count_vectorizer_feature.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import string, re, json, pickle, warnings\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from nltk.corpus import stopwords\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "from sklearn.feature_extraction.text import TfidfTransformer, CountVectorizer\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.metrics import confusion_matrix,accuracy_score, classification_report\n",
    "# importing the wordCloud\n",
    "from wordcloud import WordCloud, STOPWORDS, ImageColorGenerator\n",
    "from bidi.algorithm import get_display\n",
    "import arabic_reshaper\n",
    "from data_preparation import take_a_shower, clean_content_aravec\n",
    "\n",
    "%matplotlib inline\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>Comment</th>\n",
       "      <th>is_off</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>شوف حد ين*** أنت و إيران.</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>شوف تيفي قوداتها مع الجن غي بلاتي يجي من مكة</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>عجب يتفاوظون على شعوب عربية وعليهم وهىلا تعلم</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ههههههه هاد شوف تيفي اللي حطت عليه مشا فيها</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>الرزق يأتي من حيث لا تدري 🙌🏻❤️</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         Comment  is_off\n",
       "0                      شوف حد ين*** أنت و إيران.       1\n",
       "1   شوف تيفي قوداتها مع الجن غي بلاتي يجي من مكة       0\n",
       "2  عجب يتفاوظون على شعوب عربية وعليهم وهىلا تعلم       0\n",
       "3    ههههههه هاد شوف تيفي اللي حطت عليه مشا فيها       0\n",
       "4                الرزق يأتي من حيث لا تدري 🙌🏻❤️        0"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv('./data/offensive_dataset.csv', encoding=\"utf-16\")\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 576x432 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(8,6))\n",
    "data.groupby('is_off').count().plot.bar(ylim=0)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Clean data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.Comment = data.Comment.apply(take_a_shower)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(14000, 2)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Word Cloud"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'شوف حد ايران شوف تيفي قوداتها الجن غي بلاتي يجي مكه عجب يتفاوظون شعوب عربيه وعليهم وهيلا تعلم هه هاد شوف تيفي اللي حطت مشا الرزق ياتي تدري 🙌🏻❤️ دخلو سوق كركم شوفو شكون مكتب عليكم تعيشو عبيد فقراء معا USERIDX لسسه امبارح بحذركو ولاد الجزمه فيفي ولاد شرموطه هه😂 Hadou jouj zeal hachak machi drari الله يعفو مسكينه دغيا طارو سصحاب الحسنات باش يستغلوها حرام عليهو الله يخد الحق شوف تفي عواد متسترها شهراتها لناس تفو شحاال حامضه هاد سعييده معرفت علاه كيجيبوها فين ميكون هاد لموسم ديال طنطان يجيبوها لينا يالطيف حد مكيحملها الطلب السعودي جاء ردا الاتهامات الاسراءيليه للسعوديه بقتل الاطفال باليمن اسوه بقتل اسراءيل للاطفال الفلسطينيين جاء تقرير المتحده بلاش فقره الاهداءات وادا لابد خليها نهايه الفيديو احسن وبالتوفيق ياعمر يااربي يااربي تلطف بينا نجم كندا هه روبيني الخرا هه اللهم اشغلهم بانفسهم واجعل كيدهم نحرهم واجعل تدبيرهم تدميرهم ودمرهم تدميرا ومزقهم شر ممزق قوي متين الله ماشي سيمحمد خصها تسميك سوسو المتوحشه 🤣🤣 Tomas Shelby mzhiha 😂 اين تعليقي ياجزيره اعتقد المنطق الكلام توجد اساء وانما تعقيب كلا'"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "comments = ' '.join(list(data.Comment))\n",
    "comments[0:1000]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV0AAAC1CAYAAAD86CzsAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOz9d5Rl2XXeCf7O9c+beOF9Zkb6rMysLO+rULAEQAAEQFI0IAmR1KhbbKk1Wuolac0sNdWtJXVrzYyoVlMSDQASIAEChEfBFFDeV2Wl9y68j+ffu/7MH/dlZESGSYMCNOqpr1atzHzm3vvOPXefffb+9reFlJJ38S7exbt4Fz8fKP+1L+BdvIt38S7+/wnvGt138S7exbv4OeJdo/su3sW7eBc/R7xrdN/Fu3gX7+LniHeN7rt4F+/iXfwcod3g/f/LUhvswEMAhqIhhPivfTk/E0TMFA+BBuK/7fU1kAF+6GMoxs/sfkkpcWwP3dAIghAZShAQhhIBmJb+39xc8UMPRahIJKpQfybn8MIAVSgo142NG/hoioLy3/jcu01sOFFuZHQ3hGt7zFyZpzhfIdeRpndbF6q6+eCGUuIFAaZ226e9bUgpCWSIKhSc0ON8dZqMHqfNTCEQxDXz535NPwtI6YN0QGiAjgyugMgh1Pb/2pcGQCBDSm4TU9FI6huP+aKzgCc9klqKul9DAk5gk9ASKEIhq2cBwYIzT1rPYKnWT39tfsipE5MMDBUoleo4tgeA3fSQwK7dPRiGhqqpKMrNG9+FhSq6ppLJxn/qa7xVXGlcotvqpeyV6In1AeCFTTRhIm7TGIZSUvdcLE1DSslMo0pHLIml6diBj+17JHSDyXoZS9XpTqQ3PJbjR4ZZVRSklEhAEQLb9zFUdY0h/78Cbsv6Nes2R549zcXjYyTScSbOT/N3//DTxJLrT/yq42CoKlLCmbl5DvR2/1QXfTuQSCpeg4rfpOo1KLp1skaC8cYCpqIxnOza8HtFZ5Ks0Y1yG55Co+li2y75XPLG1ygltbrD2QszLBUbdLan2DnSjWFo3GjuSRkADtKfRIZTgAlKGmQNRHBDoztRLNObTa/y5JquRyAlSdO4iV+6OUIpGasXeX3+CmW3yZM9OzY1unPOLJPNCZJaill7mg6zC1UoeKFHSEhGz6IKlYnGONtS29ma3PZTXyMChCLw/QDX8Vuer08iaWHbLp4X4LoBhqmRSKy+9iAIEQKUlvFo1B08PyCbTXDxwiyWqbP/4OBtXZaUPiAJpYsiTIRY/7GVrY2pQOBLH4FgpjlJTs8z78wuG93Z5jm6YjvRxO05Gn4YcqG8QNaK4YchFddmwa6jKgoCwYJdp82KE4QhinA2NLpuELBQr5ONxYjpOmXbxvF9CokE8/U6CV0nF4v9N7e7uBFu2eiGYcjZNy/z1T/6PslMnIOP72FxuoRmaNi+TxCGxHR91Qo1UarQlogR1w1mq7UbnqPacEjGDGzHY6FUJ5eOk4xvPEE8PyAMQ0xDX/d9KSVeGDDvVNCEgipUtia7iGsmFa+BpW5sVKQMOVn+CfcWPo1xE0Y3CEJq1SaxmIFh6nz7B0fxvIBf/9R9N/zu+OQSX/nmWxw/NUG1apPLJnj0we188qOHiMeuXaPvB7z61mWOnhhn945u7ju0BVOfQYZFECZC7QEpkLKGUPIIsbGncRUTxTI9mfQq4z5XraNryk9tdOfsKj+aPMOF6jztVooD+V7ardTy+3bg8ebCGNvS7XTFomtNaSl2p/fghz7tZjumYhHIAAFU/Sq6YmAoBrsye0hqN17QbgaappLNxrGbLr19eVRVIQxD4nETzw8I/JD5uQpdPVkAZChpNF0SCZPpqSKmqZNvS3L82Dj1mo2iKKiaguv46Nrtb+0rze8hpU8oqyTM+zH17QTSZ6IxTt5oI6VHYzbZGCOlZ8joWaabExiKSVJPoQgVdYWhbgRFQhnghw6SEF2J3dL1qIqgO5FGUxRCKUkbFjXXQSKxVJ2UbqIIgaVqhCsilA3P49kLl3ls6zCBDLlSLLLYaCKRaIpCwjBIGgZOEHB8ZobuVJpcbPW1hVJyZanIdKXKg8O3t4hBFDJanFoimUsQS9zcLimUPk1/HlPNorXGLJQBTX9u1Ws3wi0b3UbV5vUfHqOtO8sHP/Mo3/vcc2zZ149uaMyUyvzk3CUeHB6g4XkUEnEOT0xxZanEB3dtx9Q0NGXzLc1CscYff+VF9o30kIgZfOUHb/Px9+znPfftwNBXT1wpJQ3b5Rs/Pka17vCp9x+kLZtYfj8IQxzHx9BVNFWlL96GoehIQrTWJLQUHV2J/h7KkIu110jr7XSYWxBCQaDgBDW4rnJvZSVfs+5QXKyj6Sq27eHaHh3dGQxT5+z5WcYmFvnYhw6QXHFzpZRICUJEh754ZZ4/++KL6JrKr33yXgxD46VXL/C9Hx3HMnU+/bG7lr/34qsX+MKXX6FcafLqm5ewTIN7DlgIJYtQuxHCaH3WIQotrR5zLwhYqDXww5Bzs/Pc0deNqqp4YcDUUoVjkzPs7Iw8Y9sDMhvfr6vjsNIbuRrKCaVEV1ROl2ZwQ59PDB6gP5Elo6/2Xt5aHOd4cYrR2hLv7dlJu5VEFRqz9jTtVid91sByXDCQAW7oYComovW73klHqKsnSxhK4vHVsWNVUwj8kM7uDJYVLe6eH/DSC2cxTQ0Q7NjZje8HTE0WmZsr49g+AJ1dmTWe8a3A0IaBEIGOpnQCkTebM/LUgxoVv0yn2Y0vfZbceRJakppfRaGOqViI68KLmjCQhFS9RSSSnNFHNE+ieymR1P1FLlRfIKZm2Zp6EE0Yy+EIVSh0xVOrjtkZW7vwRfPg2nNydm6ez7/5NkP5LCOFNvqzWXrSafwwjK5LUVhsNPCCkEO9vcT1tU6U6/s8deYcY8Uy+7o7SVs3H1aSUtKoNKmW6mi6xt/8v5/iiV++nx13bVnXm772jMvWjvc0DX+azth9hFLHC+vUvQmq3hU6YnejCuumvPJbNrrxpMVjv3QPbz59nPFz0yBhdnQBgLrjkjB0SraNIgRvjk+xs6OAEIIvvnWEQ/09WNr63uhVFCtN5pdqNB2PLX0Ftg2009uRQdfWN9aVmsMrR68Qt3RK1eYqo/v26QlOnJvig4/sobMtRUK7eoOuGe/YiljuTPMsNW+JRWcMXVhkjE4UoUWewDpjWa/Z1Co241cWaDQcsrkEC7MVFEUhnY2TycE9dw5x+tw0bx4Z5bEHdyx/t9F0OX5qkpEtHczMVfjyN97gwN5+PvjkXuJxEwHcfXCIP/viSxw7Ob5sdMcml/jBMyf5zV+5n7hl8Mefe47nXznHjpEHyWUSq+J0YoPto+sHnJ2ZJ24avDU6xVBbDgF89/hZKk2blGVyebGIIgS7uzs2vFdSSpbcBooQZPTY8u6mGXh85fJhLtUW+O1t95HQDAxFpc1MrDG4AO1mgse6RtAVlalGGTvw6Y63kVIdTK0N8JEy+o6CitW6H0FQQlUSgL7mumzbQ1EEprnxfFu5jl69pFhsfa9eCIGmq2grFn5dV7n3vm0sLdZo70xjGBqnTkwwONTGY0/saiXiBIoiCMNrJ5MyJAgXCGUDISxUkUIIq3Xv1k