62_VGG_CAM_prediction.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "!pip install --upgrade pip matplotlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "path_all_tfrecord = \"fp56.tfrecord\"\n",
    "\n",
    "# path_train_tfrecord = \"fp56_train.tfrecord\"\n",
    "# path_test_tfrecord = \"fp56_test.tfrecord\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'vgg_cam/model-17-1.17-53.3%.hdf5'"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dir_model = \"vgg_cam/\"\n",
    "path_best = dir_model + \"model-17-1.17-53.3%.hdf5\"\n",
    "path_best"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from fp_tensorflow import _parse_pair_56, _parse_single_56\n",
    "from fp_tensorflow import create_pair_56_dataset, create_single_dataset\n",
    "from fp_tensorflow import VGG16_convolutions\n",
    "from fp_tensorflow import create_vgg_5y_model\n",
    "\n",
    "all_dataset = create_pair_56_dataset(path_all_tfrecord, \"floorplan\", \"year\").batch(64)\n",
    "\n",
    "model = create_vgg_5y_model()\n",
    "model.load_weights(path_best)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    786/Unknown - 283s 360ms/step"
     ]
    }
   ],
   "source": [
    "predictions = model.predict(all_dataset, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = np.argmax(predictions, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([6, 6, 7, 6, 4, 4, 3, 6, 4, 3, 3, 7, 4, 4, 6, 6, 4, 4, 4, 6, 4, 4,\n",
       "       4, 6, 4, 4, 4, 3, 3, 3, 3, 4, 4, 4, 4, 4, 5, 5, 6, 5, 6, 9, 6, 4,\n",
       "       4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 3, 4, 4, 4, 4, 4, 4, 3, 4, 7, 7,\n",
       "       7, 4, 8, 8, 7, 7, 8, 7, 7, 7, 7, 7, 3, 3, 6, 6, 6, 7, 7, 7, 7, 7,\n",
       "       7, 7, 8, 8, 4, 5, 7, 4, 7, 7, 6, 8, 7, 7, 7, 5, 4, 4, 4, 7, 7, 7,\n",
       "       7, 7, 7, 7, 7, 7, 6, 7, 6, 7, 4, 6, 4, 8, 7, 8, 6, 8])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions[:128]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((50252,), array([6, 6, 6, 6, 2, 2, 2, 2, 2, 2]))"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_year = create_single_dataset(path_all_tfrecord, \"year\")\n",
    "year_true = np.fromiter((y.numpy() for y in all_year), int)\n",
    "year_true.shape, year_true[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['8928_160',\n",
       " '8928_183',\n",
       " '8928_201A',\n",
       " '8928_204B',\n",
       " '960_35',\n",
       " '960_41',\n",
       " '960_45',\n",
       " '960_49F',\n",
       " '960_50S',\n",
       " '960_53']"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_id = create_single_dataset(path_all_tfrecord, \"plan_id\")\n",
    "ids = [i.numpy().decode() for i in all_id]\n",
    "ids[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>true</th>\n",
       "      <th>prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8928_160</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8928_183</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8928_201A</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>8928_204B</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>960_35</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50247</th>\n",
       "      <td>115454_21D</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50248</th>\n",
       "      <td>115454_34E</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50249</th>\n",
       "      <td>115454_36B</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50250</th>\n",
       "      <td>115454_37A</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50251</th>\n",
       "      <td>115454_38C</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>50252 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               ID  true  prediction\n",
       "0        8928_160     6           6\n",
       "1        8928_183     6           6\n",
       "2       8928_201A     6           7\n",
       "3       8928_204B     6           6\n",
       "4          960_35     2           4\n",
       "...           ...   ...         ...\n",
       "50247  115454_21D     9           9\n",
       "50248  115454_34E     9           9\n",
       "50249  115454_36B     9           9\n",
       "50250  115454_37A     9           9\n",
       "50251  115454_38C     9           9\n",
       "\n",
       "[50252 rows x 3 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.DataFrame(\n",
    "    zip(ids, year_true, predictions), columns=[\"ID\", \"true\", \"prediction\"],\n",
    ")\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv(\"vgg_5y_prediction.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>prediction</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>true</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>16</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>20</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>12</td>\n",
       "      <td>147</td>\n",
       "      <td>14</td>\n",
       "      <td>18</td>\n",
       "      <td>29</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>69</td>\n",
       "      <td>482</td>\n",
       "      <td>36</td>\n",
       "      <td>36</td>\n",
       "      <td>25</td>\n",
       "      <td>24</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>406</td>\n",
       "      <td>1261</td>\n",
       "      <td>168</td>\n",
       "      <td>122</td>\n",
       "      <td>55</td>\n",
       "      <td>94</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>97</td>\n",
       "      <td>3040</td>\n",
       "      <td>727</td>\n",
       "      <td>396</td>\n",
       "      <td>171</td>\n",
       "      <td>131</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>23</td>\n",
       "      <td>1989</td>\n",
       "      <td>2311</td>\n",
       "      <td>1548</td>\n",
       "      <td>373</td>\n",
       "      <td>191</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>766</td>\n",
       "      <td>863</td>\n",
       "      <td>3487</td>\n",
       "      <td>2924</td>\n",
       "      <td>668</td>\n",
       "      <td>87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>155</td>\n",
       "      <td>74</td>\n",
       "      <td>940</td>\n",
       "      <td>8933</td>\n",
       "      <td>2072</td>\n",
       "      <td>151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>12</td>\n",
       "      <td>3</td>\n",
       "      <td>71</td>\n",
       "      <td>1489</td>\n",
       "      <td>6819</td>\n",
       "      <td>1618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>21</td>\n",
       "      <td>853</td>\n",
       "      <td>4025</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "prediction  0  1  2    3     4     5     6     7     8     9\n",
       "true                                                        \n",
       "0           0  0  1   16    13     0     4    20     8     0\n",
       "1           0  0  4   12   147    14    18    29     6     3\n",
       "2           0  0  6   69   482    36    36    25    24    10\n",
       "3           0  0  3  406  1261   168   122    55    94    60\n",
       "4           0  0  0   97  3040   727   396   171   131    26\n",
       "5           0  0  0   23  1989  2311  1548   373   191    25\n",
       "6           0  0  0    9   766   863  3487  2924   668    87\n",
       "7           0  0  0    1   155    74   940  8933  2072   151\n",
       "8           0  0  0    1    12     3    71  1489  6819  1618\n",
       "9           0  0  0    0     1     0     9    21   853  4025"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crosstab = pd.crosstab(df.true, df.prediction)\n",
    "crosstab[0] = 0\n",
    "crosstab = crosstab.reindex(index=range(10), columns=range(10), fill_value=0)\n",
    "crosstab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABnsAAAU6CAYAAAAqTsNSAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAC4jAAAuIwF4pT92AAD5/ElEQVR4nOzdebjlV1Un7s+qVBJCEiBAIkOEkAAxDIIytaAQZhVakMkBBRnU1nYAWxR/rQLaKiLaqC0OGAm2zSi2gKKISIiSbiZBmYcwjwkzCQmZ1u+Pc6pz6ps7nHPr3nPv9+R9ec5TtffZe6+9UlV5eGpl713dHQAAAAAAAMZp325vAAAAAAAAgK1T7AEAAAAAABgxxR4AAAAAAIARU+wBAAAAAAAYMcUeAAAAAACAEVPsAQAAAAAAGDHFHgAAAAAAgBFT7AEAAAAAABgxxR4AAAAAAIARU+wBAAAAAAAYMcUeAAAAAACAEVPsAQAAAAAAGDHFHgAAAAAAgBFT7AEAAAAAABgxxR4AAAAAAIARU+wBAAAAAAAYMcUeAAAAAACAEVPsAQAAAAAAGDHFHgAAAAAAgBFT7AEAAAAAABgxxR4AAAAAAIARU+wBAAAAAAAYMcUeAAAAAACAEVPsAQAAAAAAGDHFHgAAAAAAgBFT7AEAAAAAABgxxR4AAAAAAIARU+wBAAAAAAAYMcUeAAAAAACAEVPsAQAAAAAAGLH9u72Bq7uqqiQnJbltkhOTXCfJ15J8Icn7k7ypuy/e5pjHJrlbklsmuVaSi5J8JMk53f3JbVh/X5KbZ5LTDWdifD7Ju5O8tbsvPdQ4g5jXyySnU5IcneTCJOcmeX13f247YwEAAAAAwF5S3b3be7jaqarjkjw4ybcnuVeS628w/NIkf5vkWd39ukOMe7Mkv5LkEUmOWGNIJ3ldkqd099kLrn2DJA9Ncr8kp2dS4FnPRUn+Msl/7+63LhJnjbi3yySnB2btk2qXZ/LP75e6+98PJRYAAAAAAOxFij1LVlV/kOTxWbvYspk/T/KT3f3lLcR9RJLnJrnmHMM7yTOS/ELP8Rukql6W9YstG7kiye8k+a/dfcmCc1NVP53kmZnvhNplSX6mu39/0TgAAAAAALCXKfYsWVW9Ockd1vjq8iSfSvKZJIcnuWmSa68x7o1J7t3dFywQ8+FJXpirFmPOT/KxJCckuXGSGnz/rO5+4hzrfzbJ9db46tIkn5zGuUaSk7N2senlSR7a3ZdtFmsm5s8k+e01vvrUNOaNMrlCbuinu/v35o0DAAAAAAB73aInMdheX0zy7CQPSHJcd399d9+xu2+XSfHknkn+eTDnzknOnDdAVZ2SyYme2V/rf0tyr+4+obvv0N1fn+S0JH81mP6EqnrIAvkkk2LVbyW5d5JrdfdJ3X2n7r5tJu8RfVeStw/mfFeSp88boKrumsnJo1lnJblDd99o+s/wRknulMm1dLN+u6ruPG8sAAAAAADY65zsWbLpyZ7rJflvSZ7f3RdtMv6wTApCPzL46l7d/do54j0/yffNdL0pyX3WugquqirJHw1inZvkGzY6dTM92fOpJE9L8tebndCpqmskeUkmV78dcGmS23T3+zbOKKmq1ye560zXK5I8bK2r4KrqiEyKWA+Y6T67u++xWRwAAAAAABgDxZ4lq6oHJHn1Im/UTAs+/zfJHWe6n9/dj9xk3q2T/HuuPNVzSZLbd/e7N5hzjemcW8x0/0h3P2eDOQ9K8oruvmLDRA6ec3SS9yQ5cab717v7v24y7zuSvHKm63NJTuvu8zeYc0KSd+Xgq+bu192vnne/AAAAAACwV7nGbcm6+28XKfRM51yeq15bdv85pj42B/8av3CjQs801sW56pVqj99kzssWKfRM51yYZPh2zjw5DffyBxsVeqaxzsvkdNRG6wAAAAAAwCg52TMSVXWDTK5Km3V0d391gznvT3Lzma57dPfZc8Q6OpO3d46ednWSE7v7k4vtetM4/yHJ/5npOr+7T9hg/JGZnOQ5eqb7pO7+yByxTkryoZmuryS5/qKFNwAAAAAA2Guc7BmPL6zRd+31BlfVqTm40HNhknPmCTQ9dTM7tnLwmzfbZZjTuvlMnZ6DCz3vnafQkyTd/eEk75/pOjaJd3sAAAAAABg9xZ7xuPEafZ/bYPztB+03dvdlC8R7/SbrbYdhThvls9Ye5ipezVhGTgAAAAAAsFSKPePxbYP2Rza5guy0QftdC8Ybjh+utx2GOb1vk/FjyAkAAAAAAJZKsWc8Hjtov3KT8acO2h9bMN5w/HC9Q1JVhyV51KB71DkBAAAAAMBu2L/bG2BzVfWdSe4+6D5zk2knDNofXzDsJwbt4xecv5kfTXLyTPvSJM/fZM525zRcb0uq6oQs/s/nmCR3TPLlJF/KpBC10UktAAAAAHbWEUm+fqb9uu7+0m5tZqyq6hpJTtntfSzRud198W5vAhR79riqum6SPx50/3V3v3GTqccM2hcuGHo4/vCqOrK7v7bgOldRVackefqg+w+6e7PizXbnNFxvq348yVO2aS0AAAAA9oYHJXn5bm9ihE5J8o7d3sQS3SbJO3d7E+Aatz2sqvYl+YskJ850fynJT80xfVjIWLS6fNEcay6sqq6Z5C+THDvT/ZEkvzzH9O3OabuKPQAAAAAAsGsUe/a230ryHYO+H+3ued6qucagvegVYWud4DlqwTUOUlWV5HlJbj/TfVmSR3b3V+ZYYrtzOqR8AAAAAABgL3CN2x5VVT+V5GcG3c/o7hfNucTw1MsRC27hyDnWXNRvJ3nYoO+nu/v1c86/OMk1Z9qHmtN23aX57CQvWXDON2RywilJ8uKX/lVOufnNt2k7AFcvvdsb4JDVbm8ASOLfp6ug26/i2J138VqXbDA2f/3hN+/2Ftiiz3/sU3npk35ttmue/+CazXzjdZNrrtBfQ3/1suTfP7/bu4CrWKE/Zaujqr4/ybMG3WcmefICy1wwaA9PxWxmrVMvwzXnVlVPTvLEQffTuvvZCyxzQQ4u9hxqTlvOZ1Z3n5fkvEXmTA45XemUm988t7r1rbdjOwBXO/5aa/wUe2Bv8O/T8VPsGb9rf3XRp2nZi44/7Pzd3gLbZ9FbZVjLNfcnxxy+27uAlecatz2mqh6YyVVns3/v8VdJHt+L/T/3YSHj6AW3Mhx/WXdv6SRMVf1okt8YdP9edz91waW2O6dtKfYAAAAAAMBucrJnD6mqe2ZyHdjsr8urk3xfd1++4HLDkyYnLjj/xoP2lv6zlOkppeHpneclecIWljsvyckz7UPNaaHTOAAAAADAgqomn1WxSrmwUpzs2SOq6i5JXp6DryY7J8l3d/dWjoy+d9C+yYLzh+Pfs+gGqupBmRR2Zn+fvTTJ4xY8pXTArucEAAAAAAB7jWLPHlBV35jk75IcM9P91iTf2d1bvbB3WMi41YLzT9tkvQ1V1X2SvCgHn1J6VZLv38IppfX2sNScAAAAAABgL1Ls2WVVdWomV7UdN9P97iT37+4vHcLSbxu071RVi1zbd7dN1ltXVd0tycuSHDnT/c/Z+iml9fZw1wXnbzknAAAAAADYqxR7dlFV3TTJPyY5Yab7Q0nu291beiPngO5+T5JzZ7qOzpzFkao6Osm3zC6X5G/mnPvNSf42yTVnut+c5IHdfdE8a2zgrCSzJ51uOf1nOM++Tkpyi5mur0zXAwAAAAB2yr4V/MAe5LfmLqmqGyZ5TZITZ7o/keTe3f2JbQrz8kH7cXPO+54cfKXcm7v7k5tNqqpbZXJV27Vnut+R5Nu7+8tzxl5Xd1+c5B8G3Y+dc/pw3N8f4ikjAAAAAADYExR7dkFVXTeTq9tOmek+P5MTPR/axlB/lsmpnAO+t6qG79YM93aNJE8edJ+xWaCqulkmOV1/pvsDmeT0ufm2O5fhXv5zVR2/yd5OSPLjm6wDAAAAAACjpNizZFV1bJK/T3Lrme4vJrlfd797O2N19zuSvHim64gkz6uqa62zt0ryrBx83dkHMykarauqbpTJdXQ3mun+aCanlD69+M7X191/m+T/znRdL8kZVXX4Ons7IpPCzvVmuv+5u1+1nfsCAAAAAIDdsn+3N3A19PIkdxr0/U6S61