2. Text classification using LSTM.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b223ad7b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import time\n",
    "\n",
    "from collections import Counter\n",
    "\n",
    "import random\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sn\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "import torch.optim as optim\n",
    "\n",
    "from torch.utils.data import Dataset, DataLoader\n",
    "from torchtext.legacy import data\n",
    "from torchtext.vocab import Vectors, FastText\n",
    "\n",
    "from tqdm import notebook\n",
    "\n",
    "from argparse import Namespace\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.metrics import confusion_matrix, classification_report, accuracy_score"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e97b0f5f",
   "metadata": {},
   "source": [
    "[Workaround](https://github.com/NVIDIA/flowtron/issues/93)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8b0b7ac9",
   "metadata": {},
   "outputs": [],
   "source": [
    "os.environ['CUDA_LAUNCH_BLOCKING'] = '1'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "8ed71daa",
   "metadata": {},
   "outputs": [],
   "source": [
    "args = Namespace(\n",
    "    dataset_dir = \".data\",\n",
    "    dataset_prefix = \"faq_with_splits_\",\n",
    "    datasets = [\"lemmatized\", \"lemmatized_filtered\"],\n",
    "    model_save_dir = \".model_storage/LSTM\",\n",
    "    model_state_file = \"model\",\n",
    "    seed = 1234,\n",
    "    num_epochs = 20,\n",
    "    learning_rate = 1e-3,\n",
    "    batch_size = 70,\n",
    "    cuda = True,\n",
    "    train_column = ['long_question', 'short_question'],\n",
    "    embedding_dim = 300,\n",
    "    hidden_dim = 100\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "a3180cf2",
   "metadata": {},
   "outputs": [],
   "source": [
    "random.seed(args.seed)\n",
    "np.random.seed(args.seed)\n",
    "torch.manual_seed(args.seed)\n",
    "torch.backends.cudnn.deterministic = True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "fffc3832",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "device(type='cuda')"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "device = torch.device('cuda' if args.cuda & torch.cuda.is_available() else 'cpu')\n",
    "device"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "b1d6ea5f",
   "metadata": {},
   "outputs": [],
   "source": [
    "args.model_state_file = os.path.join(args.model_save_dir, args.model_state_file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "d5c1e0b7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ".model_storage/LSTM directory already exists\n"
     ]
    }
   ],
   "source": [
    "if not os.path.exists(args.model_save_dir):\n",
    "    os.makedirs(args.model_save_dir)\n",
    "    print(f\"Created directory {args.model_save_dir}\")\n",
    "else:\n",
    "    print(f\"{args.model_save_dir} directory already exists\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "582a81d6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Opening file: .data\\faq_with_splits_lemmatized.csv\n",
      "Opening file: .data\\faq_with_splits_lemmatized_filtered.csv\n"
     ]
    }
   ],
   "source": [
    "faq_dataframes = {}\n",
    "\n",
    "for dataset in args.datasets:\n",
    "    df_path = os.path.join(args.dataset_dir, args.dataset_prefix + dataset + \".csv\")\n",
    "    print(f\"Opening file: {df_path}\")\n",
    "    faq_dataframes[dataset] = pd.read_csv(df_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "aa44bf91",
   "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>short_question</th>\n",
       "      <th>long_question</th>\n",
       "      <th>answer</th>\n",
       "      <th>main_category</th>\n",
       "      <th>sub_category</th>\n",
       "      <th>split</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>vállsérülés/gyógytorna ?</td>\n",
       "      <td>vállsérülés/gyógytorna ? az lenne a kérdés , h...</td>\n",
       "      <td>edzeni kénealapjába véve mindent váll, hát, tr...</td>\n",
       "      <td>Egészség</td>\n",
       "      <td>Sérülések, balesetek</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>heti hányszor dehidtratálja a szervezet citrom...</td>\n",
       "      <td>heti hányszor dehidtratálja a szervezet citrom...</td>\n",
       "      <td>Nem értem miért akarod magad dehidratálni, nem...</td>\n",
       "      <td>Egészség</td>\n",
       "      <td>Bio életmód</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a doctor who ki a vörös hajú lány ?</td>\n",
       "      <td>a doctor who ki a vörös hajú lány ?</td>\n",
       "      <td>Vagy Amy Pond vagy Donna Noble</td>\n",
       "      <td>Szórakozás</td>\n",
       "      <td>Filmek, sorozatok</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...</td>\n",
       "      <td>vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...</td>\n",
       "      <td>Egy törpenyuszihoz nagyon sok türelem kell. Pr...</td>\n",
       "      <td>Állatok</td>\n",
       "      <td>Kisemlősök</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>fityma szűkület van ?</td>\n",
       "      <td>fityma szűkület van ? 15 éves fiú vagy és miko...</td>\n",
       "      <td>Valószínű. Esetleg még a fitymaféked lehet röv...</td>\n",
       "      <td>Egészség</td>\n",
       "      <td>Férfiak egészsége</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      short_question  \\\n",
       "0                           vállsérülés/gyógytorna ?   \n",
       "1  heti hányszor dehidtratálja a szervezet citrom...   \n",
       "2                a doctor who ki a vörös hajú lány ?   \n",
       "3  vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...   \n",
       "4                              fityma szűkület van ?   \n",
       "\n",
       "                                       long_question  \\\n",
       "0  vállsérülés/gyógytorna ? az lenne a kérdés , h...   \n",
       "1  heti hányszor dehidtratálja a szervezet citrom...   \n",
       "2                a doctor who ki a vörös hajú lány ?   \n",
       "3  vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...   \n",
       "4  fityma szűkület van ? 15 éves fiú vagy és miko...   \n",
       "\n",
       "                                              answer main_category  \\\n",
       "0  edzeni kénealapjába véve mindent váll, hát, tr...      Egészség   \n",
       "1  Nem értem miért akarod magad dehidratálni, nem...      Egészség   \n",
       "2                     Vagy Amy Pond vagy Donna Noble    Szórakozás   \n",
       "3  Egy törpenyuszihoz nagyon sok türelem kell. Pr...       Állatok   \n",
       "4  Valószínű. Esetleg még a fitymaféked lehet röv...      Egészség   \n",
       "\n",
       "           sub_category  split  \n",
       "0  Sérülések, balesetek  train  \n",
       "1           Bio életmód  train  \n",
       "2     Filmek, sorozatok  train  \n",
       "3            Kisemlősök  train  \n",
       "4     Férfiak egészsége  train  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "faq_dataframes[args.datasets[0]].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "cddeb707",
   "metadata": {},
   "outputs": [],
   "source": [
    "for dataset in args.datasets:\n",
    "    faq_dataframes[dataset] = faq_dataframes[dataset][[*args.train_column,'main_category', 'split']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "bf872874",
   "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>long_question</th>\n",
       "      <th>short_question</th>\n",
       "      <th>main_category</th>\n",
       "      <th>split</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>vállsérülés/gyógytorna ? az lenne a kérdés , h...</td>\n",
       "      <td>vállsérülés/gyógytorna ?</td>\n",
       "      <td>Egészség</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>heti hányszor dehidtratálja a szervezet citrom...</td>\n",
       "      <td>heti hányszor dehidtratálja a szervezet citrom...</td>\n",
       "      <td>Egészség</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a doctor who ki a vörös hajú lány ?</td>\n",
       "      <td>a doctor who ki a vörös hajú lány ?</td>\n",
       "      <td>Szórakozás</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...</td>\n",
       "      <td>vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...</td>\n",
       "      <td>Állatok</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>fityma szűkület van ? 15 éves fiú vagy és miko...</td>\n",
       "      <td>fityma szűkület van ?</td>\n",
       "      <td>Egészség</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                       long_question  \\\n",
       "0  vállsérülés/gyógytorna ? az lenne a kérdés , h...   \n",
       "1  heti hányszor dehidtratálja a szervezet citrom...   \n",
       "2                a doctor who ki a vörös hajú lány ?   \n",
       "3  vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...   \n",
       "4  fityma szűkület van ? 15 éves fiú vagy és miko...   \n",
       "\n",
       "                                      short_question main_category  split  \n",
       "0                           vállsérülés/gyógytorna ?      Egészség  train  \n",
       "1  heti hányszor dehidtratálja a szervezet citrom...      Egészség  train  \n",
       "2                a doctor who ki a vörös hajú lány ?    Szórakozás  train  \n",
       "3  vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...       Állatok  train  \n",
       "4                              fityma szűkület van ?      Egészség  train  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "faq_dataframes[args.datasets[0]].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "22b3589f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Egészség', 'Szórakozás', 'Állatok', 'Számítástechnika']\n",
      "{'Egészség': 0, 'Szórakozás': 1, 'Állatok': 2, 'Számítástechnika': 3}\n"
     ]
    }
   ],
   "source": [
    "target_names = faq_dataframes[args.datasets[0]].main_category.unique().tolist()\n",
    "target_dict = {k: v for v, k in enumerate(target_names)}\n",
    "\n",
    "print(target_names)\n",
    "print(target_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "dc2aa555",
   "metadata": {},
   "outputs": [],
   "source": [
    "for dataset in args.datasets:\n",
    "    faq_dataframes[dataset].main_category = faq_dataframes[dataset].main_category.apply(lambda x: target_dict[x])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "07dab225",
   "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>long_question</th>\n",
       "      <th>short_question</th>\n",
       "      <th>main_category</th>\n",
       "      <th>split</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>vállsérülés/gyógytorna ? az lenne a kérdés , h...</td>\n",
       "      <td>vállsérülés/gyógytorna ?</td>\n",
       "      <td>0</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>heti hányszor dehidtratálja a szervezet citrom...</td>\n",
       "      <td>heti hányszor dehidtratálja a szervezet citrom...</td>\n",
       "      <td>0</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a doctor who ki a vörös hajú lány ?</td>\n",
       "      <td>a doctor who ki a vörös hajú lány ?</td>\n",
       "      <td>1</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...</td>\n",
       "      <td>vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...</td>\n",
       "      <td>2</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>fityma szűkület van ? 15 éves fiú vagy és miko...</td>\n",
       "      <td>fityma szűkület van ?</td>\n",
       "      <td>0</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                       long_question  \\\n",
       "0  vállsérülés/gyógytorna ? az lenne a kérdés , h...   \n",
       "1  heti hányszor dehidtratálja a szervezet citrom...   \n",
       "2                a doctor who ki a vörös hajú lány ?   \n",
       "3  vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...   \n",
       "4  fityma szűkület van ? 15 éves fiú vagy és miko...   \n",
       "\n",
       "                                      short_question  main_category  split  \n",
       "0                           vállsérülés/gyógytorna ?              0  train  \n",
       "1  heti hányszor dehidtratálja a szervezet citrom...              0  train  \n",
       "2                a doctor who ki a vörös hajú lány ?              1  train  \n",
       "3  vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...              2  train  \n",
       "4                              fityma szűkület van ?              0  train  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "faq_dataframes[args.datasets[0]].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "2fff4802",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_dataframes_for_split(split, datasets):\n",
    "    return {ds: datasets[ds][datasets[ds].split == split].drop('split', axis=1) for ds in datasets}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "f4b361d8",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_dfs = get_dataframes_for_split('train', faq_dataframes)\n",
    "test_dfs = get_dataframes_for_split('test', faq_dataframes)\n",
    "valid_dfs = get_dataframes_for_split('val', faq_dataframes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "4b32f428",
   "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>long_question</th>\n",
       "      <th>short_question</th>\n",
       "      <th>main_category</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>vállsérülés/gyógytorna ? az lenne a kérdés , h...</td>\n",
       "      <td>vállsérülés/gyógytorna ?</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>heti hányszor dehidtratálja a szervezet citrom...</td>\n",
       "      <td>heti hányszor dehidtratálja a szervezet citrom...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a doctor who ki a vörös hajú lány ?</td>\n",
       "      <td>a doctor who ki a vörös hajú lány ?</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...</td>\n",
       "      <td>vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>fityma szűkület van ? 15 éves fiú vagy és miko...</td>\n",
       "      <td>fityma szűkület van ?</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                       long_question  \\\n",
       "0  vállsérülés/gyógytorna ? az lenne a kérdés , h...   \n",
       "1  heti hányszor dehidtratálja a szervezet citrom...   \n",
       "2                a doctor who ki a vörös hajú lány ?   \n",
       "3  vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...   \n",
       "4  fityma szűkület van ? 15 éves fiú vagy és miko...   \n",
       "\n",
       "                                      short_question  main_category  \n",
       "0                           vállsérülés/gyógytorna ?              0  \n",
       "1  heti hányszor dehidtratálja a szervezet citrom...              0  \n",
       "2                a doctor who ki a vörös hajú lány ?              1  \n",
       "3  vesz 2 törpenyuszi a szob 4-5 honap , de nagyo...              2  \n",
       "4                              fityma szűkület van ?              0  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_dfs[args.datasets[0]].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "9b48742e",
   "metadata": {},
   "outputs": [],
   "source": [
    "fields = {}\n",
    "\n",
    "for train_column in args.train_column:\n",
    "    fields[train_column] = {}\n",
    "\n",
    "    for key in args.datasets:\n",
    "        QUESTION = data.Field(tokenize = 'spacy', tokenizer_language = 'hu_core_ud_lg')\n",
    "        TARGET = data.LabelField(dtype = torch.long)\n",
    "\n",
    "        fields[train_column][key] = [('question', QUESTION),('main_category', TARGET)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "063f1565",
   "metadata": {},
   "outputs": [],
   "source": [
    "class DataFrameDataset(data.Dataset):\n",
    "\n",
    "    def __init__(self, df, fields, train_column, **kwargs):\n",
    "        examples = []\n",
    "        for i, row in df.iterrows():\n",
    "            label = row.main_category\n",
    "            text = row[train_column]\n",
    "                        \n",
    "            examples.append(data.Example.fromlist([text, label], fields))\n",
    "\n",
    "        super().__init__(examples, fields, **kwargs)\n",
    "\n",
    "    @staticmethod\n",
    "    def sort_key(ex):\n",
    "        return len(ex.text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "ba18423d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_dataframe_datasets(dfs, fields):\n",
    "    dataframe_datasets = {}\n",
    "    \n",
    "    for tc in args.train_column:\n",
    "        dataframe_datasets[tc] = {}\n",
    "\n",
    "        for key in dfs:\n",
    "            dataframe_datasets[tc][key] = DataFrameDataset(dfs[key], fields[tc][key], tc)\n",
    "    return dataframe_datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "4b008bea",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_datasets = create_dataframe_datasets(train_dfs, fields)\n",
    "valid_datasets = create_dataframe_datasets(valid_dfs, fields)\n",
    "test_datasets = create_dataframe_datasets(test_dfs, fields)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "8e856c84",
   "metadata": {},
   "outputs": [],
   "source": [
    "vectors = FastText(language='hu')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "63a2fc86",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([793866, 300])"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vectors.vectors.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "1a733bd1",
   "metadata": {},
   "outputs": [],
   "source": [
    "for tc in args.train_column:\n",
    "    for ds in args.datasets:\n",
    "        fields[tc][ds][0][1].build_vocab(train_datasets[tc][ds], vectors = vectors)  \n",
    "        fields[tc][ds][1][1].build_vocab(train_datasets[tc][ds])  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "b3441faa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "20"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "field_ = fields['short_question']['lemmatized'][0][1]\n",
    "field_.vocab.stoi['kutya']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "949fcde0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3362"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vectors.stoi['kutya']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "d151deba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.equal(field_.vocab.vectors[20], vectors.vectors[3362])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "36328703",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_iterators(train_ds, valid_ds, test_ds):\n",
    "    iterators = {}\n",
    "    for tc in args.train_column:\n",
    "        iterators[tc] = {}\n",
    "\n",
    "        for ds in args.datasets:\n",
    "            iterators[tc][ds] = data.BucketIterator.splits(\n",
    "                (train_ds[tc][ds], valid_ds[tc][ds], test_ds[tc][ds]),\n",
    "                batch_size = args.batch_size,\n",
    "                sort_key = lambda x: len(x.question),\n",
    "                sort_within_batch = True,\n",
    "                device = device)\n",
    "    return iterators"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "6c444ade",
   "metadata": {},
   "outputs": [],
   "source": [
    "iterators = create_iterators(train_datasets, valid_datasets, test_datasets)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "d1ec1af0",
   "metadata": {},
   "outputs": [],
   "source": [
    "class LSTMClassifier(nn.Module):\n",
    "    def __init__(self, input_dim, embedding_dim, hidden_dim, output_dim, field):\n",
    "        super(LSTMClassifier, self).__init__()\n",
    "        \n",
    "        self.embedding = nn.Embedding(input_dim, embedding_dim)\n",
    "        self.embedding.weight.data.copy_(field.vocab.vectors)\n",
    "        \n",
    "        self.lstm = nn.LSTM(embedding_dim, hidden_dim)\n",
    "        \n",
    "        self.fc = nn.Linear(hidden_dim, output_dim)\n",
    "\n",
    "        \n",
    "    def forward(self, text):\n",
    "        embedded = self.embedding(text)\n",
    "        output, hidden = self.lstm(embedded)\n",
    "        y = self.fc(output[-1])       \n",
    "        log_probs = F.log_softmax(y.squeeze(0), dim=1)\n",
    "        return log_probs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "5ff2ff14",
   "metadata": {},
   "outputs": [],
   "source": [
    "def class_accuracy(preds, y):\n",
    "    rounded_preds = preds.argmax(1)\n",
    "    correct = (rounded_preds == y).float()\n",
    "\n",
    "    acc = correct.sum() / len(correct)\n",
    "    return acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "8ea140db",
   "metadata": {},
   "outputs": [],
   "source": [
    "def show_data_to_model(model, iterator, optimizer, criterion, bar, train):\n",
    "    epoch_loss = 0\n",
    "    epoch_acc = 0\n",
    "       \n",
    "    total_predicted = []\n",
    "    \n",
    "    for idx, batch in enumerate(iterator):      \n",
    "        if train:\n",
    "            optimizer.zero_grad()\n",
    "\n",
    "        predictions = model(batch._get_field_values(batch.input_fields))\n",
    "        loss = criterion(predictions, batch.main_category)\n",
    "        acc = class_accuracy(predictions, batch.main_category)\n",
    "        \n",
    "        if train:\n",
    "            loss.backward()\n",
    "            optimizer.step()\n",
    "\n",
    "        epoch_loss += loss.item()\n",
    "        epoch_acc += acc.item()\n",
    "\n",
    "        _, predicted = torch.max(predictions.data, 1)\n",
    "        total_predicted += predicted.tolist()\n",
    "        \n",
    "        bar.set_postfix(loss=(epoch_loss / (idx + 1)), acc=(epoch_acc / (idx + 1)))\n",
    "        bar.update()\n",
    "\n",
    "    return epoch_loss / len(iterator), epoch_acc / len(iterator), total_predicted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "a8a3bfd9",
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_model(model, iterator, optimizer, criterion, bar):\n",
    "    model.train()\n",
    "    loss, acc, _ = show_data_to_model(model, iterator, optimizer, criterion, bar, True)\n",
    "    \n",
    "    return loss, acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "226d1194",
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluate_model(model, iterator, optimizer, criterion, bar):\n",
    "    model.eval()\n",
    "    with torch.no_grad():\n",
    "        loss, acc, predicted = show_data_to_model(model, iterator, optimizer, criterion, bar, False)\n",
    "            \n",
    "    return loss, acc, predicted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "d6a08b2c",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e7c8d0bb863347cbbd1ee62f3cb401c1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'long_question lemmatized' epoch:   0%|          | 0/20 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'long_question lemmatized' train:   0%|          | 0/1536 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'long_question lemmatized' valid:   0%|          | 0/330 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a0afed65e9284ebfbe2fa4d7ed5b00f3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'long_question lemmatized_filtered' epoch:   0%|          | 0/20 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'long_question lemmatized_filtered' train:   0%|          | 0/1536 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'long_question lemmatized_filtered' valid:   0%|          | 0/330 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "99f63cd2cf0f40c283e061bd2ab8e576",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'short_question lemmatized' epoch:   0%|          | 0/20 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'short_question lemmatized' train:   0%|          | 0/1536 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'short_question lemmatized' valid:   0%|          | 0/330 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "00275b0040474e48b23ec58f17971f83",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'short_question lemmatized_filtered' epoch:   0%|          | 0/20 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b1a6c909d93d4d52adde25d85d46bcbb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'short_question lemmatized_filtered' train:   0%|          | 0/1536 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4a6fa7f5eefe4b7b970d00f9d73fb9e6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'short_question lemmatized_filtered' valid:   0%|          | 0/330 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_losses = {}\n",
    "train_accuracies = {}\n",
    "\n",
    "valid_losses = {}\n",
    "valid_accuracies = {}\n",
    "\n",
    "for tc in args.train_column:\n",
    "    train_losses[tc] = {}\n",
    "    train_accuracies[tc] = {}\n",
    "\n",
    "    valid_losses[tc] = {}\n",
    "    valid_accuracies[tc] = {}\n",
    "    \n",
    "    for ds in args.datasets:   \n",
    "        best_valid_loss = float('inf')\n",
    "    \n",
    "        train_losses[tc][ds] = []\n",
    "        train_accuracies[tc][ds] = []\n",
    "\n",
    "        valid_losses[tc][ds] = []\n",
    "        valid_accuracies[tc][ds] = []\n",
    "    \n",
    "        train_iterator, valid_iterator, _ = iterators[tc][ds]\n",
    "    \n",
    "        question_field = fields[tc][ds][0][1]\n",
    "        input_dim = len(question_field.vocab)\n",
    "        output_dim = len(target_names)\n",
    "    \n",
    "        model = LSTMClassifier(input_dim, args.embedding_dim, args.hidden_dim, output_dim, question_field)\n",
    "        optimizer = optim.Adam(model.parameters(), lr=args.learning_rate)\n",
    "        criterion = nn.NLLLoss()\n",
    "\n",
    "        model = model.to(device)\n",
    "        criterion = criterion.to(device)\n",
    "    \n",
    "        epoch_bar = notebook.tqdm(desc=f\"'{tc} {ds}' epoch\", total=args.num_epochs, leave=True)\n",
    "        train_bar = notebook.tqdm(desc=f\"'{tc} {ds}' train\", total=len(train_iterator), leave=False)\n",
