCs/5/s/2zyr8zZu3MiN317P4fvv1ZV/d6r7brmpKz/n+z+b3STQjRs38t83XMdD9o+u/LtT3Xnr9V35OT/8WZr1ud0uiqbqVkEEsyuKNm7cyA03XMf+3XtPmuTWH16345Nm4Wc/md15pf1+dkZpQ4RLywccIixJTXD00Ud3/Wfef1O+1t/9sO5dUB3DgTsdayOKvIhYAnwQeDZwM/DNiPhCSunbs/0Zt956K7O/PJ+d4/oe3OWfCIkc647M5pyd9Yi+fbr+M2H2+aQu/4KOPKj7BWNKs/+/37hxI9+54ToOOqB7//4+VU2//77w49u6UxT9+KezP3f/A+HUga78s/PmiuHZnbdx40bWf+c69urr3r/du3f+c4/94aY7uvP7uXd09ufOd+EK3Ste6+jh3x7z2T6HdUrS7G8iwvy9x8/X+3EjijzgScCGlNJGgIj4BPB8YNZFHsB9Y2NsmmVP2Wxs2boVgN2WLOnaz7xvbIzZllr3jY3PuqdsNrZsHQdgtyXdW1T1vrHxWedz/1a47afdq/TG8q+H3u79erh/6+zPvfXWW9kytnNF1I5M5LRlrHs/c8vY7Avx0VH4wie7929XLyG6+BJi6xgwPrt8xrfsXBG1I+Njk//sys/cMvsbC1deeSU/Hv0x7Nal1/DuPXD/+KT2dd9Zv+s/d8s4t9566w4/tB7Ip0sfRbv3wv1jk9rXfWenPka2b8vYTuQz2sV8doP7t0xqX/edG3b9584yH0nSzttzzz3rDgFoTpF3KPCDjvbNwJN35geccsops6qub731Vu69d3aF03h1Xs8eu+/w3L322otly5bN6ufOprt1tvnA7HN6IJ/ZDQGdbU515bOlOqe3hnwA9ttvv1nFed999zE+Pr7D84Bt543NYneTnp4e9thjjx2et9vuOdYdmW0+MPucUnXOeBfzocZ8mDhnbHZF1qxy2m12+eyUsXHmNLRhyw7+DwLorWHnnbGtzGkowI7ulkR09y7RzphLTrO5+9PlnM4//3wuu+yyHZ53zz33kOY4XGN7w1InRAR77733Dn/Gs5/97FkVrW3JB2aX02zzgbnnNFM+0L7fUWn5gM+56cz2dzRbbbsx1pQib1Yi4pXAKwEOP/zwSY/N9j9+Z7pab7nlFgAOPfTQHZ5b15wamH1OO5MPdDen0vKBHOdsz2vDc262+Uyc2+3fUWn5QPdzmo+bWTujrptZpeUD9ec0H/NQJEnNEXO9a9DVICKeCvyflNLpVfvNACmlv5ru7yxfvjytW7dugSKUJEm7ant36Nu8Uq35NFtp+UB5OZWWTx0i4tqU0vKpx2sYd7Nd3wSOjYijImJ34EXAF2qOSZIkSZJapxFFXkppDDgHuBS4AfhUSulb9UYlSZK6aeod+rbfsTefZistHygvp9LyaZLGzMlLKX0J+FLdcUiSJElSmzViTt5cOCdPkiRJ0mLW9Dl5kiRJkqQusMiTJEmSpIJY5EmSJElSQSzyJEmSJKkgFnmSJEmSVBCLPEmSJEkqiEWeJEmSJBXEIk+SJEmSCmKRJ0mSJEkFsciTJEmSpIJY5EmSJElSQSzyJEmSJKkgFnmSJEmSVBCLPEmSJEkqSKSU6o5hTiLiDmDTAvxTBwE/XoB/Z6GYT/OVlpP5NF9pOZlP85WWk/k0W2n5QHk5mc/cHZFSOnjqwdYWeQslItallJbXHUe3mE/zlZaT+TRfaTmZT/OVlpP5NFtp+UB5OZlP9zlcU5IkSZIKYpEnSZIkSQWxyNuxD9cdQJeZT/OVlpP5NF9pOZlP85WWk/k0W2n5QHk5mU+XOSdPkiRJkgpiT54kSZIkFcQiT5IkqSEiIuqOQVL7WeQ1QET01h1DN0XEnnXHoF8WEUW93n2eNVdpF6ml5VOK0t4DImJ/gFToPBpfR81V4u+mxJx2VlEXfW0UEccC50TE0rpj6YYqn49HxCPrjqUbImJZRBwZEUfVHcuuqH4vL46IA+qOpRsi4hjgfRFxRN2xdEtEPCQiDo2IQ+uOZa6q10tPKRep1Wv/4JRSKuWCISKWdHzf2muAiDgOeFtEHF13LN1QvUf/S0ScVXcs3VK9fp4REadAGcVrROxbdwzdNHHtWcjv5uER8cKIeCGUkdOuau0bfAki4leALwN3Aj+pOZxdFhGPAD4D/ApwaHVsyYx/qcGqi4gvAn8G/G3Vbp3qebYWWEJ+rrX6Dlf1PPsX4LnAsupYq9/LIuJRwOeBdwFvqS6OWvU7qm7sDAPPrzuWbqguuq8HLoIyLhiq59mqiPhwROydUhpv2/MMoCrsLgFuTCltnPJYG/N5BPApYB/giOpY6/LoVL0ffAb4HeDciDij47FW5lZdA3wwIh5ddyzdUF0bfD0iXll3LLuq+t18GjgFWBkRZ3c8Vvvzra4b7K2+MGqz6qL0JcDfp5RW50NxYFvv4lcXDxcCbwPeA7w/IvZMKW2tN7K5iYgDgSHgb4HXAzcB90fE7nXGtbOq59k55OfZBUCqhgQdXG9kc1N9uH4MeCPwJ8A/RMR+KaXxeiObu+q5tgZ4L/DHwF5Aq/Kpcvgk8N6U0kV1x7OrqgvUC4G3A/8VEc+sjtd+sTBXVe/3GuBbwP7kC3Ba2kt5OvCZlNJHIqInIh4dEU+A9uVTXZwOAX9OLoj+IiKWt/mmQkTsBXwAeF9K6feAa4DxiQvdtv2OAKpRI18CvpZS+taUx1qVC2y7sfBxYAPV9UAb8wCorsveBXw0pfQG4B+AeydG+tT9fIuIPwT+NSL+aqKXcaEUNResTao7qD8C7qsOXQL8EHhCRKxKKV1YX3Rz8mJgdUrp0wAR8ezq2IcjIlr4gbU78J2U0scAImIAOAzYPSIubMuFbPU8+x7w39WhLwI/BZZGxGdSSh+tLbi5+V3gH1NKX4w8H+cU4GnAxdUwwVYVR5UDgFtTSp8AiIjHAquA0YgYbslzbTfg29WNBCLiDcCDgC8AN6WUflZncDsjIg4B3g98KKV0YUScD/wqcHkL38c6DQJXpJQ+GhGfAN4bEYPACPCjekPbaTdT9eIDl5HjPzoivptSeknLfk9PAP4upfQ5gIh4H/C8iLgOGGvpe1oPeXTS16r2C4FnAvdExHdSSm9s2e8I8u/pi9Xrpwd4IrBfSumytuVS3fD5J/LnzNeAayLi6pTS5fVGNmdjwP8A361uJPwR+fl3TkTclFJ6cV2/o4g4GXgc8KfkUW7Pj4i9UkpDC/Hv25NXr58BJ0XEbwNfTym9hNzr8q6IeEq9oe2clNJ5KaUPV3dVlwDfAE6sHmvVGyBASulHwFERcXFE3AB8HVhJ7kX6s6rnsi16gNdFxKvId1RfB/wN8PsRsbzWyHZS9Tw7v2qOAXeQL15p6cUQKaXvAQdGxNUR8W3gP8k9SFcCL6uGDTbdKHBQRJwVEZ8k3xA5kPxhexK06i7xEuAtHTfa3gWcFRHPqTGmblgP7BMRLwIuBx4MnAmsBh4Nrfod/Qw4tRpm9uWU0iDwdOCxncO02iClNJRS+r8dh/4LeBLQ29bhtCmlu8kX3Z+LiGuB7wBnkN8PnhQRz60xvLn6EXBAROxHHpb+KuA9EfGPtUY1N48AVqWUPpVSug14N/CKakRG61Sf/d8DziX3tt6YUjqNfGPhcRHx8oWOKbKTyUX0N1JKlwGfIz93nh0RL1uIOCzyalRdRDyYfMH9H9WxrwL/F2jdG3slVUM0Pw48KyJeU3dAO6u6S0dK6ankIYFfAN6QUvp+Sumz5AK2Nb+flNK7gc3kXC5LKW2u7thdDbR1OG1PSmmMPMTpuIj4vbpjmouO59pJ5JsIa4FXpZT+i/xhdWeN4c1K9bvYSv4AOxW4N6X0Ryml1wPXAq9sU29+Sum2lNI1ABGxWzXn6yPAY6tjbZ1nfD15aNYpwGhK6VdTSr9LHr7559CeG3IppSvIPZDvpbqOSSndR74Jd2+Noc3ZRDGXUvo8eYTP31XtVvxOJnTk8UbgLPLv5D0ppS0ppQ3AN+uMbxfcAexJnmbzbymllwGPBx4dEW+qNbKdlFK6OKX0qaoQCeCr5Jtb+0O75rh3PN/eA/wv4H3k5xwppfvJ16L3L3RcKbsK+HvgrRGxR0rpFvJn/NeAJ1c3DOZVa36RpZm4UEgpnUG+y/WyiDgqIvqBFwBbagxvzqqxzz0ppc3AHwCPiZatHFrdPZ34/fwHsB/w1wARcSL5IqkVQ507LkjPJl/gvTcido+82tlptPQ9YOJ3lFK6lzxv8uE1hzQnU55rV5M/aP+6engZuYel0cvEd/SgXkYeovn4eGCFwK+RL7r3qSO2XZVSmngfvg54SUQcnVo6zzildGNK6Z3AXwLfqoalQl7EaHO0ZL5xx42RPye/VgYjoj8ifh14BXnaQ+tMfHZWzfOA3aphda3SmUdK6WZywfqBiDg48rzJAfKUgVZJKd0IXEGeo39IRDyoutH4AeCuWoObo/SA/yC/T3+gOt6aUTHV822i0PtpdfjdEXFMdT39YvLw7gUTEadHxEsj4viU0jnkobHfjrx+wI/I77nnppTm/SZutOwmUVEiord6kyAi3gPcQ55f9DcppYtrDa4LqqGA7wX+VzUkoFUmeh+qIQyXA7cDhwNvTCn9S73RzV5nL0pEXEDuHXoy8I6U0hdrDa4LIuKpwD8DT0kptfICr+O5diz5zt/uwCHAedWd/VaIiIeTP1SPAu4GnkV+vXyuzri6ISI+QF4U51UT79ttFBEPIfcSXQf8gjzP9S0ppS/UGthOiI75t9Wwp4PJc6YuTCldUmtwXRARB5F7IF6fUvpu3fHsqoj4IPBU8hD7t7fpPQ1+6TP0VeQeozVAIhfk56SUhmsMcc4mXktVr9I/Ah9MKa2tOaxdEhGrgBOAvcnX0wt2vRZ5PvqLgBvIN22/l1L6kyqmlwFHpJQW7KaARd48i7zR+cQQxu09vqTzsYhYmlLa3NThTTvKZzvnL0sp3TrPYc3ZLH4/nW/ujwPuSiltaOrvZzqdz7Mq5/2a/DzbWRHx0DbeSNieiNiNPKl/NKX03Sb+jmaKKSIeRO7RewqwKaV0bRNz6LSDfCYugp4G/CSltH6Bw5uTHeT0VPIctocBn08pDTf1d9RxA2RSfDFloaVqaO2WpuYxYbp8tvP43imle+qIcWftKKfqnMOA+1NKtzf9d7Q9U24snEmeavNk4LMpz7dqtYjYhzz/+AMppRvqjmd7tve8mXKN1vk7OgBy795CPd+qQvljwEtSSj+LiCeSbwh8PaX0mciLKr0/pXTTfMeyLaaWvc5apRr+8hzyHdPHAs9IeZ7K1PNmfOI2xWzzqc6d+oHc2nymFuLVscblA9v29hpLeTGPqY81MuaZzJRP9fiki4s25Lizv6Mm5jRxM6ruOLqltHxg53Jq6HNs4jX94JTSjyJi95TS/VOLu+rcxq+sW1o+MLucOm6SNO45NtUsiojtXtc0Nbed/TyJPG/svu091gQd/98PA34MLEkp3TXl+TZxzoK+hiKv9n0/ea2Dv0spfbw6fi5wZEqplvUpWjkfpy1SnvTZC1xMnjuw3aFx1RNyCeRelsj7yzXuDWO2+VR6oIx8Ukpbm/77ibyq6e7AXwCvj4gjp54z5Xm2JKK5q7bNJp/KxHtYT1M/aCfsxO+otzp/SfVB1aicImIZedW8M2c4pzWLk5SWD8w6p23zipv0HIuIIyLid6rXwnOAL0bEhcDbIuLQNGXFyeom3HhE7BF5mGOj7GI+ffVFPr2dySk9MH86NfV3NFVEPCwi9oqIfScKBpj8Oul8b27S62d7pstnyjk9TS3wIuLhEfG6KvYzgS8DHwL+MfI86fHO31HHa2ivyFMI5ju+c8grlL4e+DBwcjywguwPgT0jYs86rrks8uZJxy/zIuC75HlQP4yIPaac11M9IbdGnvt1AXmoU6OYT7PzAfasitY/AvrI2yMc1XnClDwupFpJq6FKywdmn9NYR07zvvrWHPyMPCTlDZH3w5yk4/eyX0SsWPjwdlpp+cDschpraE4PA94REX9CXslwJXn+0z3kBRUO6uhZmfjdHAB8njwHp2l2JZ+mLlhU1O9oZ4qI6vzOIuLo2gKfRmn5kGuV90XEO8lbcbwCeCt5pdZ/nrixAL/0fPsc81znRF5B/jeBvyJv4fBC8v/330TER6s4351S+kUtNwNSSn51+YsHhsE+nPyGti/wW9Uvvr967Bigp+PvHFA9/sy64zef1uVzDHlVrEdW7YcAnyAv+354daynI4/LJvJs4ldp+ZSSE/CYjtfOPuQFO74CnNpxzpKOHC4Hnlx33Isln5JyIm/F8U3gSxMxkwuLjwDPro71duQxTB5uX3vsiyGf0nICjgXGgXeSl+A/lbx41JvIWyYd2nFu5+vnUuDYuuMvOZ+O+B5BXiXzminH3w+cPc3z7enzHNt+wEeBh5J78f6VvJLmJ8nXlE/r/L+u5f+v7l9gqV/kZYJvJt+NfxN5tbwXV0+8NwE/AZZX5x5A3jvjaXXHbT7tywd4HHmowHuB46pjh5P3xnsbsE917MAqv0bmUWo+bc+p48P0y+SlnyeK0QeRt6/4dzpufnR8wJ660LEuxnxKyYkHitOJ2CeW2n9lxzkfAl49JY9GvjeXlk/BOTW2iFjs+XQ83/ao/vwV8iiF8zrO+Uvgjzva+5NvbC3I8w3Yg7ymw+UTMZM/1/8U2L32/8O6Ayjpq+MJeQD5DsrJ5BXM/hp4B7AbeUnxc4GB6txe8lLJjfmwNZ9m59ORx68Ax5F7Io8lDwt4L3AYuYD4HPCk6twl5Iu/JvZEFpVPKTl15LBPx7FPAf+PBy7yXkoewvyUqr03eWPdxrxeSs2nxJyq9+M/m4gNeCZ5H9m3k1dr/RYdvUHkIVKNy6PUfErLiRYUEYs5nyq+Z5Hnuf161T6WPL1mNXAmeeG80zvO/0PgqQsc47HkPWFPAJ5L7sk7vO7/u5Qs8ubjl/1scgV/MbB/dexkciHxXuCgjnMnXpD71x23+bQrH/KqoDeSh/xdT96g/YnkXqFvkucZnjEljwfVHfdiyaeUnKrXy6er18aK6tgnq2OvI+8F9ISO8x8PPLruuBdLPiXlVL02NpCLhZvIQ00DOI28n9+VwMnVuROvl966414s+RScU+OLiMWaD3n4+b8DfwyMAG8k34x/OHnz9q8Dj5vyfFtSQ5x7kEeAXUa+yfGouv/vtsVWdwAlfZE3+7weeBXwA+AvOh57OrCKaqhWdSzqjtl82pcPeT7X5TxwZ/5V5JVBJ4YBPgY4oel5lJpPKTkBTwI2Ai8Afp/cG/Sn1WNvIN8Rfl7dcS7WfErIqePCbBn5Ztvzq/ZZ5Iu6363apzBl7mBDXzNF5VNqTh3xtaKIWEz5dMR1JLmXuPP59Y9VTntMPF53vB1x70aek1rrHLxfiqvuAEr5Ik9q/QLwB1X7SHJB8ecd5xxYd5zm0/58yMP6/pnJiyn8FXmIVqPesBdjPqXkVF3Eva36fk/g0eReyWOmnNfoC7lS8yklJ3Lvwm3kXqB/44H5qWcA15CHmzY2/tLzKS0nWlpELJZ8Op5Xt5PnGG/qeL49tfoMfXNH3q143tX15RYK3XMgebLlCyPikSml/yGPzX1xRLwdIKX0kxrj21nm0xAT2z1ExMER8ZCUN2a/FXhSRDykOm0YuDVN2bS9iUrLB8rIaTt7+NwLvDQiHpPy8s/fIi9HfXjnSan6pG2a0vKB8nKKiEcDzwf+F3A2cC3w3ojYO6X0r8CfA99pavxTlZYPlJdTSilFxERx+hbgLyNin5TSleRFY54E/G9gU0rp8jr2NtsZpeUTEccBLwJ+lbywzyXkPT8flFL6N/JCUl+aeL615XlXl/D/Z24i8sbL1Z4iPwHuAg4iz4HYC/iHlNJ3I294fHhK6Yr6ot0x82l8Pi8gz+W6FfgS+Q7danJud5LfDN+SUvpCTSHulNLygTJyqvZUez651/sa4Deqr7eTc/i/wO+llP69tiB3Qmn5QBk5VRea+5JzCOBl5Dv2xwKvIb9XvzKldHdtQe6E0vKBMnOCbUXEecA/kF8/f08ezvhrKaWfR8TJwF0ppf+qMcxZKymfiNgb+CDwKODNKaWRiNgNeA9wInku4c/rjLFtLPJ2QUQ8h7xK48XkN74Xkzc5fiVwCPC3KaUbqnOj6XcczKdZ+XQUqr3khWE+C/yQvDLeKvK+a6eTV9Bal1K6sol5TCgtHygrp4h4DHnC/tfIcwqvJw+XOYE83+snwAUppc/WFuROKC0fKC+niHg8D7xHfzil9IuIeCTwB8Dfp5SurzXAnVRaPlBWTqUVEaXlAxARxwIryesm/L/qZvxuwN8Bq1NK36g1wJaxyJujiHgS+S79C8kXcW8m3+U6nVxIvAr4WErp27UFuRPMp5kiYoA8Pv0g4E9SSt+PiIcBVwEfTyn9ca0B7qTS8oF259RRpC4jL3++R0rpnyPidPIqbT8GPpRS+mk1POuephapUF4+UGZOMCmv5cDfkG+QXJBSuncij5pD3Cml5QPF5lRUEVFSPh3Pt+PIcwlvAj7X9Ou0JrPIm6OIeCywhbzT/V+T53f9PXkFqmcDW1JKv6gvwp1jPs3R8Ub3aHKh+jXyZpv/CnwmpfQ/EXE4eRn+pwM3pobO84Ly8oGycqp6vM8Hvk/eLuQx1fFnAb8O3AK8py0XdKXlA2XmBJNeRyeSeyQ+Abw/pTRec2hzUlo+UFZOpRURpeUDv5TTn5H3YHx3atmw4KZw4ZW5W0/ef+h08rC/H5JXndpC3oOokQXEDMynIao3uKeQV5F6W0ppJfmu3KHkhWOOSil9nzyX8DtNLR4mlJYPlJNT9UH6CuAFKaWnAfdFxBcAUkpfBi4iF62tKB5KywfKyakaWjZJ9TqKlNK15PnS/9aW4qG0fKDMnDp15PId4F3A8cBvRMQ+NYc2J23PJyL2m3psSk5/QS5aLfDmyCJvjjou2saBR0XECvLQwN9LKX29vsjmxnwa5z+BfYBzAFJKnyfvu/YI8pv4nuSCtS3+k7LygZbnVH3AvgI4mpwHKaUnAksjYm3VHk7VvNWmKy0faH9OEdtWnX0E8IbIC11N0nFR982mDy0rLR8oM6cJpRURJeTT8Xz7FeDdked8TtKR0w0ppesWPMiCWOTtgpRSAv4J2Ju8Ge27U0rfrTWoXWA+zRARvVVP43HAMRFxAUBK6V/IwwG/mPJy6a24m1paPlBGTimlO4GPkwvTZ0182KaUTgH2jzwPpzVKywfan1N1sXY6eSGIl5O3rDl2O6f2AETEPlUPeSOVlg+Ul1NpRURp+VSxPgd4B3nfu9dEXkNhqonn274R8byFjLEkzsmbxsQLK035D5oYL9z5fUQsAXpSSls6H28S82l2PlNVRcRYROwFfAP4Vkrpt+uOa65KywfKySkinkhegv8u4JKU0rqaQ9olpeUD7c0p8tzoTwG/CTweeAJ5k+OPp5Q2VedMvI4OAP4FeE1q6IqNpeUDxeb0HPKefseRp2l8JKV0zZRzlqSUtkbEvsAzqxt0jVRSPlWP8cXAr5FXCF5O3tfzH1JK/1md0/l8+zzwR6nBW8I0mT152zFRCFQFwpMj4qTIS1dvu2My8f3E30kpbZl6rCnMp9n5bE/1BtebUrqXvKLe8oh4zERubVNaPlBOTimlb5LnFh4E/GpE7Ne2HDqVlg+0J6eI2Lu6MJtwNPDtlNJ1KaULyXuuPQ94WeSFiSZeRweQC423NKl4KC0fKDOnTlUR8X7ynqX/G/gecHZEPK7jnN6qIDqAXHDcUkOos9L2fCJij+pG6ISDgQ3V820Y+CJwGPC6iDgBfun59qcWeHNnkTdFRDyUvHoUEfFk8l2Es4F3RsRvwuRCouPuyX4R8cqmffCaT7PzgZxTROw/9XhHEXEP8IjqTbHxRWpp+UB7c4qIPSOvAEpEHB0RR2zvvKpnaA0wlFK6s0k5dCotHygnp8h7p/0L8ImIeG91+BqgNyJ+FSCltJa84uzDyXfwqS7mPgv8ZUrpioWOezql5QPF5lRUEVFSPtXz7SLgkxHx9urwxPPt1QApDy9dBwR5Sxiqz9pGPt/axiJvipTSbcB+ETFM3vvq11JKvwd8BHh1RPxGdV6acvfk80CjLvDAfJqaT0cR+mTyh+4fRsTuU8+r3rx3SymNR0RPE4tUKC8fKCanI4GzIuID5AJh2lU+U0rXpDx5v8mOpKx8oICcqt6GIeBjwIuB50fE0SmlW4DLgJMi4k0RcTLwJPKepS+qXiunAm9q0sVcaflAsTkVVUSUlE/kOYT/DHyGvL3DYET8YUrpfuADwGMj4gMRcRp56OZ/AU+JPL3mePIQzUbk0mopJb+qL6C34/vzyd3iJ1ftPcmLd3wDeFHHeQcAw8DT6o7ffFqXz5nApcDfAj8EXg/sM+WcJdWfBwLvm/p4k75Ky6eEnKrXxbuAe4F3dhzvmSaHBwGn1x33YsmnhJzIN4tfBpzdcexa4J3Aa8gXbM8FPkm+2fYE4Gnk4ae9dcW9WPIpOKdfIa9w/HLyisbfA/6weuz5wD+Qi4nTgOuq9+5PAkuAk4HH1p1DqflUMb1mIv7q2OnAqur7/YATyAvjra6ebyeRr9Ua8/lZwlftATTliwcWoTkFOK36/h/JK+XtVbX3JN9xeHLV3gtYCzy97vjNpz35kO/A7Q98GXhedexJwNXAnwC7V8d6qz8PqM7trzv2xZBPCTlNvF6q7x8LvJW8et4rOo7vXf3Z05HDWmB53fGXnk9pOdFxYUbeL/KzwGPIBcPfdDy2J3AW8B806KK09HxKy4nCiojS8qlifjjw0I72s8jDgPfaTu4D5AK3kc+3Nn/VHkCTvoDnAN8lr0w0cewT5DHQ+1Ttzg/mRwHH1R23+bQ2n/cDg8AeVftXgTuBFR3nHEhDeyJLz6etOfHADZHTgFcDv06+cPsd4EPAb5FX0fsj4EHVuQeQh2ydWnf8pedTak5VjEuAYzvah5Av7A7rOPZa4JF1x7oY8ykpJworIkrLZzv5HQFcVn3/NOCcjsd+Czim7hhL/Ko9gKZ8AX3kYuGkqr17x2OfId897akjNvNpfz4dF3WHAUdV359DLiKOq9qPJhcLN5NXawzgqzSsJ7LEfErLqbpA+FZ1MbAF+H1gKfDb5HlfPwTOqM7dF7iKZhcPReVTak7byfFJwNeAg+qOxXzKzonCiogC89kXuAB4BnA98IK6Y1oMX71owmbgbmDvaqLxGEBEHJJS+vWIeExq8MbG22E+DZJSShHxXPIciJsjYow87O9RwFuqycZPIM+L+F1yEZsi4jdTSrfXFvg0SssHysipem3sQe59fDG5d+g68ubsm4F/joh/AR6WUrqhOn85ed+r/6or7umUlg+Uk1NUKxdP81iQL0w/CPxxSunHCxrcHJSWD5SZ0wxGge9HxDPIc9f+dOKBlNIna4ppV7Qun+mebxHRQ55LPAj0A69KKV0a0Y59i1ut7iqzri8euGvfBzyk+v5d5Dv3B1XtJ5Inth5Sd7zm0/p8jiDfiX9C1f5r8oqgB5Mnvb8QOIb8Bvgd4OF1x7yY8iktJ+AN5Hk3V0/ECbwCeP52zo2Fist82p0TeWjf/+OB4fFLpjnvCOAdwHObFH/p+ZSaU0fM0+XSAzyUvHjR96gWJmp6TiXksxPPt17g0zRo0ajF8FV7ALUmn1cs+grwJfJdkqeRl6+9gHzn5IbtfeA29ct8mvtFnk/zGTrmCFZveO/paB9HHr51Qt3xLrZ82pwTD9wQeTi5IF1CHvJ3C/Ck6rHHAOupFi1q8ldp+ZSUE7kH8mPkDZcnFoWZ7qJuv4ncaehQ+tLyKTEnCisiCsxnZ55vEzfsG/t8K+1r4oNn0YmIE8lzbZ5Hvnv6OymlEyLiYcAjgcOB9Smlr7ehS9l8mptPNSxmKblnaAT415TS5oh4PvCYlNJfdJzbl1IarSnUWSktH2hnTp3P+4g4g3zz48vAL8g93ueS5w3eCxxF3kPp8zWFu0Ol5QPF5rQHeauQo4AXppTumTpMKyJ6Ut43sjelNFZbsLNQWj5QVk5VLh8hj+r5ze3l0nHuQ1JKP6zezyM1cApHafnAjp9vVfyktG3/4sY+30qzqIq8iQ/cyJtjH0gedrWFvLrZipTSxoj4lZTSf9cZ52yZTzNNV3RGxADwOvIKoT8hr6S3MqV0cZML1dLygfbnFBGHA8uqmxyPIc91+BLwfeAPyHeLzwYeXH3/i5TSt5qUQ6fS8oGycup4b574s4d8E+5XyAsobLuo6/jzAPLF7KuacFOkU2n5QJk5TSitiCghn5Kfb0VJDehOXMgv4NnkDbNfSB4KeBXV0sHAGeRVph5CS7qSzac5X8CRwFOr77c7dp68gMKLgbcBz5jp3Lq/SsunlJyqHG4gDx3dF7iR3Pu4Lw/M5Xg38C9AX93xLrZ8Cs7pdOAvgPeSR1IcUH3/JR4YpjWxb+T+5BWPG7saaGn5lJYTD3RCTPzZQ57DOsyUYYEdfx5AHgrZuNdUafmU9nwr9av2ABY02TzM7zIeWFjhndUL7AxgBXk+xHPrjtN82pkP+U79zcApVXuHhcFszjEfc+qIpZe8f9qV5FU+V5CH+30LeG3HecuqD9sT6455MeVTcE5PJi8A8QLgU+QVGU8jz8e5kDwEdeLC9MCq3Yh9IxdDPgXnVFQRUVI+JT7fSvyqPYAFSzQ/8V4N/IDJ+428mTwP5/8CA9WxxlzUmU+78gFeQl4O/dTpYp14E+9om485zSbu44F/AnYnLxDzM+DF1WNPBjYAr+44f/c64lys+RSc05HAXwJv7Th2LnBR9f1uwPHV973APwDPrDvuxZJPwTkVVUSUlE+Jz7dSv2oPYEGTzauavYZcMLxgymO71R2f+bQzH6YUAMDLyUXE06c+3vEmvj/we3XHvhjyKSEnYC/ySmtnkSfsf4m8otnHeGD5/SeRN9A+p85YF2M+Bed0CHlExWvIc2mO6XjscuDRU87fA1had9yLJZ+CczqSgoqIkvIp8flW8lcPhZqYuNoppXQH8DnynK/TI+LXOx5u3MTWTubTzHw6Jh0/MiIeERH7p5Qmtnj4QEQ8rXo8pkw+/ix5r7VGKS0fKCanRF6R8TeAj5IX7ng1sAl4Z0QcmVK6hnyXeH1dQe6E0vKBMnP6MXk/tecBm4GnR8TyiDiOXMje3XlySum+lDd0b6rS8oHCcoqIQ4BXAbcCh0XEMQAppb8BDoiIR6eUtqSUJl5DS8ibuV9eT8QzKy0fCnu+la64Ii8ijoyIp05ctE19PKV0K/B58qT451QvQFLKtxyaxnyanU+Vx3PIBcFLgK9FxMEppY+QV8+6MCKekbKJ4uFTwP9JKV1ZX+TbV1o+0P6cIi91/gvy8L9fA36eUtqQUrqbvCT/euDvIuKolNLXU0pf2d5rqylKywfKyykilkXEMSkv2X4O+cIuyD0S7yD3NPyflNL/1BbkTigtHygzp0ppRUQR+RT8fCvbrnYFNu2LWS6sQJ70fnjd8ZpPO/OZiJu8gfG3gKOB3wTuBG7jgU0/f59qXD15ONdVNHAidWn5lJJTRw5Hk3uGXlTl8saOc44kDwV6fN3xLrZ8SswJ2Ifcy/154GXAnuShZS+oHj8IeFhn7k3+Ki2fgnNaRjX0j1z8/BPwN+SFSobJq23/Wt1xLsZ8Sny+LZavIvfJi4iXAG8gz3u4YmK41g7+TuP2JppgPs3JJyL2Is8PvDPyxuwJ2Bs4FHhXSml5RKwhr6L16JTSj6q/F+SLwN1SSk0ZAlhcPlBeThHxAuCt5H37ribfPV0B/H3KQ36IiD1T7klqvNLygfJyioj9gMeSt3m4CHgGcAx5deMbagxtTkrLB8rKKSL2Ic/zOpw8ZeOfyTdM/jul9LmIOAjYK6X0g6ZcC8yktHygrOfbYlLMcM3OoS8ppQvJw7D+LiKentLkoYERsaT6c/+I+L3q7zTqRWY+zcunivFxwB9HxEuBvwJIeXP2k8j72QBcTN7w+LiJv5uymxpWPBSVD5SXU0T0kedz/G5KqR+4iVxIvAV4VUS8DaBFxUNR+UCZOaWU7kwpfY08f/CbwHfJm7bvVWdcc1VaPlBWTikPa34LsIo8j/WPyCtPvjsiHplS+nFK6QfVubVfC+xIaflAWc+3xaSIIm/iTki0e2GFbcynmflUb8bXAccC7wG+kFK6OSJ6gFHg6IhYCbyefMH31c7itWlKyweKzGkMeBB5uA/Ax4FHkDfTfhp5X8k2KS0fKDMnIM+RTil9OaX0B+T9S/+97ph2RWn5QDk5lVZElJbPhFKeb4tFEUVeVSC0dmGFqcyneflMFALVHbrvknuCnhsRR6c8Eflz5IL0SOCdKaVvVec38i5daflAmTmllH5GXszj6RFxfEppC3mozMNTSreklL7W8CJ1ktLygTJz6lTdIJnoDd/uyshtUlo+UFZOpRURpeUDZT3fStfqOXkdPUSPIY95fh5wInk1s7vJk9x/GBG/D3y7+rDdi7zJ5JtTSlfUFvx2mE8z8+nI4zjy8sbfJ98gOYc8NPAl5InIpwKfn+iVbGrxUFo+UGZOEyLiUPJwwKcA/wb8LvDalNKXag1sjkrLB8rMSapL5NVqxzvarXivnk5p+ag9WlnkRXkLK5gPzcyno3h4HnnC8XeAA4H3kufe/Do5jwPJG2c3ojCdTmn5QJk5TRURDwKeSp7o/p8ppatrDmmXlJYPlJmTJKm9eusOYGdVhcDjgOdHxHeAZwHnpZT+OyJ+k8kLKzySvLDCj2DbsKybFjzoGZhPM/OJiN1S3qA0RcSRwIuB304pXRsRZwMDwIdSSn8WEdcAP23K0NLtKS0fKDOn6aSUfk5ednu47li6obR8oMycJEnt1bo5eVUhUMzCCubTvHwiohf4XxHxlIg4njwMa3/gMICU0mrg58CbqvYXm1w8lJYPlJmTJElSt7SqJ29iWFZK6e6I+C5wD3lhhXUppY0R8TnyykUPpwULK5hPM/NJKY1FxLeAS6pDzwb+BzguIjZUcX8B+K2J3qSaQp2V0vKBMnOSJEnqltbMySttYQXzaXw++wCXkpc8fgnwPfJmx/sCtwPPBd6YUvp8bUHuhNLygTJzkiRJ6oZWFHmlLaxgPs3OZ0K1gMyJwPnkYuFfI2/Ofjh5GOq6JheqU5WWD5SZkyRJ0q5qdJHXOcyqWljhXeRhfhMLKzyJvLDCf0TEc2n4wgrm0+x8plPF/j5gDXAK8LqJoaZtVFo+UGZOkiRJc9XYIq9aWOG3yD1BdwG/AzwB+ODE8KuIeBfwsJTSi2oLdJbMp90i4qnAy4H/l1K6tO54dlVp+UCZOUmSJM1FYxdeKW1hBfNpdj47klL6t4j4ZpV364f/lZYPlJmTJEnSXDR9C4UbeaCnaD/g88BRwJsj4j3ABcBlLSogzKfFUkpj1Z9FFA+l5QNl5iRJkrSzGjtcc0JpCyuYjyRJkqT51Pgib0JpCyuYjyRJkqT50JoiD8pbWMF8JEmSJHVbq4o8yKs6lrSwgvlIkiRJ6qbWFXmSJEmSpOk1fXVNSZIkSdJOsMiTJEmSpIJY5EmSJElSQSzyJEmSJKkgFnmSJEmSVBCLPEmSJEkqiEWeJEmSJBXk/wPvOPJa3W5GdAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,8))\n",
    "plt.xticks(rotation=45)\n",
    "sns.boxplot(data = X_train_filtered)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "381045b0",
   "metadata": {
    "papermill": {
     "duration": 0.050148,
     "end_time": "2021-08-13T07:16:48.336611",
     "exception": false,
     "start_time": "2021-08-13T07:16:48.286463",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### 4.1 Handling outliers\n",
    "\n",
    "The box plots of these features show there a lot of outliers. These can be capped with k-sigma method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "ff8ffc1c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:48.441999Z",
     "iopub.status.busy": "2021-08-13T07:16:48.441422Z",
     "iopub.status.idle": "2021-08-13T07:16:48.444170Z",
     "shell.execute_reply": "2021-08-13T07:16:48.443754Z"
    },
    "papermill": {
     "duration": 0.058225,
     "end_time": "2021-08-13T07:16:48.444315",
     "exception": false,
     "start_time": "2021-08-13T07:16:48.386090",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def cap_outliers(array, k=3):\n",
    "    upper_limit = array.mean() + k*array.std()\n",
    "    lower_limit = array.mean() - k*array.std()\n",
    "    array[array<lower_limit] = lower_limit\n",
    "    array[array>upper_limit] = upper_limit\n",
    "    return array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "92d1d43d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:48.549317Z",
     "iopub.status.busy": "2021-08-13T07:16:48.548670Z",
     "iopub.status.idle": "2021-08-13T07:16:49.488764Z",
     "shell.execute_reply": "2021-08-13T07:16:49.488189Z"
    },
    "papermill": {
     "duration": 0.995019,
     "end_time": "2021-08-13T07:16:49.488905",
     "exception": false,
     "start_time": "2021-08-13T07:16:48.493886",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "X_train_filtered1 = X_train_filtered.apply(cap_outliers, axis=0)\n",
    "\n",
    "plt.figure(figsize=(15,8))\n",
    "plt.xticks(rotation=45)\n",
    "sns.boxplot(data = X_train_filtered1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "872e86fe",
   "metadata": {
    "papermill": {
     "duration": 0.050821,
     "end_time": "2021-08-13T07:16:49.590800",
     "exception": false,
     "start_time": "2021-08-13T07:16:49.539979",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### 4.2 Feature scaling\n",
    "\n",
    "Lets also scale the features by scaling them with Standard scaler (few other alternates are min-max scaling and Z-scaling)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "4ad7ad33",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:49.700059Z",
     "iopub.status.busy": "2021-08-13T07:16:49.699451Z",
     "iopub.status.idle": "2021-08-13T07:16:49.720008Z",
     "shell.execute_reply": "2021-08-13T07:16:49.719481Z"
    },
    "papermill": {
     "duration": 0.077522,
     "end_time": "2021-08-13T07:16:49.720148",
     "exception": false,
     "start_time": "2021-08-13T07:16:49.642626",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "scale = StandardScaler()\n",
    "X_train_filtered2 = scale.fit_transform(X_train_filtered1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "b4473651",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:49.843716Z",
     "iopub.status.busy": "2021-08-13T07:16:49.837318Z",
     "iopub.status.idle": "2021-08-13T07:16:50.665075Z",
     "shell.execute_reply": "2021-08-13T07:16:50.664547Z"
    },
    "papermill": {
     "duration": 0.893696,
     "end_time": "2021-08-13T07:16:50.665247",
     "exception": false,
     "start_time": "2021-08-13T07:16:49.771551",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,8))\n",
    "plt.xticks(rotation=45)\n",
    "sns.boxplot(data = pd.DataFrame(X_train_filtered2, columns=new_vars))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0fc94753",
   "metadata": {
    "papermill": {
     "duration": 0.053477,
     "end_time": "2021-08-13T07:16:50.772231",
     "exception": false,
     "start_time": "2021-08-13T07:16:50.718754",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "You can perform feature transformations at this stage. \n",
    "\n",
    "1. **Positively skewed:** Common transformations of this data include square root, cube root, and log.\n",
    "2. **Negatively skewed:** Common transformations include square, cube root and logarithmic.\n",
    "\n",
    "Please read the following link to understand how to perform feature scaling and preprocessing : https://scikit-learn.org/stable/modules/preprocessing.html\n",
    " \n",
    "Lets also plot the correlations for each feature for bivariate analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "a4cb0bbc",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:50.885414Z",
     "iopub.status.busy": "2021-08-13T07:16:50.884767Z",
     "iopub.status.idle": "2021-08-13T07:16:51.393619Z",
     "shell.execute_reply": "2021-08-13T07:16:51.394076Z"
    },
    "papermill": {
     "duration": 0.568818,
     "end_time": "2021-08-13T07:16:51.394261",
     "exception": false,
     "start_time": "2021-08-13T07:16:50.825443",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,8))\n",
    "sns.heatmap(pd.DataFrame(X_train_filtered2, columns=new_vars).corr())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "1e5f0109",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:51.532722Z",
     "iopub.status.busy": "2021-08-13T07:16:51.522600Z",
     "iopub.status.idle": "2021-08-13T07:16:51.842950Z",
     "shell.execute_reply": "2021-08-13T07:16:51.842445Z"
    },
    "papermill": {
     "duration": 0.393552,
     "end_time": "2021-08-13T07:16:51.843093",
     "exception": false,
     "start_time": "2021-08-13T07:16:51.449541",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='churn_probability', ylabel='Count'>"
      ]
     },
     "execution_count": 28,
     "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": [
    "#Distribution for the churn probability\n",
    "sns.histplot(y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e59e12c1",
   "metadata": {
    "papermill": {
     "duration": 0.054582,
     "end_time": "2021-08-13T07:16:51.952415",
     "exception": false,
     "start_time": "2021-08-13T07:16:51.897833",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# 5. Feature engineering and selection\n",
    "\n",
    "Let's understand feature importances for raw features as well as components to decide top features for modelling."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "8746dc3f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:16:52.065512Z",
     "iopub.status.busy": "2021-08-13T07:16:52.064870Z",
     "iopub.status.idle": "2021-08-13T07:17:00.041852Z",
     "shell.execute_reply": "2021-08-13T07:17:00.041264Z"
    },
    "papermill": {
     "duration": 8.035233,
     "end_time": "2021-08-13T07:17:00.041999",
     "exception": false,
     "start_time": "2021-08-13T07:16:52.006766",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(n_jobs=-1)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf = RandomForestClassifier(n_estimators=100, n_jobs=-1)\n",
    "rf.fit(X_train_filtered2, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "2a268bb0",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:00.156944Z",
     "iopub.status.busy": "2021-08-13T07:17:00.156185Z",
     "iopub.status.idle": "2021-08-13T07:17:00.262814Z",
     "shell.execute_reply": "2021-08-13T07:17:00.262150Z"
    },
    "papermill": {
     "duration": 0.165618,
     "end_time": "2021-08-13T07:17:00.262949",
     "exception": false,
     "start_time": "2021-08-13T07:17:00.097331",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "feature_importances = pd.DataFrame({'col':new_vars, 'importance':rf.feature_importances_})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "3dc97e26",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:00.394880Z",
     "iopub.status.busy": "2021-08-13T07:17:00.389545Z",
     "iopub.status.idle": "2021-08-13T07:17:00.614633Z",
     "shell.execute_reply": "2021-08-13T07:17:00.614135Z"
    },
    "papermill": {
     "duration": 0.295789,
     "end_time": "2021-08-13T07:17:00.614769",
     "exception": false,
     "start_time": "2021-08-13T07:17:00.318980",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<BarContainer object of 16 artists>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,8))\n",
    "plt.xticks(rotation=45)\n",
    "plt.bar(feature_importances['col'], feature_importances['importance'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "914c5bec",
   "metadata": {
    "papermill": {
     "duration": 0.055776,
     "end_time": "2021-08-13T07:17:00.727093",
     "exception": false,
     "start_time": "2021-08-13T07:17:00.671317",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "At this step, you can create a bunch of features based on business understanding, such as \n",
    "1. \"average % gain of 3g volume from month 6 to 8\" - (growth or decline of 3g usage month over month?)\n",
    "2. \"ratio of total outgoing amount and age of user on network\" - (average daily usage of a user?)\n",
    "3. \"standard deviation of the total amount paid by user for all services\" - (too much variability in charges?)\n",
    "4. etc..\n",
    "\n",
    "Another way of finding good features would be to project them into a lower dimensional space using PCA. PCA creates components which are a linear combination of the features. This then allows you to select components which explain the highest amount of variance.\n",
    "\n",
    "Lets try to project the data onto 2D space and plot. **Note:** you can try TSNE, which is another dimensionality reduction approach as well. Check https://scikit-learn.org/stable/modules/generated/sklearn.manifold.TSNE.html for moree details."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "e2687e7a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:00.843696Z",
     "iopub.status.busy": "2021-08-13T07:17:00.843015Z",
     "iopub.status.idle": "2021-08-13T07:17:05.399680Z",
     "shell.execute_reply": "2021-08-13T07:17:05.400119Z"
    },
    "papermill": {
     "duration": 4.617268,
     "end_time": "2021-08-13T07:17:05.400295",
     "exception": false,
     "start_time": "2021-08-13T07:17:00.783027",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAAD4CAYAAADxeG0DAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOydd3hUdfaH3zu9JpNJ770n1NARBAUUVOy994K6urq21Z9td93Vtffeu4IVRJDeIbRAeu99kun998eEIUOCoqDoOu/z+ODc3LlzM5k593vP+ZzPEbxeL0GCBAkS5I+L6GifQJAgQYIEOTyCgTxIkCBB/uAEA3mQIEGC/MEJBvIgQYIE+YMTDORBggQJ8gdHcjReNCIiwpuSknI0XjpIkCBB/rBs27aty+v1Rh64/agE8pSUFLZu3Xo0XjpIkCBB/rAIglA/3PZgaiVIkCBB/uAEA3mQIEGC/MEJBvIgQYIE+YNzVHLkQYIE+WPgdDppamrCZrMd7VP5U6FQKEhISEAqlR7S/sFAHiRIkIPS1NSEVqslJSUFQRCO9un8KfB6vXR3d9PU1ERqauohPScYyIP8KjhcHva09FHdYSJEKaUgPpQ4nfJon1aQn4nNZgsG8d8YQRAIDw+ns7PzkJ8TDORBfhVWVXRw9Tvb2GeuOTZZx7PnjyE2NBjM/2gEg/hvz899z4PFziAAmGxOSpr7KG3tx+pwH9axuox2/u+LPQx2SN5Wb2BPS/9hnmWQIEGGI7giD0JDj4UHvtzD8rIOAM4YE89ts7OJ/YWpEJvLTbvRPmR7v9V5WOcZJEiQ4QmuyIPwza5WfxAH+Ky4mTVVXb/4eFFaOaePjg/YJhYJZEZpfvExg/x+ufTSS/n000+P9mkcMm+++SYLFiz4Wc/RaIb/7N53330sW7YMgGOPPdbfsT537lwMBgMGg4Hnn3/+8E74EAgG8j85TreH7/a0Ddm+pvLQCy0HIpOIWTAzg3PGJSATi0iJUPHqJUXkxoYczqkG+R/F7T68VN5wuFyuI37M4XjwwQc5/vjjh2z/9ttv0el0wUAe5LdBKhYxJT18yPaiZP1hHTc5XM3D8wtZcdt0Pr9uCjOyo5CIgx+3/wXefvttRowYwciRI7nooosAWL16NZMnTyYtLc2/Ol+5ciUnnXSS/3kLFizgzTffBHx+S3fccQdjxozhk08+ISUlhf/7v/9jzJgxFBYWUlZWdtDXv//++7nooouYNGkSmZmZvPLKK/7XO+aYYzjllFPIy8vDZrNx2WWXUVhYyOjRo1mxYoX/GI2NjRx77LFkZmbywAMP+LefeuqpjB07lvz8fF5++eWA173lllvIz8/nuOOO8ytKDnY3kpKSQldXF3feeSfV1dWMGjWK22+/nYsvvphFixb597vgggv44osvDuVt/1GC36wgnDYmISDtMTZZx7HZQwzWfjZSiYj4MBV6teywj/VLsDrctBismB2/zersz8CePXt4+OGH+eGHH9i5cydPPfUUAK2traxdu5avv/6aO++885COFR4eTnFxMeeeey4AERERFBcXc9111/HYY4/96HN37drFDz/8wIYNG3jwwQdpaWkBoLi4mKeeeoqKigqee+45BEFg9+7dfPDBB1xyySX+xqbNmzfz2WefsWvXLj755BN/SuT1119n27ZtbN26laeffpru7m4AzGYzRUVF7Nmzh+nTpwcE/x/jkUceIT09nR07dvDoo49yxRVX+C9mfX19rF+/nnnz5h3SsX6MYLEzCBlRGt67cgLVnSZEgkBGlIZwjfxon9ZhUdraz6PflbG+uptxyXruODGHgvjQo31af3h++OEHzjrrLCIiIgDQ6313bqeeeioikYi8vDza29sP6VjnnHNOwOPTTz8dgLFjx/L555//6HPnz5+PUqlEqVQyY8YMNm/ejE6nY/z48f4mmrVr13LjjTcCkJOTQ3JyMhUVFQDMmjWL8PBw/+uuXbuWoqIinn76aRYuXAj4Vu2VlZWEh4cjEon853vhhRf6z/XnMn36dK6//no6Ozv57LPPOOOMM5BIDj8MBwN5EACiQhREhSiO9mkcETr6bVzzzlYaeqwArKnqovzNLSy6YUqwKelXQi7ff+H3DuhOJRIJHo/Hv/3ANn+1Wj3sMcRi8U/muA/UWe97fOAxf87zV65cybJly9iwYQMqlYpjjz32oNYEh6Otv/jii3n33Xf58MMPeeONN37xcQYTTK0E+Z+jvsfiD+L76DDaqe+2HKUz+t9h5syZfPLJJ/6UQ09Pz0H3TU5OZu/evdjtdgwGA8uXLz9i5/HFF19gs9no7u5m5cqVjBs3bsg+xxxzDO+99x4AFRUVNDQ0kJ2dDcD3339PT08PVquVRYsWMWXKFPr6+ggLC0OlUlFWVsbGjRv9x/J4PP5c+Pvvv8/UqVMP6Ty1Wi1GozFg26WXXsqTTz4JQF5e3s/+3YcjuCIP8j+HRi5BEAhoSAJQy8VH54T+h8jPz+eee+5h+vTpiMViRo8efdB9ExMTOfvssykoKCA1NfVH9/25jBgxghkzZtDV1cW9995LXFycP22yj+uvv57rrruOwsJCJBIJb775pn/VP378eM444wyampq48MILKSoqorCwkBdffJHc3Fyys7OZOHGi/1hqtZrNmzfz8MMPExUVxUcffXRI5xkeHs6UKVMoKCjgxBNP5NFHHyU6Oprc3FxOPfXUI/Z+CN4DP+2/AUVFRd7ghKAgvxY2p5unllXywqpq/7ZLJiVzxwk5qOTBtcvPobS0lNzc3KN9GgHcf//9aDQabrvttqN9Kr8Ii8VCYWEhxcXFhIYevG4z3HsvCMI2r9dbdOC+h/ypFgThdeAkoMPr9RYMbHsUOBlwANXAZV6v13CoxwwS5NdAIRVzzfQ0JmeE09hjIV6npDAhNBjEgxx1li1bxhVXXMEtt9zyo0H85/JzPtlvAs8Cbw/a9j1wl9frdQmC8G/gLuCOI3Z2QYL8DGxOF8UNBn4o7SBCI2d6diTHZB6+jDLI0eGNN97wyxv3MWXKFJ577rmjdEaHz/HHH099/bBjNw+LQw7kXq93tSAIKQdsWzro4UbgzCN0XkGC/GxWVXRxzTvb/I+fX1XFJ9dMJjtGexTPKsgv5bLLLuOyyy472qfxh+BIqlYuBxYf7IeCIFwtCMJWQRC2/hyf3SBBDoU+q5PHvy8P2NZvdbGt/uCqiiBB/lc4IoFcEIR7ABfw3sH28Xq9L3u93iKv11sUGRm83Q1yZHF7vFiGsd+1OT3D7B0kyP8Whx3IBUG4FF8R9ALv0ZDABAkC6NUyrpmWFrBNIhIYk6w7OicUJMhvyGEFckEQTgD+Bpzi9XqD3RZ/Ejye3+f1em5hLI+eOYL8uBBmZkfx3pUTGBGvO9qnFeQPxJIlS8jOziYjI4NHHnnkaJ/OIfNz5IcfAMcCEYIgNAH/h0+lIge+H2hZ3ej1eq/9Fc4zyAE09Vio77GgVUjIiNT8JtK6TqONleWdfLqtify4EM4cm0he3O/HmlavlnNWUSLzRsQiEQnIJMEGoCCHjtvt5oYbbuD7778nISGBcePG+Z0Uf+/8HNXKecNsfu0InkuQQ2R7Qy+Xv7mFXotv4s6VU1NZMDMDnerXcxn0eLy8s6Gep3+oAmBTbQ8Ltzfz2XWTSYv8fQ2MUMmCevH/dRZtb+bR78ppMViJ0ym5fU42px4wzOTnsnnzZjIyMkhL86Xozj33XL744os/RCAPeq38weizOvm/L0v8QRzg1bW1hz0P0+pw02m0HzRt0myw8tLqmoBtvRYn5W3GYfcPEuTXYtH2Zu76fDfNBitefJ/Nuz7fzaLtzYd13ObmZhITE/2PExISaG4+vGP+VgQD+R+MfquTXU1Dg3Zb3/AubYfC9oZernlnK/OeXsM/vy2lvts8ZB9BANEwjm/BCetBfmse/a4cqzNQoWR1unn0u/KDPON/n2Ag/4MRppIyPmXo9J6EsF9mz1rdYeLCVzexurKLDqOdV9fW8u8lZVgPkPLF65QsmJkRsC0qRE5ubLDZJshvS4vB+rO2Hyrx8fE0Njb6Hzc1NREff3jpmt+KYCD/g6FRSPm/k/NI1PsCt0Qk8Lc52eT/wqJjZYcJ8wFBe3FJG80HfCkEQeDc8Yk8c95oTiiI4ZbjM3nn8vEkhx+a/3OQIEeKg3nKH67X/Lhx46isrKS2thaHw8GHH37IKaeccljH/K0IVoX+gOTHh/LZtZNp7LWiVUhIDVcjlfyya7JSOlTZoZSKkQ1zvHC1nJNHxnHyyLhf9FqHgtPtYU9zH6VtRrRyCSMSQkk64GKxt6WPZaXttPfbmZMfzdhkPeqgIdafhtvnZHPX57sD0itKqZjb52Qf1nElEgnPPvssc+bMwe12c/nll5Ofn3+4p/ubEPz0/0E5UhN9cmO1jE3Wsa3e4N92+5xsEn9hquZwWV/VzWVvbmZfzTU1Qs2bl43zr/zL24yc+/JG+m2+CTLvbWrgufNHM2/Er3dxCfL7Yp865UirVgDmzp3L3LlzD/s4vzXBQP4nJypEwdPnjqa4oZfGHiuFCaGMStAdlSJmn9XBvxbvZbBwprbLzM5Ggz+Qb2/o9QfxfTz+fQVWp5v0SA0jEnSIRcEC7P86p46OPyKB+3+FYCAPQnyYivgw1dE+DexODx1Gx5Dtfdb9Ukune6h3it3lYUN1N3d+tpsPr55I0TDF4N8DFocLu8tD2K+o9w/y5yRY7DyK7G3p49ElZVz37ja+3d2KwTI0iP3e6DbZ6fuVzjNSK+f88YkB2wQB8uP2G/CPTgpDJg782J4yMo4fyjpwebx8vLWR3xsej5eNNd1c/uYW5j+7jpdWVdPWd3gKiyBBBhNckR8lqjqMnPfKJv9q0+XxYnO6sbvcJISpKIwP/VU7NX8u3SY7X+1s4aXVNajkYm6blc2x2ZEoj2AXpSAInD8hGYB3NjYQqZVx14m5FMTvD+T5cSG8f9UEXl9bS0ufjelZkext7fc3SB2Ydvk9sKeljwtf3YRrIGf0r8VlWJ1ubj4uM6jDD3JECAbyo8Teln5/EC9KDkMjl3Drxzv9P79mWhq3zMpEIf19/ImWlbZz/1d7/Y+ve6+Y966cwJSMiCP6OnE6JbfOyubCickopOIhFzNBEChK0TMqUcfOpl4ufHVLgHrhvHGJBx7yqFPaavQH8X28sa6Oc8clEhN6dIrKQf63CKZWjhKDv9fHZEaw8ID24pfX1FDT+fswlLQ4XLy1fuh4qhVlHb/K64lEAjGhyh+9I5GIReTFhvLcBWOYmKZnQqqe1y4pYnzq7y8/rpQNlXiGKCVIxcGvX5Ajw+9jufcnJC8uBK1cgtHuGrJaA/B6wWT/faQJJCKBWJ2Cva2B1gBRIfIh+3q9Xqo7TTT2WNFrZGRGaX41EyulTMLMnCimZoTj9YJ8GE3874HC+FDidApaDPttFO46MZdwzdD3L8jR4/LLL+frr78mKiqKkpKSo306P4tgID9KZEVref+qCXy8tQmtQkJMiIK2/v1f9IQwJcnhR19JAiCTiLlmWhqrKzpxun0XHZ1KyrSsoZOeVld2cfXbW7G7fOqSm47L4Opp6WjkEpxuD1aHmxCl9Iif3++ZlAg1714xgc21PXSa7IxL0TMq8chNUA9yZLj00ktZsGABF1988dE+lZ9NMJAfRQoTdBQm6PB6vUxOj+A/35WxubaHSWnh3D4nm+jDaPhpMVip6TShlIrJiNYSepjBsyhZz2fXTWZnowGFVMyoRB2Z0YE+K239Vm7/ZKc/iAM8vbyK6VlRKKUiXl9Xx/aGXuaNiOWMMQl+bXiX0U5lhxGX20tGlIbYw2y1/j2SFqn53dn9/hjdZjtV7SbkTjcOl2fYTt+jyq6PYfmD0NcEoQlw3H0w4uzDOuS0adOoq6s7Muf3GxMM5EeZjn4bVqebzGgNL1w4lj6LE51KeljpiL0tfVz6xhY6jHYATiyM4f9OyjuswppIJDAiQceIBN1B9zFYnP7XHExTr4WHvt5Ll8knW3x6eRUVbSYeP2ckXUYHN3+4ne2NBgASdApeu3Qc2TGHP7DC4XKzvdHAD6UdaBUSjs2OClDABBmexl4Lt3+8k421PbxySiw1XSZSwtUofi+pq10fw1c3gXNAwtnX6HsMhx3M/6gc8mVWEITXBUHoEAShZNA2vSAI3wuCUDnwb9ivc5r/e9idbr7Z1cK8p9cy47GV3PHpLjr77cTplIcVxO0uN8/+UBUQUBfvbmN7gyFgv+ZeCx9taeTWj3bw8ZZGWgxWzA4XlR1Gmnp/WZE1UiMnPTLQF0UQfGmYfUF8H0v2tNHUa2FNZac/iAM0GWy8v7nhiIyT21jTw7kvb+Sl1TU8trSCs1/awJ6WvsM+7v8666u62Vjb43/scHnoNjv43YzkXf7g/iC+D6fVt/1Pys+5X3oTOOGAbXcCy71ebyawfOBxkENgT0s/N7y/nU6THY8XFu1o4cVV1cN2Lv4cjFYXW+p7h2yv6drvMd5ndXLfF3u447NdfL69mb99tos1lZ3c8G4xsx5fzYlPruGDTQ2YDyi2NvVaWFnewYbqLrqHWXmHa+T89+yRxA+kRlQyMY+dORLlMDlsqVhAKRWzu3loYN1c24PNFejIWN9tZumeNpbtbT+kC43N6eb5FVUMjj0Wh5vVFV0/+dw/O7uaDEO2me0ufjejWvuaft72PwE/Z9TbakEQUg7YPB/fHE+At4CVwB1H4sT+F6jvMrOzqQ+zw0VurJaCuFAkA5Kzyo6hk3UW7WjmpuMyf5Edp9vjxWRzolaImZkTxUdbAjscc2P257NrOk0sHyQdHJEQyoryTlZWdAJgtLu4a+Fu0iPVjE8LB3zpmkte30KnyRfAx6foefyckSQc0No/KjGMT6+bREWbEaPdhcnuRCEVMzEtnI013QAcmx3JvMJYVpZ3MipRx4cHnOvcgtiAu5LS1n4uem2Tf1WfEKbkzcvGkxF18Jyzx+vFOIzqx+L4fSiBDhWrw0Vpq5GGHgtRWjl5cSG/eqPYxDQ9721qCNgWqpT+fjxsQhN86ZThtv9JOdwcebTX620d+P82IPpgOwqCcDVwNUBSUtJhvuzvn9ouExe/tpnGXt8toFgk8NZl45ia6VN6DOe3kaRXoTpAc9xtsiMWCT/65a3uNPHuhnqWl3UwKS2cc8YnUtVuZFuDAZEAV01LY1SSzr+/+4ClVVFy2JDAD1DVaWJ8WjhOl4dX19T6gzjA5roe1ld3E6014fR4yYrWkqT3BfXKdhOXvrnFvxoOVUp567JxVHaY6DY5MFgd3P7pLgDmj4rj7KIEPt3WhMcLs3OjOWVUoJPhp9saA1IzTb1Wvt/b/qOBXCWTcNUxafzlox3+bSIBpg+jtPm94vF4+by4mXsW7ZfCXTY5hb/OzkajOLLlLYfLjVgkQiwSGJ8azgUTknh/sy+Ya+QSdKojqzQ6LI67LzBHDiBV+rYfBueddx4rV66kq6uLhIQEHnjgAa644orDPNnfhiP2afB6vV5BEA568+X1el8GXgYoKir6vdyk/WoU1xv8QRx8wfPR78oZlahDo5BSGB9KUXIYWwfSIBKRwH0n5fkDdpfJzpc7Wnh5dQ1quZi/zs5mxjAt8QaLg79+vJMdA3nmhh4LG2q6eOfyCRisTuRSEakRauSD0hupEWpGJoSys8mX1mg22EiNUFNywNzPSK1P52xxuChuGJqu2VrXw+qKLtr6bURq5bx9+XiSw1U8uawiIKXRZ3Wys6mPSyanUN5mZO7Ta/w/+2JHC6MSdXxw9URCFBKS9OoAb3G328OOxqHplz3DpGQOZEZOJE+fO4pX19YSqpRy7fR0RibqfvJ5w+Fye2jq9c2ITAhT/ibNPPU9Fh76Zm/AtjfW13HKqDhGJx2ZclS32c6Ksg7e29RAkl7FpZNTGJ0Uxn0n5XHRxGTsXY0kh6sQi35HqpV9Bc0jrFr54IMPjsDJHR0ON5C3C4IQ6/V6WwVBiAV+nVa/PyC9wxhLtfXbsDk9aBQQq1Py3AVj2NPSh9HmIiNSQ27sfqXGdyVtPPj1/i/x9e8V8/5VE5icHtgSX99t9gfxfTT0WGnusw7Zdx/hGjlPnDOKhdubWVHeQU6MhgsnJnHlWz7995SMcObkxdBvdbG1rofcWC0nFsbywsrqgOPEh6noMPq0751GOx9vbeSW47MC3Ar3YRrwQLE6XUPuCHY0GnC4POTGDlWUiMUiTh8dz7YD8v5zCmKG/d0GE6qUccqoeGbnxyASfrnevLPfxhvr63h1TS1evFwyKYWrjkkl+ldurzfanNicQ2smBsvQ9/eXsmh7Mw99XQrA9gYDS0raWHj9ZPLiQsmJDaHUIPp9BfF9jDj7T6tQGY7DDeRfApcAjwz8+8Vhn9H/CIXDyNwumphMhHZ/N98+nbgAAUMiKtr6eXfT0Jb4leUdvuDsMIOpA+RapGIZggAHCgrkPxG00iI1/HV2NjfMyPDLyr66cSptfVbWVXVx35d7/PvePTeXs4sSKG/t54fyTiQigfPGJ1He2s+MnCiitQo21HSzvcGAQibiymPSuOvz3f7niwSYmO5rnU/Sq8iJ0VLWtr9GEKGRkRpx8JFxx+VGU91p4u0N9YhFAldPS2NSeviP/n77aO61srK8gx/KOpiYFs6svGhSfuS1hmNtVRfPD7qIvbq2lvRIDedN+HVThPE6JemRaqo79xeqVTLxEWsU6+i3Dbk4210eSpr7yYsLyjT/SBxyIBcE4QN8hc0IQRCagP/DF8A/FgThCqAe+FNcIl1uD80GKwKQEKZCdEARqK3PxuqKTu48MYePtzTSY3Fw4YRkRiSEcvXbW8mM1jC3IJaNNd08u6IKiUjErbOzmFcYg0gk8N7GhmFz6FFaBXRWwNJ7oHIphKWSeuprXDghiXc27i9OzcqLIj1KTU2nid3NfThcHvLiQsiLDRnitjdYG5wVrcXh8vDymtqAff6zpIwZ2ZE8c94YGnotSEQCbQYLVV0WFm1vZmejgRk5UczKjUYmFjMn31cqeW1tLXqVjJuOy/Drz/VqOU+fN5qnl1eyqryT0ck6bp+dM6RoOpiYUAV3zc3lokkpiIAEveqQCm9mh4tHFpfy1S5fGWd5WQdf72rhtUvHEfEz2uMXl7QN2bZoRzPnjk/8Vd0LwzW+9+reRSUUNxhIi1Dxz9NGHLHGIrFIGGIJDD5F0WC8Xm/QpfE35udKPX+OauW8g/zouJ/1in9w2vttvL6uljfW1iEIcN2x6Vw4MTkgMKyr7uK5ldUopWJm50cTopASqpTw10920Wm0s6ayC5lYxBPLKv3P2beCjdHKeXtTPffOy2NLXY+/JT5MJeWYdB18dQE0rPc9qbcW5dsncNMVW5mcPoZdTQZyYkMYn6Kno8/OY0vLyIoJQauQ8OrqGi6enMzopB83leqzOIas7l0eL31WJ5nRWn/6x+Jw8a9vi3EMyCU/2dqEAMzIjkKvlnPe+CROKoxFIhYNMY3Kitby37NG0mNxEKo8tOYnqVj0o6v24ajrMvuD+D52NvVR3WH6WYG8MD6UpXvbA7aNSvptpijlx4Xy1mXj6TI7CFFIjqg/S7hGzl/nZPPXQa6bOpU0oGlKoVDQ3d1NeHh4MJj/Rni9Xrq7u1EoDr2zO9jZeYgYLA6qOkzsau7jpVU1/u1PLqskPVITMJB4zYCMz+p088WOFgBGJoSSE6Ol02hnVJKONZVD9cwryjpIi1STHqnhlTU13HJ8FkabC7FYYEp6ONmSjv1BfB9uB5GGnZxYeAonFsb6N5fsbUMmEaNTSn0eH0Y7W+t6iQ9VERV68A9IUriaUKU0IM8dE6IgQR+YD67rtviD+D4+K25mwcxMv3pF+yO2AHKpmNhf28L1IIuan1tpP6Eghg+3NNA8YHoVFSLntFG/3ZgxrVL6o+/l4TAnLwb9pTK+29NGQpiS43KjA6wXEhISaGpqorOz81d5/SDDo1AoSEg4dDllMJAfAt0mOw9/U0prn3VIoQ7g292tAYF8bHIYiwYC+D5yYkJYX+ML3j0mB+lRQ1eXERo5ayu7uHBiEo99V8F/vitHJhZxwYRE30rYbgGFDmyGwCcqAxUMTreHJSVtZMdoeWxpub9gtr3RgCAIXHlM2kF/10S9itcuKeLuhbupaDcxIiGEh08tJCYkMOiqh1lFhyqlyCS/n1VbSria2fnRLN2zfzWdG6sd0n36U2RGa/nw6kmUtxvxeiE7WkNS+M87xu8VjULCjJwoZuREDftzqVRKamrqb3xWQX4uwUB+COxp6Wfh9maSw1UUJYexpS5QQZEfF+gLMi0rkjFJOooH2uLTI9Ucmx3JRwNjyMrbjZwzLpGV5Z1YHL4ORp+bYAQikUCfxcnLF43B6fYSopSSGaVBo5CCKglO/A8svHrQi58OUXkBr7+72cCy0nYunpQyRPXw4qpqThkV58u3H4SiFD0fXT0Jg9WJXi0lVDk0X58XF0J+bAh7Blnb/n1e7pCAfyTxer2UtvZT3mZEKhFREBf6o4VLtULCvSflMTE1nO/2tDE1I4J5I2KJHPjdHS43lR0mmnutxIQqyIrWHtRPJFGvIlH/+3CjDBLkQIKB/BDoGmiEqe+2cO64RMLVMrrNPnlhbIiCyenhVHb002t2Eh2iIDlczSsXFVHZacLj8ZIepaGstZ+bjstgS20vMaEKBAFumJFBhEaGzekhWivn1o93+gN7tFbOB1dPHFrYypsP4enQXQXqKIgpBPV+BUef1cGXO1pI0asRDZPTlIpFiA8h1xmmlhGmPngTUpxOyYsXjWVHo4EOo52CuBBGJBy60qHH7EApFQ87dOFgbGvo5YJXNvndFWNDFbx7xQTMdheLS9roszqZNyKWsclh/oCcGKbi8qmpXD41cFXp9nhZuL2FOz/f5a8J3H9yHudPSPrd2+IGCXIgwUB+CAyWez21vJJrp6eTG6OlayCYt/fbueT1LejUUk4dFU9Rsp7RSTompu0PsB1GO6+sqaEgLpTtDb0s3N5MQVwI7105gX6bk0te3+IP4gDtRju7mgxDA7lUAQlFvv+GwWRz8eHmRu44MYc+q89JcbDu+LbZ2UesYHawVWq3yU5lhwmn20N6pCbAcqC518Jnxc18vLWRtAgNNx+fwdjkn57qY3e5eWFldYBFbmufjRXlHbywsopus+93fH9zA69fOo6ZB0kV7KO2y8x9X5QEFHYf+qaUienh5BwB58UgQX5LgoH8EMiPC+Hfpxfy4Nd7MTvcRIfIue2TXX4vj3idkr/OzqKt38Yra2p45ocqipJ1PHLGCDKifIWj7GgNj5w+gnsXldBvc5EZpeGRM0YQqpJhd3notw1t8jDb3UO2/RQRGjkzc6J4fkU1l05J5h+nFrC7qY8us4N5hbGMS/l1DSobey3c9vFONg2458XpFLx+6ThyYkJwuj28srqWNzfUAb5W+0213Xy5YMpP2tY6nB66THYiNLKAdv3GHgtSsRjwvX9RWjl1XSZWlUOkVkF6pHrYyUE9ZnvARQF8q/Ru09BGrh/D7fH+fjxIgvxpCQbyQ0AhlXDO+CQmpYfjcHt4bkV1gCFTs8FKn9XJ+5sb/DnprfUGnlxWyX/PHolcIkYmETN/VDxjksLotzmJDVWiH0hdRIUouGxKKo9+V+4/pkQkkP