fVfRZc6y3d/YVNxvxikv+YK9/QuVOSs6vqCd191oFBVXXLJL+R5CGD+U/u7kvXW7yqrpnJ1Wonz3RfnuTpmbypc8t5EpnxL9Pr2jbypCSvy5XFyv+Y5B+q6r9097/O7O0OSX47yT0Ge/u5BfcEAAAAAGxJJVW7vYlttEq5sEoUe5bv9DX6fmWLa90zyVkbDejuD1TV45I8P1f+m+h2SV5bVedncgLnhEzeDhr+m+r3u/slm+zhhBx8SilJDkvy7E13v7abJfnwRgO6+1+q6heS/OZM9+lJ3lJVn0zyqUxOGd1wjek/193/d41+AAAAAAAYJcWeq4HufuH0irYzkhw189Xx089anpk9fAKmu59RVZdnUvA5bOarG+Xg6+QOuDzJz3b3s5awPQAAAAAAWBpv9lxNdPcLktwmkxM+617LluTsJKd395O6u5eyuS3q7t9Ocsckf5vkinWGXZHkb5LcQaEHAAAAAIBV5GTPknX3rl3q2N0fTPLIqvqxJN+a5BZJjk1ycSbXub2+uz+x4Jofzi5eVNndb0vywKq6fiY5nZzk6CQXJjk3k5w+u1v7AwAAAACAnabYczXU3V9O8srd3sd2mhZ0/nq39wEAAAAAzKjs4n8qvgNWKRdWimvcAAAAAAAARkyxBwAAAAAAYMQUewAAAAAAAEbMmz0AAAAAAOyMSlIr9NDNCqXCanGyBwAAAAAAYMQUewAAAAAAAEZMsQcAAAAAAGDEFHsAAAAAAABGbP9ubwAAAAAAgBW1L6t15GCVcmGl+K0JAAAAAAAwYoo9AAAAAAAAI6bYAwAAAAAAMGLe7AEAAAAAYGdUTT6rYpVyYaU42QMAAAAAADBiij0AAAAAAAAjptgDAAAAAAAwYoo9AAAAAAAAI7Z/tzcAAAAAAMAKq93eAKw+J3sAAAAAAABGTLEHAAAAAABgxBR7AAAAAAAARsybPQAAAAAA7Ix9NfmsilXKhZXiZA8AAAAAAMCIKfYAAAAAAACMmGIPAAAAAADAiHmzBwAAAACAnVHTz6pYpVxYKU72AAAAAAAAjJhiDwAAAAAAwIgp9gAAAAAAAIyYYg8AAAAAAMCI7d/tDQAAAAAAsKKqJp9VsUq5sFKc7AEAAAAAABgxxR4AAAAAAIARU+wBAAAAAAAYMW/2AAAAAACwM2r6WRWrlAsrxckeAAAAAACAEVPsAQAAAAAAGDHFHgAAAAAAgBFT7AEAAAAAABix/bu9AQAAAAAAVlRVsq92exfbp1YoF1aKkz0AAAAAAAAjptgDAAAAAAAwYoo9AAAAAAAAI+bNHgAAAAAAdkZNP6tilXJhpTjZAwAAAAAAMGKKPQAAAAAAACOm2AMAAAAAADBiij0AAAAAAAAjtn+3NwAAAAAAwIqqmnxWxSrlwkpxsgcAAAAAAGDEFHsAAAAAAABGTLEHAAAAAABgxLzZAwAAAADAztiXZN8KvXPj+AR7lN+aAAAAAAAAI6bYAwAAAAAAMGKKPQAAAAAAACPmzR4AAAAAAHbOCj3ZA3uVkz0AAAAAAAAjptgDAAAAAAAwYoo9AAAAAAAAI6bYAwAAAAAAMGL7d3sDAAAAAACsqKrJZ1WsUi6sFCd7AAAAAAAARkyxBwAAAAAAYMQUewAAAAAAAEbMmz0AAAAAAOyMmn5WxSrlwkpxsgcAAAAAAGDEFHsAAAAAAABGTLEHAAAAAABgxBR7AAAAAAAARmz/bm8AAAAAAIAVta8mn1WxSrmwUpzsAQAAAAAAGDHFHgAAAAAAgBFT7AEAAAAAABgxb/YAAAAAALBzPHMDO87JHgAAAAAAgBFT7AEAAAAAABgxxR4AAAAAAIAR82YPAAAAAAA7o2ryWRWrlAsrxckeAAAAAACAEVPsAQAAAAAAGDHFHgAAAAAAgBFT7AEAAAAAABix/bu9AQAAAAAAVtS+rNaRg1XKhZWi2AMAAAAAACuiqo5M8k1JTktyXJKjknw5yXlJ/jXJB7q7tyHO/iR3SXKbJNdLcnmSTyV5S3e/81DXH8S6cZJvSXLTXJnP+5L8S3dfsI1xlpbTdlPsAQAAAACAkauqOyR5YpKHJTlyg6GfqKozkvxud39+C3GOSfLkJD+W5LrrjHlvkt9McuahFJaq6h5Jnprk9HWGXFJVL0ryy9394UOIs7ScdopDZwAAAAAAMFJVta+qnp7kjUkemY0LPUly4yS/nORdVfXtC8a6bZJ/T/Jfs05RZOrUJH+W5O+q6tqLxJjGqap6RpKzsn6hJ0mOSPKDSd5RVQ9dNM401lJy2mmKPQAAAAAA7Iyq1fvsPX+c5Odz1b/v/2qSt2dSBDo3yfA0ytcleVlVfcc8Qarq1CT/lORmg68uyKRY8v4klw6+u38mxZFrzBNjxu8ledKgr5N8LJOr6D47+O7oJC+qqu9eJMiSc9pRij0AAAAAADBCVfWwJI8fdL8ryQOSXLu7v7G779LdN8+kuPOUJJfMjD0iyfOq6rhN4uxP8pIk15/p/nySRye5bnffrrtvmeQGSX4tyRUz474lyTMWyOkRSX5i0P3SJKd29026+w7dfXyS+2RSkDngsGkuJ80ZZ2k5LYNiDwAAAAAAjNNTBu03J7lzd7+yuy+b/aK7z+/uX0nyHUlmvzs+yX/aJM5jk9x2pv2FJN/W3X/e3f/v5Et3f767fzGTq9Vm/VhV3WKzZKrqiEzexZn1R0ke3t3vH+TzmiR3zyTnA45N8rTN4kwtJadlUewBAAAAAICRqaqTk9xm0P3j3X3hRvO6+5+SnDHo/o8bxDkiyS8Oun+2u9+1QYznJ/mLma79SZ660b6mHpfkpJn2+5M8sbuHV9AdiPOlTE7izJ5WemRVfcNGQZac01Io9gAAAAAAwPicOmh/vLvfNOfclw7aN99g7P2TfP1M+8NJnjtHjKfm4HeCHl5V195kzvBKut/o7os3mjAt0LxopuuwJI/ZJM4yc1oKxR4AAAAAAHZGreBn77juoP2xBeZ+dNC+zgZjHzRoP3e9kzazuvvcJK+b6To8yXeuN76qTkzyzTNdFyR58WZxpoYnlYZ7HlpKTsuk2AMAAAAAAOPzpUH7qAXmDsd+doOxDxi0/2GBOK8etB+4QJzXb3Yl3ezYJF+daZ+6yXs6y8ppafbv9gYAALh6u2LT/3aKMej4hRy7iy/3azh2H7/w87u9BQ7Rg/7k+3Z7C2yDNz7h73d7C2zRe/a9M8/Z7U3AYt42aJ9WVUfPWSC586D9xrUGVdXXJbnBTNfXkvzr3DucFGFm3X6DscPvzpk3SHdfVlVvTHL6YL33D8cuOaelcbIHAAAAAABGprs/noMLIkcm+anN5lXVkUmeMOgeXoN2wGmD9ge6+5J595jkXYP2zatqvUMow1jDuYvGGq63Xv9O5rQ0ij0AAAAAAOyQSmqFPnvs0Z4kP5/kipn2r1TVo9cbXFXXSfKXObjg8YrufsU6U04dtBd5FyjdfX6Si2e6jkhys52Itcb44XrbEmfBnJZm16tNAAAAAAAwUqdULVwAOr+7z9uO4N39L1X1E0n+IJNK1P4kZ1bVf07yV0nem+SiJNdPcpck35/kujNLvDrJRveInjBof3wL2/xkkpMHa17lerUkxx9irE8M2sO9r9e/kzktjWIPAAAAAABszcu2MOdpSZ66XRvo7j+sqvcm+b0kt55232n6Wc8HkzwjyXO6+4oNxh0zaM/zHtDQcM5wzVTVUUkOO8RYm8ZZp39Hclo217gBAAAAAMCIdfc/ZVLceWaSyzcZ/tHpuOdvUuhJrlrEuHjNURu7aJM11+tbNNY8cdbq36mclkqxBwAAAAAARqyq/lOSc5P8bK56QmboJkmeneTDVfXYTcZeY9C+ZAvb+9qgfdQccbYSa544a8XaqZyWyjVuAAAAAADsjH1ZrSMHV83lQZkUWRZx/rbsJUlVHZ7kBUkeOtP9+ST/I8nfZPKOzIWZvNlzxySPSfLgTN73uW6SM6rqtO5+0johhqdejtjCNo/cZM31+o5Yp/9Q4qzVv1M5LZViDwAAAAAAbM253f3OXYz/hzm40PPGJA/q7k8Pxn0qySuSvKKqvivJi3LlCZefrap3dfdz11j/gkF7rRM4mxmeehmuuV7fNbJYEWWeOGv171ROS7VKNVUAAAAAALhaqKrTkzxupuu8JA9co9BzkO5+eZL/POj+rapa6yqyYRHj6AW3udacqxRGuvuiXPWtoUVjbRpnnf4dyWnZFHsAAAAAAGB8fmrQflZ3z3tF3JlJ3jfTvl6Sh6wx7rxB+8Q51591o03WPGC490Vj3XjOOMvMaWkUewAAAAAA2BmVpGqFPrv9D3SiqirJvQbdr5h3fndfkeRvB913X2Poewftm8wbI0mq6oQcfE3aJUk+uM7wQ4q1xvj37EScBXNaGsUeAAAAAAAYl+OSXHvQ96EF1xiOH56MSa5aMDmlqo5YIMZpg/a53X3ZOmOHsW61QJy1Yq1X7FlmTkuj2AMAAAAAAONy5Bp9ixYcLh20DxsOmL7/M/sG0JFJ7rBAjLsN2m/bYOzwu7vOG6Sq9ie58zyxlpzT0ij2AAAAAADAuHxujb7hOzKbGZ7kWe+9n+F1b/ddIMZw7EZXzQ3j3LWqjp4zzt2SXHOm/b7uft96g9eItVM5LY1iDwAAAAAAO6NW8LMHdPclST416B6+4bOZew/a564z7uWD9mOmbwZtqKpOSXKPma5Lk7xyvfHd/bEkb53pOibJIzaLM/W4Qftlm4xfSk7LpNgDAAAAAADj85pB+wnT68w2VVX3SPItm6x3wKuSfHymfVKSx8wR5qk5uDz20u7+0iZzzhi0n1xV19hoQlWdluR7ZrquSHLmJnGWmdNSKPYAAAAAAMD4/MWgfZskz66qDf/ev6punuT5g+73J/k/a43v7q8l+bVB9zOr6lYbxPj+JD8w03V5kqdstK+p5yT56Ez7lkn++3qnbqrqWkn+PMkRM93P7+53bRRkyTkthWIPAAAAAACMTHe/KslrB90/nOR1VXXv4SmfqrpeVf2XJG/OVd/3+f+6+/INwp2R5J0z7eOS/HNVPWo2TlVdt6p+Ncn/HMz/403e0DmQ0yVJnjzo/k9JXlxVtxjkc68k/5zkjjPdFyT55c3iTC0lp2WZ60gXAAAAAACw53x/knOS3Gym71uT/GOSC6rqQ0kuSnK9JCdn7VeHfru7/3KjIN19aVU9PMm/JLnutPu6SZ6X5A+q6twkR033cfhg+huT/Oy8CXX3C6rq25L82Ez3w5I8tKo+luT8JDdNcv3B1CuSPKa7PzRnnKXltAyKPQAAAAAA7IyqZN+m796Px9q3ie2a7v709P2dP09y+uDrY5LcdoPplyb5pSTPmDPWu6enaV6WSbFlNs7t1pn2j0ke3t0XzRNjxk8kuTjJE2f6KslNpp+hr2ZS6NmwaDW05Jx2lGvcAAAAAABgpLr7Y0nuneQRSc7K5ITLRr6U5A+T3La7f7O7e4FY/5ZJAek3knxhg6Hvz+RKuft19xfnXX8mzhXd/TNJDlzVtp5LkvyvJLfp7hcvGmcaayk57TQnewAAAAAAYMS6+4okL0nykqo6NpN3bE5Ocp0k10jy5SSfS/LvSd41Hb/VWF9J8v9V1VOS3CXJbTK5Ju7yJJ9K8q/d/fatZ3NQrNcmeW1VnZjkrpmc6rlGkq9kUnz5l+7+8jbEWVpOO0WxBwAAAAAAVsS0cPHa6Wcn41yayXs3/7KTcaaxPp5kSyd3FoyztJy2m2IPAAAAAAA7o2rPvXNzSFYpF1aKN3sAAAAAAABGTLEHAAAAAABgxBR7AAAAAAAARkyxBwAAAAAAYMT27/YGAAAAAABYUTX9rIpVyoWV4mQPAAAAAADAiCn2AAAAAAAAjJhiDwAAAAAAwIh5swcAAAAAgJ1RlarVeeimVygXVouTPQAAAAAAACOm2AMAAAAAADBiij0AAAAAAAAjptgDAAAAAAAwYvt3ewMAAAAAAKymqkpV7fY2tk9Verf3AGtwsgcAAAAAAGDEFHsAAAAAAABGTLEHAAAAAABgxLzZAwAAAADAjqiafFZGxZs97ElO9gAAAAAAAIyYYg8AAAAAAMCIKfYAAAAAAACMmDd7AAAAAADYEftSqRV6tKdTuWK3NwFrcLIHAAAAAABgxBR7AAAAAAAARkyxBwAAAAAAYMQUewAAAAAAAEZs/25vAAAAAACAFVVJVe32LrZNr04qrBgnewAAAAAAAEZMsQcAAAAAAGDEFHsAAAAAAABGzJs9AAAAAADsiKpaqTd7VikXVouTPQAAAAAAACOm2AMAAAAAADBiij0AAAAAAAAjptgDAAAAAAAwYvt3ewMAAAAAAKymqkpV7fY2ts0q5cJqcbIHAAAAAABgxBR7AAAAAAAARkyxBwAAAAAAYMS82QMAAAAAwI6omnxWxSrlwmpxsgcAAAAAAGDEFHsAAAAAAABGTLEHAAAAAABgxBR7AAAAAAAARmz/bm8AAAAAAIDVVFWpqt3exrZZpVxYLU72AAAAAAAAjJhiDwAAAAAAwIgp9gAAAAAAAIyYN3sAAAAAANgRlRV7syerkwurxckeAAAAAACAEVPsAQAAAAAAGDHFHgAAAAAAgBHzZg8AAAAAADukVuydm1XKhVXiZA8AAAAAAMCIKfYAAAAAAACMmGIPAAAAAADAiCn2AAAAAAAAjNj+3d4AAAAAAACrqSqpqt3exrZZoVRYMU72AAAAAAAAjJhiDwAAAAAAwIgp9gAAAAAAAIyYN3sAAAAAANgRkzd7dnsX22eVcmG1ONkDAAAAAAAwYoo9AAAAAAAAI6bYAwAAAAAAMGKKPQAAAAAAACO2f7c3AAAAAADAatpXlX1Vu72NbdMrlAurxckeAAAAAACAEVPsAQAAAAAAGDHFHgAAAAAAgBHzZg8AAAAAADuiqlIr9M7NKuXCanGyBwAAAAAAYMQUewAAAAAAAEbMNW67rCbn/k5KctskJya5TpKvJflCkvcneVN3X7zNMY9Ncrckt0xyrSQXJflIknO6+5PbGWtZVjEnAAAAAACYh2LPLqiq45I8OMm3J7lXkutvMPzSqvrbJM/q7tcdYtybJfmVJI9IcsQaQ7qqXpfkKd199iHEOTnJnZPcZfrjNyU5ambI67r79K2uP4i1lJwAAAAAAGCvUuxZsqr6gySPz9qFibUcnklh6MFV9edJfrK7v7yFuI9I8twk19xoWJLTk5xVVc9I8gvd3XOu/11JfjST4s5Gxatts9M5AQAAAACHqCqTy41WxCrlwkpR7Fm+u2TtQs/lST6V5DOZFHhumuTagzGPSvINVXXv7r5g3oBV9fAkL8hV32g6P8nHkpyQ5MaZFEYy/fHnkxyZ5IlzhrlXku+cd0+Hakk5AQAAAADAnjf8i3KW64tJnp3kAUmO6+6v7+47dvftklwvyT2T/PNgzp2TnDlvgKo6JZPTL7O/1v+W5F7dfUJ336G7vz7JaUn+ajD9CVX1kAXyWc+F27DG/7NHcgIAAAAAgD1BsWd3fDiTq9xu1N3/ubtf2d1fmR3Q3Zd391mZFHz+ZDD/oVV1zzlj/WqSo2fab0py9+5+7SDee5M8bI1Yz6iqRU6AfT7Jq5L8tyQPSnLDJD+xwPx5LDsnAAAAAADYsxR7lu8pSU7t7jO6+6LNBnf35Ul+PMmbB189frO5VXXrJN8z03VJkkev9+bP9C2bn07y/pnuU5I8ZrNYSZ6T5Obdfb3u/vbu/qXufnl3f3qOuXNbck4AAAAAwCGoTJ65WZnPbv8DhXUo9ixZd/9td1+y4JzLkzxj0H3/OaY+Ngf/Gr+wu9+9SayLkzx90L1pYam739nd586xp0O1tJwAAAAAAGAMFHvGY/h2z/Wq6pqbzPmuQfuMOWO9KAe/s3OnqrrRnHN32irmBAAAAAAAW6bYMx5fWKPv2usNrqpTk9x8puvCJOfME6i7h2MryQPmmbuTVjEnAAAAAAA4VB6pH48br9H3uQ3G337QfmN3X7ZAvNcnue8G6+2G2w/aq5ATAAAAAKysqkrV6rx0s0q5sFqc7BmPbxu0P7LJ2z+nDdrvWjDecPxwvd2wijkBAAAAAMAhUewZj8cO2q/cZPypg/bHFow3HD9cbzesYk4AAAAAAHBIXOM2AlX1nUnuPug+c5NpJwzaH18w7CcG7eMXnL8T9mROVXXCFtY6ZTtiAwAAAACAYs8eV1XXTfLHg+6/7u43bjL1mEH7wgVDD8cfXlVHdvfXFlxnO+3VnH48yVMOcQ0AAAAAANgSxZ49rKr2JfmLJCfOdH8pyU/NMX1YGLl4wfAXrbPmXir2rEJOAAAAALCyqipVtdvb2DarlAurRbFnb/utJN8x6PvR7p7nrZprDNqXLBh7rQLIUQuusd1WMScAIL3bG2AbXO6XcfRe9+n37vYWOEQPffz9d3sLHKJz/+o9u70FtsG1Dj9st7fAFh2z368dMF6KPXtUVf1Ukp8ZdD+ju1805xLDUy9HLLiFI+dYc9n2ak7PTvKSBeeckuRl2xAbAAAAAICrOcWePaiqvj/JswbdZyZ58gLLXDBoD0/FbGatEy/DNZdtT+bU3eclOW+ROY57AgAAAACwXRR79piqemCS5yWZrQb8VZLHd/cil2MMixhHL7iV4fjLunu3T/asYk4AAAAAsLKqVus/fF6hVA5ZVR2b5G5JbpnkWpm8mf6RJOd09ye3Odatk9whyQ2THJbkc0nekeQN3X3ZNsZZWk7bTbFnD6mqe2ZyHdjsr8urk3xfd1++4HLDkyYnLjj/xoP2+QvO3wmrmBMAAAAAwMKq6swkj96m5T7S3SfNGfdmSX4lySOy9lMbXVWvS/KU7j57qxuqSZXwMUl+PpPiy1o+V1V/mOTp3X3hIcRaSk47ad9ub4CJqrpLkpfn4KvJzkny3d19yRaWHL6uepMF5w/H74VXIlcxJwAAAACA3TbX30FX1SMyOVHzA1n/TfVKcnqSs6rq6bWFo11VdZ0kr0pyRtYv9CTJ9ZL8YpJ/n57+Wdiyctppij17QFV9Y5K/S3LMTPdbk3znIVQjh4WMWy04/7RN1tsNq5gTAAAAAMBu+5vNBlTVw5O8IMk1B1+dn+Rfk3w8yexTJJXJqZzfWWQjVXVUJoWe+w6+uiTJ+5K8Pcnw781PTvLaqrr5grGWktMyuMZtl1XVqZlc1XbcTPe7k9y/u790CEu/bdC+U1XtX+D+wrttst5ueNugvQo5AQAAAABsxTOS/MUW5t0iybMHfWduNKGqTkny3Bx8gOTfkjyxu187M+7UJL+e5CEz455QVf/c3X815/5+J8mdZ9pXJPm1JP+9u78wjXNEku+fjj3wd+vHJ3lxVd1pnmdRlpzTjlPs2UVVddMk/5jkhJnuDyW5b3cf0nsy3f2eqjo3ySnTrqOT3DXJpvcJVtXRSb5ldrnMUdndaauYEwAAAACssqrKHrzxasv2Ui7d/a4k71p0XlXda9D11u7+902m/Womfx97wJuS3Ke7vzzY03ur6mFJ/ijJj8x89Yyqevlm/+F+VX1Dkh8edP9Ad79gEOeSJGdW1ZuS/EuS60y/+qYkj8qkiLOZpeS0LK5x2yVVdcMkr0ly4kz3J5Lcu7s/sU1hXj5oP27Oed+Tg6+Ue3N3f3J7tnTIVjEnAAAAAIAdV1X7kvzgoPvMTebcOpO/Xz3gkiSPHhZFDujuTvLTSd4/031KksfMscWnJTlspv0/h4WeQax3JvnZQfdTqurwjYIsOaelUOzZBVV13Uyubjtlpvv8TE70fGgbQ/1ZDr5P8HuravhuzXBv10jy5EH3Gdu4p0O1ijkBAAAAACzDfXLwAYRLkzx/kzmPzcG1hBd297s3mtDdFyd5+qD78RvNqarjcvBVaZ3kqZvsLZmc4vnITPummeS5kaXktEyKPUtWVccm+fskt57p/mKS+232m2lR3f2OJC+e6ToiyfOq6lrr7K2SPCuTOxsP+GAmBZY9YRVzAgAAAABYkkcP2n/T3Z/dZM53Ddrz/of0L0py4Uz7TlV1ow3GPyAHPz1zVnd/cLMg3X1Frnpt24M3mbasnJbGmz3L9/Ikdxr0/U6S61fVZtXGobcceJBqA7+Y5D8muea0fackZ1fVE7r7rAODquqWSX4jB1dOk+TJ3X3pPJvZYP+3GrSP22DsB+f4A7y0nAAAAACAQ1DJHnrm5tCNOJfpfzD/3YPuMzeZc2qSm890XZjknHnidfeFVXVOkvseWC6Tgs5z1pnygEH7H+aJM/XqHHwK6IHrDVxyTkuj2LN8p6/R9ytbXOueSc7aaEB3f6CqHpfJUbwD/yq6XZLXVtX5ST6a5IRMju4N/1X1+939kgX28+o5x33jBmOflk2O5i05JwAAAACAVfCIJEfNtM9L8spN5tx+0H5jd1+2QMzX58rCyFrrbRRrrgLM1FuSfC3JkdP2jarq+O4+f444O5nT0ij2XA109wun15mdkYP/MB8//azlmUl+bqf3tlWrmBMAAAAAwA76oUH7f81R5Bi+l/6uBWMOx6/5/npVHZ6DT9ssFKu7v1ZV5+bgW6ZOS7JWsWcpOS2bN3uuJrr7BUluk8lpmI2uMDs7yend/aTu7qVsbotWMScAAAAAgO1WVTdPcrdB9/Cdm7WcOmh/bMHQw/HD9Q44OQcfTrlojreEthprWTktlZM9S9bdu3ar4/QtnEdW1Y8l+dYkt0hybJKLM7n67PXd/YlDWH/pue10TgAAAADA1lUqtUKP9tRVH+05ZQv5nd/d523Pjub26EH7X7v77XPMO2HQ/viCcYd/N7verUzDOFv5O93hnOGa6/XvVE5LpdhzNdTdX87mdzGOyirmBAAAAADseS/bwpxN3y3fTtPnMB416D5zzunHDNoXLhh+OP7wqjqyu7+2zXHWmjNcc7tizZvTUrnGDQAAAAAAVte9ktxkpn1JJk9jzGNYGLl4wdgXzbHmdsRZK9a8xZ6dymmpFHsAAAAAAGB1Da9w+5vu/tycc68xaF+yYOy1TrsctQNx1oq1VpztiDVvTkvlGjcAAAAAANiaByU5d8E55+/ERtZSVcckecig+8wFlhieejliwS0cOcea2xFnrVjrndhZVk5LpdgDAAAAAMCOqKpMnoxZDWvkcm53v3M39jKnhyc5eqb9mSR/t8D8Cwbt4amYzax14mW45nbEWSvWWnG2I9a8OS2Va9wAAAAAAGA1/dCg/b+6+7IF5g+LGEevOWp9w/GXdfdap2AONc5ac+Yt9uxUTkul2AMAAAAAACumqk5O8m2D7jMXXOa8QfvEBeffeNBe7wq7YZzhvK3EGq65Xv9O5bRUij0AAAAAALB6HpVk9t65f+3uty+4xnsH7ZssOH84/j3rjPtgktkTR0dV1fE7FGtZOS2VN3sAAAAAANgR+6qyb4Xe7BlLLjV5XOhRg+7nbmGpYSHjVgvOP22T9ZIk3X1pVZ2b5NRBrNfNE6Sqjkxy8jyx1ujfkZyWzckeAAAAAABYLfdIcrOZ9iVJnr+Fdd42aN+pqhY5RHK3Tdbb6Lu7LhDnDkmOnGl/qrvXu8ZtGGcnc1oaxR4AAAAAAFgtjx60X9Hdn190ke5+T5JzZ7qOzpxFmKo6Osm3zC6X5G82mDL87r7zxFln7CvWG7jknJZGsQcAAAAAAFbEtCDxsEH3mYew5MsH7cfNOe97khwz035zd39yg/GvzMHv9pxeVcOr2a5iemXdDw26X7bJtGXltDSKPQAAAAAAsDoemoMLEp9O8veHsN6fZXKC5YDvrarhuzUHqaprJHnyoPuMjeZMTx799ewySZ46x/4em+SkmfZHkvzjJnOWktMyKfYAAAAAALAjqlbvMwI/NGj/r+6+bK2B8+judyR58UzXEUmeV1XXWmv89KTNs5LcYqb7g5kUWDbzlCRXzLR/sKq+b73BVXWrJM8cdP9qd1+yUZAl57QUij0AAAAAALACquqmSU4fdJ+5DUv/YpKvzrTvlOTsqjooVlXdMslfJvnRwfwnd/elmwXp7ncl+dNB919U1a9U1XEzcQ6vqh9K8i9JrjMz9t+TPG+zOFNLyWlZ9u/2BgAAAAAAgG3xqEyuPzvgLdNTLIekuz9QVY9L8vyZ9W+X5LVVdX6SjyY5IcmJg/hJ8vvd/ZIFwj0xyTcnueO0vS/JLyX5+ar6UJKvJTk5B19VlySfTfLweU8xLTmnHafYAwAAAAAAq+HRg/aZ27Vwd79wep3ZGUmOmvnq+OlnLc9M8nMLxvlqVd0/yUuS3GvmqyOSnLrOtA8n+a7uft+CsZaS0zK4xg0AAAAAgJ1RlVqhz15+tKeqvjXJKTNdl2RyamXbdPcLktxmuu5GV5idneT07n5Sd/cW4nw+yX2T/EiSD2ww9PNJfj3Jbbv77YvGmcZaSk47zckeAAAAAAAYue7+l1z1urGdiPPBJI+sqh9L8q1JbpHk2CQXZ3L12eu7+xPbEOeKJM9J8pyqum0mV7vdMMlhST6X5B1J3rAd7+YsK6edpNgDAAAAAAAspLu/nOSVS4r19iRbOrmzYJyl5bTdXOMGAAAAAAAwYoo9AAAAAAAAI+YaNwAAAAAAdkQlqZ1/RmZpVicTVo2TPQAAAAAAACOm2AMAAAAAADBiij0AAAAAAAAj5s0eAAAAAAB2RFWlanVeulmlXFgtTvYAAAAAAACMmGIPAAAAAADAiCn2AAAAAAAAjJg3ewAAAAAA2BGVFXuzJ6uTC6vFyR4AAAAAAIARU+wBAAAAAAAYMcUeAAAAAACAEVPsAQAAAAAAGLH9u70BAAAAAABWU9XksypWKRdWi5M9AAAAAAAAI6bYAwAAAAAAMGKKPQAAAAAAACPmzR4AAAAAAHZEVaVW6KGbVcqF1eJkDwAAAAAAwIgp9gAAAAAAAIyYYg8AAAAAAMCIKfYAAAAAAACM2P7d3gAAAAAAACuq9qVqhc4crFIurBS/MwEAAAAAAEZMsQcAAAAAAGDEFHsAAAAAAABGzJs9AAAAAADsiKqkqnZ7G9tmhVJhxTjZAwAAAAAAMGKKPQAAAAAAACOm2AMAAAAAADBiij0AAAAAAAAjtn+3NwAAAAAAwGqq2pfatzpnDqpWJxdWi9+ZAAAAAAAAI6bYAwAAAAAAMGKKPQAAAAAAACPmzR4AAAAAAHZE1b6VeudmlXJhtfidCQAAAAAAMGKKPQAAAAAAACOm2AMAAAAAADBi3uwBAAAAAGBHVFWqare3sW1WKRdWi5M9AAAAAAAAI6bYAwAAAAAAMGKKPQAAAAAAACOm2AMAAAAAADBi+3d7AwAAAAAArKaqStXqnDmoqt3eAqxpdf6UAQAAAAAAXA0p9gAAAAAAAIyYYg8AAAAAAMCIebMHAAAAAICdUbVa79ysUi6sFCd7AAAAAAAARkyxBwAAAAAAYMQUewAAAAAAAEZMsQcAAAAAAGDE9u/2BgAAAAAAWE1V+1K1OmcOVikXVovfmQAAAAAAACOm2AMAAAAAADBiij0AAAAAAAAj5s0eAAAAAAB2RKVSVbu9jW1TWZ1cWC1O9gAAAAAAAIyYYg8AAAAAAMCIKfYAAAAAAACMmGIPAAAAAADAiO3f7Q0AAAAAALCaqipVq3PmoKp2ewuwptX5UwYAAAAAAHA1pNgDAAAAAAAwYoo9AAAAAAAAI+bNHgAAAAAAdsTkzZ7VeedmlXJhtTjZAwAAAAAAMGKKPQAAAAAAACOm2AMAAAAAADBi3uwBAAAAAGBHVO1L1eqcOVilXFgtfmcCAAAAAACMmGIPAAAAAADAiCn2AAAAAAAAjJg3ewC4Wuvd3gCHrP0ijt5lfg1XwnkXf3W3t8Ah+tRXv7TbW+AQffrlH9rtLXCIrnXEYbu9BQBgpBR7AAAAAADYGftq8lkVq5QLK8U1bgAAAAAAACOm2AMAAAAAADBiij0AAAAAAAAj5s0eAAAAAAB2SKVqlc4ceLOHvWmV/pQBAAAAAABc7Sj2AAAAAAAAjJhiDwAAAAAAwIgp9gAAAAAAAIzY/t3eAAAAAAAAq6mqUlW7vY1ts0q5sFqc7AEAAAAAABgxxR4AAAAAAIARU+wBAAAAAAAYMW/2AAAAAACwI6r2pWp1zhysUi6sFr8zAQAAAAAARszJHgAAAAAAWEFVdWqS2yU5Mck1k1yU5DNJ3pfk37r7a4ew9jWS3DXJNyQ5LsklST6e5A3d/cFD3Pow1ilJ7pxJHkck+UKS9yQ5p7sv3sY4S8tpuyn2AAAAAADAiqiqY5P8ZJLHJ7nZBkMvqao3JvnL7v7dBdY/PslTkvxQkqPXGfOWJL/a3S+bd9111nlwkl9K8s3rDLmgqs5M8rTu/uwhxFlaTjtFsQcAAAAAgB1RVSv1zk1V7fYWNlRVD0zyp0m+bo7hRyT51iS3SDJXsaeqTk/ykiTX32ToHZL8dVX9eZIf7u5L5ll/Js6RSc5I8shNhh6T5CeSfE9VPay7z14kzjTW6VlCTjttdf6UAQAAAADA1VRVPTHJy3PVQs/FST6Y5I1J3p5kSydgqupbk7wyVy2KfDHJW5N8OMnlg+8eleQFtUCVrCbVwRflqoWey5N8KMnbknxp8N3xSf6uqr5l3jjTWEvJaRkUewAAAAAAYMSq6nFJfifJbAHi75J8R5LrdPcp3X2X7v7G7j4+yY2T/GCSl2byLs1m6x+XSQHmqJnujyR5cJLrdvc3d/fNkpyU5I8H0x+S5IkLpPOkJA8a9P1Rkpt098nd/U1Jrjtd96MzY66Z5MVVde15giw5px2n2AMAAAAAACNVVTdP8j9mui5N8v3d/Z3d/ffd/bXhnO7+ZHf/RXc/LMnt5gjzpCQ3mml/KMldu/tl3d0z6368u/9Tkv86mP/L0+LKZrlcb425v9DdP9bdn5yJc0V3/+8kd83k9M0BJyb5mTnySZaU07Io9gAAAAAAwHj9SZJrzLQf2d0vmHdyd39ho++r6vgkPzno/uHZ4ssafiPJ7Ps5107ys3Ns5+eSHDvTPjvJb643uLs/keTxg+4nTotG61pyTkuh2AMAAAAAwI6oqpX77CVV9aAk95zpekl3v2Sbw3xvkmNm2md392s2mjA9GfO0QfdjN3rnZvpWz2MG3U+dPWWzTqzXJPnnma5jkzxiozlZUk7LpNgDAAAAAADj9COD9rAYsR2G7+ecMee812ZyNdoBN0jyHzYYf9ckx8+0P5jkrDljDff04E3GLyunpVHsAQAAAACAkamqGye5/0zX27r7ndsc45gkdx90/8M8c6cnYf5x0P3ADaY8YNB+9WanembHDtqnV9XRaw1cck5Lo9gDAAAAAADj8+1JDptpv3YHYtw6yeEz7Q9196cXmP/6Qfv2G4wdfnfOvEGmb+18eKbriCS3Wmf4MnNamv27vQEAAAAAAFZT1b5MnmJZDXsslzsN2v924CdV9U2ZvH9zjyQ3SXJkkvOSvD/Jq5K8oLs/MUeM0wbtdy24x+H44XrbHeukwXpv2qE4G623K/bU70wAAAAAAGAuw2LPB6vqmKo6I8m/JvnJJN+Y5DpJjkpy0yT3SfJbSd5fVb9eVYdnY6cO2h9bcI/D8TetqmsMB1XVUZkUpbYz1nDv6/XvSE7LptgDAAAAAADjc/NB+4okZyd57Bxzj0ryC0leWVXHbjDuhEH74/NvL0nymSSXzbT3JbneGuOun6Rm2pdmchJpEcOTSsO9r9e/UzktlWvcAAAAAABga06pqs1HHez87l60kHGQmtwnNyzS/F6Sb5r+vJP8TZJXZlLMOHr63Q8mudHMnPskOTPJQ9cJdcygfeEi++zurqqLBnsdrrlW31e7uxeJtcbe1oqzVv9O5bRUij0AAAAAALA1L9vCnKcleeohxr12Dj4JkyTfPP3xc0m+u7v/efD9i6rqvyX54yTfP9P/kKp6VHf/+RpxhkWMi7ew160Ue7YaZ6M1tzuWYg8AAAAAAFcHlS2cfNnD9kwu6xUXLk/ygO5+w1pfdvcFVfWDmVybdr+Zr/6/qvqfa5ymGb5Fc8kW9vq1QfuoNcYsK86yYy2NN3sAAAAAAGBc1juN8qfrFXoO6O4rkvxYJm/8HHBqknvMEeeIuXd4pSM3WXOZcZYda2mc7AEAAAAAgK15UJJzF5xz/jbEvWCd/ufMM7m7P1hV/5iDT/fcI8lZm8QZnoqZx/DUy1p7X1acZcdaGsUeAAAAAADYmnO7+53LDtrdF1XV5UkOm+n+SpK3LrDM63JwseeOa4wZFjGOXmD91OQOv60Ue65ZVbXGtXIbGe5t3mLPTuW0VIo9AAAAAADsiKpK1eq8JrLH3h86L8kNZ9ofmF7RNq/3DtonrBNj1okLrJ8kX5eD6xBXJPnsGuM+m6Rz5aNIh0/385kFYt140B7ufb3+ncppqVbnTxkAAAAAAFx9vHvQ/vKC84fjj1tjzLAgdJMFYwzHf6S7r/K+TXdflOSj2xzrPeuMW0pOy6bYAwAAAAAA4/OuQfvIBecP36r56hpjhgWTWy0Y47RN1tuNWMvMaWkUewAAAAAAYHz+ddD+ugXnD69t+9waY96Z5NKZ9klVdcM1xq3nboP22zYYO/zurvMGme7ppJmuS3PVYtgBy8xpaRR7AAAAAABgfP42k/diDrhZVV13gfl3GLSH15ulu7+S5OxB933nWbwmDxzdZ9D9ig2m/M2gfZ+a/5Gk+w3ar+3uC9YauOSclkaxBwAAAACAHVFVK/fZK7r7vCSvH3Q/ZJ65VbU/yXcPus9aZ/jLB+3HzRMjyT2T3Gym/Zkkb9hg/DlJPjvTPjnJ6XPGGu7pZZuMX1ZOS6PYAwAAAAAA4/THg/aTqmqet3t+OMkNZtpfTvKqdca+MMmFM+27V9W9Nlp8egLmKYPu53b3FWuNT5Lpd2cOup+y2emeqrp3km+b6fpKkhdvNCdLymmZFHsAAAAAAGCcXpDk7TPtWyb546pa9+/+q+ouSZ4x6H52d39prfHTE0T/Y9D9p1V1ow329QtJ7j7T/lKS39pg/AG/mWT2+rV7JPn59QZX1Y2T/Omg+3e7+7NrjT9gyTkthWIPAAAAAACM0PRUyROT9Ez3o5O8qqoOepOnqq5dVT+T5B+THDPz1fuS/PomoZ6R5NMz7ZslOaeqvmv25E1VnVhVf5Tk1wbzf627Pz9HPp9dYy+/UVXPni3EVNW+qnpwJle/nTQz9pNJfnuzOFNLyWlZ9u/2BgAAAAAAWE21b19q3+qcOdiLuXT3a6rqF5I8fab7PkneXFWfTvLxJEcnOSXJEYPpn0vysO7+yiYxPl9V35PJVW/XmHbfNJO3cb5YVR9Kcp0kN0ly2GD6y5I8c4GUfjPJXZM8cKbvx5L8SFV9JJMTNTebxpt1UZJHdPcX5wmy5Jx23N77nQkAAAAAAMytu38zyU8luXTw1Q2S3DHJablqoee9Sb6lu9+eOXT32UkekGR4muU6Sb4pkwLMsCjy/CTf092dOU1PKz08k3d1Zh2W5ORprOsMvvtcku/s7tfPG2caayk5LYNiDwAAAAAAjFx3/36Sb0zyoly16DPrQ0l+Osk3dvf7F4zxT0luleQPk3x1g6FvTfLQ7n5kd39tkRjTOBd39/cleViSt20w9MIkz05yq+4+a9E401hLyWmnucYNAAAAAABWQHe/J8n3VtW1MrkK7RZJrp3kgiSfSfKv3f3eQ4zxmSQ/XlX/ZRrjtExOwlyS5BNJ3tDdHziUGDOxXprkpVV18yR3SXLjTE4ofTHJu5O8vrsv3oY4S8tppyj2AAAAAACwI6oqM2/dj95YcunuLyf5++lnp2JclOQ108+OmhZadrzYssyctptr3AAAAAAAAEZMsQcAAAAAAGDEFHsAAAAAAABGTLEHAAAAAABgxPbv9gYAAAAAAFhNlUrV6pw5qNRubwHWtDp/ygAAAAAAAK6GFHsAAAAAAABGTLEHAAAAAABgxLzZAwAAAADAzqhK1Qq9c7NKubBSnOwBAAAAAAAYMcUeAAAAAACAEVPsAQAAAAAAGDHFHgAAAAAAgBHbv9sbAAAAAABgNVXtS9XqnDlYpVxYLX5nAgAAAAAAjJhiDwAAAAAAwIgp9gAAAAAAAIyYN3sAAAAAANgRVZWq2u1tbJtVyoXV4mQPAAAAAADAiCn2AAAAAAAAjJhiDwAAAAAAwIgp9gAAAAAAAIzY/t3eAAAAAAAAq6mqUrU6Zw6qare3AGtanT9lAAAAAAAAV0OKPQAAAAAAACOm2AMAAAAAADBi3uwBAAAAAGBn1L7JZ1WsUi6sFL8zAQAAAAAARkyxBwAAAAAAYMQUewAAAAAAAEbMmz0AAAAAAOyISqWqdnsb26ayOrmwWpzsAQAAAAAAGDHFHgAAAAAAgBFT7AEAAAAAABgxxR4AAAAAAIAR27/bGwAAAAAAYDVVVapW58xBVe32FmBNq/OnDAAAAAAA4GpIsQcAAAAAAGDEFHsAAAAAAABGzJs9AAAAAADsiKrKvhV658abPexVTvYAAAAAAACMmGIPAAAAAADAiCn2AAAAAAAAjJhiDwAAAAAAwIjt3+0NAAAAAACwmmr6v1WxSrmwWpzsAQAAAAAAGDHFHgAAAAAAgBFzjdsuq6pKclKS2yY5Mcl1knwtyReSvD/Jm7r74m2OeWySuyW5ZZJrJbkoyUeSnNPdn9zmWLdOcockN0xyWJLPJXlHkjd092XbGGdpOQEAAAAAwF6i2LMLquq4JA9O8u1J7pXk+hsMv7Sq/jbJs7r7dYcY92ZJfiXJI5IcscaQrqrXJXlKd599CHEqyWOS/HwmxZe1fK6q/jDJ07v7wkOItZScAAAAAIDFVe1L1epcMLVKubBa/M5csqr6gySfTvJnmRQoNir0JMnhmRSGzqqq51XVtbYY9xGZnKj5gaxdFEmSSnL6NNbTp0WbReNcJ8mrkpyR9Qs9SXK9JL+Y5N+np38WtqycAAAAAABgL1PsWb67ZO3CxOVJPp7kLUn+PcmX1hjzqCSvrqpjFglYVQ9P8oIk1xx8dX6Sf53G7dkpmZzK+Z0F4xyVSaHnvoOvLknyviRvTzI8xXNyktdW1c0XjLWUnAAAAAAAYK9T7NldX0zy7CQPSHJcd399d9+xu2+XycmXeyb558GcOyc5c94AVXVKkufm4F/rf0tyr+4+obvv0N1fn+S0JH81mP6EqnrIAvn8znR/B1yR5FeT3KC7T+3ub0xy3UyuePvCzLjjk7y4qg7bgzkBAAAAAMCeptizOz6c5PFJbtTd/7m7X9ndX5kd0N2Xd/dZmRR8/mQw/6FVdc85Y/1qkqNn2m9Kcvfufu0g3nuTPGyNWM+oqk3fdqqqb0jyw4PuH+juX+7u/1fY6e5LuvvMJN+WSbHrgG/K5OTSPJaSEwAAAAAAjIFiz/I9Jcmp3X1Gd1+02eDuvjzJjyd58+Crx282d/oWzvfMdF2S5NHd/eV1YnWSn07y/pnuUzI5ibOZpyWZPZnzP7v7BesN7u53JvnZQfdTqurwjYIsOScAAAAA4BBU1cp9YC9S7Fmy7v7b7r5kwTmXJ3nGoPv+c0x9bA7+NX5hd797k1gXJ3n6oHvDwlJVHZdk9mq0TvLUOfb33CQfmWnfNMl9NpmzlJwAAAAAAGAsFHvGY/h2z/Wq6pqbzPmuQfuMOWO9KMmFM+07VdWNNhj/gCSz16Kd1d0f3CxId1+RScFn1oM3mbasnAAAAAAAYBQUe8bjC2v0XXu9wVV1apKbz3RdmOSceQJ193BsZVLQWc/wu3+YJ87UqwftB643cMk5AQAAAADAKCj2jMeN1+j73Abjbz9ov7G7L1sg3us3WW+j7+YqwEy9JcnXZto3qqrj54yzkzkBAAAAAIeoUqnsW6GPN3vYmxR7xuPbBu2PbPL2z2mD9rsWjDccP1wvSVJVh+fg0zYLxeruryU5d55Ya/TvSE4AAAAAADAmij3j8dhB+5WbjD910P7YgvGG44frHXByDn6v56Lu/uwOxVpWTgAAAAAAMBqKPSNQVd+Z5O6D7jM3mXbCoP3xBcN+YtBe72q1YZzhvK3EGq65Xv9O5QQAAAAAAKOxf/Mh7Kaqum6SPx50/3V3v3GTqccM2hcuGHo4/vCqOnJ67dp2xllrznDN7Yo1b04LqaoTsnjh6JRDiQkAAAAAo1CVqhV652aVcmGlKPbsYVW1L8lfJDlxpvtLSX5qjunDwsjFC4a/aJ01Nyv2LBpnrVjzFnt2KqdF/XiSpxziGgAAAAAAsCWKPXvbbyX5jkHfj3b3PG/VXGPQvmTB2GsVQI7agThrxVorznbEmjcnAEald3sDHKJLr7hit7fANvjc176821vgED38Znfa7S1wiI49/LDd3gKHyH8rDgBslTd79qiq+qkkPzPofkZ3v2jOJYanXo5YcAtHzrHmdsRZK9Z6J3aWlRMAAAAAAIyGkz17UFV9f5JnDbrPTPLkBZa5YNAenorZzFonXoZrbkectWKtFWc7Ys2b06KeneQlC845JcnLtiE2AAAAAABXc4o9e0xVPTDJ83Lw6e2/SvL47l7knpphEePoBbcyHH9Zd691CuZQ46w1Z95iz07ltJDuPi/JeYvMWalH6QAAAABgHfuyL/tqdS6Y2ueyLPYovzP3kKq6ZyYnRGaLcK9O8n3dffmCyw2LDycuOP/Gg/b5c8YZzttKrPUKJ8vKCQAAAAAARkOxZ4+oqrskeXkOvprsnCTf3d2XbGHJ9w7aN1lw/nD8e9YZ98Ekl820j6qq43co1rJyAgAAAACA0VDs2QOq6huT/F2SY2a635rkO7v7wi0uOyxk3GrB+adtsl6SpLsvTXLuVmNV1ZFJTp4n1hr9O5ITAAAAAACMiWLPLquqUzO5qu24me53J7l/d3/pEJZ+26B9p6pa5I2mu22y3kbf3XWBOHdIcuRM+1PTN3DmibOTOQEAAAAAh6hW8H+wFyn27KKqummSf0xywkz3h5Lct7sP6T2Z7n5PDj5xc3TmLMJU1dFJvmV2uSR/s8GU4Xf3nSfOOmNfsd7AJecEAAAAAACjoNizS6rqhklek+TEme5PJLl3d39im8K8fNB+3JzzvicHXyn35u7+5AbjX5mD3+05vaqGV7NdRVVVkh8adL9sk2nLygkAAAAAAEZBsWcXVNV1M7m67ZSZ7vMzOdHzoW0M9WeZnGA54HuravhuzXBv10jy5EH3GRvN6e7PJ/nr2WWSPHWO/T02yUkz7Y9kctJpI0vJCQAAAAAAxkKxZ8mq6tgkf5/k1jPdX0xyv+5+93bG6u53JHnxTNcRSZ5XVddaZ2+V5FlJbjHT/cFMCiybeUqSK2baP1hV37fe4Kq6VZJnDrp/tbsv2SjIknMCAAAAAIA9b5HH7dkeL09yp0Hf7yS5flXdZ8G13tLdX9hkzC8m+Y9Jrjlt3ynJ2VX1hO4+68Cgqrplkt9I8pDB/Cd396WbbaS731VVf5rkR2a6/2J66ua/H9hnVR2e5JGZ5HydmbH/nuR5m8VZZk4AAAAAwKGpqlStzpmDyX9bDnuPYs/ynb5G369sca17JjlrowHd/YGqelyS52dyvVqS3C7Ja6vq/CQfTXJCJm8HDf9N9fvd/ZIF9vPEJN+c5I7T9r4kv5Tk56vqQ0m+luTkHPx2TpJ8NsnDu/uyzGHJOQEAAAAAwJ6m2HM10N0vnF5ndkaSo2a+On76Wcszk/zcgnG+WlX3T/KSJPea+eqIJKeuM+3DSb6ru9+3YKyl5AQAAAAAAHvd6pyfY0Pd/YIkt8nkNMxGV5idneT07n5Sd/cW4nw+yX0zuc7tAxsM/XySX09y2+5++6JxprGWkhMAAAAAAOxlTvYsWXfv2qWO3f3BJI+sqh9L8q1JbpHk2CQXZ3L12eu7+xPbEOeKJM9J8pyqum0mV7vdMMlhST6X5B1J3rAd7+YsKycAAAAAYHGTN3tW552bVcqF1aLYczXU3V9O8solxXp7ki2d3FkwztJyAgAAAACAvcQ1bgAAAAAAACO2Uid7qurwJNdOclSShc/TdfdHt31TAAAAAAAAO2jUxZ6qOi7JDyT5jkzehTn+EJbrjPyfBwAAAAAAcPUz2uJGVT0hya8mueaBrt3bDQAAAAAAV1Wplfqr21XKhVUyymJPVf1Rkh/OlX+yevrxJw0AAAAAALhaGV2xp6oeneRHps0DBZ5K8oUkb09yXpILd2d3AAAAAAAAyzW6Yk+SX5n+eKDQ829Jnpzk1d19xa7tCgAAAAAAYBeMqthTVd+U5OszKfQkyTlJ7tvdF+3ergAAAAAAWEvVvlTt2+1tbJtVyoXVMrbfmbef/njgbZ6fUOgBAAAAAACuzsZW7Dl+5uef7O637dZGAAAAAAAA9oKxFXt65sdP7OZGAAAAAAAA9oJRvdmT5KMzPz9m13YBAAAAAMCm9iXZ9/9e5Ri/sZ2e4OpjbMWec6Y/VpKTquqI7r5kNzcEAAAAAABXN1V1jSR3TfINSY5LckmSjyd5Q3d/cJtjnZLkzklOTHJEki8keU+Sc7r74m2Ms7Scttuoij3d/bGqem2SeyY5Ksl3JHnZ7u4KAAAAAACWr6qemuQph7DE87r7hxaMefw05g8lOXqdMW9J8qvdfUh/f19VD07yS0m+eZ0hF1TVmUme1t2fPYQ4S8tpp4zx1NmTk1w+/fmvTSttAAAAAADADqqq05O8K8l/zjpFkak7JPnrqnpeVR2xhThHVtVfJPnfWb/Qk0yee/mJJO+qqrsvGmca6/QsIaedNrpiT3e/Kcl/yeQqt9OSvLSqjt3dXQEAAAAAwOqqqm9N8sok1x989cUkb03y4Vx5UOOARyV5QVXN/XBTVe1L8qIkjxx8dXmSDyV5W5IvDb47PsnfVdW3zBtnGmspOS3DqK5xO6C7f6+qvpbkd5N8e5J/r6rfSPKS7v7C7u4OAAAAAIAkqdqXyd/dr4YR5PKzSf5tgfGfnGdQVR2XSQHmqJnujyT56SQv7+6ejjsxyS8m+dGZcQ9J8sQkvzPnnp6U5EGDvj/K5Aq1T07j7JuOeVaSm0zHXDPJi6vqNt09LAbtdk47bnTFnqr6p5nm+UlunOSmSf4wybOr6sNJzkuy6KNM3d333pZNAgAAAADA8r2lu8/agXWflORGM+0PJfnWA8WXA7r740n+U1V9NMmvzXz1y1X13M0Oa1TV9ZL810H3L3T30wdxrkjyv6vqjUn+JclJ069OTPIzme8do6XktCyjK/YkOT1Jz7QP/Lymn5OT3GzBNWuwJgAAAAAAXO1V1fFJfnLQ/cPDosjAbyS5f5ID7+hcO5NTR8NCztDPJZl9tuXsJL+53uDu/kRVPT7JP850P7Gqfq+7P7fevCXntBR7/szZAnrmAwAAAAAAHLrvTXLMTPvs7n7NRhOmV6A9bdD92I3euZlezfaYQfdTD1yntkGs1yT555muY5M8YqM5WVJOyzTWYk9t8wcAAAAAgG1WqVSt0Ofq+dfJw/dzzphz3mszuRrtgBsk+Q8bjL9rkuNn2h9MctacsYZ7evAm45eV09KMrtjT3ft26HPYbucGAAAAAAB7RVUdkyuvLTvgH+aZOz0J84+D7gduMOUBg/arNzvVMzt20D69qo5ea+CSc1qa0RV7AAAAAACApbh1ksNn2h/q7k8vMP/1g/btNxg7/O6ceYNM39r58EzXEUlutc7wZea0NPt3ewMAAAAAAMD2qKojk5yc5HpJLk3yuSSf7O6vbmG50wbtdy04fzh+uN52xzppsN6bdijORuvtCsUeAAAAAABYDX+QSaHnGoP+