    "        valid_bar = notebook.tqdm(desc=f\"'{tc} {ds}' valid\", total=len(valid_iterator), leave=False)\n",
    "        \n",
    "        for epoch in range(args.num_epochs):\n",
    "            train_bar.n = 0\n",
    "            valid_bar.n = 0\n",
    "\n",
    "            train_bar.refresh()\n",
    "            valid_bar.refresh()\n",
    "\n",
    "            train_loss, train_acc = train_model(model, train_iterator, optimizer, criterion, train_bar)\n",
    "            valid_loss, valid_acc, _ = evaluate_model(model, valid_iterator, optimizer, criterion, valid_bar)\n",
    "            \n",
    "            train_losses[tc][ds].append(train_loss)\n",
    "            train_accuracies[tc][ds].append(train_acc)\n",
    "\n",
    "            valid_losses[tc][ds].append(valid_loss)\n",
    "            valid_accuracies[tc][ds].append(valid_acc)\n",
    "    \n",
    "            if valid_loss < best_valid_loss:\n",
    "                best_valid_loss = valid_loss\n",
    "                torch.save(model.state_dict(), args.model_state_file + \"_\" + \\\n",
    "                                                        ds + \"_\" + \\\n",
    "                                                        tc + \".pth\")\n",
    "            epoch_bar.update()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "d8cbb525",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJMAAAjICAYAAABvmu7EAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOzdd5hU1f3H8ffZpRdFpCnsWqmigBBbVIzGHjX2jt2oIWoSY4lGjS0afybR2KNGY09swYRojL3F2LChIDZEAVGK9LJ7fn+cWVnWXXaB3bk7u+/X89xnZu69c+c7s2XOfOacc0OMEUmSJEmSJKkuirIuQJIkSZIkSYXDMEmSJEmSJEl1ZpgkSZIkSZKkOjNMkiRJkiRJUp0ZJkmSJEmSJKnODJMkSZIkSZJUZy2Wt/HVV1/t1qJFi5uAgRg8raxy4O0lS5YcO3To0C+yLkaSJOWX7amCYrtNkqQ6WG6Y1KJFi5t69OjRv2vXrjOKiopivopqSsrLy8O0adMGTJky5SZgz6zrkSRJ+WV7qnDYbpMkqW5q+3ZsYNeuXb+24bPyioqKYteuXWeRvo2UJEnNj+2pAmG7TZKkuqktTCqy4bPqcq+h3dolSWqebE8VENttkiTVrlG/UU6ZMqW4X79+A/r16zegS5cug7p167ZJxe0FCxaE5d33mWeeaXfkkUeWrMjj9ezZc+PJkycvd+ifJElSIcl3e0qSJDV9jTo46dGjR9l77703FuBnP/vZ2h06dCi74IILplZsX7x4MS1btqz2vttuu+28bbfddl6eSpUkSWqUmmp7anl1S5KkhtWoeyZVZ99991332GOP7bX55pv3Oemkk3o9+eST7YYMGdKvf//+A4YMGdLvjTfeaA3wj3/8o+P3vve9DSE1nPbff/91N9tss769evXa+KKLLupW2+Ocf/753Xv37r1R7969N7rgggu6AXz99ddF22233YZ9+/Yd0Lt3743+9Kc/rQFw0kkn9dxggw026tOnz4Djjz++V0M+f0mSpFXVkO2pQw89tHTgwIH9N9xww41++tOfrl2x/umnn243ZMiQfn379h2w8cYb958xY0bRkiVLOP7443v16dNnQJ8+fQZcfPHF3WDZ3uLPPPNMu80226xvRQ0HH3zwOt/97nd777PPPuuNGzeu1dChQ/sOGDCg/4ABA/o/9thj7Sse75xzzunep0+fAX379h1w0kkn9XznnXdaDxgwoH/F9rfeeqv1Rhtt1B9JkrTCGnXPpJp88MEHbZ5//vnxLVq0YPr06UX/+9//3mvZsiUPPfRQx9NPP73Xo48++kHV+0yYMKHNCy+8MG7mzJnF/fv3H/iLX/xiWuvWraudv+DZZ59td9ddd6356quvvhtjZOjQof132GGH2e+//37rHj16LH7qqacmAHz11VfFU6dOLR49evQaH3744dtFRUV8+eWXxQ39/CVJklZVQ7Wnfve7333WvXv3siVLlrDVVlv1femll9oOGjRowaGHHrrBnXfe+cHw4cPnTZ8+vahDhw7lV1xxRddPPvmk9TvvvDO2ZcuWTJ06tdZ21JtvvtnupZdeeq9Dhw5x9uzZRc8+++z4du3axbfeeqv1wQcfvP7bb7/97l//+tfV/vnPf67x6quvvtexY8fyqVOnFnfv3r2sY8eOZS+88ELbrbbaav4NN9zQ5ZBDDvmqPl9TSZKaizqHSUcfTcnbb9OuPh984EDm3XILn67o/fbZZ58ZLVqk0qdPn1584IEHrvfxxx+3CSHExYsXVzv2f6eddprZtm3b2LZt2yWdO3dePGnSpBYbbLDB4ur2feqppzrstttuM1dbbbVygN13333Gk08+2XHPPfecdfbZZ5eceOKJPffaa69Zu+yyy5zFixfTunXr8oMOOmid3XfffdaBBx44a0WfjyRJah6aQ3vqtttu63zrrbd2WbJkSZg2bVrLN954o00IgW7dui0ePnz4PIDOnTuXAzzxxBOrnXDCCdMqhqt17969rLa6d9lll5kdOnSIAIsWLQrHHHPMOmPHjm1bVFTEJ5980hrgscceW+2www77smPHjuWVj3vkkUd++ac//anLZptt9unf//73NV5++eV3V/R1kyRJBTjMDaBDhw7lFdfPOOOMnsOHD5/9/vvvv/Pwww9PWLRoUbXPqfK3ZsXFxSxZsqTGCSdjrP6EK5tsssnC1157bezGG288/+yzz+552mmnrdWyZUvGjBnz7r777jvzoYce6rTddtv1XoWnJkmSlBcN0Z567733Wl199dXdn3766fHjx48fu/32289asGBBUYyREMK3Glg1rS8uLo7l5am8+fPnL1NL+/btv6n74osv7t6tW7fF77777ti33npr7OLFi4sqHfdb9R9xxBEznnzyydXvueeeThtvvPG8Hj161BpeSZKkb6tzz6SV+cYrH77++uviXr16LQK44YYbutTHMbfffvs5Rx999LoXXnjhlBgjo0ePXuPWW2/98OOPP27ZrVu3JSeddNL0jh07lt92221rzpo1q2jOnDlFBx544KzttttuTp8+fTaujxokSVLT09TbUzNmzChu27ZteefOncs+/fTTFk899dTqw4cPnz1o0KAFU6dObfX000+3Gz58+LwZM2YUdejQofz73//+19dff33X3XfffXbFMLfu3buX9erVa9Hzzz/f7oADDvj6r3/96xo1Pd6sWbOKe/Xqtai4uJirr756zbKylA3tsssuX1988cVrH3fccdMrD3Nr165dHD58+Kyf/exnpVdfffXHK/s8JUlq7gqyZ1JlZ5xxxpTzzz+/16abbtqvogGxqrbeeut5hxxyyFebbrpp/6FDh/Y//PDDp333u9+d/+qrr7YdPHhw/379+g247LLL1jr33HMnz5w5s3iXXXbp3adPnwHbbLNN34suuqhRNhIlSZJqUl/tqS233HL+wIED5/Xu3Xujww8/fN2hQ4fOAWjTpk288847Pzj55JNL+/btO2C77bbrM2/evKKf/vSn03r16rWoX79+G/Xt23fAzTff3Bng3HPP/fz0008vHTp0aN/i4uLqu4wDp5566hd33333moMGDeo3fvz4Nm3bti0H2G+//b7eddddZ1a02y688MIeFfcZMWLEdIB99tnn65V+opIkNXOhpiFdAG+88cbHgwYN+jKP9TRZb7zxRpdBgwatm3UdkiQpv2xPNS7nnntu91mzZhVfeeWVn9e0j+02SZKWryDP5iZJkiStqB133HGDTz75pPXTTz89PutaJEkqZIZJkiRJahYee+yxD7KuQZKkpqDg50ySJEmSJElS/hgmSZIkSZIkqc4MkyRJkiRJklRnhkmSJEmSJEmqs0YdJm222WZ977///tUqr7vgggu6HXbYYaXLu88zzzzTDmD48OEbfvnll8VV9/nZz3629rnnntu9ruslSZIKVb7bU5Ikqelr1GHS/vvv/9Xdd9/dufK6+++/v/Nhhx02vS73f/rppyd06dKlrGGqkyRJavyaanuqvLycsrJGV5YkSc1Cow6TDj/88BmPP/746vPnzw8A48aNa/XFF1+03GmnneYceuihpQMHDuy/4YYbbvTTn/507eru37Nnz40nT57cAuCMM87ose666w7caqut+rz//vuta3vsF154oe2gQYP69enTZ8COO+64wbRp04oBLrroom4bbLDBRn369Bnwgx/8YH2Af/7znx369es3oF+/fgP69+8/YMaMGY36dZUkSc1HvttTd9111+qbbLJJv/79+w/Yaqut+nz66actAGbNmlW03377rdunT58Bffr0GXDrrbd2ArjvvvtWGzBgQP++ffsO2HLLLfvAt3s99e7de6Nx48a1GjduXKv1119/o8MOO6x0o402GvDBBx+0quk5PP300+2GDBnSr2/fvgM23njj/jNmzCgaOnRo3xdeeKFtxT6bbrppv5deeumb25IkqW4adejRo0ePskGDBs29//77Vwe47bbbOu+5554zioqK+N3vfvfZ22+//e577733zvPPP99xeQ2BZ599tt2DDz7Y+a233hr7j3/8Y8Ibb7zRvrbHPvLII9e75JJLJo0fP37sRhttNP+MM85YG+Cqq67q8fbbb48dP3782FtvvfUTgCuuuKLHVVdd9cl777039r///e97HTp0KK+v10CSJGlV5Ls9teOOO84ZM2bMe+++++7Y/fbbb/oFF1zQA+DMM89ca7XVVisbP3782PHjx4/dfffdZ3/++ectRo4cue4DDzzwwbhx48Y+9NBDH9T2fD7++OM2Rx111Ffvvvvu2D59+iyq7jksWLAgHHrooRv84Q9/mDhu3LixTz/99LgOHTqUH3nkkV/edNNNXQDefPPN1osWLQqbb775/JV7ZSVJar5a1HnPo48u4e2329Xrow8cOI9bbvl0ebsccMAB0++99941DjvssJkPPPBA55tuuuljSA2hW2+9tcuSJUvCtGnTWr7xxhttamoMPPnkkx122223mR07diwH2GmnnWYu7zG/+uqr4tmzZxfvvvvucwCOO+64r/bff//1Afr27Tt/7733Xm/PPfeceeihh84E2GKLLeacdtppJQcccMD0gw8+eMYGG2xgmCRJkr6tGbSnPvroo1Y//OEPe02bNq3lokWLikpKShYCPPPMM6vdc889H1bs17Vr17K77rpr9c0222x2v379FgF079691nFra6211qIddthhbsXt6p5DCIFu3botHj58+DyAzp07lwMceeSRMy6//PK1Fi5cOOn666/vcsghh3xZ2+NJkqRva9Q9kwAOPfTQmc8///xqzz33XLsFCxYUbb311vPee++9VldffXX3p59+evz48ePHbr/99rMWLFiw3OcSQqiXep588sn3f/zjH0979dVX2w8aNGjA4sWLueSSS6bcdNNNn8yfP79oq6226v/666+3qZcHkyRJqgf5bE+NHDmy9KSTTvpi/PjxY6+++upPFi5cWAQQY/zW/atbB9CiRYtYXr70u7mFCxd+s1O7du2+2VDTc8gdN1Y9bseOHcu32Wabr++6665Oo0aN6nzMMcfUad4oSZK0rLr3TKrlG6+Gsvrqq5dvscUWs4899th199lnn+kAM2bMKG7btm15586dyz799NMWTz311OrDhw+fXdMxtt9++zlHH330uhdeeOHkxYsXh8cee6zTEUccMa2m/ddcc82y1VZbreyRRx7psMsuu8y5+eab19xyyy3nlJWV8cEHH7TaY489Zu+0005z1l577c6zZs0qnjp1aovNNtts/mabbTb/pZdeav/222+3GTJkyIKGeD0kSVIBawbtqdmzZxeXlpYuBrj11lvXrFi/3Xbbff273/2u2y2512DatGnF3/ve9+b+/Oc/X+e9995r1a9fv0VTp04t7t69e9m66667cPTo0Z0AnnvuuXafffZZtfMz1fQcBg0atGDq1Kmtnn766XbDhw+fN2PGjKIOHTqUt2zZkhNOOOHLfffdd8PvfOc7c+rSE0qSJH1bo++ZBHDQQQdNHzduXNvDDz98OsCWW245f+DAgfN69+690eGHH77u0KFD5yzv/ltvvfW8vffee/rAgQM3+sEPfrDBZpttttz9Af785z9/dMYZZ/Tq06fPgDfffLPtpZde+vmSJUvCIYccsl6fPn0GDBw4cMCPfvSjqV26dCn77W9/2613794b9e3bd0Dbtm3L99tvv1n19dxVuEIIMYSwYZ4f86IQwpchhCkNcOynQgjH1vdxG1oIYZsQwrgGOO4K/3xDCNuFECbVdy2FJoRwfQjhV1nXITU3+WpPnX322Z8ffPDBGwwdOrTvmmuuuaRi/W9+85vJM2fOLK5oM40ePbrj2muvveSqq676eO+9996wb9++A/bee+/1AUaMGDFjxowZxf369Rtw9dVXd11nnXWq/ZKupufQpk2beOedd35w8sknl/bt23fAdttt12fevHlFANtss8289u3blx111FEOcdMyQgi3hhAuyvNjfjeE8H4IYU4I4Yf1fOwjQwjP1ecx8yWE8E4IYbt6PuZK/XxDCB+HEL5fn7UUmoZqT6uwhRi/1QP4G2+88cbHgwYN8o22HrzxxhtdBg0atG7WdRSqEMLHwLExxv9kXUtd5brX944xTsjT45UA44F1Yoxf1NMxY4wx5K4/BdwRY7ypPo7dUPL1uld+nBDCrcBTMcZbq9nvfIAY4/m5RtEdMcZeDVlbYxJCOJL0t7t1Ax17uxjjkfV9bKk+2Z5qXD7++OOW2223Xd8PPvjg7eLi4mr3acrtttz7+SCgR4xxYcblNCq59/NJMcZz8viYjwOjYoxX1tPxbiXXJmnI9+D6lK/XvfLj5Npk58cYt6tmv3VJr+G6udsfU2CfQ1ZVQ7anK3++UGEriJ5JkupkHeCrlQmSQgh1H/IqSVKBuvrqq9fcYost+p977rmf1RQkNWW5D8nbABHYM8+PbVujeusA76zMHX1NJWXJMEkFLYTQOoTwhxDC57nlDyGE1rltXUII/wghzAwhTA8hPBtCKMptOyOE8FkIYXYIYVwIYYdqjr1FCGFKCKG40rq9Qwhv5q5vFkJ4MXf8ySGEq0MIrWqoc5khYlW7HYcQ+oUQHsvVOS6EcEClbbuFEMbmav0shHBaNcf/PvAYsHaum/StufV75roJz8zV0L/SfT7OvQ5vAnNXtEESQjg6hPBuCGFGCOHREMI6lbbFEMJJuW7bs0MIF4YQNsi9Xl+HEP5a8VpVDP0KIZweQvgi91r+MPe8x+dek19WOnaNr3sI4Zncbm/kXocDQ6WhZbnbcyotC3Pf0Fb8Lv1fCGFiCGFqSEOx2lZ63F/kHu/zEMLRK/JaLec17J/7uczM/Zz2rLTt1hDCNSGEf+Zew5dCCBtU2r5T7ndlVgjh2hDC06GWYYghhOLcc/wyhPBhCOHHuZ9Vi9z2ZbpxhxDODyHcUen2FiGEF3L1vhEqdT/P/U5/mKv1oxDCobnft+uBLXOv98xKz+2iSvc9LoQwIfezHhVCWLvSthhCOCH3uzQj95r4bZaklTJy5MivpkyZ8ubRRx89I+taMjIC+C9wK3BE5Q0hhJIQwgMhhGkhhK9CCFdX2nZc7j1/dkhtkk1z65cZ8l35/3ul9/czQhp+/+cQwhohtc2m5f6n/yOE0KvS/TuHEP6ce6+dEUJ4KLf+7RDCHpX2a5l7Lxtc9Qnm6vxBpdstcvtW1Py3kNp3s0IIz4QQNqruhQrVDBGr/HyX124Iy2mDVjneB8D6wMO598nWIYS1c++F03PvjcdV2v/8EMJ9IYQ7QghfA0dWV3tNwvLbm7fm2hP/ytXyfAihR0ht6xkhhPdCCEMq7f9xSG2jN0MIc0MIN4cQuufuPzuE8J8QwhqV9q/2dQ8hHA8cCpyee9yHKx3/+7nrM8PSttvc3M9h3dy2H4QQxuT2eSGEsEmlxxwSQngtV8+9wCqfJCks/7NHxe/8z8PSNu1Rle67Zgjh4ZDawi+HND1FrcMQQwg75l7/WSG1e58OuTZf+HZbbd2wbNtu9dzPZnJInyMuCrnPNiGEDXPHmhXS38i9ufXLbU/n9lnpNqyaDsMkFbqzgS2AwaQu25sBFV1kfw5MAroC3YFfAjGE0BcYCXwnxtgR2Bn4uOqBY4z/BeYC21dafQhwV+56GfBToAuwJbADcNKKPoEQQntSEHQX0A04GLg2LG3c3Az8KFfrQOCJamr9D7Ar8HmMsUOM8cgQQh/gbuDU3GswmtRYqRx4HQzsDnSKMS6pclhq6oIa0pj+XwL75I79bO6xKtsFGEr6+ZwO3EhqLJTknsfBlfbtQXqD7wmcC/wJOCx3/22Ac0MI6+f2rfF1jzFum9tnUO51uLfK87k3t74DsDbwYaW6LwP6kH6XNqxUCyGEXYDTgB2B3sD3qxz3yOqGuOW2nR9jPL/q+hBCS+Bh4N+kn/tPgDtzv58VDgZ+DawBTAAuzt23C3AfcBawJjAO2Kq6x6/iOOAHwBBgGLBfHe5TUW9P4J/ARUBn0utxfwiha+53+Cpg19zv6VbAmBjju8AJwIu5171TNcfdHvgNcACwFvAJcE+V3X4AfIf0N34A6W+WGOOtDnGTpBUyArgzt+wcQugO6csG4B+k/8Hrkt4D78lt2x84P3ff1Ug9mr6q4+P1IL1nrAMcT/rs8efc7VJgPnB1pf1vB9oBG5HeG3+fW/8XUrugwm7A5BjjmGoe826WbWPsDHwZY3wtd/tfpPfybsBrpNdiZdTYbqCGNmjVA8QYNwAmAnvk3icX5uqfRGqn7AdcEpb90nMvUhugU3W119QmqUN7E9J77DmkNtZC4EXSa1TR7vhdlcPuS2ob9QH2IL22v8ztXwScXGnfal/3GOONueu/zb0Ge1BFjLFTpfbblaR252chBYS3AD8itYduAEblAp9WwEOk36nOwN9y9VYc86nqhrjltn1cMcStGsv77AHpd3510u/DMcA1lUK1a0ifLXqQwtxlAt3q5Np897P05/IB8N3a7lfJbcAS0u/oEGAnoOLLxwtJ7dA1gF7AH6H29vSqtGFzx/dLwSaitjCpvLy83B/2Ksq9huW17qiVcShwQYzxixjjNNI/rcNz2xaTPpyuE2NcHGN8NqZJwsqA1sCAEELL3BvGBzUc/5sGSQihI6nxcjdAjPHVGON/Y4xLYowfk97Ahq/Ec/gB8HGM8c+5Y71GetOo+KC/OFfrajHGGZUaQ7U5EPhnjPGxGONi4P+AtiwbOlwVY/w0xjh/BWv+EfCbGOO7uRDqEmBwqNQ7Cbgsxvh1jPEd4G3g3zHGD2OMs0gNiiGV9l0MXJyr8x7Sm+WVMcbZufu/A2wC9fO6h/Tt4F2k8fA3hBACKWj5aYxxeoxxdu45HZS7ywHAn2OMb8cY55Ia1atqC6ADcGmMcVGM8QlSQ75yA/iBGOP/cq/xnaSGC6Tfw3dijA/ktl0F1GXS9QOAP+R+5tNJIU5dHQaMjjGOjjGWxxgfA17J1QLpf9zAEELbGOPk3M+tLg4FbokxvpZrRJ9F6sm0bqV9Lo0xzowxTgSeZOnrIBUK21MFpKm220IIW5NCnL/GGF8lfSg9JLd5M1J48YsY49wY44IYY0WPiWNJH/RfjsmEGOMndXzYcuC8GOPCGOP8GONXMcb7Y4zzcu+1F5N7Dw8hrEX6YuyEXHtncYzx6dxx7gB2CyGslrt9OCkkqM5dwJ4hhHa525W/CCTGeEuufbGQ9H4+KISweh2fD7laa2s31NQGre24JcDWwBm5n8EY4CaWtm0hfUHzUO69eEXab7W1NwEezLWzFgAPAgtijH+JMZYB97Js2w3gjzHGqTHGz0gBz0sxxtdzr+2Dlfevp9f9QNLPc99cm/E44IYY40sxxrIY422kEGyL3NKS1O5ZHGO8D3h5RR6vBsv77AHpZ39B7jFHA3OAvrnAdl/S38O8GONYUtBTm92AsTHG+3LP+Q/Urc1HLizeFTg193f9BSmgrfx7ug6wdpW/+dqsShtWTUhtYdLb06ZNW90G0MorLy8P06ZNW530YVr1b23St2gVPsmtA7iclIT/O6ThN2cCxDSR3KmkN7IvQgj3hErDaqq4C9gn1311H+C1igZUCKFPSF2Yp4TU1fgSUgiyotYBNs91E50Z0lCgQ0nfWkB649kN+CTXFXXLOh53mdcmxlgOfEr6pqTCyp6ieh3gykr1TgdClWNPrXR9fjW3O1S6/VWuoVKxrbr7d4B6e90vBjqy9BuzrqRvQl+t9Jweya2H9FpWfq3q2ohenrWBT3M/l8rHrfwaVm4szGPpa7ZMPbkGal3OErcqz2MdYP8qv6dbA2vlArYDSb2QJue6Nfer43Gr/p7OIX3jXZfXQSoUtqcKRBNvtx1B+mKnYjL4u1jaM6IE+CRW00s5t62mL91qMy0XTAAQQmgXQrghhPBJ7j38GaBT7oN2CTA9xvitIYgxxs+B54F9QwidSB+Qq+1RlGvnvQvskQuU9iQXJoU03PvSEMIHucf/OHe3FW1H1NZuqLYNWgdrk16D2ZXWVW0brErbbXntTVixtlud96+P1z2kIXZXA3vnQpyK5/TzKs+phPQ6rg18ViXEq6/2W02fPSC1aSv/HVW0W7oCLVj251eXn2V1bb66/g6sQwrUJld6fW4g9SaCNHIgAP/LDVWr6zQOq9KGVROy3DlSlixZcuyUKVNumjJlykAcEreyyoG3lyxZUnCnVC8Qn7PsxIWluXXk3oh/TnqT2Qh4MoTwcozx8RjjXcBduW+4biB1VT686sFjjGNDCJ+QGi3LfLMFXAe8DhwcY5wdQjiVmocNzSU1OipUfuP+FHg6xrhjdXeMMb4M7JXrUjoS+CvpjbI2nwMbV9zIfYtWAnxW+fB1OE51PiX1JFrZruGrYkVe928JIRxE+ubkO7lveAC+JDV6Nsp9u1bVZJZ9zUtXpvAqPgdKQghFld6MS0ln5KvNZFJ3ZOCbn21dzhBX2/Oo7ff09hjjcVQjxvgo8GhI80VcRBqqWDHJ6/JU/A0D33TDX5Nlf0+lgmZ7qqA0yXZb7n/zAUBxSPMXQeql3SmEMIj0P740hNCimkDpU6Cm+U7m8e33jcpfblR9D/g50BfYPMY4JaQ5j14nfaD9FOgcQugUY5xZzWPdRuol1YLUO2d57xMVPcuLSL06Ks5IdQhpmNj3SYHG6sCM3ONXtcx7Ygih8nvictsNy2uDLqdmSO+JnUMIHSsFSqXUX9utxvZmA6vtdV/ucwohdCX1dBoZY3y90qaK9ujF1dxnONAzhHTqsNzqUlY+GK1Q42ePWkwjDTfrxdK2Xl3a88u03Sq15yvU1nZbCHSpLiiOMU4h9e6q6Ln4nxDCM7H2M7itShtWTchyw6ShQ4d+QZ7P9CAtR8sQQuWJ85aQGgvnhBBeJr0RnUvqCk1Iky++R3rT+Jo0vK0sN563J+kbrgWkxsDyGvd3kXqwbEn6BqdCx9xx5+R6YZxIeqOozhhSD6ebSGn+MSz99uYfwKUhhMNZOlfMYFK32A+A/YF/xBhn5b7NKaNu/gqcGdI4+2eAU0hvKC/U8f7Lcz1wYQhhTIzxnVw35Z1ijH+rh2PXprbXfSppMstvvRHmvtX6I7BjpW+1iDGWhxD+BPw+hDAyxvhFSHMEDcyFJH8lTRz6F1Ij6Lx6eB4vkRoAp4cQriCNf9+DNDdQbf4JXB3S3FX/IPUI6rHceyR/BU4OIfwj99hVvykdAxwUQvgXaR6A/UjftEL6u3o5hLAz8B/SN11bkF7nxcDmwOOkv6c5LP09nQr0CiG0ijEuqqamu4B7Qgh3kb5JvoTUTf7jOjwfqSDYnlIj8EPS/+WNgcr/i/9KmgvpdNKH1ktDCOfl9h0aY3yeNMzqdyFNFPwaKVhanOupPQY4JITwDmnunOGkIdA16Uh6n5gZQuhMpffTGOPk3PvPtSGEH5PeS7aMMVZMBvwQcC1pDqLf1vJ87yH1Qu7Msl8EdiS1hb4ifQC/ZDnHeAPYKBd4vUelIe61tRtqaoPWUjMxxk9DCC8AvwnphCt9SG3Gw5Z/zzqpsb0Z0xyHDam2172i7fYtIU0kfT9wZ6wydw/pi6sHQwj/Af6XO/Z2pHbvi6TPCieHEK4h/Q/ejDRcflXU+NljeWKMZSGEB4DzQ5o8u5T0tzexlrtWtPn2AUYBP2bZNt8Y4IwQQikwizRdQMVjTg4h/Bu4IoTwK9Lf1HpArxjj0yHNh/ZijHESKdyrmA4EltOeZtXasGpC/HZMhWQ0qQFSsZxP6gHxCvAm8BapkVNxlqjepA+9c0hvKNfGGJ8ifRN3KelbpSmkrp7fnC2sGneT3pieqNQ1HNIExIcAs0lvZlXf4Cr7PanxNpX0zdo3PXpy3zztRBq//HmupstydULqMfVxLkg6gTo2KGKM43L7/pH0XPcgTfBY3Qf6FRJjfDBX4z25ut4m9d7Kh9pe9/OB23LdeQ+osm0v0kSAz4WlZwX5V27bGaQ3zP/mntN/SN+eEmP8F2mM+hO5fb41CfqKyv0c9iS9bl+SGsgjYozv1eG+X5JCxt+SGmYDSH8HC2u565+AR0kN5NeAB6ps/xXpQ8IM0hwAleeY+JT0+v2SFN59CvyC9D5SRPoG9nPSkMfhLJ2M/gnSt3dTQgiV/34qjvt47nHvJ32Q2YClY/klSfXjCNLcfxNjjFMqFtKwoUNJPUT2IE3SO5HUu+hAgNwXRReT3hNmk0KdzrnjnpK738zccR6qpY4/kOZv/JJ0VrlHqmw/nPQFxXvAF6RpCcjVMZ/0XrEe337/WkaMcTKp7bcVy7YT/kIajvMZMDZXQ03HGA9cQGoPvA9UnU+mxnYDNbdB6+Jg0iTon5N645wX0zyFq6QO7c2GVNvrfjNpftCZIXcGv0p6kXo6nxqWPSNvaYzxFVLPmqtJbZcJ5M5wl2tn7ZO7PYP0+7zc35s6Wt5nj9qMJPXKmkKa8+tuamm7VWrzXUpq8/UmfSFesf0x0u/4m8CrpNCwshFAK9LrPoM0kfpauW3fAV4KIcwhBVWnxBg/ym07nxra06vShlXTEmLtc8FJkhqxkCYUnwQcGmOs8zduIU1y/RHQsoZ5MiRJajRCCOcCfWKM9dFTR8pUCOEyoEeMsdazulW531PAHTHGmxqkMKmO7JkkSQUohLBzCKFTSJPD/5L0rXKN37BKklTIcsPijgFuzLoWaWWEEPqFEDYJyWak3+cHs65LWlmGSZJUmLYkzcVQMYTxhzHG+SGE66t0A69Yrs+2XElZCSHcEkL4IoRQ7RnKch9srgohTAghvBlC2DTfNUrLE0I4jjS8+l+V5lCSCk1H0lC7uaT5yq4A/h5C2KaGttucTKuVauEwN0mSpCYshLAtae6Wv8QYB1azfTfgJ8BupMn0r4wxbp7fKiVJUiGxZ5IkSVITluvJMX05u+xFCppijPG/pNPFr7Wc/SVJUjNnmCRJktS89SQNIaowKbdOkiSpWi2yLqA+dOnSJa677rpZlyFJkhrIq6+++mWMsWvWdTRRoZp11c6DEEI4HjgeoH379kP79evXkHVJtSovhxjTZeXrMS5dYOllXa6v6LrqHmtl1tW2b+VttV2vWrNWXKj0n9HXUo3d+uvDGms0zLFraoM1iTBp3XXX5ZVXXsm6DEmS1EBCCJ9kXUMTNgkoqXS7F/B5dTvGGG8kdzatYcOGRdtfqk2MMGsWfPnlt5e5c2HBgpqX+fOXv33hwqyfXc1atoQWLZZdqlu3vKW4eOllxVL19vLW12XfoqK0hFC3yxXZtyKMqRyMVQ76arpe1/2qu15dqLgql+Xl6blU9zpWfg2Xt722fYpyY4VCWPqaVb6sr+sVQpWvDxpiW77W12RF928K1loLOnZsmGPX1AZrEmGSJEmSVtooYGQI4R7SBNyzYoyTM65JjVCMKQCqLhiqafnqK1iypOZjtmoFbdp8e2nbNl126lT99pr2r7xUhCZVQ5CqS3Xr67quuhCoIhyQpKbMMEmSJKkJCyHcDWwHdAkhTALOA1oCxBivB0aTzuQ2AZgHHJVNpcrCokUwdSpMmfLtpbpgqKbeQMXF0KVLWtZcE/r2he9+d+m6qsuaa0KHDgYvklSoDJMkSZKasBjjwbVsj8CP81SO8qC8HKZPrz4gqrxMnpz2q07nztC1awp+1lsPvvOdmoOhLl1gtdUMhiSpOTFMkiRJkgpEjDB+PLz1Vs0h0dSp1Q8ta9s2zavRo0fqOTR8eLpesVRs69YtDT+TJKkmhkmSJElSIzV3Lrz8Mrz4IrzwQrr86qul24uKoHv3pYHQJpssGwxVXjp0aJ4T00qS6p9hkiRJktQIxAiffppCo4plzBgoK0vb+/WDvfaCrbaCTTeFnj3T3EPFxZmWLUlqhgyTJEmSpAwsWgSvv75sePT552lbu3aw+eZw5pkpPNpiizSPkSRJjYFhkiRJkpQHU6cuHa72wgvwyitLz4627rqw3XYpONpyyzRcrYUtdUlSI+VblCRJklTPysrg7beXBkcvvggffJC2tWoFQ4fCyJEpONpyS1h77WzrlSRpRRgmSZIkSfVk4kQ46ywYNQrmzEnrundPPY5OOGHpfEdt2mRbpyRJq8IwSZIkSVpFCxfCFVfAxRdDeTkccQRss00Kj9Zd17OoSZKaFsMkSZIkaRWMHg2nnAITJsA++6RQad11s65KkqSGU5R1AZIkSVIh+vBD2HNP2H13KC6GRx+F++83SJIkNX2GSZIkSdIKmDcPzjsPBgyAJ56Ayy6DN9+EnXbKujJJkvLDYW6SJElSHcQIf/87nHoqfPIJHHwwXH459OyZdWWSJOWXPZMkSZKkWowbB7vuCnvvDR07wlNPwV13GSRJkponwyRJkiSpBnPmwJlnwsYbw4svwpVXwuuvw/DhWVcmSVJ2HOYmSZIkVREj3HsvnHYafPYZHHkkXHopdO+edWWSJGXPnkmSJGnVLViQPn1LTcDbb8P226c5kbp3hxdegD//2SBJkqQKhkmSJGnVzJyZPnn/6ldZVyKtklmz4Kc/hcGD09nZrr8e/vc/2HLLrCuTJKlxMUySJEkr78svYYcd4JVXYNNNs65GWinl5fCXv0DfvmlOpGOPhfHj4Uc/guLirKuTJKnxcc4kSZK0cqZMgR13hAkT4KGHYLfdsq5IWmGvvw4jR6ahbFtsAf/8JwwdmnVVkiQ1boZJkiRpxU2alHokTZqUPn1vv33WFUl188UXsMcelH/yKdPntabt7NbcWNyabuu1pkuL1oQzWkObNtC6dc1LXbZ36QLrrQcdO2b9jCVJqneGSZIkacV89FEKj6ZPh3//G7773awrkuomRjjuOMpee4O7iw8jLlrEwN4LGdh7IS3LFsDChTBvHsyYka5XXhYsWHp9RXTpAuuvn4Klypfrrw8lJdDC5rgkqfD47iVJkupu3LjUI2nePHj8cRg2LOuKpLq76SYYNYqf83te2fxUrr0WNtlkBY8RIyxeXHPQVLFMnZqC1w8/TJevvAL33w9Lliw9VnExlJZWHzStt14KokKo15dAkqT6YJgkSZLq5q234PvfT9efemolPoVLGXr/fTj1VN7t+X1u/PJkPv07rLnmShwnBGjVKi0rOoRtyRL47LOlAdOHHy69PmpUGoJXWYcO3w6aKi579Egzhy9ZAmVly17W17qGXlq1gnbt0tK+/dLrVW/XdL3y7datDd4kKY8MkyRJUu1efRV22inNBfP449CvX9YVSXW3eDEcdhixVWv2nH4rBxxUtHJB0qpq0QLWWSct3/vet7fPmQMff7xs2PTRR2mS+8ceSz0CG4uWLdPzWZmlZcv0v2TxYvjqK5g4MT23imXu3NQDbEUUFS0bMrVrlx6jap2Vb9fH9eLi2i/rsk919ykuhkWLlu39tmDBstdXdVtxcXqd2rZduizv9ors25iGcMaYwtfalur2W17YWh/XQ1j6O1WxrMjtuuxb1MRP4l5env5WKv5elndZ3bqysvQaFRWln0fF9aq3l7dtebeh9uC+aoi/or9Pxx4LQ4bk9WVvRH/hkiSpUXrhBdh1V1hjjRQkbbBB1hVJK+aii+B//+PRY/7GhJt7cudJWRdUgw4dYODAtFQVY+q5VBEwTZ26bABR0/VV3V7d0tAfTGNcOn9VRbhU3fXabi9YsPSD1uLFMH8+zJ6drldeX9v18vKGfb4NIYSlE8W3abPs9YrLjh3Tc5s/P82BN39+es3mz1+6LF688jUUF6feZ5V7jDXE9drCoEL8+dW3Fi2W9qisurRuXfO2uiwtWy4N3RoqeKsaAlUNhCoPH24OWrb89v/rnXc2TJIkSY3Ik0/CHnvAWmvBE0+kCYOlQvLii3DxxcQjjuAXL+3HppvCd76TdVErIQTo3j0tW26ZdTUNqyIIadMGOnfOupplPyhXDZzKyr7d46C6y7rsU919WrWqPgiqbV2LFvUz7K+sbNmAqWrYVNu2RYuWHqtyb7P6vF5cXH2vkBVZarpvRa+h2gLXlb1eXJyex+LFS5eK3636ul2xVBfIVLfMm7f87RXHqE19vG5t2iwNJSsHXyt6Wds+RUXLBpFVQ8m6bqtp3xhX/Xem4nYj6mVmmCRJkqr3yCOw995pfpb//CcFSlIhmT0bDj8cSkp48aCrePu2NAe3U+tohRQVLf1A2twUF6d5qdq3z7oSNSYxpqBx4cIUVFXt6diIAg81HMMkSZL0bQ8+CAcemIbb/Pvf6axSUqH56U/TkLCnn+bqa1dj9dXhoIOyLkqSClzlHltqtowMJUnSsu6+G/bfH4YOTUPbDJJUiB58EG6+Gc46i6m9t+a+++CII+xgIUlSfTBMkiRJS91yCxx6KGy9deqR1KlT1hVJK27yZDjuuBSInncet9ySRmKceGLWhUmS1DQYJkmSpOSaa+CYY2DHHWH06HSmH6nQxAhHHZUmkb3jDsqKWnL99fC970G/flkXJ0lS02CYJEmS4PLLYeRI2HNPGDUK2rXLuiJp5VxzDTz6KFxxBfTrx7/+BRMnwkknZV2YJElNh2GSJEnNWYzw61/D6afDAQfAffel0+RKhWjsWPjFL2C33eCEEwC49tp0IsK99sq4NkmSmhDDJEmSmqsY4cwz4fzz08zEd90FLVtmXZW0chYtgsMOS8Mzb7kFQuDDD+GRR9L0Sf5qS5JUfzyXnyRJzVF5OZxyClx9dZqV+OqrocjvmFTAzjsPXn8d/v536N4dgBtuSL/Wxx2XcW2SJDUxtholSWpuysrg+ONTgPSzn6U5ZgySVMieeQYuuyz9Xu+5JwALF6YOSnvuCb16ZVyfJElNjC1HSZKakyVLYMQIuPlmOOcc+L//gxCyrkpaebNmweGHw4Ybwu9+983q++6DL79MHe8kSVL9cpibJEnNxaJFcPDB8MADcMklcNZZWVckrbqRI+Gzz+CFF6B9+29WX3st9O4NO+yQYW2SJDVRhkmSJDUH8+fDvvvCv/4Ff/hDmi9JKnT33AN33JHOSLjZZt+sfuONlC1dcYUjOCVJagiGSZIkNXULFsAPfgBPPplmJD7++Kwrklbdp5+mMWxbbAG//OUym667Dtq0gSOPzKY0SZKaOsMkSZKaupEj4Ykn4Lbb0nxJUqErL4cjjoDFi1PPpBZLm7Rff51WHXQQdO6cYY2SJDVhhkmSJDVlN92UJts++2yDJDUdv/996ml3882wwQbLbLr9dpg714m3JUlqSI4ilySpqXr11dQraccd05wyUlPw5ptpWNvee8NRRy2zKcY0xG3oUPjOdzKqT5KkZsCeSZIkNUVffZUm3O7WDe66C4qLs65IWnULFsChh6bxazfeCCEss/nZZ+Gdd1KHvCqbJElSPTJMkiSpqSkrg8MOg8mT06frLl2yrkiqH7/8Jbz9NjzySLW/19ddB6uvDgcfnEFtkiQ1I3kd5hZC2CWEMC6EMCGEcGYN+2wXQhgTQngnhPB0PuuTJKlJuPDC9GH7yiuXOV26VND+8580V9LIkbDzzt/aPHUq3H9/OoNbu3b5L0+SpOYkbz2TQgjFwDXAjsAk4OUQwqgY49hK+3QCrgV2iTFODCF0y1d9kiQ1CaNHp/mRjjgCfvSjrKuR6sf06el3un9/uOyyane5+eZ0cjcn3pYkqeHls2fSZsCEGOOHMcZFwD3AXlX2OQR4IMY4ESDG+EUe65MkqbB99FEa3jZoEFx7rZPGqGmIMQWj06bBHXdU2+2orAxuuAG23x769s2gRkmSmpl8hkk9gU8r3Z6UW1dZH2CNEMJTIYRXQwiew1iSpLqYPz9NuF1ensb6OM5HTcXtt8N996Xhm5tuWu0uo0fDxIlw0kl5rk2SpGYqnxNwV/f1aKxyuwUwFNgBaAu8GEL4b4xx/LcOFsLxwPEApaWl9VyqJEkFZuRIeP11ePhh2GCDrKuR6sdHH6Xf7W23hdNOq3G3666DtdaCPffMY22SJDVj+eyZNAkoqXS7F/B5Nfs8EmOcG2P8EngGGFTdwWKMN8YYh8UYh3Xt2rVBCpYkqSDcdBPccguccw784AdZVyPVj7IyGDEiDdf8y1+guLja3T78MM03f/zx0LJlnmuUJKmZymeY9DLQO4SwXgihFXAQMKrKPn8HtgkhtAghtAM2B97NY42SJBWWV15JPTd22gnOPz/raqT689vfwnPPpfm/1lmnxt1uuAGKiuC44/JYmyRJzVzehrnFGJeEEEYCjwLFwC0xxndCCCfktl8fY3w3hPAI8CZQDtwUY3w7XzVKklRQvvoK9tsPuneHO++sseeGVHBefRXOPRcOOggOOaTG3RYsSGdx23NP6Fl1Jk5JktRg8jlnEjHG0cDoKuuur3L7cuDyfNYlSVLBKSuDQw+FyZNT740uXbKuSKof8+al3+0ePWo9K+F996VM1Ym3JUnKr7yGSZIkqZ5ccAE8+ihcfz185ztZVyPVn7/+FcaNg8cfhzXWWO6u114LvXvD9tvnqTZJkgQYJkmSVHhGj05h0pFHplmHpabkyCNh8OC0LMeYMfDii/C736U5kyRJUv741itJUiH58MM0BGjw4FqHAEkFq5YgCeC666BNGzjiiIYvR5IkLcswSZKkQjF/Puy7b7p+//3Qtm229UgZ+frrNOf8wQdD585ZVyNJUvPjMDdJkgpBjPDjH6exPQ8/DOuvn3VFUmZuvx3mzoUTT8y6EkmSmid7JkmSVAhuugn+/Gc45xz4wQ+yrkbKTIxphOewYc49L0lSVuyZJElSY/fKKzByJOy0E5x/ftbVSJl69lkYOxZuvjnrSiRJar7smSRJUmP21Vew337QowfcdRcUF2ddkZSpa6+FTp3goIOyrkSSpObLnkmSJDVWZWXpzG2TJ8Nzz8Gaa2ZdkZSpKVPggQfS9GHt2mVdjSRJzZdhkiRJjdWvfw2PPgo33ODkMBJpaNvixXDCCVlXIklS8+YwN0mSGqN//hMuvBCOPBKOOy7raqTMlZWlXHWHHaBv36yrkSSpeTNMkiSpsfnwQzjsMBg8OE0QE0LWFUmZ++c/4dNP4cQTs65EkiQZJkmS1JjMnw/77puu338/tG2bbT1SI3HddbD22rDnnllXIkmSDJMkSWosYoSTToIxY+COO2D99bOuSGoUPvgAHnkkjfhs2TLraiRJkmGSJEmNxZ/+BLfeCr/6Fey+e9bVSI3GDTdAcbHTh0mS1FgYJkmS1Bi8/DL85Cew885w3nlZVyM1GgsWwC23wF57Qc+eWVcjSZLAMEmSpOx99RXstx+stRbceWfqgiEJgL/9Lf2JnHRS1pVIkqQKLbIuQJKkZi1GOPZYmDIFnn8e1lwz64qkRuW666BPH9h++6wrkSRJFeyZJElSlm69FR56CC65BIYNy7oaqVEZMwZefBFOPBFCyLoaSZJUwTBJkqSsfPQRnHwyfO978NOfZl2N1Ohcdx20bQtHHJF1JZIkqTLDJEmSslBWBiNGQFFR6p1U5FuyVNmsWXDHHXDwwbDGGllXI0mSKnPOJEmSsnD55fDcc3D77VBamnU1UqNz++0wb14a4iZJkhoXvwaVJCnfXn8dzj0X9t8fDj0062qkRidGuPbaNI2YU4lJktT42DNJkqR8WrAADjsMunRJE8I4q7D0Lc88A+++C7fcknUlkiSpOoZJkiTl0y9/CWPHwiOPwJprZl2N1Chdey106gQHHph1JZIkqToOc5MkKV8efxx+/3v48Y9h552zrkZqlKZMgQcegKOOgnbtsq5GkiRVxzBJkqR8mDkTjjwS+vaF3/4262qkRuvJJ6G8HE44IetKJElSTRzmJklSPowcmbpcvPii3S2k5Tj4YNhuO1hrrawrkSRJNbFnkiRJDe3ee+HOO9MZ3Dw1lVQrgyRJkho3wyRJkhrSZ5/BiSfC5pvDWWdlXY0kSZK0ygyTJElqKOXlaRbhhQvh9tuhhaPLJUmSVPhs1UqS1FCuuQYeewyuvx569866GkmSJKle2DNJkqSG8O67cPrpsNtucPzxWVejZi6EsEsIYVwIYUII4cxqtq8eQng4hPBGCOGdEMJRWdQpSZIKg2GSJEn1bdEiOOwwaN8ebr4ZQsi6IjVjIYRi4BpgV2AAcHAIYUCV3X4MjI0xDgK2A64IIbTKa6GSJKlgOMxNkqT6duGF8NprcP/90KNH1tVImwETYowfAoQQ7gH2AsZW2icCHUMIAegATAeW5LtQSZJUGOyZJElSfXrxRbjkEjjySNhnn6yrkQB6Ap9Wuj0pt66yq4H+wOfAW8ApMcby/JQnSZIKjWGSJEn1Zc4cOPxwKC2FK6/MuhqpQnXjLGOV2zsDY4C1gcHA1SGE1b51oBCODyG8EkJ4Zdq0afVdpyRJKhCGSZIk1Zef/Qw+/BD+8hdY7Vufw6WsTAJKKt3uReqBVNlRwAMxmQB8BPSreqAY440xxmExxmFdu3ZtsIIlSVLjZpgkSVJ9ePhh+NOf0hncttkm62qkyl4GeocQ1stNqn0QMKrKPhOBHQBCCN2BvsCHea1SkiQVDCfgliRpVX3xBRx7LAwaBL/+ddbVSMuIMS4JIYwEHgWKgVtijO+EEE7Ibb8euBC4NYTwFmlY3Bkxxi8zK1qSJDVqhkmSJK2KGOH442HmTHj8cWjdOuuKpG+JMY4GRldZd32l658DO+W7LkmSVJgMkyRJWhV//jP8/e9wxRUwcGDW1UiSJEkNzjmTJElaWR9+CKecAt/7Hpx6atbVSJIkSXlhmCRJ0sooK4MRI6C4GG69FYp8S5UkSVLz4DA3SZJWxuWXw/PPwx13QGlp1tVIkiRJeePXqJIkrajXX4dzz4UDDoBDDsm6GkmSJCmvDJMkSVoR8+fDYYdBly5w3XUQQtYVSZIkSXnlMDdJklbEL38JY8fCI49A585ZVyNJkiTlnT2TJEmqq8cfhz/8AUaOhJ13zroaSZIkKROGSZIk1cWMGXDkkdC3L1x2WdbVSJIkSZlxmJskSXUxciRMmQIvvgjt2mVdjSRJkpQZeyZJklSbe++Fu+5KZ3AbNizraiRJkqRMGSZJkrQ8c+bAKafAZpvBWWdlXY0kSZKUOYe5SZK0PL/9LUydCn//O7TwbVOSJEmyZ5IkSTWZNAn+7//gwANh882zrkaSJElqFAyTJEmqya9+BWVl8JvfZF2JJEmS1GgYJkmSVJ0xY+C229J8Seutl3U1kiRJUqOR1zAphLBLCGFcCGFCCOHMarZvF0KYFUIYk1vOzWd9kiQBECP8/OfQuTP88pdZVyNJkiQ1KnmbSTSEUAxcA+wITAJeDiGMijGOrbLrszHGH+SrLkmSvmX0aHjiCbjqKujUKetqJEmSpEYlnz2TNgMmxBg/jDEuAu4B9srj46+QefPgllvgrbeyrkSSlFdLlsAvfgG9e8MJJ2RdjSRJktTo5DNM6gl8Wun2pNy6qrYMIbwRQvhXCGGj/JT2beXlcMwx6ctpSVIzctNN8O678NvfQsuWWVcjSZIkNTp5G+YGhGrWxSq3XwPWiTHOCSHsBjwE9K72YCEcDxwPUFpaWo9lJh06pJENEyfW+6ElSY3V11/DuefCttvCXo2286wkSZKUqXz2TJoElFS63Qv4vPIOMcavY4xzctdHAy1DCF2qO1iM8cYY47AY47CuXbs2SMElJfDpp7XvJ0lqIi69FKZNgyuugFDddyCSJEmS8hkmvQz0DiGsF0JoBRwEjKq8QwihRwip9R5C2CxX31d5rHEZpaWGSZLUbEycCL//PRx6KAwblnU1kiRJUqOVt2FuMcYlIYSRwKNAMXBLjPGdEMIJue3XA/sBJ4YQlgDzgYNijFWHwuVNSQm8+GJWjy5Jyquzz4YY4ZJLsq5EkiRJatTyOWdSxdC10VXWXV/p+tXA1fmsaXlKSmD69HRmt3btsq5GktRgXnkF7rgDzjwzdUuVJEmSVKN8DnMrOCW5GZ4c6iZJTViMcNpp0LUrnHVW1tVIkiRJjZ5h0nJUfDltmCRJTdioUfD00/DrX8Nqq2VdjSRJktToGSYtR0XPpIkTs61DktRAFi+G00+Hfv3guOOyrkaSJEkqCHmdM6nQ9OyZLu2ZJElN1A03wPjx8PDD0MK3REmSJKku7Jm0HK1bQ48ehkmS1CTNnAnnnw/f+x7svnvW1UiSJEkFwzCpFiUlhkmS1CT95jfplJ1XXAEhZF2NJEmSVDAMk2pRUuKcSZLU5Hz8MfzhDzBiBAwZknU1kiRJUkExTKpFRc+kGLOuRJJUb846C4qL4aKLsq5EkiRJKjiGSbUoLYW5c9PUGpKkJuCll+Cee+DnP4devbKuRpIkSSo4hkm1KClJl86bJElNQIwpROreHU4/PetqJEmSpIJkmFSLijDJeZMkqQl48EF4/nm44ALo2DHraiRJkqSCZJhUC3smSVITsWgRnHEGbLQRHH101tVIkiRJBatF1gU0dj16QIsWhkmSVPCuvRYmTIDRo9M/dkmSJEkrxZ5JtSguhp49HeYmSQVtxow0tG3HHWGXXbKuRpIkSSpohkl1UFJizyRJKmgXXZROy/l//wchZF2NJEmSVNAMk+rAMEmSCtiHH8If/whHHQWbbJJ1NZIkSVLBM0yqg9JSmDQJysuzrkSStMLOPBNatoQLL8y6EkmSJKlJMEyqg5ISWLwYpk7NuhJJ0gp54QX429/g9NNh7bWzrkaSJElqEgyT6qCkJF061E2SCkiM8POfw1prwWmnZV2NJEmS1GQYJtVBaWm6NEySpALyt7/Bf/+bJt9u3z7raiRJkqQmwzCpDuyZJEkFZuHCNFfSJpvAEUdkXY0kSZLUpLTIuoBC0LkztG0LEydmXYkkqU6uvho++gj+/W8oLs66GkmSJKlJsWdSHYSQeifZM0mSCsBXX6WhbbvuCjvumHU1kiRJUpNjmFRHpaWGSZJUEC68EL7+Gi6/POtKJEmSpCbJMKmOSkoc5iZJjd7778M118Cxx8JGG2VdjSRJktQkGSbVUUkJTJkCixZlXYkkqUZnnAFt2sCvf511JZIkSVKTZZhURyUlECN8/nnWlUiSqvXss/DggylQ6tEj62okSZKkJsswqY5KS9Ol8yZJUiNUXg4//zn07Ak/+1nW1UiSJElNWousCygUJSXp0nmTJKkRuuceePlluPVWaNcu62okSZKkJs2eSXVUESbZM0mSGpnZs+Gss2DIEDj88KyrkSRJkpo8eybVUYcOsMYahkmS1Oj84hfpn/Ndd0GR35FIkiRJDc1W9wooKTFMkqRG5dFH4YYb0nxJ3/1u1tVIkiRJzYJh0gooKXHOJElqNGbOhGOOgf794cILs65GkiRJajYMk1aAPZMkqRE55RSYMgX+8hdo0ybraiRJkqRmwzBpBZSWwvTpMG9e1pVIUjP397+nEOmXv4Rhw7KuRpIkSWpWDJNWgGd0k6RG4Msv4fjjYfBgOOecrKuRJEmSmh3DpBVQESY5b5IkZSRGOPFEmDEj9Uxq1SrriiRJkqRmp0XWBRQSeyZJUsbuvRfuuw9+8xvYeOOsq5EkSZKaJXsmrYCePSEEwyRJysTkyXDSSbD55nDaaVlXI0mSJDVbhkkroHVr6N7dYW6SlHcxwnHHwfz5cNtt0MKOtZIkSVJWbI2voJISeyZJUt7deiv885/whz9A375ZVyNJkiQ1a/ZMWkGlpYZJkpRXn3wCp5wCw4fDT36SdTWSJElSs2eYtIIqeibFmHUlktQMlJfD0Uenf7p//jMU+bYlSZIkZc1W+QoqKYG5c9NZqSVJDey66+CJJ+CKK2C99bKuRpIkSRKGSSuspCRdOtRNEmVlaS6fjTaC886DJUuyrqhpmTABTj8ddt45Tb4tSZIkqVEwTFpBpaXp0jBJauYefxyGDYOjjkrdFS+4IM3p8/HHWVfWNJSVwZFHQqtWcPPNEELWFUmSJEnKMUxaQfZMkpq5d96B3XeH738/jXe9+2748EO48054+20YPBjuvTfrKgvf738Pzz8Pf/wj9OyZdTWSJEmSKjFMWkHdu0OLFjBxYtaVSMqrqVPhhBNgk01SyHH55fDee3DQQWlS6EMOgTFjoH//tO7oo2HOnKyrLkxjx8I558APfwiHHpp1NZIkSZKqMExaQcXF6UtyeyZJzcS8eXDRRbDhhmm41ciRaS6f006DNm2W3Xe99eCZZ+Dss9NcSptuCq+9lknZBWvxYjjiCOjYEa6/3uFtkiRJUiNkmLQSSksNk6Qmr7wcbrsN+vSBX/0Kdtop9Zi58kro0qXm+7VsmcKnJ5+E+fNhiy3SmcjKy/NXeyG79FJ45ZV0Frfu3bOuRpIkSVI1DJNWQkmJw9ykJu3xx2Ho0DQB9Nprp95G998PvXvX/RjDh8Mbb8APfpB6Me26K0yZ0mAlNwmvv54mMj/kENhvv6yrkSRJklQDw6SVUFICn31mRwOpyRk79tuTa//3v7DNNit3vM6dUwh1/fXw7LNpvqXRo+u35qZi4UIYMQK6dk2TbkuSJElqtAyTVkJpaZrWY+rUrCuRVC8qJtfeeOPqJ9deFSHAj36Uhm716JHCqlNPhQUL6qX0JuP889PZ8G66KYVwkiRJkhotw6SVUFKSLp03SSpw8+bBxRfXbXLtVTVgAPzvf3DyyWnepS22gHffrd/HKFT//S/89rdwzDGw225ZVyNJkiSpFoZJK6EiTHLeJKlAVZ5c+5xz6j659qpq0yY9xj/+kcbKDh0KN94IMTbcYzZ28+als7f16gW/+13W1UiSJEmqA8OklWDPJKmAPfHEqk+uvap23x3efBO++900BG6//WD69Pw9fmPyy1/C+PHw5z/DaqtlXY0kSZKkOjBMWgmdO0O7doZJUkEZOzadWW2HHepncu1VtdZa8OijaXjXqFEwaBA8/XQ2tWTlqadST62RI2H77bOuRpIkSVIdGSathBBS7yTDJKkAzJkDJ56YzqT23HP1O