notebooks/NegativeSimilarityExperiments.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.metrics.pairwise import cosine_similarity, manhattan_distances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "n = pd.read_csv('../data/training/negative/geneswap_negatives.pp_blast.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9903"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(n['contig_id'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "neg = pd.read_pickle('../data/training/negative/geneswap_negatives.domain_set.pkl')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "pos = pd.read_pickle('../data/training/positive/mibig_bgcs_all.domain_set.pkl')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "cols = np.intersect1d(neg.columns, pos.columns)\n",
    "neg = neg[cols]\n",
    "pos = pos[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4868"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(neg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>pfam_id</th>\n",
       "      <th>PF00004</th>\n",
       "      <th>PF00005</th>\n",
       "      <th>PF00006</th>\n",
       "      <th>PF00009</th>\n",
       "      <th>PF00011</th>\n",
       "      <th>PF00012</th>\n",
       "      <th>PF00013</th>\n",
       "      <th>PF00015</th>\n",
       "      <th>PF00022</th>\n",
       "      <th>PF00023</th>\n",
       "      <th>...</th>\n",
       "      <th>PF17263</th>\n",
       "      <th>PF17286</th>\n",
       "      <th>PF17319</th>\n",
       "      <th>PF17334</th>\n",
       "      <th>PF17389</th>\n",
       "      <th>PF17390</th>\n",
       "      <th>PF17391</th>\n",
       "      <th>PF17392</th>\n",
       "      <th>PF17414</th>\n",
       "      <th>PF17425</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>contig_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>BGC0000001.1</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000002.1</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000003.1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000004.1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000005.1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 2888 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "pfam_id       PF00004  PF00005  PF00006  PF00009  PF00011  PF00012  PF00013  \\\n",
       "contig_id                                                                     \n",
       "BGC0000001.1        0        1        0        0        0        0        0   \n",
       "BGC0000002.1        0        1        0        0        0        0        0   \n",
       "BGC0000003.1        0        0        0        0        0        0        0   \n",
       "BGC0000004.1        0        0        0        0        0        0        0   \n",
       "BGC0000005.1        0        0        0        0        0        0        0   \n",
       "\n",
       "pfam_id       PF00015  PF00022  PF00023   ...     PF17263  PF17286  PF17319  \\\n",
       "contig_id                                 ...                                 \n",
       "BGC0000001.1        0        0        0   ...           0        0        0   \n",
       "BGC0000002.1        0        0        0   ...           0        0        0   \n",
       "BGC0000003.1        0        0        0   ...           0        0        0   \n",
       "BGC0000004.1        0        0        0   ...           0        0        0   \n",
       "BGC0000005.1        0        0        0   ...           0        0        0   \n",
       "\n",
       "pfam_id       PF17334  PF17389  PF17390  PF17391  PF17392  PF17414  PF17425  \n",
       "contig_id                                                                    \n",
       "BGC0000001.1        0        0        0        0        0        0        0  \n",
       "BGC0000002.1        0        0        0        0        0        0        0  \n",
       "BGC0000003.1        0        0        0        0        0        0        0  \n",
       "BGC0000004.1        0        0        0        0        0        0        0  \n",
       "BGC0000005.1        0        0        0        0        0        0        0  \n",
       "\n",
       "[5 rows x 2888 columns]"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>contig_id</th>\n",
       "      <th>BGC0000001.1</th>\n",
       "      <th>BGC0000002.1</th>\n",
       "      <th>BGC0000003.1</th>\n",
       "      <th>BGC0000004.1</th>\n",
       "      <th>BGC0000005.1</th>\n",
       "      <th>BGC0000006.1</th>\n",
       "      <th>BGC0000007.1</th>\n",
       "      <th>BGC0000008.1</th>\n",
       "      <th>BGC0000009.1</th>\n",
       "      <th>BGC0000010.1</th>\n",
       "      <th>...</th>\n",
       "      <th>BGC0001423.1</th>\n",
       "      <th>BGC0001424.1</th>\n",
       "      <th>BGC0001425.1</th>\n",
       "      <th>BGC0001426.1</th>\n",
       "      <th>BGC0001427.1</th>\n",
       "      <th>BGC0001428.1</th>\n",
       "      <th>BGC0001429.1</th>\n",
       "      <th>BGC0001430.1</th>\n",
       "      <th>BGC0001431.1</th>\n",
       "      <th>BGC0001432.1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>contig_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>NEG_FAKE_CLUSTER|BGC0000001.1|AE016827</th>\n",
       "      <td>0.087922</td>\n",
       "      <td>0.033672</td>\n",
       "      <td>0.029161</td>\n",
       "      <td>0.045175</td>\n",
       "      <td>0.127775</td>\n",
       "      <td>0.066130</td>\n",
       "      <td>0.086146</td>\n",
       "      <td>0.066932</td>\n",
       "      <td>0.065357</td>\n",
       "      <td>0.046971</td>\n",
       "      <td>...</td>\n",
       "      <td>0.112938</td>\n",
       "      <td>0.150756</td>\n",
       "      <td>0.112938</td>\n",
       "      <td>0.112938</td>\n",
       "      <td>0.114377</td>\n",
       "      <td>0.152499</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.095238</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.089363</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NEG_FAKE_CLUSTER|BGC0000001.1|CP002073</th>\n",
       "      <td>0.071819</td>\n",
       "      <td>0.068761</td>\n",
       "      <td>0.089324</td>\n",
       "      <td>0.023063</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.022507</td>\n",
       "      <td>0.021990</td>\n",
       "      <td>0.022780</td>\n",
       "      <td>0.022244</td>\n",
       "      <td>0.023980</td>\n",
       "      <td>...</td>\n",
       "      <td>0.184506</td>\n",
       "      <td>0.175920</td>\n",
       "      <td>0.207570</td>\n",
       "      <td>0.207570</td>\n",
       "      <td>0.186857</td>\n",
       "      <td>0.177954</td>\n",
       "      <td>0.072932</td>\n",
       "      <td>0.072932</td>\n",
       "      <td>0.048622</td>\n",
       "      <td>0.091245</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NEG_FAKE_CLUSTER|BGC0000001.1|CP003543</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.020833</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NEG_FAKE_CLUSTER|BGC0000001.1|FQ312003</th>\n",
       "      <td>0.152676</td>\n",
       "      <td>0.146176</td>\n",
       "      <td>0.151911</td>\n",
       "      <td>0.019612</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.019139</td>\n",
       "      <td>0.018699</td>\n",
       "      <td>0.019371</td>\n",
       "      <td>0.037830</td>\n",
       "      <td>0.020391</td>\n",
       "      <td>...</td>\n",
       "      <td>0.117670</td>\n",
       "      <td>0.112194</td>\n",
       "      <td>0.117670</td>\n",
       "      <td>0.117670</td>\n",
       "      <td>0.099307</td>\n",
       "      <td>0.094576</td>\n",
       "      <td>0.124035</td>\n",
       "      <td>0.124035</td>\n",
       "      <td>0.103362</td>\n",
       "      <td>0.129315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NEG_FAKE_CLUSTER|BGC0000002.1|AP012205</th>\n",
       "      <td>0.147122</td>\n",
       "      <td>0.197203</td>\n",
       "      <td>0.121988</td>\n",
       "      <td>0.151186</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.147542</td>\n",
       "      <td>0.144150</td>\n",
       "      <td>0.149331</td>\n",
       "      <td>0.145817</td>\n",
       "      <td>0.157195</td>\n",
       "      <td>...</td>\n",
       "      <td>0.188982</td>\n",
       "      <td>0.180187</td>\n",
       "      <td>0.188982</td>\n",
       "      <td>0.188982</td>\n",
       "      <td>0.191390</td>\n",
       "      <td>0.182271</td>\n",
       "      <td>0.159364</td>\n",
       "      <td>0.119523</td>\n",
       "      <td>0.159364</td>\n",
       "      <td>0.149533</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 1551 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "contig_id                               BGC0000001.1  BGC0000002.1  \\\n",
       "contig_id                                                            \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|AE016827      0.087922      0.033672   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP002073      0.071819      0.068761   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP003543      0.000000      0.020833   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|FQ312003      0.152676      0.146176   \n",
       "NEG_FAKE_CLUSTER|BGC0000002.1|AP012205      0.147122      0.197203   \n",
       "\n",
       "contig_id                               BGC0000003.1  BGC0000004.1  \\\n",
       "contig_id                                                            \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|AE016827      0.029161      0.045175   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP002073      0.089324      0.023063   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP003543      0.000000      0.000000   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|FQ312003      0.151911      0.019612   \n",
       "NEG_FAKE_CLUSTER|BGC0000002.1|AP012205      0.121988      0.151186   \n",
       "\n",
       "contig_id                               BGC0000005.1  BGC0000006.1  \\\n",
       "contig_id                                                            \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|AE016827      0.127775      0.066130   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP002073      0.000000      0.022507   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP003543      0.000000      0.000000   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|FQ312003      0.000000      0.019139   \n",
       "NEG_FAKE_CLUSTER|BGC0000002.1|AP012205      0.000000      0.147542   \n",
       "\n",
       "contig_id                               BGC0000007.1  BGC0000008.1  \\\n",
       "contig_id                                                            \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|AE016827      0.086146      0.066932   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP002073      0.021990      0.022780   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP003543      0.000000      0.000000   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|FQ312003      0.018699      0.019371   \n",
       "NEG_FAKE_CLUSTER|BGC0000002.1|AP012205      0.144150      0.149331   \n",
       "\n",
       "contig_id                               BGC0000009.1  BGC0000010.1  \\\n",
       "contig_id                                                            \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|AE016827      0.065357      0.046971   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP002073      0.022244      0.023980   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP003543      0.000000      0.000000   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|FQ312003      0.037830      0.020391   \n",
       "NEG_FAKE_CLUSTER|BGC0000002.1|AP012205      0.145817      0.157195   \n",
       "\n",
       "contig_id                                   ...       BGC0001423.1  \\\n",
       "contig_id                                   ...                      \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|AE016827      ...           0.112938   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP002073      ...           0.184506   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP003543      ...           0.000000   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|FQ312003      ...           0.117670   \n",
       "NEG_FAKE_CLUSTER|BGC0000002.1|AP012205      ...           0.188982   \n",
       "\n",
       "contig_id                               BGC0001424.1  BGC0001425.1  \\\n",
       "contig_id                                                            \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|AE016827      0.150756      0.112938   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP002073      0.175920      0.207570   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP003543      0.000000      0.000000   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|FQ312003      0.112194      0.117670   \n",
       "NEG_FAKE_CLUSTER|BGC0000002.1|AP012205      0.180187      0.188982   \n",
       "\n",
       "contig_id                               BGC0001426.1  BGC0001427.1  \\\n",
       "contig_id                                                            \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|AE016827      0.112938      0.114377   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP002073      0.207570      0.186857   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP003543      0.000000      0.000000   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|FQ312003      0.117670      0.099307   \n",
       "NEG_FAKE_CLUSTER|BGC0000002.1|AP012205      0.188982      0.191390   \n",
       "\n",
       "contig_id                               BGC0001428.1  BGC0001429.1  \\\n",
       "contig_id                                                            \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|AE016827      0.152499      0.166667   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP002073      0.177954      0.072932   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP003543      0.000000      0.000000   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|FQ312003      0.094576      0.124035   \n",
       "NEG_FAKE_CLUSTER|BGC0000002.1|AP012205      0.182271      0.159364   \n",
       "\n",
       "contig_id                               BGC0001430.1  BGC0001431.1  \\\n",
       "contig_id                                                            \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|AE016827      0.095238      0.166667   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP002073      0.072932      0.048622   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP003543      0.000000      0.000000   \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|FQ312003      0.124035      0.103362   \n",
       "NEG_FAKE_CLUSTER|BGC0000002.1|AP012205      0.119523      0.159364   \n",
       "\n",
       "contig_id                               BGC0001432.1  \n",
       "contig_id                                             \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|AE016827      0.089363  \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP002073      0.091245  \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|CP003543      0.000000  \n",
       "NEG_FAKE_CLUSTER|BGC0000001.1|FQ312003      0.129315  \n",
       "NEG_FAKE_CLUSTER|BGC0000002.1|AP012205      0.149533  \n",
       "\n",
       "[5 rows x 1551 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "distances = pd.DataFrame(cosine_similarity(neg.values, pos.values), index=neg.index, columns=pos.index)\n",
    "distances.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "contig_id\n",
       "NEG_FAKE_CLUSTER|BGC0001417.1|AP010918    0.848528\n",
       "NEG_FAKE_CLUSTER|BGC0000500.1|CP009757    0.816497\n",
       "NEG_FAKE_CLUSTER|BGC0000984.1|CP002992    0.784465\n",
       "NEG_FAKE_CLUSTER|BGC0000077.1|CP007409    0.774597\n",
       "NEG_FAKE_CLUSTER|BGC0000134.1|FO818638    0.755929\n",
       "dtype: float64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "most_similar = distances.max(axis=1).sort_values(ascending=False)\n",
    "most_similar.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "contig_id\n",
       "NEG_FAKE_CLUSTER|BGC0001417.1|AP010918    BGC0000353.1\n",
       "NEG_FAKE_CLUSTER|BGC0000500.1|CP009757    BGC0001247.4\n",
       "NEG_FAKE_CLUSTER|BGC0000984.1|CP002992    BGC0000155.3\n",
       "NEG_FAKE_CLUSTER|BGC0000077.1|CP007409    BGC0001023.1\n",
       "NEG_FAKE_CLUSTER|BGC0000134.1|FO818638    BGC0001135.1\n",
       "dtype: object"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "similar_ids = distances.idxmax(axis=1).reindex(index=most_similar.index)\n",
    "similar_ids.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10aad9048>"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYEAAAD8CAYAAACRkhiPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAFH1JREFUeJzt3X+MZeV93/H3J1BTYvyDGHtEdnEXS0sUYFsSRoQqcjqI1OZHZewoSUEkgI2ytgv9oayqrJs/bBkh0STUqlUXZx0QOD9Y0xCblVnXJdRjmgpiLzZmAZt6gY09gKAGQj3GpV3y7R9zNr5d5sede+/MnZnn/ZKu5tznPuecZ757rz5znnPu2VQVkqQ2/di4ByBJGh9DQJIaZghIUsMMAUlqmCEgSQ0zBCSpYYaAJDXMEJCkhhkCktSwo8c9gKWccMIJtWXLloHW/cEPfsBrX/va0Q5oA7E+C7M2i7M+ixt3fe6///7vVdWb++m75kNgy5Yt7Nu3b6B1p6enmZqaGu2ANhDrszBrszjrs7hx1yfJX/Xb1+kgSWqYISBJDTMEJKlhhoAkNcwQkKSGGQKS1DBDQJIaZghIUsOWDIEkNyV5NslDPW2fSfJA9ziY5IGufUuSH/a89smedc5Msj/JgSQfT5KV+ZUkSf3q5xvDNwP/Afj04Yaq+qeHl5NcD7zY0/+xqjpjnu3cAGwH7gP2AucBX1j+kLUWbdl559j2ffC6C8e2b2m9W/JIoKruAZ6f77Xur/lfBW5dbBtJTgReX1X3VlUxFyjvXv5wJUmjNOw5gbcDz1TVt3vaTk7y9SRfTvL2rm0TMNPTZ6ZrkySN0bA3kLuE//8o4GngrVX1XJIzgc8lOQ2Yb/6/Ftpoku3MTR0xMTHB9PT0QIObnZ0deN0WjLI+O7YdGsl2BrES/8a+dxZnfRa3nuozcAgkORr4JeDMw21V9TLwcrd8f5LHgFOY+8t/c8/qm4GnFtp2Ve0CdgFMTk7WoHfjG/ed/Na6UdbninGeE7h0auTb9L2zOOuzuPVUn2Gmg34R+FZV/e00T5I3JzmqW34bsBV4vKqeBr6f5OzuPMJlwB1D7FuSNAL9XCJ6K3Av8FNJZpJc2b10Ma8+IfwLwINJvgH8KfCBqjp8UvmDwB8AB4DH8MogSRq7JaeDquqSBdqvmKftduD2BfrvA05f5vgkSSvIbwxLUsMMAUlqmCEgSQ0zBCSpYYaAJDXMEJCkhhkCktQwQ0CSGmYISFLDDAFJapghIEkNMwQkqWGGgCQ1zBCQpIYZApLUMENAkhpmCEhSwwwBSWqYISBJDTMEJKlhS4ZAkpuSPJvkoZ62jyR5MskD3eOCntc+lORAkkeTvLOn/byu7UCSnaP/VSRJy9XPkcDNwHnztH+sqs7oHnsBkpwKXAyc1q3zH5McleQo4BPA+cCpwCVdX0nSGB29VIequifJlj63dxGwu6peBp5IcgA4q3vtQFU9DpBkd9f3kWWPWJI0MsOcE7g6yYPddNHxXdsm4Ls9fWa6toXaJUljtOSRwAJuAK4Bqvt5PfA+IPP0LeYPm1po40m2A9sBJiYmmJ6eHmiQs7OzA6/bglHWZ8e2QyPZziBW4t/Y987irM/i1lN9BgqBqnrm8HKSTwGf757OACf1dN0MPNUtL9Q+3/Z3AbsAJicna2pqapBhMj09zaDrtmCU9bli550j2c4gDl46NfJt+t5ZnPVZ3Hqqz0DTQUlO7Hn6HuDwlUN7gIuTHJPkZGAr8BXgq8DWJCcneQ1zJ4/3DD5sSdIoLHkkkORWYAo4IckM8GFgKskZzE3pHATeD1BVDye5jbkTvoeAq6rqlW47VwNfBI4Cbqqqh0f+20iSlqWfq4Mumaf5xkX6XwtcO0/7XmDvskYnSVpRfmNYkhpmCEhSwwwBSWqYISBJDTMEJKlhhoAkNcwQkKSGGQKS1DBDQJIaZghIUsMMAUlqmCEgSQ0zBCSpYYaAJDXMEJCkhhkCktQwQ0CSGmYISFLDDAFJapghIEkNWzIEktyU5NkkD/W0/W6SbyV5MMlnk7yxa9+S5IdJHugen+xZ58wk+5McSPLxJFmZX0mS1K9+jgRuBs47ou0u4PSq+vvA/wA+1PPaY1V1Rvf4QE/7DcB2YGv3OHKbkqRVtmQIVNU9wPNHtP2XqjrUPb0P2LzYNpKcCLy+qu6tqgI+Dbx7sCFLkkZlFOcE3gd8oef5yUm+nuTLSd7etW0CZnr6zHRtkqQxOnqYlZP8NnAI+OOu6WngrVX1XJIzgc8lOQ2Yb/6/FtnuduamjpiYmGB6enqg8c3Ozg68bgtGWZ8d2w4t3WmFrMS/se+dxVmfxa2n+gwcAkkuB/4JcG43xUNVvQy83C3fn+Qx4BTm/vLvnTLaDDy10LarahewC2BycrKmpqYGGuP09DSDrtuCUdbnip13jmQ7gzh46dTIt+l7Z3HWZ3HrqT4DTQclOQ/4LeBdVfVST/ubkxzVLb+NuRPAj1fV08D3k5zdXRV0GXDH0KOXJA1lySOBJLcCU8AJSWaADzN3NdAxwF3dlZ73dVcC/QLw0SSHgFeAD1TV4ZPKH2TuSqNjmTuH0HseQZI0BkuGQFVdMk/zjQv0vR24fYHX9gGnL2t0kqQV5TeGJalhhoAkNcwQkKSGGQKS1DBDQJIaZghIUsMMAUlqmCEgSQ0zBCSpYYaAJDVsqFtJa+3Zsoy7ee7Ydmisd/+UNH4eCUhSwwwBSWqYISBJDTMEJKlhhoAkNcyrg7TuLeeKqH71c+XUwesuHPl+pdXmkYAkNcwQkKSGGQKS1LC+QiDJTUmeTfJQT9tPJLkrybe7n8d37Uny8SQHkjyY5Gd71rm86//tJJeP/teRJC1Hv0cCNwPnHdG2E7i7qrYCd3fPAc4HtnaP7cANMBcawIeBnwPOAj58ODgkSePRVwhU1T3A80c0XwTc0i3fAry7p/3TNec+4I1JTgTeCdxVVc9X1QvAXbw6WCRJq2iYcwITVfU0QPfzLV37JuC7Pf1muraF2iVJY7IS3xPIPG21SPurN5BsZ24qiYmJCaanpwcayOzs7MDrrlc7th3qu+/Escvr35J+atPae6tXi5+t5VhP9RkmBJ5JcmJVPd1N9zzbtc8AJ/X02ww81bVPHdE+Pd+Gq2oXsAtgcnKypqam5uu2pOnpaQZdd71azq2hd2w7xPX7/b7gfPqpzcFLp1ZnMGtQi5+t5VhP9RlmOmgPcPgKn8uBO3raL+uuEjobeLGbLvoi8I4kx3cnhN/RtUmSxqSvPwOT3MrcX/EnJJlh7iqf64DbklwJfAf4la77XuAC4ADwEvBegKp6Psk1wFe7fh+tqiNPNkuSVlFfIVBVlyzw0rnz9C3gqgW2cxNwU9+jkyStKL8xLEkNMwQkqWGGgCQ1zBCQpIYZApLUMENAkhpmCEhSwwwBSWqYISBJDTMEJKlhhoAkNcwQkKSGGQKS1DBDQJIaZghIUsMMAUlqmCEgSQ0zBCSpYYaAJDXMEJCkhhkCktSwgUMgyU8leaDn8b+S/KskH0nyZE/7BT3rfCjJgSSPJnnnaH4FSdKgjh50xap6FDgDIMlRwJPAZ4H3Ah+rqt/r7Z/kVOBi4DTgJ4E/T3JKVb0y6BgkScMZ1XTQucBjVfVXi/S5CNhdVS9X1RPAAeCsEe1fkjSAgY8EjnAxcGvP86uTXAbsA3ZU1QvAJuC+nj4zXdurJNkObAeYmJhgenp6oEHNzs4OvO56tWPbob77Thy7vP4t6ac2rb23erX42VqO9VSfVNVwG0heAzwFnFZVzySZAL4HFHANcGJVvS/JJ4B7q+qPuvVuBPZW1e2LbX9ycrL27ds30Nimp6eZmpoaaN31asvOO/vuu2PbIa7fP6q/AzaWfmpz8LoLV2k0a0+Ln63lGHd9ktxfVZP99B3FdND5wNeq6hmAqnqmql6pqr8BPsWPpnxmgJN61tvMXHhIksZkFCFwCT1TQUlO7HntPcBD3fIe4OIkxyQ5GdgKfGUE+5ckDWiouYAkPw78Y+D9Pc2/k+QM5qaDDh5+raoeTnIb8AhwCLjKK4MkabyGCoGqegl40xFtv75I/2uBa4fZpyRpdPzGsCQ1zBCQpIYZApLUMENAkhpmCEhSwwwBSWqYISBJDTMEJKlhhoAkNcwQkKSGGQKS1DBDQJIaZghIUsMMAUlqmCEgSQ0zBCSpYYaAJDXMEJCkhhkCktQwQ0CSGjZ0CCQ5mGR/kgeS7OvafiLJXUm+3f08vmtPko8nOZDkwSQ/O+z+JUmDG9WRwDlVdUZVTXbPdwJ3V9VW4O7uOcD5wNbusR24YUT7lyQNYKWmgy4CbumWbwHe3dP+6ZpzH/DGJCeu0BgkSUtIVQ23geQJ4AWggN+vql1J/rqq3tjT54WqOj7J54Hrquovuva7gd+qqn1HbHM7c0cKTExMnLl79+6BxjY7O8txxx030Lrr1f4nX+y778Sx8MwPV3Aw61g/tdm26Q2rM5g1qMXP1nKMuz7nnHPO/T0zM4s6egT7+/mqeirJW4C7knxrkb6Zp+1VKVRVu4BdAJOTkzU1NTXQwKanpxl03fXqip139t13x7ZDXL9/FG+Bjaef2hy8dGp1BrMGtfjZWo71VJ+hp4Oq6qnu57PAZ4GzgGcOT/N0P5/tus8AJ/Wsvhl4atgxSJIGM1QIJHltktcdXgbeATwE7AEu77pdDtzRLe8BLuuuEjobeLGqnh5mDJKkwQ07FzABfDbJ4W39SVX95yRfBW5LciXwHeBXuv57gQuAA8BLwHuH3L8kaQhDhUBVPQ78g3nanwPOnae9gKuG2ackaXT8xrAkNcxLQ6QBbVnGlVijdPC6C8eyX21MHglIUsMMAUlqmCEgSQ0zBCSpYYaAJDXMEJCkhhkCktQwQ0CSGmYISFLDDAFJapghIEkNMwQkqWGGgCQ1zBCQpIYZApLUMENAkhpmCEhSwwwBSWrYwCGQ5KQkX0ryzSQPJ/mXXftHkjyZ5IHucUHPOh9KciDJo0neOYpfQJI0uGH+j+FDwI6q+lqS1wH3J7mre+1jVfV7vZ2TnApcDJwG/CTw50lOqapXhhiDJGkIAx8JVNXTVfW1bvn7wDeBTYuschGwu6perqongAPAWYPuX5I0vJGcE0iyBfgZ4C+7pquTPJjkpiTHd22bgO/2rDbD4qEhSVphqarhNpAcB3wZuLaq/izJBPA9oIBrgBOr6n1JPgHcW1V/1K13I7C3qm6fZ5vbge0AExMTZ+7evXugsc3OznLccccNtO56tf/JF/vuO3EsPPPDFRzMOraWa7Nt0xvGPYQmP1vLMe76nHPOOfdX1WQ/fYc5J0CSvwPcDvxxVf0ZQFU90/P6p4DPd09ngJN6Vt8MPDXfdqtqF7ALYHJysqampgYa3/T0NIOuu15dsfPOvvvu2HaI6/cP9RbYsNZybQ5eOjXuITT52VqO9VSfYa4OCnAj8M2q+nc97Sf2dHsP8FC3vAe4OMkxSU4GtgJfGXT/kqThDfOnzs8Dvw7sT/JA1/ZvgEuSnMHcdNBB4P0AVfVwktuAR5i7sugqrwySpPEaOASq6i+AzPPS3kXWuRa4dtB9SpJGy28MS1LDDAFJapghIEkNW5vXwK1zW5Zxmaa0XON8fx287sKx7VsrwyMBSWqYISBJDTMEJKlhhoAkNcwQkKSGGQKS1DBDQJIaZghIUsMMAUlqmCEgSQ0zBCSpYYaAJDXMG8hJ6tvhm9ft2HZoWf+f9bC8cd3K8UhAkhpmCEhSwwwBSWrYqp8TSHIe8O+Bo4A/qKrrVnsMktYX/yOdlbOqRwJJjgI+AZwPnApckuTU1RyDJOlHVns66CzgQFU9XlX/B9gNXLTKY5AkdVZ7OmgT8N2e5zPAz63UzvY/+eKqXsYmaeMZZCpqFJfQrtY01GqHQOZpq1d1SrYD27uns0keHXB/JwDfG3DdDe9fWJ8FWZvFWZ/FjaI++bdDDeHv9dtxtUNgBjip5/lm4KkjO1XVLmDXsDtLsq+qJofdzkZlfRZmbRZnfRa3nuqz2ucEvgpsTXJyktcAFwN7VnkMkqTOqh4JVNWhJFcDX2TuEtGbqurh1RyDJOlHVv17AlW1F9i7Srsbekppg7M+C7M2i7M+i1s39UnVq87LSpIa4W0jJKlhGyIEkpyX5NEkB5LsnOf1Y5J8pnv9L5NsWf1RjkcftfnNJI8keTDJ3Un6vrRsI1iqPj39fjlJJVkXV3yMSj/1SfKr3Xvo4SR/stpjHJc+PltvTfKlJF/vPl8XjGOcS6qqdf1g7gTzY8DbgNcA3wBOPaLPPwM+2S1fDHxm3ONeQ7U5B/jxbvmDrdSm3/p0/V4H3APcB0yOe9xrqT7AVuDrwPHd87eMe9xrqDa7gA92y6cCB8c97vkeG+FIoJ9bUVwE3NIt/ylwbpL5vri20SxZm6r6UlW91D29j7nvbrSi39uYXAP8DvC/V3Nwa0A/9fkN4BNV9QJAVT27ymMcl35qU8Dru+U3MM93otaCjRAC892KYtNCfarqEPAi8KZVGd149VObXlcCX1jREa0tS9Ynyc8AJ1XV51dzYGtEP++fU4BTkvz3JPd1dwluQT+1+Qjwa0lmmLsi8p+vztCWZyP895L93Iqir9tVbEB9/95Jfg2YBP7Rio5obVm0Pkl+DPgYcMVqDWiN6ef9czRzU0JTzB1F/rckp1fVX6/w2Matn9pcAtxcVdcn+YfAH3a1+ZuVH17/NsKRQD+3ovjbPkmOZu7Q7PlVGd149XWbjiS/CPw28K6qenmVxrYWLFWf1wGnA9NJDgJnA3saOjnc72frjqr6v1X1BPAoc6Gw0fVTmyuB2wCq6l7g7zJ3T6E1ZSOEQD+3otgDXN4t/zLwX6s7W7PBLVmbbrrj95kLgFbmcw9btD5V9WJVnVBVW6pqC3PnTN5VVfvGM9xV189n63PMXVxAkhOYmx56fFVHOR791OY7wLkASX6auRD4n6s6yj6s+xDo5vgP34rim8BtVfVwko8meVfX7UbgTUkOAL8JLHgp4EbSZ21+FzgO+E9JHkjSzL2c+qxPs/qszxeB55I8AnwJ+NdV9dx4Rrx6+qzNDuA3knwDuBW4Yi3+8ek3hiWpYev+SECSNDhDQJIaZghIUsMMAUlqmCEgSQ0zBCSpYYaAJDXMEJCkhv0/v8FtvjDIxAMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10aac81d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "most_similar.hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>contig_id</th>\n",
       "      <th>BGC0000001.1</th>\n",
       "      <th>BGC0000002.1</th>\n",
       "      <th>BGC0000003.1</th>\n",
       "      <th>BGC0000004.1</th>\n",
       "      <th>BGC0000005.1</th>\n",
       "      <th>BGC0000006.1</th>\n",
       "      <th>BGC0000007.1</th>\n",
       "      <th>BGC0000008.1</th>\n",
       "      <th>BGC0000009.1</th>\n",
       "      <th>BGC0000010.1</th>\n",
       "      <th>...</th>\n",
       "      <th>BGC0001423.1</th>\n",
       "      <th>BGC0001424.1</th>\n",
       "      <th>BGC0001425.1</th>\n",
       "      <th>BGC0001426.1</th>\n",
       "      <th>BGC0001427.1</th>\n",
       "      <th>BGC0001428.1</th>\n",
       "      <th>BGC0001429.1</th>\n",
       "      <th>BGC0001430.1</th>\n",
       "      <th>BGC0001431.1</th>\n",
       "      <th>BGC0001432.1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>contig_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>BGC0000001.1</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.478714</td>\n",
       "      <td>0.326637</td>\n",
       "      <td>0.311400</td>\n",
       "      <td>0.165145</td>\n",
       "      <td>0.322888</td>\n",
       "      <td>0.334021</td>\n",
       "      <td>0.326802</td>\n",
       "      <td>0.337883</td>\n",
       "      <td>0.242833</td>\n",
       "      <td>...</td>\n",
       "      <td>0.330862</td>\n",
       "      <td>0.371135</td>\n",
       "      <td>0.330862</td>\n",
       "      <td>0.330862</td>\n",
       "      <td>0.335077</td>\n",
       "      <td>0.356654</td>\n",
       "      <td>0.451335</td>\n",
       "      <td>0.369274</td>\n",
       "      <td>0.451335</td>\n",
       "      <td>0.436328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000002.1</th>\n",
       "      <td>0.478714</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.336788</td>\n",
       "      <td>0.316776</td>\n",
       "      <td>0.105409</td>\n",
       "      <td>0.309142</td>\n",
       "      <td>0.319801</td>\n",
       "      <td>0.312889</td>\n",
       "      <td>0.323498</td>\n",
       "      <td>0.251870</td>\n",
       "      <td>...</td>\n",
       "      <td>0.316776</td>\n",
       "      <td>0.319801</td>\n",
       "      <td>0.316776</td>\n",
       "      <td>0.316776</td>\n",
       "      <td>0.320812</td>\n",
       "      <td>0.305526</td>\n",
       "      <td>0.392837</td>\n",
       "      <td>0.314270</td>\n",
       "      <td>0.373195</td>\n",
       "      <td>0.393179</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000003.1</th>\n",
       "      <td>0.326637</td>\n",
       "      <td>0.336788</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.258199</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.251976</td>\n",
       "      <td>0.246183</td>\n",
       "      <td>0.255031</td>\n",
       "      <td>0.280158</td>\n",
       "      <td>0.167789</td>\n",
       "      <td>...</td>\n",
       "      <td>0.419573</td>\n",
       "      <td>0.400047</td>\n",
       "      <td>0.387298</td>\n",
       "      <td>0.387298</td>\n",
       "      <td>0.392232</td>\n",
       "      <td>0.373544</td>\n",
       "      <td>0.408248</td>\n",
       "      <td>0.408248</td>\n",
       "      <td>0.374228</td>\n",
       "      <td>0.510754</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000004.1</th>\n",
       "      <td>0.311400</td>\n",
       "      <td>0.316776</td>\n",
       "      <td>0.258199</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.282843</td>\n",
       "      <td>0.975900</td>\n",
       "      <td>0.953463</td>\n",
       "      <td>0.987730</td>\n",
       "      <td>0.916261</td>\n",
       "      <td>0.935775</td>\n",
       "      <td>...</td>\n",
       "      <td>0.275000</td>\n",
       "      <td>0.286039</td>\n",
       "      <td>0.275000</td>\n",
       "      <td>0.275000</td>\n",
       "      <td>0.278503</td>\n",
       "      <td>0.289346</td>\n",
       "      <td>0.342580</td>\n",
       "      <td>0.316228</td>\n",
       "      <td>0.342580</td>\n",
       "      <td>0.395628</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000005.1</th>\n",
       "      <td>0.165145</td>\n",
       "      <td>0.105409</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.282843</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.276026</td>\n",
       "      <td>0.269680</td>\n",
       "      <td>0.279372</td>\n",
       "      <td>0.204598</td>\n",
       "      <td>0.294086</td>\n",
       "      <td>...</td>\n",
       "      <td>0.070711</td>\n",
       "      <td>0.134840</td>\n",
       "      <td>0.070711</td>\n",
       "      <td>0.070711</td>\n",
       "      <td>0.071611</td>\n",
       "      <td>0.136399</td>\n",
       "      <td>0.223607</td>\n",
       "      <td>0.149071</td>\n",
       "      <td>0.223607</td>\n",
       "      <td>0.186501</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 1551 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "contig_id     BGC0000001.1  BGC0000002.1  BGC0000003.1  BGC0000004.1  \\\n",
       "contig_id                                                              \n",
       "BGC0000001.1      1.000000      0.478714      0.326637      0.311400   \n",
       "BGC0000002.1      0.478714      1.000000      0.336788      0.316776   \n",
       "BGC0000003.1      0.326637      0.336788      1.000000      0.258199   \n",
       "BGC0000004.1      0.311400      0.316776      0.258199      1.000000   \n",
       "BGC0000005.1      0.165145      0.105409      0.000000      0.282843   \n",
       "\n",
       "contig_id     BGC0000005.1  BGC0000006.1  BGC0000007.1  BGC0000008.1  \\\n",
       "contig_id                                                              \n",
       "BGC0000001.1      0.165145      0.322888      0.334021      0.326802   \n",
       "BGC0000002.1      0.105409      0.309142      0.319801      0.312889   \n",
       "BGC0000003.1      0.000000      0.251976      0.246183      0.255031   \n",
       "BGC0000004.1      0.282843      0.975900      0.953463      0.987730   \n",
       "BGC0000005.1      1.000000      0.276026      0.269680      0.279372   \n",
       "\n",
       "contig_id     BGC0000009.1  BGC0000010.1      ...       BGC0001423.1  \\\n",
       "contig_id                                     ...                      \n",
       "BGC0000001.1      0.337883      0.242833      ...           0.330862   \n",
       "BGC0000002.1      0.323498      0.251870      ...           0.316776   \n",
       "BGC0000003.1      0.280158      0.167789      ...           0.419573   \n",
       "BGC0000004.1      0.916261      0.935775      ...           0.275000   \n",
       "BGC0000005.1      0.204598      0.294086      ...           0.070711   \n",
       "\n",
       "contig_id     BGC0001424.1  BGC0001425.1  BGC0001426.1  BGC0001427.1  \\\n",
       "contig_id                                                              \n",
       "BGC0000001.1      0.371135      0.330862      0.330862      0.335077   \n",
       "BGC0000002.1      0.319801      0.316776      0.316776      0.320812   \n",
       "BGC0000003.1      0.400047      0.387298      0.387298      0.392232   \n",
       "BGC0000004.1      0.286039      0.275000      0.275000      0.278503   \n",
       "BGC0000005.1      0.134840      0.070711      0.070711      0.071611   \n",
       "\n",
       "contig_id     BGC0001428.1  BGC0001429.1  BGC0001430.1  BGC0001431.1  \\\n",
       "contig_id                                                              \n",
       "BGC0000001.1      0.356654      0.451335      0.369274      0.451335   \n",
       "BGC0000002.1      0.305526      0.392837      0.314270      0.373195   \n",
       "BGC0000003.1      0.373544      0.408248      0.408248      0.374228   \n",
       "BGC0000004.1      0.289346      0.342580      0.316228      0.342580   \n",
       "BGC0000005.1      0.136399      0.223607      0.149071      0.223607   \n",
       "\n",
       "contig_id     BGC0001432.1  \n",
       "contig_id                   \n",
       "BGC0000001.1      0.436328  \n",
       "BGC0000002.1      0.393179  \n",
       "BGC0000003.1      0.510754  \n",
       "BGC0000004.1      0.395628  \n",
       "BGC0000005.1      0.186501  \n",
       "\n",
       "[5 rows x 1551 columns]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos_distances = pd.DataFrame(cosine_similarity(pos.values, pos.values), index=pos.index, columns=pos.index)\n",
    "pos_distances.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>contig_id</th>\n",
       "      <th>BGC0000001.1</th>\n",
       "      <th>BGC0000002.1</th>\n",
       "      <th>BGC0000003.1</th>\n",
       "      <th>BGC0000004.1</th>\n",
       "      <th>BGC0000005.1</th>\n",
       "      <th>BGC0000006.1</th>\n",
       "      <th>BGC0000007.1</th>\n",
       "      <th>BGC0000008.1</th>\n",
       "      <th>BGC0000009.1</th>\n",
       "      <th>BGC0000010.1</th>\n",
       "      <th>...</th>\n",
       "      <th>BGC0001423.1</th>\n",
       "      <th>BGC0001424.1</th>\n",
       "      <th>BGC0001425.1</th>\n",
       "      <th>BGC0001426.1</th>\n",
       "      <th>BGC0001427.1</th>\n",
       "      <th>BGC0001428.1</th>\n",
       "      <th>BGC0001429.1</th>\n",
       "      <th>BGC0001430.1</th>\n",
       "      <th>BGC0001431.1</th>\n",
       "      <th>BGC0001432.1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "      <td>1551.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.207061</td>\n",
       "      <td>0.218524</td>\n",
       "      <td>0.177992</td>\n",
       "      <td>0.202471</td>\n",
       "      <td>0.047040</td>\n",
       "      <td>0.204498</td>\n",
       "      <td>0.211378</td>\n",
       "      <td>0.206865</td>\n",
       "      <td>0.209877</td>\n",
       "      <td>0.167488</td>\n",
       "      <td>...</td>\n",
       "      <td>0.212962</td>\n",
       "      <td>0.217404</td>\n",
       "      <td>0.212244</td>\n",
       "      <td>0.212244</td>\n",
       "      <td>0.211493</td>\n",
       "      <td>0.213034</td>\n",
       "      <td>0.245220</td>\n",
       "      <td>0.221906</td>\n",
       "      <td>0.242633</td>\n",
       "      <td>0.262868</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.139103</td>\n",
       "      <td>0.145875</td>\n",
       "      <td>0.169773</td>\n",
       "      <td>0.159445</td>\n",
       "      <td>0.072437</td>\n",
       "      <td>0.159313</td>\n",
       "      <td>0.162371</td>\n",
       "      <td>0.161080</td>\n",
       "      <td>0.159298</td>\n",
       "      <td>0.132869</td>\n",
       "      <td>...</td>\n",
       "      <td>0.166483</td>\n",
       "      <td>0.164887</td>\n",
       "      <td>0.165021</td>\n",
       "      <td>0.165021</td>\n",
       "      <td>0.165130</td>\n",
       "      <td>0.163728</td>\n",
       "      <td>0.180047</td>\n",
       "      <td>0.181935</td>\n",
       "      <td>0.175283</td>\n",
       "      <td>0.221167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.100504</td>\n",
       "      <td>0.100272</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.063246</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.062994</td>\n",
       "      <td>0.066695</td>\n",
       "      <td>0.063758</td>\n",
       "      <td>0.073615</td>\n",
       "      <td>0.045170</td>\n",
       "      <td>...</td>\n",
       "      <td>0.068465</td>\n",
       "      <td>0.077850</td>\n",
       "      <td>0.068114</td>\n",
       "      <td>0.068114</td>\n",
       "      <td>0.068982</td>\n",
       "      <td>0.075731</td>\n",
       "      <td>0.092450</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.102062</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.195646</td>\n",
       "      <td>0.205152</td>\n",
       "      <td>0.160128</td>\n",
       "      <td>0.189737</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.192879</td>\n",
       "      <td>0.207079</td>\n",
       "      <td>0.195217</td>\n",
       "      <td>0.201737</td>\n",
       "      <td>0.169165</td>\n",
       "      <td>...</td>\n",
       "      <td>0.211289</td>\n",
       "      <td>0.217597</td>\n",
       "      <td>0.211289</td>\n",
       "      <td>0.211289</td>\n",
       "      <td>0.211830</td>\n",
       "      <td>0.212373</td>\n",
       "      <td>0.242831</td>\n",
       "      <td>0.223298</td>\n",
       "      <td>0.243108</td>\n",
       "      <td>0.255377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.319645</td>\n",
       "      <td>0.328377</td>\n",
       "      <td>0.306186</td>\n",
       "      <td>0.326464</td>\n",
       "      <td>0.091287</td>\n",
       "      <td>0.328118</td>\n",
       "      <td>0.338502</td>\n",
       "      <td>0.332096</td>\n",
       "      <td>0.332201</td>\n",
       "      <td>0.265742</td>\n",
       "      <td>...</td>\n",
       "      <td>0.322886</td>\n",
       "      <td>0.335216</td>\n",
       "      <td>0.322817</td>\n",
       "      <td>0.322817</td>\n",
       "      <td>0.323575</td>\n",
       "      <td>0.325560</td>\n",
       "      <td>0.384015</td>\n",
       "      <td>0.363292</td>\n",
       "      <td>0.378620</td>\n",
       "      <td>0.424665</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 1551 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "contig_id  BGC0000001.1  BGC0000002.1  BGC0000003.1  BGC0000004.1  \\\n",
       "count       1551.000000   1551.000000   1551.000000   1551.000000   \n",
       "mean           0.207061      0.218524      0.177992      0.202471   \n",
       "std            0.139103      0.145875      0.169773      0.159445   \n",
       "min            0.000000      0.000000      0.000000      0.000000   \n",
       "25%            0.100504      0.100272      0.000000      0.063246   \n",
       "50%            0.195646      0.205152      0.160128      0.189737   \n",
       "75%            0.319645      0.328377      0.306186      0.326464   \n",
       "max            1.000000      1.000000      1.000000      1.000000   \n",
       "\n",
       "contig_id  BGC0000005.1  BGC0000006.1  BGC0000007.1  BGC0000008.1  \\\n",
       "count       1551.000000   1551.000000   1551.000000   1551.000000   \n",
       "mean           0.047040      0.204498      0.211378      0.206865   \n",
       "std            0.072437      0.159313      0.162371      0.161080   \n",
       "min            0.000000      0.000000      0.000000      0.000000   \n",
       "25%            0.000000      0.062994      0.066695      0.063758   \n",
       "50%            0.000000      0.192879      0.207079      0.195217   \n",
       "75%            0.091287      0.328118      0.338502      0.332096   \n",
       "max            1.000000      1.000000      1.000000      1.000000   \n",
       "\n",
       "contig_id  BGC0000009.1  BGC0000010.1      ...       BGC0001423.1  \\\n",
       "count       1551.000000   1551.000000      ...        1551.000000   \n",
       "mean           0.209877      0.167488      ...           0.212962   \n",
       "std            0.159298      0.132869      ...           0.166483   \n",
       "min            0.000000      0.000000      ...           0.000000   \n",
       "25%            0.073615      0.045170      ...           0.068465   \n",
       "50%            0.201737      0.169165      ...           0.211289   \n",
       "75%            0.332201      0.265742      ...           0.322886   \n",
       "max            1.000000      1.000000      ...           1.000000   \n",
       "\n",
       "contig_id  BGC0001424.1  BGC0001425.1  BGC0001426.1  BGC0001427.1  \\\n",
       "count       1551.000000   1551.000000   1551.000000   1551.000000   \n",
       "mean           0.217404      0.212244      0.212244      0.211493   \n",
       "std            0.164887      0.165021      0.165021      0.165130   \n",
       "min            0.000000      0.000000      0.000000      0.000000   \n",
       "25%            0.077850      0.068114      0.068114      0.068982   \n",
       "50%            0.217597      0.211289      0.211289      0.211830   \n",
       "75%            0.335216      0.322817      0.322817      0.323575   \n",
       "max            1.000000      1.000000      1.000000      1.000000   \n",
       "\n",
       "contig_id  BGC0001428.1  BGC0001429.1  BGC0001430.1  BGC0001431.1  \\\n",
       "count       1551.000000   1551.000000   1551.000000   1551.000000   \n",
       "mean           0.213034      0.245220      0.221906      0.242633   \n",
       "std            0.163728      0.180047      0.181935      0.175283   \n",
       "min            0.000000      0.000000      0.000000      0.000000   \n",
       "25%            0.075731      0.092450      0.000000      0.102062   \n",
       "50%            0.212373      0.242831      0.223298      0.243108   \n",
       "75%            0.325560      0.384015      0.363292      0.378620   \n",
       "max            1.000000      1.000000      1.000000      1.000000   \n",
       "\n",
       "contig_id  BGC0001432.1  \n",
       "count       1551.000000  \n",
       "mean           0.262868  \n",
       "std            0.221167  \n",
       "min            0.000000  \n",
       "25%            0.000000  \n",
       "50%            0.255377  \n",
       "75%            0.424665  \n",
       "max            1.000000  \n",
       "\n",
       "[8 rows x 1551 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos_distances.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "for i in range(0, len(pos_distances)):\n",
    "    pos_distances.iloc[i,i] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "contig_id\n",
       "BGC0001240.1    1.0\n",
       "BGC0001249.1    1.0\n",
       "BGC0001107.3    1.0\n",
       "BGC0000480.1    1.0\n",
       "BGC0000474.1    1.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos_most_similar = pos_distances.max(axis=1).sort_values(ascending=False)\n",
    "pos_most_similar.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10b0aca90>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10aac0748>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pos_most_similar.hist(bins=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.11895773785772165"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos_most_similar.quantile(0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>BGC_ID</th>\n",
       "      <th>num_proteins</th>\n",
       "      <th>classes</th>\n",
       "      <th>source</th>\n",
       "      <th>species</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>contig_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>BGC0000001.1</th>\n",
       "      <td>BGC0000001</td>\n",
       "      <td>28</td>\n",
       "      <td>Polyketide</td>\n",
       "      <td>Verrucosispora maris AB-18-032</td>\n",
       "      <td>Verrucosispora</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000002.1</th>\n",
       "      <td>BGC0000002</td>\n",
       "      <td>34</td>\n",
       "      <td>Polyketide</td>\n",
       "      <td>Kutzneria albida DSM 43870</td>\n",
       "      <td>Kutzneria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000003.1</th>\n",
       "      <td>BGC0000003</td>\n",
       "      <td>6</td>\n",
       "      <td>Polyketide</td>\n",
       "      <td>Alternaria alternata</td>\n",
       "      <td>Alternaria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000004.1</th>\n",
       "      <td>BGC0000004</td>\n",
       "      <td>23</td>\n",
       "      <td>Polyketide</td>\n",
       "      <td>Aspergillus oryzae</td>\n",
       "      <td>Aspergillus</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BGC0000005.1</th>\n",
       "      <td>BGC0000005</td>\n",
       "      <td>1</td>\n",
       "      <td>Polyketide</td>\n",
       "      <td>Aspergillus parasiticus</td>\n",
       "      <td>Aspergillus</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  BGC_ID  num_proteins     classes  \\\n",
       "contig_id                                            \n",
       "BGC0000001.1  BGC0000001            28  Polyketide   \n",
       "BGC0000002.1  BGC0000002            34  Polyketide   \n",
       "BGC0000003.1  BGC0000003             6  Polyketide   \n",
       "BGC0000004.1  BGC0000004            23  Polyketide   \n",
       "BGC0000005.1  BGC0000005             1  Polyketide   \n",
       "\n",
       "                                      source         species  \n",
       "contig_id                                                     \n",
       "BGC0000001.1  Verrucosispora maris AB-18-032  Verrucosispora  \n",
       "BGC0000002.1      Kutzneria albida DSM 43870       Kutzneria  \n",
       "BGC0000003.1            Alternaria alternata      Alternaria  \n",
       "BGC0000004.1              Aspergillus oryzae     Aspergillus  \n",
       "BGC0000005.1         Aspergillus parasiticus     Aspergillus  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "props = pd.read_csv('../data/mibig/mibig_properties.csv').set_index('contig_id')\n",
    "props.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Polyketide                     0.274858\n",
       "NRP                            0.156428\n",
       "RiPP                           0.130462\n",
       "NRP;Polyketide                 0.110196\n",
       "Other                          0.103230\n",
       "Saccharide                     0.081064\n",
       "Terpene                        0.057631\n",
       "Alkaloid                       0.017733\n",
       "Polyketide;Saccharide          0.017733\n",
       "Polyketide;Terpene             0.010133\n",
       "?                              0.010133\n",
       "Saccharide;Terpene             0.003800\n",
       "NRP;Polyketide;Saccharide      0.003800\n",
       "Terpene;Alkaloid               0.003167\n",
       "Terpene;Polyketide             0.002533\n",
       "NRP;Saccharide                 0.002533\n",
       "NRP;Other                      0.001900\n",
       "Saccharide;Other               0.001900\n",
       "NRP;Alkaloid                   0.001900\n",
       "Polyketide;Saccharide;Other    0.001267\n",
       "Polyketide;Other;Saccharide    0.000633\n",
       "Saccharide;Polyketide          0.000633\n",
       "NRP;Terpene;Alkaloid           0.000633\n",
       "NRP;RiPP                       0.000633\n",
       "Nucleoside                     0.000633\n",
       "NRP;Polyketide;Alkaloid        0.000633\n",
       "NRP;Polyketide;Other           0.000633\n",
       "Terpene;Saccharide             0.000633\n",
       "Polyketide;NRP                 0.000633\n",
       "NRP;Terpene                    0.000633\n",
       "Other;NRP                      0.000633\n",
       "Polyketide;Alkaloid            0.000633\n",
       "Name: classes, dtype: float64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "class_ratios = props['classes'].value_counts() / len(props)\n",
    "class_ratios"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Polyketide               262\n",
       "NRP                      242\n",
       "RiPP                     123\n",
       "NRP;Polyketide            89\n",
       "Saccharide                78\n",
       "Other                     69\n",
       "Terpene                   55\n",
       "Alkaloid                  24\n",
       "Saccharide;Terpene        22\n",
       "Polyketide;Saccharide     13\n",
       "Name: classes, dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "similar_classes = props.loc[similar_ids[:1000]]['classes'].value_counts()\n",
    "similar_classes.sort_values(ascending=False)[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "BIO_PFAM_IDS = frozenset([\"PF00109\", \"PF02801\", \"PF08659\", \"PF00378\", \"PF08541\", \"PF08545\", \"PF02803\", \"PF00108\",\n",
    "    \"PF02706\", \"PF03364\", \"PF08990\", \"PF00501\", \"PF00668\", \"PF08415\", \"PF00975\", \"PF03061\", \"PF00432\", \"PF00494\", \"PF03936\",\n",
    "    \"PF01397\", \"PF00432\", \"PF04275\", \"PF00348\", \"PF02401\", \"PF04551\", \"PF00368\", \"PF00534\", \"PF00535\", \"PF02922\", \"PF01041\",\n",
    "    \"PF00128\", \"PF00908\", \"PF02719\", \"PF04321\", \"PF01943\", \"PF02806\", \"PF02350\", \"PF02397\", \"PF04932\", \"PF01075\", \"PF00953\",\n",
    "    \"PF01050\", \"PF03033\", \"PF01501\", \"PF05159\", \"PF04101\", \"PF02563\", \"PF08437\", \"PF02585\", \"PF01721\", \"PF02052\", \"PF02674\",\n",
    "    \"PF03515\", \"PF04369\", \"PF08109\", \"PF08129\", \"PF09221\", \"PF09683\", \"PF10439\", \"PF11420\", \"PF11632\", \"PF11758\", \"PF12173\",\n",
    "    \"PF04738\", \"PF04737\", \"PF04604\", \"PF05147\", \"PF08109\", \"PF08129\", \"PF08130\", \"PF00155\", \"PF00202\", \"PF00702\", \"PF06339\",\n",
    "    \"PF04183\", \"PF10331\", \"PF03756\", \"PF00106\", \"PF01370\", \"PF00107\", \"PF08240\", \"PF00441\", \"PF02770\", \"PF02771\", \"PF08028\",\n",
    "    \"PF01408\", \"PF02894\", \"PF00984\", \"PF00725\", \"PF03720\", \"PF03721\", \"PF07993\", \"PF02737\", \"PF00903\", \"PF00037\", \"PF04055\",\n",
    "    \"PF00171\", \"PF00067\", \"PF01266\", \"PF01118\", \"PF02668\", \"PF00248\", \"PF01494\", \"PF01593\", \"PF03992\", \"PF00355\", \"PF01243\",\n",
    "    \"PF00384\", \"PF01488\", \"PF00857\", \"PF04879\", \"PF08241\", \"PF08242\", \"PF00698\", \"PF00483\", \"PF00561\", \"PF00583\", \"PF01636\",\n",
    "    \"PF01039\", \"PF00288\", \"PF00289\", \"PF02786\", \"PF01757\", \"PF02785\", \"PF02409\", \"PF01553\", \"PF02348\", \"PF00891\", \"PF01596\",\n",
    "    \"PF04820\", \"PF02522\", \"PF08484\", \"PF08421\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "neg_domains = pd.read_csv('../data/training/negative/geneswap_negatives.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "pos_domains = pd.read_csv('../data/training/positive/mibig_bgcs_all.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "266059"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(neg_domains)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.07536298339841915\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "20051"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bio_count = sum(neg_domains['pfam_id'].apply(lambda i: i in BIO_PFAM_IDS))\n",
    "print(bio_count / len(neg_domains))\n",
    "bio_count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.3913265029544099\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "36094"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bio_count = sum(pos_domains['pfam_id'].apply(lambda i: i in BIO_PFAM_IDS))\n",
    "print(bio_count / len(pos_domains))\n",
    "bio_count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "23"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "neg_pfams = list(neg_domains[neg_domains['contig_id']=='NEG_FAKE_CLUSTER|BGC0001417.1|AP010918']['pfam_id'])\n",
    "sum([p in BIO_PFAM_IDS for p in neg_pfams])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "remove_threshold = 0.3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4868 4868\n"
     ]
    }
   ],
   "source": [
    "num_total = len(most_similar)\n",
    "filtered = most_similar[most_similar < remove_threshold]\n",
    "print(num_total, len(filtered))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "result = pd.read_csv('../data/training/negative/geneswap_negatives.csv').set_index('contig_id')\n",
    "result.loc[filtered.index].to_csv('../data/training/negative/geneswap_negatives-filtered2.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(254355, 11)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_csv('../data/training/negative/geneswap_negatives-filtered.csv').shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x112077cf8>"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10caeba20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = plt.subplot()\n",
    "ax.plot([0.890, 0.897, 0.927], label='hmm_discrete', color='red', marker='o')\n",
    "ax.plot([0.907, 0.913, 0.938], label='hmm_geneborder', color='purple', marker='o')\n",
    "ax.plot([0.799, 0.865, 0.942], label='lstm', color='green', marker='o')\n",
    "#ax.set_ylim([0.7,1])\n",
    "ax.set_xticks([0,1,2])\n",
    "ax.set_xticklabels(['All negatives', '10% removed', '50% removed'])\n",
    "ax.set_ylabel('LCO AUC')\n",
    "ax.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>contig_id</th>\n",
       "      <th>locus_tag</th>\n",
       "      <th>protein_id</th>\n",
       "      <th>gene_start</th>\n",
       "      <th>gene_end</th>\n",
       "      <th>gene_strand</th>\n",
       "      <th>pfam_id</th>\n",
       "      <th>domain_start</th>\n",
       "      <th>domain_end</th>\n",
       "      <th>evalue</th>\n",
       "      <th>bitscore</th>\n",
       "      <th>in_cluster</th>\n",
       "      <th>prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CM000950.1</td>\n",
       "      <td>SSDG_00780</td>\n",
       "      <td>EDY62461.1</td>\n",
       "      <td>13</td>\n",
       "      <td>1003</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13515</td>\n",
       "      <td>3</td>\n",
       "      <td>127</td>\n",
       "      <td>5.400000e-14</td>\n",
       "      <td>52.2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.750785</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CM000950.1</td>\n",
       "      <td>SSDG_00782</td>\n",
       "      <td>EDY62463.1</td>\n",
       "      <td>1211</td>\n",
       "      <td>2039</td>\n",
       "      <td>1</td>\n",
       "      <td>PF08241</td>\n",
       "      <td>0</td>\n",
       "      <td>95</td>\n",
       "      <td>5.500000e-14</td>\n",
       "      <td>52.4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.791825</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CM000950.1</td>\n",
       "      <td>SSDG_00782</td>\n",
       "      <td>EDY62463.1</td>\n",
       "      <td>1211</td>\n",
       "      <td>2039</td>\n",
       "      <td>1</td>\n",
       "      <td>PF08242</td>\n",
       "      <td>0</td>\n",
       "      <td>99</td>\n",
       "      <td>6.600000e-12</td>\n",
       "      <td>45.7</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.826422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CM000950.1</td>\n",
       "      <td>SSDG_00782</td>\n",
       "      <td>EDY62463.1</td>\n",
       "      <td>1211</td>\n",
       "      <td>2039</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13659</td>\n",
       "      <td>1</td>\n",
       "      <td>113</td>\n",
       "      <td>7.000000e-06</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.853821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CM000950.1</td>\n",
       "      <td>SSDG_00782</td>\n",
       "      <td>EDY62463.1</td>\n",
       "      <td>1211</td>\n",
       "      <td>2039</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13649</td>\n",
       "      <td>1</td>\n",
       "      <td>101</td>\n",
       "      <td>2.400000e-05</td>\n",
       "      <td>24.6</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.884120</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    contig_id   locus_tag  protein_id  gene_start  gene_end  gene_strand  \\\n",
       "0  CM000950.1  SSDG_00780  EDY62461.1          13      1003            1   \n",
       "1  CM000950.1  SSDG_00782  EDY62463.1        1211      2039            1   \n",
       "2  CM000950.1  SSDG_00782  EDY62463.1        1211      2039            1   \n",
       "3  CM000950.1  SSDG_00782  EDY62463.1        1211      2039            1   \n",
       "4  CM000950.1  SSDG_00782  EDY62463.1        1211      2039            1   \n",
       "\n",
       "   pfam_id  domain_start  domain_end        evalue  bitscore  in_cluster  \\\n",
       "0  PF13515             3         127  5.400000e-14      52.2         0.0   \n",
       "1  PF08241             0          95  5.500000e-14      52.4         0.0   \n",
       "2  PF08242             0          99  6.600000e-12      45.7         0.0   \n",
       "3  PF13659             1         113  7.000000e-06      26.0         0.0   \n",
       "4  PF13649             1         101  2.400000e-05      24.6         0.0   \n",
       "\n",
       "   prediction  \n",
       "0    0.750785  \n",
       "1    0.791825  \n",
       "2    0.826422  \n",
       "3    0.853821  \n",
       "4    0.884120  "
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "preds = pd.read_csv('../data/clusterfinder/labelled/prediction-domains/hmm_clusterfinder.pfam27.csv')\n",
    "preds.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_curve, auc\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x111cbb128>"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x111cb6240>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fpr, tpr, thresh = roc_curve(preds['in_cluster'], preds['prediction'])\n",
    "pd.DataFrame({\n",
    "    'fpr': fpr,\n",
    "    'tpr': tpr,\n",
    "}, index=thresh).plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>fpr</th>\n",
       "      <th>tpr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.400193</th>\n",
       "      <td>0.486395</td>\n",
       "      <td>0.921715</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.400192</th>\n",
       "      <td>0.486395</td>\n",
       "      <td>0.921795</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.400105</th>\n",
       "      <td>0.486431</td>\n",
       "      <td>0.921795</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.400093</th>\n",
       "      <td>0.486431</td>\n",
       "      <td>0.921876</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.399345</th>\n",
       "      <td>0.486760</td>\n",
       "      <td>0.921876</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.399322</th>\n",
       "      <td>0.486760</td>\n",
       "      <td>0.921957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.399001</th>\n",
       "      <td>0.486901</td>\n",
       "      <td>0.921957</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               fpr       tpr\n",
       "0.400193  0.486395  0.921715\n",
       "0.400192  0.486395  0.921795\n",
       "0.400105  0.486431  0.921795\n",
       "0.400093  0.486431  0.921876\n",
       "0.399345  0.486760  0.921876\n",
       "0.399322  0.486760  0.921957\n",
       "0.399001  0.486901  0.921957"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vals =pd.DataFrame({\n",
    "    'fpr': fpr,\n",
    "    'tpr': tpr,\n",
    "}, index=thresh)\n",
    "vals[(thresh < 0.401) & (thresh >= 0.399)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>contig_id</th>\n",
       "      <th>locus_tag</th>\n",
       "      <th>protein_id</th>\n",
       "      <th>gene_start</th>\n",
       "      <th>gene_end</th>\n",
       "      <th>gene_strand</th>\n",
       "      <th>pfam_id</th>\n",
       "      <th>domain_start</th>\n",
       "      <th>domain_end</th>\n",
       "      <th>evalue</th>\n",
       "      <th>bitscore</th>\n",
       "      <th>in_cluster</th>\n",
       "      <th>prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CM000950.1</td>\n",
       "      <td>SSDG_00780</td>\n",
       "      <td>EDY62461.1</td>\n",
       "      <td>13</td>\n",
       "      <td>1003</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13515</td>\n",
       "      <td>3</td>\n",
       "      <td>127</td>\n",
       "      <td>1.200000e-12</td>\n",
       "      <td>48.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.667345</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CM000950.1</td>\n",
       "      <td>SSDG_00782</td>\n",
       "      <td>EDY62463.1</td>\n",
       "      <td>1211</td>\n",
       "      <td>2039</td>\n",
       "      <td>1</td>\n",
       "      <td>PF08241</td>\n",
       "      <td>0</td>\n",
       "      <td>95</td>\n",
       "      <td>1.900000e-14</td>\n",
       "      <td>53.9</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.792969</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CM000950.1</td>\n",
       "      <td>SSDG_00782</td>\n",
       "      <td>EDY62463.1</td>\n",
       "      <td>1211</td>\n",
       "      <td>2039</td>\n",
       "      <td>1</td>\n",
       "      <td>PF08242</td>\n",
       "      <td>0</td>\n",
       "      <td>99</td>\n",
       "      <td>5.100000e-12</td>\n",
       "      <td>46.3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.842479</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CM000950.1</td>\n",
       "      <td>SSDG_00782</td>\n",
       "      <td>EDY62463.1</td>\n",
       "      <td>1211</td>\n",
       "      <td>2039</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13649</td>\n",
       "      <td>1</td>\n",
       "      <td>97</td>\n",
       "      <td>4.500000e-12</td>\n",
       "      <td>46.4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.814838</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CM000950.1</td>\n",
       "      <td>SSDG_00782</td>\n",
       "      <td>EDY62463.1</td>\n",
       "      <td>1211</td>\n",
       "      <td>2039</td>\n",
       "      <td>1</td>\n",
       "      <td>PF13847</td>\n",
       "      <td>5</td>\n",
       "      <td>128</td>\n",
       "      <td>8.300000e-11</td>\n",
       "      <td>41.8</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.882505</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    contig_id   locus_tag  protein_id  gene_start  gene_end  gene_strand  \\\n",
       "0  CM000950.1  SSDG_00780  EDY62461.1          13      1003            1   \n",
       "1  CM000950.1  SSDG_00782  EDY62463.1        1211      2039            1   \n",
       "2  CM000950.1  SSDG_00782  EDY62463.1        1211      2039            1   \n",
       "3  CM000950.1  SSDG_00782  EDY62463.1        1211      2039            1   \n",
       "4  CM000950.1  SSDG_00782  EDY62463.1        1211      2039            1   \n",