y6rqLUn+Lsmzu/v8Odc7ddD+2IL7GY6/aVVdo7svnu2sqqOS3GSbYw33vl7/juS0bIo9AAAAAADskH2plXpN5Cq5nFJViy5yfneftz37uYr1ri7bn+Qu08/PV9Uzkzytuy/fZL0TBu2PL7ifzyS5LFfWIvZlcuLoE4Nx108y+w/y0iSL/jMarjnc+3r9O5XTUin2AAAAAADA1rxsC3OeluSp27yPRRyV5JeSfFtV/cfuvmCDsccM2hcuEqi7u6ouSnLsBmuu1ffV7u5FYq2xt7XirNW/Uzkt1UoUe2pSOr1fkrsn+Q+ZHPc6LpN/2F9J8oUkH0nyf5O8rrtfvUtbBQAAAACA7dRJ/k+Sv03yxiTvTvL5JFdkcuLkm5M8MMmjc/D1bqcneWFVPWiDEz7DIsZWrirbSrFnq3E2WnO7Yyn2bJdpkecnkzwhyU1nv5r5+XWnn5OT3DPJL1TVh5P89yR/sIXqIAAAAAAA7AX/kOT53f2+db7/5PTzN1X135K8MMndZr5/QJIfT/L768wfvv1zyRb2+LVB+6hdjLPsWEsz2mJPVX19kucnuWsOLu709LPmtOmPN0vyu0keUVWP7O5FH2ACAAAAAGATVckW3rTZs9ZI5UFJzl1wmfO3ZTNJuvucBcZ+vKruk+SfknzLzFe/WFVndPdX15g2PPVyxBa2eeQmay4zzrJjLc0oiz1VdYMkZyU5KZMCzoHiTuXKgs4VmRylOioHv5o1O/Zbk/xTVX1rd39mZ3cNAAAAAMCKObe737nbm5hXd19cVY/K5Kq3A/WBEzJ5JuWv15gyfM9neCpmHsNTL2u9EbSsOMuOtTT7Nh+yt0yvbntZJqdzkknxppJ8PMmvJ7l3kut39/7uPra79ye5/rT/15J8LAcXiE7J2r+JAQAAAABgpXT3B5K8fNB9v3WGD4sYRy8Sa/r3+Vsp9lyzFj8SNtzbvMWencppqUZX7Enyg0nulCuLPJcleXKSU7r7F7v7td39+dkJ3f35af8vJbl5kp9LcmmuLPjcuap+cGkZAAAAAADA7nnNoH3qOuPOG7RPXDDO1+XgG8auSPLZNcZ9Ngc/z3J4JieOFnHjQXu49/X6dyqnpRrjNW7/JVcWei5N8uDu/rt5J3f3ZUmeWVVvT/KKJIdN1/rZJP9z+7cLAAAAAHD1VLUvVWM8c7C2Fcpl+I798euMe++gfZMF4wzHf6S7r/K+TXdfVFUfTXLTwdxFnl8ZxnrPOuOWktOyjep3ZlWdmOS202YneeYihZ5Z3f2qJM/MlW/83Ga6PgAAAAAArLJLB+3D1xk3LJjcasE4p22y3m7EWmZOSzOqYk+Su0x/rEyORv3+Ia73e9N1DhwPu8sGYwEAAAAAYBXcYNA+f51x78zBhaGTquqGC8S526D9tg3GDr+767xBpns6aabr0iTvWmf4MnNamrEVe75u+mMn+XB3f/pQFpvO/1CuPN3zdRsMBwAAAACAVfCtg/bwWrckSXd/JcnZg+77zhOgqirJfQbdr9hgyt8M2veZrjGP+w3ar+3uC9YauOSclmZsxZ5rz/z889u05uw619qmNQEAAAAAYM+pquskeeig+zUbTHn5oP24OUPdM8nNZtqfSfKGDcafk+SzM+2Tk5w+Z6zhnl62yfhl5bQ0Yyv2HCjMVJITtmnN2YenvrBNawIAAAAAXO3VCv5vBTwzyXVm2pck+bsNxr8wyYUz7btX1b02CjA9AfOUQfdzu/uK9eZMvztz0P2UzU73VNW9k3zbTNdXkrx4ozlZUk7LNLZiz+y1bTepqputO3IO0/kn5co3ew7pWjgAAAAAAFiGqnpyVd1hgfH7q+q3c9VTLH/U3Z9ab153n5fkfwy6/7SqbrRBuF9IcveZ9peS/NYc2/zNJLPXr90jyc+vN7iqbpzkTwfdv9vdn11r/AFLzmkpxlbsOSeTwsyB4szPHeJ6B+bXdM1zDnE9AAAAAABYhm9P8uaqen1V/XRV3aaq9g8HVdW1q+r7krwpyc8Mvj43ya/MEesZOfiwxM2SnFNV3zV78qaqTqyqP0rya4P5v9bdmz7NMi3S/Pqg+zeq6tmzhZiq2ldVD87k7/RPmhn7ySS/PUc+yZJyWpar/MLvZd19flW9McmdMynQ/EhVvaG7z1x0rar6wSQ/kisLR2/q7vO3bbMAAAAAALDz7jr9JMnXqurjmZw6uTzJ9TIphqx18OPTSb6juz+3WYDu/nxVfU+SVyW5xrT7ppm8jfPFqvpQJlfD3STJYYPpL8vk6rh5/eY0nwfO9P1YJvWAj2SS281y8FV0SXJRkkd09xfnCbLknHbc2E72JJNf6AMncSrJGVX1rKq61jyTq+rYqvqdJM890DX98enbvlMAAAAAgKuxfVUr99njjkxySpJvTnKnJCdn7TrAK5PcrrvfP+/C3X12kgckGZ5muU6Sb8qkADMsijw/yfd0d2dO0zdwHp7JuzqzDsskn2/KVQs9n0vynd39+nnjTGMtJadlGF2xp7v/Osmrc3DB5yeTfLyqzqyqR1XV7adHq46b/ni7af+ZST6e5KdzZe6d5NXd/bKlJwMAAAAAAFvza0n+KMk7MznFs5kLkrwkyT26+wHTd2sW0t3/lORWSf4wyVc3GPrWJA/t7kd299e2EOfi7v6+JA9L8rYNhl6Y5NlJbtXdZy0aZxprKTnttFFd4zbj4Ulel+R2ubLgc0ySH5x+NnKg9Hpg3tum6wEAAAAAwCh096szORiRqrpmJgWLk5LcMJO/L9+X5ItJvpDkXUne3t3zFIU2i/uZJD9eVf8lk+vWTsvkJMwlST6R5A3d/YFDjTON9dIkL62qmye5S5IbJzkik7zeneT13X3xNsRZWk47ZZTFnu7+clXdK8lzkjwkV767k1xZzFlzaq4s8lSSlyb5ke7+yk7tFQAAAAAAdlJ3fzXJm6efZcW8KMlrpp+djvWBJDtebFlmTtttdNe4HdDdX+juh2VyKuf/5soCzv8bMvM54MCYczI5bvXw7v7CkrYMAAAAAACw7UZ5smfWzDGu05LcI5OjXDdNclwmR9UuyOSY2kcyKQq9rrvfs0vbBQAAAAC4GqnUeM8crGGji6Vg94y+2HNAd787kzv6/mi39wIAAAAAALAsq1RSBQAAAAAAuNpR7AEAAAAAABixlbnGDQAAAACAvaWqUrU679ysUi6sFid7AAAAAAAARkyxBwAAAAAAYMT2zDVuVXX5oKu7+yr7W2PcdlkzHgAAAAAAwF62l4ob81526FJEAAAAAACAqb1U7EmSnv64WUGn5xizSEwFJAAAAACAbVa1L1Wr85rIKuXCatlLxZ6zc2WxZzvGAQAAAAAArLw9U+zp7tO3cxwAAAAAAMDVgTNnAAAAAAAAI7ZnTvYAAAAAALBaavq/VbFKubBanOwBAAAAAAAYsdGd7KmqR800X9XdnzmEtW6Q5H4H2t3954eyNwAAAAAAgGUbXbEnyZlJevrz+ybZcrEnya0H6yn2AAAAAAAAozLGYk+SVK4s0OzF9QAAAAAArvYqlarVeU3Emz3sVWP9U6YwAwAAAAAAkPEWe7bLbBlWAQkAAAAAABidq3ux55iZn1+4a7sAAAAAAADYoqt7secbZ37+hV3bBQAAAAAAwBbt3+0N7JaqulmSH8mV17e9axe3AwAAAACwcqoq+6o2HzgStUK5sFr2ZLGnqv5szqFPqqofWGTpJNdMcrMkt09y2LSvk7x2kT0CAAAAAADsBXuy2JPkh3LliZu1HCif3m+L6x+YfyDGBUn+fItrAQAAAAAA7Jq9WuzZaQeKPJXk4iSP6e7P7OJ+AAAAAAAAtmQvF3vmufxwqxckXp7kfUn+Icmzu/v9W1wHAAAAAIB1VPalsm+3t7FtVikXVsteLfbcbJ3+SvLBXHky5weSnLPAulckuTDJl7v7sq1vDwAAAAAAYG/Yk8We7v7Iet9VHXSY5zMbjQUAAAAAAFh1e7LYs4mP5sqTPRft5kYAAAAAAAB22+iKPd190m7vAQAAAAAAYK8YXbEHAAAAAIBxqKrh0xyjtkq5sFr27fYGAAAAAAAA2LrRneypqqOSPHym6/909/u3sM4tk/yHma4XdPelh7o/AAAAAACAZRpdsSfJ9yf5k+nPL0ly8hbXuWC6zuHT9qVJXnBoWwMAAAAAAFiuMV7j9kNJDlyM+Jfd/amtLNLdn0zyV9O1arouAAAAAADbpFbwf7AXjarYU1XXzuTqtZ52vfgQl3zhzM/vPr0iDgAAAAAAYDRGVexJcvskh+XKkz1nH+J6r8uVhaMjpusDAAAAAACMxtiKPbec+flHu/tLh7LYdP7HZrpOPZT1AAAAAAAAlm1sxZ7rTH/sJJ/dpjXPn/n5dbdpTQAAAAAAgKXYv9sbWNBscWq7XsKaXefIbVoTAAAAAOBqr6pSNbYzB+ur2q6/lobtNbY/ZQdO81SSG2zTml838/MvbNOaAAAAAAAASzG2Ys95Mz+/YVXd5FAWm86/cSbXwg3XBwAAAAAA2PPGVux5QyaFmQPFmUce4no/MP3xwNm7Nx3iegAAAAAAAEs1qjd7uvu8qnp7kttmUqD5uap6Xnd/ctG1qurGSZ6USeGokry3uz+2rRsGAAAAALg6q1qtd25WKRdWythO9iTJn2RSnOkk107y91V1wiILVNXXJfn76fwDa/3JNu8TAAAAAABgx42x2POcJB+d/ryT3CbJO6rq0VV1+EYTq+rwqvqhJG9PcqtceR3cx5M8e2e2CwAAAAAAsHNGdY1bknT3pVX1fUn+KckRmRRsrp/kz5L8VlX9U5I3JzkvyQVJjklyQpI7JrlXkuvlytM8leTiJN/X3ZcsORUAAAAAAIBDNrpiT5J09/+pqsckOTPJ4bmycHP9JA+fftZy4ELFA+O/luTR3X3Ojm4YWFlXdG8+iD3t0iv8Go7dly+9bLe3wCF6/Wfet9tbYBvc6rgb7vYWOERjvPaBg7X/bzp6K/WmBcDUvun/VsUq5cJqGe3vzO5+YZJvy+RKtwMndQ78P9ta45OZMZXk3CR36+6XLHHbAAAAAAAA22q0xZ4k6e43JTk1yX9K8p4cXNgZOvDdO5I8Psmtuvtfl7FPAAAAAACAnTLKa9xmTd/a+ZMkf1JVN87ktM8pSa6b5NgkX0ny+STvT/LP3f2p3dorAAAAAADAdht9sWdWd38iyQt3ex8AAAAAAADLslLFHgAAAAAA9o6qStV6L2+MzyrlwmoZ9Zs9AAAAAAAAV3eKPQAAAAAAACOm2AMAAAAAADBie+bNnqq6+7Cvu8+eZ9x2WSseAAAAAABbsy/JvhV658bpCfaqPVPsSXJWkp5pd9be33DcdlkvHgAAAAAAwJ61F4sb85Z5V6ccDAAAAAAAsEV77dSZQg8AAAAAAMAC9tLJnqdt8zgAAAAAAICVt2eKPd09VxFn3nEAAAAAAOyuSmXfCl3UVCuUC6tlr13jBgAAAAAAwAIUewAAAAAAAEZMsQcAAAAAAGDE9sybPQAAAAAArJaqStXqvHOzSrmwWpzsAQAAAAAAGDHFHgAAAAAAgBHbM9e4VdUHd3kL3d2n7PIeAAAAAAAAFrJnij1JTkrSSXbr0sPepbgAAAAAAABbtpeKPQfMW3Q5UBQ6lCJNZXcLTAAAAAAAK2tfVfbV6vz16yrlwmrZS8Wej2bxws21k1wnVy3YXJ7kK0kuTHJ0kmOTHDbz/YE4X0jy5S3sFQAAAAAAYE/YM8We7j5pkfFV9cgk/yNXFnremuR5Sc5K8q7uvmxm7P4k/z97dx53+zXfC/zzPTmJRJKKEFMiQozBRY1NilCdDKU1daYUrdtbRauUFm3VUJ1vZymq5g6UakuVIm615lZQgpBQSURUIpHpe//Y+8g+vzzD3s+wn7P3eb/P6/c6z1p7rfVd65yzc07291lr3TrJPZM8MsntJ/o9vbtfsdn5AwAAAAAA7IRdOz2BjaiqJyf58yTflOSCJD/c3Xfs7t/t7g9PJnqSpLsv6+4PjV//5iQ/ktHOn2skeVlV/cyclwAAAAAAALAl9pmdPdOqqnskeUFGu3IuTHJyd39gljG6++VVdVqSd2R0zNsLq+q93f2uLZ8wAAAAAMB+qrIrtZh7Dla0TGthuSzin8znZZTo6SS/OGuiZ49xv2eOi7uSPH9rpgcAAAAAADA/C5XsqaqbJLnbuHhpkj/b5JCnJLlk/PXdxuMDAAAAAAAsjIVK9uTKRE8n+XR3/89mBuvuryT59ArjAwAAAAAALIRFu7Pn6Imvv7pFY06Oc/SqrQAAAAAAmElVsqtqp6exZZZoKSyZRdvZs0clOWaLxtqqcQAAAAAAAOZu0ZI9Z018fd2quv1mBquqb05yvYyOhUuSz29mPAAAAAAAgHlbtGTPB8Y/70nOPH+T4z13/POezXfv3+R4AAAAAAAAc7VQyZ7u/miS/xgXK8l9quoPq2Y7KbFG/jDJt2eUOOok/zEeHwAAAAAAYGEsVLJn7OkZJXp6/PNjk7y3qr5zvaTPOMnzXUneN+63Z4wkecb2TRkAAAAAYP+zq2rpHtgX7d7pCcyqu99YVX+a5DG5MllzhyRvSvLFqnpXkv9M8qUkX0ty9STXTnKbJN+a5Dq5MsGz5zi4P+vuN85tEQAAAAAAAFtk4ZI9Yz8x/nlPwicZJXCul+TB42clwyRPJTklo10+AAAAAAAAC2cRj3FLjzwuyQ8mOSdXHuvWE81q4vlG14nXzk3yw939mO6e7AcAAAAAALAwFnVnT5Kku19VVW9M8kNJfizJnbJyAmtPwueKJO9N8mdJXt7dF8xlogAAAAAA+6GqyjpXrS+UZVoLy2Whkz1JMk7Y/HGSP66qQ5PcMcmNkxyR5LAkFyY5P8mnkryvuy/cmZkCAAAAAABsvYVP9kwaJ3LeMX4AAAAAAACW3kLe2QMAAAAAAMDIUu3sAQAAAAAAtl9VHZzkxCS3THLNJJckOTPJe7r7U1sc6/gkd0lyTJKDknw5yceSvLu7L97COHNb01aT7AEAAAAAYFtUKrtSOz2NLVMLupaqemWS7x9Un9Hdx21grKOSPDPJI5Mcukqb9yX5le5+/azjD8Z5UJJfTPLNqzS5oKpekuTZ3X3uJuLMbU3bZSmOcauqm1bVz1XV31bVJ6rqS1V12fi59yp9rl9Vx46fo+Y9ZwAAAAAA2G5V9YBcNdGz0bFOTnJakv+dVZIiY3dM8rqqemlVHbSBOFerqr9I8jdZPdGTJIcl+akkp1XVPWaNM451cuawpu220Dt7quomSX4jyQOSb6RUJ1OrvUb3ZyX58fHX51TV0d19+ZZPEgAAAAAAdkBVXSPJH27RWN+a5E1JDhm8dH6ST2d07NkNkxww8dqPJjmsqh7S3Wt9Xj8ZZ1eSVyd54OCly5N8NslXktw4yTUmXjsqyd9X1X26+/9NtaDMb03zsLA7e6rqwUnen+R7ctV1TPML/Bt7hsroD8L9t252AAAAAACw4349ydHjry/c6CBVdc2MEjCTSZEzkjwoyZHd/c3dfeMkxyX540H370vyxBnC/Vyumuj5oyTHdvdNuvsOSY4cj/vZiTZXT/KacYJrXXNe07ZbyGRPVd03yauSfNNkdZIvJvn3ZP2DE7v7v5KcOlH1fVs5RwAAAACA/V2lUrVEzwLd2TM+nmzP6VZXJHn2Job7uSQ3mCh/OsmJ3f36yd0t3X1md/9EkqcP+v/SOLmy3pyvtULfp3X3T3b35yfiXNHdf5PkxCSfmWh7TJInTbOgzGlN87JwyZ7x/TqvzGjbVGeU2Hltktt19w26+67jptPs7vmrPcMmuc9WzxUAAAAAAOatqg5J8qJcuTHi9zLaKLGRsY5K8n8G1Y+ZTL6s4LlJ3jFRvkaSn50i3FOSHD5RfkeS56/WuLvPypUJrT2eOE4arWrOa5qLhUv2JPnF7P2b/ZTufnh3/8cGxnrbxNfXq6pjNzc1AAAAAADYcb+S5Pjx159N8oxNjPX9SQ6bKL+ju9+6VofxzpjhTqJHVdWqW6PGd/X82KD6WevdizOeyzsnqg5P8rC1+mROa5qnhUr2jH+zfzijXTud5C+7+4WbGPK0JJdMlG+1ibEAAAAAAGBHVdWdk/zMRNX/7u4LNjHk8P6cU6bs97aMjkbb43pJ7rZG+xOTHDVR/lSSt08ZazinB63Tfl5rmpuFSvZk9It2RK7cevarmxmsuy9LctZElZ09AAAAAAAspKo6MKPExQHjqtd29xs3Md5hSe4xqH7zNH3HO2H+aVB9/zW63G9Qfst6u3om2w7KJ1fVoSs1nPOa5mbRkj03m/j67A0e3TZ0/sTX19iC8QAAAAAASLKrKrtq1xI9+8SJXWt5WpLbjr8+P8lPb3K8Wyc5cKL86e7+7xn6nzoo336NtsPX3j1tkPFdO5+ZqDooyQmrNJ/nmuZm0ZI9e7ZwdZIzt2jMyya+3r1FYwIAAAAAwNxU1QlJnj5R9fMzJjFWMrz65LQZ+w/br3WVyrxizXNNc7NoyZ4rJr7eqrkfOfH1l7doTAAAAAAAmIvxffenZLSjJUnemeRPt2DoWwzKn5ux/7D9jarq4GGjqjokV71mZbOxhnNfrX5b1jRvi7aT5Zzxz5XRxUebUlVXT3KjjHYKTY4PAAAAAADrOb5mP9rtnO4+e4vn8dMZ3XmfJJckeewM992s5TqD8qwnbn0xo9O19uQidiW5VpKzBu2undHn/ntcmmTWX6PhmMO5r1a/XWuaq0VL9nxm4uvrVdWNuvuMTYx3r1z5a9BJPriJsQAAAAAAmLArlV3Z5++5mdoKa3n9BoZ5dpJnbXoyY1V14yS/OlH13O7+2BYNf9igfOEsnbu7q+qiJIevMeZKdV/bQLJqOLeV4qxUv11rmqtFO8btX5NckCt34jxyk+M9ceLrz3b3pzY5HgAAAAAAzNOfJDl0/PXHkvzaFo49TGJcvIExLlpnzHnGmXesuVmoZE93X5rkHzLazlVJnlRVx21krKr68ST3zihx1Eles0XTBAAAAACAbVdVj05yn3GxMzq+7ZItDDG8i2YjY399UD5kB+PMO9bcLNoxbsloi9v3ZZTsOTzJP1bVd3b3Z6YdoKoel+R3MvrDX0m+luSFWz9VAAAAAACW2AOTnD5jny25O76qrp+9P9d+UXe/cyvGnjDc9XLQBsa42jpjzjPOvGPNzcIle7r7I1X1h0n+d0bJmpsl+Y+q+u0kL+vu/xp2SZKqul5GO3l+Ksldc+VlT53kWd29JW8wAAAAAADGqlK1PHf25KprOb27P7ITU0ny+0mOGH/930mesg0xLhiUh7tipjHc9TIcc55x5h1rbhYu2TP2Mxkleb4jo2TNoUl+IckvVNWey5Rq/NprquqQ7P2Lv+e1SvKa7rarBwAAAACAhVBVD03yvRNVT+ju87ch1DCJceiKrVZRo0zfRpI9V6+q6u6eIdxwbtMme7ZrTXO1UHf27NHdl2f0B/lluTJxk/HXhw3K10py9Vx5z08mXvvjJD8yhykDAAAAAMBW+fWJr/+uu7frTvqzB+VjZux/3ey96eSKJOeu0O7cXPm5fpIcmOQ6M8Y6elAezn21+u1a01wtZLInSbr7ou5+RJIfSPLx7H0s256fh0/G7T6Z5Ae6+ye7+7L5zRoAAAAAADbtiImv71dVvd6T5G2DMW60QrvbD9p8fFA+dsZ5Dtuf0d1Xud+muy9K8tktjvWxVdrNZU3ztrDJnj26+9XdfUKS707ye0k+lOSSXLmTp5JcnuRTSV6U5CFJbtndr96ZGQMAAAAAwEIYJkxOmLH/rdYZbydizXNNc7Ood/ZcRXf/Y5J/3FOuqqtnlN382jadVQgAAAAAwBp2VWVX1foNF8QyrWVKH0lyaUbHqiXJcVV1/e7+wpT9TxqUP7hG2w8m+c6J8olJXjpNkKq6fpLjJqouTXLaKs3nuaa5WahkT1XdLKMdPHv8U3ev+BvW3V9L8rW5TAwAAAAAAObngbkyWTGt2yV54UT5i0l+eNDmk5OF7v5qVb0jybdNVH97kj9fL1hVVZL7DKrfsEaXNyb5+YnyfaqqurtX6zDhOwblt3X3BSs1nPOa5mahkj1JvivJb42/7iTH7+BcAAAAAABg7rr7X2btU1XD++sv7u5/mqLr32bvxMijM0ViJMm9ktx4ovzFJO9Zo/27k5yb5Nrj8k2SnJyr3jW0kkcPyq9fp/281jQ3i3Znz2G58h6ez3f3GTs8HwAAAAAAWGavSnLhRPkeVXXvtTqMd8A8c1D94u6+YrU+49deMqh+5nistWJ9W5K7T1R9Nclr1uqTOa1pnhYt2XPO+OdO8vmdnAgAAAAAAGurJfyxv+nus5P830H1i6rqBmt0e1qSe0yUv5Lk16cI9/wkk8ev3TN7H+22l6o6OsmLBtW/093nrhVkzmuai0VL9kwmeK6xY7MAAAAAAID9xwuS/PdE+cZJ3l1V3zO586aqjqmqP0rynEH/53T3eesFGSdpfm1Q/dyq+oPJRExV7aqqB2V09NtxE20/n+Q3plhPMqc1zcui3dnzr0kuzejiqeOq6tDuvnCdPgAAAAAAwAZ193lV9fAk/5jk4HH1jTK6G+f8qvp0kiOSHJvkgEH31yd54Qzhnp/kxCT3n6j7ySSPraozMtpRc+NxvEkXJXlYd58/TZA5r2nbLdTOnnGW7B/HxYOSPGQHpwMAAAAAAPuF7n5HkvslGe5mOSLJHTJKwAyTIq9I8vDu7hniXJHkoRndqzPpgCQ3Gcc6YvDal5Lct7tPnTbOONZc1jQPC5XsGXtuRnf2JMlzquqonZwMAAAAAADsD7r7n5OckOQPk3xtjaYfSPLg7v6h7v76BuJc3N0/kNGGjw+u0fTCJH+Q5ITufvusccax5rKm7bZox7ilu/9fVT0tyfOSXD/JP1fVw7r7ozs8NQAAAAAAJuyqZNeV158svF0LvJRxMmTTK+juLyZ5fFU9OaPj1m6V0U6YS5KcleQ93f3JzcYZx/qrJH9VVTdNctckR2d06tf5ST6a5NTuvngL4sxtTdtl4ZI9SdLdL6iq85P8VpJbJ/lAVb0qo21d/7YvXYoEAAAAAADLprsvSvLW8bPdsT6ZZNuTLfNc01ZbuGRPVX1qonjZ+OeDkvzI+ElVXZDkf5JcOsPQ3d3Hb8kkAQAAAAAA5mThkj1Jjsvozp4a/7zn/p7J7WeHj59Z7FOXKQEAAAAAAExjEZM9ewyTM5tJ1izwSYsAAAAAAPumqlqqO3tqidbCclnEZM9nYxcOAAAAAABAkgVM9nT3cTs9BwAAAAAAgH3Frp2eAAAAAAAAABu3cDt7AAAAAABYDJVdqSXac7BMa2G5LESyp6pOSPIdSW6b5NoZzftLST6d5J+TvKu7L9+5GQIAAAAAAOyMfTrZU1W3SfIbSe6zRrNnJPl0VT29u189n5kBAAAAAADsG/bZPWdV9V1J3p1Roqcmnm80mXhukuQVVfWCec8TAAAAAABgJ+2TyZ6qummSv0xyWEbJnB4/yd5Jn554KsmTq+qx850tAAAAAADAztlXj3H7wyRXz94JniuSfDDJp5JcmuQGSe6c5JDsnfB5YVX9ZXefN+c5AwAAAAAwYVcqu6rWb7ggdmV51sJy2ed29lTVrZJ8W65M3lSSlye5UXffsbsf2t0/2N0nJ7l2kqcluWxiiEOTPGK+swYAAAAAANgZ+1yyJ8lDJ77uJL/d3T/S3WcNG3b3Rd39/CQPmWifJA/e5jkCAAAAAADsE/bFZM+dxz9XkrOSPGW9Dt39hiSvzpU7ge5QVfvi2gAAAAAAALbUvnhnz63GP3eSV3X3ZWs1nvCSJN8//vrgJMdldL8PAAAAAAA7oKpSS3RnzzKtheWyL+5+uebE1++fod/7BuUjNj8VAAAAAACAfdu+mOy5Rq68e+dLM/T78vjnPX0P37IZAQAAAAAA7KP2xWTP5Jwun7ZTd18xqDpga6YDAAAAAACw79oXkz0AAAAAAABMafdOT4D5qqrbJLldkqOSHJrk/CQfS/Lu7r5oC+NcK8m3Jrl+kiOTfD3JGUne092f26o4E7FOSnJ8Rmu6MMnpSU7t7lmOAgQAAAAAtlBVsqtqp6exZZZoKSwZyZ4dVlVHJ7lLkruOf75T9r5v6IzuPm6TMQ5L8oQkj09yg1WaXVRVr0ryrO7+7CZifXeSZyS5W1bZOVZV/5rkOd39xo3GGY9zuyS/nOT+q8S6vKr+LskvdveHNxMLAAAAAAD2Vft6sud2VXXZvPp29zs2GGsmVXVSkidnlOBZLfmyVbHunOR1U8Q5JMmPJXlYVT2qu18zY5zDkvx5ku+dovndkryhql6W5LHdffEsscbxnpDkhVn7z/ABSb4nyX2r6knd/XuzxgEAAAAAgH3dvpzsqYw+zN9Iv2ygb2d+vx53znRJkU2pqnsleVOSgwcvfT2jY84uSnJMkutOvHZokldW1QHd/cop4xya5F+SfPMKL5+R5OwkR2R0zNrkDpwfSXJkVT2wuy+fJtY43pOS/MYKL30hyeczSmxdf6J+d5Lfrarq7t+dNg4AAAAAACyCFY/Z2kd0RombWZ6eeGbtu6+ctnjBVgxSVcckeXX2TvScl+Qnkly7u2/d3Xfq7utllHx660S7XUlOqaoTpgz3olw10fPiJMd393HdfZfuvnmS6yX5lSSXTLS7X5JfmmFdJyZ5waD67Unu2N03GK/pBhmt6V8G7X6jqu4ybSwAAAAAYHN2pZbugX3RvpzsSfZO3kzzbLTvTvlqRomKX0/y0CTHJXnAFo39nCRHTZTPTnJSd/9xd++VUOru9yb5jiQvnag+JFdNqlzFePfQ9w+qn9zdj+ruTw3inNPdv5TkgUkunXjpKePk1DR+PaPj2fZ4Q5Lv7O73D2LtWdPfTVTvHvcHAAAAAIClsS8e4/bZ7GwCZh7ekOTNST7W3VdMvlBVN97s4FV1wyQ/OKh+XHd/bLU+3X1FVf1Ekm9JcvNx9f2q6qTuPnWNcE8dlF/f3b+51vy6+x+q6rm5ckfPwUmemeQxa/Wrqu9OcuJE1ZeSPLq7L1mpfXdfUlWPSnJakmuNq+9RVd/e3W9ZKxYAAAAAACyKfS7Z093H7fQctlt3n77NIe6XvX9vP9Ldr1uvU3dfXFUvTPInE9WPTbJisqeqDkty70H1c6ac428l+dkkVx+Xf6iqntDdX1ujz48Pyr/f3eesFaS7z66qP0jyi4NxJHsAAAAAAFgK+/oxbmzMPQflN8zQd9j2AVW1WlLwpOydVPpCd//7NEG6+/wk75yoOiTJd6/WvqquluQ7B9V/Nk2sFdp9d1UdNGVfAAAAAADYp0n2LKdjB+UPTduxu/87o/t99rhmkrtPGefD08ZZZV7fs0bbk5McOlH+eHefMU2Q7v5Mkk9MVB2eqybEAAAAAIAtVlVL98C+SLJnOV1rUD5vxv5fGpTvsMNxkuT2g/K7Z4w1PIpuOB4AAAAAACwkyZ7ldMWgfMCM/Q8clG81pzg3r6rVxhjO4bQZYw3br7YmAAAAAABYKJI9y2m4w+Y6M/Y/alBeLTGy2TjD9ldLcuNV2t5iUP7cjLGG7YfjAQAAAADAQtq90xNgW3wqyUkT5Tsneek0HavqZkmuMai+7hpxJn1zVe3q7uGOn9XcaYW66yb55Ar1w8TQmVPG2OOsdcbbsKq6Tq6aIFvP8VsVHwAAAAD2VVW7squWZ89BLdFaWC6SPcvpnUl+ZKL84Kp6YndfOkXfH1ih7rBV2v5rkktz5XFs35TkvkneuF6Qqjouyd1miDWsv3C9GOu0Xy3ORjw+yTO3cDwAAAAAAJiaNORyelNGSZg9rpfkCet1qqrrJfnpFV5aMTHS3V9L8pZB9S9X1fAunpU8Jyv/+Zs22XPxFDEmXTRlHAAAAAAAWCh29iyh7j6rql6VvXf3PKeqPtndr1upT1VdM8nrklxrhZcPWSPcb2S0m2ePOyR5cVU9srsvWyXWLyT5wVXGWy3WwYPyJWvMaSVfnzIOzGRX1U5PgU066AC/h4vuGlN9jwH7sgfe6DY7PQW2gP+aAgAA7BzJnuX1tCTflSvvkjkoyV9X1cuSnJLkwxntdjkmyf2SPDXJ9cdtz09yxMRYF6wWpLv/uapekb2TNz+U5HZV9fwkb0ty9ni8u2W0w+jbxu0uSXJ59k68rBbr4iRXnygftNqcVnG1FcbbKn+Q5LUz9jk+yeu3cA4AAAAAsM+p8Y9lsUxrYblI9iyp8e6ehyd5Q5JDx9WV5EfHz2rOSvLsJH8yUXf+OuEel+RmSe48UXebJC9bp99jkvxm9k72rBbrguyd7Bnu9FnPcCfPqgmsWXX32RkltKZWdoMAAAAAALBF3NmzxLr7bUnunuT0Kbt8IMm9k3x1UH/+OnEuSHLPJK+YMs4FSR7V3X+eq96ds1qsYXLm0BVbrW7YfsuSPQAAAAAAsJMke5Zcd38gya0y2n3zriQr3aPzoYyOV7tbd/9XrnpvzyeniHNRd/9QkrsmeWWSc1dodl5GR57dvrtfXFWHZu/j1TrJp1YJMdw5c8x6cxo4ep3xAAAAAABgITnGbT/Q3ZdmdCzbn4wTLDdMcu0kFyb5XHcPEzMnDMrvnSHWvyX5waraleTYjO4MOijJmUnO6u7JZNMwzn919/+sMvTHM7rzZ49jp53TKu0/NmN/AAAAAADYJ0n27Ge6+8Ksn+i47aA8dbJnIs4VST4zfrYiznDOw0TRem61zngAAAAAwBbbVcmuJbq/etfyLIUl4xg39lJVR2bvHTRfSfLubQp3v0H579do+8FB+cQZY520zngAAAAAALCQJHsYenCSAyfKr+zur211kKr6piT3naj6cpK/WqPL2zM6dm6Pm1fVjaaMdVySm01UfXU8HgAAAAAALDzJHr6hqg5O8vRB9Z9uU7inJzl4ovwX3X3xao3Hr715UP2oKWMN2/1Dd18yZV8AAAAAANinubOHSc9LMrlb5hXd/f6tDlJVd03yMxNVX07yq1N0PSXJ906U/3dV/d/uPmeNWNdJ8vgVxgEAAAAAttmu1HLd2ZPlWQvLxc6eJVZVt5yy3QFV9WtJnjBR/eUkT5yy//FVNVXisKpOyuhunoMmqn++u89er293/12Sf52oulaSU6rqwJXaV9VBGSV2rjVR/c7u/sdp5goAAAAAAIvAzp4dMk56HLLCS7cblA+uqvusMsznu/u0NcK8s6o+kuSVSf6+uz87mMNhGd2b89Qkd5h46Yokj54mATP2k0keXlUvS/K6JB/o7ksn4lSSuyR59PiZTDL+dZIXTRknSX4uyb9MjPGAJG+uqidP7kKqqjsm+Y0k95zoe3mSp8wQCwAAAAAA9nmSPTvn5dn7yLTVXDfJW1Z57aVJHrlG38oo2XHPJKmq85KcmeTiJNdJcoPsvcMmGSVEHtPdfzPF3CYdk+Rp4+eSqvp0kq8kOXwc5xor9HlTkh/q7p42SHe/q6qeluT5E9UnJ3lfVX0+yRfG8a6/QvendPe/rlAPAAAAAAALS7Jn/3Lk+FnNmUke0d3/vMk4ByW5xRqvX5bkBUme2d2XzTp4d7+gqi7PKOFzwMRLNxg/Q5cn+dnu/u1ZYwEAAAAAwL7OnT3L7deS/L+Mkitr+WRGx5vdYoOJnr9J8tqM7vlZy4VJXpLktt399I0kevbo7t9Icqckf5fRsXMruSLJG5PcUaIHAAAAAHZCLdWP0WFKsO+xs2eHdPdxc4jxm0l+s6quntGdPDfN6Pi2g5N8LaOdPO/v7k9sMs6pSU4d381zyyQnZHSs2+EZ7ao5J8lHk/x7d1+ymViDuB9Mcv+qunaSb01ykySHZpRUOj3Jqd197lbFAwAAAACAfZFkz36gu7+W5NTxs51xOqOkzke3M84Kcc9N8rp5xgQAAAAAgH2FY9wAAAAAAAAWmJ09AAAAAABsi11V2VXLc8/NMq2F5WJnDwAAAAAAwAKT7AEAAAAAAFhgkj0AAAAAAAALTLIHAAAAAABgge3e6QkAAAAAALCcqnalann2HCzTWlgu/mQCAAAAAAAsMMkeAAAAAACABSbZAwAAAAAAsMDc2QMAAAAAwLbYlWRXaqensWXsnmBf5c8mAAAAAADAApPsAQAAAAAAWGCSPQAAAAAAAAvMnT0AAAAAAGyLqmTX8lzZk1qitbBc7OwBAAAAAABYYJI9AAAAAAAAC0yyBwAAAAAAYIFJ9gAAAAAAACyw3Ts9AQAAAAAAllNVpap2ehpbZpnWwnKxswcAAAAAAGCBSfYAAAAAAAAsMMkeAAAAAACABebOHgAAAAAAtsWuVHZlee65Waa1sFzs7AEAAAAAAFhgkj0AAAAAAAALTLIHAAAAAABggUn2AAAAAAAALLDdOz0BAAAAAACWU1WlqnZ6GltmmdbCcrGzBwAAAAAAYIFJ9gAAAAAAACwwyR4AAAAAAIAF5s4eAAAAAAC2RaWya4nuuaksz1pYLnb2AAAAAAAALDDJHgAAAAAAgAUm2QMAAAAAALDAJHsAAAAAAAAW2O6dngAAAAAAAMtpVyq7Ujs9jS2zTGthudjZAwAAAAAAsMDs7AEAAAAAgAVXVQcluWWS45IcneTwJAcm+Z8kX0ry4SQf7e7Ltyje7iR3TXKbJNdKcnmSLyR5X3d/ZCtiTMQ6Osm3JLlRkkMyWtN/JXlXd1+whXHmtqatJtkDAAAAAAALqKoekuQ+SU7KKNGz3mf+X6mqVyb5ne7+2AZjHpbkqUl+MsmRq7T5eJLnJ3lJd/dG4ozHuWeSZyU5eZUml1TVq5P8Und/ZhNx5ram7eIYNwAAAAAAtkVVUlVL9Oz0r+hV/HaSx2W0E2WazR3XSPITST5cVc+qmm1FVXXbjHYIPT2rJEXGbpHkz5L8fVVdY5YY4zhVVS9I8vasnuhJkoOS/EiS/6yqB88aZxxrLmvabpI9AAAAAACwPC7O6Iizf0/yviRnJBnuRDkwyTOTvGjaQavqFkn+OcmNBy9dkFGy5BNJLh289p0ZJUcOnjbO2O8m+blBXSf5XJL3Jzl38NqhSV5dVd87S5A5r2lbSfYAAAAAAMDi+nySP81oh8tNkxza3bfo7rt09526+7iM7p95bJIzB30fVVU/tl6A8V02r01y7Ynq85I8IsmR3X277r55kusleU6SKybafUuSF0y7mKp6WJKfGlT/VZJbdPex3X3H7j4qo+PrPjzR5oAkL62q46aMM7c1zYNkDwAAAAAALKb7Jjmmux/b3X/R3ad39xXDRt395e7+0yT/K6OdMZOeU1Xr5QoeleS2E+UvJ7l7d/95d39j50t3n9fdz8go8TTpJ6vqZustpqoOyuhenEl/lOSh3f2JycrufmuSeyR570T14UmevV6csbmsaV4kewAAAAAA2BaVyq4leir71qU93f3h7h4e0bZW+y8n+eHsfazb9ZOctFqfcQLmGYPqn+3u09aI84okfzFRtTvJs6aY4qOTHDdR/kSSJ662xu7+SkY7cS6ZqP6hqrrlWkHmvKa5kOwBAAAAAID9RHd/NKO7fCbdao0u35nkhhPlzyR58RShnpW9k0oPraprrNPnxwfl53b3xWt1GCdoXj1RdUCS9Y6mm+ea5kKyBwAAAAAA9i+nD8rXXrHVyAMH5RdPs5uou09P8i8TVQdmdOzciqrqmCTfPFF1QZLXrBdn7JRBeTjnobmsaZ4kewAAAAAAYP9y8KB8/hpt7zcov3mGOG8ZlO8/Q5xTu/vCKeOcmuRrE+VbrHOfzrzWNDeSPQAAAAAAsJ+oqkpy50H18Fi3PW2vm+R6E1VfT/L+GcKdOijffo22w9fePW2Q7r4syb9NE2vOa5qb3Ts9AQAAAAAAllNVZZRbWA5LspZHJbnBRPljuWqiZI/hXT6f7O5LZoh12qB806raPU7OrBdr2HeaWCevMd5q9du5prmxswcAAAAAAPYDVfWIJH8wUXVFkp9a476aWwzKn5slXnefk+TiiaqDktx4O2Kt0H443pbEmXFNc2NnDwAAAAAAbMzxG9jtc053n70dk6mqmyc5dqLqwCTXTHKbJA9McsLEa5ckeWx3v3WNIa8zKJ+5gWl9PslNBmN+YoV2R20y1lmD8nDuq9Vv55rmRrIHAAAAAAA25vUb6PPsJM/a4nns8fgkT1inTSf5hyRP6+4PrdP2sEH5wg3MadhnOGaq6pAkB2wy1rpxVqnfljXNm2QPAAAAAADbYldVdi3HPTdJsixreW2S350i0ZNcNYlx8Yqt1nbROmOuVjdrrGnirFS/XWuaK3f2AAAAAADA/uNhSd5VVe+oqpuu0/bgQfmSDcT7+qB8yBRxNhJrmjgrxdquNc2VnT0AAAAAALAxD0xy+ox9ztmOiSRJd/9Mkp/ZUx4fj3atJLdL8r1JfjBXJibunuTfq+rbu/u9qww53PVy0AamdbV1xlyt7qBV6jcTZ6X67VrTXEn2AAAAAADAxpze3R/Z6UmsprsvSnLm+Pm7qnpeRse43X7c5Igkr6uq23T3+SsMccGgvNIOnPUMd70Mx1yt7uDMlkSZJs5K9du1prlyjBsAAAAAAOwHuvuTSb49yecmqo9O8nOrdBkmMQ7dQNhhn6skRsZJqcs3GWvdOKvUb8ua5k2yBwAAAACAbbErtXTPouvuc5M8c1D9yFWanz0oH7OBkDdYZ8w9hsfbzRrr6CnjzHNNcyPZAwAAAAAA+5e/SdIT5RtU1Y1WaPfxQfnYWYJU1XWy9zFplyT51CrNNxVrhfYf2444M65pbiR7AAAAAABgPzK+n+e8QfX1Vmg6TJgcX1UHzRDqVoPy6d192Spth7FOmCHOSrFWS/bMc01zI9kDAAAAAABcOqzo7v9O8t8TVVdLcscZxjxpUP7gGm2Hr504bZCq2p3kLtPEmvOa5kayBwAAAACA7VFJLdGzBFf2JEmq6vAkRw6qv7hK878blL99hlDDtm9Yo+0wzolVdeiUcU5KcvWJ8n9193/NEGu71jQ3kj0AAAAAALB/uV/2Tl2dk+QLq7T920H5x6pq3bRXVR2f5J4TVZcmedNq7bv7c0k+MFF1WJKHrRdn7NGD8uvXaT+XNc2TZA8AAAAAAOwnquqQJM8eVL+xu69Ypcs/Jjlzonxckh+bItSzsndC6a+6+yvr9DllUH5qVR28VoequlWSh09UXZHkJevEmeea5kKyBwAAAAAAFkxVvaCq7jxjnyMz2tVy84nqy5P81mp9uvvrSZ4zqH5hVZ2wRpwfTPLDgxjPnGKKf5rksxPlmyf5rdV23VTVNyX58yQHTVS/ortPWyvInNc0F5I9AAAAAABsi11VS/fsQ74jyb9V1Xuq6klVdfuqOnDYqEZuWVW/mOTjSe4zaPJb3f0f68Q6JclHJsrXTPLOqvrRqto9EevIqvqVJC8b9P/jde7QSZJ09yVJnjqo/okkr6mqm01WVtW9k7wzyZ0mqi9I8kvrxRmby5rmZff6TQAAAAAAgH3UXcZPklxSVWclOT/JJUkOT3LD8c8reWmSn18vQHdfWlUPTfKuJEeOq48c9//9qjo9ySFJbpxkmHD6tyQ/O+1iuvuVVXX3JD85Uf2QJA+uqs9ldL/QjZJce9D1iiQ/1t2fnjLO3NY0D5I9AAAAAACwHA7KKDmxnv/JaAfNH3V3TzNwd390vJvm9RklW/Y4LMntVun2T0ke2t0XTRNjwk8luTjJEyfqKsmx42foaxklev5yliBzXtO2cowbAAAAAAAsnh/IaFfOP2WUvFlPJ/lwkp9LctPu/sNpEz3fGKD7Q0lum+S5Sb68RtNPJHlMku/o7vNniTGOc0V3PynJnqPaVnNJkpcnuU13v2bWOONYc1nTdrOzBwAAAAAAFkx3fzTJR5O8oKp2JblZkptmtPPlmzI6euyrSb6S5DNJ3t/d0ySF1ov71SS/UFXPTHLXJLdJcq0klyf5wjjOencATRvrbUneVlXHJDkxo7UdnNG6PpHkXYu2pu0i2QMAAAAAwLaoJJXa6WlsmX11Jd19RZKPj595xbw0o/tu3jWHWGcm2dDOnRnjzG1NW80xbgAAAAAAAAtMsgcAAAAAAGCBSfYAAAAAAAAsMHf2AAAAAACwLXalsqv21ZtuZrdrn721h/2dnT0AAAAAAAALTLIHAAAAAABggUn2AAAAAAAALDDJHgAAAAAAgAW2e6cnAAAAAADAcqpUdqV2ehpbppZoLSwXO3sAAAAAAAAWmGQPAAAAAADAApPsAQAAAAAAWGDu7AEAAAAAYFtUVaqW556bZVoLy8XOHgAAAAAAgAUm2QMAAAAAALDAJHsAAAAAAAAWmGQPAAAAAADAAtu90xMAAAAAAGA57arKrqqdnsaWWaa1sFzs7AEAAAAAAFhgkj0AAAAAAAALTLIHAAAAAABggbmzBwAAAACAbVKpLNM9N8u0FpaJZA8A+zX/RFt8B+3yuwgAAADs3xzjBgAAAAAAsMAkewAAAAAAABaYY9wAAAAAANgWuyrZVctz/LaTxNlX2dkDAAAAAACwwCR7AAAAAAAAFphkDwAAAAAAwAKT7AEAAAAAAFhgu3d6AgAAAAAALKddVdlVtdPT2DLLtBaWi509AAAAAAAAC0yyBwAAAAAAYIFJ9gAAAAAAACwwd/YAAAAAALAtavxjWSzTWlgudvYAAAAAAAAsMMkeAAAAAACABSbZAwAAAAAAsMAkewAAAAAAABbY7p2eAAAAAAAAy6kq2VU7PYutU0u0FpaLnT0AAAAAAAALTLIHAAAAAABggUn2AAAAAAAALDB39gAAAAAAsC1q/GNZLNNaWC529gAAAAAAACwwyR4AAAAAAIAFJtkDAAAAAACwwCR7AAAAAAAAFtjunZ4AAAAAAADLaVcqu6p2ehpbZleWZy0sFzt7AAAAAAAAFphkDwAAAAAAwAKT7AEAAAAAAFhg7uwBAAAAAGBbVC3XnT21RGthudjZAwAAAAAAsMAkewAAAAAAABaYZA8AAAAAAMACc2cPAAAAAADbopJUlueem+VZCcvGzh4AAAAAAIAFJtkDAAAAAACwwCR7AAAAAAAAFphkDwAAAAAAwALbvdMTAAAAAABgOe2qyq6qnZ7GllmmtbBc7OwBAAAAAABYYJI9AAAAAAAAC0yyBwAAAAAAYIG5swcAAAAAgG1RVakluudmmdbCcrGzBwAAAAAAYIFJ9gAAAAAAACwwyR4AAAAAAIAFJtkDAAAAAACwwHbv9AQAAAAAAFhOlcqu1E5PY8vUEq2F5WJnDwAAAAAAwAKT7AEAAAAAAFhgkj0AAAAAAAALzJ09AAAAAABsi11V2VXLc8/NMq2F5WJnDwAAAAAAwAKT7AEAAAAAAFhgkj0AAAAAAAALTLIHAAAAAABgge3e6QkAAAAAALCcavwsi2VaC8vFzh4AAAAAAIAFJtkDAAAAAACwwCR7AAAAAAAAFpg7ewAAAAAA2CZu7YF5sLMHAAAAAABggUn2AAAAAAAALDDJHgAAAAAAgAXmzh4AAAAAALZFVVK1PPfcLNFSWDJ29gAAAAAAACwwyR4AAAAAAIAFJtkDAAAAAACwwCR7AAAAAAAAFtjunZ4AAAAAAADLqcbPsthX11JVleS4JLdNckySI5J8PcmXk3wiyb9398VbHPPwJCcluXmSb0pyUZIzkry7uz+/xbFuneSOSa6f5IAkX0ryn0ne092XbWGcua1pq0n2AAAAAADAgqmqayZ5UJLvSnLvJNdeo/mlVfV3SX67u/9lk3FvnOSXkzwsyUErNOmq+pckz+zud2wiTiX5sSQ/n1HyZSVfqqo/TPK87r5wE7Hmsqbt5Bg3AAAAAABYIFX1+0n+O8mfZZSgWCvRkyQHZpQYentVvbSqvmmDcR+W0Y6aH87KSZFktAHq5HGs542TNrPGOSLJPyY5JasnepLkWkmekeTD490/M5vXmrabZA8AAAAAACyWu2blxMTlSc5M8r4kH07ylRXa/GiSt1TVYbMErKqHJnllkqsPXjonyfvHcXuyS0a7cn5zxjiHZJTo+fbBS5ck+a8k/5FkuIvnJkneVlU3nTHWXNY0D5I9AAAAAABsk1qqH/vorT3nJ/mDJPdLcs3uvmF336m7b5fRzpd7JXnnoM9dkrxk2gBVdXySF2fvnMKHkty7u6/T3Xfs7hsmuVWSvx50/5mq+r4Z1vOb4/ntcUWSX0lyve6+RXf/ryRHZnTE25cn2h2V5DVVdcA+uKZtJ9kDAAAAAACL5zNJfjzJDbr7f3f3m7r7q5MNuvvy7n57RgmfPxn0f3BV3WvKWL+S5NCJ8r8nuUd3v20Q7+NJHrJCrBdU1e71glTVLZM8ZlD9w939S939jcROd1/S3S9JcveMkl173CGjnUvTmMua5kWyBwAAAAAAFsszk9yiu0/p7ovWa9zdlyd5fJL3Dl768fX6ju/CefhE1SVJHtHd/7NKrE7yhCSfmKg+PqOdOOt5dpLJnTkv6+5Xrta4uz+S5GcH1c+sqgPXCjLnNc2FZA8AAAAAACyQ7v677r5kxj6XJ3nBoPo7p+j6qOydS3hVd390nVgXJ3neoHrNxFJVXTPJ5NFoneRZU8zvxUnOmCjfKMl91ukzlzXNk2QPAAAAAADsH4Z391yrqq6+Tp/vGZRPmTLWq5NcOFG+c1XdYI3290syeSza27v7U+sF6e4rMkr4THrQOt3mtaa5kewBAAAAAGBbVJKqJXp2+hd08768Qt01VmtcVbdIctOJqguTvHuaQN09bFsZJXRWM3ztzdPEGXvLoHz/1RrOeU1zI9kDAAAAAAD7h6NXqPvSGu1vPyj/W3dfNkO8U9cZb63XpkrAjL0vydcnyjeoqqOmjLOda5obyR4AAAAAANg/3H1QPmOdu39uNSifNmO8YfvheEmSqjowe++2mSlWd389yenTxFqhflvWNG+SPQAAAAAAsH941KD8pnXa32JQ/tyM8Ybth+PtcZPsfV/PRd197jbFmtea5mr3+k0AAAAAAGAjKktx0803LO5aquq+Se4xqH7JOt2uMyifOWPYswbl1Y5WG8YZ9ttIrOGYq9Vv15rmSrIHAAAAAAA25viqmRNA53T32dsxmdVU1ZFJ/nhQ/bru/rd1uh42KF84Y+hh+wOr6mrjY9e2Ms5KfYZjblWsadc0V5I9AAAAAACwMa/fQJ9nJ3nWFs9jVVW1K8lfJDlmovorSX56iu7DxMjFM4a/aJUx10v2zBpnpVjTJnu2a01z5c4eAAAAAABYXr+e5LsHdY/r7mnuqjl4UL5kxtgrJUAO2YY4K8VaKc5WxJp2TXMl2QMAAAAAAEuoqn46yZMG1S/o7ldPOcRw18tBM07halOMuRVxVoq12o6dea1prhzjBgAAAADAtqjxj2WxwloemOT0GYc5Z2tms7aq+sEkvz2ofkmSp84wzAWD8nBXzHpW2vEyHHMr4qwUa6U4WxFr2jXNlWQPAAAAAABszOnd/ZGdnsRQVd0/yUuTvbJTf53kx7u7ZxhqmMQ4dMapDNtf1t0r7YLZbJyV+kyb7NmuNc2VY9wAAAAAAGBJVNW9krw2e2/2eEuSH+juy2cc7uxB+ZgZ+x89KK+2q2kYZ9hvI7GGY65Wv11rmivJHgAAAAAAWAJVddckf5u9jyZ7d5Lv7e5LNjDkxwflY2fsP2z/sVXafSrJZRPlQ6rqqG2KNa81zZVkDwAAAAAA26KW8NlXVdX/SvL3SQ6bqP5Akvt294UbHHaYyDhhxv63Wme8JEl3