7n2qioqgl/8Al58MQ2B2357+NWvYMmSbOvKh9mz4aij0nxVl16adTWSJEmSVkBeP0mFEHYJIYwLIUwIIZy5nP2+E0IoCyHsl8/6VkRJiXMmSQXh1FPTvEQ//nHDTa69qoYNg9degxEj4KKLYNtt4eOPs66qYZ12Wvonettt0L591tVIkiRJWgF5C5NCCMXANcCuwADg4BDCgBr2uwx4NF+1rQx7JkkF4LHH0lnaTj+94SfXXlUdO6Z5g+6+G955Jw17u+eerKtqGI88kgK+006DrbbKuhpJkiRJKyifPZM2AybEGD+MMS4C7gH2qma/nwD3A1/ksbYVVloKU6bAokVZVyKpWrNnw3HHQd++cN55WVdTdwcdBGPGwEYbwcEHp6Fgs2dnXVX9mTEDjjkmPb9f/zrraiRJkiSthHyGST2Byn15JuXWfSOE0BPYG7i+toOFEI4PIbwSQnhl2rRp9VpoXZSUpLN5f/553h9aUl2cdVYaRnXLLY1vWFtt1lsvnWXuV7+Cv/wFNt00nXFu4sT0j6eQnXIKTJ2ahrcV2s9FKmB1mWoghLBdCGFMCOGdEEIzOyOAJElaEfkMk0I166p+KvoDcEaMsay2g8UYb4wxDosxDuvatWt91LdCSkrSpfMmSY3QM8/ANdek4KJQh1G1aAEXXABPPgkLFsB++8E660DXrrDjjmno3t13w7hxUF6edbV18+CDcPvtcM45MHRo1tVIzUZdphoIIXQCrgX2jDFuBOyf7zolSVLhaJHHx5oElFS63Quo2q9nGHBPCAGgC7BbCGFJjPGhvFS4AkpL06XzJkmNzLx5aRjV+uunyawL3bbbwvvvw+uvL11eey3NAVUxzrZ9+zTH0pAhqRfTkCFpGFmrVtnWXtm0afCjH6Xazj4762qk5uabqQYAQggVUw2MrbTPIcADMcaJADHGRj3dgCRJylY+w6SXgd4hhPWAz4CDSA2Xb8QY16u4HkK4FfhHYwySYGnPJMMkqZE577x01rbHH286Zwlr0wa23DItFRYtgnffXRouvf56Gjp2zTVpe8uWMHBgCm8qlkGDoEOH/NUdYwr3Zs2Cn/wkXT7xRKpNUj5VN9XA5lX26QO0DCE8BXQErowx/qXqgUIIxwPHA5RWfLMmSZKanbyFSTHGJSGEkaSztBUDt8QY3wkhnJDbXus8SY1J+/awxhoOc5MalZdegt/9LvWA2X77rKtpWK1apXBo0CA48si0rrw8BWmVezGNGpXmjQIIAfr0WRouVfRiWnPNbx+/chC0vGXmzOVvL6s0avnSS1PAJSnf6jLVQAtgKLAD0BZ4MYTw3xjj+GXuFOONwI0Aw4YNK/BJ3CRJ0srKZ88kYoyjgdFV1lUbIsUYj8xHTauipMSeSVKjsXAhHH00rL02/Pa3WVeTjaKiFBb16QMHHpjWxQiffbZsD6YXXoB77ll6v5KSNOn37NlLw6Gvv4YlS2p/vNVWg9VXh06d0mVJSQqMVl992aWkBHbZpaGeuaTlq8tUA5OAL2OMc4G5IYRngEHAeCRJkqrIa5jU1JSWGiZJjcZFF8HYsTB6dAo4lIQAvXqlZY89lq7/6qtlezB99hn07AkDBnw7CKocFlVeOnRIx5fU2NU61QDwd+DqEEILoBVpGNzv81qlJEkqGIZJq6CkJH3BLyljY8bAb34DRxwBu+6adTWFYc014fvfT4ukJq0uUw3EGN8NITwCvAmUAzfFGN/OrmpJktSYGSatgpISmD4d5s5tOvP8SgVn8eI0vK1LlzRfkiTpW+oy1UCM8XLg8nzWJUmSClNR1gUUMs/oJjUCl1+ehmlddx107px1NZIkSZLU5BkmrYKKM+IaJkkZGTsWfv1rOOAA2HvvrKuRJEmSpGbBMGkV2DNJylBZWRre1rEj/PGPWVcjSZIkSc2Gcyatgp4904mMJk7MuhKpGbrySnjpJbjrLujWLetqJEmSJKnZsGfSKmjVCrp3t2eSlHcTJsDZZ8Oee8JBB2VdjSRJkiQ1K4ZJq6i01DBJyqvycjj2WGjdOk26HULWFUmSJElSs+Iwt1VUUgJvv511FVIzcsMN8PTTcPPNsPbaWVcjSZIkSc2OPZNWUUlJ6pkUY9aVSM3AJ5/A6afDjjvCUUdlXY0kSZIkNUuGSauotBTmzYMZM7KuRGriYoTjj0/X//Qnh7dJkiRJUkYMk1ZRSUm6dN4kqYHdeiv8+99w2WWwzjpZVyNJkiRJzZZh0iqqCJMmTsy2DqlJ+/xz+OlPYdtt4YQTsq5GkiRJkpo1w6RVZM8kqYHFCCeeCAsXwk03QZH/tiRJkiQpS57NbRX16AEtWxomSQ3m3nth1Cj4v/+D3r2zrkaSJEmSmj2/4l9FRUXQs6dhktQgpk2Dn/wENtsMTj0162okSZIkSRgm1YuSEudMkhrET34CX38Nt9wCxcVZVyNJkiRJwjCpXpSU2DNJqncPPpiGuJ17Lmy0UdbVSJIkSZJyDJPqQWkpfPYZlJdnXYnURMyYASedBIMHw+mnZ12NJEmSJKkSw6R6UFICixfD1KlZVyI1ET/7WZov6ZZb0gz3kiRJkqRGwzCpHpSUpEvnTZLqwSOPwK23wplnwpAhWVcjSZIkSarCMKkmixbBCy/UKSEqLU2XzpskraKvv4bjj4f+/eFXv8q6GkmSJElSNQyTajJnDnz3u2kC4FpU9EwyTJJW0RlnpAnI/vxnaN0662okSZIkSdUwTKpJ586py9GYMbXuusYa0K6dw9ykVfLkk3D99fDTn8Lmm2ddjSRJkiSpBoZJyzN4MLz+eq27hZB6J9kzSVpJc+fCscfChhvCBRdkXY0kSZIkaTkMk5Zn8GAYNw7mzat119JSwyRppf3qV/Dhh3DzzambnyRJkiSp0TJMWp4hQ6C8HN5+u9Zd7ZkkraTHH4c//AFOOgm23TbraiRJkiRJtTBMWp7Bg9NlHYa6lZTAlCnpJHCS6uCdd2C//eD734f11oNLL826IkmSJElSHRgmLc8660CnTnWahLukBGJMJ6KStBzjx8Mhh8DGG8O//w3nnQevvQYdO2ZdmSRJkiSpDgyTlieEOk/CXVqaLh3qJtXgo4/gqKOgf3/4+9/hjDPSuvPPh9VXz7o6SZIkSVIdGSbVZvBgePNNKCtb7m4lJenSMEmq4tNP4Uc/gj594J574NRTU4j0m9/AmmtmXZ0kSZIkaQW1yLqARm/wYJg/H95/H/r1q3G3ijBp4sT8lCU1epMnp8DohhvSGNAf/Qh++UtYe+2sK5MkSZIkrQJ7JtVmyJB0WctQt/btoXNneyZJTJsGv/gFbLABXHstjBiRwtirrzZIkiRJkqQmwDCpNv37Q6tWdZ6E2zBJzdaMGXD22enMbL/7Hey/P4wbB3/6U5rMXpIkSZLUJDjMrTYtW8LAgXUOkxzmpmbn66/hD39IAdKsWXDggWlS7eUMC5UkSZIkFS57JtVFxRndYlzubvZMUrMydy5cemnqiXTeebD99mmy+nvuMUiSJEmSpCbMMKkuBg9O88BMnrzc3UpL00ifuXPzU5aUifnz4fe/TyHSWWfBFlvAK6/AAw/AxhtnXZ0kSZIkqYEZJtVFxSTctQx1qzijm72T1CQtXAjXXAMbbgg/+xkMGgQvvAD//CcMHZp1dZIkSZKkPDFMqotNNkmXtZzRrSJMct4kNSmLF8NNN0GfPjByZDpL21NPwWOPwZZbZl2dJEmSJCnPnIC7LlZbLX2AtmeSmpunn4Yf/SidlW3zzVOo9P3vQwhZVyZJkiRJyog9k+pqyJBaeyb17Jk+YxsmqeDNnJlCpO22g0WLYNQoePFF2HFHgyRJkiRJauYMk+pq8GD44IN0GvQatGoFPXoYJqnAPfggDBiQeiGddhq89RbssYchkiRJkiQJMEyqu8GD0+Wbby53t5IS50xSgfr8c9h3X9hnH+jeHf73P7j8cmjfPuvKJEmSJEmNiGFSXVWc0a2WoW6lpfZMUoEpL4c//Sn1Rho9Gi69NAVJnqFNkiRJklQNw6S6Wmst6Nq1TpNwf/opxJifsqRVMn48bL89HH98CkzffBPOOANatsy6MkmSJElSI2WYVFchpA/bdQiT5s2D6dPzU5a0UhYvhksugU02gTfeSPMjPfEE9O6ddWWSJEmSpEbOMGlFDB4Mb7+dPojXoKQkXTrUTY3Wyy/DsGFw9tmw557w7rtwzDFOsC1JkiRJqhPDpBUxeHA6Tfq779a4S2lpujRMUqMzdy78/OewxRbw5Zfw0EPw17+mUxBKkiRJklRHhkkromIS7uUMdbNnkhqlRx+FgQPhd79L8yONHQt77ZV1VZIkSZKkAmSYtCJ694a2bZd7Rrfu3dPcxRMn5rEuqSZffQUjRsAuu0Dr1vDMM3DddbD66llXJkmSJEkqUIZJK6K4OE1YvJyeSUVF0LOnPZOUsRjhrrugf3+4+24455z0e7vNNllXJkmSJEkqcIZJK6rijG4x1rhLaalhkjL0ySew++5w6KGw/vrw2mtw4YXQpk3WlUmSJEmSmgDDpBU1eDDMnJk+sNegpMQwSRkoK4OrroKNNkrD2a68Ep5/HjbeOOvKJEmSJElNiGHSiho8OF3WMgn3pEnps72UF2+/Dd/9LpxyShrK9s47cPLJaWimJEmSJEn1yDBpRW28cZoYaTmTcJeWwpIlMHVqHutS8zR7NvzqV7DppvDBB3DnnTB6NKyzTtaVSZIkSZKaqBZZF1Bw2rWDvn1r7ZkEaajb2mvnpyw1MzNnwh//CH/4A0yfDocdBr//PXTpknVlkiRJkqQmLq89k0IIu4QQxoUQJoQQzqxm+14hhDdDCGNCCK+EELbOZ311VjEJdw0qh0lSvZo+Hc49F9ZdN11utRW89BLcfrtBkiRJkiQpL/IWJoUQioFrgF2BAcDBIYQBVXZ7HBgUYxwMHA3clK/6VsjgwTBxInz1VbWbK8KkiRPzV5KauGnT4Kyz0vC1Cy+EHXZIZ2l7+GHYbLOsq5MkSZIkNSP57Jm0GTAhxvhhjHERcA+wV+UdYoxzYowxd7M9EGmMKibhfuONajevsQa0b2/PJNWDKVPgtNNST6TLLoPdd4c334T770895CRJkiRJyrN8hkk9gcrxyqTcumWEEPYOIbwH/JPUO6nxqeWMbiGk3kmGSVppkyals7Gtt16aC2mffWDsWLjnnjQJvCRJkiRJGclnmBSqWfetnkcxxgdjjP2AHwIX1niwEI7Pzav0yrRp0+qvyrro2hV69lzuGd1KShzmppXwySdw4omwwQZw3XVwyCEwblyaE6lfv6yrkyRJkiQpr2HSJKCk0u1ewOc17RxjfAbYIIRQ7azCMcYbY4zDYozDunbtWr+V1sXgwbVOwm3PJNXZBx/AMcfAhhvCzTfDUUfB+++n6xtumHV1kiRJkiR9I59h0stA7xDCeiGEVsBBwKjKO4QQNgwhhNz1TYFWQPWzXGdtyBB4912YP7/azaWlMHUqLFqU57pUWMaNgxEjoG9fuPNOOOGEFCxdf32aJ0mSJEmSpEYmb2FSjHEJMBJ4FHgX+GuM8Z0QwgkhhBNyu+0LvB1CGEM689uBlSbkblwGD4ayMnjnnWo3l5RAjPDZZ/ktSwXinXfg4IOhf3+47z445RT46CP44x+Xng5QkiRJkqRGqEU+HyzGOBoYXWXd9ZWuXwZcls+aVlrlSbiHDfvW5oo8YOLENIeyBKTfl4suSmdja98eTj8dfvYz6NYt68okSZIkSaqTvIZJTcp668Fqq9U4CXdFmOS8SQLg5Zfhwgvh4YfT780558Cpp8Kaa2ZdmSRJkiRJK8QwaWUVFcGgQTVOwm2YJAA+/hhOOgn+9S9YYw349a/h5JOhU6esK5MkSZIkaaUYJq2KwYPhllugvDyFS5W0bw+dOxsmNWszZ8Kuu8Lnn8NvfpNCpdVWy7oqSZIkSZJWiWHSqhgyBObOhQkToE+fb20uKUlzJqkZWrIEDjggnZntscdg+PCsK5IkSZIkqV7k7WxuTVLlSbirUVpqz6Rm65RTUoh0/fUGSZIkSZKkJsUwaVVstBG0bLnceZMMk5qhq6+Ga6+F006Do4/OuhpJkiRJkuqVYdKqaNUKBgxY7hndZsyAOXPyXJey8+ijqVfSnnvCpZdmXY0kSZIkSfXOMGlVDR7sGd2UjB2b5knaeGO4804oLs66IkmSJEmS6p1h0qoaMgSmTElLFaWl6dIwqRn48kvYYw9o2xZGjYIOHbKuSJIkSZKkBmGYtKqWMwm3PZOaiYULYZ994LPP4O9/X5oiSpIkSZLUBBkmrapBg9JlNWFSz54QAkycmN+SlEcxwo9+BM8+C7feCptvnnVFkiRJkiQ1KMOkVdWpE6y3XrWTcLdsCT162DOpSfvtb+G22+C88+Cgg7KuRpIkSZKkBmeYVB+WMwl3aalhUpP10ENw1llw4IEpTJIkSZIkqRkwTKoPgwfD++/DnDnf2lRSYpjUJL3+Ohx6KHznO/DnP6fxjJIkSZIkNQOGSfVhyJA0d86bb35rU0lJmjMpxgzqUsOYPBn23BPWXDNNuN22bdYVSZIkSZKUN4ZJ9WE5Z3QrLYX582H69LxWpIYyfz7stRfMmAEPP5wmxZIkSZIkqRkxTKoPvXqlXirVhEklJenSoW5NQHk5HHkkvPIK3HXX0jP5SZIkSZLUjBgm1YcQUu+kas7oZpjUhPz61/DXv8Jll6VhbpIkSZIkNUOGSfVl8GB46y1YsmSZ1RVh0sSJ+S9J9eiuu+CCC+Coo+C007KuRpIkSZKkzBgm1ZchQ2DhQhg3bpnV3btDy5b2TCpoL74IRx8N224L11/vmdskSZIkSc2aYVJ9qZiEu8pQt6KiNKWSYVKB+uQT+OEP0w/xgQegVausK5IkSZIkKVOGSfWlb19o06bGSbgd5laAZs+GPfZIPc7+8Y80ybokSZIkSc2cYVJ9adECNt64xkm47ZlUYMrK4JBDYOxY+NvfoF+/rCuSJGmlhRB2CSGMCyFMCCGcuZz9vhNCKAsh7JfP+iRJUmExTKpPgwennkkxLrO6tBQ++yzlEyoQZ5yReiNddRXsuGPW1UiStNJCCMXANcCuwADg4BDCgBr2uwx4NL8VSpKkQmOYVJ8GD4bp02HSpGVWl5Skk7xNnZpNWVpBN90EV1wBI0fCSSdlXY0kSatqM2BCjPHDGOMi4B5gr2r2+wlwP/BFPouTJEmFxzCpPg0Zki6rDHUrKUmXzptUAJ58Ek48EXbeGX7/+6yrkSSpPvQEKg+4n5Rb940QQk9gb+D6PNYlSZIKlGFSfdp443Ta+CqTcJeWpkvnTWrk3n8f9t0X+vSBe+9N82BJklT4QjXrYpXbfwDOiDEud1B+COH4EMIrIYRXpk2bVl/1SZKkAuOn5frUoUMKIqqESRU9kwyTGrEZM+AHP4DiYnj4YVh99awrkiSpvkwCSird7gV8XmWfYcA9IQSALsBuIYQlMcaHKu8UY7wRuBFg2LBhVQMpSZLUTBgm1bfBg+Gll5ZZ1akTtG9vmNRoLV4M++8PH30Ejz8O66+fdUWSJNWnl4HeIYT1gM+Ag4BDKu8QY1yv4noI4VbgH1WDJEmSpAoOc6tvgwfDxx/DzJnfrAoh9U5yzqRGKEY4+eQUIv3pT7DNNllXJElSvYoxLgFGks7S9i7w1xjjOyGEE0IIJ2RbnSRJKkT2TKpvFZNwv/EGDB/+zerSUnsmNToxwqWXwvXXwxlnwBFHZF2RJEkNIsY4GhhdZV21k23HGI/MR02SJKlw2TOpvg0enC6rOaObYVIjMncuHHYY/PKXcOCBcMklWVckSZIkSVJBMEyqb927Q48e1U7CPWUKLFyYTVmq5P33YYst4O674aKL4K67oMg/BUmSJEmS6sJhbg1hyJBqeyYBfPaZ8ztnatQoOPxwaNECHnkEdtop64okSZIkSSoodsdoCIMHw9ixy3RDKi1Nlw51y0hZGZxzDuy1F/TuDa++apAkSZIkSdJKMExqCIMHw5IlKVDKqeiZZJiUga++gt12g4svhmOOgeeeg3XXzboqSZIkSZIKkmFSQ6g4o1uloW4VYdLEiRnU05y98goMHQpPPQV/+hPcdBO0aZN1VZIkSZIkFSzDpIawwQbQvv0yk3C3awdrrmnPpLy6+WbYemuIMfVGOvbYrCuSJEmSJKngGSY1hKIiGDSo2jO6GSblwYIFcNxxKTzadts0P9J3vpN1VZIkSZIkNQmGSQ1lyJAUJpWXf7PKMCkPPvkEttkmDWf75S/hX/+CLl2yrkqSJEmSpCbDMKmhDB4Ms2fDRx99s6qkxDmTGtRjj6X5kcaPh4ceShNuFxdnXZUkSZIkSU2KYVJDqZiEu9JQt9JSmDkT5szJpKKmq7wcLrkEdt4ZevSAl1+GvfbKuipJkiRJkpokw6SGstFGqVdMNWd0c6hbPZo1C/bZB84+Gw46CF56Cfr0yboqSZIkSZKaLMOkhtKmDfTvv0zPJMOkevb22zBsGPzzn3DllXDnneksepIkSZIkqcEYJjWkIUOq7ZnkvEn14O67YfPN05jBJ5+Ek0+GELKuSpIkSZKkJs8wqSENHgyffw5ffAFAz54p77Bn0ipYvBhOOQUOOQQ23RReew223jrrqiRJkiRJajYMkxrS4MHp8o03AGjZEtZayzBppU2eDN/7Hlx1FZx6KjzxRHpBJUmSJElS3hgmNaSKMKnKUDeHua2EZ59NPZFefz0Ncfv971M6J0mSJEmS8sowqSF17gylpd+ahNueSSsgxjS59vbbQ8eO6WxtBx2UdVWSJEmSJDVbhkkNbfDgZcKk0tIUJsWYWUWFYc4cePBB2GuvNKRt993h5Zdh4MCsK5MkSZIkqVlrkXUBTd6QIfCPf8C8edCuHSUlMH8+TJ8Oa66ZdXGNzKefwsMPp+WJJ2DRIlh9dfjNb+D006HI7FOSJEmSpKwZJjW0wYOhvBzeegs235ySkrR64kTDJMrL4dVXU3g0atQ3E5WzwQbw4x/DHnukM7U5N5IkSZIkSY2GYVJDGzIkXY4ZA5tvTmlpuvnpp0s3NSvz5sF//pMCpH/8A6ZMST2OttoKLrssBUj9+kEIWVcqSZIkSZKqYZjU0EpLoVOnb87oVtEzqVlNwv355yk4GjUKHn8cFixIk2nvsksKj3bdFbp0ybpKSZIkSZJUB4ZJDS2EZSbh7tYtjdpq0mFSjCk8q5j/6NVX0/p114XjjksB0vDh0KpVpmVKkiRJkqQVZ5iUD0OGwPXXQ1kZRcXF9OqV5kxqUhYsSJNmjxqVeiF99lkK0rbYAi65JAVIG23k8DVJkiRJkgqcYVI+DB6cTuE2fjz0709paRPpmVRWBnfeCQ88AI89luZDat8edtophUe77566YkmSJEmSpCYjr+daDyHsEkIYF0KYEEI4s5rth4YQ3swtL4QQBuWzvgYzeHC6zA11KylpImHS2WfDEUekYWxHHgn/+hd8+WUKl446yiBJkiRJkqQmKG89k0IIxcA1wI7AJODlEMKoGOPYSrt9BAyPMc4IIewK3Ahsnq8aG0z//ml+oNdfh4MPpqQEJk1KHXuKi7MubiXdf386+9rxx6chfA5fkyRJkiSpWchnz6TNgAkxxg9jjIuAe4C9Ku8QY3whxjgjd/O/QK881tdwWraEgQOX6ZlUVgZTpmRb1kobOzb1RNpiC7jqKoMkSZIkSZKakXyGST2ByoO7JuXW1eQY4F8NWlE+VZzRLUZKS9OqghzqNmsW/PCH0K4d3HcftG6ddUWSJEmSJCmP8hkmVdd9JVa7YwjfI4VJZ9R4sBCODyG8EkJ4Zdq0afVUYgMaMgSmTYPPP6ekJK0quDCpvBxGjICPPoK//Q16Li8LlCRJkiRJTVE+w6RJQEml272Az6vuFELYBLgJ2CvG+FVNB4sx3hhjHBZjHNa1a9d6L7beVZqEuyJMmjgxs2pWzsUXw6hRcMUVsO22WVcjSZIkSZIykM8w6WWgdwhhvRBCK+AgYFTlHUIIpcADwOExxvF5rK3hDcqdmG7MGDp1gg4dCqxn0ujRcN55cNhh8JOfZF2NJEmSJEnKSN7O5hZjXBJCGAk8ChQDt8QY3wkhnJDbfj1wLrAmcG1IkzoviTEOy1eNDapjR9hwQ3j9dUJIk3AXTJg0YQIcemgKxG64wQm3JUmSJElqxvIWJgHEGEcDo6usu77S9WOBY/NZU14NHgyvvw4UUJg0dy7ssw8UFcEDD6SJtyVJkiRJUrOVz2FuGjIEPvgAZs2ipKQA5kyKEY45Bt5+G+6+G9ZbL+uKJEmSJElSxgyT8qliEu4336S0FKZOhYULM61o+X7/e7j33jTx9k47ZV2NJEmSJElqBAyT8qmaM7p99llm1Szfk0/C6aenIW5nnpl1NZIkSZIkqZEwTMqntdaCbt3g9de/CZMa5bxJEyfCAQdA795w661OuC1JkiRJkr5hmJRPIaTeSZV6JjW6eZMWLIB9903j7x56KJ2FTpIkSZIkKccwKd8GD4Z33qGk+yKgkfVMihF+/GN45RW4/Xbo2zfriiRJkiRJUiNjmJRvQ4bAokW0++Rd1lyzkYVJN9wAt9wC55wDe+2VdTWSJEmSJKkRMkzKt0qTcK+zDowbl2k1S734Ipx8MuyyC5x/ftbVSJIkSZKkRsowKd9694Z27WDMGHbeGZ5+GqZMybimKVNgv/2gpATuuguKizMuSJIkSZIkNVaGSflWXAybbAKvv87hh0N5Odx9d4b1LFoE++8PM2bAgw/CGmtkWIwkSZIkSWrsDJOykDujW/9+kWHD4C9/ybCW006D556Dm29OIZckSZIkSdJyGCZlYcgQmDULPv6YESNgzBh4660M6rj9dvjjH+GnP4WDD86gAEmSJEmSVGgMk7JQaRLugw6CFi1SrpNXr70Gxx8P220Hv/1tnh9ckiRJkiQVKsOkLAwcCEVFMGYMXbvCrrvCnXdCWVmeHv+rr2CffaBLF7j33pRmSZIkSZIk1YFhUhbatYN+/eD11wEYMQI+/xyeeCIPj11Wloa0TZ4M998P3brl4UElSZIkSVJTYZiUldwk3AA/+AF06pSnibjPPhseewyuvRY22ywPDyhJkiRJkpoSw6SsDB4Mn34KX31FmzZwwAHwwAMwZ04DPub998Nll6W5ko45pgEfSJIkSZIkNVWGSVkZMiRd5nonjRgB8+alQKlBjB0LRx4JW2wBV13VQA8iSZIkSZKaOsOkrAwalC5zYdJWW8H66zfQULdZs+CHP0xzNd13H7Ru3QAPIkmSJEmSmgNP45WVrl2hZ0/44x/h1VcJ3bpx47rduOfxbnx5Sze6DOiWJsfu1g3at4cQVu5xystTt6ePPoLHH0+PKUmSJEmStJIMk7J0xhlw993wv//BF1+ww+zZ7ABQdTqjtm2XBkvduqUgqvLtqttatVp634svhlGj4MorYdtt8/jkJEmSJElSU2SYlKWf/CQtFebPZ99tp1H01TT+evUXhGlfwBdVlsmT4Y030vVFi6o/bqdOKVjq0gVefBEOO2zZx5EkSZIkSVpJhkmNSdu27HRsKSecUMpr3WHobsvZN0b4+utvh01Vl4MPhhtuWPlhcpIkSZIkSZUYJjUyBxwAJ58Mt98OQ4cuZ8cQYPXV09K7d97qkyRJkiRJzZtnc2tk1lgD9twT7roLFi/OuhpJkiRJkqRlGSY1QocfDtOmwaOPZl2JJEmSJEnSsgyTGqFddklzZ99+e9aVSJIkSZIkLcswqRFq1SrNm/33v8PMmVlXI0mSJEmStJRhUiN1+OGwcCH87W9ZVyJJkiRJkrSUYVIjNWwY9OvnUDdJkiRJktS4GCY1UiHAiBHw7LPw4YdZVyNJkiRJkpQYJjVihx6aLu+4I9s6JEmSJEmSKhgmNWKlpfC976WhbjFmXY0kSZIkSZJhUqM3YgRMmAD//W/WlUiSJEmSJBkmNXr77gtt28Jf/pJ1JZIkSZIkSYZJjV7HjrD33nDvvbBwYdbVSJIkSZKk5s4wqQCMGAEzZsA//5l1JZIkSZIkqbkzTCoAO+wAa63lUDdJkiRJkpQ9w6QC0KIFHHIIjB4NX36ZdTWSJEmSJKk5M0wqECNGwOLFae4kSZIkSZKkrBgmFYhNNoFBgxzqJkmSJEmSsmWYVEAOPxz+9z8YNy7rSiRJkiRJUnNlmFRADjkEiorg9tuzrkSSJEmSJDVXhkkFZK21YKedUphUXp51NZIkSZIkqTkyTCowhx8OEyfCs89mXYkkSZIkSWqODJMKzA9/CB06OBG3JEmSJEnKhmFSgWnXDvbfH/72N5g3L+tqJEmSJElSc2OYVIAOPxxmz4ZRo7KuRJIkSZIkNTeGSQVo+HAoKXGomyRJkiRJyj/DpAJUVJR6Jz36KEyZknU1kiRJkiSpOTFMKlCHHw7l5XD33VlXIkmSJEmSmhPDpALVrx985zsOdZMkSbULIewSQhgXQpgQQjizmu2HhhDezC0vhBAGZVGnJEkqDIZJBWzECBgzBt56K+tKJElSYxVCKAauAXYFBgAHhxAGVNntI2B4jHET4ELgxvxWKUmSColhUgE76CBo0QJuvz3rSiRJUiO2GTAhxvhhjHERcA+wV+UdYowvxBhn5G7+F+iV5xolSVIBMUwqYF26wG67wR13QFlZ1tVIkqRGqifwaaXbk3LranIM8K8GrUiSJBU0w6QCN2IETJ4Mjz+edSWSJKmRCtWsi9XuGML3SGHSGTVsPz6E8EoI4ZVp06bVY4mSJKmQGCYVuB/8ADp1cqibJEmq0SSgpNLtXsDnVXcKIWwC3ATsFWP8qroDxRhvjDEOizEO69q1a4MUK0mSGr+8hkl1OJNIvxDCiyGEhSGE0/JZW6Fq3RoOPBAeeABmz866GkmS1Ai9DPQOIawXQmgFHASMqrxDCKEUeAA4PMY4PoMaJUlSAclbmFTHM4lMB04G/i9fdTUFI0bAvHkpUJIkSaosxrgEGAk8CrwL/DXG+E4I4YQQwgm53c4F1gSuDSGMCSG8klG5kiSpALTI42N9cyYRgBBCxZlExlbsEGP8AvgihLB7HusqeFtuCRtskIa6HXFE1tVIkqTGJsY4GhhdZd31la4fCxyb77okSVJhyucwtxU9k4jqKAQ4/HB44gn49NPa95ckSZIkSVpZ+QyT6nwmkTodzLOJLOPwwyFGuPPOrCuRJEmSJElNWT7DpDqdSaSuPJvIstZfH7773TTULa50RCdJkiRJkrR8+QyTaj2TiFbNiBEwdiy89lrWlUiSJEmSpKYqb2FSXc4kEkLoEUKYBPwMOCeEMCmEsFq+aix0++8PrVvDX/6SdSWSJEmSJKmpyufZ3OpyJpEppOFvWglrrAF77AF33w3/93/QsmXWFUmSJEmSpKYmn8PclAcjRsC0afDoo1lXIkmSJEmSmiLDpCZml12gSxeHukmSJEmSpIZhmNTEtGwJBx8Mo0bBzJlZVyNJkiRJkpoaw6QmaMQIWLgQ/va3rCuRJEmSJElNjWFSEzR0KPTv71A3SZIkSZJU/wyTmqAQ4PDD4bnn4MMPs65GkiRJkiQ1JYZJTdShh6ZQ6Y47sq5EkiRJkiQ1JYZJTVRpKXzve/DnPzsRtyRJkiRJqj+GSU3YmWfCZ5/BdtvB1KlZVyNJkiRJkpoCw6QmbMcd4eGH4f33Yeut4eOPs65IkiRJkiQVOsOkJm7nneGxx+DLL+G734WxY7OuSJIkSZIkFTLDpGZgq63g6aehvBy22Qb+97+sK5IkSZIkSYXKMKmZ2GQTeP556NQJtt8e/vOfrCuSJEmSJEmFyDCpGVl/fXjuOVhvPdh9d3jggawrkiRJkiRJhcYwqZlZay145hkYOhT23x9uvjnriiRJkiRJUiExTGqG1lgjTcr9/e/DscfC5ZdnXZEkSZIkSSoUhknNVPv28PDDcMABcPrpcNZZEGPWVUmSJEmSpMauRdYFKDutWsFdd6WeSpdeCtOnw7XXQnFx1pVJkiRJkqTGyjCpmSsuhuuugzXXhEsugRkz4I47UtAkSZIkSZJUlWGSCAEuvhg6d4bTToNZs9KZ3tq3z7oySZIkSZLU2Dhnkr7x85/DLbfAf/4DO+6Yhr1JkiRJkiRVZpikZRx1FNx3H7z6KgwfDp9/nnVFkiRJkiSpMTFM0rfsvTeMHg0ffwxbbw0ffJB1RZIkSZIkqbEwTFK1dtgBnngCvv46BUpvvpl1RZIkSZIkqTEwTFKNvvMdePbZdMa34cPhhReyrkiSJEmSJGXNMEnL1b8/PP88dO0K3/8+PPJI1hVJkiRJkqQsGSapVuusA889B337wp57wr33Zl2RJEmSJEnKimGS6qRbN3jqKdhiCzj4YLjhhqwrkiRJkiRJWTBMUp2tvjo8+ijsvjuccAJccgnEmHVVkiRJkiQpnwyTtELatoUHHoDDDoOzz4bTToNFi7KuSpIkSZIk5YthklZYy5Zw223wk5/A734HpaVw1lnw4YdZVyZJkiRJkhqaYZJWSlERXHkljB4Nm28Ov/0tbLAB7Lxz6rm0eHHWFUqSJEmSpIZgmKSVFgLsuiv8/e/wySdw/vkwdizsu286A9w556T1kiRJkiSp6TBMUr3o1QvOOw8++ghGjYJNN00TdK+3Huy2WwqclizJukpJkiRJkrSqDJNUr1q0gD32gH/8Az7+OPVOeuMN+OEPYd11U+D06acZFylJkiRJklaaYZIaTGkpXHBBGur24IOw8cZw4YUpVKoInMrKsq5SkiRJkiStCMMkNbgWLVLPpH/9K53x7cwz4ZVXUqC03nopcPrss6yrlCRJkiRJdWGYpLxad124+GKYOBHuuw/69UtD39ZZZ2ngZG8lSZIkSZIaL8MkZaJly3TWt3//GyZMgNNOgxdfTJN1b7BBCpwmT866SkmSJEmSVJVhkjK3wQZw6aVpYu577023zzknzbm0777w5JMQY9ZVSpIkSZIkMExSI9KqFRxwADz+OIwfD6eeCk8/DdtvD5ttloKmJUuyrlKSJEmSpObNMEmNUu/ecPnlqbfS9dfDrFlw0EFp/VVXwdy5WVcoSZIkSVLzZJikRq1tW/jRj+C99+DBB2HtteGUU6CkJA2Fmzo16wolSZIkSWpeDJNUEIqK0tnenn8+LdttB5dcks4Cd9xxKWySJEmSJEkNzzBJBWerreCBB1KAdNRRcMcd0L8/7LUXPPeck3VLkiRJktSQDJNUsPr0geuug08+gXPPTT2WttkGttwS7r8fysqyrlCSJEmSpKbHMEkFr1s3+PWvYeJEuOYamDYN9tsP+vaFa6+FefOyrlCSJEmSpKbDMElNRrt2cNJJMH48/O1vsOaa8OMfp3mVzj8/hUySJEmSJGnVGCapySkuTj2T/vtfeOaZNOzt17+G0lI48UR4//2sK5QkSZIkqXAZJqnJCiHNoTRqFIwdC4cdBrfckoa/7bsvvPhi1hVKkiRJklR4DJPULPTvD3/6U5qs+6yz4Mkn01nhtt4abr4ZXnkF5s7NukpJkiRJkhq/FlkXIOVTjx5w8cUpULrlFvj97+HYY9O2EGC99WCjjWDgwLRstBH06wetW2dbtyRJkiRJjYVhkpqlDh3g5JPTBN0TJsA778Dbby+9/Ne/YMmStG9xMWy44dJwqeKyd29o2TLb5yFJkiRJUr4ZJqlZKy5Ocyj17Qv77LN0/aJF6axwlQOmN9+EBx6AGNM+LVumXkuVA6aBA1PvpuLibJ6PJEmSJEkNzTBJqkarVkuHulU2fz68996yIdOLL8I99yzdp00bGDBgabjUuzd07Aht20K7dtVfFjl7mSRJkiSpQBgmSSugbVsYMiQtlc2enc4YV3m43OOPw+231+24rVvXHDTVdtm+PXTtmuaDWmutdNmuXf0/d0mSJEmSwDBJqhcdO8Lmm6elshkz4MMPYd68tMyfX/fLiutffFH9PhVzOtVUT0WwVDlkqnq9SxeH5EmSJEmSVoxhktSA1lgDhg5tmGMvXpyCpblzU+A0ZUpaJk9e9vqYMfDII/D1198+RnExdOtWe+i05popoGrhfwxJkiRJavby+tEwhLALcCVQDNwUY7y0yvaQ274bMA84Msb4Wj5rlApFy5ZpWW21FPwMGrT8/efOhalTqw+cKq6/8Ubap6ys+mO0b58eb/XVl15Wvl7TZeXr7dpBCPX/ehSqJUtgzpy0zJ6dlorrlS+Li6FTp6XL6qsvvb7aagZ9kiRJkvInbx8/QgjFwDXAjsAk4OUQwqgY49hKu+0K9M4tmwPX5S4lraL27WH99dOyPOXl8NVXS0OmyZPTcL2vv4ZZs759OWnS0utz5tReR3FxCj8qB04dO678nFFVL1d2QvMYU4i2ZEnq9VV1qW393LnfDoOqC4aqrps/f8VrrU6HDjWHTbXdXn31NG+XJEmSJNVFPr/L3gyYEGP8ECCEcA+wF1A5TNoL+EuMMQL/DSF0CiGsFWOcnMc6pWatqChN6N21K2yyyYrdt6wshSQ1BU81XU6bVv28UYsXr9xzqDqheevW3w6DqguH6lNRUQrJOnZMQU/F5brrfntddftVvSwvh5kzly6zZi3/9uTJ8O67S2+Xly+/3oqebi1bpl5OlS9rul7XdRXXi4vT61JUtOz1qrfrcr26bSGkpaiofq5Xd7uiV11DX9ZVQ++/olZbLQ2dlSRJUtOWzzCpJ/BppduT+Havo+r26QkYJkkFoPJQrPqwZMmKTVpe0+WCBcsGG9WFHbWtr21d+/bLBkBt2tT/B/eVfV1jTD2nlhc+zZnz7ZCt4np16yqWisng63Kf8vK0lJXVHm6pMB13HNx4Y9ZVSJIkqaHlM0yq7mNVXIl90o4hHA8cD1BaWrpqlUlqlFq0WNprRysvhBRydegAvXplXc1SFUMLK0KmykFTXa9X3C4rS8erWMrL6+d61dsVdTfk5Yq8fg25/8rYcMOGfwxJkiRlL59h0iSgpNLtXsDnK7EPADHGG4EbAYYNG5aHJrIkqT6F4MThkiRJUiFaiWlqV9rLQO8QwnohhFbAQcCoKvuMAkaEZAtglvMlSZIkSZIkNR55+044xrgkhDASeBQoBm6JMb4TQjght/16YDSwGzABmAccla/6JEmSJEmSVLu8DjCIMY4mBUaV111f6XoEfpzPmiRJkiRJklR3+RzmJkmSJEmSpAJnmCRJkiRJkqQ6M0ySJEmSJElSnRkmSZIkSZIkqc4MkyRJkiRJklRnhkmSJEmSJEmqM8MkSZIkSZIk1ZlhkiRJkiRJkurMMEmSJEmSJEl1ZpgkSZIkSZKkOjNMkiRJkiRJUp0ZJkmSJEmSJKnODJMkSZKauBDCLiGEcSGECSGEM6vZHkIIV+W2vxlC2DSLOiVJUmEwTJIkSWrCQgjFwDXArsAA4OAQwoAqu+0K9M4txwPX5bVISZJUUAyTJEmSmrbNgAkxxg9jjIuAe4C9quyzF/CXmPwX6BRCWCvfhUqSpMJgmCRJktS09QQ+rXR7Um7diu4jSZIEQIusC6gPr7766pchhE8a6PBdgC8b6NiNUXN7vuBzbg6a2/MFn3Nz0Nye7zpZF1DAQjXr4krsQwjheNIwuP9n777DpKruP46/D0svgnSliIUiKBsFuwbUqCj2rthb1KjRxKiJiT9j1GgSa7D3xILG3ktsWKJRE1GwICoIiogICIq0Pb8/zmxYll12gd25u7Pv1/PMMzP33rnzvVPPfObccwHmhhA+XMXaKtPQXt/gNjcEDW17wW1uCBra9kLD2+YK22AFESbFGDvV1rpDCG/GGAfX1vrrmoa2veA2NwQNbXvBbW4IGtr2apVMAXqUud4d+GIlliHGeD1wfU0XWF5DfH27zYWvoW0vuM0NQUPbXmiY21wRd3OTJEkqbG8AvUMIa4cQmgIHAg+XW+Zh4LDcUd02B2bHGKfmu1BJklQ/FETPJEmSJFUsxrgohHAS8BRQBNwcYxwXQjg+N/9a4HFgF2AC8D1wZFb1SpKkus8wqWq13pW7jmlo2wtuc0PQ0LYX3OaGoKFtr1ZBjPFxUmBUdtq1ZS5H4Gf5rms5GuLr220ufA1te8Ftbgga2vZCw9zmZYTUdpAkSZIkSZKq5phJkiRJkiRJqjbDJCCEMCyE8GEIYUII4awK5ocQwpW5+e+EEDbOos6aEkLoEUJ4PoTwfghhXAjh5xUsMzSEMDuE8HbudE4WtdakEMLEEMK7ue15s4L5hfY89y3z/L0dQvg2hHBquWXq9fMcQrg5hPBVCGFsmWntQwjPhBA+yp2vXsltl/u+r6sq2eY/hxA+yL1uHwghtKvktst9D9RVlWzzuSGEz8u8dnep5Lb17nmuZHvvLrOtE0MIb1dy23r5HKvhsg1mGyw3v2Ce54bQ/gLbYGWmFWwbrKG1v8A22AqLMTboE2kgyo+BdYCmwBigf7lldgGeAAKwOfB61nWv4javAWycu9wGGF/BNg8FHs261hre7olAx+XML6jnudy2FQFfAmsV0vMM/BjYGBhbZtqfgLNyl88CLq7k8Vju+76unirZ5h2BxrnLF1e0zbl5y30P1NVTJdt8LnB6Fberl89zRdtbbv4lwDmF9Bx7apgn22C2wQr1eS6zXQXZ/sptg22wWNhtsIbW/qpsm8vNtw1W5mTPJNgUmBBj/CTGuAAYBexRbpk9gL/F5DWgXQhhjXwXWlNijFNjjP/JXZ4DvA90y7aqOqGgnudytgc+jjFOyrqQmhRjHA18U27yHsBtucu3AXtWcNPqvO/rpIq2Ocb4dIxxUe7qa0D3vBdWiyp5nqujXj7Py9veEEIA9gfuymtRUu2wDWYbrFRBPc9lFGT7C2yDlZlWsG2whtb+AttgK8owKX2BTy5zfQrLfqlXZ5l6KYTQC9gIeL2C2VuEEMaEEJ4IIQzIb2W1IgJPhxDeCiEcV8H8gn2egQOp/IOv0J7nLjHGqZAa7UDnCpYp5Of6KNK/uxWp6j1Q35yU61Z+cyVd6Qvxed4GmBZj/KiS+YX2HKuw2QazDVaqUJ/nhtT+AttgDaUN1hDbX2AbbBmGSak7bXnlD3FXnWXqnRBCa+A+4NQY47flZv+H1CW3GPgr8GCey6sNW8UYNwZ2Bn4WQvhxufmF+jw3BXYH/lHB7EJ8nqujUJ/rs4FFwB2VLFLVe6A+uQZYF/gRMJXU7bi8QnyeD2L5/4gV0nOswmcbzDZYqYJ7nm1/VargnmtoUG2whtr+AttgyzBMSklpjzLXuwNfrMQy9UoIoQmpEXNHjPH+8vNjjN/GGOfmLj8ONAkhdMxzmTUqxvhF7vwr4AFSF8yyCu55ztkZ+E+McVr5GYX4PAPTSrvG586/qmCZgnuuQwiHA7sCI2KMFX5hV+M9UG/EGKfFGBfHGEuAG6h4WwrqeQ4hNAb2Bu6ubJlCeo7VINgGsw1WquCeZxpe+wtsgxV8G6whtr/ANlhlDJPgDaB3CGHt3D8IBwIPl1vmYeCwkGwOzC7twlkf5fb3vAl4P8Z4aSXLdM0tRwhhU9JrZUb+qqxZIYRWIYQ2pZdJg+WNLbdYQT3PZVSaohfa85zzMHB47vLhwEMVLFOd9329EUIYBpwJ7B5j/L6SZarzHqg3yo2lsRcVb0tBPc/AT4APYoxTKppZaM+xGgTbYBUvU1DfzQ24DdbQ2l9gG6zg22ANtP0FtsEqVtnI3A3pRDqCxHjSqPNn56YdDxyfuxyAq3Lz3wUGZ13zKm7v1qSuhu8Ab+dOu5Tb5pOAcaTR918Dtsy67lXc5nVy2zImt10F/zzntqklqXHStsy0gnmeSY20qcBC0r8gRwMdgGeBj3Ln7XPLrgk8Xua2y7zv68Opkm2eQNo3vfT9fG35ba7sPVAfTpVs899z79N3SA2UNQrlea5oe3PTby1975ZZtiCeY08N91TRe7SQv5uxDdYg2mAUePsrtw22wQq8DVbJ9hZs+6uybc5NvxXbYMucQm7jJUmSJEmSpCq5m5skSZIkSZKqzTBJkiRJkiRJ1WaYJEmSJEmSpGozTJIkSZIkSVK1GSZJkiRJkiSp2gyTJEmSJEmSVG2GSZIkSZIkSao2wyRJkiRJkiRVm2GSJEmSJEmSqs0wSZIkSZIkSdVmmCRJkiRJkqRqM0ySJEmSJElStRkmSZIkSZIkqdoMkyRJkiRJklRthkmSJEmSJEmqNsMkSZIkSZIkVZthkiRJkiRJkqrNMEmSJEmSJEnVZpgkSZIkSZKkajNMkiRJkiRJUrUZJkmSJEmSJKnaDJMkSZIkSZJUbYZJkiRJkiRJqjbDJEmSJEmSJFWbYZIkSZIkSZKqzTBJkiRJkiRJ1WaYJEmSJEmSpGozTJIkSZIkSVK1GSZJkiRJkiSp2gyTJEmSJEmSVG2GSZIkSZIkSao2wyRJkiRJkiRVm2GSJEmSJEmSqs0wSZIkSZIkSdVmmCRJkiRJkqRqM0ySJEmSJElStTVe3sy33nqrc+PGjW8ENsDgaWWVAGMXLVp0zKBBg77KuhhJkpRftqfqFdttkiRVw3LDpMaNG9/YtWvX9Tt16jSzUaNGMV9FFZKSkpIwffr0/l9++eWNwO5Z1yNJkvLL9lT9YbtNkqTqqerfsQ06der0rQ2fldeoUaPYqVOn2aR/IyVJUsNje6qesN0mSVL1VBUmNbLhs+pyj6Hd2iVJaphsT9UjttskSapanf6i/PLLL4v69evXv1+/fv07duxY3Llz54Gl13/44YewvNuOHj265RFHHNFjRe6vW7duG06dOnW5u/5JkiTVJ/luT0mSpMJXp4OTrl27Lv7ggw/eA/jFL36xZuvWrRefd95500rnL1y4kCZNmlR42x//+Mff//jHP/4+T6VKkiTVSYXanlpe3ZIkqXbV6Z5JFdlnn316HXPMMd0322yzPieeeGL3559/vuVGG23Ub/311++/0UYb9RszZkwzgEcffbTNtttuux6khtN+++3Xa9NNN+3bvXv3Dc8///zOVd3Pueee26V3794DevfuPeC8887rDPDtt982Gjp06Hp9+/bt37t37wE33HDD6gAnnnhit3XXXXdAnz59+h933HHda3P7JUmSVlVttqdGjBjRc4MNNlh/vfXWG3DaaaetWTr9xRdfbLnRRhv169u3b/8NN9xw/ZkzZzZatGgRxx13XPc+ffr079OnT/8LLrigMyzdW3z06NEtN910076lNRx00EFrbbXVVr333nvvtT/88MOmgwYN6tu/f//1+/fvv/4zzzzTqvT+fvvb33bp06dP/759+/Y/8cQTu40bN65Z//791y+d/+677zYbMGDA+kiSpBVWp3smVebjjz9u/sorr4xv3Lgx33zzTaN///vfHzRp0oQHH3ywzRlnnNH9qaee+rj8bSZMmND81Vdf/XDWrFlF66+//ga/+tWvpjdr1qzC8QteeumllnfeeWeHt9566/0YI4MGDVp/++23n/PRRx8169q168IXXnhhAsCMGTOKpk2bVvT444+v/sknn4xt1KgRX3/9dVFtb78kSdKqqq321KWXXvp5ly5dFi9atIgtt9yy7+uvv96iuLj4hxEjRqx7xx13fDxkyJDvv/nmm0atW7cuueSSSzpNmjSp2bhx495r0qQJ06ZNq7Id9c4777R8/fXXP2jdunWcM2dOo5deeml8y5Yt47vvvtvsoIMOWmfs2LHv33PPPas99thjq7/11lsftGnTpmTatGlFXbp0WdymTZvFr776aostt9xy3nXXXdfx4IMPnlGTj6kkSQ1FtcOko46ix9ixtKzJO99gA76/+WYmr+jt9t5775mNG6fSv/nmm6IDDjhg7YkTJzYPIcSFCxdWuO//jjvuOKtFixaxRYsWi9q3b79wypQpjdddd92FFS37wgsvtN5ll11mrbbaaiUAw4cPn/n888+32X333WefffbZPU444YRue+yxx+xhw4bNXbhwIc2aNSs58MAD1xo+fPjsAw44YPaKbo8kSWoYGkJ76rbbbmt/6623dly0aFGYPn16kzFjxjQPIdC5c+eFQ4YM+R6gffv2JQDPPffcascff/z00t3VunTpsriquocNGzardevWEWDBggXh6KOPXuu9995r0ahRIyZNmtQM4JlnnlntkEMO+bpNmzYlZdd7xBFHfH3DDTd03HTTTSc/9NBDq7/xxhvvr+jjJkmS6uFubgCtW7cuKb185plndhsyZMicjz76aNwjjzwyYcGCBRVuU9l/zYqKili0aFGlA07GWPEBVwYOHDj/P//5z3sbbrjhvLPPPrvb6aefvkaTJk14++23399nn31mPfjgg+2GDh3aexU2TZIkKS9qoz31wQcfNB05cmSXF198cfz48ePf22677Wb/8MMPjWKMhBCWaWBVNr2oqCiWlKTy5s2bt1QtrVq1+l/dF1xwQZfOnTsvfP/99997991331u4cGGjMutdpv7DDz985vPPP9921KhR7TbccMPvu3btWmV4JUmSllXtnkkr849XPnz77bdF3bt3XwBw3XXXdayJdW633XZzjzrqqF5/+MMfvowx8vjjj69+6623fjJx4sQmnTt3XnTiiSd+06ZNm5Lbbrutw+zZsxvNnTu30QEHHDB76NChc/v06bNhTdQgSZIKT6G3p2bOnFnUokWLkvbt2y+ePHly4xdeeKHtkCFD5hQXF/8wbdq0pi+++GLLIUOGfD9z5sxGrVu3LvnJT37y7bXXXttp+PDhc0p3c+vSpcvi7t27L3jllVda7r///t/ec889q1d2f7Nnzy7q3r37gqKiIkaOHNlh8eKUDQ0bNuzbCy64YM1jjz32m7K7ubVs2TIOGTJk9i9+8YueI0eOnLiy2ylJUkNXL3smlXXmmWd+ee6553bfeOON+5U2IFbV1ltv/f3BBx88Y+ONN15/0KBB6x966KHTt9pqq3lvvfVWix/96Efr9+vXr//FF1+8xjnnnDN11qxZRcOGDevdp0+f/ttss03f888/v042EiVJkipTU+2pLbbYYt4GG2zwfe/evQcceuihvQYNGjQXoHnz5vGOO+74+JRTTunZt2/f/kOHDu3z/fffNzrttNOmd+/efUG/fv0G9O3bt/9NN93UHuCcc8754owzzug5aNCgvkVFRRV3GQdOPfXUr+66664OxcXF/caPH9+8RYsWJQD77rvvtzvvvPOs0nbbH/7wh66ltznssMO+Adh7772/XekNlSSpgQuV7dIFMGbMmInFxcVf57GegjVmzJiOxcXFvbKuQ5Ik5ZftqbrlnHPO6TJ79uyiK6644ovKlrHdJknS8tXLo7lJkiRJK2qHHXZYd9KkSc1efPHF8VnXIklSfWaYJEmSpAbhmWee+TjrGiRJKgT1fswkSZIkSZIk5Y9hkiRJkiRJkqrNMEmSJEmSJEnVZpgkSZIkSZKkaqvTYdKmm27a97777lut7LTzzjuv8yGHHNJzebcZPXp0S4AhQ4as9/XXXxeVX+YXv/jFmuecc06X6k6XJEmqr/LdnpIkSYWvTodJ++2334y77rqrfdlp9913X/tDDjnkm+rc/sUXX5zQsWPHxbVTnSRJUt1XqO2pkpISFi+uc2VJktQg1Okw6dBDD5357LPPtp03b14A+PDDD5t+9dVXTXbccce5I0aM6LnBBhusv9566w047bTT1qzo9t26ddtw6tSpjQHOPPPMrr169dpgyy237PPRRx81q+q+X3311RbFxcX9+vTp03+HHXZYd/r06UUA559/fud11113QJ8+ffrvuuuu6wA89thjrfv169e/X79+/ddff/3+M2fOrNOPqyRJajjy3Z6688472w4cOLDf+uuv33/LLbfsM3ny5MYAs2fPbrTvvvv26tOnT/8+ffr0v/XWW9sB3Hvvvav1799//b59+/bfYost+sCyvZ569+494MMPP2z64YcfNl1nnXUGHHLIIT0HDBjQ/+OPP25a2Ta8+OKLLTfaaKN+ffv27b/hhhuuP3PmzEaDBg3q++qrr7YoXWbjjTfu9/rrr//vuiRJqp46HXp07dp1cXFx8Xf33XdfW4Dbbrut/e677z6zUaNGXHrppZ+PHTv2/Q8++GDcK6+80mZ5DYGXXnqp5QMPPND+3Xfffe/RRx+dMGbMmFZV3fcRRxyx9oUXXjhl/Pjx7w0YMGDemWeeuSbAlVde2XXs2LHvjR8//r1bb711EsAll1zS9corr5z0wQcfvPfaa6990Lp165KaegwkSZJWRb7bUzvssMPct99++4P333//vX333feb8847ryvAWWedtcZqq622ePz48e+NHz/+veHDh8/54osvGp900km97r///o8//PDD9x588MGPq9qeiRMnNj/yyCNnvP/+++/16dNnQUXb8MMPP4QRI0ase/nll3/24Ycfvvfiiy9+2Lp165Ijjjji6xtvvLEjwDvvvNNswYIFYbPNNpu3co+sJEkNV+NqL3nUUT0YO7Zljd77Bht8z803T17eIvvvv/83d9999+qHHHLIrPvvv7/9jTfeOBFSQ+jWW2/tuGjRojB9+vQmY8aMaV5ZY+D5559vvcsuu8xq06ZNCcCOO+44a3n3OWPGjKI5c+YUDR8+fC7AscceO2O//fZbB6Bv377z9tprr7V33333WSNGjJgFsPnmm889/fTTe+y///7fHHTQQTPXXXddwyRJkrSsBtCe+vTTT5vuueee3adPn95kwYIFjXr06DEfYPTo0auNGjXqk9LlOnXqtPjOO+9su+mmm87p16/fAoAuXbpUud/aGmussWD77bf/rvR6RdsQQqBz584LhwwZ8j1A+/btSwCOOOKImX/+85/XmD9//pRrr72248EHH/x1VfcnSZKWVad7JgGMGDFi1iuvvLLayy+/3PKHH35otPXWW3//wQcfNB05cmSXF198cfz48ePf22677Wb/8MMPy92WEEKN1PP8889/9LOf/Wz6W2+91aq4uLj/woULufDCC7+88cYbJ82bN6/Rlltuuf5///vf5jVyZ5IkSTUgn+2pk046qeeJJ5741fjx498bOXLkpPnz5zcCiDEuc/uKpgE0btw4lpQs+W9u/vz5/1uoZcuW/5tR2Tbk1hvLr7dNmzYl22yzzbd33nlnu4cffrj90UcfXa1xoyRJ0tKq3zOpin+8akvbtm1LNt988znHHHNMr7333vsbgJkzZxa1aNGipH379osnT57c+IUXXmg7ZMiQOZWtY7vttpt71FFH9frDH/4wdeHCheGZZ55pd/jhh0+vbPkOHTosXm211RY/+eSTrYcNGzb3pptu6rDFFlvMXbx4MR9//HHT3Xbbbc6OO+44d80112w/e/bsomnTpjXedNNN52266abzXn/99VZjx45tvtFGG/1QG4+HJEmqxxpAe2rOnDlFPXv2XAhw6623diidPnTo0G8vvfTSzjfnHoPp06cXbbvttt/98pe/XOuDDz5o2q9fvwXTpk0r6tKly+JevXrNf/zxx9sBvPzyyy0///zzCsdnqmwbiouLf5g2bVrTF198seWQIUO+nzlzZqPWrVuXNGnShOOPP/7rffbZZ71NNtlkbnV6QkmSpGVVP0zK0IEHHvjN4Ycfvu5dd931CcAWW2wxb4MNNvi+d+/eA3r27Dl/0KBBc5d3+6233vr7vfba65sNNthgQLdu3eZvuummy10e4JZbbvn0hBNOWOuUU05p1LNnz/l33XXXxEWLFoWDDz547Tlz5hTFGMNPf/rTaR07dlz8y1/+cs1XX311tUaNGsU+ffrM23fffWfX1LZLkiTVhHy1p84+++wvDjrooHW7dOmyYPDgwd999tlnzQD++Mc/Tj3yyCN79u7de0CjRo3ib37zmy8OP/zwWVdeeeXEvfbaa72SkhI6dOiw8NVXX/3osMMOm3nHHXd06NevX/8f/ehH36211loV/klX2TY0b9483nHHHR+fcsopPX/44YdGzZs3Lxk9evT4tm3blmyzzTbft2rVavGRRx7pLm6SJK2kEOMyPYD/Z8yYMROLi4v9oq0BY8aM6VhcXNwr6zq0RK77e+8Y44Q83uf5wPHAohhj1xpe9wvA7THGG2tyvbUthLANcGOMsW8Nr7fK5zeE0AK4B/gx8DTwIHB4jHHH6q6jtoQQegGfAk1ijItW4HZHAMfEGLeupdLqhRDCE8CoGONtWdci2Z6qWyZOnNhk6NChfT/++OOxRUVFFS5ju61uCiHcCkyJMf42j/e5FXArsAZwSIzxwRpc9xHU0+/sEMI44GcxxhdqcJ23Uo3nN4RwAnAu0ApYC5gEDIwxfpLFa6RcbS+wEu3xLNucdUUIYQRl2uGq++r8mEmqH0IIE0MIP8m6jroshNAD+CXQv6aCpIrGg6jrQggxhLBe6fUY40s1HSStgH2BLkCHGON+McY7KvsCCyHcmgsDMxFCeCGEMLSSebfmGqQNUgjh3BDC7WWnxRh3rokgKbfuc1d1PZLqhpEjR3bYfPPN1z/nnHM+ryxIylru835mCKHCXfuUd+cBI2OMrWsiSKqP39kVtYFijANqMkhagVqaAJcCO+aekxm5808qWHZoCGFKvmssc/9H5MKtiuYNzQVPDVIIoVfuN8H/9pRaXjt8JdY9cVXXo6oZJkn5sxYwI8b41YresOwHrWrUWsD4Fen5s7J8DiUpeyeddNKML7/88p2jjjpqZta1VCTXK3UbIAK75/m+/Z6q2FrAuJW5oY9pregCNGcln5MV4fMnLZ9hkmpVCKFZCOHyEMIXudPlpf+0hRA6hhAeDSHMCiF8E0J4KYTQKDfvzBDC5yGEOSGED0MI21ew7s1DCF+GEIrKTNsrhPBO7vKmIYR/5dY/NYQwMoTQtJI6XwghHFPm+hEhhJfLXO8XQngmV+eHIYT9y8zbJYTwXq7Wz0MIp1ew/p8AzwBrhhDmlv5LEULYPYQwLlfjCyGE9cvcZmLucXgH+G5Fv9BCCEeFEN7P/bv5VAhhrTLzYgjhxBDCR7m6/xBCWDf3eH0bQrin9LEq/VcnhHBGCOGr3GO5Z267x+cek9+UWXelj3sIYXRusTG5x+GAsv8a5a7PLXOaX/qvTe619JcQwmchhGkhhGtD2k2t9H5/lbu/L0IIR1Xj8fk9cA5Qep9Hl3/eyyx7HDACOCO37CO56WuGEO4LIUwPIXwaQjilzG3ODSHcG0K4PYTwLXBECKFtCOGmXJ2fhxDOL339hhCKctv3dQjhE2B49Z7pKrdzyxDCGyGE2bnzLcvMeyH33L+Sex08HULoWGb+YSGESSGEGSGE34Vq9EAMIbQI6R/Mmbn3xa9CmX8FQ7meaaHcv50hhF1DCG/nXj+vhhAGlpm3zOdCCGEY8BuWPI9jymzbMbnLjUIIv81ty1chhL+FENrm5pX+M3Z47rX1dQjh7FV5zCVpFRwGvEbarerwsjNCCD1CCPfnvnNmhBBGlpl3bEjf+XNyn70b56ZX+pkblny/nxlC+BK4JYSwekhts+m5z/FHQwjdy9y+fQjhltx37cwQwoO56WNDCLuVWa5J7vP0R+U3MFfnrmWuN84tW1rzP0Jq380OIYwOIQyo6IGq6Du77PaG5bQbwnLaoOXW9zGwDvBI7jumWUjf/Q/nbjchhHBsmeWX+e6vqPbKhOW3N28NIVwdQngiV8srIYSuIbWtZ4YQPgghbFRm+Ym57+B3QgjfhdT+6JK7/ZwQwj9DCKuXWb7Cxz1U3gb6X5sg9ziWtt2+yz0PvXLzlve9vlEI4T+5eu4mhUTLe3z6AB/mrs4KITyXm77U6zw3rRXwBEva3nNzz12jEMJZIYSPQ3of3RNCaJ+7TWmb4OgQwmdA6fqX16beIffYzw7pPbnKhxAPqb34t5Deh5NCasOU/kY6IoTwcu61PTOk9ufOZW67du75K32Orwrlem9Xcp+HhiVtvrPLPb/l22r/a7vnri+vPbxpCOHNkH5fTAshXJqbVfqboPS1s0VY9vfXSrdhlR9VhUklJSUlq/yGaOhyj2FJlQsWprOBzYEfAcXApkDpPsy/BKYAnUj/MvwGiCGEvsBJwCYxxjbATsDE8iuOMb4GfAdsV2bywcCducuLgdOAjsAWwPbAiSu6Abkvo2dy6+0MHARcHZY0bm4CfpqrdQNyXzzlav0nsDPwRa4r7hG5L8S7gFNzj8HjpMZK2cDrIFKo0K6i3jMxxgrfnyGEPUmP5965db+Uu6+yhgGDSM/PGcD1pMZCj9x2HFRm2a6kL/hupADmBuCQ3O23Ac4JIayTW7bSxz3G+OPcMsW5x+Hucttzd256a2BN4JMydV8M9CG9ltYrUwshBQqnAzsAvYEqd7mMMf4fcCFQep83LWfZ64E7gD/llt0t96X+CDAmV8v2wKkhhJ3K3HQP4F6gXe72twGLcvVvBOwIlIaYxwK75qYPJu2CV7aGoZV1J48xHhFjvLX89Fzj6DHgSqADqVv4YyGEDmUWOxg4kvTabkp6HAkh9AeuJr0m1gDa5razKv8HrJs77US5H0PLE9IPiZuBn+bqvQ54OKTGe4WfCzHGJ1n6eSyuYNVH5E7bkn4UtAZGlltma6Av6Xk8J+SC3RjjuTHGc6u7DVIlbE/VIxm32w4jfV/cAewUQugC6Q8H4FHS2DC9SJ/Ho3Lz9iONH3MYsBqpR9OMat5fV6A9qffNcaTfBrfkrvcE5rH05+XfgZbAANL3xmW56X8jtQtK7QJMjTG+XcF93sXSbYydgK9jjP/JXX+C9F3eGfgP6bFYGZW2G6ikDVp+BTHGdYHPgN1y3zHzc/VPIbVT9gUuDEv/6Vn+u7/8Oiv7zq6qvQmwP6kd3RGYD/yL9Bh1zN3npSxtH1LbqA+wG+mx/U1u+UbAKWWWrfBxr6gNVME2tSvTfruC1O78vIrv9aaksSr/TnoN/iNXb6VijONJrz1IbePtlrPsdyzd9m4dY/wit817AkNIz+FM4KpyNx8CrE96D+5JJW3qXHhxH0uek4+BrcrUcGuM8YhK6nshxji0kvL/Smp3rZOr5TBSW63UZqRQrSPwJ+CmEELpd8ydwL9Jj/e5wKGV3Mf/5Np81+SWXTN32+7LvdGS21bVHr4CuCLGuBqpbXhPbnrpb4LS186/yq13pduwMcaJMcZe1alfq6aqMGns9OnT29oAWnklJSVh+vTpbYGxWdeSkRHAeTHGr2KM04Hfs+RDbSHpR+paMcaFubFzIimMaAb0DyE0yX0gfFzJ+v/XIAkhtCE1Xu4CiDG+FWN8Lca4KMY4kfQFNmQltmFX0o/WW3Lr+g/pi6P0x/7CXK2rxRhnlmkMVeUA4LEY4zMxxoXAX4AWwJZllrkyxjg5xjhvBWv+KfDHGOP7uRDqQuBHZf9JAS6OMX4bYxxHen0+HWP8JMY4m9Sg2KjMsguBC3J1jiJ9eV0RY5yTu/04YCDUzOOe+2K6E3ghxnhd7gvyWOC0GOM3McY5uW06MHeT/YFbYoxjc42Hc1fk/lbSJkCnGON5McYFuX31byhTE8C/YowPxhhLSA38nYFTY4zf5XZ3vKzcNlyee76/Af5YAzUOBz6KMf4993zcBXxAalCWuiXGOD73GruH1OiG9Pp+JMb4coxxAakBXp0xuvYnvVa+iTFOJjUCqutY4LoY4+sxxsUxjXk0nxR4rsjnQnkjgEtzr++5wK+BA8PSvf1+H2OcF2McQ2oQVRRKSSvL9lQ9kWW7LYSwNSnEuSfG+Bbph+nBudmbkn7k/Sr3HfJDjLH0H/xjSD/034jJhBjjpGrebQnwfzHG+bnPwBkxxvtijN/nvmsvIPcdHkJYg/Q9dnyuvbMwxvhibj23A7uEEFbLXT+UFBJU5E5g9xBCy9z1sn8EEmO8Ode+mE/6Pi8Oud6k1VWNdkNlbdCq1tuD9OfDmbnn4G3gRpb+wf6/7/4VbL9V1d4EeCDXzvoBeAD4Icb4txjjYuBulm67Afw1xjgtxvg5KQR5Pcb439xj+0DZ5WvocT+A9Hzuk2szLu97fXOgCantszDGeC/wxorc30r6KXB2jHFKmW3dt1yb4Nzc+2wey29T7wK8F2O8N7e9lwNfrkpxueD4AODXuedjInAJS7/GJsUYb8g977eRXstdQgg9Se3Tc3Jt05eBh6txt/sCj8YYR+cek99R/UC9qvbwQmC9EELHGOPcmDoDVMeqtGGVJ8vdbWbRokXHfPnllzd++eWXG+AucSurBBi7aNGiY6pcsjCtSfoXrdSk3DSAP5M+wJ/OhenXxxgvijFOCCGcmps3IITwFPCLmP5NKO9O4NWQjuqwN/Cf0gZUrufPpaReHi1Jr/e3VmIb1gI2CyHMKjOtMUsaSfuQ/pG4KKRd0s6K5dL1Siz12MQYS0IIk1m698fklai3tOYrQgiXlJkWcusuvc9pZebNq+B62UHCZ+S+sErnVXT71lBjj/sFQBuW/GPWKbeut5b88UIASndxXLPcfVS3Eb0q1iJ1nZ5VZloRqbFWanK55ZsAU8tsQ6Myy6xZbvma2Iby77/S9ZZ9jZVt9HxP7nksX0+M8fsQQnX+6V6V7VgLODyEcHKZaU2BNWOML67A50JFNZX/HGpM+je6VGWPg7TKbE/VK1m22w4n/bFTeuS/O3PTLiP1Gp4UKx7jrwcpeFoZ03PBBAC5gOcyUu/l0l2g2uR+4PYAvokxLjPeVIzxixDCK8A+IYQHSKHTzyu6w1w7731gt5B2mdqdXKiRu58LgP1I3/2lP2g7ArNXYLuqajdU2AatxnrXJD0Gc8pMm0Rq85Ralbbb8tqbUHXbrfx3V7WWr4nHPaRd7EaSBsaeXmabKvxeJ/1B9Xm5EC9f7bcHQghlw5LFLN0mKN9+q6xNXb6tFHNt+VXRkfQYlW+3VNh2y7XPID2XHUmvz+/LLDuZ9N5dnvLb8V0123xQdXv4aNIg9h+EED4l/Xn3aDXWuyptWOXJcsOkQYMGfUWeB/9TwfmCpQcu7JmbRu6L+JfAL3NdeJ8PIbwRY3w2xngncGfuH67rSF2Vl+mmGWN8L4QwidRoWeqfLVJ3zf8CB8UY5+R+iO5bfh0535EaHaXKBimTgRdjjDtUdMMY4xvAHiEdXeIkUjJe1Yc2pMdhw9IruX/RegCfl119NdZTkcmk3iEr2zV8VazI476MEMKBpN5mm+T+5QH4mtToGZD7d628qSz9mPdcmcKrUP65mAx8GmPsXc3bTCb9G9exkh8DtbENpe+/snoCT1bjtlNJu30BaSwkUjfj6tyuB0u/58v6nmXfa6X73Ze+bi+oaMXL+Vyo6n1S/nHoSdrdcBrV7MYtrQrbU6pK7jN2f6AopPGLIPXGbBdCKCZ9PvYMITSu4DtkMmn3kYos7zMXlv38/CXps3+zGOOXIY159F/Sj+fJQPsQQrsY46wK7us2Ui+pxqTeORV9X5cq7VneiNSzo/Rw6AeTdhP7CWmIg7ak3ZAq6tW3VNsthFC27bbcdsPy2qDLqRnS90n7EEKbMoFST2qu7VZpe7OWVfW4L3ebQgidSD2dToox/rfMrEq/10MIQ4BuIYRQJlDqycoHoxWpqO7JwFExxlcqqKlXBbertE0dQuhNmbZbmbb8qvia1JtnLeC93LTyr7HKTCW9PluWCZSqU89U0m59wP9C5bJtvqp+J1XaHo4xfgQclNvrYG/g3tyuaivadoPqt2GVJ/47pprUJITQvMypMamx8NsQQqfcfsXnkLpClw7It17ug/db0r8Ci0MIfUMI24U0UPcPpMbA4orvEkgB0imkfW//UWZ6m9x654YQ+gEnLGcdbwN7hxBahjSA39Fl5j0K9AlpYLomudMmIYT1QwhNQwgjQghtc8FH6XZUxz3A8JAGEW5CatTMB16t5u2X51rg12HJ4IltQxpTIR+qetynkfYBX0buX62/AnuW+VeLmHYTuwG4LITQObdst7Bkf+x7SANc9899Af5fTW5QJXX/G/g2pIFLW4Q0gPYGIYRNKrpxjHEq8DRwSQhhtZAGgFw315gq3YZTQgjdQxoQ86waqPlx0mv34JAGOD0A6E96TVflXtK/xluGNK7B76neoJL3kF57q4c0aOvJ5ea/DRyce7yGsfQukDcAx4cQNgtJqxDC8BBCmyo+F6YBvUIFg6fm3AWcFtKglK1ZMsZSrR/FT5KqaU/SZ1p/0q4aPyL9uHuJNF7Kv0k/+C7KfTY2DyGUjs1yI3B6CGFQ7rNzvbBkt/a3qfwztyJtSJ+vs0Ias+R/36e577EnSOP4rJ5rD/24zG0fBDYm9Uj6WxX3M4o0buAJLP1HYBtSW2gG6cfrhctZxxhST9UfhRCaU2YX96raDZW1QauomZh2334V+GPuORhIajPWxJ93lbY3a2DdVanqcV9e260xaXe8O2K5sTBZzvc6abynRaS2T+MQwt6k3Tlr0jSgQ1h6d71rgQtK3yO53yh7LGcdy2tTP0Z6De6dexxOYemgZYXl9gS4J1djm1ydvyD3+6mK204C3gTOzf1G2YKldwurzL3AriGErXNtvvNYOid4m7Qba/uQQttTy8xbbns4hHBICKFT7j05K3ebxcB0Ug+4Cl9XrFobVnlimKSa9DipAVJ6Ohc4n/Sh9g7wLmlAv9KjAfQG/gnMJX2hXB3TAMPNgItIyfyXpEHV/ne0sArcBQwFnivTNRzSIGwHA3NIX2blv+DKugxYQPrSuY0yjYLcP087kvb9/SJX08W5OiH1jJgY0lE7jmfpASgrFWP8MLfsX0nbuhtpgMcF1bl9Fet+IFfjqFxdY0m9t/Khqsf9XOC2kI7qsX+5eXuQutW/HJYceeOJ3LwzgQnAa7lt+ie5njMxxidI+6k/l1tmmUHQa8BNpPF6ZoUQHsx92e9GavB/SnoObyT9m1eZw0hdl98j/eN3L2k/d0iP1VOkxvF/gPtXteAY4wzSGAy/JDUQzwB2Lfc+qey240hB0CjSD5g5wFekxuby/J7UDflTUnhWfsyMn5Met1mksYweLHOfb5LGVxhJenwmsORIOMv7XCgNkWeEECoas+zmXB2jc3X9wLIhlyRl6XDS+B+fxRi/LD2RPg9HkML83UgDSX9G6l10AECM8R+kXZTuJH1WP0ga0BiW85lbictJ4zd+TTqqXPleAIeSek18QPpOOLV0Rm7ckvuAtaniOywXTP2LNE5k2XbC30jfIZ+TvisrHV8lpsGYzyO1Bz4Cyh+NtdJ2A5W3QavjINIg6F+QeuP8X4zxmWretlLVaG/Wpqoe96XaQOXmdScdjOXUsPQReXsu73s9197dO3d9Jun1vMptn7JijB+Qfid8kqt9TdKA0A+TdnGck9vWzZazjkrb1Ln21H6k9skM0utqmR5PK+FkUm+gT0iv6ztJbZnqGEE6AM4M0m+uu6mi7ZZr8/0sdz9TSc9H2R6Mfye1TyeS2nZ3l7ltVe3hYcC4EMJc0mN/YEzjjX1P+tx6JffcbF6uppVuwyp/Qqx6rDlJUgOX69EzC+gdY/x0BW43FLg9xujuZJJU4EII5wB9YozV+mNNUu0KIdwNfBDTUYxX5HYTgWNiOiK1VCF7JkmSKhRC2C2kXT9bkY42+C7pXylJkpaS2y3uaOD6rGuRGqrcrpHrhjScwjBSr/8HMy5LBcowSVJBCyGMK9ftuvQ0Iuva6oE9SF3tvyB13T4wd6SSJyp5TJe3O6qkjIQQbg4hfBVCqPBw97mxTK4MIUwIIbwTQtg43zWqfgshHEsaiPeJGOPorOtR/RZC+E0l7Ywnqr51g9cVeIG0C+eVwAkxxv+GNMZrRY/puOWuTVoOd3OTJEkqYCENkjwX+FuMcYMK5u9CGqNjF9LYIVfEGCsdQ0SSJMmeSZIkSQUs11Pkm+UssgcpaIoxxtdIh6NfYznLS5KkBs4wSZIkqWHrRtpFqdSU3DRJkqQKNc66gJrQsWPH2KtXr6zLkCRJteStt976OsbYKes6ClSoYFqF4yCEEI4DjgNo1arVoH79+tVmXRIxwqJF6bR48ZLL5U/l52UhhCXn1blc3WXLr7+yaVXNr2xaZetYmeWqU3N1rehty4/eUtPXV/bxXZFpKzp9RdehwtWmDTRrVjvrrqwNVhBhUq9evXjzzTezLkOSJNWSEMKkrGsoYFOAHmWudycNvL+MGOP15I7WNXjw4Gj7S9W1YAHMnJlOs2al8xkz4Jtv0nllp7lzK19ns2bQoQOsuWY6b98+nZeeWreGJk2gcePaORUVLTn3x7ukQlVZG6wgwiRJkiSttIeBk0IIo0gDcM+OMU7NuCbVMTHC998vGwhV5/qsWem2lQkB2rVbEgJ17QoDBiwdDJUPijp0gJYtDXEkKSuGSZIkSQUshHAXMBToGEKYAvwf0AQgxngt8DjpSG4TgO+BI7OpVLUlRpg3D779Np1mz676fPbsZQOhhQuXfz9t26ZQaPXV06lv36Wvl79cGgq1a5d690iS6g/DJEmSpAIWYzyoivkR+FmeytEqWLwYPv8cPvkEpkypfjD07bfVG0eoZcsUCK22Wjqtvjr06rV0CFQ+FCq93ratgZAkNSSGSZIkSVIdMXcufPopfPxxCo0++WTJ5YkT09hD5TVpsiQEKj1fa62lr1d1vtpqafwfSZKqw68MSZIkKU9KSmDq1GWDotLLX3219PJt28K668LAgbDXXrDOOul6jx6pR9Bqq6WBqB07SJKUT4ZJkiRJUg2aPz8FQxX1Lvr0U/jhhyXLNmqUgqF11oHdd18SFq2zTjq1b5/ddkiSVBnDJEmSJGkVLFgA//43PP88vPACvPrq0oFR69YpGOrXD3bZZenAaK21oGnTzEqXJGmlGCZJkiRJK2DhQnjzzRQePf88vPJKOloaQHExHH88DB6cAqN114WOHd0NTZJUWAyTJEmSpOVYtAjeeiv1Onr+eXj5ZfjuuzRvww3hmGNg223hxz9Oh7qXJKnQ5TVMCiEMA64AioAbY4wXVbDMUOByoAnwdYxxSB5LlCRJUgO3eDH8979Ldlt76SWYMyfN698fjjgChg6FIUOgU6cMC5UkKSN5C5NCCEXAVcAOwBTgjRDCwzHG98os0w64GhgWY/wshNA5X/VJkiSpYVq8GMaMWdLzaPRo+PbbNK9fPxgxIvU8GjIEunTJtFRJkuqEfPZM2hSYEGP8BCCEMArYA3ivzDIHA/fHGD8DiDF+tcxaJEmSpFVQUgLvvruk59GLL8KsWWle795w4IGp59HQobDGGtnVKUlSXZXPMKkbMLnM9SnAZuWW6QM0CSG8ALQBrogx/i0/5UmSJKnQTZyYehlNnJiur7su7LNPmjZ0KHTrlmFxkiTVE/kMkyo6hkUsd70xMAjYHmgB/CuE8FqMcfwyKwvhOOA4gJ49e9ZwqZIkSSo0334Lu+2WeiHdcgtstx3YjJQkacXlM0yaAvQoc7078EUFy3wdY/wO+C6EMBooBpYJk2KM1wPXAwwePLh8KCVJkiT9z+LFcPDB8P778OST8JOfZF2RJEn1V6M83tcbQO8QwtohhKbAgcDD5ZZ5CNgmhNA4hNCStBvc+3msUZIkSQXozDPhscfgr381SJIkaVXlrWdSjHFRCOEk4CmgCLg5xjguhHB8bv61Mcb3QwhPAu8AJcCNMcax+apRkiRJheemm+CSS+Ckk+CEE7KuRpKk+i+fu7kRY3wceLzctGvLXf8z8Od81iVJkqTC9OKLKUDacUe47LKsq5EkqTDkczc3SZJUiBYtgp//HO6/P+tKpKV8/HE6Uts668Ddd0PjvP6NKklS4TJMkiRJK2/2bNh1V7jySnjzzayrkf5n9ux05LaSEnj0UWjXLuuKJEkqHP4/I0mSVs4nn6Qg6aOP4Lrr4Ljjsq5IAlJnuQMOSC/Np5+G9dbLuiJJkgqLYZIkSVpxL70Ee+2Vun089RRst13WFUn/c/rp6WV5/fWw7bZ5vOMYYeJE+M9/4PvvoVMn6NhxyXmrVnksRpKk2mOYJEmSVsytt6ZeSGuvnfYf6t0764pUSObPT6FM8+YrdfPrroMrroBTT4Vjj63Z0pYSI3z6Kbz11pLTf/4D33xT+W1atFg6XKrqvH17KCqqxY3QSluwAGbNqv37ad06vW5CqP37kqQVYJgkSZKqp6QEfv1r+NOfYPvt4R//gNVXz7oqFZqnnoI994QePdL+ab17p/PSy+usk35cV+C55+Ckk2DnneHPNXls4BjTbp3lg6OZM9P8Jk1gww3TaN+DBqVTu3bw9dcwfXo6lV4uez5hQjr/9tuK7zeEFChVFDa1bVtxwFB+WnWWqWxajOl9X/ZU0bTKplc1rVWr9BmyvFMlz3WNKylJz+dXX1XvlI8gqVRREbRpA6uttvSpommVTS+d5ij0tSNG+O679F5e0dOiRel5KSpK59U9VXf5Ro2Wfu9V93xFb1PZqar51b1d6eNc0Wl586p72+qcr8qyZbdnZZ6Pqs5vuQV2333VXscryE8TSZJUtblz4ZBD4KGH4Pjj04DbTZpkXZUKUe/ecO65acCjCRPg3nthxoyllykNmsqETZOa9ubQQ9ahT5+W3HXXKvxmLilJh4ErHxzNnp3mN22agqP99lsSHG2wATRrtuy6qjtY0/z5aRvLhk0VBVAffQSvvpquL168khtYwxo1SkFUo0ZLn6qaFkL6XKksSCvVrFnVgVNlpxirHw5Nn17xYxoCdOgAXbpA586w0UbpvHPnFBg2qsXjGZWUpN0lKwogvvkm7VJZen3u3Oqts0WLpUOmzp2he/f0nio9lV5v2bL2tq0uKSmBOXNSQDhz5tLns2al935VgdCcOWk9VWnefNmQr1WrFCgtXpw+CxYtqvi0eHHl8xYtqt79r4zS9+3yzis7VTV/RW9XWs+KzqvubcufVzS/smUrOi97uarHcFXPe/RY9ed6BYVYNjmrpwYPHhzf9AgykiTVjs8+S/92vfsuXH556vqR510uQghvxRgH5/VOtVx5bX/NnJkCntKAacKEJZenT19q0UVdu9G4X+9lezWtt96yP45LStI6ygdHpQFH06ZQXJwCo403XhIcNW2an+2uTIwpZKho+vKur+i0qgKimvgcWLx4yY/3FT2VBnwrqnXrJYFQVacOHepHb57FiyvvGTNnTuUhyJdfwuTJy7yPgNQrrrKgqUcP6NYtfz3HqjJv3pLwp6JAaHnTZs+uOohp1WpJANS2bcU9wqo6tWlTu58dMVYcOC1evOx7uKrzsmGOGrzK2mD14JNRkiRl5vXXYY89UkP9scdg2LCsK1JDtPrqMHhwOpWzcPosfr7rx8x+awIXHPERvRblwqaHH049Tspac80UMK21FkyaBP/975LgqFmzFByNGLGkx9GAAXWzB14IhTOYd1FRCmw6dFjx2y5enIKAysImWNKjqPTUqVNh9rgpKloSWqyMH36Azz9PwdKUKem87OXXXlu2hyCk3S4rCppKL7drl74/Sk8//LD09YpOVS1Tdv7336dAaP785W9fixbpc6Rdu3RaYw1Yf/2lp1V2ebXV6sfYZSEs2bVNygNfaZIkqWKjRsERR6Qf4M89B/37Z12RtIzTft+Oa/49iJtuGkSvo8rN/PbbJT2ZyvZmevbZ9GP3kEOWBEf9+9fN4EiVKypKvWfat8+6kvqveXNYd910qsz33y8JnMqHThMnpqN8loZ4q6KoKIU/ZU/Nmy+53LbtksstWy4JfcqGQGXDoHbtKt4NVdIqMUySJElLizGNWXPeebD11nD//enffKmOueqqdDr9dDiqfJAEqUfBxhunk6RV07Jl6tm3vCN4zp2bQqbSoOnbbysPhSoLjAx1pXrBMEmSJC0xb17qjXTPPen82mv9R1d10jPPwM9/DrvuChddlHU1koA0HlW/fukkqaAZJkmSpGTq1DQ+0ptvwp/+lLp7OPim6qAPPkgHU1t/fbjzzvoxnIkkSYXEMEmSJKWBiHffPY138cADKVSS6qBvvoHddksHRXrkkXSAJEmSlF+Nsi5AkiRl7IEH0thIIcDLLxskqc5auBD23Rc++yy9bHv1yroiSZIaJsMkSZIaqhjTYDN77w0bbgj//jf86EdZVyVVKEY4+WR4/nm44QbYaqusK5IkqeFyNzdJkhqi+fPhuOPgb3+Dgw6Cm25KR9GR6qiRI+G66+Css+Cww7KuRpKkhs2eSZIkNTTTp8P226cg6fe/hzvuMEhSnfbkk3DqqbDnnnDBBVlXI0mS7JkkSVJDMm5cOpb6l1/C3XfD/vtnXZG0XO+/DwcckPbE/PvfoZF/hUqSlDm/jiVJaiieeAK22CLt4jZ6tEGS6rwZM1L22aIFPPwwtG6ddUWSJAkMkyRJKnwxwhVXpF/l662XBtreZJOsq5KWa8EC2Gcf+PxzePBB6Nkz64okSVIpwyRJkgpZSQmcdloacGaPPeCll6B796yrkpYrRjjxRHjxRbj5Zth886wrkiRJZRkmSZJUqObPT0dqu+KKFCbdey+0apV1VVKVLr88HWDwt7+Fgw/OuhpJklSeA3BLklSIZs9Oh7564QX4y1/gl7/MuiKpWp54Ak4/Pe3i9vvfZ12NJEmqiGGSJEmF5vPPYeed4YMP4PbbYcSIrCuSqm399VOHuuuu88htkiTVVYZJkiQVkvffh2HD4Jtv4PHH4Sc/yboiaYX06pUyUEmSVHcZJkmSVChefTUdsa1pUxg9GjbaKOuKJEmSVIDsPCxJUiF4+GHYfnvo2BH+9S+DJEmSJNUawyRJkuq766+HvfaCgQPhlVdg7bWzrkiSJEkFzDBJkqT6Kkb4v/+Dn/40jZP03HPQqVPWVUmSJKnAOWaSJEn10aJFcMIJcOONcNRR6dBXjf1alyRJUu2zZ5IkSfXN99+n3dpuvBF++9t0bpAkSZKkPLHlKUlSffL117DbbvDvf8M118Dxx2ddkSRJkhoYwyRJkuqLiRNhp53gs8/gvvtgzz2zrkiSJEkNkGGSJEn1wdtvw847w/z58M9/wlZbZV2RJEmSGijHTJIkqa579ln48Y+hSRN4+WWDJEmSJGXKMEmSpLps1KjUI2mtteDVV6F//6wrkiRJUgNnmCRJUl116aVw0EGwxRbw0kvQvXvWFUmSJEmGSZIk1TklJfDLX6bTvvvCU09Bu3ZZVyVJkiQBDsAtSVLdsmABHHEE3HUXnHwyXHYZFBVlXZUkSZL0P4ZJkiTVFd9+C3vvnQbcvugiOOMMCCHrqiRJkqSlGCZJklQXTJ0Ku+wCY8fC3/4Ghx6adUWSJElShQyTJEnK2vjxsNNOMH06PPpouixJkiTVUYZJkiRl6T//gWHD0uUXXoDBgzMtR5IkSaqKR3OTJCkro0fDtttCixbw8ssGSZIkSaoXDJMkScrCY4+l3dnWXBNeeQX69Mm6IkmSJKlaDJMkScq3O++EPfeEAQPgpZege/esK5IkSZKqzTBJkqR8uuoqOOQQ2HpreO456Ngx64okSZKkFWKYJElSPsQI558PJ50Eu+0GTzwBq62WdVWSJEnSCjNMkiSptpWUwC9+Ab/7HRx2GNx3HzRvnnVVkiRJ0koxTJIkqTYtWgRHHw2XXw6nnAK33AKNG2ddlSRJkrTSDJMkSaotP/wA++0Ht94Kv/99CpQa+dUrSZKk+s2/RiVJqg1z5sAee8Dzz8OVV8LJJ2ddkSRJklQjDJMkSappM2bAzjvDf/4Df/97OnqbJEmSVCAMkyRJqklTpsCOO8Knn8IDD6Qjt0mSJEkFxDBJkqSa8tFHsMMO8M038OSTMGRI1hVJkiRJNc4wSZKkmvD227DTTlBSAi+8ABtvnHVFkiRJUq3wkDKSJK2ql1+GoUOhWTN46SWDJEmSJBU0wyRJklbFE0+kMZK6dEmhUr9+WVckSZIk1SrDJEmSVtaoUbD77ilAeukl6Nkz64okSZKkWmeYJEnSyrj2Wjj4YNhyS3j+eejcOeuKJEmSpLwwTJIkaUXECBdeCCecAMOHp6O2tW2bdVWSJElS3hgmSZJUXTHCr34FZ58NI0bA/fdDixZZVyVJkiTlVV7DpBDCsBDChyGECSGEsyqYPzSEMDuE8HbudE4+65MkqVKLFsExx8All8BJJ8Hf/gZNmmRdlSRJkpR3jfN1RyGEIuAqYAdgCvBGCOHhGON75RZ9Kca4a77qkiSpSvPnp/GR7r8fzjkHzj0XQsi6KkmSJCkT+eyZtCkwIcb4SYxxATAK2COP9y9J0or77DPYfvsUJF1+Ofz+9wZJkiRJatDyGSZ1AyaXuT4lN628LUIIY0IIT4QQBuSnNEmSKvCPf8DAgTBmDNx1F/z851lXJEmSJGUun2FSRX/jxnLX/wOsFWMsBv4KPFjpykI4LoTwZgjhzenTp9dclZIkzZ0LRx8N++8PffvC22/DgQdmXZUkSZJUJ+QzTJoC9ChzvTvwRdkFYozfxhjn5i4/DjQJIXSsaGUxxutjjINjjIM7depUWzVLkhqat96CQYPgllvgN7+Bl1+GddfNuipJkiSpzshnmPQG0DuEsHYIoSlwIPBw2QVCCF1DSANRhBA2zdU3I481SpIaqpIS+MtfYIst4Lvv4Lnn4IILPGKbJEmSVE7ewqQY4yLgJOAp4H3gnhjjuBDC8SGE43OL7QuMDSGMAa4EDowxlt8VTpKkmjV1Kuy0E/zqV7DbbvDOOzB0aNZVSTUmhDAshPBhCGFCCOGsCua3DSE8khu3clwI4cgs6pQkSfVD43zeWW7XtcfLTbu2zOWRwMh81iRJauAefRSOPDL1Rrr+ejjmGI/WpoISQigCrgJ2IA078EYI4eEY43tlFvsZ8F6McbcQQifgwxDCHbkj8EqSJC0ln7u5SZJUd8ybByefnHoideuWxko69liDJBWiTYEJMcZPcuHQKGCPcstEoE1uuIHWwDfAovyWKUmS6gvDJElSwzN2LGy6KYwcCaedBq+/Duuvn3VVUm3pBkwuc31KblpZI4H1SQdHeRf4eYyxJD/lSZKk+sYwSZLUcMQIV10FgwfDV1/BE0/ApZdCs2ZZVybVpoq625Ufk3In4G1gTeBHwMgQwmrLrCiE40IIb4YQ3pw+fXpN1ylJkuoJwyRJUsPw9dewxx5w0kmw3XZpkO1hw7KuSsqHKUCPMte7k3oglXUkcH9MJgCfAv3KryjGeH2McXCMcXCnTp1qrWBJklS3GSZJkgrfP/8JAwfCU0/B5ZfDY49Bly5ZVyXlyxtA7xDC2iGEpsCBwMPllvkM2B4ghNAF6At8ktcqJUlSvWGYJEkqXAsWwBlnwA47QLt28O9/w89/7iDbalBijIuAk4CngPeBe2KM40IIx4cQjs8t9gdgyxDCu8CzwJkxxq+zqViSJNV1jbMuQJKkWjF+PBx8cDpK2/HHwyWXQMuWWVclZSLG+DjweLlp15a5/AWwY77rkiRJ9ZNhkiSpsMQIt9wCJ58MzZvDAw/AnntmXZUkSZJUMNzNTZJUOGbNggMPhKOPhk03hTFjDJIkSZKkGmaYJEkqDC+/DMXFcP/98Mc/pkG3u3fPuipJkiSp4BgmSZLqt5ISOO88GDIEmjSBV16Bs86CoqKsK5MkSZIKkmMmSZLqr0WL4Jhj4Lbb4JBD4OqroU2brKuSJEmSCpo9kyRJ9dO8ebD33ilIOvdc+NvfDJIkSZKkPLBnkiSp/pk9G3bbLY2TNHIk/OxnWVckSZIkNRiGSZKk+uXLL2HYMBg3Du68Mx29TZIkSVLeGCZJkuqPTz6BHXeEqVPh0Udhp52yrkiSJElqcAyTJEn1wzvvpPBowQJ49lnYfPOsK5IkSZIaJAfgliTVfS+/DD/+MRQVwUsvGSRJkiRJGTJMkiTVbY89BjvsAF26wCuvQP/+WVckSZIkNWiGSZKkuuvvf4c99oABA1LvpLXWyroiSZIkqcEzTJIk1U2XXw6HHQZDhsDzz0OnTllXJEmSJAnDJElSXRMjnH02nHYa7L132s2tTZusq5IkSZKU49HcJEl1x+LFcOKJcP31cOyxcM01adBtSZIkSXWGPZMkSXXD/Plw4IEpSPr1r+G66wySJEmSpDrInkmSpOzNmQN77QXPPguXXpp2cZMkSZJUJxkmSZKyNX067LIL/Pe/cNttadBtSZIkSXWWu7lJKmwLF8Izz0BJSdaVqCKffQbbbANjx8IDDxgkSZIkSfWAYZKkwnbllbDjjnD88QZKdc3778NWW8GXX8LTT8Nuu2VdkSRJkqRqMEySVLhihJtugtVWgxtugJ/9LE1T9l5/HbbeGhYtghdfTL2TJEmSJNULjpkkqXC9/nrq/XLjjTBhAlx0ETRqBCNHQghZV9dwPfNMGmy7S5d0eZ11sq5IkiRJ0gowTJJUuG6+GVq1gv33h9atUy+Yv/wFGjeGyy83UMrCPffAIYdA//7w5JPQtWvWFUmSJElaQYZJkgrTd9/BqFEpSGrTJk37059g8WK47DIoKoJLLjFQyqdrrkm7Gm69NTz8MLRrl3VFkiRJklaCYZKkwnTvvTBnDhx11JJpIaQAqaRkSaD0pz8ZKNW2GOH88+Gcc9Ig23ffDS1aZF2VJEmSpJVkmCSpMN18M/Tpk44WVlYIKUhavDjt8lZUBH/8o4FSbSoNkg47LI1f1aRJ1hVJkiRJWgWGSZIKz0cfwejRacDtikKiEODKK1OgdPHFKVA6/3wDpdpwzTUpSDr88BTwNfIgopIkSVJ9Z5gkqfDcemsKiA47rPJlQkhHdVu8GC68MC1/3nl5K7FBuOeeNEbSrrumHkkGSZIkSVJBMEySVFgWLUph0s47wxprLH/ZRo1Sz5nFi+EPf0iB0v/9X17KLHj//Gc6attWW6VQqbFfN5IkSVKhsHUvqbA8/TR88UXqdVQdjRrB9denQOncc1Og9Nvf1mqJBe+NN2DPPaFfP3jkEQfbliRJkgqMYZKkwnLzzdC5c9q1qroaNUq7YS1eDL/7XepFc9ZZtVdjIfvgg9QrrHNneOopaNcu64okSZIk1TDDJEmFY/p0ePhhOOWUFT9iWFER3HILlJTAr3+drv/qV7VTZ6GaMgV23DGFcU8/XfVuhpIkSZLqJcMkSYXj9tth4UI48siVu31RURpvafFiOOOMdP0Xv6jREgvWjBkpSJo9G154AdZbL+uKJEmSJNUSwyRJhSFGuOkm2GwzGDBg5dfTuDH8/e+ph9Ivf5l2gTv11BorsyDNnQvDh8Mnn6Rd2zbaKOuKJEmSJNUiwyRJheHNN2HcuDSY9qpq3Dj1clq8GE47LfVQOvnkVV9vIVqwAPbdNw26fd99MGRI1hVJkiRJqmWNsi5AkmrETTelo4YdcEDNrK9JE7jrrnRUslNOgauvrpn1FpKSEjjiiNQb6YYb0mMlSZIkqeAZJkmq/77/PgU/++0Hq61Wc+tt0gTuvht23x1+9rOa6fVUKGKEn/88Pe4XXQRHHZV1RZIkSZLyxDBJUv13//3w7be1E2g0bQr33JPGBPrpT+HGG2v+Puqj88+HkSPTuFJnnJF1NZIkSZLyyDBJUv13883p6GE//nHtrL9ZszQe0M47w3HHwS231M791BfXXAPnnAOHHw5/+hOEkHVFkiRJkvLIMElS/fbxx/D883DkkbUbajRrlnpA7bgjHH003HZb7d1XXXbPPWmXv113TeMkNfJrRJIkSWpo/BUgqX679dYUaBx2WO3fV/Pm8MADsP32Kby6/fbav8+65J//hEMOga22SqFSkyZZVyRJkiQpA4ZJkuqvxYtTmLTTTtC9e37us0ULeOgh2HbbtJvXnXfm536z9sYb6Wht/frBI4+kx0GSJElSg2SYJKn+euYZmDIl7XaWTy1bwsMPpzGaDj00HfGtkH3wQRovqnNneOopaNcu64okSZIkZcgwSVL9dfPN0LEj7LZb/u+7VSt49NG0y9eIEfDrX8PUqfmvo7ZNmZLGiWrcGJ5+GtZYI+uKJEmSJGXMMElS/fT11/Dgg2kMn6ZNs6mhVSt4/HHYbz+4+GLo1Sv1knrvvWzqqWkzZqQgafZseOKJdMQ8SZIkSQ2eYZKk+unOO2Hhwvzv4lZe69Zw110wfjwce2y6PGAADB8OL7wAMWZb38qaOzdtwyefpF36Ntoo64okSZIk1RGGSZLqnxjhpptgk01ggw2yriZZbz0YORI++wzOOy8NWL3ttqnGUaNg0aKsK6y+BQtg333TNowaBUOGZF2RJEmSpDrEMElS/fOf/8A778BRR2VdybI6doTf/Q4mTYLrroM5c+Cgg1LYdMUVqcdPXVZSAkcckQbavv76dAQ3SZIkSSrDMElS/XPzzdC8ORx4YNaVVK5FCzjuOHj/fXjoIejRA049NZ3/5jd1c7DuGOHnP0+76l10Ufa7EEqSJEmqkwyTJNUv8+al8ZL23bd+HKK+USPYfXd46SX4179g++1TULPWWqln1bhxWVe4xPnnp131fvlLOOOMrKuRJEmSVEcZJkmqXx54AGbNqpu7uFVl883h3nvho49Sr6VRo9KYT8OHw/PPZztY9zXXwDnnwGGHwZ/+BCFkV4skSZKkOq1x1gVI0gq5+WZYe+36PSj0uuumHkC//z1cfTX89a+w3XYwaBCcfnrqddW4lj6e58xJ4zmVPX36aQq5dt0Vbrwx9aaSJEmSpEoYJkmqPyZOhGefTUdLK4TAo0OHNFj36afD7bfDJZekwbrPOiuNr3T00dCmTfXXFyN8/fWyYVHZ08yZS9+mSRPo2RMOOQSuvTZdlyRJkqTlMEySVH/cckva/eqII7KupGa1aAHHHpvCo0cfhb/8BU47LfVcOv54OPlkWHNNWLw4DdxdPiCaODGdf/YZfP/90utu3TqNz7TWWrDFFksul566di2MYE6SJElS3hgmSaofFi9OYdKOO6YjohWi0sG6d98dXn89hUp/+lPqsdS9O0yeDIsWLX2bDh1SKLT++jBsWLrcq9eSsGj11R3/SJIkSVKNMkySVD8891wKU/7yl6wryY/NNoN//AM+/jiNrzRtGhxwwLI9i1q1yrpSSZIkSQ1MXsOkEMIw4AqgCLgxxnhRJcttArwGHBBjvDePJUqqq26+Gdq3hz32yLqS/Fp3XbjssqyrkCRJkqT/ydtAGSGEIuAqYGegP3BQCKF/JctdDDyVr9ok1XHffAMPPJAGiW7WLOtqJEmSJKlBy+eoq5sCE2KMn8QYFwCjgIq6GJwM3Ad8lcfaJNVld94J8+fDUUdlXYkkSZIkNXj5DJO6AZPLXJ+Sm/Y/IYRuwF7AtXmsS1Jdd/PNsPHGUFycdSWSJEmS1ODlM0yq6HBCsdz1y4EzY4yLq1xZCMeFEN4MIbw5ffr0mqhPUl303/+mk72SJEmSJKlOyOcA3FOAssfz7g58UW6ZwcCokA5j3RHYJYSwKMb4YPmVxRivB64HGDx4cPlQSlKhuPnmNE7SwQdnXYkkSZIkifyGSW8AvUMIawOfAwcCS/06jDGuXXo5hHAr8GhFQZKkBuKHH+COO2DvvWH11bOuRpIkSZJEHsOkGOOiEMJJpKO0FQE3xxjHhRCOz813nCRJS3voIZg5013cJEmSJKkOyWfPJGKMjwOPl5tWYYgUYzwiHzVJqsNuvhnWWgu22y7rSiRJkiRJOfkcgFuSqm/SJHjmGTjySGjkR5UkSZIk1RX+QpNUN912Wzo/4ohMy5AkSZIkLc0wSVLdU1ICt9wC22+fdnOTJEmSJNUZhkmS6p7nn4eJE+Hoo7OuRJIkSZJUjmGSpLrn5puhXTvYc8+sK5EkSZIklWOYJKlumTkT7rsPRoyA5s2zrkaSJEmSVI5hkqS6ZdQomD8fjjoq60okSZIkSRUwTJJUt9x0E/zoR7DxxllXIkmSJEmqgGGSpLpjzBh46y17JUmSJElSHWaYJKnuuOUWaNoUDj4460okSZIkSZUwTJJUN8yfD7ffno7g1qFD1tVIkiRJkiphmCSpbnj4YZgxA44+OutKJEmSJEnLYZgkqW64+Wbo0QO23z7rSiRJkiRJy2GYJCl7kyfDU0/BEUdAUVHW1UiSJEmSlsMwSVL2brsNYkxhkiRJkiSpTjNMkpStkpJ0FLfttoN11sm6GkmSJElSFQyTJGVr9Gj45BM46qisK5EkSZIkVYNhkqRsXXMNtG0Le++ddSWSJEmSpGowTJKUnfHj4R//gOOPhxYtsq5GkiRJklQNhkmSsnPxxdCsGZx2WtaVSJIkSZKqyTBJUjY++wz+9jc45hjo0iXraiSpoIUQhoUQPgwhTAghnFXJMkNDCG+HEMaFEF7Md42SJKn+aJx1AZIaqL/8JZ3/6lfZ1iFJBS6EUARcBewATAHeCCE8HGN8r8wy7YCrgWExxs9CCJ0zKVaSJNUL9kySlH9ffQU33ACHHgo9e2ZdjSQVuk2BCTHGT2KMC4BRwB7lljkYuD/G+BlAjPGrPNcoSZLqEcMkSfl32WUwfz6cVeGeFpKkmtUNmFzm+pTctLL6AKuHEF4IIbwVQjgsb9VJkqR6x93cJOXXrFlw1VWw337Qp0/W1UhSQxAqmBbLXW8MDAK2B1oA/wohvBZjHL/UikI4DjgOoKc9SyVJarDsmSQpv0aOhDlz4De/yboSSWoopgA9ylzvDnxRwTJPxhi/izF+DYwGisuvKMZ4fYxxcIxxcKdOnWqtYEmSVLcZJknKn+++g8svh+HDoXiZ3yiSpNrxBtA7hLB2CKEpcCDwcLllHgK2CSE0DiG0BDYD3s9znZIkqZ5wNzdJ+XP99TBjBpx9dtaVSFKDEWNcFEI4CXgKKAJujjGOCyEcn5t/bYzx/RDCk8A7QAlwY4xxbHZVS5KkuswwSVJ+zJ8Pf/kLDB0KW2yRdTWS1KDEGB8HHi837dpy1/8M/DmfdUmSpPrJMElSftx2G3zxRTqXJEmSJNVbjpkkqfYtWgQXXwybbALbb591NZIkSZKkVWDPJEm17+674ZNP4NJLIVR0hGpJkiRJUn1hzyRJtaukBP74RxgwAHbbLetqJEmSJEmryJ5JkmrXww/DuHFwxx3QyPxakiRJkuo7f9lJqj0xwoUXwjrrwP77Z12NJEmSJKkG2DNJUu355z/hjTfg+uuhsR83kiRJklQI7JkkqfZceCF06waHHZZ1JZIkSZKkGmJXAUm149VX4YUX4LLLoFmzrKuRJEmSJNUQeyZJqh0XXggdO8Kxx2ZdiSRJkiSpBhkmSap5b78Njz0Gp54KrVplXY0kSZIkqQYZJkmqeX/8I6y2GvzsZ1lXIkmSJEmqYYZJkmrWhx/CP/6RgqR27bKuRpIkSZJUwwyTJNWsiy+G5s3TLm6SJEmSpIJjmCSp5kyaBH//exp0u3PnrKuRJEmSJNUCwyRJNecvf4EQ4PTTs65EkiRJklRLDJMk1Yxp0+DGG+Gww6BHj6yrkSRJkiTVEsMkSTXjsstgwQI488ysK5EkSZIk1SLDJKm2zZ+fdQW1b+ZMuPpq2H9/6N0762okSZIkSbXIMEmqTTfcAB06wPPPZ11J7Ro5EubMgV//OutKJEmSJEm1zDBJqi0lJXDxxfDdd7D77vDGG1lXVDvmzoXLL4fddoOBA7OuRpIkSZJUywyTpNryxBPw8cdw6aXQuTMMGwbvvZd1VTXv+uvhm2/gN7/JuhJJkiRJUh4YJkm15a9/hTXXhJNOgmeegWbNYIcd4NNPs66s5vzwA/zlL7DddrD55llXI0mSJEnKA8MkqTZ8+CE89RQcfzw0aQLrrANPPw3z5qVA6csvs66wZtx2G0ydaq8kSZIkSWpADJOk2jByJDRtCscdt2TaBhukXd++/BJ23DEdAa0+W7QojQm12WapZ5IkSZIkqUEwTJJq2rffwq23wgEHQJcuS8/bbDN46KHUc2n48DQ4d301alTaZe83v4EQsq5GkiRJkpQnhklSTbv11nSEs5NPrnj+9tunIOb112GvvWD+/LyWVyNKSuCPf4QNN4Rdd826GkmSJElSHhkmSTWppCTt4rb55rDJJpUvt9decNNNaWDuESPSLmP1yUMPpSPT/frX0MiPEUmSJElqSPwVKNWkp56Cjz6qvFdSWUccAZddBvfdBz/9KcRY6+XViBjhggtgvfVg//2zrkaSJEmSlGeNsy5AKih//St07Qr77lu95U89NQ3Efd55sPrq8Oc/1/3xh555Bt56C264AYqKsq5GkiRJkpRnhklSTRk/Ph2t7dxz05Hcquvcc1OgdMklKVA6++zaqrBmXHABdO8Ohx2WdSWSJEmSpAwYJkk15aqroEmTtMvaiggBLr8cZs2C3/4W2rWDn/2sFgqsAS+/DKNHp3pXJDCTJEmSJBUMwySpJsyZA7fcksYQ6tp1xW/fqFEakHv2bDjppBQojRhR42WusgsvhI4d4dhjs65EkiRJkpQRB+CWasJtt6VAqToDb1emSRO4+27Ydls4/HB49NGaq68m/Pe/aTe+006Dli2zrkaSJEmSlBHDJGlVlZTAyJGwySaw2Wartq7mzeGhh2DjjWG//eDFF2umxppw4YWw2mp1dxc8SZIkSVJe5DVMCiEMCyF8GEKYEEI4q4L5e4QQ3gkhvB1CeDOEsHU+65NWyj//CR9+CKecUjPra9MGHn8c1lkHdtsN3nyzZta7Kj74AO67L+2C17Zt1tVIkiRJkjKUtzAphFAEXAXsDPQHDgoh9C+32LNAcYzxR8BRwI35qk9aaVdeCZ07p55ENaVjR3j6aejQAYYNg/ffr7l1r4yLLkq9pk49Nds6JEmSJEmZy2fPpE2BCTHGT2KMC4BRwB5lF4gxzo0xxtzVVkBEqssmTEi9iI4/Hpo1q9l1d+sGzzwDjRvDjjvCpEk1u/7qmjgRbr8djjsOOnXKpgZJkiRJUp2RzzCpGzC5zPUpuWlLCSHsFUL4AHiM1DtJqruuugqKiuCnP62d9a+3XuqhNHcu/OQnMG1a7dxPZSZPhl//Oh1t7vTT83vfkiRJkqQ6KZ9hUqhg2jI9j2KMD8QY+wF7An+odGUhHJcbV+nN6dOn11yVUnXNnQs33wz77gtrrll79zNwYOr99MUXsNNOMGtW7d3XjBlw771wwgnQpw/07AmjRqVBt7t3r737lSRJkiTVG43zeF9TgB5lrncHvqhs4Rjj6BDCuiGEjjHGryuYfz1wPcDgwYPdHU759/e/w7ff1tzA28uzxRbw4IMwfHg6Pf00tGq16uv97jt46SV49tl0evttiBFat4YhQ1KotP32sOGGq35fkiRJkqSCkM8w6Q2gdwhhbeBz4EDg4LILhBDWAz6OMcYQwsZAU2BGHmuUqidG+OtfYdAg2Hzz/NznDjvAXXfB/vun3lAPPQRNm67YOhYuhNdfXxIevfZamta0aQqsfv/7FB5tsgk0aVI72yFJkiRJqtfyFibFGBeFEE4CngKKgJtjjONCCMfn5l8L7AMcFkJYCMwDDigzILdUdzz7bDrC2m23QahoD85ass8+cMMNcPTRcMghKVwqKqp8+ZISeOedJeHR6NGpN1IIsPHGcNppKTzaemto2TJ/2yFJkiRJqrfy2TOJGOPjwOPlpl1b5vLFwMX5rElaKX/9azqy2QEH5P++jzoKZs5MA2K3bQvXX78k0IoRPv54SXj0/PPwdW4v0b594fDDU3g0dCi0b5//2iVJkiRJ9V5ewySpIHzyCTzyCPzmN9CsWTY1/PKXKVC64AJo0wYGD14SIE2alJbp1g122SWFR9tt5wDakiRJkqQaYZgkrairr4ZGjdLg1Fn6wx9SoHTZZen66qvDttvCGWekAKlPn/zugidJkiRJahAMk6QV8d13cNNNaeyibt2yrSWEtLvdLrtA167wox8tf/wkSZIkSZJqgGGStCJuvx1mzYKTT866kqRRIxg+POsqJEmSJEkNSKOsC5DqjRhTT6CNNoKttsq6GkmSJEmSMmGYVInp02GzzdKR1yUgHRlt3LjUK8mxiCRJkiRJDZRhUiXat4d334V//zvrSlRn/PWv0LEjHHRQ1pVIkiRJkpQZw6RKFBXBhhvCmDFZV6I6YeJEePhhOPZYaN4862okSZIkScqMYdJyFBfDO++koXLUwF19ddq17YQTsq5EkiRJkqRMGSYtR3ExzJgBX3yRdSXK1Pffw403wl57QY8eWVcjSZIkSVKmDJOWo7g4nburWwN3xx0wc2YaeFuSJEmSpAbOMGk5NtwwnRsmNWAxpoG3i4thm22yrkaSJEmSpMw1zrqAuqxtW+jVyzCpQRs9Oh3W78Yb05hJkiRJkiQ1cPZMqkLpINxqoK68Etq3h4MPzroSSZIkSZLqBMOkKhQXw4cfwrx5WVeivPvsM3jwQTj2WGjRIutqJEmSJEmqEwyTqlBcDCUlMG5c1pUo766+Op2fcEK2dUiSJEmSVIcYJlVh4MB07rhJDcy8eXDDDbDnnrDWWllXI0mSJElSnWGYVIV11oHWrQ2TGpy77oJvvoGTT866EkmSJEmS6hTDpCo0agQbbmiY1KDEmAbe3mADGDIk62okSZIkSapTDJOqofSIbjFmXYny4uWXU3p4yikQQtbVSJIkSZJUpxgmVUNxMcyaBZMnZ12J8uLKK2H11WHEiKwrkSRJkiSpzjFMqobi4nTurm4NwOTJ8MADcMwx0LJl1tVIkiRJklTnGCZVwwYbpHPDpAbg2mvT/ownnph1JZIkSZIk1UmGSdXQpg2su65hUsH74Qe4/nrYbTfo1SvraiRJkiRJqpMMk6qpuNgwqeCNGgVff50G3pYkSZIkSRUyTKqm4mKYMAG++y7rSlQrYkwDbw8YANtum3U1kiRJkiTVWYZJ1VRcnPKGsWOzrkS14tVX4b//hZNPhhCyrkaSJEmSpDrLMKmaPKJbgfvrX6FdOzjkkKwrkSRJkiSpTjNMqqa11oLVVjNMKkiffw733gtHHQWtWmVdjSRJkiRJdZphUjWFAAMHGiYVpGuvhZIS+NnPsq5EkiRJkqQ6zzBpBRQXwzvvpNxBBWL+fLjuOth1V1hnnayrkSRJkiSpzjNMWgHFxTBnDkyalHUlqjF33w3Tp8Mpp2RdiSRJkiRJ9YJh0gpwEO4CM3ky/PrX0L8/bL991tVIkiRJklQvGCatgAED0thJhkkFYNYs2GUXmDsX7rorPbGSJEmSJKlKjbMuoD5p1Qp69zZMqvcWLIC994YPPoAnn0wjq0uSJEmSpGqxZ9IKKi42TKrXYoSjj4bnn4ebbnL3NkmSJEmSVpBh0goqLoZPPoFvv826Eq2U3/0Obr8d/vAHOOywrKuRJEmSJKneMUxaQaWDcI8dm20dWgnXXw8XXADHHANnn511NZIkSZIk1UuGSSvII7rVU48/DieeCMOGwdVXO+C2JEmSJEkryTBpBXXvDu3aGSbVK2+9BfvvnwbavuceaNIk64okSZIkSaq3DJNWUAgOwl2vTJwIw4dDhw7w2GPQpk3WFUmSJEmSVK8ZJq2E4mJ4910oKcm6Ei3XzJmw884wfz488QSssUbWFUmSlIkQwrAQwochhAkhhLOWs9wmIYTFIYR981mfJEmqXwyTVkJxMXz3HXz8cdaVqFLz58Oee6ZD7z34IPTvn3VFkiRlIoRQBFwF7Az0Bw4KISzzxZhb7mLgqfxWKEmS6hvDpJVQOgj3O+9kW4cqUVIChx8Oo0fDrbfCkCFZVyRJUpY2BSbEGD+JMS4ARgF7VLDcycB9wFf5LE6SJNU/hkkrYcAAaNTIcZPqrF//Gu6+Gy66CA46KOtqJEnKWjdgcpnrU3LT/ieE0A3YC7g2j3VJkqR6yjBpJTRvDn37GibVSVddBX/6E5xwApxxRtbVSJJUF4QKpsVy1y8HzowxLl7uikI4LoTwZgjhzenTp9dUfZIkqZ5pnHUB9VVxMfzrX1lXoaU8/DCccgrsthtceWU69J4kSZoC9ChzvTvwRbllBgOjQvru7AjsEkJYFGN8sOxCMcbrgesBBg8eXD6QkiRJDYQ9k1ZScTFMmgSzZmVdiQD497/hwANh0CC46y5obE4qSVLOG0DvEMLaIYSmwIHAw2UXiDGuHWPsFWPsBdwLnFg+SJIkSSplmLSSSgfhfvfdbOsQ6bB6u+4KXbvCI49Aq1ZZVyRJUp0RY1wEnEQ6Stv7wD0xxnEhhONDCMdnW50kSaqP7L6xkkrDpDFjYJttsq2lQfv6a9h5Z1i8GJ54Arp0yboiSZLqnBjj48Dj5aZVONh2jPGIfNQkSZLqL8OklbTGGtChg4NwZ2rePNh9d/jsM3j22TQquiRJkiRJqlWGSSsphNQ7yTApI4sXwyGHwGuvwT33wFZbZV2RJEmSJEkNgmMmrYLiYhg7NuUayrPTT4f774dLLoF99826GkmSJEmSGgzDpFVQXJz2tProo6wraWAuvzydfv5zOO20rKuRJEmSJKlBMUxaBaWDcL/zTrZ1NCj33Qe/+AXsvXfqlSRJkiRJkvLKMGkVrL8+NG7suEl58+qraZykzTeH22+HoqKsK5IkSZIkqcExTFoFzZpBv36GSXkxfnw6cluPHvDww9CiRdYVSZIkSZLUIBkmrSKP6JYHX30FO+8MjRrBE09Ax45ZVyRJkiRJUoNlmLSKiothyhT45pusKylQ330Hu+0GU6fCI4/AuutmXZEkSZIkSQ2aYdIqKh2E295JteDtt2HPPeHNN+Guu