       "\n",
       "   pfam_id  domain_start  domain_end        evalue  bitscore  in_cluster  \\\n",
       "0  PF13515             3         127  1.200000e-12      48.0         0.0   \n",
       "1  PF08241             0          95  1.900000e-14      53.9         0.0   \n",
       "2  PF08242             0          99  5.100000e-12      46.3         0.0   \n",
       "3  PF13649             1          97  4.500000e-12      46.4         0.0   \n",
       "4  PF13847             5         128  8.300000e-11      41.8         0.0   \n",
       "\n",
       "   prediction  \n",
       "0    0.667345  \n",
       "1    0.792969  \n",
       "2    0.842479  \n",
       "3    0.814838  \n",
       "4    0.882505  "
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "preds = pd.read_csv('../data/clusterfinder/labelled/prediction-domains/lstm_skipgram.csv')\n",
    "preds.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x116c1df98>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x116c1d8d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fpr, tpr, thresh = roc_curve(preds['in_cluster'], preds['prediction'])\n",
    "thresh = np.array([t if t != 2 else 1 for t in thresh])\n",
    "pd.DataFrame({\n",
    "    'fpr': fpr,\n",
    "    'tpr': tpr,\n",
    "}, index=thresh).plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>fpr</th>\n",
       "      <th>tpr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.199321</th>\n",
       "      <td>0.161451</td>\n",
       "      <td>0.862115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.199300</th>\n",
       "      <td>0.161451</td>\n",
       "      <td>0.862199</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               fpr       tpr\n",
       "0.199321  0.161451  0.862115\n",
       "0.199300  0.161451  0.862199"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vals =pd.DataFrame({\n",
    "    'fpr': fpr,\n",
    "    'tpr': tpr,\n",
    "}, index=thresh)\n",
    "vals[thresh <= 0.2].head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tiny_aa</th>\n",
       "      <th>small_aa</th>\n",
       "      <th>aliphatic_aa</th>\n",
       "      <th>aromatic_aa</th>\n",
       "      <th>nonpolar_aa</th>\n",
       "      <th>polar_aa</th>\n",
       "      <th>charged_aa</th>\n",
       "      <th>basic_aa</th>\n",
       "      <th>acidic_aa</th>\n",
       "      <th>tiny_aa_pct</th>\n",
       "      <th>...</th>\n",
       "      <th>nonpolar_aa_pct</th>\n",
       "      <th>polar_aa_pct</th>\n",
       "      <th>charged_aa_pct</th>\n",
       "      <th>basic_aa_pct</th>\n",
       "      <th>acidic_aa_pct</th>\n",
       "      <th>molecular_weight</th>\n",
       "      <th>isoelectric_point</th>\n",
       "      <th>aromaticity</th>\n",
       "      <th>instability</th>\n",
       "      <th>hydropathy_average</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PF00001</th>\n",
       "      <td>68.482759</td>\n",
       "      <td>112.275862</td>\n",
       "      <td>86.310345</td>\n",
       "      <td>37.293103</td>\n",
       "      <td>153.689655</td>\n",
       "      <td>83.068966</td>\n",
       "      <td>35.844828</td>\n",
       "      <td>24.586207</td>\n",
       "      <td>11.258621</td>\n",
       "      <td>29.034111</td>\n",
       "      <td>...</td>\n",
       "      <td>65.367215</td>\n",
       "      <td>34.632785</td>\n",
       "      <td>14.833358</td>\n",
       "      <td>10.118048</td>\n",
       "      <td>4.715310</td>\n",
       "      <td>26821.597610</td>\n",
       "      <td>8.765065</td>\n",
       "      <td>0.136418</td>\n",
       "      <td>37.994918</td>\n",
       "      <td>0.620338</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00002</th>\n",
       "      <td>56.833333</td>\n",
       "      <td>91.907407</td>\n",
       "      <td>80.111111</td>\n",
       "      <td>35.370370</td>\n",
       "      <td>138.407407</td>\n",
       "      <td>67.425926</td>\n",
       "      <td>29.629630</td>\n",
       "      <td>20.055556</td>\n",
       "      <td>9.574074</td>\n",
       "      <td>27.839221</td>\n",
       "      <td>...</td>\n",
       "      <td>67.383447</td>\n",
       "      <td>32.616553</td>\n",
       "      <td>14.401696</td>\n",
       "      <td>10.083362</td>\n",
       "      <td>4.318335</td>\n",
       "      <td>23404.742594</td>\n",
       "      <td>8.956330</td>\n",
       "      <td>0.146652</td>\n",
       "      <td>36.005976</td>\n",
       "      <td>0.807356</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00003</th>\n",
       "      <td>73.068966</td>\n",
       "      <td>113.224138</td>\n",
       "      <td>89.655172</td>\n",
       "      <td>31.086207</td>\n",
       "      <td>154.637931</td>\n",
       "      <td>75.465517</td>\n",
       "      <td>29.396552</td>\n",
       "      <td>19.862069</td>\n",
       "      <td>9.534483</td>\n",
       "      <td>31.755878</td>\n",
       "      <td>...</td>\n",
       "      <td>67.214817</td>\n",
       "      <td>32.785183</td>\n",
       "      <td>12.772931</td>\n",
       "      <td>8.675065</td>\n",
       "      <td>4.097867</td>\n",
       "      <td>25498.628272</td>\n",
       "      <td>8.977002</td>\n",
       "      <td>0.122506</td>\n",
       "      <td>34.805930</td>\n",
       "      <td>0.847421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00004</th>\n",
       "      <td>35.963636</td>\n",
       "      <td>60.036364</td>\n",
       "      <td>40.509091</td>\n",
       "      <td>8.381818</td>\n",
       "      <td>66.763636</td>\n",
       "      <td>48.727273</td>\n",
       "      <td>29.400000</td>\n",
       "      <td>14.872727</td>\n",
       "      <td>14.527273</td>\n",
       "      <td>31.222536</td>\n",
       "      <td>...</td>\n",
       "      <td>58.188302</td>\n",
       "      <td>41.811698</td>\n",
       "      <td>25.224715</td>\n",
       "      <td>13.023254</td>\n",
       "      <td>12.201462</td>\n",
       "      <td>12479.248984</td>\n",
       "      <td>6.444131</td>\n",
       "      <td>0.057692</td>\n",
       "      <td>29.797258</td>\n",
       "      <td>0.032660</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00005</th>\n",
       "      <td>42.014706</td>\n",
       "      <td>70.558824</td>\n",
       "      <td>48.647059</td>\n",
       "      <td>10.073529</td>\n",
       "      <td>80.382353</td>\n",
       "      <td>62.176471</td>\n",
       "      <td>35.485294</td>\n",
       "      <td>18.941176</td>\n",
       "      <td>16.544118</td>\n",
       "      <td>29.263395</td>\n",
       "      <td>...</td>\n",
       "      <td>56.238461</td>\n",
       "      <td>43.761539</td>\n",
       "      <td>24.964813</td>\n",
       "      <td>13.286294</td>\n",
       "      <td>11.678519</td>\n",
       "      <td>15549.698450</td>\n",
       "      <td>7.299559</td>\n",
       "      <td>0.054210</td>\n",
       "      <td>35.749752</td>\n",
       "      <td>-0.114624</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              tiny_aa    small_aa  aliphatic_aa  aromatic_aa  nonpolar_aa  \\\n",
       "Unnamed: 0                                                                  \n",
       "PF00001     68.482759  112.275862     86.310345    37.293103   153.689655   \n",
       "PF00002     56.833333   91.907407     80.111111    35.370370   138.407407   \n",
       "PF00003     73.068966  113.224138     89.655172    31.086207   154.637931   \n",
       "PF00004     35.963636   60.036364     40.509091     8.381818    66.763636   \n",
       "PF00005     42.014706   70.558824     48.647059    10.073529    80.382353   \n",
       "\n",
       "             polar_aa  charged_aa   basic_aa  acidic_aa  tiny_aa_pct  \\\n",
       "Unnamed: 0                                                             \n",
       "PF00001     83.068966   35.844828  24.586207  11.258621    29.034111   \n",
       "PF00002     67.425926   29.629630  20.055556   9.574074    27.839221   \n",
       "PF00003     75.465517   29.396552  19.862069   9.534483    31.755878   \n",
       "PF00004     48.727273   29.400000  14.872727  14.527273    31.222536   \n",
       "PF00005     62.176471   35.485294  18.941176  16.544118    29.263395   \n",
       "\n",
       "                   ...          nonpolar_aa_pct  polar_aa_pct  charged_aa_pct  \\\n",
       "Unnamed: 0         ...                                                          \n",
       "PF00001            ...                65.367215     34.632785       14.833358   \n",
       "PF00002            ...                67.383447     32.616553       14.401696   \n",
       "PF00003            ...                67.214817     32.785183       12.772931   \n",
       "PF00004            ...                58.188302     41.811698       25.224715   \n",
       "PF00005            ...                56.238461     43.761539       24.964813   \n",
       "\n",
       "            basic_aa_pct  acidic_aa_pct  molecular_weight  isoelectric_point  \\\n",
       "Unnamed: 0                                                                     \n",
       "PF00001        10.118048       4.715310      26821.597610           8.765065   \n",
       "PF00002        10.083362       4.318335      23404.742594           8.956330   \n",
       "PF00003         8.675065       4.097867      25498.628272           8.977002   \n",
       "PF00004        13.023254      12.201462      12479.248984           6.444131   \n",
       "PF00005        13.286294      11.678519      15549.698450           7.299559   \n",
       "\n",
       "            aromaticity  instability  hydropathy_average  \n",
       "Unnamed: 0                                                \n",
       "PF00001        0.136418    37.994918            0.620338  \n",
       "PF00002        0.146652    36.005976            0.807356  \n",
       "PF00003        0.122506    34.805930            0.847421  \n",
       "PF00004        0.057692    29.797258            0.032660  \n",
       "PF00005        0.054210    35.749752           -0.114624  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d = pd.read_csv('~/Downloads/pfam_properties.csv').set_index('Unnamed: 0')\n",
    "d.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "del d['molecular_weight']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "c = d[:1000].transpose().corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "c.values[[np.arange(len(c))]*2] = None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>PF00001</th>\n",
       "      <th>PF00002</th>\n",
       "      <th>PF00003</th>\n",
       "      <th>PF00004</th>\n",
       "      <th>PF00005</th>\n",
       "      <th>PF00006</th>\n",
       "      <th>PF00007</th>\n",
       "      <th>PF00008</th>\n",
       "      <th>PF00009</th>\n",
       "      <th>PF00010</th>\n",
       "      <th>...</th>\n",
       "      <th>PF01034</th>\n",
       "      <th>PF01035</th>\n",
       "      <th>PF01036</th>\n",
       "      <th>PF01037</th>\n",
       "      <th>PF01039</th>\n",
       "      <th>PF01040</th>\n",
       "      <th>PF01041</th>\n",
       "      <th>PF01042</th>\n",
       "      <th>PF01043</th>\n",
       "      <th>PF01044</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PF00001</th>\n",
       "      <td>NaN</td>\n",
       "      <td>0.994152</td>\n",
       "      <td>0.996767</td>\n",
       "      <td>0.858155</td>\n",
       "      <td>0.916370</td>\n",
       "      <td>0.967561</td>\n",
       "      <td>0.639188</td>\n",
       "      <td>0.219116</td>\n",
       "      <td>0.940149</td>\n",
       "      <td>0.229814</td>\n",
       "      <td>...</td>\n",
       "      <td>0.479600</td>\n",
       "      <td>0.690161</td>\n",
       "      <td>0.928384</td>\n",
       "      <td>0.569138</td>\n",
       "      <td>0.961168</td>\n",
       "      <td>0.990399</td>\n",
       "      <td>0.954692</td>\n",
       "      <td>0.850712</td>\n",
       "      <td>0.767659</td>\n",
       "      <td>0.899831</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00002</th>\n",
       "      <td>0.994152</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.994071</td>\n",
       "      <td>0.868736</td>\n",
       "      <td>0.914337</td>\n",
       "      <td>0.949893</td>\n",
       "      <td>0.650648</td>\n",
       "      <td>0.262220</td>\n",
       "      <td>0.920265</td>\n",
       "      <td>0.264869</td>\n",
       "      <td>...</td>\n",
       "      <td>0.539192</td>\n",
       "      <td>0.727421</td>\n",
       "      <td>0.955501</td>\n",
       "      <td>0.607857</td>\n",
       "      <td>0.929704</td>\n",
       "      <td>0.990500</td>\n",
       "      <td>0.921738</td>\n",
       "      <td>0.866706</td>\n",
       "      <td>0.769727</td>\n",
       "      <td>0.858468</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00003</th>\n",
       "      <td>0.996767</td>\n",
       "      <td>0.994071</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.863185</td>\n",
       "      <td>0.913619</td>\n",
       "      <td>0.964204</td>\n",
       "      <td>0.640073</td>\n",
       "      <td>0.240055</td>\n",
       "      <td>0.929600</td>\n",
       "      <td>0.225174</td>\n",
       "      <td>...</td>\n",
       "      <td>0.498167</td>\n",
       "      <td>0.704774</td>\n",
       "      <td>0.945078</td>\n",
       "      <td>0.576428</td>\n",
       "      <td>0.951380</td>\n",
       "      <td>0.996520</td>\n",
       "      <td>0.940415</td>\n",
       "      <td>0.858530</td>\n",
       "      <td>0.749562</td>\n",
       "      <td>0.880369</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00004</th>\n",
       "      <td>0.858155</td>\n",
       "      <td>0.868736</td>\n",
       "      <td>0.863185</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.983022</td>\n",
       "      <td>0.920366</td>\n",
       "      <td>0.879691</td>\n",
       "      <td>0.607143</td>\n",
       "      <td>0.921873</td>\n",
       "      <td>0.588337</td>\n",
       "      <td>...</td>\n",
       "      <td>0.808621</td>\n",
       "      <td>0.936848</td>\n",
       "      <td>0.907961</td>\n",
       "      <td>0.884591</td>\n",
       "      <td>0.810347</td>\n",
       "      <td>0.844733</td>\n",
       "      <td>0.807403</td>\n",
       "      <td>0.993223</td>\n",
       "      <td>0.935176</td>\n",
       "      <td>0.787006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00005</th>\n",
       "      <td>0.916370</td>\n",
       "      <td>0.914337</td>\n",
       "      <td>0.913619</td>\n",
       "      <td>0.983022</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.967823</td>\n",
       "      <td>0.835539</td>\n",
       "      <td>0.478582</td>\n",
       "      <td>0.970691</td>\n",
       "      <td>0.522997</td>\n",
       "      <td>...</td>\n",
       "      <td>0.701140</td>\n",
       "      <td>0.867017</td>\n",
       "      <td>0.901900</td>\n",
       "      <td>0.809053</td>\n",
       "      <td>0.891569</td>\n",
       "      <td>0.892440</td>\n",
       "      <td>0.892491</td>\n",
       "      <td>0.969539</td>\n",
       "      <td>0.938878</td>\n",
       "      <td>0.879609</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 1000 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "Unnamed: 0   PF00001   PF00002   PF00003   PF00004   PF00005   PF00006  \\\n",
       "Unnamed: 0                                                               \n",
       "PF00001          NaN  0.994152  0.996767  0.858155  0.916370  0.967561   \n",
       "PF00002     0.994152       NaN  0.994071  0.868736  0.914337  0.949893   \n",
       "PF00003     0.996767  0.994071       NaN  0.863185  0.913619  0.964204   \n",
       "PF00004     0.858155  0.868736  0.863185       NaN  0.983022  0.920366   \n",
       "PF00005     0.916370  0.914337  0.913619  0.983022       NaN  0.967823   \n",
       "\n",
       "Unnamed: 0   PF00007   PF00008   PF00009   PF00010    ...      PF01034  \\\n",
       "Unnamed: 0                                            ...                \n",
       "PF00001     0.639188  0.219116  0.940149  0.229814    ...     0.479600   \n",
       "PF00002     0.650648  0.262220  0.920265  0.264869    ...     0.539192   \n",
       "PF00003     0.640073  0.240055  0.929600  0.225174    ...     0.498167   \n",
       "PF00004     0.879691  0.607143  0.921873  0.588337    ...     0.808621   \n",
       "PF00005     0.835539  0.478582  0.970691  0.522997    ...     0.701140   \n",
       "\n",
       "Unnamed: 0   PF01035   PF01036   PF01037   PF01039   PF01040   PF01041  \\\n",
       "Unnamed: 0                                                               \n",
       "PF00001     0.690161  0.928384  0.569138  0.961168  0.990399  0.954692   \n",
       "PF00002     0.727421  0.955501  0.607857  0.929704  0.990500  0.921738   \n",
       "PF00003     0.704774  0.945078  0.576428  0.951380  0.996520  0.940415   \n",
       "PF00004     0.936848  0.907961  0.884591  0.810347  0.844733  0.807403   \n",
       "PF00005     0.867017  0.901900  0.809053  0.891569  0.892440  0.892491   \n",
       "\n",
       "Unnamed: 0   PF01042   PF01043   PF01044  \n",
       "Unnamed: 0                                \n",
       "PF00001     0.850712  0.767659  0.899831  \n",
       "PF00002     0.866706  0.769727  0.858468  \n",
       "PF00003     0.858530  0.749562  0.880369  \n",
       "PF00004     0.993223  0.935176  0.787006  \n",
       "PF00005     0.969539  0.938878  0.879609  \n",
       "\n",
       "[5 rows x 1000 columns]"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Unnamed: 0\n",
       "PF00001    PF00955\n",
       "PF00002    PF00520\n",
       "PF00003    PF00361\n",
       "PF00004    PF00156\n",
       "PF00005    PF00817\n",
       "PF00006    PF00208\n",
       "PF00007    PF00236\n",
       "PF00008    PF00706\n",
       "PF00009    PF00889\n",
       "PF00010    PF00046\n",
       "dtype: object"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c.idxmax().head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Unnamed: 0\n",
       "PF00001    PF00904\n",
       "PF00002    PF00904\n",
       "PF00003    PF00904\n",
       "PF00004    PF00976\n",
       "PF00005    PF00976\n",
       "PF00006    PF00976\n",
       "PF00007    PF00747\n",
       "PF00008    PF00946\n",
       "PF00009    PF00976\n",
       "PF00010    PF00747\n",
       "dtype: object"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c.idxmin().head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x11b73ad30>"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11ce08be0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "d.loc[['PF00001','PF00002','PF00003']].transpose().plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x12547da58>"
      ]
     },
     "execution_count": 185,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1254bf908>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "d.loc[['PF00008','PF00706','PF00946','PF00147']].transpose().plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from gensim.models import Word2Vec; \n",
    "import pandas as pd; \n",
    "model=Word2Vec.load(\"../data/features/evaluation/pfam2vec_corpus-1e-02_skipgram_50dim.bin\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 187,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.3935427 , -1.3778838 , -1.3136594 , -0.17280808,  1.8323512 ,\n",
       "       -0.33866954,  1.7441221 ,  0.09685533, -0.56599593, -0.16126059,\n",
       "        0.10433929, -0.4772497 ,  0.8926293 ,  0.73920834,  0.44773048,\n",
       "       -0.33838654,  2.1351182 , -0.19783598, -1.3741797 , -0.87960106,\n",
       "       -0.4032847 , -0.9469104 , -1.2055805 ,  0.80832046, -0.6135054 ,\n",
       "       -2.7610176 , -1.1220434 ,  1.5865    , -0.01772345, -0.8797336 ,\n",
       "        0.00524217, -1.6172864 ,  0.83534026, -1.2821082 ,  0.9094383 ,\n",
       "       -1.4824384 ,  0.24430816, -1.1892785 , -0.6926669 ,  0.05665533,\n",
       "        0.00923812,  0.72111577,  0.09838675,  0.7252849 , -0.2877881 ,\n",
       "        1.8225182 , -0.17489283,  1.0115119 ,  0.30458158, -0.8383727 ],\n",
       "      dtype=float32)"
      ]
     },
     "execution_count": 187,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.wv['PF00946']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "keys = list(model.wv.vocab.keys())\n",
    "vectors = [model.wv[k] for k in keys]\n",
    "idx = pd.Series(keys, name='pfam_id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "vecs = pd.DataFrame(vectors, index=idx)\n",
    "vecs.to_csv(\"../data/features/evaluation/pfam2vec_10win_50dim_skipgram.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 216,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import normalize"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 241,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tiny_aa</th>\n",
       "      <th>small_aa</th>\n",
       "      <th>aliphatic_aa</th>\n",
       "      <th>aromatic_aa</th>\n",
       "      <th>nonpolar_aa</th>\n",
       "      <th>polar_aa</th>\n",
       "      <th>charged_aa</th>\n",
       "      <th>basic_aa</th>\n",
       "      <th>acidic_aa</th>\n",
       "      <th>tiny_aa_pct</th>\n",
       "      <th>...</th>\n",
       "      <th>aromatic_aa_pct</th>\n",
       "      <th>nonpolar_aa_pct</th>\n",
       "      <th>polar_aa_pct</th>\n",
       "      <th>charged_aa_pct</th>\n",
       "      <th>basic_aa_pct</th>\n",
       "      <th>acidic_aa_pct</th>\n",
       "      <th>isoelectric_point</th>\n",
       "      <th>aromaticity</th>\n",
       "      <th>instability</th>\n",
       "      <th>hydropathy_average</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PF00001</th>\n",
       "      <td>0.257560</td>\n",
       "      <td>0.422263</td>\n",
       "      <td>0.324609</td>\n",
       "      <td>0.140257</td>\n",
       "      <td>0.578019</td>\n",
       "      <td>0.312418</td>\n",
       "      <td>0.134810</td>\n",
       "      <td>0.092467</td>\n",
       "      <td>0.042343</td>\n",
       "      <td>0.109196</td>\n",
       "      <td>...</td>\n",
       "      <td>0.059766</td>\n",
       "      <td>0.245843</td>\n",
       "      <td>0.130252</td>\n",
       "      <td>0.055787</td>\n",
       "      <td>0.038053</td>\n",
       "      <td>0.017734</td>\n",
       "      <td>0.032965</td>\n",
       "      <td>0.000513</td>\n",
       "      <td>0.142897</td>\n",
       "      <td>0.002333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00002</th>\n",
       "      <td>0.239458</td>\n",
       "      <td>0.387237</td>\n",
       "      <td>0.337535</td>\n",
       "      <td>0.149027</td>\n",
       "      <td>0.583157</td>\n",
       "      <td>0.284088</td>\n",
       "      <td>0.124840</td>\n",
       "      <td>0.084501</td>\n",
       "      <td>0.040339</td>\n",
       "      <td>0.117296</td>\n",
       "      <td>...</td>\n",
       "      <td>0.072245</td>\n",
       "      <td>0.283909</td>\n",
       "      <td>0.137424</td>\n",
       "      <td>0.060679</td>\n",
       "      <td>0.042485</td>\n",
       "      <td>0.018195</td>\n",
       "      <td>0.037736</td>\n",
       "      <td>0.000618</td>\n",
       "      <td>0.151705</td>\n",
       "      <td>0.003402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00003</th>\n",
       "      <td>0.275416</td>\n",
       "      <td>0.426772</td>\n",
       "      <td>0.337934</td>\n",
       "      <td>0.117172</td>\n",
       "      <td>0.582872</td>\n",
       "      <td>0.284450</td>\n",
       "      <td>0.110803</td>\n",
       "      <td>0.074865</td>\n",
       "      <td>0.035938</td>\n",
       "      <td>0.119696</td>\n",
       "      <td>...</td>\n",
       "      <td>0.050753</td>\n",
       "      <td>0.253351</td>\n",
       "      <td>0.123576</td>\n",
       "      <td>0.048145</td>\n",
       "      <td>0.032699</td>\n",
       "      <td>0.015446</td>\n",
       "      <td>0.033837</td>\n",
       "      <td>0.000462</td>\n",
       "      <td>0.131193</td>\n",
       "      <td>0.003194</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00004</th>\n",
       "      <td>0.219951</td>\n",
       "      <td>0.367178</td>\n",
       "      <td>0.247751</td>\n",
       "      <td>0.051263</td>\n",
       "      <td>0.408322</td>\n",
       "      <td>0.298013</td>\n",
       "      <td>0.179808</td>\n",
       "      <td>0.090961</td>\n",
       "      <td>0.088848</td>\n",
       "      <td>0.190955</td>\n",
       "      <td>...</td>\n",
       "      <td>0.043591</td>\n",
       "      <td>0.355876</td>\n",
       "      <td>0.255717</td>\n",
       "      <td>0.154273</td>\n",
       "      <td>0.079649</td>\n",
       "      <td>0.074623</td>\n",
       "      <td>0.039412</td>\n",
       "      <td>0.000353</td>\n",
       "      <td>0.182238</td>\n",
       "      <td>0.000200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00005</th>\n",
       "      <td>0.230318</td>\n",
       "      <td>0.386793</td>\n",
       "      <td>0.266676</td>\n",
       "      <td>0.055222</td>\n",
       "      <td>0.440644</td>\n",
       "      <td>0.340842</td>\n",
       "      <td>0.194525</td>\n",
       "      <td>0.103833</td>\n",
       "      <td>0.090692</td>\n",
       "      <td>0.160417</td>\n",
       "      <td>...</td>\n",
       "      <td>0.038679</td>\n",
       "      <td>0.308291</td>\n",
       "      <td>0.239894</td>\n",
       "      <td>0.136853</td>\n",
       "      <td>0.072833</td>\n",
       "      <td>0.064020</td>\n",
       "      <td>0.040015</td>\n",
       "      <td>0.000297</td>\n",
       "      <td>0.195975</td>\n",
       "      <td>-0.000628</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             tiny_aa  small_aa  aliphatic_aa  aromatic_aa  nonpolar_aa  \\\n",
       "Unnamed: 0                                                               \n",
       "PF00001     0.257560  0.422263      0.324609     0.140257     0.578019   \n",
       "PF00002     0.239458  0.387237      0.337535     0.149027     0.583157   \n",
       "PF00003     0.275416  0.426772      0.337934     0.117172     0.582872   \n",
       "PF00004     0.219951  0.367178      0.247751     0.051263     0.408322   \n",
       "PF00005     0.230318  0.386793      0.266676     0.055222     0.440644   \n",
       "\n",
       "            polar_aa  charged_aa  basic_aa  acidic_aa  tiny_aa_pct  \\\n",
       "Unnamed: 0                                                           \n",
       "PF00001     0.312418    0.134810  0.092467   0.042343     0.109196   \n",
       "PF00002     0.284088    0.124840  0.084501   0.040339     0.117296   \n",
       "PF00003     0.284450    0.110803  0.074865   0.035938     0.119696   \n",
       "PF00004     0.298013    0.179808  0.090961   0.088848     0.190955   \n",
       "PF00005     0.340842    0.194525  0.103833   0.090692     0.160417   \n",
       "\n",
       "                   ...          aromatic_aa_pct  nonpolar_aa_pct  \\\n",
       "Unnamed: 0         ...                                             \n",
       "PF00001            ...                 0.059766         0.245843   \n",
       "PF00002            ...                 0.072245         0.283909   \n",
       "PF00003            ...                 0.050753         0.253351   \n",
       "PF00004            ...                 0.043591         0.355876   \n",
       "PF00005            ...                 0.038679         0.308291   \n",
       "\n",
       "            polar_aa_pct  charged_aa_pct  basic_aa_pct  acidic_aa_pct  \\\n",
       "Unnamed: 0                                                              \n",
       "PF00001         0.130252        0.055787      0.038053       0.017734   \n",
       "PF00002         0.137424        0.060679      0.042485       0.018195   \n",
       "PF00003         0.123576        0.048145      0.032699       0.015446   \n",
       "PF00004         0.255717        0.154273      0.079649       0.074623   \n",
       "PF00005         0.239894        0.136853      0.072833       0.064020   \n",
       "\n",
       "            isoelectric_point  aromaticity  instability  hydropathy_average  \n",
       "Unnamed: 0                                                                   \n",
       "PF00001              0.032965     0.000513     0.142897            0.002333  \n",
       "PF00002              0.037736     0.000618     0.151705            0.003402  \n",
       "PF00003              0.033837     0.000462     0.131193            0.003194  \n",
       "PF00004              0.039412     0.000353     0.182238            0.000200  \n",
       "PF00005              0.040015     0.000297     0.195975           -0.000628  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "execution_count": 241,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d_norm = pd.DataFrame(normalize(d), index=d.index, columns=d.columns)\n",
    "d_norm.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 215,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>40</th>\n",
       "      <th>41</th>\n",
       "      <th>42</th>\n",
       "      <th>43</th>\n",
       "      <th>44</th>\n",
       "      <th>45</th>\n",
       "      <th>46</th>\n",
       "      <th>47</th>\n",
       "      <th>48</th>\n",
       "      <th>49</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pfam_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PF01029</th>\n",
       "      <td>0.124140</td>\n",
       "      <td>-0.151212</td>\n",
       "      <td>0.270142</td>\n",
       "      <td>0.102252</td>\n",
       "      <td>0.047339</td>\n",
       "      <td>-0.041438</td>\n",
       "      <td>-0.204322</td>\n",
       "      <td>-0.187459</td>\n",
       "      <td>-0.283284</td>\n",
       "      <td>0.006950</td>\n",
       "      <td>...</td>\n",
       "      <td>0.222358</td>\n",
       "      <td>-0.110555</td>\n",
       "      <td>-0.001405</td>\n",
       "      <td>-0.215833</td>\n",
       "      <td>0.283268</td>\n",
       "      <td>0.001955</td>\n",
       "      <td>-0.013733</td>\n",
       "      <td>0.084766</td>\n",
       "      <td>0.124456</td>\n",
       "      <td>-0.145057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00885</th>\n",
       "      <td>0.004629</td>\n",
       "      <td>-0.011709</td>\n",
       "      <td>0.396750</td>\n",
       "      <td>0.126094</td>\n",
       "      <td>0.011542</td>\n",
       "      <td>-0.076724</td>\n",
       "      <td>-0.236473</td>\n",
       "      <td>-0.126249</td>\n",
       "      <td>-0.279889</td>\n",
       "      <td>-0.086943</td>\n",
       "      <td>...</td>\n",
       "      <td>0.212442</td>\n",
       "      <td>0.008903</td>\n",
       "      <td>0.127951</td>\n",
       "      <td>0.035227</td>\n",
       "      <td>-0.128900</td>\n",
       "      <td>-0.006543</td>\n",
       "      <td>0.043650</td>\n",
       "      <td>0.174301</td>\n",
       "      <td>0.298973</td>\n",
       "      <td>-0.224179</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00793</th>\n",
       "      <td>0.028101</td>\n",
       "      <td>-0.207178</td>\n",
       "      <td>0.149510</td>\n",
       "      <td>0.022479</td>\n",
       "      <td>0.100622</td>\n",
       "      <td>0.031000</td>\n",
       "      <td>-0.049362</td>\n",
       "      <td>-0.084984</td>\n",
       "      <td>-0.162001</td>\n",
       "      <td>0.126938</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.028869</td>\n",
       "      <td>0.064608</td>\n",
       "      <td>-0.111949</td>\n",
       "      <td>-0.072009</td>\n",
       "      <td>-0.298301</td>\n",
       "      <td>0.031149</td>\n",
       "      <td>0.060981</td>\n",
       "      <td>-0.001228</td>\n",
       "      <td>-0.189962</td>\n",
       "      <td>0.080244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00484</th>\n",
       "      <td>0.121855</td>\n",
       "      <td>0.091705</td>\n",
       "      <td>0.145550</td>\n",
       "      <td>0.050273</td>\n",
       "      <td>0.297315</td>\n",
       "      <td>-0.101037</td>\n",
       "      <td>0.044089</td>\n",
       "      <td>0.132565</td>\n",
       "      <td>-0.069157</td>\n",
       "      <td>-0.104260</td>\n",
       "      <td>...</td>\n",
       "      <td>0.050992</td>\n",
       "      <td>0.336132</td>\n",
       "      <td>0.148423</td>\n",
       "      <td>-0.040369</td>\n",
       "      <td>0.170472</td>\n",
       "      <td>-0.085966</td>\n",
       "      <td>-0.059718</td>\n",
       "      <td>-0.049444</td>\n",
       "      <td>-0.060700</td>\n",
       "      <td>0.127099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00215</th>\n",
       "      <td>0.108870</td>\n",
       "      <td>0.028832</td>\n",
       "      <td>0.217392</td>\n",
       "      <td>0.118680</td>\n",
       "      <td>-0.071785</td>\n",
       "      <td>0.047683</td>\n",
       "      <td>-0.317534</td>\n",
       "      <td>-0.052578</td>\n",
       "      <td>-0.214511</td>\n",
       "      <td>-0.176985</td>\n",
       "      <td>...</td>\n",
       "      <td>0.017877</td>\n",
       "      <td>0.012829</td>\n",
       "      <td>0.028967</td>\n",
       "      <td>-0.078334</td>\n",
       "      <td>0.156614</td>\n",
       "      <td>-0.058029</td>\n",
       "      <td>-0.022584</td>\n",
       "      <td>-0.014883</td>\n",
       "      <td>0.215639</td>\n",
       "      <td>-0.052917</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 50 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               0         1         2         3         4         5         6   \\\n",
       "pfam_id                                                                         \n",
       "PF01029  0.124140 -0.151212  0.270142  0.102252  0.047339 -0.041438 -0.204322   \n",
       "PF00885  0.004629 -0.011709  0.396750  0.126094  0.011542 -0.076724 -0.236473   \n",
       "PF00793  0.028101 -0.207178  0.149510  0.022479  0.100622  0.031000 -0.049362   \n",
       "PF00484  0.121855  0.091705  0.145550  0.050273  0.297315 -0.101037  0.044089   \n",
       "PF00215  0.108870  0.028832  0.217392  0.118680 -0.071785  0.047683 -0.317534   \n",
       "\n",
       "               7         8         9     ...           40        41        42  \\\n",
       "pfam_id                                  ...                                    \n",
       "PF01029 -0.187459 -0.283284  0.006950    ...     0.222358 -0.110555 -0.001405   \n",
       "PF00885 -0.126249 -0.279889 -0.086943    ...     0.212442  0.008903  0.127951   \n",
       "PF00793 -0.084984 -0.162001  0.126938    ...    -0.028869  0.064608 -0.111949   \n",
       "PF00484  0.132565 -0.069157 -0.104260    ...     0.050992  0.336132  0.148423   \n",
       "PF00215 -0.052578 -0.214511 -0.176985    ...     0.017877  0.012829  0.028967   \n",
       "\n",
       "               43        44        45        46        47        48        49  \n",
       "pfam_id                                                                        \n",
       "PF01029 -0.215833  0.283268  0.001955 -0.013733  0.084766  0.124456 -0.145057  \n",
       "PF00885  0.035227 -0.128900 -0.006543  0.043650  0.174301  0.298973 -0.224179  \n",
       "PF00793 -0.072009 -0.298301  0.031149  0.060981 -0.001228 -0.189962  0.080244  \n",
       "PF00484 -0.040369  0.170472 -0.085966 -0.059718 -0.049444 -0.060700  0.127099  \n",
       "PF00215 -0.078334  0.156614 -0.058029 -0.022584 -0.014883  0.215639 -0.052917  \n",
       "\n",
       "[5 rows x 50 columns]"
      ]
     },
     "execution_count": 215,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vecs_norm = pd.DataFrame(normalize(vecs), index=vecs.index)\n",
    "vecs_norm.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 244,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tiny_aa</th>\n",
       "      <th>small_aa</th>\n",
       "      <th>aliphatic_aa</th>\n",
       "      <th>aromatic_aa</th>\n",
       "      <th>nonpolar_aa</th>\n",
       "      <th>polar_aa</th>\n",
       "      <th>charged_aa</th>\n",
       "      <th>basic_aa</th>\n",
       "      <th>acidic_aa</th>\n",
       "      <th>tiny_aa_pct</th>\n",
       "      <th>...</th>\n",
       "      <th>40</th>\n",
       "      <th>41</th>\n",
       "      <th>42</th>\n",
       "      <th>43</th>\n",
       "      <th>44</th>\n",
       "      <th>45</th>\n",
       "      <th>46</th>\n",
       "      <th>47</th>\n",
       "      <th>48</th>\n",
       "      <th>49</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PF00001</th>\n",
       "      <td>0.257560</td>\n",
       "      <td>0.422263</td>\n",
       "      <td>0.324609</td>\n",
       "      <td>0.140257</td>\n",
       "      <td>0.578019</td>\n",
       "      <td>0.312418</td>\n",
       "      <td>0.134810</td>\n",
       "      <td>0.092467</td>\n",
       "      <td>0.042343</td>\n",
       "      <td>0.109196</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.142532</td>\n",
       "      <td>-0.176323</td>\n",
       "      <td>0.001416</td>\n",
       "      <td>-0.003957</td>\n",
       "      <td>0.009169</td>\n",
       "      <td>-0.047664</td>\n",
       "      <td>0.087643</td>\n",
       "      <td>-0.234951</td>\n",
       "      <td>-0.168454</td>\n",
       "      <td>0.039331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00002</th>\n",
       "      <td>0.239458</td>\n",
       "      <td>0.387237</td>\n",
       "      <td>0.337535</td>\n",
       "      <td>0.149027</td>\n",
       "      <td>0.583157</td>\n",
       "      <td>0.284088</td>\n",
       "      <td>0.124840</td>\n",
       "      <td>0.084501</td>\n",
       "      <td>0.040339</td>\n",
       "      <td>0.117296</td>\n",
       "      <td>...</td>\n",
       "      <td>0.112091</td>\n",
       "      <td>0.155145</td>\n",
       "      <td>0.071038</td>\n",
       "      <td>-0.199625</td>\n",
       "      <td>-0.143579</td>\n",
       "      <td>0.118461</td>\n",
       "      <td>0.152977</td>\n",
       "      <td>0.150701</td>\n",
       "      <td>0.058932</td>\n",
       "      <td>-0.042090</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00003</th>\n",
       "      <td>0.275416</td>\n",
       "      <td>0.426772</td>\n",
       "      <td>0.337934</td>\n",
       "      <td>0.117172</td>\n",
       "      <td>0.582872</td>\n",
       "      <td>0.284450</td>\n",
       "      <td>0.110803</td>\n",
       "      <td>0.074865</td>\n",
       "      <td>0.035938</td>\n",
       "      <td>0.119696</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.188145</td>\n",
       "      <td>-0.170506</td>\n",
       "      <td>-0.004728</td>\n",
       "      <td>-0.086949</td>\n",
       "      <td>-0.094884</td>\n",
       "      <td>-0.192082</td>\n",
       "      <td>-0.281317</td>\n",
       "      <td>0.121297</td>\n",
       "      <td>0.060522</td>\n",
       "      <td>-0.003963</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00004</th>\n",
       "      <td>0.219951</td>\n",
       "      <td>0.367178</td>\n",
       "      <td>0.247751</td>\n",
       "      <td>0.051263</td>\n",
       "      <td>0.408322</td>\n",
       "      <td>0.298013</td>\n",
       "      <td>0.179808</td>\n",
       "      <td>0.090961</td>\n",
       "      <td>0.088848</td>\n",
       "      <td>0.190955</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.056701</td>\n",
       "      <td>0.074335</td>\n",
       "      <td>0.159641</td>\n",
       "      <td>-0.012542</td>\n",
       "      <td>0.174600</td>\n",
       "      <td>0.214575</td>\n",
       "      <td>0.117087</td>\n",
       "      <td>0.175630</td>\n",
       "      <td>-0.008259</td>\n",
       "      <td>0.100392</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00005</th>\n",
       "      <td>0.230318</td>\n",
       "      <td>0.386793</td>\n",
       "      <td>0.266676</td>\n",
       "      <td>0.055222</td>\n",
       "      <td>0.440644</td>\n",
       "      <td>0.340842</td>\n",
       "      <td>0.194525</td>\n",
       "      <td>0.103833</td>\n",
       "      <td>0.090692</td>\n",
       "      <td>0.160417</td>\n",
       "      <td>...</td>\n",
       "      <td>0.117126</td>\n",
       "      <td>0.180255</td>\n",
       "      <td>-0.047497</td>\n",
       "      <td>0.206509</td>\n",
       "      <td>-0.023069</td>\n",
       "      <td>0.192936</td>\n",
       "      <td>-0.044671</td>\n",
       "      <td>-0.007429</td>\n",
       "      <td>-0.281552</td>\n",
       "      <td>0.301284</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 72 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             tiny_aa  small_aa  aliphatic_aa  aromatic_aa  nonpolar_aa  \\\n",
       "Unnamed: 0                                                               \n",
       "PF00001     0.257560  0.422263      0.324609     0.140257     0.578019   \n",
       "PF00002     0.239458  0.387237      0.337535     0.149027     0.583157   \n",
       "PF00003     0.275416  0.426772      0.337934     0.117172     0.582872   \n",
       "PF00004     0.219951  0.367178      0.247751     0.051263     0.408322   \n",
       "PF00005     0.230318  0.386793      0.266676     0.055222     0.440644   \n",
       "\n",
       "            polar_aa  charged_aa  basic_aa  acidic_aa  tiny_aa_pct    ...     \\\n",
       "Unnamed: 0                                                            ...      \n",
       "PF00001     0.312418    0.134810  0.092467   0.042343     0.109196    ...      \n",
       "PF00002     0.284088    0.124840  0.084501   0.040339     0.117296    ...      \n",
       "PF00003     0.284450    0.110803  0.074865   0.035938     0.119696    ...      \n",
       "PF00004     0.298013    0.179808  0.090961   0.088848     0.190955    ...      \n",
       "PF00005     0.340842    0.194525  0.103833   0.090692     0.160417    ...      \n",
       "\n",
       "                  40        41        42        43        44        45  \\\n",
       "Unnamed: 0                                                               \n",
       "PF00001    -0.142532 -0.176323  0.001416 -0.003957  0.009169 -0.047664   \n",
       "PF00002     0.112091  0.155145  0.071038 -0.199625 -0.143579  0.118461   \n",
       "PF00003    -0.188145 -0.170506 -0.004728 -0.086949 -0.094884 -0.192082   \n",
       "PF00004    -0.056701  0.074335  0.159641 -0.012542  0.174600  0.214575   \n",
       "PF00005     0.117126  0.180255 -0.047497  0.206509 -0.023069  0.192936   \n",
       "\n",
       "                  46        47        48        49  \n",
       "Unnamed: 0                                          \n",
       "PF00001     0.087643 -0.234951 -0.168454  0.039331  \n",
       "PF00002     0.152977  0.150701  0.058932 -0.042090  \n",
       "PF00003    -0.281317  0.121297  0.060522 -0.003963  \n",
       "PF00004     0.117087  0.175630 -0.008259  0.100392  \n",
       "PF00005    -0.044671 -0.007429 -0.281552  0.301284  \n",
       "\n",
       "[5 rows x 72 columns]"
      ]
     },
     "execution_count": 244,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features = d_norm.join(vecs_norm)\n",
    "features.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 258,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tiny_aa</th>\n",
       "      <th>small_aa</th>\n",
       "      <th>aliphatic_aa</th>\n",
       "      <th>aromatic_aa</th>\n",
       "      <th>nonpolar_aa</th>\n",
       "      <th>polar_aa</th>\n",
       "      <th>charged_aa</th>\n",
       "      <th>basic_aa</th>\n",
       "      <th>acidic_aa</th>\n",
       "      <th>tiny_aa_pct</th>\n",
       "      <th>...</th>\n",
       "      <th>aromatic_aa_pct</th>\n",
       "      <th>nonpolar_aa_pct</th>\n",
       "      <th>polar_aa_pct</th>\n",
       "      <th>charged_aa_pct</th>\n",
       "      <th>basic_aa_pct</th>\n",
       "      <th>acidic_aa_pct</th>\n",
       "      <th>isoelectric_point</th>\n",
       "      <th>aromaticity</th>\n",
       "      <th>instability</th>\n",
       "      <th>hydropathy_average</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.048595</td>\n",
       "      <td>-0.042007</td>\n",
       "      <td>-0.050449</td>\n",
       "      <td>-0.008282</td>\n",
       "      <td>-0.050177</td>\n",
       "      <td>0.030505</td>\n",
       "      <td>0.025450</td>\n",
       "      <td>0.026857</td>\n",
       "      <td>0.019399</td>\n",
       "      <td>-0.004710</td>\n",
       "      <td>...</td>\n",
       "      <td>0.021739</td>\n",
       "      <td>-0.001363</td>\n",
       "      <td>0.054349</td>\n",
       "      <td>0.045024</td>\n",
       "      <td>0.042129</td>\n",
       "      <td>0.037035</td>\n",
       "      <td>0.027147</td>\n",
       "      <td>0.023018</td>\n",
       "      <td>0.053486</td>\n",
       "      <td>-0.043780</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.086828</td>\n",
       "      <td>0.060620</td>\n",
       "      <td>0.045160</td>\n",
       "      <td>0.050223</td>\n",
       "      <td>0.082644</td>\n",
       "      <td>-0.115014</td>\n",
       "      <td>-0.143930</td>\n",
       "      <td>-0.137842</td>\n",
       "      <td>-0.123507</td>\n",
       "      <td>0.062089</td>\n",
       "      <td>...</td>\n",
       "      <td>0.033827</td>\n",
       "      <td>0.064856</td>\n",
       "      <td>-0.112828</td>\n",
       "      <td>-0.124164</td>\n",
       "      <td>-0.109055</td>\n",
       "      <td>-0.110237</td>\n",
       "      <td>-0.024248</td>\n",
       "      <td>0.043367</td>\n",
       "      <td>-0.056273</td>\n",
       "      <td>0.150751</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.062608</td>\n",
       "      <td>0.052815</td>\n",
       "      <td>0.065462</td>\n",
       "      <td>0.047874</td>\n",
       "      <td>0.072316</td>\n",
       "      <td>-0.011477</td>\n",
       "      <td>-0.026893</td>\n",
       "      <td>-0.031868</td>\n",
       "      <td>-0.017070</td>\n",
       "      <td>-0.016182</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.013492</td>\n",
       "      <td>-0.013140</td>\n",
       "      <td>-0.083511</td>\n",
       "      <td>-0.084159</td>\n",
       "      <td>-0.085048</td>\n",
       "      <td>-0.062056</td>\n",
       "      <td>-0.052057</td>\n",
       "      <td>-0.003720</td>\n",
       "      <td>-0.060900</td>\n",
       "      <td>0.074446</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.013405</td>\n",
       "      <td>-0.002510</td>\n",
       "      <td>-0.002576</td>\n",
       "      <td>-0.075736</td>\n",
       "      <td>-0.029619</td>\n",
       "      <td>0.024228</td>\n",
       "      <td>0.010364</td>\n",
       "      <td>0.006797</td>\n",
       "      <td>0.011967</td>\n",
       "      <td>0.001421</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.057813</td>\n",
       "      <td>-0.022886</td>\n",
       "      <td>0.013940</td>\n",
       "      <td>0.004095</td>\n",
       "      <td>0.001375</td>\n",
       "      <td>0.006163</td>\n",
       "      <td>0.005149</td>\n",
       "      <td>-0.049322</td>\n",
       "      <td>0.007514</td>\n",
       "      <td>-0.023404</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.035642</td>\n",
       "      <td>-0.036055</td>\n",
       "      <td>-0.028057</td>\n",
       "      <td>0.037366</td>\n",
       "      <td>-0.014076</td>\n",
       "      <td>-0.027425</td>\n",
       "      <td>-0.018246</td>\n",
       "      <td>-0.006551</td>\n",
       "      <td>-0.026390</td>\n",
       "      <td>0.017637</td>\n",
       "      <td>...</td>\n",
       "      <td>0.065558</td>\n",
       "      <td>0.034741</td>\n",
       "      <td>0.030131</td>\n",
       "      <td>0.035416</td>\n",
       "      <td>0.043072</td>\n",
       "      <td>0.017830</td>\n",
       "      <td>0.040094</td>\n",
       "      <td>0.059084</td>\n",
       "      <td>0.031010</td>\n",
       "      <td>-0.019262</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    tiny_aa  small_aa  aliphatic_aa  aromatic_aa  nonpolar_aa  polar_aa  \\\n",
       "0 -0.048595 -0.042007     -0.050449    -0.008282    -0.050177  0.030505   \n",
       "1  0.086828  0.060620      0.045160     0.050223     0.082644 -0.115014   \n",
       "2  0.062608  0.052815      0.065462     0.047874     0.072316 -0.011477   \n",
       "3  0.013405 -0.002510     -0.002576    -0.075736    -0.029619  0.024228   \n",
       "4 -0.035642 -0.036055     -0.028057     0.037366    -0.014076 -0.027425   \n",
       "\n",
       "   charged_aa  basic_aa  acidic_aa  tiny_aa_pct         ...          \\\n",
       "0    0.025450  0.026857   0.019399    -0.004710         ...           \n",
       "1   -0.143930 -0.137842  -0.123507     0.062089         ...           \n",
       "2   -0.026893 -0.031868  -0.017070    -0.016182         ...           \n",
       "3    0.010364  0.006797   0.011967     0.001421         ...           \n",
       "4   -0.018246 -0.006551  -0.026390     0.017637         ...           \n",
       "\n",
       "   aromatic_aa_pct  nonpolar_aa_pct  polar_aa_pct  charged_aa_pct  \\\n",
       "0         0.021739        -0.001363      0.054349        0.045024   \n",
       "1         0.033827         0.064856     -0.112828       -0.124164   \n",
       "2        -0.013492        -0.013140     -0.083511       -0.084159   \n",
       "3        -0.057813        -0.022886      0.013940        0.004095   \n",
       "4         0.065558         0.034741      0.030131        0.035416   \n",
       "\n",
       "   basic_aa_pct  acidic_aa_pct  isoelectric_point  aromaticity  instability  \\\n",
       "0      0.042129       0.037035           0.027147     0.023018     0.053486   \n",
       "1     -0.109055      -0.110237          -0.024248     0.043367    -0.056273   \n",
       "2     -0.085048      -0.062056          -0.052057    -0.003720    -0.060900   \n",
       "3      0.001375       0.006163           0.005149    -0.049322     0.007514   \n",
       "4      0.043072       0.017830           0.040094     0.059084     0.031010   \n",
       "\n",
       "   hydropathy_average  \n",
       "0           -0.043780  \n",
       "1            0.150751  \n",
       "2            0.074446  \n",
       "3           -0.023404  \n",
       "4           -0.019262  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "execution_count": 258,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr = features.corr()[d_norm.columns].loc[vecs_norm.columns]\n",
    "corr.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 248,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 257,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1376ab7f0>"
      ]
     },
     "execution_count": 257,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAy4AAAOvCAYAAAAqX5DWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzsnXd4VGX6sO+ZyUwmvZJkEgIkoSggJfQOCSW4ll0VXBdXRSkK0oNgAbsIspZVROz72291BXRX112lC2KhBTCokEIJkEYmvU3//jiTSWYySSbEcKK893XlgmTeOfc8b5vznPIehc1mQyAQCAQCgUAgEAg6Mkq5P4BAIBAIBAKBQCAQtIRIXAQCgUAgEAgEAkGHRyQuAoFAIBAIBAKBoMMjEheBQCAQCAQCgUDQ4RGJi0AgEAgEAoFAIOjwiMRFIBAIBAKBQCAQdHhE4iIQCAQCgUAgEAg6PCJxEQgEAoFAIBAIBB0ekbgIBAKBQCAQCASCDo9Xewv2RU2ztbfDHQpk0QJgQyGL11tpkcVrsKpk8QIEehtkc19t2Gzy9GsAs0WeYywKhYzziEz17aWyyuKVq40BfLxNsnjlHFNyYZNvSGGVqb7lame1lzz7BAAl1VrZ3GPyt/4qBpap6LSMo6Fl1OHxHbIexRkXgUAgEAgEAoFA0OERiYtAIBAIBAKBQCDo8IjERSAQCAQCgUAgEHR42v0el+ZIeGYmocmJWGoMZCzaQGX6mUZl/PvF0+uV+Si1Gop3pZH92HsAdE2dRtSMiZj05QCcWfMBJbuOeuyOf+ZeQpMHYq0xcmrRa1Q14e7pcB/l9GPvOr0e88BNxD9+F9/1nom5uKJDx9z16fsITkrEWmMge8lrVKefblTG97p4El5egFKroXR3GudWvQNA6A0jiFl2Oz49OvPj9Suo+iHbI6ecMQeMSyTm8VkoVCr0/9xO4caPnV5XaLzo8uISfK/rjrmknHMPvoDxQiG+/XsQu2a+vZCC/Jc/pGzb962KVy63fN6BdH5iNgqVEv0/d1DwemNv15eW4HtdAuaSCs7Ol7wBY/oTvfIuFGovbCYzF599n8pv0z321tHlqfsIShqEtcbAmSWvUn3Cfd+Oe2khSq2Gst1HyFkt9W1VsD8JG5fhHRuB4Xwh2fevx1JW1aIzcPxAujw5C1RKij7cQf6GTxrFHPfyYnz7STGffmA9xguFAETNv5XwOyaCxUrO6rco33vsisYccsNIYpbejrZHZ3763UNUezier8Z29h+biG71HFAqKdm8naI3tjaKufP6pWj7dsdSWsH5BWsxXSxEHRNBjx0bMZy+CEDNsVPkPrahVfH6j00k+vHZkvujHVxy5/7LUnz6JmAprSDnwXUOd8+drzvc1UdPkfvY6+3urUMd3Yke2zdQ+MqHFL31L8+94xKJWT0bVEqKP9rBpY2NvbEv1nvPPbgO0wVnb88dGyh4uXVeuDrnTjn7tlz7Xx0Oq3z3IP2ake2MS0jyQHzidRwasYDM1E10Xzvbbbnua2eTkbqJQyMW4BOvIyRpgOO1i29+TtrE5aRNXN6qpKXOfXjEAjJT36D72jlNujNTN3HY4R7oeE0THUbI2H7UXrjUau+VjjkoKRFtnI7jo+Zz5qE3iFvjPt645+dy5qGNHB81H22cjqAJUrzVJ3PInLWOiu9/8jjWOmSJWamk89NzOX33k5ycOJ+Qm8bi3SPWqUjo7ZOwlFXy87i5XHrnM3Qr7wag5tQ5Tt24lFPXLyb77ifo/Nw8ULVimMjlltEb+8xcsu9+kp+THyTkpjFoXbxhdu9PY++n8O3PiH5Y8pqLy8m+91lOTl7EuSWv0PXlJZ45GxCUlIh3XDTpo+dxdsVGuq6Z67Zc1zX3c3bFRtJHz8M7LpqgCYkA6ObfQvn+dNJHz6d8fzq6+bd4FHOXZ+aS8een+HHCAkJvHoO2R2enIuF/nIS5rJITox+g4K3P6PzIXQBoe3Qm9ObR/Ji0gIw7n6TLs/eDsnXTcFtjrjmZQ9bsta0bz1dpO0c/+QBnZz5O1pR5BN04Du/uzjGHTJ+MpbyKzKQ56N/9lKgV9zheM57LJ/uGhWTfsLDVO3YolUQ/dT9n7nmCzMnzCbpprHt3WSUZE+ZS9M6nRK10dmf9bhFZv1vUqqSlrV4A3WOzqNx7pNXxxti9GZPmE+zGG2r3nho/l0vvfIrOxRu9ahYVX7XSa3dfdXOnjH1brv0vwW+HFkeYQqG4RqFQrFAoFH9VKBSv2P9/bVvF4VOGULB5LwAVaZl4BfqhiQh2KqOJCMbL34eKIxkAFGzeS3jK0LaqCZsyhMLNXzVw+6J2casjglH5+zrchZu/IixliOP1hKfu4czTf2/VEilyxRwyZShFW78CoDItA1WQH+qIEKcy6ogQVAE+VNq9RVu/IiRlGAC1WRepzc69LLccMfsO6IHhbB7G8wXYTGZK/vM1QZOGOZUJmjSM4o93A1D6v28IGNUfAFutESzSqkpKbw2tXZxOLre83nyMOQ28k53bLmjyMPRbG3r7AVDz4xnMBcUA1GbkoPRWo9C07iRw8JSh6LfuAaCqhb5ddeQUAPqtewi296/gKUPRb5Her9+yh+AU5zpzh19dXdtjLv50P8GTnd8XPLl+uyX//ZaA0f3sfx9G8af7sRnNGM8XYjibh9+AHlc05tqsC60ez1djO/v074nhXB4m+5gq+3wfAZOGO5UJmDicko93AVD2xX78RvZvVVxN4du/B8aG7v/sI9BlPAdOGkapw/0N/r+Au63ewEnDMZ7PpzYjp3XeAZK3bv4q/c8+Al3GVODkYfV1/T8X7+ThGHPyMWS2zlvnvtrmTjn7tlz7X4LfDs0mLgqFYgXwT0ABHAQO2f//oUKhWNkWsUYXiiFX7/jdkKdHowttXCav6TLR96aQuHs9PV96AK8gv1a4w5zcxrxivHVhTmW8dWEu7mI09jKhkwdjyCum6qdzHjsd8cgQsyYqFENukeN3Y64eTVRoozLGBl53ZS4HOWJWR4VhyquP15RXhDoqrHGZujqxWLFUVKEKCQDAd0BPeu14jV7b/sqFR193fDF5glxuubyaqDCMDftWnh51pKs3tElvHcHXj6TmxzPYjGaPvM7++r5jytOjdum3ate+nadHY68bdXgwpsIS6b2FJajDglp26kIxNqhrY76bPh3VoIzFiqW8Gq+QAI/e26K/jTFfDldjO0tjqv6IrjmvqHHMkQ3KWKxYK6pRhQRKnzk2koT/vELch2vwHdKnFdGCl+t4ztc3Hs+RYc59rKLKyd3985eJ++cafIf0viJehY83ne6/lcJXPmxNqI5tmnIbzl9u+leku/mr3ltwGV64OudOOfu2XPtfHRKbtWP/dFBaOuNyHzDEZrM9b7PZ/p/953lgqP01tygUijkKheKwQqE4/Fl14+uQ7YUa/801eXZbRiqU+/52Dg5bQFrycowFpcQ/cVcLoTS/2UaZexNllD4aYhffyrl1H3nsa1Z8JWJ2s01bo3jdBuzZ9lvpbv+Ym96eJ5+r+lgGpyY9SMZNy4iYdxsKb7UHTrndMnk9GkvN9wFtz1iiH76LnIdbcUlLK/yKZvrX5eFuPLkWaaqMB+9tvb79Y74q27llZ1Ofy3ypmFOjZ5J94yLynn2b2JdSUfr7eO653Dnb7j456l6yblhM3jNvE/tyK9xt8EYumUHRu59ira71zOXBNj0pE7VkBkXvXKZX2vBluqV/frVzZ4tu92Xa2rdl2/8S/GZo6dyiFYgGXFNbnf01t9hstjeBN8H5AZS6mVPQzZgIQMWxLLyj67Nsb10Yxvxip+0Yc/VOmbhUxn7UrKjM8fe8f+yk79+bPwGkm5lC1Ixkuzvbya3RhWJwcRsauUMx5hej7RqFtksEibvXOz7TwO3rODb1YUyXSt145Yk58p4UOs2YBEDVsSy8o8OprIs3OgxTQYmzN0/vOKJRV6bO21rkbGcAU34Ral2443e1LhxTgbPTlFeEOjocU74eVEpUAX5YSp1v8DNkXcBaU4u2Z1dq0rNa9MrplstrzNOjia73anRhmApdvfomveqoMOLefJhzS17GeC6/RR9AxN1Tnfq2pkH/Uus86Nu6MIz2ujEVlaKOCJGOwkeEYNKX0RLS9hrEHBWGybVP28uY8uwxB/piKa3w6L3tHfPlcDW2sylfj1rXyfG7ly68ccz2MmZ7zMoAX0fMFqP0b+2JbIw5+WjiYqj1cB4x57mM56gwx6VB9e4iNLpwh7thfbu6veNiPBrPbfH6DuhJ0NSRRK28B1WgHzarDZvBiP7//tui15QvzU0Or7v+le9+/vId0JOg60eie7j1Xsd2r5K5sz7mK9u35dr/Evw2aemMy2Jgl0Kh+EKhULxp//kS2AUsaq0s771tjpus9V8eInL6OAACEntgrqjGWOjc8YyFpViqaghIlK4Bj5w+jqJthwCc7pMInzqUqpPnW3B/ydGJyzk6cTn6Lw8SMX28w22pqMbk4ja5uCOmj0e/7RDVJ3M40Pc+Dg2Zx6Eh8zDk6Tk6+aEmB41cMRe8/yUnJi3jxKRllHx5kPDbpHj9E3tiKa92XDZRH28Jlspa/BN7Stu/bTwl2w42uf3mkLOdAaqPZ+IdF40mNhKF2ouQG8dQvuOAU5nynQcJvTUJgODrR1Hx7Q+SLzbScXOlOqYT2vgYjBcKPI5dLre8Xh2a2AiHt2yHc78p23GQsNsae1WBfiS8v4rctX+n6vBJj3wAhX/7gh8nL+XHyUsp2XaAsNsmAODXTN+2VtbgZ+/bYbdNoNTet0u3HyJsmvT+sGn1f2+OquOZaBvEHHrzaEpdYi7dcdCx3ZDfjaTim3TH30NvHo1C44UmNgJtnI6qY5lXNObL4Wps55ofMvDuFo26szSmgm4YS8VO5zFVsesAIbdKO2RBU0dT9Z095tBAx6IL6thINN2iMeV4vnNZ/UOms/vGsZTvdP7M5TsPEOxwj6KyCbd3t2iMHrrb4j09fSWnxszi1JhZFL37GZde3+Jx8lB9PBNNA2/wjWMpd+lf5Tsa1PX1o6i096/s6Ss5OXoWJ0dL3sINnnvr3FfL3FnHle7bcu1/dXis1o7900FRNDoN7FpAoVAiXRoWg3QC7wJwyGazebSOW8MzLq50X3MfIRMGSEviLd5A5XHpsrLEnS+QNnE5AP79GyyTu/sY2Y9Iy1v2enUB/n27YbPZMJy/RObyTU47xIoWLnNKWDPL7jaQsfh1Ko9LS4IO3PkCRx3uBMdyfCW7jzrcDRly6HWOTlnhtByfze15zvaP2VvZdJN0e242QeMHYq0xcHrJa44ljfvu+AsnJi0DwK9fAvF1yyHvSePco28DEJIyjG7PzMIrLBBLeRVVP57h1J+edmzbYFU16W3vmAO9DW6dARMGEbN6FgqVkuLNOyl4bQtRS/9E9Q9ZlO88iMJbTdeXluLTJx5zaYW0vOX5AkL+MJ6IebeByYzNZqPglX9Stv2AW0dTyOVub6/N5r5fB04YRMzj90lLen60q96bnkX5Drv35SX42r1nH1yPMaeAyAXTiJx/G4Yz9TeKZ9/5BGY3R8PNlqaPsXR5do6jb59Z+qpjed8+21/kx8lLAfDtl1C/TO6eNHIeewsAVUgA3d9IRRMTjvFiEVlzX8BSWunYtkLhfh4JShpE7BP3glKF/qOd5L26lejUO6g6nkXZjkMovNXEvbIY377xWEoryJ73F4w50g6NbsFthN0+ESwWcp54h/I9aW4dTdV3W2MOThlG12dm4RUahKW8iuofz5Ax4ynHtr1U7r+w2rudm2vjtsbcUjv7eJvcOv3HD0a3ajYKpZKSLTu49PpmIhbPoCY9k4pdB1Fo1HR+cRna3vFYyio5v3AtpvMFBKaMJGLxDGwWK1gsFL78ARW7GydLzbVxwPhB6FbblyXespNLGzYTscTu3im5Y19a6nDnLFjncEcumYHNYgGLlYKX/0HFrkPN1u0v4W1IxKI7sFbXul2WuKldjoDxg4i2L4dcsnknhRs2E2n31s1fsS9K85elVPIaXbyRi+/AUuXeC2Btor5/q3On2qvpfYL27tsl1dom3e25/wUwJn9r0wOrA2HK+7lDry6g1l3bIeuxxcSlrTSXuLQnLSUu7UlziUt70lzi0p60lLi0J00lLoJfnuZ2stqblnZq24umEpcrgVz13VTi0t7I1cbQdOLS3sg5puRCzoWgmkpc2hu52rm5xKW9aS5xaW9E4vLL0FETF/m+KQQCgUAgEAgEAoHAQ1q3mL5AIBAIBAKBQCBoE7YOvORwR0accREIBAKBQCAQCAQdnnY/4xIVUtFyoXbAbJTvvguL9eq6Hj9YI991tFklwS0XagcMbhejvzLIdbRBJeN9Ywl+5bJ45RrLcPU9FFor030mAKfLW34gZXtglfEKcpVMHUyue0ABQpRGWbwBWnm8RVWteHbQL0xCl8tffl0gaA5xqZhAIBAIBAKBQHAl6cBLDndkxKViAoFAIBAIBAKBoMMjEheBQCAQCAQCgUDQ4ZHtUjHf0YOIeOQBUCop2/olJW9vdnpdoVYTtTYV7949sJSWk7d0DebcAlB7EfnEQrR9e4DVRuFzb1Bz6AePvX5jBxG1ag4KlZKSj7aj37TF2avxInr9Mnz6dsdSUsGFhc9julgIgHevbuieeRClvy/YbJz5/WJsRs+vy/Yfm0j04/YHe320g0tvbG3k7vyXpfj0TcBSWkHOg+swXSxEHRNBz52vYzh9EYDqo6fIfez1Vnl1q+dI3s3bKXLnXb8Ubd/uWEorOL9grcPbY8dGh7fm2ClyH9vgsRekdo589H5HOxe/5VLfajVRa5eh7SO1c+7SNZgvFoKXiqhnFqPtnQAqFeWf7qL4zc1NWNzT+9m7iUgegKXGyPGFGylPP9uoTK+HpxMzbSzqYD+2xc90/D1u7vXEzpiAzWLFqC/nh8WbqLlQ5LG73zN3EWV3H1n0BqVu3L1XTqfLtDFogv34LOHeRq9H3zCU4W8vZveURyk9fsYj73XP3EWk3Zu26A3K3HivXTmdWLv38ya8Q99ezFet8PZ55m6H99iijW6916ycTmd7XX+RUF/X8XOvp8uMCdjMVgz6co4v8byu5RrPlzuW61BHd6LH9g0UvvJhkw/La9I9LpEY+4P6ij/awaWNjd2xL9a7zz24DtMFZ3fPHRsoeLl17sv1+vTvQec1D9oLKSh4+QPKt33fqpj9xgwi8rG5KFRKSjdvQ/+mm3Zel+qYwy4uWuPUzlFPL0Dl74vNauPsLYtaNW+3ZR4JHX4NvZ++i4DeXTg696/kf974QX3NIde46v3s3XRKHoilxsAPTcTc8+HbHTFvj7/H8feQ4dfQ++m7CejdhWNz/0r+554/RFfOuo57+l6CkxOx1hjJWvwqVemN5z6/fvF0f/lB6SHNu9I4s+pdALquuouQyYOxGc3Unssna/FrWMqrW3TKOY8AdH92JmHJiVhqDJxcuIFKNzH794vnmr/OR6XVoN+VRtaj7zlei7kvhZh7p2IzW9DvTOP00/+vRad2xBCCl80HpZKqT/9Hxd/+6fS698DrCF46H3X3ePSPPkPN7n2O1zp/vx1TtvQZLfmFFC1b1eqYOxRiVbHLQp4zLkolEavmc3HOY5y9cQ6BvxuPJqGLU5HA26ZgKavkbMq9lP7fv+iUKu1gBU2bCsC5mx/gwn0P02nFbPD0RmmlEt0TD5Bz7+NkTXmAoBvHouke61QkeJrkzUqajf69fxOxwj4xqpTEvJhK3qoNnJ46j3N/WonN3Iqb0pVKop+6nzP3PEHm5PkE3TQWbxd3yPTJWMoqyZgwl6J3PiVq5T2O14zn8sn63SKyfreoVUkLSiXRTz7A2ZmPkzVlHkE3jnPvLa8iM2kO+nc/JWqFszf7hoVk37Cw1UkLSiWRq+dzYfYqztwwlwA37Rx022Ss5ZWcmXIfJX/7N52WSe0ckDIGhVrN2Zvmce7WhQTffj1eMREeqzslD8AvLoqvhi8hPfUt+q67z225gu1pfJPyWKO/l584y/4pj/L1hBXk/ecA16z+k8fuyOQB+MdHsX3EUtJS32bA2sbJAUDe9jS+mup+4vXy09L9vikUH8lstXfniKUcS32b/k1487ensbcZb3wrvRF27+4RSzie+hbXrXVf1/nb0/h6auO6Ljtxlq+nPMrepBXkfX6Aa1d5WNdyjec2jmUA3WOzqNx7xDOfizvG7s6YNJ9gN+5Qu/vU+LlceudTdC7u6FWzqPiqle42eGtP5ZB54xIyr1/Embsep/Oz80HViq8epZKoJ+ZxftZqsqfeT+AN4xq3821TsJRXkj1xFsXv/YuI5fa+r1ISvX45+atf4/T1D5Bz54pWzdttnUdqLhZxfNEb5H7yjefx2pFrXHVKHoBvnI69wxdzIvUt+q6b5bZc4fYjfJvyaKO/117U88Oija2OWc66Dk5KRBuv4+jIB8levpH45+e4LRf//Byyl7/B0ZEPoo3XEZw0EIDSfcc5Nn4xx5OXUpudS+cFt7QslXMeAUKTB+ITp+PA8AVkpG6i57rZbsv1XDebjNRNHBi+AJ84HaFJAwAIHtWH8JQhHJqwjEPjlnJ+42cexRzy0EIuLXqY/On34js5Ca+4rk5FzPmFFD+5juptuxq93WYwUjBjLgUz5v76kxbBZSNL4qLt1wtTTh6mC/lgMlP+v734JY1wKuOfNILyT3cCULHta3yHS4PFO6EL1d8fA8BSXIa1vFI6++IBPv17YjyXi+m85C37fB8BE4c7lQmYOIyyT6QBU/7FfvxG9Jc+z5hEak+exXDSnu2XVrTqxirf/j0wnsvDdL4Am8lM2X/2EThpmFOZwEnDKP1Ycpd98Q3+I/t7vP2m8OnfE0ND7+f7CJjkGvNwShze/fj9Al4Abb+emHJyHe1c8b+9+Cc7u/2TR1D27wbtPEJqZ2w2lL5aUClRaDXYTCaslS0fwaojMmUQF7d8DUDpkSzUgb54RzRegaz0SBaGwtJGf9d/8xPWGqOjjFYX6rE7esogcjZL7pI0ya114y5Jy6LWjRug94ppZLz+ORaD50eGo9x43cVckuY+ZoBrV0wj8/XPsbbSe97uLW3GW9qEV//NT1jsdV1yJAsfD+tarvHc1rEcOGk4xvP51GbkeORzcg+Q3Ea7u/Q/+wic7OKePKx+PP/PxT15OMacfAyZrXO3xWurNYBFqluFtwZbK1ez8unn3M7l/91HQLLL98XE4ZR9Is0j5V/ux9fezn6jEzGcOnPZ83Zb55Ga80VU/JSDzdr6FbzkGleRKYO5uEU6yl16JAuvVsd8iYqfcqCVMctZ16EpQ7i0ZS8AlWmZeAX6oXZxqyOCUQX4UnkkA4BLW/YSmjIUgLK9xx19vCItA010WItOOecRgPCUIRTYYy4/IsWscYlZExGMl78P5YelmAu27CV8qhRz9N2TyXn139iMZgBMRS2v/Kjpcw2m8xexXMwDs5nqHXvwGTfSqYwlrwBT1umrb1lFgcfIkrh4RYRhzr/k+N1cUIQ60nmge0WGYc6zl7FYsVRUoQwOxHDyNP5JI0ClxCsmEu8+PfCK6uSZNzIMU179qXJzvhtvVBimBl5rRTWqkEA03WLAZqPLe08R9+krhM25tXUxRzm7Tfl61FHObnVkGMa6MvaYVSGBAGhiI+n++cvE/XMNvkN6e+xVN4wHMOc1jlkd6T7mOm/Cf14h7sM1+A7p47EXwCsy3NmdX4SXa31HhGFuELO1ohpVcCAV2/Zjra4l4esPSNj9fxS/+wnWskqP3VpdKDUX9Y7fa/OKW5V8NCT2T+O5tPt4K9wh1OTWLwVZk1eMVhfi8fuD+nbFJzqM/B1HW/U5fVy8tXnF+FyGt6CVXq0ulNrc+rquaUNdd/nTeAo9rGu5xnNbxrLCx5tO999K4Ssfeuxz3a4pt4E7T+9+POc27S64DHdbvAA+A3rSc/sGem57lYuPve7YyfMEr6gGcwRgcjePRIZhym/QzpX2do6LARvEvvs0cf/+K6Gzb2tN2L/oPNJa5BpXWl0otTLELGdda6JCMTTo34Y8PRqdcx/T6MIwNmgPQ54eTVTjzxfxx2RKdrc8h8o5jwB460IxXHSOx9ulvr11oRjyGpTJrS/jmxBN0LBrSfziOQb860kCBiS06FR1CsdSUL9PYCm4hKpTuMefWaHREPm314l491V8xo3y+H0dFqulY/90UC77HheFQjHTZrO913JJt29u9KdGR+HcXv5lo+yTbWgSYumy5VXMuYXUHvsJm8XDCvbokjI3ZWw2FF4qfAf35swflmCtMdD1789SeyKLqm893KG93JhtNsyXijk56l4spRVo+ybQddOjZE6Zj7WyxjO3m206e92XMV8q5tTomfXeNx4jM2Xe5XuBRo8CcVcv2PC5rhdYrWSPnYEq0J8u/1hP9bdHpbM3HqBwE1Rrj/QCxNw6mqAB8Xz/+6c8fo/CbTt6/Gb6PfVnjix6w2Nfw/c20rbCe91TfybtF/JeztGymFtHE9w/nm//4GFdyzWe2zCWI5fMoOjdT7FW13rw2T1zNx7P7stELZlB0TuX6W6DF6DmWAYZk+fjndCZ2L8soeKrI9g8PqvXstv9mLOhUKnwGdSbs7cuxlpjoMv/PUftiUyqv/Ns3v6l5pHLQq5x5YYrEbOcdd1U/3Eq4+6NLh8vZtGt2CwWij7e5660q9SN8grNI9LG3W3a48+o8FLiFexH2tRHCBjYnd5vLeXAkPmtVbaqT+feeAfWIj2qGB0Rr6/HmHVaOnsjuKpoy835TwJuExeFQjEHmAPwVFRvbg92vm7TXFDkdJbEKzIcc6Hzw4rM+UV46TphLigClRJVgB/WUulhlpeef9NRLvaDFzGdy/XoA5vzi1Dr6rN7r6hwTAV6N2U6Yc7Xg0qJMsAXS2kFpvwiqg+ewFIinQ6t3HsYbZ8EjxMXc56zWx0VhrnAOWZTfhEaXbjDrQrwky5tACxG6d/aE9kYc/LxjouhJj2rRa8pX49a16CudeGYCl29ercxu/Nq4mKo9cAL9jNpDd1R4ZgL9Y3KeOnCHe2sDPDFWlpBwA3jqfr6MJgtWIrLqElOWcK/AAAgAElEQVT7CW3fHs0mLl1nTiL2ziQAyo6dxicmjBL7a1pdKIb8kibf646wsX3pvvj3fPeHp7DaT4c3RfzMSXSbMQGAkmOn8YmuP3Llowul1kO3l7+WwF6xjPlEun5X2ymIEX9L5bu717u9UT6uGa+2ld6AXrGMtnu9OwUx7G+pHGjC223mJLrMkOq69NhptA0ujWhNvHWEj+lLj0W/59tbWq7rOuQaz20Zy74DehI0dSRRK+9BFeiHzWrDZjCi/7//ehSzKb8IdXQDty7MzXiWypjcua8fie7h1rvb4m2IIfsC1ppatD27ejR/Qd13QcP6bvx9YcovQh3VoJ39pXY25xdRfSjd0c5Vew+j7dO92cTll55HWoNc46rrzMmOmEuPZaONqfe2Z8xy1nXUPSlEzpgIQOXxLLyjw6nrrd66MIz5zn3MkKd3ugTMWxeGscG47zRtPKETB/Hj9Cc88ssxj0TPnEL0nVLM5cey8I5xicc15lw93g3OPHlHh2G0t4kht5ii/0oLL1QczQKrFXVYICZ905eMWQqLUEXW7xOoIjthKdI3Wd4Vq72s5WIehrTjaHr1oEYkLlcdzV4qplAofmjiJx2IbOp9NpvtTZvNNthmsw12TVoAatNPoe4ajVdMJKi9CLx+HFV7nFeZqdzzPYE3SwMsYMoYqr+XvmgUWm8UPt4A+I4ciM1iwZjt2TWeNT9koOkWg7qz5A26YSyVu5xXPKnYdYCgW5IBCJw6mqrvpBXLKvel4d2rGwqtN6iU+A69DkPmeY+8ANU/ZOLdLRp150gUai+CbhxL+U7nVU/Kdx4g+FbJHTR1FJV2tyo0EJRSU6ljI/HuFo0xx7MzDzU/ZDh7bxhLxc7GMYc4vPUxu3o13aIxeegFqE3PQN01GrW9nQOuH0flbpd23v09Qb9v3M7mvEv4Dpeu51X4eKPtfw3G083X97n3drA/+WH2Jz9MwReHiZk2BoDgQd0xV1Q3eV+HOwL7duO6F2Zx6K71GD24dvf0ezvYPfERdk98hLwvD9NluuQOSeyOqaKmyXtZXDFX1PDfPnPZNmQR24Ysojgtq8mkBeDMezvYM/ER9rjxmitqPI7ZXFHDF33msn3IIrYPWURJWlaTSQvA2fd2sG/iw+yb+DD5Xx4m1u4NTuyO6TLqut8Lszh0t2d1XYdc47ktY/n09JWcGjOLU2NmUfTuZ1x6fYvHSQtA9fFMNA3cwTeOpXyHi3tHg/F8/Sgqv5Xc2dNXcnL0LE6OltyFGzx3t8Wr7hzpuBlfHdMJ7/gYjA1WOWuJmvQMhxu1F4G/G0vFLpd5ZNcBgm6R5pHAlNFUf29v56/T0PaKq2/nIX0xZjX/ffFLziOtRa5xde697exPXsn+5JX2mMdK3naOWc66zn//S45PSuX4pFSKvzhIp2njAPBP7IG5ohqTi9tUWIq1sgb/ROme2k7TxlH85SHp804YQMyDv+fne5533BvZEnLMI7nvbeNw8nIOJy+n6ItDRNpjDhwkxWx0idlYWIq5sobAQVLMkdPGUWSPueiLgwSPvg4An3gdCrVXs0kLgPGnk6i7xKCKjgIvL3wnTaBm37ctfm4ARYA/qNUAKIMC0fTrg+nMOY/e22GxWTv2TwdF0dypWIVCUQBMAVwPeyiAb202W3RLgoxrU9wK/MYOodPDc0GppPyT7RRv+idhC/5M7YlMqvZ8j0KjJmrtQ3hfm4C1rIK8ZWswXcjHKzqSzm8/i81qxVyop+CxlzDnNv4SNBtVbj+P//jBRD42B4VSSenWHRS9/hGdFt9JTXomlbsOoNCoiflLKto+8VhKK7iwaJ10UygQdPMEwu6fBtio/OowhWvdXylnsbrPBwPGD0K32r704ZadXNqwmYglM6hJz6Ri50EUGjWxLy1F2zseS1klOQvWYTpfQGDKSCKXzJAuibNYKXj5H1TsOtRo+wqF+7b0Hz8Y3arZKJRKSrbs4NLrm4lYbPfukrydX1zm8J5fuNbhjVg8A5vFChYLhS9/QMXuxktMqjVNX6rnN3YIEY/MAaWKso8btnMGVXuk+tatW473tQlYyirIW/o8pgv5KHy16J5bKq1CplBQ9sl2St79uNH2s0oa37xZR581M+mU1F9a0nPRJsqOnwZg9K417E9+GIBrVv2J6FtGoo0KoTa/hPP/2EPm+o8ZtuURAq7tQm2B1PVrL+o5fNd6x7YNLVym1H/NPUROkNxHFm9yJABJO59j98RHAOi76g5i/1DvPvvBV/y83jnGMZ88RvqT/3BKIJo72tDP7jXXGDjawDth53PssXv7rLqDzg285z74ipMu3tGfPMYJF6+qmevd+q6ZSYQ93mOL6+t67M417Jso1fW1q/5ETANvzgd7yFj/McM3P0Jgg7quuajn0N3rnbaf4Of+C7G9x/MvPZYbErHoDqzVtU0uY9rU1BwwfhDR9mWJSzbvpHDDZiLt7vKdB1F4q6VlifvEYymV3EYXd+TiO7BUNe3+Jb3Bf5hAxAO3YTObwWqj4K//pHx74+WQ1epm5pFxg4l81L4c8tbt6Dd+RPiiO6lNz6Ryt9TO0etT0faWlo29uGSto50Db5pA+P3TsdlsVO09TOG6dxtt/3R5UJPutswjQQPiGfTeUtTBflhrTRgKy9g3brlj29YWrnZsz3Glaua7v8+amYQnDcBaY+CHRW80iPl59ievBKDXqj8Rfcsoh/fCP/aQuX4rQQPiSXxvWYOYS/m6Qcw29xddtXtdA4Qom04q4p6bRcgEaQnorCUbqDqeDUD/Hes5PikVAL/+CfSwL4dcsvsoZx59G4CB376GUqPGXCKds6lIy+D0ivorQwK07r3tPY/oq7VNxgvQY819hCZJy0+fWrSBCnt9D971AoeTpboL6C8th6zUaijedYzMR94BQKH24pqXH8C/bzesRjPZT/6d0v0nHNtO6FLcWAhoRw4leOl8FCollZ99QcV7HxA49x6MP5+idt93aHr3ImzdkygD/bEZTFiLi8m//T40/XoT8vASadEHpYLKDz+h6rMv3DpiD+3ycKlZeTGePdyhVyDQdBvcIeuxpcTlHeA9m822381rH9hsthbXV2wqcWlvmkpcrgRN7ey0N00lLu1Nc4lLe9Nc4tKetJS4tCdyPTW2ucSlvWkqcWlv5BrLcPUtqtNc4tLeNJe4tCctJS7tSXOJS3vSXOLS3jSXuLQnTSUu7U1LiUt70lTiciUQicsvQ0dNXJq9x8Vms7lfSF16zfMHWwgEAoFAIBAIBAKJVizNLqhHvsOJAoFAIBAIBAKBQOAhInERCAQCgUAgEAgEHZ62LIfsERazPLmR1SbfpXleXvJcq22xyFPXNhkv1A5XGWTxWmTsXzaZ3CqlfKe1y2u8ZfHKOY+oZLpnTS5qjGrZ3Bpk6ttXVxMDoEa+e5nMNnm+I2uN7b6r1eGoKpNnzhb89rn6RpNAIBAIBAKBQCAjtg685HBHRlwqJhAIBAKBQCAQCDo8InERCAQCgUAgEAgEHR7ZLhXzGzOIyMfsDxTbvA39m1ucXldovIhel4q2b3fpgWKL1mC6KD1o0rtXN6KeXoDK3xeb1cbZWxZhM5o88vqPTUS3eo70wKfN2yl6Y2sjb+f1Sx3e8wvWYrpYiDomgh47NmI4fRGAmmOnyH1swxWJOfCm8YTNutVRzrtXHGd+vxDDz6d/FTFH1D04bss2il1jVnuheyEVbR/JnbvYHvON4wl1ifnsHzyPGaDr0/cRnJSItcZA9pLXqE5v/F7f6+JJeHkBSq2G0t1pnFslPVwr9IYRxCy7HZ8enfnx+hVU/ZDdqrjjnr6X4ORErDVGsha/SlV64yfQ+/WLp7v9YWalu9I4s0p6MF7sQ38kdMpQsFox6cvIXPQapgLXZ8A24X3mXkKSB2KtMZK56LUmvT1ekR4oVrLrKGcek7xdHvojoSlDsFmtmIrKyVr0GkYPvSDVd0hSIpZm6tuvQX2XNKjvLqvuImTSYKxGM4ZzBWQveRVLeXWH9srVxnUxy9G35RxTcrl7PXsP4cnSgwl/XLiRCjftHNAvjj5/nYdKq6Fo11FOPfo+AP59unLtC7NReauxmS38vPIdyo92fHebvL27cu0Ls1D5aak9f4n0B17FUlnjkbf7szMJS5bG8smFG6h04/XvJz0QUaXVoN+VRtaj9Q+OjbkvhZh7p2IzW9DvTOP00//PIy9AwjMzCbW7MxY17e71St3DGNPIfkxyd02dRtSMiY4nx59Z8wElu4565O3y1H0EJQ3CWmPgzJJXqT7hvl/HvbQQpVZD2e4j5KyW+rUq2J+Ejcvwjo3AcL6Q7PvXYymr8jhmOepbzn2CDodYDvmykOeMi1JJ1BPzOD9rNdlT7yfwhnFousc6FQm+bQqW8kqyJ86i+L1/EbH8XukFlZLo9cvJX/0ap69/gJw7V2Aze3izn1JJ9JMPcHbm42RNmUfQjePwdvGGTJ+MpbyKzKQ56N/9lKgV9zheM57LJ/uGhWTfsLDVO/Btibn8s684c9MCzty0gNzUv2C6WOD5YJU55sjH53Fh9mpOX2+POcHZHTRtCpaySk5PmkXx+/+iU13M//mKszcv4OzNC8hb3sqYgaCkRLRxOo6Pms+Zh94gbs0ct+Xinp/LmYc2cnzUfLRxOoImDASg+mQOmbPWUfH9T62LGQhOSkQbr+PoyAfJXr6R+Ofdu+Ofn0P28jc4OvJBtPE6gpMkd+7rn3I8eSnHJ6VSvOMIsUuneeQNSR6IT7yOtBELyEp9g4S17r0Ja2eTnbqJtBEL8Gngvfj6pxxLWsbxicspaYW3LmafOB3H7PUd30x9n35oI8dGzccnTkewvb7L9h3n+ITFpE9cSu3pXGIW3Or2/R3JK0cbg3x9W84xJZc7PHkAvnFRfDN8ET+nvsW169w/2uzadbP4OfVNvhm+CN+4KMKSBgDQc/UMTq/fyvfJK8het5keq2Z0eHdbvb1fnEvWMx/w/fjlFP7vIN3m3+iRNzR5ID5xOg4MX0BG6iZ6rpvttlzPdbPJSN3EgeEL8InTEWr3Bo/qQ3jKEA5NWMahcUs5v/Ezj7xQP3ceGrGAzNRNdF/r3t19reQ+ZJ87Q+xugItvfk7axOWkTVzucdISlJSId1w06aPncXbFRrqumeu2XNc193N2xUbSR8/DOy6aoAmJAOjm30L5/nTSR8+nfH86uvm3eByzLPUt4z6B4LeDLImLT7+eGM/lYjqfDyYz5f/dR0DyCKcy/hOHU/bJTgDKv9yP74j+APiNTsRw6gyGk9KRAUtphcdZq0//nhjO5WE6X4DNZKbs830ETBruVCZg4nBKPt4FQNkX+/Eb2b9NsTrcbYi5IYE3jKP8P3s998oYs9ZNzP4TXWJOHk7Zv6SYK5qIOeCGcZR/7nnMACFThlK09SsAKtMyUAX5oY4IcSqjjghBFeBD5ZEMAIq2fkVIyjAAarMuUpud2ypnHaEpQ7i0Za/dnYlXoB/qiGAXdzCqAF+H+9KWvYSmDAVwOjqp8vX2eOWh0ClDKNz8VQOvr3uvvy8Vdm/h5q8ISxnSyKv09cbWiiWPQqYM5VIr6/tSg/ou23scLNI4rjiSgUYX1qG9crVxXcxy9G05x5Rc7k4pQ8jbsg+AsiNSO2tc2lkTEYyXvw9lhzMByNuyj4ip0piy2cArwAcAr0BfDK04qyaXu61ev+46Sr77GQD93nQifjfMI294yhAK7GOqvAVv+WGpjQu27CV8qjSmou+eTM6r/8ZmNANgKir3yAsQPmUIBZsld0Va8+66ubNg817C7eP5cgmeMhT91j0AVLXQr6uOnAJAv3UPwXZv8JSh6LdI79dv2UNwimd1DfLUt5z7BILfDi0mLgqF4hqFQpGsUCj8Xf6ecrlSr6gwzHlFjt9N+UV4RTrvMHhFhmHKvyT9YrFiraxGFRKIJi4GbBD77tPE/fuvhM6+zWOvOioMU94lx+/mvCLULl51ZIMyFivWCskLoImNJOE/rxD34Rp8h/RpTchtirkhgb8b26oBK2fM6sgwzPn1MZvz3bvNTbjrCLy+dTEDaKJCMeTWu425ejRRoY3KGPP0zZa5HFzdhjx9ox1ijS4MY67euUwDd5eVf2LQ4U10umUsOS/80zOvLgyD0zaL8XbxeuvCnGPOK3b6bF1W3sHgI2/Q6dYx5Kz7yCMv2OvyF6rviDuSKN2d1uG9crSxO/eV6tsdaUxdKbe3LoTai/XbrM3To9U5b1OrC6U2r7i+TG4x3jpp5zNj1d/osfpOxqRtoMfjfybr2Q87vLut3sqT5+mUMhiAyBuHo43x7GCAty4Uw0Xn8eLt4vXWhWJo0MaG3PoyvgnRBA27lsQvnmPAv54kYECCR14AjS7UZe7Uo3Fxa1zdLmWi700hcfd6er70AF5Bfp55o5znCFOeHrVLn1W79us8PZooqU7V4cGYCqWE1FRYgjosyCMvyFPfcu4TdEhs1o7900FpNnFRKBQLgU+BBcAJhUJxc4OXn7t8rZtnI9icDzkqFO7LKFQqfAb1JnfZC5z943ICJo1wm5F7jIvX3UfDZsN8qZhTo2eSfeMi8p59m9iXUlH6+7RCdPkx16Ht3wtrjQFD5rlWeN1wpWJuIZ6mytgaxtxPitnY2phb2G6Tn+8XeLBCS+0I7qu8oTrn+Q84Mngulz7Zh27mVA/F7rSetXW990MOD7qfSx9/je7eVhybuMy2dq3v6IW3YjNbKfpkX4f2ytbGktyN+gr0bRnHlHzuy/Tai3S+ZxIZq//G14nzyVj9N3q/dP+vwN0274+L3yB25mSGbV+Dl78PVvsR+cvzuhZp+rMpvJR4BfuRNvURsp/6O73fWuqh1/12G3WdZsZ87vvbOThsAWnJyzEWlBL/xF0eet38rZX7BZePDPUt5z6B4DdDSzfnzwYG2Wy2SoVC0Q3YqlAoutlstldo4nsZQKFQzAHmADzRqQ/Tg7o4vW7OL8JLF+74XR0Vjrmw2KmMKb8IdVQnzPl6UClR+vtiKa3AnF9E9aF0LCXSacmqvYfR9ulO9XfHWwzWlK9Hrevk+N1LF46pkVcq4/AGSF4Ai1H6t/ZENsacfDRxMdSmZ7XobWvMdUhnW77yyNcRYjblF+EVVR+zV5Q7dxFeuk6YC+rdVpeYK/7rWcyR96TQacYkAKqOZeEdHU6l/TVNdFijm5+NLkfJNdFhGPM9v5SjIVH3pBA5YyIAlccld10U3rowjPnOcRvy9Gii693eujCMBc5lAIr+tZ9r//4I59e7P/sRNTOFyBnJkvdYNt7RYQ28oY29uS4xuykjeb/m2v/3COdf2NxkzJH3pBBhr+/KY1loouvbWhMd1ujG/pbqO3zaeEImDubn2x9v0imnV642rotZjr4t55iSy9155mQ63ymNqbJj2U5nDLS6MAwu26zNdT4joY0OdZTRTR/nuGm94LPv6f2i+3sY5Hb/kt7qrFzSbpeOa/rG6wifNLBJb/TMKUTfKY2p8mNZeMe4jBc381fDs8jeDdrYkFtM0X8PAFBxNAusVtRhgY4b5l3RzZyCzj6eK45l4e06Vl3cRle3rt5tKipz/D3vHzvp+/eVTcYccfdUp37dcI5Q6zzo1w3mEVNRKeqIEOlsS0QIJn0ZzSFnfcOV3ycQ/DZp6VIxlc1mqwSw2WxngfHAVIVC8SLNJC42m+1Nm8022GazDXZNWgBq0jPQdItG3TkS1F5SR9z1vVOZyl0HCLpFGmCBKaOp/v4H6e9fp6HtFYdC6w0qJb5D+mLMyvEo2JofMvC2exVqL4JuGEvFzgNOZSp2HSDkVmkCD5o6mqrvJK8qNBCUUnWpYyPRdIvGlJPvkbetMQOgUBA4dQzl//XwSHQHiLnWTcyVrjHvPkDQH6SYA1JGU/2dc8wBrYi54P0vOTFpGScmLaPky4OE3zYeAP/EnljKqx2n1OswFZZgqazFP7EnAOG3jadk20GP42tI/vtfcnxSqnSz9RcH6TRtnN3dA3NFNabCUhd3KdbKGvwTewDQado4ir88BIA2TucoFzJ5MDVZF5v2vvclxycu5/jE5RR/eZCI6XUxN+21VNV7I6aPp3hbnTfKUS50ypBmvSDVd/qkZaTb67tTK+u7U4P6Dho/kOj5f+DUPWuw1hg7pFeuNq6LWY6+LeeYkst94b3tfJ+8gu+TV3Dpi0Popo0FIGiQ1M5Gl3Y2FpZirqwlaJDUzrppY7lkb2dDfgkhI3sDEDqmL9Wnm58/5XL/kl51uP2yHoWCuCW3cOFvO5r05r63jcPJyzmcvJyiLw4RaR9Tgc16awi0eyOnjaPI7i364iDBo68DwCdeh0Lt1exOdN572xw30+u/PETkdMkdkNi021JVQ4B9PEdOH0eRfe5seG9I+NShVJ0836S38G9f8OPkpfw4eSkl2w4QdtsEAPya6dfWyhr87P067LYJlNr7den2Q4RNk94fNq3+700hZ33Dld8n6PBYLR37p4OiaHQKuOGLCsVuYKnNZjvW4G9ewLvADJvNpmpJ8HOP690K/MYNJrJuSbyt29Fv/IjwRXdSm55J5e4DKDRqotenou2dIC0NvGStdEMXEHjTBMLvn47NZqNq72EK173baPsWi/uczH/8YHSrZqNQKinZsoNLr28mYvEMatIzqdh1EIVGTecXl6HtHY+lrJLzC9diOl9AYMpIIhbPwGaxgsVC4csfULHb/SShUrm/NrAtMfsOvY6I5TM5O63p07Fyxezl1fS1kH7jBhP5yFxQKSnbuh39Gx8RvvBOak/Ux6x7wR5zWQW5LjF3Sp3JuelNx1xWqW3ytW7PzSZo/ECsNQZOL3nNsQRq3x1/4cSkZdLn65dAfN3yqXvSOPfo2wCEpAyj2zOz8AoLxFJeRdWPZzj1p6fr69rWZN4OQNxzswiZIC0nmrVkA1XHJXf/Hes5PilVcvdPoId9qdyS3Uc5Y3f3ens5PgnR2Kw2DBcucXrFJqcjYbZm3PFrZhE8YQDWGgNZi1+nss678wWOT1wOgH//BLrbl/Qs3X2U04+8Y/em4tM9Guze7IfedPKqlM1f89rtudkE2+s7u0F9X7fjL6Q3qO+EBvV91h7zgG82oPBWYy6RjqxVHsngzMpNzfquhNfaTF23ZxsDqBRNz83t2bebQy5ve7vLLeomvdesuZewpP5Yaoz8tGgj5cellYyG71rL98krAAjsH0+fv85DqVVTtOsYpx6RlowNHtqLXs/cg8JLhdVg5OcV71DxQ+MlZzuauy3e2NlTiZ05GYDC/x0k6xnne2vUND2P9FhzH6FJA7DUGDm1aAMVdu/gXS9wOFmavwL6S8vzSksSHyPTPn8p1F5c8/ID+PfthtVoJvvJv1O6/4TT9pXNTNvd19xHyIQBWGuMnFq8gUq7O3HnC6Q55s4GyyHvPkZ23dz56gL8+3bDZrNhOH+JzOWbnJIAH1XTl8t1eXaOo1+fWfoq1fZ+3Wf7i/w4Wfru8+2XUL8c8p40ch57CwBVSADd30hFExOO8WIRWXNfwFJa6dh2lbn5i2ras76jgtwvy9ze+wQA12T8r/kv6A6C4eTeX+Kav3bD+5pxHbIeW0pcOgNmm83W6FCNQqEYZbPZvmlJ0FTi0t40tRN/JWgqcWlv5Iq5ucSlvWkucWlPWkpc2pPmEpf2pKXE5bdIc4lLe9Nc4iL4ZWkucRH8sjSXuLQ3zSUu7UlziUt70lLi0p40lbhcCUTi8svQUROXZnu1zWa70MxrLSYtAoFAIBAIBAKBwIUOvHJXR0a+0xICgUAgEAgEAoFA4CEicREIBAKBQCAQCAQdHpG4CAQCgUAgEAgEgg5Pu9+5VVMrz02PFuvVl5OVmzWyeH2U8tx4COCrMcnitVrlu2dNrhvGlTLeLC5XzHLOI3IthlBpkmfO1qrkW34zTFMri1fOxR8EV45qszxjyl8tz/cjQKlMC+f8qrCKe1wuh6tv714gEAgEAoFAIBD86hCJi0AgEAgEAoFAIOjwyLfIt0AgEAgEAoFAcDUilkO+LGRLXALHD6TzE7NBpUT/4Q4KXv/Y6XWFxotuLy/B57oELCUVnJn3AsYLhaiCA4jftALf/t3Rb9nNhVVvXpa/y1P3EZQ0SHpa7ZJXqT5xulEZ3+vi659Wu/sIOaulJ8aqgv1J2LgM79gIDOcLyb5/PZYyzx62JJe357P3EJYsPeX754UbqUhv/ATlgH5x9P7rPJRaDfpdR8l49H0A/Pt05ZoXZqP0VmMzWzi18h3Kj2Z75AWIe/pegpMTsdYYyVr8KlVu3H794uluf8J46a40zqx6F4DYh/5I6JShYLVi0peRueg1TAUlLToDxiUS8/gsFCoV+n9up3Bj4/7V5cUl+F7XHXNJOecelPqX/+gBRK+8C4XaC5vJTO5z71P57Q8exwry9e3A8QPp8uQsUCkp+nAH+Rs+aeSNe3kxvv0SMJdUcPqB9Q5vwpsP4Wf31j2VuaPH25aYAaLm30r4HRPBYiVn9VuU7z3WKnfXp+8jOCkRa42B7CWvUZ3ufiwn1D3JfXca51ZJYzn0hhHELLsdnx6d+fH6FY6nwHf0mLs/O5Ow5EQsNQZOLtxApZux7N9PetK2SqtBvyuNrEffc7wWc18KMfdOxWa2oN+Zxumn/5/H7q5P30dIkuRuqr79GtR3SYP67rLqLkImDcZqNGM4V0D2klexlFe36Lwa+/avzdvW+Utud8IzMwm1j6mMRU2PqV6v1D29Po3sx6Qx1TV1GlEzJmLSlwNwZs0HlOw62uFjlmMsC347yHOpmFJJ7DNzybrrSX5OepCQm8eg7RHrVCTsj5Mwl1by05j7KXz7M2IeuRsAm8FI7vp/cPGZ9y9bH5SUiHdcNOmj53F2xUa6rpnrtlzXNfdzdsVG0kfPwzsumqAJiQDo5t9C+f500kfPp3x/Orr5t3Rob1jyAHziovhu+CJOpr5Fr5FUo+4AACAASURBVHX3uS3Xa90sfk59k++GL8InLoqwpAEAdF89gzPrt3IweQWn122m+6oZHnkBgpMS0cbrODryQbKXbyT++Tluy8U/P4fs5W9wdOSDaON1BCcNBCD39U85nryU45NSKd5xhNil01qWKpV0fnoup+9+kpMT5xNy01i8XfpX6O2TsJRV8vO4uVx65zN0K6X+ZSkp5/S9z3BqykJylr5Ml5eWeBxrnVuWvq1U0uWZuWT8+Sl+nLCA0JvHoO3R2alI+B8nYS6r5MToByh46zM6P3JXvfeFD7jw9OV5ZRvLbYhZ26MzoTeP5sekBWTc+SRdnr0flJ5Ph0FJiWjjdBwfNZ8zD71B3Br3/Tru+bmceWgjx0fNRxunI2iC1K+rT+aQOWsdFd//9KuJOTR5ID5xOg4MX0BG6iZ6rpvttlzPdbPJSN3EgeEL8InTEWqfR4JH9SE8ZQiHJizj0LilnN/4mcfu4KREfOJ0HLPXd3wz9X36oY0cGzUfnzgdwfb6Ltt3nOMTFpM+cSm1p3OJWXBry9KrsW//Cr1tmr9kdockD8QnXsehEQvITN1E97Xux1T3tdKYOjRiAT7xOkLsYwrg4pufkzZxOWkTl3uctMgZsyxjWfCbosVZRaFQDFUoFEPs/++tUCiWKhSK69si9RvQA8PZfIw5BdhMZko++5qgyUOdygRPHkbx1t0AlPz3GwJG9QPAWmOg6tDPWA3Gy/YHTxmKfuseAKrSMlAF+aGOCHEqo44IQRXgQ9WRUwDot+4hOGVo/fu3SO/Xb9lDcMqwDu3tlDKE/C37ACg/kolXoB+aiGCnMpqIYLz8fSg/nAlA/pZ9dJo6RHrRBqoAHwC8An0xeHDGo47QlCFc2rIXgMo0ya12casjglEF+FJ5JAOAS1v2EmqP2VJZ4yin8vUGDxa28h3QA8PZPIzn7f3rP18TNMm5roImDaP4Y6l/lf7vGwJG9Qeg5sfTmAuLAajNyEHprUah8fzEpFx9268uZru3+NP9BE92jjl4cn3/KfnvtwSMrvdWHvoZq6H1K9DIOZbbEnPw5GEUf7ofm9GM8XwhhrN5+A3o4bE7ZMpQirZ+BUBlC2O5rl8Xbf2KEPuYrc26SG127q8q5vCUIRTYx3LL84gUc8GWvYRPlfpD9N2TyXn139iM0iqEpqJyj90hU4ZyqZX1falBfZftPQ4W6bKMiiMZaHRhLTqvxr79a/S2Zf6S2x0+ZQgFm6UxVZHW/JiqsPfrgs17CU8Z2mhbrUHOmOUYyx0Wq7Vj/3RQmk1cFArF48BfgY0KhWIN8BrgD6xUKBSPXq5UHRWGMbfI8bspT486KsylTGh9GYsVS0UVqpCAy1U6oYkKw5ird/GHNvbn1Zcx5unR2D+jOjwYU6G0824qLEEdFtShvd66EGov1m/TkKfHWxfqUiYUQ15xfZncYrx10mSSsepv9Fh9J6PSNtD98T+T/eyHHnkBNFGhGBq0tSFP32ii0eic68WQp0fToF66rPwTgw5votMtY8l54Z8tOtVRYZjyGvavIjf9KwxTC/0r6PqR1Px42rGj5Qly9W2NLhRjg5iN+Xo0Lm2siWpQxmLFUl6NVxu9co7ltsTsyXubdbv0a2Ouc5+td+ubLdNa5IzZWxeKwaN5pEGZ3PoyvgnRBA27lsQvnmPAv54kYECCx25Nwz5E2+o74o4kSnentei8Gvv2r9HbVuR2G1y/+1zdrmPKpUz0vSkk7l5Pz5cewCvIz2OvbDH/f/bOPDyKKuv/n0qnu7NvHZJ0JwGSEBwVBcKmshMSAqP4zgA6yrwKiqIssguu4IoiKo4CgqMy42/GBcd51XEEAqiMzshORBSyEAhL9n3vrX5/VKfT3enudCeGRqe+z5MHkj5dnzqnzrlVt6ruvT6oZVm/LHX2xGU6MBIYA8wH/kcUxSeBScCtrr4kCMK9giAcEgTh0EcNZ5wYOPmS6Hgr3YnRT7WMhAd8QXDG7+YO+IrrNJYO23TCbTNJmJVB7uN/4pu0+eQ9/ieufPk+z8ke+ON0JQMbk6Ln/srhoXMp/2gf2tmTPaF2ynTmry0zIDUR3ao7OffQJg947tGXJrddH792Ew9suo+9hLXcHZ+7GQun2/Ugx7rtuA997jJbMhL8/fCPCObI5IcpePIdrnpjqRfoLta0Q7x1D0xDNJqp+GifB0xnm/ul5/bPkdtd+ZDdyXnItY1kdHHbLg6MWMiR9BXoS2tIXnOHp2BXm3TL7Tmfe7iWZf2i1Nk7MEZRFE1AkyAIBaIo1gGIotgsCILL50iiKG4FtgIcSby5Q6obiitR6aKtvyu1GgylVfY2JZKNoaQSFH4oQoMx1dR77JijYu6cTK+ZGQA0HstHpWu/cybx7V9/0js8GVBpNegt+2ioqEEZEyk99YiJxFBZe9lxE2Znovt9OgB1xwoIiNfQZq3Wamgtsefa3hkFUOui0FtstLeMtQ7UL/vkW658yfnYnDbFzcoiduZEABpy8lHromk7cmqtBn2J/bFuLa60i4vaxmdbVfz9a65852HOrX/fLd9QUoFSa5tf0R3zq7gCpYv8UsZp6Lv1YYqWbkBfVOKW1YHtg9yGtrxp56riNBgc4txmYyi2cMOCus31lb/QPZ89+a6jYmdl2dWyWhdNQ9v3dR7Usk5jramu6lL7rJs9Cd3vpVquO5aPOt6hTh1r+WIlahuf1TY+t16souKz/QDUH80HsxmlJsw6sNhRsbOyiLHEu+FYvl2eqXQa9F7GO3rGOCInDuXHW1e79blN/025/XPmdleXmq2dPQmt5fxYfywfteO5z5HtWFPa9rw2VLRfAxT/ZTcD3lnl0T5cap99XcuXq6TLa1neqrMnLnpBEIIs/x/S9kdBEMKBLr8A15iTh7qvFlViDILSn8ipo6nNPmBnU5N9gKjpEwCI/PVI6r/xbmYnR5X96XNOZC7lROZSqnfuRzN9PADBaf0x1TVZX8Fqk6GsGnNDM8Fp/QHQTB9PzU5pH2t2HUQzQ/q+Zkb73y8n7vm3d3EgfSUH0ldS/vlB4maMASBsSCrG+ib0ZTV29vqyGkwNLYQNkd5JjpsxhvIdBwFoLakm4oarAIgcPYCm0+4v5ku27SAnY7k0oP7zA/SaMRaAkDSJbXBgG8pqMDc0E5ImsXvNGEuVhR2QpLXaRWYOpTn/gls2QFNOHuokHarEWCm/bhpNXfZ+O5u63QeImiblV8SUkdRbZg5ThAWT/PbjFK/7M42HfuyU5Shf5HYbNyCpnRt18yhqnHDb8ify1zdQ/83xn4TrC3/b2F31uSb7AFE3j0JQ+aNKjCEgSUvjsTy3vNJtO/g+YxnfZyyjescBoqePAyDETS2bGloIsdRy9PRxVLup2cvR54tv7+RQ+goOpa+g4vODxFpq2V07YmxotrYjsTPGUmGp5YrPDxAx6hoAApO1CEp/l50WkOJ9PGMZxy3x7uVlvHvZxDt83GB083/DqVlrMTd7Nu7kvym3f87c7upSs4vf3mkdTF+54yCxt0g1FZrm5tzc2Eyo5fwYe8tYKnZKNWU7HiZ68nAaT57zaB8utc++rmVZvywJHV5vsP1QENSiKLY6+Xs0oBVFsdNMdvbEBSBs/BAS1tyNoPCj8v09lLy6He2y22n6Lp/a7AMIaqU0zeSAZEw19RTOX4++qBSAq/+9FUVoEILSH1NdI/kz19CSZ1+wJrP7PlnvZ+4lfNxgaVripa/SZJmO9OpdL3EiU3qFIejalPZpib84Yp32TxEZSr/Xl6OKj0Z/oYL8uS9gqmlwybpU3DqjyiX3irV3ETVhIOZmPT8s2kx9jjT94PA9z3MgfSUAoQOTLdMhK6ncc4zch6UpF8OHX0H/p2ch+Cswt+o5tfJN6r9rn7Ix0M/9GJCkZ+cQOV6aijl/yUYacySfB2avJydjOQDBA1NItUyHXL33KIWP/FHa7z+uIDBFh2gWaT1fzumVW+zuSAWpnA8QDB0/hPjH5yAo/Kj6YDelr20nbqmUX3W7pfzq8/JSAq9OxlhTL02HfK6U2IW3EDNvOvrC9oHTBf+7GqPD0y2z2ekLbkDP57ZZdM4OnzCExDV3gZ+Cyvd3U/zqh+iW30ZjTj612QcR1EqSXllMkIVbMO9FK/ea/2xFERpo5ebevoaWvPN22/cTnLcVPe1vT/msXTgdza0TwWSiaM2b1H3R8V1pd+1I32fvsdby6SWvWac0HpD9It9nLAMg+NoUktumQ/7iCGcteR2ZNYK+T8/BXxOGqa6RxhOFnLr9KbvtK/yc3xvqaZ8bDEqXPqeuvZuoCYMwNes5tWijtR0ZuucFDqWvAKR25Fd/aJu69Rh5D0vTmApKf3614X5CBvTFrDdS8MQ71Hz9vXXbAQr3dx/7PnsPEZZ4F9jE+5rsFzluE+8Um3ifscR70DcbEdRKjNXSHeOGw7kUrtpi3bbKBdtXtQw9f5x/SVxP2i9fspuMrmuq39q7iRw/CHOznlOLN9Jgqam03S9wZKJUUyEDbaZD3nuMAktNXfHqQkIG9EUURVrPlZO3YotdxyfI3/UA+p722dhJ29lTtQxw3cWPXBfWZaSWnH/+VAMgekQBA6dclnF023H5KeSq49LT6qzj8kuUu45LT6qzjktPylXHpaflruPS42w3Fzs9KVcdl0shX/nsy3bEVcelp+Wu49KT6qzj0pNy1XHpafkqr2VdWrnruPSk3HVcelruOi49rZ9Nx+XYPy7vjsugGy/LOP73Xd3LkiVLlixZsmTJkiXrZye54yJLlixZsmTJkiVLlqzLXp6vrCdLlixZsmTJkiVLlqzu6zJe5PFyVo93XJp99L60LyX4aCxAr8Amn3Cb9b47xk0+YvtyvIfCR+wWk8InXIBApW/GUSn8fDfuooeHH7pUbFijT7gGg+/yy1djTfxdryrQ43K6VMYlkNHkuxc9/BW+iXeQ2jdjTfRG39WUL8e4yPplS84sWbJkyZIlS5YsWbJkXfaSOy6yZMmSJUuWLFmyZMm67CWPcZElS5YsWbJkyZIl61JKlMe4dEU+7bgkPX0XkemDMTfryVv0Go3HCzvYBF+bTKpl4aXqPUcpfPQtu891908lafUd7L9qNsaq+suenfTUXUSkp2Fu1pO/+FWX3H6WxRhr9hyh8DGJm/jg74iaNBzMZgyVteQteg1DaXWH7zsqZGwa8Y/fAwo/qt7Ppnzzh3afCyp/El9aSuCAFEw19ZxdsA7D+TLr50pdL/pnb6R0w7tUvPF3j/y0VZ+n7iZiQpp1samm46c72ARdk9y+2NTeI5x9TFpgK+rG64lfdiuBqQmcmLLSulCVJ/JFrG19jpyQhsmNz8E2Plfb+Nz7sTuIzBiKWW+k9WwpBUtexVTn2fil3k/eTfiEIdICp0tepel757G2LnC69zBFj0vcyBtvIH7prQSkJvDDrx+0Lo7qqXwR79Cxg0lYc4+0QOB72ZRu+pvd54LKnz4vLyHomhSM1fWcmf8C+vNlhI4eiG7VHQhKf0SDkQvPbKPh356vDO0rLkDYOImNwo/Kd52z+25YQuA1KZiq6ymcJ7EVEaEkb1lJ0MB+VG7fy/nHtnrFDR49hNhH5yIo/Kj5YCeVW7d34OrWLSdgQD9MNfVcWLQWw4UywqaOQzNnmtVOfUUShf/zAK0/dsxNV+pqGxY4MJWEtQssRgKlG/5K3c5vPeaGjk0jfvUcBIWCyvd2Uba5Y6x7v7SEoGv6YayukxayPV9G0MBUEtfOt3JLNrxLrRfc7vjcpq622yFj0tCtvgf8/Kh+P5vy1ztyE15s5xYtWIfhgj03dddGyl7xjuvLmvKVz8FjhhD32L0ICj+q399F5RYnNbV+GYED+mGqruf8A89Zueor+qJ9egF+IUEgihT+z2JEvedjaXyZ2746P8r6Zchnr4pFpg8mMFnLkesXkr/8dVKev9epXcrz91CwfAtHrl9IYLKWiAmDrZ+pdBoixlxLy/nynwU7YkIaAclajt6wgIIVm0l+zjk3+bl7KVjxOkdvWECADffipo/JSV9KTsZyqrIPk7h0RudQPz/in7yPwllryM2YT8TUMaj7JdqZRN2Siam2gVPj5lL+5sdoV82y+1z32BzqvzzssZ+2Cp+QRkCSlpyR8yl88HWS1jr3Oem5uRQ+uJmckfMJSNISPl7yuelkEXlz1lH/7Q9ecX0Saxt2YJKWYxafk934fPrBzRwbOZ/AJC0RFp9r9+WQM34xxycupeX0ReIXTnP6fUeFT0hDnaTj+Kh5nFm5mT5r5zq167P2Ps6s3MzxUfNQJ+kIH58GQPPJIvLved7rWLf57IvcTnx6LgV3PsGP6QuInDqagFT73NbcmoGptoEfxtxH2R8/QffQnQAYq+oouOsZTmYu4uySV+izYYnnzvqKa8POv+MJfpywgMibnbB/l4GxpoEfRkvs+Iclttiq5+L6v3Dh6W3eMS3cuDXzODfncQom30fYjWNRObQjEdMnYaproGDiHKre/jsxK+4CoO6TLymcupDCqQu5uPxFDBdKveq0dKcNazlVRN5NS8ibsojCO1aT8Mx8UHh42vPzI+GpuZy+8wlOTpxP5NQxqB1iHWU5zj+OnUv5m5+gXSXFuvnUWU7dtJRTUxZTcOcaEp6d5zm3mz63qUvttp8fOgs3L3M+4U64kRZu7vi5VLz5MXEOXO2jc2j4ynuuL2vKVz5r19xP0V2ryZ90P+E3jelYUzMmYaptIH/CPVS+/X/ErJwtfaDwI/6l5RQ/tpHTk+dx9vZViEYvJhLxYW776vwo65cjrzsugiD8+acAR00aRtkHXwLQcCQP/7AglDERdjbKmAgUIUHUH84FoOyDL9FkDbN+nvTkLM489Y7X0+/4ih2VNYzy7V/ZcIOdc0ODaLBwy7d/RVTWcABMDc1WO0WQGjxABw1KRX+2GP25UkSDkZpP9xGWOcLOJixzBNV/2wNA7T+/IeSGgTafXYe+qITWvCKP/bRV5KThVHz4JQANR3JRhAejjIm0s1HGRKIIDbT6XPHhl0RmSfvYkn+BloKLXnN9Ees2RU4aTrmXPpfb+Fz7VQ6YpEfI9YdzUWk1HnEjJg2n8sMvAGjshNt4+BQAlR9+QYTF55b8812KNfgut1vPlKAvknK7+tN/EZ453M4mPHMElR/uBaDmn98QOvJaAJpPFGIsrQKgJbcIP7USQeXZA2hfcQGCHdmfdGRHZI6gysKu/qydbW5upfHgj5hb9R7z2hR4bX/0Zy9iOFcCBiN1n+0jNP16O5uQiddR+9FuAOp2fE3Q9QM7bCfsxrHUffqVV+zutGFiS6u1lgS1CtGL9lo6zu3c6k//RXiGPTc8YwRVf7M9zm1cvZXrp1Z51X5012fps66120EDJa7Bwq39dB9hDj6HZYygpo37uQM34zr050poyfWS68Oa8pXPgQPta6r2H/sInXidnU3oxBHUfiRx6z7/mmBLTYWMTqPl5BlaT0pPtU019V5NrevL3PbV+fGylNl0ef9cpnLbcREE4ROHn0+B37b93h2wSquh9WKl9ffW4irUDgmo1mrQF7fb6IurrEkalTkUfXEVTT+c/dmwVXFRtF6ssOFWdig6lVaD3m7fKlHFRVl/773qdoYc2kKv346h6IX3OmUqYzUYbJiG4kqUsRrXNiYzpvpGFJFhCIFqet03jdJX3vXKTzt/HHzWX7T3p83GLtZObLrLvRSxtmXrfyKfY26bQM3eIx5y7f0xFFeidNim0pFbXIkqrvsNvy/iLflrE2dnuR0X5SS3Q+1sIqbcQPOJQkS9Z1Mu+4orbdeeLR3jjmx9J2xv5R+nwVhswy2pwN/BZ/9YDYaScivX3NCEIjLMzibs12Oo+4d3HZfutGEAgYP603/XRvrvfJULj26yXvR0yo3TYLD1ubjCSaydcaVYBw3qzxXZr3HFzj9w/hHPud31uTvttr+jzyVO8itWg77YNbesC1xf1pSvfPaPtecaSyo6+Cztm01N1Us1peobD6JI77efJOnjV9Dc691TB1/mtq/Oj7J+OersiUsCUAe8BLxo+am3+X/X5WQO+Q53w5zNMy+K+AWqSFg8jaJ17/+s2IKzifMduE6n1rcxKXrurxweOpfyj/ahnT3ZE2inTFc2cUtmUvHmx5ibWjrneMHvGGunwe46Ex/Fuh3eKdsTn3UPTEM0mqn4aJ+HXGebdPDZk33rgnyT286250lut/83oH8iuofuoOihTZ3zfM31lO20gfMO4xHYy9wKGHgF5uZWWvO8vNnUjTYMoPlYLrmZ88mfupSY+2cgqD1d+6mrXOmfpmO5nMpYQO7UZcTMm+4F19V2L0G73dX2WhSJXTKTire6ynXyt0tWU77y2ZOFdJxzBX8FQUOv4sLS9Zy59UFCM64n+IaOTzi93W6n++fL3O7u+VHWL0adPU8dCiwCHgFWiKJ4TBCEZlEU3d4yEwThXuBegBWhg7k5KBmAuNlZxM5MB6DhWAFqnYa2Ie1qbRT6kiq77bRetL9rq7LYBPSJQ907hkF711u+q2HQrnXkTH4IQ3mN033yFTtuVhaxMydK3Jx81LpoG66mI7e4EpWunavWatCX2tsAVPz9a65852HOrXffgTKUVKDURVt/V2o1GMqqnNoYSipB4YciNBhTTT1Bg/oTPuUGtA/NQhEWjGgWEVv1VP75M7fM2FlZ9JqZAUDjMcnnBstnKp2mw6BrvcPdeZVOg77E84HwbfJlrGNnZRFj8bnhWD4qm5irdBr0XvocPWMckROH8uOtq936HHPnZLtY2/qj1HoQaxc+eyJf57a+uNI+zs5yu7jSaW6DdEcxaetDnF2yAf3ZEo/99hW3bbsqx3oudaxnycYZu6syllTgr7XhxkVjdNaOxPXCaOH6hQTZcaWnLV96ze5OG2ar1oLzmJtbCOjfh+bj+Z5xbX3WRneMdbEH3HzvuN31uavtNoCx2MHnOI31NSxbrkobbT3OdtzJNxC3ynuuL2vKVz4bHfLLPy4aQ2mlExubmgqVaspQUkHTge8xVdcB0PDVIQKuTqHx3zke+Xypc9tX58fLXvKsYl2S2ycuoiiaRVF8GZgNPCIIwmt4MBOZKIpbRVEcKori0LZOC0DJ2zvImbiCnIkrqNpxgJhbxgEQkpaKsb4JQ5n9hb+hrAZTYzMhaakAxNwyjqqdB2k6WcTBAXdzeNg8Dg+bR2txJccyH3TZafElu2TbDnIylkuDjj8/QK8ZYzvlmhvaub1mjKVqx0EAApK0VrvIzKE0519w6W+bmnLyUPXVoUyIRVD6E3HTGOqyD9jZ1GXvJ3Ka1KkLnzKShn9/B0DBLas4OWoOJ0fNoeKtTyjbuN2jBrl02w6+z1jG9xnLqN5xgOjp4yw+98dU14ShzL6RMpRVY2poISStPwDR08dRvfOA42Y7lS9jXbptB8czlnHc4nMvL33uZeNz+LjB6Ob/hlOz1mJudj8eoexPn3MicyknMpdSvXM/munjAQh2wzU3NBNs4Wqmj6emC7GGyyO31UlaVIkxCEp/Im8aTa1DbtdmH0AzfQIAEVNGUm/JbUVYMCnbHuPi8+/QeOikV377igvQmJOHuq8Ne2pHdk32AaIs7Mhfj6T+m++85jiq+XiutR1B6U/Yr8dQv8d+JqGGPfsJ/63UkQ3LGkXTtzZcQSBs8mjqPvP+7mh32jBlQqx14LAyvhfq5Hj0NjNvdcZVJ+lQJcZaj3Nd9n577u4DRE3reJxVifbcgOR49OdLL4nPXW23AZq+y0Ntww2/aQx1ux24u/cT0cadPJKG/0jc07es4tToOZwaLXHLN3nB9WFN+crn5u9yUfWNt9ZU+I1jaNhjn1/1e/YT/luJGzZ5FI0WbsO+I6iv6IsQoAaFH0HDr6E175znPl/i3PbV+VHWL1OCN4MVBUH4NTBSFMWHPf3ON3HTXQKS184hYvwgzM2t5C/eREOONP3qwN0vkDNxBQAhA1PoZ5mSuGbvUU4//GaH7Qw5uImcSSu9mg65J9mC4DqmSc/OIXL8YEzNreQv2UhjGzd7PTkZywEIHphCqmXK2Oq9Ryl85I8AXPHHFQSm6BDNIq3nyzm9covdXe0QtfMiDh03BJ1lWs3qD3ZTtvEDYpfMpPl4HnW7DyColdK0mlcnY6ppoGjhOvTn7Bui2MW3YWpscTrVY7Pe/WPivs/eQ/i4wZibWzm95DXrlMYDsl/k+4xlks/XppDcNh3yF0c4a/E5MmsEfZ+eg78mDFNdI40nCjl1+1PWbZtE14/bezLWfm6OcZvPERafC2x8vib7RY7b+Jxi4/MZC3vQNxsR1EqM1VJONRzOpXDVFuu2FW7YvZ+51xrrwqWvWqc0vnrXS5zIXApA0LUp7dMhf3GEokffACAiawR9np6Df1Q4prpGmk4UkjvzSeu2DWb3b5b2ZLwDlc7fWQ8bP4T41XdLU6i+v4fS17YTt/R2mo7nU5ct5XafDUsIujoZY009ZxasR19USuzCGcTOn05rYftkBAW/X4Oxstatj5eS66ppDhs/hIQ17eySV7ejXXY7Td/lU2th992whMAByZhq6imcL7EBrv73VhShQQhKf0x1jeTPXEOLwwVPYIDzKVWDxw4l9hHLdMgf7qJy8/tEL/o9LcfzaNi7H0GlRLd+OQFXSVPGXljyvDTwGAgafg0xK2ZzZsZSlzE1GBQuP+tqGxbxm/HE3D8d0WgEs0jpH96jblfHqVvNLtqR0PFDiH98DoLCj6oPdrcf5+/yrdw+L0tcY029NGXsuVIifzOOmHnTwWBEFEVKX3mP2l37O2zfXTvS0+22qzeVQscNQfu4ZWrg7bsp3/gBMRZu/e4DCColiS8vJeCqZEy1EtfgwI1ZdBvmJudco8l5O3Ipaspf4fwOd0/7rPBzzg0ZN5TYR+9F8POj5sNsKja9T6/Fv6f5eB4Ne6Sain9xOQFXS7V8ftE6a02FLNkIKwAAIABJREFU3zwezX0zAJGGLw9R9vzbHbavN7qpqR7O7RaD63vcPXl+BLju4keevIfnc7Uc2N7997R7UAHDZ1yWcfSq49IVueu4/FLlruPSk3LVcelpddZx6Um567j0pDrruPSk3HVcelKddVx6Uq46Lr9k9XDT7FKuOi49LXcdl56Wq45LT8uX7YhHQyx6QK46LpdCrjouPS1XHZeelruOS0/LXcelp/Wz6bh8+/5lfX0ccN2tl2UcfdeCyJIlS5YsWbJkyZIl62cpQRCyBEE4JQhCviAIq5x83lsQhC8EQTgqCMJ3giBM6S5T7rjIkiVLlixZsmTJkiXLYwmCoAA2ApOBq4DbBEG4ysHsUeADURQHA78DvJz2r6N89yxPlixZsmTJkiVLlqz/Rv38ZxUbDuSLongaQBCE94CbgR9sbESgbUGvcKBrq1zbqMc7LkEq37wv7at3SgFqW9Q+4fpqrInRh2MfoiMafcLVt/quz28y++a100CFb2oZoKFF5ROuL1/wbTb55v30RoNv2hGD6MN2JLDZJ1xfjkEw+ajdVip8tyJ3vY/aEdFHY6j8fXgdVG/y3dhXWZdM8YDt7C7ngREONmuAXYIgLASCgYndhcqvismSJUuWLFmyZMmSJcsqQRDuFQThkM3PvY4mTr7mOOHAbcA2URQTgCnAO4IgdKvvIb8qJkuWLFmyZMmSJUuWLKtEUdwKbHVjch5ItPk9gY6vgt0NZFm29x9BEAKAaMCzBbWcSO64yJIlS5YsWbJkyZJ1KWX+2Y9xOQikCoKQBFxAGnx/u4NNEZAObBME4UogACjvDtRnHZfQsYNJWHOPtNjUe9mUbvqb3eeCyp8+Ly8h6JoUjNX1nJn/AvrzZYSOHohu1R0ISn9Eg5ELz2yj4d/HPeaGjE0j3rKoV9X72ZRv/rADN/GlpQQOkBZRO7tgHYbzZQQOTCVh7QKLkUDphr9St7PjQmadKfnpu4hKH4y5Wc+pRa/ReLyw4z5em0x/y8KXVXuOcvrRtwDovfwW4mamY6isA+DM2r9SveeoR9w+T91NxIQ064JPTcdPd7AJuia5fcGnvUc4+5i04GbUjdcTv+xWAlMTODFlpXWxKE+V9PRdRFp8znPhc/C1yaRafK7ec5RCi89t0t0/laTVd7D/qtkeLTQaNGoIsY/cB35+1H64g6o3ttt9LiiVxD2/jICrUzHV1HFx6VqMF8rAX0Hc04sJuCoFFArqPt5D1dYPvPI3ZEwautWWxczez6b89Y45lvBie44VLViH4UL7zQelrhepuzZS9sq7Thczc6XQsWnEr56DoFBQ+d4uyjZ3rKneLy0h6Jp+GKvrpAXFzpcRNDCVxLXzLUYCJRvepdaL3A4Zk4b28Xslfz/YRYUzf9cvJWBAP0w19Zxb+DyGC2Uo42NIzd5M6+kLADQfO8XFRzd6zG1Tn6fuJnJCGiY3uR1sk9vVNrnd+7E7iMwYillvpPVsKQVLXsVU1+Qx11c11e+Z2WjSJZ9PPrCRBhftyK/+MB9FgIrKPUfIf6R9gbr4u7OIv2syotFE5e4jnH7q/3nE7U77ZWXfP5Xk1XfwHw9ruU2pz8xGky4tWvfDA5uc+hx6bRJX/kFiV+45Sp6Nzwl3ZxF/V5bV54Kn/tIp05fni7Bxg+n9xBxQ+FHxbjYlGz/qwE7asJiga6Vz5On716M/L7UjcfOnEX3bRDCZKXr8Deq+OuYV2xe5HTZOuiZA4Uflu86vCfpuWELgNSmYquspnCe1X4qIUJK3rCRoYD8qt+/l/GPubhC7VtJTdxGRnoa5WU/+4lddnqf6WRbQrdlzhMLHpNxOfPB3RE0aDmYzhspa8ha9hqG0usP3nXK7cX7s/eDviMoahmg2Y6ioI3/Ra+g95ILv2rArnplFdLq0WPGJBzZT76KWr/7DPBQBKir2HOXUI9sACLm6D1e+cA8KtRLRaOLHVW9Sd9S79lPWTyNRFI2CICwAdgIK4C1RFE8IgvAkcEgUxU+AZcAbgiAsQXqNbJbYzQUkfTPGxc+PxKfnUnDnE/yYvoDIqaMJSE20M9HcmoGptoEfxtxH2R8/QffQnQAYq+oouOsZTmYu4uySV+izYYlX3Pgn76Nw1hpyM+YTMXUM6n723KhbMjHVNnBq3FzK3/wY7apZALScKiLvpiXkTVlE4R2rSXhmPii8C19k+mACk7Ucun4hectfp9/zjq8LSur3/D3kLd/CoesXEpisJXLCYOtnF7Z+xtGJKzg6cYXHnZbwCWkEJGnJGTmfwgdfJ2mtc27Sc3MpfHAzOSPnE5CkJXy8xG06WUTenHXUf/uD0++5U5vPR65fSP7y10lx4XPK8/dQsHwLRyw+R9j4rNJpiBhzLS3nPeyk+/kR+/h8zt/zGIU3ziX01+NQpfS2Mwmfnom5roHCSXdT/af/o9eyuwAIzRqNoFRyZuo8zk57gIhbp+AfH+O5w35+6Cw5lpc5n3AnORZpybHc8XOpePNj4iw51ibto3No+Oqw50wLN+GpuZy+8wlOTpxP5NQxqB1qKspSUz+OnUv5m5+gXSXVVPOps5y6aSmnpiym4M41JDw7z/Pc9vND98T9nJm9mvxJ8wi/aaxzf+sayZtwL5VvfUzcynZ/9WdLKLjxAQpufKBLnZaICWkEJmk5ZsntZDe5ffrBzRwbOZ/AJC0Rltyu3ZdDzvjFHJ+4lJbTF4lfOM0jri9rKip9MIFJWvZft5Dc5Vvov+4ep3b9191D7vIt7L9uIYFJWqImDAIgYuTVRGcN4+D4ZRwcu5Rzmz/xiPtTtF8qnYZIb2rZIk36YIKS4vj2ugc4uXwrV6yb49TuinX3cHL5Fr697gGCkuIcfB7KgfHLOTB2GUWbP+0c6svzhZ8fvZ+eS+7/PsmJ8QuJunk0AakJdibRv8vAWNvA96Pup/SNT0h4+A4AAlITiLp5FCcmLCT390/Q+xnpBo6n8kluW64J8u94gh8nLCDyZifXBL/LwFjTwA+jpWuC+Iel9kts1XNx/V+48PQ2z3kOipiQRkCylqM3LKBgxWaSn3Puc/Jz91Kw4nWO3rCAAJvz1MVNH5OTvpScjOVUZR8mcekMj7jdPT9e2PQxxyYsI2fiCqq94ILv2rDo9EEEJcXxzXWL+HH5G1y57m6ndleum8OPy7fyzXWLCEqKQ2Op5f6Pz+T0+g/5Nn0lBes+IPWxmV7xZf20EkXxn6Io9hdFMUUUxWcsf3vc0mlBFMUfRFEcKYriQFEUB4miuKu7TJ90XIIGpdJ6pgR9USmiwUj1p/8iPHO4nU145ggqP9wLQM0/vyF05LUANJ8oxFhaBUBLbhF+aiWCyrMHR0GDUtGfLUZ/TuLWfLqPsEz7CRDCMkdQ/bc9ANT+8xtCbhgIgNjSCibpsZ6gVtGVDqNm0jDKPvgSgPojefiHBaGMibCzUcZEoAgJov5wLgBlH3yJJmuY1yxbRU4aTsWHErfhSC6K8GCUMZEO3EgUoYE0WLgVH35JZJYUm5b8C7QUdG0Guygbnxu66HPSk7M489Q7Hi8lHnBtfwxFFzGcLwGDkfp/fkVI+nV2NiHp11P7f7sBqN/5L4KulxpFRBG/oABQ+CEEqBANBswNnt2BBwgaKOWYwZJjtZ/uIyzDIccyRlDTlmOft+eY9Nl16M+V0JJb5DET2mqqPberP/0X4Q7c8IwRVP3NtqbacltvzW0/tarj0Do3ChzYn1Zbf/+xj9AM+1iHTryuvaY+/5pgG3+7q8hJwyn3MrfLbXK79qscq+/1h3NRaTUec31VU9FZwyjd/hUAdYfz8A8LRuVQU6qYCPxDAqk7JLFLt39F9GSpjdXdmUnRq/+HqDcCYKio84j7U7RfKU/OotCLWm73eSgl2/d16rMiJJC6Q3kAlGzfR6/JEjv+zkzOvvqxVz778nwR3FbPlnNk1cdfE+HAjsgcTuX2LwCo/uzfhI661vL3EVR9/DWi3oj+XBmtZ4oJHpTqMdsXuR3seE3wScdrgojMEVRZrgmqP2u/JjA3t9J48EfMrXqvmLaKyhpGuaWmpPNUsPPcDg1qb0e2f0VUlrSPpob2WekUQWqP29Dunh9tuX5BakQvGm9ftWG9soZRbKnl2k7ar1pLLRdv30eMpZZFEfxDAwHwDwui1YsnTJedzObL++cylVcdF0EQRgmCsFQQhMzuQFVxGvQXK6y/64srUcbaXzAo46IwtNmYzJjqG1FEhtrZREy5geYThdaTUWdSxmratwkYnHFtbaxcaQrqwEH96b9rI/13vsqFRzdZT0yeSqXV0Hqx0vq7vrgKtcOFklqrobW43aa1uMruYkp3VxZpe18k9eV5+IcHe8aNi6LVNt4XK1HFRXWw0dtwndl0RY4+t7rw2Y5t43NU5lD0xVU0/XDWY6Z/bDSG4vY7usaSCvwdjrN/jAZjcftxNtc3oYgIo37n15ibWkj5119J2ftnqt76CHNtg+fsOA2GYpscK6lEGdcxx/TFHXNMCFTT675plL3yrsc86zYducUVHblxznJbqqmgQf25Ivs1rtj5B84/4nluS1ybWBdXOK+pNpu2WFtqSpUYS8qnr5D07lqChl3tlc9gydufKLdjbptAzd4jHnN9VVNqbRStF2xrqhK1NqqjjW07crHdJihFR/iIK0n7/FkG/f0JQgeleMTtbvsVlTmU1uIqGr2oZVt/Wi60x9sTn1vsfNYSMeJXDPn8GQb/fY1HPvvyfKHSRrW3EYC+pBKV1ll+2bDrmvCPDPXou27ZPshtpcM1gaHYSbtpW+surgm6KkefW4srO9zEUGk16C/a152tz71X3c6QQ1vo9dsxFL3wnmfcbp4fJe5tDD38Or2mjaZo3fseccF3bZhaG0mLTfvVUlxJgEN+BmijaCmuare5WIVaK3Wqch/7E6mP/57RRzaSuvp/yX/G+/OlrJ+33HZcBEE4YPP/e4DXgFBgtSAIq7pMdTqBmsOdAsGJkY1JQP9EdA/dQdFDXizC6XSbnnAlm+ZjueRmzid/6lJi7p+BoPZunnJnm+7Id21TvG0nB0cs4Ej6cvSl1SStubPL4A53AJ3vnGfbd8t2slUPffYLVJGweJpXjbFLObriLCaIBF5zBZjNFIyZyemJs4ia/VuUCXGec7oaa1EkdslMKt76GHNTi+e89o063WbnXOmfpmO5nMpYQO7UZcTMm+51brvnOrcxlldxatRsCm5aRPEzfyTx5eX4hQR6x+pqTTskhO6BaYhGMxUf7esy95LVlJOAdrih72b/BH8//COCOTL5YQqefIer3ljqGbUb7ZdfoIrExdM42+Va7l5+Sz6HcHjyI+Q/+Q4D3vDgFWOfni+6eow9/K5btA9y25PccpoDXUfabdmDY+10JRYbk6Ln/srhoXMp/2gf2tmTPQQ7w3p+TSBx3+XQkPso/9u/0N6V5RkXfNiGdfUcKf2TMCuD3Mf/xL/S5pP7+J+46uX7urk/sn5u6uwdK9uW9l4gQxTFckEQ1gPfAs85+5Jlrud7AR6JvJZpIX3tPtcXV6LSRVt/V2k1GMqq7GwMxZUoddEYSipB4YciNBhTjTSQUxmnIWnrQ5xdsgH92RIP3LRss6QCpQ1X6YxrsXHGbVNrwXnMzS0E9O9D8/F8t0zt7CziZqYDUH+sALWu/U6JShtFa4k9X7oz2m6j1kaht9gYKmqtfy/5y26ufuchl9zYWVn0mpkBQOOxfNS6aNqeG6h0mg4DB/UOd5hUOg36kq49go2bnUWsxecGi89tEbT1p02tFx3YFpuAPnGoe8cwaO96y3c1DNq1jpzJD2Eor3HJN5ZWoNT2sv7uHxeNsayyg42/NhpjaQUo/PALDcJcU0/ojeNo/NchMJowVdXSfOQHAgakSq+deSBjcQVKrU2OxWmsrza2yVBSgUobjdEhx4IG9Sd88g3ErZqFIiwY0Swituqp/PNnnXINJQ5cbTQGR26xB7md73luS9xK+1hro53UlGTT5q9faJCVa9JL/7Z8X4C+qARVUjwtnXBjZ2URY8nthmP59m2JTtNhcGpnuR09YxyRE4fy462rO+X6qqZ0syeh+720blfdsXzU8bZthMZpTdm1Izbs1otVVHy2H4D6o/lgNqPUhFkn/bDVT9V+BfSJI6B3DGk2tTx41zqOuanl+NmT0P2+nR0QH00tp6zfb3WIpSM7QKex7l/rxSrKrT4XWHwOxVDpenIAX5wv2iTljk1ex2kwOMS6zcZQbGGHSXXlyXcd5cvcBul8r3KMdYd2U7JxF2tvFDcri9iZUk015Eg+t5+nnNRUcSUqnUPdlXaMa8Xfv+bKdx7m3HrnnfSf6vzYkfsvrvx/D3PuBdeTyfjqOCfMziTBUsu1xwoIsGm/ApzUcstF+6cwAbooq432lrHWgfqln3zLVS/N9Xp/LheJou8WY/05q7NXxfwEQYgUBEEDCKIolgOIotgIuHw/SxTFraIoDhVFcahjpwWgKScPdZIWVWIMgtKfyJtGU5t9wM6mNvsAmukTAIiYMpL6f38HgCIsmJRtj3Hx+XdoPHTSc08tXFVfHcqEWASlPxE3jaHOgVuXvZ/IaVKBhU8ZSYOFq0yItQ6uVMb3Qp0cb53BxZ2K395hHUxfueMAMbeMAyA0LRVTfROGMvuTtqGsBlNjM6Fp0jvJMbeMo3LnQYlr8x6oZvIImk6ew5VKt+3g+4xlfJ+xjOodB4ieLnFD0vpjqmvCUGbfUBjKqjE1tBCS1h+A6OnjqN55wHGzHqnk7R3kTFxBzsQVVNn4HJKWitGNzyE2PlftPEjTySIODribw8PmcXjYPFqLKzmW+aDbTgtAy/FclH10KONjQelP6JSxNOy1n9GnYe+3hP+PdNIKnTSapm9zADAWlxN0nfSeuhCoJmDgr9Cfdh1nRzV9l4faJsfCbxpD3W6HHNu9n4i2HJs8kob/SDl2+pZVnBo9h1Oj51Dx1ieUb9ruUacF2mpKhyox1lpTddn7HbgHiJrWsaZUifa5HZAcj/58qUfc5u9y7f29cQz1u+259XtsamryKBot/iqiwqyDhpWJsaj66jAUdd5BLN22g+MZyzhuye1eXuZ2L5vcDh83GN3833Bq1lrMze7fkfdlTV18eyeH0ldwKH0FFZ8fJHbGWADChkg1pXeoKX1ZDcaGZsKGSDUVO2MsFTukdqTi8wNEjLoGgMBkLYLS32mnBX669qvpZBH7B9zNwWHzOGip5aOd1PKFt3dyMP1BDqY/SPnnB4ibMcbqs8mFzyYbn+NmjKFixyEAyj8/SOSoAQ4+u7/o9cX5ok2NOXkE2Jwjo24eRY0Duyb7AJoZ4wGI/PUN1H9z3Pr3qJtHIaj8USXGEJCkpfFYnlueL3O7zV91X5trgqkdrwlqsg8QZbkmiPz1SOq/+a7LPICSbTvIyVguDaj//AC9LDXl7jxlbmg/T/WaMZYqS00FJGmtdpGZQ2nOv+Ca+xOdHyVu+9sAUZOGueWC747z+bd38W36Sr5NX0n55wfRWmo53G371UK4pZa1M8ZQbol1a0k1kTdcJfk8egBNpz2/eS3rlyHB3aBBQRDOAGakZ3sicIMoiiWCIIQAX4uiOKgzwNHeNzsFhI0fQvzqu6XpkN/fQ+lr24lbejtNx/Opyz6AoFbSZ8MSgq5OxlhTz5kF69EXlRK7cAax86fTWtg+KKzg92swVtbabV/h5/x94tBxQ9BZpres/mA3ZRs/IHbJTJqP51G3W+ImvrSUwKuTMdU0ULRwHfpzpUT8Zjwx909HNBrBLFL6h/eo2+V8esvaFrXLeKSsnUPk+EGYm1vJXbyJhhxpGr/Bu1/g6MQVAIQMTLFOJ1q99ygFD0vTD/Z/dSEhA/qCCC3nyshbscWukVO58Bmg77P3ED5Omkr09JLXrFMXDsh+ke8zlgEQfG0KyW3THn5xhLOP/BGAyKwR9H16Dv6aMEx1jTSeKOTU7U9Zt200u+//Jq+dQ4TF53wbnwfufoEcG5/7WXyu2XuU0xafbTXk4CZyJq20m0I1OqLRKTN4zDBiHr4X/BTU/m0XVVveQ7Pwf2n5PpfGL/YjqJRo161AfWUKptp6ipc+h+F8CUJQANpnl0qzkAkCtR/tovqtv3XYvr7V9cPK0HFD0D5umQ55+27KN35AjCXH6ncfQFApSXx5KQFXJWOqlXLMcM6+oxCz6DbMTS1Op0M2mZ2+tEDo+CHEPz4HQeFH1Qe722vqu3xrbvd5WcptY029NB3yuVIifzOOmHnTwWBEFEVKX3mP2l37O2zfX+E8v0LGDUX72D0Ifn5Ub8+mfNMHxCy2+LtH8jfhpWVWf8898DyGc6WEZd1AzOKZiCYzmEyUbfgr9XudnxQbWlQu49332XuIsOR2gU1uX5P9IsdtcjvFJrfPWHJ70DcbEdRKjNVSTjUczqVw1Rbrtp1Hup3bUzUF0GxSuGSnrr2bqAmDMDXrObVoI/U50jSmQ/e8wKF0qaZCB0rTIUvTEh8jz1JTgtKfX224n5ABfTHrjRQ88Q41X39v3bZCcH1O6E77ZathBzdx1KGWDaL7dqT/2rvRTBiIqVnPj4s2WX0etmcdB9MftPp8pWUK1co9x8h9+C2Lzwqu3DCPkAF9EPVG8p94h+qvT1i3HR3Y3BFIz58v9EbXxzh8whAS19wFfgoq399N8asfolt+G405+dRmH0RQK0l6ZTFBA5Ix1dRTMO9F9EVSO6JdOB3NrRPBZKJozZvUfdFx7JbJTbvdk7mtVDi/yxw2fggJa9qvCUpe3Y52mdR+1VquCfpuWEKgxd/C+eut/l79760oQoMQlP6Y6hrJn7mGlryON5xajK7b7aRn5xA5XpqiN3/JRhrbzlPZ68nJWC75PDCFVMt0yNV7j1Jo8fmKP64gMEWHaBZpPV/O6ZVb7J6KiKLrlqQ758cr/ricwH46sHALHtxqx/V3c00APXuc60yuX4381dq7rLX8w6LN1Flq+bo9z/Nt+koAwgYmc/Uf5uEXoKRizzFOPSxNbR4x/AqueHoWgr8Cc6ueH1e+Sf139tMpZ5S+767pvmzUvG/bT/SyY88ocMysyzKObjsuLr8kCEFArCiKHSffdpCrjktPy1XH5VLIXcelJ+Wu49KT6qzj0pNy1XHpabnruPS0XHVcelquOi6XQu46Lj0pX7ba7jouPSl3HZeeVGcdl56Uq45LT8tdx6Wn5a7j0pNy1XG5FHLXcelJueu49KQ667j0pNx1XHpaP5uOy5dvXd4dl3F3XZZx7FIVi6LYBHTaaZElS5YsWbJkyZIlS5asn0K+u8UlS5YsWbJkyZIlS5YsWR7Kd++7yJIlS5YsWbJkyZL13yjx8l3k8XKW/MRFlixZsmTJkiVLlixZl716/ImL0t83A/F8NRgOIFTtflrVnpKfjwbV+lJNTb4ZtO2rAfK+lNmHPvtqwLjZh+1IgA8HMftCSv77JlQRfTj9g9lHpwuTD2vKV2xfTZzjy1hrVF1ZSFmWrM4lP3GRJUuWLFmyZMmSJUvWZS95jIssWbJkyZIlS5YsWZdSZnmMS1ckP3GRJUuWLFmyZMmSJUvWZS+fPXEJGZOG9vF7pdXFP9hFxesf2n0uqPxJWL+UgAH9MNXUc27h8xgulKGMjyE1ezOtpy8A0HzsFBcf3egVV7fasqr5+9mUO+O+uJTAASmYauopWrAOw4Uy6+dKXS9Sd22k7JV3na5q7k6hY9OIXz0HQaGg8r1dlG22X5FdUPnT+6UlBF3TD2N1nbSy+fkyggamkrh2vsVIoGTDu9Tu7LgKs0ufx6YRb1n9uer9bMo3d/Q58aV2n88uWIfhfBmBA1NJWLvAyi3d8FfqvOD6kh06djAJa+6RVmF+L5vSTR1j3eflJQRdk4Kxup4z86VYh44eiG7VHQhKf0SDkQvPbKPh38e98jls3GB6PzEHFH5UvJtNycaPOrCTNiwm6FqJffr+9ejPl6GICCVl64MED+xH5fa9FD36xiXhAsTNn0b0bRPBZKbo8Teo++qYV9yENdIxrnzXeaz7blhC4DUpmKrrKZz3gtXf5C0rCbL4e/6xrV7526beT95N+IQhmJtbKVzyKk3fn+5gE3RNMkkvP4BfgIravYcpelxaeVoREULK5mWoE2NoPVdGwX3rMdV6tqhpn6fuJmJCGubmVgqWvEbTcefclLZVp/ce4exjEjfqxuuJX3YrgakJnJiy0rpitafq89TdRE5Iw+SGHWzDrrZh937sDiIzhmLWG2k9W0rBklcx1TVd1lxfspOfvouo9MGYm/WcWvQajcc7LmEWcm0y/S0rm1ftOcrpR9+y+zz+/qkkr76D/1w1G2NVvcc+pzw9m6h0yefcRRtpcMG+wso+QsGj0grjfZbPIG7mRAyVdQAUrv0r1XuOesTt98xsNBbuyQdcc3/1h/koAlRU7jlC/iNvt/t7dxbxd01GNJqo3H2E00/9P4+4SU/dRUR6GuZmPfmLX3Ua6+Brk+lnWb2+Zs8RCh+TYt3nsTuIzByKqDfScraE/MWveZVfvjrOvmxHuhPvxAd/R9Sk4WA2Y6isJW/RaxhKqztl+vp8IevnL988cfHzQ/fE/ZyZvZr8SfMIv2ks6n6JdiaRt2Riqmskb8K9VL71MXErZ1k/058toeDGByi48QGvOi34+aF78j4KZ60hL3M+4VPHOOfWNpA7fi4Vb35M3KpZdp9rH51Dw1eHvfUY/PxIeGoup+98gpMT5xM5dQzqVHt21K0ZmGob+HHsXMrf/ATtqjsBaD51llM3LeXUlMUU3LmGhGfngcLDQ+fnR7zF59yM+UQ48TnK4vOpcXMpf/NjtBafW04VkXfTEvKmLKLwjtUkPDPfc64v2X5+JD49l4I7n+DH9AVETh1NgEOsNZZY/zDmPsr++Am6h6RYG6vqKLjrGU5mLuLsklfos2GJ5/5a2L2fnkvu/z7JifELibp5NAGpCXbg7T/lAAAgAElEQVQm0b/LwFjbwPej7qf0jU9IePgOAMRWPRdf+Cvnn9rmHbOb3IDUBKJuHsWJCQvJ/f0T9H7mPvDzLtb5dzzBjxMWEHmzk1j/LgNjTQM/jJZiHf/wne3+rv8LF57ugr8WhU9IQ52k4/ioeZxZuZk+a+c6teuz9j7OrNzM8VHzUCfpCB+fBoB2/m+p+/o4x0fNp+7r42jn/9ZjbkCSlpyR8yl88HWS1t7r1C7pubkUPriZnJHzCUjSEj5+MABNJ4vIm7OO+m9/8NrniAlpBCZpOWZhJ7thn35wM8dGzicwSUuEhV27L4ec8Ys5PnEpLacvEr9w2mXN9SU7Mn0wgclaDl2/kLzlr9Pveefcfs/fQ97yLRy6fiGByVoiJwy2fqbSaYgccy0t58s99teWffD6heQt30K/5+9xyc5dvoWDVvYg62cXtv6DIxNXcGTiCo87LVHpgwlM0rL/uoXkLt9C/3XOuf3XSdz91y0kMElLlIUbMfJqorOGcXD8Mg6OXcq5zZ94xI2YkEZAspajNyygYMVmkp9zHuvk5+6lYMXrHL1hAQHJWiIssa7Zl8OxcYvJSV9KS8FFEhZ6Vsvgu+Ps63akO/G+uOljctKXkpOxnKrswyQundE51Mfni8tOovny/rlM5fbqRBCEEYIghFn+HygIwhOCIHwqCMLzgiCEdxUaOLA/rWeLMZwrRTQYqf3HPkIzrrOzCZ14HdV/2wNA7edfE3zDwK7irAoamIrelvvpPsIyRtjZhGWMoMbK/YYQG25YxnXoz5XQklvkPXtQKq1nitFb2NWf/otwB3Z4xgiq/rYXgJp/fkPoSIkttujBJCWRn1oFXswGEzRI8rmNW/PpPsIyHXzOHNEe63+2+yy2tFq5glqFKHo3DY2v2FKsS9AX2cQ6c7idTXjmCCo/tI31tQA0nyjEWFoFQEtuEX5qJYLK8weTwW3H2cKu+vhrIhx8jsgcTuX2LwCo/uzfhI6S2ObmVhoO/oi51eAx76fgRmSOoOrjrxH1RvTnymg9U0zwoFQvuDax/qRjrCMyR1BliXX1Z+2xNje30njwR8ytXZ+FL2LScCo/lHxqPJKLIjwYZUyknY0yJhJFaCCNh08BUPnhF0RkDW//viUmldu/ICLLPmauFDlpOBUffglAQyfchsO5AFR8+CWRlu235F+gpeBiFzyW2OVesstt2LVf5Vhrq/5wLiqt5rLm+pKtmTSMsg8kbv2RPPzDglDGRDhwI1CEBFFv4ZZ98CWarGHWz1OenEXhU++Al+1n9KRhlH7wlQ07GJUDWxUTgX9IoJVd+sFXRGcN77Atr7hZwyjdLnHrDrvn1h2ycLd/RfRkiau7M5OiV/8PUW8EwFBR5xE3KmsY5RZug8Vfp7EODWo/xtu/Isrir90xPpKLSud5fvnqOPuyHeluvE0NzVY7RZDao+sSX58vZP0y1Nlt1beAtmetrwDhwPOWv73t6kudSRmnwVDcflfCWFyBMta+kVHG2tiYzJjrm1BEhgGgSowl5dNXSHp3LUHDrvaY6x+nwVBcYf3dUFKJMq4jV99mYzJjqm9EERmGEKim133TKHvlXW9cbd+uI7u4oiM7ToPhoiM7FICgQf25Ivs1rtj5B84/ssnaQHfKjbXZJmAornQe6w5cKdaBg/rTf9dG+u98lQuPes71JVsVp0Fvw9U748ZFuYx1myKm3EDziULrCdgjtjaqPX8AfUklKm2Uw/5F2edYXRP+Dmxv1R2uJ991JaVDrA3FTmoqLqrdxkWsuyrpWFc68O33XRkXhb643UZfXInKso/K6AgMZdLrDYayapQaz+7HqOKiaLXNsYuVqOKcxbvSrU1XpLKNZzfZMbdNoGbvkcua60u2Squh9aJt7lShduj0qLUaWm24rcVV1o5RVOZQWouraPzhrEc8e3aUHbu12ElNa6Mc2PY2uruySNu7nv4v349/eLBHXLU2itYL9ttUO3DVjtyL7TZBKTrCR1xJ2ufPMujvTxA6KMUjrmNNSb7Yx1qlta/31mIXx/h36VTv9ewJU9t2fXGcfd2OdDfevVfdzpBDW+j12zEUvfBep0xfny9k/TLUWcfFTxTFtqu2oaIoLhZF8WtRFJ8Akl19SRCEewVBOCQIwqHtdR4+nXC8S+Fs+nFRxFhexalRsym4aRHFz/yRxJeX4xcS6BlD6LjRDnfyndggisQumUnFWx9jburq3OTOt9s5W/qn6VgupzIWkDt1GTHzpiOolR5iu8qVbJqP5ZKbOZ/8qUuJuX+G51xfsl3kTufc9v8G9E9E99AdFD20yTOmG3iHG3BO89BLzE/K7cb+eBJrp7nv4fZ/Ar7gSR56ze1iW/JTON7VunJg6x6Yhmg0U/HRvsub60O28016dq7yC1SRuHgaZ9e97xHLI7gHNd22fxe37eLAiIUcSV+BvrSG5DV3eAp2tUm33Lb8F/z98I8I5sjkhyl48h2uemOpZ1QPjrHTVUkc9i1+0TREk4mKv3meXz47zj5sR36KeBc991cOD51L+Uf70M6e7AHU2fYu4fnicpPZfHn/XKbq7B2Y7wVBmC2K4ttAjiAIQ0VRPCQIQn/A5fssoihuBbYCfJ98Y4eUM5RUotT2at8JbTSGsiqnNsaSSlD44RcahKlGGuxm0kv/tnxfgL6oBFVSPC3H8zt11lhcgVIbbf1dGaexvhbUzq1ApY22chWhwZhq6gka1J/wyTcQt2oWirBgRLOI2Kqn8s+fdcpt264dWxuNwZFdXIFSF43BgW2r1vzzmJtbCOjfh2YPfDaUSNts52qcxNoDboF3XF+y9cWVqGy4Kmfc4kqXXGWchqStD3F2yQb0Z0s88tWObXOcVXEaDCVVTm0MxRZ2WFAHn71Vd7iefNeVDA6xVmo1HfO6RLJxd4y9Ucydk+k1MwOAxmP5dq+ESHz7AaJ6hzuJKq0GvWUfDRU1KGMipactMZEYKmtdcmNnZdlx1bpoGtq2qfOAq9OgL+l88KordoyF3XAs3z6/dRr0XrKjZ4wjcuJQfrx19WXJ9SVbOzuLuJnpANQfK0Cts82dKFodakN62tBuo9ZGoS+pIqBPHAG9Y0jbu97ydw2Dd63j2OSHMJTXuGBPQjtzooWdb8dWazXoHWu6A7vdZ0NFey4X/2U3A95Z5dJn3exJ6H4vceuO5aOOd8/t4LNNrFsvVlHx2X7Jh6P5YDaj1IRZJwmwVdysLGIt/jbkSDXV1jI45RZX2tW72qaWAXrNGEfUxCGcuGWNS1/b5Kvj7Mt25KeOd5sq/v41V77zMOfWu++8+eJ8IeuXp86euMwBxgqCUABcBfxHEITTwBuWz7qk5u9yUffVoUyIRVD6E37jGOp377ezqd+zn8hpUqMSPnkUjf/5DgBFVJh14LAyMRZVXx2GIs8uLpu+y7Pn3jSGut0H7Gzqdu8nwsodSYOFe/qWVZwaPYdTo+dQ8dYnlG/a7nGnBaApJw91kg5VosSOvGk0ddn2PtftPkDUtAkAREwZSf2/JbYqMdY6MF0Z34uA5Hj050s95qpsfI64aQx12Q4+Z9vEespIGixcZYI9V50cb52F6nJmS7HWokqMsca61oFbm30AzfSOsVaEBZOy7TEuPv8OjYdOeuxrmxpz8giwYUfdPIoaB3ZN9gE0M8YDEPnrG6j/xrtZy35qbk32AaJuHoWg8keVGENAkpbGY3kec9V9bWI9tWOsa7IPEGWJdeSvR1L/zXfd8rXsT59zInMpJzKXUr1zP5rpkk/Baf0x1TVZX/1qk6GsGnNDM8Fp/QHQTB9PzU5pH2t2HbTGRDOj/e/OVLptB99nLOP7jGVU7zhA9PRxAIS44ZoaWgixcKOnj6PazfbdqXTbDo5nLOO4hd3LS3YvG3b4uMHo5v+GU7PWYm52/764r7i+ZBe/vYOjE1dwdOIKKnccIOYWiRualoqpvglDmf3FqKGsBlNjM6Fp0riwmFvGUbnzIE0ni/4/e+cdH1WV/v/3zWQmvU5IMoEASQjuKjUIqPSSgKj4dQFdxVVUECS0UAQ7KooiKqxSbXzX364KuH7VdQUSsBdA2gYEUggkkEZ6n/77404mM8lMMpMQbnTv+/Xipck9cz55znme594z9xQO9nuIw0Pnc3jofLQFpRxLetTpoEXU3mtdTF+65zARd46xahuq69A109Y10464cwwlew8D2K1LCbt5GLVn8pzq5r+3l18mrOCXCSso+fIwETNE3cAhznUNNfUEDrHozhhDyR5Rt+TLQwSP7A+AT6wGQenpcNACULhjDycSl4sLvL88RDeLrr/FXkdtbaqpx99ib7cZYyiz6AaPG0T3Bf/D6VkvueRfUvWzlHnkSra3d4zGWi4k6Xrqsy61qS/F/ULm94fgyqJnQRACEKeGeQIXzWaza0/NOH7jAuA/9no0T81B8PCgfFcqlzfvJHzJTOrTM6nefwhBpaTHa8vwvjYWY2UNeYteRp9XRODkmwhfMhOz0QRGI8Ub/kH1gZZBbDY7fMlJwNghaJ62bIe8K43Lm3YSnmLRTRN1o19fatXNXbgOfZ69ueGL78ZU1+B0O2SjyYn2uCF0f3o2gsKDsp1pFL25i8il91D3nyyq0g4heCnp9fpSfK6LxVBRLW6HnFdEyB1jCZ8/HfQGzGYzRRs/pHLfwRb1ewiO+zJg7BCiLFsSl+9Mo3jTTiIsNjfqRr8m6horRJt1eUUE3zGO8EemYzYYwGSm6K8fUrXPve2QO1vbaHI89g4cN4Tuzzwkbof80f6mtk7PoirV0tYbUvC1tPX5BevR5RYRsXAGEcnT0eY0LXjMvnc1hmbfxDvrY4Cg8UOIXv0geCgo/SiNgjd2E7X8bmpPZFGZehjBS0nMxiX49ovFWFFN9vxX0eWKPtb/p+0oAnwQlJ4Yq2rJuGc1DZkXXWrrjuhqFk5HfddEMBrJXf0OVV+1XAPgzL8Cxw2hx+qmti58YxeaZaJfV1rauveGFHwsujnJ66261/24HUWAr9XerJmrachs+YDlrJ8Ber7wMEFjB4vbIS99gzrLlqDX7XuNU0niNBXfAXFN2yF/ddS61bQiJIA+W5ej6h6G7lIJWXNfwVhRY63b5CSPAPR+cY5V91zKm9atSPulvsrJxGUA+A2II7ZxG9OvjnLhibcBCJk8nN5rZuOpDsRYVUvtqRzO3vO8Xf2tZebeL84h2KKdbaPdP/VV0m2042y0z1u0B/2wCcFLiaFc/Baz5kgGOau2taImvW5na+tb8a+4tbMJGTcIU72WjCWbqTkh6g5Oe4VjE1cA4D8wzrpNbvmBY2Q//k6LeoYe3syxSSvttsk1O56MY6XP2ocs2jrOLtlEzQlxq9yEtFc4atW22Q75wHGr9jVvLMS/X2/MZjPavMtkrthmNwAxteJg8WsfInT8IIz1Os4u3kS1Rff6/a/wywRRN2CguB2yuDXwcTItuoLSkz9seAT/fr0x6QxkP/s+Fd+ftNat8nA+BSXmxdmEjBuMsV5LVsomai1tPTB1PScSlwPgNzCOeMv2vOUHjpFj6ePBP76Jh6qpj6uPZnBupf2WucZW4rkz+7k1mzszj7RmL3Ssva95ewU+cVGYTWa0Fy9zbuU2uzc23p6O14ZejftFQt6nrRveRajf+2aXngTnM2lBl2xHlwYuHcHZwKWzcTZwuRq09lDbmTh7sPw909oDbefqdsl47lSk9C+p+rm1gUtn898XzdLR2sClM2lr4NKZtDZw6Uxae4jvbNp6kO8spLJZKnvB+cDlaiAPXK4MXXXgIk22lpGRkZGRkZGRkZGRcQPXD6iQkZGRkZGRkZGRkek4XXjnrq6M/MZFRkZGRkZGRkZGRqbL0+lvXHR6RWdLOERvlEYXwNfL/ZPPrwSdvFypS1Kjc+NcmSuIlHPTFRKtftBLOF/aT6L50lKGlLdSIpsl7GepMOqk+g5POg9TKaT5tleq9UQA/ipp7s2CROsDG/TSTarx9ZGmrWV+/8hTxWRkZGRkZGRkZGSuJvJUsXYhTxWTkZGRkZGRkZGRkenyyAMXGRkZGRkZGRkZGZkuj2RTxQLHDqbHavFgwtIPUina/LHddUHlKR5C1D8OY3k1OfNfQXexGEVwALHbVuI7sA+luw5w8antThSc0+v5hwgen2A9yKwu/VyLMr79Y5sOMjtwlAtPiQdNhd56I92X3YVPfA9OTVlpPSzKFfxHJxD1jOXwy49Subx1dwube7y6FJ9+cRgrqsldsA79pabT4pVR3Yjft4nijR84PfzSoe6YBLpbDoEs+yiVy1ta6ka/1qR7YcE69BeL8RkYT4+1CyyFBIo2/IOqve4dQNlebVub+6ZuomiDezYDxK55kNAJg8XD2xa/SW16Tsu/b0Cs9UCxsv3HOPfku3bXuz8yldhn7uOnax+wO1CsLeLWPEDohASM9VoyFm+ixom29eC4/UfJfvI9AHotn0HkzInW06Zz1v6D8v3HXNKNWfMgIRabM53Y7DcglvjGQ9T2HyPHYnPPR/9M6OShmE0m9CVVZC1+E11ReYvPO6PPCw+gtth8ZpFzm//w12QU3ipK9x8l64n3rNe6PzSZ7g/ejNlgpDTtKOee/38u6fZ87iGCxg8RD6BMeYO6k47j2XoA5YEj5D4txrMi2J+4Lcvwig5Hm1dM9rz1GCtrXdLt9fxDhIwX7XWWR/xs8ki5TR7p+dR9hCRej0lnQHuhiOyUNzBW1bmkGzAmge7PzEZQKCj9cB/FW1rmzp6vpeDbvw+G8irxINuLxfiPHETUqvsQlJ6Y9QbyX9xBzY+un0odMEbM2YLCg9IPHefsXq+n4Ns/DkN5NeeTRd2AUQPtdC+9sIOaH9Nd1pVauyMx1UjUI1OJeeY+DrqZR6SKZ6l8uyM5u+fyO4mcOcGaN8+7kTel8q/2xrLvwHii1yZbCgkUbviASjfvzSBNP/uNGkL4E3MRFB5U7NpL2fZd9jYrPdG8shzv6/pgrKgmf8la9JeKCbxtLKGzp1nLeV0Tw/k7FqE93fJvlvl9I80bFw8PotfMJeu+Zzk9fgEht4/COz7aroj6z4kYKmr4ddQ8it/+jO6P3w+AWasjf/3fubRmR7ukg8Yn4B2j4cSIZHIe3UrM2ocdlot5aS45j27hxIhkvGM0BI0bDEDdmVwyZ6+j+udf3RP28CDquXnkzFpNZlIyQVNH49XH3uaQO5MwVtaQMW4uJe98SuSqWXbXNU/OpuabI27rdrfoZiQmE+xAN9Sie3bsXC6/8ykai27D2Vwyb0shc8picu57hh4vJIPCDZfpgHYjUU/NpvprN20GQiYMxidWwy83LiRz+Vb6vOy4n/u8PIfM5dv45caF+MRqCBk/2HpNFaUmZPQAGi5ebpf24RsXkrl8G31enuNUO2P5Ng5btQdZr13a/i+OTlzB0YkrXL75NuoevXEhWcu3EufE5riX55C9fBtHLbrBFpsvbf6U4+OXcWLiCspTjxC9dIbLNodOGIxPjIaDNywkY/k2+q5zbHPfdaLNB29YiE+MhlCLzcEjriNs8lAOj1vG4TFLydvymUu6QeMT8IqJIn3kfM6v3EKvtXMdluu1dh7nV24hfeR8vGKiCBqXAIAm+U9UfZ9O+shkqr5PR5P8J5d0g8cn4BOj4bglj8S2kkfOPbqF4yOS8YnREGzJI5XfnuDEuCWkT1xKw7l8ui+c5vDzLfDwoMfzczl3/7OcmZhMyNTReDXLnaF3JWKsrOH0mLlcfuczNKvE3Gksr+Lcg2s4O2kRuUs30PP1FNc0LbrRa+aSff+znJ6wgJCpDnK2RffX0WLOjnpM1DWUVZH94AucSVrMhZSN9Nrghq7E2h2NKRDzSHAH8sjVjmepfPtK5OxL27/g2MQVHHMjb0rmXx2I5fqzFzh721LOTllC9v2r6fHifPfuzUjUzx4eRDwzn4tznubclHkE3joGVZy9zUEzJmGsrOFc4mzKdnxCtxUPAlD1+decv30h529fSMGKV9FfKvrtD1rMpq79r4vSqqcLgrBIEITo1sq0B79B8WjPF6LLLcKsN1D+2XcEJQ2zKxOcNJyy3QcAKP/iBwJGDADAVK+l9vBpTFpdu7RDJg2jZPfXANQczUAR5IcyPMSujDI8BEWADzVHMgAo2f01IZOHA9CQdYmG7Hy3dX0HxqO7UIA+T7S58vNvCUwcblcmMHE4FR/vB6Dyyx/wv2mgzbUb0OUV0pCR657uIFFXZ9Gt+PxbApOa6SYNp7xR999NuuYGLRhF5xW8VJjd3LasI9ritRvQ5RaizXTPZgD1pKEU7/wagOqjmXgG+qIMD7YrowwPRuHvS7Wln4t3fo168lDr9bjnZpHz/Ptub9cWNmkoRTu/sdH2Q9VMWxUejKe/j1W7aOc3hE0e1qIudwi1sbmmHTYba+qt5Tx8vTC7seNR2OShFO0Sba460rrNVb9YbN71DWE3izZH3Z9E7hv/h1kn7qSlL6lySTd40jBKd38FQG0b8Vx75CwApbu/ItjS1sGThlG6S/x86a6vCJ5s75/OCJk0jMtu5pHLNnmk8psT1tiqPpKBSqN2Sdd3UDza800xVf75dwQ1yyNBicMp+1jMnRX//oGAEWJM1Z86h6G4DICGjFw8vJQIKtdeuvs2z9mft8zZQUnDKd1tqzvAopuDoah9ulJrdzSmAGKem8X5duQRqeJZKt++Ejm7PUjlXx2JZXODztrGHl6qdm1OJ0U/ew/oi+5CPvq8QtAbqPriW/wn3mhXxn/CDVR+kibWu+d7fG8c2KKegFvHUPWvb9wzWOZ3Q1tD9OeBg4IgfCcIwnxBELpdCVFlpBpdfon1Z31BKcpIdbMyoU1ljCaM1bUoQgI6rK2KDEVro63LL0UVGdqijK6gtNUy7uIZqUZfYGNzoQObI9ToCprbHIjg40W3edMo3viB27rKCDX65m0d0VJX36KtAwHwGdSXvvs20XfvG1x6crM1UXW2dqPNRe2wGUClUaPNt+nDgjK8miVWL40arU0/awvKrMk3NOl6tAVl1P56oR3aoXba2oJSVJrQlmUKnJeJenAyCQfW0/f1R/AM8nNRV91M17HNdr5tYzNAz1V3c/2RrXSbNorcdR+5pCvWG4r2kr09Xs1s9mpuc35TGd+4KIKG/5GEL19k0CfPEjAoziVdVaQanY3NYi6x11U2j+eCUlSW2FOGBaMvFqfP6IvLUaqDXNQNtcthHckj4XePp+LAUZd0lc3zSEGJg9zpKKbsc2fQlJuoP3XOOlBsC1WznK1zFMuRoW3qBk+5ifpTOS7rSq7dwZgKTboeXUEZde3KI9LEs1S+3dGcDY1581XiX5/vet6UyL86Gsu+g/pyTeqbXLP3r1x8wr17M0jTz8oINYbCJk1DYYnDZwJDgeXtpNGEqbrO+jzSSOCU0fLA5b+YtgYu54AeiAOYIcCvgiDsEQThfkEQ2j+KcHRMQItvoxwUuhJboQst623xJsFBmQ6Lt1fXbCYiZSYl736Kqa7hiui2aOtWytQfzyAjKZmsqUsJf2QGgpcb56Z0QDsyZSYl77TTZifVttR2XMbDR0X0kmlccOPBvU3xFu7tvG3yd+zj0PCFHJ2wAl1RBbGr73NR11GVrtncSO5LH/DLkHlc/vg7NA9Odk3XScUtQ9p5DAieHngG+3H05sfJfu59rn1raXtlWwgLrvihu7TXt5s5QtSiaZgNJkr++a2rwu3TtSniHR9N1Kr7yXtss4uajmXd1u0bTdRj95Hrjm4X1HY1pjx8VPRYMs2tLwCulHYj7YpniXy7IzkboGDHXg4PX8DRCcvRFZUTs/p+l3Sl86+OxXLd8QzOJi4gY+oywudPd+/e7LTuTu7ndmra+r33gGsw1WvRZbr/ZUCXw2Tq2v+6KG290zSbzWYTsA/YJwiCErgZuBtYDzh8AyMIwsPAwwBPBA/gT/697a7rC0pRRYVZf1Zq1Ogtr1utZQrFMvrCUlB4oAjww1jh+qJGWyJmTabbzEQAao9n4RUVRo3lmipKjb7ZgkVdQandtziqKDW6QtcXKTvCUFCCUmNjc6Ta+oq5EX1hCSpNGIZmNvsO6kvQzTcRuWoWikA/zCYzZq2O0r990aauvrAEZfO2Lm6pq2yjrbXZFzHVN+Ddtxf16Vku2dwRbd9BfQmachOax1y3WfPAZCJnTgCg+ng2XlE2fagJRVtory1+499UxksTiq6wDO9ekXj3DCfhwHrL79UM3reO4zc/hv5yhRPtSWhmTrRoZ9lpe2nU6Jpp61poN/mYvqTS+vuCv6fR7/1VTm2OfGAyERabayw2N/Zcoz3NbbbzbQdlAEo++Y4//r/HyXtlp1PtqAcmEXWvaHPV8Sy8urduc4v2tokrbX4ZJV8cBKD6WBaYTCjVgdaFtraE33+zXTyrbNpazCVtxLNGjc4Se/qSCpThIeLblvAQ9KWVOCNi1mTCLbo1x7PscpgqSt1i4XNbeSRsxlhCJl7P6buecarZHH1hszyiCWuZOwucx7MyUk3v7Y+Tu3QDutxCl3V1zXK2ylEsF5S2qhuz/TEupGxAd8F1XSm0r1RMefeKxKtnOINs8sigfes40UoekSqepfLtK5WzwT5vFv49jevef6xVbTtbJPDtjsZyI9os1+/NUucwfWEJnpFNmp6RYQ6fCTw13TAUiTZ7BPhisrE58JbRVH/xtUt6Mr9P2nrjYjf0NZvNerPZ/JnZbL4b6OnsQ2azebvZbL7ebDZf33zQAlB7IhOv3hpU0eEISk9Cpo6iMvWQXZmK1EOETh8PQMgtI6j+wfXdb5pTtGMPJxOXcTJxGeV7DhE2fSwA/gl9MVbVWaeKNKIvLsdY04B/Ql8AwqaPpXzvoebVukXdfzLx6h2FskcEgtKToNtGU5VmX2dV2kGCp4lJPOjmEdT8JNp87s5VnB01m7OjZlPy7mdc3rzLpUELQN2JTFQ2usG3jcUmpgkAACAASURBVKaqWVtXpR4kpFF3ygjrTkPKHhHWBX/K7t3wiu2OzmbHr87Uzr5zFWdGzubMSNHm4k1t21zw3h7rwszSPYcIv3MsAAEJ8Rir69AX2z8s6IsrMNbWE5AQD0D4nWMp3XuYujO5HOz3EIeHzufw0PloC0o5lvSo04cNUXuvdTF96Z7DRNw5xqptqK5D10xb10w74s4xlOw9DGC3NiTs5mHUnslzqlv43h5OTFzBiYkrKLOx2d+i68xmfxubyyy63jGR1nKhk4ZSn3XJqS5A/nt7+WXCCn6ZsIKSLw8TMUO0OXCIc5sNNfUEDrHYPGMMJXtE7ZIvDxE8sj8APrEaBKWnw0ELQPH/fsmppKWcSlpK+d6DqKePA8CvlXg21dTjZ4ln9fRxVFjiuWLfYdQzxM+rZzT93hFFO/aQnriMdEse6eZmHulmk0eCxg4mKvkOzs5ai6ne9fV6dScy8YqJQhUtxlTIbaOoSj1oV6Yq7RCh08TcGTxlBNWWmFIE+hH73tMUrPsbtb+cdlmzSdcmZ9/WMmdXph5CPd2xbtyOp8h/+X1qfznjlq4U2lcqpurO5HK430McGTqfI5Y8cryNPCJVPEvl21cqZwN262HUNw+nrpW8aYtUvt2RWFZF29+bvWO7o7tY1Kam1DmsIT3D+kyA0pPAW0ZTs99+N7SaAwcJukP8Qixg8kjqfrJ59hMEAm4eRdUXrr6hlvk9IrS24FoQhL5mszmjIwJHo293KBA4bgg9Vj8kbj/40X4K39iFZtk91P0ni8rUQwheSnE75H6xGCuqyUlejy5XDMzrftyOIsAXQemJsaqWrJmraci0T1J6o8Lp39T7xTkEjR2MqV7LuZQ3rVsa90t9lZOJywDwGxBHbON2yF8d5cITbwMQMnk4vdfMxlMdiLGqltpTOZy953m7+n299A51A8YOQfO0ZTvkXWlc3rST8JSZ1KdnUp12CEGlJPr1pXhfG4uxsobchevQ59kno/DFd2Oqa3C4NbCzrgwYO4Qoy5bE5TvTKN60kwiLblWa2NbRry3F57pYjBWiri6viOA7xhH+yHTMBgOYzBT99UOq9rm35WJ7tW2JWHI3xlrHNlc2eDnVjls7m5BxgzDVa8lYspmaE2I/D057hWMTVwDgPzDOurVm+YFjZD/+Tot6hh7ezLFJK+22MTU7nF/QRJ+1D1m0dZxdsomaE+LuJwlpr3DUqm2zHfKB41bta95YiH+/3pjNZrR5l8lcsc1uEKBoZdpi7NrZBFtszrKxeWDaK5ywsbmPRbfiwDHONeq+vRyfPlFgMqO9eJnsR7fbfXurN7duc/zahwgdPwhjvY6zizdRbbH5+v2v8MsEUTtgoLgdsriV6XEyLdqC0pM/bHgE/369MekMZD/7PhXfn7TW7efpfN54zxcetsZzztI3qLPE83X7XuNUkjjlzHdAXNN2yF8dJffJt8S2DAmgz9blqLqHobtUQtbcVzBW1FjrNrZic+8X5xBs0c22ySP9U18l3SaPxNnkkfOWPDLoh00IXkoM5aJP1RzJIGfVNrv6vZWObQ4YN4TuT89GUHhQtjONojd3EblUzJ2NMdXrdTGmDBXV4haqeUVELLyT8PnT0eU0bS6S/ZdnMDR7y2R2YnPguCF0f6YpZ1t107OosuTsXhtS8LXonl8g5uyIhTOISJ6O1lb33tUtdFujs7XrdM6n2XQkpmwZcngzJ5rlkbbozHhWeDifCtKZvq03Of++tCM5u68lb2KGhrxiMldsazHw8Vc5vjd3tn8JguOc3d5YDrljLOHzp4PegNlspmjjh1TuO9ii/gZ965NqOrOfg/0dT/P2G3M9EY/PBYUHlbv3Ubr1I8IW3UvDyUxqDhxEUCnF7ZCvjcNYWU1+ysviYn7Ad1h/ui1/gAt3tj6V+A8Z/279ZtVFqP903ZVYANFp+Nz+aJdsx1YHLlcCZwOXzqa1gUtn42zg0tl0cld2SVobuHQmbQ1cOpPWBi6dSVsDl86ktYFLZ9LawKWzcTZw6WycDVx+z7Q2cPm90trApTNpbeDS2TgbuHQ2zgYunU1bA5fOxNnA5WogD1yuDF114CJdBpGRkZGRkZGRkZGRkXER6YbjMjIyMjIyMjIyMv+NdOGdu7oy8hsXGRkZGRkZGRkZGZkuT6e/canQSbMGQUq09dKsr5Fq7UODWbr1RBF+tZJp/7ch5dqHOq00axCknOArVXsbTVLpSvc9WliwNHnELFFbw3/nmkiDQZp7lVYvjW6Qn3TrTC5VdvzA8PbyB8mUZa4G8lQxGRkZGRkZGRkZmauJWZ4q1h7kqWIyMjIyMjIyMjIyMl0eeeAiIyMjIyMjIyMjI9PlkXSqWPwLD6CeIB5+9OuizdSk57QoEzAghj9aDqwr3X+MzCfeA+C67UvwjYsCwDPQF0NVHYcnPNqldQHi1jxA6IQEjPVaMhZvcqjtP8DmYML9R8l+UtTutXwGkTMnWk8Uz1n7D8r3H3NJN2bNg4RMGIypXkfm4jepdaDrNyCW+MaDvfYfI+fJdwHo+eifCZ08FLPJhL6kiqzFb6IrKm/xeWf0fWEW6gmDMdZrOb1oC9VO2vvav863tnfGEzsA6Ld9cYv2PjRhZZua/qMT0Dz9sHjY5859lGzdbXddUHnSY/1SvPv1wVhRTd7Cl9FfKkbZPZz41C1oz4mnTNcfP0v+k5tctlVKbaltjnrGcrjqR6lcdqT96lJ8+sVhrKgmd8E69JeKrdeVUd2I37eJ4o0fODxotDV6Pf8QweMTrIeo1aWfa1HGt39s0yFqB45y4Snx0LrQW2+k+7K78InvwakpK60HsHVl3YAxg+mxeo54WN6HqRRt/tjuuqDypNfrKfj2j8NQXs355FfQXSwmYNRAolbdh6D0xKw3cOmFHdT8mO6ybuDYwfR8djYoPCj5IJXCTf9soRuzYQm+A0Tdc4+sR3dR7OPI5GmE3T0RjCZyn36Lqm+Ou6zbiBTt7TtyCBFPzAMPDyp376HsrV32NiuVRL68DO/r4jFWVJG/dC2GS8XgqSByzRK8r40DhYKqT/dTtn2nW/b6jRpC+BNzERQeVOzaS9n25tqe4kF914nxnL9kLfpLxQTeNpbQ2dOs5byuieH8HYvQnm7ZXs50I5606O7cS2lzXZUnUeuWW/PIpcVrrbHsdU1vIp9fiMLfF7PJzPk/Lcasc+3slPbqBk4di7qZvTn/47q9IF3+kjKmpGxvKZ4JZH4/SPbGRT1hML4xkfx8wyLOLN/ONetmOyx3zbo5nFm+jZ9vWIRvTCSh4wcBcOrhDRye8CiHJzzK5S8OcvmLlqfGdiVdgJAJg/GJ1XD4xoVkLt9Gn5fnOCzX5+U5ZCzfxuEbF+ITqyHEog1wafu/ODpxBUcnrnB50NKoe/TGhWQt30rcyw87LBf38hyyl2/jqEU3ePxgUXPzpxwfv4wTE1dQnnqE6KUzXLZZPWEQPjGR/HTDYs4sf4tr1j3ksNw162Zzevl2frphMT4xkagtNp98eCOHJqzk0ISVFH9xiMtfHGpb1MODqGcf4fwDz5A1aT5Bt43Bq0+0fZvcmYSxqpbM8Q9T+u6nRK6cZb2mu1BI9q2LyL51kdsP8JJpS23zc/PImbWazKRkgqaOdqxdWUPGuLmUvPMpkatm2V3XPDmbmm+OuKcLBI1PwDtGw4kRyeQ8upWYtY59O+alueQ8uoUTI5LxjtEQNE707bozuWTOXkf1z7/+JnTx8CB6zVyy73+W0xMWEDJ1FN7x9m2tvisRY2UNv46eR/HbnxH12P0AGMqqyH7wBc4kLeZCykZ6bUhxS7fnmrlk/OU5To1bSOjto/CO72FXJOzPiRgqazg58hGK3vqMHo/fB4B3fA9Cbx/JqfELybj3WXq+IA4E3EGS9vbwIOLpZC7OeYqcW+cScMtYVHE97f+u6UmYqmrImfQQ5f/7f3Rb9iAAAZNHISiVnJ86nwvTFhF81xQ8u4e7p/3MfC7OeZpzU+YReOsYVHH2/Rw0YxLGyhrOJc6mbMcndFshald9/jXnb1/I+dsXUrDiVfSXilx/qPTwIHL1fPJmP032zRbdZrEcPH0SxqoasifOpuy9Twi36KLwIGr9CgqffpNzUx4h996VmA3GTtet+uxrcqYuJGfqQvKXu2mvRVuS/CVlTEnY3pI8E3RVTKau/a+LItnAJWzy9RTu+haAqiOZeAb6oQoPtiujCg9G4e9D1S+ZABTu+pZuNw9tUVf41Bsp+uSHLq0LEDZpKEU7vwGg+qhzbU9/H6qPZABQtPMbwiYPc1nDEaGThlK882sAao5m4hnoi7KZrjI8GIW/r1W3eOfXqCeLNhtr6q3lPHy9MLuxe1m3yUNdam9PF9o7YuoNFLrQ3j4D+6K9UIA+rwiz3kDlv74lIPEGuzIBE2+g/OP9AFR++T1+Nw102aauqC2lzb4D49HZan/+LYGJw+3KBCYOp8Kq/QP+NtqBiTegyyukISPXbe2QScMo2f01ADVHM1AE+aEMD7ErowwPQRHgQ43Ft0t2f03IZPHva8i6REN2/m9G13dQPNrzhehyxbYu//w7gpLs80NQ0nBKdx8AoOLfPxAwYgAA9adyMBSVifoZuXh4KRFUrr109xsUj/Z8gVW37NPvCU6y7+PgpGGU7voKgPIvfiRg5ADL74dT9un3mHUGdHnFaM8X4Dco3i27pWhv7wF90efmo79YCHoD1f/+Bv8J9jHlP+FGKv8vDYDqvd/he6PlSyazGQ9fb1B4IHirMOv1mGrq3NLWXchHnydqV33xLf4Tb2ymfQOVn1i093yP740t4zng1jFU/esbl3V9HOgGTGimO/EGKv8p6lbZ6PqNTEB7NgftGfHbc2NFtcsPPx3RtSXw1jFUfe66vSBd/pIypqRsbymeCWR+X7Q6cBEEQSUIwn2CIEy0/HyPIAhvCoKQLAhCh/Yn9dKE0nCpxPqztqAUL01oizLaglLrzw35LcsE3/BHdJcrqc8p7NK6ACpNKNr8pnq1BaWomtWraqbdvEzUg5NJOLCevq8/gmeQn4u66ma6ZXhp1HZlvDRqdDa6uoIyVDZleq66m+uPbKXbtFHkrvvIJV2x3hAaLtnb47i9y5rK5JfhpbF/KHGnvZWRavQFl60/GwpKUEbY26uMsCljNGGqrkMREgiAKjqCuM83EvPBWnyHXueaoRJrS2mzZ6QafUFTTOkLS1FGttTWNZYxmjBW16IICUTw8aLbvGkUb/zALc1GVJGhaPObtHX5pagiQ1uUsfNtB2V+O7pqdLa6BaUt+zkyFH1+87a235o0eMpN1J/KwawzuKarCW3qP0BX6CB3RYba93FVHZ4hAS59tk19CdrbMyLMPqYKS/Bs1tae4WoMNjabqutQBAdSvfd7THUNxH33D+IO/I2yd/+JqbLGZW1lhBpDYZO9hkLH8WxwEs+NBE4Z7dbAxTPSxh5A78jmCDX6QhvdGlFXFdMdzBD97vPE/N9fCZ0z/aro2hJ4i3v2NmpLkb+kjCkp21uKZwKZ3xdtfd32nqWMryAI9wP+wD+BCcAw4P72SzvYv775xvKCozL2P4bfMcKttx7S6bpWr+MyYqH8Hfu48NrHYDbTe+WfiV19HxkpW1zQdVRlc5sdfM6mTO5LH5D70gd0X3gHmgcnk/eKq/O129fezYtE3HETRZ/86KKmA1y013C5jLMjH8BYUY13vzh6bX2SzMnzMdm8dfrNaF8tXYf950pMmYlImUnJu59iqmvneQPt1e7omUeS6Tqq0r385d03mqjH7iPr3tUdEm5xDojTOHbhs23KS9TebVXn6O/CjE//a8BkInv0TBSB/vT8+3rqfjwmvr1xhVbuA61q25TxHnANpnotuswLrmmKlbapKzj52wSFAp8h13J+2hJM9Vp6/u1FGk5mUvfTiU7VbcR7oGiv1i17kTB/SRlTErZ3V3km6Ap04elYXZm2Bi79zWbzAEEQPIFLQJTZbDYKgvD/AKfZSBCEh4GHAVIChnCrTywA3R+YRNS9EwCoPp6Nd/cwKjkLiN/4awvtF3xr80vt3gx4R6nRFjaNwgWFB+G3DONw4qpWjZBKF0DzwCQ0MydatLPwimqq10ujRmdTL4jfEtpqi2XEv09fUmn9fcHf0+j3vnP9yAcmEzFTtLnmeDZeUWqqrXWGttDV5pfavWFROSgDUPLJd/zx/z3e6sClxwNJ1vauOp6Nd3c1jX+58/Zu+sbFKyrUajM0tfehxMecatqiLyxFqelm/dlTE4a+uMxhGUNhKSg88AjwFac2AEad+N+Gk9nocgtRxXSnIT2rS2tLabOhoASlJsz6szJSbZ2S1KRdgkoTZtVWBPhhrKjGd1Bfgm6+ichVs1AE+mE2mTFrdZT+7QunehGzJtNtZiIAtcez8IoKo/G7bFWUGn2zjSN0Bc18O0pt51+uIpVuizqjmtpapVG37OeCUpRRYeibtTWIfROz/TEupGxAd8H1bypFW2x0I9Xom+cuSxl9gUU3UPQvVz7rCKnb21BUYh9TkWEYiktblPHUhGEoKrHGlKmimoBbx1L73S9gMGIsq6T+6K9494t3eeCiLyzBM7KpzTwjHcVzCZ6abhiKSu20Gwm8ZTTVX3ztns2Foj2NKCPDMDjQVUba5BF/sZ8NhSXUHU7HWC5uHlP7zS94X9fHpYFLR3Rt7a36l3v2wtXPX41IEVNWm69ye0v9TCDz+6KtNS4egiCogADAFwiy/N4LcDpVzGw2bzebzdebzebrGwctAJfe29u0sP3LQ0TOGA1A4JB4jNV16Ior7OrRFVdgrKkncIg4dzNyxmhK9vxivR4yuj+1mfl2rxQdIZUuQMF7e62L6Uv3HCbizjEABCTEY3CmXVtPQIKoHXHnGEr2HgawmwcadvMwas/kOdUtfG8PJyau4MTEFZTtOUT4nWMB8Lfo6pvp6i26/hbd8DvHUmbR9Y6JtJYLnTSU+qxLrdp88b191sVzl788bNfeTm2uabBr78t7Dluvu9PeAPX/ycCrdxTKHhEISk+Cbh1NdZr9JgrV+w8SMk1MpEE3j6T2p/8AoAgNtC5yVEZHoOodhT7X9Qc8qbSltLnuP5n22reNpirNfsFkVdpBgq3aI6ixaJ+7cxVnR83m7KjZlLz7GZc372rzpl+0Yw8nE5dxMnEZ5XsOETZ9LAD+CX0xVtWhL7a/CeqLyzHWNOCf0BeAsOljKd/r/oJOqXRtqTuRiVeMBlV0OILSk5DbRlGZal9nZeoh1NPHAxA8ZQTVP1r6OdCPuB1Pkf/y+9T+csYt3doTmXjb6IbePpKKZroVqYdQzxgHQMgtN1H9Q7r196G3j0RQeaKKDsc7RkPt8cw2NaVu74b0DJS9olB2jwClJwFTxlBz4Ge7MjUHfibof8QvpgImjaLuZ/Eh3VBwGd8bxDUBgo8X3gP/gO6c83ztSFtliSmUngTeMpqa/c21DxJ0h0V78kjqLDEligoE3DyKqi++dcvmege61c119x8k6E+ibuDkkdT9LOrWfHcU72tiELy9QOGB79B+6LJcW/fREd1GewPbYS9c/fzViBQx1cjVbm+pnwlkfl+09cblHeAMoACeAHYJgnAOuAH4sCPCpWnHUE9I4MaDf8VYr+P04s3Wa0P3r7NuMXx25dv88a/zUXirKN1/nFKbnbQi/sf96VpS6QKUpR0ldMJghv78BqZ6HWeXNO3elJD2CkcnrgAgc+VbTdshHzhu3T0s5qm/4N+vN2azGW3eZTJXbHNJtzztKCETEkj4+U1M9VqyljTZPDDtFU5YdM+tfIs+Ft2KA8esur2euBefPlFgMqO9eJnsR7e7bHNp2jHCJgzmxoMbMdXr+HVx09S2Yftftm5jeGbl25atD5WW9m7a2jHif25yr72NJvJXb6X3/z6H4OFB+a5UtJm5hC+ZSX16JtX7D1H+0T56vLaM+APbMVbWkLfoZQD8hvUjfMlMzEYTGI3kP7kJoxtz0yXTltrmZ7YS87dnxe1Ed6WJ2ikW7bRDlH+USvTrS+n71TaMlTXkLlznev2tULH/CMETEhj442ZM9VrOpbxpvdYv9VVOJi4D4PyqbcQ2bpP71VEqDxwFIGTycHqvmY2nOpBr3n+C2lM5nL3n+S6ri9HExae2E/f+anE75I/205CRR+TSe6hLz6Iq9RClH6XSa0MK1367FUNFNecXrAcg7P4pqHpriFx0J5GL7gQg+97VGEorW1O06uY+9RZ9//4MeCgo/SiNhow8opbfTe2JLCpTD1PyYRoxG5fQ7/stGCuqyZ7/KgANGXmUf/4D1x14E4xGLjy53e0pEpK0t9FE8fNb6PHOGvBQUPnxPnRZuagX/oWGkxnUfnWQyt170axbQczedzBWVlOw9CUAyv/xOZoXl9L7860gCFT+cx/ajPOuG2w0UfTcFqLfWQMKDyp3i9phi+6l4WQmNQcOUrlrL5pXlhOb+jbGymryU162ftx3aD8MhSXi4mt3MJoofHYL0e+uEbfJbdRdfC8N6aJuxa69RK1fTlza2+I2uRZdU1UNpe9+Qsw/N2A2m6n95hdqvj7chmDHdTtkr0VbkvwlZUxJ2N6SPBN0VdyeMysDILSYy9m8gCBEAZjN5nxBEIKBiUCu2Wx26ausAxF3/tf1jKcgjcmKKz2f20UazApJdAEi/Gol0/5vw2x2tIbg6lCn7dBeIL9JVJ4ubiV7hTGapOlno0m685CDAzuwhq0DmCVqa/jvfGYyGKS5V2n10uj6+egk0QXIrwpou1AnMaHoI+kCyw3qP3q2S0ehz13PdMl2bHMvTLPZnG/z/xXA7laKy8jIyMjIyMjIyMjIXHFc28RfRkZGRkZGRkZGRubKIO8q1i6kezcvIyMjIyMjIyMjIyPjIvLARUZGRkZGRkZGRkamy9PpU8X8FK6dzHyl8ZBogTyAp0Ka139KiRbz+hmlG/9W1XtJoqv0kO4VryCRb0u5OF8qpMwjUi2e9lJKk0cMRunauqLKRxJdhYR5RCrflnITBqlyp1QxVVnrLYkuQKhXOw8V/m9CnirWLuQ3LjIyMjIyMjIyMjIyXR554CIjIyMjIyMjIyMj0+WRdxWTkZGRkZGRkZGRuZqY5ali7UHSgUuv5x8ieHwCpnot2SlvUpd+rkUZ3/6xxDWehHzgKBeeegeA0FtvpPuyu/CJ78GpKSup/U+2W9o9n3uIoPFDMNVryUl5g7qTjrVjXl+Eh7eKygNHyH1a1FYE+xO3ZRle0eFo84rJnrceY2XbByEGjBlMj9VzxBOvP0ylaPPHdtcFlSe9Xk/Bt38chvJqzie/gu5iMQGjBhK16j4EpSdmvYFLL+yg5sd0l231H52A5umHxVOBd+6jZKv9UTyCypMe65fi3a8Pxopq8ha+jP5SMcru4cSnbkF77hIA9cfPkv/kJpd1RZsT6P7MbASFgtIP91G8paXNPV9Lwbd/HwzlVVxYINrsOzCe6LXJlkIChRs+oHLvz25pxzz/IMETEjDV68ha8ga16TktyvgNiKXPhgWif+0/Ss5T7wIQ/eifCZ00DEwm9KWVZC5+E31RucvaUvhX4NjB9Hx2Nig8KPkglcJN/7S7Lqg8idmwBN8Bon+de2Q9uovFAEQmTyPs7oniac5Pv0XVN8cdSXSazSG33kT3pXfhHd+DX295lDo34lmqPCJFH4PYzz1WzwGFB6UfOM4jvTek4NM/DmN5NTnzxZhSBAcQu20lvgP7ULrrABef2u6yrSDmkahn5oh55KNULjvKI68uxadfHMaKanIXrLPmkb5pm615pO7YWfKf3OyWtlS5E6Tzr/bGsyI4gLjtj+Jn6efcJ99yy16p/KuR31Lu7GhbtzemGlFGdSN+3yaKN35AyVufuKUNom+HjE/A2Ipv+9n4drmNb/d86j5CEq/HpDOgvVBEdsobGKvq2tSUMpZlfh9INlUsaHwC3jEaToxIJufRrcSsfdhhuZiX5pLz6BZOjEjGO0ZD0LjBANSdySVz9jqqf/61XdpeMVGkj5zP+ZVb6LV2rsNyvdbO4/zKLaSPnI9XTBRB4xIA0CT/iarv00kfmUzV9+lokv/UtqiHB9Fr5pJ9/7OcnrCAkKmj8I6PtiuivisRY2UNv46eR/HbnxH12P0AGMqqyH7wBc4kLeZCykZ6bUhx3VgPD6KefYTzDzxD1qT5BN02Bq8+9rohdyZhrKolc/zDlL77KZErZ1mv6S4Ukn3rIrJvXeT2oAUPD3o8P5dz9z/LmYnJhEwdjVczm0MtNp8eM5fL73yGZpVoc/3ZC5y9bSlnpywh+/7V9HhxPihcd9fg8Ql4x2o4dtMCsldsIfYlx/4V+9LDZK/YyrGbFuAdqyF4vOhf+Zs/5cSEpZxIXE5Z6hGil85wWVsq/+q5Zi4Zf3mOU+MWEnr7KLzje9gVCftzIobKGk6OfISitz6jx+P3AeAd34PQ20dyavxCMu59lp4vzAMP91JDR22uP5NL1pyX3Y5nqfKIJH0M1jySdd+znB6/gJDbHeSRPydiqKjh11FiHun+uBhTZq2O/PV/59KaHW7Z2qgb9dw8cmatJjMpmaCpox3nkcoaMsbNpeSdT4lcNct6TXehkKxbFpN1y2K3By2S5U4kvE91IJ7NWh35r/yDi8/vcE/ToiuJf1n4reXOjrZ1R2IKQPPkbGq+OeK+NuI90idGw3GLb8e24tvnHt3C8RHJ+MRoCLb4duW3JzgxbgnpE5fScC6f7gunuWSzVLEs8/uhzacTQRDiBEFYLgjCRkEQXhUEYZ4gCEEdFQ6ZNIyS3V8DUHM0A0WQH8rwELsyyvAQFAE+1BzJAKBk99eETB4OQEPWJRqy89ulHTxpGKW7vwKgtg3t2iNnASjd/RXBk4c1fX6X+PnSXV8RbPmbWsN3UDza84Xocosw6w2Uf/4dQUnD7MoEJQ2ndPcBACr+/QMBIwYAOLTk2AAAIABJREFUUH8qB0NRmWh3Ri4eXkoElWsvy3wG9kV7oQB9nqhb+a9vCUi8wa5MwMQbKP94PwCVX36P300DXaq7LUSbC9Dl2dicaN9WQYnDKfvY1mZR29ygA6P4GtXDSwVubgYTOnkol3d9A0DN0Uw8A/1QhgfblVGGB6MI8LX61+Vd3xBq6WNjTb21nMLXyy19KfzLr7GtLf5V9un3BCfZfy44qane8i9+JGDkAMvvh1P26feYdQZ0ecVozxfgNyjedYOvgM0NWRfbFc9S5REp+hga+9kmj3zWMo8EJw2nzJJHyr9oyiOmei21h09j0urcttd3YDw62zzy+bcENovlwMThVFjzyA/4X9E8cvVzJ0jnXx2JZ1O9lprDpzFp9e3Uvfr+Za37N5Y7O9LWHY2pwMQb0OUV0pCR67Y2iL592U3fvmzj25XfnLDeo6uPZKDSqNu2WcJYlvn90OrARRCERcBWwBsYCvgA0cBPgiCM7YiwKjIUbX6J9WddfimqyNAWZXQFpa2WaZ+2Gl1+U736glKUzepVNtcuKEUVKQamMiwYfbE4bUhfXI5S3fY4TtS0sbegFGWEfaArI0PRN5YxmjBW16IICbArEzzlJupP5WDWubbNtDJSjb7gsvVnQ0FJS90ImzJGE6bqOhQhgeLfHR1B3OcbiflgLb5Dr3NJ0167yWZ9QQnKyOY2q53a7DuoL9ekvsk1e//KxSc2W5OkKzT3L21BaYvEqtLY+4G2wN6/eq66hyG/bKPbn0aT+8qHbmhL4F+aUHQ2ba0rLEWlcRRPNm1dVYdnSIBLn21Tv4M2txep8ogUfSzWaZ9HRN2WeUTXRh5xF8/msVzoQDdCbe9f1bV2eaTPvzYQ8+FafIde65a2VLlT1JbIvzoQzx1BKv9q5LeWOztCR2JK8PGi27xpFG/8oN36Ktt+pGO+HX73eCoOHHVBU7pY7pKYTF37XxelrTcuc4DJZrN5DTARuNZsNj8BTAZed/YhQRAeFgThF0EQfvm/upbrCiyFWvzK3PzgAgdl3P7q3aG2o2rt6xUcaXfkYAUXNB3aa1PEu280UY/dR+5jbk61aFPXcRnD5TLOjnyA7NsWU/DC20S/vhwPf3fOO3ChDVuxue54BmcTF5AxdRnh86cjeCldV3ah/xyeSmJTJPelf3Dk+rlc/ue3aB642WVtSfzLgWiL6hzGnIufdV/+KtiMdHlEMnvb1nUcdx2TbXc7W/LImREPknXrEgrWvE30BjfziJS5U7L7VEfi+YrKXh3/ckO/a+XOjsi2P6YiUmZS8u6nmOo6cFaKK+3ogm9HLZqG2WCi5J/fuqDpqDqJnoNkfrO48p7NEzACXkAAgNlszhUEwelTpNls3g5sBzgY9Sery0XMmky3mYkA1B7PwisqjBrLNVWUusXiZ12zb8lVUWp0ha4vkLYl/P6b7bRVUU31KjUuaGvU6CyvKfUlFSjDQ8RvdMJD0JdWtqmvKyhFFRVmV5++uMyujL6gFGVUGPrCUlB4oAjww1hRLf6NkWpitj/GhZQN6C4Uumy3vrAUpaab9WdPTVhLXUsZg0XXI8DXqmvUif9tOJmNLrcQVUx3GtKzXNQuQalpslmpCUNf1NzmEqc2N6LNuoipvgHvvr2ob0U7ctZkImZOBKDmhOhfjTV5adToCu21tQWldn7gZdPHtpR88j1/fP9x8tZ/5FS7S/iXTVurItXom9nbWEZfYGnrQLGfXflsZ9vsDlLlEan7GMQcYZtHRN2W8axqI6bcxVDQLJYj1dZpG026Jag0YdY8YqvbPI94xXRvNZZtudq5U8r7lH2d7YvnjiCFf0kdV1K1dUdiyndQX4JuvonIVbNQBPphNpkxa3WU/u2LVjUjZk0m3NLWNcez7OMqSo3OTd8OmzGWkInXc/quZ1yyWarnIJnfF229cXkbOCwIwnbgJ+BNAEEQugFuP3EU7djDycRlnExcRvmeQ4RNHwuAf0JfjFV11le8jeiLyzHWNOCf0BeAsOljKd97yF1ZAIr/90tOJS3lVNJSyvceRD19HAB+rWibaurxs2irp4+jwqJdse8w6hni59Uzmn7fGnUnMvGK0aCKDkdQehJy2ygqU+0/V5l6CPX08QAETxlB9Y//AUAR6EfcjqfIf/l9an8545bd9f/JwKt3FMoeEQhKT4JuHU112kG7MtX7DxIybQIAQTePpPYni25ooHWRtjI6AlXvKPS5ricL0eYoVNERVpurUu21q9IOETqtpc2q6AjrYnxl9254x3ZHd7GoVb3CHXs4kbhcXFD/5SG6zRgDgH9CPIbqOvTFFXbl9cUVmGrq8U8Q13N0mzGGsj2HAfCO0VjLhSRdT33WpVa1pfav2hOZeNv4V+jtI6lo5l8VqYes9YbcchPVP6Rbfx96+0gElSeq6HC8YzTUHs9sU/NK2uwOUuURqfsYxH726m2TR6a2zCMVqYcIteSRkFtGUP3Df9y2tTl1/8m0zyO3jaYqzV63Ku0gwdY8MoIaJ3nEq3cUOrfzyNXLnVLepxrpSDx3VPdq+5fUcSVVW3ckps7duYqzo2ZzdtRsSt79jMubd7U5aAHRt9MTl5Fu8e1ubvp2NxvfDho7mKjkOzg7ay2metfWNUn1HNRlMZu79r8uitDi1WTzAoJwHfBH4KTZbHbbW2zfuDSn94tzCBo7GFO9lnMpb1q3iuyX+ionE5cB4DcgjtjGbSa/OsqFJ94GIGTycHqvmY2nOhBjVS21p3I4e8/z1ro9hNbt6vnCw1btnKVvWLdfvW7fa5xKWgqA74C4pi0Xvzpq3epQERJAn63LUXUPQ3ephKy5r2CsqLHW7alwPDcwcNwQuj/zkLgN4Ef7KXpzF5FL76EuPYuq1EMIXkp6bUjB97pYDBXVnF+wHl1uERELZxCRPB1tTtMiz+x7V2No9m2S0tPoUNd/7PVonpqD4OFB+a5ULm/eSfiSmdSnZ1K9/xCCSkmP15bhfW0sxsoa8ha9jD6viMDJNxG+ZCZmowmMRoo3/IPqAy1vBAaj8/FvwLghdH96NoLCg7KdaU02/yeLqjSLza8vxcdi84UFr6DLKyLkjrGEz58OegNms5mijR9Sue9gi/rrdM6nj8W8OJuQcYMx1mvJStlE7QmxjwemrudE4nIA/AbGEW/ZDrn8wDFyLP51zdsr8ImLwmwyo714mXMrt9m9sVF6tD7/szP9S3Di20HjhxC9+kHwUFD6URoFb+wmavnd1J7IojL1MIKXkpiNS/DtF4uxoprs+a+iyxUHg5qF01HfNRGMRnJXv0PVVy3nK5vNDifXXRGbgycPp9ea2XiGBmGsqqXuVA4ZM5+z1m1qRVuqPNKZfQygcOJjgeOG0GN1Ux4pfGMXmmViTFVa8kjvDSn4WPo5J3m9tZ+v+3E7igBfBKUnxqpasmaupiEzz65+padj3YCxQ9A8bdm6dVcalzftJDzFkkfSxDwS/fpSax7JXbjOmkciUmZiNhrBaKJow9+p3n+4Rf2t5ZHOzp06g8Kpdmf6l7M+ho7Fc/+ftqMI8LH2c8Y9q2nIvGhXvzPf7mz/Mppa/770t5Y7XWlrL6Xje3N7Y8qW8MV3Y6prcLgdcq229enVvV+cQ7ClrbNtfLt/6quk2/h2nI1vn7f49qAfNiF4KTGUi29Dao5kkLNqW5PNTp5HOjuWAQbnftr6zaqLUP+/q7ru6ADwuf+lLtmObQ5cOkprA5fOpK2BS2fibODS2TgbuHQ2rT1wdDatDVw6k7YGLp2Js5tvZ9PWwKUzaW3g0plImUdae6jtTJwNXDobKfNIawOXzkSqPgbpfLutgUtnIlXudDZw6WzaGrh0Js4GLlcDeeByZeiqAxd5LzkZGRkZGRkZGRmZq0kX3rmrKyPdVx8yMjIyMjIyMjIyMjIuIg9cZGRkZGRkZGRkZGS6PJ0+VUwn4XxWqfCTaN6yTi/NPG0p5yz7qtw/sfi3jsJDmmmxeoN0/dygl26utlT4SDQfXytRHpFyDZVSId18fKlweETHVdH971t/KlXu9FZKd0CjlPH8m0GeKtYu/vtGFTIyMjIyMjIyMjIyvznkgYuMjIyMjIyMjIyMTJdH3lVMRkZGRkZGRkZG5mpilqeKtQdJBy6xax4kdMJgTPU6zi5+k9r0nBZl/AfE0ndjMh7eKsr2H+Pck+/aXe/+yFRin7mPn659AENZdZfWDhw7mB6r54DCg9IPUina/LHddUHlKR7s1T8OY3k1OfNfQXexmIBRA+m+6j4ElSdmnYGLL+yg5kfXT+5tr64iOIDYbSvxHdiH0l0HuPjUdpc1ben53EMEjR8iHiiW8gZ1J8+1KOPbP7bpQLEDR8h9+h0AFMH+xG1Zhld0ONq8YrLnrcdYWdumZsCYBLo/MxtBoaD0w30Ub2lpc8/XUvDt3wdDeZV48OXFYnwHxhO9NtlSSKBwwwdU7v3ZLXul0vYfnUDUM5bDzD5K5fLW3S10e7y6FJ9+cRgrqsldsA79pWLrdWVUN+L3baJ44wcODzNzhtT+JVUekUo3cOxgej47GxQelHyQSuGmf9pdF1SexGxYgu+AOAzl1Zx7ZD26i2I/RyZPI+zuiWA0kfv0W1R9c9wlzY7oKoIDiNv+KH6Wfm48LNBdOpJHQm69ie5L78I7vge/3vKo9VBDV2yWyrel0g4YI+oKCg9KP3Ss2+v1FHz7i/18PrnpPhW16j4EpSdmvYFLL7h/n5LCr0WbpcnZUvqXVDZL5V8yvx8kmyoWMmEwPrEafrlxIZnLt9Ln5Ycdluvz8hwyl2/jlxsX4hOrIWT8YOs1VZSakNEDaLh4uetre3gQvWYuWfc9y+nxCwi5fRTe8dF2RdR/TsRQUcOvo+ZR/PZndH/8fgAMZVVkP/gCpxMXc37pRnpvTHHd2A7omrU68tf/nUtrdriu14yg8Ql4xUSRPnI+51duodfauQ7L9Vo7j/Mrt5A+cj5eMVEEjUsAQJP8J6q+Tyd9ZDJV36ejSf6TSzb3eH4u5+5/ljMTkwmZOhqvZjaH3pWIsbKG02Pmcvmdz9CsEm2uP3uBs7ct5eyUJWTfv5oeL84HhRthIpW2hwdRz80jZ9ZqMpOSCZo6Gq8+9rohdyZhrKwhY9xcSt75lMhVs+yua56cTc03R1y31aIrpX9JlUcky18eHvRcM5eMvzzHqXELCb19FN7xPeyKhP05EUNlDSdHPkLRW5/R4/H7APCO70Ho7SM5NX4hGfc+S88X5oGH6/7VXl2zVkf+K//g4vM7XLezGR3NI/Vncsma8zLVP//quqiUvi2VtkU3+/5nOT1hASFTHeha8tevo0XdqMfs71NnkhZzIWUjvTa4d5+SxK8t2lLlbCn9S0qbr7p/yfyukGzgop40lOKdXwNQfTQTz0BflOHBdmWU4cEo/H2pPpIBQPHOr1FPHmq9HvfcLHKefx/M7u1SIoW236B4tOcL0eUWYdYbKP/sO4KShtmVCU4aTtnuAwCUf/EDASMGAFB/Kgd9URkADWdz8fBSIqhce1nWEV1TvZbaw6cxaXUuaTkieNIwSnd/BUDt0QwUQX4ow0PsyijDQ1AE+FB75CwApbu/InjysKbP7xI/X7rrK4InD29T03dQPNrzBejyLDZ//h1BifafC0ocTtnHos0V//6BgBEDATA36MAovr718FKBmxvgSKXtOzAe3YUC9Bbdys+/JbCZbmDicCo+3g9A5Zc/4H/TQJtrN6DLK6QhI9cte6X2L6nyiFS6fo3+ZWnvsk+/JzjJvp+Dk5pipvyLHwkYOcDy++GUffo9Zp0BXV4x2vMF+A2K73RdU72WmsOnMWnbvwNgR/NIQ9ZFGrLz3dKU0rel0vZtrvt5S92gpOGU7rbNX033KUPjfSqjPfepq+/XTTZf/ZwtpX9Jdp+SyL9kfl9INnBRadRo80utP+sKyvDSqO3KeGnUaAuaymgLylBZyoQmXY+2oIzaXy/8JrSVkWp0+SXWn/UFpSgj1c3KhDaVMZowVteiCAmwKxM85SbqTuZg1rm2zeGV0m0vqkg1Opu2FvVDW+oX2PZHKSrL36gMC0ZfXC5+trgcpTqoTU1lpBp9ga3NJQ5sVqN3YrPvoL5ck/om1+z9Kxef2GxN0q4glbZnc91CB/0coUZX0Fw3EMHHi27zplG88QOX7bS1RVL/kiiPSKcb2tSH8P/ZO/f4KKq7cT+TzW5uhFw2JNmEAEkgvvUGBBGVi0BIQFvxbQFtpS+CIigBIQEUW7l4QRRRsYhcrNbWX6uCtlVrFQIoVm0JEIiRCrkQEiA3srlfd7M7vz9ms9nd7GY3gTAB5/l8+Gh2zs4z3zPnfGfOzswZDGV6NDr7/qSJDLXfz3VNeIcEevTd3vBeCi42j/QEOdu2XG6Ng9dQqkcd0dnrKn+1E3znbTSf8Pw4JVe7luKRJ2fL3b7kiFmu9tVXEc1in/7XV5Ft4OJ0DnnHXx5dlPHy0xCzbAZFG9+/ctyeOJ0VsinimxBD9G/mUPzE65fVe1F44Bec7ZBuXkVzK+20f13H3HQ8l1PJi8mdvpzwRTMRfLrzDhGZ3E7WKXrkFYlIm03lWx9hbmrxzGW3TiefXcb2JVcekS9/OdvP7jdOKuPBd3vFewmQI4/I2bblcnvUrt0fp6Ke6OZxSrZ27fz7lydnO/nssuXOPhxzr7QvhauJLq+zCYIQBDwB/C8wwPJxBfAR8LwoijUuvrcAWACwInAk0/3jANDNm0bk7CQA6o8X4BPVMdLW6EJpLauyW09rid7uV0wfXSiGsip8B0fiOyicxAObLJ9rGbl3I8fveALjBaebJKsbpF9TNFFh1r/VOq319i9rmTKpjLFMDyovVIEBmGqkB3bVkVri3niCM8s2Yygqc+m51N6eEH7/HQyYnQxA4/F8NDZ1Lfmr7cobSvXWX6JB+jXbYNlGY2UN6vAQ6WpLeAhGfa1bv7GsErXONuawzjGXVqJ2E3Nr/jnMzS34JgymOSffo9jlcreVOngjtdbL6rbbptGF0ebg9R+RQNAdtxG5ai6q/gHSry2tBvR/+tR9vDK0L7n6stw5BNr7Skd9ayK1GB287WWMpZb67u+Pqabeo+/2hrenXMo80hPkaNtyuw0OXo1Oi7HCMX/pXeYvdaSW2J1PUJTWveOUXO0a5MvZsrYvmWKWq30pXF24u+KyC6gGJoqiqBVFUQtMsny229WXRFHcKYriTaIo3tQ+aAEo/cPnHJuykmNTVqL/PJPweyYCEJg4DFN9E8YK+4O2saIGU2MzgYnS/arh90xEv+cwTSeLOXT9gxwevYjDoxfRWqrnWMpjXR705XQDNGbn4TNEhyYmHEHtTcj08dRmZNqVqcnIJHTmZABCfjqW+m++A0DVP4D4P66m5Pl3aDxyskvPpfT2lIo/fsaJlHROpKRTvecQ2pmTAAhITMBU12S99asdY0U15oZmAhITANDOnETNHmkba/YeRjtL+r52VsfnXdGUnYdPbBSamAgp5rvGU5dxyK5M3b5MQmdIMQffOZb6b6WYNTER1gcN1dED8I2LxnCu3OPY5XI3fZeHz5Ao1AMlb9BdE6jbZ19XdfsOETxDOvEOumMsDf+WvKfvWcWp8fM5NX4+lW99zIXXd3s0aAF52pdcfVnuHAJSffvGdtR36N3jqHFS3+19JuSnt1H/TY7189C7xyFovNHEhOMbq6PxeJ5b58V6e8qlzCM9QY62Lbdbyl823rs6e2szMtHO7Jy/VP0DiH97NSUv9Ow4JUe77oj58udsOduXbMcpmdpXn8Vs7tv/+ihCp9tJbBcKwilRFK/p7jJb/hU506UgfsN8QiaNwNzcSu6y12nIlqaoHLnvRY5NWQlAv+Hx1ulEqw8co+A3b3Zaz+jDr3Ns6uPdmg65N90BaucPofafNIqB6x6UpgF8fz9lW3ajW34fTd/lU5uRieCjlqY+vD4OU009hambMBSXE/noLCJSZ9Ja2PFwaf7sdbR5cPXhYrwA1327E1WgP4LaG1NdI/mz19GSd9Zu/SZz1+PfQesXEDRxpDSNafoW61Sk1+19mRMp6QD43xjfMY3pF1nWKVNVIYEM3b4CTXQYhvOV5C98EVNNg3Xdam+TU2fgpFFEr5mPoPKiatc+yl/bTWS6FHPdPinmwa+k43ddHG019dJUj2fLCfn5RMIXzQRjG6IoUv7qe9TuPeTU4Yredqu8nHepwImj0K2xTIe8ex8Xtu4iPG02zTl51O/LRNCoiXklHd9r4zDVNlC8ZCPGs/YHnPClv8Lc1OJ0OmRjm/P93NvtC6DR6PpWBLnySG97/byd378dNHkUMeseAC8V+vf3UbrlA6JW/IrG7HxqMw4j+KiJfXUZ/pb6Llj0krW+dUtmor13CphMFK97k7ovsjyK9WK9N/x7J6pAP+t+zr1vHS155+zWL4rO7iPp4GLySPC0MQx+dj7eoUGY6hppOlFI7uynretWeTk/SF+Otu2K3nY7vd3R4o1e2+G15q+cfOos3sGb0/C35K8ziyVvxJLOx6mCX3c+TpnMzsWXo117q5zv597O2WYXMff2PvZycay4HDG76s+93b4ARhZ/1HUy6SM0bV/adx8kAfwffrVP1qO7gcteYB/wR1EUyy2fRQBzgWRRFKe4E3Q1cLlacTVwuVpxN3DpTVwNXK5mXA1cehtXA5fLQVcDl6sVVwOXqxV3A5fexNXA5WrG1cClt3E1cLkcuBq49DauBi69TVcDl95Gzv6sDFwuDX114OJuLrl7gVXAQUEQwi2flQMfA7N6c8MUFBQUFBQUFBQUrkrEH98PJpeCLgcuoihWA49b/tkhCMI84A+9tF0KCgoKCgoKCgoKCgpWLubej6cu2VYoKCgoKCgoKCgoKCh0gbvpkF1NYSEAEZ4IAn16/lbsi0HO+yvlQq77d71lvNxZ0+Iri7cPv5up11AJ8gUtVx6Rkx/jffFy0diqkXsTfjQIMuaRNpmex/TXyPPca22LjyxegBD/HrwX7MfGj/FE4hLg7hmXCGAq0vTHtgjAt72yRQoKCgoKCgoKCgoKCg64G7j8A+gniuJxxwWCIHzZK1ukoKCgoKCgoKCgoKDggLuH8x/sYtl9l35zFBQUFBQUFBQUFK5y+vBLHvsy7q649BqBtycSvXY+gkqF/r29VGz70G65oPFm0Mtp+N8wlLbqOunlR+cq8B8+jJgNqZZCAmWb36V2z3+64R3JwHUPSS8/ei+D8tc7ewe/kob/DfG0VddzJlXyBo4fTtSqOQhqb0RjG+fXv03Dt917M7Rc7n4TEolaa3kx4fsZXNj+QSfvwJfS8bs+HlNNPcWLN2I8X2Fdro4awLC9W6l49V2nLybs0n17ItFrHgKVF1XvZ3BhW2d3zMsd7qLFGzGes3cnZGylfHP33fHPziM0KRFTcyu5S7fSkFPYeftujOMaywsCq/ZnUfCkNFHe4BWziJw9BaO+DoDCDX+hev8xj91D189Da3GffNS1+39+l4rKV4N+fxb5v+2YpC/6wWlEP3AHYpsJ/b4sTj/z//q0FyDu2QcITRqJudnAqaWv0ejCnWCt72OcfvItu+XRj0wnbu0c/n3tPI9eBClfHpHHC9B/opRHUHmhf9d5HhmyOQ2/G+IxVddTuEhyq4IDidvxOP7Dh6LffYBzq3d2yytnzHK6Y599gBBLu85z0a4DboxjWPuLRvcfo9ChXUc9Mp3YtXM45GG7ltstm/eZBwhOSsTcbCB/2RaX3qGbF+Plq6FmfxaFqyXv4NVzCEm5CdHQRktRGfnLXsNU1+SRF2DwMw8SMlnKnQVpr9GUc7qz+4Y44jcvsbxQNoui1dILZQetnkNI8k2YDW20FpVTkLbFI3dPj49+w4cxcMNiSyGB8s1/oa6b7Rrkydn9JiSiW7NAOh/ZtZdKZ+cjm9LxvX4oppp6zi55AeP5CtTR4QzL2Ebr6fMANB8/RcmTW7sds8KVjzxPqnl5MfCZhZy+/ylOTkklZPoEfIbF2BUJvTcZU20DP9y+kAtvfoxu1f0ANJ8q4tRd6Zy6cxkF969j4HOLQOVhGF5exDy7kIL7n+KHpMWETB+Pr4NXa/H+d8LDVPz+Y6KekLxtVXUUPLCekylLKUp7lcGb07odsyxuLy+inn6YwrnryEtJJWj6BHyG2ntD7knBVNtA7qSFVL75EZGr5tot1z05n4aDR7sXr8UdbXHnJqcS7MQdanGfmriQC29+hM7BHbV6PvVfdt8dkjQSvzgdh29dQt6KHQx94SGn5Ya+8BC5K3Zw+NYl+MXpCJk8wrrs/M5/kDVlJVlTVnZr0BKaNBK/WB2HbllC7oodJGx07k7YKLkP3bIEv1gdoRZ38NjrCJs2msOTlnP49nTObvu4T3uho76P3LqEvBXbGfrCAqflhr7wEHkrdnDEWt8jrcs0UVpCJtxIy7kLnkllzCOyeC3umGcXkj/nKX6YvJiQu53kkV8m01bTwH/HS3kk+jeSW2w1ULLpz5x/9m3PfX0kZrnc7e0669Yl5K/YTryLdh3/wkMUrNhBlqVdBzu06+DutGuZ3XJ5gycn4hun49htiylYuY245517455fQMHK7Ry7bTG+Nt6ar7I5PnEZ2UnptBSUMHDJL7rl9ovVcXxsKoWPbSdug3N37PMLOf3YNo6PTcUvVkfwJMld+1U22ZOWkTMlnZbTJUQvmeFeehHHx5ZTxeTdlUbenUspnLOWgetTu9enkC9nRz31CGfmrSV/6iKC7rrd+flIXSN5kxegf+sjIh+fa11mKCqj4GePUvCzR5VBy48YWQYu/iOG0XqmFMPZckRjG9Wf/Iug5DF2ZYKSx1D14QEAav75DYFjhwMgthjAJF1e8/LRQDcmZZC8ZRiKbbwpN9t7U8ag/8DWeyMAzScKaSuvAqAltxgvHzWCxvMLVnK5/YcPw1BUitFS17WffEV/h7runzyGmg/3A1D72TeHcoTyAAAgAElEQVT0u224zbJbMJwtoyW32ONYbWM2FHXs55pPvqJ/ioM7ZQzV7e5/OrhTbsFQXEZrXvfdYVNHU77rIAD1WXl49w9AEx5sV0YTHox3Pz/qj+YCUL7rIGHTbu60rm67p42mfLfkrjvatbvuiMW9+yBhd0juqPtTKN7yd0SD9OZ0Y2Vdn/YCaKeOpmLXl0B7ffujdnCrw4NR9fO31nfFri/RThttXR7/9FwKn3kHRM86tbx55PJ7AQIc88jHnfNIcMoYqix5pPrTjjxibm6l8fAPmFu7P0ObnDHL6Q61adcNPWzXsU/P5Uw32rXcbtm800ZzwZK/Giw526k30J8Gi/fC7oOEWnJ27cFs676uz8pFE6X12B0y9WYufNAecy6qoADU4SEO7hBUgX4d7g++JGTamM7uo7lodO7dF3N8FFtarT7BR4PYzbYF8uRsv+EJtNqej/zjKwKTb7ErEzjllo6YP/uaAJtzAgUFkGngoo7UYiyttP5tLK1EHantXKbEUsZkxlTfiCokEAD/EQlck/Ea1+z5Hed++7q1A7tDE6nFUNLhNZTqUUc4ekNdetsJvvM2mk8UWk/w+rLb27Guy/Sd6zpCi6HU0dsfwc+HAQ/PoOLVdz0Ns9N6jSW2+9lJzBHO9nOHu7yHbo0ulNYSvfXv1lI9Gl1o5zKlrstEPTCNxAObSHjlEbyDAjx2++hCaT1vv14fB7ePo7uko4x/fBRBY35C4mfPMeJvTxE4Ir5PewE0Oq1dfRtKq/BxOHj76LQO9V1lPcCHptxEa2kVjf8t8tgpVx6Ry9u+XoNjn+rkDu0o4yKPdBe5Y5bL7diuW120a0Opfdu3bdeG0iqautGu5XbL5o0MpdWmbUv52N6r0WkxOOb1SPscBxD+yySqD3h+lVxj22cAQ0nn9WoiQ+1jdlIGIPxXk6k5kOXWeTHHRwC/EQkk7N1Kwp4tnH+ye+0a5MzZHVdn2korncfcXsZkxlzfZI1ZExNB/CevEvvuBvxHX+ext89iNvftf30UeW4Vw8n7CRxH7IKzMtJ/mo7ncip5MbnTlxO+aCaCj7rH2u54AXwTYoh6Yg7FT7zumVNut5N1dvp1xqlXJCJtNpVvfYS5qYfzsbtYrydlItNmU/nmpXZ7vn0lb+8lc8wSspJWYiivIW7dnO7IXa22S3f7fhG8vfAODiDrjt9Q8PQ7XPtGeh/3Oq/KzvvaeRkvPw0xy2ZQtPF9j30uV3g58ohsXudqJzvZpbvnyBhzH6vvzvnTyfcs7XrgshkUd7tdy+yWySt4cLxw+mYjh02LXjoD0WSi8sOvuiN363aR5Oz+inp0BmKbmcq/euC+iOMjQPPxXHJTUsmfnk74I7O62afkytk9d7ZdqOLUuHkU3LWU0vW/J+aVFXj187t4v8IVR48fzhcE4TNRFO9wsWwBsADgydAbmdFvsN1yY1klal2Y9W+1Lgyj5VYoa5nSStRRYRjL9KDyQhUYgKnG/sGv1vxzmJtb8E0YTHNOvtttNpTq0UR1eDU6LcYKR6/epVcdqSV25xMUpW3GUFTm1tcX3G2lDnUdqbXedmb1llWi0YXR5uD1H5FA0B23EblqLqr+AYhmEbHVgP5Pn3rkNpZJ+9DqdhZzmfP97D8igaA7b0P3hOdu3byp6GZPAaD+eD4+NrcK+Oi0GMrs3YYSvd0vTFIZ6ZVFxspa6+elf97H9e+s6jLWqHlTifq15K47no9PdNfuVkd3VIe7taSKyk8PSXEcywezGbW2v3WigL7gBdDNm0bk7CSp/PECu/rW6EJpdefWhWIoq8J3cCS+g8JJPLDJut0j927k+B1PYLxQ49QN8uURubzSeu3ziFqn7ewuk8p05e4ussZ8md2R86YRYWnXDZZ23b6m9jZrt94S+ysDGpt27TMonBE27XrE3o1kd9Gu5XLL5p07jQhLzm7IzscnKszG6yR/lertbgHz0Wkx2LSFAbMmEjplFCfuWdfJ5UjE3GmEz062xJxvf3yO0mIot399ncHhCpDGJncChM2aSMiUm/jh3rVu3XBxx0dbWgs871Py52w9at0A69/eujAnMUtl2s9HvAL9rTGbDNJ/W74vwFBchiY2mhYP84jC1UOXV1wEQUh08W8UMMLV90RR3CmK4k2iKN7kOGgBaMrOwyc2Ck1MBILam5C7xlOXcciuTN2+TEJnTAYg+M6x1H/7HSBdKmx/CE0dPQDfuGgM58o9Clby6tDEhFu9tRmZdmVqMzLRzuzsVfUPIP7t1ZS88A6NR0565OsL7qbv8vAZEoV6oFTXQXdNoG6fvbdu3yGCZ0jJLOiOsTT8W/KevmcVp8bP59T4+VS+9TEXXt/t8aClPWaNjTv4rgnUOcRcl3GIkHb3nWNpsMRccM8qTo6bz8lxkrtiq3t36R/2WB+m139+mIh7bgcgMHEYbfVNGCrsE6qhogZTYzOBicMAiLjndir3HAawezYk7I6baTx5tkt3yR/2cCRpJUeSVlL52WEiZknu/qNcu9samuk/yuKedTuVn0vuys8yCR53AwB+cToEtbfLwYNcXoDSP3zOsSkrOTZlJfrPMwm/ZyIg1bepvgmjg9voUN/h90xEv+cwTSeLOXT9gxwevYjDoxfRWqrnWMpjXR4AQe48cvm9AI3ZefgMsckj0zvnkZqMTEIteSTkp2Op/+Y7j9fvCjljvtzusj98TvaUlWRPWUmVTbvuZ8kjrtp1P5t2XWVp14evf5Cjoxdx1NKuj7tp13K5ZfO+/TnZySvITl5B1WeZDLDkr6685oYO74BZt1NlyV/Bk0YQvfh/+WHu85ib3T/HVf725+QkLycneTnVn2cyYGZ7zAmY6powVtgPXIwV1ZgaWuiXmCC5Z06keo/U94ImjiQq9eecmrvBIzdc3PFRPdC+XfvERWOwmY3TFXLn7Obvcu3PR342gfp99n25fr9NzHeMo9FyPqIK7Q9elphjItAMicJY3L0fkPscoti3//VR3F1xOQwcxPnFu2Ann3mGycy5NTuI+9M6BJUXVbv20ZJ3lsj0+2j6Lp+6fZno389g8Cvp/OTgDtpq6ila/CIAATf9hNhFq8HYhiiKnHtyO6ZqD39NNJk5t3on8e9IXv37+2nJtXhz8qnLsHg3p3HtV9tpq6nnzGLpF4Ww++9EM0RH5KP3EPnoPQAU/Hodbfrarozyu01mStZuJ/ZPT0nTD+7eR2teMeFps2nOyaN+XybV72cQ80o6CV/swFTbQPGSjZ7F5Il7zXbi/vQUqLyo3iW5Iyzuun2ZVO3KIObldK75cgemmkvnrtqXRWjSSEb/Z4s01eOyjhlIEve9SNaUlQDkPf5Gx3TIB45bZw+LXf1/9Lt+CKIo0nr2Ankrd3TLrU0ayZhDWzA1Gzi1tMN90/4XOZLU4f6f37VPM3mcKou79N0v+J/NjzD64EuYDW2cfNSz2VPk8gJU78siNCmRm/7zGubmVnKXddzOOHLfixyz1Hf+429Yp9asPnCsW7O1dULOPCKH1+I+u3onQ/+ffR7RLZfctRmZ6N/LYMjmNK7913ZMNfUUpm6yfv26b3eiCvSXTpSmjiF/9jpa8roelPeFmOVyV+/LIiQpkURLu863adfD971ItqVdn378DYZa2nXNxbZrmd2yefdnEZyUSOK/t2JqbiU/rSP/DM/YRHbyCgAKVu1kmGU65OoDx6zPk8Sun4+XRs11760BpAf0Tz/u2ZTfNfuPEpyUyIhvX8dsmQ65nRsyXiIneTkAhat2WKdDrvkiy84t+Kj5yfvS1ZaGo7kUrnJzzLiI42PA6GsJf2QmYlsbmEXOr96OqdrzyVRAvpxdsm47Q/74NIKXF9W7M6TzkWWW85H9mVS/v5eBLy9n2IGdmGobOPvoC1LMN19P+LLZiCYzmEyUPLkVU21Dz7dF4YpF6Go2CkEQvgd+LopinpNlZ0VRjHHyNTuOD54uy7BNFJ3eDXtV462S52EqOQfmNS2+snjNfffHiF5DJcgXdKBP92fCutIxm+XJYV5eP77G3diqkXsTfjQIMuYRL5nc/hqjLN7aFh9ZvAAh/j18PvUScP3pf1wRJ4BNmxf26WTrv2xHn6xHd1dc1uH6drIll3ZTFBQUFBQUFBQUFH4E9OGZu/oyXQ5cRFH8oIvFIV0sU1BQUFBQUFBQUFBQuGRczHTIT12yrVBQUFBQUFBQUFBQUOiCLq+4CILgajoaAYjwRNBmkudVMSazTK+okZFmY49nt75oZLt32Fuee4dVMj4HINczRXI++yDXM2tyPr9llilmU9uPL3fK9QyCyku+W0XkOkbK9SwmQFNr9951csm8Bnm8ai8zvmrPX5R9Kalukuf50yuKH+PDspcAd2e6EcBUoNrhcwH4tle2SOGKQ65Bi4KCgoKCgoJz5Bq0KCj0Ju4GLv8A+omieNxxgSAIX/bKFikoKCgoKCgoKCgoKDjg7uH8B7tYdt+l3xwFBQUFBQUFBQWFqxxRmVWsJ/z4bmZWUFBQUFBQUFBQULjikO1p7v4TRzLoqfmg8qLy3QzKtv7Vbrmg8SZ28zL8b4ynrbqe049swnCuAoDI1BmE/WoKmMwUr3mDuoOd7mTrksHPPEjw5ETrG3Kbck53KuN/Q1zHG3IPZFG0+k0AQn92K9HL78Vv2EBO3Pk4jd8VXBHu2GceIDgpEXOzgfxlW2jMKexUJuDGOIZa3khcsz+LwtVvARDz2C8JnXozmM0Y9bXkLX0NY7njY09dxxwyORFTFzEH2MRcbRPzoNVzCEm+CbOhjdaicgrStmCqa3LrlLN9Bd4+koHrHpLebP5eBuWvf9jJPfiVNPxvkNxnUl/EcK6CwPHDiVo1B0HtjWhs4/z6t2n4Nsdjb/+JkheVF/p3nXuHbE7D74Z4TNX1FC6SvKrgQOJ2PI7/8KHodx/g3GrP3jZtH3Mi0WvnI6hU6N/bS8W2zu5BL6fhf8NQ2qrrKFosuf2HDyNmQ6qlkEDZ5nep3fOfbnh/fHUtZ9se9PSDBE0ehbm5lcK0LTR97zx/xb7yKF6+GmoPHKV4jdSXVcH9iN+2HJ+YcFrPVlDw8CZMtY192i1X+wLod3si0WukNlb1fgYXttm/nUDQeBPzcjp+18djqqmnaPFGjOcq8Bs+jIEbFlsKCZRv/gt1V0Cf6jchkai1D4GXF9XvZ3Bhe+d4B77UEW/x4o0Yz1dYl6ujBjBs71YqXn2Xyjf+5rG3HbmOzXIcH0G+nA0Q9+wDhCaNxNxs4NTS15yej/S7MY6EV1Px8tVQtf8Yp5+UzkcGrbiHyNlJGPV1AJzZ8Beq9x/rll/hykaeKy5eXgx6diG5//c0JyYtIfTu8fgOG2hXJOyXybTVNvD9uEcof+NjBv5mDgC+wwYSevc4TkxeQu6vn2LQ+ofBy/MwgiYn4hurI3tsKoWPbSd2wwKn5WKfX0jhY9vIHpuKb6yOoEkjAWg6WUze/I3U/+e/3Q5bLnfw5ER843Qcu20xBSu3Efe8c2/c8wsoWLmdY7ctxjdOR/BkyVvy+kdkJ6WTnbyCqoyjxKTP6pbbL1bHcUvMcV3EfPqxbRwfm4pfrI5gS8y1X2WTPWkZOVPSaTldQvSSGe6lMrYvvLyIeXYhBfc/xQ9JiwmZPh7fYTF2RbT3JmOqbeC/Ex6m4vcfE/XE/QC0VdVR8MB6TqYspSjtVQZvTuu2N3/OU/wweTEhdzvx/jKZtpoG/jte8kb/RvKKrQZKNv2Z88++7bnPwT3wmYWcvv8pTk5JJWT6BHwc3KGWmH+4fSEX3vwY3SrJ3XyqiFN3pXPqzmUU3L+Ogc8tApWH9f0jrWs5c6dPbBQ54xZx5vFtDN6w0Gm5wRse5szj28gZtwif2CiCJiUCoEv9BXVf55AzLpW6r3PQpf6ib7vlal8Wd/TTD1M4dx25yakET5+Az1CHPnVPCqbaBk5NXMiFNz9Ct2ouAC2nism7K428O5dSOGctA9enXhF9KsoSb15KKkFO4g2xxJs7aSGVb35EpCXednRPzqfh4FHPnTbIeWy+7MdHkC9nAyFJI/GL03Hk1iXkrdjO0Becxzz0hYfIW7GDI7cuwS9OR4jlfATg/M5POTZlJcemrLyyBy1msW//66PIMnAJGDGM1jOlGIrLEY1tVH30NcEpY+zKBKfcjH73FwBUf/otgeNutHw+hqqPvkY0tGE4W0HrmVICRgzz2B0y9WYqP/gSgIasXFRBAajD7d+lqQ4PQRXoR8PRXAAqP/iSkGnS9rXkn6eloKRHccvlDp02mgu7D1q8eXj3D0AdHuzgDUYV6G/1Xth9kNBpNwNgami2llP5+0A32nPI1Ju50M2YL9jEXHswG0zSfaD1R3PR6LRunXK2L/8Rw2g9U2Z1V3/yL4JSbrYrE5QyBv0HBwCo+ec3BI6V3M0nCmkrrwKgJbcYLx81gsazi6IBjt6PO3uDU8ZQZfFWf9rhNTe30nj4B8ytBo/j7BxzKYazNjEn29d3UPIYqj60jXk4AGKLwbp/vXw03WpbP8a6lrNtB0+9Gf0H0nob3fTlxqOnANB/8AXBljwSPLVju/S7vyB4mv129zW3XO2r3W0o6uhTNZ98RX+H/dw/ZQzVH+4HoPaf39DvtvY+1WrtU4KPBrEbc3rLFbP/cCleoyXe2k++or9DDumfPIaa9ng/64hXWnYLhrNltOQWexyrLXIdm+U4PoJ8ORtAO3U0FbukmOuz8vDu7+/8fKSfP/WWmCt2fYl22ujuiRSuWmQZuGh0oRhKK61/G8r0aHSh9mUibcqYzJjqmvAOCfTou126I0NpLbH5fokeTaQzt77LMj1BLrejt7VU3ynBaXRaDCV6+zI23kGr7mPUkR0M+MUEil98r1tuwyWKOfxXk6k5kOXeKWv70trHW6pHHWFf1+rIUIwlNu76RlQhgXZlgu+8jeYThYgGz6azVDt4jaV61JGdvQY33p6gjtRiLLV1Vzpxa13G7D8igWsyXuOaPb/j3G9ftx4U3fFjrGv523ZHP5Xitv++2rEvl+rRWOpGHRaMsUK6xdRYUY1aG9Sn3XK1LwB1hE1/wRKvozvCWZ/qD4DfiAQS9m4lYc8Wzj/Z9/uUt2MOKXPSpyK09u3aEq/g58OAh2dQ8eq7HrmcIeex+XIfH0G+nA3SuUZriW0/rcLH4XzER6eltdT2fKTK7pwl6oFpJB54iWGvLMI7KMBjt8LVQZcDF0EQ+guCsEEQhHcEQbjPYdnrXXxvgSAIRwRBOPLXxjPOSnT6pNOPQoKrMh58tyucrld0W6bbPyv0IbfgbJ0OXqevurMpUvz8Xzh600Iu/PUrdPPu6I7crduTmKMenYHYZqbyr195InWr7L325eQzT+K1KeKbEEPUE3MofsJlF+uZ11mhS3I1uIf72FKk6Xgup5IXkzt9OeGLZiL4ePiyNqWuXajla9ue5JoeIYdbrvblcr2euKUyzcdzyU1JJX96OuGPzLoC+lQPj42iSETabCrf+ghzU4vnvkvlv9hOLcvxEWTL2S5W29ntukzp23s4PGYxWUkrMJRXE7vufo/dClcH7q7j/gHIAz4EHhAEYQZwnyiKrcAtrr4kiuJOYCfAkYH/26lnG0r1aHRh1r81kVqMZVVOyxhL9aDyQtXfH1NNvUffdSRi7jQGzE4GoPF4Pj5RYTS0fz9K2+lBc4PDFQlNlBZDmecPo/cFd+TcaUTMngJAQ7bkrbcs89FpMTjUWWupHk1Uh9dHp8VQ3rleK//2NT955zec3fS+S3fE3GmEW2JuOJ6PJspmf0VpMXQz5rBZEwmZchM/3LvWTdS267t87avTem3j1WkxVth/31iqRx0VhrHM4g4MwFQj7R11pJbYnU9QlLYZQ1GZx16jg1et02J02H/GMqmMM+/FYCyrRK2zdYd1dpdWuoy5ndb8c5ibW/BNGExzTr5b74+xri932w6//w67/GWbI6S43fRlmzxirKxBHR4iXfEID8Gor+2zbuv6ZGhfYOlTjm3M0V3mQZ8quDL6VFupQw6J1FpvO7ONV6MLo83B6z8igaA7biNy1VxU/QMQzSJiqwH9nz7t0inXsVnu4yNc/pytmzeNyNlJANQfL8AnyrafhtLqeD5Sore7CuOjC7WesxgrO/pu2Z/3cd07T3gadp9DNCvTIfcEd7eKxYuiuEoUxb+LojgdyAIOCILg2Y2ULmjMzsM3VocmJhxB7U3o3eOoyci0K1OTkYl21iQAQn56G/Xf5Fg/D717HILGG01MOL6xOhqP53XpK3/7c75PXs73ycup/jyTsJkTAeiXmICprsl6C0E7xopqTA0t9EtMACBs5kSq92Q6rtYj5HKXvf052ckrpAfqP8tkwKzbLd5htNU3YayocfDWYG5opl+idM/7gFm3U/X5YQB8Y3XWciEpN9Gcf95tzDnJy8mxxDygmzEPsIk5aOJIolJ/zqm5GzA3e/ZcwOVuX7Y0ZefhY+MOuWs8tQ7u2oxMtDMnAxB851jqv/0OAFX/AOLfXk3JC+/QeOSkx872mH2G2Hind/bWZGQSavGG/HQs9d981y2HK6SYo9DERFhjrss4ZFembl8moTM6x6yJibA+2KmOHoBvXDSGc+Xd8P646vpyt+2KP37GiZR0TqSkU73nENqZ0noDuujL5oZmAix9WTtzEjWWvlyz97B1u7SzOj7vi26Qr321uzVDolAPlPpU8F0TqHNw12UcImSGdDIYdOdYGixu9UD7PuUTF22dVa6vxtz0XR4+NvEG3TWBun0O8e47RHB7vHeMpeHfkvf0Pas4NX4+p8bPp/Ktj7nw+m63gxaQ79gs9/ERLn/OLv3D59aH6fWfZxJ+jxRzYOIwTC7OR0yNzQRazkfC75mIfo90PmL7PIz2jjE0nTzrcdwKVwdCVw/uCYLwA3CdKHa8JUcQhPuBx4B+oigOdidwdsUFIGjyKGLWPQBeKvTv76N0ywdErfgVjdn51GYcRvBRE/vqMvyvj8NUU0/BopcwFEudQ7dkJtp7p4DJRPG6N6n7ovN9nSaz6zHZkOceImjiSMzNrZxOe806deH1GS/xffJyAAJujCeufdrDL7Io+u3vAQiZNoYhz87HW9sfU10jjScKOXXfM+6q4bK4TaLTG74AiH1uPiGTRmJqbiU/bSuN2ZJ3eMYmspNXSN7h8QyzTIdcfeAYhRbvNb9fiV98FKJZpPXcBU4/vsPuio2X0PXl8iHPPUSwJeYCm5hvyHiJHJuY421iPmNxj/hmK4KPmrZq6ZeehqO5FK7aYV23t5fzXyx6u32pvFzH3H/SKKLXPihNJ/r+fspf201k+n005eRTl5GJ4KNm8OY0/K+Lo62mnjOLN2EoLidiySwiUmfSWtjxkGfBr9fR5vDrsKsu23/SKAau6/CWbdmNbvl9NH2XT63FO2RzGn6WmAtTN1ljvu7bnagC/RHU3pjqGsmfvY6WPPsDglcXMQdOGkX0mvkIKi+qdu3riPm7fOr2WWJ+JR0/S8xFi1/EcLackJ9PJHzRTDC2IYoi5a++R+3eQ53WL7po21drXQOYXcTc223bVV0DDFq/wJq/CtO30GTpy9ftfZkTKekA+N8Y3zEl8RdZFD/5BgCqkECGbl+BJjoMw/lK8he+iKmmwaXrcrq9Vc7zSG+3L5WL/AUQOHEUUZbpkKt37aNi6y4i0mbTnJNn7VMxL0t9ylTTQPGSjRjOlhP880mEPzITsa0NzCLlv3uPur2dp6t1dYzs7Zhd1XXgxFHo1limQ969jwtbdxFuibd+XyaCRk3MK+n4XhuHqVaK13jW/oQ5fOmvMDe1uJwOuanV9S1NvXls7uoI2ZvHR1+162eMejtn17dqXLrjN8wnZNIIzM2t5C57nQbL+cjIfS9ybMpKAPoNj7dOh1x94BgFv5GmgE7YsoR+1w8BEVrOVpC3ckengc/4sg9cJ7E+ROOG+/vu1F1AwBN/7JP16G7gshHYK4riPofPpwFbRFF0OyWNq4FLb9PVwOVqpauBS2/ibuDSm7gauPQ2XQ1ceptL8chAT+hq4NLbdHUy3bteWbSA64FLbyNXXcuJq5Pp3qargUtvI9cxUq66hq4HLr2JXGmkq4FLb9PVwKW3uWIGLuvn9O2By2//1CfrsctnXERRfMzF558LgvBc72ySgoKCgoKCgoKCgoKCPRfzk8tTl2wrFBQUFBQUFBQUFBQUuqDLKy6CILh6olQAIi795igoKCgoKCgoKChc5YjKrGI9wd10yBHAVMBxzj8B+NYTgVz30fbpGwd7CbmeNZHrXnw5aTPJ9wyVXPWtxiSLV+HHgSDjs3Jy8WPMI07ft3OZkPOZIjmQs32pf2R1rXD5cDdw+QfS7GHHHRcIgvBlr2yRgoKCgoKCgoKCgoKCA+4ezn+wi2X3XfrNUVBQUFBQUFBQULjKMf/4rjJfCn58cwYrKCgoKCgoKCgoKFxxuLtVrFcZ/MyDBE9OtL54qSnndKcy/jfEdbx46UAWRaullxCF/uxWopffi9+wgZy483HrS5s89YZMTsTUhTfAxltt4x20eg4hyTdhNrTRWlROQdoWTHVNfd4tZ8yxzzxAcFIi5mYD+cu20JhT2Nl9YxxDLS+/rNmfReHqt6TtXj2HkJSbEA1ttBSVkb/sNY/c/SeOZNBT80HlReW7GZRt/avdckHjTezmZfjfGE9bdT2nH9lkfbt0ZOoMwn41BUxmite8Qd3BTndKumXQ0w8SNHmU9LK8tC00fe+8bVtflnfgKMVrpPpWBfcjfttyfGLCaT1bQcHDmzDVNnrklaNP9Z84koHrpBfl6d/NoPz1D+2WCxpv6WWMN8Rjqq6ncNGLGM5VoAoOJG7H4/gPH4p+9wHOrd7pka8vuOWOWa62LVe7livmwNul/SyovNC/53w/D34lDf8bJO+ZVGk/B44fTtSqOQhqb0RjG+fXv03Dtzkee9tjlqtty1PXiUYIuhUAACAASURBVESvnY+gUqF/by8V2zrHO+jlNPxvGEpbdZ30QsRzFfQbN8Kurkuee5uGb13NK9S3YpazL8vplvN8ROHKR7YrLkGTE/GN1ZE9NpXCx7YTu2GB03Kxzy+k8LFtZI9NxTdWR9CkkQA0nSwmb/5G6v/z3255gycn4her47jFG9eF9/Rj2zg+NhW/WB3BFm/tV9lkT1pGzpR0Wk6XEL1kRp93yx2zb5yOY7ctpmDlNuKed+6Oe34BBSu3c+y2xfjG6QieLLlrvsrm+MRlZCel01JQwsAlv3Av9fJi0LMLyf2/pzkxaQmhd4/Hd9hAuyJhv0ymrbaB78c9QvkbHzPwN3MA8B02kNC7x3Fi8hJyf/0Ug9Y/DF7d6yZBkxPxiY0iZ9wizjy+jcEbFjotN3jDw5x5fBs54xbhExtF0KREAHSpv6Du6xxyxqVS93UOulQPYkamPuXlRcyzC8mf8xQ/TF5MyN3j8R0WY1dE+8tk2moa+O/4h6n4/cdE/+Z+AMRWAyWb/sz5Z9/23NcX3DLHLFfblqtdyxazZT8X3P8UPyQtJmS6k/18bzKm2gb+O0Haz1FPSPu5raqOggfWczJlKUVprzJ4c5pnTge3HG1brroe+MxCTt//FCenpBIyfQI+DvGGWur6h9sXcuHNj9GtkuI1Vddx+oFnOTX1UYrTNzPole7XtVwxy3acktEt5/lIn8Ns7tv/+iiyDVxCpt5M5QdfAtCQlYsqKAB1eIhdGXV4CKpAPxqO5gJQ+cGXhEwbA0BL/nlaCkp65L3QTe8FG2/twWwwSTu0/mguGp22z7vljDl02mgu7D5ocefh3T8AdXiwgzsYVaB/h3v3QUKn3dzZnZWLJsq9O2DEMFrPlGIoLkc0tlH10dcEp4yxKxOccjP63V8AUP3ptwSOu9Hy+RiqPvoa0dCG4WwFrWdKCRgxzON4AYKn3oz+A2ndjW7qu/HoKQD0H3xBsCXm4Kkd26bf/QXB0+y33RVy9CmprsusdV398b8ISrnZrkxwyhiqPjgAQPWn3xA4Vqprc3MrjYd/wNxq6JZTbrf8McvTtuVq13LF7O+4nz/pvJ+DUsagt+znmn927OfmE4W0lVcB0JJbjJePGkHj+Q0O8rZtueq6FMNZm7pOtvcGJY+h6kPbuh4OQPOJ07RVXGxdX/6Y5ezLcrrlPB9RuDqQbeCiiQyltaTS+rehRI8mMrRTGUOpvssyPfEaLpE3/FeTqTmQ1efdcsdsu59bS/WdEo1Gp8VQorcv48z9yySqDxxz79SFYii1ibdMj0bnLF5LGZMZU10T3iGBHn3XrT/SPh5jqR61Qzxqx/ou1aOJlOpFHRaMsUKagdxYUY1aG+Sh9/L3KXWk1q5tSbFqHcrYtD+TGVN9I6qQwB475XbLGbOcbVu2di1TzBqH/Wwo1aOO6LyfjW72c/Cdt9F8ohDR0OaRV1qvPG1MrrpWR2oxltrGW+kkXq3bug668zaaT5zuVl3L1r7k7Muy5hH5zkcUrg66HLgIghApCMI2QRC2CoKgFQRhnSAIOYIg7BIEQdfF9xYIgnBEEIQjf2/q/DyDpVCnj0RRdFvmot/Q4mydPfBGPToDsc1M5V+/6vtuGWN2Ome/g9vprP4Omxe9dAaiyUTlh564nbWtThvmoowH3+2+vnPMnuyT7iJHn/IgVqeFLsVkKnK55YxZzrYtV7uWK2ZP9rPTeDv+1zchhqgn5lD8xOseSrvh7pU2Jlf76uExyrauh8UQtep+zna3rvtQzJfvOCVnHpHxHEzhqsDd9dS3gU+BAOAL4M/AT4G7ge2W/3ZCFMWdwE6AQ1G/sLa2iLnTGDA7GYDG4/n4RIXRYFmmidJiLLd/z6XB4dd5TZQWQ5njuzDdEzF3GuEWb8PxfDRRYfbr7KY3bNZEQqbcxA/3ru2zbjljjpw7jYjZUyR3trSf6y3LfHRaDGVVduVbS/V2t4D56LQYyjvKDJg1kdApozhxzzq37o5YbOKN1GJ0cLaXMZbqQeWFqr8/ppp6j77rjPD777Br27bxqHUetG2bmI2VNajDQ6RfpcNDMOprXXrl6lPtGEv1dm1LitW+voxlUhljmaWuAwMw1dQ7ruqKccsZ8+Vu23K1azljtlunbd7UaTFWOOznUj1qF/tZHaklducTFKVtxlBU5pHTdr1ytDG56tpYVolaZxtvWOd4Syu7rOshO39DcfpmDMXdq2tZ25cMXjnccp6P9GmU6ZB7hLtbxSJEUdwiiuLzQLAoii+IolgsiuIWYHB3ZeVvf873ycv5Pnk51Z9nEjZzIgD9EhMw1TVZbyNox1hRjamhhX6JCQCEzZxI9Z7M7mopf/tzcpKXk2PxDuimd4CNN2jiSKJSf86puRswN7u/h1gut5wxl739OdnJK8hOXkHVZ5kMmHW7xT2MtvomjBU1Du4azA3N9EuU7pMdMOt2qj4/DEDwpBFEL/5ffpj7vEdugMbsPHxjdWhiwhHU3oTePY6aDPt2U5ORiXbWJABCfnob9d/kWD8PvXscgsYbTUw4vrE6Go/nuXVW/PEzTqSkcyIlneo9h9DOlNYd0EV9mxuaCbDUt3bmJGos9V2z97B127SzOj53hlx9qp3G7Dx8hnTUdcj08dQ6qevQmZMBCPnpWOq/6d6MP33NLXfMl7Nty9Wu5Yy5nabsPHxsvCF3dd7PtRmZaC37OfjOsdRbZrNS9Q8g/u3VlLzwDo1HTnrkc4xZrrYtX11HoYmJsNZ1XcYhuzJ1+zIJneG8ruP+sIbSjX+i8cgPV0zMcnnlcMt5PqJw9SF0upXEdqEgZIuiONzy/8+KovikzbIcURRvcCewveLiyJDnHiJo4kjMza2cTnvNOv3q9Rkv8X3ycgACbownrn3q1i+yKPrt7wEImTaGIc/Ox1vbH1NdI40nCjl13zPWdXc1jh3y3EMEW7wFNt4bMl4ix8Ybb+M9Y/GO+GYrgo+atmrpl56Go7kUrtrhrhpkd/em1yw6vdnLSuxz8wmZNBJTcyv5aVtpzJbcwzM2kZ28QnIPj2eYZTrk6gPHKLS4R377Gl6aDnd9Vi6nH++Y3lOjMjl1Bk0eRcy6B8BLhf79fZRu+YCoFb+iMTuf2ozDCD5qYl9dhv/1cZhq6ilY9BKG4nIAdEtmor13CphMFK97k7ovOt9DK7qJedD6Bda2XZi+hSZLfV+392VOpKQD4H9jfMe0sV9kUfzkGwCoQgIZun0FmugwDOcryV/4IqaaBuu6u6rv3uxTahd13X/SKAaue1CaMvb9/ZRt2Y1u+X00fZdPbUYmgo9amrbVUteFqZusdX3dtztRBfojqL0x1TWSP3sdLXlnu6zbvuC+HF5X+1nOtt2b7VoQXGft3o5Z5eXc3X/SKKLXduzn8td2E5l+H005+dRZ9vPgzWn4XxdHW009ZxZL+zliySwiUmfSWtgx2UXBr9fR5nCVqatbbHq7jcnVvrxVzmcrCpw0iug18xFUXlTt2tdR19/lU7fPUtevpONnqeuixS9iOFtOxJJ7CF80E4NtXf/f2k51DdBmcv5bbW/H7Aq5vJfD3WZ2/bt4b58H3VLy164P0H2ExjW/7NOXXAKefq9P1qO7gcvTwEZRFBscPh8KPC+K4kx3gq4GLr1Jn24NVxnuBi69iauBS2/jbuDSm8hV364GLgq9g1z7Wa623dXApbdxNXDpbS768Z+LQK725WrgcjlwNXBRuPR0NXDpba6Ygcvqe/r0qWrAM7v6ZD12+YyLKIprXHyeLwjCp72zSQoKCgoKCgoKCgoKCvZczJD4qUu2FQoKCgoKCgoKCgoKCl3Q5RUXQRBcPe0nABGXfnMUFBQUFBQUFBQUrnKUWcV6hLvpkCOAqYDjfKkC8K0nAn8fYw826+KR895hufCS6T5tLxnvTT/f0E8Wr0rGp6hE52++6XXUZpUsXoAQvxbZ3HJhMsuzn+Xqz3LlL4CKRn9ZvN4y5k65ENrki1ntJc/zNf4aec6Dqpp9ZfEC6EIa3BdSUOgB7gYu/wD6iaJ43HGBIAhf9soWKSgoKCgoKCgoKCgoOODu4fwHu1h236XfHAUFBQUFBQUFBYWrG9Es3wx7VzLK3IAKCgoKCgoKCgoKCn0ed7eK9Rr9JiQStfYh8PKi+v0MLmz/wG65oPFm4Evp+F0fj6mmnuLFGzGer7AuV0cNYNjerVS8+i6Vb/zNc+/tiUSveQhUXlS9n8GFbZ29MS93eIsWb8R4zt6bkLGV8s3d816M22/4MAZuWGwpJFC++S/U7fmP594JiejWLJDqetdeKp3V9aZ0fK8fiqmmnrNLXsB4vgJ1dDjDMrbRevo8AM3HT1Hy5NZuxRwwYRSRqxcgqLyofn8v+h27O7mjNi3H7/qhmKrrOffo89b97HPNEHTPLsarnz+IIoX/uwzR4Pm9wtesn0tYkvTiyxOPbqM+p7BTmcAbY7nud4tQ+Wqo3H+MU799G4B+1w3mJy8+hMpHjdhm4odVb1J3rMBj97D189AmSS/Y+u+jr9Pgwv2T36Xi5atBv/8Yeb/9AwDX7VyGf3wUAN79/Wmra+Jw0mMeeRPWz0VrifmHLmK+9neLrN5cS8zX71zayZuZ9LjHMcc/O4/QpERMza3kLt3qNOZ+N8ZxzatSzFX7syh4Uop58IpZRM6eglFfB0Dhhr9Qvf+YW2dP84g6OpyEfa9b23bTsVOUPPm6x7HKlb8AAm9PJHrtfASVCv17e6nY9mEn96CX0/C/YSht1XXSi/rOVeA/fBgxG1IthQTKNr9LbXfyiJy5U8YcdjF9GWDgg9OIfmAaYpsJ/b4sCp75s8duOfqUnF6AuGcfIDRpJOZmA6eWvkajC3eC1X2M00++Zbc8+pHpxK2dw7+vnUdbVb1H3sHPPEjIZCnmgrTXaMo53alMwA1x1pciVh/Iomj1mwAMWj2HkOSbMBvaaC0qpyBtC6a6JrdOOfsUwND189Ba9vPJR13v5//5XSoqXw36/VnkW9r2kBWz0P26Yz+ffu4vVHmwn/3HjSLitw+Dlxe1H3xO1RsO5wRqNZEvLMf3umGYauooSd9A2/kK8FYR+ewyfK+NB5WKuo/2U7VzV7djVrjykeeKi5cXUU8/TOHcdeSlpBI0fQI+Q2PsioTck4KptoHcSQupfPMjIlfNtVuue3I+DQePdtsbbfHmJqcS7MQbavGemriQC29+hM7BG7V6PvVfdtN7ke6WU8Xk3ZVG3p1LKZyzloHrU0Hl4a7z8iLqqUc4M28t+VMXEXTX7c7ruq6RvMkL0L/1EZGPd8RsKCqj4GePUvCzR7t9wMfLC926Ryh+YC35Ux8h6K4JaBzcwbOmYqptIH/yQ+j/8HfCH58nLVB5Ef3yCkpXb+X0HYsoum8VYpvnL0AMSxqBf2wk39yylB9WvMFPNjq/6/EnG+fzw4qdfHPLUvxjI9FOHgFAwprZnN70Af9JepyCjbsYtnq2x25t0kj8YyP5zy2PcnLFTq7ZON9puWs2PsTJFTv4zy2P4h8bSajFfWLBZg4nPcbhpMe48OkhLnx6yEPvCPxiI/n3LUs5ueINrnER8zWWmP99y1L8bGL+fsGrZCY9TmbS41R8msmFTzM9jjkkaSR+cToO37qEvBU7GPrCQ07LDX3hIXJX7ODwrUvwi9MRYnEDnN/5D7KmrCRrykrPTnQuMo8YisrI/+lS8n+6tFuDFtnyl8U98JmFnL7/KU5OSSVk+gR8hjnkkXuTMdU28MPtC7nw5sfoVt0PQPOpIk7dlc6pO5dRcP86Bj63qFt5RM7cKVcOu9i+HDz2OsKm3UTmpBVk3r6c4m2feOyWpU/J6LV1H7l1CXkrtjP0hQUu3XkrdnDE6h5pXaaJ0hIy4UZazl3w2Bs8ORG/WB3Hx6ZS+Nh24jY498Y+v5DTj23j+NhU/GJ1BE+SvLVfZZM9aRk5U9JpOV1C9JIZ7qVy9ikgNGkkfrE6Dt2yhNwVO0jY6Hw/J2yU9vOhW5bgF6uztm2Aczv+wZGklRxJWunRoAUvLyLWpHLuodUU/mwhgT+diCZ+kF2RoJkpmOsaKJz6INV//DsDlj8AQOC08QhqNWemL6JoxqME33sn3tHhPYq9z2AW+/a/PoosAxf/4cMwFJViPFuOaGyj9pOv6J88xq5M/+Qx1Hy4H4Daz76h323DbZbdguFsGS25xd3zjpC8Bou35pOv6J/i4E0ZQ3W7958O3pRbMBSX0ZrXPe/FusWWVjBJ90IKPhrEbkyZ5jc8gVbbuv7HVwQm32JXJnDKLR3ez74mwCbmi8FveAKGohKMZ8ug3T3F0T2G2r9K7rrPvibgVsndb3wiLSfP0HpS+gXIVFMP3bgfdMC00ZTu/gqA2qN5ePcPQBMebFdGEx6Mdz8/ao/kAVC6+yvC7xgNSLPSeQf6AdLVh9Zyx4n1XBM27SbKLO66Ltyqfn7UWdxlu79igMVtS/j0Wyn/2zceeQdMG+2R19sDb8T0Wyjz0AsQNnU05bsOAlCf1bW7/mguAOW7DhI27WaPHY5cbB6Ry9vT/AVSHmk905FHqj/5F0EO7qDkMVR9eACAmn9+Q+DY9jxisOYRLx8N3ZkYT87cKWcOu9i+HH1/CkVbPkI0tAFgrKzz3C1Dn5LTC6CdOpqKXV/auP1RO7jV4cGo+vlb3RW7vkQ7rSOHxT89l8Jn3unW1KIhU2/mwgeStyErF1VQAOrwEAdvCKpAPxos3gsffEnINKkP1B7Mtvat+qO5aHRat045+xRA2LTRlO+W9rP744VlP+8+SNgdPd/PvjcmYCwuwXhOOieo/+dB+iXZ9+V+SbdS+/d9ANTv+Rf+t1oGSqKIl78vqLwQfDWIRiPmBvdXtRSuPmQZuHhHajGWVlr/NpbpUUfad3R1hBZDexmTGVN9I6qQ/gh+Pgx4eAYVr77bba86QouxxMZbqkcd0dlrLePEW94D78W6AfxGJJCwdysJe7Zw/snXrUnSrTdSi7G045enttJK5972MiYz5vomq1cTE0H8J68S++4G/Edf162YvSPs93NbWWe3d6Rzt2ZINIgig/7wNLEfvYp2gQe/YNngowuh5bze+ndLqR5fXahdGV9dKC2lVR1lSqrw0UkHq9zVf2TYml8zPmsrw9b+H/nrPd/vPrpQWs53xN1aqsfHwe2jC6W11Gb7SjqXCb7lJxgu1NJcWOah1z5m196OmFttYu6pF0CjC6W1xN6tcXBrHGJ2LBP1wDQSD2wi4ZVH8A4KcOu8mDwCUtse+o/NxL63Af/R13ocq1z5C9r7s20eqezsjnSWRwIB8B+RwDUZr3HNnt9x7rfdyCNy5k4Zc9jF9mX/eB3BY/6HUZ+tZ+Tf1hE4It5jtxx9Sk6vtF6tndtQWoWPwyDAR6d1cFdZBwqhKTfRWlpF43+LPHYCaCJDMdi0b0OJHk1kaOcyNl5nZQDCfzWZmgNZbp1y9imwtFuPjhc2ZRyOU9EPTOOmLzZxzWYPc3ZEmH1fLqvE2/GcIFxLm03uNNc3oQruT/2erzE3tRD/r78Qf+BPVL31V8y1ypTLP0a6PXARBOHir80Jnd9P0OkqgpMyiCIRabOpfOsjzE09eK+Di3V6UiYybTaVb/bQe5FugObjueSmpJI/PZ3wR2Yh+Kh7th1Ovc7LtF2o4tS4eRTctZTS9b8n5pUVePXz89zjLJ7OhZy6BW8V/jddy/n0TZy59zECk2/t5q+oPW1j0n8Gzk0md80f+VdiKrlr/si1rzx8UW7P9rX9n+E/H+vx1ZaL8ToWifj5bZT/zaPXNHW53k6/6HfRvkve3kvmmCVkJa3EUF5D3Lo5PXJ6mkfaLlRxcuwD5P9sGaXP/p6Yzd1o23LlL2nFTtfr3i39p+l4LqeSF5M7fTnhi2Z6nkfkzJ3OuFw57CLrW/D2wju4H0fv+C35T7/D9W+kdUMtQ5+S0+titZ7uay8/DTHLZlC08X2PfV2KPdnPDhUT9egMxDYzlX/9qvecl6xPuT8WdJXrzv9xL/8Zs4Qjk6X9HP+U5/vZfoUeOBHxu+EaMJspmDCb01PmEjrvF6gHRvbM2VeQ+1awK/RWsS4fzhcEwfHnBAHIFARhJCCIoljl5GsIgrAAWACwRnsDMwMH2y1vK61ErQuz/q2O1NJWbr8qY1klGl0YbWV6UHmhCgzAVFOP/4gEgu64jchVc1H1D0A0i4itBvR/+tRtsMayStRRNl6dFmNFZ686KgyjM++dt6F7ovvei3Xb0lpwDnNzC74Jg2nOyffAq0etG2D921sX5sQrlWmva69Af6vXZJD+2/J9AYbiMjSx0bR44AXLFRab/ewdGYaxXO+kTGe3saySpszvMVVLt1Y0HDyC73XxNH6b7dI3cF4KA3+dBEDt8QJ8ozt+yfHVaWkts7/dq6XE/iqMb1SotYzuntutD+qXf/wfrn15YZexRs+bSpTFXX+8AN/oMGo5BVh+IXRwS79c2WxflJbWso79Iqi8CP/pzRxOXtWld+C8FKu3zhJzrWWZa29HzD5RoRhsyrR7M5Of6NILoJs3Fd3sKZaY8/GJ6ojHR6fFUGbfzgwOMUtlJLexstb6eemf93H9O13HDReXR6Bz2/aJjfaoT8mVv9rXa+fWhWF0dJd6kEfyu5tH5MydlzeHXcq+3FpSZX1Grf5YAZjNqLWBGPXOHxiXq0/J2Zd186YRObujvm3dGl2oXV6EzvXtowv9/+ydeXwTdfrH39M0Se8rpW16AG0p/lTOcqncRwHv3yrggauiKMoht7CuKCiKIiouAoLn7v52VcDd9dgVaMFj1RUKhVpZoQeFcvRM7zNpMr8/Jk2TNGlTSh3Eeb9evqSZb+Yzz/N95pn5Zub7fDEWlePTKwqfnhEk799gO6bBe9dz9PrfYSqtbKMbef9UImamAFB7NBeNXXxronUYnV4PNhYaHF4B00TrHHJn+PRxhE4ayk93PN2uvS3IcU5Fz5pC9D1SP1cfzUUb034/t/G1nc2mUrt+/r80+v+fBzm7uMzxXI4Kp7nE0KaNtz6c5uIy27lsqawh8KZx1P37EDSbMZdX0ZDxX3z6JUmvnSn8qujoiUsZcNjuv0NADJBh/bdLRFHcLoriUFEUhzoPWgDqf8hB2zsadWwkgtqb4JvHUJ3mOBG4Ou0AIbdLySz4+pHU/ucHAE7OWMmJ0bM5MXo2Ze98QumWnR5fAOszc9DY6YbcPIbqVCfd1AOEtujeMJLa7yTdvBkrOT5qNsdHSbolmz3X7aq2OjbSNolWHdMDbUIMRrvKIu3R8EO2o69vGkNNmuNk75p9drrXj6LO6mtVWBB4WXXjItH0jsZU4HmSaPghG03vGOn4rdq1+9pqB98maQfZadd+nYH2it4IPlpQeeE3vD9NOWfa1Tv77l6+n7iC7yeuoPTzdPTTx0g2DUmiuaYeY4njxctYUklzbSPBQ5IA0E8fQ+nudACaiioIvU56fShsdD/qT7Zv97l397ROqP/8IFFW7aAhSZjdaJtrGwiyakdNH0PZ7tZTKnRMf+pyzju81uXO5pYJ9aWfpzvourPZXNvooNtic2d0AQrf3WObgGvYnU7kjLEABCa3o13XQGCypB05YyxleyRt+3erw68fTt3x9vsaupZHnGNb2zsao4exLVf+AimPaOOj0cRJ2qE3j6Y61fGcqk47SNjtEwAIuWEkNdY8oolzzCM+CTEYzxZ7rCtX7vy5c9jFPJdLP08ndFQ/AHwT9Ahqb7eDFpDvnJLzXC58dzdHJi3nyKTlGHYfJGLGOJu2uaYek5O2yUk7YsY4DHvSqT9ewIF+D5I+bC7pw+bSVGjgyOTHXQ5aAIrf201WylKyUpZSsfsgPaZJugHJfTFX12MqcRy4mEoqMNc2EpDcF4Ae08ZRsUc6B4LHDSZ63m84cf86LA3Gdu1tQY5z6vy7e2yT6cs+TydyutTP7V0vmu1iO3L6WMp2u+jnGzzL2Y1Z2ah7RaOOke4JAm8YS+1+x8qGtfu/J/h/pcFV4JTR1H8v/VjZXFiK3zXSWxeCrxafgf+D8WTHmgqXHx2VQ34cmAQsF0UxC0AQhHxRFOO7pGq2cP7pN4j/0xqpvOXONJpyCohYPJOGrBxq0g5S8WEqca8uoe8X2zBX1VKwYH2XJG26T71Bwp/WgMqLih2SbqRVtzrtIOU7Uol7ZQlXfLkNc+VF0u2itv+wq4h4dBpiczNYRM6tesP2JMIj3dVv0PuPzyB4eVGxM1Xy9SKrr/cdpOLDvcS+spSk/dsxV9Vy5rEXJd3h/YhYNBPRbAGzmfNPbsbcmXdKzRaK1myl53vPInh5UblL0u6x6B4asnKo3XeAyh17iXl5GX32v4m5soazCyWbLdW1lL/zD+L//iogUvvlIWq/TG9fz46ytCOETxzMyAOvYW4w8t+FW23brtn3It9by/weX/EWV/9hLl4+asr2HaVs31EAflq6jSvW3o/grcLSZOS/y7Z7rG1IO4JuYjLXHvgD5gYjPy1srVg1bN96W2njEyve4kprKWbDvqMY7KqyRP5vZ18Tk3TDJw7m2gOvYXGyefi+F22ljY+veMtaDllt1T1qp3tdp3UBytMyCJs4mGHfb5LKmC5qrd6UnPYSGZOWA5Cz4s3WEqr7j9oqDsWv+i0B/XojiiJNZ0rJWb6tY9Eu5BH/4f2IXDwT0WwGs4VznYltufKXVfvsU9tI+NNqBJUX5TvSaMw5Q9SSu6n/IZfqtIMYPkyl16tLuPKrbTRX1nB6/kuSzUOvJH7uKjA1I4oiZ598A3OFZ6ViZc+dMuWwrp7Lhe/v58qNcxn+1QZEYzM/PeZ5VTNZzikZdQEq0jIIm5jM0O9fx9LQRPaiVn8PTnuJI1bt3BVv2sohV+w/0qnKZa6o3HeYkInJDPpuCxZrOeQW+qe+TFbKUgDyV26zlUOu/CLDNpcl/rnZCFo1V34oPW2pPZxN/soO7Jbz/4JDGAAAIABJREFUnELqZ93EwYw4sAlzg5ETC1v7eei+lzg0sbWf/+cPLaWnj9qqhyU8JfUzokjjmVKyl3mWs0ue3Urs22vBS0XVR3sx5hagW/BbGn/Mpu6LA1Tt2oN+/XLi97yNuaqGwiUvAFDx10/RP7+E3p++AYJA1d/20pR96qL5Q+GXg9BRhSpBEGKBV4EzwNNApiiKCZ4KZMXfLMuLcp0oKHLZ4OUlj9FegnzOPlcbIIuuqjMlmS4yosuXvLsftSDfKr+hvhdxfsQvBLNFnn6W63yWK38BlNT5yaLrLWPulAtBxtyp9pInh/lpPF9/7GJS3uAjiy6APlS+ifNXHP9cnuTZSWqX3XpJJ4CADR9fkn7scHK+KIpnRVGcDnwBpALyZHgFBQUFBQUFBQUFhUsCQRCmCoJwQhCEXEEQ3E50EgRhmiAIoiAIQ7uq6XFVMVEUPwXGI706hiAIs7oqrqCgoKCgoKCgoKDwy0IQBBWwGbgeuAq4SxCENmsLCIIQCDwGeLaadgd0qhyyKIoNoij+aP1zzcU4AAUFBQUFBQUFBYVfFXKXO+56OeThQK4oiidFUTQCHwC3umj3LLAeuCjvfHdUDvkHd5uASE8E6pq6sN5IF7CI8r2a5+PdLIuul0wTeywyvYsPEKSS591hOef1qGR6T9tkVsmiC/LlETnxlqmfNd5mWXSNJvniy18lT86W61wG+XKYnHlEq5ann+W6HwnSeFbhrDuoq9PIpq3wsxGDNP+9hbPACPsG1uVT4kRR/EwQhGUXQ7SjqmKRwBSgwulzAejkCnUKCgoKCgoKCgoKCpc69msyWtkuiqJ9idV2V2UVBMELqbjX/RfzuDoauHwGBIiieNR5gyAIX17MA1FQUFBQUFBQUFD4NSBewqvTg7QmI9DeWhBngTi7v2OB83Z/BwL9gC8FQQCIAj4RBOEWURTdrgXZEe0OXERRfLCdbXdfqKiCgoKCgoKCgoKCwi+WdCBJEIR44BxwJ2AbG4iiWAWEt/xtfeCxrCuDFuj4iUu30uvZBwmdkIzZuuBTfdbJNm38+yfYFnyq2J/B6VVvA9Bz1b2EpgzFYmym6XQxeYs3Ya6u90g3/tkHCJmYjKXBSO6iTdRl5bfVHZBAn43zpYWm9mWQv+od6ZhX3Uvo5KGIxmYaTxeRu+h1j3WDxg0mdvVDoPLC8H4qxVs+ctguaLzpvXExvv0TMVfUkD/3JYxnS1CFBJKwbQV+A/tg2Lmfs6s8XwwRIHBsMjFPz0ZQqTB8sJeSrW11e76yGL/+fWiuqOb0fEnXb2AScevmWRsJFG18n6o937tQuPRsBim+QiYk2xYUcxVffnbxVWkXX2E3XUvM0jvwTYrl2A0rqPshr1PaPZ95kOAJQ7A0NJG/eBP1P7rWjn/1Mbx8NFTtP0zBU5K2KiSAxK1L0cZF0HSmhLxHNmCuqutQU05fg3z+liuPyKUbNG4wPdfMBpUXZe+nUrT5bw7bBY038RsX4TcgkeaKGk4+usHWz4nbH8ff2s8FT77pkV4LAWOTiXlKiq/yD1Mp3bqrjW7cK0vw7ZeIubKG0/PXYzpbgu/AJGLXzbc2Eije+Feqf6Y8Ejh6IDEr70XQeCMamzn73HvUfpfVKW254vpC+xkgat7thN81CcwWCp56k+qv2rw00a7ur83XF3qNDBg1iOiV9yKovRFNzZx//j3b6vbdqftLvjbLGV8KFxdRFJsFQZgP7AFUwDuiKB4TBOEZ4JAoip90h26nqopdTEImJOMbr+foyHnkP/4GCesedtku/oU5nHx8K0dHzsM3Xk/I+MEAVH2dSeb4RWRNWkLjyfPELLjdY12fBD1HrptP3vKtJLzgWjfhhYfJW/4GR66bj0+CnpAJkm7l15kcHbeIzIlLaMw7T+yC2zwz2MuLuLVzyL13DT9NmE/oraPxSYpzaKK7M4Xmylr+O/oRSt76hJgn7gNAbDJyfsNfOLf2Pc+0nHRjn53DyfvWcHzSPEJvGYPWSTfsjhTMVbX8NHYOpW9/gn6lpNtw4jQnbl7CiRsWkXffamKfnwuqToSMXDYDwROS8YnXk2mNr/h24iv/8a1kjpyHT7yeYGt81R8vIGf2emq+/+8FaWvjo8kaNZdTK7bSa90cl+16rXuEUyu2kjVqLtr4aILHJwOgn3cb1d9kkTVqHtXfZKGf50GMyejrFpvl8LeceUQOXby86Ll2Dtm/fYZj4xcQdutofJJiHZqE35lCc1UtP456lOI3PyH2iXsBaz+/9FfOPvueZ1pOujHPPEL+/avJTplHyC1j0PZxyiMzJmOuquXEuDmUvv0x+pX3A9B4ooCcmxeTc8NC8u99mtjn5v1seaS5vJq8B57jp5SFnFryGr1fW9wps2XLI13oZ5+kWMJuHcWxCQvIvmcNPZ97BLw89Pev1NcXeo00V1Rz8oG1nJjyGAVLNtLz1U7Y/Gu8NssYX5ckclcN63pVMURR/Jcoin1FUUwURfE562dPuRq0iKI4rqtPW0DGgUvolOGU7voSgNqMbFTB/qgjQh3aqCNCUQX6Uns4G4DSXV8SOlUqWFD1VSaYpYosNYez0eh1HumGTR1G6c6vrLo5eAf5o44IcdINQRXo16q78yvCpg5vq5uRjSbaM13/QUk0nSrCWFCMaGqm4pN/Ezx5uEObkMkjKN+1H4CKf35L4MgBAFgamqhL/wlLU+crhPgNSqLpVCHGM1bdT/9NcIpD0QeCU0ZQ/pGkW/mvbwkcORAAsdFos9VLq6GzCx7LZTNI8VXWyfgqs4uvxtxzNOad50IImTIcw64vAKjrQLvu8AkADLu+IMQaYyFThmPYKX3fsPMLQqY69pcr5PQ1yOdvufKIXLr+LeeztZ/LP/6GkMmO8REyuTV+Kv75HYGjWvu5Nv0nLE2dr8TnNygJ4+nWPFL56dcEOekGTR5BxUf7JPv+9S0B17XkkSabrYJWg9jJ6oddie2GY/mYissBaQDlpVUjaDx/0UCuuO5KP4dMHkH5x98gGpsxnimh6VQh/oOSOqH76/J1V66RDcdO0lxitTm7czb/Gq/NcsaXwuWDbAMXTVQYxvNltr+N5w1oosLatik0tNsGIOKuCVTuz/BYt8lOt6nQ0OamQaPXYTxvcGzjSvfOiVTsP+KRrjpK52CvqdCAOkrn1MbOJ2YL5po6VKGBHu2/PV1Tob1umQtdHSY3un6D+nJF6utcsecPnP39Fluy9FRbDpuhbT93Jb46r+0YP5LdjvtVO2sXGtBYfaMOD8FUIhXyM5VUoNYFd6gpp69BPn/LmUdk0dWHYbQ7n41FBjR6V7p2/Vxdj3dX80ikXY7AGl+ROvdtbPEVBIDvoL703buZvns2ce5JefJIyA3XUf9jPqLR85K4ssV1F/rZk++649fo665eI1sIvuE6Go6d9NjmX+O1Wc74Urh8kG+4Kriooub8S5yrNk4/LUQ/djtis4Wyv33toWzHui4rrjsdWszC2xHNZso+8kzXddE4559JXB2bZ7vvlLAnfrY2qT+azYmU+Wj7xNLz5UVUf3kY0dNfbGWzGZc2tfml14P4ujBtV7t1ijFP4v8ia3abr0E+f8uUR2TTddGHbbvZgzad5ULttbZpOJpN9uR5aBNjiXt5MTU/cx7x6RtHzBP3kjNztWeatt3KlUe60s9d6H/F1y3CHeva25wUR/TK+8i75+mfTfcXeW2WM74uRSzyreP0S6ajBSiniqK42/rvYOAVYBjwI7BYFMViN9+z1X5eETyI//WLByDy/qlEzEwBoPZoLppoW7EBNNE6jMWOy8UYnZ6GaKJ1GIta24RPH0fopKH8dEf7ySLq/qlEzpwk6Wbmoo0Op8a6TavXYSwqd2jfVGhweAVMq9dhLG5t02P6OMImDeHYjNXt6tpjKjQ42KvW62yPPW1tiqQ2piIDqLxQBfpjrqxx3lWnMBWVodbb64a31S0sQ92BblPuWSwNjfj07UVDVq5n2j+zzZH3T6WHNb7qjkr9XGvdponWYepkfHWGiPuud9C2jx/J7g607WLMVFaJOiJUetoSEYrJUNWhvhzxJZe/5cojcum23aedbpQOk1P+amljKrT2c5DfxckjzvFV4hxfHuSRvJ8/j6ijdCS8+TtOLdqI8XRRh3py5hHHfV5YP3vyXXf8Gn3d1WukOkpH7+1PULBkI8aCjm2+WLot/BKuzRdLt7PxpXB50tGrYs/b/ftloBC4GakE2jZ3XxJFcbsoikNFURzaMmgBKH5vN1kpS8lKWUrF7oP0mDYOgIDkvpir622vx7RgKqnAXNtIQHJfAHpMG0fFnoMABI8bTPS833Di/nVYGtp/17Lovd1kpiwjM2UZ5Z8fpMf0sVbdJJpr6jGVVDrpVmKpbSAgWXovuMf0sZTvTgcgZPwgYub/Lz/d/0KHuvbUZeag7a1HExeBoPYm9JbRVKUedGhTmXqQsGkTAAi9cSQ133pencQd9Zk5aOOj0cRFSro3j6Y69YBDm+q0g4TdLumG3DCSGmtVFE1cpG3CnzqmBz4JMRjPuhyruuTntrn4vd38mLKUH63xFd7J+Aq3i6/OUvLHzzk2eQnHJi+hYs8BdNPGA+DfjraltgF/q7Zu2ngqrdqVe9PRTZe+r5ve+nl7yBFfcvlbrjwil649dZk5+MS39nPYraOodNHPLfETeuN11Hzb9co79Zk5aHpHo46V8kjIzWOodtKtTj1A6O0TAQi+YaStupI61jGPaBNibNWvPKErsa0K8ifxj6s4/8KfqTt03CM9OfNIC13p58rUg4TdOgpB440mLgKfeD11R3M81v21+bor10hVkD8J7z5F4fo/UXfop59N95d2bb4YuhcSXwqXJ0J7EyUFQcgQRTHZ+u+joigOstvm8Lc7vo++za1A7+cfImTcYFvpw5byhf1TXyYrZSkA/gMSW0sffpHBqd+/BcCgbzcjaNU0V0gj8drD2eSvbB1LWUSXL3wBEP/8bELHD8bc0ETu4s3UZUq6A1M3kJmyTNIdmEiStRxyxf4j5Ft1B3/3Ol6aVt2ajGxOrnAsB+jj7fq9y6DxQ4hd/SCCygvDh/so2rQT/dK7qf8hl6rUgwhatVQGsF8C5soa8udtwFggJaOrv9uOKtAPQe2NubqO3Jmracw547B/Ly/Xrg4cP4SYp2YjqLwo35FG8es7iVoi6VanSbq9Xl2C79UJNFfWSCUXzxQT+ptxRMydBqZmRFGk+LUPqNp7oM3+LRb3vu5um01mlVvt3s8/RLA1vk7axVe/1Jf50S6+Euzi67S1n0OnjqD32tl464IwV9dRdyyfE3c/2+prof1n5j2fe9imnb9kE/VW7av3vsKxyUsA8BuQ2FoO+YsMW3laVWggfd5YhiYmHOO5MnLnvIS5sta2b5WX68fLcvq6u/3dnre7M490ZG936nq76efgCUOIW/0AeKkwfJhG4aZdRC+7i7rMXKpS0xG0auJfW4SftZ/z5r5s6+f+/9mOKtDX1s/Zd6+mMeesw/413maXuoHjhhBtLYdcsSONks07iFw8k4asHFseiXtFyiPmyloKFqzHeKaYkN+MJ+LRaYjNzWARKf7DB1TvbVu6tdns/ne0C43tqMemEzlvGk35rZO2c2euptnpKaZcecTduQxd62f9gmno7pgEZjMFq9+m+ou2c6jc5bDL1dcAWrXra/OFXiMjF8wgYu40jHY25/326TY2u+NyvjZfbF1P4wsg+czH7g2/hKiZf8MlvQJl4Ov/uiT92NHA5SzS62ECMA9IFK1fEAThB1EUB3Qk0N7ApTtpb+DS3bgbuHQ37gYu3U17ybG76ehmurvoaODSnbR3s9OdyOVruHhTcX5JuBu4dDfuBi7dTXsDl+5GrtiW61wG+XKYnHnE3cDlckXOa7Oc/GIGLnOvv6QvbYFbPr8k/djRleJNIBAIAP6IdQVMQRCiAM9XtFJQUFBQUFBQUFBQUOgC7U7OF0VxjZvPiwRB+KJ7DklBQUFBQUFBQUFBQcGRrpRDXgO8e7EOREFBQUFBQUFBQeFXgYer0ys40lE5ZHdlJAQg8uIfjoKCgoKCgoKCgoKCQls6euISCUwBnAulC8B33XJEFwlRxsn5clHZqJVFVyPjBNMAreelZC8X5CrC4HLtt5+JpmZ5JvSaLfJNGFfJNHm6SqY8Ipe9AH4aDxfuu4xQyVXMRcZrs1z3BTVGjSy6ob6NsugCVDfIk0cULn86Grh8BgSIothmIr4gCF92yxEpKCgoKCgoKCgoXMa0V9VXwT0dTc5/sJ1td1/8w1FQUFBQUFBQUFBQUGiLfO9BKCgoKCgoKCgoKCgoeEhXqop1mV7PPkjohGTM1pWn67NOtmnj3z/BtvJ0xf4MTq96G4Ceq+4lNGUoFmMzTaeLyVu8CXN1vcfa8WsfIHTiYCwNRnIWvk5dVn5b7QEJJL02T9Led4T8J9+RtB+/k7CpwxAtFkxl1eQufB1jsfM0oLYEjRtM7Gpp5WnD+6kUb/nIYbug8ZZWjO2fiLmihvy5L2E8W4IqJJCEbSvwG9gHw879nF213WM7W+jz3Cx0EyVfH39sM7Uu7A0YkMD//GEeKh8Nhn0Z5P6+tWhczINTiXngesRmM4a0DE4++38ea8c/+wAhE5OxNBjJXbTJra/7bJwvrYa8L4P8VZKv4x6/k7Apw8FiwWSoImfh65g88HXA2GRirKt8l3+YSunWXQ7bBY23tMp3v0TMlTWcnr8e09kSfAcmEbtuvrWRQPHGv1K9p+0q392h3YI6ugd9UzdTvPF9yt78u+e6Y5LRP/UweHlRsWMvZW+01Y3dsASffn0wV9ZwZsGLmM6VoI6JICl1K00nzwHQcPQE55/c7LFu4FgprgWVF4YPXMd1r1cX49c/keaKGk7Nk+I6cPRAolfei6D2RjQ1c+6596j9Lstj3RbkyiNy5JAWej7zIMEThmBpaCJ/8Sbqf2xrs1//BOJffQwvHw1V+w9T8JRksyokgMStS9HGRdB0poS8RzZgrqrzSDdh7QOEWW0+4cbmgAEJ9LXaXL7vCCetNrcQ8+gtJDx9L/+5ahbN5TUe2yxHHgkcm0zM07MRVCoMH+ylZGvb2O75ymL8+vehuaJaWtn8bAl+A5OIWzfP2kigaOP7VHUyj8ilHTAmmeinH5LyyIeplLrKIy+35q+C+esxnXPMX0l7N1PyWufyV9C4wfRcMxtUXpS9n0rR5r+10Y3fuAi/AVIeOfnoBozWvBk173bC75oEZgsFT71J9VedW2ZOzhyWuHYWYdZrc/ZC99fmK2znVAZ5T0rX5l7LphM1cxImQzUA+ev+SsW+Ix7pXmg/q2Mi6Ju2xXa9qD9ygvNPbumUzV3JnS1EP3oL8U/fy4FO5pFLCqWq2AUh2xOXkAnJ+MbrOTpyHvmPv0HCuoddtot/YQ4nH9/K0ZHz8I3XEzJ+MABVX2eSOX4RWZOW0HjyPDELbvdYO3TiYHwT9GRcu4DcZW+Q+KJr7cQXHyJv2TYyrl2Ab4KekAmS9rktH3N0wlIyJy2nIvUwcUumdyzq5UXc2jnk3ruGnybMJ/TW0fgkxTk00d2ZQnNlLf8d/Qglb31CzBP3ASA2GTm/4S+cW/uexzbaEzZxML7xeg5cs4DsZdvou/4hl+36rn+I7GXbOHDNAnzj9YRNGARAyMirCZ86jPTxS0kfu4QzWz/xWDtkQjI+CXqOXDefvOVbSXjBta8TXniYvOVvcOS6+fjY+fr8lo/JnLiEzJRllHfC1zHPPEL+/avJTplHyC1j0PZx9HXYjMmYq2o5MW4OpW9/jH7l/QA0nigg5+bF5NywkPx7nyb2uXmg6sRp0gXtFqJXzabmy8Oea1p1o9c8yqlZT5M7ZS7BN49toxs6YzLm6jpyJjyM4Z2PiVrRqms8XUTeTY+Rd9NjnRq0tMR13n1r+GnifEJvcRHXd6Rgrqrlv2OkuI7+nRTXzeXV5D3wHMcnL+T04tfotXFx52xGvjwiSw6xEjwhGW18NFmj5nJqxVZ6rZvjsl2vdY9wasVWskbNRRsfTfD4ZAD0826j+pssskbNo/qbLPTzbuuUzYeuXUDOsjfo48bmPi8+RM6ybRyy2hxqtRlAE60jdMwAGs+WemwvyJdHYp+dw8n71nB80jxCbxmD1im2w6yx/dPYOZS+/Qn6lVJsN5w4zYmbl3DihkXk3bea2OfndjqPyKLt5UW0NX/lTJ5HsIv8FWrNX9nj51D29sdEOeUv/ZOzqf2q8/mr59o5ZP/2GY6NX0DYraPxSYp1aBJ+ZwrNVbX8OOpRit/8hNgn7gXAJymWsFtHcWzCArLvWUPP5x4Br875Wq4c1nJOpV+7gJxl2+jzoutrc58XpWtzuu2cGmTbdm77Z2RMWk7GpOUeD1q62s/G00Xk3riQ3BsXdnrQ0tXcCVIeCbmAPKJweSDbwCV0ynBKd30JQG1GNqpgf9QRoQ5t1BGhqAJ9qT2cDUDpri8JnToCgKqvMsEsVbOqOZyNRq/zWDtsyjBKdrRo5+Ad5Ic6IsRJOwRVgB81Vu2SHV+imzoMAHNtg62dl58WkY5Hzf6Dkmg6VYSxoBjR1EzFJ/8mePJwhzYhk0dQvms/ABX//JbAkQMAsDQ0UZf+E5amC6ugFT51GMU7vwKg+nAO3kH+aJzs1USE4B3gS/Uhyd7inV8Rfr10fNH3TaZg0z8Qjc0AmMqqPdYOmzqMUqu25Gt/174O9Gvt551fETZV0rb3tcpPiweuxm9QEsbThRjPSL6u/PRrgiaPcGgTNHkEFR/tA6DqX98ScN1AAMTGJltcCVpNpyfPdUVb2nYNxoIimnIKOqXrO7AvTacLMVl1qz77msCUaxzaBE66plX382/wt9O9UPyc4/rTtnEdPHkEBmtcV/6rNa4bjuXTXFwOQGN2AV5aNYKmcw+B5cojcuSQFkKmDMewS1r/t64Dm+sOnwDAsOsLQqznVMiU4Rh2St837PyCkKmO8ekOnZ3NNRdgM0DiM/eT/+yfoZPnlVx5pOlU67lc8em/CU5x9FVwygjKP7KP7ZY8YrTFlZdW45HepaDtN1DKX7Y88unXBDnpBqWMoNKWR5zyV8o1GM8U0Zjdufzl32KvNY+Uf/wNIU55M2Rya9xW/PM7AkcNsH4+gvKPv0E0NmM8U0LTqUL8ByV5brOMOSx8yjCKd0hxXZPR/rW55Zwq3vEV4VOHt9lXZ+hqP3eFruZOgPhn7ufUBeQRhcsD2QYumqgwjOfLbH8bzxvQRIW1bVNoaLcNQMRdE6jcn+G5tl5H0/nW/TYVlqN1umHR6nWO2oXlDjc1PVfexdDDb9Dj9tEUrP+wQ011lM7BXlOhAXWUzqmNnU/MFsw1dahCAz22yx1afRhN5+ztNaDVh7VtY2dv0/nWNn6J0QSPuJLkz59n0N/XEDgo0WNtTVQYTXZ2NxUa2twcavQ6jOcdj8++n3uuvJshh7bR47YxFLz0QYea6kgdJmdfR+rct7H5OggA30F96bt3M333bOLck1tsNwGe0BVtwVdLj0dup/i19z3Ws+0zSoepsPXXp+bCMte6LW3MFiw19TabNXGRJH76GvHvr8Nv2NUe62qc4troyt6oMBf2OsZ1yA3X0XAs3zY49lxfnjwiRw5ptcfxfJFyiaM9amebCw1orPlGHR6CqUR6TcpUUoFaF+yZrpPNRjc2O+QRO5vDJg+lqbCcuv+e9kjPQVuOPBKlw1Rofy6XucjZrs5lKbb9BvXlitTXuWLPHzj7+07mEZm0vZ11i1xcpyJ1GAvd56+SC8hfGn1Y6z4BY5EBjd7VeWynW12Pd2igR99tV1vGHKbRhznlERd2O1+bndpEPzCV5P0b6Pvqo3gH+3uk25V+Bul60eezjcR/sA6/YVd5ZqzNnq7lzrDJQzEWllN/AXnkksMiXtr/XaJ0euAiCEKHP0kKgvCwIAiHBEE49I/6tu8uWhu1/cx59Oxy4QjHNtGP3Y7YbKHsb193dFh2+3Ul7aztSrq1TcEL73NoyCOUfvRv9A9MvSDNtr8WuDqwjnd9IeJtpV21kRoJ3l54h/iTcf0T5D3zZ656c4nnyh70s8vK+nZNCl74K4eHzqH0b1+jn3W9J6IdarbXpuFoNtmT55F7yxIiHp2OoFV3rHkRtKMWz6Ts7Y+x1F+k2vsexnRzaTknRs0i7+aFFD73FnGvLsMrwNczDU/i2qW9rf/06RtH9O/upeB3nXvlwP2+f4Y8IkcO8XC/4Nl511lcu9Ezm718NcQtup3TnRigOWrLkEdcd7LzgbnVrD+azYmU+WTfspSIudM6l0fk0m7nOtC+rkjk4pmUvXOh+etCr1Eefrdz0j9fDutgv+7bSI3Ov7eXgyMWkDFxOcbiShJW33vBup72c3NpOcdHPkDuTYsoXPsWcRs7cb2ALuVOL18NsYtu79QPPQqXH+0+0xQE4QVggyiKZYIgDAV2ABZBENTAvaIofuXqe6Iobge2A3wffZstIiPvn0rEzBQAao/mookOt31HE61rMznV6PSrmiZah7GotU349HGEThrKT3c83aGhUbOmEjlzolU7D220jpbpXFp9GMaicof2TeedtF20ASj7+7+58v+e4MxLO9rVNxUaHOxV63WYih33ZyqS2piKDKDyQhXoj7nywiadRc+aQvQ9kwCoPpqLNqbVFq1e59Je+189tHa+bjpfTtk/DwBQcyQXLBbUuiDbhEBnou6fSuRMSbs2MxdtdLidr11oFxrQRDsdX7ErX3/DlX9+gjMb2k9apqIy1M6+LnH2tdSmPV835Z3F0tCIT99eNGTltqt5MbT9BvUl+Ibr0P/uflRB/ogWEbHJiOFP//RA14Ba38P2t7c+3IWu1KbZqusV6Gez2WyU/t/4Yx7GgiI08TE0emCz0SmuNa7sLTS49bU6Skf89t9xevFGjKeLOtQD+fKInDkk4r7r6WG1ue7Cmd4UAAAgAElEQVRorsP5IuWSDmy2O6dMZZWoI0Klpy0RoZgMVW519bOmEmW1ucZqs709TR3lEavNPr2i8OkZQfL+DdbPdQzeu56j1/8OU2mlS+1LIo/o7c/l8LY5u9CDPJJ7gXlEBu3mQifdKJ3tVSj7Y9Pow215xCF/XX8dUSs7n7+keLU7j6N0mJz6t6WNqdCqGyTlL0++26H2z5jD9LOmoLfGdc3RXIdzylVcG9ucU635y1TWeu4W/iWNfn9e6ZHNXelnaHu90MbHtBtfFyt3+vSKQtszgkF2eWTQ3vVktpNHFC4/OnricqMoii3PE18C7hBFsQ+QArzcWbHi93aTlbKUrJSlVOw+SI9p4wAISO6Lubre9vpCC6aSCsy1jQQk9wWgx7RxVOw5CEDwuMFEz/sNJ+5fh6Wh47kfRe/uJnPScjInLad890EiZrRoJ9FcU4+pxDHoTSWVmOsaCEiW3pWNmDGO8j3pAPjER9nahU0ZRkPuuQ716zJz0PbWo4mLQFB7E3rLaKpSDzq0qUw9SNi0CQCE3jiSmm9/6HC/7jj/7h4OTVzOoYnLKfs8ncjpYwEIGiLZa3Sy11hSSXNtA0FDJHsjp4+lbLdkb9nnBwkZ1R8A3wQ9gtrb7aAFoOi93WSmLJMmwn5+kB5W7fZ8balt9XWP6WMp393ia72tXejkoR75uj4zB03vaNSxkQhqb0JuHkO1k6+rUw8QeruUSINvGEntd5Kv1bGRtoms6pgeaBNibJVrPKEr2nkzVnJ81GyOj5pN2TufULJ5p0cXfYCGH7LR2ukG3zSGmrQDDm1q9tnpXj+Kuv9IuqqwINtkVnVcJJre0ZgKPBtE1GfmoI23i+ub28Z1VepBdNa4DrlhJDVWe1VB/iS+t4rzL/6ZukPHPdID+fKInDmk5I+fc2zyEo5NXkLFngPopo0HwL8dmy21DfhbbdZNG0+l1ebKvenopkvf101v/dwVhe/u5sik5RyZtByDnc2ByUmY27E50M5mw5506o8XcKDfg6QPm0v6sLk0FRo4Mvnxdm82LoU8oo2PRhMXaYvt6lTHc6o67SBht7eNbU2cYx7xSYjBeLa4Q025tet/yHHMIzePoTrNKX+lHSDElkdGUmvNIydnrOTE6NmcGC3lr9ItnuevuswcfOzySNito6h0cX1sidvQG6+j5tss2+dht45C0HijiYvAJ15P3dEcj3Th589hhe/usU2mN+xOJ3KGFNeBye6vzfbnVOSMsZRZ84j9fJjw64dTd/yMZzZ3oZ+drxfa3tEYO7heXKzcWX+8gPR+D3J42FwOW/PI0Q7yiMLlh9De5GNBEI4D/URRbBYE4XtRFK+x25YlimL/jgTsn7g40/v5hwgZNxiLtYxp3Q95APRPfZmslKUA+A9ItJUxrfwig1O/fwuAQd9uRtCqaa6Qxu21h7PJX7nNtm+zpf0xWcK62YSMH4SloYncRVuozZS0B6a9ROak5QAEDEykj7UcX+X+I5x8QionesVby/DtEw0WkaazpeQ9vt3hFwNftcmlZtD4IcSuflAqufjhPoo27US/9G7qf8ilKvUgglYtlUPul4C5sob8eRswFkgXnKu/244q0A9B7Y25uo7cmatpzHFMUtVGjVt7k9Y9SNiEQZgbjJxYuJmaTKl86tB9L3FoomRv4ECpHLJUcvEoOVZ7BbU3/7PxUQL69cZibCZvzZ+p/OZH2741Xu2/Px3//GxCxw/G3NBE7uLN1LX4OnUDmSnLAPAfmEiStYxpxf4j5Fv7+Yq3luObGI1o9fXJFdscfB2gdX2zGThuCNHWksQVO9Io2byDyMUzacjKoTpN8nXcK0vwvToBc2UtBQvWYzxTTMhvxhPx6DTE5mawiBT/4QOq93ayjOkFatsTueguzHWNLsuJenm5PqUCxg1Fv+ohBC8vKnamUrplBxGLJN2afQcRNGpiX1mKz1UJmKtqOfPYi5jOFBM09ToiFs1ENFvAbKZk41+p2d/2ZtbUrHKpGzR+CDFPt8Z18es7iVpyN/VZuVRb47rXxsX4XZ1Ac2UNp+ZLcR25YDqR86bRlH/etq+8e1bT7OIJQJMbbZAvj3RnDgHQqMxutXs+9zDBVpvzl2yi3mrz1Xtf4dhk6VVOvwGJreWQv8ig4Mk3AVCFBtLnjWVoYsIxnisjd85LmCtrbftuNLv3deK62YRabc62s3lw2kscsbO5pRxyxf4j5FlttmdY+haOTFnhUMZUJbT/fk935hE/jeucHTh+CDFPzUZQeVG+I601tn/ItZ3LvV6VzuXmyhqpJPGZYkJ/M46IudPA1IwoihS/9gFVew+41HBHd2ur3OSRwHFD0D9lLZO7M43SzTuIsOavmjQpj8S9usSWRwoWrMfklL8iFt6Fpd51/moyuY6v4AlDiFv9AHipMHyYRuGmXUQvu4u6zFyqUtMRtGriX1uEn/X6mDf3Zdv1Ub9gGro7JoHZTMHqt6n+wvVcNXc2d3cOq2nn2txn3YPWc8rIiUWbqbVem5PTXiLDdk7ZlUPef9R2Tl2xaQEB/XojiiJNZ0rJWb7NYeAT6uv+tb0L7eegqdcRuXgmotkMZgvFG/9Czb70NvuvbtC61e5K7rRnSPoWMp3yCMDIol0u3xy91KiaNenSnUgCBL+bdkn6saOBywLgZuAFYAwQAvwNmAgkiKL4244E2hu4dCcdDVy6E3cDl+6mvYFLd9LRwKU7cTdwuZxxN3DpbtwNXH4O2hu4dCdy5pH2Bi7dSXsDl+6ko4FLd+Ju4HI54+4mvrtxN3D5OZDL5vYGLt1JewOX7qa9gUt3owxcLg6X6sCl3TkuoihuEgQhC3gU6Gtt3xf4B/Bs9x+egoKCgoKCgoKCgoJCBwMXAFEUvwS+dP5cEIRZwLvOnysoKCgoKCgoKCgotMMlXHL4UqYr70GsuWhHoaCgoKCgoKCgoKCg0A4dlUN2V9ZKACI9Eag3d24l7MsBjUzvxXvJ9DaiKMr3GmRNozzv0Wq9O7dQ4sXES6a5ACaZ5j4AeMs0j0ouXQA/rTzzLoSmzqw3cvHQeMszp0dOvFXyxZcg03wPORc7l2t+oK9ZvuuFXMg1R0/h8qejUUUkMAWocPpcAL7rliNSUFBQUFBQUFBQuJyR73eLXzQdDVw+AwJEUTzqvEEQhC+75YgUFBQUFBQUFBQUFBSc6Kiq2IPtbLv74h+OgoKCgoKCgoKCgoJCW2SdgJL03Cx0E6VF1P772BZqs/LbtAkcEM+V1kURDfuOkPP71kJmsQ9OJeaBqYjNZgxpGeQ9+5du1716+yL8EqMB8A7yo7m6nvSJj3tsc89nHiR4whBp4bjFm6j/8WSbNn79E1oXjtt/mIKnpIWXVCEBJG5dijYugqYzJeQ9sgFzVZ1HuolrZxE2MRlzQxPZCze7tDlggN0iV/syyHtSsrnXsulEzZyEyVANQP66v1Kx74jHNsevfYDQiYOxNBjJWfg6dS60/QckkNSyaN2+I+Q/+Y7D9uhHbyH+6Xs5cNWsNotNdYduz8fvJGzqMESLBVNZNbkLX8dY7PzGpGuCxg2m55rZoPKi7P1Uijb/zWG7oPEmfuMi/AYk0lxRw8lHN2A8WwJA1LzbCb9rEpgtFDz1JtVftXnY2a5u7Gpp4UvD+6kUb/mojW7vjYvx7Z+IuaKG/LkvYTxbgiokkIRtK/Ab2AfDzv2cXbXdY017ej37ICETkm0LQdZnuY5t20KQ+zM4vUqK7bCbriVm6R34JsVy7IYVtkUkPbFZLl9fiK4qJJDE7Y/jb/V1y6KQncF/zBCiVj2MoPKi4sO9GLbtbKMdvWEpvv36YK6o4exjL2A6J9msvaI3+rXz8QrwA1Ek/38XIRo9m0sjl68BAsYmE2Nd1LX8w1RKt+5qox33yhJ8+yVirqzh9Pz1mKzaAOroHvRN3UzxxvddLop4qemCfP3sP3oIkU/OQVB5UbljD4btLnTXL8OnXx/MlTWcW7jOQTfq2QWoAvwQLSKnblvosa6cvg4Yk4z+qYelxRh37KXsjbbasRuW2Gw+s+BFTOdKUMdEkJS6laaT5wBoOHqC809u7pS2XPcEAWOSiX7augDlh6mUurL55VZ/F8xfb7O5b9oWm831R05w/sktvwibLzVEparYBSHb6mq6iYPxi4/i+2se4/iy7VyxfrbLdlesf4jjy7bx/TWP4RcfRdiEQQCEjLya8KlDOTh+GQfHLqVg66c/i+6xhzeSPvFx0ic+Tuk/D1D6T89XQw6ekIw2PpqsUXM5tWIrvdbNcdmu17pHOLViK1mj5qKNjyZ4fDIA+nm3Uf1NFlmj5lH9TRb6ebd5pBs6cTC+CXrSr11AzrJt9HnxIZft+rz4ENnLtpF+7QJ8E/SEWm0GOLf9MzImLSdj0vJODVpatDOuXUDusjdIfPFhl+0SX3yIvGXbyLBqh0wYbNumidYRMmYAjWdLfzbdc1s+5uiEpWROWk5F6mHilkz3TNjLi55r55D922c4Nn4BYbeOxicp1qFJ+J0pNFfV8uOoRyl+8xNin7gXAJ+kWMJuHcWxCQvIvmcNPZ97BLw8PEW9vIhbO4fce9fw04T5hN46Gp+kOIcmujtTaK6s5b+jH6HkrU+IeeI+AMQmI+c3/IVza9/zTMsFwROS8YnXkzlyHvmPv0H8Otf+jn9hDvmPbyVz5Dx84vUEj5f8XX+8gJzZ66n5/r+ei8ro6wvVFZuMnH/pr5x99j3P7XTS1q9+lIIHniZ3yqME3zwGTR/Hfg6ZPgVzVS25Ex7C8O4/iFgxS9qg8iLmlWUUrtrMyevncvrulYjNHk6glcvXVu2YZx4h//7VZKfMI+SWMWidbA6bMRlzVS0nxs2h9O2P0a+832F79KrZ1Hx52HNNOXWt2nL1c9TquZyZ/RR51z9C0E1j2+pOm4K5upa8SbMpf/fvRCx/wKYbvWE5RU+9zskbHqXgnhWd0pXT19FrHuXUrKfJnTKX4JvHttEOnTEZc3UdORMexvDOx0StaNU2ni4i76bHyLvpsU4PWuS6J8DLi2irv3MmzyPYhb9Drf7OHj+Hsrc/JsrO38bTReTeuJDcGxd2etAim80Klw2yDVzCpw6laOfXAFQfzsE7yB9NRIhDG01ECKoAX6oP5QBQtPNrelw/DICY+yZzetPHiEapWoeprPpn0bUn4pZrKf77t56aTMiU4Rh2fQFAXUY2qmB/1BGhDm3UEaGoAn2pO3wCAMOuLwiZOrz1+zul7xt2fkHI1BEe6YZPGUbxjq8AqMlwb7N3gC81h7MBKN7xFeFW3a4QNmUYJTu+BKA2IwfvID/UTtrqiBBUAX427ZIdX6Kb2urv+Gfu59Szf+5UOZqu6pprG2ztvPy0iHim7T8oiaZThRgLihFNzZR//A0hkx37KWRyaz9W/PM7AkcNsH4+gvKPv0E0NmM8U0LTqUL8ByV1QrfIplvxyb8JnuzYfyGTR1C+a79V91sCR0q6loYm6tJ/wtJk9EjLFaFThlO260sAajuI7Vqrv8t2fUmoNYYbc8/RmHe+U5ry+vrCdC0NTdSm/4Sl6cIqhvkO7Ivx9HlMZ4rA1EzVZ18TOOkahzaBk0ZQ9bd9AFR//g3+1w4EIGB0Mo3HT9F0XHryaK6sAYtns0Pl8jWA36AkjKcLMZ6RtCs//ZogJ+2gySOo+Eiyuepf3xJw3UC7bddgLCiiKafAY005dUG+fvYd4Khb/c+vCZx4rUObgEnXUPW3NEl39zf4WXX9RyXTdCL/gnTl9nXT6UJMVu2qz74mMMXZ19e0an/+Df522l1BrnsCv4GSv202f/o1QSlO/k4ZQaXNZkd/dwW5bFa4fJBt4KLVh9F4rsz2d1OhAa0+rE2bpkKD7e/G861t/BL1hIz4H4Z8/hyD/76awEGJP4tuCyHXXImxtIqG/CKPdAE0UTqM51v3ayo0oI5y3K86Kgyjnbax0IAmSidtCw/BVCK9rmQqqUCtC/ZMVx9Gk51uU6EBjZM9GiebndtEPzCV5P0b6Pvqo3gH+3ukK+1X56Rdjlavc2ij1eucbC5HY20TNnkoxsJy6v972mPNi6EL0HPlXQw9/AY9bh9NwfoPPdQNw1jYGl/GIhe+jrJrY7Zgrq7HOzTQo++6Qx2lw3i+9btSbOmc2oS1tjFbMNfUoQoN9Gj/HaGJCqPJTt943oAmypXdhnbbdEpTJl93RbereEfqMNlpNxeVoY507GfvKB2mwlKbtqWmHlVoEJreMSCK9Hz3GeI/fg3dw7d7rCuXrwHUkTpMzrHtZLNDG1tsByH4aunxyO0Uv/a+x3py64J8/ewdpaPZTtdUVIa3s26kDlORnW6tVTc+BkSIe+dZ4v/xB8Iemuaxrpy+Vtv7EWgubOtrdaRrXwNo4iJJ/PQ14t9fh9+wqzulLdc9gRQ79v3s4noRqXM8n63+BsnmPp9tJP6DdfgNu8pDayXksvmSxCJe2v9dosg2cJEqKjvh/Iu64KqNdZO3F94hARy+/vfkPvNn+r25+GfRbSHiNyM79bTFnbSztuBSu4sB5IE9rttIjc6/t5eDIxaQMXE5xuJKElbf2wltV7t19reL74kiXr4aYhfd7vGg4WLptlDwwvscGvIIpR/9G/0DUy9YuE33ufC11MaD73ou60rYRRsP99+hvqtj9+C86tIByOTrLul2EZc+bNOo7UeiiOCtwm/oVZxbsoFTdzxOYMq1nfjlWC5fu96vZzlbJGrxTMre/hhLfWMnBGXWdbffto1cal/sfvb0GiWoVPgOuYrzS1/i1J3LCUy51vY0pmNZGX3tCg+vFc2l5ZwYNYu8mxdS+NxbxL26DK8AX891LqF7Ao/ytdXm4yMfIPemRRSufYu4jb8QmxUuGzpagDID+BvwviiKns2Wlb73MPAwwOLAIdzkmwBAzKwpRN8zEYCao3n4xIRThfQoUKvX0VTkOPm56bzB4Vdyn2gdTUXl1m3ltvklNUfywGJBrQvEZGg7cfti6gIIKi8ibhxOesrKDn0Rcd/19JiZAkDd0Vw00a37Vet1mJwmfBsLDQ6/+mv0OozFkraprBJ1RKj0K0NEKCZDlVtd/awp6GdOstqci9ZOV6vXYbSzB6Rfv+1tltpYf9Uoa9Up/Esa/f7cvt1Rs6YSOVPyd+3RPLTROlp6RasPa6PddN7ZZqmNT68otD0jGLR/g+2YBu1dT+b1v8NUWtltus6U/f3fXPl/T3DmpR3t2g0t/Rfeus8oHSZnX1vbmAoNoPJCFeSHubLGo++6w1RoQBPd+l0pthy/ayqS2piKrLqB/tLrHBdI5P1THWJbGx1ObcuxR3sQ29GtMXYhyOXrruh2leaiMtR22t5R4ZiKDS7a9KDZ2s9egZK2qaiM+oM/Yq6QXqut/eoQPlcnUvddZoe6cvkapF/91c6xXeIc21Ib59j2G9SX4BuuQ/+7+1EF+SNaRMQmI4Y//fOS1QX5+rm5qAxvO111VDjNrmyOstMNkHSbi8qoT8+y6dZ9dQifq/tQ/5+OdeX0tanIgFrfw/a3tz7chbbBpa8BzEbp/40/5mEsKEITH0NjVq5bPbnuCexpLnSML3WUjuY214syNPpwm8321wtnm7XxMTRc4jYrXD509MQlFAgBvhAE4aAgCIsFQYjuaKeiKG4XRXGoKIpDWwYtAOfe3dM6sf3zg0RNHwNA0JAkzDX1GEscb0aNJZWYaxsIGiK9Dx01fQxluw8BUPp5OqGj+gHgm6BHUHu7HLRcbF2A0DH9qcs5T1Nhxxffkj9+zrHJSzg2eQkVew6gmzYeAP/kvpir622PPFswlVRgqW3AP7kvALpp46nccxCAyr3p6KZL39dNb/3cFYXv7rFNpjfsTidyxlgAApOTaHZnc10DgcmSzZEzxlK2Jx3AYT5M+PXDqTt+pl2bi97dTeak5WROWk757oNEzBgHQIBV2+SkbbJqB1i1I2aMo3xPOvXHC0jv9yCHh83l8LC5NBUaODr5cZeDloupC+ATH2VrFzZlGA2559q1uYW6zBx84vVo4iIQ1N6E3TqKylTHfqpMPWjrx9Abr6Pm2yzb52G3jkLQeKOJi8AnXk/d0RyPdbW9W3VDbxlNlQvdsGkTrLojqfn2B4/27Y7i93bzY8pSfkxZSsXug4RPGwdAQDuxba5tJMAa2+HTxlHRTgx3hJy+vlDdrtLwQzaa3jGoYyNB7U3wTWOo3edYIKRm3wGCb5MG8EHXj6LuP1I/136dgfaK3gg+WlB54Te8P0057Z/LLcjla4D6zBw0vaNRx0YiqL0JuXkM1U7a1akHCL1dsjn4hpHUfifZnDdjJcdHzeb4qNmUvfMJJZt3enxDK5cuyNfPDVnZNptRexN04xhq9n3v0KZ23wGCb5N+FAuaOor67626/87A54r4Vt1h/TDmejbnRG5fa+20g28aQ01aW1/btO18rQoLshWaUMdFoukdjamg/dfH5bonsKf+hxxHm28eQ3Wak7/TDhBis3kktW5s1vaOxvgLsPmSxHKJ/3eJIrR5PGi/URAyRFFMtv57NHAXcBvwE9JTmA7rpu6PnOFWoO+6B9FNGIi5wchPC7dQkymVxBu2b72txHDgwASu/MNcVD4aDPuOkv2EVK5WUKu4cuNcAvr1QjQ2k7vmz1R8c8wjo7uiC3Dla3OpOpzD+T+lutx/oLf7ibc9n3uY4HFSKeb8JZuot5Z9vXrvKxybvAQAvwGJrWUAv8iwlUxVhQbS541laGLCMZ4rI3fOS5gra237bjC7f4DWZ92DhI4fhKXByIlFm6m12pyc9hIZk5YDEDDQrhzy/qPkPSGVH7xi0wIC+vVGFEWazpSSs3ybw8BH1cHrPgnrZhMyfhCWhiZyF22hNlOyeWDaS2TatBPpY9Wu3H+Ek1Zte4akbyFzygqPyyF3RfeKt5bh2ycaLCJNZ0vJe3y7w9MYrXezW93gCUOIW/0AeKkwfJhG4aZdRC+7i7rMXKpS0xG0auJfW4RfvwTMlTXkzX0ZY0ExAPoF09DdMQnMZgpWv031Fxlt9u8luPZ30PghxK5+EEHlheHDfRRt2ol+6d3U/5BLVepBBK1aKods1c2ft8Gme/V321EF+iGovTFX15E7czWNTjc7JrOqXX/3fv4hW2yfXPy6raRxv9SX+TFlKQD+AxJJaCmH/EUGp3//FgChU0fQe+1svHVBmKvrqDuWz4m7n7XtW+XlOot2t6/d0RXd/v/ZjirQ1+br7LtX05hzto2Gn9Z1HgkYN5TIJx9G8PKiclcqZVs+pMeie2jIyqF23wEEjZqYl5fhc7WkfXbhemmyNRB863h0j0wHRGq/PETJi++22X99k1oWX2u83VegChw3hGhrqdyKHWmUbN5B5OKZNGTlUJ0mxXbcK0vwvToBc2UtBQvWYzxT7LCPyEV3Ya5r7FSp3O7W9Va5vzvo7n4WvFznEf+xQ4n8vbUc8q69GLZ+SPjCe2jMyqF2v6QbvWEZPldJZXLPLX7Rpht0y3jCH5mBKIrUfXWIkvXvtNm/yeQ6j/wcfezlxuaAcUPRr3oIwcuLip2plG7ZQcQiSbtm30EEjZrYV5bic1UC5qpazjz2IqYzxQRNvY6IRTMRzRYwmynZ+Fdq9re9kW5wc05B994T+GjcX6cCxw1B/5S1HPLONEo37yDC6u+aNMnmuFeX2GwuWLDeZnPk4pmIZjOYLRRv/As1+9Lb7L/R6P5+pDttBhh27u+evGspO5V3jL+k338L+fCLS9KPHg9c7D5TASnAHaIozupIoL2By+VKewOX7qS9gUt30tHA5XKkvYFLd+Nu4NLddDRw6U7cDVwuZ9wNXLobdwOX7qa9gcvlSnsDl+7G3cClu3E3cPk5cDdw6W7aG7h0J+0NXLqb9gYu3Y0ycLk4XKoDl44iK9v5A1EUzcBu638KCgoKCgoKCgoKCgrdTrsDF1EU73S3TRCEWaIotn3+rKCgoKCgoKCgoKDgFvESLjl8KdOVcshrLtpRKCgoKCgoKCgoKCgotENH5ZDdlR0SgEhPBEK1TZ09JoULRCvK8464nO+m1xg1sug2NMv3/q7oshB+9+MtyPc+vp/61zf/wdgsz1wAueZQyUmDUZ45CL9Gfo3xJdccveoGrSy6AP5ao2zaCpc3Hd19RQJTAOfFFgTgu245IgUFBQUFBQUFBYXLmV9fzZmLQkcDl8+AAFEUjzpvEAThy245IgUFBQUFBQUFBQUFBSc6mpz/YDvb7r74h6OgoKCgoKCgoKCgoNAW2V7UDxw7mNjVD0mL5X2QSvGWjxy2Cxpver26GL/+iTRX1HBq3ksYz5YQOHog0SvvRVB7I5qaOffce9R+5/mq1HLpyq3d69kHCZmQjKWhibzFr1OfdbJNG7/+CSS2LA64P4PTq6TFGMNuupaYpXfgmxTLsRtW2BYWvNRtTlw7i7CJyZgbmsheuJnarPw2bQIG2C26uS+DvCelQnm9lk0nauYkTIZqAPLX/ZWKfUc81k5Y+wBhEwdLC34ufJ06N9p9bdpHOPmktFhbz2UziJo50aZ9qhPactoc/+wDhExMxtJgJHfRJpc2+w9IoM/G+VKM7csgf5Vkc9zjdxI2ZThYLJgMVeQsfB1TsfMbqm0JGJNM9NPWRdQ+TKX0jV0O2wWNN7EvL8G3n7RYXsH89ZjOldi2q6N7kLR3MyWvvd+phQnl0gUIHJtMzNOzEVQqDB/spWRr23Oq5yuL8evfh+aKak7Pl84pv4FJxK2bZ20kULTxfar2fO9CwTVB46RzGZUXhvddn8u9Ny7Gt38i5ooa8udKuqqQQBK2rcBvYB8MO/dzdlWH6xa3IeD/2Tvz+Kaq9P+/0zRJ9zZNaZuUrS3L1xFZyir71lKcEX8zgI4yPwRkk4JsBXHHbRBEBRUQxm3G34wjMOMXHUelgMqoo+wVEehCoUD3dN+SNL2/P26aJmmSpoWagvfzevHS9qXpAM4AACAASURBVD497/s895zn3HPvPeeMSyDGsjlh6QdpFO9oGe9uLzfH+9LSTZiu2Me7T9o2Cre0Pd7eyp3eZF8Lt9sTs1EnDkEwNlB/qZALK1/DXFnrMbv7Mw8QOnGwuDnhyteo/dE527o54aHj5D4psuVhQcTvWI2qWySGy0VkL96MuaKm03K92aYAYp+bh9rST2W66KcC+8fR29JflB08SU5TP7X294QnD0VobMRUUknW8tcxepCzvZW/OqOkVcXap2tZVewaqD50e24R2fc/zdlJS1FPG4Nf7252Jpp7EjFXVPPT2MUUvfkRukfuB6ChtJLsec9zLmk5l1ZupceWlZ2f62V26MQE/GK1pI9KIWftG8RuWOjULvaFReSs3UH6qBT8YrWEThgEQO25XDLnb6Lqu59uGJ/VkwbhH6fl6O3LyEzdSa+NC5za9dq4gIzUnRy9fRn+cVrUEwdaj13d9S9OTF7Diclr2nQD38Q+dvsyMlPfoNdG5/HutXEBmak7OWZlD7Jhf8LJyWs42Qa2N30Om5iAX5yWkyOXkr1mB3EvOPc57oWFZK95g5Mjl+IXpyXM4nPe9n2kT1pFemIqpWnH6bZqZutQHx90zywmZ856MpNSCJ02FlUv+/qlvjsJc0U1GRMWUfLWPqLXzbE7rn18PtVfHffYT69yLeyuzy7iwv1Pc25yCuppY1E5tKlwS5s6O24RxW99hHad2Kbqzl/i/J2rOH/HCrLvX0/XPy4BuYddgKUtZ81+mrMTl6K+y0lb/n0iDeXV/DRGbMsxj4pcwWAkb/Nfufrcu23318KOscQ7IzGFMCfxDrfE+/z4RRS/tQ+tQ7x1T8yn6su2x9trudOL7GvlVh5O54cJKzg9eRX1F/LQLZveJrYqVsfp0Uu4+PAOemxY5NSux4bFXHx4B6dHL0EVqyN0grhPtjbld1R+fZrTo1Oo/Po02pTfdV6uN9sUzf3FiduXkZX6BvEu+qn4jQvITt3JCUt/0ZSzr27fx6mJq0mfvIayNuRsr+QvSTeVvHLVAwb2xnCxAGNuIYKpgbKP/0No0jA7m9Ck4ej3HgKg/N/fEDyqPwB1Z3JoKCwFoD4jFx+VApnSsxdH3uJ6m62eMoySvV8CUH0iA3loIIpItZ2NIlKNPNif6uPinqMle79EnTxcZGZdpT47z2NeZ/A5YspQCnd/BUDViUx8QwJRRobZ2Sgjw/AN8qfK4nPh7q+ISB7Woqy2SjNlKEW7v7RhB6BwYCsiw5AHBVjZRbu/RJM89Jq43vQ5PHkoxXtEdrWF7dTn4ABrHSve8xXhFra5us5qJw9QgQcPogIG9MZ4KR/TZbF+VXx8mJDE4XY2IYnDKf/HQQAqPv2GoJEDbI6NwHi5gPqM3Db56i0uNLWpfIyXbdqUAzs0cTil/7BtUyJbqDeCWZwN6qNSehTjJgU6tuWPWrblsKThlFractknzW25sc5AzdGzNBrat8pQwEAx3k0+l398mJAkh3gnDaesKd7/doh30giMuQUYMtseb2/lTm+yr5Vb8VW6tZ5VH89AqdV4zA6bMgz93i8AqGmFXXP8PAD6vV8QZskjYVOGod8j/r1+zxeEJdvXk87E9WabAgi36aeq29FP2eZsnwAVggcJxVv5S9LNJa8MXJTRGox5Jdafjfl6FFH2yU0RHY6pycbciLmqBrk62M4m7I6R1J3JQTA2dGqu99nhGGzZeXqU0eEtbIz5erc2bZVXfdaGY8hr9seQr0epDW9pk+/aRjcvmYRDm+nzyoP4hga2ga2xYxvzS1E5dN4qrcaBXWrXwYvsl+j9yhKP2V712aGOieXa+6zUajA6np9NHeu+7j4GH9tJl9+NJffFv7fK9I3WYMpvZpoK9CiiHepXlAZjvmP9CkHmr6LL4ukUbX3fYx+9zQVQOLLzS1qyozUu21TAwD70TXudvp+/ypXHtltvBDzh2rZlU74Tn6PDm21ctOX2SBFl408T2zGPRDnzuTnehe2Mt7dypzfZ15Pb5d6JlB860Qa2fY4Q65l9uQpHdr4epaUuKiLCMBWJnyuZispQaEI7LdebbQpa9lMGF/2Uvc/2/VT3dfcy5PgbdJk+htxNH7TK9Fb+6rRq7OT/Oqm8857N2TYUgsPwWebEyMbEr083dI/MJveR7Z2f63V2y3IFT9jX+kijk/ncwh2nNqJR3rv7OTJ8GScmrcFYWE7c+tnXhG7pt2ub/Hc/5+jwpZyYlIqxsIzY9fe3H/yz+ey6XKuNsz+0Mcl94W8cH7KI4n8eRjt3qidQJ0hP6pdA1MpZlLy9j8ba+tY5nYUrFuy03NbZ4n9qT2VwPnEpGdNWE7lkBjKVh/uXeFKnnZ6bZ8W7Z7fXZ4HolbMoeesa4u2t3OlN9nXi6h6ajtDQiP6fh9vAdlasQx7xpD60Vd7gerNNuSq6Df0UQO4L73Ns8GKK//EftPOS2wt1MOmA/CXpplJrG1AOAV4ErgKPAG8Dw4AMYKEgCE4/gpfJZAuBhQCPqfszPain3XFjvh6lLsL6s1KrwVRUamdjytej0EVgKtCD3Ad5cCDm8ipAHJHH7nqESyu3YLxU4LGz3uJ6gx01J5kusxIBqDmVhUoXQXUTW6dpMfHZ6PCEXKnTYCxofaKdO/3cPmvnTkE7azIAVaeyUOma/VFpNRgL7NnGPL3dEybRxvLUrKTC+vv8vx6g33vrWmEnEz1rkoWdbcdWasMxOLANLdjh1vOzZRf89QC3vvdIp/Q5ek4yURZ2dbpYx6rcsA35epSO51dobwNQ8uHX3PLeo1ze7P4JXkN+CQptc/1SRGusnxc2yVRQglIbQYND/QoY2IfQqSOJXjcHeUggQqOAYDCi/8snbpne5DaVa8fWRmByZOeXuGxTTTJkXaGxrh6/Pj2oO53VOtehLSu0mpbcAtHGHbc9MhWI/tixHfNIgXOfAwb2IfSOkWgf8Tze3syd3mJfb27EzPGETR7CuXueapUdef9UO7ZtjhDrWStsmzxiKilHEakW33pEqjHpK3Alb3Gb5I02FT03mShLP1Vt6aeac3Z4y5yd5+hzSxuAkg//wy3/71Euv7jbLd9b+UvSzaXW3rhsBzYBnyBuOLlTEIRQYJ3lmFMJgrBLEIQhgiAMcRy0ANSmZ6KK1aLsFolM4Yv6zjFUpB2xs6lIO4JmxkQAwu4YRdW3PwAgDwkk/t0nyNv4HjXHznnqp1e53mAXvvsZPyau5sfE1ZR9doSIGeMBCErog7my1vpau0mmojLM1fUEJfQBIGLGeMo+P+JYbJv0c/uc/87n1onl+s+OEnX3OACCE3rTUFWLsajczt5YVI65po7ghN4ARN09jpLPjwLYzQ2JmDqMmnOXW2F/Zp1Mr//sCJF3j7eyzVW1mBzYJgd25N3j0VvYtt8Za6YOp9YN25s+F7z7GemJqeKE+k+P0GWmyA6ysJ353FhdR5CF3WXmOEo/E9l+sVqrnTppCHVZV92yAWp/yETVU4eiaxQyhS+hd46l8oB9/ao88D1h08WOOnTqKKr/K9avC3ev4/yY+ZwfM5+Stz+iePsejwcP3uJCU5vSoewWZW1TlWnfO7CPED69ZZtSdouyTmZVxHTBLy4G45VCj7g16Zmoetq05Wkt23J52hHCLW1Z/etRVH3zg8d+uVNteiZKm3iH3TmWSgd2Zdr3qJvifccoqi0+Z9+9jnOj53NutBjvom2tx9ubudNb7OvJDR0/CF3Kb8mYs4HGutbnYBT9+VPOJK3iTNIqyj7/Hs2MCQAEumE3VtcRaGFrZkyg3MIu338UzUzx7zUzm3/fmbhN8kabKnjnM9InryF98hpKbfopdznbXNOcsyPvHk/p5005O9pqFz5lqGc520v5q7NKaOzc/zqrZC1eDdoelMlOCoIwyPL/uYIgdHd2zJ1Odr/LKSBkwmBinnpAXCb3g4MUvr6H6FX3UXs6i8q0I8hUCnpsWUnArXE0lFdxcelmjLmFRC2bSVTKDAw5zZMOs/+wngYPnnB4k/tzsI0Ncpfsnn9cQOj4QTTWGbiw8nXr0pj90l7ix8TVAAT2jyeuaXnLL05w6bE3AVAnD6fnc/Px1YRgrqyh5kwO5+971lq20tfsNZ+rjEqX7F4bHkA9YaC4JPGKbVSni8tbJhx4kROT1wAQNMBmaeBDp8h+VFzesu9rywjq1xNBEDBcLiZzzU67QYCslff18RvmW9gGMlZspzpdjPegAy9y0sqOty6HXHbopJXdx8JGgPrLRWSu2WnXoQjOP7rqcJ99Ze4zWewf56OeMAhznYGslduosfg8IG0z6YmpAAQOiKe3ZTnkskMnybHUsb5vrsE/XofQKGC4UsyFh3faPdkL9nN+AxQ8fjDaJy3LEu85QPG23USunEXd6UyqDhxBplTQ7ZVV+P0qDnNFNbnLNmG6bN/ZRS6/l8ba+jYtk/tzcM2Nzq9z8ITBxDw5H5nch9LdB5rb1A9ZVB6wtKlXVuFvaVOXlr6I8XIh6t+OJ3LJDDA1IAgChVv/TsX+71uU3+iCGzJhMF3XN7flgtf2oF0tcissbbnnlpX494vDXF5FTorYlgFu/XYX8uAAZApfzJU1ZM1aT32m/cDYV+66fgWPH4zOshxy2e4DFG3bTZQl3k0+d3tZ9NlcLsbb6BDvqBX3Yq5xHu86o+tPTjoyd7Ymb7GvhTvgm23IVAoaysSn5NXHM7i4bqe1bB+Z+9zZ/fmFVnbOqteotbBv3f8yZ5JWARDQP755WeIvTpD7+J8AkKuD6fVGKsqYCIxXS8ha9CLm8mqXrJ+LK/dxXrc7uk3Vmdx/ShW3YT5hln4qy6afGnDgRdJt+qlelv6i/NBJLjT1F2+m4t9LB5acnb12l13ODlS5yNkdnL8ABl76yHUn2Ymkv3Ncp15iQPPxV50yjq0NXP4LPAWEApuB5YIg/K9MJhsHvCQIwpDWAK4GLpKuv9wNXDpS7gYuHS13A5eOVGsDl46Uu4FLR6q1gUtHytXA5WaWq4FLR8vVwKWj5W7g0tFyN3CRdH3V2sDlZpSrgUtHq7WBS0fK1cDl55A0cLk+6qwDl9bWl12M+KlYIzAFeFAmk72LOOfF+SYRkiRJkiRJkiRJkiRJknSd5XbgIghCOuKApUnLLf+QyWRzEee9SJIkSZIkSZIkSZIkyVN14nkknVnXshzy09ftLCRJkiRJkiRJkiRJkiTJjVpbDtnVEhYyIOr6n44kSZIkSZIkSZIkSZIktVRrc1yiED8Vc1wYXoaHn4l5a8K4WfDenKKwgPZuLndtKjH4eYWr9tLEQwB/eYNXuN6cSPxLVI3BO5NMG72YR7y1IEFpvXfySIjCe5N5vTVh3FuTtsF7dVvu4735yN66H6k0eqlvVnrnXgR+mYswtFWdecnhzqzWBi7/AoIEQTjleEAmk33ZIWckSZIkSZIkSZIkSZIkSQ5qbXL+A26O3Xf9T0eSJEmSJEmSJEmSJEmSWqq1Ny6SJEmSJEmSJEmSJEm6npI+FWuXvDpw6fHsA4RNTKCxzkD2ytepPX2hhU3AbXHEN+3Oe+gEl54Qd20N/83txKy+B//eXTlzx8PWnX09Veyz8wiblEBjnZGsFa9RczqnhU1g/zh6WXb5Lj94gpwn3gag29rfEz5lGDQ2YtJXkLn8dUyFjtOAWipw7GCin1iITO5D2Qf70e/cY3dcpvRFt3k1/v16YS6r4spDL2C6WgSAqm9PtM8txScoAASBnP+zAsFo8tjfvs/PIWKSuKv5mYd2UOXE3+D+sdz66hLkfkpKDp7k/GPvAhB0aw9ueXEBcpUCocHM2XVvUXnSs3iHjB9E96fng9yHkvfTKNj2zxY+x25ZQUD/eBrKqrjw4GaMV0Sfo1OmE3HvZDA3kvvkn6j8qsUXi27V/ZkHCJ04WNwJeeVr1P7ovH5Zd0I+dJzcJ8X6JQ8LIn7HalTdIjFcLiJ78WbMFTUecYPHDaLr+gXibsh/T6Nw+z9a+NzjlZUE3Cb6fDHlRYxXiggeMwDdutnIFL4IpgauPv8u1d+e9tjfXxq3ST2efQD1xATMbvJIoE0eKbPJI92fmI06cQiNxgYMlwrJXvka5spaj7jeyCEAQWMT0D21AHx8KPsgjeI39todlyl96frSKvz7xWMuryJ36SZrHgFQ6LrQe/82ira+73QXeXfq/fxcNJPE3cV/emg71S7yyC2vijtt6w+eJPOxdwC4ddcKAuJ1APiGBNBQWcvRSWs94t6oeUT9m5HErLoHv95d+enXa627sXvqc9f1C0Dug/595+2q55aV+N8Wj7msipwlYruShwUTt/NhAgb0Qr/nEFee2NUmf9sba3lYMPG71hJo4TbtKu+pOkMe8cb9SJ/n56Cx9M1n3fTNv3p1ibVNZVj65n67lrdoU0cmPewR11v1K2hcAjFPitzSD9Io3tEyf3V7uTl/XVq6CdMV+/zVJ20bhVvanr8k3Ry6luWQr0mhExPwi9WSPiqFnLVvELthoVO72BcWkbN2B+mjUvCL1RI6YRAAtedyyZy/iarvfmozO2xiAn5xWk6OXEr2mh3EveCcHffCQrLXvMHJkUvxi9MSNlFk523fR/qkVaQnplKadpxuq2a2DvXxQbv+QXLnPUXWlAcJvXMsyl7d7M9r5hTMFdVkTVyA/p3/JfLhueIBuQ8xL6eS/8Q2LkxdwqX71iE0eL5bfcSkgQTERvPNiOWcTf0Tt2xy/gXgLZvmczZ1F9+MWE5AbDSaiQMB6PPkLC5s3st3kx4me9Nuej8xyzOwjw/dn1tExv99hjMTlhF+1xj8ene1P7ffJ9JQUc2Pox+k8E8f0fXR2QD49e5K+F2jOTNxGRl/eJruzy8GH8+ra+jEBFSxOk6PXsLFh3fQY8Mip3Y9Nizm4sM7OD16CapYHaETEgDQpvyOyq9Pc3p0CpVfn0ab8juPfe723CKy73+as5OWop42Br/e9tdZc08i5opqfhq7mKI3P0L3yP0ANJRWkj3vec4lLefSyq302LLSY39/cVyLwiYm4B+r5ZQlj8S5ySMX1u7g1KgU/GO1hFnySMXhdNInrOD05FXUX8gjZtl0j7k/ew4B8PFB98xicuasJzMphdBpY1E55BH13UmYK6rJmLCIkrf2Eb1ujt1x7ePzqf7quGc8G2kmDSIgNprvRjzEudRd9N0036ld300LOJe6k+9GPERAbDThljxyZuEWjk5ay9FJayn+5HuKP/neY59v1DxSdy6XrAUb295PWdpV1uynOTtxKeq7nLSr3yfSUF7NT2PEdhXzqNiuBIORvM1/5epz77aNaeG2N9aCwUjei3/jyrPt43ozj3jrfkQzaSD+sdH8d8RyzqX+ib4u+ua+lr75vyOW42/TN/+4cCtHJj3MkUkPU/TJEYo/OeIZ2Iv1K8aSvzISUwhzkr/CLfnr/PhFFL+1D61D/tI9MZ+qL9uevyTdPPLawEU9ZRgle78EoPpEBvLQQBSRajsbRaQaebA/1cczACjZ+yXq5OEA1GddpT47r13s8OShFO/5ysLOxDckEEVkmAM7DHlwgJVdvOcrwpOHAWCurrPayQNU4MHiGf4D+mC8lIfpcgGYGqj412GCJ4+wswmePJyKfx4EoPLTrwm8fQAAQWMSqD93EcM58UmMubwKGj1/x9gleSj5ew4DUHFc9Ffp4K8yMgzfIH8qjmUCkL/nMJFThwIgCOAb7A+IT3UMHj4ZDhzYG8PFfIy5hQimBkr3fU1Y0nA7m7CkYej3fAFA2SffEjy6v+X3wynd9zWCsQHj5SIMF/MJHNjbY5/DpgxDv1cst6aV+lVz/DwA+r1fEGa5xmFTms9Lv+cLwpLtz9uVAgb2xnCxwOpz2cf/ITRpmJ1NaNJw9HsPAVD+728IHiX6XHcmh4bCUgDqM3LxUSmQKT17KfpL4zZJPWUYxW3MI8U2eaTiq3Qwi22p6ngGSq3GI643cghAwIDeGC/lY7osxrvi48OEJNrXzZDE4ZT/Q8wjFZ9+Q9DIATbHRmC8XEB9Rq5nQBtFJA+hwJJHKt3kEXmQP5WWPFKw5zBdLHnEVpHTbqfww2884t7IeaQ+60q7+qlAx3b1Uct2FZY0nFJLuyr7pLldNdYZqDl6lkZD21dpu5ZYN9YZqD56lkaD518CNKkz5BFv3I90SR7qUZvy9aBNRU0bQUGb2tTPX78CBor5y2jJX+UfHybEoX6FJA2nrCl//dshfyWNwJhbgCGz7fmrM0po7Nz/Oqu8NnBRRodjyCux/mzM06OMDm9hY8zXu7W5HmxDvr7FDYtSq8GYp7e3sWF3X3cfg4/tpMvvxpL74t9bZfpGaTDlNzMbCkpQRNkzfaM1mPKLxR/MjTRW1SJXh6DsGQOCQPd3niF231Y0Cz17KtwklVZN/dVmX+rz9fhp7ePopw2nPr+02SavFJVWTNwZT/yZ3k/+gTEnttH7qf9L1vPve8RVasMx2vhsLNCj1Dq7xiVWn82Vtfiqgz36W7fsaPvrZ8rXo3CoOwrH+pWvRxktXhNFRBimInGAZioqQ6EJbQPX5rzz9S2usyI6HFOejc9VNcjVwXY2YXeMpO5MDoLRs+Wef2ncZn64Pf8a8kjkvRMpP3TCY+7PnUOgKUc0c00FehTRDvGO0ti3qaoa5OoQZP4quiyeTtFWz9qvo1TacOqv2vuscmiTKm04BptY1+e1tAkbcQvG4grqcgo84t7IeaS9Uji0K5Hbsl0ZW2lXbdW1xPqauJ0gj3jjfsSxb3bdppr7ZoNN39yktrYpb9UvRZSm+Ro2cR2vs62Nk/xV2M78JenmkduBi0wmC5LJZM/IZLIzMpmsQiaTFctksu9kMtmcVv5uoUwmOyaTyY79b23L7zUtRi1+JQhCqzYeP5p0f35OirUv1+kK9zYmuS/8jeNDFlH8z8No5071BOrJmTk9L5mvnIAhv+Lqqs1cvGctwYm3E2jzFKI95XoUa4tJ1zmJZDz5Z/6TkELGk3/mV68svgauo4krGw/+tm3oltfYg3rQZnnAdRdrAL8+3dA9MpvcR7ZL3Fb5HlxDD/KI7qHpCA2NlPzzsIdYL+QQEewE60m8BaJWzqLk7X001rZ3b4d2xtrBJPK3ozx+2+KKK+URF0bX3D1eS6yvK9breeTnuR9pX5tyNIn67UgKP/Roaz2X2J+lfrU3XwsC0StnUfLWteQvSTeLWnuf+lfgQ8RNKO8GAoG/A4/LZLI+giA86uyPBEHYBewC+F73O2utjJqTTJdZiQDUnMpCpYug2nJMqdO0mJxqdHiKqdRpMBZ49pmSo6LnJBM1azIA1ekiu8pyTKXVYCwotbM35OtR6prZKq0GY6G9DUDJh19zy3uPcnnzB275DQUlKLQR1p99oyMwFeqd2HShoUAPch98ggMwl1dhKiih9siPmMsqxfP/6hh+t8ZT8226S17XuUl0/cMkACpOZeMX0+yLn1aDwSGO9Xn2b2H8dOFWG+3d46wT9Qs/+o5fvez8O29Hidev2WdltAaTQ5ybbEz5os/yENFnT/7WUZH3T7WrX7bXT6H1oH7ZXGNTSTmKSLX4tiVSjUlf4bnPOpvz1mowFdmftylfj0IXgclyneXBgeLnf4hPwmJ3PcKllVswXvLs6dkvjRs1J5lIy3WuPpVlz9dpMLYxj0TMHI968hDO3vOUW663cwhAQ759HlFEa6yfyTTJVFCCUhthzSNN8Q4Y2IfQqSOJXjcHeUggQqOAYDCi/8snLnkxc6egs+SRqlPZ+MVEUMF5qz+OecSQp0dlE2s/nQaDTVxkch8ifz2Mo4nrWvW1STdyHmmvTA7tSuQ6XmfRxlm7aq+uJdbXzPVCHvHG/UjXuUnWNlVp6ZubehfXbaq5b1bpwu24TW3qSOIjHp+Dt+qXqaAEhSPX8TpbbBy5AQP7EHrHSLSPeJ6/Ors68+dYnVmtfSrWUxCEdwVBuCIIwsvANEEQMoG5gIezlZtV+O5n/Ji4mh8TV1P22REiZowHICihD+bKWuunOU0yFZVhrq4nKKEPABEzxlP2uYeTzxxU8O5npCemipNhPz1Cl5njLOzeNFTVYioqd2CX01hdR1CC+D10l5njKP3sKAB+sVqrnTppCHVZV1vl1/2QgbJnDIquUaDwJfQ3Y6k+aD85terg94T+TkxoIVNHU/PfHwCoPnwCVd+eyPxUIPchYNhtGDIvu+VdeWc/3016mO8mPUzxp0fRzhwLQOhg0V+jg7/GonIaqusJHSz6q505lmKLv4aCMtQjfwVA+Jh+1F7wrGOoSc/EL1aLslskMoUv4XeNpjzN/vqVpx1BM3MCAOpfj6Tqm9PW34ffNRqZ0hdlt0j8YrXUnMp0yyv686ecSVrFmaRVlH3+PZoZYrmBbupXY3UdgZb6pZkxgXJL/Srff9R6XpqZzb9vTbXpmahsfFbfOYYKB58r0o6gmTERgLA7RlH1rXid5SGBxL/7BHkb36Pm2DmPeL9EbuG7n3E6cTWnLXmkSxvzSBebPBI6fhC6lN9yfs4GGuvcf7Pt7RwCUPtDJqqeOhRdo5ApfAm9cyyVB+zjXXnge8Kmi3kkdOooqi155MLd6zg/Zj7nx8yn5O2PKN6+p9VO/+o7nzdPqP/0CNGWPBIyuDdmF3nEXF1HiCWPRM8cS8lnx5p9HXsbNZl5dp++tKYbOY+0VzXpmah62rSraS3bVXnaEcIt7Ur961FUffPDNTGbuO2N9bXIW3nEG/cjV97Zb51QX/zpUbs25apvNlfX27Wppr4Z2t+mvFG/atMzUdrkr7A7x1LpwK1M+x51U/66YxTVluucffc6zo2ez7nRYv4q2tZ6/pJ0c0rW4nWo7UGZ7FtgrSAIX8tksjuBpYIgTLEcOy8IQt/WALZvXBzV848LCB0vLq15YeXr1iUE+6W9xI+JqwEI7B9PXNPyg1+c4NJjbwKgTh5Oz+fm46sJwVxZQ82Z3X11JwAAIABJREFUHM7f96y1bLPg/tOs2D/ORz1BXIIwa+U2atJF9oC0zaQnporsAfH0tixlWnboJDkWdt831+Afr0NoFDBcKebCwzvtnraGBTh/lRk0fghRjy9E5uND+d40SrZ/QJcVf6DudCbVB79HplQQ81IqfrfGYS6v4sryTeJkfiD0rgloFs8EBKq/PEbRxndalH+1Osilv/+zYR6aiQMw1xn5afkOKtPFpR5HHNzId5blE0MGxHHrq0vw8VNQcvAU5x8VGWHD+tL3uTnIfOU0Goycffgtqn5o/gRQrTC45IZOHEy39fPAR47+gwPkv7YXXeq91KRnUZF2FJlKQezWFQT0E33OXvISxtxCALTLZqC5ZzKYzeSuf4vKL1rOPxDcXOfuzy+01q+cVa9ZlyK9df/LnElaBUBA//jmZUy/OGFdwlOuDqbXG6koYyIwXi0ha9GLmMurrWX7yl0/KgmZMJiYpx4Ql/X84CCFr+8hetV91J7OojLtCDKVgh5bVhJwaxwN5VVcXLoZY24hUctmEpUyA0NO8yTP7D+sp8HDtz03M9fQIHfJ7/nHBYRZrnO2TR65Le0lTtvkkXibPHLR0pYHfrMNmUpBQ5n4JLH6eAY563Zay250U786MocABPs5H0gFjx+M9knLcsh7DlC8bTeRK2dRdzqTqgNHkCkVdHtlFX6/isNcUU3usk2YLhfalRG5/F4aa+udLidaXOvv0uc+Gx6w5pGzy7dTZckjQw9usi5tHDwgjlssy6rrD54i49G3rX9/y9YlVBzPJO8vaS3KDlG4HjjeqHkkLHk4PZ6bj294KObKGmrP5JAx6xlr2XIf93mk6/rmdlXw2h60q++j9ocsKiztqueWlfhbfM5J2Wz1+dZvdyEPDkCm8MVcWUPWrPXUOzzsclW3ryXWt/13F/Jgfys347711GdesStf7uP8luDnyCPGVvJIR92PVJsVLrl9N8wjfOIAGi19c1ObGnZwo3Vp4+ABcZblkBWWNtXc/9+y9UEqj2dy9S8HWpStVrr+rKqj65erPjJ4/GB0luWQy3YfoGjbbqIs+avygMjt9vIq/G+Nw1wu5i+jQ/6KWnEv5hrn+Qug/8WPPfk23+sqmjTu2uc+dKAiD37VKePY2sClP/Am0Af4EZgnCEKGTCbrAtwrCMKrrQHcDVw6Uq0NXDpSrgYuHS13A5eOlLuBS0fL3Q1HR8rdwEXS9Ze7gUtHyt3ApaPlauDS0XI3cOlIuRu4dLS8lUfcDVw6Wt6q264GLj+H3A1cOlLuBi4dKXcDl46WN/vIG2XgUjihcw9cor7onAMXt3NcBEH4ARjm5PfFMpns2j52lCRJkiRJkiRJkiRJkiQPdS3LIT993c5CkiRJkiRJkiRJkiRJktzI7RsXmUzmajaWDIi6/qcjSZIkSZIkSZIkSZIkSS3V2nLIUYhLITuu+ScDPFo03F/Z9l10r4e8+W16ZZ3KK9zogFqvcBvMXtvHlCqjd74dbmjwXv3ylXnns9gGL7apyEDv1G1vqqZe6RVuhH+dV7jezNn1prbttH69ZDR5J3+B9+bX1DV4r7/w923bxpTXS12U3slftV7qHyV5KC/mvBtZrWXrfwFBgiCccjwgk8m+7JAzkiRJkiRJkiRJkiRJkiQHtTY5/wE3x+67/qcjSZIkSZIkSZIkSZIkSS3lnffjkiRJkiRJkiRJkiT9QiVIuyq0S14buASNSyDGsglR6QdpFO/Ya3dcpvQVNyHqF4+5vIpLSzdhulJkPa7QdaFP2jYKt7zvchMiZwoel0DMU/ORyeXo/76foh3/aMHt/vJKAm7rRUNZJZeWvojxShEBA3rTbUOKxUhGwZb3qfj8uzb73ePZB1BPTMBs2Syv9vSFFjaBt8VZN8srO3SCS0+8BUD3J2ajThxCo7EBw6VCsle+hrmy9W9ng8YmoHvKsmHdB2kUv9Ey1l1fao517tJNmK7ax7r3/m0UbW1brAGCxw2i6/oF4iZXf0+jcHvLePd4ZSUBt8XTUFbFxRQx3sFjBqBbNxuZwhfB1MDV59+l+tu27dAc/9xcwieJsc5Yvo3q0zktbIL6x9F3awo+fkpKD54g+3FxY68eqTOJnjUZk74SgJwNf6Ps4EmP2b2fn4tmkriZ2U8PbXfKDu4fyy2vimz9wZNkPta8qVjXB5KJmZeM0GBGf+AE2c/+9ab2+dZdKwiI1wHgGxJAQ2WtdUNDdwoam4D2yYVi3d69nxJndXvzKvz69cJcXsXlZRsxXS1CERNJ77QdGC6IO9bXnTpP3uPbPPbVW9wm9Xj2AcImJlg33XSWRwJs8ki5TR4J/83txKy+B//eXTlzx8PWjfZa9dlLORu8m7e9kbObFPvcPNSTBtFYZyRz+evUOGlTgf3j6G1pz2UHT5LzuLjhZ/e1vyc8eShCYyOmkkqylr+OsdBxumrn8znuuXmEW3w+78LnoP5x9LHmsJNcePxtu+MxD04j7qnZ/PdXc2kobX3nhpDxg+j+9HyQ+1DyfhoF2/5pd1ym9CV2ywoC+ov91IUHN2O01O3olOlE3DsZzI3kPvknKr9q8WW9S7W3TfkP6E3XDUstRjIKt/yNyhvkfiRkvHhPgNwH/fvO7wl6blmJ/23xmMuqyFnSfE8Qs242MqUvgrGBK8+3/Z5A0s0h78yS8/Eh5pnF5MxZT0ZiCmHTxqLq1c3OJPzuJMwV1Zwfv4jit/ahXTfH7rjuiflUfXm8zdyuzy7iwv1Pc25yCuppY1H1duDek4i5opqz4xZR/NZHaNfdD0Dd+Uucv3MV5+9YQfb96+n6xyUgb1v4wiYm4B+r5dSoFHLWvkHchoVO7WJfWMSFtTs4NSoF/1gtYRMGAVBxOJ30CSs4PXkV9RfyiFk23SOfdZZYZyalEOok1mpLrDMmLKLkrX1EO8Ra+/h8qr9qY6wt7G7PLSL7/qc5O2kp6mlj8HOIt8YS75/GLqbozY/QPSLGu6G0kux5z3MuaTmXVm6lx5aVbUKrJw3CP07L0duXkZm6k14bFzi167VxARmpOzl6+zL847SoJw60Hru661+cmLyGE5PXtOkGXjNpEAGx0Xw34iHOpe6i76b5Tu36blrAudSdfDfiIQJiowm3sMNG3UpE8hCOTEjlyLjV5O74+Kb3+czCLRydtJajk9ZS/Mn3FH/yfetQHx90Tz/IxblPkTVlCaF3jnNetytryJy4EP3b+4h+eI71mPFSAdm/eYjs3zzUtsGDt7gWhU5MwC9WS7olj8S6ySM5a3eQPioFv1gtoZY8Unsul8z5m6j67qc2+eyVnG1heytveyVnW9TUnk/cvoys1DeI3+icHb9xAdmpOzlhac9hE0X21e37ODVxNemT11CWdpxuq2beMD4fu30Zmalv0MuFz702LiAzdSfHrDlskPWYUqdBPbY/9VeKPYP6+ND9uUVk/N9nODNhGeF3jcGvd1c7k4jfJ9JQUc2Pox+k8E8f0fXR2QD49e5K+F2jOTNxGRl/eJruzy8GHw/r1zW0qfrzuWTeuZLMO5aTM/spuj6fcsPcj3R7bhFZs5/m7MSlqO9yck/w+0Qayqv5aYx4TxDzqP09wdnE5VxctZWeW9t2TyDp5pFXBi4BA3tjvJSP8XIhgqmB8o8PE5I03M4mJGk4Zf84CEDFv78haOQAm2MjMOYWYMjMbTPXcLGZW/bxfwhNtOeGJg6n9B+HACj/9zcEjxK5Qr0RzOJ7PR+VEtqxsJN6yjCK934JQPWJDOShgSgi1XY2ikg18mB/qo9nAFC890vUyeI5VnyVbj2HquMZKLWa1n0eIMbaZPG54uPDhDj4HJI4nPKmWH/qEOvEERgvF1Cf0bZYQ1O8CzDm2sQ7yX4/09Ck4ej32sa7PwB1Z3JoKCwFoD4jFx+VApnS8xeEEVOGUrj7KwCqTmTiGxKIMjLMzkYZGYZvkD9VllgX7v6KiOQW+622WRHJQyjYcxiAyuOu2fIgfyqPZQJQsOcwXaYOBSDm/iQuvbYPwSiugGMqqfSMewP7bKvIabdT+OE3rTL9B/TBYFu3/3WY4MQRdjbBk0c055FPvybQpm63V97iNkk9ZRglbcwjJTZ5pD7rKvXZeW1ieitnN7G9lbe9kbObFD5lKEW7m9