1.Quora.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Quora Question Pair Similarity"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Description"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Quora is a place to gain and share knowledge—about anything. It’s a platform to ask questions and connect with people who contribute unique insights and quality answers. This empowers people to learn from each other and to better understand the world.\n",
    "\n",
    "Over 100 million people visit Quora every month, so it's no surprise that many people ask similarly worded questions. Multiple questions with the same intent can cause seekers to spend more time finding the best answer to their question, and make writers feel they need to answer multiple versions of the same question. Quora values canonical questions because they provide a better experience to active seekers and writers, and offer more value to both of these groups in the long term."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<b>Problem Statement</b>\n",
    "\n",
    "- Identify which questions asked on Quora are duplicates of questions that have already been asked.\n",
    "- This could be useful to instantly provide answers to questions that have already been answered.\n",
    "- We are tasked with predicting whether a pair of questions are duplicates or not."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Sources/Useful Links\n",
    "- Source : https://www.kaggle.com/c/quora-question-pairs \n",
    "\n",
    "<b>Useful Links</b>\n",
    "- Discussions : https://www.kaggle.com/anokas/data-analysis-xgboost-starter-0-35460-lb/comments\n",
    "- Kaggle Winning Solution and other approaches: https://www.dropbox.com/sh/93968nfnrzh8bp5/AACZdtsApc1QSTQc7X0H3QZ5a?dl=0\n",
    "- Blog 1 : https://engineering.quora.com/Semantic-Question-Matching-with-Deep-Learning\n",
    "- Blog 2 : https://towardsdatascience.com/identifying-duplicate-questions-on-quora-top-12-on-kaggle-4c1cf93f1c30"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Real world/Business Objectives and Constraints\n",
    "- The cost of a mis-classification can be very high.\n",
    "- You would want a probability of a pair of questions to be duplicates so that you can choose any threshold of choice.\n",
    "- No strict latency concerns.\n",
    "- Interpretability is partially important."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Machine Learning Problem"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Overview\n",
    "- Data will be in a file Train.csv \n",
    "- Train.csv contains 5 columns : qid1, qid2, question1, question2, is_duplicate \n",
    "- Size of Train.csv - 60MB \n",
    "- Number of rows in Train.csv = 404,290"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Example Data Points\n",
    "\"id\",\"qid1\",\"qid2\",\"question1\",\"question2\",\"is_duplicate\"\n",
    "\n",
    "\"0\",\"1\",\"2\",\"What is the step by step guide to invest in share market in india?\",\"What is the step by step guide to invest in share market?\",\"0\"\n",
    "\n",
    "\"1\",\"3\",\"4\",\"What is the story of Kohinoor (Koh-i-Noor) Diamond?\",\"What would happen if the Indian government stole the Kohinoor (Koh-i-Noor) diamond back?\",\"0\"\n",
    "\n",
    "\"7\",\"15\",\"16\",\"How can I be a good geologist?\",\"What should I do to be a great geologist?\",\"1\"\n",
    "\n",
    "\"11\",\"23\",\"24\",\"How do I read and find my YouTube comments?\",\"How can I see all my Youtube comments?\",\"1\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Mapping the real world problem to an ML problem"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Type of Machine Leaning Problem\n",
    "It is a binary classification problem, for a given pair of questions we need to predict if they are duplicate or not."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Performance Metric\n",
    "- Source: https://www.kaggle.com/c/quora-question-pairs#evaluation\n",
    "\n",
    "<b>Metric(s):</b>\n",
    "\n",
    "- log-loss : https://www.kaggle.com/wiki/LogarithmicLoss\n",
    "- Binary Confusion Matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train and Test Construction\n",
    "We build train and test by randomly splitting in the ratio of 70:30 or 80:20 whatever we choose as we have sufficient points to work with."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<script type=\"text/javascript\">window.PlotlyConfig = {MathJaxConfig: 'local'};</script><script type=\"text/javascript\">if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}</script><script>requirejs.config({paths: { 'plotly': ['https://cdn.plot.ly/plotly-latest.min']},});if(!window._Plotly) {require(['plotly'],function(plotly) {window._Plotly=plotly;});}</script>"
      ],
      "text/vnd.plotly.v1+html": [
       "<script type=\"text/javascript\">window.PlotlyConfig = {MathJaxConfig: 'local'};</script><script type=\"text/javascript\">if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}</script><script>requirejs.config({paths: { 'plotly': ['https://cdn.plot.ly/plotly-latest.min']},});if(!window._Plotly) {require(['plotly'],function(plotly) {window._Plotly=plotly;});}</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from subprocess import check_output\n",
    "%matplotlib inline\n",
    "import plotly.offline as py\n",
    "py.init_notebook_mode(connected=True)\n",
    "import plotly.graph_objs as go\n",
    "import plotly.tools as tls\n",
    "import os\n",
    "import gc\n",
    "\n",
    "import re\n",
    "from nltk.corpus import stopwords\n",
    "import distance\n",
    "from nltk.stem import PorterStemmer\n",
    "from bs4 import BeautifulSoup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Reading data and basic stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of data points: 404290\n"
     ]
    }
   ],
   "source": [
    "df = pd.read_csv(\"train.csv\")\n",
    "\n",
    "print(\"Number of data points:\",df.shape[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>qid1</th>\n",
       "      <th>qid2</th>\n",
       "      <th>question1</th>\n",
       "      <th>question2</th>\n",
       "      <th>is_duplicate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>What is the step by step guide to invest in sh...</td>\n",
       "      <td>What is the step by step guide to invest in sh...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>What is the story of Kohinoor (Koh-i-Noor) Dia...</td>\n",
       "      <td>What would happen if the Indian government sto...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>How can I increase the speed of my internet co...</td>\n",
       "      <td>How can Internet speed be increased by hacking...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>Why am I mentally very lonely? How can I solve...</td>\n",
       "      <td>Find the remainder when [math]23^{24}[/math] i...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "      <td>Which one dissolve in water quikly sugar, salt...</td>\n",
       "      <td>Which fish would survive in salt water?</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id  qid1  qid2                                          question1  \\\n",
       "0   0     1     2  What is the step by step guide to invest in sh...   \n",
       "1   1     3     4  What is the story of Kohinoor (Koh-i-Noor) Dia...   \n",
       "2   2     5     6  How can I increase the speed of my internet co...   \n",
       "3   3     7     8  Why am I mentally very lonely? How can I solve...   \n",
       "4   4     9    10  Which one dissolve in water quikly sugar, salt...   \n",
       "\n",
       "                                           question2  is_duplicate  \n",
       "0  What is the step by step guide to invest in sh...             0  \n",
       "1  What would happen if the Indian government sto...             0  \n",
       "2  How can Internet speed be increased by hacking...             0  \n",
       "3  Find the remainder when [math]23^{24}[/math] i...             0  \n",
       "4            Which fish would survive in salt water?             0  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 404290 entries, 0 to 404289\n",
      "Data columns (total 6 columns):\n",
      "id              404290 non-null int64\n",
      "qid1            404290 non-null int64\n",
      "qid2            404290 non-null int64\n",
      "question1       404289 non-null object\n",
      "question2       404288 non-null object\n",
      "is_duplicate    404290 non-null int64\n",
      "dtypes: int64(4), object(2)\n",
      "memory usage: 18.5+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We are given a minimal number of data fields here, consisting of:\n",
    "\n",
    "- id: Looks like a simple rowID\n",
    "- qid{1, 2}: The unique ID of each question in the pair\n",
    "- question{1, 2}: The actual textual contents of the questions.\n",
    "- is_duplicate: The label that we are trying to predict - whether the two questions are duplicates of each other."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Distribution of data points among output classes\n",
    "Number of duplicate(smilar) and non-duplicate(non similar) questions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x22cc6988d30>"
      ]
     },
     "execution_count": 6,
     "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": [
    "df.groupby(\"is_duplicate\")['id'].count().plot.bar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "~> Total number of question pairs for training:\n",
      "   404290\n"
     ]
    }
   ],
   "source": [
    "print('~> Total number of question pairs for training:\\n   {}'.format(len(df)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "~> Question pairs are not Similar (is_duplicate = 0):\n",
      "   63.08%\n",
      "\n",
      "~> Question pairs are Similar (is_duplicate = 1):\n",
      "   36.92%\n"
     ]
    }
   ],
   "source": [
    "print('~> Question pairs are not Similar (is_duplicate = 0):\\n   {}%'.format(100 - round(df['is_duplicate'].mean()*100, 2)))\n",
    "print('\\n~> Question pairs are Similar (is_duplicate = 1):\\n   {}%'.format(round(df['is_duplicate'].mean()*100, 2)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Number of unique questions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of  Unique Questions are: 537933\n",
      "\n",
      "Number of unique questions that appear more than one time: 111780 (20.77953945937505%)\n",
      "\n",
      "Max number of times a single question is repeated: 157\n",
      "\n"
     ]
    }
   ],
   "source": [
    "qids = pd.Series(df['qid1'].tolist() + df['qid2'].tolist())\n",
    "unique_qs = len(np.unique(qids))\n",
    "qs_morethan_onetime = np.sum(qids.value_counts() > 1)\n",
    "print ('Total number of  Unique Questions are: {}\\n'.format(unique_qs))\n",
    "#print len(np.unique(qids))\n",
    "\n",
    "print ('Number of unique questions that appear more than one time: {} ({}%)\\n'.format(qs_morethan_onetime,qs_morethan_onetime/unique_qs*100))\n",
    "\n",
    "print ('Max number of times a single question is repeated: {}\\n'.format(max(qids.value_counts()))) \n",
    "\n",
    "q_vals=qids.value_counts()\n",
    "\n",
    "q_vals=q_vals.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x = [\"unique_questions\" , \"Repeated Questions\"]\n",
    "y =  [unique_qs , qs_morethan_onetime]\n",
    "\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.title (\"Plot representing unique and repeated questions  \")\n",
    "sns.barplot(x,y)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Checking for Duplicates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of duplicate questions 0\n"
     ]
    }
   ],
   "source": [
    "#checking whether there are any repeated pair of questions\n",
    "\n",
    "pair_duplicates = df[['qid1','qid2','is_duplicate']].groupby(['qid1','qid2']).count().reset_index()\n",
    "\n",
    "print (\"Number of duplicate questions\",(pair_duplicates).shape[0] - df.shape[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Number of occurrences of each question"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Maximum number of times a single question is repeated: 157\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20, 10))\n",
    "\n",
    "plt.hist(qids.value_counts(), bins=160)\n",
    "\n",
    "plt.yscale('log', nonposy='clip')\n",
    "\n",
    "plt.title('Log-Histogram of question appearance counts')\n",
    "\n",
    "plt.xlabel('Number of occurences of question')\n",
    "\n",
    "plt.ylabel('Number of questions')\n",
    "\n",
    "print ('Maximum number of times a single question is repeated: {}\\n'.format(max(qids.value_counts())))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Checking for NULL values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id              0\n",
       "qid1            0\n",
       "qid2            0\n",
       "question1       1\n",
       "question2       2\n",
       "is_duplicate    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "            id    qid1    qid2                         question1  \\\n",
      "105780  105780  174363  174364    How can I develop android app?   \n",
      "201841  201841  303951  174364  How can I create an Android app?   \n",
      "363362  363362  493340  493341                               NaN   \n",
      "\n",
      "                                                question2  is_duplicate  \n",
      "105780                                                NaN             0  \n",
      "201841                                                NaN             0  \n",
      "363362  My Chinese name is Haichao Yu. What English na...             0  \n"
     ]
    }
   ],
   "source": [
    "#Checking whether there are any rows with null values\n",
    "nan_rows = df[df.isnull().any(1)]\n",
    "print (nan_rows)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Empty DataFrame\n",
      "Columns: [id, qid1, qid2, question1, question2, is_duplicate]\n",
      "Index: []\n"
     ]
    }
   ],
   "source": [
    "# Filling the null values with ' '\n",
    "df = df.fillna('')\n",
    "nan_rows = df[df.isnull().any(1)]\n",
    "print (nan_rows)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Basic Feature Extraction (before cleaning)\n",
    "Let us now construct a few features like:\n",
    "\n",
    "- freq_qid1 = Frequency of qid1's #ie, number of times question1 occur\n",
    "- freq_qid2 = Frequency of qid2's\n",
    "- q1len = Length of q1\n",
    "- q2len = Length of q2\n",
    "- q1_n_words = Number of words in Question 1\n",
    "- q2_n_words = Number of words in Question 2\n",
    "- word_Common = (Number of common unique words in Question 1 and Question 2)\n",
    "- word_Total =(Total num of words in Question 1 + Total num of words in Question 2)\n",
    "- word_share = (word_common)/(word_Total)\n",
    "- freq_q1+freq_q2 = sum total of frequency of qid1 and qid2\n",
    "- freq_q1-freq_q2 = absolute difference of frequency of qid1 and qid2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\ndef common_words(row):\\n    finalset = set(row[\\'question1\\'].lower().split(\" \")) & set(row[\\'question2\\'].lower().split(\" \"))\\n    #print(finalset)\\n    return 1.0 * len(finalset)\\n\\ndef total_words(row):\\n    set1 = set(row[\\'question1\\'].lower().split(\" \"))\\n    set2 = set(row[\\'question2\\'].lower().split(\" \"))\\n    return 1.0 * (len(set1) + len(set2))\\n    \\n    \\ndf[\\'freq_qid1\\'] = df[\\'qid1\\'].map(df[\\'qid1\\'].value_counts())\\ndf[\\'freq_qid2\\'] = df[\\'qid1\\'].map(df[\\'qid1\\'].value_counts())\\ndf[\\'q1len\\'] = df[\\'question1\\'].apply(lambda x: len(x))\\ndf[\\'q2len\\'] = df[\\'question2\\'].apply(lambda x: len(x))\\ndf[\\'q1_n_words\\'] = df[\\'question1\\'].apply(lambda x: len(x.split()))\\ndf[\\'q2_n_words\\'] = df[\\'question2\\'].apply(lambda x: len(x.split()))\\ndf[\\'word_Common\\'] = df.apply(common_words,axis=1)\\ndf[\\'word_Total\\'] = df.apply(total_words, axis=1)\\ndf[\\'word_share\\'] = df[\\'word_Common\\']/df[\\'word_Total\\']\\ndf[\\'freq_q1+q2\\'] = df[\\'freq_qid1\\']+df[\\'freq_qid2\\']\\ndf[\\'freq_q1-q2\\'] = abs(df[\\'freq_qid1\\']-df[\\'freq_qid2\\'])\\ndf.head()\\n    \\n'"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\"\n",
    "def common_words(row):\n",
    "    finalset = set(row['question1'].lower().split(\" \")) & set(row['question2'].lower().split(\" \"))\n",
    "    #print(finalset)\n",
    "    return 1.0 * len(finalset)\n",
    "\n",
    "def total_words(row):\n",
    "    set1 = set(row['question1'].lower().split(\" \"))\n",
    "    set2 = set(row['question2'].lower().split(\" \"))\n",
    "    return 1.0 * (len(set1) + len(set2))\n",
    "    \n",
    "    \n",
    "df['freq_qid1'] = df['qid1'].map(df['qid1'].value_counts())\n",
    "df['freq_qid2'] = df['qid1'].map(df['qid1'].value_counts())\n",
    "df['q1len'] = df['question1'].apply(lambda x: len(x))\n",
    "df['q2len'] = df['question2'].apply(lambda x: len(x))\n",
    "df['q1_n_words'] = df['question1'].apply(lambda x: len(x.split()))\n",
    "df['q2_n_words'] = df['question2'].apply(lambda x: len(x.split()))\n",
    "df['word_Common'] = df.apply(common_words,axis=1)\n",
    "df['word_Total'] = df.apply(total_words, axis=1)\n",
    "df['word_share'] = df['word_Common']/df['word_Total']\n",
    "df['freq_q1+q2'] = df['freq_qid1']+df['freq_qid2']\n",