5jplAe7xD9OvjUILfOzuaehbt59LsK1DIxD80v4ObCmCM68f5gbKju9gdxgBaDjXc31vPAKQW09dl4b3Ngo5Pd5aG8zXjQQF7dYaKuy4xSJmZmdhTt/XaS9Co+LW6iqsNEYUIob23wHVOnknLVtDT+852vwCsS4O/zfOmSA3PfcTrlkLsVtUzsd278KarajSzc0cKG6m7mFcYwJz+GhJ+ZQ+802lhb2cXXu1opiA9l3ohYsg4Y1hEkyKEQDOQ/g6RwNRaHi/JBXYn7aOixEK6W02/dH+C/39tOt8kRkFo4WMFsQqqeq45J4/u9bYRr5MwrjOWrHc3kxPhmXu7L29Z0mvh6VyurKzo5oSCGOfnRJOoDC34ZURpevaSIFoMVpUzilwP+GG63B/ER8A8Zzu97Y00PVocLqViEWi4Z0mKukIrptzpxebx0Gm30Wpx8uaOZMUlh3P/VXq6ZnsYrq2voH2jzFwS476Q8QlVS0iM0iEUCbo+X00bH8+LKav/F1OOFB7/ey7iUMAoPGIiREKbixQvG8pePdvi8YgYseA+la7Ktz8pV72yjdsAauLihl+KGXh49a+Qhe8m7PV7eXF/Hcyt8nZXLyzr4aGsDn1wzOeAzUtVh9Dd1FcSHDtvUFSRIMJD/TFQyCWcWJfDgV4FmRkXJehZub/Y/FgmQF6tFIT204FjdaWLh9iYmp0dgsDj41+JSFszI4JkfqtlS18Os3Ghm5UVzy8c7KG31XUi21veyobqbJ88dhVYhpctop9lgJVQpJTlcdUhTdjr6bays6OCTLU3kxYVw9rhE0iI0OFxuQg/BLrXVYGVtVRfrq7soStZzfE40b60PXHWfWBCDWi5Bo5By5wk53DmofT89Uo3H6+XU59YRE6rggonJPPLtXhL0amq7fL7jfVanP4iDz47gm12tvHHpOGq6zTx3/mg+L24mMUzlL0IPpq3fRuEw5z4xPZwvbphCp8lOuEZ2yLr2yg6TP4jv45vdbdw4M5Oc2EPLrzf1WnhldWAHbVufnbI2oz+Ql7f1c94rm+gZ+J3kEhHvXzXhkGoKP0Zzr5WWPgs6lYzUcLXfWjnIH5dgIP8FnFgQQ6vBxlvr65BJRFw8KZlwrZSZOVFsruvhyqlpuDwedEoZZa1GJmf8dHExUiuny+Tgy50+/fnZRYl8s7uVinYTALua+lhX1UVKuNofyMG3kmvoNuP2woL3i2nosaKSiXnglHxOGRn3o5PlHU43G2u6qeuyMDUzAkGALbU9PPDVHjqNDi6ZnMK8whi/XO9AzA4Xjywp8zc9LdzewsRUPf84tYB7vyjB4/XZxp45NgGH20NHv51IjZy/zcmmxWAlXCMnNULFDe9vx+3xUtNlZlt9LwtmZtDQbfF1wEpE2JweMqI0hKmk7Grqw+7y0Gd1sretn3Nf3ojXC6MSdSikwpAh1iKBIemSbpOdkuY+Oox2EvUqCuJCf5Yt7HBqIEHAZ5pziHgB7zCtSYNbs1eUd/qDOPjSUK+urmXk+bpfHHy31PVwzTvb6DE7kIlF3HdyHmeOTfj9jHEL8osIBvJfQGyokjtOyOaiSUm0GKz89eNdrKroZHZ+NKePiedvn+7GNJBDV8vEvHHZOAriQ3/0trsgLpQ5+dF8N9C8Eq9T8PFWk//nqRFqMqI0TMuKZFlpu791H3yB5a8f76Chx2eba3G4uf3TXWTHaH90xubKik7+8tEO/+SX5HAVc/Jj2Fzr08nf/+UeXG7PQRuIajvN/iC+j421PdwyK4vFN0/D5fGQHK6iodvCrR/vYNneDjKjNZwxJoH11V0cnxfD3xftCWiysrs8ON0ettb3MjUzguWl7UzNCKfP4qDL5OD6GRnsbDQwOy+ah77e61ed7Gg0UNVh4olzRnHHZ7voMTtQSEU8fGqhv+AM0G918s9vS/mseP/d0z1zc7l8auohFy0zozRDNPRnFyWS+jOahBJ0Si6fkspLq/d3CUdq5WQPGgDSarAOeV5DrwWnx8svUUh2Gu3c+vEO/8XB4fbw90UljEgI/dHPSZDfP8FAfqi4nOB1+QzsAYlYRJJeTVOPlW6znXOKEghXyzHaXOTEaP0+42aHmw82NyKXNHHVtHTSD2J4FBWi4F+nF3LqqHj2tvYH5Ekvn5KC0e7im92t7Gg0cNeJuby2tpZmg5VTRsYicvRT1m4KOF5sqIIes4OvdrWglUvIiw0JcFjsszh4bGl5wPiu+m4LKpk4wE3x1TW1nDY6fljJoucgxj5uj5c+q4PiegMN4SpeXlPjnxla0txPTWc5j545ArPdjUYu8V/09iEWBGq7zJxYEMPtc3K45aOdWJ0+Bc+Gmm7umZvLpDS93+Y3IUzJ/FHxCALIxALf3jSVZoONMJWUlHB1QFF4S11PQBAHeGxpOTNzow76tzmQqBAFz14wmh/KOtla18PMnCiOyYz80bufA5GIRVw+NZWMKA0LtzczMlHHqaPiSB50MZiZG+Uv5O7joonJKH/h6rnLZKexZ+jFodlgDQbyPzjBQP5TeDzQuAHWPQOmNhh/DWTNAZUvT5kdo+Xh+QU8trTCf0t/2uh4wtQyvh8wWuoy2THaXHy8pZHzxiehkosDOitdbg+1XWZ6LQ4yozUs3dtGbZeZMUlhWJ0uesxOFu3wBZ9ei5OytlIenJ9Ps8HKydlawuoXExuailou4ZSRcbi9XtIj1Fz19lb/yn1yejiPnzOKmIFgbnN5hp1473B5EAsCLq8XiUjgrklKNLVLcHkcSGLyIdrXRep2ewhRSJiaEcHaKp/1gEiA88cnIRELfLC5kW93t3L1tLQhg58tDjdlbUbWVXVx7bFp3P/l/npDTIiC3NgQHjtzBP1WJ3aX2x/E9/Ha2lrmj47j3HGJrCjv5NTR8Ty/ogq7y4NWLuG5C8YwbZhJQI09lmFnhNpdHizDjIX7MVIjNFwxVcMVU395+3p0iIKzihI5qyhx2J8XJet56txRPPpdOTanm2umpXN83kGHcP0k4WoZsaEKWvtsAdt/dc+bIL86wUD+U7Ruh7dOAc/AF33RtXDaSzDyXAAUMjGLdjQH5GUXbm/mttnZ/kA+LkU/0A5u5rjHV6FXy3jglHyOy4nCC3yytZEHv96L0+0lXC3jufPH4MWLweJEEOCmD3YEnJLL48Xu8iCXiLH1thC1/n6ePP5T1hv0PLmsgpNGxPHNrtaA9Mv66m621PYgl4iIDVWSGaXmluMzae+3Y7K7+HpXK70WBzqVFNfAMv1f01WcvOt6RL0DRTmpCi7+kr2SLN7f1MCayi6uPiaNMUk61lZ1cfW0NF5fW8fZL20kLULNnSfmYLS5UEhFfiVJkl7F2UWJqOVipmdHkRmp4b0rJrC+ppsIjYy82BCUUhH3Liqhy+zgumPTh/xJxCIBiSBw+dRURibouPWTnf70jNHu4uYPt/PVjVOHWOO299twuHwXoMHF05wYzY/a6B4t1HIJ80fFMzUjArfHG3BH9UuIClHw37NHcvXb2zDZXYgEuPPEHLKij4wtbpCjRzCQ/xQNG/cHcUGgdvK/WdKYxIatmzg+L5oxSUO9VwAcbjexIQrOHZ9IR7+Vfpub7/b4fK07jXauf6+Yz66bhEQk4t4v9g9x6DY7+NtnO/n8uim4PV4aeszo1bKACwX4VpGPf1+BdkY4oxSh6M01vLiqF4/Xl1ZZMoyHdkW7kWd+qALglYvGUtLcz+fbmwlVSrnymFRGJeqI1spRSMX0WR3Mka3cH8QBnBY8ax7nE9XfeHejLzd+z6ISCuND+PcZhdz44Q6qO3xqjpouM49+V86CGRlcOjmFF1fVoJKJuXRyCq+vq6W1z0ZMiJz0CDU6tYyTCmOwu9w880M13WYH509IYnuDAalYNCT9csvxmeg1cvT45oMeaGTWa3HSabSjlIrRKCT+IRvhGhmfbWvilllZfLqtibI2I+NT9dw+O3uIHUFrnxWjzUVMiIKQX8l58FA5kta1k9Mj+ObGqTQZrOjVMtIihm+YCvLHIhjIfwrJ/tvOzpE3cGNJJiXtvsAdHaKgss3EmKQwNgxMiN9HXkwINx6Xwbqqbk4fE8+NH2wfcuiKdhOhSinpkWoMFqdfOtfQY6WrsZSczX+npOARLp6UzH8GNQulR2qICfF9uf+7vo+TT/sPvV1Wf6fixppuZuRE+gun+8iI1LBgZgY6pZSydiOfbGsCfL4n/1lSzodXT6St3866yi5m5EQi7R46kFnUXYHNG5iP393cT3231R/E92FxuPF4veSozXxzRR49aNhW38vEtHAyozTkxmp54vtKtjcaUMvE3HRcJr0WBzsaDexoNHDL8Zl8uLmBBTMyaDZY6TLZOXlELLlxIXy1s4WmXisj4kP8OvJ9hCqlVHeauOG9YkYm6rhhRgYF8aGkhKu57+Q8bv9kF8dkRTI9K5JRCaEUDvKIcbo9/FDazt0LS+g2OxidqONfpxceVFZotrnotfp81bWK39GA4h8hOUJN8s/0dg/y++aIBHJBEG4BrsSnqtoNXOb1em0//qw/CEkTQREKtj6qtOMpad9fLEoIU/Lk8krumZtLY6+Fpl4rggCXTUnF5HDR2mfjhIIYuox24nVKKjsCA2CCTkGfzUVOTAgRWjkRGhnPrahCr5Khr/wc6tbgSWzhmz1Kbp+TTZ/VSYhSSpRGhsXhRiSAye7iieoELhmjQ7e+HIPFN+h4Rk4UVoebNVVdhKlk3D47m9fX1foHLp82Op7pWZGsqtjvM72z0cATyyq4Zlo6z/xQTf6k0UP016a881i+cWhuXSEVBaRQ9jEuxMCEZWfQXXg1/2iawfpag/9n545L9KdxzA43/1pcxkPzC/yukd/sbmVkoo5HlpTxwCn5bG/sZXNtNy+trvH/HhlRGu46MYd/LynD6failIq5bXYW//q2jG6zg5a+NjbV9vDxNRN9759CyrtXjsdkdxEVoiAtQuMfpNzUa2FPc79/pb64pI3tjQbuXribty4bP8QTvKS5j398u5dNNT2MSQrjvpPzgkXDIEeFww7kgiDEAzcBeV6v1yoIwsfAucCbh3vs3wXReXDpN1D5PSjiGDxf2u314vXCo9+Vc1ZRInq1FJVUQpxOESDru/KYVC6bksJ9X+zxB65pmRHU91j5+6IS//EiNDKum57OeGULUateAyDXtReDOZ9Hvyvn6mlpNPVYeHt9HYlhKu49KY/nV1QTHqLAgIYXLhjDzR/uoMNo592N9fx9Xh4n5McQF6bk0SVl7BmkP1+4vZlbZ2UFBHKFVIxKJqbf5qTZYOWp8jDuP+bfJGx9BBxm7GOv4lthKlMyFAHNT1MywlFIfSvq/yzZf+dwwZhI8vb813cRVBYEBHGAzbXdLJiZyTsb6tk+MEB6cGFTIRXj8Xi54dh0ttb1UNLcz6zcaH8QD1FKCFNJWVXewaNnjsTh9qCWiXltbU2AAVmP2cH66m7uG0hhySUiXrtkXIA51p6WPi59YwudAwXgSenhnDMukY+2NFLcYKC13xYQyNv7bVzzzjaaBySCW+t7ufzNLXxxw1Tiw368eNhisFLW2o/D7SEzSvujVgxBghwKRyq1IgGUgiA4ARXQ8hP7/7GIKYSYQjKMdvJiN7F3ICC63F7/NJ13N/pkYg/Nz+fOz3cHyPpeW1vLSxeO4fY52ZjsLiRiETEh8gBvFYAuk4NEnZxJP1wODt/qPWHzw7w15XE2yiexpcFIqFLGHSdkIxOL6DE7+Pu8XDqMNjKiNLg9Xp44eyQdJjsSkcCHmxs5JisSY5sxIIjvQ6uQcOPMDF5dU0tKhG8yULhaTsvAJJxlNRZ2dmRwxch3SQ6TkpWRw9+eXMtJI2L52wnZdPXb0allPpnjzhayorU8fvZIGnosxGklzG5+Hm3tYgDsnsAGlhtmZNDeb+Px7yvIitZyz7xcHl9agXJQJ+xFE5Np6LawurLD383oHpA8XjQxGYVUzO5mA8nhaqRiAZfH5zIZp1MxMS0Cs8PNW+vrAOg9oLHmroW7WHj9FCI0cuwuN8+tqPIHcfB5xkxI1SMWCehVMkIUEsw2Fw29ZiQiEVan2x/EB//96nvMPxrI67rNXPP2VsoH5KJauYT3rpoQXMkHOSwOO5B7vd5mQRAeAxoAK7DU6/UuPXA/QRCuBq4GSEr6daeP/1pEauU8csYIPitupqLdiCAIPHxqAZ9sbaS+28JFE5NIi1STGKaivH1/4PR6ocPo4F+Ly/zbLp+SguMA9z0At1eg8tRv6LKLkdm6SWlchNrTT3pUCN1WDxEaBa+vq6Oxx8LMnCjOGZfItKxIwtQyttf38u6mBtZWdmFyuPi/k/IQiwREIoGCuBBKWvoDXsvqcPPRlkaeOncUVqcbm9PDySNjCVPJ9hdmTXYeWWfnttlZaPvtXH9sul/mZ5a7kIlFuD0eQpRSei0ONtV0cWJhHFLBS2fMMYTUfYeov4F0ZwUxISNp63cwMyeK4vpef12hqdfKzkYDd8/NQS4RceqoeHJitby5vo49A+d8TJbPllYkCBybFUFrn5Vlpb67o401PWyo6eaSSSk8MMg6YWxyGCeNiKXTaPe3+++jscdKv9VJhEaO2eaiuN4w5G/RZbSjU0r55+kF2F0e7lm4neVlHYhFAhdNTObkEbFDZoK63cNr6/exsbrbH8TBp7J5cVU1T54z6g/ng26yOSlrM9I50CGbFa1FJgm2+x8NDvtdFwQhDJgPpAJxgFoQhAsP3M/r9b7s9XqLvF5vUWTkUI3vH4XoEAW9JjuZkRq+29PKTR9uRyYWcc+8XL7Y2cJ17xaTG6vl+kGyObnEZxY1mO/2tHNWUeBMPqVUjE4t48ZvO7jwgyqe2yNmR+6tnL4ll4ZeG+lRWh74ai97Wvrpt7lYtKOFtzbUIZUIlLX288XOFgwWJxdOTOaW4zKp6TRjd3lYW9nFpZNTSB8ocKllYv5xcianiNfx8YRa4t0tNHSZaey1YLK7qWg3smBmBgqpCKlY4Iwx8TT0WHlnUyMzc6JYVdHBvV/s4bkV1fz7uzLSIjV8tKWR+h4LE9IiuPXjnVz3wU5O+lbGd2NfAG0M8dv+wxszXJw9No7xqfohxeFuswOPF/ptLnY2GXhkcZk/iCeEKekx+1bL66u6uPKYNH8Q30d1p5keS6DPyrb6Xk7Ij+GOE3L8Ovx9jE0OI1LrKxiHKKXMyhvqXz42JYx3rhjPjOwoPi9uYnmZ7zX3GV6NS9UjGdRsdMaYeN7eWEfHIIWR0+2hvK2fdZWdVHea6DQOLR2VthqH1BZ+75jsLp5dUcWZL27guveKOfnZtSwuaf3pJwb5VTgSqZXjgVqv19sJIAjC58Bk4N0jcOzfFLvTjUgQkB5kVeF0e6jpNDE7PwaRIJCkVyKIBMLVMha8v90/jHjRjhZm5UVRlBxGTZeZe+bmIBMLiASfI9+svGgK4kJIDlfz8kVjeXFVNREaOaeOjqe81cic/FhOKPDJCP+zpAyHy4PB6sTs8AwZeLx0TzvlU4385aMdNPb6bvU31/Xw2JkjkEnE/iLg8tIO5o+K469zskkU91Dw3ekIfT5VijckgcaxL3Ht4j40cglXT0tDJHh55rzRbG8wsKK8w+/vMi0rgqpB6hSv15c6Oj4vmnidkoe+KfUrSOwuD7ctN5F05vsoe/ZiVkQxMzeahDBfKsR5wOo1XC3D4/Fw5dRU/r2kjH6bi3idkuuOTae9z8bHV09CIhHweLwB3ac/RrhGRl5cCI+eOYL7v9yD2eEmO0bLQ/Pz/SoTiVjEpZNT2dtqZFeTgQsnJpMVrSFCLScmREFVp4lvdw+Vc+5t6ee+k/LoMNlRSMRsq+9lRXkHtxxvJypEgdPtYWFxM3cv3I3L40UuEfHMeaPJiNJQNajwffKIWLQ/w+vlcGnrs+H2eIgNVQ5rhXwoVLYbeXHQEHKvF/6+sITRiWEkHYKD5G+Jx+Olw2RDKRUPO7bwf4Ej8elpACYKgqDCl1o5Dth6BI77m9FvdbCqopPX19URqpRyzbQ0xqXohxgT7W4y0Gl08NGWBna39DE2Sc+END29FueQALu8tIPXLymitc/OkpI2uk0Onj1/NBVtJoobe3liWSXgW23eNDOTtzbUgRfe2ljvz9VGaeVcNCmZijYjH21pGrY5JlQpxWh3+oPncblRjEzQ4fH4XP/2BUuH28Mn25ooae7jo8zl/iAOIPQ3keWqQCREY7K7eGp5Bf8+YyS1nWaeX1kd8HoHWtCCTxcfppLhdHuHaLrNDjeL2zS8vT6KM8cqKG+rZ1ZeDGcXJfLepgb/fqMTdSwv66DZYGVkQig3zMggUitna10vD39dynXHpvPIklJ/CuukEbF8tXP/CjAnRovsgL9XWoSa9EgNCqmYM8cmMD5Vj8nmIla33wt+H+lRGl6/tIjqDiMPfVPGW+vr0Mgl/PuMQp5fWU12jHaI6ig1Qs2/FpcFFGijtHL0Gt+xqztN/iA+KlHHCfkxfLenjZNHxCITi3h+VTXTMiMxO1zsbOpjVKJuyHtb1WFiSUkr2xsMnFgYy7TMiF/cGNRvdfDFzhYe+64Cu8vN1cekccHEZKJ/wfGGG8BhtLvoG2ZAytGkqdfC2xvq+XBLA3GhSu6em8vk9PD/OcfHI5Ej3yQIwqdAMeACtgMvH+5xf0tWlHdy84c7/I9XV3TyybWThtiF1ndb+OfiMn+gXVHeQWOvhYsnJQ85ZqhSyqqKLvLjQ2jtt1HS3E9rvw2lTMzqii7/fk29PhvYiycms7W+J6Dg1mG0095vQ0AgXCOj22RnZEKoX7UBcONxmfx90R7mj4rnq10tpISrefz7Co7PjQowi9qHRCwgshuGbJe7TURrkzl7XAJhajnrqjqJDVHw19lZPL+i2h+sYkIUQ3Tb80bEsrK8g3mFvgA1+KIWqpT6DaI+3NLIhROT+Me3pdx/Sj73npRLXZfFP+DhnYGC8cQ0PY8tLWfBzEze39zApLRw9GopF0xIwu318tKqWm6bnUVKuJrtDQbSozQkhSkZmxzGTTMzWLq3nSkZEZw7LtEf9BwuD2KRQKxOSehBGnwMFieb63pJj9QwJy+GJL2S9TU97G7uZ15hHFvrev2NWSMTQnF7PFw2JYXWPiuRWgXb63u5+fgs/wDqjn47Lo8XtUzM7LxoHlmyv0aiU0n512mFvLKmhm92tyIWiUiPVFPdYcLt9ZIaocHqcHPZG5v9d1rLyzq4YmoKd56Qe9C7xh9jS10v9y7a33z29A9VRGjlXDwpxbfBaQFBDJKfbkBKClcN+VunRqiICz287tMjidvj5c11dby61tfU1m81ctmbW/j8usmMHOai+UfmiNzPeb3e/wP+70gc67fGbHcFONCBL/2xuqLLH8hrOk209dmwOT0BgRZ8KyatQsLIxFB2Nu4PsPfMy6XDaMdkc3Hl1DQ+29ZIfmzosB2XxQ29pEepqWgfbmCFlRPzo7E6PfxrcSlPnTsaicjXvWh1uBEEAZVMTH6clsL4XD7Z5lvl7mzq4/wJybyxrjZgpNmlk1OwdMQRIHgTRHTHzeTscTLEIoH7v9z/ZQ9TSblmehomuwudUopSJuJfpxfy2ppa2vptvtWlRITF4SY3RsvfTsjmie8rMDvchCglLJiZgVImJjbUZy+bHqnB6fbS1mejocdCXbeZ0hZjQEBo67OjU8mIDZHz3PmjsLu8PPT1XnotTgrjQ7l4UhLrq7pYuKOFjCgN3+xqocvk4NbZWdwyK4trpqejlIr9aYPqDhP//b6cJSVtpISrufKYVFoNVk4ojCU/ztcM1N5vY8H72wO8WC6bnIJmYDj0499XcNGkZDRyCRq5BIfLw2fFjSyYmcl3e9to6mlj/qg4EgYpVmJDFUjFAieNiPU3X+3DYHFS1m701wH6LA4eWVzmv0sZnaTjmmlp/iC+j7fW13PBhOShs1wPgR/KOoZs+2hLI2cW6lHVLoUNz4I8FKbeAsmTQXLwNERGpIYXLxrDHZ/tptNoJytaw3/PGnlEu1APl45+G+9vbgjY5vZ4qWg3BgP5/xoiQUA9zCR31cC2LbU9XPrGZswON8+cN2qY5/suBnedkMO2BgNGm5PMaC2Ld7cGFOTuPCGbR77dy8zcoaZHY5PDWFHawbyRcWyp6w0IvCMSQnltXS0PnpLP0+eORq+W8d+lFWyu841TEwS4/+R8nG4v66t9Mr2JaREopCLuWbibW2dlUdlhwmRzMm9EHN1mO2+bJ3DR+LuILnsbpyqKxqK7WNUup73fysryzoBz67U4iQ5R8P73FX6TrfEpYVw4MRmJyLeydHq8jE4Ko8vsoN/q8zEXiQTsTg/Pr6jiL8dnsbm2m0dOLyBEKeH2OdmEKCXIJSIyItUUxuuobDf6R8Ql6ZWIgNFJYVidbuY/t86fD9/d3IdELDA+OQyT3cWOAf05wPb6XlxuT0Bh2exw8fA3e1lR3kl+XAgJYSqeX1nFWUVJnP/KJhZeP5m0SA3lbcYhhlrvb27gyXNHIQi+1NRrAyu7v87O4rNtTdwzN5dr3yv23518vLUJu8vD/52ci16tIC3SN6mps9/OqkF3Yftwu70D3uZejsmKZMH7+7t/tzcYArzI9zE4cWVz+jpnD3UqUeow3ZyZURpkLZvhsyv2b6xbBZcuhuRJBz2WSCQwMyearxaEYLA6idIqhqSrjjZyqYhIrZz67kDF0s/xnv+j8L/3G/1MrE4Xl05OYUJaOEqpmA3V3RQ3+LywDRYHf1+0G/NAc8n3e9qZPzKOL3bul8lfNCmZEIWEy9/ayn/OHMGra2pQSsVDVBVPLa/iksnJVHWYOWVkHF/tasHr9Q12PiYjgswoDZXtRm6d5Rvi/N7GBk4fE09jj4WqDjONvVZEgs/DZF8QB1+R6dkVVczJj/Gv+mJDFL5VZ5+Nfy0uI1GvJH/AjAovyMLi+UfHbGKTptBhgYSeKD7aUs+c/JghuX7wTejpNPmCeEaUhglp4ahkYmxON51GOx9ubWLvIIXJOeMS+e/SCv/zBeC6Y9Pp6Ldz2Rtb/a+RpFdy9bQ0Xl1bx4iEUP4+L5fqDhP5cSFcOS2Nmz/cwfxRcUOKmtsbDMOmsyZnRGC0u9APkvG199nYVNvDvSflsr3BQFmbkZNGxFEYF0Kf1SefS4vUDBnEDL7g3dhj5Z65uXy/tx2r083lU1LIitaSHxfKjobeITWBr3a2MLcglowoNRKRiDs/243T7eGMMQkBd35SsUCcTsHoJB1/OT6TdqOdG2dm0Gtx8snWRt9AbquTOJ3Cr+sHuGBCEtFaBasrOnh+ZTVmu4urp6UzPSvyJz1hZuZE8uraGtr6fH9LtUzMZZOTkCy5Zf9OqnBfJ3Pl0h8N5OCTZ7b129HKJUPSVU63h/puMzanhyS96qj41ejVcu6dl8dV72z1f4ayojUUxof++BP/gPypA7nJ7uKpZZUBns/XTEvj7/NyyYkNoaHbHKD5XbynjUfPHEl6lAa7y01MiJKSZgNt/Q4sDjf3f7mHe+flYrK7h7yW1elGKhbx5c4WipLDuOOEHCI0MhJ1Kr7Y0cz7W/YXH8elhPGP0wp4eXWNv8DWbLCiVUixOYceu8tkD/gitfbbAvZzu73kxIZw5dvb/Nvmj4qjxCBlc20Pr07IZnFJG9/vbef0MfG8uma/UZZCKkKnlOL1+nLXebEhvLiqGqfbZ5V725xsSgcNWGjqtdJssJKkV2GwOLjymDTWVHVicbj4vrQj4ELR0GOlxeA71zWVXVR1mLhgQhJrKrtYUtJGy4DFwYFEh8hxe7xMTNOzscZ3UZuZHUlDt4WzXtzAg/MLmJQWjkgkoJCKuWZaOs/+UEXvQKG2qsNEY4+VB07OY59oIytaM2QY8+y8GH4oa2dLXS/jU/XcNiuLadk+mWJlR/OwM04jNHI21HTxze4WZuRE+S1jG3st3DAjg5XlHURpFVw4MQmL3eXr1u21csdnu/B4fRfCv52QzcPflOJ0e3jzsvF8saOZbfW9nDIyjpk50exu6ePi17f4X/PGD7bz3PmjmTcibsj5eDxedjUb2FDdg0wi8Ox5Y+i1OLA63OTEhpCll4I8BKQqyqY8wYr+OJrNImaF6Rhrd6E5QDbrdHsQgPJ2Ize8X0xdlwWZWMSdJ+ZwzrhE1HIJvRYHb62r49kVVbg8Xiak6vnn6YWH7Pf+S2jrs1HW1o/T5SEzWkvKwN3HuBQ9z543mj0t/cglIjxeL71mx+/S7fJw+FMH8upO0xDj/lfX1nL6mHjA5zo3MU1PbZeZ1Ag15xQl8v6mejbX9SKXiPyruAsmJJEdo+HUUfGUNPczIjEUpVQcoGbIjwuhptMn29ta34tIJJAcpoJ0+HBroDnVlrpepmRE+IO4XCJCKhZhtruQS8RDio3H5USxrb4n4BgKmQSRABdMSGZCmp47Pt0V8PMvdrTw3PljuGZaGg98uYezxiXxyOIyWg02FszMYG1lF0lhCi7NspPf+B9yT57HJncU//x2f8GuusvMm+vrOCYjgtWV+1MHe1v6WTAjHa1SyuNLKzh/QhKlrf209w/VUBttLpQyMXaXh9Y+G6EqGaFKGbPzo3l7Qz0lzX0BjTcSkcCts7Jo7LFwdlEi10xLo77bworyTn+x9LI3tvDljVPIiQkhSisnUa/0B/F9fFvSyhlj41HJJGyo7iIzSsu7V0zglTU17G7q44SCGMan6LG7PZw/Pom0SE2A3WuoUkpbn5XC+FB/SkYQ4KaZGfxnaTn9Vpc//w7w7e42tHIJJ42IZUJaOFe+vX+VOCM7ill5vulQTb1WFu9u45JJyZw0Mp7UCDW3z8kJOPeXD6jpyMQiOox2XlpVTXu/jelZkYxN0aORS9jW0Mt5L2/0W0OoZGI+unpi4DDqqbdQ1dDIq505hCikbOvo5t2d9Twh0nHaaF+vg9nuYn11F6+vrWV8qp51VV3UDTRZOdweHvx6LyMSQilK0bOjwcCTyyv9h99U28Mb62q5/+T8X0UtUt9t5vp3i/0Tm0KVUt67cgIF8aFUdhi54f3tSESC/z1YVdHFW5ePP2jR+0CsTl/D2OLdrYSpZczOix4yzPto86cO5Cbb0GECbo8Xk91NVbuR1ZWdXD4lldWVnSglYkSCT/Xx4Px83/AFjxe1XExNh4kLJiTzyOIyLA43l0xK5vGzR/LksgoqO0wcmxXJCYWx3DUwdFguEXFCfgzdJjtNvVYOuDsHfOqQJL2KOJ2COfkxPL+ympNGxLK3pY+7Tszh7Q31NPVamJMXw/zRcTyyuIwbZ2bgBZwuD1lRah45YwQvrKxGp5L600ODqWjvZ+nedgoTwli0vZk7T8yhxWBFIgj8c34OkpbNyHtqkNWuYOTuD9g9c8OQY2yu6+G66ekBgXxCip4wtQwRvpSKWBDY0dDLzJwo3j7gwhkbqghYBVsdLv75bRljknTcPTeXRxaXMTs/mhcuGEOXyafPLq7vRSkT+7xYvF7a+m2MTAxlbkEMHUY7BquTqnYTBouT9zc1MCsvCplYxNlFCeg1MsSCwJa6HpxuL5e8vhm7y8OYJB1Pnjuax84cicXp+km9cX5cKK+uqWVEQiiz8qJxuj1kRWt5e0Md/Vbf58ru8gQYiRntLsamhPH3RSUB6aIV5R3cMivL71a5vdHAv88YgUYuZvHAVKjc2BDGp+qJ0ynRKgJrOlcek8qLq6pp7/elTF5fV8fjZ41kfJqe3U19XHdsOjKJiA83N9JssPJtSRvpURp/bt2TOJlKSzclO6to6rUyMyeKGdlR/HdpBdMzI9Fr5Kyv7uaqgTu6sSl6tg7TCdvYY6EoRe9Psw3muz3t/OW4LCK0R74YurGmO2DsXp/Vyatranj0zJE0DRSLXYO+ZDsaffWH4QK51+ulrttMv9XXwxChlbOuqpsr39qvqH51TS2fXDuJgt9RiuZPHciT9CrC1bKAyesp4SokAvz1052cPCKO6wYVs5RSEa9dOo6/fbrL/wFRycS8cvFY1lZ2Y3G4CVNJ0Sik3PzhDm6dnUmSXsWG6h42Vnfz4oVjaDXYsLs9vLCqmvPGJ7K3pZ9xKYGe5vGhClr6rNw4M4PvSlp5cGA25dsb6vnLcZm4PV7uPDEbtUxCi8GKVCxw0aRk/r24HIfbQ5hKSpJeRYvBSm2XmdY+m0/a1rm/kUcuESEIAqWtRialhTMqUcdra2oZlahjbkE0N364k+ouD2pZOvdPe4lTKu4hUjP0g18YH8ropDAum5LCFztaGJWoY3Z+NHcvLPHfUaRHarjzhBxWVnRy7rhEvtjRQqhSytXT0/hikPnW3MIYf7G1uMGAye5i3ojYgVU5rK7swmx3kh2tZURiKA9/vYebjsvig82N3Dgzgzc31FHaaiRUKSUzSsPLq6oJ08gJV8u57+Q8XlxV7f+7nTc+EanI5/lS0W7k612trKvs4rwJSYT+iFpjHx6vl9vnZNHQY/V53USGUt1pYkJaOFMzI/lsWxMLtzfx5Nmj+KS4iaYeK9OyIgcWCkMXEK5BKafRSTr0ailPHpD2m5ETyRNnj+L43GheWl3jv0DIJSJ/EN9HdZeJV9fW+H2BJCLBf2Fs7LFw/5d7uGxKKrmxIWxvNXPThzv8PQdf7mxhVl40mZEaXB4vTreH19buvwto7LEMaWoC/FLPlGGKqqMTdb9a09OB+n7wFcUtTjcxw8ghs6I16IYJ4nanm0U7Wrj/yz1YnW6S9EpevqiIZwc8/PdhdbpZX90VDOS/FxL1Kl6/dBwPf1PKjsZexiaHcdroeNqMNiSCiC11Pbg9XrRyCZdNTUEiElHfbWFOfgxvrq/D7fFicbhZuL0Zvcr35T8uN5pF25sJ18joNDp4ZPF+Y6zv9rTx0KkFvLa8kk6jne31vYhFIsal6EmP1LClrpfcWC0XTEiiscdCa5+Niyan0G12sL2xj3idkiS9ipUVnWyt6+X7Ut8K7s4BG9d9q7xei5NnV1Rx5TG+MWRf7mjhjhOz+by4mT0t/cTrlNwwI51QpYS/z8vBbHdjcbi57+RcZBIRD39TRnWXL+CZHW5uX9ZH1gk3oxB5OGlELF8PpDlClBLOG5fEde9uI0Qp5a4Tc6jqMPL93o6AL1d1p4ktdT1YHC5MNhcvXTSGLXW97G4ycHxeNJPSwhmXqqfDaKem0+y/+FS0m7hmWhpzC2J4bGk5NV0W3B4v66t7WFvVzc3HZ3H/l3uZnh3Joh3N/u7TPquTuxbu5j9njMDicLGz0cCOJgNNvVZSwlWcOy6JTpOdyg4zYpHvDmJGdhRerxev14sg/Hi349a6Hm54v5j2fjshCglvXDaOxSVt/tqCWCTw8PwC7C431Z0mThkRi8Xh5tW1tbg9kYxK1AWobeQS0YB6BSI1cu47KY/2fjtvbwy8e1lR1kllh4lxKXo+vXYyqyo66bM6h6hFBAHkErE/iINvRfrptiZm5kZREB/KI4vLKG7o5eOrJ7G7qW9Il+3y0nZeumgsUSEKXG5PwJzQJSVt3HliDo8vrcA4cFG6bEoK+XE+N8kxyTpm5ESyosx3UdarZdx0XOavNsBiYmp4QF0H4JSRcYQqpeTFarl2epq/CzVEIeEfpxYOGSQCUNZm5I7P9qcgG3qsfLS1Edswd7NO1yG0Ff+G/KkCudvjZWejgeWl7YhEAsflRhOplpEaoWJMko69rf3c+fluwlQyrpueRrfJwdjkME4oiOGZ5ZX+8WBJehXXTk/nuRVVJOlVpEdqyI0Joa7bgsfjRSwSmJUTTXlbPzcfl4kXL7WdZr7e3YrZ7uKBk/PoNvum2OTGallb1UVdt5mzxsaTE63l6R+q2FDt8yKRigWePnf0QC5ZRF2XhZRwFU8sq0QhFXHDjAzkYtEQZUdrn41IjZyLJyUTopRitrm4aGISTQYbXUY7//imFKvTze1zsnliWSV3npDNy6trmJ4dNcRcC6DBFQoODz1mB7fMyiJE4dNS/2txKS6Plx6zgzs+28Ubl47j2RVVQ55f3m7ipuMy+by4iUve2MIF45NICfcpOyak6fl4axNL9rShlkm4eFIyxQ297G7qQyQINBmsjEoM46SR8VS0Gflmd6vvQuH1rY4yojRD9PleL5jtbp75oYrTx8SzqbYHich35/KPb0rxeH3eNnfNzeHZFVW099tRy3za85NGxB3U/Km93+YP4rGhCuYWxtJjcrB10B2V2+PlX0tKefLsUVz+1lb+cWo+4Wo5/zytEIvDzYkFMTyxrJL11d2kRai5aloaNqeLFy8YTW5sKIl6Fbtb+gL+pnq1jPmj4nAOqFkK4kMpiA+lqsNEWVt/QM1GIhIYrvO+2eC7y3thla9bt6rDTHOvmSm6bj463kqPEMZTO6Cs00aYSkZGpK+hTCIWceUxaays6EQlFZOoV/H2ujrevGw8JoeLUKWEjCitvzAaG6rk8bNHUdFuxOpwkxGpIUF/+MVFk91FeZuR1j4rCTol2TFalDIJY1PCuG12Fs/8UIXD7WH+yDh/nStEKeOmmZnMLYylz+okSa8KGHA9mMYey5BtH29p5P5T8rl9UI1JIhKYlBEO+O6kPHiRiY+u4dmfKpAXN/Ry7ssb/amSF1dV8/jZI/l4a2CzRo/ZQZhKRmufjUsnJbN0b3vAjMeGHgtOt4djsyPJitby1PJKbE4P45LDOD43isxoDSqpz9f76R8q8Xp9KYi/HJeJgJdkoZ0wGSTo4B9L6ilKi+auuTnct2gPKpmEKekRjE0OQyYW+QyklpRRlBzGZ8XN5MeFcN54n3vk5VNSeWt9HZdMThnyu0Zp5URq5ejVMiQiEW63m2dX7E8t7KPL5CBcLSNOp2R6dhTxOiXxOuUQi1ZVdAbr6030Whw88X0FC2ZmDLnl9HihtsvMjOxIttT1MiZJx/SsSFweLzkxWirb+v0NL+8O/JsfG0JTr47FA4HYZHfx/Mpqbp+TzVljE3h/UwNb6vcHyTPGxDMiIZSyViMquRiZWITB7CA6RD4kvdBv801d2lZv8Le/f7Or1V+TOHV0HC+vrvE/z+xwc9snO0kOVw3p6t1He7+N9n47k9LCGZ2k450N9by5vo45+dGMTU71a837rS4aeiw8ND+f0lYj3+2pIi1SzQkFMT6/nrxoTh0dx/qqHh74ag9XHZNGWZuJx5dVMjUjwmcpkBI20Gmq5owxCbyypoY31tUxJknHP08vJCcmhKQwJW0GK/8+YwSLtjfTbPDluIdrGJpXGEtdl4ldA53BiWEKsntXIfvianA7QCSmYMo/uWpHJhdPzSA1cn/AG5cSxsdXT2RrfS87G/uYkKYnKkSGzuWbxmSyDwxIGWgIClPJmJAaPux7+EuwO928ua6WxwbJWh8/ewTjUvTIJWKuPzaDk0bE4fJ4SAxTBaz+VXLJIdkED2dVoFPJmJim57nzR/Pm+jrC1TIun5pGQVwom2q6eW1tLQaLk0unJDM146cloL8Wf6pA/t7G+gC1h9Ptxeb0DFGBAChlYqJDFPTZnFS0D83BdfTbOD4nir8Pmre5pb6XmFAFZxXF09Hv4PPi/fnf3c19ZERpmJEexd2L69nQYCY3Wsm9xyfw0PJWxiSFIRWLcHu9PLZ0fzpmbmEMKeFqYgYmne9t7Sc9Uk10iByZRESXycF3e9q4eloar62txT3QEn7PvFyueWeb/wL0l+MzyY/VDgnkvhFrGfznu3Kaeq2kRqi448Qcbv9kp3+Fd+qoeHa02siODiE5XIVIEKFVSIjQyOg6wHOj3+ZCKhZx03EZtPXZeXK570IWoZHxj9MKh7zXZ4xN8K8QB+N0exCJCAji4BuIsWBmBjOyI/lkayM3zMzgrfW1XDs9g/8sKfMXtU4eGetPXxQ39HLP3FxClZIA75hwjXzI++HxQnmbcdhAbnO6sTrc3HxcBol6Fbd9sn+V9u3uNs4fn0RCmJKmXisJYUpa+2yo5WL/RavTZKekuY/Lp6aikIr55zelGO1u7jspl4XbW/znW9FuYlVFFy9cMJqPtzYRq1Pw8Del/hV6cYOBexft4Y1Li9jV1MfFb2zG44Wi5DByYrTE6RSEKHyNV29vqKPH7ODEglhUcjHWQXr5l08KR/bFxb4gDuBxk7jubl44Ywnm0MDRdnaXh1fW1LJ0YKD4kj1tLCttJ0Ij54sdvr6KyenhPHbWSOJ0Pz5Y41Bxujy09VuRikUYrE7++/3+IH7hxGQ2VPdw1+claBUS7p6by5z8mCEuoz+HnJjANIxCKuLfZxSSqFeTqFczKz8aEQISsYji+l7Of3WT/7O8ua6Hp84dxfxR8Yf3S/9C/lSB3DiMSmVvSx8XTUzmzYEBBODTWL+9oZ5NtT0UJYcxOz86wF8cYEZOFN3GoSPPfijrIC1SPexrbanrIVorZUODr+hY2m7lmk9refr0dGr6bdw4M4Mb3t+OIPhaoE12F9/ubuPB+fl8M5CXPiYjgpdW13DVMWn+oFXS3I/D5eHGmRl4vF7Gpej568c7A+4inv2hipcuGst3e/c3KoWrZYxK0vH40gp/QKvtsvDhpjr+eVohtd1mpGIRG2u6WbSjmb8cn8m6qi6umZZOe5+Vv87O5u+LSvwf5lNHxVHTaWLRjhb+cWoBTy/fv2LvMjl4cVU1D56Sz9+/8Kk2EvVKsqO1pIar6DLZA1IJqeHqYYuCHi+MiA+lssPERRNTiNDKOT43CqPVyYdXT6TLZEcjl1DS3McjS8o5Z1wi8TolfVYHcaE+BdC+xhyz3YV+YDDGYMQigdouE+39diI0clIj1AjA0pI2bvpoB3q1jBOH0bevruxkfKoe6OGKqal09Nv5aEvg3Z7Z4UYQoN1o55jMSDKitVidnoCcOfjqCj0WJ3fPzeXLnS1DUmdb6npo6bPx0upq/x3G1oGLXkqEmknpcl5ZXc3cQl+ueHVlJ1/ubOGNS8fx6JkjSNSryBRV+geY7H+D3bS3NrCrO5yChDD/5pousz+I72NdVTd/OT7T/3h9dTc7Gg1HJJA39lp4cWU1H21pRKuQcNvsbEYnhlHc0EtsqAIB/A1wdpODWz/eyXtXKpiSEYHV4aKqw0SP2UGiXuX7+/1E3QNAq5Ry48xMTiiIodfsS8OkDborGZw+WVvVNWTx99Kqao7PjT6si8kv5U8VyC+cmOz3lN5HrE7FzoZenjxnJK19dqRigY013f528a31vVw0MZlTR8Xz9a4W5BJfvrCuy4x4mERkdoyWUKVk2IaDsUlhGO2BHYRGuwuj3U2EVklxg4HsGC3H50axs6mPUKWU1HA1YSqp/3zGpeh5cnklG6q7+cdpBYSppPRafHcNFe2VXD4lBZvD5W+n34fL46Wy3cRz549ma71PB6+Uirnpg+1cNS09IC8eopLz/MpqqjsDv+T7ipALPijmqXNG8fyqap4+dxSdRgcauZhdTQZcHrjpuAxah9GMb28w8MTZI/nmxmOwOFwYbU4MVgdzC2OZlB6OSBB4bW0tiXoVSXolX+1qJSZE4TeqApiQqueTbU1+qd7s/GgeP2skmgMGHytkYs4uSqC6w8RHA81WC2Zm0Npn4+ppaXy7u5W9LX38fW4udy3cjd3lQRB86aowlZR5T6/F4nATr1Pw4PwCttX30mG0c9eJOXxe3IRONbRYlhapZnxKGHKxiJLmPuaPigtYIOwjUa+ios1IZpSGHrMDnWr423GpWKCx14JyGAuJ6BA5Grkk4GK9D4vDTXK4mnPGJ/HSIKvZc8clMi5Fv79FvdcEyjCwDrrrEctoJ5zytsCFi2c4jSxDrYT3NUD9UoxWJ168fLS5wZ+G67U4uWdRCXefmENxg29494ryob4xxfW9jEzQ8fKaav8iQikV88olRUzNiDik11fLJYxKDKPHZKfP5sJkdw07VPtAp03wjSY8hOvFr8KfKpCPT9XzysVFvLy6GpEgMH9UHO9vbmBaZiR3LyzhrKJEFm5vot/qQiTARZNS0Kuk9FgcFKWEMW9ENA6XT471l492cubYBCanh7N+oDCplom5Znoazyyv4qQRsczIjmTFgJwuNlRBblwI2xt8HXpfDmrz12nVVPc46DDamJ0fzb8GNd2oZWKeOncU95+ch1QsIjtGi1gkYLK7WF/VzW1zsjFanSwr7eC43CgkIgGZROwfQbePEIWv4+6RJWXcMy+HJ76v9M+1dHsCLy4VbUYmZ4QPCeRJehVLStqwuzy09ds5Z2wiDpebleXtrBzkJTIyIZRrpw+13C1KCcPt9Y07U0rFaBQSFry/3a9x18ol/PvMQirajXSbHby3qYFbZmWxo9FAaWs/41P0nFgYE6BFX7qnndqZFgrjQzE7XPSYHGgVEvJiQ5iUHh5Q/1BKxXy5s4UIjYzpWZGY7G5KWgy8eOEYOvrtSMQi5BKB2z7Z5X9vLp6UwvXvFfvTTIIA98zNpa3fFiDBU8nEXDA+CafbQ4JeyfLSTu5eWMJFE5N5ec3+YDomSce2ul4+3NLIsdmRdJnsaOVS5uRH+y9OADOyI3G6Pcx7ei2T08OZnRftXxGLRQL/PK2QuAFXyRsb9nu0CILP714uEXPt9HSmpIdT22UhUa9kRIIu0GckLBnOeB0+u9wXzOVaaqc8yqPr3dxzUqAnUGqEeohMNi9WS1NvYIEwJ2ao4+ahYLK7WFXeydM/VDAzO4qF24dOi/Tl4TVEa+VcOTWVDqOd7/a0+z+n0aEKytv7A+4E5VIRG6u7UEpFJIapftIC2Ov1sqGmm7s+3019t4WilDAenF9AXmxgqmlqZjjK5YFNfzfMyDhk35sjzZ8qkKvlEmblRXNMZgRut5et9T2Utho5NjsKi8PNmopO5uT5PEsun5rKirJO/4dELhHx7PmjeeCrEs4el4RULPDptibmFcZyy6wsorRyQpVS/vbpbvqsTkpa+rlxRjrHZkfRbbbTb3Xx+NIKHG4Pt8/OIjlcRVOvlXPHxtLc70Yjl6BXyVha0uZvItGppISpZNR0mjHZnZS1mdjZZKCm0+y/jf58ezOXTk7h5uMzeWt9HctKO5hbEM2/Ti/k4a/30tJnIzZUwVXHpPH0D75uO4PFFTCcOFmvDhjUcNLIWBLDVGyO7vHXB47NiqTTaPcHNIPFgUQs4r/f13LzcZmUtxtp77eTHaPlyqmprK3s5MIJSby/uQGP13ch++usLE5/bh19NhdpEWpGJeoCGpWMdhc/lHVy5th4PtzcSEaUhkcWl5EfF0JhfCj13WZcbl/qaJ/eXCOX4PV42VTTzXMrqlhd2UV2jJa7T8zBeYB/ypKSNm45PpPX19XxWXEzx2RGEK4JQavwWQ5vqu3m8imp/lVuXKiC6oEpS/vwemFVRSdWh5tpmRH8dVYWHq9Pm9xndbKhutufyxWLBPa29vHQ/HwaeqxEh8hJDlf5G2tKW/uZnRfDOxvrOWtsAjcfl0ltl5mxyWGMT9Vz9osbMNldLN3bzqy8aP46K4sEvZK4UCVdRhvf7GohK0rDs+eN5uU1NWhkEq49Np3RA4XdMJWMaVlRTMv6kS9FxkzMl/5Aa1Md27rEPLfRwwUTE5l4QKFSJAj85fhMVpZ3DnQehzMnP4anB3z1VTIxf5uTzcgEn7ba4XLjHlAGHQpban2SToCYECXxYcqAO7FQpZQorZxrpqXzWXETL66uQS4Rce74JBLDlNT3WBiXog+wi4gOkXPF1DReXFXNsyuqiQ1V8PS5oxmXOnwhG6Cm08zlb27xa/S31vXylw+38+HVE9Gr9zczFcTr+PiaiSzd206v2cG8EXGMTtId0u/6a/CnCuT7UEjFIIWpmZF8eOU4mnqtyCUiarrMHJsdyZlj41HLxAErUrvLwwsra8iLDcXudCMTi3C63SwuaUUtl5ARqabFYOWOE7Jp7bMhEQmIBYHHv68IWBmDb4Dw1PRwRiXpyIrW4nB56DI5yInVYnK4mJQeQW6slq31vXQa7YSqpJS3G2k2WDllZNwQlc07G+vRKiS43F6umJrKlztaSNKruem4TLpMdtr6fEOO9+WcFYMGHJ8yMpbtDb3cOisLu8uDXCJCIRHxwFd7ePTMkdR1m4kJVfDFjhY+HEhRFMSH0NJnQ6uQcNroePqsTmZkRxETqiQhTElNl5n3NjeSHxfCjTMz8Xi9KKVi3tlQT99AkNSpZLQPM/bMYHFgsrn5YmcLfzk+i/RIDcUNvTjdHi6bksoLK6s5fWw8KpkYl9vLA6fk88GWBr8hFviKlde+W8x/zx4Z0Fm5u7mP+DAlj5xeSKfRzpI9bTz6XQW3zc7yp0AqO0yEKHwpC6lEhN01VENsc7qJDVUwLkVPUWoYkRoFu5v6uGdhCYUJoYQoJVwxNQ2X2+PrtHV7MNmcvD7gYrmP9n47arnYny6SiUVcMimZKenh9Fqcfo02wPd72/l+bztPnDOSC17d5K+PjEnScd9JuXx49UTEgnBIWu1Oo426bgsKqZi0SDXq6DRSIlKQGqxMGS8Qr1MOySkvKWnjb5/tIjpETmaUlu/3tjN/VDxPnT+all4rggB9FicNPRbMdjdPLK8YsHBO5djsSEJ+olP2s+Imv7/5mspO7j0pj5LmPhxuD9dNT8ft8bK2qptRiTpiBwr/dpeHt9bX8fR5oxidGEaiXoVx0HfNZ95W7r8Qt/bZuP69Yr5YMOWgefx9Rl+DqWg30WKwBQRygMIE3c9q1d/X9PVr2BT8KQP5Phq7DGSWPc8YpRzmncLfvq7j9XV1jE0OJXGYHHezwUJWdhSjE3V+mdmCmRl8V9LGFzuaeeCUfAwWB3E6BW63F4VUzPnjE3lhUJ4SQCSI2NZgIDlCw1vr69EoJExKC+cf35TRNlD0fOCrvf4899e7Wrn+2HS21PUGfLn34R7Qrq+s6CQlQs28ETE09lj4eEDV8VlxM1anr8h28aRkItQyFszMIEThc62747PdAcd74JR8LA43/1pcygUTkjHbXZwxJoFRiTpEAnQaHbyzsZ4XLxzDiytr2NFk8D/38ikpJA9ohve09Pv9tuePimN3y36b2D0tfVw/I4N1VYGzO88Yk4Ba7ksNPfF9BZlRGoqS9XQarayv7mZ7o4EbZ2YwMiGUiyal8MzyCo7Li/EH8X1YnW7qus3ccUIOnxU3Udtp5pZZWSSEqejotyGXiiiID6XFYEUhFfu9cT4vbuLuebk8+NVe3zSiCclDio2njIwjUa/kxve3c9roBP52QjbPrayiosPIqaPjuP5YnzRz34VTJRNz25xs3B4vrgPSWC+uquGty8ZxxwnZdJkcfLqtkTNe3MC/zyhELRMH3LFIxb4xd9cdm47L7WVxSSvFDQY21vaQqFcfkhd4eZuR697dSs2AT8r54xO5dVY2EVr5QfXV3SY7Tyzz3WW099v9cs2S5j6yorW4PV6ufW8bdV0WRAKcXZSITCxiV1MfN324g2fOG83JI4caeu2jtsvEqEQdCWFKQhRSPtnWxPMrq/nHaQVEaeX8a3GZv9lrycCEpTFJOoobDADUd1k4ZaRPLZIdo+WR0wv5vy/34PEyxNWyc2Bhc7BAHjpMvUIlEw8xD/s5mO0uNtZ08/q6WhQSMVdMTaUoRX9EB1UfkUAuCIIOeBUowGeZfLnX6x1qzPE7ot/qpLmqhJStT4HHTdq8KTxzZi4OJDT12ZEP8ybPyY9hVm4UpW1GXrtkHA09Zhp6LJS3mzhjbDwauZgeM3yzq40QpYTJ6eFMz4rk+9IOqjpMiAQ4d1wSSXoFZ45J4OFvS/3H/mBzA7fPyaa510q8TjmkWPnepgYWzMjAaHMSpZUH/HxqRgRdAzaz2+p7uWFGGi43JOhVOF1eXrhwDDWdZhRSEbub+/hwSyNlbUb+fcYIzA5XwKo1N1ZLol7JU+eO9uemp2ZGIhMLfLCpgY4Bdcn0rAgMVmdAEAff3cEjpxeSpFfRMKjBwuPxMiN7v9eK3eVhZ6OBu+fm8MqaWgTg+mPT2dlkYFlpBxdNTKahx8IXO1qo6jRx++xsXl5TQ3a0hugQBR9cPYmaThOlbSaOzfYOMSkDn7z08aWlPHRqPpFaBS+sqGZbgy8llR2t5YSCaKZnRTIqIZRnzhtFs8FKTZeFfouDSyenIAgCgsjn976kpA2H28NxOVF8tauVKI2c647N4NGl5Zw/IZGSZl8DT2lrPyFKaYDixuJws7PRQFqEmowoDW9dNo4XV9dgc7g5Pi+aV9ZUc/qYRG79eKf/Ofd9sYcH5udzz8IS7C4PMrGIZ84fzafbGllW2oFcIuK88UnUdpkx292+foCfCOQOl5uXVlf7gzjA+5sbmZETzay8wJy42+Nlb0sf5e1GYkMUQxQa+/axOtw8trTcb6Dl8cKHWxq5bXa2P/312toaZuVF++6ED6C2y8xFr26iacCqVyTA3XNzeXJZJemRGswOtz+I7+Ob3a0smJnhD+QpEfsXXXKpmLOLEhmfqqet38ZzK6oCLsJKqZgw9cG13plRWi6cmMS7G/cPpLjvpDzidArqB1RcP1eVs6G6myvf3u/V8kN5Bx9eNZEJaUdOZ3+kVuRPAUu8Xu+ZgiDIgN+9R2R7vw2rxQge35e/za3l2s/28sjphWyt60WElxcuGEN5uxGP10vfQE745o92ck5RIje8X8zz54/xF2UmpobT0mfjP9/t14Av29vB8xeM4epjUnF7vGgUUtr7bWyr66Vy0ABj8AWd1j4bqZFqWvsCtc3gG1UWGypndWUXl09NZVeTgfI2o+/2PiWM/yzxve7END09Zid3LyzxPzcjSsNZYxMQBN8XJypEzoKZGZS2GJCIJfzztEKkEt8giHSFGbfIwddlXeTEhhCmkvL1zhZGJem4cGIyIkEgP05Lc58NsSBwwYQkPt7aSLxOicvjpanXSqfRzpljE2josVDa2s+YpDBUMjEeLxyfG8Xysg4UEjHpkWrm5MVw2uh4WvtsXPXWVtoHLlDP/FDFLcdn8vCp+fRZXSzc3kxMiJwH5xcQE6qgx+wrap4+Jg61XMyD8/P5+6IS/wrsxIIYdjT4GoE21vQQpVX4gzj4bFjHpYSxrLQDvVrGa2tr6be5iA1V8OD8fF5fW8uGAYtcsUhgQkoY541P5O7PS3C4PUhEAvNGxPqChOCbAvTiqhrqui3DDljoMtlZMCODkQk6KjqMCECIUsrTyytJCVfz3Z7AztQOo52le9r45JpJ7Gruw+pwsbG6m+8H5KM2p4c31tVx2+xsFFIR+h8JTvvos7pYM8jcbB8V7cYhgXxTbTcXv7YZl8dLhEbG5VNT8fS3oZLAW3scdPQ7KIgPpdfiYG3V0GNaHC7/sHGNXDpspynA9oZefxAH3/6fb2/mk2smkhMbMuz5erwg4DvgpDT9EM2/SCSQFqkhXqfk7hNz+cfAgkkkwD9PLyDlIHce4Pub3DY7m7mFsXT020kOVxGhkfOPb0p5b1MDKpmY20/I5tSR8WgPofnH6fbwxrpA+wCvFxbvbvt9BXJBEEKBacClAF6v1wEMHW3yG+LxeHG4Pf4VgNfrparDRG2XGZ1KSla0FgSwhGbhip+ApLsMiUJDuFpGk8HKjOxI7C4Pt3y8A5vTtxq6dXYm729qpMfsQCoRkIpFlLT2MT0rkh2NBtRyCa+sCUyhONweNtR0kxmlZnlZB8tKO5CJRVw4MYnC+BDW1wSmFWxON0abC2FgatHg2+pLJiezvLSD2QUxlLb2MyU9gsxIDcvKOnzpAqOdeJ2SkYk6Hvp6b8BxqzpMaBQS7hkU3FdVdPHqxUXcvXA3Tb0+462rp6Uxx/E2JZnX02t1+S1rZ+VGkxahxun2Eq2V8e6mBpYMKCyS9Cqev2AsX+1sQSYRMS4ljKQwJVWdFrRyMfNHxXLbJ7v8t+PjU/XcfFwmE1L1xOkUJIer2dloYGV5hz+I7+PdjQ18dM0EGnt8Him7m/q47t1i9GoZ509IIkIjw+rw8NRAMPzv2SPpsziID1ORqFfS1GPF7fXy36UVfkfCwextNZIa6fvbjEkKY2VFJ619NhZub+b+U/L51+JS8uN0KKQiVDIxBouTUJWUTqOdSK0ch9vNMRkRuFwe5hXGUtNp5oeyDn9RcDAn5McQG6oYaDM30Wd1khcbQnaMlrY+G+M0YUPOL0wlIztGw5c7m2nvt7PrgAlG4LtAnD46jkjtT8/KDFVKmJoezsIdgYqQzOjALlCT3cmj35UzLlXPuJQw0rQejnWtRFf2T3BaOWvMdRgLLiIuNgSzw0VRsn5IMFcOXLgFAa49Nh2ZZPjcveGA+hH4UjmRWgWCIJAZrSFJr6ShZ//iZt6IWCam6ZmQNoHsKC3hB3FUlEvFXDQpiQlpetr7bSSE+ew0fkpTrlPJmJweQavBisvrZU15p9+8rN/m4t5Fe0jSq5ieFeV/Tq/ZQVlbPwaLk5QINVnRPnWZgE85cyCyYbYdDkdiRZ4KdAJvCIIwEtgG3Oz1egOWnIIgXA1cDZCUlHQEXnZ4Slv7eX9TPdsbDZw2Op65BbFUd5q44q2t2F0eLpmcwufFzXxe3IxOJcV23AvMtn/PZ7s6uW12ln+e5Jvr6/zpBofbw+NLK7n22DReX1tHTnQIp4+Jx2xzMSM7CovNxYaqLv8qYTChSim7mvv9E4Mcbg+vr6vjsTNH+Fcs4FstZEZpMNrdvLm+jltmZfmaPgw2pmZEMCUjnJouMwve3+6/zT1/fCL3zstla30vNx+XSZ/Vyd6W/mFHhPWafV8YuUREcriK3Bgt/11a7m8Ecrq9PLeimmmnncSGRjMrBuntvy9tpyA+hDfW1/GX4zL9QRx8dgWfFTfRarByTFYkrX02zHY3MSEKCnRWbOYWrpiSQpfZycLiZjbX9jAjO5KJaeHYnG421Xaz4L1izp8wdOKPWi6m2+RgS20P/XaXPy3TbXbw4Nd7eWh+vr+1v7LDxF8/3smbl40jLVLFirIuPtrSSGa0hnPGJdLWZxuiPR6REMo3u1oZmxJGw6BxYFvqevl2Vyuz8mJ48vtK/3QkuUTEHSfk8NWuFq6amgoCnDs+kTNf2oDXC0+fO4pbjs/Cixev1zcqzuuF08fED/QvCMikIpL1Subkx/DB5noK4nXMLYxFKsLfEwADcsaJycilEmblxfDvJWWkRqiHjC3LjtEwInHoRWA4ZBIx1xybzvZGA3UDxzm7KNGvctmHzeFhUlo45e1Gnvmhim9OtKL74a/+n2s3/RdteCwkXoVaJuHOE3O47M0t/nm2Z45JQCuXcHZRInMLfb7uB2NkQmiAYgoG6jgDwTk2VMlrl4zj8+ImNtb0cNKIWOYUxBzyYAiF9NDa8/fR0W/z3TXX9/L49xXYnB5OHR3PeeMT+WDz/rkBG2t6/IG8x2znwa/2smjgAikRCbxySREzsqOQiEU+BVx5p/93lIoFJqbq8Xi8/rmyh8uRCOQSYAxwo9fr3SQIwlPAncC9g3fyer0vAy8DFBUV/SrWYQ3dZi58dZPflrakuR+5WMxLa6qxuzwk6pU4XG6/+qLDaOdvi8rQnH8yo71muswODGYHyeHqIf7dDrcHjxdum5PFnQt3cf74JIx2Nw9/W8q4ZD1zCqIZkagLMNeRS0SkhKtYOMiqdR9b6np54BRfIFLLxZw2OoGV5e2UtpqYPyqeh78pJT8uhMgBz5TdzQZeXl0bkKt8f3MjRSl6nlhWiVgQcLg95MWGcGJBbIBOXSISCFFIOHNsAnE6BWWtRo7LjeamD3cMOa86dzhbanuHbN9a18uoRN2QoiLA7qY+/jYnm4e+2etv2ZdLRLx04Rhu/6yGTlMzIQoJd8/NIUliIDIuCkEQ2N5goLXPSqfJgcfrJVIj9wdN8AWZH8o6UcnELD5geLHX63OnG5zft7s8VLabBmaU+ixwX15dw+LdbTx8WkGArn9cShgyiQir083Y5LAA061JaXpGJ4ays7k/4HzsLg9dJhujE3Xc8MF2vAOySt/vXso17xbz5YIpZEVreHWNiWMyIxEEX87Y4/Vy8aRk7vxsN5UdvuLelcek8e/F5ZS29vP4WSP59NrJ7Gnpx+31kBsbQk6MT7s8NjmMv52QQ3OvleKGXv/dRUF8CFMyIof8PX6MnJgQPr5mErUDWv60SM2QQl64RoZULGJ5aQeJeiWxncuGHmjbmzDqfJCpKYgP5Z3Lx7NkTxuCILCxupsV5b4O54K40B8t6hXGh/LaJUX8e3EZ3WbfvNczxiQG7JMZreWOE3NxuT3DKj5aDFaaei3olDJSI9VIB/axOlxUd5ox2Vwkhat+Mre9vaGXG9/fzrnjkwJsMj7e2sgVU1MDLK+TBl1I9rYY/UEcfM1393y+m0ULphClVZAUpuLZ80azvLQDsUhgVKKODqON2i4z6VFHZmrSkQjkTUCT1+vdNPD4U3yB/Denot0U4C0O0G600ThwWzY5PWLILE2A7Y19hCikZEVrKG83khnt+3APLljJxCJGxofSYbRzz9xcPi9u9ndb7mrqY3VlJ/efkufrXNzdilYhJTc2hI+3NpIRqaa2KzAnHq6R8crqGnJjQxiXoqe8zUhWdAirK7oIV8t4+txRmOwuksNVlLYaiQ1VDVtwsjk9/G1ONg63h7hQBa+uqSEr2meF++3uVuJ0Ss4cm0CYWord6d7f8SYT+xtapmSEMz5Fj9vrJTlCQ26sz/DJ5fH4UyIjE3UoZSLfwIVBY+nAF2gMVmeA74rd5eH9zQ08cXI89y9ro6rTwn1f7GXhWeGk6MSUt/Xz/MpqxqX4VpMvr67humPTMdvd9FmdTEzT89b6Ohp6LNxyfBaRGpl/xbcPpUw8xH7VZPflgadmRlDebuSyySmYHG76bU4umpjM6KQwEsKUeL1eeixOHjmjkPVV3f5VYU6MhpNHxnHvl3sH2u0DCVP78qX7aO2zsXRvO/85YwQtBisGiwOFVMLV09K57r1t1HZZEAR44uxR3PTBDv9nakejAYPFwfzRcXyytQmL0834KM2wX2yJWMTEtHBcbg+FCaHUdJpRSkXkxIYMa/T0U0SFKH60Mcbp8bCqwnfB67M6sagTGLLmj8gC8f6URs5AmuW+L/awp6WfSWl67j0p7ycHScgkYmbmRDM2OQy70/Oj5zVcEC+u7+Wad7bRabIjEQnceWIO509IwuHy8MLKar8dQ6RGzuuXFh1ULthltHPzh9vxAjVdQ72VVld0UpSi57s9baRHqJmYvv+z0W0eatXR0mfDZHMRpfUpZe5eWEKYSorL4+WTbU0khCl58pxf1jw1HIcdyL1eb5sgCI2CIGR7vd5y4Dhg708979dguJb58lYjE1L1bKrtodNoJy5UQZfJTkyIArlEhF4tY0xiKC4PbCyto9PsYl1lNw+cks99X5RgdrhRSEU8NL8AuVhgT0s/USFyfxDfR2WHiQ6jHaPVyZyCGDZVd/PfpeXkxoZw7fR0ttb3+m+b82JDKIgPod/mwuHyoJSLaek1U9dl4r9nj2RjTQ93L9zNghmZPPNDFRtreogJUXDrrCxeX1fnV4OIRQJGm5N/LS7zP37q3FF0mexsKO7m/lPy+XZ3G//8tpQ7T8jh692t/vPd5ym9t6Ufo93FEwONHZ+FynnkzJGY7S7fOLpwNavLOyhKCeP2T3cxKS2c08fE+w3BMqM0jEwMHTKBHqCp10afTcN7Yyt5uCaTryos1HsjSfTIeOjrYtZWdWFzuv3He3JZJRq5hIsmJvHa2lr2tPQTE6Kgot3IueOTuH9AUgaQoFOSGakJuLgdkxlBc6+FMclhPDqo6JwQpuT/Ts5jZ5OBpwZ1/QE8ftYI8uNDeCptFAqpCAGBsjYjcwpiiBxGBTLc2nLbQGv4S6trmFsYQ3ZMCDmxvpVvfbcFlUxCe79tiHdMXbfFL8sLG6blfzBWh4t1Vd28sLIKtxeuOzbtsCRxP4ZMLGZ8qp4djQb6rS6KpWOI1cQhMg2sOqUqmHQDiANff2yynvevnECf1YVe7RuwYra7UEjFw343AUqaDWys6cHr9TIhLZxIrfyQfFHA51L6t892+u+aXB4vD39TypikMKxOX4ryuunpyKU+m+et9b2kRqiH2DmAT/zQ0GNFI5cQrh76d0+PUnNeUSJnjo0nLzaU+LD9q/uUcPWQ9NDk9HCiB+oWvWYnfVZnQD9JU6912Pm7v5Qj9Um4EXhvQLFSA1x2hI77s8iO0Q6ZhJMTq+XEwhj+9ukuVld08sTZI2nstdLaZ/P5VSgkpKqdqOqWcnLHq7gkWmoKbmBpl5Z/nzGC8jYjbq+Xb3e3MCEtgvc3N3DzcZnDvn6Lwde2/fGWRkx2FxdNSmZkfCjflbRzyeQUNHIJ0SFypCIRfxkopI5NDiMtUsXMvBgeWVzGBa9uJl6n5O/z8lhe2u4fLtzWb+Ohb0q5Z24OD35dSrhaxm1zsnlhkJuf2+Pl9bW13HRcJnq1bGCgsRGn20u/zRXwQbO7PDy1rJL/OzmPWwbJ3pr77Ly2phaT3eX3ZHnxwrGsqeyi02jn610tPHBKPkl6FR6vlxaDjX9+WzpkriTA9KxI7lvawIeTpVyZ1ss3lXL0IRq+L233F8e21vf6lQIqmRiH2826qm6//vyccYm8taGOyzRy7j0pD4fLg0YhoanHyuvravnbnGwsTjehCgluL+hU0gCLA/B9abpNDgrjdUMmQn2wpZGJaeF812ggVqfk462NeL0+Q7G/z8vl1llZfLi5AbPDzTXT0gK+wPsYmaDze5N8u7uNGdlRTE4PJz5M5S9C9tuG1i3kEt9g6/+cOYLKDiMGi5OC+BC/jNDqdFHWaqS+24JcKuLz4ma2DUjurnmnmNcvLWJmzn61Sc9AQNMfgp78pzhzbALf7m6lqdfKX3+wIZ3zGjN1bQhuB22KdJ7bJKWwqZ5p2ZEBPRehKhmhKhlNvRbe29TAwu3NjEwM5ZyiJDxeD6FKGWmRGsQigR2NBp78voKChFA6jXZeWFnNq5eMY0zyoeX8u012qg5Qf4HPd93qdHHjTF+3875Ar1fLGJWoY3TS0OPrVFK/gZpcKvK7WIKvXnHt9Ay/FfKB5MZqeea80dy7qIRei5OilDD+7+R81AN2CGr50EKvRCQM66HzSxG8B7re/AYUFRV5t27d+tM7/gJqOk2squhkb2s/M7KiGJ+qJ0Irp8/ioMNoZ0V5R8AA4bxYLZ9Mqkf97YKA47Sd/jkT398vizo2K5Jus4PdzX1MzYhAJhHxw6CC4JgkHQlhKr7c2eK/WIxO1PHcyiqitAoyonwt3OFqGdkxWq54ayujEnUUxocyPiWMx5dVBqRf5BIR10xPC/CNAJ+mNUQhoWqg6/TFA5qNkvQqxiT5Lg4xWjkJejX/XlJGqFKKIPjUKvvIjNJw0shYnvi+MuAYerWME/JjeH+zT0t7xdQU1HKJ/1xunZXF44MsRQEunphIflwoTy6vwup0c9roeLpNDr7c2cJnx5uIs9fwleZsXF4PfRYXnxc3B+SfAf5zRiF6tZydTQYaeyxkRmtZW9mFXi1lXmEsYrHAf5dWcEJ+DE8f4IUuEQl+vfx/B6wQBvPg/HyfD41YxPMrfUOKL56cQmF8KHVdJmJ1Sm75aGfAc/LjQsiJ0SIW+eZ97mo0EBkiZ1NND+8NFDJjQhRcPS2Nh77Z679QXjQxmfI2I4+fM9JflFtS0srikja/5SvA3+floJZJuGuQmmhuYQwPn1pAmErGx1sauePz3QE/M9vd/rTHjJwoXr+kiH6bkyUlbf/P3ltHt3Vme/+fI2bJlszMFDt2mJlTSJu2KfOUGaZTmNJ0pjCdcqed4pSZ26TBpmFGx3HMzLYsy2L6/XEcxYrTuTP37fveO7N+e62utpJ8dCSds5/97P0FXlhfQygU4pa5OSwelYBRI6fH7sHtDxBnUIV7x/9stPW7qO60I5VIiDcqqe1y0NzvpMPm5t1tjXgDQaZnm3nxwjERImIen597vyqPkHGO1ipYMS6FN7fU8+KFZWgVUvocXj7e3cy22l6So9RcNiWdbrub+5YU/lPn1zfo4ZxXt0Xg4gG+uH4ywaAoeXz8Gj4eV03L4Pennfr4ays6ueGDvYRCcPnUdNLN2rBuT07cf90Gaet34fD4iTOoInTJ9zT08tGuZr4Y9n1cMTWdM0YnUJb6y4PgU4UgCHtDodC4kx//j2N2ZsboTimqb9QoGPT4RyTGbJME7b7XRrw+qu1nfrf4MnoHPfxc1UPPoId0i5bDrTa21PRwwYQUbpiVRUOvg7RoLR5/gLe2NgDQ0u9i47Eu/MEgcqlI8x6e+C6bnMboZCOz82P5Zn8rBQn6ET10jz9I7CkgZUa1nCNtNnQqGTmx+gjRLhDVAL/e38rqIwH+dslYtlR3c9eCvKGtbYjMGB17G63kxekZk2bCexLzDUTPyOGaFYOeAGOGVTEV7QMRzvYxeiXnxbZScOxRUpf+kVX1YmJpH9J5SXbvQxWfx9LMeJa8uBWFVOCuhXn87svD4eQ3LdtCr8NDUpSG7FgdWTFaVHIphQl6dtb38cA3R7DoFJw7LgWPLzBCFCzNoqGp18nne1vCfefjoZZLsQ6pDMbqlDx4egEef5BHvq3glY21JJnUXD0tY8T3cKRtgJtmZ4MA22t6+MtQ++mcMUk8tbyEXocXlVzCE6tO2OyNT49iQnp02OHoeCLfUdeH3e3nzgWiFIJaJkWrlIXbYsdj5eEOLp6URpJRzSMnwUhXHu7g9vm54UQepZYPDRf7Iti5v/vyMEa1HKlE4OFvj9A96OG8cSlcOzOL1H/BqSfRpCbRpKa2a5Cr390TvkbjDEpumZvD02uOsbmml9puB2PTTiTyA802vj5pwN/n8KKQSUg1a9hc3YOAaP69c2iw3mJ18eSqSh4ZkjDwB4JIBOEfojqidUqeXD6aq9/dg83lQyLAnQtyyYs3QCjEKxtH8jEq2my/aOc3Jz+WlbdMp7nPiVmnFA1i/gURrF8aplr0KhyeYb+9XMqehl4sun99vvFL8R+XyE8VTb0OjrQNEKdXjOhL2bwh/OroEV9Ef0hHdaedzBgtU7LMmDRyCocICjaXj492NZMdq+WGWVk8+E1FRP9TJZMyIT2a0ckmZBLJiOr1vR2NPLZsFG1DBKBorQKNQhohZAViVT68VbSgIBaNQsr3h9rDzM5zxiZz6eQ0VpV3cFpJAn0Ob3joGAgGSIrS8PB3Fbi8fpaPTWZUooHarkF2NfRR0zXIjFwLZ5Ul8fWBVkIhyIrRMj49mieGJZh5BbGsOtzOHfNz+WhXE9tqe7h5TjaXlerxtpWT7jtK4k9/AZ+TMboE3hm4iHabm8mZZm6blcLuDi1BIZ4CX5DkKDVH2gaI0sh55PQicUglldDR7yI3Ts89X5wwth6dbGR8ejRvDMkh9Dm8PL7yKPctKeC2eTl8uqeZo+12ylJNrBiXzO4Ga9jT8oqp6Wyq6iY5Si2igL6vYPnYZAZcPrF631QXRn+09rtwn0JTpSTZyNcHWll/tItrZmSG1f8+39fK1GwLAqJufGaMjrruQe5akMe22h7e3lpPfrw+YutcmmLi79saInZxL1xQekp8e5/DS3DID/bkCOt1SAQunCjCeD/b2zzidZ/sbsbm9tE2JCv7wc4mFDIJDywt/MV+9S/F2orOiEKjc8BDY5+DlGg1zX0uAsEgHn+ADpsbCQJHOwaQSSQjdkWCAAsK43hjcz23z8/hvZM8Sf3BEHKpwLqKTv6+rR6dUsblUzMYmxb1i7uJ8RnRfHfztDBqJStGG9abObMsiQ0n4fmXj03+xR68VCKQE6f/p6rvfyZsTi+CIJBu1nLNjEw+2NnEmiOdqBVSnltRSsqvYH93PP7jE3lVp51L39zJdTOz+Hp/M+eMTQ7DDwH2tzoZOOdmohs2QWjowlMaaIyaiNQh4c+rTyThSRnR/P60Aty+IFqljOpOOzJJpLBSVoyOzgE3SrmUP62s5NoZIyu9YEgU7kqP1rC9tofarkHunJ/LH4ahIZaVJrH6SAcPLC3gcKuNTIuOKK2MV3+uj6Dnf763hZcvLANEtthxxbhxaVG4fUHu//rEtv3FDTX8dmEeHTY3jb1OWqwuLpmcxubqXm6ZI3qLyiQC0Vo5Y1KjUMhEX9PmPidf7m/DqJazaFQ82TFaajsHOd+7Fv2WP0R8NmXNKs6YehU5idlkmLVc9t6hIXhgLyXJRu5ZmMeV7+xhbUUX8QYlMqmEYDBEgknFkbaBCMee0SmmsHnA8O8uEAzx/Poq5hXEcce8HFKNctZU9lGQYABa+XRPMzfPyeam2VmEEHB6/dyxIJeVhzuo73GwsCie88en8tqmE22pTdU93DAri79tqiMQDBFvULGsNIkPdjZy/hAVfnZebFjGtb7XwQvra5AI8NIFZZg0Cj7Z3UxpShRVnXZMGgW1QzDDpj4nm6u7WT4mia/2txIckjgoTTGFB/HHQy4VGHD5WH+0i9w4XYQ7lUouoSDewK1zc5iVFxPGR6dGa1ANDfSOM1uLkw2oFTKm51jY22hla00vn+1p4ZrpmST8ixTzgyfJMADUdztINmnQK+WiXs/nh/j2YBvRWgW3z8vlrgW5NPU5qWgfYF9TP9mxWqwOLwa1nNDQzvBUph4ahSyCzr6mopNPr53MuH+ARU+N1pxypzEt28KdC3LDNP1rZ2YxM/dfg2v+d8Lm8rKuopOXfqolNVrNBRNSeXdHI32DXn63OJ+p2WbSLb8O7PB4/Mcn8lWH2/EFRHPg1GgdSUO9uC3V3aREa5ieY2GdQyB99keM8h0kKNOwI1jEMVciX+yL7B3vqO/jiqnpfLG3kTNLE8iO1eHw+Ll1bi7eQACpREK/08uq8nZm5MZQ2z1IcpRmhDlCVoyOPY1WloyK5/a5uTy7rpozShO4a0Eebn8ApUzCngYrvYOiq7w3EGJNRScTM6NJi9aw+aTPWNvtoCjRwIHmfqxOLzNyLcwviGPLSYJUAN8ebOP0UrEvfnZZEpXtA1w4MYWPdzUjEQQWFMXx4c4m9Co5/kCIJ1ZVcsvQcNfm8oVNGp5aXkyfJ5mTaxdnbBkfl9uQKXxsqeqJUJI71GKjpd/FCxeUUd1pJzVag9MbwOULMDsvhufXR37f/U4fFp1ihHqkPxDk3LEpotBTsJlcl52cOXPoGnAz4PbjDwT5fF9LGHZ627wcXttUF65w/76tgQWFIuRt75Ac8O76PhYVxfH8ilJcvgCHW2209buYkG5GIRWxv5kxoiDZT5XdYfJXMAQ3f3yAF88vJcagjNh9JZnUFCYaaOsXCSYGtZyb5mQD4m7jm/1tLBoVj0ImYXN1DxkWDXcvzOPR745SlGTg+llZvLe9kX1N/aSZNdy/pID5hXEsKUkIv0coFGJeYRwKmYi40SilNPQ6CIUEXlpfgycQZFq2hetmZrKhsgvNfzFg8/gDtFldhIB4o4oQYsth1Unm1pOzRELXOWOSeWdbYxhHbXf7cfoC7GroY0+jlfHpUbx84RikErjhg31MzTJz0+xs3tnWwFXTMnh6zbFwW2pBYSwV7ZHop2BI3BEMT+StVic76vuoaBtgbFoU49OjTslsNeuU3DQ7m7PKkgiFxN9DIhGwOryEQqFfZSgMIpO8uc+J2x8g2aRhS3UPdw7ZAK4Yl8INH+wLo63u/7qcJ84u/v8T+T8bLVYnfQ4vRrWcrBgtdrePBUXxXPv+XlQyKePTo+mwucSe5MRU3jhmpjjpbPQyGW9sqeOh0/VolVJidErOGJ2ILyhWqwBquYTuQS/xBhVRWiVf7m8NV2oSAX67KJ+XN9QwNi2K93Y08PAZRby/o5HyNhtjUkWt6RfXV2NQyVh9pINFRfE09jrZUtMTrsDkUoHnVpTy4a6msELgtwfbWFgk6qkP16BIjlJzoMnK4lHx3Lcknz+vruIva6pYMT6Fk8OiUzI6xcTNc3L46VgXh1psLC6K4/zxKShkEvocHg62nHBwX16WhFxChImCQSUjK0ZHm7WImIwFaOrXiC/WmKktvIntXzv43aIUXvrpxDwiNVrDolGip+I72+qZkmUJo2Vm58UyNcssSidwAiK5pqKDPy4bxV2fHyIUEhPKxIxodEoZBYkGXlhXgyJPR1tKGt98fpCK1gEunJCMTCoNJ3GAYGhkm2Ld0U5euKCMggQDX+9v5eJJqby7vYklxeJv4fL60avl5MXpeX9nY/h454xN4qlzSnhg2E4nEAyhVcr4aGfkYK2130Vtt4OHvz3CA6cV4PEFeX9nI3qljCiNAn8wJFK6zRrKUrJZUpzAp3ua6Bhwc+HEVO789CCz82O5aU42HTY3j3x3hOIkY0RFvb+5nyve3h2uxHVKGa9eMoYDTTYun5qORCIglwj0Orz84cwijP8A5tjY6+C5dVV8c6ANg1rO7fNyqe4UUVsXTUhlY1UXKrmUstQopmVbSInWIABfHzjRD18xPoW3t9SHWzprK7qo7Xbwt4vG8t5VEzHrFEgFgfd3NtFt93DXgjxcQ7LQJclGXt9UP+K8hrvx9Dm8/GXNMeKNajQKKZuqutnXaOXuRXkoTyEDIAhCeE4x6PGz/mgnf1lThS8Q5MbZWSwtTiTqFNo4w3/bQbcPnUp+ypbUoNvHp3uaeWr1Mdy+IDNyYlhQJCKJTBo5bTYXJ9M/Xt9cx5LihF/VqPk/LpEHgiHWHe3k7s8PMuDys6w0kQsnpnKkdQABuHZGJsEgxBuVxBvVfLiziYe+PUJJionRKaawP2Bzn4unlpfQ2ufiTz9WEhjq3/3hzFHMK4jFpFHw+d4WjGo5l01OY8W4FPpdPqK1ClzeAGnRai6amEpjr4NAIMiysgTOLE2krd/F8+uqOasscaj36CQYgoPNNiZlmllakohSKiE7TuzRnyzzuvpIJ/cuyWdzdQ8SAc4Zm4JRLeO7Q+1MyDCTH28IV5lZsTpMGjn9Q/h1mUTgyqlpXPPu3ohKuTQ1CpvLxx9+OMr1szJ54fwyuu0esmO17Kjr49uD7RQmGFg+Joln11Zz+/xcDrXa+P6gjUcWP0fc5Faamhs45ovH70vhsWVSFDKBW+fm0NLvQkBknb63vZEQIS6dnE5BgoHb5+Xw3PpqlhbH8fLGWoKhEBdPTOWzvS0EgiFOK0nEolPy24V5xOiU/HSsOzwcjNYquHlONhIB7v2qPGwx9pe11Zw/PlICQnqKnqhGIeNAcz/rKjp5/OxiPtjZiDcQoHfQy56GPh5fXszLG2rod/oiFoXP97aSadFF7BK0CinBUGgEOQnEvvD1s7IYcInkrrPLkihMNPLyTzUcarEhEcSqLc0siqWNTokCGpFJBS6dnI5aLuGHQ+3hHrXd7ed4PR4KhXhve2OEVOugx0/foI8XN1SHHzeoZTy1vIQOm4ej7QPkxelHDBEDwRDvbGsIi8D1O3089O0Rfrswj79tquOuBbloVVIGXCJjtiTZiFohw+b0khqtCTN+xeQVqTNf1+1gwONj6pDdWigU4oIJqbywvjoC7//ZtZO5ZmYGW2tPFCkKqYS5hSc0TRp6Bok3qnlzSz0ef5DUaA2XT0mnocchDjn/Qexp6OPWYWzmB74+gk4pZ1nZqQ2Ta7rsvLe9kY1V3czIieHSyWkj+ucHm208+v2Jluim6m4sOgWFCQYaex2nXFz