40gcZi9T9jaANkE7+KSoYFZ542s52qX+Fc9SRvFV9FSsnB3D28svg8C84cUoaoQqPb6iOhXXsuVGHghg08aeMENRp+EVNN4EmHMPRpBEVMJUF//CDjjbd5feEv2Zl5kjZzaMOx3HDMVtxrxw+o2Daz1Rq7OzvIqmudqEzLiKpCrGvAQqDmuCzUG1RsZ12j67k+iqLgNB1iyWuGMAxCXvv+EUpzFR75pXtpVJt4rr/h9YfSJZAuzWCWqjtGIG0kktHqd1EVCz9soCtJVGFSdcdQzRiGmtrweFdxy0ZXURW23zlM93AH85NL7L5vhC/8q78FoOl5zFZr3NnfgyIE3zt1jkuLS/Rl0vz+A/eQsSwuLy2tHiAv4MrUItsG2hFCMDLYzj///Q/Qlomjqgp7tnUjBOveACEEnW0p/tnvvq9lpNtWvT85W+L1E6M8dGgrnW03HoyU3k5cy+FLFy+0Ga+foD+xrxXLFUgpqftLFN0pMloPM+M2INi9v59E0mJupkzvQBuFjmvb+UfuH+Enz5/mzbev8NC9I2iagucFvPDKeb71g6N88iOHeOPtK9x3aAvveXQXpqHh+yHlapPpmTKGrjI81I6UklK5yde+dZi2fJLd27vRdZXdO3o4eXaKSsUhn73xb5RSslhvEDcNMpbJYzu20JFKsrW9jbsGe2/4/VXHAp6eOktM1flA7y40JUp8HF2aZDCZJ2fG+c7ESX5v+4Ocrczxt6NH+OTgQbri18an7DbZkmpnzq4wWS9zZGkCQ9X41NBeRDiP6y2gCKt17Q6qkgU0hNDwghksfTuquGYopZScPzfDa69dZHCwwCOP7txwHKanS9RrDttGOtkk2bwhhBCk0zESSROtFT44cOfQTYybQ6n+VWzvNKqSRFM6MLQhDH0Llr4bIVYvFE33MLnkr1Ft/oCa/WOyiV9h2p6k5C3hS59tyZ1YSgxXOpiKxcOFJ9acM2vklv8e09JMNU/Q9CsAeLKJaP0XsRx0iuEEoQzoim3njtxHUcXmztLNIGUaJAyDvV2d675/vZFeD6oQ5OIxetIp+rJrt2CBH3D4Jydo723j1Gvnef9vPIzeWnhVTeWBjxxibmyRRDpGpi2JbmzMXnLDKlX3ErqaxlSz5M3dqEo0F5v+AnawQFofRlVubRdz00Z3bnyR069fQBLFvhRVQdM1qqUaO+/aCsDBvh72dHdyfGqG0aUSn7nnIP3XDcyuztWeU7Xh8KXvvsk/+733L8e9OvLXtiors8S243HuyhzZdIyB7vzy+4VcgrOX5xibLjLQnVsexM62NAKoN90b/r5mUMFUk0gZcKn2JrpiMtE4ScEcjLZWQCMocbH2OhONk3SYW9nR/ySGZmK1vKOOrrWTIBYz+OCT+/jCl1/hb7/zFqmkRbHc4EfPnKJWd/j8X7+MZel0tad5+tlTSCmpVG3GJ4tcvDLPUrHG/r39fP27b3NlfJEjJ8ZRFYXP//UrdHdlmFuo4DgeQSijcIofEDd06q6Loaroa7yJyFMYyGfoSt/YSG8GRQge6dyKqWo4YcDF6gK7sl0sODXazAQHUn0s2DWKboMd6U4+d+FVvnjpDT7Yt4eBRI6y2+S1hSvc1z7M24sTTDbKdMZSzNlVvFCiCwvXv4Ku9SLxCMMa0eLnIYSJImKI67zcubkKf/anz3HkyBiPP76L+x/Yhq6vneYTE0t88S9fZnGxxv/6rz+Nrt845hqGEsfzURWB0TqmUASaouI0XRRFRLmNurP8p6opWHETZQWzR6CTif8iKflelqqfo9r8PqpaIB37MKa2HVDxgkkADK0fKW2a7luEskEQLhHKJr2xfs5WT5HQ0igkaQY+KT1FXI3iik3fQyKJqWtpbh3Wdjqs7UDkdS+5o8TVPDFtkzjSO4CEYdwwvHgjCCFIm2uNnOf62HUHK2HSrNnYDYdMIUV4XViwvFDl6S+9yCf+wQew4iZCiZypxekSJ14+S8+WTrbs7UczNASCZrBIID3y1h6UFTtHS81jqfnbSvLdtNEVioioMpoSGV1FQdVVOvrb2P/wrmufAzRVoZAxiLfiyg3fRle05djpSkgpmV+qMbdUo7dj45seSsmliUW++ZNjfPzJ/avem5wt8cNXTvPoXSMMdF9b0Qd7cmRSMVxv7RbCCRr40kETBnZYY7R2hA5rC82gymTzFMOJuzDVJG7YRBU6ZW+WicZJFpwxtqbu5Wz5Bfbl3rtpEu4qBvraqNZsvvX9ozRtD0PXePSB7RTakpw8M8Xo+CLffOoIfhCiKIJ43GCwr40nH91FzNL58fNn+Mu/eZXe7hwf+9ABqjWboycmeOPtK9QbDg/dN0ImHaNiO5Rtm0IiwVKjQT4eX2N0FWXzkMGtojse3bPJRomX5y6xLd1OdyzDD6dOU3abTDXKnCrN0G4lebJnB1XP4SuX3yJnJghlyKnSDC/PXcILQ+KazvHiJPe2D6ErBpa6E13rRBVpJGHki4nNs9mjVxYYG1vC8wImp0rMzJTp729b87nXX7/Eq69cwDA0pqaKDA4WNv2di5U6b50ZJ2EZ3LGtZ9noXkWtVEc3dZKayvzEIoEX4PsBVtykc7CAsWKeCKGhqZ2U6l8hZu4nYT2AH8xg6TuRBATBJJXGd7CMXWhqB6a+m2rzBwRhEVPbhkCQ1QsMxPZjqTpLTh1D0XADFWSIG9pUPYclp8GWVBtxbfUcjWKtNlrLa4upueW//zQIpcT1A1RFrJl3ADFdX9foSikJW/9f/72m53Fmbp6MZbGlLY+CILZOrLeyWOPisVHuef9+Hvn4PQglCk8EfsCFo6Mk0jG6hzsQQlArN5ifWMSMG6iqgtNwOfnKOSbOz3Dp2BiJdIzebV3oaor22EE0EUdXUqvm3e3S7eAWjG6hJ0fbR+5cFR8ZPz/N4lQRM37tpqqKwtaODKP1OtP2HPWggYpCTLVI6Qk0oa66eEURNGyXL377dT75/oMMdOdQWzcmlJJKzWZ8psjxc1M898YFwlYCKmgZqKbt8czr51ko1mlv0bKklHh+QKVuk4ibWIbGVPMM3daO5XOfr77M2cqLGEqMpN7GvH2Fk+UfowqdhJbnzaWv0xPbgaUmyZndvDz/JWJqGk0xOVt5EU/aaxIU60FKOHFmkofvH+G9j+7Gdn1MQ6O/N49laRzY0UO53EBPmFSKNWKpGLGYQTYdoy2XRFEEd+zpo1RukE7F6OnKEIaSJx7eSblq43k+Xe1pljybhXKD+VpE3cnFLbKxq9vytTUuPy0NZ7xepOY57MpGVDtDUUnqFjXPYW+um7RucrG6wIuzl/iF/j2kdYvBZB5VKFypLeKFIYai8njXdpqBR1wzMBWVimfTn8hhqREDRlHX34puBE2/xqGdm6tw/vwsfX1rPZLZmTK1mkMsFnL82AQDA21rkoEr8fqpMfo7shQyCSxDp1K3iVv68lz1HJ/xs1M0qzaNapN0W4p6pYGma+S7shjWasPXdI+iqd0krPsQqNTsF1iqfwFkiJQuqpLB9k5QbT6N708ihIVl7CNu3Y8QMZq+x4niNN2xNPNODU0oLDoNCmaCsmfTZiYwFJVAysjI4mD7C0hC4moHdX+ajDGMEApxLbvpmF4/Fm7oYLS21FfHTErJVLnCn71+mLv6e3nv9q1rDKila8vjter4wOfffJs7uju5q79v1XuXF4v822de5NP79zKczyEExNdhKTlNhyPPnkI3NYZ29eE5HgvTJc6+dZFzb13mwY8eIplNUJwrk21PgxDoho6iKtgNh9nRBe794AHe+MFR5icW6R7uQFUN4tr6VNKfBjfv6bYC22EYMnp6iu/9+bM06w7Du3s58cp5PvMvPgHAWGOKscYM3tVKGAl26CClZGuyn7yx2ptNxAw+/Ng+vvidN3j12BUSMYNEzERKaNoujusjRLRN3721m/HpJf7n//MpkgmT7kKGUrXBQqmOqgj+5X/8Hp2FNJVqk2KlgQQeODBMd0eGi/WXaTeH0DARAoYSd7LoTJA1OhlJPUBIQCgDFBRUoeNLF12xMJU4w4lDdFsjqK1MsB9GMTDjOu9AShlVL7Umouv6nDo3zY+eOcWv/uIhMppCX0ceXVcJw5CZi7PIhguVBqm4gWFobN/bv2orCjDY38bgdd5af2+e/hX/bvMD+vMZglbCRlMERqsIpeZPUvdmaLfuYME+Rpu1B5VrCQYv9FCFEhHTUdYYKN8P8L0A3dCWC2AW7BpTzQoj6SgWH0pJXyKDHXg0AhdFKIzWlnise4T724dXPZy7Ml0/M+5lR3sas8UuKJcajI8tEgThcsz1KiKvGWzb4/jxMT70C3esuiY/CDk9OsubZ8Z5cN8wmhoR+AvZBI4bcGlqkc58kmcOX2BLT4G7d/ShKIIgCEm3JdF0jTAIQYC5TnLO8ydIWI+giGg7mLAeImbcgSQEGRDKJgCqkkLKACFUhIhHIRWhEFc1HunaigJsJ9q5XKgsENN07oqlW0VAPoYCFfcSbljCC2sE0qWhTKMInZuNY4eElL0lFp1ZdMVEU3QMYaAKlTYzWhSlhMlyhbcmJunLplvO0TVjLYTYsNhBSsn3z5xnrlrnzt7eVSHFQIbMVWtcWFik4XrEDB1TXWu2Cj057nh4Jz/8wgssTC+iahod/W3svnc7935wP2/88CgnXzlHo9Lkng8cZHBnL5eOjaLpGmbMQDc0/uJf/S2KotA30hU5dTco9rpd3HIiTQhBKhvnsV+6l6HdvdSrTf7oH/3F8vtDiV6GEmsTMkW3gqGsXaEMXePDj+5lZKDAy0cuMz5Tomm7GLpGujdPT3uGbYPt7BvpIZUwmVuq8sxr5zl1aQbPC9gx1Mln79xKNh3j2dfPcWlikVTSYttgB3fvHeDAzj4SMYPLiwqHl75Fh7UFBRVfemSMdtrMAeI3iGWpaMRFPWIDCAuDWsSJDUNQr23VpZQcOTFOtWYjhODCpTl+/Pxp7rtrC/t29zJ7eQ4ZSpo1m4WpJUrzFbKFNEbMwK5HSbnbyOdE46iprGRlrETVHWOu+TYxrcDx4p9yX8c/J6Fdu0eX65comO3U/RoFox1TNZcpWp4XcOzoGEePjHHf/dvYvqMbTVOwVJ2z5VncwCehG0w3KmxNFWgGHn87eoRz5Tke6tzKE93bVxmzIJRUbJt8IqrOqjvuhh4QRAZ/bq6Cogja29M3rHostKdItHZeQ0MF7r9/2xqDC5BImstx3M7ODL4fYhjXjj1brPL9185g6hpHL0zS057lqddOU64PU8gkmC/VuDC5gBCCS1ML3LdnkI6BzUMUqyHxg1lUJQUorTh1DFXECcIitca3MfRtpGJPsHJSOIGNIkLsoEFIgBPYtJtdCKFwR75n1Rli6NS9aZywSM7cia7cHp+57le5UD2FqqjE1SS6NHBoElevMSIURXBnXw9//MlfJGEYmJqG4/ucm19gMJclbVnoGxhdIQSH+no4PDHNTLVKT+ZaorUnneaB4QEmyxVKtk3c0FtzfTUMy+C+Dx3kvg8dZLY5RsYoYKnRHFtwpjjw0RGyxur789Av3o1h6SAEH/v77+OJX3mA7/7JT8h2ZFA3YEu9E7gto1vozVPojRJZU5fnMMwbHyZnbEzQN3SVfdt72bf9xtnzbDbOB96zm1/6wAF0ZfXgb+2/Nqh+EFJr2Mur5o70wxwtfZ9zlZdRhEZG72Q4eSeFG9BgorLaOjKYIwxGESIDBCBdQmmjrjK6MD1b5vtPn2BmvoJl6tyxp49PfvQQ6VySVCbB4nSRTHuatp4cdt0mlU+SuolqtZ8GqjDQ1SS6EsdU0pEWA1D1qsw6MzT8OtIsMNGcoOgW6Y8PkDNyBEHI2TPTfPEvX+bYsXHOnJ3mt3/nEXbs6GI41cb+fC8vz15CU1Qe7tzKAx3DKELhf9zzHtzQx7xO18LzAxqex8W5JfLD0QNxZaHIlo48sQ2M7sREkb/4wosEQcgvfPgABw8OrmtEr8KydPL5KCzT1pZkcGh9QxiPG+i6imUZfOzjhzCM1XNYSujIJsmmYtRtl7lilfffs5OJuRJLlTogiJk6YSjpLdy4+OR6mMYuKo1vozqZiC6GSsw4gKlvR1GSJGPvaxnklSGPkFlnEkMxcUK7xaapUDC7NlyrA+mgKwlUcfsxW0uNsTN9B2k9ypeEMqTiFYmtMLoAuqrSvSI5e2Zunv/tmRf5R48+yKG+HlRlrT4DrV/4oV3beXtymtNz88tGV0pJ3NAZzuU4PTcfec5C3DAZt+jOENdSy0a35M5jqfE1RjeRiROGkpMvn2X83DQzV+bRdJWuwXaUDWLPDc+jZrt0ppNMFMu0JxOY6yRqN8NPLYKQyiX48N9dS1G5Eaqeg66oWOtsFTbDaK3I63NjPNq9lf5kdt3PBGHI2cuzfOe5ExzY2cejd20jrucYSOyn6VfwpUtffA9Z4ybiNbKJDEsINQ/SBukhlAxgcn3RgaIovP/xPWzf2sn0TImYZbBtawe5THTzhSKWF6ufJ1TFxFCSGEoaTUkghEpIyKwzy7HyUUzFxFRjzNjT5I08eT9PzshRKjX4i794kWPHxpASDr91mYGBPFu2dGCZOk907+Dhzq2R2AnXeJVRNdL1tCePYqOJ4/kYmsrYYgnH92m43qaySpcuzfHaqxdpNl2mp0r80ifv5tHHdrUKEtZHV3fmhh5xLGagaSpCEcvFDithOx6WofOxh/eten3P8DsT4zO1ERLmAzTdoyAb6PowmtoBqEhpE0gXVWgEYY1QOkh8VJEmpWUQCHTFxFIsLHXzKqi4Fm3/lQ1Kh28GujDQVvDrFaGQNdYmJ6/HfK3BRLnCxYVFDvX1INiY+jmcz3Gwt5uvHDnBWLGEJGLZLDWaXFoq8tDQAPl4rHWMzXMSkSjStftvqjH0DRKFAjAsHafp0t7XxoHHdlPoza37WYDxpTLHJmf49KF9fP3oKT5xYDe961DXNsNt34nZsQUaNZverZ1kC2lee+ooiWyc7QcH1yQNrkfNc/jShcPcVejnUHvfpp+9Hm1mnDvyPWSMjVduKWF2scprx67Q25Fd9Z6hxoiJNNomvEMpJRVvEV/65FQHGcyhGHtb9KQQIYxWvGo1K0IICBWXWHeZR7Zsp+QtEFfXJ3nfLKSUuGGVULotDqWBqW7G8vBxghKqMLGDIkm9G0UYyyRuTVxLrpmKyR2Z/cTVOKZiktTuJqEllreNhq5SaL/muaRSMXbv7kNrbb2uqrTdDCaKZc7NLBCEsmVsXeKGsRxm2AgrR+7ixVk+/7kXWFys8aFfOEAms77B6ejIrOuprEQspm9KE8skY4z030q44BqCMMT1fGLm1cpAyVK5sapwRwiFuHk/cfNeQK7SU5ChjxfMEcoaflghlDZBWCVt3U/O6F91roS2Me2v7NjENANjHSbB9ZBSYgc+0/UqQ+ncKo9UCLFp0vjM3DyFRIK2+Gpmya7OdkxV5eJicfm1jXSCVEWhkIjz9LmLnJ1fwAsCsjGLoVyOJ7YN856RrStYC5s/U7pioqDghx4Vf6mVf1mfSysUwfZDW9h2cBikRL1BubYXBJQaUbx9ulzF8YNNP78ebtvonn3zErPji+Ta01SWajz9Vy9y4LHdlBer3P3kPoQg4rqtGJxQSi5Xl/je+Gnemh9nd7aTuufwH0+9zFSjwn0dgzzUNUybGUdTVEIpeW76Ik9PnecX+nfyQOcweTPO4YVJFuwaD3dtWZeaoiqCe/YN8q//0S+ST8cxW1vH3thuIFqpb8Q8mLIv0fBr5HMPIZR2YHWJX/T3aBLUvBJ22CCrF6j5Jd5cfJp0V57n5r7K/tyjDMV33ZThDWV0AwUCJyjhhjUsNctU/QVS+gA1f5qsuQ1DSRNIhwX7BIrQaLfuoO7PAiClx5JzloTWRcm9SEx7bxReUJIR+VyJIVBQhUJCjZPUEmRa2eusll/lISZTFr/5mw+RSlq8+upFPv7xQ9z/wLZbUtiKrknSn8/SmU62yoMjY6oqCmEo143RXcXgUIHdu3s4fHg0UrSaKfNXX3qF6ekSn/zUPfT15dYY2K6uDKq6+TValhGV9gbru9mFTJxc6tY0CQAWy3W+/P232drfxvsfiKiU33/pNM++cZ7/+b/70HJFmxM41AObNiNL3W+gCI+YahJKiaqkSJp3AgJJAEiQAYqS2PjELUgpafo+r82M8ecnD/N/P/QwpqPStD0ObOvZ9LtPj13g3775AjvyBT48vJN7u/opWHE0ZW1ydSV+ePYCH9w5Qlt89XgVEnGe3L6V6WoNKaME8z994pFlVs1V+GHIxYUljk/P8v98/xP0ZzPLSbiUaaKryqrnfGdHgc7UoeV/F91ZMno7S+4MR4rPU/GW6LIG8KWHEzYJZciezL3kjPWpkkKITedLEIZcWliiP5dFV1WsFnsiF4+h30ay7baN7r6HdjL/5Vc49uJZPMcj35Vlz30jPPM3r9E1UGDywgz3fOAA1go6WdP3eGX2MltTbVhKpMMQAvd2DCKl5IeT53hu6iKP9WzjQFs3IVBym3ygbyc/mDjLvnw3OTOOFwa8MDfKjmwHvYm1Xp8QgkTMYPtgpB8QyABVqGjrJPJWIpKWkwgEKjqqUCMdA7G55z7rjDHVvMyB7COAQBEqCpGG6fWaDdfOFeCGdUz1Wjxw0T5J019AVUwW7ONoIkbaGGCm+Sa+dKl6Y2jCwg0qNP15ZptvUbD2EkqP0dqP6LAO4IY1Gv4cVW+CQLosOWdpenPorfJEVRitBUeQ1rL4XsDsbBlVjRgVV+Ohqho9aB0daX79Nx7kFz92iGw2jnUbmrLNwCMgRNMVlpo1QiQdVoqkbq1LZ1uJoaF2PvFLd+O4PrMzZSqVJvW6w1PfO8rMTInf+u1H2LGje9Vi0d6RumF4wbJ0NFUlCNYvA1UUhdvh8T//5kV+8PJpPvPRe5Zfe+PkGGevzHH60ixtA9EjF8gQJ3AJwoBpe5EOK0/dtyl7VbYl+1vlzbeOmufyhdNv8+cn32KhWecfHLifrG4yu1Td9HuBlHz70hnGqiXGqyWeGbvEjnyB3959iEf6humIJTa87zPVGsWmjWT1zkRTFO4b7OfFy6PLrw3lsgShZKxYoua6SCkZLZb54uEj7O/p5p7+PgxNZXSpxJsTkzwwNMC3T53BC0I+uX8PuViMuK6v0mWYbo7S8GuoQqNg9rArfTed1gBFdw4vdIhpSbL6re9apJTUHJfz84v8+2de5g8/8iSKItBbEyOfiKEpN95FXI/bNrq5jjQf/uzjnHj1HPWS5KO/9x7a+9qoFmvMjM4zenaK3feNrDK6qhAYisarc6PEVJ035sc5XZojbAmkDKfy7Mx2kDZMhBDIMGSuWaPpR7qm83adnBnn3o4Bji5OUvOcDa8vkAGBDDBVk+nmFF1WF/oNjGfZW6DuV+iORQmhm5FyDGVAUsuS0zuQSBShtIL4CeJqClWsX2bohU2mG68wlHr/8msl9wKXq9/DUtvIGdvwwhozzTeJa+3UvAmSei9Vb5xGcxZTzdIeO0Ddm6LkXiSp9bDknIlq/VEJ8YlrBeabRxEIes0RANLGMIrQ8byASqXJ9HSJhfkqHZ1pHMenWGyQyyUoFJJRzFMIkkmL5AaynZshkCFN32PJqVFymzQDDz8M0BUVgSCp31ggRFEE99y7lXxbktmZMsePj/PKy+eZni7z1ptXcN2A3/nso+zb17d8rFwuEXnjm8T+TFND0xScGxcr3hLS64yTpiqEMoprKxj4MkACuqIxbS+iCZUg9HGkiy9vfbt6FVJKputV/uL028w360D0+7OpGLHixka36jq8Mj3Gm7NRFZwEfBlycnGOf/HKj/jI8E7+zs797GnrXJeuZWkaL14eZUdHgYwVLaRl2+bkzBw/OneBJ7dvW74PQgjcwOOFS6P84Ox5JsoVHN/n3oE+fuPQgeVdz4uXR/n6iVN0ppKcmV1AVQQ1x11X6rHD6uPVhaew1DhbknspmD2oQsVULSw1gale1bW4NTQ9j8NjU1xeKpI0jWXd36uJs1wshvbz9HQBzLjBoSf2Lv+7OFfBtT0MUyfflcV1vFWftzSdjw3t5a72/qjCrLrIlWoRTVHojmfZme2gP5nFaK0eVc9hZ7aDiXoJLwyXV9GsGaM9lkTdZCDrQZ2aX6Un1sucM0vBLKCzudGt+kXeWvoJH+r+LYRQUK5LlIUywA7qaIpBM6iR0nIsubMcK73YigF7DCZ2EFdTaELHUMwNwxiSgJJzkao1QVLrRQjoit+LHSzRbt1BV/yedb8H4AQVpuovYqk5DCVJd/y+5USJG7bkNTWDmmcT1wwUoeCFAV7okzHuIgwN6pUmlUoT1/XJZOMkkxaWFeL7AY7tbeiBSimp1x0uXZxjdrZMEIR0dWfZsaN7jViMFwaM1ReJawaqEmlO5M0EKd1cVSUlpcT3g3XLdSEyvNu3dzEy0smhu4bYs7eP7z91jKNHRjl5YoLP/dnz/M7ffZQ9e3oRIhK5EUJEVZMbPBS6of3UPMxyzWZ0apE7VrBu8pn4mvBLOmGRipvs395LfEW+I5Qhdb+JpZroikbFq992lwUn8Dk2P8P3rpyl6q52RsJQ0rjuWbyKquvw9Ysn+dMTb7JkN9a8b/s+X794iovlJT675y4e799CQl99nw/19fDHr7xO0/MYyGWxPZ/RYonLS0Ue3zbMPdcVPMR0nQ/u2k5HKsF4sUTasrh/cGAV8+HRrUN0pBLc3d/LYDZLiKRng7L1nNHB3sz9JLQMBbN72cCm9c2TfVen+Ebr/lK9ydsT02RjFu3JBIv1JqGMchLFRhNVFbclsv5TsxeklFSLdV596giLU0UGdvQwtLuPzsEC6fxaKpShamxJR4OxLXPN5V+w6/xw4iy277E710kgJSqCR7u38Mb8OBcqi2SNGE7g8+rcKGXXxrquA8WcPYeu6JTcIqeqJ9ma2AYWLS3RGw9Oh9mPRDLrjKNyTYO0GdQJQg9fesza42T0NmadMXan76Xml4ipSe7IPszl2omI5qKlWvHTjdX5VWGS0Ls4svAfUIWBoujE1Q7y5i6yxuaC3AKFkABTyRJKF4EgkCEzzSKqUKj7DhkjTsVrIhzIGyl8GVBy6yw4VRKihjqvYRoa27d3LWfyIeLlBkGAuk6s3PMC3nzjEj/+8UlGryxQLjcJQ0k2F2fPnj4+/JGDbNvWuTyJNaHSGctgqRptUtIfz5HULSJB7mjGh6HkJz85RaPh8NGP3rn57xaCeNzkwQdHGBxs43vfOcJTTx3j5MkJvvC5F/nMbz3E7j292E13+bMbJdQMQ0W9ydi0lJJTl2bYOdy5zCd2HJ8vfOt1bNdba3SvexBTSQtVVdZQixShkNKvhRHS+q2HFKSUXCgt8tULJ3hh8grni4u44TVvuVRrMufV1tXzDcKQFyav8F+Ov8lYtbThOfww5O25Kf53+wXswOfDwztXPXsPDQ8yWizxjROnqToOCcNgpNDGLx/Yx6Nbh7Cu+91CCPLxGO/dvo2SM4EQCjFNp+LNktYjtkVfNkNvizrWn7sWQgylXMOCUFDpjg2jKfpNebShlFy6NMfbx8Z5+IERujrXT0zn4jEODfRwanqOwbYcf/3mMWqOg66qHJuYIZeI8dDWoRue73q8I31z4imLOx/fQ+AHxFMxEpnYDbPHV3G1RUdM1dmSaiNlmEig4tl8+eIRXp8bw1Q1fnXrQb43fpofT54nY1j88taDdMauX/kkF2rn6Yv1MRQf5mTlBFPNSQx1cxWgojtLyVsgqWXpMPuo+xUSWoq6W2XenuBY6UVGUgepekXK3gIVbxE7rNMIqtS9MpONCyw6M6T0LFW/tOwha0Lf0HNRhclg8kly5gheGG0Fk1ovMa2wrhhy3Yu2ngnNRBUxOmKPElPjrQmoMlabZdGtIYCK16RSbtBpZZEy5Hx1mryRJGcmWXQrWFYbWwYLqOpa6UNdV9dk9YMgZHq6xDe/8RYvvniOhfnqKqnCYrHOxPgS585O87GP38Ujj+6IYqaKQpu5niERy1nsZ35yij//0+fo6s7y6KM7yWTWZzNIGXI1a61pKoODBX79Nx+ipzfHX3zhJY4eHePzn3+R3/mdR6jXHVRV0N2d2dCL0fWb83Rdz+erTx/l2dfP8+//p19CNVsl6kjeOj2+Ri8km4qtMXCJmLFh1v52IaVkyWny3ctn+crZY1woL+L4AbqqoiuRNjJAwjLozqVIWAZhKJe9cCklVypFvnzuGOPVMind5Fd23MHj/Vs4tjDDF88cYbxaxlRVtmTynC8ucrlS5N+88Twp3eQ9A1uX+bJpy+Q37jrA+3ZswwtCdFUhZZrk4zG0VieNsJUrUYRoaS14+NLFDqsoQkULDBr+0rLRhfVDQ98fP8P7+nZQtBuEMqQzHjk45g2ocysxMbnE57/0MkeOjjHQl6ejPb1iXFyCYApJiKEK7up32dNVAJos1kwCmcbSe7HdSZKxXjKxWw+73ZLRjQbPAwSq0AmljyBSGyv0bMxtu4qG7xJTdXwZ4oYBZotuNNOsEFN1dmTbaQQui06NvJngMyN38cnhO9AVlaRucnfYz4f6d2GoGgnNQL3upmT0LPPOG1ysnSeppbgjc4Aeq5dJewJ1RajAC4MoWdYqba54S7y28BRu6NAdG+JA/FEkIVfqp/jhzBeJqymem/saosV6sNQ4KS3HC3Nfxw0dFKGhCIWp5iWWnBn25x4GIK6lNowjR6WzUYghxMcJiuTS29c1uBEJ3qXsNql6C2SMGCW3SVK3SWomvYkMA4l2+uPRzkFyrR9cIEOqXpOcEbEX+uMFnMAjpt04ISalxPMCjh4d46+++DInT0Yxv3w+QTJpoagKruNRLDao1x3On5/hT//kWWo1mw988I41QuDr4dVXLzI3F0kMnjwxwQMPXlW/kkCkQ4BQ8INFFGGhKHEifodCOh3jgx/aTzab4E//5DmOvD3K5z73Apalo6oKO3du3BZK05RNWRhSSlwvYHymyAtvXWRitrSKTmwZGnFLX7N/0jWVXHr1PYxbBmKDc0kpo9JtKZeTlzeClJJL5SL/4ejLfO/yOXwZkjdj3NXZzuN9WzhbnOfL544DkZdarNkcOT8ZVc1lo92nL0MOz03x8tQY7bE4/+Suh/nQ8E7ims6hjl50ofAfjr6KE/jc19XPtmyBH4yeY65Z49+8+RwD6Qy78h34YcB0s0zeTGDGBD1mhkCGaELBDj1kKImrBlPNMnHVoGAlCaXPkjuGH9q4YQNNMSgHTa5vQ+QGPrqi0gw8luwGvgxp+C4gGa8VafgenfFbV8qbni5z+Mgo9brDsRPj7N/Xf02Y3juF651ESh9V7SIMl4hpHUhp05eNg6ygaW0EoYfeCputvC9wY12Tmza6TlAmlEGLA6qTMvpp+osYaoow9NCVJH7YRBUGqrLW0AQy5K3FMfZle5ixq8w2o7LRjBFjulnGCXzazAQhkobvkjcSxHWDeCt+5ASRwb5eMQmi2NiSu8Ro4wpbElvZlhxBaxn02UaVgt6HIrRl2sqiXacZePQlsuhCZTCxi8HErjXHfbj9Y/BT9HPcmbprUx1SXU3SFb83Ci8IdcMyzUCGNAOXObuGE/ioQiGm6lRcm1grsaEKZVUERV/x50olNFUoN82t9f2Ql146x+c/9yLjY4u0t6c4eOcQjz66k527erAsnbnZCq+8cp5nnjnNhfMzLC7W+NIXXyaXT/DIIzs2rR4DMFu0wsXFGseOjXPX3Vta1WEBtncWL5hACAsFCy+YQ1PbAImu9WFqg5imzoMPbUc3VD73Z8/z5huXEQJGRrrYuWvjCkddVzfcjYWhZHaxwotvX+LZNy5w9soclrF2i1zIJtbUdQghyF/nrcfN9RefMJRML5R5+8wknh+wZ2sXW/oKN0zOXCwv8a9ee4bnJy+Tt+Lc29XHh4d38mjfFnRF4a/OHuXLHF8+R6XeZN+WrlVFKHXP5eXpMVRF4dd2HeCDQztItp41S9N439AIT105x9vzUyzZTf77/fdR9xyen7zCaKXEfzr+Bn94/5NU/CY/nDrNfe3D+GFAzXNYdOv0xbMcLU4QhpKPDx4EWE4ShgQ4YRU/jLKYfuiQNnpau5lrOFOaY0+ui9lGjS9fPIIT+DzYNdRqJGusWzrebLoIRSzPq/Xguj7NRnTut4+O8alP3L1sdFW1i7i2jSCcQxFJVLWdIFhEyiaq2oGUDUBtqd2lsQMbUzFb3TbqGIqBeYOd9U09fYH0KDkXccIqhpLAUNO4QTVqV+HPYAdF4loHtr9E1txKUlnrYUgJBTNJzXepejamohFIuezxJjSTQIakdYukZq7aFla9GlP2DFsTQ8ttdlYdG0kzaCKA3lgfqqLiBgHNwKPqOYzXS2QMq1XJoxK0VsyuWHpNKfE7iRtteVShk9Q3b9IppcQJAvww5I5cNykjyg7XfJem75LepEjkZiClZHx8ic7OzKoqLyklr716kc9/7kWmp4rs29fPBz90Bw8+uJ1E0lye0AODbfT15dixo5u/+tIrHD58hVKpwXe//TZ33z1MIiFwg1kUYaKrhTXezI6d3Tz3/BkadYeLF+eYmSkzMNCGxMcLJqPJjYqipDGE2Ypnu8vC5hAZ0LvuGqZUbPCFL7zI4kKNnt7chsUTAKq6sad7cWKBv37qLSbnyuzZFmlNXJ5cXPO5tmyCpfLq5JMiBOnk6vOahrrGI5ZSUqw0+M9ffZnn3ryA6wcc3NHHP/7M4wz3tm1oMIp2kz87+RavzoyzPdfOr+88wPsGt9HeonRJKemIJ4lrOg3fI2Hp7Cp0Lv/mq7B9n4ulRXbkCjzWt4WUsdpQtMcSJHSDUEpKjs32XIHf2n2I8WqZ86VFXp0e4+WpUe7s6ubOtgF0RaXhuzQCj7ofJfLqnrPcZbvhuwStxV5XLPriB9eMR8hq5sZ8s843KyexVI3tmQKDqRwZI8Yb8+NM1svsy69+dhoNl+dfPEu52uR979lDNhNfdxw1TcEwNGzHY2qmhG17yy2pFKUdEOjaluXPq+q13NPVbiyKkmbWnmGyOUlaT6MKjVAGUT+/d8LoCgQpo49YaKMJiwCPqjeFqWYQQsVUo3p+TY9hKOu7+5qisCvbhR8GZI1YywAqXK4t0p/IkTOveQdu6K/K+lf9GhWvumG1qCpU+uP99MevVes0fJeJepmcGaMvkUUVYnnLHdN0/DD8qQWVf17QFIWcESOhX5PTS+kmqU266d4MpJQcPTLG1772Br//956gt/eaAPyxY+N88S9fYm62zIMPbedTn7qHbSNd61ZxKarCvjv60XQVzw84emSMS5fm8byIGuWHRSBEVZKoYrU3f/fdw3zlK6/RqDuMjy1y5fJ8JLOITty4E00trOpyuxGCIGJVeG7UxXdqssjUVJGeDcJe0VZ+7eu1hsOXvvcmiqLw9z79ECMD7Xz7uRNMzZfXfLYtk6BUaa56TQjIXEcb0zV13Yf/paOXOT+2wG985B7K1SZPv3aWbz57gv/h7zy6YSz6tZlxnp+4zHA6xz+75zHu6epbReMSQpAxLdpjCUarJUCsy+CQgB34ZAwLXVHWcGybvheF4YQgpmkoQnBPVx8f2bKL/3T8deabdX4ycYmHeoc4mO9f8T2Xstckpho81DlCTNWZt6ss2jW2pDbeNgohUK8zRztzHXz10jHaY0nu6xggb8aRRI5aQjMYWCED4Dg+L716nr9s9Q4sV5r8+q/cv6qh61XohkY8HjW+feKRnav614lb6PjdCBpk9SymaqIKDU1oG1a+rcRNGV1FaMS1a9UcgfSi7pfCvOWabk1RSba8SyklPfEMCW31hV6//S2YeZbcIreChG4wmMphqdq63qwfhmtiwjeDermBoiqYrXjlz7o9dJQk0NblR64H1/Px/RDL1Jcf3FKlSaVuM9izWvdhbGyRr33tDV579SIf/vBBentzSCk5d26GL/7FS4yPL/LY47v4zc88REdHetPkqKoq7NjRzUc+eidzcxXCIGzFrW2CsIap9SzLGK5ER2eGBx4Y4WtffZ3FxRqXL89zz71bMU0NVYmudzNjK6XEcXyef+4M3/j6WwA88OAIJ09M8NT3jvKZ33p43RCHqq4vvvLykcvMLFT5J7/1BAPdeVRF0NeZXa5qXIlCLsGl6zxgIQTZ6zxdTVvr6Yah5JWjl/nl9x/kfffvxPV9YpbOi29fYmahQs86gv5LdoPnJi6zaDf4x4ce4t6u/nVLfDOGSUf8qtFdH5qi0B5LcHhuiv/j6Kv8w4MPsi3btnydZ4oLzDfr6IrCcCZauGKazqe37+V7l89ypjjPmaV5LpQW2d9+zeO0VH1ZdyNjWEgJTuizK9tN8hadhHYryXv7ttNuJShY14ozVjpoECV6jx4f48//4kWmpqPf/NQPjrF1uIP3PLY2bGjoKrGYDkU4uH+QRNxYls+8FXSaXS2DG30vvC48shFuy9VThY6hJH4qEQ246rFZN+S6GYpBI2guezw3A11RSenmhuGDG5U2rsRVnVzf86ks1pi6OMPFo2ME/s0N8s8LsiX6fnl8kZn5Mp4f4Ach0/NlGg2Xat3Gdq5xcM+fn+XY0XGCIGR8fJEwlIyOLvDFv3yJU6cmefSxXfze7z9BZ+fmWgZSSoIgpFisU6tGYZ5du3vRdQhlA4kbUdvWmdSKIvjQh+6g0J4mDCODv7hQbS1oN56ejYbL0z86wX/5z8/gOB6/+nfu5x/8wfs4eHCIV1+5wLGj4+tyjhVl/QXz6Lkpzo3O8b9//id89UdHePPUOL4frssJb8+l1sxdISCbXr2t1dXV8XaAcq3JQrHOw3duJWbpZJIxfuUDh0jGTY6cnVj3t54tLvDG7AQN3+PPTx3muYnLzDfqOL6/6jdmzNgNE0wp3eCRvmG8MOAHV87zT1/8Pq9Nj2EHPot2g+9ePsNopURSN3mwZ2j5e13xFL8wvANNKEzVKlwuF5HSRcoAKQOQDaQsAY3WeETiR3kzcdO5hOVxUxR2ZNpXGdz1oCiCfC5BLhtfTpAODRbYt2f9mL6ha8RafGlVVSIpzebT63426trtRb/tOsTU2Couf1RQdeM5e1tW0wkq1Lyp2/nqLcFUM8sxT4HAC71VmrxShpTcS4Ry446e7wikgiy2Ua/YlBeq1MuNSJh6xUSvedM4wdot6DuNpN6zqnR4JWzXZ26hSiYd4/L4Irqm0nQ8zlycpS2b4MylWQq5BPt39ZFKmNhNF9uOEgoXLswyPV3ia199g8NvjXL/AyP8zmcfJZPZvD2O1yojPnduhh88dYyjR8ewLJ3HHt9FLBZDUbrxgoVNvdWu7iwf+cgBPv/5Fxm9ssDSUp3evs3V2KSExcUq33/qGF/58mskkiaf/OTdfOAD+4knDP7Or9/P//KH3+T558+wbaST9HWMgo2M7vse2MF8scqpizP80ZeeIwglW/sLNB2P05dm6e3ItCrZfOaLNZYqDc6PzpNNR92Qqw2HSt2mkF2tNXu91fWDkEbT4bUTVziwow9TV7Edn858itHp9Xd1i80GM42osuzo/DR/8Oy3ua+rnwd6BhlMZ8maFnHNwA2D5VLVjWBpOo/3beG5icu8Mj3G4bkp/uDZ7/Jbe+6k4tp859IZ3DDg7q4+7um6VtwghOD9Q9v5LyfeoOzYzDfm8b0AISwgJPAvIpQUqrYNVb053rGUksWlGvlcYtXiLoS4qd2oEIItwx38wf/tSb745VdpNFx+73cepbBBM1pdV1cV8oRhhZrzAqn4+9d8NpQVGs6bmPp2DO1qq/pr570d3JbRXXLO8PbCf7ytE94KehMPsr/td7EDB01o+NcZ10B6vLXwR7g/Y2OnyDjDpT/A0BKMHBwillibvLpUeYqJ+vM/0+sA2N/2+/Qm7l/3PRlKag0H1wtoyyVIxE1KVZuRoXY0VYkMg6LgeT5gkkrHSKVjFJfqHDs6xueDkFdeuUBPb5b3vGdPtAW7/hyt7fzcXIXp6RJXLs9z9OgYJ46PU6+7WJbOE+/Zzd49faiqgheU8IJZVH3rhr/JMDTuvXcrzz57hsmJJRqNzWtzgyDkypUFvvPtt/nJj09SaE/xqU/dw6OP7cSydOp1ByEEu/f0cfzYOJcuznHg4OCqYyjK+jHd/dt76WnP8MaJMS5OLFCsNLjvjiFePHyRf/eFn7Clrw3L0Kk2HM6NzrFUrvNvP/c0fZ1ZVEVhsVzHcQP++e++d8W51i452VSMVNLij7/yEnu3dZOIGVRqNpPzZdpz6xurbdk893cP8PzEFezAp+F7/GTiEj+ZuISlarTHEuSsGEndoOg00RSFiWqZOwpd64YhhjM5/u7euyk7NqeW5phr1vh3b72Aqii4QUAhFuc3dx1crhC9is54kq2ZPMcWZvHCBkL0IGUVGTZQ1HYQFmKdUFIYSk6cmqC9kKK7K7v8+pXRBb7+7cM8/MB27jwwuByDDsKQktukzbo2HjONCu1Wcg1zQVEE20e6+N3ffjQKW3bnNkyUXk2kXYUQJqpY35GJwmMLSDm04lXZ8nyV2yovvm1Pd9E5cztfvSVkjGHswGbRLaIJjYR6PXE+pOicww5uLd57qzBEiu4tBSw9vaFsZc2f/rmMyWbedDxmcM/+oVWvjQxtnLzo7EzT1ZWhuFRndrbC7OwpAKoVm29+4y2effY0qZSFZekYhhaVk9YdqjWbhfkqs3MV5ucqOE60GMbjBo88upOPffwusrk4UvoEYRmBtqZr70oIIejuyXH//dv4m6+8ThBuHLZZWqrz/aeOcuTIKCdOTOK5PoMJk7GxRb7w+RdxbJ963aFabTIzU2Z+vsrCwlrdARGJsq57jvZckg8+tBs/CGg0XSxTp6c9w+e++RqvHx+l3nptZKCdx+/axttnJnnmjfP4fkghm+Dj79lPd+FaTFZVFNLJ1ToTuqbyK++/kz//xmu8dOQylqGxb6Sboe48ufRGRrfAP7rzIbZm2vj2pdNM1CrL79mBz3itzHitzHA6x962Tuabdf7k5JscX5xhb1sn3YkUWTMWJdF8j4rrMFOvkNSviiBJAikJgha1S0qeHruAG/jcUegmH4uKcXRFYVu2jfOlRVJmF5o+EtG9ZBOxgVCPlJLjJyf4sy+8QFdHmt/+zYfp6swwNr7IX/zVK7z0ynnOX5wjCELuunMYTVPwZMBkvUSblWDBrrHkNLhcXWJfvpuMEmOmWCVu6nRkkssc5/4b7JAgirGv1mMWa1g1y/dJ7cTUdyKETktDHS+YxXZPYBl70NXN2Ufrnv+Wv/FzxlVj62ruzzxptSGEwIqbqzq6/v8SKpUmr756ge0jXQwNt29K0l75Xm9vnoGBAqdPrQ4Vzc1VlgsWriqOqaqClFH7nCBYbRSFiFrefOKX7ubhR3ZQKKS42jctlG4kwi3Xr/2/injcYGCwjXQ6tmHrnjCUXLkyz1//1as0m+5yVdyFC7OcPz9LEISEYbhK2K27O0t6nSo3IdYvDF8ZG9VUZVnAZteWTv7hrz/GQqmO5wdoqkIuHactE+dDD9coVaOS6Jil09+VXeVl9XRk+HuffnCNjb9//zC9nVkqNRtNVdA1lT/521fYtWV1M86r16QKwa58Bz2JFI/3b+H7V87x47GLjNfKy2XVhqLyQM8gHxgc4Uq1yPGFGc4uzS97wJaqLQvauEFA0/eoODbBOkmgJbvJF88c4cdjF+lJphlO5xjJFRhKZ9EUlfZ4gt5kpjWeCoiNDe6pM1N84UsvcfL0JGfOTqPrKr/66fv40Y9P8sprF3Bcn/MXZnjl9YsMbW9nyi8znGrDlyHjtSLnyvMsOnWc0OdsSeWu3AAN16PuuOSS8VUMDSklC4s12vKJdXMRqqpgGCu9d4EQ+qrvR78pumEN51UUkSGT+DgCE6Skbr8AqGhW5y17u++I0dWVOIZy6y1Lrocvbdyg2tIQjaApGiESO3Bbze42pnopQsdUsjelDrYZQhnghCXCGxiKjSBQMNT0smD4TwM3rC6XCW+ESxfn+NpX3+BTn7qHoeF2alUbGUoSSRNFVQj8MDIyimB+tkJbIYWmqySTJnv39vL6axcplxt0dKQjwRvXx/cCfD8kCEI8L8DzonuiKALD0ND1yFvo7s7ywIPbefiRHbQXUuiGtmxcBCqm1ksoG3CDiVmvO1y+OI9uqMTi63vFQRAyPrZIvX5N0EWIKK58lUmi69ryQqEbKvfdv5Xt29fv9qDrKvHY6so8XzosOFfI6N1U/QUK5iAld4qUXqCj3Vwu+135ndQ64aZQykgGF4ll6mztKywvRL4ftK5VZWtfAdf3mV+q8YVvvY6iCPZuu+Y92bZHudLEtl0G+tuQYYglNe5q72VvWye/t++eZe7sZK2MH4Z8fOtutmTzfGBwO1O1Kot2g9lGjdlNxl8TCnkrxt5CJzXP5ezSAjXPxQkCRqslRqsl3pidwFBUDFXFC0Pu7+5nV/7G1UOhlMy3dkZBEKK07lUmHeOhB0Y4cXqSE6cm2b6tkycf301Z2lypFdEUtVUx6nCxukDD98iZcSxVR1MVBFBIJzBX0Bh9P+C1Ny7xN19/g3/xTz+yblxX0xSMlXoQQkEIM2rNhYIXTNCwXyGT+BQQoqldeP44UtqEMkRV2zD1nbj+BUJ5F6q4taq4n9roChS2pj/C3e3/6IaflaHEdTxUTV0Wc16J0epPeGvh31P3Z6LPE5XsIg1EmMEJAkpujb5Edt3jt5k7ub/zn5MxhpZfC/wQoUSdX6WUq1a+SjEyZunrYmgVd5yXZ/+QefvY8muhlNi+jxsExHVtUx3NuNbOnYX/nqHUe5EyxJPeuvy9sCU/GYmqr5/5fGv+jzhd+utVC9H1uHJlgVKxQXdPFgC74TJ2cY5Y3KB/Swdz00UUVUHTVCrFBrmWEJEQggMHBunrz+N6Pr/9O4+w745+Ll2aZ3JiibnZCsVSnWbTIwgi7y6RsGhrT9LTn2FkpIvOnhSpVszNC/1lPeLo+AqKSKCp+eW+bOshDCWXL83z4ovn6OzMrEl6XYUQkE7HuPvu4WsqYpqCoauYpo5l6cTiBulUjFxbgu7uDAMDhXWPJ4TgkUd3Lsf3gla+oOLNU/MWSGh5FpzLmEocN2zgBHUq/jxtRj+K0DDWiVmuRMP1qLQ0ZlOmSaoVlqrVHV4+ehnb8cilox5dlycXefbN81TqDv/i996HseLZOHJklHjCjIyVolCvOYxPLHLvPVtJp2MkdIPuRGpVsusqPrv3LhK6wTcunmKiWqbqOfit0I2qRHz1lG6Qt+LsyBX4xMhe7u/qp+a5vDw1xvdHz3F6aZ7peoWKG33XD0MavkfKMLmnq5/uxI0NjqooPPTACKGUfPmrr9PZmeEzv/4gyaTFju3d/JN/+EH+5utv8ORju9mzO2Ic7Mh1IKVktlnlQmWBjw3eQUcsmrdSQqneZHKpTDK2utrv7LkZPv/FlxibWOLK6OK6RldVlVV8c4GGlB6VxndQlAQN52WEiNNwXiWUTTx/FF0dQkqPUuNvUJUcoayjigy300n25xpeCEPJ4mwZ1/aw4gaGpZNIxdBWtPZeiabvMV4rMdOoMt2oEBAu6zXcDFzH4/LZGTp7cyRSFrOTRTK5BNVKk7aONMXFGtVSg+6BNto6NvbUJZKy49D0Q6ZqFbZk83Qlbq6ZpB3aTDYusy21Z817Nb9KyV0gpsaJqQmS+q31WrqKhYUqjuMta96alo5h6dRrNs2Gg9300A0V1/GJJ02UFSr5Xd1Zdu/uYWJ8CRB0dmbo7MxQdKN2RQWjfc32yQ4cSl60pS2GJZRA4Ic+Fa9G3siS0q+NjUBgaauTWNejXrf50Y9OMDm5xN33bCGfT+I0XexGlBC7qlanaSqPP7Gbhx7b0dJgDtBEJPByO6Gn973/Wv+zkjtL1ZtnyZ1AILCDKnZQxWntMhShUXQm8EOHvNGHsY5GRqnRpOa4ND2PmuMyW6nhBQFDbTl2dXWgqVHFWKna4Icvn+HC+AKuFy1mQz15fvsX72Xftp7IIw4lS0u1SErQ8Wg2PRp1hxDIZuKbthq6ipim85ndd3Jvex9PnzzDWLlE0/fI5JNRaMSK05NMM5JrYyCZXe7OkDJM3j80whP9WzhdnOfN2QkulJaYrJWZb9RZspt0JpI83rflBldwDZqm8tjDO8ll4iSzMbLZOI7nR80suzL8vc8+hifDVaI8Qgi64mm64qufTSEgk7Dobcus6Rxz6swUYxNLhEHIlbEF7rpzaM21qIqySpRIURLEjDuo2S8ghIJl7CcIijTdIwhhoqmdxK37EMJCShfbPY6mdhCz7kQRm7ebWncsbvkbm+Bqr/ur24frIcOQhaki4+dnUTWFVC5BW2eG/u1dJNZpjSIEWC0pyKwZyTr2xG8+jFFcqHH6yBiNqo2iKfhegNP0OPzyedraUyQzMRbnKoSh3NToQtQCKGUYND2fot28aaMLkjlnkq2p3dd+V2t1dMMmZX8JTzoIoZDk9oxuve7g+yGuG3lr6WycfYeG8P2AiSsLdPXlSaatNYpiEIULDt01zCsvXyCTjRGEPhKYs2doBg1yRhvIgIZfx1BMzFaDPzf0qHhVklqColti3lkkplpYgbnK6N4IYRjykx+f4tlnTpPJxNm+vYtk0mJppsTUpVmQsPehHVS8JnHNoOG71H2HQEp8GRCEIVuS7bdV6HIVrutj1yWKoZE1ugnDEN/W6IvtJa114EmHojuJppgoqJjq+t6dlFCxnah3nKrSl8vg+gFpy1oOuWRSMT72+B3s2drNyQvTzBfrpFMWe7d2s3tL16oijFBK+vvydHdnb6hhsREUIRiMZTjQzHIonsdzA/bsGqLQlbnhQlVZrNMj43y6fw9sUZipV1nymyw0G5iqxtbMtaSV7we8+vpFypUmTzy2a5kHexVL9QaaorB//wBX5ouML5WJGToCQcI0cAOf6VKVtGVi6hrpmImlb5x8df2AhuORjV8X2mnF6oNQcvnKwvpjoohVLdYFBgnrYeLmvdTsZ1CVHH4wQz71O6vKzQEK6b+/6ZjdDN4xo1txbb51+RRvzk0S1/Ro+6JqmJqGLlQe7d3C7mw7vVs6aOvKRhJ9ukp5sYZcIRO4EjFVp6fVjqczlsQJ/TVdZjdDOhvn3sd3ousaV87NYCUMGnWbRNKkqz9PKhunszdPtrC5kRAI4rrOTM1mvlGjJ3nzMRxTsVCFSt2vUnQX6DC7mXOmmbHH2JbcQ0yNo6DesGfbZvC9KLk1O1NmZORa/FJRFHJtSVLrCGtfRRhKxkYXURRB70CGi/WzGK0us27osOTML+sKq60uwtP2LM2gSVKLkzMyFN0SaS1FwWxDv4WdSLPp8vSPTvKVL79Gve5wzz1b2LO3D0UReE6kUGUlTEZrixwpjlEwk1Q8G1PV8WVAm5GMpAOJFvvRczOYMYPugWvi1bVyg8krC/QOFUhm4pHWxMU5FEVBagpdXRlc16c4LenoHCCfT1AqNSgV67S3dxKLWZgyRCDI6j2oQkdv1d8v2FfIm33YQZ1Q+uQSbcQNHV1TNy34sUydvdu62bWli7rdQNECkkYGJ6wSSGgGZVQ08m1xtHcgV2JaOn3D7Vw6M83s5BLb7+jf1OBWK03CIGRmskiukKJaaaAbGrm4wdbO9XUhPC/gx8+exvMDHntk56r3gjBkYqkS9X8TgplyDZBk4hZpy2Kx1mC6XCFtmXh+gOP5bOsqbGp067aL7Xprkq59PTna2pJMTBYZn1gkCMO19DJVQVu3WacC6CgigaFtQ7C+BnHNddFVdVXLoLrrRtox5o3zOO+Y0Y1rOgcKPeTMOO2xBIVYHFOJlH4EgrRhoqkqbSv4eRC1/Vm56myEq5Utt4JYwiSWMCNh8VZSCWDr7l7MlvzfzUIVgo54ki3ZPIYaNc2Mftv6iAxB9LBqioET2jSDGs2gTjOoAYI5e4qYlliOJ94uTCu6jW++eZn7HxhZ/l2KIshswPmUUmLbHi+8cJZvfuMt7tg/QCKjMVEvoit6K8YsmGyOMZwYoewVCQmJq0l6rE4kkUi5KlTiaqwllRn1Wbu6RXQdn8tX5onFDJLJlqC4lNQbDqdOTvLss2c4eXKCWtWmqyvDe9+3j66uaJFNZhOMnp7ESlhoisKhtiGcwMdSG7RbSdrMJJpQaAYeilAYuzDLF/5fP+DRDx9YNrq1coO//bPnmZso8hv/4wdIZuLMThb5yv/5E/bcPUxusEAYhiws1DBNHbvpMjra5MrleVw3wDB1kimrJTZeWJPVfmXhy9zV9nHG68dQhMah/EfXCJVvBlUBwwpxghq+NKn7i9hBNYqFo2KpaXxpk9X70NZR7luJIAg5c3oSKaNqwJVzW9NV2jqjcNr8dGlZn2I941mv2oxdnKdesxECUpkY1XKTRt1lYEv7hsY6CEPGJ5bo6c6uYQwoQrC9qw0pYb5SJ5+Mk4mZKELgBSGqIujPZ0DAUq2B6wfEjNVsAsf1aTZdYpaBogpK9SaWEWmo2J6P2dK3GNnWRV9vnonJIqVyk8WlGh2F9JrrWe/ZF0IjYT2IQMPUd8B1Rvf84gLPj47Sk07x0MAggQxRWhrPl0tFaq7LfX39a457Pd4xo+uHIYaq0hVPYqgafihRRdjSsZXL2rXXQ1+npv1ngVjimljM7UASVQTN1mtUXIdCPE7aMDfkela9ElW/QkJNUvcrvDD/PQbi2xiIj9AIalyqncJUY2xP7qPN7LopoYyNUCik0HWV11+7yIkTE+zd27fhgnK1uGF8fJFnnjnND79/HMPU+MhHD5I209xl3s/V5EAgAxacWSw1RslbQhd6SwvCXHU8fIEfhDQ9D7sZFUgk0zEWF2v8r//qW8zPV4jHTRIJE9+PerM5jr9M70qnY3zoFw7w8CM7lq9bSknXUDuzowv07ugklrBQEMvCLFfv49WFOAwkzZqD715bwIIgpLRQo1ZpLu+mZBDSbDiEQcg9924hCCRWzMCydAI/RDZd7n9gBE1T1+wOrp87AoGlJBBCWaOQdTNwwjpNv0jdL7LgXELKkLiWIwx9mmEJkLhhk8wNlOg8L+D4sTH+4x/9CBD8/X/wXvbt61v1bKVzCfYcGsJzfRpVG8/1MdYJN8WTJrv2912TgRSCvqECSDbUBAaQIVSqTdraki3jeO3YQohlr7W/bTX7Y+VCJqUkaRprxXeaHt/4zmG+9d0j/P5nH+PRh7aTiVs4no+qKJybmmdPfyeqELTlE+za2c3R42M4jsfcXGWt0VU27v6riLXPYcOLurO8NjHBh7fvoBCPE0jJVKVKWzxGybbxgpsX0HrHLF7Nc3llZozLlSUypkXOjJEzY2hC5eTSDB8e2sXufOeND/RTQsoAx59DEQaqkkARt969dj0oCHJW1InUCVbXuq+HRlDjcv0MutDJ6m1sTe5GEzo1v0xGz7M9tY/d6UO3pHi/EbZv7yaVtpibrfC5P3ueT336Hnbv6SOdjqEoYplfWyrVmZkuc/LkBM8+c4YLF2bJ5eJ8+pfvjZS9hFi1pRII2s0uan4FgcBU126d6lWbidFFmg2HarlJNp+IHlLAiuk88OAIr7x8nqmpEsXiWupbe3uKD35wP5/4pbtWJYdSuQRW3CAMQqrzNZLJeOuaVuPqvU1lY7R1rn64TEsnkY7i9lcRS5jkClF46GrH387ODGEomZ+vkEpZGJtosTpBHTuokdByZIwuAumT0PI0g8pN8TWllCzUGyRMHSeMhNEL5nB0vepqPZOmXyHEQwkmCAMXofYglLWqabbt8dwzp5mcKEZdtb9/jN7eHNlcfDmOL4TAbro0aw6mpa8ZSSklFdchbazX4mj9sWh6HoEMSehGpP7l+iwV60xNl8hvsMNas3Ct+Pdy88oV7wdByLET4/zgRycQArYOR952IZ2gkE7QaHUWvcaYERzcN8BTPziO5wdUqva617ARF3w9TFaqFO0mPek0h6eneHhwKNoFVissNRscn5vFVDXu6Lw5+/aOGd1CLMFndh5a83rT93hjboIlu0EQhpQrTSxTX1dy7Z2AF5aRcglNSZFQbj67ejMIZEhHPMFso7YcXtgIWb2AqiZoBg2GEzsJCTEUEyewsYMmnVYfhvLT83gBtu/oYtu2TuZmK5w4McHSUo09e/vo6ckRjxu4rk+lYjM7W2ZsdJGxsYWoeqqQ4hc/diePPbZr3SSbEAJNaGSNPFlj/UqfMJRRi59MnGTKIhY3ibUU2LLZBL/+Gw9y4MAAb7x+iWPHxpmYKCJEZGxHRrq4576tPPjAyLKI9Mpz66ZO52Bhlfj2RrDiJsnr9HP15fDAtftktAzx9VCUiLlxIzSDKpONk/TG95A1urHDGoYSwwlqAATS50rtMIOJg2jK2jENpeSZS5c50N1FT9YCBDFt/fPGtGgRCepPIdV2CJcQ5sOrPrMwX2VxscZDj+xkfq5KtdrkwMFBnvnJKQqFFA8+vH1ZZyD0Q/q2tDM/XSJcp+rvK2eP82u7DqyKVa4HPwyZb9SZrFWYqlV439BI5PVbBlNTJQ4fucKWoQLx+E8nPRqGIZevzPOt7x1hYqrIpz9xF4W25KrnLm4axM3VtmTr1g7a2pLMzVXWFYS/ylm/WXQmk7w8PgZAxXGouy4Jw2CmXmO0VCKu6yw2Ggxmszd1vJ/p3n6sWuLU0iyhjAjR9YbLxHSJWEzH0FTa21LvuPFVhYWu6EjpoqxolROEIadn5nn6zAWCUPLJO/cymM/e0rFzVoycFaNQqxK7wcS01BhDqX1rXo+pCbLG5l1KbxXJpMXHP3EX58/PMjtTZmKiyMREEcOI+p1FzIZg1VZu69YOPvFLd3P/A9s27Eu2Htwg4O2ZKe7tjWJX6WycdHb97wsBiYTJPfduZeeuHiYniywt1qNYYTpGV1eGtnxyw5i+EIJEeu2xK8U6nhus8myvFkSshKoqxFMWxoqST7HO524FqtBIagUMJUZW78b2q6iKgdoysKH0OV15jpiapie+c833hRAsNRrYvk9SX79oYy1ChLYN6Z1a9Wql3OCbX3+TY0fH+ejHD/Fbn32ESsXm2NExvvutw8Rb+YzHntiNrqvkO9LkO9Kkc4l1efKXykscW5jhvu7o3tZcl7dmJ3m4L/LsZupVYppOKCU/GY+q1E4uznF3Vx8Z1aCvJ8ubb4/y/R+eIB4z+cD79pKIm7e807Qdj/GJJd4+MsqLr5zn9NlpTFOjv69tw47RKyElZDMxajWbvt61O4OI433z19T0PJKGQU8qRX8my1KzQSglPak02/Nty2yt/vTNsY9+Jkb36sN9sbLIn55+g0d6hulPZRmfXGKpVCflW5SrTRzXZ8tg+231jt8IobTxggpeWCZlXuPGFhs2f/7KW+zv6+bMzDwz5eotG92ruEoX2ygZ8V8De/f28Qf/w/v43J89z9jYIr4fVZK5boAQ0VZa01RSKYvHn9jNBz+0n+6u7CqDtBFWGusgDDlfXFw2ujcDIQSZTHyTppMSkDfcnkspmZ8u8aV//yOGd3bzi791zesTQqzrvSSS1nI8H1rd1TbpV7YwXcKKGSSzq+UZQxlQ8eaRSLJGF27YwJcOnrTRRWxZgFsRGn3xPVypH142umOlMinTIGtFXndMj9g9YRhi+z6+lOiKgtkSC78ZuK7Pd779Nt/8xmEaDYelxRr/07/4KKmUyTM/Pkm53KRatTl3dpqHH9mxKnTT0Z1dN2Lw4S07eXr0An3JNKqi8L1LZ/HCgF1tHczWqzw7cZl9hS4Wm3VOLsyRNizyVgw78CmYce67dxuHj44xNVPiC196iVffuMhD94+wd3cvhbYUmnZNTjWSTIz0nxcXa0xOl5icKjI+scT4xCLFUpNKtUmz4SIEOEHI57/4Ehcvz/HYwzvp78svc22v7qCvjC5w6vQUJ05NMDq2yN2HhulcR5f4ZnWwvSDg8PQ0r02M88SWLewqtKMqCn3pn45R8o4bXTfwmahXyJsxLpYXUYXCvZ0D9CYydG9LsT2MWnRX6zamod90G+ybhaZkMRSFwLNb7cmjmK6uKgzms3h+QMoySVq3tvWREAlzKwLXjqrqmrUmqdzNc1J/llBVhbvv3sKWLR28ffgKx46OMTNTwXY8EgmTnu4sO3Z1s3//AG1tKQzj5gsKKq5Dw/MoxOIgWKVYZfseuqLeUozseoT41LxZknoXUgatRp+rPTHP9ZkZW+TH3zjMi98/Tnur+u4qhCLW7e4bS5pYceOGnwMoLlT5z//Lt3ngfft49CMHVuVIA+lxsvw0U82zUSm6MIlrGQ7lP4ovXUw1mgcKKgPxOzhbfQmItuLfPX2W8wuL