Sublevel Model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data_path = '/Users/apple/Desktop/Machine_Learning/project/Audio/data_audi/data_prepared'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/apple/anaconda3/lib/python3.5/site-packages/sklearn/cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from scipy import interp\n",
    "import matplotlib.pyplot as plt\n",
    "from itertools import cycle\n",
    "import pandas as pd\n",
    "from sklearn import svm, datasets\n",
    "from sklearn.metrics import roc_curve, auc\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "from sklearn import tree\n",
    "from sklearn import cross_validation\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.naive_bayes import BernoulliNB\n",
    "from sklearn.metrics import roc_curve, auc\n",
    "from sklearn.metrics import roc_auc_score\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data = pd.read_csv(data_path+'/sample_selected.csv',index_col = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data.to_csv(data_path+'/sample_selected_R.csv',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "data[:100].to_csv(data_path+'/sample_selected_test.csv',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>vowel</th>\n",
       "      <th>stress</th>\n",
       "      <th>pre_word</th>\n",
       "      <th>word</th>\n",
       "      <th>fol_word</th>\n",
       "      <th>F1</th>\n",
       "      <th>F2</th>\n",
       "      <th>F3</th>\n",
       "      <th>B1</th>\n",
       "      <th>...</th>\n",
       "      <th>F2@50%</th>\n",
       "      <th>F1@65%</th>\n",
       "      <th>F2@65%</th>\n",
       "      <th>F1@80%</th>\n",
       "      <th>F2@80%</th>\n",
       "      <th>nFormants</th>\n",
       "      <th>Speaker</th>\n",
       "      <th>Democratic</th>\n",
       "      <th>Republican</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>IH</td>\n",
       "      <td>1</td>\n",
       "      <td>SP</td>\n",
       "      <td>IN</td>\n",
       "      <td>MADSEN</td>\n",
       "      <td>430.4</td>\n",
       "      <td>1927.0</td>\n",
       "      <td>2641.6</td>\n",
       "      <td>436.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1972.7</td>\n",
       "      <td>422.3</td>\n",
       "      <td>1806.1</td>\n",
       "      <td>398.1</td>\n",
       "      <td>1745.1</td>\n",
       "      <td>5</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AE</td>\n",
       "      <td>1</td>\n",
       "      <td>IN</td>\n",
       "      <td>MADSEN</td>\n",
       "      <td>IN</td>\n",
       "      <td>562.9</td>\n",
       "      <td>1890.9</td>\n",
       "      <td>2478.6</td>\n",
       "      <td>141.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1846.5</td>\n",
       "      <td>599.8</td>\n",
       "      <td>1715.6</td>\n",
       "      <td>448.1</td>\n",
       "      <td>1670.9</td>\n",
       "      <td>5</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>IN</td>\n",
       "      <td>MADSEN</td>\n",
       "      <td>IN</td>\n",
       "      <td>381.9</td>\n",
       "      <td>1717.8</td>\n",
       "      <td>2673.1</td>\n",
       "      <td>100.4</td>\n",
       "      <td>...</td>\n",
       "      <td>2201.0</td>\n",
       "      <td>389.9</td>\n",
       "      <td>2218.7</td>\n",
       "      <td>389.9</td>\n",
       "      <td>2218.7</td>\n",
       "      <td>5</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>MADSEN</td>\n",
       "      <td>IN</td>\n",
       "      <td>MADSEN</td>\n",
       "      <td>430.7</td>\n",
       "      <td>1675.7</td>\n",
       "      <td>2497.5</td>\n",
       "      <td>160.5</td>\n",
       "      <td>...</td>\n",
       "      <td>1657.5</td>\n",
       "      <td>413.4</td>\n",
       "      <td>1608.0</td>\n",
       "      <td>413.4</td>\n",
       "      <td>1608.0</td>\n",
       "      <td>5</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AE</td>\n",
       "      <td>1</td>\n",
       "      <td>IN</td>\n",
       "      <td>MADSEN</td>\n",
       "      <td>AT</td>\n",
       "      <td>540.0</td>\n",
       "      <td>1959.1</td>\n",
       "      <td>2548.0</td>\n",
       "      <td>207.9</td>\n",
       "      <td>...</td>\n",
       "      <td>1879.4</td>\n",
       "      <td>576.4</td>\n",
       "      <td>1722.0</td>\n",
       "      <td>392.6</td>\n",
       "      <td>1629.6</td>\n",
       "      <td>6</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>IN</td>\n",
       "      <td>MADSEN</td>\n",
       "      <td>AT</td>\n",
       "      <td>408.8</td>\n",
       "      <td>1570.2</td>\n",
       "      <td>2677.6</td>\n",
       "      <td>281.2</td>\n",
       "      <td>...</td>\n",
       "      <td>1581.7</td>\n",
       "      <td>386.0</td>\n",
       "      <td>1598.8</td>\n",
       "      <td>386.0</td>\n",
       "      <td>1598.8</td>\n",
       "      <td>5</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>MADSEN</td>\n",
       "      <td>AT</td>\n",
       "      <td>LEAST</td>\n",
       "      <td>451.4</td>\n",
       "      <td>1470.7</td>\n",
       "      <td>2545.6</td>\n",
       "      <td>205.5</td>\n",
       "      <td>...</td>\n",
       "      <td>1467.8</td>\n",
       "      <td>451.1</td>\n",
       "      <td>1443.7</td>\n",
       "      <td>451.1</td>\n",
       "      <td>1443.7</td>\n",
       "      <td>5</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>IY</td>\n",
       "      <td>1</td>\n",
       "      <td>AT</td>\n",
       "      <td>LEAST</td>\n",
       "      <td>THE</td>\n",
       "      <td>398.6</td>\n",
       "      <td>1924.7</td>\n",
       "      <td>2737.6</td>\n",
       "      <td>88.2</td>\n",
       "      <td>...</td>\n",
       "      <td>1960.9</td>\n",
       "      <td>348.6</td>\n",
       "      <td>1953.8</td>\n",
       "      <td>348.8</td>\n",
       "      <td>1953.3</td>\n",
       "      <td>5</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>LEAST</td>\n",
       "      <td>THE</td>\n",
       "      <td>COURT</td>\n",
       "      <td>391.6</td>\n",
       "      <td>1717.5</td>\n",
       "      <td>2518.5</td>\n",
       "      <td>111.6</td>\n",
       "      <td>...</td>\n",
       "      <td>1828.2</td>\n",
       "      <td>366.1</td>\n",
       "      <td>1878.8</td>\n",
       "      <td>366.1</td>\n",
       "      <td>1878.8</td>\n",
       "      <td>6</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AO</td>\n",
       "      <td>1</td>\n",
       "      <td>THE</td>\n",
       "      <td>COURT</td>\n",
       "      <td>ASSUMED</td>\n",
       "      <td>498.3</td>\n",
       "      <td>1096.9</td>\n",
       "      <td>2200.7</td>\n",
       "      <td>118.1</td>\n",
       "      <td>...</td>\n",
       "      <td>1124.8</td>\n",
       "      <td>513.4</td>\n",
       "      <td>1169.5</td>\n",
       "      <td>505.3</td>\n",
       "      <td>1209.2</td>\n",
       "      <td>3</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>COURT</td>\n",
       "      <td>ASSUMED</td>\n",
       "      <td>THEY</td>\n",
       "      <td>444.8</td>\n",
       "      <td>1607.0</td>\n",
       "      <td>2658.3</td>\n",
       "      <td>184.9</td>\n",
       "      <td>...</td>\n",
       "      <td>1618.2</td>\n",
       "      <td>448.8</td>\n",
       "      <td>1629.0</td>\n",
       "      <td>448.8</td>\n",
       "      <td>1629.0</td>\n",
       "      <td>5</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>UW</td>\n",
       "      <td>1</td>\n",
       "      <td>COURT</td>\n",
       "      <td>ASSUMED</td>\n",
       "      <td>THEY</td>\n",
       "      <td>423.4</td>\n",
       "      <td>1420.0</td>\n",
       "      <td>2738.7</td>\n",
       "      <td>91.9</td>\n",
       "      <td>...</td>\n",
       "      <td>1228.8</td>\n",
       "      <td>424.0</td>\n",
       "      <td>1143.6</td>\n",
       "      <td>424.3</td>\n",
       "      <td>1133.9</td>\n",
       "      <td>6</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>12 rows × 46 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               name vowel  stress pre_word     word fol_word     F1      F2  \\\n",
       "0   JOHNPAULSTEVENS    IH       1       SP       IN   MADSEN  430.4  1927.0   \n",
       "1   JOHNPAULSTEVENS    AE       1       IN   MADSEN       IN  562.9  1890.9   \n",
       "2   JOHNPAULSTEVENS    AH       0       IN   MADSEN       IN  381.9  1717.8   \n",
       "3   JOHNPAULSTEVENS    AH       0   MADSEN       IN   MADSEN  430.7  1675.7   \n",
       "4   JOHNPAULSTEVENS    AE       1       IN   MADSEN       AT  540.0  1959.1   \n",
       "5   JOHNPAULSTEVENS    AH       0       IN   MADSEN       AT  408.8  1570.2   \n",
       "6   JOHNPAULSTEVENS    AH       0   MADSEN       AT    LEAST  451.4  1470.7   \n",
       "7   JOHNPAULSTEVENS    IY       1       AT    LEAST      THE  398.6  1924.7   \n",
       "8   JOHNPAULSTEVENS    AH       0    LEAST      THE    COURT  391.6  1717.5   \n",
       "9   JOHNPAULSTEVENS    AO       1      THE    COURT  ASSUMED  498.3  1096.9   \n",
       "10  JOHNPAULSTEVENS    AH       0    COURT  ASSUMED     THEY  444.8  1607.0   \n",
       "11  JOHNPAULSTEVENS    UW       1    COURT  ASSUMED     THEY  423.4  1420.0   \n",
       "\n",
       "        F3     B1  ...   F2@50%  F1@65%  F2@65%  F1@80%  F2@80%  nFormants  \\\n",
       "0   2641.6  436.0  ...   1972.7   422.3  1806.1   398.1  1745.1          5   \n",
       "1   2478.6  141.0  ...   1846.5   599.8  1715.6   448.1  1670.9          5   \n",
       "2   2673.1  100.4  ...   2201.0   389.9  2218.7   389.9  2218.7          5   \n",
       "3   2497.5  160.5  ...   1657.5   413.4  1608.0   413.4  1608.0          5   \n",
       "4   2548.0  207.9  ...   1879.4   576.4  1722.0   392.6  1629.6          6   \n",
       "5   2677.6  281.2  ...   1581.7   386.0  1598.8   386.0  1598.8          5   \n",
       "6   2545.6  205.5  ...   1467.8   451.1  1443.7   451.1  1443.7          5   \n",
       "7   2737.6   88.2  ...   1960.9   348.6  1953.8   348.8  1953.3          5   \n",
       "8   2518.5  111.6  ...   1828.2   366.1  1878.8   366.1  1878.8          6   \n",
       "9   2200.7  118.1  ...   1124.8   513.4  1169.5   505.3  1209.2          3   \n",
       "10  2658.3  184.9  ...   1618.2   448.8  1629.0   448.8  1629.0          5   \n",
       "11  2738.7   91.9  ...   1228.8   424.0  1143.6   424.3  1133.9          6   \n",
       "\n",
       "            Speaker Democratic Republican label  \n",
       "0   JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "1   JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "2   JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "3   JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "4   JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "5   JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "6   JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "7   JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "8   JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "9   JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "10  JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "11  JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "\n",
       "[12 rows x 46 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[:12]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "data[['name', 'vowel', 'stress','word',  'F1', 'F2',\n",
    "       'F3', 'B1', 'B2', 'B3',  'dur', 'F1@20%', 'F2@20%',\n",
    "       'F1@35%', 'F2@35%', 'F1@50%', 'F2@50%', 'F1@65%', 'F2@65%', 'F1@80%','F2@80%','label']].to_csv(data_path+'/sample_selected_R.csv',index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['name', 'vowel', 'stress', 'pre_word', 'word', 'fol_word', 'F1', 'F2',\n",
       "       'F3', 'B1', 'B2', 'B3', 't', 'beg', 'end', 'dur', 'plt_vclass',\n",
       "       'plt_manner', 'plt_place', 'plt_voice', 'plt_preseg', 'plt_folseq',\n",
       "       'style', 'glide', 'pre_seg', 'fol_seg', 'context', 'vowel_index',\n",
       "       'pre_word_trans', 'word_trans', 'fol_word_trans', 'F1@20%', 'F2@20%',\n",
       "       'F1@35%', 'F2@35%', 'F1@50%', 'F2@50%', 'F1@65%', 'F2@65%', 'F1@80%',\n",
       "       'F2@80%', 'nFormants', 'Speaker', 'Democratic', 'Republican', 'label'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "% matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 0, 2])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.stress.unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Vowel Level Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "vowel_count = pd.DataFrame(data.groupby(['label','vowel'])['vowel'].count())\n",
    "vowel_count['Party'] = list(map(lambda x: 'Democrat' if x==1 else 'Republic',vowel_count.index.get_level_values('label')))\n",
    "vowel_count['Vowel'] = list(vowel_count.index.get_level_values('vowel'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x104ae50f0>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "sns.set(style=\"whitegrid\")\n",
    "\n",
    "# Draw a nested barplot to show survival for class and sex\n",
    "g = sns.factorplot(x=\"Vowel\", y=\"vowel\", hue=\"Party\", data=vowel_count,\n",
    "                   size=8, kind=\"bar\", palette=\"muted\")\n",
    "g.despine(left=True)\n",
    "g.set_ylabels(\"survival probability\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def distribution_vowel_figure(vowel, data):\n",
    "    data_AA = data[data['vowel']==vowel][['label','vowel', 'stress', 'dur',\n",
    "                          'F1', 'F2','F3', 'B1', 'B2', 'B3']]\n",
    "    data_figure = pd.DataFrame(list(data_AA['label'])*6,columns=['label'])\n",
    "    value = list(data_AA['F1'])+list(data_AA['F2'])+list(data_AA['F3'])+list(data_AA['B1'])+list(data_AA['B2'])+list(data_AA['B3'])\n",
    "    data_figure['value'] = value\n",
    "    data_figure['feature'] = ['F1']*len(data_AA)+['F2']*len(data_AA)+['F3']*len(data_AA)+['B1']*len(data_AA)+['B2']*len(data_AA)+['B3']*len(data_AA)\n",
    "    sns.set(style=\"whitegrid\", palette=\"pastel\", color_codes=True)\n",
    "\n",
    "    # Load the example tips dataset\n",
    "    #tips = sns.load_dataset(\"tips\")\n",
    "    plt.figure(figsize=(15, 10))\n",
    "    # Draw a nested violinplot and split the violins for easier comparison\n",
    "    sns.violinplot(x=\"feature\", y=\"value\", hue=\"label\", data=data_figure, split=True,\n",
    "               inner=\"quart\", palette={1: \"b\", -1: \"y\"},size=10)\n",
    "    sns.despine(left=True)\n",
    "    plt.title('Important feature distribution of vowel'+vowel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA34AAAJaCAYAAABwY5WAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11b80b390>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA34AAAJaCAYAAABwY5WAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDwAAAAAS7v8C+7+cDwZWXX4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11a934668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for vowel in data[data['word']=='AMENDMENT'].vowel.unique():\n",
    "    distribution_vowel_figure(vowel,data[data['word']=='AMENDMENT'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from scipy import stats\n",
    "np.random.seed(12345678)\n",
    "def ttest_main_feature(vowel,data):\n",
    "    data_p = data[data['vowel']==vowel][data['label']==1][['stress', 'dur',\n",
    "                          'F1', 'F2','F3', 'B1', 'B2', 'B3']]#,'F1@20%', 'F2@20%','F1@35%', 'F2@35%', 'F1@50%', 'F2@50%', 'F1@65%', 'F2@65%', 'F1@80%','F2@80%','nFormants']]\n",
    "    data_n = data[data['vowel']==vowel][data['label']==-1][['stress', 'dur',\n",
    "                          'F1', 'F2','F3', 'B1', 'B2', 'B3']]#,'F1@20%', 'F2@20%','F1@35%', 'F2@35%', 'F1@50%', 'F2@50%', 'F1@65%', 'F2@65%', 'F1@80%','F2@80%','nFormants']]\n",
    "    print(stats.ttest_ind(data_p, data_n))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/apple/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:4: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "/Users/apple/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:6: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ttest_indResult(statistic=array([  2.09467112,  17.61977419,  20.19927834,  34.11501114,\n",
      "        25.15047303,  12.25442664,  -3.55723305,   6.18501249]), pvalue=array([  3.62008892e-002,   1.84971757e-069,   1.10716553e-090,\n",
      "         1.06357520e-254,   1.81126465e-139,   1.61445265e-034,\n",
      "         3.74826580e-004,   6.21595767e-010]))\n",
      "Ttest_indResult(statistic=array([ -2.04576937,  11.56726154,  10.35200777,  30.38547853,\n",
      "        11.47681165,   9.26120101, -20.39167906,  -2.86136938]), pvalue=array([  4.07803182e-002,   6.17075228e-031,   4.15582625e-025,\n",
      "         2.37100459e-202,   1.76203785e-030,   2.03976378e-020,\n",
      "         2.44032648e-092,   4.21857191e-003]))\n",
      "Ttest_indResult(statistic=array([ 12.88117432,  26.3372738 ,  39.78905033,  48.09778345,\n",
      "        43.85714597,  13.68275241, -11.48111323,  16.51863158]), pvalue=array([  5.78187053e-038,   8.00983605e-153,   0.00000000e+000,\n",
      "         0.00000000e+000,   0.00000000e+000,   1.29764177e-042,\n",
      "         1.64810927e-030,   2.74062473e-061]))\n",
      "Ttest_indResult(statistic=array([ -0.80185184,  20.88923871,   9.69506421,  41.73530056,\n",
      "        20.58361416,   4.99222811,  -0.09595119,   5.00411295]), pvalue=array([  4.22639375e-01,   8.00884761e-97,   3.19097322e-22,\n",
      "         0.00000000e+00,   4.54898522e-94,   5.97238187e-07,\n",
      "         9.23559408e-01,   5.61550649e-07]))\n",
      "Ttest_indResult(statistic=array([  1.66608354,   3.20692211,   9.96441715,  18.0141031 ,\n",
      "        12.04899187,   6.51659244,   6.19676481,   0.52873727]), pvalue=array([  9.57002466e-02,   1.34210058e-03,   2.24305735e-23,\n",
      "         2.02362758e-72,   2.08291330e-33,   7.22993060e-11,\n",
      "         5.78834735e-10,   5.96989017e-01]))\n",
      "Ttest_indResult(statistic=array([  5.34948714,  11.06391503,   7.36072881,   4.98466901,\n",
      "        14.636864  ,  -2.08913256,  10.36763124,   5.13373483]), pvalue=array([  8.83889640e-08,   1.94690024e-28,   1.84246633e-13,\n",
      "         6.21671110e-07,   1.82589318e-48,   3.66982083e-02,\n",
      "         3.57989775e-25,   2.84556599e-07]))\n",
      "Ttest_indResult(statistic=array([  0.55159216,  12.39681032,  13.9673512 ,  23.84938006,\n",
      "         8.01611086,  12.40163135,  -3.16551235,  -1.59107675]), pvalue=array([  5.81228657e-001,   2.83342223e-035,   2.63464660e-044,\n",
      "         1.79318076e-125,   1.09947318e-015,   2.66815668e-035,\n",
      "         1.54842002e-003,   1.11594478e-001]))\n",
      "Ttest_indResult(statistic=array([  4.76255124,   4.07666568,  11.63187813,   6.62504689,\n",
      "         9.91539535,  -3.86033636,   1.8484806 ,   2.40076341]), pvalue=array([  1.91752286e-06,   4.57641490e-05,   3.14415883e-31,\n",
      "         3.51021397e-11,   3.76808163e-23,   1.13388417e-04,\n",
      "         6.45393723e-02,   1.63649248e-02]))\n",
      "Ttest_indResult(statistic=array([-1.6956238 ,  4.13251005,  6.76677932,  8.82624087,  0.90799551,\n",
      "        7.07134068,  3.10482787, -1.1166578 ]), pvalue=array([  8.99689858e-02,   3.59933455e-05,   1.34444557e-11,\n",
      "         1.14838016e-18,   3.63888915e-01,   1.57278520e-12,\n",
      "         1.90592001e-03,   2.64150871e-01]))\n",
      "Ttest_indResult(statistic=array([ -4.13111093,  13.26716428,   9.44827208,  26.22482313,\n",
      "         6.42385424,  14.14691689,   7.88180321,  -3.55189837]), pvalue=array([  3.61252211e-005,   3.82442483e-040,   3.50208456e-021,\n",
      "         3.61337278e-151,   1.33348290e-010,   2.11799032e-045,\n",
      "         3.25307182e-015,   3.82605196e-004]))\n",
      "Ttest_indResult(statistic=array([ -1.12969066,  14.15970425,  19.84971438,  38.33997763,\n",
      "        18.04880834,  21.27073266,  -1.03204085,  -0.53902776]), pvalue=array([  2.58607791e-001,   1.70141937e-045,   1.31672151e-087,\n",
      "         0.00000000e+000,   9.07021403e-073,   2.65965087e-100,\n",
      "         3.02054084e-001,   5.89868229e-001]))\n",
      "Ttest_indResult(statistic=array([ -3.62637688,   4.54562523,  15.83482906,  25.32994625,\n",
      "        13.96555458,   0.50080785,  -2.77666602,   2.19680704]), pvalue=array([  2.87498439e-004,   5.48050235e-006,   1.93963611e-056,\n",
      "         2.52575486e-141,   2.65599365e-044,   6.16506921e-001,\n",
      "         5.49247808e-003,   2.80353940e-002]))\n",
      "Ttest_indResult(statistic=array([  1.30305475,   9.99341528,  20.94060942,   2.31230811,\n",
      "        13.47249453,  13.63142386,  10.36488611,   3.81858076]), pvalue=array([  1.92559866e-01,   1.68192100e-23,   4.20017314e-97,\n",
      "         2.07632661e-02,   2.52246906e-41,   2.90791473e-42,\n",
      "         3.71739955e-25,   1.34324040e-04]))\n",
      "Ttest_indResult(statistic=array([ -4.19016612,   7.36659791,  11.95723652,  22.66217672,\n",
      "         9.0038219 ,   8.28101687,  14.09154869,  -1.11787084]), pvalue=array([  2.78946124e-005,   1.76135502e-013,   6.20897413e-033,\n",
      "         1.81790742e-113,   2.21021295e-019,   1.23344541e-016,\n",
      "         4.64433342e-045,   2.63624361e-001]))\n",
      "Ttest_indResult(statistic=array([ 0.82921542,  3.31594136,  4.25875319,  6.71939478,  3.1674437 ,\n",
      "        0.82269964, -3.84139344, -0.55003272]), pvalue=array([  4.07007343e-01,   9.17431522e-04,   2.07929991e-05,\n",
      "         1.95023944e-11,   1.54365640e-03,   4.10703422e-01,\n",
      "         1.23286409e-04,   5.82312338e-01]))\n"
     ]
    }
   ],
   "source": [
    "data_fill = data.fillna(method ='backfill')\n",
    "for vowel in data_fill.vowel.unique():\n",
    "    ttest_main_feature(vowel,data_fill)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'data_fill' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-18-b0e3c4a7529f>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0mvowel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'AA'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m data_p = data_fill[data_fill['vowel']==vowel][data_fill['label']==1][['stress', 'dur',\n\u001b[0m\u001b[1;32m      3\u001b[0m                           'F1', 'F2','F3', 'B1', 'B2', 'B3']]#,'F1@20%', 'F2@20%','F1@35%', 'F2@35%', 'F1@50%', 'F2@50%', 'F1@65%', 'F2@65%', 'F1@80%','F2@80%','nFormants']]\n\u001b[1;32m      4\u001b[0m data_n = data_fill[data_fill['vowel']==vowel][data_fill['label']==-1][['stress', 'dur',\n\u001b[1;32m      5\u001b[0m                           'F1', 'F2','F3', 'B1', 'B2', 'B3']]#,'F1@20%', 'F2@20%','F1@35%', 'F2@35%', 'F1@50%', 'F2@50%', 'F1@65%', 'F2@65%', 'F1@80%','F2@80%','nFormants']]\n",
      "\u001b[0;31mNameError\u001b[0m: name 'data_fill' is not defined"
     ]
    }
   ],
   "source": [
    "vowel = 'AA'\n",
    "data_p = data_fill[data_fill['vowel']==vowel][data_fill['label']==1][['stress', 'dur',\n",
    "                          'F1', 'F2','F3', 'B1', 'B2', 'B3']]#,'F1@20%', 'F2@20%','F1@35%', 'F2@35%', 'F1@50%', 'F2@50%', 'F1@65%', 'F2@65%', 'F1@80%','F2@80%','nFormants']]\n",
    "data_n = data_fill[data_fill['vowel']==vowel][data_fill['label']==-1][['stress', 'dur',\n",
    "                          'F1', 'F2','F3', 'B1', 'B2', 'B3']]#,'F1@20%', 'F2@20%','F1@35%', 'F2@35%', 'F1@50%', 'F2@50%', 'F1@65%', 'F2@65%', 'F1@80%','F2@80%','nFormants']]\n",
    "stats.ttest_ind(data_p, data_n)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from mne import Epochs, pick_types\n",
    "from mne.io import concatenate_raws\n",
    "from mne.io.edf import read_raw_edf\n",
    "from mne.datasets import eegbci\n",
    "from mne.event import find_events\n",
    "\n",
    "from pyriemann.stats import PermutationTest\n",
    "from pyriemann.estimation import Covariances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.69911020775865129"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(data.label == 1)/len(data.label == 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/apple/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:9: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "/Users/apple/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:13: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "/Users/apple/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:18: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "/Users/apple/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:22: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n"
     ]
    }
   ],
   "source": [
    "data_AH = data[data['vowel']=='AH'][['Speaker','word','label','vowel', 'stress', 'dur','pre_seg','fol_seg',\n",
    "                          'F1', 'F2','F3', 'B1', 'B2', 'B3','F1@20%', 'F2@20%','F1@35%', 'F2@35%', 'F1@50%', 'F2@50%', 'F1@65%', 'F2@65%', 'F1@80%','F2@80%','nFormants']]\n",
    "# test vowel by word model \n",
    "feature_V1 = data_AH\n",
    "feature_V1 = feature_V1.dropna()\n",
    "from sklearn import preprocessing\n",
    "vowel_encoder = preprocessing.LabelEncoder()\n",
    "vowel_encoder.fit(feature_V1.vowel)\n",
    "feature_V1['vowel_num'] = vowel_encoder.transform(feature_V1.vowel) \n",
    "\n",
    "word_encoder = preprocessing.LabelEncoder()\n",
    "word_encoder.fit(feature_V1.word)\n",
    "feature_V1['word_num'] = word_encoder.transform(feature_V1.word) \n",
    "\n",
    "\n",
    "pre_seg_encoder = preprocessing.LabelEncoder()\n",
    "pre_seg_encoder.fit(feature_V1.pre_seg)\n",
    "feature_V1['pre_seg_num'] = pre_seg_encoder.transform(feature_V1.pre_seg) \n",
    "\n",
    "fol_seg_encoder = preprocessing.LabelEncoder()\n",
    "fol_seg_encoder.fit(feature_V1.fol_seg)\n",
    "feature_V1['fol_seg_num'] = fol_seg_encoder.transform(feature_V1.fol_seg)\n",
    "\n",
    "feature_V1.index = feature_V1['Speaker']\n",
    "label_speaker = feature_V1.groupby(['Speaker'])['label'].mean()\n",
    "speaker = np.array(list(label_speaker.index))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/apple/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:4: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n"
     ]
    }
   ],
   "source": [
    "from random import shuffle\n",
    "p_l = list(feature_V1['label'])\n",
    "shuffle(p_l)\n",
    "feature_V1['label'] = p_l"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAl8AAAJaCAYAAAARciKuAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1198b1cf8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "speaker_pred = dict.fromkeys(list(pd.Series(feature_V1.index).unique()))\n",
    "for key in speaker_pred.keys():\n",
    "    speaker_pred[key]=[]\n",
    "    \n",
    "X, y = feature_V1.drop(['Speaker','label','vowel','word','pre_seg','fol_seg'],axis=1), feature_V1['label']\n",
    "\n",
    "cv = StratifiedKFold(n_splits=3)\n",
    "classifier = tree.DecisionTreeClassifier(max_depth=8)#, min_samples_leaf=300)\n",
    "#classifier =svm.SVC(kernel = 'rbf', probability=True)\n",
    "#classifier = GradientBoostingClassifier(n_estimators=50)\n",
    "mean_tpr = 0.0\n",
    "mean_fpr = np.linspace(0, 1, 100)\n",
    "\n",
    "colors = cycle(['cyan', 'indigo', 'seagreen', 'yellow', 'blue', 'darkorange'])\n",
    "lw = 2\n",
    "\n",
    "i = 0\n",
    "plt.figure(figsize = (10,10))\n",
    "accuracy_all = []\n",
    "for (train, test), color in zip(cv.split(speaker,label_speaker), colors):\n",
    "    train_s = speaker[train]\n",
    "    test_s = speaker[test]\n",
    "    clf = classifier.fit(X.ix[train_s], y.ix[train_s])\n",
    "    probas_,label = clf.predict_proba(X.ix[test_s]),clf.predict(X.ix[test_s])\n",
    "    # Compute ROC curve and area the curve\n",
    "    acc = []\n",
    "    prediction_byspeaker = pd.DataFrame(y.ix[test_s] == label )\n",
    "    prediction_byspeaker['Speaker'] = list(prediction_byspeaker .index.get_level_values('Speaker'))\n",
    "    prediction_byspeaker = prediction_byspeaker.groupby(['Speaker','label'])['label'].count()\n",
    "    for s in prediction_byspeaker.index.get_level_values('Speaker').unique():\n",
    "        p = prediction_byspeaker.ix[s].idxmax()\n",
    "        acc.append(p)\n",
    "        speaker_pred[s].append(p)\n",
    "    accuracy = sum(acc)/len(acc)\n",
    "    accuracy_all.append(accuracy)\n",
    "    fpr, tpr, thresholds = roc_curve(y.ix[test_s], probas_[:, 1])\n",
    "    roc_auc = auc(fpr, tpr)\n",
    "    if roc_auc < 0.5:\n",
    "        fpr, tpr, thresholds = roc_curve(y.ix[test_s], 1-probas_[:, 1])\n",
    "        roc_auc = auc(fpr, tpr)\n",
    "    mean_tpr += interp(mean_fpr, fpr, tpr)\n",
    "    mean_tpr[0] = 0.0\n",
    "    plt.plot(fpr, tpr, lw=lw, color=color,\n",
    "             label='ROC fold %d (area = %0.2f) (Accuracy by Speaker = %0.2f)' % (i, roc_auc, accuracy))\n",
    "\n",
    "    i += 1\n",
    "    \n",
    "plt.plot([0, 1], [0, 1], linestyle='--', lw=lw, color='k',\n",
    "         label='Luck')\n",
    "\n",
    "mean_tpr /= cv.get_n_splits(speaker,label_speaker)\n",
    "mean_tpr[-1] = 1.0\n",
    "mean_auc = auc(mean_fpr, mean_tpr)\n",
    "#speaker_pred = pd.DataFrame(speaker_pred).T\n",
    "#speaker_pred.sum()[0]/len(speaker_pred)\n",
    "plt.plot(mean_fpr, mean_tpr, color='g', linestyle='--',\n",
    "         label='Mean ROC (area = %0.2f) (Accuracy by Speaker = %0.2f) ' % (mean_auc,sum(accuracy_all)/len(accuracy_all)), lw=lw)\n",
    "\n",
    "plt.xlim([-0.05, 1.05])\n",
    "plt.ylim([-0.05, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver operating characteristic example')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Word Selection\n",
    "* 1 Frequency\n",
    "* 2 Balance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count     24209.000000\n",
       "mean        129.395101\n",
       "std        1501.856421\n",
       "min           1.000000\n",
       "25%           3.000000\n",
       "50%           8.000000\n",
       "75%          32.000000\n",
       "max      151130.000000\n",
       "Name: word, dtype: float64"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "word_freq = data.groupby(['word'])['word'].count()\n",
    "word_freq.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x24937b828>"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAskAAAHfCAYAAACiUkX2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x117702208>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "% matplotlib inline\n",
    "np.log(word_freq).hist(bins=100,figsize=(12,8))\n",
    "plt.title('Histogram of word frequency (log10)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count       236.000000\n",
       "mean       7218.436441\n",
       "std       13348.329311\n",
       "min        2002.000000\n",
       "25%        2436.750000\n",
       "50%        3401.000000\n",
       "75%        6645.750000\n",
       "max      151130.000000\n",
       "Name: word, dtype: float64"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "word_freq[word_freq>2000].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "word_selected = list(word_freq[word_freq>2000].index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8183"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "word_freq[word_freq>2000]['GOVERNMENT']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x11a4976d8>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD8CAYAAAB5Pm/hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n3AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x119875d30>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "word_freq[word_freq>2000].hist(bins=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>vowel</th>\n",
       "      <th>stress</th>\n",
       "      <th>pre_word</th>\n",
       "      <th>word</th>\n",
       "      <th>fol_word</th>\n",
       "      <th>F1</th>\n",
       "      <th>F2</th>\n",
       "      <th>F3</th>\n",
       "      <th>B1</th>\n",
       "      <th>...</th>\n",
       "      <th>F2@50%</th>\n",
       "      <th>F1@65%</th>\n",
       "      <th>F2@65%</th>\n",
       "      <th>F1@80%</th>\n",
       "      <th>F2@80%</th>\n",
       "      <th>nFormants</th>\n",
       "      <th>Speaker</th>\n",
       "      <th>Democratic</th>\n",
       "      <th>Republican</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1512</th>\n",
       "      <td>MARKDOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>FIRST</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>437.8</td>\n",
       "      <td>1627.8</td>\n",
       "      <td>2460.9</td>\n",
       "      <td>190.2</td>\n",
       "      <td>...</td>\n",
       "      <td>1618.9</td>\n",
       "      <td>418.0</td>\n",
       "      <td>1604.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>1604.0</td>\n",
       "      <td>6</td>\n",
       "      <td>MARKDOLSON</td>\n",
       "      <td>12.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1513</th>\n",
       "      <td>MARKDOLSON</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>FIRST</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>569.1</td>\n",
       "      <td>1702.9</td>\n",
       "      <td>2964.0</td>\n",
       "      <td>189.9</td>\n",
       "      <td>...</td>\n",
       "      <td>1687.1</td>\n",
       "      <td>593.5</td>\n",
       "      <td>1683.4</td>\n",
       "      <td>589.7</td>\n",
       "      <td>1651.1</td>\n",
       "      <td>5</td>\n",
       "      <td>MARKDOLSON</td>\n",
       "      <td>12.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1514</th>\n",
       "      <td>MARKDOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>FIRST</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>599.9</td>\n",
       "      <td>1602.3</td>\n",
       "      <td>2819.4</td>\n",
       "      <td>137.2</td>\n",
       "      <td>...</td>\n",
       "      <td>1612.7</td>\n",
       "      <td>555.2</td>\n",
       "      <td>1627.3</td>\n",
       "      <td>555.2</td>\n",
       "      <td>1627.3</td>\n",
       "      <td>5</td>\n",
       "      <td>MARKDOLSON</td>\n",
       "      <td>12.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4250</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>ISSUE</td>\n",
       "      <td>458.0</td>\n",
       "      <td>1676.4</td>\n",
       "      <td>2783.6</td>\n",
       "      <td>188.7</td>\n",
       "      <td>...</td>\n",
       "      <td>1624.9</td>\n",
       "      <td>374.2</td>\n",
       "      <td>1398.9</td>\n",
       "      <td>374.2</td>\n",
       "      <td>1398.9</td>\n",
       "      <td>5</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4251</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>ISSUE</td>\n",
       "      <td>546.0</td>\n",
       "      <td>1492.4</td>\n",
       "      <td>2653.4</td>\n",
       "      <td>142.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1496.2</td>\n",
       "      <td>543.7</td>\n",
       "      <td>1498.1</td>\n",
       "      <td>543.7</td>\n",
       "      <td>1498.1</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4252</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>ISSUE</td>\n",
       "      <td>469.7</td>\n",
       "      <td>1364.1</td>\n",
       "      <td>2659.1</td>\n",
       "      <td>116.9</td>\n",
       "      <td>...</td>\n",
       "      <td>1365.4</td>\n",
       "      <td>426.2</td>\n",
       "      <td>1365.8</td>\n",