X5qr3n00dayqulqSm0wTa4X6bVnTvEn2AAAAAADAAquqW2R0VNs1J6o/muQ7u/srmxj6g4Pynatq90oNV3HSOuOt9dqJM8S5Y5KrTZS/0N2rHeM2jLOda5obyR4AAAAAAFhQVXWjJP+U5DoT1Z9O8u3dvan7ZLr7Y9l7x82hmTIJU1WHJvmWyeGSvHGNLsPXvn2aOKu0fcNqDee8prmR7AEAAAAAgAVUVddP8tYkx0xUn5Xk27r7rC0K87eD8qOn7Pfw7H2k3Hu7+/NrtH9T9r635+SqGh7NdhVVVUkeOah+/Trd5rWmuZHsAQAAAABgW1TV0j37iqo6MqOj246fqD4nox09n97CUH+W0Q6WPb6/qob31gzndnCSpw6qT1mrT3efl+R1k8MkedYU83tUkuMmymdktNNpLXNZ0zxJ9gAAAAAAwAKpqsOT/EOSW09Un5/kO7r7o1sZq7v/M8lrJqoOSvLSqvqmVeZWSX47yc0mqj+VUYJlPc9McsVE+Ueq6gdWa1xVJyR54aD6V7r7krWCzHlNczHLpUMsgaq6TZLbJTkqo7MIz0/ysSTv7u6LtjDOoUnulNEf/iMyerN8JckXMtra9tktjHVwRmcq3jKjC8guSXJmkvd096e2Kg4AAAAAwD7ib5PceVD3m0muXVX3mXGs93X3l9dp84wkD0hy9XH5zkneUVU/091v39Ooqm6e5LlJvm/Q/6ndfel6E+nu06rqRUkeO1H9F+NdN7+1Z55VdWCSH8pozUdMtP1wkpeuF2eea5oXyZ4dVlVHJ7lLkruOf75TksMnmpzR3cdtMsZhSZ6Q5PFJbrBKs4uq6lVJnrWZRExV3SvJzyS5X5ID1mj3iSR/kuT3N5pkqqqjMsr0PjKjxNVKbd6XUSZ3vTMaAQAAAAAWxckr1P3yBse6V5K3r9Wguz9ZVY9O8oqMjldLRpsK3lZV5yT5bJLrZHR30PCsu9/r7tfOMJ8nJvnmjD4rT0YnlP1ikp+vqk8n+XqSm2Tvu3OS5NwkD+3uyzKFOa9p20n27ICqOinJkzNK8KyWfNmqWHfO6JzD9eIckuTHkjysqh7V3a9Zp/0wzsFJ/jBXvQhrNTdL8utJHldVP9jd/z5jvJOTvDbJtddpesckr6uqP0/ymPW27wEAAAAAcFXd/arxcWanZPR58h5HjZ+VvDDJU2aM87Wq+s6MPv+998RLByW5xSrdPpPke7r7v2aMNZc1zYM7e3bGnZN8b7Y/0XOvJO9YIc7Xk5yW5H1Jvjh47dAkr1zrHMQV4uxO8tdZOdFzfpIPJvn3JJ9b4fWbJnlLVd1xhnjfmuRNuWqi5/wkH8jojX354LUfzWhd+84NagAAAAAAC6S7X5nkNhnthlnrCLN3JDm5u3+uu3sDcc5L8u0ZHef2yTWanpfk15Lctrv/Y9Y441hzWdN2s7Nn33NBrrr9bGZVdUySVyc5eKL6vCS/kOTl3X3BRNs7JXlekm8bV+1KckpVfai7T5si3M8k+e5B3anjWO+c/INfVccmeVKS/5Mrk43XSPLnVXX79c44rKprjtc1mWU9I6Nj6v52T6zx+p+R5HET7b4voy2AvznFmgAAAAAA9kndvWPf1D6+J/2Hquonk3xrRqc4HZ7k4oyOPju1u8/agjhXJPnTJH9aVbfN6Gi362d0fciXkvxnRve2b/renHmtaTtJ9uysr2a0u+bfk/zb+OcbJ3nbFoz9nOy9zezsJPfs7o8NG3b3e6vqO5L8WZJHjKsPSfKCJPdfK8j4IqxfGFS/McmDunu4uybj+4B+ZnyXzp9PvHRCkockeeVa8ZL8XPbeqfTpJN/a3Z8fxDkzyU9U1Wcz+rXY45eq6sVTXDgGAAAAAMAquvt/MjqBaR6x/iPJhnbuzBhnbmvaao5x2xlvSHLrJEd09726+ynd/ZfdfcZWDF5VN0zyg4Pqx62U6NljnCX9iSSTZxreb3y/0FrunuSaE+VLx7GukugZxHtZkn8cVD9grT5VdVRGO4ImPWaY6Bl4bkbb6/a4RpKfXSsOAAAAAAAsEsmeHdDdp3f3aeMEy3a4X/betfWR7n7dFPO6OKPLpSY9dp1uwwux3rNO8mXSXw3KN12n/fdn7yPu3tHdb12rw/hYt2cPqh/l7h4AAAAA2H61hD9gXyTZs5zuOSi/YYa+w7YPqKq1jvs7clD+3AyxPjsoH7FO+wcOyqdMGedtGR33tsf1ktxtyr4AAAAAALBPk+xZTscOyh+atmN3/3dG9/vscc2MjmpbzVcG5UOmjbVC23NXa1hVhyW5x6D6zdMEGe/u+adB9Zp3EQEAAAAAwKKQ7FlO1xqUz5ux/5cG5Tus0faDg/KdZjgi7S6D8r+t0fbWSQ6cKH96nJia1qmD8u1n6AsAAAAAAPssyZ7lNLwL6IAZ+x84KN9qjbb/L3sfx3ZMRnfrrKmqjkzymImqTvLiNboM53DaejHWab/WmgAAAAAAYGGsdRcLi2u4k+c6M/Y/alBeNTHS3ZdX1dOSvHyi+k+q6svd/Q8r9amqo5P8VZJrT1T/fnevddzcLQblWe4GWqn9jarq4O6+eMZxAAAAAIBpVTL9QUALYImWwnKR7FlOn0py0kT5zkleOk3HqrpZkmsMqq+7Vp/ufkVV/a8kPz+uOizJ31fVPyd5Y5LTk1ya5HpJvjXJw5McOjHEy5L8zDpTGyaszlyn/dAXk1yWK//M78rouLuzZhznKqrqOrlqgmw9x282LgAAAAAAJJI9y+qdSX5kovzgqnpid186Rd8fWKHusPU6dfdTq+r9SV6Q5Ebj6nuPn9V8OMmvdvdrp5jXcA4XTtFncn5dVRclOXyNMTfq8UmeuUVjAQAAAADATNzZs5zelNFOmj2ul+QJ63Wqqusl+ekVXpoqKdLdr0lyhyQvmaL5fyZ5YZLXTzP2CnPYyPFrF60zJgAAAAAALBzJniXU3WcledWg+jlV9aDV+lTVNZO8LqOjzYYOWS9mVe2uqqdndITcI6eY5m2S/HmST1XV/adof/CgfMkUfYa+Piivuy4AAAAAYONqCR/YFznGbXk9Lcl35cq7ZA5K8tdV9bIkp2R0hNpFSY5Jcr8kT01y/XHb85McMTHWBWsFqqpvymg30eQ9QWcm+Z0kb07y6YwSLddN8i1JHpcrj3c7Oskbqur/dPf/XSPMcCfPQWvNaRVXW2fMjfqDJNMcRTfp+Ey/qwkAAAAAAFYl2bOkuvusqnp4kjckOXRcXUl+dPys5qwkz07yJxN156/WuKoqyV9l70TP3yX5ge7+6qD558bPa6rqJzJKkuxJhv9uVZ3W3f+8Sqhhwmm402caw508ayaxptXdZyc5e5Y+o182AAAAAADYPMe4LbHufluSuyc5fcouH8hox80wSXP+Gn1+NMl9JsofTfLQFRI9w7n9UZJfnaiqJL+3RpdhYubQFVutYpyU2pZkDwAAAAAA7CTJniXX3R9IcquMjk57V5LLVmj2oSRPSHK37v6vXPXenk+uEeIJg/KvdPdFU07v+dk7kXRCVZ20Stvhzpljpoyxx3Wz9062K5KcO+MYAAAAAMAMagl/wL7IMW77ge6+NKNj2f6kqg5NcsMk105yYZLPdfcw6XHCoPzelcatqiOT3H6i6oqM7u6Zdl4XVtXbknzvRPU9kpy6QvOPD8rHThtnlfZndPdW3dkDAAAAAAA7RrJnP9PdFyb52DrNbjsor5jsSXJcslcq+7zu/sqMU/r0oHz0Ku2Gcx4mpNZzq3XGAwAAAACAheQYN/Yy3q1zt4mqryR59yrNrzYor3RE3HouHZQPWKXdRwZtj6uq688QZ3g83Adn6AsAAAAAAPssyR6GHpzkwInyK7v7a6u0/dKgfK2qOmjGeMOdPOes1Ki7v5rkHYPqb58mQFVVkvsMqt8w1ewAAAAAAGAfJ9nDN1TVwUmePqj+0zW6nJW9d/McmOTuM8TbleTkQfXpa3T520H50VOGuleSG0+Uv5jkPVP2BQAAAACAfZpkD5Oel+RGE+VXdPf7V2s8vv/nXwfVPztDvB9Jcsyg7q1rtH9VkgsnyveoqnuvFWC8q+eZg+oXd/cVU88SAAAAAAD2YZI9S6yqbjlluwOq6teSPGGi+stJnjhF978YlL+rqn5xiph3SfJ7g+p/7u4zV+vT3Wcn+b+D6hdV1Q3WCPW0JPeYKH8lya+vNz8AAAAAAFgUu3d6AvurqjopySErvHS7QfngqhreN7PH57v7tDXCvLOqPpLklUn+vrs/O5jDYUnum+SpSe4w8dIVSR49Tq6s58+SPCnJzSfqfrmq7pzkOUn+rbt7IubRSR6X5ClJrjaI+QtTxHtBkkckud64fOMk766qn07yhj2xquqYJM8Yx5r0nO4+b4o4AAAAAACwECR7ds7Ls/eRaau5bpK3rPLaS5M8co2+leSe4ydVdV6SM5NcnOQ6SW6Q5KBBn8uTPKa7/2aKuaW7L62qByV5V5IjJ156wPj5clV9Nskl47Ucu8pQT+jude/R6e7zqurhSf4xycHj6hsleX2S86vq00mOGMc5YND99UleOMWyAAAAAIAtUFUZ3bSwHJZpLSwXyZ79y5HZOyEzdGaSR3T3P88yaHd/tKrukeQVSf7X4OVrjp/VfDXJT3f3S2aI946qul+S12bv9RyRvXcoTXpFkkdN7jICAAAAAIBl4M6e5fZrSf5fksvWaffJjI5Vu8WsiZ49uvsjSe6c5MeTvG+KLl9M8vwkJ8yS6JmI989JTkjyh0m+tkbTDyR5cHf/UHd/fdY4AAAAAACwr7OzZ4d093FziPGbSX6zqq6e0Y6Xm2Z0fNvBGSVIzkzy/u7+xBbFuyTJKUlOqaojM0r+HJvRjpsDk3wlyTlJPrAVMbv7i0keX1VPTnJikluNY12S5Kwk7+nuT242DgAAAAAA7Mske/YD3f21JKeOn3nFPC+je3XmEeuiJG8dPwAAAAAAsF+R7AEAAAAAYFvU+MeyWKa1sFzc2QMAAAAAALDAJHsAAAAAAAAWmGQPAAAAAADAAnNnDwAAAAAA26LGz7JYprWwXOzsAQAAAAAAWGCSPQAAAAAAAAtMsgcAAAAAAGCBSfYAAAAAAAAssN07PQEAAAAAAJZU1ehZFsu0FpaKnT0AAAAAAAALTLIHAAAAAABggUn2AAAAAAAALDB39gAAAAAAsG3ccgPbz84eAAAAAACABSbZAwAAAAAAsMAkewAAAAAAABaYO3sAAAAAANgWNf6xLJZpLSwXO3sAAAAAAAAWmGQPAAAAAADAApPsAQAAAAAAWGCSPQAAAAAAAAts905PAAAAAACA5VRJKrXT09gyy7MSlo2dPQAAAAAAAAtMsgcAAAAAAGCBSfYAAAAAAAAsMHf2AAAAAACwLapGz7JYprWwXOzsAQAAAAAAWGCSPQAAAAAAAAtMsgcAAAAAAGCBSfYAAAAAAAAssN07PQEAAAAAAJZVjZ9lsUxrYZnY2QMAAAAAALDAJHsAAAAAAAAWmGQPAAAAAADAAnNnDwAAAAAA28KNPTAfdvYAAAAAAAAsMMkeAAAAAACABSbZAwAAAAAAsMAkewAAAAAAABbY7p2eAAAAAAAAy6lSqaqdnsaWqSzPWlgudvYAAAAAAAAsMMkeAAAAAACABSbZAwAAAAAAsMDc2QMAAAAAwDapJbvnZpnWwjKxswcAAAAAAGCBSfYAAAAAAAAsMMkeAAAAAACABebOHgAAAAAAtkVluW65Waa1sFzs7AEAAAAAAFhgkj0AAAAAAAALTLIHAAAAAABggUn2AAAAAAAALLDdOz0BAAAAAACWVY2fZbFMa2GZ2NkDAAAAAACwwCR7AAAAAAAAFphkDwAAAAAAwAJzZw8AAAAAANuiavQsi2VaC8vFzh4AAAAAAIAFJtkDAAAAAACwwCR7AAAAAAAAFphkDwAAAAAAwALbvdMTAAAAAABgOdX4x7JYprWwXOzsAQAAAAAAWGCSPQAAAAAAAAvMMW4AAAAAAMBMqurgJCcmuWWSaya5JMmZSd7T3Z/a4ljHJ7lLkmOSHJTky0k+luTd3X3xFsaZ25q2mmQPAAAAAAAsuKo6OqOEyF3HP98pyeETTc7o7uO2IM5RSZ6Z5JFJDl2lzfuS/Ep3v36TsR6U5BeTfPMqTS6oqpckeXZ3n7uJOHNb03aR7AEAAAAAgAVUVScleXJGCZ4bzCHeyUlem+Ta6zS9Y5LXVdWfJ3lMd18yY5yrJTklyQ+t0/SwJD+V5OFV9ZDufscsccaxTs4c1rTd3NkDAAAAAACL6c5JvjfzSfR8a5I35apJkfOTfCDJZ5JcPnjtR5O8sqpqhji7krw6V030XJ7k00k+mOQrg9eOSvL3VfUt08YZx5rLmuZBsgcAAAAAAJbPBVs1UFVdM6MEzCET1WckeVCSI7v7m7v7xkmOS/LHg+7fl+SJM4T7uSQPHNT9UZJju/sm3X2HJEeOx/3sRJurJ3lNVV1jmiBzXtO2k+wBAAAAAIDF9tUkb0/y60kemlGC4gFbOP7PZe/dQ59OcmJ3v767e09ld5/Z3T+R5OmD/r80Tq6sqaqutULfp3X3T3b35yfiXNHdf5PkxIx23+xxTJInTbOgzGlN8yLZAwAAAADAtqglfPYxb0hy6yRHdPe9uvsp3f2X3X3GVgWoqqOS/J9B9WMmky8reG6SyftzrpHkZ6cI95Qkh0+U35Hk+as17u6zkvz4oPqJ46TRqua8prmQ7AEAAAAAgAXU3ad392ndfcU2hvn+JIdNlN/R3W9dZ16d5NmD6ketdc/N+K6eHxtUP2tyl80qsd6a5J0TVYcnedhafTKnNc2TZA8AAAAAALCa4f05p0zZ720ZHY22x/WS3G2N9icmOWqi/KmMjqabxnBOD1qn/bzWNDeSPQAAAAAAwFVU1WFJ7jGofvM0fcc7Yf5pUH3/Nbrcb1B+y3q7eibbDsonV9WhKzWc85rmRrIHAAAAAIBtUbV8z37m1kkOnCh/urv/e4b+pw7Kt1+j7fC1d08bZHzXzmcmqg5KcsIqzee5prmR7AEAAAAAAFZyq0H5tBn7D9sPx9uJWPNc09xI9gAAAAAAACu5xaD8uRn7D9vfqKoOHjaqqkOSHLvFsYZzX61+W9Y0b7t3egIAAAAAALCgjq/Zz3Y7p7vP3o7JbIPrDMpnztj/i0kuy5W5iF1JrpXkrEG7ayeZ/IW8NMmsv0bDMYdzX61+u9Y0V5I9AAAAAACwMa/fQJ9nJ3nWFs9juxw2KF84S+fu7qq6KMnha4y5Ut3XurtnibXC3FaKs1L9dq1prhzjBgAAAAAArGSYxLh4A2NctM6Y84wz71hzI9kDAAAAAACsZHgXzSUbGOPrg/IhOxhn3rHmxjFuAAAAAACwMQ9McvqMfc7Zjolsk+Gul4M2MMbV1hlznnHmHWtuJHsAAAAAAGBjTu/uj+z0JLbRBYPycFfMNIa7XoZjzjPOvGPNjWQPAAAAAADbplI7PQU2bpjEOHSWzlVV2Viy5+pVVd3dM4Qbzm3aZM92rWmu3NkDAAAAAACs5OxB+ZgZ+183e286uSLJuSu0OzfJZGLnwCTXmTHW0YPycO6r1W/XmuZKsgcAAAAAAFjJxwflY2fsP2x/Rndf5X6b7r4oyWe3ONbHVmk3lzXNm2QPAAAAAACwkmHC5IQZ+99qnfF2ItY81zQ3kj0AAAAAAMBKPpLk0onycVV1/Rn6nzQof3CNtsPXTpw2yHhOx01UXZrktFWaz3NNcyPZAwAAAAAAXEV3fzXJOwbV3z5N36qqJPcZVL9hjS5vHJTvMx5jGt8xKL+tuy9YqeGc1zQ3kj0AAAAAAMBq/nZQfvSU/e6V5MYT5S8mec8a7d+d5NyJ8k2SnDxlrOGcXr9O+3mtaW4kewAAAAAAgNW8KsmFE+V7VNW91+ow3gHzzEH1i7v7itX6jF97yaD6mevt7qmqb0ty94mqryZ5zVp9Mqc1zZNkDwAAAAAAsKLuPjvJ/x1Uv6iqbrBGt6clucdE+StJfn2KcM9PMnn82j2T/Pxqjavq6CQvGlT/Tnefu1L7Pea8prnYvdMTAAAAAABgOVWNnmWxL66lqk5KcsgKL91uUD64qob3zezx+e4+bY0wL0jyiCTXG5dvnOTdVfXTSd7Q3T2eyzFJnpHkcYP+z+nu89YYP0nS3edW1a8l+bWJ6udW1bFJfrW7Pz+OsyvJ9yT5nSTHTq4jyW+sF2eea5oXyR4AAAAAAFhcL09yoynaXTfJW1Z57aVJHrlax+4+r6oenuQfkxw8rr5RRnfjnF9Vn05yREaJlwMG3V+f5IVTzG+P5yc5Mcn9J+p+Msljq+qMjHbU3Hgcb9JFSR7W3edPE2TOa9p2jnEDAAAAAADW1N3vSHK/JMPdLEckuUNGCZhhUuQVSR6+Z5fMlHGuSPLQjO7VmXRAkpuMYx0xeO1LSe7b3adOG2ccay5rmgfJHgAAAAAAYF3d/c9JTkjyh0m+tkbTDyR5cHf/UHd/fQNxLu7uH0jykCQfXKPphUn+IMkJ3f32WeOMY81lTdvNMW4AAAAAALCguvu4Ocf7YpLHV9WTMzpu7VYZ7YS5JMlZSd7T3Z/colh/leSvquqmSe6a5OgkByU5P8lHk5za3RdvQZy5rWm7SPYAAAAAAAAz6e6Lkrx1/Gx3rE8m2fZkyzzXtNUc4wYAAAAAALDAJHsAAAAAAAAWmGPcAAAAAADYFjV+lsUyrYXlYmcPAAAAAADAApPsAQAAAAAAWGCSPQAAAAAAAAtMsgcAAAAAAGCB7d7pCQAAAAAAsJxq/CyLZVoLy8XOHgAAAAAAgAUm2QMAAAAAALDAJHsAAAAAAAAWmGQPAAAAAADAApPsAQAAAAAAWGCSPQAAAAAAAAtMsgcAAAAAAGCBSfYAAAAAAAAssN07PQEAAAAAAJZT1ehZFsu0FpaLnT0AAAAAAAALTLIHAAAAAABggUn2AAAAAAAALDB39gAAAAAAsE1q/CyLZVoLy8TOHgAAAAAAgAUm2QMAAAAAALDAJHsAAAAAAAAWmGQPAAAAAADAAtu90xMAAAAAAGA51fhZFsu0FpaLnT0AAAAAAAALTLIHAAAAAABggUn2AAAAAAAALDDJHgAAAAAAgAUm2QMAAAAAALDAJHsAAAAAAAAWmGQPAAAAAADAAtu90xMAAAAAAGB51U5PAPYDdvYAAAAAAAAsMMkeAAAAAACABSbZAwAAAAAAsMAkewAAAAAAABbY7p2eAAAAAAAAy6lq9CyLZVoLy8XOHgAAAAAAgAUm2QMAAAAAALDAJHsAAAAAAAAWmGQPAAAAAADAApPsAQAAAAAAWGCSPQAAAAAAAAtMsgcAAAAAAGCBSfYAAAAAAAAssN07PQEAAAAAAJZTjZ9lsUxrYbnY2QMAAAAAALDAJHsAAAAAAAAWmGQPAAAAAADAApPsAQAAAAAAWGCSPQAAAAAAAAtMsgcAAAAAAGCBSfYAAAAAAAAsMMkeAAAAAACABbZ7pycAAAAAAMByqlSqaqensWUqy7MWloudPQAAAAAAAAtMsgcAAAAAAGCBSfYAAAAAAAAsMMkeAAAAAACABSbZAwAAAAAAsMAkewAAAAAAABaYZA8AAAAAAMAC273TEwAAAAAAYDnV+FkWy7QWloudPQAAAAAAAAtMsgcAAAAAAGCBSfYAAAAAAAAsMMkeAAAAAACABSbZAwAAAAAAsMAkewAAAAAAABaYZA8AAAAAAMACk+wBAAAAAABYYLt3egIAAAAAACynGj/LYpnWwnKxswcAAAAAAGCBSfYAAAAAAAAsMMkeAAAAAACABSbZAwAAAAAAsMB27/QEAAAAAABYUpVU7fQkttAyrYWlYmcPAAAAAADAApPsAQAAAAAAWGCSPQAAAAAAAAtMsgcAAAAAAGCBSfYAAAAAAAAsMMkeAAAAAACABSbZAwAAAAAAsMAkewAAAAAAABbY7p2eAAAAAAAAy6nGz7JYprWwXOzsAQAAAAAAWGCSPQAAAAAAAAtMsgcAAAAAAGCBubMHAAAAAIBt4tYemAc7ewAAAAAAABaYnT37maq6TZLbJTkqyaFJzk/ysSTv7u6LdnBqG1ZV10pyUpLjM1rThUlOT3Jqd39pJ+cGAAAAADBvVXV8krskOSbJQUm+nCs/B754J+fG9pDs2WFVdXRGb7q7jn++U5LDJ5qc0d3HbTLGYUmekOTxSW6wSrOLqupVSZ7V3Z+dYey3J7nnZuY34V+6++QZYt8uyS8nuX9W3qV2eVX9XZJf7O4Pb80UAQAAAAD2TVX1oCS/mOSbV2lyQVW9JMmzu/vcec2L7SfZswOq6qQkT84owbNa8mWrYt05yeumiHNIkh9L8rCqelR3v2Y757WKS6ZtWFVPSPLCrP1n+IAk35PkvlX1pO7+vU3ODwAAAACYQdXoWRb76lqq6mpJTknyQ+s0PSzJTyV5eFU9pLvfse2TYy7c2bMz7pzke7P9iZ57JXnHCnG+nuS0JO9L8sXBa4cmeWVV/cB2zm0Vb5ymUVU9Kclv56qJni9ktKYvDOp3J/ndqvrpzU4QAAAAAGBfUlW7krw6V030XJ7k00k+mOQrg9eOSvL3VfUt2z5B5sLOnn3PBRllVzelqo7J6A1+8ET1eUl+IcnLu/uCibZ3SvK8JN82rtqV5JSq+lB3n7ZOqCcnueYGpnj3JL80Ub40ySvW61RVJyZ5waD67Ume3N3vn2h3p4x2/kweMfcbVfWv3f1vG5gvAAAAAMC+6OeSPHBQ90dJfqW7P598IyH0wIy+if7YcZurJ3lNVd2mu4fJIBaMZM/O+mpGO1H+Pcm/jX++cZK3bcHYz8koO7vH2Unu2d0fGzbs7vdW1Xck+bMkjxhXH5JRUuX+awXp7vdtZHJV9ahB1RunPCPy1zM6nm2PNyR5SHfvdQTcxJr+Osn9xtW7x/236o4hAAAAAIAdU1XXSvL0QfXTuvt5kxXdfUWSv6mqf0vyriTHjV86JsmTkjxzm6fKNnOM2854Q5JbJzmiu+/V3U/p7r/s7jO2YvCqumGSHxxUP26lRM8e4zf7TyT5r4nq+43vF9pSVXWNJA8aVL9kin7fneTEiaovJXn0MNGzx7j+UeN2e9yjqr59lvkCAAAAAOyjnpLk8InyO5I8f7XG3X1Wkh8fVD9xnDRigUn27IDuPr27TxsnWLbD/bL3rq2PdPfrppjXxRkdfTbpsVs4rz0entHOoT3OTvKmKfoN/yP0+919zloduvvsJH+wzjgAAAAAAAtlfDTbjw2qn9XdvVa/7n5rkndOVB2e5GFbPD3mTLJnOQ2PKXvDDH2HbR9QVVt93N8jBuWXd/dla3Woqqsl+c5B9Z9NGW/Y7rur6qAp+wIAAAAA7ItOzN5XeXwqo/vNp3HKoPygLZgPO0iyZzkdOyh/aNqO3f3fGe202eOaSe6+FZNKkqq6WfY+ii2Z4gi3JCcnOXSi/PFpj73r7s8k+cRE1eFxbw8AAAAAsNjuNyi/Zb1dPZNtB+WTq+rQFVuyECR7ltPwfMXzZuz/pUH5DpuYy9BwV88HuvvDU/S7/aD87hnjnrrOeAAAAAAAi+T2g/LUn5l29+eTfGai6qAkJ2x+SuwUyZ7lNLwL6IAZ+x84KN9qE3P5hvEZkj8yqH7JlN2HczhtxvDD9luyJgAAAABgdbWEzz7EZ6Z8g2TPchru5LnOjP2PGpS36k1+7+x9xNwlSV4+Zd9bDMqfmzH2sP1wPAAAAACAhVBVh+Sq13n4zHQ/tnunJ8C2+FSSkybKd07y0mk6ju/Uucag+rpbNK/hEW5v7O7hkXGrGSaszpwx9lnrjLdhVXWdXDVBtp5bThZO/+Qnt2o6AAAAAGzACp/PHLQT81