0v7Ao/J/Ef1wir+0a5KYP9+ELhEg0qpiaJaoSThpCd3yyp5lgEK6cls6BZlt4y9hhc+PxBWjocVDTPchne1qI0si5clpGmHTiC4R4+LsjPLm8hGitguk5MXx1oBX34Q7Gp0fz6s+19Ax6sWgVPHxmEXd9dhCNQsaNs7N4fVM9HQNukqPUPHPeaDbXdIeHmH/bVMfiUfFMyTLj8wdZeVik7ftPweQEMKjk3DI3G4kgUN1pRyIIXD09E4kgUN9tZ2lxPANuH7vrrfzhzCJ6B704vH6SjUoUAw18fkEKp73biCCIYk758Xo8/iDRWgWvbBT7xtNzLGypOQGxrOyws6texZuXj+OVn2qYkCGqzt36ZQ2jk43kxo/FLwT5ansTAy4fN8zO4rM9zbT2u7hxdg5/GibZ+9qmOpRzsqnvcfDOFeOp63ZQbHITK/fgkJt5+PRCmq0ufq7qpqrTTpZFhzlLwQ/DCE19DrEPeemU9AifyAG3f4RMaZfdQ0G8nqPD5AUumZTGNwda6Rhw88DX5dw8JwuLVoVcJiARoKXPRWlqFF+c1KMHUa4h3ayhvsdJjE7JtTNFF5oLJ6ZGeHQaVDK6Bty8OASVHJWo5/KpGdR1DXKoRWwhBEMilj3BpGZ9ZSe/XZjH7fNyUEglfH2gFbcvwEUT00g3a/D4g6gUJ6rTYDBEW38kMqMo0cCq8vaI5D4wpHq4tqITq9PL21dMGOFh2TvoOaWUhNXp47qZmTyxqjIsW/Hx7mZeu2QsC4riMWoU/P60Qi57S1RKlEpGDiUFAdRy2bD/F1gxhD56b0cj0VoF9y8toGTIXeij30zi24Nt6JRSlpYkMHoYG7Pf5eOvw7gTTX1Oegc9tA5xBbJjdb/IDl1V3j7isfd2NLKkJD7CWPn493Hzh/vD18x7vY3squ/lw99MipALPtmUHUSZ2jNGJ1LRPoBBLQsblRyPW+fl/OoGzf9xibx70MPM3Bia+pzcODuLrTW9+IZkR4djj5/88Ri3z8tBKZNQlGhkTJqJG4YWgHiDit8uyuNPK4+ypbonopfq9gVRyST8dKw7TP891GJjQ2UXz60oBURx+l11fSSZRHGkZ9dWh1EtLVYXD3xdzrnjkiPO++eqbi6cmMol7+8CYGNVN4+fXUxOrI7qrhMDr9HJRvoGRfLElCwLi0fF85t394aT/qNnFrGkOIHP9rSwuqKDgkQ9n+xupq5HtCwrSzFx7VQdq26ewpbaPsw6JSsPtzMx08ztQxdYz6CHTIsuYugEIimpom2AsiQtsTIH75XbqO12Uts9SLxBye3zcqntHuT2eTk8saoSjz/I6GQjW6ojkQMgEoG8/iCDLh9nG49h2HkHDLTiTxzPrqIHeHCzM1zhzsixnNLU93CrLdzuGh6rDnfw1PJinltXTb/LRyAY4oqpGbT0OwkGRYLH5pqecOVoc/kwaRS02Vy89FMNTm8AvVLGg6cVUJpqYv1Jsg7FSUYWj4pna00vjX1OXtxQzekliVw8KZU0s4bP97aQYdYwOcsSRv/MyY9ldLKRD3Y0YVTL+f1pBby5+UQLorXfxa76PtpsbgoSDFzz3t7w+725pZ5b5mbz/o5GXt9cx6NnFDE7X+yLXzI5LWJYatLI6RoYSRk/3mbssnt45NsjfHbd5Ajst0YpugXtbeyP+LsojYw+h2+E9tDz66uZkmVGp5IzOdPMNzdNpaZzkFiDkhXjkvlkGPzziikZZMZEDiOTotT8dlE+l09JRyGTRCTHyVlmJmedWuL1uPjb8bh2RiYbq7p5eSi558TqePXisaeUOog3jBzyJplUyISRi099jyNi4Qc41jlIXY8j4lzjDCP77DkxOs4ek0S6RYtaJuHNy8ZzsKWffqePGbmWiIXp14r/mETeOeCmrnuQHXW9tFhd5MXpidGp+PZgG/cuKWDbKTSTt9f1MjrFxPQcC39Zc2JI1THg5tM9zSwsimd7bS8LiuLCidygkiGVCCPQFB5/EMeQe/373zdidfo4bXQCCUb1CGp2r8NLXryeGL2SbrsHs1bBI2cW0TC0fdYqpFwzM4vqTju3zs1md4M1fK4pURqkUgmvXTKWLruXe744FHHsHruHZ9ZWhdsp3XYPRzvsSCUC9yzKY2+jld99W0VpsonLp6ZT02Gn2eoirteBQirh998cAQi3LYZvCqK1cs5KcRC99wVkx7axOHU+m0vO5tZ1g3QMeAghVrqCIIQrwm67h8LEkVveLIuW/AQ9uYoeDJ9fDIGh9k/bbib47uP8UU/y3kERy364dYCZeSPRBlOzLWgUUmblWiLMn7NidfiDIeYXxiGXSdhY2cXuhj6WFMZQmhbNte/vj6D2K6SiTG2fwxvupds9ft7c2sBvF+ZR0zlI45AswkUTUxmdYsKiU5JgVNPY6+CC8SlkWETYW3asnhXjkqnuHOTyv+/G7vFjUMsoSjSEZRAAttb2cMf8XJ4c0pGPN6iYnmPB4fGPkMAFceA3Ni2atRWdXPfBPr6+YSqjU0xMy7bw5NnFvLSxBrlEwkUTU3H7ghHa6yAuPsc1zxt6HTg8/ohErlPKuWdRPpe8uSv82xUmGrC5/CPMuUEsaMLKnRKBokQjRYliRZ0Tq2d+UTwNPQ4yY0QJB5V8ZKqRSoR/CUHj8ftJMKm4aU42G491idDHUCjMAgZRKuPrA63cuSAPBrvE7YBWvHYWFsXx9tb6cHWslEm4fErGKbHqvyT0pTgJBplm1kQUejqljCUlopb5lVNPINb+kU/orxH/EYm81erk9U2iqP/GIbKE2Aro4+JJaXh9ARKNIy+YRJOaQy22ER59AEeHFAJDoRCNQwnWpJHzhzNHsbO+b8QPCiKk8LjBA8Bne1qYlGEekRAVUglNvU6WFCeQEqVibGoUW2p6wtut62dl8/rmOmwuH+eNS6HZ6iDdrGVXfR+f7WlBEOCVi8ZEOJjH6JQsK0tkVJKR0hQTP1eJspnBoanlmaWJfLW/NTxM3VjVzZH2AZ4+t4R3djSyYnwKdd2D3D4/l/2NVn6u6mZZ2QlNlSumppNiVPDMjk4mxVzEJGks8YdeZlHiUc4b9Xs+OSxqMf9wuJ3fLswLn1ebTdSBtugUYXx7tFZBUpSaB785whvn55F9kvyvrPsIp48NYQ2JyCCTWoS4XTwpjY93NeEPhsiP1zM62cRlb+3mrxeNYVx6FO02NxadqKXT7/QyNdtMY4+Dc8elUNkxgFqlxBMIcuvcnDC6RBDgmpmZKKWSETdvZYedZ9dVcfmUdFy+AJOyzOTH68MkEYNafkoRJk8gRIpZwzljk3l9cz2zcmP5/lDktt4XCNEz6EWnlFGcbMSkkbOrwcOdnx7kvFMMqROM6rBoWCgEtd2DjE4xYdIoWDEhlQVF8UgEkfjW5/DwhzOLePmnWuQygXPGJPNzVXf4GjytJIGYITErt89PXbcTty9AdoyWL68XFRiDhEiP1tA16EEukfDu9saIVt+1MzN/cVhn0SuZVxB3yuf+u9FqdfLSTzVhv9rTSxKZmx93So2frTU93JZUjXTV7SCRw+z7oeB0ChONfHbdFA40W/EHQ4xONlF0iiIDIMuiY1lZUgSx6czSRLJiIk0pjGoF2TFaZuRY8AdDhIDn1lXzt4vH/Hof/p+I/4hEvr+5H4NGzsbtkZVMp91DSpQGk0ZOcrSGL/a1hFdjjULKpEwzmRYd8afYHmXFaHH5/FwxNYODzVbOG59Ks9WJVICCOD0lS408t66a2m4HUolASZKBmmGY3+Px1f4WbpiVxUs/nejr3T4/B7VCSppZy466Xj7a1cL49Chm58dy54Iceh0e9CoZS4rjKUuN4psDrSOc3LvsIpFGIZWQE6dj8ah4Xt9cz+ub65mYEc1dC/L482oR2qVXykiOUkdQpkGslms6B7l+VhZ3fXYw3MpYPCqeGKkSfyDEn84axaDbz9qjnby9Vaw6PgZWjFrEw6lHUDdtZO40J0etRvQqGX0OLzvq+iIYpy9uqOaZ80Zjc/npsLnxB0O8sL6GYAie29TOhOzT0TWuh4nXgiCl3TwRJ6mMSw+xvbaP1Uc6uHFWFtOzzczIsVDeZqOx18mf1xwjEAyxt7GP/HgD/qCI4BmdbOS69/eREq1mWo6Fd7adIJ6kmTWcNzaZ3y7Mw6iRo5ZL+XJfC4NuP2PSTCMW3XkFcfx5zTHuXJAXwW49VQx6/Oxt7KWyfZBQKMSM3Bi8/qDIajzFwh+rV3LPojy21vbi9gXYXttLbpyOcWlRfH+wPQxZ1SqlLCtNZHNND3a3T7zmBIFvDrSSE6ujIMEQgbyI1iq5ZHI6S4oTCARD7G7sY1N1NzfMyiLZpGZqtpmGXic2l4/arkEe+KYcfyBEXryoGXLyQhIMhvjomkm8uaWergEPs/Ji2FLVjV4lY25+7C+aRvyjaLE6cXkDJJrU/7Bf3DHg5ljHAHXdjgg98K8PtPLQaYXMzY+NIFkBLMozIF15KTiGdmnf3AAaM+QtIi9eT178f0340apk3Ls4j3kFsRxqsVGcZGR8etQIwa2UaA1j0qIjdsanlST8U+/xa8Z/hNbK65tqGXD7+evG2hFSry+cX8axjgEUUgkhQayi+p1eZBIJb26px+byMT3HTEGCkdc31xEKiVP+Py8fjVEt44kfK7lyaga3fnKAJKM4qDzUamNrTQ8WnZLZ+TE4PQHK22ykmUUW5vvDGGtjUk1cNSWddrsHjUJGr8PDqsMd5MfrOdZh58gwKnxOrI7fn1ZAW7+bI20D/FjegUWn4LzxKby7vZE+h5crpqYTDIVIjtIgQUweIeCR7yLp3LNzY8iI0SKVCEzJstDW7+L335SPgEK9dEEpf1xZOaIHfcd8kUjx7PmlNPc6R1DJBQFWzu+nYMvNHF22ik8bRenf44O9O+bnkG7WUtvjQC6R4Av4cftC4bnC8Ugwqvh6rpW4QCcDLRWU597EIaucvU39rK3oZEqWmTSziMX/dE8zS4oTI5yYTi9JGBI1O4FzPn98Csc67EzOMofRDcdDKhG4a4HY0rhyajp5cXp6HV5e+bmGG2dlh1sxg94Ay8uSUMql9AzNXQqHWge/FDtqe9hS04NCJsUXCJJkUpMTq8WgVnCwpZ+7PjtETqyOM0oT8QWC5MTq6Hf66BjwkBKl5vfflHP3wnyeXn2MK6elo5JLUckkFCUZeWVjLTVdg0zLsVCcZKSibYDP9orO8+9dNeEf2ob1Ojy8vaWe1zbVMyXbTIZFy7vbGwkEQ+TF6TmrLIknfhR/33PGJvP4WcXIT9qdeP0BXtlYK+LDG6x4/EEEAb68fsopRah+KZxeP98dbOfnY10UJBqQCgIlyUbGpEWNoMO321zc9vEBfIEQKrkkQooCYGq2mfuW5PPmloZwkTK/IJYHs2pIWXdd5BuXrICzR0px/Brh8vo50jZAQ6+TGJ2CokTjfynf+9+N/2itlcIEI39adZSzy5IieteFCXrcvgCrKzq5dFIaD357hFvmZvPyTzUMZw9vru7lyikZJBpVqOSiY/b2ul6cXj9un6jJHQqJ7YWqrkEeG8bA/PZgG7fNy+HvQ1Xf+PSosA6JXCpw/vhUbv30IMvHJrO5qjs83LpgQgpfnoQSqO4axOYStcCP05e7Bz384fsKfrswH7lM4Jk1VeHBU1q0mkfPHBV2ezkeWoWU6bkxrDvaSXmrjaY+J6eVJHDRxLQIWvTiUfHEGVSnHCSGCPHnc0rQDUHrRjwfAn9Ign/8tezvFjjcYkM/rFpptbqp7LCzqryDtGgN187MQqeUjZD7XDE+hZ2BeEYLUh60FvPzR2KSnpxp5sqp6by1tYGJGdEMevy09ruJGuZyftPsbMalRbGnycrUbHMYqvnJnmZunZuDPxiKaAdcPDFVFFBz+bhrQR46pRSVXOCp1ccwaxU0W138WC6yaVcdbuexlUcRgMfPLsF+Ckuy4+ELBGmxunB4A9hcft7bIe6+ZBKBly4sY01FFzvqRGnkYCjEo9+fENhaWhzP+eNTsLn9zMiJ4YdDbXgDwbAY2k1zsrnh/X3Yh+YsX+5rpW/Qiy94wr3qqR+P8c5V49Epxe/G5Q3Q2OdAQCDdrGF3fV94RzguLTrCNOFYp51dDX1hV/qfh1puonPPiZZXt93DSz/VREAsQyHRRvBfSeSHW218c6AVi04ZMZd69Iwi0RN2WL+6vNXGznpRwvm0koQRiTzToiNaq+SPy4q5eloGoRCkmzVof3x75Bub0v/pc/xXQ62QMS49+h+yT/9vx39EIh+dYuL88Snsb+7ntnk5NPY6KUo0UJJsZNDt5/dLC+iye/j9aQU4PH6WFidGOOjkxulw+gI8PFTVlqaYwtjgW+dm4w2EwlvxkwkLomuMF71Sht3jZ3eDlb9eUMLMVDnZFjUBmYRb5+WQYdZGQNn0p9BEBrFi/OQk44ZgSHx8T4MVh1dkfiaZ1LTb3Gyv7SXjpL7dxZPSeHFDdVj/fPWRTlqsLu5akEe6RUuP3UOGRUtqtNh3LU0xcaC5P+IYCQYVv/3iEDfOymZUkoF0iybslA4wNcNIuiWA7GgTy+0PUDj9L5z1gUiI0Q6ZVrT1u0Tz4JZ+qjrtbK7u4d7FBXy+t5l+p48V41OYlm1hf00L6/1x/Fx3YkC3va6XoiQDRrWcdpsb1ZDH4eojHdw+L4c+p1dsTXwp+oFOyTJz98IT7aRgKMTmqm4Wj4rn+0PtnFWWJNLFd574nDfNziaImJh6HV6So9ScNSaJh747woDLz+hkI4uLE4ao53Jsbj9quYS8BENY0KxzwM0bm+v5+zbxulhWmsTyMUl8sa8VfzBEZbudd7Y3hCWLv9zfGrGQ/XC4g9JUE+9tb+K3C/NGsFslAuEkfjw2VnVz85zs8MLV0OvA6QmgU8pptbp4enUlXx1oEzHsM7MYcJ+w+Tv+38NjT0Mfj501ikmZdqxOL3/8voIHzyikOMkUfo1GISMlSj1CadCk+ddILfXdDiZmmnn2JP2hP648ytQcC1nDBO9sQ9dvx4Cb5GgNySY1LUNwyySTmsWj4kkYmn1F7JbGXQVHvgLf0LmqTFB4+r90nv9u8R+RyHUqGZdMTmdWnpPaLjtTMs1YXT5u+nAf18/K5skfK8MDzdIUI79dmM+kzGg2V/dQnGSkNNlEdfcgSSY1V05Lp7bTQVGSAX9QrIwOttgwaeT8dkHuP3U++t6DLPHvhI9foHXx2/TYc5EK8PvTCnhmbTU2lw+7y8fcgtgIaNvMXAuxOsUIzWwQE3lN1yDnjk0m3qiirtvBolHxpJk1tFhdTMu2hDW+VXJJOIkfjyNtA+xvtvLC+hrGp0cxIyeGh74tJ92i5doZmby4oZqKdjsGlYzfTM9kT4OVstQoKjsGUMokLB+TTEOPg4r2AeanKzlbfwT913dAKIQSyPVV8tiZE+iye5AI4A0EcPoC3P9VOVdPz+CbAzUsK03C6fXzwNICDjbb+GBnEx/sbOLexXnsqg8AkUiLirYBsmK0JJjURKlkxOhFVqo3EGR6TgzXvLsn3CraViu6/EzJMtNt92BQyTjUamN8RjR3zM9FJZeMwElvqenmlrk53DI3G5lEgj8YZHSyiTc216OQSlg8KiFCPOzr/W3cvTCPj3c388ezijFpFGyq6o5o9Xy2t4UbZ2eHRdA6BtyY1Ao6fG60SlkYTTQ8XN4gTX1O/rjyKJdPSY9oY52KzKSSS/AN21IuH5uMZQgSt+5oR9hcORQS5QhumydKLHj8wVOaURQlGvh0dwu7G/q4enomWqWbynY7eqUcp9dPoklNlFbBQ2eM4sq/7w63LyekRzMq6R+3m04Os05J60nY9+PnNuiOXLDy4vXcMCsLiUTgWLudiyamEmtQIZMK5MTqwlZ4IyJ5LFy1FtoPgkQKiaUQc0IvPxQK0drviqDt/7vHf0QiPx4p0Ro8vgCrj3TwzvZGChONfHewLQKVcqDZRn2vg5pOOxMyo7HaPfhDIXJidfxmRgaPfX+Uexbn8+SPlVw4MZWDQ8SNfqcPpy/IGaWJEWw0pUyCRacIV03xBgVZnoOgFG8sRe9REo0lvLm1AblUwg2zstha08N3h9q4fGo6hQkGWqwuSpKNhEIh3thSz61zs3llUx3lrWL/PDlKTbpZw4rxKexusPLZ3hN4+EkZ0cTolURrFTy1vASH14/xFGgCuVQgFAKLTsH1M7Posru5ZkYmq4904vD6+eOyUVS02xn0+FHJJdT2OJBJBGbmxWJQSbnxwwOkRmu4abKFefav6A3G01t8Deaj74LPxUBnI73y0Ty/vpozRidS2z0YhoX1OjxcOyOLV3+uZVSSkV31fRHb5N9+cZhnV5Ty2d5Id/f8oYFRbqwOo0bO1dMy+NumOvzBIAlG9Yh+/9aaXp48u5jRKSbSLRrKUqL48UgHKpkEyUkJMUavZHZ+HFf+/cSsZlZeDItHJWBQyShNMbGmIlJfxB8M0dDrYHeDlfoeB26fiNS5cXY2X+9voTQlCrc/wOGWfm6YmcWAx4/PH+C+Jfnc8vEBjnXYR+x+huPg221udjX08fAZhXy6u4UorYKpQwv0cQkIgBtmZbG+ohOpROC8cclcMtSSCIVCI9AxTm8AlUxKYYKBivYBaroGmVcQG7Y8NGsVXD4lnZ+OdbOlJsjLP9Xw4GkF9Dm8LHp+E25fkOIkA0+fW8q0bAvf3DiVmq7BIUil8V+2mitJMtBhc6FVSCOw6VkxWpKjRDjnweZ++hxeYvRKNld1c7htgHSzhkmZuUO/rfYfvMNQxI8S/zkp+hwePtndPDRsD3HNjEwumZR2ShJRq9XJttpeDrXYGJcWxcTMaOJPgX773xC/WiIXBEEK7AFaQ6HQab/Wcf/ZONzSz5qKTrrtHmbmWojRK0kagheeHB02N5kxOlYeauOCielc9/5e4g1KlpUlkx2rY0+DlTiDisbeSI3wt7bUc//SQv5wZhE/HesmRq/k9OJY9lQ2kGbWMCFBxpVp3SRWrQFdLMjVbNXM4/FhkMTHV1Xyl/NGIxBCJZWKMLoUEzd/tB/X0IX945FO/nr+KGp7XKgUCrJjddz68X4ePL0wjAU+Hjvq+7h9Xg7PrqsmJ07HX9ZUMa8glll5MRF45OMORVdPz+CeLw4TCIW4enoGggB3fXaIJ5YXE6URleke+PpI+O921vfx0gVlrBiXwhf7WvDK9FxaPZODrXYyzNn8/rRrGVf9PN3aHPCKSSndooloXSWaNNz16UG8gSCjU0y8dJIpRCgEdpeP0cnG8MJZlmpifmEcNV2DHGi2khyliahUtacQf4o3qChM1NM96KHN5qKmaxCTRoFSLqFzwBOhYX7G6ETe2hLZJtt4rJvzxqXwp7OKRf2Qkbwa8W9LE/nLmmNhUTKDSsaTy0v4y5oqNEopl0xKo7bbjkIqJSdWx5jUKN6/agKbqrpZMT6FNzbXsb2uj0Sjitvn50a0U7bW9DA3P5ZLJqcyJy+WOKOa51aUsq/JSlOfS9xBphhZMT4Vjz9IgkEVHkwKgsCkTHNE0gdw+QK8fcV4jnXY8QWCZFo0XDwpjYPN/dg9fm7/9CCZMVounZzGu9sb6bZ7+XBXU7gAOtw6wBOrjvKHM4ooSjSQG6f7L5EqvQ4PFa0DYR3/ggQDKrmUOKOapcUJJEWp+ePKo9R2ORiXHsWjZxTh9Aa48u+7Ixy+frcon7oeBw294twoxaymqcpJdqzuvzRTBhFxs7+5n+8PteH1BylLjQpj90FUBE2OUrPiJFmHfqeX+74qD2u/v7ejkXPHJvPImUX/kkb5/6v4Nc/oVuAo8I8FD/4vRHmrjfP+tgOX7wSF+P4l+Xy5r5UZOTERoj4gqro19Tq5eHI6z6yt4uwxyYxNNdHY5yQYCiGVQLPVyXljE1lV3hn+uzabm4aGai6Pb6I/ZQJddi/BkMAleUEulW9F37QBeb0B8pbA+kcIpE7jo8qRmtJrjnTQ0OPk0smpfLK7hdx4fTiJH48Pdjbx8owgDx/U4/YHGHD7T4l3hxNwuePojHVHu1hWmsSdC3Kxu/2UJhup6x5EIsAfV564iJ/68RiPnllEVecgb29p4IHT8uke9HDTnGwcHj+f72nB7hGhh3qVjKfPHc1TqyvD2iP1vU5u/qKaB5beTpxWSlfNACVJxojqV6OQIhFCXDszE5lEQCmTkmRSj9heW50+lhQncEZpElEa0Sj6870tbK7u4fdLC/n20InfsCBBT7fdw+QsM9uHKvvjpgJ7G63sbuiPoPPft6SAI202bpmbzU+V3VR32SlI0I9QVQSobB8gLVrDaaMT8PpC7BuyVgOxep5fEEdrv4s3Np9YBAbcft7e1kC8UcWWmh7u/vwQTy4v4f6vDqNRSrluRhYXTUqlpnOQZ9ZUctmUDC6dnE5N9yA5cVpm58Ww/mgXydEaFo+K55m1VeHn44xqkqI0JEVFMiN1yhDlbTZ+quxCKZNQlhpFXryeM0sT+X6YNktenI75hXHEGVTh6tnlDfDgNxUcarXh9gXw+IOUtw4wNz9uiKQjGiEMj41V3Xy6twV/MEQgGCI3Tkd2rJ50i0Y05B4W/U4vf/z+aMQw/+lzSzhnrAhrjNYpmZMfR2lKFAMuH2adAr1Kzo/lorWcIIgGKoNuPx/samRxcQKf721hT6OVDZXdvLShhgSjirevGB9ur9hcXgSEEdj2gy39rPjbdvzBEIUJhggV0uPx2Z4WzhmTjHQYRLSmazCcxMOv29vCFVMzTklw+5+OXyWRC4KQDCwF/gjc8Wsc81+JrTU94SQO4k1tc/m4YGIqZq3oDLS+shO9UsZNc7JJMqnosXsQgEfOKOKF9dVMzIiiNNlExZBC4ZojHeiUcp5YVsDn+zvY02hldqaO05UH6fAmYHX4mJMfQ1Ofi15lCocGZlGaN4cZyVLUe15FmjaD9uIbSKxWc3Lv16xVcrDZhlmnwuoc6eUIYnUltB8gWjuPJJNaTJASGJMaxb5hrL1MiwaHV7zpVMOqpOOLV1GinpIkI5trek9ZSaw83M4d83P4fG8rjb0uHl8lzhNidAqePm80VZ12FFJJOHkOF5ACcHgDtNq8CIKa93c08uIFpcgkEhYVxbG6opMHlhbw8LcVYTLQGaMTuHlONg98XR5GlEzPsVDdaWdDZRc3z8mmsdfH8+tFca6LzRr+vOZYhDbImaWJfH+4nQyLltvn5+IPBFHIJBzrGGBeQRwPf3c04hyfX1fFH5aNotnq4rxxyQiCSISZkWNh0zAVSYVUglQqwe72ovVbKc1K5s/Li/EEQgSDIeIMSj7e3TSk0hgZ1Z12lhQnsGVos7G/yUq0VkGX3cNTq4+REqVmaqyTQLFFNPrwBVg8Kp7ZQzunyVlmOgdEhFIwJCJhJKegjh+PvU1WLnx9RxhFYlTL+fiaSRQkGPjwNxOp7hxEIgjkxOlGtD9sTh8z82JIt4hWZoIg8NqmWmwuH5MyookzqEbo/OTE6qjvcbC5pofHlo3imTVV1PU4KE4y8MezisPa6ADHOuwjEFmPfFuBWatEo5BS2WGnocfBtBwL49Ki0avkHG0foKnPxfWzssiN1XOwpR+1QnR2kg0l2KwYLS1D11+7zc0HOxq5c0EePx0Tk7tMCrfMyWFWXmwYm76qvCP8WboHPaf0Ay1KNEQkcQD/KQTQgIjZxP+m+LUq8ueA38IIeepwCIJwDXANQGpq6i+97L8VJ9uV3TYvl1Xl7RxttyMIcPf8XK6YkkYQkAIf7GrihyHc8ey8GObkx/G7Lw7j8AaYlm0hP17PG5eO468baylvG2BmdjRfXJJF3uE/4/fEU5u6iEKVit99cRgE+M20DPIzU/l4fyvP73HxhzOeJEqroM/pZVFxkFVHOsLVtEEtIyVajUkj51BLP4WJBnLj9EglQgQG/opSPe3qqVTX2Uk3a7l5Tjb3fV3OuWOTyY/Xs7/ZSmGCgTn5cRxosXLznGxMGhn3Ls7nmbVV4rbbqOKcsSm8trmWsWlRuLwjL8JYvYpttb1cOyOTi9/ceYL9NzqRp1cfo7prkGWlSXQOuInSKFDKJBG4bBATYK/Di1wqUN5qo8vuIUav4t0rJvDXISGx4/HtwXbGpEbx6sVjOdjSj1wqwer0UJYSxfghiVqtUspjy4pQSKUopAJNfU6SotREaeRYnT5SozVMyTKP8Ct96pziiPc6Hg5vgGAwRH68Do1cJpo3uKWcNz4FlVzC+spu0s1aLpqYisXfzoK+L1Ad+5pAfQGasXdxxjde+hw+EowqfjM9E71q5G0zJcsSscBqFNII16Gttb20x2j505oT7aGVhztIidYwryCOt7c1oJJLmF8Yh0QQiDeoyIkTERyBYAgBwkM5nz/I65vqIqCANpePjce6KUgwkGBUh9Ecx+NYxwD7m/rxB0MkmdT8bVNdmCmaZFJz1bQMChMMTMyIZnttT0QfXaOQcsGEVB5fWclV0zK494vD4ZnQ4dYBrntvL1/fODXcZz7VTsfu8WN1ennw26pwMfD2tgZ+v7SACRnRnP/aDkwaBeeMTaayY4DSFBPPr6/GoJIzKsmAViHl4klpEcPnvY1Wttf2cvsnB8KP3fjhft6+Yjyz80TFxOGJt9vuIVavDEtjAIwdauF9tqcZo1oelgnOjNGOMHIfnx5FuvnXc/X5NeP/OJELgnAa0BUKhfYKgjDrl14XCoVeA14DkRD0f/q+w2NqtoXn1leHCQ7VnXaOttspTNBz5bQMvL4gh1ttjIrT0Ov2h5M4INqa/Xji4nD7AjT3OWnpdzE9N4bxGdG8srGWRqub302/ndd2WxmvCfHsuhNMrsdWVvLEWUXcvTAPm8vPp/taOdRiY2JGNGeWJvLQ6UXYnD48/gAh4OWfapmdF8MPh9tZWpxA14Cb+5cUsK/Rii8YYtnoBH6q6ebD3Z1Mz4khzqhiZ0MfeqWMv26sJdGoYmZuDNlxWrz+AEaVnL2NVqI0Mfxc1cXV0zPJjdMx6Pbx3aF2DrUMMCsvlpIkLSvL28P2YjqlqAGy7mgnTq8/Ynho0iio7hokSiMnzqDkb5vqSDNruG6mqGro8gWwOn0sH5PElpoerpqazisXjUEhk+Lw+DnSPsCWmh7KTzGjcHgD+PwBVpW34/IGuWpaBrd9ciD8/jfMyqI0xcjm6h4cngCXTk7j5Z9quHp6BhJBoNvuodvu4cIJqXyxTyTFXDAxlXi9im67C5VcEtGGKkw0UNfjQCmTkBunRymXcMvH+8OQwGtnZDIhIxqfy8HsipeQVf0AgNSxmeT2fTw8+UNuWeej3ebm24OtXDcziyunpvPBziY8/iATM6LJjdOF2zkGlYxorTLCxi3RpKLzFNv6bw+08e6V45FJRUbmtwfb8AdCzC2IhVCIrdXd1Pc4cfkCFCToKUszIRUktNlGIj+6hh0/EAxR2z1Im9WFWiHl6nf2hJOvUibh7oV5YT5Ea7+LBKNKHHr/UMmgx8/cglhuny+ireIMKh5fVYk3EEQmFUbAIdtsImfgeCLPsGhHLPijkgyEQqERO7pP9jTTMeBGo5RxyaQ0nlkrGmdLBLhhVjbrjnYyISOa2Xmx3PLx/ohjXjYlfYS5MsBX+1rCiXzJqHj+vq0hDPl8+aca3r58PANuPxIB3P4Al7y1K/z82DQTL104hgSjmlcuHstHu5rYWtPDnLxYlo9N/od67v+T8WtU5FOBMwRBWAKoAIMgCO+HQqGLf4Vj/1MxOsXEh1dP5K2t9UzKiMblCzIrPxaXx8/z66rJsGiZkRvDTZ+Vc+2MzPDfKWWSCMOBBKOKiZnmCFfz0hQTz64o5at9rbiVFpaW6nl7mFQpwKJR8TT3ezjW1c66o50nesg9Dio77Fw5NZ0nVh3l7oV5OD0BSpKNFCcZ0SllfHeojXPGJNNidRJnVLK3sZ+bPt7PnfPzCIYgSnCQN7CNOYrD3Do9k8+6U/iiTkqUVsETq47hC4SI0si5f0kBrTYX22r72Fbbx5PLi+l1+LhscjplKVHUdTtIjdLw4vll7G/uJxQSK7yXN9Zw0+xsqoYUFuP0Sm6ak41OJeOh0wtptbpYd1ScE2gVMsw6BfML48RFIMnI9wdbGZsmelke99ZcUBjHstJE3trawNj0qBEiUBadgsd/rGRpcSJ58Xp+9+WhcBKfkBGNUi7ljysr8QdCLCtNpCTZSMeAizc21+H0BnluRSlf7Gsl06Ll8inp+IMh9jVaKUsxUdc1wFNnF7Gtrp+PdzczISOaeQVxPDmkKX/RxFQU0hOJfm+jlb2NVlQyKbkWOd/FXENm4jkUHH4SRW8leB1kCa2ACRC1tdVyKYkmNS9cUEqP3cvhVhvZcTpunpONRackO1bHbcN0r9PNGpQyySnV93LidKREa5maHcPlb+8OP37XZ4d45eIy9jb0h79XQYA/LhtFfryeC8ancn9recSx5hWe0DdZV9HBTR/tx6wV2cfDk6/HH2R3Qx9FiYYwsqjf6cMfDIUF3tYf7QpDY++YnxO2ZDtVe04hlXCopZ9Mi5bkaNE8+83Lx/PA14dp6HGGf4OWU8AOlTIJHTY3Z5Ul8fJPNWGfz2AI/rqxhlvn5aKUSclPEBmof9/WQDAk+shOz7GMGOwCEa2k0tQoPrx6Iu9sa8DlDXLZ1DRKU01oFDK67W5Oe3FLBK5/b2M/P1V2kWTSkBmjQa+UkW7W8v1hUTLh7oV5I+YV/xvi/ziRh0Khe4F7AYYq8rv+XyZxEDHWEzPN5MTq+Hh3M8+uq8IXELeQD5xWwN6GPj7a1cS5Y1PCgu7LxySREq0hZpgk5dKSBN49KUkfaO6noddJXryeFqsTuURCcpQGo1rO9JwY/IEQsQYlD3xdzi1zs0dUHAea++kYcPP70wqp73GwvbaPaTlm0ixaTBp5WPdiX1N/BCxt0OMnJ0bDg4m7iF79JwDkwGVJY5m48FWWvn0C6WB1+nh2XTVPnD2Km+dkIwgQpZHz+KpKrpiSzjvbG4jVK1lzpJO8eD2nj07gu4NtePxBfjM9ky3VPVidPu5akENylJbffSn6bAoC3D43h3FpUbRYXZxZmsiD35xAtOiVMv560RgqO+38cRjbdU1FJ+PTo6loH+CWuTm0WEUEiVQicOXUdIxqOU19Ll75uZZb5+aEk6pEEFtdr2ysxaAW3VU+3dPCqGQj+XEGxqZGkxmjxe0NhPvbf9tUh0wi8OiZo7jj04PhWcn1MzP48DcTeO3neh5fdTR8s26q7mZmbgyCIO7G4g0qzDoFW2u7eWadmBQEQcmri55m4faLwG3DIxETg0kj5+4Fedz52UHsbj+ZMVqWj0nm56ouEk1qVh5u55WLx3D/l+VcPCkVk0ZBv9NLv8vHoCeA0+uhIEHP0XZRHlWrkHL5FJGK//3BSOglQJfdy1tbTwxVQyF49PsKrpqWSc2Qz+vHu5tRy6X8bomIk/77tgZSo9R0D4ooHb1KNoJTANA76A0PBuVSQexDCyM3ykqZhJm5seTE6QkEQkTrFFwyOZX3tp+ohK+Yms5X+1spSDCQHK1BEASmZVv44ropdNk9vPxTDUdabcwrjAsT547H+PRopmZb2NfUP6LSD4ZEzHdWjJYYvYp7Fudz/oRU/IEgadFatCoZF05IDV/LIGLsTx+dGD6GXCphcpaFSZnmMLHueLh8wXCLZXi029zc91U59y8t4KPdzeHXtFhdlKaYuHyYquHJ4fD4abO5UMulYWei/xfxvw9H898Mp9fP4VYbTw3DeLf2u3hpQw0FCQYuGLoA9jVaeej0ArbX9hEKhciO1TEpM5oddX2o5VIGvSNdzQPBIB/ubOLVS8ZQ3jbAtdMz+fZQG8+sraIwwUCCSbzRT8Yqg1hFWXRKPtzZxJ4hqct9TVbuWZTH/iYRMikR4PTRiWRYtGHSikIm4doSCdE7n4k4nrR1L/HuuhHv09rvorx1gBc31KCUSbhzfi5/u3gM/mCIYx0WVg6hbw632jh7TBIlySbKUk089M2RMJ73xtlZ3P9VeTixhkLwzLpq3rxsHP1OL98dikw2do+f8rYBNlWNlF2t7LBz14JcrE4f10zPpMvuJtGkZm1FB2NSo7h1yAc0w6Ih06KlrsdBQYKeKK2Cc8el0Dvo5cIJOhKMKm77+ED4RtUopPzprGJi9EoeW1aERBBIMIo6JcMH3q9vbqAo0RRWwzweHTY3UzItZFp0bKjs5FBLP6eXJJATZ2BnvTX8uR/c6qFs1JVEdW5nfZ8FEBPnn9ccY8moBKJ1Clr6XNjdPp5aPpoXNlTT0Oug1epid6OV3Y1WJmeaGZVk5KNdTVwzI5M3tzRwwYQUFhXFExhaWfxDFaj5FP6RgcAJqVizVkFpqon2fjfRmuN9YxkPnlZIVecAh1tsvLmlnlvn5fDJnmYae52sGJ/CgNNHYpSaH0/y3Dx9dCLvbG9gTGoUp5Uk8MbmOtIsWpaVJob9NwHuWZTPqCQjo1NM4ccSDCqSjGrsngAahZR1FZ3UdjtGXP9mnXLINzOLr/a3cf9Xh7ltfg7ba3tp6nOyrDSJZWVJGFQy5FIJn+xWRMw45FKBqdkWUs0iblwpk44YNJs0Mu5ZlEebzY2AgFmnOCXBRxAEpCc9HGdQsqw0KWIwK5UIYRXMlzfUcOvcbGp7nBxotlLeOsB3h9q5dHL6Kd+jtmuQR747wqbqHgxq8bc5rSThlBK+v3b8qu8QCoU2Aht/zWP+s7G/oS/Cgf14HGkT+8PPr6vm5YvKSInWEAgGWVqcQPegh72NVs4uS2ZGjgWTRsG8/FjWDmNbquQimcTjD2J3+5EIAr1ODx8M+TN22z2MHZqEH223MznTzPa6E2SX00sSiVLL2dNoZdGoeMammkgzaylvtbGmQkyuwRB8c6CNW+Zmo5RJiNErCYVCjEvWw9aRfVVp0MPJP12CURXeusokYoX14oaaoc8fw9+vGM++RitFiQbkMgmPfFdBUaKBq6aLGhVbqntw+wIjGKUAVqeXs8Yk8+fVVSOe6xl0k5+gj/jMMTol07Kjeei7CgZc/vBFvb22h5JkE/d8eSjcP9YqpDy7opSHvj0yZMJRRecwY4TfLsxDp5Th8Yvn5fQG2Fbbw876Pr4YEkq6ZW72iN/eHwzh8QdJiVZH7JLOKktm0Ovjz6uPhRP/iz/VclZZUkS13GX30Ju9nPK4M1DYddy1QECnlDEhPZrGPkdEb3ZZaSKz82JYVpaERafk8inprDvayfa6XtpsLq6bmcWMHAu76/siFPyitQpevUiUO100Kp53tzeGz0kuFciL06OWSzmrLAm1Qsr22l7SzBoyY3Tc/NH+cBvkxQtKuf+rcq6Yms6LG6rD321lh50zSxNxdTu4c34uX+wT4YM3zs5iUoYZu9uHVinjTyuP4guEKEoycO7YZBaNisfu9pNu1lKYaBhhS5YcrcHq8vG3YcPmdLPmFxX/orRKPtsrKo/+4fujlCQbybCIBKDjWPAZuTE8t6KUWz8+QK9DlPd9cnkxZcMWkFPFT8d6+MP3R9EqpISGro9l7Yk8fe7oMNrll0Ipk3LL3BwEAb4/1E6SSc2FE1N5Z3sDOqWMW+flsK6ikyPtA0zMMLOwKB6vL3jKJO7xB3huXVUYBTXg8nPXZ4dIN2v/n2iw/EdU5P5AEIcvwKkmqFkxOlqtLuwe0fThlY213Ls4n9s+ORi+EUwaOddMz+S+r8q5aU42MXolK8s7yDBrOaM0kZc21HBmWSK9Di9bqnsoSzWFj9896CFGryRWr2T1kQ4unJDKhMxo+p1ekk0a4o1KrC4fF09MRSGT0O/yc+hgG/UnkY0AGnqc/PGsUcTqVYRCIZ7a0cPT2aehqfn+xItUJgZ0mdy9UMUza8UBr36I6PPs2mo0Cil3D2l2HN9Sf7W/jfoeJ7fOzear/W2MS4vio99M4IOdzfxwuB0BkSBj0StJMEaKaEkEMTF32d2cPz6ZP64cRh+XCMToVLh8gQhz5utmZvLA1ycq/QGXnwe/OcItc7Kp63FGDAEd3gDrK7t45PQirC5vRBIH0R1nSXFChNhXv9OHQSUHxAStlElGyBqMS4tibUUHZ5UlMTAkRDYjx4LHH8TrD0VU7wDfH2rjN9Mzw4l8dm4sf9g6yLyCeD7bW8+EjGgEQTRo+MtJOiFfHxCF0+7/qpxb5mSzubqbWXmxeHwBNlf3YNEpuPzvu7l8SjqJJjXba3vJi9czNduCdggBI5dKuGlONn0OL8FQiDi9ik93N/HqxWV8faA9vHBUtA+wtbaHy6ak8/KQuXVN1yB2j1hkDP9uAX441M6Ns7MwqGXctTCPQDCETCIBQWB8ehRvb23gptnZZFi0rK/s4qI3RYeqvHg9L11QdkqJWblUwtXTMpiYHk3ngAeTRk5evP4XCToyiYBafgIae6jFxqEWG0uKEyJeNy0nhu9unkbngJtorYI083/N4GzudXDdzEzUcimBUAiFVMr22p6hwayYyEMhkQ/w0c5Gega9XDQplcmZFnQqGekWLblxep5dEcdPlV08vUY0Ub5uZibPr68OSyqsPtJBu83FM+eOPuV5dNs9YdvI4VHX4/j/E/k/G4IgoJRLae93cdnkNN7ZLt70BpWMiyel8sSqShRSCX0OL5dMTmP1kc4I155+p49mq5MEo4qXNtTw+NnFzM6LpbrDyhd7W7h+SjyNdgGDSo7H70ejkKGQSsKDmZc21PDYsiKUcik2lw+zVsG2mm621fRQlmaiqtNOYYKBYEjg6TXHyLBoKUs1hSn4xyPVLErTPrHqKNmxelYe7WfqzGtYoEsnpvF7PDHFdI2+gZcOBDnY0s4tc3Kw6BRkWLQ8vqqSqdkW8uJ1ePzBEX3RA839SASBRaPiWF/ZhdXpDdO5Q4jJqCDBwN0LRYu7nkEvKrmIbmjrd6JXyrAJot/g+qOdGFRyLpiQylOrK2m1ujhvXApLixNQySVEaeQ4vAFSozUsH5OELxhCIZWQaFKH2xfDo6nXyetb6piSZRnxnMsXQCmPrKzmFMRy35eHAXHQZlDJuX5WFltqesLytWeVJXHJm7v48UgnZq2C/HgDMXoFHTZPRFI5HhqFDJlEQBBEDfKLJqZy+du72dvYz/IxyYxLj+KDnY1cMOHU0NnjPfi/b2/gnDHJvLW1gVvn5vD0uSVc/e4e3L4gL6yvIStGx6y8GLJiRQPn40YFvYMeDjb3Y9LI2VHXS9PQLuKc8WIPeHgMuPwMLzZ31PUxvyCO44WiIIjfi8cv4uszLDr6XV5+98Xh8HUfo1Py/tUT+N3ifNr6XbRY3XwzrKVyrMPOG5vreOj0QjTKkZIPXn+Qrw+28e2QONeKcSncNi+XeONIqrtZp+TuhXncNgwmaNEpKD6FTkuiSf1PMTaPx/yiOO7+7FBYVVQpk/Dc+aURQ9lDLTYueG1H+H7dWNXNSxeWcVqJ2EvPidXx2uY6zhidSMeAm1ari9RozQhdnEMtNtz+U+PI9UoZ2bG6sAn18M/5/yL+IxK5VCKQpoNvB9w09Dq5b0k+BpWchl4Hz62rxhcI8sgZhQiChFWH2yPIDgqpBIlE3EpHaxVkxej4uaqb+q4BPhm1k8vNR9C0u9g99mlSHTtZGLUKtyOZRRct5IpVTiZnWihIMBAC3tnWyL4mKwqphN8uymPl4U4Kk4yY1GJ1sX5IAL++x8HyMckRDMf8eD0+f5BmqwuTRhFuFdz/8yCvRs9lTuoijEYTiU4jQayUpBj5+7Z6rE4fj5xRxDXT0znSbueF9TXcPiSSNDzkUgG3L8iNH+4Xh5d9I9tQW2t6mJJl5rFlo3B5A+iUMg40W2nxBfn7tgbuW1zAG5vqmJhpxuHxU9s1GLb8+nhIsfHexfnIh9QZL5iQwl/WVIW/79vmZXNaSTw/HYs0ApiYKUI8p+fEoJZLI6rlK6akkxKtJsOiRSIILB+TRFu/iwdPL2T90S5m58bw0e5mKtoGGJsWxZQsMwdbbJSlmPjjWUX8fKyHdLOGlGgtAgKf7Glmek4MGRbRPPl43DNFxyLJjxSvOI3XdnSy8lA7N8/J5m8/1/HhriaOtNk4e0wy9d2OEX9blGigYWiH5fYGw87p3x1sY0qWOQIKWds9SG33IG9cOpZJmWZarW6a+hzY3X7a+l0c67SztDiR+l4H5a02AsEQcqkEfzByBzG8H91idbFifApub4Cb5mQjlwgMegJEaUS39i/3taBTySKKl+5BD2srOrl2ZhYquZS1FZG/CcD2uj4aepwUDiVch8dPfY+DECF21Vv5dpg418e7mxmbHsW5Y1No6nPyU2UX22t7mJ4Tw8y8GOYXxvHeVRPYeKybBKOKGbkxZA5TOvzvRn2PM5zEQUTkfLiziSlZZroGPChkAtuGKvTh8defapiVG4PbF6TfJe7wjnXYmV8QR4gQqdEjB5VyqXDKIgBAp5Jz54JcbvrwBERyVl4Mo/4LDftfK/7tE3lLn5PN1T2kqFzhnumkGC+nm6vITp6CSp7OqEQDNreP+74sRxDg2hlZlLfauGZGJoFgiDSLFoNSRohQGEb3+mIDps3Pgc+Fd8xVpNn3ELvyKgAUgOHgmzxx9hbu+Kqa93c2IQhwVlkSZp2CtRWd/H1bA8+eX0rngJsd/aLh8pmlSeHzfn59FZdMSiMlWoM/EKSm28Frm+u4eU42h1psXDo5LUwwae5z8U6fizcuzcblC5Jh0eD1h/jNjEx21fdR3mrj/PHJPPqD2PaoaLdHqCHOzovl7DFJ9DhEudf6bgcz82IiTHsBMmJ0vLGlnp5BLy+cP5ojQ2y774Yqd6NGTlKUmg2VXSRHqcmM1XDbvBwaekS3mQyLhh/LO1g4Kp57l+Rz35eHIxbN59bV8M4V47hpdjaf7G4mEApx7thkKtoG8PjFxeLuhXnsqu+jsc/B7LxYUs0asixaHj69iM/2NvPyTzU4vAEWFsVx9bQMGnqdWIdaKsehhCDS5o+02Wjtd1HXM8jcgjjquoNUdQ5S2+3g7vnZ6CRe2nusTDS7GNP6Kq3py7nrG5HhuLvByu+XFvDW5ePoc3jx+IL8eKSd6bkxXDMji72Nfexv6mdsWhSJJnVYL+XM0kTWV4qzj1iDErvbj0wiRHwPKrkEt188l3e3NzC/MI5bhsEVX/m5llvn5XDB+BTe3FLLldPSeXmYw1RatJrAEBkoJ07HA0sKsDp85MXr+WRPc1ivBuCh0wvRq+S0nQL6V9U5yHvbG7C5/BQnj0w4o5ONdNndFGKkvd/Fn1Yd5buD7UzOMo8g4QGsr+hkXkEc93xxKCyd8OORThYUxvHMeaOZnhPD9JyR3qv/J3EqbH5zn5N9jVbu/6ocXyDI+afYRUkEUWTs79vqI9y7kqPUfHj1RAxqOQuL4lh95IRExy1zc0j7BUJQbfcgD31Tzg2zsvAHQ8ikEnrsI8/t/1b8Wydym9PLVwdaidEpsKLjlrnZSCUCs8w2Yn5+lYTBh9BMf4lK5zjq+9zhlfJYp53Hlxfz6HcV3Dg7m2fWVNHaLxJJ7lqQy7crYsnd8TvwuUAdRUfGWaQOdxxRGXFlLeHVra00D1XOoZAo+n/XgjzWVoj638FQCL1KHkaiRGkVxOiUdA968AVCfLy7mcfPLqa6axC720e0RoE/IGJ52/rdXDwxlc/3tSCTSLhpThYg4anVR2nuc2FQy7huRhYzcmLoGfTw1f62sJ3Y6iMdnDM2mTvm55Jh0fDFvlZu/mg/AHMLYpmVF0ucUUVylDpc+aebNcTqldw6RkG83IdZZuO+zS0RVZxSJiE/3sCConi6BjxsquphV72VPocXrVLKp3vE/nZevB6ZIIRt9YZH96CX3Dgdj55ZhFwq8Oza6rBLUp/DS1u/kyunpiERBNptHu796jD+YJB7F+czryAOo1qORiFFP0SCOtxq4/SSRF4bJiUrCOIN+dqmOq6ZkUnPoAe3V3TkOY6QeWJ1NflxGv40S0+22s/60LU8usGJzXWiz97a7+KyKekcbrVx0Rs7cXoD1HY7OH98Kn2Doo77+LQoVpZ3kGnRMTsvBqc3QFXnIAqphAWF8Ty1+igPnV7IQ98eIRgS+8X3Li6gx+5BJhHQKqVhXfHhsfJQOxdOSOH88en0DHr48zkl7KzvJS1aS0GCHqc3wONnjyI1WsuNH+yl3+Xn9vm5EUkcRLf7Ny4bx6FmG/tP0pzPjtXx1f5W5hXGs6/JysUTU3h/p7izEs2xDWiGeuTbanv47qC4oDf1OpmeYwkvmsdjXHo0dd2D4SR+PNZUdFLX44ig8f9S2FxeDjbbKG+1kWrWUJpiQiWX0tDjQCGTkGXRhecKABMzonnxpGPMyY/lvq8Oc+nkdJ788RhquXQEQemmOdn0Ony8dpK/QIvVRXWXuPA/emYRy8ck09TnJDdOT0my8RcHqM19TtpsngiDbYBzx6WcUlnx145/60Re1TVIZccAyfmxDLj8vLKxFl8gxIsSgafmvsiZh2+kcMOVHJ2+klDIEt6276wTp//TsmP4YGdTuL1RkmzCFwjxQ5uWg4UvUjbdh1II4ldFiy7vUjn1s/9KtXYMSrWOHR8eGHFODo8fqURgTKqJTpsbpVzKbfNyqOka5K8/1XDVtAwUQ9K3iSY1j6+qpLpzkKJEAy9eUMZj34vmFl8faGV0soG3Lh/PwaZ+ojVyHv2+IozAGHD5+cvaKh4+o4hEk5r3dzZy5dQMnhu6kD7f20KGWcs5Y5MjCDnrj3Zx/cws3tpcx9ljkxmdbKTP4aPP4WaxdA+ZB+8GzwDo4vh00fOc96MMp9fPTbOz8fkCTIseIEFopMtg4oMqKYtGxfPaprqIhD8+PYpgMESiURWx7U02qemxe3hiSH1uUVE8D5xewHvbGtnTaOWWudl8vb+NN7Y0AKKuzPkTUnhjcz0PfVvB7fNyyI7RUtM1yFtbG3jx/FIQRDjcLXOy+WxvCyaNnDsX5PH1/haumZHJX4cqeBAHt/ctKeDxVSI5qDglmiNuI2s7tKREaxid0hXh/zg5y4xMKiHeoOKPy0ahkkvpsLnIiNGx9mgHG471UNU5yGklCeiUNqI0Im77nkV5GFRyXthQjUQQKEwQvTA9/iDtNjef7G5idn4cJo2cbw+0s2LCSLPlaK2C3kEvvQ5xmF6SbGJKZjRbanuHTKaVGFRyXt9US//QgPNUbvcDLh+qIeu5Syen8fneFmRSgQsnpLKvycrh1gEyLDpquuxcPDGNx88yMugJYFDLqOkeDItSba87sXtr7XeRYFJHIIKyY0WvUd8vaJScymVq0O3jcKuNFquLOIOK4iQDn+xuiWBaP3ZmEe/uaAybhp9dlsQ9i/OIGyJXyaUSHjytgNc21WN3+1hWloTDG6Dd5qHf6UMpk/DW1nr+cu5o9jRa6R30sHxsMuPTo+mye37R/QogzqBmQdE/16+POYW1m14pw/QLBtW/dvzbJnKfP8iHOxvJidWjlEn53Zfl4YvIHwzxuw0DFM+5lZzNt5Fn8PL1ISu3zsvhxfXVlKaYcHr8JEerw+JSMTolkzKjIyQu08wa7l6QhxIJaZNup0MSy5f2Uby5pobSFBGHveWkakqrlJEXp+e6mVm8tqku3L4YkxrFpZPT+evGWspSTNy9II/L/r4rfM5H2gZ45PsjXDlVhAOq5VKitHLsLh9GjRyLTkVTX6Q7SyAYwusP8PGuZgZcfnbU9fKX80azuaqbRJOa2XmxETfF8ajqtHPWmCQSTWp8gRCCAKcn2sn8+iZxwQIY7KRw6618euG3bGiVMjk9ikLHdtR7rwWvA+Rqxkx7mvbEPNr6Xaw83I5GIdKsv9rfyo2zsnjsrFE88FU5Ro2cs8ck0WJ10TPo5a4Fefzt51rWVHSwfEwSWbE6YvVKmvpcYaw9iHj7GbkW5hfGsbaik03VPTy2rJCyVBMTMqN5dVNdWKY4Vq/k4TMKUcukBEJB0sxaXN5AhOZ1MISIkDmjkANN/XTY3Hy254Rr0xVT0smP11PbPchFE9No6XNyrGOAdpubd7Y30jPoYdGoeDQKGbF6FfcsymNUopHL3t5FMES4BQWi2l+X3cO9iwu4/G2RHl+WYhL7yONSeGZNFZdPTcfu8WNQyTFp5OHhmkSACyeksu5oJ8+sFRfma2Zk0jXgJiVaw9H2ATZUdmHRKblqWgZtNjdVnYNIJZIRleeZpYnsqO3h8VWVlCQZePSMIo60D/DV/tYwQmhPQx93z0pgRuAntHU/4okfQzB9KUtL8sJGFGPTovh8mMPVi+urefTMIqxOH75AkK4BD81WF1OyzJQkGyPkoydnRZNxkoa4LxBkc00PFW0DyKQSvj7QylllSTy77gQiKCdWx64GaziJA3y5v5W5BbEsLVHj8Ph5anUlObF6zh6ThMcfZP3RzrD1oUwqIRgK0TPoxeMP8vAZReHj1HTZaR7C2h+HEoOYB3LjRvbu2/tFUptMKpAdqx+RuHNiddw5PzeMaJJKBP50dnEYA/9/O/5tE7nV5WVTVQ8XTtQQgoiL16xVUJxsZECnYde8L/ikWo1eKceklvHBpcV8e6SXomQTexqt4YHj0pIE3t3eGPEejb1O+l0+OjvdZOYu5phNyovfihor2+t6uX9JAbXdjjBc77xxyVh0IlljR11vRA96X5OVslQTk7PMzMmPpd3mHFG9HOsYRC2XkhKtYUtND0c7/Kyr6OTsMckEg6GIm/14xBlUVA/B/nbU9bF8TDLTcyz8eU0VFp2SMammEVvg3Hg9mWYtn+xpFttRgwJSeysEAzizluI0ZBLdsBKJtZZAfxuFCUXoPa2ov7tOTOKANzoffyCIXurhwokpzCuIQ6uU0t7vpigplb9urGVpSQLPnDeaZquLuz8/oU1jUMl48pwSylttDLh99Dk8HOsQb5KTY3dDH3lxolOMQiZhQ2UP7+9oJBgKccWUdBYWxfPn1cfosnt47Puj3Dg7i7e2NnLfknzWVnSOON6Ay4dGIWN+URzXvrcv4rn3djTy2iVj2dNoFeccXYPcv6SAJ3+sDPe439hcz5VTxXbLqvIObpmbPcLgAsRC44XzR/PC+lrsHj+XTk7DrFUgl0qQSQSyYnW4fUERS/5zLdfOzMLtDQzJrepp7HNG9NV1Shk7ehy4/cGwmFWX3cMTP1Zy1wLR4u797Y08euYoPt3dRH2vkzn5sVw1NYNL3toJiLOTln4Xb29tiDjX4mQj891r0f38IACq2lVw7DO49BtQisiO6dkWpmeb2TxUuKRbtHTZPeEdIIjDPbNOyZPLi/lyXyt7G/spTjYSrREt6IbL3R5s7ueRbyvCsrKLRsXjC4Qi7uOCBNGE4vZ5OfiCIZQyCRuPdXO41cbSEtHAunfQy9H2Nm6anc1fN54owkwaOQqpgC8QYnSykQkZJ1QPD7fYuOiNHQy4/Zw3LoUbZ2Wxo76PshQT54xNRimX8uPh9jA7Ns6g5M7PDoXbkKOTjbxwQVkEPFKtkHHltAym51joHvSQbNKQfYoF4f9W/NsmcqNKzti0KJQyCYNuf9hx5JJJaShkEnbU9fJ5i4l0iw5PyEZ+gp62fjdxegMrj3TT0Odm4ah4ChMM/GllJRqFdIQG8/HYUNlFvCEFp/dE/zQUgj+vPsZFE1MZnx6Nze3ji70tfLqnhTGpYl/v5KhsH2BcWhTPra3ilbPTyIvTc0ZpIi5fAIVUQnu/i2STmq/2t/L+jkYUMgkXTkhlT4OVa2ak8/DpRdz12cHwDX7NjMyww4xJI+eehfkc67Szt8FKh82Nw+unNMVEVoyO2m4x2RclGggFQySYVJg0cn7z7l5y4nT8fsF4rAu/5s/7oL7Rx9n5i7i48DASnZnGPicyZwO5XvEYvfkX8br8Qj7a7OQWhY+nVx8JI00unJA6VNGm8sOhdqZlB/l0WCUH4iDyQHM/7+9oxOkNcFZZElFaBSlRanbURQ5g8+MNfLGvhfkFsUzIMEcsCE/8eIzHlo3ijrk5PLO+mn6XD7NOSa/DQ1WnnUmZ0SSY1PgDQSSCwIc7m1hQGMcDX5dz85yRyB5/MMSBln7+ulEcfkkE6Hf5Rki6frm/ldNLEnlvRyPWQS9ZFi21PSd4ATE6JW02N+6eANVd4next9Ea1jUBuG1eDl/ubeHmOdmsPNQeRlFcMDEVu9vH94damZBxAo7pDQSZkmXmraG20/DrMBQKYVTLuWF2Fj8caiXNouXsMckIhIgxKMiJ1dEz2Ic/GKKt3830HAubh4griUYV109JQPfBI5FfRk8VdB0Fg5jIk6M1vHjhGGq7B4fmOD5+8+7e8MuzYrRhnW5RXKxtyLxFTIhHWgd4/oIy1Aopbl+AV36ujdAG/7G8g3n5sRHSwh0DLk4fncgj31WEf4OLJqYyISMam9PLgMvPbxfmceNH+1l9pIPfLc6nutNOhkXLlGwL/Q4f7105gbx4fbhPHQqF+HBXU3h+8+meZgwqGY+cUchZY1Jo63dx04f7whou2TFaJmaaIwhnB1tsbKru4ZKTqm2tUkbpv2BE/WvGv20iV8qlXDMjk83V3Wyu7ua3i/JYe6STnsETwPwjbQNYdApunZvD74c0Qo5DA59ZW0UwBOePT+blC8vwBoKcPy6F94Zts1RyCTFaORdOTEVA7IMdN9YFcRewt9FKKBQiwaQJ//j1PQ7OHpM8wvV7dKqod/7IGYWM6f+KCyacHtbOnp0Xy9RsM89vqCFaq+CeRfm8vLGGt7Y28OTyYo6226npHuSmOdkEgiEUQ/ZlA24/d8zPweby8/a2eq6dkcWmqm4WFMaRYdYikwjcMjcbty+AVCKI+uIyCU+vruJASz8A5a0D7Glz88J6Z7gi+tseD31l4ykzmPjhUAvnZRpBrgG/m72xy3l1jY2LJqbyxub6CLjgh7uauHNBLu9sb0SrkHKgxYbDM9Jcw+sPhivwr/a3ct+SfHoGvWE3dxBJPYIg4vw7Bjxsqu5iVm4MVqc3PNT74XA7549PIT9ex9RsC61WF8vHJJMareHLfa1hyKdSJuHpc0v4sbyDW8drKY0WSSd9ESQiEw3DEnIwBIpT7BKi1HJ0SnGh/nxfC8+fX8YHO5vY22hldIqRSyalhfuzALEGVUQSB3hnWwPLypL4w/cVzMqL5YOrJ1DVOciza6swaRTcOT+f1UfauXNBLh5/kNxYHT8d6yYpSh02jTgeFp2Sly8s47r394VnFV/ua2VRUTxnjUnhzgV5XPzmTty+IJ/uaeaqaelcPzOLEJAZoyXB2wSn6BUTiuy5mzQKxqaJ5JZBt493rpjArgaxkhUEgQ6bG61CRm23g84BT7h1kx+vR62UYPd4USvU2Jw+djf0jXi7qs5B7lmUR2aMqCS5qCiep4fBV0E0XFlYFM+K13ZQ2WFnWraZNy8dx6M/VLDycDu/W5zPuLSoX3QwEiUrIn+LAbefTdU9nDUmhfJWW4QQV0q0doQxOcCh5n6YlHbK9/ifiH/bRA6QG6tlV30fOfEGXtxQw7MrSrnsrV0Rr+kZ9EboN3gDQd7Z3sDSkgT2NFjZXNNLtkXLREMPxfrdxE0p4dNKP1lRMi6dloXTB49+VxFOcOPTo1gxPoVPdjejUUi5YmoGPxxqJ9agDvcGrU4fOqWMSRnR7Bhqr0zIiGJiRjRpZg2hYIA+fR7P/yAy88xaBQUJ+rCsKIhb6d/MyOTZtVWUt4r6zB/sbGZ4yCQCfz63hL/9XBfuBf959TH+sKyIz/e2cO+Xh7l9fg7PratmwO1HLhW4bV4O6WYtL5xkt+bxB0bojH95sIspOfFcNzMLly9A1bIfyN16O33KRO6YL0MlFzWi1x/tjHDSEeUMfMTolRxo7uecsckRGjhSiai3PZyFmBqlIcmkZmq2GQkCzVYnfQ4vgiBQkmxkSpYZAdF1Pkav5PenFfDqxjpidEpe3FDNnXMyiDPpuPPzw9T3OMhcNiqcxI+f03Pranh2sZmS1ZfCYSvvzH2DF45Fs79VRCmcOToxDNkcfq4nD21vmZdLhlnNlCwz+5pt9Dm8nD46geIkI/ubrVz/wT5kEoG3LhvPgsLYEUNIQRCll6dkmcUBoT/IznprhBfs1poenjqnhDs+PQjAivHJLC2OZ1JmNPd9VR7Wri9LNeHyBdjV0BcxcAZYXdFBa7+LcenRfHPjVKo6B9ErZRQkGiLNJvzpUHox7H/3xGPG1AjD4pNDp5IzPVeUWL7+/b1h3e6ZuRYunCgmOKVMwu3zczncYuNou52Pd7WwrCyJOIOS6dkxES5OAKWpRoqSTPiDMCvXQrPVdUpt890NfWHizZaaXnodXt6/aiImtRyd6h8PF+VSCeeNS4m4XgEWjRJZpn0nSVQcabOxeFT8iIV4avZI8tr/ZPxbJ3Kry8+LG2q4fEoaj5xRRLvNhUwiGQH+P1nLqrnPxZ3zLQgIrCrvINmkInuqkfxDr3ITfi7KXIDa08tW3x95cUNjRILb3WDlnLHJjEuLIkanwOMPsqwskcoOOxMzorl4Uhrt/S4G3H7OHZfMrPxY3L4A1Z2D3P9VOeeNS+GZtVU8fvYorE5RhnRBUVzEIAlE9cNAMMgNs7KI0irw+EdWtf5gCIkgRAz0Ogbc+AIhVh/p5OJJabz6c114G+kLhHh6TRUvXVA2QrNbKhkJq9KrZCRFqbnvq3Jqugax6BQ8f+7HrN3ezPrKE0iYG2Zl0TngCaN/lDIJK8an8NzaauweP6nRGm6Zm826ii5MajmLi+N5fZhVWoxOSXmbjZd+quU30zM41GKLmC/ctSAXuVQSsdBtrenhtnk5SCUSvj3Yhrq/msZgXrhabR+2bZ+WbWF8ehT+YAinwsino15ltKyZzKYveTB3LtL5k3l4fTcXvrGTv18xDrU8l/I2G7lxevxBEe0hk0rod/rIidMRCoU41jmI1x9EIRVo63eSFy9qkri8QUIh8bt+6acaXr14LLVdg0gFgWAIBtw+ojRyNlR28Zt394qDzYlpCCcJTPiDIcpbbZi1Cq6flcVPx7q4/ZODLCqK4+ULy6jrdhCjV6JVyLjhw32nJIGp5aIxx9H2AV7fXMfuhj4WFMaPNEeQKWHWPZAwGso/I5Q2FXf+2XgUcUPiveD1B+gd9KJVysKqicFgiI92NkWYL/xc1cOEDDPnj09Gr1Lw5uZ6ugfFyvyZtVUc6xjg6XNLuWVuDpUdA9R2OxAEuGpqBpMyzbT3u7j63T1cOCGVzgF3BEQWhnSETtI6sbv9ONx+YrQKdjf0samqG51SxvQcC4WnIOTMKYjltnk5vPpzLXKJhNvm5zAxQ9xpZMdG9rXjjSpm5MTQYnWxvrILiQAXT0pjcpZ5xHH/J+PfOpF7A0GWlsRj1il5dl01oxIMXDEllb9tbgi/JitGO2KVXTE2iV31vXy6R6xw+xxervjcztcLHmLUxquIKv87A4UXUtHjDTP2hkdrv4uv97Vy18J8Hl91lHabG4VUwpXT0nF6/Dy7rpoYvRKvPz5CIwRE9b14g4o1R7rCqou/FAaVnHe2NdBsdXHNjMwRrYDcOF3Yvk4pk3C4xca+Jmu4ijFp5HSdJNMZCkGbzcVlk9P526YT2Gu9UkpBvJ6jwyjGd8zP5clVx8IaKj2DXva2DEYkcRDbBBdPSuOd7Q3cOCsbjUKK1xckWqfA7vHj8PrDdnGjEo0kR2nIjtXSbfcwOsXIwqJ4nhpCC+lU8hFEpdc313PRxEhSh8cfRCMXXXhunZuDNtmAxC9QnGTkcKstzMA7rSQBfzAUxvd+vFvJzXOy2BlK4eH+NA5V2JmZ3cnppSmUpkXRbvMglwrU9zjQKWWMTjESb1BT3mZDIoEOm4v9TTZ+HGaCffv8HB5fVUlrv4vTSxJJjlLz7cE2WvtdtPQ7uefLQ+FkNy3bTGq0lgPNYmsoGIL3d4jDWYlAxOBUJpWIuu5b6sM7gs/2tbKzwcr0HAvPr6/m6XNHY1TLMajljEo0UD6scvzN9Ewq2mxsrunly6F24IDbx/eH26nrdjAmLYoZuTEii9GYDBOupj5jBXsa+2mvdBEMNVCWYiIpSsOrP9fyw6F2cuN03Le0gIkZZly+AFtrI3cwQNjkeXSKKZzEj8cPhzu4Za4oC/3JNZNp7HOiUUjJsGhRyaUcaO6n2+5BEEQI7j2L8nlraz3NfS6MajmPnFHEX9ae2LncODubvkEPt3x8gImZ0cTolfx1Yy2BYIgXN8j47LrJxOqVHO0YwOb0kWHRkhdv4Na5OZw7NhmJRIhwUxqVZOS5FaU8+n0FMqnAkuIErnp3DzNyLNw6N4dorYKFRXEj7PP+p+PfOpEnmzQUJRh5ZAh73TXg5t7F+Tx5dhG76q0UxipQqbX0OnwY1DIGXH5Kko2cMy6Fi96IbMH4gyGq/DGMGvp/qd9FffcgM3NjRzjXaxQypufG8PLG6jBixRsI8urPdTx/fikg2medahGo6xkkOUrNxqou/nBmESq5lDVHOrloUiovrD/R7tApZWRYtGHC0Qc7GrlrYR6rDndwuNXGxIwopuWI7MxDLTa21/WytDiBP541im67F5lEwOb0EatXRiRzQQCXN8j+5n7uWpAnmiNLxJ7f4lHx3D4vh5Z+J5lmLSFBYG9TJOLlVJ6FDm+AokQD9y0u4FjnAN8dbGdBUTzLxyTTPeCmom2AHw618/CZRfzph6O02dxMzjTz8BmFmDQKrnt/b7hF6z/F8Y/r14x43O3n2XVVhELwrkbOMytGc2ZpIueNF1E+t87NRimTRrR1uu0e/AF4YUNVGAG0sqKHBquHstQovL4gU7LNJEdr8A7hvh/5bh8KqYTx6VFcODGVx1edOF5OrA6FVMLlU9JF/1Kri0STCplE4PzxKWyu6omoWGP0KjZVj5T97bZ7iNKcEP7SKWUY1XLcvkBEWwegqc+JeWg3KBHgkslpvLOtgbPLkjh3XAqt/S7SojWsLG/n+fXVnDsumcIEAwa1DLvbz9NrRIjcl/tbGZcexd8uHotZp6Sue5AbPthHZYe4cKdGa4jVKXlraz0/V4kJ+2CLjUvf3MWX10/B4fExNdsSAQ8ESLdo2F3fd8pd3vCK2qJXYjkJxhellqNVSPEFQmgUMp78sZJlpUlYSpXkx+uYlWNhR10vH/c1s6Q4nu21PeE2ybFOO+lmDecPQQoHPX42VXdT1WEPz7VkEoE3LhvHrLzYUxpEqORSlpUlMSkzmj6Hj/Nf3w7Apuqe8BA2w6Il3vjP68H8v4h/60TuCwT5YEgVTioRuH1+Lo//WIlrSLBJITFQmKxFKoEbZ2Vh9wRINKr5uaqHKK18hNKeWjl0UQkCakM0MxOiaLB6cXj8bKnpwaiWc9u8HPqdXhaPij/lwGbA5Wd+QSy9g15mDEMHHI/SFBMf7GgiFIIDzTam51j4zfQMdAopadEavj/UjkWnJCtWR/Mw3LjDG+Dzvc0sLIqnOFmsOo+3Gu5emMf2ul5xQDQqjlc21nLP4nw+2d3EtTMzeX5Yj/zuBXk09jmx6BRkxmh5clUlTVYnN87OpsPu5ofDHZQkG8mN1bOr0RpeAI+HIDCiLTM2LYrvD7WHZXkFARYWxbHhqCjOVZxkJC9Bzwc7msJJaXtdL9vrevnzOSURczaZRBiBhZ6RYyHeqIpwtInSyPEHg+G/tTp9/FjeSZxeyQs/1BClkfPsitEjkgyIVenJMM6Kdjs3zMrG4w9woLkfmVSCUS3jz6uPcXpJAukWLXsarLT1n0iqBrWMs8ck8dTqY+HzGJMaRUGCnqunZ1CcbOSDHZFWZHXdIvnrZNndzBgd187MpKnPSZxexbQci8gWPAW1XhBAKghoFFL6HF5eGpp3PLOuGplE4IULSvn5WA+5cXq21vTy1b5WrpuVhUwi8Nw6USHzssnpyGUCUkGgqdeJWadk7dHOcBIHccGo7h6MWIjgxJD/iR8ruW5mJqUppvBA8NyxyUzLttBh87Cv0Up+vD5CSOo3MzJI/Qe+l6lmDY8vL+G+Lw9zw6wsGnqdtFpFl6FJmRZ0aiW/XZTH0pIEnB4/174fCSFt6HVyZumJxaHf4ePnYZr0/mCIB74u5+sbpo5YRIZHvFGN0xsYoSYJYHeP7Nv/T8e/dSKXSUT3cBCdZb7e3xr+4mu7HdR2O3g81cJ3B9spSNBTkmyiZ9DNuzsauHZGVsRwKSdOR6/EzN5F35AQpeO7ZhUZShW5cUr0ShkrJqRg0Sqwuby8sbmeV3+u46yyJObmx/LKMF1mmURgTkEcm4e8Fp89bzTvbGugMNFISrSIWbd7/CRHqVlWaEDXvgNpdxSb+y1ERVu4dkYmXXYPWqUEQZAgH8LCApSlRPHW1oaIJCQRxJ708UGrxx+ktd/FC+uqOX10IjE6JfcvKaCl34VEEPhsbwt6lYwlxQm8tKGa+4acZV79uZb9Q9v9Y5129jdbuWN+HvcszOf335SHt/wub4C/nDual3+qoabLwczcGC6YkEKX3YNGISXOoCQ7Ro9KLmHA7SMEBIFxadG8NqyVczy67R6un5kZ/g4/2NnESxeW8exa0ahhZm4MuXF6bvxwP3+7ZCxrjnSGFR8f+a4i4lgNPQ4WFMTyp7NG8aeVleyus5KXMFIjO3qoui9KNDAnPxZBgIq2AexuX9jmL86g5PZ5uaSZNURpFbw4lCzHZ0SFGcJnlCTy9taGiIVoX5OVC8an0COBvQ0id2DNMDz7wRYbN8zK5kBzf7iQmJlrQa8UF41XLhpDjF7B7gYrXQMeihIMnD8hhY+H6ZifVZbEz1U9LC2OHzFb8QdDrDnSxb4mK+PTo8iOFaGnWoUUty9IvEHFI2cWcbiln6/3t9HU52R9ZSd/u2QcB04aAAIcbLaRG6sbsfCoFVIumZSGzSWipgZcfnQqGcVJBqq7HHy6pxlBgKunZTI7L5Z2m4t5BXFMyjQj/wc64YIgsHhUPDmxovz07PxYMmM0KGUnUlW0Vsn0nBgqThpAnjjGiX+Xppp45efaiOdbrC4cHn84kXt8ASo77DT2OrDolBQmGjBpFCQY1SwojIv4/ZQyyYg++v+G+DXMl1OAd4E4REXU10Kh0PP/p8f9Z6LT7uGmGalc+b6VDItuRPUL0NLvorLDTmWHnZ+revjz8mKunJqBRiHC0ao7B9GpZJQmG+kc8LC+J5ZJ0RZ2tjTy5s5yZuXFUJRo5KFvyrl2RhZddg/LxyYhFQQOt9pQyCTcOjeb59fXcOmkNJQygTs+O4F1VsgkTMo08/m+FhIMKq6dmUVunA4pEG07SuHOOxhMmMRg0W+pDQTZ19TPzvpezh2XwtOrK7hvSQHfHWync8BNYYKBqs5Bdg3tBEYlGTi9JJFd9X0kmtQsK00kNUrc8tk9fj7c1USvw0OL1XXKqXtjr5OmPhep0epwEj8eNV0OjnXYkUoEfr+0gF6nj4J4PZ/vbeHnY93cuSCX6q5BNh7rZktND8lRapQyKbsb+gkhYNTIWH+0i8wYHeWtAzx8WgHj06PC29PjEaWRkxWj5aULyzjWbifWoMTm8pJm1rBifDK+Ie2ZGTkW2vtdlCQbeWVjLXctzCM7VkdF+0AYwTE+I5pjXXZkEgmPnVWEXiln1eF2VoxP4bM9zQRDkGnRYtEpuHtBLo19rnA/dVq2OUIb5oIJqfQ6vCwsiue9YUSxD3Y08dtFeXy0qwmjRjGiBwxim62my0GsXolaoeaM0Yl8f6iNEDC/II4Om4v3rprAwWYb7TYXFe127v3yMHctzCVWp2BLbR+PfX8CKXXXglyeWzE6TM0fcPtJMKqJ1StG7CpBHFI7vX5WHu7g4klpZMVoSTdrUUgFJEIKdw9xEc4Zm0y/08e3B9vYVtvLnILYEZraEzOjyLLo2DBM5mFqlpn9TVY+HFpc3tpaz72LC3j602NcPyuLzCEWZygEr2+uQ6OQYtEpuWVuzj+sgo+HXCqhIMFAQYLhH74uwyJKUAxfzKZkioYZE9KjuHlODlFaOYIQia6ckxdDjOHEeawq74iQ2F0xLoX7lhRg1Mi5d0k+URo53x5sJytGywOnFY5wKfrfEL9GRe4H7gyFQvsEQdADewVBWBsKhSr+qz/8Pw2HO4BeKeGNS8fR1u9ieo6FbbW9Ybf1UEgUcDreHuhzeGmzuXlhfTXBEEPVSSHt/W5CwNOrj/HHs4u549ODYcjip3taGJ0sqgmmRGtotjoxqOX4AkFm5cVidXgwahT86exiarvs/H3YTZ9p0eL0BnhnewOXTBKZfVVddsakRrGvycoVuxR8NvNpvrem8dRXfYCYoK+dkcm3B1pp6HXy2A9HmZ5tIcMSjVmnZElJPBXtA7h9Ac4YncSfVp5Acqyt6OTNy8bxp2Wj6LR7RCMLp5eLJqayrqKLn6u7wxe0RiHlroV5/PWnWi6dfGo8rCDAKxtruWJqOh/tauKssiR+Grqhr/9gH3fMz8PtDzAxI5ofDrXTNejhQLOVfU1WrGOTefiMQtYe7WJ0ipEuh5frZmZR2+2gtd+FIIi6GRkxWtHy7kgn+5v7mZMfS0G8npxYHRsqu8Pb4lm5MaSZtRzrHODKaRnsqu8l3qDikslpuL0BVHIJvQ4viSY1D3xVjt3jRy2XctOcbPY19nHznBxkEoGp2WZ6Bj1Y9Er+vOYEHXxLTS85sXoePL2Qv6w+NlTZtnP9rMwI1FOvw8vjKyt5bsVoNAoZCwvj+HGYQp5UImBUy5mTH0tzn5M4g4o1Rzq5cbbopbqjro8EkxqPLxRBbgJ4+adaxqRG8fqmuojW0tNrqnj9krHkxOp4+LsKBj1+nJ4AvmCQh04rZFttbxhrrVfKSDCq8PqDLCiMpTTFgDcQ5O7PD/L42SURok7vbm/kxtnZaBVSKtsHuGpaBuePT+GTPc2EQiLbclmpSH3/60VjUMokBIIh3P4At3x0IHycUAi+OdDKzLwYXtxQzTtXToj4XE5vAJNGhlmnxBcIcqRtgJouO0a1glGJBhL+Bf3x4XEc1fPQ6YXsa7SSatbi9vpp7nOhUchINWtEj4ELyvh8TwtLSxJEOGmUms1VPfxwuI2kKM0QL0CGRiHCaf3BIJuquxifbibDouOxZcXcNi83ArHzvy1+DfPldqB96L/tgiAcBZKA/+uJvKrLzh2fHuTmOdmMSdZxw9QEZuTE8My6E4MsnVLGzXOyeXyVqDnS6/CG2wQdA25e2ViHRafgjS31PH3eaHoGvfQ5vGGTYH8whFImJdGopLZ7kLpuB+8P63s+eFohFe0DxOiUNPY6EQS4rNTAGLOXqGgzD27o5vkVpexpsPLqz3W4fGLSeWBpIeeOT6FRn8pfVkZiul/bXMetc3NYU9FFIBgK+05OTZLT2hvkiqnpJJnUYT3o4xEIhlh5uJ1ttb20WF1IBHjmvNHc++VhXL4gV0/PQC2ToFHK6Hf6eGJVJR5/kB31vZxWkhA2mgCYlBlNZYcdly+ARCJw9fRMPtjRSJxBSbfdQzAk9srPLk2iz+GjySqqR967pICPdzXz5f5W4owq1h/tYmqWmbx4A7d9sp+FRQmkRmtIMKl4a0sD7+1o5LQh/W2A7bW9TMky4/EHI3qbG6u6GZ8RzehkI5/uaSYrRke8UUV5q404g5InfqzlyeXF3D+UxEE0pXhhfbU4JxiSmVUp8pEKAo29kbo1x99jbJqJ62dlEwyFqGgf4IOdTVw5NT1C6lSvkqGQSWm1ubhyWgaCILD6SAdJUWrunJ9Hx4A7PL+YVxDDtTMzqet2oJJLObs0kfpuB85TeMN22T009TlHaOoA9Dm9TM22cOf8XN7YUs+RtgEyLVrUCgn3LM6nw+ZGLhUYnWxiW20PF05MZVd9H+uOdpEXb+DaGVmsq+hgYVEcOXF6ugY8fH+oje21vYxNi2J2XiwHmvuZXxjH2WOS0KtkxBlUfLK7OTwD0CqkPHJmUVgSYnjY3X60ChnBkHjP/eHMIh774Wi4ajaq5RzrGKDf6eO69/eG78GyVBO3zMkhGApRlGQk/p9Eg7RYnfzmvT30O33cuSCXQ6021h3tCpPTlDIJAqCQSZmVF4MvIDI61XIp03MsEVBWg1rGNTMy0CnlPPljZXgRHZ1s5OWLxpAcpflvLzb/r+JX7ZELgpAOlAE7T/HcNcA1AKmpp3ZZ+VeixeoM90hzNP8fe28Z39aZ/ft+xWBJtmVmZscUZmZumzIzTWHKOOV2yjzl6ZQxbZM0bZKmSRpmcBwzM0i2mKX7YjtKFGfO/5zPmXP+M/fe9c6ytb2lvfd61rPWD2xMaP4HqsZf2Jz+fkgP2eryUtVlJitGg1hECDkIBOecW2ZmU9VlxmB102N2ctXkDLbW9gWHPBKxiE+uHkuPxTWCrfnqb3VcMSmdWK2C2h74cK6MiI13IK05gT17KffOeoy3/2jCFwhw66xsIsU2Wvst/F7dy7S8GBwQbA2cjECAs1L8Mzz1jFV18b51Msfah/Cfxdzu9C2kPwBv/N7I6DQ9a4528fyvtTyxrIhv9neQHadhRq7gVBMIwJScqKDfaFpUGGaHh0/3tJKmVzM5M4qqbhMrK5LpMztJjVJzrMOEViHD7PLwwA+Vwf+5t9nI/QvyeXGj8PDfOD1T0AlpMnDZBOF76hhy4OwXnH/aDQ6SI1VcNDaFvw0vdF3D7bAzY1fjALlxGnrNLtYMS6omR6q4bmoGM/JiqO2xjpDOFWCKwm2eHqVGLpHw4sZarpiUPuL4xUk6ZufHAdBmtHHe6CS+O9hJhFrOHXNyONYxRE6sliiNnFu/PITT4yc9Ss11UzNYWByPzx9AKRXz5b42Xr2gFK1CKqhffn2EAasbkUiABCZFqAiTS0ds+eN1Snx+PxMz9IxOj2BsuqD7fdIMWCwSMa8wnvwELU63D7cvwM2fH8JgcxOuknHLjCxEw3/3zvDM4VDbEJHqfl5YVUJ2rIa/bWtkQ1UvyZEq7pqXS2OflZl5guNSy/ACopCK+fza8bQa7CEicja3jxc31PHoksIRUMlFo+L5en87ieFKEiNUlCRHBHfIDw7PHVaNSWZXgyHkfYfbhjjWaeKVTXVMyNTz2oXl/1PQvlaDPfic210+7G5fCMP4+mmZQVTK3mZjsHVy4dgU3t8eOqsxO4RW1T92t4TshI52mDjWbiJOK9gZ/rtW4/AvTOQikUgDfA/cEQgERkwhAoHAe8B7AGPGjDm71uX/QthcPkwOD3lxWqaJjqDa/xbEl9AyOFLMvdfs5PbZ2WiUUq75x4GQ341KCqehz8JF41K5+7ujzC2Ioyw1ImRS7/MH+HxPa9Bk+fQwO72kR6lp6rdy79Rootaeg8jYBCIR+1Ov5uZvTz0I1d21vL8onIs7b6Gz7A76YuZgEylHQAQj1TIyosOCGh5RGjkvLM3EYW1nKCyS65IcfNKsIzdOG4JDl4hF5MZr+fa0nmGv2cn03FMsNKPdzWUTUmkcsDE7P5ZXNtXRNGDD5fWRER3GuaOTeHJtNU0DNvLitFw6IRWFTMyaI90c6zzVR79yUjpSSYD1x0LZeYEA1PdZuWpyOhIRbKrpCyJHfj0uqB3W9lrw+QM8sDCfxn4bT/1cTVKEihfOK+GZ9dX0W1yMSg4fsWhWpEbSMegIwZl3DDo41DZEgk6Jzx8Iau6cDJlERKxOwZzCWM4tT+a2rw4LfXenl9Lk8CDVP0aroCwlklu+OMS8wjhWlCcRH64iLSoMl9dPUoSSgvhkGvttPPdLTTAZtRjsDNo9vL+9mRaDnWdWFvHAogJe2VRHVZeZeJ2SR5cW8ex6gW/w3h9N3DUvl41V3Ty8qIDnfq3B4wugUwnyt1tq+rlrQQ5H2szc8NlBnB7BQPqxpUWseGsnr1xQJpiDDNjY2TDA1VPScXj8SMUittb28fCSohCTChAQPe1GB1/uawtei45BB8//Wsun14yjecAWTOIgLH6vba4/q61dj9lJq9HOQ4sL+PlYN0N2D0tLE+kecvCnWTkkRaoYGBabEolEPH1a5RujUZzVCMI//GXuaTJS1WUOJnKPz/9PB6MRqlO97492NvOnWTl0DAoon6WliczIiwmaRv982j0qFYtGFE4g6KSczby9Y8jOs+urcfv9TMiIojQlgpSzuAf9d8f/2Gb6fzJEIpEMIYl/HggEVv8rjvlfhT5MxlMrioVtYMMa4cW+EyxNHQkNmpUfS4vBRuuAjSsnpgd7ntEaOedUJPFbdR9SsYgrJqWTGaMhKUKFThW6xu1sNFISKwvqZ5yMiZl6osIUtBodiM2dQhIHiC3kx9aR2OdPa0X4NQkkbbmDsL6D3Pr5If6yrJDCYXRFbpyG22bn8MD3ldw7P5d1i9y8tTyVe9Y0cfFaG/ftErHXkcSkzGgMRgNvXlDE7IJYlpcl8vbFFfxjV0vI/5tfFBccMC4vS2R/i5E7vjnKG783cMNnB1lamohataaC3wAAm3pJREFULuGjnS2oZBJ+PNzJ2Aw9t87KFgwg1p2grs8aksQBvtjbhhjxCD9NECQHRqdGkhChGgH/+/FIFzPzYum3uDjYOsRLG+soSNCROMzge2FVCTPzY5GKxSFyonlxWmbmxwTJSafH8U4TFqeXXY39/GVZYfAaySQibp6RzRu/N3BOeTK1vZYgAujjXS1kxWq4c04OL59fyoqyJJ7/VUjQv1b18lt1H0N2N29vaeSjHc30WdwcbjfRbXLy5sUVPLqkkEvGp6KSSajrtQSt15Ijwnjul5rgcLnH7OTub45y++xTzMtwlYz6Phtf7Wvn6+sn8s6lFTy0qIAv97cRq1Pi8gR46ufqIMSz3ejghV9rOX9MMj8c7uD1zfUsfn0Hz/5SQ7/FTVO/lVd/q2dXkxGHx8tZ4NvIJOIR18Ll9WN2ekPMtk9Gi8F21jZHql6F0ebi2fU16JQyrpiYxkXjkllamsgrm+q44dODLHljB+//0cSg3ROyqG6vH2BOYVzI8cQiAXd+59xcVDIJFoeb2h4zT/98gvPf2c3fdzafFYKZFRvGLTOygp/jxY21JIQreWRJAavGpBCjPXXu2tNo+xtO9HJuRXLIsdRyCZnRapYNe3ieHh5fgI92tfDZnjbu/OYIe5oMONwjWdb/3fGvQK2IgA+B6kAg8PL//in9z4XR6uKVTXXIpWJWFo4ili3g9zKx8+88OeMGXt0vyICuGpPM4fYhfq/u44nlReTFa3ntgjIi1DJ6TE5e2lRHcZKWpEgV721vYsgukE9un53Dm783BE2MLyxSU7HjBj5Y+Vee2GqkccDGrPxYLhyXyi/HBR/Q3V0+MuUacFvBbUd/lgF9tDKA2CY8UHFdm4lQn8MdXx3h/cvHcLB1kFaDnWfX15AYoaSqx0K+Xs09v3VjsLmDBhVP/VxNbqyW22emYehuISM6AoPVzQsba7l0Qhpf7G2jz+JkVUUyGpWMhj6BDJETqwkx2D1ZzawoS6Tb7EKrlNLUb6dyOGknhiu5YlI6kjM1DhCQGeFqGReNTWFvszHYIlDKxMxLF3NwwD7CqR4EgwGRSMAbf3+og7vm5fL2lkb6rS4UUjF/npuLVinh7a0NrBqdwsLiBEQigU1b22M5q5filOxoBu1ulpUloVXIuG5aJiBgrX8aHhoP2QWS1OlwztWHBHepq6dkjNhu/17TR5pezZj0SGEAub2JqDA5K8qTuOfbo9jcPhLDldy7QCBVDdo8DNoFI4gzFxv3sPoiCANki9PLTTOy2Fbbj8vr4/G1J4LJdF/zILU9FqbnxoTMCGp6Lfxpdg5isSg4sJSKRXy8q4W75+VisnuYVRDL6oMdXDQ2lb+ftqAnhCvRKqQjdisgVLZjz+LyfuGYVIoSdTy7chSPra3C5fUTo1Hw8vllpOjVzC2MI0whDeq+P/hjZXA2EQjAS5vqmJgVRXZsGA19wu62stPEjNwYtKOT+fFIJwnhKi6dkMbrmxswOTxcPy2TtKgwrvj7PnpMwg71cPsQ1d1mnlxRHPRCBWjos1KSHMHbl5RjsnuJ0ynIjAkjPXokNHB5WSJf7G3D7fPTb3HRYrDxxPIi1h3rJkajoCw1AoVMwjVTM7C4PPx0pItwlWDovb7yFIrH4wuwo2GAsemRZ/0//53xr2itTAYuAypFItGR4dceDAQC6/8Fx/6ncaLbgsHmRiyCmuh5xGi/Q2TpIrzhJy61tjD50o/5pNLB+mPddJmcPLaskNc2n2Jijk2PZGJmFItGJVCQoOX+7ytxeHwUJ+mIUMl5748mrpyUzt+2NbK0JIFlOW4kP+xiavdcvsm/ENPyG/CJFfzRZWN8ZhSbq/t447CXqROfIvWPO2GwmWWjTXx27BS5RSYRcUmmAzYJPWW3JonRaXq8fgPV3WbyhmFNt83OISNajVwipl9dQXXvPq6clIbN5WVrbT9Wl5d9LUZu+MLMx5cU0d7pYFeDgeVlibi8Pu6Zn0u70YFGIUEtF7bsarl0xG4ChJnBpKxofjjcQV2vhaWlCVR2mpiUFUVpcgSf7mnlsglpRKplwUUNBNx+55CTqg4jn5+bwJZmG2opzI4yEhhq4Zn1Xq6ZkkG8ThkiVzorP5b9LYNcOj4Vt8/PRztPaXG4vH6e/aWGj64cQ6RaHpRQAEiJVHHR2GSGbC6WliSwrrKbQEAgWC0sTmBzdQ9P/1zNy6tK2VjVE1J9yiQCDTsvTkOEuognh6F9armEO+fkIhaPXKhS9Wp8/kCQaTpk93D15AxeGVbNBOgyOfloZzNXTEwjLUqNze1Fo5CgVUiDSe1khCkkKGVibpiWxU9HOsmN05AepWZPk3FERby1rp/bZ+eEJPLkSBVSMRxoGWROQSxlKZFYXR7CFFKcXh/58VrcXj9tgw7CFFLunJNDVZeZomFp2Rc31nLdtEy+2d9OXryOhn4LY9MikYpFZMZrefn8Up5ZX43F6eXyiemsrEhCIZNw/tgUxmXoGbS7SYpQkRChonPQTteQk+YBK2UpkSSEK4NOQafHwVYjL68q45GfjnO0w0RShIqCRB3lqeFkxoRhc3lxe30sKUnAaHMTq1VgsLmDSfxkfHewg+umZZITKzwfuxoGuOPrI/RZXMglYq6ZkkFGtPqfJteylAi+vXEim070YHf7iA9X8dmeVvRhcur7LOxqHGDxqAQSI1Q8d24Jt8/ORSEV89wv1SOUD50eHzLJ2ZUV/zvjX4Fa2QGMfBL+D4fT62d2fiwryhOpMzuJXvgtmb4mpG4TUlMrXYZBojXhrBqbQn6slt9qekMemP0tgywpSWTtsW6iNXI0Cil3z89lT5MRg9XFhWNTyYnVctmEdLbU9lGqEVE0XG1HdGwlUhMGhz5GU3YbH/RMoywlnln5MbRqizAuyEJmacEvkfPOxUUcaDejFPuZpmxi1J47hBNQ6+mOncbR6iEmZ0eTH6/lRLeZHw51cvnENPRhcqq6zDi9Cj66YgzfHGiny+TkysnpNPVbWV/Zg9XlZWebnfHpkVidPl7eJNDVRSK4dWY2KpmYe78/xtfXjWP1oU6m5cchOaNHOD03hvo+K1qVnNd/b2BJSSLXTc1EHyYLDro6hxw8tLiAX4/30NBnZWJWFHMK4mjut/LVwS5WRHl5qOMh8LngaB/fT/4J8PLp7lbumJNDfZ9VWCRK4onXqWg12hGJRIxJ1/PhGfraAH1mF/ctyOeT3S1UdZkZmxbJFZPSWX2gndvijqEtm8SM3CjEYgnx4UpsLg+FieHcPDMbg83NReNS+dvWRqYmS7ki3Ugq3Ug80BDIYX3lAE8sK6J9yIHfH+DVzfXcuyA/RKdEIRWzvCyRxHAlLq8/uAtw+/wjTCTajQ6GHAJm//bZOSSq3Nw8MyuE7XnR2BTC5CKunpzB94c66Bh0sKw0UZCUnZxBepQ66GoDQruhNDmcO+fm4vP7kUvE5MVreW9bI7ML46nvtfDixlOzl/NGJ2NzeflwZwu3zc5m7dEuChMEj8msGA1N/VbaBwVI3uKSBHY1GpiTH8fYjEiu++QA5amR3DUvj/W3TcXj8xMfrgr2lyXDJhgno9/i4o6vj4RIvf55bi7Tc6ODNP6T4fYF0CilfHLNeLpNDg40G3jkx0pmF8TjcHsJV8n4cn8TPn+A5EgVs/Njz7qoikWi4K7QaHPz6E/HidLI0SoFydy/bWskPlxJQaIuxLziZIhEIkpTIihNicBodXH3d0dDFvrXLywjcRiVopBKSB/GwS8pTeKno6EzoPlF8cSH/3vprMB/KLPT7vaSEyUnckwya4920W1yoqpIplE1mh3NA4yJLqB9QMnrw44hC4vjz0rVbui1cu/8PCRiQfjo8dPkag+1DXHP/Fx+ONzBgNVNe3Yivuy5EJGGRK6CP14Avxd9w/cUjlvJX34VaPBpUWqunJTOc1ttuLxu4DDlKeHcPCOLCKKxTbwbsUzOLkcKt/1oxuHx0dBnpWXAxmUTUllRnkRSpIpbvjiM0ebm7nm5HG4bRK9RMNht5s3fG7huambIgLR9yMl725uCiSMQgHe3NfHuZRXcPS+PLrOb/a0mVAoZDyzM5+NdLXQOOZiWE8PS0gScHh9f7xeq3x8Od1KcpGPKaTKdWbEa7v72GKXJ4RQnhbO9foBNJ3p56+Jy/jQrG1+CmhbRXSTVfYrM2keMwk9ShIrMmDA+3tWCVilldn4MsToVd313LLiQvLiqhDidYgSpxeTw8MKGWmYXxDEjL5Y+sxOz08NvtQaWJGcx4duxjI3KwpS+iHcbZ/HOIRuvXlhOjEZBcqSKR36s4uEFGczt+QjVljeDx80vuYKK+Cv4+kB7UJ9Do5DSOehgbLqeSyak4fL4hjVPXDQN2Pj+UAe3zcpmSpb+rDuaqDA5VqcXf0DYNchxMz1JTNYlFfSYXcTrpOSprCz+5FhQZnZpSQJRYXImZUVj9/i4dmoGdref7w92UNtr4bGlRbzxe0PQLFkhFfPmxeU8sWIUJoeHv55h3/fdwQ6eO3cU2+r6WXOkS5ixrK4M3ssTs6L44aZJPPNLdTABV3WZ2d4wwKyCOL7e387xLhPf3zSJpEg1Xp+f+l4BAZQSqQoxD67qCtXrBnhrSwMfXzWOE90W+oer5DvmZGN1+TDYXJgcHqq7zTi9Af6ytAiby4tMKubu04hzHYMOPtzZzPPnlJAdo6Gh/9TzetXkdFKGESgGq4vzx6ZyoMWIUibhwnGpfLSjGbPTQ1O/jfLU0ERucXrw+gJEDrN59RoFz51TQlWXGaPNTWZMGIX/hHg0KSuKv11SwYc7moc1bdIZnxGJRCzC7w/QYrDRZ3ERq1WQER2G6CwtyP9b8R+ZyDsNNpR4uOubKmxuH6vGJNNncdFvdVGQoCM6Ss37P1dz0bgUItVyBm0uZufHsGhUPGKRCJFIsPxKiVTTb3Egl0npt7hG6HF/uruN2cM3ellGPDtkzzH98J+h9ufg39SPupu7fmoIVmqtBjuf7m5lSUlCUKjncLuJpgE7G3qVSMQVTMmJ4U/fHw75X3ubjVw7NYMfj3RicsYEt5rp0WFsqe2nqsvCzPwYlMMDyXlFcdT0WGgesOMLBEZM4t0+P51DTv6+qwXZsA7Nb9V9fLW/ncUlCcRqlRxoMfLTkS7y40OZavW9VuYXxgd/PimUdbTDFOLS3m9x88bvDXyrU/L48gXsVs0gdbochUzO3IJejnSYWFAcj1YhZWJWFDd9dijkPB9fc4KnVhbzwOpK7G4fIpGgibOhqgevP8CGqh42CH4gTMqKosfsxGJxC4YHA/WED7zGeVNS+Ls0hWd+rua+BXn0mZ1cPSWdWFc7qgNvhXwu1bF/MG7GErwZKbQbHVw9JZ0Bqxuv309SpIqXNtZidngRiQSq+yvnlzK/KJ7vD3UyMSuK9Cg1V01KD/af5RIxN0zP4u2tDSwrTcTm9nH96nZ8/gB/mhzHPGk1fYNS3jsRz70L8nhpYx3pUWpUcklQCgCE6jsrVsNVU9JxugXc/umO9y6vnzd+b+CNC8uxurxn9YAwWN3cNjsHh8vLB9ubQ+7l3Y0G2gbtIxJwQ5+VxaMEHW6hVWJDp5Tx1f42nvtFQNMkhit557LRlCRH4PX5zzp4dHn9aJQSbpuVQ6/FSX6cln0tRg62DuIcXhhf+U1oSanlEu6Yk4vU4x9xnD1NRgIieO/y0fxe08exjiFm5ccxKTsKmVRMIBCgacAWQoL75Xg3d8/LI1anYGttH+XDDj1Oj5ft9QO8sqkOq8vHtVMyWFyaQFSYglid8n/K2T5MIWXhqARm5sciQjCzAQFl82tVD3/+5ghOjx+VTMKrF5Qxryjuvy2Z/0cm8iRPMxtbLNjcPtKi1GRGhWHz+Ph0Vyt2t5fFoxJ5eGkR+5qNeHx+eswu0qM1DNrdQXxtml7NHXNzsDq9PPTDcW4anoCfHjKpIOB055wc1h3tYuGoBNyaRE5f89u9OvxnOKk0DdhYVJIQ/HlqTjQSsYjtDQP0ml1Mz4vlzjk5eP0BbMNUepfXj8nhoTgxnMbh4dC98/Np7rcxKSsKiVjE53vamJgVRVGSlnEZevyBAF/ua+fJ5UVoFNIQY4FwlYw+s5PlZYn0mJxsOtHLlOxoNlT1hBgOP7m8KKgjfjJcXj+xOgXZsRoa+qwopGLEIihKDGdiVhQ9Jic9JgdOj5B8r5uWwRNrq+kccnDB2BSOtg8FceBH2oeYlBVFRWrEiL6xxeXF7fHx7MpR+AkwaHMjk4hHiP6LRdBrcaKUickQ9eILT6Om7CGavNFo1GqunhyJ2SkkYJfXL2i5y5xndb3J1MFmk58HFuWHVK0KqZiHFxdQ1SXMKvITtBxsGeTzva34A8LQrThRxyUT0njjojLajQ6SIpSC3dzyIrx+Qmje96xt5v4Fpby8qR63b4AH46O5aXomBQk6rvp4f8g5He0wMSs/lrVHurhrbg77W0M/PwgokkG7m5RIFSmRqqAqJgjDzF6zk0/3tPLpNeN4f0fziPf3mp0jcOsgzA9OhlImFgaL604lyi6Tk4d/rOTTayYwYHXRNeQkXCULMXyYnR8LAaFaLojX8dHOJg4Of4aqLjO5cRrOHZ3Mtwc6sLt9HO80MTFz5IB1dFoELo+P2l4rEWo5103NpDAxnOpuMxuP9yKTiPj+cGfIezy+AN0mh4DzHkanNPVbOdI+FDTlAHh0TRVun59lJQnE/s8oF/ZVw4m10HMUZeFyyJwBslhAcAC78+sjwXvH4fFx5zdH+Pm2qSNMpv9vxX9kIpf1HEKCgEyYUxAHInjzd8HZfmqO0BKwOjwcbhsK6iVvqe1nbHokC4vj+eV4D63DynL7mo34/AFkYtEIpb/rp2Yil4qp7bWwvqqHKK2C2IRzGKf6DhxCdROtGFlZRIXJUUrFaJXC0EkfpsDi9LCsNIles4ONVT38OIweidEquHd+PlWdQ6RHqZmaE41YJAg7bTrRw4ZhwR6FVMx9C/J5aWMt71xawfEuMxMyoihOCidiWJXx3T+a6Le4iNMpuH5aZhANcO/8PDac6KHX7OCBhfkc7zQxKjmcgngtH+xoZtGoBFL1KtKjBK3mziE7yZEqHl1SiNnhQSWT8MnV4/h8bxv/2NVCcqSaO+fkUN1lYkp2NBureoOLQUK4kq/3h5J5djUauHVmFgnhypA5hUwiotPk5G9bG7l9dg7Pb6hlQqaeS8enIpGIUcskbK7pZUyaniNtQ7xx/igiW3azc+ybPLbHx4LiBMRWEQUJWt78vSGo/a6Uick5pwhfZBaSwdMEk7QJbO3XMCYtgj8aBkKqVoGiLWVHwwBf7ReYozdMy+TpFcW0Ge1IJWKkEjEQoKbbQpRGQUKEitcvLOeT3S04vSPvg611/ZSlRnCsY4i8OC07GgbO2mcHAc3j8PjoGHIRox3Z551fGM8z609w7/x8rpiUzo76AQ60DjIqOZzZ+bG8MNyTb+qzMa8wLnh/nQy1XMLMvFh+P801aVx6JMmRaooSdSRGqDjQMhgUFDsZYXIJcwri+fZAO7FaJX/f2cyf5+Wyp9HAiW4Lk7KiSNGr2NEwwJqjXVwxKT2YxE9GXa+VhcWnCpt+q4AcuXxCGp8MX7NojZx75+dz3ScHaRo2B5GKRbx3+Wju/PoI10/LYlfjQBBzfnp4fAE2HO/h1QvLMFhdPL6mivizMDG/PdBBhFrGOeXJZ+3FB2OwFT47F8zDi0bNOph2L8y4H8QSei3OEbt3u9tHn8X5/yfy/5UQ+TwUBRrQh6UhFgn9tQvGpmB3eXlri0B3f2pF8QjR+/0tg9wxJycoDNQ16CBuWMTn/e3N3DQji64hBw63j7x4LeuOdbO32UhOrIbrp2ayv2UQp1uNZO635NKK2w+q6DIunWAM0vZlEhH3LcznRJeJ9y6t4Kn1pzDFMVoFTy0vCpHe7Le4ONAyyOKSeB5be4KWATuz82M5tyKJqz4+RV5yef18vreVu+bmUtlp5p0/GhGLRFwwJoVuiYgwhZR75uUhEgkV3iub6oMV+tpjXdwzL58+i4MwuZTOIQcvb6ojUi3n6skZfLO/nfsW5PP53lY+2d3C7IJYBu0eHl97gn6LiwXFcRAQBc0UWgw27vzmCLfPzqEgQcfHp7my/7OtpccX4Ibpmby3rYkuk1DV3TxDQHA8tLiAzdXCghWjUZATp+W1zfXYXF4uHJvCjLwYuoacfH2oG2POJfzWYGJBsY4PdzTjH2bM1vSeWjycHj+f7OskdvpHFFX+FXXHdhwJ4zmScytP/WoiL655hL75heNSefSn40F4XmO/lVd+q+OJZUW8vKk+iKy5cGwKK8uTWHusC4VUTGO/lcPtQ2d1jIlUy3F4vNwzP4/2QTuBAHy1vz3IoD0ZCeFKsmI1yKUSHvjhGC+cW8JjSwt5+bc6LE4v8wrjyI7V8O3BDuqG4ae3zMxidFoEu5uMIXRzkQgWjkpg0C7It+pUUm6blYNULKYiNYLxGXoOtw+RERWG3ePjjq+P8OoFZWyp7eOZ9TW8cn4pMVoFy8sSUcok5MZpeGpdNX0WFwuK48mN07K70UCMRsG4DBkHWgf59mAHIpEw9Iw8i248QJTm1OuTs6JRy8WMz4xi1ZhkDDY3fSYnuxoNwSQOQnvrtd/qmZEXQ/dw0XXX3DwOtJ5qEUnEIqblxnDV5HSyY7UcbDFyuGOIVWcRtopQy/j5WDeTsqKDw82zRu/xU0n8ZOx6DcouBn0GsVrlCKlllUxCrPa/bwj6H5nI3fEVZFZ+yo4l4TjcA5h1ufzcpePFTadEkM6mCndmTI82EaaQ8e0hgcr/woZapudEUZQUznO/1gS3ofV9VmYXxDI7P4ZItZzP6wdZUDyFyk4Teqebul4rf56bi9vnRyYW0WG0kxurYU+zMUR1sN/iYnNN3whdk4IELX/+5mjwxvjpaBdmp5dJWVHsajSgkIqpSI3E4/ejkEt44rQH990/mnhqRTFKmRidQsLBdhOfneFKpA+T88H2RmYXxFHZKeiHgOB2/vT6av52SQWPrz0RhAmmR4Vx33fHgkktO1bLW1samJUfS3lqBDaXF7VcSqxWwYGWQUqTw9k//HA19lkZmx4Z0o+dXxRHuEpGWqSaG6ZnYrS5kUkk5MRq+MvSQj7d3UqqPoxbZ0WQqldx73enKP//2N2KSCRifWU30/NiQKpiZbmO6m4zt8zMpmXAxoAlVHYBBC3tb+rV5CQ+hC7BysZmN7+vFZJ9VZeZx5YVhSj6iUSMwFh3m5y0GOwhCodf7W9nfIYerUJKjEbOF/uMdAw6yIrRhOC0FVIxi0sS2FbXz4sb6rhjTg4fDLc8HltaSH6Clu31A+THC+gSlUzCrsYBSpMj2FIriIWtGp2CSi4hRiPn78OLpVYpPLIdRgezC2PxBgTz5TVHu0iPUhOtUXDDpwd5/twSxqRHEqWWM+jwYHU5KUuJ4KHVlSjlUv6o68c+fK5NA1Z+HS5uTnSbuX1WDs9vqMHsFGz6bpqRxXO/1LChqoc75+QSq1Vw/+pT1wiElo3L68fq9DAjL4atp3234zL0aBRS4nQKLh6bSk6chpc21vLEumoSw5XcNieHt7Y0ML84njOjy+SkOFFHZJgcj0/oTT+9opi6XisGm4vLJqQxJl0fRNlIJWLMDgERc3oLSCIWMa8ojve2NQVNv/9p/FMjauH1zOgwXjq/lLuGn1mlTMzLF5SOtND7vxj/kYl8UByJcrAF9aF/oAaipEoWL/yM104jezT1WylPjeDwaf3W8lTBJV0lk3Dz1BSMBHCIFbxzSRrb6vpRyQVBnWfX1464ln1mF+kZwoOSn6Dj+k8PUp4SQW68ln3NRvadRhvPjdPw5zm5HD5tMHgyKjtN3DhdGOhNyopmTHokYXIpf5qVw46G/iDlfkttH3+alY1ELGJchp7t9QIRYe1RYcs8MTOK0emRDNnc7G400Gq0ERUm5/rJKXx74JQlnEwiYnZ+HE/+fIJVY1NC7N1ORk2PJUQsXywO9QG1OL1MzooiRqvgpdMUA88bncSVk9OQisX8+Zuj9FlcrDnaxQvnlTC/KJ69TQZy4rSYHB7e3tqAQiZh7WlwLplExHuXjaEiTc8T604Qq1OwojRpxPltqe1jQmYUv1R2MyU7mtu+PBxU+5ueG8O03OgRlnpzC+MIV0qJ0Kn5udLGtsZTFbvXHyAQCPDk8mI+3dMCQFnSyApOIRU01c+MziEH9X1WxqTruXF6FvW9Fg60GLlxRhZ2l48whYTCRB33fVdJv9WFSiYhLSqMO4Z9NWN1Cj7e2UJego76XivrjnWTFR1GSUoEmTFhwe/4w+HEr5AKWOndjQZGJUXw13NH8fX+9qACYUVqBG9eXI7N6eXZ9TWEq2RU95hRy6V8d6gjOHNYWpLAtdMyeHxtdcjniQpTBAfa2bEaHlhdGWz/tBntfLK7laWliXx3sIOXN9Xx6JKCEbISIAx/LU4vF4xJoSI1kqouM4WJOooTdSRHqLhrbg5xOhWVnSZEw4jlLpOTx9ec4OopGWgVI9PRqtHJrK/sYnZBPKPTIrityEVxz99YZWslMOoCFPEZIBYxYHHROGBFLBLx57k5vLutkeunZQrsWFGAVH0Yb29t4N4F+f915RxXBGExYDvNyWn8TYIhNcLzsag4gYJ4HX0WJ3E6JelR/z9q5X8pBiwuJD1HETdvO/Wi10nqgadZUfAk3x4XKuC1x7p58bwSylMiONZhoigpnHMKwsBp4pJxFdz3fRWx4Uqiwtz8XFlHcqQKj8+P2eFlak40tb2hfd6xGXr2NxtYXp5E56CdMLmEE91mFo4aWUWMS9ejlIrIPYsA/dh0Pb9UdnHR2BQcXn9IYrxiUjpDdg81PRZ0SimFCVpkEnHwbwxWF8VJAr74UOsgb21pIEaj4LZZORhsLrbVDTAvL5KPzk1hS6cYt9ePRinlvT8EaGJCmIj5+eFU6Kw4fWLeP+6jacBBhEoWsk2UikUhokhrj3bxxPIibvsyFGnz3cFOpuXE4PX7uWFKBhFqMb1WH9vr+1HKJLQZHexsNODx+XlxVSlN/TbumJNDn9nFl/vb8PgCtA/aaR6w8uZF5VhdnrPupFIiVZgdbhaNSuSFYYnZk7Gtrp85BbG8cVE5f9T1cahtiOKkcNKi1HQYHdjcPq6ZksmeJmPwM8ZqFQxY3by/vYmbZ2QxKUuP2tHPrdPTeXNbS/DYt8zMprl/JGxVLBJR1WWmxWDj092tmJ0ebp6eTUqUir2NBnLjNGyp6QtWfnfMzeGB1ceChKo4nYKLx6WGSMo2DthYVpZIcaLgJykSwbEOE7/X9OHy+skZlitoHrAyYHWHDIQPtQ1R2jqI3x8gIAqgVUqRiUUYrK6Qv1t7rJvJ2dEhO4esGA16tQx/QJDAlUnEI3r4zQM2Fp82vNcpZTyzchQ3fHYwiEJaXpZInFZBUqSKpgEb4SoZK8oS+fZAO0ari1n5sdT12njkpxMEArCiPImiJB2rD3XiGB6arznaxb3z8/h8bxtGm5vLJqZx8fhUZuXH8o+dLTw3VUHOuovAOVwg1a+HxS/TlH4BN352MAgxnlcQxxsXl/PbiT7KUyKI0Mg40DzI0ytGUXEWvaRAIBCahPUZcNlPcOwr6DwIJRdCzjyQnEqX4mF8fda/icnEf1QiDwQCbDrRw3yXYcTvJAO1jMmS8e0wqitVr0KtkBCtVbBqdBLZOh/fHepiS4uTiZnQZLCzvDwp+DCdFMz59mAHb15UHmQPysRizhudTEOflbLUSF7aWIfT4xcGPk0CNX12fiybh4dIObEaYrQKbvriCK9dWMZlE9L4cl8bXn+AWfmxyCRihhxeLp0Qz6UfhvqGfranlVtmZlHTY+GKSekcah1i22n+jo39Nm6ekcUvx3uDrL8+i4tH1hznkSWFTMmOweEX41DoSI9yct/3lbiHK60Lx6YQ6+vh0cB7yHZ9CxIFs8pv5yPnDBIjlJxTkcQ3w2iWXyp7+NOsnKD066Ddjd8fOOuQrnnAxnRNJ0XG75HV19CRsYrwmHK+qbEHF8O75+Xx/C81dA4POrNjNdw0PSto6jCvMJ77Vx/j8onp5MVrSdWraBtmCiqkYs4bncJT605wRZr+rMJGMVoFuxsNVHUJw7eJmYIx8D9qWmkz2llaEs/frxxDfZ+NIbsbl9fPO9sahzU6hLZHbbeVa6ZkMDEnjjajDa1ChtMreJEeaBuk3SjIAl8+MY0ItYz7F+Rz93dHyY/Xce7oZN75o5FBu4flpYmc6LYQq1Ny17w8AoEAR9qHQlixvWYXA1Y3MRpFsG1TmKBlUlYUj689ESQmTcuJ5uJxqYItXoOBgkQd725rQnKW1kB1txmFVMy5Fcm8+ls9k7Kig9LNp8fhtkHuW5jPgNWFRCzG5/cTo1Py2oVlONy+s4pa6cPkaBUSsmM1XDslgyk50TT12XhqeTFWl4e0qDAqUiMRi+HJtSf4YXjQKhGLeP7cEvY1GzjUNhRsLQF8c6A9qIVuc/uQScSc6DbTZrRz84ws5hfHkx4VhkQsIjlSTUVqJFR+cyqJn4xtz9EmHhfCE9lY3cvCUfE8c86o4GtzC0YWXHa3l71NRj7d04pKJuHyiWmMTosUhtrxRRD/JEGG3b95/Ecl8l6zkzd+b2DB8pwRv/PkLyMuMZk75+jJiglDq5Tx2NoqmgfsaBVSHlqYzc52N3KJGOewBojvbL0wwOMPcMHYZHLitDg9Pn6rFoR2HvmpKvg3T66r5ukVxVhdXqRiEa9dUIZKBslqL5re/ZTP1vH+riaun55NhFpQatvfMsiHO5p5aFHBWbfsPn+A9Kgw3r6kgg+2NxGtUSA9TQGpPCUCsVhEdmwYo5Jz6be4+GxPK4GAgG09ne335zm5fHrNWLY3GJBLxOiUUsKbPkN2/BvhD7xO4vb/lZtXFFOnzuHc8mQyYzRYnF7UcoHd9vDiAoYcHpIjBEW7MxmIMVoFo2Oh7OfLwSUkn+SO/SydeB/molUc7TCRFqWm1WALJnEQ8Msz8mIYn6GnvtfKD4c6eXpFMV8f6OD5DbVcNTkd7bD58ElYZrJeTUGClsnZUexsEBby7FgN51YksbW2j6/2C4vQiW4z2+sHmJEXE1QU7DW7ONA6yOi0SK746ERIRQ+QFRPGzoYBznt3D5Myo5iWG02H0YFUIqZz0MGDC/MZcnjoMbnYXNPLx7tamZIdzZyCOEYlhYeYO68+3MnF41Lptzh5aaOgN+I+C6Klz+IiMUJJv1VAGV0yPo1djQYWFCcMi4aJMNo9RIfJuWJSGluq+2g12qnpsbBqTHLwOzgZhQk6fjzShdsX4KVVJbi8XirSIkbsLAsTwzHZPeyoH+DSCWmoZBIkIhFahZTbvzrC+Aw9F4xNCRLEZBIRfz23hMnZUVw0Lg2dSsbv1b1cfZqKqFgE3904kfZBezCJn7yf//prDZdOSOPQaQPKk3GgxUhRUjhj0iKZkh1FVFgxsVoFJckRI9iTAsrk7L3rjrPotx/pGGLlGeJYZ8aeRkPI5/jleDdvXFTBumNdzCqIZVp2NHH/ZibL/yz+oxL5SUnLPnU2igWvoN76GDhNeLPm0Vt6C90GH1kxGqI0ch784TjNA8IFtri8PPhTDXfMyeXlTXVcOC6Vtce66TW7yIoJC5GsTYtSc6R9kClZ0bw8PDxNj1KPsEoD+P5QB9dOzaC218KRNgMymZRquRSdoowx2h4eKxmkye1Hp5JR2WFCp5Ty4KICNp3o4fbZOSP0SxLDlRxpHyJao+Bw+9AwQy5HMK7QKpiWG8OdX5/Cxo5KCg8+dPYzBnWvbq7jKW0xKRFK7lt9nJeWphPV/NOIz6Dt3E6BQs3aTjUv/D6EPxAIFiF3zMnlzd8bSAhXMjUnmovHp7GnycDeJgPFSeHML4onwbgzmMRPhu7gmyy84FwaKpLot7jOakTQNehgXmE8z/5SLeDp3b4gmuiD7ULlplNJef3CMrQKGbsaB3jkhyqeObcYj9fP5JwYmgesfL63jeKkcO5fkM/Lm+pw+/y0GOzEaJVcOTGdbrMTjVKKWCTmq33t3DknlxdOW/DGpkdypG0oOJzNS9AikYj5cn97cM6QolexsjyJ1zefMgDZ0TDAXfNyQ/DUJ+OX491Bos3xDhOXTUzj4BmJrGQY/jktNwaTw4PV5SFKo+Ava6qC7YqcWA3XT8vA7PQyOScajy/AmLRIPP4A4zL0wbnM+Aw94SoZPn+ABYVx/HysmwlZ0WREh5GmV9M6nOhGp0WSEa1GHyanIi2CXY0GpGIxIpEFlUzCOeVJrD7ciT8Q4M9zc1HLJZQkh2N3+Wjss5Ebp8Ht8wV79ycjTC5lX8sgZ0P09VlciEWCVsyZkRevJTtGQ3WPGZ1SxqUTzu5UFYyEUlBowXXavGPyPTQbNUAoQq0idWQL5fRw+3whOwQQWolbanup7DTxy/EeLh2fyiNLC0PEuv5d4z8qkUdpFNy3MJ9Gkw9j9DKMk4oIl3mptmkpFSeRFuWnvs9K04CV5tNgTCBcpJNthu8PdvDQ4gK21vRy9ZQM9jcb2dNkpDQlnJl5sfRZnMikIsamRbK/dRC724fmLIMYrVLGoM1NQrgKkQh+PNwZ3Pqr5RI+X6olTm7n2b1tiESCf+fGE708tKiAt7c2ctvsHL450E51t4WylHCWlyXR2C8QcF6/sIyPdjTz6/EenlhWhEwq4rlfakP+f2WniTkFsUzKihrhKuMPCIOk/Dgt10/NICc5Gm9nCdKB+pC/Eys0aFZfxnmaRKTTX+KeLcL3ppZJiNcpGJ0WSYJOSWWnKbjtvX5qBu9sa+KlTbUsWnCWW0gso9viZl5hLFmiHn7piRwhPjQmXU9Tv5VbZ2Xj8fgQi0T8aVY2UolgT/fxrhbMDi9mh5cfD3fxR90A9w/rl98xJ4dnf6kNqjR2DDo43mniwnEpfDJstZcfp2X98e4guxaEtkhatIqLxqVQ2WlibJqeqTnRIVVZpErG8U5TMImDoKdidXpHLLxhculZcc2pejUlyRHcppHz5b52DDYXz55TTI/JhVQiIlIlQ6+R8/GuVloNNhaPSqA8JZJHf6oKYb429FsJV8l5+EdBfXB0chgLM+WMzY6nc1DBHXNyGJUUzr4mAyl6NU8uL8LrD/B7bT+VXWaun5bJ7MJYtAoZMomI9Kgw2o0OPtvTyoy8WL490BEcWBYkaLllZjarD3eyv2UQg9XNqjHJXPDenuDC/viyIs6rGInBLkuNYMPxHlaWJ40wnKhIjaCmx0xunC5EjiFaIxe02tcIu1yX189z54xC/j9KmrEFcMU6Akc+h4F6TAUXclhSytj0iKCWOcCi4vizKjqeHt1nYagCqE+7pl/sa+OKSenk/Bt6dJ4Z/1GJHATzWpPZRMOgn85ANFKtmp8ONPD0b+3kx2u5fXYOXUNOojXyEW5A0cMY19peC5/uauXZc4tpNdjpHHIwJSeapHAlW2r72VDVg1gEz6wchUou4Y/6AXLiNKjlkmDlKxGLmJEXQ0KEis3V/WTHhTqN290+PjghYV6WibmFcUglYkx2N5eMS6UwUcf6ym6eWV/N3MI4zqtIZsjuxusLsLPBQPNAG2IRXDU5g3EZkSilEmp7LWdtx8SHwWNjPVywJrQq1iikeH1+mg02UqPULPvbfn5edSVFzb8HyUyemCIGE2cTK3oP2WADEziKVpHPhKwoChK0fLanDX2YjJXl8aitrRjlibQM2IjVKVlensiyskQOmltYrIkF6ymiSVfFn7ljfS+95jaeWJrPlOwIKjvNbKruRSwScU5FEpWdJn443IlSJubRJYXc8sWhYALIidVw5aR0DrQIAmbTcmIYnRbBkN3DJ7tbuXh8ajCJn4yOQQeRauH6jk2PRCQiJIkDfL63jbHpelIiVYSrZBzvNJMXr0UlkwQld+v6rLSfZavebXISo1UEE3lypAqZVERmTBgFCVqqu4UqUSYRsaI8iQd/qEQqEXHv/HxEInh9cwPdJoFdeeWEVAri4zmvIgm5VIw+TM6+FiNd5tDkMj5Dz2d7WvH6A7y/UMP4rk/QVW3FMziOxytu5of+CHKG77uTLNU75+QCAtT1na2NnFORjC8QIDsqjBPdZr7e385Vk9PZ3WgIQZ1Ud1toN9gJV0oxOb1cNC4lRPgrEIAn1p5gXIae66ZmhhidG60u8hN0fLyrhWfOGcXzv9ZitLkZlRTOhWNTUcokfLC9keVlSURr5ESFyWnstwVnMAC/VfcyaMsnLvy/qH4Ty6gKpPPUumoO/zSEy9uEVinlmZXFNPTZEImEFuyZxKbTY3fjAPtajMwvig8xL5GKRYxL1weLgQCcdS707xj/cYl80GpH5erH5NTRbXbywY7m4A1Z02PhoR+Pc+HYFK6bmsmLG2uDcMRrp2QwaPNwx5wcAgHBmOJvW5toNdpYWpLIhzuauX5aJhuGSS/+ADzwQyUPLshnQqYem8vL8+eWcLh9CI9PEIUSi+D1zfVMy4mmY2jkkKje6CUuUsJHu5pQSMWo5RK+PdjBlZPSWVaWxJH2IdZX9rC9boCnVxaz+nBncCfhDwjws/KUCFoGLGyu6WNGbixbak8lTIVUTJlzP7nbHuDNVTu5b009HYMO4nVKrpuWwf4WI2KRiCnZ0ehUUn43xVM7+h/kijrxiGTsMMfx2Q9Ofiq9mfjm1YTL/OQNGx+f3kbY1WjgmwsS8Qek/FzZJlhnzcnh7S2NiEQgmfUeZfad6GytDKbN58WaaEqSNeTEajjUYaY0VU9mrJo/JWSTG6fl/T+aONphIi9Oy1WT0zHa3Fw6IY1vD3Tg8Pio77Oyakwyy0uT+KN+gHEZeiJUMmK0wnAwRivn7nmC8bNSKuHXqm6Od5qJ1yl4eHEBcTrlWYd2J2Vpn99Qxz3z8yhI0PHpnlb+PDeXp4f1O7bXD3DtlIwQTRkQZBaiwxR4fAEKE3RMzYmm3+Ki1+zi7nl5tBnsDDrcpESqg8nX6w9Q2TlEr9kZZLQGAvD33W1MT/RxoHqQb07Y+cvSQup6rdwwJZOASITD40Mtl2ByuNlc3cdV5VpmVN6HbECoXmX164nuOUxYyUesOeJj04neICHJTyAo12CwuflwRxPnlifh8fmJCpMTrVFgsrvP2u5qMdj54voJODw+ekzO4LNzMrx+QUYhVa/m5VWlGGxuojVyMmLUiBDz6/E9vLG5nkeXFCIVi1DJJexsHCApQkVRUjhSsYhUvRp/AN7e2hhy7PKUyLNaqXl8fup7LbQa7cNkMQ3vbGtiz2lwX4vTy+5GI5treuk1u3hsWeFZrRJBWHTuX13JFRPTsDo9PLSogINtg6hkYuYVxYdU6ivKEkmL+v975P9HQuIw0uGL4pWNVSwqSRiBZTXa3MTpFLy7rYmbZ2TjCwRI16uRScS0Gm3U9FhYWBSHPyD0DPVhcvrMTj65qoKvD4y0LVs9LID/e00fRYk6VlYkkaBT4g9Az5ADtUxCSayUxPAIVp9RAZ5TFs/7OwTyjcvrx+X1U5YSQUO/lQ1VPTy3slgYNA0nKZN9ZMV9ottMSqSKw22DXD8tC61SytbaPlKj1Nw/NZrcTbeD00SucSuvX7iYpgEbfcOoiMRwFYlaCfPN3zFjnI99mgzuWG0FwoePbkGnktKctoq3rdNpapRw7ugkDreH9nM9vgAHjWrUKm/QZk0lPdWqunGjnWjNePRhU1kSncCobBlrjnbxW3UvZckRWJxe8mN13Le6khumZ3K0w8S4DD1FiToe+vE4Pn+AeJ1g0vD0z0LP3O3z88qmOnx+QTN9TkEc6VFqbp2ZTcegI6iZA3DzjCxGJYXjDwR4eVMdXl+Aty4uH9EKSY4U3G3unCPMJxLCFfgDfvISNLx9cQUtBhu+QAClVMI5FUmsOdKFRCzi/DEpbK8fQCUXs2hUPD0mJ5WdpiCxSi2XcN+CfD7c3syfZucwMy+WiVlR/FzZTbRGGWJOcDK6jRauL5EypUSATRptLrSqSDbX9AURSY8vK+LcikTyfQ3IqqpCD2DpZqrezI9DMVw6IS1IAlPLJDyxvJAv9rbh8wc4f2wquxoHyIoOo7rbQlZsGKn6MGYXSEb4ok7Jicbj87O3yUh6lHqEZIVWIUWrlHHVx/vpGLSzsDiB9Gg1qXo1o9Mj+OGWyVR3mwXp2+gweswu8mK19Fic7GsepLHfyhfXjedAizHEOCNSLePOuTmo5COT76YTvdx62m7t2ikZ6NUjE77D4+OmGVlEquVMOIuOy8kYcnhoNQimMy8Mw3pz4zR4fAHq+6zcND2L0WmRjE6NZG5hLErZf0aK/M84y9OizaFga6MBq8tHYYJuRE9OJBKYiReNT8Xh9qGWS1lztIsj7UN8e140i9J1vLqnl3XDD1d5SjiXTkinva2V8uQoVodCpSlKDOdAi5C82o12lFIJD/5wHLPTw+tzNXyQuAH1zt9wpc1g7OUXcN53BmwuH5eMTyEnPoLrpkl5bhgGFq6Scd3UDNqMdk50mXH5Aojwg0vEF/vaiY9Q8nBJAR/taKZruIJTySX8bVsT9y/I56VNdaTq1Vw4LpX5yT5Gb1gKXhdNU1/isHQCDNhJjlRjsLpoM9iZn6tlWsNfiaj9FoCsmVmIRMoQstP1UzO57rumIJ1/R6ORO+fkCNXvaYuk1QOd5lNVXE54aLU2YHXj8vjJjA7j8XXVwfcebh/irm+P8vjSQh5aVECMVkFRoo5pOTEhKJses5PvDnYwtzCOX6t6ECFieJ2gsd/KnXNy+HxvG2PS9Ty/IRRW98H2Zp5eWczao13MyIthfWUP7/zRyHsX5PH0pjaOdpoZkxbJReNS2dtkYOOJXi4aJ9i0Tc+NoXXATkqkikmZUXj9AVxeHymRSuYWxCGTinl2fTWN/TYkYhH3L8xHKZOE4P/tbh/v/dHE3fPz8PkDVPcIRt4LixOI0yoYkxbJzjP8R6PjEmmRh3Hfl0eCbZ0dDQZum53N0Y4hhuwenllfzXuXVRBljwCReJhdeCqaTH6+PtDOjdOzGJ+px+MLBHeW103NxOLycve3x7h7Xh6PrT0RbAuur+zhrYvLmV0Qy+bqPiRiEVdPTidGI2fVO7vx+AJEqGU8vKiQlzbV0msWpFqfWF5EQ5+FdqOdBxbl8/X+dtYc7eLjXS08ubyYcel6RqdFEq9TUtdr4fG1VbQY7MglYq6YlMZdc3NRyyS8vKmeeYVx3Dk3F6/PLxhMn8Wbs8NoDyEnAXywo5l3L63gkz1tTMjUMzpNT5/FSV6clsfXnkAuEfP3K8cSpzt7JR2tUVCWEhFi0n06dLHNaEcuFfPNwXa21ffz9fUTiFD/8zbNv0v8SxK5SCRaALwGSIAPAoHAc/+K454ZVoeHN/9oZdDu5a55ubyzrYFLJ6QFe1oAl4xLRSwW0dDezbJkB6KAnw6lhoryMHK33MC6ghdYV3mqB3q43URFmpl5hVlsqe2jIjWSQ21CRZoZrWZmXgxrhiFVl01M47XN9ZgcHq4q0zG/9mHkvUcAUPTXkt76B6sv/5Lval1srOrl77tamZQVxefXjqOmx4JULKLP7GLziV7+srSQHfUDZMZqePg0XeZNVb3cPT+P536pYWV5Iie6BGztBzua+eu5JUTKvKSJ+0hp/RFEIo7N/YJL19kxO4XEUpig48pJaVwxLoHRu25E0rw1eOycmne4Z8aLPL9FgOqJRaBThqomAnyyu5X7FuRz7/fCeemUUkRiEdHDujQRahnZ9qOcXxTHN1WnHoLHZ4ajlARCFgAQcNOdJidOj58ttb1cPTmDAetI4k9Vl5mlJYlMy4nmw50txOkUKKQSlpQkcPd3xzDa3BQk6kawbt0+P61GO3/UD3Dn3FzWV/ZQ22NFKpVxz9RoHLJcvjvQTueQg5+Odg2Lj9UFP/fVk9ORScW0GR24fX7e/6OJWQWxRKjkKGXiIKrJ5w/wzf52LjsLuqJzyEF6lJrt9QM0D9hI1aspSwlHKhZz+cR0mgZswR75NRMSaRoCk3NohB3e6kOdzCmI47uDHQKBSSRivymSjNE3EXaaLK855xy+bVHTMWjji72t5MVr+b2mn4vGpWByuGkesNE15Bxu0XhGoJpe31zPc+eWcPG4VDQKKcmRSi77cB9ef4CV5Umk6tX0mJ28cF4Jh9qGSI5Q8cKvtSwsSWBuYRyrD3UGE6DZ4eWOr49w3/w8Xt5UzyNLClh7rDsIVXX7/Ly/vZlvb4gPiottPNHLxmFBOIBFoxI4MzpNjrOigjy+AB9fNZaPdjbz9tYGkiJUlCaHkxQh+L6+urme+HAlmTEjyTo6leD1u+sMHSaA/HgtVV1mdjcaiNUqSIlUYXd7/7+RyEUikQR4C5gLdAD7RSLRmkAgcOJ/99hnht3jo2lAUC20urwc7TCjkku5a16uoImskJITqyEuMMDLiveRbBdMmaelTsE04Slqsj/A747mgYUufP4A2+r62dts5I+6fq4YpWR3Qz+pURqm5eYMJ4sAfWYn710+GqvLi3fY8w9gWrQJec2RkPOT9Ffj7a/nrS2nbr5djQYmZkbx0mk6MA8szMdgdVGeFsHHO0Op5V5/AJPdzUOLCwhXyoLJ1B8I0NBr5ibn+ygPfQCR6VjmvMDrRxSYnacGnSe6zVhcXiweKX65htM3q8rufVxYeADFkpnIJWLiNFKsZiNnhlgkKEFeOj6VrFgN6VFhHO0YonvIyeUT0/j+YCc+xxAPSH9j+fwVDHgVpMvNZBq+4nDC/SPkUkUiyI3VYnZ6iNEq8AWEGcOZkR+vRSYR8e3Bdu6al8sfdf04PX6yYzVBFInPFxhhpRatkWMb/vmkNdvMvBh2tFjZ3ThEWaqIxaWJ+PwBloxK4JPdrSGL10c7W7h7Xh4vbqwlVa9mWVkS729vYk6BoC2TG6cJJq36PusIY24QzIBbDDY+GtZEqeu1MiEzilc21eH1Bzh/TAr6MDnjktX4fB6u+bKa66ZmnvW79w9/eQnhSmq7LaRER/CmYTGLZ5ST6mujU5zI6r4E1lXahq+5hVkFcZgcHt7Z1sQTy4qGdxCCdZf/LHwJrz/ARzuaWXusG7EIHl5cyNVTMuk1OdlS18cPw3Kx4zP0LC9L5N0/mmgcsFE8bB13Eip6MgIBsLp9uH1+ei2uEMmKk9E+aGdKdjQ5sZqQHv24dD3p0WfRKQkwAh4sFYuI1Sl4fXM9O4ax9B2DDh5bc4LbZufwym91JIYr6TE5CFfJiNKcMs61ubzY3QLJK1Itw2hz8+4w61kfJueCsSk8s76aqyen4/UH2N9i5K0tjVw2IY38f2I+8e8S/4qKfBzQEAgEmgBEItFXwHLgX57ItUop55TG8ea2lqA2xJ4mY1CfJDFcycz8WK4L24GkZk3wfR6Ph52DOn6rN7Pm6Ckc9jVTMjDY3ExNEpP88+V8XnE5VxzKYu2woFVypIrlZUnsbGxlaUkiVpeXqDA5Bpsbn2hknw6JHF94OrfPFqrd5gE7Px7pRD5sKtxutPPtwQ621vZzyfhk7G7/WbG3IpGINL2algEbWTFhNA3YuGpyBjJ7H8qBapj1MHidDGmyqOntH/H+hj4r725r5MkFf2F+02bwOE4eGGdMCeZ2L+nRaopEzVhkzhG90FVjkvmtuo/CeC2NfdagkP+KsiRBMXFKOursFLTV7zO59mvhTVIlx2Z9QseQi0vHp/LpsBokCH3NDVU9fDbcUx6foWdRSQJXTxZMGgIBocq/YGwKz66vwe3zs7/FyIaqXow2d4gY0Vf727hzbi4f7WymY1Cogq+YlM5LG+uQSUQopWLmFMRSnCRUaEq5lNd+E5QgJ2bquXxSOt+dMcsA8A23LdqMdjTDwlSba/rIT9AxNSeGeUXx+Hx+NEoZuxsN3Ds/j9c21+Py+onXKXl4kYC8ORnJkSrqeq1BKvxJLZhzKxKZkStgwlVyyQhD5FVjknl/exM5sRoum5CGx+dHLRNjFus4pkqmVTWZW788TCAQyn3oPY1wddLQYVJWNJtr+ojWyJFLxMGZBsCSkkQ+GsZR+wPwt22NXDExDZfPz/HOU4XB3mYj03Jj6LUIu4k0vRqlTHJWZ6eT7ZHWAVvI4ncy4oYNHV4+v5Qv9rVxtN1EWUoEMVo5zQM2ylLkDFhcHO0Yon3QQVqUmkvGC8zo+j4r0Ro510/LRC4RBZP4yfD6A3h8fu5bkM8fdf1c/ME+UvUqnl45ikmZURxoG+SFX2toNTpYXpbIspIEJmZGoVPKcHh8aIaZq9Nyomkx2INyv9XdFjZV97L6pkkkR/73iWL9V/GvSORJQPtpP3cA48/8I5FIdD1wPUBqaur/8j+xubx8uruFxTlKpLIc4nWKsySgFHrNDqJUUD/tdcxSPXFhMppl2TT3+1lzNFSj+ZPdLdw9N4dZ7EdcXYlm0z08PP97Vg4/54tGJfD1/jYGrG621fXz7qUV3LsgnyfWVvFti5LxOStRWNvpyLoI8GOPKeO679rpHkZMlKdEcMWkdOJ0Shr6rUzLjWZshp4+sxOFVEKMVsklE9I4+t2p1opcIqY4UcdNnx9CBNy7IA+VTMKGqh7umx5PQDWdRm8Mlf5MXC1W7p2Xw5Pra0PaGUkRavqtbp7f2k3Ooq/I3PMgbpmOhrFP8GWNks/31uMPwA9LxJQffYznln3N1kYz/RYXFWkRHGwdZGxaBG6/P5h8QZAvuH12DvowGQ0eDTuLXidf1ILc76JRlMrDv/tZUW5jWUkSEzKj6TE70CllxGoVHGob4o45OdhcPj7e1czETD1zC2KZmRtFr8VD04CNlzfWBZNNfa91eDjpprHfxrj0SPa1DDJo9/DixlqeWlGEViGjttfCX3+tIU6n5JrJGSREKGg3CltyfyAQ9B0F2N1kJDJMzsryRH44HHovJJ7G4DtZ1WsVUhxugQBTEK/l0SWF7G42kJ+gw+Xxcc2UDMJVMhLDlShkIgoTtBwY1uLWKWUM2kPhr8XxalalWMlTmLl3ko5P9rTxp9k51PdaGLJ7WFaWSJ/ZyeUT02k32nnq52pEIvjH1eMoS4kgEPCjkIr56zklbK3r45fjPahlEq6YmM4Lp7FL06LVLC9L5LsDnfxlaRF+v5+/XzmGY50mttf3s7QkifXHu0N2Nf0WQcfnjd8bODMOtBhJ1as5Z5j9W5AYzlMrirn580NBZMuSkoSgQN2Gql6eWlnMU+tOBHvRl09IDRpBd5uc7G40kBmjYVtdP51DDg61DvHqhaU8+0st3x8SWn/jMvQUJ+ooSgxnfnE8UpGIwgQdHj/Bgur0iNEoWH24gyPtAuKozejg6o/38/m147nsg33Be+uD7c3kxmm4//vQ/vvotEgeX17Iktd3hhy3z+wavh//353I/6ciEAi8B7wHMGbMmP9ldGZdr4UUXwd5P9yI9dztbG8Y5NaZ2bQM2DE5PCwaFq9yuD38zTaT97c34fEFSI6Q8+L5EaR4ndwxJ4cjbUNsHZ6We3wBysKM5Pxya/D/JPp7yYhOYmFxPB2DjiAW3eML0GVy4vUFeO7cEnRKKZ2qv/L5gS6++L2bvDgNefGyYBIHYdC3vCyRz/e0YnP7gkmhLCUctVzKuz9VEa9T8sSyIvY2G9EopIzP1NPUb+OKiWl8tLOF536p4b3LRrOwOIHO9kbkkjQu3KFn0C4gDhTSIZ5eUcw93x9DIRVz8TjBzzAQgH6LkzWGDFo1z2NzwzRHHJ/uOWUx9muPhlEiKfmG3/jRUki/1TcMp4xiVFI475/FGPlAixGRSIS2XIYvPJlPW8PYVttPl0mo+sdnRnHtpwcwOTykR6l5fFkRd35zlEXFcRQlhdNrcvL08iKkUgkvbqzjUPsQUWFybpmZHdICKErU8flwVb/maBd3zslhWVkiPYM2KsL6kXj7uOKbHiZlRXHNlEx6zU6e/aWGZ84pJlmv5sWNtdw6K3vE+W+u7uPTq8exv2WQjkEHOqWUp1cUgwieXlnMV3vbUMkkFCfpuGFaFm1GO9dOSWdMup4r/r4/mAzunpdLuEpGZkwYv1f3sfZYN7PyY7gzJ4ZXfqunttfCyvIkNg33gWdmqHg2ZS/xm54Hv4+btQmMn/EWF6yrZXRapMAStXt4ev1IfZQekwODxUXHkJ0v9rXjDwhEm7curiBcKeX+1ZXBXnu0Ro5OKcPp9nP7nGzqeq18vreVPU0GPL4AK8uTGbS72dkQ2iMuS4lgzZEupmRHj2ChTs6OZnJWFJkxmqDdWVFiOO9cOppWg52kSBUbq3qC0sxiMbi9fi4Zn4ZUIkIuFTM+Q8/Hu1rY12xkZl4sl05I5dlfaoMEqBPdZhr6bMEkDrCv2YhMLOKG6VkECHCwZZCrPz5AmELCrbOyeern6mALb3puNLlx2mASPxkeX4D6XmvIbiQ/XovB6h6BET/YOojd5UMqFo2QcZD9V9K3/83xr0jknUDKaT8nD7/2L41+i4spUSaY/zQycxu+gIZn1tdw1eR05B4xnUNOnvulmj/PzQtBQ6yoSOKZn2s4NkwgmZEbw0XjUvhyXzsZUWqyO38E/6mtrTIqmYcW5fHGlkaOnsZGzI7V4PUFePaXGlxePwuK47hkXBoTc+KwuMDl83OsQ9gqzi6IxeX1I5OIMViddJmc3DQjk41Vvbx6XiHVXWZEHjvPLM7gtu/r+cvaKlaUJQotmGF7qtGpEby0qoTOIScej5cn1p3gH3P8bDEnMWg/tfV1ef1squrmk6vHsrPBILgfDQ+ZFo9KQCUTM7skk0AAzC4Pt80WktuOegPvHxxk2vIXGDP4Cy8Xq/Ep9eDzYtWG8XXjUNA9/vRIi1LT1G9FKhExYBZYrTdMzyJKI6fP7MLqdAcHVGKRiD/q+nhyeRFrjnax5kg3S0sT0KhkfLWvnXC1nIcWFfDF3jbe3dbEyvIkPtvbxqJR8Vid3pCKUSIW4fX7udH7OWZbBEfVy7hjTg6dg46gmUi8TikIUQ3rk58NCZEVE4bFZufOOTl0DDooStTxlzUn6BwSRLFunJ5FQYKWPouL2786jD8gJLkItTyYDMakReIPCKbDFpeX2flCYnpnWxOTs6K4Z14eJqeH5Egljy8r4qOdzdxW5CT+t2dPnYilm4pjj/P7BU8QZqmh0RqBXB4VQk4SvkPI19hRe9w8t+HUxvdQ2xBbavvw+nycU5EcbNMMDRtX3zs/D6vLS3W3GYfbT/uggzFpeloMNpRSCffOz+eDHU0MWN2UJoezaFQCz/9azWsXlnP5xDS+2Dss9JYXy4KieJL1amxuL/WdJlweH0abm70tRo51mMiJ1TArP5YxaZEgEmF1enn1t7pgEXTlpHTu++4YyXo1FWmRNPZb8fr8vLSqlPu+P4bL62fRqPiQZHsydjYauHt+HoN2N68P7xbcdj9f72/n+XNL6DY5kUvFTMzUo5ZLR9jQgdCSVcslqGQSbpyRxeG2QVRnwZnHaBW0GmycU5HMNwdOfddFiTpy/83Znf+KRL4fyBGJRBkICfxC4OJ/wXFDIjM6jLBuJ5h7yXS1E6NdAUCEWs6eJkNQWF4kCjA/NxyFNMCJfj8GqyuYxEGw33p4cQHXT8tgfqaKqJ++DP7OMfFu9trisfrt3LcgD4vDgy8ANquF9GgNO1ut3DozWyDpDA+2DrcbuWFaFhKxiPw4DYMObxCaJhGLeGJZESvKEojRKpiVG8kd350aHSRHKHh/VSYXf17PD4e7+NNwBRmtkQsPhMuLTCKm3+bh1pnZaFJVdO9vGfHddJhcBPxCwum3uhi0uVlQnMC4DD1SieA7GiaX8OX+VnY1GhGJYGlJIgsLInFqw/mgfz5X17+Lpk6YK4QrdFyx8ku+6IojXqcMGk7E65TMLYzH4/fT1GfjnT8aGRrGacdoFDyypCA47AOID1dSlBTBHV8fweMLBNmwD55mPLylto8HFxXw9M/VTM+NIStWQ16slhM9ZiZnR+Fw+5mVH8vOxgEuGZtEp3gKV2+R02kSCCVFiTqum5rJl/vaeGRJASe6zUFWX5vRztScaFIi1URp5IhEkBKp5savKokKk/PCqhLe2NwYtKk7SVR586JyjnUMBSu2I+1DpESqyI/X0mNycvWUDG7+/FQ//LfqPsJVMkF8q9HA5Jxo1h/rIipMzkc7mnnunFEkGzeMuG6inqPYAnLeas1mcq6Uqrb+ECy9SAQPLiogqXMd641lI96/q8HAtVMz6Bxy8HNlN7fNFqpUl9eP2enF7vbyzram4PWr7rYwPTeGBcXx/PXXGq6fmkmYQsrOhgE2V/fw3LklfL2/HX8gwOsXlZMQrsTj83OgdZBuk5Mv9rXxw+FOJmcLC85v1UIf2eH2kREdhlImwWRzUZEWGSTk6JRSylMjiNMp+KNugDd/byAxXMkN07OwOD2cNzqZziEHV0/OQC6VhChCgqDHkqZXs/U0IhwIw+RXf6vngyvGEKtVBIeajy8rCnqnRqhlPLSogAGrmysnplOUFM6z60/QMSSQneYXxbGhStgxnXxW1x7pZFZhPMVJOo60DxGnE3SGDrUNYnZ4yY3TUJQUHtR9+neJ/+1EHggEvCKR6FZgAwL88KNAIFD1X7ztfzlS5RZEMfm4YgoI87s4f6iSRdfEoZAMckVxGh6vn5vGlZJgrUTZ+CoiuwPHjJv53RvL0QQdJ7pPVbGN/VZGJYUjD9OxZ/461LoI+pxSgUykVRKvkrG5uo/1x3swO9w8urSQ+9fUY7C5uH12Dk/8fIK6Xiuj0yK5dWYOz/9aywvnjUKkVfD+j6c+us8f4Jn11Xx2zThUUhH3f19JSZKOpaWJmJ1epBIxXiSUJoczKSuKCZl6cmPLONg2RG2vZdgH0BV0h/ktOZxbpqfyj8Oh7ixzCmL55mA7VpePaI2ce+bnEaaQsrvJwPeHOvH5A8wtiKMgIZxdjULbZc3RLuYlRmLrH2K6pgvVwVPDYVxmInc8gS/hWa6anI5MIsLs9CIGPD4fDreAJLllZjYuj5/fa3o51DZEdY+FeYVxqOUSdjUaaDfaaTPYg33UitSIIAnkZAQCcLzTRGlKBLuaDESq5QxYnfxR18fcgjjEIhHfHWync8hJZJiSzxti6TSdgq1VdZk5pzyRx5fk88hPVRhtbl67oJRZ+bH8dKSTv10ymsfWVtEx6EAuEXPTjEweWVxIn9VFICDiQOtIdEVNr4V5hfEc6zAFt9j7Woz8ea7QTjlziAeCLvrcwji6TU6KEnRcMyWTlzfVcemENF7cVMdfSiKJOeM9nthivqj28XWlma8rzdw+O4c2o433LhtN15CTyDAZGSob2rV/I7f07RH/szQlnN2NA2yvN3DRuBTitUpm58cSo1WQGROGweYOJvHTz/O80ckM2T1Ehsl5Ym0VGoWMR5cW8qfT9OZ3NBh4cVUJfWYX1T1mMqLDgkiWWXmxrD7ciUomQEPTotQhdnMXjUvhi+vGYXH40A6bgH+2pzUoTNZlcvLEuhN8ctUYKpLCUKtUpEcLxgwfXTWWlzfWcrB1kGm5MfxpVg56jYKM6JFQwtLkcDJjwkJErRYWx5MeNYlmgy3olmQ/zbXpnvl5PPVzNZtO9DKnIJb7FuSREK4iN05DhFrGr1U93P3tUWQSEXlxWhYWx/P0uhMM2NwM2T14fH7ev3wMswviRpzPf2eIAv9EyvX/ZIwZMyZw4MCB//oPh8PX34CvuxK5qVlQ2mvbDRNuxWftw6zNYUCkp8oRwWRpDTHfnxvy3pZZb/NqVyEZ0WG8saWBW2dmY3N5OdZhYnpuDMWJOtZVdge1uKPC5DyypJAIlZRAAPosTsRiMTqFFLfPzwM/HB+ulEUCM1Qt56JxKYCI5gEbRrsbrVLGmiOdQZr348uKyIvXYHJ4sDi93PPdsWBv76rJacTrVLy1tQG3188VE9MFRb4OE1qFhOI4FR63i2+OGcmM1TI5K4rWAQtvbWvG6fZzyYRUUvRqPtrRjEwiZmFxPDFaORanl7+sCQUOXTEpnfWV3cHB6KWlOpbkqkka3EvKjgdCv3SZiu0LfmUIHaMHf6E9bBRHPUm8tLEOl9ePVCzi5plZ/Hysm1n5sWyt7SdVr6bf4iRcJWdKTjQfbG/i0glpQc33itRIwhSSEJ0OgHMqkhibruevv9bg8vh597JyvtrfwS/He5CIRJw3OpmlJYnU9Jj5/lBnyKI8KimcSyekYbC6GLC6ONwmzCVEIhF5cWE89XNNUN/7ZLx+YRlH2ocw2Ny0Gmwj+qp3z8tjR4Ow4JxERC0oiiNCJWNpWSI2l48ttX2sO3pqYDg6LRKtQkpBgo41R7t4ZEkBKpmEXrOTox0mjEYDz8ZvJeLAq8LqpdZzZMq7nP+zL9hSiNMpmJUfS2qkml+rurlxejaRYivjt1xKT8pCHuydwZYmAa0Sq1Xw5PIiTnRbiFDL0CllbK7upTQlgk92t2J3e3lkSWGIkzwIleczK4rpt7ooS4kQ2n82N2uPdo2AFM7Mi2F5WSLHO82kR4exq3GAkuQIuoccdA45mFcoLLQP/3RccOI5LR5aXIBOIWVSVhQmh5clb+7gzLhtdjYiBL/c22fncMHYFMIUUup7rfSYHCTr1WTHaBCLRXQO2rnr26PB6xGplvHpNeMpTgofcdyT8cz6at47wxFrQXE8LQO2IKs1Ra/irrm5LClJZHeTgcvO8Ah45fxS2gcdtBvtxIUrsbm8bKntY/VNk/+Hei7/p0IkEh0MBAJjznz9P4LZGbB0I6/5Caw9QhIffxNYupFsfYZIxyCRygjiln1IoGb9iPem1H6MWfwYfww6eGSxsPU/2UPe22xkWWkCPafBqAw2QXnvz3NzMdhc9FrcHO0YIl6nZE5BLFOzo0mPCSNcJWNLTR9+fwCVTMR9q6tC8K7PnTOK+xaoqOu1kBsbRsuAnTajje8OdgaTuFgE0RpliAHAu380EastYEFhLG5fALfXj90nJUxh5sPtzRgsLiZk6nntgjL6zE4cHj93nvawHmkf4u9XjmVL7UjCw4EWI08sK0QuleDy+omQQ/egicyYkUNBW9oc3t5n4poJKhJ3PYpn4tNcs8MRdNlZXBJPfrx2WHNGxKSsKMwOLz1DDixuHwMWF08sL0YsFgWNao+0D/LAooKQRC4Ri1g8Kp61R7tJjlANY5CHgrR2byDAV/vbKUjUgUjQPDmZyCdlRZERHcb9q4WFMUaj4JULSqntteBweXF4/COSOAgs0h+PdHH/gjyWlSZy3/fHgv3ccyqSONQ2iFwqxuYSKrkYrYKLxqfSPGDj1d/q2d8ySMKw1+THO1sw2FxcOyUDg9XFT0e7mJwdzZf729k27FuZHqXmgYX5tKtzMaYtwGvqJaDP4LKvukP6wu7huUp+go6JWVFsr+9nZqoUz6TbSVhzEy8X9NEwby5OkYrY9Dwu+rQSo92DUibmorGpXDguhbe3nmoV7W02UpyoC/kOLh2fSqxOQdOAVfCcDZOTOixhcWbIpRI+3NESFCh7+fxSHv2pKojB/626j5fPLx2RxAHMDoGZ+tdzS0iOVJ4VZZIRHUaESsa1UzNI0avYXN1LQ58NtVxCdY+ZzdV9vHJBKYtGJRIfruKF80poMzhweX1kxWpIi/ofO9afTW/HZPeEaLqcW5HMU+uqKU2JHAGnzI/XUtlpCmkXjkoKZ3yGPuhp8O8S/xGJ3GPuRRqTCyd+EF5IHgvr7jilg+0cQrfpbhx5K0e+Vx6B3e7nYKuZG6ZlBpP4yVh7rJs/zcoOITBUdppwenx0DTqDmuSAICm7vIjOIQfPrK/hkmI11yUcp7bdHZLEAV7aVMeCojh2Nhq4b34eRquLyDB5SP8vVa+mpntkogH427Ym1h/vQa8WWiUOt4+75uWy9mgXJckRtAzYUcrFQQ/P02NDVTeFCVrWhhZjlKdE0G9x8eLGOiwuwZ39kvGpvLh/gD9PeIT4A8+D14U/aQzd5XcwplNL3ZCYyPP24jc0BB/g88ekUJoSzl9+OhH8PNEaOTfPzCZSI+ePI10M2j2kRKlJ16v56IqxuLw+GvssRKllPLOymMoOE2PSIwlAsA+eF6+jNCUcAjAmTdCkOdpuoqbHzLH2IcpTIkmPUjMzTxAPm5gVFUKV77e6ePaXGm6cnsUz66v5y5JCbpgQS02/kz2tFipS9Xh8fpIj1Rhtbv6y5gTPrCzilfPL2N9qRCISsb1+gINtgzx3ziiG7G4uGJtCpEpGmEzC6kOdHBveZXWbnDz/aw1/u2Q0TQNW3tzSwJyCOI60DzE5OzpkWNZisLOxqge90k9apI4hr578QDhef+i1u3FKCrkJEexu6GdqXiyH2028tKmPMckJvLXiG2L6dzE2cBxvbAl37u7HaPeQEK7k2qkZfLq7lY93tzA1O5p75+fxwkYBxvfupaM50GKkzeggJ06DCAhTSJFKxLy8qQ6PL8Ar55eweJSwk1tzpJt+q6AhvmhUPLd/dQQQFtzmAdsIFvCJLqHtcrpstFouIRCA+xfks/pQB7U9Fm6ekc0zv5xCmczIi2FLTR9NAzZunJ7F1/va2XbaAn9ORRKFCYLvaUK4ijVHuthc08fETD1XT8n8L5M4CLyHn46EfsfnVCRxuH2ICJWMspQItjcMYLC78fr8pJyhmT6nMI53zhD3quw0cfH4VGK1Cv6d4j8ikQdU0TB0mo622zrCzIDBZjxpU1AdfAc8w8laJKYh+0r2rhNssHxn4o3+SYxJE2yr3v0j9CIabG4G7R7WV/YQq1Vwa/QhErY/weEpIw0bTHYPYQopTf02bv3yMLfPyWHA4ubGaVn8bZtw3CGHwHQ8PYoSdVR2mvh5uCI12Nzcv7qSDy4fw7cH2vnz3Fx+ruzmpyNdRKhlXD8tk36rm64hBxKxiD6LC39AhEomofC02UC8TsnCUfFc/tH+4PewoaqXGI2CsuxkfvWsYOLKWWgUcvYaVWw7ZmZrXQuPLytizYkuLi7LJlzVhMXpIT9OQ2WHKWRRGrC66TDa+a26L6iN/uhPVTy4MJ8PdjQzaHdzTkUS+Qk6eszCQOyRn6qC/ctRSeGUpUTw+uZ6/jwvlw3Hezh/TDIFiYLSYHKEimNdJuQSCTdMz2DV6KQRMsUg9Mzr+yw8NTeBMeb1LOj+kKGkmewov4pPDvajVkgRiQSvysnZUXy5v502g4Orp2RgsDpZXpbI7XNy8Pv9DDk8PLZGOMcnlxcFk/jJ8PgCdA458PkDVHWZmZAZxcLi+BAt85Oxv3WIspQI3t/dzJLiWEoTtXx7cSofHTTRbPKzoCgGvdRLhNRDXkI4vx7vCZJSDnTYGf8VfLl4GhO3Xoh5yqPU9wnGCReNS+W5X2qCc4g/6gdwev1Mz4lBLBbx5pYG6noE67mNJ3oESG6kire2CPfgg4vyMTm9NPTbCVNIuX1ODl2DDoqTw/l0d0vIZzjb47Ohqod7FuTz/h9NVHaayIxS89CCDKr7Paw71kXlMLnoq/3t3DMvD61SitPr51j7EGuGjbhbBmwhSRzgh8Od3DYrh8PtQ+xvNvL3XcK5tBnt7Go08O1NE4k/i56K0+NDBChkEsZm6HnzonJe+70ery/AstJEnG4fR9qGCBDgt+pe/AHBlDpFryYhQsUjiwv4YHszaoUEnVI6AoYIoFfLBTu4f6P4j0jkCl0MLoMeRfYcaPgNbH0gVYL3tK2TRMbBwTAmLXkTaf9xLG4R9Zqx3LtTRiDg4pYZGRT6a8mOCaPhtOp5SUkC0Ro5ErEInz9Ail7FdVMz2N88eNaL6PQIONOLChUkVP4NgFxpH3KJNpQ5V5oQfBA9voDQ+/2jiefPG8W8wjg21/QRrpQxIVNQyTu5rZuaHc0/TtOOORmtBht58Ro21/QF+/m9ZhdPrqvmtQvLONAyiMfnJytGQ3q0Oig0VdVtFujTLi9/+amKKyelhzi8/FrVAyL4bE8bd8zJoabbzO817aTo1dw2O4eDDV08FPkbytXv8/rsj3lwhwQfnNU7s67XQoxWEWJy8dX+dsZn6Fl7rJuv93eQqldT3W2mx+QK0f+o7DQNwzZ9HG0fwuTw8MPhLpaVJtIyYOetrY3B3ZRELOLBRfnEahX8eW4uHp8fpUzCT0c6kUnEdA05uVG1C/Vv94NExv6CZ7j1u1OQ1O31/Ty0sIAhp4d9zcIArt/iwury8vbWRrJiNVw9KZ1ttf2sKEuiacBKi8F+VmibWCQKIjQGbW7Oq0gO4RKcjPLUiKDxwbrjfRSn6PnjxAA3zsyiw+Sl1+zE6pbw0bpmHlhSgD8Q4JaZ2cgkIsQiEW6fn361HEfyZPpV2UzJiaamx4LH5w8mcY1CSlqUmuouM1dOTsfp8fHFXgM2ty+kYjba3MwpjEUhFqOWC6JuVpcXrULKrbOymZgZwfFua4g6os8fIE6nGAGPPLcimW21fcwtiOWR2Qlk1b5L+IFaXBVvBJM4CACD5zfU8tKqkhDLROCssEOB7Svl3IqkEb9vH3TQ1G8LSeQ2l5cdDQO890cTMomIG6dnMSEziiWliUzLjcHm9tAyYKdryMlDiwv4/mAHbm+AZaUJrKxIDl7DSdnRWFxeOgcdQbni01uBUWFyChP//ej6/xGJXJJQxIA/jLCYYpRZc5EYGhDNexp+vQ/8XhBLcMx/kW5vDAetTiYdX4077yIMXgUXF8nJTsmirs+Kru573isfzc/e0RzscDAxK4oUvRqTw82nV4+l1WhnwOLmli8OkxGl5popGSFMt5MCU/OK4jH2d4JUqKYLD/6FTxe/zUtHxLQOeZhbEIdEIg5BN5zsQR5uGyIQCPDJ1WMBuP3LI6wak4JCJiYQEHDaaVEqanpCkRG+QIAes5ttdaEwLBDc1lcf6sDm9iEWwZPLi+kadPLB9uYRVevS0tBLnh4VRofRwcTMKA63DQVRJY39Vp77pYavLitA+dVfAZi+6yq+Kr+d4+pUKtIi2HEGqWR2QRzPn8akBFArJHj9fi4Ym0KMVoFELGZFWRL3rw5F3gB4fQGKEoUeZFGiDrfXT7xOwaDDFdIS8/kDDFhd/Haij91NAlVbJBK28lqllKbWdtQH3wHAkzKZj2pDP3MgAI0DVmKGIWvjMvRoFBIcbi9hCik76gc41DrIo0sKeWFDLQUJOrJiwrhheiZv/F5PQbyOPouLwgQdEhEo5FJumJbJpKwoesxO9jQaWTQqPtjnz4/XkhWjCdnmOz0+jvU4aDS4eW1zPUN2D3KJmBtnZOL1+tnfPEhWrAaPzx8sCACeXvJXdtT0o9N4Bcje8H11+cQ01HIJtT1WpuZEMz5Dj8nhoWPQMWKIqVHI6Bp0cv/CfG749GAwMVtcXl7fXM9fzy3h010t3L+wAIPNSSAguCEVJuh4cFE+B1sHg5Vsv8XJr1U9NPRZyYuIpj95IQc1M9Ep5UFd9NMjRqsMzkxOhtnhJSFcGdRsl4pFPLqkkGOdJo60m5CIRdw2O5s3fm8ItmbOtF/b3Wjghk8PBn/e02Tki2vHMyk7Gp1Khk4lIyH8FC9iXIYep1tA1QA0D1hpM9i54+sjQenjbw928NbF5SRFqNjdZCA3TsvVk9NJOQu/4r87/iMSOUBcUjqWATnNigy0GRDtaEZ07ke4vAF84WkMhWWQb/chFesxJ39KeN8RpnubcGWMpk2qY3JUJC2WO0jzNnK1u47L87IYkEVwos/Bp7tbWVqaQEpkGEN2D9dOzUAhlZAcoeTRJYX8crybrBgNOXEaXt1UT7JexQVjUzHE30vUptsQWboY/9u5/D3/fKqm3UmzQ8V9359KVGPSImk1CBVRVoyGRaPiOdBs5IcjXVwxOZ2XN9UFb9DbZmdz3bQs7v/+WLDSKkuJoNVgZ8DqIl6nHDGUCVNIgg+GPyBsd2cVxIbIF5wMxWlECKVMzIryJB796Tg3z8wOEmtOhscXoNEEFZp4RNYesBtI2f0ompwjWHL+MmwGIfSCL5uYhj5Mjkwi4vSidXlpIlKJmL/vbKHNaCdMLuGBRQVcMCaFt87oP4YpJNwwPZN7vj0WTC5Z0WHcOTd3xOfQKmTBJA5Ccn5/exPLShMpStZjlc6gadQk/JpEFJUjyR+BACRHKrl/QS7p0YJZQZvRzoKieO6ck0NDnxWNQkJuvBZ9mJyuIQfjM/Q8uLCAjVU9zCmIY0JGJE0DNtoHHRwYHoCuPdbNwdZBJmTquWNODnE6JS0DVl797VQvf2x6JNXdFlaUJfLutqYgFt/t8/PG7w3cOz+PvHgtC0fFs7fJGJSiyI7VYPVJKctMwDfs3ymTivD4sjnQOhh0u9lS28fhtkFumpHNnMI4uk0OjrSbUEjF3DhM3sqMCaPX4hyhvmhz+/AHAiwsSeRo+xClKeF4fQH6LC5+rephbkEcMWoxu5sGaTVYMTt9NPRZ8foD3PCdiYcW5vPB/mYUh05w66zsEAnne+blECH38fW1o9neOERlp5mNJ3rpHHLw0OIC1ld2c7htiFtmZvHhjuag525jv5WS5HCWjEpg7bFu5hbGkhUr9Mjtbi82p5f1x0K9BABWH+pgUnb0iNdBKKxkKmERbOq3cskHezl/TAqDdqHdedmENLx+wf91YlYUNpeXFoN9BJzz3yX+YxI5gDY6kZP8qkAgge5BOx6/D19AhN3tJUrmpcUM3d4YXLLZpMWp2VbXz/b6GiZkRTElO5rOsHH4/AGcPh9qkRiZRMxTK4qRiUUc7zAxIT0Cj8/PH3V9HGu3sag4gek5xby1pRGvL8CNM7IwWF2oZGKatNOxL/qUyLYNiCJS6YmfwZAonFitiA8ur6DH5MLk9NBisPPtgQ4KErSEq6Q883M11cMVd0u/jaeWF9NtdqKQismP0/LlvlaeXTmKVqMdqURMml7FHV8fRSoW8eAigfRyMslnRKlxevwhbSCb20dVp5llZYl8d/AU5VkhFZMepeadSyoYcnhIi1JzYLi6sji9RKrlI/q7PVYP+ya9w/jfzhN2P8oIWvOvp6XPQVlyOAuKBNy4FjuRIjPfXJTK+lYxvRYXcwviqO4x8dOBjmC7xeb28chPx3n/sjHMzItha10/apmE66ZlMi5NzxtbGkKSS+OAjT6LC6VMHIKOUMpG9igNNjdj0vUYrU5elVzFBxs7EYvMPLy4kG0NxuBiKZeIKUzQYbR78fjg9q+OBBfCdce6GJMeyfb6AX4+1s1VkzP4cEczm6t76be4CFfL2d5gYHuDgbVHu3j+nEKe31DHuRVJ1HRbMA+vYifF3MQieOG8Ui6bkMbRDhPlKRHIpWLe/aOJW2Zmj0gMJwXEpBIRL26o5YKxKdw9Lw+r00tmTFjIuabpVdy7IJ+EcBW7GkMX4b3Ng1w83sPW2j7GZ+iZmhODzx9gzdEudjUOUJQYjtHmHiGmpZCK8QUCwfbbF/sEOOaHOwR9nKq2Pp749RSkL02v4tULy6jrtZCmV+Py+pmZF0usToHd7eXdS0fTZbKjV8l4an0t/VZhJ3NORRJWl5dPrh6LTimj1+zk3PIkrp+aicnhCSbxk3Gsw8TlE9OYVxTHmHQ94UoZe5sNvLyxjlaDnYXF8VwxKZ1/7GphVr6gWikCNhzvZmyGHn3YPx9O7mky0GN24g8IpLWbZ2Tx3DCDGwTjiWm5Maw91h2iyfPvFP9Rifz0EIlEJOpHTq7jo72YHB68fj8mh5fpuTFMyNSjlEmQicUY7e7h3wfYdKIXi9PDpePTCFNKCVPK8AZgcNDARE0v6GR8/IeJJrOIu+bl8vmeNrpNDi4el8S4aC8Gm4tb9+qxuFaRGK7g2pR0Bk1O9BoFMWoZEmM9eclZ6JQyChN0pESq+OlIVzCJA0zLi0EhFQtWXBo5J7pNZMRoUSsk/Hi4k/ZBB5OyolhQFM+vVT28v72J22fnIhaBQiYmKULFLV+EumHMzIvlRGs3D4xXcEV2NG8f9WFwiVhYHM8jP1Vx5aQ0XthQx8LieFaUJ7JqdDK/VvZw7ZQMnj9NfKkiNYI2o4NtfTL+kb8KzYkvObJ0PZd+04HVdQrlc8fsLG5SbECx7SmipUoixtzNW57RfLrbwUXjU2noCzV8DgSEKuvSCWmUpUbg9QWI0Sj4bF8rLYZQ9A9A84CNp1eM4rXN9Rhtbi4ek0BihHKEJsbU7Gg+39PKyookPtgtkFf8AcFE94llRVR1mZFJxIzL0KNViHlraxMTMqNCtvmXTkjjsbVVmB1e7pmfx8M/HQ8Oh78+0CFU/Ik6qrrMGGxuei1uFFKhL69VSblgbEoIOSYA6JQSSlMiBGd3EUFNEsuwrO+Z+u16tdCSuGd+Hvd+dyyojpgQruTe+XkYhyv4LTV9NPRZ/6mvpFwiJitGw6vDOH6JWMSCongyYsLIi9PyztYGbpudzWub6/H4AsglYp5YXsSPhzpYXhjOwW4XHYNOvjnQzryiOFbkqbl5tZDgV5QlkRkThtvnx+sL0D3kpLnfFlQOBZiWE01FaiRqqZguswudSkq/1cWJbjO+AwEKErSYHF4e+bGKVqNgQHHLzCwK/olkbKpezbgMwdLuQIsxRAjr77taWFmexLLSBBRnGH9cNiGVBxYVoJafPd0NWN0EAkKVPisvhrVHu0PuibpeKwuK47l9VjaF/6Zytv+xifyfhUIuJXb4giVGhv7O4/PTbXKSKRGhkgTIiw3D6QtgdXq57/tK2ox2lDIxTy4r5IafncCpamlf4wB/XZLKoFdBuNiFXOxHqVRw88wIRCJBGMjq8tJlciLDw9ojJiblpHPTJ8fQKWVMyY4iJ1rFsrJE7B5BQOvicSkkRggKf+uOdjGvOIGPdrYOD/pkPH9uKQ6PFxEiwhQSylN1NPdbsbu9mJ1ePtvTSn68lldXjeKTve04PX4uGZ9KnsbJ1R1vof7mF1JEYl4su4Yfws7n0XUn8AfAMVzZHuswkRkdxpH2Ia6akk6aXsVrF5ZR32dFLhHTOeTg6/3t6MPknJh1DzElN7K3Vzmi7/nhjlZGn7uY0pRd6Fp+JX7no1w0432WbAjjllnZZ5U8dfv83PPdMS4cm4LL6+eFjbUopGLmF8XTcsawtyQ5nPo+C7fNziY5QoXf56HX5ODBRQV8vKuFjkE78woFUa51R7sxWN0hmugNfVYe+amKn/80WVBGdHr4en8XGoWU/Hgt2bEaGob1saUSUbAlJRYxAun06/Eerp6SHhxcmpw+7piTw6YTvVxRnsaOhgFumZnNr8e7CVNIOac8Cac3wP3fVwaTzuJR8Ty2tBCX18dTy4t5+MdKHB4/Do+PP83M5PXfmxCLRPxwuDNE4rbb5GTQ7ubLvW2YnV5WlieRGKFiR30/ZSkRHDlNG+j22Tl4fH6iNXLumpfL9wfauXRiOj8c7hy2oJNz7ZRMttb18dDiAtzeAHa3l1R6eKaok1pPNFPSo9FpdXy4txulVIJW6sfl9bF4VAL9Vic/HjklqfT6RWVBqOLJ+KN+gHlF8QzaBSLcjdOz2NU4wA+Hu6jttXD11HQ+2tFMq/GUAcUrv9Xz+oVlTM+NZlvdqRnMwuJ4fqnsQS4RIxbBobbBEUPQdce6+ODyMVz58f6Q1z/d08a5FcmUpZ6REIZjXIZgDfePXS08taKYv6w5OzH98klpwZ76v1v8vy6R/49CJhGHCEHp1CoMNhcDFievXViGzeVFKhHR1Gfl5hlCn87l9TM1JxqZTMqkVw/xl6WFlCRq6TS5iFGJSdXAX39vYWud0J+ckKlnwZwkvjtk44PtFp5eUYzXF0CnkuHyeJAN1PFsVgf6+u+RrtuPYeZz+OIXct/CfLw+Px9dOYbuYWeX7fUDVHYOcfvsHPa1DFLbNcBrsRuoDJ/F+51SxqWFc2W+jyLnXiLVaXh8AjolxfAd6qZfhA8Z8BN2+H3GTytBKo7B4z9lqzU9NwaJRMT84niSIpTc/MVh7p2fx5tnSJkuKo7n3b0GZhfE4fCMdPaRSkRsanbiSL+JeW2/gd9LYv8OUvQr6TM7uXteHg//eDxY5awak8yuRgNGmxuZRIxYLAr2iQMBWFmexLpjXajkEq6alMHqw53saTJw34J8vt7Xzv7WQaI1cm6YlsmCojikEjGNfRayYzSMTg0nN07LrTOzEYuEiv2D7U0UJ4XjcPsw2j08+MOp+cXWun6eWl7MQz8epzBWyfL4QabNtqCOSee4a2T7JlanwGgTzjUqTGDQvr21kYcWFSBGQP+Eq2RMy4nG7vFR3W1md5MxJOn8XNnDuIwocuO0dA05uWxiOjqlhIxoDdXdZhYUx2N3e9l2FlJXm9GBViXDaPfw7cEOChJ0bDzRx5WT0pmaI+hr58ZpeWljHa9tFuCSErGIp1cW8/W+9uACNGB18/yGGu6Yk8vOBgPdJgdLM8VkDe3j3qYytjZbASsiEfx1ZTHHOs1UDklZVppIalQYr5zGrwBoGbCPcG4CYSH8en87pckR7GwYoDQ5gorUCJoHBPGuw6ctPiejvs/KZRPSmZ4bS22vmaQIFS0GO3/f1UKURs6R9qGzOgqFDyfZs52H2enF6fFR1WWmqd9KlEZOcWI4aoUUn8/P8+eV8Mbv9bzxez1LShJCSEAApckRRGmUIw/8bxL/n0rkZ4ZYLCJGqyRGG3qBSpIjkInFrBqdjNMrmD809Nn44PLRNPXb8CMiTqfk6o/3c9sYFXeUSrmyIgexQkO4RkWGtJfHxot44oCcu749RrxOyYOLcplhX49KbME32IXEa8E66T4Go8dicweo7xviSPsgt87MIUPnp6nHyMR0HYsK9fgCAeYUxKCSiWmUT2DcpvOoSByNXx2G/GAVGyr+xsYTfcglYm6dkkj05pEM17iBPaRFXSB4Yh7voSI1grEZkTy5rprM6DB6TQI6YUNVL3fOyeHjXS2YnV6WlCSgVkjZXNNPYoSa2P+nvfOMjqu82vb1TO+jkUaj3qtVLHcbF7BxwQUMhFBDD5CEEkIIEEryJtQUSggQaugQCPCCgzE22MZgDMa4NxXLalbvGmmKpp3vx8hjj2UgL3xBsnOutbw8c6ZtjbT2ec5+7n1vi3ZYl975U9J5Y3MjB5L0LEgeD41fMmDO5MS8eLpdfp5eX8NTF01kS30PSqWCT/d1RHw3jFoljT0eHGYt7f2DvLSxnqIkC1fNymZ2QTw3vbmTui43cwocrN7bxuYhi9XOAR/3vV/Bg+eUsam2h4mZcdy3opyr5+Ty05e3RE4a8SYtNy4oIDfeyNqKdhJj9Fw/Nw8IX6E9u6GWbQ29/PWcYmZ7VmP535vCszEVKhLnP8xJubl8XH1IGXP93PDq++JpGSRadfx1bbhssam2C5shEbNWRZ/HHykxXHtybpQc8yAHy0L3rCjHqldz3pR0Ln3+y0gSumR6BnPHOKLsCADyE8ws33lI/bJ+X0d487TVSWGimYdW7+OnJ+VEGcUFQxLPrK+lKMkSlThDUvg7mJUXR687APh42T+faQV66vsPUNvpQpLgTx9U8cezxnLjGztYUpqEVTc8bVS3DzA2xcLOwySHaTY9Vp2Kmbl2NtV2MybJwpQsGzPz4mhzDmLUqChIMFF5hHdNdryR297eRZ/HT7bdSE68iY1DG7mBkMTq8nZOzLMPmx503cl52M0abltciMcfJCPWQFXbAGvK28lzmFixqyViWZBqC8uMw2qiJrQqJVMyYzFqVZi04RPWil0tWPRqfr2okBNy4ob9zKOJ/+pE/lUcrKVlHTbzryAxXBubUxi+L0kSL/14Kit3t/LPBj+nloYHz4ZVITbyGeTJqQ3UzS5mWZWX29/ey6uOPG6frCDG7KMp+Ufcs2GAqjX13PcDCxlDDQlef5Aut4qkeBsmMYhXZeS+9yqxmzScMT6ZTvV4ms5aRmz3NgIKLavTruOmdeFOvFsXF/L8pmYKEydjbD1C3pc8jt+OKyIQkihLtWIzaHh/dytXz86mOMnChc+GL0e3H+ilqcfNI+ePD296ra+JyOaW72zm53PzuHpODq19Xhq63ZSkWNlc10PHwCBJej30OAlZ03GlnkTdFy5KUiz4QyG+qOnCORiMTHuHsIJnSqqRpqFhvncvL6fLFZ43ed7kNN7d2czi0iRcviBmnWrYlUJIgj5PgNPHJXPR3zcxNtXKil3R9c2OgUF8gRC/Wbabn8/N494VFREteKxRw3Un59HS62GKuRvL8lsODTgOBYhd+yt+sfhdJmSFteoOk5bNdV10DgwyJSuWP62sjKy0TToVMXo1v1yQz+/fDXvcCAGnpIXYkW1lfU10M5FFr+bjyg7cviAXTcvgqY/3R60kX/isnofOLeNHU9N5c0sjWrWCK2dms6m2K6omnhYbHiJxoNsTcf3zH0WX3djjZlHxcKOnwkQzTb0envm0JqqkdNviMdz3fsWQ1NNHZ/8gPxifgkmnwuMPDpMQtvV5uXh6Jlvre9mwv5NJGbbwFKINtazf18m5k9NIsGjZWNNNvDlcartvRQW/X1rMvSvKI541505OZV1FO+1D+wblrf3sX1XFT2dns+NAH3ubnNyxZAwflrdzxrgUYgxqmvu8JFl0rK1oI96sZX1VB9Nz7Tz20X48/iCXTc9kwOvjqY/3c+n0TDLiDBg1Su5+rxyjVsWYRDNzxyRw+zthV0770Gr9d0uLmDcmgcRRusF5OHIi/5YIIShLi6EsLeboT0gqw5hURjHgVHRS3trPrqY+Hi63sbA4h0Sziuum9+IwCJzd1bidCopycshOi+VAYyNPburilS8bKRna4f/98nI+2NuGRqlgYoaNa046mySjRIxXye+WhksxiWYtA94AG1VLmWNehaI/XMP0JU6gyTaV3yzbTWOPm/mF8fx0ZipXzkintsuF2xfiqYsmsqG6g1AgwMn5Nsr31yL0NmoP02/3uP28sbmRG+blkh1v4osPK1ld3k4wJKFVKTit1E6H52be74znrn+04Q9KlCRbufrEHFz+IPs7Xfxyfj7V7QMkx+gojtdQtmwBGXlnssN/BrcsLECvUdLr9vP4uv00D+mKp2XHsqAo4ai19l6PH6fHz5MXTUCnUnLXe8MnDLb0eciOD3ejHt7Q0+3y0dbnZVFpIjV160kMHSHX9Htwdjbz0Jp2hICLT8hkbWUnHf2DOCw60mLDjVdqpeDMsgQ21fexbsiW12HWkhWjJLnmNe4ozuIWr4XtzW7MWhXXnJzLezubw5keUKsUUbXwg/R7A8wrSiDPYSQ73kRIknjlsIlNyVYdcSYtVW0DbGvoYXpOHDNy7aTE6IfNTZ0/JoHseBMFCSZqO934giGWlCZi0ampGuyPkqqGpPBewPScONbv62Rmrp1lO5r5bH8XqTY9Z45P4deLCnl9UwMTMmzMLnTQ0uulscfDlCwbZ5TGkd22ikocrN/XyZwCB219Xl7/8pBtwZwCB9Nz4vjDygr+cFYpFa39lCZb0KgVXPpctKGeLxiiwGEm2aqn3xvAolMxvyiBOKOGihYnX9R0YdCoOLnQwQOrKrjzjFJ+9MwXkdff9V45vz21iKtOyuGe98KLheJkCw+eU8aWhh6Kkiy4fUGevngiWpWCjyo7CAQlsuONI2KM9W2QE/n3wMQMGzfMy2P7gT6aet04LDomZcWiIxa6q8FihJhMMIWNTtNSU/m1QcXZBSp6vRKJul5+dXI6T2xoRqUUzCmMx2rUcuGLm7lt8RgqWpzYjBoqWvvJcxgRIo09C/5BSqiJ5v4AK1pjeO61FhaXJnLz3Axm1z6I4cWXAIgd/2Oqxt9GU3sXv8jrxOzcRzAQw/Qx6ewTVnRaDXe8c8g/vM3pJbvvc9y6ZK6alc2uJiehIcncw593UJBQGOlM1akVxBjU7G3tZ94YB/3eAE98vB+7SYMqZOACxZeIvnrUnXsh4UzKW/tJjtHT2T8YJcvbWNPN3DEJ/HrRGH71xo7IBuTsgni6Bwbx+o28u6WZkCTxk5NyeHzd/qiuxKlZcSRZdWw4Ys4jQHOfB68/iFsRB2r9ofmmEJZa+mOAAcalxqBXKyIKk9IUK8UOLUG3gtlxfbh8/Tw6pMOflBnLvrZ+7tjdSrxpLL+ZpubZnLW0l2TTlTKfX761m/b+QW45pZCVu1vp6B8kPdYQVYLRqhQkmLX8+PkvuWJWNg+trubO04q4clZWZPXa7w3w6Np9XDEzm7JUK2admj+vqqQgwczNpxTw/Gd1dA74WFSSyPlT0tl2oJeiZAuXTM9Ep1KypaGHPc19Rze9GlLUzClwMCEjJqICaezxoFIInvy4micunMTGmi7ufHdvVKfvX84qJNPVjTbex1/PK6PH7R/mxPlRZTs3zMtjXVUH5a39PLq2mrvnOUiy2zBolFFdv8XJFnY39/H4x4dkj2eOT2FiRgyPfHSoF+HLum7uO7OUUCjEzBw7n+4/tMfwv9saSTDrIiXBjv5Bmvu8ZMWFT2zNvR72tQ9Q4DBw9uR0qtpc7Gnqx+uXSI3Rkp/41S6Lo4Fjwsb2eEGSJCQpXJv/t3C2QsMGOLAJKWMmrfapKHRmEoZakzfXdfPHlZWYtEqWliWTatMzJtGMyX0A/IPs9jmo7RmkuddLS5+HaZkxTOxaTvzHt4TjsWWzpuwhrljpYmyKhQumppNmhqQYI+6gErcvRLxZTUWbi401XcSbtExOM1HgL8fgaaXRNokNbWoq2waw6NSoFIL0WAPVHQP4gxJxRg1vbW3kN6cWcdfyvZw5PpmCBAtKhUDj7SS9fQ2OgQpei72a3606VHLJthuZlR/PC0P+GgAPnD0Wu1nL/vYBetx+1EoFe1ucTMywce+K8qjV5wNnj+Xu98rRq5X8bHYOOfFGdjc5GfAF+Oua6PLMnacXEwyG8PqDnGXeg2PNL8DTA4Y4+hY9RgVZ+IMh3q/x88qmsCa/MNHMdbNSWbDzetRdFeBz8/L09+kO6jFowjX/5w+LXQh4a1GICWsv5KMzNnPZa+GkOCHdxtwxDt7f1cIFU9N54bM6KtsGSLTouHFBPvd/UElRkpXCRBMOi442pxe3L0iqzcBrmxqoGWq7v2VhARPSbfzuX3s4rSwZly+ISafCbtSEvXwkiT+srIw6uf1oajp2k4a/rdvPTacUcN/7FVHf4W9PLSIvwcQja/axqS569NvfLgiX3XY3OUmL1bPjQB9vHNavkGrTc9+ZJdz1XjltzkH+57ThdroHP+PO5Xu5ZWEBf1xZyasLBbm+St7SnMb9H1QRHBqu8fgF47n61W3DZJYHXwfhZH/p9Ew+qgzvFc0pdNDZ7+XeoXFys/PjmZYTx4pdLdR0uLhjSSHBkESSVY8vKPHuziacbj9nTUxjTUUb7w75wAgB9/9wLLOyY3HYvtmo6z/NV9nYyon8GMc1GMA1GCDOpI1MSfpavH3QuQ8CgzQrk5j/dNWwy/q3fnYCu5r62NXYR3qsgcIEE5/XdJNhN9LtGiQYlEiN1ZOfYObnr27HL0k4PeHk+viPxhOSwiuejoFBnJ4ADrOagcEgHQM+PqnqQAB3nVHCHe/s5pzJaTz+0f6osW4AN8zLi/iYl6ZY+MGEVO5fVcWdpxdz29u7GAyEKEw0kx1vjLTCH2RuoYOrZmXx1rYmVu5p5ZaFhehUCnYc6EUoFLyx+QACOG9KOifkxPL7ZXtZOj4FhYAJFie5Rjd2fxuhPcuoL/4ZRe+dSfOkW9inKUSKy8ek15K/8VYste+FP9AQx1snf8RNb+3mohMyeXtb47Cu2nvnWLjgi9NpXPQiN26N44sht029Wsn5U9KYnhvLB3vaiTfr6Pf6mJIZR6xJQ0OXG7tJyzWvbo3UpBUC7lhSxBPr9nPhtHRUCsEJuXa21PVwz2EOg3kOE3efUcKXdd3c/0G0yuRg48sja6vJTzDxw4lprNjVTL83wMKSJFr6PCwpTaLX7eP1zY3YDGqKkq2k23Rsb+zjhc8OnXhPyo9HrRSRiUEmrYrzJqfxzFBT0V/OLeNv6/ZHWVYUJJj5+dxcWvu8tDk9JIhezm17EJOvg9tiHyDBZsUXlDghOxZ1/wHO/efwzs1fLcjn/g+qUCkEv1taHHXlqFIIfnNqEdXtA1S393P93Fy6hmwQrHoVnQM+mnq8KBVw74qKqH6Em04p4JG1+yJXKtl2Iw+dW0ZZ2tHli98n/xE/ciHEn4HTAB+wH7hMkqTe7/KeMv83jFoVRu3/4deos0Jq+O/ANhjg53OlKD/0C6amke8wU5xsZVLGAC19HpKten6abqOiqYdkk5IEqwG72YCExJ1nlFDb6UKvUaJSCq55dRsLS5JIsGhp7HYzJTOGwkQze1vCjStLxyWTYNaybHsjN8zPx2ZQ4z6Kt7NVr2ZsqpWSZCvzihwAJMfo+MPKCv501li8/iAOizaycjocIaCx10Ovx8+PZ2ThD4R4eWM9505O463NjfxoagYCCbNeTWuvl7lFCSzf2Rzl53LNzDx+2VdNquiAoI/kL+4iGWhY8Heq3GloOFSrr5/1APe+X0VIgu6BQRxmHU5PtBLDaLHRcOKDCHsBP5ttZWp2LJ9UdXJCdiwzcu1c/Owm7CYtM3Lt9Lh8rNlbwezCeNZWtHNifnzUxmJICk/6+Z/TxjAYkPiwoo0DvV5Oyo/npLz4yHDxfe0DNPZ6Il3Ah3NwtQvhhpc/r6rg2jm59Hn8bKzpZE5BAje/uZNut49zJ6VxcmE8K3e3kRFr4KUjdP4fV3Vww7y8SCI/Z1Iqq8sPTXGq6XCxZGwyJV0udjb2UZZqJT3WQI8r7DPz6lnxFK29FkVPDV3jrmZ9VT8HtrQzOdPGD8YYMVd/SHrsJBq6D5Xb7CYNWfbwCnlyZizvDamEVArBguLwfoBEWO1VlmblyU/qWFfVjlGj4vq5efS6B+kfDBKSGGaO9/7uFmbk2lkz9PN09A8etfw0mviuNfIPgVuHxr39EbgVuOW7hyXzfaDXqsLdlWkxNHS7SbLqKU62RJoeSlKsURNYEq16elw+1lW289qWZty+IP86zA/9yllZXDc3b2hotZaceBPTc+0MBqE0TcnDq/expqIDvVrBX84bz9/X1+Dxh1hckhjVEWjQKAmEJGL0ajQqBa98Xs/Vc3K46ZQC1ApB/2CAbrePsaZe5uTH8s72pshltxAwLTuOGIOG60/OY+XuVv64soKLTsjE5w9x7dw82p1eOgYGWba9mSWlScQY1MN86p/+vJWzZ19Oquew1b7exi5fErs9Vt7lFyyZ9RNCkqA9lE2XK3wy/GBvG/f9oJRfvbEjElO23YjRame7fx4vrWpgYYmSGdl26jvdLNvejFKp4KyJqZi0Kj7cG7YWvnxWZrjbUqUYNsSgMNHMzNw43P5wU9VBlm1v4pHzx7OvfSAyXKLd6SXBrI2avQowv8jBlvpDJRN/UKLX42fZ9mZ+PDOLP6+qRK9WctOCAuq63Lz8RQOLipPwBYNH7STVqZUkWnScPi6ZomQLz26ow2HWcs7kNLLjjTy8eh8IQX6CeWhwdAdnTUgl06YlrfwZBmIKqJj4J+qUmZxjgX1tA0zJshFQqLF2fMkTM4q4b28cG+t6KUm2snRcMvva+nnm4kmoFLBsRzN6tZKbFxbw5pZGVuxqJd9h4tbFhWwbGlQN4T6Le1aUc//ZY3l7WxM5hynTDqIUIqoRbHFpEjH64X49o4nvlMglSfrgsLsbgR9+t3Bkvm+MWhVTs+OYmv3v6WTf29XCHe/sjrjRHc7T62u5cUE+u5v6WF3exk0LCsiwH5o+fs8ZpSys7qS208X9qyqZlGnj3MnpqBWQaTeybHsz2XYj501J49N9nZxUEE+cQcOikgQQApNOhU4lSDCriTdpUCjVzGhdxrMXnsG/dncyGAgxI9dOnEHJCxvq2Hagl9sWF/DyFdNQBz0kKHroDKqoaPFxoNtDa58Xk1aF2z/cXCwQChFSalEZdRCXQ9BRinPitdz0YhdTstQ4PUqu+jCsgFlS2hdp21cpBC29Xm5cUIDHH0StVOD0+Pnbumoum57Fl3U9JFp01HW5WTZ0ErQb1fS5/ZHZrI09HnY19XHPmSXUd7m5YEo6y7Y3U5Rk4bqTc6lq66emw8WX9Y1RMbt9YQvgGxfk87d1+6ntdFGQaEajVPDQOWN5c2sTFa39nFzoYGyKlZveOnQSSLHqmJlrx6xToRoq0f14VhZPr6+JuAF+UtXJ1bNzmJ1/aNUPkGTRMTkzlimZsby/u5Wgu5dlS5UkatxU+rvZ71Jz9Zxc3t7WxNqKNoqTrVx8QgbPbajlwdOzUfafxMONuTy9vBWoQgj43WnFuLwBnljfxCX511B24CX+OOFsehZMoMMj0KiUvPh5LX9ZU41GqeAnJ2bzy/l5PHnYsOmq9gFu+OcOLjkhc9jvt7HHQ2mylRiDZpiU8rIZmTz5SQ0WvYolpUn8YHwKCvFv7muNEP8/VSuXA69/1YNCiKuAqwDS09P/P36szPdFt2uQJz7e/7XPSY7RM6fAwcw8O5MyYqMeS4zRs3hsEjXtLmYXOMi0G7Dq1Gxt6EGhEPz85FzizVr2NDuZX+RgXJoNqyEs/6podVLX6UShgKIkCw6rGveAH3NGGTNrn2e2oYOenB/gMsVw7b8aaHX6OGtiKtsbnVh0GoSALrWdpBgdkzN9WPUq5hU6MGiVqJWKYU1O55bFkpwg4NM/EpjyM/bHzabRb0Gn7uWTqg7uWFJEc6+HVqeXD/a28sDZZTz5SQ1NvR46BgZ5/rO6KAlgTryJVXvD5YY+jz+yYgbodft5d2f0JJtASOJAjweFgLe2NvLb04qwGzXsbnby2EfVLC1LPuqglBDw+pcNXDAlnTiTht1NTv65+QAOs5YrZ2VzYr6dD3a38fdPa7l8ZhZKIcKDNuxGbnpzB90uP79eFG6WUClEJIkf5IXP6rj/7DKMWhWb67sZmxLD/OIEzn96I6k2A3fOczCp6mG0e8KpIEFrJmb2s/zwnQBTMuO4YmY2s/Pj2VDTya2LCpn0xTVsTb+EpzceuvKRJHjgw0oeOW88le0D7Axl8Xn8bRgGVTz/RgW1nS6MmrDRWn23h/KWfh75qJrHLhg/zISs1+3HqB2+mk6w6Gjp9bCvzcmNCwqoaHXi84c4s8zBeGMn+T8sYzAYwqJVIgG5CaPTY+Ug35jIhRCrgcSjPHS7JEnLhp5zOxAAXvmq95Ek6SngKQhvdn6raGVGFKUIz98E2FDdxaKSxKiNxhy7kRk5cZw1IfUr38OoUVGaeqhc81FlO5c9d8gbQ69W8vpPpjE2NSbqdYWJFgoTLTR0uXhodRXLtjdj1Kh48Jxx5OamohEh3Fo7y3a04rDoyU+0smpPK90uH2dPTCEQkvh0XyeiQXBqWRL9Db3c9/4OBgMhbj4ln8cuGM8725vZ0+xkSUk8pYlGPm71kH/iA7xQruD5t6tIjdHz0DllvLujmeU7m7l9yRi0KoFKoeBv6/aTEWfglOJEipIsvPh5XVQJYlFpIi8OKVm0KgWTMuLZ2tALQEOPB6teHbEpOEiiRcvVc3JRKQT5CSZW7mqNdGauq2znshlZPLzmkCGZRqmgJNnCo2urOXN8Krua+nhmfXjDsbHHw7X/2MZNCwqYlBXLqr1tkQarGIOay2dkRawH6rtcTEiP/v4PIgFbG3qp7XQxf0wCkzJt3PDPHREjtAyfM5LEARjsZ8z2uzmz8B5e393Jp9WdaFSCwkQzHn+ITTMep7d+D4f7GkHYo7zbHXZndPkC7G5yUtvpigzIcPmC/GX1Pm5ckE95S1iN4wtKw4zUhIBsuylqIEZZmpVEq5ZgKMS4NCs6jYqZuXEkWbW0On00h8xkxBpGra/K0fjGRC5J0ryve1wIcSlwKjBXGgkJjMz3htWg4Zfz87nm1W1sqe/hvMlp/Gx2Drub+piSGcui0sT/Uxecxx/ksSPKMx5/kM/3dw1L5AChkMQrmxp4e1t49do/GODKlzbzyhVTmTHkO32OWs17u1oitsFXz8mNXBlMyz7kTZ3nMHNqaRJuXxCVUuD2Bfn1wkL0GiXdLh9NvR60WWVs7nQxo0BDaUY8PS4/Bq2SU8cmEQhKeP0B+juamazax42T8vi400AwFCJR5+OBs0t5dVMj/UPmVpvre3B6AzjMWs6fmoE/EGRRSdjR8sM9rdyyaEyU6iLJqqO1b5BHP6rGolMxNdNG58BgpEHF6Q2wraGHX87P57P9ncQZtZw6NikyntCgUfL21kOmVhDe4PT4gygV8JfzxrHjQB96tYK0WEOUUdQ/Nh3gqhOzKUm2EmOIPsFcdEIGCWYt509JByHxm3f2RMkWle7hg0/U7bsozTp0ue4LSjz/WR2banuYkB7DzfNLUCu3RG3KZtuNSBJ8uq+TkGQnK97Ie7uGb2wfvgnp9Pi5+ISMKJ+Uy6ZnEmtU88j54+ga8vZJtoZtJtQKgUatwKbXkDe04rYZR6+fytfxXVUrC4GbgZMkSRpuKCFz3DGn0MFLl09h5Z5WEi065hc7uGVh4bd6r1Ao7Lh3JF81obzb5RuWnCDs4jgj105Fq5N7lpdT0drPqWMTOWN8yldKxhQKEWXBcDhJMXqSYvS8ueUAN70ZtjpItOg4udCBPxhiaVkSvkEvSd56VE1vISnVKK1pfFLlQaNS8vqXXv50Vik2g4ZYo4adB3qZmRvHWRNS6HAOUtnipK7LjcOs5eHzxofHuUkSj54/nr0tTuLNWroGfJGk7PQG2NLQS2ackXiLlo01XfiDEp/s66S81ckTP5rIox9V87NXtgJw/uQ0QoSVP0dOro81aghJEje8vh2VIuwkeMWs7KgGHIBet481FW3cc0Ypm+u7qW4fYHZ+PAODARRC0NTrRiHEMNmo3zy8bOrOmMvaA+EkrVUpKEw0849N4Q7VrQ29VHS4efTcEm7/VyWdAz7yHCbOn5JOTYcLly9ArFFN18AgyVZdpNv3IAevEGfm2ml3ehEIHrtgAt2uQcw6FRadmj+trOSy6emcXpKATn/IlzzvaHWGY5TvWiN/FNACH4rwZsBGSZJ++p2jkhm1GDQqZuXHMys//ju/l1Gr4icn5UTZnyoEzMo7+nsbtEryEkwRH46DpMTo6HX7uPGfOyLufs99Vs9rXzay/LqZ5DiOnrC/iYbDlCytTi+vbmog1abnvCnppMQagTjImgqAxevnOjrZ0xIe9P3cZ/UUp1hJMGvRqZXoNUq2H+jFoFGxv2OAiekxdLp8VLf3U9k6wAd7w3r3mo5+xiSaI34tB3nhs3ruPL2I2k4Xty0eQ6/bT4xBTbJVh0JIXDUriwkZNtJjDby8sZ6s+PCm8b0rDklLEy1aEixa/EFp6F84eTf3epg7xhGR2+U5TOQlmHl/VyvvbG/iwR+WMX+Mgz+8X0FQAoNawUXTMzFqwpuBB1fKM3Pt7CUO6+y7sX56NwS8hBJK6ZtxG5qNEmdNsDA+PYaGLnfUKn9Xo5P0EjuLS5OINWqYmGHjyhc2kxprYHFpEiqlgs11PVw+pKY5uDF50bR0MmLD5S6TVkVlWz85DiMpMTrijGpSYnTYjFqm5cR9pRf58YLcECQzovR5fHxS1clzG2qxGTRceWI2kzJsXzmlfEt9Dxc+80Wk3lmaYuVvF07A6faz5JFPhz3/75dMYu6Y4WZR/w6fVHVw8bOboo5dPzePX8zLQxxFxXDvir089Ult1LHMOAPPXDKJMx/7LLJ6HZtq5ZxJqQghePGzeix6FZdMz0QlYMAXIs6o5scvbI6qsccY1Dx10UQGAyHanV5UCvCHiEz7+fv6Wn57WhHv72rFqFXx1tZGblqQT7xZR3XHAGadKuyJo1aQGqNndXk7q/a0RbxnTilOiKyCy1ucvLW1kZAUHiBR3+Vib4uTh84dx7Of1rK5voeSFAvnTExDpxHYDFoUQtA1MMjynS0c6HZxbZmCCUlqlterSU1OxqBREpKgxzXI7cuiyzF3n1GC3ajip69sB8J7LVecmE1jj5scuxFPIIQvEMKoURJn0uL0+IkzaUmN0aJRKzBq1Ni+ZgLQ8cR/pCFIRua7YtVrOK0smQVFCSiEQK06egI/yMQMG8uunUF12wB6jZIxSWYSrXrqgq5hE94h7DL4bRmXFsPvTivi/g+qcPsCnDUhlR9OTDlqEodwyz1EJ/JTihPJtpv46/njuf61bTi9AarbB7AZNCwsSeL0smSCkkRTj5det494s4K3tjZy9Zxcnvx4P/6ghEGj5J4zSjDrVAhveLpVvzdIc5+XPIeZFz6rx6BV4fGHx8HZjBruPL2YPo8fvUbJhurOiDTy+nn5rNrTRnX7AJdOz6Clz0tFSz9Ly5J5YUMdyTY9iVYdF03LwGHWIknQ6vTgD4a4+c0dvHT5VFZXtFPd3s+dy/dyyfRMbAYfXn8okqxLU23Epqdw9aoKylt6gQ5MWhXXz8ujx+VjYrqNzfU9CAFnT0zDpFMSlMJj4+q7PezvdPHQh1XcsWQM7f1epmXFoVQKelx+Um160uNGvlV+tCGvyGWOCyRJ4tVNDdz+9qENw3MmpXL7kqLIwIFvS2OPG38gRLJNP2x6++F0DgzyxLr9PLuhlpAE03PiuOfM0kgHYn2Xi/Z+L3ajlky78agnhEF/kO0HetndFLZHUAhBtsNEZpwh8vw2p5eugUH6PH621PegUSlIsemxm7SMSTBT3tofnmFr0vDchrrI5vC1J+fy0uf1UQ6Ql07PxGZQ88jaahYUJ3BKcSJJVh2uwQA7m5ysKW/DoFEyb0wCr21q4OLpmayr7GBtxaFNzfMnp7F4bBKBYIh4k4Zl21t4cWM9P56ZhValINakITVGz13vlVPT4WJpWRKLSpJwDQbocfkoTLZQmGima8DHnmYnbl+AnHgjhUmWr521+d+I7LUic9zj8QXY2+KkrtNFgkVHUfL3nwgGA0HqOl34g9L3ImFrd3rpdvmIN2uJMw3/WTdUd0YsXa+fmxclVwRQKwVXzsrmb+vCG6vFyRYeOHssW+p7I/7cEN67uP/sMuxGDd5AiD3NTj6qbGdsagzTsmKpbOsnz2HigQ+rmJlrx2HREQyFKEuNYW1FO//a3sy1c3OJN2lJtelJNOtQqxQ4LLp/zyNIBpBLKzL/Beg1KiZmxDLxiEak7xOtShkZQvJ94LDocFi+WjI3McPGc5dO5qlP9h+1MUalUETprqdkxdI/GODVTQ1RzwtJUNHaz+KSRP68bE94pKBNz/p9HSRbtRGfeYNGGfFMz4wzcNrYZH61oIDTypJxevxk2Y3kOkxfWZ6S+XbIiVxG5jhGp1Yyp9DBjNw4mnu9vPhZHY29hyR8F05LZ+XucFNXSoyO8yan4RoMoFcPT/p6tYKCRDOXz8zknvfK+bwmbB+wdFx43yAvwcwLl01hX/sAIUki12EiaaivYNq/aQEh8+2QE7mMzH8BGpWSTLuR5y+fytqKNqpa+5lblEBxkoU5BQ4CoRC5DjPJMXoOdLu5ZHomWxp6IuoSg0bJ9Jw49BoVZ45PZUaOHV8wRKJFF6Uw+qYrBJn/DHIil5H5LyLXYSL3CF39kSqQtFgDoVCIpy6cyPp9nZh1Kk7Mj48yVpOT9ehCTuQyMjLDyLCbyLCbmF98HLU/Hsd8vWhXRkZGRmbUIydyGRkZmWMcOZHLyMjIHOPIiVxGRkbmGEdO5DIyMjLHOHIil5GRkTnGGRGvFSFEB1D/PXyUHej8Hj7n2zLa44PRH6Mc33dntMcox3eIDEmShhn2j0gi/74QQmw+msHMaGG0xwejP0Y5vu/OaI9Rju+bkUsrMjIyMsc4ciKXkZGROcY53hP5UyMdwDcw2uOD0R+jHN93Z7THKMf3DRzXNXIZGRmZ/waO9xW5jIyMzHGPnMhlZGRkjnGO+0QuhPizEKJCCLFTCPG2ECJmpGMCEEIsFEJUCiGqhRC/Hul4DkcIkSaE+EgIsVcIsUcIcf1Ix3Q0hBBKIcQ2IcTykY7laAghYoQQbw79/ZULIU4Y6ZgORwhxw9Dvd7cQ4h9CiBE3GRdCPCuEaBdC7D7sWKwQ4kMhxL6h/22jLL4RzzHHfSIHPgRKJEkaC1QBt45wPAghlMBjwCKgCDhfCFE0slFFEQBulCSpCJgGXDPK4jvI9UD5SAfxNTwMrJQkqRAoYxTFKoRIAX4OTJIkqQRQAueNbFQAPA8sPOLYr4E1kiTlAWuG7o8UzzM8vhHPMcd9Ipck6QNJkgJDdzcCqSMZzxBTgGpJkmokSfIBrwGnj3BMESRJapEkaevQ7X7CCShlZKOKRgiRCiwBnhnpWI6GEMIKnAj8HUCSJJ8kSb0jGtRwVIBeCKECDEDzCMeDJEmfAN1HHD4deGHo9gvAGd9nTIdztPhGQ4457hP5EVwOvD/SQRBOigcOu9/IKEuUBxFCZALjgS9GOJQj+QtwMxAa4Ti+iiygA3huqPzzjBDC+E0v+r6QJKkJuB9oAFqAPkmSPhjZqL6SBEmSWoZutwIJIxnMNzAiOea4SORCiNVDdb4j/51+2HNuJ1wyeGXkIj22EEKYgLeAX0iS5BzpeA4ihDgVaJckactIx/I1qIAJwOOSJI0HXIxsSSCKoTrz6YRPOMmAUQhx4chG9c1IYb30qNRMj2SOOS5mdkqSNO/rHhdCXAqcCsyVRodwvglIO+x+6tCxUYMQQk04ib8iSdL/jnQ8RzADWCqEWAzoAIsQ4mVJkkZTImoEGiVJOngl8yajKJED84BaSZI6AIQQ/wtMB14e0aiOTpsQIkmSpBYhRBLQPtIBHclI55jjYkX+dQghFhK+BF8qSZJ7pOMZ4ksgTwiRJYTQEN5k+tcIxxRBCCEI13bLJUl6cKTjORJJkm6VJClVkqRMwt/d2lGWxJEkqRU4IIQoGDo0F9g7giEdSQMwTQhhGPp9z2UUbcYewb+AS4ZuXwIsG8FYhjEacsxx39kphKgGtEDX0KGNkiT9dARDAmBoNfkXwmqBZyVJumdkIzqEEGImsB7YxaEa9G2SJK0YuaiOjhBiNvArSZJOHeFQhiGEGEd4M1YD1ACXSZLUM6JBHYYQ4vfAuYTLAduAKyRJGhzhmP4BzCZsDdsG/A/wDvBPIJ2w/fU5kiQduSE6kvHdygjnmOM+kcvIyMgc7xz3pRUZGRmZ4x05kcvIyMgc48iJXEZGRuYYR07kMjIyMsc4ciKXkZGROcaRE7mMjIzMMY6cyGVkZGSOcf4fldUiYVt7MUAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pca = PCA()\n",
    "pca_components = pca.fit_transform(X_train_filtered2)\n",
    "sns.scatterplot(x=pca_components[:,0], y=pca_components[:,1], hue=y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "ad202938",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:05.526727Z",
     "iopub.status.busy": "2021-08-13T07:17:05.525821Z",
     "iopub.status.idle": "2021-08-13T07:17:08.092361Z",
     "shell.execute_reply": "2021-08-13T07:17:08.092840Z"
    },
    "papermill": {
     "duration": 2.631052,
     "end_time": "2021-08-13T07:17:08.093002",
     "exception": false,
     "start_time": "2021-08-13T07:17:05.461950",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAAD4CAYAAADxeG0DAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOyddZQcZdaHn2q36ZEed9dMdOJOlAR3CW6Ly+IsLMvCLru4LguLB4eQIEFCSAJxnWQy7u4zrdPe9f3Rk046Exb4yMLC9nMO59A11W9VV6dvvXXf3/1dQRRFQoQIESLErxfJL30CIUKECBHipxEK5CFChAjxKycUyEOECBHiV04okIcIESLEr5xQIA8RIkSIXzmyX+Kg0dHRYnp6+i9x6BAhQoT41bJr164+URRjDt/+iwTy9PR0du7c+UscOkSIECF+tQiC0Hyk7aHUSogQIUL8ygkF8hAhQoT4lRMK5CFChAjxK+eo5MgFQbgRuBQQgTLgIlEUHUdj7BAhQvxyuN1u2tracDhCP+efE5VKRXJyMnK5/Aft/5MDuSAIScB1QKEoinZBEN4FzgJe+aljhwgR4pelra2NsLAw0tPTEQThlz6d/wlEUaS/v5+2tjYyMjJ+0HuOlmpFBqgFQXADGqDjKI0b4leO2+tjyOlBr5YjCAJVnWaqusxIJRJGJYWTEa39pU8xxL/B4XCEgvjPjCAIGAwGent7f/B7fnIgF0WxXRCEh4EWwA58KYril0c4ucuBywFSU1N/6mFD/Aqo6DDz4sYGdjUPsnhUPHPyYrno5R3Y3V4AYnRK3rhsMrlxYQA099vY02Kkz+pkdFI4o1MiUMmlv+RHCAGhIP4L8GOv+dFIrUQCJwIZgBF4TxCEZaIoLj90P1EUnweeBygpKQl55/7GaR8c4sKXt9NjcQKwqa6fpj5bIIgD9FqdrK/qwaBVUNVlpqLTwobqXjbW9QHwzDnjWDo68Rc5/xAhfk0cjdTKfKBRFMVeAEEQVgDTgOX/9l0hftPU91oDQRzAoFPQbhy5YBYbrmLZi9uo7LQAsKAwjlPHJ/HB7nb++lkVGdFasmJ0KOVSXB4fDX1WesxOEsJVZMbokEpCs8UQIY6G/LAFmCIIgkbwPw/MAyqPwrghfsUopMEpkd3Ng8zMiQ7alh2rY3tDfyCIA6yp6CYvPoyHTivm+nk59Jjt7GgaYMWuNlbuaeOM57ZwwcvbOfUfm/m2tpdQY5RfngsvvJD333//lz6NH8wrr7zCNddc86Peo9Ppjrj9nnvu4auvvgJgzpw5gYr1JUuWYDQaMRqNPPvssz/thH8ARyNHvk0QhPeB3YAH2MNwCiXE/y45cTqmZEaxtWEAALPDQ2qUht8vzOWfGxpQyiTcvjifP39aMeK9HUY7FR1mlDIpkVo5r25uxieKXDwjnSfOGovbKxKullHVYWG3SkpKpIZYvfrn/oghjhJerxep9OiuhXg8HmSy/7wDyX333XfE7atXrwagqamJZ599lquuuuo/eh5HpSBIFMU/iqKYL4riKFEUzxNF0fn97wrxW8agU/Lw6WN49IwxXD4rk38um8CUTAPTMw28dskkVlw5jbn5scw6bJYOEKlVsKaym4QIFc9taMDu9uL0+PjH+gZ2tRh5el0t9b021lb38NAXNWypH2BzXS8PfV7JuztaqOwwMWBxYnN5foFP/tvntddeY/To0YwZM4bzzjsPgG+++YZp06aRmZkZmJ2vX7+e4447LvC+a665hldeeQXw+y3ddtttjB8/nvfee4/09HT++Mc/Mn78eIqLi6mqqvrO4997772cd955TJ06lZycHF544YXA8WbOnMkJJ5xAYWEhDoeDiy66iOLiYsaNG8e6desCY7S2tjJnzhxycnL405/+FNh+0kknMWHCBIqKinj++eD56I033khRURHz5s0LKEq+62kkPT2dvr4+br/9durr6xk7diy33HIL559/PitXrgzsd+6557Jq1aofctn/Lb+IaVaI/w2SIzUkR2oAaOyzcc2bu9nfYQZgdHI4T5w1lvOnprO1cYDabisAx46Kp23QTka0ltJW44gxdzQOcObEVO78cH9g29aGAf56SjFJ4WrMDg8b6/oxaBUkRKjotTiRSARGJ+lJNRz58TjED6e8vJz777+fzZs3Ex0dzcDAADfddBOdnZ1s3LiRqqoqTjjhBE477bTvHctgMLB7924Abr/9dqKjo9m9ezfPPvssDz/8MP/617++87379u1j69at2Gw2xo0bx9KlSwHYvXs3+/fvJyMjg0ceeQRBECgrK6OqqoqFCxdSU1MDwPbt29m/fz8ajYaJEyeydOlSSkpKeOmll4iKisJutzNx4kROPfVUDAYDNpuNkpISHnvsMe677z7+9Kc/8fTTT3/vZ3zwwQfZv38/paWlAGzYsIHHHnuMk046CZPJxObNm3n11Ve/d5zvIxTIQ/wsfL6/KxDEAfa1mfiyvJsrZmfxr/NLaOi1oVVKQYQznt9KnF7J6KSIEeOMT4tk87Cq5VBWl3Vy8Yx0Lnt1Fx6fP2+eHx/G2ZNS+PMnlWTFaHnuvAlYHW6MQ258ol/+mJ+gRxJaMP3BfP3115x++ulER/ufpKKiogD/TFYikVBYWEh3d/cPGuvMM88Men3KKacAMGHCBFasWPFv33viiSeiVqtRq9XMnTuX7du3ExERwaRJkwJFNBs3buTaa68FID8/n7S0tEAgX7BgAQaDIXDcjRs3UlJSwpNPPsmHH34I+GfttbW1GAwGJBJJ4HyXLVsWONcfy+zZs7nqqqvo7e3lgw8+4NRTTz0qKaBQIA/xs7CxbmRxw+b6PmbnxnDZ6ztpHbATrVPwzDnjmZgeyY6mQRIiVMTolPRa/Zm6SI2c1EgNnUdQvyhlUlbuaQ8EcYCqLgt2t49nzh2HxyuyfEszsXoleXFheEXY3jzAXz+voiBBz0ljEylMDD84oMcN+ECmPOrX4reIUnnwOh1YgJbJZPh8vsD2w8v8tdrgYrADY0ilUjyef58WO1xnfeD14WP+mPevX7+er776ii1btqDRaJgzZ853WhP8FG39+eefz/Lly3n77bd5+eWX/9/jHErINCvEz8LCwrgR2+bmxXL1m7tpHbAD0Gd1cf5L2/nj8YU8fuZYfD6Rp84Zy8OnjeZvpxZz1Zws7vukgqIkPbJDZtESAZYUx/NNzciZeqxWwYDNxetbm5EI/hm9V4Rus5OkCDXXz8vmy/JOzvnXNva2DiJ6XND4DbxzDrx6PJSvBLt5xLj/qxxzzDG899579Pf3AzAwMPCd+6alpVFRUYHT6cRoNLJ27dqjdh6rVq3C4XDQ39/P+vXrmThx4oh9Zs6cyRtvvAFATU0NLS0t5OXlAbBmzRoGBgaw2+2sXLmS6dOnYzKZiIyMRKPRUFVVxdatWwNj+Xy+QC78zTffZMaMGT/oPMPCwrBYLEHbLrzwQh5//HEACgsLf/RnPxKhGXmIn4V5+XFsru/ni3L/Y/exo+IZnRxOfa8taD+nx0ef1cVJ45IC28QMkcZeG/1DTp48aywKucDLF07k430duL0iBQl61lZ0ceLYJF7Z3BR4n0wiIJVJ2FzRy5y8WMx2Nw09VhAEOkx2eq1ODBo5D502ht+/u5d9bSZaBuzEKtNRj/8zHSYXviElse1WBpw20gw6cuN0/9OVjkVFRdx1113Mnj0bqVTKuHHjvnPflJQUzjjjDEaNGkVGRsa/3ffHMnr0aObOnUtfXx933303iYmJgbTJAa666iquvPJKiouLkclkvPLKK4FZ/6RJkzj11FNpa2tj2bJllJSUUFxczHPPPUdBQQF5eXlMmTIlMJZWq2X79u3cf//9xMbG8s477/yg8zQYDEyfPp1Ro0Zx7LHH8tBDDxEXF0dBQQEnnXTSUbsewi+hwy0pKRFDHYL+97A5PDT22xAESDdo6bU4WfLktwy5DlZ7CgJ8dPUMYvVKtjcOsKt5gNHJEaQbNJz5/Fbc3uHHdqnAixeU0GG0o5ZLUcgkqGRSPtvfxYd72onSKrhtcR5l7Sb6rS4+3teJXiXjunk5PPxlNQ63/5F/2eRUpmQaiNMr2Fw/wGNf1fL4mWO475NKBmwuAOL0Ss6dnMbTX9fx0oUTmXEEpc1vlcrKSgoKCn7p0wji3nvvRafTcfPNN//Sp/L/YmhoiOLiYnbv3k14ePh37nekay8Iwi5RFEsO3zeUWgnxs6FVyRiVFE5RYjhapYw0g4YHThrFoWuNNy/MIzVKzcNfVHPtW3t4ZXMzN727l0fX1HBMfmxgP49X5MVvG3lvZzuiCK9sbmZ/hwmvT+TBU4pZUBjHij1tZMeG8fG+TgBOGJvEC982BII4wPJtLfhEEY9XJFavIt2gYUNNXyCIgz8N02NxEq6R84dVZfRbQ+raEP8/vvrqKwoKCrj22mv/bRD/sYRSKyF+MQRBYOnoBPLiw2gbtBMbpiInTkdTv433drUF7buxrp8b5+cEUjMACpmE00bFs6fVSFaMv5T/i/Jueq1OJmVEEalVEKM7uAgXoZHTbR4ZhC0OD4M2F/W9NhLC1bQODI3Yp8NoJzZMSXmHGZvTg0EXWgT9T/Pyyy/zxBNPBG2bPn06zzzzzC90Rj+d+fPn09x8xLabP4lQIA/xi+H1iYgiFCaGBylG3F7fEfc/IEiZmB7JpAwDE9MjueqN3YHUzFvbW/nneROo6bbQY3ZQnBROSqSK3DgdNd1WukwOMqK1NPYdzMsLAujVMso7zOxqGaS5f4gLp6Wzs3kw6Nijk8N5dp1fZRMdFgriPwcXXXQRF1100S99Gr8KQqmVEL8IZW1GbvtgH6c9t4XlW5vpMR+UeaUbtIxPjQjaPzNaQ1GinvtPLCJcrWD51iY21fUF5dcBXt3cSIZBS6xehd3lpaLTwhklKSwtjmdH4wDXz8shIVwFgFou5bbF+Tz9dT1FSeE09tkoStRTkBDG2RNTUMokqOQSLpyWTnWXhaJEPTcvykWjCM1/Qvx3EfoXGeJnp67HwtkvbMPq9GuFy9pN9Jgd3DA/F4lEIEKj4I5j83l3VxsVHWbOnZxKZrQWl8dHv83JV5XdjE+NIE6vQqOQBgXzRUXx3PlhGWaHf2yVXMLvF+bR1D/ExIwodjQNcNfSApweH20DQ2yo7uHKOZmEK6W8cN4EmvuHuOGdUtIMWn43O5MxKRHIJLCjycg5k1IobzNRnBSBzyeGColC/NcQCuQh/uN0muzsazXSY3GSE6fD4fZxyQx/9d2Whn62Nw7wz28aOGNiCsmRGkRR5NXNzUTpFJxRksLn5V2kRWlIiFDTNmjnxQtK+Hx/F1/s7+LK2VnU9lj5aG8H8XoVrQP2QBAHcLh97GoexCeKvL+rDa1CSkyYEr1SRq/FydVzs+k2O1DKZDg8Xl7a3MSVc7LwiTDk8nHjO3s5f2oaLo8PlVyCIBG4+JUdaBVSjimIJSdWx6gjVKCGCPFzEgrkIf5fdBrt1PZYkAgScuN0xOpVR9yvx+zgpndK2TLsgnjFrEz2t5vZVN+HIMDionhOn5DMF+VdSIf12V0mB3PyY6jqsvDHj8oDY8XolDxx1lh+t3xXIFjvaB7kd7MzWVwYx/FjEvm8oguJAHF6FQM2F06Pj36rkwiNAoCJGVFMTI9iY20vs/NisDrcqBUSLC43OqUct8fHk2vrgj5DfnwYlZ0m2o1ObvugLLB9bVU3fzy+CJlEID/h6CkQQvxyfP7551x//fV4vV4uvfRSbr/99l/6lH4QoUAe4kdT023hkld20Dror8gsTNDzzLnjyIgeaUpV1WUJBPF4vQq728umen8FpkIqQSGTMCc/hmPyY5FKBL6u7KbdaCc1SoNJ7w6yws2M0bK/wxw04wZ4bUsz9580iltX7OWJM8eRHRtGS7+N+HA1dpe/X+hzG+opiA/jspkZdBmH+HBPB//Y0IBEgBfOL+GPH5Uz5PTywMmjuPPDMgaH3MgkAhdNz6Cqy4JWKeflTY2BYy6bksqY5Ah2Ng8wYHOhkstID/Uf/VXj9Xq5+uqrWbNmDcnJyUycODHgpPjfTiiQh/jRfLC7LRDEASo6zayr6iVjxshAbnMeDLoFCXp2DatBorQKrj0mm+VbW1hV2sGVszLY02qkbXAIvUrOfZ9U4PGJzC+I5eLp6by0qYn8BD0Ot3fEMaSCgMXhZvklk/lobycvb2oK/G1Mcjh3Lc0n3aDF4/WxtaGf1fs66RpeXPWnUDy0DX+eva1GrpqThcnhQSoI7GsdZEKGgUnpUXxR3oVSJuGOY/Np7LPxwe42xqdGUt9r44rXd/LKRZNIiAj5ov8crNzTzkNfVNNhtJMYoeaWRXlB1cD/H7Zv3052djaZmZkAnHXWWaxatepXEchDqpUQPwq318f2hpH+GntaB4+wN2TF6lDK/P/MIjUyipP8KYhzJqfy0BfV1Pf67Wuz4vS8tLGR3Lgw3t7RitvrlyauqejBK0JsmJIu4xCRGjnROkVg/IRwFfeeUEhOjI5ei4vXtwRrdPe2majtsfHE2hr2tpmIDVNR33dQJz4mORyF9ODP4LlvGtjbZiIlUk12rJZlU9MZkxROZZeZ08ancMG0dJ5dX8+rW5rZ2jDAs+vrAb8mfm1VDxUdZnw+kbJ2E+/tbOWTvR009wfbEIT4aazc084dK8poN9oRgXajnTtWlLFyT/tPGre9vZ2UlJTA6+TkZNrbf9qYPxehQB7iRyGXSlg6OiHwWiJAVoyWY4vij7h/blwYyy+ZxPjUCPQqBRnRWlKjNAgQpDaxuzwkR6qp67GOGGN7Yz+Xz8xkTEokBq2C383O4oIpaXxwaiRfT9jEKfuvIa/rI5S4gtwPD6BVSPnLycWkRqqI0ipYVBRPcqSaMycmc0x+LPW9NtRyCVMzDczNi2VtZQ97WoxYHR4ueXUnX5R3oVXI0Ktl5MbqgnqRAqwqbeeY/Fh6LU4e/LySfe1GTnl2E29sa6G2x8rqsk4aekd+rhD/Px76ojqoiTeA3e3loS+qf6Ez+uUJBfIQP5rFo+I5YWwikzOiuHlhHsXJ4dT0WKnsPLJL4MQMA69dPIkr52TSOjDErNwYMg7LJ1udHixOD4lHSE0UJ4WTFKkmIVzFwJCLjGgtZ+cLjP/2ctSbH0FoXI9h7U0U9H3JrNxgH5RonQLjkJtz/rUNmUzK3avKcXu83LW0gKpOM499VUtFh4nHzxwHiPRZndy6OI/Fo+J4dXh2PyMnmub+Ifa1mzjCfQIBAZVcgtPj5ZuaPtoH7Zw3JY2MaC1PfV3L3z6v5rb399HQFwrmR4MOo/1Hbf+hJCUl0draGnjd1tZGUtJPS9f8XIQC+f8ANqcnKFf9U0mO1PC3U0Zzyvgk/v5FNSv3dPD4V7Wc9fxW6nosR3yPTiUnLlzNxTMyaOyz8lFpB4tHHZzFv7SpiQdOGoVGISUn9mCuPUanZF5+HDU9VnY29eN0+7j9gzL05loEc2vQMWI33MbFJQbOn+oPoieOTeSBk4t58PMqRBGe+rqOx88cw4ljE/n9u3vZ22ZmckYU8wvjufKNXWxpGKCs3cSfPq6gosPC6ROSefbccdhdXiK1CkrSIonVK0kMD1bonDkxhdgwFS9vakItl2IacpMfr+fDPe2BwL+jeZB3trfiO9KdIMSP4kg3+3+3/YcyceJEamtraWxsxOVy8fbbb3PCCSf8pDF/Lo7KYqcgCBHAv4BRgAhcLIrilqMxdoj/P1anhw3VvTy7vg6JANfMzWFGTjRa5U//2u0uD099HSzTM9nd7G0zkR0bNmL/PouTvW1GarutXDs3B7VCgs3pZXFR3HBqQ0pL/xCmITf3nVhEt9mBANT12rju7T04PT6OH51Aj8XJJTMyGLDtJPEI59Xf3cagTc+TZ42lqtPEtW/uwTVc8j9gc7GjaZC9rUbGp0aysa6PRUXxbG8cGDHT/mRfJ0uK43lzWysbD+lIdMex+fzz/Al8vr+big4T07Kj6TY5uH2FX5Z42cxMpBLY1TxyHeHLim7OLEkhMzbUcu6ncMuiPO5YURaUXlHLpdyyKO8njSuTyXj66adZtGgRXq+Xiy++mKKiop96uj8LR0u18gTwuSiKpwmCoAA0R2ncED+BbfX9XP3m7sDrK5bv4pWLJjInL/bfvOuH4RVFsmK0TEyPoqzdFMhtuz0+9rYa6TDZSQpXkxmjRQSeWlfL61uakUoE3F6RMyemcEZJErtbjLy2pRmlTMIF09J5eXMTWpWMum4LPRYnu1uMgWN+vK+Th08fzZ0r9qOYlkCBPgXpIbPywVEX8VolXDQzjh6znds/LMd7SIQemxJBTbeVb2r7uGJWJmkGDW6vF4Vs5IOpWi4lTq8MCuIAj66pYVzqJG5ZlIfN6aG2x8LuZiMXTU8nRqdEIvgNvg5PHQEUJITx/Lf13LmkEL1aDkBtt4XyDhMgUJSoJydu5E0wRDAH1ClHW7UCsGTJEpYsWfKTx/m5+cmBXBCEcGAWcCGAKIouwPXv3hPi52H59pYR297f1XZUAnmv1YlWKWNH0wAlaVEcNzqB5zbUE6lRcOIzmwL7XT8vh8xoLTqlnGuOycbtEQlTyXhzewtLi+MDnX6cHh8Z0RokAnxU2sFti/O56d3SEcc12d24vD7+utlK3LynmWr/lvD+PfSkLmGNaxTnTU+iqc9GYoSKv5w8ikfX1NBtdjIxLZLr5uews2mASE0KUolAfLiK1CgtWqU8qNRfEODY4viAJPFQnB4ftd1WJqRFoVXKGJsSiVQiwer0YHV6+LKim7oeK/edWMj0bAOb6vyddBLCVYxOiuDBz6tYNiWdUUnh7G83cfEr25mVG0t8uIodjQOcMzmVoqRQcdH3cdK4pKMSuH8rHI0ZeQbQC7wsCMIYYBdwvSiKQZorQRAuBy4HSE1NPQqHDfF9RGsVI7ZFHWHbj6W+18qFL+0IqDfaBtsZlxrBW5dO4Yo3dgXt++z6Ol6+cCJ6swyL04NGKWVX8yAXT8+gx+xkapaBtZXdnDslHYVUwt3HFfLhnnZ2NfczLcvAt7V9nF6STEyYEqkgkBmtQyKA2yty9Zc24vRTyY1byO8zcyjd0sxxKQriw1R4fD7KOszcuiifLrOdMckRXPPmbuxuH388vogHP6sMFBZdOjODJ84ay8baPobcXkYlhtNpsiOVSNApZQFPGPDP6tsGhxhyeQhT+WfVObE66nosvL6lGQF/W7sZ2TGMSgxnYnofPlHE4vDw6JoaJAKBJ4DVZR1cOSebVzY30dw/RLpBw5QsQyiQh/jRHI1ALgPGA9eKorhNEIQngNuBuw/dSRTF54Hnwd8h6CgcN8T3cNakVFaWtge66ihlEk7+ibOYXc0DbGsYGCHB29NixOxw03OY37fbK9JhcvDXz6qQSwWOHZXAguH+nQ6PjxWbmvjjCUVc9touZmRHE66Scs6kVFRyCZnROubmx/LSxkaa+v3a7xPHJvLw6aP5/Xv7EEXot7o4fmEC7+5oI1KrJCFcyab6fu7/tCroPKZmGihIDCcuTMVb21uYkmnA5vKwtWGAf33bSG6sjo/3daKSSXhvZxuXzszAbHdxy6JcPtzTQU23henZBsamRIIIf1y1n5sX5ZMYoUYqEUiK0JAdq8Mnwqnjk0iO1BChUfDYmlq+PSQ9c9nMDNIM/sxjbJiKx9fW4nB7WVIcj1Yp46mvaxmfGklSZKiwKMQP52gE8jagTRTFbcOv38cfyEP8woxPjeD9301jU10fgiAwPdsQKMj5sdT1WGgbtHPjO6VcOC19xN8lAujVcvQqWVAJfYRGTl2PBYVUwu3H5vP2jhY+2ttBjE7JTQtykAhCwPDq/KlpNPXbGLC5cHm95MWFsbqsKxDEAVaVdjA+NZKXL5zI/nYTaoWURL2aiAIFXUY7/TY3TX0jG0O0DAwxJTOKCWlRpBrUrKnoRqeUceeSfF7f0oxKLg3qCvSvbxspSYtgdk40tyzKpbLTn8v2+kS2Nfazqa6fKZnRnDExhT0tg5z1/NbAgukHu9t467IpTMk08OCpo9ne2E9Vl4XxqRFMSItCKZMCYNApiderOHlcEh/sbsNs97B0dAKdJnsokIf4UfzkQC6KYpcgCK2CIOSJolgNzAMqfvqphfipCILAmJQIxqRE/KRxqrvMnP3CNs6ZnMrgkJvqbkuQBwrAJTMyKUzU8/x5Jdz0bikdJgdJEWpuPzaPu1bu58yJybyxrTnQbLnX6uTuVeX86YRCdjUPcsex+Vzx+i48PhGJAOdOTiUuR8WeVuOI8+kw2vl0XwfXz8/B5fFR1mFiyOVFJZfy9Lo6/1PHtuD3zMmLprHXhinWFWSKtat5kFsX55MSqWZBYRxrKg52IJpfGM8fVpUzPTuKCI0Sq9PLP9bXB9QSG+t6OXl8Eu/saAtSvYgivLm1mSmZBpIi1ZwcmXzE65pm0HDSuCT++tnBp4cXNzaSEK6iJD3qB38/IUIcLdXKtcAbw4qVBiDU1uM3xIaaXgZsLuQSf+HL6rIuzp6UwuQMAz0WB9Ozo5maZUApk1KYqOfxs8Yy5PQSoZFT3WXmloV5hKlkvL41ePHV4xPpsbg4b0oaN7+3D49PJCNay7mTU1lb1cOfP63kounpfF3VE1g0BH8AfOHbBtoG7NzxYRk5sWFcND2dT/Z10jZoJydWx62L8vjH+npsLg93HpuPWikjKyaMNw9bAPaJYLa7idIquHRGBiePS8Lq9BCtU7CneZBLZmQwKknP3lYjr2xu4rQJyaRGaXB5fYxLieCtbS1H1OgfqcL0cPLiwvisrHPE9uVbm5mfH0t6TEimGOKHcVQCuSiKpcCIzs4hfht0mRxMzohCp5Lxu9lZCIJAl8nOM+vq+MNxhSwsjEMhk9JlsrOhppcOowOJ4C/nn5Jl4Lq39rB4VDx6tQyz3YNGIeWY/FgkgkBJWgQ2l5eWQX865MyJKTywuhJxOA7uaTFy2+I8djUP4nD7OGV8EtsbB5iSaeCz8i7CVH41zCNfVtPUP8RlMzJoGrCxp3WQh04bzcCQm6+reviqsptj8mPRKqQjPp9OKeOsF7bSafLn9y+flcmioixSIjV8Xd3D0+vqmJ4dzRNnjeWNbS28f0g/0d8vzGVyRhRfVHQFzhng3Clp33tdlXIpiREjlbpRWgV//7KaS2ZkMCEtNDP/ubj44ov55JNPiI2NZf/+/b/06fwoQu6HIQI43B7KO8w09Q8RrVUyKkmPQadkcVEcCpmEP39SGdh3amYU71wxhZaBIT7f301RYhiNfTb++FF5oEt9SpSaeL0KmURgVWkHV8zKYmNtL6eOT2ZddS9l7UYmZ0aRGK7imPxYTHY37YP2oIAI8FlZF38+cRQdRjsGnYIvyru5fl42XSYH50xKRSnzq128PhGv14fN5WVvq4nZuTEYtEpSotREaORsrO3jlsV57G0zBcbWq2SoFVKidSpOL/EbJlV1WqjvsfLE2lo21PgXKrc2DDAmJZzMw6x6X9rYyMXTM7hzSQFb6vuRSQTOn5pGSVrkD7rmU7MMROsU9Fn9+XmpxN+Q+q+rq2jotfHW5VOI1Px0pVGI7+fCCy/kmmuu4fzzz/+lT+VHEwrkIQJ8sb+b1sEhwpQy3tzWTEqkhntPKMIQpuLVzcGuglsaBthS38/DX9YAcEZJMp0mRyCIA7QO2Om1Ojl/ajp6tQyX28uyKWkMDrmZlRPNqeOTqO+1oZVLuXBaOqv3dyITRrZPU8olrCrtYGNdHy9dUMKc3Gje29nGij3t3Lo4n68qutnZPEicXsndSwtZPCqe6DAlt39QRpfZQZRWwVVzsnlxYwPv72zj9mPz6TY5UMgllKRFUt7hL9V/Zl09Xp9ISVokJrsnEMRjw5Rkxmip6bYyJzdYgz845Mbp9fHo6hqKk8L568nF3ysfbB0YYl11D9/U9PpVORdMZHfLIO1GB5EaOa9tacbjE6nqsmAccocC+ZHY9y6svQ9MbRCeDPPugdFn/KQhZ82aRVNT09E5v5+ZUCAPAUBNl5l2o523d7QSrpZz3pQ04vUq6nushKnlI9zmgEDpuyD489bbG0eWpTf02fi6qodei5MHTiqi3+bC5vTP/Pe2Gbl8ZiZGh5s/f1JJ66Cd2xbnoZJLmJ4dzZjkCFxeH1kxWj7Z28ndSwt46MtqThmXzI7mQf50wig+3tcR6HjfbXZy3dt7WHHlNG5+b29gljtgc/Hommoum5nJU1/X8cGuNm5elEuf1YUogtfroyAhjEdOH80fVpazs3mQAZsTQYCr52ZjHHJT3WXmpLGJQT4wADOzo0kKV3HhtHRMQ67v9fsw2V3c9WEZ39T6bxJfVfYwMyeaOxbn8cGedjqM9kDOfXSyniitPPDe1gEbXq+ITi1jwOYmWqc8KnUBvzr2vQsfXwfu4YItU6v/NfzkYP5rJRTI/8cZsLkwDbn4sqI7MLtuG/QvIv7p+CIEAY4fk8CcvBjWV/cG3qdRSAlTysmLC2PJ6AT2tho5dlQC/9hQHzR+apSG3mHN+e4WI4nhaj4p66Sxz8bUTAPf1PYxPcsQaFTxwreN/PnEUexqHuTRNTWBca6ak0VTv43KTgtZi7TMzI6mz+occfPwidA6aA8E8QM43D4UUgknj0tkTl4sV72xB1EUuXVRPlVdFv6xoYFLZmTw4KnFNPbZcHtFbl+cx4d7Oqjq8huB7WgapLbHyrmTU3l7RysLC+OYlRvDHSvKiNDIuWtJAU63l72tg3xT04dSLmFaVjSjDpmhN/TaAkH84LWU8U1tPyt2tyOVCJxRkkLbgJ1LZmYQrlbQa3GwrrqHZ9fVY3Z4OHFsIhaHh31tRv5+2mi/tv1/ibX3HQziB3Db/dv/RwN5yP3wfxSH28sX+zs5+dlNPP9NA2/vCHYSFEX/bDpKp+Drql6umJXJmSXJhKvllKRF8uy545FLBSZnRJEVoyU/Xo9aIeHqOVkoZRIiNXKum5fNNzUHg3+4Wo5MKtDY55cgTsqIYk1FNxrlwQXIAZuLDpNjxPn885sGZuXG8PdTi6nvsbJ8WwsDNhcxOuWIz6ZTylDLgxc1JQKMTQlnfGok179ditcnMr8gji8rulhT2cP5U9PY0tDPNW/u4ZEva3jq6zpSo7SBIA7+fHpWjI5pWQaePXc8OXE67hg2yzIOubnl/X18UtbJ2qoetjUO8JfVVZzxzy3sbz+Ykz88/6+SS8iL1/Hg51UMDrnps7p4dn09i4rjArLRnc2D3Pp+GU39QwzYXLy8qYlwtZxBm5srl++mZ7jb0QE6jHZaB4eCfGZ+U5jaftz2/wFCgfx/lL2tRq5Yvpvm/iF6rU4MR3hE1yik1PfauOndvZz9wjaMdjd/OqGQmxfl8t6uVt7Y1oJPFOkw2smK1VHTbaXP6uS5ZRN46YIS3tvRxrZDZsxDLs+IJs05sTpsTg+TMw6qMzxeH4fj9YnUdFvRqWSsLO0A4ON9HVwyIwPJIWn1eQWxaBQSbl6UG7T9ilmZJEaoA2X1AHnxYexuMaJRSFHKpOw7ZBG03WhncMgdeB2ulnPD/Fw+3tvB1W/u4Zb396JTyIg/QtNpAYFzJ6eiHfZvWVt5UJueGaNlYvrBGfTYlIggPf4BPtvXFfj/0kOMww6wpqKbadkGOk0O2od9uM12F69taWLx498w7+ENPPhZFZ2mn+bR/V9J+JF1+d+5/X+AUCD/jSKKIsYhFy7PyNw2EOQquKGml7MnpQYFvhidksIEPR8PB02AL8q70SplXPtmKZ/u66K2x8p7u9pQyaVc/3Ypn+zr5J2dbVz/9h7K2s1cNiuTrBgtEgHm5MVQEK8nJVLD4lHxRGjkJEaoKEzU80V5N1fPzebmhbmcODaR/PiwEbnflCg1HUY7y7c0kxrll+wZh9y8s7OVGxfkcu/xhTxzzjiWTU7lzOe38ea2Fq6fn8t187J58YISJmVE0W9zkRunC3xOp8eHViElNkxJ2+DIatCmPhvjUyMAOG1CMk9+XUv/cPWn2e7hkTU1nF4SHDyGXF6eWFvLPR+Vc/ux+f7ztB+8IURoFDx82hhuW5xPSVoki4riAyX7h5IRc9A98Uh58GidEuOQG5VcQviwk+KOpkHuWVWO2eHB5fXxwrcNfLJ3pE79V8+8e0B+2FqEXO3f/hM4++yzmTp1KtXV1SQnJ/Piiy/+pPF+TkI58t8gDb1W3tzWwpcV3RQnhXPZrIwRedRDZ+Bur8jz3zTw+JljaRmwI5OASi6lsd/GpIwoag9pv9ZnddJrPeinMj3bryABv7rjounpmOxurE4PUonApPQoHjixCJlMQvugHbVMwrRMA8ePTuCW9/cFHAe3NmznkdPHMCk9CpvLw99PHc3T6+rY2+b3Dl9SHM/fP6/G6fHx7Lnj2FDTy5DLS2OfjeVbmnn8rLHYnP7gClDfa+Ox4f+/ak4WFrubz8q7WHHlNB49Yyx3fVjGJ3s7uHRmBs+ur+fEsSM9aGwuD5fOzKRtcAiFVILxkBk6+G8EhzIpI4qWAf8NodfipLTVSGa0loWFwW3w0qK1nDAmgeouM8+uq+eyWZloFVJsw9ciXC3n2GL/e7xeHyq5hDi9ku5hHxuZROC40Qk8sLqS+08cRbrBH/QPXcM4wPu72jhncgpapXzE3361HMiDH2XVyltvvXUUTu6XIRTIf2PYnB7+9HF5QDrXMjDE1oZ+3rl8CtmHeF2XpEeSHKGizegI7BehUfDU2lqsTg8nj09GIhEYlRTOCWMS+Wivf2Yukxx8iFNIJZwyNo5/fOOXJl4xO5O/fVYdULNkGjTce2IRpe1m+sxOZuRGU9nl9xk36BScOi6JT/d3BTxOHvysimfPHYdKKkGilHDzwly21PdT3mnmgU8r8Ykglwo09du4dVEuMolAn81NRrSWu1eWU5AQxpDzCOoajw+nx0ef1UVNt4UTxyYyLiWCQbsbuQBqhQyDVsGZJSm8t6sVn+g32dKr5CgkEpYUJ2BzeFDLpUHqHalEYFaOv22dxeGhtNUYVCxU12Pl76eNPqK/TVm7KZAiemZdHZfPzgJEkiM1jEkODzTnkEgEKjstnDo+GblUglQikBWjxer0sOLKaeTFhyEZfsSIDRu5XpASpT5ie7pfPaPP+J9d2DwSoUD+G6Om2xII4gfot7ko7zAHBfLMGB3LL53CjqYBmgeGhjv0WJmUaWBiehR3rSzD5vQilQjcMC+Hq+dmMiY5AplEwqhEPYlqD3cWdJG670bECXfxvCCwpb4/EMQBrjkmhze2tTA2JYKCRD13riijw+S/cQgCPHzaGDJitNT2WHlreytSicA3tX3+HLLbR1OfjeQoNSa7OxCMrpmbTbzMTpZe5OQ3/R4nN87Poa7XSo/VwflT03n6kM5FUonAxIwoWgZs3D7cQebtHa2kRmkYlxrBx3s7eHdHK5MyolDJpVxzTDYauZT4cDU6lZT2QTvlnSYGbU7uPq6Ae1aVB/xg7j2+kDEp4UzMiOL9Xa18eFgX9+PHJAY8U3w+MRBwgSC9vcnu5rE1NShlEr7+/WySIg+mWgRB4KxJqZz9/FYcHi8SQUCrlPL2ZVMoTAy+QeTG6UiOVAd81LUKKYuK4lEeoXFGiN8WoUD+G0MiCMilQsC69gA+UeTdna0cPyYBtdz/tadHa0mMULGtcYD6Xis1PX7Nd2Ofjd8vyOPxtTWY7R4e/aqGe44r5K6V5Rw3OoFr5mYz1b6B8NW/A2CuF6IWPMEjXx/0MZFKBBBgVFI4/1hfz0XTMwJBHPzqjWfX13PWxBRy48IoSAhjfkEcy7c2s7NpAKVMyvphxcvYlHD+ed54jENu1lX34oqUo9ZF4BPb/N7kw1HebPdQ1Wnm+nk5fFXZTaRGzrIpaexvMyKTCfh8Ape+dtAv/eaFeUxMj2RcagRrK3tIM2i4cHo6dT1WRFHkj6sqAguJCwrjkEqtPHHWWCq7LGgUUmbmRKMYdjKckxvLlXMy+de3jXh9ImeUpLCkOAGz3c3m+n6Wb20iJkzFsimpjE+NJDdOh1ImCUrPnD81nfjwkTr0sSkRrLhqGvvbTUglAsVJ4UfsJJRq0HLCmESUMgleUUQiCKQbNIFzPIDb60MAZNIfFuBFUUQ4QqFWiP8c4uHypu8hFMh/Y+iVUk4vSeHNbQeD6pjkcAw6Jee/tJ3sGB3jDykfV8iklKRF8db2FlaX+ZUSnSYHZW0mLpyezlNf1yGK/pTKtCwDL29qYkysjLDdTwfG8KmicIkyjh+TwO4Wf3GOTxQRAIvDg93txeUZqUQZHHLR3O+XON4wP4flW1sYHHKzpWGAa47JDgTy0lYTVV3WQM4b4N0rkkg3aGjqH0IjlyKVCHh9Il9V9rClvp/fL8hFIgGLw83T6+t56uxx3PhOadDxH/uqhgdOGsUHu/0z6V6rk/IOM39YWsDH+zoDQRz8KpGbFuRS3WUJzPgnpfurQlsH7GTEaLloeganT0jBJ/pTGkqZlJV72rnhkON+uq+T9343ldHJ4bxx6WSeWFtLY5+NsyamcMr4ZP8N8AgUJOgpSNB/5/d+YB9BgO0NA9hcXqZkRgWldawOD5vr+3h5UyNapZxLZmRQkh6J/N8EdJVKRX9/PwaDIRTMfyZEUaS/vx+VaqQi6rsIBfJfCV1mO819Q6gUUrJidPRbnWys7WNvm9HvPphpQCmX4PD66LM4uWlBLm2DduL0SswOD6Wt/gDb2G8LCuQA7cYhPtvfFbTN4vRw4HcrEcBodyEgMCY5nF0tRo6XDi+WKvW0FF6BR4S0KA1/WFrAq1uaUEglpBu0VHSacXtF9GoZEoGgfO3S4gTWVfdQmKAnTCnn2+FCmYxoLe2HtVnTKqRcMC2dt3e0sGRUAk6PlyfOGsdjX9Xw4Z427juhiDe2NdNjcXL86EQSItTU91q59+NKVHIJDrd3xFOK1yfSZgw+jt3tRauUUnaIFPEA/TZXQG54w/wc3t7eyvu7D6ZTlk1J464l+agV/p+Vxe7m2fXBDapdXh/f1PQwJiWCkvQonj+vBIfbS+Qhi88+n4jF6UanlH9nYP8u8uP15McfOeBvrOvld8sP9nD9uqqbd6+Y+m8tc5OTk2lra6O3d+RCaoj/HCqViuTkHy6nDAXyXwGVnWYue21nIPd5/4lFfLC7nT2tRmQSgTSDlr9/Uc2u5kFm58YwIS2Sv39RRZRWiWlYaXHlnCwAorUjF8TkUgkqmXREGb5UEFDKJFwxO5OPSjuo77Vy9dxsNtX10T3hGhJMt7Np2gvcstpEp6mb7FgdN87P5YZ5OWgUUrY09DE2JQKZROCdHa3ctbSAFbvb6be6WDwqHrvbS9ugnbMnpbBit3+RUCWXcM3cTO78sDxwHoLgV4j0me08t2wCFR0mtjUOsrG2l+uOyaG620KH0c7Ni/IwOzzE6BSY7G5MdjcvnDceELA6PURpFUHNI8KUMo4UJq1OD+PTIviivDtoe2qkmspOM4Lg178//lVt0N+Xb23mjJJkRidHBM5bJZdw8fR0wlRyJAK0Ge2YHR5aBmykRmlRK6SoD3FkrO+x8ub2Fr6u6mFGtoHzpqaTexQaMjvcXl74tjFom0+Eryq7/20gl8vlZGRk/OTjh/jPEgrk/+U43X5d8qGNgPtsrkDDhdNLknlvZ2ugi05jn42iRD3nTUnjlc3NTMuKYtmUdBp6rdx/YhFFSSNna6lRGm5akMsDqw+6G45JDic/Xs9jZ47lkS+rAw0hYsOUXDIjgzJfItaTP+XSV+sCC3d1PVb+srqSx84cw8o97ehUct5YXcW9JxSxtrKbz8q6uPaYbHosTp5dV0+X2UFRgp4pKRomMYBplIJ0oQtBqmNBYRyf7+8iPlzFeVPS2Fzfx7SsaC57dScen4hKLuGG+bnc+3E59x5fRFP/EFe8vgu3V0QuFbjn+CLi9Coe+6oOt9fHcWMS+ePxhTz0RTVtg3YSw1XcsaRgRMFMYYKe8g4zY1MiqO+1UddjRRDgnEmp5MeH+Quk5ucGZJOHYz5EM65TyblpQR53rCijc3h9oChRz1Vzsui1uEiNOqgV77c6KWs3Ud5uIlKjQCoReH1rC5vr+3nrsikjCqm+j4ZeK9VdFmRSCYUJemLCFEdc9FSEFkJ/E4QC+X85RrubrQ39QdsOXQeJCVMGtUIDKO8ws2xKGvef5Ld+vebN3fhESAxXMTY1kujDytoFQeDMicnkxOn4trYPrVKK2e7hiuW7iNEpOXl8EvW9DSwdHU+kRkF1twWpINDrUAapL8BfEbm1oR+NQka4Wk670c7dq/YzLiUCnVLGuqoeEiLUnDkxBUGA0Yk63t3Zyt/s/4TmzQD0nvQ2kzJSyIrRopJJaTfaGZcayd+/qA6UnTvcPp75uo6zJ6didXn52+dVgdSJxycy5PQEdd55bE0NNy/MZXZuDNOyDDjcXj7Z287svFhuWZRHn9WJXiUnN1ZHaZuRryq6uXlBLnq1nAiNApkUjn3iW8LVCiwON4+dMTZIIQKQHKkm65BmED6fyFeV3YEgfuC7qey0MGZ41g4wYHPyx4/K+WRf5/D3AbcvzscnithcXva3myiRS9Gr5bg8Xio7LTT0WYnSKClKDCM6LDjIl7UZOfdf2wIt9zKitbx4QQm/m53F5vqD/5aUMgnz8uMI8esnFMj/y4nQyJmRHR34kYNfrnYgiEi+YwGq2+xAJpHw7PqDJlYdJgcPfFrBC+eXoFPJsTo8lLUbaTfaSY/SEqaS8cHuNmxOTyAoekWRsSnh/PWUYpIj1Fz95u5AgLj3+MIRx1XLpfhEfwXj5ro+bpifwyubmyhtNbKwKA6DThlISZw+IYkN1b2Eq+V8mPsgUxK/JkEjUuMI555VB1Mr1x6TjSAwwjvE4vSgkElwur1B6o/MaG1Quf0BNtT0IpNKiNIqSI1Sc9r4ZMo6zDz5dR06pQyHx8uDp4zmjmML8Ipi0CLgruZBvD4CqZlVezu4bXE+7+1spbTVyLjUCK6bl0PCIe6HDo+XXcPOjIfSZbaTHHVQYljVaQn6fkURnv+mgRPHJvLSpiYALpiaxk0L8thc38dVb+4O3MwXFMZx9Zws9rWbiFArGJ8azkubGoP6pjb22dhU18cZJSm8ffkUPt3XgU4pZ/GoeEYn//96uIb47yIUyP/LUcr82uaKDhMNfUMIAkgF+MeyCazc086gzcUx+bF8XdUTeM+ioni2NQwwIT1yxHjbGgcw2t0MDrn4YHc7L21s5C+nFFPRZcbp9mF1ePD4RFKi1CybnIZSJqGh14ZCKvCPvR2BAKGSSzCaTFwxK4N/fuPPvY5LieCqOVlolFJEUSQmTIHbK/LASaPQKmW0DlhRy+WcPzWNmDAl5R1m9rQa+cvJo7h5VTmXTp7JrU23EpadABxMAbX0D3Hi2MSArFIllwSuTXasln6LC82wr4lUIhATpgyUrR9KmErOgM1JdqyOJ9fWUpigp2VgiFsW5RKhUZAWpWFSRhQSiYDksOx5apSaNIOG5uGnnzUV3VgcLu45rgiPTyQpXI3hsIIcjULG0uIEKjstQdtHeJrbg50awb+wqlUe/Hm+uqWZ+QVx3L1qf9AT2Zrh6t0DTpHnTkqlosNy+HDU9VpRyqVMyTQwJdMw4u8hft2EAvmvgPx4Pe9eMY2mfhtqhZTMaC1qhYxRiXqcHh+DNhfbmwbY3TxIrF5Ffa+Vr6u6OWlc4oixxqdGoHP3s7NbZG1FN79fmMdfV1fRb3Ny+axMzp+Sxvu727h0Rib3fVIRmAVfPSeL+l5/qf7kZDV/Gd1D+v4H6Bh/I+PPnohKpWJlaQdXLN+FTinjohkZJOhVJEeqWV3WSa/FRZTSx1lTMrE6PKzY044PkVsX5dFndZEZreWVHb2cN+tk7JJg75Hi5HAsdjd/PaWYPquLQZsLnVJKcXIETf02UsN8bD5Xi72vBU9YMh+2qclOjubDPe2BBVyZRGBRURw2p4ceiz/VoVZI2dtmom3QztVzszHZ3VR0msmODUOnDP5pxISp+MeyCfz9syo21/czPi2CO5cUjCjKOZwTxiSxr83ElxXdSAQ4f2oahYn6IG12Qrg6IJ88+D35pY2H0m1xjLDnBb8u/ABfVHRxwdR0qtcEB/OZ2TH/9jxD/LoRfqzw/GhQUlIi7ty582c/7m+Z+l4rV76+k9NLUojWKemxOmnosaKUS3hti19THqNT8uJiFaNXn0DpnFfY4ivktS3Nw/auUQzYXBQnhROhlXHPygrqeg96rGTFaJmSaeCNbS18vNRD8drzaZjxCLdW53Bavopam5oXNwdbz965JJ8Oo4PmfitGu4crZqThReCmd/cxPTsaQYCNtX2IIlwyI4PP9nfywnFRfF5v55GN/lzu3LwYjh+TSL/NRbROwR0rynC4fdxzXCF/+7yKglgl/8zaStyOvwWO2znlHq5vmMii4mRcXh/9VhfZsTr2NA+SEx/GC982cN6UNCwOD69taeaWRXk8/GV1YAHz0pkZXHtMNuHqkWZVQy4Pg0NuIlRytKofNg/qtjjY32aix+JEAD4qbWdSpoEzS5LZ0TTIcxvqOWlcMq9saqTD5GBmTjRLRiVw18qyILnmh1dN4/GvatlwiDWwVCJww/wcHvnyoMb+X+eX8G1tL29sa0Eh8y8Kn1GSTMRR7DTkGz4xyY+UR4b4aQiCsEsUxRH9kY/ajFwQBCmwE2gXRfG4ozVuiB+Gx+vjrElpfLinnR6Lg4WF8WTE6EiLUlOYGE64QmRM32ckfnkf+LykDm6nNb2EuXmxFCeH8+y6Wq6em4NcJlDWZg4qhgG/CdWNC3LxeEVSu14AbQwbPEX8eWwfsYMbeGz/tBHnNGhzoZQJ5MXreW5DA1e+ZeTJs8bh9PiCUkFjU8LJidXhK4rnq245JQVJvJCZhkSQ4PX5+OeGBkwODwXxYTjcPsYkh7Opvg+nx8eVRT7iNj4UdNyE7X/h+gUreXBnB8ePTkAmFQhTyRidEs6OpkEumJrOt3V9TEiL5LjRCbyyuSlIhfKvbxuZXxB3xBSERiFDo/jhPxurw82z6+qCWuVdPy+HeL2SDTV93PFhGaIILQO1HD8mgby4ME4Yk0CX2UlWjI7aHis6pYw/nVBEYYKeu48rRPFZFWsqu0mOVHPD/ByeWHtQBnkgtfSH4wq5eEYGUolAUoT6qBXzuL0+djYN8OqWZhwuL+dPS2NKpuFHXZMQR5+jefWvByo5NLkZ4mdjyOXlL6sr8QzPlF7f2szF09N5e3sLUomE54/xkfjtbYH9vRIpz3xdi1QqwenxsmxKOrU9Fh5d04lBp2BBYVzAKOsAzf1DTM2KQtobj2XKLSRJpBRsvBZ7/ESyo2YH3PkOEKaSU9ttITfer4MWRb8mPidOR223f7avVfgrUW96b2/gffF6JQ+eOporXt/FlXOy2NViZHxqRMAfPFwtp384xaDzmUE8rGrU5yFR4eC0CTk43F5y48LQKmT0WJzDqRARCSIDViexelXAsfBQDm/WcDg2p4eabgvdZgcpkRpy4nQjSuEBKjrNI/qdPv9NA5fMzAhcE/Br19/a7n+imZUbQ2FiOO9cMZVOo50wlYzUYYfD7FgdT549jl6LA61Shsvjo7l/iDe3tRAfruK2xfkUJeqRSSWkGbQcbUpb/YqYA08K62t6eemCEo4pCKlffkmOSiAXBCEZWAo8ANx0NMYMEUyXyU632UmUVkFK1Ej/6qpOcyCIH+C9nW1cOy+bbrOT8v5usqRy8PqDYYN2HFXdVqZmGciLD+PpdXVcMDWdHouTHouTuXmxnDwukY/3dhKtU3L13CzcPpG2QQdV6edRPeBllqoR7IOom77ipgVXsKtdEpAj5sfrSDNoeHxtLZmH9LlUySXcvbSQHU0DVHVaWDo6nscOK6zpMjup77Xi9PgCnX7K2k1cNy+HjXV97Goe5JKZGZS2Gql3G5iujgT7IeoQbQwNbgN//Oig8uWMkmRquq2UthqRSgT+uWwCOXE6nG4f39b0sq89WOVy4Bp7vL6A3C9CraAwUY9WKeVf3zbw1Nd1nDwuiYQIFd2WCGZkRaMYPt8Dgb71CDcJu9uLRi4d8X0BxOtVgfx8lFZxRC9ytUIaCOwANy3I5bwpaaiGJYr/ST4r6xzhpvjipkZm5sb821L/EP9ZjtaVfxy4FRhpqDGMIAiXC4KwUxCEnaFy3x/HtoZ+jn96Eyc+s4mlT37LF+VdQQtjdT2WIzZHjtDKidereHFjI0/tFemcdGfgb9qwcKZnG6joMBGtU2K2e5ANP5ZPyojC6xMx291cPiuTufkxPPRFNav2tJMTo6HNJsHpFtFGGEChBdHHhI2X89HcPp6Zr+KxU3K5dGYmX5V3MzY5Auuw0sWvJBG45q3djE4K59pj/H04Dy2iOYDT7UMi+HXymdFa3F6RbQ0D3Dg/h1i9igGri9sW5fFalUj13OfxRWUD4I3Oo2vJS1z/WfC/sfd2tTE717/g5/WJPLOuDq9PJDc+jL+cUhxo7qCSS3jg5FEUJISB18P6ynZOfGYj179dyq0f7OXz/Z1squ3DK4r88fhCdrcM8uTaOi5/bRdPravDbHcx5PTw/DcNXPbazsCYh5IZrUUtl7K/3cSiooMzWZlE4L4Ti3508Y8gCMTqVf/xIA5HNtqSSw/X+MCA1Unb4BDO72hsEuLo8pNn5IIgHAf0iKK4SxCEOd+1nyiKzwPPg3+x86ce938Ff0HPnkAzB7PDw7Vv7uHT62YEHPD2t5sRBGFEgcrVc7JxuL3MzYtla0M/N9eP4YaFK3BoU3hjrwm7y83FMzKRCALRWgVapYxTxiWxp9WISi4hPVoXpEOv6bGgUyv403t76TA5cC3O49jpfyPtmxthqJ/cb69DO/1+Lt0oMODw8btZWYxNE+i3OLh8ViZapYx/bqjH5vJS2WWhdXCIyelRuL0iz6w7eByZRCA9WotPhB6Lg1sW5bGuupfKTjMIAtcck82G6l5qanq4fl4u1T4R+8K30YsW1rX6sHbpsTqDZ/miCCIH/9n12ZzsbTWSGaOjKFHPC+eVYHG4MegUpEZpkdj76K3bzd2rZfhE/0LxRdMz+MvqKuxuL9OzopBLpYGKV49P5Kmv6yhKDCc+XMkTa2tRyiR0m/3n/8rmJloH7BQnhXPR9HTWVnTxZUU38wpiuWlBLhFquV8+KfiNvnRK2X+lSdWxo+J5aWNj0NPEJTMyAgHe4/Wxsa6Pe1btp93o4LjRCdwwP5eM6KOf5glxkKORWpkOnCAIwhJABegFQVguiuKyozD2/zzdZkdQRx7wGy+1G+2BQN48MMRTa2u5dKb/BzXk9DA6OYLYMCXPrKvD5HBz5ZxMfD6RJoWGO94pC/wQd7cM8sfjC3n8rLH8ZXUVFZ1+ydv2xgGmZEaxoDCONRXdxOtV3H9yEc+sqw3Y0ZZ3mNnuyOa2k1Zj7m5iUIjiiX0ClT12f0PmhDDqeqysKO1g0OYKdMABiNP7W5Wd/a/tnDYhmUtmZLCmops4vZJlU9LYUN3NnUsK0Ktk3PRuKaeMT2ZpcQLf1PQiyzEE8vfbGweZkBZBZrSOVaXd3LQwl0GTI6ijDvhtCHotB18vLIxn9f5OxqREsLtlkE6jg7J2EzqljBsX5JLSuYmhgUG6zH7N9ynjk3jm67rAk09BYjjLtwbnvgGa+qz4fCJTMqPY2jCA2e5hS3kXUzINnDhWRUv/EBliGzemtTI2Lod6k3+B8qvKbiZnGviivJu/fFbN/IJYzpqYQsYhlaL/DYxJjuDdK6by0d4O7C4PJ41LDrTDA6jqsnDJqzsDT4yrSjvweH08esZYlPKRawghjg4/OZCLongHcAfA8Iz85lAQ/2n0WZ1UdZox2t3E61XkxOqC2q0JAkHd46u7/Pnx5zY0IJcKKKQSMqK13PhuaWAxrbzDzN1L82notY7Izb6xrYUrZ2cGgvgBtjYMcP28HNZUdHPnknxEH5S1mQPmU1+Ud3PHkny+6PHw6hYT/baDgfKaudnc+1E5t86I5A+LMinvcdI2aOfjvR1EaOQYtEru/agC8Lcji9EpmZZt4PQJyZS2GVlcFA+CgMvt5Z7ji3hlUyNfVXazpDgBq8PDk2eNpdfiRMTvu33fJ36fmE/2dXDS2CTuOLaA93e1UdpqZHJGFGdPSuWRNdXEhCk5oySZLpOD0yeksOxf2wI3ppk50URoFLy/q40lCalUKjKYk+tgXXUfMqkEi/NgtWRdj5XChPCAbe8B3D6Rq97czaUzM+gx+xdTJ2cY0CpleEWRudnhpDc+ReT+l8hWR+KccDmnl01ldGoM7+xoDSy81vda2dU8yIsXlBB+FGWDPxWJRGB8WiTj0yJHNMsAv+nX4RW4q/d3cetix39k8TWEn5Bm6L+MPouTu1ftD9jKyiQCj5/p7zFpcvitZf+wpIDsQxYQp2YaAl7ibq9IhFpGy8BQIIgfNzqBvPgwOk1O8uLDWFIcH9gf/C3bDEofcqnA0uJEUqLUuDw+PtrbQeGwAZdWJcPnE7nt2HwUUglqhZQukx2z3c3MjDAmJmSzt9vNgM3J9CQ5aeEO/rA4k9JWC09tqMTl9ZEfF8ZjZ/oDsM3pDsrr91qd7G4ZZFFRHE63D6dX5NvaXqZlG7hzWKIH8PKmJi6fmYlMELC5vKRGaYjQyInUyBkccrO/3YxxyM3ti/OJ0so5bUIyJemRlHeYuHRGBjFhSj7c005xUjgf7+0IanbxbW0fN87PYeWeDtoGI/hgdxt/PbkYnyjg9vrQq2WY7f5g/k1NL385uZimflugbH/xqHh/+gd4fUszfz6piPd2trGjKTjYv7poMbN5CeyDKLc8ymWzPqberRihntnZPEhT/xBjjnIgN9ndNPVZEQSBjGgtYaofnlv3+UR2twzy+tZmBm0uzpuaxtTMaHTDmvoj5eljdMogh8cQR5+jGshFUVwPrD+aY/6vUd5hDvIG9/hEHvy8ilcunsT2xgE8PhGj3U2b0R4waDohU8KUE71Yhxzsc8aztlMRKFGfmmXAJ4pBBSMnjklkYnpkIMCcMCaRRE8b959YyIubmllZ2o5OKePaY7JwD1notzpwuDzc8M5e3F6RNIOGi6al8/hXtdhcXpaHKXlmUTi/K70IovPw5FzBhNdFLpuZxSNfH7ROreq28O6OVtw+H6eXJJMbp6NmWIaYFKHi8pmZXPdWaeCJ4XezM2kbsHN4zdrq/Z1kxWhRSCW8urmRK2am8eeTRtFvdWF1ehiwuWgesPFpWRcLCuN4e3srG+v8Xuf/OHc8x49OoNfqOqIPSrfZyehkfcCIrN/mwuX1EaGRc9eSAv78SSVquZQrZmeyr83IdfNy0KtkWBwevqrsDniqOz0+MqN1R/R8aXdp/I9VoghKPTPyE4kwj1QiAUddCdLcb+MPK/cHznN+fiz3nlhEcuSRj384+9qNnPX81sB39E1tH8+cM56loxMAKEzUMz3LwKZhcy5BgD+dWERs2I9bwA3x4wjNyP/LGBhyjtjWNmhnQ01vkP/1zqZBnj13HJHOdsI/vIDwrn0AjA1L4Oyz3mOnIxKtQsqUjKgR8r5Vezt4/IzRpEWpyU8IB1EkUubivc2tgcBqdXp45MtaXr1wAhMzZOxsNvL7BXms2tvO4lEJ3P/pQc16r8XJdV+aWbXkUeKqXsXe34rZEY/jCIqFbY0DXD4rg5ouK6eXpPBleRcdRgf3nlDI79/bG5T2eW5DA4+fOXbEGAatAqvTw9++qOLZc8bx4sYmdjQPolPKuGRGBtsbB8iO1XLv8YVsaxwIBPFxKRF8tr+TJcUJvLSxiQlpkXQcYlYFkBihIjkyKtDVR62QMDY5AtOQmy/2d7JsSipFieHc/N7eIKOuWxblBd0YihL1fFvby5TMqBE9VFOUQwEBuWvefUTGpVEc5mJuXgzrqg+qbc6cmEJ69A8LsD+Uz/d3BYI4wFdVPczJj2XZlLQf9P5va/tGpOae21DHnLwYtEoZcXoVj54xlv0dJoxDbrJitBQmhkpL/tOEAvnPiNvjo6HPitnhITlSTcIh/RlLW418sKuNokR9YLJ2gDm5MZR3mDh3ciqRWgWlLUY21fdR0Wlh+uA6GA7iAIKlE0XpK0jzb+PqudlEfsdjebJ5D3/1fUBT9I181qGhXlnAzpZdnDQ2kVm5MZiG3CRHqXl/dwcfHNJU+PZj8xlyeUb8mLvMDrr7Bomr+gRb4cWIogPFEWaThYlhZMTo+NtnVfRanczOieGvpxSzo2kgkLY4FLvLS2yYkp7hhUqJAKeXpDBocyIRBJ5ZVx9wHLQ6PTyxtpabF+ZhtrtxeXxsGg7ixUnhnDI+iX6rE4fbS3W3haWjEyhKtFHe4W8WcfbEFLJjdEglAqOS9OTFhdE6YOfNbS1IJQJnTkwJVLw6D2tdt3JPO+dNTuWVLc1MHTamenhNNXctKaCxz0bLgB2JAJfPSCMrqoehGXcgpExCnT4Z8LtFPnByMVsb+ilrNzEhNZKJGVFHtWJSFEXWVHSP2L6+uucHB3LpEZQ0UomEQzfHhauICw/NwH9OQoH8Z8Lq9LB8azMPDXtqx4Yp+ceyCUxIi2R/u4kz/7kFp8dHTqyOmxfm8ermJhYVxZNm0JBu0OAT4Q8r99NjcTI1y8Cti/LosTjwdewdWQzQug1NgZu/f1HNZTMzRsgSMw1q0kzrkFetJDz5GLSKGZR1DnH5zHScHpGb3j1YZXnXknwmGyPZ1uifbb63s5XLZ2WO+HzhajmRPiPGuX+lVZXD2RONVHdZWFgYx5fDwUOvlnHelHR2Ng4EAnOHaYiabgu9FueI85RJBPRqGRdOS0cll2IcchGmklPdZaYkPQq9SobR7iYvPriDjsfnQ6OU8uz6em5emIsoQk23lX6bi2fW13PF7Cw0CimPf1XDcaMTuWJ2FvU9Vr6p6eXN4erK584dT2WXmSfWDrdq88Irm5u4dVEePt/IcglBgEWj4glTy/2LwUMuHjtjLN3DXY2GXF60ShmZMVqUslHAMSPGSIxQc8r4ZE4Z/8NbfP0YBEFgbl4MOw9LKc3Ijv7BY8zMieaJtbVBN7Kr52aHSvR/YUJX/2eissPMg4c0OuixOLnrwzKunJNJt9kZ+GHU9lh5cWMjfzm5mPs/rQgEtjm5MczOjeG9XW1sqe9HLhE4ZXwSjfqJZPHqwQNJpDROuZ+yXg9/WFpAj8XJhVPT2dLQz55WIzNSlVyd2UfM+n+ATEVT9Cwqy0yYhjwsm5LKBS/vCDrvR9fU8tBpowOB3DjkZsDq5Mb5OTy+thZR9C+W3rskm+2WMF7e66b88zIeOGkUDrcPtULKkuJ4fD5w+3w8ubaWuxem4rGHMyrKx4ScZL5u8bKytJ3bFufz4sZG2gbthKvl3LEkn0iNArPdwzPr6+izOgOVo7uaB/nD0kIUUmFEy7W8+DBidArm5sWgkEh4cl0d41MjefDzKrw+kTe3tXDzwjxe29LE3jYjcqkQaMB8gNI2I9sbB0Z8j+UdZo7Jj0UpkwQFs0tnZNDUb0MEXt7UyCnjk/nTx+X0WV1MzzZw7wlF5MR+d8s2n0+kvtdK6+AQ0Tol2bG6Hx0c7W4vcolwxKKdAxw/JpH+ITcrdrdhHHIzOSOKufmx37n/4YxK8lsHfLy3nUGbm5PGJTHxCHbJIX5eQoH8P4w/nWILchI8QFWXhca+oRElz4kRKlaVtgfNTtfX9HLD/BxkEgGPT+Sb2j5Gp0Swpy+Z68dcSVTZCyB6qVr8Dr9bI9LUvx+ApHAV95xQRJpBw59KXMRs+TPKtVsAaJ16N5sr27jC0ECcuB1J/zjunZnNvd/aAse1u704D7FJXTYllZm5MTy/oZ4b5uXg9onkx4dR1mZCJTcwJsXJwqIESluNTMsy8P6udjbW9XHV7Cy+qOhi0ah4on29PGS8EWfYdLY7r6M4KZzxqZE8uqaav582hvIOM3aXl4eHn14eOX1M0LUAqOi00Dxg4/UtzVx7TDaPr63FbPcwvyCWGJ2SLyu76DE5sbu8XDU3mwGrMyCLG7C5aOyzcv7UNHwi1PaM9O82DbkpStSzu8UYtD0/PoyStEhevXgSq/a002dzccq4JN7c1sKxxfF4fSI3LsjjkS+rA5azm+r6uf+TSv6xbPx3Buf1NT1cuXx34OZw4/wcLp2ZGeRJ/l30WZ2sqehi+dYW0qI0XDIzk/GpESMKikpbjby0sZHaHgtXzs6iJC2SzBhdUOPn70MQBMamRDA2JeIHvyfEf55QIP8PciCd8vAX1Vw/P2fE37NjdXi8IgUJYYxNCae01a9wyIkNO+JssNvsoChJzzH5cWjkElKjNDxf18+FvYu4ds5SYsNUfNKuoam/KfCedpODjbW9aBVSDPlZlI17goTxgxRUPEZ7yvFcUPEkkd+87N+5Yjlnpc/DPfdOdnRDTbcFnwg6hZTkSDVnlCTTOmDnwc+qOG1CMmsqevisvJO/nzqa3S2DQTK7cyen0jpgZ0qmga0N/QgSgZaBIZ7/poHUpZmkzbiHXa40tjQ5eHFjJcsmp3LRtHSeWV8XuA4HqOw0o5YHN4cWBJAIAoNDbp5dX8+jZ4zF4/OhkApUdprZ02yk1+JkXkEcK3a3MScvFo1cypDbi17tdzA8oD2/c0nBCD/wwkQ92bE6Pt7XiXHYrCter0IQ4MRnN3HRtHTcPh/dZgef7uvg2OJ4tjUO8PHeDuRSCWeUpJAdqws0cN5Q00uP2Ul69MifXIfRzq3v7wua4T/2VS2zcmMYl3pwtmu2u6jutjJgdZEWrSEnNgypRGDF7nb+MtxvtbzDzNqqHlZcNY2iQ7zSa7otnPPC1oDLY2VnFVfPzQoaP8Svl1Ag/w9S1XkwnbKlvp9zJ6fy5vYWRNHfxPj2xfm8sa2FLfV9nDY+mevn5fLmtmamZRqQSQXe29kWNF5ypIasaC0un0hz/xClrSZOL0mmtHWQa9d08MRZ2dT1+nO8OqUMg05B26CdPquTgpwYzvjXbjw+cbgn5F84TmYkstyflvGFJVE28UG6wkbR3jpEU38/kzMNnDAmAavDzanjk+mzOHlvl/+cxqZG0mNxcO3cHASEEVrpd3a0cuOCXHY0DTAl04BUQqB93Gs7eyg6bixuqYy9rY0YtAqWb2shNy4sEDQPxWh3c0x+DJ8eon0/eZy/TRxAn9WFxe7mwc+r+MPSQu77pCJwrBc3NnL+1DQsDre/6tPmYm+rKShP/NqWJm4/Np/ydhNOj49pWQZkEoHmfhuXzczwt65Ty2kbtPPk2jpcXh+PfVXLLYvy+GB3O51GB2nRWlaV+qtNnR4fr29t5sYFuexsGkStkOLximiUI7XUTreXbrODk8Yl8em+zqD+nj2HVKaahtw89GU1y7f6veXlUoHnzy+hKEHPPzfUB4/p8VHRYQ4K5FWd5hENo//1bSNnT0wNajsX4tdJKJD/B2kftHP5rEw0Cilen0hCuJpnzhlHU98QaQYNf1i5n6WjEzA7FDyypoaEcDWXz8pkR0M/509KoqLDRHmHBUGAsyamMDkjim2N/fxzQ0Og5dqnZZ3cd2IRJWlRbKztY3p2NHnxYfhE6DTaOWVcEhPTo7hkuPs8+BUxf/u8ijMviwtYwO6Z/DhvdsRSVdoe6ExT021ld/Mgty7Ko91op7TVGPTZOowOWgaGiNKOLALx+EScHi9JEWqyYnX8/fOD6wPJehkqrQ6PzU6cXsmopHBUcgktA0MsHZ0Q5LsiCP6Z8IDNxd9OLUYhk+By+1hX08ue4fPRq2X4RJEei5MukyMQxA+wqrSDSRlR3P9pJalRau5aUsCWhoGAXLBt0M4Dn1Zy88Jc1Aop2xsHWL2/i/QoNdfPz8PqdPPGtpagjj0SAbpNdjKjtWTG6Fhb2RN0TLVcSmK4ipPHJdFvc7GgMA7tYWmV5n4bj3xZw8f7OtApZJw3NY1dzYNsaxxAIkBK1EFVU2WnKRDEwX9T/OOq/bxwXgmq4dL3A0WWPjFYf97Qax3RJBtAJZeyr92EUi4l5rA2df+N+Hz+71ghkxzRFfJ/mVAg/w+SEKHioS+raRu0MzPH3xEnw2BAHifB6vQQofH7aq8s9S+0DQ65ufm9vXxw8ShyK57m9SQzzUWTUQpeMjIMNMilDLl8QY11wT/rXFgQR5hKTrhazts7Wqk7pKT/ytmZLCmOx+XxsaWhH+OQvwv8LjvEHLuC7PaPeLNJQ3KcZsSiX22PlT6biwUFcXx8iD95mkHDx3s72Nk8wLmTUzFoFfTbDrYhK0mLpLHXyqkTUnhwdWUgkBQmhHH2lEz+srqGik4zS0cnYHN52N1i4fypaayr6uXK2Vl8tr8TvVrOiWMTyY3Vkh2r441tLexpMTIt20BxUjhflnchl0q4YV4OOxoHEUX/gurhhKlk1HX78+AtA3aquiwckxfL5+WddJn8s960KDVxYUrKOy18PKwtP3tyGr9/r5SCBD0J4apAID9udAI5cWGYhlycNDaJgoQwbE4vHSa731tmTzuXzcrgr59VBao+P9zTzlNnj+P4Mf72ez6fyPKtzQHPGIvTw7Pr67l1UR772008cHIx2YcsjvbZRrZ4O29qOle9uZvzpqbh8vgCLd+0Siljhpsql7WZOPdfWzlncirxehVdh/isnzs5lXs/Kue6Y3JYNvWHyQ9/KTqMdt7c1sxrW5qJ1Cq489gC5uTFhPxbhgkF8v8gfRYnS4oTKEoIY2VpB0+urSPhZBVGuxufCHPyYviotIMbF+Ti9fmQSiS0DQ4Ra29AufUxlEAUL/oH2x2H6uRPOVJnLbfHh8XpJjtMTb/VFRTEY8OUxOpV7Gsz0WFycPH0dEYnR3DPqnIa+vyLmstKziM5Vst3me31mB08uqaGi6dn8Nw39Rw/OpHGPhsen0hhQjh2l4er52azqa6Xik4Lc/JimJEdTX2vjds+2MeN83Nxe70Yh9wUJoXzWVkXkVoFV8zO4p0drczOjabL5MDu8pIcqUKrlHHVnCw6jA5e3dzEXUsLuHtleUCyuKq0g6IEPY+dOYb6XhtfV3VzxsQ03t7ZitXhGRGwLp6RwaOHVLbKpRKyYnW8/7tpVHdZkAh+t0Wn28NTw08DhQl6djQN4PaK7GszcfdxBWyu7yc/Pgy5VMJjw82OBQFuX5SHQi7F4fZRmKAnOVKDWiFldFI4fTYn+9v9N4CHvqhmSmYUMWEq+m1OVpYGN+4A8Ikiq2+YSVpUsC9JWpQ2qL5gXEoE2xoGqO+1Ea6W8cCnVViHvWC0CinTsqLx+kRe29KE2eHhpY1N3LQgF5vLQ3P/ELlxYWxt6KfH4mT5tmZOHp8UWFjtMNoxDfv8/JiF0P8kH+xu4+nh78bs8HDF8l28d8VUJmZE/cJn9t9BKJAfRbpMDva1G+k2OciM0fFtXR8rdrcTp1exbEoaHq8PlULKw6vKmZ0bw7SsKK6ck8X9n1YGFrrGpkSgzRy50Im1G4ljgJK0jBHSt4umZxCvdOEQ5IiS4K/0vKlp/HV1FS6vD5VcQoxOyWtbmgNBHGD5zm4eO7WQD/f1cEx+bFAbtunZBso7zPRanKRHa7j3+EI+2tvJruZBEsJVjEuNoN3owOX1cfcUBZ4hCbdsMfPezrZAKuePH5Vz5ZwswtVyfvf6rkDq4+N9Hdy1pIBXNzcxM8efEtpY20d2nJTyDjN720wcPzoRn49AED9AeaeZ+HA1+XF6ZufEUN1t4aHTRvPommrOmpSKXCpgHHKTFxfGa1ubA0FOLhWYmmUgKdKftjhQml7TbeYf6+tJjdLQMjCERiHFcsiTz1Nf13HFrExy4sK4+s3dge2Xzsjkvd3tgZunXiXjibPGUtdjY2DIRWK4mpPHJfHUsHPipro+5ubFoVXKyI3TBTkygt+l8fAgDpAXr+OJs8bxh5VlmO0epmRGsaq0gzSDhu2Ng4HPB2BzeflwTzt5cWFUdvlvIi6vj2fX13He1DRKW418vK8jcFNIjdIglwp4vD7WVvZwx4dlDNhc5MeH8fDpYxiV9O8bTB+O2+ujrsdK2+AQMToVOXG6H6S++S76rE7eOCStdIDStsFQIB8mFMiPEn0WJ7e9v5cNtX3kxYWRnxAWWPxqGRjir59V8viZYxm0ufD6RL6u6mFWTjTv7GgNCsqlrUbMcxPRH1be6YvOo4co2geHeOi00f7SdpODpaMTyI7R0dhjxur2Ync5A+6E4P9Rubw+dEoptyzKo6F/iD2HSeoA2vuMnJvjpcyuZ3xqBK0DNlKjtHSYHLyxzf8j6jQ5qOm2MDMnmlm50UzOiGJ3ixG310diuIpw6x5qjSI+bwx/mhVGrwNe3uuf3UklAuUd5qD8tSj61RxTM6MoTNLzzw0N6FQyNtT0BGaxpa1GxqZEcOLYxMD1BH8+2OsVsbo83PpBGfW9VrJjddy5pJCabgvROgUvbWzid7MzmZZloM/iJCVKw00LcoOkc30WJ11mB51GO1vqB/jdnCx2twyyt83IDfNz2TzsGWIccvPYV7U8edbYwNeikkuQS4WgJyCzw9+yrdfiZF+biX2Y2FjXxxWzMrE6vTywuoponZIZOTHctCCP3c3bAmqcMcnhjE87sopEIZNywphExqdGYHF4SAhXMeTysrfNRL91pK1DU/8QSrmU0yYks7+9InBuCqkEu9sb+AxKmYSzJqbgE0Xqe2xc9ebugHqnqsvCTe+W8s7lU3/UzPzL8i6ue7s0MM5NC3K5dGbG/7toSCWXkBAR/JQFEKX578/r/1yEAvlRoMtkp6zdxIZhD4uZudEjZhCi6Lc+zTnEtbBlYIjGQ2bGB9hqiWbxsc+gW3sbOC34IjLYPf6vnP9WHRdNT2d2jhaDVoHR7uHhL2qwu73cvDAXj0/kpU1N3DA/h71tRhp6bURpFFw1J4tUg9+Pu6LDxOiU8BGLc1EReqbqupiu60JQavg6LIObV9UFbjKRGjker8jqsq6Ac+IrF05kX5uRL8q7EYBvz40lP9bC64Pvo9/+FmijOWnO3dxaloRCKgnkcA/F7fVx3OhETHY3a6t6uH5eTlDABn8wP3lcUtD2U8cn88iX1RxTEEf9sEa/rsfKnz4u58SxSXSanPTbXLy/q53jR8ez/NLJxIer6Le6+Gx/Jxa7h5QoDXeuLKOpb8j/tDA7kw/3tHHZzExERGLClDx6xhhe+LYBqSBw0YwMOk3+5hBl7SbC1fIRXvHgX1coStQHFmOHXF70ajm1PVZ6LU4a+mxMyohiQlokH187ndpuKxqFlPwEPXHf0x3oUHOr86elc9eKMiZmRPFNbbCfyxkTkqnuspAZreXCaem8ua0FuVRAp5LzxJlj2dE0gFcUkUsl3PZBGY+fNRaLwz3Cgram20qX2fGDA3nrwBB3rCgLGufRNTXMzo1hzP9Te65Tyrl5YR7nv7Q9MG5KpJoJaZEYh1x0mhyEqWQ/2Pjrt0gokP8EfD6RDbW93Pr+Ps6amBLY3m91EadXBhz0DiCTSlhd1sWF09J4ZXMz2xoHmJ0XE2QpC5DqaULX+TWDZ37EjoYeSk1aXlxtZWJaJHlxYby7qx2728vsvBhSIlUUJIbjdPsYl6hHo5Ay5PKSGKEmUq1gVKKemh4rt39QRmK4ihPGJhKpUQz3k/QX2Zw8LolwrZpG7Rg0rRuQy2JpNHq5a0kBn5Z1kmbQMDYlImAhUNtjpTBBzwe72/D54PllE/jXxkZeq5Nxu2wlQvnr/g9i6SJ93dU8cepKzvismfOmpPPR3o4gH5nzJiaSbN7FmvYUpmcZvrOTTEqUmhvn5zAw5CZaq0A5PEvLidNx3wlF9FqdZMfqMA23jYvWKojUyFlSHM87O9tIMWhRyCRc9toOqrqsnD4hmWc31NM8/B2Z7G4e/rKG6+blBPLfv5udye3HFrCgMA5B8AcU45CLmTkxvLW9hfXVvRQnhY+Qic7IjmZtZbCnicXh4ZPhRVST3c2W+n5m58WSHRsWtKj5Y8iK0fHP80to6bcRoZbz7Pp6RBGun59NnF7JCU9vxOnxkRcXxu3H5jE9O5r0KA1nvrB1hFb/y/IujhteiD2USI1/Ab2xz0ZttwWVXEp+fNh3tqMbHHKNWIwHf3rkpzAl08CKK6dR1WVGq5BRnBzOkNPDVW/spqLTTLhazv0njWJRUdwRm2D/1gkF8p9Afa+VK17bhcvrQyoRUMkluDz+NMY1x2Rz6/v7AlWbuXE6BmwuPi/vYkpmFPedUERypBqvT8Tt8fFVVQ96lZy/LojGJRe503MJ6n0CsZE5vFfZiNPj44yJKdz4binTs6IZlRSOz+dDIZNyz6pyrjkmmz9/UkGX2cG5k9OI1inJSNNgdrgDN4oOk4O8uDC/WmRUAlE6BSmRGmq6LcSGKXltRyvjUqayYU8vC7IFLMMz6G2NA7y7sw1BgDuOLeDTfR0cNyaRv31Whccn4vb5OHdyCnJ7L8I3K0Zcp3BTNVfMWozZ7ubZc8bx/s4WnB4vF+X7mFh1HyvDz6ei08TcvFg+29/J1EwDWxr6A++fmmmgx+JgbUU3EVoF0zJTeejLGrpMDq6bl8ODn1Vy7uQ0/vVtI2EqGVOzDPSanTxyxhjquqxcMiMDtVzK3lYjkzOjaegdoiA+jHGpEagVMup7rLy7s5UeixOfz+/LfvyYRI4d5bdmPdSvO0KjoN/qYm+LkfFpkdR0W7hyThavbGrC6fGydHQCE9Iief2Q7kExOiWO4fTJyeOS2NdmYkt931ExxQpXyylOjqA4OSJwvhqFlJveLQ08TVV3W7jvk0qePXc8WTG6YUvZ4EAerVOSHxfGJTMyeHGj33pYKhH4yynFDA65OPv5rYEAPSY5nKfPGX/EJuDxehVJEeqAuRj41yV+6mxZKhEYkxIRmNWb7G6ue3tPoBnKgdcfXzPjR+f0fwuEAvmPpMPol695vT60KhlSiQBeeG9HK3ceW4Db5+Ot7a1srOvnpgV5KGQCUVolSpmEhl4rSeFqihLD0SiktJvsaBVS5ubHMi07mvmKfVQoDCx7a//w0QZRyrq4cUEu7+30e2rffVwhdpeXIZeXhHA1f/2smni9CqPNr1a5a2kBD39RHWirtqgojiXF8RQk6GkbHKJ5wMZHezuYmRNDXa+Vh76oBvzqiyWjEpDbe7gxrQG9LobrNimDStRFEVoHbBQl6Xnwsyq0CimXzMzE4/XRY3ExLTkGX8J4umKm4paFkVj/NvKuPUh00azY1ErTgIMwhZS/nVJItGBB5rGyznMJf1pj5Lpjsnl0TQ0+ES6Yls7YlAhqeixMSo/C5/MRqVYyKy8WtVyCXCZFKRU4bnQCr2zyB51onQKlTEJypCbIe/2upQWUd5j486eVTM6IIj5cxWNnjKGsw8QbW1vQq+Usm5LKdfNyeGVzE4WJ4Tx19ngyojXkxR/ZfrWm28K+DjP7huWIqVEaLpyWxujkCF74tpH3d7Vxy6I8KjvNZMfqGJ8ayZ6WQW6cnzPc0UjCkMvDtoYBxqVGEPEdDpXdZgdSQSD6B2q85VIJWxr6+HB3O3qNnNsX5/PMurpAZ6NeixOZVMKlMzP4uqonsBgdppRxTEEsYWo5N8zLYfGoePqtTtIMWtKi1Nz07r6gWfbeNhM7mgaOGMhj9SqeOmcc17+9h9YBO5EaOX8/bXRQI5SjQY/Zwd7DnipEEZr6bf82kFd2mvm2tpdBm4tZuTGMT438TUgYQ4H8R9DQa+WyV3dSP5zXjtIquGVRHg63F5PdTbfFQUGCnh6LA7PdQ1WXGavTw8baPv48S8vxymbOOTGd9UNhvLy5iZpuC1MzDSwpTqDX4sCSN4EXvgjOrTs9Pn9lZnwYY5LDERF4ZXM93WYnNwyX/WfGaKnqMnPnkgJcHh9/PL4It9dHh8lBmEpKjE7JfZ9UkBmt47jhBgDv7zqYDjijJJmStEhMdjflbSYuHngP+VAuWuWJI66BRiHjs/2dnD81jZK0SG77oCygmDgmP4Zjsh/k72sbGXJ6Oav4fi7Pq2avKZmFRTGsr+lha8MAF7y6hytmZ7Kz0cGF0zO5aYGFSI0i8PTy6uYmdEoZyZFqoosUaFUyrnxjV+DvUVoFfz2lmLI2E51mByqZFLPDw6zcGB5dUxN0vo98Wc3183K449h8vqzopq7HyqjEcCx2Dxan/7+/fV7N7YvzuX5eDtOzo9F9j8Li8GYPLQNDvLalmefOiw60fttc38eopHBunJ+L0e7iibW1zCuIw+XxsaHGX5H63IYGbpifw1kTU6jptmK0u4ZnzEpW7ungHxvqUUgl3Lwwl8Wj4tF9TyefT8s6uPPD/YHXYUoZl87MCPjRH3CJLEmL4oMrp7G9qR+lTMrE9CgKEvw3rTC1nInpB5UggzYX+ztGNsc40trOAcanRrLiyml0m51EahQBhdDh9FudWBweYvTKEcVS34dMKuGCaWms2NUe1IIvWvfdN73qLjNnPr8lYJf8jw0NvHhBCfMK4n7Usf8bCQXyH8Hayp5AEAe/+VJ1l5mydhMVnf6CE51SxtVzs/jb59XkxoXx6Joa/jZHy2mV1yI1NbPz1M3c+1F5YMb8TW0fg0Nu/nxcNs1Wf5plBCIsHpWAQafgurdLA5I1r09EJZdQ22PlziUFPP11LQnhasJUsqAuQ5fOzOCPxxWysrSD37+3l+fPK+HBzyppNzo4qySZuflxtAzYiNQqcMgNVGScT451FyeOTWRjXV+QSmNieiSTMvzqmS31/YEgnhiuYvGoBG59/6A3+vK9JnT6yXxc1o0o2njw1GLOn5qGcchNpFaBUirliuW7EEW4YX5OkJ+K1emh2+wgPVrL/asrg4zFBmwuKjpMDLnczM6NYX11LwatgiH3yEYWDrePxAg1N7xzsH/pvjYT9x5fGCTj7DI7yI3XIYoizX02wjXy75wpFyToSTdogtZAlk1JQ6+Wce/xhexuMRIbpmRhURxZsTrcXh+vXTKZuh4r935UHjRW68AQt68oY/2w3cCCwljOKEnB4fZSkBDGprp+bn5/H1E6Jcf8G5fCPotzhAukxenBK4qEq2XcuaSQ4uGZquSwNMW/I0Ij58QxiTyzPtgG4LvUNQeICVMR8x1dgXw+kY11fdz5YRltg3Zm5UZz99LCQDPxf0e/zcnHpcM3OZmEy2dlsqGml53NgyybnEp+/HePsb1xpOf941/VMjkj6ntvkv/thAL5D6R9cIjS1pGtwco7zcTqVYFAbnX6Cy7i9SpcXh8yicBEeR1Skz9n2m6TBHWTByhrN9EzJLK6rJMFhfHsPaQ9mESAiRn+KkmVXBqkO357eyu3LMrj470d1HZZqO+1ccLYpMBi3QFe3tTEbYvzmJZloKrLQku/jZsX5qFWSHG4vVz86o6AD8cp45JwZU2gghK+3tHBbYvzaR0YQq/2L3pplVLu/bgChVSKQecPdJMzopiZEx1Uwn+AT8t7OG18Cm6fj6ve2IPN5WFObiwFCWH0W53849xxWJ1eOo0Onjx7LPd+VE670UGcXskVs7IYsDmP6L9idXhoNzoYnxqJBL+M8bwpaSPMtZIj1Vjs7hHt4laVdjD5ELVHlFZOeYeZloEh7v+kkqxYHfefOOqIOuUIjZxlU9LotTjpNvsXWXVKGZe8spM+q4s0gwa70/+UZtAqSI7SMCM7Gs8RbtJpBm2gmvaymZmUtRu57LVdAMwriOXsSSm8tb2VL8u7/m0gFxERj7A9MVzN6utmfees+PsQBIEzJqbQ1G9j9f4ulDIJN8zPZXzK/99sq7bHwiWv7ghIUb+p6eNO135eurDke/uHrqvs4d6PKwKvH1lTw9PnjOO2xXnkxYehV3+3usZ+hBv9kZQ6/VYnVV0WbE4PmTHa//di9M9JKJD/QLY3DpAbFxZk3AQwKyeGt3e0Bm2zOT2EqWSoZFLkUgkq98HAHC0Ed6oHvy+Hye5hfJKG0co2tpzmpdkTxYctGqbkxFLbbaUkPYrGPis6pSwwC+61Onl0TQ1vXTqZBz/357q9RyhR9zeyULGjqZ9FRfH021ysr+nlmrlZ/GFleZCZ0oo97SwsiuMPK/fTZ3Wxen83CeEqDFoF8wpi6TI5qe6yIpcKXD8vl/XVvczOjeGJtbVcOnNkw4mUSA0GnYJ7DpmJrqvuIVIrRy2X8G1tf0CnHhum5KHTRuNwe6nsMvP2Dr9v+NmTUvjL6oNeLYIAEzOiiOqxkWbQoJD6GzFHqmU8e+44/rCynHajnfx4HbcuzqfL5ODyWZm8tqUpYBWgUkgDs/HEcBVOj8gz62pJjlRzwbR0XtzYyF0flvHkOeOI1irptTj5prYXl8fHnHy/aqXb7CRCI+ejve1cPTc7YFtrsru5fl4OH+5p57inNrGwKI5rj8khK1YXpPEvStSTGaPl6rnZhKtldJkcbG04WAy2trKHq+b4m2AcKR99ALfHh8Xh4Xezs/jzJweDnFYhZVxqxP87iB8gzaDl4dPHcuOCIeRSCSmRGiRHKjH+gTT02Ub44exoGqDL5EAllyIVhCOO73B7ee2QReQDbK3v5/6Ti7/3uBPTo0a4XF4xO4vwQ568ukwObv9gH+uH019quZTXL5lESfp/d+HRTw7kgiCkAK8BcYAIPC+K4hM/ddz/NvpsTiq7LCybnMp7u/xVi0uL45mUEcWzhz12njI+mbgwJW63iyR9Pqh1oLgZbP0UtSxnSeG5rK44qMq4em4WX+5r4ZHMXei/uAdEkQSJjIwFz1AblsJXFd28srmJGTnR/PH4Qu5YURZwMfzDkgLK2s3MzIlhc30/KRGaoGABkBmtZVN9H3FhKuL0CqIlQ7zeaqTD5AhosA+l3+oKBCXwFwJ1mhxcPz8Hq9PDTQtycXt9RGrk/GFpPp0mf2MMj9dHZrQ2UDWqDhSkjMyxflPTy73HF3HNW3sC23osTp5dX8+dS/P5praXxYWx3PBOKQsL4/j9wlze3dmKXiXn8lmZCKKIUi4BRHY0D7K2soenBXj38ilce0w2aQYN39b28bvXd+Py+ojWKfj9wjwe+LQSQYALpqYxaPM3fRhyeXn+m3q/MiI5gjEp4Tx4yihaB+2c8dwW1Aop505OZVNdH9saB3lpUyM3zs/lno/KAzdV2SGBZ9mUtKAF51WlHXQYHbx0QQmvXjSRR9bUYHd5mJRh4Nq39iCKMDs3+oit7qq7LRQn6ZlXcOTZ+KDNxYsbG3huQwOz82K449h8Ntb1kW7QcnpJ8ncu2P5Y1Arpj5qZujzegId8SpQmaF0hQj1y1n3CmAT2tpm448MykiLUXDA1nXGHearLJALJEeoRDa0TfmBbueKkcJZfMpl/rK+j1+rk4ukZI67rvjZjIIiDfxb/l9WVvHrxpO99WvglORozcg/we1EUdwuCEAbsEgRhjSiKFd/3xv9WrE4PVZ1meixOkiLU5MWHMSYpggc+rSIrRsclMzKQSATi9UrGpkTwt1OLeWZdPXKZwO2L8lFIBT4u60QuFbggvgnD59eDpRMiUomYdi23MsDxY4rptXlx+0Q+K+viirwh9BvuOVjN6fMQt/5mdkx9h0/LjABsqO5lWqaBh04bjVQiED5cZHL/p5U8dFoxD58+mm9rerl5YS4f7mlnX5uJkvRIZufG8tAXVSRHavjnmXnkvjGH3Nl/5xujnUnpkWw/zII2JkxJpEbO4CEpDZVcQrhazord7YH8u1wqcNeSgoAi4YVvGzh/ajonjk1EIZOgV8nZ2tDHmOSRj+HpBm3QItUB9rQYsdo9TElSce0Kf853ZWkHUVoF9580itgwJaWtg0glEgZsLu7/tJKpmQYum5nJC982sKm+H6VMgsPtDbrB9lld7Gke5IZ5OagVUmxOL5MyDSx67Bs8PhGJALcfW8Cq0naue6s0sFCYFKmmstPCo2tqefj00cwriOPrqh4GbU4ePn0ML21sJEqrYHZuDP1WF69tbUaAEemzHU0DtBntFCdH8I9zJ1DRaeK057YEvu66Hhuzc6MDRUQHKEmLYnFRHBkxR1Z97GkdDHiQrK3sYWNtH79fmMvF0zP+baeg/ySdRjvPrq/jze2tSAS/182lMzICOfP8+DCOH5MYMGEzaBWMSvI3tAbYySCf7+9ixZXTKDpEgSKTSrhkZgZrq3oCT1MRGvkPXqyUSSVMzTIwIS0Ct1c8om3A4VYQANVdFqwOz287kIui2Al0Dv+/RRCESiAJ+NUF8pZ+Gy39NjY3DAQFgb+cPIqTxybx7Lnj+cvqSl7Z3MT5U9M4Jj+OCI2CMyemsrAwHongbxF2wcs78PhEXjw9k8o+L44xj5EjNpO2835Y/yCtc99Hr1HzytY6jh8dx87mQW7LtjIikeu0oHIbAy8vmZHJ/Z9WcvXcbMraTURoFGyp9+d4LQ4Pz6yrp9/m4uN9nczOjeW2xfl0muyBpgMGrYLI1q/AYSRrw7V4TvwUX1E8g0NuanusqOQSbpyfy+CQi1sW5QUUEBdPTycnVke70R60iOr2iry4qZEnzhzLqCQ9+9vNvLK5CbVcyq2L87h71X58IkzNjA78Hfw65+NGJwTNYg8wNjWCmm4LpY39QdsHbC7+9W0DWoWU2h4bZruby4Z7h25p6Pe3dwtTkhzhX+w9VId+gPJOMzF6JUNuKd1mB9OyojhhbCIrdrczOzeWL8q7Ag6HFqeHx77yN3Ku7Kz2WwL4/A6VaQYtWpWcOXkxHDsqHrlUYFvjAK2Ddn6/MJf4IxTLqOQS1MMyN7VCitnuCfq62412YvWqQMEVQElaBEuK40kzHLlICqCiIzhVNycvBq9PZENNL3lxYb+I1/iaym5eH65s9gL/3NBAQbyek8YlAf5mzRdMTePU8UmIoki0Tsnlr+8KGsPp8bGvzRQUyOGgImZfuwmFVMKYlPAfncNWyKR8l0gmK2bktT62OOHfqmH+GziqOXJBENKBccC2I/ztcuBygNTU1KN52KNCZaeZ817cxnlT0/nHYUb9f/q4gsnDMsHJGVG4PD7i9CokEgGXx0u/1eU3WXK6Wb6lGY9P5Pypafxjax87W0VAJFydyfL5z1O85mz0rh4arbGcPSkVRJHnlo1H52sFqQJX7Bhas88BUSSl4zPqHHrA3yHH4fZy+7H5PP11HR0mB8ePTkAqEfj7cWkURDlIi5DTb3PhE/156HXVPVw3LxvwL5peOTOFmI9O838oj5N4+rC7I/jTCUU09tnoNjvY0zLAieOSUckkPHn2WMJVcp5YW4NBp8RuCva6AGgdsNMyMMT0rGhOG5+MVCIwOOTiybW1AaXJh6UdXD4jE7vHh83pIU6vpLFviP0dJi6YlsZrW5oRRUiKULOoKI7ydiO5kSODfE6MhugwFdFhKr6q6MZ7SCQsbTVy/0lFrNzTwRflfu394RyTF8vS4ngeX1vDDfNzaTc6kEsl/G52JnFhKv70yci5h8Pjn1mfPzWd57+pp773oGrJ7fVxxawsJBJ/V6ID13xJcTzTsw1sqjt4M7l5YR6phwTVxAg1cqkQlCt+e3sLNyzIJUwpQ6eUoZRL6DY7CFd/t4Lm0Jn6qeOT6LE4+dvwekmUVsGrF02iOPnoFMg09FpZXdbJpro+Fo9KYF5B7IhCH6/Xx8o9I10d11R0c9K4JFoHbNyzqoJ11X6LiLl5sdy5JH+EpBNAKh35b0AQBMI1cjINWvQaOSlHuSx/dHIEfz6piAdXV2FzeZmVE8NV/8feWYa3eWVd+xYzmJmZHYeZGcvMzAxvGWamzDhlZkwxSZOGGiYHzcwoyWL8fjyOEsUuzEznG+q6rv6oLHikSOucs/faa03LQCb915xufit+NyIXiURa4FPg2kAgMKSjFwgEXgJeAhg1atRwDfZ/KT7a0UyP1Y3H5x+6Mfb6aTc58Pr8QYvSys4BHG4f66u6SAxXs7Wuj06LE6vLi0wilD12NB/5GMwOD89W6Hg6dToyXSTvbWpicnYUWVEapGIRVZ54+k7YwcF2Cw+uFpqnp4+6g7yoMGA/OTFaEsIEo6S2QUK1ub18MDeAYcNVSHqreCN9AT/mn8U1qwQFjUwiojjBwC1zsymJ8DFm8/ngHiQikZht3TL2NJtYW9kdkpizsaaPi6ek8/iqKtIjNZw1PoV+uxvlMKPPU7IiMShluC2dLI7qRtFZTocqAwrTeHybUJrJiNSwub6X5j47G2tCd8qnjUnirkX59Nnc9NvcPPx9JS+fmktszybe0kXROSA8h14lZWlxNG9ta0ciFnHJ1HQK4vWcMjqJD7c3Mz0nigGHL3hi2N9q5rwJqby9RVhYp2RFEqmTc9sX+7lkSgYvb2hAIxeTG6fnlQ11nDshlaRwVdC24DDkgwQTpZOHkDjAU6urWVgkLKZH78K/3dfBqaMTefSkYpr67MglErbW95EVo2NKViQikeA0eO+SAh74toIBl5conYJLpmbw4HcVPHJiMZe9uyvYhF5aEs+di/OH3RWWJRuZmSc4ViYfpYAB4RTzxKpKnjtjJCr5bxt68fsDHGgzs79N2DwUJxpIj9LSZXFy+bs7qegQTgub6/rY3tDHQycWh2jAJRIxI1OMQT39YRyWPq462BUkcRA2HFOzI7l1fg6Xv3ukX6JXSSlJNA65vh0NfVz01g767R7EImGBPHtC6q9q/38rNAopZ45NYWp2NE6PEIzyjzg3/v/C73KFIpFIhkDi7wYCgaEz2v/m8Pj87BmcYAwEhG7/0TXOWL2Shl47b2xqYEJGJNsb+vhuf4dQ/5uYysE2Cx/vbEEpE3PZtEwGXN7gSPbR2N/tpXP+tTh1qYxO7eb1jQ1cOSOTh76vCNb8FhbFMT0nih8OdfH2tlZuX6jn+TPKUMrEdJideH0Bbpqbw6aaHh6aoiD8/SXgEchHX/ERc5y9zM26lhXVA5w2JpkOs5OPd7awYFwXlujROFMWEV37MY78U3lmt4ip+XqeWVMTcp1WlxfvoPqlrsfGG5sauHBSGt/t7+DhE4t56odqWk0ORiQZOWdCKh29Jm7TfIXsm1cAyAQuSJ1DS8GVlPeK0CikfLyzhbPGpQwh8tQIDUlhKnY29uPx+blvaQF9AzZG9mzivXnHUytOQywSkWhQsKXNhdnpRYRwPO4acFHdMcCTp5RS2TFAh+UICa840Mk9i/O5ckYm6ZEaPtvVyiMrBFnmDR+Xc+v8XB78roLI6h5unJvDfV8d5M2TktH07kfr7addmshWVyLZMVpumps9bH3U4wtQ3mzmxk/KOWNsMuPSw4OqkwitgkdWVNI94AqeTDZUd/Pt1ZOJ0Mq59bO9SMVi7j+uEK/Pj9Xp48kfqpieG8WTP1SHKIm+LG/j+LIEpuZE4/b62dPcz3f7O5BLxcwriOWRE4qp67GGKF4OY3eziQGn5zcT+faGPs58dWvwpBClU/D+RePosjiDJH4YX+9t5/JpGeTHh+74TxyZyPLyNjoHo+pSI9TMzhfq2KsrQj1ohNu6+OtZI3njvNF8t6+dWINwMss5RhPeZ3Nxy6d7g70bfwAeXlHJmLTw31VVIhKJQk5P/wn4PVQrIuBV4FAgEHj8H7+k//+QmepZlu5nd7NwvL1+TjavbqinzewkI0rDDXOyeWZ1DeFaORanJ7jr8wfA5PDyU41Qp3Z6/JQ39XPNzCwahpl8m5MbwZ92W1EphN34HYvyeOqH6hAb22/2tXPDnGx+GHQnXFvZxcTMSHY19gdvA7hhTjb23qogiR+GqmE1lyy+mckFhSQYFOA0U5xoYFUghTcqougecHFS2bMUKg3s7TjIxBxhqOjYKDDJUWoBu8tHRpSW4kQDX+5u5ZwJKUTrhCDi6z7cw8OTJch2vRryeE3DSq4+7nJatEX0WN2cPiaREclG7liYh83to7LDgs8fGLT4reChE4rIjdbQ1Gfjpc09eMdez7ub29jdIpDv3PwYwrVyNg/aym6q7eWhE4rIidXx+e4W1lX1cO2srJAhH7PDw75WM6sOdobEtAFUdw6QGKaipd+BCPj4rCxyttyCtHYVAMlA+szH2eKbxyMrqrhyRiYGlSxoygVwfFkCH+xo5Nb5uUjFIgrj9Vw8JR2/X1BYHB1Zd/j70dwnDBHta7UwvzCWlj47n+9uRaeUcdfifBKNSk57ZduQ707ngHAK29HQxxmvbg2eGl/bWM9Hl4xnVGoENtfQzcPMvBjCfqYscyycHh/P/VgTUu7pHnCxqbaHzJ9ptg4XRpITq+eTSydQ1SmEduTE6og3ChLIKVlRISWnw7ep5VKm5UQzLefntfL9Ns+QUxEQknP6v4rfo/AzETgLmCESifYM/rfgd3je/3/ormSOezVnlxowOTw8t6aGm2ck8eXFZdwwJ5vrPizn0GCSzOEJvMPotDhDVm8/wo9rR2M/505IRTFYW5ueE4VBo8TmlZAUruGe5QeFKcK+UIdEIITYUyM0qGWSEBIHeHZNDV3K1KHvRaGn0w5muxuL3UGUeS8nFoXz528P0dLvwOX18862VrY3msmM1vLt/nbOGhf6POPSw0M8tv9yfCG3fb6PF9cJqpC/fFvB2spuVh3swOL0Ig14hjZqAYXIy+mvbOXGT8pJDNdw5Xu7+dM3h3hiVRUuj+Ck+O7WJooT9Xh8fq76sJwbPz3ApHQjlT0udrccId8VBzsJ1yhQyo58ZddXdVOSZGRdlbCQvr+tiZvnHalFy6ViZuREEakdSmQ6pQzboGrG7PDgaNkbJPHDiPrpXmID3RQl6Hl9Yz1XzsgUdooxOq6dlUVZopEzxqby1Opq7vzyALd8to/6bhuTBmP9VMd4eIhEwtHdqJZTmmQgLVLDoyurqO22safZxA0fl1PdZWN8uuACObcgJvheUiO0+PwBXv+pIeSj9vgCrDjQQXXnAH12NxdMSgs2kUckGbl0ym+v77q9flpNQ0mxy+IiM0ZLflyolHFJaTwpERqsTg9tJgeOo04RSeFqZubFMD03OkjiAPMKYilLNgb/f2SykTn5v011EqaRkxk9tBl59PP/O+Fgm5kPtzfxyY5mqgajBv9Z+D1UKxuBv3864N8AfpmGuN1PcGfces6dcTrigJek3X+ha+7zfNsM18zKQquQkhSmwuTwYP3RGyS6um7BqGpPswnXYBP0YJuFriYTGVF24YclEWqonRYX0TolSpmYvxxfxIaqbkqTjEMmIg8TQGKYimidIhiuezRcXj/NDiUFSVNRN68L3m6beg+ayAySbG7u+uoAdpeGa2d5h/DsygOdnDsxlWfW1GBxuHnllBxaOzqI0cmwSfTc8GUdWoWUK6dnYHV5h1jyflneynWzsoEOfuzSMiVuNPL27cG/B4wptEoSOH2MD6fXx+s/NeA+yo98dUUXpclGlpTEMzM3mms/OjJCb/eJ2NYQ6q8NUN9jJd6gCurUFTIJLq8PsUg4HXVaXDy6oopTRicxLSeKz3a10GN1s6w0HolIxJrBRVivkhKpVdBv95AXpyMxTE3YMbVxAJxm2nv6uWluAXcvP8ifvzlEWZKRCyanURyvp9/h4YaPyoOTpx5fgPu/OURqpAZfIMAlU9N5anV18H2dMz6VeKOKcI2c+5YUcu2He0JeLhAQph7PnZDCyoOd7G4yMS49nP9bkEuMTkEgEMDhGSrZzIzWctZrW+kwu8iN1XHZtAzyYnVMzIwMGXb5NehVMs4en8Ldx9gITMyMIFqn5LkzylhxoIMttb3MzhdIurpzgD9/U0F5i4mpWVFcPzeb3F/QrqdEanjlnFHUdtkQiQSfoHDNb1OEhGvkPHxCCRe9tYNemxuJWMTNc3N+cSx/OHh9/n+6NHNPs4lTX9ocPOnqlVLevmAsarmEMI38d1fB/PtX8f8/oFaUTELSNNTNa0lvF2RQpin3c/1KE4tLEzjQZsHi8HDvVx1IxSJOGZ1EVrSW7/Z3cMbYFB7+rpI/LSvE4vQiEQnyutd/aqC22xaUMd6zOJ+nVh/xwlhcHEeHxcmy0nhsLi/VXUK4wLWzskg0qrl2VhYWh4dn1tTwxnmj0SulIQ50pUlGVlWZaDDewPyM09F5e3EZMrFFFCD2QWaMliunZ2JyeIhUDV1nYw1KwjVyXjijjEh/N6O/mQo+F/g8dEz6E6+fuRSRVMFNn+zjjHHDq4wOKw3e22dh8ry7mZHwNfLaFdgTxrMl6hQuea+BOIOS+5cWDgl1BmHiNEIrp6bbSiAg+NRcMyuL+m4rpUlGDrWH7mJSIzT8WCGQsUQsYlRKGFnRWs4Zn8rrmxoAYYAjLVLDpe/sDP6INtX2cvfifAoT9OiUchLClGys7uXyaRm4fX6+3NPGrSOzQaoA7xEdsTt1BjVOA4v1Sh47qYTKDgsyiZjUSDWf7m4lSqcMNp6Pxu5mE1Ozo9hU08N1s7Jx+/zIJWJSIlTB3WNmjBaDemjdvTDBwAPfVQYHtaq7rBxos3DBpFSSIxI5b2JaSJ9BJAIRomCIdEXHABUdA+hVUlZcMwVw09RvRymTkBqhGVYdcjTmF8Zid3t5ZUM9GoWUW+blBBOV0iI1XDo1g0unZgCCWdg5r28PLmQrD3XS2G/nvQvHEvELRBWuURCe9vcRWVlKGMuvnESLSQgDyYjU/uYTR1XnAJ/vamVrfS8LiuKYVxj7TwmjCAQCvLulMaRcaXF6+WRnM6sOdiGXinjkxBLGpkf8bq/5P0/kvVYX39Z5iMh9DFW+h2yNHf9AJ42yDHLivLy8vo4ZeTHBMXIh0LaR+5cWMD49nIxoLS9vqKO228r3+zsYkxZBToyWpaXxfLO3HZ1Syg1zsvnsGCL7el87V8/I4pEVldy3rJCGHhsen5/6HhsDTsGvZXl5G7PzotnV1MfDJxbzyoZ6DrVbmJQVyZSsKOxuL0+v7uFpl44p2WkcNyKBG1/Zi88foDTJyIzcaNSODsbK6siPTuZgl0A6ErGIM8am8PTqKu5dWsDV33bx9ry3Sal6g464WdxSkU5cXx97mkx0W130Wt2kRWpCHO+WlcajkUsYnRpGSoSGQHg0x5cv4KIxZ1PVH+CVVc1kx+jw+QMMOL1MzIhkY03oLjszSssdX+7nxjk5XDc7myitnK4BF6sOdnHGuGSyY7RUdQqENiEjgoJ4PfMKY1HKxEzJjiLFqCIn3oDPH8CollHXYyM5TI3N5R1S8391Yz2Li+Mw2d14/cJUqkImxu720dRrY11/Iu7Zb5K5809Iequwpi9kQ8L5zMpMJ2twh1mWEkanxcnpL2+httvGmcMk04MQn9bca+OeJQU09NppNzlIjdAwKu3IYJRaLuWK6Zlc9NaO4I7dqJYBoiHTtgfaLHRaXJgcHsalR/DqOaN4bWM9MqmYCyelhUzxHobb66ff7uHyd3exu9mEVCziyhmZnDsh9WeljCDY0F42LZMTRiYiFYsJ/4VkoIYe2xAfnMqOAVr6Hb9I5P8oEsJUf7PtQLvZwYVvbqdp8OS1q8lEebOJh08s+c2N4N8Kvz9AU9/QWn7XgAuDSkZl5wAXvbWDr66a9IszAn8L/ueJvNXk4MPtzcGdlVYh5dpZudgGvJQmGciL0/HOMMGvP9X2Mi07km31/Zw/MQ2ZVMzo1HC0CgnxYSpijUrOHJtEjKmcZoWYiZmRTM2JpqXPzsc7W4I/XpvbR2OvjU01PexsMnHd7GyeWFXFuPRwHj+5mKxoLW9vaeLp1TXcvjCPcyem4vH6MTs8fLu/nb8cX4RYJAyZnP/GjuD17Wk2kRim4op0JykrruPlUbexv3QkVr+MHNUAvgg3V8zIRATE6VVsdMRTm/cnGkxuNtVX8sSY1OAu+r1tTVwxPYN+m4f6Hiszc6Owe/w8vKKS0iQDJrubdVXd7G+zcs2XVh49sZhLp8rZ22KmMF5LlGSAq8aH4/R42dFoQq+ScuOcHBxeYdzf7w/w5A9VBAJCWUloLlczpyCWBUVxFMTr6bS4WF7ezpi0MNr6nXy+q5WTRyWyvakfjVzKM2tqiDUo6be5GTOM0ZVELGLA5SUnRsedXx4pHeTG6rhwUhpqhZSPm1KIy36Gkigpe00K3t/ZwWMZAao6BrA4Pbi9foGkB8ngiz1tXDc7iydXVTPg8iIRi7hgUho7G/tJHZHAeW9sp9PiYmRyGH86rhDDUYZOzX02ttf3csu8XFr7HajkEqGUNkw9H0CvlKGWS1BIJczMi2FKdhQihGnF6s6BEA8eEAy4vtzTGpwU9foDvLiulkmZkb9J4RH9M86FR2M4yZ9ULEL9OxPj74HqTmuQxA/jq73tXDEj8xdLQX8PJBIxp49NYWt9qASzJNHIqoOCasfi9NLSZ/+DyH8v7GjoDzkeW11ettX3cda4FG79bB8quYT8OD37Bv1CJmVGMiYtHKNahlwqQSUTE62Ts/JQF3qllEidgsvf3UUgIOQK3jAnl+vePlL/HZkSxkkjE2kzOWgzObhuVhZWl4+ylDDOm5SGz+dnaqqai3PMSA1ZdFncfLxTyJF8Z0tzcLemVUi5bnYWV72/m9NGJw27A9pW38cV+XoIBEjY/mcSgLayG9jom8L6BgsFCUacHj8LiuO4/5tDHFcaz4DLy4zcaDpMzqCu2ucP8PTqGqJ0Ct5aEkanx0WEv5fjjxOhtezEqYhgv7SIBGM2SeEqDrRZeHmDEPjwY2UXnxsUfDBiP68k9bN3yvm0WEU8+UNVcBz69DHJFMQLk58Oj49XNtSzuDSBd7Y0opJJePCEIh76roIYg5K1lcJ49o1zstnR2M+Wuj4WFcVRlhzG3MIYNlT1oJBJhkhIjx+RgNXlHWL1WtEhxJe9sLaGfa1HmqtTsiK5YU422+r7WFfdw4ZBl8REo4rrZ2fz4PcVWF1enl1Tw+ljk0mP0tBmcvLd/nbOGick1Z88KgmpWMRPtb3c/Ek571wwNliz3t8qfEb+gBCn5vb6sbl9PHZSMTNzo1ldcaS5PTMvmtIkI4qjdPxHl0iyYnS8d9FY3vipgeouK9Nzo4nQyAgg9AMsDi8nj0oiSqfg9U3CqW5KdtQ/TCJZMVpOHpXIR0dF3V07K4vUn4nr+1diOBMusQjEw8lufgdMyYrivqUFPPdjDVKxmBNHJrKhpicoRRWJwPg3BFr/Gv7niby+J/QYOzc/hjkFMdR0WTl5VCIfbm8mP16PxelhaXEcHRYX725roqXfgUIq5pnTRiACsmN0ZERq2VrfSyAAOTE6bpybzZ1fHAhpNO5s7Of4EQmkR2koSTRy9Qe76bG6KUowsKfZxGljEnkseRP9EbN5e78QCzc3PwaZJPTIbXV5WV/Vw2ljkojVK0PMhQ6jJMlAl1tMnkwNHju29AX8pXcqX1daARtf7+tkdGoYp4xK4vrZ2Xh9fhRSCU6vj6fX1HDT3ByeWFXFgMuLSARnjIxB7Omnt9/JNNkWWPMnALqnPsV79f2sqjJz7aws3twU6lDXYnZRIctj1paTUcct5rbPQ7XEH2xv4uqZWcER/o7BaUYQvNQrOswsKY3nox3NiEUiLp+WQVGCgds+28e9SwvZ22Li0qnp/HV9LVvr+9ndbOKqmVk09doxO9xMy4mm2+piRmo0r/3UMORz6rW5Q0gchB9494ALly8QJHGAFpMQtF0Qr+dAm4V+u4c+m5vjRiSgkA5w89xcZFIxz6ypCS5US0vjcbh9tJudGNRyvD4/7WZH8Ed9tKdNU5+DkiQDcwti2N1kojDBwIhk4xCt9rEoSjAwOjUcj8+PRi6h0+JCLBLxwPHFVLSZiDOqqe22sb2+n2/2djAqxchfzxr1N5VAugcc7GoS5JwahYR5BXHcMjeHBUVxtJsF9VZRguFX6/A/h0AgQEOvnZZ+O+EaOZlR2t8tvSc7WlDdHI6GA2EDkRLxz9GLh2nknD0+lQWFcQQCAVYd7AxKZwGun5VNxs9IOv8e/E8Tuc3lpTQpjLe3CAnjdy3Op93kZNWhLuEfvc3MuRPTWFfVybLSRLptbtZX9zAxM5KkMDXPrKnmnuUHmJEXHSy/FCUYOGd8KqNSw9jTbA7qf49Gj9VFbY9tMOdTwl+OK8DnF5QoarmM7qJL6HEGqOqs4Z0tjZQkGllQFD9kl9nQa2NcejgPr6jkxJGJTMuJCsojo7QKjitNJKAQ03PCxxi2PUFd6il8/W3owrW9oZ+JmZG8tL6OOxflIxaBwy3G4fHx9JpqzhiXjFQiRioWMSZexp82mLh7rA9WvgCAK34szzRnsKpKOLG4vX78w0gRD9/idgy18fUHCLEWTY1QE62T85fjCkkwqugecLHiQAdXTs8kPUqDyebBZHdzy/xcbvxEUI2IRYLrYAAR2+r7ePC7ChLDVDx5cilun5/Z+TGo5RJOKEsI2UHKJeKgr/phqAftX59eU8P0YXTNe1tNXDUjk001fUzNjmRCRgQxBhW5cXr6rC6ueG93iPnSl3vauHleDlql8HOTSsRoFNIh2aQJRhUWp4ePtjfzw/VTOXn0b7OyCAQCbK3v44HvDnHVzCweW1kZ7BFoFVIeO7mY//tsP16/n/MnprH6UBc7Gk3UdFl/kchdXh82lxeDSo5ELGJTbR/XfLAn+PcPtjXz5vljflH7/bfgp5peLn57B3a3D5EIbp6bw9njU/+hycqargHqe+zolVKeOq2E9VW97G7sZ3peNBMzIkJOOf8MHI7pO64sgbx4Pa0mB7F6Jblx+mHVaH8v/muJ/GC7hZ0Nfbi8flIjNUhFIrJjdShlYiraBzA7hJSadZVdXDwlHZlYxAs/1gbLLN/v7+DSqekoJHDTnBze39bMh4MEsLmul2idgnMmpPLS+rqQ5lGEVk52jJbmPjvbG/qYnRfDyoNHdqBikWBeVZxoEBJjFudR02XnqdXVODw+UiLU3LEgj3u/Phi0Ad3dbKLpqwOcOiY5GIwLMDkrircHfUo+3tHCoyeVUJYcRoJRRaxByec7m+m2ullfY2Zq+jUsUiYCh4Z8VoEA2N0+DrSaSRgMhC5ONLC3xcyL6+oAKIjTkxeZyN3ZjWSaWsBpAqAndSHfrDuyOKyt7GZpaXyISiVKqyDHJ/iJJ4SpiNL66T4qVT0xTIXF4SEpXEWsXsn5E9PYVNtDUbwerz/ATZ/uJU6vxOn1s6Oxn7PHp+D1+XlkRVWw2eYPwFubG7lxTg7b6oUJx+4BFxZLH+e/X8lpY5IYlWzk9LHJyCRivt7bTmKYiosmp+P2+siN1VHRMcBJoxLJj9NjdXk5d8LwJDIyOZy3NjVw6uhkjitLDPlbU799yHg6EMwTPYxRKeH02dwUJhhYX9VNTqyO3FgdT/xQxbLShL9JntbYa2fVgQ5yYnXsbuoPafRaXV5WH+riprk5PLKikid/qOamuTnsazUPCVQ4GgfazDz/Yw27m0zML4rjtNFJvL9taAzhmoouxv0O6otOi5MbPt4TnGgNBOCh7ysZlx7BiOS/L8Ria10v57y+Lfh5HLY6uGBS2j98vX8rVHIpI5LD/u738mv4ryTyvS0mTn1pS/BLoZCKuXFuDk+vqWZJaTzVnVZh2vFgJ2EaOUlhKkQihkjJ3t7cyOljU5BILEPkc10DLlQyCQXxeuq6hedbWBSHSiZBp5KSFqlGJhahUkjw+AKsreoiTq/knAmpROkUVHYOEKtXYnZ4eXJ1VfDLNisvhoPtliCJH0avzU1OjBaNXILT62dxcRxyiSh4zctGxOP3+xmfqCRKZufmVU2cMz6VK94T/CvW1g6gUvVTlhwWQjRZ0drgZJzHH0AhFfPYqipumJ3DmNRwNtf1MjI5jHlFsaza10pp6Qi29+aSn1uH5uCHaMy1pEYUU9Ul6MwPtlsoSNBz67xcfjjUyYgYCSdENpG8/n5c6XNQaMO5d2kqr2+sZ3eziRHJRk4ZnYRSKsHrDxBnUGJ1eem1uanqtlGqkHHD7GzaTE4OtJkZnx7BrkYThQn6oJ78aBydAnPNWB2FPSvJiy0gNULDvrYBtjb0UxBv4IKJSg60m7npk3IunJzOwuI4zh6fwme7Wvl4cMEWieC+JQWcOjqJD3c0EwhAcaKB/HgdX+xpJWEY6ZrJ7qYsOWyI+2KCIVRlkRGt5QRlIh1mByOSjby4tobl5W1MzY7isl8xafL5A/gCfuQSyeB30cny8nZeOCGNxzd2Dbl/n83N57taOHFkIi+tr8Ph8ZEZpf3ZQOTmPjtnv7qN3kE1zKsb6xGLhNPWsXAPY0Xx98BkdwdH+o9Gp+Xvm9o02d3cvfxAyKJ2tNXBfxv+K4l8+Z62EK8Kl9fP7qZ+rC4vOoWUogQDt3y6L/j3jEgNF00dmm7j8QWQikVE6RTDjiIb1TIumSIMfcwtiOWB746k2Fw4KY0xaUYufns3o1LCuHxaJr1WF8+sqeGGOdkUJRi476uDnDU+NeTLppZL8AcYkmQCkKoYYMXESjzaBPxpydSbPYxODUcsEnafb2+q48WxPTQocrlgYhoub+iP7Lv9HVw2NYOxaWFsqu0lP15PtE7JM2uEBmB2jA6pRMS9Swrosrj4bn8HWTFa1lR28f72Jl46exQf1Zl5bGUDf556BktHGDBWL+fuKWdy7peO4Gh3XbcN1WA6UmlyOF22AVrGv8VWs5HEVgWPrijnrsUFjEwN42CbhR6rm8dWVqJRSDl7fAo3HZX7qVcK2vL3BneD5S1mypKNjEkLIzNKS80xUr3iGDlXjdExIsxFWesrGNsbOXfCAu5afuioidkWbluQxw+HuvD5A6zY38Gdi/PoMrtCzMMCAXjux1rOGJvMIycWo5BKqOocoK7bxvOnlzHiqAlFv1+Y0g0E4KzxKTT322npdyAWwUmjkviyvJX8eD3JRzUYo/VKovVKipOEBcLu8qKSC6HTw8Ht8bG72cSbmxrotro4Z0Iqk7MiidIqGJsgJ+/A45yecRKbQp0BKEsO47FVVYwZ3DnHG5S8dHYZ7WYnX+5pRSoRMzYtgvx4Qb1R220NkvhhvLW5kT8fV8iuQU8iEE6XM3/jVObPoblPOMG0mhzcsTCP7/d3hPwbJBj/vhr2gNNL5TDTlEefBP+b8B9F5DWdA5gcHmEnnPDzzZ+OYYY0THYP+XF6cmJ13PvVQSI0ctIiNfgDAa6emUmXxc31s7MxOzy8sakBnz/AshEJbK3rIT1Sw7IRCcGdGgipJCWJBqwuL7cvyOOyd3eFvN4rG+vJjikmNULNjsb+kC+nRCTC5vKyqDie1Ah1iD/I/lYLEVoZp41O4p2tR46yZ49NpLDuVdR7XsWXfzwrteP5/lAXy8vbCASEH+sdiws5+7P91PY0opA2c/3s7BAjJxCULOlRGiZkRJIWpeG1jfUUJRhYUBTHpztbaDHZeePc0by1qZFWk4NW05GTwfb6PgwqGSIR3L7OypvRC7h75tlYAkpePSeN5n47Pn8AlUzCvV8dxOryMiEzgidWCcZRN8wJ556vhOZvfY+Nj3e0oJCKiTEo8fgCzMmPDfmMQZBpmR0eRKIjLgC7mkxYnF4un57BX749RI/VjVQs4sbpSUzadS1z+w6CrRsCAXrG3ca+toEQ2wOAHw51MjYtXEhVClfz2oZ6FpXED/nedA44yYvXs6aik/e2Hon0W1vVxSeXTsColmNzefl0V4sw+ZkSxtKSOKZmRxGuEWrLayq62Nti5sxxqSFEfhg2l5edjf089J1gnXrh5DROHpVEzFGOirub+qnuGuC2z/bjHVzgtzf08/jJJRw3IoG7xsvQfPAOkzJtPDjjQp7Z5UYshksnJvH1wR4UUjF+f4DpOVHMzIuhqc/O6S9vCS6+armEjy4ZT2GCYViXS58/QF6cnsdOKuGjHc2o5RLOHp/yD5VV2k0OLnt3Z7DJDXDtzCwa++xYHB7uWVxAdszf1xCM1CqYkROq/gFICf/3U9T8HviPIHKv18/6mm6W72mj3+5hYmYEXQNOMqM0JEUI/9AOtweby4dOKeO4sgS+3tcOgEEp48LJqUzNiiDKVkX4rj/zRpoB8/Q5uM1dmJUJnPXBnmDMVnK4itvm59Jv9+Dy+Lh8eiZVXVY83gBXzcikvNlESoSaeQVxPPFDFRuqe7lyRuYQogBoNdm5bnY2TX32wSYg9FqdyGVi3t3axO4mE/cszuPGOTk8urISl9fP5roenju9jB0Nfdw4JxtfIEBmpJox5u9Rt3XCgscwx06gq8XNl3uO+D7vaupneXk77kHXQpfXzwPfVfDc6SNwewPUdA0wIzeaRSXxfL+vnRfW1XLXojxSIjS4vX4eXVkZ/FHXdtuGPYGAEFJ85thk3t7SRFWXnTM+bOKRE4u586M9wXg4vUrKVTMy6RpwUddlDaozVDIJV83IxB8QTH4ePKGILbW9mAcnViVigiR1NAIBwQPi6L/0Wt20mhxcNDmdCI0chUxMkd6BvCUAVuHH64kuZKdmCl7b0Of0eP2oVII2e2p2FPd/c5BlIxKD4/6HMSs3hu4BJx9uD11gOsxCQG9imJqDbRbuGtSmb67tZXZeDB/taA4xnxKL+Nnhmj3NppAm4mMrq9ApZZw7IRWAyg4Ll7y9k4XFcUM+n+fX1jIrP4ZI5WBiTs3nnNq2jjk5J0PAy0HPLHY2wJ+OKyA1XMMFk9IwquXcu/xAyPXZ3T5WHuigMMFAVoyWyZmRbDhqgOuyaRlkx+goiDewuDQOMaJ/eMz9YLslhMQBXttUz5vnjUGvlJEaqUHyd2aDquQSbpmfS6/NxZ5mM2q5hDsW5lGQ8Ptqxv9d8B9B5Dsa+7j83V0UxhuYnhtNj9WNVuHE5vJx86f7GJ8RychkI3qVlJZ+B+FqGe+fV0aixkO/V0GPQ8TG2j5cTjXL8s8nsns78R3r8FX/wEdRf6EsOYy9LWb6bG6a+hxoZQFOSOlE0rAWOnSow2bh82sgABMyIsmL0/FjZScbqoU6qMnuIc6gDHFh0yqklCWH8cWeVqo6raREqIkzKFlUHI/f7+P4kljGp+rpMQ8wMsLL6tN07HAm4fSJ+Ms3hzh3YipapRSJWMSNn+7jrkl5TAmzsbY/l0939JMd4+W2Bbk8sao6WBfeWt9HfpwhxFO7sc/O8WXxROsUiEViHG4vEzMjOb4sAblEUKd4fQEitXI+3tmCXCKmzexgcUk85UdlI8olYmL0SnqtHjKitNy7pIB2s4OCeD17W8whGZ8Wh+BFo5SKmZoTRXKEhnCNoJV+YW0tHl+AMLWMa2Zlsau5n1NGJ/PZrhZWHezkxJGJwQYrCOQfrVOEkOu0nCjKW/rRymXEG1XUdluRikW4PEre0d7CmUuvxeV2c8AdS71DxahUNR/taA55jvMnpyJCxLLSeLbU9XLdrGzaTA6ePKWUh1dU0mZysKAwjuxYLS39zmHNhA5zTNMxxmfb6nu5ekYWj62qCt5287xc0n9GX72hunvIbe9saeSEsgR0ShmH2i3Y3b5hNc8KqRiJCMQRGRCWBv31YO8jvPxFApE5JJRdyTdXR5IepQ2SotvnG7bE0D34bxihVfDQicXsbOyjpstKcaKRsuSwoKzwcG3+H8VwqfZWp5cwtfx30aJnx+h48/yxtJkcqOUSksPVw8p0/xvwH0Hk5a1mwtVyxqSF88iKyuDtI5KNZEZpeXxVFVfPyCAv3oDP6ydK7qKmx8knHW5GpYYTb1Ti9Phx+GR0EEVrxHw6LS7CJ52Dus1MT1MHx41ICNbS52prMH54SvAsnzbawQvt09lQI5QoDCoZ/zc/N3jc/3RnC7cvzOPtzQ1UdlpJDhe8Uqq7rOTHG1DKxOTFGXhpfR0fbG/m5VPzuOmzg0ETKYVUzKun5fJjZTctZjfHlSXgcPuI0im488sDONw+3j7koyplCq9vEXbhu5pMhGvk3L+0gBsHa8qFg9rmw1DLJSSFqem0OHl0ZRVVnVZkEhGXTEknTCPnqvd3BWuhhQl6zpsoGPR/vquV1EgND51QRGu/kxi9gsQwFTXdVp5ZU41tUB5WkmhgclbkkMYsCKWT2fnRNPXa+GZfB0tL43ls5RFi67d7eHl9PVOyI/nrujpuX5jHTzW9iID7lxYEVSUnlERR0+vmimmZ1PVYGZseTnqkhg3VPdhcPjotTpbvaaOux8akzEhcXhGv7HRw1YxMXtlQj8PjIzVCzQtnjuSTHS1YXR6umJ7J6oouPt3VQoxOyUVT0nn+xxoaeu1o5BJunpvDpOwoRASY+fh6cmN1HDcigY93HtmVJ4apyI4RzJqidUcUJtfNyuJgu4V1VV3cvTgfuVQsaJjjDT+riR4uPDglXI18sOEpk4ixugSCU8slIf2fa2ZmoVHIQBELp74LG5+EhvWQPgPRxKtJix6q0JBLJJw9PpXtx2S1Lsw/UiaJN6qINyYMe72/F7KidSFlRRCCy+OMvy1M+bfAoJIFZxJ+T7i9PvY0m1hxoBO1XMLsvBiKEg3/soXiP4LIxcDC4nje2hw6aLK7ycS0nGhyYnTIpRIue2cXX1w6hge+rWVrk9Do+HRXKwuKYrG5vBQnGllb2UVmlJZPB8OJTx2dRHK4mh8OdjCnIJb7FuWg3no9baNuYUARS2z/DnarxgdJXKeQcuLIRGwuL/csLuCtzQ1C6r1BwQllibh8fhIMSkx2DzkxOi5/dxcXT03n3q8O4vMHeGheLF/ubg1xAnR5/Xyyz8zeNittg0lEp4xO5rkfa+ge1CNPy4nm5Q11Ie+/z+am3+FhfHoEzf12FhXHhUgd/29BLt8faMfi8AY9Szy+AM/+WMuflqlCGlr7Wy0sK02gpnOAE0YmopJJUMsl9FidvLShluk50Swujsc+uIuK0SmZkRvDeW9s54rpWSFZngBzCmLos7lZfaiLsuSwELI7jFaTgylZUWyu7eVP3xzizLHJ+PwB/AQYnRpGQ6+dTQ0DdA64KEk00NxvgwBc9NbO4I9fJILbF+Tx528PYXZ4gprwLbW9nDEuiR0N/Vw0OZ2RSUY0Cgnb6nv5ck9bkJQtDiv/99k+nji5hOZ+BxaHh8dWVVGcZCQ/Ts+JZYl8vLOF3Fg9V83IZG+LmRHJRhYWxQXlhAUJek4bk0S7ycnmut5gT2JHowmVTMJbF4z+RS305KworpqRSQDw+QJ8vbeNy6ZlBDXOBfEGYvQKXtlYx9Uzsmjss+HxBlg2Ip6RKUfJ2WIKYOlzgjRUGQbSnyewyZkRPLY4lee39KCQwjWlYkZWPgoJt4DuH2tg/lbkxOp498KxPLaykppuGyeUJXDG2JR/urb798C2hn7OOsoX/qX1dXx0yXhKBg3G/n/jP4LIixIMbKnvC/GSOAyf38/C4jie+KEaiRhMNkeQxA/ju/0d3LEgjx8qujhtdBKbanvx+QOcNzENq9PDkpJ4jGo5A04vAx7YPOIxrv94L702N2NTTmGyMQJoQCkTc/2cbJ5dU0OvzY1MIuKxk0s42Gbh5k/2Ea1XcNWMLF5aX8ueZmHKccDlxe314/MHkEvElBpdrKgdWrO1OD1My47EqJazs7GfvS0mzp+Yxu1f7BcGJDg8Thz6WJvLywWTUumzudnXauayqelIJGIyorRo5RKaeh38cLBhyOt1D5MWbnV6kEnFITvnM8cmkzno9Nhrc/HwCcWCjl6r4KX1dbi9Afa3mrlwchofbGvG5w9w/qQ0LA5PsERS3WUla5imVZxByea6Xo4bkcDqii5yYvUkGJVc/u7ukGP3lTMyeeOnBuxuH0qZNEjipUlGZuXF4PL6uGNhHtE6BQ8P5lXOyo8hMUzFrDwhFHt1ZRe3fLqPiyan89Xe0ExJnz9ATbeVp1fXcGKRkS/nu5D7u1DIwrh2VjZ5cXqWl7cRo1dw35J8UiJD30u4RsFtC/IobzZx5quhoRAOj4/WfgejU4e8/SCa++28s6WRfrsHtVzCvUsKKDkqZzM1UsM7F4xlXVU3Tb12FhfHU5JoQDNcqrtUDtpfl9cZna2csHkZs5LnIPa70W1YDl4nFC7+h4i8uddGdbcVuURMdqzuVz1bRqWG88q5o7G7fERo5MOO0v+7we318dK62pCJbZfXzw+HOv8g8l9CcZKRgEgIZ/jxqGAHpUyMWCTCqJbh8wfIjtFidYUSnUwi4urBTrjL46em24ZRJWN7Qz/bG/q5aU4WSpmEWL2SrgEXV7+/m4dPLOLORfnE6BTUdg+QG2fgiUgDXp+fVzc2BHeyiWFqfqrp5cPtgpqh1+bmyvd2ceOcHPY0m4N1ValEzLScKEYkh2F2dXJSoYE11aaQ65yRE019r40njvIC+f5AB+dPSuO1jfWMiIYLJ6Xw3