       "      <td>426.2</td>\n",
       "      <td>1365.8</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4321</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>ISSUE</td>\n",
       "      <td>417.6</td>\n",
       "      <td>1372.1</td>\n",
       "      <td>2422.7</td>\n",
       "      <td>164.9</td>\n",
       "      <td>...</td>\n",
       "      <td>1319.0</td>\n",
       "      <td>386.5</td>\n",
       "      <td>1281.8</td>\n",
       "      <td>386.5</td>\n",
       "      <td>1281.8</td>\n",
       "      <td>5</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4322</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>ISSUE</td>\n",
       "      <td>516.9</td>\n",
       "      <td>1463.1</td>\n",
       "      <td>1805.8</td>\n",
       "      <td>196.2</td>\n",
       "      <td>...</td>\n",
       "      <td>1476.4</td>\n",
       "      <td>529.2</td>\n",
       "      <td>1475.6</td>\n",
       "      <td>529.2</td>\n",
       "      <td>1475.6</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4323</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>ISSUE</td>\n",
       "      <td>467.3</td>\n",
       "      <td>1502.8</td>\n",
       "      <td>1783.8</td>\n",
       "      <td>138.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1508.7</td>\n",
       "      <td>442.7</td>\n",
       "      <td>1505.5</td>\n",
       "      <td>442.7</td>\n",
       "      <td>1505.5</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4353</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>429.3</td>\n",
       "      <td>1469.0</td>\n",
       "      <td>2503.6</td>\n",
       "      <td>231.2</td>\n",
       "      <td>...</td>\n",
       "      <td>1489.5</td>\n",
       "      <td>308.8</td>\n",
       "      <td>1544.6</td>\n",
       "      <td>307.7</td>\n",
       "      <td>1537.6</td>\n",
       "      <td>5</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4354</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>583.1</td>\n",
       "      <td>1534.3</td>\n",
       "      <td>2611.3</td>\n",
       "      <td>255.7</td>\n",
       "      <td>...</td>\n",
       "      <td>1534.8</td>\n",
       "      <td>583.3</td>\n",
       "      <td>1532.5</td>\n",
       "      <td>583.3</td>\n",
       "      <td>1532.5</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4355</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>519.2</td>\n",
       "      <td>1540.1</td>\n",
       "      <td>2605.3</td>\n",
       "      <td>244.5</td>\n",
       "      <td>...</td>\n",
       "      <td>1541.4</td>\n",
       "      <td>509.9</td>\n",
       "      <td>1537.7</td>\n",
       "      <td>509.9</td>\n",
       "      <td>1537.7</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4756</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>SIX</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>604.3</td>\n",
       "      <td>1430.9</td>\n",
       "      <td>2554.9</td>\n",
       "      <td>490.6</td>\n",
       "      <td>...</td>\n",
       "      <td>1274.0</td>\n",
       "      <td>624.8</td>\n",
       "      <td>1161.7</td>\n",
       "      <td>615.8</td>\n",
       "      <td>1160.1</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4757</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>SIX</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>547.5</td>\n",
       "      <td>1764.1</td>\n",
       "      <td>2632.0</td>\n",
       "      <td>202.3</td>\n",
       "      <td>...</td>\n",
       "      <td>1737.9</td>\n",
       "      <td>543.1</td>\n",
       "      <td>1681.8</td>\n",
       "      <td>539.2</td>\n",
       "      <td>1670.6</td>\n",
       "      <td>5</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4758</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>SIX</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>507.3</td>\n",
       "      <td>1516.7</td>\n",
       "      <td>2578.4</td>\n",
       "      <td>242.9</td>\n",
       "      <td>...</td>\n",
       "      <td>1523.6</td>\n",
       "      <td>482.9</td>\n",
       "      <td>1518.2</td>\n",
       "      <td>482.9</td>\n",
       "      <td>1518.2</td>\n",
       "      <td>5</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4810</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>THE</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>402.6</td>\n",
       "      <td>1566.4</td>\n",
       "      <td>2333.9</td>\n",
       "      <td>103.5</td>\n",
       "      <td>...</td>\n",
       "      <td>1466.9</td>\n",
       "      <td>363.4</td>\n",
       "      <td>1400.6</td>\n",
       "      <td>363.4</td>\n",
       "      <td>1400.6</td>\n",
       "      <td>5</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4811</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>THE</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>525.2</td>\n",
       "      <td>1651.0</td>\n",
       "      <td>1765.3</td>\n",
       "      <td>182.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1643.7</td>\n",
       "      <td>528.8</td>\n",
       "      <td>1634.2</td>\n",
       "      <td>528.8</td>\n",
       "      <td>1634.2</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4812</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>THE</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>307.9</td>\n",
       "      <td>1224.6</td>\n",
       "      <td>1971.6</td>\n",
       "      <td>235.7</td>\n",
       "      <td>...</td>\n",
       "      <td>1243.7</td>\n",
       "      <td>312.3</td>\n",
       "      <td>1264.3</td>\n",
       "      <td>312.3</td>\n",
       "      <td>1264.3</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5645</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>410.0</td>\n",
       "      <td>1432.8</td>\n",
       "      <td>2563.4</td>\n",
       "      <td>161.5</td>\n",
       "      <td>...</td>\n",
       "      <td>1427.1</td>\n",
       "      <td>371.5</td>\n",
       "      <td>1400.3</td>\n",
       "      <td>371.5</td>\n",
       "      <td>1400.3</td>\n",
       "      <td>5</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5646</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>565.1</td>\n",
       "      <td>1473.8</td>\n",
       "      <td>2062.9</td>\n",
       "      <td>120.2</td>\n",
       "      <td>...</td>\n",
       "      <td>1471.8</td>\n",
       "      <td>548.4</td>\n",
       "      <td>1476.1</td>\n",
       "      <td>545.0</td>\n",
       "      <td>1475.2</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5647</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>522.3</td>\n",
       "      <td>1417.0</td>\n",
       "      <td>2412.4</td>\n",
       "      <td>263.7</td>\n",
       "      <td>...</td>\n",
       "      <td>1439.5</td>\n",
       "      <td>537.2</td>\n",
       "      <td>1456.4</td>\n",
       "      <td>537.2</td>\n",
       "      <td>1456.4</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5672</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>QUESTIONS</td>\n",
       "      <td>409.3</td>\n",
       "      <td>1342.7</td>\n",
       "      <td>2799.0</td>\n",
       "      <td>101.5</td>\n",
       "      <td>...</td>\n",
       "      <td>1495.1</td>\n",
       "      <td>365.9</td>\n",
       "      <td>1654.7</td>\n",
       "      <td>360.8</td>\n",
       "      <td>1664.6</td>\n",
       "      <td>5</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5673</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>QUESTIONS</td>\n",
       "      <td>521.7</td>\n",
       "      <td>1949.7</td>\n",
       "      <td>2829.1</td>\n",
       "      <td>172.8</td>\n",
       "      <td>...</td>\n",
       "      <td>1936.7</td>\n",
       "      <td>519.4</td>\n",
       "      <td>1775.6</td>\n",
       "      <td>519.4</td>\n",
       "      <td>1775.6</td>\n",
       "      <td>5</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5674</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>QUESTIONS</td>\n",
       "      <td>491.8</td>\n",
       "      <td>1887.4</td>\n",
       "      <td>2617.9</td>\n",
       "      <td>116.4</td>\n",
       "      <td>...</td>\n",
       "      <td>1840.8</td>\n",
       "      <td>476.0</td>\n",
       "      <td>1798.7</td>\n",
       "      <td>476.0</td>\n",
       "      <td>1798.7</td>\n",
       "      <td>5</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5852</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>ISSUES</td>\n",
       "      <td>377.0</td>\n",
       "      <td>1340.7</td>\n",
       "      <td>2403.0</td>\n",
       "      <td>167.6</td>\n",
       "      <td>...</td>\n",
       "      <td>1325.4</td>\n",
       "      <td>341.9</td>\n",
       "      <td>1310.4</td>\n",
       "      <td>341.9</td>\n",
       "      <td>1310.4</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5853</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>ISSUES</td>\n",
       "      <td>525.6</td>\n",
       "      <td>1607.2</td>\n",
       "      <td>2023.5</td>\n",
       "      <td>290.5</td>\n",
       "      <td>...</td>\n",
       "      <td>1604.4</td>\n",
       "      <td>528.9</td>\n",
       "      <td>1576.6</td>\n",
       "      <td>528.9</td>\n",
       "      <td>1576.6</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5854</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>ISSUES</td>\n",
       "      <td>445.3</td>\n",
       "      <td>1479.1</td>\n",
       "      <td>1940.3</td>\n",
       "      <td>124.3</td>\n",
       "      <td>...</td>\n",
       "      <td>1496.6</td>\n",
       "      <td>450.4</td>\n",
       "      <td>1511.9</td>\n",
       "      <td>450.4</td>\n",
       "      <td>1511.9</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6026</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>DECISIONS</td>\n",
       "      <td>414.8</td>\n",
       "      <td>1367.4</td>\n",
       "      <td>2472.4</td>\n",
       "      <td>237.2</td>\n",
       "      <td>...</td>\n",
       "      <td>1331.1</td>\n",
       "      <td>327.1</td>\n",
       "      <td>1300.1</td>\n",
       "      <td>327.1</td>\n",
       "      <td>1300.1</td>\n",
       "      <td>5</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6027</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>DECISIONS</td>\n",
       "      <td>509.7</td>\n",
       "      <td>1532.1</td>\n",
       "      <td>1952.7</td>\n",
       "      <td>333.2</td>\n",
       "      <td>...</td>\n",
       "      <td>1531.9</td>\n",
       "      <td>506.5</td>\n",
       "      <td>1530.0</td>\n",
       "      <td>506.5</td>\n",
       "      <td>1530.0</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6028</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>DECISIONS</td>\n",
       "      <td>423.9</td>\n",
       "      <td>1392.8</td>\n",
       "      <td>2336.9</td>\n",
       "      <td>212.9</td>\n",
       "      <td>...</td>\n",
       "      <td>1393.4</td>\n",
       "      <td>401.9</td>\n",
       "      <td>1398.9</td>\n",
       "      <td>401.9</td>\n",
       "      <td>1398.9</td>\n",
       "      <td>6</td>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3107718</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>RIGHT</td>\n",
       "      <td>491.5</td>\n",
       "      <td>1587.0</td>\n",
       "      <td>2649.8</td>\n",
       "      <td>224.1</td>\n",
       "      <td>...</td>\n",
       "      <td>1576.6</td>\n",
       "      <td>425.6</td>\n",
       "      <td>1663.5</td>\n",
       "      <td>425.6</td>\n",
       "      <td>1663.5</td>\n",
       "      <td>4</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3107719</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>RIGHT</td>\n",
       "      <td>635.2</td>\n",
       "      <td>1658.8</td>\n",
       "      <td>2684.5</td>\n",
       "      <td>96.1</td>\n",
       "      <td>...</td>\n",
       "      <td>1662.8</td>\n",
       "      <td>580.3</td>\n",
       "      <td>1636.5</td>\n",
       "      <td>571.7</td>\n",
       "      <td>1628.2</td>\n",
       "      <td>6</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3107720</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>RIGHT</td>\n",
       "      <td>551.1</td>\n",
       "      <td>1668.6</td>\n",
       "      <td>2444.8</td>\n",
       "      <td>101.1</td>\n",
       "      <td>...</td>\n",
       "      <td>1642.1</td>\n",
       "      <td>533.1</td>\n",
       "      <td>1593.2</td>\n",
       "      <td>533.1</td>\n",
       "      <td>1593.2</td>\n",
       "      <td>6</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3107950</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>TENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>CASES</td>\n",
       "      <td>404.9</td>\n",
       "      <td>1610.7</td>\n",
       "      <td>2639.6</td>\n",
       "      <td>133.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1604.0</td>\n",
       "      <td>378.9</td>\n",
       "      <td>1598.6</td>\n",
       "      <td>378.9</td>\n",
       "      <td>1598.6</td>\n",
       "      <td>5</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3107951</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>TENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>CASES</td>\n",
       "      <td>580.8</td>\n",
       "      <td>1700.4</td>\n",
       "      <td>2622.7</td>\n",
       "      <td>71.4</td>\n",
       "      <td>...</td>\n",
       "      <td>1726.8</td>\n",
       "      <td>562.4</td>\n",
       "      <td>1747.2</td>\n",
       "      <td>558.1</td>\n",
       "      <td>1730.5</td>\n",
       "      <td>6</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3107952</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>TENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>CASES</td>\n",
       "      <td>495.8</td>\n",
       "      <td>1947.5</td>\n",
       "      <td>2616.9</td>\n",
       "      <td>66.5</td>\n",
       "      <td>...</td>\n",
       "      <td>2017.7</td>\n",
       "      <td>491.7</td>\n",
       "      <td>2059.4</td>\n",
       "      <td>491.7</td>\n",
       "      <td>2059.4</td>\n",
       "      <td>6</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3108019</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>457.0</td>\n",
       "      <td>1358.3</td>\n",
       "      <td>2559.2</td>\n",
       "      <td>118.2</td>\n",
       "      <td>...</td>\n",
       "      <td>1348.5</td>\n",
       "      <td>429.5</td>\n",
       "      <td>1344.5</td>\n",
       "      <td>429.5</td>\n",
       "      <td>1344.5</td>\n",
       "      <td>5</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3108020</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>555.0</td>\n",
       "      <td>1778.3</td>\n",
       "      <td>2121.4</td>\n",
       "      <td>100.6</td>\n",
       "      <td>...</td>\n",
       "      <td>1772.1</td>\n",
       "      <td>549.2</td>\n",
       "      <td>1752.7</td>\n",
       "      <td>545.6</td>\n",
       "      <td>1744.7</td>\n",
       "      <td>6</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3108021</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>567.4</td>\n",
       "      <td>1745.0</td>\n",
       "      <td>2685.9</td>\n",
       "      <td>118.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1747.4</td>\n",
       "      <td>563.2</td>\n",
       "      <td>1747.0</td>\n",
       "      <td>563.2</td>\n",
       "      <td>1747.0</td>\n",
       "      <td>6</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3108989</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>FIRST</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>459.8</td>\n",
       "      <td>1473.1</td>\n",
       "      <td>2631.1</td>\n",
       "      <td>80.6</td>\n",
       "      <td>...</td>\n",
       "      <td>1433.7</td>\n",
       "      <td>392.5</td>\n",
       "      <td>1402.4</td>\n",
       "      <td>392.5</td>\n",
       "      <td>1402.4</td>\n",
       "      <td>5</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3108990</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>FIRST</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>586.0</td>\n",
       "      <td>1748.2</td>\n",
       "      <td>2670.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1743.0</td>\n",
       "      <td>573.4</td>\n",
       "      <td>1755.4</td>\n",
       "      <td>557.8</td>\n",
       "      <td>1756.2</td>\n",
       "      <td>5</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3108991</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>FIRST</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>544.3</td>\n",
       "      <td>1716.3</td>\n",
       "      <td>2691.4</td>\n",
       "      <td>56.1</td>\n",
       "      <td>...</td>\n",
       "      <td>1735.5</td>\n",
       "      <td>535.7</td>\n",
       "      <td>1726.8</td>\n",
       "      <td>490.2</td>\n",
       "      <td>1719.2</td>\n",
       "      <td>5</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3109819</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>TENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>REGULATION</td>\n",
       "      <td>422.7</td>\n",
       "      <td>1537.4</td>\n",
       "      <td>2569.8</td>\n",
       "      <td>135.1</td>\n",
       "      <td>...</td>\n",
       "      <td>1543.3</td>\n",
       "      <td>391.7</td>\n",
       "      <td>1550.9</td>\n",
       "      <td>391.7</td>\n",
       "      <td>1550.9</td>\n",
       "      <td>5</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3109820</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>TENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>REGULATION</td>\n",
       "      <td>564.1</td>\n",
       "      <td>1630.3</td>\n",
       "      <td>1940.7</td>\n",
       "      <td>69.4</td>\n",
       "      <td>...</td>\n",
       "      <td>1632.2</td>\n",
       "      <td>548.9</td>\n",
       "      <td>1636.4</td>\n",
       "      <td>548.9</td>\n",
       "      <td>1636.4</td>\n",
       "      <td>6</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3109821</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>TENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>REGULATION</td>\n",
       "      <td>500.4</td>\n",
       "      <td>1633.7</td>\n",
       "      <td>2130.9</td>\n",
       "      <td>100.8</td>\n",
       "      <td>...</td>\n",
       "      <td>1628.6</td>\n",
       "      <td>488.3</td>\n",
       "      <td>1619.7</td>\n",
       "      <td>488.3</td>\n",
       "      <td>1619.7</td>\n",
       "      <td>6</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3109918</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>WHICH</td>\n",
       "      <td>545.2</td>\n",
       "      <td>1446.6</td>\n",
       "      <td>2665.1</td>\n",
       "      <td>59.7</td>\n",
       "      <td>...</td>\n",
       "      <td>1421.6</td>\n",
       "      <td>557.1</td>\n",
       "      <td>1383.3</td>\n",
       "      <td>557.1</td>\n",
       "      <td>1383.3</td>\n",
       "      <td>5</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3109919</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>WHICH</td>\n",
       "      <td>314.1</td>\n",
       "      <td>1549.0</td>\n",
       "      <td>2344.0</td>\n",
       "      <td>114.4</td>\n",
       "      <td>...</td>\n",
       "      <td>1548.6</td>\n",
       "      <td>307.9</td>\n",
       "      <td>1547.4</td>\n",
       "      <td>307.9</td>\n",
       "      <td>1547.4</td>\n",
       "      <td>5</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3109920</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>ELEVENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>WHICH</td>\n",
       "      <td>591.4</td>\n",
       "      <td>1935.2</td>\n",
       "      <td>2685.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1933.8</td>\n",
       "      <td>608.4</td>\n",
       "      <td>1924.8</td>\n",
       "      <td>608.4</td>\n",
       "      <td>1924.8</td>\n",
       "      <td>5</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3110751</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>THE</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>IN</td>\n",
       "      <td>417.0</td>\n",
       "      <td>2074.6</td>\n",
       "      <td>2816.0</td>\n",
       "      <td>69.5</td>\n",
       "      <td>...</td>\n",
       "      <td>2015.9</td>\n",
       "      <td>390.1</td>\n",
       "      <td>1942.1</td>\n",
       "      <td>390.1</td>\n",
       "      <td>1942.1</td>\n",
       "      <td>5</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3110752</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>THE</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>IN</td>\n",
       "      <td>428.6</td>\n",
       "      <td>1967.1</td>\n",
       "      <td>2515.7</td>\n",
       "      <td>73.5</td>\n",
       "      <td>...</td>\n",
       "      <td>1982.4</td>\n",
       "      <td>405.6</td>\n",
       "      <td>1985.7</td>\n",
       "      <td>401.6</td>\n",
       "      <td>1982.2</td>\n",
       "      <td>6</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3110753</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>THE</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>IN</td>\n",
       "      <td>399.0</td>\n",
       "      <td>2029.5</td>\n",
       "      <td>2943.7</td>\n",
       "      <td>88.7</td>\n",
       "      <td>...</td>\n",
       "      <td>2020.7</td>\n",
       "      <td>398.2</td>\n",
       "      <td>2005.1</td>\n",
       "      <td>398.2</td>\n",
       "      <td>2005.1</td>\n",
       "      <td>4</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3116417</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>FOURTEENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>453.2</td>\n",
       "      <td>1580.1</td>\n",
       "      <td>2526.9</td>\n",
       "      <td>106.8</td>\n",
       "      <td>...</td>\n",
       "      <td>1566.9</td>\n",
       "      <td>422.5</td>\n",
       "      <td>1549.5</td>\n",
       "      <td>422.5</td>\n",
       "      <td>1549.5</td>\n",
       "      <td>5</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3116418</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>FOURTEENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>588.6</td>\n",
       "      <td>1596.9</td>\n",
       "      <td>1944.2</td>\n",
       "      <td>77.5</td>\n",
       "      <td>...</td>\n",
       "      <td>1610.2</td>\n",
       "      <td>530.6</td>\n",
       "      <td>1613.1</td>\n",
       "      <td>518.5</td>\n",
       "      <td>1607.9</td>\n",
       "      <td>6</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3116419</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>FOURTEENTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>480.1</td>\n",
       "      <td>1732.8</td>\n",
       "      <td>2648.5</td>\n",
       "      <td>84.1</td>\n",
       "      <td>...</td>\n",
       "      <td>1737.6</td>\n",
       "      <td>477.5</td>\n",
       "      <td>1740.2</td>\n",
       "      <td>477.5</td>\n",
       "      <td>1740.2</td>\n",
       "      <td>5</td>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3117429</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>FOURTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>VIOLATIONS</td>\n",
       "      <td>492.3</td>\n",
       "      <td>1407.8</td>\n",
       "      <td>2647.5</td>\n",
       "      <td>412.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1393.0</td>\n",
       "      <td>436.1</td>\n",
       "      <td>1382.4</td>\n",
       "      <td>436.1</td>\n",
       "      <td>1382.4</td>\n",
       "      <td>5</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3117430</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>FOURTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>VIOLATIONS</td>\n",
       "      <td>706.7</td>\n",
       "      <td>1566.0</td>\n",
       "      <td>2641.8</td>\n",
       "      <td>255.6</td>\n",
       "      <td>...</td>\n",
       "      <td>1560.0</td>\n",
       "      <td>708.4</td>\n",
       "      <td>1543.9</td>\n",
       "      <td>708.4</td>\n",
       "      <td>1543.9</td>\n",
       "      <td>5</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3117431</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>FOURTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>VIOLATIONS</td>\n",
       "      <td>534.9</td>\n",
       "      <td>1437.2</td>\n",
       "      <td>1813.2</td>\n",
       "      <td>201.8</td>\n",
       "      <td>...</td>\n",
       "      <td>1439.0</td>\n",
       "      <td>511.1</td>\n",
       "      <td>1440.3</td>\n",
       "      <td>511.1</td>\n",
       "      <td>1440.3</td>\n",
       "      <td>6</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3117488</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>FOURTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>429.5</td>\n",
       "      <td>1464.8</td>\n",
       "      <td>1872.1</td>\n",
       "      <td>247.7</td>\n",
       "      <td>...</td>\n",
       "      <td>1410.9</td>\n",
       "      <td>435.0</td>\n",
       "      <td>1360.2</td>\n",
       "      <td>435.0</td>\n",
       "      <td>1360.2</td>\n",
       "      <td>6</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3117489</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>EH</td>\n",
       "      <td>1</td>\n",
       "      <td>FOURTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>675.7</td>\n",
       "      <td>1607.1</td>\n",
       "      <td>2805.4</td>\n",
       "      <td>248.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1588.3</td>\n",
       "      <td>627.1</td>\n",
       "      <td>1558.1</td>\n",
       "      <td>627.1</td>\n",
       "      <td>1558.1</td>\n",
       "      <td>5</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3117490</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AH</td>\n",
       "      <td>0</td>\n",
       "      <td>FOURTH</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>SP</td>\n",
       "      <td>460.0</td>\n",
       "      <td>1341.9</td>\n",
       "      <td>2634.2</td>\n",
       "      <td>244.4</td>\n",
       "      <td>...</td>\n",
       "      <td>1523.0</td>\n",
       "      <td>605.3</td>\n",
       "      <td>1868.6</td>\n",
       "      <td>393.1</td>\n",
       "      <td>1416.7</td>\n",
       "      <td>5</td>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>24.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3551 rows × 46 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                    name vowel  stress    pre_word       word    fol_word  \\\n",
       "1512          MARKDOLSON    AH       0       FIRST  AMENDMENT          SP   \n",
       "1513          MARKDOLSON    EH       1       FIRST  AMENDMENT          SP   \n",
       "1514          MARKDOLSON    AH       0       FIRST  AMENDMENT          SP   \n",
       "4250      THEODOREBOLSON    AH       0    ELEVENTH  AMENDMENT       ISSUE   \n",
       "4251      THEODOREBOLSON    EH       1    ELEVENTH  AMENDMENT       ISSUE   \n",
       "4252      THEODOREBOLSON    AH       0    ELEVENTH  AMENDMENT       ISSUE   \n",
       "4321      THEODOREBOLSON    AH       0    ELEVENTH  AMENDMENT       ISSUE   \n",
       "4322      THEODOREBOLSON    EH       1    ELEVENTH  AMENDMENT       ISSUE   \n",
       "4323      THEODOREBOLSON    AH       0    ELEVENTH  AMENDMENT       ISSUE   \n",
       "4353      THEODOREBOLSON    AH       0    ELEVENTH  AMENDMENT          SP   \n",
       "4354      THEODOREBOLSON    EH       1    ELEVENTH  AMENDMENT          SP   \n",
       "4355      THEODOREBOLSON    AH       0    ELEVENTH  AMENDMENT          SP   \n",
       "4756      THEODOREBOLSON    AH       0         SIX  AMENDMENT          SP   \n",
       "4757      THEODOREBOLSON    EH       1         SIX  AMENDMENT          SP   \n",
       "4758      THEODOREBOLSON    AH       0         SIX  AMENDMENT          SP   \n",
       "4810      THEODOREBOLSON    AH       0         THE  AMENDMENT          SP   \n",
       "4811      THEODOREBOLSON    EH       1         THE  AMENDMENT          SP   \n",
       "4812      THEODOREBOLSON    AH       0         THE  AMENDMENT          SP   \n",
       "5645      THEODOREBOLSON    AH       0    ELEVENTH  AMENDMENT          SP   \n",
       "5646      THEODOREBOLSON    EH       1    ELEVENTH  AMENDMENT          SP   \n",
       "5647      THEODOREBOLSON    AH       0    ELEVENTH  AMENDMENT          SP   \n",
       "5672      THEODOREBOLSON    AH       0    ELEVENTH  AMENDMENT   QUESTIONS   \n",
       "5673      THEODOREBOLSON    EH       1    ELEVENTH  AMENDMENT   QUESTIONS   \n",
       "5674      THEODOREBOLSON    AH       0    ELEVENTH  AMENDMENT   QUESTIONS   \n",
       "5852      THEODOREBOLSON    AH       0    ELEVENTH  AMENDMENT      ISSUES   \n",
       "5853      THEODOREBOLSON    EH       1    ELEVENTH  AMENDMENT      ISSUES   \n",
       "5854      THEODOREBOLSON    AH       0    ELEVENTH  AMENDMENT      ISSUES   \n",
       "6026      THEODOREBOLSON    AH       0    ELEVENTH  AMENDMENT   DECISIONS   \n",
       "6027      THEODOREBOLSON    EH       1    ELEVENTH  AMENDMENT   DECISIONS   \n",
       "6028      THEODOREBOLSON    AH       0    ELEVENTH  AMENDMENT   DECISIONS   \n",
       "...                  ...   ...     ...         ...        ...         ...   \n",
       "3107718      SETHPWAXMAN    AH       0    ELEVENTH  AMENDMENT       RIGHT   \n",
       "3107719      SETHPWAXMAN    EH       1    ELEVENTH  AMENDMENT       RIGHT   \n",
       "3107720      SETHPWAXMAN    AH       0    ELEVENTH  AMENDMENT       RIGHT   \n",
       "3107950      SETHPWAXMAN    AH       0       TENTH  AMENDMENT       CASES   \n",
       "3107951      SETHPWAXMAN    EH       1       TENTH  AMENDMENT       CASES   \n",
       "3107952      SETHPWAXMAN    AH       0       TENTH  AMENDMENT       CASES   \n",
       "3108019      SETHPWAXMAN    AH       0    ELEVENTH  AMENDMENT          SP   \n",
       "3108020      SETHPWAXMAN    EH       1    ELEVENTH  AMENDMENT          SP   \n",
       "3108021      SETHPWAXMAN    AH       0    ELEVENTH  AMENDMENT          SP   \n",
       "3108989      SETHPWAXMAN    AH       0       FIRST  AMENDMENT          SP   \n",
       "3108990      SETHPWAXMAN    EH       1       FIRST  AMENDMENT          SP   \n",
       "3108991      SETHPWAXMAN    AH       0       FIRST  AMENDMENT          SP   \n",
       "3109819      SETHPWAXMAN    AH       0       TENTH  AMENDMENT  REGULATION   \n",
       "3109820      SETHPWAXMAN    EH       1       TENTH  AMENDMENT  REGULATION   \n",
       "3109821      SETHPWAXMAN    AH       0       TENTH  AMENDMENT  REGULATION   \n",
       "3109918      SETHPWAXMAN    AH       0    ELEVENTH  AMENDMENT       WHICH   \n",
       "3109919      SETHPWAXMAN    EH       1    ELEVENTH  AMENDMENT       WHICH   \n",
       "3109920      SETHPWAXMAN    AH       0    ELEVENTH  AMENDMENT       WHICH   \n",
       "3110751  JOHNPAULSTEVENS    AH       0         THE  AMENDMENT          IN   \n",
       "3110752  JOHNPAULSTEVENS    EH       1         THE  AMENDMENT          IN   \n",
       "3110753  JOHNPAULSTEVENS    AH       0         THE  AMENDMENT          IN   \n",
       "3116417      SETHPWAXMAN    AH       0  FOURTEENTH  AMENDMENT          SP   \n",
       "3116418      SETHPWAXMAN    EH       1  FOURTEENTH  AMENDMENT          SP   \n",
       "3116419      SETHPWAXMAN    AH       0  FOURTEENTH  AMENDMENT          SP   \n",
       "3117429  JOHNPAULSTEVENS    AH       0      FOURTH  AMENDMENT  VIOLATIONS   \n",
       "3117430  JOHNPAULSTEVENS    EH       1      FOURTH  AMENDMENT  VIOLATIONS   \n",
       "3117431  JOHNPAULSTEVENS    AH       0      FOURTH  AMENDMENT  VIOLATIONS   \n",
       "3117488  JOHNPAULSTEVENS    AH       0      FOURTH  AMENDMENT          SP   \n",
       "3117489  JOHNPAULSTEVENS    EH       1      FOURTH  AMENDMENT          SP   \n",
       "3117490  JOHNPAULSTEVENS    AH       0      FOURTH  AMENDMENT          SP   \n",
       "\n",
       "            F1      F2      F3     B1  ...   F2@50%  F1@65%  F2@65%  F1@80%  \\\n",
       "1512     437.8  1627.8  2460.9  190.2  ...   1618.9   418.0  1604.0   418.0   \n",
       "1513     569.1  1702.9  2964.0  189.9  ...   1687.1   593.5  1683.4   589.7   \n",
       "1514     599.9  1602.3  2819.4  137.2  ...   1612.7   555.2  1627.3   555.2   \n",
       "4250     458.0  1676.4  2783.6  188.7  ...   1624.9   374.2  1398.9   374.2   \n",
       "4251     546.0  1492.4  2653.4  142.0  ...   1496.2   543.7  1498.1   543.7   \n",
       "4252     469.7  1364.1  2659.1  116.9  ...   1365.4   426.2  1365.8   426.2   \n",
       "4321     417.6  1372.1  2422.7  164.9  ...   1319.0   386.5  1281.8   386.5   \n",
       "4322     516.9  1463.1  1805.8  196.2  ...   1476.4   529.2  1475.6   529.2   \n",
       "4323     467.3  1502.8  1783.8  138.0  ...   1508.7   442.7  1505.5   442.7   \n",
       "4353     429.3  1469.0  2503.6  231.2  ...   1489.5   308.8  1544.6   307.7   \n",
       "4354     583.1  1534.3  2611.3  255.7  ...   1534.8   583.3  1532.5   583.3   \n",
       "4355     519.2  1540.1  2605.3  244.5  ...   1541.4   509.9  1537.7   509.9   \n",
       "4756     604.3  1430.9  2554.9  490.6  ...   1274.0   624.8  1161.7   615.8   \n",
       "4757     547.5  1764.1  2632.0  202.3  ...   1737.9   543.1  1681.8   539.2   \n",
       "4758     507.3  1516.7  2578.4  242.9  ...   1523.6   482.9  1518.2   482.9   \n",
       "4810     402.6  1566.4  2333.9  103.5  ...   1466.9   363.4  1400.6   363.4   \n",
       "4811     525.2  1651.0  1765.3  182.0  ...   1643.7   528.8  1634.2   528.8   \n",
       "4812     307.9  1224.6  1971.6  235.7  ...   1243.7   312.3  1264.3   312.3   \n",
       "5645     410.0  1432.8  2563.4  161.5  ...   1427.1   371.5  1400.3   371.5   \n",
       "5646     565.1  1473.8  2062.9  120.2  ...   1471.8   548.4  1476.1   545.0   \n",
       "5647     522.3  1417.0  2412.4  263.7  ...   1439.5   537.2  1456.4   537.2   \n",
       "5672     409.3  1342.7  2799.0  101.5  ...   1495.1   365.9  1654.7   360.8   \n",
       "5673     521.7  1949.7  2829.1  172.8  ...   1936.7   519.4  1775.6   519.4   \n",
       "5674     491.8  1887.4  2617.9  116.4  ...   1840.8   476.0  1798.7   476.0   \n",
       "5852     377.0  1340.7  2403.0  167.6  ...   1325.4   341.9  1310.4   341.9   \n",
       "5853     525.6  1607.2  2023.5  290.5  ...   1604.4   528.9  1576.6   528.9   \n",
       "5854     445.3  1479.1  1940.3  124.3  ...   1496.6   450.4  1511.9   450.4   \n",
       "6026     414.8  1367.4  2472.4  237.2  ...   1331.1   327.1  1300.1   327.1   \n",
       "6027     509.7  1532.1  1952.7  333.2  ...   1531.9   506.5  1530.0   506.5   \n",
       "6028     423.9  1392.8  2336.9  212.9  ...   1393.4   401.9  1398.9   401.9   \n",
       "...        ...     ...     ...    ...  ...      ...     ...     ...     ...   \n",
       "3107718  491.5  1587.0  2649.8  224.1  ...   1576.6   425.6  1663.5   425.6   \n",
       "3107719  635.2  1658.8  2684.5   96.1  ...   1662.8   580.3  1636.5   571.7   \n",
       "3107720  551.1  1668.6  2444.8  101.1  ...   1642.1   533.1  1593.2   533.1   \n",