g2p5++XJ97rbCe46tm3u9zTnefvX6zNV07e280ujR738U+jW37zJT5k+xZTu/M3selPbiqntjdl07R9wdWqDtssxOqqsOTfN+g+iUzDDGcw4UzTmHYftNrmvD4JM/czAAPe/DwlwYAAACAHXbDJB/Y6Uksuod939J/7vX6DfR5dpJnbTLu8PPNr3V3zzjGdn5mypw5xm05vSmjTO4e10vyhPU6VdX1kvz0Ci9txZv8YUmuPlH+YpK/n6H/cA4Xzxj/onXGAwAAAIBJw9NvYF+y2c9LE5+ZLhXJniXU3WcledWg+jlV9aDV+lTVNZO8Lsm1Vnj5kC2Y1vAIt5d392Uz9D94UL5kxvhfH5S3Yk0AAAAALK9v2ukJwBo2+3lp4jPTpeIYt+X1tCTflSvvkjkoyV9X1cuSnJLkwxllbo9Jcr8kT01y/XHb85McMTHWBZuZSFXdJMm3DqpfMuMwF2fvnUGznpl6tRXG2yp/kOS1M/a5bZJXTpQfkuRjWzYjYBrHZ++t1g9McvoOzQX2Z96LsPO8D2Hf4L0IO++WSf5yovzenZrIgjs9yW12ehJzcGSSayb5XGZPtJyzBfGHn29u5I6p7fzMlDmT7FlS3X1WVT08yRuSHDquriQ/On5Wc1ZGZ0b+yUTd+ZucziOy92Vh7+/u/5hxjAuyd7JnmLlezzArvakE1qTxZWozXX62wqVtH+vuj2zVnID1rfA+PN37EObPexF2nvch7Bu8F2HnrfA+3LLPj/Yn3X1xEv/92n7DP5+zfl6abONnpsyfY9yWWHe/LcndM/13An0gyb2TfHVQf/5G51CjvyWHyaWXbGCo4X9oDl2x1eqG7f2HCwAAAABYVMPPN69eK2Qs1+Ez0yUi2bPkuvsDSW6V5HFJ3pVkpXtyPpTkCUnu1t3/lave2/PJTUzh5CTHTZQvSfKKDYwz3DlzzIz9j15nPAAAAACARXFukp4oH5jkOjOO4TPTJeIYt/1Ad1+a0bFsf1JVhya5YZJrJ7kwyee6+9xBlxMG5c2cT/qIQfmN3f2lDYzz8SR3mygfO2P/YXv34wAAAAAAC6m7L6qqzya50UT1sUm+OMMwPjNdIpI9+5nuvjDrv2lvOyhvKNlTVYclecig+sUbGStXnfMwIbWeW60zHgAAAADAIvlY9k72nJDk32fo7zPTJeIYN/ZSVUdm7x00X0ny7g0O95Dsfe7jfyf5hw2O9cFB+cQZ+5+0zngAAAAAAIvkg4Py1J+ZVtX1s/f1G5cmOW3zU2KnSPYw9OCMznfc45Xd/bUNjjU8wu3l3b3SnUHTeHtGx87tcfOqutEqbfdSVccludlE1VfH4wEAAAAALKo3Dsr3qaqasu93DMpv6+4LtmBO7BDJHr6hqg5O8vRB9Z9ucKzjktxzUP2SjYyVJN19cZI3D6ofNWX3Ybt/6O5LNjoXAAAAAIB9wLuTTN7HfpMkJ0/Z99GD8uu3YkLsHMkeJj0ve5/x+Irufv8Gx/rRJJNZ5Pd1939ueGYjpwzK/7uqjlqrQ1VdJ8nj1xkHAAAAAGChdPcVueo32D9zvd09VfVtSe4+UfXVJK/Z2tkxb5I9S6yqbjlluwOq6teSPGGi+stJnrjBuJVRsmfSSzYy1qTu/rsk/zpRda0kp1TVgSu1r6qDMkrsXGui+p3d/Y+bnQsAAAAAwD7g+Ukmj1+7Z5KfX61xVR2d5EWD6t/p7nNXas/i2L3TE9hfVdVJSQ5Z4aXbDcoHV9V9Vhnm89291qVZ76yqjyR5ZZK/7+7PDuZwWJL7JnlqkjtMvHRFkkd399lrrWENd09y/ET5kiSv2OBYQz+X5F9yZaLyAUneXFVPntyFVFV3TPIb2fsoucuTPGWL5gEAAAAAsKO6+9zxN/L/2kT1c6vq2CS/2t2fT5Kq2pXke5L8TpJjJ9p+PqPPUVlwkj075+XZ+8i01Vw3yVtWee2lSR65Rt/KKNlxzySpqvOSnJnk4iTXSXKDJAcN+lye5DHd/TdTzG01jxiU39Dd521ivG/o7ndV1dMyyljvcXKS91XV55N8IaN1XX+F7k/p7n9doR4AAAAAYFE9P8mJSe4/UfeTSR5bVWck+UqSGyc5YtDvoiQP6+7z5zBHtplkz/7lyPGzmjOTPKK7/3mjAarq6kkeOqh+yUbHW0l3v6CqLs/oP2IHTLx0g/EzdHmSn+3u397KeWzSOUmePSgD8+V9CPsG70XYed6HsG/wXoSd533IQuruK6rqoUlenOT7J146IMlNVun2pSQP6e5Tt3t+zEd1907PYb9UVZ/JdDt71vLS7n7kGjGelOQhSe6ctRN7n0zyJ0l+v7u/tpkJVdUPJ3nZRNV/J7lhd1+2mXFXiXX7JL+a5Luz8v1TVyR5U5JndPeHtjo+AAAAAMC+pKoenOQZSW6/SpMLMzox6tmbuMaDfZBkz35gvNvmDklumtHxbQcn+VpGO3ne392f2MHpbVpVXTvJt2aUpT40o/9gnZ7kVBeLAQAAAAD7m6q6aZK7Jjk6o6s8zk/y0Yw+M714B6fGNpHsAQAAAAAAWGArHX0FAAAAAADAgpDsAQAAAAAAWGCSPQAAAAAAAAtMsgcAAAAAAGCBSfYAAAAAAAAsMMkeAAAAAACABSbZAwAAAAAAsMAkewAAAAAAABaYZA8AAAAAAMACk+wBAAAAAABYYJI9AAAAAAAAC0yyBwAAAAAAYIHt3ukJwKKqqtskuV2So5IcmuT8JB9L8u7uvmgHp7ZhVXWtJCclOT6jNV2Y5PQkp3b3l3ZybrCSeb0Pq+rQJHdKcrMkRyQ5KMlXknwhyXu7+7NbGOvgJCcmuWWSaya5JMmZSd7T3Z/aqjiwleb4XrxWkm9Ncv0kRyb5epIzMnp/fG6r4kzE8nciG1JVleS4JLdNckxGf3d8PcmXk3wiyb9398VbHPPwjP7M3jzJNyW5KKP3x7u7+/NbHOvWSe6Y0XvxgCRfSvKfGb0XL9vCOHNbE8tp2d+L87KMa2J+lvF9WFW7ktw0ozVdfyLGeUk+muQD3X3pZuMMYvq3KbCu6u6dngNsiao6Osldktx1/POdkhw+0eSM7j5ukzEOS/KEJI9PcoNVml2U5FVJnjXLB8BV9fYk99zM/Cb8S3efPEPs2yX55ST3z8o7/i5P8ndJfrG7P7wlM2QpLfr7cIVY90ryM0nul9GHWav5RJI/SfL7G/1gu6qOSvLMJI/M6B/vK3lfkl/p7tdvJAb7jyV8L353kmckuVtW35n+r0me091v3GiccSx/J7IhVXXNJA9K8l1J7p3k2ms0vzSjP0e/3d3/ssm4N87oz+zDMvpmhKFO8i9Jntnd79hEnEryY0l+PqMPz1bypSR/mOR53X3hJmLNZU0sp/3gvXiT7P13/B2SHDLRZKb/F1wnlvciG7KM78Oqul6SByf5jiQnZ5TgWc1FSf4yyW919wdmibNCXP82BabX3R7Pwj4ZfVfDXyc5K6O/tNd6PrPJWHeeMs6e54IkD5th/LfPMPZ6z5tniPuEjP5xNc24lyb5Pzv9++7Zt55leh9OxDk4yYs38N77RJI7byDeyUnOmSHOS5MctNO/955961nS9+Jh4zXN8j788yQHb3Bd/k70bOhJ8vsZfZfyRv7d9tIk37TBuA/L6Dt7p4lzRZLnZfwNfzPGOSLJm2dY0+lJbr0vr8mznM+yvheTfE9GH+ZO8+/Ft2/Rr6X3omejf3aW7n2Y5PUZJVVmXc/lSX49G/x/t/i3qcfjmfFxjBuL7s5Jvne7g4y/u/9NGX0APOnrGf3P7EUZbUe+7sRrhyZ5ZVUd0N2v3O45Dkz1Xc1V9aQkv7HCS19I8vmMvlP7+hP1u5P8blVVd//upmfJsliq92FV7c7ow+XvXuHl85N8JqN/SF8vyQ0Hr980yVuq6tu6+31TxvvWjNZ1yOCl85N8OqOj3G6YvXcW/WiSw6rqId3d08Rhv7Bs78VDM/rOy29e4eUzkpyd0QfQx2fv73L8kSRHVtUDu/vyaWKN4/k7kc24a1b+DuLLM/oz9MUkBya5UZJrDNr8aJJbjv/uuGDagFX10CSvzFW/y/ecJJ9Lcp0kRyepPV0y2pVztSRPnCHOIUn+MaMdBJMuyejvxK8nuUn23pV6kyRvq6oTu/uTM8Say5pYasv6Xrx3kvtOO6fN8l5kk5bxfXjSCmMno/8v/Pw4zsEZ/f139YnXdyX52SQ3r6oH9wxHnfq3KbAhO51t8ng282R0vNJq39Xw1UH5MxuMcUxGHyhNjvWlJI9Lctig7Z2S/NOg7deSnDBFnDsmuc8GnmcP4l2S5NpTxDsxyWWDvm9L8s0rrOntuep3jNxlp3//PfvGs0zvw3H/n11hHe9Kco8MvvMrybFJfjtX/S6vjyQ5cIpY18xVd0d8JskDJ2ON1/9HK8zrSTv9++/Zd54lfC++coV1/FmSmwzaHZXR0RbD7yB99gzr8neiZ1NPkvdO/Jn4ckbf1XzfJIcP2h2Q0W7Od6zw5/svZ4h3fEY75ib7fzDJvQbtbpHkr1aI9X0zxPrDQd/Lx++5a060OSijY0jPG7R9f5ID9rU1eZb3Wdb3Ykb/3lzt7/hh/Ldv8tfQe9GzqWcZ34dJzp1o/99JXpBREvbgQbsDkzwgyYdXiPPCGdbk36Yej2dDz45PwOPZzJMrP9j6n/FffC9I8pCMvkPk5MFfeJ/ZYIyXDsb5YpJbrtF+V5KXDPq8cRt/DV4xiPXXU/Y7ddDvb7PK1uKM/gf+jYP2/7LTv/+efeNZpvfh+B/nww+q3pB1PqjKaCfB8B/zPzBFvF8b9PlUkhus0f4XBu3Pz8SHbZ79+1my9+K9VnhPrZnczOhM+Esm2l+U5Jgp1+XvRM+mnow+2Pp0kkf///buO1qWolzc//OSox4kiGAADAhGBDMiBsxZrllUTL97zfGqmLjmnH56rxkxAEbMiqgogooSREVBiQIiWRElnvf7R/WGnt4TesKevWd4PmvNWqdrqrqq53RNz+63qwpYv0X+NYGPdjnP79Oyvubvv6PoMe0N5enlZl1/BtZqUc+tWXyzqef1DbgN5cZePf8zVtIx+Zrv1xz3xfdX+S8Avge8iTK125aUQGt9n4eN+RnaF32N9ZrHfkgJ9vyW8tu6TZ9dj/J3ZL2eK4BbtTwmf5v68uVrpNeyN8CXr3FelCc4dgTW6PLe7o2L3Wkj7P8mLJ4f9VEtyq0HnNgod88lOP7rU56SrtfziBblHtwocz6w+YAyW9D5NEsCeyz3OeBr+V/z1A8pT2c1f5D3DL40yn6vUfYLA/JvzuLRFvcbUCYo01rVy7xluc8BXyvjNWd98fuN/Ae3bGNztOvHW5Txmuhr7BfwUIacj59yc+tXjfPo8y3K3YbOEaWXAzsMKLMecFKjrme3qOugRpn9W5R5ZvP7hgGjXad5TL7m+zXHffE2wM17vPf0xv4OG+Pzsy/6Gvs1j/2QMvPCot/YA8psSJlCbqi/3fC3qS9fvsZ4dZtvUpoZmXlyZp6QmauXqIqHQsfaVr/PzINbtOsy4N2N5OdMsF0LHk/nOh/nUtZRGORZje0PZ+Z5/Qpk5rnARwbsR9dBc9YPt29s/zIzzx7YwuIrje1bDMj/BMri8wt+mpk/7FcgM5NyM7tu74iIbvl13TIvfTEiNqIEXuve0rKN76M8BLHgyRGxQa/MFa+JGltmfjszrxiyzNWUEXh1D2xRdG861w04MDP/MKCuyygLUdf1PWcjYhPgMfXdAG9s0b5PU9bVWnAzytTD/UzlmDT/5rEvVuV+n5knt2jTuOyLGts89sPM/Pqwv7Ez81KguXZOm2Pyt6mkkRnskfq7d2P7m0OUbeZ9eLXw+yQ9rbH9+Ryw4F9ErMviHxifallfM9+DI6LbwovSJE2zH96gsf2XIeo6o7G9akD+Rza2P9mynh9TpkVYsCVwt5ZlpXFMqy/ek86g0l8z81dtKsnMi4HDa0nrU56O7MprolaAwxvbm7YIUD6isd32+nEQcGlt+84RsVWf/M0A72GZecqgSqqbYZ9uJD9qQLFpHZPUy0rui9M0j8ek2TGP/bB5TDftl9nfppLGZbBH6q95If5N24KZeQ5lpM2CTYB7TaJRABFxS8qifXX7tSi6O2U48YITM/P0Hnk7ZOZpwJ9qSRuz+OafNGnT7Id/b2yv3zVXd8285/fKWI1c2K2RfEibSqrRPYc2kh/Wpqw0pmn1xWY9x7etp0e7mjcB6nbHa6KW10Vd0q7fK3NEbE/nyNFLgSPbVFQ9YVzPG5SATi/N91pdpyo/aGz3vE5N+ZikXlZyX5yKeTwmzZx57IfNY+p5PJXd8beppDEY7JH627SxfeGQ5S9obO80RluamqN6js3MNjfE7tjYbvVjqOaIAfuTJm2a/fC4xvYuQ0yRdpfG9lF98t4GWLu2fWp1M7wt+6GWw7T64jT7/B0b214TNW1bd0lrnsN1d2xsHzVoVHfDMOds871h+sfRlHUTFmwVEZu3rGcpj0nqZSX3xWm5Y2N7Ho5Js2Ue+2HzmPodT7c2+NtU0lAM9kj9NedkXXPI8ms3tncYoy3XiIg1gKc2kvdrWbzZhhOGrL6ZfyLHJPUxzX74czqnY7sxZW2dviLiBsCza0nJ4ils+rXBfqhZMK2+OOl6bhURvfZhX9Rya45wO33AOgdTOWcjYm0Wrz3Xuq7MvBxori/Sq3/YD7USrMi+OGXzeEyaLfPYD5vHdNKA/LNwTJJWMIM9Un/Np4m3GLJ88wnGSV1o70vnNDdXAJ9vWba5AP0wa5J0y9/cnzRpU+uH1cKgr24kfywiHtSrTERsDXwH2KyW/OHM7DfF1aT74c0iYr0h9yENa1p9cdx6mvnXBbbtkddropbb3o3t7wzIP61zdjs61+v5d2b2nJ50zLrsh1oJVmpfnKZ5PCbNlrnqh9XDRns1kmf6mCStfAZ7pP6ai9DeuW3Bak2d5nysNxy7RUVzCrdvZeag4cALmjfBzhyy7rMG7E+atKn2w8z8AvCOWtJGwHcj4ocR8ZKIeEREPDginhERnwROBO5ay/9Z4MUDmjZuP/wbUJ+iYA0WT30lTdq0+mKznjtVI1rb2qVLWq+6vCZq2UTEQ1i8ftt+A4pN+pztNbVas55muVHq6tU/pnVMUlcrvC9O0zwek2bEnPbD51IenlhwJfCFAWX8bSppLAZ7pP4Ob2w/tprWoo0ndknbaMz2EBEbA49pJO83xC6abbh0yCY08499TNIAU++Hmfkq4PFAfTHM+wLvBb5OeSLrU5SnzxYW0DweeFxm7lWNEOpnrH6YmQn8e8A+pUmbVl/8BeWP4QXXAx7SppKI2Aa42xB1eU3Usqim//xoI/ngzOy33htM/pxdOyLWXYJ6upWZVj/sdUzSIjPQF6dpHo9JM2Ae+2FE3Bx4eyP5w5k5KHjjb1NJYzHYI/X3HTpvOG0JvGhQoYjYEnhhl7cmcaF9HLBBbftvwHeHKN9sw2VD1u8NZk3bsvTDzPwiZWH3/Vpk/x3wbkogqI1x+yHYFzV9U+mLmfkv4AeN5P9pGVh6C91/37a9yew1UUuuGqn2Ocq6cAv+Tvd+0jTpc7bbPidRT7e6ptUP+9UlXWNG+uI0zeMxaYWbx34YERsAXwY2riWfDry+RXF/m0oai8EeqY/MPAs4sJH8loh4VK8yEbEJcDDdp1RafwLNak7h9vnMvKprzu6a63r0W/Cwm8sb25M4Jqmn5eiHEbFWROxDmU7q6S2aeVtgf+CUiHhYi/zj9kOwL2rKptwX39PY3gn4dESs1S1zVddrgCf1eLtXXV4TtRzeBTy4kfbczGwzL/+kz1noft5O8zo1rWOSmmahL07TPB6TVr656ocREcBngDvWkq8CnpyZl7TYhb9NJY3FYI802KuB82rb6wBfjYjPRMRuEbEqItaNiJtHxAuB33Pt+h0XN/b1z3EaEhHbAbs2kvcbcjfNJ0PWGbJ8c1jzKE96SsOaWj+MiOsBhwFvBlZVyWcCrwDuQJlSal3gppSp3n5UK7418M2IeP6A4xm3H4J9UctjKn0xM3/E4jnNnwwcGxFPiYitI2LtiNg8Ih4eEYdSRvVA+aO4+VRjr7q8Jmqqqn7x0kbyOzPzoJa7mPQ5222fk6inW129+se0jkm6xgz1xWmax2PSCjan/fA9wJ6NtBdl5hEty/vbVNJYDPZIA1RPMj+ezrlPA9gL+AlwEeUC+mfgA8CNqjxnAa9s7O7iMZvztKruBcdk5m+H3EfzhlfzyZFBmk+GjBXAktqYVj+snsT6CnDPWvK3gR0z892ZeXxmXpKZV2TmXzLzi5l5P+A/gayV+WBE3LfPIY3bD8G+qGUw5Wvic4FfNdJuC3yWEoC9AjgX+AZwv1qeZwP/almX10RNTUQ8CXh/I3k/4FVD7GbS52y3fU6inm519eof0zomCZi5vjhN83hMWqHmsR9GxKuAlzSS983MjwyxG3+bShqLwR6phcz8MXAv4OSWRY6lLObeHKZ78ahtqG5C79VI3m+EXTUv9ht2zdVbM78/HjQVU+qHewH3r23/AfiPQUPuM/P/KCOBFgTwoT5FxuqH1feBP+S1LKZ1TczMfwL3ZvEIn17+CeydmfuzeH7yXnV5TdRUVFN8fobOh3a+CjwrM7N7qa4mfc5elZndnvodt55uZdoGe5bqmKRZ7IvTNI/HpBVoHvthRDwXeFsj+YOZ+cYhd+VvU0ljMdgjtZSZxwI7UJ40/hll3tWm31AWq75bZp7E4jUK/jxGE3YHtqltX0H7G2B15za2b9w1V29bD9iftGSm0A+bi82/KTO7LdzZzTvovKG8Y0Tcs0fecfvhDYH62iWrgfOH3Ic0smldEzPz35n5ZMpUcAfQ/Ty/EPgIcMfM/HREbEjnFBZJWX+rG6+JWnIRcR/gS3R+b/8AeGJmXj3k7iZ9zp7XNdfieprlRqmrV/+Y1jHpOm5G++I0zeMxaYWZx35YjVJqjt75DPDiEXbnb1NJY+m5yK2kxTLzSuBjwMeqm0k3ATajTGfzl8xs3oTasbH96zGqf1pj+1uZecEI+zkRuFtt+6ZDlm/m/+MIbZBGtlT9MCJuQOdCmquB7wzRrksj4sfAo2vJuwHd5mc+sbE9bj883ScnNW3TvCZm5lHAkyJiDcr5vzllDvMzgbMysx5satZzUmb+o8euvSZqSUXEXSlTDdanYTkSeHRmDrvoMkz++tHrnD2FEsRd+Htx/YjYPDOHuRHWtq5pHZOuw2a4L07TPB6TVpB57IcR8UhKYKf+MP1XgGcOOUppgb9NJY3FYI80osy8lMEXzts1tkcK9kTERixe5O/To+yLxW1u3hQbZIcB+5OmZsL9cBs6pxK4MDP/PmSTTm1s93oS2n6ouTKta2JmrgZOq16TqMe+qCUTEbcHvkvntILHAg+p+swopnLOZuaVEXEysH2jrp+0qSQi1gW2a1NXl3T7oSZqlvvilM3jMWmFmMd+GBH3Bw6i897q94EnjTBKqVcb7IeShuI0btISqUYJ1J/I+DvlqZVR7Enn3KvnAN8bcV/HNbbvMWT55rRUzf1JK8aQ/XDdxna3aakGubKxvWaPfL9v5N0mIm7UI2839kPNlAlfEwd5aGP7u33yHtfY9pqoiYiI7SnT0mxSS/4D8MARHiSoO66xfeeIGOYBvmHO2eZ7w/SPnem8rv41M3tNJdOsZymPSdcxc9IXp+W4xvY8HJNWgHnsh9V03V+n81p3OKOPUurVBn+bShqKwR5p6TwWWLu2fUBm/mvEfTWncPt8Y8qaYRxGmWJnwa0i4mZtCkbENsAta0mXVPuTVqph+mFzWsRNI2KdIetrNe9zZl4C/LSRvEebCiIigPs3kr/ZqnXS8pnkNbGniLge8JBa0kWUqTR6OQyviZqw6hw6FNiilnwqsMeQ06Atkpl/BE6uJW1IyxtB1XSLd6/vDvhWnyLN91pdp3rk7XmdmvIx6TpkjvriVMzjMWn5zWM/jIg7Ad8GNqgl/xp42BDrvfZyGP42lTQGgz3SEoiI9YB9GskfH3Ff2wD3biTvN8q+AKp1PQ5pJO/dsngz3/fGfGpFWjIj9MOz6BzNszZwryHqWwPYvZF8cpesC77R2H5my6ruA2xb2/4b8MuWZaWpm+Q1sYV96JwH/nP91rPymqhJq0Zp/pDOBZXPAu6XmWdNqJpRrx+Pp3P6nF9n5tl98n+Hzuvi7hHRnJptkeqhhKc3kr8+oNi0jknXEXPWF6dpHo9Jy2Qe+2FE7EiZqu36teTfAQ/qs0Zka/42lTQugz3S0ng7UH/64guZecyI+9qLznVEjs7M343csuKTje3nRcTm/QpExBbAfw3Yj7SSDNUPq7mif9FIfvkQ9T2Vzj9koPxx08uBdD61tVtE3LdfBdUNtDc0kj9drWMirVSTvCb2VC36++Ja0kXAm1sU9ZqoiaimK/wBcPNa8nmUp5eba7qN41OUJ5AXPCEimnP0N9u2HvCqRnLfczYzLwQOru8GeGOL9u1NWQdvwemUp7r7mcox6bph3vrilM3jMWkZzGM/jIhtKce0WS35z5Rjas4SMQ5/m0oamcEeqYWIuHXLfGtGxFuBF9WSLwJeMmK9QQn21O03yr7qMvPbdN7U3hT4ZESs3S1/NZXVJ6t8Cw7PzO+P2xaprSn1w881th8UEa9rUeddgA81kn+UmWf2KlOtXfD/N5I/ERFb9anq1cBute2/A+8a1D5pkqZ1TYyIm7edd72aN/27QH3qxf/us0bINbwmahIiYmPKeoq3qSVfDDwgM/8wybqqh36+WEtaB/hMNY1ht7YF8H46p3Y5hXKDbJA3APUHCp4aEU/slbl64vndjeQ3DXqyeMrHpDk2x31xKubxmDR989gPq7/RDgXqf6udQRmldM7wLe/N36aSxhGZOTiXtIJVN3jW7/LWHej8Y/NvwFN67ObszDyhTx3nURZUPwD4bmae0Xh/I8oaAa8Cdqq9tRrYMzO/Nug4etS7G/CTWtIVwI2qJy3HEhG7VvuuB30PA15Wf+I6InYG3kPnVHJXA7tmZnMUhK6j5qUfVj+gfwfcqvHWN4G3AEdl7cIZEVsDzwVeSefinKuBe2Rm3+nVqifefg9sWUs+HXgh8M2FuiLixsBrq7rqXpmZBnt0jXnpi9V+3k2ZYuOzlNEFx2bmlbX3A7gLZbqOZ9J5PftqVVerH7peEzWuiPgxi6fyfD3w8xF2d3RmXjSgvlsAv6FzvYDfAC/OzMNq+W4FvA14TGMXj8vML7VpTER8FHhOLWk15Zr4voV2VtfPJwPvpXMB7uOBndusNTnNY9L8mvO+2FyzccEDgFfUto8HXtYj7ymZecqAeuyLGsu89cOI2AA4is7g1dXAC4A/tTiGpp/1m2a4qtPfppJGYrBHMy8iTqNzephRfCYzn96njvPpfEriQuBM4DLKQoNb0fk0MZQL7LMz89OjNioiPknnvKtfycw9R91fl/2/EnhHl7fOBv5KOa4bdXn/ZZn53km1Q7NvnvphNez/Z8ANurx9EeUJriuAGwI37bGbF2Rmc9ROr/p2o8z7vF7jrYspi5euqupZs/H+14FHt72ZreuGOeuL76bzZtUVlD7xd2Djqp7rdyn6HeCxg/6I7lKf10SNLCIm+V18n/rNqT51PgH4Ap3T/UKZJucMSn+8cZf3P5SZL2zbmOom10+AXRpvLfTJy4Ht6Fz7AOB84J6ZedIQdU3lmDS/5rwvTuLY9s3MN7aoy76okc1bP4yyjvIkp57bNjNPG5TJ36aSRtFqagxJi9yA7jeCF5wJPC0zfzRqBdUf1v/RSN5v1P11k5nvjIirKT8g6jeSt6JzePKCq4GXZ+b7J9kOaURL0g