/MFD9zGUy6EpkQrr+cVFPvfm25Rsm6F8jl8/sJ/OVJKpSgVdUWlPJlrddB1EuLT697YWMtPU8DyfrSMdtHekSaUsfvlX7+drX3mdzq40v/CRg60Y7rVGo6qqrHu/BtIZJmpl/s0bzxOGkl1tHVRdhz96+xWypsV0vcrh2SkSus6WbJ7vXDrDfd39tMcSaJrCQ/eP8PaRUd546zKVqs1bb49y7MQEmqZgGhqppIVp6cvMmWbTw3V9giAkaLFewiAkCOVy9eG+vX0U2pIcOT7O3HyFb377bZ76wXHiCYN8NkEYSqo1m1rNIQhC/CCiT+ayCZ58fPcGLIWbS6hrisKd3d3c0dmJoW5MA4w6mkclxDcjQfCOG92K6/BX547wkaFdTNYqDKfznFqaY2++M2oy2bqmXObmtDZXQkpJM4jaiFzfbeIqhFAwtS5C6WD70xhqHk2kSFsmv3zoDv7m8Alc3ydh6ITSwwtLqMJCFbENlYYgMriVSo1mxWbq8hxtXVl8z3/HjG6z4WAYOkEQrJtVvhlUgxodHWmefP8eHn3vDnRFXyO+LaUkkDbRjZAEoUvU5Xj95JEXBLw5Pcmrk+P80s49DKSzq95/YXyUXW3t9KbS634/DCVN10NTlVV0KtePsv2GpiKlxJc2ZXccL2yQ1nuJadfOU6s0eev5s/zob96gUbeJxdeGpNRWmfP1sOJmpH3cgqKKDQsNZieKTF6e5/yJcR758H5WuoO6YvFg+6+v+z0/dMkbflTyHTosOKOYSuTVX14qMpDLoCiCLx05xj9++MHoAVYUJssVhIBf3r+Pp89fZLJSYaHR4CtHj2PpOr924A62tOURShbpjyG0oWvXo6s88eQeNE3l7JkpfuEjB2lvj4SG3veBfQwOReXPA4NtrQq3kOmpEieOT3DnoSHaW8VAV4VyGp7Hy1Nj/Maug9zb3c/ZpXksTediaZFH+4dvqnNJZ0eaz/7WI2QycV5/6xKLi3Vc18d1I7H5Yqmx6feFEKSSJu3tafp6ctx79xYevG8buqHxk2dP852njnLh0iyNpkuj6bKwUFtzjETCZLC/jUceimQiNz7XDX9Oy1lT11S9rUQoPeygSCAdDCWNqd44xPCOG902K85v7DiIBJacBh8Z2s0rM6NM1Mtsz/4UrXWJElmjtSJXqoscaOulfR2ZRSl9XH8eVcQRQmuJWEQD2JlO8qt338FfvnaEpXqT7oxDw7uCoRaI6YOblid4jsfSdIPAC7ASBkEQEEvefiKs2XDQNHWZ1nPm7TGGdnQxM77EroMbT5aVsAOHolshoycxVYPx5jRZI03Dt6n6NdqMHKpqEIQOgbTRlBih9Gn6MxhqFoFC3ZskpnVhqusnyspOlP3dks0xVavSk1y9tVpo1jmzCF3JJNp1q7zrB7x9cZLz0ws8tHuIoY7oHDXb4YUTlylkEhza1kfDX8ALbULpR4UiK54Ix/Z45YcneOPZM9zzxC627e3ju198ec11qlokcHP902TFjWVPD1p196a27lPX1ZcnW0gxfnHuhuyUldAUg5q7yJnKcxTdKQLpcSD3YQCOTs8wnMvSn8nwndNn8YIQQ1OpOQ77ujq5sLjIt06doSORwPZ9vnXqDAd6uhkrlTk+M8uWtjxq/FPrnjcWM3j8id08/MgOrBUaBIahse+Oa+EfKSVTUyX+4nMv8PbhUbq7P0ahZaCnahVemhzlUrmIpgg+vGUnoZRcKC3SFkswWatELeJvoiBOCMHwYDuf/cwj3HlgkGMnxrl0eZ75hSrVmo3r+AStxKthqBh61Kssn0tQaEvR0Z6mpzvL0GAbW4baSSSuxYPf/+ReBgfaeO7Fs5w6PcXUdIlm00UiicUMOtvTDPS3sWN7FwfuGGB4sLBJFZ8glbQY2RqJ299shLDsXkLhaif0ObywQSg9qu4oCb33pgWu3nGjK4RgIJVjplElZ8bYkS3wzOQFFu3NV7mbPbYfBozVi2xLF9Y1ul5YREpQRZy4PojakpvzgoCT03Ms1RtIJJqqoIoEltZFIJsom7RKh+hhbevOkdpALu5WsTBdIpGOk2+P6EtLcxW6B9tYnL15QfZmYHOhdoXhRD85I4MX+hwrnYl4kYpJVo8MpC8b1N0xNDVJKF3coEzZOUPG3EkzmEYV5oZG1w9DTizMUXFstuXb0BWF9vi1XcrD/UMcnpnCDyXX58NK9Sb/6QevIiUc3NKz/PpMscpXXjrGvsEudvZ1oOkxkqIDTYm6/eordA0mL89z+ew0T3zsTg49sgPP9elYp9mkEALTMtYk5a5qfKz8nGGur/WRaUsysq+Pt188d1OMiZXQFZOYmkEYCp2xbeTNSLilkIjz1eMnCaXkkeEh4rpOXzrNU2fPU3UcLE1DENGSvnfmHCnT4MTMHDO1KiOFG2vDGqZ2w7i86/q8+vJ5Xnz+LI7jMz1dZPfeXjRNJabp5GNxNFXlUEcPcS0Kxz3evwVD1birs+em+/NBtJa15RO857Hd3HNomJnZMqVyg0bTxfMCwjB69nRdRShFLMsknUqRTgXkcr3omksom6hKuGrx1XWVfXv6GB4sMDFVZHGxhu34ICWmqZHLJehsT5PLJW6YKBUC7j40zJahdoYGC5t+diVm6i+jKUlSRj9Vb5yF5lFC6QIKqhIjpt3csX5m7IU2M87v7r6HrBk9QG6wOnt+O1CIqtI0oW7Yi0hT0sTVNureWULZjaZE239VCNoScc7PLTBTjkStFcXCdmduaqXTTBXdbFD3RjHUAsY67cQ3gtN0OX9igp7BAtlCCiklbzx3Fk1T0HSV+ekSmXwSTVNRb1Bj74QubuhxtnqJJafEaGOKRtAklJJmYOOGHqZqsC05uDzOupIkaQxHiSopiWs9BNJubYfyq1ge1yMfi/OxkV00fY++VAZT07i7u5cgDLEDn7emp8jFYusSw/0gpFhtkrqOOmS7PqVak7OT8yxW6gx1bmxcJi/P89rTJ2nWbAxTp29LO5m2BPWqvRxjh0jBzowb6IZ27XUJhqm3vMDoeFeNsxAsywzSUqCToWTPXcOceXts+fzLtldKQuS6vdIAElqekdR9LWK/WOaN3t3XS1cyGekoZKMww+7ODtKWRcP1SJoGCUOn6rgEYUguHqPctPHDkO70rUkGAq0wQpkrV+YZGemiozON4/iMjS4ui82fPz/LI4/tQlUjOcf3DERdPVYWndxMOetmuBqT3UgxTsoAxzuBlC5BeCmKiQoHL/AJwzpC34mQseWE21VDmkxa7Nx+rVik4S+hCRNdWb/d+kbo6szQdR1FMAxDJidLZDIxUimr9Vo0AwLqaEqchN6DH9oE0iGudZE1t6MpcXQljq7cXKjxZ2Z0dVWlPZag5NpkzRiGqnKhsshwKrdmG3ozcAMfXVExVY24trFXqggz0tVVu9GV7PLrgZSUm00G8lmOTsxQadotvdcOHH8zpdGrkJEUY1hFqlluar9FS05SVdi2p4/J0XlmJ4vRdldCveqQycfxnMgDuCpBuRkmmzPkrTK60OiNd3F3/g4W3RIVr8pgone5M6kmtOVxVoSOsaaEOppwKpsXZxiqymAmu+q1uG5wcn6Wr587TU8yxUP9g+sKzuiqSnsmge2tLnNW1UjPtDObumFCc8uuHvbdu5Wjr17kjWfPoGoq6VyCrr4cP/76W/QMFQDB9OgCbz57GkVRaNYdegbbcG2fk29dxoqbWCsMfzIT45UfncRxfHqHCqiqwtJchamxRY6/epFdBweWjbntecvG9sTcLIe6I499vQc8io1HCIMQu+kig5DBRDoy/K1jxnSdkcJq2mDXCvvamby9PEHUer7Ev/u338W2Pf7ef/ckHZ1pTFNjYLANRYnUy86cmsJ1fAxTo+zY6EqUJLJ9n4xltSqsAvKxGKYWmQiv1U1is/jmzV5jBAVT301UMB8dW6AShCVCUUERSYIg5M03rzA7W+ZDv7AHlKDVcVe2JEQV7KCEG9bosPasq5VwK5iYKPInf/oc73nPbh56cDuu6/Pc82dJJAweeGAbg6kPoQgVgUZH7C6QEIaRJOut0BB/pjzd+Wadb105TdmNbuyXzh7hD/Y/SM689SqsE6UZ9mQjXmNHLEVqw3bOkqY/Rii9VRQkVVHIxmIcm5xlrlpDUxUkPo4/s6mnd+2oIVL6rT/DW1pVxy7MMryzh8APuXhqkljCpLQUCf3opkquPYXn+gigvXvt1nkl/DDgRPkc/fFuhhJ9aELFVE16Y53LOhc/ayhCsK+ji30dm/NMdU2hLZ1gcnF1yERXFeKGxlBnjuwN4uK9w+38/X/5Cc4dG+PIS+dp1B12HRzk8Atn+dN/810aVZtQSgxDo70nixUz+NN//R3qVRspJW2daf7uP/sIqey1OTe8s5s3nzvDF/8/P1zWVDYsnXx7iqHt3Tzx8UMoSkSPu1QqUnFsBjJZ3CCgZNsYqkrC2HzOOLbH2Jkp7IaD5/jsf3jnqjDH7UBKSa1qU687kRC9trqjdRCEvH34CsePjfPk+/bS1x/tIExTZ+euHrp7ckxOLDF6ZYFazcFKGhyfnSVtWlRdhzPzCzy5dQvfv3CeS8Uinz14J/2ZLIKIaWHpGjsLt5eXaQYus80SlmrQYaVbO9VWtdyKz2lqB6gdANi2w+uvX2RhscbB+3LoqRq+bBCGPlJINGFhqWlMJb3uvN9sZx15/XJV9WEoJY2Gw5XLC9x7z1bK5agjS6GQ4sD+wWXvF0DFAAHTU4soikJ3d+6mY8M/U6PbGU9yb2c/371SjyQaM3mc4PbEXSpukx9OniGuGuzIdtARS+GHzXU+KUgZe9eIfiutRNr7dm3j0sJSK3suUIRBIG3C0EHZRP8guqkCPywSypuPAyFhdmKJiUvzqJrC/vu20Ttc2DAunL+BxGSX1U4htoUuq7DsyW605V0Jr8UW0NcJXwThNTX/dxKaqtCWiq8xupqqYq4QNPHDEC/00ZRIG/d6GKbG3ru3sPfuqMJQSsnW3b1s3z/AwnSJMJQk0zG27e0jlY1z7tg481MlJJKBrR3su3vLqvHu7Mvzqd9/nJ0HBpkeW0RKSSobZ2BbJ9vv6CeRieG1qq5mazVqLfFv2/c5uTDHtlx+Q6MbhpLF6SKu7ZHKJejd2snCVPGWKqA2wvxchR88dYzZ2Qqf+Z2HKRRWhx88L+CVl86TSJhs39FNLneNF93Tm2PfHf1MTizhuj6XRue4Ela4uFRkX2cnZdvG8X1+cOECoYSkbnB4eprXJydRhOByschDA4PrGl0n8KJMv1ifhlh2G5ytTlF06gwmC7Sb6ZvS/o5CQTqappI2O4mbHahKlBg21Gu7AT90WLesudLk1Okpto900da2evdw6vQk58/P8tCDI3S12ivF4waJhMXQcAHDUFHVVpeUdXQ4gOVKT98P6b6Bs7QSP1Ojqykq+/Jd/GDsHCXXps2K3zT5+3rsyXVzqjhDZyxFd2xzwxR5fGt/mh+GnJiaY65aJ2boKEInpg9Rc08TyAYKm211BaqSwtKGUJVbiLUJ2La3j0unp2jrSNPes1aF6VaQ1pP0xW+2kilCpWHz46MX6G3LcM/21UUNC5U6Pzh8lvt3DjLcmX9HDa+mqORTawsiVEWs4vrWPJuxepG+RJa8eWMqoRCCnqECPUOR8teSO0fFLxFXY5iKyQd23bvq8+ONS6hesEokvrMvT+d1TQyr5QaGGSlXjZZLQNRLrD+doS0Wp+F5VByHzA3inYqiUCs3GD87zfZDwwzs7LmlCqj10Gy6vPDcWf72q29wx4EBVEVZc6/m5ypcOD9DZ1eGbSOdq+ZZNhNnz94+XnnpPNVqk7nZCgNDCba15RnMZIjrOpeLRXrTaXKWhUAQyBCv1Q+vJ5Via3792PtUs8jp8gQPtEe6x11WFmNF8q0ZuFypzTOYaGc40XnTnG5FEaRTFo7jkV1R9aheFyrT1nGWgiDk1Vcv8uWvvMav/Mp9PPmePavuwcx0mb/92pucOTPN7/z2w/T15SOjGzdIxCP9CV3XSCZN2ttTJBLrn8PQVdoLqZv2cuFnaHSvuvZuGETyhkLhyb4R9Ns0OO1Wkke7t93y+eHa9sJUNba258nETPwgbAmuW8S0AdQbBsGjDGVc34q4hZi0EIKOnty6GfefNSLSNpwcm+X/+N7L/O777l1ldEv1Jn/9wlG++tIxXj83zh/++vtJrxCF3ow2dVPkclUhn1zf6OorWAaaouJuwr3eCBWvyNHSqzSDGiPJfVEXDm1tmfWV+lnU5O4NO3NIKVmaq/CNP38BIQQf/uxDvD03RWcixd09veSsa8ngpudhaRs/NooiaOvOopsaY2emaFbtdeN9643tRo1Ew1By+uQk3/7mYcJQ8vAjO0mv0Jm4eqxjR8eo1x3aO9Loutpq8dMSglEEI9u76B/Ic+rkJKEXcndv1OJnrFTitYlx+jMZPrht5JbjtlWvGdHPmkXKbgM/DEjpMSSSuGpS8ho80rGLU+UJKl6DdvWa07Tec3oViqK09BvWdnO+ERRFYXCwwAMPjLBluH2NUdx/oJ8PzO3juefO8l/+5Fl+//cep7MzgxXTl51mTVPWaIJcRRhKZqZL5PIJ8vlbi8H/zMqAG77H2wtTvD47znSjwlvzk8Q0nR3Z9k0n7fUIZWS0b7XjZrlpc2pmjoFclt5serneOmEYGFrUEVXiUXWOoYgYMX1zbqxARLSyG1DL1sNV43fVy5dS4oTBcvJJE2u9lp8WQRhSadj84O3zfPmFI8yX62sedE1VUJVIcakzm1qlYQpwebbIXz57mONXpgmlZFtPgQ/ftYv9w92kYjeuqVcVQSZurWE2qMq1xoBFp8miW2PJqd8SNQngePl1klqavZm7SWtZLjfOEtV6tYwVIapQMRUTRSg0/FrUKFKNLyccIWI+nHrrCi8+dQxNV3nPJw7xoW3bUYWynEiCqPW4bt6clkAym6BzoLCuroTvB7z80nlKxTpdXRly+SSJhInaitEKEc0VCdhNj1dfOc83vvYms7NlRrZ3k0iYLC3Woso6GWXdi8U6T//oJK4bcPHCLP/y//G3jIx0cuDOIXbu