2CzzbKuSJIkSZKkBs8waRV5RLcatmABjBoFW28NG20Er7wC114Le+yRdWWSJEmSJAlonHUB9V2XLulkz6RV9MUXcP31cN118OWXaXe2Sy+FI46A1VfPujpJkiRJkpRjmFQDBg40TFopMaaeRyNHwn33waJFsMsucNJJsNNOacBtSZIkSZJUp+T113oIYVgI4cMQwoQQwlkVzB8RQngnd3o1hFCcz/pWVnExjBuXshBVw/ffw403pt3YttkGnnoKTjkFPvoIHntsyZHbJEmSJElSnZO3nkkhhCLgKmAHYArwRgjh4Rjje2UW+xQYEmOcGULYGbgeqPOjLhcXw/z58OGHMGDACtzwo4+gSRPo1au2SqtbPv4YrrkGbroJZs2CDTdMu7YdfDC0apV1dZIkSZIkqRryuZvbpsCEGOMnACGEUcAewP/CpBjjq2WWfw3onsf6VlrZI7pVO0waNw623DKFSa++Cn361Fp9mSopgaefTruyPf546nG0zz5pV7att4YQsq5QkiRJkiStgHzuS9QNmFzm+pTctMocDTxRqxXVkH79UiZU7SO6TZ8Ou+0GLVumMGXYMJg2rVZrzLtZs+Dyy6Fv37Tb2ptvwu9+B5Mmwd13p93bDJIkSZIkSap38tkzqaLkIFa4YAjbksKkrStdWQjHAccB9OzZsybqW2lNmkD//tUchHv+fNhrL5g6FUaPToNQb7stDB8OL7wArVvXdrm169134aqr4O9/T2MjbbklnHde6o3UtGnW1UmSJEmSpFWUz55JU4AeZa53B74ov1AIYSBwI7BHjHFGZSuLMV4fYxwcYxzcqVOnGi92RRUXVyNMihGOPTYdwexvf4NNNoFNN4V77oG334b99oOFC/NRbs1auBDuvReGDk2HtrvtNjjwQHjrrbStBx1kkCRJkiRJUoHIZ5j0BtA7hLB2CKEpcCDwcNkFQgg9gfuBQ2OM4/NY2yorLk6djaZPX85CF12Ueuz84Q8pOCo1fDhcey08+SQcd1wKneqD+fPhggtg7bXT9kyaBH/6E0yZkgbZ3njjrCuUJEmSJEk1LG+7ucUYF4UQTgKeAoqAm2OM40IIx+fmXwucA3QArg5pPJ1FMcbB+apxVZQdhPsnP6lggfvug9/8BkaMgLPPXnb+McekEOb3v4fu3VPgVJfNmQN77gnPPQc77piO0rbLLlBUlHVlkiRJkiSpFuVzzCRijI8Dj5ebdm2Zy8cAx+Szppqy3DDpzTfh0ENhiy3gxhsrH3j6//4vBUrnn58CpZ/+tFZrXmnTp6fg6L//TbvrHXpo1hVJkiRJkqQ8yWuYVMg6doQ116zgiG6ffw577AFdusCDD0Lz5pWvJIS0u9vUqXDiibDGGrD77rVZ9or77LPUE2nSpLQ9u+6adUWSJEmSJCmP8jlmUsEbOLDcINzffQe77ZZ2CXvkEejcueqVNG6cBuQeNCgNYv3aa7VW7wp7/33Yaiv48kt4+mmDJEmSJEmSGiDDpBpUXAzvvQcLFgAlJXDIISldGjUKNtig+itq1QoefTR1ddp1VxhfB8Yi//e/YZtt0pHbXnwxXZYkSZIkSQ2OYVINKi5OWcsHH5AG2X7wQbj00jS+0Irq3Dkd3a1RIxg2LPUGyso//wnbbQerrQavvLJkgChJkiRJktTgGCbVoNKMZc7IW+Gii+D44+GUU1Z+heutl3ooTZsGw4en3eXy7d57032vs04KktZdN/81SJIkSZKkOsMwqQb16QPbNxnNZjcdlw7pduWVlR+5rbo23TSNoTRmDOy3X+r6lC/XXw/77w+DB6dd29ZYI3/3LUmSJEmS6iTDpBrUeNLH3FuyN1ObrwP/+Ac0aVIzKx4+PB3l7amn4NhjIcaaWW9lYoQ//hF++tO0i90zz8Dqq9fufUqSJEmSpHqhcdYFFIxZs2DXXWncOLJv80d5vV27ml3/McfA55/DuedCjx7whz/U7PpLlZTA6afDZZfBiBFwyy01F4pJkiRJkqR6z55JNWHRorQ72Mcf8+Sx9/Pvb9arnfGyzzknhUrnn596KtW0hQvhyCNTkHTyyfC3vxkkSZIkSZKkpRgmraoY0yDbzzwD111Hx32GAGmIoxoXAlxzTTo63M9+Bg89VHPrnjcP9tknBUjnnQdXXJGOJCdJkiRJklSGacGqGjkyBTxnnAFHHvm/I7rVSpgE0LhxGpB70CA46CD4179WfZ2zZ6exkR59FK6+Gn73u1UfOFySJEmSJBUkw6RV8cQTcOqpsMceacBq0jjVPXrAO+/U4v22apWCnzXXhN12gw8/XPl1TZsGQ4fCq6/CnXfCCSfUWJmSJEmSJKnwGCatrLFj4YADYOBAuP32pXYJGziwFnsmlercGZ58Mt3vsGGs1CBNn34KW28N48fDI4/AgQfWfJ2SJEmSJKmgGCatjK++Sj2CWrdOIUzr1kvNLi6GDz6A+fNruY711oPHHkv1DB8Oc+ZU/7Zjx8JWW8GMGfDPf6ZASpIkSZIkqQqGSSvqhx9gr71ST6CHHoLu3ZdZpLg4HeDtvffyUM8mm6QxlMaMgX33TUdkq8qrr8I226RxkV56CbbYovbrlCRJkiRJBcEwaUXECMcem8KYv/0tBTkVqPVBuMsbPhyuuw6efhqOOSbVWZknn4Sf/AQ6doRXXoEBA/JUpCRJkiRJKgSNsy6gXrnwwjQ+0vnnw377VbrYeutBixZ5DJMAjj4apkyBc89NI4Cff/6yy9x1Fxx2GGy4YRo8vEuXPBYoSZIkSZIKgWFSdd17L/z2t3DIIfCb3yx30aKilNfU6hHdKnLOOSlQuuCCtPvd8ccvmXfVVXDyyfDjH6fd89q2zXNxkiRJkiSpELibW3W88Ubq0bPllnDDDWmsoSqUHtFteXuc1bgQ4Jpr0m5vP/tZCo1ihN//Hk46KQ0a/uSTBkmSJEmSJGmlGSZVZcoU2GOPtEvYAw9A8+bVullxcTpQ2hdf1HJ95TVuDHffDYMHw4EHptO558Lhh8N991W7fkmSJEmSpIoYJi3P3LmpN8/cufDII9C5c7VvmvdBuMtq1SrV2717OtLbL38JN9+cgiZJkiRJkqRVYLpQmZKSND7SO+/Ao4/CBhus0M0HDkznY8bALrvUQn1V6dwZXnwR/vOftNtbNXbNkyRJkiRJqophUmUmTYJ//Qsuuwx23nmFb962LfTqlVHPpFJrrplOkiRJkiRJNcQwqTJrrw3vvQft26/0KoqLMziimyRJkiRJUi1yzKTl6dBhlXYPGzgQPvwQ5s2rwZokSZIkSZIyZJhUi4qL09BL48ZlXYkkSZIkSVLNMEyqRZke0U2SJEmSJKkWGCbVonXWgdatDZMkSZIkSVLhMEyqRY0awYYbGiZJkiRJkqTCYZhUywYOTEd0izHrSiRJkiRJkladYVItKy6GWbNg8uSsK5EkSZIkSVp1hkm1zEG4JUmSJElSITFMqmUbbpjODZMkSZIkSVIhMEyqZW3awLrrGiZJkiRJkqTCYJiUB8XFhkmSJEmSJKkwGCblwcCBMGECfPdd1pVIkiRJkiStGsOkPCguhhhh7NisK5EkSZIkSVo1hkl54BHdJEmSJElSoTBMyoNevWC11QyTJEmSJElS/WeYlAchpHGTDJMkSZIkSVJ9Z5iUJ8XF8M47UFKSdSWSJEmSJEkrzzApTwYOhDlzYNKkrCuRJEmSJElaeYZJeeIg3JIkSZIkqRAYJuXJBhuksZMMkyRJkiRJUn1mmJQnrVpB796GSZIkSZIkqX4zTMqj4mLDJEmSJEmSVL8ZJuXRwIHwySdpIG5JkiRJkqT6yDApj0oH4X733WzrkCRJkiRJWlmGSXnkEd0kSZIkSVJ9Z5iURz16QLt2hkmSJEmSJKn+MkzKoxAchFuSJEmSJNVvhkl5VlycxkwqKcm6EkmSJEmSpBVnmJRnAwfCd9+lo7pJkiRJkiTVN4ZJeeYg3JIkSZIkqT4zTMqzAQOgUSPDJEmSJEmSVD8ZJuVZixbQt69hkiRJkiRJqp8MkzLgEd0kSZIkSVJ9ZZiUgeJimDQJZs3KuhJJkiRJkqQVY5iUgYED0/m772ZbhyRJkiRJ0ooyTMqAR3STJEmSJEn1lWFSBtZcEzp0MEySJEmSJEn1j2FSBkJwEG5JkiRJklQ/GSZlpLgYxo6FxYuzrkSSJEmSJKn6DJMyMnAgzJsHH32UdSWSJEmSJEnVl9cwKYQwLITwYQhhQgjhrArm9wsh/CuEMD+EcHo+a8u30kG433kn2zokSVLhq0YbbEQI4Z3c6dUQQnEWdUqSpPohb2FSCKEIuArYGegPHBRC6F9usW+AU4C/5KuurPTvD40bO26SJEmqXdVsg30KDIkxDgT+AFyf3yolSVJ9ks+eSZsCE2KMn8QYFwCjgD3KLhBj/CrG+AawMI91ZaJZsxQoPfggzJmTdTWSJKmAVacN9mqMcWbu6mtA9zzXKEmS6pF8hkndgMllrk/JTWuw/vhH+PBD2HdfWLAg62okSVKBWtE22NHAE7VakSRJqtfyGSaFCqbFlV5ZCMeFEN4MIbw5ffr0VSgrO7vsAjfcAE8/DUcdBSUlWVckSZIKULXbYCGEbUlh0pmVzK/37S9JkrTq8hkmTQF6lLneHfhiZVcWY7w+xjg4xji4U6dOq1xcVo48Ei64AO64A85aZjhMSZKkVVatNlgIYSBwI7BHjHFGRSsqlPaXJElaNY3zeF9vAL1DCGsDnwMHAgfn8f7rrF//Gr74Av78Z1hjDTjttKwrkiRJBaTKNlgIoSdwP3BojHF8/kuUJEn1Sd7CpBjjohDCScBTQBFwc4xxXAjh+Nz8a0MIXYE3gdWAkhDCqUD/GOO3+aozCyHAFVfAl1/CL36RAqUDD8y6KkmSVAiq0wYDzgE6AFeHEAAWxRgHZ1WzJEmq2/LZM4kY4+PA4+WmXVvm8pc00KOHFBXB7bfD9Olw2GHQqRNsv33WVUmSpEJQjTbYMcAx+a5LkiTVT/kcM0lVaN4cHnoI+vaFvfaC//4364okSZIkSZKWZphUx7RrB08+CauvDjvvDJ98knVFkiRJkiRJSxgm1UHduqVAaeFC2Gkn+OqrrCuSJEmSJElKDJPqqPXXh0cfhc8/h113hblzs65IkiRJkiTJMKlO22ILuPtueOst2G+/1FNJkiRJkiQpS4ZJddxuu8F116Xd3o45BmLMuiJJkiRJktSQNc66AFXtmGNg6lQ45xxYYw246KKsK5IkSZIkSQ2VYVI98dvfwhdfwMUXp0Dp5z/PuiJJkiRJktQQGSbVEyHAyJHpyG6nnQZdu8IBB2RdlSRJkiRJamgcM6keKSqCO+6ArbeGww6D557LuiJJkiRJktTQGCbVM82bw0MPQe/esOee8PbbWVckSZIkSZIaEsOkemj11dPR3dq2hZ13hokTs65IkiRJkiQ1FIZJ9VT37vDUUzB/Puy0E0yfnnVFkiRJkiSpITBMqsf694dHHoHPPoNdd4Xvvsu6IkmSJEmSVOgMk+q5rbaCUaPgzTdh//1h4cKsK5IkSZIkSYXMMKkA7LEHXHMNPP44HHccxJh1RZIkSZIkqVA1zroA1YzjjoOpU+Hcc2HNNeGCC7KuSJIkSZIkFSLDpAJyzjnwxRdw4YWwxhpw0klZVyRJkiRJkgqNYVIBCQGuugqmTYNTToGuXWHffbOuSpIkSZIkFRLHTCowjRvDXXfBllvCiBHwwgtZVyRJkiRJkgqJYVIBatECHn4Y1lsPdt8dXn4564okSZIkSVKhMEwqUO3bw9NPp8G4d9oJnn0264okSZIkSVIhMEwqYN26wYsvwrrrwvDh8NhjWVckSZIkSZLqO8OkAtelCzz/PGy4Iey5J9x7b9YVSZIkSZKk+swwqQHo0AH++U/YbDM44AD4+9+zrkiSJEmSJNVXhkkNRNu28NRTsO22cPjhcN11WVckSZIkSZLqI8OkBqRVK3j0UdhlFzj+eLjssqwrkiRJkiRJ9Y1hUgPTvDncfz/suy/84hdwwQVZVyRJkiRJkuqTxlkXoPxr2hTuugtatIDf/ha++y6FSiFkXZkkSZIkSarrDJMaqMaN4dZboWVL+OMfU6B02WXQyL5qkiRJkiRpOQyTGrBGjeCaa1KgdNll8P33cO21UFSUdWWSJEmSJKmuMkxq4EKASy6B1q3hD39IgdJtt6WeS5IkSZIkSeUZGYgQ4LzzUg+lX/8a5s1LYyo1a5Z1ZZIkSZIkqa5xhBz9z1lnwRVXwAMPwJ57plBJkiRJkiSpLMMkLeWUU+DGG+Gpp2D4cJg7N+uKJEmSJElSXWKYpGUcfTTcfjuMHg077gizZmVdkSRJkiRJqisMk1Shgw+Gf/wD3nwTtt8evv4664okSZIkSVJdYJikSu21Fzz8MLz3HgwdClOnZl2RJEmSJEnKmmGSlmvYMHjiCZg4EYYMgcmTs65IkiRJkiRlyTBJVRo6FJ55Br76CrbZBj7+OOuKJEmSJElSVgyTVC1bbAHPPZeO7vbjH8P772ddkSRJkiRJyoJhkqpt443hxRehpCTt8jZmTNYVSZIkSZKkfDNM0goZMABGj4bmzdPub//+d9YVSZIkSZKkfDJM0grr3Rteegk6dIDttoMDDoDLLoPXXoP587OuTpIkSZIk1abGWReg+mmttVIPpTPPTMHSPfek6U2bwqBBsPnmaZylLbaA7t2zrVWSJEmSJNUcwySttDXXhL//PV3+4ovUM+lf/0rn11yTeisBdOu2JFjaYgvYaKO0m5wkSZIkSap/DJNUI9ZcE/beO50AFixIA3SXhkv/+hfce2+a17RpCpS22GJJD6YePSCE7OqXJEmSJEnVY5ikWtG0KWyySTqdckqa9uWXS4dL110Hl1+e5q255tLh0qBB9l6SJEmSJKkuMkxS3nTtCnvtlU4ACxem3kul4dK//gX33ZfmNWmSei9tvjmsvz707Jl6L/XsCW3bZrcNkiRJkiQ1dIZJykyTJjB4cDqddFKaNm3a0uHSDTfAvHlL365NmxQqlQ2Yyl7u3j31jJIkSZIkSTXPMEl1SpcusMce6QSweHHaPW7yZPjssyXnpZfffBOmT192PV27Vh429ewJnTpBo0b53TZJkiRJkgqBYZLqtKKidDS4bt3SLm8VmTcPpkxZOmQqvTxuHDzxBHz//dK3ado0hUs9eqRxnU44AdZeu/a3R5IkSZKk+s4wSfVeixbQu3c6VSRGmDmz4rBp0iS49FL4y19g993h5z+HoUM9spwkSZIkSZUxTFLBCwHat0+nH/1o2flTpsA116Sjyz30EAwcmI5Ad/DBKaiSJEmSJElLOGqMGrzu3eGCC1KPpRtvTD2Zjjkm7QJ39tnw+edZVyhJkiRJUt1hmCTltGgBRx8NY8bAc8/BNtvAH/8IvXrBQQelo8xJkiRJktTQGSZJ5YQA224LDzwAH3+cdnl74gnYYgvYbDO44w5YsCDrKiVJkiRJyoZhkrQc/8/efcdJVd3/H3+dXUC6SFdQsVBEDVEIdgF7iSVqjL3FIEn8Rk1Ro8avP2OMSUy+0aixR00s0Whswa5YYsfYsIINLEjvCMue3x9nNgzLNmB37s7s6/l43MfM3HvnzufOzO6cec+55260Efz+92lcpcsug9mz4aijUm+lX/4Svvwy6wolSZIkSSoswySpATp2hB/+EN5+G8aOhSFD4NxzYYMN4Pjj4dVXs65QkiRJkqTCMEySVkFZGey9dzrs7e230xhLd9wBW20FI0bAXXdBRUXWVUqSJEmS1HQMk6TVNGgQXH55OgTu4ovhk0/g4INh003hd7+DWbOyrlCSJEmSpMZnmCStoS5d4Cc/gYkTU8+kjTaC00+Hvn3h+9+Hl19OYystXZp1pZIkSZIkrblWWRcglYrycvjWt9L02mtw6aXwl7/AlVcuX6dDB+jaFdZZZ/ll/vXa5q29djrETpIkSZKkrBkmSU1gyBC47jr4zW/g0Udh5sw0zZqVpqrr77+//PaiRbVvL4QUKNUWOnXpkq7XdLn22tDKv3RJkiRJUiMp6FfMEMJewCVAOXBtjPGiastDbvk+wELguBjjK4WsUWpM3bvDYYc1bN2vvloxaKp+WX3eJ58sv13foN+dOtUdOFVd1jSvQ4cUZkmSJEmSBAUMk0II5cDlwO7AFOClEMK9Mca38lbbG+ifm7YB/py7lEreWmtB795pWhUxwsKFKVSaPbthlx9+uPz2vHl1b79VK2jfPtWXP7Vps/K8uubXt6xt2xWnmua1amWwJUmSJElZK2TPpOHAxBjjBwAhhNuAA4D8MOkA4KYYYwSeDyF0CSGsG2P8vIB1SkUlhNR7qEOHNOj3qqqogDlzag+eZs1Kh+B99VXN05IlKZCaPn3l+fm3Y1zzfS0rqztsqi2MWmutNKZV9amsbNXm17WsoiLt89KlDb9clXUhhXGtW6/aZUPXbd06hXWtWqX9yb+s7Xpdy8vLDf4kSZKkUlXIMKkPMDnv9hRW7nVU0zp9AMMkqYm0agXduqWpqcSYwpaaQqb8afHi2qe6lucvmz695uXLlq04VVY23f7WpSq4aWgY1K5dug7Lw6XFi2Hu3NrDp/zry5Zls5+Qgrf8gCk/XGrq603FgKxuRx0FF1+cdRWSJElqaoUMk2pqglfvq9CQddKKIYwGRgNssMEGa1aZpCYVwvIApTmprKw5ZKo+r75llZUpMKkvGMriML3Kyob3glq2LIV+FRXLr9c0b3WXV8nvpdYU15tKIR6j2H3961lXIEmSpEIoZJg0BVg/73Zf4LPVWAeAGOPVwNUAw4YNs4kvaZWVlaWpuYVcjamsbPm4VJIkSZLUGMoK+FgvAf1DCBuFENoAhwH3VlvnXuCYkGwLzHG8JEmSJEmSpOajYD2TYowVIYSTgYeAcuD6GOOEEMKY3PIrgbHAPsBEYCFwfKHqkyRJkiRJUv0KeZgbMcaxpMAof96Vedcj8MNC1iRJkiRJkqSGK+RhbpIkSZIkSSpyhkmSJEmSJElqMMMkSZIkSZIkNZhhkiRJkiRJkhrMMEmSJEmSJEkNZpgkSZIkSZKkBjNMkiRJkiRJUoMZJkmSJEmSJKnBDJMkSZIkSZLUYIZJkiRJkiRJajDDJEmSJEmSJDWYYZIkSZIkSZIazDBJkiSpxIUQ9gohvBtCmBhCOLOG5SGEcGlu+eshhK2zqFOSJBUHwyRJkqQSFkIoBy4H9gYGA4eHEAZXW21voH9uGg38uaBFSpKkomKYJEmSVNqGAxNjjB/EGJcAtwEHVFvnAOCmmDwPdAkhrFvoQiVJUnEwTJIkSSptfYDJeben5Oat6jqSJEkAtMq6gMYwfvz46SGEj5to892B6U207eaope0vuM8tQUvbX3CfW4KWtr8bZl1AEQs1zIursQ4hhNGkw+AA5ocQ3l3D2mrT0t7f4D63BC1tf8F9bgla2v5Cy9vnGttgJREmxRh7NNW2QwgvxxiHNdX2m5uWtr/gPrcELW1/wX1uCVra/mqNTAHWz7vdF/hsNdYhxng1cHVjF1hdS3x/u8+lr6XtL7jPLUFL219omftcEw9zkyRJKm0vAf1DCBuFENoAhwH3VlvnXuCY3FndtgXmxBg/L3ShkiSpOJREzyRJkiTVLMZYEUI4GXgIKAeujzFOCCGMyS2/EhgL7ANMBBYCx2dVryRJav4Mk+rX5F25m5mWtr/gPrcELW1/wX1uCVra/moNxBjHkgKj/HlX5l2PwA8LXVcdWuL7230ufS1tf8F9bgla2v5Cy9znlYTUdpAkSZIkSZLq55hJkiRJkiRJajDDJCCEsFcI4d0QwsQQwpk1LA8hhEtzy18PIWydRZ2NJYSwfgjhiRDC2yGECSGEU2pYZ2QIYU4I4dXcdG4WtTamEMJHIYQ3cvvzcg3LS+11Hpj3+r0aQpgbQji12jpF/TqHEK4PIXwZQngzb17XEMIjIYT3c5fr1HLfOv/um6ta9vl3IYR3cu/bf4YQutRy3zr/BpqrWvb5vBDCp3nv3X1quW/Rvc617O/f8/b1oxDCq7XctyhfY7VctsFsg+WWl8zr3BLaX2AbLG9eybbBWlr7C2yDrbIYY4ueSANRTgI2BtoArwGDq62zD/AAEIBtgReyrnsN93ldYOvc9U7AezXs80jg/qxrbeT9/gjoXsfyknqdvHlQXgAA+5tJREFUq+1bOfAFsGEpvc7AzsDWwJt5834LnJm7fibwm1qejzr/7pvrVMs+7wG0yl3/TU37nFtW599Ac51q2efzgJ/Wc7+ifJ1r2t9qy38PnFtKr7FTy5xsg9kGK9XXOW+/SrL9ldsH22CxtNtgLa39Vds+V1tuGyxvsmcSDAcmxhg/iDEuAW4DDqi2zgHATTF5HugSQli30IU2lhjj5zHGV3LX5wFvA32yrapZKKnXuZpdgUkxxo+zLqQxxRifAmZWm30AcGPu+o3AgTXctSF/981STfscY3w4xliRu/k80LfghTWhWl7nhijK17mu/Q0hBOBQ4NaCFiU1DdtgtsGqlNTrnKck219gGyxvXsm2wVpa+wtsg60qw6T0AT457/YUVv5Qb8g6RSmE0A/YCnihhsXbhRBeCyE8EELYvLCVNYkIPBxCGB9CGF3D8pJ9nYHDqP0fX6m9zr1ijJ9DarQDPWtYp5Rf6xNIv+7WpL6/gWJzcq5b+fW1dKUvxdd5J2BqjPH9WpaX2mus0mYbzDZYlVJ9nVtS+wtsg7WUNlhLbH+BbbCVGCal7rTVVT/FXUPWKTohhI7AncCpMca51Ra/QuqSOwT4E3B3gctrCjvEGLcG9gZ+GELYudryUn2d2wD7A3fUsLgUX+eGKNXX+mygAri5llXq+xsoJn8GNgG+DnxO6nZcXSm+zodT9y9ipfQaq/TZBrMNVqXkXmfbX7UqudcaWlQbrKW2v8A22EoMk1JSun7e7b7AZ6uxTlEJIbQmNWJujjHeVX15jHFujHF+7vpYoHUIoXuBy2xUMcbPcpdfAv8kdcHMV3Kvc87ewCsxxqnVF5Ti6wxMreoan7v8soZ1Su61DiEcC3wTODLGWOMHdgP+BopGjHFqjHFZjLESuIaa96WkXucQQivgIODvta1TSq+xWgTbYLbBqpTc60zLa3+BbbCSb4O1xPYX2AarjWESvAT0DyFslPsF4TDg3mrr3AscE5JtgTlVXTiLUe54z+uAt2OMf6hlnd659QghDCe9V2YUrsrGFULoEELoVHWdNFjem9VWK6nXOU+tKXqpvc459wLH5q4fC9xTwzoN+bsvGiGEvYAzgP1jjAtrWachfwNFo9pYGt+i5n0pqdcZ2A14J8Y4paaFpfYaq0WwDVbzOiX12dyC22Atrf0FtsFKvg3WQttfYBusZrWNzN2SJtIZJN4jjTp/dm7eGGBM7noALs8tfwMYlnXNa7i/O5K6Gr4OvJqb9qm2zycDE0ij7z8PbJ913Wu4zxvn9uW13H6V/Ouc26f2pMbJ2nnzSuZ1JjXSPgeWkn4F+S7QDXgMeD932TW37nrA2Lz7rvR3XwxTLfs8kXRsetXf85XV97m2v4FimGrZ57/m/k5fJzVQ1i2V17mm/c3Nv6Hqbzdv3ZJ4jZ1a7lTT32gpfzZjG6xFtMEo8fZXbh9sg5V4G6yW/S3Z9ldt+5ybfwO2wVaaQm7nJUmSJEmSpHp5mJskSZIkSZIazDBJkiRJkiRJDWaYJEmSJEmSpAYzTJIkSZIkSVKDGSZJkiRJkiSpwQyTJEmSJEmS1GCGSZIkSZIkSWowwyRJkiRJkiQ1mGGSJEmSJEmSGswwSZIkSZIkSQ1mmCRJkiRJkqQGM0ySJEmSJElSgxkmSZIkSZIkqcEMkyRJkiRJktRghkmSJEmSJElqMMMkSZIkSZIkNZhhkiRJkiRJkhrMMEmSJEmSJEkNZpgkSZIkSZKkBjNMkiRJkiRJUoMZJkmSJEmSJKnBDJMkSZIkSZLUYIZJkiRJkiRJajDDJEmSJEmSJDWYYZIkSZIkSZIazDBJkiRJkiRJDWaYJEmSJEmSpAYzTJIkSZIkSVKDGSZJkiRJkiSpwQyTJEmSJEmS1GCGSZIkSZIkSWowwyRJkiRJkiQ1mGGSJEmSJEmSGswwSZIkSZIkSQ1mmCRJkiRJkqQGM0ySJEmSJElSgxkmSZIkSZIkqcFa1bVw/PjxPVu1anUtsAUGT6urEnizoqLixKFDh36ZdTGSJKmwbE8VFdttkiQ1QJ1hUqtWra7t3bv3Zj169JhVVlYWC1VUKamsrAzTpk0b/MUXX1wL7J91PZIkqbBsTxUP222SJDVMfb+ObdGjR4+5NnxWX1lZWezRo8cc0q+RkiSp5bE9VSRst0mS1DD1hUllNnzWXO45tFu7JEktk+2pImK7TZKk+jXrD8ovvviifNCgQYMHDRo0uHv37kN69uz5tarbixcvDnXd96mnnmp/3HHHrb8qj9enT58tP//88zoP/ZMkSSomhW5PSZKk0tesg5PevXsve+edd94C+PGPf7xex44dl51//vlTq5YvXbqU1q1b13jfnXfeeeHOO++8sEClSpIkNUul2p6qq25JktS0mnXPpJocfPDB/U488cS+22yzzYAf/OAHfZ944on2W2211aDNNtts8FZbbTXotddeWwvg/vvv7zRq1KhNITWcvv3tb/cbPnz4wL59+255wQUX9Kzvcc4777xe/fv337x///6bn3/++T0B5s6dWzZy5MhNBw4cOLh///6bX3PNNesA/OAHP+izySabbD5gwIDBo0eP7tuU+y9JkrSmmrI9deSRR26wxRZbbLbppptuftppp61XNf/JJ59sv9VWWw0aOHDg4C233HKzWbNmlVVUVDB69Oi+AwYMGDxgwIDBv/rVr3rCir3Fn3rqqfbDhw8fWFXD4YcfvuEOO+zQ/6CDDtro3XffbTN06NCBgwcP3mzw4MGbPfLIIx2qHu+cc87pNWDAgMEDBw4c/IMf/KDPhAkT1ho8ePBmVcvfeOONtTbffPPNkCRJq6zBPZNOOIH133yT9o354FtswcLrr2fyqt5v0qRJbf/973+/16pVK2bOnFn24osvvtO6dWvuvvvuTqeffnrfhx56aFL1+0ycOLHts88+++7s2bPLN9tssy1+9rOfTVtrrbVqHL/g6aefbn/LLbd0Gz9+/NsxRoYOHbrZrrvuOu/9999fq3fv3kvHjRs3EWDGjBnlU6dOLR87duw6H3zwwZtlZWVMnz69fNWfCUmS1BK0hPbUH/7wh0979eq1rKKigu23337gCy+80G7IkCGLjzzyyE1uvvnmSSNGjFg4c+bMso4dO1b+/ve/7/Hxxx+vNWHChLdat27N1KlT621Hvf766+1feOGFdzp27BjnzZtX9vTTT7/Xvn37+MYbb6x1+OGHb/zmm2++ffvtt3f+17/+tc748ePf6dSpU+XUqVPLe/XqtaxTp07Lnn322Xbbb7/9oquuuqr7EUccMWNVnzdJktTMD3OrzUEHHTSrVatU+syZM8u/853vbPTRRx+1DSHEpUuX1njs/x577DG7Xbt2sV27dhVdu3ZdOmXKlFabbLLJ0prWHTduXMd99tlndufOnSsB9t1331lPPPFEp/3333/O2Wefvf73v//9PgcccMCcvfbaa/7SpUtZa621Kg877LAN99133znf+c535jTZjkuSJDWSpmpP3XjjjV1vuOGG7hUVFWHatGmtX3vttbYhBHr27Ll0xIgRCwG6du1aCfD44493HjNmzLSqw9V69eq1rL6699prr9kdO3aMAEuWLAnf/e53N3zrrbfalZWV8fHHH68F8Mgjj3Q+6qijpnfq1Kkyf7vHHXfc9Guuuab78OHDJ99zzz3rvPTSS2+v3rMnSVLL1uAwaXV+8WoqHTt2rKy6fsYZZ/QZMWLEvEceeWTSu+++22aXXXYZWNN98n81Ky8vp6KiotYBJ2Os+YQrX/va17565ZVX3rrzzjvXPvvss/s8+uijcy+++OLPX3311bfvvffezrfddts6f/7zn3s+//zz763B7kmSpBJV6u2pd955p81ll13Wa/z48W/36NFj2cEHH9xv8eLFZTFGQggrNbBqm19eXh4rK1N5ixYtWmFYhg4dOvy37l/96le9evbsufTOO+/8sLKyknbt2g3N2+5K9R977LGzfvOb36x32223zdtyyy0X9u7du97wSpIkrazoxkyqbu7cueV9+/ZdAnDVVVd1b4xt7rLLLvPHjh3bZd68eWVz584tGzt27DqjRo2a99FHH7Xu1KlT5Q9+8IOZp5566tRXX321/Zw5c8pyv+bNufLKKye//fbbjdp1XZIkqak1Vntq1qxZ5e3atavs2rXrssmTJ7caN27c2gBDhgxZPHXq1DZPPvlk+9x6ZUuXLmW33Xabe+WVV/ZYujR1bqo6zK1v375L/v3vf7cHuP3229ep7fHmzJlTvu666y4tLy/niiuu6LZsWcqG9tprr7l//etfu8+bN68sf7vt27ePI0aMmPPjH/94g+OOO2766u6nJEktXdGHSWecccYX5513Xt+tt956UFUDYk3tuOOOC4844ogZW2+99WZDhw7d7Oijj562ww47LBo/fny7r3/965sNGjRo8G9+85t1zz333M9nz55dvtdee/UfMGDA4J122mngBRdc0Gx+cZQkSWqIxmpPbbfddou22GKLhf3799/86KOP7jd06ND5AG3bto0333zzpB/96EcbDBw4cPDIkSMHLFy4sOy0006b1rdv3yWDBg3afODAgYOvu+66rgDnnnvuZ6effvoGQ4cOHVheXl5zl3Hg1FNP/fLWW2/tNmTIkEHvvfde23bt2lUCHHLIIXP33nvv2VXttl/+8pe9q+5zzDHHzAQ46KCD5q72jkqS1MKF2g7pAnjttdc+GjJkiL/aNILXXnut+5AhQ/plXYckSSos21PNy7nnnttrzpw55Zdccslnta1ju02SpLoV5QDckiRJ0qrafffdN/n444/XevLJJx3fUpKkNWCYJEmSpBbhkUcemZR1DZIklYKiHzNJkiRJkiRJhWOYJEmSJEmSpAYzTJIkSZIkSVKDGSZJkiRJkiSpwZp1mDR8+PCBd955Z+f8eeeff37Po446aoO67vPUU0+1BxgxYsSm06dPL6++zo9//OP1zj333F4NnS9JklSsCt2ekiRJpa9Zh0nf/va3Z9x6661d8+fdeeedXY866qiZDbn/k08+ObF79+7LmqY6SZKk5q9U21OVlZUsW9bsypIkqUVo1mHS0UcfPeuxxx5be9GiRQHg3XffbfPll1+23mOPPeYfeeSRG2yxxRabbbrpppufdtpp69V0/z59+mz5+eeftwI444wzevfr12+L7bfffsD777+/Vn2P/eyzz7YbMmTIoAEDBgzefffdN5k2bVo5wAUXXNBzk0022XzAgAGDv/nNb24M8K9//avjoEGDBg8aNGjwZpttNnjWrFnN+nmVJEktR6HbU7fccsvaX/va1wZtttlmg7fffvsBkydPbgUwZ86cskMOOaTfgAEDBg8YMGDwDTfc0AXgH//4R+fBgwdvNnDgwMHbbbfdAFi511P//v03f/fdd9u8++67bTbeeOPNjzrqqA0233zzwZMmTWpT2z48+eST7bfaaqtBAwcOHLzllltuNmvWrLKhQ4cOfPbZZ9tVrbP11lsPeuGFF/57W5IkNUyrBq95wgnr8+ab7Rv10bfYYiHXXz+5tsW9e/deNmTIkAV33nnn2kcdddTsG2+8sev+++8/q6ysjD/84Q+f9urVa1lFRQXbb7/9wBdeeKHdNttss6im7Tz99NPt//nPf3Z944033lq6dClf//rXB2+11VYL6yrtuOOO2+j//u//Ptl3333nn3rqqeudccYZ611//fWTL7300t4ff/zxG+3atYtVXb5///vf97700ks/3mOPPRbMmTOnrH379pVr9sRIkqSS1ALaU7vvvvv8ww477J3c9ruff/75va+55popZ5555rqdO3de9t57770FMG3atPLPPvus1cknn9xv3Lhx7wwaNGjJ1KlTVzqcrrqPPvqo7TXXXPPR3/72t0+AGvdhyJAhi4888shNbr755kkjRoxYOHPmzLKOHTtWHnfccdOvvfba7ttvv/3k119/fa0lS5aE2vZXkiTVrtn3oDn00ENn/v3vf18H4K677up69NFHzwS48cYbuw4ePHizwYMHD37//ffbvvbaa21r28YTTzzRcZ999pndqVOnyq5du1buscces+t6zBkzZpTPmzevfN99950P8L3vfW/G888/3xFg4MCBi771rW9tdMUVV3Rt3bp1BNh2223n//SnP13/ggsu6Dl9+vTy1q1bN9LeS5IkrblCtqc+/PDDNjvttFP/AQMGDL700kt7v/POO+0Annrqqc6nnXbal1Xr9ejRY9m4ceM6DB8+fN6gQYOWAPTq1ave49bWXXfdJbvuuuuCqts17cPrr7/etmfPnktHjBixEKBr166VrVu35rjjjpv16KOPrv3VV1+FK6+8svsRRxwxvUFPoCRJWkHDeybV8YtXUzryyCNnn3POOes/88wz7RcvXly24447LnznnXfaXHbZZb3Gjx//do8ePZYdfPDB/RYvXlxnMBZCaJR6nnjiifcfeOCBTnfffXeX3/72t+u9//77b1544YVfHHjggXPuueeetbfffvvNHnzwwfe22mqrxY3ygJIkqXS0gPbUySefvMEpp5zyxZFHHjnn/vvv73T++eevBxBjXOn+Nc0DaNWqVaysXN7R+6uvvvrvSvk9wGvbh9x2Y/XtdurUqXKnnXaae8stt3S59957u44fP/6tendIkiStpNn3TFp77bUrt91223knnnhiv4MOOmgmwKxZs8rbtWtX2bVr12WTJ09uNW7cuLXr2sYuu+wy/1//+leX+fPnh1mzZpU98sgjXepav1u3bss6d+687MEHH+wIcN1113Xbbrvt5i9btoxJkya12W+//eZdccUVU+bNm1c+Z86c8gkTJqw1fPjwRb/61a++2HLLLRe8+eabtf6qJ0mSVGiFbE/NmzevfIMNNlgKcMMNN3Srmj9y5Mi5f/jDH3pW3Z42bVr5qFGjFrzwwgud3nnnnTYAVYe59evX76tXX321A8AzzzzT/tNPP61xfKba9mHIkCGLp06d2ubJJ59sn1uvbOnSpQCMGTNm+hlnnLH+kCFDFjSkJ5QkSVpZsw+TAA477LCZ7777bruqLtnbbbfdoi222GJh//79Nz/66KP7DR06dH5d999xxx0Xfutb35q5xRZbbP7Nb35zk+HDh9e5PsBf/vKXD88444y+AwYMGPz666+3u+iiiz6rqKgIRxxxxEYDBgwYvMUWWww+6aSTpnbv3n3Zb3/72579+/fffODAgYPbtWtXecghh8xprH1X8QkhxBDCpgV+zAtCCNNDCF80wbbHhRBObOztNrUQwk4hhHebYLuN+vqGED4KIezWWNsrNiGEs0II12Zdh9QSFKo9dfbZZ392+OGHbzJ06NCB3bp1q6ia/+tf//rz2bNnl1e1mcaOHdtpvfXWq7j00ks/+ta3vrXpwIEDB3/rW9/aGOCYY46ZNWvWrPJBgwYNvuyyy3psuOGGNfb4rm0f2rZtG2+++eZJP/rRjzYYOHDg4JEjRw5YuHBhGcBOO+20sEOHDsuOP/54D3ETIYQbQggXFPgxdwghvB9CmB9COLCRt31cCOGZxtxmoYQQJoQQRjbyNhv19Q0hnBdC+Ftjba/YhBA2yL1v6x3fTqUvxLhSD+D/eu211z4aMmSIH7SN4LXXXus+ZMiQflnXUWxCCB8BJ8YYH826lobKdavvH2OcWKDHWx94D9gwxvhlfes3cJsxxhhy18cBf4sxNusv/IV63vMfJ4RwAzAuxnhDDeudBxBjPK+e7X1EI77HC/3+WxW5BuLfYox9m2jb58UYRzb2tqU1ZXuqefnoo49ajxw5cuCkSZPeLC+v+ftQqbXbcp/lQ4DeMcavMi6nWcl9lk+JMZ5TwMd8DLg3xnhJI23vBnLtkRDCcaR2xY6Nse2mUqjnPf9x6morhBD6kZ7DfvVs7zxg0xjjUY1dX2Nsr7E15Xex3LZHxhg/auxtqzCKomeSpDptCMxYnSAphNDwcdPUrPlaSlL9Lrvssm7bbrvtZueee+6ntQVJpSb3JXknIAL7F/ix/Wyq2YbAhNW5o89p6bB3j4qdYZKKUghhrRDCH0MIn+WmP4YQ1sot6x5CuD+EMDuEMDOE8HQIoSy37IwQwqchhHkhhHdDCLvWsO1tQwhf5P+DDyF8K4Tweu768BDCc7ntfx5CuCyE0KaWOlc4RKx61+MQwqAQwiO5Ot8NIRyat2yfEMJbuVo/DSH8tIbt7wY8AqyX63J6Q27+/rmuwrNzNWyWd5+Pcs/D68CCVW2UhBBOCCG8HUKYFUJ4KISwYd6yGEL4Qa7r9rwQwi9DCJvknq+5IYTbq56rEMLIEMKUEMLpIYQvc8/lgbn9fi/3nJyVt+1an/cQwlO51V7LPQ/fqdp+bvl3cvOrpq9yv9JWvZcuDiF8EkKYGkK4MoTQLu9xf5Z7vM9CCCesynOVt41a35M5Xw8hvB5CmBNC+HsIoW3efb8XQpiYu9+9IYT1qj3fPwwhvA+8X9PzUE9dK+xbyDuEr7HfuyGEDsADLH+vzg8hrBeqdRdvwHv3p7U9V5JUn5NPPnnGF1988foJJ5wwK+taCugY4HngBuDY/AUhhPVDCHeFEKaFEGaEEC7LW/a93Of9vNz/9K1z81c43DvkHUaU99l+RkiH3v8lhLBO7jNwWkhth/tDCH3z7t81hPCX3GfRrBDC3bn5b4YQ9stbr3VIh/R/vfoO5ur8Zt7tVrl1q2q+I6S23ZwQwlMhhM1reqKqf9ZV39+62gwN+Kyv2t4kYGPgvtxn4Vq5z8N7c/ebGEL4Xt7654UQ/hFC+FsIYS5wXE2116aez+sbQghXhBAeyNXy7xBC75Da1bNCCO+EELbKW/+jXNvh9RDCghDCdSGEXrn7zwshPBpCWCdv/Rqf9xDCaOBI4PTc496Xt/3dctdn57UXFuReh365Zd8MIbyaW+fZEMLX8h5zqxDCK7l6/g6sVjsh1P2doU0I4abcsgkhhGF599sspPbL7Nyy/fOW3RBC+HMIYWwIYQHw3ZqehzpqWmHfQgi3heV/e43+3g0h/BXYgOXv1dNDCP1y222Vu299793ba3uuVPwMk1Sszga2Bb5O6rY9HKjqHvoTYArQA+gFnAXEEMJA4GTgGzHGTsCewEfVNxxjfB5YAOySN/sI4Jbc9WXAaUB3YDtgV+AHq7oDIX25fiS33Z7A4cAVYXkD5zrgpFytWwCP11Dro8DewGcxxo4xxuNCCAOAW4FTc8/BWNKHQH7gdTiwL9AlxlhRbbNUHeJWQ80Hkp7Pg3Lbfjr3WPn2AoaSXp/TgatJH5Tr5/bj8Lx1e5M+5PsA5wLXAEfl7r8TcG4IYePcurU+7zHGnXPrDMk9D3+vtj9/z83vCKwHfJBX92+AAaT30qZ5tRBC2Av4KbA70B/Yrdp2j6vpELfcsvPyDnGr8T2Zt/qhuedtI+Br5BqKIYRdgF/nlq8LfAzcVu2hDgS2AQbX9zzkq2/f6rI6790Y4wJWfK92jDF+Vm27DXnv1vhcxRjHeYibJNXoGODm3LRnCKEX/LdXxP2kz5Z+pM+/23LLvg2cl7tvZ1KPphkNfLzeQFdS75vRpO8bf8nd3gBYBFyWt/5fgfbA5qTPlP/Lzb+J1Caosg/weYzx1Roe81ZWbF/sCUyPMb6Su/0A6bOuJ/AK6blYHbW2Gaj/sx6AGOMmwCfAfrnPwq9y9U8htVEOAS6sFl4cAPwD6FJT7bW1RxrweQ3pc/UcUvvqK+A50nPUPfeYf6i22YNJbYcBwH6k5/as3PplwI/y1q3xeY8xXp27/tvcc7Af1cQYu+S13S4htTk/DSkgvB44CegGXAXcmwtL2gB3k95TXYE7cvVWbbPWtkKM8aOqQ9wa8J1hf9LfShfgXnLv5xBCa+A+4OHcPv8PcHNue1WOAH4FdCK9x+t8HqrUt28NsMrv3Rjj0az4Xv1tDdut771b43NFeoB+HuJW3OoLkyorKyvrPwes6pR7DivrXVGr4kjg/BjjlzHGacD/A47OLVtK+uK9YYxxaYzx6ZgGB1sGrAUMDiG0zn1oTKpl+/9tlIQQOpEaMLcCxBjHxxifjzFW5P4BXgWMWI19+CbwUYzxL7ltvQLcSfpHXLUfg0MInWOMs/IaRPX5DvCvGOMjMcalwMVAO2D7vHUujTFOjjEuWsWaTwJ+HWN8OxdCXUjqVbNh3jq/iTHOjTFOAN4EHo4xfhBjnENqVGyVt+5S4Fe5Om8jNUQuiTHOy91/AikwaJTnPaRfCG8hHRN/VQghAN8DTosxzowxzsvt02G5uxwK/CXG+GYuDDlvVR6v2n7W9J6scmmM8bMY40xSI+TruflHAtfHGF/JNTZ/DmxX9ctczq9zta/qa7km+5b1e7em50pqzmxPFZFSareFEHYkhTi3xxjHA5NIX2Yh/RC3HvCzGOOCGOPiGGNVz4YTSV9wX4rJxBjjxw182Ergf2OMX8UYF8UYZ8QY74wxLsx9zv6K3Od3CGFd0g8NY3KfF0tjjE/mtvM3YJ8QQufc7aNJX6Rrcguwfwihfe52/o+AxBivz7UtviJ93g0JIdR59sLqGtBmqO+zvrbtrg/sCJyRew1eBa5lebsW4LkY490xxspV/Lyv7/Ma4J+5NtZi4J/A4hjjTTHGZcDfWbHdBvCnGOPUGOOnpIDnhRjjf3LP7T/z12+k5/07pNfz4Fzb4HvAVTHGF2KMy2KMN5JCsG1zU2vgj7nX4B/AS6vyeDn1fWd4JsY4Nvcc/ZX0wza5x+8IXBRjXBJjfJwU2OYHnffEGP+dey1rPLFALVZ73zJ+79b2XKkE1BcmvTlt2rS1bQCtvsrKyjBt2rS1SV+q1XjWI/2SVuXj3DyA3wETgYdDCB+EEM4EiGlA4lNJH2Zf5rqGrkfNbgEOCunQuYOAV6oaUSGEAbmuoF+E1N34QlIIsqo2BLbJdSmdHUKYTQoPeueWH0wKsT4OITwZQtiugdtd4bmJMVYCk0m/QFSZvBr1VtV8SV69M4FQbdtT864vquF2x7zbM3IfLlXLarp/R2i0573ql6CqX816kH4NHZ+3Tw/m5kN6LvOfq4Y2pKur8T2ZJ/8sfAtZ/hxVfy3nk34ZbozXck32Lcv3bm3PldSc2Z4qEiXYbjuW9KNO1QDwt7D8ULf1gY9jDT2Uc8tq+8GtPtPyvySHENqHEK4KIXyc+/x+CuiS6xm1PjAzxrjSYYcx9V79N3BwCKELKXSqsUdRro33NrBfLlDan1yYFEIoDyFcFEKYlHv8j3J3W9U2RH1thvo+62uzHuk5mJc372Mar91W1+c1rFq7rcHrN8bzHtIhdpcB34rpx+OqffpJtX1an/Q8rgd8Wi0IWeW2WwO+M1Rvi7TNHfa1HjA5137Jf/zGaret7r5l+d6t7blSCajzhayoqDjxiy++uPaLL77YAg+JW12VwJsVFRVFd2r1Zu4zVhy8cIPcPHL/0H5C+qDZHHgihPBSjPGxGOMtwC25X7muInX5PLr6xmOMb4UQPiY1XFb4dQv4M/Af4PAY47wQwqms+AtPvgWkf95V8j+8JwNPxhh3r+mOMcaXgANyXWZPBm4nfVjW5zNgy6obuV8j1gc+zd98A7ZTk8mknkSr2z18TazK876SEMJhpF+GvpH7ZQtgOqnhs3nuF7bqPmfF53yD1Sm8rvdkPXetep9X7UMHUpfuxngt69u3pnjv1ldrQ967UtGxPVVUSqbdFtJ4KIcC5SGNXwSpt0WXEMIQ0v/yDUIIrWoIlCYDm9Sy6YWs/PkwJe929f/1PwEGAtvEGL8Iacyj/5B+jJoMdA0hdIkxzq7hsW4k9ZJqReqdU9fnQVWv8jLgrbj8rKZHkA4T240UaKwNzMo9fnUrfPaFEPI/++psM6zhZ33XEEKnvC/lG9B47bZaP6+bWH3Pe537FELoQerpdHKM8T95i6raor+q4T4jgD4hpNMS52ZvwGoEow39zlDNZ8D6IYSyvEBpA9JZl/+76eoP1cCSPqfufWuq925d9TXkvasSVmeYNHTo0C8p8FkfpBq0DisOsltBajCcE0J4ifRP7lxSd2hCGoDxHdI/17mkrqrLQjpeuQ/pV67FpH+qdTXqbyH1YNmO9CtOlU657c4PIQwCvg9MW/nuALxK6uF0LSm9/y7Lf8G5H7gohHA0y8fB+TowP1f7t4H7Y4xzcr/oLKNhbgfODOl45aeAU0jdf59t4P3rciXwyxDCqzHGCbmuynvEGO9ohG3Xp77nfSppQMuJ1e+Y+2XrT8Dueb9sEWOsDCFcA/xfCOHkGOOXIYQ+wBYxxodIz+VfQgg3kRpC/7s6hdf2nmzAXW8Bbgsh3EL6xfVCUnfyj+q4T63PQzX17durNP57dyrQLYSwdkyHPdZUU1O9d6XM2J5SRg4k/f/dEliSN/920lhIp5O+oF4UQvjf3LpDY4z/Jh2q8oeQBvR9hRQsLc310n4VOCKEMIE0ds4I4OU66uhEanPNDiF0Je/zJsb4eQjhAdI4Pj8kfY5sF2OsOqHE3cAVpHFcahqvJd9tpB7IXVnxR8BOpM+SGaQv2xfWsY3XgM1zgdc75B0CXl+bYXU/62OMk0MIzwK/DulkKwNIn7mNcer5Wj+vY4xvN8L261Lf817VXllJrufKncDNceXxH68B/hlCeBR4MbftkaR2w3Ok7wk/CiFcTvq/Oxx4YlUKX43vDFVeIIU6p4cQfg/sQBpX6ht13KfW56Ga+vatqd67tdbXxO9dFQF/HVMxGEv6J141nQdcQGq4vA68QWroXJBbvz/wKKlB8hxwRYxxHOnXuItI6fwXpIHx/nu2sBrcSvpwejyvezikQYuPAOaRPtBqHeSYNIjkEtI/4hvJ656dS/D3IB2v/Fmupt/k6oT068dHuS/jY2jgP+YY47u5df9E2tf9SAPnLanzjg3b9j9zNd6Wq+tNUu+tQqjveT8PuDHXfffQassOANYBngnLzwzyQG7ZGaTg5fncPj1K+gWVGOMDwB9Jg59PpIZB0BuotvdknXK/CP2C1KD6nNSYP6zOO9X9PORvu759a/T3bozxHdLf1Qe5+lY4zLQp37uS1AIdSxob75MY4xdVE+mwoSNJPUT2Iw3G+wmpd9F3AHI/Ev2KFMrMI4U6XXPbPSV3v9m57dxdTx1/JI1/N510VrkHqy0/mjRmyzvAl6TDi8jVsYj0GbgRcFddDxJj/Jz0Gbs9K7YRbiIdevMp8Fauhtq28R5wPukz+33gmWqr1NpmYDU/63MOJw2C/hmpN87/xhgfaeB9a9WAz+umVN/zfh1pTKLZIXcGvzx9SSdiOTWseDbeDWKML5PG/7mM1NNpIstPxrGENDzFcbll36Ge900tVvU7A3mPvz+pbTydFIQek2v/1Kau56H6tmvdtyZ87/6a9AP+7FDDmaVpoveuikNowNhakqQWIIQQgf55hwZIkpSpEMK5wIAYo70dpDwhhBuAKTHGc+pbV2oKDn4lSZIkqdnJHRb3Xeofq0aSVGAe5iZJJSqEcFa17uHVD/GT1AKEEK4PIXwZQqjxDGUhuTSEMDGE8HoIYetC1yhVF0L4Hmmw5QfyxlCSSlYIYYNa2m3zQwirdRIYqSl5mJskSVIJCyHsTBoL46YY4xY1LN8H+B9gH2Ab4JIY4zaFrVKSJBUTeyZJkiSVsFyvjpl1rHIAKWiKMcbnSaeOX7cw1UmSpGJkmCRJktSy9SEdTlRlSm6eJElSjUpiAO7u3bvHfv36ZV2GJElqIuPHj58eY+yRdR0lKtQwr8ZxEEIIo4HRAB06dBg6aNCgpqxLktQAVSPX5F9Wn1ddqOk/fyPXUN+yhl6u6fYbuqz69YbcXt371LeNVV2vVy/o0KFh21hVtbXBSiJM6tevHy+//HLWZUiSpCYSQvg46xpK2BRg/bzbfYHPaloxxng1cDXAsGHDou0vSU0pRqioqHtatqxpp4oKWLp0+eM15Pqq3qeysubHbsh8NZ4Qlk9lZSveXpV55eUrz6sK7+q6bMg6tV3+8Y+w555N8rTU2gYriTBJkiRJq+1e4OQQwm2kAbjnxBg/z7gmSY2kogK++mrlacmS1Z+WLm34evWFQXVNlZVZP3s1CwFat4ZWrdK0KtfXWgs6dlw+v1WrFEiUl9c81basofPLahjYZnV629Q1r+pxqh67ruursm5Db6/psvwgRw1nmCRJklTCQgi3AiOB7iGEKcD/Aq0BYoxXAmNJZ3KbCCwEjs+mUql0VVTAwoUrTgsWrDyv+vyaQqD8MKiu5VXrNEUg07o1tGlT/9SqFbRvv2Jw0phTbQFMY041BUI1BTRSS2OYJEmSVMJijIfXszwCPyxQOVKzUVkJixYtnxYuXPF2TVP1dRoaEC1Zsur1tW4NbdumnizVpzZt0mWHDtC1a+3L65qq1mlIKJQfHrVubS8OSYZJkiRJkpqZZctSMFM1VQU1DbmdH+rUFRKtTsBTpW1baNcuhTnt26epQwfo1CkNhJs/r+p6/lTT/Px57dqlHjCS1Fz5L0qSJEnSGokR5s+HuXNhzpyGXc6fX3s4tKpBT1nZ8jCmQ4fl19u1g54902X+VLWsrqm2ddq2tWeOJBkmSZIkSS1cZSXMng3TpqVp+vSGh0Jz5sC8eQ0bm6dTJ+jcOU0dO6bQZ731VgyAqq6vyu211jLgkaRCMkySJEmSSsyyZTBz5vJwqK7pyy9TeFTXacbbtoW1104hUNVl//4r3q66rGne2mun8Ki8vHDPgSSp6RgmSZIkSUUgxhT8fPABfPLJyoFQ/u2ZM2vvKdSlC/TokaaNN4ZttkmHglXN69EDundP61WFQ23aFHJPJUnNnWGSJEmS1EwsWQIff5wCo0mTll9WXV+wYMX1Q0hn86oKgTbbDEaMWDEYqh4StW6dzb5JkkqHYZIkSZJUQLNmrRwWVV1Onrxij6K2bVPvoY03hl12SZebbAIbbph6E3Xr5qFjkqTCM0ySJEmSGlFlZQqFaguMZs1acf0ePVJAtMMO6XKTTZaHRr17pzOVSZLUnBgmSZIkSY3gnXfgL3+Bm26CL75YPr9VK+jXLwVE3/jG8qCoqsdRp06ZlSxJ0moxTJIkSZJW09y58Pe/pxDpuefSIWf77gv77AObbprCovXXT4GSJEmlwo81SZIkaRVUVsJTT6UA6Y47YNGiNPD1734HRx2VDk2TJKmUGSZJkiRJDfDJJ3DjjXDDDWn8o86d4eij4YQTYPjwdGY1SZJaAsMkSZIkqRaLF8Pdd8P118Ojj0KM6axq/+//wUEHQfv2WVcoSVLhGSZJkiRJeWKE8ePTYWy33AKzZ8MGG8C558Kxx8JGG2VdoSRJ2TJMkiRJkoBp0+Bvf0sh0htvQNu2qffRCSfAqFFQVpZ1hZIkNQ+GSZIklboY4V//gjlzoEMH6Nhx+WX+9XbtHPRFLU5FBTz4YDqM7b770u3hw+HPf4bDDoMuXbKuUJKk5scwSZKkUnf11TBmTP3rhZCCpZqCpobM23RT2HLLpt8fqRG8807qgXTTTfDFF9CjB5xyChx/PGy+edbVSZLUvBkmSZJUyj77DE4/PR2jc+WVsGABzJ+fpqrr1S+rz5szBz79dMV5ixat/FijR8NVVxV+H6VVMH8+HHggPPYYlJfDvvumw9j22Qdat866OkmSioNhkiRJpexHP4IlS1LvpE03bbztLluWQqX8gGnttRtv+1ITOfVUePxxuPDC1Aupd++sK5IkqfgYJkmSVKruuQfuvDN9a27MIAlSl47OndMkFYl//hOuuw5+/vM0SZKk1eM5KSRJKkVz58IPfwhf+xr89KdZVyNl7rPP4Hvfg623hvPOy7oaSZKKmz2TJEkqRWedlb4933WXA8GoxausTIe0LVwIN98MbdpkXZEkScXNMEmSpFLz3HNwxRXwP/+TznEutXCXXQYPPwx//jMMGpR1NZIkFT8Pc5MkqZQsWZKO5enbFy64IOtqpMy9+WY6oeE3vwknnZR1NZIklQZ7JkmSVEp++1uYMAHuvx86dcq6GilTX30FRx6ZTjR43XUQQtYVSZJUGgyTJEkqFe++C7/8JRx6KOy7b9bVSJk7+2x4/fWUrfbsmXU1kiSVDg9zkySpFFRWwujR0L49