iZ+IYEoHBoU4rIMORBAdb2XLT7SzTJYpsyVzevDucToELw0Hlv+qyx8bmqHT4DxD8zh5xn3wPBM38Dm+qXpZ8q3fc1YQ51OyxpGPo9XwBQ9sm3BI/ub/n9cEr3fY1gbMB4uQjDxXwCB/b2iHstbUqoN1hjLFMpETz01VbeuM6BjvcEH7W8JwhLGk6p5Z6g7BP7ewJT0z3B+bbfE3RGCY2yTv2vs8orAxdFlAZTXon1Z1O+HkWUxrWNuRFzVQ1ydQgyfxVdFk+ncOv7bedGazDl23JLUERrWtq04AYDEDCwD33TXqfv569y5bHt1kbrqZTR4Rht/Dbm6VFGh7e0yde7tQGIvHci5YdOtMr0dfS5QN/S5ygNxnzXsS5qR6xFXzT2/jq7ztHhLuPdpLA7RlJ3Jsd6I+8RWxuOIa85joZ8PUpteEubfNc2unnJJBzaTJ9XHsQ3NNBjtkobTv3VZr8N+XpUDmyVA7s+r9kmIF5L2PD/YfCnzzPow/UED4z3iHsj+9yksBG3YCyuoC6noFWm2J6bn6o25Jc4zyNNNuZGGqtqkatDAFB2iyL+463Evr+BgKG3euaoF7lNUkaHY7hOecRTeStng3fztjdytrVcrcahPZeicrg5VGk19uz8UrsbyO7r7mXI8TfoMn0MuZs+8IzbiXw2uvDZPoc1+xyeNARDfik1P11qAzO8uf8DjAVO8mZ0uH0fWVmLrzrYo791pWtpUwD+A/vQZ/82+nz+GlcfvzHuRxQO9wSmfCf3I7bn5eaeoPbHtt0TSLp55HbgIpPJQmUy2QsymeycTCbTW/6dtfwuzN3fupXMyUjO8YmBC5volbMoeWsfjbX17QG3kyv+p/ZUBucTl5IxbTWRS2YgU7VxjfT2+u3wpEz30HSEhkZK/nm4XcwWT2dcnFfUylmUvN3eWOM03G2JN4Bfn27oHplN7iPb28h2X65rG9Eo7939HBm+jBOT1mAsLCdu/ey2wF2W68n5yXx98A0L4vjUx8h65j36/cnDV+I3sM9NivztKI/etrhUC6Zzm4biUs6Pnkv2ncvJf/5Nur2Sik+Q/43BbW+bbs9rYnfl/Sw5G7yat72Rs63lOkN7Vs+alPvC+xwbvJjif/wH7bxkD7ne89l5sZ757OOvpNuK6VzycIDmrsAWLzCctjkP/9Yltv1tCqDuVAYZSSlkTVtF5IMzb5D7EWfFtQi2W6Rfn27EPNqOewJJN41ae8+2GzgEjBcEoQBAJpNFA/cDe4BEZ38kk8kWAgsBngi/jRnBPeyOmwpKUOgirD8rtBpMRaVObUwFepD7IA8OxFxeRcDAPoTeMRLtI3OQhwQiNAoIBiP6v3zSqrOmghIUWltuhPXVo9Um3znXVoasKzTW1ePXpwd1p7PcMqPmJBM5SwxT9akslDZ+K3WaFpMljfl6uydmSp0GY0GzTcTM8agnD+HsPU+16i9Yngbb+hytsX6CZfW5oASlNoIGZ7GeOpLodW2PtdUXW3+dXed8vct4K6I1xO56hEsrt2C81PoTeO3cKWhnTQag6lQWKl1zHFVaDcYCe7YxT2/3NE+0EWNtKqmw/j7/rwfo9946t+yYuVPQ/WGShZ2NX0wEFZy3lmsosL/OBge2n06DwXJ+hrxS6xyPqpPZ0NiIQhOMSd9ykunN4jOATO5D5K+HcTTRPbdJpgI9Cm0X68++2ggneUS0aarbPsEB1vplNor/rf8xG2NuAcrYGOpbac/e4kbNSaaLJY/UnMpCpYug2nJMqdNgamMeaau8lbOt5f6MedubOTt6bjJRsyZZ2NmodBqavFBpw1u0Z0OeA9uJDUDJh//hlv/3KJdf3N3pfNbOTSZ6VnMesc1hSm24XY5o8tk+h4k++/WIxq97JAmHNlt+r2HQ/k2cmvoIpuJyl3zRFxt/ozWYHPOmxcaUb6lfIWJ79uRvXela2pRdPLJvnPsRk8M9gUKradmWC0QbV/cEcX96hIsrPLsn6OySVhVrn1r7VKynIAgbmwYtAIIgFAiCsBHo7uqPBEHYJQjCEEEQhjgOWgBq0zNR9tSh6BqFTOFL2J1jqUw7YmdTmfY96uliMgu9YxTV3/4AQPbd6zg3ej7nRs+n5O2PKNq2x+MOsDY9E1WsDmU3kau+cwyVafaTgCsPHCF8+kQAwu4YRZWFq+wWZZ38pojpgl9cDMYrha0yC9/9jNOJqzmduJqyz47QZcZ4AIIS+mCurMVUZJ8oTEVlmKvrCUroA0CXGeMp+1yMTej4QehSfsv5ORtorDN65vMPmahsYh1651gqDzjE+sD3hDXFeuooqv8r+nzh7nWcHzOf82PEWBdv9zzW0BRvLcpukdZ4Vzhc54q0I2hmtIy3PCSQ+HefIG/je9QcO+cRL/+dz60Ty/WfHSXq7nEABCf0pqGqFmORfedlLCrHXFNHcIL4TXLU3eMo+fwogN38jIipw6g5d9kt++o7nzdPLv/0CNEzxwIQMrg3Zlfs6jpCBovs6JljKfnsGADFnx5FPbofAP5xWmQKX6eDlpvJZwD12NuoyczDkO9Zx1/3Q4Z93f7NWKoO2LfnqoM2eWTqaGosdVseHmKdRKvoFoWypw5TrmcdoTe4he9+xo+Jq/nRkkci2phHImzySHvkrZzdxP4587Y3c3bBO5+RPnkN6ZPXUPrZESLvbmKL7dnk0KZMlvYcZGnPkXePp9TSnv1io6124VOGUpd1tVP6nP/OZ5ycvIaTk9egt/E5OEHMI658DrbxWf/5UWrP5fJ9vwc4OnQJR4cuwZCv52TSWreDFoCa9Ez8bPqp8LtGU+5Qt8vTjqCZOQEA9a9HUvXNaevvw+8ajUzpi7JbJH6xWmpOZbbqM1xbm1J0ta/XqrgY6ypn7uTt+5Ga9ExUPW3uCaa1vCcoTztCuOWeQP3rUVR9Y3NP8OcnyHvB83sCSTenZO4mdclksv3AAeDPgiAUWn4XBcwBEgVBmNwa4IeedzoFBI8fjM6yDGDZ7gMUbdtN1MpZ1J3OpPLAEWQqhbgM4K1xmMuryV22CeNl+w4nasW9mGvqnS6t2Sg4n1gUPGEwMU/ORyb3oXT3AQpf30P0qvuo/SHLyu3xishtKK8Sl9W8XIj6t+OJXDIDTA0IgkDh1r9Tsd/5ykf1Jtcvsnr+cQFh4wdZlzFtWor0trSXOJ24GoDA/vHNy5h+cYKLj70JwMBvtiFTKWgoE29iq49nkLNup7XsQJXJuc/jB6N90rIc8p4DFG/bTaQl1lUHjiBTKuj2yir8fhWHuUKMtckh1pHL76Wx1nmsG8yux78hEwYT89QD4nLIHxxsjvfpLCrTLPHespIAS7wvLt2MMbeQqGUziUqZgSGneRJx9h/W06CvsCu/yqh0ye614QHUEwaKy2qu2EZ1urjUY8KBFzkxeQ0AQQNslgY+dIrsR8WlHvu+toygfj0RBAHD5WIy1+y0uxFvcFG/mtRnwwNoJg7AXGfk7PLtVFnYQw9usi7zGzwgjlteXYLcT4n+4CkyHhWX9JQp5NyyZQlB/XogGBvIevo9yr4+Yy3bV+a6zd6oPgPcsnUJFcczyftLWouyIwOdL7MZNH4I2icWIPPxoWxPGsXbdxO5wlK3D4p1u+vLq611+/JDGzFdLiQkeSSRK2YhmBvBbKZoy9+oOuT5jf3Pwa2pd123e/5xAaGWPHLBJo/0S3uJH23ySJxNHrlkySPq5OH0fG4+vpoQzJU11JzJ4fx9z1rL9le6ziPeyNnQ8XnbWznb3Oj+2WHchvmETRhIY52BrBXbqU4X2QMOvEi6tT3H08vSnssPneRCU3t+MxX/XjpoFDBcKSZ77S67tzFyH9ePezvSZ1MrPsdvmG/JYQYybHwedOBFTtr43LQcctmhk9YcZquhR7dzcsrDdssh+/s6nxMROnEw3dbPAx85+g8OkP/aXnSp91KTnkVF2lFkKgWxW1cQ0C8Oc3kV2Utewpgr1m3tshlo7pkMZjO569+i8ouWcz2Uvman3Pa2qbDfTiDywRkIDQ3QKFD46t+p3N9yOeRao/vPxzryOivlzn0OmTCYruub7wkKXtuDdrXYliss9wQ9t6zE3xLrnBTxniD6oZb3BFmzWt4TACRc3td5Z5bbKG/khGv4frfjpfv2i04Zx9YGLmpgHXAXEIX4pWEh8BGwURCEVh+Nuhq4dLTcdYIdLXedYEfK1cClo+Vu4NLRcjdw6Ui1dhPfkXI3cOlIedNnVwOXm1nuBi4dKVcDl47WLzFntzZw6Ui5G7h0pFobuHSkXA1cOlquBi4drdYGLh0pVwOXn0M3ysDl6u0TO/XAJea/hzplHN1ma0EQymQy2TtAGvCdIAhNn1Ujk8mSgc86+PwkSZIkSZIkSZIkSZIkqdVVxR4C9gFLgR9lMtldNof/2JEnJkmSJEmSJEmSJEmSJElNau39+AJgsCAI1TKZrCewVyaT9RQEYSvOF7aTJEmSJEmSJEmSJEmSpOuu1gYu8qbPwwRBuCiTycYjDl564OHAxVfune9oZT7e+3RQ7iW2t75Z9hYXQOal+R4GL30T7035eynWAJV1Kq+xvSXBS3M+vFW3jWa5V7gA3spgJsF78z18zL+8Z4+GBu/UbW+1ZWOj99qUykvzem4kScsht0+tZc0CmUw2sOkHyyDmN0AEcFtHnpgkSZIkSZIkSZIkSZIkNam1gctswG6TAUEQGgRBmA2M7bCzkiRJkiRJkiRJkiRJkiQbtbaq2BU3x765/qcjSZIkSZIkSZIkSdLNLaHxl/e55vWQ1z7UDxw7mOgnFiKT+1D2wX70O/fYHZcpfdFtXo1/v16Yy6q48tALmK6KO8Oq+vZE+9xSfIICQBDI+T8rEIye7T0QOGYwUY8vQib3oXz35+h3OeFuSsWvXy/M5VVcXb4B09UiQqaNRzN/utVO1TeWnP/zEIazFzz2OWhsArqnLBtBfpBG8Rt7W7C7vrQK/37xmMuryF26yeozgELXhd77t1G09X2nG7i59NlLsfYmO3hcAjFPzUcml6P/+36KdvyjBbf7yysJuK0XDWWV4oZ1V4oIGNCbbhtSLEYyCra8T8XnLTf2ak09nn2AsIkJ1o29ak+3rCcBt8U1b+x16ASXnhA3UQv/ze3ErL4H/95dOXPHw9ZNwTozF6D7Mw8QOnEwjXUGcla+Ru2PztmxrzyEj5+SikPHyX1SZKt/M5KYVffg17srP/16LbVtYMc+O4+wSQk01hnJWvEaNadzWtgE9o+j15alos8HT5DzhLjxZbe1vyd8yjBobMSkryBz+euYCsta/H1n4gLEPjcP9aRBNNYZyVz+ukt276ZN+g6eJOfxt+2O6x6cRuxTs/n+V3PtNulzp2u5xvKwIOJ3rEbVLRLD5SKyF2/GXFHT6X2Oe24e4RbueRfcoP5x1g0RSw+e5IIDN+bBacQ9NZv/toEL0Pv5uWgmiRsE/vTQdqqdsIP7x3LLqyJbf/AkmY+9Yz3W9YFkYuYlIzSY0R84Qfazf/WI2+v5uWgmJWCuM3DuoW1OuUH94/ifV1Msm8meIMuGG/NAMjHzplq5F579f52aC7/MNhX/3FzCLfHOWO463tYNiw+eIPtxMd49UmcSPWsyJn0lADkb/kbZwZOtMoPHDaLr+gXiBpR/T6Nwe8u+uccrKwm4LZ6Gsioupoh9c/CYAejWzUam8EUwNXD1+Xep/va0x75KunnknZmBPj5o1z9I7rynyJryIKF3jkXZq5udSdjMKZgrqsmauAD9O/9L5MNzxQNyH2JeTiX/iW1cmLqES/etQ2jwcBKYjw/R65dwef6TZE9dTMhvxrXkzpiCubKa7MnzKX3nQyLXzAOg8qMvyZm2jJxpy8hLfQnT1cI2DVrw8UH3zGJy5qwnMymF0GljUTmw1XcnYa6oJmPCIkre2kf0ujl2x7WPz6f6q+OeMy1cr8Tam2wfH7o+u4gL9z/NuckpqKeNRdXbnht+TyLmimrOjltE8VsfoV13PwB15y9x/s5VnL9jBdn3r6frH5eAvG3NJHRiAn6xWtJHpZCz9g1iNyx0ahf7wiJy1u4gfVQKfrFaQicMAqD2XC6Z8zdR9d1PNwS3ia2K1XF69BIuPryDHhsWObXrsWExFx/ewenRS1DF6gidkABA3blcshZsbDM7bGICfnFaTo5cSvaaHcS94NznuBcWkr3mDU6OXIpfnJawiaLPedv3kT5pFemJqZSmHafbqpmdmgugnjQI/zgtJ25fRlbqG8RvdM6O37iA7NSdnLh9Gf42bAClTkPY2P7UXyn2mHut11ib8jsqvz7N6dEpVH59Gm3K7zq9z03cY7cvIzP1DXq54PbauIDM1J0cs3DVDlx1G7kAmkmDCIiN5rsRD3EudRd9N813atd30wLOpe7kuxEPERAbTfhEcVpq2KhbiUgewpEJqRwZt5rcHR97xA2fNAj/WC3fj1hGRupO+mxa4NSuz6YFZKTu5PsRy/CP1Tpwh3J0wmqOjlvF5R0fdWou/LLb1NHbl5GZupNeG53Hu9dGMd5HrXXbOu2Zq7v+xYnJazgxeY1HgxZ8fOj23CKy73+as5OWop42Bj+Hvllj6Zt/GruYojc/QveI2Dc3lFaSPe95ziUt59LKrfTYstJjXyXdXPLKwMV/QB+Ml/IwXS4AUwMV/zpM8OQRdjbBk4dT8c+DAFR++jWBtw8AIGhMAvXnLmI4Jz4ZMJdXQaNnSzP497fnVn5ymOBJt9vZBE0eQcU/D4jcz74mwMK1VchvxlH58Vdt8jlgQG+Ml/IxXS5EMDVQ8fFhQhKH25ebOJzyf4g+V3z6DUEjB9iqeewfAAAgAElEQVQcG4HxcgH1Gblt4nor1t5kBwzsjeFiPkZLrMs+/g+hDrEOTRxO6T8OAVD+728IHiVyhXojmEWOj0oJ7VhISz1lGCV7vwSg+kQG8tBAFJFqOxtFpBp5sD/VxzMAKNn7Jepk8Rzrs65Sn513w3ABwqYMQ7/3CwBqWmHXHD8PgH7vF4QlD7Owr7SLHZ48lOI9YlusPpGJb0ggisgwB24Y8uAAq8/Fe74i3MI1V9dZ7eQBKo+vt7e4AOFThlK0+0sbdoBzdlAAVRZ20e4v0SQPtR6PfWYOF599DwTPwdd6jcOmDEO/R/x7/Z4vCEu2b5Od0WeNDbeqndz4Z+aQ00YuQETyEAr2HAag8rhYx5QObGVkGPIgfyqPZQJQsOcwXaaK7Jj7k7j02j4Eo7hbvKmk0kPuUAotddsd1zfIn8pjos+Fe74iYqp4nXX3J5H72v/eMFz4ZbapiClDKdwtxrvqhPt4N/lcuPsrIizs9kjsmwsw5tr0zUn25YUmDUe/17Zv7g9A3ZkcGgpLAajPyMVHpUCmvLFX9xSEzv2vs8orAxffKA2m/BLrzw0FJSiiNPY20RpM+ZYnF+ZGGqtqkatDUPaMAUGg+zvPELtvK5qF0/FUvtEaGmy4poISfB25URpMBTbcapFrq5Bfj6XyX20buIj+2LL1KKLt2YooDcYmG3Mj5qoa5OoQZP4quiyeTtHW99vEtPrjhVh7k61wjHV+SctYR2sw5TnGOhiAgIF96Jv2On0/f5Urj223DmQ8lTI6HENeM9+Yp0cZHd7Cxpivd2vTVnmLK5arwZjXXK4pX4/CoVyFIztfj9LhurSda++zIV+PUmtfplJrf26GfHufu6+7j8HHdtLld2PJffHvnZrbVK7BrtxSVA5slVbjEOtS6/mFJw3BmF9K7U+XPGbCtV9jRUQYpiLxczhTURkKTajnbG/57MA1uuAa8u3PzZZryC+lpo1csdxw6q/a1zGVNryFjS27Pq/ZJiBeS9jw/2Hwp88z6MP1BA+M95hruGpfb1vjGuy4OkKH30LCp39k4IdPd3ou/FLbVLiDz3qUDvFWOsbbwUY3L5mEQ5vp88qD+IYGts6M1mC07aPy9S3uCRTR4S775iaF3TGSujM51kGqpF+W2j1wkclkn7abKvNkQpITG0FA5isnYMivuLpqMxfvWUtw4u0Ejmz5VqQtZdqfmnsbvwF9aawzYMhsY0fkpFzBcUjrgh21chYlb++jsba+bUxXZbY0csq9tlh7k936dXYea/E/tacyOJ+4lIxpq4lcMgOZSuEh13XZHl3r9rze6QxccL6rUxvbVruwHpTptBbamOS+8DeOD1lE8T8Po507tVNzXRXc8jo7Ywv4+CvpumI6uZs+8JzXSpl2Jh1wjV2xfw6fnTcXz7ndVkznUnti7argNuQxma8PvmFBHJ/6GFnPvEe/P3n6aY2zPOJo4jrXiNxATkx9lOxn3uNXf1rVyblO0b+ANuW67ri3EY3y3t3PkeHLODFpDcbCcuLWz/aA6eR3bajTAH59uqF7ZDa5j2xvnSfpppTb92wymSzB1SFgoItjyGSyhcBCgKci+nF3SHe74w0FJSi0Ec0nER2BqVDvxKYLDQV6kPvgExyAubwKU0EJtUd+xFwmvgau/uoYfrfGU/NtujtXrGX62nAV0RE0FJXa2ZgKSlBE23CDRG6TxLctX7bKasHOt/dZEa2xvva0ZSu1EVa2PDgQc3kVAQP7EDp1JNHr5iAPCURoFBAMRvR/+cQjn70Ra2+yTQ5chTYCk2Os80tQ6CIwOcTaVoasKzTW1ePXpwd1p7PcMqPmJNNlViIANaeyUOkiqLYcU+o0LSZeGx2e0Ct1GowFnk/O9jYXIPL+qXZspa65XIXWA7ZWg9Hhunii6DnJRM2aDEB1uuhz05VTaTUYC+zLNOTr7c5N5YJb8uHX3PLeo1ze7PwGxFtcgOi5yUTNmiSyT2Wj0mls2OEt2XmOsRZt/HpEo+oeycBDm63nNHD/JtKnPoKpuLwF93peY1NJOYpItfhkOFKNSV/h0l9v+qydm0y0hVtl4dqWaXDCtX06r7Lh+nWPJMGGO2j/Jk654ALEzJ2C7g/NbL+YCCo4b/17g0NbdWT76TTW8zPklVL8yfdiWSezobERhSYYk77lpHHd3Cno/iDW7cpTWahiHOptaz7b5BFDXiklVm6WhRtincTdGbjw/9k70/AoqrRh35VOd/a1Q5JOiJCEZUZZw6ayE8LiOqPAzIgfimzKIlsQxhUVRRYVRwHBdV7fmVFgFp3xFUiCyqgzJBCIoEIWlgDZ93QSujud+n5Up9OddNKdhFAB676uXJDU6brreerUcrrqnPPzPKZ0c6eis5zDqk9k29VtR/k2tqjbTfk2lTS58v+UzICP1rXptsYS0XRt1ui0mJrfg+WXtnptVodrid79ey6s3Ibxgt1MHdclyqhiHcPZE5c0YCvwarOfrUBgax8SRXG3KIrDRVEc3rzRAlD3fSaa3pGoe4aB2p2Au8ahTzliV6Y65QgB90knFf/pY6j5z/cA6A+n49G/N4KnB6jc8B45EEPWRZeCrTuZiaZ3hNXrf+c4qlPsR4zSpxwh4D7pwPafNoba/37ftFAQ8J8+lqrPD7vks6X2+yw8LG5B7U7A3eOoSk61K1OVfITA+6WYA6aPRm+J+eysdZwZO58zY+dT8v5nFO/Y61KjBeTLtZzu2owsPKIj0ERJuQ66eyxVSfbequRUgu+fBEDgHaOp/k7yaqLCrJ3x1ZE98IyJxHip0Kmz8MP9nEpYzamE1ZTvTyVkxgQAfOP6Ya6qtT7Ob8RUVI5ZfwXfuH4AhMyYQPmB1Oar7bZegKI/fsEPU1bxw5RVlB84gnbGRAB82nA36Ovwsbi1MyZS0QF3wYf7yUhIlDq2f5FKj5njAfCN60t9dS2mIvubBVNRBQ36Onzj+gLQY+Z4yvanAeAZrbOWC5oynLrsy93OC1DwwX4yJq8hY/IayvanEjprglO3uabJHTprAmUH0qg9nUvagHkcG7GYYyMWY8gv5cSUJ1q9kb6a+7jiYBramdLntTOd73u5Ys7/YD/HJ6/h+OQ1lNp4/eL6Ym7D62fjLbV4jwyYR9qIxaRZvMfb8AJc/uAAafFPkBb/BMVfpBI+U5ouzX+Y5DY2cxuLKjDr6/AfJrnDZ46jZP9RAIq/SCNozAAAvGJ0CGp3h40WgLwPDnA0fg1H49dQ8kUaYZa67T9MyrUjb72NN2zmeEosdbvki1QCxwxs5nXceJDLCz/PYyr/gwPWzvSl+9MImyXl2y+u9Xzb1u2wWeMpOSDl27Y/TMj0kdScdn59lq7NOjRRodZrc2WS/TZXJqWindHy2qzy9yH2w2fI2/QRNUdPO3Up3LgILR6H2i4UhFPAr0VRzHKw7KIoilEOPmbHj7F3OhT4ThhO2NMLEdzcqNiXRMmOT+ix4kHqTmahTzmCoFET+WoinrfEYK6o5tLyzVInbyDg3oloH50JiOi/OkrRpg9arF9wcxyXz/jhhD1lGQ5530FKd35CyPIHuXIyC/0hyRuxNRHPm6UhiS+v3GT1eo8cSOiauZyf2fYj6Pp6lcO/+00Yhu5Zy3DIe5Mp3r6H0JWzqTuZRXVyKoJGTdTrq/C8OQZzpZ7cZZsxXbS/aQ5d/jsaaq84HA5Z5ea4L0ZX57otutptbC3XE4cR+ex8BJUbZXuSKXxrL+GrHqD2+2yqklMRPNT0en0VXrfEUF9RLQ2HfLGQoF9PIHTxDDDVI4oihW98TOXBIy3WbzC13Smw98sLCJggDWN6duVb1qGFByS9yqmE1QD4DIolpnFY4i/TufDUuwAETRtF7w3zcdf6Y66qoeaHc5x54EWX8t2VXjeh7dcTbnppodV9btWb1iGNbzn4Gj9MkY4Z70GxTcN6fplO7tPvABA4bRS9NszHPTgAc1UNtT+cI3P2C9Z1mxpa/44l+uX5BE0cirnOQPbK7dRkSN7BSVvJSEiUYh4cS1/LsMTlh45zzhJz/3fX4BUbgdggYrhUzNm1u1p86yiXVxRb/zYuZuN8AicOoaHOQPaKHegb3clbyJi8BgDfwbH0sQxjWnHoOGeffK/Feoal7SBj6lq7oVs1qtZH7+vMPlYF+dHn7UQ0kSEYL5eQvWgL5gq9dd1Gs+Nj+VrE3FYvttiN8wmyeDNtvEOTt3Dcxts4HHL5oePkOPCOSNvB8WZek9j2d4f9Ns5DO2kw5jojPy3fQXWGNJLliJTNpMU/AYDf4Bh++YfFluGBT5D5pDREr6BW8ctti/Ed0AvRWE/28x9R/s0P1nW7tfGKaN+N8wieNARznZEzy7dbvcNTtnA0fo3V+4s/NA6Te4IsS8yC2p1fbHsM3wG9aTDWk/P8R1R8c6rNOK+VV93GOexGPKbqzG1fp/psnGep20bOrNiO3pLvuOQtpFtjthkO+dAJa93u/+YyfAf0RhRFDBeLyVqzy67h46cxOnT6TxxG5HPzpOGQP0lpujafzKYqyXJt3rYSb8u1+fzSrRhzCwlbNpOwJTMwnGsaxCXnwfXUO3jKNDT30+viUcaFuMnduAs89EpP7pZ5dNZwmQGcFEXxjINlvxJF8R/OBK01XLqa1hou14LWGi5dTWsNlxuZ1houXY2zhsuNiLOGS1fSVsPlRqWthktX0tZNVlfirOHSlch15nTWcOlK2mq43Ki01XDpSuQ6ppw1XLqS1hou14LrpeFyfkhCtz4Ie59I6pZ5bPOsKYriPkAQBCFeEATfZos70FNcQUFBQUFBQUFBQUGh/bTZcBEE4XHgU2AZcEoQhHttFr/clRumoKCgoKCgoKCgoKDQiLPniAuAYaIo6gVB6A3sEwShtyiKb9DKaJ8KCgoKCgoKCgoKCgpXG2cNF5UoinoAURTPC4IwAanx0gul4aKgoKCgoKCgoKDQbrrz7PTdGWcNlwJBEIaIongCwPLk5S7gfWCgK4LyGq9ObmLHkLObulGUp5Opr8oki9fQIF+nWndBnj2tkrGjulyYZeosDuDvZZDNLRdGkzzHlUYtT0diT1GZBfvngJw3ay7NidwlXnmC9hHk6yAv172fwo2PsyFN5gB2s/yIolgviuIcYFyXbZWCgoKCgoKCgoKCgoINbT5xEUXxUhvLvr36m6OgoKCgoKCgoKBwYyM2KD0uOsLPb3IEBQUFBQUFBQUFBYXrDlln0Yve8AhB8UNpqDOStfwtak6ea1HGZ1AMfRtnJE45zrmn37dbHvHYPUQ/N4cjN8+1m6nWGTEbHiHY4j7Titt3UIx1NuSylOOcbeaOfOweYp6bw3/a4e730sNo46XZtn96fCfVDrx+g6K5+Q+LcfPUUJpynMynPgRgwO7leMdGAODu7019VS2p8Wtd8vZ6cR6Bk+JoqDOQs/Itak+ebVHGe2AMsY0zqh9K58Iz0gy5wXfdRuTq3+DVtyc/3LHWOhu7q8RumEtwfBzmOgOZy7ejbyXX1tl5U9LJefoDabsTZxI+ezKm0ioAzm38M+Upx13yRr/4CIHxcTTUGcle8War9auPZWbzipR0zj0j7eOoJ35L8NSR0NCAqbSSrOVvYSosdznmzuQ76pk5BCUMRzTWc+VCIWdXvom5qrbLvZ3dz71enEfQJGk/t+b2sXGX27hvssTcYKzHcKGQHBdj9h0Xh+7ZheDmRvmeg5S8vc9uuaBxp+fWVXgO6IO5opqLyzZhulyEOjKUvkk7MZy9DEDdiTPkPb3d5Vjl8gL4TxhKz/ULQOVG6V+SKNzx1xbu3ttW4jUwFnN5NecWb8F4qQhVoB8xu9biPbgPpXsPcemZ3e3yyhmz77g4Ip5bILk/SaLYkfvVVXgNiMVcUU3u0s2YLhdZl6sjetD34HaK3vgLJe/8vdt7O+NWR4bSL3mHNd+1x8+Q9/SObh+z7/g4Ip+V6nXZJ0kU72zpjXqtyXth6WZMl+y9/ZK2U7jt2uW60zHLeEz5jBtG+DMLEVRulH9ykNJde1u4I7auxmtAH8zl1Vx6/BVrzB79e6PbsBQ3X28QRc79agWi0fU+tnLe+ylc/8j2xCUofiheMTrSb1tGduLbxG5a6LBc7KYF5CTuIv22ZXjF6AicNNS6TBOhJXDcIK5cKu6Q++hty8hKfJs+rbj7bFpAVuIujlrcQc3cQe10a+OH4BUdzn9uXc7pxHfov3mew3L9N8/np8Td/OfW5XhFh6OdNASAUwvfIDV+Lanxayn6PJXiz1Nd8gZMisMzWkfG6CWce+Jtojc6jjf6lUWce2InGaOX4BmtI2CiFG/t6Vyy5m+m+r8/uhxrI425TrttGVmJu+izaYHDcn02LSAzcRdp1lwPsS67vPtfpE9eQ/rkNS43WgInxeEZo+P47UvJWbOTmFccxxzzykJy1rzN8duX4mlTv/J2fEpG/CoyEhIpSzpG1KqZLsfc2XxXHc7g+4krODl5FVfO5hGx7P5r4u3Mfg6cFIdXtI4TFndMG+6zT+zkxOgleEXrCLS4Kw9nkGETc6QrMbu5EfH8Y5yf+xzZUxcTcPd4PPpE2RUJmjUFc1UNWZMWUvr+p4Svfdi6zHihgJy7Hifnrsfbd9GXy2txR21YRPac5/lp0lKC7h2LZ197t/a3CdRX6Plx7KMUvfsZkU8+BIBoMJK39U9c3vBh+5zdIOaIFx7l3MPryZqyhIB7xjl2V+rJnLiIkvc+JXzdw3bLdU/PR//1sevDexXcxgsFZN+5nOw7l7er0SJnriMt3syEJQQ68AZbvGcmLKL4vU/RNfNGPDOf6q+ufa6hE/VLxmNKt/4xch95juypjxFw9zg0zdyBM6dirtSTPWkBpR/8g9C1c6UFKjciX0sk/5ntnJ2+mAsPrEOsd31ADznv/boboih065/uimwNl+CpIyja8xUA+vQs3P29UYcG2pVRhwai8vWm+lgmAEV7vkI7bYR1efQLD3P+xY/aPUyJ1sZd3UF37AsPc66d7h7TRlCw9zAAVceycPf3QdPMqwkNxN3Xi6qjWQAU7D1Mj+kjWqwr7J5bKfi7a92MgqaOpGTfVwDo0zNRBfigDg2yK6MODULl54XeEm/Jvq8ImjYKgCvZl7mSk+dynLaETB1B4Z6vgcZctx5zY64L93xNyLSRHfI1EjxtBMV7Ja/e4nW4j/28rTEX7/2aYIvXrK+zllN5e0A7qlhn8135dQaYpdHS9Mcy0ei018Tbmf0cNHUkxe10F7cSc7WLMXsN7ofhQj6mi4WIpnoq/3UYv4Rb7cr4Tb6V8r+mSI4vvsHn9sEdiq87eAF8hvTFcL4AY67kLv/s3wRMsT9WAqeMomzfIQDKP/8Wv9GDAGioM1CT9hMNhvaPNCRnzN6D+2K0df/zMP4Jo+zK+CeMosLq/hZfG7d/wq0YLxZwJTP3uvBeDXdHkS3XQySv0eKt+Odh/Kc0804Z1VS//q+Zd8qtGHMLMGRd+1x3NGY5jymvwf0wXsjDdLEAGt2Tm7tHUfk3yV31xTf43Ca5fcfGceX0eQynpack5opqaHB9dE857/0UbgzabLgIguAvCMJGQRA+EgThgWbL2vE1Tks0Oi2GvFLr74b8Mjya3ax46LQY85vKGPPLrDc0wVOGY8wvo/bHC512G1txG/Ltt8/Wbcgvo6adbg9dEFcu266zFA9dcLMywRjyy5rK5JXhobO/AQy89ZcYiyupO2c34FuraMKDMeSVWH835pWiCQ9uUcYu1w7KdASNLrjZfi5F0yxmjS64Wa7ty0Q8Mo24Q1vp9/pjuAf4uOZtFrO0Tvt9rNFpMTbfNpuYb1r3AMOO7qLHfePI3fKxS15H7s7ku8fvJlFxKP2ae9uLJjwY41Vyh7oYszpciym/6Ru3+vwS1GH2+1gdZlPG3EBDdS2qIH9pe6LCiP3nG0T/ZSPeI25xHqTM3ka3bZ5N+aWow5u5bfeFuQFzdQ2qIL92eRx55YrZPVyLKd8m5gIHMYdpMeY3j9kfwcuDHo/eT9Ebf2mXU05vZ90g5bvPv7YR/fFGvEfcfE28nYlZHabF1LxeO6pfLep1k7dQhlx3KmY5j6kw+5jrC1q63cMduzW9I0EUuemDF4j+9A20C117I6AROe/9FG4MnD1x+QBposm/Ar8VBOGvgiB4WJbd2tqHBEFYKAjCUUEQjn5a2/I9d6lQyz+JzVvPjp5UiSJuXhp6rrif3M2fONn81rbP8XpddUetuJ8LHXI7DNrpxjUvEvbr2yn8+3ft0Dpap3Nvux4ztMPdYrUOy0iF8j48SOqoZaTHr8FYWEHM+jkuap3n2uGDUJsiua/8mWPDF1H8t8Po5k53yWuRO1C3P98Rj9+PWN9A6d8OX1Nvh3Ah3+2JucTVmJ06HZepLy7jzJi55Ny9nPyX3iXq9UTcfDsx78C18rpy7nJ4nnFd4TLXLOYO1mtRJGzlbEre/5SG2iuu++T2dtJdX1zG6dGPkH3XCvI3vEvUtnbkuxvl2qXzhygSvnI2Je/Jk+tO72cH67T3Oi7TFceUg0IO3YK7Cu/hN3N51VbO/+YJ/BJua9+TIBnv/bobYkP3/umuOOucHyuKYmNz+h+CIDwFHBIE4Z62PiSK4m5gN8C34TOsNTJ87jTCZscDoD+Rg0eElsYuVR66YIwFZXbrMeTZf0uusZTx7BWOx02hDDm01fJZLUMObiZj+u8xFVc43Cbd3GmEW9zVFrfteg0O3LbfAnjYuD1vCiXOxj304GZOtOLuOXcKEQ9K3qoTOXhGaqm0rlOLocC+w7fkbfoW2iMiGKNNGUHlRuidI0lN+L3DOBsJe3gaPWYnAFBzIhuPiBD0jfFGaFt0NDc2eyKhidDaeduDbu5UdLMnA1B9Itsu1x46bYv9bGyR6ya3qaTS+vf8PyUz4KN1rXrDH55GmMWrz5BibqpfLb2G/FI0zbet0L4MQMnfv+GXHz3Jxa2tnyyvdr5DZk4gcPJwTv/muVadXeFtD2EPTyPU4tafyEYTEWK/3g7EHDR5OD85ibkRU0Epal0P6+/uuhBMRWUOy9QXlILKDTc/b+nVBsBslP69cioHY24BmuhIrpzM7rZekL6Jts2zWqfFVNjSrYkIwWRxq/x8rO6OImfM9fklqHU2MYdrqW8RcwkaXYjV3Riz95B+BEy/nfB1D6Py90FsEBENRkr/5/Nu6+2sG1rm2yM6kjoX8i1XzKaCEtTN63WL+iWVaV6vvYf0I+CO29H9/trnunMxy3hMFdjH7B4egqmw1EGZlm5TQQm1qacwl0sD5ui/PornLbHUfJfRqk/Oez+FGw9nT1w8BEGwlhFF8SWkBslhwLUX720o+GA/GZPXkDF5DWX7UwmdNQEA37i+1FfXYiqyr3imogrMNXX4xvUFIHTWBMoOpFF7Ope0AfM4NmIxx0YsxpBfyokpT7RZcfM/2M/xyWs4PnkNpTZuv7i+mNtw+9m4Sy3uIwPmkTZiMWkW9/E23Jc+OGjtUF/8RRrhM6V5O/2HSTEbm3mNRRWY9VfwHyZ5w2eOo3h/mnV50LiB1GTl2b1O5ojCD/dzKmE1pxJWU74/lZAZUry+cf0wV9ViKrK/qTQVlWPWX8E3rh8AITMmUH7Atc7/zcn/4IC1M33p/jTCZo0HpFy3GrNNrsNmjafkgBSzbX+YkOkjqTl9sVVvwYf7yUhIlDrUf5FKj5njLTG3Xr8a9E31q8fM8ZRZcu0ZrbOWC5oynLrsy23GfDXzHTBhKBFLfk3mwxtpqGu7P4Kc+7nww/2cTFjNSYu7RzvdPRzEfMaFmBup+z4Tj94RqHuGIajdCbhrHNXJR+zKVKccIeh+6YIZMH0MNf/5HgBVsD+4Sac2dVQYmt4RmHJde/VSLi9ATUYWHr11aKJCEdTuBN0zlsok+/1XkZRK8IxJAATdOZrqb793ef2tIWfMtd9n2bvvHkdVsn3MVclHCLS6R6O3uM/OWseZsfM5M3Y+Je9/RvGOvS7f0Mrl7ay7eb49ekdgdDHfsuU6IwuNjTfw7nFUNavXVUk29euO0ei/k7w5s9Zxesx8To+RvEXbr12uOxOznMdU3feZaHpHou4ZBha3PqWlO+A+ye1v49YfTsejf28ETw9QueE9ciCGrNavyyDvvZ/CjYezJy7/BCYByY1/EEXxj4IgFAJvdkZcnpxOUHwccf99i4Y6A9krmrrMDE7eQsbkNQCcXfsOfSxD4lUcOu7yqFLO3MHxcQy3uDNt3EOTt3Dc4s5e+451OOTyq+AuTT5OSPxQbjvyBg11Rn5cvtO6bGTKJuvQxqfXvmsZDllNacoJSlNOWMuF/ep2Cl3slN9IRcoxAuPjGPzdDhrqDJxd+ZZ12YCkVzmVsBqA8+t2EdM4TO6X6VRa+hkETRtF7w3zcdf60/+jp6j54RxnHnjRJXdZcjrB8UMZ8d83paGnVzSNfhKXvIV0S66z1r7TNBzyoRPWXEc/8//wHdAbURQxXCwma80ul7zlKekExscR95/tmOsMZK9s8g5O2kpGQiIAOet209cyHHL5oePWvhW9nnoQr9gIxAYRw6Vizq51zQudz3fvl+YjeKj5xSfSkwf9sUzOr3Pul3M/N7qHWNw5Nu6BSa9y0uI+t25X01DMX6Zb8x1tifmXNjGfcxazuYG89W/T+48vILi5Ub43CUNWLqErZlN3MovqlFTKPzlIz9dW0/fQbsyVei4+vgkAn5EDCF0xG9HcAGYzeU9vx1ypb9snt9fivvjMbvr873oElRuln6RwJfMiutUPUPt9NpVJqZR+nETvbSu5+d9vY66o5tySrdaP3/LdblR+3tLN4dRRZM9ezxUnNx3dIea8594m+n+el4aN3ZssuVda3MmplH+SRNTrq+j35S7MlXpyl212ff3dzdtJt8/IAYStnI1oNoO5gcvtrdty5frZt4n5n+dB5Ub5HskbZvFWJadStieJqNdW0f+rXfoFGcMAACAASURBVJgrukeuO+2V8ZgqeH4nN334IoKbGxX7JHePFQ9SdzILfcoRKvYcJPLVRPocegdzRTWXlksxN1TpKXv/H0T//XVARP/VUfRfpbXts0HOe7/uRkM3HrmrOyO0eLeweQFB+AUQCRwRRVFv8/fpoih+4Uxg+6rYtUTO1/OMokoWr6/K9XHUryaGBnniBXAX5NnTKuHnN5qJnBH7erZ/JKzrHaNJnuNKo3Z9aNOrSXceflPh6iHnQFAude3oEq88QbvJeJ0qr+lEn8FOMrpg33VxMsn85bRufSPR76f93TKPzkYVWwZ8CiwDTgmCcK/N4pe6csMUFBQUFBQUFBQUFBQacfaq2EJgmCiKekEQegP7BEHoLYriG7QyKJOCgoKCgoKCgoKCQusoT5k7hrOGi6rx9TBRFM8LgjABqfHSC6XhoqCgoKCgoKCgoKBwjXDWcCkQBGGIKIonACxPXu4C3gcGuiLwUsvT70LWTk/18mjVKnneTTfLmGuNTDHL+e7wz5F6s7MBEG88rtQ7Oz13DW5u8tRtU72MfeVU8vSVk6vvg5zI+S2zXP1r5OpbI2d/IpPyNEGhi3B2NzAHsBtjTxTFelEU5wDjumyrFBQUFBQUFBQUFBQUbGjzKz1RFC+1sax9Y/IqKCgoKCgoKCgoKCA2KE+lOsLP7/0LBQUFBQUFBQUFBYXrDnleogb8Jwyl5/oFoHKj9C9JFO74q91yQeNO720r8RoYi7m8mnOLt2C8VIQq0I+YXWvxHtyH0r2HuPTM7nZ7b3p+PqjcKPlLEgXb/9bCG71tBd6DYqkvr+bsY1ut3tjdT+Bj8eY+/U6H4o7Z8AjB8UOlyRiXv0XNyXMtyvgOirFOfFmWcpyzT78PwE2JswifHY+ptAqA8xv/7NKkTHLlupHoFx8hMD6Ohjoj2SvedBizz6AY+lgmgqxISefcM1LMvZ6ZQ9CU4YjGeq5cKCB7xVuYq2pdirkj+xkgfMn9hPxuMpgbyH32Haq+PuFI0aZbrrrdEa/f2MFErpuDoHFHNNZz6aUP0X93stvHC+A3Po7I5+YjqFSUfnyQop0t3Te9thLvgX2oL6/iwlLJ7T24L1Ebl1gKCRRs+wuVB/7b5V7fMUOIWDcHQe2OaKon7+UPrTOAt4fOHFNRT/yW4KkjoaEBU2klWcvfwlRY3mUxdzbXjdz0wjwCJg2joc7AuZVvUnvqbIsy3gNjiH79cdw8NVQeOkbus+8BoAr0JXbnajyiQjFcLCLn0a2YK2tciFmq24LKjdKPHdftXq+vxHugdB45v6TpmLLdz5dfat8xJbmvr7rdea98uZbrHPZzjBmgz0tz0cbHYa4zcPrx7ehbuQ/6xR+WoPLUUJqSTvZTH1iXRc6bRuQj0xHrzZQmp3P2xf9t9zYoXL/I88TFzY2oDYvInvM8P01aStC9Y/HsG2VXRPvbBOor9Pw49lGK3v2MyCcfAkA0GMnb+icub/iwQ96bNiwi8/+9wA8TlxF871g8+/a0KxLy2wTqK/WcGvMYhe98Rs8n5zR5t/yZSy92wGshKH4oXjE6jt62jKzEt+mzaaHDcn02LSArcRdHb1uGV4yOoElDrcsu7/6c45PXcHzyGtdmkpUr1xYCJ8XhGaPj+O1LyVmzk5hXHMcc88pCcta8zfHbl+IZoyPQEnPF4QxOTFhBRvwqruTk0XPZfS7F3NH97Nm3J8H3juGHScvIfPB5bnrpUXBrx2EiY93uqLe+rIqcR17ip4TlnF/1Br3fWNn947W4e764iLMPPc/pyUsIumccHs3cwb9JwFyp56fxiyh+7zN06yR33ZkLnLl7FWfuWEHOQ+vp+fJiULm4nzvhNZdXcfaRDZyZ+ji5q7Zx0+vtyLWFzh5TeTs+JSN+FRkJiZQlHSNq1cwujblTubYQMCkOj+gITo5ZzPm1O+m1cZHDcr02Psr5tTs5OWYxHtERBEyMA0C35D6qvjnJyTFLqPrmJLolrp1HojYsIueh5/kpfilB9zio25aYfxwn1e2I39sfU6enLOfCyjfota2d+/k6rNud9cqZaznP2T+rmIHg+KF4Res4cusyMhN30W/zAofl+m1eQGbiLo7cugyvaB3Bk4YAEDj6FkKmjSBt4mrSxq/i4s7POrQd3QFR7N4/3RVZGi4+Q/piOF+AMbcQ0VRP+Wf/JmDKSLsygVNGUbbvEADln3+L3+hBADTUGahJ+4kGQ/tn0pa8+VZv2affEDhlVDPvSEr3fmnxfoffmCavPu0nGgwdHyVNO3UERXu+AqA6PQt3f2/UoYF2ZdShgah8vak+lglA0Z6v0E4b0WGnXLluJHjaCIr3fg2APj0Ld38fxzH7eaO3xFy892uCp0nbWPl1BpilEX+q0zPRRGhdjLlj+zlwyijKPv0G0ViP8WIRhvP5+Azp63K88tbtjnnrfjiHqbAMgCtncnHzUCNoXHsYK2f98m7czxct7n/+m4AE+/0ckDCKsr9K7or/+xa/0YMBEK8YrfXKzUMD7ThJd8Zb98NZ6ossuc5sX64b6ewxZdbXWcupvD1cil2uXDcSOHUkpfuk47UmPRNVgA/q0KBmMQeh8vOi5tgZAEr3fUmgJebAqU3He+neLwmcZr/trcdsU7f/2bJuB0wZRek+25ibjqn6wo7v5+uxbnfeK0+u5TqH/RxjBgiZNoJCy/mr6ph0/tI0O39pQgNx9/Wi6qh0/irc+zUh06Xti3hoCrlv/gPRKA3faiqp6tB2KFy/yNJwUYdrMeaVWH835ZeiDtc2KxPcVMbcgLm6BlWQX6e8Gl0wxvwmr7GgFI0u2L5MuE0ZcwPmqlrcO+lt8msx5JU2+fPL8NDZx+2h02LIbypjyC9DY1Mm4pFpxB16lb6vL8Y9wMepU65cN6IJD8Zg4zfkl9rFA1JejHml9mXC7fcLQOhv4yk/5PwpU2f2syufbQu58n21vIF33E7tqXPWi8K18nYEdbgWU76tu8SBW4upFbf3kH70T3qL/gf+wKWndlhvurra20jAHbdT98NZl3PdyNU4pm5a9wDDju6ix33jyN3ysVOnXLm2xhNuH49Uz+yPS3V4MMZ82/NrKRrLNqpDAjEVSa/DmYrKUWsDXHTanAvyS1GHtazbzvZz4B23U/eD68eUtN7rs2531Ct3rmW5H/kZxgzgoQvGcNn+3OTR7BrroQu2vw/KayrjHRtBwKhfEvfFywz5+/P4DYnt9DYpXF+02XARBCFcEISdgiBsFwRBKwjCekEQTgqCsEcQBF2HrY4GUmjxXMpBoU4/umq5zpZaF8p01O5K3G2Uyf/wAGmjlpIen4ixsJzo9Q+5IHXB2SW5tqzZUdDN/A7H1Wjmj1x+P6LZTMlfD7tidaZsYz93cv/Lle+r4PXsF0Xkk3PI/f2Oa+rtOM7rlsODzlKk9kQmZxKWknnPakIXz0DwUF8TL4Bn3ygi1j3Exfbk2rrqzh9Tua/8mWPDF1H8t8Po5k53xerU2TW5bl3fImYX8nK1nU73c78oIn7fzmOqNfl1ULc77JUz1935nH2jxdzKel27PkuFBHc33AN9SJ/+JDkvfMTN76y6GhslC2KD0K1/uivOnrh8CPwIXAS+BOqAO4F/A2+39iFBEBYKgnBUEISjf9Ofb7HclF+KJiLE+rtap7W+qmItU2BTRuWGys8Hc0W104DawphfikbX5NWEazEVlLVeRuWGyt+7U17d3GkMTd7C0OQtGArK8bB51UmjC8bQzC99s9BUxkMXjNFSxlRSCQ0NIIoU/CkZv6F9nPrlyHX4w9MYnLSVwUlbMRaW4WHj99BprfE0YsgvtXsFzEOnxWizjT1mTiB48jCylmxzyd+Z/ezKZ9tCrrrdWa86XEvMO7/n/IptGC/YTd3Upd7OYCooQa2zdYe0dOeXoHbiNmRfoqHuCp79el0TrzpcS+/dT5K7ahvGXNdyfbWPqUZK/v4N2jtvdeqXI9ehD03nloOvccvB1zAVlNvFI9Uz+wEFjM2ePGlsYjaVVFhfLVOHBmEqrXTqNzar2xqdFlNR85hL29zP0bt/z4WV7Tum4Pqt2x31ypprOe9HfiYxR8ydyvCULQxP2YKhsAyPyGbnJmf3QRFajAXllmVllHx+BIDq49nQ0IBa69+h7VK4PnHWcAkTRfFNURRfAQJFUdwkimKuKIpvAq2ekURR3C2K4nBRFIff59u7xfKajCw8euvQRIUiqN0JumcslUmpdmUqklIJnjEJgKA7R1P9bftH3nHk9Yxu8gbfO4YKB17tzIkW7+1Uf9u+kTqak//Bfmtn+tL9qYTOmgCAX1xfzNW1mIoq7Mqbiiow19ThFyf1qwidNYHSA2kAdu+xa6ePovb0Rad+OXJd8OF+MhISpc6/X6TSY+Z4AHzj+lLfSswN+jp8LTH3mDmesv1SzIEThxC59Ff89PArNNS59k5tZ/ZzRVIqwfeOQdC4o4kKxTNaR82JLJdjl7Nud9Sr8vch9o/PkPfKR9QcPX3NvJ2lNiMLj+gINFFhkvvusVQlHbErU5WcSvD9kjvwjtFUW0bw0kSFWTsOqyN74BkTifFSYZd7Vf4+xHzwLPmb/4eaoz+5HOvVPKY8o5selgdNGU5d9uUujbmjuS764xf8MGUVP0xZRfmBI2hnSMerT1w/zFW11le/mmIup0Ffh09cPwC0MyZScUCqixUH06zHu3Zm09+dx2xTt+9uWbcrk1LRznC8n2M/fIa8Te0/pprc11fd7rxXnlzLdQ77OcWc98EBjsav4Wj8Gkq+SCPMcv7yHyadv4zNzl/Gogrq9XX4D5POX2Ezx1NiOX+VfJFK4JiBAHjF6BDU7taRVhV+HghiG4/SBUHIEEVxsOX/G0RRfNpm2UlRFAc6E6RH3etQ4D9xGD3Xz5OGAfwkhYI396Jb/QC132dTmZSK4KGWhuIbEIO5oppzS7ZizJVOgrd8txuVnzeC2h1zVQ3Zs9dzJcv+Jr5BdPyYK2DSMKLWPwJuKko/SSb/zX1EJP6OmoxsKpPSEDzURL+xAm+LN2fxq1bvwP/sRuXnZfVmPrCeK1kt5+isq2+9k1zsxvkETRxCQ52BzBU70GfkADA0eQvHJ68BwHdwrHU45PJDx8l5UhrSs9+by/Ad0BtEuHKxiKw1u+xuWHzUjgcO6OpcX2kjXoDol+cTNHEo5joD2Su3U2OJeXDSVjISEqVtHxxLX8vQreWHjnPuqXelvHz3Fm4aNfXl0jc91emZnF3bNPyiRmV26OzMftYtm4H2N5PBbCZ3/XtUfZneYv1uQuvHTVfn+2p7wx+fSdiSGRjO5VnXlT17PfUufCt9reJ1c3Ocb7+Jw4h8dj6Cyo2yPckUvrWX8FWSuypZcvd6fRVet8RQX1EtDd16sZCgX08gdPEMMNUjiiKFb3xM5cEjDh1X0xu2bBahi2dgtMl1zv97zmGua42tv2bTmWOq/7tr8IqNQGwQMVwq5uzaXXbfeHprHJ9HujrXpnpVmzm/6aWFBEwYKg2HvOpNar+XYr7l4Gv8MEV6XcR7UGzTcMhfpluHrVcF+dHn7UQ0kSEYL5eQvWgL5gq9dd3uKsd9MfwnDiPyuaa6bY35ZDZVlrrda9tKvC0xn18q1e2wZS2PqZwHWx5TQhvnkeutbrvqFVu5Nnd1riW345i7+hzm8DXxaxBzW29KdnXMVUZNq+6+G+cRPGkI5jojZ5ZvpzpDGtp8eMoWjsZL90F+g6XhkKVpIU6QZbkPEtTu/GLbY/gO6E2DsZ6c5z+i4ptTduufULi3+77nZMOpmLu68dhdMODsv7plHp01XF4ANouiqG/29z7AK6IoznAmaK3h0tW01nC5FrTVcOlKWmu4dDXOGi5dSWsNl66mrYaLwtWntYbLjUxbDZeupLWGS1fjrOHSlbTWcOlq2mq43Ki01nC5Nm55vK01XLoaOYe0bavh0tUoDZerQ3dtuLT5qpgois8CPQVBiBcEwdfm79nAu129cQoKCgoKCgoKCgoKCuB8VLFlwKfAMuCUIAj32ix+uSs3TEFBQUFBQUFBQeFGRBSFbv3TXXH2js9CYJgoinpBEHoD+wRB6C2K4hu0MuKmgoKCgoKCgoKCgoLC1cZZw0XV2L9FFMXzgiBMQGq89MLFhou3lzzvS8v5bqenuX0Tyl0tjCZ53hH38+zYDLpXA7NMY40bTPL165GraqtkfB/faJZnP8vZV04uqgwecm/CNadBnlM2ZnnmgJYVFfL0JwJwk+lwFmQ6a5tlPH8Fe12Rza1wY+PsrFkgCMKQxl8sjZi7gBDA6YhiCgoKCgoKCgoKCgoKVwNnXxvPAey+ixJFsR6YIwjCri7bKgUFBQUFBQUFBYUbFDnfDLqeabPhIopiy0lKmpZ9e/U3R0FBQUFBQUFBQUFBoSWyvajvM3YYoU8tQlC5UbH3AGW799otF9Tu6LYk4nlLH8wV1eSt2IjpchH+d08geP791nIe/aM5/+vHMfx01mVv2NMW754DlDb3atyJ2JyI5wDJe3m55JVcvQl/cRkqX2/EBpHz9y1HNLreh8d3XBy6ZxeCmxvlew5S8va+Fu6eW1dZ3ReXbcJ0uQh1ZCh9k3ZiOCvNcF134gx5T2932es/YSg91y8AlRulf0micMdfW3h7b1uJ18BYzOXVnFu8BeOlIlSBfsTsWov34D6U7j3EpWd2t2JoO+aI5xZIMX+SRLGjmF9dhdeAWMwV1eQu3WyNuV/yDmvMtcfPkPf0Dpe9fuPjiHxuPoJKRenHByna2TLmm15biffAPtSXV0mTqF0qwntwX6I2LrEUEijY9hcqD/y33XH3enEegZPiaKgzkLPyLWpPtqyf3gNjiN22DDdPDRWH0rnwjDTBVvBdtxG5+jd49e3JD3espcYy0Z6r3qBJcZjb8PrYeMttvDc9M4eghOE0GOsxXCgkZ+WbmKtqXXbf9MI8AiYNkyYIXPkmtaccx2ydIPDQMXKfldyqQF9id67GIyoUw8Uich7dirmyplvHHP3iIwTGx9FQZyR7xZvUnDzX0jsohj6WSSArUtI598z7AEQ98VuCp46EhgZMpZVkLX8LU2F5i893N3fshrkEx0u5zly+Hb0Dr++gGPq/0ThxXDo5T38AQK/EmYTPnmyd5frcxj9TnnLc5Zjlcvd5aS5ai/f04617f/GHJag8NZSmpJP9lOTtnTgT3YNN3rMv/5mydsTc76WH0cZLE43+9PhOqh24/QZFc/MfFuPmqaE05TiZT30obdMtvfjFlgW4eagR682cWfceVcddO5fI5e1MrgEi500j8pHpiPVmSpPTOfvi/7rkBfnqV8yGRwiOH0pDnZEzy99yeCz7DoqxTkpdlnKcs0+/b7c88rF7iHluDv+5eS71ZdUuxyxHvn3HxxH5rHQ/UvZJEsU7W94TRL3WdE9wYelmTJeK8Brcl54bl1oKCRRu+zNVHbg2K1z/yNMz0M2NsOcWc2nBs5y941H87xqPJjbKrkjAzKmYK/WcTZhP2Yd/p8eaRwCo+udXnL93GefvXUb+mlcxXS50udGCmxvh6xdzcf6z5Ey3ePvYewNnTMVcpSdn8nzKPvg7oRYvKjcitq6h4Nm3OHvHY+Q+uBaxvh2TH7q5EfH8Y5yf+xzZUxcTcPd4PJq5g2ZNwVxVQ9akhZS+/ynhax+2LjNeKCDnrsfJuevxdjVacHMjasMisuc8z0+TlhJ071g8+9p7tb9NoL5Cz49jH6Xo3c+IfPIhAESDkbytf+Lyhg9d9zWP+YVHOffwerKmLCHgnnGOY67UkzlxESXvfUr4OvuYs+9cTvady9vVaMHNjZ4vLuLsQ89zevISgu4Zh0ezmIN/k4C5Us9P4xdR/N5n6NZJMdeducCZu1dx5o4V5Dy0np4vLwZV+w6TgElxeEbryBi9hHNPvE30xoUOy0W/sohzT+wkY/QSPKN1BEwcCkDt6Vyy5m+m+r8/tssbOCkOr2gdJyzemDa8Z5/YyYnRS/CK1hFo8VYeziBj4gpOTl7FlbN5RC673+HnW4vZIzqCk2MWc37tTnptXOSwXK+Nj3J+7U5OjlmMR3QEARPjANAtuY+qb05ycswSqr45iW7Jfd065sBJcXjG6Dh++1Jy1uwk5hXH3phXFpKz5m2O374UzxgdgZMkb96OT8mIX0VGQiJlSceIWjXTJa+c7qD4oXjF6Ei7bRlZibvos2mBw3J9Ni0gM3EXabctwytGR9AkazdJLu/+F+mT15A+eU27Gi1yuYPjh+IVrePIrcvITNxFv82Ovf02S94jty7DK1pHsI330q5/cTR+DUfj17Sr0aKNH4JXdDj/uXU5pxPfof/meQ7L9d88n58Sd/OfW5fjFR2O1uLu8+xszm3dR2r8Ws5u3kOfZ2Z3a29ncx04+hZCpo0gbeJq0sav4uLOz1zygnz1q9F79LZlZCW+TZ9Njo/lPpsWkJW4i6NW71DrMk2ElqBxg7hyqdjleEGmfLu5EWm5J8hMWEKgg3uCYMs9wZkJiyh+71N0lnuCK2dyybp7JVl3LOfcnOfo+dKSdl+buxsNotCtf7orsux1z0H9MF7Iw3SxAEz1VH1+GN/Jt9mV8Y2/lcq/JwNQvf8bvG8b3GI9fneNp+pfX7vs9XLg9Ytv5p18K5V/k7xVNl6fMXEYzpzDcFr6RsJcUQ0Nro+O4jW4H4YL+ZguFiKa6qn812H8Em61j2fyrZT/NQWAyi++wef2ljG3F58hfTGcL8CYK3nLP/s3AVNG2pUJnDKKsn2HACj//Fv8Rg8CoKHOQE3aTzQYOjZqmPfgvhhtY/7nYfwTRtmV8U8YRYU15m/xvQoxew/pi+F8PkaLt/yf/yagmTcgYRRlf5Virvi/b/EbLXnFK0YwS/vVzUPToSG8gqaOpGTfVwDo0zNRBfigDg2yK6MODULl54X+WCYAJfu+ImiatI1Xsi9zJSevQ97idnqLbbyVX2dYY68+lolGp3XZHTh1JKX7vgSgxom75tgZAEr3fUngtJFNn98rfb5075cETrPfX90t5uBpIyje+7XFm4W7vw/q0MBm3kBUft5N3r1fE2yJ16yvs5ZTeXu0q57J5Q6ZOoLCPZK32uLVNPNqQgNx9/Wi2uIt3PM1IdNGtlhXe5HLHTJtBIWWXFcda9tbddTi3fs1IdM7H3OPaSMo2HvYRXcWAAV7D9Nj+ghpoQgqPy8A3P29Mbj4VE0ub2dzHfHQFHLf/AeiUeqWayqpcskL8tUv7dQRFO35ysbr7fhY9vW2eov2fIV22gjr8tgXHubcix+1u8OEHPn2HiLdEzRemyv+eRj/Kc3uCaaMaroP+r+mewLxisF6rhY8NIhKB5GfLe1uuAiCENpZqTpMS31BifX3+oIS1GHalmXyLd8gmBtoqK5FFeRvV8b/jnHtari4h2upz2/ymgpKcG/mdQ/TYiqw8eolryY6EkSIev9Fov/xB4IXzHDZC6AO12LKb/pGpD7fccymVmLWRIUR+883iP7LRrxH3NIurzHPJub8UtThzbzhwU1lzA2Yq2tQBfm1JzyHuIdrMdnl24E7TIsxv7m7KeY+/9pG9Mcb8R5xs8tedXNvfomDmLWYWonZe0g/+ie9Rf8Df+DSUzusJ0tX0YQHY7DJuTGvFE14cIsyxvzSNsu0F43tfuykN/R3k6g4lN4OtxZjXtN6pXpmv151c3d+KRrLflGHBGIqkm5wTEXlqLUBLnrlibn5Pjbkl7Zo9Gh09jkx5Nt7b1r3AMOO7qLHfePI3fKxS1453RpdMIbm69QFtyyT33qZiEemEXdoK/1efwz3AB+XvHK6PXTBGC7br9OjmdejuTfPvkzkI9MY/uVW+m9rX8weuiCuuOQus3GX4aGTGu6Zz/yRvs8+yOj07fR57v+R89Jfurm3c7n2jo0gYNQvifviZYb8/Xn8hsS65AX56pdGp7XzGvPL8Gh2LHvotM28ZdbjPXjKcAz5ZdT8eMEln/16r32+1WE2110s1wlH90Etrs3SPYHXkH70O7idfgfe5PLT7b82K9wYtNlwEQQhuNmPFkgVBCFIEISO32UJDh5BNW89Oyhj28L2HNSfhjoDxqz2HLDOvUIr2yaoVHgNu5m81Vs4/9s1+CXc5vApULtoEbPjMvXFZZwZM5ecu5eT/9K7RL2eiJuvl2uOVtbptNDV+DLDyT5srUxjzKdHP0L2XSvI3/AuUdvaEbML+9mxV/qn9kQmZxKWknnPakIXz0DwULvobX3dLsXd2aR38Lhq7o14/H7E+gZK/na4HW5Hq3Xt2OoUMsXsSiwOH7TbFMl95c8cG76I4r8dRjd3ukteWd1tHDNtl5EK5X14kNRRy0iPX4OxsIKY9XNc88rqdnQsO/c2Hu+X/3iQ/45axtFJkjf2+XbE3MHzWGORng8nkPnsH/k2bglZz/6RX77+6HXnbU+uBXc33AN9SJ/+JDkvfMTN76xy0et4vdeifjk+Nbl2X+DmpSFqxf1c2PyJSy4Hdqfqq57vjp6vLWXqTmSSOWUJ2fesIvSxme2/NnczunLW+6vx011x1jm/BGjeMogE0pEO6xhHHxIEYSGwEOD50FuYFXCT3XJTQQnu4SFNGxEegqmorGUZXQ/qC0tB5YabnzcNFU2dzvzvHEf151852Xx76gtKcNc1edXhIdQ78KrDe1BfYPH6emOuqKa+oITatJOYy6XHoTVfH8Xzlj7U/ifDJbepoBS1rkdTzDpHMUtlrG4/yQ1gNkr/XjmVgzG3AE10JFdOZjv35peiibCJWafFVNjSq4kIwWTxqvx8rN7OUJ9fgtou31rqW7hL0OhCrDHbupvH7BEdSZ0rMRc08+pCWsacX4LaScyG7Es01F3Bs18vp96wh6fRY3YCADUnsvGICEFvWaaJ0Lbo/Gxs9i25JkKLscD1ztm23lCLV38i225fayK0GNvpDZk5gaDJw/npN885dYc+NN0uveP/2AAAIABJREFUZk1E03qleubErdNitOwXU0kF6tAg6WlLaBCm0spuF3P4w9MImz1Z8mZI+7ixxnjotBgL7OuYIb/ULiceNvHaUvL3b/jlR09ycWvrNyByuXVzp6KzeKtPZOPRfJ3NvMa8UrtvjKUylidpJU37NP9PyQz4aF2r8crpjpg7lYgHJW/ViWw8Itv2Gpp7beqXqdjG+7/JDPzftmPuOXcKEQ/GW9w5eEZqaVyDh06Lodk5ovnTHY+IYKtbN2u8tcN80Wf/5ZevOe53Jqf3aubakFdGyedHAKg+ng0NDai1/tYO882Rq37p5k4jfHa8xZtj59XogjE4i1kXjLGgDM9e4XjeFErcoa3W7Rl6cDMnpv8eU3GFQ7ec+QbLtbn5/YijezBn1+Yc16/NCjcezl4VewI4A9wjimK0KIrRwCXL/x02WgBEUdwtiuJwURSHN2+0AFw5mYmmdwTqnmGgdsf/znHoU+xHh9AfOkLAr6UDzG/aGGr/833TQkHAb/pYqj5vxzfCQJ0Db3Vzb8oRAu6TvP7TxlD7X8mr/3c6nv2jETw9QOWG94gBGLNzXXd/n4mHxS2o3Qm4axzVyUfsylSnHCHofumEFjB9DDWWmFXB/uAm7Sp1VBia3hGYcgtc8tZkZOHRW4cmKhRB7U7QPWOpTEq1K1ORlErwjEkABN05mupvv3e0qnZT+32Wfcx3j6Mq2d5dlXyEQGvMo9G3ErNH7wiMLsZcm5GFR3QEmijJG3T3WKqS7HNdlZxK8P1SzIF3jKb6O8mriQqzdvhTR/bAMyYS46VCp87CD/dzKmE1pxJWU74/lZAZEwDwjeuHuarW+hpUI6aicsz6K/jG9QMgZMYEyg+kNl+tS96TCas5afH2aKe3h403YMJQIpb8mjMPb6Shznm/pqI/fsEPU1bxw5RVlB84gnbGRAB82nA36Ovwsbi1MyZSYXFXHExDO1P6vHZm09+7U8wFH+4nIyFR6tT+RSo9Zo63ePtSX12Lqcj+ZsFUVEGDvg7fuL6Sd+Z4yvanAeAZrbOWC5oynLrsy93Snf/BAWuH49L9aYTNkrx+Fq+xmddYVIG5pg4/izds1nhKDkhe2/fnQ6aPpOb0xTZjlsud98EBa2f6ki/SCLPk2n9Y6956fR3+wyzemeMp2e/Ae4fzmC99cJDU+LWkxq+l+Is0wmeOc+o2669Y3eEzx1FscRsKygm8XXrFNmjsAGrPtn7+lMt7NXNd8kUqgWOkebG9YnQIavc2b6Llql/5H+zn+OQ1HJ+8htL9qYTOmmD1mls5lm29obMmUHogjdrTuRwZMI+0EYtJG7EYQ34px6c80WqjBeTNN0jXZo3NPUHg3eOoanY/UpVkcx90x2j0lmuzuqf9tdkjJhLjpaI2fQo3JoKzDk6CIPQEXgcuAs8BGW01Wppzut8dDgU+44cT9uQiULlRue8gpW9/QsjjD3LlVBb6Q0cQNGppOOSbYzFXVpO3cpPUqR7wHjmQHolzuTCr9UeTrYXlM344YY3DMO87SOnOTwhZ/iBXTjZ5I7ZavBXVXLbx+t8zkZBHZyGKIjVfH6Vo8/sOHWaz4/ag74Th6J5ZgODmRvneJIp37CF0xWzqTmZRnZKKoFHT87XVeN4cg7lSz8XHN2G6WIj/tNsJXTEb0dwAZjNF2/5M9aGWN3ZGk8qh13/iMHqun4egcqP0kxQK3tyLbvUD1H6fTWVSKoKHWhoOeUAM5opqzi3ZijFXulm/5bvdqPy8EdTumKtqyJ69nitZ9idltXvr75n6TRiG7lnLcMh7kynevofQlZaYk6WYo15fZY05d9lma8xhK2cjms1gbqBw25+oTklrmesGx48z/SYOI/LZ+QgqN8r2JFP41l7CV0kxVyVLMfd6fRVet8RQX1EtDYd8sZCgX08gdPEMMNUjiiKFb3xM5cEjLdZvMLX9sLL3ywsImDCUhjoDZ1e+ZR3SeEDSq5xKWA2Az6BYYhqHQ/4ynQtPvQtA0LRR9N4wH3etP+aqGmp+OMeZB160rrutI7b3ywsItHhzbLwDk17lpI031sZ73uId8u12BA819eXSt1v6Y5mcW9c0z6xKaPtccdNLC60xn1v1JrUW9y0HX+OHKdKx6j0otmk45C/TyX36HWndQX70eTsRTWQIxsslZC/agrlCb123uY3H1l0Zc1sjq0S/PJ+gidKQsdkrt1OTIXkHJ20lIyFR8g6Opa9lSOLyQ8c5Z/H2f3cNXrERiA0ihkvFnF27q8W3nW3Rle56sfXvs/psnEfQxCHS0K0rtqPPkEZ0jEveQvrkNQD4DrYZMvbQCXKelIae7v/mMnwH9EYURQwXi8las6vFjVJbdKW7oY2q3XfjPIInDcFcZ+TM8u1UW7zDU7ZwNF7y+g2WhoyVhqs9QZbF+4u3JC+iyJWLxWQm2nvNTr477L/xEYInDaahzsiPy3da3SNTNpEav9bqloYlVlOacoLMJ6XhagNG9qffhocR3FU0GIycWfse1d+3HO72WntVtH696EyuBbU7v9j2GL4DetNgrCfn+Y+o+OaU3frd2nj7pSvrl9DGWTt243yL10Dmih3oLcfy0OQtHLd6Y63DIZcfOm712jIibQfHp661Gw65rfMmdG2+g72uOHT6TRhGhGU45PI9yRRt30OY5Z6g8doc9Zp0bTZXSPcExouFBP56IqGPzUCsr4cGkcI/fEzVQcfDIQ86/8/u+56TDelR93brEQbiLn7aLfPotOFiLSgIdwNPAb1FUQx3VdBaw6WrkXPAidYaLl1Naw2XrqathktX01rDpatx1nDpSuSq2s4aLl2JswtwV9Gdh4TsKtpquNyotNVw6UqcNVxuRNpquHQ1bTVcupK2Gi5diVznTWi94XItUBouV4fu2nBxetYUBOEXgiDEA18CE4HJlr9P6+JtU1BQUFBQUFBQUFDohgiCME0QhDOCIGQLgtCic5cgCB6CIHxiWX5EEITenXU6G1XsceBTYBlwCpgiimLjs8CXOytXUFBQUFBQUFBQ+Lkh9wSTnZ2AUhAEFbAdmA7cDPxOEITmc1fMA8pFUeyD1O1kU2fz5uyJywJgmCiKvwImAM8IgrC8cZs7K1dQUFBQUFBQUFBQuO4YCWSLonhWFEUj8DFwb7My9wJ/tPx/HxAvOBzT33WcvaivEkVRDyCK4nlBECYA+wRB6IWLDRd9raYz29dhVG7yvTqocpPnHd46kzxjmteZwMO9Xha3v58879F6iSZZvAANZnm+M3BTyXdMXS7v/ISo1xuebmZZvFdEefrKAbjL1P/Bz12+41ku5Oq/pVbJU68BamS6RhplPKaCNPJcI/UGee79FK4pkUgDdzVyCRjVWhlRFOsFQagEtEjTrXQIZ09cCgRBGNL4i6URcxcQAgzsqFThxkKuRouCgsKNg1yNFgWFGxW5Gi0KriH3BJPOfgRBWCgIwlGbn4XNQnBlJu2rPtu2sycucwC7u1JRFOuBOYIg7HL8EQUFBQUFBQUFBQWF6xVRFHcDu9socgmIsvm9J5DXSplLgiC4AwGA62P/O6DNJy6iKF4SRdHh7FGiKH7bGbGCgoKCgoKCgoKCwnVJGtBXEIRoQRA0wG+Bz5qV+Qx4yPL/GcAh0dV5WFpBvskoFBQUFBQUFBQUFBSuOyx9VpYCBwAV8L4oij8IgvACcFQUxc+A94CPBEHIRnrS8tvOemVruPhPGMpNz88HlRslf0miYPvf7JYLGneit63Ae1As9eXVnH1sK8ZLRagC/Yjd/QQ+g/tQuveQdeZtV/EbP5Se6xdIs8h/nEThjr+28PZ6fSXeAyXv+SVbMF4qwm/sYCLWzUFQuyOa6rn80ofovzvZLrfv+DgiLTPGln2SRPHOfS3cUa+twmtALOaKai4s3YzpUhFeg/vSc+NSSyGBwm1/puqA4xljWyN6wyMExQ+loc5I1vK3qDnZcgZln0Ex9G2cnTflOOeeft9uecRj9xD93ByO3DzXbnbetujofgYIX3I/Ib+bDOYGcp99h6qvT7gcr/eYYYQ++Ri4uVG5bz/l7+6x96rVhG9KxOPmvpgrqshftZH6vEJQuxO2/nE8B/SFBpGil9+mLu17l72N7rCnHrW6y97Z68C9Gs9bJHfeqo3UXy4CdxXhG1bgeXMsqFRUfZpC2e49rVha4jN2GGFPL0JQuVGx5wClu5t5Ne5EbE7Ec0AfzBXVXF6+EdPlIvzvmYB2/v3Wch79ozn3q8cx/HS228cM0PeluWjjh9JQZ+DHx3egd1C3/QZF80vL7M+lKcfJekqa5fuW3Svwjo0AwN3fm/qqWtLin+hyL0DPedOIfGQaYr2Z0uR0cl78k8sxR7/4CIHxcf+fvTMPj6rI9/d70unuLGTtkKQTAiQBvI7KvqggSyAhouKdAdSRuYiCogRkB5dBcUWRcXBUUEdnnOu94wI6V+c3CkRwdxRkiRGBLAQCZO/sW3en+/z+OJ1Ob+l0J4QT9XyeJw8kXV1vf6rqW3Wqz6kqrC0mClY+32k8D9m2jIAgDbX7DlO0UYrnpPW3ED1zPFitmA115K94AXN5jU/cS564jf7TR2FpMfLDvTtoyD3t0fPlf7oHVZCGyn1HOPmgtKFMv18N5FfPLEYVGkTr2Uq+v+cFLI0tPnuWq54HPrqIiLQxWFuMFK16nuYf3OMi5IoUkv94LwFBGur2H6L4Iel08ajrryZx9c0EDR3Aj9etp/n7Qp/99pStiuxH6o41aJNiMZ6toPDurVjqmnziDnpsEZFpo7G2GClc9QLNuZ65qduWS+1r/2HObJS40ddfReKamwkeOoBjszbQ5KPn8KnS2IwqAMObnsfmwdtWEXxFKpaaBoqWPmO/Jkh5eQMhtmuCcxu9PdnSuVIev4No2xh5spMxst/wFPsJ9tX7jnDKZYxMvGc2KQ8v4N9+jJHDnliIzhZTx+/dQUMn7fpXf1pqb9d5D74ufZ7LBvEfz9xJgFaN2Gbh5H2vUX+k75e3XP1XX9PP4YBjURQ/BD50+dtDDv9vBeZdSKY8x/YGBDDw8SXk/dejHJu2nOgbryFo6ACnJDG3pNNW18gPk+6h/M8fMOCBBQCIRhMlz/ydc4+93i1u0uNLKLztEY5PX0bU7GsIGprklER3czqWukZ+nHw3Fa9+QML90h2utup6Cu94ghMZKziz6jkGbVvlNzvx0bspWriJvPQsImdPRjvEmR19UwaWukZOTl1C5Wvvo79vIQCtJ4vJv2EV+bNWULTgYQY8kQUq36suavooglP0HL5qOQVrXyL1adf1VZJSn76TwrUvc/iq5QSn6IlMG2V/TZOgI3LycFrPVfrlubv1HDR0ANE3TuJY2nLyfvcIA5+QLop95cZuzOL8Xb/n9A13EX7dVDSpA52ShM+diaWukdOZd1D73/+g/9o7AIiYdy0AZ268h3OL7qf/hjvBn537AgKIeyiLc3dupOj6JYR5YEfMzcBa30jRzEXU/O3/6L9GYodlXoOgVnN69lLOzLmXyJtnEZgY6zM3ftNSzi5+iMJr7yb8+iloXNpX5NyZWOobKZyxmOq//oPYdRK3/oNPKZq9nKLZyylZ+wfM58v9mrTI5hnQTR9FSHI831x5LyfWvsIlWxZ7THfJljs5sfZlvrnyXkKS44lOk/YcOXbXNg5OX8/B6eup/Ne3VP7r24vCjZx4GTGZYzkwbS0HpqyheMc/ffYcmTaaoBQ9R65eRuG6HaQ85TmeU566i8J1L3Hk6mUEOcRzyfb3yZm+mpz0tVRnHyJptW9jSsz0kYQm6/nyypX8uPbP/KoTz7/asogf1/6ZL69cSWiynhib58ueXUL+42/y76nrKf/wIIOzbvDZs1z1HJE2Gm1yArmTlnJ6ww4GbV7iMd2gzXdzesMOcictRZucQMS00QC0nCim4M6nafjmR5+9Xii2Pus31H+ZS+6kLOq/zEWf9RufuUHJenImZlG0/iWSN3tuX8lPLaFo/Q5yJmYRlKwnYprUvppPFJO/eIt/nm1jc8GCRzietoyoGz2Mzbek01bbyI/XSGNz4gPS2CwaTZRs/V/OP/667zwXtY+R3121nPy1LzGkkzFyyNN3kr/2Zb6zjZFRLmNklJ9jpG76SIKT4/n3lSs4sfbPXLJlkcd0l2xZzPG1r/DvK1cQnByPztauhzw0n6KtuzgwfQOntrzDkI3zfQPLWN5y9V+Kfj6SZeISOnIoxtOlmIrLEc1tVL//JZEZzjuoRWaMx7DzEwBq/vU1YZOGA2BtMdJ48DhWo//bV4aMHIrxdJmdW/PPL4jIGO+UJiJjAoZd+wGo/fArwiZK3JZjRbSVS+uJWvOKCdCqETS+37AKGTkU05lSTGcldu0/PyfcxXN4xgRq3t0HQN2HX9Hv6hEAiK1GsEg77ghaDf4+Hhg9cxwV73wKQOPhfALDQ1DHRjqlUcdGouoXQsOhPAAq3vkUXeY4++vJjy7k9GNvgB/sntRzZMYEqt//EtHUhulsBcbTpYSOHOoTN2j4JZiLSzGfKwNzG/UffkZo2lVOafqlXUX9+x8D0LDnC0KulAYCbepAmr+R7uxYquuw1jdKd198VNDwYZiLS+zshg8/o9/0K53Z06+i7v8c2FfZNu4TRQJCgkAVgBCkQTSbsTY2+8QNHj4M05kSzGdtnv/1OWHTXTzPuJK69yRu/e4vCblqhFs+4ddPof6fn/nsV07PADGZYynb+bnk6VA+geGhaFzatiY2ElW/YOq/ywegbOfn9L92nFtesbOvovwfvi3d6yk38bYMzjz/PqJJ2vvEXFXvq2WiM8dRuVOqIymeQz3Hc1gIjbZ4rtz5GdGZUl/neJdDFaL1eX+X/pljKbF5rjtUQGB4iEfPgf2CqbN5Ltn5Of2vHQtA6BA9Nf8+DoDhs1zirnPue71JrnqOnDkewy6pf2o6nIcqIhR1bJRTGnVsFKqwYJoOnZS87fqESFtZtxaco7XQda2qb+opO3JmR99q2PkJkZmuu5R6VtTM8VTt+hSAxi647e2ratenRNnyby0477fnUNex+QP3sTkyYwLVtrG55l8dY7O1xUjTweNYjSa/mI7SOYyRDd0cI1MfXUiRn2Nk/8xxPrXrwM7atQiqsGBAupNo9PHOg5zlLVf/pejnI68TF0EQMh3+HyEIwmuCIHwvCMLfBUGI6y5Uo4/GVNqxhbOpzIBGH+2cJt4hjcWKpb6ZwKieneWgiddhKnHglhpQx+mc0qjjozGXOHAbmlC5cCNnXU3LsSL7RYcvUsfpOvIFzJ7Yjmns7HAAgkcOY9jeFxm253nO/367fSLjizR6HcYSg/13Y2k1Wr0zW6vXYSrtSGMqrUZjSxOdMRZTaTXNP57xmSlxu1/Pvry3MwXG6mgr6/jWq628yq2sA+N0tJVWdnAbmgiIDMd44hT90q4CVQCBiXFoLxtKYHx/nz0HxsVgLnVgl1UR6MqO1dHm4Nna0IwqMpyGPV9ibW4l9Yu/k7r/v6n+y3tY6xp948Y75AmYPXHjdJjLOjxbG5vt7atd4ddNpv7/+TdxkcszgFYfTev5Dt/GUgNal3ai1UdjdGjbrSXuaSKvvBRTZR0tRR73Irng3JBUPZET/oMxHz3BqH9sImxkqk9ckGLGWOLM1rjEs0avw+QU8wY08R2fb+B9tzLmu5fp/5vJFD/zlk/cIH00recd/JRWE+TiOUgfTWtpx4YxrSUdaRpPnKN/5hgA4m+YQFCi82f2JrnqWRozOvI0lxpQxzvnqY6Pduk7DWjifffWW2x1TCTmCulC1lxRg1oX4SPXuX2ZSpzbTnsaJ66HNP5I7TI2S17dx2ZTF2Nzd+U6Rpo6GSMd25fRZYw0llbT5OcYqdVHOcVU5+26I6aMJdVo9dJEMm/j3xj60O+YePhFhjz8XxQ+8aZPXDnLW67+qy9K7OM/fVVd3XF50uH/fwBKgRuQdhLodDtkx72f32s67SmF21/cvqTw8HhOz/Yh8Ih1z9TTY0EOSYKGJZFw/wKK79/uJ9tTvr6wpTQtR/PIy8iiYPZqYu+Zh6D14yAtj9m6sj28TxQJCNYwYOUcire87TvPS6a+13MP6t9jnj6UNSJ17+2hrbySgTufJ/b+u2k9+iOipYcHpvniGZHgKy4Bq5XCyfM5NWMh0bf/BvWAeB8hXbcvj4fVOqQJGnEJ1hYjxnz/Bl+PuiiewRffXcU0QOyvJ/r8LfyF4AqBAQRG9uPQtQ9S8OgbXP5n3x897aoeO/l0Tp6Ln/o7h8YuofK9z9Hffq3PbHesb/0IwA8rXyLp9plcufdJVP2CsfrxxY9s9ezDmOFLfXRLcrF70H92n+kpOx+Og7hAV1ee7fg+RiatnMOZCzRG+tKu25MMWJhO3kN/46vRWeQ/9Dcu/ePd3cZerPLuS/2Xop+m/FmcP1YUxfbDKP8oCMJtnSV03Pv5uwH/6dbUTaUGNPoY+++aeB3msmqPacylBlAFoAoPwVLr22K3zmQqNaBJcODqdZgrnLnmUgPqhBjMZTZuWKidq47XkfzK/ZxZtQ3TGd++sbPnW1aF2oGt9sS2pfHEbpex8BzWllaChg2iJbegU1787ZnEzZ8OQOPRQrQJOtpz0uqjMbmUt7HE+VsPjS1N0KB4tANjGbl/q+29Okbu3ULOtfdjrqz16rkn9ezLeztTW3mV012SwLgY2lzKuq2sikB9f9rKq+xlbbWVdeVTHYsNk/7+LOYzvj/20FZehVrvwI6Poa3C4JYmUB9jZweEhWCtbSDs+qk0ffEdtFmwVNfRcvhHgi4fKj2C1RW3TMqzXep4d8/msirU8f1ps7WvgH7OMSXdbfnUZ69yeU68fSYJv5PadsPRQoISY6hDelRGq9dhLHN+XMJYYnD69jQoQYfRoS0JqgBirxvPwfT7vPq8kFxjSbV9nUXDkUKwWlHrwjAbPPdx8QsziZs/A4DGnAK0CTEO8axzj+dSA5qEDrZWr8NU7h4/Vf/4kkvfeICzWz1fdCXdnkHi79IAqD9a6HSXJEgf7ebZ8Q6L5DmaVlua5oISDt8sfRcWkqKnf/oovEmueo697Vr6z08HoOlogVM5qvU6t4XAplLXvtNzWfuiC8k2V9Wijo2S7rbERmE21HXKjVuY6cTVJsTQft9Tk+ADN0GHqaz7C6TNLmOz5NW1/5LSeBsf/ZH+9kzi53e0L22C8/hn9DBGOrYvrcMYGTQwltEOY+SovVs42skYOeD2DHu7bo+pOnuenbXrjpjSJkTby1p/0xT7Qv2KD77h0mc9r4Ny1cUub7n6L0U/T3V1xyVWEITVgiCsAcIF56lyt9fHNOXkE5SsR5MUi6AOJPrGSdRmH3BKU5t9AN28aQBEXXc1DV/5t4OXJzXn5KN14EbdcA11Lty67APo5koDdeSsiTR8Le0opQoPJfX1jZQ8/QZN353oFlszOAH1gDgEdSCRN0ym3oVdn/0tUXOkDi1i1kQabWz1gDj7Ynx1Yn+0KYn2nbc6U9lfd5MzYx05M9ZRvfsAsTdNBaDf6KG0NTRjrnDuUM0VtViaWug3WlrPEXvTVKr3HKT5RDEHL1/EoXFLOTRuKcZSA0cz1nc5aYGe1XNt9gGib5yEoAlEkxRLULKepqP5XTIBWnNPoh6UQGBiHKgDCZ81haZPnHdha/zkG8JvlDrSsJnX0PxNDgBCkBYhWAtAyNWjEC0WTIXFPnEldh7qQQmobeywWVNo3O/C3v8NEf/pzm4rrSTkSmndiRCsJWjEf2A6ddYnbktunr19oQ4k/LrJNOxz4e77lojfSNzwzEk0f+OwW5ogEH7tNdT/63Ofvcrl+fxf93QstP7oAPHzJkuexgzF0tCMyaVtmypqsTS2ED5Gatvx8yZTtfs7++tRk6+gKb/E6XGM3uZWfnSQqEmXAxCcokdQB3Y6aQEoe303OelrpQWpHx2g/7wpgPd4tjZ2xHP/eVOo3n0QgKBkfYf3jLG0FJzvlHv2r3v5Zvp9fDP9Pio++o4Em+eIMUNo68RzW2MrEWOGAJAwbzKVNs+aGNtjiYJAyqpfc/ZvH3fKBfnqueJvH3EsYzXHMlZTs+dbdHOl/il09DAs9c32x6/aZa6owdrYQujoYQDo5k6jds8Bt3x90YVk1+49aO9bdfO8f6by13fzQ/oafkhfQ83uA8TMnQpAPy9cS2Mr/WzcmLlTqemmZ5DGCu1gh7F5tvvYXJt9gGjb2Bx13UQavvJvt0dXlf51N0dmrOPIjHUYHMbIsNFS++psjAxzGCMNtjHy28sXcXDcUg7axsgjXsbIc3/dy4HpGzgwfQOVHx10atedxZSlsdWpXVfaYtlYVkPk1b+SyuSay2k+5dsXqhe7vOXqv/q6rKLQp3/6qgRvC70FQXjY5U/bRVGsFAQhHtgiiuKCrgCe7rgARKSNIWnTHRCgwvD2x5Q+v4uEtb+lKaeAuuyDCFo1yc+tJOTyFCy1DRQu/QOm4nIArvj3K6jCghHUgVjqm8i7dROt+eec8lcFePYVPm0MiQ8vkrZDfnsf5S/sJH71rTTnFlCffQBBq2bQtlWEXJZCW20Dp5dtxVRcTtzyecRlzcVY1PHte+HvNtHm4VssVYDn9SdhU8eQYNsOueadj6l48R3iVs2nJTef+o8ldtKzqwm+LAVLbSPFy7dgOltO5K+nEXvPXMS2NrCKlP/pLer3um+H3NCq7aQWIGXzYiKnjcTaYqRg5XYac6QtE0d8/Aw5M9YB0G9EKkNsWz3W7j/CqQdec8tnzMHt5Mzc4LTVozaw80c+elLP+uVz0d08AywWije9Rv0nh93yDw9r9cgNnTyO/vcvgYAA6t/bS/XLb6Fb/l+0/pBP0yffIGjUxD+9Hu2lqVjrGihdsxnzuTICE+IY8OoTiFYrbRUGyn//R9pK3CeJopegDp08jtgH7oIAFXXvOrLzaPrkWwSNGv2WdWgvTcVS10Dp6qeRxk/HAAAgAElEQVQwnytDCAlC/+RqaUcuQaDuvb3U/OVdt/ytFs/s0CljiXvQth3yrr0YdrxNzIrf0ZqbT+N+iZuwdS1Bv5K22z6/6mlpMT8QMv4KYtfdzul5qzv1FaDqvK/obc/nazp/tnrY5kXo0kZgaTFxfMV2GnKkHdHG7dti3/I2bEQKl/5pKaogDYZ9R8l7oGMb00ufW0rdoXxK/ju7U8aF5gpqFZduW0q/ywchmtooeOQNar485pR/UEDnjygmP7mYqGnSFqoFq16kqT2es7eSk74WgNARqQy1bSdas/8IRQ++CsAlr64jODUB0SpiPFfJqQ0vO33j2Wjt/DHU/9h8OzFpI7G0GDm24iXqbZ6v3PcU30yX7mSEj0jh8j/dQ0CQhqp9RznxgLQl8cA7ryXp9gwAKj48QP7jzs/jB+J93V5v1nNYYOebvQx84i4ipkrbMBetft6+pfFle5/lWIYULyHDUzu2JP7ksH2b/sjMCQx6fDGB0RFY6ptoPlZE3vxHvfq8UGxVVBhDXlqLJjEG0/kqCpY8g6W2Y/2YtwuTwU/eaeeeWvWCfUvjy7P/wA/pawAIHZ5KSvt2yJ8c5oytfUVlTmDw44sJ1IVjqW+i6VgRJ299zJ63WuW5XYdPG8OATR1jc9nzO9GvuZXm7wuos43Ng7etItg2VhRlbbWPFZd9/QqqsBD7NUHB/E205rt/AdJk7rxtp25eTJRtjMxzGCNHffwMRxzGyPbtkGv2H6HQwxg57uB2jriMkSZR1Sn3ks13EJ02AmuLiR9X7LC36/H7nubA9A2A1K6l7ZDVtnYtxVTE+EsY9vhChEAVVqOJkxteo+H7jq2FozSex0fo/fJubev8gZ7e7L8Ari59t+9edTvoa/2cvryUpM+Wo9eJC4AgCP8BJALfiqLY6PD3TFEUd3cF6Gzi0tvqbOJycdi+L5y/kPI2celNeZu49LY6m7j0trxNXHpbnU1celveJi69LW8Tl5+rvE1celPeJi69qa4mLr0pbxOXn6vk+ka1s4nLxZC3iUtvytvEpTflbeLS2/I2celt9dULblcpE5fuqatdxZYD7wPLgR8EQbjR4eUnPb9LkSJFihQpUqRIkSJFnUkUhT7901fV1ZT4LmCMKIqNgiAMBnYJgjBYFMXn6GTjB0WKFClSpEiRIkWKFCm60Opq4qJqfzxMFMXTgiBMRZq8DEKZuChSpEiRIkWKFClSpOgiqauJS5kgCCNFUTwKYLvzcj3wF+AKXwBGGZ9zlEsamZ7hNYrd3uitZ5Kxjpua5FnX80vUhTiaorsKVcm3jkouaQLl6UdEkzzfSQXKtDYQQKuWp3315Z17ektyPoISpu3eae89VZNRIws3UCVfTAVY+vTyjT4h+Wrnp62urnQXAE7764mi2GbbTWxyr30qRYoUKVKkSJEiRYoUKXKQ16/KRVE85+U1f46ZVqRIkSJFihQpUqRIkaJuS6ZnixQpUqRIkSJFihQpUqTId8m6ACX58TuImj4Ka4uJ/BUv0JRb5JYmdHgKQ9sPfNp3hKLfSweKDVx/C9GZ4xCtVsxV9RSseAFTeY3b+3uD3a6Ee2aT/PACvv3V7U6HTXnTwEcXEZE2RjpQbNXzNP9wyi1NyBUpHQeK7T9E8UPSIVdR119N4uqbCRo6gB+vW28/jMwXDX3idnTTpQPFfrx3O40e/IYNT+bSP0l+DfuOkP+gdMjVZa+sJCQ1AYDA8BDa6pvtB775IjnKOmzKKAZsulM6XOutbMq3Ox9qKGgCGfTHVYRckUpbTQOns57BdK6CsGtGkHDfAgR1IKK5jfNPvE7j17k+e5WTLRc3fKrERRWA4U3P3MHbVhF8RSqWmgaKlkpcVWQYKS9vIGTEEAw793Nu4ys+Mx016LFFRKaNxtpipHDVCzTneo6p1PbD8vYf5sxGKaair7+KxDU3Ezx0AMdmbbAftNeXuXK2bbn6TZCnvPtNGU2i7dDg6rezqdyxy+l1QRMoHRp8uXSo65llWzCfqyB4xFAGbF5mSyRQvu3v1O9xPzTYm8KmjCbx4cUIKhWGt/ZSscO9ngc+u4qQK4bQVlPPmWVSPYeMGErS5iw7u2zbm9T5wZaPK2efLY9nkC+mutu226VO6M+w7Bcp3/YmVX/+h1+eBz22iKi00Vi8xHKoQyzXOMTywI0LiEofi9XUhvFMOYWrnsdS3+wXv69IVPa46pZku+MSNX0UwSl6Dl+1nIK1L5H69F0e06U+fSeFa1/m8FXLCU7RE5k2CoDz29/naNoacmasoyb7EEmr5100NoAmQUfk5OG0nqv0mRuRNhptcgK5k5ZyesMOBm1e4jHdoM13c3rDDnInLUWbnEDEtNEAtJwopuDOp2n45kefmQC66aMISY7nmyvv5cTaV7hky2KP6S7Zcicn1r7MN1feS0hyPNFpIwE4dtc2Dk5fz8Hp66n817dU/utbn9mylHVAAEmPL6Hwtkc4Pn0ZUbOvIWhoklMS3c3pWOoa+XHy3VS8+gEJ998GQFt1PYV3PMGJjBWcWfUcg7at8tmrrGyZuQULHuF42jKibvTAvSWdttpGfrxG4iY+IHFFo4mSrf/L+cdf953nooi00QQl68mZmEXR+pdI3uy5fSU/tYSi9TvImZhFULKeiGlS+2o+UUz+4i1+x5RcXDnbtlz9JshU3gEBJD56N0ULN5GXnkXk7MlohziXdfRNGVjqGjk5dQmVr72P/r6FALSeLCb/hlXkz1pB0YKHGfBEFqj8GG4DAhjw2BJO3fYIJ2ZkETV7MlqXeo621fPxKUuofO0D9PdJ9dxy8gwnb1jNyVkrKbxtEwOeXOo7W0aunH22LJ6RMaZ60LbblbBxMQ2fHvKdaVNk2miCk/UctcVyipdYPrV+B0cnZhGcrCfSFst1n+eQM20luTNW03qqhMTlc/z+DIp+2pJt4hI9cxwV73wKQOPhfALDQ1DHRjqlUcdGouoXQsOhPAAq3vkUXeY4ACyNLfZ0ASFaRHzfwaKnbIDkRxdy+rE3/NpqKXLmeAy7PgGg6XAeqohQ1LFRLtwoVGHBNB06CYBh1ydEZo4HoLXgHK2FJT7z2hWTOZaynZ8DUH8on8DwUDQufjWxkaj6BVP/XT4AZTs/p/+149zyip19FeX/8H15kxxlHTJyKMbTZZiKyxHNbdT88wsiMsY7pYnImIBh134Aaj/8irCJwwFoOVZEW3k1AK15xQRo1Qga329MysWWixvqyv3AnRuZMYFqG7fmXx1ca4uRpoPHsRq7v9NP1MzxVO36FIDGLmKq0da+qnZ9SlTmBMlvwfluxZRcXDnbtlz9JshT3iEjh2I6U4rprFTWtf/8nPCMCU5pwjMmUPPuPgDqPvyKflePAEBsNYJF2jNI0GoQ/fQr1XMHu+afXxCR7syOSJ9A9buO9dzONtnZAVoNfgyNMnPl7LMvvmeQL6Z60ral167EVFyGMb/YL78gxXKln7Fc6RDLdZ/l2Mu84VAeGr3O78+g6KctvycugiBckFai0eswlhjsvxtLq9G6NECtXoeptCONqbTaqZEOvO+3jD30Ev3nXEPxlrcvGjs6Yyym0mqafzzjMxNAE6/D5MA1lxpQx0c7pVHHR7twDWjie1bkWn00reer7L8bSw1o9dFuaYwO3NYS9zSRV16KqbKOliKnjea8So6ylsq5w6+p1IA6zpmpjo/G3J7GYsXS0IQqKswpTeSsq2k5VoRo8n2rVLnYcnHVLlypTbtzTV1wuytNfDRGR98lBjQuMaVxjSkPaX46XBnbtkz9JshT3uo4XUc5YmvbrmXtmMZe1uEABI8cxrC9LzJsz/Oc//12+8WWT+x4HeZSR3aVh7jyxJbqOWTkMC7JfoFL9vyJcw/6zpaLK2e7lsszyBdTPWnbQrCW/nfPofy5N/1itkvjOB7Qs1iO/W0atfsPd+tz9AVZxb7901fldeIiCMJTgiDE2P4/VhCEU8C3giCcEQRhipf33SUIwneCIHz3frP7s4tSIvc/uX0r5enxP4c0xU+9yXdj7qby3S/Q35HpzcoFYwcEaxiwco5fE6Wu8nRKInj8cP6zugK7+fWUxvnX2F9P9OtuS+foXi5rH8q5K79Bw5JIuH8Bxfdv/2mw+zLXYyPwHeGd7563e/vy+CF/olxPWcrXti9KvwnylLcvfbGXNC1H88jLyKJg9mpi75mHoFX7A+8mW/qn+WgeJ9OXkTd7DbFL5/rBlokrZ7uWraw7Q1+Ma5Hut+34VfOpeu19rM2t/nN7wnaJ5YR75yC2Wal67/PufQ5FP1l1dT/1OlEU77P9/xngZlEUDwqCMAz4OzDW05tEUXwFeAXgq/i59tYWf3smcfOnA9B4tBBtgo72ZWRafTSmsmqnfIwlBqc7LBoPaQCq/vEFl/7PA5x95p1OjVwodtCgeLQDYxm5f6vtvTpG7t1CzrX3Y66sdePG3nYt/eenA9B0tABNQkeear0Os8uGAqZSV64OU7m7566UePtMEn4n+W04WkhQYgx1nLR/ZmOZM9dYYnD6picoQYfRoUwEVQCx143nYPp9dCW5yrpdplIDmoQYh/x0mCucmeZSA+qEGMxlBlAFoAoLxVIrfUp1vI7kV+7nzKptmM74fndJTrZcXLMLV2rTLtwyKY0nbncUtzDTKaa0CTE02l7TJPgQUwk6TGW+b+QhN9ctz4tYz3LGstzlbS6rQu3atl3L2pbGW9s2Fp7D2tJK0LBBtOQW+M7WO7Jj3OOq1Ad2gX9subhy9tkX27Pc46PdczfbdsjIYUTMuhr9/QtRhYciWkVEownDf/+rU17cwkxibbHceLTAua4TdG4bK3UVyzHzphI1YyzHb37Yq09FP0919aiYWhCE9slNsCiKBwFEUcwD/D6yvOyvu8mZsY6cGeuo3n2A2JumAtBv9FDaGpoxVzgHm7miFktTC/1GDwUg9qapVO85CEBQcrw9XfTMcbQUnL8o7OYTxRy8fBGHxi3l0LilGEsNHM1Y32lHUfG3jziWsZpjGaup2fMturnTAAgdPQxLfTPmCueANVfUYG1sIXT0MAB0c6dRu+eAV2+edP6vezoW1H90gPh50nmh4WOGYmloxuTi11RRi6WxhfAxkt/4eZOp2v2d/fWoyVfQlF+CsbTrSZRcZd2u5px8tMl6NEmxCOpAom64hrps5zKsyz6Abm4aAJGzJtLw9fcAqMJDSX19IyVPv0HTdye69NpX2HJxm3Ly0Q524M5259ZmHyDaxo26biINX33vF8NV5a/v5of0NfyQvoaa3QeImTsVgH5eYsrS2Eo/W0zFzJ1KTTdiSi6uoy52PcsZy3KXd3NOPprBCagHxCGoA4m8YTL1LmVdn/0tUXOki9CIWRNptJW1ekCcfZG2OrE/2pRETA47MvnC1iYnoEmKs9dzfbbzpij1Hx8geo57PWuSnNlBKYmYzpX/BLhy9tkXz7Pc42O75+627cKb7uPEpMWcmLSYqr98QMWLO71OWkCK5dz0NeTaYrm/n7Hc3yGWI6aOIiHr15xcuBlrS/fXR/YFWRH69E9fleBt0aAgCMuBG4CngMlAJPAeMB1IEUXxv7oCON5xcVXK5sVEThuJtcVIwcrtNOZIW1SO+PgZcmasA6DfiFSG2LYBrN1/hFMPSFviXfLqWoKHJIBVxHiuksL1r3i8G9MbbEeNObidnJkbnLYg1KgsnXIHPnEXEVOlbYmLVj9v39L4sr3PcixjNQAhw1M7tkP+5DDFv/8zAJGZExj0+GICoyOw1DfRfKyIvPmP2vNuaOv8FvWwzYvQpY3A0mLi+IrtNORIj/CN27fFvrVx2IgULv3TUlRBGgz7jpL3QMeWi5c+t5S6Q/mU/He2W95awfszvb1Z1iEas0dm+LQxJD68SNpa8+19lL+wk/jVt9KcW0B99gEErZpB21YRclkKbbUNnF62FVNxOXHL5xGXNRdjUcdC3sLfbaLNUOfVY19g9za3s64ifNoYBmzq4JY9vxP9mltp/r6AOht38LZVBF+egqW2gaIsiQtw2devoAoLQVAHYqlvomD+Jlrzz7oxzBZVp74HP3mnPaZOrXrBvtXt5dl/4If0NQCEDk8lpX2b3E8Oc+bBVwGIypzA4McXE6gLx1LfRNOxIk7e+phP5d3bXE2g536kt+u52dR5P9KbsRwY4L0f6c3yDu6kHwmbOoYE25axNe98TMWL7xC3aj4tufnUfyyVddKzqwm+LAVLbSPFy7dgOltO5K+nEXvPXMS2NrCKlP/pLer3um+TaxU7vzgImzaGxIcWI6gCqH7n4456/r7Azh70R4ndVtsgbdF7tpyoX08ldulcMLchiiLlz71F3V7fd4Lsba7YieeL0W8KgudOrLc9Nxk1nZZ3b8ZUWJCxU25327aj4lb+FktTq8ftkL31I4OfvJNIWywXOsTyFdl/INchllMdYvm0LZZHfvUiglZNW43ks/FQHkX3veyU/5Ul7/Xdq24H7Y+7qQ+vJIG08nf6ZDl6nbgACIIwFbgHGIb0aNlZ4P+Av4qi6Lm3d5C3icvPVd4mLr0pbxOX3lRXE5feVGcTF0UXXj1eatUDeZu4/FzV2cSlt+XtgqM31dXEpTfV2cSlt+Vt4vJzVWcTl4uhziYuvS1vE5felLeJS29Lrn4ElInLhVJfnbj4smdgGdJ6lW9FUWx/xBhBEDKB3b31wRQpUqRIkSJFihQp+jlKOYCye+pqV7F7gfeB5cAPgiDc6PDyk735wRQpUqRIkSJFihQpUqSoXV3dcbkTGCOKYqMgCIOBXYIgDBZF8Tk8b9KnSJEiRYoUKVKkSJEiRRdcXU1cVO2Ph4mieNq23mWXIAiD8HHiItcaBDmfx5dLsaHNsnDbLH6fY3rBZG6TZ+2DRc7ntGXiyhlSKpmeTZeznlvMvp/+fSEVqpVnpx6jTH4BGmVagyDneg+VTGuKzFb5xosgmdafyhVTDa1+b/56wSSX55+S5FvV99NWVz1ImSAII9t/sU1irgdigCt684MpUqRIkSJFihQpUqRIUbu6mrgsQFqcb5coim2iKC5A2h5ZkSJFihQpUqRIkSJFinpdXu/Ni6J4zstrX134j6NIkSJFihQpUqRIkSJF7pLtoeKwKaMYsOlO6bCpt7Ip3/6u0+uCJpBBf1xFyBWptNU0cDrrGUznKgi7ZgQJ9y1AUAcimts4/8TrNH6d6zM3fKrERRWA4U3P3MHbVhF8RSqWmgaKlnZwE+9bgKAJRDS1cc5Pbk/YqsgwUl7eQMiIIRh27ufcxlf84vabPBr9Q3dBQAA17+yl6qVdbtwBW1cTdPkQLLUNnF3+NObzFagTYxmavQPjqfMAtBw9ScnvX/SLHTZlNIkPL0ZQqTC8tZeKHe6eBz67ipArhtBWUy8d7HWugpARQ0nanGVLJFC27U3q9rgf4OZNAx9dRETaGOmwz1XP0/zDKbc0IVekdBz2uf8QxQ9JB3upIvuRumMN2qRYjGcrKLx7K5a6Jp/Zgx5bRFTaaCy2A7aac93ZoVek2A/Yqtl/mDMbJfbAjQuISh+L1dSG8Uw5hauex1Lv2/qlQY8tIjJttP1gL0/cEAdurQM3ycYVTW20ninnlB9cOT3LVc9y+QVIfuwOIqePxtpiomDl8zTlFrmzh6cwZNsyqZ73HaZoo3Sg7KCNC4jKaK/nMgpWvuATW85Yhp617ejrryJxzc0EDx3AsVkb7Afe+aKelHXS+luInjkerFbMhjryV7yAubzG7f2dsh+/g6jpo7C2mMhf8UKn7KG2wwlr9h2h6Pd/cXo94Z7ZJD+8gG9/dbvT4YTeJGfbTnn8DqJtnk924rnf8BSG2TxX7zvCKRfPiffMJuXhBfzbD89y9CNyx1RP2tfA9bcQnTkO0WrFXFVPwYoXMPnQtuX23JekbIfcPcmzSi4ggKTHl1B42yMcn76MqNnXEDQ0ySmJ7uZ0LHWN/Dj5bipe/YCE+28DoK26nsI7nuBExgrOrHqOQdtW+c0tWPAIx9OWEXWjB+4t6bTVNvLjNRI38QFn7vH0FZxe/RyDn/OD20O2aDRRsvV/Of/46/4xbdyER+7h9O0PUzBzKRE3TEE7xJkbdVMGlvom8tPuwvCX94nfsND+mulMGYXX30vh9ff6PWkhIIABjy3h1G2PcGJGFlGzJ6N18Rxtq+fjU5ZQ+doH6O+TPLecPMPJG1ZzctZKCm/bxIAnl4LK9+YakTYabXICuZOWcnrDDgZtXuIx3aDNd3N6ww5yJy1Fm5xAxLTRAOizfkP9l7nkTsqi/stc9Fm/8ZkdmTaa4GQ9RydmUbT+JVI23+UxXfJTSzi1fgdHJ2YRnKwnctooAOo+zyFn2kpyZ6ym9VQJicvn+Ow5KFlPjo2b7IVbtH4HOROzCErWE2Hj1n+ew/cO3AQfuXJ7lqOe5fLbzg5K0XPk6mUUrttBylOe2SlP3UXhupc4cvUyglL0RKZJ7NrPczg6dSU501fTWljCgOU+eJYxlqHnbbv5RDH5i7fQ8M2PfnF7WtYl298nZ/pqctLXUp19iKTV83xmR00fRXCKnsNXLadg7UukPu2Znfr0nRSufZnDVy0n2IENoEnQETl5OK3nKv3yLFfbbvf83VXLyV/7EkM68Tzk6TvJX/sy39k8R7l4jvLTsyz9iMwx1dP2dX77+xxNW0POjHXU+Nq2Zfas6OchWWo9ZORQjKfLMBWXI5rbqPnnF0RkjHdKE5ExAcOu/QDUfvgVYROHA9ByrIi28moAWvOKCdCqETS+3TgKdeV+4M6NzJhAtY1b8y9nrrmde9I/bk/Z1hYjTQePYzX6v0tH8IhhGM+UYj4rcev+3+eEpV/plCZsxpXUvLsPgLqPviT06hF+czxJqudSTGcd6jl9glOaiPQJVL/rWM8SW2w1gUXacyNAq/F7S6vImeMx7PoEgKbDeagiQlHHRjmlUcdGoQoLpunQSQAMuz4hMnN8x/t3Su837PyEyEznz+1NUTPHU7nrUwAau2A3HsoDoHLXp0TZGHWf5di9NxzKQ6PX+cyt8pNb1Qm30Q+unJ7lqme5/AJEZ46jcudnNnY+geGhqGMjXdiRqMJCOtg7PyPa5tmJfTgPTULXbDljGXretlsLztNaWOI3t6dlbWlssadThWj98h49cxwV73zqwA7xzO4XQoONXfHOp+gyx9lfT350Iacfe8OvbTblbNs6B88N3fSc+uhCivz0LEc/IndM9bR9ObbtgBAtog8fQm7Pin4ekmXioonXYSqpsv9uKjWgjnPu3NTx0Zjb01isWBqaUEWFOaWJnHU1LceKEE1tPnHVLlxzqQF1vDvX5AO3+QffuReS7a/U8TrMpR3fPLWVVrmXdZxDGosVa0MzqqhwADRJcaT+8zmS39xMyLjLusF29FzlwbOu03oOGTmMS7Jf4JI9f+Lcg9vtnZYvktqYwYFtQB0f7cKOxlTakcZUakBj+3zqmEjMFdJtb3NFDWpdhB/saOf2XWJA48LWuLI9pAGI/W0atfsP+8w1XiBufz+49nxl8SxPPcvltz1fx3o2lhrcLg41eudyMZZ2wr5lOjX7j3TJlDOW4cK27Z5wu1PWA++7lTHfvUz/30ym+Jm3fGfrdRid8q1G68LW6nUubbva/vmiM8ZiKq2m+cczPjNB5rbt4tnUiWdjqXO5OHo2llbT5Lfni9+PyB5TPWxfAAPv+y1jD71E/znXULzl7S6Zcnvua7L28Z++Kq8TF0EQDguC8HtBEFIvKNXTY32u344IHhI5JAkalkTC/Qsovn/7heV6SuTCTXzAT+4FYl8wuZW15zRtldWcnHQ7hTesoPSJV0n641oC+gX7AfLkx/d6bj6ax8n0ZeTNXkPs0rkIWnWP0K5swSP7AhS4L/l6SuNS2Qn3zkFss1L13ufd5oo94Bp85XaW70Xx7CnLi1DPcvnFNz8en5x2+XiJK+YgWixUvesLW8ZY7iTv7rRtf3Uhyrr4qb9zaOwSKt/7HP3t1/oB94T2re8OCNYwYOUcny4m3blytm1P2fruOWnlHM50y3PX3Avfj8gdU57wvpV1u4qfepPvxtxN5btfoL8js7tQlyS96FnRz0JdPesUBUQCnwiCUAa8CbwtiqLXe+6CINwF3AXwYNRw5vQb7PS6qdSAJiHG/rtGr8NcUe2UxlxqQJ0Qg7nMAKoAVGGhWGqlRXbqeB3Jr9zPmVXbMJ1x2q3Zq8wuXLVeZ3/8y56mTErTGTflz/dzeqV/3AvB7q7MZQbU+v723wP1Me5lbUvTZuMGhIXYuRaT9G/rD4WYisvQJCfSmlvgI7sKtd7Rc4y759KqTuu5XcaCc1hbWgkaNogWL+zY266l//x0AJqOFjg9AiOVt/PCQZPLt6cavQ6T7fOZq2pRx0ZJ357FRmE21Hn1Grcwk1gbu/FogXP7TtC5LVp0YyfoMJV1pImZN5WoGWM5fvPDXXIdPWsTYmh0yLNLzx64kTPGcqILrpye5apnufwCxC/MJG7+DImdI9Vze5Ro9TpMZc5xZSw1OJWL1sEzQP95U4meMYZjN23qkg0XP5bhwrdtX3Why7pdVf/4kkvfeICzWzu/sI6/PZO4+dMl9tFCtAk6B3a0O7vEtW1LaYIGxaMdGMvI/Vvtn2nk3i3kXHs/5spaN66cbVt/eybxNs8NNs+OfowePDveGdA6eA4aGMtoB8+j9m7haCee5RwvQJ6YulDty1VV//iCS//nAc4+806f86zo56euHhWrEUVxrSiKA4E1wFDgsCAIn9gmJx4liuIroiiOFUVxrOukBaA5Jx9tsh5NUiyCOpCoG66hLvuAU5q67APo5qYBEDlrIg1ffw+AKjyU1Nc3UvL0GzR9d8IPq9CUk492sAN3tju3NvsA0TZu1HUTafjKgfu3jZQ85T+3p+yeqOX7PLSDE1APiENQBxJx/WQaPv7WKU3Dvm+JmiN1ZhHXTt6eLFIAACAASURBVKLp3zbP0eEQIDURdVIcmsEJmIt9n7BJ9ZyAJinOXs/12c7s+o8PED3HvZ41SXH2hXfqxP4EpSRiOlfulVfxt484lrGaYxmrqdnzLbq50wAIHT0MS32z/VZ+u8wVNVgbWwgdPQwA3dxp1O6R6qR270F086T36+Z1/L0zlb++m9z0NeSmr6Fm9wH6z50KQD8vbEtjK/1s7P5zp1JjY0RMHUVC1q85uXAz1hbv65rKX9/ND+lr+MHGjfGTG+OBm+cDV07PctWzXH4Byl7fTU76WmmR90cH6D9vio09lLaGZswVzhdm5oparI0t9Bs9VGLPm0L17oMARE4bSeKy/+T4wqd8YsPFj2W4sG3bH13Isg5K1tvTRWWMpaXgvHf2X3eTM2MdOTPWUb37ALE3tXvunG1p6mDH3jSV6j0HaT5RzMHLF3Fo3FIOjVuKsdTA0Yz1Hi/gQd62XfrX3RyZsY4jM9ZhcPAcNnooFi+ewxw8G2yev718EQfHLeWgzfMRL57lHC9Anpi6UO0LICg53p4ueua4Ltu2XJ77suR+FOyn+qiY4HZr0PFFQTgsiuJol7+pgHTgZlEUb+8KcGTgjR4B4dPGkPjwImk75Lf3Uf7CTuJX30pzbgH12QcQtGoGbVtFyGUptNU2cHrZVkzF5cQtn0dc1lyMRR03fQp/t4k2l284OrMVPm0MAzZ1cMue34l+za00f19AnY07eNsqgi9PwVLbQFGWxI2/151bMN+d603dZQNc9vUrqMJCENSBWOqbKJi/idb8s075a9QWj9x+U8ei33gnQkAANTuzqdz+DrEr59OSm0/DvgMIGjUDnl1D0K9SsNQ1cvbepzGfLSc882piV85HtFjBYqFi299p2O/eIbdZOp//hk0bQ+JDixFUAVS/83FHPX9fQP3Htnr+42qCbfV8ZtkzmM6WE/XrqcQunQvmNkRRpPy5t6jb+61b/uY2VafsgU/cRcTUUdL2lqufp9m2Beple5/lWMZqAEKGp3Zsb/nJYYp//2cAVFFhDHlpLZrEGEznqyhY8gyW2kZ73hbR+zaGg5+8k0gbu3DVC/btV6/I/gO56WsACB2e2rF16yeHOf3gqwCM/OpFBK2athrpW6bGQ3kU3feyPW9v5MFP3mn3fMqBe3n2H/jBgZviwD1j447wwD3twO3qoYje9KwSOqfLVc+96RfA6oWd/ORioqaNwtJipGDVizTlSOwR2VvJSV8rsUekMtS2RW/N/iMU2dijvn6BAE0Hu+FwHqc2dGyxHqIxe2T2diwbzd4fAOhJ247KnMDgxxcTqAvHUt9E07EiTt76mD1vb/Xck7K+5NV1BKcmIFpFjOcqObXhZadvrMUu+pGUzYuJnDYSa4uRgpXbaWxnf/wMOTPWAdBvRCpDbNvV1u4/wqkHXnPLZ8zB7eTM3OC0NbAqoPNLk95s22ar9+9LUzcvJsrmOc/B86iPn+GIg+f27ZBr9h+h0IPncQe3c8TFc5DK8xgJvduPqAM9c3s7ppqMGq9l3ZP2dcmrawkekgC2tl24/hWnth2q9Txp7W3PACPPfPCT2Gf4w7hb+vQWA7PK3+qT5djVxOUtURRv6Qmgs4lLb+tCLFX4qamziUtvy9vEpbflbeLSm+pq4tKbkossZ0h5m7j0puSsZ28Tl95UZxOX3lZXE5felFz13NXEpTflbeLSm+pq4tKb8jZx6U11NnHpbXU1celNdTZxuRhSJi4XRn114uK1BxFF8RZBEP5DEITpgiD0c3xNEARfVmIpUqRIkSJFihQpUqTIQSJCn/7pq+pqV7HlwPvAcuAHQRBudHj5yd78YIoUKVKkSJEiRYoUKVLUrq7uzd8FjBFFsVEQhMHALkEQBoui+BzyPbGiSJEiRYoUKVKkSJGiX5i6mrioRFFsBBBF8bQgCFORJi+D8HHiItdaE7meDwdQBchjuqlVnudZ1TI9NwzyPTuslWnNBcjXtgNk9NzQqpWNLZfkWoMg13PxgoztS641VCoZ+0651tcEBvp+cPOFVptM62tMMsWUXOvVQN41az8VWZWv/7ulrqK4TBCEke2/2CYx1wMxwBW9+cEUKVKkSJEiRYoUKVKkqF1dTVwWAE4Hd4ii2CaK4gJgcq99KkWKFClSpEiRIkWKFClykNd7eaIonvPy2lcX/uMoUqRIkSJFihQpUvTzllVZKt4tyfYQYvjUUQzYdCeoAjC8mU359nedXhc0gdJhjFekYqlpoGjpM5jOVaCKDCPl5Q2EjBiCYed+zm18pRNC59yBjywGVQBVb2ZT9uJ7btzkbSsJGZ5KW00Dp+7ZauemvrKeUBu3/eApfxQ2RfIsqAIwvOXZ86A/riLkCol9OkvyHHbNCBLuW4CgDkQ0t3H+iddp/DrXL/agxxYRmTbafqBYc+4ptzQhV6R0HCi2/zBnNkoHTUVffxWJa24meOgAjs3aYD+MzBfJVc9hU0aT+PBiBJUKw1t7qdjhzh347CpCrhhCW029dMjVuQpCRgwlaXOWLZFA2bY3qdvzjV/sflNGk/iQ5Ln67Wwqd+xyYyc9u5rgy1Ox1DZwZtkWzOcqCB4xlAGbl9nZ5dv+Tr0f7F+iZ4Dkx+8gavoorC0m8le8QFNukVua0OEpDG0/sG7fEYp+/xcABq6/hejMcYhWK+aqegpWvICpvMbt/Rea266Ee2aT/PACvv3V7U4H5XWlQY8tIiptNBYv8RzqEM81DvE8cOMCotLHYjW1YTxTTuGq57HUN/e6556UNUDyY3cQOX001hYTBSuf75Q9xHYQZO2+wxRtlNhJ628heuZ4sFoxG+rIX/ECZh/Zcvad3RmrAOKz5hDz2xlgsVL80J+p/+yoz1yAgY8uIiJtjHQY46rnaf7Bs2f7YYz7D1H8kORZFdmP1B1r0CbFYjxbQeHdW7HUNfWa3wsxNsNPM6ba1Z1+5Jd4PaLo5yF5VqoFBJD0+BIKFjzC8bRlRN14DUFDk5yS6G5Jp622kR+vuZuKVz8g8YHbABCNJkq2/i/nH3+9W9yBjy8h778e5di05UTfeA1BQwc4JYm5JZ22ukZ+mHQP5X/+gAEPLOjgPvN3zj3WDa6NnfT4Egpve4Tj05cRNduD55vTsdQ18uNkyXPC/ZLntup6Cu94ghMZKziz6jkGbVvlFzoibTRByXpyJmZRtP4lkjff5TFd8lNLKFq/g5yJWQQl64mYNgqA5hPF5C/eQsM3P3bLsxz1POCxJZy67RFOzMgiavZktC7caFtZH5+yhMrXPkB/n8RtOXmGkzes5uSslRTetokBTy4FlR9hEhBA4qN3U7RwE3npWUTOnox2iAv7pgwsdY2cnLqEytfeR3/fQgBaTxaTf8Mq8metoGjBwwx4Ist39i/RMxA1fRTBKXoOX7WcgrUvkfq057ad+vSdFK59mcNXLSc4RU9kmtS2z29/n6Npa8iZsY6a7EMkrZ53UbgAmgQdkZOH03qu0me/AJFpowlO1nPUFs8pXuL51PodHJ2YRXCynkhbPNd9nkPOtJXkzlhN66kSEpfPuSieu1vW7Z6DUvQcuXoZhet2kPKUZ3bKU3dRuO4ljly9jCAHdsn298mZvpqc9LVU+8GWs+/s7lgVNHQA0TdO4ljacvJ+9wgDn7gbAnyPqYi00WiTE8idtJTTG3YwaPMSj+kGbb6b0xt2kDtpKdrkBCKmjQZAn/Ub6r/MJXdSFvVf5qLP+k2v+u3x2MxPN6agm/3IL/F6RNHPRrJMXEJHDsV4ugxTcTmiuY2aD74gImO8U5rIjAlU79oPQM2/viJs4nAArC1Gmg4ex2r0/1RWiVtq51a//yWRGRNcuOMx7PzExv2asEkd3MaDx7Eau7dLR4ir53+6e47ImIDB5rn2ww7PLceKaCuvBqA1r5gArRpB4/vNsqiZ46na9SkAjYfzUEWEoo6Nckqjjo1CFRZM46E8AKp2fUpUplQ2rQXnaS0s8duzXPUc0l7PZx3KOt25niPSJ1D9rmNZjwBAbDWBRdrNKUCr8fu4+JCRQzGd6WDX/vNzwl3aWHjGBGre3QdA3Ydf0e/qdrbRzha0GkQ/tuT7JXoGiJ45jop3PgWg8XA+geEhqGMjndKoYyNR9Quhwda2K975FF3mOAAsjS32dAEhWkQfzfeUC5D86EJOP/aG31svRs0cT6Wf8VzpEM91n+XYy7zhUB4ave6ieO5uWQNEZ46jcudnDuxQz+ywkA7POz8jOnO8G1sVovW5jcvbd3ZvrIrMmED1+18imtowna3AeLqU0JFDfWZHzhyPYZeUb1MXnpsOnQTAsOsTIm1lHTmz43MZdn5CZKbz577Qfns6NsNPN6age/3IL/F6RNHPR7JMXNTxOkwlVfbfzaUG1PE6lzTRHWksViwNTaiiwnrE1eijMZV2cE1lBjT6aOc08Q5pLFYs9c0E9pAr5evs2VRqQB3n7tnchefIWVfTcqwI0eT7lpKa+GiMjuwSA5p4T74NXtP4K7nqWR2vw1zqyK3ywNV1WtYhI4dxSfYLXLLnT5x7cLt9QPKJHeeQLzbPrvUc54kdDkDwyGEM2/siw/Y8z/nf+87+JXoG0Oh1GEs62q2xtBqty0WDVq9zbtul1U4XFgPv+y1jD71E/znXULzl7YvCjc4Yi6m0muYfz/jEc2I7xgw9i+fY36ZRu/+wb1yZyrrdj2MfZiw1uF0cavQ6TE6fz9nzwPtuZcx3L9P/N5MpfuatbnEvVt/Zk7HKl/d6Zcc7l6PUbzu/X+3qudSAxtbfqGMiMVdIj+GZK2pQ6yK6Zso4Ntvz/gnGVHf7kV/i9UhflNjHf/qqvE5cBEEYKwjCJ4Ig/I8gCEmCIGQLglAnCMJBQRBGeXuvV3laj+T2bYGHRD0uSfc83bE+pLkwaPeMPbAdPQcNSyLh/gUU37/dT7YnTz6we1rgfaie/Snr5qN5nExfRt7sNcQunYugVfuB7i5bStNyNI+8jCwKZq8m9p55frB/iZ7pxLYr28P7HNIUP/Um3425m8p3v0B/R2avcwOCNQxYOcevC3fnfLtZ3i6BlXDvHMQ2K1Xvfe4j1xP2IpQ1IPjg2eMyV4ckxU/9nUNjl1D53ufob7/WV7AH7EXoO3s0VvVwDPOh3/alPvyTjGNzJ3n39ZjqUT/yS7weUfSzUVf397YDDwORwNfAKlEU0wVBmG577SpPbxIE4S7gLoAHI4fzm36DnV43lxrQJMTYf1frdZhttx7tacqkNOYyA6gCUIWFYqn1ffGqJ5lKDWj0HVxNvA5zWbXHNOZSGzc8pMdce74OnjV6HeYKF8+lBtSdeFbH60h+5X7OrNqG6YzTDtUeFbcwk/7z0wFoOlqANiGGxnZ2gs5tYarJ5RtMTYIOU5nvC2c9Sa56NpdVodY7cmPcuaVVnZZ1u4wF57C2tBI0bBAtuQW+s109u9ZzmQ/sQv/YvyTP8bdnEjd/OgCNRwvRJuhoz0mrj8bkEtPGEpe27SENQNU/vuDS/3mAs8+806vcoEHxaAfGMnL/Vtt7dYzcu4Wca+/HXFnrkR23MJNYWzw3Hi1w7ksSdG6L3LuK55h5U4maMZbjNz/skXehPbuqq7IGiF+YSdz8GRI7R+rDOtg6d3apAU1CB1ur12Eq98T+kkvfeICzWz1f8PWFvrMnY5Uv73VV7G3XOnl2LEep3+7Cs0NZm6tqUcdGSXdbYqMwG+p61W939VOPqe70I05efmHXI4p+PurqUTG1KIofiaL4JiCKorgL6T/7gKDO3iSK4iuiKI4VRXGs66QFoCknH+1gPZqkWAR1IFGzr6Eu+4BTmtrsA0TPTQMg6rqJNHz1vV/GPKkpJ5+g5A5u9I2TqPXA1c2bZuNeTcNX/u2W0Zmac/LROrCjbnD3XJd9AJ3Nc+SsiTR8LXlWhYeS+vpGSp5+g6bvTvjEK399Nz+kr+GH9DXU7D5AzNypAPQbPQxLfbP9Vn67zBU1WBpb6Td6GAAxc6dSs+eAa7Z+Sa56lso6AU1SnL2s67O/dUpT//EBoue4l7UmKc6+OFyd2J+glERM58r9YmsGJ6AeILEjb5hMvYvn+uxviZojDVwRsybSaGOrBziztSmJ9l2CFM8dKvvrbnJmrCNnxjqqdx8g9qapAPQbPZS2hmbMFc6DtrmiFktTC/1GS8/5x940leo9BwEISo63p4ueOY6WgvO9zm0+UczByxdxaNxSDo1birHUwNGM9V4vNspf301u+hpybfHc38947u8QzxFTR5GQ9WtOLtyMtcX7GjK5yhqg7PXd5KSvlRbUf3SA/vOmdMm2Nnaw+8+bQvXudrbeni4qY6xXdl/pO7s7VtVmHyD6xkkImkA0SbEEJetpOprvlVfxt484lrGaYxmrqdnzLbq5Ur6hXjxbG1sItXnWzZ1Grc1z7d6D9s+lm9fx997y21391GOqO/1Iu36J1yN9UdY+/tNXJXhbDCsIwr+R7rhEAFuBFaIo/p8gCFOAP4iiOLYrwOGkGz0CwqeNYcCmRdJWfG/vo+z5nejX3Erz9wXUZR9A0KqlbXIvT8FS20BR1lZMxdLF1GVfv4IqLARBHYilvomC+ZtozT/rlL9V9Lw/dkTaGJI23cH/Z+/N45sq0/7/92mapAttkwZK01KgLeDXkbVsKohQaEHH5RkFHcUHUdmkbGXXEcWFQREVB1nV0edxZlRgZn7o46gUUBl1FGSpFYUuFAp0T/ctSZPz++OkadImbQqWU/G8Xy9e2pw755PrOtd1nXOfc+77xk+F6f39FGzeQ9Ty+6hNz6Yy7QiCVk3sq0sIcujmzH/JqTvoPztRhQQ6dTPvX0tDVuulblR+nn0aOmE40U8121z02m4il95PXUY2VQ6b+2xKJei6OBorqjm7QLK558Jp9EyZijm3eUBazgNraWxxJ8vSqPJ2GOj7x9mEjR+Gvd7MmdTXnFMIDkx7iR+SlgEQPDieuKbpBz87xrk/vAGAfspo+j43C39DKLaqWmpP5nL6/med+1arbF51O/s4+3nxdciE4UQ/OQtB5UfZrv3Nvv4+m6r9Dl+/spRAh6/PLXgRy/ki9L8bT8T8qWBtRBRFil59j8p937bav5/gPW9Cxg8nyjE1cPmu/RRv2UXP1OnUZ2Q5tWNelrRtFTXkLdyA5XwRut9NIOLRqYiNjWAXKfrTe1Ttaz01sLfYvpptrm7QetWOWz8L3YSh2OvNZC/ZSk26FNtD9r9I+qQVAHQbEk8/x3SiFQePc+ZxaWrNa95YTmC/KLCLmC+UkLNyp8cnBD+3rivDj2wlffKqVtOYqvy8nzr6/nE2Okc+57jk86C0l8hwyed4l3w+68jnoV9tQdCqaSyX9GqOZpK7eodz3za79/tZnelroY34Aoj94yz0E4ZhqzeTnbqF2ibttI2kJy2XbB4ST3/HdMjlB4+T67D5mjdWEBgfhejQPrNqh5u2qg3tzqydbR3jyzlXGRdOxXDvJLDZyFv7JlWftR5zIXqpIwC9181x2py7dDN1Dpuv2/cyJ5OXAhA0OL55OuTPjjmnIVbpQ+i3fTma6O5YLpaSPfdFbBU1zn17O85X4tzc2EZs/xJzyhVPdSRI433Cgqv1egRgdP4/fhELpPwj8v4u/f7bXYV/65J+bK/jMgTYgNT5SgUeBR4ELgKzRVH8uj0Bbx2Xzsbbxd2VwFvHpbNpq1B0Jm11XDobbx2XTtdt5yKrM5ErtuW0ua2Oy9VKWxe1nUlbF1mdSXsdl86krY5Lp+rKdIyh7Y5LZyLncW6r49KZyJVTbXVcOhu5rkdA6bj8XHTVjkubY1xEUUwXBGEJEAVcEEVxMbAYQBAE30dWKigoKCgoKCgoKCgAYPc4CYFCe7Q3q9gi4J/AQuAHQRDudNn8x878YQoKCgoKCgoKCgoKCk20N6vYbGCEKIo1giD0BfYIgtBXFMVX8TL7pIKCgoKCgoKCgoKCws9Nex0XlSiKNQCiKJ4VBGE8UuelD0rHRUFBQUFBQUFBQaHDdOkBLl2Y9jouhYIgDBVF8QSA48nLbcCfgUG+CMj1Cp+fjCHhr5JnwGWdpQML9/2MyDk4Xy7kfDVVtMs0OF+muAb5BvTKNYAZ5LszZBblGUislXECTvFXGF9yTfKhlnFCAqtMg+StcuWUTOcKkHcyF4Wrm/ayaQbgtrqQKIqNoijOAMZ12q9SUFBQUFBQUFBQUFBwob1ZxVpPhN687auf/+coKCgoKCgoKCgoXN105UUeuzLyPL9UUFBQUFBQUFBQUFDoAO2Ncek0Qm4eRq+1s6VVW99Lo2jr3922Cxp/+rySStCgeBrLqzmb8iKWC8WE3DSEqNUzENT+iNZGLq57m5qvM3zWDR0v6aLyw/SuZ92+m1IJHBSPrbya3PmSrkoXQtyOVQQN6Ydp90EurNnZYZu7jUsg6qnZ4OdH+ftplGzf00q710tLCRwYj62imrwFG7BeLHZuV0f1oP++LRS/+i6lr/+zQ9qxzz2MfuIw7PUWsha/Rm1Gbqs2wYPj6O9Ynbf8wHFyn/iz2/aoR+8g9qkZfPubh1qt8u0NufwdcnMC0U/NQlCpML23j+JtrXV7v5xK0KB+NJZXSavIXygmaEh/YtanOBoJFG56l8pPW6/k3hZyHWe5cqrbuASMT86R7N21j1JP9m5cSsDAftgqqjm/8AWsF4tRR0fQP20b5jMXAag/cZr8J7b4rNtE7LMPo5uYgL3eQvaSzV5ju59jRfWKA8fIXSPFdszK3xM+eRTY7VhNlWQtfg1rUblvujLlFECfZx9Bl5jgXOW7LuNMqzZBg+KaV/k+eIxza6TVtsNvu4HoZfcS2L8XJ29d5Vy12hf6r3sIw0RpxesfF22lxoPNIYNjufZPks2mA8fJ+sNbzm29HplC9MNTEBttmPYfI+fZv/qsLZe/+zz7CPrEBGxt+DrYxdflLr7uvWYG+qQR2C2NmM8VkZO6GVtVnc82937mEcISh0sr2Kdupu4Hz8fZuYL9waPkPSlpq3TdiN+2DG1MBObzxeTM24itstZnm690fMl5bgaIe+5hwh3xddpLfHUbHMcAR3yVHTjOmRbxFf3oHcQ9NYP/dCC+Lienrtu5hKD4KAD8Q4NorKrjyMSVPumGjh9G76dngcqP0nfTKNzyD7ftgsaf2E1LCBosnS/OPLoRywXpPBWZcjfd75sENjt5T75O1RcnfNJsQq64Vrg6kOeJi58fMc/NJefBp/lp4gL0d9xEQP8YtyaGe5OwVdbw47h5FL/xAVGPPQhAY1kVOQ+v41TyYs6lvkqfTakd1s2e8TQ/JS5Af6cH3d8n0VhRw483SbrRj0u6otlC/sa/cvG5ty/Z5qhn5pE7cy1ZySmE3TEObT93bf09ydgqa8icMJfSN/cSuXqm23bjE7Oo+eJoh6X1E4cRGGfk2A0LyV6+nfgX5nhsF//CbHKW7+DYDQsJjDOiSxzm3KaJMqAbN5iGCyW+C8vlbz8/ej07lzMPPs2pSSno7xiHtoVuuCO+frp5LiVvfoBxtaRbf/ocp29fyulbl5Dz4Fp6/XE+qDqQJnIdZxlzKurpRzn70FNkT55P2O03e7a3qpasxDmY/ryXyFXN9lrOFZJz2yJyblt0SZ0WXWICAXFGjt+4gJwV24h73nNsxz0/h5wV2zl+4wICXGI7f+te0icuJT1pOWVpR4lZOs0nXdlyCghLTCAg1kj6mBRyV24ndr1n7djn55K7chvpY1IIiDUSNkHSrjuVR9asDVR/82OHdA0ThxEUG8k31y/i1PKdXLNhlsd212yYzanlO/jm+kUExUYSnjgUAN2Y6+g+ZQSHJyzn8M3LyNv2oc/acvlbl5hAYKyREw5fx7Xh6zMrt3FiTAqBsUZ0Dl9XHkonfcISMiYtpeFMPtEL7/ZZOywxAW1sFBlj53N21Tb6rJ/rsV2f9fM4u2obGWPno42NImxCAgDGlLuo+jKDjLEpVH2ZgTHlLp91r3h8yXlupjm+vrthIVnLt9PPS3z1e2E2Wct38J0jvvQt4kvfwfi63Jw6OWcTRyau5MjElZR89C0lH33rm7CfH72fm0vmfz/DyQkLCb/zJgL693Jr0v33STRW1vDD2Ecpev0Dej0+A4CA/r0Iv3MsJxMXkvnA0/ReNw/8fD9HyhXXClcPsnRcgob2x3y2EEteEaK1kfIP/01Y8ii3NmHJozHtOQhAxb++ImTMYADqT+bSWFQGQENmHn5aNYLGtwdHwS11P2itq0seTZlDt/yjZl17vZnaIz9hN1suzeYh/bGcK8B6XtKu/PAQoUmj3dqEJo2m4u8HAKj8+Cu63TjEZdv1WM4X0pCZ12Ht8MkjKd71OQA1x7LwDw1CHaFza6OO0KHqFkT10UwAind9jmHKSOf22GdmcvbZd0D0faYQufwtxVcBlvMu8dXC12FJoyn7u2t8Sb4WGyxgk9489dNqOjxfoVzHWa6cChwyALOrvf93iJCk693ahEy6nnKnvV8S7GLv5RI+ZSQlu78AmmI72HNshwRR44jtkt1fED5F8o2tpt7ZThWk9fl4y5VTAPrJoyjd06SdiSosGHWEvoW2HlVIoNPm0j2fo58ixWFD9kUacvI7pAnQfcoICncfAqDqqORrTQubNRE6VN0CqfouC4DC3YfocYtkc/SDyZzbvBfR0giAtbTKZ225/K2fPIqSDvq6xMXXlV+kO+tJ9dFMNEaDz9q6yaMw7fkMgNp2tGuPngbAtOczdI7Y1k0ehWm39H3T7s/QTXGvQ23ZfKXjS85zM4DBJb6qLzG+4p+ZSW4H4+tyc8qViDtuoOifvg09Dm46Rzr8Xbb3S3TJ7vGhS26On/KPviZk7GDH56Mp2/sloqURy/lizGcLCB7a32eb5Yrrrohd6Nr/uiqydFw0kQYs+aXOvy0FJtQ93Qu6OjIca1Mbmx1bdS0qfYhbG92tN1J/Mtd5ImwPdQtda4EJdWRrmgBRVgAAIABJREFUXUs7upeCf6QBa4GLdqEH7Z4GLAUttUMRArX0mHc3xa++e0naGqMBc77J+be5oAxtixOo1mjAUtDcxlJQ5jzJhiePwFJQRt2P5zqkK5e/1S19XVDqQdfgNb6Chg7gmrTXuObTP3HhD1udFx6+INdxljOnrAXNdxgbC0pb6/Z0aWOzY6+uQ6UPlX53TE/iP3yV2HfXEzTyOp80XdFEhmN2sdtcYGp1cagxGrC4xb8JTWS48+/eq+9n+Hc76HHXOPJefM83XZlyClrbbMl3t6epjZu2hzYdRWsMp+Giu6+1xvBWbcwuug35zW2C4o3oRv8/hn+8jmH/XEvI0HifteXyt8a1PnF5vo64L5GKg8c6oO0et1L9dN+vuqV2gQmNo96ou+uwFkuvPVqLy1EbwnzUvfLxJee5GVrHl8VLfLnGtrlFfJkLyqjtYHxdbk41obv+WiwlldTnuk0C6xWNMbz5HARYCk1ojJ6OsYu/q+rw14f49N02tWWKa4WrhzY7LoIgdBME4RlBEE4KglApCEKJIAjfCIIws53vzREE4TtBEL77e81ZDw08fKnlXQpPC2W4NAkYEEPUYzPIe2xrWz+l47qeGv0c05F7sEf0yWaRnqnTKf3zXux1DZeo7Wm3LbU9fE8U8QvU0GvJ3eRteP9n0b0y/vbsR/cm3nXrTmRyOmkBmXcsI2L+VARtB9bHkes4y5VTnvAxthpLyjg99iFybl9Mwbo3iHllOX7dAjskJXjxZXvyrnbnPf83jo6YS8k/DmF86BYfhT3JXoGcgkuPsctOrMvLK8HfD39dN47e8geyn3mHga934JVE2WrYJdrcwtdRi+5GbLRT+o9DHdD2tNsWse3L7+socsSXnOdmvJnje3zFLLmbc5eUz5eXU01E/G6Mz09bvOm2dre3Nj58t2PSVyauFa4a2nsf5K/AP4HJwD1AMPAe8IQgCANEUXzc05dEUdwJ7AQ43vvOVtFmKTChieru/FtjNGAtLnNrYy0woY7qjrXQBCo/VCHB2CqkwW7qSAOxOx/jXOomLOd8u8PQtE9XXbXRgLWohW6h1MaT7uXQWFCK2uiiHWlwvp7TrF2KxtidxhbaQUMHEHbLjUSunokqNBjRLiKaLZj+9yOvepEPTaHn9IkA1JzIQRtloMkKrTEcS6G7tjnf/U61xtEmoE8k2t4RDD240fFdA0P3bSD9lsewllS0abNc/rYWtvC1sXtr3YJSr/HVhDn7Avb6BgIG9KE+I9sn7St9nJuQLacKTaiNPZx/+xu7t9Z1tGmy1y8kyKlrs0j/bfghB0teIZrYaBra8XXkzCn0nD4JgJr0bLRR3V1i29A6tgtMaKKaY1trNGBpcUwASv/5Jde+8zjnN3q+AJEzp3rOnEKP6UkA1J6QbK5p2m+UodWEApYWT540UQYshb5NOuBK9EOTiXpAsrn6RA4B0d2p5LTzd5tb7NOcb3K7Ux0QZcDs8Is5v8z5Dn718Ryw21EbQrCaPOe7XP7uOXMKEQ5f15zIds+rKAOWDvq6+7Tx6CeN4Kd7n/JopysRD97idpxd41aqn+1ou8S2tbQCdYReuisdocdqqvSqK1d8NSHHucL40BQipzfHtjbKPXbMHuLLNba1LvEV0DuCBJf4GrZvAye8xNfPmVMAgsqPiN+O4kjSap9tl46fS1xHGrC2sLepjbXA4e9QqW778t2WyBXXXR27bMsM/7Jp71WxvqIovi2K4gVRFF8G7hBFMQt4CLjkEVF16VloY41oYiIQ1P7ob7+JyrTDbm0q0w5jmJoIgO7WMVR//T0AqtBg4t9eQ/4L71D73akO6damZ6Ht66J7R2vdirTDhDt09b8dQ/VX31+qmW7UfZ+Ftm8U6l49EdT+hN0+jqr97tpV+79Fd7dU0MJuGUPNfyTtM/es5vRNszh90yxK//wBJVt3t3sxW/jWJ6RPWkH6pBWUfXKYiHvGA9AtoT+N1XVYi90LqrW4AlttPd0SpHdVI+4ZT9mnR6g7lceRgY9wdOR8jo6cj7nAxInkle12WkA+f0vxFYUmpqczvqrS3ActVu0/TPjdreNLE9PTORhfHd2DgLhoLBeKfNe+wsfZ3eYrn1P132e623vbOKr3u/u6+sC36J32jqXWYa8qPNQ5qFMd0xNN3yisee13mgrf/oT0pOXSgPqPD9Nj2s1A27Ftr2mO7R7TbqbskyMABMQane30ySOoz77oXVfGnCp6+xN+SFrGD0nLKP/kMN2nNmkPwFZV53x1olm7HFtNA90SBgDQfep4yj893HK37XLxrU+bB/9+fJjIadK6w6HD+2OrrsPSwmZLcQW2mnpCh0s2R04bR+kn3wFQ8vER9GMHAhAYZ0RQ+3vttIB8/i56+xMykpaR4fB1jw76uoeLr8PGDyMq5Xecnrkee337YzCK/+djTiYv5WTyUso//RbD1AkABLehba+pJ9ihbZg6gQqHdsW+IximSd83TGv+3JvNcsRXE3KcKwre+oTjk1ZwfNIKTC7xFZIgxba3+ApxiS+TI76+HfgIR0bO54gjvo63EV8/Z04B6McNojYrH3NB250HV2rTswhwOV+E3zmWCg/+boof/W9vpPqrDOfn4XeORdD4o4mJICDWSO2JrDb15IprhasTodUjYNeNgvA1sFIUxS8FQbgdWCCK4mTHttOiKF7TnoCnJy4AoROGE/3UI9LUre8foOi13UQuvZ+6jGyq0g4jaNX02ZRK0HVxNFZUc3bBRix5RfRcOI2eKVMx5zYP/st5YC2NLXrd3swKnTCcXmubdQs378a47H7qvs+m0qHbd1MqgQPjsFVUk5si6QJc9/VOVCFBCGp/bFW1ZE9fS0PW+VYaan/PYyJCxg/H+KRjmtzd+ynZsouI1OnUZ2RRvf8wgkZNzCtLCfhNHLbKGvIWbsB63v2iOWLxfdjrGjxOk1tVr/VsNBC3fha6CUOx15vJXrKVmnRpisoh+18kfdIKALoNiaefY6rHioPHOfP4m632M/zIVtInr3Kb6jFQbfWq29n+9vPzfKBDJgwn+slZCCo/ynbtb46v77Op2u+Ir1eWEuiIr3MLXsRyvgj978YTMX8qWBsRRZGiV9+jcl/rmVpUXnSh849zo83z/YbOzim1v82jbrfxIzCumY3g50f57jRKtu4iYonD3gOSvb1eXua09/yiF7CeLyJ0yo1ELJmOaLODzUbxpr9RfdDzSait2I794yz0E4ZhqzeTnbqF2qbYTttIetJyAIKHxNPfMR1y+cHj5P7hDQCueWMFgfFRiHYR84USzqza4XYnXxS93xHrzJwC8PfzPraq7x9nEzZemkL1TOprzilnB6a9xA9JyySbB8cT1zRd7WfHOOewWT9lNH2fm4W/IRRbVS21J3M5ff+zzn3X2rw/iB+w/hEMiUOw1Vv4afFWqtOlaUxHHtjgnIY1ZEgc1/5pPqoADaYDJ8h8XJoyVlCruHbTfLoN7INoaST76Xco//Kkc99aoe2xZJ3pb1U7vtY5fJ3j4utBaS+R4eLreBdfn3X4euhXWxC0ahrLJa2ao5nkrt7RrCu0/fpL73VznMc5d+lm6hza1+17mZPJSwEIGhzfPG3sZ8fIe+J1ad/6EPptX44mujuWi6Vkz30RW0WNc9/2NmK7M+NLrfJcR67EubnW6v213/j1s9A74ivTJb6G7X+R4y7x1TQdcvnB4+R4iK+RR7ZyvEV8WUXv94gvJ6cArn11PpVHs8j/37TWPlV77yyHJQ4nZu3D4KfC9P5+CjbvIWr5fdSmZ1OZdgRBqyb21SUEOfydM/8lp7+NC6diuHcS2GzkrX2Tqs9aj91qq3Z2ZlwDjLz4z1/Eo4y/Rj3Qpd9/m57/ly7px/Y6LkOA14EBwA/AI6IonhYEoQdwnyiKf2pPwFvHpbOR83VIbx2Xzqati7vOpK2OS2fjrePS2bTVcelsvHVcOhtvHZcrgVyx3dbJt7Npq+PSmbTVcelM2uu4dCZtdVw6Vbedjktn0lbHpTPx1nG5ErTVcelM2uq4dCZtdVw6Gzlr5y+l4/KXLt5xeaCLdlzaPEOJopguCMKDQDTwjSiKNY7PSwRByLwSP1BBQUFBQUFBQUFBQaG9WcUWIQ3OXwD8IAjCnS6b/9iZP0xBQUFBQUFBQUFBQaGJ9t4JmA2MEEWxRhCEvsAeQRD6iqL4Kl5mGVVQUFBQUFBQUFBQ8E5XXuSxK9Nex0Xl8nrYWUEQxiN1XvrgY8el2qK5vF/4CyRIlG/MhxxYbCrZtDXI8760n4zvpss11sRile84y4Ug53GWaSxAsCyq8o0zkbTlGisnn81yjQ+Us47oAs2y6Naa5RlbE6SV71qktuHXd+2ncGVob8RYoSAIQ5v+cHRibgO6A4M684cpKCgoKCgoKCgoKCg00d4TlxlAo+sHoig2AjMEQdjh+SsKCgoKCgoKCgoKCt6Q73nrL5v2ZhW70Ma2r37+n6OgoKCgoKCgoKCgoNAaeSbsdxD/3EOET0zAVm8mc/EWajJyW7XpNjiOaxwLPpUdOEbOE28B0Gf5NCKnT8JqqgIgd/3fKD9wvEvrho4fRu+nZ4HKj9J30yjc8g+37YLGn9hNSwgaHE9jeTVnHt2I5UIxAJEpd9P9vklgs5P35OtUfXHCJ80mYp97GP3EYdjrLWQtfo1aDzYHD46jf9PiWgeOk/uEtMhV75W/J3zKSES7HWtpFdmLX8NSVN7q+97o8+wj6BMlf+ekvkZdxpnW2oPinAu4lR88xrk10sJevdfMQJ80ArulEfO5InJSN2OrqmtXM3T8MHqtnQ0qP0zvplG09e9u2wWNv7SY2aB4bOXV5M5/EcuFYkJuGkL06hkIGn9ESyMX1r1NzdcZPtsK0O3mBKKflLTL3k+jZNueVtoxLy8lcGA8topqzi3YgNVxnAHUUT0YkLaFok3velyA0qvuuASMT86RFr7ctY/S7a11e21cSsDAftgqqjm/8AWsF4tRR0fQP20b5jPSqvH1J06T/8QWn3Uv1dcqXQhxO1YRNKQfpt0HubBmp8+arsQ++zC6iQnY6y1kL9nsNbb7ORagrDhwjNw1UmzHrPw94ZNHgd2O1VRJ1uLXsPoY23LpyunvPs8+gi4xwbkYo6dcDnLJ5QqXXA6/7Qail91LYP9enLx1lXNRQ19tlqN2htws+VpQ+WF6z7Ov+7ySStAgSfdsSnMdiVo9A0Htj2ht5OK6K1dHAof0p9f6BY5GAkWb/kbVp9/4rvsrrCPdxiUQ9ZRj0eD30yjxZPNLzb7OW7DBafOA/VudNtcdP03+E1s7pC3H+REgeNxwItfMQVD5Uf7+Pkw7dreyOWrjMgIH9sNWXs2FRc9jvSjllPaavhifW4BftyAQRXL/awmixffxNL2feYSwxOHSApSpm6n7wXMdcS5AefAoeU9KNqt03YjftgxtTATm88XkzNuIrbLWZ22FXz7yrIoE6CcOIzDOyJEbFpK1fAf9XpjtsV2/F2aTuXwHR25YSGCcEX2ic8gNF3f+H8cmreDYpBU+dx7k0sXPj97PzSXzv5/h5ISFhN95EwH9e7k16f77JBora/hh7KMUvf4BvR6fAUBA/16E3zmWk4kLyXzgaXqvmwd+vh+6JpuP3bCQ7OXbiX9hjsd28S/MJmf5Do45bNYlDpPs3bqXE4nLSJ+0gvK0o8Qsneazti4xgcBYIyfGpJC7cjtx6z1rxz4/lzMrt3FiTAqBsUZ0EyTtykPppE9YQsakpTScySd64d3ti/r5EfPcXLJnPM1PiQvQ33kTAf1j3JoYfp9EY0UNP940j+I3PiD68QcBaCyrIufhdfyUtJizS1+l76upPtvapB39zDxyZ64lMykF3R3j0PZz1w6/JxlbZQ2nx8+l5M29GFfPdNsetWYW1Z8f7bBu1NOPcvahp8iePJ+w229upau/JxlbVS1ZiXMw/XkvkauadS3nCsm5bRE5ty3q0MXG5fhaNFvI3/hXLj73dsdsdUGXmEBAnJHjNy4gZ8U24p73HF9xz88hZ8V2jt+4gACX2M7fupf0iUtJT1pOWQdiWy5dOf0dlphAQKyRdEcux7aRy7krt5E+JoWAWCNhjlyuO5VH1qwNVH/zY8eE5aqdDl/nPPg0P01cgP4OD76+NwlbZQ0/jpN8HfWYex05lbyYc6mv0mfTlasjDafzyLo9laxbF5M74yl6rUsBle82/+rqiJ8fUQ5fZyWnEObB13qHrzMnzKX0zb1EutRsy7lCsn+7mOzfLu5wp0WW86PDZuPaR8l7+CmyJz9K2O3j0LSwWTdtMrbKGrITZ2N66/8jYtVD0gaVH9EvL6dgzRbO3DKfc/evRmz0fSKRsMQEtLFRZIydz9lV2+izfq7Hdn3Wz+Psqm1kjJ2PNjaKsAkJABhT7qLqywwyxqZQ9WUGxpS7fNbuaohd/F9XRbaOS/fJIyna9QUA1cey8A8NRhOhc2ujidDh3y2Q6qPSWpdFu76g+5RRv0jd4KH9MZ8twJJXhGhtpGzvl+iSR7u10SWPwrT7MwDKP/qakLGDHZ+Ppmzvl4iWRiznizGfLSB4aH+ftcMnj6R41+cA1BzLwj80CHULm9UROlTdgpw2F+/6HMOUkQDYauqd7fyCtIgdCGn95FGU7GnSzkQVFow6Qt9CW48qJJAah3bJns/RT5F8U/lFOtikN0Grj2aiMRra1ZR8Xej0dfkH/yYs2f346ZJHU7bnIADlH31FyBjJ1/Unc7EWlQHSBYCfVo2g8f3BZNDQ/ljOFWA5L2lXfHiI0BbHOTR5NOV/PyDZ96+v6HbjEJdt12PJK8ScleezJkDgkAGYzxVgdehW/t8hQpKud2sTMun6Zt2PvyTYRfdSuRxf2+vN1B75Cbv50ld3Dp8ykpLdUj7XOPLZY2yHBDXH1+4vCHfks2tsq4K0PldruXTl9Ld+8ihKO5jLpS653JB9kYac/A7rylU7g1r6+sPWvg5LHo3J4euKf7nXkcamOpJ5ZeuI2GB21kxBq0EUfa/Xv8Y6EjRE8rXT5g8PEZrUwtdJo6lw2uxesy8HOc6PIB1ny7l8rOcLoek4T2p5nEdT+Q/J5qqPvyT4Bsnmbjcl0HDqLOZT0hNmW0U12H0fraGbPArTHilXa9uxufboaQBMez5D56idusnNuW7a/Rm6Ke7HSuHqR7aOi8YYjjnf5PzbXGBCYwxv3abAe5uoh6eQcHAjA155FP8w3ybxlFPXUlDq/NtS6EE30qWNzY6tqg5/fYhP321b29DC5jK0LQqc1mjA4mKzpaDMrQj2Xn0fI45up8fdN5G34X3ftSPDseS7/PZ8E5pIT3ab2mwDEHFfIhUHj7WrqY40uGlaC0yoIw0t2rj8LpsdW3UtKn2IWxvdrTdS90MuosVtfoq2tXsasLbU7mnw3sapHYoQqKXHvLspevVdn/Wa7TFgLShx/t1YUOpZt6mNzY69ug6VPhQATUxP4j98ldh31xM08roO6f4cvr5UNJHhmF30pVx119cYDVha5rxLfPVefT/Dv9tBj7vGkffie11aV05/t7T5cnK5Q7oy1U5NC19bPOVyZLiHXG5dR+pPXrk6AhA4dAAD9m1hwKebufjEVufFbbu6v8I64h9pwOoSI9ZCD9o9De7x5eJrTUxP+v3fJmLfW0/QyN90SFuO8yOAf093mxsLWx9n/0jPx1nTNxpEkd5vPUPs3lcxzPHxKY/THve6KB1rd3vULW0uMKFxHBN1dx3WYum1WmtxOWpDWIf0FX75tNlxEQTBXxCEuYIgfCIIwveCIKQLgvCxIAjzBEG4vInJBQ/LwLS8MeSxjdQo/+19HB69kGMTV2ApqiBu7Yyureth2ZtWN8I86EptfPhux6Rb34XztCqPS5u859/lu+HzKPn7vzE+PKUD2t592WabFgclatHdiI12Sv9xyAdNT7tr5ew2JQMGxBD9+AzyHuvYo/9LtlcUiUydTumbe7HXNXRM0xs+HuPGkjJOj32InNsXU7DuDWJeWY5ft0DfNH4GX18Ogg/+9rjglEuTvOf/xtERcyn5xyGMD93SpXVl9bfH+tTxXL4EYQ+6LZt0Qu30xdftnE8CBsQQ9diVrSMA9ScyyUxOIfuOpUQ8Og1Bexmn66u9jlxqXDtsPjXmYbJvW0LBc28Qs6kDNrex33bbXM750es+WzXy+NsEfxVBI37DxaUbOXvvSkKSbujYUzcfjrUv9VXh10t7z67fASqAtUDTDGO9gAeBvwD3evqSIAhzgDkAy0ISuCMoDgDjQ5MxTp8EQPWJbLRRzT18rdGApbDMbT+WfJPbkwGpjaOnXVrp/Lzgr/sZ+M5qr0bIpeu2zwITGmN359+aSAPWlrqONtYCE6j8UIUGYauo9um7LYl8aAo9p08EoOZEDtooA9VOe8Jb2WzOd79jrPHQBqD0n//m2r88zvkXd3nV7jlzChHTkxza2WiiXH57lKHVwH5Li7vVmqhmfwN0nzYe/aQR/HTvU23a3IS1wOSmqTYanK9/OdsUSm2shQ5fhwRLj7yR7v7Fvf4YZ5dswnKu0CfN5v2Wom6pXdxSW2rTUjto6ADCbr0R42MzUYUGI9pFRLMF0/9+5IOuCbWxh/Nvf2N3D7pSm0aHrl9IkNNmm0X6b8MPOVjyCtHERtOQkd2+7mX6+lKInDmFno58rknPRhvV3SW2W+ezucCEpmXOF3mK7S+59p3HOb/R8xNFuXRdudL+7jlzCj0cuVx7QrK5xrFNE2VoNaFAe7l8KVzp2um2T9fa5SmXC0wecxmkOhK78zHOpV7ZOuKKOecC9voGAgb0od6XfP4V1ZEmGgtKUbvEiDrS4HzNr1m7FI2xu9NmV+2WNmtjo9v0tdznR3A8YXGx2T+yO9Yik4c2rY+ztbCUusM/YCuXJiiq+eI7Aq6Lp/brdK96EQ/e4lZHXOuidKzbsdmldlpLK1BH6KWnLRF6rKZKfqnYfVrGXaEl7b0qliCK4qOiKH4jiuIFx79vRFF8FBjm7UuiKO4URXGEKIojmjotAAVvfeoc1G765Ag977kZgJCE/jRW12EprnDbj6W4AlttPSEJ0jvJPe+5mdJPjwC4jUvpfssoak+d92qEXLqu1KZnERBrRBMTgaD2J/zOsVSkHXZrU5F2GMO0CQDof3sj1V9lOD8Pv3MsgsYfTUwEAbFGak9ktalX+NYnpE9aQfqkFZR9cpiIe8YD0M1hs7WFzVaHzd0cNkfcM54yh80BsZHOduGTR1KffbFN7aK3PyEjaRkZScso/+QwPaY2aQ/AVlXnfMzbrF2OraaBbgkDAOgxdTzln0q+CRs/jKiU33F65nrs9b69w1ybnoW2b7Ov9XfcRKUHX4dPTQRA/9sxVH/1PQCq0GDi/2cN+c+/Q+13p3zSc6UuPQtN3yjUvXoiqP3R3T6OqhbaVWnfor9b6lSG3TqGmq8l7Zx7VnNq7CxOjZ1F6Z8/oHjLbp86LQD132eiddENu20c1fu/dWtTfcBF95ax1P7HYXN4qHPAsjqmJ5q+UVjzfLvQuhxfXyqFb39CetJyaWD7x4fpMU3K57Zi217THNs9pt1M2SdNsW10ttMnj2gztuXSdeVK+7vo7U/4IWkZPzhyuXsHc7m7Sy5fKle6djZRl56F1kVXf3trX1emHcbg8LXu1jFUf+1SR95eQ/4LV76OqHv1dA7GV0f3QBsX7ZxhrT1+TXWkibrvs9xtvn0cVftb+Hr/t+icNo+hxovN2r5RWNqxWe7zI0jHWdM3WooVx3GuOdD6OIfdJdkc6nKcaw4dQ3tNX4QALaj8CBo1CHNW29dBxf/zMSeTl3IyeSnln36LYaqUq8Ft2GyvqSfYYbNh6gQqHDZX7DvizHXDtObPFX49CG0N3BME4RvgJeDvoijaHZ/5AdOApaIotjsq6lDkNK8C/dY/gn7CUOz1Fk4v2UJNujQlXsL+Fzk2aQUA3Ya4TEt88AQ5j0tT4l2zeSHdBvZFFEXM50vIWrGjVQdELt0gf8/TAoYlDidm7cPgp8L0/n4KNu8havl91KZnU5l2BEGrJvbVJQQNjMNWUU3O/Jew5BUBYFw4FcO9k8BmI2/tm1R91vpdVnOj9wdocetnoZswFHu9mewlW6lJl6YiHbL/RdKdNsfTz2FzxcHjnGmy+Y3lBPaLAruI+UIJOSt3ut1dVvm1/f503z/ORjd+mHMK1aZpUAelvURG0jIAggfHN0+h+tkxzv7hDQCGfrUFQaumsVy6q1VzNJPc1c1rn2pUnmczCZ0wnF5rH5GmMX3/AIWbd2Ncdj9132dTmXYYQauWptZ0+Do3ZSOWvCIiF02jZ8pUzLnNg4izp6+lscVdHX+Vd5tDxg8nyjGNafmu/RRv2UXP1OnUZ2RRtV/Sjnl5KYHXxWGrqCFv4QYs54vc9tFzyX3Yahs8Tofs5+c5pbqNH4FxzWwEPz/Kd6dRsnUXEUsk3eoDhxE0anq9vIyA38Rhq6zh/KIXsJ4vInTKjUQsmY5os4PNRvGmv1F9sPXJwGJV/ay+Brju652oQoIQ1P7YqmrJnr6WBg8nwYY2Yjv2j7PQTxiGrd5MduoWaptiO20j6UnLAQgeEk9/x7TE5QePk+uIr2veWEFgfBSiI7bPrNrh8UmjHLoB/p7HRHS2v602z8cZpFwOc+TyGZdcHpj2Ej+45HKcSy6fc9isnzKavs/Nwt8Qiq2qltqTuZy+/1nnvtuqI51dO1Vecip0wnCin2r2ddFru4lcej91GdlUOXzdZ1MqQdfF0VhRzdkFkq97LmxdR3IeaF1H2rL5UuuI7ncTiHh0KmJjI9hFiv70HlX7Wk+H/GusI2p/z/4OGT8c45OO6ZB376dkyy4iHL6u3i/ZHPPKUqfNeQs3OG3umTod0WYDm52iTX+l+sCRVvuvNXt/Va8zz4+hgWavut3Gj6DnE3MQ/Pyo2JNG6db36bEIwjnjAAAgAElEQVTkAeozsqg58C2CRk30S8sJuE7y94XFG6TB/EDYnRMwzJsGiNR8/h3FL7zV2uYGjVft3uvmOOtI7tLN1Dlsvm7fy5xMXgpA0OD45umQPztG3hOvA6DSh9Bv+3I00d2xXCwle+6L2Cpq3PY/8uI/fxHPMt7s9UCXfv/tkQt/6ZJ+bK/j0hd4AZiA9MoYgA74DFgtimLrhQta0FbH5WrFW8els2mr49KZtNdx6Uy8dVw6m7Y6Lp2NtwuOzsbbBceVoK2Oy9WKt45LZ9NWx6UzkbOOeOu4dL6uUkeuJN46Lp1NWx2XzqStjktn01bHpbP5pXRcXu/iHZfZXbTj0ubVgCiKZwVBeBnpqUsOcC1wPfCjL50WBQUFBQUFBQUFBQWFn4M2Oy6CIDwF3OJolwaMAr4AVguCMEwUxXWd/xMVFBQUFBQUFBQUFH7ttPf+xVRgKKAFCoFeoihWCYLwIvAtoHRcFBQUFBQUFBQUFDqAfC+K/rJpr+PSKIqiDagTBCFHFMUqAFEU6wVB8Mnnco33kJNGuzzreppFeXS1Mmafv+bXF19yvSOuUcszngjALnbJV207FYtMY03qbPKMJ9J2YPXtn5sgmeqITaZzBYDZKtP5Qi3P2C2AqnqtLLqynZut8tXsBpnql8LVT3vZZBEEIcjx/8ObPhQEIQyls6igoKCgoKCgoKCgcIVo79baOFEUzQBN0yE7UCMtQqmgoKCgoKCgoKCg0AF+hS8S/Cy0N6uYx7n0RFEsBUo75RcpKCgoKCgoKCgoKCi0QLbFEULHD6P307NA5Ufpu2kUbvmH23ZB40/spiUEDY6nsbyaM49udK7+G5lyN93vmwQ2O3lPvk7VFye6vG4TfZ59BH1iAjbHYlN1GWdatQkeFOdcbKr84DHOrZEWguy9Zgb6pBHYLY2YzxWRk7oZW1WdT7r91z2EYaK04NOPi7ZSk9F6NuuQwbFc+ydpAUrTgeNk/aF5Ualej0wh+uEpiI02TPuPkfPsX322Ofa5h9FPHIa93kLW4teo9aAdPDiO/o7FL8sPHCf3iT+7bY969A5in5rBt795iMay6nY1u92cQLRj8bay99Mo2bbHbbug8ZcWbxsYj62imnMLNmB1WV1aHdWDAWlbKNr0rsdFILuiduj4YfRaK+ma3k2jaOvfW+n23ZRK4KB4bOXV5M5/EcuFYlS6EOJ2rCJoSD9Muw9yYc3Ojtk7LgHjk3Okxdt27aN0e2t7e21cSsDAftgqqjm/8AWsF4tRR0fQP20b5jPSqvH1J06T/8SWDmlfaj6rdCHE71xJsMPmpsXNOlsXfr115LqdSwiKjwLAPzSIxqo6jkxc6bPNctSRkJulnBJUfpje85xTfV5JJWiQdJzPpkg5FXLTEKJWz0BQ+yNaG7m47m1qvs7w2Va5tXs/8whhicOlBQJTN1P3Q+v4ChoU17xA4MGj5D0pxZdK1434bcvQxkRgPl9MzryN2Cpr29WUs2aDPPEF8uWUnPEV99zDhDt8fdqLr7sNjmNA0yLgB45zpoWvox+9g7inZvCfDvha4epAnhFjfn70fm4umf/9DCcnLCT8zpsI6N/LrUn33yfRWFnDD2Mfpej1D+j1+AwAAvr3IvzOsZxMXEjmA0/Te9088PPRDLl0HegSEwiMNXJiTAq5K7cTt36Ox3axz8/lzMptnBiTQmCsEd2EYQBUHkonfcISMiYtpeFMPtEL7/ZJ1zBxGEGxkXxz/SJOLd/JNRtmeWx3zYbZnFq+g2+uX0RQbCThiUOl3z3mOrpPGcHhCcs5fPMy8rZ96LPN+onDCIwzcuyGhWQv3078C55tjn9hNjnLd3DshoUExhnRJQ5zbtNEGdCNG0zDhRLfRP38iH5mHrkz15KZlILujnFo+8W4NQm/JxlbZQ2nx8+l5M29GFfPdNsetWYW1Z8f9dlO2bX9/Ih5bi7ZM57mp8QF6O+8iYD+7rqG3yfRWFHDjzfNo/iND4h+XHrbUzRbyN/4Vy4+93ZHrQU/P6KefpSzDz1F9uT5hN1+cyt79fckY6uqJStxDqY/7yVy1UznNsu5QnJuW0TObYs63Gm5nHwWzRbyX/wbF569NJuVOtLxOnJyziaOTFzJkYkrKfnoW0o++tZnm+WqIzHPzSXnwaf5aeIC9Hd4yKl7k7BV1vDjOCmnoh6TcqqxrIqch9dxKnkx51Jfpc+mVJ9tlVs7LDEBbWwUGWPnc3bVNvqsn+uxXZ/18zi7ahsZY+ejjY0ibEICAMaUu6j6MoOMsSlUfZmBMeUun+yVrWYjU3whY07JGF9Nvv7uhoVkLd9OPy++7vfCbLKW7+A7h6/1LXyt76CvuyL2Lv6vqyJLxyV4aH/MZwuw5BUhWhsp2/sluuTRbm10yaMw7f4MgPKPviZk7GDH56Mp2/sloqURy/lizGcLCB7av0vrNqGfPIqSPZ8DUHMsE1VYMOoIvVsbdYQeVUggNUczASjZ8zn6KdJvrPwiHWxSOFUfzURjNPik233KCAp3HwKg6mgW/qHBaCJ0bm00ETpU3QKp+i4LgMLdh+hxy0gAoh9M5tzmvYgWaTYYa2mVzzaHTx5J8a4mm7PwDw1C3UJbHaFD1S2IaofNxbs+xzBlpHN77DMzOfvsOyD6tshs0ND+WM4VYDkvHeeKDw8R2uI4hyaPpvzvBwCo/NdXdLtxiMu267HkFWLOyvPZTrm1pdgudMZ2+Qf/Jix5lFsbXfJoyvYcBKD8o68IGSPFtr3eTO2Rn7CbLR22N3DIAMznCrA67K38v0OEJF3v1iZk0vXN9n78JcEu9l4Ol5PP9nozNUd+wm7u+GxSSh25tDriSsQdN1D0z698tFi+OuKWUx+2zqmw5NGYHDlV8a/mnKo/mUtjURkADZl5+GnVCBrfX3CQU1s3eRSmPVLs1rYTX7VHTwNg2vMZuimjmr/viH3T7s/QTXHPDW/2ylWzQZ74AvlySs74Mrj4uvoSfR3/zExyO+hrhasHWTouGmM4loLmITKWQhMaY7h7m0iXNjY7tqo6/PUhPn23q+m67TvfZR/5JjSRnvRNbbYBiLgvkYqDx3zS1RrDabjYrGsuMKFt8du1xnDMLroN+c1tguKN6Eb/P4Z/vI5h/1xLyNB4n3QBNEYD5vzm/ZoLytC2uFDSGg3uNheUOS+mwpNHYCkoo+7Hcz5rqnsasLr42VpgQt3T4L2NzY6tuhaVPhQhUEuPeXdT9Oq7Put1BW11pMEttqwFJtSRLXRd48+pG9JhrZa61oLmu16NBaWe7W1qY7Njr65DpQ8FQBPTk/gPXyX23fUEjbyuQ9qXk8+Xg1JHJDpaR5rQXX8tlpJK6nMLfdIFeeqIpkVOWTzlcmS4h1x2jy/drTdSfzLXeePnl6Hd7EeplrgfQ3XL+CowoXHUG3V3Hdbicum7xeWoDWHtaspZs0Ge+JL2KU9OyRpfLXxt8eJrV5vNLXxtLiijtoO+Vrh6uOSOiyAIHXsR3v3brT5p1XEWvLXx4btdTtf7vlvtxFMb3NtELbobsdFO6T8O+Sp8abqOJoK/H/66bhy95Q9kP/MOA1/vwKNhj9IttT18TxTxC9TQa8nd5G1433c9uHQ/iyKRqdMpfXMv9rqGjmnKre3Fh+026owbVj4e38aSMk6PfYic2xdTsO4NYl5Zjl+3wA4IXU4+Xw5KHemQbosmEb8b06GnLd6lO7uOeN6fe5u27Q0YEEPUYzPIe2zrVaUt+BKDHdKUsWaDPPHlXbhFk07IKRnjy3Np8t3XMUvu5twl+VrhaqHN53uCIHi7FSgAt7bxvTnAHIDHdEO4K7iv23ZLgQmNsbvzb02kAWthmcc21gITqPxQhQZhq6j26bvekEO358wpRExPAqDmRDaaKJd9RBmwFJV70De4tylsbtN92nj0k0bw071Ptakb/dBkoh6YCED1iRwCortTifRYX2s0YC501zXnm9zuegREGTA77DPnlznfna0+ngN2O2pDCFaT5wFxkQ9Noef0iQ6bc9BGGWhqqTWGY2nhN3N+C5sdbQL6RKLtHcHQgxudv3vovg2k3/IY1pIKr7ZbC0tRu/hZbTRgLS7z2MZa6DjOIcHYKqoJGjqAsFtvxPjYTFShwYh2EdFswfS/H3nV6wra1gKTW2ypjQasRS11pTYtdS8Ha6EJtbGH829/Y3cP9kptGh26fiFBTl2bRfpvww85WPIK0cRG05CR7ZP25eTz5aDUkUurIwCCyo+I347iSNLqdm2Wu45YWuSUxlMuF5g85jJITyNjdz7GudRNWM75/nRJDu2IB2+hhyO+ak9ko4lq9qNUS9qJL6MBi6PeWEsrUEfopactEXqspsp29eWom3LFl5w51cSVji/jQ1OInN5sszbK3Y9mD752tVnr4uuA3hEkuPh62L4NnGgnl7sqXXkcSVemvScuJcB3wFGXf985/kV4+5IoijtFURwhiuKIlp0WgNr0LAJijWhiIhDU/oTfOZaKtMNubSrSDmOYNgEA/W9vpPqrDOfn4XeORdD4o4mJICDWSO2JLJ+MlUO36O1PyEhaRkbSMso/OUyPqeMB6JYwAFtVnfORehPW4nJsNQ10SxgAQI+p4yn/VPqNYeOHEZXyO07PXI+9vu3xCBff+rR50N7Hh4mcNg6A0OH9sVXXYSl2T3JLcQW2mnpCh0vv20dOG0fpJ98BUPLxEfRjBwIQGGdEUPt77bQAFL71CemTVpA+aQVlnxwm4p4mm/vTWF2HtYW2tbgCW2093RIk7Yh7xlP26RHqTuVxZOAjHB05n6Mj52MuMHEieWW7BaouPQtN3yjUvXoiqP3R3T6OqhbHuSrtW/R3S4U07NYx1Hz9PQA596zm1NhZnBo7i9I/f0Dxlt0+d1rk1K5Nz0Lbtzm29XfcRKWH2A6fmgiA/rdjqP7qe5/t8kb995loXewNu20c1fvdB4hWH3Cx95ax1P5H0lWFhzoHpqtjeqLpG4U1z/cLvMvJ58tBqSOXVkcA9OMGUZuVj7mg/c5aV6gjWpfjrL+9dU5Vph3G4Mgp3a1jqHbksio0mPi315D/wjvUfneqXVvl1i7+n485mbyUk8lLKf/0WwxTpdgNbiO+7DX1BDviyzB1AhWO+KrYd8QZ+4ZpzZ+3Z++VrptyxZecOdXElY6vgrc+4fikFRyftAKTi69DEiSbvfk6xMXXJoevvx34CEdGzueIw9fHfchlhauL9kZUnQEmiqLYasSbIAjnL1nVZidvzesM+OtT4KfC9P5+GjLPE7X8PmrTs6lMO0Lpe/uJfXUJA7/chq2impz5LwHQkHme8g+/4rqDr4HNxrkndoLdx36rXLoOKg4cRTcxgaFfb8XumMa0iUFpL5GRtAyA3NU7nNOYVnx2zPkOeuy6WQhaNde+L90lrTmaSe7qHe3qmvYfxzAxgRu+/RO2egs/LW5+tDvywAbn9ImnV73BtX+ajypAg+nACUwHjgNQ8O5Brt00n1FfbES0NPLTIt9nfyrffwz9xAQSvnkNe72Z7CXN2kP2v0j6pBUAnFn1Ov0cUx9WHDxOuUP7krDZyX9yO3H/+zSo/CjftR9zVh49U6dTn5FF1f7DlO1KI+blpVzz+Q5sFTXkLdxw6XpdQdtm5/yanfT7y1ppesv3D9CQeR7jsvup+z6byrTDmN5Lo++mVH7z7+3YKqrJTdno/Pp1X+9EFRIkXTRMHk329LU0ZPmQ4jY7+Wu30/d/nkHw86N8dxrmrDwilkj2Vh84TPn7++j18jL6H9yJrbKG84teACB41EAilkxHtNnBZiP/iS3YKms6ZPOl5jPAoP/sRBUS6LQ58/61NGRd6FTdX3MdAej5X5fwmhjy1ZELa3YS/457TkUuvZ+6jGyq0g5jej+NPptS+c2h7TRWVHN2gZRT3R+8FU1fI5GL7iFy0T0A5DywlkYfnj7IrV154ChhicMZ9NU2aTrkpZud267b9zInk5cCcPaxHc3TIX92jEpHfBVs+Qf9ti+nx30TsVwsJXvuiz7ZK1vNRqb4QsackjG+yvcfI3xiAiMcvs508fWw/S9y3OHr7FWvO6dDLv8ZfK1w9SC0eo/TdaMgpABfiqKY7mHbQlEUN3v4mhvf9fqvX920D412eWaZrrPJsyyPVpDvgWdIgMc1Uq9qGm3yxJdGbZNFF6DBItuSU7Kh1JErR5Cm47PL/dKRq45o1b4P5P65qW7QyqJrFuXxtV4r3/mxxqKWTfumwj2/iDXpN8c80KWvjxee/0uX9GOb2SSK4hZAKwjCSABBEH4jCMJSQRBu9aXToqCgoKCgoKCgoKCg8HPQ3uD8p4BbAH9BENKA0cDnwGpBEIaJoriu83+igoKCgoKCgoKCgsKvnfbeCZgKDAW0QCHQSxTFKkEQXgS+BZSOi4KCgoKCgoKCgkIHsHfJF7G6Pu29eNkoiqJNFMU6IEcUxSoAURTrUWZyU1BQUFBQUFBQUFC4QrT3xMUiCEKQo+MyvOlDQRDC8LHj4ifIM/bI4yJHVwiVnzyDmIO18gwwvfwF/S4duQZbqvx+ff12q00lm7ZcdUROVDLZHB5wGQv5XQZ2GW8/1sk0kPjXWEfqZRy0LZe/w9VtTzveWcgV1wC6wF/fxDkKV4b2Oi7jRFE0A4ii6JrxauDBTvtVCgoKCgoKCgoKClcpv77bFj8PbXZcmjotHj4vBUo75RcpKCgoKCgoKCgoKCi0QJ7JxRUUFBQUFBQUFBQUFDqAbKu6hY4fRq+1s0Hlh+ndNIq2/t1tu6Dxp++mVAIHxWMrryZ3/otYLhSj0oUQt2MVQUP6Ydp9kAtrdnZIN+RmSVdQ+WF6z7Nun1dSCRoUT2N5NWdTJN2Qm4YQtXoGgtof0drIxXVvU/N1xi9Cu9u4BKKemg1+fpS/n0bJ9j2tdHu9tJTAgfHYKqrJW7AB68Vi53Z1VA/679tC8avvUvr6Pztkc7ebE4h+UjrOZe+nUbKttXbMy83a5xZswHrBXXtA2haKNnVcO/a5h9FPHIa93kLW4teozcht1SZ4cBz9m1bnPXCc3Cf+7LY96tE7iH1qBt/+5iEay6p91u7z7CPoExOwOVY2r8s401p7UJxzZfPyg8c4t+ZNAHqvmYE+aQR2SyPmc0XkpG7GVlXXpXWbtHWJCc7V3D1pB7loV7hoh992A9HL7iWwfy9O3rqK2u9zfNbt/cwjhCUOl1b5Tt1M3Q+edZ2rfB88St6Tkq7+thuJXnovAf178eNvV1J3hXRVum7Eb1uGNiYC8/licuZtxFZZ26W1Q25OIPqpWQgqFab39lG8rXX96v1yKkGD+tFYXsW5BVL9ChrSn5j1KY5GAoWb3qXy0298thXkO1/A5dWR3it/T/iUkYh2O9bSKrIXv4alqNxn7V9bHZGrhshls5w5BRD77MPoJiZgr7eQvWSz19jut2mB5O8Dx8hdI8V2zMrfEz55FNjtWE2VZC1+DasPsS3n9UhXQ3lV7NKQ54mLnx8xz80le8bT/JS4AP2dNxHQP8atieH3STRW1PDjTfMofuMDoh+XhtSIZgv5G//KxefevmTdnAef5qeJC9Df4UH33iRslTX8OE7SjXpM0m0sqyLn4XWcSl7MudRX6bMp9Zeh7edH1DPzyJ25lqzkFMLuGIe2n7uu/p5kbJU1ZE6YS+mbe4lcPdNtu/GJWdR8cbRj9jq0ox3amUkp6Dxohzu0T4+fS8mbezG20I5aM4vqzzuurZ84jMA4I8duWEj28u3EvzDHY7v4F2aTs3wHx25YSGCcEV3iMOc2TZQB3bjBNFwo6ZC2LjGBwFgjJ8akkLtyO3HrPWvHPj+XMyu3cWJMCoGxRnQTJO3KQ+mkT1hCxqSlNJzJJ3rh3V1aFyAsMYGAWCPpDu3YNrRzV24jfUwKAbFGwhzadafyyJq1gepvfvRZs0lXGxtFxtj5nF21jT7r53ps12f9PM6u2kbG2PloY6MIm5AAQP2pPLJnv3DFdY0pd1H1ZQYZY1Oo+jIDY8pdXVvbz49ez87lzINPc2pSCvo7xqFtUb/CHfXrp5vnUvLmBxhXS/Wr/vQ5Tt++lNO3LiHnwbX0+uN8UHXg1CPX+YLLryMXt+7lROIy0ietoDztKDFLp/ms/WurI3LVENlsljOnHDYHxBk5fuMCclZsI+55zzbHPT+HnBXbOX7jAgJcYjt/617SJy4lPWk5Zb7GtpzXIwpXDbJ0XIKH9sd8thBLXhGitZHyD/5NWPIotza65NGU7TkIQPlHXxEyZjAA9noztUd+wm7u+CwdQS11P2ytG5Y8GpNDt+Jfzbr1J3NpLCoDoCEzDz+tGkHj+wMrubSDhvTHcq4A63lJt/LDQ4QmjXZrE5o0moq/HwCg8uOv6HbjEJdt12M5X0hDZp7PtrrabDlXgMWhXfHhIUKTW2gnj6a8SftfLbSTr8eSV4g5q+Pa4ZNHUrzrcwBqjmXhHxqEOkLn1kYdoUPVLYjqo5kAFO/6HMOUkc7tsc/M5Oyz73R42jT95FGU7GnSzkQVFow6Qt9CW48qJJAah3bJns/RT5F8U/lFOtikezHVRzPRGA1dWrdJu7SD2qUu2g3ZF2nIyfdZrwnd5FGY9nwGQG07urVHTwNg2vMZuimjHLoXZNHVTR6F6f9n78zjm6ry/v++bZN031LaJqVAW8BRkR1R2SmF4oz6ewZwRvFRQARlka0g7riCgIoLKOgo8zijI6gz6jiiBVTUGdkKFR2BthQKdN+bLlnv74+bpkmaNClabtX7eb140TbfnPf9nHvON/fk3nPOTun9VTs/IzrTtV90N7aUv9r6cs2HXxLllkeiMkZS/a5z/pL6sthicrSrAI0aOrlYmlyfF/Dj84jV0OyICwjVIHbC/K8tj8iVQ1rZF9uznH0KIDZzBBU7v7B7ziMoMsxz244IbfO88wti7XnEuW0Hhmr8OgY5r0cU/XLU4cBFEIRAQRDmC4LwmCAIo9xee+BCoapELabitrn95pIqVIlat5jYthirDWtDI4ExEReKBEDtxjWVVKFKaM81++BGX3sNzd8XIpos3Z4dlKjFXOJU16Ue6jpBi6nEnRuJEKKhx53TKH/uLX9ttivX7H6e3T07x3hgl10gW63TYiyucvxuLKlG4/ZhotFpMZW0xZhKqh0fOLGTh2Mqqabpv2c6z3Zuu4CpuAp1Ymz7GGe2hxiA+JsmUrs3p1tzW8s1/kTszkjqV21lSrnEtUyVO7ekCnWi/4OyruCq4qIxl0uPVZjLa1Bpo7o1W+WeR0oqPeRsT31Zyl+hg/tzSfaLXPLJ85y7f4vjossfyfV5AT8+jwD0Wn0Tww+/TI9pYyha/7b/7F9ZHpErhzjKvcie5exTrX6c69tYUtVuwKXWueYaY4mr516rb2bYoa30+P1Yijb8zSdTzusRRb8c+brjshUYB1QBzwuC8IzTa16fLxAEYZ4gCIcEQTj0nuG0hwAPb2r3rbaHoB+7rYE/XE8bwDiFBPdPRn/vrRTdu+XnwfZQpugXVyRh2UwqX3sfW9MF7uvgpVx/YhKXzaTyTz+G7alYd7aH94kiASFqei6d1qmLDNdyL9C3WwPX3z0N0WKj8r193ZvrpVy/2tmP7dR+9CvBn3r5uXBlY19oX5b+azp6khMZizh5/QriF0xH0HRifwm5Pi+8FetnHmlV0bq3ODTsTire/RLdnMxOsH9leUSuHOKt3C73LGOfwr8c4XFnJaeQonVvcnj4fCre24du9lR/oB6QF+l6pBtK7Ob/uqt8PW90pSiKAwEEQXgR2CIIwnvATXhp0wCiKG4DtgHkJN/Qzr+5pAq1Ps7xu0qnxWx/FMoRUyrFmEurIDCAwIgwrLX+T472JJMbV63TYi5345ZUofLCVSVqSdl2L2eWbcJ0pvRnwbaUVKLSOdV1otbx2JmDW1qJWheHxY0bOrg/UVOvIXH1LAIjwxBtIqLRRNX/feQX21xaicr9PLt7tse4ew4d3J+oa69Bd6//7MTZmSTMTAfAcLQAjV5La4vR6GIxlbqyjcWu3zCp7THBvRPR9Ipn8N6N9vdqGfzpenKn3ou5otYjO2FWJvEzM+zsfNdzrde2m5Brcvt2S63XYipti4mbMZ6YScP54Q8Pe/UrJ7eV3cPObjyaj0Yfh8GpXPeJmr7Y/ir+tqkuXLW+rUwpl/jg6rSY3PrAxeaaK2tRxcdIdzziYzBX1XVbNtj7qXMe0cW1z9klnvuys4z557A1txDcvzfNx/J9cqVyL+7nxU+VR9xV+fcvufQv93F2ww6v7F9bHpErh7Sy5cqdIE+fSpyVScLMSZLnXKm+29q2tn3bLqlyyTUaL7mz8u9fcekb93F2Y8df9sl5PaLolyNfd1zUrT+IomgRRXEekAvsBcIvFNqYm4emjw51cjyCKoiY68dQl33AJaY2+wCx0ycCEPPbUTR8/e2F4hxqys1Dk+LEva49ty77AFo7N/raUTT8W+IGRoaRtv1Bip96g8ZDx3827KZv89D00aPqmYCgCiLqurHU73bl1u/eT/Q06YM6auooDP+RuKduXM2JMXM5MWYula99QMWWnZ1KEk25eaid2NHXjaXezXN99n5iWtnXjsJg91xw42qOj57L8dESu3yzb3bp67vInbSS3Ekrqd51gPgbxwMQPrQfloYmzOWugw5zeS3WxmbCh/YDIP7G8VR/cpCm40UcHHA7h0cs4PCIBRhLqjg6eZXXQQtA2fZdHMtYwbGMFdTsOkCP6a3s/ljrmxyP5rSxa7AaWggf2h+AHtPHU/OJVDdR44egX/g/nJi1Fltzx8/my8VtZX+XsYLv7Oy4TrLjnNidUfmfP+b7ycv5fvJyaj7Zj3b6BADCOuDaDM2E2bna6ROolZlb++lBtDOk92tn+D4eOdnQmr/0qJMTHPmrPnu/S0z97gPETmufv9TJCY6Jw6qkHgSnJmE6V+aT2aqL/XnxU+URgOCURCpkSZsAACAASURBVEdc7JQRNOef75D9a8sjcuUQOT23So4+Vbp9F7kZWdKE+o8P0GPGOLtn723bZmhr2z1mjKN6V2vb1jniYiYP99m2Qd7rEUW/HAntbtM5vygIfwH+IoriLre/zwVeEkXR571JT3dcACInDKPnmtulpYHf3kPpCzvRrbiZpm/zqcs+gKBRSctbDkjFWttA4cKNmIqkjnn5v7cRGBGKoArCWt9I/sw1tOSddTt2z8cTOWEYSQ+3ccte3Eni8ptpOpZPvZ3be9MyQi9PxVLbwOlFEjdh8QwSFk7HWNg2+a/gljVY/Pi28mKxgwI9P+MaMX4Yuofsyw/u3E3F5h3EL5tJ87E8GnYfQFCrSH52OcGXpWKtM1C0eD3ms65JMH7JTdiaWjwuP9jREygR44ehty+HXLNjN+Wbd5BgZ9fvljwnP7OckMtTsdZKbJMbO2HpTVgbPbMbWjRe2alr5xI9YTC2ZiP5S7dgyJWWxxy0ewO5k1YCED4ojb72ZUxr9x7h1H1/alfOsINbyJ1yj8tyyIEBHT9P3OfJO4geP8SxrGfr0pxXZD/NsYwVAIQNTGtb1vOzHE7f/yoAg7/ejKBRYamReIbDJylcvbVD3sXger3F6sSOsrNPObEHZD/Nd07sVCf2GTs7JnMkfR6fS5A2Emt9I43fF3Li5sccZQcI3htZryfmObiFy19wLGl8+afP8P3k5QCEDkxrWxr4sxyKHngFgOjMkfR+fC5BsVFY6xtp+r6QkzMf9V3RP5IbGBNB35ezUCfFYTpfSf78DVhrDV5ZF5OtCrJ6ZEZMGEbSQ3MRAgOo3rG7LX99m+/oy72flfqypbZBWrr1bBkx/zOe+AXTwWxBFEXKnvsbdZ/ub1e+zea9hXX150Wz2ftH2Y/JI5e8mkVIXz3YRIznKihYtc3lW20lj7TndlUOgY4ff+lKz8Eqz3NSu7pPNZk6vkRLeXIuMROGYG02kr9sM42tbTt7I7kZWZLnQWn0sy+HXLP3CIV2z5e8upKQND2ivW2fumerS9uOCPY8gOvq6xGAKwo/9PVx1S20vvct3fmJLFad+Uu3rMcOBy4AgiBcCYiiKB4UBOEyIBM4Loriv/wBeBu4dLW8DVx+yfI2cOlq/RSP7V+oOhq4dKV8XXD8EiVnl+po4KLop5W3gUtXq6OBS1ero4FLV0rJIxdXcmURbwOXrpavgUtXytvA5WJIGbj8NOquA5cO57gIgvAwMBUIEgQhGxgJfA6sFgRhiCiKT3T9ISpSpEiRIkWKFClSpOjXLl+T86cDgwENUAr0FEWxXhCEDcB+QBm4KFKkSJEiRYoUKVLUCf367rf+NPI1Od8iiqJVFMUmoEAUxXoAURSbUepckSJFihQpUqRIkSJFF0m+7riYBEEItQ9chrX+URCEKPwcuJisgT/i8C5cgTI+E99o8W9X+59awYHyPJtusslzjgHiwppkY8slq83X9w1dIzmfx29sUfsO6gLJ6VmueRdyySxTuwaICWv2HdQFkqsvS2x5Hl+Xay4mQLNRnj5lscpznkNkmlsDUNoUKhv7CtnIii6GfF1hjxVF0QggiqJztlEBt3XZUSlSpEiRIkWKFClS9AtVt56Z343V4cClddDi4e+VQGWXHJEiRYoUKVKkSJEiRYoUuUm++9SKFClSpEiRIkWKFClS5KfkmYxhV+/Hbidm4lCs9g2fmo6dahcTdkWqY8Onmr05nHlQ2tir14O3EpMxHJvJgvFMGQXLXsBa7/98h16P3k7UxGHSBm7LXqDpu/bs0CtS2zZw23uYoockdmB0OGkvrUCTHI/xbDkFd27EWtfoF7fvE7PRpkuej9+9GcOxwnYx4QNT+c3zCwkMVlO1J4f8+193vJZ0eyZJc6YiWqxU7c7h1GN/8YsrZ12nPT6bWLvnk0u8e77EvoFb9Z4cCh6QPPfOmkHizEmYq+oBKFz7JjV7jvhkho0dRuKD8xACA6h5+1Oqtu50eV1QB6HfuIKQAX2x1jRw7u51mM+XA6C5pA+6xxcREB4Kokjh/1uKaDL77VcudvjYoegftm/s9XY2FS+/047b8+nlhAxIw1rbQNGi9Q4ugErfg36fbqb8ube8buzVnfy2So6+HDl+CD3XSBurVr2VTdmWd9t57rNpGSFXpGGtaaBwwQZM58oJjI4gdes9hA7qS9XOvZx7cFunvLYq5fE5xKQPwdZsIm/JizR66FNhA1PpZ+9TNXuOUPjAawD0WvVHYjNHINpsmCvryV/yIqaymnbv9yS58iZAymNziE4fiq3ZRP7SF7x67mvfLK92Tw6FD0qek1f9kdgpV4LNhrmqjrwlL2L2w7OcbVuu/hwxbihJD89FCAyk6m+fUv5S+7bd65llhF7RF0tNvbQp4rlywkcPRr/6VgRVEKLZQvGT2zHYd3r316/uoXmS3x2fUunJ78blBA/oi7W2gbOLn8J8vhxVUjz9sl/CeEraub356AmKH9jsN7dVcuWRXo/MhcAAKt/KpnTze+08p2xaSujANCw1DZy6ayOmc9I5Tlw4jbibJoHVRtFDr1D/xdFO+Y0YJ+UwITCAqr95zmG9n11G6BUS+/RC6TxHjBnkcp7PP7Edw7+PdYp9yROziEuXNr78/u6XaPDQlyMGpnD58wsIDFZTuecIJ+7fDkD45b25dMMdBGpUiBYrP6z+E/VHCjrF7y6yKQ+LXZBku+MSPXEoISk6jo5aSOGql0ldO89jXMq6+Zxa9RJHRy0kJEVH9IQhANTtyyV3wlKOTVpOy6likhZP85sdNXEomhQ9x0Yv4PQ9L9F77XyPcb3X3snpe17i2OgFaFL0RE0YCoBu4e+p/+oYx0YvpP6rY+gW/t4vbmz6EEJSdOy/ajEns7bSf/0dHuP6r7+Dk1lb2X/VYkJSdMROHAxA9KjLicscwcEJKzg4bjlnX/rAL66cdR2TPoSQVB0Hr15MXtZW+j7l2XPfpyTPB69eTEiqjhi7Z4Dz2/5JzqSV5Exa6deghYAAdGvuomjOw+RPuYuo68ai7pvsEhI9YwrWOgP5E++g6vV/EH/PbOmFwACSnsmi5MHNnJq6gDM3r0a0dGLRA7nYAQHoH72TwllryJu8kKjrx6Jx48bcOBlrnYGTE+ZT+af3SVw9y+V13QNzMXxx2H+vcvq1S5a+HBBA8uPzyb/1EX6YuIiYG8YQ3M/Vs/aPGVhqDfx3zJ2Uv/oBSfdJUwJFo4nijX/l/OPbO+XTWa19KufqxeRnvUzaU577c9pTd1CQtZUce5+Knij15/Nb3ufoxBXkTlpJTfZhkpfP8IsrV94EKYcFp+o4cs0iCla+ROo6z55T182jYOXLHLlmEcFOnou3vE9u+nJyM7Ko9tezzHlErv7c87H5nLrtEY5PWkjM9WPRuLXt2D9kYK0z8MO4+VT86QN0q6W2ba2p59Scxzkx5W6Klm+i17PLOuf3kbs4Pfth8qcsIOq6cZ791jeSN3EeVa+9T+I9bX5NZ0op+N3dFPzu7gsatMiVR3o9Pp+T//so309YTOwNYwju19MlJO6PGVjqDHw3+i7KXvmAnvfdCkBwv57E3jCa7ycu5uQtj9DriTshoBOXc/YcVnDbI/yQvoiY6z3kMPt5/u9YKYfp75XOs6W6noI5T3B88hLOLHuO3ps6cZ6BuPTBhKYk8vVVS/gh6xUuXX+7x7hL18/lh6xtfH3VEkJTEtHarwn6PzSTUxvf4Zv0eyhYv4N+D87sFF/Rz1+yDVxiplxJxTufA2DIOUlgVBiq+BiXGFV8DIERIRgOnwSg4p3PickcCUDdF7lgldYLaDh8ErVO6zc7esqVVL3zGQCNPtiNh08AUPXOZ0RnXtn2/p3S+6t2fka0/Zh8KS5zBGU7vwCg/nAeQZFhqOOjXWLU8dEEhYdQf0jyXLbzC+KmSlz9bZMpeuEfiCZppRBzZb1fXDnrOm7KCMp2SJ4bcjr23GBnl+34gjh7XV+IQgb1x3SmGPPZUjBbqPvnPiImXeUSEzFpJHXv7QGg/uOvCLt6EADhY4bScvw0xuPSN0DW2gaw+b8Kjlzs0EH9MJ0pwXy2DNFsoe7DfURmuLbLyIyR1L4rces+/prwawY5vXYVprOltJws8turnH5bJUdfDhvcD+PpUkxFUl3XfPAlUZNd22v05JFUv7MXgJqPviZi1EAAbM1GGg/+gM144btKx04ZQfmOzwEw5OQRFBmKyq1PqeKjCQwPdfSp8h2fo80cAYDV0LaCVkCoBtHPb/3kypsAsZkjqLDnToM9j3j0HBHalsN2fkGsne3sOTBU49esWDnbtlz9OXRwP4ynSzDZuTUffkmUGzcqYyTV70ptu/ZfXxMxSuI2f38KS3k1AC0niwjQqBDU/j3UETKoP0Znv//cR0SGe11fRY3D71eEOfn9sZIvj5Q48kj1+18RPdn1fdGT28qt+ejfRIweaP/7SKrf/wrRZMF0thzj6RLCBvfz22+oew77sH0Oi5o8kqp3nM+zxG7+vhBL2YWdZ4AemSMo2bkPgDof10F1h/IAKNm5j/ipUv4SRQiKCAEgKDIUo593ixX9ciTbwEWdGIupuG1+v6m4CnVibPuYkqoOYwDib5pI7d6cTrC1mIrbyjWXVKFyK1flzi6pQp0oXbCr4qIxl0udxVxeg0ob5RdXo4vFeL6tTGNJFRpdbPsYJ66xuC0mNE1P1MhLGfrxkwz++yNEDE7ziytrXetiMRa7ela7eVa7e3aL0c/JZOjejfR/9i6CosJ8MoMStJhL2vxaSitRJbgOtoIStZhLKqRfrDZsDU0ExkSi7pMEokiv1x8l5f3n0M7z/+6SnGypzDauubQKVaIrV5WgxdQaY7VhbWgkMCYSIURDjzunUf7cW53yCvLWNcjTl1WJWpf+JDHd6tq5zznqOqJz5rxIrdO69alqNG5fJmh0WjfP1S5fOPRafRPDD79Mj2ljKFr/tn9cmfKmxI7F6FTnUo5w9azWuR6fscQ1h/VafTPDDm2lx+/HUrThbz6ZsuYRmfqzyp1bUumhbWsx+2jbUddeQ/P3pxxfsvnHrXD8bilpX9eqBM91DaBOTiDtw+dIeWstoSMu94vpLDnatloX23b+AFOph8/GxFjXc1zfRFBMhF/v7ZDtlsNMJVXt6zsx1ud5jr72Gpq/L/T7PANodDG0OF0HtZRUEex27MG6WFpKqttiiqvR6KSB5MkH/0y/h25hTM5m+j38v+Q/0fl23l1k6+b/uqs6HLgIghAqCMIqQRBWCoIQLAjCLEEQPhAEYb0gCOE/iix4WENeFH3HuH1Vpr97GqLFRuV7+zrB9lSsa7mCP8fXabUvs12RHriiPUgICiAoOoycqfdR8OgbXPbKcj+xcta1J7b/x1e8/VMOjFxMTvpKTGW1pK659cKY7YM8MoWgQEKHX8b55Rs5/YdVRGRc3blv9uRid9BuOjw2USRh2UwqX3sfW1OLfyxfZbYP8sj90XXtpegu78t+MD17vnCkz6LbnWsP73OKKVr3FoeG3UnFu1+im5N5wdyLkzf9K9djS3QKKVr3JoeHz6fivX3oZk/1B+rPkXk8rq7IIxelP3vx45vb9mNwv2T0q2/j7L1bLoDfEddzjKWimhOjZ1Nw3RJKnniV5GezCAgP6RxLlrZ9odcDfr63c+jOn+f+yejvvZWiTp/nC23b0n89Z2Vw8qE/8+XQhZx86M9c9uydneQr+rnL1/297cBZIAT4CPgB2AhcB7wE/K+nNwmCMA+YB3BP1GD+X2gKAAmzMomfmQGA4Wg+an2c4z1qvbbdBFGT27dqar0WU2lbTNyM8cRMGs4Pf3jYp9H426bSw85uPJqPWt9WrkqnbTdRsx1bp8Vkvz1qrqxFFR8jfbMSH4O5qs4rVz97CvpbJgFQfzQfTVJbmRqdFlNptUu8dIfFKcbJs7G4msqP9gPQcCQfbDZU2kjHxHVnyVnXutlT0M2UPDcczUej79izyd2zro1trmyr25K/7mbAG6t98i2llah0bX6DEuMwl1V5iOmBpbQKAgMIiAjFWtuAubSSpgPfYa2R6tTwxSGCL0+j8d+5Prlysi0lrlxVotZxO79V5tJK1Lo4BzcwIgxrbQOhg/sTNfUaElfPIjAyDNEmIhpNVP3fR93Sr1x9uVXmkiqX/iQx3etaijG71fWFKnF2Jgkz0wEwHC1Ao9fSWppGF+sxj7h6bh8DUPn3L7n0L/dxdsMOj1w56zpxViYJ9jxiyM1Ho49z8uwhd5ZUuRyfxont6vkrLn3jPs5u7PhOk6x5RKb+bHbzrNLFtW/bJZWovLRtVaKWPtvuo2j5JkxFpX55lbhVqHQ9HL8H6eIwl7fvU57qGsBqkv5v+a4AU1Ep6pQkWo7ld8iUO49I5Tl9LidqMbt/NtpjzCX2uo6UPPvzXp9s52sCnbZ9fZdUdXieU7bdy5llmzCd8X2ee86eTM9bpPxVd7SAYKfroGCdFmOpa123FLvehQnWxzpidDeOc0zUL/vgGy57xvN8JEW/XPl6VKy/KIorgIXA5cBiURT3AasAr18fiaK4TRTF4aIoDm8dtACUbd/FsYwVHMtYQc2uA/SYPh6A8KH9sdY3OW61tspcXoPV0EL40P4A9Jg+nppPDgAQNX4I+oX/w4lZa7E1+35evPzPH/P95OV8P3k5NZ/sRzt9AgBhHbBthmbC7Gzt9AnU2tm1nx5EO0N6v3ZG2989qfj1TziUvpJD6Sup/PggCTPGARA5rB+WhiZM5bUu8abyWiyGZiKHSc+rJswYR+WugwBUfnyA6NHSnrAhqToEVZDHQQvIW9clr3/imExftesgCTdKniOGevdsbWwmYqjd843jqPxE8uz87Gvc1CtpPH7WJ7/525Oo+ySh6pkAqiCifjcWw579LjENe/YT9XspkUZOHU3jf6TVbwz7ctBc0gchWAOBAYReeQXGPN9MudlN3+ah6aNH1TMBQRVE1HVjqd/t2i7rd+8neprEjZo6CoOde+rG1ZwYM5cTY+ZS+doHVGzZ6ddFjlx+5erLrWrMzUPTR4c6OR5BFUTM9WOoy3Z9X232AWKnTwQg5rejaPja/9WVPKn09V3kTlpJ7qSVVO86QPyN4wEIt/cps1ufMtv7VLi9T8XfOJ5qe58KTkl0xMVOGUFz/nmvXDnrunT7LnIzsqQJ9R8foIc9d3bk2WZo89xjxjiqd7V61jniYiYP79Bzq+TMI3L156bcPDQpetTJEjfmujHUZ7t6rt99gNhpUtuOvnYUDfaVwwIjw0h9/SFK1v8fjYd+8NsrSHXt4vd3Y2nY3b6uYxx+2+o6MDbSMTFdlZyAuo8esx+Dpu6QR4JT2vJI7A2jqfWQR1rLjfntNTR8fczx99gbRiOog1AnxxOcoqPxaJ5PZquk8+yUw65rn8Pqsg+gne75PKdtf5Dip96g8dBxv3jnXv+Ub9Lv4Zv0e6j4+CC6GWMBiOrwOqiFKPt1kG7GWCrsfdlYWkPMNZcBEDtmAE2n/B8gK/plSGh3i875RUE4KoriYPvPr4miOMfptVxRFH3e+/5G/3uvgD5P3kH0+CHY7Ev0Nn4rLWl3RfbTHMtYAUDYwDTHEr21n+Vw+v5XARj89WYEjQpLjfQNgOHwSQpXb3WUHSh0fN+01xPziLKzC5e/QJOdffmnz/D9ZOkRrNCBaW1LH36WQ9EDr0hlx0TQ9+Us1ElxmM5Xkj9/A9Zag6PsRov3G1n91t5O7MTBWJtNnFiymYZcacnF4Xs2cCh9JQARg6TlkKWlgY+Sd5+05KKgCuI3m+4ifEAfbCYLBY+8Qe1X3znKDg70vmpNV9a1yRbYYV33XXs7MRMGY2s2cWLpZgx2z0N3byBnkuQ5fJDTcsh7j1Jg93zJC4sJH9AHURQxnq0gb+VWlyQXF+Z5Webw8cNJeGAeQkAAte9kU7nlbXosvYXmY3kY9uxHUKtIejqL4MtTsdY2cG7JemkSLhB1wwS0d84ARAyfH6L8qdc9Mrypq9lWm+fvGyLGD0P3kH351J27qdi8g/hlM2k+lkfD7gMIahXJzy4n+LJUrHUGihavx3y2zKWM+CU3YWtq8bh8amCA56deL0ZdN7aovVV3l/Zlb54jJwyj55rbpaVE395D6Qs70a24maZv86nLPoCgUUnLIQ+QPBcu3IipSKrry/+9jcCIUARVENb6RvJnrqHFw0Vts1nl1XPq2rlETxiMrdlI/tItGHIlz4N2byDX0afS6GvvU7V7j3CqtU+9mkVIXz3YRIznKihYtc3l7oW6gzzSlXVt9tKuW5Xy5FxiJkhLqOYv20xjq+fsjeRmZAEQNiiNfvblkGv2HqHQnsMueXUlIWl6RLvnU/dsdfEcHer5sSq5+jJ0fX+22jw/ChcxYRhJD81FCAygesduyl7cSeJyqW3X75badu9nlxNyeSqW2gZpOeSzZSQsvpH4BdMxFRY7yir434exuN19CAr0nkd0D96BEBBAzc5sKrbsIH6p3e8eyW/PZ1Y4/J69+ynMZ8uIzLyG+KUzEa02sFop3/QmDXs9Dxyajd77VFe2bcHL9UjUxGEkr5kDAYFUvb2bkhfeQZ91E425+dRlH0TQqEh5bimh9jxSsOBpRx7RLZ6O9g+TwGqlaM2fqP+s/dzTwADv10GRE4aR9HBbDnOc52P51NtzWO9Nywi1n+fTi6QclrB4BgkLp2N0Ps+3rGl3niuNwV7Zv1k7B+3EQVibTfx3yUvU268JrtrzFN+k3yMd36BULn9+AQHBKir3HOXEfVL/ib7yEi55fBZCUCA2o4kf7vkTDd+6LqecUfa2P895yq5He8/s1ushP3Tmr92yHn0NXF4FloqiaHD7exrwZ1EUR/sCdDRw6Ur5Grh0pToauHSlOhq4dKV8DVy6Ut4GLr9kdXSx05XydhF/MdTRwKUrJafnjgYuXamOBi5dKV8Dl66Ut4FLV0uuviyx5bkm8TZwuRjqaODSlfI2cOlqdTRw6Wp1NHDpaikDl59G3XXg0uEVtiiKcwVBuFIQBFEUxYOCIFwGZAIngDEX5QgVKVKkSJEiRYoUKVL0q1eHAxdBEB4GpgJBgiBkAyOBz4F7gMHAE119gIoUKVKkSJEiRYoU/ZLUnZcc7s7y9UzTdKQBigYoBXqKolgvCMIGYD/KwEWRIkWKFClSpEiRIkUXQb4GLhZRFK1AkyAIBaIo1gOIotgsCIJfg0VNkDzPS8uphNBGWbg1hk6uXf8TKTpYnufDASxWeZ4R92uLhy7ST7AtxgVJzufxm63yzBuz+b+v2k+uKI3vFfy6QjVGjSxcjX8fKV2i2iZ5nseXcy6mXGTBLBMY+TybZMpf6gD5rr/iNPJdFyj6ZctXbzIJghAqimITMKz1j4IgRKHc5VKkSJEiRYoUKVKkqNOSaX2Mn718DVzGiqJoBBBF0XmgogJu67KjUqRIkSJFihQpUqRIkSIn+VpVzOjl75VAZZcckSJFihQpUqRIkSJFihS5SZ4HL4GIcUPoueYOafOjv2VTtuVdl9cFdRC9n11G6BVpWGoaOL1wA6Zz5USMGYR+9a0IqiBEs4XzT2zH8O9j3Z4LEDZmGAkPzEcIDKB2xydUbdvZjq1fn0XwgL5Yaxs4v2Qt5vPlRF4/Hu3caY44zSUpFP6/uzH+cMpvdspjc4hOH4qt2UT+0hdoPFbYLiZsYCp97Zu31e7JofDB1wBIXvVHYqdcCTYb5qo68pa8iLmspt37PSl83FCSHroDAgOofjubipfeaec5+ZnlhAxIw1rbwJlF6zGfKydkUD96rl1kDxIo2/Qm9Z9847ffC+W2SqXvQf/szZRtesvj5m0dsscORf+wfeO4t7OpeLk9u+fTbeyiResxn3dl9/t0M+XPdY4tV13L6Rkg7fHZxKYPxdps5OSSzRg8tO3wgU4bnO7JoeABaTOz3lkzSJw5CXNVPQCFa9+kZs8Rv7h9n5iN1s49frd37m+eX0hgsJqqPTnk39+2CWHS7ZkkzZmKaLFStTuHU4/9xS+unDms3xOz0aZLm/T99+4tHj1HDEzhUvsGulV7jpDn5Lnn7Zkkzcl0eC547K9+s1Men0NM+hBszSbylrzoNYf1s5/nmj1HKHxAymG9Vv2R2MwRiDYb5sp68pe8iMnPHCZX7gTo/djtRE8c6tg4uOlY+5wfekVq28bBe3M486C00Wjs764macUfCOnXk++vvcex6bC/3JiJUtv2xg1z4tY4cXs9eCsxGcOxmSwYz5RRsOwFrPX+7bcll185PcuVv+TyLGf+6m6yyTbr6ucteWbbBgSQ/Ph8Cm57hB/SFxFz/RiC+yW7hGj/kIG1zsB/x95J+asfoL9XejLNUl1PwZwnOD55CWeWPUfvTcu6P9fOTlyzgLNzH6Jg6p1E/m4c6r6u7OjpU7DWGyiYNJfq1/9O/Mo5ANR/8DmF1y+m8PrFFGc9jfl8WacGLdEThxKcquPINYsoWPkSqevmeYxLXTePgpUvc+SaRQSn6oieOASA4i3vk5u+nNyMLKqzD5O8fIbfnpMevZPCWWs4mbGQ6OvHonHzHHvjZKx1Bk6Mn0/Fn95Ht3oWAC0nisi7bhl51y6h8NaH6fnEQgj0s7n+CG6r9A/OpeHzw/7x3Nh6Oztv8kKiPLBj7OyTE+ZT+af3SXRj6x6Yi+GLTrLlqms5PQMx6UMISdVx8OrF5GVtpe9Td3iM6/vUHZzM2srBqxcTkqojZuJgx2vnt/2TnEkryZm00u8P/dj0IYSk6Nh/1WJOZm2l/3rP3P7rJe7+qxYTkqIj1s6NHnU5cZkjODhhBQfHLefsSx/4Z1jGHKZNH0JoSiLfXHU3x7O2ccn6uR7jLll/B8eztvLNVXcTmpLo5nk4ByZkcWDcCope+tBvdut5zrl6MflZL5P2lOcclvbUHRRkbSXHfp5bc9j5Le9zdOIKcietpKYTOUy23AlEf1jTGgAAIABJREFUTRxKcIqO3FELKVz1MilrPbNT1s2ncNVL5I5aSHCKjqgJErvpeBF5c9fT8M1//Wa2eg5J0XHUzk3tgHtq1UscHbWQkBQd0XZu3b5ccics5dik5bScKiZp8TSP7+8ufuX0LFf+ks2znNdgin4xkmXgEjq4H8bTpZiKyhDNFmo+/JKoyVe6xERNHknVO3sBqP3X10SMGghA8/eFWMqqAWg5WUSARoWg9u/GkVxcgJCB/TGdKcZ8thTMFuo/2kdE+tUuMeGTrqLuvd0A1O/6itCrB7UrJ/J346j/8Au/uQCxmSOo2Cm9x5CTR1BkGKr4aJcYVXw0gRGhGA6fBKBi5xfEZkp1YzU0O+ICQzV+L80SOrgfpjMlmM5K9V374T4iJ4909TN5JDXv7gGg7l9fE36N5FlsMYJVmlYlaNSInVhK68dwpdeuwlRUijGvyG+mgz1IYpvt7LoP9xGZ4cbOGEltK/tjN3bGVZjOltJysnNsuepaTs8AcVNGULZDatsN9ratdmvb6vhogsJDaLC37bIdXxCXeWW7sjrFzRxBmb1P1R/umFt/yM7d+QVxUyWu/rbJFL3wD0STtGyZubLeL66cOSwuczilO/f59BwYHkL9oTwASnfuo8fUEQAk3TaZMy+832nPALFTRlC+43OgNYeFes5h4aGO81y+43O0mRLbOYcFhGoQ/UxicuVOgJgpV1L5zud29kkCo8JQxce4sWMIjAhxsCvf+ZyYTKnvteSfp6Wg2H+gE7eik9wKJ27dF7mOfNJw+CRqnbZb+21ly+FZrvwF8niWM38p+uVIloGLOlGLqbhtioyppApVgmujVyXGYm6NsdqwNjQSGBPhEhN97TU0f1/o+CDsrlyAoEQtlpI2trm0kiA3dlCCFnNphYNtMzQRGBPpEhP527HU/7NzAxd1YixGJ9/Gkqp2SUat02IqrnKNSYx1/N5r9c0MO7SVHr8fS9GGv/nFVSVo2+oSMHuqb+cYR31LnkMG96f/p5vp/8kLnH9giyNJdiVXCNHQ485plD33ll8sdwUlajG7nOcqVInt2aYS7+zyC2DLVdcgn2cAtS4Wo3u71cW2jynxHqOfk8nQvRvp/+xdBEWF+cXV6GIxnnctU+PG1bhzi9tiQtP0RI28lKEfP8ngvz9CxOA0v7hy5jCNLpaW8655xJfnFhfPOqJH/oZhHz/BkL+v8dszSPnJ9TxXo3HLYRqdFpMT21RS7ZLneq2+ieGHX6bHtDEUrX/bP65MudMT21TsWm5rjItnDzGdlTox1rWN/Qhu/E0Tqd2b4zdXDr+OcuXwLFP+cvi5yJ7lzF/dUWI3/9dd1emBiyAIJ3801dMScO7f8nraKMMpJLh/Mvp7b6Xo3i3dn+sN7sYWPLLbYoIHXYKt2Ygx70znyD7K9XJ0Lr6L1r3J4eHzqXhvH7rZU/0F++R2FNN89CQnJy8k//rlxN81A0Gj6nJu4rKZVP7pfWxNF7gGvYdy293B8MJOWDaTytcukC1XXXsp96J49lquPzFSUPH2TzkwcjE56SsxldWSuuZWf8HeiuyQ21ovQlAAQdFh5Ey9j4JH3+CyV5ZfKLZb5bCO2JLncA5PvZ/8R99gwCudeNTDI9qd7eF9TjFF697i0LA7qXj3S3RzMv3DypU7JbgHtB/1/WMvOS40l7hx9XdPQ7TYqHxv3wVzL4pfb+XK5Pni5K+Oy+0w5sd4ljV/KfqlqMP7bIIgNNDWZFpbU2jr30VRjPTyvnnAPID7YwYyLbyPy+umkirU+jjH72qdFnN5tUuMuaQKlT4Oc2kVBAYQGBGGtbYBAFWilpRt93Jm2SZMZ0r9cyojF8BSWkmQro2tSozD4s4urUSV2AOLnR0QHupgQ+vdls/94iXOyiRh5iQADLn5aPRxtJak0WkxlbqyjSVVqPVt33xodFpMZa4xAJV//4pL37iPsxt9f2NpLq1E5VTfKk/1bY/xVN+OYys4h625heD+vWk+lt+l3NDB/Ym69hp0984iMDIM0SYiGk1U/d9HPrkAlpJKVC7nWeu4ve3MVuviHOfZhT31GhJXd54tV13L4Vk3ewo6e9tuOJqPxr3durVtU3GVy7fzUow0QdpcWef4e8lfdzPgjdVeufrZU9DfInHrj+ajSeqYa3Tn6tu4xuJqKj/aL3k4kg82GyptpGOSrTdd7ByWNHsK+lvSpeM8WkBwUhx1nHB4Npa6TjR39xys12K014uxuJoKh+cCu+cIzFWubbBVibMzSZgpsQ1HC9DotU45LNZjfTvfDVF7iAGo/PuXXPqX+zi7YYdnroy5M2FWJj1mZgDQeFRiG1r96LXtJvab3O4AqZ3aWGeUMCuTeDvXcDTftY3pte0WMvDFjZsxnphJw/nhDw/75Mrht5Uth2e58pecnl3Kk+kaTNEvR77uuGwH/gH0E0UxQhTFCKDI/rPHQQuAKIrbRFEcLoricPdBC0BTbh6aFB3q5HgEVRAx142hLvuAS0xd9gG00ycCEH3tKBr+/S0AgZFhpG1/kOKn3qDx0HH/ncrIBWg+dhJ1Hz2qngmgCiLyt2Np2OO6cpNhz36ifi8ltMjM0TR9823bi4JA5NQx1H/k3zc5pdt3kZuRJU0K/fgAPWaMAyB8aD8sDU2Yy2td4s3ltdgMzYQP7QdAjxnjqN51EIDgFJ0jLmbycJrzz/t1DE25eQ7PgiqI6OvGUu9W3/XZ+4mZJl2cRF07CoO9vlU9ExwTxFVJPdCkJmFyWvWrq7gFN67m+Oi5HB89l8rXPqB8806/By0ATd/moXFiR103lvrdbuzd+4luZU8dheE/EvvUjas5MWYuJ8ZI7Iot/rPlqms5PJe8/oljMmrVroMk3Ci17Qh72za5tW1TeS3WxmYi7G074cZxVH4itW3n58njpl5J4/GzXrnFr3/CofSVHEpfSeXHB0mw96nIYd65FkMzkcPs3BnjqLT3qcqPDxA9+goAQlJ1CKogn4MWuPg57Pzrn3AwfRUH01dR8fEBEmeMdXi2eqtrJ8+JM8ZSuesQABUfHyRm9AA3z54HLQClr+8id9JKcietpHrXAeJvHA90nMOsjW05LP7G8VR/0prDEh1xsVNGdJjD5MydZdt38V3GCr7LWEHNrgPETW/13B9rfRPmctcLS3N5DVZDC+FD+wMQN308NZ8ccC/Wp8q27+JYxgqO2bk9Osnt4cSNGj8E/cL/4cSstdiaTd3Sr5ye5cpfcnpulZzXYIp+ORJ8TcQVBGEYsAFpAPMikC+KYqq/gCO9bvAIiJwwjKSHb5eWxHt7D2Uv7iRx+c00HcunPvsAgkZF703LCL08FUttA6cXbcRUVEbC4hkkLJyOsbBtEl7BLWuwVNV5wsjCDdaYPbLDxg0n4X77csjvfErVS28Tt+QWWo7lYdi7H0GtQr8xi+DLpCVjzy97SprMD4ReeQXxK2dzeob3R0pqDCFeX0t5ci4xE4ZgbTaSv2wzjbnSUpGDsjeSm5ElHd+gNPrZl/Ss2XuEwvtfBeCSV1cSkqZHtIkYz1Vw6p6tLt8KhWu8J62I8cPQ25fordmxm/LNO0hYNpPmY3nU75bqO/mZ5YRcnoq11kDR4vWYzpYR/T8TiL9rOqLFAjaRsuf/Rv2n/i/Re6FcZyUsvQlrY4vH5Xk93kF3Yusesi8NvHM3FZt3EG9nN+w+gKBWkfzscoIvS8VaJ7HNbuz4JTdha/LM9tZlu7qu5fRc0xzsld137e3ETBiMrdnEiaWbMeRKK+4N3b2BnEkrAQgf5LSc6N6jFNwnLel5yQuLCR/QB1EUMZ6tIG/lVpcLB1sH6bHf2tuJnTgYa7OJE0s202DnDt+zgUPpEjdikLQcsrSM6VHy7FxBFcRvNt1F+IA+2EwWCh55g9qvvnMpP8pLv+rqHFZj1Hj13H/t7WgnDsLabOKHJVscnkfsWc/B9FUOz5c+v8C+BPRRTt73mt1zIJduWkD4gN6IJgv5j7xBzVffO8rWCB3Pq0pdO5foCYOxNRvJX7oFQ2sO272BXMd5TqOv/TzX7j3Cqdbz/GoWIX31YM9hBau2ueQwQfB+orsydwZ2wAXo8+QdRI2Xlp8+texFxxK/A7Kf5ruMFRJ7YBqprcsDf5bDGTs7JnMkfR6fS5A2Emt9I43fF3Li5sccZXdE7vPkHUTbuQVO3Cuyn+aYEzfNiXvazh389WYEjQpLjTQoNRw+SeHqrW11LZNfOT2bbIFeuV2Zv9QB1g4cd61nTZBn9sW4BhtS9P7PYk/6e/vc3J2nkrD29Jvdsh59DlwABEEIABYBM4A0URT1/gK8DVx+yfI2cOlqdTRw6Up1NHD5paqji/iuVicX/frJJKfnjgYuXamOBi5dLW8Dl65WRwOXrpSvgUtXqqOBS1fK18ClKyUXWc4rIbk8dzRw6Ur5Grh0pbwNXC6GlIHLT6PuOnDxOTlfEIQrgWGiKD4P3Ae8LAjCtV1+ZIoUKVKkSJEiRYoUKfrZSRCEWEEQsgVByLP/H9NBbKQgCOcFQXjRV7m+Juc/DEwFggRByAauBL4AVguCMEQUxSc66UORIkWKFClSpEiRol+1bN160eGfRKuBPaIorhMEYbX993u8xD6GNL7wKV+790wHBgMaoBToKYpivSAIG4D9gDJwUaRIkSJFihQpUqRIkbNuAMbbf/4z8DkeBi72ufQJwC5guK9CfT0qZhFF0SqKYhNQIIpiPYAois2AfA8kK1KkSJEiRYoUKVKkqLsqQRTFEgD7//HuAfY59E8DK/0t1NcdF5MgCKH2gcswJ1AUfg5c5JpIHBAg3y04i0WeiXhx0Y2ycE1GX82o66TWyLNzrtXS6b1bfzKJtm45X65L1SLK06fklMHUiU1Af0JFq42ycOtNalm4AGqZJslbxV9fXw6QcUECq02evB2pkadPNRrl61NWm/Ldti919wfFnPdktGubKIrb3GJ2A4m01/1+YhYA/xJF8azHDX89yNcV51hRFI0Aoig6t0IVcJufB6VIkSJFihQpUqRIkaKfieyDlG0+YiZ5e00QhDJBEHSiKJYIgqADPG0QdzUwRhCEBUA4oBYEwSCKotfdVDscuLQOWjz8vRKo7Oi9ihQpUqRIkSJFihQp+lXqA6SbHOvs/7/vHiCK4szWnwVBmAUM72jQAn4sh6xIkSJFihQpUqRIkaKfTrZu/u8n0DogQxCEPCDD/juCIAwXBOHVCy1UtskJkeOH0HONtMt31VvZlG151+V1QR1En03LCLkiDWtNA4ULNmA6V07EmEEkrb4VQR2EaLJw7ontGP59zG9uxLihJD08FyEwkKq/fUr5S+25vZ5ZRugVfbHU1HNmkcQNHdSP5LUL7UECpZveou4T/3dyBwgfOxT9w/bdxd/OpuLld9qxez69nJABaVhrGyhatB7z+XJUSfH0370F46nzADQdOUHxA1v85oaOHkb8fXdBQAB17+yi5tUdrlyVisSnstBc1g9rbT0ly9diKS4DVRAJa+4meEA/sImUP/kyzQe/7bRn3UPzJM87PqXSk+eNywke0BdrbQNnFz/l8Nwv+yWH5+ajJyh+YPPPwnPYmGEkPDAfITCA2h2fULVtZzvP+vVZDs/nl6zFfF66g6q5pA+Jjy0mMDwU0SZy+vdLEE3+bWgaNnYYiQ/OQwgMoObtT6na6oG7cQUhA/pirWng3N3rXLi6xxcREB4Kokjh/1va7bmtuuSJ2+iRLu1s/t3dL9Fw7HS7mIiBKQx4/i4Cg9VU7DnCifv/DED4Zb24bMNcAsOCaTlbwbd3vYjV0NytuQCpj88hNn2ItNv2khdpPFbYLiZ8YCr9W3fb3nOEUw9IO9j3yrqRxJnpmKvqATi99k1q9hzxybzQnB0YHUHq1nsIHdSXqp17Ofdgh08eeFXfJ2ajTR+KtdnI8bs3Y/Di+TfPLyQwWE3Vnhzy738dgD5ZM9DdMsnh+dSTb1Lth2eAlMfmEJ0+FFuzifylL3is67CBqfTdtEjaXXxPDoUPSnWdvOqPxE65Emw2zFV15C15EXNZjd+e5WLLxe392O3ETJTOccGyF2k6dqo994pUx07uNXtzOPOgtIt8rwdvJSZjODaTBeOZMgqWvYC1vskvLkDK43OIsfepPC99KmxgKv3sfapmzxEKW/vUqj8SmzkC0WbDXFlP/pIXMfnhWc7rEbk8R44fQq9H5kJgAJVvZVO6+b12nlM2LSV0YBqWmgZO3bXRkUfStq0izJ5Hih54pdN+FV1ciaJYBaR7+PshYK6Hv28HtvsqV547LgEBJD8+n/xbH+GHiYuIuWEMwf2SXUK0f8zAUmvgv2PupPzVD0i6T5pSY6mup2DOE/yQsYTTy5+jz3PLOsXt+dh8Tt32CMcnLSTm+rFo3Lixf8jAWmfgh3HzqfjTB+hWS9zmE2c4cd1yTly7lILb1tDzyQUQ2InqCwhA/+idFM5aQ97khURdPxZNX1d2zI2TsdYZODlhPpV/ep/E1bMcr5nOlJL/2yXk/3ZJpwYtBAQQ/+BCzs97gNPXzSPyt+NRp/VyCYmcPgVrnYHTmXOo/b+/0yNrDgBRM6YCcOaGuzh3+730uOeOzm2fHhCA/pG7OD37YfKnLCDqunGePdc3kjdxHlWvvU/iPa6eC353NwW/u7tTgxa5PSeuWcDZuQ9RMPVOIn83DrWb5+jpU7DWGyiYNJfq1/9O/EqJTWAA+o0rKX3oRU5dexdFt9yDaPFz9+GAAHRr7qJozsPkT7mLqOvGtufOkDznT7yDqtf/Qfw9sx3cpGeyKHlwM6emLuDMzau7P9euuPTBhKXo+Oqqpfw36xUuW98uFwJw2frb+W/WK3x11VLCUnTETRwMwOXPzCfv8bf4z/hVlP3rIH0WXtetuQAx6UMISdVx6OrF5GW9TN+n5nmM6/vUHeRlbeXQ1YsJSdURM3GI47Xz2z7iyKSVHJm00q9By4/J2aLRRPHGv3L+8e1+e3RXbPoQQlJ07L9qMSezttJ//R0e4/qvv4OTWVvZf9ViQlJ0xNrrG+Dc1n9yKH0lh9JX+j1oiZ44lOBUHUeuWUTBypdIXee5rlPXzaNg5cscuWYRwak6ou11XbzlfXLTl5ObkUV19mGSl8/w27NcbDm5ISk6jo5aSOGql0ld65mbsm4+p1a9xNFRCwlJ0RE9QeLW7csld8JSjk1aTsupYpIWT/OLC219KufqxeRnvUyalz6V9tQdFGRtJcfep1o9n9/yPkcnriB30kpq/PUs5/WIjJ57PT6fk//7KN9PWEzsDWMI7tfTJSTujxlY6gx8N/ouyl75gJ733QrY88iGNzn32PZO+VT0y5MsA5ewwf0wni7FVFSGaLZQ88GXRE2+0iUmevJIqt/ZC0DNR18TMWogAM3fF2Iuqwag5UQRARoVgtq/G0ehg/thPF2C6ayd++GXRGWMdImJyhhJ9bsSt/ZfXxMxahAAYosJrNLNswCNutPLQYQO6ofpTAlmO7vuw31EurEjM0ZS++4eAOo+/prwawZ1DuJBwQMvwVxUgvlcKZgt1P/rC8ImXu0SEz7xaurf3w1AwydfEnqV9GGvSetF0zdHAbBW12GrN0h3IvxUyKD+GJ09/3MfERlXucRETLqKGofnrwj7uXse2B/TmWLMZ+3sj/YRke7GnnQVde9J7PpdXxF6teQ5bPRQjCcKMR6XvvWy1jaAnyuzhAxy5db9cx8Rk9zreiR170l1Xf/xV4TZueFjhtJy/PTPituqHpnDKd65D4C6w/kERYaijo92iVHHRxMUHkLdoTwAinfuo8dUaan4sL46av7zAwBVXxwj4beueai7cQG0U0ZQvuNzABpy8giKDEXlxlbFRxMYHkrD4ZMAlO/4HG3mCL8Z7voxOdvWbKTx4A/YjKYL5sdljqBsp7Q3Wf3hPIIiw7zWd/0hyXPZzi+Im+p/vXpSbOYIKuxcQ47E9VjXEaEY7HVdsfMLYjMlrvNdtMBQTac+N+Riy8WNmXIlFe98bueeJDAqDFW860bbqvgYAiNC2rjvfE5MpvQ5WvdFruMzuuHwSdQ6rX9gINapTxkuoE85ew4I1SD6YVrO6xG5PIe1erbnker3vyJ6sqvn6MlXUrXzMwBqPvo3EaPb8ojh4A/YjJ27I9+dZUPs1v+6qzocuAiCMNDpZ5UgCA8IgvCBIAhPCoIQeqFQVaIWU3Hb3H5zSRWqRK1bTGxbjNWGtaGRwJgIl5joa6+h6btCRJN/S+KqErWYS5y5lR64WsxeuKGD+3NJ9otc8snznLt/iyNx+KMgd3apB88JWkwl7uxIANTJCfT95yZS/raW0BGX+c+N12IprXD8bimrRJXgyg1K0GIpqXDhBkRHYjx+ivCJV0NgAEFJCWgu70dQYg+/2VJ9O7FL2rNVCU4xVhu2hiYXz2kfPkfKW2sJHXH5z8JzUKIWi8t5riTIA9tc6uTZIHlWpySBCMmvPUbKP54n9o7p/nMTXNuXpdSD50TPda3ukwSiSK/XHyXl/efQzvP/m0q5uK0K1sXScr7K8XtLSTXButj2MSXVbTHFbTGG4+fokSmt9J543UiCk/y72JGLC6DWaTEWt7FNJdVo3C7SNDotxpK2GGNJtcuFnH5OJkP3Pk2/ZxcQFBXmk/lT5ewLlUYXi/G8s58qNG71rdHFunoudo1JmpPJ8M82csmmu/zyDKBOjMXo5NtYUtXuglit02Iqdj02dWIbt9fqmxl2aCs9fj+Wog1/84srJ1tOrnMbMxW7lumIcTrHnmIA4m+aSO3eHL+4rX6Mxa79xVOfcmG79aleq29i+OGX6TFtDEXr3/bJlPN6BOTxrNbFtl3jAKbSKtQ6T+fYyXN9E0E/UR5R9MuQrzsu251+Xgf0RdooJgR42dubBEGYJwjCIUEQDr1nOO0hwMOb2m344iHIKSS4fzJJ991K0b2deGzKY5luXE+PBdlDmo6e5ETGIk5ev4L4BdMRNJ3YZ8FDuaJfbBFLRTXHR80h/3dLKXn8VZI3ZREQHtK1XETq3vsES1kFvXa+QPy9d9Jy9L+I1s49yuMB7sb2HGOpqObE6NkUXLeEkideJfnZn4tn323M41rloogQGEjIsMsoXrGB039cSUTG1Y67Mb6x/jzO5oUbFEjo8Ms4v3wjp/+wioiMq/2/8yUXtwO1P9cegwD4bunLJM+ewlWfPklgeAg2P78EkZPruen6zy7Z/gkHRy4iJz0LU1kNKWv8WNn+J8jZP06e+rR7iPd+f/7Pn/LNyMUcmrgSU1ktaY/c6h/VS1/t+Mhw8V207k0OD59PxXv70M2e6hdXTrZsnv3gesvbztLfPQ3RYqPyvX3+ccFLivK/TwEUrXuLQ8PupOLdL9HNybxQqFtIF12PeMVffM/+9WM/ilb0q5GvZ6ycW1A6MEIURbMgCPuAXG9vcl77OSf5hnZNzlxShVof5/hdpdM6Hv9yxJRKMebSKggMIDAiTHqUBOlbiNRX7uX00k2YzpT6sOBcZiUqnTM3rj23pBKVF26rjPnnsDW3ENy/N83H8v1iW0rc2IlaLO08V6LWxWHxwLaapP9bvivAVFSKJiXJL7alrNLljkFQQhyWcleupbSSIF0PLGWVDq7Nzq1Y1zaRNvnNZzCfKfbLr+SnCpXOia2Lw1ze/jyrdD0cngMiQr16Vqck0dLNPUvlOp/n9mxzaSWqRCfP4ZJnS2klTQePYa2RJhA3fnGI4Mv70vQfr13NhevcvoIS4zCXVXmIaV/X5tJKmg585+AavjhE8OVpNP67e3KTZ08m6ZaJANQfLXC5WxGsi8VY6jpB1PlOB0CwPpYWe0xTfjE5f3gSgNBUHT0yhuBNcnEBdLMzSZwpzXFsOFqARt/GVutiMZa6tjHpbkNbjEYXi8keY66sc/y99K+7ufyNeztkw4/P2Rci/ewp6G+RtgioP5qPJsnZj9bhp1XtPOu1mOz1ba5o81zyl91c8Rfvq20mzsokYabENeTmo9HH0erCI7ekCrXe7djc6gag8u9fcekb93F2o/dvpuViy8VNmJVJ/MwMiXs036WNqfXadpO9TW53f9RO5xggbsZ4YiYN54c/POyR5+J5diYJ9j5lsPepNs+xHtuXC9tDjOT5Sy79y32c3bCj3WvOkuN6RG7P0vlzOseJWsxu5bXGmEvsniNDf1Qe6c5SxmMXJl93XKIEQfi9IAjTAI0oimYAURqWX3CdN+bmoemjQ50cj6AKIub6MdRlH3CJqc0+QOx06SIh5rejaPhaWtkpMDKMtD8/SPG6N2g8dLxT3KbcPDQpetTJCRL3ujHUZ+93ianffYDYaRI3+tpRNPxb4qqTExyT31RJPQhOTcJ0rsx/9rd5aProUfWU2FHXjaV+t6vn+t37iZ4mJZWoqaMw/MfuOTYSAuzs5AQ0ffSYivwbsLUcO4Gqt56gpARQBRF57TgaP3NdfcTw2TdE3iB9aEVMGUPTN9KFoxCsQQjRABB6zRBEqxVTQZHfnpu/Penq+XdjadjtWt8Ne/YT4/A8mkYvntV99Jh/Dp6PnURt94wqiMjfjqVhjxv7/7N33vFRVXn/f9+ZzEx6m7RJaEkoq/RuAwKhBHfVRwXcFX8KAtJEWihiL8gC1lVB0F3cx+exgGVdH6Wjoq50iLFACp30SW9T7++PO5nMJDPJBIw36v3sy9eSuefe9/2eer/3nvM9ew8SdpvEDk27gdoDks3VXx3Dv1cigr8O1CoCh/bBnOMbu+67LLTdEpzcsD+NpHpv87wOu03K61CXvK7efwxdr26N3GF9MWVf6LDcC1t2cSB1JQdSV1K0/Qjxk0cCEDa4O9aqWsxF5W7pzUXlWKvrCRvcHYD4ySMp3nEEAG2UNC0RQSBp8a1c+OeeDscFyN+yw7mY3rjjEDFTUgAIGdQDW1UtliZsS1E5tpo6QgZJ67NipqRg3HkYwG0eu37icGpPtp7nV9JnX66IaV4EAAAgAElEQVTytux0LqYv2X6Y2MmjAAgd3KOF/K4jdLBkc+zkUZTskGx2XQ8TdeMwalqwueDNHWSMS5cWl28/RLSDGzxI4nrKa3t1HcGOvI6ePIpSB9c/0eBMFzF+CHU5l1q0WS62XNzCN3eQOW4pmeOWUrbjENGTUhzcntgqa7EUuTsulqIybNX1BA/qKXEnpVC2U6qHYSkDiZ9/K6emrcFe1/p6qoItO8gYu4yMscsodWlTLdlsq2m0OWZKCqU7G2xu3Dg8csLQVssZ5Hkekdvmmoxs/BMb+5HIW26g3EM/op88GoCIP15H1Te+R41V9PuQ0OzToOtBQdjS5KeVoigWCoIQB/yvKIrNwpw1lacvLgChowfT6fEZCGoVxvf2UvDyNgxL76T2uxwqdh9C0Gmk0Jp9krCVV3Fm/rOYzxcS98BkYudPwnSm8S14ztTHsRor3K6vUnm2K2T0YBIenYmgVlG6dQ+Fr2wjbonErdwjcbu+sISA3klYy6uk8IMXCom4NYWYeZPAYkUURQpfepeKXQc9MtTe2CmDMTzqCIe8bQ/Fr24lZvFU6jKzqdpzCEGrofMLS/C/OglbRTXnF6zDcqGQ0LTriF08VZqyZLNT+OL/UrX3cLPr6/w9L1oLGjmU6Adng0pF5Ye7KN30LvoF/4/677Op+fwAglZD3Nrl6K5Kxl5RRf7SNVguFuAXH0unN1Yj2u1Yi4wUPvwC1rzmG5+aTd4/3AWnDMHwyCwElYqybbsp3rCVmEUOm/dKNnd6fqnT5gsPrHXaHLNoKqLNDjYbRS++TdW+Q82ur9V5nmLT3jbbrN59/qBRQ4h9yBEO+f1dGDe+R9TCu6jPzKZ630EErYb4Z9Pxv1oKe31p8VppcTsQevNoouZMQRRFar48QtG6fzS7vmj3PD0rOGUIsQ/fh6BSUf7+bko2vEf0oruoy8ymeq/ETXguHf/eUpu6uHCdkxt2y2j0cyYDItVfHKFobdOm712/BPditfc5zn9YM52oMQOw1Zn4YeFrVGZIIVSv2ftXDqRKb9ZD+yfR529zUflrKdl7gpOrJE6XWRPpPH08AEWfHSL76Xd8tru9uQGC9+ljyWtmEjF6APY6E1mLNlCdkQvAwD3rOT52GQDB/ZOd4ZDL9h0nd5UUNrbnywsI7tMNRKi/UET2sk1uDyxBGs/9yOX22QC9/7MZdUgggsYPW2UNOVMfp76Jk1pp1raY3z3WzCByzABsdWZOLXyVKkd+D9m7niOpks0h/aVwyFII6BNkO2z+wysNNovUXygmK32Tm+OjVXlfH5D4zEwiRkthr3MWv0qNI6/7736WjHHpUp71T6aHIzRw2b7jnHlI2qKg1xvLCEiOR7SLmC4Wc3rFJo9vrDsauz25KsH7M0e3Z2YRnjIQuyMccs13Erfv7ufIHLdU4vZLdoZDLv/8GGcd3AHfvIqg02Atk97OVx/N4szKTW7Xt9m999tJa2YS7mhTOS5tqv+e9WS4tKnujjZVvu84px31q9cb6QR0jweHzbnLN7vZHKTz7Ei19/NIjanlNtWeNuv8PPdfYWMG0/nxe0GlxvjeHvJffp/49L9Qk5FDxe7DCDoNiS8tItDRj+TOe87Zj/T9djPqkABnP5J15+PUZ19sxhhy8V9tCAUqnxZ3+3OH/ujywtl3O2Q+tui4AAiCMBywi6J4WBCEq4E04KQoip/5AvDmuLS3vDkuv4S8OS7tLW+OS3urJcelveXNcWlvteS4tLe8OS6/ZbXkuPxW1ZLj0p7y5ri0t1pzXNpTLTkuin5eteS4tLdaclzaU94cl/ZWa45Le8qb4/JLSHFcfh51VMelxSdOQRAeAyYCfoIg7AaGAV8CKwVBGCiK4upf4B4VKVKkSJEiRYoUKfrNSHldcnlq7VX5JGAAoAMKgE6iKFYKgrAeOAgojosiRYoUKVKkSJEiRYraXa19N7WKomgTRbEWyBVFsRJAFMU6FGdRkSJFihQpUqRIkSJFv5Ba++JiFgQh0OG4DG74URCEMHx0XIxm/yu4vcuXTvj9+VX2Wnny2tKq/9t+Sgosbz1RO0ijlW/33t/jGpeAGnnmS2tkXPuglmktgNmmloWrkXHtQ1yUPOFWBRnXYtos8vTbKrV8NldWyjNG1lvkWQeqD62VhQuQXRkmG/vXIlEJiHxZaq01jRRF0QQgiqLrCK4BfNi1TJEiRYoUKVKkSJEiRYquXC06Lg1Oi4ffS4CSdrkjRYoUKVKkSJEiRYoUKWoi+eLYKlKkSJEiRYoUKVL0O9Tvb0HDzyNZHZdeq6cRlSptcvXDAxupyjzTLE1Iv0R6/20ean8tJXuPc+qhNwEI7t2Vq9bPQq3TIFpt/LTy71Qez/WJm/z0dCJTB2GrM5G18FWqPXCD+yXR66WGjcyOkfuwtHFc1/TJxE0di8VYCcCZNW9Ttve4zzbLxe6+ejp6B/fkA965f/jbfNT+Wox7j5HzUONmgAkz0ki4dyKi1YZxzzFOP/U/Ptt8ReV8dVeuWj8TdZA/9ReKyZz7MrbqulaZAdcPIWrlHAS1msoPtlP+961ux/0H9yFqxRy0PZMoXPYMNbu/dh4zvLYaXb8/UH/8BwrmP+qznU72dUOIXDEPQaWi6qPtVPzjPXf2oL5ELp+LtkcSRStWU7vnKwC0vZLRP/QAquBAsNkpf+NtanZ+6Tv3+iHoV8xFUKuo/HAHFX9vwh3cF/1yyeai5c9Qs/sr57G4javR9buK+uPfU3h/22yWi9ugpKfvJTJ1IPY6M6cWvkKNl7rd09mmjnP6YfeNPRPm3kzSY3fz7dXTsZb6tt6h61MziBgjtancxa9Qm3m6WZqgvknOzfLK9h3j3CPS5m1dHrmbiHFDsJutmM4Vkrv4ZWyVvs9H7/LkDMLGDMZeZ+LM4pep/b45O7BvEokvPIDKX0vFvqOcf1Riq8ODSd64FF3nGEwXisid8yy2ipoOb3Pi0/cS4SjnbC/lHNQviR4Nm27uPc6ZJuUcP/dmEh+7m4M+lrNsbUqmPgQg8IYhRD04B9RqKt/fTvkbHvrOB+eg65lEQfoz1Oxq7DtDbhlLxJw7ASh77W2qPt7ju80y9iMhowaR8NhMBLUa47u7KNr4gdtxQetHl+cXE9i3O9aySmkjyItFBPbvQec18x2JBApefIeKnQfaxJarTQWNGExMw0bJ23ZSunmbu80aPwzr0/Hv3R1beRV5i9ZguVRE6E0pRM683ZlO1yuRs7c+gOmn5vftTb2fvofYVGkz2RMLN1KRebZZmj+snEKnySPRhAexPXm68/ek2TfSZepoRKsdk7GSjMWbqLuoTAD6PUm2VdVRqQMITIzjm2sW8lP661y1bobHdFetm8lP6Zv55pqFBCbGoR8zAICej07l9LPvcyB1BbnrttLjkak+cSNSBxKQZODwtQvITt9E97WzPKbrvnYWWembOHztAgKSDEQ4uACXNv8fx8Yu49jYZW1yWuRiR6YOJCDRwMFrFpCVvome6zxze66TuAevWUBAooFIBzf8+t5EpQ3l8OilHB61hAsb/+2zzVdazlc/P5ucp9/mQMoyij47RLf5N7UOVamIfng++XMf5vzNswi+cTSapC5uSaz5xRQ9/BzVn33e7PTyLdsoenCdzzY2ZetXLaBw3iou3jqToDQP7IIiih9ZT/X2fW6/2+vrKX54HZdum0XBvFVELpuLKiTIZ27UQ/dTMO8hLtwyi+CJKR5sLqL4kWep/mxfs9PL39xG8arLsFkurkMNberItQvITn+N7mvv85iu+9pZZKdv4oizTQ10HtPG64kY2Y/6i8U+c8PHDCIg0cCJ6+dzZvlrJK3xzE3862xOL9/IievnE5BoIHy0xK3Yn0HG6EVkjl1C/ek8Ehbc7vF8TwobMwhdYjyZN8zj7IqNdF0z22O6rmvmcHbFRjJvmIcuMZ6w0YMAMMy/jcqvM8m8YT6VX2dimH9bh7e5oZyPXbuAnPTXSPZSzslrZ5GbvoljjnIOb1LO4W0pZxnblCx9iIMd/fB88mY/zPmbZhFy42g0yR76zlXPUfWpe9+pCgshct5dXPzzQi7e8QCR8+5CFRrsM1e2fkSlotNTszl9zxOcHDufiJtHouvR2S1J5B3jsFVU89Oo2RT//d8YVkpLfOtOnePUTUs4deMicu95nE7PzAO1749VsrUplYrYx+ZxcdajnL5xDqF/GoU22d3msMkTsFVUc3rcTErf/IjoZfcCUPnJF5y9ZQFnb1lA/rLnsFwqbJPTEpM6gOCkOPZdu5iM9Nfpu9bzM0HBrmN8NfHhZr9XfH+WryY8xJdjVpD/fwe56pE7fWYr+m1INsclOm0o+dv2A1BxNBu/0CC0MeFuabQx4fgFB1BxJBuA/G37iZk4FABRBL+QAAD8QgMxFZb5xI2aMJTCrdIbqKpjLXOrjmYBULj1S6LShl2mpfKzo9KGUrhN4la2kteVRxzcbV8SNVHixt8znvMv/wvRLEV2spRU+sy+0nIO6m6g7NufADB+mUnMH4e3ytT17YXlfB7WiwVgtVK9/QuCxlzrlsaaV4g56wyivfnH2rqDJ7DXtv5VxyO7Ty8sF/KwXpLYNTu+IDDlumZsS/YZsLtHFLGeu4T1/CUAbMVGbKXlqCLc88ort4nNNdu/JGh0c64564zUeJqo/uAJ7DVtj0AjF7dB+glDKdr6BdDQpgLRNKlfmphw1MGBzjZVtPUL9GlDnceTn5zGmafe8nh/3hQxYRjF70vc6mNZqMOC0MRENOFGoA4JoNrBLX7/CyLSpPpb8WUG2KS6V3U0C61B7zM7fMIwjO9LD401rbBrjp4CwPj+54Q7+pHwCcMwbpPON277nPC01tuU3DZHupRz9WWWc+KT0zjbhnKWrU3J1IcA+LvabJH6zmAvfSdN+s7A6wdT++0x7BVV2Curqf32GIE3DPHNZhn7kcABPTCdzcd8oRDRYqXsk68IG+feJsLGDaf0A8lhKv/sG0Ku7w+AWG921mmVTktbg0TJ1ab8+/XEfC4PywWpnCs/3U/wWPdyDk69hoqPpC9mVTu+JvDa/s2uE/KnUVT+X9u+6MVNGMyFrdLXsvJjOWhCA9HFNK+j5cdyMBU1jxpq/OZHbHVmAMqO5hBgiGwTvyPJjtih/+uoks1x0RkiqL9kdP5dn2/Ev0kF9DdEUp9f2pgmrxSdQWrUWY/8kx6P3sWIY6/S47H/R87qd3ziag2RmPIauaZ8I9omXK0hElO+9zTx96YxaN+z9HxhLn5hvr/NkoutM0RiuuR+TV0Trq4pN68xTWByPGHDr2LQ9mcY8NEThAxI9okrXffKyrn65AWi06TBL/ama/BPaL1j9ovRYy1ofKtqLSzBLybK53u+EqljorC5sG1FJfjFtp2t7dMLQaPBeiHPp/R+MVFNbC5GHev7g+HlSi5ug7QGvVubMueXomsyeOsM+iZtqtQ5wEeOH4Ipv5SaH8+1jRsXiTmvcXqCOc+INi6yeRoXrqc0ADF/GUP5vmNtYOsxu9hsyTeiaXJdTVN2vhFtnGSzJiocS5H0osdSVIZG71vYUlltblLOJi/l7G6zezmb80upbUM5y1W35epDANSxeiyuNheUoPax7/SLjcKa736ur/ctZz+iidNjyW+s15b8EjRx+uZpGuq+zY6tqgZ1RAgAgQN60mv3K/Ta+TcuPrTB6Uj4IrnalCZWj7WgkWstKEHTJL81sfrG8rTZsVfVoo4IdUsTeuPINjsu/oZI6l3acl1+abNnAl/V5c4UivZlXNa5in69atFxEQThfkEQohz/7i4Iwn5BEMoFQTgoCELfK0M334tCbPomRfCwX4UjSadp48h69J98NWg+WY/+k6tfmOMj1vs1W04jJcp7cxeHhi/gWOoyzIXlJD1+t29cWdme8rp1bkN5CH4q/MKDODZxFblPvsXVry/xkeuN7Xs5/7DoNTpPH8/wXWvwCw7AbvZhP48W8rDd1UI++ip1VCTRq1dQ8uizvt+3p61dfgmb5eI24H3he0mjCtDSedHtnFv3nocElwH2pV43afDxD9yOaLVT8uH+NrA9Xdb9ukJ7tIEOZnPzfsQTWirnTotu53xby1m2NiVTH+KF7fNnhCvJL1n7kcus144ktSeyODXufrJuXkrMvEkIOk0b0DK1qcvkutZD/369sNeZMGe37aXPzzU+J9x+A+H9k8jd8Embz1X061Zri/PniqL4iuPfLwEviKL4kSAIKcBrwPWeThIE4T7gPoCFIYP5Y4D0hr7T9PF0uisVgIoTuW5vz/0NekwF7tO96vPc3877x0c60ximjHIu4C789wGuft7zPG8Aw/QJGKaOBaDqRA66+EauzqDHXFDqlt6cZ3R7myelcbyhLKlw/p7/v3vo89ZKr1w52fHTJxB/l8StPJGDLqFlrqkpN76Ra8orpeTTg5INx3PAbkejD3UGCWiqn7Oca3PyOHbHMwAEJhmIGjeQ1mQtLMEvLtr5t19sFNZiYwtn/HyyFRajdmGrY6KwFfnOFoICiX3lacpeeRNT5k8+n9fc5mhsRaUtnPHzSA6uYXoacVOl+lV1ItetTWkNkZhaq9uGSMwFpfh3jcO/SwyD9j3r+F3PwF3rODHxQSzFzacoxE5LI2bqOACqT+SgjW98m6yN12NuMl3VnG90m7qhdWlTAFGTU4gYO4Sf7nisVZtj7plItINdcyIHrYvNGoMeS2tsgx5zoZQvlpJyNDER0teWmAgsxgq8SU6b46anEeso52pHOTcsp28oQ1eZ8pra3FjOui4xDHAp5wG71pHhpZwbJFebkqsPAbAVlKBxtTnOd7a1oISAYf3czq079J1v58qU1wCWghI0hsZ6rTFEYSl0Z1vyS9DER2EpMIJahTokCFu5e3AHU85F7HX1+PfsSl1mjleenG3K1Wa/uEauX1wUlib5bSkowc8QjbVQslkVEojdxebQP46k6tMvfOJ1mz6OLlPHAFB+4jT+Lv1XgCGS+gLfpvo3KGpEH3os/C/+c9uTvr3M7KDquJOxOrZamyrm6tjEiKL4EYAoil8AId5OEkVxsyiKQ0RRHNLgtABc3LKLA6krOJC6guLthzFMHglA2OAeWKtqMTeZz2guKsdaXU/Y4B4AGCaPpHjHYQBMBWVEXHc1AJEj+lB7usCrEflbdjoXtBt3HCZ2yigAQgZ559pq6ggZJHFjp4yiZKfEdV2fETVxGDUnL3jlysnO27KTI6nLOJK6jJLth4mdLHFDW8zrOkIdeR07eRQljrwu2X6I8BukD2wBSQYEjZ9XpwV+3nLWRDk+TQsCiYtv4+I/d3vlNsj0/Sk0XRLwS4gFPz+CJ6ZQ83nbIr1crkw/NLDjwM+PoLQUar/81reT/fyIfeFxqj/ZTe3uNryNxmFzVxfuxFHUfOEj9wokBzd/yw6Oj13G8bHLMO44RMyUFEBqU7aqWixN6pelSZuKmZKCcedhak+e52CfGRweOo/DQ+dhyjdyfPxyrw+zhW/uIHPcUjLHLaVsxyGiJ0nc4EE9sVXWOqdfNXLLsFXXEzyoJwDRk1Io23kIgLCUgcTPv5VT09Zgd8zXbklF/9zOD+OX8MP4JZTtPIh+0mgAglpg26vrCHKw9ZNGU+5gl+86jH6ydL5+cuPvHc3mgi07yBi7jIyxyyh1KedgR9/prZyDXcq51FHOh/vM4OjQeRx1lPOJFsq5QbK1KZn6EIB6p82xoGlb31n7zVECrxuMKjQYVWgwgdcNpvaboz6dK1deA9RmZKNLjEfbORZB40fETSOo3H3QLU3lnkNE3i49eIffeD1V/5EcMm3nWOdifE1CNP5JCZgvFrbIk7NNNag+Mwttt3g0naRyDv3jSKr3updz9b6DhN0qvfwMSbuB2m9dnFBBIGTiCCo/9a2Ond2ym/1jH2T/2Acp2HGEzlNGABA+qDuWqlqPa1m8KbRPN/qtn8nhe57F3Ib1top+OxJa+gQtCMJqIAF4EvgzUAt8CKQCt4ui+KfWALtj7/AK+MOae9GP6Y+tzsyPCzdSmSFFprhm71oOpK4AILR/Er3/Ng+Vv4aSvSc4tUoK0Rs+rBe9np6G4KfGbjLz04q/U/VdY3hMneB9nmn3NTOIGD1ACp+66FWqHdxBe9ZzbOwyAIL7u4Qk3neC3FVS+MFeLy8guE83RFHEdKGY7GWbmj2It6T2ZNtbcN97rJlB5Bgp/OCpha9S5eAO2bueI6kSN6S/FA5ZChl7gmwHV9D48YcX5xLcpxt2s5XcJ96i/Ovvnde2tOL/Xkk5d541kc7TxwNQ9Nkhcp52X8uUFO057wNHDEW/Yo4UWvOjXZRvfoeI+Xdj+iGL2i8OoOvTk7gXH0UVGoJoNmMrKePCf0kRXeL/+RzaxE4IgQHYyyspevQF6v7jPgCr1N7rV8ANw9AvnwsqFVX/2knFG28TPu8ezD9kUfvlt2h79yT2hcdRhQYjmizYjKVcum0WQX9MJfqJdMy5jZ/eSx5dj/mUe5hv0e5p6gAEjBiKfrkUTrTqo52Uv97E5t49iX3pMVQhDTaXcvFWyWbDm8+hTews2VxRSfGjzzez2au9vwD3UnFos98alLxmpqNNmchatIHqDCm/Bu5Zz3Fnm0p2hkMu23fc2aZcNfTwBo5PWOEWJlej8l7O3Z6ZRXjKQOyOMKY130ncvrufI3PcUgCC+iU7w5iWf36Msw+9AcCAb15F0Gmwlkms6qNZnFm5ye36asF7g+6y+j7CHOwzS16m1sHuvet5fhgvTeUM7JfcGA7582Ocf/h16boRIXR/LR1tQhTmSyXkzF6PrbzaeW2b6Ll+tbfNNnvL/UjSmpmEO8o5x6Wc++9ZT4ZLOXd3lHP5vuOc9lDOgw9vIKNJORuiPT8AtXfdFlSey7i9+xAAm8VzfgeOHCqFkldJfWfZpneIvP9u6n/IovZzqe80/K2x77SWlHHhZsnmkNvGE3HfXwAo2/QOVR/tanZ9ldqLzb9AP1JZ6e+RHTJ6MAmPzkRQqyjduofCV7YRt+ROar/LoXLPIQSdhq4vLCGgdxLW8iopHPKFQiJuTSFm3iSwWBFFkcKX3qVi18Fm16+3eJ/c0p5tKjy43is3aNQQYlfNBrWKivd3YXztPaIeuIv677Op3ncQQauRwiFfnYytooq8xWulxfxA4LC+RKdP59wU79PGsyu9r53rs2Y6MaP7Y6szcWLRJioczwQj96xh/9gHAbjqkTtJuPU6/OMiqC8o4/zbn5P17Adcs3UVoVd1od7xZarukpHD9zzrdv2bCt7x3ol1IM3tNqVDf3TZeHZrh8zHFh0XAEEQpgFzgWRAB1wA/gWsFUXR+xwDh1pyXNpTLTkuv1W15Li0p1pzXNpT3hyX9lZLjkt7y5vj8ltWS45Le6olx6W91ZLj0p5qyXFpV24rjkt7ypvj0t7y5rj8EvLmuLS3vDkuv4S8OS7trZYcl/ZUS45Le6slx6W9pTguP486quPiS2v6EbhfFMXDgiD0BtKAn3xxWhQpUqRIkSJFihQpUuSujhxyuCOrRcdFEITHgImAnyAIu4FhwJfASkEQBoqiuPoXuEdFihQpUqRIkSJFihT9ztXaF5dJwACkKWIFQCdRFCsFQVgPHAQUx0WRIkWKFClSpEiRIkXtrtYcF6soijagVhCEXFEUKwFEUawTBN8WkUTp5JljqZZxbrrHsOu/gGpNbYgf/zMqzM/3aCY/tyxmtSxcQSXffHy55qb7aW2ycAEsojw222zyTfHVyLTGxSRTXqtknDZRbAyWhauSqYxBvrVMOj/5+pEKk1Y2thwyVgbKxo5WmWRj/1r0+1uJ/fOotRHKLAhCQ80f3PCjIAhhKHmuSJEiRYoUKVKkSJGiX0itfXEZKYqiCUAURVdHRQPc0253pUiRIkWKFClSpEiRIkUuatFxaXBaPPxeApS0yx0pUqRIkSJFihQpUvQblqhEFbssyRNcHAgZNZBOj89CUKswvrubwg0fuB0XtH50fWExgX2TsZZVcXb+eswXiwgZ0Z/4lXcjaPwQLVYurX6T6v9k+swNHjWIhEdngVpF6Xu7Kd74fjNu5+eXENAnGVt5FefuX4flYpHzuCY+mp67X6XwxXcoef2jNtkcPHIQ8Y/NApWKsvd2U/xac3an5xrZ5+9fh+WSO7vHrlcpeqnt7K5PzSB8zCDnJle1maebpQnsm9S4ydW+Y5x7RNq8LfJP15Kw9A4CenTihxtXODfI8kVylXPQiMHEPDQbQa2ifNtOSjdvc+dq/KTNtXp3x1ZeRd6iNVguFRF6UwqRM293ptP1SuTsrQ9g+ql5fnlT4A2DiX1oDqhUVLy/g9LXm7I1xK1din/vHtjKK8lbsgbrpSLwUxP39CL8r04GtZrKj/dSunmr7zaPHEzcI/chqFWUvbcL46YmXK0f8c8uJaBPd2xlVVx84K/O+qXr1Q3D0/ejCg4EUeTMfy1CNFt8tjdm1VynvWVvuN+zZG86uqsle/OXrMGaVwgaP2IffwD/Pj3ALlL0zGvUHf7OC8W7eqyejj5V2sDtxwc2UJ15plmakH6JXOXYXNW49zjZD21xHus0I42Ee9MQrTaMe46R+9T/+sTtvno6+tRB2OpMnHzgVY/c4H7Spq5qfy3GvcfIceEmzEgj4d6JTu7pp/7HZ5sTn76XiNSB2OvMZC98hRoP7KB+SfRo2HRz73HOPPwPALos/zORaUMR7XYsJZXkLHwFc2FZs/M9Sa68BvnyW66+E6DLkzMIGzNY2mh08cvUfu+Z7dxodN9Rzj8qsdXhwSRvXIqucwymC0XkznkWW0WNzzZHjJHy2pvNQS42l7nY3OWRu4kYNwS72YrpXCG5i1/GVlnbKlOusaJBctWvX2s/0qD4uTeT+NjdHLx6utumri1Jjvql6LcjmVZhquj89Gxy73mCn1LvJ+LmEfj36OyWRH/HOGwV1fw4cpy/uHAAACAASURBVA5Fb/yb+AelmWnW0kpy713NyfELObf4Jbq+uLhN3IQn53Bm2uNkjZtP+M0j0XV350ZOGY+toppTKbMp/vvHGFZOczse/8hMqr7wbUfxpux4Bzt7/HzCPLAjHOys0bMp+fvHxDVhGx6eSfWXbWeHjRmEf6KBjOvnc2b5aySuuc9jusS/zubM8o1kXD8f/0QDYaMHAlB78jzZM9dRdeDHtoFlLOfYx+ZxcdajnL5xDqF/GoU22Z0bNnkCtopqTo+bSembHxG97F4AKj/5grO3LODsLQvIX/YclkuFbXJaUKmIfXQ+F2c9wpk/zSbkjylok7u4syeNx15ZzZkJMyj757+IXiqxQ9JGIGg0nL15Huduf4DwO27ELyHGZ67h8bmcv/cxcibMJeymkWib1K9wh805Y2Zh3PIvYlZMlw6oVSQ8n07+I69yeuI8zt25EtHq4wJalYqYR+Zz6b6HOXvTfYR6sDd0ksQ9m3Yv5f/9EdHpkr1hkycCcO6WuVyc8SDRK2a1ObKFPnUggYlxHLjmAU6mb6bXupke0/VaN4uT6Zs4cM0DBCbGETlmgJQn1/cmKm0Ih0anc2jUUs5v/MQnbmTqQAISDRy8ZgFZ6ZvouW6Wx3Q9180iK30TB69ZQECioQl3KIdHL+XwqCVc2Phvn22OSB1IQJKBY9cuICf9NZLXem7PyWtnkZu+iWPXLiAgyUD4GKk9X9rwMSfGLCVj7DLKdh+l85LJPnHlymuQL79l6zsdbF1iPJk3zOPsio10XTPbY7qua+ZwdsVGMm+Yhy4xnrDRgwAwzL+Nyq8zybxhPpVfZ2KYf5tP3PAxgwhINHDCYXNSCzafXr6RE9fPJyDRQLjD5or9GWSMXkTm2CXUn84jYcHtHs93k1xjhUNy1a9fcz8CoI3XEz6yH/UXi33mylK/FP2mJIvjEjigB6azBZjPFyJarJR98hVh44e5pQkbPxzj+/sAKP/sG0Ku7wdA3Q9nsBaWAlCfdR6VToOg9e3DUeCAHpjP5WO+IHHLP9lP6PjhbmlCxw+n7IO9AFR89g3B1/V3OXYN5vMFmLLPt93m/hLb4mBXfLKf0HFN2OOGU97A3t6EPe4azBcKqM9qOztiwjBK3v8CgOpjWajDgtDERLil0cREoA4JoPpoFgAl739BRJp0f/U5l6jPzWszV65y9u/XE/O5PCwXCsBipfLT/QSPvdYtTXDqNVR8tAeAqh1fE3ht/2bXCfnTKCr/78s22ezfryeW83lYLkrsqs++JDj1mibsa6n4l4O98ysCr5UGIUQRVaA/qFUI/lpEiwV7tW9vkgL6u9tc8X/7CRnrzg0ZO5yKD6X6Vbn9a4IcNgePGET9ybOYTkpv22zlVWD3LfaGf79eWM7nO+2t/OxLgsY0yesx11L5sYu910j26pK7UHvghMQsrcBeWS19fWmDotKGULBtv2TT0Wz8QoPQxoS7pdHGhKMODqDySDYABdv2Ez1xKAAJ94zn3MsfI5qtAFhKfNtBPSptKIXbvmyV6xccQOURqU0VbvuSqIlS/Y+/ZzznX/5Xm7kAkROGUrT1CwCqj2XjFxqIpglbExOOOjiQKkd7Ltr6Bfo0yWZbdZ0znSpQ5/N0BbnyWmLLk99y9Z0A4ROGYXz/cwBqWmHXHD0FgPH9zwlPG9Z4/jbpfOO2zwlPcx9vvCliwjCK22hzsYvNFV9mgE3qP6qOZqE16FtlyjVWNEiu+vVr7kcAEp+cxtmn3gLR9ylPctSvjip7B/+vo0oWx0Ubp8ec17hExpxvRBPrXvk0cZFYGtLY7NiqalBHhLilCb/xOup+OONstK1JE6tvvCZg8cR1TePkhiIE6IieczuFL73jq5lu8ovTY8l3YRcY0cQ1Z5vzvbOLLpOtjYvE5JrfeUa0cZHN0pjzjS2maTtXvnK2FjRyrQUlHsvZml/s5NqralFHhLqlCb1xZJsdF7/YKCz5jW+frAUl+DVh+8XosbqUs72qFnV4KFU7v8ZeW0/yV2+TvO+/Kf3Hh9grqn3kutcvTzZLdbC5zdpuCSCKdNnyJIkfv4T+Pt/fYPnF6LEWuNhb6IHbJK9tVTWowkMxnTxN8JhrQa3CLyEWXe8e+MVF+8wG0Bkiqb/UaLcp34jOENksjcmlbtfnNaYJTDYQPvwPDN6+moEfPU7IgGSfuaZLjdf0hWty48YTNvwqBm1/hgEfPeEzF0Br0GPKc2WXomsyeOsMevf2nF/qNsB3WfkXhhx9jejbR3B+3Xs+ceXKa+d1ZchvufpO6bp6zC7lbMk3omlyXU1Tdr4RrWNc0USFYymSpgBaisrQ6MN85Ea699tXYHPMX8ZQvu+YD0x5xooGyVW/fs39SOT4IZjzS6n98ZzPTJCnfin6balFx0UQhA8FQbhLEISfN8i9p9kgTT12T1NGXJL49+xM/IN3c/7BDW3gerqmL1yRuMVTKfn7x9hrL3NfGg/XFX1kxy6eSsk/ZGBf6cKxX1k5u+aJf79e2OtMmLPb1il7VNNs9MRGJKBvL7DbyR05ldNjpxE5/TY0neJ8Y/g0xcpzvgh+agKHXM2lJc9y9o7lhIy7lqDrmn+B8pXra92q+HAn1sJiumx7mZgH51B/4kdEW1v3eLjcNu045KfCLzyYoxMfIufJt+jzuq/TTDzZ3TSJ97yRuEEcm7iK3Cff4urXl/jI9Yj2kOceznNJc/6v73Bk8ByKP/gKw71pVwJukqQ98toz+xfJb7n6TvCp/xR86evazL3Mcm5ic/wDtyNa7ZR8uN8HpqfL/QJjRQs38Mu0519nP6IK0NJp0e0+v/Rwv6YM9UvRb0qtfU8djvTF6G+CIOwB3gE+FUWxxR0HBUG4D7gP4KGIftwe3M3tuDnfiDY+yvm31qDHUlTqlsaSb0QTH4WlwAhqFeqQIGkKC6CJ05O4+UHOLX4R87kCH8x0XLOgBI0LV+OJ60jTlBs4oCdhN16H4cFpqEODEO0iosmM8b8/9YltzS9BY3Bhx+mdn7dd2VpDFFZP7InXEbfSd3bstDSip44DoOZEDrr4KBre3Wvj9ViaLMY15xvd3shq4/WYC3xbsOtNcpazX1wj1y8uymM5+xmisRZKXFVIIPbyxoWFoX8cSdWnX7TFXMDxxcHQ+NXALy4Ka5GxWRo/QxTWwhI3dsifUqj56ghYbdhKK6g79iP+fXpI07Ba4xa41y+/uCgshUYPaaKd9UsVEoitvApLQQm1h77HViZNM6j+8gj+vZOp+U+GT/a6fiXxi43C2iSvrc68LnGWcUNeF/91szNd57efx3Ku9Wk1CdMnEH9XKgBVJ3LxT4iiAmmqjM6gx9Sk3kpvKBvrtn+8HlNBqeNYKcWfHpSudTwX7HY0+hAsxuaLTOOnTyD+rrEAVJ7IQZfQeE2dQY+5wN3uplydS5sy5ZVS4uTmOLihWIyep3rETU8jdqpkc/WJXHTxehruUGeI9Mh2a88e0gCUfPQVV/3PKi6s9xwEQq68BvnyW86+M+aeiW5sbXzjdTUGH9gGPWbHuGIpKUcTEyF9bYmJwGKs8MqNnZZGjINbfSLHvd+O1zcL3tCazVGTU4gYO4Sf7njMJ7vlGCvkql+/hX7Ev2scui4xDNj3rPO+B+xaR8bEB7EUlzfjyl2/OqqUqGKXp9amihWJojgJ6Ap8AswCLgmCsEUQhPHeThJFcbMoikNEURzS1GkBqM3IRpdoQNs5BkHjR8RNI6jYfcgtTcXuQ+gnjQEg/MbrqfqPFG1IHRpE8puPkLf2LWqOnPTdUgdX2y0eTadYBI0f4TeNpLIJt3L3QSJulxp22I3XU+3g5k5ZyckbZnLyhpmU/OPfFL26zWenBaD2u2x0Luywm0ZSuacJe89BwhvYE6+n+luJfXrKSk6NmMmpERK7eEPr7MI3d/D9uKV8P24pZTsOETUpBYDgQT2xVdY6pxA0yFJUhq26nuBBPQGImpRC2c5DTS/bJslVzvWZWc5yRuNH6B9HUr33gFua6n0HCbtVGjxC0m6g9luXaFaCQMjEEVR+2vY3OfWZWWi6xqNJkNghN46iel9T9gHC/svBnjCC2gOSg2DNLybwGulLhxCgw7//HzCfvuATt+67LLTdEpw2h/1pJNV7D7qlqdp7kLDbpPoVOvEGahw2V+8/hq5XNwR/HahVBA7riynbN2595ik0XePxc9gbeuMoaj5vYu/nBwi9pbm9gr8OIUAHQOB1AxFtNsy5ra/hurRlJ4dTl3M4dTnF2w8RN3mkZNPgHtiqajEXuQ+c5qJybNV1hA6W1s/ETR5JyY4jABRvP0zEDX0ACEgyIGj8vD5I523ZyZHUZRxJXUbJ9sPETh7l5Fq9cK0u3NjJoyjZcRiAku2HCL+hbxOu9/npBVt2kDF2GRljl1G64xAxU1IACB4ksS1N2Jaicmw1dQQPktgxU1Io3Smx/RMbv+JFThhKXc4lr1y58hrky285+86if27nh/FL+GH8Esp2HkQ/aTQAQS2w7dV1BDnY+kmjKXewy3cdRj9ZOl8/ufF3bzZnjltKpsPm6DbaHO1ic1jKQOLn38qpaWuw17X4jtMpOcYKuerXb6EfqT15nsN9ZnB06DyODp2HKd/IifHLPTotIH/9UvTbktDs06DrQUE4JorioCa/RQJTgCmiKI5pDXC8yy0eAaGjB5Pw2Awp9OF7eyl8ZRtxS+6kNjOHyt2HEHQaur64mMDeSVjLqzh7/7OYzxcSu2AysfMnYTrT+GY2967HsTZ5m6RWeV5aFJIymHhHOOSyrXsoenUrsYunUpeZTeUeidv5+SUE9E7CVl7N+QXrMF8odLtG7KK/YKup9xqS2NusnZCUwRgedYRD3raH4le3EuNgV+05hKDV0PmFJfhfnYStQmJbmrBjFv4Fe61ndq1J4xkMdHtmFmEpUhjT04tfcYbl7LP7Ob4ftxSAoH7JJDWE9Pz8GOceegOAiLThdHt6Jn76UGyVNdT8cIZTdz7lvLbWz/v0nvYu5wB/zyF7g0YNIXbVbFCrqHh/F8bX3iPqgbuo/z6b6n0HEbQaKRzy1cnYKqrIW7xWWtgOBA7rS3T6dM5N8f7ZXVB5bzdBI4cSs+o+UKmp+GAXpZveRb/g/1H/fRY1nzvY65ahu0pi5y/5K5aLBQiB/hieWSJF5RIEKj7cRdk/Pmh2fZvF8/uG4JQhxD58H4JKRfn7uynZ8B7Ri+6iLjOb6r0SN+G5dPx7J2Err+LiwnVOm8NuGY1+zmRApPqLIxSt3dLs+n5az+UcNHIo0Q/OBpWKyg9d7c2m5vMDCFoNcWuXo7sqGXtFFflL12C5WIBffCyd3liNaLdjLTJS+PALWPOKPDIuloZ6/B2g55oZ6Mf0x1Zn5qeFG6jKkKLADd27jsOpywEI6Z/EVX+b5wgneoKsVVJIT0Gj5qoX5xHcpyui2UrOE29R9vUPzmurWngj1mPNDCLHDMBWZ+bUwled3CF713MkdZmT+wdHaODSvSfIXvV3B9ePP7w4l+A+3bCbreQ+8RblX3/vdn2N4J2dtGYm4aMHYK8zkbNoA9UZUnvuv2c9GWMldnD/ZLo7wpiW7zvOaQe71xvpBHSPB7uI6WIxucs3u71pNYne32fJldftnd8Bau99WHv2naoWyhigy+r7nOwzS16m1sHuvet5fhgv9U+B/ZIbwyF/fozzD78OgDoihO6vpaNNiMJ8qYSc2euxlTeumbOJ3qeXdntmFuEObq6LzX13P0emi83JLjafddg84JtXEXQarGWSU1p9NIszKzc5r63zMl6091gBUGHSerW5vduzHNyW+hC4sn7EVYMPbyBjwgq3cMjensGgfesXwDV5H7YtPKVMmt7t9g79yWXL2Q86ZD625rjsF0Vx5JUAvDku7a2WGk17q40RXX82teS4tKdaclzaW94cl/ZWS45Le8ub49Le8ua4/BJqyXFpT7X2MN2eau2ho73UkuPSnpIzr1tyXNpTrTku7amWHJf2lDfH5ZdQS47Lb1Fy9SEg7zOY4rj8POqojkuLa1xEURwpCMIw6Z/iYUEQrgbSgJOiKH72i9yhIkWKFClSpEiRIkW/IXXkkMMdWS06LoIgPAZMBPwEQdiNtFj/C2ClIAgDRVFc3f63qEiRIkWKFClSpEiRot+7WosqNgkYAOiAAqCTKIqVgiCsBw4CiuOiSJEiRYoUKVKkSJGidldrjotVFEUbUCsIQq4oipUAoijWCYLg01cum12eKXKiTPO0AWos8qw10cg0p9RkVWOzy5PfV7plwZVIrrVMcslkadtu1D+nInQm2dhyqU6m/A5UtW3zvt+CArTyrJWTU3aZ1rjIKbnWMlllGh8BAjTy1O0ys78s3F+T7HI+wPyK1VprMguCEOj49+CGHwVBCEOZnqfIIbmcFjn1e3NaFClSpEjRr0tyOS2KFLWnWnulN1IURROAKIqujooGuKfd7kqRIkWKFClSpEiRIkWKXNRaVDGP8zNEUSwBStrljhQpUqRIkSJFihQp+g1LmSh2efr9zfFRpEiRIkWKFClSpEjRr06yrbYNTRlIlydmglpFyTu7KXj1Q7fjgtaPxBcXEdgvGWtZFafnPov5YhHq8BCSNy8nqH93jNv2OXcJbgu30+OzQK3C+M5uCje470wuaP3o9uJiAvomYyur4sy89U5u0qYVBDq4Fx/ZfFl2Jz19L5GpA7HXmTm18BVqMs80SxPcL4meLzXsknuc0w9LO093SZ9C3NRULMZKAM6ueZuyvcd94nZ9agYRYwZhc+xUW5t5ulmaoL5Jzp1qy/Yd49wj0g65XR65m4hxQ7CbrZjOFZK7+GVslbU+25z49L1EOGzO9mJzUL8kejhsLtt7nDMNNi//M5FpQxHtdiwlleQsfAVzYVmrTDnLOWSUxBbUKozvemZ3fWExgX2lun12vsQOGdGf+JV3I2j8EC1WLq1+k+r/ZCrcDsiW02aAxKfuJTx1EPY6MzmLXvbaprq/eL+08/TeY5x5RGpTnZf/mcgJw8Bux2KsIHvhK1h8aFMgbz8iBzt41CASHpX6kdL3dlO88X2344LWj87PLyGgTzK28irO3b8Oy8UiAvr3oNOa+x2JBApffJvKnQd8tlVOdsioQSQ8NhNBrcb47i6KNjav212eX0xg3+5Yyyo5d79UtwP796DzmvlObsGL71DxK+A2qOtTMwgfM8i5m7un+hXoUr/KXepX5J+uJWHpHQT06MQPN65w7gTvi65kfGxQ/NybSXzsbg5ePd1t9/qWJOcY2XP1NPSpA7HVmfjpgY1UebA5pF8iV/9tHip/Lca9x8l66E0A+mxeSGByPAB+oYFYK2s5lLqizfeg6NcrmbZIVtHl6dlk/b8n+WH0AiJvGYF/j05uSaL+PA5rRTXf3zCXwtf/TadVdwMgmszkrX+bi0+9eVnczk/PJufuJ/hpzP1E3DIC/x6d3ZLo/zwOa3k1P46YQ9Eb/yZh1T2N3Gf/l0tPXwbXoYjUgQQkGThy7QKy01+j+9r7PKbrvnYW2embOHLtAgKSDESMGeg8dmnzpxwfu4zjY5f57LSEjxlEQKKBE9fP58zy10ha45mb+NfZnF6+kRPXzycg0UD4aIlbsT+DjNGLyBy7hPrTeSQsuL3NNh+7dgE56a+R7MXm5LWzyE3fxDGHzeEOmy9t+JgTY5aSMXYZZbuP0nnJ5Nahcpazg517zxP8lHo/ETd7YN8xDltFNT+OlNjxD0psa2klufeu5uT4hZxb/BJdX1yscDsiW06bkdqzf5KB49fdT+6yjST91XObSvrrfeQue43j192Pv0ubytvwMRmpS8gYl06pr20KefsRWdgqFQlPzuHMtMfJGjef8JtHouvuXs6RU8Zjq6jmVMpsiv/+MYaV0wCoP3We7JsWk33jQs7c/RidVs8HdRuGW7nYKhWdnprN6Xue4OTY+UTcPBJdk7od6ajbP42aTfHf/41hpVS3606d49RNSzh14yJy73mcTs/M6/hch8LGDMI/0UCGo34ltlC/zizfSMb18/FPNBDmqF+1J8+TPXMdVQd+bBP3SsdHAG28nvCR/ai/WOw7WMYxUp86gIDEOL69ZiEn01+n17oZHtP1WjeTn9I38+01CwlIjEM/ZgAA39/3EodSV3AodQVFnx6i+NNDl3UfHUF2xA79X0eVLI5L0IAemM7mYz5fiGixUvrx14SPH+6WJnz8MIzbPgeg7NP/EHJDPwDsdSaqD/+E3dT2aBkSt8DJLfv3V4SNH9aEO5zS9/c5uN8Qcn0jt+bwT9hN5jZzG6SfMJSirV8AUHUsG7/QQDQx4W5pNDHhqIMDqTqaBUDR1i/Qpw29bCZAxIRhFL8vcauPZaEOC0ITE9GEG4E6JIBqB7f4/S+ISJPKpOLLDLBJsRmqjmahNeh9Zke62Fx9GTbbquuc6VSBOkQfGpOc5RzYlP1Jc3bY+OEYHezyzxrZdT+cwVpYCkB91nlUOg2C1rePor837u/VZoDItKEUb/sSaGhTQZ7bVEhgY3ve9iWRadI9urYpdaDO54nWcvYjcrADB/TAfC4f8wWpnMs/2U9ok3EqdPxwyj7YKzE++4bg6/oDINabnDxBp0VsY9hTudiBDWPzBZe6Pc6dGzZuOKUfuNbtBq7ZyVXptG2awC8Xt0ERE4ZR0sb6VeJSv+pzLlGfm9dm7pWOjwCJT07j7FNvtWlvADnHyOi0oRRs2w9A5VGp/9I2sVkbE45fcACVR7IBKNi2n+iJzZ+DYm++hoKPvrms+1D061WLjosgCEmCIPxDEISnBUEIFgThdUEQvhcEYZsgCN0uF6o1RGLOb1zbby4wojVEuqeJc0ljs2OrrMUvIuRykQBo4vSY8xq5lnwjmjh9kzSRjWlsdmxVNaivkNsgrUGPKc/o/NucX4quyQCqM+gx5TemMeWXug2y8femMWjfc/R4YR5+YUG+cV1tAsx5RrRxnvLb2GIagJi/jKF83zGfuNDcZpMXm93YTWzusvIvDDn6GtG3j+D8uvdaZcpZztombHO+EU1sc7alFXb4jddR98MZRLNve2r83rhysuW0WeJHYnLhm/KNzR7EtQY9Zrd2596eu6y8k8FHNhF920jOr3/XZ65s/YgMbE2svrEMcfQjTcvZNY2znEMBCBjQk567XqXnzpe59PAG58O1L5KLrYnTY8l35ZZ46Ds9caW6HTigJ712v0KvnX/j4kMdn9ugpm3qSupXm7hXOD5Gjh+COb+U2h/PtYkr5xipM0RQf8m9b9I1ef7TGSIx5Zc2pskrRWdwdyTDr7kKc3EFdWcKrvieFP261NoXlzeBw0A1cAA4CUwEdgD/8HaSIAj3CYJwRBCEIx/WnPWUotkvzV4WeNgo44r36vG090ZzsIc0V8htuLIv/BbS5L+5k8PD7+dYajrmwjISH/cxIrUncDOuR7DbX/EP3I5otVPy4X7fuOA5O9tgM8D5v77DkcFzKP7gKwz3pl0W8xcrZ5/KuGW2f8/OxD94N+cf3KBwOyJbTpsBwYf27HGbIZck5//6NkeHzKb4w/0Ypk/0Fdwqt/36ERnYl8t0pKk7kUXW+Pnk3LyEmLmTEXRt2JhYNvblcqX/qz2Rxalx95N181Ji5k36FXC9X7v5ONV6/WqzrmB8VAVo6bTodp9e5vnC/eWehS6vrJsmib31Ogo/+s/PcUOySezg/+uoam2OQogoihsBBEGYJ4ric47f/y4Iwv3eThJFcTOwGeBIp/9qZr0534jWEOX8Wxunx1JQ6jGNJd8IahXq0EBs5b4tOvMmS74RbXwjV2PQYyl051oKpDSWAgc3JOiKuIbpacRNTQWg6kQuuvjGtxpaQySmJnab8oxub1x0hkjMjjSWkgrn7wX/u4febz3olRs7LY2YqeMAqD6R42a3Nl7fbIG7uclbW228HnNBY5qoySlEjB3CT3c81qrNcdPTiHXYXO2wuSEHXe1xtdmN7SENQMlHX3HV/6ziwvqtLfLlKOcGmZuwtQY9lqIm7HwjGi9sTZyexM0Pcm7xi5jP+f4m6ffG/b3ZHDctjdipYwGozshBFx/l0qb0zdtUvhFtvGs/osdc6KlNfc1Vb63iwrOeH37k7EfkZANYCkrQNO1HmpazI01L/Ygp9yL2unr8e3alLjOnQ7MtBSVoDK7cqOZ9Z74P3JyOz42dlka0o37VnJDaVLXjmDZe3yxgRWv1y1f9XOOjf9c4dF1iGLDvWce5egbsWkfGxAexFJe3eA+/9BjZafp44u+SbK48kYt/gp6GpxmdQY+pST5Kz0GNX2F08ZFueS2oVcT8cRiHxnl/BlL021VrX1zsgiD0FARhGBAoCMIQAEEQugPqy4XWZGTjn2hA2zkGQeNH5C03UL7bfYFV+e5D6CePBiDij9dR9U3bI+944uq6NXIjbh5BhQdu5KQxDu71VH3z3RUx87fscC6mN+44RMyUFABCBvXAVlWLpci9g7EUlWOrqSNkUA8AYqakYNx5GMBt7qt+4nBqT17wyi18cweZ45aSOW4pZTsOET1J4gYP6omtshZLkXtHYSkqw1ZdT/CgngBET0qhbKeUN2EpA4mffyunpq3BXtf6vNaCLTvIGLuMjLHLKHWxOXhQD6wt2BzsYnOpw2b/xDhnusgJQ6nLudQqX45yblBtRjY6l7odcVNzdsXuQ+gd7PAbr6fqPxJbHRpE8puPkLf2LWqOnFS4HZQtB7fgzR1kjEuXFtRvP0T05FFAy23KXt3YpqInj6J0R0ObMjjTRYwf0mKbkrMfkZMNUjlru8Wj6RSLoPEj/KaRVDYp58rdB4m4XXogC7vxeqod5azpFOtcIK5JiEaXlID5YpFPXDnZUt2OR9s51lm3K3cfdOfuOUTk7c3rtrazO9c/KQHzxcIOyy18cwffj1vK9476FdXG+hXlUr/aop9rfKw9eZ7DfWZwdOg8jg6dhynfyInxok2yxAAAIABJREFUy1t1WuCXHyMvbtnlXFBfvP0wcZNHAhA6WLLZ3MRmc1E5tup6QgdLNsdNHkmxo/8CiBjZl5rsPLfpZIp+PxJaWrgnCEIqsAGwA7OAxUA/IAyYJYrix60BPH1xAQgbM5jOj98LKjXG9/aQ//L7xKf/hZqMHCp2H0bQaUh8aRGBfZKwlVeRO+85zOelzqjvt5tRhwQgaPywVdaQdefj1GdfdLu+SvBsV+jowXR6fIYUxvS9vRS8vA3D0jup/S6Hit2HEHQaKQSgg3tm/rNObu//bEYdEujk5kx9nPrs5s5DjcX7Z+rkNTOJGD0Ae52JrEUbqM6QwiYO3LOe42OXARDcP9kZDrls33FyV0khF3u+vIDgPt1AhPoLRWQv2+TWyWlU3uf1dntmFuEpA52hHhvCNfbd/RyZ45YCENQvuTHU4+fHOPvQGwAM+OZVBJ0Ga5n0tqX6aBZnVm5yXttmb9n/TVozk3CHzTkuNvffs54MF5u7O2wu33ec0w6be72RTkD3eLCLmC4Wk7t8s9sbqQCN5yAN7V3OHmcNuLATHmtkF76yjbgld1KbmUOlg931xcUE9k7CWl7F2fslduyCycTOn4TpTOMiz9y7HsdqrPAO+x1zf+s211m8fxBPfGYmEaOlcKI5i1+lpqFN7X6WjHHpAAT1T6aHIxxy2b7jnHG0515vLCMgOR7R0aZOr9jk1qa89Z3Qvv1Ia2pPdqDWcz8SkjKYeEdI4rKteyh6dSuxi6dSl5lN5R6pnDs/v4SA3knYyqs5v2Ad5guFhN86mpi5kxCtVrCLFP7tXSp3tS1Eb3uz7aLnTixk9GASHp2JoFZRunVPY93+LsfJ7fqCxLWWV0lhiS8UEnFrCjHzJoHFiiiKFL70LhW7DnpkyMU1tdCmuj0zizBH/TrtUr/67H6O713qV5JL/TrnqF8RacPp9vRM/PSh2CprqPnhDKfufMp5bWsLY+SVjI+uGnx4AxkTVriFQ/Y2PkL7j5FlZn+v7F5r7iVyTH/sdWZ+XLiRqgwp9PSwvWudoY1D+ic5wiFrMO49QdaqLc7zr3ppLpVHs7n033s8Xj+18L0WRuiOozu6en4+7ih679y/OmQ+tui4AAiCMBywi6J4WBCE3khrXH4URfEzXwDeHJf2VkuDb3urJcelPdWS49Keas1xaU+11DG3p1pyXBQp+jnUkuPSnpKz75RL3hyX37K8OS6/ZbXkuLSnWnJc2lNyjY/QsuPS3lIcl59HHdVxabEVC4LwGJKj4icIwm5gGPAlsFIQhIGiKK7+Be5RkSJFihQpUqRIkSJFv3O19vphEjAA+P/snXd8FHX+/5+zm930uiHJJrQkEE46oUsndAV/d4pX9BSQIr0YrKBYkBNQsYHg3emd99UTvPPU4xACKJ4NQhULpBAIIX3T67b5/bGbzW6ySSYJYaI3r8cjD2Xns/Oc12fen/fOZ2Y+n48nkAt0FUWxTBCEbcBxQOm4KFKkSJEiRYoUKVKkqMPVUsfFLIqiBagSBCFdFMUyAFEUqwVBkOe9JEWKFClSpEiRIkWKfsLqzKvTd2a11HExCoLgI4piFTC07kNBEAKxDdhvGaCWp38jyvj+rt//2PvSapV8fktq5HmP1ixjfKlkSnaWFich7DiFeNbIwrVY5TvP3prWLUx5vdTu9bLaKDnHjZXWeMrCFd2vuHNDpJYpj5hkzJ0B2rat9t5eadQWWbhyjmOKCmz/8gKKFLlTSx2X8aIo1gKIoujcA9EAElc/VKRIkSJFihQpUqRIkaL2qdmOS12nxc3nhUBhhxyRIkWKFClSpEiRIkU/Y3Xm1ek7s+R790ORIkWKFClSpEiRIkWKJEqeSc0B/wnxRD2xEEGtxvD3Q+Tv+ofLdkHrQfcX1uIzoBfm4jLbYlNZ+fgM6k23LcvthQRyd7xL6UHpC3v5TxhC102LbIsu/T2JvJ2NuT1eXIvPgFjMxeVcXm7j+o8bROTD9yBoPBBNZq5tfouKr8630rM8bDk9+02IJ8q+iFrRe0kU7Hq/EbvbC+vw7h+LpaScKyu2YsrKx3tQb7puWWEvJJC34x3KWnGeAWKfmU9IQjyW6lpSVr9GxfmMxsc3MIY+9oW9io6cJn2DbZGrHolzibhrCiZDGQAZW96h+MgZyezem+ejS7AtZvbDqp1u2f4Do7npZRvbcOQMqY/Z2P32rMEnNhIAjwAfzGVVJCc8KInba/N8dHbPF1Y17fkXLy9H7aXFcOQ0aXZuz8S56O+u93zp2XcoaoXnuM3z0CXYFkX8cdUuypvwbFtUzOY55bG3AOi/Z3Ujz3ULkTUnuWI7YOIQuj+5ENQqCt9NIve1fzbiRu9Yg89AG/fS0u0Ys/JRB/kTu+dBfAf1wrDvKJkb3pDM7Ayeu26ytWXDu+65PXesxXtALJbicjKWbXN4jtn9ED52z1kb9/xkPAPEPLOAkIQhWKuNXFz9KpVNtKk4Rx45w6UNf3bZHrV0DjFP3MPXfee7LBDYkuTKYdHPLCDY7jm1Cc++A2PoXbdQ8pEzZDTwHLl0DtFP3MPxVnhuT/4CiLpvBlELZiKaLRgOn+bS03+TxJUrtuVsU23NYQARy28n9LdTwGIl8/E3KDt2VjLXd9xQwh5bgqBWUbLvIEV79rlyNR7otyXi1a8XlpJystdswXQtn4DZEwlZeLujnGefaC7/chW1P15qtXdFP23J88RFpaLr00u4dO+TXJiynOA54/Hs3c2lSMivp2IpreDHCUso+NNH6B+2DampvniFi7PXcXHWGtLv3UTXZ5eBWqINlYpuzywh/d4n+TFhBcFzxuHVgKuzc38Yfz/5f/yIyEdsXHNRGekLNnNh2mqurH2JHjvWttqzLGyZPUc9dT8Z8zaRMnU5QXPG49mrwXm+cxqW0gouTlxCwZ8+RP/wPABqLmaSOnstqbNWk3HPE3TdvFz6eQaCE4bgHaMnefRKUhN30+u5RW7L9XpuESmJu0kevRLvGD3Bkwc7tl3b829OT1nP6SnrW9Vp0SUMwSc6gm9GreJC4h76bF3otlyfrYu4kLibb0atwic6ghA7+/vFO0hOeJDkhAcp2H+cgv3SVp8OSRiCd7Se46NWkpK4m7it7j3HbbV5Pj5qJd7RegcXIGv3vzmZsJ6TCetb1WnRJQzGOzqCr0et5kLiG/TZel8TnhfyY+Ievh61Gu/oCHR29neLX+JEwkOcSHiI/P0nKNh/omWojG2q+zNLSPn9U3w/aSUht43Dq3dXlyKhv5mKubSC78YuJe+Nj+j66D0AiLVGsre9Q9bTb0nndRLP3Z5ZQto9T/Lj5BUE3+aG+5upmEsq+GGcjRv16L31nrf/H9ee+Yl5pj6PnBy9ktTE1+n13GK35Xo9t4jUxN2cdOSRIY5t2kgdweMHUpNV0Cb2jc5hddzTo1eSlvg6sU14jn1uEemJuzlt5wY18BzUSs/tzV9BY/oROmM4yZMeIHnCOq7u+kgaWK7YlrlNtTWHefXuSshtY/l+8kpS7n6S7pvvB5X0a7DwJ5aRtehxLs26n4BbJ6CNdfUcOHc6ltIKLk1dSNFbH9Bl/QIAyj7+jMu3reTybSvJWf88pmt5P/lOi7WT/3VWydJx8Rncm9rLORiv5iGazBR//F8Cp450KRM4dSRF/zgKQMl/vsR/zCAAxBojWGxVqvLU0ppXBG3cXIyZTtxpI1y500ZieN+ZOxCA6u8zMOcVAVCTkonKU4Oglf7ASi623J6NV+rPc8nHnxMwzfU8B0wbSfE/jgBQ+p8v8bu57jzXOs6z4KlFbOVUR6HTh5O39xgA5adT8QjwRRsW5FJGGxaEh5835adSAMjbe4zQGSMa7au1Cp0xjNx9nwNQdqppttrPm7KTqQDk7vucLjOHN9pX2JzR5H3wpUTucPL2HWuR6+HnTdlJu+d9xwid2X7PXWYMl+TZQ4Ln8DmjyJXgWa7Y9q3LX3Zu0YdfENQgroOmjcCw71MAivd/hf9YG9daXUtF8o9Ya9s2E5+8np24HzXmBk0bSZGdW7y/nmutrqUy+UestW2b0UnOHKabPpz8vZ8BdXnEB02DuNaEBaH283Hkkfy9n6GbUR/XsU/NI+Ppt1s9XZtcOSzEyXNFGz1HPzWPy6303N78FXnvNDJf+Rei0TYbn6mwTBJXrtiWs021J4cFTRtJ0YdfIBrNGK/mU3s5B9/BvSVxvQbGYbySjelqLpjMlO3/HL8po13K+CWMovSDwwCUf/IFPqMHNdqP/60TKPv3sVb7VvTzULMdF0EQVIIgLBAEYb8gCOcEQTglCMLfBUGY2B6oJkKHKad+bL8ppxBNhK5xmWx7GYsVS3kl6mB/AHwGx9En6VX6HHyZrMd2Oi5wW5I2Qocxu55rzDGgCW/IDWmSW6egWTdT/X2GI0F2ZracnjXhTucQMLljh7s7zwEAeA+OI+7Qa8QdfIVrG6SfZwCtPoTabIPj37U5BrT6kMZlcpouE7lgBvFHtxP34lI8An0lsz31IdRcq/ddm2PAswHbswG7JrtxmaBRN2EsKKU6I1cyt/aaq5+WuLUNuFELZjDs0+302dFaz8HUSGIXObGL8NQHu5RpjWfZ2pQ+BKNT/jLmuomtCKcyFiuWsio8GnDbIrk8axpwTTkGNzk7pL5ME9y2SNa8rde55BFjThGeele2p17XII8UobWXCZk2jNqcIip/uCKZWc+WJ4c19FzbhGdjjmu9OHs25hRR1UrP7c1fPrGRBI68ifgDzzL4gyfxHxwriStXbMvaptqRw6R8tylpwnWYc+u/a84tdHtNYM4pcHCt5VWOa4I6Bcwar3Rc/ofV0hOXPwHdgS3Ap8B++2cbBEFY2dSXBEFYLAjCSUEQTv6jwl3ycjO3eMM7M+4m9bcXqTqbwsWpK0iZ8wBhy+5A8NS0YKNpbGu4AF5x3Yh85B4yH9kpjSk3W1bPbT3PtjLVZ1NImbactDnrCFs6V/p5bnK/0o8v+61DnBi5ktMJ6zHmlRCz6R7p7HbGd53CfjlG8tOWpriNbni64dY9zbr2l0N8M3IlJyfbPMc+2fGeGxYJ/+XN5H3wVZuRNya221rPrUBIR3eePOI2BqQj2sXuoBzmdm2ZRmz3ZVTeWrqtuZ0rW99rFbNZ+I3IYW53Kd1z1zW3k9kmz+3LX4KHCo8gX07PfJT0p96m7xvr2oq9MbEtZ5tqVw5rR25r4zWBc/x5DeyDtboWY2rrbwZ0Nomi2Kn/OqtaemY+VBTF+fb//0IQhG9EUXxcEITPgbPAK+6+JIriHmAPwNkecxq5N+UWotGHOv6t0YdiyityLZNTiCYyFFOuAdQq1P6+WEpcB/jVpmVhra7BK64H1efTWrBiu1OnjaznavU6TPkNuYYmuZoIHdF7HuHK2h0Yr0i7Cy43W07PplzbOayTxh07V8J5Tpd2nvXzp6O/awoA5WfT8Iysv5PjqddhzHVlG7MNLncSbWWKbcdVWOr4POf/DtP/7Yeb9Ro1fzqRdyfY2el4RYVSykXHfmvt+3V4asD2itRR63R8glpF2C0jSJ7aPDdy/nQi77Z5LjubhmdU854bcj0jnTwXOHn+22EG/K15dtf50xyey86m4xWlo24PTXuuvzPnGRniYDt7PjH1kWa5dZK1TTnlL22EDlPD2LKXMeXYuQE+jeK6LZLLs6kBV6PXNc7ZubYyzbXltuhGe9bPn0HEXfVt2TmPaPUhLu0U3LQpfQjG3CK8ekTg1T2M+KPb7Z/rGHJoK2dnPoKpoKQJtjw5LGL+DMLtnivsnuvOXJ2fhp61etd6qfPs2T2MwU6eBx/ayrkmPF/P/FWbXUShfTxg+Zk0sFrR6AIckxM0JbliW/Y21cYcJuW7TcmUW4hHRP13PSJC3V4TeOi7YM6zcVX+PlidPAfcMp7y/Z+1xq6in5laeuJiEgQhFkAQhHjACI71XdrcHas6l4pndCTabuEIGg+CZ4+jLMl1AHLZ4ROE3D4ZgKBZYyj/6lsAtN3CHYO0NVFd8IqJwpiV1wquHm23MAe3NMl1EHBp0gl0dzTmqgN8iX1rI9nPvU3lyQtt9Hzj2XJ71vaMRNPVdp6DZo+nrAG7LOk4wbfbfjADZ42hws7WdHU9z54xUY4ZTZpSzpsHHQNRDZ8kE37nBAD843tjLq/CmO/6w2nML8FSWY1/vO393PA7J1B4MBnA5d3q0JkjqLxwtVn2tTcP1g+oP3CCiLnjAQgY2htLU+yKagKG2tgRc8dT+MlJx/bg8QOoTM12ebXKnbLfPOgYTF94IJnwuRMc3KY8m5244XMnUPiJG8+zWvac9eYhx4D6ggPJLp6brO+KGhfPBXZ2azzXSa7YrjyXipcTN+S2sZQ04JYknUA3d5LN1y03U/5l62ayakpyevbs6cSd05hbknSCEDs3+JYxlH/5bVttuuhGe8558xPOTFnPmSnrMXxygrA7JwK2PGIpr8LUIK5NDfJI2J0TMRxMpupCJsf730fy8GUkD19GbY6BM9MebLLTYmPLk8Ny3/yEc1PWc27KeoqcPPvZuU159nPyXGT3nNz/Pk4NX8Ypu+ezzXi+nvmr8MAJgsYOAMA7Ro+g8Wix0wLyxbacbao9Oawk6QQht41F0Hqg7RaGV7SeyrOpkrg151Mc1wRoPAi4ZTwVR1xnC604epzAX9o6s/4zxlL1tZNnQcB/5jjK9n/eVuuKfgYSmnscJAjCZOAtoAbQAL8RRfG4IAhdgPWiKLY4R6u7Jy4A/pOGEvX4QgS1iqK9h8l7dR8R635H1bdplB0+geCpoceL6/DuF4O5pNw2HfLVPIJ/OZGwZXeAyYwoiuS99HdKDzWedUkU3T2HhYBJQ4l64j7btJrvHannnk+jLMnO3bEWHzv38ortGDPzCF85l/Dld1Cbke3YV/rdmzAbSt1yOhO7o7lqVdNjT/wnDiXSPh1y8d7D5L+2l/C1d1F9PtVxnru9YDvPlpIKMlduxXg1j6BfTiJs6R2IZjNYRfJe/jtlhxpPh1xS49Uku9eW+wieNNg2jema16g4Z5uBJP7wNk5PWQ+A3yCnqUSPniX90T8B0OeVlfj174koitReLSB1/W6XH1FzE/FVp7gt96GbPAhLtZEfV++k3M4efmSrY2pj/0Ex3PTyMvu0nmdJebR+OtGbXlpG6alUsv+a1GjfqmbuGfTech8hkwdjqTZycfVrDu6wI9s4mbDewf3Fy3XTp54l1e75F6/aPCOK1FwtICXR1bOlhfscfbYsIGTyIKzVRn5YvcvBHnHkOcfUxv6DYuzTIWvsnuunMr3ppaWUnUrl2l8PN9p3iGeNW2ZHx7bF6v48B04eSrdNC0ClxvDeYXJeeZ/IxN9SeS6N0qRkBE8N0S+twad/DJaSctKXPY8x03aDZcDXe1D7eyNoPLCUVZLyu03UpGY1YqhV7s9zR3tu6ichYNJQum6q5+a+sg/9A7acXWrn9tyxFm+754zl2x2e+321B7W/j8Nz2l2bqEl1vZB2+1rWDfJcYWz6NdTYLQvteaSWlDU7qTiXDsCQw9s448gjsY7pkIuPnnHkEWcNT97JmekPuUwNLLp9X6heHZnD1M3kkZgtCwmye05z8jzo8DbOOXnuZeeWHD3DJTeehybv5FwDz6Zmcmd78peg8eAXO5bi178nVqOZ9CffpuSL71z2H6B1P5i9o2O7KXU019pMXbcnh+lX3oHu11PAYiFz058o+/R0o/37+biva98Jwwh/dAmoVZS+fwjD6+8Ruupuar5LpeLocQStxjYdct9YLKXlZK99zjaYH/AZMYAuifO5cmfzrwH+IuU/zTesTqLbut/aed/HAj7M/HenrMdmOy4AgiCMBsyiKCYLgtAXmAFcEEXxP1IATXVcOlpNdVwUXX8113HpaDXXcelItdRx6Ug113HpSLXUcelINdVx6Wg11XG5EWqq49LRkuvV5uY6Lh2t5jouHamWOi4dqeY6Lh2p5jouHa2mOi4/VzXXceloNdVxuRFSOi7XR52149LsGBdBEJ4AZgIegiAkASOAY8DDgiAMEUVx8w04RkWKFClSpEiRIkWKFP2Pq6XB+XcAgwFPIBfoKopimSAI24DjgNJxUaRIkSJFihQpUqRIUYerpY6LWRRFC1AlCEK6KIplAKIoVguC0JkX1lSkSJEiRYoUKVKkqFNKuYhum1rquBgFQfARRbEKGFr3oSAIgUisc6uM74jLJZVM76bLVdcWq3xjHzxVFlm4HnKOcRFkGuNile91XDnHmsglucaayNWeBZniGkAr0zg9y/9gHhFk9CznmA85VGtu6RKv4+RRq1yWK+oYtRTV4+1THyOKonMUaoB7O+yoFClSpEiRIkWKFClSpMhJzXZc6jotbj4vBAo75IgUKVKkSJEiRYoUKfoZS5RpZr+fuuR7x0eRIkWKFClSpEiRIkWKJEq2FyADJg6h6ybbwoSGd5PI2/kPl+2C1sO28NKAWCzF5WQs24YxKx91kD8xux/CZ1AvDPuOkrVxz0+CC+A/IZ6oJxYiqNUY/n6I/F2N2d1fWIvPgF6Yi8tsi25m5eMzqDfdtiy3FxLI3fEupQcbL8bYOT3b2IJaheHv7tk9XlyLz4BYzMXlXF5uY/uPG0Tkw/cgaDwQTWaubX6Liq9at/p4j6fvI2hyPNbqWtLXvkrV+UuNyvgMiCF2x0r7ImqnubLRtphZyK2jiXrg13j37sr3sx6i8tv0VrGjn15AUEI81mojaWteofJ8RqMyvgNj6LVjhY195DQZG20LUHZ78DeETB8BVismQympq1/FlFcs2XPw5HgszXj2dfJc7OS5+8Z7CJ46DKvRTO2VPNLXvoKlrEq652cWEJwwBGu1kdTVrzbpuXfdQn1HzpCxwea5+4O/IWTGcESrFVNhGWmrX8UowXPAxCF0f3IhqFUUvptE7mv/dNkuaD2I3rEGn4G2+Lq0dDvGrHwAIpbfTuhvp4DFSubjb1B27Kxkr3Jx69hytefuT91H4OShWKtryVj7ClXfuW9T0S+uQuWlpfToKTIft8WXOsiP2F0P4NktjNqr+aTfvx1LaaVkz3LVt5JHbnAekcFzW+NLHeRP7J4H8bW3qcwNb0j22R4utD+uoX05u06RS+cQ/cQ9HO8732Wh0aYk13WQop+P5HniolLR7ZklpN3zJD9OXkHwbePw6t3NpYjuN1Mxl1Tww7j7yf/jR0Q9ahtSI9Yayd7+f1x75q2fDtfO7vr0Ei7d+yQXpiwneM54PBuwQ349FUtpBT9OWELBnz5C/7CNXX3xChdnr+PirDWk37uJrs8uA7XEUyez527PLCH93if5MWEFwXPcsO2efxhvY0c+YmObi8pIX7CZC9NWc2XtS/TYsbZV6MDJ8XhF6zk3ZjkZD75O9JbFbstF/2EJGQ/u4tyY5XhF6wmcNASAqguZpC7cSvk3P7TadtDkeLxi9Jy5eQXp63cR8wf37Jg/LCZ9/eucuXkFXjF6gibb2Nk7P+RcwjrOTU2kKOkU3dbNlcz1jtZz1u45phnPlx7cxdkxy/GO1hNk91z6+TnOTVrD+SnrqLmUTdTK2yV7Dk4YgneMntOjV5KW+Dqxz7lnxz63iPTE3ZwevRJvJ8/Xdn7I2ckPcG7Keoqlelap6P7MElJ+/xTfT1pJyG3j8Ord1aVI6G+mYi6t4LuxS8l74yO6PnoPAF69uxJy21i+n7ySlLufpPvm+0ElvU3JwrWz5WrPgZPj8YyO5PzYZVx+aBc9tixxW67Hlvu5/NAuzo9dhmd0JIGT4gHQL/8VZV+c5/zY5ZR9cR798l9J9ixXfSt55MbmEVk8tyO+xFoj2dveIevptyR7vB7cducR2p+zAbSROoLGD6Qmq0CyZ1mugzqprIid+q+zSpaz7ju4N7WXczFm5iGazBR/9F8Cp41wKRM0bSRF7x8FoHj/l/iPGQiAtbqWyuQfsda2flVWubgAPoN7U3s5B+NVO/vj/xI4daRLmcCpIyn6h41d8p8v8R8zCACxxggW29wIKk8trYkn+T07sT9uzA6cNhLD+86ebezq7zMw5xUBUJOSicpTg6CV/oAwePoICt//DICK0ymoA33RhAW7lNGEBaP296biVAoAhe9/RvAM2zmpSbtGTXp2600DITOGU7DvmJ2dikeAL5qwoAbsINT+Pg52wb5jhMyw1Y2lotpRTu3jKfl8B08fQUErPRc4eS49ds4RZ+WnUtDqddI9Tx9O/t46dioeAT7uPfv5UG5n5+/9DN2M4Y08q3w8Jb3761vXpuzxVfThFwRNc21TQdNGYNj3KQDF+7/Cf+xA++cjKfrwC0SjGePVfGov5+A7uLckr3Jx69nytOeg6SMwvG/zVNlCfFWeugiA4f1PCbLHddD0+jox7PuUoBmudda8Z3nqW8kjNziPyOC5PfFlra6lIvlHrLUmyR6vB7e9cQ3tz9kA0U/N4/LTb0ueAlGu6yBFPy/J0nHRROgwZteP7TflGNBE6BqUCakvY7FiKa9EHez/k+TWsU05zuxCN2wdpibYPoPj6JP0Kn0OvkzWYzsdDVgKVy7P2gZsY44BTXhjdlOe6xQ062aqv89ANJpbwQ6h1pmdbUAbEdKojDHH0GyZtqghuzbH0OjHW6vXYcw2uJZxYnd/+HcMPbmbLr8aT+a2v0vmGq+T57DfTqbk6GlJ3Do/tS5+ivBs4NlTr3Nl5xS51Ev3h3/LsFOv0+X2cWRufU8CMwSjU5sy5hrQ6t35dYqvsio8gv0lfbezcaEztOf682djux67pmF85RjQ2o9PExqEKd/22o4pvxiNLlAaV8b6VvLIDc4jMnhuT3y1R3LGtY3fvpwdMm0Yxpwiqn64Ipkp13WQop+Xmu24CILgIQjCEkEQPhEE4VtBEM4JgnBAEIT7BUHQtJnqbir1Rj12N4Xa28OWi9vkfhvsWGiaXXU2hYtTV5Ay5wGeRAfmAAAgAElEQVTClt2B4Cmx+uX0LIXdjGcAr7huRD5yD5mP7Gwlu/F+RSns62BccOvJdb9uVxNwKpL5h3c4NWwJBf/8HP38mVLBLXKleI5cdTui2UrhPz+XxoUmwrsh2x26vkzmH97l5ND7KfjHf9EvmNEmaOPQbqqMhO92Oq7br3eq9iwl9q8H+MbVt5JH3HE7Ko/I47k98dUeyRjX7nchOWervLV0XXO7pBtMEqANinTAdVAnlSiKnfqvs6qlJy5vA4OBTcAs4BbgSWAQ8LemviQIwmJBEE4KgnDynxWXG2035RjQRoY6/q3R6zDZXwtylMl1KqNWofb3xVLS8sCv5iQX17bfQjR6Z3ZoY3ZOIZoW2LVpWVira/CK6yGNK6NnYwO2Vq/DlN/Qs6FJz5oIHdF7HuHK2h0Yr+S2yAufN4P+Sc/TP+l5THlFeDqzI3WNBmkaG9zN00bqMOZKG7zaUBHzZjAoaTuDkrZjbMD21Osw5rr6rs0xoI3UuZZpcF4ACj/4At0to5rkhs+bwYCk5xmQ9DzGvCLX+o7UNRrg3pLn0LkTCZ4yjLQVL7bsef4MBh3exqDD2zDmFuPp4ieksefsBmw3ZWye/9usZ1cvTn4jdJga7M+ljFqFOsAHS0m5pO92Ni7c+PYcdu9M+h16gX6HXsCUW+wSszZ2C/HlFNemwhLHK0easGBMhlJJx3Cj61vJIzc4j8jk2dVL2+KrPZIjj1yvnO3VIwLP7mEMPrqdock78dTrGHxoK5ourq+aNZRc10GKfl5qqeMSL4riUlEUvxFFMcv+940oikuBIU19SRTFPaIoDhNFcdiv/Ho22l55LhXPnnq03cIQNB4EzxlHadIJlzIlSScIuWMyAMG3jKH8y29b663TcAGqzqXiGR2Jtlu4jT17HGVJx13KlB0+QcjtNnbQrDGUf2Vja7uFOwahaaK64BUThTErTxJXfs9O7NmN2aVJJ9Dd0dizOsCX2Lc2kv3c21SevCCJl/fWJ3w39QG+m/oAxZ+cIPSOiQD4xcdhKatyvKZSJ1N+MZaKGvzi4wAIvWMixQdPNNytJOW+9QnnpibaBoUeOEGXuRPs7N6Yy6sw5Zc0YJdgrajGL972XnKXuRMo+iQZAK9ovaNc8LRhVKdda9bz+akPcN7uuUsrPXdx8hw4cQiRy3/JxXlbsFa3PA4i981PODdlPeemrKfokxOE3VnHbtqzpbLec9idEyk6WOc5wlEuZPrwZj3XqfJcKl5O8RVy21hK3MS2bu4kAIJvuZnyL887Pg+5bSyC1gNttzC8ovVUnk1tkSknt459I9tz/l8O8P20dXw/bR3FB4+ju8PmybeZ+LJWVONrjy/dHZMoscdXyaFkR53o5tZ/LsXzjaxvJY/c4Dwik+c6tSe+2iM58sj1ytlVFzJJ7n8fp4Yv49TwZdTmGDg77UFMBSUNkS6S6zpI0c9LQnOPgwRB+AZ4HviHKIpW+2cqYC6wThTFFkdXnu52m1tAwKShdN10n22a3PeOkPvKPvQP/I6qb9MoTTqB4KmxTenZPwZLSTkZy7djzLQFab+v9qD290HQeGApqyTtrk3UpF6VZPhGcFUq93XqP2koUY8vRFCrKNp7mLxX9xGxzsYuO2xj93hxHd79YjCXlNumAbyaR/AvJxK27A4wmRFFkbyX/k7poeON9m+1un2I3uGe3b414MSOeqKe7fB8Po0yO7vHjrX42D1fXmFjh6+cS/jyO6jNqB/Ymn73JswN7tIazeom2T2fXUTgxCFYq2u5tPZVx1Sk/ZOe57upDwDgOzCWmLppTD89zZXH/ghA8IyR9HxmIR66ACxllVR+n8HF3z3t2LdFbMY0EP3sQoInDcFSXUva2teoPGdjD0razrmpiTb2oFh626f0LD56hgw7u88f1+MdG4loFanNKuDSQ7td7oSphKbbbM9nFxFk95zu5HlA0vOcd/Ic6+T5sp07+MvXEDw1mIttd7cqTqWQ8fDues/W5u9zxGxZSNCkwVira0lbs5OKOs+Ht3FuynoA/AbF0ss+tWbJ0TNcevRPds+JePeKBLvn9Af3uHj29HA/vilw8lC6bVoAKjWG9w6T88r7RCb+lspzaZQmJSN4aoh+aQ0+9thOX/a8I7b1K+9A9+spYLGQuelPlH0q/V38G8Ft6jx3dHtu7jx337zY0aYy1r1ClT2++h16ge+nrQPAZ2Bs/XTIn552TBGrDvan1+uJaKNCMV4rJG3JNiwlFY59C83EdUfXd3OelTxyffOIVUbPWrXFLbM98TXg6z2o/b0dbSrld5uoSc1q1uP14EqJ61pz8xPatCdnO2to8k7OTX/IZTpkX0/3ndaOvg4CGHzlo+aDrJNoereZnfd9LODg1QOdsh5b6rj0BJ4DJgPF2F5QDAQ+BR4WRbHxpN8N1FTH5eespjouHa2mOi4dreY6Lh2t5jouHamWLjg6Us1dcHSkWuq4dKSa6rj8nPW/dp6b67h0tOTy/L+YR1rquHSkmuq4/FzVUselI9VUx+VGSOm4XB911o5Ls1EtiuJl4NcAgiDosHVcdoiieHfHH5oiRYoUKVKkSJEiRYoU2dRsx0UQhI/cfDy57nNRFOd0yFEpUqRIkSJFihQpUvQzlZT1yhQ1VkvPEbsCPwB/xDYhnQAMxzbuRZEiRYoUKVKkSJEiRYpuiFrquAwDVgOPAetFUTwrCEK1KIrHpAK8tPK8my7ne7RmizzvS2s85FmMSa2SbxGocqNWFq5VxhslFnnWjUUryPd+uKdGHracU9nXmuR5P13j8b81DgCgzCRPHpFTovsVUTpccuaRapnGfPhpTbJw5VyeUfs/mEcU3Ri1NMbFCrwoCMI++3/zWvqOIkWKFClSpEiRIkWKFF1vSeqEiKKYBcwVBOEWoKxjD0mRIkWKFClSpEiRop+vrMoYlzapVU9PRFHcD+zvoGNRpEiRIkWKFClSpEiRIreS7bUv3/FDidi4GEGtovi9Qxh273PZLmg9iNz+AN79e2EpLidr1R8wXcsHwLNPT/TPrEDl5wOiSMb/W4NolPYOqd/4ePSPLwaViuK9hyh8/f1G3K7b1+HVvxeWknKurnwO07V8NFFh9E7aRe0l20q81Wcvkr3htVZ59p8QT9QTCxHUagx/P0T+rn80Ynd/YS0+A3phLi6zLbyUlY/PoN5027LcXkggd8e7lB78RjLXb3w8kU8ssnl+L4kCd56fX4d3/1gsJeVkrtjqqGsATWQXeh96jfyX3qXwjQ9a5bmt51kTFUbsodcx2uu76uwFcje2rr5jn5lPSEI8lupaUla/RsX5xssO+Q2MoY99ca2iI6dJ3/AmAD0S5xJx1xRMBtsDxowt71B85Ixkdq/N89HZ2RdWNc3+xcvLUXtpMRw5Tdpjbzq2Rd03g6gFMxHNFgyHT3Pp6b9J4sZtnocuwbZ424+rdlHuhus/MJq+Ly9D5aXFcOQMKY+9BUD/PavxiY0EwCPAB3NZFScSHpLsOeaZBYQkDMFabeTi6lepbMJznKO+z3Bpw59dtkctnUPME/fwdd/5LouZNSW5YttvQjxRjy8CtYqi95Io2NWY2+2Feu6VFVsxZeXjPag3XbessBcSyNvxDmWtaMt16v7UfQROHmpbCHLtK1R9d6lRGZ8BMfULQR49RebjtoXj1EF+xO56AM9uYdRezSf9/u1YSitbZMqVv+Rm9948H12CbTHGH1btdNuW/QdGc9PLyx1tKtWpLXe9bwZRC2Y42nL60/93Q9j99qxp1J6TEx6UxJUrj7Qnh3RPvJOIuxIcOftyK3O2HGz/CUPoummRbTHZvyeRt7NxXPd4cS0+A2IxF5dzebktrv3HDSLy4XsQNB6IJjPXNr9FxVfnJXu9Hp7r1NqcLde1n6Kfj+TpuKhU6Dct5cq9GzDlFhLzwYuUH/kGY1r9Cs5Bc6djKa0gbfIiAm4dT9hD87m26jlQq4h6IZFrDzxP7YUM1EH+iGaJg8BUKiKfXErGPRsw5xqI+deLlB8+Tq0TN/jOaVjKKkmdvJjAW8cT8dA8rq7aCoDxSi7pt65qs+euTy8h/a7HMeUaiPvoeUoPn6DWadXqkF9PxVJawY8TlhA0exz6h+/lyoptVF+8wsXZ68BixSMsmD4HXqL08AmwSBh6p1IR+dT9ZPx+I+ZcA7EfvkCZO8+lFaRMWkLgreOIeHgeV1dudWzXb1hIxbFTbfLc5vMMGDNzuDR7Zeu5QHDCELxj9CSPXol/fG96PbeIs7MebVSu13OLSEncTfmpFPq/8yjBkwdTfPQsANf2/JusXR+3mh2SMATvaD3HR60kYGhv4rYu4vTMxuy4rTZ22ckUBrzzKCGTB1N09CxBY/oROmM4yZMeQDSa0YQGSOLqEgbjHR3B16NWEzC0N3223sfJmRsaleuzdSE/Ju6h7GQqg955GN3kwRiOnuW7xS/V18um32Mpq5Lsua6+TzrqezHnZj3SqFyv5xaRaq/vfu88RvDkIRQftf3AayN1BI8fSE1WgTSoXLGtUhH11P1k3L0RU66BXh+9QFmSKzfEzr04cQmBs8ehf3gemSu2UnMxk9TZa21tuUswcQde5gepbdmuwMnxeEZHcn7sMnzj4+ixZQk/zm58Ydhjy/1cfmgXlacu0vvtjQROiqf009Pol/+Ksi/Ok/vaP4lY/iv0y39F1rNvt+hZlvwlM1uXMASf6Ai+GbXK3qYWcmrmY43K9dm6iAuJu+1t6pEGbXkYJyYltqotXw/294t3OMr02vR7zBLbs1x55HrkkGt79nNtl7tVHDohW6Wi2zNLSLvrCUw5Bvp8vJ3SpBPUOMW1zh7XP4y/n6DZ44h85F4uL9+GuaiM9AWbMecV4RXXndi/beL7EQtuuOe25GxZrv06qZpbAF5R05JleiLvQXEYr2RjupoLJjOl//4c/ymjXMr4TxlJ6T+PAFB24At8Rw8CwG9cPDUXLlN7wXZnwFJSDlZpP0Leg+KovZKD6WoeYh13akPuKIr/YeOWHvgC35sHtctrnXwG96b2cg5GO7v44/8SOHWkS5nAqSMp+sdRAEr+8yX+Y2xsscbo+KFVeWppzWuRPoN6Y3T2/PHnBDTgBkwdSYnD85f4OXkOmDoK49VcalIyW+25Pee5vQqdPpy8vbbJ78pPp+IR4Is2LMiljDYsCA8/b8pPpQCQt/cYoTNGtJ89Yzh5+2zsslPNs8tO2tn7jhE608aOvHcama/8C9Fom5HPVChtWFmXGcPJ3fe5RG4qALn7PqfLzOGN9hU+ZxS5H3wp1TK66cPJ3/sZUFffPmgasDVhQaj9fBz1nb/3M3Qz6tmxT80j4+m3JU/lJVds+wy2cevacsnHnxMwrQF32sj6PPKfeq5YU+toy4Kntk0/XEHTR2B4/1MAKk+noA70RRMW7FJGExaM2t+bylMXATC8/ylB9tgOmj4Cwz7b9w37PiVohuuxN+VZjvwlNzt0xjBJbUrdRJuKuncaV175sNVt+XqwnRU2ZzR5EtuzXHnkeuSQtkoOti2uczFmOsX1NNffn8BpIzG87xzXAwGo/j4Dc14RADUpmag8NQja1t2HliNny3Xtp+jnJVk6Lh7hOkw5hY5/m3ML0YTrXMtE6DDl2HvxFivW8irUwQFoe0aBKNL9zaeI/vAldItvl8zVOO8TMOc05mrC3XMBtN3Cif34JaLf3YLP8H6tsWxn13s25RSiidA1LpNd6GBbyitRB/sD4DM4jj5Jr9Ln4MtkPbZT8h1Dj4bcXENjbrgOY05DbgCCtydd7r+d/JfebZVXB7sd5xlA2zWC6I9epsc7f8BnWOvqW6sPoTbb4Ph3bY4BrT6kcZmcpstELphB/NHtxL24FI9AX8lsT30Itddc9+vZgO3ZkJ1dX8YnNpLAkTcRf+BZBn/wJP6DYyVyg6mRxC1y4hbhqXe96A0adRPGglKqM3IlcQG0ep1LfRtzivDUu55rT72uQX0XobWXCZk2jNqcIip/uCKZKVdsa8Kd2ilgyjG4zyON2rItrr0HxxF36DXiDr7CtQ3S23KdtBE6jE51bcoxoIlwPc+aiBCMOc7nw4DWXjea0CBM+cW27+YXo9EFtsiUK3/JzfbUh1BzrZ4tpS3XuLRlPUEjf8HQA5sZ8sEmyW35erDr1Nr2LFceaW8Ogbqc/Ty9X1zWqpwtB9vWjuvPr9FdHokIaTKu6xQ062aqv89wdI6lSpacLdO1n6Kfl9rccREEYU+bqYKU+ePdlBFFBA81PsP6cm3ddi7/+kH8p45u31ORhncK3B2aKGIuKOLi2Pmkz15NzuY/0u3FRFR+3q0AuffjWsRdGdt/qs6mcHHqClLmPEDYsjsQPDUSsY332egur1uuSPjauyj884dYq2qksSSw3RRyyzYXFJE6bh4Zc1aR9+wfidqxvnX13UxdNl/GVij7rUOcGLmS0wnrMeaVELPpHulsN54a3ZBq5rwIHio8gnw5PfNR0p96m75vrGszV0qMNSwS/subyfvgK4nMJncruW2pvLV0W3M7V7a+1yqmbLHdTNxIKVN9NoWUactJm7OOsKVzpbdlx77dfNaAL0g5xvZCb0T++omzbW3Zj1MzHyPtqbfp/8baG8auU9gvx0h+2tIebnvzSHtyCEDOWwdJHrmC0wmJGPOKid50b+dmS2I2f3694roR+cg9ZD6ys2VeQ3wnydluCrllXvdrv04gK2Kn/uusavbZoiAIIU1tAmY1873FwGKAJ0L7c2dAd5ft5txCNPrQ+oOICMWUZ3BTpgvmXAOoVaj8fbCUlGPKLaTqxHdYim2P3CuOncSrXyyVX51rzgpgvyOr71LP1Ydiyi9yW6YhF8BitP235rt0jJm5aKOjqDmf1iLXtl9Xzxp9KKa8BuycQjSRoZjsbLW/r4Ndp9q0LKzVNXjF9aBaAtuc04AboXM8YnY+Nq0+1OG5juszOI7AmTcT8fA81AG+iFYRsdaI4a/SJpZrz3kG5/pOw3glp8X61s+fjv6uKQCUn03DM7L+To6nXocx19W3MdvgcofJVsZ+J7qw1PF5zv8dpv/bDzfrNXL+dCLvtrHLzqbhGdU8u7YhO7KeXZtdROH+4zYfZ9LAakWjC3AM/HRW1/nTiLw7wc5NxytKR92Re+p11Nr36cqtb9aekSEOLoCgVhF2ywhOTG38rnND6efPIOIuG7v8bLpLfWv1IdS25FkfgjG3CK8eEXh1DyP+6HbHcQ85tJWzMx/BVFDSJF+u2Dbl2tqpg6vXuckjEtpyuvS2HHbvTLrcNRWAyrNpaJ3qWqPXYcpzPc/GHIPL3WCtXofRXjemwhI0YcG2py1hwZgMpbQkufKXHOyo+dMdbar8bDpeUaGUYnvlruk2VV/XXpE6R+zXZhdR4GjL6fa27I/J4H4Q8/VkQ317Tp7afP6SK49crxwCrjk79/8O0+/tzssGext1yiNad3kkx9BkXGsidETveYQra3dgvCLtqZbsOVumaz9FPy+19MSlADgJnHL6O2n/C2vqS6Io7hFFcZgoisMadloAqr9NQdszCk3XcNB4EHjreCqOHHcpU37kOIG/sjWwgJljqfz6WwAqPj+NZ5+eCF6eoFbhM2KAyyDN5lT9bQqePSPRdA1HsHPLDzfmBt9u4wY6cdUhAaCyVZemWzjanpGYMqW/SlN1LhXP6Ei03Wzs4NnjKEtyZZcdPkHI7ZMBCJo1hvKvbGxtt3BQ29lRXfCKicKYlSeN+22qq+fZ4yk7fKIB9zhBDs9jqLB7vnTnw1wct5CL4xZS+OePKNi5T3KnBdp3nl3rOwJtz0iMLdR3zpsHOT1lPaenrMfwSTLhd04AwD++N+byKoz5rgnVmF+CpbIa//jeAITfOYHCg8kALu90h84cQeWF5mMs+82DnExYz8mE9RQeSCZ8ro0dMLRptrmimoChdvbcCRR+YmMXHjhB0NgBAHjH6BE0Hm47LQBZbx7iRMJDnEh4iIIDyUTMHd8i11JR4+BGzB1PgZ0LEDx+AJWp2S6vgTSlnDc/4cyU9ZyZsh7DJycIu3MiYKtvS3kVpgZsU4P6DrtzIoaDyVRdyOR4//tIHr6M5OHLqM0xcGbag83+AIJ8sV11LhWtEzdo9njKkhpwk5zyyKwxVNjbsqara1v2jInCmJVPS8r/ywG+n7aO76eto/jgcXR3TALANz4OS1mV49WvOpnyi7FWVOMbHweA7o5JlBy0HWPJoWR0c23f182t/7wlz3LkLznY1948SHLCgyQnPEjBgRMubcrSZJuqdmlThZ+cBKDgQDLBY/sDzm256ZmXricbpLdnufLI9cohgMv4DN3MkVS1kLPlZENdXOvRdgtzxHVpgzxSmnQC3R2N41od4EvsWxvJfu5tKk9eaJF1vT23NWfLde2n6OellkZzXQISRFFsNHpVEIS2R4zFSu6Tu+j+1tMIKhUl7ydRm5pJlzV3U30+lYojxynZe4io5xPpdfQNLCXlZK22zQRkLaug6M//IvqDFwGRis9OUvFZcvM8J272ptfp+ZenEFQqivfZuGFr7qL6fCrlR05Q/N4hur7wAL2P7sFSWsFV+wxXviP6E7bmLkSLFSwWsje8hqW0olWesx7fTcxfNyGoVRTtPUxN6lUi1v2Oqm/TKDt8AsN7SfR4cR03HduNuaScKyu22djDbiJ62UYwmRFFkawNr2MpbnnaQYfnJ14n+q9P2qaM3XfY5nmt3fPhExS/l0S3F9cR9+luLKUVZDrNutQuteM8+wzvT5c1d4PFgmixkrPxNaytqO+iw6cJSRjC8G9esU31uKZ+KuX4w9s4PWU9AKkPvVE/HfLRs44pLKM3/h6//j0RRZHaqwWkrt/dKrYuYQgjj7+CpdrIxdX17GFHtnEyoZ79i5frppk8S5GdnfPup/xix1KGH3seq9HMhVXSpoE2HD5DaMIQRh9/CWu1kR9W73JsG3HkOceUpBce+qN9GlMNhiNnMRw56ygX/v9ubuVrJTYVHz5NSEI8w755FWt1LSlr6l9dGHJ4G2fs9Z320BuOqTWLj55p1XSljSRXbFusZD/+OjF/fRLUKor32rjhdm7Z4RMU7U2i2wvr6PPZbiwl9Vzf4X0JW3oHotkMVpFrG1933EGUqtIjpwicPJQBX+6yTYe87hXHtn6HXuD7abZXCy8/srt+OuRPT1N69DQAOa/9k16vJ9LltwkYrxWStmSbJM+y5C+Z2YbDZ9AlxDP6+MtYqo38uLo+rocf2eqYXvjiQ3/kppeX2ac2P4vB0ZaPctOOZYw4th3RaOZHiW35erABwv9fa18Tky+PtDeH1OVsRKi5mt+qnC0L22Ila+MeYt+2xbXhvSPUpNjj+nwaZUn2uN6xlr6fv465pJzLK2xPOELvnYW2p56IVXcSsepOANLv3oRZwtPT6+W5TZLr2q+TSuzEr2N1ZgnNzWojCMJy4AtRFBs9ixMEYaUoiq+4+ZqLfoi9RZYzYxWlvEvZMTJbZJnzALVKnkagVsk3s0dhpY8sXKuM+cYiz5waaAX5pp4M8q6VhSvnbJW1Jnlmq9d4/LSnGG2Limq85D6EGy7R7QCGjpeceUQu+WnlWWukwtjK8XPXUTrfatnYfdP3y3cB2ApN7DqlU/dcPss63CnrsdkrIFEUX2vYaREE4a/2bS12WhQpUqRIkSJFihQpUqToeqilwfkNV1MSgEmCIAQBiKI4p6MOTJEiRYoUKVKkSJGin6OsygKUbVJL7yJ0A74H/ohtEj4BGAY838HHpUiRIkWKFClSpEiRIkUOtfSy/FBsM4k9BpSKovgZUC2K4jFRFI919MEpUqRIkSJFihQpUqRIEbTwxEUURSvwoiAI++z/zWvpOw2VU+HXjsNruywyDTwEaN36tddP8gzlBZOMde2NPAM9VTLOBuIl0+DWWlGeSQEAtJ5ytSr55KGWZ9ILi1We86zRyDdou7uv9NmYrqfknERGJciTw2pq5Bsw7uUlzyD5qiqtLNzw4FbMfHqdlVcsz7Wfop+/JF3riqKYBcwVBOEWoHVzdypSpEiRIkWKFClSpMghZYRL29Sqm/SiKO4HpK9AqEiRIkWKFClSpEiRIkXXQfK9+6FIkSJFihQpUqRIkSJFEiXXsAgA4jbPQ5cwBEt1LT+u2kX5+YxGZfwHRttX59ViOHKGlMfeAsCvXw9+sW0RKk8NotnCxYf/RNmZdEncPpvvpYud+92qXZSfv+yW2//lpai9tBQcOcPFx/5i4/btTt9tC1H7elFztYBvl76KpUL6Qkt9ndjfrtpFmRt23CO/JmrueDRBvhyKmef4XKX1YOCrywkcGI2puIIzi1+i+mrBT8JzWMJgLNVGzjXhuc8jdzo8H4yZ7/g8ZNQv6Pv0Pfj37c6ZJS+T++8Tkrm9N89HlzAEa3UtP6zaSUUT8XWTffV6w5EzpD72JgD99qzBJzYSAI8AH8xlVY6VqqWo1+b56BLisVTXcmHVa27ZfgNj+MXLy+0rXp8mzc4GiLpvBlELZiKaLRgOn+bS03+TxI1+ZgHBCUOwVhtJXf0qlW64vgNj6F23EvKRM2Rs+LPL9silc4h+4h6O952PuUj6KuNy1LfP2KGEP3Y/qFSUvv8JRW/sc9kuaDREPPcAXv16YykpI3vdFszX8sFDTcQza/DqGwtqNWUfHqFoz17JXuXiAviOH0rExsUIahXF7x3CsLsBW+tB5PYH8O7fC0txOVmr/oDpWj4Ann16on9mBSo/HxBFMv7fGkSjtPf+/cbHE/nEIlCpKH4viYLX32/E7fr8Orz7x2IpKSdzxVZM1/LRRIURd3gntZeuAVB15iLZG3a6QzTtedxQwh5bgqBWUbLvIEV7Gta3B/ptiXj164WlpJzsNVswXcsnYPZEQhbe7ijn2Seay79cRe2PlyRxfcYOI/SR+23n6v0DlPzR9Vx5De1P6CP34xkXQ27is1Qe+g1g/VEAACAASURBVMKxzf+2KQTf/zsAil9/h/IPD7fKs8/YoYQ9utQRY8UN2LYYS8Szry3GctZtwZydBxoPwjetwqt/b7CK5D/7OtXJ37aCK49nv/Hx6B9fbIuvvYcodBdf29fh1d92jq+ufM4RX72Tdjniq/rsRbI3vCaZK6vnCfFEPb4I1CqK3kuiYFdjz91eqG9TV1ZsxZSVj/eg3nTdssJeSCBvxzuUHfymlZ7liS+AmGcWEGL/nbrYxO+U38AY4uy/U0VHznDJ/jvVPfFOIu5KwGSwjVq4vOUdio+caRW/s8iqvCzWJsn2xEWXMBjv6Ai+HrWaC4lv0GfrfW7L9dm6kB8T9/D1qNV4R0egmzwYgF6P30XG9vc5kfAQl7bupdfGuyRxQxMG4xut54tRa/gh8Q36bl3otlzfrffxQ+IbfDFqDb7RekLt3H4vLCH1mXf5euKD5P0nmZ7LZ0v23CVhMD7Reo6NWsN3iW/Qvwl2/qFTfDXjsUafd/3dJMwlFRwbtYaM3fvps/F3krhye/aNjuCzUWs5n/gG/Zs4z3mHTvPljA2NPq++Vsi51a+T/c8vJTMBdAlD8ImO4JtRq7iQuIc+TXjus3URFxJ3882oVfhERxBi9/z94h0kJzxIcsKDFOw/TsH+45LZIQlD8I7Wc3zUSlISdxO3dZHbcnFbF5GSuJvjo1biHa13sIPG9CN0xnCSJz1A8oR1XN3VcDkl9wpOGIJ3jJ7To1eSlvg6sc8tdlsu9rlFpCfu5vTolXjH6AmaPMSxTRupI2j8QGqypHWI6yRLfatUhD++nKxFG8m4dQn+t0xEG9vdpUjgHdOwllWQMf0+iv/yL7o8sAAA/xnjEDQaLs9ZxpXbVxH061l4RIVJMysX187Wb1pK5oInSJu+lMDZ49H26uZSJGjudCylFaRNXoThzX8R9pD9RoBaRdQLieRsfI1LM5dx5XcPI5olDohXqYh86n4y5m0iddpyAueMx7MBN/jOaVhKK0iZtITCP31IxMPzHNuMV3JJu2U1abesbnWnBZWK8CeWkbXocS7Nup+AWyegjXVlB9o9X5q6kKK3PqDLelt9l338GZdvW8nl21aSs/55TNfyJHdaUKnosmE52Us2kDl7Ef6zJqFpcJ7NOQXkP/o85fs/df1qoD8hy+4m6zeryfr1KkKW3Y0qoBWDlVUqwjYu59riDVyevZgANzEWcIfN8+UZCyj56wd0SVxgr4uZAFy5bSlZ9z1Cl4cWgSBxEgC5PKtURD65lMvznyBt+jICZ09wH19llaROXozhzx8S8dA8xzbjlVzSb11F+q2rWt1pkdNzlL1NpUxdTpCbNhVib1MXJy6h4E8fore3qZqLmaTOXkvqrNVk3PMEXTcvB3UrLufkii/qf6dOjl5JauLr9Grid6rXc4tITdzNSfvvVLDT79S1Pfs5M2U9Z6as/8l2WhS1XbJ1XLrMGE7uvs8BKDuVikeAL9qwIJcy2rAgPPy8KTuZCkDuvs/pMnO4baMIan9vwHaHtjavWCJ3GNl2bumpNDwCfJrkltq52fs+p8vMYQD49tJT/PWPABiOnSf8lhGSPYfPGMY1O7vEzvZswK7bVptf4vb7WXtt38/9+DihY/tJ4srreSjX9v3X4UvTSs/VVwsp/yET0dq6OxOhM4ZJii91U/HlpLA5o8n7QHrHKXTGcPL2HWuRbYvtFADy9h0jdKatXiPvnUbmK/9CNNpm0jIVSpsPI2T6cPL3fgZAxelUPAJ80DTgasKCUPv5UH7Kxs3f+xm6GfWeo5+ax+Wn34ZWLowlR317DYzDlJmNKSsXTGbK/3MMv4RRLmX8EkZT+i/bHdDyg//FZ7Sto4QoovLxArUKwUuLaDJhraiS5FUuLoD3oDiMV7IxXbWxS//9Of5TXNn+U0ZS+s8jAJQd+ALf0YNsxzQunpoLl6m9YLu7aSkpB6u0mct8BvXGeCUH09U8RJOZ0o8/J2DqSJcyAVNHUvIPG7f0wJf43TxIsq/m5DXQ1XPZ/s/xmzLapYxfwihKP7DX9ydf4DO6Mdv/1gmU/Vv6LP5eA/pgyszGbD/PFQc+w2+yK9ecnYcxJaNRPfqMGUrV16exlpZjLaug6uvT+IwdJp09sA+mzBxHjJX95xi+Ddh+k0dT9qFTjI2yxZhnbHeqvjkLgKWoFGtZhe3ueCf27D0ojlrn+Pr35/hPbRjXoyh2xNcX+F6v+JLJs89gW5sy2j2XfPw5AdMatKlpI+s9/6e+TYk1tWCxHYvgqUVsZb6WK74AdE6/U+Vt/J1S9L+tZjsugiCoBUFYIgjC04IgjGmwrfHt8VbIUx9MzTWD49+1OQY89SENyoRQm1NUXya7CE99MAApG/9C78fvZszp1+j1xO9J3/yuJK6XPsSFW5NThFcDrpc+hBonbk12fZmKC1l0mTEUgIjZI/GK0kniSmVL/b5osWIqr0YT4n9duB3pubodntsqT30INdcKHf9uOr6cji27cZmgUTdhLCilOiO3VexaSbHtVMaJ7RMbSeDIm4g/8CyDP3gS/8GxkrhavY7abGduEZ5613PlqddhdOIac4rQ2suETBuGMaeIqh+uSHTq6udG17dHeCimnPonQ+bcQjzCXf16hOkw59iPy2LFWl6FOiiA8oNfYK2qIfa/7xB79K8U/fmfWEulTR0qF9fG1mHKqa9nc24hmobsCF398dWxgwPQ9owCUaT7m08R/eFL6BbfjlTZ9lnPNeUa0ES4cjXhOoxOni3llaiDAwDQdgun1793EP33LfgM7yuZW7dfc27znjXhOsxuPDsrYNb4VnVc1OE6TLmu51kdFirpux7hofXHQ12MSPsu2OPHmZ3n5jw38Gwpr0QVFEDthUu2C2+1Co+ocDz79cYjooskrlyeNc4xC5hz3J9jd3ENtviK/fglot/dgs9waTf06iSb53AdpmynNpVjcO85232b8h4cR9yh14g7+ArXNux0dGQkHbdM8QWNf6eMTfxOufw+Ov1OAUQumEH80efp/eIyPAJ9JbM7m6yInfqvs6qlJy67gQmAAXhZEIQXnLb9qqkvCYKwWBCEk4IgnPx3dVPjTtw8Wmx418DN48e6Il3nTSXl8b/wZfxyUh//Cze9eH9zPppVo7sV7p562st8t+Z1us2fzqhDz6L288ZqbN/6Eq29U9LUcbWb20GeBTc7brdniWQ34AZF3JVx/WfYL8e06mlLU+xGlt3Gtq2Q4KHCI8iX0zMfJf2pt+n7xrq2YiWfZ5W3lq5rbidz63vSWNLgDYp0VH03vT+39YyI94A+YLWSPv4uLk2ZR8j8X6HpGtH5uZJeyXB/LgQPNT7D+nJt3XYu//pB/KeOln7Xupl4bfbYRBFzQREXxiwg7dY15DzzR7rtSETl5y2N28x+W3N8XgP7YK2u/f/snXd8FGX+x9+zyW4aqRuSbEKABIKnICU0pZNAAOvvFPAUT0FpUqSFpoKoKArYRYr97nfnCZye3s8TCMWCHtIjIJBCCZBGNr3uZnd+f8xmk012k91AmADz9uUL2H1mPvN55jvPM8/O83wHQ6oLg3K7de1k29VIe9pcbafqG5GiL7dTnXOZ9lveJWTpdCqP/o5ocnJKoJyem9rWwf6rL+dzetAk0u+dQ9bLHxL55lWIr2vhuZlxXVOm4mgKKQkzSbtvPiFPjUPwcOH9OHLFl8PdOn8/kvXpdg70n8Xh+EQMOQVErXjcaW2FG4OmFuf3E0WxO4AgCO8B7wuC8CXwMPZDCwBRFDcBmwB2hT5kjch2kxIIfzQegOKj6XhGaKl57ZeHTktVtu10r6p6v8h6hAdhsJTRjR9qXaif+80+bn1jmkMTkZMSiHg0zka3Bk9dUAPduk8bADzDg6i0lClPy+TwQ68A4B2to+3IXjRGh0kJRFq0C53QbozKrHw8I7RUZuUjuKlQ+3phLLD/a628nkdaPRcdPYNXhJYaNVc9u0LEpFHW+Co5mo5nRDBFnAYai6869RKupSq79qmT4KYi5O5+HBi5pEnt8EmjCH90BADFR9PwqFPfHjothjr7taftEa61xnZVZj55ljUeJUfSwGxGrfWzLkasS9ik0YROkDyXHk3HI1xLzXJ6D12QXd26v1xpLGU8O4Th0T6EnrvXWo+5547VJI9ZivFywyl8IG99g+VXQl3tL33uYcFU5+oblHHXBVOdkwduKlS+3pgLS/C9ZxhlPx2EahOm/CIqDv+OZ7cYaepEK9UFy9MGXe0vuu5hwRhz9HbKtKU6W2/VNhWWYMzOo3z/cUwFUhyV/nAQz66dKPsluWndLFtddZiW6hzb2DJm56HRBVt13Xx9pOlogMkg/Vl5PB1DRjYeURFUHEtzyrMxOw/3sHqecxtqu+vaUp2jt6nvGvzuHkLJt987pVeDKTsPdZjteTbVO8+OqM7Ow6tfd5ttK/Y7v4C5OifP5lds99Bgqut5rrZ6zrPWd43ny69uspaL/PsbGM9nOqUrl2djtt72mtLZO8d6u3ENDeNLExVBpZPxJZ/nPNThda4pndZuXKvDgzHauaZqqEq/iLmiEs8uHZy+pq51fOkmjSZsQm1f4RFu2wdVNdU/1unLjHm1L4vN/ttOuv51qVOeFW4cmnriYn3dqyiK1aIoTgWSgd2Ay69FvfjJDvbHL2Z//GIuf3eAsHFDAPDrHUN1STmGemscDLmFmEor8estzZ8MGzeEy9sOAFCVXUDAAGnKQeDgbpSfcdzxX/hkB/vil7Avfgm53x0k3KLr37uzQ93q0kr8e3cGIHzcEC5vOyhVSLBl+oEgED3vj1z4rPEMIuc/2cHe+CXsjV9CzncHibBoB1i07a3rcETu9kO0Gy9tH3Zvf/R7T7RSz0nsjV/K3vilFs+Dm+3ZFS59sr12gfd3+23iy+Qwvips4ivP4hkgcMjtlKVm2kxXdETmJ9s5GL+Qg/ELyfvuAKHjhlq1Hdd3rXbouKHkWWI777v9BAy6HQCvaB2C2t3uoAUg+5NtJI9YSPKIheRv20/I+GEAtImVdI31dI25hZjKKmgTK+mGjB9G/vYDlJ/K4EC3JznUdwaH+s6gKkvP0YRFDgctIG99A1QeS0HdIRx1RCio3fG9ayilu20z65Tu3of//0gDSt9RgynfJ92kV2ddxvsO6WmD4OWBZ48/YDhzoVXrAlT8loKmYwTqdpK2/z1DKN1lm8igZNev+D8g3ST4jRlE2X+lG6nSHw/jcUtHBE8PcFPh3e92qlKd0y7/LRWPjuGo24UiqN3xv3cIxTttM/wV7/yVgAclXf8xAym16LoF+YFK6mrUkaF4dAzHkOH81MvKYyloLNqo3fG7ewilu+rX96/4/9FS36MHUf7fOjePgoDvmMEUf/uj05oAlcdPo+4QgbvlPLcZM4yyPc5lbir/+RDeA3qj8muDyq8N3gN6U/7zIee1j51G3SHcqu1319AG2qV79uF3f8MYEzw9ELw8APAe0AvRZMKQnuGcrkyeK35LsY2ve4ZQsrNhXAda46s2ruvHl6ZjOEZX4ksmz+XJqda4FtTuBNw7hOKketdUUh3Pdw2k9BfJs7pdqHUxvjqiLR7RERgu5jpr+ZrHV9Yn26yL6fV1+infWKmvcNRP+dbpp/Tbpf6x7noY7Zj+lJ9yvv1sbYii2Kr/b60IjR2cIAj/C/yvKIrb6n0+GVgvimKTzybrPnGpzy2rniAorgfmCgO/z1lPSbKU7aXfrtfYH78YAN8e0ZZ0yGr0u46S8oyUPtW/3y10WTkRwd0Nc5WB04s/ouS32pR6JscPhPjDqkkEx/XEVFHFiTkbKLbo3rHrVfbFS7/2+vWIpts7T6Hy1JC36yinLLrtp4whclICALn/2U/qyoZraxqbSNXVom2uqOK3ORsosmgP2vUqey3atyx7hPAHBuIZFkhldgEX/7aH1LVbUXmo6fHeTPxu74ixsJQj096h4nxtY9XY47OW9GxspK5rPLeN6yGlgJ6zsY7nVeyNl34t+cOyRwh/YIDV84W/7SF17T/x7xlN70/mow7wwVxppCq3iB+HLrTu2wvHj6i7rHoSbVwPTBUGTs553xpffXettqba9e0Rza3vzLCkJD5KyjO1qYFvfXsGRYdSyfxLUoN9q5qYShCz6kmC4qQU0KfnrLNq99m1hoPxC63af3inJt3jUVKf+QiQUrv+4a2naNOtI2ZDNekv/JXCvcet+1YLjrWjV00mYLgUX2lz36c0WZqq2WPnGpJHSLptenSisyXNZOHuI5yx6Nal94H3SR612CYdcpXY+O8cLVnfEYH20zL7DOlLyDNTQeVG0T93kL/xH2hn/5nK4ymU7fkVQaNGt3ohHrd2wlRUQtb8VzFezEbw9kT3ynwpk44gUPTlDgo+/mej/q61rslov77bDOtD6HNTEVQqCrcmkff+F7Sd+ygVx1Ip3SVpR7yeiGfXaEyFJVycs1pa2A743z8c7fRxgEjp9wfJfe2TBvs3me3r+g7rjW65JR3ylp1cXreZkHkTqDiWSsnO/QgaNZFvzsfztmhMRaVkzF6N8UIOfqMHEDpvgjSdxGQm562/UbLrQIP9q9WOr2WfoX0IfWYauKko2roD/YYvCH76USqPp1K621LfaxLxvE2q78x5r1k9e/e7nbaJkzg/3vGUS3d3+9reQ/oSvGQ6gkpF8Vc7KNj4OUGzHqPyRArle/bh0a0LuneWo/LzRTQYqM4r4MJ9UpYk3wcSCJz6MAAFGz+n5KsdDfZvFh23nT5D+tJ26TRQqSj+sm6MpVK2Zx+CRk3Ya4vwuLUT5qISshaswngxG/fwUNp9+DKi2Ux1rp6c596kOrPhTa3KQTvS0p4rK+3fNrQZ1gfdsikIKhUFW5K4/P5mQuZa4muXFF/t3lhgja8LT79mja+QuRMQTWYwmch96++U7LafNt/T037q75b2XF6uafAZSNdUuCUdcsHmneSu20yo5Zoq3rkfwUMtpUPuGo2pULqmDBdyCPjjcEKeGotYXQ1mkZx3/kHxjoaDLU8vx6nOWzq+cgoc/7bdadVkAi39VEqdfqrXzjUcqdNP1aRDLth9hHRLP9Xl3dm06dYRRKi8kEvqwo0NBj6Ds7c6n+ZMRu4IH9Z6RwfAvszvW2U9NjpwsbuBIPxFFMXHnC3f2MClJWls4NLSXNmql+Yj10t5mhq4tCSNDVxakqYGLi1JYwOXlqSpgUtL4mjgciPjaODS4roOBi4tTWMDl5bG0cClpWls4NLSOBq4tDSOBi7XAkcDl5bG0cClpWls4NLSNDZwaWmUgcvVobUOXBq91xUEof4LJARguCAIAQCiKN7XUgemoKCgoKCgoKCgcCPSmjN3tWaa+pE+EjgBfIiUZkMA+gCvt/BxKSgoKCgoKCgoKCgoWGlqTkBv4BDwLFAkiuL3QIUoij+Iouh8QnwFBQUFBQUFBQUFBYUroNEnLqIomoE3BUHYYvkzp6lt6uOvrrqCw2s+ooxzhwWZ5g7LNTddrrnSIN8c8ZvRs0aQbw1CWZk8c8TlRM71D3JgrJZvDZVZvPniS64+0k3l/IsSrzYVVfKsr5HLc0GRtyy6AD5q+dbXKNzYODUIEUXxIjBOEIS7Aft5WRUUFBQUFBQUFBQUmkRU1rg0C5eenoii+C3wbQsdi4KCgoKCgoKCgoKCgl3kezavoKCgoKCgoKCgoKDgJHK9+gO/Yb1o/8JkcFOR93kS2eu+tPle0LgT9dZcvLt3orqghDNPrbW+GTZs5oMEPzwCTGYyln9A8Q9HXdZv/+KT+Mf1xlxRxdl571J+/EyDMt63RxP15tOoPDUU7T5ExnLpBUhuAW3otH4BHpEhVF3IJX36WkxFZa3ac4eXniQgLhZzRRXp896j/Jh9v53emm15MeFhzi+T/AbdcycRCx7CK6YdJ+5aTNlv6S5py1HX17PnwHsGEDH/ITxj2vH73Ysod0H7evV8JddUuxXSC9z0nyeR877tyxwFjTsd35qH1+2dMBWUcHbGGgwXc3EL8CV642K8e3RGv2U3F5dtcsmrXLo12s1pR9wCfOm0aRE+Fu2M5z64LnRrtOU6z3K12XJq30zX8s3sWY7+Qm7PrYnW/Hb61oxMq7lVtF85jZQ/v8iJ4bMJun8wnjHtbIoE/2kk1UWlHB/0FDkffEO7Z6R3XnrGtCPo/kGciJtNyqMv0P7l6aByzYZ/XCweUeEcGzSDc4vX02HVNLvlOqyazrnF6zk2aAYeUeH4D48FQDfzAYr3HuPYoJkU7z2GbuYDrdqzf1wsnlE6kgfO5OyiDUStmmq3XNSr0zi7aD3JA2fiGaXDf3gvAMpPZZA6eTUl+353WrOu9jWv6+vcc8WpDNKmvOay9vXsubnXVOTKaaQ99gIn42YReP9gPGMibYpo/zSS6sJSfh88ndwPvyHimccBEKsMZK79G5dWfuqyV9l0LdrNbUfEKgOZa/7OxZeaoS2XrkVbrvMsWz8lc39x01zLN7nna95fyOxZ4cZAloGLT88Yqs5lYcjIQTRWk//1XgIS+tuUCUjoh37LHgAKvv0F30HdLZ/3J//rvYiGagwXcqk6l4VPzxiX9ANG9UO/Vdp32eEU3Px9UIcE2pRRhwTi5utF2aHTAOi37iFgdL/a7S3Hpt+yh4DRtsfe2jwHjupH3tbvAShtwm/poRQA8rZ+T6DFV2XaJSrTM53Ws/EkQ13D9e25Mu1is7SvZ8/Nv6ayrddUwTc/4Z/Qz/a4EvqTv3U3AAXf/ozvQOmaMldUUXbgJOYqg8te5dKt1W5eO2KuqKL0wEnMVa5n+5FLt1ZbrvMsT5stp/bNdC3fzJ7l6C/k9qxwYyDLwEWjC8KQlWf9tyFbj0YXZFsmrE4ZkxlTcTnugb5ObdukfpgWQ6be+m9jlh51mO0+1GFBGLJqyxiy9GjCtNJ3wQEYcwukbXMLUGv9m9aU0bMmLIiqzDrbZ+rRhNnT1jdapjnIUdeS7vXrufm616/n5pxndZgWQx2/kqa2Xpmg2jImM6aSMtwCfV0z10p04craketRF+Srb1nbbFn7i5vnWq7h5vR87fsLuT23NsyIrfr/1kqjAxdBELwFQVgkCMJCQRA8BUGYKAjCN4IgrBYEoU3zZRvmj28w1U9wVMaJbV2Xb7ATwY6+60KNi14zz3b3KzZZhqsRuLLUNYpn6y5vYM9OaNotdKV25dJ1sF/n25HrUdeu9DWqbzn7KTn7Czuf3ajXsgv6N55nGfoLuT0r3BA0tTj/U+AC4IWUBvkksBa4F1gP/NneRoIgTAWmAiwN6MEDPh1tvjdk6dHogq3/1oRpMWbn2y1jzNKDmwo3P29MhSVObWuPkMfH0HbCSADKjqahCa8d5at1Wow5BXb0a8todFoMOZKOMa8QdUig9MtKSCBGfVGT+tfac+jE0TZ+PcKDKa3ZPtwJv+FaDNm2ZZxFrrq+UTy7wo3iuTnXlDFLjya89rqQNG3r0JgtlTFmW64pXx9MhSWum20FunBl7cj1qAvy1bcc/ZRc2jfjtXwzepazvwB5206FG4empop1EUVxATAT6ArMFkXxR2AR0MPRRqIobhJFsY8oin3qD1oAypJT8YzSoYkMQVC7E3T/IAqT9tuUKUzaj3bccAAC7x5Ayc/HrJ8H3T8IQeOOJjIEzygdZUdTmzSa+9l3nEiYz4mE+RRs/xXtWGnfPrFdMBWXWx/z1mDMLcBcWoFPbBcAtGOHU7hdOsbCHQesx6YdV/t5Y1xrzzmfbuP4yAUcH7mAgm37CR47DIA2jfg1lVbSxuI3eOwwCpzwZQ+56vpG8ewKN4rn5l5THh1rr6nA+wZTZOeaChobB0Dg3QMp+fm3ZnltDbo12s1tR65H3Rptuc7zte6n5NK+Ga/lm9GznP0FyNt2tkZEUWzV/7dWhMYOThCEo6Io9rT8/WNRFJ+o812yKIoOBy81HGz3P3YF/ON6E7niCVC5of9iJ1nvbiU88WHKktMoSjqA4KEm6u25eHeLxlRYQvqM1zFk5ACgmz0W7UMjwGQiY8VHFO853GD/omjvmWQt7V+eiv+wXlLqw/nvWlPOdt3xBicS5gPg3b1TberDPYetaTzdAn3pvCERTUQwhkt5pE1bg6mw1LpvQbBfpy3t2WR2PA7t+MoUq98z896zpi7slvQ6x0cuAMCneyeia9Ie7jnM+Wc/BCBwdH86rpyMu9YPU3EZZSfOcvqRl6z7Vjnwey3q2tzIeb5ePQeM7k+HlZNxD/LHVFxG+YmzpEx48Yb23NR5dlOZ7Wr6De9NuxVPIrip0H+xi+x3t6Bb8Ajlv6VRlLQfwUMtpda0XFNnZ661XlNdf9mEm683gtodU3EZaRNWUJl6oVGP11LX0Xm+knbk9v9uws3Xy6qd8sgKKlMvOuW5pXUbi6+Wru+WqGtn2uyWqu8r7SNv1Gu5sT7yRvRsNLk59Ast21+o3UyyeAaIvfB14zeArYReYQNb7+gAOJL9c6usx6YGLh8Cc0VRLK33eSfgM1EUBzUl4Gjg0tI0NXBpSRwNXFqaxhrllqSpG9qWpLGb+JZE8XxtcdTx38jIdZ7l4ma8puRErj5SzmtZrj5SLs9NDVxaEkcDl2uBMnC5OrTWgUuja1xEUZxc/zNBEP4iiuJjgiAMbrnDUlBQUFBQUFBQULgxac2Zu1ozjQ5cBEH4pv5HwHBBEAIs/76vRY5KQUFBQUFBQUFBQUGhDk1lFYsETgAfIiWkE4A+wOstfFwKCgoKCgoKCgoKCgpWmhq49AbmAM8CC0VRPCoIQoUoij84KyDXPFo55yx7a6pl0S2plGc+q5xzWeWawat2l2+editO9tFilFR6yKIr58oaL3d52hGDTPPi5VzjIldvIWfbefOtGgOTTPcFlUa1LLo+aqMsunBzrhtzFVGZKtYsmlrjYgbeFARhi+XPnKa2UVBQUFBQUFBQUFBQuNo4NQgRRfEiME4QvojAFQAAIABJREFUhLuB4pY9JAUFBQUFBQUFBQUFBVtcenoiiuK3wLctdCwKCgoKCgoKCgoKCgp2kXXaV/sXn8Q/rrf0wqd571J+/EyDMt63R9e+8Gn3ITKWfwRA4D0DiJj/EJ4x7fj97kXWl0U5Q4eXniQgLhZzRRXp896j/Jh93U41L13afZjzyyTdoHvuJGLBQ3jFtOPEXYutL2xyhjZDYgl/fgqoVBR8kcTlDVttvhc07rR7fT5e3TphKiwhY9ZqjJdyrd+rw9sSs2MduW9/Tt4HXzmtCxD10hMExMdirjCQNvddyo6dbVDGp3s0nd+aJXnedZizyz4GIHLRnwga1Q/MZoz6IlLnvIcxp6DB9vbwG9aLdiumgJsK/edJ5Lz/zwaeO741D6/bO2EqKOHsjDUYLubiFuBL9MbFePfojH7Lbi4u2+SSX7l0Qb7z3GZoLBHLJc/5XyRxeX1D3cg3anXPz1qN8aKtbpekdeS85Xp8yakdtfIJAuN7Ya4wkDrnPYexHfP2TFSeGgp2HeHsc1Jst1/0J4JG90U0mzHmFZM25z0MTsZ29MonCLLonnag26Z7NF0suvm7jnDGoltDxFP3Ef38Y/z3tklU55c4pes3rBftX5gMbiryPk8ie92XNt8LGnei3pqLd/dOVBeUcOaptRgsdR0280GCHx4BJjMZyz+g+IejTmnWIFc7AlK7HRgXi6mRdtunTrtdUKfdbr/sMQJH9sFsqKbqfA7p897FVFzutK4c/YWcbWdz4sstwJdOmxbhY9GteTljS/v1HdyDiCWPIWjcEQ3VXHz5U0p/OeaSNsgXX3K2I9dTfMGVt1+tDfPNuGD1KiDP25gA/7hYPKLCOTZoBucWr6fDqml2y3VYNZ1zi9dzbNAMPKLC8R8eC0DFqQzSprxGyb7fXdb1jNKRPHAmZxdtIGrVVLvlol6dxtlF60keOBPPKB3+w3sBUH4qg9TJq13WRaUi/MXpnJ24gtSEmfjfNwSPzpE2RQLHJ2AqKiVl+DTyPvqasCUTbb7XPTeZ0h8OuaYLBMTF4hmt48iAWaQvXE/0q/Y9R786lfSFGzgyYBae0ToC4iTPme9/TXL8fJJHJpKfdIjI+eOc9hy5chppj73AybhZBN4/GM8YW8/aP42kurCU3wdPJ/fDb4h45nEAxCoDmWv/xqWVn7rsVzZdi7Ys51mlIsKimzJyJgF2dIMsuqeHTePyR1+jq6cbvmwyJd+7Hl9yagfG98IrWsfhO2eTlriBTq/Zj+1Or00hPXEjh++cjVed2L70/tccjVtA8oiFFLgQ2zW6B++cTWriBjo70O382hRSEzdy0KIbaNEF0IRrCRzSncqLl503rFLRfuU0Uv78IieGzybo/sF4xrSzKRL8p5FUF5VyfNBT5HzwDe2eeQwAz5h2BN0/iBNxs0l59AXavzwdVM53AbK1IxZtrygdRy3tdnQj7faZRes5OnAmXlE6AiztdtGPySQPn8uxEfOpPJNJxOwHndKVs7+Qq+1sbnyJVQYy1/ydiy81T7e5fqvzi0l/4mVOjpzDuflv0/HteS7LyxVfcrYj11t8XWn7pXDjINtZDxjVD/3WPQCUHU7Bzd8HdUigTRl1SCBuvl6UHToNgH7rHgJG9wOgMu0ilemZLusGjupH3tbvAShtQrf0UAoAeVu/J3B0f4vupWbpeveIwXA+C+OFHERjNUX//hG/kf1tyviN7E/hP3cBUPTdz7QZ0KPOd3dguJBNZUqGy9pBo/tyeYuUCK70cCrufj6oQwJsyqhDAnDz9bZ6vrzlB4IsdW0qrbCWc/P2wNlEGD49Y6g6l40hQ/Jc8M1P+Cf0sykTkNCf/K27ASj49md8B3YHwFxRRdmBk5irDC77lUsX5DvP3j0lXYNFt/DfP+KXUE83oT8FNbr/qaebcAeGjGyqUl2PLzm1g0b1JXfz90BNbHvbj+023pRYYjt38/doR/cFbGNb5e3hdJYXbR3dkmboAnR6cSJnX/qrS2nipNjOssZ2/td7CahX1wEJ/dBvkdrWgm9/wXdQd8vn/cn/ei+ioRrDhVyqzmXh0zPGaW252hGQ2u3LLrbbl+u020U/JINJyqNVcigFjU7rtK4c/YW8bWfz4stcUUXpgZOYq1zPZHUlfitOnMWYkw9A5ekMVB5qBI1rk0nkii9525HrK76utP1SuHGQbeCiCdNiyNRb/23M0qMOC7Ipow4LwpBVW8aQpUcT5lyD4Fg3iKrMvNp9ZurR1NPV1Ne1U8ZV3MO0GLNqdY3ZetT1vKhDtRhqypjMmErKcAv0Q/DyoO30B8l9+/Nmadf3XJWlb9CwanS256Mqy9Zz+yWP0PvgRto+MISMNf9wSlcdpsVQR1c6x/U8hwXVlrF69nXaW2vSBfnOszpUi7G+59CGusYGnmt1c5oZX3Jqa3RaqmziNh+PerHtodPWa0fybeK//ZKH6XNoA20fHEzG6i+apWtwoFuVZXtsNbpBCX2oysqn7PfzTunV6gbVxg5gyNaj0dlrv+rUdXE57oG+Tm3bqLZM7YjV01Vqt0MejqNw92GndeXoL+Rqw64kvq6Eq+U34K4BlB8/i2hwLZW4bPElUztyPcbXlbZfrRGxlf/XWnF54CIIQspVUbaX4rveLwaCYKfQlc4JtLNPsf4+7ele6Ulsrq4oEjpvAnkff425vLKZ0k3Xo92M63WKZLz6dw71mcblL39EN2mMk8L29lm/Hu0dm3O7b3W6IN95duZacVAmbN4E8j5qfnzJq21vt/W17WxXp0zGq59zsPd0Lv/zJ3RPjHZO1pkYc1BG5aUhcu6DnHdykNTUThuGtqMyTmzbmLJc7Ygk3qS2M+12+NMPIlabyfvyx2brXpv+wt4ur0UbdiXxdVVlXfbr2SWSiGceI2Pp+83Qlye+ZGtHrsv4aoG4U7guafR5qiAIJdSGak3UeNd8Loqin4PtpgJTAZb69+SPPh0BCHl8DG0njASg7GgamvDaEb5ap22wUNNQ7xc9jU6LwfJI2BVCJ4620fUID6a0Zp/hTuiGazFkO7+I1B7VWXmodcHWf6vDtFTX82LMzkOjC6Y6Ww9uKtx8fTAVluDdswv+YwYQtmQibn4+iGYRscqA/i+OE7yFTRxN6IQRAJQmS55rlux56LQYsm21q7L0NufDw0Fd5321l1v/+gwX1jbdWBqz9GjC63jWaa2P9Gs9S2WM9TxfCXLpwrU/z3X3qa7vObehrtqOZ++eXfC/awC6pa7ryqEdNmk0oRPiASg9mo5HuLZObAc1jO3M+u1IwzIAeV/9xK3/+wwX1my2q6ubNJowi26JRbfuPqvs6Nb99bTm2Dw7hOHZPoTY3Wstn2vptWM1R8csxXi50KFvqGmbautaE6bFWE+3powxy1LXft6YCkuc2rY+crYjoRNHE2Jpt0uPptlc05pwbYMkCk2128HjhhE4og8nH3q+Uc+tob+Qqw27kvi6Eq7UrzpMS/QHSzk39y0M57Od0pQrvlpDO3I9xldz2i+FG5Omnrh8CvwLiBFF0VcURV8gw/J3u4MWAFEUN4mi2EcUxT41gxaA3M++40TCfE4kzKdg+69oxw4HwCe2C6bicoy5tg2FMbcAc2kFPrFdANCOHU7h9v0um8z5dBvHRy7g+MgFFGzbT/DYYQC0aUTXVFpJG4tu8NhhFDRDty7lv6Xi0TEcdbtQBLU7/vcOoXin7T6Ld/5KwINSg+Y/ZiCl//0NgDPjl3B68GROD55M3sffcPn9LU3eVGZ/uo3kkYnSQtjv9tN23FCL5xiqS8ox5to2bMbcQsylFbSJleaMth03lPxtBwDwjNJZywUm9KEi7ZJTnsuSU/HoqEMTGYKgdifwvsEUJdl6LkzaT9DYOGnfdw+k5OffnNp3a9SFa3+erbrJqWjq6AbcO4Tiep6Lk34lsEb3roGU/iLppo9fwqlBkzk1SNLNXee8rhza2Z9sI3nEQpJHLCR/235Cxg8DGo9tU1ltbIeMH0b+9prYDrOWCxrVt9HYzvpkG0dGLOTIiIXo6+j6xsZgakTXt46ufvsByk9l8Gu3JznQdwYH+s6gKkvPkYRFTd5sgBTbnlG1sR10/yAK7cS2dpzUtgbePYCSn49ZPw+6fxCCxh1NZAieUTrKjqY2qidnO5Lz6TaOjVzAMUu73dbFdrttnXbbf1gvwmf+kdMTV2GuaHxufmvoL+RsO5sbX1eq21y/bn4+dPpsGZmv/pWyg6ec1pQrvlpLO3K9xVdz2q/WjlkUW/X/rRWhwWPv+gUEoTewBmkA8x6QJopitLMCByL+6FCg/ctT8R/WS0qHPP9da0rjrjve4ETCfAC8u3eqTYe857A1xWLA6P50WDkZ9yB/TMVllJ84S8qEF637Not2Jy0A0PGVKVbdM/Pes6ao7Jb0OsdHLgDAp3snomvSW+45zPlnPwQgcHR/Oq6cjLvWD1NxGWUnznL6kZds9u/tYX9xou+w3uiWW9LkbtnJ5XWbCZk3gYpjqZTs3I+gURP55nw8b4vGVFRKxuzVGC/k2OwjZM7DmMsr7aaMLanUOPQc9cpkAof3wlRRRdq8dZQlS557JK0leWSi5LlHJ2IsaUwLdh/hrMXzLR8uxKtTOKJZpOriZc4s3mjzS6unu+P5xH7De9NuxZMIbir0X+wi+90t6BY8QvlvaRQl7UfwUEspF7tFYyos4ezMtRgyJM9df9mEm683gtodU3EZaRNWUJl6waHWtdRVu5sdarf0eXZ0yfoO6024JSVxwead5K7bTKhFt3in5Dnyjfl4dY3GVCjpGurphs59GFOZfd3GaGntkkoPh9rRqyYTMLwn5ooq0ua+T2lNbO9cQ/KIhQC06dGJzpZ0ooW7j3DmGSmN6S0fJuLVORwssZ2+aJNNbDs+y9Bp1WQCLbopdXR77VzDkTq6NWlMC3YfId2iW5e+B97nyKjFDdKYejm4rvzjehO54glQuaH/YidZ724lPPFhypLTKEo6gOChJurtuXhbYjt9xuvW2NbNHov2oRFgMpGx4iOK9zSci28wuTn03JLtiEpovC/q+MoUAiztdnqddvv2pNc5Vqfd7lSn3T5n0e758zoEDzXVBVIdlx5K4eySjdZ9O+4tWra/ULuZHOq2dBvmqI+8kvi6/b+bcPP1suqmPLKCytSLNvt3dJ6b6zfs6XGEzhxL1dnaJAhpE1ZQrS9qoNFYbLdkfBnNjn8jbsl2xEftOFHC9RhfzrRfAH0u/quxS7rVcGtIv9Y7OgBO5u5vlfXY5MAFQBAEFTALGAd0EkUx3FmBxgYuLUljA5eWxtHApaVpbODSkjQ2cLlRaWzg0tK04h9CWozGBi4tiXxn2fHApaVp7OauJWlq4NKSyNVbNDZwaWnk6iPlPM9yxXZjA5eWpLGBS0sj5z2YMnC5OrTWgYtTOQNFUTQD7wiCsAU43rKHpKCgoKCgoKCgoHDj0pozd7Vmmlqc/42djz1qPhdF8b4WOSoFBQUFBQUFBQUFhesSQRCCgC+AjsA5YLwoig0ylwiCsBq4G2ndfRIwR2xkOlhTT1zaAb8DHyJlFxOAvsDrLjtQUFBQUFBQUFBQULgZWALsEkXxVUEQllj+vbhuAUEQBgADge6Wj/YCQ4HvHe20qYFLH2AO8CywUBTFo4IgVIii+EOzLCgoKCgoKCgoKCjc5LTmzF1XifuBYZa/f4Y0GFlcr4wIeAIapIcjaiCHRmh04GJZ2/KmZW3Lm4Ig5DS1TX1MMi3QMsm0GA7kWyRfLcrjubLapZC4qpSZ5dH2NMq3qFauyJZzoXpbv3JZdOXsV0wmec60yiiPaU+NfEk+qozytCNuKvmuKrsvP7wGVMsU1wA+MiXOqTTIE1/ubvLFl6FankQIClePuu9ktLBJFMVNLuwiVBTFLABRFLMEQQipX0AUxf8KgrAHyEIauLwniuLJxnbq7OL8i8A4QRDuBopdOGgFBQUFBQUFBQUFhesIyyCl0YGKIAg7gTA7Xz3rjIYgCJ2BW5GWpgAkCYIwRBTFHx1t49LPAKIofgs4/2Y6BQUFBQUFBQUFBYUbDlEURzj6ThCEHEEQdJanLTog106xPwL7RFEstWzzHXAH4HDgIt8zWwUFBQUFBQUFBYWbELGV/3cV+AZ43PL3x4Gv7ZTJAIYKguAuCIIaaWH+lU8Vayk6vPQkgXGxmCxvqi0/dqZBGZ/bo61vqi3YfZjzy6Q3xrZf9hiBI/tgNlRTdT6H9HnvYip2fh581MonCIzvhbnCQOqc9yg7drahdvdoYmreVrvrCGef+9jm+/Cn7iPq+cf49bZJDd567VD3pScIiI/FXGEgbe67DnU7W948XbjrMGeXSbqRi/5E0Kh+YDZj1BeROuc9jDkNMsvZpdPKSQTFS3WdMmcdpXZ023SP5haL3/xdh0l/7hMAOiSOI2zCCIx6aZbg2VV/p2DXEad05fR8y8sTCY6X3vJ94un1lNjR9e0eRdd3ZuDmqSFv1xFOP/upVBddO3Drmim4eagRq02cXPIRxUfSnfYcvfIJgizxddpBfLXpHm19G3L+riOcscRX+8TxhE2It9b3ORfqW664lsuzz+DehDw7DcFNReGW7eRv2mLzvaB2R7cmEc+unTEVlpA5dxXGS7n43TuMoMkPWst53BLFuT8+TdXJhm2QI93Q5yy6m7ejr6+rcSd8dSKe3STdS3MkXUmrI2EvzcatjTeiWeTcA3MQDc7PvW8zJBbd8qmgUlGweQd5G7Y20G63dr5V+8Ls1zBeykUdEUJM0nqqzlwCoOLoaTKfW+e0rt+wXrRbMQXcVOg/TyLn/X820O341jy8bu+EqaCEszPWYLiYi1uAL9EbF+PdozP6Lbu5uMyVKdISPkN6E7ZsKoKbioIvdqDfaKe+1y7Aq1tnTAUlXHz6VZv61q2chaqNN4giZ/9nrtP17TtU8iy4qdD/w77nDm/Ow/v2TlQXlHBupuTZd3APwpc8hqB2RzRWc+nlTyn95ZhLntsMjSViuVTf+V8kcXl9w/Mc+cZ8vLp1wlRYwvlZqzFerP1BUx3eli5J68h563PyPvjKed0hsYQ/P0WKry+SuGwvvl6v1c2Ytdpa1zW6MTvWkfu2a7qy1rVMngHav/gk/nG9MVdUcXbeu5Qfb9gGed8eTdSbT6Py1FC0+xAZy6X7oMB7BhAx/yE8Y9rx+92LKP/N+T5KrvjyG9aL9i9MBjcVeZ8nkb3uywa6UW/Nxbu7dJ7PPLUWg0U3bOaDBD88AkxmMpZ/QPEPR53WVZCFV4HNgiA8iTRAGQcgCEIfYLooipOBrUAccAxpof42URT/3dhOZXviEhAXi1eUjqMDZ3J20QaiV021Wy7q1WmcWbSeowNn4hWlI2B4LwCKfkwmefhcjo2YT+WZTCJmP2h3e3sExvfCK1rH4Ttnk5a4gU6v2dfu9NoU0hM3cvjO2XhF6wiI62X9ThOuJWBIdyovXnbJs2e0jiMDZpG+cD3Rr9rXjX51KukLN3BkwCw86+hmvv81yfHzSR6ZSH7SISLnj3PJ74E7Z5OauJHOr02xW67za1NISdzIAYvfwLie1u8ubfo/Do9YyOERC10atMjlOTi+J95RYfx8xxxOJn7ArauftFvu1tWTOZm4iZ/vmIN3VBhai+cuyydwZu1W9sUvJn31ZmKWTXDac019H7xzNqmJG+jsIL46vzaF1MSNHLTWd218Xdr0LUdGLOSIC/UtV1zX1b6mnlUqQp+fwcUpyzlz13T87hmKplOkTRH/caMwFZVyZuRk8j/9irYLnwCg+N/fc+7+2Zy7fzZZC1/HeCnH6UELKhVhK2ZwYfJy0sdYdDvb6gaMHYWpuJT0EZPJ/+QrQiy6uKkIX7uQ7OXvceaup8h4dDFitQuJHlQqwl94inOTnidt1Az87x2KRz3twPEJmIrLSI2biv7jrwlbPNH6neF8Nun3PE36PU+7NGhBpSJy5TTSHnuBk3GzCLx/MJ4xtrraP42kurCU3wdPJ/fDb4h4RvqhTawykLn2b1xa+anzevW0dSueIuOJ50kb9RT+9w5pWN+W85wWNwX9J/8iZPEk6Qs3FRFvJJK1bB1nxszg/CNLnK9vi+f0x1/gZPwsAu+z4/mhkZiKSvl9iOQ5fKnkuTq/mPQnXuZUwhzOz3ubDm/Nc9lzxIvTOTtxBSkjZxJw35AG5zlofAKmolJOD5vG5Y++Rrdkos334csmU/L9IZd1wy26qQkz8bejG2jRTRk+jbyPviasnq7uucmU/uC6rpx1LYtnwD8uFo+ocI4NmsG5xevpsGqa3XIdVk3n3OL1HBs0A4+ocPyHxwJQcSqDtCmvUbLvd9eEZYyv9iunkfLnFzkxfDZB9w/GM6adTZHgP42kuqiU44OeIueDb2j3zGMAeMa0I+j+QZyIm03Koy/Q/uXpoFImDbVmRFHUi6IYL4pijOXPfMvnBy2DFkRRNImiOE0UxVtFUbxNFMX5Te1XtrMeOKofl7d+D0Dp4RTc/H1QhwTalFGHBOLm60XpoRQALm/9nsDR/QEo+iEZTFLGjJJDKWh0Wqe1g0b1JXdzjXYq7n7eqEMC6mkH4NbGmxKLdu7m79GO7mv9PurFiZx76a8upR0KGt2Xy1t+qKPrY1/X17vW85YfCBrdDwBTaYW1nJu3B84+yQse1ZeczZJuiUVXU09XExKAexsvq9+czT8QbNG9EuTy3HZ0X7K2SFMkiw417rnoYCoAWVt+JGSMdI5FEdx9vQBw9/OmysmnPADaOvFV0sz4ag5yxTXI49mzexcM5zMxXsgGYzXF3/5ImxF32pRpE38HRV/tlI5r21687+zRYD++9wyl+P+cz/DuZUfXN76e7og7KPpS0i2uo+szKJaq02epOiU9jTIVloDZ+cw/Xj26UHU+C+OFHERjNUX/9yO+I++w9TPiDgr+uQuAou/24jOgoWdX8ekZQ9W5bAwZkm7BNz/hn2DbPgQk9Cd/624ACr79Gd+BUlp+c0UVZQdOYq4yNEvbq4dtfRf934/4jqjvuT9FX0qei7/bi4+lvtsMjqXy1Llm1bd3fc//bujZP6E/eovnwv/Ueq44cZbqnHwAKlMyUHmoETTOT3Dw7hmD4XwWBst5Lvz3j/gl9Lcp45fQv/Y8/+dn2tQ5z34Jd2DIyKYqNcNpTQDvHpKuNb7+/SN+I+vpjuxPoTW+6umOvAPDhWwqU1zUlbOuZfIMEDCqH/qtewAoa+I+qOzQaQD0W/cQYOkfK9MuUpme6bKuXPEltSNZ1vOc//VeAurpBiT0Q79FqpOCb3/Bd1B3y+f9yf96L6KhGsOFXKrOZeHTM8Zl760Jsyi26v9bK40OXARB6F7n72pBEJ4TBOEbQRBeEQTB+0qENWFBGDLzrP82ZOrRhAU1LJOlb7QMQMjDcRTuPuy8tk5LVWbtfquy8vGoN/Dx0GlttbPyrYOjoIQ+GLLyKf/9vNOaIPmpquO5KkvfYMCl0Wkx2Bybref2Sx6h98GNtH1gCBlr/uGcri6onl89Gl1QwzJZjsuEPzGa2N1r6fLmU7j7+zilC/J59tAFUnmpdp+VWXo863n21AVRmZVfWyYzHw+d1GmkLPuMmOWPMvjwOmKe/zNpL3/ulG6Nn7r1bXAQX7b1nW9TL1J9v07MmzOcrm+54tqe9rXwrA7VUp1dG1vV2XmoQ7UNy2RZnh6ZzJhLynEL9LMp43fXEJcGLu5hWqqzanWN2Xm419N1D9VizK6jWyrpaqIiQITIj18i6l/vEDRlrNO6AOowLcas2qdh1Vn2PRsdeNZEhtLp328T9fkqvPt2dUm3bnttzNKjDqunW7dNN5kxlZThFujrij27uIdqMWY1fp7dw+x71nSMAFGk/ScvEvX122inOv9kXlPPsyFL37Cuw4IwNuE54K4BVJw4i2hwPt2zOlRbu18s9W3vPDfQ9kPw8qDt9AfJedv5NqsGqR7rxrad8xyqxZDlWDe3Gbpy1rVcnqHGd22bKF1Xtv2Uuv59UJYeTZjzP9TaQ6740uiCausRMGTbuRcJC7Kt6+Jy3AN9ndpW4eagqScun9b5+6tAZ+B1wAvY4GgjQRCmCoJwUBCEg/8qbzjP3VKo4Wf1R3h2E83blgl/+kHEajN5XzpMQGBH2550fW170iIqLw3t5j5IxuovnNer2aUTnu2m1q9TJOPVv3OozzQuf/kjukljnBVudJ+Oy0iFMj/dwf7+szkcvxBDTiHRKx5zThcZPdvZa8Nz7Lhe2k0cScryz/gpdiYpyz/jtjenO6nrKGydiy+ArE+3c6D/LA7HJ2LIKSBqxeN2CtsTtrfLlo9rkMlzM9uQunXi2f0WzBVVGFJdGaw5EdMOjk1wc8Or921kLljDuT8txHfknXafArmEk/VcfTmf04MmkX7vHLJe/pDINxNRtfFyTsOZ82u3XpzbfePazrxwxEF9u7vh3ec2Ls1fy7mHFuE78k7nn0A5FdONe/bsEkn40sfIWPq+c5qN7tcZbZGweRPI++hrzOWVrmk62Kdz7aZI6LwJ5H3cXF07n8lY19fEMzjl25k+1HVdmeLLbr/c9LFJZZzYVuGmoKnnqXUjJR7oK4qiURCEH4FkRxvVzf28L/wBa2iFThxNyISRAJQeTUMTHmzdRhOuxVBvOo6h3q/zmnAthuzaMsHjhhE4og8nH3q+CRsQNmk0oRPiLdrpeIRrqVl27KELwpCdb1O+KrOetqWMZ4cwPNqH0HP3Wsu2WnruWE3ymKUYLxc21J04mtAJUra40uQ0PMKD6+hqG+pm6dGE1+p66LQYcmzLAOR9tZdb//oMF9bav9HUTRqFzqJbcjQNj/r7rKdryNTb/EIulZHq2phXZP0862876fbXJXY15fbcblIC7R6VznHR0XQ8I2r36anTUpVtG1+VmbZPYTzDg6xldOOHWhfq53yzj9vesD/3uAbdpNGEWeKrxBI2nNDjAAAgAElEQVRfNWh0QVTZiS/b+q6Nwbr1nf23nXT961KHunLFtZyeazBm5+EeVtuGuIcFY8zNb1hG15bqHD24qVD5emMurE044Hf3EEq+/b5JrbpUZ+fhrqvVVYcFU21HVx3Wlupsi24bb0yFJVRn51F+4BimAikJQdkPB/Hs2pny/zpsTuvtV49a17bWs86eZ6mMVdtX0gYwGaQ/K4+nY8jIRhMVQeWxtKZ1s/Q27bVap8WY01BXEx6M0aLr5utj1b0SqrPzUOvqneccvZ0yDT0bs/Mo33/cWt+lPxzEs2snyn5pur4N9TxrdNqGdZ2lR+3AszpMS9SmpZyf9xaG89kueTZm56GuX9/2YsyOtnfPLvjfNQDd0om4+fkgmkXEKgP6vzT9JoPqLNu6VodprdOw6upqdMHWurbRHTOAsCWu68pZ19fac8jjY2hruQ8qO5pm0/dJ11UT90EO+kdXkCu+JC91znOYFmP9exFLGWOWRddPupad2fZ64ypl7rrpaOqJi78gCH8UBOFBwEMURSOAKP0c4XKN53y6jWMjF3Bs5AIKtu2n7dhhALSJ7YKpuBxjru0Fa8wtwFRaSZvYLgC0HTuMgu37pQMb1ovwmX/k9MRVmCuanjud/ck2kkcsJHnEQvK37SdkfI12DNUl5RhzbW/OjLmFmMoqaBMrzaEMGT+M/O0HKD+VwYFuT3Ko7wwO9Z1BVZaeowmLHN7cZX+6jeSRidLi8u/203bc0CZ1zaW1um3HDSV/2wEAPKN01nKBCX2oSLvk0G/WJ9uti+n12w4QOl7S9bXoGurpGix+fS26oeOHkrdd0q27NiR4TD/KTl1wqCun54uf7GBf/GL2xS/m8ncH0I0bAoB/b8eeq0sr8e8t6erGDeGyRbcqu4DAAbcBEDS4G+VnGu8Msz7ZZl1Yrq8TX76xMZgaiS/fOvGlt9R33bUh2jH9KW+kvuWKazk911B5LAVNx3DU7UJB7Y7f3UMo3bXPpkzp7l/x/6M0iPYdPYjy//5W+6Ug4DtmMMXfuvC0Fqiwo1tSX3fXr/g/IOn6jR5E+T5Jt/Snw3jeEoXg6QFuKrz7dsOQ5vw88YrfUvCwaAtqd/zvGULJzl9typTs+pXAB6UBpf+YQZRZPLsF+VkXs6ojQ9F0DMeY4dxNXllyKh4ddWgiQxDU7gTeN5iipP02ZQqT9hM0Ng6AwLsHUvLzb/Z25TIVv6Wg6RhhrW//e4ZQuquhZ/8HJM9+dTyX/ngYj1s61tZ3v9upSm06tgDKk1PxiKrj+d6GnouS9qO1eA64ayAlv1jq2s+HTp8uI/O1v1J28JTLnsuTU60xJqjdCbh3CMX1tIuT6pznuwZSatFOH7+EU4Mmc2rQZPI+/obcdVucuqkEKP8t1Ta+7h1C8c56ujt/JcAaXwMptdT1mfFLOD14MqcHS7qX33dBV866vsaecz/7jhMJ8zmRMJ+C7b+iHTscAJ9G7oPMpRX4WO6DtGOHU7h9f4P9uuRZpvgqS07Fs855Drp/EIV22hHtOKlOAu8eQMnPx6yfB90/CEHjjiYyBM8oHWVHU6+oHhSuT5p64vIjcJ/l7/sEQQgVRTFHEIQwIK+R7ZqkcNchAuJj6fnL+5gt6ZBruD3pdY6NXADA2SUbremQC/cctq5liXp5MoKHmlu/kJ62lB5K4eySjU5pF+w8TGB8LLH73sNcUUXa3NpHyz12riF5xEIAziz+gM6W1K2Fu4+4lE3Lru6uwwTExxL733WYKqpIm1eb1adH0lqSRyYCkL5kEzGW1MAFu49YPXd49lG8OoUjmkWqLl7mzGLn/ObvPExQfC/67ntXSlU7t1Y3ducaDlv8pi7+oDYd8u6jVr9Ry/5Mm24dEUWRqguXSV3onK6cnvN2HiE4vhcDf30bU4WB3+est353x67X2Be/GIBTiz+k6zszUHmqydt1lLxdUnrFkws2csvKiQjubpirDPye6HwK14KdhwmKj6WPJb5S6sRXr51rOGKp77TFH1hTAxfUia+a+kaEygu5Tte3XHEtm2eTmZwX1xP50UpwU1G0dQeGtAyCn36UyuOplO7+laIt29GtSSQ66UNMRSVkznvNurl3325UZ+dJi75dwWQm+4X1RH68UkqHXKM751Eqj0m6hVu2E742kU47P5TSIVt0zcWl6D/+iqgv30IURcp+OEjp9wdc0s5csYGOn72IoFJRsCWJqtQMQuZOoOJYKiW79lPwxQ7avbGAmN2bMBWVcuFpSdunXzdC5k5ANJnBZCLzuXWYikqd1r2wbBOd/3eFlK72i11UplxAt+ARyn9LoyhpP/p/JNHxrXnc9tMGTIUlnJ251rp511824ebrLd0kjepP2oQVVDo5gKip7/afvoSgUlG4VfLcdu6jVBxLpXTXrxRu3kHE64l03v0BpsISLs5Zba3v/I//RdRXbwIipd+7UN8mMxeXbaLTX209h81/hPJjaRQn7Uf/RRId3prHbT9uoLqwhHOzJM/Bj9+FpqOOsKfHE/b0eADSH11Btb6oMUUb7czlG4j+ywvgpqJg806qUjMInSed5+Kd+8nfnETkG/O55fuNmApLyZi92rl9N6X7/Aai/vKClBp4i6QbYtEt2bmfgi+SiHxzPl32bMRUdPV0Za1rOTwDRbsO4R/Xm9t/Xi+lQ57/rvW7rjve4ESClGDp3NKNtemQ9xymyNI/BozuT4eVk3EP8qfLX56j/MRZUia86JxnmeIrY9kHdPnb86ByQ//FTipTLhCe+DBlyWkUJR0g7x87iXp7Lt32rsdUWEL6jNcBqEy5QMG/f6br7vfAZOL8c5tcSmyicOMgNJjL2dQGgvAXURSdXuRQd6rYtcRkli9NniDI8/ivWpTHs7sgX+NRZpbnVUSeggspbK8yckW2nF1EWz/n39F0NZFzDrXJJM+ZNhjdZNH11Di/oPpqU2WUpx1xU8l3VTm1dKgFqJYprgHc3eSp70qDPPHloZbvmjJUy9OOAPS5+C+Zots1orQ9WvVcsbP65FZZj41eTYIgfGPn4zhBEAIARFG8z873CgoKCgoKCgoKCgoKV5WmfgaIBE4AHyKtaRGAvkiZxRQUFBQUFBQUFBQUFK4JTQ1cegNzgGeBhaIoHhUEoUIURedfeqCgoKCgoKCgoKCgYMWsZBVrFo0OXERRNANvCoKwxfJnTlPb1KfEpL6Cw2s+8s2iBS9BnnmlRa6dmquGp4wL5HQ+ZbLoajzkmzuskmkNlVGmtQ8A+SVOvm/kKuMmU12DfPPTiw0aWXSNJvniK9BfnjVU7u7ytZ2iKM/0dZNJvmnzhip5+sjianmuqQ4yxTVAjj6g6UIKCs3AqatYFMWLwDhBEO4Gilv2kBQUFBQUFBQUFBQUFGxx6ecHURS/BZxL2K2goKCgoKCgoKCg0ABXs/oqSMg5o0pBQUFBQUFBQUFBQcEp5JnwaeGWlycSHN8LU0UVJ55eT8mxsw3K+HaPous7M3Dz1JC36winn/0UgDZdO3Drmim4eagRq02cXPIRxUfSndLt8vJEtBbdk43o3vbODFSeGvS7jpBi0e22aQ7encIBcPfzprq4nP2Wlxk6Q9RLTxAQH4u5wkDa3Hcps6Pt0z2azpaXMRbuOszZZR8DELnoTwSN6gdmM0Z9Ealz3sOYU9Bge3t0Xfk4ofE9MVUYODpnPUXHzjUo84cl42k3bgjqAB++6zTJ+nn0tLtoP2E4YrWZKn0xyfM2UnHR+fePynGe2wyJRbd8qvRCsc07yNuw1eZ7QeNOu7Xz8ezWGVNhCRdmv4bxUi7qiBBiktZTdeYSABVHT5P53Dp7Eg7xHtSbkGeeApWKoq3bKPhws622Wk3Ya4l43BaDqbCYrPmrqM7MAbU7oSuexrNbDJhFcl/ZQMUB598+7j2oD8FLp4ObG8Vbv6Owni5qNaGvLsSjawzmwmKy579i1Q1ZMQePrpJu3qr1Lun6DO5N6HPTpJcxbt6OftMWW78ad8JXJ1rr+tKcVRgv5eJ33zC0kx+0lvO4JYqz//M0VSfPOK0NV3ZNdVj2GIEJfRAN1VSezyZt7nuYip2bF97hpScJiIu1vkC3/FjD4/a+Pbr2Bbq7D3N+2UcABN1zJxELHsIrph0n7lpM2W/OtV0AvkNjiXh+MoKbG/p/7CB3/T9tvhc07rR/Yx7et3emuqCY87PWYLiYi3ePGCJXzbQUEsh+63OKtu9zWheg88uT0MbHYqqo4tTT6yi1U9dtukfzh3dm4uapQb/rMGnPfmL9LuLJ0UQ8MQax2oR+52HOvPS/TmvLUd9yXcteA/ugXfwUgpuK4i+3UfTRFzbfe/a+He2i6Wi6RJO76BXKkn6yfhe2/mU8ut9K5ZHj5Mxa7rRmXe3gJdMR3Nwo/ud3FH5k69mzdzeCF0vaOQtfoSxpr/U73YaX8ej+ByqPnCB7pmvactU1yNtfyHU/0lR/4dm7G8FLp+PRJZrsxFco21F7nn3vH0Hg9EcAKNjwd0q+3umSZ7nu/RRuDGR74hIc3xPvqDB+vmMOJxM/4NbVT9otd+vqyZxM3MTPd8zBOyoMbVxPALosn8CZtVvZF7+Y9NWbiVk2wSldbXxPvKLC+O8dcziV+AG3ONC9xaL73zvm4FVH9/jUt9kfv5j98YvJ/XY/l7/d77TngLhYPKN1HBkwi/SF64l+dardctGvTiV94QaODJiFZ7SOgLheAGS+/zXJ8fNJHplIftIhIuePc0o3JL4nbaLD2H3nPJITP+D21+x7zt5xmJ/GPNfg86Lj5/hp1LP8ELeYrP/7lVuXPeKkY5nOs0pF+AtPcW7S86SNmoH/vUPx6BxpUyRwfAKm4jJS46ai//hrwhZPtH5nOJ9N+j1Pk37P0y53QqhUhCybyaWpz3Hu3qn43T0MTaf2NkX8xo7CVFTKudFPUPiXr2ib+AQA/uPGAHD+/qe4+ORS2i6e4vxb4lQq2j43k8xpz5Fx7xR87xqOur7ug6MwF5eSMXoShZ99iXaBdC78x0q6F/5nOpmTl6BdNNUl3bAVM7gweTnpY6bjd89QNPXqOmDsKEzFpaSPmEz+J18RslDyW/zN95y9bzZn75tNZuLrGC/luDxoudJrqvDHZI4Om0ty/Hwq0zNpN/sBp3T942LxjNKRPHAmZxdtIGqVfd2oV6dxdtF6kgfOxDNKh/9wSbf8VAapk1dTsu93l/yiUtHupWmcefwFTo2YSeB9Q/CIsa3voIdGYioq5eTQaVz+6Bt0Sx4HoOL0eU7fO5/Td80l/fEVtHtlBrg53wUExffCK0rHr3fMJiVxI11WT7FbrsvqKaQkbuTXO2bjFaUjyHItBwzsSvDovhwYvoADQ+dzYb2914TZR5b6lvFaDn52FtkznuXC/VNoM2YY6mhb3eqsXC4vW0vpf3Y32Lzw0y1cfqaZbzm3tCNZTz1Hxn1TaHPXcDval8l97nVK/7OnofYnW8hd2gxtueraov3/7J15fBRF+v/fnWRmcl8TkkxCgCQEfypyhEuRGwLBg/3ucuguroKCKIfcoK73hSgqrgfC6up+/a664LG668qt4rFyBbKIQg4CAXJP7mtmMtO/P3oymZnMlUDoLPbn9fKFma7udz9PVT1V1d1VJVd7IVt/xIf2oqW4nLIHX6Duc8d89osII3rhbZy7dSnnbrmP6IW34Rce6rPNcvX9FF0+km3g0iNzGMXb9gFQcziXgPAQ1LGOq1CoYyMJCA2i5lAuAMXb9hE7dRgg7WgdECatNBQQHozBxzcPPTKHUWLl1nrh1lq5Jdv20cPKtVfctGsp+eQ7X00mOnMY5duklaTrsyS2yomtio3EPyyY+sM5AJRv+5rozOEAmOubbOn8gzX4upJe/JQhnN0qPZGrzspDFR6MJrb9ih/VWXkYyqrb/a7/7ifMTUYAqg7nEaSL9g2MPPkcNLAfhjPFmM6WIppaqPnnPsIyrnVIEzbpWqo+2iPd1xffEjJyoM82eVLggCswFRZjOlcCphZq//U1IROuc0gTOuE6aq1PqOp2fEPwtVJA1qT2ovGHowCYK2uw1NZLTxF94V5zBabCIlqs3PovviLUBbfu77sAqN/ZxlWl9qLxhyNt3Lp6NP37+cQNGtAP45kiTGet9n6+j7CJTtxJ11LzsWRv7fZvCb6uva/DbxpL7T86vsr6hdapmq+zwSyt7FSXlYM6QesTN2rKcCo+/MrKzcE/IgRVbJQTNwr/sCAbt+LDr4jKHAFAc955mvOLOmxv8KA0DKeLMVrLdtU/viEiY4RDmoiMEVR+JHVoq//1HWHXS/4Wm402W/00ap/jR6tiModRavW199gp2Vy67Wtipkq+TrhjMoWv/B3RKK2WZqrwfZ0XOfwtV13W2NfllhYavviakPEjHdK0FJVizCmQAqSTmvcfxdLQudWknNn1X3zVzuZWtuhiNcmm/UexNDa1+92b5PI1yNteyNUf8aW9sJUxp3wOvn4Ijf/OwlJTh6W2nsZ/ZxE8amiHbJaj79cdZUHs1v91V8k2cNHoomg+r7f93VysJ9CpQxyoi6a5uLItTVElGp3UWOU8/BfSHrmN0Vmvkfbo78l7+v1OcQ3FejROXI0uGoMd12DHbVXktVdiLK+hqaDEJy6AOj4aQ1HbJ1aGYj1qnWNHSa3TYixyvD91fNv99br/dww5tJkevxlD4fMf+MQN1EXTbHfNpuLKdr72Vb1+N46yvdk+p5cjn1XxWkzF5ba/W4orUMU5+lkVZ5fGbMFS14h/VDgA6qQ4Uv/xMsnvryN42NU+2woQEKulpcSOXdqeHRCnpcWOba5rwC8yHMOJU1Lj4e9HQGIcmqvTCIjv4RPXP06LyZ5bUoF/bIxTmpi2NGYLFivXeNKJe5Xv3IB4LS3FbWXaVFJBgAt7Hbj1bb5uVfiNY6j9Z8cHLhejTrUq9taJVO090imusaj9NdXx0RiL9R7TdFRS2bbzd3EFqnht+zSt92YtX/5RYQAED+rHFbte5Yodf+TcH163DWR8kUYXjcGn2GmXpqgtTXBqAhEjriT9i2cY9MnjhA1K9Zkth7/lqssBsTFO3HL843wbUF+oXNkc4BRHLhX3Uvga5G0v5OqP+NJeuFNAXExbPljPDYjzvYzI1fdTdPnI4xwXQRAWAx+IolghCEJf4M/AAOAkME8UxWOdR7d/ldtuhQVXr3utSXrOySDnkb9Q9vkB4qZdy1Uv3UPWzKc6xW331MoF1zlJ3K9HUvrJ9z7w7C/rne3yBbddksJn36Pw2fdIXPJrdHOncnbD31yd4Qz2yvVFidNHETkwhe9//UQHzpIrn9tBvd0WiCIt5ZWcHDUXc3Udgf1T6f3GQ+RmLsRS7+NTRJdlxwd7Ean5eAfq1CR6bXuFlqIymo/+hGg2d5rb7pG6m3JQ+/EO1Cm9SNr2KiYrF1+5PtQnb+U+cOAVWJoMGHLP+Mj0/dpu7rCdaxKXTkc0m6n4aJ+vYBdY3/L5wtS5+NWKbTyaw8mMxWj69qTXC8uo/eowosHUaXa7MOLBL0KAHwGRIWRNfZCwwX256k8r2D9skY9oGfwtW112dclL9PTzIrUVF4N7SXztTpeqvZCrP3IhdeWCy2c36RMo+q+Vt8n594qi+Kr1/18GXhJF8RNBEMYBbwDXuzpJEIS7gbsBloYN4cYg6claz7mT6XnbRABqjuYTmNj2VCNQp8VQ4vjKr7nIcSQemBBtS6ObNdY2Wav0sx+46sUFbo3oOXcyCVZurZVbYz2mccG1f0oIoEmIxmiXRvD3I/bG4RzIeMAts1XxczKJmz0JgPrsPDQJMdTZsY0llQ7pDcV6h89VNDotxlLHNAAVn3zLle8+6Hbg0mduBr1mTwCg+ugpAu2uGaSLprmkY69XY0b3J23p//D9b57AYvS8MZ5c+dwqU4kela7tiVuALgZTWaXLNC0levD3wy8sGHO1lDNmo/Rv84/5GAtLUCcn0nwszysXrE8o7Z72BcTF0OLEbimpIEDXg5bSCvD3wz8sBIuVXf7sFlu6pPdexHTGt89bzCUVqOy58TGYy/ROacpRxffAbOX6hYVgqZG4Fes329Il/vUljGfO+2ZvSQUBuranbar49vaarPdm83Vom6+h9W3LVz7x4OLXqR4zxxE9aQjHZz3mkRs3J5MeszMAaDgqceutx9QJ2nYLZRid3v6oE7QOcaQzMpVUoLL3ty4Gk1N8MBVXoEqIwWT1t39YiIO/AQx557A0NRPYrzdNHsp2wtwpJNwm+br2aB6aRCc/Ovu6SI/GzmaNnc2GokoqPt8PQN2RPLBYUGnDMeldfzImt7/lqsvtuT0wl7VvA7pCLm0u13s4owu5l8DXcOnbCzn7I63ypb1wp5aSCoKGD3A4t+mA58UQ5O4TdFcpyyF3Tt4+FbMf2MSKovgJgCiKXwFh7k4SRXGLKIpDRVEc2jpoATj39k5+mLiWHyaupfyLg+hmjgEgYkgaLXWNGJ3mVxjLqmmpbyZiiPS9qm7mGMq3HwTAUFJF1MirAIge3Z/GU+5fkZ57e6dtAlv5FweJt3LDPXDN9c2EW7nxdlyAqDHX0JBb5PD61p1K3tlOdsYqaUL9FwfoMXMsAKHpEtvkxDaVVWOpbyI0XWL3mDmWSis7MFnXdg+Th9KU575zefrtXeyb9AD7Jj1AyfZDJM0aDUBkel9MdY0u57K4U3j/Pgx4fh4H79iA0Yfv0uXK51Y1/ScHTZ8EVD3jEFQBRNw0hrrd+x3S1O3ZT9R0KZBGTB1Fw7+lwOsfHQ5+UrVQJcWh7pOAqdD3zwGbj51E1TuBgMQ4UAUQfsNYGr50XLmp/ssfCP+V1BkMmzKaxh+kT++EQA1CkAaA4JGDEc1mjPmFvnF/PImqd6KNGzp1XDtuw5c/EPY/UkcwdPJoGve35wZdlw5mMyYfuU3HclBbfY0qgPAbx1C3x8nePfuJ+I1kb3jmKBp/sGvkBIHwqaOp/dzHNx1c3DoVOX4QiYv/h5/nPIvFOo/LnUrf2c6PGSv5MWMlVdsPEDNjnJXbD3NtI6Yyx8bXVFaFub6Z0HRpvlDMjHFU7fB9MQ9XaszORZOcgDpJKttRN4+mdpdj2a7dfYDo6dJDi8gbrqfue8nf6qQ422R8VWIPAlMSMZ4r9cgrensHhyau5tDE1VR8cZA4q689xc6W+iZb7IybOZYKq68rvjhA5KhrAAhK0SGoAtwOWkB+f8tVlw22uhwPAQGETB1Lw1f/7rQdHZHhx5OoelnjSIDrONIVksvXcOnbCzn7I63ypb1wp8bvDhM8cgh+4aH4hYcSPHIIjd8d9mqznH0CRZeXBE8jPkEQngYSgSeAW4FG4GNgIjBdFMWbvAF2xd3iFvD/1t2JdsJAzE1Gflq6idpsaUWha/es5wfrkn7hA1O4+o8L8QtUUbHnKCcflJbWjBx+BVc8NQchwB+LwcjPa9+i7j9tS+p5GpFdse5OoicMxGLl1lm5w/esty0lGDYwxbr8oAr9nqPkPNi2pOeVL99L7eFczv+v6yUAg/zcv5FIfmYeUeOlZQDzlr9GQ7a0jN/AXRvIzlgFQMjAVNKsS7dW7T1CwR/elO77zdUEpSYgWkQM58o5tXazwxNPvah2y+2/bi6x4wdibjJwdNlmaqw2j9m9jn2TpCc1Vz78OxJ/PZLA+CiaS6oofO9LcjZ8xLVbHyT8yl40W59wNp3Xc/CODbZrB4qev5PvynzWhTS4ZIaOG4ru4fkIfn5UbdtF+etbiV02m6ZjudTtOYCgVtHzxZUEXpWCuaaes/etx3S2lPDMkcQum41otoDZTNnG96jb277zo9a4z+OQMcPo8cAC8POj9uOdVG7+AO2S39P8Yy4NX/6AoFYRv34NmitTsdTUUbxyHaZzJQQkxNHzzacRLRZayvSUPvQSLUVl7a7vJ7iuUsFjhknLmPr5UfvJTqo2v0/04ttpPp5Do5Ubt34N6iv7Yqmuo2TVM7RYuQl/ehosIi1lesoeftEl12Tyd23v2KHE/cG6HPKHO9Fv+hsxS2+j+Vgu9Xv3I6hVJGxYReBVqdJyyMvXS5P5geDh1xC7ei6nZ65w60+Aqvogt8cupE4N/v5V/NQqWqqkp6Z1WTmcWtv29Nbfja8B+jwzn4hxg7E0GTi1/FXbErv9d73AjxkrJe6AVFJal+f9MoszVm5U5gj6PDWPAG045toGGo4XcPJ3TzpcX6NyXcbCxg8h8ZF5CP5+VG7dTemr24hf8Tsa/5NH7e4DCBoVvV9aQdDVKbRU10nLIZ8tJerX44hdOANMLYiiSOnLH1Czc3+761c3a9zanLbuLqInSMuqn1z6mi12Dt3zPIcmrpbub6C0HLJfoJrKPUfJfVBaklhQBfD/Nt5LaP8+WIwt5D/+LtXf/mi7dpC/5897utLfURGuJ7N3dV0OCHAdO4NGD0O7RloOue6THVT/6X2iFt2O4XgOjV/9gObqfsS9/Ch+YWGIRiPmikrO/VpaaU33zguok5MQgoOw1NRS/siLNH3fvmMpiq5X3goePQzt2nukpZg/2Un1Fid2/37Eb3wEv/BWdhVn/0diJ/zlBdTJPSV2dS1lj7zUjm02u+Z2ta8BjAbXH5l0dXtR2hDikgtd2x/prXX/gNJbe6Hp3w/dH9vyuaWiirPTpHwO+81kou7+LQBVm9+n7pOd7a5foG+/AFCrurJPAJBR+rcOLCsnnxKjru7Wr1zOVx3vln70OHABEARhLnAPkApogLPA34H1oijWeDoXPA9culJy7qzpaeDSlfI0cOlKeRu4dKXcDVy6Wp4GLl0tdwOXrpa7gculkKeBS1fK08Clq+Vu4NLV8jRw6Up5G7h0pdwNXLpa7gYul0LuBi5dLXcDl0shdwOXrpangUtXytPApavlaeDS1fpvGbjoIq/q1gOX4uqfuqUfvfbvRVF8WxTFEaIoxoiiGAYcFkXxQV8GLYoUKVKkSBJPV/4AACAASURBVJEiRYoUKVJ0MeRtVTFXO4RNaP1dFMVpXXJXihQpUqRIkSJFihQpUmQnb+9NewI/AW8iLUYnAMOAF7r4vhQpUqRIkSJFihQpuiwlduNNHruzvA1chgJLgT8Aq0VRPCoIQpMoij7vFNcjsOO76F4MyfX9LkC9USULt4efQRaunKpqDJSF698kX8CRq2z7+8n3PX540C+vbDcZ5IkjIQHyzK2Rs3xV1QTLwpVrvpqcEmS0ucUiz+xXuea91tXL0z4ChPn7uj+UIkUdk8eBiyiKFuAlQRC2Wf8t9XaOIkWKFClSpEiRIkWKFF1s+TQIEUXxHDBTEIQbAe8beShSpEiRIkWKFClSpMillA0oO6cOvT0RRfFz4PMuuhdFihQpUqRIkSJFihQpcinZPvsKG5tO4qPzEPz90X+wk7JNHzkcF9QB9HpxOcHX9KWlqlbaRO1cGcED00hat8iaSKBk4/vU7PB9Z9+wsYPp+dh8BH8/9B/sovT19tzeLy0n+JpUWqrqOL1I4oaNHkjC/bcjqAIQTS2cf/od6r8/1mG7U566k+iJg7E0GTm59FUajhW0SxM6IIV+L7du4HaEUw/92eF44r3TSHn0dv591VxaKut84vZ+8i6iJqRjbjKQv/xVGo+dapcm5JoUUq2bt1XtzeLMw9LGcb0evp2ojKFYjC0YzpSSv/wVzLW+73sgF/tCfN1r1SziZ0+07ex9et17VO054rPNyU/eSeTEdCxNRvKWveKSHTIghb7WTRGr92RR8LDETlpzK9FThoPFgklfQ+7SVzGVVrU73yX3qTuJstqc68bmkAEppFltrtpzhIJWm9fcSnTmMESLBVNFLXlLX8XoIxfkyefQMenoHrkb/Pyo2rqTijc+dDguqAPouWEFgf37Yq6u4+yS9ZjOl6FKjCVt1yYMp84D0HT0JEUPveazrXJxW9XribuImDAES5OBguWv0Phje18HX5NC8kv34ReopmbvYQofkXwdddNIElfcQmBaT366cQ2N1o0cuzM3fJwUt/H3Q/++67jdZ+Nygq5JxVxVR8FCKW77R4aRsnktwQP7ot+2l3MPb3FDcK/eT95F5IR0LB7KdbBdua62K9fRN11H4spbCErryfEb1to2zfRVF+Jv/8hQUjetRJMUi+FsGfn3bMBc49u+V3Lkc/i4wfR6fB74+1Hx/i5KXvvY4bigDiB54zKCB0ht86l7N2A8J200Gb9oOjG/nQRmC4WP/Inar4/6xLSXXO3UhbQVvR++najJQxGNLTSfKSFv2as+c+XsC8nZH1H03y95Zqr5+dHzyQWcuuNxTkxaRNS0MWjSkhySRN+Sgbmmnp/HLqD8rc/Q3X8HAE0nz3Dy5hWcvGEZ+Xc8Rs9nFoK/j2b4+ZH01ALy73icnycuJmraaAKduFor96cx91D25mckPCBxWypryb/zaU5MXsqZ5S/Te+PyDpsdNXEwQSk6Dl23hNxVb9B3/d0u0/VdP5/cVZs5dN0SglJ0RE0YbDumTtASNWYAzefKfeZGTkgnKFnH0esXUbDmDVLWueYmP7uAU2s2cfT6RQQl64gcL3Fr9mWTPX4ZxyatoPlUEYlLpnd79sXw9fktn3Nk0mqOTFrdoUFL5IR0AlN0HBm5mPzVm0h51jU75dm7yV/9BkdGLiYwRUeklV30+qdkT1xBdsYqKncdJmnFzA7ZnHXdEvJWvUGqG5tT188nf9Vmsqw2t3LPv/4pRyesJHvSaqo6wG21+ZLns58fCY/fy+m5j5I3ZSERN49F09exPkfNmoy5toHcCXej//OnxK+dYztmPFNC/k33kX/TfR0bPMjFtSpiQjqa5ASOjVrI6bWb6L1ugct0vdfdw+m1mzg2aiGa5AQixqcD0HSikLz566n74af/Cm5r3M67/XF+nrCYqF+5iNu3ZtBSXc9Po6W4nfigFLdFg5GiDX/l/FPvdIxpVcSEdAKTdWRby3Wyh3JdsGYT2dcvIjBZR4S1XDeeKCR33nMdt5kL97du0W+o/fYYx0YtovbbY+gW/eaScDuVz35+9HpqATm/f4Lj45cQ/avRBKb1dEgSc2sGLTX1/DjqXkr/9Bk9H7wdgMC0nkT/ahTHJywh57bH6fX0PeDXsa6NXO3UhbYV1fuyOTpuGdkTV9CcX0TPJb7lsZx9ITn7I4ouD8kycAkelIbhdDHGs6WIphaq/vENERkjHNJEZIyg8qO9AFT/6zvCrh8IgNhsBLO0+oyfRk1HVpOTuCUYC+24k4c7ciePQP+hPXcAAE3HC2gprQSgOacQP40KQd2xF1baKcMo2/oVAHVZuQSEB6OKddxdVhUbiX9oMHWHcwAo2/oV2sxhtuOpT8yh4Ml3oQPfRkZNGU75hxK3PisH/4gQVLFRTtwo/MOCqLdyyz/8iqhMKU9qvs62+bzucA5qnbbbsy+Grzur6MxhlG+TFt6rz8olIDzENTssuM3mbV8TnSmVRXN920p8/sEan8t4tJ3N9Z2w2Z7rF6zp0FKNcuRz0MB+GM4UY7LGkZp/7iMs41qHNGGTrqXqoz0S44tvCRk50Gebuhu3VZFThqP/8EsAGrz4uuHwSQD0H35JpLV8Needozm/6L+GG+Ictz9rH7cjJ4+g0hq3qz5vi9uWJgMNB3/GYjB2mAtSua7oYLmusCvXzXnnO2UzXLi/I6cMR79NOl+/7UsiMx3b2K7idiafQ1r7BNY8rvz0WyInO95v5OQ2e6o+/56wUQOsv4+g8tNvEY0tGM+WYThdTMigtA7x5WqnLrStcOBm5aBO8I0rZ19Izv5Id5MFsVv/110ly8BFFa/FVFxh+9tUXIEqXts+TZE1jdmCua4B/6gwAIIH9eOKXa9yxY4/cu4Pr9sKsTep47UYi9q4xmI9qjhnbrRbbqsibxhJ0/ECRGPHljhU67QYivR2/Eo0TpVOo9NiKG5LYyiutFXM6MlDMRRX0vDTmY5x46Md7S7So46Pbp/GjusqDUDsbydQvTer27Mv1NcACXdmkr73BdJeWkhARIhP3FZ7DHY2G4r17YKrWqfFWKR3TGNnc6/7f8eQQ5vp8ZsxFD7/gW9cJ5sNbmx28LWTzb3u/y1DD79Bj+mjKXzubz5xQZ58luJI25vHluKK9vU5zi6N2YKlrhH/qHDpfpLiSP3HyyS/v47gYVd7N1JmbqukONbmR1OxHpWTH1XOvi7Wo46/sAZeLq7KKW5L3PZx2+glbndGznX5Qsp1x9kX5m9VTCSmMulTT1NZFSptxCXhdkZqXTRGuz6BsUSPWufKz3Z5XNtIQFSYT+d65cvVTl2EtsLGvXUiVXt9+zJAzr6QnP0RRZeHPA5cBEH4WBCE2wRBCL24WBf7UDi/QRBcpZH+aTyaw8mMxeRMW0nswhkIGh/3O3C1/UUHuACB/ZJIeOB2Ch943Teml0u357tO4xekJmnZdM50oDPpEeyL3U4j7oT7piO2WKj4eF+3Z1+IrwGK39nBwRGLyZq4CmNpFcmP3eETV2J7t9nlTix2SQqffY/DQxdQ/vE+dHOn+gh2hfXdZon7PoeG3EP5R9+guzPTNy7IW8Y8Ml2naSmv5OSoueTfvJTip98k6aVV+IUGdY55qbk+lG1fymCH1Y25rgv/hWGly7a/bvs65b1cd47t6rLdw98Xn+vKz85J3KXx4VyveLnaqQtvKwASl05HNJup+MjHuClnX6i7tBWK/mvl7f3eCMAC/FEQhN3A+8Dnoih6fO8uCMLdwN0AD0UPYHpob4fjppIKVLoY298qXQwm66tHW5riClQJMZhK9ODvh39YCOZqx4nohrxzWJqaCezXm6ZjeV5MsT4VSmjjqnVaTGXOXL1bripeS/KWBzizfCPGMyVeeQC6uZnEz54IQN3RfDR2r3LVumgMJY58Q5He4Sm5RheNsaSSwN7xBPaKJX3vBuvvWgbvfI6jUx/AVF7djhs3J5PY2RkA1B/Nc7Q7Qdtu0rXR6UmPOkGLsaQtTczMcURNGsrPtzzq1Wa52BfL1wCmihrb7yV/3c3V7z7gkR0/J5O42ZMkm7Pz0CTE0FpaNTqt7bo2drHe4bW+RqfF6FQHACo++ZYr332QsxtcD1jj52YSZ7W53mpzGze6PbfIydcu0kjcb7jy/x7k7PNb3dosZxkDMJXoUel62P4O0MW0r8/WNC3W+uwXFmyrz2aj9G/zj/kYC0tQJyfS7EMckYMbe8dUelh93XA0z6HsqHTados3tPO1m/LlTXJx7WVyitsSt72/1V7aC18VNyfTwWZNQgz11mPqBB9sdirXHdHF9LepohpVbJT0tiU2CpO+BneSO5+l69nFj3gtJqe41JrGVGzN43CpTvlyrivJFb8udlvRY+Y4oicN4fisx7za7GDLJewLyd1WdFcpyyF3Tt4+FSsTRXEG0Bv4BzAfOC8IwtuCIEx2d5IoiltEURwqiuJQ50ELQGN2LprkBNRJcQiqAKJuHk3trv0OaWp3HyB6+gQAIm+4nrrv/wNIn1m0TsZXJfYgMCUR47lSn4yVuDrUSbE2bs2uAw5panYdQDujPdc/PITUdx6maP27NBw64RMPoPjt7bYJ3vrtB4idNQ6AsPQ0zHWNmMocBx2msmrMDU2EpUvf6MbOGod+x0EaTxSyv/9dHBy2kIPDFmIo1nNk8hqXgxaA0ne2cyxjJccyVlK1/QA9Zkjc0PR+mGsbbZ8QtHGrMNc3E5reD4AeM8ZRtUPyTcS4wSQs+jUn56zD0uT9W3G52BfL14DDd8