6qGtkCKeMOjozHD2zDSdK3QrulMpPrFrN7qq3rQ84Upcqc8zVl/g7rZt1H2Hb08eJqbqSEAXKgEhaivJuC87cG3cpaRcaVKr2nR2Zta0IBJCrFsIE4aSRtNtFRRppFJWJErfqnSr1x3CUNLVleFXf+U+YrG1SoNdnVl++Zfv5X3v3cuXv/waf/yfnuHX/s79xCxjOQSnaSrWOrTU6Nqgty+PdhNa4NfjZ+bpXigv8J0rp3midyu/v+deSk6T/3zydfYVuvilLXtv2sjYwSIKGoaaRqAgCW7IqYUoy3piapYriyU+eXAvxv+Xvf+Mkiw/zzvB3//6Gz4y0vss77uqvQUa3fAkAYhONNKQGkoUtaOVNNJ80c7ujvacObM6Z86e3RnNSDtDiSApUgJBgBIAwjUIoL03VdXlbXqf4c31dz/cyKiMjMisrDaQzu4+fbq7KsyNGzfufe/7f9/nfR5FxvV9zs4vossyw+kkktDIGI8S0qm49FHAbQZWPwxZa1QZjqcp2g2qrs3l4ir7Uj3IQmYikcH2PfymBkHZsek1YruaX+4G2/W4MLPMH//0HebWSiQMnUSXkydh6Pz2Mw9ydHSAM/uGOzQw/tfvvsIb1+bIxA2EELxyaZrnz9/kCw8c4ZceP8G+wRwxXdtxaSWEwNAUEmZ7BisJ0bq4Y6rGXMMlo8Xwgp01ScMwAARe6LRoWVPxI1wuv0tAwMnUw5ufCoAV1FlszLAvfhRdNik4a1wpv4cTOBxPPcBIbOrO/sgSDz19FNfxSffEGZ7s7Sqssv2crdcd1tcr9PUlW0peYRjSqFrEmvZTm5Sjrbh8aZGv/usXmJvd2LJtUFUlGhpommwGfkA+X2vZMAFcu7rE/+X/9Ockk0ZL0KZWs2nU71jSe17I+lqF9bUKr716AwBdV0gkDWzLI5E0OHT4Du1KlWUyZmdz27JcNE1pW5bX6jaO45FKma0ywZdGH8T2XW5Ulql6Fp8bOoUuqziBx4CRZs2ucDDZWRJzXZ9vf/s9XnrpKv/n//bLTEy0r1AkiTZd3s3ju7JS4s++/gZvvHGLyckc/+Xf+gT79w9Qrzt897tn+e73ztJouBw9MszP//xpjh4dJpFov1lKksA0NMxhjd/9u5/iG994iz/909dIpc1WKSGyau8eIoUQOz53N3wsQVcIwX29w9zXe4cQH1c1fvnASf6ncy/zlanjKHsIupu0kIa/ToiPF1qAIKl2Op9uR1zXODrYz+u3Z1mpVBlOJ7m0tMoPLl3jwYlRhtPJ5r5GTrwfB25X8tQ9FwGUnAbD8TTXSmtcyC9T9WyM5olZsOsU7AZuEInTrDaqPDtykKG7+MFt2nxvvSiCIODc7UX++CfvkkkY/De/+AniusY//8bzXbdh6ipPn9zH4kYp4lZuKS9M9PdwbHyAQ8N9SJLg+uI6Pzl/g+++fZkLs8t85vRBTu8bZqIvSzZhdrWp0VWF+LaAv1XlyfY9evQYN8vrVFyLrN5ZjnADh1V7kazay7I1R0xOEBAwbE7Qpw/yduFFVu2FtvdYfoMb1QsMm5MoQmXNXuZY6kFKbp6Cu84IU22vN2Iaz3zl/l2P93ZcvbzIt7/1Dr/+m4+3gpjvBVx7b4ajD+1DSALd7EwQzr47zdoWjV+IlLGikVkPqnbHe+IJHUWRKBWj5nGlYlHZphUrSYJjx0cwTY0b11coFmtsLm5s28O2o7HZp5851taJ74YwhJn5DSZGc60ldqVqMbdQwPU8jsaH2XqPtgOP2foGI2aW0VgOy3eQm43RuLJz0zHXk+D4sRFM844+xB1zStGhhua4Pm+8cZNyqcFnP3uC8+dm+fa33+W/+q8+Q61m4bo+Tz55GENXuH5jhd///Z/y+OMHefjhfUxO9hGLaR03z3hM5+e+eB9f/epLFAq1VnYtSeIDZbJ3w8/UDbjfjHM427dnUpMX1Gj463hBnSB0UISx54EEgOND/bx2a5bpjQIVy+YvL1yhJxa5kG7UGgyl7518fi/QJYVZq0BWMzmY7mOxVqbi2hzK9CELiZii0vCkO0LQTQ+pQTO141I7DEPylTquHyA1hTt6U3eaT/lqg7euz3PfviG+8shxelNx5tdL9O2idbFcqPDVH7/Nrz11ui3o/tYzD5CM6a1A+ujhcc7sG+YvXn2fvzp3g3/93JuM9WaYHMjSk4gx3pfhs2cOMZi9c1x1VSbZJdPdzJKkJs1NFlLbdw7CgLpXJaGmCAg4V3wNSciU3DwZpQdFUpmuXcENHEw5gSlH32/TgcOU46TUHp5f/Q41r4SpJJipXyOhpBgx2wOuFwQULQtdlknqe68rX7myyNxsvq1mK0Q0knz2xcuUN6pdtXsHhzIkkgaW5QKQzsQYGsrgOB7FQo1qzcb3AmRZIpUyOXBwgIcf3Y+qyly6sMCtW6ssLRap1qxItczUyPYkOHR4kC995QFSaZOrlxe5cWOF27fWWFkuUSzUsG0PWZF4/ImDbaPREJ1X84sF+nqTGE3tj3yxzrWbK5w4MkI2G+PVN28yOpSl3nA7hKpSqskXhk+3/p5QDW4t5RnrS5OVktQsh4bjkjB0jGaGaBgqzz57jKeeOkQiEdnMX768SLVqcd994xiG2sG2sRouN2+u8tu//RRjYz1cvLjAv/mDF1hZKTE0lOFLXz6DaWgoisTKSomf/OQyL7x4hTffusW+ff0MDWX45CcOMzZ2J6v2/YClpSKKImGaGktLRQYHIuNORZZaKwjX96k6DtktqwLL8/CCgMRdKIRb8TMNuhnd5DcPndl791JomHIfkqIQEkQTH+HeSwHZmMlDE6O8fnuOpKHx2aMH6UvE+drb55krFD/2oDuRzJLSDOKKiirJVFybY9kBcnqsrbt7L7i2tE6tcBM/COhJxDg22t/2/OxagdevzvKpk/vRmvXIuKHRl9rZK+3771xhbr3YEZjT8fZsSJYkTkwM4gcBy8Uqb12f4+byBjeXN5CE4NOnD/LlR461vUdXlI7SRqTcvzktJxMSElc1YlsyIi90WWjc5rB6H5rQ6NH6SaoZDkknkYSEIlS80MUPfVJKlqzWRyKILpQwDLGCOl7gMBU/jCKpyEJGEzppLUd8m5uv4/tc3VgjY5gc693ZB2w7ZmfWo/HwLXQkSZY4cGqcC69dZ+76MpPHOldljz1+EEkS/Ns/epn5uTy27RKGIc88e4wDhwYJ/JAgiKbqDENFSsocGR9CkgSPPHaA/EaVcimSRhVCoKoysZhGrjdJb28yMhkdyfLoYwfJ56tUKhaNhoPn+UiSxOEjQ12be4VSnWK5wehwlunZdYSA/r4U5y7OceLoCCtrZY4cHGR+qRiZTO5wXGzXi/R382XScZ1izSIVM1grVVFzcivoQhR4NzPpfL7Kj/7qIo2Gw9S+fgxDQ5Y7ucgLiwVqNZswDDl6dJgjR4aZnl5nYqK3teQvFGvcvLnKwmIBx/bIZGL4XkC1arUxEcIw5NKlBb7zl2d55OH99PYleeed2xw6OEgyaSLJEkEY0S1XajUur67ymQN3GvrrtRqO75PYgdnRDR9Z0N3LiK8qyeS6LB93giw0TGWb4Pc99JuEEDy+f5wTwwMR/cTQqTkuMU0lCEKWSmX6k4l7kiXcdJ4ViGb5Y3d+a0YzWvuS0gxSWueybnME1XU81paK+F5A/0i2TQd2E6os88SRCQw1EpM2ttUewxAqDZs//uk7fP+dKxwa6aM/nWB+o8Rob5py3UKRZVRZwvMDzk0v8YN3r/L4kUkgumAgKl1sLvX8IMRyPa4vrvPjc9d55+Y8fhBycnKItWKVM/tHODjcy/HxATRFwXJcQqJmXsP1iBtaW0Nq67ipLsn0G0kantv2PYIwYMNZYcNeoe5XMOU4++JHWhltN+iygRe4vJn/KSv2HKfSjzIRO7hjEzYMQ6aLReKaStVxsDyP8VSaNxbmMFWVJ8YmeH91mVuFApbrcbi3l8VKhSfGxkkbBhsb1YgBI23NdAVmwmBoqo/5G8sU18pcPzvD1PERlOZvlUgafOLpo0xO9fHnf/YGb7x2g9u3VlldLfPsp4/zW3/rE6iGjBcEqLLM9eI6M/UCDd9l0EzSO55mWM5hykpTQawIKEhbGDhCCOIJnXhi75l7f2+Sr/2Ht8j1JFjfqOC4PplUDNNU+dELl3j/0gJLK2XOnBzbkfpZadi8cmGa0b40MUPFD0JuLm4wnEtRrlsM96QIghDP8yP9hbY6q4QQMD29RqlUZ6A/4vNu12BIJAz++T//S3K9CTKZGCvLJS5cmOell65i2S6VskWtbiPLEvedGucf/5MvMDyUQVUjvu3WoPuDH7zP1/7sdXw/akbWaja1ms0v/MIZUqIpA9kMOqaisF6vU3Wike2YqvLG/DwFy2IklWo7x3fDRxJ0wzBktR5F/OFEkobnUXMdYqpKQr1TQ6m7Lq/MzfDU+CRe4BOEIUktKnDv9CM6nk/dcUjo+gcSO9cVhb5k9DUXi2W+d/EqxYaFLEn87y+/xd9/+jFy8b3fCLzAY81ew5RN6n6dYWN419fvJWtyLI9yoUpxvcLizAayIpHKxroG3bHeNAPpnXmBpyaH+G9/9Rn+4tULXJ5f5fWrM/SlExwa6uXbb1zi9auz7BvMMdmfZXatwCuXZyjWGnzj1feZWy9yZLQfSQhKdQvLcSnWLFZLVWbXilQaFpm4ybGxAX75yVOcnhri6y+d5+svn2O9XGNuvciLF2+jyhJBCKulClfm13js8DhDPVvq06L93mnKKofT7f5SmqQxbE7w4vr3yGn9nEo/iiHFCIMoC5Tk7owPWSg82PNJIERCuivr5fL6Kl4QUHddDvTkkCXBtY0NemIxao5DzXFpuC511+Xy+hpxVaNkW6QNg3rdafmCpbZZBY3sG6Dnb2bwHA8jpndoaiiKxL79/fyD//rzvPfJab73nfe4fGmRH/3wfU7eN87wqV5ulTc43eyLrFvR9bXRqCNLEvtSPZjNmr/rXkWSsghxEPCb6nouEADqXY+B7wesbVS4cXuVL376JPsmeu/4Ujaff+f8LKahMTXey4P3TbRq8kEQUKxaJGM6qiJjOx6qKlFp2LieT8KMRopdzycdj5pvlUqDt9+Z5sknDqFvEetJpUxOnBjlnbenqTYnCuMxveUQDZBI6Py933uGH//4EjdvrVIs1llYLDI8nKFQrKMoMv0DKfbvH+CRR/YxNtrTkS1vxepqmVwugSxH9LDeXJKjR4dINm9W2Wwcv8lV1mSZmuvwJ+fOcjOf57fP3I8XBpiqwtX1dU4N7o0//5EEXT8Mubi+ylqjxgMDw7w4P03NdRlJpHh0eLQlByiI6mcLlTLr9Rrr9TpnBocwFJXeWPfAV25YXF/eYDyXZiiT+lAk875knNNjw6xVbxDXNU6NDLb0RPcKy7eYrc/So/Ww4WwwaAwif8iZ73rVorRRZWAsx4GTY7uvFuTu0zGt5xWZhw6OcXJikOuL61xf2sDUFA6P9PEfXr/IT87d4L1bCwRBSNI02DfYwy8/cZKLsyucvbXIixdv43p+M4MT6KpCwtAZyCR4+OAYTx6b4hMnplp12l964iReEPDC+7f48bkb1CwHb/MkVWSOjPbz+NHJtmaaICpVRJQ50RxmaYckZIblffSYI6R6EtGE0UYVPaZRK9bpGcp0/f5CiD3/HkIIvnjwcNtjVcfmoZFR+mKxVp33iwcPk+pS65UkQaVicfniPBNN3drWtiVBbJcR583XxmIaTzx5iCNHh/nm19/gRz+8wMZ6lYPKMH1mAk1SOJztbzs+BbveNrknhNrkjvu47kVU9SSedxPXvYSmPYQsj+waeD0/4PqtVXw/4MC+/o4ESFFkDh8YoFK1yPXE0fU7o7ML62X+/Kdn+cpTJ5ka6qE3HefZ0weBaNV0eXaVM/tHGM6lcD0fWZJYLhY5d26Wpz/ZfuwlSXD82Aj79/e3At3+/f1MTt6ZABVCMDSU4W/8jccJw5AbN1b4oz9+hX/0Dz/bMaG3F/zWbz256/OPPrIfy/Nwfb/1nSUhSOk68+USF1ZW0RWZ+/YYcOEjCrqyEOTMGGXHxvY9EqrGRCrDtcI6L8/P8KtHTkYqY4RYnsfNQp6laoWNeh1JCA725HYMurqi0BM3PxJNAVWWOTUyyIvXb5Ov1SOpSP/e6GJxJc6oOUpKTaFJH42/m6rJxJKRrcxHxdc1NJWTk0OcnLxDDfovPvUA9+8bYTFfIghDMnGTwyN9TA5kqdRtzt5eZKlQxnE9QqJml6kppGImwz0ppgayHfzcdMzgt555gEcPjTO9WqDcsKKgDRiqwsHhXk5OtJ+QuqrwieNTJEx9V+uc0nqFuauLDE72UVwvs7FYoG+kB0VTyO0wbOKHAdJdSj67IaHpPDQ80vr7yC7WLNlsnGtXl3n91Rucvn+SoR1uBHeD6/osLxUpFOrIisTQUBpTUUhrRldnlW4MjwgernMWRZnE9xdwnDcBF9P8CuwybalrCk89enDXfcxlE3zuU8c7Hr82u8pP373BgZFexgeyKFuszXVV4fT+OyvBzaZYEIQUCjUqFYtMprNc1D+QIp6IaIqyfMfTrlRq4PsBiYSO43hcu7bMf/zWuxw4MHBPPn8Q+bAJIdDvQvtSFJmFYoGsYSBLEhKCsUyayUyWlK7zxPg4GcNgf/ZnXNMVCAxFIa6qHMjmON47gNec51+oRNSY28UCA/EEA4kEA/EEw4norlSwGgzEd57oCMIQ1/dJx40PPUoJ0V3KDyKZvi8eP7Tn6RvX86nUaiiyTK86gKGqpNS9eSUFYUjRaXCrvIETeBzJDKAIiUvFZR7sHSeRjpG4x5NmJ9h+npo7T0KdRNsy+ROEPrmUwdMn91FybpLW9rcCk+OXySQ0njo5iOPHaXgrZPSDXaXqwubvsfW46arCqakhTk0Ndby+G3RV4ZHD40RZ7s6vi6djjB4awrVdzLjB1ImxiMPapeyyiZlyAVmSuLCxgi7LHEj3Mpn6eKYB9x8Y4M3Xb3Lu7Ax/8edv8it//VH6+lN7Hgn1/YDZmXVefvEqr716ndu31/jUp45x7PgIcc0g2aX+3