       "3107950  404.9  1610.7  2639.6  133.0  ...   1604.0   378.9  1598.6   378.9   \n",
       "3107951  580.8  1700.4  2622.7   71.4  ...   1726.8   562.4  1747.2   558.1   \n",
       "3107952  495.8  1947.5  2616.9   66.5  ...   2017.7   491.7  2059.4   491.7   \n",
       "3108019  457.0  1358.3  2559.2  118.2  ...   1348.5   429.5  1344.5   429.5   \n",
       "3108020  555.0  1778.3  2121.4  100.6  ...   1772.1   549.2  1752.7   545.6   \n",
       "3108021  567.4  1745.0  2685.9  118.0  ...   1747.4   563.2  1747.0   563.2   \n",
       "3108989  459.8  1473.1  2631.1   80.6  ...   1433.7   392.5  1402.4   392.5   \n",
       "3108990  586.0  1748.2  2670.0   89.0  ...   1743.0   573.4  1755.4   557.8   \n",
       "3108991  544.3  1716.3  2691.4   56.1  ...   1735.5   535.7  1726.8   490.2   \n",
       "3109819  422.7  1537.4  2569.8  135.1  ...   1543.3   391.7  1550.9   391.7   \n",
       "3109820  564.1  1630.3  1940.7   69.4  ...   1632.2   548.9  1636.4   548.9   \n",
       "3109821  500.4  1633.7  2130.9  100.8  ...   1628.6   488.3  1619.7   488.3   \n",
       "3109918  545.2  1446.6  2665.1   59.7  ...   1421.6   557.1  1383.3   557.1   \n",
       "3109919  314.1  1549.0  2344.0  114.4  ...   1548.6   307.9  1547.4   307.9   \n",
       "3109920  591.4  1935.2  2685.0   93.0  ...   1933.8   608.4  1924.8   608.4   \n",
       "3110751  417.0  2074.6  2816.0   69.5  ...   2015.9   390.1  1942.1   390.1   \n",
       "3110752  428.6  1967.1  2515.7   73.5  ...   1982.4   405.6  1985.7   401.6   \n",
       "3110753  399.0  2029.5  2943.7   88.7  ...   2020.7   398.2  2005.1   398.2   \n",
       "3116417  453.2  1580.1  2526.9  106.8  ...   1566.9   422.5  1549.5   422.5   \n",
       "3116418  588.6  1596.9  1944.2   77.5  ...   1610.2   530.6  1613.1   518.5   \n",
       "3116419  480.1  1732.8  2648.5   84.1  ...   1737.6   477.5  1740.2   477.5   \n",
       "3117429  492.3  1407.8  2647.5  412.0  ...   1393.0   436.1  1382.4   436.1   \n",
       "3117430  706.7  1566.0  2641.8  255.6  ...   1560.0   708.4  1543.9   708.4   \n",
       "3117431  534.9  1437.2  1813.2  201.8  ...   1439.0   511.1  1440.3   511.1   \n",
       "3117488  429.5  1464.8  1872.1  247.7  ...   1410.9   435.0  1360.2   435.0   \n",
       "3117489  675.7  1607.1  2805.4  248.0  ...   1588.3   627.1  1558.1   627.1   \n",
       "3117490  460.0  1341.9  2634.2  244.4  ...   1523.0   605.3  1868.6   393.1   \n",
       "\n",
       "         F2@80%  nFormants          Speaker Democratic Republican label  \n",
       "1512     1604.0          6       MARKDOLSON       12.0        3.0     1  \n",
       "1513     1651.1          5       MARKDOLSON       12.0        3.0     1  \n",
       "1514     1627.3          5       MARKDOLSON       12.0        3.0     1  \n",
       "4250     1398.9          5   THEODOREBOLSON        0.0       57.0    -1  \n",
       "4251     1498.1          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "4252     1365.8          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "4321     1281.8          5   THEODOREBOLSON        0.0       57.0    -1  \n",
       "4322     1475.6          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "4323     1505.5          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "4353     1537.6          5   THEODOREBOLSON        0.0       57.0    -1  \n",
       "4354     1532.5          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "4355     1537.7          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "4756     1160.1          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "4757     1670.6          5   THEODOREBOLSON        0.0       57.0    -1  \n",
       "4758     1518.2          5   THEODOREBOLSON        0.0       57.0    -1  \n",
       "4810     1400.6          5   THEODOREBOLSON        0.0       57.0    -1  \n",
       "4811     1634.2          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "4812     1264.3          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "5645     1400.3          5   THEODOREBOLSON        0.0       57.0    -1  \n",
       "5646     1475.2          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "5647     1456.4          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "5672     1664.6          5   THEODOREBOLSON        0.0       57.0    -1  \n",
       "5673     1775.6          5   THEODOREBOLSON        0.0       57.0    -1  \n",
       "5674     1798.7          5   THEODOREBOLSON        0.0       57.0    -1  \n",
       "5852     1310.4          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "5853     1576.6          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "5854     1511.9          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "6026     1300.1          5   THEODOREBOLSON        0.0       57.0    -1  \n",
       "6027     1530.0          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "6028     1398.9          6   THEODOREBOLSON        0.0       57.0    -1  \n",
       "...         ...        ...              ...        ...        ...   ...  \n",
       "3107718  1663.5          4      SETHPWAXMAN       56.0        0.0     1  \n",
       "3107719  1628.2          6      SETHPWAXMAN       56.0        0.0     1  \n",
       "3107720  1593.2          6      SETHPWAXMAN       56.0        0.0     1  \n",
       "3107950  1598.6          5      SETHPWAXMAN       56.0        0.0     1  \n",
       "3107951  1730.5          6      SETHPWAXMAN       56.0        0.0     1  \n",
       "3107952  2059.4          6      SETHPWAXMAN       56.0        0.0     1  \n",
       "3108019  1344.5          5      SETHPWAXMAN       56.0        0.0     1  \n",
       "3108020  1744.7          6      SETHPWAXMAN       56.0        0.0     1  \n",
       "3108021  1747.0          6      SETHPWAXMAN       56.0        0.0     1  \n",
       "3108989  1402.4          5      SETHPWAXMAN       56.0        0.0     1  \n",
       "3108990  1756.2          5      SETHPWAXMAN       56.0        0.0     1  \n",
       "3108991  1719.2          5      SETHPWAXMAN       56.0        0.0     1  \n",
       "3109819  1550.9          5      SETHPWAXMAN       56.0        0.0     1  \n",
       "3109820  1636.4          6      SETHPWAXMAN       56.0        0.0     1  \n",
       "3109821  1619.7          6      SETHPWAXMAN       56.0        0.0     1  \n",
       "3109918  1383.3          5      SETHPWAXMAN       56.0        0.0     1  \n",
       "3109919  1547.4          5      SETHPWAXMAN       56.0        0.0     1  \n",
       "3109920  1924.8          5      SETHPWAXMAN       56.0        0.0     1  \n",
       "3110751  1942.1          5  JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "3110752  1982.2          6  JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "3110753  2005.1          4  JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "3116417  1549.5          5      SETHPWAXMAN       56.0        0.0     1  \n",
       "3116418  1607.9          6      SETHPWAXMAN       56.0        0.0     1  \n",
       "3116419  1740.2          5      SETHPWAXMAN       56.0        0.0     1  \n",
       "3117429  1382.4          5  JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "3117430  1543.9          5  JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "3117431  1440.3          6  JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "3117488  1360.2          6  JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "3117489  1558.1          5  JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "3117490  1416.7          5  JOHNPAULSTEVENS       24.0        3.0     1  \n",
       "\n",
       "[3551 rows x 46 columns]"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[data.word == 'AMENDMENT']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/apple/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  if __name__ == '__main__':\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "1184"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(data[data.word == 'AMENDMENT'][data.vowel=='EH'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "E = list(EH) \n",
    "A = list(AH)\n",
    "len(E)\n",
    "len(A)\n",
    "E == A*2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "word_AMENDMENT = pd.DataFrame({'EH':EH,'AH':AH})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>AH</th>\n",
       "      <th>EH</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Speaker</th>\n",
       "      <th>pre_word</th>\n",
       "      <th>fol_word</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ALANGURA</th>\n",
       "      <th>SP</th>\n",
       "      <th>OR</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DAVIDCFREDERICK</th>\n",
       "      <th>SP</th>\n",
       "      <th>SP</th>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JEFFREYALAMKEN</th>\n",
       "      <th>SP</th>\n",
       "      <th>SP</th>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">KANNONKSHANMUGAM</th>\n",
       "      <th>FOURTH</th>\n",
       "      <th>PURPOSES</th>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SP</th>\n",
       "      <th>SP</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                    AH  EH\n",
       "Speaker          pre_word fol_word        \n",
       "ALANGURA         SP       OR         1   1\n",
       "DAVIDCFREDERICK  SP       SP         5   2\n",
       "JEFFREYALAMKEN   SP       SP         3   2\n",
       "KANNONKSHANMUGAM FOURTH   PURPOSES   5   2\n",
       "                 SP       SP         1   1"
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "word_AMENDMENT[word_AMENDMENT['AH'] != 2*word_AMENDMENT['EH']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "EH.index = EH.index.droplevel(level = 'vowel')\n",
    "AH.index = AH.index.droplevel(level = 'vowel')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/apple/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  if __name__ == '__main__':\n"
     ]
    }
   ],
   "source": [
    "EH = data[data.word == 'AMENDMENT'][data.vowel == 'EH'].groupby(['Speaker','vowel','pre_word','fol_word'])['word'].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/apple/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  if __name__ == '__main__':\n"
     ]
    }
   ],
   "source": [
    "AH = data[data.word == 'AMENDMENT'][data.vowel == 'AH'].groupby(['Speaker','vowel','pre_word','fol_word'])['word'].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Speaker</th>\n",
       "      <th>word</th>\n",
       "      <th>vowel</th>\n",
       "      <th>plt_vclass</th>\n",
       "      <th>plt_manner</th>\n",
       "      <th>plt_place</th>\n",
       "      <th>plt_voice</th>\n",
       "      <th>plt_preseg</th>\n",
       "      <th>plt_folseq</th>\n",
       "      <th>style</th>\n",
       "      <th>glide</th>\n",
       "      <th>pre_seg</th>\n",
       "      <th>fol_seg</th>\n",
       "      <th>context</th>\n",
       "      <th>vowel_index</th>\n",
       "      <th>pre_word_trans</th>\n",
       "      <th>word_trans</th>\n",
       "      <th>fol_word_trans</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1512</th>\n",
       "      <td>MARKDOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>T</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>F ER1 S T</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1513</th>\n",
       "      <td>MARKDOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>F ER1 S T</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1514</th>\n",
       "      <td>MARKDOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>F ER1 S T</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4250</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>IH1 SH UW0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4251</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>IH1 SH UW0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4252</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>IH1 SH UW0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4321</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>IH1 SH UW0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4322</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>IH1 SH UW0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4323</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>IH1 SH UW0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4353</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4354</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4355</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4756</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>S IH1 K S</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4757</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>S IH1 K S</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4758</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>S IH1 K S</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4810</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IY0</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>DH IY0</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4811</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>DH IY0</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4812</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>DH IY0</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5645</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5646</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5647</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5672</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>K W EH1 S CH AH0 N Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5673</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>K W EH1 S CH AH0 N Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5674</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>K W EH1 S CH AH0 N Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5852</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>IH1 SH UW0 Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5853</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>IH1 SH UW0 Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5854</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>IH1 SH UW0 Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6026</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>D AH0 S IH1 ZH AH0 N Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6027</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>D AH0 S IH1 ZH AH0 N Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6028</th>\n",
       "      <td>THEODOREBOLSON</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>D AH0 S IH1 ZH AH0 N Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3107718</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>R AY1 T</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3107719</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>R AY1 T</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3107720</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>R AY1 T</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3107950</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>T EH1 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>K EY1 S AH0 Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3107951</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>T EH1 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>K EY1 S AH0 Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3107952</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>T EH1 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>K EY1 S AH0 Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3108019</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3108020</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3108021</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3108989</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>T</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>F ER1 S T</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3108990</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>F ER1 S T</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3108991</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>F ER1 S T</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3109819</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>T EH1 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>R EH2 G Y AH0 L EY1 SH AH0 N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3109820</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>T EH1 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>R EH2 G Y AH0 L EY1 SH AH0 N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3109821</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>T EH1 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>R EH2 G Y AH0 L EY1 SH AH0 N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3109918</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>W IH1 CH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3109919</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>W IH1 CH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3109920</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>AH0 L EH1 V AH0 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>W IH1 CH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3110751</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IY0</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>DH IY0</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>AH0 N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3110752</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>DH IY0</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>AH0 N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3110753</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>DH IY0</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>AH0 N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3116417</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>F AO0 R T IY1 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3116418</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>F AO0 R T IY1 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3116419</th>\n",
       "      <td>SETHPWAXMAN</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>F AO0 R T IY1 N TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3117429</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>F AO1 R TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>V AY0 AH0 L EY1 SH AH0 N Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3117430</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>F AO1 R TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>V AY0 AH0 L EY1 SH AH0 N Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3117431</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>F AO1 R TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>V AY0 AH0 L EY1 SH AH0 N Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3117488</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>labial</td>\n",
       "      <td>voiced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>two_fol_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TH</td>\n",
       "      <td>M</td>\n",
       "      <td>initial</td>\n",
       "      <td>1</td>\n",
       "      <td>F AO1 R TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3117489</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>EH</td>\n",
       "      <td>e</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_one_syl</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>3</td>\n",
       "      <td>F AO1 R TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3117490</th>\n",
       "      <td>JOHNPAULSTEVENS</td>\n",
       "      <td>AMENDMENT</td>\n",
       "      <td>AH</td>\n",
       "      <td>@</td>\n",
       "      <td>nasal</td>\n",
       "      <td>apical</td>\n",
       "      <td>voiced</td>\n",
       "      <td>nasal_labial</td>\n",
       "      <td>complex_coda</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>internal</td>\n",
       "      <td>7</td>\n",
       "      <td>F AO1 R TH</td>\n",
       "      <td>AH0 M EH1 N D M AH0 N T</td>\n",
       "      <td>SP</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3551 rows × 18 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Speaker       word vowel plt_vclass plt_manner plt_place  \\\n",
       "1512          MARKDOLSON  AMENDMENT    AH          @      nasal    labial   \n",
       "1513          MARKDOLSON  AMENDMENT    EH          e      nasal    apical   \n",
       "1514          MARKDOLSON  AMENDMENT    AH          @      nasal    apical   \n",
       "4250      THEODOREBOLSON  AMENDMENT    AH          @      nasal    labial   \n",
       "4251      THEODOREBOLSON  AMENDMENT    EH          e      nasal    apical   \n",
       "4252      THEODOREBOLSON  AMENDMENT    AH          @      nasal    apical   \n",
       "4321      THEODOREBOLSON  AMENDMENT    AH          @      nasal    labial   \n",
       "4322      THEODOREBOLSON  AMENDMENT    EH          e      nasal    apical   \n",
       "4323      THEODOREBOLSON  AMENDMENT    AH          @      nasal    apical   \n",
       "4353      THEODOREBOLSON  AMENDMENT    AH          @      nasal    labial   \n",
       "4354      THEODOREBOLSON  AMENDMENT    EH          e      nasal    apical   \n",
       "4355      THEODOREBOLSON  AMENDMENT    AH          @      nasal    apical   \n",
       "4756      THEODOREBOLSON  AMENDMENT    AH          @      nasal    labial   \n",
       "4757      THEODOREBOLSON  AMENDMENT    EH          e      nasal    apical   \n",
       "4758      THEODOREBOLSON  AMENDMENT    AH          @      nasal    apical   \n",
       "4810      THEODOREBOLSON  AMENDMENT    AH          @      nasal    labial   \n",
       "4811      THEODOREBOLSON  AMENDMENT    EH          e      nasal    apical   \n",
       "4812      THEODOREBOLSON  AMENDMENT    AH          @      nasal    apical   \n",
       "5645      THEODOREBOLSON  AMENDMENT    AH          @      nasal    labial   \n",
       "5646      THEODOREBOLSON  AMENDMENT    EH          e      nasal    apical   \n",
       "5647      THEODOREBOLSON  AMENDMENT    AH          @      nasal    apical   \n",
       "5672      THEODOREBOLSON  AMENDMENT    AH          @      nasal    labial   \n",
       "5673      THEODOREBOLSON  AMENDMENT    EH          e      nasal    apical   \n",
       "5674      THEODOREBOLSON  AMENDMENT    AH          @      nasal    apical   \n",
       "5852      THEODOREBOLSON  AMENDMENT    AH          @      nasal    labial   \n",
       "5853      THEODOREBOLSON  AMENDMENT    EH          e      nasal    apical   \n",
       "5854      THEODOREBOLSON  AMENDMENT    AH          @      nasal    apical   \n",
       "6026      THEODOREBOLSON  AMENDMENT    AH          @      nasal    labial   \n",
       "6027      THEODOREBOLSON  AMENDMENT    EH          e      nasal    apical   \n",
       "6028      THEODOREBOLSON  AMENDMENT    AH          @      nasal    apical   \n",
       "...                  ...        ...   ...        ...        ...       ...   \n",
       "3107718      SETHPWAXMAN  AMENDMENT    AH          @      nasal    labial   \n",
       "3107719      SETHPWAXMAN  AMENDMENT    EH          e      nasal    apical   \n",
       "3107720      SETHPWAXMAN  AMENDMENT    AH          @      nasal    apical   \n",
       "3107950      SETHPWAXMAN  AMENDMENT    AH          @      nasal    labial   \n",
       "3107951      SETHPWAXMAN  AMENDMENT    EH          e      nasal    apical   \n",
       "3107952      SETHPWAXMAN  AMENDMENT    AH          @      nasal    apical   \n",
       "3108019      SETHPWAXMAN  AMENDMENT    AH          @      nasal    labial   \n",
       "3108020      SETHPWAXMAN  AMENDMENT    EH          e      nasal    apical   \n",
       "3108021      SETHPWAXMAN  AMENDMENT    AH          @      nasal    apical   \n",
       "3108989      SETHPWAXMAN  AMENDMENT    AH          @      nasal    labial   \n",
       "3108990      SETHPWAXMAN  AMENDMENT    EH          e      nasal    apical   \n",
       "3108991      SETHPWAXMAN  AMENDMENT    AH          @      nasal    apical   \n",
       "3109819      SETHPWAXMAN  AMENDMENT    AH          @      nasal    labial   \n",
       "3109820      SETHPWAXMAN  AMENDMENT    EH          e      nasal    apical   \n",
       "3109821      SETHPWAXMAN  AMENDMENT    AH          @      nasal    apical   \n",
       "3109918      SETHPWAXMAN  AMENDMENT    AH          @      nasal    labial   \n",
       "3109919      SETHPWAXMAN  AMENDMENT    EH          e      nasal    apical   \n",
       "3109920      SETHPWAXMAN  AMENDMENT    AH          @      nasal    apical   \n",
       "3110751  JOHNPAULSTEVENS  AMENDMENT    AH          @      nasal    labial   \n",
       "3110752  JOHNPAULSTEVENS  AMENDMENT    EH          e      nasal    apical   \n",
       "3110753  JOHNPAULSTEVENS  AMENDMENT    AH          @      nasal    apical   \n",
       "3116417      SETHPWAXMAN  AMENDMENT    AH          @      nasal    labial   \n",
       "3116418      SETHPWAXMAN  AMENDMENT    EH          e      nasal    apical   \n",
       "3116419      SETHPWAXMAN  AMENDMENT    AH          @      nasal    apical   \n",
       "3117429  JOHNPAULSTEVENS  AMENDMENT    AH          @      nasal    labial   \n",
       "3117430  JOHNPAULSTEVENS  AMENDMENT    EH          e      nasal    apical   \n",
       "3117431  JOHNPAULSTEVENS  AMENDMENT    AH          @      nasal    apical   \n",
       "3117488  JOHNPAULSTEVENS  AMENDMENT    AH          @      nasal    labial   \n",
       "3117489  JOHNPAULSTEVENS  AMENDMENT    EH          e      nasal    apical   \n",
       "3117490  JOHNPAULSTEVENS  AMENDMENT    AH          @      nasal    apical   \n",
       "\n",
       "        plt_voice    plt_preseg       plt_folseq  style glide pre_seg fol_seg  \\\n",
       "1512       voiced           NaN      two_fol_syl    NaN   NaN       T       M   \n",
       "1513       voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "1514       voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "4250       voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "4251       voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "4252       voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "4321       voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "4322       voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "4323       voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "4353       voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "4354       voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "4355       voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "4756       voiced           NaN      two_fol_syl    NaN   NaN       S       M   \n",
       "4757       voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "4758       voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "4810       voiced           NaN      two_fol_syl    NaN   NaN     IY0       M   \n",
       "4811       voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "4812       voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "5645       voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "5646       voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "5647       voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "5672       voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "5673       voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "5674       voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "5852       voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "5853       voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "5854       voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "6026       voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "6027       voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "6028       voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "...           ...           ...              ...    ...   ...     ...     ...   \n",
       "3107718    voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "3107719    voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "3107720    voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "3107950    voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "3107951    voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "3107952    voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "3108019    voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "3108020    voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "3108021    voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "3108989    voiced           NaN      two_fol_syl    NaN   NaN       T       M   \n",
       "3108990    voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "3108991    voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "3109819    voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "3109820    voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "3109821    voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "3109918    voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "3109919    voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "3109920    voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "3110751    voiced           NaN      two_fol_syl    NaN   NaN     IY0       M   \n",
       "3110752    voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "3110753    voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "3116417    voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "3116418    voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "3116419    voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "3117429    voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "3117430    voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "3117431    voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "3117488    voiced           NaN      two_fol_syl    NaN   NaN      TH       M   \n",
       "3117489    voiced  nasal_labial  complex_one_syl    NaN   NaN       M       N   \n",
       "3117490    voiced  nasal_labial     complex_coda    NaN   NaN       M       N   \n",
       "\n",
       "          context  vowel_index        pre_word_trans               word_trans  \\\n",
       "1512      initial            1             F ER1 S T  AH0 M EH1 N D M AH0 N T   \n",
       "1513     internal            3             F ER1 S T  AH0 M EH1 N D M AH0 N T   \n",
       "1514     internal            7             F ER1 S T  AH0 M EH1 N D M AH0 N T   \n",
       "4250      initial            1  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "4251     internal            3  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "4252     internal            7  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "4321      initial            1  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "4322     internal            3  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "4323     internal            7  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "4353      initial            1  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "4354     internal            3  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "4355     internal            7  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "4756      initial            1             S IH1 K S  AH0 M EH1 N D M AH0 N T   \n",
       "4757     internal            3             S IH1 K S  AH0 M EH1 N D M AH0 N T   \n",
       "4758     internal            7             S IH1 K S  AH0 M EH1 N D M AH0 N T   \n",
       "4810      initial            1                DH IY0  AH0 M EH1 N D M AH0 N T   \n",
       "4811     internal            3                DH IY0  AH0 M EH1 N D M AH0 N T   \n",
       "4812     internal            7                DH IY0  AH0 M EH1 N D M AH0 N T   \n",
       "5645      initial            1  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "5646     internal            3  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "5647     internal            7  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "5672      initial            1  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "5673     internal            3  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "5674     internal            7  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "5852      initial            1  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "5853     internal            3  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "5854     internal            7  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "6026      initial            1  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "6027     internal            3  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "6028     internal            7  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "...           ...          ...                   ...                      ...   \n",
       "3107718   initial            1  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3107719  internal            3  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3107720  internal            7  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3107950   initial            1            T EH1 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3107951  internal            3            T EH1 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3107952  internal            7            T EH1 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3108019   initial            1  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3108020  internal            3  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3108021  internal            7  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3108989   initial            1             F ER1 S T  AH0 M EH1 N D M AH0 N T   \n",
       "3108990  internal            3             F ER1 S T  AH0 M EH1 N D M AH0 N T   \n",
       "3108991  internal            7             F ER1 S T  AH0 M EH1 N D M AH0 N T   \n",
       "3109819   initial            1            T EH1 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3109820  internal            3            T EH1 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3109821  internal            7            T EH1 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3109918   initial            1  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3109919  internal            3  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3109920  internal            7  AH0 L EH1 V AH0 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3110751   initial            1                DH IY0  AH0 M EH1 N D M AH0 N T   \n",
       "3110752  internal            3                DH IY0  AH0 M EH1 N D M AH0 N T   \n",
       "3110753  internal            7                DH IY0  AH0 M EH1 N D M AH0 N T   \n",
       "3116417   initial            1    F AO0 R T IY1 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3116418  internal            3    F AO0 R T IY1 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3116419  internal            7    F AO0 R T IY1 N TH  AH0 M EH1 N D M AH0 N T   \n",
       "3117429   initial            1            F AO1 R TH  AH0 M EH1 N D M AH0 N T   \n",
       "3117430  internal            3            F AO1 R TH  AH0 M EH1 N D M AH0 N T   \n",