baqSNoPHHWI7vkne1kZ6ySJtR/cYAeM8dabU6jxQ3bUt9EqJXdY+ZYKrdL7MBknS1d1OShNOWdd899ezvZk1aTPWk1lXY2e+KaG9q4sbPGUbmjlRtvSxc9ZZhHLshbxgCa/pODpk8Cqp5SHIm4aQx1ux3jSN2e/URNlwZ2EVNH0fBva32ODrdN4FUlxaHuk4Cp0LeHEXJwy/7yBccnr+D45BVU7diPdsZ4AEI8+NpS30SI1dfaGeOp3nGg3XW7K9deDdm5aPrYxe1p7eN29a4DRFvjdtSN11P33X86zSt9Zzs/ZqzkR2u5julguY6xK9cd1cX0d/XOg2hnSudrZ3rOB7nzuSE7l0C7tjn6V6OodpHHrfZE3TiSuu+O2X6P/tUoBHUA6qRYApN1NBzN9cqUK35dzLYicvwgEhf/Dz/PedbnuAmXvi8kd1uh6PKS4GnEJwhCliiK6U6/RQOzgFmiKE7wBjjae5pLQNj4ISQ+Mg/B34/KrbspfXUb8St+R+N/8qjdfQBBo6L3SysIujqFluo6aTnks6VE/XocsQtngKkFURQpffkDanbub3d9UXT5gpXw8UNIfPQuaQnAv+1p4x7Lo3aXlbtxOcFW7unFGzAWlhK3ZCZxi2ZgKGibdJh/22O0uHiKVW90/+la6rp5RI0fhKXJQM6y16nPlpaKHLz7eY5MWg1A6MBU2xK9VXuPkP/gW+2uM+zg6xyZstZhOWSVn/u5Pn2emU/kuMG2JT1bl+W8ZtcLHMtYCUDIgNS2JT2/zOL0H94EYNB3ryFoVLRUSaz6wzkU3L/ZLetSsk0W92PvC/F1v1eWENq/D4jQfLaM3NWbHRoUf8Hzk5LkZ+YRNX4w5iYDectfo8HKHrhrA9kZqySbB6aSZl3ismrvEQqsNl/x5mqCUhMQLSKGc+WcWrvZ4Smcu7INkLJuHpFWm/PsbB64+3my7Wzua7W5eu8RTlltvuLNVQT1TQArN3/NFgeuv4fyBV2bz6GBrhuo0HFD0T08H8HPj6ptuyh/fSuxy2bTdCyXuj0HENQqer64ksCrUjDX1HP2vvWYzpYSnjmS2GWzEc0WMJsp2/gedXt973hdCm6TwX0c6fX03URYfV2w4hXbkrNX73yR45NXABA8ILVtudovsyh86E8ARGaOoPdT8wiIjsBc20Dj8QJyZj/hk91dyfVUvsLHD6HnY21xu+SVbehWSu1FjTVu99m4nKD+KZir6yhYJMVtgKu/34J/WDCCKgBzbQN5sx+jOdfxQYTJ7O+W3eeZ+TabT9mV6/67XuBHu3KdYleuz1jLdVTmCPo8NY8AbTjm2gYajhdw8ndP2q7t5yWOXIi//aPC6PvGKtSJMRjPV5C34HnM1fVuWReL6y2fBTc2R0wYQtJjd4KfP/q/7ab4lQ9JWPVbGrLzqNl1EEGjIvnlZQRb8zh/4Qu2PNYtmYH2lklgNlP42FvUful63kOLh/aiK+OXxUPMvpC2YvD3r+KnbuPWZeVwam3bkt9BKvdzT7q6L2Ro8VynurI/cm3Rx+4d3o0UE96vW79yqajN6ZZ+9DZw2SeK4pgLAbgbuHS1PHXuulqeBi5dKU8Dl8tVngYuXSlvA5eulFxl29vApSvlbuByOcvTwOVylJzly9PApSvlbeByOcrdwOVSyNPApSvlaeDSlfI0cOlqeRq4dLX+WwYu0WFp3ToAVNbldks/eqzFrgYtgiD8b9fdjiJFihQpUqRIkSJFihS1l8fJ+YIgfOb8EzBeEIRIAFEUp3XVjSlSpEiRIkWKFClSpEhRq7ytKpYEHAfeRFqLTgCGAi908X0pUqRIkSJFihQpUnRZSllVrHPy9sHnEOAw8AegRhTFr4AmURS/FkXx666+OUWKFClSpEiRIkWKFCkCL29cRFG0AC8JgrDN+m+pt3PaX0OeuT1yTgBUyzTJVB1gloUr50MDs0yTLc0yLv4gut6SrMslyGhzi1mefP4lTp5WyRZH5KxT8kiuxUXklJx1Sq4SFqqRZ3GReoNaFi7IuzCAostbPg1CRFE8B8wUBOFGoLZrb0mRIkWKFClSpEiRostXFtkemfx3q4NvT8TPgc+76F4UKVKkSJEiRYoUKVKkyKV+ee+pFSlSpEiRIkWKFClS9F+nDr1xuZgKGzuYno/Nl3Zt/WAXpa9/5HBcUAfQ+6XlBF+TSktVHacXPY/xXBlhoweScP/tCKoARFML559+h/rvj3WAm07io/MQ/P3Rf7CTsk3tub1eXE7wNX1pqarlzGKJGzwwjaR1i6yJBEo2vk/Njh86bHfvJ+8ickK6bcfYxmOn2qUJvialbcfYvVmceVja2Tz6putIXHkLQWk9OX7DWttus95tlsfXAOHjJDb+fujfd83us3E5QdekYq6qo2ChxPaPDCNl81qCB/ZFv20v5x7e4obgXslP3UnUxMFYmozkLn2VhmMF7dKEDEghzbqLfNWeIxQ89GcAeq25lejMYYgWC6aKWvKWvoqxtMpndspTdxJtZZ90ww4dkEI/K7tyzxFOWdmtSrx3GimP3s6/r5pLS2WdT9zUp+YSPTEdc5OBnKWvUe+Ge4WNm0X+Q28D0HvVTOJnT8Kkl74GLVj3HlV7jvhsc/KTdxI5MR1Lk5G8Za+49Xdf6w7Q1XuyKHhYsjlpza1ETxkOFgsmfQ25S1/F5IO/O1ufQ0cNcijbRc+8Q/33//HZ1tCx6SQ+IpXryr/tonzTh+24SS+uIKh/KubqOs4sfg7TuTKCBqbRc91iayKB0o3vUduJONLribuImDBE2tl8+Ss0/ug6jth2Nt97mMJHpDgSddNIElfcQmBaT366cY1tV3RvkjN2yhnDej95F1ETpDrlLmaH2MXsKruY3evh24nKGIrF2ILhTCn5y1/BXNvoM1uOOiUnV05fy9E2dzaOtEqV0IN+u16jdOP7VPzpE59tbZU8MVu+uqzo8pBMM179SHpqAfl3PM7PExcTNW00gWlJDkm0t2RgrqnnpzH3UPbmZyQ8cAcALZW15N/5NCcmL+XM8pfpvXF5h7g9n1zAqTse58SkRURNG4PGiRtt5f48dgHlb32G7n6J23TyDCdvXsHJG5aRf8dj9HxmIfh3zH0RE9IJTNaRff0iCta8QfK6u12mS352AQVrNpF9/SICk3VEjB8MQOOJQnLnPUfdDz91yGZZfG3Hzrv9cX6esJioX7lg35pBS3U9P42W2IkPSmzRYKRow185/9Q7HWNaFTVxMEEpOrKuW0LeqjdIXe/a16nr55O/ajNZ1y0hKEVH5ATJ1+df/5SjE1aSPWk1VbsOk7RiZofZh65bQu6qN+jrht13/XxyV23mkJUdZWUDqBO0RI0ZQPO58g5zD163hNxVm+m7fr5bbs6qzRy0cQfZjp3f8k+yJq0ma9LqDg1aIiekE5ii48jIxeSv3kTKs65tTnn2bvJXv8GRkYsJtPN30eufkj1xBdkZq6j01d8XUJ/NVbWcuvMpTk65j8IVG+n1UsfiSOIT91Aw5zFyMhYROW0Mmr5O3FmTMdfUc3LcAsrf+hTd/XMAaD5ZSO7Ny8m9YSkFtz9Kz6cXdSqOaJITODZqIafXbqL3ugUu0/Vedw+n127i2KiFaJITiBifDkDTiULy5q/vcByRLXbKGMMiJ6QTlKzjqDVmp3iI2afWbOLo9YsIStYRaY3ZNfuyyR6/jGOTVtB8qojEJdM7xL7kdUpmrly+lqtt7mwcaVXCw/Oo++qw70w7yRWzZeuPdEOJotit/+uukmXgEjwoDcPpEoyFpYimFqr+8Q0Rk4c7pImYPAL9h3sBqP7Xd4RdPwCApuMFtJRWAtCcU4ifRoWg9u3FkcQtxnjWjpsxwpGbMYLKj+y5AwEQm41gllYL89OoO7UMTdSU4VR8+BUA9Vk5+EeEoIqNckijio3CPyyI+sM5AFR8+BVRmdI9Nuedpzm/qENMuXwNEOLM/qw9O3LyCCqt7KrP29iWJgMNB3/GYujcaizRU4ZRtvUrAOqzcgkID0YVG+mQRhUbiX9oMHVWX5dt/Qpt5jAAzPVNtnR+wRrEDmS41o5d1wk2QOoTcyh48t0OLdkWM2UYpVu/tuOGoHbiqmMjCQgNsnFLt35NTObwdtfqqKIzh1G+TWLXW9kubQ4LtpXt8m1fE21l2/vbP1jjU/26kPrcdPwULWWdjyPGM23c6n/sI3yyIzd88giqPtoDQM2/viN0ZGscMdjiiKBRd6pxiJwyHP2HXwLQ4CWONBw+CYD+wy+JtPq6Oe9cJ+OIPLFTzhgWNWU45R2M2eV2Mbvm62yb7XWHc1DrtD6z5ahTcnLl9LVcbXNn44h07FqMhSUYcgs7xG2VfDFbnrqs6PKRx4GLIAgpgiD8WRCEpwRBCBUE4U+CIPwoCMI2QRD6dBaqjtdiLKqw/W0s1qOKcwwyqvhoTK1pzBbMdQ34R4U5pIm8YSRNxwsQjb4tu6eK12IqbuOaiitQxTtztW65wYP6ccWuV7lixx8594fXbUHSV6njozHY212kRx0f3S6NsVjvMU3HmPL4WrquI9tUrHfh7+i2NG7YnZFap8VQ1OZHQ3ElGqeGTKPTOvq6uNKhset1/28ZevgNekwfTeFzf+s02+iGbSh2vL9WdvTkoRiKK2n46YzPTIkb7WSzHrUuun2aYvdpEu7MJH3vBvq9dC8BESG+s53KtnRdR5vVOi1G5/uzK9u97v8dQw5tpsdvxlD4/AdemRdan1sVccNImo6f8j2OxNldE2u5dq5Tca644QAEDepHv52v0W/HK5x/qDNxxNGPUr1yzGeVcxwp1qOO970j5yw5Y6ecMUxtH5+4sJgd+9sJVO/N6hD7UtcpubndxdeXom2+kDgiBGnocc90Sl9+v9N8OfJZzrqs6PKRtzcu7wAHgXrgB+AEMBXYDvzZ3UmCINwtCMIhQRAOfVR/2kUCFyc5P3kUXCSySxLYL4mEB26n8IHXPVvgDdwBbuPRHE5mLCZn2kpiF85A0Kg6wHZ97XZPXF3xL2TJPNl87SPbZZ50DOMru72vXZxnl6bw2fc5NOQeyj/6Bt2dmb6jffK56zR+QWqSlk3nTAcGSh7B7dztvg4UvbOTAyOWkDVxNcbSalIeu70DaO91y+UeCnZJCp99j8NDF1D+8T50c6f6QvXK9Fq205JIuP8OznakbPtgq6c0TUdzyJm8iLxpK4i9d2Yn4oiL35x97cs9Xij0UsVOWWNYJ+12qngJ901HbLFQ8fG+DqDlqFPyceX0tTxtc+fjSPzy2VS89SmWxuYLwMuQz3LW5W4oiyh26/+6q7y9ZwsTRXETgCAIC0VRfMH6+1uCICx2d5IoiluALQBHev2qnfXGYj3qhBjb32qdFpP1s41WmYr1qBJiMJXowd8P/7AQzNXSBGVVvJbkLQ9wZvlGjGdKvFvZes2SClS6Nq5KF4Op1Jlb4ZbbKkPeOSxNzQT2603TsTyPzLg5mfSYnQFAw9E8NAkx1LfanaBtN5nN6PTUQ52gxVji+6RwZ8nl69br2rNVOm17f5dIaTz521fFz80kbvZEAOqP5qNJ0NJ6JY0uGmOJI9tQ5ORrF2kAKj75hiv/70HOPr/VLVs3N5N4K7vOyra/rsEF2/4tTOv9BfaOJ7BXLOl7N1h/1zJ453McnfoApvJqF9wp6GZPsnLzHLganbadPcZ23LbyZaqosf1e/Nfd9H/3frf2AsTPySTOyq7Plsp2m7/bsw3FetTO91fqyt/fcuW7D3J2g+fB24XWZ1W8lj5bHqRwxUaMhR2MI87l2rlOlfgQR/J9jyOxd0x1iCP2fpTqlZc44sbXvkqO2OlgyyWMYXFzMom1+rr+aJ4jO0HbbpEObzE7ZuY4oiYN5edbHvXKlqtOycWV09dyt80XEkeCB/Uj4oaR6B6Yg394CKJFRDQY0f+v590q5I7ZcvZHFF0+8vbGxSIIQj9BEIYBwYIgDAUQBKEv4N9ZaGN2LppkHeqkWARVAFE3j6Zm1wGHNDW7DqCdMQGAyBuup8664o9/eAip7zxM0fp3aTh0ohPcBNRJcTZu7a79Dmlqdx8genp7rjopzjahVJXYg8CURIznSr0yS9/Zzo8ZK/kxYyVV2w8QM2McAKHp/TDXNmIqcwx8prIqzPXNhKb3AyBmxjiqdhxwvmwHbb70vgZoyM5F08eOPa09u3rXrF2eqgAAIABJREFUAaKt7Kgbr6fuO99XdnJWydvbyZ60muxJq6ncfoDYWeMACE1Po6WuEVOZY8ffVFaNuaGJ0PQ0AGJnjaNyx0EAApPjbemipwyjKe+8R3bx29s5Mmk1RyatRm/HDktPw+yBHWbH1u84SOOJQvb3v4uDwxZycNhCDMV6jkxe43LQInF32CbT67cfJG7WWBu3pa4RoxPX6MSNmzWWCqvN9vNhYqYOp+HEWY82l7yzneyMVdLkzC8O0GOmxPbkb0t9m797zBxL5fZWf+ts6aImD/Xqb7iw+uwfHkLK249Q/Nz/0nDoZ68sZ666TwKqnhI38uYx1DqV69pd+4maLg1kI2643rZimaqnYxzRpCRitFslyJ3K/vIFxyev4PjkFVTt2I92xngAQjzEEUt9EyHWOKKdMZ7qC44jlzZ2OrIvXQwrfWc7xzJWcswas3t0MGb3sIvZEeMGk7Do15ycsw5Lk/f5enLVKbm4cvq6O7TNnY0j+bPu58SoeZwYNY+KP39G2WvbvA5aoLvEbHn6I4ouHwmeJocKgjAReB2wAPOB5cAAIAK4WxTFv3sDuHrjAhA+fgiJj94lLYn3tz2UvrqN+BW/o/FYHrW7DiBoVPTeuJzgq1Noqa7j9OINGAtLiVsyk7hFMzAUtE2Ey7/tMVr0NQ7XFwTXdoWNH0LiI/MQ/P2o3Lq7jfufPGp3W7kvrSDIyj2z+HmMZ0uJ+vU4YhfOAFMLoihS+vIH1Ozc75JhMLl/kdXnmflEjBuMpcnAqeWv2pZN7L/rBX7MWAlAyIBUUlqXXPwyizN/eBOAqMwR9HlqHgHacMy1DTQcL+Dk7560XVsdYHbJ7Gpfe3qjGD5+CD0fa2OXvLIN3UrJ3zVWdp+Nywnqn4K5uo6CRRIb4Orvt+AfFoygCsBc20De7MdoznXsTDeZ3H9ykrJuHpHjB2FpMpC37HXqsyVfD9z9PNmTVgMQOjCVvtalgav3HuHUg9Lylle8uYqgvglgETGcKyd/zRaHp1HevtBPXTePKCs7x449ePfzHLFjty6HXLX3CPlWtr2GHXydI1PWOiyHLLp+gQ9A33V3WblGTi57jfpsaUnP9N3Pk2Xj2i2HvPeojXvFK0sI7d8HURQxnC0nd/Vmh4FPgODZ6uRn5hE1fjDmJgN5y1+jodXfuzaQnbEKgJCBqaRZl9as2nuEAmvZvuLN1QSlJiBa/X1q7WYHfwerTS6Zna3PcUtmEbtwBkb7sv37R9uVbT93cWTcEBKsy5hWbd1N2WtbiVs+m6ZjuTZu0osS11xdT+GS5zCeLSXy1+OJvXcGYksLWERK//gBtTtdLw3sKY70evpuWxwpWPGKbUnjq3e+yPHJKySfDUhtWw75yywKH/oTAJGZI+j91DwCoiMw1zbQeLyAnNlP2K6tchNHujp2iqL7ct3VMczQ4v45XJ9n5hNp9XW+Xcy+ZtcLHLOL2al2Mfu0tVwP+u41BI2Kliqp/tYfzqHg/s22a1s82AxdW6fk4rqrU9C1vgY3nz7ZsbuqbQ5yF786GUfsFbfst5gbml0uh1xvUHuwuGvzOUjlev5JV9dlgMGFn3quWN1EIcF9uu/3WEBD4+lu6UePAxeXJwjCP4Fpoij6NLvS3cClq+Vu4HIp5KnD0ZVyN3Dpasn5KaSngUtXqmPTqS+uPA1culLeBi5dKXcDl66Wp05WV0uuOOJu4NLV8jRw6Wp5Grh0pbwNXC5HyVmn5PK2u4FLV8vbwKUr5W7gcimkDFwujrrrwMVjyygIwmcufh4H/F0QBERRnNYld6VIkSJFihQpUqRIkSJFdvL2SC8JOA68ibSugwAMA17wdJIiRYoUKVKkSJEiRYpcqzuv3NWd5W1y/hDgMPAHoEYUxa+AJlEUvxZF8euuvjlFihQpUqRIkSJFihQpAi9vXKzzWF4SBGGb9d9Sb+e0A/jL8128nHNc/AR5vmdVqeT5Nr1Fpu/DAdSqzq9jfyFqMsgzt+aXqjoZv9X+pUmueRdGi7fnaJefjKJ8sVO4KBtm/XdJI9M8PaNMbWSLKF+dkrMPpujylk+DEFEUzwEzBUG4Eajt2ltSpEiRIkWKFClSpOjyVUcXx1IkqUNvT0RR/Bzwvli4IkWKFClSpEiRIkWKFF1E/fLezStSpEiRIkWKFClSpOi/TvJsFACEjkkn4dH54OdH1d92Uf7Ghw7HBXUAPV9YQVD/VMzVdRQufg7T+bYdplUJPUjb+RplL7/vcuMlT1zdI3dL3K07qXDF3bCCwP59MVfXcXbJekzny1AlxpK2axOGU9LusE1HT1L00Gsds3lsOonWzaYq/7aL8k3t2Ukvttl8ZvFzmM6VETQwjZ7rFlsTCZRufI/aHa43rXOlkNFDiHtoAYK/H9Vbd6Dfsq0dN+G5VTabzy9dZ/O15oo+xD+5BP/QYESLyOnfLEU0+j6H55eYzwC9nriLiAlDpA0Cl79C44+n2qUJvialbYPAvYcpfETaCDLqppEkrriFwLSe/HTjGtvmgl3N9Y8MJXXTSjRJsRjOlpF/zwbMNQ3d3uaUp+4keuJgadPNpa/ScKygXZrQASm2zT4r9xzh1EN/djieeO80Uh69nX9fNddhs8/uyL1Qdq9Vs4ifPRGTXvrq9/S696jac8Qnbu8n7yJyQrptg8DGY67z2LZB4N4szjws5XH0TdeRuPIWgtJ6cvyGtbYN/rq7zXLmc7+n56CdKG0Q+PN9m6hzwQ4bkMxVf1yIX6Aa/Z4j5PzhHemeru7N/3t+Pn4aFWKLmZP3v0XtEd98nvb0XLQTpc0Yf7rvderdcK/84yIbN/cPbwNw9ZZlBKcmABAQHkxLbSMHJ67p1lyA5KfuJMqaz7lu8jlkQApprZsG7zlCQWv5WnMr0ZnDEC0WTBW15C19FWNplVdm+LjB9Hp8Hvj7UfH+Lkpe+9jhuKAOIHnjMoIHpNJSVcepezdgPCe1j/GLphPz20lgtlD4yJ+o/fqoz7a2KvWpuURPTMfcZCBn6Wsu/R06wG7D4j1Z5D8k+bv3qpnEz55kq1MFPtapsLHpJD46D8HfH/0HOynb9FE7m3u9uJzga/rSUlUrbWR7rozQUYNIuP92BFUAoqmFomfeof77/3TY5u4k8Rc4z+xiSJ43Ln5+JDxxDwVzHiN38iIipo1B0zfJIUnUrMmYa+rJGb+Airc+Jf7+OQ7HdQ/No/7rwx3nPn4vp+c+St6UhUTcPNY1t7aB3Al3o//zp8SvbeMaz5SQf9N95N90X8c7s35+JFptzslYRKQLm6OtNp8ct4Dytz5FZ7W5+WQhuTcvJ/eGpRTc/ig9n14E/j5mnZ8f8Y8t5Oy8R8ifeg/hN41F7cSNnDEFc209+ZPmUfn2J8SuvlM64O9HwobVlDzyKqduuJfC29YitnRgAYBfYj4DERPS0SQncGzUQk6v3UTvdQtcpuu97h5Or93EsVEL0SQnEDE+HYCmE4XkzV9P3Q8/XVKubtFvqP32GMdGLaL222PoFv2m29scNXEwQSk6Dl23hNxVb9B3/d0u0/VdP5/cVZs5dN0SglJ0RE0YbDumTtASNWYAzefKuz33YrHPb/mcI5NWc2TSap878BET0glM1pF9/SIK1rxB8jrX3ORnF1CwZhPZ1y8iMFlHxHiJ23iikNx5z3U4j0E+m+XMZ+3EQQQlx/Pva5dyYtWfuOK5u1ymu+K5efy8agv/vnYpQcnxaCcMku7pkdkUbPiQAxPXcuq5rfR9eLaP3MEEJ8fzw7X3cWLVFq54bp4b7nxOrNrMD9feR3ByPNFW7vG7N3Jw4hoOTlxD+ef7Kf98f7fmQls+Z123hLxVb5DqJp9T188nf9Vmsqz5HGnN5/Ovf8rRCSvJnrSaql2HSVox0zvUz49eTy0g5/dPcHz8EqJ/NZrAtJ4OSWJuzaClpp4fR91L6Z8+o+eDtwMQmNaT6F+N4viEJeTc9ji9nr4H/DrWnWu1+eB1S8hdtZm+6+e7TNd3/XxyVm3moK1sD7IdO7/ln2RNWk2Wr3XKz4+eTy7g1B2Pc2LSIqKmjUGT5tQPuiUDc009P49dQPlbn6G7/w4AzFW1nLrzKU5OuY/CFRvp9dLyDtmr6PKRLAOX4IFpGM8UYzpbimhqoeYf+wjPGOGQJjxjBNUf7QGg5ovvCB050O7YtRjPltCcU9ghbtDAfhjsuf/cR1jGtQ5pwiZdS5WN+y0hdtwLUfAgyWajlV39j32ET3ayefKINva/2mwWmw1gllZDETTqDk3oChrQD+OZIkxnS8DUQu3n+wibeJ1DmtBJ11Lz8W4Aard/S/B1EjdkVDqGkwUYTkhPYczVdWDxfVWWX2I+A0ROGY7+wy8BaMjKwT8iBFVslEMaVWwU/mFBNBw+CYD+wy+JzBwOQHPeOZrziy45N3LKcPTbpPP1274kMtMxr7qS3VmbtVOGUbb1KwDqsnIJCA9GFRvpxI3EPzSYusM5AJRt/Qpt5jDb8dQn5lDw5LvQgXolF/disTujqCnDqfhQ4tZ7yeN6K7fiw6+Ispaj5rzzncpjkM9mOfO5R+YwSrbtA6D2cC4B4SGondjq2EgCQoOoPZQLQMm2ffSYamWL4B8WBEhvIAw+vAEAiMkc6hPX3x3XTrHTrqP0k++6NRcg2i6f6zuRz+b6Jls6v2CNT0/SQwalYThdjLFQaqcqP/2WSKc+QeTktphc9fn3hI0aYP19BJWffotobMF4tgzD6WJCBqX5bC9AzJRhlG6VdrWQyrb78tVqc+nWr4mxxuzOKLjVZmvbXPWPb4hw6hNEZIyg8qO9AFT/6zvCrpfa5qbjp2gpqwSgOacQP40KQS3bR0OKZJQsA5eAeC2m4grb36YSPap4rUMaVZwWY2saswVzXQP+UeEIQRp63DOdspff7zBXFa/FVNz21KuluAJVXHuuLY3ZgqWuEf+ocADUSXGk/uNlkt9fR/CwqzvGjtNiKrKzuVjvml3U3maAoEH96LfzNfrteIXzD71uG8h4U0C8lhYHX1cQ4MQNiNNiKrGzuV6yWZ2cCCIk/flJkv/+R6Lnz+iQzb/EfAZQx2sxFultf5uK9ajio53uMRpjcVsaY7EetZNvLjVXFROJqUzq3JjKqlBpIy4Zu7NS67QYiuyvWYlG53hNjU6LwY5rKK5EbU0TPXkohuJKGn4681/BvRhsgIQ7M0nf+wJpLy0kICLEN258NAa7GGYs0qN2ymO1cx67SNMZyWazjPms0UXRfN7+uno0uminNNEYiivb0hRVotFJg8mch/9C2iO3cX3Wa/R99PfkP+1bLNXoomk+35bP7rlt99Zc1D5N5LVXYiyvoamgpFtzoX0+G9zks2P8cixfve7/LUMPv0GP6aMpfO5vPjCj29o+wFiiR61zVZ/s2sfaRgKiwnw61xe+o80u+E7+dk4j1akN9HvpXp/qlMq5T1Bc0b5PEO+qHxTmkCbihpE0HT+FaGzxbqiiy04eBy6CIPgJgnCnIAifC4KQLQjCYUEQPhAEYdwFUYX2+wS0e4vgIg2iSNzy2VT8+VMsjRdp/452XNdpWsorOTlqLvk3L6X46TdJemkVfqFBvnPc2ONrmqajOeRMXkTetBXE3jsTQePrPiLeuYIbruDvT9CQqyha+Tynb11NWMZ1trcxvqF/gfns4doOSXwpDx2VXFwZ2a4u6Wte+wWpSVo2nTM+dDK6C/dC2QDF7+zg4IjFZE1chbG0iuTH7ug02Kf6fBG+45bLZjnz2ZfY7TpPpH97zskg55G/8F36InIf+QtXvnRPl3Kdszn219d36K2HfFx3aN/LF0Dhs+9zaMg9lH/0Dbo7MzsFbRcO3eavD+d6xXv3pac+SdE7OzkwYglZE1djLK0m5bHbfYG6vZ6v9xWYlkTC/Xdw9oHXfeB1b4mi2K3/667y9p7tLeAMsA6YgbSHyzfAQ4IgXCOK4iuuThIE4W7gboBHtNcwI6y3w/GW4gpUuhjb36p4LS2llQ5pTCUVqHUxtJTowd8P/7AQzNV1BA/qR8TUkcTfPwf/8BBEi4hoMKL/X++rNJtK9Kh0PdqM18VgKnPmSmlauX5hwdInUoDZKP3b/GM+xsIS1MmJNB/L88pttUeVYGezTuuCLaUxOdlsL0P+OSxNzQT2602TD+yWkgoCHHwdY3vd6sCNt7M5VLK5paSCxoPHMFdJk+8avj5E4NV9afx3tk82/5LyOfaOqfSYnQFAw9E81AltT5FUOi0mp880jMV6h6d1ap0Wo5NvfNHF5JoqqlHFRklvW2KjMOlrLhm7I9LNzSR+9kQA6o7mo0mwv2Y0hhLHaxqK9A5PTzW6aIwllQT2jiewVyzpezdYf9cyeOdzHJ36AKby6m7DvZhsAFNFW76W/HU3V7/7gEsmQNycTIc81iTEUN/KTfAhjxO0GEt8+0TJWXLZLGc+95w7mYTbJHbt0XwCE7XU2K6rxeDkS4PTGwdNQrTN37pZY20T9cs++4ErX3Q97wwgce4UG7fuaD6BiTHUcNILt83mwAStg18Efz9ibxzOwYz73TLl5ALEz80kzprP9dZ8bm1t7cuOPdsxfrVPA1DxyTdc+X8Pcvb5rR75Ul1pax/V8VpMTtdrTWMqtraP4VI75cu5rqSbOwXd7EkA1B3NcyjbGp22nT3GdmW7rT7b16niv+6m/7vefW4qceoT6GIwOfcJit33g1TxWvpseZDCFRsxFvr+Rk3R5SVvn4oNEUXxMVEUvxVFcRkwWRTFXcCNwEJ3J4miuEUUxaGiKA51HrQANP4nF02fBFQ94xBUAUTcPIba3Qcc0tTu3k/kdCmoREy9nvp/S6tHnJp1PydHz+Pk6HlU/Pkzyl/f5lNnFqDpPzmO3JvGULfbcQJf3Z79RNm4o2iwcv2jw22T31RJcaj7JGDqQMVpzM5FbceOvHkMtbucbN5lx77hetuKGaqecbbJ+KrEHmhSEm0ri3i1+ViOjYsqgPAbx1C3x3FFsvo9+4n4jRTMwjNH0fiDxK3/JovAK5IRAjXg70fwsP4Y83yfb/JLyueyv3zB8ckrOD55BVU79qOdMR6AkPR+mGsbbZ9gtcpUVoWlvomQ9H4AaGeMp3rHgXbXvZTc6p0H0c6UztfO9H4/ctlc/PZ22yRr/fYDxM4aB0BYehrmukZMZY6dQlNZNeaGJsLSpW/AY2eNQ7/jII0nCtnf/y4ODlvIwWELMRTrOTJ5jdtOpVzci8kGHL7d104dQeOJs265pe9s58eMlfyYsZKq7QeImSFxQz3ksbm+mVBrHsfMGEdVJ/JYTpvlzOdzb+/kwMS1HJi4lvIvDhI/cwwA4UPSaKlrxOjENpZVY65vJnyIxI6fOYby7ZLNhpIqIkdeBUDU6P40nnIfx86/vaNtYvsXBxy4ZrfcJgduxfZDtuNRY66hIbfI4TO27sQFKHl7O9mTVpM9aTWVdvkcmi752l0+h9rlc6W1fAUmx9vSRU8ZRlPeea/8huxcApN1qJNiEVQBRP9qFNVOfYLqXQdsMTnqxpHUfXfM9nv0r0YhqANQJ8USmKyj4WiuV2bx2ztsk+n12w8SN2ssIJVtt+XLrmzHzRpLhdVm+/kwMVOH0+ChTrWqMTsXTXIC6iSpbY66eTS1uxzb5trdB4iePgGAyBuup87aD/IPDyHl7f/f3pmHSVZU6fv9ml12EHQAWUUQEARsAWVQQMUVXFhEQAZFcRlkcUFcQBDHAYQZQAUURDYVEERwROlm36GbHRpkkU3xh2Kz0yzd3++PiOzKKqq7GqmIyKo67/Pkk3VvVtUXNzPy3hsR53xnPx4+5CSenjRlSK1g9DLUissLklaxfY+k9YDnAWw/J+lfX0eaPoO/7n8MK510QLKrPWMiz931AEvvtQPP3nIXT068lqmnTeB1/7M3b7joWKY//hQP7H7IvyzXT/c7x7DiiQeiceOYesaEpLtn1r3gWqaedj7LHf5lVr3wJ0x//Cke/NLBACz41rVYes8d8PQZMH06f/3Wj5j++FNDCA7Q3u8YVj7pAJhrHFNPT8f8mnzMT0y8ln+ePoHXHb43q118LNMf6zvmBcevwdKf3xq/+CLMMH/59jEzV0HmRPdvBxzN6352ULJD/vX5PH/3A7x6jx2ZdstdPHXhNTx2xh9Z5gdfYZWJxyU75L3SMc944ike/dlvWOms/8U2T18yiacuvu7lHfNY+5yBxy+YzKKbrc+brjg6WQPv3bcwueb5h3Pbe/YG4L59j+2zBr7oeh6/8HoAFnvvBqxw0K7MvcSivOGkb/HMbX/mTzscWFz34R+dxeuP+QpLbb85z//lH9y926E9f8xTJ17PEpuvx1uu/iEznn2OP+3ZFz6w7sRDueFdXwXg7n1+OtOuduqFN8yxq1Sv6Q6H9krf3omF1loRDNMefIS7vnrsHOk+dsFkFtt8Pda58sfMePY57t3rhzNfW2vCYdz67i8DcN/Xj2Xljh1y12e8+Hs3YMWDdmXuJRdhtZO/ydO3/Zk7P/Hdnj7mlp/zoxNv4NWbr8tG1xzBjGef5/Y9jp752lsvOJhrN98HgDv2OS7bIc/DoxfcyKMXJFvcKV8+ljcc9B9o7rmY8dzz3PGVn8yx7pKbr8dG1xzJ9GefZ8oefcc8/oJDZloM37nPcbzxyC8w1/zzZt2+Y37Nh19+uFYrXUif8+Kbr8d6+XO+u+tzXmfiodyUP+d79/kpr8+f82Ndn/MK39yRBV6/DMwwzz30d+752hy819Nn8MC3f8obTt0fxs3Fo6dNZNqfHmSZr2zP0zfdzeMTruMfv5rISkfsyVqXH830x57kni8cBsC0Pz3I1HOvYM0LfwjTp3P/t37ysoxzAP458XqW2Hxdxl99VLL63rPPQXO9iYdyfT7mu/b5aZ8d8oU3vuQ7ZZvnHvz7nH2nps/gof2OZeWTvoPmGsc/T5/ItLse5LV7f4Jnbr6bJyZey6OnTWCF/9mbN15yLC8+9iT3/2e6Dr165w8w74r/xmt3347X7r4dAPfstD8vDhEZ0Mv0cjhWL6PZvXGSNgN+DjxHGuRsb/tqSUsBX7U9pEn6LSt9qMkn80rGVa+UGTMGC4YtzzzzvAyr4mHkxRfnaqIL7T7nZ5+b0xyjYDiYNr1dHxtrzDvu5d0ADRfPzxh79ZCfd8Nz5xisITGf2vTt+eZuk0T+zIvtrlOLzPdcM+03339Om5uwl8k88y7b01/CF57/S0++j7NdcbF9oaQVgCVt/wNA0km2PwnMeWWnIAiCIAiCIAiCV8BsBy6Szun6ufPjZpIWA7C9ZbmmBUEQBEEQBMHoo6eXW3qYoXJcXgfcBhxHeo8FjAcOK9yuIAiCIAiCIAiCmQzpKgZMBr4JPG77YuBZ25fYvqR044IgCIIgCIIgCIA5K4ADLAecAfwQeKBi8ZvPNiy800Q7jnlsaI813TjmsaEbxxy6o1U7jnnsaMejtx9zZONi+yHb2wDnAae8rJHRK+OzFbV6RTuOeWxojzXdltpxzGNDO4559Ou21I5jHjvaQQ8zVI5LP2z/HzBnVQCDIAiCIAiCIAiGibFnnB8EQRAEQRAEwYij1wcuc1bqd3RpxzGPDe2xpttSO455bGjHMY9+3ZbaccxjRzvoYWSHk3QQBEEQBEEQBL1Nr6+4BEEQBEEQBEEQxMAlCIIgCIIgCILeJwYuQRAEQfAykTTfnOwLgiAIho+eHLhIWkvStpI+2XlU1l9a0vKdRyXNxSW9VdImnUcN3ZZI+oCkr0nar/No3abStOzbrb9XNZF08JzsK6S9zZzsK6C7x5zsK6R9wZzsK6B78pzsK8RVc7hvWGnct5tot+pfWeftc7JvmDX/U9LiJTVmo31mvjZXvz9UYsfOvUC+B3tr7XYEvU3PJedL2h94J7AG8HvgfcDltreuoL0lcBiwDPAIsAIwxfaahXV3BfYAlgNuBDYErrK9WUndrD0/8GlgTWD+zn7bnyqsewzwKmBT4Dhga+Ba258uqZu1lwL2IfWx7mMu+n437ttNtBv2r+ttrzdg38221y6pOxvtl+yrpHuD7XULas5P+h5fROpfyi8tApxn+42ltLN+v2OWNBdwi+01Cmq+FliWVIz5E/Q/5mNsr15KO+v3Wt8upt26f+U2VP8+SzoI+DhwPfAz4I+udLMm6V3ALqT7kDOAn9u+o5L20cAMYDPbb8yDt/Ntj6+hH4wMXlYBykpsDawD3GB7F0mvId3Y1uC7pC/rRNvrStoU2L6C7h7AeOBq25tKWh04oIIuwMnAHcAWwIHADsCUCrpvs712vugdIOkw4KwKugCnAqcBHwA+B+wM/L2Cbsu+3Uq7av+S9HngC8Aqkm7uemlh4MpSuln7fcD7gWUlHdn10iLAiwV1tyfdQK8k6ZyulxYGHi2lm9kN2JM02TOZvhvLJ4AflRKVtC/wDWABSU90dgPPU95GdQvgP0gTTYfR/5i/UUq0cd9upd2kfwFI2gh4G7CUpL27XloEmKuktu1vSfo28B7SIOKHkk4Hjrd9T2HticBESYuS7n8mSHoQ+Clwiu0XCspvYHs9STfktkyVNG9BvWAE0osDl2dtz5D0oqRFSCsfK1fSfsH2o5LGSRpn+6JKS/DTbE+ThKT5bN8habUKugCvt72NpK1snyjpF8AfK+g+m5+fkbQM6QZrpQq6AEvaPl7SHrYvAS6RdEkF3ZZ9u5V27f71C+A84PvA17v2P2n7nwV1Af4KTAK2JN1kzdQG9iqoeyXwMPBq0o10t+7Ng/7FMGH7COAISbvbPqqk1gDd7wPfl/R92/vW0s3aJwInSvqY7TMrSrfs2020W/WvzLzAQqT7pIW79j9Bmggqim1L+hvwN9LEx+LAryVNsP21ktqSlgR2BHYCbiBN9m1MmuR7Z0HpF/KqqXM7liKtwATBTHpx4DJJ0mKk0f1k4Cng2kraj0laCLgUOFXSIxScKe3ioXzMZ5ORpu6uAAAgAElEQVRmN6aSboJq0Jk9eUzSWqST5IoVdH+Xj/lQ0nK4qbf60DnmhyV9gPReL1dBt2XfbqVdtX/Zfhx4XNIRwD9tPwkgaWFJG9i+pqD2TcBNkn4DPG17etaeCyiWtG37fuB+STsAf7U9LesuQOrX95XS7mKGpMVsP5a1Fwe2t/3jwrrXSlo0f+7kPv5O22cX1gVYX9IFA475y7a/VUKscd9upp2p3r+6JrV+nr9j1ZD0JdIg4R+k6+JXbb+glHdyF1Bs4CLpLGB10mr5h2w/nF86TdKkUrqZI4HfAEtL+h5pgFjk+xSMXHoux6UbSSsCi9guOmvYpbcgMI20HL0DsChwqu3S4RbdbXhH1v2D7ecr6O0KnAmsDZxAmmHaz/YxpbW72jAfMH/n5qOC3geBy4DXAUeRlv4PsH3ObP9weNuwIhX7divtVv0rhxqs14kLzxf8SaXzTLLW1cC7bD+VtxcixWm/rbDuJFII5vN5e17gihrx4ZJutP3mAfuK5te01J2VjurkMrXs2020G3/ObwC+QppwmTnZWzInUtKBpLCwlwyYJL3Rdslw2/fb/v2AffPZfq6U5gCt1YHNSfdhF5Q81mBk0osrLp3ZlFXJybySNrF9aWld2093bZ5YWm8gkpYG/pw3Xws8UFrTdmeV4xLqhS0ByeWKrgR5Sdg+qbSu7d/lHx8nmQNUo1XfbqXdsH+pO5k1h8nVOt/N3xm0ZO2nJL2qgu7c3ZMdtp+vGB8+TpK6bmjnIoXaFNcdZF+tz3mu7hu6vMJVww65Zd9upd2qf0FKUD+GtPIxvZLmSgMHLZJOtr1ThRv5g0gGLt1cBdQYGC9BCmP+Zde+eQrn1QQjjJ4buGgWDltADYetDUkz8G8knRTnIoV8LFJYd6Cb2fKkhOaibmZd+h/gpa5PBxbWHNTlCig+cFE7p6uWfbuldvX+Bdybwy2OzttfAO4trNnhaUnr2b4eQNL69OV0leTvkrbsrBxK2ooUalKDPwKnK7kFmmR68YcKupMkHU5K1DawO/3zi0pyCnCBpBOy9qeoM+HVsm+30m7VvwBetH300L82rPS79ueB2volBdXnlreApHXp7+BWY+IFUtj464CpWX8xUkj3I8BnbNf6bgc9TC/Wcek4bN1ve1NgXeo4PgH8kOSicRewALAraSBTmo6b2Z9srwS8C7iigm7Hlng70gVfwDYkG+jSbE1aDv6b7V1Ijle1iredTFrR2oK0ErAcKZG5NC37dhPthv3rcyRHoL/kxwbAZyvoQnJBOkPSZZIuIznY/WcF3c8B35D0oJIL0D4kV6Ya7ANcCHwe+CJwAQXj8LvYneQkdhppZnxa1i+O7UNIs9NvJE3AfDfvK03Lvt1Ku1X/AjhX0hck/ZukJTqPEkKS9pX0JLC2pCfy40nShOZvS2h2sQXwA9L18HDSZOphwN4UdMsbwB+A99t+te0lSROap5MGyKXz5YIRQs/luEi6zvZ4STeSrPGeGyy+tZD2JNtvUZcvvaQra8SmZ92bgHXz8vu1tosXXuoca9fzQsBZtt9TWPda22+VNJkUrvUkcKsL18zJ2jc42V13jnkekk9+6TouLft2E+1W/as1uU+tRhqs3VEz1CG/x+okUFfUnZd0Ez8DuLNGjl6X9qLAjAbH/FrSzfsM4Drbf6upP5Zo1b8k/XmQ3bZdLPRVDdzyurRru+V1a0+y/ZbB9tW6Vga9T8+FitHWYeuZfHK8UdIhJIvRBSvotnIzg3a2xC0dtlo5qbXs2620m/QvSSsDR5BWMk0Ki9vLdvGwlhyK+AWSfaiByyQd4+z2VVB3SWD/jq6ky4EDa5iL5HDAY4B7SIO1lSTtZvu8wrrjSQX6Fs7bjwOfqhFSksMv9yOtBAg4StKBtn9WWLdl326i3ap/AeQoiCpIWt2p2OMZkl6SU9IJPy2kvaPtU4AV1b9uTUf78FLaXfxT0j7Ar/L2dsDUHCoXtsgB0IMrLt1oEIctSYvbnlpIbwXg/5HyW/bK2j+2fXcJvS7dBUk3eOOo7GamVOTqKFLYVidO/Ke29yut3dWGFRngciVpTdu3FdLrOF29Cfg5yenq27aPLaE3izZU7duttFv1LyVnrx/Rl+T5cWB32xuU1M3ap5NWEE/Ju7YHFre9TWHdCaTJj47uDiRr4HeV1M3adwAf7JwrJa0C/J/LV5G/Gfii7cvy9sakc3aNKvJ3klzcHs3bSwJX2i5ag6tx326i3aJ/SdrM9oWSPjrY67aHvWCypJ/Y/qykiwaXLOpktpvtY5XyTwcTL14UW9Kr6Zt8ESnv9QCSkc7ype/FgpFBTw9cBkMV7CZno32m7Y810L3K9kYVdF5iSyzp3bYnlNYepC0tP+ednYrM1dZtecw1bFyr9S9J1wy8mZJ0te0Nh1trEO2bbK8z1L4CupNtrz9g30tCLwppX2p7k65tAZd07yuke4Xttw+1r5D2BcD73N9++velB4qN+3YT7Rb9S9IBtvdXMl8YiF3YzCUIgsHpxVCxodDQv1KMqnbBXcw/9K+8cpxsPQd6tR8MVB+40PZz3oMGdti0Pebi2pX710WSvk4KOTAp5OD/lJNqXbbS+A2SNrR9NYCkDahjtnGRpI+TklkhGWD8XwVdgNsk/T5rm2TCcF1ntrrE7HTmWknHklYAOp/zxZ0wm5KhNaTk9Gsk/TZrb5Xbs3fWLhVa07Jvt9Ku3r9s75+fdxnu/z0Ukr5IirqoVnBT0pGze932l0ppd7VhKZLpwkAXyuLul8HIIVZcRoB242OuUuRrEN045jGgXeq91uAJtR1KJ9ZOISXmd+owLQ9MIcVou1QYk5L70IL01ZqYC+jUprIL2rrPYla6Q7HZ6VmE1HTrlgytGTSkpku8SGhN477dRLtV/8rai5LClzqrO5eQcseKFUxWg4Kbknae3es1ohAknU9yCPwKycFuZ+DvtvcprR2MHEbiiktQl5E1sh0exuIxt6LIez1UQm3hEMj3DqFdJJ/I9sJD6BbLGxtqVlrSvra/X0B3tgVkS4Z9DjUwkXSU7d0L6Dbr2620W/WvzM+AW4Ft8/ZOwAnAoLkvw0T1gpstwqMHYUnbx0vaw/YlwCWSLmndqKC36MU6LkMxqsNpeky3CEq8bohfq2alOghj8XMeVX1sDji41D+2ff/sHqQaFC04uZEupNCeFuzRSBegeJ7NLCjWt3tYu2T/WsX2/rbvzY8DKB823im4ubmkzUihkEULbkr63/x8rqRzBj5KanfRcfx8WNIHlAphLldJOxgh9NyKi6QfACfMZmZw85rtGUCR5UpJKwEPO9ulSloAeI3t+/Kv7FRCdw65b8jfeJnYtqSzmU0l4FLJnpLGAVvbPn02vzas+QgaolhZV1x4sb4taUPgNuc6F5IWBtawfU1p7SG4r5HuWBwkxjGPDeJzHl6elbSx7csBJL2dPpv3UnSKx36edGznA8cV1uxMbPygsM7sOCiH5n2Z5Ea5CMnhNQhm0nMDF+AO4CeS5iYtx/6yO5a0RPKfpFsYPGRFdMWk2z5/uLUzZ5CqEXeYnveNz7q3FtIdMgnQdqnl8Ksljbd9XaH/PyhOxT3/k74E5sF+Z7irnE8m9S+R8h2m5p8XI+VBrJR1SybVHg1057A83b2vlHbD/jUULcMBW2nHMY8N4nMeXj4PnJhvqAX8k5R7UQzbM0jn56NL6gzQnJyfL8kOeauT3tcqxT5zONyqtn9Hsj+ebRhoMHbpuYGL7eOA4yStBuwC3CzpClLth9klYr4SPljo/84pc3efGGw/n08cNfiM7R91aU+V9BmgmHtJZlNgN0n3k26i+w0SCzNB0ldISYCd5OViN++duHBJxwDn2P593n4fULzGRmZmvHRu04w8OVCaVv0r6C1G40x8L2uPNYq917ZvBNaRtEjefqKUlqTTbW87q8nUGtdHNSr2aXu6pC2B/ympE4x8em7gAjNH3qvnxz+Am4C985fn48Otl+POW/J3SVvaPgdA0lak465B9STAzPsqaMyKjgPNF7v2mfJxy+Ntf26moH2epO8W1uxwr6Qv0TeD9wWgeKVt2vWvobivoXarG9qWeWNnNNKtYUM9K45opHtfI92W2sX6l1Jh0U5RREu6nOQqVqJAdCcnq+Vk6mHAph5Q7BMoOnDJXCnph7x0UrGkrXkwwug5O2RJhwNbkhJYj7d9bddrd7pAVeJsIzq7ULFiNqJZfxXgVGCZrPkg8ElXqBIr6VBgRdIMi0kWhA/a/nJp7ay/NP392h+Yza+PaCT9EbiMVNncwI7AJra3qKC9NHAksFnWvgDY0/YjhXWb9S9JawFr0L9/nVRat0t/0L4taYmCoXmLA6sO0L20hNYA3fmBT/PS+gvFi/TlGeKBugdW0F2KlIswsI8VrznRsm+30G7cvyYAl5LO2wA7AO90wUKjkg4eaAE82L5C2k2KyWatwaJqXOM7FYwcenHg8ingV7afGeS1RV3QO701khYifSZPVtQcR0oC3JyuJEDb02f7h69cd0vSzM4ywCPACsAU22uW1O3Sb3HxXYK+egAmXQwPLJzb0pSG/Wt/4J2kz/j3pBW+y21vXVI3azfp25J2Jc3YLgfcCGwIXFXpRvoMUn7iJ4ADSTd3U2wXdfXK4ZevIoWeHkcqunmt7U+X1M3aTWpONO7bTbRb9a+sPdn2+gP2TbL9loKaL6mrJenmkqFiysU8gXeTzlndxT7vrDWZGQRD0XMDFwBJy5K+ODND2WrMGnbpV1kFkLSj7VOUKy0PxOUqLzdH0k2k2f+JtteVtCkpafuzFbSbXfiz/kK2n6qk9TXbh0g6isFjpotXQ25BjhFfB7jB9jqSXkMaMH2ognaTvp2PeTxwte03S1odOMD2diV1s/YN+Vhvtr22pHmAP5YeNHXpdZ4XAs6y/Z6Sull7su31u28oJV1i+x2FdVv27SbarfpX1v4BMIk+Q5etgTVtz7YA6b+o9XlSGO/KpByTDgsDV9jecbg1u7SbFfnsasNrgP8ClrH9PklrABvZPr60djBy6LkcF0n/DXwcuJ2+CtCd2enS2oPOlJKWp0uwYH4erHBc0RFlDyQBvmD7UUnjJI2zfZGkWjUAtqbv4rtL5+JbWlTS27LOQsDyktYBdrP9hYKyU/LzpIIaL6EH+tez2YDgxZxU+wjlc5g6tOrb02xPk4Sk+WzfoWRyUoNO/YXH8mrm30ghgqXp2NI+I2kZ4FGyS18F+tWcAP5KnZoTLft2K+1W/QvSivHe9NkFzwU8nScchzuU/BekXJLvA1/v2v9k6ZV5D1HksxI/J7nJfjNv/4m0qhkDl2AmPTdwAT4CrGb7uQba3yWFV/SbKS0lZvvY/ONE2/2SSJW84kvSOgnwsTw7eilwqqRHgBcrabe6+P4PsAVwDoDtmyQVjRu2fW5+nlkVOYdvLeSC7ji071+TJC0G/JRkR/0UcO3s/2TYaNW3H8rHfDbJOW8q6Wa6Bj/J+TXfJvXvhYD9Kuj+Lh/zocD1pEFy8UmITKuaEy37divtVv0L24NNLM5E0pqedd25l6v1OMkKePv8vzvRHwvllfriOaAt84mAV9s+XdK+WfNFSUXDioORR8+Fikk6D9imVijNAO1Jtt+SQz3WzTe319p+a2HdweJZX7KvkHaTJEBJCwLTSHkPOwCLkup9lHBqGaj9Y+AbpJW9L5MuvjeWnnGSdI3tDTphD3nfTbbXKambdX5BisOfTrrhWBQ43PahpbVbI2lFYBHbN1fSa9a3u9rwjqz7B1eowdALSJoPmH8050EOpHbf7hXtXqLEtVrSh4DDaZAD2jif6GLgY8AE2+spFU4+uHT4ZTCy6MWBy5mkMJ4LgJmrLjVi8SVNBD5MWqZ9NemEMd7222b7h/+63kakwpN70t+7fBHgI5VuaKsnAfYSNS++kn5Nuhj9kLSy9yXgLS5g8T2I9o0572EHYH2SG9LkUp+zZu3UB8Awh1fMqg1NHLZaUytHbxDdVu5eTRy2Ws5Mt+zbrbRb9a+h6J6IGsb/2TIHtGU+0fok98u1gFuBpYCtx/rgOOhPL4aKnZMfLdiKNFO6F30zpSVPjPOSlrznpn+eyxOkPIxidCcBSuo+KSxMwdoHg9zQir6q8sMdLzxQe5azYpLWc3mv+M+RajssCzxEctj64mz/YviYJ1+APgz80PYLkorNWnTCKyQdSIpHP5m+FYjZhl4MB5qFwxbpZqC09oak0KE3kr7jcwFPlx6sDZKjtzxp5rTGLO2g7l4VdAc12gBqWAOfTHp/t6BrZrq0aOO+3US7Vf+aQ0qcR1vmgDbLJ7I9Oa8Wr0a6Xtxp+4Uh/iwYa9iOx4AHacVjic6jgt4KDY5xUdLJ6JekZejOo/jxNvxcL5rN48LW7St87F8C/kK6uVP+rC+roHvNnOwroHsLaWb2xry9OnBapfd6EvB64AbSoGUX4HsVdG8CliSZTkC6yftJpWO+ecDzQsD5lT7nccBNefs1wLmVjrnzPneOeZ4a55HGfbuJdqv+NYdtu77A/5yYj/GofI0+Ariy0vHsCixOsu2/lzQJslsl7ZtIYdyrtP5c49G7j55ZcRnChciuEza1G2nm7FlgBn2rAaUTt59RKtQ3cBm82CyWeyMJcB3g3/PmpS68HGx705L/fygkHTnI7seBSbZ/W1Lb9pGkJfgO9+fwg9JMz+FpvyJ9l7anzy2wJC0dtrB9t6S5nOrVnCDpygqyLWdpW7l7NXWPy8+1Z6Zb9u1W2i3d44aiRA5Z7eiPbi6wPZVkLrIygKRa7/WWwHbA6ZJmkBzFTq9xPxKMHHpm4EKfC9EU4Ktd+wUcUqkNXyH5s/+jkl6HU0lf0A/SVcishvCskgApHF4iaQ/gM8BZedepkn5i+6iSull7HuDzpBklgIuBY11+SXp+0gzlGXn7Y8BtwKclbWp7z5Lig8WIU/5i+AnSbOERpIHLFXlfaVo6bD0jaV7gRkmHAA/TZ31ekpZOfYO5e/20gm5Lh62O09W36HO6+nYF3ZZ9u5V2q/6FpI+QVtIez9uLAe+0fTaA7Q2HW9P201lrEeDc4f7/Q3AmMDCs+tek3Mii2L6fdL93iKRVSd+ng0kr10EA9GZyfrNkcUl/AD5q+5nSWgN0mxQyyzqtiuXdTCos1TlBL0iq8l3jcz6OFNbRsQjeCZhue9fCuhcC77H9Yt6em5Tn8m7gFttrFNRuVmG8NYM5bElaPM8qltBbAfh/pPyWvbL2j23fXUKvS3dB0sz0OBq5meV2vMTdS9K7bU8orLsiA4w2NIxWtf9Ce3Z2lw15IY2qfbsXtGv3r46xyYB9w56QP+D/Dxr9YbvYaqJSwdo1SQOH7snjRYCvuoKjWW7HisC2pJWX6aRQxMNqaAcjg55ZcWmVLD6AfYErJV1DXUezVoXMoF14iegfMjQ976vB+AGhhxfmAVxpliXNvHcuuAuSKgRPl1S6btHb3Fdh/ABJh9G32jXsSPqa7UMkHcXgBSiLuwR2aV0yyO4LeOms4nDp3Z9/nAYcMPB1SWfa/lgB3afzjzPoG5R3615le6Ph1h2kHc/Rdf7MHAwUHbjYvm+Q3SdT6HOeA/ZgkM9hOKndt3tBu0H/GjfIvtL3Ti2iP1YjRX0sBnyoa/+TpOiI4uR7r3lIUQnb2L63hm4wsuiZgQsNK8Z2cSxwISkBcUYlTWhXyAzahZecAFwj6TekActW1KuOO13SKrbvAZC0MnXyLg4hhQ9dTDrmTYD/yjPlEwtr144R77grTSqo8UqoNUgejFo5GAOZf+hfKUar97vl5xzHPDp0J0k6HPgRaRJmd1JYYknuAapGfuQ8y99K2sj2VTW1u9jZ9h2NtIMRQs8MXAYmizfiRdt71xa1/bv84+OkUJ6aNEkCtH14voHfOO/axfYNpXUzXwUuktSZzVmR5PxUDEkihYX9Hngr6UL7Ddud+PCvzupvh4lOjPgh9F10i1UYt31ufi464/wKaBkj20o7jnlsaMcxDy+7k3ItTiOdt2vY2LeK/gD4iKTbSJNdfyDV1dvT9imlhbPZQ0/W6wl6h54ZuPQIF0n6LCkZrvtkUXTFR9JSpKXYFen6TFyhkFnjJEBIF4JODG8triCtrm2et48l1SIohm1LOtv2+kBRB7FZ8AOSIcG/k471MuDo0qKSJpCW/B/L24sDv7K9RWntIGhMy5WPYJjI18ivD/mLw0ur6A9IeZhfy6YEDwHbkEoGFB+4zCoXs7RuMLKIgUt/Om5H+9J/Bqd0aMdvSTeSE6kTsjSTWSUBUviYJe1HOiGemTVPkHSG7YNK6mZOIhX5/G7e3p4UC79NYd2rJY23fV1hncE4kRSr3LFE3p70PmxbWHepzqAFwPZUJevt1kQ4TV3uG+5/mFcxl7P94Gx+rYRVLZLGkSp6nz6bX6uVmzmQsdi37xvufyjpf23vKelcBs/T23K4NbtoEv2RmSc/vx/4pe1/pq9aFarmYgYjk55zFWuJpG1JDilPSPo2Kcnwuy5cUX0w15JaSLqL5O5V1QJa0hRgXdvT8vYCpEJeb6ygfdOA5PxB9xXQvR14A3A/8DR9TjE1nNRaHfNk4CMdH/7sGHPWQOfAAro/AE6YlaOUpCUq5s4N1H6P7fML/N+VgIcHfKde00lal7SW7VuHWzf/7y+SHMy6V9a2t/3jEnpdupPzKmZ1JF1qe5Ohf3PY9JaY3eud/lyyb0vaELjN9pN5e2FgDdvXlNRu0b8kre++Su4vYRbmBMOl/T3SdaJq9EfW/m/gw6TJzLeSkvV/Z3uDCtrX2N5A0tXAR0m5mLfaXrW0djByiBWX/nzL9umSNiZZ1B5GCqcp/YX9naT32/59YZ3BqJ4EmLmPFMM6LW/Pl9tSgxskbWj7agBJG1BndvR9FTRmRatj/iZwuaTORX4ToKjVduYOUp2NuUlGEL/stk8tdHM1WPFcGDBALTFoyZwBvK1re3reNz7rFhm0ZD5j+0edjbyy9hmg6MCFtquYEyR9hZT70HF0K3lzOZnUvwQsD0zNPy8GPEA22yh8c3s0/V3Dnu7eV1C7ev/Kg5a5svaOpXRmQXf0x8wmUcHYw/bXs7PoE9nx8mlSLmwNmtXrCUYOseLSRcebXdL3SXU1flHSr13Sk/RdiBYkhTV0rJFte5ESugPasC7Z4YuKSYCSzibdUE0gvQfvBi4nFcEsqp9Xe1YjXewh3QRMIYXKFV8ByaFS3YmHxaoCd91Mz0PfMZtUaPR222uV0u5qw9KkwcqNpON+xPalpXWz9mok44XtSQO1n9q+qJDWCrN7vcsmuQizqDdRfFUt69wMrON8Qck3fDe7cO2HxquYfx5kt12w1kbWPQY4pzPRJel9wLtsf7mkbtYarI8Vr7PWqn9lrT8CH3KuVTMWkPQ2Xppze1LlNjSpBxX0PrHi0p+/SDoWeBdwcP7iDObhPizYXrjU/34ZtEoC/E1+dLi4ovZ7K2rNRNKWpFW8ZUgDtBVIA6aSF98PFvzfQyJpV1I9i+VIA5cNSeYAm1XQngtYPT/+AdwE7C1pN9sfH2690gOTOeDvkra0fQ6ApK1Ix12DPwKn55tqA58jORKVptkqpu2SduKzY7ztz3W14zxJ353dHwwj90r6En3GHl8AatTaaNW/IEUHXCHpHPqvrB1eSlDSPCQzlU4o4sXAsbZfmOUfDZ/2ycAqpPN1J+fWpJzIarhRPaig94kVly4kvYp0U3uL7bsk/RvwpoKhHd3aHyVZAxu4zPbZpTWz7pW23zb0bwavFKUil5sBE/PK3qakOO0aoVNNyCs+44Grbb9ZqTrzAba3K6x7OLAlqSDe8bav7XrtTturFdDsrKC+5CUqrKBKWgU4lTQwFvAg8Enbd5fUzdrjgN1ITn0dy9jjbFcxG6m5ijlAdy1gjQHaRW/w8grAZSSXJwM7Apu4glNffp+PJJ3HTPp+7Wn7kcK6zfqXpP0H2W0XtOiVdBxplbxjJ78TMN32rqU0u7SnkPKWeu7msGQETDByiIFLDyDpx8DrgV/mXdsB99gu7RXfLAlQ0tuB75BWHeam7+auVXG+4kiaZPsteQCzru0Zkq61/dbWbSuFpOtsj5d0I7CB7edqmFFI+hTJdvkl+VuSFu0OPxhtKBWUVSeBejQzq1XMSiFE+wPvJA1cfk9a/bnc9taFdZcA9ifNxptUPPjAGonbYxFJ29g+Y6h9w6zZxEwl65wBfMn2w6W1Xi6SrndhY5eg94lQsd7gHcBaXfG7J5JCt2rQKgnweFLRy8lUtoBuyGP5pvJS4FRJjwAvNm5TaR7KyZZnk5KZpwJ/HeJvXjG2fyZpWUlvpn+c9qW1Bi21VgEk7Wj7FEl7D9jf0S0Z0nK67W1nZUxQIdfku6Tww36rmIU1O2xNKs53g+1dJL2GgkVdO+QByh6SFrL9VGk9AElfs32IpKMY/HMukpPYA/0L0rVx4CBlsH3DyXRJq9i+B0DSytS7Tr4auF3StfSfzCxp/xwEc0wMXHqDO0kJ4p0Y+dcBN9cQbhin/bjt8xppt2IrkovaXsAOwKKkGjqjFtsfyT9+R9JFpGMuHpuuZOn5ceB2+sdpFzcFaJDLtGB+HixnrvSS+h75uVUu1Qu2H5U0TtI42xdlR6QaPJtXTV9UKuD7CBVcn3Li9HHAQsDyktYBdrP9hYKyU/LzpIIag9Gsf2XTg/cDy0o6suulRSg/4fRVUkHse0nRCCuQTEZq8J1KOv8K97VuQNCeCBXrAZSsYsfTVyF2PCmB+RkoO9PRKgkw31jORSou1T2rU7RmTjA2kHQnsHZO8Kyt3SSXSdLbbV8x1L5C2gfb3meofQV0J5JqTnyfNFP8CCl5vXjeXg7x/QZpgPxl4CngRttFbzAlXUNa7TmnE+8v6dYaDoED2jEOWMj2EzV1a5EHhG8mTS7t1/XSk8BFtqcW0h1HWkWcTHKCFHBHi3NZbSRNImWIkwMAABHXSURBVLmc/qLU+xuMfGLg0gNoFgWuOrhsoasmSYB59n0gtl3cbaoV2YDhYGBp0sWoStL2WETSecA2tUJpBmg3yWUaLP67Vkz4LLRr2OQuSFrFFH2rmKfafrSk7iDtWBFYxHbxlXL1Fem7oWvgUiv/4RckR6/ppBvrRYHDbR9aSG9WhhcA1Dh35tW0pztGAEpuhfMNlj83jJpX2d6o1P+fhebltjce5D2vdp2S9HrSytJ2pNW9E4Dze9EoIGhHhIr1ACUHJnPA+AEXvAvzDVdRbG9aWqMHOYRUD2DKkL8ZvFKeAW6UdAEV6xNlquYySdqIVHhyqQF5LouQVjWLIenzJEvclZVqbXRYmAoFTm0/3bV54ix/cRiRNMuBoKT1KqwaP5jDxSxpXuBL9IVylWYN209I2oFkSLAPaQBTZODiXDJA0oHA34CT6Ruk1ioncD6pREJnEmSBvK/kqt75kj4GnFXrpt32xvm5WZmG7ID4TUnfJoUH/gyYIelnwBFhQBFADFya0gszHDRKApS0KH3OOACXkJxxRq3bE/D/YtBSjXPyowW1c5nmJeU7zE3/m7knSCFFJfkFcB4pVOvrXfufLHmTMatzJnXOnYfN5jVTvkbR54AjgGWBh0g30cUdKDPz5PDiDwM/tP2CpBo31lvY3qBr++gcMndIBe35u1dubT+lVDqhJHuTctdelNRZURwTq/OS1iaturwfOJNk8b4xqd5cUTfKYGQQoWJjHEmbk5Zj+yUBulB18S7dM4Fb6R+ito7tj5bUbUEOEYPkHvdaksNW9yrAWS3aFZQlh5h0O5qVthhfwY2LYNZyUgvaoFR8ch9SMdcPkExlTrH974V1rwR+BPyKNDjcHvhipVymK4DdOytpktYnDdqKhnJl2+tV6f99ahmdURxJk4HHSK6jZ3bn9Ug6azTeHwQvnxi4NCSfmGZJhRudZkmAGqSWx2D7RgOSTsg/dmaEu7HtT1Vu0qhlCPtUV8oD2I20wvIsMINKNYokLQV8jeRe1n2zUzxvTNKHgMNpU09lHaBz43xpjTyTrNvK2OTIQXY/Dkyy/duS2rNoz9y2i7ps5RyiI4C3k77XV5AKX95XUjdrjycNmDo27v8GbGd7ckHNXUmOasuRKthvCFxpe/NSmr2ApJVt39u6HUFvE6FibZnMS29mZ4Y7UNhaMycNH5Znjqpc7Lt4VtLGti+HmQUpn63chip0XIaU6vPsYfuxvL04sw87CV4+HfvUKSRL0Q6iTlgJwFeANW3/o5Jeh1OB00ix4Z8Ddgb+Xkn7IBrUU5G0B/AZkjshpJyin9g+qrQ2cDTJ2OTHeXunvK90dfP5gdXpqyPyMeA24NOSNrW9Z0lxSR9gwOCYwrbueYCyVUmN2WhfJ2l1+k/uFR2cks5j44GrbW+a9Q8orNkL/EXSJ4AV6b9aParLBgQvjxi4NMRdNVQGWxauRPUkwMzngRNzrgvAVNKN1mhm7c6gBcD2VEnrtmzQaMN91Z5fPzBsKl/8a3AP2cq8MkvaPl7SHjmk5JJstV6DVvVUPg1s0EnSz5pXATUGLk2MTYDXA5t1VjkkHU3Kc3k3hQsXSzoGeBWwKamWzNb02fiX0GtS+HJAG15FyjlZwfZnJK0qaTXbvysoO832NElIms/2HZJWK6jXK/yWtHo4ma5w6iDoJgYuPcCsloWBGsvCrZIAp5BmwFcBFiOdrD5M/ZWfmoyTtLizP30erMZ3cBhp7XKV2Re4MicP13Q068wCP5xnxf9KOqfUoKqTWheiv5nIdF4ajlmKVtXNlyWdsztGJgsCy9ieLqn0zd7bbK+dra4PkHQYfatdJWhV+LKbE0g30p2clodIq10lBy4PSVqMlA85QdJU+kLVRjPL2X5v60YEvU3cNPUGzZaFbS/caLXnt6QkvOuBv1TUbclhpBvaX5NmD7cFvte2SaOOJi5XAziW5IBzCynHpRYH5RXML5NWHBYhOZvVoLaTWocTgGsk/YY0YNmKlNhbg+7q5pDCW2pUNz+EZPV9MemYNwH+S6mmzcTC2p1w3mckLQM8Cqw0m99/Rdg+Nz9XsbqeBavY3k7S9rktz0oqOji2/ZH843eUap4tCvyhpGaPcKWkN9kuunIYjGwiOb8HkHSd7fGSbiSFPTxXK1G9VRKgGlR67gUkrUGySxVwge3bGzcpGGYkXVnD7agXqe2kljXXI9mlAlxm+4bSmll3ftIgsXOunAD8j+1pBTVFOle/CLyVdB651naV2fhcX+Mo0jnsR3n3cba/XVh3AqmgbHd+4K9sb1FSN2t1oh+usL2epFWAX7pwQdmxRJeZytykSdR7SavVnQiQooVsg5FFrLj0Bi2XhVut9ozJmZU8UInByujmIkmfBc6lf6hYaZfApUiJ6ivSf/BQ3LVuVk5qFDYY6W5Cl24tTiLVyvlu3t6eVCBxm1KCti3pbNvrk1ata/MDUn7iv5NyiS4jGRKUZqlB8gOXrqALqd7YH4DXSTqV5Gz2H5W0xwofbN2AYOQQKy49hqR3kJeFbT9fQa/Jao+k20lJpn8mZlaCUYSkP3dtzjzBVrBDvpJ0IzmZrlwL22eW1M3adwEb1XZSk7QfaaBwJukc8mHgDNsHVdC+aaC99mD7Cuj+CPi57etK6sxC+3TgSeCUvGt7YDHb2xbWnQx8pFMXKNsjn2V7vZK6XfpLkqIRRJrkq+0YOCaQdLLtnYbaF4xtYsWlx2hQYKrVas/7KmgEQQv2IU08PJFDa9ajb1a+JK+yvU8FncFo5aS2PbBuJzxL0n+T8uaKD1yAGyRtaPvqrL0BdQwgNgV2k3Q/8DR1J31WGzAwu6iSk9o3gcu7XPI2AT5bUjCHIHbTcSxcXtLyzgUpg2GlX90nSXMB6zdqS9CjxMBljNMqCXCgVW0QjCK+Zft0SRuTLGoPI4XTbFBY93eS3m/794V1BqOVk9p9JFORTl7JfKRBVA02AD4p6YG8vTwwpROvX3Ag0XLSp8lgzfYfJL2FNFi5kRQmV7ru1+xqbJmU5xMMA5L2Bb4BLCDpic5u4HngJ80aFvQkESoWBEEwjEi6IRdh/D5wi+1fdPYV0nuSvsK1C5Iu9h1r5BrW5ki6FricAU5qpd2gJJ1NytGbQHoP3p3b8UjWLzZwkrTC7F4vPTmTczxmOkF2wqgKaXWSp+chFWJ8IG+vANxe2mhlFiYyV9mOwcMoQtL3be/buh1BbxMDlyAIgmFE0u9IFt/vIoU5PEtyfiqa+9CSVk5qkmZbtLaxjW4RJG1JWg1YhjRAWwGYYnvN2f7hK9NsPUi7hT4TmTd3TGRsb1dSN2t3ClAub/uzklYlhcyVrOMyZsmOcf3KM9i+tF2Lgl4jBi5BEATDSL7ReS9pteUuSf8GvMn2+RW0P0qyBjbJGvjs0ppZ93vA/VR2UhuL5JySzYCJeWVvU2B720VzPlrSuGTAaSTDi0/aXkvSAqTVnuLaY41YWQvmhMhxCYIgGEZsP0NXNXHbD9OX2FsMST8mOfX9Mu/6nKR32/5iaW3gE/m5O8yjuB2ypLcD3yGtOsxNX6J6LRvmFrxg+1FJ4ySNs32RpINbN6owLUsGVC9AOYZpVow7GDnEwCUIgmB08A5gLedldEknknJOimO7WPX0ITge2IsBFtCjnMckLQRcCpwq6RFSQcpRS+NK8s/nVZbO92oVulYVg2Flmu1pkpA0n+07JK3WulFBbxEDlyAIgtHBnSRnq06+weuAm2sIS5qHVJhwk7zrYuBY2y/M8o+Gh8dtn1dYo9fYiuSithewA+km/sCmLapIg5IBUYCyHi1X1oIRQuS4BEEQjAJyjYvxwLV513hSdfNnAGxvWVD7OJLjVCcZfidguu1dS2lm3f8G5iKF5nXn1kSNjWDYiAKU9aldjDsYOcSKSxAEwehgv4ba4we4pl1YqTBhpzbOW7r2jeoaG9mA4WBgadKNdCevp7jt9Rjm7fStJgKEo1ghcuHPjsHIFTFoCQYSKy5BEATBK0LS9cA2tu/J2ysDv7Y9sPp48AqRdDfwIdtTWrdlLJBX9cYDp+Zd2wOTot7I8CNpP2Ab+sxNPgycYfugdq0Keo0YuARBEIxgJF1ue+OuQpQzX6JeAcrNgROAe7PuCsAuti8qrLsoKQehMxt+CXCg7cdL6rZE0hW23966HWMFSTcDb7Y9I2/PBdxge+22LRt9SJoCrGt7Wt5eALje9hvbtizoJSJULAiCYARje+P8vHALfUnjSEU2VyVVVRdwh+0azks/A24Fts3bO5EGUB+toF2VHCIGMCnXFjmb/nk9Zw36h8FwsBjQqUm0aMuGjHLuIxWenJa35wPuadaaoCeJFZcgCIIRjKQlZvd6jSKQkq6yvVFpnUF0X1KEsFZhwtpIOiH/aNLgsBvb/lTlJo0Jcv2W/wYuIr3vmwD72v5V04aNQiSdTQrLm0Dq5+8GLgceAbD9pXatC3qFWHEJgiAY2UzmpTezne3iRSAz50v6GHCW686GPStpY9uXw8yClM9W1K+G7V1gZn2ePWw/lrcXBw5r2bbRjO1fSrqYdEMtYB/bf2vbqlHLb/Kjw8WN2hH0MLHiEgRBMErIqy+rksItgDp1L3J+zYKkQojTqJRfI+nNJAvmTvjOVGBn21Xq17RA0g221x1qXzA85MHwjbaflrQjsB5whO37h/jT4F9A0rzAG/LmnRVqQQUjjFhxCYIgGAVI2hXYA1gOuJFUd+JKYPPS2rYXHmzQVIEpwCHAKqQ8hMdJTkSjduACjJO0uO2pMHOwGtfychwNrCNpHeCrpLyqk4B3NG3VKETSO0kTEfeRJj9eJ2ln25e2bFfQW8TJLgiCYHSwBymc5Wrbm0paHTighnDDQdNvgceA64G/FNbqFQ4DrpT0a1Io4LbA99o2aVTzom1L2go40vbxknZu3ahRymHAe2zfCSDpDcAvgfWbtiroKWLgEgRBMDqYZnuaJCTNZ/sOSatV0m41aFrO9nsr6PQMtk+SNIlUZFPAR23f3rhZo5knJe0L7Ahsku2Q52ncptHKPJ1BC4DtP0mK9zroRwxcgiAIRgcPSVqMZJM7QdJU4K+VtFsNmq6U9Cbbt1TQ6hnyQCUGK3XYDvgE8Gnbf5O0PHBo4zaNViZJOh44OW/vQDIfCYKZRHJ+EATBKEPSO0gJ63+w/XwFvd8AuwB7klYCppJmT99fWPd24PXAn0k1TTqmAFEcMAhGGJLmA74IbEz6Ll8K/LhSTahghBADlyAIgmDYqDlokrTCYPvD8Sl4pUi63PbG2TGv+0apimPeWCOH4J1oe8fWbQl6mxi4BEEQBEEQBE2R9EfgQzVWiYORS+S4BEEQBEEQBK25D7hC0jnA052dtg9v1qKg54iBSxAEQRAEQdCav+bHOGDhxm0JepQIFQuCIAiCIAiCoOeJFZcgCIIgCIKgCZLOpb8BQj9sb1mxOUGPEwOXIAiCIAiCoBU/yM8fBV4LnJK3tyflvQTBTCJULAiCIAiCIGiKpEttbzLUvmBsM651A4IgCIIgCIIxz1KSVu5sSFoJWKphe4IeJELFgiAIgiAIgtbsBVws6d68vSLw2XbNCXqRCBULgiAIgiAImiNpPmD1vHmH7edatifoPSJULAiCIAiCIGiKpEnAp4EHbN8Ug5ZgMGLgEgRBEARBELTm48CywHWSfiVpC0lq3aigt4hQsSAIgiAIgqAnkDQO+CBwNDAD+BlwhO1/Nm1Y0BPEiksQBEEQBEHQHElrA4cDhwJnAlsDTwAXtmxX0DuEq1gQBEEQBEHQFEmTgceA44B9unJcrpH09nYtC3qJCBULgiAIgiAImiJpDWBdYAW6JtZtH9isUUHPESsuQRAEQRAEQWsOJ624XA+Eo1gwKLHiEgRBEARBEDRF0q2212rdjqC3ieT8IAiCIAiCoDVXSnpT60YEvU2suARBEARBEARNkHQLYFL6wqrAvaRQMQG2vXbD5gU9RgxcgiAIgiAIgiZIWmF2r9u+v1Zbgt4nBi5BEARBEARBEPQ8keMSBEEQBEEQBEHPEwOXIAiCIAiCIAh6nhi4BEEQBEEQBEHQ88TAJQiCIAiCIAiCnicGLkEQBEEQBEEQ9Dz/HyUdDff+gh2dAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x137102dd8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,1, figsize=(15, 15))\n",
    "sns.heatmap(corr, annot=True, ax=ax, fmt='.2f', vmin=-1, vmax=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 222,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>40</th>\n",
       "      <th>41</th>\n",
       "      <th>42</th>\n",
       "      <th>43</th>\n",
       "      <th>44</th>\n",
       "      <th>45</th>\n",
       "      <th>46</th>\n",
       "      <th>47</th>\n",
       "      <th>48</th>\n",
       "      <th>49</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pfam_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PF00008</th>\n",
       "      <td>-0.059567</td>\n",
       "      <td>0.021568</td>\n",
       "      <td>0.124386</td>\n",
       "      <td>-0.113212</td>\n",
       "      <td>0.155973</td>\n",
       "      <td>-0.210214</td>\n",
       "      <td>0.139286</td>\n",
       "      <td>0.013310</td>\n",
       "      <td>-0.002099</td>\n",
       "      <td>0.137781</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.093701</td>\n",
       "      <td>0.084840</td>\n",
       "      <td>-0.047832</td>\n",
       "      <td>-0.048769</td>\n",
       "      <td>-0.272294</td>\n",
       "      <td>0.099531</td>\n",
       "      <td>-0.129423</td>\n",
       "      <td>0.071061</td>\n",
       "      <td>0.056498</td>\n",
       "      <td>-0.040379</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00706</th>\n",
       "      <td>0.126280</td>\n",
       "      <td>0.070211</td>\n",
       "      <td>-0.034045</td>\n",
       "      <td>0.228256</td>\n",
       "      <td>-0.053936</td>\n",
       "      <td>0.369830</td>\n",
       "      <td>-0.041520</td>\n",
       "      <td>0.079236</td>\n",
       "      <td>0.007799</td>\n",
       "      <td>-0.018709</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.112229</td>\n",
       "      <td>0.138848</td>\n",
       "      <td>-0.037965</td>\n",
       "      <td>-0.020669</td>\n",
       "      <td>-0.037405</td>\n",
       "      <td>-0.116916</td>\n",
       "      <td>-0.022610</td>\n",
       "      <td>0.091931</td>\n",
       "      <td>0.165369</td>\n",
       "      <td>-0.031008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00946</th>\n",
       "      <td>0.054680</td>\n",
       "      <td>-0.191446</td>\n",
       "      <td>-0.182523</td>\n",
       "      <td>-0.024010</td>\n",
       "      <td>0.254591</td>\n",
       "      <td>-0.047056</td>\n",
       "      <td>0.242332</td>\n",
       "      <td>0.013457</td>\n",
       "      <td>-0.078641</td>\n",
       "      <td>-0.022406</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001284</td>\n",
       "      <td>0.100193</td>\n",
       "      <td>0.013670</td>\n",
       "      <td>0.100773</td>\n",
       "      <td>-0.039986</td>\n",
       "      <td>0.253225</td>\n",
       "      <td>-0.024300</td>\n",
       "      <td>0.140542</td>\n",
       "      <td>0.042319</td>\n",
       "      <td>-0.116485</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 50 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               0         1         2         3         4         5         6   \\\n",
       "pfam_id                                                                         \n",
       "PF00008 -0.059567  0.021568  0.124386 -0.113212  0.155973 -0.210214  0.139286   \n",
       "PF00706  0.126280  0.070211 -0.034045  0.228256 -0.053936  0.369830 -0.041520   \n",
       "PF00946  0.054680 -0.191446 -0.182523 -0.024010  0.254591 -0.047056  0.242332   \n",
       "\n",
       "               7         8         9     ...           40        41        42  \\\n",
       "pfam_id                                  ...                                    \n",
       "PF00008  0.013310 -0.002099  0.137781    ...    -0.093701  0.084840 -0.047832   \n",
       "PF00706  0.079236  0.007799 -0.018709    ...    -0.112229  0.138848 -0.037965   \n",
       "PF00946  0.013457 -0.078641 -0.022406    ...     0.001284  0.100193  0.013670   \n",
       "\n",
       "               43        44        45        46        47        48        49  \n",
       "pfam_id                                                                        \n",
       "PF00008 -0.048769 -0.272294  0.099531 -0.129423  0.071061  0.056498 -0.040379  \n",
       "PF00706 -0.020669 -0.037405 -0.116916 -0.022610  0.091931  0.165369 -0.031008  \n",
       "PF00946  0.100773 -0.039986  0.253225 -0.024300  0.140542  0.042319 -0.116485  \n",
       "\n",
       "[3 rows x 50 columns]"
      ]
     },
     "execution_count": 222,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vecs_norm.loc[['PF00008','PF00706','PF00946']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x12528d940>"
      ]
     },
     "execution_count": 223,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1252a32b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = vecs_norm.plot.scatter(x=0, y=1)\n",
    "vecs_norm.loc[['PF00008']].plot.scatter(x=0, y=1, color='lightgreen', ax=ax)\n",
    "vecs_norm.loc[['PF00706']].plot.scatter(x=0, y=1, color='pink', ax=ax)\n",
    "vecs_norm.loc[['PF00946']].plot.scatter(x=0, y=1, color='red', ax=ax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>pfam_id</th>\n",
       "      <th>PF00008</th>\n",
       "      <th>PF00706</th>\n",
       "      <th>PF00946</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pfam_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PF00008</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.038059</td>\n",
       "      <td>0.198470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00706</th>\n",
       "      <td>-0.038059</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.061612</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00946</th>\n",
       "      <td>0.198470</td>\n",
       "      <td>0.061612</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "pfam_id   PF00008   PF00706   PF00946\n",
       "pfam_id                              \n",
       "PF00008  1.000000 -0.038059  0.198470\n",
       "PF00706 -0.038059  1.000000  0.061612\n",
       "PF00946  0.198470  0.061612  1.000000"
      ]
     },
     "execution_count": 224,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vecs_norm.loc[['PF00008','PF00706','PF00946']].transpose().corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 217,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "index = np.intersect1d(d_norm[:1000].index, vecs_norm.index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 226,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>PF00001</th>\n",
       "      <th>PF00002</th>\n",
       "      <th>PF00003</th>\n",
       "      <th>PF00004</th>\n",
       "      <th>PF00005</th>\n",
       "      <th>PF00006</th>\n",
       "      <th>PF00008</th>\n",
       "      <th>PF00009</th>\n",
       "      <th>PF00010</th>\n",
       "      <th>PF00011</th>\n",
       "      <th>...</th>\n",
       "      <th>PF01034</th>\n",
       "      <th>PF01035</th>\n",
       "      <th>PF01036</th>\n",
       "      <th>PF01037</th>\n",
       "      <th>PF01039</th>\n",
       "      <th>PF01040</th>\n",
       "      <th>PF01041</th>\n",
       "      <th>PF01042</th>\n",
       "      <th>PF01043</th>\n",
       "      <th>PF01044</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PF00001</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.994152</td>\n",
       "      <td>0.996767</td>\n",
       "      <td>0.858155</td>\n",
       "      <td>0.916370</td>\n",
       "      <td>0.967561</td>\n",
       "      <td>0.219116</td>\n",
       "      <td>0.940149</td>\n",
       "      <td>0.229814</td>\n",
       "      <td>0.631372</td>\n",
       "      <td>...</td>\n",
       "      <td>0.479600</td>\n",
       "      <td>0.690161</td>\n",
       "      <td>0.928384</td>\n",
       "      <td>0.569138</td>\n",
       "      <td>0.961168</td>\n",
       "      <td>0.990399</td>\n",
       "      <td>0.954692</td>\n",
       "      <td>0.850712</td>\n",
       "      <td>0.767659</td>\n",
       "      <td>0.899831</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00002</th>\n",
       "      <td>0.994152</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.994071</td>\n",
       "      <td>0.868736</td>\n",
       "      <td>0.914337</td>\n",
       "      <td>0.949893</td>\n",
       "      <td>0.262220</td>\n",
       "      <td>0.920265</td>\n",
       "      <td>0.264869</td>\n",
       "      <td>0.639796</td>\n",
       "      <td>...</td>\n",
       "      <td>0.539192</td>\n",
       "      <td>0.727421</td>\n",
       "      <td>0.955501</td>\n",
       "      <td>0.607857</td>\n",
       "      <td>0.929704</td>\n",
       "      <td>0.990500</td>\n",
       "      <td>0.921738</td>\n",
       "      <td>0.866706</td>\n",
       "      <td>0.769727</td>\n",
       "      <td>0.858468</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00003</th>\n",
       "      <td>0.996767</td>\n",
       "      <td>0.994071</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.863185</td>\n",
       "      <td>0.913619</td>\n",
       "      <td>0.964204</td>\n",
       "      <td>0.240055</td>\n",
       "      <td>0.929600</td>\n",
       "      <td>0.225174</td>\n",
       "      <td>0.620210</td>\n",
       "      <td>...</td>\n",
       "      <td>0.498167</td>\n",
       "      <td>0.704774</td>\n",
       "      <td>0.945078</td>\n",
       "      <td>0.576428</td>\n",
       "      <td>0.951380</td>\n",
       "      <td>0.996520</td>\n",
       "      <td>0.940415</td>\n",
       "      <td>0.858530</td>\n",
       "      <td>0.749562</td>\n",
       "      <td>0.880369</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00004</th>\n",
       "      <td>0.858155</td>\n",
       "      <td>0.868736</td>\n",
       "      <td>0.863185</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.983022</td>\n",
       "      <td>0.920366</td>\n",
       "      <td>0.607143</td>\n",
       "      <td>0.921873</td>\n",
       "      <td>0.588337</td>\n",
       "      <td>0.904199</td>\n",
       "      <td>...</td>\n",
       "      <td>0.808621</td>\n",
       "      <td>0.936848</td>\n",
       "      <td>0.907961</td>\n",
       "      <td>0.884591</td>\n",
       "      <td>0.810347</td>\n",
       "      <td>0.844733</td>\n",
       "      <td>0.807403</td>\n",
       "      <td>0.993223</td>\n",
       "      <td>0.935176</td>\n",
       "      <td>0.787006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00005</th>\n",
       "      <td>0.916370</td>\n",
       "      <td>0.914337</td>\n",
       "      <td>0.913619</td>\n",
       "      <td>0.983022</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.967823</td>\n",
       "      <td>0.478582</td>\n",
       "      <td>0.970691</td>\n",
       "      <td>0.522997</td>\n",
       "      <td>0.874438</td>\n",
       "      <td>...</td>\n",
       "      <td>0.701140</td>\n",
       "      <td>0.867017</td>\n",
       "      <td>0.901900</td>\n",
       "      <td>0.809053</td>\n",
       "      <td>0.891569</td>\n",
       "      <td>0.892440</td>\n",
       "      <td>0.892491</td>\n",
       "      <td>0.969539</td>\n",
       "      <td>0.938878</td>\n",
       "      <td>0.879609</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 941 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "Unnamed: 0   PF00001   PF00002   PF00003   PF00004   PF00005   PF00006  \\\n",
       "Unnamed: 0                                                               \n",
       "PF00001     1.000000  0.994152  0.996767  0.858155  0.916370  0.967561   \n",
       "PF00002     0.994152  1.000000  0.994071  0.868736  0.914337  0.949893   \n",
       "PF00003     0.996767  0.994071  1.000000  0.863185  0.913619  0.964204   \n",
       "PF00004     0.858155  0.868736  0.863185  1.000000  0.983022  0.920366   \n",
       "PF00005     0.916370  0.914337  0.913619  0.983022  1.000000  0.967823   \n",
       "\n",
       "Unnamed: 0   PF00008   PF00009   PF00010   PF00011    ...      PF01034  \\\n",
       "Unnamed: 0                                            ...                \n",
       "PF00001     0.219116  0.940149  0.229814  0.631372    ...     0.479600   \n",
       "PF00002     0.262220  0.920265  0.264869  0.639796    ...     0.539192   \n",
       "PF00003     0.240055  0.929600  0.225174  0.620210    ...     0.498167   \n",
       "PF00004     0.607143  0.921873  0.588337  0.904199    ...     0.808621   \n",
       "PF00005     0.478582  0.970691  0.522997  0.874438    ...     0.701140   \n",
       "\n",
       "Unnamed: 0   PF01035   PF01036   PF01037   PF01039   PF01040   PF01041  \\\n",
       "Unnamed: 0                                                               \n",
       "PF00001     0.690161  0.928384  0.569138  0.961168  0.990399  0.954692   \n",
       "PF00002     0.727421  0.955501  0.607857  0.929704  0.990500  0.921738   \n",
       "PF00003     0.704774  0.945078  0.576428  0.951380  0.996520  0.940415   \n",
       "PF00004     0.936848  0.907961  0.884591  0.810347  0.844733  0.807403   \n",
       "PF00005     0.867017  0.901900  0.809053  0.891569  0.892440  0.892491   \n",
       "\n",
       "Unnamed: 0   PF01042   PF01043   PF01044  \n",
       "Unnamed: 0                                \n",
       "PF00001     0.850712  0.767659  0.899831  \n",
       "PF00002     0.866706  0.769727  0.858468  \n",
       "PF00003     0.858530  0.749562  0.880369  \n",
       "PF00004     0.993223  0.935176  0.787006  \n",
       "PF00005     0.969539  0.938878  0.879609  \n",
       "\n",
       "[5 rows x 941 columns]"
      ]
     },
     "execution_count": 226,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x = d_norm.loc[index].transpose().corr()\n",
    "x.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>pfam_id</th>\n",
       "      <th>PF00001</th>\n",
       "      <th>PF00002</th>\n",
       "      <th>PF00003</th>\n",
       "      <th>PF00004</th>\n",
       "      <th>PF00005</th>\n",
       "      <th>PF00006</th>\n",
       "      <th>PF00008</th>\n",
       "      <th>PF00009</th>\n",
       "      <th>PF00010</th>\n",
       "      <th>PF00011</th>\n",
       "      <th>...</th>\n",
       "      <th>PF01034</th>\n",
       "      <th>PF01035</th>\n",
       "      <th>PF01036</th>\n",
       "      <th>PF01037</th>\n",
       "      <th>PF01039</th>\n",
       "      <th>PF01040</th>\n",
       "      <th>PF01041</th>\n",
       "      <th>PF01042</th>\n",
       "      <th>PF01043</th>\n",
       "      <th>PF01044</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pfam_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PF00001</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.008835</td>\n",
       "      <td>0.313048</td>\n",
       "      <td>0.065194</td>\n",
       "      <td>-0.029711</td>\n",
       "      <td>0.159311</td>\n",
       "      <td>-0.006431</td>\n",
       "      <td>0.141462</td>\n",
       "      <td>-0.001737</td>\n",
       "      <td>0.132200</td>\n",
       "      <td>...</td>\n",
       "      <td>0.493384</td>\n",
       "      <td>0.074946</td>\n",
       "      <td>0.180001</td>\n",
       "      <td>0.038872</td>\n",
       "      <td>-0.080016</td>\n",
       "      <td>0.266598</td>\n",
       "      <td>-0.162764</td>\n",
       "      <td>-0.276095</td>\n",
       "      <td>-0.000242</td>\n",
       "      <td>0.056733</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00002</th>\n",
       "      <td>0.008835</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.047111</td>\n",
       "      <td>-0.034849</td>\n",
       "      <td>0.105328</td>\n",
       "      <td>-0.034632</td>\n",
       "      <td>-0.463499</td>\n",
       "      <td>0.021370</td>\n",
       "      <td>0.291554</td>\n",
       "      <td>-0.229693</td>\n",
       "      <td>...</td>\n",
       "      <td>0.016806</td>\n",
       "      <td>-0.067179</td>\n",
       "      <td>0.021102</td>\n",
       "      <td>-0.075207</td>\n",
       "      <td>-0.101602</td>\n",
       "      <td>-0.183917</td>\n",
       "      <td>-0.075302</td>\n",
       "      <td>-0.124115</td>\n",
       "      <td>-0.298236</td>\n",
       "      <td>0.153560</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00003</th>\n",
       "      <td>0.313048</td>\n",
       "      <td>0.047111</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.002882</td>\n",
       "      <td>-0.072720</td>\n",
       "      <td>-0.116677</td>\n",
       "      <td>0.057312</td>\n",
       "      <td>-0.098613</td>\n",
       "      <td>0.141267</td>\n",
       "      <td>0.311389</td>\n",
       "      <td>...</td>\n",
       "      <td>0.471532</td>\n",
       "      <td>0.156334</td>\n",
       "      <td>0.398627</td>\n",
       "      <td>-0.189345</td>\n",
       "      <td>-0.116935</td>\n",
       "      <td>0.105283</td>\n",
       "      <td>-0.068048</td>\n",
       "      <td>-0.056163</td>\n",
       "      <td>0.101469</td>\n",
       "      <td>0.048376</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00004</th>\n",
       "      <td>0.065194</td>\n",
       "      <td>-0.034849</td>\n",
       "      <td>0.002882</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.299206</td>\n",
       "      <td>0.130528</td>\n",
       "      <td>0.069408</td>\n",
       "      <td>0.170381</td>\n",
       "      <td>0.243991</td>\n",
       "      <td>-0.052491</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001301</td>\n",
       "      <td>-0.097694</td>\n",
       "      <td>-0.209356</td>\n",
       "      <td>-0.000954</td>\n",
       "      <td>-0.041439</td>\n",
       "      <td>-0.144040</td>\n",
       "      <td>-0.098941</td>\n",
       "      <td>-0.006326</td>\n",
       "      <td>0.088771</td>\n",
       "      <td>-0.166710</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00005</th>\n",
       "      <td>-0.029711</td>\n",
       "      <td>0.105328</td>\n",
       "      <td>-0.072720</td>\n",
       "      <td>0.299206</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.192628</td>\n",
       "      <td>-0.086249</td>\n",
       "      <td>0.145821</td>\n",
       "      <td>0.082489</td>\n",
       "      <td>-0.283238</td>\n",
       "      <td>...</td>\n",
       "      <td>0.048513</td>\n",
       "      <td>-0.066837</td>\n",
       "      <td>-0.168083</td>\n",
       "      <td>-0.121035</td>\n",
       "      <td>-0.288996</td>\n",
       "      <td>-0.150944</td>\n",
       "      <td>-0.026100</td>\n",
       "      <td>-0.117201</td>\n",
       "      <td>-0.099841</td>\n",
       "      <td>-0.002272</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 941 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "pfam_id   PF00001   PF00002   PF00003   PF00004   PF00005   PF00006   PF00008  \\\n",
       "pfam_id                                                                         \n",
       "PF00001  1.000000  0.008835  0.313048  0.065194 -0.029711  0.159311 -0.006431   \n",
       "PF00002  0.008835  1.000000  0.047111 -0.034849  0.105328 -0.034632 -0.463499   \n",
       "PF00003  0.313048  0.047111  1.000000  0.002882 -0.072720 -0.116677  0.057312   \n",
       "PF00004  0.065194 -0.034849  0.002882  1.000000  0.299206  0.130528  0.069408   \n",
       "PF00005 -0.029711  0.105328 -0.072720  0.299206  1.000000  0.192628 -0.086249   \n",
       "\n",
       "pfam_id   PF00009   PF00010   PF00011    ...      PF01034   PF01035   PF01036  \\\n",
       "pfam_id                                  ...                                    \n",
       "PF00001  0.141462 -0.001737  0.132200    ...     0.493384  0.074946  0.180001   \n",
       "PF00002  0.021370  0.291554 -0.229693    ...     0.016806 -0.067179  0.021102   \n",
       "PF00003 -0.098613  0.141267  0.311389    ...     0.471532  0.156334  0.398627   \n",
       "PF00004  0.170381  0.243991 -0.052491    ...     0.001301 -0.097694 -0.209356   \n",
       "PF00005  0.145821  0.082489 -0.283238    ...     0.048513 -0.066837 -0.168083   \n",
       "\n",
       "pfam_id   PF01037   PF01039   PF01040   PF01041   PF01042   PF01043   PF01044  \n",
       "pfam_id                                                                        \n",
       "PF00001  0.038872 -0.080016  0.266598 -0.162764 -0.276095 -0.000242  0.056733  \n",
       "PF00002 -0.075207 -0.101602 -0.183917 -0.075302 -0.124115 -0.298236  0.153560  \n",
       "PF00003 -0.189345 -0.116935  0.105283 -0.068048 -0.056163  0.101469  0.048376  \n",
       "PF00004 -0.000954 -0.041439 -0.144040 -0.098941 -0.006326  0.088771 -0.166710  \n",
       "PF00005 -0.121035 -0.288996 -0.150944 -0.026100 -0.117201 -0.099841 -0.002272  \n",
       "\n",
       "[5 rows x 941 columns]"
      ]
     },
     "execution_count": 227,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y = vecs_norm.loc[index].transpose().corr()\n",
    "y.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 228,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1.        , 0.02090249],\n",
       "       [0.02090249, 1.        ]])"
      ]
     },
     "execution_count": 228,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.corrcoef(x.values.reshape(1, -1), y.values.reshape(1, -1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 240,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PF00001</th>\n",
       "      <td>0.257560</td>\n",
       "      <td>0.422263</td>\n",
       "      <td>0.324609</td>\n",
       "      <td>0.140257</td>\n",
       "      <td>0.578019</td>\n",
       "      <td>0.312418</td>\n",
       "      <td>0.134810</td>\n",
       "      <td>0.092467</td>\n",
       "      <td>0.042343</td>\n",
       "      <td>0.109196</td>\n",
       "      <td>...</td>\n",
       "      <td>0.059766</td>\n",
       "      <td>0.245843</td>\n",
       "      <td>0.130252</td>\n",
       "      <td>0.055787</td>\n",
       "      <td>0.038053</td>\n",
       "      <td>0.017734</td>\n",
       "      <td>0.032965</td>\n",
       "      <td>0.000513</td>\n",
       "      <td>0.142897</td>\n",
       "      <td>0.002333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00002</th>\n",
       "      <td>0.239458</td>\n",
       "      <td>0.387237</td>\n",
       "      <td>0.337535</td>\n",
       "      <td>0.149027</td>\n",
       "      <td>0.583157</td>\n",
       "      <td>0.284088</td>\n",
       "      <td>0.124840</td>\n",
       "      <td>0.084501</td>\n",
       "      <td>0.040339</td>\n",
       "      <td>0.117296</td>\n",
       "      <td>...</td>\n",
       "      <td>0.072245</td>\n",
       "      <td>0.283909</td>\n",
       "      <td>0.137424</td>\n",
       "      <td>0.060679</td>\n",
       "      <td>0.042485</td>\n",
       "      <td>0.018195</td>\n",
       "      <td>0.037736</td>\n",
       "      <td>0.000618</td>\n",
       "      <td>0.151705</td>\n",
       "      <td>0.003402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00003</th>\n",
       "      <td>0.275416</td>\n",
       "      <td>0.426772</td>\n",
       "      <td>0.337934</td>\n",
       "      <td>0.117172</td>\n",
       "      <td>0.582872</td>\n",
       "      <td>0.284450</td>\n",
       "      <td>0.110803</td>\n",
       "      <td>0.074865</td>\n",
       "      <td>0.035938</td>\n",
       "      <td>0.119696</td>\n",
       "      <td>...</td>\n",
       "      <td>0.050753</td>\n",
       "      <td>0.253351</td>\n",
       "      <td>0.123576</td>\n",
       "      <td>0.048145</td>\n",
       "      <td>0.032699</td>\n",
       "      <td>0.015446</td>\n",
       "      <td>0.033837</td>\n",
       "      <td>0.000462</td>\n",
       "      <td>0.131193</td>\n",
       "      <td>0.003194</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00004</th>\n",
       "      <td>0.219951</td>\n",
       "      <td>0.367178</td>\n",
       "      <td>0.247751</td>\n",
       "      <td>0.051263</td>\n",
       "      <td>0.408322</td>\n",
       "      <td>0.298013</td>\n",
       "      <td>0.179808</td>\n",
       "      <td>0.090961</td>\n",
       "      <td>0.088848</td>\n",
       "      <td>0.190955</td>\n",
       "      <td>...</td>\n",
       "      <td>0.043591</td>\n",
       "      <td>0.355876</td>\n",
       "      <td>0.255717</td>\n",
       "      <td>0.154273</td>\n",
       "      <td>0.079649</td>\n",
       "      <td>0.074623</td>\n",
       "      <td>0.039412</td>\n",
       "      <td>0.000353</td>\n",
       "      <td>0.182238</td>\n",
       "      <td>0.000200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PF00005</th>\n",
       "      <td>0.230318</td>\n",
       "      <td>0.386793</td>\n",
       "      <td>0.266676</td>\n",
       "      <td>0.055222</td>\n",
       "      <td>0.440644</td>\n",
       "      <td>0.340842</td>\n",
       "      <td>0.194525</td>\n",
       "      <td>0.103833</td>\n",
       "      <td>0.090692</td>\n",
       "      <td>0.160417</td>\n",
       "      <td>...</td>\n",
       "      <td>0.038679</td>\n",
       "      <td>0.308291</td>\n",
       "      <td>0.239894</td>\n",
       "      <td>0.136853</td>\n",
       "      <td>0.072833</td>\n",
       "      <td>0.064020</td>\n",
       "      <td>0.040015</td>\n",
       "      <td>0.000297</td>\n",
       "      <td>0.195975</td>\n",
       "      <td>-0.000628</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  0         1         2         3         4         5   \\\n",
       "Unnamed: 0                                                               \n",
       "PF00001     0.257560  0.422263  0.324609  0.140257  0.578019  0.312418   \n",
       "PF00002     0.239458  0.387237  0.337535  0.149027  0.583157  0.284088   \n",
       "PF00003     0.275416  0.426772  0.337934  0.117172  0.582872  0.284450   \n",
       "PF00004     0.219951  0.367178  0.247751  0.051263  0.408322  0.298013   \n",
       "PF00005     0.230318  0.386793  0.266676  0.055222  0.440644  0.340842   \n",
       "\n",
       "                  6         7         8         9     ...           12  \\\n",
       "Unnamed: 0                                            ...                \n",
       "PF00001     0.134810  0.092467  0.042343  0.109196    ...     0.059766   \n",
       "PF00002     0.124840  0.084501  0.040339  0.117296    ...     0.072245   \n",
       "PF00003     0.110803  0.074865  0.035938  0.119696    ...     0.050753   \n",
       "PF00004     0.179808  0.090961  0.088848  0.190955    ...     0.043591   \n",
       "PF00005     0.194525  0.103833  0.090692  0.160417    ...     0.038679   \n",
       "\n",
       "                  13        14        15        16        17        18  \\\n",
       "Unnamed: 0                                                               \n",
       "PF00001     0.245843  0.130252  0.055787  0.038053  0.017734  0.032965   \n",
       "PF00002     0.283909  0.137424  0.060679  0.042485  0.018195  0.037736   \n",
       "PF00003     0.253351  0.123576  0.048145  0.032699  0.015446  0.033837   \n",
       "PF00004     0.355876  0.255717  0.154273  0.079649  0.074623  0.039412   \n",
       "PF00005     0.308291  0.239894  0.136853  0.072833  0.064020  0.040015   \n",
       "\n",
       "                  19        20        21  \n",
       "Unnamed: 0                                \n",
       "PF00001     0.000513  0.142897  0.002333  \n",
       "PF00002     0.000618  0.151705  0.003402  \n",
       "PF00003     0.000462  0.131193  0.003194  \n",
       "PF00004     0.000353  0.182238  0.000200  \n",
       "PF00005     0.000297  0.195975 -0.000628  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "execution_count": 240,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vecs_norm.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}