    "df['freq_q1-q2'] = abs(df['freq_qid1']-df['freq_qid2'])\n",
    "df.head()\n",
    "    \n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>qid1</th>\n",
       "      <th>qid2</th>\n",
       "      <th>question1</th>\n",
       "      <th>question2</th>\n",
       "      <th>is_duplicate</th>\n",
       "      <th>freq_qid1</th>\n",
       "      <th>freq_qid2</th>\n",
       "      <th>q1len</th>\n",
       "      <th>q2len</th>\n",
       "      <th>q1_n_words</th>\n",
       "      <th>q2_n_words</th>\n",
       "      <th>word_Common</th>\n",
       "      <th>word_Total</th>\n",
       "      <th>word_share</th>\n",
       "      <th>freq_q1+q2</th>\n",
       "      <th>freq_q1-q2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>What is the step by step guide to invest in sh...</td>\n",
       "      <td>What is the step by step guide to invest in sh...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>66</td>\n",
       "      <td>57</td>\n",
       "      <td>14</td>\n",
       "      <td>12</td>\n",
       "      <td>10.0</td>\n",
       "      <td>23.0</td>\n",
       "      <td>0.434783</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>What is the story of Kohinoor (Koh-i-Noor) Dia...</td>\n",
       "      <td>What would happen if the Indian government sto...</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>51</td>\n",
       "      <td>88</td>\n",
       "      <td>8</td>\n",
       "      <td>13</td>\n",
       "      <td>4.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>How can I increase the speed of my internet co...</td>\n",
       "      <td>How can Internet speed be increased by hacking...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>73</td>\n",
       "      <td>59</td>\n",
       "      <td>14</td>\n",
       "      <td>10</td>\n",
       "      <td>4.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>Why am I mentally very lonely? How can I solve...</td>\n",
       "      <td>Find the remainder when [math]23^{24}[/math] i...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>50</td>\n",
       "      <td>65</td>\n",
       "      <td>11</td>\n",
       "      <td>9</td>\n",
       "      <td>0.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "      <td>Which one dissolve in water quikly sugar, salt...</td>\n",
       "      <td>Which fish would survive in salt water?</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>76</td>\n",
       "      <td>39</td>\n",
       "      <td>13</td>\n",
       "      <td>7</td>\n",
       "      <td>2.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id  qid1  qid2                                          question1  \\\n",
       "0   0     1     2  What is the step by step guide to invest in sh...   \n",
       "1   1     3     4  What is the story of Kohinoor (Koh-i-Noor) Dia...   \n",
       "2   2     5     6  How can I increase the speed of my internet co...   \n",
       "3   3     7     8  Why am I mentally very lonely? How can I solve...   \n",
       "4   4     9    10  Which one dissolve in water quikly sugar, salt...   \n",
       "\n",
       "                                           question2  is_duplicate  freq_qid1  \\\n",
       "0  What is the step by step guide to invest in sh...             0          1   \n",
       "1  What would happen if the Indian government sto...             0          4   \n",
       "2  How can Internet speed be increased by hacking...             0          1   \n",
       "3  Find the remainder when [math]23^{24}[/math] i...             0          1   \n",
       "4            Which fish would survive in salt water?             0          3   \n",
       "\n",
       "   freq_qid2  q1len  q2len  q1_n_words  q2_n_words  word_Common  word_Total  \\\n",
       "0          1     66     57          14          12         10.0        23.0   \n",
       "1          1     51     88           8          13          4.0        20.0   \n",
       "2          1     73     59          14          10          4.0        24.0   \n",
       "3          1     50     65          11           9          0.0        19.0   \n",
       "4          1     76     39          13           7          2.0        20.0   \n",
       "\n",
       "   word_share  freq_q1+q2  freq_q1-q2  \n",
       "0    0.434783           2           0  \n",
       "1    0.200000           5           3  \n",
       "2    0.166667           2           0  \n",
       "3    0.000000           2           0  \n",
       "4    0.100000           4           2  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "if os.path.isfile('df_fe_without_preprocessing_train.csv'):\n",
    "    df = pd.read_csv(\"df_fe_without_preprocessing_train.csv\",encoding='latin-1')\n",
    "else:\n",
    "    df['freq_qid1'] = df.groupby('qid1')['qid1'].transform('count') \n",
    "    df['freq_qid2'] = df.groupby('qid2')['qid2'].transform('count')\n",
    "    df['q1len'] = df['question1'].str.len() \n",
    "    df['q2len'] = df['question2'].str.len()\n",
    "    df['q1_n_words'] = df['question1'].apply(lambda row: len(row.split(\" \")))\n",
    "    df['q2_n_words'] = df['question2'].apply(lambda row: len(row.split(\" \")))\n",
    "\n",
    "    def normalized_word_Common(row):\n",
    "        w1 = set(map(lambda word: word.lower().strip(), row['question1'].split(\" \")))\n",
    "        w2 = set(map(lambda word: word.lower().strip(), row['question2'].split(\" \")))    \n",
    "        return 1.0 * len(w1 & w2)\n",
    "    df['word_Common'] = df.apply(normalized_word_Common, axis=1)\n",
    "\n",
    "    def normalized_word_Total(row):\n",
    "        w1 = set(map(lambda word: word.lower().strip(), row['question1'].split(\" \")))\n",
    "        w2 = set(map(lambda word: word.lower().strip(), row['question2'].split(\" \")))    \n",
    "        return 1.0 * (len(w1) + len(w2))\n",
    "    df['word_Total'] = df.apply(normalized_word_Total, axis=1)\n",
    "\n",
    "    def normalized_word_share(row):\n",
    "        w1 = set(map(lambda word: word.lower().strip(), row['question1'].split(\" \")))\n",
    "        w2 = set(map(lambda word: word.lower().strip(), row['question2'].split(\" \")))    \n",
    "        return 1.0 * len(w1 & w2)/(len(w1) + len(w2))\n",
    "    df['word_share'] = df.apply(normalized_word_share, axis=1)\n",
    "\n",
    "    df['freq_q1+q2'] = df['freq_qid1']+df['freq_qid2']\n",
    "    df['freq_q1-q2'] = abs(df['freq_qid1']-df['freq_qid2'])\n",
    "\n",
    "    df.to_csv(\"df_fe_without_preprocessing_train.csv\", index=False)\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Analysis of some of the extracted features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Minimum length of the questions in question1 :  1\n",
      "Minimum length of the questions in question2 :  1\n",
      "Number of Questions with minimum length [question1] : 67\n",
      "Number of Questions with minimum length [question2] : 24\n"
     ]
    }
   ],
   "source": [
    "print (\"Minimum length of the questions in question1 : \" , min(df['q1_n_words']))\n",
    "\n",
    "print (\"Minimum length of the questions in question2 : \" , min(df['q2_n_words']))\n",
    "\n",
    "print (\"Number of Questions with minimum length [question1] :\", df[df['q1_n_words']== 1].shape[0])\n",
    "print (\"Number of Questions with minimum length [question2] :\", df[df['q2_n_words']== 1].shape[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualization: Word Share"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\scipy\\stats\\stats.py:1713: FutureWarning:\n",
      "\n",
      "Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtUAAAHkCAYAAADxfdP+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3Xl83VWB9/HPuUtu9qRLuu/QhbZAC2VViqIiCILj8LjMOM8sOhUUdVQcwF0cUcbtEYZRquKGoxYYx4KCw760pXSlbbqRNk2bLmmaPbm56+88fySppbSlaXPv+d17v+/Xq682N7c53+Ymvd97cn7nGGstIiIiIiJy6gKuA4iIiIiI5DqVahERERGR06RSLSIiIiJymlSqRUREREROk0q1iIiIiMhpUqkWERERETlNKtUiIiIiIqdJpVpERERE5DSpVIuIiIiInKaQ6wCnYuTIkXbKlCmuY4iIDNqaNWsOWWtrXOfIJv2fLSK57GT/387JUj1lyhRWr17tOoaIyKAZYxpcZ8g2/Z8tIrnsZP/f1vIPEREREZHTpFItIiIiInKaVKpFRERERE6TSrWIiIiIyGlSqRYREREROU0q1SIiIiIip0mlWkRERETkNKlUi4iIiIicJpVqEREREZHTpFItIiIiInKaVKpFRERERE6TSrWIiIiIyGlSqRYREREROU0q1SIiIiIip0mlWkRERETkNKlUi4iIiIicJpVqEREREZHTpFKd59LpNI899hjRaNR1FBEREZG8pVKd51avXs1dd93FkiVLXEcRERERyVsh1wEkszo6OgBobGx0nEREJPsWL379bYsWZT+HiOQ/zVSLyCnr7e3lt7/9LevWrXMdRURExCnNVIvIKVu+fDk/+tGPqKgo55FHHnUdR0RExBnNVIvIKRtYVtTV1U0qlXKcRkRExB2VahE5ZXv37j3856amJodJRERE3FKpFpFT1rhnD0Fj+/6si2FFRKSAqVSLyClJJpPU1dXxpjFxDLBt2zbXkURERJxRqRaRU7Jjxw4SySRnD08yrtyjtnaT60giIiLOqFTnOWut6wiSp2prawE4syrFmRUJNm/ahOd5jlOJiIi4oVKd55LJpOsIkqdqa2sZVgwjij3OrErR1RNlz549rmOJiIg4oVKd5xKJhOsIkqc2bniF6ZVxAKZX9b1427RJS0BERKQwZbxUG2OuMsZsM8bUGWNuO8b7/8EY02yMWd//6yOZzlRIotEooGUgMrQOHjxI86EWplf17U09ptSjvOgvS0JEREQKTUZPVDTGBIF7gXcAjcAqY8xSa+3mo+76O2vtzZnMUqja2toA6OnpcZxE8slAeR6YoQ4YOLMiwcYNr7iMJSIi4kymZ6ovBOqstTuttQngt8D1GR5TjtDe3g5Aa3+5FhkKdXV1BA1MKk8fvm1qZYrGvfuIxWIOk4mIiLiR6VI9HjjyyqXG/tuO9tfGmA3GmIeMMRMznKmgbNiwAYDWVpVqGTr19fWMKfP4bV0pD2wvBWBCWRprLQ0NDY7TiYiIZF+mS7U5xm1HL+59BJhirT0HeBL4xTE/kDGLjDGrjTGrm5ubhzhm/jo8U93aQjwed5xG8sXOHXVMLE2yuzvE7u6+VWQTyvvWV9fX17uMJiIi4kSmS3UjcOTM8wRg35F3sNa2WGsH2t6PgfOP9YGstYuttQustQtqamoyEjbftLa2kkwmscEwXjrN9u3bXUeSPBCNRjnQdJAJRyz9ABhd4hEOws6dOx0lExERcSfTpXoVMN0YM9UYUwR8AFh65B2MMWOPePM6YEuGMxWMLVv6PpVeuOw1b4ucjldffRWAyf0z0wMCBiaWpdm+XceVi4hI4cno7h/W2pQx5mbgz0AQuN9aW2uMuQNYba1dCnzSGHMdkAJagX/IZKZCsnlz3yYrNhSB4kptdyZDYuAnHlMqU69735SKJC9t24bneQQC2gZfREQKR0ZLNYC19k/An4667ctH/Pl24PZM5yg0qVSKJ596ChsMgzEky0fz0sqVdHZ2UllZ6Tqe5LBt27YxrBiqil6/9/nUihRP743R2NjIpEmTHKQTERFxQ1NJeerZZ5+l6cCBw0s/EqPnEI/F+P3vf+84meSyzs5OXnjheeZWH3vbvNnDkhgDjz/+eJaTiYiIuKVSnYc8z+OXv/oVtnQYNlTcd1vpcFLVE1ny4EOHT1kUGaylS5cSjye4atKxS3VNiceCkXGW/uF/9HUmIiIFRaU6Dy1btozdDQ3Expzzmk0N42PPpae7i6VLlx7/L4scRzwe5+GHHuScEUkmHrXzx5HeNSlGd0+UP/7xj1lMJyIi4pZKdZ7Zv38/3/72d6CkmtTwqa95n1c+inTVBH76058evohR5GT99Kc/pa29g6snnngG+oyqFDOrUzzwq1/S2NiYpXQiIiJuqVTnkZ6eHm697Ta6ojF6zrgCzOsf3t6pl5EMlnD75z9PU1OTg5SSix588EGWLFnCOyb0MnvY63f9ONo/zeoiHeviXz93C21tOs1TRETyn0p1nkilUnz1q19j9+499JzxVryS6mPez4ZL6DnzbXR2R7n11tu07lXe0NNPP829997LBTUJ/nZ6FHOsc1KPMrbU4zNnd3Do4AFuu/Vf9XUmIiJ5T6U6D1hrufvuu1m16mViky8hXTnuhPf3SobRM+2t7GrYxVe/9jUSiUSWkkqueemll/jmnd9gZnWKj87uInAShXrAmVUpPj6nk+2vvspXvvJlFWsREclrKtU5Lh6Pc8cdX2fp0qUkxpxNsmbmSf29dNV4YpMu4eWVK/nsZ2+ho6Mjw0kll3iexy9/+Utuv/02xpUk+dTZnRQFB/9x5o9M8k8zu1m9ejU33fhRdu/ePfRhRUREfEClOoe1tbXxL5/+NM888zTxCQuIT1gwqL+fHDWL3mmXs7G2lhtv+pguKhMAuru7+eIXv8j999/PJaPifPG8NsrDrz/o5WRdPi7Ov57bQeuBPdz40UW88MILQ5hWRETEH1Sqc9SuXbv46I03sXXbdnrPuILE2HM4qcWuR0mNOIPojKs40NzCR2+8ifXr12cgreSKnTt38tFF/8zKFcv50PQePjq7m8gpzFAfbc7wFF9b0MaYcA9f+tKX+MlPfkI6ffxt+URERHKNSnUOWr16NTd97GM0t3XSPeNqUsOnnNbHS1eMpmvWtXSng3z2s5/lscceG5qgkjOstTz22GPcdNON9LTs57b5HVw5MXYqr9OOa2Sxx+fnt3P52BgPPPAAt9zyWVpaWoZuABEREYdUqnOItZbf/e53fO5znyNKMV2zrsUrrxmaj11cSfesa0iUjeauu+7innvuIZV6463TJPdFo1G+8Y1vcNdddzGtNMrXFrQyszozj31RED58Vg8fmdXN5g2v8OF//AdWrlyZkbFERESySaU6R8Tjcb7xjTv54Q9/SKJ6Mt2zrsFGyod2kFCE6IwrSYyezcMPP8wtn/sc7e3tQzuG+Mqrr77KP3/kwzz11JO8d2qUW+d1MCxy6uunT9bCcXG+uqCNcq+DW2+9lfvuu08v4kREJKepVOeApqYmPv7xm3nyySeIjz+P2BlvhWA4M4OZAPFJF9M79TJeeWUDixZ9lLq6usyMJU4tXbqUj910I9HW/dw+r4P3TO0d1JZ5p2t8WZqvnt/GW8fF+M1vfsMnbv44zc3N2Qsgp8UY82ljTK0xZpMx5jfGmGLXmUREXFKp9rna2lo+8s+L2LGrgej0t5MYN++ULkgcrNTI6XTPfBcHO7q56WMf4/nnn8/4mJId1lp+9rOf8b3vfY+zqmL824JWZp3EKYmZUBSEf5zVw81zu6iv284nbv64dqHJAcaY8cAngQXW2rlAEPiA21QiIm6pVPvY9u3bueWWz9GVsHTPupZ09aSsju+V19B91ruJF1Xz1a9+leXLl2d1fBl6nudx991384tf/ILLxsb49NmdVBRlfrnHG7lwVILb5rXR3XaQT9z8cXbs2OE6kryxEFBijAkBpcA+x3lERJxSqfaphoYGPnvLLfR6AbpnXHXcY8czzYZL6ZlxJamSEXzlK1/Rlns5LJVK8c1vfpPf//73XDWxl4/M6iHoo/8BplWm+cL8doh18KlPfoJNmza5jiTHYa3dC3wH2A3sBzqstf975H2MMYuMMauNMau1rEdECoGPnlJlQFNTE5/5zGfpjiXpnvHOob8gcbCCRUSnv4NkuJzbbr+dbdu2uc0jp+QHP/gBTzzxBDdMi/LBM6PZWEU0aOPL0nxpfhtltofPfvYz7Nmzx3UkOQZjzDDgemAqMA4oM8Z86Mj7WGsXW2sXWGsX1NQMzS5FIiJ+plLtMx0dHXz6M5+htaOL7ulXYourXEcCwIaL6Z5+JTEb4pZbPqd1rzlmw4YNPPLII1w9sZfrpvT6slAPGFnicfv8doJegu9//3tY6355irzO24F6a22ztTYJ/DdwqeNMIiJOqVT7zH//93+zb+9ees58G17pCNdxXsMWldE9/Z1098b4+c9/7jqOnKRkMsl3v/NtRpbAe6dFXcc5KcMjHjdM7Wbt2nU8+eSTruPI6+0GLjbGlBpjDPA2YIvjTCIiTqlU+0gqlWLpI4+SqppAumKM6zjHZIsriQ8/k2eefVZ7WOeIJUuW0LB7D383vXNIjhzPlivGx5lWmebe/7iHrq4u13HkCNbalcBDwFpgI33PJYudhhIRcUyl2kdWrFhBW2sLiZpZrqOcUHLUTNKpFI8//rjrKPIG0uk0Dy75HeeOSDB/ZNJ1nEEJGPj7GV20d3TyxBNPuI4jR7HWfsVaO8taO9da+3fW2rjrTCIiLqlU+8gfli6FSDnp6gmuo5yQVzIMr2IM//OHP+B5nus4cgJbt26lvaOTS0fnZt+ZWplmbJll+bJlrqOIiIickEq1j2zZsoVE5Xgw/n9YElUTObB/P93d3a6jyAksX76cgIFzRuTWLPWR5o2IsX79enp6elxHEREROS7/t7cCUlxcgrG5MfNrbBqAkpISx0nkRJYve5GZ1SnKwrm7g8b8EQlS6TSrVq1yHUVEROS4VKp9pKysFNIJ1zFOikknCYXDhMNh11HkOPbv30/9rgbmjcjNpR8DplelKAv3XXMgIiLiVyrVPlJeVoZJ58iP6dMJSkpKXaeQE3jppZcAmDcyN16oHU8wAOcMj7NyxXLS6bTrOCIiIsekUu0jo0aNIhTvhBw47CIY66SmZqTrGHICK1YsZ0yZZWxpbiwpOpF5IxK0d3axdetW11FERESOSaXaRxYsWADxbgIxn+//nE4Q7G7iogsvdJ1EjuPgwYOsXbOW+SNirqMMiXNGJAkF0DaOIiLiWyrVPnJhf0kNtvv7CPBQ536w3uG84j+/+93vsF6ad4zPj1JdFrZcNibG44/9iUOHDrmOIyIi8joq1T4yatQoJk2eTLjT36U62NFIpLiYuXPnuo4ix9DW1sajjyzl0tExRpbk/tKPAddM7iWdTrNkyRLXUURERF5Hpdpn3nTppQS7myDl0x0brCXSsYcLL7hAO3/41JIlS0gkklw7udd1lCE1qsTj4lFx/vCH/6Gtrc11HBERkddQqfaZyy67DDyPUPtu11GOKdB9EJuIsnDhQtdR5BgOHDjAQw89yCWj44wty59Z6gHXTYmSTCT42c9+5jqKiIjIa6hU+8ysWbMYMWIkobYG11GOKdy2i2AwyCWXXOI6ihzD4sWLwUtxwxlR11EyYlyZxxXje3n0kUeor693HUdEROQwlWqfCQQCvOUtlxPu3At+27PaWoo6drNgwQLKy8tdp5Gj1NbW8vTTT3P1xCgji/NvlnrAX03tpThk+eF//qfrKCIiIoepVPtQ3xKQNKGOva6jvEagtxViXVr64VP3//SnVEfg2kn5tZb6aBVhy/WTe3h51So2bNjgOo6IiAigUu1Lc+fOpbSsjFDHHtdRXiPU3pfn4osvdpxEjlZXV8eatWu5ckIPxSHXaTLvivExyotgyZLfuY4iIiICqFT7UigU4uKLLqKos9FXpyuGOxqZMXMmI0aMcB1FjrJkyRIiIcNbx/l015ghFgnCFeOiLFu2jMZGf29BKSIihUGl2qcuvfRSbKKXQE+z6ygAmGQvge6DvOnSS11HkaO0tLTw9FNPsXBMlLKwf16EZdrbx8cIGnj44YddRxEREVGp9qsLL7wQYwyhDn/MwgX713dfdNFFjpPI0VavXk0qnWbh2MKYpR5QHbGcOyLOimUvuo4iIiKiUu1XlZWVTJ8xg3DnPtdRAAh17qO8ooIZM2a4jiJH2bBhA2Vhw8TytOsoWTezOsWBg80cPHjQdRQRESlwKtU+duEFF/Qt/0gl3AaxlqKu/VywYAGBgL5k/GbDK+s5szJOwLhOkn0zq/q2ndy4caPjJCIiUujUkHxswYIFYC2hLrez1YHedmyipy+P+Ep3dzd7GvcyvSrlOooTk8rTRIKwZcsW11FERKTAqVT72Jw5cygqihDs3O80R7B/Ccr555/vNIe8XldXFwBVRfl72MuJBANQXvSXz4OIiIgrKtU+Fg6HOeecswl3H3CaI9i1n9FjxjBmzBinOeT1env7DnopCRXOrh9HKwlaotH8PJZdRERyh0q1z82fPx8TbcMkHZ2SZz2Kups4/7zz3IwvJzRQJouDhVuqI8G0SrWIiDinUu1z8+bNA/pmi10IRFuxqTjz5893Mr7IGzGA9dEhSSIiUphUqn1u5syZRCLFBLuanIw/MO65557rZHw5sYqKCgC6kwW49Ue/7lSIqqoq1zFERKTAqVT7XCgUYvbsswj3uNmHN9jdRM2oUYwaNcrJ+HJilZWVAPSkCvdbuSdlDr+4EBERcaVwn4lzyDnnnIOJtkI6md2BraUo2sy555yT3XHlpA2Uya5EYc5UexZ6EvbwiwsRERFXVKpzwNy5c8Fagt3Zna02iW5svKdvfPGlUChEdVUlbfHC/FbuSBg8CyNHjnQdRUREClxhPhPnmLPOOguAYE9zVscNdveNN3v27KyOK4MzatQoWgu0VLfGggBaniQiIs4V5jNxjikvL2f0mLEEoq1ZHTcQbSEYDDJ16tSsjiuDM3rMWFoTYdcxnGjpfzGhUi0iIq6pVOeIWTNnEO7NbqkORluZPGUK4XBhFrZcUVNTU7Az1QPLXrT8Q0REXCvMZ+IcdOaZZ0KsE1KJ7AxoLeFYKzOmT8/OeHLKhg8fTm/Skki7TpJ9HYkAoWBQFyqKiIhzKtU5Ytq0aQAEYu3ZGTCdwCZ6tfQjBwwbNgzoK5iFpiNuqK6qJBAovH+7iIj4i56JcsSYMWMACCS6szJeIN4FwNixY7Mynpy6gVLdWYilOhlg2PDhrmOIiIioVOeK0aNHA2DiPVkZb6C8D4wr/jWw5j1dgCd1exbC4SLXMURERFSqc0V5eTmlZWVZm6keKO8DM+TiX57nAWAK8PyXAOBZz3UMERERlepcMnz4cEyyNytjmWQvAV0AllMK8ZvZGLC2AKfoRUTEdwrxeThnFUeKMV52tngwNkVRUQRTiNOfOaa7u++nF5Fg4ZXLSNDS3dnpOoaIiAgh1wHk5BWXFIOXneUfeGkikUh2xpLTsnPnToIGxpQW3p5648vSrNrVRCwWo7i42HUcEZHXWrz49bctWpT9HJIVmqnOIcWRCAGbpZlqL0UkogvAckFdXR3jyj1CBfjdPLE8jbWW+vp611FERKTAFeDTcO4yfQtIszOYtRijLw+/s9ayo247E0qTrqM4MbEsBfS9sBAREXEp463JGHOVMWabMabOGHPbCe53gzHGGmMWZDpTrorFYniBYFbGsoEQ8Xg8K2PJqdu3bx+HWto4syrlOooTNSUelRF45ZVXXEcREZECl9FSbYwJAvcCVwOzgQ8aY2Yf434VwCeBlZnMk+uivTEIhLMzWDBELBbLzlhyytatWwfA7GGFOVMdMDC7Ks66Nau1C4iIiDiV6ZnqC4E6a+1Oa20C+C1w/THu93Xg3wG1uBPo7e3FZnGmOhGPqaj43Jo1a6guhnEFeJHigNnDk7S0tbN7927XUUREpIBlulSPB/Yc8XZj/22HGWPmAxOttY+e6AMZYxYZY1YbY1Y3NzcPfdIc0NPdjQ1m6eLBYBGe59Hbm519sWXwrLW8sn4tZ1XFC/LglwEDs/Tr1693nERERApZpkv1sZ7qD099mr4r4b4PfPaNPpC1drG1doG1dkFNTc0QRswN6XSazs4ObLgkK+N5/eO0trZmZTwZvObmZlrbOgp2PfWAmmKPyohh69atrqOIiEgBy3SpbgQmHvH2BGDfEW9XAHOBZ40xu4CLgaW6WPH12tvbsdZii0qzMp4N942jUu1fW7ZsAWBaZWGXamNganmcLZtrXUcREZEClulSvQqYboyZaowpAj4ALB14p7W2w1o70lo7xVo7BXgJuM5auzrDuXLOQLm1oezMVFvNVPve1q1bCQZgUnlhl2roe2HRsHsP0WjUdRQRESlQGS3V1toUcDPwZ2ALsMRaW2uMucMYc10mx843TU1NAHhZmqn2ispeM674T+2mTUwuTxPWduKcUZnCWnt49l5ERCTbMn5MubX2T8Cfjrrty8e571synSdXNTY2AuAVV2VnwFAEE44cHlf8JZlMsnXrVt46JuE6ii+cUZnCALW1tZx//vmu44iISAHSHFeO2Lt3LyZcDKFI1sZMRypVqn2qrq6ORDLJ9AK/SHFAWdgyvtxj06aNrqOIiEiBUqnOEY2NjaQjFVkdM11Uye49KtV+tHbtWgBmVBXmoS/HMqMqwcYNG3QSqIiIOKFSnSN21teTjmRp6Uc/r6SKlkPN9PT0ZHVceWNPPfG/TK9KUx3R4TwDFtQk6I3Feemll1xHERGRAqRSnQNaWlroaG8nXToiq+MOjLdjx46sjisntnPnTnbuauCS0TqY50izhyWpjsCTTz7pOoqIiBQgleocUFdXB4BXOjyr4w6MNzC++MOTTz5JwMCFo3SR4pECBi4a1ctLK5bT1dXlOo6IiBQYleocMFBq01ku1TZciikqUan2kVQqxZ8f+xNnD09QWaSlH0d705g4yVRas9UiIpJ1KtU5oK6uDoorsrrzBwDGkCwexrbtr2Z3XDmuFStW0NLWzhXjY66j+NKUijRTK9M88oc/YK1edIiISPaoVOeALVu3kSzJ7nrqAenSEeyq30kyqV0m/OCRR5YyrBjOGa7H43jeMraXnbt2sXnzZtdRRESkgKhU+1x3dzcH9u/Dy/JFigO8shGk02l27drlZHz5i4MHD7Jq1SouHxMlqO/c47p4dJzikOGPf/yj6ygiIlJA9NTsc4fXU5e5mqkeCcD27dudjC9/8fTTT2Nt37phOb6SEJw/spfnnn2GREIXc4qISHaoVPtcfX09AF5Jdi9SHGAjFZhg+HAOcefJJ/6XM6rSjC71XEfxvUtGJ+iJ9rJy5UrXUUREpECoVPvc7t27MaEibLjETQBj8Iqr2L17t5vxBYBdu3ZRt2Mnl4zSBYonY86wJJXas1pERLJIpdrnGnbvJl1cBcY4y5CKVFK/q8HZ+AJ//vOf+/em1tKPkxEMwEU1vSxf9iKdnZ2u44iISAFQqfa5Xbsasn48+dG8kmoONR8kFtMsqQupVIrHH/sT545I6FjyQVg4VntWi4hI9qhU+1gikaC15RBepMJpDi9SgbWWAwcOOM1RqF5++WXa2jtYOFYvagZjckWayRUef/qTdgEREZHMU6n2sZaWFgC8ojKnOWz/+IcOHXKaoxBZa3n4oYeojMC5I7Q39WAtHNtLXd0ONmzY4DqKiIjkOZVqH2tubgbAFpU6zeGF+8YfyCPZs3z5ctasXcu1k3oI6bt10BaOjTGsGO65+27S6bTrOCIiksf0NO1jAzPDNux6prqvVGumOrvi8Tj/cc/djC/3eLuOJT8lkSB88IwuXq2r47HHHnMdR0RE8phKtY+1t7cDYMPFboMEQphQhLa2Nrc5CsySJUvYf6CJvz2zW7PUp+GiUQlmVqf48eL76Orqch1HRETylJ6qfay7uxsAG4w4TgKEiujp6XGdomDU19fzy1/+ggtqEswdrrXUp8MY+Lvp3XR1dXH33Xe7jiMiInlKpdrHuru7McEwBNw/TF6w6HDJl8xKJpN84xv/Rkkgxd/P1Od8KEyqSHP9lChPPPEEzz77rOs4IiKSh9y3NTmu7u5uCBW5jgFAOhDWj86z5Fe/+hV1dTv4hxldVBZpX+qh8u7JvUytTPO9737n8M46IiIiQ0Wl2sd6e3uxgbDrGH0CIaK9va5T5L3GxkYeeOAB3jwmxoKahOs4eSUUgEVnddEb7ea+++5zHUdEhsrixa/9JeKISrWPeZ7n9Hjy1zAGL+25TpH3Vq5cied5/NVU/7+AeWB7KQ1dQRq6gty5tpIHtrvd+vFkjC9Lc3FNjBXLl/V9f4mIiAyRkOsAcnx9T/r+KNUWQ1olJOPWrFnDqFJLTYn/P9e7u0P0pvtel29tz53X57OHJ3nhQA91dXXMmDHDdRwR8YujZ7kXLXKTQ3JW7jwTFiDP87C+makO4Hk6PCOTUqkU69etZXZ13HWUvDZ7WN9uKmvXrnWcRERE8olKtY/1nQDnk1KNIa3lHxkVi8WIxeJEgro4MZNCAQgGoLW11XUUERHJIyrVPuZ5HtYvpdoYrUHNsPLyci5buJAXDpTQm3KdJn89s7eYtAdXXXWV6ygiIpJHVKp9zHcXKmr5R8a9//3vJ5qEF/Y7PkUzTyU9eHJvKRdcsIBp06a5jiMiInlEpdrHfDVTjcHztCwh02bPns2cObP5895S9OkeeiubimiPw/ve937XUUREJM+oVPtYa1s7NuiPw19sKEJnR4eWgGTBX//1DTRHDbWtPtmjPI88s6+UiRPGs2DBAtdRREQkz6hU+1Q6nWZv4x68kirXUQDwiqtIJOIcOnTIdZS89+Y3v5nKinKe3RdxHSWvNHYHebUjyLXvvg7jl2VVIiKSN1SqfergwYMkk0m8Yv+UaoDdu3c7TpL/ioqKuOrqd7G2JUJnQuVvqDy3P0IoGOSd73yn6ygiIpKHVKp9qqGhAcA/pbp/xnwgl2RWWVkZaQ/2R4Ouo+SNvT37VyICAAAgAElEQVQhysr8f+qjiIjkJpVqn1q9ejUEgqRLh7uOAoANlUCknFWrVrmOkvdWr17NL37+cy6oiTOjSnvrDZX/M62HaHcXX/3KV0il9Hk9XcaYamPMQ8aYrcaYLcaYS1xnEhFxSaXah9LpNE8+9TTJyvHgkwsVMYZE9RRWrVpFZ2en6zR5a//+/Xztq19hXFmafz6r2zc7KuaDqZVp/nFmF+tfeYX77rvPdZx88APgcWvtLOBcYIvjPCIiTqlU+9DGjRtpb2slNdxf++gmR0wjnU7zwgsvuI6Sl7Zv385tt/4r6XgPn5rbQXHIdaL88+axCa6c0MuDDz7I/fffTzyuI+FPhTGmElgI/BTAWpuw1ra7TSUi4pZKtQ89/fTTmGCYVPVE11FewysdASVVPPnUU66j5JWuri6+//3v89GPLqK9aQ+fnNPB6FJtXZgpHzgzyqWj4/zyl7/kn/7xH1i5cqXrSLloGtAM/MwYs84Y8xNjTJnrUCIiLqlU+0w6nebZ554nUTkegj7bp9gY4tWTeWX9ejo6OlynyXme5/HYY4/xd3/7Nyxd+gfeMb6Xb13YyuzhWu+bSaEA3Dinm3+d14nXsY9bb72VL33xizQ1NbmOlktCwHnAD62184Ee4LYj72CMWWSMWW2MWd3c3Owio4hIVqlU+8yWLVvo7GgnNWyy6yjHlKqejOd5vPTSS66j5LSdO3fyyU/czF133UWNaeeOBe18aEaUsrCOUcyWucOT/NsFrfyfaT28vOJF/u/ffYhf//rXuojx5DQCjdbagWn+h+gr2YdZaxdbaxdYaxfU1NRkPaCISLapVPvMsmXLwARIVU1wHeWYvLKRmEgZL774ousoOWv//v184uaPs7tuM/98VjdfOK+dyRVp17EKUjgA754S45sXtXF2VQ8//vGP+eEPf+g6lu9Zaw8Ae4wxM/tvehuw2WEkERHnVKp95vkXXiBdMQZCPj1NzxjilRNZ+fLLusjrFKRSKb7+9Tvwkr18+bw2LhsbJ6AdPpwbWezxybO7uHJCLw8//DDLly93HSkXfAL4tTFmAzAPuNNxHhERp1SqfcRay769e0mVjXQd5YTS5TUk4nFaWlpcR8k5P//5z9m8eQv/OKOLUSW6GNFv3n9mlMkVHt/65p1oHfCJWWvX9y/vOMda+x5rbZvrTCIiLqlU+0g6ncZaCwGf76Vm+r5sksmk4yC5pba2ll//+gEWjo1x8eiE6zhyDOEAfGx2B/FoN9/+97tcxxERkRyiUu0jhy+QMj5/WEzf0dm6oGtwrLUYDJ3JAJ6uR/StzmSAtAWrx0hERAbB5+2tsAzM/Fqfl2ob0Ez1qZg7dy6f+pd/Yf2hIh7YXqbS5kP7owF+sKmKsePG86Uvf9l1HBERySH+bm8FJhKJEAgEMKmY6ygnZJJ9+crKdNbDYF1//fW8733v48m9xfx5T7HrOHKEzoThuxuqCRZX8K27/p3KykrXkUREJIeoVPtIUVEREyZOJBj19/U+wd42wkVFjBs3znWUnHTjjTeycOFCflNXxrIDRa7jCNCbMnxvQxXtySLu/Oa3GD9+vOtIIiKSY1SqfWb6mWcSjre7jnFCgd5Wpk6dSjAYdB0lJwUCAb7whS8wb965/HhLBaubVaxdiqfhexsqaegJ89WvfY05c+a4jiQiIjlIpdpnpk2bho11Qcq/u0OEY+2cecYZrmPktEgkwjfu/CYzZ87kP2sr2NjisyPpC0TSg7s3VrK9I8TnP/8FLr30UteRREQkR6lU+8zAj50DiW7HSY7DS2ETvVr6MQRKS0v5929/h8lTpvKDTZXs7NTMf7Yt3lzOxtYwt9zyOd72tre5jiMiIjlMpdpnBi6OMil/nlY4kKuqqspxkvxQUVHBJz/1LyTSsKfb5/uT56G6rgjz5s3jmmuucR1FRERynEq1z+RKqdbOCEPn5ZdfJmDgvJH+XfKTrxaM7KV200a6u336kyEREckZKtU+c7hUp1WqC4G1lqefepLZw5JUFGnj6my7aFSCZCrNsmXLXEcREZEcp1LtM/F4X2n17QEw/bliMX/vpZ0rHn74YfYfaOKiUf58EZXvzqhMUVNi+fUDv6K5udl1HBERyWE+bW6Fa/fu3QB4xf5cs5wu6cu1Z88ex0lym7WW++67j//4j//gvJEJLh2jUu2CMfDhmZ0c3N/Ixz92Ew0NDa4jiYhIjlKp9pm/lOpqx0mOI1SMKSo5nFMGL5lMcuedd/Kb3/yGK8bH+OTZXYT1nejM7OEpPj+/nXjnIW7++MfYuHGj60giIpKD9FTuMw0NDZhIGYT8eyBIKlLJrl2a0TsV8Xic22+/jSeeeIIbpkX5+xk9BIzrVDKlIs2Xz2ujzHbz2c98hhUrVriOJCIiOUal2mc21W4m6ddZ6n7pkuFs376dZDLpOkrO+dGPfsTq1Wv48KxurpvSi1Gh9o2aEo8vzW9jfEmMr9/xNfbt2+c6koiI5BCVah9paWmhcc9u0hVjXUc5oXTFWBKJOFu3bnUdJacsX76c3//+97xzYi+Xj9Maaj+qKLJ8Yk4npGLcccfXSKVSriOJiEiOUKn2kVdeeQWAlO9L9RgA1q9f7zhJ7jh06BDf+uadTK7weN8ZUddx5ARGlnj808wutm7dxs9+9jPXcUREJEeoVPvIunXrMKEivLIRrqOckA0XY0uHs3bdOtdRcsYPf/hD4tEebprdoYsSc8CFoxJcPjbGf/3Xr7UjiIiInBQ9vfvIho0bSZbWHN4L2s+S5aPZXLuZdDrtOkpOqN9Rx+xhccaVea6jyEl6x4QY1sKuXbtcRxERkRzg//ZWIKLRKLsbGkiX17iOclLSZTXE4zEVjpPU2tpKdZEKdS6pivQ9Xq2trY6TiIhILsh4qTbGXGWM2WaMqTPG3HaM999ojNlojFlvjHnRGDM705n8aNu2bVhrSZflSKnuL/9btmxxnMT/UqkUHV3dVKlU55SKsMUYlWoRETk5GS3VxpggcC9wNTAb+OAxSvN/WWvPttbOA/4d+F4mM/nV5s2bAXJmptpGKjHh4sO55fh6enqw1lIWtq6jyCAEDJSFDV1dXa6jiIhIDsj0TPWFQJ21dqe1NgH8Frj+yDtYazuPeLMMKMjmUVdXB8WVECp2HeXkGEOyZATbtr/qOknO0JbUIiIi+SuU4Y8/HthzxNuNwEVH38kY83HgM0ARcMWxPpAxZhGwCGDSpElDHtS17a/W+f7Ql6N5JcNoaNhGKpUiFMr0l5KIiIiIf2W6CR1rcu51M9HW2nuBe40xfwN8Efj7Y9xnMbAYYMGCBXk1mx2Px9m3txFv7LmuowxKunQYqaYke/fuZfLkya7j+Ja1efXlWnA8T2vhRTJm8eLXvr1okZscIkMg08s/GoGJR7w9ATjR2b+/Bd6T0UQ+1NDQgLUWr2SY6yiD4pUMB2Dnzp2Ok/hbXV0dAKNLtf1grhldkmJHnZY4iYjIG8t0qV4FTDfGTDXGFAEfAJYeeQdjzPQj3rwGKLhnsIFSmnulugqMUal+A2vXriVoYGZV0nUUGaTZ1XG2bN1GNKpTMEVE5MQyWqqttSngZuDPwBZgibW21hhzhzHmuv673WyMqTXGrKdvXfXrln7ku507d2ICIbziStdRBicQgpJqduzY4TqJr61du4ZpVSmKtew858welsTzPDZs2OA6ioiI+FzGn+attX8C/nTUbV8+4s+fynQGv6vbsYN0SXVOnKR4tGRxNa/WqVQfz8GDB9m+bTvXT0m4jiKnYHpVikgQXnjhBS6++GLXcURExMdyr8Xlobq6OlI5tvRjgFcynOaDTXR3d7uO4kuPPvoo1lrePCbuOoqcgqIgXDwqxpNPPqGvcREROSGVasdaW1vp7Og4fNFfrkmX9uWur693nMR/UqkUf3xkKeeMSFJToh0kctVbx8eIxxM88cQTrqOIiIiPqVQ7dvgixdLcLNUDF1fqYsXXW7FiBS1t7Vwxvtd1FDkN0yrTTK1Ms/QP/+M6ioiI+JhKtWMDF/nl2s4fA2xRGSYU0cWKx/DMM89QGYFzR+Tnrh+9KUNxcTE33HADxcXF9Kby98zIy8b0Ur+rgYaGBtdRRETEpwZdqo0xZZkIUqh27tyJiZRhwzlyPPnRjCFVUk2dSvVrJJNJXlqxnHnDYwTytGtGU4Zrr72Wm2++mWuuuYZoHpfq+SP7XhgtW7bMcRIREfGrky7VxphLjTGb6dsaD2PMucaY/8xYsgJRv2sXyUiObaV3lHTxMBoadruO4SuvvPIK0d4Y543M310/SkOWRx99lHvuuYc//vGPlIby9+TIEcUeUyvTvPjiC66jiIiITw1mpvr7wDuBFgBr7SvAwkyEKhTWWvbs2YNXXOU6ymnxiivp6e6io6PDdRTfWLVqFeEAzBmen0s/AEpCllgsxsMPP0wsFqMkj0s1wLwRcbZs2apdQERE5JgGtfzDWrvnqJt07vJpaG9vpzcazYNS3Ze/sbHRcRL/2LVrF2PLPCJB10lkqEwuT2GtZfdu/VRGRERebzCleo8x5lLAGmOKjDG30L8URE7Nnj19r1HypVSrbPxFQ/1Oxpbk7yx1IRpb1jeHoIsVRUTkWAZTqm8EPg6MBxqBef1vyyk6cOAAAF5RheMkp8cWlQN/+fcUung8TlPzIcaV6Qc5+WRUsUcwoBePIiJybCd1TLkxJgj8nbX2bzOcp6A0NzcDYItKHSc5TYEAJlJ2+N9T6A4dOoS1lpHFOvAlnwQDMLwYfZ2LiMgxndRMtbU2DVyf4SwFp7m5GROOQDDsOsppS4dKOXjwoOsYvtDV1QVAeVilOt+Uh9KHH18REZEjndRMdb9lxpj/AH4H9AzcaK1dO+SpCkRzczNeUea2/Y7sfolgtAWAkq1/wisdTnzSxRkZK11UyoEmlWr4S6kuy/PdMApRWTBNV2en6xgiIuJDgynVl/b/fscRt1ngiqGLU1gONjeTDpVk7OMHoq2YdN/FcqGuA6QyNhLYcCmtLVprCtDT0/eaM9+3mCtEJSFLU7dmqkVE5PVOulRba9+aySCF6NChFmx4uOsYQ8KGS4lGe4jH40QiEddxnCop6XuhFEvn7wmDhSqWNpSW6VBZERF5vcHMVGOMuQaYAxw+U9tae8fx/4Ycj+d5dHS0440a7zrKkPDCfUWytbWVsWPHOk7jVmVl3wmZ3UmV6nzTnQpSU5nbW2CKiEhmnHSpNsb8CCgF3gr8BLgBeDlDufJeZ2cnXjqNDef4zh/9bH+pbmlpUak+XKoHdbaS5IDuVJAz+h9fEfGpxYtff9uiRdnPIQVnMM/6l1pr/y/QZq39GnAJMDEzsfLfwJ7ONoMXKmaTjfTtVd3U1OQ4iXsjRowgYAzNvSrV+STlQWsMampqXEcREREfGsyzfm//71FjzDggCUwd+kiFYeBIb684P2a9vEjfATY6qhyKi4uZMGEcu7sHtbpKfG5fT5C0B2eeeabrKCIi4kODKdWPGmOqgW8Da4FdwG8zEaoQHC7Vkdw+TfGwQAhTXK5S3e/M6TPZ3VPkOoYMoYb+F0kq1SIiciwnXaqttV+31rZbax8GJgOzrLVfyly0/NbY2IgproBA/sxmJosq2LNHpRpg2rRpHOqF3kzuYyhZ1dgTJBwOMWHCBNdRRETEhwa7+8elwJSBv2eMwVr7ywzkyntbtm4lGal2HWNIeSXD2bHjVZLJJOFw7p8SeTqCwSAAAW0AkjcMYDCHH1sREZEjnfRMtTHmV8B3gDcDF/T/WpChXHmtvb2dvY2NpCtGu44ypNLlo0kmE2zfvt11FOdisRgAYV2rmDciQUsimSSdTruOIiIiPjSYmeoFwGxrrY6JO00bN24E+kpoPhl4kbBx40bmzJnjOI1bvb29REJGM9V5pDjY919fLBajTAfAiIjIUQYzj7YJGJOpIIVk48aNEAiSLhvpOsqQsuESKKliw4YNrqM4t3XLZkZEPNcxZAiNKO57PLdu3eo4iYiI+NEbzlQbYx4BLFABbDbGvAzEB95vrb0uc/Hy0+o1a0iX1UAg/9ZmJspGs/6VV0in0wW79rS+vp4NGzfx/jOirqPIEJo3IkF5ESxd+gfOP/9813FERMRnTmb5x3cynqKAtLe3s3PHDlLjz3MdJSPSleOIHtrO9u3bOeuss1zHcWLp0qWEArBwbPyN7yw5oygIl43p5X9ffJGWlhZGjBjhOpKIiPjIGy7/sNY+Z619DlgNvND/5/1AFbA8w/nyzpo1awBIVY5znCQz0pV9R5SvXr3acRI3kskkf378cS6siVNRpMsP8s0V42Kk0x6PP/646ygiIuIzg1lT/TxQbIwZDzwF/CPw80yEymdr1qzBhCJ4ebaeeoANl2DLRhRsqd62bRvR3l7Or0m4jiIZMLrUY2KFx9q1a1xHERERnxlMqTbW2ijwXuAea+1fAYW9xcMp2Lx5C8mykWDyd6+1ZNlotm7dhucV3oV6AxdpzqxOOk4imTKzMkHtplpSKZ3sIyIifzGoUm2MuQT4W+CP/bcV5pVopyiRSLB7z27Spfm9FjNdOpx4PMa+fftcR8m6jRs3MLbMUqmlH3lrRnWSWDxOXV2d6ygiIuIjg9mn+lPA7cDvrbW1xphpwDOZiZWfGhoa8NJpvNLhrqNklNf/oqGurq7gjnRubm6mMqzDQfJZdVHfT2Cam5uZNWuW4zQi4tTixa99e9EiNznEF056ptpa+7y19jpr7V39b++01n5y4P3GmHsyETCfDMxs5ftMtVdSDSbAq6++6jpK1l1wwYXUdYToSerUl3y1rqWIYDDAvHnzXEcREREfGcqFvW8awo+Vl/bu3QvGYCMVrqNkViAIxRUFufzjzW9+M2kL61vCrqNIBlgLaw4VM3/+fCoq8vz7WEREBiV/r5bzoaamJkykPK8vUhyQCpdy4MAB1zGy7qyzzmLEsGpWHYy4jiIZsKcnyMGo4bLLFrqOIiIiPpP/7c5HmpqaSIVLXcfICltUzoGmJtcxsi4QCPCOd17FupYimnv17ZVvnmwspigc5vLLL3cdRUREfGYon/W1iPQN7D/QhBcudx0jK7yiMtrb2kgmC29rufe+970YE+CJxmLXUWQIdSYMyw4U886rrqK6utp1HBER8ZmhLNU/GMKPlXfi8TiHmg/iFRfGOkwvUom1tiDXVY8aNYq3vOUtPHeghF5tZZw3nt5bTNKDG264wXUUERHxoTfcUs8Y8whw3E13rbXX9f/+86GLlX92796NtRavZJjrKFkx8O+sr69n8uTJjtNk33ve8x6efvppNrQUcdFona6YD1Y2912gWIhfzyIi8sZOZqb6O8B3gXqgF/hx/69uYFPmouWX+vp6oH+7uQLglVSBMYf/3YVm+vTpADT16nykfOBZOBgNMGPGDNdRRETEp95wptpa+xyAMebr1tojL3l/xBjzfMaS5Zn6+nowAbxIleso2REIQXFlwZbqkpIShg+rpikacx1FhkBbPEDSg/Hjx7uOIiIiPjWYNdU1/acoAmCMmQrUDH2k/LRu/Xq8spEQKJwdIZIlI1i3/hU8z3MdxYkJEybS1DuYQ0vFr5r6d3IZN26c4yQiIuJXg2l4nwaeNcY8a4x5lr4jyj+VkVR5prOzk21bt5KsLKwn5FTVBLo6Ow6fJFloKior6fUK50VUPutN9W1uVFVVID9pEhGRQTupaTRjTADoBKYDs/pv3mqtjWcqWD5Zs2YN1lpSVRNcR8mqdFXfj8pffvnlglyLGggE0E6T+cH2P47G6PEUEZFjO6lpNGutB3zXWhu31r7S/0uF+iS9/PLLmFCkb/lHAbHhEmzZCF5audJ1FCeMMdjj7psjuWTgcQwU0PItEREZnME8Q/yvMeavjaZqBiWdTvPisuUkKscVxPHkR0tUTqB20yba29tdR8m6YDBI0urbJR8k+y8LCAa1m4uIiBzbYFreZ4AHgYQxptMY02WM6cxQrryxadMmujo7SA2b4jqKE6lhU7DWsnz5ctdRsm7kyJG0xdBsdR5oi/f9VzlixAjHSURExK9OulRbayustQFrbdhaW9n/dmUmw+WD559/HhMIFtx66gFe6XAoruC55wtv98XRo0eTSENXUrPVua4lFqSstITy8nLXUURExKcGtd+XMeY6YGCv6mettY8OfaT8Ya3lueef79v1Ixh2HccNY0hUTWL16tVEo1FKS0tdJ8qa0aNHA3AoFqCyKO04jZyOQ7HA4cdTRETkWE56ptoY8y36ttDb3P/rU/23yXE0NDRwqLmZVPUk11GcSlVPIp1KsW7dOtdRsmrgOOs93dqrOtftiRYxecpU1zFERMTHBrOm+l3AO6y191tr7weu6r9NjmPVqlUApCoL+xS2dPkoTDB8+PNRKCZOnEhlRTnbO1Sqc1lLLEBLL8ydO9d1FBER8bHBbkdRfcSfdQrCG3h51SooqcZGCnwdZiBIsnwMK1e+7DpJVhljmHv2ObzaGXEdRU7Dq/0vilSqRUTkRAZTqu8E1hpjfm6M+QWwpv82OYZEIsH69etJVBTWKYrHk6oax/79+9i/f7/rKFk1d+5cDvQY2uO6WDFXbW0PUxyJcMYZZ7iOIiIiPjaYUn0NcD99Zfq/gUustb/NSKo8sH37dpKJBOnKMa6j+EK6/8XFhg0bHCfJrosvvhiAtYeKHCeRU+FZWNdSzAUXXkgopGU8IiJyfIMp1T/r//064HvAvcaYTw19pPywadMmANLl2jEAwCupxoSKqK2tdR0lq6ZOncq4sWNYe0hLQHJRfVeIthhcdtllrqOIiIjPDWaf6qeBbwBfAn4CLABuylCunFdbWwvFldhwieso/mAMydIaNmzc5DpJVhljuGzh5dS2hYmmtAQk16xpLiIQCHDJJZe4jiIiIj43mC31ngKWAe8HtgEXWGtnZSpYLrPWsmHjJpJlNa6j+Eq6fBQNu+rp6elxHSWrLr/8ctIevHxQS0ByiWdhxcESzjtvPhUVFa7j+I4xJmiMWWeM0XkFIiIMbvnHBiABzAXOAeYaYzQNewz79++no72NdPko11F8JV0+CmstW7ZscR0lq8466ywmT5zA8/v17ZJLNreFaemFd73rGtdR/OpTQGF9M4uInMBgln982lq7EPgroIW+NdbtmQqWy7Se+tjS5TVgDBs3bnQdJauMMVx9zbXUdQTZ2xN0HWfITCpPURL0KAl6zKpOMqk85TrSkHp+X4SK8jLe9KY3uY7iO8aYCfRdvP4T11lERPxiMMs/bjbG/A5YD7yHvp1Ars5UsFy2adMmTKgIr6T6je9cSIJF2NLhbCywddUAV155JcFggOf3588Fix+aEWVyRZrJFWk+f14nH5oRdR1pyHQnDWsORXj7O64kEsmfx2wI/T/gXwHveHcwxiwyxqw2xqxubm7OXjIREUcGs/yjhL5dP2ZZa99mrf1a/8WLcpQNGzaSLK0BM9izdfJfsqyG2s21pFL5Nav5RoYPH87FF1/C8qYSUsetIeIXLzUVkfTgXe/SobFHM8ZcCxy01q450f2stYuttQustQtqanR9iYjkv5PeeNVa++1MBskXbW1t7NpVT3r8ea6j+FK6Ygzxg1vZtm0bc+bMcR0nq66++mqWLVvGxtYw80cmXceRE3jhQAlnTJvK9OnTXUfxozcB1xlj3gUUA5XGmAestR9ynEv8ZvHi1769aJGbHCJZoqnUIbZ27VoAUpXjHSfxp4FDYNasOeEkV166+OKLqa6qzKslIPloT3eQ+s4gV+sCxWOy1t5urZ1grZ0CfAB4WoVaRESlesitWbMGE4rglY1wHcWXbLgYWzaSVatXu46SdaFQiCvfeRXrD0XoSGjPar96bl+EUDDI29/+dtdRREQkh2S8VBtjrjLGbDPG1BljbjvG+z9jjNlsjNlgjHnKGDM505kyxVrLypdXkagYo/XUJ5CoGMvm2lqi0fy5sO1kXXPNNaQtvKDZal9KpGFZUwmXLVxIdbUuNH4j1tpnrbXXus4hBWz5cvj2t+G558Ba12mkwJ30mupTYYwJAvcC7wAagVXGmKXW2s1H3G0dsMBaGzXG3AT8O30HzOSc+vp6Wg41k558qesovpaumkD6wEbWrFlTcMc/T548mXPOOZtn6zbwrkkxApqw9pVVzUX0JOHd73636ygi8ka+8hX4t3+DYBDq6mD3bvjgByGU0WqTeUevRQetR88RmZ5OvRCos9butNYmgN8C1x95B2vtM9bagSnLl4AJGc6UMStWrAAgVT3JcRJ/S5ePxoQihz9fhea6667nYNSwsTXsOoocwVp4cm8p48eNZf78+a7jiMiJLF0Kd9wBf/M3fTPVV18NL74IDz3kOpkUsEyX6vHAniPebuy/7Xg+DDx2rHfkwp6ny5YtxysbiS0qdR3F3wIBEpXjWLZ8OZ5XePvLLVy4kDGjR/HQznI8/bTSN9YeCrOjI8j7P/BBjNGPEER8KxqFT34S5syB+++HkhJ4z3vgiivgmWegwA4YE//IdKk+1jPTMWuEMeZDwALgmFv3+X3P0/b2drZs2UyyaqLrKDkhVT2JjvZ2tm7d6jpK1hUVFfHhj/wzDV0BXmoqch1HgLQHD+6sYML4cdqbWsTvvvlNaGiAe++F8BE/8Xvve2H8ePjFL6Cp6bV/p6cHHnsMvv99+PKX+8r4+vVahy1DKtMLjxqBI1vmBGDf0Xcyxrwd+AJwubU2nuFMGbFs2TKstaSGaenHyUhVTQAT4MUXX2T27Nmu42Td2972Nn77m//i4fqdXDCqlbCua3XqhQMR9vUY7rj1RkK5vh5TJJ8cvb5450747nfhQx+Cyy9/7fvCYfjIR+A734E774QbboBIBHbsgM99Djo7++5nzF/K9Pz5cNtt8L73Zf7fInkv00/lq4Dpxpipxpgi+vY0XXrkHYwx84H7gOustQcznCdjnnvuOSiuxCsZ7tTKz74AACAASURBVDpKbghFSFWM5Zlnn8MW4ExBIBDgpo99nOZewyO7SlzHKWidCcODO8uZPfusgrtwViSn9PTAj38MEyfCPfcc+z7jxsEXv9g3Y/3AA/DTn/attX73u+GJJ6C5GeLxvgsb//M/IZmE978fbr8dCnA5ogytjE7JWGtTxpibgT8DQeB+a22tMeYO4P+zd99hUZ5Z48e/zzSGKk0sFKVY0Nhb7N2NMTHN9GZJTCwpm8S0zW/3fbO778bN1pTNrptsNr1XNfYS7DX2BoggShWQDlOe3x+3xIYywMw8U+7PdXERYZg5AR3OnOfc5+xUVfV7RLtHCPDFuT7GHFVVp7oyLmerrKxk165d1LdNFa+AJYdYIzqRl72ZrKwskpKStA7H7QYOHMjEiRNZvHoVA9rW0ynUpnVIfun9Y8HU2vUsWPCs7KWWmk0OanADVYX0dPjmGzh7FlasgKuNvIyMhKefFolzSAi0bQvz5198m+RkmDMHHn4YHnsMXnkFhg+HBx5w7f+L5NNcfp1TVdUfgB8u+divL/hvr9+wsGXLFmw2G5aIzlqH4lWsEQmQvZkff/zRL5NqgMcee4ydO7bz7yN2/mdAKQbZBuJW2wtNbC8MYNas6SQmJmodjiRJDfbtEz3Ty5aJRLq8HIKD4cEHYeDApr9er4du3Zq+ncEgKtbh4SKxTk2FQYNaH7/kl+SvcCdYs2YNBIRgD/a8A5SeTDUGYQttz5q1a/2yBQQgLCyMp55+hpwKHYuzZRuIO1XUK7yfHkqXlBTuvvturcORJAnAaoUZM6BPH/jgA2jTBnr1Ej3Ur7wCQ4Y4/zEVBX77W0hMhI8/htJS5z+G5BfkiZxWKi8vZ8eOHbL1o4UskUnkZm8mIyODLl26aB2OJkaOHMn48eP5fu0a+kXX01m2gbjFe8eCqbbqef6FF+ThRMmr+GzLicUiDgx+8w089xw8+6z75k4bDCKZ/93vRC/2pe0ikuQAWalupbS0NNH6Eemf7QutZYnsDIqOtWvXah2Kpp544gnCw8P59+EwLPKsjMttLRBtH9NnzCA5OVnrcCRJUlV4/32RUL/2mqhKR7r54H+7djB1Khw4INpPJKmZZFLdSqtXr4bANtiDorQOxTsZzFjDOrJq9Wq/XATTICwsjGeefY6TlTq+zZJtIK5UVifaPrp378Zdd92ldTiSJAFs2ADbt4s2jMce0y6OceOgQwf4/HOordUuDskryaS6FQoLC9m7dy91EUmy9aMVLFHJFBcVsd/Pt2ANHTqUyZMnsyQniMxy2Y7gCqoK7x4NoU418MILL8q2D0nyBCdPwmefQY8e8OKL2sai14sRe8XFYsmMJDWDTKpbYc2aNaiqiiVKXj5uDWt4AoreyKpVq7QORXPz5s0jOiqKfx8Oo162Vjvd5gITPxWbePjh2XTq1EnrcCRJstnEBsSgIJg5E3QekJakpooDkb/7Haxfr3U0khfxgL+93mvFipXYQ2JQzWFah+Ld9EbqwxNYs3YtdXVeuVDTaUJCQnj2uec5XaXwdVaQ1uH4lNI6hQ/SQ7mmZw9uu+02rcORJAlgzRpRqb77bggN1Tqa8+65B7p0EXHl52sdjeQlZFLdQpmZmZw4kUW9PKDoFJaoZGqqq9m6davWoWhu0KBBTJkyhWU5gaSfle0JzqCq8J8jIdgUI889/wJ6vV7rkCRJysqC778X4/P69dM6mouZzfDFF2JG9tChsGmT1hFJXkAm1S20evVqUBSskXJhhDPYwjqimALFzG+JuXPn0rZtNO8ckW0gzrC5wMTeMyYeeng28fHxWocjSQ5btOjyN5/x9NOi3eOuuzzzXFKvXrB2rYht1Cj4zW/EHG1JugKZVLeAqqqsXrMGa1hHVKOc1OAUio668M5s2bKFqqoqraPRXHBwMM8seJbTVQrfnZB/x1rjbL3CRxmh9OiRyq233qp1OJIkgehV/uYbuO4694/Oa45rr4U9e8TymZdfhpEjIS9P66gkDyWT6hY4dOgQRYWFcja1k1kjk7BYLGySl9kAGDx4MNdddx1Lc4I4USHbFVrq/aPB1KkGnnvuedn2IUmNcHs13GaDp56ChASYMMHFD+YEYWHiMOUnn4gE+7nntI5I8lAyqW6BtWvXgk6PNVxOD3AmW0gMBISI1hoJENNAwsPDefdoKHb/3OTeKj8VG9lRFMD06TPktA9J8hTvvQc//QQLF4LJpHU0jrvrLjFD+8MP4dAhraORPJBMqpvJbrezbv16LGGxYPCiJwNvoCjUh3dm165dVFRUaB2NRwgNDWXuvPlklev58XSA1uF4lXobfJQRSkJ8HHfeeafW4UiSBFBRAb/6lTj8543/Lp99FoKDRX+1JF1CJtXNdOTIEUrOnMEa0VnrUHySJbIzNptNTgG5wPjx4+nd6xq+yAqh0uKBh3k81PKTgRRWKzz+xJNyyYskeYpXXhEj6v76V888nNiU6Gj45S/hyy/lKnPpMjKpbqYNGzaAosMaLicIuII9uC2KKYi0tDStQ/EYiiISw2qLwtfH5exqR5yp1fF9dhCjRo1k4MCBWocjSRJAdjb8+c9w771iuYq3evJJMbXkyy+1jkTyMDKpbgZVVVm//kesoR3AIC/Fu4SiUBeewNZt26itrdU6Go+RkpLC1JtuYu1pM6eq5GG7pnyRGQQ6I3PnztM6FEmSGjz3nEhGvX39d2SkeFGwYoVjty8uhhMnXBqS5BlkUt0M2dnZ5OWdxhqRoHUoPs0a3glLfT27du3SOhSPMn36dMzmQD7LlNXqqzlermdzQQC333EH7du31zocSdKMR824zsyEzz6DBQvAF2bFT5oEO3ZAUyNgMzLg978X7S5yxrXPk0l1M2zevBkAa7hMql3JFtoexWD6+fstCeHh4dx3//3sKTZxqET2CDdGVeHTjBDCw0K55557tA5HktzCbodTp2D/fvG+rk7riC5ht8Pnn0PHjuKgny/4xS/EE87hw1e+zYYN8Le/if//2lpZrfYDMqluho0bN2EPjkY1BWsdim/T6akPi2Xjpk3Y7Xato/Eot912G+1i2vJxphyx15idRSaOlBmYMeshgoPlv1PJd506BYsXi06K+fMhLg569xbvg4Jg2DBYuhRKSrSOFDE+78QJ+L//E5MzfMGgQdCmzdVH6732GgQGwgsviEOZV0vAJZ8gk2oHlZWVcfjwISxtfOCylRewhidwtqyMo0ePah2KRwkICGDO3HnkVOhYe0r29V+ozgafZIaQ1LkzU6ZM0TocSXKJgwdFJ8HLL4ukWacT+1PeeksUg//1L3j+edFpsHixmF733nuirVcTdrsItH17sZXQVxgM4ht/6JCoWF/Kbhcrznv2FP/v8fFw5Ij745TcSl5DdtDWrVtRVVVO/XATa5s4UBQ2b95Mamqq1uF4lNGjR9Ovb1++OriHITH1hJpkyRpgaXYgxTUKLz0pR+hJvufMGdGSvHcvhIfDLbfA8OEQGio+P3v2xbf//e/F9LqVK2HjRti5U3zNQw+JRNxt9uwRZfWZM8HXNppOmgRffSVGBHbocPHn9uwRlwm6dxd/7t4d1qzxwN4cyZlkpdpBW7ZsQQkIxh4UpXUo/sEQgC2kHZtkX/VlxIi9J6ix6fhSjtgDoKhGx9KTQYwdO5a+fftqHY4kOdXOnfA//yO6B269VSTM1113PqG+kshIsQTw5Zeha1eRlI8f78aqtaqKKnW7dqJdwtdMmiTeHzx4+ecaNgM3FIW6dxfr2dPT3RObpAlZznGAxWJh2/bt1IUleNewels9ZrOZG264gSVLllBpq9c6omaxtInneOYOCgsLiYmJ0Tocj5KYmMitt97KV19+yYgOdXRp47+nylUV3j8Wgt4QwJw5c7QOR5Kcxm6Hb7+FZcsgORlmzYKoK9R1rjbZIzJS9F1v3iwS68GD4fvv4ZprXBP3zw4fhtxcmD7dzeVxN+ncWbxgOHRItIJcaM0a0frRpo34c0qKqNTLFhCf5oN/y51v79691NbUeF3rh2Kt54YbbmD+/PlMmTIFxepdSbXt3Pd7y5YtGkfimWbOnEl0dBTvHg3F6sfnObcXmth7xsishx6SL74kn2G3wyOPiIR6+HCxxO9KCbUjFEXcz48/ikEUw4fD1RbXOmUUX1oahISALy9g6tEDjh0Di+X8x+rqxOSP8ePPfywgAJKSxG0lnyUr1Q7YsmUL6PTYQjs0fWMPohpMLFmyBFVVWbp0KarBu1oF7OY2YA5j85Yt3HTTTVqH43GCgoJ44slf8tJLL7Esx8yNnf1vWU6VReGjjFC6pKRwyy23aB2OJDmFqsLcufD223D99TB1qvMukg4ZAtu2wdixonth5Urn3C9wcfZdWioawCdMAKPRiQ/iYXr2hHXrxDzqhlaPLVugpkb8v+flnb9tQoJ4oWGzeXd/+aWvsi5t6PdjslLdBFVV2bhpk9iiqPeyJwa9idraWr766iuxnVBv0jqi5lEU6sPi2L1rl9yueAUjRoxg5IgRfJsdTEG1//1z/jwziPJ6hQXPPisPJ0o+45lnxBSPF15wbkLdID4e1q8XnQuTJsHJk869fwA2bRLl9lGjWJTWXbx5whIaZ+vSRSTIF47W++YbMJlg9OiLb9uxo6hoZ2e7N0bJbfzvt3Az5eTkUJCfLxe+aMQaHo/FYmH37t1ah+KxHn/iCYwmM/89FtLoZCdfdbTMwLrTZm6bNo2uXbtqHY4kOcVf/iLeHntMHEh01TGeuDhRYG3TBt54QxSWncZuFyNHevSAtm2deMceyGwW/dINSXVNDbz/Ptx2G4SFXXzbhgkhcl61z5JJdRMa+nmtbeI0jsQ/ye2KTWvbti2zH3mUgyVGNuV72dWIFrLY4d2jYbSLacuMGTO0DkeSnOLTT+Hpp2HaNDGL2tXn4uPi4IcfRI/166+LfNAp0tNFlj58uJPu0MP16CEOZBYUiGHhZWWNt0S0by/eX21hjOTV5PXSJmzduhU1KBI1IETrUPyTTk99aAe2bN2Gqqoo3jR9xY2mTp3KypUr+DjjML2jSgjz8dnVS7IDOV2lsPDXzxAU5F1nBSTfduKEmNgRHg69ejl+uHDxYrj/fhg5Ej74wH0tt716waOPiuV/778vcsFWP83u2iX6qHv1av7XemNvyODBsGIF/OMfosWjW7fLWz9AbJMMC5NJtQ+TleqrqK6uZv/+/dSHxWodil+ztYnjTHER2bIP7Yp0Oh0LFjxLjU3HZxm+nWTmVetYnB3EuHFjGTJkiNbhSNLP1q8Xa8NvuUUcBExMhD/9CeqbGLy0cqWoTvfrB0uWiI4Cd0pNFfOvd+8