NojA0HxBiU2l4+uARcer59xaRF0WV3029yIgEMdFhxuL3lx+iGe58PtkAsSDEPSzN/bdmR8flt9P2eNSyE/Tk9mtJYX1wtEvepgJwlGFedNTKEsOZz6Hiv3fR06ePRVeRvXzcri6TU1+PwB9EopF00WckGtLi9/PWsk/TY3LSbHkNrpW5sbOGFEImsqu4KSysQwFRMzI3l05ZFS2/yCWM6dkEacUcGnO1t54Lsuzh6fQqfFFYzeM9ndRGoVQ8pBhxORPtln4vR4D6Wb74XYv5IQpuP8SWmcOT75F2vDOqWMWINyyPQt8It+2wdazdzwUXlwTN/u9vF/n+0jI0pL2VG77awYHVkxR/luu+1gagdlOCj/DmWHzw3WLgwH3wm93TO0TPZr8PuFxfyn2h5Mdg86pYxXNtaRGqHhmdNKSfoVpYhGLg3J/fx3RyAwfH3f4f59NPV/D/4jPj21XMrY1Aj0ChlxBhXf7GsnI0rDhZPTWF7exuljktEppJgdHvxiKakR6mAQwtnjU3l/a1NwcGZXUz/zCmMZnRpGj9VNABFXvLuL0mQjCwtjmH5cBmmyHsLE/ZTb4kmJNPLC2lo6LE5cHj8XTErj2R9raDU5mJothBcci0idnKtnZFKWqGVksoEui5PJmZGYHB6q/IkYVFLuWqThox0tiERw0shEjGopH37fHPI8JruHaJ2Ccyek8uzmHo4fkcBNc3PY3dRPQpiaMLWM8mYTFoebeYWxvL+9meXlwm5TLII/LytiZm40uXF66nqsQWXOwqI4DGopN8/NQe/toVTaRKzCxT6vbUgAxdHj8yqZhB6rmz99c4gTRyZywcRUXh30Lmk1OXB7A1zz4W7Om5AW7B8cNyKBtEgNLq+f3FgdL5xRxt5WMz5fgGfWVAdPWftazTy7pobrZmUN+TwVUgkJYUoae+2cMTaF5eVtLCqO4/Wf6kPu992BDgoTDexpNgdlZ+EaOVvqeoPDL9/t7wjmpB5+rxlRWpLD1VwyJZ2VBzux+XyIK7+BvhqIHwH8tgZfZrSOm+flhgQzTM6KpCB+eKVEZYeF1RWdIY1iEJQ7zf32ECIPQecBnDveZpt+Lu/VtqBTqTh9fCpKmYQorSI4Fv6LMCZB7mKo+OrIbUoDROX8+mOPwe7mfk59KVTKePWMLB78voLNdX2/SuT/aVDIJFwwKS2kxyASCeXEfxX+I4gcBPeyggQD9y0t4PxJqSwvb6fD5GTA4aWhx8b1c7J5ZX0dUrGYy6els691QLATjVAPmdhccaCDq2ZkIhGLeXxVFbPyYlDKxNz2xUFAGEJ56fgkcFpApeOaacm4bGb29ErY12riyhmZvLOlkbw4PZdPy2R/m5lv97XjD8BNc3N4f2szu5tNqOUSbpmTSVmSgV2tNl5aX4fL60cpE3PdrGyKEwyYnR4eXlHJq+eMHvZ9K6Vi1HIJEzIiaTc7SQhTUZRgYFeTCVGkhqxoHVqlFJvLGyRxOBxMW8EJIxMpTtDz1Mmlgh+yWka4Rs5rGxsYEebg5I6/IG/5CYCccXcSqS0JIZZonYKBQXngpdPS+eGgUAv/ZGcLL589kvXVPVR3WYnSKShMMGBZ72VTbQ8LCuPwBwJ0D7j4fPeRqdg/LyskWqugy+ri9LEpfLi9iR6rmwSjiutmZxOtk3PNzCxe31QfXHjOnZCKUSkQ6cc7mrl9QR4yiSik6XcYERo5comYy6Zm8PXeNrKitcwvjCM7RssPB7uwurx8tquFm+fmopCJESPEr93xhaDPPnlUInFGz5DnBSE+72C7MMCUEqEmN1bozRyNE0cmkhOro6bLSrReQXGC8Wf9TJaXt+Py+oIOhYdxeMDr/W1NzC2ICd3R23rhs0vZWng/53xjQ69yc92sOK7+YDfNfQ6Sw1U8dlIpo4ex8g2BXANz7oPwVNj/GcSWwNSbICLjlx93DPz+AG9uahgiZazvtRFnUFLV8c+NOPtXYVJmFH89q4xXNzSgkku4eEo6pUn/nPH734L/GCI/DKlETGa0jlNGSfippodlIxKwOr1EaOVMz43G7vYil0qYXxjNshHxtA+TQSgWiQgEwOMTtN3FicaQI7rF6eXxTSbOHJfC15ubGZkaTs+Al6IEFZnRWqRiMVdOz6SiYwCRCLoHnNwwJ4cN1d1sqesNekDb3T7u/rqSF84s48/fHAo2OJ0eP0+vrubs8al8NNh0M9tdnD0+JZgoBELHvdXk4KnVNWjkEh44voi7lx9Ao5CSGqHhm33tdA+4uGpGJm730Jp3v92DUibhpk/2ce+SAl5cVyuEOHxYTqvJwZVzbUESB4jf9QSvznmX27co2d9moShBz5Uzsmjtt3P/0gJ0KhlPrDpS+ukecHHX4nxsLi/+gGCyFGdQsr2hn+NGJDAxM4IbPz4ypTkqJYzKjgHe3ir4w6jlgvc4AdjV2B9sQOpVUh44roiVBzsZnRpOlFZOn83F0yflEa+Xo3D1sMeiDzoQHoZSJqbV5OCZNTUY1TIeOr6Yaz7cjdPjJ0av4Nb5OfRa3fgCgpdMpEbGvlZLUPUD8PaWJkpm6sjMXQzhGVR3WlhT0c3W+j6KEw2YHR5e/6kBkQieOnUES44ZIlJIxeiVUiI0cnQKaUjGKAg9DYVUjFQi5kCbmf2tZi6flinkmHr9iEVw1YxMPtnZQkXHADKJiBNHJh15AlMjXrmO1yqEn+4po5J58ofqYCh0U5+Di97ewVdXTiLp15Lgw9Nh1n0w4RpQaEH2t2df+gOBEIOww7A4PGgV0mG94f8boFVKmVsQx/ScaESIfnNK0T8LvwuRi0SiecBTgAR4JRAIPPh7PO8vISFMzcmjkzE73JQ39qNSSFlYHIfd7eODbU3EG1WMzxBIICNKE5K+fWKZQP6HR6ydx4yyS8UirpoQzVVf7OeqGVn8+ZtDnDcxjXe3tbB10JBpYmYEl0xJx+8PMDc/mkPtA9w4O5tTXt465Fpb+hwhKhUQAiUkEqEumxurI9agot3s4pqZWWxv6CMlQs3YtAhu/3xf8P53LT/AGWOTee7H2pAab06MlnCleMhYf26sjsYeGy6vn3CNjJm50Xj8/uC0psZnJgRuKyUrT+bquT+ytz+aPpubVzbUMTNdy7m9D9FpLOEv08Zw21obGrkEi8PDWa9uQyIWcebYZNpMDi6flsFzP9by+e5WUo+xCJ2aExXSSLW7Be/xe5YUcP83R2rqFoeXZ3+sYUpWFDF6BZe9s4tzBodj3t/ZxcSMMOweN0tK4tErZWweLJ0In00NSpmw0Db12blkSgYBAry2sYH7vj7ELfOyKYjXc/4b28mN0xExTP36k3oZc056gkBAyT3Ld/NTrTAYs6aii7Fp4cwtiGXFgQ7u/GI/ZcnGEDOstVXdXPL2zuC/w5KSOC6floleJeX7/Z18tKOZrGgtF01JZ1lpAmsru2nrt/PimWVUdlqJNyjZUd/H/MJY5hTEYLF7ONRuwexwE6tXESnSoBRLEA3+Mytk4iCJH4bJ7qHV5Ph1IgcQi0Eb9ev3+xlIJWLOHp8SMj0MQiN6TFo4KpmER76vIEKnYFJGJNl/Y77mvzuOPZH9q/APE7lIJJIAzwGzgRZgu0gkWh4IBA7+o8/9W2BQyZmUHU2LyY7HG0AugVvm5VLfa8Ph9qGUS7hncT57W82UN5spSTIwLj2S6s4BvtnbxnkTUoeknd80Xou/r468OMFYSz84ln6YxEGw3MyN1bOptocbZqahkwXoMNt5+tRSPt/dGpJ6Hx+m5IppGfgCAVYc6KS+RyBC3aC3yMSMCNZXdfPk6hoUUjGFCQZ+qulFr5SFTLaZ7B4K4w0h04czc6MpVvcStfsZ3jj+Mrb2qrG5fdR0WZmWE8UD31aQH6enqsPKzLwY2sxHFoA2WTJpR8++A7bMJby7p4/x2fG0mRwopBI0WjVmUxhxW//MyQmjCTv+KQKqcOp7bEzLjqIwUbimufkxuP0Brp2VhVEtI0Ij46/rj+ieh5t+bTc78Q8zzXmofYD7lhTw6IpKpudGc7DNEjSi2lzXS0migZNHJaGUiblmZhaxegX3fHVwMHQihxfX1Q6ZNH3guwoqO6xsqu1jbkEs2+r7KIgT1CFFCQYK4/X4AgEiNHKe/qmbGXmiIIkfxtb6Pq6dlcWKAx2YHR52N5mwu314fX4CwG2f7QtZTJeXtzM6LZy6LlswV7SiY4A1FV18ccVEXj93FF+Vt/PyhnrmF8aiV8lotTh5a2sTGVEalpYmsPTZn3D7/GgVUu5YmMv05PmcL/eytkZo0h67gEvEIsIG80DtbiG56J8p6ZuYGcnjJ5fw3I81yCRiLp+WQWGCnqpOK2e9dkTFE6GR88ml41DKpIhEImKH0c//I7C5vLSZHChlkt+2iP0X4ffYkY8BagKBQB2ASCT6AFgK/H8hchDq58nHNFQyY3S099up7bHR0GNDjIiTRyWgkctQSiBaJ+fU0cloFBICiLhuVhbP/liDxxdgsqEHpy+ARh5Gu9lJeqSGg22WIa97qN3CdTMyKG8boCDeQJvJjlEt5/yJaZw1LgWpWER1l5VdDSbe3tKISAQnj0qiONHA9JwoCAgNL5fPz6HBWqLL62fnoC/LjsZ+ChIMQUP6cI2ccI2MexYX0GNzEaVVMC7CTvIHU+jNPoWDnS7e3NaNSAwXTUpnR0M/EzIimJAZySMrKonWKbhqRiYLimL5dl8Hf9op47kZL5K+7R6wduLJXUZv2TWUNSp4bk0NA4ONyI01PQSmncbZ2hVIW7eTODHAbrOgBDltbDIvrK1hRm4Me1vNPPj9kRLV6WOS+OuZI3lkZSVVnQNDHAAB0iM1g1mVoRiVEobb66eux87pYyNDgqtBMM+aUxDLvMJY/vTNQR44rpgLJ6cjAjrMDtKjtJw1PhKvT5B+NvbayYjSkhOr45GVlVw6JYPl5U6SwtXcvTif7Q19bKztYVxaBJnROroHXMO6/cGRdS8vTkeryc5L62vJiNKRGK6ia8BFhEbOKaOTEItFyCRiEo0q7v8qVMVjcwvDTFe8tzu40G2q7eXmuTn0D6qilpTE89Tq6iBJW11eHvq+Ev+sOUzQdfH6KbH81OLi8mnpPLPmSEnu9gV5RGjlfLSjmTc3NRClVXDptAxGp4b/3SPvvwSDSs7xZYnMzItGjAidSobZ7uaRlaEKqHHpEXyxp53XNtYjFou4fnYWS0sTfjZD1OxwIxWLf5MfeW23lXuXH2B9dQ86hZTbF+axpDQe9X+QGuYfwe/xLhOAo+UWLcDY3+F5/2HEhQkjufW9NlrNDjx+PyqZhFGp4exsNPHaTw1cODmNpj47p4xM5LVzRuNyu0gOlBPY/R5xissoyIvh7S2NLCuNZ9NRCR8A+XF6vIjJiNJx/UflaORSzp2YSrpigFxPNVJXP3kRGTzapgiWVt7eIrjI7Wnq543NginWuqoeTh+bzIoDock5eXE6ttT1IRLB+RNSyY83sKWuD7FYxMc7WpBKRIxcEMa+ic9SJcvjgW+OKGgeW1XFfUsL+H5/B8//WMNZ41LQKqXolFIiNAoePL6Ipn47+7TZNE37GKnfxbPbbZxkMRKh9QVJ/DCe2m5nXulZBHwe7ltnYnujMJwlEsGDxxehVUhDHCUB3tvWTFaMjli9ktGp4Wyt6+XaWVm8sLYWl1eoW18/JxuH28uV0zP46/o6PL4AyeFq5hfFUtFhYUlp3LC+L/lxenJidDT323n0xFLsbi8vrqthdGo4o1LCSQpT8cTgiLxGLuGORfmMTQvH4/czJTMq2Mfot7tZvqeVjkEL1ea+Fmq7baRHqpmYGcGIZCO7j3L8K0k0UN9jozjRwAlliVR2WNjXamFGrmA/PD49nNn5sTy6sjJI0BdNTmNyVgSrK45IZ8UiqOqyDmnYvrm5gZvn5rKryYTHHxjigNlv92D1irh7u5T11bXkxepJi1Bzw5xsZGIxxYkGBpwelu9px+b20mF2cqDNwsaaHj69bMLvqnMecHpo6rMjl4hJjdCEZJJ6fAFsziPvTS2XkB6lCVmQ715+kBi9inmFsaHv0eZm5YEOXtpQh1Yh5ZqZWUzMjPzZyViX18ezq6tZP5jkNODycutn+0iP0v7X1uiPxf+35UokEl0MXAyQnPzbkk9+D8QaVZw2Opm6bhu7m/pRy8V0mB0sKoolPVLD3lYT+XF6Tn91Gy+fPZKv9nYSU5pKQd8+ro17n+rw8zFOTcXuhdIkQdoGMCLJiFohQa+Ucu2He5BJRNyzMIv9tY3k1T2Kpnld8BpuX/QiKysjg0f9jVWddFiO1DV7bW6idXJSwlU0DvqkROkU5MXqidAoKE0y8uiKyqDUTzZoNdtudnL8h3WMSEohUusfMu68pqILXyDANbOyeHFtHd1WF3qVlMunZbLqYDsKqRSvN0AgIKbDImZLk5VOu2DbeiwOk+n+iLls33Kkth4IwOOrqrh9Qd6wA1sOjy9k2jQ5XM3zZ5Sxt8VEgcHNaP8O5B21FEWmM/fsAta0+GkzOXliVTU3zMlmVEo4Tf12xmdEBE8mOTE6ZuRFc9HbQgK9RCzihjnZXDo5nXe2NXHG2BSe/bEm+Jo2t4+X19cxNTuSWIOKS6ekYq/fxokpWojQBkn8MHY19TM5K5L6HhtnjEliQkYEW+r6GJcezqTMSHqtblxeYUedFqXhullZZERreXFdDU+eMoI/f3sohKBf3lDPYyeXhBB5WoQKyTALlFgkwjNofCYTDy2bhKlldA+4KE0ysq6qm4PtFg62WyhvNfPAcUVc/t6uYNiGUibmprk53P/1Ibz+ADsb+383Iq/vsXL75/vZVNuLVCzi0qnpnD8pLaiwidQpuGhKGvcPzhOUJBrZcoxHO8A3+9qGEPnqii5u+ezIpuD8N3fwwcXjftZpsdfq5ttjJotB2KX/QeS/Ha3AUW11EgdvC0EgEHgJeAlg1KhRPx9N8k+AdHDKzKCWsepgJx/tqGdMajjHlyUwuyCWvgEnxYkGegacnDU+lQPdNvTz3iCu+VvG7r+HkXnH0xk7HaM6kQsmpSGTSPD6/Ly5uYGCOD0qmYiPFin50/YmrsrsQbN/Xcjra9bcxjWj3+XOH91IxCKyYnSMSBa8Y97Z0ojN7WNtRRfvjG9nj3I0tWZB9vfe1ib6bG6MahkHjsoa9PgCvLu1icQwNfML48iM1tLUa+eiKemYbO6g/W1KhJpwtZznfqylb/C4bnF4eWJVFXcuyufTnS18f6CDF84cyU0flwPCIhGjV6BXSrEMyg5j9ArunpWA3BVLuzgGCG2Sdg+4CNPIyYzWUtN1xB9cLhEP0aU39dnZ32ahvK6Nc8M/I+zAW8JnBCiLzmVi8U18bfMEm5uXvbuLrGgtl03LYEpWFBtrujljdAJXfrA3+NxCOHQ1T5xcyvkySYj17mHU9dhYUBxHdZeVC7ofQrr/Q5Aq2bp4zZD7Hq4+SMQiDrYPsL6qiwsnp3Oo3cJpg81shVTMn48r5K4vD2B3+0gwqrh7cQG+QGBY7xmfz88rZ49ke0M/o1LD6LQ4SQlXo1NIQ04/509Ko7FHmIFYXt7GtbOyeGZ1TbBG/siJJTjcXrqsbi6bms5rPzUgEYu4c2EeKw52BkkcBHXUtvq+oLpHIRteWdHYY6O224pSJiE7RverOnSfP8A7W5qCJ1SvX7B9GJEcxsy8I1rqpSUJKKUS3tjUQEqEmkCAIf4u6cdMydpcXt7bGmrFAbC28ueTiDQKKdkx2iG5q1G/RU//X4Lfg8i3A1kikSgNgcBPBU7/HZ73d0eMXsmZ41I4cWQiUvERG069SkZqtA6P10/vgAODUoo1oKe15Gq8RWCw1yOWSojXS7jmo308cUop4RoFXp8fq9vH/eMlRFV/SKPlFFR+69AXtvcRpfBx3exs/H4/YRoFAw4P3+9v54Y5Odz/zUFOLI0iafk8xOPv4fmDI7h7ST6njE7C5RUaaceiud/O4pI49raYQ4zExqWHM68wlq11vRTEG5CKRUESPwyX149cKg6WF1r67EHSXlqaQL/NzdWD4cW5cTqa++x8vN+MI62Egrihu8TZ+TG4PD4umJTGu1sb2d9qId6g5PaFeWw9Rs2woDCWHQ19nJPlIWz9WyF/0+17A3PYQr7YA5dOyQjWzqu7rFz/UTkLCmO4e7KezrZ9HNsfdXr8eP0BDCoZVtdQfXlxooHUcBWlCTqkKzYKN3qdZFX9lcmZpwW9dEAwbjrUbiYlQh0Ma+4acPPm5iP+8C6vnyd/qGZRsRAK3Wpy0G93s6vJRHaMNuhtcxg9VuFvx5clcOFbO7A4vIRrZDx4QhHb6vtoMTmYnRdDdecAmdE6nj99BBancJ+3LxiDPxAg3qAi5ShXQL8/wOljUwgEAjy9pnpYXX33gIswtRy9SsqolKG7030tZs58dWtQ+TIuPZzHTiol4WfCLUCoXX8/zA54b4sphMgjdQrOGJfC0tJ4ZBIxVZ1Wvt3XHly4wjVy5g/uxt1eH3tbzGyp62VyVhRTs6N4cV1dcIIy7Gfq6CDIdO9YlM/Zr24Lnkin50RR9F9qWTsc/mEiDwQCXpFIdCWwAkF++FogEDjwKw/7l+LnQk9lUjGxYRo27Wrh9s/38ugJxeTGaECupb+/n5/qxTx92ggOtJkZmRLGnIJYlFIRJZo+dA0rOD3/NJpECeRKZOA7sjPyZc0nNimDG97cGxzf1sglXD0zi5UHO3jylFISnYcg4Eeqi+KaWdlc/PbO4JDI3Yvzh1zrwqI4FFIJ3+4L/UFtqevjudNHMK8ghts+38+5E1KDjnkzcqMpSTTgCwTQKyQopGJB96ySsrAojqIEA2sruzh/UhoPfVfBbQvzeH9rI/kJBoqTwvmuyUxGfzevn1nCHV9X0WZyMDMvmrFpEexuNhFvUHH7gjxMdg+xeiWf7WqhNCWMB+L0tFscpIRraOix8cyPNVycMvw4s9zvxGSX8OD3FTxz2oiQv313oJM7SrRIfO0oZfEhyUpCiIPw/mfkRnP5tAxe3ViPy+snKVzFstIEbv1sP1KJiPfmPUXZqhOFxx14kxuWnsnsggIae+0khakwqGREahXc/82Rfv2x8lGAln4H0foju76WPge1XVaunJ7Fg98dos3sRCEVc+HkdL7Z106nxYlRLQv+u/bZPFzx3m7OGpdMariah1dUcMPsHPa1mShNCuOxlVV0W12MTg3nxjnZfLe/nUidklEpYaRGahCLRSSFqzE73OxpMnF8WeKQ78OCojgAbl+YR84x0j+Xx8eza6pD5Itb6vrY1dT/i0SuUUgZkWwccvLJitHidPvwEwhpMmqVMlweH3KpiJfPGUVrvwO5VExBvJ70wST5n2p7Of+N7cFTVrhGzsWD6Vs6hZTJ2b8skRybFsHXV02irtuGViklN1b3i8HS/234XWrkgUDgW+Db3+O5/h0wtyCWKJ2CdZXduPwBvt3Xx9wEJwty02iwuBiRFIbPJ+yQ3t3SxKhxRrD1sMj9HV9JllA16w0ydt6PpK8Gb94yKrIv47O9XSEeHDa3j8rOAcwOL5FaOU7C6Z73It7kyVQctJAdpWVarmAIJRWLePD4Ih5dWUm/3cOc/JhBx8fhj8qSgI8tjWbsbh87Gvp48pRSITVJBM+srqHb6iJap+DORXlIRCI+2t6MyeHh2/3tXDw5nYe+O8QFk4XoO41SxrNravAHhN2aJCYMD2Kun51F94Cbr/e2cd/XB7l5Xg4GtZQbPy7nniX5dFicJEeoaewV9PuNPVZSwtUUJRqQSUTstIYx0ZiK2NQQvG6/MZUdA2GAcETe32pmfkEs1d1WFhTFIpeIGTBoie3cynMLirlxVT/9gzYGjx+XQ3LvOkxRcTy7xsKysngumZJOcoSaHQ39PPR9BV5/AK8/wHOHFDyfPAVF03oABky9vLLNyjOnj+CRFZXsa7VQmKAj3qAK7qyHS80pSzZy6KiSV0qEmoqOAb7c08ri4niSI9S0mhx8Vd5GU5+d1Aj1sOlV+1stnDomiePLEnlydbWQDrW5MegZvr2hjwe/qyAhTIVe6aBnwMWcgpggCRpUcs4cn8qK/R1cOyuLj7YLoRYXTk7juLKEYbXyIKhg9raah9xe1z3MqfIoKKQSLpuWweba3qDv0BljkpFJJJz56lY8Pj+XTs1gclZUcOr4rc0NPLKiEn9AiEh8+exRweu3Oj08vrIqpAzXZ3Ojkkm4f1kBo1PCyY379d11fJgKtVyCXilD90+wrv13xv+GNudvhEYhZXJWFJOzhF1AfpyBj3Y089SnlczKjWb0xDAcXj+JYUryYrKx++x4S88iZs9zXBD2HR1552Bd/BJ2VHT4dFgcbhr7hh5FewZcTEiP4ECrhQAGXm/JJtfmQCmVcPzIRO74Yj+5sXqidAqSjEqunJ6Jzy9M0r24ro7TxiRSlmwMyVFMDVdSKKrDlpDGrngdY9IjuOK9XXh8ATRyCdfOzua5H2voGnAx4PRiUMkoSQojKVyNXinlk10tLBmRyPvbmoSsyaNSiLbU9ZEeqaWm285D31dw0qikYB3S6vTyxMoqUiM19Ns8NPbasbm8GDVy/IEA4VolUrGYpl4bN87Jwe310bvoNQzbn0LevAFP8mQOZF7CC58fIcZYg5JJmeHsbrbw9OpqvP4Ar/8k5/o5p1DT0smbJ8Qh6qkmMiaSuJVLoK+Ok43peAqfwKuUsaayiwGnlw+2h3rY1Jn8uJKTUIhE2CbexouH5Jw4KpEbP95L9WCNv2fAzfVzcpidH41OKWPA6eXN80bz/rYm1lR0kx2t5voZqVz83l6idAoumpxGtE5JlE5OXpyBVzbUcea4FP66/oj1cGOfnYsmp/PFnlD3xSnZUdz71cFgaeTpNTVcNzubA+0WSpOMBAIB9jSbuGRqOvd9dZA2s5OPdzXz6IklwVT2BUWxOD0+Pt7ezNLSeKZkRzMi2fizp08Ao9TDgtwwXt3aHnL7b2mIFsQb+OKKicHausvj45zXtwf/ftm7u3j57JHMzo+lomOAh46SpZrsHm75pJyPL51AhFaBxxeg3+4e8hoiEZw1LvVXrwUEKfBfvjnET7U9lCQauXtJ/r90ZP7/N/4g8t+A3Dg9dyzM54rpmWgUUtpNds54dRtTsyO5dkYmUr+Y/rG3YshdjMTeRbQxjW5ZODaJAZ3fT59TzLyiONZVhw6XTMqMJFwr5+u97awddHX8dl8HY1LDKYjTc9uCPHY09tNmEpJ4MqO1vPZTAw09Nmbnx5AQpua++Sl8slfPhppeihKNnFwajcK6jRER8Vw5PZMr3tsdrCfb3D6e/7GGE8oSqe22sr2+nzWVRwaXFhXHEaVVUN9txezwsr9l6G5tR2MfcqkIfwA+3N7MDXOy+eFQF7F6BaeMSWHlgQ7e29bErLxYihIN1HfbUMjFKKUiYvUK1AoJF721k0umpLP0YxMpxgsYm3UhWzsCjDNHoVbYcNv9ZERpMNk9OL0BHj8qaafX5uaVDfWUJRs577MOPl8aQdwXJ4FTuFapqY5x6hZaI/M4SZNExzAp7EuLoqlRzac3chnbB6LwS50UJRh5fNB+YHZ+DJMzI/mqvA1/IBAsVyQYVVw9M5OFOTrGmL9HYh/g/ImZ9Ns9PL+2FpPdw81zc/D4/Jw2JpkEo5IEg5LWwV24YrCccO8Swcfe5fVz0sgkwjWyIfVtmVjEzXNzWF/djUQk4s6F+dR2WYO+QbVdNi57ZxdfXjkRo0pGZccA9V1WzhyfQmmSkYJ4w7AntvoeGy39dsI1cjI91Zyl201teg5r6waQS8RcNTGG0t9YW04KVwcHb276pHzI39/c1MCM3Bjahm0+2+mzuYnQKgjTyLlwcjr3HGU2JhbBqNTfRsQ9Ay4uf3cn9YNN4t3NJs59fftvsyn4L8EfRP4bIRaLgjW3tCgdj51UwiVv70SEiJEpRjIiNWToMmix6QmQjDSgoaHDwkc7W/H6A8jEYq6YnsmnOwXHw4smp2N3e2k1OYIkfhjbGvo4Z0IKt3y6Lyjp29tiZsDpoa7bSkOvnfpeG2NSS9nc7EQsFnPbgjx6rC6eWttIUlgqEzOVSCWBIU3Bw/4rI5KMIZa5AF/vbeeDC0Zw42cV9Frdw3b9ixON/FRzREZmtnu4fFoGuXF6Tv7rluDt5S1mXjl7FCaHh9Z2BxMzI3ljcyPxRhVJYSrijSrazE7azHA4L6Smr4n7lhRQ32Oj3+7hmTXV3LVoaH+gvsfGwuI4em1uant8JDtDF5xwqQeTXMIL62oRi0RcNWhyZnf7WFIaT5fNx/E/igEP+XFWrpuVxc7GPuINSvrtHo4bkcDNn+zl4inpIYtIq8nBt/s6sLm8KAtGMsZZw7M/HtEQquUS9Cop2dE6/IGAcJ0l8ajlEvyBAKOSw6jptvLw9xXMzItBLhXzxqZ6lpYmEKNX0Dkog5RJRGTH6LjwrR3B515f3RP8LI4bkUBqhBqvP0BTn506n5+zDvug7xR2sh9cNI6xx6g8Ntb0cMlbO4IJTzfPSOKchs95ThVG8+y5yP1OUhofRxJ4GwiVBP4aDMN4oxtVcsQiYQE8FhlRmmAICMCiojhEwOs/1ROplXPtrGyKE42/6bWb++1BEj8Mk91DQ6/tDyL/A7+MyVmRvH/xWH441EWXxYVSKuHir2u5ZmY2+3d2s77qEGePTebSqZlYXR4e+LaC8hYTcwtiGZUSxpqKLr7d38E1M4fatoKw8zxWl/3JzlbOGp/CS+vr+Mtxhby4rpYD7QNE6RTIJOLgMX5rvWDX+sQppUPChCM0cuxuLzr98OPR0o59TEgx8tEeB3aPj5JEQzC7M96gJD1SwyeD5lZj08IpSjRQ3TlATZeV+5cWcOdgCPEZY5J58LuKYKni673tnDshlexoLWPS8qnosJAWqeH4MiFiTy4Rc6jNjFImZldTP7ubTRhUMsKGCSxODhfi6wAkx/qESGQ4w3O55sNysmK0TMuJ4s2fGlhSmsDkzAge/L4yRCI5NTuSuh4bP9X0cv2cbKQiEW6vjxPLEjENc9zf22JiQVEcr1ZYKZg1GmggSqfghLIEJmZGsnJ/BwSE3arV7aHX6uKlXYIatyhBT16cHpvbF+JUubW+l4mZkXy2qxW9Ssq1M4XG5rHY3tDHFdMy2FLfF3SUfHVjPfcvLQz5dw4E4PPdrSFE3mVxcuNH5cE+TSAAD61uZty8Cxmx9jxya78W7pg6CeR/u7/5opJ43t7SGFSNSMQizp6QgkgkIidWxx0L83jwO6FPEaGR8/CJJSGujpE6BedMSGVZaTxSyW+b5jwMrUKKVCwaEkytHy54478UfxD53wmRSESUVslbmxuwOLyMSw/nzkUFrD7UxfFlCcTolOQlGLjji3009dk5bkQiMXoFGVEaXB4v03KiyInTURirYVyqgS0NR3aV2TFa1HKhthlnUHLamGS8foHsonRy7lmcT5beT5SaoMfJp8f4oh9upt63tJD7vxY8SPRKKfcvK6RnwIVIBEnhqpCg5rwYDWm2rVwQn862pghe/6mBZaUJLBuRQJhajkou4d6vBKJODFMxLj0iJP19fHoEy0oT+GJPK7EGJRcVQElpHyL8HPAlc9/mNkoW53H/1wc5Z0Iqp4xK5OHBBhgQDEKenhtNbqye8ekRVHYOcMnUdF5eX4c/IPimnDshlUdWVFKSaEBvjMA2/iY0+9/Bp0+ipewGTl9uo8PiosPiZF+rmTPGJrOvxczoFCNnjE3m2TU1ODw+ji9LpHPAxQvr6njguCIqO6y8sakejy9AYphq2EW2NCmMQ+0WtAopzYFoTijzEK5R8MXuNlYe6OTSaRl0mR18tLOF8ekRzC+M5Ys9rfj8UNdt47gRQ3Mwc2J0JIepuHJGJrmxOnY09GFyDLXS1Sik6FTSoIUDCMZj721rZGp2ND8eVSY7Nqi5z+YetszUIToqTUiugZl3CU6ICKoWqUT8m8b6SxINfHzpeNZXdePx+ZmaE03J4I5ao5ByzvhUJmcJnvyJRvXPqmIMvyAz/DmkRmq4fnY2Dx8lwz17fAoZ0X9H4MZ/KP4g8n8ACWEq7l1cwHUflZMVreO+rw8SqVVwsN3Cc2truXZWVtB18aMdQsNNLhHz8qk5vLSlg7PGJ+MwdfKXgha+jYtnVbOYsSkGihL0SKRSEo0qzp+UxgPfHQr6Pf9pZgQnybej+fEDHsk5nY5JM9jSJR82Yd3rC5Afq+bB44vwB4Rj+D3LDzAuPYJonYIzxqSwr9WM3+8nM0ZHuEbOS20STpHX8t5UMV85S3F6fGys6SEvTs8rG+o4Y2wKGoWEnBhdiEUtCGZWj55YTHKEihSDhBF9B0jceAeoI5CX/R8PLJqKTqfgnPEppEWoufnTUD34F3vaePb0ETT3CdFnHRYnDT02vP4AV8/MIj1Sg04pZUtdH9fPziY3VsfBfjvNseeRlH42XVY3l39chdd/ZFLT4vCQHqmlvNnEle/v4eZ5OSwoikMhE7PyQCdNfcKRXCkXh2SitvQ76DQ7OHlUIh/tEBbJ1Ag149LDefD7Cu5fWkhjn5OiBAP3fHVEpnjzJ3u5dX4ue1vM7G0xs6lWzwWT0ihvNuPw+JiQEUlxop69LUJTN1avJCdWF2wGikTw7GkjmJojYU1FV1DJIRGLOGVUYlD7fzRqumwsKTkSDScWMWTBiNQqSA5X0dQXWq9OTM2Gs74AtxUicyAqm+4BJ6sPdfHetibSIzWcOzGN0l9pgIpEIooTjT9bDpFJxeTE/nN03TKJmLPGpTAi2Uhzv51YvYrCBD3av2FX/5+O/513+k/C/KI4UiI0WJwettT3Ut1pDQ7LHDvVCOD1+0mw7uOFiLV4B0bSFjWW7t54LrK/ysV6GxIztBc/yEnv1nDTvBz+uq42SOKJYUrmulYhbSvnp8lvsa8nQOMhOzKRm9sX5oXsjvUqKRmRanTYEYukPL1G8Fu5fUEevkAAXyBAU6+dkiQDayu7aey1kxenR69M5e0WHZeEd/L897X4/AHEIihLDsPp8fPqRiGV5+qZmcPGXTWbHDy9WhiPTwvP5o0pj1MvSefyVXbs7gpkEhGXTs2gqc8+ZFAJhHFri9PL7IJYvipv48lTSllf1c1L6+vIjtERppGxuDieQAAufGtH8Cg/tyCGaTnRKKQSTiqNI1KrIDFMhdnh4UCbmek50Zw8KglfIMDHO5qJ1asI18hp7rcTCBAS7HAYX5a3s7Aolpvn5RBnUGJ3+ShvMfGXZYVsqOnlpFEJPPht/ZDH1XZbidUrB08EFk4fm8KeZjMLCuPQKKW8es5oDrRZcHl8pEVp+eFgJ3lxWmbnxxJvUAo79AgNH148ni/3tCITi1lUGseIpDA8wzhFzsmP4YSyBCSiAE5vgBPKEoMWzYcRqVPw+MmlXPrOTnqsbhRSMfcsLiA7PgJk00Pu+/GOluDudm+LmRUHOvn88gm/SQL4r4JOJWN8RiTj/9UX8i/CH0T+D0Ipk4REcp3/xnbcXj/ROgVur58IjTwkrX5JSTyNKj26MYW0egys3N/EovQwqkpuQY6HVqeMdXsc+AIBDrZbaOw70sQ5I1+Bqu8Qryfey+PvV+H2CeqOU8cks7W2l5fOGslnu4SyxuTMSGTuPga8enQqKYuK43jo+wpcXj/ZMVpOG5OMUS0LemEAfL+/g4+ONzApqgFH2HgeOE7B7mYTXn8AnVLK7QtyeWZNDRanF5fHx8TMiJDGp0YuIXDU6lXf56LGOJEbl9cHVRkeX4DnfqzhsZNKhgRDSMQiIrRy7l5+gDsX5fHdvjYqOwbY3tjH/csK2VjdTbxeSaRWzh1fHAjxlVlxoJORKWE8fGIR725pJC1CQ0XHAK8PTmaCMN25pERQ83y4owWjSsYdC/Oo7BgY9kQjFglZmVe8txsQds9SiYiMKC09A07CVTJiDAoqO0NTcPRKGXb3kYWh0+JkW30f2+r72NnYxwWT0hAhWOfGh6mRiARtdYfFSYfFxV1f7ufm+bmIEGSFBpWM9CgtErGI4gQj9yzJ5+HvBVOu8ekRJISpOO3lrbx89kgmZf384Myo1HCWXzmJNpODMLWc1EjNkLJJh9nJi+tqQ25zeHwc6rD8WxP5/zr+IPLfERMzI/nwknGUN5m5d0kBzf127l6Sz7rKbio6BphXEEthggGv30+jW0rngIvaPh+dmQkoJQEeWVXDnmYz4Ro5Z45NZk1lF9NzovlucBza5ApwMP08Hvy6Ifiatd02Vh3sRKeQIBbB7Pxo1HIJKqkfmTKSx9fUMjEzkmfWHDGRquq0YrJ5WH5Mmrzb52dDfxizFJ2k13/GVXtHsa9NaAzKJCJePHMkT51aikwixuPzMyM3mmidgtUVXWRH6zi+LCGkTgnQ7xIHg4UPwx8Al8/PNTOzeGp1NQfaLERpFdwyP4f6wWGU+m4b9y8twunxceOcHJbvaSU9SktBopHdzaZgSeRoeHx+vD4/84viCQQCGFUywtSy4OvvbTFz/sQ0rv1wDyAQc3mLiXuXFNBpcXLSyMSQlKLrZmfj9Hh59MRivtwjSBGXjUjA6fVz8ZQ0wvr3csnEHDbX9gZPTeEaOXqVLGh5MCkzkoqj4s5WHeoiP97AU6urmZUXzWXTMrjm/d20DCZZaeQSHj6xhLNf24bd5WNRcTw6pZTMaA1njktFqxTqzUUJBtZVdbO32cyTg+qjGz/ey/IrJxL9M41sgHijoBj6OYjFDCtblIp/OQGnpd9OdacViVhEToyOmN/Za/wP/DL+IPLfETKJmNGpEZQlh9PUZyMrRkuiUcW4tAgQQbRO+HI39Fhp7nfw9uZGjiuN47EVFdw8N5eTRyVR3mKmz+Zmd7OJRUVxGNVyfP4Aqw518l2dl/TYRKAu5HW31fdxy7wcVhzoDBJRWqSaK6ZmcPrIOGr73Zw9PoUwtRyRSPD9qO2xopAOTRUacMPpmzS8MSOPM3Lg1kGu9/gCfLC9mfFp4cQYlNjcPmJ1CsZnRHDiyEREwOa6PkbESLko24vR102vNBatQVAp5MbpKE4wUtdjZW1lNzqFlCve28WsvBim50Zjtnt4fFU1i4uFkXK9SsY9XwmGVMnhahYVx/H82lrC1DKun53DqMFTUHqUlr0tJio6BiiMN/DQisqgd7xKJuHmeTnc9/XBYJmrw+KkIF7P4pJ4arut6JUy/IEAX+9tZ2p2FK+eM4o9zSY8Pj/L97TRY3MzMtlIVrSWA+0W7vpyP/cvLaCipZcH94sYn9LBS2eNoqJjAI/Px4jkMOq7rcwvFBbtogR9SMkLQKOQMDkrkn67m021vUESB6FJ/fnuFlIjNOxtMfPFnlbOHp/CqxsbmJQVRUaUFpFIRKfFFSxhHUaHxSlMuh5F5H5/gOY+Oy6fn4Qw1a+m1UfrlNw4J4dbj3IfjNTKfzZAGgRP/XNe30aHWehN5MRo+etZo0iN1PzsY/7A74s/iPyfAIlYRNpRrm7KY348qZFaEsPUbK3r5f5vhWnR7U396JRSbpmXS5vJgUomwe0PEAj4WFoSx5i0cJr7HbhEQ7XdeXE60iI0Ibvu+h47Ln+AaI2aCWE6rv+4nMZeYRc7NjWM8yalCdOXgyWAl9bXIRaJ0CuFHeznrQbGJsiAI1K9AYegzb1vMJZNrxR2rZe9s4vjyxIZEavk4tTN6NY9HHyMff7TPHnKPJ5fV89LG+rIi9Px5CmldFpceHyB4GnjMGRSMZnRWvyBQLAc09Rnx+MLoFNI6bd76LO5uG620Eg22z2UJRtICdfQ2GcPCQBxeHx8v7+DiRmRbKzpIVIrRy4Rsbgknge/qwjez6iWcfGUdB7+vpLZeTFYXV421/VSGK/nyhmZ3LP8IFkxWsamh3PxpCTWHGjn3V2CQqSmy8oXe7t45KQiWvodHGof4MkfqkgKU7OhugefP8BVMzKDJ5V5BTFE6xSEq+WkRqrpGXBx8qgkovUK6rqtrDzQSavJEaK9XlfVTWmSkYGjlCzxw+x4M6I0ROmOqD4sDg8fbG/m8VWVOD1+ZuZGc+ei/F8l2AVFcUTrFaw80ElKhIYZuVHBcfpjEQgE+HB7c5DEASo7rfxY2cV5kWm/+Dp/4PfDH0T+L4JUIubcCWnkxxtYV9lFmFpOWqSGx1dV0dLvwO31kxqh5tpZmaQYFSAS8ciKSqblRAUTfgB0Cik3z8vlxR9rhkwH9lnd7G8xMyMvOkjiYhHMLYzjyvd2B3W3kVo5f1pWSI/VHayPdjik/NAS+vVYWprA3UdN3yWEqZBJxJw7MQ2/P0CqUYRu5VNHHiDX4XA4uHdjBTWDJZP9rRZu+WwvfzmuCLlEHGJGFa6RMzolDKlYNCQRqNPiJEwjZ8DlJVwr56/r6lhf3SOYUk1KQyGVIBWLWFgUxzf7jmiwG3ptLCmNx6iWUZpkJFav5KljdrImuwerUwhF/rGqizfOHc3FU9L4Zm87t362lwePL0avlGGwN6I0H+KDPaESuQGXlw6zi9wYHY//ILgQHl03t7m9LCiKIyVcTV6cnqs/2M2Nc3J4c1MDdy3O46HvK2npF6Z3b1uQR9eAiw+3H3FaTAxT4Q/4g5K9mi4rX5a3cfXMTF5aX4fT4yfOoOTRk0oIU8up67bSZ3Njdnj4y7dHeiCrK7pIiVBz+8L8X5QU6lUyZuTGMCM35mfvcxgen58dR8khD2Nvi+lXH/sHfj/8QeT/QkTqFCwoimNSZgQXvLkDiUjEVTOy8AcCWBweyhI0tA54kfochItsXDc7G7vbS3KYmuNGJDDg9CILeEiVW8iI1rL1KK9nkQgKEww4PD7kEhGnjU4kgIgBp4dVBztDhid6rG5quqy8u6UpaDG6uDSe9gEveZ02PL4AS0ri8PkDQeJVySQsLU3gji/2B5/ns11K3h17H2k/3ULV/A/Z6slAoVBQ0x2aHGRxeKnvsXHtrCyeX1uL1eUlXCPnxjnZQQ+YYxU/mdFaNtf1cvnUDGJ1SjKjtVR2DHDfskKqOgZo6rPT2Genx+ri9DHJvLdNIMKZudEsLIzl/q8P8fXedi6bmo7dM1Sh4vb5mZARwZi0cN7a3Ei0TsFxZYmcPzGNpAgVOqWcrnY77TItCmn9kEXT6vJS020bNpfU4vDg9vrQKCTBhdLh8XHqmCRu+mRv0MXxQJsFm6uBK6dnMDBYY1fJJCwqjiMjSkeUTonV5eGOz/expb6PpHAV509MQyoRMb8glswYHV/uaeP/PtuHVCzixFGJQ67l673tTM+NRq+UoVVI+Km2l6Y+O5OzIhmZHIb2bxyikUslLCmJZ+8xdg7TcqJ/5hF/4J+BP4j83wCNvfag4f6W+j5kEhGjUsIJU6eApQUSYxmp6SataT09UePwRhTy7s52PtnZwq4z5OhXPkDZyDfw+uGbvW3EGpT83/xcXtlQR6/NRbxBxe4mM712N1fPyOStwzPxR8Hp8ZMXp6PX5uakUUnEGbXc8OlWpmZHIRWL2d7Qz5ljk1lQFMuqg53MKYjhw2PMqNrMTvaK81FN/gu37tSyq6mKK2dkIpOIgs3Aw4jRKTnQZubqmZkYlDIqOwd48PsKxqdHkherY2lpPN/sbUcmEXPR5DT8Pj/Xz87moe8reH5dLREaOY+dXMJNH+8NugSWJhkpSTISrpahkIqYnR/LouJ4qrusdFtdXDQ5nXCNnHPGpVLdbSVusDzR3GenLNlImEoWYu702e5WHji+iHu/Psit83N4enUH1V1WThuTHJRhgqAv77G6eX5tLTfOzeFA2xFduUQsYmx6BPtbzby/rTlo/aqWi9HIpSFWvAANvXaUcik3zc0hOVxQtHQOuChOFGrU7SYnWwZDwJv7HDy/VlgY8uL0iEQibvi4HJ8/gEQswjCMA2BGtJYX1tayqVaI3ftkZwst/Q5e2VDPQycUccrovz29a35hLAfbLHy+pxWxSMT5E1OZkDF8CMTPwWR3I5eKfzVjs9PspM/uJkqr+NUAjP8l/EHk/wZQy6UhZQaPL8Dmul5OHpVIzYCe7fssjElJZ0y2mISqlcgsOzij+BQmZUYit61G0r6Tcebv8CbNISUik7RIDfd/fYjGPju3zsvlzuX7gzvcR1dWcdHkNB5dWRVyDWPTwpmSFYkvEECrkLKjqY9b5uXS0CP4VWys7uHK93eTFa3l9gV5tJudbHD0HPtWGJDHcEAdza4mgWC+39/BWeNSggENAMeNiGdddRf7Wy1cNjUDl9cvuDAmGmnus+Pz++mwOHnt3FGY7B68Pj9J4WrOfm17ULseb1Tx0Y7mIIkD7Gk2MSkrEofHx5OnlKKQibn+43LOn5DCNbOy6LK4MDu9lCUZ+XR3Cx9uF8ofJYkGpmVH8eGO0OlYu9tHh1lojn6+u40fBz1xarus3Dgnh3azA51Silgk4q/r6/D5A6yv7ObB44v4aEczWqWUM8Yk4/EFeP2nhuBu/bgRCURqFViGmd5UySQcaLXwwrpaFFIxz5w2gpl5MShlwk9VKZOEpDcdhkEpo83kCDauff4AvVY3pUlG9gwOEekUUubkx3Df18JC8+K6Ws6fmBZcDB76vpLpOdG/qHoZDglhav5yfCGXTctALBKRHK6mzezgs10tdJidlCQZB6MRh9JN94CTb/Z18PpP9URp5Vw7K4ex6eHIJKEqmUAgwE81PVz/UTldAy6Sw1U8cUopI4cJy/hfxB9E/m+AlAg118/JDmnAHTcinrFp4cQalNjdPrJjtNQN5FIVnUggECDQDzaXC2+4YG6UuOFmTtU+Su2CDzjkUdHYZ0clk9BjcwVJPEonRLi1m51cNSOTD7c3o1NIuXRaBjub+nllg7DL1Cmk3LEoT3j8gJNv97UHrXIPtguWpE+cUoLXHwjZmUrEImIjDFhdRwiqpstKpFbOjXNy0CmlaBUSVh3qCibM3PrZPm6dn8sTPwhyvLPGp/DoikoSjCrsbh/P/VhLY5+Nm+fmhgwgpUVqKB+mDttpdjImLZwr3ttNSriah08oxub2Ut9t45GVlUTrFJhsbg61H6lhl7eYsXt8BBg6bOMPBChOMvDM6iPa6rVV3Wyq7eGRk4p5bGUVU7KiOHV0EisOdHCw3UJyuBqNQorN5ePmT/fy0AnFXDI1nUBAiPCL0MpZeaCTTouL40YkBH1TAC6aksZng94sLq+fL3a3MiNXKFM099mp6bJy09ycoKcNCMNQuXF62kwORKIjg2hvb2nkzHHJXDk9A7PDS32vjSdWHfH9dnr8iI+qldvdXjz+AM19NjbW9LK/1cz4jAjGpoUTpftlclfKpGTFCMEVbSYHF7+1g8qjUpIeObGYk0YlDXnc8vJ27h9cWBp77Zz92lY+uWwCZcmhzof1PTYufnvnUQ1wB5e+s4vlV04kzvDzcsr/FfxB5P8GkErEnDEmmeIEA419dmL0SooTDETqFMSHHXFvSwrX4A8EqO0W3AHFIuiXxaCfeC2iTU+DtYuM78/CsuhbJGIRXr8fpUyCViHl8mkZtJkc9Ns9ZEZrsTq93DQ3B5VMjMnuCZI4wMmjk6jqGGB9dQ/nT0pj19pQuaPD46PN5MTu8nL+xFS+399BjF7JxVPS+ammm6LEsBD99pa6PiQiEWeOS+HJH6qHDNDUDpY5fjjUxdTsKC6YlEpGtJar3tsTPKVYnJ4QY6Q9zSbGpocHm7iHUZRgwGL3cPn0TDxeP5UdA6RHaXh5Qx2BgJARebSu+zDWV3ZzyqgknvihmmWl8SSEqYKa6KY+OzmxWmoHG7YKqZhb5uXi8cHikgQ+3N6ECBEnj0oSTkPfHAyZFB1wekOkgieUJbCrSdDCT8uO4oY52fj9AeKNKt7Y1BCS+dnc78DrD9DQM8Cayi5EiIjSKfjTsgI6LS5B3poeQbhGjsvj4/KpGbywrhZ/QLjOpDA1sQYV6VES7vhif8himBimYsDpoSTRQGOvnWtmZqKUiLnk/V3saDQB8O7WJs6bmMqt83NRDBOuMRwOtJlDSBzgL98eYnJWFLFHqW16rS5e3RD63fIHYHeTaQiRt/Q7hvQlugdctJmcfxA5fxD5vw10KhkTMiOZ8Cv3S4vU8vXedgri9RjVcm5e28ntYxZSMicak7GQj+ukfLOqhUsmp/L8unoCAbhqRibPra0Jkss3+9q5Y2EeafoAtSZviMviuPRwGntt/HBIkNc19tqCUXFHI9agwKiWEaNXkBurIzlczfbGPsRiCX/+5hD3Lyvkyz1tHGq3MDU7ikmZkby1uUGwLu0MfU96pQzb4DUcaLPgD/gBUYii5avydu5alI/F6cE56JaYF6ujpc/B5johyf3McSnEG5X86ZtDNPTa0Sqk3DhXIMnDi0pV5wBLS+OHKC0KEwzsbjbx0lkjeei7Cj7d1YpIJJDulKwoJmZEsrfZTH68nomZEXy+u4XpubE89+MRgn5+bS23L8jFccxnpZSJiTMoaTcLUW+jUgVlTlOfnbVV3ayt6kYkgkdPLAmZdAU4a3wKTo+PJ36oDlHj3DAnmw+2CaWl504fQUqkBpPDzQ+Hurh6ZhYeXwCRCF7/SQg+LozX89AJRTy8QlDI5MXpuWtRHtvr+yhMMHDuxFQ2VvfQ2GdnTkEcDb12eqzCRPKbmxo4bUwy2TGhUXE/B/8wsXgDTm/Iv6fT40MsAqNGHvRYPwytYuiCEaGRh5w2QFiowtT/Ow6Hv4Q/iPw/DBFaBedMSOO7fe1Ud1o5d3wq4jA1Zo8IZ1cj6WmTyLJ3Yvf6eeGMMhoGE1yO9RJ5e0sjiye20qWdgeWosOLRqeEsL2/jxjk5uLw+wtQyLp+WEVJTn5kbzdfl7Xy7vwOpWMQdC/O47qNy2s1OonQKLp+ewV1f7ueG2dlcMS2DPU39rKvqZmRqOIXxeva1mhhwCq8ZccwkZFlyGDVdA8HhqSPvW05V1wDvbBHUKBKxiFvn5zIqNYyzxiXTNeAiWiujor4Jo1IoF1hdXu796iCvnj0qWCvutbkRiUSMz4hg82AK/LScKIxqOQqpiC/3tFE1aHMbCAjWwSNTwqnrsXLX4nw+3tHEaz81cNHktGEDiDfV9pIXpw+qOHJidETrlNwyNweJVEyfVcjXLEo0srfVHCzxLC2Jp8/q5vrZ2XyyswWHx8fFk9OZnhNFRcdACIkDvLKhnjsX5XPzJ+WsPtTFwuJ4IrQKbG5vcNLzMFLC1Ty6sopVBztZXBJPpE6BUSXjpfV1/FTTyw1zsrnuwyPBEAqpmOtmZfPg90Kpzx9g2ADwo9FpdrCuuofv9ndQEq/jullZPL2mJlizP2VkHHF6JSa7mx8ru3ljUz2xeiVXTMvgyvd3Bwk6UitnZErYkOfPiNZy89ycEHOx+5YUkBrxx9AR/EHk/5FIi9Rw+fRM7C4vdd1WOgdcGCMyscnTMXpFnDPoWX7nl/sZnx4xrHOd2+vHFFbEt7s6KE40cMX0DF7ZUC+UQMam8MB3h4LOhCePSuSFM8qCYcM/VnQF48qUMgltZifXzsyi3eJk1cFOttf3cd2sbB5dWYXHF+DW+Tlsb+ijtttGaoSah08ood/mxuXz4/EFeHyV8OOcXxiLze2loddOQbyB00Yn8f6gMmZyViSPHbWY+PyCZ8vTp5RyqGOAKREmMho/RmLr4pysNHaWTObiFQ7cPj/NfQ7OHJuMWAS7mkx8s7ed/1uQy5SsSFxeP7sa+3l9Uz33Lingqvd3D/msKjsG+GRnC6//1MBtC/L4oaKbVzc2MDYtHOgNuW+sQcn0nOhgRJ/L66OyY4Anfqii3+5hQkYEl0/L4LbP9zEyJZzZ+TGIRSK21/dhCvPww8EO/ryskB2NfXy4vYkem2vYqUqzw0PPoBLncABItE7J4ycLgSf9dg9yiZi7F+cjFov4sryNQIBg7N21s7L4sbKbadlRQ4ayXF4/bWZHsDw2IzeKlPCfJ0y318fza2t5c1ANtammhzGpYdw+P4uXNzZxYo6cU2IqkYmKWHmgk5s/FVwzyzHT2GvjtXNHc6DVjFYpJStKS5/NTb/NRdhRfuVKmYSzx6cyLj2CTouTBKOa7BhtSI3/fxn/EJGLRKJHgMWAG6gFzgsEAqbf4br+wG+AWiGlMNFI4eEbAgHae/u4/rMKcuN09FjdfLW3ndw4PQqpOETjfPGEeOpdanY31xCmkbOxpocLJqWRF6fj/m8OhdjLfrSjhbFpEajkYry+QJDE5RIxN8zJ5tk1NfTa3IhFcOa4FHJjddz7leCBfsu8HJ5YVR00DmvotfN/n+/j+VNLOPetnSwrTeSSKRlkRmv5fFcL9w5awq482MmflxVyR5SGWIMKlVzC2eNT+HJPWzD13WT3IBaLmJiiJq7ua9ao5/F2q5RwB5yV4+OOqWHctaYXqUTE82trSI/Scs3MLFIj1Nzx+f6gZh7ghJGJNPcJC8i6qtDEpjCNHJtb0LZvqO5hRJKQk3r+xFS0CmmwNKVTSMmO0XH3VweCqhSX188Nc7KDpZ1Ntb3IJGLOGJtCXY8tWDtPiVAzJSeSE0cmct4b24O9gJp1dTxxSskQCWdxooHdTSZyY7XMOEqzPSYtgq+uOmKMlRapobnfjlwiHlbjLpEIvZRjIZeKyY7RMSEjgmUjEtAof54qmvocvL1FIPETRyYSZ1AK/YRAgLdm+cjceAUO9QlsbzLTYXFww5xs1ld1s72hn4oOK9UdA0zJjuLCN3bQNahCmp0fw5+WFRJzlIJGo5AGM0r/QCh+2Qnn17EKKAwEAsVAFfB///gl/YG/GyIRWq2e3DgdA04vF01OIzFMxa6Gbl46rYAFRTGUJRu5d0kBKTHhXPPJQYoTjWyq7aW+x8bza2spbzEPm7HYZnaQEqFGJRejGDRVmlcYy3tbm4Ik7Q/AW5sb0SqluLx+ZBIR0TpliPsjCATc3O9Aq5Dx0Y5m3t7SyLb6PtYcE3n38c5mylLCeXl9HRe8uYPv9nVw+bQMksKF5tbi4jg6LE5e2NTOi8451JHInhYrXx0yc/pyG6mxgg/MptoepmRH88OhLp5aXc2+VjPpUUd2mPEGJfMLYnh6TQ2TsiKJPYo8ZuZF09BjCx79HR5fsOm3trKLv55ZxktnjeSls8r49PIJZMfo0Py/9s47vK3y+uOfq70sS/Lee8axnb0TQgbZYe9VNmWWUGb76y5QoEDLaCmUvfdMSAKB7O1MO473HvKUbA1r3N8f11HiOJTSQBPK/TwPD/GVfO+RbJ/76rznfL8aFXanl4FAkOtnZPDVUa9rbYWdXrcPrUrB9KxI7p6fy9gUKz39ftIijZiOSpqPrjrIUxeNDo3tj02xsnBkHKvL2vD6g6QfZaCQaDUwPi2CrJgwVEoFyTYjP52ZMeQ5TrePSek2NlR2MG9E3JDHlAqBU3OjuXFmBoVJ4Qh806pXenPGp9lwDfj56xeVfLq3ld9+WsEvt6rpyLuY1/Xncc7fN/PnVRU8vPIg+fHhoU8aiVYDf/ykLJTEAVaVtg0bMpL5eo5rRS6K4sojvtwMnH184cgcL2E6NfNGxPL0umrWVth58KxCYoUu0tdcxhRDJM3jr6BJYaDPryYjykRTj5u0SEOoI2NzdSfzRsTy6REftwUBBAQauz3sru/i10tGcN/yMpIjDEMsyw7h9gZYODIWXyCI86huE5BkAhItGq6Zmsj9n1UP+/5DiCJ8sKuJ3U3SH7S9z8ufPivn5llZrCptZVyqbYi5RbhezdXT03l0dQX+oMjutgHSIgzUd7vRqhScPSaRteV2TDoVt8zOwuH20+bw0OrwcNNru1g2NxuAv1xQzLaabpJsBt7d2TikPfCU7CgeWX0QpULg/HHJrCpr46uDHYxMCOeC8Unsbezhznk5qJQKBI49wn7Iqm5zdSc3z8rizncOT77q1coh2iwgzRWMTrZy25xsajr72d/k4IEVBwiKUpfON6FUCFwyMYW8ODObqzrJig5jcmYEwaDIeyVNtDnc/OGMAj7Z04JJq+LCCcn84r29VA/6YCZZ9bxwxfiv1VtJshk4f1wyUWFa/vLF0Pr85gYXVXMv4Y/PDS1ZvbSplltmZVHf6SIlwjBskxegpXf4gkLm2HyXNfIrgDe+7kFBEK4BrgFITv7202My/z5jU23o1Uqc/X2oRTc3ftLGsmlPkqywE6UXGFHzPiXGKdwwJY0ur4Iwg46ttV043H521vfwj0vG4PEH+bK8negwHZdPSeWt7Y0sKowlNTIMgiLPXTYOly/AymhTaHPwEN6ANOCzqDCOg+193LswD4fbx4ub6ujsH+Da6Rno1CpyYsJ5/vJxtPS6CTdoeH1b/ZDywaWTUrjnvX1Dzh0IiuTEmEiLSKfSLg3mrKuws6Wmi163j6AohvwrFUo1WrWCcalWdColvW4vS0ZGccNre3F6/QgCXDQhhYo2J25fgLpBo41NVZ08srqCnJgwzh6biNsXwBcIctGEFByeAS6fnMbUzAieWVfNukE99pqOfjZVd3J6cQL3D27IXTs9nVHJFqZlRrKuUhqeUisFLp2UwkOfHeScMYmhmvUh3L4APS4fZr0qtEH9y0X59Hn9aFUCLT1u1lbYMWpUXDQxmT6Pn5Ze9ze24NmMWubmxzI3P5YBf4DGbjeCIHDF1FRqO10caHFw66ws0qON3P3O4SQOUgvkpqrOr03kWpWSm2Zlsruh55hmKt0ehvlpBkXJou3lq8ajUMDs/GjeKxm6KMj4muvJDOcbE7kgCKs5tqX2vaIofjD4nHsBP/DK151HFMWngacBxo4de4wft8x3hVIhUJhkYU8jbKi0Y9GruO4daeVrMaj5yynTmRLcieDaR1/yTB7d1cdlk1JRCAIGjYIwLeTFhpEXZ6ar38sTX1QSFKVxmXiLDpVC4O2djZi0KpbNzebe9/fR0SfVyK+cmoZKIZBkM/CrD0tD5ss6tYKHzymirtNFcZKFa18toaNvAJ1awS8X5pMcoeMfl47lw13NODw+lhTFo9coMWpVeP2HSzPzC2J5c3sDXxw4XK64Zno6rQ4PdZ1SLVgQBHQqAYVCwe8/kTovLpmYzKWjLFz1RlmoNj4hzUa4Xs0lE1PY3+zA7QtQbe8PTRWWtzm5f/kBxqfZsGjUWA0qREQCQTc7G3pCSfwQdqc35LUK8MKmWtSqdKxGDXecloNZr8bp8fFBSTNXTUsjP85McoSBXpeP6o7+0PfptUoePKuIVoeH/DgzceE6fvnBftaUtzMyIZxbZmWSG2vm1x/u59xxSTR0uf7tXurmHjcr97fS1T+APyhi0ig5d1wyxUlS7bnXPTSWQ9R3D9d/P5K4cD0mrYppWZGsq5BuWkk2PVEmLR5fgEiTJtTOCJLee5hOxcXPbMXp9fPLhXk0drvZVtuNVqVg2dwcCv+NTxsyEt+YyEVRnP2vHhcE4XJgETBLFI91P5Y5URQmWkiPNDItK4rnN9ayqrSdrGgTQkQcqs9uhoE+wjf8mXPPXsPV79TR2CPVKBcVxnHW6ARW7GtjbYWdgsRw5ubH8NqWeq6Zns7Nbx8uZ7xb0sTvlxZwoNVJRrSRHpePX324n+tmZISSOEhThC9uquPCCck8vbaKqZlRvL+rCY8vyC8/2McLV4zn0dUVJFr1nDYilsZuFxlRYdx5Wg53v3fY23NKZuQQoS6AFzfVcsWUNP65oYbM6DB+MjmFyDAd/1h7uGzzypZ6ZuVEUtcl9SzfNiebHXXdPPllJXFmHbfMysJqULOjvptYsy7UOx8Iimyq6uTSiSno1UqsBqnun2DVDdNyB6lsdAhRhPxBV51u18BgP7meyyancO/7+/D4gqgUAtefksGne1upskvGDJFGDR/taebKKWmMSrHyyd6WkLHy3qZe9jb1MiM7iuwYE6IoHlNT5esoa3Hwzs4m9jb1olcruXxyKjvquzlthLRWC9eruXJqWmgG4FBbaGqEkY4+L5GmofomDV0u3L4A8eE6wnRqfr+0gC/L21EoBPY3O7AaNXS7Brj51Cye31hLdUc/yTYDv1yUx+1v7QndVH//aRkXjEvi3gV5mPVqUiOMckfKt+B4u1bmAXcAM0RR/Ne3bJkTgkmnpiDBwn1nFHLxhF72NTmo7hogfdKvSNj4S+hrI9u5hbfPyKHW7qDXlM5L29u5/a093DkvG4NWwYGWPl7YWMvPT8sZYp0GksfmxqpOVpa2csdpObyxrQG9RhnqLDmS+i4X6ys6+PJgBzfPygwdtxo0lDY7ODU3miSrni8OtLN8Xyv+oMjV09L47dICWh0etCoFqmP8cXt8QeItkiHCOzsa+OkpGZzx1KYhzwmK0OUOkBNtwGbSs622K7RybO71cN/yMp64aDSjkiw8+Fk5y+Zms76ik6Ye16D6oIkWh5eGbje+QBCXN8BlkyQNmUnpkmqiWa/Cc8Tk5DljEml3uIkxa3EPBChICCcqTMfSx9eHxLL8QZEnv6zil4vy+KrcztSsSGo6+smIMrG+sgOjVkljVz83npqJUhAQBPD6gny0u5n7zyrE4R4Ysjn7r+jz+nh5cx17B/cc3L4AT31VxR/PKCAYFFEoBA62SUqSQVHkwvHJeANBjFoVH+xqIjpMy6w8SdrWPeDn/V3N/OGTMvq8fiak2fjDGQVkRoeR0mnk8ue2ha6rVyu5dXYWxckWFhTG0efx4/EFh3i2iiK8urWBiyamfG0JR+brOd4a+eOAFlglSJ6Hm0VRvO64o5L5zlGrFCTZDDy+ppLVZe08rA/n2lHPckaBlZf2+5ibZWKCrYVOcwDTtCTq7T1Y1QFGJVnJiwtnW00XH+xqOmYLmy8QxKRRkWQzoB8cPjpWcplXICknCgJDPDIvm5yCSinw1Fc19Lh85MeZuWt+LvctP8Cz62u4dXY2T6ypZE5eDIuL4oe0/AGMSrZQY+/n2Q21LBwZh6AQiDJphwhqZUSZMAT7eWheNDU+C7sbeilrcYQ+7gdFKG2WdFJ+tSSfdQc7GBEn2dc5PdIQ0VvbG8mMllbBH+9u5tdLRjA9K5I3tjeG9NPn5MVw1/xcfIEgLT0eipOtJFj0aFVKzHo1JfXdwwSvAkGRBIueGdlRGLVK3qvpYt/g5t++pl4umpjCza+X0DPYwhgVpuXO03L43celFCSYaeh2s7Wmi1HJFuYVxJEZfexE2Ovysb5yuNCZ2xegtMVBuF7NFc9tDTkWbavt5sxRCVS397G5uotTc6M5JSea0uZeDrb1YXd6yYgysruxly01XTy88iB/OmskT35ZOez8zb0evjjQTo/Lx3Uz0ml3eobFYTWosRo0w47LfDPH27WS+c3PkjlZiDBp+eMZI7lgfC92p5e0KCPh8eFcHxtkR103HcEwXAdWc/vWDKJNau6casViVFLtkGriWpWCG2ZmhlZ0IJUSJmdEYDVoeHtHA5dOSmFHfTdrytu5aVAyt8/rZ3FRHFaDhsZuN+eNTWRrjVRf1qoE8uLMXPPSjtBGWWmLA+Uu+8GI7AAAP5hJREFUePnseHxd9cSniMw7S0tnWAytfUH+b3E+r2yu42BbH1MyIxiVbOXhleXo1UqmZkZgd3j5zdIRvLy5jn1NvYxOsTJvRAwRxgC3fdpMhb2WcL2aa6en89aORmoGa8IqpYJXttTz68X5LCqM54+flhFE6iPv6HNyzthE3thWT7XdxZVTU1m1v5UIk3bIQM2qsjayY00YNAqum5FO8lGTh1FhkrRBzxE+piqFQGffAM9vrOGiialcOyODHXXdvL6tnor2Plbsax3yfLvTS8WgGJkI3Dcotvb5gXbe2NbA69dMpMfto7zViValoCAhnJQII2E6NTmxYexrGtoh0u8NsPSJDTx4duEQ2zmA93c1ccPMTL6q6CAv1szq0lZufK0ktCl9+eRUQBIeW1naxj0Lcun3DpUoOPR7Mj7VxrhUG43dLqLDtNwyKzNk8qFWCvzp7KJ/6Scq8/UIJ6KsPXbsWHH79u3/9evKfD1vbW/g7R2NLBsFKn0YO+wKanr8jEuzkRlp5EBbH++VNJEWYaAg0cLbOxox61RMzYrkpU11tDu9LJubzWtbG7hgXBL1XS4iTFpGJoQjIhKmUXKgrQ+nx0ecRU+EUUu3awCtUkGrw8OvPyodFtOqOW1krfsZojWV/sVPM+MVB1OzJM2TjGgjmVEmkiMMtDo8mLQq8uPM+AMBKu39gIBKIRBp0iIIkqnxfcsPDBFzUghw6+xs/rzqIOPTbCRa9Wys7GROfgw767u5c14ud7+7N6QhDnDX/Fz+8rnkAvSLhXlsrOzki8H69SHGp0nSqsvmZDMhfbgu99qDdm58dScOjx+dWhqH31jVwehkK88P6poYNJLYmVqp4PMD7XxxYOg1ZuVFMzHNxv0ryofV6R+/YBS3vbk7pG2SaNHx4pUTCIpwoMXBHe/sCWnnzMiOIlyvxj5oyFHR7qTa3s8ne1sQRWnj/PpTMuh1+bh0Ugpn/W3jMLmH2+ZI72FubBhvXDORrw7aufkIn1KFAG9eO4lRyVaCQRFfMIhBo8I14KeirY/OPi9JNgO+QBC9RkWKzSDXx78GQRB2iKI49ujj8oi+DAAzc6JxuH38dGUVWpWbn83J4qXtlby/q5kpGRF09HmZkR2NXwzS4fBw97wcPH6R+q5+rpuRgcWg5q5399Lj8vG7T8qID9dxxqgEdGpo6ZVqs519A8zOj0GnUvDqljpW7JfUs361OH9YPNFhWswuSR9c6K5Fs/ERzhz5c97f38miwjie21AbEvZSCPDrxSPY3+wgK9qEw+Pn6bXVzM2PJdqsZWNlJ6fmRQ9T5AuK0ubebXOyqe3s592dTVw/I4OoMC2V9j521ncPSeIAr2+tZ25+LO/vamJTdScT023DEnl2tInVZe1U2fto7fXQ2T9AVoyJkQnhWAwapmdH8fFNU2l1eFApBG56bRfnj0/iz6sOcuusLF7eXB8qC+nUCh49r5gvDrQPEY0qTAgfMpl6JG0OzxCBqsYeD1tqOml3eHl+Yy13zstFp1bQ6vBS2tyLVqXArFOF5AmKkyzcMDOTx7+o5MLxySwaGUdyhIGq9r5hSRwkdyWtSsGvl4wg3KDhlJxonrhwFM+ur8Gi13D19DSKkywoFQJKhYB6cA7R5Q0QYdJg1qm46929bKnpCnWsXDAuibBvsYn7Y0dO5DKAZDt35bR0FhfHoxQEnF4fapWAayDAF+V2frEwjwdWHOD04gTiLHrWVXQyNtVKvzfAhgo7F09OJdmmD5UAfEGRkYkWGro83Pv+vtBH8d2NvdwxL4ctgy43AAdaHZw9JpG3d0iJW6UQuG+6lpgNz4aeo6lfx7zFv+Qfm714fQGunJrKG9saCdNJEr3L97WwsaqLh88pRAyKzMmLGTKsdNGEJOaNiGFvk2NIco42a3luQw09bh8XT0im2zXAU19VsbQ4/pi+lt2DPd4AkSYtAVEkO8bEwcGbRHqkkQiTllaHB1GEW97YFfreh84uxGLQUNHeR1a0ieIkCw6PpJni9Qcx6yTxsCNr+x5fkA92NfP0JWPY09iLTq0g3qJjT2MvO+t7WDgybsjrPDUnkmizlltmZbG/uTd0s2vp8bCv2UH3YFfRzbMy+ftXUlfPtTPSeevzw6Yauxp6KEwM58kLR1GUFM6AX6TH5SPGrCM+XDdErVClEMiLDeP1ayaGNH3MejULC+MZlWShtrOfdodUCsob7OAZ8Af56mA7v/pgP21OLwsKYhmTYmVLTRdef5A/flrGiHgzUzIjv+a3VeZo5NKKzDFpc3h4bkMNerWSoAjRZg0ubwCzTk2P20e8RcffvqpmXKqN5zfWAnDu2CTGpFro8/hJjzSyqrSNwkQLTT1uVEppFd7m8JJolTS0397RyMUTkhkIiPS6BhidYiVMpyIn3E/x8jNQ9h62pHOlzmZF/v3c9u5Bbp2dxfqDdoqSrfR5/aw7aGfZ3JzBVkI9qZEGAqJIhFFLeZuTSyamsKOum/WVHWRGm4i36Pnr5xUsLorHZlAxNSuKkvoe3tvVREPX4ST/xIWjuPn1XUNKFxdPSGZtRQc97gGun5HBnz4r54LxSYxPjcDrl6RZ39/VxLiUCNZVdrBjcKpzQpqN6DAtH+05rGJ49pgErpicxsbqTty+AG9tb2RUsoUPdg0djBkRbyY2XMfng0l53ogYfjY7m35fALVSYE9jLx/tbuaM4gQOtDl5YWMtQVG65pgUK09+WcXzPxlHlb0/ZOIQH67jpzMzeGNbA2mRpmETujkxYTx9yWj+9Fk5n+5rxaxT8+sl+SRa9Nz8+i5aej2YdSruWZjHuBQrGdFDJW5r7H385PlttPR6mJsfQ7RZy5KieIqSrJTUd3PmUxuHDA8tGBlLXacrNOH56yUjQvV3mcPIpRWZb4VZp+ZAq5Mvj9IJeeTcIgRB4MVNdSwuisesV/K3i8fQ1O3C5Qvw/IZasqNNWA0aytv6eHWrNLmoUSq4a34u9y8/gFalYMAfJMGiRxAE3twuSdN+ViqVWv5xURG9Sadi631uMJgESjJvRK83cUp2FC5vgO31PWwfdC0CWFXaypKieDKjTYTpVDz8WTk3zMpiYrpkprxy8Nw76rpJsul59LxiXt1az3slnczOj+UvXwzttADo6PPy0DlF/HN9DXan5OZTmBROXryZWLO0Kr52ejpmvaTL/l5JE5/sbWF8agQjE8N54ojujckZETxylLzs2zuaGJVsJSXCgMWgweH2YTNq+OCoOKZnRfHK1sM3tRX725iTH0u4Qc3oZAtmnZpokxan1x8qXwQDIltqukiPMvLrxfk8s66ayyalsmBkLKtL27lyWhqrStuYnBFJVJh2WCKfkhnBvmYH6VEmbpiZyfK9Ldz97l5eumI8H9wwhVaHB6tBQ5LNwLHYUtOFCPx8sCX1q4N2+r0BwvVqqtr7hk2Ari5t54qpqaFEHm/5dnZzP3bkRC5zTPQaJcvm5LCnsTfU77uwMI7YcB32Pg83nZrFgVYHb21r4pTsSHLizKzY18aYZCvFSeH4A2JoNQpSHfWNbQ3MHRHDKTlRvLujiRk5UawaTLB5cWGcmiv1KDc6fFSHX8msM87BrPDweasRuzeClWsqyYszh1rX1EqBK6akodcoUSslk4Hn1tdw94I8FhfF8+KmOmZmR7GqbKiTRUOXm4r2PjYO6pEjSi2MJUfcGEAqEZTUdXPVtDRqO/opTrKwr7mXCKMWu9NLfryZt3fUc+7YZN4racLh8XPTqVl8tq+VP3xSxkUTUkKfVr6urU5A4OXNdSybm8OIeDN6jZJfLc7nyS+r8AwEuHhiMi29nmG16bqufv7yViVPXDiaX3+4P1SOiQvXceOpWTyySpL83VLdxYA/yPrKTtZXdvLmNRO5+dRMrn95Jw3dbr462MElE1OYkGYLlbsyokyMTAznhlelmrlBo+S+M0ZS0d5Hr9tHtFn3jb6eLb1uzh2bxH3LD4Q+0by+rQGdWsmkYxgzx1l0tDul1zA7L5qiRHmq89sgJ3KZr2VkYjgf3jiF6g7JJSgrKoxwg5r8ODN/+bySZzdI9nAlDT1YDWoePa+IQFCkptOFSjFcWLO2s5+75ufy9vYGIkwalhTGs7+pl8mZESgFgafXSr6YF4xPJkyrZvZrPVgMau6eH8/e0jZKWxyUtji4bU4OAD89JZM3tzfQMlizNWqU/GxONk09bsm4QoQEqx4BhrlxHmpjXzAyljaHh1tmZfGbj0qp6ehHq1Jw3YwMGrtcvLi5jg92N/HwOcXc8sYuelw+1EqBq6el89LmOq6elsZNr5WEBnxW7Gvlznk5PLq6gmmZkVS0O4ke1CVPiTAMsaZLsumpaHVy9pgkzv37ptA5ThsRwy2zMqm299Pv9VNlH2pNF6ZV4Q+IJNn0fHmwfUhNvaXXg93pJSpMutnkxIZR2e4MyeAu39/KgoJYzhyTiMsbIEynYu1BO+eMTeT88cm4vH6SbHpe2HhEWWsgwF++qKQgwcx7JY1kxYSR8g2GDuNTI9hS2zmso+b1bfVcOCGZMSkWdgzaySkVAvcuyEOtFDh3TBLZsWHYjHI/+bdBTuQy/5JEq4FE69CPzz1uHy9urh1yrNvlo9XhJTZcLxkcu4ZPdp6aG01dZz/L97eiVAikRho4a3Qi7U4vjw9apmmUCjLUXZwW0cnC0zy06jIo7/eRF29mZKKFNoeHbbVd3DIrk36vP5TEAfoHAmyp6eKi8UmE6VWMiDfzXkkTS4uHGhxLpR81P5uTxZhkK5+XtXGwrY+xqVYWFcYRGFQFPGt0AgoBFo6M51cf7j+8kRuQpjHvnJfD3iYHHl+Q9EgjgiBQZe/j4z0tzMiOorTFwSUTUhAEuPG1Em6fm8PW2i521fdQnGxhaXE86w528Pe1VaEkDvDZ/jZGxIfzzw21LC6K49JJKawsbWddhZ2cmDBOH5XAI6sPkhtrHuZZCtDY5SLWrEMhwBmjEthe18207GiUgsCoJDOPrq5gQ9VhnZjHzivmgRUHhmxiXjUtjYRWfWhjuMrex6LCWD7Y1cy+pt5vTOTFyZZhHT8gdSNFGjU8fuFo9jf14vD4yYw2kR9nRqU8XlXtHy9yIpf51ghIq6gjlQoBbEYNM7Kj+HBXE29sb+Bns7N4bmMtPS4f07IiyY0N4+M9LczMiWZDVQdev8j6yg7cA4fLBo/O0jGv5DoUW5uIBfJMsQTHPsVVn0tJYXyalRSbgUiTllbHcOnTkfHhfLSnhfdKmtCqlFw4IZlIk4Z7F+SxpaaTnMHVXmefF4VCwZaaLty+IOVtziHJDaQbll6tJDZce8yk5PYFMGmV3Lswj72NvQRFkfPHJ7G5soPJGTaSIgxUt7tC78t9yw8wKsnCnPwYDrY5MWiUnDM2gUueHS4FbNAoef3qCfQPBHB4fNx4SgZXTkmlucfDh7ubcHr8lDY7uGpaGluP6AACOK0gliSrHpVCwU9e2BbyEI00aRifNmrY66zrcg3zzdxe08XPT8umyt6P0+Onss0ZEhM78qbzdRi1KiZnRpATYwq1fQqCpOQYFKHa3o9CEBiTYv3Gm4LMNyMncplvTaLVwA0zM4dYryVadOTGSu1lTo+ffU0O2nq9nDkqgYnpNj7e28IjqytQCPDc5ZIErkWvpqzFwbSsKKCT6DAtY/rXonAeXj0Lfa0Ud68i0TqHxm43W2u6uXhCCn/9ooKzxyQxJTOCrv4ByloOlzDe2Sl9v9sX4Nn1NdxxWg4RJg1alZJXt9SH3HpGJoRz/rgk1lfamZIZOaxbJMGip38gwNgUa8g8+UjUSgXpkSZufr0kVEL4ZG8Lj184GgUiP3tjd6i2Pa8gllNyoviy3E5JQw9/vaCY7n4fexqk17/6qDp+fpyZN7c38m5JE7mxYSwqjOefG2ro6h9gckYE98zPpbTZQWaUkbPHJPLBriYEBK6ensbsvGhMWjWXP7d1iBF0R98A+5t7h5lpH13+GJdqZWRCOMve2kMgKBIdpuX3pxfQ0uNGo1SQE/vvmTAnWg08c9k49jb10uMaICdW8i+9/pUdbKuV9k+sBjUvXTmBAlnp8LiQE7nMt0ahELhoQjIZUSZW7m8lL87MrLyYUAfDoT90e5+Xf26o5eXN9dw5P4dRSdJk394mByqFgEoBy+bm0OeVtFmizVosPcMnPG3de4i3LKaxW1oVOz1+smPM+IMi7oEgaRFGzh6TRLhexVNfDjeqaO51E2vWDjMw3tvUy+LCOJYWJ+AeCNCTHUlJfQ+z86MZm2JDp1Jy3Yx01ld2cPW0dB5ZdTDUGXLTzEx8/gCrSluHJEJRhI93NzM2xcrZoxOJCtPy/MZaVuyTRMW+LLdz/rhE3AMBfvexJKn789Ny6OofYGd9N0aNkrsX5KEUBN4dLActLornwSOMJjZWdWLSqgjXq7nxtV3kxoZx9bR00iKMLCqMpb7bTUO3i7kjYtjb1DskaTs9fix69ZBjh4ZyDmnAzMiO5qGVh6/X7vTy51UH+b9F+bx81fhjeoh+HUk2w5DOlvd2NoaSOEglub9/VcXD5xahUSmPdQqZfwM5kcv8R9iMWhaMjGPByLhhj41MCOe3S0bw2OeSV+f4NBuiCD3uAfJiw+h2+VhX0cG6io6QVstD5xSyu7GXft0itFUrhpyvPmkR+9dJZRSzTkVqpJE2h2dIcvvyoJ1/XjaW7BhTyO3oECkRxiH119QIA0FRUmP0B0W0QGGCBafHz/TsKF7ZXE9ps5OlxfFMTIvgmfXV7G9u4KKJyaiUCpSCQHOvi8snpfHkV1XDXr97IMBHe1rY1dCDRqngjnk5/P6TMoxaJS9eMZ6gKFJS3xOazHxoZTmzcqNZNiebsalW3AMBttYeLpe4fcO1S1aVtXHzqVlcMjGFcIMaUYQDrb30+/z89fNK+rx+Ti9O4M/nFnHrG7tC5ZCpWZHMyI7i/uUHKG91Mq8glmmZkYxKslDS0IPHF8CoUXHr7CxeGjQCATjQ6kSjUlCYGI4g/Ofj85J8wlB2NfbQPxCQE/lxICdyme8crVqJSadiUVE8YToVO+u6+f0nZZJZhFpJvEVPik1PXZeboAibqrs4c3QiD35WzsCENK4efyumHU8CIq7R17BXPx6DppvCRAuXTkqhw+kJObYfQgDaHF4umZTCzvpu2hxSJ0dGlJGMKCMVbX2cXhzP3BGxdPR5Jaeh2DD0KgVd/QP0eyUPzd99XBY654EV5fzl/GLmFcQRbzHQ3ONhdVkbroEA9yzI4/pXd3DnvNyQLskhJmdGcv9y6TwDgSDbarsZmWAm3qLnxld3Mjs/hrTIw3VhUYTVZe2U1PcQZ9HR2usFgZDeuUE9PMGl2AwUJYZzz3v7QhK/D59bxI2vHrZUe2N7A2E6FcvmZBMURTKjw0i1GYgJ1/P8T8bh9PqJNGpRqxQ0dbt4Z0cjuwd9MnVqBXfOy+V3H5cSFCEz2sTLm+sQBLh5VhZpkf+Z1OzoZMuwY4sK47HI4/jHhZzIZb4XxqRY+eeGmpDSns2oYVZeNDajBq8vwEUTUogwaWjq8aBUSN0eiVY92zpU1BkXM3H8qQjAC2UifrGPa6ankWDVIwK1nS7UysOrwknpEUzKiODxNZUIgiTi5PMH6ewfwOMPUN0unXtEvJn7lh8IDZ2YdSpunZ3N+7uauH1uDpuquoa9jjaHB19QZO1BO0at5Kc54Asw4A9gdw4QH67jz+cW8UFJM0FRZP7IWF7eXM+RZWeHx8dvloygvsvNFVPTEETJru5IOzeQnI4e/KycJUUJfHGgjTtOy+HZ9TVEmjRMSo9gU7W0SalSSHZxAIHBO4hZp2bnUd6gILV8JtkM3Le8DI8vSFqkgb9eMJqChHBMusPJs6S+J5TEQdrQ/HhPC9Ozotjd2MNFE5K579MDDASCBILw4DmFIRPqb8OYFCu3zs7iyTVVDASCzM6L5rxxSce1ypeRE7nM90RKhJFnLxtHeasTXyBIVrQpJOnqCwSJC3eysrSFVJuBn721F7VSYNncHCranJh1au5aM3Tzzx8Useo1XPrcVsw6NeePT+aJNZWoFAJTsyKHlFnufGcvf71gFP5AEI8zQKRZi9Pjo7HbHUriS4riyYg24fUHOGdMIo+vqeC0/Fg21xzu6DDrVfgC4hAj5AdWlPO7JSP45Yf7ueGUDB767CAz86I5b2wik2097GyyDzMSPnt0Ir/9qJRdg4ky0arnD0lW/nL+KL4st9PY7WJ6VhRryu20ObxkRRvRqeN4Y1sDd83LJSCKRIVpuWNeDgkWPb5AELtzgJqOfi6dlMLDKw/S5/UTHTZ8SOeUnOghjko1HS5+8f4+7j9zJBnRJlyD5Z0Wx/CunLrOfh46p5ANlWE8supgSIjr4z3NLJub/R91m1gMGhYXxpFsMyCKIgWJ4aTKXSvHjZzIZb43Ysw6Yo4xAahWKhiREE55m4OgCE+em8dn+9tp7+zhojHx9PmV7GvuDW2KzciOIt6iZ1+zA19ApLN/gD2NPfxiYR7d/V621QxfSX+4q5HFRQm0ODy4BwJUtDnRq6Vf9yVF8XT2e/lwldSlohDgrvl5xFt0ocEZgFOyo1i5v23IMYDSVid/PL2AlEgjm6o7eXFTHXedEot736dUq2bz89Ny+HhPM/6AyFVTknF6fKEkDtDY7ebjvS1o1UoWF8bxxJpKfvdJKb6AyIKRsejUSj4/0MZ1p2TQ7R6gs2+AtRV2RiVb+P3HZdj7vKgUAjfPyiJ8UMDL7QugUSlIsuqJCtMyMT0Ch1sqFx1CECAtwkhTj5stNV3srO/h2fXViCIsm5s97D2ckR1FXaeLf6wbuoEcadIO8Sb9NpS29HLRP7aEOocsBjWvXDWBEfFy18rxICdymRPGzJwYVpe10uqCcRmRJBpFVBotSjHAiDgzkzMiEQTocQ3Q4fTiOqLffF1FBxurOvnHJWNo7B7uNhOn8yM620iw2nhsdSW5sWFcOjkFnVpBZrQplMRBkrN9bkMNv1iYx7I5Odj7vKTY9MSYdexq6GFqdiSiCP9YW81AIEiiVc9TX1XR6/Fx48xMEEF0NrBVPZ4uvx41Qc4ek4RWJWBS+VlbdTiJXzA+ibhwPUFRJMKowd7nITs2jOJkC0qFgtQIA4GgSHqkiU1VHSwqjKe1x8WfzirkgRUHQlOc/qDIn1cd5C/nFwNSkrY7PTx6fjHPbajlb19VMSc/hkiTlnsW5NLY7SY+XD8oMaAhJcLATa+WhDZcn99Yyy8X5fHY6gqcXj+LC+M5d2wSn+1vDW2EHrrOb5eOIOoYq/9/h492tYSSOECPy8f7JU1yIj9O5EQuc8KwGjWcMzaZdodUJ2/pdbPorxvIjjHxf4vy2T5Y89WrVdR3u1ErBDKiDnelBIIidZ39nDc6mhX7W0M2dFqVgjMTulEM9LHSXUyrw0Orw8P+ZgePnFtIWWvfsFhaej00dru5f8UBIk0abpyZybUv7ww9Hheu46ppaby+rQG9WknDYCvkfcsP8NfzR+HoNqMyRfDqp+WhRBUfruP203IYnaxAKSg4JTuKV7fW89qgkBjALxbmsXxvK6IoYu/z8uDZhdz21u7Q5umHu1v428WjCQZFqo7R8eHxBbltThYiAg1d/fz9q2pWlrZx3rgkHG4fV7ywDVGE+YPmyh8Pqi++V9LEbXOz+c2goce22m6SbQaW3zKdAX+AeIserVpJSqQBu8NLU4+bPq+fzCgT+d+i/fBoytucw4+1Dj8m8+2QE7nMCeeQAJMoQtagtvfDKw8yIt7My1vqUQjwi4X5gMj1M9Il2Vv3AGmRRnwBEYvfzjtzvWx1SmJM4012CjbdwcFxvx1qkdbnxdhZxsSkHP56hEkDwJSMCNZWSEqPkzMieWmzpMgoCHB6Xhhz471kxbuYnlnImoOdoaEaUZR6sysdGpz2niGrzeZeD3Wd/eTFmnlnRyMZUcbQzekQf/migt8vLaCsxcGkDBsvbWoYElcgKPLxnhYuGJc0TKsFwKRVsrWmm5ZeD6eNiOG9kmoMGiURRg1vbDt8w1i+v5Wrp6URZdJg7xvA4fHj8weJMKnp7JNi1qmVJFiHWq3FmvXEmvVEh+lweCRN8uNpEzx9VMIwt6OzxiT+x+eTkZATucxJQ2SYlicuHM3zG2pZW2EnNlzPK1dNAOD/PtgXWpEqBLh7fh6bqzp5c0cjv12UzZKuNRSUvSOdyOvEmz4XV+JUrANukm0G6rtcZEWbUNliCXZU8cg5hfxheTl2p5epWZHMzo3mN4Na3RFGDT0uqX/6sdkmTqu5D+36jaDUYBlzK62W0/nT2YVsqe7ize0NaFQCi4oT+O0x7Oqq7f3UdbopbXEyv2B4z73T46e5x4NKqeDR1VUYtcOTZCAo4gsEuW1ONr/8YB8Otx9BgKumptE/EKDP62dEvJmCBDNx4TqsBg2lzcPlC0rqe1g2N4f6Lqnu3eP2sWhkPC9sqkOrUnDO2ER63QPsbuhlb1MvKTYDo1OslLY4uOfdvbQ7vYxJtvKHMwrIjfvPVuVTMyO4e34uf/2iElEUuWlWFlNlA4nj5jsxlhAEYRnwEBAliuJwm+6jkI0lZP4V/kAQp9dPmFaFSqng87I2rnxh6O+LSaviF4vyuOudvQDcN8vKWZrNaPa/Sd2Ye3msNp73d7WG9FacHh8pNgMPrzpIUJRGwx88qxCdWoFGpWBXYy8mrYoYs44+j499TQ42VbXxcvw7WPa/NOTarYtf5fQVOs4dl0hxkgWFUoBAkHJ7P/d9emDIcx84ayS/eH8fs/JiWDQylvLWPhCkFsLSFicen58rJqdyzcs78QdF7jgtJ2SmDNInggfPLuT5jbU0dLk5f1wSapWCWLMWpULg7ncPd6TEW3TcNieb+5cfYHFRPM9tqB0SywXjk/hqUF/+zNGJGLVKsmPCaOx2MybFSn6cmafXVXP/Edf/3dIR/Pqj0iHTq0VJ4bx4xQTCj6P3u6XHjYhUspJbD/99vjdjCUEQkoC5QP3xnktGBiQ3+yP1u48l0uT2BYgwakIdJXd/3k3/wqW0x09D1W3i3ZLq0POeXV/Dc5eP45qXtof6u7tdPm54rYQnLxzNpc9tCR3PiDIyMzcah8fPsimRWNasGHZtU08ZF01cSmqkkXdLGpmeHYXPLzIqSRpYem1rPUqFJHXrGgiQbNMzKc2GUqFAREStVOL0+PAFApwxKpF+XzBU3/9oTzN3z89lfWUHGqWCSRkR9Lp9oX78v6893EHys9lZQ+Jq7vEQZdJyxZQ0kmx6NlQetqBLiTAQaz5s0xZt1lLX0cdV09JDYli1Hf0hHfNDtDm9w7RYdjf00ubw0O7wsLO+mwF/kKIkCwXx4f+2aXKcRf/NT5L5t/kuSiuPAHfAMGMTGZnvhOwYE0aNkv4j9EEWF8ahUyu5Z0EeBweV+XJizESatDz2+XC3n4Zu1zC1Rq9fUj08Mk9V2ftZUqTimXU1+DxWpsQUoalePeT7mhSxPLLyIAsK4oi3Grjznb2oFQpumpnJ0qJ45uTFoFEJCIJAR5+XB84spKbTxU9f3Rmqf6dHSjcMENGpFSH98H1NDkqbHUxIs3Hd9HS21HaRGWUa1gIJSOOsg1gNai6ckEJjt5s/fVaOQoDzxiVx2eRUuvoHaHd4h7gg9bp85MeHU93eh8WoIcasw+s/fENZWhxHeqSJGLOOq6els7q0ldn5sWhUCnY3dOPyBrj8+a2hPQi1UuDVqyYyLs327/5YZb5DjksAWBCEpUCTKIq7/43nXiMIwnZBELbb7fZverqMTIismDBevXoi80bEkh5p5OZZmVw5NY3sGBMLRsZx7tgkihLD2dvUi9WgJit6+Pi4Ra/GeFTvc5hWhesYOiZefxCVQuD9/d3szvgp6K2hx/zpsyFxLFMyI8iNM/P02mq0KgV/PreQxh435z69mUv+uZXH11RRUt/NT18pYfm+Vp5cUzlkE7O6ox+zTkWbw0tdRz+/XTKC5EFxqegwHUuKE1hZ1oZKoaC208XjF44mNeKw+NT0rEiseulTy/hUK79ZOgLF4Fj/WaMTCIrw2tYGnvqyih6Xj5c214VW1gpB8gF9e0cTpz22jkV/Xc9n+1qJs2i5d2EeT100GrtjgEdWV3DXu3t5v6SJn8/L4bWt9Tz1ZSXRZh3Nve4hG8m+gMjf11Yx4B/+fsp8/3zjilwQhNVA7DEeuhe4B6ms8o2Iovg08DRINfJvEaOMDEVJFv5yQTFuX4Bw/VD3mBizjpQIA83d/VTZ+/nJhHg2V3eG1Pwyo01Ud/Rx65xsnlxTSbdL8sb81eJ8ylqGbgoKglR/9w8mvevWBHnpjA+IHqjH7lWypjOCv75Yy6+W5NM/eP6zRydS3+3mje0NCIIkGWDWqXF4JLcdEbD3DQx7Tb6gSJxRw1vbGrhrfg7zC2IpTAwnEBTZWtNFaYuTnfVSl4tisFb+0Z5msqLNBIJBNlR18vTFo2nodnPza7tC552cEcGiwjg+3tNCU4+bSek2/EGRz/a1EmHScMaoBMrbnFS0S21/dqeX617ZwVs/KWTVvmYmZkaxsfrwhKu9z8vbO5rIijZR0tDDuzubsBo0JFj0Q3TaW3o9+IMisrfPf59vTOSiKM4+1nFBEEYCacDuwc2KRGCnIAjjRVFs/U6jlJEBNCrl17a+2YxabEYtUUoX5p1P8Mx5V7K+0YdSELD3DfDY55WMTAjnLxcU0+8NEAiK7KrvwaRTcd7YRD7Z20qkScNVU9PQqKXadIRBw8LCOFa2OPH6s8iKNlHr7mTZ3Gz+9mUlt8/NBSApwsDBVkkP/boZGawqa+NAq4Mkm55fLMhjV0MP8wpieXtHYyhehSA5FYXpVMwfGYvXL6JWKfD5g6AQyIoJIypMy5gUK89tqBkcAKrg3LGJIbOOmdmRaNVK/nxUXXtjVScPnVNEhFHDiHgzq8ra+arczvTsKHpcAzyw4gApEUbGptpYsU/6U02NMNLhClDb5SbD6eVoyludjE+zhQaDPtvfyrhUG027DmvHXzYpFYNGboQ7EfzH77ooinuB6ENfC4JQC4z9d7pWZGS+L2K0A7Djb+S3l7Eu/g88+qU0gm7Wqbh6Wjr+gMgza6sYnRqBSafikdUV/HpxPma9GrtzgD98eoCgKHLu2ERmZEdxy+u7QrV5lULg7gW5vL+ziWVzcwiIQa6elk6/x09cuJ5LJqVw//IDIU2Sqq+q+cmUVPY39XLZpFTC9Sre2tFITJiOm2dlsaa8jXd3ShOmxUnhXDUtHVGEu97ZE7pmolXPdadk8PgXlbT0uhkIiPS4fGiUCuaNjGN7XfeQvYNDNPW4aO11E2fWkRsbxhvbGnhz++G+8kSLnnaHtDl6zYx09jc5+GBfJ1dMTSdMpwIahpxvfJqNPY09oa9zYsM4e3QCe5t7cHsDXDsjg1l50cicGOTbp8z/FoIKVFpMdZ9zg0LB5MseocUFVqOW3v4BttT0cv3MLNaW24kdHETq7B/gH+tqhpzmpc31FCVahiRJf1Bkc3UnU7MjeWjlQWZkRTEh3caAP4hWraCmwxVK4od4a3sj545N5IbXSnjhJ2PJiQljX7ODzdWdbKrq4idTUgmKsLGyA6NayevbG4Zcs7HbzYA/yJVT00iyGTDrVDx2fjEqQeDlrfUYNErGpliHDBrFmLVkRYexo7ab17c3sLAwjhtnZoZ8UQ0aJeePT+KGV0u4fW42D31WHtrkXL6vld8syeeiCcm8vq2BQFBkQpqNgkH/U5BMrm+amUlxspV3r5uMPygSYdJ+hz9EmW/Ld5bIRVFM/a7OJSPzH2NJIjjjLhSrfomuZhWTagrwZ55GzeQHEA16ChPCWV9hB0EkyapHp1Yc0/TXpFWFJGKPpMM5gMfnoK7TxYuddexr7uWu+bmUtzqJOkYyM2lVITeeNeUdVLQ5USoEsmPCWDAyjrd2NKAQBK6amsb+Fge1HcPNlJ1uH3WdLp5dL91sUmx67j+zkDCtilVlbSybk0O4Qc36ig4yo03cNiebO97eTbRZR7fLxxNrqrhgXBJ/PKOANocXhQBPrKnkznmSO9OhJH6IN7c3MjXTxh9OLyBcryYz2oRJqyItyojHJylZZsVILlDhBrkifjIgr8hl/rcQBAJFF9GsTMTYuJa+sDS2CkXc8UwZD5w1kmBQ5L2SZn6xKI83tzdw57xc7E7vEL1vgMunpDJwRILTqhRcNS2d6DAtXf1exqXaeGFjLTvre6ju6MfrCzAqyUJBnJl9R2ygXjghOZSATToVIxPD+Xh3CzNyokImFlqVgv3NDk4flcApOf5heiTFyRZeP6IsUtflZmVZG4uL4vjiQDvv7GzkwgnJLCmKJyFcS323hwsmpFDW4mByRiRalYKn11VjNWp48svDjkYN3W4umZhyzLfxgjEJKFRqIsO0obq33Pt98iIncpn/OdSmCLoTT+XarVHUd7pwevspTAxnQpqNHvcAN56aSVmLA4VCwe8+LmVaVhTzCmKYkROFAESFaXlpUx0BUeSuebl8uLuZs0Yn8NRXVXQMdp/o1Arump/LezubyIoysb6yg998XMopuVHcMCuT7TXdRJu1fLa/jV63jxizlvzYMHY29LCoMJbVpZLeSGqEgYsnpvD2jkbufW8vy+Zkc/GEZN7c3ohOreDaGRlUtfdx9IeD0mYHY1OsPHPpWNZXdvDQZ+VSDX9+Lnubenlj++GN1axoExeNTyYrZmhbpscXoDjJMqxH/YxRCXxZ1cOs3GgMGhWuAT+1HS68/gCpkcYhw1oyJwffyYj+t0Ue0Zf5b1DX2c/Btj50Ksn5/ZA4V1W7k6YeNwN+kete3hFqNdSpFdw6O5vGbjed/V6W721Fp1Zw+5wcnF4/j31eMeT8c/NjmJETyYe7WthyhCb6iHgzt8/NprRF8rl0DZYvdColf159kGunp9Pn9fPKlnruOC2Hh1aWDxlKuvnUTLRqBWmRJh5ZVc6c/NghK2mAZXOyeX1bPaePSuSJNYcHfW6elcnjX1Ry1DAm9585EoUAW2u7+GxfG+lRRs4YnUAwKBIUYXdjD70uH5MyIviy3M6Wmi6mZ0Xy+9ML+NtX1by6VRrcLkoM58/nFpNxjF59me+frxvRP66BIBmZk5mUCCNz8mOYlh0VSuIAGdFhTM+ORqMSePzCUVwxJZVbZmVx3xkjeXptNS9vriPJauB3S0fw5EWjGZMSTkvvcM3zph43Zp16SBIH2N/soLnHg06twGbQYHd62dPYS7JNj1opsLnSzty8SCx6NX1e/7Ck+8KmOnJiw6jp6MPtCxJj1nLW6AQUgtTnvqQonjaHh5QII5urO0mw6Dk1N5pkmwEBYdj5QGp33FzdRVqEkQsnJnPL7Cw6+wawGTWs3NfC0qJ4+rx+Hl55MPR6NlR1sqOuO5TEAXY39vLyEcNFMicHcmlF5kfL5IxIKtudWPRqul0+PtjdzOWTU2npcRNh1BAXrqOl243Xpx1ilnyIuSNivlbwSakQeHFTLXWdbhKtem6dlcUb2+v501mFZIaLbDpQw+VTUo9Zpogwatha3UVubBh3z8/FalATCMLionhEUcTrC3LdKzuZkR3FGaMS2Fnfza6GHiak2YgP13H2mARaej3Udrho6pGuX9PRT3Ovm4Uj4/AEAtz73j5aej2MTbZwy5xs6rtczMiOItFq4MPdUktkpEnD3qbhKoqfH2jn1tlZ8kbnSYScyGV+tKiUCnLjJGeadRV2VuxrZVVpG7FmHWsP2vm/xSOo6uznqbXVJFkN3Dwrk9e3NuDxBbh0UiqZUSZWlrYyKSOCTVXSRml0mJbTRsRgNaqp65SmHhu73by0uY4pmRGs2N/K4pGxREZEcNv7FVw7PZ0km56GLum5wqBGyqOrK7hnQS7+oMgLG+vYWtuFSavi8impxJp1FCaEs3BkLK9trWdnfQ8AZS1OEq160iJMVNv7mZkbxZhkK409bh5bXYE/KFLa7OCyyam09HpQCLCwKJ7rX95J36BT0Pg0G+eNS+KNbQ1kRJmIDR/eiTMpIwKjVk4dJxPyT0NGBmmMP8Wmp67LHRo7L2t1kGQx0NjtprHbzYFWB/ML4tCqFZj1KkpbnHy0u4VLJ6VSmBBOklVPfbebPY09GDQqLpucygsbawGpJHHGqAS+OGCnpsvNmGQbkzMieHZ9DVdOTcOgVaISJKGtt3c04vYFSI/U89KWRrbWSqWOPq+fx7+o5LHzirh1ThY9Ll8oiYPksbmjrpu1FdJM3s76HtZXdDA1MzK0D+DwSFrmANOzo/hod0soiQNsrelibn4Md5yWg8sXoNrez7SsSNYNnjPZpueKKanHbNmUOXHIiVxGBogL1/PMZeN4t6SJzVWdLCyMY0Z2FC5vgPy4MEpbnHS7fKF68e1zc9hU1cmsvBie31jL5ZNSeHFTHQfbJdnYzdVdnJIdxfSsSNZWdJATE8b2um6q7f24BwJU2fs5c1QCZ41OpMc9gFGj4k8rDtA1KEQ1Md2KVq3m87K2YbG2Ob387M3d3DxrqIxtcZJl2IZsbaeLpcVDV9XKwUyeZDXwQX0TR+P0+DlnTAJn/W0TbQ4vc/Jj+NnsLIIijE21kBP7n1u9yXw/yIlcRmaQrJgw7pyXiz8QHLLivO/MQm5+rYS6Lhc6tdRPvnxfC/ubHfz0lAxMWhURYdpQEj/Elwft3Do7i92NvVw8MZm9Tb08cl4RX5S187evqgiK0tj/PQvy2FbbxSWTUmnpcZEbbSA91sIb2+rJiDJx4ChPS48vSFCE7bXdzMqL5vOyw9ZpwlEWdoeOHWJCmpUwvfRnv7O+mykZkSzfN1QaaUS8mUSbkWVzc7jj7T2sKm1jVWkbVoOaJUWTj+ctlvmekNsPZWT+DTqcXpp73ASCIre/vYsq++EJzL8vjUOnUnDZO8NXt09fMgalAhq63Dy3sZa6ThejkiwsLIzjgRUH8AVExqZYMWiU1HT089wlxfQOBClt6WdnXRfFyRb++OmB0PTljOwoIkwa3t0pXeu8cUlEhWlp7HIxJtnKttouPhw0WAYoiDdz5/xcSuq7MWpVdPcPDPpuKvAHxNBG6Ae7milrdXDr7GzOHZuIzajF6faxtbaLlftbSbQZmJMX8x9bvMl8N3xd+6GcyGVkviWf7WtlZWkbnf1eFoyMozDcgwE313/aw/6Ww6vnWbnRXDE1DafHx61v7BridJQZbWJkQjjvlTSREWVk2Zxs9BoVW2o6aenxkBsXRkVbH6OSLXS7fPgCQeItekxaJTe9totzxiaSaNETEEV0KiVef4Cn19awpDieSJOWijYnObFhTEyzcftbe+j1+Jg/MpbxKTYsBjVlrU72NvXyeVk7CgHOHpPIosJ4HvqsjEkZUVw7Ix2bUdZPOdn43qzeZGR+TDjcPv702QFUCoHb5+awsrSVO3c2IQjw28UjaHF42NvkYEKalegwHZuqOlArFcPs6irb+5hXIMn83zM/lyp7P4+vqQxpqH+wG26bk81rW+spSLDw5vYGfn5aDj6/ksfOL+Yf66p5a3B602pQc+PMTNy+AG9sa0CvVvLHM0fy3s4mWns9FCSGE2nS0O8NcM/7+4g0ajl3XGKoJBMUJX2V3FgzB9v72dPkZGJ6xKCDkcwPAXnrWUbmW6BVKUi0Gihv6+Pql3aQHmXi/HHJmLQqXtvWwNSMCDw+P3/7spqfv72Hzj4vseHDNUoMGiValcAf5sSx/mAbvR5/KIkf4p2djeTEmok1a7lmWhpWg5oXNtVR29Ef8vAEyX90e103p+VLidftC9DU7WZthZ23djQyPtVGuF7Nh7ubCQRFsmJMrK8Yrja9saoj1C9/dF1e5uRGXpHLyHwLtGolN56ayaaqTgYCQR5YUc6ENCtvXDOJeIsOu8PL3ZNN1PUa0OhNWIw6en0KzhyVwLuDMrCZ0SZum51JY1sncZEG3jzQwZSM4WUMnz+IXqNkQrqNhz4r5+l1NUSHaWk8wpXnEKUtDq6elsbYVBsKhRCqoYOkxf7p3sN187qufiamR4RMIg6REmFkW60kh5seNXwASubkRU7kMjLfkrEpVt796WQOtDrQqpQUJJhJi5S0R7bXdtPb2sXpu69D0ddCwJbFzlNeQK0U+P3pBXj9AaLDdNzwWgmiKGmHn16cgEGjQqtSDJGUPX1UAvEWHWJQ5ILxyUzJcqNVKog0aUJllUNMzojg72urWVAQR0qEgcWF8exvllbtMWYteXHmUOJu6HJz3lgjiVY9jd3STSEz2oRWpaDX7WNxYRzFSZbv/42U+c6QE7mMzLdEEAQKEsIpSAgf9lh0mJaffuWGGX9jtFCBxu8gzGfnrNEFbKjqItmm54kjjJjbHJKtWkW7gzvm5bCluouu/gHOGpNAdJiO8lYH9v4BfvH+/pBpxWWTU/jtknx+/4nkRjQjO5IpmZE0drtZsb+VokQLE9JthB1yRRJFpmdHsaq0DXufdL13djbwiwV5aFQKNGolcWYdrQ4Ps/NiyIg2Ea5X/3feTJnvBLlrRUbmO8Q94OeFTXU8sELaEDVqVPxm6Qjm5MfweWkb8RY917+yk/ajfDH/eEYBNqMGjUpBhEHDiv2tvLatgfPHJfHOjqZQAgbIiwvj7vl5dAwee6+kiXUVHSybm82mqk4EQeo1v31ONk6PH6NOiQBsqu5CIQgIgrRpOzIhnDNGJ/433x6Z40TuWpGR+S+g16i4fHIKUzIi6OgbIMGqJyPKhFIh4PIFaHV4uGB8Eo99flh6ViFAr9uH3enlta0N3L0gl6e+qgYkPZgjk/jMnGjiwnX85PltBIIi+XFmFhXGsa6igw2VHdw4M4NL/7mNaVmRNHS7aeh2EWHSEGHUMCvdQK+jlw0tCiZnRjIpPeK//v7IfD/IiVxG5jtGp1YxMtEy7PjkjEhW7GtlQloEP5niZ8W+VmxGDWeMSuDVLfUsLY7nj2cU8PTa6tD3tPS4SY80Ut3RD8DYVCsPflYeery0xUGiVU9RYjhRJi2tvR5E4MLxyTy0spyK9n6SrFpeneMn7osHUTkbmVx8GQ7zeagVkd/3WyHzX+K42w8FQbhJEIQDgiDsFwThT99FUDIy/4sk2QycPSaBoBhEAUzNjCTRqudPK8ppd0r2cbFmHTbTYXnYj/e0cMmklJC/aI9rYNh5t9Z2UZRkYXSKFW9AcjX604pyzhgllU1+NyFI0scXoWreDs5W1OseQNj2NM9tqOZg63CZWpkfHse1IhcEYSawFCgSRdErCII8QSAj8y+wGrUUJ9vQqpV8Vd5BSUMPM3OjWFQYT0uvm8ue28Izl45jzQE7bl8Arz/I02ureficQjbXdBETpht2zpEJ4SwYGcs/vqoh3qrjpc2SsJfD40OrUpAaqIfg0B512/4XURUvZmOVimxZBOsHz/GWVq4H7hdF0QsgimL7NzxfRuZHj0mrYmyKjTCtmtHJFrz+AGa9hp+/vRtfAO55by+PnFeE3elFo1ISHabB45PaFstaHJySE8WX5XZAMqG4bkY6O+t6iLUcTuIAGpUCXyDIgGJ48scQgd0jsnt/G2ePScCkk00ifsgcbyLPBqYJgvAHwAPcLorituMPS0bmfxtBEMiNM4dEqIJBkcfOH8VfP69gX7ODtQc7uGhiMqkRBhQKgboOF/EWA5lRRoJBkfPHJeH0+EmNMLCroZsYs5YHVx6unSdadFj1GoIilPhTybRloug6vMFaM+YeXvncxaKiONRK5X/99ct8t3xjIhcEYTUQe4yH7h38fhswERgHvCkIQrp4jJ5GQRCuAa4BSE5OPp6YZWT+51AoBMal2vj7JWNxenzYjJohUrqHEn59Zz8dfQNYjBqiTBr0Gskz9Jl11fxiQR52p5cUm5ZxMdDl1/LH0wsIN2roH/EaqubtODpbaNDn8vtdetSqAAtHxqFVy4n8h85x9ZELgrACeEAUxTWDX1cBE0VRtP+r75P7yGVkvju8vgDVHf30un0kWfUkWA3HfF4wKFJS383+FgeiCLmxYYxOtqBWyYn8h8L31Uf+PjATWCMIQjagAYar8cjIyHxvaNVK8v4NnXCFQmBMqo0xqbb/QlQy/02ON5H/E/inIAj7gAHgsmOVVWRkZGRkvj+OK5GLojgAXPwdxSIjIyMj8x8g65HLyMjI/MCRE7mMjIzMDxw5kcvIyMj8wJETuYyMjMwPnBOiRy4Igh2o+69fGCI5OdsjT9a4QI7tP+FkjQtO3thO1rjg5IotRRTFqKMPnpBEfqIQBGH7sZrpTzQna1wgx/afcLLGBSdvbCdrXHByx3YIubQiIyMj8wNHTuQyMjIyP3B+bIn86RMdwNdwssYFcmz/CSdrXHDyxnayxgUnd2zAj6xGLiMjI/O/yI9tRS4jIyPzP4ecyGVkZGR+4PzoErkgCA8OmkXvEQThPUEQLCc4nnmCIJQLglApCMJdJzKWIxEEIUkQhDWCIJQOGmvfcqJjOhJBEJSCIJQIgvDxiY7lSARBsAiC8Pbg71iZIAiTTnRMAIIg/Gzw57hPEITXBEE4hv/bfy2WfwqC0D6omnromE0QhFWCIFQM/t96EsV2UuWMY/GjS+TAKqBAFMVC4CBw94kKRBAEJfAEMB/IBy4QBCH/RMVzFH5gmSiK+UgOUDecRLEB3AKUneggjsFjwApRFHOBIk6CGAVBSABuBsaKolgAKIHzT2BIzwPzjjp2F/C5KIpZwOeDX58Inmd4bCdNzvg6fnSJXBTFlaIoHrIU3wwknsBwxgOVoihWD0oCvw4sPYHxhBBFsUUUxZ2D/3YiJaSEExuVhCAIicBC4JkTHcuRCIIQDkwHngVJ5lkUxZ4TGtRhVIBeEAQVYACaT1QgoiiuBbqOOrwUeGHw3y8Ap/83YzrEsWI7yXLGMfnRJfKjuAJYfgKvnwA0HPF1IydJsjwSQRBSgVHAlhMcyiEeBe4Agic4jqNJA+zAc4Nln2cEQTCe6KBEUWwCHgLqgRagVxTFlSc2qmHEiKLYMvjvViDmRAbzLzjROeOY/E8mckEQVg/WAo/+b+kRz7kXqXzwyomL9ORHEAQT8A5wqyiKjpMgnkVAuyiKO050LMdABYwGnhJFcRTQz4krEYQYrDcvRbrRxANGQRBOWkOYQZexk64v+mTOGcdr9XZSIori7H/1uCAIlwOLgFkn2JquCUg64uvEwWMnBYIgqJGS+CuiKL57ouMZZAqwRBCEBYAOMAuC8LIoiidDYmoEGkVRPPTJ5W1OgkQOzAZqDpmiC4LwLjAZePmERjWUNkEQ4kRRbBEEIQ5oP9EBHclJlDOOyf/kivxfIQjCPKSP5UtEUXSd4HC2AVmCIKQJgqBB2oD68ATHBIAgCAJSrbdMFMU/n+h4DiGK4t2iKCaKopiK9H59cZIkcURRbAUaBEHIGTw0Cyg9gSEdoh6YKAiCYfDnOouTYBP2KD4ELhv892XABycwliGcZDnjmPzoJjsFQagEtEDn4KHNoihedwLjWYBU81UC/xRF8Q8nKpYjEQRhKrAO2MvhWvQ9oih+euKiGoogCKcAt4uiuOgEhxJCEIRipE1YDVAN/EQUxe4TGhQgCMJvgPOQSgMlwFWiKHpPUCyvAacgycO2Ab8C3gfeBJKRJK7PFUXx6A3RExXb3ZxEOeNY/OgSuYyMjMz/Gj+60oqMjIzM/xpyIpeRkZH5gSMnchkZGZkfOHIil5GRkfmBIydyGRkZmR84ciKXkZGR+YEjJ3IZGRmZHzj/Dx8C0JSHDVO4AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.scatterplot(x=pca_components[:,1], y=pca_components[:,2], hue=y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01b9b83e",
   "metadata": {
    "papermill": {
     "duration": 0.063551,
     "end_time": "2021-08-13T07:17:08.224795",
     "exception": false,
     "start_time": "2021-08-13T07:17:08.161244",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Let's also check which of the components have high feature importances towards the end goal of churn prediction."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "586937b3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:08.359345Z",
     "iopub.status.busy": "2021-08-13T07:17:08.358392Z",
     "iopub.status.idle": "2021-08-13T07:17:20.982182Z",
     "shell.execute_reply": "2021-08-13T07:17:20.982618Z"
    },
    "papermill": {
     "duration": 12.694325,
     "end_time": "2021-08-13T07:17:20.982792",
     "exception": false,
     "start_time": "2021-08-13T07:17:08.288467",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<BarContainer object of 16 artists>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "rf = RandomForestClassifier(n_estimators=100, n_jobs=-1)\n",
    "rf.fit(pca_components, y_train)\n",
    "\n",
    "feature_importances = pd.DataFrame({'col':['component_'+str(i) for i in range(16)], \n",
    "                                    'importance':rf.feature_importances_})\n",
    "\n",
    "plt.figure(figsize=(15,8))\n",
    "plt.xticks(rotation=45)\n",
    "plt.bar(feature_importances['col'], feature_importances['importance'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a5aa1235",
   "metadata": {
    "papermill": {
     "duration": 0.065189,
     "end_time": "2021-08-13T07:17:21.113066",
     "exception": false,
     "start_time": "2021-08-13T07:17:21.047877",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# 6. Model building\n",
    "\n",
    "Let's build a quick model with logistic regression and the first 2 PCA components."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "a7f08df2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:21.245608Z",
     "iopub.status.busy": "2021-08-13T07:17:21.245033Z",
     "iopub.status.idle": "2021-08-13T07:17:21.363943Z",
     "shell.execute_reply": "2021-08-13T07:17:21.363462Z"
    },
    "papermill": {
     "duration": 0.185674,
     "end_time": "2021-08-13T07:17:21.364082",
     "exception": false,
     "start_time": "2021-08-13T07:17:21.178408",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(max_iter=1000, solver='sag', tol=0.001)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr = LogisticRegression(max_iter=1000, tol=0.001, solver='sag')\n",
    "lr.fit(pca_components[:,:2], y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "775e57ca",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:21.506723Z",
     "iopub.status.busy": "2021-08-13T07:17:21.499855Z",
     "iopub.status.idle": "2021-08-13T07:17:21.518635Z",
     "shell.execute_reply": "2021-08-13T07:17:21.519564Z"
    },
    "papermill": {
     "duration": 0.089994,
     "end_time": "2021-08-13T07:17:21.519838",
     "exception": false,
     "start_time": "2021-08-13T07:17:21.429844",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8981053233093448"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr.score(pca_components[:,:2], y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "321b0b34",
   "metadata": {
    "papermill": {
     "duration": 0.064818,
     "end_time": "2021-08-13T07:17:21.699674",
     "exception": false,
     "start_time": "2021-08-13T07:17:21.634856",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "The model has 89.8% accuracy, but let's build a pipeline to fit and score the model faster.\n",
    "\n",
    "The steps of this pipeline would be the following, but this is only one type of pipeline -\n",
    "1. Imputation\n",
    "2. Scaling\n",
    "3. PCA\n",
    "4. Classification model\n",
    "\n",
    "You can change this pipeline, add addition transformations, change models, use cross validation or even use this pipeline to work with a Gridsearch."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "c8d915ec",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:21.834791Z",
     "iopub.status.busy": "2021-08-13T07:17:21.834131Z",
     "iopub.status.idle": "2021-08-13T07:17:21.835947Z",
     "shell.execute_reply": "2021-08-13T07:17:21.836435Z"
    },
    "papermill": {
     "duration": 0.071699,
     "end_time": "2021-08-13T07:17:21.836592",
     "exception": false,
     "start_time": "2021-08-13T07:17:21.764893",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "imp = SimpleImputer(strategy='constant', fill_value=0)\n",
    "scale = StandardScaler()\n",
    "pca = PCA(n_components=10)\n",
    "lr = LogisticRegression(max_iter=1000, tol=0.001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "7d0defd3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:21.971470Z",
     "iopub.status.busy": "2021-08-13T07:17:21.970871Z",
     "iopub.status.idle": "2021-08-13T07:17:21.974585Z",
     "shell.execute_reply": "2021-08-13T07:17:21.975035Z"
    },
    "papermill": {
     "duration": 0.073281,
     "end_time": "2021-08-13T07:17:21.975259",
     "exception": false,
     "start_time": "2021-08-13T07:17:21.901978",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "pipe = Pipeline(steps = [('imputation',imp),\n",
    "                         ('scaling',scale),\n",
    "                         ('pca',pca),\n",
    "                         ('model',lr)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "8972c5fd",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:22.109239Z",
     "iopub.status.busy": "2021-08-13T07:17:22.108575Z",
     "iopub.status.idle": "2021-08-13T07:17:22.525661Z",
     "shell.execute_reply": "2021-08-13T07:17:22.526440Z"
    },
    "papermill": {
     "duration": 0.485922,
     "end_time": "2021-08-13T07:17:22.526693",
     "exception": false,
     "start_time": "2021-08-13T07:17:22.040771",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pipeline(steps=[('imputation',\n",
       "                 SimpleImputer(fill_value=0, strategy='constant')),\n",
       "                ('scaling', StandardScaler()), ('pca', PCA(n_components=10)),\n",
       "                ('model', LogisticRegression(max_iter=1000, tol=0.001))])"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe.fit(X_train[new_vars], y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "441b2354",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:22.717605Z",
     "iopub.status.busy": "2021-08-13T07:17:22.716972Z",
     "iopub.status.idle": "2021-08-13T07:17:22.754258Z",
     "shell.execute_reply": "2021-08-13T07:17:22.755024Z"
    },
    "papermill": {
     "duration": 0.111563,
     "end_time": "2021-08-13T07:17:22.755300",
     "exception": false,
     "start_time": "2021-08-13T07:17:22.643737",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training accuracy: 0.9001232164860087\n"
     ]
    }
   ],
   "source": [
    "train_score = pipe.score(X_train[new_vars], y_train)\n",
    "print(\"Training accuracy:\", train_score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "b3db02c7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:22.947486Z",
     "iopub.status.busy": "2021-08-13T07:17:22.946583Z",
     "iopub.status.idle": "2021-08-13T07:17:22.959536Z",
     "shell.execute_reply": "2021-08-13T07:17:22.960316Z"
    },
    "papermill": {
     "duration": 0.087891,
     "end_time": "2021-08-13T07:17:22.960566",
     "exception": false,
     "start_time": "2021-08-13T07:17:22.872675",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test accuracy: 0.9\n"
     ]
    }
   ],
   "source": [
    "test_score = pipe.score(X_test[new_vars], y_test)\n",
    "print(\"Test accuracy:\", test_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "beec9d12",
   "metadata": {
    "papermill": {
     "duration": 0.070399,
     "end_time": "2021-08-13T07:17:23.152879",
     "exception": false,
     "start_time": "2021-08-13T07:17:23.082480",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Let's make a confusion matrix to analyze how each class is being predicted by the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "012c5a58",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:23.299159Z",
     "iopub.status.busy": "2021-08-13T07:17:23.298119Z",
     "iopub.status.idle": "2021-08-13T07:17:23.420848Z",
     "shell.execute_reply": "2021-08-13T07:17:23.421562Z"
    },
    "papermill": {
     "duration": 0.200569,
     "end_time": "2021-08-13T07:17:23.421831",
     "exception": false,
     "start_time": "2021-08-13T07:17:23.221262",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[50022,   271],\n",
       "       [ 5322,   384]])"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(y_train, pipe.predict(X_train[new_vars]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "84ba50bf",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:23.573764Z",
     "iopub.status.busy": "2021-08-13T07:17:23.571151Z",
     "iopub.status.idle": "2021-08-13T07:17:23.608185Z",
     "shell.execute_reply": "2021-08-13T07:17:23.607354Z"
    },
    "papermill": {
     "duration": 0.109186,
     "end_time": "2021-08-13T07:17:23.608435",
     "exception": false,
     "start_time": "2021-08-13T07:17:23.499249",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[12507,    67],\n",
       "       [ 1333,    93]])"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(y_test, pipe.predict(X_test[new_vars]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3ff7895f",
   "metadata": {
    "papermill": {
     "duration": 0.068268,
     "end_time": "2021-08-13T07:17:23.788669",
     "exception": false,
     "start_time": "2021-08-13T07:17:23.720401",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "We can see a high amount of type 2 error. Due to class imbalance, the model is clearly trying to predict majority of the cases as class 0. Understanding how to handle class imbalance in classification models might be the key to winning this competition :) (hint!)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "407cfc13",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:23.933119Z",
     "iopub.status.busy": "2021-08-13T07:17:23.932523Z",
     "iopub.status.idle": "2021-08-13T07:17:23.954810Z",
     "shell.execute_reply": "2021-08-13T07:17:23.955721Z"
    },
    "papermill": {
     "duration": 0.09953,
     "end_time": "2021-08-13T07:17:23.955991",
     "exception": false,
     "start_time": "2021-08-13T07:17:23.856461",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.58125"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "precision_score(y_test, pipe.predict(X_test[new_vars]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "3eb37be6",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:24.161533Z",
     "iopub.status.busy": "2021-08-13T07:17:24.160724Z",
     "iopub.status.idle": "2021-08-13T07:17:24.197024Z",
     "shell.execute_reply": "2021-08-13T07:17:24.198010Z"
    },
    "papermill": {
     "duration": 0.123943,
     "end_time": "2021-08-13T07:17:24.198308",
     "exception": false,
     "start_time": "2021-08-13T07:17:24.074365",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.06521739130434782"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recall_score(y_test, pipe.predict(X_test[new_vars]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1d1e1eef",
   "metadata": {
    "papermill": {
     "duration": 0.067672,
     "end_time": "2021-08-13T07:17:24.385001",
     "exception": false,
     "start_time": "2021-08-13T07:17:24.317329",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# 7. Creating submission file\n",
    "\n",
    "For submission, we need to make sure that the format is exactly the same as the sample.csv file. It contains 2 columns, id and churn_probability"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "ad672d46",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:24.530611Z",
     "iopub.status.busy": "2021-08-13T07:17:24.530007Z",
     "iopub.status.idle": "2021-08-13T07:17:24.533218Z",
     "shell.execute_reply": "2021-08-13T07:17:24.533650Z"
    },
    "papermill": {
     "duration": 0.080814,
     "end_time": "2021-08-13T07:17:24.533810",
     "exception": false,
     "start_time": "2021-08-13T07:17:24.452996",
     "status": "completed"
    },
    "tags": []
   },
   "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>id</th>\n",
       "      <th>churn_probability</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>69999</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>70000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>70001</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>70002</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>70003</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      id  churn_probability\n",
       "0  69999                  0\n",
       "1  70000                  0\n",
       "2  70001                  0\n",
       "3  70002                  0\n",
       "4  70003                  0"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "020d3949",
   "metadata": {
    "papermill": {
     "duration": 0.068381,
     "end_time": "2021-08-13T07:17:24.670628",
     "exception": false,
     "start_time": "2021-08-13T07:17:24.602247",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "The submission file should contain churn_probability values that have to be predicted for the unseen data provided (test.csv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "5cb95d9d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:24.811739Z",
     "iopub.status.busy": "2021-08-13T07:17:24.811179Z",
     "iopub.status.idle": "2021-08-13T07:17:24.837592Z",
     "shell.execute_reply": "2021-08-13T07:17:24.838025Z"
    },
    "papermill": {
     "duration": 0.098877,
     "end_time": "2021-08-13T07:17:24.838199",
     "exception": false,
     "start_time": "2021-08-13T07:17:24.739322",
     "status": "completed"
    },
    "tags": []
   },
   "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>id</th>\n",
       "      <th>circle_id</th>\n",
       "      <th>loc_og_t2o_mou</th>\n",
       "      <th>std_og_t2o_mou</th>\n",
       "      <th>loc_ic_t2o_mou</th>\n",
       "      <th>last_date_of_month_6</th>\n",
       "      <th>last_date_of_month_7</th>\n",
       "      <th>last_date_of_month_8</th>\n",
       "      <th>arpu_6</th>\n",
       "      <th>arpu_7</th>\n",
       "      <th>...</th>\n",
       "      <th>sachet_3g_6</th>\n",
       "      <th>sachet_3g_7</th>\n",
       "      <th>sachet_3g_8</th>\n",
       "      <th>fb_user_6</th>\n",
       "      <th>fb_user_7</th>\n",
       "      <th>fb_user_8</th>\n",
       "      <th>aon</th>\n",
       "      <th>aug_vbc_3g</th>\n",
       "      <th>jul_vbc_3g</th>\n",
       "      <th>jun_vbc_3g</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>69999</td>\n",
       "      <td>109</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6/30/2014</td>\n",
       "      <td>7/31/2014</td>\n",
       "      <td>8/31/2014</td>\n",
       "      <td>91.882</td>\n",
       "      <td>65.330</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1692</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>70000</td>\n",
       "      <td>109</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6/30/2014</td>\n",
       "      <td>7/31/2014</td>\n",
       "      <td>8/31/2014</td>\n",
       "      <td>414.168</td>\n",
       "      <td>515.568</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2533</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>70001</td>\n",
       "      <td>109</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6/30/2014</td>\n",
       "      <td>7/31/2014</td>\n",
       "      <td>8/31/2014</td>\n",
       "      <td>329.844</td>\n",
       "      <td>434.884</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>277</td>\n",
       "      <td>525.61</td>\n",
       "      <td>758.41</td>\n",
       "      <td>241.84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>70002</td>\n",
       "      <td>109</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6/30/2014</td>\n",
       "      <td>7/31/2014</td>\n",
       "      <td>8/31/2014</td>\n",
       "      <td>43.550</td>\n",
       "      <td>171.390</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1244</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>70003</td>\n",
       "      <td>109</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6/30/2014</td>\n",
       "      <td>7/31/2014</td>\n",
       "      <td>8/31/2014</td>\n",
       "      <td>306.854</td>\n",
       "      <td>406.289</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>462</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 171 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      id  circle_id  loc_og_t2o_mou  std_og_t2o_mou  loc_ic_t2o_mou  \\\n",
       "0  69999        109             0.0             0.0             0.0   \n",
       "1  70000        109             0.0             0.0             0.0   \n",
       "2  70001        109             0.0             0.0             0.0   \n",
       "3  70002        109             0.0             0.0             0.0   \n",
       "4  70003        109             0.0             0.0             0.0   \n",
       "\n",
       "  last_date_of_month_6 last_date_of_month_7 last_date_of_month_8   arpu_6  \\\n",
       "0            6/30/2014            7/31/2014            8/31/2014   91.882   \n",
       "1            6/30/2014            7/31/2014            8/31/2014  414.168   \n",
       "2            6/30/2014            7/31/2014            8/31/2014  329.844   \n",
       "3            6/30/2014            7/31/2014            8/31/2014   43.550   \n",
       "4            6/30/2014            7/31/2014            8/31/2014  306.854   \n",
       "\n",
       "    arpu_7  ...  sachet_3g_6  sachet_3g_7  sachet_3g_8  fb_user_6  fb_user_7  \\\n",
       "0   65.330  ...            0            0            0        NaN        NaN   \n",
       "1  515.568  ...            0            0            0        NaN        NaN   \n",
       "2  434.884  ...            0            0            0        NaN        NaN   \n",
       "3  171.390  ...            0            0            0        NaN        NaN   \n",
       "4  406.289  ...            0            0            0        NaN        NaN   \n",
       "\n",
       "   fb_user_8   aon  aug_vbc_3g  jul_vbc_3g  jun_vbc_3g  \n",
       "0        NaN  1692        0.00        0.00        0.00  \n",
       "1        NaN  2533        0.00        0.00        0.00  \n",
       "2        NaN   277      525.61      758.41      241.84  \n",
       "3        NaN  1244        0.00        0.00        0.00  \n",
       "4        NaN   462        0.00        0.00        0.00  \n",
       "\n",
       "[5 rows x 171 columns]"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unseen.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "da9eb3a2",
   "metadata": {
    "papermill": {
     "duration": 0.06952,
     "end_time": "2021-08-13T07:17:24.977086",
     "exception": false,
     "start_time": "2021-08-13T07:17:24.907566",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Lets first select the columns that we want to work with (or create them, if you have done any feature engineering)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "530c384a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:25.121684Z",
     "iopub.status.busy": "2021-08-13T07:17:25.120439Z",
     "iopub.status.idle": "2021-08-13T07:17:25.139846Z",
     "shell.execute_reply": "2021-08-13T07:17:25.140254Z"
    },
    "papermill": {
     "duration": 0.093498,
     "end_time": "2021-08-13T07:17:25.140420",
     "exception": false,
     "start_time": "2021-08-13T07:17:25.046922",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(30000, 16)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "submission_data = unseen.set_index('id')[new_vars]\n",
    "submission_data.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d584eab9",
   "metadata": {
    "papermill": {
     "duration": 0.069588,
     "end_time": "2021-08-13T07:17:25.279867",
     "exception": false,
     "start_time": "2021-08-13T07:17:25.210279",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Next, lets create a new column in the unseen dataset called churn_probability and use the model pipeline to predict the probabilities for this data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "ab2459d7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:25.422640Z",
     "iopub.status.busy": "2021-08-13T07:17:25.421696Z",
     "iopub.status.idle": "2021-08-13T07:17:25.457451Z",
     "shell.execute_reply": "2021-08-13T07:17:25.456659Z"
    },
    "papermill": {
     "duration": 0.108369,
     "end_time": "2021-08-13T07:17:25.457668",
     "exception": false,
     "start_time": "2021-08-13T07:17:25.349299",
     "status": "completed"
    },
    "tags": []
   },
   "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>id</th>\n",
       "      <th>churn_probability</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>69999</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>70000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>70001</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>70002</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>70003</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      id  churn_probability\n",
       "0  69999                  0\n",
       "1  70000                  0\n",
       "2  70001                  0\n",
       "3  70002                  0\n",
       "4  70003                  0"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unseen['churn_probability'] = pipe.predict(submission_data)\n",
    "output = unseen[['id','churn_probability']]\n",
    "output.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d4cc053f",
   "metadata": {
    "papermill": {
     "duration": 0.124336,
     "end_time": "2021-08-13T07:17:25.701056",
     "exception": false,
     "start_time": "2021-08-13T07:17:25.576720",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Finally, lets create a csv file out of this dataset, ensuring to set index=False to avoid an addition column in the csv."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "cd0ce11a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-13T07:17:25.868564Z",
     "iopub.status.busy": "2021-08-13T07:17:25.867848Z",
     "iopub.status.idle": "2021-08-13T07:17:25.914141Z",
     "shell.execute_reply": "2021-08-13T07:17:25.913627Z"
    },
    "papermill": {
     "duration": 0.12837,
     "end_time": "2021-08-13T07:17:25.914318",
     "exception": false,
     "start_time": "2021-08-13T07:17:25.785948",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "output.to_csv('submission_pca_lr_13jul.csv',index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "afc24e68",
   "metadata": {
    "papermill": {
     "duration": 0.08214,
     "end_time": "2021-08-13T07:17:26.080507",
     "exception": false,
     "start_time": "2021-08-13T07:17:25.998367",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "You can now take this file and upload it as a submission on Kaggle."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8a688423",
   "metadata": {
    "papermill": {
     "duration": 0.070782,
     "end_time": "2021-08-13T07:17:26.223247",
     "exception": false,
     "start_time": "2021-08-13T07:17:26.152465",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.4"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 57.533667,
   "end_time": "2021-08-13T07:17:27.204430",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2021-08-13T07:16:29.670763",
   "version": "2.3.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}