       "3117431  internal            7            F AO1 R TH  AH0 M EH1 N D M AH0 N T   \n",
       "3117488   initial            1            F AO1 R TH  AH0 M EH1 N D M AH0 N T   \n",
       "3117489  internal            3            F AO1 R TH  AH0 M EH1 N D M AH0 N T   \n",
       "3117490  internal            7            F AO1 R TH  AH0 M EH1 N D M AH0 N T   \n",
       "\n",
       "                       fol_word_trans  \n",
       "1512                               SP  \n",
       "1513                               SP  \n",
       "1514                               SP  \n",
       "4250                       IH1 SH UW0  \n",
       "4251                       IH1 SH UW0  \n",
       "4252                       IH1 SH UW0  \n",
       "4321                       IH1 SH UW0  \n",
       "4322                       IH1 SH UW0  \n",
       "4323                       IH1 SH UW0  \n",
       "4353                               SP  \n",
       "4354                               SP  \n",
       "4355                               SP  \n",
       "4756                               SP  \n",
       "4757                               SP  \n",
       "4758                               SP  \n",
       "4810                               SP  \n",
       "4811                               SP  \n",
       "4812                               SP  \n",
       "5645                               SP  \n",
       "5646                               SP  \n",
       "5647                               SP  \n",
       "5672             K W EH1 S CH AH0 N Z  \n",
       "5673             K W EH1 S CH AH0 N Z  \n",
       "5674             K W EH1 S CH AH0 N Z  \n",
       "5852                     IH1 SH UW0 Z  \n",
       "5853                     IH1 SH UW0 Z  \n",
       "5854                     IH1 SH UW0 Z  \n",
       "6026           D AH0 S IH1 ZH AH0 N Z  \n",
       "6027           D AH0 S IH1 ZH AH0 N Z  \n",
       "6028           D AH0 S IH1 ZH AH0 N Z  \n",
       "...                               ...  \n",
       "3107718                       R AY1 T  \n",
       "3107719                       R AY1 T  \n",
       "3107720                       R AY1 T  \n",
       "3107950                 K EY1 S AH0 Z  \n",
       "3107951                 K EY1 S AH0 Z  \n",
       "3107952                 K EY1 S AH0 Z  \n",
       "3108019                            SP  \n",
       "3108020                            SP  \n",
       "3108021                            SP  \n",
       "3108989                            SP  \n",
       "3108990                            SP  \n",
       "3108991                            SP  \n",
       "3109819  R EH2 G Y AH0 L EY1 SH AH0 N  \n",
       "3109820  R EH2 G Y AH0 L EY1 SH AH0 N  \n",
       "3109821  R EH2 G Y AH0 L EY1 SH AH0 N  \n",
       "3109918                      W IH1 CH  \n",
       "3109919                      W IH1 CH  \n",
       "3109920                      W IH1 CH  \n",
       "3110751                         AH0 N  \n",
       "3110752                         AH0 N  \n",
       "3110753                         AH0 N  \n",
       "3116417                            SP  \n",
       "3116418                            SP  \n",
       "3116419                            SP  \n",
       "3117429    V AY0 AH0 L EY1 SH AH0 N Z  \n",
       "3117430    V AY0 AH0 L EY1 SH AH0 N Z  \n",
       "3117431    V AY0 AH0 L EY1 SH AH0 N Z  \n",
       "3117488                            SP  \n",
       "3117489                            SP  \n",
       "3117490                            SP  \n",
       "\n",
       "[3551 rows x 18 columns]"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[data.word == 'AMENDMENT'][['Speaker','word','vowel','plt_vclass',\n",
    "       'plt_manner', 'plt_place', 'plt_voice', 'plt_preseg', 'plt_folseq',\n",
    "       'style', 'glide', 'pre_seg', 'fol_seg', 'context', 'vowel_index',\n",
    "       'pre_word_trans', 'word_trans', 'fol_word_trans']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def word_selection(frequency_cut, word_freq_df):\n",
    "    return list(word_freq_df[word_freq_df>frequency_cut].index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def word_dataset_generation(word_selected, data_ori):\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Word Model - Embedding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/apple/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:8: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "/Users/apple/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:12: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "/Users/apple/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:17: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "/Users/apple/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:21: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n"
     ]
    }
   ],
   "source": [
    "data_featureV1 = data[['Speaker','word','label','vowel', 'stress', 'dur','pre_seg','fol_seg',\n",
    "                          'F1', 'F2','F3', 'B1', 'B2', 'B3','F1@20%', 'F2@20%','F1@35%', 'F2@35%', 'F1@50%', 'F2@50%', 'F1@65%', 'F2@65%', 'F1@80%','F2@80%','nFormants']]\n",
    "# test vowel by word model \n",
    "feature_V1 = data_featureV1[data_featureV1['word'] == 'AMENDMENT']\n",
    "from sklearn import preprocessing\n",
    "vowel_encoder = preprocessing.LabelEncoder()\n",
    "vowel_encoder.fit(feature_V1.vowel)\n",
    "feature_V1['vowel_num'] = vowel_encoder.transform(feature_V1.vowel) \n",
    "\n",
    "word_encoder = preprocessing.LabelEncoder()\n",
    "word_encoder.fit(feature_V1.word)\n",
    "feature_V1['word_num'] = word_encoder.transform(feature_V1.word) \n",
    "\n",
    "\n",
    "pre_seg_encoder = preprocessing.LabelEncoder()\n",
    "pre_seg_encoder.fit(feature_V1.pre_seg)\n",
    "feature_V1['pre_seg_num'] = pre_seg_encoder.transform(feature_V1.pre_seg) \n",
    "\n",
    "fol_seg_encoder = preprocessing.LabelEncoder()\n",
    "fol_seg_encoder.fit(feature_V1.fol_seg)\n",
    "feature_V1['fol_seg_num'] = fol_seg_encoder.transform(feature_V1.fol_seg)\n",
    "\n",
    "feature_V1.index = feature_V1['Speaker']\n",
    "label_speaker = feature_V1.groupby(['Speaker'])['label'].mean()\n",
    "speaker = np.array(list(label_speaker.index))\n",
    "feature_V1 = feature_V1.dropna()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "\n",
    "label_speaker = feature_V1.groupby(['Speaker'])['label'].mean()\n",
    "speaker = np.array(list(label_speaker.index))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Base model Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAl8AAAJaCAYAAAARciKuAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3Xd8zdcfx/HXvTc7EoQkRkQS4xubCIqitVc31aJFBx1K\nS5dqUaO01aXa0qlGVfujS221xYq9vkjsIJEE2cm99/v741xEBAlJpPV5Ph4eufc7zvd879Xm7Zzz\nPcdkGAZCCCGEEKJomG93BYQQQggh7iQSvoQQQgghipCELyGEEEKIIiThSwghhBCiCEn4EkIIIYQo\nQhK+hBBCCCGKkNPtroAQt5umaRZgMNAT9d+EC/AXMELX9QxN06YBu3Vdn1iA17wfaKvr+iBN0+oD\nc4FzwI9AVV3XB91kud8AU3Rdj9Q07VvgZ13XlxVAfdcDHqjPRgN2OXbt0XW9162Wn8v1+gLddF3v\nms/ztgP3AA9ePF/TtC5AE13XRxR0PQuTpmnhwP90XQ/Ksb0vMAqYBtyj6/o92fZ9BLwEhOi6fiLb\n9pVAK6CKruvR2bbfA6wAXtN1faLj73o7IC5Hdb7XdX2SoxyA1rqu2x1llAXidF03aZoWBERx+e+H\nGUgGPtV1/RfH8aOAkcBTuq7/kK0unsAZYKXje+sLfAYczlGXnbquP+ko5wWgrq7rp7OVsxsYCAQC\nQxybA4G0bPf1ErAa6Af8AATrun4EIYqIhC8h4CugNNBG1/Xzjl8Cs4BvgScK44K6rv8J/Ol4ez+w\nQtf1Zwqg6HbAVMc1CqI8HGU1A3D8ct2t63r9giq7IF2sl6Zp2Tc3AnxuS4UKRzKQ4viZfHGjpmlu\nwJPA/1Dh480c5x0DegOjs217EhV4svvkBv/QuAt4Cxh7jf1p2f9+aJpWGViuaZpN1/W52eryBCr4\nXPRI9vtxWHODAO4NTNc0rYOu61dMWqnr+nRguqMO08jxDyhN01KAWMfbnNcVolBJ+BJ3NEeY6AWU\n13X9AoCu6ymapj0HNM/l+KeAAagWIB9ggq7rX2maVg71P/qyjkP/1nX9nets7wt0A2aj/vVu0TTN\nHVjK5RabcsAUIBSwo1q0JmmadhfwAeAKlAeW6rr+tKZp44AKwCxN054E3gcm67r+P03THkS1NpiB\nJGCIruubHK0HQY5yKgMngd66rp/Kx2d4D6qFIgUogQo77YG3HZ9TKvCqrusRjuOHo37RmoEjwAu6\nrsfkUnR5TdMWOe7pKPCsruunHa0vk3Vd/5+jvEvvNU0zAN9sdWsCPOf4fM/ruj48R91bAB+iWvUy\ngbd1XV/k+H4ecnzu1Rz30EfX9X3ZzrUAp4Gmuq4f0jRtGPCcruuVHfuXAR+hWoG+Qn3OJuBHXdc/\ndPzdWwPsc+xrhWqxewU4z+XWo5y2AmuBLYAl2/bHUK1OHwNLNE0bret6arb9M8kWvjRN8wDuBvLb\nMjoGeE3TtGW6rm+40cG6rh/VNG0E8BqqhRdgEfCgpmkB2Vro+jjqGJqPusxEhcGhQH5bphejPv/F\nuq6fzee5QtwSGfMl7nQNUV1nF7Jv1HX9dLZ/pQOgaVoJ4Fmgs67rDYAeqBCEY3u0ruthQAugmqZp\nJa+z/eJ1ZqEC1pxcuu++BA7ouh4KNAX6a5pWFdVFOkLX9SZATeB+TdMaOoJFDNBL1/WN2eod6rjG\nI7qu1wNGAH9omubtOKQF0N1xnRRUWMmv2sDjuq7XRXXxvJftc+oPzNM0zdMRCusAjR2tIwtQLYy5\nqQ4MdJS5CxXw8sXxOVz8fHMGrzKoVqLBjmv0AWZqmhbsOKQV8JKu67WBjeRoSdJ13Ybqnu7o2NQR\ncNE0rbrjO64HLEe1oq7Qdb0OKtD31jTtMcc5AcAYXderA/6o7sSWuq43QoXB3O4pWtf1Abqur9R1\n/f1su14AZuq6vgU45bif7LYB6Y5ACvAwqvXVmuO4VzRN257jT53sVUAFqVnZ/g7dyA7U935RFvAL\n6h8+aJoWCHgBu3Oc1yKXuvTLtj8deBx4R9O0sDzWRd2Erj+i63qSrusdb3y0EAVLWr7Enc5OHv8R\nout6sqZpXYEumqZVA+qjWnpA/Ut+geOXyDLgTUcX5rW25+WSbYHXHdc+jwo4aJrWB+isadpbqFYC\n92z1yE1rYPnFsT66rv+jaVosKniCGmNzMXxu4+a66I7run7U8bodqiVtebb7tANVga5AY2CLY58F\n1eqUm2W6rh9yvP4O2HwT9bqeJsChi0FV1/U9mqatQ40ZM4DIbK0yW1FhJaffgOc0TfsRKAf8hLr/\nBNTfCWdU4GrvuMZ5RxdYJ2ADKvhEOMpqAyzJNn7pay4Hu+tyBI96qJZUUGMHB2uaNiVHd9wMVOvX\nRlQ4ewV4NUdxN+p2RNf1bzRN64D6B8LLeaiigWo9zG468D2qhfYJx/ucbtTtiK7ruzRNexv4SdO0\nhtc7VojiQlq+xJ1uI1BD0zSv7Bs1Tauoadrfjq7Ai9sCgO2o7rm1qG41AHRd3wwEo35hBgGbHK1R\nuW7PY92sqF9aF68f4mhpWA10BvajupBOorqzrsWSvRwHMyoYgBqIfJFxg7KuJfuYGQsq7NW/+AfV\nNbTbse/9bNvDyaV718GWo75Z16ijy03U92I9b/VzWYq6hy7ASsf79qhxfP9zlJfzvOzXyNB1PXvL\nU/Zjc7ZIXc+LjuMjNU07AgxCtRx2ynHcLKCbo3XPW9f1nC1N+fEsqnWwdx6ObUSOblTHfxsWxwMn\nPVDB9abouv45cIibaB0V4naQ8CXuaI6xRrOA7y92oTh+fgnE67qe/RdwOOppqbHAElQrDpqmWTRN\nmwC8o+v676huwT1A7Wttz2P1lqGexsLRjbUcNf6oEfCGruvzUN1WVbk89sfK5V/sFy0HOmiaFuIo\nqzVQCRU8C8NyoL2juxNN0zoDO1EtdIuBZ7J1V41Gtcbk5l5HiyGortCFjtdxqO8CTdNqAnVvUJ/c\nPhNQLU6hmqY1dpRVC2iJClF5out6OrAKNZ5uieN1U1RX7mJd15NQLVwvOq5REjXIfWkuxS1BfW4B\njvd981IHTdNKocZ7ddV1PcjxJwA1HuqKVinH3/edqBana33ueaLreiIqeL13g/pVB95BjX/LaQbw\nCap7PeFW6oP6b6UL6r8HIYo1CV9CqLEye4H1mpqqYKPjfc6nBZcAJ1BjXvahxjbFof5n/ylQ3/GY\n+xbU4/E/X2d7XgxEtcrtBNYB43VdjwTGA1sdZb7p2HfxF8481Lil9hcL0XV9r+Me5znOmQDc5+jK\nLHCO6/UHftY0bQdqgPb9uq4no8Z3zQc2aJq2BxWc+l6jqJ2oULybK6cNGIsKKbtR4W31Dar0Dyp8\nfp6jnmeB7sDnmqbtQrW89NN1/UB+7hfV9Vgd+McR1ncA6xzBDNS4pjaOa2xCfUfTchai6/ouVDfz\nck3TtgBuebx+H2CvrusrcmwfC7TWNC1n2J8ONOPaLU25jfmaktuBuq6vQg3wz84923lbUfc6TNf1\nv3MpYiYq8E67Rl1yG/O15Rp1iUN9FjfbEipEkTEZRs5WdyGEEEIIUVik5UsIIYQQoghJ+BJCCCGE\nKEISvoQQQgghipCELyGEEEKIIiThSwghhBCiCP1rZriPjIyUxzKFEEII8a/RsGHDXCet/teEL4CG\nDQt35YjIyMhCv4bIP/leih/5Toon+V6KH/lOiqei+F4iIyOvuU+6HYUQQgghipCELyGEEEKIIiTh\nSwghhBCiCEn4EkIIIYQoQhK+hBBCCCGKkIQvIYQQQogiJOFLCCGEEKIISfgSQgghhChCEr6EEEII\nIYqQhC8hhBBCiCIk4UsIIYQQoghJ+BJCCCGEKEISvoQQQgghipCELyGEEEKIIiThSwghhBCiCEn4\nEkIIIYQoQhK+hBBCCCGKkIQvIYQQQogiJOFLCCGEEKIIFWr40jStiaZpK3PZfp+maZs1TYvQNO3Z\nwqyDEEIIIURxUmjhS9O014FvAbcc252BT4D2QCugv6Zp5QqrHkIIIYQQxUlhtnxFAQ/nsr0GcEjX\n9URd1zOBtUCLQqyHEEIIIQQAMfEx2O3221oHp8IqWNf1uZqmBeWyyxs4n+19ElCysOohhBBCiDuX\n3bCz9/Rhpq3cytlj8+nQZB/pmc40Yt1tq1Ohha/ruAB4ZXvvBZzLy4mRkZGFUqGivobIP/leih/5\nToon+V6KH/lOil5KVjp7Eo6wIyGa9ftTaXLUyiHndViT3Tj16POUi/nptn4vtyN87QOqaZrmAyQD\nLYGJeTmxYcOGhVkvIiMjC/0aIv/keyl+5DspnuR7KX7kOykahmFwIO4Yaw9vY030NnbGHCTlVAV8\nt9biiUY/M8MpjfjFRwGwLxhJm0cD8Pct/ExxLUUWvjRN6wmU0HX9a03ThgCLUWPOvtd1/WRR1UMI\nIYQQ/34pmWlsOLqLtdHbWXt4O7HJCQCkxQSSuaYvA4N/J6bTd8w0BVC3SnUyKgWRviSKV19cz/nz\nlW5r3Qs1fOm6fgS4y/H6p2zb/wL+KsxrCyGEEOK/wzAMDifEsCZ6G2sPb2Prif1Y7bZL+zPi63Ji\n1aM8bl/BQw+8xMduocRmladKSFmad2iBYYfz9k3A+tt3Ew63o9tRCCGEEOKG0rIy2HxsD2sPb2dN\n9DZiLsRd2mc2mQioUJ2o1NYc/Lkx4Qd38d1DPVkTYGUYDUjbehKP08m0mNMDk8lAHx/Hc92023g3\nl0n4EkIIIUSxcfzcadZEb2dt9DY2H99Lpi3r0r7S7l40Cq5HenAD/hfTgO3ve+AbEcs3nQfStOPv\nvEUToilJ+uZjpG04TvdpA/FwMYhdloxpowfaB7638c4uk/AlhBBCiNsmy2Zly/G9jsHy2zmaeOqK\n/bX8Q7g7pAFVQuqz0K8KX6wxk/YiWFZZeanZJEa+OYLfXSvQm7ZkGSbcdhzh7NrjhDTTqBQWQlpM\nFtufO8XEZU9hMm29TXd5JQlfQgghhLgt1kZv48OVMziSEHNpm5erB02D6tIiuAHNgutxwLMUHxsw\ncBkYo4G10KrKSr549SVK+h5mKE3YhmrRqnU2jdUrjmM2m9nXrh/LJhu4/XGSVvfVJqRe8VlMR8KX\nEEIIIYrUkYQYJq6YwZrD2wCoWNKPDlpTWoTUp26F6mC2MBfoYsCmBcBoYCNULHmCz/q9ysM15/An\nQbxPB1JwxtfTzrsd+1PFvTZDGUZSwzas829B+oDp2D2c6P37vbfzdq8i4UsIIYQQRcJqtzFp9Wxm\nbV2I1W6jhIs7/Zs+Qs+wjjhbnDgHfApMMuD4n8AYIBJcLBkM7/oJb7YaQypWhpqasdxQ00W0rVaW\n2kmNaFLpHuI5z93DumI3GRhPL8MAur3ejDIVvK5dqdtAwpcQQgghisS2E/v5cct8TJh4uE5rBt79\nKGU8SxENTAK+tUPKPGAssEOd89hdC/mi22B8OMg6oxwjLY2Js7nj6WLnjXu7sX5mJG/PfIt90fsJ\nHdwUk8nOjqMlMebuw6eCFw+/2vT23fA1SPgSQgghRJHItFkBaFK5NiM69Gcd8DHwuw2MX1Gha486\n9q4a0cx85hWqGH+SZlgY79aQn9Orgg0aVCzBu+1H8dm4j5kzZw4ly5YitH8jMslgr8mPo5224Aw8\nMfYe3Dxdbs/NXoeELyGEEEIUCcOwA3DGZKIxsMUKzAbGAbo6plpwKrMGTyDc+ACTLYN9Fl+GuYRz\nONUbJ7PBC83b0rtBH14dOpR58+bh5ubG2DkTiXVP5qTZmxUfZVE5JpXgev60frLubbrT65PwJYQQ\nQohClwh8sHcNAJs9/Ij5HszvgT1K7Q8KMvj6td9o6/QKpqRj2AwTP/jW46v46lhTzYT4uPJel3eo\n4V+VXbt2MX/+fDw9Pek3fTrfhYTjZt3NuqSKNH7/W+zA0x+1w2Ix37b7vR4JX0IIIYQoNIeAz4DZ\nJ/ZTfvcmEra2J+7zPnAM7EDVqvDBG/t4wHsQ5uPLADjhU4nhpmpsj1NTSPQMC2dwi0G4OjljBU7W\nqUODaVOpXC2TD8rV5ZzZCSx1ad15BvZ0K426VqN+m+Dbdcs3JOFLCCGEEAXKANagxnP9CRhpdsoO\n19H/+Iys82UACA2FUcMu0L3SGMzbP4XzVgxXb/4ICeb9g1VIzXLG19PC6E5DaBrUkFXp6Qx9th8n\nunbF+nAnXmlgpayRxtPnI1k4KpWgufswn0rGZDHx1Idtb+Pd35iELyGEEEIUiCzgF+ATIBIgFSxf\ng9N7WZyNewCAGjXtjHzHRPfaszCvfQ22ngZMJNYMZ3SqK//sDQCgXfVqvNrudb5296JnWhpJTz+N\n26pV+MYfo087V7xIx9tSmgstdxC8M5HAmr4cO5lE5xfCqRRa9vZ8AHkk4UsIIYQQtyQB+BqYDJwE\nSAaPr8A0EVJiwYYrbuWPMOTNNMY8moF5xUBYvE6dXL4OazVXRm6qwNkUdzxdTAxr8zQBNdvQxmRi\nX2oqPn364LZ+PdXah/PYx08A6ZSzlKWz5938Fh0JwLG9cXh4u9JzVMvb8hnkh4QvIYQQQtyUg6jx\nXD8AqQAXwO8LSP0IkuPVMZW0eMxNv6dT61jGlk/B9NNUMOzg4UdasxZ8EnuUOSuqABAWUJ4xnYbx\nc0k/HgGy0tKo1KsXtk2b8PP358vhE9hqiaacUxlauYfjbLoyxgTV8aOkr2fRfQA3ScKXEEIIIfJt\nKvA8anwX56DKJDjzKcQmqv133QXPvZLA5OMvcZ9xkLfOR2GKTQSTBcL6sLfqQYYtTeVIQhWczPBC\n84dp06gbT5rNrHBc43k3N/y7NiXzhbvoV+URqoZUJdgegrvZ9bbcc0GR8CWEEEKIfIkCXgGMeKj3\nKURNgqgLal+LFjBiBLRpA5/OGck0YxG1TImQAVS6F9s9jfg+ehFTftWw2s2E+JTkvS5vssM/mAbA\nOcAvMZH3EhOoWT6ZHd3UU4vpbqr8Wwtepxw/TbdQxq2T8CWEEEKIPDOAvnGQ9hE4fQE7ktX21q1V\n6GrVCkg5Q+zcAbxy8g8wgc2zApZ7h3HCbx7DF0ayPaYGAD3D7qVfi34MdXZhpqP8DvHx2Po+waEX\nW5BVpipmTDRyrU1Nlyq3WPNDwFAAzp9vSsmSt1jcLZDwJYQQQog8OX0a+kyEtV8BqWAFOnSAd96B\n5s0BWxZs/RJj3Qj8Mi+QaZjZW+kh6j34Ir8feJ73p9dSU0iUcGdMx1ewBtWlMXAUcAfGxMXxd48e\nNHi5JZWbVsXZcKJjieZUdPK7xZqnAA8B54GHiIvrQWDgLRZ5C4rn1K9CCCGEKDZOnoTBgyE4GJZ8\nBKRC/S6wYQMsWuQIXsdWwIwGsOJlTJkXWGOU58USvSjfeSpDFn3NyMX1Sc1ypn31Bvyvz+csDqrL\nPajgFQ4sPX2aP7t1Q9d1Dv+5l5L2Ejzi1bYAgpcBPAvsBjRgGtLtKIQQQohi6dgxeP99+PZbyMx0\nbHwQmr8Naxo6IsyF47DqVTjwCwA272CGp1ZlYVZpnq3dhZ4zh3M2xaCESyZvtnmUrjV7EGMy8S7q\n/OHASGD4r5Op1FXDebEzX0/4kjKlymAyFURI+gy1gKQnMA/wLoAyb42ELyGEEEJc4fBhGD8epk2D\nrCwwmaBpd4h4G0rUhZ8AkzUDIj+CDePAmgpO7tDkLT5O8mfhjuX4lSjNNxt+A6BhQCxjO52kQske\ngImLOa4yMBaIscZR5Yn6hJgM7nk6jLI+BTVJ6irgVcfraUDNAir31kj4EkIIIQQABw/Ce+/BjBlg\ns4HZDD17wkvD4WFHbpkABEYvgBWD4dwhtbF6N2j1EVGZZn6a9hoAscmJOJktDLy7DE+G/4LF/Bw5\nu/uMw4cZNutbqr/UGMNkUNulKppbSAHdzUngUcAGvAZ0K6Byb52ELyGEEOIOt28fjBsHs2eD3Q4W\nCzz5JLz1FmiaGjF1Cuh+LooXVrwC0X+pE31qQOvPoXIbsmxWBs0Zil3N/EVImQDGdxlIqF8P1Lir\nTldc0+nQIXiyF66fPE4GmQQ4+dPMrV4BdTVmoMJWLHAv8F4BlFlwJHwJIYQQd6jdu2HsWPjlFzAM\ncHKCfv1g2DCo4pjZYQUwKyuVcZvG8+bmDzHZMsDFC5qOggYvgcWZE+fO8NJvH3Li/BkAHqnbhtfv\n7YOb83lgM+ACtL503ej9+ynz2GNUa1iRcqEV8cKDdh53YTbl7znA7cuiyUjNymXPW8AGoBIwh+IW\nd4pXbYQQQghR6LZtU6Fr3jz13tkZnn4a3ngDgoIuH5dqGMw9OJd9K4dSOemY2ljzSWgxAUqUv3Tc\n0D8/ITr+BAB3B9dnRPtnHXuWoFq9WqEGvMOePXsY9NhjWBIS2JcSSiunMMq5+eJqcsnXPaybu48P\ne/6G3W5Q8+5K7F17HCdXi2Pvb46fPwK++Sq3KEj4EkIIIe4QmzfDmDHwl6PX0NUV+veH11+HgIAc\nB8fv5eQ/g5h8bDkAdt/6mNtMhorNryr3XFrSpdcVS2afGmKB4+flLseUlBQqhlfGNbA1CcPep4ab\nW77vY/G32/hiwN/Y7Qbtn2nA+rn7AGj3VH3HETbHz+B8l10UJHwJIYQQ/3ERETB6tJqTC8DdHZ57\nDl57DcqXz3FwxgWIeBdj2ySq2a0kuJXmQvNxBNXtD2bLVWVfmw1Y7HjdmbNnz1K2bFkqhgXxyMd9\nyTLDdOd0IH/h638frGfaGyoQ9nq3FScPxJOcmE6D9iHc07M28DlwDNXVWSpfZRcVCV9CCCHEf9Tq\n1aqla9ky9d7TE158EYYOBb+cc5cadtg7E1a/DqlnMDAxte4ATjYfyxiPm5n6YQsQDwSzadM5nnji\nft745G1szT0wzAYRzsHEmfO+xo9hGEx7czlzP4gAYMDnHQnQyjBr5Cpc3Jx44ctOmEw/AoMcZ3yJ\nhC8hhBBCFDrDgBUrVEvXqlVqm5cXDBoEL78MZXPLUWe2wT8DIWY9ACfKN+X+NpM55x/Grhtcb/tJ\nnbjkxFz2qC7HdevC6dOnF+UbBJLe2Bkzdio6BzPbvSGV8/hko81m54vnFrDk221YnMy88uP93PVg\nKAPrTAXg8ZEtKV/lH+BpxxmfZHtd/Ej4EkIIIf4DDAOWLFGha73KUJQqpZYFGjwYSpfO5aS0eFj3\nNuyYChjg4c+plh9QrWZv0k1mlnJxmHzuzqac49W/PsVm2GlUqRabj+/Jtnchq1a58tRTm0hPz6Lr\nwEcwO1uo6RJCgFsYRh6DV1aGlQ97/cb6uftxcXNi2P+60ahLNaYNW87p6ESC6vjx8KvngJ6AHRgF\nvJynsm8XCV9CCCHEv5hhwN9/q9C1ebPa5uMDQ4bAwIFQMreePbsNdn0La9+C9AQwWSBsEPamI+nh\nWpJ0oC/Q9jrXtdptvPHXZ8QlJxIWEMq9VcOzha9YoqO30a+fL5lZVnr16sXQuwdw0HacUOcgjuQx\neKUlZzLuoV/YvuwwniVdGTH/MWrdHciRXWf4beIGTCZ4bXZZLE7dgSxgCDAibx/cbSThSwghhPgX\nstvhjz/UmK5t29Q2X1949VV4/nnV1ZirmAhYPhBit6r3ga3h3klQthbfAGsAf+CjG1x/0urZbDmx\nj7Kepfig62CWH9yUbe9igoOtvPB2PVzufoynQrrjZHGihiXvTx9eiE9lVOfZHNgUQyl/T0Yv7klI\nvXLY7Qaf9/8bm9VO3wneVK7VD0gDngEmcrsXzc4LCV9CCCHEv4jNBnPnqnm6djkGZJUrp55cHDBA\nDarPVcppWPMm7PlRvS8RAPd8rJYGMpk4CbzuOPRzwOc6dViqb+DHLfNxMlv48L6X8S1xuU8zavMe\ndpc7SLVaHpTv9iRJhhP7rIdp5FQrz/d49uQFRrSfxbG9Z/EPKsWYpb2oUFXVaOGUSPQNJ6nTKplH\nXvsMSAJ6AFP4NwQvkPAlhBBC/CukpMCsWfDpp2o5IICKFdXEqM88o6aPyJUtC7ZPhvWjIPMCWFwg\n/DVoMgycVVIzgBeAC8ADXH8VxOj4k4xYNAWAIa16ERYQemlf+r5YFi9ey6af7Ixb9TxJFifKmksR\n5hp6RRmO6pNblU8ejOeddrOIPXqewFq+jF7ck7IVvQGIj0nix2H/4B+cwLsLZ2EyxwNdgRlAfqbB\nuL0kfAkhhBDF2MGD8OWX8MMPcP682hYYqJYA6tdPTZR6TUeXw4pBEL9XvQ/pAvd8CqWrXnHYr8Cf\ngDfwBdduP0rJTGPoHx+TmpVOx9Bm9Ay7PHnqpqVrSF6kA/DM+42JsTTBBSfaeTTFYrocjAxgpON1\n3xzlR207xciOszkXm4LWpCKjFjyOl8/liDZ10CLcS8QxMWIWLu5xqHUbfwGcr/MhFD8SvoQQQohC\nlJgI06bB999DbGz+zjUMiIu7/L5ZMzVPV7du4HK91XguHINVr8KBX9X7UlVU6KrS9apD44GXHK8/\nACpesy4GoxZNJTrhJCFlAhjZvv+lRbBnzpzJnM9+AKDHgAB8W7Tmgh1aeYRT0lLiinL+RM0A5g+8\nmG377jXHGN31Z1IvZNCgXQhvzeuOe4nLN7nxrwPsXhXJB2t/pLR/HNAE+IPc28+KNwlfQgghRCHY\nsQPGjQtk0SJIS7v5ctzcoGdPFbrCwm5wsDUdtnwEG98Dayo4ucNdb0PDIeCU+0zyQ4FYoCXwbK5H\nKDMiF7DkwAY8Xdz55IEheLio8mw2G3PnzgXAo0Uwrw4+iY/ncKKtM6niXOmKMuzAO47Xb3F5GotN\n8w8woftcMtOt3N29BkNnPIiz6+WIkpacyY/D5jJ68XQCtDigDmoesWs9VVC8SfgSQgghCkhWllqs\nevJkWLsWLi7q3K6dCk9NmkAeZ1m4xNv7OuO5sov+G1YMhnNR6n317tBqIngHXvOUJailp12BbwDz\nNY5bGRXJp6tmATCm0/ME+VQAICsrC2dnZ6ZPn87Y7z7hjJaIv/cGLCYnarq0u6qcX4FdQADQ37Ft\nxaxdfNLnD+w2g479w3j+y05YLFfW5Ocxi3hxyhSqhp3CMKpiMi3h+o8EFG8SvoQQQohbdOoUfP01\nTJ2qXoO2yjscAAAgAElEQVSa6qFz51hGjfIjNPT659+SxEOw8mUVvgDK1ITWn6spJK4jGRjgeD0K\nqJ7LMefSkvho5Qz+3LMagL6N7qNNtcYATJo0idWrVzNjxgy8vLyo1TmMYP80FqZ60MVjPiaTxxVl\nWbk81usd1IqOf32+iamD1PqP3Yc158lx917qyrzoUORh6rcdQq27j5GVWQFnl2VAuRt/LsWYhC8h\nhBDiJhkGfPMNvPIKpKaqbTVrqslNe/eGAweOExqacxHFApKVAhvHw5YPwZYJLl7Q7F2oPxAsNx6A\n/g5wBKiP6nrMzjAMFukRfPDPjySknsfF4sxzzR6hb6P7MQyDjz/+mI8//hiTycT69etp2boVaWXS\nAajiHHFVgAKYBehACNDXMJj17mpmv6tC3VMftuXhV5tedU56Siop5x+hQbso0pJK4e61Aqic98+o\nmJLwJYQQQtyE2Fg1xcNff6n3992nZpVv1Sr/XYv5Yhhw4H+waigkHVfbavWBFhPAM28tQhuBz1CT\nM3zHlc8Knr5wlnHLvmd1tJqENTygBu+0f5YgnwoYhsGECROYPHkyZrOZTz/9lDZt2rA5fTeGk8GF\nxDhCvVcCda+4XibwruP1O3aD7wcvZv7kzZjNJgZ+05X2T9W/qo6nos5ybG9Xmty3jZTz7jg5LyP3\n9rl/HwlfQgghRD7Nnw9PP60CWMmS8NVX8PjjRXDhs3vU1BHH/lHv/RpA68lQsVmei8hEzQVvoFq8\nLo7htxt2ftm+lM9WzyY1Kx0vVw9eadWLh+rci9mkxmB99NFHTJ48GYvFwueff84DDzzABXsK2zPU\nFBMHdkVgCjKuuuYPwGEg1DA48+TvrJm1GycXC6///DDNHrq6TzZy0UHOnuhDh2c2kp7iQnLCXPyD\nG+b5Hos7CV9CCCFEHqWkwNChamwXwL33wo8/QqVK1z/vlmWch4h3YeskMGzg5gN3vwd1ngFz/iYX\n/RTYDVRFjfUCiDp7gtFLvmZ7zAEAWldrxLA2/fArceWg9vbt2zN9+nTef/99OnVSc3yl2TPwNHuQ\ndj6dCwlXz6WRDoxxvK4zYT1rZu3GvYQLw39/lPptrlxuyDAMfh2/Dkxv8eiwCKyZzhjGH/gHd8zX\nPRZ3Er6EEEKIPNi8GXr1UpOeurjAe++psV7maz0iWBAMO+ydAavfgNQzgAnqPQ/Nx4B7mZsqco3j\n50jAyWZlysY/+Hbjb2TZrJT1LMWwNk/RtnrjS8fb7XaWL19Ou3btqFu3LhEREXhmW8PI38mHHiXa\nMyd6aa7XmwqcBHwPJpAy/B9KlnFn1MKeVG9U4apjJ/b+Hb/AL+kzfg12mwWz0/9wd/lvBS+Q8CWE\nEEJcl9UK48fDu++qdRVr11bL/NSte+Nzb8mZrWoB7FMR6n2FZuopRv8bTfZ1fYZhYM5M4+yZaHos\nn0ZU/AkAHqnbmpdb9sLb7XKwstlsDB06lF9//ZW3336b559//lLwshsGezIPEeoSjLPJiaxM61XX\numCzM9pqB1cnKg9Zgm9Fb8Ys6UmlGr5XHXs6OpESpb6jz/hlGIYJs2UGcP8t3WtxJeFLCCGEuIao\nKHjiCYhw5J8hQ2DcODXxaaFJi4e1w2Hn14ABHv7Q6kOo0fuGI/nthp2E1AucSYonNjmR2KQEziTH\nE5uUqH4mJ3IiKYHaWen84DgnsHQ5RrR7lkaBVy58bbVaefnll/ntt99wd3enTp06V+zfn3WYdenb\nico6Tv3MUL7b+DvlHHOe2rLszJ24lg/MJhLeaE6pjSdpcCCesev64hdYMte6O7v9xPNfqOkyTKYp\nQFEMors9JHwJIYQQORiGWktx8GBITlYLWP/4I7RpU4gXtdtg1zcqeKUngNkJGgyCpiPB1fuqw8+n\nJbM/9jD7Y4+w74z6efzcGax2240v5eSKn5cPD2l38UyTh3BzvnKtoqysLAYOHMj8+fPx9PRkxowZ\nNGnS5NL+DCOTTem7AQgmgIHz3ud8ejL3BoQAcGxvHN9NWM/Ow2rhooFHzvHGxqcpUepaqfU3fMoP\nAuB/7z9Etzf6X+O4/wYJX0IIIUQ2Z8/Cs8/C77+r948+qp5m9CnMCdVProd/BkLsNvU+sA20ngRl\namIYBrFJ8eyLPcL+M0dU4DpzhFNJZ3MtqrS7F35eZfArURq/Ej74e/ngV8Ln0uvBJXxY4OrBnyYT\n912jOitWrGD+/Pl4eXkxc+ZMwsPDL+0zDIMVqZtJNzLwN5fhoz9ncPzcGXxSSxP/nh0eVOE1c0pn\nsnzcaWkYjO5R65qLdat59h/DZLLz07v3sHp2W7q9cZOf47+EhC8hhBACFRjmzFGD6E+fVsv6fPGF\nGmRfaPN2ZSbD8hdh73QA7CUqcazRu+x319i/Zzv7zvyGHnuExLSkq051c3JF8wsk1C+YUL8gQv2C\nCCkTcFUrVk55eT6gffv2vPvuu4SHh1O//pVzcG3L2M8RawwuOLMpYg87Yg7idN4Zt898qdz1EADu\nWhMi69cGYIzJdJ3gtRZ4EMgkKaE/P42qSICWhwr+y0n4EkIIcceLjFRdjOvWqfctW8L06VC5kCdT\nP7xmPDv2rGG/KZz9JWqhpxuk/rMYWHzFcd5unoT6BVHDL5hQ/yBC/YKpXLo8lgJ81DItLY1BgwYx\nYMAAwsPDeeaZZ3I9LsQ5gINZx4jaGMOCneswpZsp820gd7WsyQtfJgLwmfsQkoD2qAW7c7cV6AKk\nAU9xLnYM6tnI/z4JX0IIIe5YZ87A8OHw/feq5cvPT00h0bcvWPI3fVae2ex2VkVFMjNyAZEnooDG\nasbTJLU8j2+J0ipk+QUR6q8CV3nvsrku2VNQUlJS6NOnDxEREezfv58VK1bg5HRlREi1p+NucsV8\nzsKJd5KZU3kxmKDKolBem9KDJvdXx2T6ktP4M4nmAIy95hX3AR2AC0B34GsgsbBur9iR8CWEEOKO\nk5kJkybB6NGQlATOzqrl6+231Yz1hSElM43fd69k9tZFHD93BgBPrDTjFDUa9iM0qBGhfkGU8Syk\nClxDUlISTzzxBJs3b8bf358ffvjhquCVZs9gXvJyOGpmUce9nPI+A09D+Sx/Zi1+EzdPF9TS2ZFM\nYBxpWHgAaJTrFQ8DbYGzQCdgJmqhozuHhC8hhBB3DMOAv/9WU0YcPKi2de0KH30E1Qtp2cCY83HM\n3raIeTv/ITkzDYAK3r70rNuKh9Y+TglnZ2i1Ot8z1ReECxcu0LNnT7Zt20b58uX55ZdfCAkJueIY\nm2Hnz9iVJLunkhibxoX4NLwfcOIs0PnuuxzBC2A3x/HhK54HYHSuV4xBBa8YoBUwF1DnH4o8BYBX\nGfeCvs1iR8KXEEKIO8K+fWow/WLHcKrQUPjkE+hYCBOoG4bBjpgDfLn7D7auPIjdUOsdNqio0bth\nF+6tGo4lZh2YrFCm/m0JXgAuLi54eXkREBDAr7/+SmBg4BX7z54/z/+ilkMVG+mnsjg0JJGh0x5k\nqnU2nIaGATWzHb2RcQwnE1d6kHNpbVAtXe2AaFSb2J/A5aD19xdbAGjTp14B32XxI+FLCCHEf1pi\nopqdfvJkNUN9yZLq/QsvqO7GgpRls7Ls4CZmbvmb3aejAHAyW+gYehe9G3amVrkqlw8+u0v9LFsn\nl5IKhw01ssocH0+KxYJbqVJ8//33JCYmUqHC5eV+DMNg1c+7iay6C8/qzmTEWXH/sxSTVj8M7gav\nfH4Yi8lM/YqXmwujieY7xmLGzqirnqk8D3QE9gK1gIXA5bnLDm09xf6IE3iWdOWeXrUL7f6LCwlf\nQggh/pNsNvjmGzWOKz5ercE4YACMGQO+V69uc0supCczd+c/zN62iDNJCYB6QvFuv9q83KkP/l65\nTBJ2MXz5FvY6Rcpp4AkgIjaWMj168I2HB51nz8bb2xt398stUMf2xTJl4GJ2/nMEvw6e1H6rPB1L\nN6Pm6+rRz3WHt2Mz7NQpXxVPl8vnjaYpVpzpQyyh+GW7cipwHxAJVAGWAleuS3mx1attv/rZujH/\nuyR8CSGE+M9ZuVINoN+5U71v1Qo+/RRyTFl1y44mnmJW5EL+2L2KdGsGAEE+FegV1omuNVuwb9ee\n3IMXQJyjckXQ8rUU6A2cPXUK/0cfxRQdjb16dTIyMi4dk56SyexPVxFTL4aUOsl473DnsUdb06Z5\nXSyWyy1ZW47vA6BhQI1L2/aTzAzuw4ksRpD9gYEM4GHUct4BwDKg/BV1S0pIY9VParb8zi80LMjb\nLrYkfAkhhPjPOHwYXnsN5s5V7ytXhokT4ZFHCm6iVMMw2HR8D7MiF7I6aisGajzXXZXr0LthZ5oH\n18NsusH8W4YBZ1XgwLfwwpcVGAFMAMwnT1Lp0UfJOnKEGjVqMGfOHMqUKYNhGET8sZ95i9cQ+IY3\nZUp74FPHk5HvtMenzNXLGm05vheA8EqXx3uNIgU7JXiW3wnhkWxX74Was8wXFQGDripv6ffbyUy3\nEtahChWrlblq/3+RhC8hhBD/eikpMH68CloZGeDhAcOGwdCh4F5AD89lWrNYuH8dsyIXoscdBcDF\n4kznGs3p3bAz1XwDb1BCNknHIPOCWjTbw+/Gx9+EY6ilqdcDZsOg7vPPc+bIEerUqcNPP/2Ej48P\np6MT+Xr4Imyd06g6oTQAPsml6FqpBR7mq9dhTM1MZ++ZaMwmEw0qqqnodwBz8MeVdN5mO/AIYAee\nRT3NWBK1hFDoVeXZ7QYLvooEoMuL4Vft/6+S8CWEEOJfyzDgp5/gjTfg5Em1rVcvmDABAgIK5hrx\nKef5dccyftm+hPjU8wD4eJSkR/32dK/X9ubm5SrkLsc/gH6owfUVgJ9MJspPnMi4ceOYNGkS7q6e\nzB69ml/Hr8OrsQtN7w/ElGnibq8G1PQOueaErjtiDmC126jlH0IJVw9AtawBPM9XBKChZox9GZgG\neAALgNz7eyMXHeJ0dCJ+lUsS3rlqgd1/cSfhSwghxL/S5s1qXFdEhHofHg6ffQbNmhVM+QfjjjNr\n6wL+3ruWTFsWANV9A+ndsDOdQpvj4nQLj0peGmxfsOErA3gdmOR43+bQIVrPn0/LwYMxVa/Ojz/+\nSOTiKKYMnkl6qTQy063UrVyDhlnVCC0TjJfZ47rlR55wjPeqpMZ7bUJNGOFBKm8yAbVW4zvA56j5\nu/4Arv2FzJ+sBtp3fiH8inFl/3USvoQQQvyrnDoFb70F06ap9/7+qqXrySfVE423wm7YWXd4BzMj\nF7Dh6K5L21uGhNE7vDONK9UqmGV+4gp+molDQA/UiolOwGv79/N3jx58cfYswf7+tGvRhW9eWULE\n/P2E/VAB/3a+1Dym0apB3ufVujjeq6FjvNc7ju2D+Ax/rMA8YBxqxvpfUBOq5u5UVAJbFx3C2dVC\nu6cK+EmIYk7ClxBCiH+FjAz1xOLYsZCcDC4uatLUt94C76vHhedLWlYGf+1ZzU9bF3I4IQYANydX\nHqjdip5hHQnyqXCDEvLprKPbsYCmmZgNDACSgGBg/O7djHnsMRITE2nRogWmI+V57tkvsVpsNP0t\nkNJN3XHFmdA6ee+bTcvKYNepQ5gwEVYxlNWokVzeZPEaH6KeYnwTMKG6HB+4bnkLvorEMKDlY7Uo\nWfb6LW7/NRK+hBBCFGuGAX/+qQbPR6l5S7n/frUkUNVbHCZks9v5duNvzIpcyPn0ZAD8vXx4vEFH\nHq7TmpLuJW6x9rmwZkCCDiYz+NS88fHXkQoMAr5zvO8ODN6xg/49e3Lu3Dka129KyV13M3POGlx8\nLbRdXh2nQPAwudHVsyU+lryPV9t56iBWu41QvyC83Dx527F9CP/gQyKXF8b+EjWxxbVlZVhZ+v12\nALoOzH0FyP8yCV9CCCGKrT174OWXYdky9b5mTbUkUPv2BVP+rzuW8uW6XwGoXa4KvcO70LZaY5wt\nhfjrMWE/GDYoXR2cb/5RzD3Ao6g5412Bz4D+wK+6zrlz56hSti4ZC2pzkkTKVynNfb/XI6bSaUqa\nS9DFsyXeZs98XS/SMb9XeKUaLEPN3OUDvMqwbEe9Dzx3w7LiY5JITkynTEUvqoUXcKviv4CELyGE\nEMVOQgKMHAlffaVmqi9VCkaPhueeK7glgWKTE5i05mcAxnZ6ga41WxTMeK4bOeV4QsD35sY5GaiW\nrkFAGmoChzlAlZQU7G7ueMQFUyv9QZx3+uId4kGnUQ14rPu9OLla2JKxh9ouVXOdRuJGIh3jvcIC\navKWY9sXLMaTbY53Q1DD/fPOyfn2rGl5u0n4EkIIUWxYrTB1KowYoQKY2azWYHz3XShbtmCv9cE/\nP5KSmcY9VRoWXfACOLxQ/QzKf/PdBdTYrp8d7/sCk4Gta9YQ/uwA6js9QNIeD3yqB9L4hyA8W1rI\nMiVjuNoxm5xo7HZz6yZmWDPZ6RjvFRdYm01AZ9bTgwcdR5QEJt5U2XciCV9CCCGKheXLVRfjbsfE\n7/feq6aOqFMIU2GtjtrK0gMbcXd25c02/YoueFkz4KijDzWoY75OjUQ9zRgFeAJfodZqXPDXYp5/\nYQBWexanK0bRbm5XfNpc7M40qOxcnizDhust3OKuU4fItGVRzbcyE1zdqcd25tEZE+mOI7qgBtqL\nvJDwJYQQ4raKjoZXX4XfflPvg4PVYPoHHyy4JYGyS81MZ/zy7wF4ofmjlPcu4Ca16zmxGqyp4FsP\nvCrm6RQDNW/Xa0AWUA/VzVjNbvDhsG/4bMZYLO5myibWpNeT/bC2ScaCGc0liPquGt7mm39owDAM\nNh/fy6erfwLANawj6exnLe1x5TxQGTgK3HXT17gTSfgSQghx29hscNddEBcHnp4wfLiaPsIt/0OS\n8mxqxFxiLpxF8wuiZ1j+Wp9u2RFHl2Nw5zwdHg88hZrIFOBFVOdezI7TPP3sVE412Mazvw/lwq4U\nBt31DIE1fdmTGUWIc8BNjeu6yDAMNh3bzZSIuWw9sR8AL/cSHK8RwFJaU5Y4oANwxHFGkzyVey42\nhW1Lo9n4h37TdfsvkPAlhBDitsnKUsHLyQl0HSrmrTHopqVlZTAzcgEmTLzT7hmczEU84Dt6gfoZ\n3OmGh65Frc14AiiFGmTf8UIG08euZNeFg9T5ripNAtVUFdVCPankVRaTyURt15uff8MwDDYc3cWU\niLlsP6kCkrebJ70bdsYjvCFNndpSiRPYaYGZH4CKqJnsc5+o1ZplY3/ECSIXRbF1cRRRW09fsb/m\n3ZVuuq7/ZhK+hBBC3HYWS+EHL1ADx612GyXdSlCnfBGvJXguGhJ1cC0JFZpe8zA7MAG1ZqIN1aH3\nk2FwbM4enhuyFP+RUOsxNTmql+FJmHso1V0qYzblPr2/DTiFWmj7Wn8uGIa6tmHHCKwFgbUAMJtM\nmDChmxJYzj1UJYqzNKQsfwGbUZ2iDVCTXSinoxPZuliFrR3/HCEtKfPSPmdXC7VbVSasQxUadqxC\npRpF2OVbjEj4EkIIIYrCxaccK7cHc+6/fk+jBtE7huTzOtDn4Al+3bCGreOOcfTcTvbOi6ZUWC8e\nqdmZ6h7BpJhM7OfKMHU02+uTgPVGdbs4uM50ZUugHfDiAvPpRB12c5iaBLII9XTjRgDSkpuwa+UB\nti6OZuviKGIOJlxRRqUaZQnrUIWwjlWo3TIQV/cCmivkX0zClxBCCFEUDl/scsx9vNdS1LzwsYAv\n8FVyLPG7I1hZLQPv+53w8yzBmpdWwm74fYPBL41COAacy8Ol/YHA7H8Mg+Qz0azduojjh7djyUil\ntJsXT4R34dH67fBwuTheLA0L92FiMwYhBLEUjDJE7zzN1sVH2bqoL3vXlsWaNefStTxLulKvbTAN\nO1alQfsQ/ALzPov+nULClxBCCFHYrOlw/B/1OvjqQf5fAS84Xlcw7PQ9t5I4czzUAAMzW49lEvnl\ndADOjxhBzIABl851J0ewyvEnALgYpQzDYFVUJFMj5rH3TDQAfh4l6dviMbrXyx66ADKBbsBq7Lby\nbP77C9bN3cy2JdEknk4G1HgtkwmqN65Aw45VCOtQheqNK2JxusUVzv/jJHwJIYQQhS3trApgnuXB\ns9xVuz8APO0ZpJhdiTGZOeFagrJZ51nrEsKO9HJkPnM/thMnCBs7lrb9+lGZy+GqDDeeYctu2Flx\naAtTI+ahxx4BoKxnKfo2uo9u9dri7ux6xfHWrExSzj1MSd8FJCeW4PW7H+HY3s2X9vtUsBLWYSdh\nHbyp3/YrvMvcWQtj3yoJX0IIIURRyWVQfIaRRZO0vTTLOsTUY3VwnriPUxlpNLuvOtMfrY1naQtH\n5sxh8+bNdO/ePV+Xsxt2/jm4makRczkQdwwAX8/S9Gt8H4/UbYubs8ulY88cOacGyi85SNMHP6L1\nE1tIOe/K8Da9iTlUnvptAx1jtzKpXOteTCYnYDcgwSu/JHwJIYS4YyRnpN7uKlzhWNYpFlzYwD1O\nakh8iz/WUdVWiqc+6cL3s6by7vCfef/99wkKCiIoKCjP5doNO8sObOLriHkcPOsIXSVK81TjB3i4\nTmvcnF1IT81i89KDbF0czbbFUZzQ4wGDZz9ZSOsntpCZ5syyaeN4YtwD1G5VGTcPZ9TTjfc4fr4E\nVC/Qz+NOIeFLCCFEgdiyBQYNUvN25ZXdXnj1ySk2OYGX5n0AQDXfwCK77lfAOmAmEId6mhGg4hmd\nhs47MTvBQUtZ1v3pzOjmAbQbXpnx48fzxRdfYDabefzxx2nQoEGermWz21l6YANfR8wjKv4EAP5e\nPjzV+AEerH0Pp/efY8FnkWxdHMXu1cewZtounevh7cpLX2+iRY8NGIYLLu5/8cDgnOtPzgVWA2WB\nd27lY7mjSfgSQghxy3btgvbtITHx5s6vfXPrPedZzPk4+v86luPnzlCtbCAfdB1UuBd0WFS6NG+j\npiIFyACWZNkImrSJei57MT/pzWIXjb9Tq2Du5kYTZzOjRo3i22+/xcnJicmTJ+cpeNnsdhbr6/km\n4jeiE04CUM6rDL20LgTEBLBz0hH6L/mKhJikS+eYTFCtUQXCOoQQ1qEKoU3nYLHMBSyYTD8DOYNX\nOmqRI4AxqKlfxc2Q8CWEEOKWHDgA7dqp4HXffTBxYv7XZKxcuXDqBnD83GmenTOWU0lnqeEfzJRu\nb1HK3avwLuigA+McNzbEsa1Uho2+rb4lNiKWjTXLsuSZu9juXokP3eFh4KMRI/juu+9wdnZm6tSp\ndOjQ4brXsNptLNq/nm82/MaRhBiwQYWE8oSdq0PqNju/bonEMCIvHV+6XIlLYat+uxBKlr04Xutr\n1KxiAN8DD+VytU9QywnVAZ65qc9EKBK+hBBC3LSjR6FtWzhzBtq0gV9+Kdx1GfMrOv4k/X8ZS1xK\nIvUqVGPyw2/i7eZZ6NdNA7oDaRYLjwNPnU3FADbV7Eztr7040N/Gpvl92e7uzjPAq47zwsPDmT17\nNlOmTKFNmzbXLN9qt7Hw/+zdd1zV1R/H8de9l8sGAcE9EAcuXDjB3NvUHKmllWnmLLWc2a+yTEvN\nzNWwtDQzF5Z7b8WNigtBQZbKUPa46/v741wUFRUEcp3n4+HDO853XCjv+3HG51w8xC9H/Ii6Go/1\nZXvKXK2IVbAdumQjZxC9XxaWGqo3LSsmyrf3oEKt4qgeSMYrgKHmx/OBt3O44nXga/Pj75HxIX/k\nT0+SJEl6Itevi8AVEQG+vvDvv89W8AqKucaQ1V9zOz2J+mWrM6/7+PvqWBWeD4FAoFxqGr2WXmDS\n9HUMWPYGl+t0wNYItXZ14WcbG4oD041GAs6epW7dunTt2hUfHx9cXXPedkdvNLD+9H6WLNtC8ik9\n1kF2lIopeud9HUZKVXa5E7a8WrhjY2+Z47mEDYiwpQDTEFt35+QTIBXoBjw8FEq5I8OXJEmSlGdx\ncWKo8coVqFcPNm0Cu8LvUMq1c9evMGztNJIyUvF1r83sbh/fU1ahMP0J/Aq4HY+m+bt+LD5/G/e2\nlgTU7obKZKSRQ3NaaksC8J3BwOejRrFx40YWLVpEu3btcgxeeqOBpeu34jf/ECZ/SzTpjmQNnNo4\nWFK7lbs5cFWkhIdzLu90N6J/zgBMBCY9pN1J4HdAC8zK5bmlR5HhS5IkScqTxERo3x7On4fq1WHb\nNijyDO0gcyryEiP9viVVl07LSvWZ8eooLC3+m/0ELwEjb6dT85M9lP/5JLcUKFalCPX+KI1OZaLe\n+W1849uXBKCDXs++ESPYtGkT9vb2ODk9OIE9JTmd+XP+4cAfF1BdsUSFLRqgqKcdLbrXpkHHSlRt\nUgYLreaBYx/tCNAVsQRgOKLXKycKMMr8eBTwH29G/oKS4UuSJEnKtdRU6NwZTp2CihVh5054yAjZ\nU3E0/Bwf+s0kw5BJe88mfN1pBFrNf/NVl2JSGLD0LA3G78QqNg2NhZrGb7rz3vzO7DUdQhV3jpgk\nPSsB28xMHIcOZdP27Tg6OrJ8+XLq1at351yXT0Xx84xNXFp/HVW6GhWWYK1QvVtp3hvbgSr1Sz/8\nRh7rLNARMYz4FjCPh9fIX4UolOEGfJqPa0rZyfAlSZIk5UpGBnTvDocOQZkysGsXlCz5tO/qrgNX\nA/jo39nojHq61mjOF+2HoFH/N3sMhgXeZMzwLbgejACgWvPyjFzQgbiMCNwcnOiR5MWt3e/ToO8R\nALovX86e7dtxcnJixYoV1KpVi/QUHXv+OsuqefuJO5cKgAo1Kg8DLd6tyZAPu2DvmN85a5eBtojt\nuLsjVjY+7GeUzt0VkFOBZ6h78zknw5ckSZL0WCYT9OkDO3ZA8eIieBVmeYi82hV8jPEbfsBgMtK7\ndlsmtXkXdQ5b+RS0tORMln++j/Vzj6EYFTKL2/HWd215682aZCg6TkYEUFPRYaVS823dMVxzKEtd\n4Ld33mFqaCh9+/bFKtOFuUM2sHv5WQypouqsydqIha+B3h+8whtd2hZQiAwH2gAxiAC2gkfHgO/M\nx4WAaJMAACAASURBVNQGBhXA9aUsMnxJkiRJj+XvD+vXg5MTbN8OVZ6hXWW2XDzE5M0LMCom+nt3\nYmyLt3Iop1CwFEXh4OoLLBqzg1vRyShqFddGNmD4Vy1428kaRVE4kH6SGKckDqSfwkVdgTme7+M8\n6kO+Hj8Bo6MrvmW78/Pbh7l66sad82aWT8O+jYrBIzvToXaTAgyQekTR1AjAF1gHWD2ifRQw3fz4\neyCvc8qkR5HhS5IkSXqslBTxd8OGUKvW072X7NYF7mHKtl9QUBjcuDsjfHsXevCKDIrjp5FbOb0z\nFICMhqU4/mMnutQryTBzm2B9OFcNUahNKrytvWidlIHTgLexOn6cKbsDKRPanoxUPQBGGwNp3om4\ndbRhdK/utKpcvxB67S4jyr4WAzYCj1ua+gmQhhiabFnA9yIVWvjy9PRUAwsR/ZWZwHtBQUEh2d4f\nC7wBmIBpQUFB6wrrXiRJkqT82bZN/P0srWr8O2Ab03ctAWBk0z4MbpxTVfaCk5GmZ/W0g6ydcRiD\n3oSDiw3Gb1qxaVBdPNUqfkJMW7+mv87e9BMAlIstysL0NG52ex2r6GC0BnscLnqTYdSTUSGV1Ea3\nKdvamXHN+9Gikvd/MFTqyuO3BToGLAUskaUlCkdh9ny9BlgHBQU18fT0bIwYPO4G4Onp6YSoQVcJ\nEb9PI/pAJUmSpGfM0aPwww+gVsOECU/7boQ/jm9g9r7lAIxt8RZv1e9cqNc7uuEyv3y4jZthCQC0\nHlQH/29as9rVFmtgNWAPGBUThzICMGHCy7IyETHJLOrSBVIisDQ4UDutLbd99dxqGEyVWmX41Od9\nmnvUK/TeutxTgNHmx2MAj6d4Ly+uwgxfTYGtAEFBQUc8PT3rZ3svFbiGCF52iN4vSZIk6Rmj08Gg\nQWLC/fjx4O39tO8INpzffyd4TW4ziN512hbatW6GJfDzh1s5tiEYgAq1i9N/YUfG+5RlP+CI6Dnw\nMrfXqNR0sn2FYH043lbV+ah4CEpmOlYGRyZ4OdO59e+ML/shQ5u8wysedZ+h0JXlb8AfKI4YepQK\nQ2GGL0cgMdtzo6enp0VQUJDB/DwCuICYxTf9/oMlSZKkp2/6dFFMtVIl+OKLp303wj+BewD4uMVb\nhRa89JkG/Gb5s+rrg2SmG7BxsOStqS2pNbw+nS3UXABKAVuAWkCk4SYR+hs0tq6Fk8aBEjFFGJZw\nhGPVmtCg5ACcj93Ep8pK3BxtWd5v6jMYugCSgKyuza8RX+NSYSjM8JUEZN82Xp0teHUESgIVzM+3\neXp6HgoKCjr2qBOePHnyUW8XiP/iGlLeyd/Ls0f+Tp5NBfl7CQmxZurUaoCaceOCuHAhpcDO/aTS\nDZkERAWhQoW7walQ/ju8cjSOTd+eJz5c1Nry6lCK9qOrcqOsPQ1NBmKxxCM9nbnBwej1evbYpHO5\n5A0UtcKt0Fj2HD7Gqh9+J9WnOSxsSJF0I1qTmOCuMsGpU6cK/J4fx8bmMtWrQ3p6BhcuPPgz02pj\nqFRpNLa2EaSleXLxohdiW6EX19P8N6www9choAuwyjznKzDbe7cR1dsyg4KCFE9PzwQePwMQ70Lu\n7z558mShX0PKO/l7efbI38mzqSB/L0YjDB8OBgMMHQrvv+9ZIOfNr13BxzAqJuqU9qRZY98CPXdc\nVBK/fbyDAysvAFCmalGGLehI7VYV2A0MQfQqNAP+sbHBycuLgMxLXM4MRUGB2/Cl32Ki/zqKKTED\nTeg1KkZdo5aNC0FEoULB0lL7H/+/owB+wDgAbGxcc7h+IPA+EAlUxtZ2M97eL/Zcr//i37BHhbvC\nDF/rgLaenp6HEQtA3vX09PwICAkKClrv6enZBjji6elpAg4COwrxXiRJkqQ8mDsXjh0Tley//fZp\n381dh0LPAODrXrvAzmk0mNgw7xjLP9tHeooOK1stb3z2Ct3GNEZrqeEvYACiUtbriHWA1sD29CNc\n1UcCcP7cVfZsOUzi2kBMSZnoatcm6c8/mRMayeoAUcerbIlYILebXheEq8AHwGbz8wbAb/e12Qn0\nRMRKX+BfoOh/dYMvrUILX0FBQSZg6H0vX8r2/ufA54V1fUmSJOnJXL0KkyeLxz/9BI7PyNQfRVE4\nFHoagKYV6hTIOc8fDOfH4VsIC4wBoPFrnrw/px3FyjuhADO5u8HOaMSyfTWQmJ5CcEgEumI6du09\nTvi162RuuIwpKRNNvXrEL1/OZEdHtCuPo880UrG2C0Xs0vhvwlcmokTEVCADsS3QNETfXfZiqb8D\ngwED98ZKqbDJIquSJEnSHYoCgwdDejq8+abYRPtZcSU+khvJ8bjYFqFqcfd8nSsxNpUl43ex83fR\nk1a8ghND53WgQefKd9p8AnxjfvwdMFoxsS/lBBciQ1m+YyupunS0WgvqlvRkyptDue0dxeRff+X4\nr79Syd6eycCcY3EA1G5eKl/3m3t7gOHc7evohwhiJbK1UYAp5j8ghiS/4eF7PEoFTYYvSZIk6Y7F\ni2H3bnB1hTlznvbd3CtryNHHvdYTFyM1Gk1sWxTA0k92k3I7AwtLDa9P9KHXRF+sbLR32qUDMxBz\nZv4COulS+enmRtSOakxuJgwqI43Le9HerS7qBB3eZatxuWw1zrRujaJS8TOiDyn0WDwAtVuUEqOA\nheYmMBb40/y8CqLOeev72ukQ87v+QISteYiwJv2XZPiSJEmSAIiOho8/Fo/nzgU3t6d7P/fL75Bj\nyMnrLBi2meDj0QDUa1+RIfPaU7ryg3OcjIgClHZAF1M6i2PXo3W0ID0jk6jAWH7pMRmL23r69u1L\nYmIiLi4ufN6sGZkqFQMQG/Kk3E4n+mIiFpYaajQpVkjhywT8AkwCEhCRbzKiN+v+vRsTEfO7dgG2\niJpeXQrjpqTHkOFLkiRJQlHE6sbERHj1Vejb92nf0b3SdBmcirqEChWNy3s9/oBsUm6ns+zTvWz+\n8QSKAkVLOzB4Tjt8e1Z7bL0tG0XHqqTtaO0tSExMoammLr5t6hAQEEDvfv1ITEykdevWXGrYkL2I\nzXuyNuQ5uycMRYFqPmWwttM+/CJPLAAxtTqrSlMHYD5QMYe2EUAn4Bxif8dNQP0c2kn/BRm+JEmS\nJNasgX//BQcH+PFHeNZqgB6LOI/eaMCrZCWcbXO3AkBRFHYvO8visTtJjE1DY6Gm25hGvPFZM2zs\nLXN1jnSVJZHXYjBYG7C9YYNv8zocP36c/v37k5KSQseOHfly4UK8LMX5vufuWsHTO64AUKdqJKx7\nNa8f+RGSgM8QQ4YmRLnXHxC9Wjn94k4DnYFooCpi9WOFHNpJ/xUZviRJkl5y8fEwcqR4PHOmKC/x\nrDmcxxITYedi+HH4Fs4fCAegZrNyDFvYkfI1iuXq+Ah9NKXUdugiw1mzYxcO1nasH/g9ADt27CAl\nJYUuXbowb948Bmq13ALaKib6RR+ByH0QsZfTa6sDLtTW/gixoiQF7h3y9LnvpSB2kRyDCFJqxBrM\nL7m3pnn29isQqxxTMFco478tdyHlRIYvSZKkl9xHH0FMDDRvLlY6PmsUReGgeb6Xr8ej53ulJWey\nYsp+/p1zFJNRwamYHQNntaFlf69cb+lzVR/J/rQjjFZZsuLocYzAgPqv4qC1AWDSpElUqlSJHt1e\nZU9CEH+61cTamMmPy7xRxZ8HIOZ2EaJjm2JrraNyp05QvgWUbQ72T7rqMQQYCWwzP28E/AQ87Oex\nF1Ek47j5+RvAEh6cByY9DTJ8SZIkvcS2boWlS8HaGhYtAvUzWG0g9FY0UYkxONk4UKN4TvOZzDXA\n1lxk0ZjtxEclo1ar6DyiPm9NbYm9U+5rV4XoItiVfhQFhZNJWozhwbjZO1M2xZ4WzZvz53djqMAl\neqv2kfbrMIb2E/Otvjj0GRXjz4NLVSjbgjPH6gAxlGtUFk2Hr/Lx6TOBbxF1ujIRm8F8g6jPldMv\n6zxif8ZN5uclED1jgx7SXnoaZPiSJEl6SSUnw5Ah4vGXX0Llyo9u/19SFIWAqCDWBe5he9ARAJqU\n90KTQzqMuhzPTyO3ErBDLCes0rAUwxd2opJ3yTxdM0QXzq70YygoVNVWYe6mRTQmBt/YZIZOexed\nUWH1zEGM900m1cKWwe0WcdWpIrWSI/moeH0Yeh3sRD2t00vWATF4NMxPtfhdiDIQl83P30aUfc1p\n6DQKUbd8CWIemD2i5+sjxJpN6Vkiw5ckSdJL6pNPIDwcvL1hzJinfTdCbMptNpzfzz/n9nLt9vU7\nr3uXqcaoZm/e0zYzXc+qaQdZO8Mfg86IvbM1A75pTbv36qJW523FgKIoBOpCUFCoY1WNL86FMSl5\nB0VDwhm2yRmDScWguimM61qOYzUG0L/2EIItHbACFjmUQev5+j3nOrMzFACPRq5P8FO4gQhNK8zP\nqyFqdrXIoW0ioiLZ94jqZBbAMMSE/NzNb5P+ezJ8SZIkvYQOHoQFC8DCAn77Tfz9tOiNBg5cDWBd\n4B4OhZ7GqJgAcLNzpmvNZnSr2YLyzvf2Yh3beJmfP9zGzdAEANoOrMOAb1pRxO3JenlUKhWd7Jpy\nTh/JOFMxbh75HkLjGLrRGaOiYmif1kz86ju+tnPjC0QdsJqIAqz3F764di6GhJhUXEo54Fo+L/dj\nRMzjmowIVdaIEPUxcP/qTJ257ZdAvPm1XojhyWeoC1PKkQxfkiRJL5mMDHjvPVHba+JEqF1we1Tn\nydX4KP4J3MOGCwe4lZYIgIVaQ+tKDXmtZgt8KtTGQq2555iYawn8MmobR/4VQ3HutYoxfGEnqvuW\nfaJ7iNDf5LwuhNa2jYhTWTLQ0oOYfcspm5FIx5KJ/OHqQts+H9Jr4mRaqFQcNh83BhFzcppNdmZX\nGAB12lTI9SR/OIHosTphft4ZUUri/pIQWSseJ3G3amtTxHBk41xeS3raZPiSJEl6yUybBkFBULUq\nfPrpf3vtVF0624L8+SdwD2eig++87lG0DN29WtK5WlOK2hV54Di9zsi67/xZ+dUBMtMN2DhY0v/L\nFrw6sgEaiyebSH7DEM+2tEMYMLJXF8ogq8rcSIrDY81yPEpn4mpj5N/hrqx5/1PqAsmIilp/AG0e\ncd7T5iHHOm0qAPrH3EUi8CmwABGsygBzgdd4sGbXXu5dwVgVMRm/Sw5tpWeZDF+SJEkvkfBwUcsL\nxOpGq/+g8sD9k+czDJkA2Fna0KFqE17zaolXiUoP7SU6vSuUn0ZsITJIDK8161uDQd+1pWipnGpb\n5c4tYyJb0g5gwEgRbXn6WFbiNlB16gSS/M5wu5oLca85M6zLEtaYj+kJ/MzdIqo5MeiNBO4NA6B2\n6wqEXb/8kJYKYnufjxBzvDSIml1fICbLZ5fTCsYpwEDk1/jzSf7WJEmSXiITJ4phx759oWnTwr3W\noybPv1azBW2qNMLW8uFlIOKjk/nt4x3s/1vUzirjWZShCzpSp3X+qrMnmVLYmLqfTEWP1qIUA23q\nk6FS0WDODKKWbwWgaqMa1H5nK9EOpbFHDAC+w+P7l4KORpGRqqdsNVeKlnIg7HpOrS4DI4Cd5uc+\nwI9ArfvayRWMLyoZviRJkl4SR47AihWiptc33xTedTINOmbuWYrf2d25mjx/P6PBxMb5x/nzs72k\nJ+uwsrGgz/9eoftHjdFa5f9rK82UiUExYtS48aFtY/QqNe0WLuTczB8AcB07krljJgHgk5nAMisn\nPHJ57nuHHO+XAUxH1OnSAS6IlYrvcm8NLrmC8UUnw5ckSdJLQFHulpP46CMoX75wrnMz+RYf/fsd\n525cwUKtoVWlBnSv2TLHyfM5uXAogoXDNxN2NgaAxt2qMHhOe4q7O+X73m4Zk3BWO1DMoijB9i2Z\no7ZFr9LwaXQ0f83+DlSQ9tkkzr4/Eo3JwOf+U5jk/VGevijP7HpY+NqO6O0KMT9/FzFfyy1bm6wV\njF8BcebXeiKm9lfJ02eVnm0yfEmSJL0E/v5b9HyVKCGGHgvDqchLjF3/PfFpiZRydGV2t4+pVjx3\nQ4SJsaksmbCLnUvEHo7F3Z0YMq89DV/Nf+hQFIULuqscygignrUXs6w8Wakpghb4E+hTsgR+nwzj\nhl1p0vq8QaWM2yxf256GxkzwzX11+rTkTIKORKHWqKjZXKRbrTYW6AOsMreqgRhifCX7HZLzCsYZ\nQJMn/+DSM0uGL0mSpBdcejpMmCAeT50KDk8+Tz1HiqKw+swOvt39BwaTkYblajDj1VE42zo+9liT\nSWHbolP8MWk3KbczsLDU0GuCD69P8sXKRpvvezMqRg6mB3BRL3qklpkyWQk4Kgp9pk8ntlQpGvbq\nxoWBYwEYqEvhh99rYJ96HXpsztO1zu8Px2gwUbVJGeyKWABzqVHjEyAVsEXM3xoDZP9c+4BxyBWM\nLxcZviRJkl5w338PERFQpw4MGFCw59YZ9EzbtZh1gXsA6O/diTHN++VqiDHk5HUWDt/M5WPRANRt\n58HQeR0oXSU/W/LclWbKYFvaYW4a41GjZqtNfdZZlqeUotDt88/597ff2KjVEtOmDRq1hnGx4UyP\nWgOp16FCR/EnD7Lme7V6Ox1oCASg0QB0RZSPyD7Wex6YCGw0P5crGF8m8jcsSZL0ArtxA6ZPF49n\nz8YcBgrGzeRbfLx+NoHXQ7Cy0PJ5uyF0rv74JZQpCRks+3QPmxeeQFGgaGkHBn/fDt9e1fJQlPTx\nbhjjuGmMx1Jlw0I7X05qnKlmMtFs8mT+XboURavl9s8/Y6uPo+mGxUx9/VPY6ycOrjcqz9cLOnqO\nYQs20HHICcRQYjlCQkZRqdJH2VrJFYySDF+SJEkvtE8/hZQU6NYNWrYsuPNeigljxNpviEtNoKSD\nK9+/9vj5XYqisOfPQBaP3UlCTCpqjYrXRjfijc+bYetQcAXHUk3p2Klt8NCWoYRNfUZYlCRabU1T\nRaH0+PFsWLECxcqKlEWLKGVxA7u1S/ikxwQ0SaEQFwiWjlA2Lz8shdSE35i87jOci6eiKBaIMPUZ\niYmXzG0etoLxf0DxAvvs0vNBhi9JkqQX1OnTsHix2LdxxoyCO2+aLoNx6+cQl5pAg7I1mNFlFC6P\nmd8Vdi6Gn0Zs4dz+cACqNy3L8IUdcfcquOChKApndJc5nnGOV+2asc/CjbctK6BHrBn0VKmYX6wY\nttbWuCxZwls2iWzdt4GG5WrgW6EOnPxenKhCJ9Dcv5fiw1wChmPntAc74No5T8rXXIPY+RFUKj2i\nStiXyBWMUhYZviRJkl5AiiJKSigKjBwJVQrwe37Gnj8IT7hBFbdyLOw5EUuLR0+MP+x3kW/7+GE0\nmCjiZsvAWW1o9VatAh1iNComDqSf5JI+DIBVxtuMthBlHN7W6zkXGcnaChVQjRtHv969GedsR+/F\nYjhwTLN+4l5C1omTVe6eiyumA18jerP0pCc78POHrSlXcyzla9YkawVj9eofA5HmY+QKRkmQ4UuS\nJOkFtH497NkDLi7w2WcFd96dl4+xLnAPVhZapnf+4LHBy6A38tvYnRgNJtoNqsPAmW2wd7YpuBtC\n9HjtSjvKVUMkGjRcs23I19oyAPTU6dgyYgRqf3/KrFnDsqpVaeHuzuTNC9AZ9XSs6kP1Eh6QFgNR\nh0SPl3uHx1xxMzASCDVffxDjfD0ICzQyN6Ai2VcwWlsDeCJWMHZFrmCU4N6SupIkSdILQKeDsaJy\nAl98Ac7OBXPem8m3+HL7LwCMbtaPSq5lH3vMvhXnuBmaQOkqLoz4uXOBBy+AIH0YVw2RaLHA364F\nX2vLoAXqZ2ay5/330W7ejKXRyPL0dFog5qttunAQC7WGkU37ipOErAcUKNcarB42hBoJ9AI6I4KX\nF3CQ6OBvCQs04uhqiXutwUALROmIEly79glwDuiGDF5SFhm+JEmSXjALFkBICFStCkOHFsw5TYqJ\n/21ZSGJGCr7utXmjbvvHHmM0mlg97RAAvT9pikZTOF85lbXl8bD0YLttY5ZYuGALOKSnc3XgQKx3\n7MDWyYlNq1bRrG5dAObs/wsFhb5121PGybxdz5V/xN+VchpyNCAmylcD1iJWJc4CTgK+nNkVAEDt\n1qdQqzciVjB+CYQQF9cDOcgk3U/+FyFJkvQCiY+HL78Uj2fNAm3+65QC8OfJLRwNP4ezjQNfdhyW\nq/lah9ZcJDIonuLuTjR/s2bB3Eg2CcZkrFWW3FRbMdrGm4uI2JMCWMyahf3evTgVLcrqv/+mevXq\nABwOO4N/2FkcrGwZ3NgctHTJcG0HoIKKXe+7ij9iVeIZ8/PuwA9AWbJWMJ7eGQZUpU6bUGA4Yg9G\nuYJRejgZviRJkl4gU6ZAQgK0bQudOhXMOYNirjH3wAoAvugwFFe7x++zaDIprJx6AIBeE32w0BZg\ngTEg3ZTJptT96FQqZto156LaDivMwQv4ePRoQkNDmTRxIlXMqw1Miok5+/4CYGDDbjjZmEv9h24F\now5K+YJdVmi6hdju5xfzc3fEqsVXEXswihWMRmM8Z3aL/Zpqt/4JqFegn1N6McnwJUmS9IIIC7Ni\n4UJQq0VB1YJYTJih1zFx0zz0RgOv125Di4reuTru6Pogrp2LpWhpB9oMqJ3/G8nGqJjYkeZPspJG\nhNqZEJU1KkCXlET5WbP4a+JEfBwcRJ2NbDZdOEhQ7DWKO7jwZr1s1etDsoYcX0OsUlyKmDAfi9gK\naCzwKWCD2KPxE+AKAFdOdSI1wYYSHs6UqCCDl5Q7MnxJkiS9IObMKYPRCEOGQM0CGuWbs/8vrsZH\n4u5Sio9bvJWrYxRFYeXUgwD0HO+D1qrgvmoUReFA+imijbEkqqxZYOuLXqVBlZBAtX79SDh9mlVJ\nSfjMmXPPcZkGHQsOis2tR/r2wVprruNl1EHoJvG4Sg2gJWK1IkBzYCFQ3fzaeOCY+T2xgvH0zqLA\nHuq0yd0G4pIEcsK9JEnSC2HHDjh40AkHh7tzvvLr4NUAVgRsxUKt4ZvOH2CjzV0V+lPbrhBy8jpO\nxexoP7huwdyM2fHMC1zSh6JDzc+2PiSqbSh26xYNe/cm4fRpypUrx9ispZ7Z/B2wjevJcVRxK0fn\n6q/cfSNiLxgToZ0rFOmKCFluwB/AHkRPWBfECsZjiD0YfyZrBeOZXWEAMnxJeSJ7viRJkp5zBoMo\nqAoweTIUK5b/c8anJvLZ1p8AGNm0z2O3DsqiKAp/fyXmenUf2xgrmwKa8Q8YgSWWHrjoo1lrU5tQ\ni6J0iI0lrW9fQi5dokKFCqxatYpSpUrdc1xiegqLjoihxdHN3kSjVmfdLNz8AgYARbKqzw9BVJ9P\nBwZz7x6M4xDbBtkDkJmu58LBcFQq8GrpXmCfU3rxyfAlSZL0nPvtNzh3DkqVymTUqPzvkagoCl9s\n+5n4tETql63O2/VfzfWxgXuvcfFwJA4uNnQaVj/f9wJijtdpXQhjLCtxQG2Dyr4NWpWKnwDfmzd5\nPTqaypUrs3LlSooXf3CV4a9H15GcmUqjcjXxcc+afxYOCZ2g0Xnx1FAVLJYghhhnALO5uwfjUHJa\nwXjxUAT6TCMV65agiKttgXxW6eUgw5ckSdJzLDER/vc/8fjDDyOxtq6Y73OuPrOT/VdP4WBlx9cd\nh9/tKcqFrF6vbmMaYWOf2/0RH86oGNmQ5s8Nw3XKmJLBxhtPlYrlycnUc3CAmjVZuXIlpUqVwtXV\n9YHjoxJjWBGwDYAxzfuhUhmAH8D0P3DOEAsXE98Ft/nAb4ghxsfvwZiWlMn2304DUFsOOUp5JMOX\nJEnSc2zaNIiNhaZNoXXrhHyfLzQ+iu/2LgPgf23fo4Tjg4HmYS4fj+bsnjDsiljR5YMG+b6XVFM6\na9KOkm6MJVWl5ZClB0OA8RER9Ovdm379+jFy5Ehq1ar10HMsOLQKvdFAp2pNqVY8GrHFzzkx4zkI\nSBkD3o2BWmStYARfYCY57cF463oy6+ceY8uPJ0lNzESlgibdq+b7s0ovFxm+JEmSnlOhoZC1qK8g\nSkvojQYmbppHhkFHl+rNaF81bxtAXzsXA0CDVytjV8T6ie/DBGzR3+RK+hEsFR1JKit+tGvGDxon\n6oSF0bt3b6KiotiyZQuDBw/GyirnodaLN0PZdOEgrnZGPm3rD3wg3kiygB0GsGwOrx5EVK+HR+3B\nGHU5Hr9Z/uz64ywGnRGAms3L02dyU6o1KfPEn1V6OcnwJUmS9JyaMEHs4/jWW9CgAZw8mb/zzT+4\nkksxYZQuUoyJrQfk+fiMVD3AExVUVYAA4C/gb0CltmSCoueCRXG22DRgv9qGzJAQevbpw40bN6hf\nvz7Lli17aPBSFIU5+5bxWs2rTGh1EVvLFFC0cLEM7AgFZ0fous+csUoAU4CB3P+1GHQsirUzDuPv\ndwlFwdzT5UmvCb54Niqd588pSSDDlyRJ0nPp4EFYvRpsbMTQY34dDz/PH8c3olapmNZpBPZWuZ9A\nfuFQBH4zD3N0/WUArOxyv8IxBFiBCF3JxgS89NeJsq4GGmdm2rWilsaZ8yoVaQkJdH39dWJiYmjc\nuDF//PEH9vb2Dz3v6Sg/hvjMo16ZrPlbLeGYCxxcK2qlvpYE2gdXMIIIbqe2XWHNt4cJ3HsNAAtL\nDa3erkWPsY0p45n7oVhJyokMX5IkSc8ZkwnGjBGPx4+HMvkc9UrKSGHylgUoKAxp3JM6pT0fe4zR\naOLo+sv4zfTnkn8kAForEVDe+KzZI4+9gagTv5y7JUvr6iMZnnYULSYiNE6c15bkTQsXpiE6pyyd\nnHj//ffZu3cvS5Yswdb2YeEwFZPyBV6lZmOhNpGuL4KNdhYErhPBSw10VYPjgysYjQYT+1eeZ+2M\nw4SdFUOoto5WdBxaj66jGlG0lMNjfy6SlBsyfEmSJD1nli+HEyegVCkYNy5/51IUha92/MrN5FvU\nKlmZwU16PLJ9Zrqe3UvPsu67I0QH3wLA3tmazsPr8+oHDXAunnNvVCKwDtHDtQsxrwvAXlEYzEpS\nHQAAIABJREFUobuMe8ZZAA5r3Qm2cGM+MAIIDAwkMzOT+vXrM2zYMAYPHoyFxcO+uv4FPkCtigBg\n0/nqtKcqRI6EC5miSZt6UGYF2VcwZqTq2LH4NOu+O0LMtUQAXEra03V0IzoOqZev+WuSlBMZviRJ\nkp4jqakwaZJ4PG0a2Nnl73wbLxxge9ARbLXWTOs8Agt1zvO1kuLT2LTwBBvnHScxNg2A4u5OdBvT\niLYD6+RYViID2IwIXBsBc/xBiyjo8CbglhFAkE6sMvSz9uKApScrVSpeA06dOkW/fv1QFIUNGzZQ\nuXLlhwSva4jJ9BsACI51ZculEgyxisDiwIW7zer1Bq+Vd54mxqWxcf5xNs0/TlJ8OgClq7jQY5wP\nrd7yKtBtkSQpO/lfliRJ0nNk1iyIigJvbzHRPj8iE24ybafYfHpi6wGUdSrxQJsbobf5Z/ZRdiw+\nTWaamFBfsV4JeoxrQtNe1dFY3FsDzAjsRQSutYgeLxBDhy0Qgasn4GJ+3V/tiAE1i20bcU1bhl2A\nD3D8+HH69+9PSkoKnTp1onz58jl8Ah1ipeIUREFUB87faI5JOcSH5c6JGwBo/DqU6A8VOgNwMyyB\nf2YfYfuvAWSmGwDwbFSanhN8aNzNE7W6AHYkl6RHkOFLkiTpOREVBTNmiMfffw95qH36AIPJyCeb\n55Omz6BdlcZ0rdH8nveDT0TjN9OfQ2suYjIpAHh3qEiPcU2o1dIdVQ51LdYAo4GobK/VQwSuPkDW\n1LSbhluEmFJRWZblbatK3NKWxFFtx2FEsYfDhw/zzjvvkJaWRrdu3Zg7d24OPV77gWFAVs9WB/TG\nDGqU2Ai3QVmrQqUo0Ggi+E4H4OqZG6yd4c+BlecxGcVnqt+pEr0m+FDjlXI5fiZJKgwyfEmSJD0n\nJk+GtDTo2RNeeeXx7R/l1yPrOBMdTDF7Fz5t+x4qlQpFUTixJQS/mf53VvlpLNS06u9Fj7GNcfd6\ncOsegCRgFPC7+XlFROB6E8heflSn6DmWcY5zuhDUWDDDoihX1LbUVduxGVHwAeC3334jLS2NXr16\nMXv2bDSa7EOhscD4bFdzN19lO1qNidRkDaaVVjhkpkHFrig+UwncE8aabw9zapsY3lRrVLTs70XP\n8U0e+pkkqTDJ8CVJkvQcOHkS/vgDLC3v9n49qTPRl/nF3w8VKqZ2Go6txoadv5/Bb5Y/4edjAbBx\nsKTjEG+6jmqIaxnHh57rEPAWEApYA98h+qOy9yEpikKoIYqD6QGkKRmAip2WFQlXWdIeWA04mNup\nVCrmz5/PsmXLeO+991Df6d4zIbb/mQDcBiyBRsBxIAxF0bDmdCXK7k2msekmRuea+Gd8ztrGvxN8\nPBoAK1st7QfX5bUxjShW3il/P0RJygcZviRJkp5xinK3tMSoUeDh8eTnStWl88mm+RgVE/1qdSTi\n7yQW/DCPW9HJALiUcqDb6IZ0eP/Rq/z0wJeInQ9NQF1E6YhqObSNMd5ie5q/eKJxYZqNNxEaJ94B\nFiEm4G/bto0ff/yRpUuX4ujoyPvvv5/tDGcRm1ubz4EnEAMcMD/vwex91XA5sZZ6hni2nmmG3/Hu\nRF/ZBICjqy1dPmhA5xH1cSwqN8CWnj4ZviRJkp5xfn5w4AC4uYmhx/z4ZtfvRCbG4KYvytE+0exP\nCAegfE03eoxtQrM3aqK1fHSF+mCgP6JGlwrRF/Uloi8qJ8UtiuKpdee4xpmJlhVRVCo+NR+jAjZu\n3MiIESMwGAysWbOGgQMHmo9MBr4AfkBM5XcyXyXI/L4vMIMLN0oQc+Btyvq7MfBgVxKS7YFkirs7\n0X1sY9q8Wwdr29wXfpWkwibDlyRJ0jMsM1MUUgX48ksoUiTv51AUhbjUBFbt3cn6S/tQ6VVo5jiT\nmWCgVkt3eoxrgneHio+dcK4AvyIm1acBZYFlQPOc2ioKRzLO4mnpjoOmCH/aNmAxosbpj8AQc7t1\n69YxatQojEYjI0aM4N133zVfyQ8xkyzKfFQxRG8XZN+DMS4yie8HT+X2nhaczxRbDXnUKU7PCT45\nrsaUpGeBDF+SJEnPsLlz4epVqFED3nvv8e3T9ZlciYvkctw1gmPDCY4N52J0GCnGtDttnDeWoFXL\nuvQY24TK9Uvl6j5igcGIMqYAbwALEX1ROTmaGcgZ3WWC9RGscujIJpUGG2AlosYXwD///MOHH36I\nyWRizJgxfPzxx6hUYcBIRIUwgCKIghUxiGr0U4BBRFy8zdqZG9j951lMejEnrVaddHp9O4i6bT3k\nykXpmSbDlyRJ0jMqJgamThWPZ8+G7NUWTIqJyIQYguNEwLocG865yGBi9iSgoDxwLnWaBssYa+oW\nqcbnywZS0sPlgTYPsxV4F7EtUBFE6HrzEe0DM4M5nRmEChV+NvXZpNLgiii02ihbu+rVq+Ps7Myg\nQYMYNWoYMB34ClGeVQsYEMHLDrHC8SMuHr7Nmm/X3tlHEpVC01rn6dr6ONVnBoFGDi9Kzz4ZviRJ\nkp5RU6ZAUhK065KMc9Vw/joVYQ5a17gSH0m6PvOBYzRqDa6KM/ogMAar0d6wxlXnwmvv+tB5St4m\nnKcjIs988/NmwFIgp3KnWa7oIzmUcRqAjTb12agtgQciwFU2tzl27BgNGjSgSpUq7N27FxeXs0Bt\n4FLWp0BM6dcA72MyfcaJzUms+XYNFw6KrYO0VhoqdXPjYoUtTHRbh+LoLoOX9NyQ4UuSJOkZoTca\nCL0VxWXzcOFedTjVPg3nRpHbvLfqwfbF7F2o7FqWym7lKG1ZnIBlwVxdlURqvA5LxFY53T9uQsu3\nvLCyyVswCQD6ARcRfVBfAWMRcehhFEUhMDMYgB1WNdlo6U59RI9XVjWtX375hSlTpjBy5EgmTRqE\ni8tYxMwxzGc3mv/0QK/7iv0r9Kyd+c+dEhh2TtZ0Hu5NxxHe9N/wCS7JKQCorHPfkydJT5sMX5Ik\nSf8xRVGISbnFZfNwYUic+DvsVjQGk/FOOwtzF5OVxooqxcpS2bUcld3K3QlcTjYORAbFse67I6xY\negx9pji2mk8ZeoxrQqOued8qx4io1fUpou+pKqKERL1HHGNSFHSKHmu1Jdi9wmrdNXZbetAJMccr\na6vt+fPnM326qDZfuvRVxMT5BMSaR8V8dV/SkqexbZEl/36/g7jIJACKlnbgtY8a035wXWwdrNhy\n6TA3kuNp7OAAKYC1rNslPT9k+JIkSSpEeqOBizdDuRx7zRy0IrgcG05yZuoDbVWoKOdcgsqu5aji\nVo5Zk8sREViOSyeKUcH93lV7Fw5FMH/mZo6uv4yigEoFVZsXZ+DXHanuW/aJ7jUceAexNyPAcGAm\n8KiBygyTjl3pR8hQdMTatWSEygKTVUUGAT8hvmQURWHOnDnMmjULlUrFrFlO9O37a7azKIAnCTFT\n2TDXhY0LjpKakAFAuequ9Bzv80AJjL9ObgGgvXt1OOcHVjJ8Sc8PGb4kSZIKyb4rJ5m5ZykRCTcf\neM/JxsHckyV6saq4lsejaGlsLe8WNp1+FXTxkNV5ZTSaOLr+Mn4z/bnkHwmIuU+t3q5F948bczPl\nGtW9nyx4/Y0oY5qIKOqwBOj0mGPijAlsSz1MspKKUWXJNFMKJk0RvgA+Q0HFXGAugYEZzJoFarXC\nnDm36dkz++6Pxbh+5TP8ZpVn1++B6DLERtfVfcvSc4IPDTpXfqD3LvB6CGevB+NgZYe3W0nxopXz\nE31uSXoaZPiSJEkqYGG3opm1ZxkHQgMAKOXoRr0yVe8OG7qVxc3OOdflEDIz9Gz5+SzrvjtCdPAt\nAOydrek8vD6vftAA5+JiYO/myWt5vtdEYARiaBFEGYhfEQHsUUJ0EexNP44BIykaZ6bZ+pCktuVX\nYBCpwHuISAe1asE339ji6GiiW7cM8xlsCTk5gTXfVuHw2hBMJvG5GnWtQs/xPo/svfvr1FYAeni1\nxMogtg6SPV/S80SGL0mSpAKSqktn0ZF1LDuxCYPJiL2lDUN9etG3bnu0mrz/c6uYTICa8T5LMNwS\nvWfF3Z3oNqYRbQfWwcb+YTXlH+8y8BciaEUhhha/R9TyelwkDFYM7Mg4ixoj/lp3VtjUQ6vSsB7o\nxBWgB4pylqlTi9KiRXleeeUUb72VBtRBUaZxeqcLa769xJld4UAwFlo1rd6pRY+xTShX3e2R145N\nuc32IH/UKhV96raDs3PEG9ay50t6fsjwJUmSlE+KorD54iG+37+c2JTbAHSr2YJRr/SlqF3ee2Ru\nhN7mn9lHiYtqAhQh6VYGXvVK0GNck3xVbY9GTID/CziR7fX6iJ6vKg85TkHsrviPYmAdas6oLChj\n50MlQxz+lpVor1IxBajHVuANTKYEJk0qx59/Gvj772j8/VXY283jwJqm+H17jCsB4uo29pZ0GFKP\nbqMbPXLz7uxWnd6BwWSkVeUGlC5SDDITxBuy50t6jsjwJUmSlA8Xb4byze7fOR0l9husUaIik1q/\ni1fJSnk+V/CJaPxm+nNozUV0Jgv05o17Jqx4jXZ9yj5R1fYExEY9y4E9cKf8qgPQE1EstRUPlpAw\nIraxXgf8AySaUhmWepAqFsUItanLKxpnumuc2QDYoyAKpH6K0agwbpwXK1fGY21tzbx50Tg6KrxX\nycCNK+sBcCpuR9dRDek01Bt7Z5tcfxadQc/qMzsB6Fevo3gxQ4RdudpRep7I8CVJkvQEbqclMf/Q\nKtae2YWCgottEUa98gZdazZDrcp9z5SiKJzYEoLfTH8C94o5WxoLNemer2O4aEOtWtC2dznykrvS\ngU2IHq5NgM78uiXQGRG4OgP3xx4dsBsRuP4FspYJVDbE8knaYWwVHe4GhcWKHntVVt2wJGAAsA6D\nAcaMaYafXwjW1ta80+ZjWjQfCUBMaAIlK7nSY2xjWr9TG0vrvH/9bL10mNvpSXi6lce7TDXx4p2e\nLznsKD0/ZPiSJEnKA4PJyJozO1lwaBVJGaloVGreqNeRoT69cLDKffV4RVE4sOoCf3914E4BURsH\nSzoO8aaEbxNe7WGHRgOLF4M6F1nOABxxcGA+oqcryfy6CmiNCFw9eHAvxhRgCyJwbcp2HIAH8E7m\nFYplBAAK5SxK0Nq2MVZ3glcQ8BqiMr0jirKUhHg/tJoIPOI6ceyXeJSFouX4lT1o0r0GGs2TDZkq\nisLyU6K8xBv1OtztBcw093zJYUfpOSLDlyRJUi6diLjAt7t/53JsOACNytVkQqsBVHQtk6fzhF+I\n5aeRWzm7JwwAl1IOdBvdkA7v10NjZU2dOqAoMH48eHs//DwKcAzRw7USuFnl7qyt+ojA1Qe4f+vs\nOGA9InDtALJvUlQL6G7+U96YwqqMAEwo1Lb0pJG1F+o7XXD/Am8Byeh01bhwbB67fkkmYXVFKuKM\nlaEo9dpXNJeJUGjaqzrwZMELICAqiEsxYTjbONCpmu/dNzLMPV9ywr30HJHhS5Ik6TFuJscze99y\ntl46DEApR1c+bvE2rSs3yNM8rPQUHSu+3M+/3x/FaDDhWNSGt6e1ovWA2ncKiE6aBEFBULUqfPZZ\nzue5hAhcfwFXsr1eLiODgdbWvMGDk+fDEXO31gH7AZP5dRXgy93A5QEYFSMalQY09jSz8UaFCk9L\nd/MRJuAL4CsUBQJ29Wf4B9HExo+mcnwXLLGlQ59m9Bzvg0edEoiAln9/mXu9etVug5VFtlWecsK9\n9ByS4UuSJOkhMg06lp3YxKIj/5BhyMTKQsvAhq8xoEEXrLW5L/OgKAqH1lxk0ZjtxEclo1JBhyH1\neGdaKxxc7s68OnECZs4U1eoXLwbru/VWiURUzfoLse9ilpJAX0QvF+fPU9/cVaYg9mVcZ/5zMtsx\nWqAdImx1BUpke++WMZGtaYdoZO1FRW1ZqlpWyPbubaA/JtMWjvxbnZXfvM62kC0kW0diobGmST8P\nhkzpSYkKBdsLdT0pjt3Bx7FQa3i9dtu7byiKHHaUnksyfEmSJN1HURT2XT3FzN1/EJkYA0Cbyg35\nuMVblCry6DpU94sMiuPnD7YRsOMqAJXrl2TYwk5UaXDvYKBOBwMHgtEIo0dDkyZwC1iLCFz7uLtS\nsQh3Vyq24O5KxePAUe4GrsvZzm8HdEQErs7mc9z/mS/qQzmcfhoDRs5lhuBhUSZbz9459Jk92POn\nA34zR3HtsgNXXTaRYh2NvbUjfy5bTgOfR+0A+eRWnt6OUTHRwdOH4g7ZNtA2ZIBRBxor0OZ+1aQk\nPW0yfEmSJGUTdiuaGbv/4FDYGQA8ipZhYqt3aFTeK0/nyUjVsWraQfxm+mPQm7B3tuad6a1o917d\nHCedT58OgYHgXh3qToduiInwevP7Vojq828iQlRWp5gesRejH7DGy4uYbOcsiujZ6g604cHVjVky\nFT37009wRS+2LPLUutPUpu6d4JWauIItPy9i/Zxu3Lou6nHFlN5PihKNm5sbq1evpnLlynn6+eRW\nuj6TtWd3AdnKS9y5cTnkKD2fZPiSJElCVKf/xd+PP09uxmAy4mBlyzCfXvSu0y5P1ekVReHIv0H8\nMmo7seGJALQbVId3preiiJtdjscEBMJXx4E/4Oab8I75cmqgLSJwdedub1Uad+dvbUAMBgJgaUlZ\n7s7fakru/pEP1UdxRR+JFgtesalHFcvyANy6nsD6H2ax+UeFtKSWALh7udJzvC/lm/Rj1OhRzJkz\nh4oVK+b655NXmy4cICkjlZolKlKr1H0BTw45Ss8pGb4kSXqp3UpLYuflI/zs70dcagIqVHT3askH\nTftS1O7+wblHu37lFj99sI2TW0IA8KhTnGELO1GtyYOrIRXgCPCnCX4pDcaN4vV0oCHQD+jN3flY\nt4FliMC11dwuSzVE2Kp68SL9q1V77PZAACZF4ZYpEVeNE57a8iSakqmqrUARjT2RQXH4zdrL7qXn\nMeisAKjZXEP7Ee05f9Of5m/WRK1Ws379+icq/JpbiqLc2cfxzft7vUCudJSeWzJ8SZL00olPTWRb\nkD+7go9xKvIiJkXMpvIqWYmJrd6lZsm89eRkputZ880h1nx7GH2mEbsiVvSf2pJOw7wfGGI8z92V\nimEgurdcwCIEJpSBAdaQVRs/GvgRMaS4F1HLK0tD7vZweZpfO5mWlqvglWRKYXfaMeKNibzu0BZH\ntT2NrL0IOhrF/G83c+SfIBRFTPz36XGFnuN74VbZhz59+nDhwgX0ej1jxowp1OAFcCz8HFfiI3Gz\nc6adZ+MHG8hhR+k5JcOXJEkvDUVR2HLpENN2LiE5MxUAC7UGH3cvOldvSoeqPnmqTg9wbONlfv5w\nGzdDRRBo9XYt3p3RGufi9nfahCNWKi5H7JGYpZge4ueDcRmsnwYdK0EwMAPRw3UkW1sNolhqd8R8\nsLxVFhPun1Rvq7Im1ZhO0NbrrJ3hz7l9osK+haWBVm+fpsfYZMp4LiU21opevXoRFBSEh4cHffv2\nfYKr591yc6/X63Xa5Dz0mzXsKHu+pOeMDF+SJL0UbqclMW3nYrZfFpGmUbmavObVglc86uWpMn2W\nG6G3+WXUNo5tCAbA3asYQxd0pOYr5e60SUEMH67PdpwT0Avoa4T/tYKYg9BlABzuAOMQPWNZrIH2\niMD1KmIC/ZMyKEZ2pPlzzXAdgAqa0mg22fD1136EBYpp+raOJjoNO0TXUf64lOwLrOLGjdv06dOL\nkJAQqlSpwsqVKylWrFg+7iR3IhJusP/KKbQaC3rVan33DUWBmNNwYSlc+ku8JrcWkp4zMnxJkvTC\n23/lFFO2/0JcagK2WmvGtXqb7jVbPtGwmS7DwNqZh1k97RC6DAM2Dpb0/7IFr45sgMbibq9ZBqKH\najciRHVFTJzvgPiHd8wC8D8ImhKwYbaYOA8inL2KCFztESUi8kNRFFQqFRYqDRYqCyzRYrvPkb8/\nOHFnQYBLSVu6jj5NxyGrsSuiAPOBwQCcPXuWq1evUq1aNf7++29cXV3zeUe583fANhQUOlb1oaid\nEyRHwcXlcHEZxJ2729DZE2oO+E/uSZIKigxfkiS9sFIy05i1dxnrAvcA4F2mGl92GEYZpyfruTm5\n7Qo/jdzK9ZBbADR/syYDZ7ahaCmHe9rpEZPldyMmzB9ADBPuAkYA667CrUmirfFHKOksdkjsjqjb\npaVg3DDEcygjgBY2DbC4ZcHNX9PY92co8RdSACjjWZQe44rTsv9otFaRiI2I1gKN/8/eeUdHVX1t\n+JmSnkAqnQChXDqBoBRp0oMo8KF0BelFRWzoDwUErChWilJE6YgiiPTeWwDpl16SQAohPTOZcr8/\nTiChJZOQ0DzPWrPmzi3n7knI8M4++7yb9PR0nJ2dad26NTNnzqRu3br4+vpmd7t8IyU9jb+ObMYN\nC4O9TbCkNVxcz02nM1c/qNwNqr4CxZ4iV13HJZJHACm+JBLJE8n+y8f5aNVUIhNjcDY48XrjbvQK\nCc11TRdA9KUEZoxYy84/TwJQuoo/QyaHUvPZsnecawNeQWSyfIC3gFHASsQ0JBoiqZQKlbvCrI5Q\nj/vpengnqXYTu02HOWURNVyLd2xkfduTmNNEyb5SrySdRzakfoct6PXdEXKxMbAYKMb58+fp2bMn\nY8eOpXXr1rRu3Tofo8sBu409W79lZPpmWusjcN35p9hvcIag9kJwlQsVryWSxxQpviQSyROF2ZrO\nD9sWMTdsJRoaVYqW45PQYblufg1gSbfx16TdLBy/DXOqBVcPJ3qMbcrzbzx9sxdjVjSgN6K43gAk\nAe9lOV4bKD0Dlm8Ef3/Y+gPkzi8/Zw6bT7PPdBQLVjQLnP3+Gqe+isWeplG3XQVeHNmQao2LotO9\nDszIuOp14GvAiTNnztC1a1euXr3K9OnTadWqVYGvagQg9hgcn4N2Yi7NkyNE00kNKN4Aqr0ClbqA\n24PJvEkkBY0UXxKJ5Inh+NVzjFo5mXNxERh0evrX/z8G1O+UK5PUGxzacJ5pw1YRrl4DoHGXqvT7\nuhX+pQrdcp4d0eh6NTAJiMjYb0Poh8bA/yGmFQ2Xodrb4vgPP0BAPigvq8XGpeMxnAm7yq7VR3Fp\nb8S7gzPRa5M59n4Upks2mnavTud3G1C2RlFEl8imwF5ENdpPiFwdnDx5km7duhETE0P9+vX55Zdf\n7lN4mchsinQXUqPh5AI49htEHwDEzyxc82CzcxW69ZyD0a/yfdxfInk0keJLIpE89lhsVmbu+Yvp\nu5ditdso51uCCaHDcu3XBRAbkcjMt9axbfFxAEpW8mXwj6HUbhUEQDyif+KujMceIOG2MZ4CBiKK\n7G9Ul2katB8MSUnQoQN07Zr792m12Lh4NJqzB65yJuwKZ8KuEJ12nUpj/Dgz8xpxO9NwWqfHZ64b\nujNOtOxQhw4j6lEk8IZZ7BZENVo0UAbhICb6MV6+fJmXXnqJuLg4GjVqxC+//IK7e+5Xgd7KaIT4\nqsHNiVVLGpz7W6xWPL8aNJvY71IYKnXhq2gdc67GM7xeDym8JE8sUnxJJJLHmnPXIvhw5WSORYnG\n1b1CQnm9UXdcnXJXE2S12Fj+/V4WjN1KWnI6Lm5GunzUmMpv1Wefi5EfEWLrxF2u9UaIMh3wK/Dy\nXc6ZOxdWrgRvb5gyJecacUu6EFo3RNbZsCucPxyNNV2IFSdvPZX+F0DDvoHojTp8Snlim6vRrENd\nytcpjpdv1k6OGvADogLNhnAMWwhkrlwsUaIEzz77LHFxcUyfPh03t/ttVL0T+ArQg/YTRGwXgktd\nDOmJ4hSdIbOOq/zznE+4xpxf3sbV6ELnrPYSEskThhRfEonkscSu2Zl/YDXfb1uA2WqhuJc/40KH\n8HRgtVyPdXjzBaYNW8Wl47EAeHdSiP6mNQPLeJN427nOQAjQIONxFng/49gv3F14Xb0Kw4eL7W++\ngRIl7jwn6kI8B9eeuym2LhzJFFpZKV7BlyrDi+LxEmiuGjoNqjiX56mnqnHceJTgkKDbrkgFBgFz\nM16/C3zKjY//gwcPUqRIEUqWLMk333yD1WrFxcXl3j8sh0gF+gAaXG4Iq3tA4oXMw0VDhOCq3A3c\nM1ee3mgl9FzVxhR280QieVKR4ksikTx2RCREM3r1NPZfFlODHas3491nX8Ezl2apZ64k8dW76wmf\nJ3yjUsr7cOyHtsSEVrh5TmkyhVYDIBi4IU0WABmOEfyAKLa/HU2DoUPh+nVo0wZ63+Wk84ejeKf+\nrJurEW9QspIvFUKKUz6kOBVCihMUXAxPb1f2mY4RZj5OSUMRGroF42e4Vw/KCwgDi0MIx7BZiGlH\nwd69e+nVqxcBAQEsW7YMf39/DIY7FxLkirQ4SH4RAk5DDPDHDpFs8ywFVXtB1ZfBr+odlyWaUlh+\nbCsA3Wu3vb8YJJJHHCm+JBLJY4OmaSw7uoUvN/1KSnoavu6FGdNmIM3Kh+RqHJvVzl+T9zFj9BYM\niWZsrkbO/O8ZIt5tSG1XI68ghFZ9oOQ9xliOyHJpiDzSa/c4b8kSWLoUPD3h55/vnG60Wmx8++py\nzGlWqjcJpF4HRQiu2sVwLyRkXrpmYWfaIXyd3fDElWAXhQCDD2WMxbMpiF8HdAPigPLAX0D1m0d3\n7NhB7969SUtLo2XLlnh730d/RFs6nFspphXNy6GLTaxEWO8GlbuILFeppqC/t7BbemQTJquZeoHV\nqRhQOu+xSCSPAVJ8SSSSx4LYlHjGrZ3OlrNhALSo+DQftuqPr3uhHK68leM7LjNl6EouHI4WdhDt\nK9L2uzZ8HORDMGJaMSc0oB8iofM+mdmvO2KOhWHDxPaXX0Jg4J3n/DlxF2cPXKVImcKMXtENd69b\np/yirNfYkLaHRHsKkbYYunm2xUlnpKzTXeYub0b3JfA/hAJqh+gqmSmutmzZQt++fTGZTLz44otM\nmjQp9xkvTYOre8VKRXUhmOKEO+wrGcevdYIX54BTzh79NrudhQfXANAjJDR3cUgkjyEG39PFAAAg\nAElEQVRSfEkkkkee9af2MH7dDOLTkvBycef9Fq/yXJVGubJBiI9O4Zf31rPhV9Ha2lrWm4Pft+HL\n5yvddbowJ2Iznj/N5pw334SYGGjaFAYNuvP4xWPRzP9YTLW9MaP9LcLLrtk5aD7JfvNxNDT89N60\ncK+XrUmsXp+KmFZckrFnNDCGrBauNpuN8ePHYzKZ6NGjB1988QV6fS4sXhMuwIm5cHwOXD+Vud+/\nOoR6gfcuIBgCFuKYlIUtZ8OITIyhVOEiNC5X2/FYJJLHFCm+JBLJI0uiKZnPNsxm5YntANQvU4Nx\nbQdT1MvxFtM2m51V08KYM2oTKQlmjM4Gao9syLgPnqGomxPd7zPGe8m/v/+GefPAzQ1mzIDb9Y3N\naufbV//Gmm6jzYDaBLe8tVD+lOUi+8yizXYt50o87Vodgy677NRpKlfuA5wDvBAF9i/ccZbBYGDO\nnDksXLiQ4cOHOya8zAlwaomYVgzfmrnfvShU6SnquAJiQNcakf76FUeFF8C8A6sA6F6nLYbcCEGJ\n5DFFii+JRPJIsvPCv4xZ/RPRyXG4Gl14q2lPugTnzm395O5wpg5dxdmDVwEIaVueQT+0pU8FX+zA\nG+RGIjhOfDwMHiy2P/kEKlS485ylX+/i9L5IAkoXot9XrQDRFui6PZGSxiJUcirLZWsUVZzLUcpY\nNIc7rgB64eaWAFQBlgLKLWf8/fffrFu3jm+++YbixYszYsSI7Ie0W+HCWpHhOvsXWE1iv9EVKnQS\ndVxlWoLeCCQi2oiDyLTVzCHeTE7FXGT/5eO4O7nSoXozh6+TSB5npPiSSCSPFKnpJr7dOp9Fh9YC\nUKtERcaHDqWMT3GHx0iITeXXDzaydsZBAAJKF2LAd21o0FHhsE7HOsTav7vMBOYL77wDkZFQvz68\n8cadxy+fiGHemC0AvDa9PU5eBvabjnHIrGLQGejhFYqLzplW7vVzuJMdGA+MBeD69Wfx8VmGyHxl\n8ueffzJ8+HDsdjutWrXi+eefv/twmgbRh0SG6+QCSI3KPFa6mRBcFTuDy+11dm8Dl4C6wMgcYr6V\nG/YSL1RvilcuV6tKJI8rUnxJJJJHhkMRKh+tmsql+KsY9QaGPtOFPk897/BUlM1mZ93MQ/z6wUaS\n4tIwOunp9E4Duo5qhKuHyHFNyji3L6LxdX6zbh3MnAnOzjBrFtxex26z2fm2799YzDZa9Q3Gp7kL\nC5JWkaqJzFIpQ1Esmg2XHBN8CYj1ln8jJj8/4dy5VoSE3Cq8Fi1axNtvv42maYwYMYL27dvfOVRS\nBJyYByfmQOzRzP0+iphSrNoLCpW5RxyrED0iXRDTjY7/t3I9NfHmlHL32m0cvk4iedyR4ksikTx0\n0q0Wpu5cwux9y7FrGpUCApkQOgylyL3+w7+T0/sjmTJ0Faf3RQIQ3LIcg35oS+nKmS7uEQhvLj3w\nZv6+BUC0DhowQGyPHQtVqtx5zrJv96DujsC3hBftv67N2tRdAAQYfGjgWosSRkcaPh5HdIs8jZCQ\nC4A2QNgtZ82bN4/33hOtvd977z2G33B6BUhPhjNLxbTixfXc7MHo6geVuwvRVeypHKz4rwP9M7bH\nA3f6d2XHH4c3YLZaaFyuNmV977V6UyJ58pDiSyKRPFROxVxk1MrJnIq5hF6no+/THRjS8EWcjU4O\nXZ8Ul8Zvozax+qcwNA18S3gx4JtWNHqp6h31YT8AFuBF4HYf+Pzggw/g4kWoXVtMPd5OxKlrzP1w\nM+jg9Z+fI6hwScqnlaaIwZeazhUdrGdbgnCPT0HUVi3lXu+mWLFiODs7M3LkSAYPHgx2G1zeJATX\n6T/AkiJONDhD0PNCcJULFa8d4k0gEuGK9paD1wgsNiuLDq0DRKG9RPJfQooviUTyULDZ7cze9zdT\ndizGardR2rsoE0KHElxSyfliwG7XWD/7X2aP3EBibCoGo54Ob9aj2+jGd3hlASQBP2Vs30UX3Tdb\nt8LkyWA0iulGp9u0o81m57u+f+NR04mWsyuiKCXQ6XS0dKvnoOiyAR8Cn2e87g5MR1Sv3cqpU6eo\nVKkSLVq0YMuWLQS6J8HW94VFRHJE5oklGgrBVakLuPnm8h0vB34DXIHZQO58wjac3kt0chzlfEvQ\nsKzjBfoSyZNAgYkvRVH0wBSgFmAG+quqeibL8VDEshiAA8AwVVW1gopHIpE8Oly8foWPVk3h38jT\nAHQNbs2bTXrg7uzq0PVnD15h6rDVnNwVDkD1pmUYMrktZaoVuec1sxDNr58B6t1n/LeTmgr9+ont\nDz6A4OA7z1nx4z6SghJpMCkQvZPGIbNKE7c6Dgqva0APYC1C5HwFDOduRhfff/89EydO5PuJE+hU\nKZ7AY79B9IHMEwqXgyoZdVw+FXP7VrPEMzBj+zOgUq5HmJ9hL9GjTttcrWCVSJ4ECjLz1RFwVVW1\ngaIo9YGvyViLrCiKFzARaKaqaqyiKO8B/ohOYBKJ5AlF0zQWH1rHpC3zMFnNFPH05eO2g2hYtpZD\n1yfHm5j70WZWTtmP3a7hU8yTfl+3pGn36tn+B24Fvs3YLois15gxcOYMVKsGo0bdeTzyTBzrDu6j\n1g9ixWZN50rUc63h4OiHEP0ZLwABwGKg2R1naZrGgnm/sWDREnQ6MK8ZBhEZ04ouhUV2q+orUPKZ\nHOq4HOE1IApogjDsyB1Hr5zl38jTeLm4075qk/uMRSJ5/ChI8dUIWA2gqupuRVHqZjnWEDgCfK0o\nShAwQ1VVKbwkkieYqKRrjFnzE7suCIf5dlUa8UGLPhRy9czxWk3T2DjnML+8u4H46BT0Bh0d3qxH\nz4+b3ux/mB1/IqRLBeAeJgt5Zs8emDRJmKjOmgUut4Vjt2tMn7aSapNEVq6Bay1quTiaKZoHDADS\nEDYOfyJafWdBs6OFb+fz8R+y4J+z6HUa37aJp3N1C5RtLwRX+eeFP1e+sARYiJju/IWs7vmOMv+g\nyHp1qvGsw9lOieRJoiDFVyHEWugb2BRFMaqqakVkuZ4FgoFkYJuiKLtUVT11l3EkEsljjKZprDyx\ng882zCLJnIq3mxcftuxHKyUnDyvBhSNRTBm6iuPbLwNQtVFphkwOpVzNnIxHM+6PmKQDGEFuK5Ny\npm9fsNvh3Xfh6afvPP7PlP0c+uUSz3QqQ8Mq1R0UXhbgXeC7G3cBJiPqqzK4floUzh+fw7qDV/jx\nHz8MOo0fuvvQofv7ULkbuN97GjZvRANDMra/JC/LFmKSr7Pm5C70Oh3dpL2E5D9KQYqvRG51+tNn\nCC8QBQP7VFW9CqAoylaEEMtWfIWFhWV3OF94EPeQ5B75e3n0cOR3kpieypxT6wiLEX/awX7l6a20\npnCyU47Xm5ItbP75NHsWXcRu0/Dwcab18MrUeq4kcZZw4sLCHYrzoKcn+xSFwlYrtQ4fJky7/9JS\nDSAkBIDjxyEw0ESHDscJC7t17MiEBGaP2o0l0UbRvYVxdrUQdj779200xhEU9D5eXgew241cvvwO\nsbGdgWMYLAn4RK/DL+ofPBOP3LymaZWidL8eSMW6LSnVrCNhGnDiMnD5vt9r1ncdFDQSH59YEhOf\n5vTpp7jd2sIR/jq/A6vdRm3/Clw9c5mr+Rrjo4n8/Ho0eZi/l4IUXzsQGf7FGTVfR7IcCwOqK4ri\nj6iBrY9YtpMtIRkfdgVFWFhYgd9Dknvk7+XRw5HfyeYz+xm3di5xqQl4OLvx7rOv0LF6sxyLqzVN\nY+vCY8x8eytxV5LR63W0f+0peo1vhqd37qeoxmc8v2408kydOrm+/q4xZtnW6WDePFeeeebWsS+n\nR7En/iy1ZhTHZZkXL4+4i7npHexFZLnCgeLo9UsoU6ouZdJXCtf5cyvAbgHAbvTki6M16dpnGEEN\nuvLV64YC/ltZAGwEvChUaDEhIY57sN0g3WrhnT0/AzCseXdCAqvlb4iPIPLz69HkQfxeshN3BSm+\nlgKtFEXZiViS86qiKG8BZ1RVXa4oygfAmoxzF6uqevReA0kkkseHZHMqX276jWVHNwNQt1QVxoUO\noWThnKfALh2PYdprqzm86QIASv2SDJkcSoU6jrcWysophCGCM6JEPL+wWBD9o4Fhw6BRo8xjNs3O\nAfMJwtKOo3fRYbsOg75xxMdqJjAUSAetIVz9HxybB+rzYIoTp+j0UKY1tsq9eGf6Xhav/JN/Tkxh\n8+YuGAu0H/UVYFjG9iQg98ILYI26i7jUBCoFBFK3dO4MWSWSJ4kCE1+qqtqBwbftPpnl+EJE1aZE\nInlC2HfpGKNXTyUyMRZngxPDm3SnR5226HXZK4O05HQWjNvKsm/2YLPaKeTnRp8vWtDy1WD0+ryv\nzPsGkaV6GXCsQswxJk4E/ie2P/ssc3+8LYn1qbuJtcejoXHu+zheqNwY74DsFhWYEbYRP4kq2RP1\n4HgMXM+SKfOvDlV7Q5UeWF2L8Oabb7J06VLc3Nz4/PPPMRoL8nu0huiCeR1oC/TL2yiadrOPY3dp\nLyH5j+PQX6yiKB5AecTUobuqqikFGpVEInmsMFnS+WH7AuaGiVVs1YoGMaHdMIL8SmZ7naZp7Fhy\ngukj1nItIgmdDtoOqkPvT5vj5et2XzHFIKw/Ibfe69lz/Dh8/DE3xZdnFl1l1BlItKdgjdbY++ol\nqhQvS+OPsptaiwRzBzi1H47rIFwD9ohD7kWhSk9hghpQC3Q6LBYLrw0bxooVK/Dw8GDOnDnUq5ff\nrmW38xuif2RhRHVI3kTTochTHI86h7ebF+0qN8r5AonkCSZH8aUoSguEMbQB0UPiqKIoPVRVXVvQ\nwUkkkkefY1fPMmrlZM7HRWLQ6RnY4P/oV68jTobsP17C1Vh+en0NB9edA6Bi3eIMmdKOSk/l3OMv\nCTibwzm/ASagHbntOHhvbDaxutGSnrlP0zQOmk9S26Uynnp3fDcEsLzPZtydXRjyZ2g2o22H1P+D\nOTFizTeasIOo0EnYQ5RpCfpbf4apqamcO3cOLy8v5s6dS926de86cv4RjsjKgVh5WSrPI83PEOYv\n1myBq5Oj7YskkicTRzJfnyI8u1apqnpVUZQmiMpLKb4kkv8wFpuVGbuXMn33UmyanSDfkkxoN5Rq\nxcpne50pJZ1Fn2xn6Ve7sFrsePq40vuz5rTuXxuDIefCpSMIn5prDsb5toPnOcJ33wlfrxIlRUdD\ngNOWS+w1H8VJZ6RoVBHm9dmONcnOoAVt8S5yZ+sfMY03BXgTdlqF8PJToO5IqNgZXArdcYXJZEKn\n01G4cGEWLlzIlStXqF69ej6+s7uhIZpmJyDWTr2S55GuJsay4fReDDo9XYJb5VN8EsnjiyPiS58h\nugBQVfX4jW2JRPLfJCIllq/mj+Z41Dl06Hg55Dlea9Q124yGpmnsXqby8/C1xFwSFoCt+wXT+7Pm\nFA64m0i5k4uIqqNriJJv7xzOb4gQavnB6dOZ7vXTpsELgJNmY69JLOQ2YuTHgf+QlpRO/Y4KTbre\nbboxDeGT9auYFz2sE0X0z/8JfnfPz6WlpdG3b19cXFz4+eef8fPzw8/PL5/eVXbMQKyJ8gV+Jq/T\njQCLDq3DptlpozSgqNeDiF0iebRxRHyFK4rSHtAURfFGLHm5VLBhSSSSRxG7Zmdu2Cq+2z8fq91G\niUL+jA8dmuPKtStn45j2+hrCVon2rkHBRRkypR1VGjg+jXUNaIPIODVFtM94UN7odjv07w8mE/Tq\nBe0zauFbmE+RrKXhp/fm8vx4Dqw5i6ePK8OmtrtLQflF4P+AA6C5waZyoB2H2kPvKbxSUlLo3bs3\nu3btwt/fn8jISMqWLVuA7zRrrDcq5X4EiuV5pDSLmT8ObwCgZ0h207ASyX8HR8TXIMRkf2lEmcVG\nRL8LiUTyHyI8PprRq6cSFn4CEK1h3mn2Mp4u7ve8xpxmYcnnO1jyxU4sZhsehV3oNeFZ2g0JcWiK\n8QYpQHtABWoAf/HghBeITNfWrVCkCHyb0SSykN1EG7NYwF0tpQIfj/gdgEHfB+NT7OptI5wEXgVi\ngSA48xpcfgtcfaHB2LveMykpiZdffpl9+/ZRtGhRFi9e/ICElx3hNZYMdAa63ddoq07sIMGUTLVi\n5alZPK+NvCWSJwtHxFctVVW7Z92hKMr/IZqMSSSSJxxN01h6dBMTN/5GqsWEn3thepZvQb82XbK9\nbu+KU/z0xhqizscD0KJ3Tfp80QKfojn3csyKFegK7AYCERmvnKYb85OLF2HkSLE9ZQr4+YlqqOL2\nRKw6PWUMxfm9/x5SEsw8/Xw4zXq2zma0tmCdCVsyVvs1HAduvnecpWkaAwYMYN++fRQvXpzFixcT\nFJT7Vj65xw58gviO7Y+oTcvbdKOmaSw7uoWvNs8BoIe0l5BIbnJP8aUoSlfABRinKMro2675H1J8\nSSRPPDHJ1/l47c9sO3cQgFaV6jGqZT/OnTh9z2uunr/Oz8PXsPdvcU7ZGkUYPDmU6o0Dc33/Gw5T\n/wB+iAqknNdC5h+aBgMHQnIydO4sHjdQjUUY4xXKH78fZt8/p/HwtjJs2gJ0Oh9EnVRWDEBPYBSE\nTYSE8+BXDWoNuut9dTodb7zxBlFRUfz6668EBub+Z5d7TgADge0Zr6cBeesNGZsSz7i109lyVjh8\nt65UnzZKg/wIUiJ5Isgu8+UFPJPxnLVm1QqMKsigJBLJw2etupsJ62aQYErGy8WD/7V8ldDKz9wz\ne5FusvLHxJ38/ukO0k1W3Lyc6TWuGe1fewpDHu3XPwRmAW7ACqByHt9LXpk9G9auBV9f+PHHzP2X\nLFfQGYthu2rml0GbABj47XL8ShiAA0DZuw+YfAX2fCK2n/32DiuJa9eusXXrVjp16kTDhg1Zv349\nBkN+twK/HTPwOWJhezrCjvZHxJRj7lmn7mbC+pnEpyXh5eLOyOZ9aF+1scx6SSRZuKf4UlV1BjBD\nUZQWqqpueIAxSSSSh0hCWjKfbZjFqpM7AWhYthZj2wyiqNed02M3CFt9hmmvr+HKGdEGp2mP6vSd\n2BK/El55juMHhBwwAL8jGsA+SCIjYcQIsf3dd1Aso+b8kuUqq1K3M9wQwLbBUaTEmwgJPUPzVw4B\nS7in8ALY/j+wJEP5F4SPVxaio6Pp2rUrp06JJuSdOnV6AMJrB6KE90TG6wHAF4BPrkdKSEvm842/\nsPLEDgDql6nBx20GUayQfz7FKpE8OThS85WiKMoywBMx+W8AyqiqWrYgA5NIJA+eHef/ZcyaacQk\nX8fV6MLbzXrxUq2W98xaRF9KYMaItez8UxSeB1b1Z/DkUGo2K3tfcSwm09pzBvDcfY2WezQNBg+G\nhAR47jno2VPsT7Ob2WU6BMClo1Bs+SncC6Xz+s9/odMNIdts0dX9cGw26J2g6de3HLpy5QpdunTh\n3LlzVKpUiWeeeaZg3thNEoD3EVOLAJUQdhJN8zTajvOHGLPmp5v/bt5q2pMuwa1ktksiuQeOiK9Z\nwESgD/A9N9dKSySSJ4XUdBOTtszl93/XAxBcohLjQ4cS6HN3iwGrxc7vn+9g4fhtmFMtuHo40WNs\nU14Y/jRGp/vL1mxC9GLUgM8QHzwPmoUL4e+/oVAhsdJRp4PLlqtsSttHqmbC0+5OSqf9GIH+k1bi\nX6osouF0NhzKmLesMxx8KtzcHRERQZcuXbhw4QJVqlRh0aJFBezjtRTRZjwS8V/A+4hKktyvH01N\nN/H1lrkscfDfjUQiETgivsyqqv6iKEpZRGfVVxAm0xKJ5AngYITKR6umcDk+CieDkaENX6L3U89j\n0N+9TuvQhvNM7beN2IuixWvjLlXp93Ur/Evd6cyeWw4BHRCVR28AI+97xNwTHQ2vvy62v/4aSpUC\nm2Znm+kAqZqJ4gZ/jr8dg/FqGjGtg2jV9yQQRo7iJT1JPBe/dQL1n3/+4cKFC9SoUYP58+fj63vv\n6d37IwIhuv7KeN0Ake3Km1P+gfATfLhqKhEJ0Q79u5FIJJk4Ir5MiqL4Iix26ququlFRlIIuRJBI\nJAVMutXC5B2L+XXfCjQ0KgUE8km7YVQKKHPX82MjEpn51jq2LT4OQMlKvgz+MZTarfLHAuE8EIro\n29gF+Ib78VTPO6+/DteuQYsW0LlPElbNHaPOQAu3eoRbo0leZmPHzO1YvJw5PL09Ot1k8rIUwG63\no9frGTBgAC4uLnTs2JHChQvn/xvCjmjP+z6QiFhD9RnCaT/3QslsTWfy9sX8tv8fNDSUgDJ80m4Y\nFQMexIpMieTJwBHxNQlYhJhu3KsoSk/E1zyJRPKYcjL6Ah+unMLp2EvodTr6Pd2RwQ1fvGszbKvF\nxvLv9rLg462kJafj4makUd8gXvu6M04ujnyE5EwMwr3+KtAc0RT7YeRPli6FxYvBwwPGz77MHyn7\nqOIcxDNuwRQ1+uES58rQ16YCcOKrVpgCCwO9c32fM2fOMHjwYH744QeqVKlC7965H8MxjiOK6Hdm\nvH4BmExeG2Qfv3qOUaumcO5aOHqdjv71OjGoQeccm6hLJJJbyfEvRlXV3xVFWaKqqqYoSl1EZeaZ\ngg9NIpHkN1a7jdl7lzN15xKsdhuBPsWYEDqUWiUq3fX8hJgU/td8DhePxgDQoJPCgG9aczn2bL4J\nr2REQf1pIBhRkeSSLyPnjrg4GDoUdDqNL1Ye518vkeEzaWY0TUOn0zHttVUkxqZRq8VZ/hlQO+PK\n3OXnTp6PpOt744iNjeW7775j2rRpOV+Ua8yItaKfARZEe6AfEd+hc59PtNiszNyzjOm7/8Rqt1HW\ntwTjQ4dIx3qJJI9kZ7IagGjuFYeYAbAiusI2QJhMF30QAUokkvzhQlwkH62ayuErwvy0a3Br3mzS\nA3fne9cqHd9xmYtHY/Ap5snwWc9TN1QUil+OzZ+YLMCLwD6gHLAKuP/Ksbzx1ltwPcnK+yv2YqgV\ngQ6o71qLms4V0el0bF9ynO2/n8DVw8zrM1byaR5W8h2NNtLtzR+4nphCkyZN+Oabb/L/jbANke1S\nM14PQvh45a0vwNnYcD5aNYVjUecA6FknlNcbd8PN6WFIZInkySC7r67zEOUX/oCzoihLgQWIgoER\nDyA2iUSSD9g1O4sOruXbrfMxWdMp4unLuLaDaVC2Zo7Xapp4VuqXvCm88i0uoB/CtT4g4/lhrZFb\ntQp+/RUqNUigTMNInHCilXt9SjuJiBJiU5k6dAUAr365lqJlP8z1PU5EpNL1d3/izSk0b96c6dOn\n4+qanx0q4wkM/ASROwRRh/Yz0DhPo91oov7DtoWk2ywU9/JnXOgQng6slk/xSiT/XbITX+VVVS2v\nKIoXsAsYivA9nKSqavoDiU4ikdwXVxNjGbPmJ3ZfFAuU21dtzMjmfSjk6vGQIxPl33MAD2Al8LAm\nsBITYeyP1wA/+nfyo6VHPfz03ngbhEGspmlMe20lCTFmajQ7T+jg4sDrub5PKV9nyvlYKVI+hKkz\nZuDikl+ZIw3R7e11AgKuAE7AB4gucHm7x+1N1DtWb8a7z76SbRN1iUTiONmJr0QAVVWTMlY7dlZV\nddeDCUsikdwPmqax4vg2vtg4myRzKj5uXnzYagAtKz39sEMDRB3DRMQH0J9A3YcUR6rdxJR/D/Lq\n/HACxzZkxIiSGI2lbzln6de72bboBC7u6QyfuRu9fhtaLuqmDh8+TIUKFfByM7Kg8zVcO/bFKd+E\nVzgwDFgOQHJyTTw9FwBV8zSapmn8eWQjX22ac7OJ+ug2A2lWPiSf4pVIJJC9+NKybEdJ4SWRPB7E\npSYyft10Np7eB0Cz8iGMbj0AP4+81fzkN/MRxaQAvwCtH0IMmqahWi6yNekQ3jUtmFMMDH7dgvG2\nT8Q9y1V+eU8YiI749S+KBf2MaPHtGNu3b6dPnz7UqVOHX7vY8XLRwJgfTj12YCoiw5WEqAb5AlWt\nS0hI3oRXdHIc49ZMZ9v5W5uo+7g/rCo8ieTJJdvG2oqiNEas+PbI2L75dU9V1a0FHZxEIskdm87s\nZ9za6cSlJuDh7MbI5r15oVrTR6bNyzoyHesnAr0eVhypuzhnjQAjHFtXjGLhdWjx5q1TsecPRzGx\nx59oGvQav4FGL3YFGjl8j82bN9OvXz9MJhMlS5bE2Xgxn6I/CgxEVIMAdESsZCxJXl2AVp3cyafr\nZ5JoSqGQqwf/a9GXtpUbPjL/biSSJ43sxFc4MC5jOyLLNoisWPOCCkoikeSOJHMqX278leXHtgDw\nVOlqjGs7mBKFAx5yZJmEIYwOLIjM1zsPMZYSxiKcSohhzohgzGog+/bdKjKuRyUz7vmFmFKsNO3x\nL11HGRBZJsdYt24dAwcOJD09nZ49e/L555+jX/HSfUZtAj5BNL62AMURnl2d8jxifFoSn6yfyVp1\nNwDPlK3F2LaDKOJZUC77EokEshFfqqo++yADkUgkeWPPpaOMWTWNK0mxuBidGN64O93rtEWve3Ta\nvJwF2iE8vXoisl4PmpPp53HWORHkVIqEsPJ88EJp0uJd2LsXnJwyz0s3WZnQcTExlxJR6l9m+Mwd\n6HQHAMemC1NTU3nnnXdIT0/n1VdfZfz48fmQQdqCyHadyng9BOHhlXdH/K1nD/Dx2p+JTYnHzcmF\nd5q9TOeaLWS2SyJ5AEhbYonkMSXNYub7bQuYf2A1ANWKleeT0KGU8yv5kCO7lShEXVd0xvMsHqx7\nvV2zs9t0mMPppzFioLCzH/36upF8zYUPPoA6dTLP1TSN7/r9jbo7goDAeD78az7OrksQWSbHcHd3\nZ/bs2axdu5b33nvvPsXMdeA9YEbG6yrAdOCZPI+YbE7lq81zWHpkEwB1SlVmfNshlPKW1o0SyYNC\nii+J5DHkyJUzfLhqChfiIjHqDQxs8H/0q9cRo/7RaruahMh4nQNCgCWA8wO8v8mezvq03YRbo9Cj\no6FbMF+NcUNVoXJlGD361vMXfbKdLfOP4uphYfTf8/ApOgxo69C9/vjjD65cuZc22mQAACAASURB\nVMJrr71G7dq1qV27ds4X3RMN8dN6HSFfnRHWEe9zP/7/+y4dY/TqqUQmxuJscOK1Rl3pFdJONsOW\nSB4wUnxJJI8RFpuVn3f9ycw9f2HT7AT5leKT0KFULZY/za3zk3REjdcBoALCy8vrAd4/zW7i9+R1\npGomXHUutHFvQMShACZOBJ0OZs2CrB6n238/ztyPNqPTaby7YDHlapYFJjh0r4ULF/LOO++gaRr1\n69enbt37Mc+4jLBVXJHxuhHCLLVKnkc0WdL5ftsC5h1YBUCVouWYEDqUCv6lc7hSIpEUBDmKL0VR\nfIAvgfKITiBfAW+rqnq9gGOTSCRZOB1zmQ9XTeZk9AV06Hil7nO81qgrLsYHmUtyDDtiVeN6oAjC\nvb7IA7q3TbNj0Olx07tS3BhAij2VFu71cLF60KYv2GwwYgQ0aJB5zen9kXzTexkgHOzrPX8V0ezI\n6a73yMpvv/3GBx+IYvz333//PoSXDZiCyHAlIxotfYloFZT3zFTWLKlBp2dA/U70r99JNsOWSB4i\njvz1TQfWAk8jPhGuAHMRvXAlEkkBY7PbmRv2Dz9sX4TFZqVEoQAmhA4lpHTeMyEFiQa8jehF5omQ\nMA8iL6dpGqctl9htOkx7j6b4GgrRzK0uRgzodDo+/gyOHIHy5WFCloRWbEQi4zssxpxmpVXfQ3R6\newfwO6LbZPZ4zJzJBxlzl6NHj2bQoEF5jP4IQmTtyXj9f4iGIiXyOJ7Ikv606w9m7VkmsqS+JZnQ\nbijVipXP85gSiSR/cER8lVNV9WdFUYZktBUapSjKvwUdmEQigfD4KD5aPZUD4ScB6FyzOW83exkP\nZ7eHHNm9mQh8i8gZLQXqZH96vhBvS2KH6SCXrVEAqOnnaeBWCyed+Ig7fDhTcM2YAe4ZXXJMKemM\nf2ERcZFJVG8SzdCpy9DpBiOS/A5gsQAwYcIEXn311TxEbgLGIzJcVoRX148I7668c3uW9OUQkSV1\ndXr0sqQSyX8RR8SXVVGUwmQ43iuKUhExqyCRSAoITdP44/AGvto8hzSLGX8Pb8a0HkiT8g9CyuSd\n34CRWbZbFvD9NE1jr/ko/5pV7Gg440RDt1ooTmVvnmO1Qt++4nnIEGjWTOy32zUmvbKMsweuUiyo\nEB/88TlOzq7ApBzve+XKFShenJTBg9ndqBHVq1cHWzqkXIGkcEiOyHgOF89J4RB75LZRNiPsI04j\n/KuHIuwj8u4ob9fs/LJ3OZN3LH4ssqQSyX8VR8TXGMSnRKCiKH8BDYC+BRmURPJfJib5OmPX/MT2\n84cAaK3UZ1TLfni7Pchy9dyzGuiXsf0t0O0B3FOn02HW0rGjUdmpHPVcq+Omd73lnEmTICwMAgPh\niy8y988bvZmdf57Eo7ALY1a0o7D/m4AvcO+soqZpfP3118yYMQOnhQuxBAdT/dCrsD0cUqK4tSvb\nXXDxhiKBQH9gZsbOqojqjoa5ffu3cDn+Kp8fXMiZhAgAOtdswdvNej3SWVKJ5L+KI+JrHbAfqIdw\nGRykqmpUgUYlkfxHWX1yJ5+un0WCKVm0eWnZj9DK9/ef8oNgL9AZMXE2EhhegPdKtqeyOW0/dV2q\nUszoz9Mu1VGcylHUeKcru6pm2kn8/DN4ZejXTXMPs+iT7egNOkYu7kzpKjn3vdQ0jc8++4zJkyej\n1+swXLyIJTgYovaLE3R68CgBXqXAsxR4lszc9ioFniXAaxcY2iNcz5yBDxE/sbxPB2qaxu//ruPr\nzfMwWc0EePgwps1AGgfdj9WFRCIpSBwRX5eAP4G5qqruyelkiUSSexLSkvl0wyxWn9wJPF5tXk4h\nVt+kAr0RE2cFxZn0S2xNO0A6otaqvbEJrnoXXPV3el/ZbGK60WyGPn2gTRux/8TOy3zXT9g4DPi2\nDXValweuZXtfTdP4eMxops+chVGv8WO7OF7o0EEcfGkD+CjgURT09/pIvYSYVvwn43VjhH1EZcff\n/F2ISrrGmDU/sevCYQDqFanCxJfeprCb532NK5FIChZHxFd1xJfaTxVFKYlYxDRXVdWzBRqZRPIf\nYfu5g4xd8zMxKddxc3Lh7WYv8+Ij0uYlPiol2+NXgDZALBCKmDwriKjNWjrb0g5yxnIJgDLG4jR1\ny97SYfJk2LkTihUTU48A0RfjmdBxMdZ0G88Nrcvzrz2V883tNuZMfIPpM//CSa8xrf112jQOzjwe\nmF2bWxuigH4UkIJoBzQRMUGbd/sITdP458R2Pt/wC0nmVLzdvBjVsh9+yU5SeEkkjwE5iq8MP68Z\nwAxFUeoCPwEfOXKtRCK5N6npJr7aPIc/Dm8AILikwoTQIZT2LvZQ4zKnWdi++Dgrp4ah7hH1Q25e\nd06LJSAE1wWED83vOOKKlTcOm09zxnIJIwYaugVTxalctuL03DnIsN5i2jTw8YHUJDMft19IQkwq\ntVt5MfA7G5mZqMQ7B9E0OLMMdnzIS9ox1pXzpXcjf5r0n8aK8s/nELEG7ES0EN+bse9F4Hty06ro\nbsSlJjJh3Qw2nBbjNgmqw5g2A/H38CYsLOy+xpZIJA8GR0xWA4CXEPWzvsB8oFMBxyWRPNEcCD/B\nh6umEpEQjZPByGvPdOXlus891DYvkWfiWDUtjPW//EtSXBoAHt6utOxTixffv7XuzKzT0RH4F6iE\nkDAe+RyPVbORbE/F2+BFbZfKJNlTqONSBW9D9gsPNA3694fUVOjWDTp0AJvNzsTuf3DxaAyllGuM\nXPwpBqPpLldnfCRe2oRt6/v8sOw4fWun4BlQjgFzpjC/bGu66vTEZZx9axdNDTgMLMx4XMjYXwqY\nDLyQtx9EFjad2c+4tdOJS03Aw9mNd599hY7Vmz0SWVKJROI4jmSvDgGLgbdUVd1fwPFIJE80Zms6\nk7cv5rf9/6ChoRQpyyehQ6kYEPhQ4rFZ7exdcYqVU8I4uO7czf0V6xan3dC6NO5aDVf3W/NZNmB0\nuXJsRuRw1gD++RxXrC2ejal7sGLjRc9WOOucaO7+tEPXTp8OmzaBvz98/73YN3vkDPb9E4WXbyqj\nV8zB07sW4HfnxVEhsK0N1vNreWOVN8vUQsyxNSZ6yQrCswjj6kBP4BUAVDIF18ksg5XIOOMD7sc+\nAiDJnMqXG2ez/NhWAOqWrsq4toMpWfhB9Q2QSCT5iSPiq7SqqtLXSyK5T05EnWfUysmcvRaOXqej\nf71ODGrQ+aG0ebkWmcTaGQdZM/0gseFiys3Z1UjTHtVpNySEinXv7qyuIVYybvDxoRDCXqJsPsZl\n1zT+TVfZZzqKHY3Cek9S7SacDY5NaF6+DO+8I7Z//BECAk6zduanLP06CIPRxgd/7KBEhbmICdMs\nxKmw40M4NZ7THmXot/sZzqrnsXt6cvT9saTr9QQCPTIeNbgILEKUwB7KMpA/mRMFjbifuq4b7Ll4\nhNGrp3E16RouRieGN+5O9zpt0etkM2yJ5HHlnp/6iqIcUFW1DsJkNat5jQ7QVFU1FHh0EskTgNVu\nY9aeZfy06w+sdhuBPsX4JHQYNUtUfKBxaJrG4U0XWDk1jN1/qdis4jtVyUq+tBtSlxa9a+Lpk70n\n1GeICTQnu51lej018zG+NLuZtak7uWKLBaCac3nqu9a86VKfE5oGgwdDUhJ07GihS5dRHN68lMmD\newIwbJobNZut5xZbh8TLsOtjos/9w+JKnZnXeSvqx7Nw270We6FCWOfOpX9ICD2ABlxBz++IDNeu\nLHcuhGgH1B1oTn6Vw6ZZzHy7dT4LD64BoHqx8kwIHUo5v5I5XCmRSB517vkpkSG8UFX1jq9XiqLc\nua5bIpHcwYW4SEatnMzRq2JxcPfabRnepDtuTg/uTyg53sSGX/9l1bQwwk8KSwW9QUfDzpVpN6Qu\ntZqXdahmaBZizZ4OGH/+PM3K52+PQGedExbNirvOlWZudQl0yl1h+ty5sHIleHubmTLlKa6cjeCz\nzgOxWQ10fKsGrftladmTGktS2CT+SrnKPKUL61tNw6Y3Yrh4kYADBzB6e/P+ggX0r1kSJ6YjBNdm\nMpt7uCNquLoBbYH8+X2mppvYc+ko288fYsuZMGJSrmPUGxjUoDN963XAqJffeSWSJwFHCu53qara\nIMtrPcJ0tUZBBiaRPM7YNTsLDq7h+60LMFnTKerly/i2Q6hX5sH92ZwJu8LKqfvZMv8o5jQrAL4l\nvGg7sDat+9fGv6TjdUgrEI1wQBgn1IuPz5cYLZqV/abj1HGtjIvOmdYeDXHCiNtdfLuy4+pVGD7c\nChj55ptBeLmd5p3mb5AU585T7Svy6pei2N2cnsTqc/8w3+DE8vofYXISmT6jycRzLgZ6limDsnAm\nLvYNVKv2AbAWYR0LImPWDiG42pMfSww0TePi9StsP3+IbecOEhZ+AovNevN4Rf9AxocOoUrRnJt8\nSySSx4fsph03As0ytrPWfFmB5QUblkTy+HIlMZbRq6ey99IxAJ6v2oT3mvemkGt+rwe8E1OqhW2L\njrFyahin90Xe3B/cshzthoTw9POVMDrlLnuyC+iCKLQfhbAKzQ9Dg2hrHBvS9pBgTyZNM9Hc/WkK\n6XP/M9K0SIYOjeT69bq0abOaXj3X8/Fz4wg/aaJ4jSJUmd+JT7Fz5PoZ1rsFcL1yZuOjxunJvGTR\nsb53L+qHFKLb+0noqqwEzBlnGBBOZt0Qza5zdsLPCZMlnbDw42w7d4jt5w9yOT6zYYgOHbVKVKRR\nudo0DqqNUqSMrO2SSJ5Aspt2bA6gKMp3qqoWZLcQieSJQNM0/j62lS82ziY5PQ0ft0J81Lo/LSo6\ntkrvfog4dY2V08LYMPtfkq8LCwVPH1davhpM6KA6lKx0l5V9DnACkeNJQ9iCjs+HWO2axiHzSfab\nj2FHw1dfmGAXJQ8jmYFvWLLkGEu3zcG1XSr15xrp8eZPpK47QHoRD+b93ZUZXhlZNJ9KANS6fpoe\nOgPdvEvhY15Fr16j2b8/kfDzNgYNisHXVwOaIGq4OgMB9/2ew+Oj2XH+ENvOH2TfpWOYrOk3j3m7\nedGwbC0aBwXToExNfNzvb2WkRCJ59Mku89VeVdUVwAFFUV65/biqqr8VaGQSyWPEtZQExq+bzqYz\nwo3l2Qp1+ajVAPw8ChfYPW1WO3uWq6ycGsah9edv7q/0dAnaDalL465VcXHLu+1pOCLnEwc8D0wj\nf9zrd5oOcTT9DADVnStQ37UmRp1j2bg04DgaRzjEEQ5wIL0eW5u8CzFgwp3ZkwtTfcpqbM4G9i99\nCR9/M03Or6ZG7BFqmOJ5usyzVC6tgW4R8fF/0r2XgYMHnSlRwsrixaXw9R2JWK1Y6r7eo8Vm5UDE\nSbafO8j2c4c4Fxdxy/GqRYNoVC6YxkG1qVas/EP1d5NIJA+e7Gq+nkKUejS7yzENkOJLIgE2nN7L\n+LUzuJ6WiKezGyOb9+H5ak0KzPjyWmQSa6YfYPXPB4mLTALAxe2GTURdKoTcn4M6wHVEGflloAGi\n3Px+1vBpmoYNO0adgRrOFblsvUoj19qUdrrVzd+O8JqPy3hcAo5keZxBw44OqC0ezkBRMKRCw21n\nKTRcrAx8YUQMcw8G47crDgqXhZavQJkE0PUEorFYoHt3fw4fdqZ0aS9+/30mpUs/cx/vEKKS4m5m\nt3ZfOEKqJdPE1dPZjYZla9EoKJhnygXj73H/05cSieTxJbtpxzEZz6/e2KcoSiGE79exBxCbRPJI\nk2hK4cuNv/L3cWF8+XRgNca1HULxQvltOSrEy78bL7By6n52/6Vitwn3l1KKH6FDQmjxSs42EY6S\nBnQAjgFVEN/A3HM5hg2IRwioaLuZs6aDpGs20twbcs3gSZxnW/7R6W6KrBuP62SuJ7wbBmxU4RQ1\nOInTySrMe68KLqdh/bcxfP/SYlJtGl1bbOHlgA1QxBeatoCAU6Abl2WUijg5daNnTy+SkpaxaNEi\nSpbMvX2D1W7jcORptp8X2S015uItxyv6B9IoKJhG5YKpVaLSQ/Fzk0gkjyaOrHbsBzQG3gUOAkmK\nosxRVfXTgg5OInlU2XPxCB+tnkpUUhwuRifebNKTbrVb53txdPL1NNbPFjYREadEUxuDUc8zL1am\n3ZAQaj7rmE2Eo9gQzu3bEK1z/gYsiNqvrCLp3yJF+Ou2fVkf8QCaxtOWS7xkOoSnlo4JA5/Zk4k2\neEE2MXsh+pj5AsXQqM4RavATNdhBZVRc6U18/ASqtfCHSPh+/CFm91lCapILL3TYSa8PtkNNP3C7\nBmzIGLU00I2oqNacP+9B/foN6NULOnd+DTc3x0XrtZQEdl74l23nDrLzwmGSzJmNx12NLtQvU51G\nQbVpXC6YYgUgwiUSyZOBI1/FhiJqbrsDyxAG17sBKb4k/0liU+IZsuQzbJqd6sXK80m7YZT1vbsj\nfF45tS+SlVP3s23hsZs2EX4lvWg7qA5t+tfGt3j2/Q1vx8S9hVLWx04gAlHbFQ9UuNeApUtnez9f\neyp90vZR0RoNQIwhgPNudXnJ4HlTWN3t4UPW5tw7gDeAAxmvGyPWXgbzzjtgSL3EX0PGcnJDEo1f\nvU6rlw9SosqNrotmoAhinWY3oAFXrkTRpUsXIiMjWbhwIU899VSOwsuu2Tl29Rzbzx1k2/lDHMvw\na7tBGZ/iNA4KplG52oSUqoKzsaBai0skkicJh/LgqqpeURSlHfC9qqpWRVHyZ35DInkMSTKlYNPs\nlCjkz689xuWb8aUp1cLWhcdYNXU/p/dfubm/dusg2g0J4an2lTAbRVPny2QKpmvkLKrSchmLBqQg\nmuPcTSRZo6KoUrToPUWUEzp+t15Hp3OmgWstFKcyucjQhQPvIVr3gMhaTUQIKR2bV0VTz/Y/pi6Z\nzbViXnSoltVzzAexQrEb0JQbH3Hh4eF06dKFixcvUrVqVYKCgu5594S0ZHZePMz2cwfZcf5frqcl\n3jzmbHDiqdJVaRRUm0blggn0KXbPcSQSieReOCK+jimKsgIIAtYrirII2FuwYUkkjz7ORuc8CS+N\nW4vK1csJbN54gbDD0SS7GrF0q4Y2oh7eNYviXN6HP92cmJFxbnp2A2eDE6KN9L3E0jGE1NEhZM7/\nZez34u7dCcPCwwkpWvSWfVesMajpFyjvVhed3o02Hg3x1RfCTe/qYJQm4CtEE6NUwBUhwkYC7mA+\nh+XC6zSutppms+ygg2LEk5bkjCW9PYX8+gKtuKV9EHDx4kVeeuklIiIiqFWrFvPmzcPHx+fmcU3T\nUGMu3sxuHY48hV3L7KhWopA/jYPq0KhcME8FVnug3QkkEsmTiSPiqy/QEDiiqmq6oihzgVUFG5ZE\n8njgyFRe1sc1RFG5LesgpQtD71oO3c+Newuouz1uCC537m0T8RcwOmN7GplO9o6SZjezy/Qvpyyi\n4LyEsQiVnMtQ0ljEwRE0YCnwNnAhY9+LCBnoDba5kPw1eJ3CKcMOzJquZ8/yymxdWIOmPd+lYaeQ\ne47+008/ERERQZ06dZg3bx6FChUi2ZzK7otHbxbLx6Rcv3m+UW+gbunKN41Oy/mWKLCVqxKJ5L+J\nI+LLGVHzNUlRFCOwCdhIZs8NieSJQAMOIya9spvKu44QMycR4iYvGJPTcbqWitO1NFwSzJT0cqZy\nmcKUC/DIth4qv+f7tyEm6OzAWHInvDRN40T6OfaYj2DWLOjRU9ulMkFOufHIOoooI92Y8boG8DmQ\nCNpw0FaCwQqFARtcO+LDT9P7c2yxBwlRenqNb5at8AIYO3bs/7N333FR138Ax1832FMQAcUBjnNv\nRRT3yJlbc1Vqv0rTSlMzzVXOXGWaDU3NUtHcmZqZG/deHC7EhaKAzAPu7vv74wMupsqh5ef5ePTg\n7r6f7/f7uQPjzWe837i6utKqe3vW6nex5/IJjt8IwWh+GAJ7OBagvm81Av2q4l+sIo42T7u/U5Ik\nKfdyE3zNRcwB9EP88fw/xB/IfSzYL0nKN+eBZWn/Xc5FexsgPR+7KzmPPrmaFaKO3OTE0lOErA5B\nczcRdaoZn7LutB5QkyZvVsbRNbdTc3nnDKI0dDLwHg9Hv3IrhVQOJ58lWUnFR+tJoG01XDW53QgQ\nBYwD5iPGAQsgkpvGgNIVVIkPh+qugvl6EYbMmMH369rQxmURKffv07BnRbqPDsz