+P165xex2sZa8Vy8x8cIfREbCnDlQVCS+iVd7ZdKhg0yqfZhMqq/ip59+wmazYZOtH5qynntRs337do0j8WyJiYnceeddbMg3c6TUNy9CqSq8dyyUAHMg8+bN1zocyQu464zz8uXwySfQu7co1K5YAaNGwYIFYh700qWXf43dDn/8I0yZInaDLF9+eRuuu0yYIJL6r79u5X6SjAwoL4f+/Z0Wm1fo2hVmzIChQ+HBB698uw4dRE+1Px2A8SMyqb6KHTt2oOiN2ELk7FstqQEhEBTBNplUN+mBBx6gfbsY3kv3zdnVWwtMHCox8PDsR4hqzdBeyS+sWCHmO5865drHycoSAx8GDRKtFP37i6kaS5aInmWdDm64QSSub74plrC89ZaoZj/3HNx0k6hyR0a6Ns6rURSRC0ZHw7//DWfPtvCOdu9ueeuHtxs0SPxwr/bc1LEjVFS4/i+lpAmZVF/FT3v2YAmJAZ0Xz5P0EfUh7Tmw/wBWq/9uDnSE2WzmiSd/yalKHctPuvkasotVWxU+zgylW7eu3HjjjVqHI3mB3/5WVINd3Tm2dq1o2bjvvst7oSdPhn374M9/FkXc+fPF+b25c+H4cfjnP+GLLyAiwrUxOiIwULwoqK4WM7JttmbeQUPrR8+e7u9h8RYNE0BkC4hP8s1rxE5QUVFBTnY2to79tA5FAmwhMdQVHub48eNyfFoThg4dyvDhw/hu22aGt68jIsA3LjN+lxVIeT0s/OVT6L15cYLkFhs2iFHJAPn5rnuc8nLR7jF69JXzSJMJnnpKvGVlwZ49ok0kKcn1Ez6aKzYW7r0X/vtfscq8WTuVjh8Xky8GDHBVeN7vwgkgkyY5734dOeApl7S4nEyqr+DQoUOoqipbPzyELaQdAAcOHJBJtQPmzZvPg9u383lmMI/0qNQ6nFY7XaVjZW4g118/he7du2sdjuQFXnlFjEhWVTHpzFU2bBAV3TFjHLt9YqJ404ojudfQoaKvesUKUXR2+Cl3924wGPyz9cNRoaGiPURWqn2SbP+4ggMHDoCiYAv28cH1XkINCEEJCGH//v1ah+IVOnbsyJ133sWm/ADSz3r3a2dVhY8zQjAHBvLQQw9pHY7kBfbtE73MTzwhZjG7qlJtsYit0z16iO2EvuSOO8SLknffFe0gTbLbRVLds6foI5Eapyjie7RvX+vu58L1lN44htBHefdvWxfKzMyEwHDvW03uw+oDI0nPyNQ6DK9xzz33sOyHpXycYePX/cs87jKzo/aXGNl3xsi8eTOJ8ITGU8njNUzaePRRsdhv/35RTXZ219Dy5aLb4d57L/54YzmOt115N5th5kwxneSTT+DJJ5v4gu3bxcKXm292S3wXWpR2+dWr2aOOuD0Ohw0dKtZmVleDnLPvU2Sl+goyMo9jCQjXOgzpAvbAcE6fyqW+qaGvEgBBQUE89PBsMs/q2V7onZsW7Sp8mhlCxw7tuVmDX9aSdzp9WqzfjooSFWSbDYqLnf84a9eKQRepqc6/b0+QmCimlmzfLhLrq/ryS/GqpXdvt8Tm1UaMEJc5duzQOhLJyWRS3YiamhoKC/KxB8mqmCexB0Zgt9vJzc3VOhSvMWnSJJI6d+aLrBAsXjhib0NeALmVOmY/8ihGo7xqJDkmP//8kIWGtgxXtICsXy8OG/ryX83rroPkZLHbJCfnCjdSVZFU9+ghK6+OaFjfvnFj45+XM6y9lkyqG5Fz7pnDbpaVak9iDxQvcrKysjSOxHvo9XoenTuXwmqFNae8a8RVnQ2+PhFCj9TujG5s5a8kXUFe3vmkumHYgrMPK5aWwt69zTjE56X0erHTxGaDBx4QrdOX2blTzC30t4UvLRURITYCbdhw+ec2bRLDy3/4wf1xSa0mk+pG5J8radjNoRpHIl3IHiB+HgWuPMrvgwYPHkz//v1YnB1MjReN+V6da6a0Fh6dMxfFWxvCJU3k5Z1PpoODxcAFZ1eqN2wQBcXWJNWXnjXz1PNmbdvC3/8OP/4oFsNc5ssvxdSPPn3cHpvXGjFCLIq5cBj4kiViQ1B1NXz3nei7kbyKTKobUVhYCIDdFKxxJNJF9EYUY8DPPx/JcQ8/PJuKelhx0jtO5VdZFJbkBDNk8GB6yx5NqRlU9eJKNYgWEGe/Fv/xRwgI0HY8njvNmAHjx8Ozz4qe9Z+pqtheM2GCeAUjOWbECLFZsWGiVUYGTJsmJoP89reQkgLvvQcnT2obp9QsMqluRFFREYreAPoArUORLmE3BlNUVKR1GF4nNTWVkSNGsOxkEBX1nl/1XZZjpsoCDz38sNahSF6mogJqai5Oqtu3d36lev16McTBl/upL6QoYvtjfT08/vgFn/jpJ7HRZto0zWLzSiNHivcNlzzmzROv0hYvFu0hc+aI3pt167SNU2oWmVQ3oqioCNUU7HmrriRsxiDy82X7R0vMnDWLWhssyfHsanV5vcKKU0GMHj2aLl26aB2O5GXy8sT7S5PqykqoqnLOY5SViVzS0YUvviIlBX7zG/jqK9GdAJyf+uEP03mc2a+TkACdOoktRfPmwcqV8Lvfnf+LGxIietR37RKvZCSvIOdUN6KsrAybrFJ7JNVg5uzZs1qH4ZUSExOZMGEia9auYnJ8DeEeur58SXYgFpvCzJkztQ5F8kINSXVDTzVcPAEkObnl992QQ+3bJ4qL5eUXJ+/O4DF91WlpF/959igAnn5ajNebNw/GjlEJ++ILGDdOzC+UmqdhAPhbb4kEeu7ciz8/ZAhs2dL6RTGS28ikuhEVlZWoeu+c6+vrVL2JqkonlZv80IMPPsiaNWtYnB3I/V0dWZPmXiV1OtacDmTipEl06tRJ63AkL9RYpdpZSXWDzEzQ6ZrXT+0xyXIrGY3isOK118IdE86wPCODtCELOLIISOvu2UtXPM3QobB1qxitl5x8+Xaibt0gPFzcRvIKMqluREVFJaresy+R+yvVYKK2pga73Y5OJ7uXmisuLo7JkyezYtlSJifUEm32rOHVi08EYkfH9OnTtQ5F8lKNJdUNiziddZHr1ClRCTf5ae1l8GB47DF4/bVINjGME/1u0Tokj9XotscLt2sqyvn+6kvpdOKbvXo1FBWJMSySR5NZSSOqqmSl2lOpehOqqlLlrOZIP/Tggw+i6PR8l+VZLxyLanSszzMzZcoNdHD2NXXJb+Tni/Ne4ResGTCZxNrtigrnPMbp0xAb65z78la/+61KB10BDwe8T1WQFyR73jC7sDHXXiuGg3/6qdaRSA6QSXUjLBYL6PRN31Byv3M/F4vFonEg3ismJoapN93Mhnwz+dWe8xTwbVYgOr2B+++/X+tQJC/WMKP60nPmoaHOSaprauDMGejYsfX35c1Ccw7yhn0uh+uSWbtW62i83NUS/thYiIuDDz7QJjapWTznN6oHsdvtcvKHxxI/F3uja70kR91zzz0YjUa+yfKMlcKnq3RsLDBz88230FZe4pRa4dIZ1Q2clVQ3zGj290o1X37JTXxHn9Q6Fi+GkhKtA/Jh114LO3bA0aNaRyI1QSbVjbDb7ajIpNojKTKpdoaoqChuvW0aWwsCOFmp/VWZr7OCCDAFcM8992gdiuTlXJ1Unzol3vt1Uq2q8Nln5HcdxR33BWC3w2efaR2UDxs8WPRXy2q1x5NJdSNUWan2WA0vdmwXrnaVWuTuu+8mODiIL45rW63OKtezvTCAO+68k4iGE2WS1EL5+Y0n1WFhzkuqAwIgMrL19+W19uyBI0fIGHQP0dFwww3iQ/ty/fmb4kJt2oiNlR99JPqrJY8lk+pG6A0GFFX+xfVEDT8Xo7+sMXOhsLAw7r7nXvYUmzhapt0goC+OBxMWGsKdd96pWQySb6irE20IF86obtBQqW5tTnL6tOin9uvhQx99BEYjWQPEFsUJE8QLmU93plBdr/2VL590//1w4oRY5Sl5LH9+WriigAAz2K1ahyE15tzPxWw2axyIb7jtttuIigjn8+PBqBrsgjlYYuBAiZH7H3iQ4OBg9wcg+ZSGVeRXav9Q1dZtVVRVUan269YPm00sLZk8mbpgUZk2GOCee+BMlZmXlwzQOEAfNW0axMTAn//s3Pv11qkoHkom1Y0wBwSg2GRS7YkUmVQ7ldlsZvrMWaSXGdhV7N4xknYVPs0MpV1MW6ZOnerWx5Z8U2MzqhuEhor3rWkBKS8XSblfJ9VpaaJcf++9F324a1cYnpzPqyt7s+OEPGzsdGYzzJ8PP/wABw9qHY10BTKpboQ5UFaqPZbNik6vx2CQe4ucZfLkyXRKiOezzBCsbux62pxvIrtCx+xHHiUgIMB9Dyz5rKtVqsPCxPvWJNXykCKi9SMkRDRSX2Ja/0w6hlfz4H/HUGuRbSBON3cuBAbCX/6idSTSFcikuhER4eHorLVahyE1QrHWEhbWRuswfIrBYGDO3HkUVCusPeWeKwB1NvgiK5Ru3boyduxYtzym5PsaKtVX6qkGUW1uqYak2m9nVFdWijEft98OQZcfcA4y2Xj7/jQO50Xw6+8HXv71aWnn32S7QfNFRcHMmfDhh3KGoYeSSXUjoqOj0cuk2iPpLNVER0VpHYbPGTJkCP379+Pb7GAqLa6ffLMsJ5DSWpg7d55cNy85TV6eGNwUE3P555xVqQ4LO5+g+53PPxeJ9UMPXfEmv+iZy+yRh/nTqt5szmznxuD8xLPPgl4vZxh6KPnbrBGRkZEo9XINtifSW2uIjpZJtbMpisK8efOptih87eL15WdqdSzJCWL06FH06dPHpY8l+Zf8fJFQN9YdFhQkEu7WJNWFhY1Xwf3G229DaioMHXrVm/1p2lYSIiuZ/t/RchqIsyUkwK9/LWYY7t/v+se79CCjvLpwVTKpbkR0dDSqzQI2uQrb0+gs1UTJSrVLJCcnc+PUqaw9FUiuCxfCfJYZBDojjz46x2WPIfmn4mKIjm78czqdqDC3pv2jsBD8duHnwYOwZYuoUjexxyHUbOHdB9eTXhjOi98MdlOAfuSpp8TBgU8+EXMkJY8hk+pGdDh3ykVXe1bjSKSLWOtR62vo6LcNja43c+ZMgoOD+TAjxCUj9o6WGdhaEMBdd9/9878zSXKW0tKrL2VpzVbFigrx1lhriT/Y98Q72PRG3lfvd6hgObZbHvPHHODva3tx8LRc6uRUJtO5GYZn4LvvtI5GuoAcodCIzp07A6CrKcMefIWyh+R2utoy4PzPR3K+Nm3aMGPmLF577TV2FxsZ0NZ5V2vsKnycEUJ0VCR333230+5XkhqUlMDVnh5ak1RnZor3/lip1tnq6bLtA7L73ERtqOPfgIW3bWP9sY78Z3M3Xrp+NxFB9S6Mspm8vY2ha1cYPRrWroUBAyA5WeuIJGRS3ajY2Fh0Ot3PSZzkGXQ14ufRqVMnjSPxbVOnTuW7b77mk8yT9I4qweik61mb8gPIKtfzq1/NITDQtX3bkn8qLYX+/a/8+dBQsZSuJTIyxHt/rFR3yt1EYGUxR0Zc+YBiY4JMNj6fvZq+v72NdzZ155fj96E/93yyKK37ZbefPeqIM8J1Hk9PvG+9VfRVf/AB/OpXIDcNa87l7R+KolynKMpRRVEyFEV5vpHPj1IUZbeihRUWzAAAIABJREFUKFZFUaa5Oh5HGI1GOnTs+HMSJ3kGfW0ZeoOB9n59Usj1DAYDc+c/RmG1wupc54zYq7XCF8dDSO3ejfHjxzvlPiXpUiUlEHGVToPWVKobkmp/rFR3z1xKRWQCp1InNPtrUzuUce/gdNILw/nqpyQXROfHzGa47z4x9mbpUq2jkXBxpVpRFD3wJjARyAV2KIryvaqqhy64WQ4wHXjGlbE0V/du3Ti9cRtysJ7n0FedITk5WS5+cYMhQ4YwZMhgvt29nZEd6ggxtq7B+oecQMrq4P8ef0KO0JNcor5ebDu8WlIdFga1tVBTI3ZoNEdGhvh6f1vmGlKZR1zeTnbd8BtUXcsOMF+bVMiJklDWHIkjNryK4ckFDn9toxXt2S0Kwzf17AnDhsGKFeIyTUKC1hH5NVf/dhsMZKiqelxV1XrgU+CmC2+gquoJVVX3AW7c5da01NRU1LpKOVrPU6h2DNXF9OzRQ+tI/MYjjzxKrRWW5bQui6iwKKzIDWLUqFH0kD8/yUVKS8X7pg4qAhQVNf/+MzL8s0rd7fhyAI4On9Gq+7m9fyap7Uv5aHsXjhXIBV5ONW2a+Mv93ntgs2kdjV9zdVIdC5y84M+55z7m8Rp++esrW/DsKzmdrqYU1WaRSZkbJSUlMWbMWFaeCqKivuULYZbnmKm1wfTp050XnCRdoiGpbqr9A6DA8ULpz/wyqVZVUrJWcbpdP6oiW1cB1evg4RGHiQ6p5c0fe5JTEuKkICWCg8U0kNxcWLNG62j8mquT6sZ+E7foOrKiKLMVRdmpKMrOopaUGZopJSUFvcGAvqrQ5Y8lNa3hxU1qaqrGkfiX6dOnU28T7RstUV6vsPJUEGPHjiUpSfZTSq7TsLX5apXqhq2Khc18Wq+pEdsUfeaQ4oXrwtPSrniztmcO06byFOmJE53ysMEBVp4ct58go5W/r72G/LPywLLT9O0LffrA4sVi1J6kCVcn1blA/AV/jgNOt+SOVFVdpKrqQFVVB7Z1Q7nAZDKR2r07hsoWlDQkp9NX5BMeHkFsrFdc6PAZnTp1YsyYsazNC6SuBVcVfzwdQJ0VHnjgQecHJ0kXaE6lurlJ9fHj4r2/Vaq7ZK3EqjORFT/KafcZGVzHk+P3oyjwt7W9OVMZ4LT79nt33SUW83z8MS5ZNCA1ydVJ9Q6gi6IoiYqimIC7gO9d/JhO069fP3RVxWDzoNma/khVMVUV0L9/P5QmNnlJznfzzTdTY4Hthc375WdXYX1+EH379JGzxSWXc6RS3dKk2h/H6Sl2K8nZ68iJG4rF5NxWjXZhNTwxbj+1Fj1/W9ub8ho5Cs4pIiPh5pvhwAHYuFHraPySS5NqVVWtwHxgBXAY+FxV1YOKorysKMpUAEVRBimKkgvcDvxLUZSDroypOfr27Quqir5CVqu1pNSVo9ZViZ+H5Ha9e/cmPi6W9aebd2DxUKmRomqFG2680UWRSdJ5jlSqAwLEW0uTan+qVMfl7SSwroz0zs5p/bhUfEQVj409QFm1ib+v7UVVnZzq5BRjxkD37vD55y07PCC1istnW6mq+oOqql1VVU1WVfX35z72a1VVvz/33ztUVY1TVTVYVdUoVVV7ujomR/Xs2RO9Xo+hIl/rUPxaw/e/T58+GkfinxRF4YYbp5J+1kBeleNPGRvyAggNCWbkyJEujE6ShIakOjz86rcLDW1ZUh0ZKc6D+YuUE6upNYVysuMQlz1Gctty5ow+SF55EG+s7ykTa2fQ6WD6dDAY4O23oa5O64j8ihwYexVms5lusq9ac/qKfMLahJMg529qZsyYMQDsLTE5dHu7CvtLAxg2fAQBAbJnUnK9khJxELGpMfYtTapTUloem7cx1FbS+eQGshLGYNc79m++pXp0KOOh4UfIOhPGLW9NwmKTLX6tFhEBM2bAyZPwzjtg96iJxT5NJtVN6NO797m+aqvWofgtY1Uhffv0lv3UGmrXrh2xHTtwsMSx3sfsCj2V9TBw4EAXRyZJQmnp1fupG7Qkqc7MBH8aXtN573cYbbUua/24VP+EYu4bnM6qw3E8/cVQtzymz+vdG+64A/buhS+/1DoavyGT6ib06dMHVLscracRpa4Saitk64cHGDR4CEfOmrA6UPQ4WCqS7wEDBrg4KkkSmlpR3iA0tHmtplYr5OT4V1Kdsv0jKoNiyI/p5bbHHJGSz1MT9vH6umt4Z2M3tz2uTxs3TrytWQPr1mkdjV+QSXUTrrnmGhRFkYcVNaI/13rTu3dvjSOR+vbtS50VciqbXlV8rMxIQnwckY6UDiXJCZpTqS4qcvyK+MmTYkldYmLr4vMW5vJC4g6tJKPzBFDcmyIsvHUbE1NzmfPxCLZl+dGpUFe6/XYxv/qzz+B7rxm+5rVkUt2EkJAQ2nfoiK6mVOtQ/JKuugS9Xk+iv/xG82DJyckAnKpq+jDRqRoTKV26ujokSfqZo5XqsDBRfS4rc+x+s7LEe395Ckre9Tk6u430zhPc/tgGvcqnD68hNryK2/81kWI5w7r1dDqYNQs6dRLtII5uXLTZYNcu2LoV0tOhqsq1cfoIedTWAclJieTtPkCt1oH4IX1NGXHx8RiaOn0kuVzHjh0xGg3kVl29Ul1rhaJq5GxqH6YoSjzwPtAesAOLVFX9u5YxNadSDaKv+mq3X7RIvG8Y97t1q3+M1EvZ9hFn4npTGpF88Scu3bw4ynkLYS4UGVzHl4+sZtgfb+Led8ZxS58sdLL81zoBAfDYY/DuuzB1KixZAmPHNn7bU6fgr38VBxwvfOVpMMBtt4mvk+ebrkj+VXVAYmIi1JwFewtWykmtYqw7S7I/NTN6ML1eT6eEBE410f5xulp8Xl5d8GlW4GlVVVOBa4F5iqL00CoYVW1eTzU4flixuFjkEP7QyRRWmEG7rK1kDL5X0zgGdCrm9Ts3sfJQPEsPdNI0Fp8REgKrVkFCAkyYAL/7nbhkA+If0OnT8N574pLM3/4Gv/iF2Mz4m9+IhDw1VbSQvPWWqGJLjZLlPwd07twZVBVdbTn2IAeetSXnsFtRa8tlxdODJHTqzIG841e9Td65pFqOQPRdqqrmAXnn/rtCUZTDQCxwSIt4qqvBYnEs8Q0LE+8dTarPnBH3q2/6KIHXS9n+MaqikDHobjiQpWksD488wubj7Xh/S1cSo8u5pqNswWy19u1h2zaYMwf+3/+DhQvFx86cEZd6jEZ49FF46ilo+L27aBF07Ag9e8Lq1WKSyOLFYnOjdBmZVDsg/Nw2AcUmh6i7k2IV6+HDm9rmILlNXFwc62rAagfDFa5zFVTr0SkKHTp0cG9wkiYURekM9AO2XfLx2cBscP0LrIYV5a6qVEdFtSwur5GWBqpKyvq3yWvbhyqNEupFad0v+vPAhCLWHI7lP5u68+Lk3USHyN/BrRYWBh9+CHfdBStWwLJlYgh7cjIMGiQS6sYoCkycCHl5sHy5qFx3k1NaLiXbPxxgNp9bz2z3rlnV9qBIVL0RVW/EGtoee5CXXb889/3++fsvaS42NhZVhaKaKz915NfoiWkbjcnk2qURkvYURQkBvgKeVFW1/MLPqaq6SFXVgaqqDmzr4mZkR1aUNwgOFvmBo2P1ioshOrrlsXmL6JKjhFecJCPR/QcUr8RksDN75CHsqsJra3txtsaxOflSExQFbrwR3ngDnnsOHnpI9EqHhDT9tXfeCTExoj/bYnF9rF5GJtUOaEjqFC9bAFOXcC22oChsQVHUdL+euoRrtQ6pWRSZVHuc+Ph44HzfdGPyqo3EJcg+SF+nKIoRkVB/pKrq11rG0lCpdqT9Q68XlWdHKtX19VBe7h9JdZcTq7HpjByPH6N1KBdpF1bL/DEHKKsJ4K9relNRKxNrTQUEwN13i1eymzZpHY3HkUm1A35es+xllWqvd+77LSueniMxMRFFUcipbLxzzGqHU1U6unTp4ubIJHdSxHrTd4DDqqr+Ret4mlOpBmjXzrGk+swZ8d7X2z8Uu5XkE2vIib2W+oBQx74oLe3yNxdJiSln3pgDFFea+ePKPhw44LKHkhzRvbtoF1m+XFarLyGTagdUVFQAoBpkcudWevH9rqys1DgQqUFgYCBxsR3JqWg8qT5drcdqh5SUFDdHJrnZcOB+YJyiKHvOvV2vVTDNqVSDuHrtSFJdXCze+3qlumPBTwTVlpDhprXkLdGt3VmeHLefWoueIUPgo4+ufvtFad0ve5OcRFHghhvEq9ktW7SOxqPIg4oOKD73zKoagzSOxL/YTeL73fD9lzxDl67d2LflVKOfyz6XbMuk2repqroR8Jhhtc2tVMfEwE8/NX07f0mqu2Stos4YQk6sZ7cIpsSU89L1P/F94bXcd5+YEPf661pH5adSU8X4vRUrxIi9K4zHaZj33mD2bDfEpiFZqXbAmXPXAFWTTKrdSmdE0Rt//v5LniE1NZXiGiipvfzpI/2sgeCgQOLi4jSITPJXpaXid3qog50Ljlaqz5wROy8axvD5Ir21ls4n08hKGIVN7/kbDNsE1rN2rZgI98EH0L8/5J0N1Dos/9MwDaS4GJYu1ToajyGTagcUn5v+rxrkgTm3UhRUUxBFRUVaRyJdoG/fvgAcKbv8QteRswH06dsXvT8M9ZU8RsPiF0cXvcXEiGVx9fVXv13DOD1f3ujXOXcjJmuNR7d+XMpohJdfhvXrxUHSP67sy5F8OXrV7fr2Ff/wXntN60g8hmz/cEB2djaY24Diw8+sHspqCuPEiWytw5AukJSUREhwEIdLay/6eEmdjvwqhVv79tMoMslfObqivEFMjHhfWAhXu6jiD+P0uh5fTkVQO06366t1KM02cqTYZTK8bz1/X3sND484Qv8E17QLNtaTPXvUEZc8ltfQ62H0aPj2Wzh0CHpotlTVY8gs0QEZmcexmOWrYC3YAsM5eTIHq1VOXvEUer2e3n36cvjsxZeKD5eK1+gNlWxJchdHV5Q3aNdOvG+qBeTMGd9OqoOrC4nL28mxpF94bdGoc2d4dtIeEqMq+PfG7uw56eOjWjzNyJFizN4bb2gdiUfwzn9FblRXV0d+3mnsgTKp1oI9MAKbzUZubq7WoUgXGDhwIIXVChb7+Y8dKjESGhIsDylKbteaSvWVVFWJ9ee+nFR3Ob4SBZVjSddpHUqrBJpsPDbuAAmRlSzamMr+U1626MybhYSIudXvvw9nz2odjeZkUt2EnJwcVFXFHtiMMojkNPYg8X3PytJmba7UuP79+wNQZRFPIaoKB8vM9B8wEJ0vN6BKHqm5lWpHkuqGzzXc1ueoKl2PL+d0TB8qQmO1jqbVAo02nhi3n9jwKv6Z1oODp+XvbLd57DHxKvTdd7WORHPyt18TCs7tsrU7OhBfciq7SaxNLXTkqL7kNp06dSIqIpwqqzgZll+to6QWBgwYoHFkkj9yRaW6YY25rybVMcUHCa846fVV6gsFmWw8OW4/HdpU85ZMrN2nf38YNgzefBPs9qZv78NkUt2EknNbBVSjHNmjCb0JRWf4+ecgeQZFUejbfwDVVvEUcvSsWB0s+6kld7PbxSSP5lSqQ0LAbG66Uq0o0LZt62P0RN2OL8eiN5OVMEbrUJwqOMDKk+P20S6shjfWX0NaegetQ/IPjz8OGRliy6Ifk0l1E36eUS2Tam0oCgQEyVnVHqhnz57YVLDYIfOsgdCQYOLj47UOS/IzZ8+K9qPmVKoVpelZ1YWF4j6NxtbH6Gn01lqSs9eSlTAaiw8uNQsxW3lm4l56dCjlo+1d+Hh7CrUWme641K23QseO8Oqr4h+kn5J/y5pQWlqKYjR77cloX2DTB1DasDJN8hg9zo1PqrEqZFSY6NHzGhRHBwVLkpM0XMRqTqUaHEuqfbX1o/PJDZgsVRxNnqx1KC4TaLQxd/QBJqaeJC29A/+7ZCCL9yb4c77nWkYjvPiiGB7ux8tgZKbYhPDwcFRrHdhtWofit/TWWsLD5fQVT5OcnIyiQLVVx+lK3c9JtiS5U3NXlDeIjYWcnMY/p6q+nVR3O76c8uD25MX00ToUl9LrYFr/LBZM2kuA0cbUf1zH2L/cwNbjPvqD1drs2dC1KyxYAH46Blcuf2lCfHw8qCq6unI5AUQLditqbYVsK/BARqORAFMAlfW1qIjDi5Lkbg2V6ua0fwCkpMDKlaIn+9KBNcXFUFPjm0l1cEkOsfm72N3rAb+5ApvctpyXJu9GReHlpf0ZuvBmbu6bxe9v2sHGjPaX3d7vl7q0lNEIf/wj3HwzLFoEc+dqHZHb+ce/qFZoSOZ0tXL+ohZ0teUAMqn2UAFmM1ZVtHzEXW01nSS5SEsr1SkpInE+ffryz6Wni/cNS2J8SfdN/wHgaJLvtn40xqBXMert/Grybqb2PsHyA/Fc8/LtfLYzSfZbO9PUqTBuHDz/PGT73zZkWaluQkJCAgC6qmKI6KxtMH5IVyVWzjb8HCTPEhBwfqtibKz3z7qVvE9rKtUgBhZc+nqwIan2tUq1YrfRbdM75HYYRGWIf07FMBvtTOmVw+gup/l+X2fWHo1jz8lopg89Srf2snjWaooCb78NvXvDjBlwx2pxKSgt7dwNzl0FmD1bsxBdSb48a0JQUBDDhg/HXHQExVKjdTj+xW4jMH8fnTsnkpSUpHU0UiNMJtPP/x0YKCfkSO7X0kp1ly7ifUbG5Z87dkzkAb62TTHu4HJCSnM5knKD1qFoLsRs5Z7BGSyYuAej3s5f1/bmhwPx2OVBxtZLTIS//hXWraPHj29pHY1byaTaAY/Mno1it2I6vVfrUPyKsego1JYzZ86j6PV6rcORGiGr05LWSkogMFDMnW6OuDgwmc5XpS+Ung5RUeBrTzupG/5NdVg7TsQN1zoUj5ESU86Lk3czMKGI7/Ym8o/1PSmpCmj6C6WrmzULxo9n4OJfY6ryn+ldMql2QKdOnZgyZQqmoiMo53p8JRez1hOYt5d+/fozePBgraORruCOO+7QOgTJz5WWNr9KDSJhTkpqvFKdnu57rR9BZadJ2L+EY0Ono+pk5+eFzEY7s4Yf4a6B6RzKj6D/729l5wkfu0zhbooCf/4zAdWl9Fv2f1pH4zYyqXbQ9OnTCTCZCDyxEWwWrcPxbaodc/YWVEsNc+Y8Kmcfe7CwsDCtQ5D8XHNXlF8oJeXypFpVfTOp7rb5XXR2G0dGPKR1KB5JUWBstzwWTNyLqsKIV6fy8fZkrcPybn36cGzodK5Z9xqhlXlaR+MWMql2UFRUFAsWPIOhsoCg9FUysXYV1Y75eBrGkkxmzpxJ165dtY5IugrZRy1praSkZZVqOJ9UX7gQJD8fqqp8bPKH3U63Te9wqttYymNStI7GoyVGV7DrV18zJLGQe98Zz0vfDsRub/y2i9K6X/TmtxYtuvjtAjum/hYVhWuOfKFRcO4lk+pmmDBhAi+99BLGqkKC01eArV7rkHyL3Y45cz3GkuM8/PDDPPDAA1pHJDXhwoOKkqSF1lSqu3SB6mrIu6CItvfc0Zn2l48v9lqxR9YQVpzFkZG+OXHB2aJD6lj15A/MGn6E3y/rz7R/TaTOKtOllqiOiCWn9w0kZ69D8YMlerKxqpnGjx+PXq/n5Zd/S/CxlVR1mQgGeaih1ew2Ao+vx1CazZw5c7jzzju1jkhygEyqJa2VlED//i372gvH6nXsKP573TqxwyLZh678p25YRG1wFFl9b9E6FK9hMtj59/1p9OxYyjNfDmFndlseG3uANoGymHYlV6rWZw68k6TdX9G+cC8Q5N6g3Ey+9GqBMWPG8L//+z8Ya84QcmQpusoirUPyakptOcHHlmMozWb+/PkyofYiMqmWtNbanmq4uK963ToYMkRMBvEFgWfz6bT3O45d+wB2oywANYeiwC8n7GfJ/BUUVgTyp1V95GSQFsjpNQWLIZCU7LVah+JyMqluoZEjR/Lqq68SHWQg+MgSTLk7wQ8ubTiVqmIsOEjooW8JtlXw0ksvMW3aNK2jkppBd+l+Z0lyo/p60f/c0p7qhAQwGM6P1Ssvh127YOxY58WotdQNi9DbLBwa438ro51l8jUneWLcPipqjby6sg9FFc2c3+jnbKYgTsQNJzHnR7BatQ7HpeRvxFbo378/7733XyZfdx0BefsIOfz9zxsApatTassJOroMc842Bg0cwPvvvceECRO0DktqJjmZRdJSw+KXllaqDYaLx+pt2AB2O4wZ45TwNKez1pOa9k9yrpksDyi2UnLbCp6asI86q56/ruktK9ZXceAAvPoqLF16PofO7DQOc305HD6sbXAuJpPqVgoJCeG5555j4cKFRJp1BB9ejCl3l6xaX4mqYiw4ROih7wixlovv3Suv0LZtW60jk1pALuWRtNSworyllWq4eKzeunWi7WPo0NbH5gk6//QNwWfzODhmvtah+ISEyEqeHL+f6noDf13Tm7JqH+kRcpKqOgOvr+vJ66+LKTrffw+//z2cOgW5HQZh1ZngyBGtw3QpeVDRSYYMGcL77/2XN998k+XLlxNQdoKa2AFYwzuJxiwJfUU+5twd6CqLGDhoMAsWPEOMrw2D9TOyUi1pqbWVaoAePWDVKkhLE0n10KFiQ6Mv6Ln+DcqjkzjZ8zqtQ/EZCZGVPD5uP39b04s/ruzL/LEH6Nim+rLbZWTAP/4Ba3/sQWWtkWuTChiZkn/FdKCxQ36zR3lPAmqxKSzamEp6YRumTRNXew4fhvffh/fegxeGmyiK6kaHzEytQ3UpWal2otDQUJ5//nkWLlxIfHQbAjPWEnz0B3SVhVqHpiml5iyB6asJOvID0QF2XnjhBf74x4UyofYBMqmWtOSMSvVzz4lJHzfeCD/95DutH1En99AhYyMHx8wDefbBqZKiK3h6wj4sNoU/rujLzuxoLDbxXJiZCU89JV6svfUWFFeaqbXq+Wh7V/6zuRu1Ft/8Wfzy82EcyY/gvsHpTJwoJuj07i3+XWVnQ2ZRGAVte0FOjjgM4aNkpdoFhgwZwoABA1i2bBlvv/MOusNLsER0pi5uIKrZfzbQKZYaTKd+wlR8FLPZzH0PPcS0adMwm+UhD18jk2tJCw2V6tYk1dHRsGIFDB8uDir6yiHFHuvfxGIK4uiwGaIMLzlVp6hKnv/FHt5Yfw3/3tiDQKOVN9Zfw5FHxGuYmTPh5Zdh8YLd2FVYdiCBxfs7UV5rYp4NfKlz7t1NXXlzfU8mpp5kWHIB0O3nz117LXz7Law9Gsu4xGvg0Mciy/ZRMql2EYPBwI033sj48eP57LPP+OSTTzEe/Jr66O7Ud+yLavThxNJmwVRwEHP+fhTVztSbbuLBBx8kojW/+SSPpCgKgYGBPPjgg1qHIvmhhkp1a9o/QEwBWbMGvvpKJNfeLqCqhC7bPiL92vupD5bPu64SFVLHrybv5nB+ODuzYwgLrOeR5yK4+Wbo3Pn87XQKTOmVQ5vAej7Y1pWFC+HFFzUL26nSC8J47LPhjO12ilv7Zl32+YAAGDECVq+K5lDPgVwHopzvo2RS7WJBQUHMmDGDqVOn8p///IcffviBgJJMajr0xRKT6luX5VQVQ8lxAk/thLoqRowcyezZs4mPj9c6MslFFEXh+++/x2g0ah2K5IcaKtXh4a2/r65d4YUXWn8/nqDbpv9gsNSI1g8387d13Qa9Sq/YUnrFlooe6NlX3lo5PDmfI/nh/PrXMYwZA8OGuS/ORnu2W7lg02JTuO8/4zDq7bw3fT3LDjT+u37sWFi9Clac6M5T7dpdPBjex8ik2k2ioqJYsGABt99+O6+/8Qa7dm4j4MwxauIGY2sTq3V4raarOkPgya3oKgpI6dKFxx97jN69e2sdluQGMqGWtFJSAm3a+Nal9NZS7DZ6/PgPTncZRUmcfA72JIoC9w5O54w+hrvvhj17Wte6pLWXlwxg+4kYPp+9ivjIqiveLjISeseeYfuJGNR+KSh794Cq+uQQB5lUu1nnzp3506uvsnnzZl57/XUKjq3AGpFAbdxgr+y3Fn3TuzAVHSM0rA2PLljA5MmT5VIQSZJcrrTUu5MSV4g/sIyw4iy23bpQ61CkRgSabHz6qahSP/QQfPkleFJquWjRJR9I697oFJK1Rzry+2X9mD70KLcPuKTto5Ee/h4dOrAnN5qMtkPpUrUJjh6F7r53VUNmPhpQFIXhw4fz/nvv8fDDDxNcXUjowW/EfGubRevwHGO3Y8w/SOiBrzGfyeT222/n448+ZMqUKTKhliTJLVqzotxX9fz6t1QGtuVERZRIbuQhRY8zaBD84Q/w9dfwz39qHU3zFZabufedcXRrV8Ybd29y6Gu6tS8DYJ1tlPjA1q2uCk9TslKtoYCAAO69914mTZrEv/71L1avXo3pbA6V3a4Hgwdva7LbCUpfib78NP0HDuTxxx6jU6dOWkclSZKfKSmRleoLtck/Snzednb0noWqk7/ePdlTT4nDsY8/DsodqegU9epfcFkJWRt2Ozz43zGUVptY8cQPBAc4tna8XWgNbQLrWJ/fndlGI+zb5+JItSH/1XmAtm3b8tJLLzFx4kReePFFgjLXUd1lkmceYlRVArI3oy8/zVNPPcWNN94ox6lJkqSJ0lKIi9M6Cs/R48d/YNMZOJJyg9ahSE3Q6eDTT+Gee2DOxyMZkZzH+NRTdAirbnarsTsXx/x5VW+WH0zgrXs20DuuxOGvUxTo1u4s647FonaMRdm71yXxaU0m1R5kyJAhLHjmGRYuXEhAzhbqOg3zuEZ+Y/4BTMXHuO+++5g6darW4UiS5Mdkpfo8fX01Xbe8R1b8aGoCZU+MN2jTRqzy/tWUPSxc0ZeNmR0ID6wjIriOQKOVLcfbkRhdwcBORYzvfooAo13TeLdlteXFbwdzW//jPDLqcLO/vmu7MrafiOFoyjC6713hk4cVZVLtYSZPnkxubi4fffQRdnMbLO2v0TqknxlKszFzVGNpAAAgAElEQVTn7mDMmDHMnDlT63AkSfJjqioPKl4oeefnBNSc5VCXm7QORWoGvR5euXU74YF1HMyL4FhBOJV1BqrrDaw8HMfpsmAAwsz13DEwk5eu/4lOUZVuj7O4MoA7/z2BuIhK3r4/rUW5cLd25/qqlXF0P/Mx5OVBx45OjlRbMqn2QLNmzWL//v3s27cda0Rn1IAQrUMC1U5gVhrt27fnhRdekIcRJUnSVFkZWCwQE6N1JJ6h+4ZFlLbvTn6MHKPnjSKD6xiZks/IlPyfPzZ71BFq6vWsO9qRL3cn8cHWLvx3czfGdjvF1N7ZmAzuqVxbbArT/jWR/LOBbFiwmPCglq0ZbxtSS1xEJesqBjIHYO9en0uqZWbkgQ4ePMjBQ4ewhXVANQVpHY6g6KiPSCQ/P5+NGzdqHY0kSX6uoEC8b99e2zg8QcSp/bQ/voUjI2f73OV0fxdosnF9r5P858EfSf/tZwxJLGTV4Xj+b1k/TpYEuyWGJz4bxo/HOvLOA2kM6lzU4vtRFBjTNY/1p7uggk8eVpRJtYfJy8vjxV+9hM0YRHXyOFA850dU12ko9tD2/OEPr3D4cPP7qSRJkpylIalu107bODxBatoirIYAjl37gNahSC4UH1nFA9ce48lx+6i2GHhlRT82Zrj2VeWqw7G89WNPFkzay71DWr8JcXhyPkWVQWR3HCYq1T7GczI2ierqal544UUqq2upTJ7oeWP1dHqqk8diNQTywosvUlhYqHVEkiT5KZlUC/r6arps+4Cs/tOoC4nSOhzJDVI7lPHr63fRtV0ZH2zryic7krHZnX+FYvt2+HJ3MncMyOSVW7Y55T77JxQD8FPHKbJSLblGbm4ub7/9Nvff/wAnsk9QlTQGNbCN1mE1SjUGUpUynrMVVcyYOYvXXnuN9PR