8z8w9VAOYLwO0bb29C5xPLTZcAL8zM/Yao76cR8VDgS3Qezyo6R0XUfQHY20CPVoglvyZW1gG27/P+VcA7gTe0GUXQ5DVRsyYzD6xGuH2SzhF9m1evbt5NGY06TD3/iogHUq5T9bXl+vXJ04BHDBPoqeqayjFJkzSP5+08HpPm2zyes/42lTQK1+yR2nkrZcjxoJtHf6b8WNh+Aje1HkN5cnnBOZR5bycqM99DeVLz23TOyV63GvgWZRqO90+6DVJLU+uHmfl74M7As4CjWxT5G+VH+I7DBHpq9f0I2BH4X+BffbIeSxmx8OTMvHzYeqQJmVZf/Brl5nLfqTuASykPQ9wuM/cZJdCzwGuiZk1mHgDclvIQwJV9sv4U2D0zXzHKgwLVFMJ7UKZz+3OfrBdSviNul5m/Hbaeqq6pHJM0SfN43s7jMWm+zeM5629TScNyGjdpCNVom52AW1CGAa9HuTF7JnBMZo4yX+uKERGbAbtSpuLYkHID7WTgiMw8fznbJi1Yjn5Yra1zZ8p0aquAtSlTSZ1HWUdkYnVGxPrAPYAdqrquAM4CfpmZ/W6wSVM1rb5YPaV5a0pA9MaUByGupvS/PwC/ygGLv49Yr9dEzZRqMepdKYtOb0yZWvEMyjl71oTruh1wJ8r0MWsCF1DWvPtlfW2tCdQztWOSJmUez9t5PCbNt3k8Z/1tKqkNgz2SJEmSJEmSJEkzzGncJEmSJEmSJEmSZpjBHkmSJEmSJEmSpBlmsEeSJEmSJEmSJGmGGeyRJEmSJEmSJEmaYQZ7JEmSJEmSJEmSZpjBHkmSJEmSJEmSpBlmsEeSJEmSJEmSJGmGGeyRJEmSJEmSJEmaYQZ7JEmSJEmSJEmSZpjBHkmSJEmSJEmSpBlmsEeSJEmSJEmSJGmGGeyRJEmSJEmSJEmaYQZ7JEmSJEmSJEmSZpjBHkmSJEmSJEmSpBlmsEeSJEmSJEmSJGmGGeyRJEmSJEmSJEmaYQZ7JEmSJEmSJEmSZpjBHkmSJEmSJEmSpBlmsEeSJEmSJEmSJGmGGeyRJEmSJEmSJEmaYQZ7JEmSJEmSJEmSZpjBHkmSJEmSJEmSpBlmsEeSJEmSJEmSJGmGGeyRJEmSJEmSJEmaYQZ7JEmSJC2JiNgmIrL22m+52zQJEbFf47i2We42SZIkSbpuW2u5GyBJkiT1EhFrA3cAbglsCWwIXAZcBJwLHJOZZy1fCyVJkiRJWn4GeyRJ0lyqnrQ/dQpV7ZuZb5xCPdcZEbEmsCfwFOD+wHoD8p8NfAv4TGYeufQtlNRPRKwBHA7co5b8+cx8yhj7vA/wQyCqpCuAXTLztyM3VJIkSZojTuMmSZKkFSMiHg6cBBwIPIwBgZ7KVsBzgCMi4tiIuN8SNlFzoDEF22HL3Z55k5mrgb0po/AWPDkiHjbK/iJiA+ATXBvoAXizgR5JkiTpWgZ7JEmStOwiYu2I+F/gG8B2XbJcBZwDHA8cW/376i757ggcGhHPW6KmSmohM08EXt9I/mhErBphd2+l83vhOODto7VMkiRJmk9O4yZJkubVOcAeLfM+AHhFbft44GUty54yTKO0WLUuz1eAhzfeuhr4HPBV4IeZeWmXcrsDjwCeCGxae3vzpWqvlJlPB56+zM2YBe8FHgvctdreqkrbu+0OIuIewAtqSVcCz8jMKyfVSEmSJGkeGOyRJElzKTMvAw5tkzcibtxIuigzW5XVRLyDxYGenwHPzcwTehWqbvb+APhBROwD/DfwYmCDJWqnpCFk5tURsTdwDLBulfyMiDgoM78/qHxErAd8ks4ZKd6WmcdNvLGSJEnSjHMaN0mSJC2bag2PlzSSvwHcv1+gpykz/5GZ+wC7UNb8kbQCVP1430byxyJi4xbF3wjcurb9W+AtE2qaJEmSNFcM9kiSJGlZRMS6wMcbyccDe2bm5aPsMzP/ANwF+OmYzZM0Oe8Cjq5t37RK6ykidgZeXku6ijJ92xWTb54kSZI0+5zGTZIkacIiYmtgZ8r6FJsCFwNfy8yzl7NddRGxJmUUzC2ALShTLJ0HnAocMWqwZUhPBbasbV8FPH3ctTgy8+/Aj4YpExE3pQSJbghcH7iQsu7TEZl53jjt6VHfbYHbADcG1gROAw7LzHMHlFsLuBtwO2AT4B/An6uyl02wfZsC9wK2A9ajfBYnAUdm5upJ1TNEezYA7glsTTlfrwbOBU4AjsnMnHabxhURqyjHtBWwGfBPyjEdm5kTH50WEVtQ/k+3BdYGzqd8fr/IzKsnXV9dZl4VEc8Afg2sUyU/JyK+mJmL+mq1HtenKH1jwbsy8+hm3ka5AG4P7EA5TzakHOeZwOGZ+c9xjiMiNqL021tTvtvXB/5e1XHMUvy/VfXelnJMNwI2Av4G7O+6RZIkSaoz2CNJkjSkiKjfWP5JZu5epT8UeCmwO4tHUJ8FHFzl24YSVFnwmWrB97b1Hwbce2E7M2OIstsArwMeBdygR7Z/RcRXgddn5qk98kzCSxvb38jMY5ewvg4RsQbwdMo0crftkW11RBwFvDUzv9lyv9vQ4/83Ip4CvAy4Y5eiV0bE/sArMvOixj7Xrsq9FNi8S9lLIuItwHsy86oWbdwPeFotadvMPK1q+zuBx9B5o33B2RHxXuD9Sx0gqNp5T2Af4L5cu+ZL07kR8VFKMOCSPvs6jFq/qbl3o0837ZuZb2zsaz+6fH599lEvuxvwBmA3evw9FhF/Bj4CfLjtSJaIOA24WbV5emZuU6XfCng78Ei6z+xwQUS8FfjQUgYPMvO3EfFm4H8Wmgx8IiJul5mXNrK/hhK0WdBtKrhrRMTmwKuBJ1ACIt1cERHfA16Xmce3bXf1+T0ReBAlQN7zb+iIOAf4X8pneVGvfI0yuwM/riXtm5lvrAK7/wk8lxJgavoK5UECSZIkCXAaN0mSpLFF8SHgW5Sb0ivyN1ZEvBY4Edib3oEegA2ApwB/jIhnLlFbdqA8qV7XnNJtyVSjr35FWfy9V6AHyv/l3YBvRMR3Wq4z0q2+NatAzmfpHuiBMtrimcAR1c3rhbKbAT8D3kb3QA/AxpQb+gdUo7ZGaeN9KNPo/QfdAz1QRqG8u2pjv3NoLBGxYUR8kXLcD6Z3oAfKCI7XASdFxJ2Xqk3jioh1qnPgJ5TviX4P3t0CeC/wu4i4dZ98g+rcEzgOeDS9v5c2Bd4DfC0i1hu1rpbeXrVnwbZV2jWqUSyvqSVdTZm+retow+o76mRK0LZXoAfKiKJHAMdGxOvbNLZaU+xEytpBd2Pww5JbUoJSvx3nXIyITSgBoA/SPdAjSZIkLbIib0RIkiTNmLcCz69t/xv4I2WNinOWpUU1VaBhP+BNXDuF0oILKDf4f00ZfVS3DuXJ+5csQbPu1dj+F/CDJahnkYjYFjgSuFPjrdXAKZTP4vQuRR8M/Li6ETusD1KmrVtwHnAMZcH55k3sHYDPV21dHziEMsXcgjMogaoTgeaIlD2BV43Qvh2Bb1CCRlT7PZXen8Vdge+PGvzqp5pu7CeUoFPTmZR+dRzQHDmxJXBYROw66TaNq1qf6tt0ngML/kr5nE8CmiNrbgn8LCJ2GqHOhwIHUqYao9r3ScBRlGkDmx5KGdW1ZKqRQ8+gTNm44HkRcS+4ZnrJT9P5PfXezDyq2/4i4k3AJ7j2vF3wD+D3dD/WNYB9I+IDLZrcLfj1b8rneDTl/+1UyndH3daUc7EZ0G5jLUpfrJ/HF1G+p4+nTBsnSZIkLWKwR5IkaTw7Aq+s/v0nyg3qG2TmDpm5S2beiLK+Sutpg5bA6+mccupK4APAbTJzs8y8Q2beOTNvTBlR8FE6gwjvjIh7TLhNzWDPcVOaFmwt4ADKAvELrqKMLrhJZt68+iy2oXwWn2jsYmfKNE3DuDfwX9W/DwHukplbZObOmXl7ynotb6DzM9+juln/PmAnys3kD1OmC7tZZt4lM29dHcdXG/XtUwVMhvERylogCXwI2CYzt2t8Fp9ulNmFCQcHqqn1DqR8zgvOA14B3Cgzb1L1q50on9uudK7PtAFldNOmXXb/MmCP6lV3fC2922v/cY+LEhC+fyPtYOAOmblV9TlvTwlYvZISUFiwKfClar2Ytq5PGUW2JiVAtjewaWZun5l3zcxtgVtRAlB1z4uIJR1JkpnHUUapLQjgk1Vg8+WU82rBiZTvr0WqNYBeW9815f/qzsAmmXnb2rFuXdVZD6a9MCIe16LJV1M+p+dSgm8bVZ/jLtX/23bAKspoyPqaPRsAX6jWERrGs7g20HMoZV2nhe/pO1DW6tqDznNEkiRJImZwLVNJkqSJioin03kj+5p1eHrk7/YD6nDgIW0WAJ/mmj1VkOZwrn3I53zgwZn56wF1PB74Qq3c8dWNxomIiF/ReVP3w5n5/F75J1jvyyjTkC24AnhkZn6vT5lnAx9rJD86Mw/ukX8bOv9/F3wIeFH2+AFeTS1VX5fkRMoN+dXAEzLzyz3KrUkJIt23lvySzHx/t/xVmf3oDABCuVm+V2Z+rk+5FwH1/SZw98z8ZY/82zDEuR4R/03ntF6/BB6Rmef2KbMGJXhZP38+mJkv6lOm67pbbQ2zZk81ndcvKUGNBf+TmW/os/+dKUGs69WSP5CZL+5T5jSuXbNnwTHAgzLzvB5l1qRMP/mgWvL7M3MpRvPV612HMirmdrXkL1KmWVsYTbMauFdmHtml/HaUUXEbVEn/Bh6bmd8dUO9uwPe4drTTucDNMvOyHvm3BVZnZrfRbd3yr0dZS+chteSH9GtXlzV7Fiz5/4MkSZLmiyN7JEmSxncR8Lg2gZ5l8Hqu/c23mhLY6BvoAcjMgyjreCy4fUQ0RyaMo7neS8+b+ZNS3dhuBgBe0y/QA5CZHwf+r5H8siGr/znw4l6Bnso76VxwfXtKgOAdvQI9VfuupnOEA5Qp54b14X6BnqquD1BG3iwIFn+mI4mIDbh2lByU6c0e0i/QU7VpNfBi4Be15L0jYtUk2jUBL6Ez0POtfoEegMw8GnhOI/lZEXH9Ier9B/CYXoGeqp6rq/bVjXLuDCUzr2DxdG6Po3PatA90C/RUXsm1gR6AvQcFeqp6f0oZPbRgC8qInF75T20b6KnyX0aZqq8+1doz2pavORJ46QjlJEmSdB1msEeSJGl8H87MZV+bp6laL+KBtaSD+tw87eYddN6MfexEGlY0gz0XT3DfvTwQuElt+3TKiJA29qGsK7Rg1yHX43hDFZToqbpRfEgj+VJaTJWWmT+nM2A27Bovl1GmkmvjVXSuUfKYCQVW9qLzvHhjZl7YpmAVtKhPDbYRnef+sqg+l3q/SVoGCquAaz2AtSHwpCGq/782gYrM/COd00zecsgp40ZSBbTe1ePtP1P63CIRcQPKubLg55l5YLe8PXyczr4yye81qnO2HngaZQrM1w8IDEuSJEmLGOyRJEka3wHL3YAemk/of3aYwpl5AWUR8gXNdXbG0VxQ/dIJ7ruXeze298/Mq7rmbKhu4B7cSN6tZb0XU9beaON3je0fZGbbBdnrZTevppRq69tDBFZOB35SS1oXuNsQdfVSn/rqKjpHELXxQzqDUJM8X0d1d2Cd2vbPMvOkXpm7+FRju+05B3DQEHmPq/17DcoaN9OwL3BCIy0pI3V6rUmzO9dOwwbDf69dSee0afeopgKcpPrUhVtHxOZDlP0bnetQSZIkSa0Y7JEkSRrPJcAflrsRPTRvdg+cvq2LM2r/vvUIi433cklje8MJ7befuza2h72h+sPGdtsAxzFDPKV/QbNsy3Ldyl6va67uDhsib7f8dxmyfIfqvLpnLemkzPzHMPvIzEvp/AyGGXm1VJbrnLsS+M0Q9TSnyhtmuriRZeblwPsayUdm5uF9ik36e+16tAhuRcSqiHhmRHw6Io6JiL9GxKURkc0X8OpG8c2GaNuvHdUjSZKkUay13A2QJEmacaev4BtzzZvd544Zq1mTcmO07UiTfi4EVtW2p3Fzubl4/fFdc/XWvHl+05bleq6Z0sW/GtvjlF2/a67umiOKhs2/7ZDlm25I5xRuO1Y3zcfRnCpwOYx1zmXmKRFxCdeOhLtJRESL75wLq6nt2mqOrBvm3BlXc3TdoNF2ze+1oyYQg74B8Jdub0TEhpQpDl9IGcU2ilVD5D11cBZJkiRpMUf2SJIkjWeo0QdTtukS7HNSQZnmlGFbTGi//WxS+/fqttOW1ZzfZ3/9XDZkPZMqO8wd8OaooGHzrxqyfNNKPlfH0TxHmudQG/XPek0WT4HYzTjnDQx37kzb1M6ViNgM+DnwCkYP9DBk2ZV8TZEkSdIK5sgeSZKk8Vy53A3oY9US7HNSDwv9Ediltr3ThPbbT33R+eYomDaaox/a3HSfFcN+Hs3PYqOuudpbNWb5blbCg23Nz2WUtam6nXfX5YDAqiXYZ69z5UvA7Rppf6Gs+XMCcCbwT+DfdK4XtRfw1BHbspKvKZIkSVrBDPZIkiTNr3/RuW7Lgxk8RdIg54xZfsHhwFNq2ztFxJpDTj01rH9y7RP8G4xQvrmuUHPdoVk27OfR/Cz+OWb9zWDTCcCLxtznv8csPwnNz2WUtanm+bwbRfNceQYl6DKOResbRcQjgN1rSZcA/wkckJmrm/kbZe83ZnskSZKkoRnskSRJmr5x1yJpe2P+fDqDPcdkZnMh9uXSXIB9A+B+wCFLWOdFXBvsWSMiNsnMi4Yo31xkfZiyK90wC8jD4qm0Lh6z/ub0ZpGZh465z5WgeY6MMgVZvczVGOxpnisnZOZRS1DPExrbz83MA1qWXQnrRUmSJOk6ZiVMbSBJknRd01xPY9jF0Ddvma+50PcthqxnyWTmHyhTudU9e4mrPb2xfYchyzfzN/c3y247ZP7m1FbjLip/Dp0jcW4WEWuPuc+VYKxzLiK2o3O6wDMyc9xg8ayb1vfa3Wr/vgD44hBlbzPhtkiSJEkDGeyRJEmavuZ6GzdsW7BaMHybltl/3Ni+b9t6puS9je1HRcSwAZhh/KKxPezn0czf3N8su/eY+ccaWZGZVwJH1JI2AO46zj4HVVn7dyxhPZ5zkzet77X69/Kf204xGRHXA3ZemiZJkiRJvRnskSRJmrLM/DdwXi3p9hHR9nfZ44eo6nuN7eessNES+wN/q22vBew3bhsjYoOI6Ba8+Elj+ykR0Wpa44jYBHh0I/mno7RvhXpoRLSaeioibkZnsOdyJhOEaJ6vL5jAPnu5tPbvUdZvausXwBW17V0jYpiRKHs3tpvn8HXRoXSuPfaEiBhlerxB6kHAdYYotzew3oTbIkmSJA1ksEeSJGl5HFP79ybAHoMKRMT1gf9uW0FmHk3nU/A3Ad7ctvxSy8zLgec2ku8IHBgRw9xcvUZE3JyyHtB9urx9CHBGbXtb4Pktd/0mOoMCh2dmcxq6WbYesG/LvG+n8++Ir2XmxRNowyfoXPtnz4h46AT2282FtX9vs0R1UH0uX64lBfDuNmUjYk/g7rWkfwJt14yZW5n5N+CztaQNgQ8vQVXn1P59m4hYNahARGwNvGEJ2iJJkiQNZLBHkiRpeXynsf2OiOg5wiAiNgQOogRshvE6YHVt+5UR8fqIaD11VUTcOCLeFRF3HrLugTLz68AHG8mPAQ6JiFu33U9ErBcRLweOA+7Uo66rgQ80kt8eEfcbsO+9gf9qJL+nbdtmyPMi4sn9MkTEC+lcuD5Z/JmOJDP/DryjlrQGcEBEPGKY/UTEzhFx0IBsv6/9e7OI2H2YOob0Pjr74CMj4rX9CkTEHSnBr7pPZGZzCsjrqjcD/6ptPz4iPjpMkDgibhARr42Ih/fIcmTt3+sAbxuwv82BbwGr2rZBkiRJmiSDPZIkScvj83TerLwDcGhE7FTPVAUxHgP8CnggZUqoM9tWkplHAPs0kvcFfhURT6imJ+sQEWtGxA4R8ZyI+D5lQfSXU56gXwqvYHHw697AbyPiUxHxsG6BsIhYOyLuGxHvq9r4LmCjAXV9APhlbXtd4LsR8ZaIuFFj/9tFxEcpN93rwbGDqiDVPDmdcoyfjYgPRkRHULH6LD7J4sDOxzNzkuvIvJPOc2Fj4OCI+Gr1f71us0DVR+4SEa+JiKOBXwOPG1DPIY3tr0XE2yPiPyLiARFx/9pru3EOKDN/TQn41L0pIr4cEbdtHMsmVdDyCOD6tbdOBvoGiK5LMvMU4JmN5OdQvjOeHRGL1kGL4uYR8dSI+Crle/RNQK8p4PZvbP9/EfGZahrD+n43johnAsdTRiYC/GG4I5IkSZLG12qOckmSJE1WZl4QEa+nc0qnuwPHRMRZwNmU4Mq2wPq1PC8EngjceIi63h4RWwAvqSXvTJkSanVEnAFcUKWvAm7E0q5j0mzfFRHxaMpUTM+qvbUW8IzqdVVEnEtZ6+hqYEtgC3r/nv1rj7qujognUaa3u2mVvDbwGuBVEXEqZYqvzek+vdcxwH+2PrjZ8V/AgZTgyguA51efxQX0/iyOpQTqJiYzV1f/PwcDu1fJQVkv6dHA5RFxOnARZeq5VZS+sOaQVe1PCYJuVm2vovcUifsCbxxy/037UAK696+lPRZ4bEScTenvGwPbUc7HuguAx2XmpegamXlgRGxFCfIuPMR4K+BjwMci4i/A+ZT1fVZRvjM2HmL/h0TEd4CH1JL3AvaKiFMo30WrKN/R9RFFXwD+hNO5SZIkacoc2SNJkrR83gd8vEv61sCdgR25NtCzGnhxZn50lIoy86XA0+hcEwXK78FtKMGfnYGb0z3Qc0mXshOTmVdk5rMpN/RP65JlLWAryg3zO1X/7hbo+Tlwz36fUzUq4J50rpsE5bO4OeWz36ZL0e8Cu2fmRf2OZUadADwSWJgmLCiBh16fxa+APZZiWrFqOrc9gPdSbtTXrUu5oX9XyrlwM7oHev4yoI4LKcGWc8dtbxvV+lQPBT7X5e2tgF2A7Vkc6PkTsGtmNs9VAZn5Xkowpltw9ybATpRz+JZ0D/RcTv9z4MnAUV3St6Ocg9vTGeg5kBKcliRJkqbOYI8kSdIyyczVmfkc4OnAGX2yHg7cIzPHWhslM/en3Lh/HXBSiyIXURaX3wvYMjOPG6f+NjLzYMqN2SdRgiuXtyh2JmVU0C6ZeY/MPHJQgcw8k3IT+Fl0rt+yKCtl2rdHZuZDMvOSFu2ZSZn5Y0oA5cuU0VPd/BV4JXD3zLygR55JtOWqzHwZ5Wb6x2gXlDmtyvsAugeomnX8FLg18Hzgm5SpAC+hc32diakCmk+ljFj6EYsDWXUnAy8DbpuZf1yK9syLzPw+JfjyQspUajmgyD+Bb1NG6N0oM5tTSNb3fTGwG2W6t7/32efvgSdl5hMz84r2rZckSZImJzIH/RaWJEnSUouIoDyFvhNlaqmgjE44MjNPXaI6F0YQbUFZt2I1ZWTHWZQ1J07OzCW58d1WtUbLHSgBoBtSRh1dTplq7W/A0ZnZdcq2Ieu5KeVJ/RsC16MEuv5K+fynMvpjmiJiP8pIrwXbZuZptfc3A+5FuYm+LiXYciJwxHKcE1X/uE312owyfdbllBvwpwInZObZ027XOCJiFbArZWTPpsCllHP6uMw8cRmbNtMiYnNKX96S8rmuQfleO4fyvfanzLxyhP2uR5lqcwdgE8r6aWcDv8rMNsFzSZIkaUkZ7JEkSZKuYwYFeyRJkiRJs8Vp3CRJkiRJkiRJkmaYwR5JkiRJkiRJkqQZZrBHkiRJkiRJkiRphhnskSRJkiRJkiRJmmEGeyRJkiRJkiRJkmaYwR5JkiRJkiRJkqQZZrBHkiRJkiRJkiRphkVmLncbJEmSJEmSJEmSNCJH9kiSJEmSJEmSJM0wgz2SJEmSJEmSJEkzzGCPJEmSJEmSJEnSDDPYI0mSJEmSJEmSNMMM9kiSJEmSJEmSJM0wgz2SJEmSJEmSJEkzzGCPJEmSJEmSJEnSDDPYI0mSJEmSJEmSNMMM9kiSJEmSJEmSJM0wgz2SJEmSJEmSJEkzzGCPJEmSJEmSJEnSDDPYI0mSJEmSJEmSNMMM9kiSJEmSJEmSJM0wgz2SJEmSJEmSJEkzzGCPJEmSJEmSJEnSDDPYI0mSJEmSJEmSNMMM9kiSJEmSJEmSJM0wgz2SJEmSJEmSJEkzzGCPJEmSJEmSJEnSDDPYI0mSJEmSJEmSNMMM9kiSJEmSJEmSJM0wgz2SJEmSJEmSJEkzzGCPJEmSJEmSJEnSDDPYI0mSJEmSJEmSNMMM9kiSJEmSJEmSJM0wgz2SJEmSJEmSJEkzzGCPJEmSJEmSJEnSDDPYI0mSJEmSJEmSNMP+HyxRe/mzL/pmAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 2100x1500 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(7, 5), dpi=300)\n",
    "ax = fig.gca()\n",
    "\n",
    "c = ax.pcolor(crosstab.transpose(), cmap=\"BuGn\")\n",
    "\n",
    "ax.set_aspect(\"equal\")\n",
    "ax.set_xlabel(\"True Completion Year\")\n",
    "ax.set_ylabel(\"Prediction\")\n",
    "\n",
    "ax.set_xticks(range(0, 11, 2))\n",
    "ax.set_yticks(range(0, 11, 2))\n",
    "ax.set_xticklabels(range(1970, 2021, 10))\n",
    "ax.set_yticklabels(range(1970, 2021, 10))\n",
    "\n",
    "fig.colorbar(c, ax=ax)\n",
    "\n",
    "fig.savefig(\n",
    "    \"vgg_5y_heatmap.pdf\", bbox_inches=\"tight\", pad_inches=0,\n",
    ")\n",
    "fig.savefig(\n",
    "    \"vgg_5y_heatmap.png\", bbox_inches=\"tight\", pad_inches=0,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 눈 먼 정확도는 딱 한 구간 틀린 게 많다는 걸 보여주지 못함..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(29027, 50252, '57.76%')"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "correct = df[df.true == df.prediction].shape[0]\n",
    "total = df.shape[0]\n",
    "correct, total, f\"{correct/total:.2%}\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16457, '32.75%', '90.51%')"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "one_off = (\n",
    "    df[df.true - 1 == df.prediction].shape[0]\n",
    "    + df[df.true + 1 == df.prediction].shape[0]\n",
    ")\n",
    "one_off, f\"{one_off/total:.2%}\", f\"{(correct+one_off)/total:.2%}\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}