XHJJ1tVImXvsMfj97+H73zdblSSpsdkzSZKkUnD99fDUU3DttdC7d9bVSJmaOROOPTYNtn3xxVlXI0lS6TFMkiSp2H3xBfzsZzBiBJxwQtbVSJmKEcaMgalT4d57U2c9SZLUuDzMTZKk+syaBdOmZV1F7U45BRYtgquvdoRhtXg33QR33JGGD9t666yrkSSpNBkmSZJUl3//Ox0rM3hwGsm3ubn/frj9dvjFL2DAgKyrkTL1wQdw8smw886ps54kSWoahkmSJNXmuutg1Cjo3BnatoVddoHXXsu6quXmzUujC2+xhd+c1eJVVMDRR0N5Ofz1r+lSkiQ1DcMkSZKqq6iAH/0ITjwRRo6EF1+EcePS4Cu77AL/+U/WFSbnnAOffgrXXANt2mRdjZSpX/8ann0WrrgCNtgg62okSSpthkmSJOWbORP22gv+9Cc47TQYOxbWWQc22SQFSh07wq67wiuvZFvnCy+kGn/4Q9h222xrkTL24ovw//4fHHFEmmq1cCF88knqYfjeezB9egqPJUnSKvFsbpIkVZkwAQ44ACZPhuuvh+OPX3H5xhunQGnUqBQoPfooDB1a+DqXLoXvfQ/WWw9+9avCP76UtRjTYZ4zZrDwk+lcefgMfthlBr/ecjqcOyOFRDPyLquuL15c8/bWXhu6dYOuXeue8tdZZx1o3bqw+y1JUjNhmCRJEsB996UuDR06pMBou+1qXm+jjVYMlB55BL7xjUJWChdfDG+8Affck8ZzkkrJxx+nv7EZdYRCM2akUBVoD1xfdd+fk85oWBX8dO+ejnnbeuvlt7t1S0HQ4sWpJ2LVNGPG8usffpguZ82Cysraa+3UqfbAqXdv6NcPNtwwXfq3KkkqIYZJkqSWLcY02Mo556QvnHffDX371n2ffv2WB0q77QYPPwzbbFOAYoH330/H8xxyCOy/f2EeUyqkl1+G445L18vLlwdA3bpB//7psM7cvFc+7sb/XtadfY7qxvd/kVuvS5fGG327shLmzq09dKo+vfHG8uvVD59bZ50Vw6Xq17t0aZyaJUkqAMMkSVLLtXAhnHAC/P3vcPjh6ext7do17L4bbghPPpkG6N5jD3jooaYfuyhGOOmkdGa5Sy9t2seSsrL77jBxYgqMOndOPY1q8PnnsMeWsOHW8N3rgKYYg76sLIU8Xbqkw1wbKkaYNg0++ij1tProo+XT+++nHo0LFqx4n7XXrj1o6tcvhVG1PBeSJBWaYZIkqWWaPBkOPDCdme2ii+D001f9i9r6668YKD34IGy/fVNUm9xwAzzxBFx1Fay7btM9jpSlzp3rPSQsxjSk2cKFcPPNzfBkhiFAz55pGj585eUxph5ONYVNH34Ijz8O8+eveJ+OHVcMmnr3hlatUi+s6lNZWePMa906Pbn5lzXNM+SSpBbHMEmS1PI8+ywcdFD6JnrvvfDNb67+tvr2TYHSqFGw554pUNphh8artcrUqfCTn8BOO8GJJzb+9qUictllqTPgFVfAoEFZV7MaQkg9r7p3h2HDVl4eYxqvqaaw6eOP4amn0uF3zUV5ecNCp+rzOnZc3vOrS5fUO6u22+3bG1pJUjNimCRJalmuvx7GjEm/7D/+OAwevObb7NNn+RhKe+4JDzyQQp/GdNpp6bCYq69OPQekFmrCBPjZz2DffdOfckmqGkS8a9c0lltNvvoqjem0bNmK05rMy59fUZGmJUvSYOdLly6/vqbzFixIYdmkSTB7drq+ZEndz0mrVnWHTTXN69oVevVKY2kVy//NGGHOnPQDwowZ6aQQVe8FAzVJzYhhkiSpZaiogJ/+FC65JA2a/fe/p8Z5Y1lvveWB0t57w9ixsPPOjbPtBx6AW29NA28XZTcMqXF89RUceWQ6Cu6661r49+q11sq6gsa1eHEKlubMSZdVU123P/98+e3qY1DlKyuDHj1SsNSzZ7qsmvJvVx2a2NjHTVb1NJs6dfn0xRcr3s6fagvW2rRZ8YyBNU01LevYsYX/sUhqCoZJkqTSN3MmfOc78OijcOqp8LvfpV+5G9u666ZAaZddUqD0r3+l8ZTWxPz5qfvFZpvBGWc0RpVS0TrnHHjtNbjvvvTdXyWkbds0DlTv3qt3/6VLU7BUFS7NmpX+91cFNF9+ufz6pEnpcuHCmre1zjorh0w1Xe/YMQ20Xl9I9OWXqb7qystX3N7gwemyd+/lPaoWLqz5zIEzZqTxtcaPT7dr2xdIn3e1BU2dOqXeTx06pJ5PdV1v3z5NxdLLS1KTMkySJJW2t96C/fdPA25ff30atbcp9e6dBsneZRfYZx+4//50fXWdey588gk880zp9USQVsHjj8Pvfw/f//6aDXOmEtW69fJxqBpq/vwVQ6aarr/+ero+a1bD68gPiL72teXhUPWpa9fGC2YWL14eoFWFTTWFUDNnwpQpKZWdOXPlgd4bol272sOm6vM6d17e46tq6tEjHYbY3HpLLVkC06engLBqmjkz1bnWWitPbdvWPD9/eakOUL90Kcybl8ZuW7gwhZIeitniGCZJkkrXffelY2Lat08BT1OeaS1fr17p8XbdNX3rve++dH1VvfRSOizv+99vmkG9pSIxaxYccwwMGAAXX5x1NSoZHTumaeON6193yZIUKuUHTvPnLz98rmpaZ51svky3bZt6x67qmT4rK2HRohQILFiQpqrrNc2ra/mMGenHj+rza9K6dXruqsKlmgKn/NsdOqz6c7Jo0YrBUH3TnDmr/hgNUVfYVHXZvv3ykG5NL9u2rfk9GGN6TubOrX2qCojqmxYvrnlf27RJfwP5vd/yb9d2vUuX1FOvMSxduvz9VzXNn1/37bKylcdfqz7V9ry2YIZJkqTSEyP85jdw1llp8Np//hPWX7+wNfTsmbpSVAVK994Lu+/e8PsvXQrf+176RfvXv266OqVmLsZ0pOfUqXDPPem7klRwbdqks3f27Zt1JY2rrGx5b6IePRp/+0uXpt4+VUFc1TRt2oq3J05M82rrKdW+fc2hU+fOqfdQTeFQbUFWq1ZpW1XT0KEr3s6funVL9/nqq7qnxYvrX6e29RYvTodHVoV6VZcLF6bB8FdHu3bLp/Ly5SFRQ7bXunUayL5z59TjqHPnFFIOHJiuV5/atUvbzu8ZV3V98uTUC27WrFRDXdZeu/bQadmyhgdD9Q3mX9NzVVmZXou6tGlTd9hU0wkBqqaOHdM2Ylw+Vb9d01TfOvnLe/Zc/jgFYpgkSSotCxfCiSemAasPOyyN0pvVt88ePZYHSvvtl74J77lnw+77f/+XGmB33ZUaKFIL9de/wu23p0x16NCsq5G0Slq3XrUeUwsXrhw0Vb/92Wfp8/HLL1Nw0K7d8vCne/d0oorawqEePdJnarH0MFm6dOWQqbbL2pYtW7Zi+FMVENU2NdUh9UuXplCpptCpptuTJy+/3qrV8tCzY8fl13v1gk02WX67pnVqul01L38MsMWLVx7wv77pk0+WX6+tt1ah3HEHHHJIQR8yxKpEq4gNGzYsvvzyy1mXIUnK2pQpcOCB8MorcOGFacDq5tBgnDEjnUHu7bdTL6m99657/UmTYIst0np33VWYGpu5EML4GOOwrOvQcoVof334IQwZAlttlXLZxjoKQlIJiDH1JmnbNutKpLrDqKoedyGsPNU2f1WXjxgB/fo1ya7V1gazZ5IkqTQ89xx861vpl7h7721eI/R26waPPZYCpQMPTIHSPvvUvG6McNJJqTv1n/5U0DKl5qSiAo4+Ov1ofNNNBkmSqgnBIEnNR9u2aWpBpxr1vI6SpOK2ZAlceSWMHJm6LD//fPMKkqp07QqPPgpbbplCr/vvr3m9v/41BU8XXQR9+hS2RqkZuegi+Pe/4YorYMMNs65GkiTlM0ySJBWnd96Bn/0sDYb6/e/DzjvDiy/C4MFZV1a7rl3hkUfScTsHHZR6UOWbNg1+/ON01rmTTsqmRqkZePFFOO88OOKINEmSpObFMEmSVDwWLoQbb4SddoLNNoM//jFdHzsWHnoohTXN3TrrwMMPp0FgDjkE7r57+bIf/zidEeWaa5YPCCm1MPPnw1FHpY55l1+edTWSJKkmBR0zKYSwF3AJUA5cG2O8qNrytYG/ARvkars4xviXQtYoSWqGXnklBSy33JLClgED4Le/hWOOKc5j07t0SYHSXnvBt78Nf/97OrPI3/4G557bvHtXSU3szjvTWcKfeCL9qUiSpOanYGFSCKEcuBzYHZgCvBRCuDfG+Fbeaj8E3oox7hdC6AG8G0K4Oca4pFB1SpKaidmzU3h07bXwn/+kQQ2//W343vdgxx2bx1na1sTaa6feVHvtBYcemgbpHjgQzjor68qkTB17LAwdmk5oKEmSmqdC9kwaDkyMMX4AEEK4DTgAyA+TItAphBCAjsBMoKKANUqSshQjPPNMCpDuuAMWLYKvfz0d63LEEaXXTaFz5xQo7b13Gmn4jjtgrbWyrkrKnEGSJEnNWyHDpD7A5LzbU4Btqq1zGXAv8BnQCfhOjLGyMOVJkjIzdWo69/e118J776WQ5dhjUy+krbfOurqm1alTGpT7gw9g882zrkaSJEmqVyHDpJqOR4jVbu8JvArsAmwCPBJCeDrGOHeljYUwGhgNsMEGGzRupZKkprdsWRo36Npr01nNKirS4WtnnZUGpu7QIesKC6ddO4MkSZIkFY1ChklTgPXzbvcl9UDKdzxwUYwxAhNDCB8Cg4AXq28sxng1cDXAsGHDqodSkqTm6uOP4S9/geuvh8mToXt3OOUU+O530xnaJEmSJDVrhQyTXgL6hxA2Aj4FDgOOqLbOJ8CuwNMhhF7AQOCDAtYoSWoKS5ak3kfXXpt6IwHssQf84Q+w//7Qpk229UmSJElqsIKFSTHGihDCycBDQDlwfYxxQghhTG75lcAvgRtCCG+QDos7I8Y4vVA1SpIaWYxw883wk5/Al19C377wi1/A8cdDv35ZVydJkiRpNRSyZxIxxrHA2Grzrsy7/hmwRyFrkiQ1kS+/hO9/H+66C7bbDm64IfVGKi/PujJJkiRJa6CgYZIkqYX45z/hpJNgzhz47W/hxz82RJIkSZJKhGGSJKnxzJ4NP/oR/PWvsNVW8PjjsMUWWVclSZIkqRGVZV2AJKlEPPxwCo5uuQX+93/hhRcMkiRJkqQSZJgkSVoz8+ensZH23BM6d4bnn4fzzoPWrbOuTJIkSVITMEySJK2+p5+GIUPgqqvSGdvGj4dhw7KuSpIkSVITMkySJK26xYvhpz+FESPS7SefhIsvhnbtsq1LkiRJUpNzAG5J0qp5+WU45hh4+20YMwZ+9zvo2DHrqiRJkiQViD2TJEkNs3RpGlh7221h7lx48EH4858NkiRJkqQWxp5JkqT6vflm6o30n//A0UfDJZfAOutkXZUkSZKkDNgzSZJUu2XL4Le/haFDYcoUuPNOuOkmgyRJkiSpBbNnkiSpZhMnwrHHwrPPwre+BVdeCT17Zl2VJEmSpIzZM0mStKLKSrj8chgyBCZMgL/+NfVIMkiSJEmShD2TJEn5Jk+GE06ARx+FPfaA666Dvn2zrkqSJElSM2LPJEkSxAg33ghbbAHPPZcOaXvwQYMkSZIkSSuxZ5IktXTTp8N3vwv33gs77QQ33AAbb5x1VZIkSZKaKXsmSVJLFiMceWTqhXTxxfDEEwZJkiRJkupkzyRJasluvRUefhguvRT+53+yrkaSJElSEbBnkiS1VDNnwqmnwvDh8IMfZF2NJEmSpCJhzyRJaqlOPz0FSo88AuXlWVcjSZIkqUjYM0mSWqKnnoLrroMf/xiGDMm6GkmSJElFxDBJklqar76C0aOhXz/43//NuhpJkiRJRcbD3CSppbnoInj3XXjgAejQIetqJEmSJBUZeyZJUkvy7rtw4YVw2GGw115ZVyNJkiSpCBkmSVJLESOMGQPt28Mf/5h1NZIkSZKKlGGSJLUUN9wA48bBb38LvXplXY2kAgoh7BVCeDeEMDGEcGYNy9cOIdwXQngthDAhhHB8FnVKkqTiYJgkSS3BtGnw05/CjjvCd7+bdTWSCiiEUA5cDuwNDAYODyEMrrbaD4G3YoxDgJHA70MIbQpaqCRJKhqGSZLUEvz4xzBvHlx9NZT5r19qYYYDE2OMH8QYlwC3AQdUWycCnUIIAegIzAQqClumJEkqFn6jkKRS98gj8Le/wZlnwmabZV2NpMLrA0zOuz0lNy/fZcBmwGfAG8ApMcbKwpQnSZKKjWGSJJWyRYvg+9+HAQPgrLOyrkZSNkIN82K123sCrwLrAV8HLgshdF5pQyGMDiG8HEJ4edq0aY1dpyRJKhKGSZJUyn75S5g0Ca68Etq2zboaSdmYAqyfd7svqQdSvuOBu2IyEfgQGFR9QzHGq2OMw2KMw3r06NFkBUuSpObNMEmSStUbb8DvfgfHHQejRmVdjaTsvAT0DyFslBtU+zDg3mrrfALsChBC6AUMBD4oaJWSJKlotMq6AElSE6ishJNOgi5d4OKLs65GUoZijBUhhJOBh4By4PoY44QQwpjc8iuBXwI3hBDeIB0Wd0aMcXpmRUuSpGbNMEmSStFVV8Fzz8FNN0G3bllXIyljMcaxwNhq867Mu/4ZsEeh65IkScXJw9wkqdR89lk6c9uuu8JRR2VdjSRJkqQSY5gkSaXmlFNgyZI06Hao6SROkiRJkrT6PMxNkkrJ/ffDP/4Bv/oVbLpp1tVIkiRJKkH2TJKkUjF/Pvzwh7D55vDTn2ZdjSRJkqQSZc8kSSoV554Ln3wC//43tGmTdTWSJEmSSpQ9kySpFIwfD5dcAmPGwPbbZ12NJEmSpBJmmCRJxa6iAkaPhp494de/zroaSZIkSSXOw9wkqdj96U/wyitw++3QpUvW1UiSJEkqcfZMkqRi9skn8ItfwL77wiGHZF2NJEmSpBbAMEmSilWM6extMcLll0MIWVckSZIkqQXwMDdJKlZ33gn33w+//z1suGHW1UiSJElqIeyZJEnFaM4c+NGPYOut06UkSZIkFYg9kySpGP385zB1Ktx3H7TyX7kkSZKkwrFnkiQVm+eegyuvTD2Shg7NuhpJkiRJLYxhkiQVk6VLYfRo6NsXfvnLrKuRJEmS1AJ5bIQkFZOLL4Y334R774WOHbOuRpIkSVILZM8kSSoWkybB+efDwQfDfvtlXY0kSZKkFsowSZKKQYwwZgy0aQOXXpp1NZIkSZJaMA9zk6Q1sXAhnHMOzJ0L3bunqUePFS+7d4dOnSCE1X+cm2+GRx+Fyy+H9dZrvPolSZIkaRUZJknS6lq6FA45BB58ENZdF6ZNS/Nq0qZN3WFT9evduqX7AMyYAaedBttsAyedVLj9kyRJkqQaGCZJ0uqorITjj4cHHoCrrkpnWIsR5s2D6dNTsDR9eu3XX3klXc6aVftjdO6cgqVly2D2bLj6aigvL9guSpIkSVJNDJMkaVXFCD/+cTr07IILUpAE6TC2zp3TtPHGDdtWRUXqeVQ9bKoeQP3kJ/C1rzXdPkmSJElSAxkmSdKq+vWv4ZJL4JRT4Kyz1mxbrVpBr15pkiRJkqQi4NncJGlVXHMNnH02HHkk/OEPazaotiRJkiQVIcMkSWqou+6CMWNg773hL3+BMv+FSpIkSWp5/CYkSQ3x+ONw+OHpjGp33AGtW2ddkSRJkiRlwjBJkuozfjwccAD07w/33w8dOmRdkSRJkiRlxjBJkury3nvpsLZu3eChh6Br16wrkiRJkqRMGSZJUm0++wz22ANihIcfhj59sq5IkiRJkjLXKusCJKlZmjUL9twTZsyAceNgwICsK5IkSZKkZsEwSZKqW7gQvvnNdIjbAw/A0KFZVyRJkiRJzYZhkiTlW7oUvv1teO65dNa2XXbJuiJJkiRJalYMkySpSmUlnHACjB0LV10FBx+cdUWSJEmS1Ow4ALckQRpk+yc/gb/9DS64AEaPzroiSZIkSWqWDJMkCeCii+CPf4RTToGzzsq6GkmSJElqtgyTJOmaa1KAdOSR8Ic/QAhZVyRJkiRJzZZhkqSW7a67YMwY2Htv+MtfoMx/i5IkSZJUF781SWq5nngCDj8cttkmnbmtdeusK5IkSZKkZs8wSVLL9MorcMAB0L8/3H8/dOiQdUWSJEmSVBQMkyS1PO+/D3vtBeusAw89BF27Zl2RJEmSJBUNwyRJLctnn8Eee0CM8Mgj0KdP1hVJkiRJUlFplXUBklQws2bBnnvC9OkwbhwMGJB1RZIkSZJUdAyTJLUMCxfCfvvBe+/BAw/A0KFZVyRJkiRJRckwSVLpW7oUvv1tePZZuP122GWXrCuSJEmSpKJlmCSptFVWwgknwNixcNVVcMghWVckSZIkSUXNAbglla5ly+Ckk+Bvf4MLLoDRo7OuSJIkSZKKnj2TJJWmigo49li45RY45xw466ysK5IkSZKkkmCYJKn0fPUVHH44/POf8Otfw5lnZl2RJEmSJJUMwyRJpWXRIjjoIHjwQbjkEvjRj7KuSJIkSZJKimGSpNIxfz7svz+MGwfXXAMnnph1RZIkSZJUcgyTJJWG2bNhn33gxRfhr3+FI4/MuiJJkiRJKkmGSZKK3/TpsOee8MYbcPvt6TA3SZIkSVKTMEySVNy++AJ22w0mTYJ77oG99866IkmSJEkqaYZJkorX5Mmw667w2WcwdiyMGpV1RZIkSZJU8gyTJBWnSZNSkDRrFjz8MGy/fdYVSZIkSVKLYJgkqfi8804KkhYvhscfh6FDs65IkiRJklqMskI+WAhhrxDCuyGEiSGEM2tZZ2QI4dUQwoQQwpOFrE9SEXjtNdh5Z1i2DJ580iBJkiRJkgqsYD2TQgjlwOXA7sAU4KUQwr0xxrfy1ukCXAHsFWP8JITQs1D1SSoCL74Ie+0FHTrAY4/BgAFZVyRJkiRJLU4heyYNBybGGD+IMS4BbgMOqLbOEcBdMcZPAGKMXxawPknN2dNPp7O2rbNOum6QJEmSJEmZKGSY1AeYnHd7Sm5evgHAOiGEcSGE8SGEYwpWnaTm65FHYM89oU8feOop6Ncv64okSZIkqcUq5ADcoYZ5sdrtVsBQYFegHfBcCOH5GON7K20shNHAaIANNtigkUuV1Gzcdx8ccggMGpRCpZ4e/SpJkiRJWSpkz6QpwPp5t/sCn9WwzoMxxgUxxunAU8CQmjYWY7w6xjgsxjisR48eTVKwpIzdfjscdBAMGQJPPGGQJEmSJEnNQCHDpJeA/iGEjUIIbYDDgHurrXMPsFMIoVUIoT2wDfB2AWuU1FzceCMcfjhsuy08+ih07Zp1RZIkSZIkCniYW4yxIoRwMvAQUA5cH2OcEEIYk1t+ZYzx7RDCg8DrQCVwbYzxzULVKKmZ+POf4Qc/SANu3313OnubJEmSJKlZKOSYScQYxwJjq827strt3wG/K2RdkpqR3/8efvpT2G+/dJhb27ZZVyRJkiRJylPIw9wkqXYxwi9/mYKkb38b7rzTIEmSJEmSmiHDJEnZixF+/nM491w45hi45RZo3TrrqiRJkiRJNSjoYW6StJLKSjj1VPjTn2DMGLj8cigz55YkSZKk5spvbJKys2wZjB6dgqQf/xiuuMIgSZIkSZKaOb+1ScrGkiVwxBFw3XXwi1/AxRdDCFlXJUmSJEmqh4e5SSq8RYvSINv/+hf87ndp0G1JkiRJUlEwTJJUWPPmwQEHwLhxcOWVcNJJWVckSZIkSVoFhkmSCmfmTNhnH3j5ZfjrX+HII7OuSJIkSZK0igyTJBXG1Kmwxx7wzjvwj3/AgQdmXZEkSZIkaTUYJklqepMnw267wZQpcP/9sPvuWVckSZIkSVpNhkmSmtbEibDrrjB7Njz8MOywQ9YVSZIkSZLWgGGSpKbz5pupF1JFBTzxBGy9ddYVSZIkSZLWUFnWBUgqUS+9BCNGQFkZPPmkQZIkSZIklQjDJEmN76mn0qFtnTvD00/D4MFZVyRJkiRJaiSGSZIa14MPwl57QZ8+8MwzsPHGWVckSZIkSWpEhkmSGs+dd8L++8PAgal3Up8+WVckSZIkSWpkhkmSGseNN8Khh8I3vpEG2+7RI+uKJEmSJElNwDBJ0pq7/HI47jgYNQoefhi6dMm6IkmSJElSEzFMkrRmLroITj45Hd52//3QoUPWFUmSJEmSmpBhkqTVEyOcdRb8/OdwxBHwj39A27ZZVyVJkiRJamKtsi5AUhGqrIRTToHLLoPRo+GKK6C8POuqJEmSJEkFYM8kSaumogK++90UJP3kJ3DllQZJktTMhRD2CiG8G0KYGEI4s5Z1RoYQXg0hTAghPFnoGiVJUvGwZ5KkhluyBI48Mh3S9v/+H/ziFxBC1lVJkuoQQigHLgd2B6YAL4UQ7o0xvpW3ThfgCmCvGOMnIYSemRQrSZKKgmGSpIZZuBAOPhgefBD+8Ac47bSsK5IkNcxwYGKM8QOAEMJtwAHAW3nrHAHcFWP8BCDG+GXBq5QkSUXDw9wk1W/uXNh7b3joIbj6aoMkSSoufYDJeben5OblGwCsE0IYF0IYH0I4pmDVSZKkomPPJEl1mzEjBUn/+Q/ccgscdljWFUmSVk1NxyPHardbAUOBXYF2wHMhhOdjjO+tsKEQRgOjATbYYIMmKFWSJBUDeyZJqt3nn8PIkfD663DXXQZJklScpgDr593uC3xWwzoPxhgXxBinA08BQ6pvKMZ4dYxxWIxxWI8ePZqsYEmS1LwZJkmq2aRJsMMO8OGH8K9/wX77ZV2RJGn1vAT0DyFsFEJoAxwG3FttnXuAnUIIrUII7YFtgLcLXKckSSoSHuYmaWVvvAF77JHO3vb44zB8eNYVSZJWU4yxIoRwMvAQUA5cH2OcEEIYk1t+ZYzx7RDCg8DrQCVwbYzxzeyqliRJzZlhkqQVPfcc7LMPtG8PTz8NgwdnXZEkaQ3FGMcCY6vNu7La7d8BvytkXZIkqTh5mJuk5R5+GHbbDbp3h3//2yBJkiRJkrQSwyRJyT/+Ad/8JvTvD888A/36ZV2RJEmSJKkZMkySBNdcA9/5Thobadw46NUr64okSZIkSc2UYZLU0v32tzB6NOy5ZzrMrUuXrCuSJEmSJDVjhklSSxUjnHkmnHFG6pV0991p0G1JkiRJkurg2dyklmjZMvj+99PhbWPGwGWXQXl51lVJkiRJkoqAPZOklmbJEjj88BQknXUWXHGFQZIkSZIkqcHsmSS1JAsWwEEHpbGRLr4YfvKTrCuSJEmSJBUZwySppZg1C/bdF154Aa67Dk44IeuKJEmSJElFyDBJagk+/zydre3dd+GOO1LvJEmSJEmSVoNhklTqPvgAdt8dpk6Ff/0Ldtst64okSZIkSUXMMEkqZW++CXvsAYsXw2OPwTbbZF2RJEmSJKnIeTY3qVQ9/zzsvDOEAE8/bZAkSZIkSWoUhklSKXrkkXQ4W9eu8MwzsPnmWVckSZIkSSoRhklSqbnzznTWtk02SUHSRhtlXZEkSZIkqYQYJkml5Lrr4NBD4RvfgHHjoHfvrCuSJEmSJJUYwySpVFx8MZx4Yhpw++GHYZ11sq5IkiRJklSCDJOkYhcjnHUW/Oxn8J3vwD33QIcOWVclSZIkSSpRrbIuQNIaqKyEH/wArroKTjoJLr8cysuzrkqSJEmSVMLsmSQVq8rKFCBddRWceSb8+c8GSZIkSZKkJmeYJBWjGOHkk+Haa+Gcc+DXv4YQsq5KkiRJktQCGCZJxSZGOO201BPp9NPh/POzrkiSJEmS1IIYJknFJMYUIF1yCZx6Klx0kT2SJEmSJEkFZZgkFYsY0yFtF18MP/wh/OEPBkmSJEmSpIIzTJKKxfnnw4UXwujRcOmlBkmSJEmSpEwYJknF4MIL4bzz4Ljj0lhJZf7pSpIkSZKy4TdSqbm7+GI4+2w48sh09jaDJEmSJElShvxWKjVnl14KP/sZHHoo3HADlJdnXZEkSZIkqYUzTJKaqz//GU45BQ46CP72N2jVKuuKJEmSJEkyTJKapWuvhR/8APbbD269FVq3zroiSZIkSZIAwySp+bnxxnTGtr32gjvugDZtsq5IkiRJkqT/MkySmpNbboHjj4ddd4W77oK11sq6IkmSJEmSVmCYJDUXd9wBxxwDI0bAPfdAu3ZZVyRJkiRJ0koMk6Tm4O674YgjYLvt4L77oH37rCuSJEmSJKlGhklS1u6/Hw49FIYNg7FjoWPHrCuSJEmSJKlWhklSlh56CA4+GIYMgQcegE6dsq5IkiRJkqQ6GSZJWXnsMTjwQBg8OIVKXbpkXZEkSZIkSfUyTJKy8OSTsN9+sOmm8Mgj0LVr1hVJkiRJktQghklSof3737DvvtCvX+qd1L171hVJkiRJktRghklSIb3wAuy9N/Tpk4Kknj2zrkiSJEmSpFVimCQVyvjxsOeeKUB6/HFYd92sK5IkSZIkaZUZJkmF8NprsPvusM46KUjq0yfriiRJkiRJWi2GSVJTe/NN2G036NAhBUkbbJB1RZIkSZIkrTbDJKkpvf027LortGkDTzwBG22UdUWSJEmSJK0RwySpqbz9NowaBSGkHkmbbpp1RZIkSZIkrTHDJKkpVAVJAOPGwcCBmZYjSZIkSVJjMUySGlv1IGnQoEzLkSRJkiSpMRkmSY3prbcMkiRJkiRJJc0wSWosb70Fu+ySxkgySJIkSZIklSjDJKkx5AdJTzxhkCRJkiRJKlmGSdKaMkiSJEmSJLUghknSmqgaI8kgSZIkSZLUQhgmSaurKkgqKzNIkiRJkiS1GK2yLkAqSvlB0rhxMHBg1hVJkiRJklQQ9kySVpVBkiRJkiSpBTNMklbFhAkGSZIkSZKkFs0wSWqoCRPSWdvKyw2SJEmSJEktlmGS1BD5QdITTxgkSZIkSZJaLMMkqT4GSZIkSZIk/VdBw6QQwl4hhHdDCBNDCGfWsd43QgjLQgiHFLI+aSVVYyQZJEmSJEmSBBQwTAohlAOXA3sDg4HDQwiDa1nvN8BDhapNqlFVkNSqlUGSJEmSJEk5heyZNByYGGP8IMa4BLgNOKCG9f4HuBP4soC1SSvKD5IcbFuSJEmSpP8qZJjUB5icd3tKbt5/hRD6AN8CrixgXdKKqgdJAwZkXZEkSZIkSc1GIcOkUMO8WO32H4EzYozL6t1YCKNDCC+HEF6eNm1aY9QnwZtvGiRJkiRJklSHVgV8rCnA+nm3+wKfVVtnGHBbCAGgO7BPCKEixnh39Y3FGK8GrgYYNmxY9VBKWnVvvpnO2ta6dRojySBJkiRJkqSVFDJMegnoH0LYCPgUOAw4In+FGONGVddDCDcA99cUJEmNziBJkiRJkqQGKViYFGOsCCGcTDpLWzlwfYxxQghhTG654yQpGwZJkiRJkiQ1WCF7JhFjHAuMrTavxhApxnhcIWpSC/fGG7DrrgZJkiRJkiQ1UCEH4Jaal5tvhu22M0iSJEmSJGkVGCap5Vm4EL73PTjqKNhqK3jhBYMkSZIkSZIayDBJLcvbb8M228C118JZZ6UeSX37Zl2VJEmSJElFo6BjJkmZuukm+P73oUMHePBB2HPPrCuSJEmSJKno2DNJpW/BAjjhBDj2WPjGN+DVVw2SJEmSJElaTYZJKm0TJsDw4XDDDfCLX8Cjj8J662VdlSRJkiRJRcvD3FS6brgBfvAD6NQJHn4Ydtst64okSZIkSSp69kxS6Zk/Px3SdvzxsO226bA2gyRJkiRJkhqFYZJKyxtvpHGR/vpXOO88eOQRWHfdrKuSJEmSJKlkeJibSkOMcP31cPLJ0KVLGhtpl12yrkqSJEmSpJJjzyQVv3nz4Oij4cQTYccd02FtBkmSJEmSJDUJwyQVt9deg2HD4NZb4Ze/hAcfhF69sq5KkiRJkqSS5WFuKk4xwjXXwI9+BF27wuOPw4gRWVclSZIkSVLJs2eSis/cuXDEEXDSSSlAevVVgyRJkiRJkgrEMEnF5T//gaFD4fbb4cIL4YEHoGfPrKuSJEmSJKnFMExScYgRrrgCtt0WFi2CcePg5z+HMt/CkiRJkiQVkt/E1fzNmQOHHgo//CHsums6rG2nnbKuSpIkSZKkFskwSc3b+PGw9dbwz3/Cb34D998P3btnXZUkSZIkSS2WYZKar9tvhx12gCVL4Mkn4fTTPaxNkqTVEELYK4TwbghhYgjhzDrW+0YIYVkI4ZBC1idJkoqL38zV/MQIF10E3/kODBuWBt3eYYesq5IkqSiFEMqBy4G9gcHA4SGEwbWs9xvgocJWKEmSio1hkpqXpUth9Og0uPbhh8Ojj3pYmyRJa2Y4MDHG+EGMcQlwG3BADev9D3An8GUhi5MkScXHMEnNx5w58M1vwrXXwtlnw9/+Bm3bZl2VJEnFrg8wOe/2lNy8/woh9AG+BVxZ14ZCCKNDCC+HEF6eNm1aoxcqSZKKg2GSmodPPoEdd4THH4frroMLLnB8JEmSGkeoYV6sdvuPwBkxxmV1bSjGeHWMcViMcViPHj0aqz5JklRkWmVdgMT48bDffrBgATzwAOy2W9YVSZJUSqYA6+fd7gt8Vm2dYcBtIQSA7sA+IYSKGOPdBalQkiQVFcMkZeu+++Cww9K4SM8+C5tvnnVFkiSVmpeA/iGEjYBPgcOAI/JXiDFuVHU9hHADcL9BkiRJqo3HEdXiiy/gG9+AW2/NupIS9qc/wYEHwuDB8MILBkmSJDWBGGMFcDLpLG1vA7fHGCeEEMaEEMZkW50kSSpG9kyqRc+eMGkSPPZYOqmYGtGyZfCTn8All8ABB8DNN0OHDllXJUlSyYoxjgXGVptX42DbMcbjClGTJEkqXvZMqkVZGey8M4wbl3UlJWbBAjj44BQknXoq3HmnQZIkSZIkSUXEMKkOo0al3kmTJ9e/rhrgiy9g5Mg0TtKf/gT/939QXp51VZIkSZIkaRUYJtVh5Mh0ae+kRjBhAmyzDbz1FtxzD5x8ctYVSZIkSZKk1WCYVIctt4SuXeGJJ7KupMg9+ihsvz0sXQpPPw3f/GbWFUmSJEmSpNVkmFSHsjIYMcKeSWvk+uth771hww3TGdu23jrriiRJkiRJ0howTKrHqFHw4Yfw8cdZV1JkKivh7LPhu9+FXXaBZ56B9dfPuipJkiRJkrSGDJPq4bhJq2HxYjjySLjwQvje9+D++6Fz56yrkiRJkiRJjcAwqR6bbw7duztuUoNNnw677w633QYXXQRXXQWtW2ddlSRJkiRJaiStsi6guXPcpFXw/vuwzz4weTLcfjt8+9tZVyRJkiRJkhqZPZMaYNSoNGbShx9mXUkz9swzsN12MHs2PP64QZIkSZIkSSXKMKkBHDepHrfdBrvuCt26wfPPw/bbZ12RJEmSJElqIoZJDTB4MPTo4bhJK4kxDbJ9+OGwzTbw3HOwySZZVyVJkiRJkpqQYVIDhJB6J40bl/ITkZ6I006Ds89OZ2575BHo2jXrqiRJkiRJUhMzTGqgUaPSuNIffJB1Jc1AZSV8//twySVw6qnw17/CWmtlXZUkSZIkSSoAw6QGctyknGXL4Lvfhauugp//HP7wh9R1S5IkSZIktQiGSQ00aBD06tXCx01auhSOPhpuuAHOPx9+9SuDJEmSJEmSWphWWRdQLKqPm9TiMpQlS+Cww+Cf/4Tf/AZOPz3riiRJkiRJUgbsmbQKRo2CTz+FiROzrqTAFi+Ggw5KQdIllxgkSZIkSZLUghkmrYIWOW7SggWw334wdmwaJ+lHP8q6IkmSJEmSlCHDpFUwYACsu24LGjdp3jzYe294/PE0TtLo0VlXJEmSJEmSMuaYSaugRY2bNHt2CpJeegluuQW+852sK5IkSZIkSc2APZNW0ahR8Pnn8N57WVfShGbMgF13hfHj4R//MEiSJEmSJEn/ZZi0ikp+3KSpU1Ni9tZbcM89cOCBWVckSZIkSZKaEcOkVbTpptCnT4mOm/TppyktmzQJ/vWvdJibJEmSJElSHsdMWkVV4yY9+miJjZv08cfp0LYvv4SHHoIdd8y6IkmSJEmS1AzZM2k1jBqVjgZ7552sK2kkkybBzjunsZIeecQgSZIkSZIk1cowaTWU1LhJ77yTgqQFC+Dxx2GbbbKuSJIkSZIkNWOGSath441h/fVLYNykN96AESNg2bKUjG21VdYVSZIkSZKkZs4waTVUjZs0blwaN6kovfJK2onWreHJJ2GLLbKuSJIkSZIkFQHDpNU0ahRMmwZvvZV1JavhhRdgl12gUyd46ikYODDriiRJkiRJUpEwTFpNRTtu0tNPw267QY8eKUjaeOOsK5IkSZIkSUXEMGk1bbQRbLhhkY2b9NhjsNde0LdvOrRtgw2yrkiSJEmSJBUZw6Q1MHJkymQqK7OupAHGjoV994VNNklFr7de1hVJkiRJkqQiZJi0BkaNgunTYcKErCupxz//CQcemAbZfuIJ6Nkz64okSZIkSVKRMkxaA0UxbtJtt8G3vw3DhsGjj0K3bllXJEmSJEmSiphh0hrYcMM0dlKzGzcpxnTGtpNPhiOPhB12gIcegi5dsq5MkiRJkiQVuVZZF1DsRo6Ee+5J4yaVZR3Nvf023HJLmj74ANZaC446Cv78Z2jfPuPiJEmSJElSKcg6/ih6o0bBzJnwxhsZFTB5Mvzud7DVVjB4MFx4IWy8MfzlLzB1Ktx4o0GSJEmSJElqNPZMWkP54yYNGVKgB50xA/7xj9QD6amn0rzhw+GPf4RDD4V11y1QIZIkSZIkqaUxTFpD668Pm2ySxk065ZQmfKAFC+C++1KA9OCDsHQpDBwI558Phx8Om27ahA8uSZIkSZKUGCY1gpEj4a67mmDcpKVL4ZFHUoB0990pUOrTJ6VWRxwBX/86hNCIDyhJkiRJklQ3w6RGMGoUXHcdvPZaGrpojVRWwrPPpgDp9tvTIW3rrJPOynbEEbDTTs1gpG9JkiRJktRSGSY1gvxxk1Y7THr99RQg3XorfPIJtGsHBxyQAqQ994Q2bRqpWkmSJEmSpNVnmNQI+vSB/v3TuEmnnVbPyl99BXPmwOzZ6TRwTzyRQqQ334Ty8hQcXXhhCpI6dixE+ZIkSZIkSQ1mmLSqlixJQVC16Re9ZvP+I7OpPHM2ZXNXXv7fadGilbe5ww5w+eXw7W9Djx4F2hFJkiRJkqRVZ5hUmzlz0iFm1cOghQtrXP3o3GW8uBWs0wW65E19+qx4O3/aYgvYcMOm3BNJkiRJkqRGY5hUm7XWgi+/TIHPeuvVHgblpqlfdWHjoV04/6L2/OSnnmFNkiRJkiSVJsOk2rRtCy+91ODVewHrD4QnxsFPftpkVUmSJEmSJGXKc8w3opEj4emnoaIi60okSZIkSZKahmFSIxo1CubOhf/8J+tKJEmSJEmSmoZhUiMaMSJdjhuXaRmSJEmSJElNxjCpEfXuDZttBk88kXUlkiRJkiRJTcMwqZE5bpIkSZIkSSplhkmNbNQomD8fxo/PuhJJkiRJkqTGZ5jUyBw3SZIkSZIklTLDpEbWsycMHuy4SZIkSZIkqTQZJjWBUaPgmWdg6dKsK5EkSZIkSWpchklNYORIWLAAXn4560okSZIkSZIal2FSE3DcJEmSJEmSVKoMk5pAjx6wxRaOmyRJkiRJkkqPYVITGTUK/v1vWLIk60okSZIkSZIaj2FSExk5EhYuhJdeyroSSZIkSZKkxmOY1EQcN0mSJEmSJJWigoZJIYS9QgjvhhAmhhDOrGH5kSGE13PTsyGEIYWsrzF16wZf+5rjJkmSJEmSpNJSsDAphFAOXA7sDQwGDg8hDK622ofAiBjj14BfAlcXqr6mMGoUPPssfPVV1pVIkiRJkiQ1jkL2TBoOTIwxfhBjXALcBhyQv0KM8dkY46zczeeBvgWsr9GNHAmLFsGLL2ZdiSRJkiRJUuMoZJjUB5icd3tKbl5tvgs80KQVNbGdd4YQHDdJkiRJkiSVjkKGSaGGebHGFUMYRQqTzqh1YyGMDiG8HEJ4edq0aY1UYuPq2hWGDHHcJEmSJEmSVDoKGSZNAdbPu90X+Kz6SiGErwHXAgfEGGfUtrEY49UxxmExxmE9evRo9GIby6hR8NxzsHhx1pVIkiRJkiStuUKGSS8B/UMIG4UQ2gCHAffmrxBC2AC4Czg6xvheAWtrMiNHpiDphReyrkSSJEmSJGnNFSxMijFWACcDDwFvA7fHGCeEEMaEEMbkVjsX6AZcEUJ4NYTwcqHqayqOmyRJkiRJkkpJq0I+WIxxLDC22rwr866fCJxYyJqaWpcusNVWadyk//3frKuRJEmSJElaM4U8zK3FGjUKnn/ecZMkSZIkSVLxM0wqgJEj4auv0kDckiRJkiRJxcwwqQB22gnKyhw3SZIkSZIkFT/DpAJYe23Yeus0bpIkSZIkSVIxM0wqkFGj4IUXYOHCrCuRJEmSJElafYZJBTJyJCxZ4rhJkiRJkiSpuBkmFciOO0J5ueMmSZIkSZKk4maYVCCdO8PQoY6bJEmSJEmSipthUgGNGgUvvggLFmRdiSRJkiRJ0uoxTCqgkSNh6VJ49tmsK5EkSZIkSVo9hkkF5LhJkiRJkiSp2BkmFVDHjvCNbzhukiRJkiRJKl6GSQU2ahS89BLMn591JZIkSZIkSavOMKnARo6Eigr497+zrkSSJLUUIYS9QgjvhhAmhhDOrGH5kSGE13PTsyGEIVnUKUmSioNhUoHtsAO0auW4SZIkqTBCCOXA5cDewGDg8BDC4GqrfQiMiDF+DfglcHVhq5QkScXEMKnAOnSA4cMdN0mSJBXMcGBijPGDGOMS4DbggPwVYozPxhhn5W4+D/QtcI2SJKmIGCZlYNQoePllmDcv60okSVIL0AeYnHd7Sm5ebb4LPNCkFUmSpKJmmJSBkSNh2TJ45pmsK5EkSS1AqGFerHHFEEaRwqQzalk+OoTwcgjh5WnTpjViiZIkqZgYJmVg++2hdWvHTZIkSQUxBVg/73Zf4LPqK4UQvgZcCxwQY5xR04ZijFfHGIfFGIf16NGjSYqVJEnNn2FSBtq3h222cdwkSZJUEC8B/UMIG4UQ2gCHAffmrxBC2AC4Czg6xvheBjVKkqQiYpiUkVGjYPx4mDs360okSVIpizFWACcDDwFvA7fHGCeEEMaEEMbkVjsX6AZcEUJ4NYTwckblSpKkImCYlJGRI6GyEp5+OutKJElSqYsxjo0xDogxbhJj/FVu3pUxxitz10+MMa4TY/x6bhqWbcWSJKk5M0zKyHbbQZs2HuomSZIkSZKKi2FSRtq1g912g6uuggkTsq5GkiRJkiSpYQyTMnT11dChAxx4IMyalXU1kiRJkiRJ9TNMylCfPnDXXfDxx3DEEbBsWdYVSZIkSZIk1c0wKWPbbw+XXw4PPghnn511NZIkSZIkSXVrlXUBgu99D155BX7zG9hqK/jOd7KuSJIkSZIkqWb2TGomLrkEdtwRjj8eXn0162okSZIkSZJqZpjUTLRpA//4B3Ttmgbknj4964okSZIkSZJWZpjUjPTqBXffDV98AYceCkuXZl2RJEmSJEnSigyTmplhw+Dqq+GJJ+BnP8u6GkmSJEmSpBU5AHczdMwx8J//wB//mAbkPvbYrCuSJEmSJElK7JnUTP3ud7DLLnDSSfDii1lXI0mSJEmSlBgmNVOtWsHf/w7rrgsHHZTGUZIkSZIkScqaYVIz1r17GpB71iw4+GBYsiTriiRJkiRJUktnmNTMDRkCf/kLPPss/M//ZF2NJEmSJElq6RyAuwgceii8+ir8+tew9dZpHCVJkiRJkqQs2DOpSPzyl7D33nDyyfDMM1lXI0mSJEmSWirDpCJRXg633AIbbZTGT5oyJeuKJEmSJElSS2SYVES6dIF77oFFi+Bb30qXkiRJkiRJhWSYVGQ22wz+9jd4+WUYMwZizLoiSZIkSZLUkhgmFaH994f/9//gppvgkkuyrkaSJEmSJLUkhklF6pxz4MAD4ac/hccey7oaSZIkSZLUUhgmFamystQzaeBAOPRQ+PDDrCuSJEmSJEktgWFSEevUKQ3IXVmZeiktWJB1RZIkSZIkqdQZJhW5TTeF226DN9+E4493QG5JkiRJktS0DJNKwJ57wq9/DXfcAb/5TdbVSJIkSZKkUmaYVCJ+9jM47DA46ywYOzbraiRJkiRJUqkyTCoRIcB118GQIXDEEfDee1lXJEmSJEmSSpFhUglp3x7uvhtat04Dcs+dm3VFkiRJkiSp1BgmlZgNN0xjJ733Hhx9dDrTmyRJkiRJUmMxTCpBI0fC//0f3HtvGkOpoiLriiRJkiRJUqkwTCpRJ58MJ5yQzu626aYpXPKwN0mSJEmStKYMk0pUCHDNNWkMpQ02gB//GNZfP531bfLkrKuTJEmSJEnFyjCphJWVwQEHwFNPwYsvwj77pB5KG22Uzvg2fnzWFUqSJEmSpGJjmNRCfOMbcOutMGkSnHIK3H8/DBuWxle67z4H6pYkSZIkSQ1jmNTCbLgh/P736VC3iy+GDz6A/feHwYPhqqtg0aKsK5QkSZIkSc2ZYVILtfba8JOfpJ5Kt94KnTrBmDFpfKVzz4WpU7OuUJIkSZIkNUeGSS1c69Zw2GFpTKUnn4Ttt4cLLkg9mE48Ed56K+sKJUmSJElSc2KYJCCd/W3nneGee+Cdd+D44+GWW2DzzdPA3Y89BjFmXaUkSZIkScqaYZJWMmAA/PnP8Mkn8MtfwiuvwG67wVZbwU03wZIlWVcoSZIkSZKyYpikWnXvDuecAx99BNddBxUVcOyxsNFGcNFFMGtW1hVKkiRJkqRCM0xSvdq2hRNOgDfegAceSGd++/nPYf314X/+Bx5+GObPz7pKSZIkSZJUCIZJarAQYK+94JFH4NVX4ZBD4KqrYM89YZ11YLvt4MwzU+A0d27W1UqSJEmSpKZgmKTVMmQI3HADzJgBDz4IP/sZlJXB73+fBuxeZx0YPjzNv/9+mD0764olSZIkSVJjaJV1ASpunTqlnkl77pluL1gAzz0HTz6ZpksvhYsvTr2avv51GDECRo6EnXaCrl2zrFySJEmSJK0OwyQ1qg4d0pnfdtst3V60CF54AcaNS+HSlVfCH/+YwqUtt0zh0ogRsPPO0KNHlpVLkiRJkqSGMExSk2rXLvVEGjky3f7qK3jxxeU9l669Fv70p7Rs882Xh0sjRkCvXllVLUmSJEmSamOYpIJaa610iNtOO8E558CSJfDyy8vDpZtugiuuSOsOGpRCpaFD0xnkBg9OYzFJkiRJkqTsGCYpU23awPbbp+nnP4eKCnjlleWHxd16azpjXJXevZcHS/mTh8hJkiRJklQYhklqVlq1SmeBGz4cTj8dKivhk0/grbdWnG68EebNW36/bt1qDpnWXTeNzyRJkiRJkhqHYZKatbIy6NcvTfvss3x+jPDpp/D22yuGTLffDrNmLV9v7bVTqLTZZiuGTOuvn7YtSZIkSZJWjWGSilII0Ldvmnbfffn8GOHLL1fuyfSvf8H11y9fr0OH5QHTgAGw8caw0UZp6tnT3kySJEmSJNXGMEklJYR0FrhevWDUqBWXzZixck+mRx9Ng37na98+9YTaaKMVQ6aqqXPngu2OJEmSJEnNjmGSWoxu3WDHHdOUb8EC+Ogj+PDDlaennlpxbCaArl1rD5o23DCdsU6SJEmSpFJlmKQWr0MH2HzzNFUXI8ycuXLI9MEH8OqrcM89sGTJ8vVDgD59VgyY+vVLA4H37p2m7t2hvLxQeydJkiRJUuMyTJLqEELq0dStGwwbtvLyykr47LMVQ6aq648/ngYJj3HF+5SVpXGZevVaHjDVNq29tuM3SZIkSZKaF8MkaQ2UlS0fCHynnVZe/tVXMGUKfPHF8mnq1BVvv/VWuly6dOX7r7VW/YFTr16p51Pbtk2/v5IkSZIkGSZJTWittWCTTdJUlxhh1qwVQ6bq04cfwnPPwbRpK/d2AujSBdZbLwVL+VP1eR07NsmuSpIkSZJaCMMkqRkIIQ3s3bUrDB5c97oVFSlQqgqZPv985enpp9Nl/nhOVTp1qj9wWm+9dNY6D7GTJEmSJFVnmCQVmVatloc+danq7fT552lcp/ywqer2Sy+ly4ULV75/u3bLH6dnzzRuVPfuy8eQqn57nXUcWFySJEmSWgLDJKlE5fd2qulMdVVihLlzaw6bqqb33oPp02HGjNQzqrbHW2ed2sOmmm5365YOBZQkSZIkFQ/DJKmFCyGdNW7ttWHQoLrXjRHmzUuh0owZywOmmm5/9hm88UaaV1PPpyodO6YQqmPHlacOHWqeX9fy9u3TwOiSJEmSpKZhmCSpwUJIYyl17gwbbdTw+y1eXHsANWNGOhxvwQKYPz9NU6Ysv141v7Ky4Y/XocPKQVP+vKrrq3q7TZtVf84kSZIkqdQYJklqcm3bQp8+aVodMaZAqipgqh401TTlL5s3L01ffJHmVy1buLDmM+PVplWr2sOmtm3TIXtt2qTLxp7atTPMkiRJktQ8GCZJavZCSGFKu3bQo0fjbTdGWLRoefiUHzTVdL22ZTNmpLDrq69qnpYta5x6W7deMcBqrMu2bT1znyRJkqSGM0yS1GKFkMZYat++aR9n2bIVw6UlS2oPnmqb8kOv6pdTp64ccC1Z0vD6ysrSc1DVqyr/sqZ5DVlWfZ3WrdPZ/kJIU1lZw683dN1WrdLjtW5d96VjakmSJElrxjBJkppYeXlhQqt8S5fWHDzV1LOq6pC/qpAr/zL/+ty5Na+Tf1nb2f6ak7Ky5UFXfcFT/mXr1imwqrqsPjXG/Px51R+vpsevaZ69zCRJktTUChomhRD2Ai4ByoFrY4wXVVsecsv3ARYCx8UYXylkjZJUClq3hi5d0lRIlZXLQ6jqQVOMaXmMDbu+qutWVKQQbcmS2i/rWlbb5fz5y29XVKw81TR/6dLCPu/5ystrD57KytLysrKap9qWNXT+7rvDmDHZ7bskSZIKo2BhUgihHLgc2B2YArwUQrg3xvhW3mp7A/1z0zbAn3OXkqQiUFaWxmBq2zbrSrK3bFnN4VNdAVT1ZdXnrc46+ZeVlTVPy5bVv6yiov77bLll1s+6JOn/s3ffYVJV9x/H34elCAIqAjZEFKlqiEKwRIXYO2o09ho1xpBETaImJmo0JppfNMaosffeK2psWKNRjA0VxQqiAtJBhGXP748zK8MyuyywO3dn9v16nvvszJ07d753ZnfnzGfOOVeSiqGYPZMGA2NjjB8ChBBuBYYB+WHSMOD6GGMEXgwhrBxCWCPG+HkR65QkablVVKSlTZusK5EkSZIaVjGnIV0LGJd3fXxu3dJuI0mSJEmSpIwUM0wqNCVoXIZt0oYhHBNCeCWE8MqkSZOWuzhJkiRJkiQtWTHDpPHA2nnXuwETlmEbAGKMl8cYB8UYB3Xp0qVBC5UkSZIkSVJhxQyTXgZ6hRDWDSG0BvYH7q+xzf3AoSHZDJjufEmSJEmSJElNR9Em4I4xVoYQhgOPAhXA1THG0SGEY3O3XwqMAHYBxgJzgCOKVZ8kSZIkSZKWrJhncyPGOIIUGOWvuzTvcgR+VsyaJEmSJEmSVH/FHOYmSZIkSZKkEmeYJEmSJEmSpHozTJIkSZIkSVK9GSZJkiRJkiSp3gyTJEmSJEmSVG+GSZIkSZIkSao3wyRJkiRJkiTVm2GSJEmSJEmS6s0wSZIkSZIkSfVmmCRJkiRJkqR6M0ySJEkqcyGEnUIIY0IIY0MIpxS4PYQQLszd/kYIYZMs6pQkSaXBMEmSJKmMhRAqgIuBnYH+wAEhhP41NtsZ6JVbjgH+VdQiJUlSSTFMkiRJKm+DgbExxg9jjPOAW4FhNbYZBlwfkxeBlUMIaxS7UEmSVBoMkyRJksrbWsC4vOvjc+uWdhtJkiQAWmZdQEMYNWrU5BDCJ420+87A5Ebad1PU3I4XPObmoLkdL3jMzUFzO951si6ghIUC6+IybEMI4RjSMDiAWSGEMctZW22a2+83eMzNQXM7XvCYm4PmdrzQ/I65YBusLMKkGGOXxtp3COGVGOOgxtp/U9Pcjhc85uaguR0veMzNQXM7Xi2X8cDaede7AROWYRtijJcDlzd0gTU1x99vj7n8NbfjBY+5OWhuxwvN85gLcZibJElSeXsZ6BVCWDeE0BrYH7i/xjb3A4fmzuq2GTA9xvh5sQuVJEmloSx6JkmSJKmwGGNlCGE48ChQAVwdYxwdQjg2d/ulwAhgF2AsMAc4Iqt6JUlS02eYtGSN3pW7iWluxwsec3PQ3I4XPObmoLkdr5ZDjHEEKTDKX3dp3uUI/KzYddWhOf5+e8zlr7kdL3jMzUFzO15onse8mJDaDpIkSZIkSdKSOWeSJEmSJEmS6s0wCQgh7BRCGBNCGBtCOKXA7SGEcGHu9jdCCJtkUWdDCSGsHUJ4KoTwTghhdAjhlwW2GRpCmB5CeC23nJZFrQ0phPBxCOHN3PG8UuD2cnud++S9fq+FEGaEEI6vsU1Jv84hhKtDCBNDCG/lresUQngshPB+7ucqtdy3zr/7pqqWY/6/EMK7ud/be0IIK9dy3zr/BpqqWo75jBDCZ3m/u7vUct+Se51rOd7b8o714xDCa7XctyRfYzVftsFsg+VuL5vXuTm0v8A2WN66sm2DNbf2F9gGW2oxxma9kCai/ABYD2gNvA70r7HNLsDDQAA2A17Kuu7lPOY1gE1ylzsA7xU45qHAg1nX2sDH/THQuY7by+p1rnFsFcAXwDrl9DoDWwObAG/lrfsrcEru8inAubU8H3X+3TfVpZZj3gFombt8bqFjzt1W599AU11qOeYzgF8v4X4l+ToXOt4at58HnFZOr7FL81xsg9kGK9fXOe+4yrL9lTsG22CxvNtgza39Vdsx17jdNljeYs8kGAyMjTF+GGOcB9wKDKuxzTDg+pi8CKwcQlij2IU2lBjj5zHGV3OXZwLvAGtlW1WTUFavcw3bAh/EGD/JupCGFGN8BphSY/Uw4Lrc5euAPQvctT5/901SoWOOMf47xliZu/oi0K3ohTWiWl7n+ijJ17mu4w0hBOBHwC1FLUpqHLbBbINVK6vXOU9Ztr/ANljeurJtgzW39hfYBltahknpDXxc3vXxLP6mXp9tSlIIoQewMfBSgZs3DyG8HkJ4OISwQXEraxQR+HcIYVQI4ZgCt5ft6wzsT+3/+MrtdV4txvg5pEY70LXANuX8Wh9J+na3kCX9DZSa4blu5VfX0pW+HF/nrYAvY4zv13J7ub3GKm+2wWyDVSvX17k5tb/ANlhzaYM1x/YX2AZbjGFS6k5bU81T3NVnm5ITQmgP3AUcH2OcUePmV0ldcgcA/wTuLXJ5jeH7McZNgJ2Bn4UQtq5xe7m+zq2BPYA7Ctxcjq9zfZTra30qUAncVMsmS/obKCX/AnoC3wU+J3U7rqkcX+cDqPsbsXJ6jVX+bIPZBqtWdq+z7a9ald1rDc2qDdZc219gG2wxhkkpKV0773o3YMIybFNSQgitSI2Ym2KMd9e8PcY4I8Y4K3d5BNAqhNC5yGU2qBjjhNzPicA9pC6Y+crudc7ZGXg1xvhlzRvK8XUGvqzuGp/7ObHANmX3WocQDgN2Aw6KMRZ8w67H30DJiDF+GWNcEGOsAq6g8LGU1escQmgJ7A3cVts25fQaq1mwDWYbrFrZvc40v/YX2AYr+zZYc2x/gW2w2hgmwctArxDCurlvEPYH7q+xzf3AoSHZDJhe3YWzFOXGe14FvBNjPL+WbVbPbUcIYTDpd+Wr4lXZsEIIK4YQOlRfJk2W91aNzcrqdc5Ta4pebq9zzv3AYbnLhwH3FdimPn/3JSOEsBNwMrBHjHFOLdvU52+gZNSYS2MvCh9LWb3OwHbAuzHG8YVuLLfXWM2CbbDC25TVe3MzboM1t/YX2AYr+zZYM21/gW2wwmqbmbs5LaQzSLxHmnX+1Ny6Y4Fjc5cDcHHu9jeBQVnXvJzHuyWpq+EbwGu5ZZcaxzwcGE2aff9FYIus617OY14vdyyv546r7F/n3DG1IzVOVspbVzavM6mR9jkwn/QtyI+BVYEngPdzPzvltl0TGJF338X+7kthqeWYx5LGplf/PV9a85hr+xsohaWWY74h93f6BqmBska5vM6Fjje3/trqv928bcviNXZpvkuhv9Fyfm/GNlizaINR5u2v3DHYBivzNlgtx1u27a/ajjm3/lpsgy22hNzBS5IkSZIkSUvkMDdJkiRJkiTVm2GSJEmSJEmS6s0wSZIkSZIkSfVmmCRJkiRJkqR6M0ySJEmSJElSvRkmSZIkSZIkqd4MkyRJkiRJklRvhkmSJEmSJEmqN8MkSZIkSZIk1ZthkiRJkiRJkurNMEmSJEmSJEn1ZpgkSZIkSZKkejNMkiRJkiRJUr0ZJkmSJEmSJKneDJMkSZIkSZJUb4ZJkiRJkiRJqjfDJEmSJEmSJNWbYZIkSZIkSZLqzTBJkiRJkiRJ9WaYJEmSJEmSpHozTJIkSZIkSVK9GSZJkiRJkiSp3gyTJEmSJEmSVG+GSZIkSZIkSao3wyRJkiRJkiTVm2GSJEmSJEmS6s0wSZIkSZIkSfVmmCRJkiRJkqR6M0ySJEmSJElSvRkmSZIkSZIkqd4MkyRJkiRJklRvhkmSJEmSJEmqN8MkSZIkSZIk1ZthkiRJkiRJkurNMEmSJEmSJEn1ZpgkSZIkSZKkemtZ142jRo3q2rJlyyuBDTF4WlZVwFuVlZVHDRw4cGLWxUiSpOKyPVVSbLdJklQPdYZJLVu2vHL11Vfv16VLl6ktWrSIxSqqnFRVVYVJkyb1/+KLL64E9si6HkmSVFy2p0qH7TZJkupnSd+ObdilS5cZNnyWXYsWLWKXLl2mk76NlCRJzY/tqRJhu02SpPpZUpjUwobP8ss9h3ZrlySpebI9VUJst0mStGRN+o3yiy++qOjbt2//vn379u/cufOArl27fqf6+ty5c0Nd933mmWfaHX744WsvzeOttdZaG33++ed1Dv2TJEkqJcVuT0mSpPLXpIOT1VdffcG77777NsCJJ564Zvv27ReceeaZX1bfPn/+fFq1alXwvltvvfWcrbfeek6RSpUkSWqSyrU9VVfdkiSpcTXpnkmF/PCHP+xx1FFHddt00017H3fccd2eeuqpdhtvvHHffv369d944437vv76620AHnzwwQ4/+MEP1ofUcNp33317DB48uE+3bt02+tOf/tR1SY9zxhlnrNarV68NevXqtcGZZ57ZFWDGjBkthg4dun6fPn369+rVa4MrrrhiFYDjjjturZ49e27Qu3fv/sccc0y3xjx+SZKk5dWY7amDDjqo+4Ybbthv/fXX3+CEE05Ys3r9008/3W7jjTfu26dPn/4bbbRRv6lTp7aorKzkmGOO6da7d+/+vXv37n/22Wd3hUV7iz/zzDPtBg8e3Ke6hgMOOGCd73//+7323nvvdceMGdN64MCBffr379+vf//+/R577LEVqx/v97///Wq9e/fu36dPn/7HHXfcWqNHj27Tv3//ftW3v/nmm2022GCDfkiSpKVW755JRx7J2m+9RbuGfPANN2TO1Vczbmnv98EHH6zw/PPPv9eyZUumTJnS4r///e+7rVq14t577+1w0kkndXv00Uc/qHmfsWPHrvDCCy+MmTZtWkW/fv02/M1vfjOpTZs2BecvePbZZ9vdfPPNq44aNeqdGCMDBw7st+222858//3326y++urzR44cORbgq6++qvjyyy8rRowYscqHH374VosWLZg8eXLF0j8TkiSpOWgO7anzzz//s9VWW21BZWUlW2yxRZ+XXnqp7YABA+YedNBBPW+66aYPhgwZMmfKlCkt2rdvX3Xeeed1+eSTT9qMHj367VatWvHll18usR31xhtvtHvppZfebd++fZw5c2aLZ5999r127drFN998s80BBxyw3ltvvfXO7bff3vGhhx5aZdSoUe926NCh6ssvv6xYbbXVFnTo0GHBCy+80HaLLbb4+rLLLut84IEHfrW0z5skSWriw9xqs/fee09t2TKVPmXKlIr99ttv3Y8//niFEEKcP39+wbH/O+yww7S2bdvGtm3bVnbq1Gn++PHjW/bs2XN+oW1HjhzZfpdddpnWsWPHKoBdd9116lNPPdVhjz32mH7qqaeu/dOf/nStYcOGTd9pp51mzZ8/nzZt2lTtv//+6+y6667T99tvv+mNduCSJEkNpLHaU9ddd12na6+9tnNlZWWYNGlSq9dff32FEAJdu3adP2TIkDkAnTp1qgJ48sknOx577LGTqoerrbbaaguWVPdOO+00rX379hFg3rx54cc//vE6b7/9dtsWLVrwySeftAF47LHHOh588MGTO3ToUJW/38MPP3zyFVdc0Xnw4MHj7rvvvlVefvnld5bt2ZMkqXmrd5i0LN94NZb27dtXVV8++eST1xoyZMjMxx577IMxY8a03mabbfoUuk/+t2YVFRVUVlbWOuFkjIVPuPKd73znm1dfffXtu+66a6VTTz11rccff3zG3/72t89fe+21d+6///6Ot9566yr/+te/ur744ovvLcfhSZKkMlXu7al333239UUXXbTaqFGj3unSpcuCH/7whz3mzp3bIsZICGGxBlZt6ysqKmJVVSrv66+/XmRahhVXXPHbus8+++zVunbtOv+uu+76qKqqirZt2w7M2+9i9R922GFTzz333DVvvfXWmRtttNGc1VdffYnhlSRJWlzJzZlU04wZMyq6des2D+Cyyy7r3BD73GabbWaNGDFi5ZkzZ7aYMWNGixEjRqzygx/8YObHH3/cqkOHDlXHHXfclOOPP/7L1157rd306dNb5L7Nm37ppZeOe+eddxq067okSVJja6j21NSpUyvatm1b1alTpwXjxo1rOXLkyJUABgwYMPfLL79s/fTTT7fLbddi/vz5bLfddjMuvfTSLvPnp85N1cPcunXrNu/5559vB3D77bevUtvjTZ8+vWKNNdaYX1FRwSWXXLLqggUpG9ppp51m3HDDDZ1nzpzZIn+/7dq1i0OGDJl+4okndj/88MMnL+txSpLU3JV8mHTyySd/ccYZZ3TbZJNN+lY3IJbXlltuOefAAw/8apNNNuk3cODAfocccsik73//+1+PGjWq7Xe/+91+ffv27X/uueeucdppp30+bdq0ip122qlX7969+2+11VZ9/vSnPzWZbxwlSZLqo6HaU5tvvvnXG2644ZxevXptcMghh/QYOHDgLIAVVlgh3nTTTR/84he/6N6nT5/+Q4cO7T1nzpwWJ5xwwqRu3brN69u37wZ9+vTpf9VVV3UCOO200yacdNJJ3QcOHNinoqKicJdx4Pjjj594yy23rDpgwIC+77333gpt27atAthnn31m7LzzztOq221nnXXW6tX3OfTQQ6cA7L333jOW+UAlSWrmQm1DugBef/31jwcMGOC3Ng3g9ddf7zxgwIAeWdchSZKKy/ZU03LaaaetNn369Ip//OMfE2rbxnabJEl1K8kJuCVJkqSltf322/f85JNP2jz99NPObylJ0nIwTJIkSVKz8Nhjj32QdQ2SJJWDkp8zSZIkSZIkScVjmCRJkiRJkqR6M0ySJEmSJElSvRkmSZIkSZIkqd6adJg0ePDgPnfddVfH/HVnnnlm14MPPrh7Xfd55pln2gEMGTJk/cmTJ1fU3ObEE09c87TTTlutvuslSZJKVbHbU5Ikqfw16TBp3333/eqWW27plL/urrvu6nTwwQdPqc/9n3766bGdO3de0DjVSZIkNX3l2p6qqqpiwYImV5YkSc1Ckw6TDjnkkKlPPPHESl9//XUAGDNmTOuJEye22mGHHWYddNBB3TfccMN+66+//gYnnHDCmoXuv9Zaa230+eeftwQ4+eSTV+/Ro8eGW2yxRe/333+/zZIe+4UXXmg7YMCAvr179+6//fbb95w0aVIFwJ/+9KeuPXv23KB37979d9ttt/UAHnroofZ9+/bt37dv3/79+vXrP3Xq1Cb9vEqSpOaj2O2pm2++eaXvfOc7ffv169d/iy226D1u3LiWANOnT2+xzz779Ojdu3f/3r1797/22mtXBrjzzjs79u/fv1+fPn36b7755r1h8V5PvXr12mDMmDGtx4wZ03q99dbb4OCDD+6+wQYb9P/ggw9a13YMTz/9dLuNN964b58+ffpvtNFG/aZOndpi4MCBfV544YW21dtssskmfV966aVvr0uSpPppWe8tjzxybd56q12DPvqGG87h6qvH1Xbz6quvvmDAgAGz77rrrpUOPvjgadddd12nPfbYY2qLFi04//zzP1tttdUWVFZWssUWW/R56aWX2m666aZfF9rPs88+2+6ee+7p9Oabb749f/58vvvd7/bfeOON59RV2uGHH77u3//+90933XXXWccff/yaJ5988ppXX331uAsvvHD1Tz755M22bdvG6i7f55133uoXXnjhJzvssMPs6dOnt2jXrl3V8j0xkiSpLDWD9tT2228/a//99383t//OZ5555upXXHHF+FNOOWWNjh07LnjvvffeBpg0aVLFhAkTWg4fPrzHyJEj3+3bt++8L7/8crHhdDV9/PHHK1xxxRUf33jjjZ8CBY9hwIABcw866KCeN9100wdDhgyZM2XKlBbt27evOvzwwydfeeWVnbfYYotxb7zxRpt58+aF2o5XkiTVrsn3oPnRj3405bbbblsF4O677+50yCGHTAG47rrrOvXv379f//79+7///vsrvP766yvUto+nnnqq/S677DKtQ4cOVZ06daraYYcdptX1mF999VXFzJkzK3bddddZAEcfffRXL774YnuAPn36fL3XXnute8kll3Rq1apVBNhss81m/frXv177T3/6U9fJkydXtGrVqoGOXpIkafkVsz310Ucftd5qq6169e7du/+FF164+rvvvtsW4Jlnnul4wgknTKzerkuXLgtGjhy54uDBg2f27dt3HsBqq622xHFra6yxxrxtt912dvX1QsfwxhtvrNC1a9f5Q4YMmQPQqVOnqlatWnH44YdPffzxx1f65ptvwqWXXtr5wAMPnFyvJ1CSJC2i/j2T6vjGqzEddNBB037/+9+v/dxzz7WbO3duiy233HLOu+++2/qiiy5abdSoUe906dJlwQ9/+MMec+fOrTMYCyE0SD1PPfXU+w8//HCHe++9d+W//vWva77//vtv/fnPf/5izz33nH7fffettMUWW/R75JFH3tt4443nNsgDSpKk8tEM2lPDhw/v/stf/vKLgw46aPqDDz7Y4cwzz1wTIMa42P0LrQNo2bJlrKpa2NH7m2+++Xaj/B7gtR1Dbr+x5n47dOhQtdVWW824+eabV77//vs7jRo16u0lHpAkSVpMk++ZtNJKK1VtttlmM4866qgee++99xSAqVOnVrRt27aqU6dOC8aNG9dy5MiRK9W1j2222WbWQw89tPKsWbPC1KlTWzz22GMr17X9qquuuqBjx44LHnnkkfYAV1111aqbb775rAULFvDBBx+03n333Wdecskl42fOnFkxffr0itGjR7cZPHjw12efffYXG2200ey33nqr1m/1JEmSiq2Y7amZM2dWdO/efT7Atddeu2r1+qFDh844//zzu1ZfnzRpUsUPfvCD2S+99FKHd999tzVA9TC3Hj16fPPaa6+tCPDcc8+1++yzzwrOz1TbMQwYMGDul19+2frpp59ul9uuxfz58wE49thjJ5988slrDxgwYHZ9ekJJkqTFNfkwCWD//fefMmbMmLbVXbI333zzrzfccMM5vXr12uCQQw7pMXDgwFl13X/LLbecs9dee03ZcMMNN9htt916Dh48uM7tAa655pqPTj755G69e/fu/8Ybb7Q955xzJlRWVoYDDzxw3d69e/ffcMMN+//kJz/5snPnzgv++te/du3Vq9cGffr06d+2bduqffbZZ3pDHbsaVgghhhDWL/Jj/imEMDmE8EUj7HtkCOGoht5vYwshbBVCGNMI+13i6xtCaBtCeCCEMD2EcEcI4aAQwr+XZh+NJYTQI/f49e81uuR9ZnY8TUEI4dIQwh+yrkNqCorVnjr11FMnHHDAAT0HDhzYZ9VVV62sXv+Xv/zl82nTplVUt5lGjBjRYc0116y88MILP95rr73W79OnT/+99tprPYBDDz106tSpUyv69u3b/6KLLuqyzjrrFOzxXdsxrLDCCvGmm2764Be/+EX3Pn369B86dGjvOXPmtADYaqut5qy44ooLjjjiCIe4lYAQwrUhhD8V+TG/H0J4P4QwK4SwZwPv+/AQwnMNuc9iCSGMDiEMbeB91uv1DSH8NITwZe41WTX3c72l2Udjaej2eNbHk7XG+pyghhdiXKwH8Ldef/31jwcMGOAbbQN4/fXXOw8YMKBH1nU0pBDCx8BRMcbHs66lvnJd3nvFGMcW6fHWBt4D1okxTlzS9vXcZ4wxhtzlkcCNMcYrG2LfjaVYz3t9HieEcAjwc2CLGGNlgdu/3UcI4VpgfIzx941Vc43H7gF8BLSKMVbmXt8zYowjC2x7LTAyxnjtEvbZYM99zfqWd38NLYRwOOl/0paNtO+hMcbDG3rfKn+2p5qWjz/+uNXQoUP7fPDBB29VVBSe77vY7bbc//sBwOoxxm+K9biloNjvxbnHfAK4P8b4jwba37Xk3rMb872qIRXrea/P44QQWgEzgM1ijK/XtY9c2HVjjLFboxRcuL6Ruce8sq72Qq62M2KMQ5ewv2tpwOe+qX9eaMzPCfmfm9TwSqJnklTC1gG+WpYgqSF7p2gR6wDvFSMMKafXsJyORVLzddFFF6262Wab9TvttNM+qy1IKrZcUL8VEIE9ivzY/m8vbB1g9LLc0ee0UawGrMAyviZLo5xev5D4eV+Nxl8uNbgQQpsQwgUhhAm55YIQQpvcbZ1DCA+GEKaFEKaEEJ6t/icXQjg5hPBZCGFmCGFMCGHbAvveLITwRQihIm/dXiGEN3KXB4cQ/pPb/+chhItCCK1rqXORLqk1ux2HEPqGEB7L1TkmhPCjvNt2CSG8nav1sxDCrwvsfzvgMWDNXFfca3Pr98h1E56Wq6Ff3n0+zj0PbwCzl/YNLYRwZAjhnRDC1BDCoyGEdfJuiyGE40Lqtj0zhHBWCKFn7vmaEUK4vfq5CiEMDSGMDyGcFEKYmHsu98wd93u55+R3efuu9XkPITyT2+z13POwX/X+c7fvl1tfvXyT+wal+nfpbyGET0Pq2nxpCKFt3uP+Jvd4E0IIR9bj+fkjcBpQ/Zg/rvm65217DHAQcFJu2wdy69cMIdwVQpgUQvgohPCLvPucEUK4M4RwYwhhBnB4CGGlEMJVuTo/C2nYY0Vu+4rc8U0OIXwI7Fq/V3qxWtcPITwd0tC9ySGE22pssl3udZ8aQrg4hFDds61FCOH3IYRPcq/z9SGElXK3VQ+5+3EI4VPgSaD6tZyWe042r6OmRY4thPCzkDeEL/e7vl2N5+7GvOubhRBeyP1OvR7yutXnXrMPc7/HH4U0VLEfcCmwea62abltF+kqHkI4OoQwNvc7fH8IYc2822II4dhCz5Wk8jB8+PCvvvjiizeOPPLIqVnXkudQ4EXgWuCw/BtCCGuHEO7Oved8FUK4KO+2o0N6z58ZUptkk9z6RYY35/8fDAvf308Oafj9NSGEVUJqm03K/e97MITQLe/+nUII1+Tea6eGEO7NrX8rhLB73natcv/zv1vzAHN17pZ3vWVu2+qa7wipfTc9hPBMCGGDQk9UKPCenX+8oY52Q6ijDVpjfx8A6wEP5N5P2oT03n9/7n5jQwhH522/2Ht/odprE+pub14bQrgkhPBwrpbnQwirh9S2nhpCeDeEsHHe9h+H1DZ6I4QwO6T2x2q5+88MITweQlglb/uCz3uovQ307Xt37nmsbrvNzr0OPXK37RZCeC23zQshhO/kPebGIYRXc/XcRgqJ6np+egPVQ56mhRCezK1fbBh/CGFF4GEWtr1n5V67FiGEU0IIH4T0d3R7CKFT7j6F2jtLalNvn3vup4f0N7nUbYWQ/D2k9tf03Gu2Yd4mq4QQHso9Ty+FEHrm3XeLEMLLufu9HELYIu+2kSGEs0MIzwNzgBtIYfVFuefjIupQ89hCal8elbutZlut+rmrbtvV1eYt2FYNS/ickNumX+64poX0GWqPvNuuDam9VvC5UuMyTFJjOBXYDPguqcv2YKC6m+avgPFAF9K3DL8DYgihDzAc+F6MsQOwI/BxzR3HGF8EZgPb5K0+ELg5d3kBcALQGdgc2BY4bmkPIPdm9Fhuv12BA4BLwsLGzVXAT3K1bkjujadGrY8DOwMTYoztY4yH594QbwGOzz0HI0iNlfzA6wBSqLByod4ztXXVDGlM/++AvXP7fjb3WPl2AgaSXp+TgMtJjYW1c8dxQN62q5Pe4NciBTBXAAfn7r8VcFrIjVWnjuc9xrh1bpsBuedhkaAjxnhbbn17YE3gw7y6zwV6k36X1s+rhRDCTsCvge2BXsB2LEGM8XTgz0D1Y15Vx7aXAzcBf81tu3tIjc4HgNdztWwLHB9C2DHvrsOAO4GVc/e/DqjM1b8xsANQHWIeDeyWWz8I2KdGDUMLDXHL3XZ43hC3s4B/A6sA3YB/1th8N+B7pL/HH5H+viA1eA8HfkBqOLcHajYyhgD9cvepfi1Xzj0n/ylUW32OrS4hhLWAh4A/AZ1Ir/NdIYQuub/NC4Gdc39/WwCvxRjfAY4F/pOrbeUC+90G+AvpOVgD+AS4tcZmBZ+rGOO1DnGT1EgOJb1f3ATsGEJYDVIoDzxI+l/Vg/S+c2vutn2BM3L37Ujq0fRVPR9vddL/1nWAY0ifB67JXe8OfM2i7wU3AO2ADUhtor/n1l9PahdU2wX4PMb4WoHHvIVF2xg7ApNjjK/mrj9Mei/vCrxKei6WRa3tBmppg9bcQYyxJ/ApsHvu/eSbXP3jSe2UfYA/h0W/9Kz53l9zn/nv2d+qR3sT0nvR70ltrG+A/5Ceo865xzy/xm5/SGob9QZ2Jz23v8tt3wL4Rd62BZ/3Qm2gAse0cl777R+kdudnIQWEVwM/AVYFLgPuDymUaw3cS/qd6gTckau3VjHG90i/e5DaH9vUse1sFm17t48xTsgd856kNs2awFTg4hp3/7a9U1ebOoTQGbiLha/JB8D382qotb0QYxyZN8RtB1K7qjfp92Y/Fv0bPgD4I6ltNxY4O/f4nUhtpAtJz+/5wEMhhFXz7nsI6W+7A6md9ywwPPd8DC/45NXj2OqhrjZvwbbqkj4nhDTE8YHcfbuSpqq4KffZsVrB5yq3f78UbERLCpOqqqqqfAGWU+45rFrihuXjIODMGOPEGOMk0h/3Ibnb5pM+xK0TY5wfY3w2pom7FgBtgP4hhFYxxo9jjB/Usv9vGyQhhA6kxsstADHGUTHGF2OMlTHGj0lvYEOW4Rh2Az6OMV6T29erpH+u1R+I5+dq7RhjnJrXGFqS/YCHYoyPxRjnA38D2pI+EFe7MMY4Lsb49VLW/BPgLzHGd3Ih1J+B7+Z/kwKcG2OcEWMcDbwF/DvG+GGMcTqpQbFx3rbzgbNzdd5KelP5R4xxZu7+o4HvQMM877mg5mbSnAKXhRACKZA4IcY4JcY4M3dM++fu8iPgmhjjW7nGwxlL83jL6HtAlxjjmTHGeTHGD0kh2/552/wnxnhvjLGK1MDfGTg+xjg7N9zx7zWO4YLc6z2FFHQsi/mkDwFrxhjnxhhr9rQ6J8Y4Lcb4KfAUqZEN6W/1/NzvwCzgt8D+YdEecWfkal/a38flObaDgRExxhExxqoY42PAK6S/dUj/TzcMIbSNMX6e+32sj4OAq2OMr+Y+HPyW1JOpR942tT1X0vKwPVVCitluCyFsSfr/fXuMcRTpw9uBuZsHkz74/ib3fzj///tRpA/6L8dkbIzxk3o+bBVweozxmxjj1zHGr2KMd8UY5+Tea88m9x4eQliD9D52bK69Mz/G+HRuPzcCu4QQOuauH0IKCQq5GdgjhNAudz3/i0BijFfn2hffkN7PB4RcT9n6qke7obY26JL2uzawJXBy7jV4DbiShW1byHvvX8r3yyW1NwHuybWz5gL3AHNjjNfHGBcAt7Fo2w3gnzHGL2OMn5FChJdijP/LPbf35G/fQM/7fqTX84e5NuPRwGUxxpdijAtijNeRQrDNcksrUvtgfozxTuDlpXm8ZfQT4NQY4/i8Y92njvZOXW3qXYC3Y4x35o73AmBZTrIznxT29AVC7rE+z7v97hjjf3OPfxML2yO7Au/HGG/I/c7cArxLCg6rXRtjHJ27ff5S1LTMx5YLwetq8y6prVqbzUhfdp6Ta3s/SQrZ88Pp2p4rNbIlhUlvTZo0aSUbQMuuqqoqTJo0aSXSB/fmYk3St2jVPsmtA/g/UmL875CGqZwCENOEa8eT/rlPDCHcGvKGn9RwM7B3SEPn9gZerW5AhRB6h9SF+YuQuhr/mRSCLK11gE1z3SmnhTRk5iDSt3mQvkXZBfgk12Wz1uE+NSzy3OQCh3Gkb86qjVuGeqtr/kdevVNI3W7z9/1l3uWvC1xvn3f9q1xDpfq2QvdvDw32vJ9NelOt/sasC+mb0FF5x/RIbj2k5zL/uapvI3p5rEPqOp3/e/E70jec1cbV2L4V8Hne9peRvlmBhjuGk0iv9X9z3X9rDvnLbwjMYeHrXOhvtSW1H8/SWJ5jWwfYt8bzvCWwRi443I/UC+nzXLfmvktRU/7f3yzSt4D5fyO1PVfS8rA9VSIyaLcdRvpip3qC9ptZONRtbeCTWHiOv7VJwdOymJQLJgAIIbQLIVwW0pDnGaQhzSvnekatDUyJMS42LDCmHh/PAz8MIaxM+iBZsEdRrp33DrB7LlDag1yYFNKw6HNCGoI0g4U905e2HbGkdkPBNmg9rEl6DmbmrfuEhmu71dXehKVru9V7+4Z43kMaYncRsFdMXyBXH9OvahzT2qTncU3gsxohXrHab/fk1fMO6YvsutpvtbWpF2nf5I5lqV//XChyEamH1JchhMvzglmof9sNGu73cXmObUlt3iW1VeusKfeZqVrN47XtlpE652OprKw86osvvrjyiy++2BCHxC2rKuCtysrKkjt9+3KYwKITF3bPrSP3Rvwr0pvMBsBTIYSXY4xPxBhvBm7O/SO9jNRV+ZCaO48xvh1C+ITUaFnkmy3gX8D/gANijDNDCMdT+/Ca2aRGR7X8N+5xwNMxxu0L3THG+DIwLNf1cjhwO+mNckkmABtVX8l9i7Y28Fn+7uuxn0LGkXoSLWvX8OWxNM/7YkII+5O+Yfhe3jcok0mNng1y367V9DmLPufdl6XwJaj5WowDPoox9qrnfcaRvo3rXMuHgQY5hhjjF6RvAqu/5X48hPBMXPJZMar/VvMfv5LU8KyeLyP/eJbmd3NJx7akv78bYoxHU0CM8VHg0ZDmwfgTqXdY9eS1dVnkeHPDC1Zl0b8/qcHZniopRWu35f6H/QioCGn+Iki9tFcOIQwg/S/sHkJoWeA9ZBxQ27wgc1j8/+v4vOs1/1f+CugDbBpj/CKkOY/+R/rgNw7oFEJYOcY4rcBjXUfqJdWS1Dunrv+n1T3LW5B6P1S/Rx1IGia2HSnQWIk0DKlQ+LrIe0cIIf+9o852Q11t0DpqhvTe0SmE0CEvUOpOw7Xdam1vNrIlPe91HlMIoQupp9PwGOP/8m6qbo+eXeA+Q4C1Qkin2Mqt7s6yB6OFFKp7HHBkjPH5AjX1KHC/WtvUIYRe5LVv8tryS19ojBcCF4YQupI+S/wG+MMS7laz7QbpOXwkf9c1H6qeJS3SditwbEtqu9Xa5l3OturaIYQWeYFSd9LZspWxOsOkgQMHTqTIZ5VQyWkVQsifOK+S1Fj4fQjhZdI/r9NIXaEJafLFd0lvGjNI3wosCGnc61qkb7jmkhoDdTW4byb1YNmc9A1OtQ65/c7K9Vb4KTBp8bsD8Bqph9OVpNT7xyz89uZB4JyQTiNfPafKd4FZudr3BR6MMU7PfZuzgPq5HTglpHH2zwC/JP3jfaGe96/LpcBZIYTXYoyjQ+qmvEOM8Y4G2PeSLOl5/5I0J89ibxi5b7X+CWyf960WMcaqEMIVwN9DCMNjjBNDmktnw1yYcDtp4tDrSY2g0xvhuKrrrvZfYEYI4WTSWPV5pPH1bXMB4yJijJ+HEP4NnBdC+APp92ddoFtMwwRuB34RQniQ9AZd329JFxHS3Bn/iTGOJzUEq4eOLsktwMkhhIdJr1f1fFKVofC805NIH7TWY8lv4ks6ttdIQ+oeJs1PtA8LG0I3Ai+HNBfV46RvujYj/f7MBzYFniD9n5iVd6xfAt1CCK1jjPMK1HQzcGsI4WbSt5J/JnX//3gJxyItF9tTqsWepP9fG5HeT6rdTpoL6STSh7tzQgin57YdmPtAfCVwfkiTUb9KCpbm53pqvwYcGEIYTZo7ZwhpqHBtOpD+n04LaT6Wb99Pc+9jD5Pm8fkZ6X/u5jHG6klz7wUuIfXw+OsSjvdWUi/kTiz6RWAHUlvoK9IH1T/XsY/XgQ1ygde75A1xX1K7obY26BJqJsY4LoTwAvCXkE640pvUZjy47nvWS63tzZjmAmxMS3rea7aBvhXS8LC7gJtijTluSF/w3BNCeJzUbmoHDCW1e/9D+qzwixDCxaT/i4NJw8obypfAqiGElWKaxgFSG/nsEMJhMcZPckHYFjHG+2rZR11t6odIk1nvDdwP/IxFQ5V6CSF8j/RZ51VSO2ku9Wu7jQD+GUI4kPS/4odAf9LvUm1qfS1rWNKxvUZqN3YHppOmCwCW3OZdQlu11s8JwEuk5+ekEMJ5pDmcdidNPaGM+e2YltcIUgOkejmD1FPgFeAN4E3SP8nqsyn1In04nEV6Q7kkpgmG2wDnkL5V+oLUJfLbs4UVcAvpjenJvK7hkCbqPRCYSXozq/kGl+/vpMbbl6Rv1r799iH3zdMOpHG+E3I1nZurE1KPqY9zQdKx1LNBEWMck9v2n6Rj3Z00wWOhD75LJcZ4T67GW3N1vUXqvVUMS3rezwCuy3V7/VGN24aRJsx7Liw888bDudtOJr2xvJg7psdJ354SY3yYNJb7ydw2i02C3gCuIs2NNS2EcG9u2N/upIbeR6TX8ErSt3m1ORRoDbxNevO8kzRnA6Tn6lFS4/hV4O5lrPN7wEshhFmkN/9fxhg/qsf9ribNb/EM6XjmkiY2LCjGOIf0QeD53HOyWR37XtKx/YH04WcqaV61/LkzxpF+L35HCrDGkb6ta5FbfkX6u5xC+pBUPcn+k6QekV+EEPL/L1Tv94nc495F+oDWk0Xnu5KkYjqMNPffpzHGL6oX0tCXg0g9RHYnTWb7Kal30X4AuQ+1Z5P+d84khTqdcvv9Ze5+03L7uXcJdVxAmr9xMumsco/UuP0QUpD/LjCRNC0BuTq+Jv1PXZclvIfFNB/Mf0jzROa3E64nDVv5jPRe+WId+3gPOJPUHngfqDnvSq3tBmpvg9bHAaRJ0CeQeuOcHtN8fsulHu3NxrSk532RNlCN27qRegQfHxY9I2/3GOMrpB4oF5He48eSO8Ndrr27d+76VNLv87K2fQqKMb5L+pzwYa72NUkThN9PGuI4M3esm9axj1rb1LnPHfuSPrd8Rfq9WqzHUz10JLWVppJeh69Ic6ku6fi+Is219avcfU4Cdqvxeaimf5DmiJoaQriwjn3XeWy53/nbSJ/xRrF4gFVXm7eutuoZ1PI5Ifc7swfp+Z9MCq8Pzb3OyliIS553TpKk5ZLrRv4R0KqWIX+SpBIUQjgN6B1jbIieOpKakBDCSODGGOOVWdeipqfOYW6SJEmSVEhuWNyPKTDHpSSpvDnMTVLZCeksEbMKLAct+d6qrxDCpbU8z5dmXZukhUIIV4cQJoYQCp6hLCQXhhDGhhDeCCFsUuwaVXpCCEeThiE/nDeHkrRMQgi/q6VN8fCS7636CiFsVcvzPCvr2lR6HOYmSZJUxkIIW5Pmibk+xrhhgdt3Ic2XtgtpHpF/xBhrnU9EkiTJnkmSJEllLNdrZEodmwwjBU0xxvgi6dT0a9SxvSRJauYMkyRJkpq3tUjDlaqNz62TJEkqqCwm4O7cuXPs0aNH1mVIkqRGMmrUqMkxxi5Z11GmQoF1BedBCCEcAxwDsOKKKw7s27dvY9YlSZIyVlsbrCzCpB49evDKK69kXYYkSWokIYRPsq6hjI0H1s673g2YUGjDGOPlwOUAgwYNira/JKl2McKCBWmprFx4ua51CxZAVVXh9cuyVFWlOqp/Vi9Le72ubaqqFi7Vj1nXumXZpubjL83l+mxb83Ur9Fouz7r8pdC6utbX5z7XXw977lmvX8ulVlsbrCzCJEmSJC2z+4HhIYRbSRNwT48xfp5xTZJKzIIFMH8+zJuXlurLlZXpcs2fhdYty221BTCFApklbVPoem2BT33WN5dzXVVUQIsWC5ea1wutq+82ISy8HsLC69WXq7epub76cm33y79cveSreb0h1tV8rJrr6lq/pPusu279XquGZJgkSZJUxkIItwBDgc4hhPHA6UArgBjjpcAI0pncxgJzgCOyqVRSTVVVC8OZefPgm2+WfHlptstfqsOfQoFQfdZVVRXveQkBWrWCli1TmFC9VAcQ9b1ec13r1oteb9Fi8ceoqCi8blnWF7qtUI3Ls9QMYgoFKUu6Xtc2ar4MkyRJkspYjPGAJdwegZ8VqRypJFRVwddfw5w5hZfZs9Pt33yzMKBpjMuVlQ1/bC1bQps2KYyp/tm6dVqqL1f/bNcOVlpp0XWFtqttXcuW6Xr1Un295s+lva2Fp5GSMmeYJEmSJKnkVFXBrFkwc2ZaZsxY9Ofs2WmpLRAqFBBVX547d9nrCiGFNG3apECltssrrgidOtW9Tc2f1Uv+9fpcrr5uECOpoRQ1TAoh7AT8A6gArowxnlPj9pWAG4Huudr+FmO8ppg1SpIkSWoc1QHQjBmLhz/VP5e0rvryrFn1f9zqXjY1l+pAp9BtNberua5t28IhUPUcLpJUzooWJoUQKoCLge1JZw15OYRwf4zx7bzNfga8HWPcPYTQBRgTQrgpxjivWHVKkiRJWlSMqbfOjBkwffqy/5w5s36TErdqBR07QocOC3927gzrrbfourout2+fQqC2bdPwKElSwynmv9XBwNgY44cAuTOGDAPyw6QIdAghBKA9MAVohJHCkiRJUvNUWQmTJ8OkSWmZOHHh5epl8uTFw6D585e873btUqCz0koLf66++qLXO3ZcdCkUBLVp0/jPgyRp2RUzTFoLGJd3fTzp9LP5LiKdnnYC0AHYL8ZYxPMCSJIkSaXlm28WD4PqWqZOLbyfEGDVVaFLl9QLaO21Fw+Gav7Mv9yhQ+pRJEkqf8UMkwqNHK7ZyXVH4DVgG6An8FgI4dkY44zFdhbCMcAxAN27d2/YSiVJkqQmoKoKJkyADz5YuIwdC59+ujAcmrFYSzmpqEihUJcuafnudxderrl07ZrmDqqoKOrhSZJKVDHDpPHA2nnXu5F6IOU7Ajgnd4rasSGEj4C+wH9r7izGeDlwOcCgQYPqMfJakiRJanrmz4ePP140LKq+/OGHi55ZrKICevRIy7rr1h0OrbyyZ+6SJDWOYoZJLwO9QgjrAp8B+wMH1tjmU2Bb4NkQwmpAH+DDItYoSZIkNbjZsxfvXVR9+dNPYcGChdu2bQs9e0Lv3rDzzulyz56w/vrQvbuTSUuSsle0t6IYY2UIYTjwKFABXB1jHB1CODZ3+6XAWcC1IYQ3ScPiTo4xTi5WjZIkSdLymD8fHn8cXnll0cDoiy8W3a5TpxQQbbYZHHTQwrCoZ880YbWnlpckNWVF/V4jxjgCGFFj3aV5lycAOxSzJkmSJGl5VFXBc8/BLbfAHXfAV1+l9WutlcKhXXZZtHdRz55pCJokSaXKTrKSJEnSUooRXnsNbr4Zbr0Vxo+Hdu1gjz3gwANh223TdUmSypFhkiRJklRP77+feiDdcgu8+26av2inneDcc1OQ1L591hVKktT4DJMkSZKkOkyYALfdlnohvfJKms9o663h+ONhn31g1VWzrlCSpOIyTJIkSZJqmDoV7rorBUgjR6ZhbZtsAn/7G+y3H3TrlnWFkiRlxzBJkiRJAmbPhgceSEPYHn44nZmtVy847TQ44ADo0yfrCiVJahoMkyRJ0rKJEebMgZkz09KuXTp9lVRC5s+Hf/879UC6774UKK25Jvz852ki7U02ScPaJEnSQoZJkiQ1J/PmLQx/lmaZMWPxdbNmpXOiVzvmGLjssuyOTaqnqip47rkUIN15J3z1FayyChx0UOqBtNVWUFGRdZWSJDVdhkmSJJW722+Hk0+Gzz+Hb76p331at4YOHRZdOnWCHj0WX1+99O3bqIchNYQJE+AHP4D33kud6YYNSwHSjjumX3tJkrRkhkmSJJWrKVNg+PA0AcygQfCjH9UeBHXsuOh1P1WrDM2dC3vvDZ99BjfcAHvtBSuumHVVkiSVHsMkSZLK0b//DUccARMnwllnwSmnQEvf9tV8xQg//Sm89BLcfXcKkiRJ0rJpkXUBkiSpAc2enXoj7bgjrLQSvPgi/P73Bklq9v75T7j2WjjjDIMkSZKWl2GSJEnl4sUXYeON4eKL4YQTYNQoGDgw66qkzD3xBJx4Iuy5J/zhD1lXI0lS6TNMkiSp1M2blz4hf//7aYLtJ5+E88+Htm2zrkzK3IcfpunC+vaF66+HFrZ+JUlabvZ5lySplI0eDYceCq++CocfDhdckIa3SWLWrHS2thjhvvvS3PKSJGn5+d2MJEmlqKoq9T4aOBDGjYN77oFrrjFIknKqquCww+Dtt+H226Fnz6wrkiSpfNgzSZKkUvPJJ+lT8tNPwx57wBVXQNeuWVclNSlnn53O2nb++bDddllXI0lSebFnkiRJpSLGdDqqjTZKw9quvhruvdcgSarhvvvgtNPSCNDjj8+6GkmSyo9hkiRJpWDixHQ+8yOOgE02gTfeSJdDyLoyqUkZPRoOPhi+9z247DL/RCRJagyGSZIkNXX33QcbbgiPPALnnZfO1tajR9ZVSU3OlClpwu327dM0YiuskHVFkiSVJ+dMkiSVvk8/hc8+q3ubJXVPWNLtnTrBeutBRcXS1bY8ZsyAX/4yDW3beON0XvMNNyze40slpLIS9t8/zUc/ciSstVbWFUmSVL4MkyRJpe2GG+DII9MnycbWpg307Qv9+8MGG6Sf/fun00S1bOC31JEj4fDD0yfjU09NE8C0bt2wjyGVkVNOgcceg6uugs03z7oaSZLKm2GSJKk0xQjnngu//S384Adw0kl19y6Kccn7q+u2iRPTOcZHj4b//AduuWXh7a1bQ58+iwZMG2yQQqZWrZbuuObOTeHR3/8O668Pzz8Pm222dPuQmpkbbkgjQH/+85QtS5KkxmWYJEkqPQsWpE+N//oXHHAAXHNN6jVUTLNmwbvvLgyY3n4bXnoJbr114TatWqWQqWZPpl69CodMr74KhxyS9nXccfDXv8KKKxbvmKQS9PLLcPTRKVM+77zcym++gc8/hwkT0hDYzz5beHnCBPjiizShUufOiy+rrrr4OidfkiRpEYZJkqTSMmcOHHhgmpT6pJPgL3+BFhmcT6J9exg0KC35Zs9eGDJVB02jRsEddyzs/dSyJfTuvWjA9O67cNZZ0LVrmmh7xx2Lf0xSU1dVBZMmfRsMTX9nAs+d9RnXtJrAD1t8Rqvv5QKjyZMXv2+bNrDmmmnZYIMUOE2eDB9/nH5Om1b74664Yu1BU6EQatVVix9wS5JURIZJkqTSMXky7L576gH0z3/C8OFZV7S4FVeEgQPTkm/OHBgzZmEvprffhv/9D+68c2HIdMABcNFFabJvqbmqnkE7vydR9c/PP4f587/ddCXglwQWrNqVVlPXgu7d07DQtdZKoVH+z06d6h4KO39+Oh3cV1+l/zU1l/z1Y8emnzNm1L6/jh1TONy1K6y22sLLNa+vthqsvHI2obgkScvIMEmSVBo+/BB23hk++SQFMHvvnXVFS6ddu3RGto03XnT911+nkGnePBg8OJvapKbkv/+FQw9Nlzt2XBgIDR36bTAU11iTP1+3Fpc+sCZ/v3l19jlgKecmK6RVqxTsrLZa/e8zb14KoAqFT5MmpbnWJk6E996D555L6wvNz9ayJXTpsuTgqXpx2J0kKWOGSZKkpm/UKNhll9Rz4Ikn4Pvfz7qihtO2LXz3u1lXITUd222XAtY110zDSQu45GL4/QNprvp9Dihyfflat4bVV09LfSxYkHo4ffnlwqBp4sTFr7/3Xlr39deF99OxYxpKt+KKKaiu/tkQlxv6zJSSpLLku4UkqWl7+GHYd980D8nTT0PfvllXJKkxrbRSWmoxciQcf3wa8XrmmUWrqmFUVCzsXVQfs2fXHjh99VUaPjt7dvr5xRcLL1evry2MqkurVilY6tAh9Qrr3n3hss46Cy8vadigJKmsGSZJkpqua65Jp2n6znfgoYdgjTWyrkhShj7+OGXLvXrBjTc2g2mGVlwR1l03Lcuiqgrmzl08ZKrrcvXPGTPSXFWvvw4PPJD2U7O2/KCpZti01lqp51ZWqocTlmLgNWtWGtL92WdpSOPKK8Mqq6Sf7duX5jFJKjuGSZKkpifGdGaz00+HHXZIcyR16JB1VZIyNHs27LknVFamkzl27Jh1RSWgRYuFQ9iWR4xpvqdPP00hx6efLlw++SSdTGDixEXvE0L6AiA/YMoPnNZeO/XUWpqAa2m3bd06PU63bulnoWWllYofzlSHRR9/XHgpdDbCai1apFCpeqkOmWouta1v184wSlKDMEySJDUtlZVw3HFwxRVpEt4rr0zDLiQ1WzHCEUfAm2+mToq9emVdUTMTQpogvEuXxc9UWe3rr2H8+MXDpk8/TfPe3XNPmrB8edQ2z1OnTikwqjmH1Ndfp7MDjhsHTz2Vzgi4YMGi+2zfftFwqVDwVMvcXbWaObP2sOiTTxYPi9q0gR490jJw4MLLa62VnrNp0xYuU6cuen3aNHjnnYWX58ypu7ZWrRYPl1q1SsFb69YLLy/PuvzLAN98k5a5cxv+clVV+r2snrtsjTUWXs5fmsqwzG++Sa/hlClpmTp14fDX1VZLx1IKE+zHmELRSZPS71379gsDTNtszYZhkiSp6Zg9G/bbL31aPPXU1DupKTT+JGXqnHPgjjvgr3+FnXbKuhoV1LZtSvlqS/qqqtIHz+reTOPGpQ+k9Z1EvG3b5X8/qKxMc0tVB0w1lzfeSHNT1Tzj3sorLx4wdeuWejWNH794YPTVV4vef4UVFgZE3/te6p1Vfb1HjxQkNNSYzXnzYPr0wqFTzTBq6tSFwyCnTk0nuZg3Ly21XW4srVql56lNm7QUurziiouvDyH1ivviC3j+efj88xTYFNp/oZCpUPjUtm3dtca48DmrDoVqW2puM3v2kp+Ljh1rP5NjzXWrrNIwvztVVel3YtKk+i2TJxd+nmHRYGmVVRZfalu/yirpdV0WMS7smTh7dgq6qi/Xti7/elVV+n1qyKVNmxQU1rf+BQtSHfk/l2Zd9f+kIgqx0OlJS8ygQYPiK6+8knUZkqTlMXEi7LZb+gb7kkvgJz/JuiI1ISGEUTHGQVnXoYWK1f568EHYYw844IA0T5L5shrVvHmpB1NtgdO4cYv3LMoPiwotXbuWxy9u9QfeusKm6svVP2NcPBSqGQi1bt1wYVqMab6vL75Iy+efL7xcc5k4cfHgEFKYUx0yde2aAreaIdH8+bXX0Lp16glVn2WVVVLIWdtE+9XXJ08uXGvLlqk3U12BU9u26f6FAqH8yzV77VXr0CE9RufOC3so5i8rr5yCmeqgcurUwsu0aWm7uqywQuHwacGCukOhJfXIK/S8tW+fAsoVV0y/f9W93vKX5c1KqkPSVq1S8FNbINQQmcwdd8A++yz/fgqorQ1mzyRJUvbGjk3dDSZMSEMh9tgj64okNQHvvAMHHggbb5xGvJbD53E1ca1bLwyBalM9pG/69NRLqVzCoiUJIX0Ib9mEP0KGsPCMkH361L1tZWUKUgoFTdUh1OuvpzCmUyfo379+4VBjzEu1YEHq8VZX4DRxYmpPTZxYdw+oVVZZGAStvz5svnnhkKg6QGrIYXfz5xcOnGoLoT77DN56K/3OVQc/K60Ea6658Hp+KFTfdfU5OUCMqd6aAdOyLPPmpcCqoiIt1Zdr/lyedZtu2nCvUz014f8EkqRm4aWXUo8kgCefhM02y7YeSU3CtGkwbFj6HHfPPUseeSIVTfWQPpW2li1TD6RSOFNs9bxKXbvChhsuefvZsxcGTHPmLOxVtOqq2c5p1KrVwqCqqQth4RxgnvGhIMMkSVJ2HnwQfvSj1JB75BEb55KA9CX8gQem6WeefDKd/EuSVE8rrgjrrpsWqZE00OBUSZKW0uWXp24HG2wAL7xgkCTpW6eeCg8/DBddBFtumXU1kiSpJsMkSVJxxQh/+EOaYHunnWDkyDRZpCQBd94J554Lxx4LxxyTdTWSJKkQwyRJUvHMnw9HHAF/+hMcdRTcd1/qii1JOVtvDSecAP/4R9aVSJKk2jhnkiSpOGbOhH33hUcfhTPOgNNOax5nv5G0VLp2hfPPz7oKSZJUF8MkSVLj++IL2HXXdIrdq66CI4/MuiJJkiRJy8hhbpKkxlNZCbfdBptvDmPGwAMPGCRJkiRJJc4wSZLU8KZNg7/9DXr2hP33h1at0kTbO++cdWWSJEmSlpNhkiSp4XzwAfziF9CtG/zmN7DeenD//fDuuzBoUNbVSZIkSWoAzpkkSVo+McKzz6YZc++/H1q2hAMOgOOPh403zro6SZIkSQ3MMEmStGzmzYPbb4e//x1efRVWXRV+9zv42c9gjTWyrk6SJElSIzFMkiQtnSlT4LLL4KKLYMIE6NcvXT/4YGjXLuvqJEmSJDWyooZJIYSdgH8AFcCVMcZzatz+G+CgvNr6AV1ijFOKWackqYAxY+CCC+C66+Drr2H77eGqq2CHHaCFU/BJkiRJzUXRwqQQQgVwMbA9MB54OYRwf4zx7eptYoz/B/xfbvvdgRMMkiQpQzHCk0+moWwPPQRt2qQeSMcfDxtumHV1kiRJkjJQzJ5Jg4GxMcYPAUIItwLDgLdr2f4A4JYi1SZJyvfNN3Dzzakn0htvQNeu8Mc/wrHHpsuSJEmSmq1ihklrAePyro8HNi20YQihHbATMLwIdUmSqk2aBP/6F1x8MUycCBttBFdfnc7OtsIKWVcnSZIkqQkoZpgUCqyLtWy7O/B8XUPcQgjHAMcAdO/effmrk6TmbPToNJTtxhtTr6Rdd4UTToBttoFQ6N+3JEmSpOaqmGHSeGDtvOvdgAm1bLs/SxjiFmO8HLgcYNCgQbWFUpKkunz6KfzqV3DnndC2LRxxBPzyl9C3b9aVSZIkSWqiihkmvQz0CiGsC3xGCowOrLlRCGElYAhwcBFrk6TmZe5cOO88OPvsdP300+HnP4dVV822LkmSJElNXtHCpBhjZQhhOPAoUAFcHWMcHUI4Nnf7pblN9wL+HWOcXazaJKlZeeih1Pvogw9gn31SqORwYUmSJEn1VMyeScQYRwAjaqy7tMb1a4Fri1eVJDUTY8fC8cenMKlvX3jsMdhuu6yrkiRJklRiWmRdgCSpkc2eDb//PWywATz9NPztb/D66wZJkiRJkpZJUXsmSZKKKEa46y448UQYNw4OPhj++ldYY42sK5MkSZJUwuyZJEnl6O23YfvtYd99YZVV4Jln4IYbDJIkSZIkLTfDJEkqJzNmwK9/DQMGwKhRcNFF6edWW2VdmSRJkqQy4TA3SSoHMcJNN8FvfgNffgk//jH8+c/QpUvWlUmSJEkqM4ZJklTqXnsNhg+H55+H730P7rsPBg/OuipJkiRJZcphbpJUqqZMSSHSwIEwZgxceSW8+KJBkiRJkqRGZc8kSSo1CxbA1VfD736XAqWf/Qz++Mc00bYkSZIkNTJ7JklSKXnpJdhsMzjmGOjXD159FS680CBJkiRJUtEYJklSKZg4MU2qvdlm8NlnabLtp59OZ22TJEmSpCIyTJKkpu6aa6B3b7j++nS2tjFj4MADIYSsK5MkSZLUDDlnkiQ1ZRddBD//OQwdCv/6F/Ttm3VFkiRJkpo5wyRJaqr+9a8UJO25J9x+O7RqlXVFkiRJkuQwN0lqki6/HI47DnbfHW67zSBJkiRJUpNhmCRJTc3VV8NPfgK77AJ33AGtW2ddkSRJkiR9yzBJkpqS666Do46CHXaAu+6CNm2yrkiSJEmSFmGYJElNxY03whFHwLbbwr33wgorZF2RJEmSJC3GMEmSmoJbb4XDDktnbbvvPmjbNuuKJEmSJKkgwyRJytodd8DBB8OWW8IDD0C7dllXJEmSJEm1MkySpCzdfTcccABsthk89BCsuGLWFUmSJElSnQyTJCkr990H++0HgwfDww9D+/ZZVySpTIUQdgohjAkhjA0hnFLg9pVCCA+EEF4PIYwOIRyRRZ2SJKk0GCZJUhYefBD23Rc22SQFSR06ZF2RpDIVQqgALgZ2BvoDB4QQ+tfY7GfA2zHGAcBQ4LwQQuuiFipJkkqGYZIkFdvDD8MPfwgDBsCjj8JKK2VdkaTyNhgYG2P8MMY4D7gVGFZjmwh0CCEEoD0wBagsbpmSJKlUGCZJUjH9+9+w116wwQbp8sorZ12RpPK3FjAu7/r43Lp8FwH9gAnAm8AvY4xVxSlPkiSVGsMkSSqWJ56AYcOgb1947DFYZZWsK5LUPIQC62KN6zsCrwFrAt8FLgohdFxsRyEcE0J4JYTwyqRJkxq6TkmSVCIMkySpGEaOhN13h1694PHHYdVVs65IUvMxHlg773o3Ug+kfEcAd8dkLPAR0LfmjmKMl8cYB8UYB3Xp0qXRCpYkSU2bYZIkNbZnn4Vdd4V1101BUufOWVckqXl5GegVQlg3N6n2/sD9Nbb5FNgWIISwGtAH+LCoVUqSpJLRMusCJKmsPf887LwzdO8OTz4JXbtmXZGkZibGWBlCGA48ClQAV8cYR4cQjs3dfilwFnBtCOFN0rC4k2OMkzMrWpIkNWmGSZLUWF58MQVJa62VgqTVVsu6IknNVIxxBDCixrpL8y5PAHYodl2SJKk0OcxNkhrDf/8LO+6YAqQnn4Q11si6IkmSJElqEIZJktTQRo2CHXZIcyM99VTqmSRJkiRJZcIwSZIa0v/+B9tvD6uskoKkbt2yrkiSJEmSGpRhkiQ1lNdfh+22gw4dUpDUvXvWFUmSJElSgzNMkqSG8NZbKUhq1y7NkdSjR9YVSZIkSVKjMEySpOX19tuwzTbQunXqkdSzZ9YVSZIkSVKjMUySpGURI7z0Ehx/PGy5JVRUpCBp/fWzrkySJEmSGlXLrAuQpJLy1ltwyy1w663w4YfQpg3ssguccw707p11dZIkSZLU6AyTJGlJPvooBUi33JLCpBYt0vxIf/gD7LUXrLRS1hVKkiRJUtEYJklSIV98AbffngKkF19M67bYAv75T9h3X1httWzrkyRJkqSMGCZJUrWpU+Huu1OA9NRTUFUFAwakIWz77w/rrJN1hZIkSZKUOcMkSc3bnDnwwAMpQHr4YZg3L02ifeqpKUDq3z/rCiVJkiSpSTFMktT8zJsH//53CpDuuw9mz4Y114Sf/QwOOAAGDYIQsq5SkiRJkpokwyRJzcOCBfDssylAuvNOmDIFOnWCgw5KAdJWW0FFRdZVSpIkSVKTZ5gkqbxNnw5nnZVCpAkTYMUVYdgwOPBA2H57aN066wolSZIkqaQYJkkqb8OHpyBpt91SD6TddkuBkiRJkiRpmRgmSSpfI0bAjTfCH/4AZ56ZdTWSJEmSVBZaZF2AJDWKGTPgJz9JZ2M79dSsq5EkSZKkslHUMCmEsFMIYUwIYWwI4ZRathkaQngthDA6hPB0MeuTVEZOOinNkXT11dCmTdbVSJIkSVLZKNowtxBCBXAxsD0wHng5hHB/jPHtvG1WBi4BdooxfhpC6Fqs+iSVkZEj4bLL4MQTYdNNs65GkiRJkspKMXsmDQbGxhg/jDHOA24FhtXY5kDg7hjjpwAxxolFrE9SOZgzB446Cnr2TGdxkyRJkiQ1qGKGSWsB4/Kuj8+ty9cbWCWEMDKEMCqEcGjRqpNUHv7wB/jgA7jySmjXLutqJEmSJKnsFPNsbqHAuljjektgILAt0Bb4TwjhxRjje4vtLIRjgGMAunfv3sClSipJL70EF1yQJt4eOjTraiRJkiSpLBWzZ9J4YO28692ACQW2eSTGODvGOBl4BhhQaGcxxstjjINijIO6dOnSKAVLKiHffANHHglrrgl//WvW1UiSJElS2SpmmPQy0CuEsG4IoTWwP3B/jW3uA7YKIbQMIbQDNgXeKWKNkkrV2WfD22+nibc7dsy6GkmSJEkqW0Ub5hZjrAwhDAceBSqAq2OMo0MIx+ZuvzTG+E4I4RHgDaAKuDLG+FaxapRUol5/Hf7yFzj4YNhll6yrkSRJkqSyVsw5k4gxjgBG1Fh3aY3r/wf8XzHrklTCKivhxz+GTp3SfEmSJEmSpEZV1DBJkhrceefBqFFw++2w6qpZVyNJkiRJZa+YcyZJUsMaMwZOPx322gv22SfraiRJkiSpWTBMklSaqqrgqKOgbVu4+GIIIeuKJEmSJKlZcJibpNL0r3/Bc8/BNdfAGmtkXY0kSZIkNRv2TJJUej75BE45BXbcEQ47LOtqJEmSJKlZMUySVFpihGOOSZcvu8zhbZIkSZJUZA5zk1RarrsO/v1vuOgiWGedrKuRJEmSpGbHnkmSSsfnn8MJJ8BWW8FPf5p1NZIkSZLULBkmSSoNMcLPfgZz58KVV0IL/31JkiRJUhYc5iapNNx5J9xzD5x7LvTunXU1kiRJktRs+dW+pKbvq69g+HAYOBBOPDHraiRJkiSpWbNnkqSm7/jjYcqUNPF2S/9tSZIkSVKW7JkkqWkbMQJuvBF++1sYMCDraiRJkiSp2TNMktR0zZgBP/kJ9O8Pp56adTWSJEmSJBzmJqkpO+kkmDAhTb7dpk3W1UiSJEmSsGeSpKZq5Ei47LI0X9Kmm2ZdjSRJkiQpxzBJUtMzZw4cdRT07AlnnZV1NZIkSZKkPA5zk9T0/OEP8MEH8NRT0K5d1tVIkiRJkvLYM0lS0/LSS3DBBWni7aFDs65GkiRJklSDYZKkpuObb+DII2HNNeGvf826GkmSJElSAQ5zk9R0/PnP8Pbb8NBD0LFj1tVIkiRJkgqwZ5KkpuGNN1KYdPDBsMsuWVcjSZIkSaqFYZKk7FVWpuFtnTql+ZIkSZIkSU2Ww9wkZe/882HUKLj9dlh11ayrkSRJkiTVwZ5JkrL13ntw+umw116wzz5ZVyNJkiRJWgLDJEnZ+fJLGDYMVlgBLr4YQsi6IkmSJEnSEjjMTVI2Jk6EbbaBTz+Fhx+GNdbIuiJJkiRJUj0YJkkqvsmTYbvt4KOPYMQI2HrrrCuSJEmSJNWTYZKk4vrqqxQkvf8+PPggDB2adUWSJEmSpKVgmCSpeKZOhe23h3ffhfvvh223zboiSZIkSdJSMkySVBzTpsEOO8Do0XDvvemyJEmSJKnkGCZJanzTp8OOO8Lrr8Pdd8POO2ddkSRJkiRpGRkmSWpcM2em8OjVV+HOO2G33bKuSJIkSZK0HAyTJDWeWbNgl13gv/+F22+HYcOyrkiSJEmStJxaZF2ApDI1ezbsuiv85z9wyy2w995ZVyRJkiRJagD2TJLU8ObMgd13h+eeg5tugn33zboiSZIkSVIDMUyS1LC+/joNZxs5Em64AfbfP+uKJEmSJEkNyGFukhrO3Lmw557wxBNwzTVw0EFZVyRJkiRJamD2TJLUML75Js2L9O9/w1VXwWGHZV2RJEmSJKkR2DNJ0vKbNw/22QcefhguvxyOPDLriiRJkiRJjcQwSdLymT8ffvQjePBB+Ne/4Oijs65IkiRJktSIDJMkLbv589ME2/fdB//8Jxx7bNYVSZIkSZIamWGSpGVTWZkm2L77bvj732H48KwrkiRJkiQVgWGSpKVXWQmHHAJ33AF/+xscf3zWFUmSJEmSisQwSdLSWbAAjjgCbr0Vzj0XfvWrrCuSJEmSJBWRYZKk+luwAH78Y7jxRjj7bDjppKwrkiRJkiQVWVHDpBDCTiGEMSGEsSGEUwrcPjSEMD2E8FpuOa2Y9UmqQ1UVHHMMXHcd/PGP8LvfZV2RJEmSJCkDLYv1QCGECuBiYHtgPPByCOH+GOPbNTZ9Nsa4W7HqklQPVVXpTG1XXw1/+AOcZs4rSZIkSc1VMXsmDQbGxhg/jDHOA24FhhXx8SUtixjTmdquuCL1RvrjH7OuSJIkSZKUoWKGSWsB4/Kuj8+tq2nzEMLrIYSHQwgbFKc0SYtZsACeew4OPhj+9a80P9Kf/gQhZF2ZJEmSJClDRRvmBhT6BBprXH8VWCfGOCuEsAtwL9Cr4M5COAY4BqB79+4NWKbUjM2dC08+CffcA/ffDxMnQuvWqUeSQZIkSZIkieKGSeOBtfOudwMm5G8QY5yRd3lECOGSEELnGOPkmjuLMV4OXA4waNCgmqGUpPqaPh1GjIB7700/Z82CDh1gl11gr71g552hY8esq5QkSZIkNRHFDJNeBnqFENYFPgP2Bw7M3yCEsDrwZYwxhhAGk4bhfVXEGqXm4fPPU8+je+5JPZHmz4fVVoMDD4Q994RttoE2bbKuUpIkSZLUBBUtTIoxVoYQhgOPAhXA1THG0SGEY3O3XwrsA/w0hFAJfA3sH2O015HUEN5/P/U+uuceePHFNLF2z57wy1+mHkibbgoVFVlXKUmSJElq4orZM4kY4whgRI11l+Zdvgi4qJg1SWUrRnj11RQe3XsvjB6d1m+yCZx5ZuqBtMEGzoMkSZIkSVoqRQ2TJDWyykp45pkUHt17L4wbBy1awNZbwz/+AcOGwTrrZF2lJEmSJKmEGSZJpW7OHPj3v1N49MADMGUKrLAC7Lhj6oG0227QuXPWVUqSJEmSyoRhklTKRo2C7baDadNg5ZVh993T8LUdd4QVV8y4OEmSJElSOTJMkkrVxIlp4uwOHeCOO2DIEGjVKuuqJElNUAhhJ+AfpJOgXBljPKfANkOBC4BWwOQY45AilihJkkqIYZJUiubPh333hcmT4bnn0qTakiQVEEKoAC4GtgfGAy+HEO6PMb6dt83KwCXATjHGT0MIXTMpVpIklYQWWRcgaRkcf3yaaPuqqwySJElLMhgYG2P8MMY4D7gVGFZjmwOBu2OMnwLEGCcWuUZJklRCDJOkUnPllXDJJfCb38ABB2RdjSSp6VsLGJd3fXxuXb7ewCohhJEhhFEhhEOLVp0kSSo5DnOTSskLL8Bxx8EOO8Bf/pJ1NZKk0hAKrIs1rrcEBgLbAm2B/4QQXowxvrfIjkI4BjgGoHv37o1QqiRJKgX2TJJKxWefwQ9/CN27w623QkVF1hVJkkrDeGDtvOvdgAkFtnkkxjg7xjgZeAYYUHNHMcbLY4yDYoyDunTp0mgFS5Kkps0wSSoFc+fC3nvDrFlw332wyipZVyRJKh0vA71CCOuGEFoD+wP319jmPmCrEELLEEI7YFPgnSLXKUmSSoTD3KSmLkb46U/hv/+Fe+6BDTbIuiJJUgmJMVaGEIYDjwIVwNUxxtEhhGNzt18aY3wnhPAI8AZQBVwZY3wru6olSVJTZpgkNXX//Cdcey2cfjrsuWfW1UiSSlCMcQQwosa6S2tc/z/g/4pZlyRJKk0Oc5OasqeeghNPhGHD4LTTsq5GkiRJkiTDJKnJ+vhj2Hdf6N0brr8eWvjnKkmSJEnKnp9OpaZo9uw0pG3BgjThdseOWVckSZIkSRLgnElS0xMj/PjH8MYbMGIE9OqVdUWSJEmSJH3LMElqas49F267Dc45B3baKetqJEmSJElahMPcpKbk4Yfhd7+D/feHk07KuhpJkiRJkhZjmCQ1Fe+9BwccAAMGwFVXQQhZVyRJkiRJ0mIMk6SmYMaMNOF2q1Zwzz3Qrl3WFUmSJEmSVJBzJklZq6qCgw9OPZMefxx69Mi6IkmSJEmSamWYJGXtj3+EBx6ACy+EoUOzrkaSJEmSpDo5zE3K0t13w5lnwhFHwPDhWVcjSZIkSdISGSZJWXnrLTj0UBg8GC65xAm3JUmSJEklwTBJysKUKTBsGHTokCbcXmGFrCuSJEmSJKlenDNJKrbKSth/fxg/HkaOhDXXzLoiSZIkSZLqzTBJKrbf/hYeewyuuAI23zzraiRJkiRJWioOc5OK6aab4G9/g+OOg6OOyroaSZIkSZKWmmGSVCyjRqUAaeut4YILsq5GkiRJkqRlYpgkFcPEibDXXtClC9xxB7RqlXVFkiRJkiQtE+dMkhrb/Pmwzz4waRI8/zx07Zp1RZIkSZIkLTPDJKmxHX88PPtsmi9pk02yrkaSJEmSpOXiMDepMV10EVxyCfz613DggVlXI0mSJEnScjNMkhrLddfBz38Oe+wB55yTdTWSJEmSJDUIwySpMdx1Fxx5JGy7Ldx2G1RUZF2RJEmSJEkNwjBJamiPPgoHHACbbgr33gsrrJB1RZIkSZIkNRjDJKkhPfss7LUXbLABjBgB7dtnXZEkSZIkSQ3KMElqKK+8ArvuCt27p95JK6+cdUWSJEmSJDU4wySpIYweDTvtBJ06weOPQ9euWVckSZIkSVKjMEySltfYsbDddtC6NTzxBHTrlnVFkiRJkiQ1mpZZFyCVtPHjU5A0fz48/TT07Jl1RZIkSZIkNSrDJGlZTZyYgqQpU+Cpp9Kk25IkSZIklTnDJGlZTJ0KO+wAn36aJtseODDriiRJkiRJKgrDJGlpzZqVztr29tvwwAOw1VZZVyRJkiRJUtEYJklLY+5c2HNPeOkluOMO2HHHrCuSJEmSJKmoDJOk+po/H/bbL52x7brrYO+9s65IkiRJkqSia5F1AVJJWLAADj8c7r8fLroIDj0064okSZIkScpEUcOkEMJOIYQxIYSxIYRT6tjueyGEBSGEfYpZn1RQjHDccXDzzfCXv8DPfpZ1RZIkSZIkZaZoYVIIoQK4GNgZ6A8cEELoX8t25wKPFqs2qVYxwm9+A5dfDr/7HZxSawYqSZIkSVKzUMyeSYOBsTHGD2OM84BbgWEFtvs5cBcwsYi1SYX96U9w3nkwfHi6LEmSJElSM1fMMGktYFze9fG5dd8KIawF7AVcWsS6pMIuuABOOw0OOwz+8Q8IIeuKJEmSJEnKXDHDpEKfxGON6xcAJ8cYFyxxZyEcE0J4JYTwyqRJkxqiPmmhq66CE06AH/4QrrwSWjhXvSRJkiRJAC2L+FjjgbXzrncDJtTYZhBwa0g9QDoDu4QQKmOM99bcWYzxcuBygEGDBtUMpaRld9ttcPTRsNNOcNNN0LKYfyaSJEmSJDVtxfyU/DLQK4SwLvAZsD9wYP4GMcZ1qy+HEK4FHiwUJEmN5sEH4eCDYcst4a67oE2brCuSJEmSJKlJKVqYFGOsDCEMJ52lrQK4OsY4OoRwbO5250lStp56CvbZB7773RQqtWuXdUWSJEmSJDU5RR2/E2McAYyosa5giBRjPLwYNUkAvPgi7L47rL8+PPIIdOyYdUWSJEmSJDVJziosvfEG7LwzrL46PPYYrLpq1hVJkiRJktRkGSapeXvvPdh+e2jfHh5/HNZYI+uKJEmSJElq0gyT1Hx9/TUMGwYxpiCpR4+sK5IkSZIkqcnznOdqvn77W3j33TS0rU+frKuRJEmSJKkk2DNJzdMTT8A//gE//zlst13W1UiSJEmSVDIMk9T8TJsGhx+eeiOdc07W1UiSJEmSVFIc5qbm55e/hM8/hxdegHbtsq5GkiRJkqSSYs8kNS933w3XXw+nngqDB2ddjSRJkiRJJccwSc3HF1/AT34Cm2wCv/991tVIkiRJklSSDJPUPMQIxxwDM2fCDTdAq1ZZVyRJkiRJUklyziQ1D9dcAw88AH//O/Tvn3U1kiRJkiSVLHsmqfx99FGadHvoUPjFL7KuRpIkSZKkkmaYpPK2YAEcfjiEANdeCy38lZckSZIkaXk4zE3l7YIL4Jln0jC3ddbJuhpJkiRJkkqe3TRUvt56C373O9hzTzjssKyrkSRJkiSpLBgmqTzNmweHHAIrrQSXXZaGuUmSJEmSpOXmMDeVpzPPhNdeg3vvha5ds65GkiRJkqSyYc8klZ8XX4S//CVNvD1sWNbVSJIkSZJUVgyTVF5mz4ZDD4W114Z//CPraiRJkiRJKjsOc1N5OekkeP99eOop6Ngx62okSZIkSSo79kxS+Xj0UbjkEjjhBBg6NOtqJEmSJEkqS4ZJKg9Tp8KRR0K/fnD22VlXI0mSJElS2XKYm8rD8OEwcSLcfz+0bZt1NZIkSZIklS17Jqn03X473HwznHYaDByYdTWSJEmSJJU1wySVtgkT4Kc/hcGD4be/zboaSZIkSZLKnmGSSleMcNRR8PXXcP310NJRm5IkSZIkNTY/fat0XXEFPPww/POf0KdP1tVIkiRJktQs2DNJpemDD+DEE2G77eC447KuRpIkSZKkZsMwSaVnwQI49NA0rO3qq6GFv8aSJEmSJBWLw9xUev72N3jhBbjxRlh77ayrkSRJkiSpWbFLh0rLG2/AH/4A++wDBx6YdTWSJEmSJDU7hkkqHd98AwcfDJ06wb/+BSFkXZEkSZIkSc2Ow9xUOk4/Hd58Ex58EDp3zroaSZIkSZKaJXsmqTQ89xz89a9w9NGw665ZVyNJkiRJUrNlmKSmb9YsOOww6NEDzjsv62okSZIkSWrWHOampu9Xv4KPPoKnn4YOHbKuRpIkSZKkZs2eSWraRoyAyy+HX/8attoq62okSZIkSWr2DJPUNM2bB//4Bxx0EGy4IZx1VtYVSZIkSZIkDJPU1FRVwa23Qr9+cPzxMHAg3HMPtGmTdWWSJEmSJAnDJDUlTz4JgwfDAQekuZEeeQQeewzWXz/ryiRJkiRJUo5hkrL35puwyy6w7bYwcSJcfz28+irsuCOEkHV1kiRJkiQpj2FSHebPh6+/zrqKMjZuHBx+OAwYAP/5D/zf/8F778Ehh0ALfzUlSZIkSWqK/MReiwkTYOWVUycZNbCpU+Hkk6FXrzQ/0q9+BR98kM7YtsIKWVcnSVLZCSHsFEIYE0IYG0I4pY7tvhdCWBBC2KeY9UmSpNJimFSLNdaAlVaCkSOzrqSMzJ0L550HPXumXkj77QdjxqTLnTplXZ0kSWUphFABXAzsDPQHDggh9K9lu3OBR4tboSRJKjWGSbUIAYYOhaefhhizrqbEVVXBjTdC376p99HgwfC//8F118E662RdnSRJ5W4wMDbG+GGMcR5wKzCswHY/B+4CJhazOEmSVHoMk+owZAh8/jm8/37WlZSwxx6DgQPTPEirrgqPP57O0jZgQNaVSZLUXKwFjMu7Pj637lshhLWAvYBLi1iXJEkqUYZJdRg6NP18+ulMyyhN//sf7LBDWqZNg5tugpdfTmdskyRJxVTo1Kg1+11fAJwcY1xQ545COCaE8EoI4ZVJkyY1VH2SJKnEFDVMWtLkjyGEYSGEN0IIr+UaKlsWs76aeveG1VYzTFoqH38MBx8Mm2wCo0bB3/8O774LBx7oGdokScrGeGDtvOvdgAk1thkE3BpC+BjYB7gkhLBnzR3FGC+PMQ6KMQ7q0qVLI5UrSZKaupbFeqC8yR+3JzVqXg4h3B9jfDtvsyeA+2OMMYTwHeB2oG+xaqwphDTUbeTING9SKPS9npKvvoI//xkuuiiFRqecks7YtvLKWVcmSVJz9zLQK4SwLvAZsD9wYP4GMcZ1qy+HEK4FHowx3lvEGiVJUgkpZleRJU7+GGOcFeO3012vyOJdsItu6FD47DP48MOsK2mivv4azj03naHtggvgoIPSJFN/+YtBkiRJTUCMsRIYTjpL2zvA7THG0SGEY0MIx2ZbnSRJKkVF65lE4ckfN625UQhhL+AvQFdg1+KUVrshQ9LPkSNTXqI8b74Ju+4K48aln+ecAxtumHVVkiSphhjjCGBEjXUFJ9uOMR5ejJokSVLpKmbPpPpM/kiM8Z4YY19gT+CsWndWpAkg+/WDLl2cN2kxn34KO+0ECxbAU0/Bgw8aJEmSJEmS1AwUM0yqz+SP34oxPgP0DCF0ruX2okwAWT1v0tNPp3mTBEyZkoKkWbPgkUcWnvZOkiRJkiSVvWKGSd9O/hhCaE2a/PH+/A1CCOuHkKa5DiFsArQGvipijQUNGZI64nz8cdaVNAFffw3DhsEHH8B998FGG2VdkSRJkiRJKqKizZkUY6wMIVRP/lgBXF09+WPu9kuBHwKHhhDmA18D++VNyJ2Z6o43Tz8N665b56blbcECOPhgeP55uPVWeyRJkiRJktQMFXMC7iVO/hhjPBc4t5g11Uf//rDqqmkS7sMPz7qajMQIv/wl3H13Omvbj36UdUWSJEmSJCkDxRzmVrJatFg4b1Kzdc45cPHF8Otfp1BJkiRJkiQ1S4ZJ9TRkSJoz6ZNPsq4kA9ddB7/7HRx4IJzb5DqOSZIkSZKkIjJMqqchQ9LPZtc76ZFH4KijYNtt4ZprUjctSZIkSZLUbJkM1NNGG8EqqzSzMOmVV2CffWDDDdNcSa1bZ12RJEmSJEnKmGFSPbVoAVtvnSbhbhY++AB23RU6d4YRI6Bjx6wrkiRJkiRJTYBh0lIYOhQ+/BDGj8+6kkY2cSLstBNUVsKjj8Iaa2RdkSRJkiRJaiIMk5ZCs5g3afZs2G03+OwzePBB6NMn64okSZIkSVITYpi0FL7zHVhppTIe6jZ/PvzoRzBqFNx6K2y+edYVSZIkSZKkJqZl1gWUkoqKNG9SWfZMihGOPTbNj3TZZbDHHllXJEmSJEmSmiB7Ji2lIUPg/fdhwoSsK2lgp58OV18Np50GxxyTdTWSJEmSJKmJMkxaSkOHpp9l1TvpssvgrLPgxz+GM87IuhpJkiRJktSEGSYtpe9+Fzp2LKMw6b774LjjYNdd4dJLIYSsK5IkSZIkSU2YYdJSqqiALbcsk0m4X3gB9t8fBg2C226Dlk6hJUmSJEmS6maYtAyGDoUxY+CLL7KuZDm8+y7svjusvTY8+CCsuGLWFUmSJEmSpBJgmLQMhgxJP0t2qNuECbDjjtCqFTzyCHTpknVFkiRJkiSpRBgmLYNNNoH27Us0TJo+HXbZBaZMgREjYL31sq5IkiRJkiSVECfJWQYtW6Z5k0ouTPrmG9h7bxg9Gh56KKVikiRJkiRJS8GeSctoyBB4+22YODHrSuqpqgqOOAKefBKuvhp22CHriiRJkiRJUgkyTFpGQ4emn888k2kZ9XfSSXDLLXDOOXDIIVlXI0mSJEmSSpRh0jIaODCdAG3kyKwrqYe//x3OOw+GD0+hkiRJkiRJ0jIyTFpGrVrB979fAvMm3XYbnHgi7LMPXHABhJB1RZIkSZIkqYQZJi2HIUPgrbdg8uSsKylg7ly4/HI49FDYaiu44QaoqMi6KkmSJEmSVOIMk5ZDk5w3aeJE+OMfYZ114Cc/SWdsu+8+WGGFrCuTJEmSJEllwDBpOQwaBG3bNpGhbm+/DUcfDd27wxlnwPe+B088AS+8AKusknV1kiRJkiSpTLTMuoBS1ro1bLFFhpNwxwiPPw7nnw+PPJJ6Hx1+OBx/PPTtm1FRkiRJkiSpnNkzaTkNHQpvvglTphTxQefOhWuuge98B3bYAV57Dc46C8aNg0svNUiSJEmSJEmNxjBpOQ0ZkjoIPftsER5s0iQ488w0H9KRR0KLFnDttfDxx/D730PnzkUoQpIkSZIkNWeGSctp8OA0uqxRh7q9/TYcc0yaD+n009NkTY8/nnokHXYYtGnTiA8uSZIkSZK0kHMmLac2bWDzzRthEu4Y0wTa558PDz+cEqtDD03zIfXr18APJkmSJEmSVD/2TGoAQ4akTkJTpzbAzr75Js2HNGAAbL89jBqVhrZ9+ilcdplBkiRJkiRJypRhUgMYOjR1JHruueXYyaRJaRLt6vmQAK6+Gj75BP7wB+jSpSFKlSRJkiRJWi4Oc6tNZWWaVbtly7S0alX4csuWbLpuK1Zv3ZIX/92S3bfLuy2EJT/OO+/ABRfA9dens7TtvDOceCJsu2397i9JkiRJklREhkm1mT0bttmmXpuuAHwOcFFuqVZRsUjotFgg1aIFfPBBmnipej6k/v0b/FAkSZIkSZIaimFSbdq1S6doq6yE+fPTz+ol/3ru8oj7K3ni35X86fT5tG1Z97aLXD/8cPjJTxzGJkmSJEmSSoJhUm1atUoza9fTCn3h/Edhm0Gw666NWJckSZIkSVKGnIC7gWy2GbRuDU8/nXUlkiRJkiRJjccwqYG0aweDBxsmSZIkSZKk8maY1ICGDoVRo2DmzKwrkSRJkiRJahyGSQ1oyBBYsACefz7rSiRJkiRJkhqHYVID2nxzaNkynQROkiRJkiSpHBkmNaAVV3TeJEmSJEmSVN4MkxrYkCHwyiswa1bWlUiSJEmSJDU8w6QGNnQoVFbCCy9kXYkkSZIkSVLDM0xqYFtsARUVDnWTJEmSJEnlyTCpgbVvD4MGOQm3JEmSJEkqT4ZJjWDoUHj5ZZgzJ+tKJEmSJEmSGpZhUiMYMgTmz4f//CfrSiRJkiRJkhpWUcOkEMJOIYQxIYSxIYRTCtx+UAjhjdzyQghhQDHrayjf/36aN8mhbpIkSZIkqdwULUwKIVQAFwM7A/2BA0II/Wts9hEwJMb4HeAs4PJi1deQOnaETTZxEm5JkiRJklR+itkzaTAwNsb4YYxxHnArMCx/gxjjCzHGqbmrLwLdilhfgxoyBF56Cb7+OutKJEmSJEmSGk4xw6S1gHF518fn1tXmx8DDjVpRIxo6FObNgxdfzLoSSZIkSZKkhlPMMCkUWBcLbhjCD0hh0sm17iyEY0IIr4QQXpk0aVIDldhwttwSWrRw3iRJkiRJklReihkmjQfWzrveDZhQc6MQwneAK4FhMcavattZjPHyGOOgGOOgLl26NHixy2ullWDjjZ03SZIkSZIklZdihkkvA71CCOuGEFoD+wP3528QQugO3A0cEmN8r4i1NYohQ9Iwt7lzs65EkiRJkiSpYRQtTIoxVgLDgUeBd4DbY4yjQwjHhhCOzW12GrAqcEkI4bUQwivFqq8xDBkC33yTJuKWJEmSJEkqBy2L+WAxxhHAiBrrLs27fBRwVDFrakxbbQUhpKFuQ4ZkXY0kSZIkSdLyK+Ywt2ZnlVVgwAAn4ZYkSZIkSeXDMKmRDR0K//lPGu4mSZIkSZJU6gyTGtmQIWkC7pdfzroSSZIkSZKk5WeY1Mi22ir9dKibJEmSJEkqB4ZJjWzVVeE730mTcEuSJEmSJJU6w6QiGDIEnn8e5s3LuhJJkiRJkqTlY5hUBEOHwtdfwyuvZF2JJEmSJEnS8jFMKoKtt04/HeomSZIkSZJKnWFSEXTuDBts4CTckiRJkiSp9BkmFcnQoWnepPnzs65EkiRJkiRp2RkmFcmQITB7NowalXUlkiRJkiRJy84wqUiGDEk/nTdJkiRJkiSVMsOkIunaFfr1M0ySJEmSJEmlzTCpiIYMgWefhcrKrCuRJEmSJElaNoZJRTR0KMyaBf/7X9aVSJIkSZIkLRvDpCKqnjdp5MhMy5AkSZIkSVpmhklFtPrq0KeP8yZJkiRJkqTSZZhUZNXzJi1YkHUlkiRJkiRJS88wqciGDIEZM+C117KuRJIkSZIkaekZJhVZ9bxJDnWTJEmSJEmlyDCpyNZaC9Zf30m4JUmSJElSaTJMysDQoc6bJEmSJEmSSpNhUgaGDIFp0+DNN7OuRJIkSZIkaekYJmWget4kh7pJkiRJkqRSY5iUgbXXhvXWcxJuSZJUHCGEnUIIY0IIY0MIpxS4/aAQwhu55YUQwoAs6pQkSaXBMCkjQ4bAM89AVVXWlUiSpHIWQqgALgZ2BvoDB4QQ+tfY7CNgSIzxO8BZwOXFrVKSJJUSw6SMDB0KU6bAW29lXYkkSSpzg4GxMcYPY4zzgFuBYfkbxBhfiDFOzV19EehW5BolSVIJMUzKSPW8SQ51kyRJjWwtYFze9fG5dbX5MfBwo1YkSZJKmmFSRtZZB3r0gDvucKibJElqVKHAulhwwxB+QAqTTq7l9mNCCK+EEF6ZNGlSA5YoSZJKiWFShk45BZ59Fs46K+tKJElSGRsPrJ13vRswoeZGIYTvAFcCw2KMXxXaUYzx8hjjoBjjoC5dujRKsZIkqekzTMrQMcfAYYfBH/8II0ZkXY0kSSpTLwO9QgjrhhBaA/sD9+dvEELoDtwNHBJjfC+DGiVJUgkxTMpQCPCvf8GAAXDQQfDhh1lXJEmSyk2MsRIYDjwKvAPcHmMcHUI4NoRwbG6z04BVgUtCCK+FEF7JqFxJklQCWmZdQHPXti3cdRcMHAg//CE8/zy0a5d1VZIkqZzEGEcAI2qsuzTv8lHAUcWuS5IklSZ7JjUB660HN90Er70GP/0pxIJTYkqSJEmSJGXPMKmJ2GUXOP10uP56uOyyrKuRJEmSJEkqzDCpCTntNNh5Z/jFL+Cll7KuRpIkSZIkaXGGSU1IixZw443QrRvssw9MnJh1RZIkSZIkSYsyTGpiOnVKE3JPngz77w+VlVlXJEmSJEmStJBhUhO08cZw6aXw1FPw+99nXY0kSZIkSdJChklN1GGHwbHHwrnnwt13Z12NJEmSJElSYpjUhF1wAQweDIcfDmPGZF2NJEmSJEmSYVKT1qYN3Hln+rn33jBrVtYVSZIkSZKk5s4wqYlbe2249VZ491348Y8hxqwrkiRJkiRJzZlhUgnYdlv485/h9tvT0DdJkiRJkqSsGCaViJNOgr32gt/8Bp55JutqJEmSJElSc2WYVCJCgGuvhZ494Uc/ggkTsq5IkiRJkiQ1R0UNk0IIO4UQxoQQxoYQTilwe98Qwn9CCN+EEH5dzNpKQceOcPfdaSLuffeFefOyrkiSJEmSJDU3RQuTQggVwMXAzkB/4IAQQv8am00BfgH8rVh1lZoNNoCrroIXXkhD3iRJkiRJkoqpmD2TBgNjY4wfxhjnAbcCw/I3iDFOjDG+DMwvYl0lZ7/94IQT4MIL4eabs65GkiRJkiQ1J8UMk9YCxuVdH59bp2Vw7rmw1VZw9NHw5ptZVyNJkiRJkpqLYoZJocC6uMw7C+GYEMIrIYRXJk2atBxllaZWreD222GllWDvvWH69KwrkiRJkiRJzUExw6TxwNp517sBy3xOshjj5THGQTHGQV26dFnu4krR6qvDHXfAxx/DoYdCVVXWFUmSJEmSpHJXzDDpZaBXCGHdEEJrYH/g/iI+fln6/vfh/PPh/vvhnHOyrkaSJEmSJJW7lsV6oBhjZQhhOPAoUAFcHWMcHUI4Nnf7pSGE1YFXgI5AVQjheKB/jHFGseosRcOHw4svwu9/D4MGwQ47ZF2RJEmSJEkqV0ULkwBijCOAETXWXZp3+QvS8DcthRDg8svhjTfgwANh1ChYZ52sq5IkSZIkSeWomMPc1IhWXBHuvhvmz4d99oG5c7OuSJIkSZIklSPDpDLSqxdcfz288gr84hdZVyNJkiRJksqRYVKZGTYMfvc7uOIKuOqqrKuRJEmSJEnlxjCpDJ15Jmy3HfzsZ2n+JEmSJEmSpIZimFSGKirglltgtdVg113hkkucQ0mSJEmSJDUMw6Qy1bkzPPggrLde6qG03nrw97/DnDlZVyZJkiRJkkqZYVIZ22gjeP55eOIJ6NMHTjwRevSAc86BGTOyrk6SJEmSJJUiw6QyFwJssw089RQ8+ywMHAi//W0Klf74R5g6NesKJUmSJElSKTFMaka23BIefhj++1/Yems44wxYZ5109rdJk7KuTpIkSZIklQLDpGboe9+De++F116DnXZKw9569IBf/Qo+/zzj4iRJkiRJUpNmmNSMDRgAt98Oo0fD3nvDBRfAuuvC8OHw6adZVydJkiRJkpoiwyTRrx/ccAOMGQMHHwyXXQbrrw9HHw0ffph1dZIkSZIkqSkxTNK31l8frrwSPvggBUk33AC9e8Ohh8K772ZdnSRJkiRJagoMk7SY7t3h4otTr6Rf/ALuvBP694f99oM33si6OkmSJEmSlCXDJNVqzTXh/PPh44/h5JNhxIg0z9Kee8Irr2RdnSRJkiRJyoJhkpaoa1f4y1/gk0/g9NPh6afTGeF23hmeew5izLpCSZIkSZJULIZJqrdOneCMM1Ko9Oc/p95JW22V5lr69a9TsLRgQdZVSpIkSZKkxmSYpKXWsSP89rdp+NsVV0CfPnDhhSlYWnNNOOYYePhh+OabrCuVJEmSJEkNzTBJy2zFFeGoo9JcSpMnwy23wNCh6ecuu0CXLrD//nDbbTBzZtbVSpIkSZKkhmCYpAbRsePC4GjSJHjooXT2tyefTOs7d4Zdd4Urr4SJE7OuVpIkSZIkLSvDJDW4FVZIPZOuuAI+/xyeeQZ+9jN4+204+mhYYw3Yeut0priPPsq6WkmSJEmStDQMk9SoKirSXErnnw8ffgj/+x/84Q8wfTr86lew3nrw3e/CH/8Ib7zhmeEkSZIkSWrqDJNUNCGk4OiMM+D112HsWPjb36B9+xQmDRjgmeEkSZIkSWrqDJOUmZ49U++k555Lw+Euv7zwmeFuuy2dOc5eS5IkSZIkZa9l1gVIAKutluZTOvro/2/vbmPkqs4Djv8f7669CzY2+AVccAxBNgkVausQQ9+SSIUKrCpu+6FxVAmaVkFEIDUfKhUpUpSPTas2aioEgpYEqqiJopbiIJOkaio1igSxTQDjEhsHcGxsDDHExrGd9eLTD+eOZnb2zu7send2597/Tzqa+3Jm9zw+e9mHZ8+9AydP5k+Ie/zx/MlwDz+c+6xZAzfdlNvmzfDhD8OKFfM6bEmSJEmSasdikhacxifDbdsG587Bnj3wzDPN9q1vNft+4APNAtNNN8ENN8DQ0PyNXZIkSZKkqrOYpAVtaAg2bcrtM5/Jx06cgJ07m8Wlp56CRx/N54aHc9/WAtP69fl5TZIkSZIk6cJZTFLfWb4cbrklN8jPUjp4cPzqpQcegC99KZ9fsybfFtcoLnl7nCRJkiRJM2cxSX0vAq6+OrdPfCIfK7s97sknm+9p3B63eXNeyXTddXDppfMxekmSJEmS+ovFJFVS2e1xP/95vj3uhz+ceHscwKpVsHFjLixt3NjcvvbafPucJEmSJEmymKQaWbECbr01N2jeHrdnD+zfD/v25ddvfxu+8pXm+yLyc5fai0wbN8K6dbBo0byEI0mSJEnSvLCYpNpqvT2u3cmT8PLL44tM+/fDD34Ap041+w0Pw4YNE4tMGzfCypW9ikSSJEmSpN6xmCSVuOQS+NCHcmuVErzxxvgi0759eXXTE0/A2Fiz78qVuai0YUNewfS+9+XXRlu+vLcxSZIkSZI0GywmSdMQAWvX5vbRj44/d+4cvPpqcxVTo9D0ve/BkSNw/vz4/suWjS8utReb1q2DkZHexSZJkiRJUjcsJkmzZGioeYtbu7ExOHoUDh1qtp/+tLn9ox/Bm29OfN/KlZMXnK68Mn9fSZIkSZJ6xWKS1AODg80CUCdnz8Lrr48vODWKTq+9Bt//fv5EulYRcMUVuV1+OaxZk1/LtlevzuOQJEmSJOlC+L+W0gIxPAzXXptbJ6dOTSw2HToEx47ltndvfh0dnfjeiLzSqVPBqX1/eHjuYpUkSZIk9S+LSVIfWboUPvjB3DpJKX8a3bFj+da5RqGpfXvXrrz97rvlX2fZsmZhadUquOyy8rZyZXN76dJctJIkSZIkVZfFJKliIvInxS1fXv78pnZnzjQLTZ2KT6++Crt3w9tvw+nTnb/W4GDnolN74am1XXIJLFo0e/8GkiRJkqS5YzFJqrmREVi/PrdunD0L77yTC0vHj+fXTu3112HPnrzdaQUU5ALYsmW5AHbJJTN/HR52ZZQkSZIkzTWLSZKmZXgY1q7NbTpGR/MDxMsKUO+8k2/NO3Gi+Xr8OLzySvPYmTNTf4+hoWZxqbXQtGwZXHwxXHRR5zbZ+ZERGBiY0T+XJEmSJFWOxSRJPbF4cX7A95o1M3v/uXMTC05TvZ44AQcP5v3Tp3P7xS/g/Pnpf/8lSyYvRi1dmotW3balS/P7XEklSZIkqd9YTJLUF4aG8jOXVq68sK+TUi5MNYpLs9GOH8+ftPfuu7lN9lypVosWdV+EGhnJ/waNtnjx+P3pHGs/bkFLkiRJ0nRYTJJUKxG5kLJ4MaxYMTff4733xheXptveemv8/ujo3IyzYWAgr7waHm62kZHO+zM5t2RJLlwNDk7dBgZys8glSZIkLUwWkyRplg0MNJ/bNBtGR/ODz8+da7bR0fH70zlWdvyXv8ztzJn8vc6ebW6fOZOfa9V6rvX8XBkY6L74NDhYvuqqfWVWp3PdbE+20muqVWAWxyRJklQlFpMkaYFrrKRaiFLKhalOhabW7bGxvGprbKz7Np3+rcWxM2fys7LKimfthbS5XvnVMFUBanAwF5waDcbvlx2bbp/Gqq/WFWCdjs1k/4Yb4JZbevPvKUmSpPljMUmSNGMR+Ra2JUvmeyQzl1IuWk1VeOp2lddMV4+NjeWxpNQcV2trPzbdPufP5+/13nvNIl1ju2x/qj5lPv1pi0mSJEl10NNiUkTcBvwjMAD8c0rpb9rOR3F+C3Aa+LOU0rO9HKMkqV4imrfMjYzM92j6x/nzE4tNg/6JSpIkqRZ6lvZFxABwP3ArcBjYGRHbU0r/19LtdmBD0W4CHiheJUnSArJo0cK9/VKSJElza1EPv9dm4EBK6ZWU0ijwdWBrW5+twGMpexpYERFrezhGSZIkSZIkTaKXxaQrgUMt+4eLY9PtI0mSJEmSpHnSy2JS2Ycipxn0yR0j7oqIXRGx66233rrgwUmSJEmSJGlqvSwmHQbWtexfBRyZQR8AUkoPpZRuTCnduHr16lkdqCRJkiRJksr1spi0E9gQEddExGJgG7C9rc924I7IbgZOpJSO9nCMkiRJkiRJmkTPPs0tpTQWEfcC3wEGgEdSSnsj4u7i/IPADmALcAA4DXyqV+OTJEmSJEnS1HpWTAJIKe0gF4xajz3Ysp2Ae3o5JkmSJEmSJHWvl7e5SZIkSZIkqc9ZTJIkSZIkSVLXLCZJkiRJkiSpaxaTJEmSJEmS1DWLSZIkSZIkSeqaxSRJkiRJkiR1zWKSJEmSJEmSumYxSZIkSZIkSV2zmCRJklRxEXFbROyLiAMRcV/J+YiILxfnX4iITfMxTkmS1B8sJkmSJFVYRAwA9wO3A9cDn4yI69u63Q5sKNpdwAM9HaQkSeorFpMkSZKqbTNwIKX0SkppFPg6sLWtz1bgsZQ9DayIiLW9HqgkSeoPFpMkSZKq7UrgUMv+4eLYdPtIkiQBMDjfA5gNu3fv/llEHJyjL78K+Nkcfe2FqG7xgjHXQd3iBWOug7rFu36+B9DHouRYmkEfIuIu8m1wAKciYt8Fjq2Tuv18gzHXQd3iBWOug7rFC/WLuTQHq0QxKaW0eq6+dkTsSindOFdff6GpW7xgzHVQt3jBmOugbvHqghwG1rXsXwUcmUEfUkoPAQ/N9gDb1fHn25irr27xgjHXQd3ihXrGXMbb3CRJkqptJ7AhIq6JiMXANmB7W5/twB3Fp7rdDJxIKR3t9UAlSVJ/qMTKJEmSJJVLKY1FxL3Ad4AB4JGU0t6IuLs4/yCwA9gCHABOA5+ar/FKkqSFz2LS1OZ8KfcCU7d4wZjroG7xgjHXQd3i1QVIKe0gF4xajz3Ysp2Ae3o9rknU8efbmKuvbvGCMddB3eKFesY8QeTcQZIkSZIkSZqaz0ySJEmSJElS1ywmARFxW0Tsi4gDEXFfyfmIiC8X51+IiE3zMc7ZEhHrIuJ/IuKliNgbEX9Z0udjEXEiIp4r2ufnY6yzKSJei4g9RTy7Ss5XbZ6va5m/5yLiZER8tq1PX89zRDwSEW9GxIstxy6LiP+KiJeL10s7vHfS636h6hDz30XEj4uf28cjYkWH9056DSxUHWL+QkS83vKzu6XDe/tunjvE+42WWF+LiOc6vLcv51j1ZQ5mDlacr8w81yH/AnOwlmOVzcHqln+BOdi0pZRq3cgPovwJ8H5gMfA8cH1bny3AU0AANwPPzPe4LzDmtcCmYnsZsL8k5o8BT873WGc57teAVZOcr9Q8t8U2ALwBrK/SPAMfATYBL7Yc+1vgvmL7PuCLHf49Jr3uF2rrEPPvA4PF9hfLYi7OTXoNLNTWIeYvAH81xfv6cp7L4m07//fA56s0x7Z6NnMwc7CqznNLXJXMv4oYzMFStXOwuuVfnWJuO28O1tJcmQSbgQMppVdSSqPA14GtbX22Ao+l7GlgRUSs7fVAZ0tK6WhK6dli+13gJeDK+R3VglCpeW7ze8BPUkoH53sgsyml9L/A222HtwKPFtuPAn9Y8tZurvsFqSzmlNJ3U0pjxe7TwFU9H9gc6jDP3ejLeZ4s3ogI4E+Af+vpoKS5YQ5mDtZQqXluUcn8C8zBWo5VNgerW/4F5mDTZTEp/wI/1LJ/mIm/1Lvp05ci4mrgN4BnSk7/ZkQ8HxFPRcSv9nZkcyIB342I3RFxV8n5ys4zsI3O/+Gr2jxfnlI6CjlpB9aU9KnyXP85+a+7Zaa6BvrNvcWy8kc6LKWv4jz/LnAspfRyh/NVm2NVmzmYOVhDVee5TvkXmIPVJQerY/4F5mATWEzKy2nbtX/EXTd9+k5ELAX+HfhsSulk2+lnyUtyfw34J+A/ezy8ufDbKaVNwO3APRHxkbbzVZ3nxcDHgW+WnK7iPHejqnP9OWAM+FqHLlNdA/3kAeBa4NeBo+Rlx+2qOM+fZPK/iFVpjlV95mDmYA2Vm2fzr44qN9dQqxysrvkXmINNYDEpV0rXtexfBRyZQZ++EhFD5CTmayml/2g/n1I6mVI6VWzvAIYiYlWPhzmrUkpHitc3gcfJSzBbVW6eC7cDz6aUjrWfqOI8A8caS+OL1zdL+lRuriPiTuAPgD9NKZX+wu7iGugbKaVjKaX3UkrngYcpj6VS8xwRg8AfA9/o1KdKc6xaMAczB2uo3DxTv/wLzMEqn4PVMf8Cc7BOLCbBTmBDRFxT/AVhG7C9rc924I7IbgZONJZw9qPifs9/AV5KKf1Dhz5XFP2IiM3kn5XjvRvl7IqIiyNiWWOb/LC8F9u6VWqeW3SsoldtngvbgTuL7TuBJ0r6dHPd942IuA34a+DjKaXTHfp0cw30jbZnafwR5bFUap6BW4Afp5QOl52s2hyrFszByvtU6ndzjXOwuuVfYA5W+RyspvkXmIOV6/Rk7jo18idI7Cc/df5zxbG7gbuL7QDuL87vAW6c7zFfYLy/Q15q+ALwXNG2tMV8L7CX/PT9p4Hfmu9xX2DM7y9ieb6Iq/LzXMR0ETk5Wd5yrDLzTE7SjgLnyH8F+QtgJfDfwMvF62VF318BdrS8d8J13w+tQ8wHyPemN67nB9tj7nQN9EPrEPO/FtfpC+QEZW1V5rks3uL4VxvXbkvfSsyxrb6t7Bqt8u9mzMFqkYNR8fyriMEcrOI5WId4K5t/dYq5OP5VzMEmtCiClyRJkiRJkqbkbW6SJEmSJEnqmsUkSZIkSZIkdc1ikiRJkiRJkrpmMUmSJEmSJElds5gkSZIkSZKkrllMkiRJkiRJUtcsJkmSJEmSJKlrFpMkSZIkSZLUtf8HHUndBhqNb1QAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1440x2880 with 8 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "rows = len(args.datasets) * len(args.train_column)\n",
    "\n",
    "fig, axs = plt.subplots(rows, 2, figsize=(10 * 2, 10 * rows))\n",
    "\n",
    "for i, col in enumerate(args.train_column):\n",
    "    for j, ds in enumerate(args.datasets):\n",
    "        idx = i * len(args.datasets) + j\n",
    "\n",
    "        axs[idx, 0].set_title(f\"Loss values for '{ds}' '{col}'\")\n",
    "        axs[idx, 0].plot([*range(args.num_epochs)], train_losses[col][ds], color = 'b', label='Train loss')\n",
    "        axs[idx, 0].plot([*range(args.num_epochs)], valid_losses[col][ds], color = 'r', label='Valid loss')\n",
    "        axs[idx, 0].legend(loc=\"upper center\", bbox_to_anchor=(0.0, 1.1), ncol=1)\n",
    "\n",
    "        axs[idx, 1].set_title(f\"Accuracy values for '{ds}' '{col}'\")\n",
    "        axs[idx, 1].axis(ymin = 0, ymax = 1)\n",
    "        axs[idx, 1].plot([*range(args.num_epochs)], train_accuracies[col][ds], color = 'b', label='Train accuracy')\n",
    "        axs[idx, 1].plot([*range(args.num_epochs)], valid_accuracies[col][ds], color = 'r', label='Valid accuracy')\n",
    "        axs[idx, 1].legend(loc=\"upper center\", bbox_to_anchor=(0.0, 1.1), ncol=1)\n",
    "    \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "61f0821e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9bf7c07e7d1146c4b699a16b3e76edce",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'lemmatized long_question' test:   0%|          | 0/330 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ed8ce7931a9e4dbe8266a07dc7f6cc71",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'lemmatized short_question' test:   0%|          | 0/330 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "91d0d55f0f934065a6782a0d7686ff02",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'lemmatized_filtered long_question' test:   0%|          | 0/330 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "68981d2b45b04a12934271b2df6e060f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "'lemmatized_filtered short_question' test:   0%|          | 0/330 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "predictions = {}\n",
    "test_target = {}\n",
    "\n",
    "for ds in args.datasets:\n",
    "    predictions[ds] = {}\n",
    "    test_target[ds] = {}\n",
    "    \n",
    "    for tc in args.train_column:\n",
    "        _, _, test_iterator = iterators[tc][ds]\n",
    "    \n",
    "        question_field = fields[tc][ds][0][1]\n",
    "        input_dim = len(question_field.vocab)\n",
    "        output_dim = len(target_names)\n",
    "    \n",
    "        model = LSTMClassifier(input_dim, args.embedding_dim, args.hidden_dim, output_dim, question_field)\n",
    "    \n",
    "        test_bar = notebook.tqdm(desc=f\"'{ds} {tc}' test\", total=len(test_iterator), position=0, leave=True)\n",
    "\n",
    "        model.load_state_dict(torch.load(args.model_state_file + \"_\" + \\\n",
    "                                                            ds + \"_\" + \\\n",
    "                                                            tc + \".pth\"))\n",
    "        model = model.to(device)\n",
    "\n",
    "        _, _, predictions[ds][tc] = evaluate_model(model, test_iterator, optimizer, criterion, test_bar)\n",
    "    \n",
    "        temp = []\n",
    "        for batch in test_iterator:\n",
    "            temp += batch.main_category.tolist()\n",
    "        \n",
    "        test_target[ds][tc] = [fields[tc][ds][1][1].vocab[val] for val in temp]    \n",
    "        predictions[ds][tc] = [fields[tc][ds][1][1].vocab[val] for val in predictions[ds][tc]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "9acada35",
   "metadata": {},
   "outputs": [],
   "source": [
    "accuracy = {}\n",
    "\n",
    "for tc in args.train_column:\n",
    "\n",
    "    accuracy[tc] = {}\n",
    "\n",
    "    for ds in args.datasets:\n",
    "        pred = predictions[ds][tc]\n",
    "        accuracy[tc][ds] = round(accuracy_score(test_target[ds][tc], pred), 3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "6afc8993",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 792x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1, 1, figsize=(11, 5))\n",
    "\n",
    "acc_df = pd.DataFrame({f\"{tc}\": accuracy[tc].values() for tc in args.train_column}, index=args.datasets)        \n",
    "acc_df.plot(kind='bar', ax=ax)\n",
    "\n",
    "ax.axis(ymin=0.8, ymax = 1)\n",
    "\n",
    "for tick in ax.get_xticklabels():\n",
    "    tick.set_rotation(0)\n",
    "\n",
    "for label in ax.containers:\n",
    "    ax.bar_label(label)\n",
    "    \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "074e9c5f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x1152 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(len(args.datasets), len(args.train_column), figsize=(8 * len(args.train_column), 8 * len(args.datasets)))\n",
    "\n",
    "for i, dataset in enumerate(args.datasets):\n",
    "    for j, train_column in enumerate(args.train_column):\n",
    "        axs[i, j].set_title(f\"Confusion matrix for '{dataset}', '{train_column}'\")\n",
    "    \n",
    "        pred = predictions[dataset][train_column]\n",
    "    \n",
    "        cm = confusion_matrix(test_target[dataset][train_column], pred)\n",
    "        cm_df = pd.DataFrame(cm, index=target_names, columns=target_names)\n",
    "    \n",
    "        heatmap = sn.heatmap(cm_df, annot=True, ax=axs[i, j], cmap='Reds', fmt='g', annot_kws={\"size\": 15}, cbar=False)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "13488a58",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification report for 'lemmatized' 'long_question':\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "        Egészség       0.96      0.97      0.97      7648\n",
      "      Szórakozás       0.92      0.91      0.92      5474\n",
      "         Állatok       0.97      0.97      0.97      7022\n",
      "Számítástechnika       0.88      0.92      0.90      2889\n",
      "\n",
      "        accuracy                           0.95     23033\n",
      "       macro avg       0.94      0.94      0.94     23033\n",
      "    weighted avg       0.95      0.95      0.95     23033\n",
      "\n",
      "============================================================\n",
      "Classification report for 'lemmatized' 'short_question':\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "        Egészség       0.90      0.94      0.92      7648\n",
      "      Szórakozás       0.87      0.92      0.90      5474\n",
      "         Állatok       0.95      0.90      0.93      7022\n",
      "Számítástechnika       0.90      0.82      0.86      2889\n",
      "\n",
      "        accuracy                           0.91     23033\n",
      "       macro avg       0.91      0.90      0.90     23033\n",
      "    weighted avg       0.91      0.91      0.91     23033\n",
      "\n",
      "============================================================\n",
      "Classification report for 'lemmatized_filtered' 'long_question':\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "        Egészség       0.97      0.96      0.96      7648\n",
      "      Szórakozás       0.91      0.94      0.93      5474\n",
      "         Állatok       0.96      0.97      0.97      7022\n",
      "Számítástechnika       0.93      0.89      0.91      2889\n",
      "\n",
      "        accuracy                           0.95     23033\n",
      "       macro avg       0.94      0.94      0.94     23033\n",
      "    weighted avg       0.95      0.95      0.95     23033\n",
      "\n",
      "============================================================\n",
      "Classification report for 'lemmatized_filtered' 'short_question':\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "        Egészség       0.90      0.95      0.92      7648\n",
      "      Szórakozás       0.91      0.91      0.91      5474\n",
      "         Állatok       0.95      0.91      0.93      7022\n",
      "Számítástechnika       0.90      0.85      0.87      2889\n",
      "\n",
      "        accuracy                           0.92     23033\n",
      "       macro avg       0.91      0.91      0.91     23033\n",
      "    weighted avg       0.92      0.92      0.92     23033\n",
      "\n",
      "============================================================\n"
     ]
    }
   ],
   "source": [
    "for ds in args.datasets:\n",
    "    for tc in args.train_column:\n",
    "        pred = predictions[ds][tc]\n",
    "        report = classification_report(test_target[ds][tc], pred, target_names=target_names)\n",
    "        print(f\"Classification report for '{ds}' '{tc}':\\n{report}\\n{'=' * 60}\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}