0rcOSJMnPyKRaaDigeHjkbK1DkdwoxGxl/pgDTOiey/pjsfx9bS8q65x3TG7/fvjvf8XEjvdnrHPahN9esSXodXZ2BwyFI0egttY5d+wh5EFFjVRXV7N+/XqW/vADBw8cAEXBGhZLfZdJ2NrEah3eVdkDI6jseh31efv5+tvv+Prrr0lMSuL6yZOZOHEi4eHhWocoSZKPKygQkxOi/Lw4m5r2L3FAsctIrUOR3Eyvg9sHHCcuopIPt3XlD8v78ejIQ8RHVjn09YsWAZfOuB4lCsj/+hfExsKcIQedOsov0GQjtX0Zu6u7g80Ghw5B//5Ou3+tyaTajex2O3v37mXZsmWsX/8j9fV1ENiGurgBWKJSUE3uOXTgDPbgaGpTxlJrrcNYcpzMgnTefPNN3vrnPxk2dBiTJ1/HkCFDMBjkXzFJkpyvoADatvXMHVnuEpm7j3ZZW9ly+1/kAUUnamyZihYcjWNoUiHtw2p4K60Hf1jRj6m9TzBrxFH0uia2NDZixw54912Ij4cnnoCgnbZWxdaY/gnFrMxMEn/Yt08m1VLz5OXlsXLlSpb+sIzCgnwUg4m6iM5YortiD27r3U+GhgAsMalYYlLRVZdiLE5n0/adbNy4gTbh4fxi0iSuu+46kpKStI5UkiQfkp8vWz9SN8gDipKQGF3B/7t+Nx9tT+GbPUkMXRjCq7dtY3TXPIe+3m6H7/Z1ZvlBSE6G+fMhyEXDx/onFPP+1q7kGeLp8OGHUF8Ps32jfUkm1S6gqirZ2dls3LiRtA0bOHb0KAC2sI7UJ43GGt4J9L73rbcHRVCXMJi6uIHoz+ZiKU7n8y++5PPPPyc+PoHRo0cxYsQIunXrJlebS5LUKgUF/p1U6+ur6bJVHlCUzgs1W3hk5GG2nzjDikNxjPnzjUxMzWXemIPY7KC/wtdlFYfy5e4kMoraMGIE3HUXGI2ui/Pnw4qR4+mQu8V1D6QB38vsNGK32zl8+DAbN27kx7Q0Tp86JT4eEoMlbgCWyGTPWOLiDjodtogEbBEJ1FlqMJRkcaIsm5MffcSHH35IZFQUo0aOZMSIEfTt21e2iEiS1GwFBdDdM67SayJ5x2eYasvlAUXpIooCQxILeePuTbyxrid/XdOLm9/6BWHmelKPQ9euEBYGBgOcOgWHN13DwbxIQs31PHDtUYbd183lF8/7xp8BYLd5GNfnfiHWo/oImc20gsVi4aeffjpXkd5IWWkJKDpsoe2xdBqKNTzBq/qkXUE1BmJp1wNLux5grcVQdpKC0hy+W7yEb7/9lqDgYIYPG8bIkSMZNGgQgYGBWocsSZKHU1VZqU5t2KAoDyhKjQg02Vjwi338csJ+lu5P4Hc/9OPQoRi2XTIZLybUzJRrspnUIxez0QZKN5fG1dCL3S60mlXlg3mpqgrOnnXpY7qTTKqbqbq6mm3btrFhwwa2bNlKTU01it5IfVgs1qReWNvEed58aU9hMGON7oI1ugu1NiuG8lNYynJYvX4Dq1atwmg0MnDgIEaNGsnQoUPlFBFJkhpVXg51df6bVMsDipKjDHqVm/pmU1AeiDoyhqIiqK4Wbczt2kGbvTs1iSs+spIjBYniD7m5msTgCjKpdpCqqqxcuZLXXnudqqpKFFMgdWHxWOM6YQvrADr5rWwWvQFrRCesEZ2oVe3oKwowlGazZfc+tmzZjNFoYtasmUybNk22h0iSdBF/n1EtDyhKLaEoEBOjdRRCQkQlO7NjOEMkUTKpdpyiKNcBf0f0yL+tquorl3w+AHgfGACcAe5UVfWEq+NqjsLCQv70pz+xfft27KEx1HYfhS0kxqNWiHs1RYctrAO2sA7UqUPQVZ8h4PQe/vnPf7Jm7VpeeP55OT1EkqSfNSTV7V27odkjGeqq5AFFya1cMWIwIbISgM3mcdyYm+30+9eKS5NqRVH0wJvARCAX2KEoyveqqh664GazgFJVVVMURbkLWAj/v717D4+qvvc9/v4m5EoS7iogt0AABRUlCREQUJAiVKRutIh6xCqWU8W6rdb2ObZ1e9qeat21tuxNpVZta7u11dYigldUUElIuISLAkGDikFAEEKAJCT5nT/WQiMSMiHJrJnJ5/U8eTKzZs3MZzHDmm9+87vwzdbMFaq6ujoWLVrEf8+fT1X1ESp7j+DIKWeomG5NZtS178rhAeNpt7eUraX5zJ49m2uvvZarr76ahNYckiwiUaEtt1RnFv3NG6A45ttBR5FYtGxZWJ6ml19Ur0gcx6XbfxaW5wyH1m6pzgW2OufeBzCzJ4HLgPpF9WXAPf7lp4F5ZmbOBTsc9OOPP+b++++nuLiY2ozuHB4wCpecEWSktsWMmi6ZHMjoTtKHBTz++OO8/vob/OAHdzG4LQ/5F5E2XVSfsfxhPut+Bp8MGO1tCFMRJC1owYKgEwQuLamGLu0rWW3Dvf/QlZWQnBx0rGZr7SbXnsBH9a5v97cddx/nXA2wH/jKd1pmdpOZFZlZ0e7du1sprsc5x+23f4/i4mKqep7HoYGToragrkvtTF1q56BjnDSXkEJl/3EczhzLtm2lzJ07l0OHDgUdS0QCtHOnt5Ji165BJwmvzh8Vc2ppAe+OvkkDFCXq9epUwcbqAd7KM++80/gdokBrF9XH+19/bAt0KPvgnFvgnMt2zmV369atRcI1xMy47rr/RUJCAsm73iV+30eN3ylCVfXOo6p3XtAxmiWuYhcpZWswM66++mpSW2uZJxGJCjt3egV1fEOrWcSoowMUS87XAEWJfr07V7D9cFcOkAbFxUHHaRGtXVRvB3rVu346UNbQPmbWDugA7G3lXI2aPHkyjzzyCP169yB16yskbXsLao8EHattcXUkfryG9puep1tGMg899BCzZs0KOpWIBKwtzlHdruogWQVPUDr8CqraR++3jyJHHR2sWBw/HNatCzhNy2jtoroQyDKzfmaWCMwAFh6zz0LgOv/ydGBp0P2pj+rTpw+/mz+fGTNmkPjpFtLffY64g58GHatNsMpy2m9eTFLZGiaMH89jjz7K2WefHXQsEYkAbbGo7l+kFRQlthwtqld3uDBmWqpbdaCic67GzG4BXsSbUu9R59xGM7sXKHLOLQT+APzZzLbitVDPaM1MTZWYmMicOXPIzc3lpz/7GbyzENe+C9XpPanteDq17U/xOvdJ8zhH3KE9tNu/nYT924k7uJuU5BS+d/fdTJgwIeh0IhJBdu6E/v2DThFeg5cv+PIARZEo1yGlmozkKla3y4V1v/WWSo3ysQKtPk+1c24xsPiYbT+ud7kSuKK1czTXeeedx+OPPcaiRYvIzy9gw8YN1H2yDmuXSHV6D2o79KSmw+ltflnyprAjlcSXf0y7/dtJPFCGqz6MmZE1cCB5IyYxZcoUTm1rzVEi0qg201Ltz+zR+bOtnFpawNtXPBj1RYdIfb07H2T1wSGwfw+UlUHPY+eyiC5aqq4JMjIymDlzJjNnzqSiooLVq1dTUFDAivx89m57CwCX2pnqjJ7UdjjdWyAmro2NpDkRV0fcQb81uvxj4ip2AZCWns6IC0YyYsQIcnJy6NSpU8BBRSRSVVR4yyy3ZFEd6TOcDdnyT2riEylxAzSFnsSUXp0qeGlHTw6TTEpxsYrqtiotLY0xY8YwZswYnHOUlpaycuVK8vPzWb9+PbWfrPdasdNOo7bD6V4rdlJa0LHDzo4cJn7/0dboHbgjXmv0wEGDOD9vMrm5uQwaNIj4tjaMX0ROSlubozqpaj9ZpS9T0m8iVUnRObWrSEN6dz5ArYtnA0PJWbcOJk8OOlKzqKhuAWZGZmYmmZmZzJgxg0OHDn2pFfvTD972dkzpSHVGD2oyelKbfhrEx+DqgHW1xFfsIn7/xyQeKMP8gZ3pGR3IGzeaESNGkJ2dTceOHQMOKiLR6P33vd99+wYaI2wGb32edrVVbBh4edBRpIW0xrLf0aqPP1ixsPPXyImBwYoqqltBamoqo0ePZvTo0Tjn+OCDD1i5ciWFhYWsXbuWIzvfgbh4atNOoSajBzUZp3sLtERjXznniKvc7/eNLiOh4hNc7RHi4uM588wzyc2ZSm5uLgMHDiROAzpFpJlKSrzfWVnB5ggHq6vhzC3P8vGp5/JZpzY2MlPahM7tqzg14xD5qeP5zrp/Bh2n2VRUtzIzo2/fvvTt25crr7ySqqoq1q9fT1FREfkFK9lWuookVmGJKVSndaemQ09qM3riEiN4gZOaKtqVl3mt0RVlUOn9pdm9ew9GXDiFnJwchg0bRvv2GrQpIi2rpARSU6F796CTtL5+Hy0j/dBOVmTfEnQUkVZhBudn7iT/w7Ngx+aoX65cRXWYJSUlkZ2dTXZ2NnPmzGHPnj2sWrWKwsJCClYWUl7qfbfpUjtzJL2HV2SnnwpxAb5Urs7r0lFeRkJ5GXEHd4NzpKSmkp0znJycHLKzs+nRo0dwGUWkTSgpgQEDovOLvSapq+O89X/is4w+fNBzVNBpRFpNXr9dPLu2H3voQJeiIhgdvdNGqqgOWJcuXZg4cSITJ06krq6O9957j6KiIlYWFrJu3Tpqd27AEpKp6tyf6lMG45I7hC2bVR8iYfdmkvaUQFUFZsbgM84gN+eSzwcYtmunt5CIhE9JCZx1VtApWl+/Nc/QeX8pr476EU6zSH3ZV6ZrUR/lkETozDF5md5MYAXkMfnVV1VUS8uIi4sjKyuLrKwsrrrqKiorK1m7di1Llixh+ZtvkrhzI7UZPajuNpiajr1bZ9EZ57wW6d2bSNj3ITjH8Oxsvj5lCtnZ2aSnp7f8c4qIhKCmxhuoeHmsj9mrq2P4ov/gs4zevN/7wqDTiLSq7D67iYuD/FOnMXnpE/CTnwQd6aSpqI5gycnJ5OXlkZeXx549e1i8eDH/WriQT99biiWmUtl1IEe6DmyZqfpqKkn4tITkTzfD4XLS0tP5+je/yaWXXkrPKJ83UkRiwwcfeIV1rA9SHLjij3Qu28irI+9WK7XEvPZJNZx9Nqz4bBysuBkOHoQoHZOlojpKdOnShWuvvZaZM2dSUFDAs//6F4UrV5K0o5iaDr2o7jaIupSmT1MXV32QhN2bSfxsG66uliFDhjJt2mWMGTOGpKSkVjgSEZGT0xZm/kg+sJu8Z+5gx4DRvNd3fNBxRMLi/PPhiT/2o/ZILfFvvglf+1rQkU6KiuooEx8fz8iRIxk5ciQ7duxg0aJFPLfoecpLXj7px0xOSWHS1EuZOnUqmZmZLZhWRKTltIWi+vy/305C5QGWX7MASnYHHSciaZ7n2JOXB/Pnt2NTu7MYsnSpimoJv+7duzN79mxmzZrFypUrKS8vb/JjpKSkkJubS2pqBE/hJyKCV1SnpcXuaooDCv5CVsETrJryI/Z1P0NFtbQZeXne7/x+VzHk1b8HG6YZVFTHgISEBEaN0pRLIhLbSkq8VupYnE7vtC3LGPunb/HxoAtZM/nuoOOIhFVWFnTrBkuTJ3PD6h9CWRlE4TS9KqpFRCQqlJTA8OFBpzhJx05nNmbM5xdPK1nOxPnTKO+aycvffoa6dolhDicSLDO45BJYtPBMal0c8QsWfLWovummYMI1gdaNFhGRiHfkCGzbFnv9qQe+9RhTHhxPZXo3loy4h+pV670CPELnFBZpLVOmwN598eTn3QYPP+xN9RNlVFSLiEjEKy2F2trYKaoTD+3jwj9cw7g/fYsdA8fy7F35VKS1gbXXRRowcSLEx8PzvefAJ5/AmjVBR2oyFdUiIhLxtmzxfsdCUd1l7xYu/+m59C96ksKp97Jk7hKq23cKOpZIoDp2hAsugEXv9ofMTHjtNXAu6FhNoj7VIiIS8ZYtg4QEOPvsoJM0T9b7L3BBwX9S2eFUFt75Jrsy84KOJC1JS6g3y5QpcOedxof3/pjeP57l/ccfOzboWCFTS7WIiES8F1+E0aO9KfWi1ZDNz3Dhiv/Hzm5D+cdF89i1vfqL/tPqQy3ClCne78VdroWhQ+Fvf4MPPww2VBOoqBYRkYhWVgbr1sGkSUEnOXlDN/2dUUW/obTXBSy58H4qk5u+Aq5IrBs82Ov58fQ/4uD66yE9HebPh507g44WEhXVIiIS0V56yfsdpYus0X/lXxm5ah7v9xrLK6PvoS4+IehIIhHJDG68EV59FYo/6w3f+Y439c9990F+ftDxGqWiWkREItoLL8Bpp0Vnf+rum19n3OOzKDtlGEtH3Y2L01Amka9YsODznznJj9M+6QgPvHwO9O4N3/8+pKbCRRfBc88FnfSE9L9bREQiVm0tvPwyXHpp9K2k2KlsIxPnT2P/KVm8NOqn1MVrUZdYtmCZBiW2hE7tq5k9ehPzXhvCz6etpNcpp3iF9VNPwbRp3hzWN94YdMzjUlEtIiIRq6gI9u6Nvv7UqfvKuOQ3l1CbkMILcxdTvfGDoCNJGxLtBf5t49fz29eG8NDSoTwwvQAyMrwp9q64AmbPhupqr2tIhFH3DxERiVhPPum1UE+YEHSS0KWU72Tyry8m6dBelsxdTEWXPkFHEokqfbpUcFXOe8x7bQiF27p5G9PS4Nlnva+tbr4Z5s0LNuRxqKgWEZGItGYN/Pa33iQAXbsGnSY0Kfs/YcqvLiJ9zzZeuHkRe3qfG3Qkkaj04JVvc1qHw3xj/sXsLE/xNiYlwdNPw2WXwdy58JvfBBvyGCqqRUQk4tTUeN0mu3aFBx4IOk1oTitZzuU/H+4V1HMXs2PQuKAjiUStrmlVPPu/X2LvwWQu/93FX8yql5jozV/9jW/Ad78L997rDb6IAOpTLSIiTWZmk4CHgHjgEefcL1ry8X/9a1i92vvs7BThK3in7ivjnBfvY8hr8yjv1p8XbnmePaXl8IkWdBFpzPH6f980ZhMAw3rt4Y/Xv8Y1j17EGWd4f2Bfcw0kJiZ6Axevvx5+8hNv8aRHHoG+fcOc/stUVIuISJOYWTzwX8DFwHag0MwWOufeaann2LIFpk6F6dNb6hFbhtXVklSxhw67t9Llo7X02vgCPd95ibi6WjaPuoH86Q9wJCUDSlVQi7SEK4aXMrTHM9z0ypXccAP8+7/DJZdATk4Cg2f8mR79p9Pxvh/SMXM4GZdPIP7yyyA3F3r3xiUkUlUFmzbB+vUwZAicd17rZVVRLSIiTZULbHXOvQ9gZk8ClwEtVlQvWOAN8A/nNHqX//Q8kg7uwVyd91NXB0cvuzriamtIrCz/0n3Ku/bj3THfZsP42zjQtV/4worEsOO1Xr/xBixZ4o1VfP55r6EaDJjm/wDPQPIzh6klnlriqTvmMe66q3WLanPOtd6jtxIz2w1ofqLQdQU+DTqExCy9v5qmj3OuW9AhmsPMpgOTnHM3+tevBUY4526pt89NwE3+1UHA5rAH9UT7+zPa84OOIRJEe34I9hhCOm9HZUt1tH8ghZuZFTnnsoPOIbFJ76826Xjtx19qoXHOLQAWhCdOw6L9/Rnt+UHHEAmiPT9ExzFo9g8REWmq7UCvetdPB8oCyiIiEhFUVIuISFMVAllm1s/MEoEZwMKAM4mIBCoqu39IkwX+FazENL2/2hjnXI2Z3QK8iDel3qPOuY0Bx2pItL8/oz0/6BgiQbTnhyg4hqgcqCgiIiIiEknU/UNEREREpJlUVIuIiIiINJOK6hhnZpPMbLOZbTWzHwSdR2KHmT1qZrvMbEPQWUQaO9eZWZKZPeXfXmBmfcOfsmEh5B9jZqvNrMafJzzihHAMt5vZO2a2zsxeNbM+QeRsSAj555jZejNba2ZvmtmZQeQ8kVA/881supk5M4u4KepCeB1mmdlu/3VYa2Y3BpHzeFRUx7B6SwlfApwJXBWJJwGJWo8Dk4IOIRLiue4G4DPn3ADgQeC+8KZsWIj5PwRmAX8Nb7rQhHgMa4Bs59zZwNPA/eFN2bAQ8//VOXeWc24YXvZfhTnmCYX6mW9m6cCtQEF4EzauCXXLU865Yf7PI2ENeQIqqmPb50sJO+eqgaNLCYs0m3NuGbA36BwihHauuwz4o3/5aWC8WTgXQT+hRvM757Y559bBV1ZejhShHMNrzrlD/tV8vPnNI0Uo+euvUd+eYxY8igChfub/X7w/CirDGS5EUV23qKiObT2Bj+pd3+5vExGJJaGc6z7fxzlXA+wHuoQlXeNi4Vzd1GO4AVjSqomaJqT8Znazmb2HV5TeGqZsoWr0GMzsXKCXc25ROIM1Qajvo3/zuxE9bWa9jnN7IFRUx7ZGlxIWEYkBoZzrIvl8GMnZQhXyMZjZNUA28MtWTdQ0IeV3zv2Xc64/cBdwd6unapoTHoOZxeF1ffpe2BI1XSivw3NAX78b0St88Q1U4FRUxzYtJSwibUEo57rP9zGzdkAHIqf7Uiycq0M6BjObAPwfYKpzripM2ULR1NfgSWBaqyZqusaOIR0YCrxuZtuAPGBhhA1WbPR1cM7tqffe+T0wPEzZGqWiOrZpKWERaQtCOdctBK7zL08HlrrIWf0sFs7VjR6D3/XgYbyCelcAGU8klPxZ9a5OAUrCmC8UJzwG59x+51xX51xf51xfvH7tU51zRcHEPa5QXofu9a5OBd4NY74TUlEdw/x+g0eXEn4X+FsELyUsUcbM/gdYAQwys+1mdkPQmaRtauhcZ2b3mtlUf7c/AF3MbCtwOxAxU4yGkt/McsxsO3AF8LCZRdS5PMTX4JdAGvB3fyq0iPnDIcT8t5jZRjNbi/ceuq6BhwtEiMcQ0UI8hlv916EYr1/7rGDSfpWWKRcRERERaSa1VIuIiIiINJOKahERERGRZlJRLSIiIiLSTCqqRURERESaSUW1iIiIiEgzqaiWiGZmbzfz/rPMbF4z7r/NzLo2J4uZTTOzM082g4hIW2dm48ysSUtrn8x9RJpDRbVENOfcyKAzHNWMLNMAFdUiIiEys/gIyNAu6AwSXVRUS0Qzswr/d3czW+YvGLDBzC44wX2uN7MtZvYGMKre9sfNbPpxHnuc/9j/NLN3zOx3ZvaV/xtH9/cvf9/M1ptZsZn9wt8228wK/W3PmFmqmY3EW/Hpl372/v7PC2a2ysyWm9ngFvinEhGJCP758Vb/8oNmttS/PN7MnjCzq/zz5wYzu6/e/Sr8RT4KgPPNbJKZbTKzN4HLG3nOsf45dq2ZrTGzdP+mNDN72n+cv5iZ+fv/2D9fbzCzBfW2v25mP/c/P75rZt3883mh/zOqoQwiKqolWswEXnTODQPOAdYebyd/+dL/wCumLyb0FuJc4HvAWUB/TnACN7NL8FqfRzjnzgHu92/6h3Mux9/2LnCDc+5tvCVW73TODXPOvQcsAOY654YDdwD/HWJGEZFosAw42vCRjVfYJgCj8Zb2vg+4CBgG5JjZNH/f9sAG59wIoAj4PXCp/1inNfKcdwA3+58RFwCH/e3nArfhfRZk8kVDyzz/fD0USAG+Xu+xOjrnxjrn/hN4CHjQOZcD/BvwSJP+JaRNUVEt0aIQuN7M7gHOcs4daGC/EcDrzrndzrlq4KkQH3+lc+5951wt8D94J/+GTAAec84dAnDO7fW3D/VbntcDVwNDjr2jmaUBI/GX6QUeBrqHmFFEJBqsAob7rcVVwAq84voCYB9fnKNrgL8AY/z71QLP+JcHA6XOuRLnLf38RCPP+RbwK7+FvKP/2OCd27c75+rwGmP6+tsvNLMC/3x9EV8+X9f/3JgAzPPP1wuBjHqt4CJfov5CEhWcc8vMbAwwBfizmf3SOfenhnZvYHsN/h+S/ld9iSe4T0OPAWAN3P44MM05V2xms4Bxx9knDtjnt6aIiMQc59wRM9sGXA+8DawDLsT7FvBDYHgDd630GzY+f6gmPOcvzOx5YDKQb2YT/Juq6u1WC7Qzs2S8bwiznXMf+Y01yfX2O1jvchxwvnPuMCKNUEu1RAUz6wPscs79HvgDcF4DuxYA48ysi/914xX1btvGFyfzy4CEerflmlk/vy/1N4E3TxDnJeBbZpbqZ+vsb08HdvjPe3W9/Q/4t+GcKwdKzewK/75mZuec4LlERKLRMrwuGcuA5YfM1ZEAAAF2SURBVMAcvJbifGCsmXX1ByNeBbxxnPtvAvqZWX//+lUnejIz6++cW++cuw+v68iJxqocLaA/9b89nH6CfV8Cbqn3PGoQkQapqJZoMQ5Ya2Zr8Pq1PXS8nZxzO4B78L5ufAVYXe/m3+OdzFfidROp3xqxAvgFsAEoBf7ZUBDn3At4XwMW+V8J3uHf9CO8ov5lvA+Eo54E7vQHz/THK7hvMLNiYCNegS8iEkuW43VtW+Gc2wlUAsv9c/QPgdeAYmC1c+5fx97ZOVcJ3AQ87w9U/KCR57vNH3RYjNefeklDOzrn9uF9HqwHnsXrXtiQW4FsM1tnZu/g/XEgclzmdVUSabvMbBxwh3Pu643tKyIiInI8aqkWEREREWkmtVRL1PLnMk06ZvO1zrn1QeQREZHWZWbXA989ZvNbzrmbg8gjUp+KahERERGRZlL3DxERERGRZlJRLSIiIiLSTCqqRURERESaSUW1iIiIiEgzqagWEREREWmm/w+IUxTGW/SX8wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 864x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 8))\n",
    "\n",
    "plt.subplot(1,2,1)\n",
    "sns.violinplot(x = 'is_duplicate', y = 'word_share', data = df[0:])\n",
    "\n",
    "plt.subplot(1,2,2)\n",
    "sns.distplot(df[df['is_duplicate'] == 1.0]['word_share'][0:] , label = \"1\", color = 'red')\n",
    "sns.distplot(df[df['is_duplicate'] == 0.0]['word_share'][0:] , label = \"0\" , color = 'blue' )\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualization: Word Common"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtIAAAHkCAYAAAATociHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3XuYZHV97/vPd1V199xnuDQjzDACMiAYPLIZRxO830K2gmbnZhI8aHwe1K07BnIRs/djziHbxNs2cZ+DZ8s+YkxyFDDRHUC8xWsUlZkBBGcAZxiYYbjMDMz90peq9T1/rLWqVlWtqq6qqerqXvV+Pc881XVZ1T+0n+5Pf/v7+/7M3QUAAACgM8GgFwAAAADMRwRpAAAAoAsEaQAAAKALBGkAAACgCwRpAAAAoAsEaQAAAKALBGkAAACgCwRpAAAAoAsEaQAAAKALxUEvoF2nnnqqn3XWWYNeBgB0bNOmTc+4+/ig1zGb+J4NYL7q5Hv2vAnSZ511ljZu3DjoZQBAx8xsx6DXMNv4ng1gvurkezatHQAAAEAXCNIAAABAFwjSAAAAQBcI0gAAAEAXCNIAAABAFwjSAAAAQBcI0gAAAEAXCNIAAABAFwjSAAAAQBcI0gAAAEAXCNIAAABAFwjSAAAAQBcI0gAAAEAXCNIAAABAFwjSAAAAQBcI0gAAAEAXCNIAAABAFwjSADoyPT2tb3zjG5qenh70UgAAGCiCNICO/Ou//qv++q//Wt/85jcHvRSgbT/5ibRmjXTw4KBXAiBPCNIAOrJ7925J0p49ewa8EqB9Dz8sPf64tHfvoFcCIE8I0gAwBMzsMjN72My2mdl1Gc+/28weMLP7zOyHZnZh/PhZZnY8fvw+M/sfs7/6E+deewsAvVAc9AIAAP1lZgVJN0h6vaRdkjaY2W3uviX1si+4+/+IX3+FpE9Kuix+7hF3f9FsrrnXwrD2FgB6gYo0gK44pb35ZL2kbe6+3d2nJN0s6c3pF7j7odTdxZJy9X9wEqD5sgXQSwRpAF0xs0EvAe1bJenx1P1d8WM1zOy9ZvaIpI9J+sPUU2eb2b1m9n0ze3l/l9ofSYCmIg2glwjSAJB/Wb/1NNRm3f0Gd3+epA9I+i/xw09JWuPuF0u6VtIXzGxZ5icxu9rMNprZxr1zbFcfFWkA/UCQBoD82yXpzNT91ZKebPH6myW9RZLcfdLdn40/3iTpEUnnZV3k7je6+zp3Xzc+Pt6ThfcKFWkA/UCQBtAVeqTnlQ2S1prZ2WY2Kumtkm5Lv8DM1qbuvlHS1vjx8XizoszsHElrJW2flVX3EBVpAP3A1A4AHaE3ev5x95KZvU/SNyQVJN3k7pvN7HpJG939NknvM7PXSZqWtF/SVfHlr5B0vZmVJJUlvdvd983+f8WJYfwdgH4gSAPAEHD3OyXdWffYh1Ifv7/Jdf8s6Z/7u7r+Y/wdgH6YldYOMyvEO77viO+fbWY/NbOtZnZL/KdGAAD6goo0gH6YrR7p90t6MHX/o5L+xt3XKvoT4jtnaR0AgCFERRpAP/Q9SJvZakUbV/7f+L5Jeo2kf4pf8nnFu8MBAOgHNhsC6IfZqEj/raQ/k5TUAU6RdMDdS/H9zIMBpLk9kxQAMH8w/g5AP/Q1SJvZmyTtiWePVh7OeGlmjWAuzyQFAMwfVKQB9EO/p3ZcKukKM/v3khZIWqaoQr3CzIpxVXqmgwEAADghVKQB9ENfK9Lu/kF3X+3uZyk6AOA77v77kr4r6Tfjl10l6V/6uQ4AwHCjIg2gHwZ1suEHJF1rZtsU9Ux/dkDrAAAMAcbfAeiHWTuQxd2/J+l78cfbJa2frc8NABhujL8D0A+DqkgDADBrqEgD6AeCNAAg96hIA+gHgjQAIPfYbAigHwjSAIDcY/wdgH4gSAMAco+KNIB+IEgDAHKvaUX62WelG2+c9fUAyAeCNAAg95pWpL/yFeld75Ke5IBdAJ0jSAPoiPO3ccxDTSvSpVLtLQB0gCANoCtmNuglAG1rWpFOniiXZ3U9APKBIA0AyL2mB7IwzgPACSBIAwByr+mBLFSkAZwAgjSArtArjfmE1g4A/UCQBtAReqMxHzXt4ODscAAngCANAMg9KtIA+oEgDQDIvaYV6eQJgjSALhCkAXSE3mjMRzNWpGntANAFgjSArtArjflkxh5pKtIAukCQBgDkHhVpAP1AkAYA5N6MB7JQkQbQBYI0gK7QK435hANZAPQDQRpAR+iNxnxEaweAfiBIAwByj82GAPqBIA0AyD0q0gD6gSANAMg9DmQB0A8EaQBA7nFEOIB+IEgDAHJvxh5pWjsAdIEgDQDIPSrSAPqBIA0AyD0OZAHQDwRpAEDuzXggC60dALpAkAYA5F7TinSr1o6vflV6+um+rgvA/EaQBgDkXscV6TCU3vxm6bOf7fvaAMxfBGkAQO51vNkwDKPHJib6vjYA8xdBGgCQex0fyJK8sFTq67oAzG8EaQBA7nV8RDhj8QC0gSANAMi9GQ9koSINoAsEaQBA7lGRBtAPBGkAQO51fCALFWkAbSBIAwByb8bxdwRpAF0gSAMAcm/GA1lo7QDQBYI0ACD3qEgD6AeCNAAg97o6kCXrcQBIIUgDwJAws8vM7GEz22Zm12U8/24ze8DM7jOzH5rZhannPhhf97CZ/ersrvzEzXggS7PWDirSAFogSAPoijeU9jCXmVlB0g2Sfk3ShZJ+Nx2UY19w94vc/UWSPibpk/G1F0p6q6QXSLpM0qfj95s3Oq5IN5vmAQApBGkAXTGzQS8BnVkvaZu7b3f3KUk3S3pz+gXufih1d7GkJHa+WdLN7j7p7o9K2ha/37wx44EsVKQBdKE46AUAAGbFKkmPp+7vkvSS+heZ2XslXStpVNJrUtf+pO7aVRnXXi3paklas2ZNTxbdK133SBOkAbRARTqHfvSjH+nKK6/UxMTEoJcCYO7I+hNCQ3+Ou9/g7s+T9AFJ/6XDa29093Xuvm58fPyEFttrXR/IQmsHgBYI0jl04403ateuXXrmmWcGvRQAc8cuSWem7q+W9GSL198s6S1dXjvnzDj+jtYOAF0gSOdQiW/8ABptkLTWzM42s1FFmwdvS7/AzNam7r5R0tb449skvdXMxszsbElrJd09C2vumRkPZKEiDaAL9EgDwBBw95KZvU/SNyQVJN3k7pvN7HpJG939NknvM7PXSZqWtF/SVfG1m83sVklbJJUkvdfd51XCpCINoB8I0gAwJNz9Tkl31j32odTH729x7Yclfbh/q+uvppsN6ZEGcAJo7QAA5N6M4++Y2gGgCwRpAEDuzTj+jtYOAF0gSAMAcq/rijStHQBaIEgDAHKPijSAfiBIAwByjwNZAPQDQRoAkHszjr9jsyGALhCkAQC5N+OBLM1aO6hIA2iBIA0AyD0q0gD6gSANAMi9rg9kIUgDaIEgDQDIvRnH39HaAaALBGkAQO7NOP6OijSALhCkAQC5R0UaQD8QpHPIG0ouADDcqEgD6AeCNAAg9ziQBUA/EKQBALk34/g7jggH0AWCNAAg92Y8kKW+8tysUg0AKQRpAEDudV2RLpcz0jcARAjSAIBcuPlm6QMfyH6u6x7prOcAIEaQBgDkwte+Jn3hC9nPdX1EeNZzABAjSAMAcsG9+d7AGcffNWvtkNhwCKApgjQAIBfCsHnmnfFAFirSALpAkAYA5EIYStPTzZ+TujiQRaIiDaApgjQAIBe6qkg3e4KKNIA2EKQBdIQj6DFXtQrSVKQB9ANBGkBXzGzQSwBqtNpsOOOBLGw2BNAFgjQAIBfCsPn5KYy/A9APBOkcolIIYBg1y8TSCR7IQkUaQBMEaQBdoVcac02SfbNyb9dHhEtUpAE0RZAG0BH+4oG5Ksm+WSPw2GwIoB8I0gCAXEhCclbu9TgoU5EG0EsEaQBALrRs7ZiOwjA90gB6iSANoCP0RmOuahWkXVbzmoaLCNIAukCQBtAVeqUx17SsSHv09drVHGlaOwA0QZAGAORCOxVpNhsC6CWCNAAgF1ptNgxViG45kAVAD/U1SJvZAjO728x+Zmabzez/jB8/28x+amZbzewWMxvt5zoA9B690phrmlWk01+qTTcbckQ4gC70uyI9Kek17v6/SXqRpMvM7KWSPirpb9x9raT9kt7Z53UA6DF6pDHXNJsjnQ7PVKQB9FJfg7RHjsR3R+J/Luk1kv4pfvzzkt7Sz3UAAPKvWUU6nYm72mxIRRpAE33vkTazgpndJ2mPpG9JekTSAXdPvjPtkrSqybVXm9lGM9u4d+/efi8VADCPtdPa0XZFOn0RQRpAE30P0u5edvcXSVotab2kC7Je1uTaG919nbuvGx8f7+cyAXSIHmnMNc02G7asSLdzIAutHQCamLWpHe5+QNL3JL1U0gozK8ZPrZb05GytA0Bv0CONueaEKtK0dgDoQr+ndoyb2Yr444WSXifpQUnflfSb8cuukvQv/VwHACD/TqhHmoo0gC4UZ37JCTld0ufNrKAotN/q7neY2RZJN5vZf5V0r6TP9nkdAICc62r8HRVpACegr0Ha3e+XdHHG49sV9UsDmKfokcZc02z8XToT1+Tl9NcwFWkAXeBkQwBdoUcac02zzYZNK9LJnSD+UdgscVORBtAEQRoAkAvt9EjXVKSTOyMjrV9IkAbQBEEaAJALHW82TJ4oxl2O6RYOWjsAtIEgDaAr9Ehjrul4/F27QZqKNIAmCNIAukKPNOaajivSyZ2ZWjuoSANogiANAEPAzC4zs4fNbJuZXZfx/LVmtsXM7jezb5vZc1PPlc3svvjfbbO78va1s9mwZY80FWkAHer3HGkAwIDFs/xvkPR6SbskbTCz29x9S+pl90pa5+7HzOw9kj4m6Xfi5467+4tmddFdaGf8Xcse6fqKdKEQhWsq0gCaoCINoCv0SM8r6yVtc/ft7j4l6WZJb06/wN2/6+7H4rs/kbR6ltd4wjo+kGWmHumkUk1FGkATBGkAXaFHel5ZJenx1P1d8WPNvFPS11L3F5jZRjP7iZm9pR8L7MTkpHTaadJXvlL7eNfj7wjSALpEawcA5F/Wbz2Zf1IwsyslrZP0ytTDa9z9STM7R9J3zOwBd38k49qrJV0tSWvWrDnxVTdx+LC0d6/02GO1j3dckZ5ps2GhUG3vAIAMVKQBdIXWjnlll6QzU/dXS3qy/kVm9jpJ/1nSFe4+mTzu7k/Gt9slfU/SxVmfxN1vdPd17r5ufHy8d6uvk+Tamuqymm82bPtAlvqKdBBEQZqKNIAmCNIAOkJLx7y0QdJaMzvbzEYlvVVSzfQNM7tY0mcUheg9qcdPMrOx+ONTJV0qKb1JcdYlubY+SDdt7ShXf+nraLNhEETPUZEG0AStHQCQc+5eMrP3SfqGpIKkm9x9s5ldL2mju98m6eOSlkj6UvzL0k53v0LSBZI+Y2ahouLLR+qmfcy6ToO0h66ku6WjHmkq0gBmQJAGgCHg7ndKurPusQ+lPn5dk+vuknRRf1fXmWatHU3H35WS3wGSlqT4rypJebpVkKYiDaAFWjsAdITeaAxaxxXpVBBu2SOd1dpBRRpACwRpAF2hVxqD0vFmw1L0hCmUd7LZ0CyqSBOkATRBawcAYF7pvCIdPVFQWe4FVVo7Wm02dI8q0hKtHQCaIkgDAOaVmXqkm03tKKisMCxUn2inRzoIqEgDaIognUP0sGI28HWGQUlybX2huK2KdKdTO4KAijSApuiRBgDMK81aO5r3SFeDNJsNAfQSQRoAMK90Ov4umiOd9Eh74wWtTjZk/B2AFgjSADrCtA4MWscnG85UkeZAFgBdIkgD6Ao90hiUjjcbltJTO1JP1G825IhwAB0iSAPoSBKgqUxjULo7IpyKNIDeI0gDAOaVzg9kiV5YVKm2It3OZkMOZAHQAkEaADCvnEhFOjNIt6pI09oBoAWCNABgXul2s2GgsPaadg5kobUDQAsEaQBdoUcag5LV2uFezcUN4+/KTTYbttvaQUUaQBMEaQDAvJJVkU4H5NZHhKefaGOOdFZF+vrrpUsv7f4/AEBucEQ4AGBeaVaRTnTdI11XkT6uhSo9fkBLj+2Wbryx+tzXvy498siJ/UcAyAUq0gC6whxpDEpWRTr9ca8OZPmTnX+oy3f+39nN2JOT3f8HAMgNgjSArtAjjUHJqki3DNLlJhXpGTYbPl06RU+Vxht7pAnSAGK0dgAA5pVOK9KZB7LceKO0bVv08YYNmW8YKlBZhdq+keR1k5PR4/xCCQw1KtIAgHml49aOuCIdHciSCr5JQC4Uotu6inSoQGUvNFakkxEh9eNBAAwdgjSArtAjjUGZabNhq/F3YVZrR9MgXVBZQfOB1bR3AEOPIA2gI/RGY9C6rUgHCrN7pIMg8w3LSUW62VnkBGlg6BGkc4igAyDPOt1s2PRAljZaO0IZFWkATbHZEAAwr9RXpG+8UTp6tPH5RM2BLK16pJttNqQiDaAJKtIAOkJvNAat05MNmx7I0tZmwyB7/J1EkAZAkAbQHVqIMCitNhsGwUxzpFNft8kbJD3S6cDs3nr8nUSQBkCQBgDML60q0sVi6yCdfm1Fs82GXqAiDaAlgjQAYF5pVZEuFluPv6u5bsbWDouCND3SAJogSAPoCr3SGJRW4+8KhYyKdNikIp2+KOMNw2bj76hIA4gRpAF0hR5pDMpMFekwrHuurrWjoSKd1SNdmdrR4kCWiYkT/C8BMN8RpAEA80pWRTqR1aWxcWN0W1R04f/8n21cFIYKPa5Iu2endirSwNAjSOcQf3LHbODrDIPSqrWjWKx9jaRK8E0q0g0XFQrap5MUlhpPNpQUHcqS/nqntQNAjCANoCu0dmBQZmrtkGqDdNIjHSiseW3ywf6pxVqtXbp98znpixTGPyIbDmWhIg0gRpAGAMwrrcbfJe3O6SBtYe3Ujvogvfv4Mh3XIj19aFH1ori1Q4qDdF3bhySCNACCNABgfmmnIp0egRd6k6kd8QeHpxc0vF8y/k5S46EsBGkAsWInLzazVZKem77O3X/Q60UBmPvokcagzHQgS/o10QujmxmDdLk2LJc92oRIRRpAM20HaTP7qKTfkbRFquzYcEkEaWAI0SONQUkybd2J3pKyg7TPVJGeGover1VFmh5pABk6qUi/RdL57s53DgDAwLSqSCeT7Gor0q2D9JHpKEiH6aEe9T3SWSNCCNLA0OukR3q7pJF+LQQAgHZ0HKTbrEjXt3YkFemw/lAWKtIAYp1UpI9Jus/Mvi2p8t3D3f+w56sCMOfRI41Bydps2GqOdJi0faju7PBKkB6N3jdMtSu1U5HmZENg6HUSpG+L/wEAPdIYmFYnG2YF6ZnG3x1pUpEuNxt/R0UaQKztIO3unzezUUnnxQ897O7Tra4BAKDXWo2/S1o7asffxc/N1NrB+DsAHepkaserJH1e0mOSTNKZZnYV4++A4UIlGoPW8RHhzTYbxhcdnhxteL+otSMVpBl/ByBDJ60d/03SG9z9YUkys/MkfVHSJf1YGIC5jR5pDEqnmw2Tr9WGI8JjlR7p+qkdzY4IJ0gDiHUytWMkCdGS5O6/EFM8gKFFZRqD0s7JhrVBOrpt3iPdRkWaIA0gQycV6Y1m9llJ/xDf/31Jm3q/JJwoAg76iUo0Bq1n4+9maO0oN5vawWZDALFOgvR7JL1X0h8q6pH+gaRP92NRAAA003FFuq5Huv6itnqkqUgDyNDJ1I5JSZ+M/2EOo2KI2cDXGQal84p0/FwcpCvX1QXpxjnSTTYbUpEGEGu7R9rM3mRm95rZPjM7ZGaHzexQPxcHYO6ihQiD0k5Funb8XeuK9JFKRdprnqs5kCVr/B0HsgBDr5PWjr+V9B8kPeCUogAAA9JxRTp+XdM50pMj8fvVVqTLoiINoLVOpnY8LunnhGgAmH/M7DIze9jMtpnZdRnPX2tmW8zsfjP7tpk9N/XcVWa2Nf531eyuvFGrIJ09tSN6MjkiPB2kQ5mOVo4IT32SZj3S7gRpABWdVKT/TNKdZvZ9SZXvHu5OzzQwhPidev4ws4KkGyS9XtIuSRvM7DZ335J62b2S1rn7MTN7j6SPSfodMztZ0l9IWqeo23hTfO3+2f2vqOp0s2HSI90wR9pdR7Sk8rKGzYZBRpBOv4ggDQy9TirSH5Z0TNICSUtT/wAMEQL0vLRe0jZ33+7uU5JulvTm9Avc/bvufiy++xNJq+OPf1XSt9x9XxyevyXpsllad6bOD2SJbi0Oxs2DtFUvqO+RJkgDyNBJRfpkd39D31YCYF5gk+G8tEpRe15il6SXtHj9OyV9rcW1q7IuMrOrJV0tSWvWrOl2rTPKqki3PCLcoyeDQJV+6ehx1+FUPah+mkeoJq0dCYI0MPQ6qUj/q5kRpAFIIlDPM1n/Z2X+acHMrlTUxvHxTq919xvdfZ27rxsfH+9qoe3IqkgnMivS8essqLuuLkiX457o5AXlMLogVNCY3hcsiII0f6EBhlonQfq9kr5uZscZfwcMr6S1I8xKMZirdkk6M3V/taQn619kZq+T9J8lXRGfHdD2tb02NSV9+cvZObXz8XfRbVD/K0GzHukwrPlNoWb8XXK7aFH0cU3pG8CwaTtIu/tSdw/cfaG7L4vvL+vn4gDMXVSk55UNktaa2dlmNirprZJuS7/AzC6W9BlFIXpP6qlvSHqDmZ1kZidJekP8WF994xvSb/yG9NBDjc9lVaSTj1sdyNLQIx2Gda0d1Yp0mPrxWDP+LvlEixZFt7R3AEOtkx5pmdkLJZ2Vvs7dv9zjNQEAesjdS2b2PkUBuCDpJnffbGbXS9ro7rcpauVYIulL8S9JO939CnffZ2Z/qSiMS9L17r6v32s+frz2Nq3z8XfRbSGo3r/xB8/XJTt+rMPVIVTVkw2zgnT9ZsN0kF5SrWoDGC5tB2kzu0nSCyVtVnW7hksiSAPAHOfud0q6s+6xD6U+fl2La2+SdFP/VtcoyavlcuNznY6/S9qRzLzmtfJQhxT9YbWo6UoLSMsgnW7tkDjdEBhynVSkX+ruF/ZtJQDmBVo6MBuSsJwVpDsef5e0fdQFaXPXkbi1Y3lwpHIAS3SqYaG6lqyK9MKF0S2tHcBQ62Sz4Y/NjCANQBKBGv3VKki3qkhn9khXT2SpYQp1KA7SywpHWvdIt2rtADC0OqlIf15RmH5a0cmGJsnd/YV9WRkAYGg1a+1wbx2kzaL2jvoe6UBlBfEvf2GqOfGwlmqsWFbRyq17pMt1ZXCCNAB1FqRvkvQ2SQ+odqQ9AAA91awinb7fKkinx9+5SyZvmIhtCnXYlmmsWFYQhjWtHQ1B2qdrPxFBGoA6C9I7453dAMBR4eirZhXp5L5ZBxXpUAoUVgZJ1/ZIL9GCkbIKU2HNZsOGHmnG3wHI0EmQfsjMviDpdqk6L4jxd8BwoTcas6FZRToJyCMj2SPuzLKfM3nla7capMO4taOkYMorJxl2NLWDIA0MtU6C9EJFATp9TDjj7wAAPTdTa8foaHT6YSLJt0GQUZFWXJGuD9KKpnYsGCkrsBlaO9hsCCBD20Ha3d/Rz4UAAJBo1trx2c9Gt0kY/sxnonw8U2uHyRUEtdfKXYd9SdQjbd7eHOnk9sEHo9s77pCejE9Mv/rqrv97AcxPbY+/M7PVZvYVM9tjZrvN7J/NbHU/Fwdg7qE3GrMhCdC1Y+wajwJPvhzT/dL1QVpyBXLJan/kmcJqj3SrirQVG1s7Rkai2/SuRgBDp5M50p+TdJukMyStUtQr/bl+LArA3EevNPqpWUW6WZBu2drhpkBh0tlReQ9z12Et0YJiWYFcZa/2SNdsNrSRxs2Go6PRbX3SBzBUOgnS4+7+OXcvxf/+TtJ4n9YFABhiM/VIJ0eBh3XDWLPH33k8/q7ulz8PNakFKhbCqEe62RxpKzaWvgnSANRZkH7GzK40s0L870pJz/ZrYQDmNlo80E8zVaSTIF2fb7OndpgCC2X14+/kKqugwFwFc4XKDtKhZYy/S4I0rR3AUOskSP+BpN+W9LSkpyT9ZvxYU2Z2ppl918weNLPNZvb++PGTzexbZrY1vj2p2/8AALOLAI3Z0KwiXd/aUd+6nH2yYfSkZ8yRDhUoMCmwsDr+Ln68spasHmkq0gDUQZB2953ufoW7j7v7ae7+FnffMcNlJUl/7O4XSHqppPea2YWSrpP0bXdfK+nb8X30WLn+JxAAzBPtBun6Hmmz6LnGI8IbK9LyUGUFMvNoakeTinRNkKa1A0DKjEHazD5mZu/OePwaM/toq2vd/Sl3vyf++LCkBxVtVHyzpM/HL/u8pLd0unDMjCANYL7qtLWjPkinr3O3KCJnzJFOWjui8XfVIF272TCjIl0oRDsbCdLAUGunIv0mSTdmPP4pSW9s9xOZ2VmSLpb0U0kr3f0pKQrbkk5rcs3VZrbRzDbu3bu33U+FGEEa/USLB/rpRCrSQdB4fHg0taN5a0fb4++S26zxIACGTjtB2t09zHgwlNTW/CszWyLpnyX9kbsfandx7n6ju69z93Xj4wwI6RTjydBPfH2hn7odf5ddkZbMvLG1I65IJ60dZa82XtcG6YwjwpMgzWZDYKi1E6SPmdna+gfjx47PdLGZjSgK0f+fuyfHie82s9Pj50+XtKf9JaNdQdDJXlIAmDtmGn/XbLNhEGQEaSnzQBaFoTzZbChvUZEeoSINIFM7SetDkr5mZm83s4vif++Q9NX4uaYsKll9VtKD7v7J1FO3Sboq/vgqSf/S+dIxE4I0gPmq2x5pKQrS9a0dkldydPWaKDibuYKgdrNhbY90xvi7rDl7AIZOcaYXuPvXzOwtkv5U0n+KH/65pN9w9wdmuPxSSW+T9ICZ3Rc/9ueSPiLpVjN7p6Sdkn6rm8UjW9K7evTo0QGvBHlESwdmQ7dHhCc90rUHskiBVSvSlWuSKra5ChbWnGxYU5FWix5pWjuAoTZjkJYkd/+5qhXkTGb2f7n7f0o/5u4/VPM+6te2tUJ0LAk6hw613Y4OdIxAjX7qdLNhImntmJioPhYdEd7YI520cgQmubnCZkE6a7MhFWkA6uxAlplc2sP3Qg8cPHhw0EsAgK70dvydao5qnozIAAAgAElEQVQIb2jt0ExzpFNvWL/ZkCANDLW2KtKYX8L475UEaQDzVaebDetbO2p6pJU9tSN5SRC4rFVFOqu1I+sIRQBDh91oOTQxOSlJOnDgwIBXgjxKevDDsGEqJtAzzSrS6fNQ0vdbVaRDt6giHQR111jlmsBcZVWDdMOBLFSkAWToZZCmYXIOCMNQhw5Fleh9+/YNeDXIo6Q3mh5p9FO7Fel2grQ8Gm/nQf1mw7hHutLa0awiXWCzIYBMvQzSn+rhe6FL09PTCuOfIMePzzjmGwDmpBPtkU7/wSScmpaFZdkz8Qm5W7ZEj6fH3yVBOgxbbzakIg0gZcYeaTO7XVGLWSZ3vyK+/bveLQvdGhsb06JFi3Xs2FGdcsopg14OAHRlpqkdrYJ0EDRWpM2qmw2TAF3JxCYFQVitPGdVpOvnSAcBUzsAtLXZ8BPx7X+Q9BxJ/xjf/11Jj/VhTThBS5Yu0bFjR7Vy5cpBLwU55vVzx4Aeanf8XbpQnHQbNfRIK+mRrm1HyqxIl8sNQTq0gM2GADK1cyDL9yXJzP7S3V+Reup2M/tB31aGrhXjnzDj4+MDXgnyjB5p9NNMrR1ZPdLNgnTSI10Zf5e8V2qOdGCqCdI1mw05kAVAE530SI+b2TnJHTM7WxJJDQDQc91sNkyCdP34u1BBNP7OktfWtngE5ioEYfMeaWXMkeZAFgDqbI70NZK+Z2bb4/tnSbq65yvCCUu+z7PZEP1ERRr9lATh+pyafH/L6pFuVpF2JQeyBJX7kioHsEStHanAnHUgS7OKNEEaGGptBWkzCyQdkrRW0vPjhx9y98l+LQzd8/jHxOQk//egf+iRRj+dSEW68WTD+Ihwq78m3drRYmpH1mbDdI90+pMDGCptBWl3D83sv7n7L0v6WZ/XhB6ZmJgY9BKQQ0mAJkijn050s2H9yYZSOuwmUzuqrR2tNhvWzJGuH3/nHj1XqPZUAxgenfRIf9PMfsP4e+6cVy5H3/CpSKMfkpYhvr7QT53OkQ7D2h7p2tYOU2BeaUcK42vKqdaOQuA1rR3JZsPRYrn5ZsMkPNcvEsDQ6KRH+lpJiyWVzey4ol/p3d2X9WVl6FoScKhIox+SIH3s2LEBrwR51m5Fup3WjtCD6LlKb0f1cSmZI63M1o6RQhgdHZ51RHhQPQkRwHBqO0i7+9J+LgS9E4bRN/wyVRL0QfJ1VWKTFfqo04q0VM21DePvJJlCeTL+zsOaaxtaO9zrgnTjZsNSGOjRo2dobXpRAIZOR0eEm9kVZvaJ+N+b+rUonJhFixZJkk466aQBrwR5tHjxYknSkiVLBrwS5Fmnmw3rWztqx9+ZzCQL6ivSta0dWT3SlSBdt9nw1nvX6gXf/KQOaDlBGhhibQdpM/uIpPdL2hL/e3/8GOaYQlyW4Yhw9EPyi9rChQsHvBLkWbOKdBKcszYbJhrH30VTO7xubEfNgSxJj3Tc2pH0SI8UQpW9sSL9zNGFmg6LOqrF9EgDQ6yTHul/L+lFHv9NzMw+L+leSdf1Y2E4cVSk0U/sO0Y/dTP+rllrR7XybMkD0TV1c6SzKtKjhbLK5bqpHWYqhdHzNW0fAIZOR60dklakPl7ey4Wg96Y5uhZ9kIy9CwkP6KNWmw2DQA0zoaXm4+8UT+2on/Vcu9kwe450w2bDeAHT5VSQpiINDK1OKtJ/JekeM/ueookdr5D0wX4sCr1x+PDhQS8BOTQ1NVVzC/RDq82G6YEZ7Yy/C2XxyYbxZsO4Il1O90ibZ55sWGntSOZFx5+oJkjzSyUwtDoJ0m+UdJOk/ZJ2SvqAuz/dl1XhhCQFGoI0+mHv3r2SpGeeeWbAK0GeJUG4fjhMs4p0q/F3UhSWqxfFUzvi1o5AriCQXIG8VJZlbTZMPkncQ5K0dpTSM6YBDJ1OgvTnJL1M0hWSzpF0n5n9wN0/1ZeVoXvxDxYOzEA/7Nm9O7rds3vAK0GezVSRrrQ7NznZMP3a0AOZPGNqRxSGLT4iXJK8HMpSmw0PHBvT6ORodTG0dgBI6WSO9HfM7PuSXizp1ZLeLekFkgjSwBCZLkW99/Tgo59abTbMau1IB+n0OSnRQSuWbCuMXpu8l5IeaVehEPf+T5cVpCrShSB1XHgYVirStHYAkDoI0mb2bUUnG/5Y0r9JerG77+nXwgDMTStWRNNgTjrp5AGvBHnWavxdodBeRbpcjg5ucZnMXB7U9oMkITyZ2iFJ5ZKrWBOkXdNJkE4q0mYqlaMLqEgDw62TqR33S5qS9EuSXijpl8yMQbJzUfwDgfFk6IcVK6LhPcuXM7gH/TNTRXqmHun0tR5vNqwff1d/RLgUVaRVF6RrKtJZrR1UpIGh1UlrxzWSZGZLJL1DUc/0cySN9Wdp6FbyR8zR0dEBrwR5RJDGbGgWpJN50VlBOj1HWkpXq6OTDRvG3yWbDeMjwiUpLNWOvyuYq+x1QdpM08yRBqDOWjveJ+nlki6RtEPRBI9/69O6cAKSnxUEafRD0htd5s/Z6KNmrR2teqQTQVB7bcP4u7qTDS3VI12erj3ZsFhI9UiXy9WpHWw2BKDOpnYslPRJSZvcvTTTizE4yQ+JQlKWAXooGX+X3AL90OmBLOk50s1aO9xq03dta0d0cUNrh3k0Rzr5JMyRBpDSdo+0u3/c3X9KiJ7b3F0HDh6UJO3YsWPAq0EeJWPvdu9mjPx8YmaXmdnDZrbNzK7LeP4VZnaPmZXM7Dfrniub2X3xv9tmY72djr+TGls70kE6MMniF1Sq2Ml15goKcZCemGreI52uSIdsNgTQWUUa88ATTzyh48eOSZJ+vnnzgFeDPNq7Z0/NLeY+MytIukHS6yXtkrTBzG5z9y2pl+2U9HZJf5LxFsfd/UV9X2hKpxXpZuPvoudMksuTFmlPTjZM5kh7ta96croxSNf3SLPZEECsk6kdmAe2bIl+LpaWr9a2rVs5xhk9d+ToUUnS0fgW88J6Sdvcfbu7T0m6WdKb0y9w98fc/X5JcyIVtqpIp8fftdPaESpQYKHMaivSoVKtHXFFujxZqumRLgRemTedrkgnQZqTDYHhRpDOmSNHjkiSyotOVrlc5nRD9FQYhpqcjH45m5yaVkiAmC9WSXo8dX9X/Fi7FpjZRjP7iZm9pbdLy3YiB7I0BumkVF2bvqs90qnWjoaKdJhZkWazIQCJIJ07yUiyYOKQgiDQ4sWLB7wi5Enyi9nSkShAT0xMDHI5aF/WUHnPeKyZNe6+TtLvSfpbM3te5icxuzoO3BtPdDNqkk1LdbtyWo2/a9raEfdI11+UBGwzKShGF4VTpYzWjuS0luqBLE1bO/grIDBUCNI5s2zZMklSMHFQi5csVRDwfzF6JwnSi0e85j7mvF2SzkzdXy3pyXYvdvcn49vtkr4n6eImr7vR3de5+7rx8fHuV6uZx9+1e7KhFAVpyaW61o5yqiJduWayNkgH1saBLMknevRRafFi6d57T+i/HcD8QcrKmWpF+mAlVAO9smTJEgVmevpYQWamJUuWDHpJaM8GSWvN7GwzG5X0VkltTd8ws5PMbCz++FRJl0ra0vqqE3cimw0zx9+ZUv0g1cel+IjwjIq0xQe1lNMV6fgT1UztSNL8rl1RCf1nP+vF/wQA5gGCdM4sXbpUkmQeasUKTp5DbxWLRZ166smSpFNPOVkjIyMDXhHaEY8tfZ+kb0h6UNKt7r7ZzK43syskycxebGa7JP2WpM+YWTL25wJJG83sZ5K+K+kjddM++qLd8Xft9UgHUU3aApnC1IEsjZsNkx7psgpRy4d1MLUjPqxIT7Zd7AcwzzH+LmfSVejlVKTRB6efsUp79j6r08/oZK8aBs3d75R0Z91jH0p9vEFRy0f9dXdJuqjvC6zTqiI9Otp6s2FNj7R7XJH2OEhXW8OTaRyW3mw4VZJGo3aOIIhCdmWz4kwHsiQN3U880ZP/DQDMfVSkc2ZsbKzy8YIFCwa4EuTVqaeOx7enDnglyLNWQdqsg4q0x6E4DtKBwtSBLNEFgbkKxbhVY6qcau2oq0jXHMiS0SNNRRoYOgTpnNm2bVvl44cefniAK0Fe7Xp8Z80t0A+tNhum50jPuNmwXI4DsylMKtL1B7JItRXpOEgHgVUDe/LJaO0AkEKQzplNmzZJkiaf80I9+cQT2sPpc+ihqakpPfLIdknS9u2PcuAP+qZnJxumNg7WV6ST1o4g8Oo1lSBd7ZGOXlvIPJAlM0jT2gEMDYJ0zmzadI980ckqnXJOfH/TgFeEPNm+fbtK5bJeunJSpXJZjzzyyKCXhJzqZrNhEoZrKtJJkJaqPdKVa1Lj74I4MMetHWUrxi0kqSCd6pHOPJAl6ZF++mkOaQGGBEE6ZyYmJxQWRuWFUUnM+UVv7dixQ5K0/rSoEv3YY48NcDXIs5kq0vWbDdNnotQH6fRmwyA9tUOp1o44MFcq0laINxtmBOlmR4QnFelyWeKvgcBQIEjnzPPOOUfFif0Kju+TJJ1zzjkDXhHy5IwzzpAkPfBsNPZu1Somd6A/ejb+LnW4SrOpHUE8L1qSwul4s6Elmw2T1xaan2xYv9lQok8aGBIE6Zw599xz5dOTKh7YJYkgjd4699xzFZjpu08ukJnp3HPPHfSSkEPujRPlEmGYvdlQqlap63ukoyPCG3ukk6O/rSZIVyvSDT3SYZia2pEcyJJRkZYI0sCQIEjnTFIhLB5+UsuWr+DkOfTUwoULtWZNdNL0mjNXa9GiRQNeEfIoyaVmUW51r30ua/xd8rjUrCJtlYp05ZpKRVqpHukwPpClmL3ZsH5qRzCSnfrZcAgMBYJ0TrkVpNSfMIFeec7pZ9TcAr2W5NLk4Mx0e0d9RTrd2lEsT+qqa05S4ZFfVK+Lx9+lD2Sx1MmGprAmMNf3SDdsNqwcER4HaSvS2gEMMYJ0zkwn38iDokrTpdYvBjrk7nr4oQclSQ8/9GBl0xbQS0kuHR2tvZ98nLXZ0F0qlCY0duyAgiej1rba8XdRgaGmtUOBgrjgEARt9EhnzZG2jNaOZcuoSANDgiCdM8ePH5ckeWFUU1NTKtU3GAIn4PHHH9f+Awd1zrJpHTh4SDt3cigLei/JpVlBuuX4uzgUF7xUvS6Z2iHJzaLWjvi9ygoUWPTJCklFuhRW5khHgX2GOdLpinTy/fa5z6UiDQwJgnTO3HvvvbLiqKZPPVel0rS2bNky6CUhR+6//35J0pvPOl5zH+ilVhXp+iCdPtkwUByKw+nqdTUHsmRVpKNrgro50tWKdOP4u9AK8nijYmZFmiANDA2CdI6EYagf/uhHmlp6hkorzpSCQHfdddegl4Uc2blzp0YC6UWnTGukUJ0rDfRS9xXpJEhPVa9LNhuaUkeERxeECipBOWnhSFo7qpsN4zWkNhtOa6SynoYgXShIq1fT2gEMCYJ0jmzdulX79+1TacUaqTCq0pLT9W8//NGgl4UcWbJkiaZDaSqUpstiKgz6IgnOzTYbJv3RyVQPKd4DmFSXw1LltdXxd6qOv4uuUFmFakW6yRzprJMNp220utb6zYbFonTGGdKzz0ociAXkHkE6Ryobv5JSjZnC9JBV4AQtW7ZMkrT7eDRfbPny5YNcDnKqndYOqTFIV3qky6nWjnK54UAWd8k8qlQnvdGV8XfTdT3S8fOVEwzDMPo4WWu6Il0qRek/OaiI9g4g9wjSOXLeeedp+YoVKu7fKZWnVTz8lF7+sksHvSzkSBKknz5WqLkP9FI74++kKFBnt3bU9khH4++qFelkt2FZhWoVu36zYV2P9LSNVjYbpivSJRupbe0YGYkq0hJBGhgCBOkcCYJAL7v0Uo0efkLFAzulsKxf+ZVfGfSykCMnnXSSJGnn4SjJrFixYpDLQU41q0inxjhLim6zNhsGcUU6Pf5OaqxIl1VI9Ug3Tu1I90hXAnMYatrTFenUEeE/+1lUlU72pvz93/fwfxUAcxFBOmde8IIXyEtTUVVa0gUXXDDgFSFPTjvtNEnS1oNRqXDlypWDXA5yqn6zYTJVLsmrSUW6sbWjeUU6MMkVbzaUZPFmw0JyTd0c6bJqD2QppSrSJUttNkwfEZ6Uy5cuje4fPtyr/0kAzFEE6Zx58sknJTOVl0d/Wnz66acHvCLkyfj4uCTpFweLNfeBXmq22TAJ1EmPdBDUVqSTNo1COWNqhyQPotaOMH5tWYVKUK6MuSt5ZmtHyYpNNhumKtLlchSki8XahQPIreLML8F88uijj0oLV6i86JTK/ec+97kDXhXyYnR0VCevWK59Bw7qpBXLNDY2NuglIYfqK9Jf+IK0cqU0MRHdz9psGIZS4DP3SKfH32W2diRTOzJbO+Ie6Wbj75IgnZTMCdJA7lGRzplfbN2m6bHlCheukMy0bdu2QS8JOeLuCuPkEoYhR4SjL+or0unOCanV1I7owqweaasbf2eqbe2oHBFe6ZEO6lo7RipzpGumdqguSKfPLydIA7lHkM6RJ554Qnv37FZ56XOkoKhw8bg2bto06GUhR5599lkdOHhIKxeWdfDQET3zzDODXhJyqH6zYTtBWkr1SKdbO8pluUxq6JGube1IxuAlVeyyFepaO1KbDdMV6WS+dHJxuiLN+FEg9wjSObJhwwZJUml5NMN0etkZevihh3SYDS/oka1bt0qSXnH6RM19oJfqWzvqg3TW+LswVHXjYCk6CKX2iPBqRTr0aLNhzYEs6Yp0vBGxprUjaBKk0weyJEGaijQwNAjSObJhw0ZpwVL5WDTbt7Rsldxd99xzz4BXhrxIgvPLTp+USbQOoS/qK9JJWE4eT4+/qznZ0LNPNgwVyFTfI117IEvliPByWDnEpfZAlpHq1I6a1o5CY2uHWRSoCdJA7hGkc2TRooWy0pQU/xAJpo5KkhYsWDDIZSFnLP4niR5p9EWzinTy5ZYef1c7RzoKrvUVaY+ryzKLK9Amk2pbO4LojUIF0uRkqiI9Q2uHZYy/k6JATZAGco8gnSO//uu/Li9NauSZrZK7xvZs1hmrVunFL37xoJeGnFi7dq1c0r89NSaP7wO91myzYfJ4082G3tgj7eX44qSKLY9bO+JZ0aqtSJdVkCYmKkG6YbNhXUW6pIzxdxIVaWBIEKRz5MILL9SFF75AC/ZsUeHIbgVH9up3fvu3FQT834zeOO+88yRJ339yQc19oJdm6pFuFqQLSUU6I0gn7SBJcFY8taNh/F1ckS5njb+LB1nXbjbMGH8nEaSBIUHCyplXv/pV0sQhFZ99RJL0yle+crALQq6ccsopWrpksfZMFLR0yWKdcsopg14Scqjd8XfpzYY14+9KU5XXh9P1YdblbqkjwuO+6nSQjivSNT3SVmwSpDOmdkgEaWBIEKRzZseOHbKRMZVWnFm5D/TKXXfdpcNHjurksbIOHzmqu+66a9BLQg4122w4U0U62WyY7pH20CuvlaLWjugUxLgiraRHOh5/V9faUbPZMA7SSWtHYFEYryw43SNNkAaGAkE6Zx544OeaXnSqwsWnSZI2b9484BUhL44fP65P/e3faNWSUB996QGtWhLqU3/7Nzp+/Pigl4acSQJz/RHh9ePv6jcbZrV2hKWw+mJFrR0eX5B1RHizinQ5XZH2KEiPFaN501SkgeFFkM6RY8eOaefOHSovPk0+skBauFxbtmwZ9LKQE//wD/+g3Xv26u3nHdZYQXr7eYe1Z+8z+vu///tBLw050+xAlnY3G6ZbOxoq0haNv6u2djQP0maSxZ+rtkc6CtILRkq1Felk/J0UBWkOZAFyjyCdI3v27ImOcF6wXJJUGl2mp556esCrQh4cPHhQt9xysy59zoTOXxGFifNXlPSy50zo1ltv0YEDBwa8QuRJ/WbDdFiWWgXpePTndLW1I6lIp1s7QpnMs1s7GjcbJq0d1Yp0yZMgXVbZqUgDw4wgnSN79+6VJPnoIklSOLpIe+LHgBMxNjamsdHRygSDRGDS6MiIxsbGBrMw5FK7Fen6zYZJa4eVS5UxzmG5dtZ5ciCLKcxs7Uj3SNe3dhw6En3iB/ecLEmaLgdsNgSGHEE6R5555hlJUjgSBWkfWaTDhw5qampqkMtCDixYsECvevVrdPfeBZqIinKaLEt3712oV7/mtVq4cOFgF4hcaRakW202DMPqZkNNT1dybGX8XRK+44q0XLWtHUGT1o54TSWNKChPR28ft3aMBGFjawdBGhgqBOkcWb48aukIJg5WbhcvWcocafTEZZddpsmStHFvlG427hnVRMl12WWXDXhlyJv6zYatgnTNZkOPg+v0tIIgHn+XtHaoek1Ska6dIx1/DgXS9LRCr53aUbaCCmG8iTFu7SgWQpUVNB4RniySHmkg90hYObJu3TotWrxYI/u2S+VpjR7cqde8+lUqFoszXwzM4KKLLtLpK1fq7j1RG8fde8e08rRxXXTRRQNeGfKmm4q0VJ0jrVKpoSKdNEmbQrkaTzZMjggvqxDfxkE6qM6RLtRVpIsFZ/wdMOQI0jkyOjqqV7/qVRo9sFPFfY/KyyW99rWvHfSykBNmpkKxoJE4WIwErmKxKDOb4UqgM/UV6fo50unxd+nn0hXpJMcmPdJWCdKKD2Txuqkd8fvEPxarPdLR42UVFYTJZsNoASOFMPqYHmlgaBGkc+ass86Sl6dVPLhLkrRmzZoBrwh5US6XtXv3bp22MAoNpy0MtXv3HpUJC+ix7sffxQ38qdaOxgNZwrgG3eKIcKnS2mGpzYbFuLWjVN8jTZAGhhZBOkfcXbfddrt88amaWnWxJOmrX/3qgFeFvHj22Wc1XSrrtIVROBhfUFapXK5scgV6pX78XboPWmpxRHiy2TDV2lE9kCW+sdqKdBKUa8bfSQpltUeEq9oil1Ski4VQZY8XEYa1QToZGwIg1wjSObJp0yY9/vhOTZx2gcKFJ6m87Ax9+Sv/S6V49ilwInbu3ClJGo8r0ivjQJ08DvRKkj/rNxtmVaTDMBWwUxXpSo90XJFWurVDzQ9kSXqk648IL1s6SBdl5iqYV16fzJimIg0MF4J0jnz961+XjSxU6eRzJEmTKy/U/n3PatOmTQNeGfLgnnvuUcGk5y2LNlyds6ykQhA9DvRSp5sNkyCd2SNdOZAlbvFQKI/jdLq1w+p6pMuVqR3R46UkMMcfF8xl5lFFOv6c0SII0sAwIUjnSKlUko+MSUH0jdzjedJUpNELG+7+qc5dXtI/b1+kf/zFIi0oSmuXlbRhw92DXhpyppPxd+kgHSj+XlcqVcffVQ5kCWquSSrS6a2yQeA1mw1rKtLp1g4VVQhCBVatYCuZ158+IpwgDeQeQTpHFi9eLCtXD19JPl68ePGgloScOHDggLY9sl2/dNKUdh4paueRKFT80slT2rbtEY4IR0/1siKddSCLK5CpdmqHFGffpLXDo6kdlhGkyx6oELgKQaoiTWsHMJQI0jmyZMkSaXqq8lPFSlPVx4ET8Oijj8rdtWpJbTBYtTgKD9u3bx/EspBTzTYb1o+/SzYbVnukZx5/J3n0eo9bO4JqkA6C2s2GtRXpVGuHF9tr7eBAFiD3CNI5smrVKnlYkk0ekiQFx/fJzLRy5coBrwzz3YUXXqglixdpw57Rmsc37BnT4kULdeGFFw5oZcij+op0EpRn2mxYCBtbOyqbDWOVnujM1o7G8XeVHmmrndpRCFyBucpOawcwzAjSOXLJJZdIkooHn5AkjRx6Quc///launTpIJeFHBgbG9Pr3/Cr2rB3TEnL6ZFp04a90eMLFiwY7AKRK+2Ov2vokc6Y2lHpkU4l5lCNB7Ik71vZbKhkjrQqbSCJskebFKMeaTYbAsOMIJ0jq1at0mkrn6PCoSel0qSCo89o/YtfPOhlISfe9KY3qRRKB6eibxs/enpM02H0ONBLnYy/q+2RjoO0uwoFrxl/l3R2RD3S1akdVtMjbSpb9ElDt9TnqQ3S0ypWKtKlMH6cIA0MJYJ0jkxOTioITIXScVlpQnLXdPLNHThBW7dulZlpohQlkscOR8eDb926dcArQ96c8GZDRQes1B7IYqlr0geyVD9vEEhhIWrhSFo7JMWV58bNhoy/A9DXIG1mN5nZHjP7eeqxk83sW2a2Nb49qZ9rGCaf+9zn9PRTT2nijH8nX7Bc06ecq1tuuUUPP/zwoJeGeaxcLuvTn/60PvKRj+iCk6a1clEUDn5/7VFdeNKUPvrRj+qGG25gzCJ6pt3xd8lmw8rjXv0aDMyjHuly8mQ1MbuiedLpOdLJ+4VBHKRVrUgHdRXpkhfi1g5X2eP3TXqk0zsh2WwI5F6/K9J/J+myuseuk/Rtd18r6dvxfZygzZs36+ZbbtHU+HkqL18lSZpY8xKFIwv14b/6K01NTc3wDkCjw4cP64PXXadbb71Vr199XH/ywoMqxLlhyYjrT154SK9ffVxf+tKX9MEPXqfDhw8PdsHIhaSQW4yLwEnFuX5qR8Nmw6yKdF2WNfOaHul0RbpQkMpBVAYP3SrPmbnKlu6RLlTnSLcaf+dOVRrIub4GaXf/gaR9dQ+/WdLn448/L+kt/VzDsPjsTTdJIws1eeb66oPFMR0/86XauWOHvvOd7wxucZiX9uzZo//4nndr08YNesf5R/S2846pWPcdoxBIbzvvmN5x/hHds2mj3vPud2n37t2DWTBaMrPLzOxhM9tmZg0FDDN7hZndY2YlM/vNuueuiv+KuNXMrur3WtOBOV3YbdbakSiE1Va2SpAuJ0eAJ0eEu+RRj3TmZsO4Il1uaO2ofvGX4taOIN3aUV+RLtT1TgPIpUH0SK9096ckKb49rdkLzexqM9toZhv37t07awucj56zcqUsLEmpqomkyk8ZRuChU/fdd58e3/WE1o1P6lVnTLZ87avOmNSLxye164knde+9987SCtEuMytIukHSr0m6UNLvmln9zMKdkt4u6Qt1154s6S8kvUTSekl/0e+WvLdeaa4AACAASURBVPSmwiCo3k9uq5XiNlo7Kk9Wr4kq0klrR/Xz1rR2eG1rR1g3taNgdUE6Cczp8XfpxwHk0pzebOjuN7r7OndfNz4+PujlzGkve9nL5KUpFQ4/VfN4cf8OLVm6VBdddNGAVob56nWve52uuOIK/XTPmL60fVFN5S/NXfqn7Qv1k91juvzyy/WGN7xhdheKdqyXtM3dt7v7lKSbFf11sMLdH3P3+yXVN/b+qqRvufs+d98v6VtqbNnrqXRFOl11do/up3uka1s7SgrjIwwLllSk4zdNVaS9prWjriKdbDZUuiLtKtX1SEebDdW8Ih3UBWwAuTSIIL3bzE6XpPh2zwDWkDuXXHKJxhYsUHH/juqDYajRQ7v0sksvVbFYbH4xkCEIAv3RH/2RLr/8ct2xY6H+afvChjCdhOjbdyzSm970Jl1zzTUKgjn9+/mwWiXp8dT9XfFjPb22V39FTCrPhULtAYHlcjWfRp+vbo50WFJ5dGF0bRDWjr9LrpFLXj2QpeGI8Mr4u6CmR7q2Il2oHBEeeiCXsnuk048DyKVB/MS7TVLSY3eVpH8ZwBpyZ2xsTOc+73kKJg5VHrPShLw0qQsuuGCAK8N8FgSBrrnmGl1++eW6fcci3b5jYc3zt++ohuhrr72WED13WcZjTf7G0P21vforYn1rR7pHulWQLnhJ5WJ0OFCgqCJdnm4cfxem5kg3tnYU4udV2yPtta0dydQOKT7EhR5pYCj1e/zdFyX9WNL5ZrbLzN4p6SOSXm9mWyW9Pr6PHli4cKGC1K51xcflcuocTsT+/ft14MABSdKjh2v/srEjvn/wwAHt379/1teGtu2SdGbq/mpJT87CtV2pb+1oFqST8XfpI8JLo4uij4NQYShNx3PP05M+qq0dxZpfE5Ie6eR0w9rxd42bDZO2kLIK2XOkJYI0kHN9/Xu/u/9uk6de28/PO6zGxsZqgrTFQXpsbGxQS8I85u6688479f98+gZNHj+m3zrnqH5tzYQ+dt+yymve84LDOmvnAv2vu36o//2eTXrPf3yv3vjGN8osq4iJAdogaa2ZnS3pCUlvlfR7bV77DUl/ldpg+AZJH+z9EqvSmwqTsCxFQbqQ2k/dWJGeVilp7bBQU2VpcjrumY6v86AgefT1LSmztSMJ0pWKdFB3RHgYqBCUKtcSpIHhxd9hc2LLli265957FRZGK495/PGXv/wVHTt2bFBLwzy0a9cuXXvNNfr4xz+uVSOH9F/X79flZ000jL8rBtLlZ03ow+v368zRw/rEJz6ha6/5I+3atWswC0cmdy9Jep+iUPygpFvdfbOZXW9mV0iSmb3YzHZJ+i1JnzGzzfG1+yT9paIwvkHS9fFjfZNUnpMgna5Ip39Ha5gjHZZUHon+ApdsNpyaji4oFqIXRUHaU0G6+n7V1o76irRqg3SltSO6X1KxeZCmRxrINYJ0Dtx///269o//WMfDoo6d9fLK4z62RMfPeaXuf+B+/emf/pmOHj06wFVivvjRj36kP3jHO/TQ5vv0jvOP6IMXH9Tpi1qf0PacRaGuu/ig3nH+ET20+Wf6g3e8Qz/84Q9nacVoh7vf6e7nufvz3P3D8WMfcvfb4o83uPtqd1/s7qe4+wtS197k7ufG/z7X77WmNxXWB+lWFenASyqNRBXpQFFrR6UiHb+fxz3QNj1VeY9EEEihFRoq0gUL6042rM6RluKQnfRIM/4OGCoE6Xnu3nvv1Z/+6Z9pQmM6cv6vyceWaGznTzS28yeSpNIpz9Pxc16lzVu26Npr/5iT5zCj73znOxq1af31+v169arJmopdK4FJr141qY+s36cxm9a3v/3t/i4UuVUuZ5+03WyzYaWnWmWVR6qtHeWyNFWKLhgpJhXpolym0ckj0XsEdePvrFgJzZUgHbimw2onZDkrSNPaAQwlgvQ8ViqV9Od//uea8IKOnH+ZfHSxJCk4tk/BsepfXksnn63jZ79CDz/8kD7zmc8MarmYJxYtWqSCmU4ea12FbuakMVchMC1evLjHK8OwSFee05sN68ffpfunpagKXaq0dkRBejJu7SgE0Zt4IdpMODIZ/YUu/Yti1CNdbKxIB7VzpMthdCALmw0BEKTnsWKxqJe//OWyqaMKjrWemFA89IRkple84hWztDrMV4sXL9axUruT0bIdm3aCNLpW39qR3mzYco60QpXjqR2BynGPdHRBMQ7SYVCQyzQydaTyHon61o5Kl0YQqlQ3/i6qSMf3CdLA0CJIz3PXXHON1jz3uVr86PdlU9k90MW9v9DIM1v1tiuv1Pr162d5hZhvFi1apOmyVOquIK1SKE2VRZBG19IV6fQR4VlBOgyrFWuTV3qkC0mPdCl6o2LcwuGFqLWjGFekG042zOqRDlylMB2kTUHgCoKMinT6NwCJzYZAzhGk57mFCxfqL6+/XqOBtOjRHzQ8bxOHtGjnT/Siiy/W29/+9tlfIOadNWvWSGqcGd2ux+LrVq9e3bM1Ybh0vdlQYWpqR22PdDK1Iwyi8XZJRTpIlaSj1o7GIF1s0tpBjzQAgnQOjI+Pa9GiRVL6MJYKl3tZa848U4X0TyCgiUsuuUSBmR54dqSr6+9/dkSBmdatW9fjlWFY1Fek060d9ePv6oN0tSIdtXZMlmpbOyoV6anj8XvUVaRVqGw2rLZ2uKbrKtKFIKxt7YindvwfX3+p7npkJUEaGBIE6Rz4x3/8R+3fv0/HVze2bfiC5Zoaf75uv/12bdu2bQCrw3yzdOlSXXDB83X/vu4O8nlg/5ie//zztWzZsplfDGQ4oYr0aO34u6m4tSPZbBjGmw2Lk8fi968m8+atHWHtEeFhdLR4TUU6buH4q29eoi/e/TyCNDAkCNLz3I4dO3TLrbdq+pRzFS45LfM1k6v+nVQc06f++39XiX49tLB3717dcccdOnL0mB49VNDxUmcnFB4vSdsPFnT02HHdcccd2rNnT59WijxrNv6uXG7cHNhYkY5bO5LNhsn4u7i1w8zkCjQynWw2rO+RDjI2G7qm4yAdyuRKKtK1c6RdpulyQc8eXcCBLMCQ6OsR4egfd9cdd9yhGz79aYUqaHJ1iz+jF8d0fNUleuD+H+ld7363PnjddTr33HNnb7GYs0qlkh544AHdfffd+umPf6ztjz0mSTppgfT61cc1GnQ2vWM0kN5w5nFt2L1Dn/jEJyRJ55x1ll7yy7+s9evX66KLLlKxyLcdtNZs/F1WRTp9sqHJVYqndtS3dlSOAreoD7o4PVF5j0ShEIXixgNZXOUwemxaUctT1hzp6cICqSztOzpGRRoYEvxEm4eeeuopfexjH9e9996j8rLTdXzty+TxD49mpsfPlxfGtH3nj3X1u96lt115pa688kqNjHTXB4v5a8+ePVFw/ulPtWnjBh07PqGCSeetKOl3njepF54yrdWLyzUBo12FQPr9tcf0e+ce0xNHC/rZsyO6f99W3XrLY/riF7+oRQsX6JJ1L9ZLXvISrV+/Xqedlv1XFAy3Vq0dM46/q5xsmFSkCxrTROXr2SSFChrCcvK5yqke6cpmw0KokkevL8U/NtNzpEsqSmGo6dEoSP//7L15lBxnefb9e6qql5mefV81o12WZMuLLIFt2bKxsDEyCQE+m5iXwAnbl0AwX968hCVAWMIaiF9MiIFw4mDAGJsYIxuM9w1bi2VJlmTtmhnNvq/dXd1V9Xx/VO/do82z+/md06d7uquqnx6Nuq6++rrve+B0QjoUchecf/r3bIVCMT9QQnoe4TgOv/3tb/mP/7iLiO0QbrqCaOVKzlbxWGXNjBbV4G/dzt13380zzz7HZz/zj6xYsWKaV66YTaLRKPv372f79u1sf+lFTra0AlCeBxtLw1y0LMLq0ih5U/huIAQ0FNg0FNi8vSlMyIKDQx72DYTZt+s5nnvuOQAWNzex8U1vZuPGjaxdu1Z9sFMApy82PLOQTkY74u3vvERT9pFpBYWZQjoqskW2LpLt79Idaffx+LEimvvcadGOTCH9/vfD6Cg8+ui5/2IUCsWcQwnpecTjjz/OHXfcgTT8TFzwDqT/PIq5DD/hpddgF1bTcvJP/O3HP85vH3zQ7fqhWJB88AMfoL2jA4CA4fCeJSEuqYhSf56u8/mQZ8BllVEuq4wi5QQdEzqv9Hv4/amT3HtvK/feey91dXX84he/mJkFKeY0mY50ah/pMxUbJrp2yJgjbev4MJP7AFJoieJBkSLMdR1smWsgi8SWZ4h2kBTSk0Y7gkHYtg2WLj2/X4xCoZhzqGLDecSGDRvYuPFNCCtM4PiTaGM9534Q6eDp3k9e+048Hg8f/MAHlIhe4Nz63vdy8bp16LrGhKXxu7YCfn0inyc7fPSGZvYtoC+k8WSHj/tP5PO7tgLGoxq6rrFu3UW8973vndG1KOYumY70ZNGOnMWG3sz2dzqeLEc6paAwc7JhjtiH27XD/SEe7dBS+0gLV1xHdVdIj4R8yUmIqcWGTz0F4TCYSWGvUCjmN8qRnkeUlJTwjW98neeff55/+7c7GDj0MJGKFZiN68Hwn3F/bbyX/LYXERMDXL5xI7d/8pPU1dXNwMoVs8nWrVvZunUrwWCQ3bt3s3PnTra/9CKvHHE7atQEJBeWhrmwLMoFpVF8U9hu3LTh0JCHfYMeXh3y0z3hipGaqkreepNbgHjJJZeoKYiKNFK7dpxrsaEdey/UpBvtiNg6PhFJO76DNmm0I7eQllixYsMI3sR9cVFvax6wISKSLSMHzQBVkO5IP/ywe62EtEKxYFBCep4hhGDTpk1cdtll/Pd//zf33XcfvpE2xlfehJNXMul+nu79+E/toKy8nNs//WU2bdqEmKnv9RVzgvz8fK666iquuuoqpLyd9vZ2V1Rv386zr+zmsfYohgYrS6JcWBbhorIoDQW5hvycno4JnX0DHvYNejk87MFywOf1cPEll/KeDRvYsGEDDQ0N6u9PMSmTFRtmtr+LRzvij2s4RA4dAUDv6cQORzCHg3hICmlNgBQiKaQzHO5UkR1fg6Elu3ZEcMWyrjmJYsOEkNaShkaWkJZSCWmFYgGihPQ8JRKJMDY2hiPl2QkSIUAIotEoo6OjyLPdT7EgEULQ2NhIQ0MDV155JUePHmXbtm289NJLHBj0cGDQw73AZy4Z4YLSs++De2jI4F9eKU67b+PGjWzdupUVK1ZQVVWl/u4UZ2SyaIeUuTPScTQcHM0AXUeXFrajEXE8GY60zHCdM0aET+JI21JDAlHhBUn6iHDhnkqjWtKRHgjFInNxIX3wILS1QXGxEtIKxQJCCel5hmVZPPTQQ/zkP/+TYDBIpGo1Zt0lYHhPu1+0eg12QQ3OqZf4zne+w28feohP3X47q1evnqGVK2YT27bp6uqitbWVlpYW2traaDl5gra2NkLh5Em9yCeozYtSH7BoCFgsLTq3YRJLiiz+asU4pyYMOicMOkMet1vI9u0A+H0+mpoW0bx4CYsWLaK5uZmmpiZqa2vVCHtFgtM50pldOyDdkXaEDprmdvCQAtPxZGSkQZ4u2iFzDWRxnyCKh0hcSKcWG2ruqTTVkR4IulnthJCOu9Fvfzv85jfn9XtRKBRzDyWk5xH79u3ju9/9Hi0tJ7GL6giveStOXulZ7+8EyplYeRPG4AmOtezkb/7mb7jxxhv52Mc+RknJ5LEQxfwhGo3S0dFBS0sLra2trnA+eYJT7e1Eo0lRXOqHurwomyos6gI2dfk2dQGbIu+5DWDJxKvDWxpMSOmSMBYRdAZ1OiZ0OidCdPaNs6PtKI+Gk/t5PAaNDQ00NS9OiOumpiYaGhpUS7w3IGfb/i5+O1VIS01PONKOI4hIA69IjXa47nJmfCN+20HkdKQhLqR9ifuSQto1MlId6cFQnnuAeLHhH/7A7RX30Ld3LT83f+m+KPXtjEIx71FCeh4wMTHBXXfdxUMPPQS+AkJLr8MqbTq/N2EhsMqXMlqyCG/nHv7wxz/y/Asv8Mm/+zuuv/569bX7PKKrq4sDBw4kRXPLCdo7unDiqgKozIe6vAjX19rU59vUB2xqAzb5xvkJ5nuO5NM65gqQf9ldxKICi/etCJ52n0KvZKXXYmVJursdtARdE67A7gjqdI4eYf/2Fp5+CuKr0zSNhrpamhYvSYjrNWvWqCLZBc65DGSJbw9usaEjYkIaK+FIewkl9jE0SUQaOQeyuO3vckc7wC00tGLt79yuHbHnFzkc6XgLvLgjfeoUB4yL6BuodEW0ZYH6kKhQzHuUkJ7jvPjii3znO//KwEA/keo1mPWXgj4Fb766h0jj5VgVy3BaXuBrX/sajz3+OP/77/9eTZubB0gp+eu//muCQVfECiQXlkVZ12hRH3Dd5dp8e0o7cAC0jRuEbFdkHBp+fa3z8g3J0mKLpcXpAtu0oSuo0xkT2a1jJ3n+uXaee85VLXl5eTzyyCPqQ98C5mzb3+WKdsi4kJY2UgpMx0sRI8S7vXoNG0dqmLGiQbKiHSJ7smEs2hHBiylSunbEJxvGHOnUrh0D4/50IR0MYuLDdGLv36aphLRCsQBQfaTnKMPDw3zlK1/hM5/5DP0hm4kLtmIu2jg1IjoFJ6+UiVU3EW7cyM5dL/P+9/8Vv/3tb9NcTcXcQwjBt7/9bbZu3UpRYQESwZFRH30hHY8GddMgomcKn+6u36NBX0jnyKgPiaCosICtW7fyne98R4noBU5q+7tMIZ1ZbBi/H+LFhq6Q1qRrU4ccHx6RzEh7dHfjcQoSx4+T2f4udSALQFjkYQo3+2zoTnaxoUjWqgwGY470K6/Aj34Ew8NEghaRidgHx3BKtkmhUMxblCM9R/mnf/onXn31VRxvgNCSzeeUhT5nhEa0+gLQdJzWP/G9730PIQTveMc7pu85Fa+bNWvWsGbNGm6//Xb27NnDM888w7PPPM1L+8fwGYKLy8JsqIpwUXlkXohq04Z9A1529HrZM+jDtKC4qJDrb7yGzZs3c/HFF2MY6i3rjUBqtCO1j/RZFRvGu3bgCtaQ44117XDdYm+GkM4uNsyRkY4J5ojwE8aNbxhatpBOdaR3tlQSsr2c7ChiNUAkgqlnONIKhWLeo85Kc5QPfehD/OKXv2TH9u0E9v8PdmENkYrlWKWLQZ+6fzZhjuHpP4pv4BiY4wQKCrnxhreyadOmKXsOxfRiGAbr169n/fr1fPKTn2Tv3r08/fTTPPfM02zfP4ZPh3XlJlfWmFxSET3zAWeYPf0enu/2sXfQj2lJiosKeevbNnPNNdco8fwGZbJiQylzFxvGM9KpXTt0GRfSfjxaMj4Ud6RHRTHIHO3vnOyMdDzaYQofpvDHjiMTPahtzRXHcSGd740yYXpwNAPNsd0FWhYRPERk7O9ZCWmFYkGgzlBzlHXr1rFu3Tr6+vp49NFH2fbww3SffA5xajtm6WKiFStwAhXnV3DoWBhDrXj7j6KPdiKE4NLLLmPr29/OlVdeidd7+lZ6irmLYRhcdtllXHbZZXzyk59k3759/PKXv2THzp3s7PXyg01DFHheX2eOqWQiKvjevkIkgssvX8+tt97KunXrlHh+g2PbEP8TOJv2d6nFhvGMtOa4dwYdP14tO9oxrJeDlWNEeEpGOjPaYQp/Iludy5GORzuK/FEmIkZMSFuJnLTpeIk4SkgrFAsJdbaa41RWVvK+972P2267jb179/LQQw/x5JNP4u07jAyUM7H8rUhP3lkfzxhsIb/1BaRlUlBQwLs/8AFuvPFGampqpvFVKGYDXdfp6elh3769FHjhw6vG5pSIBgh4JJ+6aIwfHypi3769XHvttaqftOJ1FRvGM9K67brQYelNy0h7DVdgj+plYIFIUdJnan8XET7CJDPScWzhLjY+PrwoL0LfWB6OrrtCOuK234tIA9NWQlqhWEgoIT3HkVLS1tbGyy+/zI4dO3llzyvuA0IgtXMvPJSahjS8YJmMj4/z8CO/p7+/n/Xr13PppZdSVFQ0xa9AMRuMj4/z3e/+K08++RQXlFp8dPUYZb65WUB6cUWUr14+yH8cLOJb3/oWO3fu5O///u8pKCiY7aUpZonJMtJn3bVD09CtuJD24xHJaIc3NdpB9ohwGx3nE7fD97MHsphaakZa4kh3ATbpjnSxP8LJ/kKcgIEmrYRojjge5UgrFAsMJaTnIMPDw+zevZtdu3axfcdOBvr73Af8RUQKm7Eb6rAKa8Hwnf5AObBLFjFW3IgwxzBGO+gc6eDhP/yRbdu2IYRgxYqVbNhwOevXr2f16tVqGMY8wrIsOjo6OHr0KP/54x/R09vLu5cE2doUSvv6ei5S6pN8+uIRHm7188DTT/HawQP89Yc+zIoVK6ivr1dRjzcYqV07dH1yIZ1rIEvCkSZFPOfo2jGiuUOotIw+0o4jcLa+A76fq4+0j3As2uHRHaKxVpCZGemivAhRW2dC5CMcO+FIm46BIzUsdAwlpBWKBYE6O80RhoaGuP/++9m+YwfHjx1DSokwfEQKa7GblmMV1SH9U+QWC4H0FxH1FxGtuoCwdNDG+zBGO3ito5PD99zDz372M3x+P5dcfAmbNl3FTTfdpFqOzSGGh4c5ceIEx48fd6+PHaWlpZVILItZmSf53KWjLC8+txHfs4km4ObmMBeUWvzwoORrX/saAF6Ph+bmJpYuW86SJUtYunQpS5YsUdM4FzDnGu1IKzaMde2IZ6QBPKmi2nAPNiaKYsdIxp3izxV/vmSxYTLaEZExR1p3sJyYIx3v2kHSkQYYFOUUpwjpuBsdwauEtEKxQFBCeg7gOA5f/OKX2PfqPuyCaqy6S7CK6mLFhDPQ6ltoOIXVRAqridRfClYEY6yLyEgHL+05wEsvvYiUkq1bt07/WhRpRKNR2trakoL5+HGOHzvK4NBwYpsin6AxP8J1NRaNBTaNBRYNARtjnnaJX1Zs8c2Ng7RP6JwaNzg1rnOq/zVeaDvG73+f3K6stISly5YnhPXSpUtZtGiR+hZlAZAr2iHlmftIp002lKmOdHJEuEd3BfaEiPWRJltIJ4oXMxxpU/gJy9Riw5gjHStOTBQb5rnPNyArKJUdSUc6lo828ZGvhLRCsSBQQnoO8Lvf/Y59+/YSar4Kq3LFbC8HDC9WaRNWaROmlOQf+QM/+MG/s2HDBjX1cJqQUjIwMJAmmE8cO0rrqVPYtqsSDA3qAw4XBCIsWmYnRHOxd24VEE4FhgbNhTbNhXba/SMRkRTX42FOHRrklZd3YsWElK5rLGpszHKvKyoq1Dcq84hcjnS8Bd6Z2t8lJhs6KXEOkfw7imekR/QyIN2r0HX3WJmOdDwjHcWb0rVDJvpLJ6IdcUc6JqQHKUVzWiESwUbDlilFiUpIKxQLAiWkZ5ne3l5++B//gV1Uh1WxfLaXk40QhJquxDj4IN/97nf5+te/rgTJFDEyMsLPf/5zjh45wvHjxxgdG088Vp4HDfkRbmqIu8w21Xnz12WeKoq9kuKyKGvL4iJpAsuBnpAeE9c6bWNHeeWFVh5/PLlfYUGApcuWsXz5Cm677TYVC5njpDrS8T7SCbGcI9qRENk5JhtCRkY6Fu0Y8Na5+5wm2pHZ/i4ifJj4MDQHIZLPb5Ee7Sjyu883QJkbMYlEEo+B60grIa1QLAze4Kfl2ef+++8nHAqBY+HpO4Qwx8+800whHbTxXjz9R3A0g5deeoljx47N9qoWDO3t7Tz44IO8smdPmojO90C51yJgSAQQtgVDpsZAWCM6NxtvzBqWAwNhjSFTw7QFQkDAkJR7LfJTEh5j4xPs2bOXBx98kI6OjtlbsOKsSHWkPR5XKMeacJyxjzRCSxvIAmBo2V07TEtP7kPy2Lky0sloh49ITEi722f0kcbA0Bx8sRZ7QQJu144MIR3Bq0aEKxQLBOVIzzLvfOc7cRyH5194ge7WF4EXkYFyIkUNWCWNOIHK8xu6cr7YUYyRDoyRU3hH25GREJqmsXbthVx99SYWL148c2tZ4KxZs4ZHHnmE/v5+uru7My5dnOjs5KVTAzhOUj0LoMQPFT6LSr9Nhd+mIs+hwu9Q6bcp8zt4FtDH46gDg2GNvrBOf1ijP6TRH9bdn02D4TCkBls0TaOqopyaxXVcXVNLTU1N2qWiokJ1AJkHpDrS8ch7KORen679XVL5pkc7fKR27XBFblxIZw1kyeFIx4VzVLjRjngP6awR4dKLrsnE4yb+hCMdj4S49ytHWqFYKKgzyixTW1vLxz/+cf72b/+WtrY2XnzxRV740584sP9VnK69CG+eK6qLG7GK60Gf+kIqYY5hDJ/CM9yGPt4NjkMgUMCbNl3BFVdcwYYNGygsLJzy51W4kwjjIi8XlmVNmdCu8NuUzzGh/XqE8lIllBcsqe3v4oNW4wZurmLDuCMt44HnjGJDIyUjrWugCQfTig9dSf6FxTPSWcWGsW1MfJh4E108EkI6XmyIB11zEo+HhT8xkCXLkVZCWqFYEKgzzhxBCEFTUxNNTU3ceuutjI6O8tJLL3H//fdz5MgRPP1HQdOJli8l3HzV1DynOUbg2BOI4CDgTsK7fssWbrrpJtasWaMEyRzgfIR2T08P3V1dnOjq5KW2PhyZFAqpQrsu3+LWZUECMzjtcCIquPdYPp1BI7dQFoKqygollN/gpEY7MoW0yHCQ49tDypTC+ETBGKkjwsGNd5jRc3CkdfevNF5smHSk3ceTjrQHQ5OJXtUmvoSQVo60QrEwUWekOYDjOHR3d9PS0sLJkyc5efIkJ06e5FRbG9Fo8gQgvPlIMZX/ZALH8GF4fMioiW3bPProo7zwpxdZumQJixc3s3jxYhYvXkxzc7OaejgHiEajDA8PMzw8zNDQUNrtkZERhoaGGBoaZHhwkOGRkTQRDa5oHQrDRNQgaAnCtphRIW3agmOjHvrCOhE7+3FHSoZHRgAIh8OMjIzQ09NDa2srI+wsGAAAIABJREFUpaWllJSUUFJSknZbtbtbeKRGO87Gkc4Z7Uh1pEn/Y/MaDmNh9+/mdMWGSUfavWPcV8aEU4QnFvUQiWhHvBuHx412aEkhLWSs2FD4E58alSOtUCwclJCeQaSU9Pb2Zgnm1pZWIpHkm6rwFRD1F+OUrcTOK8HJK8XJK5nyWIf0FRBc+TaQEhENoYWG0ELDREJDjB7vZN+Bg0gr2X+1uKSUpUuXsLg5KbCbmprUKOfXgWVZjI6O5hTESbE8xNDgIMPDw0wEQzmPo2tQ5BUUeWwKPRZLPJLCKocij0ORV1LkdSjyxK69Dj4952GmnTK/w9c3uj2wTRtGI5p7iYrYbcFoNMRYZIzR3g46OwwOWzojpsSepNAykJ/niuuyMkpKStNEdqboLioqUo72POB0jvTpMtKpjnRqRtqrWSTfydyphG4pb3a043R9pA8tuoGunhKMUIYjHR8RLt1oR6I4MdWR9hZC7G3e1PKVkFYoFgjqjDKN2LbNI488wmuvveYK55ZWwqFg4nHhCxD1FeOULsXJK8XOK8Xxl4DhPc1RT4+v7SX04AAAeYcewckvw1z0ptPvJATSm4/tzccurgdi7/dSIiITCYEdDQ8zdKiVV/bsRdpJt6esvIJlS5fQ3NzMpk2buPDCC897/fMdx3EYHx9Pc4tzucfDQ4MMDw0xOj6BlNmOsBBQ6BUUeRwKDYsGr8PqMofC6rgYlu5jMYGcb8gZrUmdCnw6VOY5VOaduRWJlBC0BKNRwVhEYzQaE90RjbFoiNHIMKPtbbS0GuyLaIxFJDl+rQghKCoIUFxSQmlZ+aQud/zngoICNG0OhcrfIJzOkT5t1474HZqWFu0wRLqQ9upJhzoz2gHJDiGZQtqSGlFHT0Q93MecpCMtXUdaCLdA0RXSMUfayE8I6YhHCWmFYqGghPQ0Mjw8zB133IFlWUjdR7R8CU6V6y7beaVg+M58kHNECw4ibNeJMca6eV0DooVA+gqwfQXYJY3JuncpEZFxtNAQemiInokBBnfsYMeOHbS0tPCtb33rdb6K+YNt23z961/nxPFjDA8NMTI2lhigkkkgJoyLPBaVHoelRZKicofCmCiOC+RCj0OBR6ad4OcKIUvg9/vZunUr27ZtI2TNzAhyISDgkQQ8ktr8MwtvR8J4VDCWJrhjrnc0xGh4gLHWkxw+ZjAa1ZiI5I636LpGcWEhJaWlLF6ylM985jPK0Z4BzteRlpN07fBqFkkLI9lLGrKjHZBLSLvb246G5YhEtMN9TCYcaTcjHRugpDuY0pdof2cayd7lplGghLRCsUBQZ4RppLy8nB/84Ad845vf5OSJE4hoCKvuYqQnb7aX9voQAukrxPYVgubB138UgJtvvpmPfvSjs7y4mUVKSSRiEgoFGZuYyCmidQH1BTYVPpsyn0Opz6bU58Ruuxf/PPmfGLQEW7du5eMf/zhSSp59+L7ZXlJONEEs0mJTH3DvC1swZGqJy6CpMWTaDJoa/aZOx7iOnaGnbdthbGICn99HNBrJfiLFtHC2jnRmsSGpXTvSJhumf+CLFwNCerQjU0gnhsIIt9+07QgsWyPPmzyeLmTCkY5KI+FeezSHCN5k1w4jP7GP604rIa1QLATmyel7/rJy5Up+/KMfce+99/Jf/3U33gO/IdiwAat82cz2h55qrAi+9p14+w5TW1vH//k//8wll1wy26uacQzD4Mtf/grgiuqRkRH6+/vp7++nr68vcd3X10d/Xy9H+voYnwhmHSffI1xR7Y1miez47ULP7Mc38g3Jtm3bkFLy8MMPU23M/nhyKWEsKlLEsZZ+O+JhyNQIRrPXWhDIp7KykurFVaytrKSyspKKiorEdUVFBcXFxWqa5wxzru3vzpSRzhTSXj3VkU7eHz92vMY7fnwhYs6zI7AcLdHeDlzn2ZJJRzoupA1dpkU7TCOQ2CfiCYDZf8bfg0KhmPsoIT0DGIbB+973PjZt2sQ3v/UtDh54Drr2YJY0Y5Utxskvnx+i2rEwhtsxhk7iHWkHx+KWW2/lAx/4AH6/f7ZXN+sIIRL52mXLlk26XSgUYmBgICmwU0R3b28PB/v6GOoezuq4YWhQ6odSTzRdaPtjt70OJT5nWseI5xmS8HiYBx54wP25ZHqFtOXAcCRFHIczhHLUw3CYrImPmhCUlpZQUVnJkqrqhCiujInlyspKysvLycub598OLVDOtv3dpF07NA1NJh1fQ8vs2pGakT6zIw2ukLYcDcsWifZ38fuTjrSejHZoDqb0JCcb6sm/NVPPV5MNFYoFghLSM0hTUxN3fv/7PPHEEzz22GPsevllnO5XwV8UE9XNc09UO5Y76XDwJN6RU0g7SmFRMde+/W3cfPPNLF++fLZXOO/Iy8ujoaGBhoaGSbexLIvBwcFJ3e1TvT3s7h4gEk3vjyuAIh9sqQ/yjubcHT7mAw+15PFYRz6jZnqfaQCvx0NlRTkVzVUsqqzK6SKXlZWpLPM85lzb3w0NgYaNES8i1HX0lJZ3HpEupM8U7ch0pCFWVCgFUUdLiGWIRTviGWnHk3CrDd0hIr1JR1pPiXboeSraoVAsENSZZobRNI0tW7awZcsWRkdHef7553nyqafYvXs3Tvc+V1SXxpzqvLLZEdU5xHNBYSHX3nQjmzdvZt26dUqkTDOGYVBVVUVVVVXOxy3L4tChQzz33HM88fjj9A+4nVokUOR1KPflaNI8j6jw2xR5HUbMpCVYXl7OW97yFq6++mpWrVql/gYXMKmOdLxN+OmKDQ8fhs0Fu8nXYjn2TCGtTS6kzybaATHn2XYz0p7Mrh3Eu3YkM9KG5hCxvGjSBtMkkpfhSCshrVAsCNSZaBYpKiripptu4oYbbuDkyZP8+te/5vHHH8fXtQ9f1z7IK2Z82Rakf+YGoXh6D5HXsSvRP7q0rIz3vOc9vO1tb6OkpOQMeyumC8uyOHz4MHv27OGVV15h/6v7CJvuv1FjocOWBpNVJRarSqMUzuCAleniipoIV9REGIsKDg97eG3I4NBIH/fddx/33XcfPp+XCy+8iEsuuYSLL76YlStXKmG9gEh1pHXdvX26YkOAW0r+gBPSEztpJMVyppD2nmOxIYChSWypZTvSmkyOCHf0RIcPQ5dEoslPAWYgKaQjml8JaYVigaDOPDOAlJKxsTG6urrSLp2dnXR0dtHb24Od0UZM+AuxPIEZd6SlZmD7itDFODIaZnBwkLvuuou77rqL/PwAtbW11NfXUVtbm3aprq7G55v6dn5vVCzL4siRIwnh/Oq+fYRjJ976AoerKk0uKLFYWRKlyDv/hfNkFHok6ysjrK+MAEHGIoJDwx5eGzY4dGgHP961CwC/z8eFFyWF9YoVK5SwnsekOtJCuPGO0znSgQC8Ne8FHDMppNOjHafp2pFy/2Tt7yDmPDsid0Y67kg7BnmpXTtkTEiHQq54jmFqKtqhUCwU1JlmijBNk+7u7iyx3NHZSWdnV9ogFgDh8WN7C7C9BTgVFyB9hTixi/QGQJud0XNWxTKsilihnBVBi4yhhccQkTEi5jij/WMc79qHCP8J6aSfnErLyqivq6eurjZLaFdUVKjBFmfB2NgY9913H7954P7EFMP6AocrKyJcUBpl1QIXzmei0Cu5vCrC5VWusB6NCetDQ2Fee207O3fuBNxph+/8i3dxyy23UFhYOLuLVpwzqY40nFlIb9wI/lMhpDaZkM5dbChEeiec0wppIbFiXTuy+0i7O0Ycg8K0jHRMSEcirniOoRxphWLhoIT0edDa2sqTTz5JZ0wkd3R2Mjw0mLaN0AykvxDLU4BT2IRTURgTywU4vkLQz3964YxheHGMcrcAMpP4WHFzDGGOoZlj9EbG6W/t48Cxk8jweNrmumFQVVVNQ8zNXrRoEVu3blXdPmIEg0Huv/9+fnXvL5kIhri80mTD4girSqMUv4GF85ko8ko2VEXYUBUBJhiJuFGQ7T0m99xzD//zmwe45db38q53vYtAIHDG4ylmn3gHjtSiwlQhnXp/dTU0NsLmzaD9VxQr1j0DTUuLduhaeluXuCOd2rEj9diTZaQtR8N2tPTJhiIz2uFGrgxNEooLaSCi+WLHlJhCCWmFYqGghPR58PDDD3PffemDKKTuwSptJhorEpSevLnVfWOqSRkrTmF19uOOjYiMY4x2YQyeQI5109XZQVdnB+AWXa5du5ZVq1bN8MLnFuFwmAcffJBf/PweRsfGuaQiwl+sCdJUOL+LBWeL4hRh3TYW4oGTEX76059y/6/v471/eRt//ud/rlrezXFyCWmPB8bG3Nupb6ulpfD5z7u3NcfCEdmOtI6VNSU0npHOfIfO7NqR3v7OwYy6x0/PSDtp0Q5dcwWy60gnT7GmcE2DQl+UiPDlFtLd3VBTk32/QqGYsyghfR589KMf5YYbbuDgwYMcPHiQV/cfoP1UG57+o3j6jyLzS4nmV+IUVGIHqnDyipMTtxYydhR9oh99ohdtvA9vsB8ZcSMtHq+XVStXsmbNGlavXs3q1aupqKiY5QXPHqZpsm3bNu752X8zNDzChWVR/mJ9kKVFMzNy+43AokKbT100xvFRg9+ciHLXXXdx36/u5bb3/S9uvvnmN1ymXwhxI3AHoAM/kVJ+I+NxH/DfwGXAAHCLlLJFCNEMvAYcjm36kpTyY9O1Tjv2GTIz2hHXnfokqTfNtpDxHHKKkPZhJkeHx5jMkT5dtMPQJKblbpCVkZZJRzousj2aQzTVkRY+NOGQ57UwhQ/CGUL6ySfh+uvh+HFYvDj3i1QoFHMOJaTPA13XWbp0KUuXLuXmm28G3GzroUOHOHjwIAcOHOTAgQNM9B8BQBherPwKrEAldkEldkEVGPM/0iBCI+gTvejjvXiC/YjgoDtmDqitq+fCDVclRPOSJUtU8VcK3/zmN3nyyScBKPQ4FHlt9vR7aB/XKfc7VPjdceLe2YnKz2siNu7Y77DOQFijP6xR5LUp9OgMDY9w5513cuDAAb74xS/O9lJnDCGEDvwA2AK0AzuFEA9JKQ+mbPbXwJCUcpkQ4lbgm8AtsceOSykvnom1ThbtiDNZqYVIdaQ1LSGkvUQI+4rTtvUap4929Pe7z5m6Bl2TBCPue1h2145YRtrWUyYbpjvSEXz4DAefYRMhhyP98svu+2dHhxLSCsU8QimbKaKwsJDLL7+cyy+/HHA7dbS3tydc6/37D3Dy5Ks4XQ5CNwg1biRasWJ+xj/sCP6WF/AMngQgLz+fNatXs3r121m9ejUXXHABxcXFZzjIG5t3v/vd1NbW0tPTQ3d3F0e6unixZyhrmmGJX1DmtajwWzGB7Yrs+O38GR7RvajAonXMVRdNhTaLCmbeQQ9agv6wFhPJeuL2gKkzYHoYDme4jEJQXlZK08paamrcDjNXXHHFjK97ltkAHJNSngAQQtwL/BmQKqT/DPhS7Pb9wJ1iFmajT+ZIx5lMSGuOheOJndJS2t/5MBkrqE3b1qvHiw0zjhE79iuvwJo12ZMNTcv920/tI21oKSPCU4W0Jok6KdEOfHgNG6/hYOYS0seOudfj6fUlCoVibqOE9DQhhKCxsZHGxkZuuOEGwM3DHj58mLvvvpvdu19AH+0k3Hzl/Cg8jKFN9BM48TRaZJz/9Vd/xXXXXUdjY6PqyHGOxJ36VCzLoq+vj+7ubnp6ehKX7u5uOrq7eKWrj2g0XbjmewTlfpsKn+VexwR23NUu8sgp/az2vhVB2sbdt43PXjo6dQeOISWMRkWam+wKZZ1+02AgrBOMpgtlj6FTVVlJ9dI6VtbUUF1dnbjU1NRQWVmpvg2BeuBUys/twMbJtpFSWkKIESBeabxYCPEKMAp8Xkr5XK4nEUJ8BPgIwKJFi85roXZyOGGCs3KkbQvpzc5Ie4lkCekzRTuOHYMPfjD9+LrmJKMdmY60496fGu3IdqS9+Aw75kh7skeEHz3qXk9M5H6BCoViTvKGP7vMJH6/n3Xr1vHtb3+bX/ziF/z0pz/FExxgYslmnMAczwtLiafnIP6OnZSXlfOlL36NCy+8cLZXtaAwDCPRLjAXjuMwNDSUJbJ7enro6erkcE8PwVD6ydmjQ7lfUuGNUp4isCv8DosK7Rl3tMF1lNvG9CxHuT/iYSAsiGbUWebn+amprqZheR2XxcRxdXU1VVVV1NTUUFpaqj7InZlcH6cy//En26YLWCSlHBBCXAY8KIRYI6XM+iQlpfwR8COA9evXn9cfVzzakfpP6klGjSd3pO0oTo72dz5MxgIZQnqSaEfqsdetSz++rknC0Vi0IzMjredh5xfiBLWEI+3RnByOtINXdzBlLPQtZdIWjzvSSkgrFPMKJaRnENu2GRgYoLu7m+rqai6//HJ27NhB4NA2gsuuxy5umO0lToq/9QU8fUfQNI233XgDQ0NDHD16lJqaGtWnd4bQNI3y8nLKy8uz3Ow4Y2NjdHd3c/jwYfbu3cvevXvp7u2leyL7W4/lxVH+6bKpd5XPxPf2FnJ4xJPzsaqqKtatW8e6detYuXKl+vuaOtqBxpSfG4DOSbZpF0IYQDEwKKWUgAkgpXxZCHEcWAHsmo6FnsmRnrTY0LGQKV074tEOLxEm8ivTtk107cj46JB67IsvdkePJx6Ltb8DN7aRuF9IbH+AyDe/B59IPmboEkvqOAg0JBHpSTrS0uOKaMtyPyWEQnAq9oWBinYoFPMKJaSnECklg4ODicEsqdcdnV309fZg2+l2m/AFsDwFoM3tfwrHW4AMlCPNMe655560x/Ly86mpqaG+ro6amhpqa2vTrvPz82dp1QuTuDPd3d2ddunp6aG7s4Oe3l7MSDRtnwKvcJ1oXzQR/1hdGp3kGaaXD6ya4MCgJxnbMA36wgbjEUlvby+PPfYYjz32GD6vh+qqKqpr3b+rzItyos+JncByIcRioAO4FfjLjG0eAv4KeBF4N/CklFIKISpxBbUthFgCLAdOTNdCz1RsOFlUSbOt3I60YScHtcTwxDLSp3OkL7ooXUinxjk8mY60I4jaWuxnJ237CF78mBwfKiFoGgwIH76CWD7aNF0hfSLl16kcaYViXjG31dscQ0rJyMjIJEK5k96eXqLRSNo+wpuP7Q24Ewwr1yBjA1kcXwHSWzBrEwzPlUjdxUTqYkX7lolmjrvDWCLjRMwxxgbGOdl9AGG+hLTTc7wFhYXUxoR1psiuqalRQ1kycBwn8YEsVSR3dXXR3dVJT29vVla60Ceo8FlU+yzWVDtU5LkRjkq/Q7nfJm8O/U+vD9jUB7L7ZIcsGAjr9MUjHyGN/vA4/YfbOLTPYMzMyEZ7DKqrqqiZRGiXlZUpoR0jlnn+OPAobvu7n0opDwghvgzsklI+BPwn8DMhxDFgEFdsA1wNfFkIYQE28DEp5WD2s0wNZyo2nMyRnrRrR474Urxrx2TFhosWuT2qU9FTRHd6tMPBloJIQkjLtG1MfPgxiUoPhi4xNEnEiX0jY5pQUJDMR4NypBWKecYcOr3OLWzb5ve//z3Hjx+nq6uLzq4uerp7MM30DGrqqG9ZvgLH6wpl6SvA8RaCPsO/YjuC3+9n69atbNu2jXE7cuZ9zhXDh2P4cAKTTDy0wghz3B0vbo4TMccZ7h3jaPs+MF8AJ11EFRUXU1dbR22tK7Df/OY3vyHy111dXbz66quJrHN3dzfdnR309vURtTJ+RzGhXOe3uKg2mXOOd/HwL4D/yXkGNBTYNBTYQLZbHrZIZKr7E/nqCfqPnOLIqwajmUI7VoRYU1efyFXX1NRw4YUXTppDX8hIKR8BHsm47wspt8PAe3Ls9wDwwLQvMMb5tr/THAtHS3btSDjSOSaDes8w2fDiHI3+jIxOHYl9NEnU1hKOdCLaoSWFNEBEetA1B0NziMSz0/HOHXEhrWnKkVYo5hkL4PQ7PQwPD/P9O+/EzKysBqTuxSqqxS6qw84vR/oKkYZ/TrSyE1aEre/Yysc//nGklNz3u0dneAEC6clDevJwSMklSscV1+Yo+ngvxmgn2ngfAsnoyAijIyMcOvQa4I5gfyMI6c9+9rOcPHky52MFHocVxRYrSqKsKLaozXcLA+fAn9is4U8T2kmkdAsYu4M6R0YMDg97ODJiMB6Fjq5uOrq607Zvamri7rvvnsmlK86B821/J+xJMtK+7P80nkRGOne0I7PQEEAXqdGOlIy0JglHBZFYa7x4tCO+TVxIW9LA0CSG7mDaGUL62DEoL3eVvBLSCsW8QgnpSSgvL+c3DzxAR0cHnZ2ddHR0uJfOTtrbOxgcaEMOtSa2F7oHx1+E5S1A+orc+IbfvZbewIxNNpSGl23btiGl5OGHH0YaM5hPtqNo5pgb+TBH0cLubSM6BuGxxLAWAI/HQ82iRhobGqirq6Ouro76+nrq6ureMG