06ufPn+fzMZ/z\n9mcfcCrqCntdw/lt1bgHx9UqFdWK6KjvV41A32qU8SgmR7ckSco3uQm+auj1+kfnRAbpdLpzluqQ\nJOWH64jEocsQ+VPSeQNVeXzK7sn/khBjNWUBfTb3iIsSaSI2fX+UmxfELjxbrZq6ncrSakANKjV8\nmkXoeSsckUQ1BugIzCN32evNipnQ1KsoKNiorGlkVwsjRvy0Prl8LybgR2AMYnxRhfgkb6S9nvbS\nDcSHG1Ecqk1h5LruzF1nprHdUlLu30dXpwgfLWyX4Z7h0RGs+Hs980d/RUqCgRNjw3FsXBKAAnbO\nBPpWJTCtjI+LnWMu+itJkpT3chN8qXU6nater48B0Ol0rsgpR+lfKApYjQi4diGmGUHMaHUBeiL2\nx+W0hP5KNscURSH08E02zz/K7hVnSTGIfyoeRZ157d3qtOhf9anTROS1KETgdQORvGEZOb9ngCjT\nfXYkHSbSFI2PqxsAxa2eJpv+LmAwIlc9iG0AqYgJXCDaEU7Hi6DL5A0BY6Fhfw4esWLmTIUqbMQ2\n6ToexVz4fF03rG21JBtTOHr9PHsvi8zyl85fIHbNGZRkI9a+btTq0oRGuloE+lajvJevLFItSdJL\nITfB1yzgsE6n25D2/HXEdiRJ+lcwINYyrUD8qgcxddgOEXC1TnueW+mZzAvYPQyiDAkp7Fp+lj/n\nH+HSsYgHr1d/raRIE9GmNBrti//Fn4hYwHkeqIhI3JebCc9zKZfZm3QMMwoOKjvsU6xzPumBMOAd\nHiY8TZcKRl+4YAP7QyA6HmxcofZIqDYYrOxJToZ+/cBP2UthzmDrYMWAZS3Ydms/e4NPcPDqGQzG\nZHG1W7HErj2Lkmykar1aLPzhJ7wKPH9RbEmSpLyWY/Cl1+sX6XS6w4hBATXQSa/Xn87hNEl6aSwA\nliJ+eFsgAq6OgPMzXm/Tub0AtNAFcC3kLn/OP8o/S06ScF8EAU5udjTvV5WW71WncCm35+5/XjEC\n3RFpSositiwXyPYM4VpqBHuSjqIA5az9qGNbmTOJp3I4S0FkpBmGSJaqPHKsOBhaw5HrcHCjeElr\nD7U/glrDwfZhryZOhKhz56jBTlCBsX8M/9s//rE76TyKU9+vGqVsPBmzazi1atVizpw5WFnJhKeS\nJL2csgy+dDqdGuiP+AM5WK/Xz8u3XklSHklFJCwACEJMLz6PqMRYgsNOokbN0VHXWbb18INjujpF\naD2gBoFdy2Nj93L94lcQi+r/QKxb2wrkdk9iKkY0aKhio6OWbYUcWociMob9hJjYTGcH9ISkzrD/\nTzj5A5hTQa2Fyu9Bnc/B4WHC0sj4aH758yJzJ3lTh/UAxLSOIK7oPRys7ahTvJIoUl2iKjE3Iyld\nujRqtZoqGzZQqFAhtFpZR1GSpJdXdv+Hmg9UQexGH6XT6XR6vf6LbNpL0ktnOWJxeVlE4tDn9Zd+\nP0azCfsQJ/Rbb2Fjb0WjXhVpPaAGJas9zfqn/DUG+BkRAv2BKJidE0VRUKlU+Fn54Obogos6qwXq\nV4GViIndY08ccwfGQXIvODIbjnaF1ARABeV6Q90J4OqHyWzm9A09e6+cYM/l45y/dY3w2aOoqfyO\nBiOqwGQ6fOJPg5I1qFZE96BI9Y4dO3jnnXfo1q0bkydPpnDhvC3zJEmSZAnZBV8NgPJ6vV7R6XRT\nEYl4ZPAl/WuYERmjQORIf94VVyaTmUVbNoEV2B1xodX7NXh7ahMcXPI/TcTTmAtMQiyqXwkE5OIc\ns2JmS+I+fK18KGftmyGFhFZ7F/gWEXAFZ3IFZ2AqpPaBE9/DodJgSKu76NcOAicR5VCc4LCT7N37\nB8Fhp7hviH9wdsyOjlS6HYwtcfj5ezNze1+srB/fFvDXX3/x3nvvkZKSgqIoD4JFSZKkl112wZdB\nr9crAHq9/p5Op1OyaStJL50NiIXlRRHrvJ5Hwn0D4/ovJaLOHVTJaj58tzMdBuYmjHmxViFKU4OY\nCGyby/OCDScJN0YQaYrGz6oINiprxD7JNcAKKlfegQhv4WEp7FREiPshmEfDmbWwvyzEp00/+jSA\nwCkcx51vt6/g2PUQlEfWgvm4FKK+X3WKpNZmzoizuBCBi3cBvvyjZ4bAa9OmTQwcOBCj0Uj//v2Z\nMGGCDLwkSfrXyC74ejLYMmfaSpJeQgoPt+QO48lqf0/n5sUovnw9iDM+Ir1dvcJV6dD35Q+8dgK9\nEZ/FJKBvLs87m3yJMykXUaPmNfvq2KjSpxS3kp5lRlG0qFT+iBLf19PObA7KLAg9B/vqQvQF8XKh\nahA4mQi3Gny9ZzmbQ8RImZVGSw2fcmmJTqtSvIA3ZrOK5sV24KmEoLK2Yer27rgUfDzbfEREBIMH\nD8ZoNDJgwABGjx4tAy9Jkv5Vsgu+iut0up+zeq7X6/tZrluS9Hz+Qey1K4hIcvCsTmy/wtSuvxMX\nnURKlzgA+jRplQc9tKyTQHtEIe4PgM9yed6Z5IvsNYi0sw3squGl7QAcTTuqQewXbUps7EZcXfem\nve4HykwIs4G9b8KdtLS1BUpD3S8x+LVn8ZFNLFr/CQZjCjZaK96u9Tpv1WqLg/XjBZM+e/M09jf3\noqDi0xWdKVouY6oILy8vZs6cycWLFxk2bJgMvCRJ+tfJLvga+sTzXZbsiCTlpfRRr48RdRiflqIo\n/DH3MD8N+QuzSaFUz4Jcdz6Hh0MBahXNacffixWGSKIai9jd+Q25y14PkKSIdBl1bCtT1vokIvDy\nQizZbwUsBMbi6poMOACfw80A2DMOrqf9L8KxMASMhwpvk2Q28+7KiZy6pUMNbAAAIABJREFUJUbB\nWujqMKRBLwq7ZAyqtq28zullG9EA9d97jfodSz52fPny5bi5ufHaa6/RsWPHXH8ekiRJL5ssgy+9\nXr8kPzsiSXnlMCKdpxNi1OdppaaY+H7QZrb+JEZwun5WjxuNw+AktCpXF436xSdLzUok8BoQATRC\n5DfLKXu9oijEK4k4qR2oaVOeolovvLTuiEwzIMbNPBH58MX6rXv3WuOufAJ7v4FLaeNqtm5Q+zOo\n+gFY2WFWzIz+41tO3bqAt1NBJrX+gBpFM99nGXElhtm9V6LBhE2Zmoz8vtZjxxcvXszo0aOxtrZm\n586dFC9e/Kk/G0mSpJeFTIYj/eekj3oNQBS+fhr3IxOY3Pl3zu4Jx9pWy0c/t6Nut7I0+345AG3K\n18/LruapBMSC+lBEjph15Jy93qwo7E46whXjTdo7NMZN45wWeB1EJEd1RWQE65x2Rk2I+RQ2/wy3\nmwEKWDlAjSFQcxjYuDy49uxdv7H9wiGcbOyZ13kkJQtmnlksMS6ZIYFBqFMTuG/ly+rdrz12/Kef\nfmL8+PEAfPbZZzLwkiTpX08GX9J/ynlgLaJc0JCnPPfKqdt8+XoQd67ex62wE5+v60aZWoXZdeko\nMUlxlHT3Qefxcv7iTwW6Ita5lUBkr3fJ7gQgVTHyT+IhrhhvoEVDgjkJN0163v+ZaV97IdKzAgkf\nwgEjnOqJuzkV1FZQ5X3wHw0Ono9dO+j4X/xyZBNatYaZ7YdmGXiZTGa+6LCWuJt3iMedAT91ppDn\nw5HF7777jkmTJgEwadIk3n777Vx+IpIkSS+vXAVfOp3OASiJqIhrr9frEyzaK0l6RtPSvvZFrFTK\nreC1Iczqsw5DQiplahfm83XdHhTATi8n1LZ8/ZdycbeC2FSwGbHBYCuQU7rXOHMiWxP2cdccgzVW\ntHIIxFtbMO3oFUQJci1wHAx34UgJOLoAjImAinuebXBvNxdcSmS49u5Lx5j6zyIAxrV4F/9iFbPs\nx6IR2znzzwVSsMO95Rv0fOvhAnxFUdDr9ahUKqZNm0avXr1y9XlIkiS97HIMvnQ6XVPgB8TSkQDg\njE6n66nX6/+ydOck6WmEA78hMk0Nz+U5iqKwYuIefhsrFos37l2JwT+1xdpW/NOIS05k56UjALQu\nVy/P+5wXRgK/IDYWbALK5OKcw4Yz3DXH4Kx2pJV9PQpoHq10+Q1gBnMFOBIMh1VgCBOHSnWAehMJ\nu2rAPZPA6/ztK4z44xvMisJ7AZ15vWLDLPvw18LjrJt1ADNqQh27cHCxqIOpKAqxsbG4uLgwa9Ys\nunfvTt26dXPzUUiSJP0r5Gbl8GQgEIjR6/URiMz307M/RZLy3wxEFqo3AL9ctDckpjKt+2p+G7sL\nlQrentaUob+0fxB4AWwPPUSyMZWaRcvj5Vwwm6u9GF8DXyH+ivodqJ1D+1RF5OkKtKtGeWs/Ojk0\nfSLwigZlgXi4+rQoLmZQoGgj6LEf2q+Fgpnv9oyIvcvgNV+RlJpMm/KBDKibdSXN07uuMu/9PwE4\nQ2ve/awEnp4i8Jo8eTKtW7fm1q1baDQaGXhJkvSfk5tpR7Ver4/Q6XQA6PX6c+mPJellEQmkhQyM\nzE37a/eZ2H4ll45HYOdkzYjlnajVpnSGdpvO7wGgTfnAPOtrXlnBw3VtCxGJILJiVswEG05y23SP\n1x0aYa2yooFdjccbKWaIHAiFEkS5xnDA0wsCl0Dx5pDNlGt8ciKD1nxFZEI0NXzKMb7Fe1lO0d68\nGMXkTqswGc1cpg4xTtUYOFAEXuPGjWPhwoVotVrOnj2Lt/fLWy9TkiTpWeUm+Lqu0+naAopOp3NF\n7N4Pt2y3JOnpfAMkIXb7Vcqh7fn915nUcSUxtxPwLlmAMRu6U6x8xrxTt+PucTj8HNYaK5qV9rdA\nr5/d38CbaY+nPfI4MwZzCtuS9nPDeAc1Ku6YoiiiLfSwgaLAlc0QPBLanxav6YF2ZaD0aVBlXx8g\n1WRk+MZvuHA3nBJuhZnd/hOstVaZto2PMfBFuxXERSWh8izN+dtNGfkBODubGTVqNL/88gvW1tb8\n8MMPNGvWLJefhiRJ0r9LboKv9xC/24oClxEplN61ZKck6WnEIopHA4zKoe3fi08y971NGFNMVG5S\ngpErO+Psnnka1s3ng1FQaFiyOs62DnnZ5edyDOiI2OH4Mdmvb4syxbIlcS+x5gTsVDa0sK/7yMJ6\n4Ppe2PsZ3NgL5RDJ0QxAM2dQ/0VOhZkURWHK9p8JDjtJATtn5nX6FBc7x0zbmoxmpnVfzfWQe3iW\n8mDJxY7Y2qr5+GOYPXs2v/zyCzY2NixYsIAmTZo8xSciSZL075Jj8KXX6+8APfKhL5L0TOYD94GG\niB0hmTGZzCwasZ11sw4A0HZQLd6Z1RytVdYpSB9OOb48ub0uIaYX4xH/KGeSdfZ6RVHYnniAWHMC\nBdWuvOZQDyd1WqB55yTsGw2XN4nntm7QSANEpiUHWwTknFZj0aENrD71DzZaK+Z0HI6Pq2eWbX/8\neCvH/7qMi4c9iRXewHTRhvf7g6cn9OzZk82bNzN27FgaNGiQuw9DkiTpXyo3ux2vkLHINnq9Pjdr\nmiXJopKA2WmPs6pfGB9jYHqPNRzdcgmNVs2Aea1o+W71bK8bGnmV0MhwXGwdCfStmpddfma3Ednr\n7wDNgMVkvmNGURRMmNGqNDSxr82p5AsE2lXDSqWFmEuwbyyELEckSHWEGkOhVnmwfiPtCu8CnXLs\nz9aQ/XyzZzkqVExuPYjKhTOumUv3x7zDbJp3BK21hr7zutGyhytabSp+fr9hMvXB29ubrVu3otHk\nlI9fkiTp3y83046NHnlshZjxsLFIbyTpKS1GBCXVECWfn3Qj9B5fvh7Edf09nN3t+Gx1Vyo1zHlE\nJz23VwtdHaw0Lz4XcRzQGjHyVR1YQ+YTgrHmBHYmHsZZ7Ugj+5q4a1xpbF8L4m/BgS/h9E9gNoLG\nGqoMAP9RYO+ByGIPoozQNzn250LMdWbu/h2AIQ170qxM1mvijm+7zI8fbQXgo4VtWflPUUymFKpW\n/YCZM/8kMlLPlClTZOAlSdIrIzfTjlefeGm6Tqc7Aky0TJckKXeSeJjz5DMyTr+d/OcKkzv/TkKM\ngRKVCvH5+m54+RbI8boms5k/z+8DXo4pxxREcZ9jiBQafyKWZj1KURTOpVxmv+EkRkxEm2NJMidj\nl5IIh6bB8TlgTAKVGir0hbrjwDk9CP0cuJn2eC05FSUKj47g2zPrSDGl0rVKM96s2TbTdvHRSZwP\nvs6MXmsxmxS6japH2aaV+bl/Mm5u73HnzjZcXFzo1q3bs30wkiRJ/1K5mXZ8dAGGCqgA2GXRXJLy\nRQqinM4VQEfmk2Rz3vmDhBgDddqXYejSDtg75W7A9sj1c9yJj6KISyGqFs5NylLLMSOy9W8DCiGy\n1z+5qirOnMjOxMPcMN0BoKSVD4FaHXaHZ8HhryA5RjQs3QnqfQnu5dPOPACMQeydBGhM1qvmhJik\nOD5YM5X41CQCfasysmlfVCoVcVFJXDx667H/bl+JeXBe3c5laT6oMR07JuHo+A62tjtxdXVlxYoV\nVKqU0/5USZKk/5bczKdMeOSxAtwF3rJMdyQpZyagDyKbuzuiEE5mE1axdxMBGLKkfa4DL3g45dim\nXOALLyc0HFgGOCJGvEpl0saoGIkw3cVWZU2gdWVKhWyGA50gIUI0KNYUAieDd21EOLcJkZp19yNX\nKYUoxZ21ZGMKH6+bQXh0BEWNntQNqcNX3dZw6WgEt8NiMrS3ttXiW8WTSo2KU759ffz9VUREnMHD\nYx+uru6sWrWC8uXLZ3InSZKk/7bcBF9Ber3+e4v3RJJywYxYDr4SMfW2FTEUm52nCaAMqSn8HXoQ\nePGJVWcAsxALLdcA6SlRFUUhzHiTa8YIGtjVoIDGmeZ2/hS6sgv7Pc3g/mXR0LMm1J8CxZshxgp/\nQQRdZ9OuZJ32ui9iUvPJycyHzIqZsVu+5/gNPZ43vLBb7MnymD0PjtvYafGt6kWpGt6UrC6+Fivv\ngUarZs0aaNzMTGIiBATUYsiQn6hSpThlyrzYUUVJkqQXJTfB1yBABl/SC6cAQ4GfEfPem3gYkOSV\nXZeOkpCSRAWvkpRwK5zHV8+9pTzM37UEaJ72+K4phuCkE9w0RQLgpy2CT/gRSuwdBXfTEqS6lYXA\nSVCqI6jiESHcbOB62lWKAE0RwZg1Yuww68ALYN7elWw5H4zbPk9s/yhIkslI5SYlaPpmZUrW8KZo\n2YJotI/vvVQU+PJLGDcuFje3twgM7MOGDZ2wsWmexV0kSZJeDbkJvq7pdLp/gIOINc4A6PX6LyzW\nK0nKxDjEPjwrxLJwSyyF/+Pciy8ntA7ol/Z4Fg+T7J1I1nPAcAoAG5U1tZJsKLy5C9wQmwNwKgp1\nJ0D5PqC+h1hI/x2QPiVYHhiBWNeVXgVyBmKvaNbWnt7Bwj3rcf/dB/vjLphRCHzbj+ELeqDRZF4e\nNjER+vWDlStjcHfvhbX1CeLibqMorclpQb8kSdJ/XW6CrwOPPH6xC2CkV9Z04EtEXqsViHxXeS0q\nMZbgsJNoVGpa6rJfeG4p8xFDzWbEyFd67cY4cyJHDGK6sKLRlVr75mFzYb04aFcQ/EdDlfdBey3t\nCouB5LSz6yOCrtaIFXMNEGlp26e1zdr+sFNMWbGEQot8sb5lh62DFUOWtMe2RGKWgdeNG9C+PRw/\nHkWhQm+g0ZylWLFirFy5EltbGXhJkiRlGXzpdLq39Hr9Er1ePyGrNpKUH1YXLMiUtMeLyE36z2fz\nl34/RrOJQN+quDu4WugumVMQ41ST056PB8Yi1nepVCqc1Pa8Zi6OIXQFpffOEI2snaDGJ1BzKFiH\nAL0Rq8MUxN9JHXg40pVuFOLvqaKICdys/566EHmNkV99j/uS4mgStRQu7cbotV0pXqEQR48ezfSc\nQ4egQweIiIilcOGuKEoIvr6+rFy5ksKFX9w0riRJ0ssku5GvjxDLTSTphfkNmFqsGADzyL6A9PN6\nsMsxn6ccU4D/IVZgaYAfgP5AqmLk78QDFDXZUvHgTxQ9vRAUE2hsoOpAqD0S7I8CrwM7065mjdgL\nOgwo+8SdtiLKcGsQeyjdsuzTnbgoPnzvWxzXeaFSVNRqW4pPlnbE0TXrkatly8RUY3IyNGzoRPXq\n/hw7ZiIoKAhPz6zLDkmSJL1qXnzqbknKwnpEThNFpWIqMNCC9wqPjuDUrQvYWdnQqGRNC97pcbFA\nF0QeL3tgFWJy0GBOZkv8LiKU+9xOiqV0yK/YoEDF/hAwCpyDEUWG0hbZ4wwMAD4EMhthuoUIykBk\nj8k6wIyKiuV/Lb9Bc1gswu/6eT36TGiMWp35KJnZDGPGwOTJoFbfpFcvA4sW+aHRTCQuLg4XF5en\n+kwkSZL+67ILvirodLrLmbyuAhRZ21GypL+BbogVSn1v3eJTb2+L3u/P82LUq2np2thb58+6pFuI\nQOsEIoHqJqC6onAu6TyHkk5h0GpwiL9L2y2TsCnRGuqOBPe9iGSo4WlX8UasDHsXyCrISc+MFgk0\nAUZm2aebl6L4oPF3mK5pUWzNDPmlPc26Zl3bMi4O+vSB9evByuoapUt3IyQklYiItRQtWlQGXpIk\nSZnILvi6iPjdIEn5ah9iKXgKMBh48+ZNsGDwpSgKm0OCgfybcjwPtAKuAqWBLYCfKYV7IT+zu0gB\n0GrwvnWWJhdP4tTuZyi0F5FwIjrtCmURS/J7kXOp1WnAdsAD+JXMU9KCyWRmaNufSL2mYPJKYfy6\nHvj7Z50ENSwMXn8dTp+GAgXC8PHpxr17NyhRoipOTtmnrpAkSXqVZRd8pWRS11GSLOoYIuJPBN4G\nvgaOW/iel+5dJyzqJq52TtQuVtHCdxPBZTtEGOUPrDTGk3B9A2wbjXtsGNVrdMfNbE3Jsi1Qtb2D\nCEUNaWcHAJ+mXSHz3YYZ7zY27fFSxEhZ5r747BfiQ1IwuaQyelsX/CtmHXidOOHAyJFw9y6ULn0R\nO7vu3L0bQc2aNVm6dCnOzs656JskSdKrKbvga1++9UKSEKNBr/FwHdRP5C68yMy2RScwxKeg0arR\nWGV/lb9DDwHQqGQNtOrMR4XyyhqgJyIJxOuKwtTra9luG09yATvcbDR4upentl8tKHoIVG8hkk6A\nCLY+Beo9xd2iEFnCTIhdj1kn6Fi/Zw+H54ShRkObL6sRWLFKlm1//hnef78MRiO89hq4uIxn794I\n6tSpw5IlS3B0dHyKPkqSJL16sgy+9Hp99gmAcqDT6dSIDI9VEL9r3tHr9RczabMJWC9LGL3ariCW\nj98FWiJ2OT7LbhBFUVgxcQ+/jd0FwBtj6mNjZ5XtOdsviHJCzcr4P8Mdc+9bxBZiBZh57R8qxe1j\nV5GygB1Fb4Vg16g7FD0AqmFpZ1gh9ncORyRIfRoKYs/kNcT42sQsW565dYm5H/yBTbIjXg2cGDS4\nY6btjEYYMQJmzwZQ8/HHMH063L//DTNmzGDMmDHY29s/ZT8lSZJePZbc7dgBsNXr9QE6na4OMBMx\nf/KoiWS33116JdxAFLu5iUj/uRqRMOFpmYxm5g/azJYfjqFSwftzW9FmYPY7F8OjIwiNDMfR2g5/\nC005mhFL3KcD1W4fY0XwGA7VbsulImXRphqomxJKuTIbUalPpp3hBLyHCNV8nvGu8xC58l0QaWkz\nD0Bv3L/DkPFzsDldELWdiim/9s203f378MYbsGULWFlB//5bMZv/xGyejru7O1OmTMn0PEmSJCkj\nSwZfgYh1xOj1+gM6ne6x34I6na4L4vfSZgv2QXrJRSKWkV8BagEbESkXnpYhMZXpPdZwcEMo1rZa\nhi3rSN2OT+a5yii9iHaDktWx1mY/QvYskhGlgo5E6QnaN4ZuoasAuO/hxo2y/jQp8DPOmvRVbZ7A\nx8D7wPMkeT0OfJL2eAFQItNWsYYEPvh1GpogsSOx35RmeBTNuDvxwgWxsD4kBAoWhGnTjjF16mAS\nEhIoW7YsAwYMeI6+SpIkvXosGXw5I2qYpDPpdDqtXq836nS6ioilL114uBo4R1ll1c5L+XEPSYhX\nq3lPp0Nvb0/JpCSm6PVcMJkybZvd9yUxJoVlQ49w7VQMds5W9JxVA5tiCbn6Xm44sQMAX03BPP/e\nx6vVTPe0pf2ZWXxzbSs7Gg8m3HgJtzJGqpT+gxpWv6FWKRgMxYiIeJOoqFYoig1w6ZnvqVYnUK5c\nH2xtU4iM7Ex4uC+Q8X0ZzSZmn/yd6OUmnO5b4V3emSIB2gyfwaFDTowc6UdsrJaSJZP44IN1TJ48\nhqSkJAICAqhWrZr8N/OSkd+Pl4/8nrycXuT3xZLBVyxi/iSdWq/XG9MevwkUAf5B/FmeotPpwvR6\n/ZbsLlijRg1L9POBo0ePWvwekpCAWP6tB0oBu+3s8K6aeT6p7L4vt8NiGNdrGdf1MXgUc+GLLT0o\nWs4jV324FXuXKzsisNXa0KdpJ+ysckrZkHu3Eu/y16Gp/L5rLld9/VnXYRJGrS3HmnekvcMYVCoQ\na7E+xdb2dUqU0FCiRF7c+U1EDrBKeHgsxcPDLkMLRVEYu+V7Lp2KpFCwL2qNis+Wdcevitdj7b77\nDj78EEwmMfI1ePBxBgwYS1JSEvXr12fp0qVYWeX9aKH07OT/w14+8nvycsqP70t2wZ0lg6/0HfUr\n09Z8pafiRq/Xj0h/rNPpxgMROQVe0n9HMtAR8QPig0io+ixZvC4dv8X41iuIjoinROVCTNjcE/fC\nuc8vtf2C2OUY6Fsl7wKvlDjuHJ2N45EZ9DSnsKdpf0KLNQKgtNUe6tv9hErVBrH7sD55W6t+CSKd\nhD0QBGQMvAB+PLCGDad24b26FCpFRYehdR4LvFJT4aOPYP588XzkSJg0CQ4eVGE2m+ncuTO9e/eW\ngZckSdIzsmTwtRZortPpghG/YfrqdLqhwEW9Xr/BgveVXmJG4A1EOZ1CiNSfxZ/hOse3XWZyp1Uk\nxadQuXEJRq/tioPL02WmT1/vlSe7HI3JcOp7Ug5MolBSJBSDI407EGrVCC3JBNotRmdVBJXqEFDp\n+e+XgZ6HBZjmAuUybbXp3F6+27cKpz0F0d60wbOEKz3HNXhw/N496NoVduwAGxtYsAAaN76BWl2E\ngIAANm3aROnSpTlx4oQF3oMkSdKrwWLBl16vNyNWDj8qJJN24y3VB+nl8y5iD54rIgAr8wzX2PHr\nKb7uuxGT0UzDHhX4eNHrWNk83Y/y3YQYTtwIxUqjpb5ftWfoRRqzCc4theBxEB+OtoyaG3VLUcTt\nItWV9SQkFaaSjStumkVA0We/T7YMiGJMiYillG9n2urItXOM2/o9migr3P/2xoSZgfNbYesg9pae\nPw/t2sGlS+DlBevWQUzMVgID3+fLL7+kd+/elC2b8yYGSZIkKXuysLaUb8KARYjJsC1A5ac8X1EU\nVk/fz+JPtwPQaVgAb09rmmXB5+z8c+EQCgoBxSvjaPOMualuH4fNveH+OagAkbWLsFf9PvfN3nQ1\nT8FB3ZeG9gOBAs92/VwbBpxCrJ77nsymMsOibjJk/UxSjUYq/12D6GQDDd6oQI2WpQD480/o0QNi\nY6F6dVGr8fjxjQwaNAij0cjFixczXFOSJEl6NjL4kvLNP2lfX0MsNX8aJpOZBUP+YuO3h1Gp4J1Z\nLWj/8bNPF6ZntW9WpvazXeDyn/B3V6iQiLGamovahuxN6ovRZIu9ykyCeQ8OassWAxfWIHJ6WSHy\neWVc8xaVGMsHq6cSa0igRkRV7hw24OBqy/++boGiwKxZMHw4KIqYcly8GLZuXcuHH36I2Wxm4MCB\njBo1Kh/eiyRJ0qtBBl9SvkkPvpo85XmpySamdV9N8OoQtNYaPlnanvrdKjxzP2KS4jhy7RxatYZG\nJbNPwpqpc5PA8Dn0BZNWw96kdwhJagpASSsf6tvWwFb9LGlin9ZVRBZ7EClcM+7cMaSm8NHa6Vy/\nf4dyjn4k/2oFpNL3q6bYuzrSr58ItgAmTIAxY+Ds2TMMHjwYRVH4+OOPGTZsGCpVXm4MkCRJerXJ\n4EvKFwrPFnzFRyexdNAhrh6PxsHFhtHrulG5UYnn6suOi0cwKWbqFKuEi91T1CFUTsCdXlD23IOi\nk1vNLQgxt8QREw3talDGqng+BSqpiLqNMYhNxR9maGFWzIz+cy6nbl3A26kgNQ7UZNeds5QPLEqV\nttVo0gSCg8HODn75Bbp0EedVqFCBDz74AFtbW4YMGZIP70WSJOnVIoMvKV/ogVuIHY65rVJ4J/w+\n41stI/xcNO5FnJiwuQclKnk+d1+2P5hyzM20pQLsBPNUUP8FnqCYYXt0Rz53GU0LdQ0+tU8mUUnE\nQ2PptV2PGgfsRyTrWERm67y+2b2cvy8cwtHajiGF3+bbxZvRWqlpPrQNdeqoCA8HHx/YsAGqVYPf\nfvuN6tWrU65cOUaOHClHuyRJkixEBl9Svnh01Cs3v9LDTt9mXKvl3LsRRyE/R6bt7Jtp6ZunFZec\nyIHw06hQ0aRUdlOOJkS2lK+Aw2KkKxVSzloz0msRfnbWvJsYy1sOZjRqGxzIuwStOdsGTEV0ahng\nnqHFyhPbWHx4I1q1hq9afsTS9sEAVOxQjw69PUhMhDp1YO1asbPxxx9/ZMKECRQsWJDdu3fj4vL8\nn7UkSZKUORl8SfniaaYcT+0MY2L7lSTGJlOxQTHajtflSeAFsOfyMVJNRmr4lMPdIbP6iUnAL8AM\nIG2HX5IGjpm4dsmHqW1/ppImBlAoo3me+ovPKgLojRiRm4BI1Pq4vZePM2X7zwCMaf4/wlfEcj3k\nHrYF3ZiyKhAz8Oab8MMPYGsLc+fOfVAYe+jQoTLwkiRJsjAZfEkWZwZ2pD3OKfjaHXSWWW+ux5hi\nol6XcnyytAOnz57Ms76k73JsmmGXYzTwHTAHuCNeMhZG2Xcf1YkEDnkGsKb9OCoRgxo1DexqUNa6\nRJ71K3fMiPJBd4DGQMYdiPo7Vxm+8RvMisL/6nSkpnVFBk36EYCdd9ugqLR8NQ2GDQOVCmbPns2M\nGTNQqVRMnz6dHj165OcbkiRJeiXJ4EuyuFNAFFAM8Mum3brZB1gwdBsA7QbX4p3ZLdBo1HnWj8QU\nA/uuiMzszUqnB1/XgNnAj4iKkwDV4WYDlFXfoTKmsMm3NYPaLObT5APYY8tr9nXx1Gac6rO8rxBT\njh7Ar4DmsaO34+4xaM00ElMNtC5Xj4F1uzIs8FeMKSauUYVkxxJsWA5t24r2ycnJ7NixA7VazezZ\ns+mSvuJekiRJsigZfEkWl9N6L7NZ4efhf7Nu1gEA+n7VlE7DAvJ8wXdw2EkMxhQqeZfC0+kWMBSx\nZiq93ntz4FO4bEBZ2w4VCvPrfcqsWl/yu8aK4tr62KpscFRnXjPRsoKBz9MeLwEKP3Y0ISWJwWu+\n4k58FNV9yjLhtfdZMP4Uofuvkow9icWasX8TVKwoktUmJydja2vLr7/+yuHDh2natGl+vyFJkqRX\nlgy+JIvLbr1XarKR2W9vYPeKs2it1Hy06HUa97JE7UP4O/QA1X3u8EXLS8CktFfViJQNw4FqmGMu\nkfxnTWxRmNH2Gxy9vPkx5Rw17KrAC1njBWLcsAdiE8AwoNVjR41mE8M3fI0+8irFC3gzu/0n/LYw\nhdVf/o0VoJRtzv499hQsKAKvsWPHEhISwi+//IKzs7MMvCRJkvKZDL4ki0oFdqU9bvzEsYT7BiZ1\nXMWpHWHYOVkzek1XqjbLbmLyWZlJNa2md82JVPS6m/aaHSJB6VDAF4Do1ESiN3SmeEocv7SejKOX\nyFDvqs7PnYxPUoB3gHCgNg+DxrSjisKU7YvYF3aSAnZOfNvxU2ZMcuKPSevwIQnrwr6sPlEJGxsw\nm8189tln/Prrr1hbW3P69Glq137GDP+SJEnSM5PBl2RRR4F4RAGd6389AAAgAElEQVRtn0dev3sj\nlvGtlxN26g4FvByZsLkHflW9LNADE9AMK81OKnpBXLItTjafAoOAgg9aHVcUwra9R+uYUFa3/ZIk\nz1JoUNPU3h8/K5+sLp4PvkOkvHDmRugPzBsQRGx0IimmFFJMqSSmGIhJisdLVZJiLoUZ9PkfxMQo\n+BCB2krLtztbY2OjwmQyMXz4cIKCgrC1tWXBggUy8JIkSXpBZPAlWVRmU47h5yIZ13IZkddi8dG5\nM2FLTzxLWGpKbyWwk4QUR77dU5pCjsPo59/zsRY/A6dOzOPr879ypUg1oguVwk5lQ0v7ei9kYb2i\nKNxLvE9k/D+U8RiCRg1LDndg7YzfMf6TcRTOGrEG7da1KADSE0W8+WUDipR2A2DMmDEPAq/FixdT\nv37GFBWSJElS/pDBl2RRTwZfZ/eG80W7IBJiDJQN8GHsxu44u9tb6O4m4AsA5u6twvLjPqx5u8GD\nowZgMHDuxj62Bo8GoEjVT2lqX4dCWjec1Q4W6pcIsKKT4giPvkV4dAThMRFcTXt8LeY2ZiWWFX22\noVGn8vtJP2btSsYpLgZXPEmtHY97Wxu8nAri6eSOVVQ5vp3sRXQMFCsGs2ZCKZ2G4hULPbhft27d\n2Lp1K3PnziUgIMBi70uSJEnKmQy+JIsxAPvSHjcGgtecZ3rPtaQmm/B/vQzDl3fC1t7Kgj1YAYRg\nSC3CyhOF8XUrTMmCYgoxDOgM3EiI4K9Do1jZ9Wsa37iBb9nulLJgj0CkvOj56yiuRN3Mss2UNqco\n4RZHRKw3UYlfMLl1cS5HRbF180l6NG7B22PEIvlFi+C90ZCaCi1awIoVUCCtylFKSgp//fUXbdu2\npWrVquzbtw9bW1sLvztJkiQpJzL4kizmACIAqwIcmHeYHwZvQVGg1fs1eP/blmi0eZfDKyMj6aNe\nf4U2w2g2PKjluBnoBcSZUvnj0GccaPgeZo0VYWWapy29t6zr929zJeomWrWG0h7FKObqRbECXhQv\n4E2xAl6Uct+Jg00QYIeX89+8GyCqYa5y3vfgGiYTjBgBs2aJ5x9+CDNngjbtX7TBYODdd99l+/bt\nfPHFF/Tv318GXpIkSS8JGXxJFpM+5ei1N5zvB20BoM/ERnQbFZgPRZuXA6Eoih9z9lgBBpqUrs14\nREimAL+dm8mVai1BpaaiqjD17PN3AbqvWxFW9JnyxKuhiB2YAN+SWRny5GRo1w42bxbB1nffwf/+\n9/B4UlIS/fv3Z9euXbi6uuLvn5sC4pIkSVJ+kcGXZDHbzQqoVdybGoy3RsWHC9rR7O0q+XDnh6Ne\nV6P6Exl/Cm+XQgwtVIKtiESvyy+tIKZYSQBqpThTvWDdfAgIhTtxYmG8Wv3kyJ8B6I7ItP8G0C/T\n85cuhX33wN0dVq+Ghg0fHktMTOStt94iODgYd3d3goKCKFeunAXehSRJkvSsZPAlWcSduGQO2Fmh\nMisUPnqTMRvfoGYrS6+mSrcMURS7JCtPeQIQVro2epUKd2Bj5Gnq/PEOOwPexNO9JuVLds2nfoHJ\nbObbvUEANClV84mjI4ATiCJMP/BkPYBLl8TXu/egQgXYuBF8n5gn3bZtG8HBwRQqVIigoCDKlClj\ngXchSZIkPQ8ZfEl5Ljoinv9N3IN5bivcj9xk+oY3KFOrcM4n5gkj8CUAZmU060MPA3CtjD+1gVkJ\nZ6m4rS8qYwKN7iWgqtU3n/olrD+zk5A7YXg5ufNWrXaPHFmHmGa0AoIA58fOmz8fFi4EHVCqJCwL\nBudHmiiKgkqlon379kRFRdGgQQNKlixp8fcjSZIkPT1LrniWXkE3Qu8xLGARZ4qJbFM9SxbIx8AL\nRMHpi5gpTa+IAOLjokhxdONN75IsSL7EaeM51jfoR6pXTVTN5kM+TTUCxCUnPhj1+rhhL+ys0nN2\nhfNwinEa8HBELDUVPvgABg4ERRGvder0eOAVHR1N9+7dOX78OAB9+/aVgZckSdJLTAZfUp4JOXCd\n4XUXcTsshvh2pQF4vUB+FqFOJX3Uayxj2Bl6FAD/0rUZbrxJcJJ4XiVkO1ZtgsDKUvnFMvfT/jVE\nJd6nahEdLXXpubaMQE8gGmgLfPygfVQUtGwpFtRbW0O3tNnRR5eK3bt3j27durFv3z5GjRqFkh6h\nSZIkSS8tGXxJeeLgxlBGN1lK7L0kKnQrR2TZglgDdfO1F0uBy1ygDFOVNyh44RAAb5Yrz98J+1FU\nKmocW0Wlku+AqyVqSGbtavQtfju2GRUqPm381iOL+8cjsqEVARaRvs7r/HmoXRv++Qc8PWHXLqhW\n/fFrRkZG0rVrV86dO4efnx8LFy7Mt00DkiRJ0rOTwZf03Lb8eIxJHVaSnGSkeb+q1F7WCUWlIgDI\nr7ElE6lEMRGA8YyleeRN1DG3cbN3Ico+ErMKKp7dTE3biuDbKp969dDMnb9iNJtoX7Eh5b3SA7+/\ngcmIf4a/kV5rcvNmqFNHLLCvVg0OHxbPH3Xnzh26dOmCXq+nTJky/P777xQunJ/Tu5IkSdKzksGX\n9MwUReG3cTuZ+94mzGaFHmPr8+GCtuzUiB+rJjmcn1fuAt+wBDeucJ6yVOEN2oQeBKCxX1XabJlM\njWOrqHc7ElWdz/OpVw8Fh51k16WjOFjbMbj+G2mv3gb6IDKOjQUaoigweza0bQuxsdClC+zZA0WL\nZrymk5MTnp6elCtXjlWrVuHp6Zlv70eSJEl6PnK3o/RMTEYz897fxF8LT6BWqxg4vzUt3xXzYpkV\n07aUQ0APUtieNuplZiwj0NDxwiG8PN1pGr8fuxvB1EooCb2OgCp//95INRmZvmMpAP+r05GCDq6A\nGXgTiAAaAp+TnAwDBohyQQDjx8OYMY+v7wJI0cSRlJKAnZ0dixcvxmAw4Obmlm/vR5IkSXp+MviS\nnpohIYWp3VZz5M+L2NhpGbGiE/6v6wARTpxDTDdaOl/8SqA38BZLKMFVUilHBbpx5d4Nos336da2\nGTeirmKMcEb7+hqwdbVwjzJadfJvLt+7TlFXT3pVT5/unA78hZhm/I07dzR06gT79oGdHSxZAl0z\nST0WGRNBqPt6lmw/QO+4xjg5OWFvn7+bBiRJkqTnJ4Mv6anE3ElgQtsVXDh8E2d3O8ZsfINyAT4P\nju9I+1ofsLZwX0QO+xSmpo16WTEW0LDtwkEaBlZHrVbjFXkBbdP54FHZwr3JKCYpjvnBqwD4pFEf\nrLVWwH5gdFqLxZw6VYR27SA8HHx8YP16qF4947V2bznErOWfk6qNR6Xyzq+3IEmSJFmADL6kXLt1\nKYqxLZdz62IUniVcmbClBz66go+1yc8pRwPQl0W4E46ogSiGi84ZLlDWyw9NYiz+yU5Qrmc+9Caj\n7/atItaQgH+xijQqWQORTqIHYAKGsm5dG3r3hoQE8PeHtWvB+5G4SlEUzuwOZ8EXa1h3dh5GTRKO\nyd5MHDkbJyenF/KeJEmSpOcngy8pVy4cucn41su5H5lIyWpejP+zBwW8HDO0y8/gy5pkRjMp7dk4\nQMPFqDB8y4uRuMCQvdg0+CEfepLRhchrrDq5DbVKxfDGb6Xlcv0fcBVFqcmUKdMYnTYA1rs3/PQT\n2NqK5yajmX2rz7N2xn5Cj9wkxGMlRqskfD3KsWTpYkpW8snirpIkSdK/gQy+pBwd2XyRqV1/x5CQ\nSrXmfny2ugv2TjYZ2oUBlwEXoFo+9KsLP1OMayRTERu6ALDnwFySK1XEFBlB2ZqTQGPpyc+MFEVh\n+o4lmBWF7lVbUNqjKDAfWE1Skgf9+//D8uVaVCqYMgVGjBCJ9pPiU9j28wnWzz7I7bAYAFwKOjCw\nx6eEJgbz9ZxZ2NnlZ9JaSZIkyRJk8CVl6+/FJ5nzzkbMJoXGvSvx4cJ2WFlrMm2bvt6rEZB5i7yU\nzAAmAxDFOLxRg34l/5w7wTn9LSbWa47K6cXkvdp56SgHw8/gbOvAwHpdgVPAEG7e9KZDh9McPuyE\noyMsWwbt2kHUrTg2fnuYP+cfJSHGAICdXxIF/VOYveBTbO2tEWkpJEmSpP8CGXxJmVIUhZWT97L0\n850AdBlZl7cmN8k2g3p+TjnCAry5zkkq40AnjHdPc+DaGs5pPNEqCg2rvpHzJSwgxZjKzJ0itcSA\nul1xtVMD3TlypCLt22/j5s0ClCgBGzeCszqSb/ofYMevpzGmmAAoX68oZTs4MPXH0STsSqDltvq0\nb9/+hbwXSZIkyTJk8CVlYDKZ+X7QFjZ/fxSVCt6d05J2g2ple45CfgZfBkgb9ZrAOKYnx3MydC5n\nanWmhectDJdSsbe2tXgvMrPh7C6uxdzGz60IXas0A94lKKgyb7+9BIPBlgb1FSYOvcrqkQc4vOkC\nIKYc63YqS8dhAUQr4bz55pskJSXRoUMH2rRp80LehyRJkmQ5MviSHmNITGVGzzUcWB+KlY2G4cs6\nUrdTuRzPCwVuAh5ABUt3kgXATc5TmXVKe77b8RY7arUE4MSpUD6o1t3iPciMoigsO7YFgHcDOqFR\nBTFmjC8TJ45BhZlejc/gfv8A0zreAsDGTkuzvlVpP8SfwqXc2L17N3379sVgMNC1a1dmzpyJRmP5\nCVxJkiQpf8ngS3og9l4iX7QLImT/dRxcbRm7sTsVAovl6txHR70sW9rZAEwBYA7jGXFoOie9vDBq\nbbh46Rp3bkfTwC+TRFn54GD4GS7du46HQwHqFomha1dH1q/5FF/VAaoXOETMjvvEAC4e9rQdVIvW\nA2viUvBhktTr169jMBjo2bMn06ZNQ/1kentJkiTpP0EGXxIAt8NiGNdyGdf19/Ao6syELT0pVt4j\n1+fn35Tjj4gxtqokXHVkZFgQm1uOApOZ4IOnqVO8Es62DhbvRWaWHdtC8QKxfFY7lsYB7iSeS6QZ\nX6NVkjFEQZEybnT8JIDGfSphY2f14Lzo6GgKFChAz5498fX1xd/fXwZekiRJ/2Ey+JK4fCKCca2W\nEx0RT4lKhRi/uQcFizg/1TV2pn21bPCVRPqoF4kD+fqPHqQ6FcAlKYnz1xJISEiiaRlLFzXK3PWY\nUGoWXUijuOJ8XLsfXobtqDEDUD6wKJ2GBVC7XRnU6sfHBTdu3Mjw4cP5+eefqVu3LgEBAS+i+5Ik\nSVI+ksHXK+7E35eZ1GkVSXEpVG5cgtFru+Lg8vSL1aPSvvrmae+e9AMQAUo1WPM9boZ7bPaqRVmn\n15m5dxAalZrGJWtatAcZxaMoS7lxaB67h3cg4pQDHoSiALXalaX7qADK1sk8KeqaNWv46KOPMJvN\nBAcHU7du3fztuiRJkvRCyODrFbbjt9N8/fYGTEYzDd6owJDFr2Nl87L+SCQCU8XD4wWJjzvK3gYf\n0r/CKL64copUk5HaxSpQwP7pRuyejQIcxGRcwN5Vp1gzowaXjolF/ia0FKhUlWmr/Cmqc8vyCkFB\nQXzyyScoisKQIUP45JNP8qHfkiRJ0svgZf1NK1mQoiisnr6fxZ9uB6DD0Dr0m94sw5TYy+V74DYk\nFocd29jf9BPCfP1pRjgHQg8C0Ky0v4X7EAksJTFuCdsWOrBudgCR4e0ASMaeq6qaDJ5Ri8FD7bO9\nyq5duxg6dCgAI0aM4KOPPrJwvyVJkqSXiQy+XjEmk5kFQ7excc4hAPrPbE7HoXVecK+yp1IZgGni\nybYbXC5Rm0u+/qSiYbPaj+Jhc1Choknp7HORPRsTsA1YyL2b29n4bQ02f9+GhBhR5ifF2o2QlAAi\n7HyZ8kMEg/tkH3gBBAQE0KJFC/z9/Xn//fct0GdJkiTpZSaDr1dIisHIzD7r2Pf7ebTWGoYseZ2G\nb1R80d3KkYfH78AduGNN0nVbdnf/EIDdtpVIvaIn1ZRK1SI6PBwL5OFdw4BFwCKunjWwdkY9dv72\nIcZUkXerSKWibL8SwKX4Mth4XqfOoK/5qNfYbK8YFBREs2bNcHd3Z+HChXJHoyRJ0itKBl+viPjo\nJCZ2WMmZ3eHYO9vw+bpuVG5c4kV3KxcS8PJaIh7uTWFf488xWFlTRFOI49alcAndBECz0nmxyzEZ\nWIdI4rodUAheW46pXftjNqlRq1XU61IWVckAxswsgtEI3lXP49b5K/q16IRWnXVC1Dlz5jBt2jQq\nVKjApk2bsLKyyrKtJEmS9N8mg69XQOS1+4xruYzwc3dxK+zEhM098K3s+aK7lUvfYWUVLVJ73fag\nZrMeGDQ3aWhXE4ypOF05DkDT5wq+TgELgV95uG/ThkvHezGztx9mk5lmfavQaUQgX811Y17aDKhn\n440UbLkMDycXOlVunOmVFUVh1qxZzJo1C5VKRb9+/WTgJUmS9IqTwdd/XNjp24xrtZx7N+IoVr4g\n4zf3pFAxlxfdrVyKB+NE0ELqIVu0bVfi6lyKtpQCQHPpMJrUZEp6+lHYJfcJYR+7Ph0Qo1zpqgH9\nibrVni/a/U5yYhxN36pMn+nt6NYd/tkOKm0qPp1/xK3mXjpVbswH9brjbOuY4eqKojB16lTmzp2L\nWq3m66+/pnPnzs/QT+nf6ODBg3z88ceUKiV+XhMSEvDx8WHGjBlYW1sTFRXFtGnTuHnzJiaTCW9v\nb0aOHImHh/hZPnLkCPPmzcNoNJKYmEinTp3o1avXY/e4du0agwYNomzZskybNi1DH65fv87QoUNZ\nuXLlY6/PmDEDPz8/OnXq9OA1g8HA8OHDuXfvHg4ODkybNg03t8d37KakpDB69OgXXoHBbDYzfvx4\n9Ho91tbWTJw4keLFiz/WZuLEiRw7dgwHB5F0+bvvvmPOnDmEhIQAEBkZibOzM0FBQYwcOZIJEyZg\na5sxzc7EiRN555138PLyAv7P3nmHRXF9f/hdlg4CAopUA9g1FuyKaGyxxKgoYAOxRJNYokbF2Auo\niEajRpTYsYJiwZpYYlcUrFhRxILSEVj67vz+WJm4LsYS/Wry2/d5eNiduXfunTuzu58599xzEP00\nly9f/iFP8R8zYcIEOnXqhJub21vVCwkJ4fTp02hpaSGRSBg9ejS1av1z95RX3Yvvm0uXLhEQEIBU\nKsXV1ZXhw4er7A8ICODmzZtkZ2eTn5+PiYkJYWFhrF27lr17lTMpLVu2ZPjw4dy6dYs//vhD7Rjv\nBUEQ/hV/Fy5cED40/4s2/pdcPhoveJoGCp2ZKYxzXSNkpeV+sLa0BOWFKn6fB80ZIwgCguIxwu7E\n9cJB2WkhT54v7q6+d4lQO8hLCDy78x0bmCgoe20qCML3giBEC4IgCPm5hcKohivFcbt6uUiw/6xA\nAEHQNs4QnIdNFnw3TxOuP733t0dPS0sTGjRoINjb2ws7d75rHz9N/muflQ/B2bNnhVGjRqlsGzNm\njLB//35BoVAIvXv3Fv744w9x36lTp4Tu3bsLxcXFwoMHD4SuXbsKKSkpgiAIQl5enuDh4SEcO3ZM\n5Xg7d+4U5syZI75/+bo8fPhQ8PDwUOtbUFCQsH37dpVtq1evFhYvXiwIgiDs2bNHmDVrllq9FStW\nCIcPH36T0/+gHDx4UPDz8xMEQRAuXrwofPvtt2plevXqJaSlpZVav7CwUOjZs6dw8+ZNQRAE4fjx\n48KSJUvUyl28eFFlfBMTEwVvb2+hV69ewoMHD96orx/rs+Ln56d2v7yOO3fuCF5eXoJCoRAEQRCu\nX78udOnS5b3051X34vvm66+/FhISEgSFQiEMHjxYuHbtWqnlzp49K94DDx48ED97crlc8PLyEm7c\nuCEIgiCMHTtWSEhIeKe+PL/2pWoajeXrP8qJsFgWeO+iuFBOM/dqjN3YHV39f9HlLkwE6S8ARKcM\n4vFn+ugXJ6NAIB8YLS9G624MAG3fKap9PLDg+esDgHLFp0IhsMh3N3fOJ2LlaEa1gV/SoJmcApke\n+jbxNBy+iondO/Fl1aZIJKWH5ij5cJmbmxMWFsadO3do3779O/RRw/uiM7DvPR+zE7D3LcoXFhaS\nnJyMqakp165do0yZMrRt21bc36xZMxwcHDh//jwXLlygW7duWFpaAqCvr8+qVaswNPxrNW1iYiLB\nwcHk5+fj4OBA3bp1mT59Oqampujp6TFr1iyV9g8ePEhwcDDm5uYUFRXh5OSksj86OprBgwcD4Obm\nxrJly1T2C4LA7t272bFjBwAHDhxg48aN4v5ffvmFO3fuMH/+fHR0dPD09MTGxoaFCxcilUqxt7dn\n5syZFBQUMGnSJLKzs8nIyMDDw4M+ffqIx5HJZGqrgBs3bqxifYiOjqZFixYA1K1bl2vXrqmUVygU\nJCQkMHXqVFJTU+nZsyc9e/YU92/YsIHmzZtTtWpVceznzp3L999/r2LRCw0NZcCAAeL7bdu20aZN\nG/T19dm0aRN+fn4AhIeHs3nzZhQKBW3atGHEiBHitpycHLp06cKIESNo3rw5p06dAmD06NH06tWL\nx48fs337dhQKBSNHjuTu3bv8/vvvFBcXU6ZMGZYsWYJCoeCnn34iMTGRoqIipkyZwoYNG+jSpQut\nWrXi7t27BAYGEhISojIOmzZtYtWqVcjlcgICAjh79iz379/Hz88PuVxOt27d2L59O7q6ugCYm5uT\nmJjItm3bcHNzo3r16mzbtg0Ab29vHB0diY+PRxAEFi5cSLly5ViwYAHnz59HEAR8fX3p2LEjUVFR\nLF26FFBaVAMDA0V3C7lczoQJE6hcuTJDhgwhNDSUPXv2IJFI6NSpEz4+PkyYMIHMzEwyMzNZsWIF\npqam4nU7ePCgyjkGBgZiY2MDQE5ODoWFhTg4KHMSu7q6cubMGWrWrMnLHDx4ULwHioqKWLlyJVKp\n0m+3uLgYPT09ADp27MjGjRv56aef1I7xT/gX/RpreFN2/XKOlaN/RxCg87AGDPnlS6TSf9HKOkGA\nuPZQQ05qchViHL4EoIVBfR5q6eMF3Eu4imNBLuaWDriUtX6HRvxQOtj3pUR4yYsVbJp+jBNh19Er\no0NK45oMGmQFghZla0cx7efHDHGbgoGO3iuPqlAomDBhAlKplNmzZ+Po6Iij44eN+6/h0+Xs2bN4\ne3uTlpaGlpYWnp6eNG3alH379mFvb69W3t7ensTERJKTk6lWrZrKvjJlyqi8t7GxYciQIdy7d48+\nffrg7u6Or68vPXr04NChQ8ydO5fx48eL5YOCgggPD8fMzIwhQ4aotZ2TkyO2YWRkRHZ2tsr++/fv\nY2xsLP6I3r9/n5CQEAwMDJg6dSonT57EysqKgoICwsPDEQSBDh06sGnTJiwsLFi0aBE7duygZs2a\ndO7cmfbt25OUlIS3t7eK+DIyMiI0NPRvxzUnJwdj47+m+qVSKcXFxWhrK3/ScnNz6devHwMGDEAu\nl+Pj40OtWrWoVq0ahYWFbNmyRRQVJfXNzc25ffu2yrhHRUUxZ44ypZlCoWDPnj1s3boVbW1tOnfu\nzA8//IBMJuO3335j9+7d6OrqMnfuXBITE8VtV69e5Y8//kAmk73yfExMTAgODkahUBAdHc3atWvR\n0tJi0KBBXL16latXr2Jra8vChQu5ffs2p0+fxsPDg82bN9OqVSu2bdumIi5LcHFxYciQIRw7doyg\noCDmzp2Lu7s7Y8eO5cSJEzRu3FgUXqAUX8HBwWzYsIFff/0VfX19Ro8ezZdffikeb+bMmWzcuJEV\nK1bQokULHj16xJYtWygoKMDT05PmzZtz584dgoKCsLKyYvny5Rw4cIAuXbpQXFzM2LFjadCgAX37\n9iUuLo59+/axadMmJBIJvr6+uLq6AtCkSRN8fX1Vzqdfv37069fvje8LIyMjHj58qFausLCQI0eO\nEBkZCYCOjg7m5uYIgsC8efOoUaOG+L1dtWpVlixZ8so23xWN+PoPoVAIrPU7TMT8MwD4zm1Nj/HN\nXmmh+WSJCYAaseQpTDigNxMFUFXnM07r2PEdIANq3FHGKfN6J6vXcSCc/Nwy3D73A7EnjnPpz3vc\nOvuY4jwFgkTgonVl4rcoM1U263WOsGBnbM3+vi25XM7YsWMJCwtDX1+fAQMGUKVKlXfon4b3zdtY\nqN4nTZo0YeHChWRkZDBw4EDs7JSppqysrHj8+LFa+YSEBJo1a0ZycjJPnz5V2Xfz5k0EQaB69eql\ntpWcnMxnn30GQMOGDVmwYIG4LzU1FWNjY8qWVYZjqVevnlp9Y2NjUSDIZDJMTFSzRWRkZIiWOAAL\nCwv8/PwwMjLi3r171K1bF0D80UpPTyc5OZlRo0YBSgtI8+bNadmyJevWreP333/H2NiY4uJilXbe\nxPL1Yl9BKYxKhBeAgYEBPj4+GBgo4/E1adKEmzdvUq1aNc6cOUPDhg3VxGz58uXJzMxU2aZQKERx\ncuLECWQymZiNQqFQEBkZSeXKlalcubLoLzZx4kQuXbokbtPS0mLixIlq4y0Igvi6ZMy0tLTQ0dFh\nzJgxGBoa8vTpU4qLi7l3757ou1WlShWqVKmCIAgEBASQlpbGqVOnxMDNL9KggTLdWr169Zg3bx7G\nxsY0bNiQkydPEhERwffff69SPiEhAWNjY1FwXr16lSFDhtC4cWNxHEEpwo4cOYKVlRWxsbF4e3sD\nSotRYmIiVlZWBAQEYGhoSFJSEi4uLgDcunULY2NjcnNzAbh9+zaJiYmiyHr27BkPHjxQGZMXeZ3l\n6+X7orT7GODMmTNUq1ZN5R4oKChg4sSJGBkZMW3aNHF7uXLl1O6L94FGfP1HKCooZtGA3RzbHItU\nW4sfVnehtXftj92tt+fBUcibCgaQKWtKIboY5EtZb+LC2udFeinkPI67wDOgzVtEtc9Oz+P6qQRi\nT/xG7IlviIu2Q150QKVMkXkRcdJmxN/ugI5eMQG/pDBu6OvbKC4uZtSoUezYsQMDAwPWrl2rEV4a\nRMqWLUtQUBA+Pj7s3LkTFxcXUlNTOXLkCK1bK0X+8ePHSUhIoFGjRtjb2zNs2DA6deqEubk5MpmM\nqVOnMmzYsFeKr/Lly/PgwQPq16/P+fPnRSEGYGZmRnZ2NsJyincAACAASURBVOnp6Zibm3P16lXR\ngbwEFxcXjh07Ru3atTl+/Dj169dX2W9hYUFWVhYA2dnZLF68mD///BOAAQMGiGKiZNqubNmyVKhQ\ngWXLllGmTBkOHz6MoaEhq1evpm7duvTp04ezZ89y7NgxlXbexPLl4uLC0aNH6dSpE5cuXVL7rN2/\nf5/Ro0ezY8cOFAoFMTExdO/eHYDTp0+X6oT+7NkzLCwsVLbp6ekhl8uRSqVs27YNf39/WrVqBSin\nPv39/Vm1ahX37t2jsLAQXV1dRo4ciZ+fn7gNYOTIkUyaNIni4mJkMhk6OjrExcWJ7ZSM2c2bNzl0\n6BDh4eHk5eXh7u6OIAg4Oztz9epV2rZty8OHD1m0aBELFiygS5cuBAQE0Lx581JXUV+5cgUXFxcu\nXLhA5cqVAfD09OS3334jIyNDzbp669YtNm/ezPLly9HT08PR0ZEyZcqI03HXrl2jQoUKxMTEUKlS\nJZycnGjcuDGzZs1CoVCwbNky7Ozs8PX15dChQxgbG+Pn5yfeGzVr1iQkJAQPDw9atGiBk5MTlSpV\nYuXKlUgkEvF788CBA6UaDV5n+SqxzD548AB7e3tOnjxZqrP86dOnqVOnjvheEAS+//57GjdurGYV\nzsrKUlt48j7QiK//ALJn+QS4h3PlyH0MjHWZGOFBvXZOr6/4qZH1EA56QD/lB9XaaBw15A0ZpKvg\nukSKAbAEqPXoBkPysqlY1ppKlqUnrQZIfZRF7IkH4l/CtZTne5Rf1IJEoMg2jwLHXOTOBVhXrMu5\n37xJfWKIrS3s2qVN/fpvNqX5ww8/sHPnTvGHo+RJUYOGEipVqoS3tzf+/v4sXryY5cuXM3v2bFas\nWAFAhQoVCAkJQSqVYmdnx7hx4xg+fDhSqRSZTEbPnj1p2bLlK4/v7+/PTz/9RHh4uDjtXYK2tjZz\n5sxh0KBBmJqaqliJSujduzd+fn707t0bHR0dFcsZQMWKFUlPT6e4uBhjY2NcXFzo3r07hoaGmJiY\nkJycLFr2QCkoJk2axJAhQxAEASMjI+bNm4dEImH69OlERkZiZmaGVCoVhcub0q5dO06dOkWvXr0Q\nBEE81zVr1uDg4ECbNm3o0qULnp6e6Ojo0LVrV1F8xMfH061bN5XjKRQKkpKSxJWpJbi4uBAbG4ut\nrS2XL19m4cKF4r769etTUFDA/fv3+eabb+jXrx8SiYQvvvgCW1tbcZtMJqNLly5YWVnh4+ODl5cX\ndnZ2orXm5TE2MDDA3d0dXV1dypUrR3JyMr169WLixIn069cPuVwuWtLc3d1p1aoVu3btKnWcLl++\njI+PDxKJRByjOnXqkJCQoLZyFqB9+/bcvXsXDw8PDA0NEQSB8ePHixaiHTt2sHbtWgwMDJg3bx5m\nZmZERUXRp08fcnNzadu2LcbGxnTt2hVPT09MTEywtLQkOTlZbENfX5/p06fj5+dHeHg4TZs2pXfv\n3hQWFlK7dm2srP5ZGKQZM2YwduxY5HI5rq6u1KlTh8zMTCZPniz6ocXHx4tTqQCHDh0iKiqKwsJC\nTpw4AcCYMWOoV68ely9fpmnTpv+oT6UhedH0+SkTHR0tvPwk9gHaUHva+zcwo8sWzu+5g5mVETP2\n98a53rv4QP0zpIACKH7++q0pLoCtblAximsN2yMXqnBG7xdGSbQoAGoCW5//n31oNVsv/c6gxl0Z\n2aI3oHxyeXQrTRRa1088JOm+qqlY0FZQtdEj6raK56yuAfFmn+Faoy6uTnVJuvg5gwbokpMDjRrB\nzp1g/RbDuHPnTiZNmsS6detEU/9/mX/rZ+W/zoe+LitWrMDJyYl27dp9sDY+BseOHSM2NlZtGu7i\nxYvs3buXyZMnv/OxP+Q1SUpKYvz48axbt+6N6ygUCnr37s2qVatU/KNeh7e3N9OnT8fZ2flduvrJ\n8abX5ccff2TUqFGl+mi+YRul+v1oLF//AW6fU/qOzPq9778oeOpLHP0B0qN43P1zTuUPQEDKPO0M\nCrQt6J6SwoZy5TAEFIKCI3fOA1BV7syuRedEwfUsJVflkIK+gvzPZBQ45lLgmMvQXucZ3OIayTnV\naJ7/O04WdoCEuXNh0iSln3/fvvDbb/DcVeRvyc/P58qVKzRq1Ihu3brxxRdfiKtyNGj4L9K/f38m\nTZpEmzZt/jPpsQRBIDIykpkzZ6rtq1evHrt37+bp06dq07Qfm4MHD7J06VICAgLeuE5JXDgvL6+3\nEl7/X7l58yYODg7vJLxeh0Z8/YcoW+Ff+mG6tgaurCCjlT2/F49CQMoBvWqkaluwGaj84AGG5cpx\n/2oS23YfI0UvA+1MXRZ/cQAJfz1UaJmCrGIW+Y5KwVVknU+5MmVp61SXdlXK0+wz5bL48sbrKG9s\nT34+DB4MGzeCRAJz5oCfn/L168jLy2PgwIGcOXOG1atX07p1a43w0vCfR19fX2068t+ORCJh/vz5\nr9z/ovP1p8SXX36pMnX2Jtjb279yivJ1vM4P779ItWrV1Pzi3hca8aXh45IUDYe+45mFFZGVJ1Mg\nlOGxti6P9WoRA1QColH6tY2qv5KULxOhJehfKYO+jTZ5FXPItEunwDGXYstCtLQk1LapQgvHurg6\n1aNquYrPHTc9gULAG2jEkyfQrRtERYGRkVKAde36Zl2WyWT079+fM2fOYGlpWarvhgYNGjRo0PAq\nNOJLw8cjLw129wB5ARfbdiZXKIuB9DEywxGclEh4MZqWLDOfoiI5ebWVq61kLpnEuaUBUNagDF84\nNsbVqR7NKtbG1OBlC+BdYBugB8whOloptB4/hooVYfduqP2GC0Ozs7Px9vbm/PnzWFlZERYWpuak\nq0GDBg0aNPwdGvGl4eOgkMPePghZCVyxd6Gu7Xbsi2N5pD2X/hJ1l/3MvByKbPORly0CoLpzRdyc\n6uHqVI+aFZzQkvyd/8mvgAD0JizMFl9fyMsDV1eIiIByb5EWcs2aNZw/fx5ra2vCwsLUooRr0KBB\ngwYNr0MjvjR8FBSnp1H49DSRX88h0U7GDxJ/9HWMcKaVWtlCeREzDq4g93Ol1evrmi2Z1fG7N2wp\nB1iFQiFhxozZlPjUDhoEy5bBW6xuB2DYsGFkZmbi6+srprDQoEGDBg0a3ob/xnIVDf8q0uN2URyz\ngD0dJpFW3hkbuQmCIEGfGWplFYKClTf2cz05nrzn4uurmi3eorX1yGTFeHoeYeZMa7S0YNEi5YrG\nNxVeaWlpDBo0iMTERKRSKVOnTtUILw2v5dy5czRt2hRvb2+8vb1xd3dn5MiRYuDN9PR0/Pz8xPQ6\nP/74IykpKWL9CxcuMGDAALy9venRo4dKHsUSHj58SNeuXcUcgy/z6NEjPD091bbPnz+fiIiIUuv8\n8ccfYhT30pgwYcLfpsr5X7F06VJ69uxJr169uHLlitr+K1eu0KdPH3r37s3IkSMpKChAoVAwdepU\nvLy88Pb2JiEhAVDmpXwx6OmL7Nq1iz/++EN8HxISgqurKwUFBR/mxN4TERERf7uQ4FUcO3aM/v37\nM2DAAHx8fNi9e/d761Pr1q0/+Lilp6czcOBA+vTpw6hRo8jLy1PZHxERgbe3N7NmzcLT05PPP/+c\nrKwszpw5g5eXF3379mXkyJHk5eWRn5+vEiT2faKxfGn4n3I8/TZ19vfnaMthpJZzxkCrgI6Gc5BI\n2gCuauUXn9jChZRbGOoZkls2F2meNvXtSo/wrY6CBw+20bXrSS5dqoepKWzdCm+zQCg5ORkvLy9u\n376NXC5n7dq1b15Zw/97StILlfDjjz9y5MgRvvzyS4YPH87AgQPF5NqnT59m6NChhIeHk5iYiL+/\nPytXrsTS0pL8/Hx8fHywt7dXic4eExND06ZNmTBhwnvpr7+/PydPnnxlFP19+/ZRs2ZNjIyM3kt7\n70psbCxRUVGEh4fz5MkTRowYwfbt28X9giAwZcoUFi9eTMWKFQkPD+fx48fExcVRWFjI1q1buXTp\nEnPnziU4OJgBAwYwduxYtcTUubm57N69m1WrVonbIiMj6dSpE3v37sXd3f1/ds7/K6ZPn86uXbsw\nMTEhJyeHrl270rx5c7Xo/58qy5Yt46uvvsLd3Z2QkBC2bt2qkiPS3d0dd3d3oqOj2bNnDz169MDE\nxITp06ezceNGLC0tWbBgAeHh4fj4+FCvXj127twpZkh4X2jE17+Ix3fSiIq8g6BQVeEFuUUfqUdv\nThEwrbiA3pE9uFu9NfGfNUYHbboa/oSxVgaUYvXaevF31kTtRirRolmV2hx6dBaze+Zoa71ZGNcz\nZ87TvfsmkpIqUKmSQGSkhLdZNfzkyRO8vLy4e/cuVapUYd68eW9eWcMnxfTOm7mwr3TLxrvSoFMl\npu/t/cblCwsLSU5OxtTUlGvXrlGmTBlReAE0a9YMBwcHzp8/z4ULF+jWrZuYS1FfX59Vq1ZhaGgo\nlk9MTCQ4OJj8/HwcHByoW7cu06dPx9TUFD09PWbNmqXS/sGDBwkODsbc3JyioqJS/RVdXFxo27Yt\nW7duLfUcQkND+fXXXwFl0umSiOH5+fkEBgaio6PDd999h5mZGW5ubri5ueHv7w8oUxzNnj0bQ0ND\npk6dytOnT8nIyMDNzU3M/1jC0KFDxfx/AM7OzkyfPl18Hx0djaurKxKJBBsbG+RyuZg6CZQRzM3M\nzFi3bh23b9+mZcuWODk5sXXrVlq0UFrO69aty7Vr1wBlYms9PT0x/2MJkZGRNG/eXHx/7tw5HBwc\n6NWrF+PGjRPF1+XLlwkICEAQBKysrJg/fz63bt0iICCAnJwcnJycmD9/Pt98840YqHTz5s2kpqbS\nvXt3lTGrU6eO2rg6OjqybNkyDh06hFwup3fv3kgkEu7fv4+fnx9yuZxu3bqxfft2lUwBly5don//\n/uTk5DBixAgqVqzIuHHjxKTio0aNYuDAgdR+YcWRhYUF69ev58svv6RSpUrs378fXV1dlixZwr17\n90hLSyMrK4vJkyfToEED9u/fLyYCr1+/PmPHjuXp06dMnz6dgoICMjMzGTZsmMq9vnnzZk6dOsXP\nP//MpUuXWLhwIVKpFHt7e2bOnElkZCTbt29HoVAwcuRIMcL8hQsX+OWXX1TuFV9fX9q0aaNybwwd\nOhQANzc3fv75Z7UE3QD37t0jLi5ODCUSGhoqft6Ki4vR01Mu+erYsSODBw/WiK//r5zecZOfvXeS\nLytdaGlpSdDRe6fY8h+c+0BvoNrNTdRMi2VXk14AtDW8jZn0LtAeaPZX+fRE5h8N5UT8RQC8q7Tj\n2FPll6TZ3Td7+lq/Hr75xoXCQh3atIknLMyRt0nP9fjxYzw9Pbl//z41atRgy5Yt/5onPw2fDmfP\nnsXb25u0tDS0tLTw9PSkadOm7Nu3r9TAjfb29iQmJpKcnKwWX+jlRNA2NjYMGTKEe/fu0adPH9zd\n3fH19aVHjx4cOnSIuXPnMn78eLF8UFAQ4eHhmJmZqeWvK6FTp06cO3eu1H35+fk8efJEFDh37twh\nKCgIKysrli9fzoEDB+jSpQspKSmiCPD09GT27NlUqlSJ8PBwVq5ciYeHB3Xr1sXDw4OCgoJSxVdJ\nyqVXkZOTg5mZmfjeyMiI7OxssW8ZGRlcvHiRKVOmULFiRb799ltq1apFTk6OSnBRqVRKcXEx2tra\nVK1alaioKJVxj4qKUrFuhYeH4+HhgZOTE7q6uly+fJk6deowZcoUFi5ciLOzMxs3buTu3bvitszM\nTG7evMndu3dfeT4vjtnGjRvVxrVly5YcP36c8PBwCgsLWbBgAaNGjcLd3Z2xY8dy4sQJGjdurJai\nycDAgJCQENLT0/Hw8ODQoUPo6+sTFxeHpaUljx49UhFeAMHBwaxdu5YxY8aQnp5Or169xPyI+vr6\nrF+/njt37vDjjz+yfv16lixZwvbt2zEwMGDcuHGcOnUKiUTCgAEDaNy4MTExMSxZskQUX6Ghody4\ncYNffvkFLS0tpkyZwqZNm7CwsGDRokXs2LEDbW1tTExMCA4OVulbgwYNXhtvLCcnR/yslNwXpbFr\n1y6VvI/ly5cHlNPu586dE+9JU1NTMjIyyM7OVvsM/hM04usTRxAEtgacZMOUPwFo2LkydtXURUCl\nBtYYmer/j3v3eiKAQcAzQcH680FoCQJd5Q4kGtXFTnvg81LTAcguyCXkzHY2xRygWCHHSNeAUW69\nMc7WYu2tg0jytCjz8O8Dmcrl8NNPEBQEoMPw4cv4+WdPSsk5+7eUJJKtXbs2mzZtomzZsm93AA2f\nFG9joXqflEw7ZmRkMHDgQDH3oZWVFY8fP1Yrn5CQQLNmzUhOTubp06cq+27evIkgCK+cEkxOThaT\naTds2FAlGGpqairGxsbifVyvXr23Ppdnz56pfA6srKwICAjA0NCQpKQkXFxcALCzsxNFwN27d5kx\nQ2nVLioqwtHRETMzM65evcrZs2cxNjYWfeBe5HWWL2NjYxW/M5lMpvLDaGZmRsWKFcUwMC1atODa\ntWtq9RQKhZjnsly5ciQlJan0IyMjQ3zoevbsGcePHyc9PZ3Q0FBycnLYsGEDderUIS0tTUy7U5Iz\nsWRbdHR0qXkUX/QjenHMShvX+Ph4ateujVQqxcDAQEx31LBhQ06ePElERIRaaiRQ5p+USCRYWFhQ\npkwZMjMz8fDwICIiAhsbG77++muV8s+ePSMxMZFx48Yxbtw4kpKSGDFiBDVr1gSU9zNA5cqVSU1N\n5cGDB6Snp4tiXiaT8fDhQ+rXr09wcDDbtm1DIpFQXFwstnHmzBmkUilSqZS0tDSSk5NFoZOfn0/z\n5s1xcHDA0dFR7XzexPJVco319fWRyWSYmJioHScrK4vExETxfEpYu3YtBw4cYOXKlaLlC8DS0pLM\nzEyN+Pr/Qn5uEb8M2M2JsOtIJNB/bht6jGtaarb3T418YCzKIA8As+7tIbFGC2xvSTCs2gs76W8o\nVyK2Q65ozM5rh1lyYisZeVlIkND98y8Y4eqFhZEZMyOWAWBwvQxailevEcnKgj59YO9e0NaWs2TJ\nML79tghQ/1J6FYmJiZQvX54KFSoQHh6OkZGRJnK9hn9M2bJlCQoKwsfHh507d+Li4kJqaipHjhyh\ndevWABw/fpyEhAQaNWqEvb09w4YNo1OnTpibmyOTyZg6dSrDhg17pfgqX748Dx48oH79+pw/f14U\nYqAUI9nZ2eLU3NWrV986XU7ZsmVVhMvkyZM5dOgQxsbGKk7JL6YdcnR0JDAwEBsbG6Kjo0lJSSEi\nIoIyZcowc+ZMEhISCAsLQxAEle+111m+XFxcCAoKYtCgQTx9+hSFQiFavUBpQZTJZCQkJFCxYkUu\nXLhAz549cXBw4OjRo3Tq1IlLly5RpUoVsc6zZ8/UrNvm5uai5WT37t306NFDXNyQl5dHmzZtSE9P\np3z58ty/f5/PPvuMkJAQHB0dxW2AuE1XV5eUlBScnZ25fv26mET6xTErbVydnJzYvHkzCoUCuVzO\nkCFDWLFiBZ6envz2229kZGSUGon96tWrgNKylpubS9myZenQoQOrV6/GzMxMTcgUFhYyatQoNm3a\nhLW1NeXKlcPS0lIUhrGxsXTt2pXbt29jZWWFnZ0d1tbWrF69Gh0dHSIiIqhevTq//PILHh4etGzZ\nku3bt7Njxw6xjWXLljFp0iQ2b96Ml5cXFSpUYNmyZZQpU4bDhw9jaGjIkydPSk1f9SaWLxcXF44d\nO4a7uzvHjx8vNX/j+fPnqVWrlsq24OBgYmNjWbt2Lfr6qoaMrKwslfvrfaARX58oKQ+f4d8tjLsx\nTzEoo8u4Td1p9FWV11f8BLiNMp78ZUAHCBIEahRe4Ua1tjxzcOVrqQ6wD4CEdDfG75nIzeT7ANSz\nrYpfa1+qW/311BOVeBMAg6smaOuULr7u3YMuXeD6dTA3F9i27Wu++GIfcPGN+x0XF4enpyfNmzdn\n0aJFmsj1Gt4rlSpVwtvbG39/fxYvXszy5cuZPXu2KDQqVKhASEgIUqkUOzs7xo0bx/Dhw5FKpchk\nMnr27EnLli1feXx/f39++uknwsPDkUqlzJ49W9ynra3NnDlzGDRoEKampqK1523Q1dXF0tKStLQ0\nLCws6Nq1K56enpiYmGBpaUlycrJanenTp4s+SQABAQE4OzszZswYoqOjMTAwoGLFiiQnJ4tC5E2o\nVasWDRo0wMvLS1zBCEofrdzcXLy8vAgICODHH39EEATq1atHq1atUCgUnDp1il69eiEIgsoYXbly\nhdGjR6u007hxYy5fvkzDhg0JDw9X8fs0MDCgffv2hIWFMWPGDCZOnIiWlhblypXD19cXKysrJk6c\niEwmw8nJCV9fX3R1dZk5cybW1tbiNNfLlDau1atXp0WLFvTu3VtMjK2rq0udOnVISEgo1bIGiAs1\ncnNzmTlzJhKJBD09PRo2bEh6errK1C0orX+TJ09m+PDhaGtrI5fLadWqFa6urly8eJEbN27Qv39/\n8vLymDVrFubm5vj6+uLt7Y1cLsfW1paOHTvSoUMHAgICWLFiBdbW1mRkZKi0M3nyZDw8PGjatCmT\nJk1iyJAhCIKAkZER8+bN48mTJ298L7zMd999h5+fH2FhYZQtW1a0AI8fP55Ro0ZhY2NDfHy8yvin\npqby66+/UqNGDb755htA6evVp08fsrKyMDExef+LTARB+Ff8XbhwQfjQ/C/aeBOun34o9LVaIHRm\npjDIaYlw/1rSx+7Sa9ESlBdqiyAIRs9fOwuCcEEQhEtJB4XgzDBhZcp6IT0vURCEPEGhMBAEAaHN\nsq+F2kFeQvvl3wv7b5wSFAqFeEyFQiFsWXFUqB3kJdTx7y30tJwrnIq4odb20aOCYGEhCCAI1asL\nQlzc2uc9cHvj/t+4cUOoXbu2YGNjI/To0UPIzc1957H4/8Cn8lnRoMqHvi6RkZHCmjVrPmgbH4OM\njAxh6NChattzcnIEHx+ff3TsD3lN5HK54OnpKWRnZ79VvWnTpgmnT59+qzqLFy8WNm3a9FZ1PmXe\n9Lps2LBB2Llz5z9po1RNo4nz9YlxeN1lfmq1nswkGbW/+IyfowZSsWbpT0ifIoMBGdALiAFsilM4\nq6t86vki8Qn6UnMiY/2RSPK4mWxGVoEJ3zbtwc6BP9OhWjNx6iH5wTOmdtjEsg1Kc7VtljUrrnxP\ns+6qpvWQEGjXDtLSoFMnOHtWgbNzwPO9I9+oz9euXaNnz56kpqbi5uZGaGgoBgYG/3wwNGj4j9G5\nc2diY2M/iThf75O1a9eqWb1A6bDdrVs3Dh48+BF69fc8fPiQ7t2707VrV5VFBK9j4MCB5OfniysI\nNbya/Px8YmJi6NKly3s/tmba8RNBLlewZvxhdv58FoCvhjdk8M/t0Nb5NFcwvooclBkUNwEIAr/n\nnEOQaFHn6h7ijXvy7erReNc/DEDiswbsGrgQaxNLsb4gCBxceZFVP/5BXnYhBSOVX/IjfHtgbv2X\ns2NxMYwZA0uWKN+PHQtz54JUehC4A9gDr8+UnZeXh7e3NxkZGbRp04aQkBC1+X4NGjQokUgkBClX\ns/yneHm15Yu87xAD7wt7e3t27dr11vVWr179Tu2NGDHiner9m9HX11dZuPI+0YivT4CczHyCekcQ\nfeAuUm0tvvu1Ix2GuHzsbr0zkud/SCR0vHiAK9pZ5MdeZ0yOMrJx68qpz/9PBf4SXkn3M1nyzR4u\nHYoHoJ5XRSLtr6OtJaWF81+rszIywNMTDh1SRqkPCYH+/QEUwNTnpYbxJre3gYEBgYGBREREsHjx\nYrWl2ho0aNCgQcP7RiO+PjKPb6cx6+utPLqVhomFAT9t9+DzlhU/drf+MQlFTzDKfIbZxRU0VhTT\njY6YG5ri17oFtqZbAVNAafZWKAQOrIhmzfjD5OUUYmJpyHe/diCx8mN2HxaoZe6Ika5yGvDWLaVj\n/Z07UL487NgBzcQQYaHABcAapfh6NefOnSMxMZHu3bvTvn172rdv/2EGQoMGDRo0aHgJjfj6iCTd\nz2RM49XIMvP57PPyTN7lSQXHf388qToFCezPj0Ir6Q7fCMUcldjRsn5vhjTtQRm9Nc9LtQO0eXov\ng8WD93Dl6H0AXD2q8+3SjpiVN2JImNLfq365ygD8/rvS4vXsGdSpA7t3w18pFnOAn56/ngu82gfi\n5MmT+Pr6UlBQgJ2dHQ0bNny/A6BBgwYNGjT8DRrx9RGJOXgXWWY+VZvYMuv3vhiW0Xt9pU+QHJQT\nfgCO98/QxzgBdKTUe3gGCVC98zJaVytxWNwPgELRkb3LzrPW7zAFuUWYljPku2Udce1ZA4DMvGwu\nPFROOdYxr8Qvvyh9vBQK6N5dGcFe1cd0LvAEaAj0e2Vf//zzTwYNGkR+fj6enp5iYEgNGjRo0KDh\nf4VGfH1ESgIcO9ax+tcKL4CSlL7l0hPplXMB3bIVkNy/TYPre8C2Bdai8MoHjgAwp0c2Z3YeAMCt\nV02GLumAqeVfeeuOxl1ALihoYl+HRfOqsHOncvuUKTB9OqjG30sA5j9/vQhesYj3jz/+YMiQIRQW\nFtKvXz/mzJlTaiA/DRreByUpSkqirMtkMuzs7Jg/fz66urqkp6cTGBhIYmIicrkca2trJkyYQLly\n5QBlNO9ff/2V4uJicnNzcXd3V4vn9PDhQ4YPH061atUIDAxU68OjR48YM2YMYWFhKtvnz5+Pk5OT\nSuqc7Oxsxo0bR05ODkVFRUyYMKHUSPgTJkxgypQpHz259tKlS/nzzz/R1tZm4sSJamly1qxZw7Zt\n28TgmDNmzMDGxoZx48aRlpaGkZERgYGBmJubM2PGDIYNGybm9nuR5cuX4+rqKgblnDZtGpcvX2Zn\nyZfSJ8qSJUuwtLSkd++3y+6wY8cOduzYgVQqRRAEBg8ejKur63vpU/PmzTl16tR7OdarSEhIYMKE\nCUgkEipXrsy0adNUvudDQkI4ceIE2dnZCIJAamoqJv4ABgAAIABJREFUp06dYs+ePaxbtw6pVEqV\nKlWYPn06aWlpBAcHi3Hk3ieaXx4N/4g9hXnKKPbyYqqfXImDfQUKFOATs17pdN9QmVsuOz2PuzFb\ngTzuXbbmzM4szKyMmBjhwfjN7irCC+Dw7SgALkc2YufOcujrw5YtMHPmy8ILYDxQgDKDZLOXd4qc\nPHmSwsJCBg4cyNy5czXCS8MHp0mTJoSGhhIaGkpERAQ6OjocOXIEQRAYPnw47dq1IzQ0lE2bNtGj\nRw+GDh2KXC7n4cOH+Pv7ExQURGhoKBs3bmTXrl0cP35c5fgxMTE0bdq0VOH1tqxZs4YmTZqwYcMG\n5syZw8yZM9XK7Nu3j5o1a3504RUbG0tUVBTh4eH8/PPPYgqjl8sEBgaK418SJb5KlSps2rSJbt26\nsWyZMnuGt7d3qavanjx5wu3bt0XhlZeXR0xMDM7Ozq/MgflvJjs7m2XLlrFy5UrWrFlDYGAgEydO\nRKFQvL7yJ8KcOXPEKP2CIHD48GGV/UOGDCE0NJQpU6ZQoUIF5s6dS35+PosWLWL9+vVs2bKFnJwc\njh49Srly5TAyMiIqKuq991Nj+dLwTsgVCjbFHmOIQy3QNcDq7E6+rOBAKgYUpD3jxhk5cek9iDuT\nx93oJSTdz+SbhftwdoELeyvxRb/P+WZRe0wsDNWOnV2Qy+n7VxAUEi7saEC5coXs26dLgwal9eQE\nEAYYoJx6VCcvLw8DAwOmTZtGo0aN6NSp078iRZOG98fw7YFiovb3RQvHeizt4ffG5QsLC0lOTsbU\n1JRr165RpkwZMdkwQLNmzXBwcOD8+fNcuHCBbt26iZYYfX19Vq1ahaHhX5+XxMREgoODyc/Px8HB\ngbp16zJ9+nRMTU3R09Nj1qxZKu0fPHiQ4OBgzM3NKSoqwsnJSWV/SQR2ALlcrpLbroTQ0FB+/VWZ\nNCwqKoqlS5cCynhIgYGB6Ojo8N1332FmZoabmxtubm74+/sDyhRHs2fPxtDQkKlTp/L06VMyMjJK\nTaz9utyO0dHRuLq6IpFIsLGxQS6Xi6mTSoiNjSUkJISUlBRatWrF0KFDiY6OZvDgwQC4ubmJ4svJ\nyYl79+6RkZGhkr9y8+bNfPnll+L7/fv307RpU9zc3Ni4cSONGzcG4OjRo+JY1KhRgxkzZnDs2DGW\nLl1Kbm4uDRo0YMaMGbRt25b9+/ejp6cnWh9tbW2ZP38+Ojo6eHp6oq+vz8aNG8U2f/nlF8zMzPD3\n9+fKlSsUFRUxYsQIrly5gpWVFX379uXZs2cMGDCAiIgIlXE8dOgQ+/fvJz8/n8mTJ5OVlUVYWBiL\nFy8GoFevXixevFiM9m5oaIhcLmfz5s188cUXODg4cOjQIbS0tJgwYQKCIPDkyRNyc3MJDAzE2dmZ\n0NBQ9uzZg0QioVOnTvj4+HD79m3mzp2LQqEgKyuLyZMnq7h4/Pzzz2RnZzN16lQOHDjA2rVr0dLS\non79+owdO5YlS5Zw8eJFcnNzxawIAAcOHFAZG4Bx48apWD1jY2Np1KiReI1PnTpFu3bteJmoqChM\nTExo0aIFCoWCLVu2iPEdi4uLxfv/q6++YsmSJeIx3xca8aXhrTn/IJZ5R9dxtHYbFIWGOIVH8UOM\nHpm3FBy7fBfjp9lczO3/vLQyNZCegTZNuytDSNRtNwzPid1eefxF4THIBTmy+Oq41DBl5swrNGhQ\nu5SSCqDkS3sc4KBWYvv27QQGBhIWFsZnn31G586d3/m8NWh4W86ePYu3tzdpaWloaWnh6elJ06ZN\n2bdvH/b29mrl7e3tSUxMJDk5WS1X38tJfW1sbBgyZAj37t2jT58+uLu74+vrS48ePTh06BBz585l\n/PjxYvmgoCDCw8MxMzMTEyG/SEkC4pSUFMaNG8fEiRNV9ufn5/PkyRNR4Ny5c4egoCCsrKxYvnw5\nBw4coEuXLqSkpLB9+3Z0dXXx9PRk9uzZVKpUifDwcFauXImHhwd169bFw8ODgoKCUsXX63I75uTk\nqKTGMTIyIjs7W0V8de7cmT59+mBsbMzw4cM5evQoOTk54jiW1CnBycmJmJgYlSTNUVFRKlOz4eHh\nzJw5UxSDSUlJWFhYMGvWLMLDw7GwsGDp0qU8ffpU3Hb//n3OnDmjlij9RQoKCggPDweU05whISEY\nGBgwdepUTp48iYGBARkZGWzbto2UlBQ2bNiAh4cHY8aMoW/fvuzZs6fUQKC2trbMnDmTO3fuMH78\neCIiIvD39+fZs2ekpKRQtmxZlTQ7UqmUNWvWsG7dOgYPHkxRURHffPMNffr0AZT3Z2BgIMeOHSMo\nKIixY8eyb98+Nm3ahEQiwdfXF1dXV+Li4vDz86Nq1apERkYSEREhiq/AwEAkEgnTpk0jMzOTJUuW\nsH37dgwMDBg3bpw4Lenk5CQmEC+hQ4cOdOjQ4ZXjCKjkCX35Gr/I7t27Wb58OaDMrVnyoBMaGkpu\nbi7NmzcHlGnBYmJi/rbNd0EjvjS8MQ8TkglYtY7YqAS00stT5+lNDB4qzbFXuhiT/IcMg3wBbZ1C\nqjqn4dzxKyo1sMW5vjX21bKQak8ETKnSsEepxxcEmDcPVl0/h+nnUEm3MTuPwfXrRa/o0TqUcfRt\nUU49qrJlyxbGjh2LIAj88ccfYs4uDf//eBsL1fukSZMmLFy4kIyMDAYOHIidnR0AVlZWPH78WK18\nQkICzZo1Izk5We3H+ubNmwiC8MrE2snJyWIy7YYNG6pMo6WmpmJsbCxadUrz5QK4desWY8aMYfz4\n8WpP+s+ePVOxCllZWREQEIChoSFJSUnij6udnZ1oQbt79644JVhUVISjoyNmZmZcvXqVs2fPYmxs\nTGFhoVo/Xmf5MjY2VomyL5PJVMSpIAj0799f3NayZUuuX7+uUk8mk4mCE5R5DTMzM1X6kZGRIf4o\n3717lzt37jB3rtLCLpFI2Lx5M3379sXExERMyj18+HBSUlLEbffv32f48OFq5yiUOP2iTEBegoWF\nBX5+fhgZGXHv3j3q1q1LfHw8devWFftZEo3fyMiIuLg4IiMjRSvei5Ss5K5cuTIpKSlIJBK+/vpr\n9uzZw6NHj+jZs6dK+aSkJPLz80Ufp/j4eAYPHiwmp27SpAmgvH9mz57N7du3SUxMxNfXF1DeIw8e\nPKB8+fIsW7YMfX19ZDKZGIE/NTWVW7du4fB8mfqDBw9IT08XHwZkMhkPHz5UG5MS3sTy9aI7ycvX\nuIS4uDgMDQ2pWPGvsE4KhYKgoCDi4+NZsmSJKOCkUilSqRSFQvFeXVU04ktDqWQ8zSEu+onKX9pj\n5ROEKSUJcLPQMtKhTi8bbBcYoJ2jg2nYStyL9yFtFQgNO71wxPDn/9tT2m2Xnw/ffAMbt+ZTc/pl\nANbNasSrs/xkAyVP5nMBVR+U9evX89NPytATEyZM0AgvDR+VsmXLEhQUhI+PDzt37sTFxYXU1FSO\nHDlC69atATh+/DgJCQk0atQIe3t7hg0bRqdOnTA3N0cmkzF16lSGDRv2SvFVvnx5Hjx4QP369Tl/\n/rwoxEA55ZednS1OzV29epUKFSqo1I+Li+OHH35g0aJFala3knN4UfBMnjyZQ4cOYWxsjJ+fnygm\nXvyBcnR0JDAwEBsbG6Kjo0lJSSEiIoIyZcowc+ZMEhISCAsLU7FWwOstXy4uLgQFBTFo0CCePn2K\nQqFQsXrl5OTw1VdfsW/fPgwNDTl37hw9evTAwMCAY8eOUbt2bY4fPy6KClAKhxIBVYK5uTlZWVkY\nGxsTHh7O6NGjxUUPiYmJeHl58f3335OVlUVmZqY4Pfj111+L2wBxm66uLsnJydjZ2XHz5k1xOq1k\nzLKzs1m8eDF//vknAAMGDEAQBJycnDhw4IBYZtSoUaxatQpPT0+Cg4OxsrJSOf8Srly5QpcuXbh1\n6xY2NjYA9OjRg7Fjx5KXl8ePP/6oUj41NZUJEyawYcMGTE1NsbW1pWzZsujo6ADKKb0GDRoQExND\n5cqVcXJyolKlSqxcuRKJRMLatWupUqUKw4YNY/78+Tg7O7N48WLxQcPS0pJVq1bh7e3N8ePHqVWr\nFtbW1qxevRodHR0iIiKoXr26ONX5Mm9i+apRowbnzp2jcePGHD9+XBSML3L69Gnq1Kmjsm3q1Kno\n6uqybNkylbYFQUBbW/u9+whrxJcG0hKzVUTW3egnpD/JUSunbaRFjlU2tK7LjVY1sa1vzfHK5hzO\nP8XD4qdU1zOkuWIvz/RNMK398rTG/uf/O6od98kTZfiIc+egfMNLaOkWUtu6MhVM1L9M/mIO8BRo\nDPRR2bNjxw5ReE2dOpWhQ4e+6VBo0PDBqFSpEt7e3vj7+7N48WKWL1/O7NmzRaFRoUIFQkJCkEql\n2NnZMW7cOIYPH45UKkUmk9GzZ09atmz5yuP7+/vz008/ER4ejlQqZfbs2eI+bW1t5syZw6BBgzA1\nNUVbW/2rf8GCBRQWFhIQoMyNamxsTHBwsLhfV1cXS0tL0tLSsLCwoGvXrnh6emJiYoKlpSXJyclq\nx5w+fTp+fn7I5XIA0X9nzJgxREdHY2BgQMWKFUlOTsbKykqt/quoVasWDRo0wMvLC4VCIVpqIiMj\nyc3NxcvLi9GjR+Pj44Ouri5NmzalZcuWNGrUCD8/P3r37o2Ojo6KdfDGjRuMHTtWpZ1GjRpx+fJl\nLC0t2bt3r0o6HxsbG6pVq8bBgweZNm0aQ4cORUtLixo1avD555+L23Jzc2nUqBGff/45gwcPZsiQ\nIdja2pZqkTE2NsbFxYXu3btjaGiIiYkJycnJuLu7c+bMGXr37o1cLmfYMGUQ6bZt2zJz5sxXpnx6\n9OgRPj4+FBYWigsorKysMDIyom7dumr3Qc2aNfHx8aF///7o6+sjl8vx8PAQ/QOPHz/O4cOHUSgU\nzJkzB3t7e5o2bUrv3r0pLCykdu3aWFlZ8fXXX/P9999jYWFBhQoVyMjIENuQSCTMnj2bQYMGERYW\nhq+vL97e3sjlcmxtbenYUf034m3w8/NjypQp/Pzzzzg5OYk+ewMHDmT58uXo6uoSHx+vcr/Fxsay\nbds2GjRoQH9luhR8fHxo164dt27dEq2O7xPJi6bPT5no6GjhxaeUD9QGH7qNF9m3PJpl3+2jw1AX\nhi//8L5IgiCQ9lgptO7GPCEu+ilx0U/IeKoutAxN9HB2qUCl+tbi/yPPzjIj/iL3vKaggzKWvFVx\nGjtkR9BGSq/TERhf38iChuP50e3F1Vf5gDmQBySijECvJCYGunaFR4+UAVM7+C/mbNJpxrTsS/+G\nSh8G9esSD1RHucLxDKD6ZJOeno6npyd9+/ZlwIAB/3zgNKjxv/6saHgzPvR12bNnD6mpqeI003+F\nuLg41qxZIwrPEh4/fkxgYKDooP4ufMhrkpeXR79+/QgPD38ry8zQoUOZOHGiyrTb65gwYQKdOnXC\nzc3tXbr6yfGm12XevHm0bt2aBqWv+HqTNkpd3aWxfP1HEQSB1EdZL1i0nnI3+gmZyTK1skameji7\nWCuFVn2l0LJ2NkdLS/WeKYjR5mEHpRVpClAb2FtwHYDPBUuMbmymUEuHX11+QNWYfQyl8KrHi8Ir\nPFyZkzEvD1xdYXNYIR5hSsfGNpUb/83ZlYSW6MuLwmv37t106NABc3Nz9u7dW+pqLQ0aNLw7nTt3\nZvz48chkso8ebuJ9Ehoayg8//KC23dbWlqpVq3L16lU+//zzj9CzVxMTE8O0adMYNWrUGwuv/Px8\n+vTpQ4sWLd5KeP1/JSUlhZycnHcSXq9DI77+AwiCQMqDZ6pCK+YJz1Jy1coamelTqb41lZ6LrEr1\nrangVPaNQi9ssatOkWl5bLPTmFDGgkKhiBxFLtpIqX1pOxJBwYYa/XlibPNSTdUpR4VCGa+rJDTP\ngAEQHAxnHl0ltyifauU/w86sPKVzDNjGi6ElBEFgwYIFLFy4kE6dOhESEqIRXho0fAAkEskrp7j+\nzZQWJ6yEkim+Tw0XFxciIyPfqo6+vr5aOIo3pWShwf8nypUrV2q8u/eBRnz9yxAEgaT7mc+nDp+K\nPlpZaXlqZcuYG/xlzXpu2bJyNHunGFdHgCPlKyKRF+N54yQ06spliQ7Jxu25mxuP5LLSCXV+w7Ho\nqtX+S3zJZODrC9u2KYOlzp8Po0aBRAJH7pwHoG2VV1m95MDo56/9ADsEQWDu3LksXboULS0tTQwv\nDRo0aNDwyaMRX58wgiCQFJ+p6gwf85TsdHWhZWJhgPNzS1bJX/mKpu9FiGQDJVG7yp/ZzqnifPrn\nXWSvXjWytAyYcWkt+sX57HPqgrVFDSap1L4H3AbMePiwCV27wsWLYGKijFj/om/lo8wkAGrbVH5F\nT9YCFwF7YByCIDBjxgx+++03tLW1Wbp0aamxbjRo0KBBg4ZPCY34+kQQBIEndzPUhJYsM1+trGk5\nQ1FgOde3ppJLBco5vB+hBVAIXEbpyn4G2IMyebZU9ozyUbtp2P0rahXGUUmewRPdRoy/pIx4/WXD\n8XRSO5rS6pWW1o6GDbVJSgJnZ4iMhFesmEcqKc1/IYu/QksEAobEx99jw4YN6OjosGLFCpVI1Bo0\naNCgQcOnikZ8fQJERd7mxJZYZM8K1PaZWRmprDisVN8aSzuT9z61dhyIRCm2olGuT3yZmvcv0bxx\nLWqZa6ONNiMN6mH5f+ydd1hU19aH32HoIEVABAUjFkw0FqyosaaomKgoWBIsqBgVc21EY0XEQjAa\nGyixRaPciGBvCZrYC2LsghUsdOkgbWZ/fwwcGcGIid6Ub97n4dHZZ5dz9jkzs2bttdfv8kYoSAcb\nZ+S1OjzXQgGo4gumT+9FcjJ066YKtK8kJc1LWAikAM4IMRCZTJUBeePGjRQWFqrJtGjQoEGDBg1/\nZzTG19+A9ARVqgfzmsZqwfD1WtpgYVvtjccwJQHPZw9yBFqh8lulA36ArX4GxU3qI5SCntU6YCmr\nBtGleXJaf6kK3JK4ihCjkcnOUVioy969PRg3Dr79Fkrz9VUZXd1HwDIAFIpvmDp1Ci1atGDo0KG8\n9957r37BGjT8Dzh37hwTJ06kfv36gCrbdu3atVmyZAm6urqkp6cTEBBAQkICCoUCGxsbpk+fjpWV\nFQAXLlxg9erVlJSUkJ+fj6urq5Tgs4yHDx/i7e1No0aNKhXXfvToEZMnT2b79u1q5WW6guWlc/Lz\n85kyZQpZWVkYGBgQGBhYaeLO6dOnM3v27L98t+OqVav49ddf0dbWZsaMGWpZzkElFL5jxw7pGubN\nmyflq/r55585dOiQlOdr+fLluLi4SPeqPLt378bQ0FDSBwwJCWHz5s0cOXLkb72xJyIignv37lXI\nXfYyjh07xoYNG9DS0kKhUDBgwAA++eST13JO3bp1k7Qt3xTp6elMnTqVgoICatSowaJFiyTNRlDN\ny86dO8nJyUFXV5ebN29y6tQprl27xpIlS9DW1sbZ2ZlJkyZRUFDA3LlzWbx48ev/HhZC/CP+Lly4\nIN40/4sxyrM/+IJwwU+44CdObL/+Px27PDFCNclWQoj9Qoi00vIxpeUthRBpBTliRdI2EZTxo1h6\ncYuqwo1tQixBiPUNhVAqSlvlCyG+EkqlthAC8fBhLdG7936xevWLx49PTxQbz+0RnVeNEk0DB4qo\nB+XnolhkZHQUQiCKiz8V48aNE7a2tqJBgwYiNTX1tc2Bhlfjf/1e+Sdy9uxZMXHiRLWyyZMni4MH\nDwqlUikGDx4sfv75Z+nYqVOnRL9+/URJSYl48OCB6NOnj/SMP336VLi5uYljx46p9bdr1y6xaNEi\n6fXz9+Xhw4fCzc2twrkFBgaK8PBwtbKNGzeKlStXCiGECA8PF/Pnz6/Qbv/+/WLz5s1Vufw3yrVr\n14SHh4dQKpXi8ePHwtXVtUKdKVOmiKtXr1Yonz9/vvjoo4/U7k1WVpYYPXp0hbp5eXnC09NTrax3\n795iwYIFFebvRfxV75Xw8HARGBj4yu26dOkisrKyhBBC5OTkiG7duom0tLSXtKoaXbt2FQUFBa+l\nrxcxf/586d6sXbtWbNy4sdJ6Fy5cEL6+vuK///2vEEKIPn36iNu3bwulUikGDRokYmJihBBChIaG\nioiIiD90LqX3vlKbRuP5+ptgYPLX/4IyBylmKxJYC+gKwfC70Qz8eT154im2NpaMbNRXJcQY9bWq\ncqupINMCfgG8gDsIIWP16vEEBi5k40YTShVUAJXBfzvtAUdunefonShupT6QjpkZVKOOeVkusFxg\nEGZmJykuNmb8eBn79+/C2NiYLVu2SJprGjS8HBfgwGvusxewv8q1i4qKSElJwdTUlGvXrlGtWjW1\n5fL27dtjb29PVFQUFy5coG/fvtIzrq+vz/r16zE0NJTqJyQkEBwcTEFBAfb29jRv3hxfX19MTU3R\n09Nj/vz5auMfPnyY4OBgqlevTnFxseQFKmP48OFSJvqEhIRK319btmxh9WpVjOf58+dZtWoVoMof\nFRAQgI6ODmPHjsXMzIxOnTrRqVMn/P39AZXE0cKFCzE0NGTOnDkkJSWRkZFRqbD2y7Qdo6Oj6dix\nIzKZDFtbWxQKhSSdVMb169cJCQkhNTWVLl26SEoXTk5OvP/++/z4449SXRMTE/T09IiJiVGTVtq7\nd68ksAwqb6a9vT2DBg3Cx8dH8hxevnyZBQsWIITA2tqaJUuWEBsby4IFC8jNzcXBwYElS5YwevRo\nfH19qVevHqGhoaSlpdGvXz+1OWvWrFmFea1bty5BQUFERkaiUCgYPHgwMpmMuLg4SUGgb9++kqB5\nGZcuXWLYsGHk5uYyYcIE6tSpg4+PDzt27ABg4sSJeHp6qnkNLSws2Lx5Mx999BH169fn4MGD6Orq\nsnLlSu7du8eTJ0/Izs5m1qxZtGrVioMHD7Jp0ya0tLRo2bIlU6dOJSkpCV9fXwoLC8nMzGT8+PFq\nz3poaCinTp1i6dKlXLp0iWXLliGXy7Gzs8PPz4+9e/cSHh6OUqnkiy++wNnZGVB5g5cvX17huS0v\nhh4dHS3d606dOrF06dJKkwLfu3ePO3fuMHfuXADefvttMjMzKS4uprCwELlcDkDPnj0ZNWoU/fr1\nq9DHn+GNGV+Ojo5aQBDQDFVGzFGxsbF3yh2fBAwqfXkgNjb2xYlW/oUoFEqS72e8vOKfHQdVmojU\n36mT8NzrbGBk6f873DjMtfRo0vIzaWztwLS2w1W7EeN+htRLYGgN77iUttgAwM2bjRk5MoTMzPYc\nOQJlnvzsgjw2nN9N5K1zPCzd2QhgrGtAp3ot6d6gDe3faoqhrj6QCPQGLlJQYMrYsZ356aejmJiY\nsHXrVknEV4OGvzNnz57Fw8ODJ0+eoKWlhbu7O87Ozhw4cAA7O7sK9e3s7EhISCAlJaWCvmJ54WhQ\nydt4eXlx7949hgwZgqurK8OHD6d///5ERkayePFivvzymeB8YGAgYWFhmJmZSULGzyOXyxk6dCi3\nbt1i48aNascKCgpITEyUDJzbt28TGBiItbU1a9as4dChQ3z88cekpqZKRoC7uzsLFy6kfv36hIWF\nsW7dOtzc3GjevDlubm4UFhZWany9TNsxNzcXMzMz6bWRkRE5OTlqxpeLiwtDhgzB2NgYb29vfvnl\nF7p27UqvXr04d+5chT4dHR05f/682ryfP39ebWk2LCxMktvR1dXl8uXLNGvWjNmzZ7Ns2TLq1avH\n1q1buXv3rlSWmZlJTEwMd+/efeH1lJ+zrVu3VpjXzp07c/z4ccLCwigqKuKbb75h4sSJuLq6MnXq\nVE6cOEHbtm3VDC8AAwMDQkJCSE9Px83NjcjISPT19blz5w6WlpY8evSownJtcHAwmzZtYvLkyaSn\npzNo0CBJGFxfX5/Nmzdz+/ZtpkyZwubNm1m5ciXh4eEYGBjg4+PDqVOnkMlkjBgxgrZt23Lx4kVW\nrlwpGV9btmzh5s2bLF++HC0tLWbPns22bduwsLDg22+/ZefOnWhra2NiYqImbwXQqlUrtmzZ8sJ5\nBNWzUfZeKXsuKmP37t1qgueOjo58/vnnmJmZ4ejoKP04MTU1JSMjg5ycnArvwT/Dm/R89QX0Y2Nj\nnR0dHdsB3wB9ABwdHR1QpSdvCwjghKOj487Y2Ngrb/B8/hZkJOfy07rfOBTyG6kPsqRyw2oVs2P9\nGXJQmULLUYnxVIWyM/ABHgA2ybH0MkvGuHZDOtk7Mcj+I7TKdiKWeb3e7wLaLYAUFApd/PxmsWjR\nNN5/X5eDB8HUVFWtWFHCpF1LuPDoJgDmBiZ0a9CK7g3a0Ma+CTry8o/iDVQJWR8A9bh9+2saNozl\n/Pk7hIaGVviw0KDh5VTdQ/U6adeuHcuWLSMjIwNPT09q164NqPT1ysSGyxMfH0/79u1JSUkhKSlJ\n7VhMTAxCiBcKa6ekpEhi2q1bt1bTLUxLS8PY2Bhzc3MAWrRo8cJz3rx5M3fv3mXMmDFERkZK5VlZ\nWVL7smtYsGABhoaGJCcnSz+IateuLRkBd+/elRKYFhcXU7duXczMzLh69Spnz57F2NiYoqKiCufw\nMs+XsbGxmsh3Xl6e2hejEIJhw4ZJZZ07d+bGjRt07dr1hddtZWVFcnKyWllGRoYktp2VlcXx48dJ\nT09ny5Yt5Obm8sMPP9CsWTOePHkiiWSXxeWVlUVHR1eI1Ss7xzLKz1ll83r//n2aNm2KXC7HwMCA\nWbNmAar7fPLkSSIiIhg3blyFMVq2bIlMJsPCwoJq1aqRmZmJm5sbERER2NraVojlysrKIiEhAR8f\nH3x8fEhOTmbChAk0btwYQBKpbtCgAWlpaTx48ID09HTJmM/Ly+Phw4e0bNmS4OBgduzYgUwmo6Sk\nRBrjzJkzyOVy5HI5T548ISUlRTK+CwoK6NChA/b29tStW7fC9VTF81X2bOjr65OXl1ephmZ2djYJ\nCQnS9WRnZ7N27Vr279+PtbU1X3/9NRs2bGBE9r1fAAAgAElEQVTUqFGAShA8MzPzH2N8dQQOAcTG\nxp51dHQsn5//IdAjNjZWAeDo6KhD5Rvs/jXEnnvMrmVnORMRQ0mxEgBLOxPSHmZjaKJHg9bPZ4X/\nY8QDK4HvUHmwABxQWbm/hwyVNfwzEAIYK4sZU3AGYzMjrLTM6WPf9ZnhlfwbZESCqxzqqlz3sbHv\n0adPCLGxjZgyBQICoNRrC8C3x7dy4dFNLI3MWOwyAafabyOvVBLjV1R2exZPn7bh8eNVFBVpMX16\nP4YPH46NjU0lbTRo+Htjbm5OYGAgQ4cOZdeuXTg5OZGWlsbRo0fpVromf/z4ceLj42nTpg12dnaM\nHz+eXr16Ub16dfLy8pgzZw7jx49/ofFVo0YNHjx4QMuWLYmKipIMMVAt+eXk5EhLc1evXqVmzZpq\n7deuXYu1tTV9+/bF0NBQWnYpfw3lDZ5Zs2YRGRmJsbEx06ZNk4yJ8lI3devWJSAgAFtbW6Kjo0lN\nTSUiIoJq1arh5+dHfHw827dvRwihFtD8Ms+Xk5MTgYGBjBw5kqSkJJRKpZrXKzc3l969e3PgwAEM\nDQ05d+4c/fv3/90+s7KyJEOrjOrVq0uekz179tC/f3+mTZsGqHQVu3fvTnp6OjVq1CAuLo633nqL\nkJAQ6tatK5UBUpmuri6pqanUq1ePGzduSOLO5eessnl1cHAgNDQUpVKJQqHAy8uLtWvX4u7uznff\nfUdGRkYFTynA1atXAZVnLT8/H3Nzc3r06MGGDRswMzOrYMgUFRUxceJEtm3bho2NDVZWVlhaWkqG\n4fXr1+nTpw+3bt3C2tqa2rVrY2Njw4YNG9DR0SEiIoK3336b5cuX4+bmRufOnQkPD2fnzp3SGEFB\nQcycOZPQ0FAGDhxIzZo1CQoKolq1ahw5cgRDQ0MSExMrlUyqiufLycmJY8eO4erqyvHjxyvVb4yK\niqJJkybSa319fQwNDaVl/Ro1apCeni4dz87OrnTzyZ/hTRpfJkBWudcKR0dH7djY2JLY2NhiIM3R\n0VEGBAK/xcbG3npZh9HR0W/oVN/sGCn3cljz6SkUxUpkWtCoszWtB9iTGpfLoW+yqdvGnMtXLv2p\nMa4ZGrLV2pqj5uYoSj/EnHJyGJKczHtZWchf0h4gV0uLgY0bI9PRYXbGMUzMjMjKyqX5kzpcVpad\nn4JGymEYDQd0FRQWVmP27MUsWfI5urowZ04cn3zyhEvlLudM0g1+uHkQuUwLL8deaKUWcCn1twrj\nV69+gDp1/NDSKuHx486MHKnLgwfD1MRuExKeXyTV8Ffxv3g//pO5desW6enpavPUvXt3pkyZwsSJ\nExk7dixbtmyRPFQWFhZ4e3tzqfTN07dvX4YPH46WlhZPnz6la9euGBsbq/UXFxdHUlKS5F3ZtGkT\nGzduRC6XM3r0aK5du0ZeXh6XL19mxIgRDB48GCMjI7S1tYmLi1Prq0GDBtKSk1KpZNiwYRXusb6+\nPkePHsXU1JR27drxySefYGRkhKmpKVlZWTRs2JC8vDyp3cCBAxk3bhxKpeoHp5eXF6ampmzevJlj\nx46hp6eHtbU1kZGRr/zlZmtrS+/evRFC8NlnnxEdHc2pU6coKCige/fu9OvXjwEDBqCtrU2TJk3U\n5q6ye3P8+HHc3d3VymrWrMn+/fvR0tJi8+bNjBs3Tu14ixYtWLZsGYMHD+aLL75AJpNhZmZG06ZN\nKy1zdnZmxowZVK9enerVq6NQKKR7VNZvZfPaoUMHHBwc+PjjjxFC8P7770uGVUxMDB9++GGFexUX\nF0dKSgr9+vWjoKCAzz77jIsXVdq5derUITs7u9Kl0MGDB+Pp6YmWlhZKpRInJycMDAxISEjg5s2b\nuLq6UlhYyPDhw7l//z5dunShX79+KJVKrKysqFmzJo6OjsyePRszMzMsLCxISEggOjqawsJCLl68\nyMcff8zs2bOpVq0a7u7uDBkyBCEEBgYGjB07Vu25flU6dOhAcHAwGzdupFq1aowfP57o6GiCgoJw\nd3fH0tKSEydOUKNGDbX+3dzcGDhwIDo6OhgZGTFmzBiio6PJy8tDS0uLmJiYVz6X3+VFkfh/9q9h\nw4ZLGzZs6F7u9aPnjus3bNhwW8OGDdc2bNhQ/rL+/sm7HcOXnBYu+ImZ728RyfGZUvmsD38QLviJ\nyO8v/aF+i4UQ24UQzuLZRGkLIT4VQvyRKxld2kfP4jSxNjNMLEvaIjqHjCpX45IQxU2l0WKu9hK2\ntgkChGjRQogbNyr2eTP5vmizzEM0DRwofvzt8AtGVgoh/KV+s7O9RZ8+nwhbW1vRokULsWvXrj9w\nNRreJJrdjn9P3vR92bt37wt3j/2TycjIEGPGjKlQnpubK4YOHfqn+n6T90ShUAh3d3eRk5PzSu3m\nzp0rTp8+/UptVqxYIbZt2/ZKbf7OVPW+/PDDD3/4O+j3djtWTQr9j3GK0s1zpTFfV8sOlHq8dgOX\nY2Njx5QtP/5buXFCtZuv62fvUsNeFQT1NLeIq7/GI5NBq54Vc8v8Htmosl7VB9xRJUY1A6ajiu/6\nAajoaP19fgLWCYEuEKhtQU+Djhz6+QxZ2bnAU+ArVa/aVyAHotZ2odG7+0lMtOGrr+Ds2YoZ6zOf\n5jBp1zcUlBTRp0kX3Jp9UMnIxah2SM4CZGRmBjB48H2ioi5ga2tLeHi4FCejQYOGvxYXFxeuX7+u\ntvz4b2DTpk1MmjSpQrmRkRF9+/bl8OHDf8FZ/T4PHz6kX79+9OnTB2Nj4yq38/T0pKCgQNpBqOHF\nFBQUSJ66182bXHbcCXzg6Oh4GlVI0QhHR8fJwB1Ajiqvp56jo2OZut9XsbGxZ97g+fwlKJWCGycf\nAtD4PXup/FLkPUqKFDRyro2pVdWSFcYBK4B1qALqQWWATUSlvVj1t5862YC3KGJi3mlq6dahsW5d\nFNo1SEx6Qrs6ycC7wF2EkKG8pI38ZAmeC1dQpw5s2QKV5TlVKJVM37+ShOxUGls7MPN9z0qS1OUA\nbsBhQB/YxsKFZ/jtt9+ws7MjLCwMOzs7tbV3DRo0/HXIZDICAwP/6tN47Ty/27I8rzvFwOvCzs6O\n3bt3v3K7DRs2/KHxJkyY8Ifa/ZPR19dX27jyOnljxldsbKwS+Py54vKLpvpvauy/E49j08h+8pTq\nNsZY1322NTpq320AWru83Ot1BliKSqhHWVrWBZiMKnvRn3VfzlDmMTjvBDbKHIwL8ijRsUdGOn49\nztGnSRwAxcWNifzWmZ5a6zhwsydOH77LihXPdjM+z6qTP3Im7grmBiYs7TMZPe3nd3M+Lj37y4AV\nKnGjtsya1Z3c3FxmzpxJrVq1/uSVadCgQYMGDX8/3uSyowbg+olnXq8yz49SKYjar0p51rp3g0rb\nCSAcaAe0B3agulkewEVU6Uw/5s/fwL2KDOxyj2KjzMFQy4S+xl3Qlv2IlqwxfZrEUViixY0b/tSp\nE03THFWSSsPOPnz//YsNr59jz7Lh/G7kMi0CP/4PNU2eT9Z4tfTKLgMNSEnZx5dfhvP06VNMTEwI\nCgrSGF4aNGjQoOFfiybD/Rvmemm81zvllhzvXkwkIykXy9om1G1qXWm7X4EBpf+vjsqFOB54PQkp\nVDxW5nEv91dMKUEm12eg0Sn0ZJOAm8hkEBVvzfgpAUSFD+Md6+vUMk1AoW9Dl8+6vLDPO2kPmX1I\nlRhvUudPaW3f+LkakUB/VIudHUhMDMHdfQz37t1DR0dHbWejBg0aNGjQ8G9EY3y9YcqMr8bvPctm\nHbW/dMmxd4MXinU+Kf33PVTJ0gwrrfVnULBNFoWBbjy2IhUXg6+Ry8oS4Vly+84C3nPuTWGaLXp6\nEDQtEpQgd3j/OQHtZ2QX5DFp9zc8LS6k19sd+Kxlr+dqbAJGAyWAG48eLcLd3YP4+HjeeecdJk+e\n/NqvUoMGDRo0aPi7oVl2fIOkPswiJT4LQxM96jSpIZW/SrxXDV6/4ZVScpskRVt8ZF0Zrz+J3gaL\nkMtqAON49OhnPh93n0aOoylMs0XfNo7oaOj8Vmmma/v3K+1TKZTMPLCaBxlJNLSyZ86HXuUMSwHM\nA0agMrx8iI8PwNV1MPHx8TRr1ozt27dXSHCoQcM/mXPnzuHs7IyHhwceHh64urryxRdfSBnd09PT\nmTZtGh4eHgwZMoQpU6aQmvpMCOzChQuMGDECDw8P+vfvz9atWyuM8fDhQ/r06SMl/nyeR48e4e7u\nXqF8yZIlREREVNrm7t27tGzZksLCwkqPT58+/W+x23HVqlUMGDCAQYMGceVKRXGUjRs34uLiIs3/\nvXv3UCgU+Pv7M2jQIFxdXfnll18AWL58OXfu3KnQB6hkaH7++WfpdUhICB07dnzh/PxdiIiIYMmS\nJa/c7tixYwwbNowRI0YwdOhQ9uzZ89rOqVu3bm983tLT0/H09GTIkCFMnDiRp0+fqh2PiIjAw8OD\n+fPn4+7uzrvvvkt2djZnzpxh4MCBfPrpp3zxxRc8ffqUgoICtQTCrxON5+sVKAFOoJJ7rgp3Snc5\nWnSw44BcZefmJeZw+0Iicn1tHnWrS/IL2l74syf7AlJK0tmbdwlt2Ri6GX/NYy1XWpb0YfU2e4JX\ny7l5vjSDvEyJVZe9NO9/lMaNAuHIr6py++6V9htyJoLj9y5iom/Esj5TMNApEwovAsag8nppocq/\nP47MzMtkZ2fj5OTE1q1bK5WA0KDhn06ZvFAZU6ZM4ejRo3z00Ud4e3vj6ekpad6dPn2aMWPGEBYW\nRkJCAv7+/qxbtw5LS0sKCgoYOnQodnZ2dOrUServ4sWLODs7M3369Ndyvrm5uQQEBFTQCCzjwIED\nNG7cGCOjqu3QflNcv36d8+fPExYWRmJiIhMmTCA8PLxCnYCAALVM5hEREZSUlPDf//6X5ORkDh48\nCMCIESOYOnUqISEhan3k5+ezZ88e1q9fL5Xt3buXXr16sX//fjXdx38Lvr6+7N69GxMTE3Jzc+nT\npw8dOnT4x/w4DgoKonfv3ri6uhISEsKPP/6oJqzt6uqKq6sr0dHR7Nu3j/79+2NiYoKvry9bt27F\n0tKSb775hrCwMIYOHUqLFi3YtWvXP0dY+9/IWsD7pbWe0eTEA+oAP79nx5rSMrsDd2gKJHSvS39D\nnZf28TpvUKoig315xykGastjiVfW4cf/OtJ9ylvkJ6miyWTaRVi0PE13t3v07VaL9xv6QtJ5KM6F\n6m9DtYqB8L/ejSb49A5kyFjs8gW1zcri2LJQxXcdQeW/+y/p6R2oXh2aNWvGjh07eOutt14pR40G\nDX8EFxc4cOD19tmrF+x/BcnIoqIiUlJSMDU15dq1a1SrVk0yvADat2+Pvb09UVFRXLhwgb59+2Jp\nqdqsoq+vz/r16yX5E1CpPQQHB1NQUIC9vT3NmzfH19cXU1NT9PT0mD9/vtr4hw8fJjg4mOrVq1Nc\nXCwJB5chhGD27NlMnjy5Up1AUIkir169GlCJTq9atQpQ5UMKCAhAR0eHsWPHYmZmRqdOnejUqRP+\n/v6ASuJo4cKFGBoaMmfOHJKSksjIyKhUWPtl2o7R0dF07NgRmUyGra0tCoVCkk4q4/r164SEhJCa\nmkqXLl0YM2YMJ0+epGHDhnh5eUnXC2BiYoKenh4xMTFqMj179+6lQ4cO0utz585hb2/PoEGD8PHx\nkYyvy5cvs2DBAoQQWFtbs2TJEmJjY1mwYAG5ubk4ODiwZMkSRo8eja+vL/Xq1SM0NJS0tDT69eun\nNmfNmjWrMK9169YlKCiIyMhIFAoFgwcPRiaTERcXx7Rp01AoFPTt21cS5y7j0qVLDBs2jNzcXCZM\nmECdOnXw8fFhx44dgCrFhqenp5peroWFBZs3b+ajjz6ifv36HDx4EF1dXVauXMm9e/d48uQJ2dnZ\nzJo1i1atWnHw4EE2bdqElpYWLVu2ZOrUqSQlJeHr60thYSGZmZmMHz9e7VkPDQ3l1KlTLF26lEuX\nLrFs2TLkcjl2dnb4+fmxd+9ewsPDUSqVfPHFF1JOsqpoO0ZHRzNmzBgAOnXqxNKlS9WMrzLu3bvH\nnTt3mDt3LqB6tsvebyUlJejpqRwIPXv2ZNSoURrj66/kWum/TQH736tYiij1fDXqaE9Z/lFRGu9l\n3bsBvV/SXgdVDq/XgcrwOkYRxSRoazPa8FvOJtTix+8syU+yRb96Ol3d7uDtpU+3d9ujr9PlWePL\npUuOdSouOcalJzBzv+qDYsJ7A+lQt1npkYeocuxeA6yBfdy8acSgQV35/PPPGTt2rNovUg0a/o2c\nPXsWDw8Pnjx5gpaWFu7u7jg7O3PgwAHs7Owq1LezsyMhIYGUlJQKWn3Pi/ra2tri5eXFvXv3GDJk\nCK6urgwfPpz+/fsTGRnJ4sWL+fLLL6X6gYGBhIWFYWZmJgkhl2fVqlV07ty5Uo1AUBkCiYmJkoFz\n+/ZtAgMDsba2Zs2aNRw6dIiPP/6Y1NRUyQhwd3dn4cKF1K9fn7CwMNatW4ebmxvNmzfHzc2NwsLC\nSo2vl2k75ubmYmb2LHWPkZEROTk5asaXi4sLQ4YMwdjYGG9vb3755RcyMjKIj49n7dq1REVF8dVX\nX0nLuY6Ojpw/f17t+s+fP6/m3QoLC8PNzQ0HBwd0dXW5fPkyzZo1Y/bs2Sxbtox69eqxdetW7t69\nK5VlZmYSExNTqZRPGeXnbOvWrRXmtXPnzhw/fpywsDCKior45ptvmDhxIq6urkydOpUTJ07Qtm3b\nCh5LAwMDQkJCSE9Px83NjcjISPT19blz5w6WlpY8evRIzfACJImpyZMnk56ezqBBg/D2Vrkd9PX1\n2bx5M7dv32bKlCls3ryZlStXEh4ejoGBAT4+Ppw6dQqZTMaIESNo27YtFy9eZOXKlZLxtWXLFm7e\nvMny5cvR0tJi9uzZbNu2DQsLC7799lt27tyJtrY2JiYmBAcHq51bVbQdc3NzpfdK2XNRGbt375au\nC1R6jgA///wz586dk55JU1NTMjIyyMnJ+ccIa//reFz6ry/wMhs4N+Mpg6+lINeVs721LbpAcWEJ\ng3+6RwGwoVd9arykj9fJuadXKBTFXMWS24bFyGUKTGW65F1Q5bj99YA5bdu2qbzxg8rjvfKKnjJp\n91Jyi57SvUEbPNv0KT1yCVUOrwSgEXCQa9dyGTTIjYyMDE6ePImXl1cF4V4NGt4Ur+Khep2ULTtm\nZGTg6ekpqTVYW1vz+PHjCvXj4+Np3749KSkpJCUlqR2LiYlBCPFCYe2UlBRJTLt169ZqySHT0tIw\nNjbG3NwcUGkSPs+ePXuoWbMm4eHhpKam4unpqRZnlpWVJbUvu4YFCxZgaGhIcnIyTk5OANSuXVsy\nAu7evcu8efMAKC4upm7dupiZmXH16lXOnj2LsbGxFANXnpd5voyNjdXizvLy8tS+GIUQDBs2TCrr\n3LkzN27cwMzMjC5duiCTyWjTpo0kfA1gZWVFcrJ6IEhGRoa03JaVlcXx48dJT09ny5Yt5Obm8sMP\nP9CsWTOePHlCvXr1APj0008BpLIy3c3nKR9HVH7OKpvX+/fv07RpU+RyOQYGBsyaNQtQ3eeTJ08S\nERFRqbeyZcuWyGQyLCwsqFatGpmZmbi5uREREYGtrS2ffPKJWv2srCwSEhLw8fHBx8eH5ORkJkyY\nQOPGql3r7dq1A1Q6oGlpaTx48ID09HTJmM/Ly+Phw4e0bNmS4OBgduzYgUwmo6SkRBrjzJkzyOVy\n5HI5T548ISUlRTJ0CgoK6NChA/b29tStW7fC9VTF81X2bOjr65OXl1dpSEt2djYJCQnS9ZSxadMm\nDh06xLp16yTPF4ClpSWZmZka4+uvokzSuSoZqG6ceogQ0LCNLbr6qmm+eiyegrxi3mpaQ5IZ+l8Q\nl57Arl+Ok9u6BTvsOtMnKxTMwKTkLfJzdejQAdq2rXwHI0U5kHgWZHKw6ywVCyGYczCYe08e4VC9\nFvN7ji0NsD+MKklGLtAJ2MVvv8Xx6aefkpWVRffu3QkJCdEYXhr+X2Fubk5gYCBDhw5l165dODk5\nkZaWxtGjR+nWrRugEnaOj4+nTZs22NnZMX78eHr16kX16tXJy8tjzpw5jB8//oXGV40aNXjw4AEt\nW7YkKipKMsRAteSXk5MjLc1dvXqVmjVrqrUvH1TerVu3CpnQzc3N1QyeWbNmERkZibGxsVpQspbW\ns31cdevWJSAgAFtbW6Kjo0lNTSUiIoJq1arh5+dHfHw827dvRwihtvP7ZZ4vJycnAgMDGTlyJElJ\nSSiVSjWvV25uLr179+bAgQMYGhpy7tw5+vfvj6mpKceOHeOjjz4iJiYGGxsbqU1WVlaFuKbq1atL\nnpM9e/bQv39/aXPD06dP6d69O+np6dSoUYO4uDjeeustQkJCqFu3rlQGSGW6urqkpqZSr149bty4\ngbW1dYU5q2xeHRwcCA0NRalUolAo8PLyYu3atbi7u/Pdd9+RkZFRqceyTHw7NTWV/Px8zM3N6dGj\nBxs2bMDMzKyCIVNUVMTEiRPZtm0bNjY2WFlZYWlpKRmG169fp0+fPty6dQtra2tq166NjY0NGzZs\nQEdHh4iICN5++22WL1+Om5sbnTt3Jjw8nJ07d0pjBAUFMXPmTEJDQxk4cCA1a9YkKCiIatWqceTI\nEQwNDUlMTFSbkzKq4vlycnLi2LFjuLq6cvz4cVq2rCi2FxUVVWHlJTg4mOvXr7Np0yb09dVzwGdn\nZ7+y8PvL0Bhfr0CZ8VWVXFvPUkw8W6As2+XY5gWJVV83D4uTuFJ4mzU7d3DNzIY4+97oKBV8U5r0\n9IHqFJky5Xc6eXQclCVg4wx6zwzGDef3EHn7PMa6BizrOwUjXQNUwkefAwpgCLCBlJQsBg8eTE5O\nDj179iQoKOiFwbwaNPybqV+/Ph4eHvj7+7NixQrWrFnDwoULJUOjZs2a0g+T2rVr4+Pjg7e3N3K5\nnLy8PAYMGEDnzp1f2L+/vz9fffUVYWFhyOVyFi5cKB3T1tZm0aJFjBw5ElNTU7S1X/2jX1dXF0tL\nS548eYKFhQV9+vTB3d0dExMTLC0tSUlJqdDG19dXikkCWLBgAfXq1WPy5MlER0djYGBAnTp1SElJ\nkQyRqtCkSRNatWrFwIEDUSqVzJkzB1DFaOXn5zNw4EAmTZrE0KFD0dXVxdnZmc6dO1NUVMTcuXNx\nd3dHCCF55QCuXLlSQd+xbdu2XL58mdatWxMWFsbXX38tHTMwMODDDz9k+/btzJs3jxkzZqClpYWV\nlRXDhw/H2tqaGTNmkJeXh4ODA8OHD0dXVxc/Pz9sbGykZa7nqWxe3377bd577z0GDx6MUqlk8ODB\n6Orq0qxZM+Lj4yv1rAHSRo38/Hz8/PyQyWTo6enRunVr0tPT1ZZuQeX9mzVrFt7e3mhra6NQKOjS\npQsdO3bkt99+4+bNmwwbNoynT58yf/58qlevzvDhw/Hw8EChUFCrVi169uxJjx49WLBgAWvXrsXG\nxoaMjAy1cWbNmoWbmxvOzs7MnDlTisEzMjLi66+/JjExscrPwvOMHTuWadOmsX37dszNzSUP8Jdf\nfsnEiROxtbXl/v37avOflpbG6tWreeeddxg9ejSgivUaMmQI2dnZmJiYvP5NJi9S3P67/b1JZfgy\nfm+MYiGElhBCJoQoqkJfU9tvEC74ifP7bwkhhFAqlcKz7grhgp+4cfrh6zjd3+V+0WOxNnOHCM7c\nLgYenisMcjMEQoivhRBC/CSEQPz00/vCwUGIkpLf6ejoRCGWIMTJ2VLRqfuXRPMlg0TTwIHil9tR\nQogMIYS3eHa7ZgghFFL9lStXis8//1wUFVVl5iryv7j3Gl4NzT35e/Km78vevXvFxo0b3+gYfwUZ\nGRlizJgxFcpzc3PF0KFD/1Tfb/KeKBQK4e7uLnJycl6p3dy5c8Xp06dfqc2KFSvEtm3bXqnN35mq\n3pcffvhB7Nq168+MUalNo/F8VZEUVLqK1qgC4X+PwqfF3I5KQCaDt9urgmofxaSRfD8TE0tDGrZ5\nnXnqK/K4JIWf8k+jRKCbdYZuLePoYDSbmqj2HsIDqe7EifC7K4AP1IPtH2WmMG3fCpRCMMa5H13q\n/4YqsD4VlV56EODFiRMnMDIywsnJCW9vb5RKZaVuZA0aNPxzcHFx4csvvyQvL+8vTzfxOtm0aVMF\nrxeoArb79u3L4cOH+eijj/6CM3sxDx8+xNvbm4EDB77SjnFPT09q1Kgh7SDU8GIKCgq4ePHiGxGT\n1xhfVaQsNLYq8V63zidQUqykbjNrjM1Ua8fnS5ccW/ash1z+5oyQp8pYjuRfQIkuTXQP0MFu04sS\n0pOba86IEb/TWV4SpF0DbUOwacfT4kIm7/6G7II83nOow+ftA4DfSiu/B6wAmvPLL78watQodHV1\nOXToEHXq1NEYXho0/AuQyWRv5Ivor+b53Zbled0pBl4XdnZ27N69+5XbPR/LV1UmTJjwh9r9k9HX\n11fbuPI60RhfVeRV4r1unCzVc+xYTlLojcZ7KVFpJq4iQ3mXEvElNvI7GGcd5CstP/L0zOgDlO1V\n3LgRoqLk2Nr25Xc/Vx4cUf1r1xmhpYPfodXEpsZjbwYLey1DS1YM2AGBgDsg46effmLMmDEUFRXh\n5uZW6XZ6DRo0aNCg4f8zGuOrirxasL0qv1dZsH1uxlNunHqIXFuLFh/We41nlQ18D6wCbqnOT1uX\nHrpHmX0gh52O/yWjZhfaAd+WtoiPh9GjVfKM9++/pPv4shQT3dl6cRkHbp7HQKeYZX0jMdGXA18B\n0ygTQNq/fz/jxo2jpKSEkSNHMm/evBdqV2rQoEGDBg3/X9EYX1WkqsaXQqHk5ml14yv68F2UCkHT\nrnWkZcg/x01gNSrDSyV2lFLSjmTFZ7yt7cqXEas5ZVCNjCZd0AM2oorGAli5EhQKGDIESlMOVY4Q\nEK/aeh6l9z1Lf2oEaOHX4xL1LQehMicHxREAACAASURBVLzeKlddEBoaSklJCWPHjmXmzJkaw0uD\nBg0aNGioBI3xVUXKYr5eZnzFXUnmaU4RNR3MsbBVJWQrW3JsVQUh7RejAPah8nJFlivvTKHwJvKp\nIdnKPC6knOe39EQSPacC4I8qzSlAdjZ8953q/5XElpbjITyeDrmPSdI3wOe4AwqhxfDWNfjQMQpQ\nz3eiUCiQy+V899137Nq1i0GDBmkMLw0aNGjQoOEFaIyvKlLVBKtlS45l8V6KEiXRB1WyEn8s3usJ\nsB7VLsL40jJD4DPAGyGacPzpObKVDzERxqw8uJ3EbsMpNDLDGShvY23YoDLAOnWCVq2eH0cAZ1Et\nUIZDtIJCocVkra5k5OnTrk4TvnhvBipx7GeEhoYSGhrK1q1bqVatGoMHD/4D16hBw7+PMomS+vVV\nP7ry8vKoXbs2S5YsQVdXl/T0dAICAkhISEChUGBjY8P06dOxsrICVNm8V69eTUlJCfn5+bi6ulbI\n51S2461Ro0YEBARUOIdHjx4xefJktm/frla+ZMkSHBwc1KRzhBB06tRJSs7avHlzpjyXBLCoqIiZ\nM2cSEBDwl26iUSqV+Pr6Ehsbi66uLv7+/tSpU0etjr+/PxcvXpR2ZQYFBVFQUMDUqVMpLi7GysqK\nxYsXY2BgwLx58xg/fryk7VeeNWvW0LFjRykp59y5c7l8+TK7du168xf6J1i5ciWWlpav/Jm8c+dO\ndu7ciVwuRwjBqFGj6Nix42s5pw4dOnDq1KnX0teLiI+PZ/r06chkMho0aMDcuXPVntWQkBBOnDhB\nTk4OQgjS0tI4deoUe/bsYePGjWhpadG/f3+GDBlCamoqwcHBUh6514nG+KoiVV12fD65aszZR+Sk\nP8WmfnVqNXwVVfhLwEpgG1BQWuYAjAdGAOYohJIzBZe4W/wQHbQ5c/IKSbUbkdGkM/qoLzeWlMC3\npYFfkyeXH6cI2IHK6IpSFWXIEbdhoaw11/OqYWtiRUDv/yB/7sP2+++/Z8aMGQD89NNP9O/f/xWu\nT4OGfz9l8kJlTJkyhaNHj/LRRx/h7e2Np6enpHl3+vRpxowZQ1hYGAkJCfj7+7Nu3TosLS2lZJl2\ndnZ06tRJ6u/ixYs4Ozszffr0P32uDx48oHHjxqxZs+aFdTZt2kTPnj3/8t3LkZGRFBUV8eOPP3Lp\n0iUWL15cQQfw+vXrrFu3Ti0z+YoVK+jXrx99+/Zl5cqV/Pjjj1KS0G+++YZFixap9ZGYmMitW7f4\n/PPPAVVW+4sXL9KwYUPOnTtH27Zt3/zF/g/JyckhKCiI/fv3o6urS3JyMm5ubvz6669/+T2vKosW\nLWLixIm0bduWOXPmcOTIET744APpuJeXF15eXkRHRxMSEsLUqapVoq+//pp9+/ZhaGiIi4sLLi4u\nWFlZYWRkxPnz52nT5gXye38QjfFVRapifAkhyhlfKs9X1P5nuxxfvhSnQGUIrQTK/zroAXgDPSnv\neUpXZnG96C5ayDBMM+Do/VskjFgCqJYbHcv1sHOnKti+fn3o3RsgDQhBFTtWdnXVgTEQ/ZgwTrFL\nvIWetg5L+0zGzEBd02rdunWSGvzcuXM1hpeGvzcRLnD/wOvts24vcK26aGRRUREpKSmYmppy7do1\nqlWrJhleAO3bt8fe3p6oqCguXLhA3759JU+Mvr4+69evx9DQUKqfkJBAcHAwBQUF2Nvb07x5c3x9\nfTE1NUVPT4/58+erjX/48GGCg4OpXr06xcXFODg4qB2/fv06ycnJeHh4oK+vz1dffaVWRwjBnj17\nJKmYQ4cOqWk/Ll++nNu3b7NkyRJ0dHRwd3fH1taWZcuWIZfLsbOzw8/Pj8LCQmbOnElOTg4ZGRm4\nubkxZMgQqZ+8vDzJ2Cmjbdu2aiLI0dHRvPfee4DKQ3ft2jW1+kqlkvj4eObMmUNaWhoDBgxgwIAB\nzJgxAyEESqWSxMREycvn4ODAvXv3yMjIUNOvDA0NVcvvdfDgQZydnenUqRNbt26VjK9ffvmFVatW\nAfDOO+8wb948jh07xqpVq8jPz6dVq1bMmzeP999/n4MHD6Knpyd5H2vVqqU2Z/r6+hXm1czMDH9/\nf65cuUJxcTETJkzgypUrWFtbS9JtI0aMICIiQm0eIiMjOXjwIAUFBcyaNYvs7Gy2b9/OihUrABg0\naBArVqyQsr0bGhqiUCgIDQ2la9eu2NvbExkZiZaWFtOnT0cIQWJiIvn5+QQEBFCvXj22bNnCvn37\nkMlk9OrVi6FDh3Lr1i0WL16MUqkkOzubWbNmSdqfAEuXLiUnJ4c5c+Zw6NAhNm3ahJaWFi1btmTq\n1KmsXLmS3377jfz8fEkVobJnDsDHx0dNHPz69euSodSpUydOnTqlZnyVcf78eUxMTKTnyNHRkZyc\nHLS1tdXkrnr37s3KlSs1xtdfQQGqxT8doKJT+hmJdzPITM7D1MpQ8nKVxXu1rtKS43ygTO7CBJWH\naxzQUK3WE0UWFnJTrOTmvGfghFahjFEH5pPYdRhFxua0B57PWrN0qerfSZMSkMvnAj/wzKP2TmmL\nT+FpAZevvksA7QGY86EXb1urC5xu3LhRMrwWLFjA8OHDq3BtGjT8/+Ps2bN4eHjw5MkTtLS0cHd3\nx9nZmQMHDlSahsXOzo6EhARSUlIqaPU9L+pra2uLl5cX9+7dY8iQIbi6ujJ8+HD69+9PZGQkixcv\n5ssvv5TqBwYGEhYWhpmZmSSEXB4rKyu8vLzo2bMnFy5cwMfHh/DwcOl4XFwcxsbG6OjoSK9DQkIw\nMDBgzpw5nDx5EmtrawoLCwkLC0MIQY8ePdi2bRsWFhZ8++237Ny5k8aNG+Pi4sKHH34oGXvljS8j\nI6OX6vfl5uaqJRaVy+WUlJRIskn5+fl89tlnjBgxAoVCwdChQ2nSpAmNGjWipKSEPn36UFhYyPjx\n46U+HBwcuHjxoppI8/nz59WWZsPCwvDz85OEvpOTk7GwsGD+/PmEhYVhYWHBqlWrSEpKksri4uI4\nc+ZMBaH08pTNGaiWOZ+fVwMDAzIyMtixYwepqan88MMPuLm5MXnyZD799FP27dvHxx9/XKHfWrVq\n4efnx+3bt/nyyy+JiIjA39+frKwsUlNTMTc3V5PZkcvlbNy4ke+//55Ro0ZRXFzM6NGjpftjZ2dH\nQEAAx44dIzAwkKlTp3LgwAG2bduGTCZj+PDhdOzYkTt37jBt2jQcHR3Zu3cvERERkvEVEBCATCZj\n7ty5ZGZmsnLlSsLDwzEwMMDHx0dalnRwcJAExMvo0aMHPXr0+N1no7zhZGRkJGlzPs+ePXvUvLwN\nGjSgf//+GBgY8MEHH0iC3PXr1+fixYu/O+YfQWN8VYEylSkbno94UqfM6/VORztkMhnJcZnEX0vF\noJqumsbjiymL6RoLBADqH7YlQsHpgsvcKLpLT8MO1NGx5W2dukw6sJTHtRxVy41CsFEmo3zS+jNn\nlJw9q4W5eTbDhjUA8kuPuAD/QZUBTPWwpl4IZIqiFSXIGeLUg97vvFfhLJ2dnbG0tGTatGlqH5oa\nNPxteQUP1eukbNkxIyMDT09PapduMba2tubx48cV6sfHx9O+fXtSUlIqfFnHxMQghHihsHZKSork\nyWndurVacsi0tDSMjY0lr06LFi0qtG/SpIkkeN+qVSuSk5PVvsgyMjLUYqIsLCyYNm0aRkZG3Lt3\nj+bNmwMqMW2A9PR0UlJSpASmBQUFdOjQgc6dO/P999/z008/YWxsTElJidp5VMXzZWxsrCbyrVQq\n1fQqDQwMGDp0KAYGBoDqPsTExNCoUSN0dHQ4cOAAp0+fZtq0afzwww+AyvjMzMxUG7f8Nd+9e5fb\nt2+zePFiQJVwNjQ0lE8//RQTExNJlNvb25vU1FSpLC4uTu3cyxClQuTl5+xF83r//n1pfq2srKRs\n/EZGRty5c4e9e/cSFBRUYYzWrVsDKsMiNTUVmUzGJ598wr59+3j06BEDBgxQq5+cnExBQYEU43T/\n/n1GjRoliVO3a9cOUD0/Cxcu5NatWyQkJEg/wLOysnjw4AE1atQgKCgIfX198vLyJEM5LS2N2NhY\n7O1V34cPHjwgPT1d+jGQl5fHw4cPK8xJGVXxfJVfHs3Ly5OMqPLcuXMHQ0NDKU4wJiaGX3/9VRL2\n9vHx4eDBg/Ts2RO5XI5cLn/tKi3/jEXcv5g/Gu9VtuTY4kMHdHR/T8PneVrzvOGVryxgb96v3Ci6\nixZa5ItCACJvnePnhzd4/KHq4V0gk5Xzk+UAq1i69DAAn3++ClXs6XggBtXuyQ8oM7yKC/OYGnWW\nVAxoaVmTyZ0/k3oSQvDrr78ihKBRo0acOHFCY3hp0FBFzM3NCQwMZNasWaSkpODk5ERaWhpHjx6V\n6hw/fpz4+HjatGlD7969CQsLIz09HVB9icyZM6dS8eoyatSowYMHqs+gqKgoyRADMDMzIycnR+rv\n6tWrFdqvWrWK77//HlB9Gdna2qqFSlhYWJCdnQ2oYoNWrFjBsmXL8Pf3R09PTzImyr6gzM3NqVmz\nJkFBQWzZsoXPP/+ctm3bsmHDBpo3b86SJUvo0aOHmhECzzxf5f+eN16cnJw4fvw4AJcuXaJhQ/XV\ngbi4OIYMGYJCoaC4uJiLFy/SuHFjfH19OXv2rDRO+evLysqSDKgyqlevLl1zWFgYkyZNYv369axf\nv57vv/+e8PBwTE1Nyc7Olgw3f39/EhMTK5RduXIFXV1dUlJSEEIQExMjjVM2Zy+aVwcHB+me5eTk\nMHLkSADc3d0JDg7G2tpaLbatjCtXrgAQGxuLra3qG6x///4cOnSIqKioCkLtaWlpTJ06laysLEDl\nOTM3N5e8ndevXwdUsYYNGjTAwcGB+vXrs3nzZrZs2YKrqysNGzZkwYIFfPHFFwQEBNCwYUPpHlta\nWrJ+/Xru3LnD8ePHqV27NjY2NmzYsIEtW7bw2Wef0axZM7U5KU+PHj0qPBvlDS9QLfueO3cOUL2n\nWlXcXcbp06elcUDlVdbX10dPTw+5XK5234UQaGtrv/aYN43nqwpU1fi6cVI9v1fU/jsAtOnd8IVt\nqkK6IosDeSfJFfkYywz5yKg9VnJzsp7msvDIRhK7DaXY2JwOqPxYUAzMAYK4f786ERF30NEpwtvb\nHHgEmFc6TuDueVxSmlJDq5jAAXPRkaseDyEECxYsIDg4mEmTJjF16tRKf01o0KDhxdSvXx8PDw/8\n/f1ZsWIFa9asYeHChaxduxaAmjVrEhISglwup3bt2vj4+ODt7Y1cLicvL48BAwZU+LIsj7+/P199\n9RVhYWHI5XIWLlwoHdPW1mbRokWMHDkSU1NTNS9RGV5eXvj4+HDs2DHkcnmF4PM6deqQnp5OSUkJ\nxsbGODk50a9fPwwNDTExMSElJUXy7IHqy3PmzJl4eXkhhMDIyIivv/4amUyGr68ve/fuxczMDLlc\nTlFREbq6ulWeyw8++IBTp04xaNAghBDStW7cuBF7e3u6d+/Oxx9/jLu7Ozo6OvTp04cGDRrg4eGB\nr68vq1evRktLC19fX6nPmzdvSsHXZbRp04bLly9jaWnJ/v371eR8bG1tadSoEYcPH2bu3LmMGTMG\nLS0t3nnnHd59912pLD8/nzZt2vDuu+8yatQovLy8qFWrVqWfoS+aV1dXV86cOcPgwYNRKBTScun7\n77+Pn5/fCyWfHj16xNChQykqKsLPzw9QeV2NjIxo3rx5heegcePGDB06lGHDhqGvr49CocDNzU2K\n/Tt+/DhHjhxBqVSyaNEi7OzscHZ2ZvDgwRQVFdG0aVOsra355JNPGDduHBYWFtSsWZOMjAxpDJlM\nxsKFCxk5ciTbt2+XNjwoFApq1apFz549q/oYVMq0adOYPXs2S5cuxcHBQYrZ8/T0ZM2aNejq6nL/\n/n2sra2lNrVq1WLgwIEMGTIEHR0d7O3tJVmp2NhYyev4WnmR4vbf7e9NKsOX8aIxlpWehPfvtE1P\nzBEu+In+RotESbFCPM0tFH31FojeMj+RnlRVxfnhpSNtUCuNfnpDBGduF+E5kSJP8VQqn3VgtXgr\nPEAghNBXKkWsdOS4KJu6iRN/FCCEh4fid0fedeUX0TRwoGgZOEBcObFMKlcqlWL27NnC1tZW2Nvb\ni3379lXxWl4f/4t7r+HV0NyTvydv+r6sWbNG/PTTT290jL+C27dvixkzZlQof/TokZgwYcKf6vtN\n3pP8/Hzh6uoqFIrf/3x/Hi8vLxEXF/dKbaZNmyaOHTv2Sm3+zlT1vgQEBIioqKg/M0alNo1m2bEK\nVCXH1/VSPcdGzrWRa2tx+ch9igsVNGhti7l11RXny5OvVAXEt9BrREf9Fnxi1AVDLVWG/FP3LxNx\nJ5rHH44GYKHacqNqSTIry4V169wBmDTpxbf6etJd/H9WZV+doXeHd9uNBVRxFDNmzGD9+vXo6ury\n3Xff4eLi8oeuRYMGDf98hg0bxqFDh1AqlX/1qbxWtmzZwn/+858K5bVq1cLR0bHSZdq/mosXL+Lu\n7s64ceOqvCRWUFCAq6srjRo1qpAXTUNFUlNTyc3NrXTp8s+iWXasAlVZdnxez/H8K+1yLEP1gSYE\nnC+8ys2ie/Qz6o6p3Jgmes+y4+cVPWX+T9+R0G2YtNz4RSW9rVvXm9xc6NoVKomvBSA9P5vJu5dS\npFQwgDu4th4A2nqqazh/ns2bN6Onp8f69evp2rXrK1yLBg0a/m3o6+urBfL/W5g3b94Lj5XfEfl3\nwsnJib17975SG319/QrpKKpK2UaD/09YWVlJy7WvG43xVQWqYnzdOPlsp6MQoly8V1WNr3zgMELA\n6QJLrhbFoIWMJ8pMTOXqnrOVJ34k1sqezMadMKhkdyMUUFIiZ/ly1RZp9aSqzyhRKvDZ+y1JOU9o\nShrTtGOg2bNdRu3atWPRokW89dZbaokdNWjQoEGDBg1/HM2yYxV4mfGVn13I/UvJyLW1cGxXm3uX\nkkhPyKG6bbX/Y+/M46qq1v//PhxmkEFAFAQT5ynNMRzrZmnOoZBDIA5plpoTooYTIkqQXsUUueIQ\nJjdx1hxy6CuJOVKalCOKA8gso4cD56zfHyd2HA+Y3qi897ffrxcvPXuvvdbea2/Yz3meZz0fPNrV\nfcZR/oUQmZxSBfKTWoURCt6y7IqHib769Q8PrrH1l1PcrxRuNDTvDrFz51Du3atD06bQr1/VI/7z\n5DYu3PsZB2MFn5GIaWs/yoxtmD59ulRrxc/PTza8ZGRkZGRkahDZ8/UM/J6o9i/f30erFTTr5IK5\npYkUcny2qvagK3YaxiX1AJLVHVBixFuWXWlgUk+vVWm5msVH1vPg9dGUW9vTnarCjeUIEc9nnx0A\ndALaVaUDHPolkdiLX2NsZESE+D/qKEpRvzyZjz78kIMHD3Ly5EkSExOlOjkyMjIyMjIyNYNsfP0O\nhUAROilr22raVA45QqWq9v0bV3PEk8QA6bQwzSC1zJH2Zi1wMzH0mP3rzG5+tK/Ho1Y9pHCjoV31\nLYmJTTl/vjMODgI/P0Pj71pmKouO6Ja3z3J3pH1qOqXuA5k4+1OOHj2Kra0tMTExsuElIyMjIyPz\nJyCHHX+HyiHH6nxYlZPt8zKKuH4uDRMzJW3fMKzQa0gpyaUnKRcmmCkCGWT1WpWG19XMO/zr8nEe\nvDkegGUKBVWbdl+xYoUuyWvSJAWVpOAAyH9cxPS9n6EqVzOoRTeGZ8TxuAzGfVXO0aNHsbOz46uv\nvqqyAraMjMyzc/bsWTw9PfH19cXX1xcvLy+mTp2KWq0GdBXgAwMDJXmdmTNnkpWVJR1/4cIFxowZ\ng6+vL0OHDjWo7A1w7949Bg8eTGBgYJXncP/+fXx8fAy2R0REGCReazQaQkJCGD58OF5eXnz77bdV\n9jlnzhy96vJ/F2vWrGHYsGEMHz5cKiZamU2bNtG/f39p/lNSUqq9xlWrVnHz5s0qx9m7dy9Hjx6V\nPkdHR9O9e3dKS0v/nAurIXbt2kVERMRzH3fy5ElGjx7NmDFj8PPzY9++fTV2Tv/4xz/+9HnLzc1l\n7NixjBw5kmnTpvH48WO9/bt27cLX15clS5bg4+NDmzZtpIKqGo2GqVOnSgV8VSoVgYGBBoWAawLZ\n8/U7/F6+V1lpOdfP6gKTLbu7cW6/zuv18j8aYm71+0UD08u38p3Km3vlL9PXakiVYcpyrYZFR9Zz\n9zVfKdw4pcre1Ny6dZE9e6IxNdXy0Uf6trVGq2XO15E8yM+khXNDPnExRnE1G2XdDhhb18HBwYF/\n//vftGzZ8nfPW0ZG5vepkBeqYObMmZw4cYI+ffowefJkxo4dK4lrnz59mokTJxIfH09aWhohISFs\n2LABR0dHVCoVfn5+uLm56eVgJiUl4enpyZw5c/7wue7du5fy8nL+/e9/k5GRwaFDhwzaHDx4kFat\nWmGlk8r420hOTubcuXPEx8eTnp7OlClT9HQoK9qEhYXRunVraduuXbuqvMYxY8Ywa9YsoqOj9foo\nKSlh3759xMTESNv2799Pv379+Prrr/V0H/9XWLRoEXv37sXGxoaioiIGDx5Mt27dDKr/v6isXbuW\nAQMG4OXlRXR0NF999ZWe/rCXlxdeXl5cvHiRAwcOMHToUGxsbLh79y6BgYE8fPhQkl0yNzfnlVde\nYc+ePVLR1ZpCNr5+h9/L97p5MR21qhz3lo7YOFhy7sB14NlWOWqFmlOPCwEbHJRNqc4RGXvha85Y\n1+ZRy6eFGwG+YdWqMQhhxMiRUPcJB9raxO2cvnMJe4tarBg0jfJtPckpMcKh/yzWj3iH9PR0PUkS\nGZn/FfoDB2u4z37A8yhGqtVqMjMzsbW15cqVK9SqVUsyvAC6du2Ku7s758+f58KFCwwZMkTSFTQ3\nNycmJgbLSq7stLQ01q1bh0qlwt3dnXbt2rFo0SJsbW0xMzNjyZIleuMfOXKEdevWUbt2bcrKyqSq\n5RWcOnWKpk2bShXp58+fb3ANsbGxfP7554CuFM2aNWsAnYcgLCwMExMTJk2ahJ2dHT179qRnz56E\nhIQAOomj0NBQLC0tWbBgAQ8fPiQvL4+ePXtK+o8VVFSGr6BCyLqCixcv0r17dxQKBS4uLmg0GnJz\nc/UkdpKTk4mOjiYrK4vXXnuNiRMnVnuNNjY2mJmZSfqPFezfv59u3bpJn8+ePYu7uzvDhw8nICBA\nMr4uXbrE0qVLEULg7OxMREQE165dY+nSpRQVFeHh4UFERATvv/8+ixYtolGjRsTFxZGdnc0777yj\nN2dt27Y1mNeGDRuydu1ajh07hkajYcSIESgUCu7cuUNgYCAajYYhQ4awc+dOPaWAH3/8kdGjR1NU\nVMSUKVNo0KABAQEB7NixA4Bp06YxduxYPYkeBwcHvvjiC/r06UPjxo05dOgQpqamREZGkpKSQk5O\nDgUFBQQFBdGxY0cOHTrE5s2bMTIyokOHDsyaNYuHDx+yaNEiSktLefToER999JHesx4XF0diYiIr\nVqzgxx9/ZOXKlSiVStzc3AgODmb//v3s3LkTrVbL1KlT8fT0BHTe4FWrVuk9K/7+/npi6BcvXmTi\nxIkA9OzZkxUrVugZXxWkpKRw8+ZNFi5cCOgM7ZCQEP71r3/ptXv77bcZP368bHz91fxegdXkXyWF\nWnZ3p0yt4YcjKcCz5Xv9ov6KHK0r1opHtDPzq7LNndw0IpMO8cBXJ/VRfbgR8vL2sXHjCkCXaF+Z\n4zfOseHsHowUCsIGfoz1nQRGxuRSoq3L9kn/oLaZmWx4ycjUMGfOnMHX15ecnByMjIzw8fHB09OT\ngwcP4ubmZtDezc2NtLQ0MjMz9YwA0OnPVcbFxYUJEyaQkpLCyJEj8fLywt/fn6FDh3Ls2DGWL1/O\n7Nmzpfbh4eHEx8djZ2cnCRlXJi8vj9TUVNavX8/58+eZO3euXqhTpVKRnp4uGTg3btwgPDwcZ2dn\noqKiOHz4MAMHDiQrK0syAnx8fAgNDaVx48bEx8ezYcMGvL29adeuHd7e3pSWllZpfFVILlVHUVER\ndnZ20mcrKysKCwv1jK/+/fszcuRIrK2tmTx5Mt9+++1Tr7FZs2acO3dOb97PnTun592Kj4+X5HZM\nTU25dOkSbdu2Zf78+axcuZJGjRrx5ZdfcuvWLWnbo0ePuHr1Krdu3ar2eirP2Zdffmkwr7169SIh\nIYH4+HjUajWfffYZ06ZNw8vLi1mzZvHdd9/RpUsXA4kmCwsLoqOjyc3Nxdvbm2PHjmFubs7Nmzdx\ndHTk/v37BtqI69atY/PmzcyYMYPc3FyGDx8uaWuam5vzxRdfcOPGDWbOnMkXX3xBZGQkO3fuxMLC\ngoCAABITE1EoFIwZM4YuXbqQlJREZGSkZHzFxsbyyy+/sGrVKoyMjJg/fz7btm3DwcGBf/7zn+ze\nvRtjY2NsbGxYt26d3rl17NiR2NjY3302Kn5XKp6Lqti7d6+eZuiTv28V2NrakpeXR2FhocHv4B9B\nNr5+h98LO/4mpu1GckIqj4vUNGjtRJ0GdtUcoeOxtphzpbqiql0trDBRmBm00Qotwd/8i9s9R1Fu\nZUcPqgs3ghAlhIS0objYmt69S3j55d++IafkPCDooE7xflrPkTSzqc+IdwfzY7oprk61KCpRYSjJ\nKiPzv8PzeKhqkoqwY15eHmPHjpW0D52dnXnw4IFB+9TUVLp27UpmZiYPHz7U23f16lWEELRo0aLK\nsTIzM6UvUJ06ddIrhpqdnY21tTX29jpd16pyOu3s7HjttddQKBR07tyZO3fu6O3Pz8+Xjq+4hqVL\nl2JpaUlGRgbt27cHoH79+pIRcOvWLamAaVlZGQ0bNsTOzo6ffvqJM2fOYG1tLeXAVeb3PF/W1tZ6\neWfFxcV6L0YhBKNHj5a29erVi59//vmp1+jk5ERGRobeeeTl5Unhtvz8fBISEsjNzSU2NpaioiK2\nbt1K27ZtycnJoVGjRgCMGjUKkhHfiwAAIABJREFUQNp28eJFaVtlKucRVZ6zqub19u3bvPzyyyiV\nSiwsLAgKCgJ09/nUqVPs2rWLDz/80GCMDh06oFAocHBwoFatWjx69Ahvb2927dqFi4sLgwYN0muf\nn59PWloaAQEBBAQEkJGRwZQpU2jVqhWge54BmjRpQnZ2Nnfv3iU3N1cy5ouLi7l37x4dOnRg3bp1\n7NixA4VCQXl5uTTG999/j1KpRKlUkpOTQ2ZmpmR8q1QqunXrhru7Ow0bGuZMP4vnq+LZMDc3p7i4\nuEoNzYKCAtLS0qTr+T0cHR159OhRjRpfcsL97/A040urFfyS+FuyfUVh1Wepal8sdmGueISr8U0a\nGntX2WbHpeOcMLfiUcvuWAjBRqq+YULArFn3WbFiCgqFlnnzfjO8CktLmLYngpIyFX2aeTLAoys+\nXgP58a4Kd1stu3btwt3d/XfPV0ZG5j/H3t6e8PBwgoKCyMzMpH379mRnZ3PixAmpTUJCAqmpqXTu\n3JkBAwYQHx9Pbm4uoHupLViwgMzMzGrHqFOnDnfv6r4Mnj9/Xs+TbWdnR2FhodRfVXI5HTp04OTJ\nk4DO0KtXT7/Ujb29vZ7BExQURGhoKMuXL6dOnTqSMVFZ6qZhw4aEhYURGxtLQEAAvXr1YteuXdSq\nVYvPPvuMsWPHolKpDBKa169fT2xsrPRT2fACXXX3U6dOodVqSUtLQ6vV6nm9ioqKGDBgAMXFxQgh\nOHv2LK1bt37qNebn5xvkNdWuXVvynOzbt4+hQ4eyceNGYmJi2L59O4mJieTm5lKnTh3JkIuOjubo\n0aNVbjM1NZUWVfz888/SOJXnrKp59fDw4Oeff0ar1VJWVsaYMWNQq9X4+PgQHx9PTk5OlZ6bivuc\nlZVFSUkJ9vb29O3bl8TERI4ePWpgfKnVaqZNm0Z6ejqgM0gdHR0lwzA5ORmA69ev4+zsTP369alX\nrx4bN24kNjaW9957j7Zt27Jq1SoGDx5MeHg4Xbp00bu/a9euxcbGhri4OOzt7albty5r164lNjaW\nDz74gC5duhjMSQUVnq/KP5UNL9A9GxX3OCEhgQ4dOhj0c/78eb1cwN+joKBA7/mqCWTP1+/wtJyv\nuz9nUZSnwrG+DY5uNpzbX5Hv1bSK1pXR4KgMwcc6hVKxAYXCxKDFw4Jsws/t5cEoXb7E8mrCjRoN\nTJgAGzc2xcREzdatR3n9dZ3+olZoCTr4Oal56TRxdGfRa8OZOHo4yddSaGhXzvbgd3DxkJPrZWT+\nCho3boyvry8hISGsXr2aqKgoQkNDpRBb3bp1iY6ORqlUUr9+fQICApg8eTJKpZLi4mKGDRtGr169\nqu0/JCSEuXPnEh8fj1KpJDQ0VNpnbGzMsmXLGDduHLa2thgbG/7p9/HxYeHChfj4+CCEMJDcMTU1\nxdHRkZycHBwcHBg8eDA+Pj7Y2Njg6OhYpWG4aNEiKScJYOnSpTRq1IgZM2Zw8eJFLCwsaNCgAZmZ\nmTg7Oz/zXLZu3ZqOHTvy7rvvotVqWbBgAaDL0SopKeHdd99l+vTp+Pn5YWpqiqenJ7169UKtVld7\njZcvX2b6E/kaXbp04dKlS3Tq1In4+Hg+/fRTaZ+FhQVvvfUW27dvZ/HixcybNw8jIyOcnJzw9/fH\n2dmZefPmUVxcjIeHB/7+/piamhIcHEy9evWoU6dOlddW1by2aNGCHj16MGLECLRaLSNGjMDU1JS2\nbduSmppapWcNkBZqlJSUEBwcjEKhwMzMjE6dOpGbm6sXugWdsRUUFMTkyZMxNjZGo9Hw2muv0b17\nd3744Qd++eUXRo8ezePHj1myZAm1a9fG398fX19fNBoNrq6uvP322/Tt25elS5eyfv166tWrR15e\nnt44QUFBeHt74+npySeffCLl4FlZWfHpp59Kxt9/wqRJkwgMDGT79u3Y29tLHuDZs2czbdo0XFxc\nuH37drXz/yQFBQXY2NjU/CKT6hS3X7SfP1MZvoKqxnjp1xO4WUX7r9eeF/0JFp+O2CnuXc0S/QkW\nw2uHi/Ly6hXmtVqt+EkVJ9RaMyGEhxBCXWWbj3YuF3Y/nxIIIXpqtaKqHlUqIYYNEwKEsLAoFocO\n9RVC3Jf2R53eIV4Of1d0WzlK3N36uhCfKUXKJ0rh86qDSF9qI0T+86na/138Ffde5vmQ78mLyZ99\nX/bv3y82bdr0p47xd5CXlycmTpxosL2oqEj4+fn9ob7/zHui0WiEj4+PKCwsfK7jFi5cKE6fPv1c\nx6xevVps27btuY55kXnW+7J161axZ8+ePzJGlTaNHHZ8CoLfwo71qtj/W76Xu1TVvsPbjVAqq5/W\n62V3OKVSsq94IULMAwy9XoeuJnLAyJhHLbphKQQbq1jdWFwMgwbBjh1gY6Pmm2/eom/fEiqWBiTc\nSmJd4g4UQEDecXbsPo8Q0LBNd75aNZu6H/0INrKqvYyMzLPTv39/kpOTX4g6XzXJ5s2bDbxeoEvY\nHjJkCEeOHPkbzurp3Lt3j3feeYfBgwdjbW39+wf8SkWot2IFoUz1qFQqkpKSGDhwYI33LYcdn0Iu\noAbs0FW4f5KfK1Y69nBn/ZTDwNPzvcpEOWdUFwBoY3oeheLfhmOWFBCSuIMHIxYBunBjoyfaPHoE\n/fvD6dPg5ARHjszilVcSAV1SfWpeOvMOrkEgGJX/I8t3armbb4PFa0v44N1pyMjIyPwnKBQKwsPD\n/+7TqHGeXG1ZmZouMVBTuLm5sXfv3uc+buPGjf/ReFOmVLfc638Xc3NzvYUrNYlsfD2Fp+V7ZaY+\nIuteAdb25tR2qcXPp+5hpFTQoc+TptJvXC69xmMBdZQ3aGLyFmBYhPXTE5v5pZs35VZ29BKCj54o\nupqRAX36wKVL4OYGR48+olmzdYASGEqJWsX0PZ9RWFpCp0d32LEjjweFxrRr147h7/n/hzMhIyMj\nIyMjU1PIxtdTeNpKx4qQY4tublw6loKmXEvrXg2wtq9aD/GxtpQfS3WrW7qYH0Gh+Magzf/dusi/\ny8t41KKbbnXjE+HG1FR48024cQOaNoWjR8HdfQdQDryFEE4sPLKKWzn3qZubxcUdd8gsNqZDhw5s\n3bq1yiW3MjIyMjIyMn8tcs7XU3hagdXKeo7nnkFIWyVU2CnTcDP+AVdjb0C/rldhaQkLv/s3D94c\nB8CnCgWV609fvQrdu+sMr3bt4Lvv1Li7fwt8/muL4Ww5f4Bvrp3BijKG5/xAdomSV7t0Ydu2bbLh\nJSMjIyMj84Ige76ewtM8Xz+f+tXz5enKrk9PA0+XFLJXJuBlNQ01HkCywf5/JmzjUpchUrjxw0rh\nxqQkXagxOxu6dcvgwIEA7Oz2ABWVey04k9qMVd9FIso0hJic5R+D3uald8fS5VVPPUkSGRkZGRkZ\nmb8X2fh6CtUZXwU5Jdz9ORtTc2O0AgpyHlPXw576zR2r7OeGOpX6xhFYGIEZUwBzvf3n7yYTU5JP\nfotuWAotGxVGv7okNXz33VUGDGhEQYE5ffseYufOoVhaVqi0twL68yD/bWbv24D6YSFle35E6/sS\n9N3M60rDlZQyMjJ/DWfPnmXatGk0bqzziBcXF1O/fn0iIiIwNTUlNzeXsLAw0tLS0Gg01KtXjzlz\n5uDk5AToqnl//vnnlJeXU1JSgpeXl0E9p3v37jF58mSaN29OWFiYwTncv3+fGTNmsH37dr3tERER\neHh46EnnREdH89133wG62kbZ2dkkJibqHadWq/nkk08ICwursgjmX4VWq2XRokVcu3YNU1NTQkJC\naNBAf/V2SEgISUlJUn2mtWvXUlhYyLx589BoNAghCA4OxsPDg8WLF/PRRx9JWpqViYqKonv37lJR\nzoULF3Lp0iX27Nnz51/oHyAyMhJHR0dGjBjxXMft3r2b3bt3o1QqEUIwfvx4unfvXiPn1K1bN4Nn\nqqZJTU1lzpw5KBQKmjRpwsKFC/We1YrnvLCwECGE9JxfvnyZ5cuXI4TAycmJ8PBwCgoKWLdunVRH\nrkaprgbFi/bzd9T5GvTr4DufaPf9nquiP8EisNcWsWXecdGfYBE19XCVfeaUPxLrHm0Xm/I3CLW2\nvhCiWG9/iVol3vpinjAuyhMIIdYI7a8jvie+/nq4MDcvESCEt/dXorTURggxQAixVgihq9H1WF0q\nfDYHiBYz3hbuDesLFxcXMeH9cX90Kl4o5JpSLx7yPfl9zpw5I6ZNm6a3bcaMGeLQoUNCq9WKESNG\niKNHj0r7EhMTxTvvvCPKy8vF3bt3xeDBg0VWVpYQQojHjx8Lb29vcfLkSb3+9uzZI5YtWyZ9fvK+\n3Lt3T3h7exucW3h4uNi588m/bL8xYcIEkZCQYLB9/fr14vjx40+56r+GI0eOiMDAQCGEED/88IP4\n4IMPDNoMHz5c5OTk6G2bPXu2NOcJCQnio48+EkIIcevWLTFnzhyDPtLS0sT06dOlzyUlJWLAgAFi\nxowZ4syZM890rn/X78p/UperoKBA9O7dW5SWlgohhHj48KHo0aOH0Giqr135PHTt2rVG+nkaEydO\nlO7N/PnzxTfffFNluwsXLkjPuVarFYMGDRJ37ujeq9u3bxe3bt0SQggREREhzp49+x+dy9PqfMme\nr6dQXc5XRYmJVj3cOLtPV9W+unyvsyqdvEMjk9OYKKbyZNGKqNM7ONup/6/hRi2TFEuBBXz1lQ/v\nvbeV8nITxo8/TVSUDUrlQ+C3hH4hBEuORvPTjz9RuOcKWrWWfn3eIHLN2j945TIy/1ss6h/HhYM3\na7TPjv0as+jrZ/cqqNVqMjMzsbW15cqVK9SqVUsSGwbo2rUr7u7unD9/ngsXLjBkyBDJE2Nubk5M\nTIxeCkFaWhrr1q1DpVLh7u5Ou3btWLRoEba2tpiZmbFkyRK98Y8cOcK6deuoXbs2ZWVleHh4UBXf\nfPMNNjY29OjRQ2+7EIJ9+/axe/duAA4fPqwnvL1q1Spu3LhBREQEJiYm+Pj44OLiwsqVK1Eqlbi5\nuREcHExpaSmffPIJhYWF5OXl4e3tzciRI6V+iouL+eCDD/TG7tKli54I8sWLF6Xza9euHVeuXNFr\nr9VqSU1NZcGCBWRnZzNs2DCGDRtGYGCgpM+n0WgwM9Pl3np4eJCSkkJeXp6efmVcXBx9+vSRPh86\ndAhPT0969uzJl19+KUnhfPvtt6xZswaAli1bsnjxYk6ePMmaNWsoKSmhY8eOLF68mN69e3Po0CHM\nzMwk76Orq6venJmbmxvMq52dHSEhIVy+fJmysjKmTJnC5cuXcXZ2ZtSoUeTn5zNmzBh27dqlNw/H\njh3j0KFDqFQqgoKCKCgoYPv27axevRqA4cOHs3r1aqnau6WlJRqNhri4OF5//XXc3d05duwYRkZG\nzJkzByEE6enplJSUEBYWRqNGjYiNjeXAgQMoFAr69euHn58f169fZ/ny5Wi1WgoKCggKCpK0PwFW\nrFhBYWEhCxYs4PDhw2zevBkjIyM6dOjArFmziIyM5IcffqCkpERSRajqmQMICAjQEwdPTk6mc+fO\nAPTs2ZPExETefPNNnuTcuXPSc56SkoKdnR1btmzh+vXr9OrVS/r9GDBgAJGRkVKfNYVsfD2F6sKO\nFSsdXZo6cOdyJuZWJrTpZViwNL08m9TydIxR0cHsBKAvCJr88Bar8zPJ7zUKC62WTUYXMGIR0dET\n+OCDKIRQEBAgCAvryhMVJwCI++EIexKOULD7CqJMy+C+r7F6/cYqpUNkZGT+es6cOYOvry85OTkY\nGRnh4+ODp6cnBw8exM3NzaC9m5sbaWlpZGZmGmj1PSnq6+LiwoQJE0hJSWHkyJF4eXnh7+/P0KFD\nOXbsGMuXL2f27NlS+/DwcOLj47Gzs5OEkKti/fr1rFixwmD7nTt3sLa2xsTERPocHR2NhYUFCxYs\n4NSpUzg7O1NaWkp8fDxCCPr27cu2bdtwcHDgn//8J7t376ZVq1b079+ft956i4yMDHx9ffWMLysr\nK2JjY586r0VFRXqFRZVKJeXl5dLfvpKSEt577z3GjBmDRqPBz8+P1q1bS3OakpJCWFgYn3/+udSH\nh4cHSUlJelqB586d0wvNxsfHExwcLAl9Z2Rk4ODgwJIlS4iPj8fBwYE1a9bw8OFDadudO3f4/vvv\nDYTSK1MxZ6ALcz45rxYWFuTl5bFjxw6ysrLYunUr3t7ezJgxg1GjRnHgwIEqC4G6uroSHBzMjRs3\nmD17Nrt27SIkJIT8/HyysrKwt7fXk9lRKpVs2rSJLVu2MH78eMrKynj//fel++Pm5kZYWBgnT54k\nPDycWbNmcfDgQbZt24ZCocDf35/u3btz8+ZNAgMDadasGfv372fXrl2S8RUWFoZCoWDhwoU8evSI\nyMhIdu7ciYWFBQEBAVJY0sPDQxIQr6Bv37707dv3aY8GQggUv74wraysJG3OJ9m3bx9RUVGATkD9\nhx9+YP78+TRo0IAPPviA1q1b4+npSePGjUlKSnrqmP8J8lu6GjRAxa9K3UrbVcVqbl5Mx8hIQWF2\nCQCvvOWBiZn+VAohOKu6DEBbswNYGk0AfvtjUaYpZ863W7k3WFfcL8KohIZ48emnMwkM1OmHhYbC\nnDmKKg2vi9e/I+LEZpS1LencCOq/1I2V0V+gVCpr4vJlZP6neB4PVU3y6quvsnLlSvLy8hg7diz1\n69cHwNnZmQcPHhi0T01NpWvXrmRmZhq8rK9evYoQghYtWlQ5VmZmpiSm3alTJ73ikNnZ2VhbW0te\nnVdeeaXKPm7evImNjY1B/hToXlCVc6IcHBwIDAzEysqKlJQU2rVrB+jEtAFyc3PJzMyUCpiqVCq6\ndetGr1692LJlC9988w3W1taUl5frjfMsni9ra2u9KvtarVbvS6eFhQV+fn5YWOgiBa+++ipXr16l\nefPmnDlzhsWLF/Ppp5/qef+cnJx49OhRtdd869Ytbty4wfLlywFdwdm4uDhGjRqFjY2NJMo9efJk\nsrKypG137tzRO/cKRCWx6Yo5q25eb9++Lc2vk5OTVI3fysqKmzdvsn//ftauNYx4dOrUCYAmTZqQ\nlZWFQqFg0KBBHDhwgPv37zNs2DC99hkZGahUKinH6fbt24wfP14Sp3711VcB3fMTGhrK9evXSUtL\nw9/fH9CJk9+9e5c6deqwdu1azM3NKS4ulgzl7Oxsrl27hru7OwB3794lNzdX+jJQXFzMvXv3DOak\ngmfxfFXO7youLq5ypf/NmzextLSUnnM7OzsaNGgg5Wf26NGDK1eu4OnpiVKpRKlUotVqazTPUS41\nUQ0ZgBaog74A0LWzD9CUa2nYzplLx+8A0KkKIe0yyjFRFGCuKKCt2XfAR3r7N53bx3evvIXG0pZe\nWg0TxUjmzv1IMrw+/xzmzkXf8FIXwc+xZPy7D5MiF6AuLMXf6BrbVyxiZXScbHjJyLyg2NvbEx4e\nTlBQEJmZmbRv357s7GxOnDghtUlISCA1NZXOnTszYMAA4uPjyc3NBXQvkQULFlQpXl1BnTp1uHtX\n55U/f/68ZIiB7uVSWFgo9ffTTz9V2cfp06fp2bNnlfscHBwoKCgAoLCwkNWrV7Ny5UpCQkIwMzOT\njImKF5S9vT1169Zl7dq1xMbG8sEHH9ClSxc2btxIu3btiIiIoG/fvnpGCPzm+ar886Tx0r59exIS\nEgD48ccfadpU/2/wnTt3GDlyJBqNhrKyMpKSkmjVqhVnzpxh6dKlbNiwgTZt2ugdk5+fLxlQFdSu\nXVu65vj4eKZPn05MTAwxMTFs2bKFnTt3YmtrS0FBgWS4hYSEkJ6ebrDt8uXLmJqakpmZiRCCq1ev\nSuNUzFl18+rh4SHds8LCQsaN05Uk8vHxYd26dTg7O1O7dm2De3b5ss4BcO3aNVxcdDGcoUOHcvjw\nYc6fP28g1J6dnc2sWbPIz88HdJ4ze3t7yduZnKxbqZ+UlESTJk3w8PCgcePGfPHFF8TGxuLl5UXT\npk1ZunQpU6dOJSwsjKZNm0r32NHRkZiYGG7evElCQgL169enXr16bNy4kdjYWN577z3atm2rNyeV\n6du3r8GzUdnwAl3Y9+zZs4Dud6pjx44G/Zw+fVoaB3QeveLiYlJTUwHdYpcmTXTVC4QQGBsb1/gC\nE9nzVQ2/l+/VtIsrxzddAqBTP8N8L1OFCf0sIykSlzFVTAV+s75vZd8nLOce+Z5eWGg1xLCFyR++\nTVTUJJRKwZYtCqRFTdpyuHscfo6FG7tRl6nwu9WJBwfuY2VnxuitYRi3HWYwvoyMzItF48aN8fX1\nJSQkhNWrVxMVFUVoaCjr168HoG7dukRHR6NUKqlfvz4BAQFMnjwZpVJJcXExw4YNM3hZViYkJIS5\nc+cSHx+PUqkkNDRU2mdsbMyyZcsYN24ctra21aYm3L59m27dulW5r0GDBuTm5lJeXo61tTXt27fn\nnXfewdLSEhsbGzIzMyXPHuhenp988gkTJkxACIGVlRWffvopCoWCRYsWsX//fuzs7FAqlajVakxN\nDRU/quPNN98kMTGR4cOHI4SQrnXTpk24u7vzxhtvMHDgQHx8fDAxMWHw4ME0adKEmTNnUlZWxpw5\ncwCddyU4OBiAX375hVmzZumN07lzZy5duoSjoyNff/21npyPi4sLzZs358iRIyxcuJCJEydiZGRE\ny5YtadOmjbStpKSEzp0706ZNG8aPH8+ECRNwdXWt0iNT3bx6eXnx/fffM2LECDQaDR99pPsy37t3\nb4KDg6uVfLp//z5+fn6o1WrpOp2dnbGysqJdu3YGz0GrVq3w8/Nj9OjRmJubo9Fo8Pb2ljyECQkJ\nHD9+HK1Wy7Jly3Bzc8PT05MRI0agVqt5+eWXcXZ2ZtCgQXz44Yc4ODhQt25d8vLypDEUCgWhoaGM\nGzeO7du34+/vj6+vLxqNBldXV95+++1nfg6qIjAwkPnz57NixQo8PDyknL2xY8cSFRWFqakpt2/f\nxtnZWTrG1NSUpUuXMnPmTIQQvPLKK7z22muAznCt8DrWKNVl4r9oP3/1ase9vw7c/4k2n/SOFf0J\nFpvmHBP9CRbTOv7LoJ9LqusiX/Ptrz3YCiHypH3lGo3w3hEqlMWPBEKI1eWpYsSIOAFCmJmVi337\nhBBarRAPk4T4droQ6+oKEYH04ze5q3Cp7ypcXFzE7LmBQqvV1vAsvHjIK+tePOR78mLyZ9+XqKio\naleP/Tdz48YNMW/ePIPt9+/fF1OmTPlDff+Z96SkpER4eXk992rECRMmSCv7npXAwECD1bb/zTzr\nfQkLCxPnz5//I2NUadPIYcdqqCrZXlOu5er39wHITdMl8T0ppH2/PIPTqh/ZWXSPMmEGfIxOmlvH\nv388wjetX0NjaUu3slIOD7lJXNxwrK1VfLsvjYF1lsOWNrC1PVxcCcUPwb4JdA1mjjaAY7vvgFbg\nM3oky5cukxILZWRkZP5sRo8ezeHDh9FqtX/3qdQosbGxfPzxxwbbXV1dadasWbVh2r+TpKQkfHx8\n+PDDD585JKZSqfDy8qJ58+ZV5vXJ6JOVlUVRUVGVocs/ihx2rIaqRLVTfnyIqriMeo3tuXxCFxvu\nXCnfq1xoSHisWxXxsuluTBQm6IyvX/vMz2RR+i3y+0/GXKNB/d49vjvakSmvRxIyaic2PyUAv+Y/\nmDtA8+HQ0hfqdubi3V/YtmUQCOg38h1WLP1UNrxkZGT+UszNzfUS+f9XWLx4cbX7KkJ8Lxrt27dn\n//79z3WMubm5QTmKZ6ViocH/Tzg5OUnh2ppGNr6qoSrPV0WJCfdWTpzde53a9azxeOW3tZAXS3+m\nQFuEvVEe7cz2AbMBXRKkEIK5J7/kdm9domTzyDvMMAli8OK9WBir4BGgNINGg3QG10t94dcK9VlF\neQQeisRmaBtaq5z4V9iaP/fiZWRkZGRkZP40ZOOrGqpKuK8wvoyMdB6njv0aS//P1eRzqfQaAL0s\nPkOpMAdmSMfuSz7JvmaeaCxtqJuUzYmiDti/oltRQv1eOoOr6TAws9U7j5iNMazb8wXq7nXp2Pxl\n/uUzv8avVUZGRkZGRuavQza+quFJz5cQQlrpmHFbt3y4cr6XucIMD5P6mCqOUdf4OhAA6OrDZBc/\nIvDeL+S/PQmTEjUJ517F3iKfx26NsOh7AmzcqzyH9evXSy5Pt2aORHw0HROlfMtkZGRkZGT+m5Hf\n5NXwZM7Xg+s55GeVYONkye1LGRibKmnX+7cCfZZG5vS2NEcrFqCTAJop7ZufEMfNHr4ArE6cSpPi\nWzxuXReLt5JAYbjcGGDNmjUsW7YMAJs3mhA9fTmOVnZVtpWRkZGRkZH570Fe7VgFpUAOOsvU6ddt\nyd/pvF7OL9kiBLz8+ktYWJui0pZyuDiRHE0+EIKRQgAfALoaIkevnyW2bk801ta8kXqM8Zf+haq7\nJRZ9DlVreK1atUoyvKzfbMKSmUG87NKkyrYyMjIvJmfPnsXT0xNfX198fX3x8vJi6tSpqNVqQFcB\nPjAwUJLXmTlzJllZWdLxFy5cYMyYMfj6+jJ06FCDyt4A9+7dY/DgwQQGBlZ5Dvfv38fHx8dge0RE\nhEHidWFhIePHj2fUqFH4+/vrnUtl5syZo1dd/u9izZo1DBs2jOHDh0vFRCuzadMm+vfvL81/SkoK\njx494v3332fEiBFMmjSJnJwcQJdwn52dXeU4UVFRetqRCxcuZMiQIX/ORdUgkZGRxMXFPfdxu3fv\nxs/PjzFjxuDv78+pU6dq7JyqqyFXk6SmpjJixAhGjhzJwoULDVbmCiHo0aMHS5YswdfXV1pAcuTI\nEYYOHcqwYcMkqaesrCw54f6vJP3Xf+vxm3Vake+lKdPdyIqQ47nSK9wpT6Nclc8Aq52AGbqQIxSo\nihifmEPxmC5YlhYTc2Qcoo8C8zafAtUXbXP3eAmFsRFWbzRm5LsjGNa2d7VtZWRkXlwq5IUqmDlz\nJidOnKBPnz5MnjyZsWP+aUcxAAAgAElEQVTHSuLap0+fZuLEicTHx5OWlkZISAgbNmzA0dERlUqF\nn58fbm5uehXok5KS8PT0lIqG/hF27dpF06ZNmT17Ntu3bycmJsag34MHD9KqVSusrKz+8Hh/hOTk\nZM6dO0d8fDzp6elMmTKFnTt3GrQJCwujdevW0rawsDA6dOjABx98wOnTp1mxYgVLly6VXsIVX3or\nSE9P5/r165Lc0ePHj0lKSqJp06acPXtWEtb+X6GwsJC1a9fy9ddfY2pqSkZGBt7e3vzf//1fjVd4\n/7NYtmwZ06ZNo0uXLixYsIDjx4/rCWvfvXuXVq1a8f7770uySRqNhs8++4ydO3diaWlJv379eOON\nN3BycsLKyopz587Jwtp/BVWtdKzI90q7qZPn6NS/MZnlufysTsEIBV3Nd//a8n10ZhuM+ud57o7X\nVetd9u08GhTehUb9gA8NxhRC8NNPP9G4eVP2FCVhN6YjbRu3ZO4bY2r+AmVk/j9j8s4wvrv9Q432\n2aPhK6wZWrXHqSrUajWZmZnY2tpy5coVatWqJRleAF27dsXd3Z3z589z4cIFhgwZIukKmpubExMT\ng6WlpdQ+LS2NdevWoVKpcHd3p127dixatAhbW1vMzMxYsmSJ3vhHjhxh3bp11K5dm7KyMj1dQ4Cm\nTZuSkpIC6ISrq6qCHxsbK4lRnzt3jjVrdCuvVSoVYWFhmJiYMGnSJOzs7OjZsyc9e/YkJCQE0Ekc\nhYaGYmlpyYIFC3j48CF5eXn07NlT0n+soKIyfAUVQtYVXLx4ke7du6NQKHBxcUGj0ZCbm6snsZOc\nnEx0dDRZWVm89tprTJw4kZs3b0q6iO3bt5e8Gh4eHqSkpJCXlyfpXwLExcVJFdIBDh06hKenJz17\n9uTLL7+UjK9vv/1WmouWLVuyePFiTp48yZo1aygpKaFjx44sXryY3r17c+jQIczMzIiIiMDDwwNX\nV1ciIiIwMTHBx8cHc3NzPS/nqlWrsLOzkySKysrKmDJlCpcvX8bZ2ZlRo0aRn5/PmDFjDLyZx44d\n49ChQ6hUKoKCgigoKGD79u2sXr0agOHDh7N69WpJXNvS0hKNRkNcXByvv/467u7uHDt2DCMjI+bM\nmYMQgvT0dEpKSggLC6NRo0bExsZy4MABFAoF/fr1w8/Pj+vXr7N8+XK0Wi0FBQUEBQVJwtoAK1as\noLCwkAULFnD48GE2b96MkZERHTp0YNasWURGRvLDDz9QUlLC0qVLadSoEfBs2o7JycmSodSzZ08S\nExP1jK/k5GQyMjIICQnB0dGRuXPn4uHhwcGDBzE2Npa8oRVfMAYMGEBkZKRsfP0VPGl85aQV8jAl\nD1MLY1RFZbi3dKTOS3bsLj4OQBtTB2orYwBTQPfH+JOIexxo/Bo4Kqh/NYcpybqHHfMtgH59LiEE\nCxYsYPPmzbR97w3uOahwqVePFYNnYGpsgoyMzH8nZ86cwdfXl5ycHIyMjPDx8cHT05ODBw/i5uZm\n0N7NzY20tDQyMzNp3ry53r5atWrpfXZxcWHChAmkpKQwcuRIvLy88Pf3Z+jQoRw7dozly5cze/Zs\nqX14eDjx8fHY2dlJQsaVsbe3JzExkX79+pGfn2/wklOpVKSnp0sGzo0bNwgPD8fZ2ZmoqCgOHz7M\nwIEDycrKYufOnZiamuLj40NoaCiNGzcmPj6eDRs24O3tTbt27fD29qa0tLRK46tCcqk6ioqKsLP7\nLQfWysqKwsJCPeOrf//+jBw5EmtrayZPnsy3335LixYtOHHiBC1btuTEiROoVCqpvYeHB0lJSbzx\nxhvStnPnzuHl5SV9jo+PJzg4WDIGMzIycHBwYMmSJcTHx+Pg4MCaNWt4+PChtO3OnTt8//33BkLp\nlSktLZVCXVFRUURHR2NhYcGCBQs4deoUFhYW5OXlsWPHDrKysti6dSve3t7MmDGDUaNGceDAAQYO\nHGjQr6urK8HBwdy4cYPZs2eza9cuQkJCyM/PJysrC3t7e8nwAlAqlWzatIktW7Ywfvx4ysrKeP/9\n9xk5ciSgez7DwsI4efIk4eHhzJo1i4MHD7Jt2zYUCgX+/v50796dmzdvEhgYSLNmzdi/fz+7du2S\njK+wsDAUCgULFy7k0aNHREZGsnPnTiwsLAgICCAxMVG6H0FBQXrX07dvX/r27fuUJ0P3Pq2ogVnx\nXFTGycmJCRMmUKdOHYQQBAQEsHPnToyNjfnmm28IDg6mV69e0pePxo0bk5SU9NQx/xNk46sKnky2\nrwg52tWxIjM1n04DmnCt7A5ZmjysFBZ0MN+CrjjqWKA+n0aUEXreDWaByeNyvnNrrzO3LOzAyFFv\nLK1Wy9y5c9m6dStGxkbcfPQAFzcPor2DcK6lL/IqIyPzn/E8HqqapCLsmJeXx9ixYyXtQ2dnZx48\neGDQPjU1la5du5KZmWnwsr569SpCCFq0aFHlWJmZmZKYdqdOnfSKoWZnZ2NtbS15dV555RWD49es\nWcP48eMZPnw4V69eZcqUKXpFPPPz8/W8Qs7OzixduhRLS0syMjKkl2v9+vUlncZbt25JBUzLyspo\n2LAhdnZ2/PTTT5w5cwZra2spB64yv+f5sra21ss7Ky4u1jNOhRCMHj1a2tarVy9+/vlnJkyYwNKl\nS/H396dHjx7UrftbnUYnJydJCLuCvLw8yft469Ytbty4IRUbVSgUxMXFMWrUKGxsbCRR7smTJ5OV\nlSVtu3PnjoEweMU5VtCwYUPp/w4ODgQGBmJlZUVKSgrt2rXj9u3bkr6gk5OT5L2zsrLi5s2b7N+/\nn7Vr1xqM0alTJwCaNGlCVlYWCoWCQYMGceDAAe7fv8+wYfq6wBkZGahUKhYsWADotD7Hjx8vhede\nffVVQPf8hIaGcv36ddLS0vD39wd0z8jdu3epU6cOa9euxdzcnOLiYqytrQHdc3jt2jXc3XUr/O/e\nvUtubq70ZaC4uJh79+4ZzEkFz+L5qhweLS4uNtDQbN26NUqlkp9++omOHTuSkZEhGWxvvfUWvXv3\nZs6cOezZs4ehQ4eiVCpRKpVotdoaDb3+dwRx/2KerPFVEXJUFev+SHQa0BQPE1damzamq4UzpopY\nwBgh5rBwIQSGm4DOM0+k+Wxeeqwz3rCopzeORqNh1qxZvxpeSqwHtsC59Uus9w7ipdouyMjI/G9g\nb29PeHg4QUFBZGZm0r59e7Kzszlx4oTUJiEhgdTUVDp37syAAQOIj48nN1eX5lBcXMyCBQvIzMys\ndow6depw967ub8358+clQwx0Ib/CwkKpv6rkcmxsbCRjxcHBwSCp3t7eXm9bUFAQoaGhLF++XPIi\ngP7Lr2HDhoSFhREbG0tAQAC9evVi165d1KpVi88++4yxY8eiUqn0DBHQeb5iY2Oln8qGF+hChqdO\nnUKr1ZKWloZWq9XzehUVFTFgwACKi4sRQnD27Flat27NhQsXGDx4MJs3b6Z+/fp6obD8/HzJgKqg\ndu3aFBQUADqv1/Tp04mJiSEmJoYtW7awc+dObG1tKSgokAy3kJAQ0tPTDbZdvnwZU1NTMjMzEUJw\n9epVaZyKOSssLGT16tWsXLmSkJAQzMzMEELg4eEh3bPCwkLGjdMV6/bx8WHdunU4OzvrXX8FFQsR\nrl27houL7p0ydOhQDh8+zPnz5w2E2rOzs5k1axb5+boalK6urtjb22NioovAJCcnA7pcwyZNmuDh\n4UHjxo354osviI2NxcvLi6ZNm7J06VKmTp1KWFgYTZs2le6vo6MjMTEx3Lx5k4SEBOrXr0+9evXY\nuHEjsbGxvPfee7Rt29bgOaqgb9++es9FbGysnuEFurDv2bNnAd3v1JPSQGvWrGHLli2A7guNi4sL\nxcXFvPfee6jVaoyMjLCwsJDGF0JgbGxc4zlvsuerCp4MO1Z4vgqyH2Ntb07zV10xVijpbvEKMAbQ\nIsRYPv64AZGRwA4Bjgr+oT3HBKOVUNIeSAJLJ71x9uzZw1dffYXS1Birgc1xaOJK1LB5NHEyDEfI\nyMj8d9O4cWN8fX0JCQlh9erVREVFERoaKoXY6tatS3R0NEqlkvr16xMQEMDkyZNRKpUUFxczbNgw\ng5dlZUJCQpg7dy7x8fEolUpCQ0OlfcbGxixbtoxx48Zha2tbZT7Xxx9/TFBQENu2baO8vNwgZ8zU\n1BRHR0dycnJwcHBg8ODB+Pj4YGNjg6OjY5WG4aJFiwgMDESj0QBI+TszZszg4sWLWFhY0KBBAzIz\nM3F2dn7muWzdujUdO3bk3XffRavVSp6a/fv3U1JSwrvvvsv06dPx8/PD1NQUT09PevXqRWpqqrQy\ntE6dOnpz9MsvvzBr1iy9cTp37sylS5dwdHTk66+/Zu/evdI+FxcXmjdvzpEjR1i4cCETJ07EyMiI\nli1b0qZNG2lbSUkJnTt3pk2bNowfP54JEybg6upq4JEBnUevffv2vPPOO1haWmJjY0NmZiZeXl58\n//33jBgxAo1GI0ke9e7dm+DgYMLDw6ucp/v37+Pn54darZby25ydnbGysqJdu3YGz0GrVq3w8/Nj\n9OjRmJubo9Fo8Pb2lvIDExISOH78OFqtlmXLluHm5oanpycjRoxArVbz8ssv4+zszKBBg/jwww9x\ncHCgbt265OXlSWMoFApCQ0MZN24c27dvx9/fH19fXzQaDa6urrz99tvP9hBUQ2BgIPPnz2fFihV4\neHhIOXtjx44lKiqKCRMmEBAQwIEDB7Czs2PZsmVYW1szcOBARo0ahbGxMc2aNWPQoEGAznCt8DrW\nKNUpbr9oP3+mMnwFFWP849dBvxFCFOY9FgMUwWKg0RLRn2AREbBDbC/4RqSVZQkhbgkhlEKrVYqZ\nM28JEEIxslwghLDUFItU4SaEqCPEL1FCRCDE3qF642k0GtHH30u0+LiP6PJPP/Hjg2t/+jX+N/JX\n3HuZ50O+Jy8mf/Z92b9/v9i0adOfOsbfwY0bN8S8efMMtt+/f19MmTLlD/X9Z96TkpIS4eXlJTQa\nzXMdN2HCBHHnzp3nOiYwMFCcPHnyuY55kXnW+xIWFibOnz//R8ao0qaRw45VUDnn65fT9xACzGvp\nchjqjbYmR/uIlLL7wDJAw4kTvnz2mQcmDcpQrC8FYKViOu7cA2LhcZmuQ0sn1Go1c+bM4fbt20T8\nXyzprUywfsmJyHdm09alKTIyMjIvKv379yc5OfmFqPNVk8TGxvLxxx8bbHd1daVZs2ZVhmn/bpKS\nkvDx8eHDDz985pCYSqXCy8uL5s2b06BBgz/5DP/7ycrKoqioyCB0WRPIYccqqBx23PdrvtfjQjXm\ndYwpci0CoI2ZBUJsRqs14oMP5lHLVoPYe4Mi65b01v4f7xtFA3OAt6BEt3pDpbRnwvjxHD9+nCMJ\nx9G844GJ0piVg2fSyb3VX36dMjIyMs+DQqGoNsT130zFooCqqAjxvWi0b99eb0HEs2Bubm5QjuJZ\nqVho8P8TTk5Of1qRVdnz9QSFv/5YAHb8lu8ltIIO893RKrQ0MK6HqfqfKBTlxMWNoKioMR6bv6Go\nbUustEXEGPmhwBP49aY9zuJxmYKxn+ni5RbWlqi7OWNspCR84DS6NfwT4skyMjIyMjIyLySy8fUE\nFdXtXYAyVTnXz+n8YApjcPAyB6B+iR1GRhvRahVs3vwJ4zec4kqf7gD802ga7hQCcYBuhUjJo4f4\n7anNyR9uY2VbC7N3mmPiXIul/Sfzjyad/toLlJGRkZGRkflbkY2vJ6ic73XjQhrlag1GSgUKEwWN\nyt2pXVaHiwejMTVVc/iwD0tXObCmmQUai1q8Kb5hHDFADPBbPF1VkEXeYyNq2VphMrgJxo5WLOoz\nkbebd/0brlBGRkZGRkbm70Q2vp6gcr7Xpqhi7tIOrUZgbGbLyQ0dWfZmU4a/Gw1Al1fnEVjwHY8a\nd8RaFLJBMR4FHwK6isgFBQWUlpZS2yiXMUMtMX6nBcYOVsx9YwxD2rz2d1yejIyMjIyMzN+MbHw9\nQYXxVTsHPvuyOaVYY93UFE2vl1i6TODrG465eSllZUPZlf6IU+3eAipWN9oDuqrSeXl5vPvuu0ya\nNImDeYIIy84o7S2Y0WsUw1/pU/XgMjIy/zOcPXsWT09PfH198fX1xcvLi6lTp0oV3XNzcwkMDMTX\n15eRI0cyc+ZMsrKypOMvXLjAmDFj8PX1ZejQoQaVvQHu3bvH4MGDpdpVT3L//n18fHwMtkdERBgk\nXj969Ij333+fESNGMGnSJEnjrjJqtZqAgAC0Wu1zzUVNU1Hb691338XX15fU1FSDNiEhIXh5eUnz\nX1lmZvPmzUREREifFy9eTHZ2dpVjRUVFceXKFenzwoULGTJkSA1ezZ9DZGQkcXFxz33c7t278fPz\nY8yYMfj7+3Pq1KkaO6du3brVWF/VkZqayogRIxg5ciQLFy40eFajo6Px9fVlyZIlDB48WO+cHj9+\nzPDhw7l16xagW+0oJ9z/RUjGV6EWUODMdTym1ObtLSpWHD3HlClRAOSrJxNobo3GohZ9OMw44oCv\nAHNyc3Px8fHh8uXL/PDTj8zLb4oWIya9+g6jOxnqb8nIyPxv8uqrr0qVuHft2oWJiQknTpxACMHk\nyZN58803iY2NZdu2bQwdOpSJEyei0Wi4d+8eISEhhIeHExsby5dffsnevXtJSEjQ6z8pKQlPT0/C\nwsL+8LmuX7+eDh06EBcXh6+vLytWrDBos3nzZt5+++0ar/b9vBw7dgy1Ws1XX33FzJkzq1yJl5yc\nzIYNG6T5r1WrFiqVilmzZrFt2za9tr6+vnpyTBWkp6dz/fp1WrduDehezklJSTRq1Eiqov6/RGFh\nIWvXrmXDhg1s2rSJsLAw5s2b97cb28/DsmXLmDZtGtu2bUMIwfHjx/X2T5gwgdjYWObPn0/dunWl\nZ+enn35i1KhRkrwR6FY7WllZce7cuRo/T7nUxBNU5HxZXMvGDHMcaufg6q1TVB/d42tMlI8RYhAj\nfi4kr9Nr2Ih8/qV4HwVrgeZkZWVJ2mgubq6Uv+2OsDLG3/g2E7sZfgOVkZH5K+gPHKzhPvsBXz9z\na7VaTWZmJra2tly5coVatWrRu3dvaX/Xrl1xd3fn/PnzXLhwgSFDhki6gubm5sTExGBpaSm1T0tL\nY926dahUKtzd3WnXrh2LFi3C1tYWMzMzgwr1R44cYd26ddSuXZuysjKpankFN2/elDQD27dvb/CN\nXwjBvn372L17N2Cos7dq1Spu3LhBREQEJiYm+Pj8v/buPLyma33g+PckMkkEMYsYomxKDVFCqlw1\nXKqmEDUk5uJRqk2RiiDmqrGiaqjpBulPDNfQitZVXLNSMdxaqoSoSEJCpmaQnN8fJ/ZNZEArid68\nn+fJ0+69z95rnbUS+z1rrbPfvlStWpUlS5Zgbm6Ok5MTM2fOJCUlhSlTphAfH09sbCweHh564mYw\npVIaPXp0trJdXV2z5Uc8e/Ysb775JgBNmjTJNjIFppGxmzdvMm3aNO7du0efPn3o06cPKSkp9OzZ\nEzc3N65fv66/3tnZmevXrxMbG5stf2VQUJD+hHSAffv20apVK9q0acPmzZtxdXUF4IcffmD58uWA\nKb3NjBkzOHz4MMuXLycpKYnXX3+dGTNm0KFDB/bt24eVlRULFy7E2dkZR0fHbG1mbW2do13LlCmj\npyhKS0tj3LhxXLhwgUqVKjFw4EAePnzI0KFDc4xmHjhwgH379pGcnIyfnx9xcXFs3bqVZcuWAdCv\nXz+WLVumJ9cuWbIk6enpBAUF0a5dO6pXr86BAwcwMzPjk08+wWg0EhERQVJSEvPnz6d27doEBgay\nd+9eDAYDb7/9NoMGDeLq1at8+umnZGRkEBcXh5+fX7ZUTosXLyY+Pp5p06YREhLChg0bMDMzo1mz\nZkyYMIGAgAB++uknkpKS9KwIuf3OQc7cjpcvX6ZFixYAtGnThmPHjtGxY0eedPr0aezt7fXfo9TU\nVL744otsyegB3nnnHQICAvRrvigSfD3h8chXyukIKpJO9UGlMbc2o3oJB8qYLwEgJNydHzLXbC02\neONEOzKMXiSmJOI5yIsrV67gVKs6jzpXJ8PaQD9+4cNSCXqmdSFE8XDy5Em8vLy4f/8+ZmZm9O3b\nl1atWvHtt9/i5JQzjZiTkxN37twhKiqKevXqZTuWNXE0mNLbjBw5kuvXrzNgwADc3d0ZMmQIvXv3\n5sCBA3z66afZbiQLFiwgODiYMmXK6ImMs6pfvz4HDx7k1Vdf5eDBgyQnJ2c7HhYWhp2dnZ7nLyws\njNWrV2NjY8O0adM4evQolSpVIiUlheDgYIxGI507d2bLli2UK1eOpUuXsnPnTho0aEDXrl3p1KkT\nkZGR+rTrY7a2tgQGBubbrgkJCXqyZgBzc3MePXqkp8tJSkrC09OToUOHkp6ezqBBg2jYsCH16tWj\ndevWuT7rytnZmXPnztG+fXt93+nTp3F3d9e3g4ODmTlzpp7oOzIyknLlyjFr1iyCg4MpV64cy5cv\n5+7du/q+sLAwTpw4kSNRelaP2wxM05xPtquNjQ2xsbFs27aN6OhoNm3ahIeHB97e3gwcOJC9e/fS\nrVvOWRVHR0dmzpzJL7/8wqRJk9ixYwezZ8/m4cOHREdHU7ZsWT3wetyO69evZ+PGjYwYMYK0tDTe\ne+89vX+cnJyYP38+hw8fZsGCBUyYMIFvv/2WLVu2YDAYGDJkCK1bt+batWv4+PigaRp79uxhx44d\nevA1f/58DAYD06dP58GDBwQEBLB9+3ZsbGyYOHEix44d0/vDz88v2/vp3LkznTt3zrMdAT1JNph+\nl7JON2e1e/duVq5cqW8/Th7+pFdeeYVz587lW+YfIcHXEx4HXw9O/0Zl8zhqDjd9CmpoeQZIJDW9\nE17pjUm3LkVn9jGMfxP+YB/vb/+Ym7ERpNaBElGlSOxUBTNrAz2rVMAn4v8wOMh0oxBF59lHqF6k\nli1bsmTJEmJjYxk2bBjVqlUDTPn1fvvttxyvv3nzJm5ubkRFReW4WV+5cgWj0Uj9+vVzLSsqKkpP\npt28efNs02j37t3Dzs5OH9Vp2rRpjvNHjhzJnDlzGDJkCG+++SaVK1fOdjw2NlYfiQNT8m0fHx9s\nbW25fv26nv+uVq1agGlNW1RUFB9++CFgerr6G2+8Qdu2bdm4cSPfffcddnZ2PHr0KFs5zzLyZWdn\nl+0p+xkZGdnyFNrY2DBo0CBsbGwAUz9cuXIlR0CbVYUKFfRE2Lm9519//ZVffvlFn6YyGAwEBQUx\ncOBA7O3t9aTcY8eOJTo6Wt8XFhaWre6PGbMkE3/cZnm1640bN/T2rVChgj5CaWtry7Vr19izZw8r\nVqzIUUbz5qZHGdWpU4fo6GgMBgPdu3dn79693L59mz59+mR7fWRkJMnJyXquzBs3bjBixAg9MGnZ\nsiVg+v2ZO3cuV69e5c6dOwwZMgQwJSe/desWFStWZMWKFVhbW5OYmKgHyvfu3UMpRfXq1QG4desW\nMTEx+oeBxMREfdova5s89iwjX1mnxBMTE3PNoXnt2jVKliz5TE/5Nzc3x9zcnIyMjBc63S5rvrIw\nkhl8GY2En7pD9Zq3wQzsMqxwKjEPAN9f3+Z+rSbY85DVxve5G7eGIatmo46dx9LcgjK1K1PFyxX7\nMqV5t3FHpqUdxswANBhchO9MCFGUypYty4IFC/Dz8yMqKgoXFxfu3bvHwYMH9dccOXKEmzdv0qJF\nC9555x2Cg4OJiYkBTDeRadOm5Zq8+rGKFSty65bpodBnzpzRAzGAMmXKEB8fr18vt3Q5P/74Iz16\n9GDDhg1Uq1Yt2zQRmIKCuLg4wLQ2aNmyZSxZsoTZs2djZWWlBxOPb1Bly5alcuXKrFixgsDAQEaP\nHo2rqyvr1q2jSZMmLFy4kM6dO2cLQuC/I19Zf54MXlxcXPT1b+fPn6du3eyp2cLCwhgwYADp6emk\npaVx7tw5GjTIP4vIw4cP9QDqMQcHB/09BwcH89FHH7F27VrWrl3Lxo0b2b59O6VLlyYuLk4P3GbP\nnk1ERESOfRcuXMDS0pKoqCiMRiNXrlzRy3ncZnm1q7Ozs95n8fHxDB8+HIC+ffvy5ZdfUqlSJRwc\nHHK8pwsXLgCm5NBVq1YFoHfv3oSEhHDmzJkcidrv3bvHhAkTePjwIWAaOStbtqw+2nn58mXAtNaw\nTp06ODs788orr/CPf/yDwMBA3N3dqVu3LnPmzOGDDz5g/vz51K1bV+/j8uXLs3btWq5du8aRI0eo\nVq0aVapUYd26dQQGBuLp6Unjxo2ztUlWnTt3zvG7kTXwAtO07+P1eEeOHMk1NdDx48f1cp7GaDRS\nokSJF77OUUa+snhobk4KUPHGA+5FGykd/TshLRPZGnsDgyGOyN9bsaz6MACW8BHWie8z4MtArq4/\nTEZ8Cp/2+pBe3Xr894KRZ+HCRbApD7Vl5EuI4uyVV17By8uL2bNns2zZMlauXMncuXNZtWoVAJUr\nV2b16tWYm5tTrVo1Jk6cyNixYzE3NycxMZE+ffrkuFlmNXv2bCZPnkxwcDDm5ubMnTtXP1aiRAnm\nzZvH8OHDKV26dLZRosdq1aqlf2uyYsWK2c4HqFGjBjExMTx69Ag7OztcXFzo1asXJUuWxN7enqio\nKH1kD0w3zylTpjBy5EiMRiO2trZ89tlnGAwG/P392bNnD2XKlMHc3JzU1FQsLS2fuS07duzIsWPH\n6NevH0ajUa/r+vXrqV69Ou3bt6dbt2707dsXCwsLevToQZ06dfK95s8//8yECROy7WvRogWhoaGU\nL1+eb775hl27dunHqlatSr169di/fz/Tp09n1KhRmJmZ8eqrr/Laa6/p+5KSkmjRogWvvfYaI0aM\nYOTIkTg6OuY6IpNXu7q7u3PixAn69+9Penq6nvKoQ4cOzJw5M8+UT7dv32bQoEGkpqbqa/gqVaqE\nra0tTZo0yfF70EcLGU8AAB49SURBVKBBAwYNGsTgwYOxtrYmPT0dDw8PfX3gkSOmLC0ZGRnMmzcP\nJycnWrVqRf/+/UlNTaVRo0ZUqlSJ7t27M2bMGMqVK0flypWJjY3VyzAYDMydO5fhw4ezdetWhgwZ\ngpeXF+np6Tg6OtKlS5d8++lpfHx8mDp1KosXL8bZ2Vlfszds2DBWrlyJpaUlN27coFKlSs90PaWU\nPur4QuWVcftl+ynIzPCPfX3pkhGj0dh8w3mjm/l6Y1fDTOP7nfcYjcbSRqMRY9eI9UaMRmMX4zfG\ne4k9jH+fP8JYrU5NY9WqVY1d33nHGBcX99+L3TpkNK6tYzQuxGg8+GGB1/1/WWH0vXg+0icvp4Lu\nl5UrVxq/++67Ai2jKPzyyy9GX1/fHPtv375tHDdu3J+6dkH2SVJSktHd3d2Ynp7+XOeNHDnSGBYW\n9lzn+Pj4GA8fPvxc57zMnrVf5s+fbzxz5syfKSPXmEamHbOIzvzkVfbwTer2SqH9pdr0WPYz8JCL\nCU35pvIQSvOARb/PZPAXFbi05nsyElN5vUVz/u/rr00LYpNj4bv3YOvfIPYXKPcqNJ9YpO9LCCFe\nhMGDBxMSEvKXevTAswgMDGT8+PE59js6OqJpWq7TtEXt3Llz9O3blzFjxjzzlFhycjLu7u7Uq1fv\nmdY7FXfR0dEkJCTkOnX5Z8m0YxbRmfPaaftjqLHAAhtHCypbnwTA22o2AJ+meDN3R0vUmTMYk9Jw\ndWvJ5n9swsbaGlQwHBwHSZFgbgmuU6C5D5SwKrL3JIQQL4q1tXWuz8P6q5sxY0aexx5P8b1sXFxc\n2LNnz3OdY21tnes3PZ9Fbs9S+19XoUKFAnvIqgRfWURbWGAZlcijBAMVOtiC0YizxUFOpTTlgFUX\nOqft4uQ2Ixci7lK7fRPc3x7KoH6eWKdGQ8j7cD3zD8GxNXRcDeVy/1aSEEIIIYovCb6yiLawwOHo\nLWq9nY65lRmVUNiaPWCK1WeUfnQPq6Ur+deGq9R8tyVrRkzFyb4ChH4J/54MaQlgaQ9tPoNG74FB\nZnSFEEIIkZMEX1lEW1pSZs8tnHuZHtBW1/oIJ40t+FdGW9ov7c2PX1zGmJZOoxgHnB7dh697QYRp\nWpJXesFbAVDKsQjfgRBCCCFedhJ8ZRFtYYHtNw+4m3Aflw5XcbY4RS9jEI2XjODngJ/gUQZv/f0t\nPncvDYFNISMNbKtA+y+gTq+irr4QQggh/gJkbiyLmBQjltGJNHX8F90cPuWiQUN98QPRAQfhUQYd\n2jdng9tpSpyZYwq8Go+GoT9L4CWEyNWpU6f0p5EX5rlCiJebjHxlSgcyjkfQ8N0EOn9ievLwitCW\n2MXdIL68He2c7VnbaBdmDwGHetBxDVRrXaR1FkIIIcRfjwRfmaKAqhtv4LKqEiHmc7E9NJ1zJ+5g\nbWPBsoHQNuM0ZuYW0GIyuPrK4yOE+ItxdMy5HnPgwIF89tlnf+h4brkZn8Vbb73Fvn37sLKyYuHC\nhTg7O9OzZ089BU1aWhrjxo3TE2n//vvvjB07lh49etC9e/c/VKYQ4uUiwVemO0CjyvfA3AKLmIv4\neCViUe1nlvaMpV2JCHB0Mz0+onz++cGEEOJ5/etf/yI2NpZt27YRHR3Npk2bcHNzIykpidGjRzNo\n0CDat29f1NUUQrwgBRZ8aZpmBqwAGgMpwAil1LUsx98DRgGPgNlKqb0FVZdnEZ6aTouhsUBFti26\nhDEDOlZ6QDubBGjzhWl9lzw+Qoi/rKeNVP3Z43+EMTPh8I0bN/T8cRUqVOCjjz7i1KlTnD59Gk3T\nSE1NfeFlCyGKTkFGEz0Ba6VUK+ATQH8ssqZplYEPgDeAvwPzNE0r0nm8I5tCMa9Vnkepj/j5wCW6\nupqxavDrGIb+B5qMkcBLCPFCWFpaEhUVhdFo5MoV0/pSZ2dnPYVNfHw8w4cPB+Bvf/sby5cvZ+nS\npURGRhZZnYUQL1ZBTju2BkIAlFInNU3LmhypBXBMKZUCpGiadg1oBJwpwPrkq2KF9WDWhutH/8Mb\nzaxZPX0+1HEHg6GoqiSE+B9w7Ngx3N3d9e2hQ4cycuRIHB0dsbe3B6B9+/acOHGC/v37k56eni2l\nTfny5Rk3bhy+vr589dVXGOTfJCH+8gyPh71fNE3TvgK2K6X2ZW7fApyVUo80TfMEXlNK+WQe+wfw\nD6XUgbyud/bs2YKpaKZ/XZuFm3aadbsa88HffUi3KFWQxQkhhBDif1yzZs1y/bRUkCNfcUDWCMZM\nKfUoj2OlgAdPu2CzZs1eXO1yXPuf7D/4HeumdyqwMsQfc/bs2QLte/H8pE9eTtIvLx/pk5dTYfTL\n2bNn8zxWkAuZjgFvA2ia1hK4mOXYaeBNTdOsNU0rDdQHLhVgXZ5J+dLliroKQgghhPgfV5AjXzuB\njpqmHQcMwFBN07yBa0qp3ZqmLQP+jSkAnKKUSi7AugghhBBCvBQKLPhSSmUAo5/YfSXL8TXAmoIq\nXwghhBDiZSTPTxBCCCGEKEQSfAkhhBBCFCIJvoQQogCcOnUKTdP49ttvs+3v1q0bn3zyyQsvb8eO\nHYwbNw4vLy+8vLzo0aMHM2bM0I+Hh4frx/v164e/vz8JCQn68QMHDujnenh4EBISkms5u3bt4vvv\nv3/h9X9e58+fx8PDg379+rF8+fIcxx88eICrq6v+njZu3Kgfy8jIYMSIEQQFBQGglMr1GgCxsbFM\nmzZN346MjKRx48bs27fvBb+jF++NN9547nNiYmIYN24cw4cPZ9iwYfj5+ZGc/GKWZAcEBOhtXpCW\nL19Onz596NevHxcuXMh2LDo6Gi8vL2bNmoWXlxevv/46QUFBxMfHM3r0aDw9PXn33Xf56aefAPj8\n88+5du1absX8KRJ8CSFEAXF2dmbv3v9mTlNK8fvvvxdYeW5ubgQGBhIYGMjOnTv5+eefuXjxIsnJ\nyYwZM4YRI0YQGBjI119/TePGjfn4448BOHfuHBs2bGDlypUEBgayevVqFi9enOOmk5SUxO7du+nY\nsWOBvYdnNX36dBYtWkRQUBChoaFcvnw52/H//Oc/vPPOO3p7DB48WD+2dOlSHj58qG9rmsbNmze5\ndetWjnKWLl3KgAED9O0dO3YwaNAgtmzZUgDvquh99dVXuLm5sXbtWtatW4eNjQ1ff/11UVfrmV2+\nfJnTp08THBzM4sWLs30AAVP6rsDAQKZOnYq3tzevvvoqffv2Zf369bRs2ZJNmzYxb948Zs6cCZge\nivzZZ5+98HpKYm0hRLGw8mFwjn31LWrRtuTrf+j46NIeTy2zXr16hIWFERcXh729Pbt376Zbt25E\nREQAsG/fPjZs2ICZmRnNmjVjwoQJ3L17F39/f1JSUnjw4AHvv/8+HTp0oFu3brRo0QKlFAaDgRUr\nVlCqVN4Pg05MTCQ+Pp5SpUpx6NAhmjdvTuPGjfXjvXr1IigoiPDwcIKDgxk8eDC2trYAlC1bluDg\nYP0J/I/t2bNHH01JSEhgypQpxMfHExsbi4eHBwMGDMDLy4uyZcsSFxfH6tWr8ff35+bNm2RkZPDh\nhx/i6upKSEgImzdv1q/7+eef4+DgoG9v2rSJ/fv3Zyt7/vz5VK1aVS87NTWV6tWrA9C6dWtOnDhB\ngwYN9NdfunSJy5cv4+npiYODA35+flSsWJGQkBAMBgNt2rTJdv0uXbqwefNmJk+erO9LSEjg4sWL\n+g3caDSya9cutmzZwpgxY7h69Sp169YlOTmZyZMnc+fOHdLS0pg6dSr169dn8uTJXL16FSsrK6ZO\nncqNGze4fv06EyZMICUlhS5dunDw4MFsbRYQEICfn1+Odg0NDWXOnDkYjUYqVarEvHnz6NWrF/v3\n78fc3JwFCxbQsGFDunTpotc/NTWVjz76iIiICDRNw9/fn/79+zNr1izq1KnD4cOHOXToENOnT9fP\ncXR0ZP/+/dSoUQMXFxd8fHwwGAzcvn2b8ePHU6FCBSIjI2nTpo1+7alTp5KSkoKVlRWzZs2iSpUq\nLFq0iEuXLpGYmEjt2rWZN2+eXsbNmzfx9vZmzpw5ODo6MmXKFGJjYwHw8/ND0zTatWuHs7Mzzs7O\nTJkyRT931KhRJCUl6du1a9fG399f3z579iytW7fGYDBQtWpV0tPTiYmJyfb79bgvZ82axcKFCzE3\nN2fIkCFYWloCkJ6ejpWVKeOhvb09VlZWXLlyhXr16vGiSPAlhBAFqGPHjnz//fe4u7tz4cIF3nvv\nPSIiInjw4AEBAQFs374dGxsbJk6cyLFjxzAYDAwdOhRXV1fOnTtHQEAAHTp0IDExka5duzJ16lQ+\n/vhjjhw5QteuXbOVdfz4cTw9PYmOjsbW1pbRo0dTs2ZNvv/+ez1QyapatWrcuXOHqKgonJycsh0r\nXbp0jtefPn1aT5V08+ZNunbtSqdOnYiMjMTLy0sfIerWrRsdO3Zky5YtlC1blrlz5xIbG4unpyff\nfPMNYWFhrF69GhsbG6ZNm8bRo0fp3r27Xo6npyeenp55tmlCQgJ2dnb6tq2tLeHh4dle4+zsTMOG\nDXFzc2P37t3Mnj2bsWPHsnfvXpYtW8YXX3yR7fWaphEQEJBt3/nz56lVq5a+feLECerWrYuDgwO9\ne/dm8+bNzJgxg6+//hpHR0eWLFnC1atXOX78OKGhoTg6OuLp6UmpUqU4fvx4jmA2q8dtdvny5Vzb\nderUqSxZsoTatWuzefNmbt26RbNmzTh69CitW7fmyJEjjB8/Pts1k5OTmTBhAo6OjowfP56DBw/i\n4eHBzp07mTRpEtu3b2fUqFHZzunfvz9WVlasXbuW8ePH06xZMz04++2331i7di2lSpViwIABXL58\nmTVr1uDl5UXbtm05ceIECxcuZMaMGdjb27N+/XoyMjLo2rWrnpv0xo0bbN++nUWLFlGzZk0WLFhA\ny5YtGTBgAGFhYUyePJmgoCAiIiLYsWMHZcuWzVa/VatW5dmGYPrdKFOmjL5ta2tLfHx8juDr3Llz\n1KlTB2dnZwC9b6Kjo5k4cSK+vr76azVN4/Tp0xJ8CSHE83raSNWfPZ6Xbt264e/vj5OTE6+//t8U\nt7du3SImJoaRI0cCppGq8PBwmjVrxpdffsm2bdswGAw8evRIP+fVV18FoEqVKqSkpOQoy83NjYUL\nFxIeHs6IESOoWbMmAJUqVcqx9gUgLCyMqlWrUrVqVSIiIrLdXM6ePUv58uWpUaOGvi82NpZy5UwP\noy5fvjwbN27ku+++w87OLls9HwcsV69e5ezZs3rZjx490q/h4+ODra0t169fp0mTJtnq9bSRLzs7\nOxITE/VjiYmJOQKbli1bYmNjA5gC4GXLlvHPf/6TyMhIBg8ezG+//YaFhQWOjo60adOGChUq8OBB\n9kQrsbGxlC9fXt/eunUrt2/fZvjw4aSlpXHlyhUmTJjA9evX9ZG0unXrUrduXaZNm5Zj344dO/Rr\nPZna73Gb5dWu9+/fp3bt2gAMHDgQAA8PDwIDA8nIyMDNzU0fuXmsatWqODo6AtC0aVNu3LjBwIED\n6dWrF8OHD+fu3bvZRgvBtFaxZ8+e9OnTh9TUVNasWcPcuXPx8fGhXr16emDTqFEjbty4wdWrV1m1\nahVfffUVRqMRCwsLrKysiImJwdvbm5IlS5KUlERaWhoAR44coUSJEpibmwOm35GTJ0/qa+ji4uIA\n0+jrk4EXPH3kK7ffjdxGiI8ePZojWFVK4e3tzaRJk2jRooW+//Fo34skwZcQQhQgJycnkpKSCAwM\nxNvbWx+hqVatGlWqVGHdunVYWFiwY8cO6tevz+eff46Hhwdt27Zl+/bt7Ny5U7/WsybVdnJyYvr0\n6YwfP55vvvmG9u3bs3LlSi5cuECjRo0ACA4OxsHBAScnJ9zd3Vm0aBGurq6ULFmS+/fv4+vry+ef\nf57tug4ODsTHxwOwbt06mjRpwoABAzh58iSHDx/OUU9nZ2cqV67M6NGjSU5O5ssvv6REiRIsW7aM\nQ4cOAaY1NU8GIk8b+bKzs8PCwoJbt27h5OTE0aNHGTt2bLbX+Pn50alTJ95++219SnLSpEn68YCA\nAMqXL68HSHFxcTlGR8qVK6cHAzExMYSGhnLgwAE9cPDz82Pnzp3Url2bixcv0qFDB8LDw1m6dCmN\nGjXi4sWLtGnTRt/31ltvER0dDZBjjdrjNsurXStWrEhYWBg1a9Zk9erV1KpVi44dOzJ37ly2bdvG\nhx9+mKOd7t69S1RUFBUrVuTcuXP07t0bGxsbXF1dmTNnDj169MhxzsaNGwkPD6dv375YWlpSp04d\nrl+/DsCvv/7K77//jqWlJRcuXKB37944OzszbNgwXFxc+PXXXzlz5gxHjhwhIiKCpUuXEhMTw/ff\nf6/38eDBg6lRowaTJk1i06ZNODs70717d7p168b9+/cJDjZN75uZ5b4k/WkjXy4uLixYsEAPLjMy\nMnL0K5hG4FxcXPTta9euMX78eJYuXZpjhOvhw4f6h44XRYIvIYQoYG+//Ta7du2iVq1aevDl4ODA\nkCFD8PLyIj09HUdHR7p06ULnzp2ZM2cOq1atokqVKvpamOfl5uaGm5sby5Ytw8fHh5UrVzJ37lwe\nPHhAeno6mqaxePFiwDQq0rdvX4YNG0aJEiVITk7G29s7x03I1dWV0NBQmjdvTrt27fD392fPnj2U\nKVMGc3NzUlNTs72+X79++Pn54enpSUJCAgMGDMDOzg4XFxd69epFyZIlsbe3Jyoq6rnf34wZM5gw\nYQLp6em0bt2axo0b8+DBA/z8/Fi+fDkff/wxvr6+BAUFYWNjw+zZs/O9XmhoKK1atcq2r3Hjxixc\nuBAwfcuzU6dOeuAF0LdvXyZNmsQ///lPpkyZgqenJ+np6fj6+lK3bl18fX2ZOXMmJUuWxNfXlxo1\nahAUFET//v1p0KCBvsYuq7zadcaMGfj6+mJmZkaFChUYMmQIYBpZDQkJoU6dOjmuVaZMGWbPnk1k\nZCRNmzalbdu2er379++fbcQoa7vOmDGDLVu2YG1tTdmyZfH39yctLQ0LCwvGjx/PvXv36Ny5M/Xq\n1cPHx0dfo5icnMyUKVOoVq0aK1as0AM4JyenbH3s5uZGSEgIa9asYfTo0UyZMoWtW7eSkJCQI4h+\nXg0bNuT111/n3XffJSMjQ/+m6p49e0hKSuLdd98lJiYGGxubbB9mFi1aRGpqKnPmzAFMAf6XX34J\nwIULF/joo4/+VL1yMBqNf4mfH3/80VjQCqMM8fykX14+0icvp4Lul4SEBOOgQYMKtIyi4u3tbbx1\n61aO/VOnTjVevnz5D1+3oPtk9erVxuDg4Oc6JzQ01Dhx4sTnOic8PNzo4eHxXOe8zJ61X2JjY42j\nRo36M2XkGtPIoyaEEEI8E1tbW3r27JljPdZf3ZUrV6hevXqOLx0AjB8//qV9rMQnn3zCjz/+mO3L\nCk+zadMmpk+fzrhx4wqwZv87NmzY8OJHvQCD8Ym59pfV2bNnjc2aNSvoMijoMsTzk355+UifvJyk\nX14+0icvp8Lol8wycl2oKSNfQgghhBCFSIIvIYQQQohCJMGXEEIIIUQhkuBLCCGEEKIQSfAlhBBC\nCFGIJPgSQgghhChEEnwJIYQQQhQiCb6EEEIIIQrRX+ohq0VdByGEEEKIZ5XXQ1b/MsGXEEIIIcT/\nApl2FEIIIYQoRBJ8CSGEEEIUIgm+hBBCCCEKkQRfQgghhBCFSIIvIYQQQohCVKKoK1AUNE0zA1YA\njYEUYIRS6lqW4+8Bo4BHwGyl1N4iqWgx8gx98hHQL3PzW6XUjMKvZfHztH7J8ppvgF1KqZWFX8vi\n5Rn+VroA0zM3zwHvK6Xka+0F7Bn6ZQLQH8gA5iqldhZJRYshTdNcgflKqb89sb8bMA3TvX6dUmpN\nYdWpuI589QSslVKtgE+ARY8PaJpWGfgAeAP4OzBP0zSrIqll8ZJfnzgDAwE3oBXQSdO0RkVSy+In\nz37JYjbgUKi1Kt7y+1spBSwA3lFKtQTCgPJFUcliKL9+KYPpvtIK6AQsLZIaFkOapk0CvgKsn9hv\nASzB1B9tgZGZ9/9CUVyDr9ZACIBS6iTwepZjLYBjSqkUpdRD4BogN/qCl1+fhAOdlVLpSqkMwAJI\nLvwqFkv59QuapvXB9El+X+FXrdjKr0/cgIvAIk3T/g1EKqWiC7+KxVJ+/ZII3ARsM38yCr12xdev\ngHsu++sD15RSsUqpVOAo8GZhVaq4Bl/2wMMs2+mappXI41g8ULqwKlaM5dknSqk0pdQ9TdMMmqYt\nBH5SSl0tkloWP3n2i6ZpDYEBmIbtReHJ79+v8kA7wAfoAnyoaVrdQq5fcZVfv4DpQ+R/ME0FLyvM\nihVnSqntQFouh4r0Xl9cg684oFSWbTOl1KM8jpUCHhRWxYqx/PoETdOsgc2ZrxlTyHUrzvLrl0GA\nI3AQGAJ4a5rWuXCrVyzl1yf3gTNKqbtKqQTgCNCksCtYTOXXL12AKkAtoDrQU9O0FoVcP5Fdkd7r\ni2vwdQx4G0DTtJaYhukfOw28qWmataZppTENTV4q/CoWO3n2iaZpBmAXEKqUGqWUSi+aKhZLefaL\nUmqSUso1cxHrBmCxUiqkKCpZzOT379dZoKGmaeUzR11aYhptEQUvv36JBX4HUpRSyZhu8mUKvYYi\nq5+BOpqmOWiaZgm0AU4UVuHF8tuOwE6go6ZpxwEDMFTTNG9M87+7NU1bBvwbU3A6JfOPRRSsPPsE\nMMe0INIq85tcAJOVUoX2h1KM5fu3UrRVK7ae9u/XZGB/5mu3KqXkw2PheFq/dABOapqWgWl90fdF\nWNdiS9O0AYCdUmp1Zv/sx3SvX6eU+q2w6iGJtYUQQgghClFxnXYUQgghhCgSEnwJIYQQQhQiCb6E\nEEIIIQqRBF9CCCGEEIVIgi8hhBBCiEJUXB81IYR4gTRNqwlcJeczpboppcLzOMcfQCnl/yfKHQIs\nBm5l7rIBDgNjsj6k9xmvNRP4MfOxAD8opdpl7j+vlPpTDyrVNO0QUA1IyNxlD1wHBiqlIvM57z0g\nQSkV9GfKF0K8XCT4EkK8KHf+bJDyB+1WSg0B0DTNHDgODAdWPc9FlFJZ0yT9Lcv+F/WeRiilDgFo\nmmYGbAO8MaUCyssbwKEXVL4Q4iUhwZcQokBl5oAMAOyAisA8pdTKLMctgHVAw8xdK5RSazRNq4Qp\ngHLClIh4slLqQH5lKaXSMxNKN8y89lDgY8CI6envY4GUPMrbgCnQcck895RSylXTNCOmZO63gKZK\nqUhN0xwwZb6oAbQHZma+5gbwnlLq/lOaxRZTHsZTmWV5ZNbTBrAChgElge7AW5qmRQDnn7c9hBAv\nJ1nzJYR4UapqmnY+y8/EzP0jgNlKqeaYkj4veOI8N8BBKdUU6Aq8mbn/c0xPnW6GKQhZpWlaKfKh\naVo5oBNwQtO014ApQFul1GtAIjA9n/IAUEp9kPlf1yz7HgHBgEfmrt6YnmheBvgU+Hvm9fYD8/Oo\n3leapoVmBlInMT3hfEnmKNho4B2lVGPgM/4bWO0Gpiml9v+R9hBCvJxk5EsI8aLkNe34MdA5M+3N\na5hGwLK6BGiapu0HvgUeB20dgHqZa7HANLJUG9MIUFbdNU07jymlixmwAwgC3gf2ZBmFWg2sxxQs\n5Vbe02wClgDLgf6YAjtXTImSf9A0DUypsGLyOH+EUuqQpmluwHZgp1IqFVNlegHdNNNF/gbklr/0\nWdtDCPGSk+BLCFHQtmJKLLwH+BpT4KJTSt3XNK0B0BFTYuJzmdvmwFtKqRgATdOqAFG5XF9f85VV\n5ohSVgagRD7l5UspdSYzCW9zoJpS6oSmaT2Ao0qp7pllWpMzuHzyOscz88du0TTNBbAGTmMK7o4A\nFzBNjz7pWdtDCPGSk2lHIURB64hp6mwX0AX0hfFk/n93IBD4BvgA0zcCnYCDwJjM17yKaYSs5HOU\newjTqJhD5vZ7mEao8iovq3RN03L7cLoZ07qrx98+PAW00jStbub2VGDhM9RtMaZvPI4C6mJakzYX\n+AFwxxRoATzivx+S/2x7CCFeEhJ8CSEKmj9wVNO0/2BaXxUG1MpyfB/wO3CZzBEgpdRFYBzQUtO0\nC8D/AZ5KqfhnLVQpdQGYBxzWNO0KpvVZfvmUl9UuIDRzJCurTUCTzP+ilLqLaXH8Vk3TLmJarP/x\nM9QtBdO0pT/wK6apwyuZdYrGtJAf4ADgq2laH/5kewghXh4Go9FY1HUQQgghhCg2ZORLCCGEEKIQ\nSfAlhBBCCFGIJPgSQgghhChEEnwJIYQQQhQiCb6EEEIIIQqRBF9CCCGEEIVIgi8hhBBCiEIkwZcQ\nQgghRCH6f3w+wNr4SvhSAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x24ca9fda0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Decision Tree\n",
    "% matplotlib inline\n",
    "speaker_pred = dict.fromkeys(list(pd.Series(y.index).unique()))\n",
    "for key in speaker_pred.keys():\n",
    "    speaker_pred[key]=[]\n",
    "    \n",
    "X, y = feature_V1.drop(['Speaker','label','vowel','word','pre_seg','fol_seg'],axis=1), feature_V1['label']\n",
    "\n",
    "cv = StratifiedKFold(n_splits=10)\n",
    "classifier = tree.DecisionTreeClassifier(max_depth=8)#, min_samples_leaf=300)\n",
    "#classifier =svm.SVC(kernel = 'rbf', probability=True)\n",
    "#classifier = GradientBoostingClassifier(n_estimators=50)\n",
    "mean_tpr = 0.0\n",
    "mean_fpr = np.linspace(0, 1, 100)\n",
    "\n",
    "colors = cycle(['cyan', 'indigo', 'seagreen', 'yellow', 'blue', 'darkorange'])\n",
    "lw = 2\n",
    "\n",
    "i = 0\n",
    "plt.figure(figsize = (10,10))\n",
    "accuracy_all = []\n",
    "for (train, test), color in zip(cv.split(speaker,label_speaker), colors):\n",
    "    train_s = speaker[train]\n",
    "    test_s = speaker[test]\n",
    "    clf = classifier.fit(X.ix[train_s], y.ix[train_s])\n",
    "    probas_,label = clf.predict_proba(X.ix[test_s]),clf.predict(X.ix[test_s])\n",
    "    # Compute ROC curve and area the curve\n",
    "    acc = []\n",
    "    prediction_byspeaker = pd.DataFrame(y.ix[test_s] == label )\n",
    "    prediction_byspeaker['Speaker'] = list(prediction_byspeaker .index.get_level_values('Speaker'))\n",
    "    prediction_byspeaker = prediction_byspeaker.groupby(['Speaker','label'])['label'].count()\n",
    "    for s in prediction_byspeaker.index.get_level_values('Speaker').unique():\n",
    "        p = prediction_byspeaker.ix[s].idxmax()\n",
    "        acc.append(p)\n",
    "        speaker_pred[s].append(p)\n",
    "    accuracy = sum(acc)/len(acc)\n",
    "    accuracy_all.append(accuracy)\n",
    "    fpr, tpr, thresholds = roc_curve(y.ix[test_s], probas_[:, 1])\n",
    "    roc_auc = auc(fpr, tpr)\n",
    "    if roc_auc < 0.5:\n",
    "        fpr, tpr, thresholds = roc_curve(y.ix[test_s], 1-probas_[:, 1])\n",
    "        roc_auc = auc(fpr, tpr)\n",
    "    mean_tpr += interp(mean_fpr, fpr, tpr)\n",
    "    mean_tpr[0] = 0.0\n",
    "    plt.plot(fpr, tpr, lw=lw, color=color,\n",
    "             label='ROC fold %d (area = %0.2f) (Accuracy by Speaker = %0.2f)' % (i, roc_auc, accuracy))\n",
    "\n",
    "    i += 1\n",
    "    \n",
    "plt.plot([0, 1], [0, 1], linestyle='--', lw=lw, color='k',\n",
    "         label='Luck')\n",
    "\n",
    "mean_tpr /= cv.get_n_splits(speaker,label_speaker)\n",
    "mean_tpr[-1] = 1.0\n",
    "mean_auc = auc(mean_fpr, mean_tpr)\n",
    "#speaker_pred = pd.DataFrame(speaker_pred).T\n",
    "#speaker_pred.sum()[0]/len(speaker_pred)\n",
    "plt.plot(mean_fpr, mean_tpr, color='g', linestyle='--',\n",
    "         label='Mean ROC (area = %0.2f) (Accuracy by Speaker = %0.2f) ' % (mean_auc,sum(accuracy_all)/len(accuracy_all)), lw=lw)\n",
    "\n",
    "plt.xlim([-0.05, 1.05])\n",
    "plt.ylim([-0.05, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Classification Tree built on word \"AMENDMENT\"')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Base model Logistic Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Base model SVM with RBF kernel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [Root]",
   "language": "python",
   "name": "Python [Root]"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}