7hN7/5TQ4dOkRvby+9vb309fXR29NDb083A4ND7O7X2N2fVBB+Q1DmdyjzRN1rn9uFozR2XeabWxGOqSJkwaCpMxjWGDTdDh+Dpub+HPUwGNYIW9m9oyvLS6mqrqGquprKykqqqqqorKzkggsumKVXojgbzrf93aRdO/y5hHQ8I51xjNixcznSmjaJIy3c9ncRKyPakelIOwYej4OhSSZiojvNkV6+HHp6VLRDoZhnLMDT7tQRCARYsWIFK1asyHrMNE26u7vp7OxMXNrb22nv6KC7+zVsKyXDKjTwF2J5C3BSRLb0FeL4S6bWyda9hIODPPBA7JvYwpKpOza4Yjk0lBDMWngULTKGERlDmsG0TfMDBdTX19FQvyIhkuOXioqKN3x+taqqiqqqqpyPWZbF4OCgK65jQjshtnt72N/Tw1D3CDJjgEu+R1DmsynzWZT5kmK7zGcnxLdvDsXyzdgUQlckuy3xhkyNAVNj0DQYNLP7RgshKC0ppqq6mmVVSZEcv1RWVlJWVqZ6R89T4m+d5xPtSHWkExnpvOwdvJM40uvXwzvfCddem3381E4dWe3vUooNU/tIQ6ojrZGnRdMGu6Q50ps3Q1sb7N8PP/pR8ok/8pHcL1ihUMwJ1JnmPPH5fDQ1NdHU1JT1mG3b9Pf3J5zsuNA+1d5BR0cr4d6k4DQb1hOpvWgml/66CLy2DS00lPi5tKycxsX11NdfmuYq19fXq7ZlrwPDMBLCcM2aNTm3iUaj9Pf309fXlyW4e3t62NPbw3DnWNZ+Bd6Y2PZalPkdyn02iwps1lVMXxePvQMet3d03FmOuCJ5PJKdXy0pKqSyqpqm6mouTxHH8euKigo8ntzt8xTzn1DIvU5t9nPGYkMp0Rw7zZFOTDbMzz7NaZobwcjMSNfWwm9+k3tdeooLndb+TpOxYkM98XPqNiY+orofy9ES0Y74tphmsvXdsmXw/PPZEw8VCsWcRgnpaUDX9cRktUsuuSTtMSklv/jFL/jxj38M+aVEyxbP0irPD7P2IvJanqewIMB3//VfWb58+Wwv6Q2Lx+M57QAXcL856e/vT3GzU2Mk3exracVxHDya5IebBvFOg1sdseH/vlpI1BFomsbi5ibqqqq5OMNFrqqqoqKiAp/PN/WLUMwb4kI6Ly9535na34lYAXPOrh05oh3gutLn8l3g5I60g2Unox1Gjq4dluFzhbTuFhua0dg6w+Fk67vly8Hng8g0FIgrFIppQwnpGcS2bX74wx9y//33Yxc3EFyyGYz5Mx4cwCpfyoSvEHH8Cf7uk5/ky//8z1x++eWzvSzFJPh8Purr66mvr0+7f2RkhDvvvJPjJ07SUODwoVWj0yKiAbw6fO7SEX5yqIj2cVi8ZCmf+MQnKC4unp4nVMxrgrEv7M7FkdYcNw+Ss2uH4WTvgJuTTs09nwk9ZVtPhjtty+w+0p5UIa37seyYI61JIlZMwptmsmPHsmWukFYZaYViXqGE9DQhpWRoaIjW1lba2tpoaWlh//79HD16lEj1GszGy2esAHGqcQqqGFt1M86xx/n0pz/NlVdemYi5NDc309jYqHpDz2GeeeYZvvfdf2VsdJQ/bw7yjuYQxjT/KS4psvny+iEeasnjd088zq6dO7j9U/8fmzdvnt4nVsw7cjnSqUmenI50rHd9zoz0JELaa2RHO06HfoaMdFYf6ZigDuPHMnzYUYGhxR3p2H8400w60kuXukJaRTsUinmFEtKvEyndaWwtLS0JwdzS0kpLaysT48l8qjC82P5izOarsCqzixfnG9JXwPiqt+M7tYPndr3Kc88/n+jKIYSgsqqKxc3NCXHd1NTEokWLVG56FjFNk3/5l6/xzDPP0lzk8L/Xj7KoMHswynRhaPAXS0Ksr4rwn4dsvvSlL3H11Vfzuc99TsU5FAlO50hrWm4hrdluvj+Rkda0ZEbayP037jnHaMfkGWkno9hQxo6fzEhbhp+oqaFrEkOXRC0NCQjThO5uV0CXlrovVAlphWJeoYT0WWJZFl1dXQnB3NraysmTJ2lrO5U2pEV487B8xdh5dThlJTj+Epy8EqQnf070mZ5SdA9m85WYAI6NZo6ihYbRwiN0hIbo2X+M7Tt3pQ1gKSktY3FzUlzHL6WlpYiF9vuZY3R0dPDMM88SMCS3rx2hzJ/bqZtuFhXYfHLtKJ/fWcKzzz7L+9//fpYtWzYra1HMPU5XbHi6jh2Q4kj7/ejlpTAA3tMI6fNxpA3NSXsrTzjSGX2kU9vfWboP2xYYupN4PIIXX1xI19S45wflSCsU8w4lpCdhaGiIBx98MCaYW2jvaE9raSd8BUR9RTjFS3DyXLHs+IuRnrzTHHUBo+k4eaU4eRlzdeODWMLD6KFh+sIjDB5tZ8+rB5BWsqgmP1BAc3MTi5ubWbt2LTfccMMbvj3eVLNkyRK+/OUv87WvfZUv7y7l9guHaZ5BRzpOy5jOv71aQlTz8s///DklohVpxB3pXMWGkw5jiWekU6Id4l++xiVf7Wdt3RATZvapzqvV2Jq8AAAgAElEQVQ7yLPX0VkTC+NM2kc6rdgw3rXDSWSn04R0dbV7sHixoZQLz3hRKBYoSkhPwu7duyeZfiawA+WuaPQGkN4C99rwIzX168xGgO5BevJxpERqOtLwYXsK0Cd60SLuFK/gxDgHDxzg4IEDPP3001xzzTXk58/gMJk3CFdffTW1tT/gs5/5R766Gz5ywRgbqmauS8COXi8/fq2IotIy7vz6N1TXF0UW5+VI2/Fiw/RKxN2fd3vZ/ejZVVn7rG/qw3bOXqzqGf2hk/enFxsm+kintr/T4l07ZEJom/gojAvpxbHuTT6fK6Kj0fQKS4VCMWdRym8S3vKWt3DppZfS09NDX18fPT09idZhPT09dPf0MNx1LGsghvD4cbwBbE9+utD2BnC8BW7EYxqdVie/DBkcAMDOL8fJL5u253KfJIoWmUBExmPXE+61OY5hBRHmBNKx0nYxPB4qKyupaV5BdXV12jCNeNtAVaw4fSxfvpy7fvRjPv/5z3Hn/te4oTHEtXVh6gLTF/XonNB4qtPPo6fyWL16FV/5ylcpLy+ftudTzF/Ox5HO6tpxFly7svOc1qWL9P7QibUZNmZUz+4jneJIh4yC2L5OQmhHiOWhe3rgzW92DxavFTBNJaQVinmCEtKnobS0lNLSUlatynYzwM1Nx4dhpArt3t5eurp76O1tJ9ib0cpICIQ3H8sTwPEEkL5AmuB2/MWgn/+wCXPRm9CCgwCEVt103sdJLDcSRDNHYwJ5HGG6QlmPTqBFJ5DR9DyfEIKS0jJqG6uprl6VJpLjQrm4uFjloWeZsrIy/u3f7uCOO+7gkUce4dFTeTQWOmysDLOxyqQ6//WL6p6gxvZeH9v7/Jwa0xBC8La33cinPvUpvEokKCYhGHQFc2bLOyEmGcZCjq4d00CmQI5TVRhmYMKfiI8kig1TMtIhLZB4LNWpZmIC+vrcjDQkX7RpgirMVijmBUpIvw4MwzjjQIxgMJhTaMdd7b6+dqxocqKcHagguPodM7H8M6KFhsnf/z8Ikg5MoKCAqqoqamuyRXJ8oIYazTw/8Hq9/MM//AMf/OAHeeaZZ3jyiSe4/+BB7j+Rz+Iimw2VYTZWRajIO3tR3R/S2N7rZUefn5OjrqhZvfoCPvGW67nmmmuoqKiYrpejWCCEQq4bnfpZWwhXY57RkZ5WIR3rD52Rka4tdi30juFA2nZxwR0SeYT0FEc6JSNNR4cb5YhnpOPfxKmCQ4Vi3qAUzzSTn5+fc5T4qVOnePzxx/njHx+jq6sToelEihuJVq+epZVm4/iLMRdtxDtwDG2iHyEEK1esZMuW67n66qsJBAKzvUTFFFBRUcG73vUu3vWud9HT08PTTz/Nk08+wa8OH+FXxwMsLbbZWBVmQ1WEMl+2qB4yBTt6fbzU6+f4iCtkVq5Yzsf+8i1ce+21VMdFgkJxFgSD6fnoOKcV0rH2d1KbPiGdnFiYHu2oKXKFdNugK5bjznX8urdwOR3l5XDM3TfhSGv50NoaO0iGI62mGyoU8wYlpGeQgYEBnnrqKf742GMcOXwYALuojkjzVVilzXNvyqEQRKtXE61ejRYaxhg8we6DR9m9+2W++93vcsUVV7BlyxY2btyIx3P+cRTF3KG6uppbbrmFW265hc7OTp566imeevIJfnH0BL88GmBVqYXlSAwNhk3BDw8WcWjIQAJLlyzmw//P9WzevDlrkqJCcbbEHelMTpcGyuraMQ1kDlqJU1OcLqTjQlkTrjt9pOE6qmu6YsdIyUh7AtDSEjtIDXv3wk+ev4X/y52IcBiFQjE/UEJ6mgkGgzz//PM8+sc/svvll5FSIgPlmI0bsMqWIL3zozOFk1dCpP5SInWXoE304Rk4zrMv7uCZZ54hECjguuuuZcuWLaxdu1a1rVsg1NXVcdttt3HbbbfR1tbGT37yE5599lmaCiwMTTJkarw2ZLBp0yY+/OEPs2jRotlesmIBcDpH2rKy74fJu3ZMJXpi0EruaEfSkU4d3OIQdTRsx31P9GhOwtGOeAugda+7YXU1/3Un3LnnKr5IORWZjvTXvw4bN8J1103561IoFK8PJaSnAcuyePnll/njH//Ic889TyRigr8Qs+YirPKlOHkls73E80cInIIqzIIqzMaN6KMdRAeOs+2RP/C73/2Oqqpq3vrWLWzZsiUrzqKYvyxatAjLsijPA7/hCoHmQpvyPLAtS4loxZQRCuUW0h4POJPE9c+na8e5YkzS/q660O3Xd2oonpFOHSUusWwNK9ZmT9dlclCLEYChWDF6dTWvvOLeHKaEitSMdDQKX/gCvO99SkgrFHMQJaSnCCklhw8f5rHHHuOxx59gdGQY4fFhlizGKl+KXVC18Brsaxp2SSN2SSNhO4ox1Ern4HHu+fnPueeee1i2fDk3vPWtXHfddarV2TxnfHycnTu285baEC1j7tuGELChIsTjO3cwNjamxr8rpoRgcPJoR0pddhoz2rUjo/2dz+NQFggzOOFH1xy0lLd5j+ZgORpWSo9pPTXaAVBQgAwUsGeP++MwJenFhidOuFb8wMD0vDCFQvG6UEL6dWLbNr/61a/Y9vDDdHZ0gNCIlizCWnYZVnEDTONXjXMK3YNVsQyrYhkiGsQYOMnRzmMc+8EP+Pd//3cuvewy/vK97+Wyyy6b7ZUqzoPt27cTtWwuq4zgpOiI9VURfn/KZvv27Vx//fWzt0DFgiEYzN35zeuFyaLDM9m1I9ORBrfgcHDCnxX7MHSHqC0SjrShSTyxaIdpxIR0TQ0tLTAy4v44QnG6kI7V09DfP3UvRqFQTBlKSL9OBgYG+K+77yYSf+OTDt7xbvTIBMbgCRxfEY6vEOkrdK+9gYXlTDsWmjmOMEfRzDG08BiaOYYRHUeEx5C4bv3Lu3Zh6LoS0vOUeOeN7b1e3r8imLj/pR5v2uMKxeslFIKqquz7162D0dHc+8xE147Jig3BzUkf7CrDmyGkE460k8ORNmL5lerqhBsNORzpQ4fca+VIKxRzEiWkXydVVVU8+D//Q0dHB52dnXR1dcVud9He3k5fTytOarBP08FfiOUpSIhsx58U2sy1MeNSIiwzKZRTLkZkDGlOpG3u8/mpraujsWEtdXV1iUttbS018RZPinnH2rVrec973sOvf/1r1pVFWVcRZW+/h8fa83j3u9/NhRdeONtLVCwQJis23Lx58n1msmuHJ6P9HUBNkZuT9hp22v3xjHS82NBIzUjrsRdZU5MupEUZRNqTd8QdaSWkFYo5yRxTbfOT/Px8li9fzvLly7MesyyL3t5eOjs70y7t7R10dp4kHA6lbS98ASxvAY63EMcfE9q+Qpy80tc18fBMiMgEWmg4IZJFTChr5hjSSq8gLy0ro76pjvr6dWliua6ujpKSEjW1cIHyoQ99iJd37eTHh1v4tH+YHx8uYklzMx/+8Idne2mKBcRk7e9OR9yRns6uHcZpHOl4C7ysaIfmYDmCqB0rNtScxHESjnRNDa+8AiUlMDwMfVol+08MsDZ+kLiQHhwE2558vKNCoZgVlJCeZgzDSIjMTKSUjIyMZInsjs5O2traGOk4ltjWDlQSXH3ztKxRhEYI7P9NYoKhEILaunoaVyxLrL2+vj4xxdEfn76leEPh8/n4py98kY9+5CN8YWcJmmHw+S98AZ/PN9tLUywgJnOkT4d/YhAA0zt9Ba+ny0jHW+B5jeyMdKoj7UmZbGhqsU8L1dXseQiWL4ddOyXDohyPlRIGP3zYFc+27SptVbitUMwplJCeRYQQlJSUUFJSwurVq4lGo+zatYunn36a48ePu9sYXiLFjUSqLpi2dUh/EWbTm/AMnkQf60ZKia5prFq1imuuuYbFixcrl1kBwOLFi/nKV7/Ks88+y9VXX82SJUtme0mKBcb5ONJ5oz3uvv7SaViRy+ky0oloh54R7dAkYUvHslPb38Ucad01JAYKmjh1CjZsgAN7LYYow7Bi31SOj3Okv5TGi5eTt+dFN96hhLRCMaeYNSEthLgRuAPQgZ9IKb8xW2uZTSKRCLt27eKpp57i+RdeIBQMIgyfOy68rhm7qH76O38IQbTqAqJVFyAiQYyhFlqGWrj77ru5++67qW9o4Lprr2Xz5s0sWbJEieo3OBs3bmTjxo2zvQzFAkTKyftIn468sV4ivgJsY/q+LdNF7hHhkHSkc3XtsEwjZ7GhKdxPC/vDywBoaIA8j81QtDThSEc7+7iEV/invCf4R/7M7dyxYsU0vDqFQnG+zIqQFkLowA+ALUA7sFMI8ZCU8uBsrGe2eOSRR/j+9+8kFIqJ55JFROubsYvqZq1tnvTmJ8aCi2gQY6iVtqEWfnbPPfzsZz+jvqGR73z7W9TW1s7K+hQKxcIl3t7uTEK6smUnFa27eO2a/xcA/1gv4cIcrT6mkLhIzuzMAW77O8iOdmR37Ui2v4torujvctyONyUlkOe1GI6WYMSEdF9rkCABDvZVugf8+c9h/374yEem+NUpFIrzZbYc6Q3AMSnlCQAhxL3AnwFvKCG9d+9eQpEowRVvxS6snXM9p6UnP+lUR0N4eg7Q0b6Pjo4OJaQVCsWUE4x1VjxTtGP1Mz9k2Y6f89rVHwMhyBvrJTTNQtprOLz/TYdZXTuU9djkjnT6ZENDd9BizrYpXCHda7lxlKIiyPdYjFKMJxbt6O1wiyhPRuvdA06kd0lSKBSzjzZLz1sPnEr5uT12XxpCiI8IIXYJIXb19fXN2OJmiuLiYgRgT/HgFie/DCe/bMqOByA9eTiBCsBdt0KhUEw1cSF9Jkc6f7gD3YrgH3eHlOSN9RIqml4hDXDl0h5K8yNZ95fkR/AadpZbbWRNNpQIAT4fRIRbpNsXLkLT3Nec57UYkUWJjHRvjyu6TwzH3s/Hx6frpSkUivNktoR0rpBtVvBMSvkjKeV6KeX6ysrKGVjWzFJcXIy0o+gj7YjoJCO7zgNz0ZswF71pyo6HlIjwCPpYN6CEtEKhmB5CsRq7MznSgeHO2HWHu/1oL6HC2RsKJIQb78juI+1gpU02jMVDvGDmlcDixfQNGZSX44ppr8WoLEg60gOuwdI5UkBY5ClHWqGYg8xWtKMdaEz5uQHonKW1zBrxjgf5R/4IuD2ko/9/e3ceHVd1JXr4t0tTabQkW5ZteZAtGeMREwwGzBQCYQ6EDs0UAjQdmgS/Tpqm0yHvsVYgnQ4kL48kTUNiwtQQCKQZmiGYMMViNJ7xhC3JtmTJxjLWrNJQw35/nFtSyYMsC1kqSftbS6tunXvurVPH0q3tU/ue488mnJrbOaoc8Y8CGcD/74Q7SAjU4WutxReoJbG1loTWOjTsFjwYlZ1DTs7RuzPeGDNyHcmIdPRxX8E8/M17j3pqx+HMyG8gN729W1miL0IwJkc6OvNHSgp0HHci/GwVNTd1reSYmhSmMeLlSIfD1DR23TxZkTaTGfuNSK/60XPIs8/wpdJnDlwxNxSCxAM/4h96CC66CA4yI6sxpg8GK5BeAUwXkalANXAVcM0gtWXQnHLKKbz44ouUlZVRXl7Otm3b2FpaSmXFJsJhb2TDl4Cm5hBKzekMsMNpOfBF705XRdqbSAjUdgbNyW31aFvXGrxp6RlMLy6muPgMioqKKCoqorCwkKSko7cwjDFm5OrNiHRCRyv+gMtTTq+rJiVQiy8SHpDUjp488+038fmUZ1YUdZZFc6Sb2pLwJ4Y6Y93UVGhuTYCcHPbuhegXrqnJIZoj6fiC7VBZSY2O6TzXtpSZzGje2O01v/vgXBLqx/PBvn0wpqsuNTUwbRq88AKce25n8Z497j7Fu++GO+/s/z4wZiQalEBaVUMishh4HTf93SOquvEwhw1L2dnZLFiwgAULFnSWBYNBdu7c2Rlgl5WVU1ZWRsPnpZ113Oh1TszodQ7aw/Li0hEgobUWX6AuZpTZ3cgiIkwoKOCY40/oDJiLiorIy8uzqe6MMQOmNyPSaQ1dX16m11eT2lgD4GbtaD3UUUdfTvqBudPRWTtK94xiWl7XIEVBAVS7QXX27oV589x2WpL75i+gqSRv2kQNl5KSGKI9lMj2hGJo+bjzHKqwpSGfTFJhx47ugfTq1S4NZMWKboF0lbfyeGVl/7xnY8wgziOtqn8G/jxYrx/PkpKSmDZt2gGLXdTW1lJeXt75U1paRuXOjUTC4UOc6UCpaWkUFxczvfg0pk2bRnFxMYWFhbZaoTFm0PVmRDqaH+22q0ltcoF06yAH0gcTXcVwd2M6J0/b01k+aRKsW+e2a2q6j0gD1JNNx/ulVCcWMjazjd0NaWyXad1ypPeV1tKguTSTTrh8OQkxgzFs3uwed+zo1h4LpI3pf7ay4RCSm5tLbm4uJ554YmdZR0cHlZWVVFRU0NFx4IhIVFZWFkVFReTn59soszEmLvVqRNoLpIMp6aTFBtJZ+VCz72g38YhEVzEEl0MdNWkSvPIKBINQVxebI90VSBfuW8du/0Qy/R0Ewz62hydDW1eOdOnr5UAuYRLZs/FzuqU8b/Jmkt0vkI6OglsgbUz/sUB6iEtOTqa4uJji4uLBbooxxnwhvQmkozN17J18Ahl1Vd1HpImvQDo6r7Q/McTk3KbO8kmT3Oj71q3ueXREOi3ZfbtYTzYANZrHVH8HPoFtzRPd9HfqgvPSDz7vPF/Vp83dA+noiHRFRbf2xI5Iqx54f6Ix5sgN1vR3xhhjTDe9Se1Ia9hFKCmVugmz3Yh04x4i4qM9vX/nzu8P0dSO6WMbSIj5tJ3kzVm1erV77Aqk3Yh0A26K0bpQJpn+DsZktLK9dRxEIp3LP5Zt6JoytWp7sOvkql0j0hUV7phoPS+QDgSgtrY/3qExxgJpY4wxcaG3I9It2RMIZBfgD9SRUVtBW8YYNM5WhoWueaNnjKvvVr5/IH2w1I7KUXPoCCeSmRJkTEYb9cEM6sh2edJLllBaJmT53Ch3VWlMcviePS5fZMYMaG93zz3R1A6AnbFLohlj+swCaWOMMXGhVyPS9btoyS6gZY/LFx6z+T3afOlQUjIALTwymSlBBGXWuO7LikcD6VWr3OPBbjbcmHuGO4c/yOh0N/q8nakuvaOjg7K2AmamVpAkQSobs1nyO5fy0bBiK/fwrzR95TJ30pj0jqoqKCx025YnbUz/sEDaGGNMXAgE3BoiPU1Vn96wi0D2BFrS3HRv2Y2VBFLjL60DYNaEOu66ZCUFOYFu5fn57j2uWeOedwbS3oh0rYxm45jTAcj0d5CX6aVzUAzNzWhVNaVMZ+yoVvKSGtit+SS/+WdYsoQXf7qRO7iHC1/8Ns2kd95wqOpGpKOj308/DUuWuB8A6upYvGgNP/1J92XOjTE9s0DaGGNMXGhtPczy4Kqk11W7Eek0F336NExbSvbANPAI+QTysw6ck8/nc3NJNze77Vzv/wEJPkhJDPHBMTewPW0O4Eakx2UF8EmETcyClhY+L6ungWyycxPI9QeoYiKZLbtZUnIsy7aOI4EQ7++exuU8j27fAUBDg8sKmTTJ/Wdl/xzpil+9wAMfHMdTj8TZHILGxDkLpI0xxsSFQKDn/Ojk1gYSg60ERk3oDKQBWv05A9C6/hVN7xg9GhJi0rtTk8IEgsk0tSUDkOUPkpSgFI+pZwNzoLmZsi1udo9Ro5MYldZBNQVktLhc6MqWMRyTUM4lcyt4g69SvclNuxfNj87JcT913bNNePT5USg+tlSm0h6z0nkkAuXl/f/+jRkuLJA2xhgTF1pbezf1XSB7AsGkdDoS3fB1qz8+R6R7Eg2k8/K6l6clhwgEE2lqc/ktGSluRo7ZBV4gvWULpRvdmgFjs9rIzIIqJpLR/BkAZcHJTPXvZka+u8Fx7SYXkEdn7ChqWcc4f323Eenf/hYe3nQKqQQIR3x8+mnXvj/9CaZP77oxcn+rVkEPSxgYM+xZIG2MMSYuBAKHv9EQoCW7wNX3RqWH8oh0NGc5KjU5RGtHIk3tSaQkhkhOdDnLcwrqKKOYtnWfUpo0C59EGJ3eRmYGtOMn3BhAWwJU6BQKshqZmNOCEGFN5WigK5C+aOli5ta8SW1t12IxpasbqYpM4Af8HID167va8847Lr/6d7/pmm4vautWWLDABeLGjFQWSBtjjIkLh0vtSGuIBtJu+ZGWVHfDYdsQDqQPGJFOCtEaTKCpLZlMf9f80LPH1xEmkS3MYFnuZcybWEtigpKT7oaDmxvDtO906R35OSH8SWEmJX3GmtopoNqZ2jGjbjlF7Zuor4ewyxBhdUkzuezjmhmrSKKDDSu7guYP3nWVnvqDy+mO9crTbvq9t97qjx4xZmiyQNoYY0xcONzNhul1XmrHKC+Q9mbuCAyjQDonrZ2qunQ27c7uFkjPKXBJze9Oupb39xRz8dzKzvoATU1C8kaXf+Gf4oa5izM+Y01kHuzdS1UV5Kc3kUyQyVSiKtTXuxzotdtHcQGvUXHWDRzLp6z/yEXMjY2wYbOPr/I6zSE/t94S6jbTx6sPuKn1SpZFOoNyY0YaC6SNMcbEhcONSKc37KItLYdwsou2W1JdFDqUR6T3T+249LgdLJiyl+b25M4gGdzqiIm+CPe1/gMR9XHxPBfERuvUtyazo208iRImN9ulbUzJbmAHU6lbX0V1NRSEK9l1zJmMHuWm2aupcQuz1Heksyh1Nbv2JTOX9az/RGHJEpYvB1Xhdn7JTDax/M2uZc4bq5soqZlBIdupb/B1SwcxZiSxQNoYY0xcONyIdObecgJeWgfAnrw51GdOoiV19AC0rn8VFUFmJsyc2b08wx/ipkVb+N8XrOaKL23rLE9OjDBjXD3bPh9FXmYrJ07ZC0CWvwMfEcop4qP0rzA2q40EnwukJ4xxaR9rSxrZvqWdiW1lbD/+cgrGtJNJIx+9sJuNr24HYF7eZ7SnjGJ6cgU7W/OoDyTz4bshhAgLr57GjUlPsmVPTmeKyH3XrSZEEnem/RKAe/69d/NP790LDz3EQUewg0F44gk3Em7MUGGBtDHGmLjQ04h0Wv0uJm5+g52zL+gsq5x4Ks9+7UkiCckD1ML+k5UFu3bBFVccfP/k3GZy09u7lc0e79I7LppTic/79E7wwdy8z/gV/8SKjvmMH9XSWX/MODev3g9/PZ7N5Smcw5vsmH8ZdRPncQOPsbJyLKu25TCfNUTyxwMwcZRL69iwK4cPHi9lDhvIGp3E6RPKSaad956uhJISNn7USDb1TPhSPlPZxs6VXUuR09gId9xx0HnzbrsNbr4ZHn64e7kq3HILfOtbcOedve5GYwadBdLGGGPiQk8j0se++xC+SJhNZ94ysI06ijIyQKT39WdPcIH0xfO6r+99/Zk7yM0MEggmMSG7axXFtKxkxmc08nHDDM7mLW49YTktuZPZlX883+VBQppAVVM25/E6NWPc0PiYsa5Bjy+bykfVEznF9zEceyx1U0/gcp7n4+1j+bwmzButp7EwazM7Jy1iUcJyNlWmE4kAqjx13uOcd89ZfHV2NU/+sivA3roVnnpKSU6K8KMfuUVhojnXl1wCjzziUl6WLIHdu/vYqcYMMAukjTFmhBCR80Vki4iUicgPD7I/RUSe8fYvF5HCmH13eOVbROS8o9G+Q41ISzjIzPeWsHPWeTTlFR2Nlx4Svn78di6eW8F5s3Z2K09PCbH4yxsoHN3YOWodtfCcLHJzIvzX99fg+/k9AAST0sga6+c8lgKwKG0ttdnTANCCicxnDb//eB4N4UxOn1wBSUlUjzuB6xOfpDnk56dvLsRHhAuP34X6Epk2oZX6cBZPnPl73v7Os1z30Xcpzzqe8vaJXH/7GO466RWWfLOEW09bR0qklZeD51O3L8Ll57UQiUBJCbz6KlwyeS1v/7iEYFD5xS9i3kQo5O6KNCYOJQ52A4wxxhx9IpIA/CdwLlAFrBCRl1R1U0y1m4A6VS0WkauAe4ErRWQWcBUwG5gAvCkix6hqv87VsHAhzJhxYPmUdS+TXr+L9655sD9fbsiZW1DHy4tfP+i+sZlt3HH+2gPKlyyB1lYfBZNvdwVl7mHVvBu5rqyEttYpVJ/xbXw+Fw7szZvN6sWPUvH+T9hansDZl7i7IiMJSQRPP5spJTv5LJzH45mL2TXxOwDknTyN45t3cMN7f0/6e83MSK9meeVEXvzbpfzsjXbuX7GQU9et4u2ORdyc+gT5haNZvPk/+M3K77Fr7XbKQ1O4kNd4rvIykm4K8U15nAd/fSWT9qznpPq/8OJf0qiJjKY4ay8L5nVw+j/MIm32VD5bVc2yD5PZ8Xk6Z86p5aTzc0nMH03bW++z7fVSGnYHCDW1UnRyHuNvOA8ZnQu7dtFRsZtQ1WekjkpGzjzDJaq3tLiUlIYGaG93Q+PjxnV9ZRAMuuUg09LcVwm9EYlAUxOkp7t12QeQalfTOzrc6pkJCe5t7NkD+fmuSWVlrolz57pvCJYuhSlT4OST4ZlnYNkyuPpqOOecI/v2ZCSxQNoYY0aGk4AyVd0GICJ/BC4FYgPpS4Efe9v/DdwvIuKV/1FV24HtIlLmne/D/mzg0qUHL5+17AGacyZROfei/ny5EWH/6fWidufPh3y4hr1AUrd9MncOhXPnULj/MRNO5IYL9kDbDvbmfquzPDnZx00/KWT10hpefSuFF9/MIHOUj9a8Qq47v4Z7/jKZFR3z+Jv0pcz86hSWp32Hk6bv5mer/oP/rP46J/Ix/2fGs/xx9rNcN2kZP1n5GqUfHsNtT50MLCCJIDmJjdTUj4YS8JWEUQRlblfjXgT5twjptNDCdWjsF+4bIePhJvy0EWQ8DbiVMBMIkUUjGewhgo920mknlzAJpNCOn89IkXZ8REDVe80WVPaBdP9CX+ha4EZEXf2I0oafIEF8oiRIBB+KTyL4iCAoPnFnVYQ2TUER/NJOBB9tmkIiIVKkg1b1067JpEkriRKmKYDow0IAAAoYSURBVJIOQIYEaNMUGjWDNGklXQLURrJp0xRG+9zqljWRMSQQYmzCPj4P5xAkmUSCZPgC1EdGAZAqrd7ru/eVSJAQSfiljUcf9XNC8nrenvBNsnzNLkqHrscj3T6ax4mA3+/+lxAIuJ/XXoNFizhaRGMbEMdEZC9QMdjtGELGAJ8PdiPMsGW/X0dmiqoeIqQZGCLyDeB8Vf177/l1wEJVXRxTZ4NXp8p7Xg4sxAXXH6nqk175w8Brqvrf+73GzcDN3tMZwJZ+aLr9rlkfRFk/WB9EHe1+6PU1e8iMSA/2h9BQIyIrVXXBYLfDDE/2+zUkHeyL2f1HUg5VpzfHoqpLgCVH3rRDs98164Mo6wfrg6h46ge72dAYY0aGKmBSzPOJwK5D1RGRRGAUUNvLY40xZsSxQNoYY0aGFcB0EZkqIsm4mwdf2q/OS8D13vY3gLfV5f+9BFzlzeoxFZgOfDxA7TbGmLg1ZFI7zBHr169XjdmP/X4NMaoaEpHFwOtAAvCIqm4UkbuBlar6EvAw8IR3M2EtLtjGq/cs7sbEEHBrf8/Y0QP7XbM+iLJ+sD6Iipt+GDI3GxpjjDHGGBNPLLXDGGOMMcaYPrBA2hhjjDHGmD6wQHqYOdwSwMZ8ESLyiIjUePMNG3PUjNRr2cH+xkQkV0TeEJFS7zFnMNt4tInIJBF5R0Q2i8hGEfmeVz7S+sEvIh+LyDqvH+7yyqeKyHKvH57xbh4e1kQkQUTWiMgr3vO46QMLpIeRmCWALwBmAVd7S/sa018eA84f7EaY4W2EX8se48C/sR8Cb6nqdOAt7/lwFgL+WVVnAicDt3r//iOtH9qBs1X1OGA+cL6InAzcC9zn9UMdcNMgtnGgfA/YHPM8bvrAAunhpXMJYFXtAKJLABvTL1S1BDebgzFH04i9lh3ib+xS4HFv+3HgsgFt1ABT1d2qutrbbsIFUAWMvH5QVW32niZ5PwqcDURXFR32/SAiE4GLgN97z4U46gMLpIeXAmBnzPMqr8wYY4YSu5Z1l6+qu8EFmcDYQW7PgBGRQuB4YDkjsB+8lIa1QA3wBlAO1KtqyKsyEv42fgX8AIh4z0cTR31ggfTw0qtlfI0xJs7ZtcwgIhnAc8D3VbVxsNszGFQ1rKrzcauJngTMPFi1gW3VwBGRi4EaVV0VW3yQqoPWB7Ygy/Biy/gaY4YDu5Z1t0dExqvqbhEZjxudHNZEJAkXRP9BVZ/3ikdcP0Spar2I/BWXM54tIoneiOxw/9tYBHxNRC4E/EAWboQ6bvrARqSHl94sAWyMMfHOrmXdxS7dfj3wP4PYlqPOy4F9GNisqv8vZtdI64c8Ecn2tlOBc3D54u8A3/CqDet+UNU7VHWiqhbirgNvq+q1xFEfWCA9jHj/M4suAbwZeFZVNw5uq8xwIiJPAx8CM0SkSkRGwt3iZoCN5GvZIf7G7gHOFZFS4Fzv+XC2CLgOOFtE1no/FzLy+mE88I6IfIL7z+UbqvoK8K/AbSJShssXfngQ2zhY4qYPbIlwY4wxxhhj+sBGpI0xxhhjjOkDC6SNMcYYY4zpAwukjTHGGGOM6QMLpI0xxhhjjOkDC6SNMcYYY4zpAwukTdwRkQ++4PE3iMj9X+D4HSIy5ou0RUQuE5FZfW2DMcYMdyJyloi8cpg6F4jIShHZLCKfisj/Haj2GdMbFkibuKOqpw52G6K+QFsuAyyQNsYYj4gkHGH9OcD9wDdVdSYwB9h2NNpmTF9ZIG3ijog0e4/jRaTEm4x/g4ic3sMxN4rIVhFZhpvMP1r+mIh8I+Z59Nxneed+QUQ2ichvReSAv4dofW/7ByKyXkTWicg9Xtm3RWSFV/aciKSJyKnA14BfeG0v8n6WisgqEXlXRI7th64yxpgB4V3//tHbvk9E3va2vyIiT4rI1d71cYOI3BtzXLOI3C0iy4FTROR8b2T5PeDyw7zsD4Cfquqn4BbqUdUHvPNOEZG3ROQT73GyV/6YiDwoIu+IyDYROVNEHvFGtB/br133etfkN0XkJBH5q3fM17w6fhF51Htfa0Tky175DSLyvHdNLxWRn/dTN5shyAJpE8+uAV5X1fnAccDag1USkfHAXbgA+lx6PxJ8EvDPwFygiB4u6iJyAW6UeaGqHgdEL5zPq+qJXtlm4CZV/QC3lO2/qOp8VS0HlgD/S1VPAG4HHuhlG40xJh6UANHBjAVAhogkAacBpcC9wNnAfOBEEbnMq5sObFDVhcBK4CHgEu9c4w7zmnOAVYfYdz/wX6o6D/gD8JuYfTleW/4JeBm4D5gNzBWR+THt+qt3TW4C/g33+fF14G6vzq0AqjoXuBp4XET83r75wJW4z48rRWTSYd6LGaYskDbxbAVwo4j8GJirqk2HqLcQd0Hcq6odwDO9PP/HqrpNVcPA07gPhEM5B3hUVQMAqlrrlc/xRpjXA9fiLtbdiEgGcCrwJxFZC/wOt/SrMcYMFauAE0QkE2jHLWO+ABcQ19N1DQ7hAtszvOPCwHPe9rHAdlUtVbes8pNfoD2nAE9520/Q/fr9snf+9cAeVV2vqhFgI1Do1ekAlnrb64Flqhr0tqN1TvPOjTcqXgEc4+17S1UbVLUN2ARM+QLvxQxhFkibuKWqJbiLcTXwhIh8q6fqhygP4f2ei4gAyT0cc6hzAMgh9j8GLPZGLO4C/Aep4wPqvdHp6M/MHl7LGGPiihdk7gBuBD4A3gW+jPs2r7KHQ9u8wYrOUx3By24ETuhtE2O2273HSMx29Hmitx30gu1u9byAO1pHeni92POGY44xI4wF0iZuicgUoEZVHwIeBr50iKrLgbNEZLT3VeMVMft20HUhvhRIitl3kohM9XKjrwTe66E5fwH+TkTSvLbleuWZwG7vda+Nqd/k7UNVG4HtInKFd6yIyHE9vJYxxsSjElxqWgkukL4Fl3L3EXCmiIzxbii8Glh2kOM/BaaKSJH3/OrDvN4vgB+JyDEAIuITkdu8fR8AV3nb19Lz9buvSrxz47VhMrDlKLyOGcIskDbx7CxgrYisAf4G+PXBKqnqbuDHuK8a3wRWx+x+CHeB/xiXAtISs+9D4B5gA7AdeOFQDVHVpbi855Veesbt3q47cYH8G7gPiag/Av/i3aBShLsY3yQi63CjLJce5r0bY0y8eReXlvahqu4B2oB3vWvwHcA7wDpgtar+z/4He2kQNwOvejcbVvT0Yqr6CfB94GkR2Yy7VkfT4v4Rl/r3CXAd8L1+eH/7ewBI8FL3ngFuUNX2wxxjRhjp+mbDmJFDRM4CblfViwe7LcYYY4wZmmxE2hhjjDHGmD6wEWkzpHhzkabsV3ydqq4fjPYYY4z5YkTkRg5MzXhfVW8djPYYcyQskDbGGGOMMaYPLLXDGGOMMcaYPrBA2hhjjDHGmD6wQNoYY4wxxpg+sEDaGGOMMcaYPrBA2hhjjDHGmD74/wz2Y+GbG27XAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 864x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 8))\n",
    "\n",
    "plt.subplot(1,2,1)\n",
    "sns.violinplot(x = 'is_duplicate', y = 'word_Common', data = df[0:])\n",
    "\n",
    "plt.subplot(1,2,2)\n",
    "sns.distplot(df[df['is_duplicate'] == 1.0]['word_Common'][0:] , label = \"1\", color = 'red')\n",
    "sns.distplot(df[df['is_duplicate'] == 0.0]['word_Common'][0:] , label = \"0\" , color = 'blue' )\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}