x0KrnsVq/EjfH+BWu0PCUOXMKjhebOEoYcQd45Z0LTD+Shu8F7TZmdutekwvIcSoCxLlEsNXnvtJs88cxRNU3Acn/PnZ3n+hSsMDERi5tsxPb3Gd75zFtf1cD2flZUSgwNpnn76yD1Tu+aWC6iKwuTI3YNlEIZ8++oVBIJHx8Y43t9/1/fsho+mkSYif6WheBLL83hxbpr1Rp2laqVFdl+tVblVzPPA4DB9LVqSTMWxSWjdL6KKZXO5skp/KtHRMLoXlBoWr9+e5dhQP89dukHdccjGzKjTTaTcZXseqizvyCpwfZ9bS2VcP+DgSC+GtvfJtJpn80fX3mS5UWY4nuZScZknB/bxxtoMJ7PDKPeYMQehhxuUkYSOF9TQ5WxTayLADxo0vCUMua8VdIPQp+GtIQsNXc5g+wUCPEQIAgUvsAiEQ8m5hRtEqlteYHUEXc8PeOXSbV6+PM3vfu4RelPxD3Thbr6n4dVwQ5e4kiSqwcr4oU9IgCJUYkmDWNLA96Kl3U7eZtF3jOT+nMDHcR3O9A1zubDK2fXFPQXdMAyjG3zoo0lKW7NvJzz+xCH+8tvvkt+o8YPvnefWzVWe+fRx7n9gkmTKvPP+MGrABkFIo+6wvFzk9s013n9/jutXlyiVGoRhyLOfPsFv/s0niHW5qWyfatwKVT2IJI+gqifYrJZHg0QuICO2OBrMrRb49z96j6984gQHR/fO1NgJhqagKhIPHhnbxamhHamUybHjI/zJn77K889fJhbXKRRq1OsO9903zqefPd5Vh/fAgQE+/4WTvPLKdfIbVZ55JpJgHB3trhmyqcW7mawtr5fpzcSRZClifhAxXTwv6hHs5Ne4L5slbRgQhvTFo8zc8qtISNhBDZrzBIqkY8h3L3F8NDXdIODC2ipFu8GtYp681eDLB47y3ZtX+cxkRK94dGSMkBBZ+NjuFUJcdPXIrndzSUjYXiTB9mGMD+uOw3yhzFMHJjk5MshCqRxRWurVyAKoXMBQVPane8iZ3Q+JLMuk4yaFaoN7LAMTV3T+7tHHCUNYtSr8y0sv88nBA4zGMnihf/cNbEPFucW69SYj8c/hBhWC0EUSKpa3jB/aTanKLTPtSDhBGT+wcIISTlBmofo8kpDJ6IdwgxpB6JHS9rHaeJsgdNvev4kbS+v8i+++yuJGmVTM4Pc+/2hX1bK9Iu+sc770FvdnH6PsFunVB1hszLJkzfFY7lniSlOAfpfPCJuB8mp5CQnBkfQwIVD3HG6WNhiI3f0icAOfNavMxdICb2/c4jenHmcslrvr+8bGe/j5L93Pn3/tDRoNh/PnZrl8aRFdV8hkYxhG1ET2PB/H9qjXbWzbi0wY/RA/CIjHdfYfGODzXzzFo48fJJk0OgKI43mUGjYNx2U8l2l7zg8CbE9hvSLTnxrCUDv5t5v/rzYc3rw0y7mbi+iqgvyExNhAFu1D/IampqKpCqmYvucCYCym8Yt/7UHS6RjvvjtNsVhndLSHT37iMCdOjGIYnWO7EPmf3X9mgpMnRiNOuyx3yFtuhe14bBRrjAxkCEN4/+oiJw4OoSgyiyslDk32E4bw3uV5xgYzDHVxDoaoxDmcTBKGAQEBll+l6q1T9wpIQsYLHGJKFpPkzy7oQuScGQleZ1lv1LldKjSVpKIffXM5Gp0EDhIakthdVzcIo2DrNd1CP2jgHUqn+J0nHgSigYmfXL1Jod4gEdfYaER1Zbk5mrrTiaNIgoSpsVKotGQS94ogDJmvFVmsl0lrBmXXIggjOpB/D7xjAC90CHBJqJNU3VmcoEQYeriBjRABijDQ5CxSczkZmT7WcYMqNXcJxy6jS2lUOYEbVFltvIMuZUjrBwlCBy9oEFeHumoM96UTPHhglFuJDR4/MvGhAi5AxSsRk+M4gY0QEm/lX8aU42zYa7iBs6dtVD2bF1Yu829vv8ynBo5xLD0ScVclmT7z7kIkTuDx6up1/v3Ma5wtzOAGPgdTg4zGelp9hK2d/E00XBdFkfjSVx7Atlye/8ll1tbKramy7dNiW6HrCv39KSamennk0QM8/Oj+rk0gx/OQhES+1uD68jpeEHYEXdvzWSiUeH9uhacOT3YE3c2AlC/X+Mbz53E9j9/+wkOcu7HIf/9HP+K/+fVPcXyqexMoDMFynOY4d/eVnaYqu7KKXM9nOV8hpqvkmjxwIQSJrMFnv3KSX/mVh6h6DrIQJJoa2V4QINE9q5ckCV2PPi8anQ92NEWo1h1efe82kyM9SEKKXIYlQbFSj7LebIJStUGtydS5G2p+EcevUveKuKGFLsVRJB3XbzQHfPZGz/toGmmSxJm+qJ5n+x6qJFGyLfZleqi527+MYO9k2xBdkSk1bBzPx9Q+WNAtNSy+8/4VVFlCElFHfpOfe6pvsKuG63bIkoSuStiud8/Nt7Jr8W+vvx2JgAhQhEyPHudL4wN3f/M2VL0So/I4ujyOFdRJKQfQZIOis0pOH0IWCuG27DkIXVQpTq9xEoC4OkwY+rhBFU1OtWlZ7E//tZbI0HbkkjF+7wuPMr9e4vg97Lsb+NQ9l1hzSGQTphwjpsTJar2sWktokk5ciTMR348u3/0ELjg1vjn7Fl+feYN1u8ITfYcilwdES9/B2aLfvMkbTuh3dBIuFRf4n6/+kOnaeut172xM85XRB9hcbVZsG0FEGSpYDdK6wVqTIjmVyfLXf+MxDh4e4p23bnHr5irLyyXqNRvPi+zOTVMjlY7R15dkYDDN6GgP+/b3c/joUNdGEkTmna/fmuNgf47eZJyK5ZCNdyYphqIQhhGveTfZ1GLV4vyNxYix8egxbsyv05PaWWrVdj3euTLH21fnkCWJBw6Pct+BEcxtTtWqIu+aDG2UavzLv3iFx05M8PNPHG99Xt1zWbeqWJ7LfK1EzXMYjaepujaqJDOZ7GkF4Z1QchvokoLtu6Q1k7xdJ6HqJNVo9RwzVAZySW5Mr6GqCgcmesll4liWy8hABkWRuDm7jiILkvG7189tv0oQevQbB9DkLT0YY+/qh/AxTKTpssIz4/uouQ6O7zNfLUfDBC2FLZsgqCBLdxcCNzWVsViGSsNmbqPI/oHcB8p2ZUmi4bhcyRc5OtSHKkkkdZ2hxL1RTExNpScZQ4goe92rNrDlu+TtOr84eYqv3z5Lv5Fojbxa3gYhAaayN0V+N7BZteaQJYWGXyMmJ7HtBgLIaUNNMvlWOpFAlzPocqYlGiMEhMhRLbhbNrGLdGUmbpLpcvHbfp2iu0y/PknF28CQk2hSdCIX7QbLjQqTyR6W6mU0SWEwlmQifoAhc4z3Cq8TEvJQz5PN+u7dsWaV+bOZ1/nm7FuU3Ebz2PiRyJAsoQiJrGay0qhyY32DtGGgyhLrtTrDUopYMyP85txbbQEX4HJpgYprkdZMqo7D+ysr1Nxo/LPQsDBVhVhzbDskqlF+8ukj3Hd6nOWlEvl8FctyCZrSipqukIgbpLMxenriJJPmrvYuru9zcy3P9eV1JnoyQMQxH0hHXmvFeiQ4Xm3YCCHI1+qUG516DVsx3Jvid37+EWRZYrgvza9/+n7WSzWmhjsbSX4Q8OLZm/zBd9/g5sIGQghePHeTX33mDJ97+HCbgJGqyLtmunXb5fVL02iqzOMnp+htZrtmc5Cg6jmMxNPUPRdDjtT/dFnZk9iT7XtcL69ScGoMmmmKdp1ePcHRzCC6rBIzNZ56cD+ziwV0XaGvJ8Fb52eYWyrw4IlxpsZ6SSUMwnBvTcW0GiUa3fR77wUfedDdaNT5jzcuk2/UqbkO51aX6dFNnpnYx3gq0+QUmoS4d51iUySJTJO1sFKq3nMtdRNxTeVLp47yF2cvUrWiZetep0c2ERKdYGP9WdZLVRq22yHkshN0ScGQFX6yeJ0nB6a4WFzG8j1SNJ163QVq7gJp/SCy0HZVUcto/ehyDD/0yKp9mEqCyF1K3jVYAm1czeiQf3RyklVvg/PF53g098u8vv4NTmaeJatOUXFtblfy2IHHuY3FVuYZU1RSmoGMwpR5CE3RUTwDy3XQNCU6N4Qg8IOWTf0mlhpF/ujWS3xv4RxV784y3gk8vMAn77rcKucJwhBdlrm9UYiy2+b5I4D9vTmWrRLv5ac7vkvBqXOxNM+x5Bi3iwX643HiWpaSZZHWDTRZpsc0sTyvdeMVQpDNxluZ69ZTequIy/ZzPZKg9Nsu/BAfQYCpKfzkyi0sx2M4k2ShUOLy4irr1XqT8ypwPJ/eZJxs3OyaBGxeY4amct/BiAoXBiFxUyOTjIwfN7WlN4PnrYUNXr0wzekDI/zWFx4mkzRZXC+RNPWOa1CRpV1ZRXFD4+jEAGevL3BtdpXcicmoXCGrjCUy0cpTUroel036X91zI21eJepVbK7C+o0EpqwSEKIKiTARuahsD9gDvUmk5gDW/vE+xoay9OeSzedSUXlxD6yLDxtsW9v5SLayBd9qBtxfP3qKl+dnUCSJsVSaf/Xem/zjhx6lRy8hmkHFD1aQpT52M3nUFYXeRMTV3am7uB1B6CKQWhmfEIK+ZJyfO3GYN2dmOdjXQyZ2bwfQ9wNWihVuLK4z3p/BvAdZt7Rm8rcPP0bVszmQ6uVcfnFLA03CCao0vJsE+OhymrS283JFFRoDxlhLge1uY8ibF7VE+8jlbuLR3bcTZcl+6CELuYNsHzaNOwHsoI4XuNwor+MGAVPJHlKaQd6uE1c0qp7dWjpWSxbnXlrm5AOTrFfWseoOtarFykKRQydGcF2Pk/dPtn3W1fISzy2+3xZwIQq6buijyyqW79HwI4W3fbkeIFqZxDWtxai5XV2j5nXWjq3A4fX1mzzae5CT/QPIIipLbVIbNycoE1rYdtsKgoBCvkYiYbQEZcIw0hR4+cVrfOKTh8lk2xkfBXed88U3cQMHXTaaZot1UGSePHmMAX2MMAzRFJkwjLLQSH8lbOlMyJIUaRRvq7HXLYea5dCTiiE39zkMQ/7jyxdYXC/z93/pSTw/4Os/PUvdcviNzzyAqalcvL3MUC7Fr3zqdCR4JAl8PyCEjqxWlnYPuj2pGD/3+DH+H197njcvz3Ji3xCpuIEkRJvWxnb4YcC7q4sczPRiBx4Nz8P1K6w3auzP5MjqJrqsosvdG4et34QQoUQJXEhILhtvuzltL5f8LPCR65YdzOYYTaZZqddYrFY43T/E0+NTrFt1iraDIo+gyENIUhrPXyMMd2+YbMoAKnJ3/6VOhNj+GnV3Bi+o4QV13KCCIGAkY/ILJ4c50O/e9XO3Q5YlBrIJ+jNxDFW5J8sdWQj6zUSroXY43Y/ZPFkUYQABklBpeCuo0t25v5KQkYWCtAfxcze0uFF9k3VntumkGsELbfLOAlawN1X+vLPAqn2Lm9U3qXgbHc/LQiEmpzHkBDE5hSypjMYzHEj30mcmEAj6jAQJVSej3amHra+UiSV0FmbWKW5UWV8pUSk1UFSJuem1lk7tVmiSjNHhbhzJC7qBj6moHEjnWkLg+3t72N+bYyrXQ38yQU9z7LvsNtqOSeuYBT4XS/NUvAaarLSscTaD2+ZFu9VXLggCrl9b5n/853/JD39wnno9Wu6vrZX5g99/gX/zv/+U3//ffsriYrEtMDS8GnElyWcG/1pzxHyJZwe+zERiCkdUSBo6KdPAUFVMTSVh6KTM6LHN5+K6Fp2TW4VhgoDvv3GFP/nhO1hb9HiL1QbPvXmVicGIRje9nOe1C9MM96bRVQXH81gpVnj/1hLFSh1ZioK1qshoTenOMAxbE2Obz+8EVZG5/9Aojx6f4LWLM6wWqzu+dis26XEvL05Td12mywVmKkWqrsN0udBWq9+KqmdTdhtU3EgqteTUuVpeYsOukndqbNh7/Pzm6mQ395QPio88031iZIJbxTwL1TJnBoZ4YGCYMAw5lusjo5tEXNIyQVBBUw+0cQg/KrhBmbJ1HlMdQRYx3KCIoYzg+OvIIkYQ2qhyBpm9f7YgOoF60wn0e+QMN3yXb89e5GJhCUNW8QK/1RE3lBxjic8CUHMXMeS9T7bcDWEYUvOKXCm/jCHFeaDnF+jVI+V+y69xq/o2A8Z+xuMncQObG5U3sIMaCSVHTh8jpfa1MuSqV8AOalhBDTewmt5kC6zbM4zEjiKhYMhJFKGhSpHHluV7DOpJbN9jrVHFVDQs32WxVmIqlaPXiON5AQ88dgBjW3D1PZ+l+QKxLoItMVknqZosW6W2x93Axwt8BNDwXL4/e5Vf3ndyx+MTNGuy3bBhV7lSWuSxvt0ntSAKuNeuLvPHX32Js+/NcOvWGpIkePKpw3z/L8/y0gtXqNcdXnnpGuMTvfzSrzzc4ZLwUaNhu7x1eZbVQrUVIAGuzKxStx1O7BskCEKuza4RBCEHRnojh+wwJBUzuDS9zDdeOMcvP30fEwM9IKDasJlfLXJjfp3h3jQPHhlD2hZ0N0sVW5kt/dkEDx8d5/2bi9Stuyc7ISGKJNFnJtifzpE1TPalo+vCD4OmI3PnNRgSUnLrlJ06duAyFutlySqwapfwwmi0XZMV+ozd+wZhGDI/m+et12+i6QqKElmsq6qMrEj09qXYfzCHqjQIgwpCHr2rvsVWfORBVxKCA9kcB7J3eI6O7/OLh46TNUyCsIzjXmsu/ac+srHXrVClFCn9eNTFFxK6PIAsdFQpgyIlEEh7pndshRCiTSt2r6i4Nu+uz/E3DzzIxeJK0zKm89DHlAE+ysVHSMiaNc2Z7BeZrZ1joXGFrDaELFQMOYEiaeSdBUZiR6I3CPACl8XGVS6XX0STTCbjZzicepwR8yhuaLFi3SSjDVFyV5itn6fmlfBCl1HzKKacRIiIJO4GIVcKS3iBz1K9zEK9zKCZJKubXC6u0mcm6DXi9A+lO4SqIZJIHB7vTno3Fa3Vod6KzfKCH4a8vTpPWjNa/mbdEFO0HbO0vF3lUmmBR3sP3PUc9b2AC+/PceniAp4XUC41uH5tmSefOkxPLoFpatTrDv0DKfbt7/9IJivvBlmSUBWZhbUi66UaiixRtxzeuDQTZe4IZlcKvHVllrVilQu3llo9ir50gr50gm+9dIGZpQL7R3qRZYm1YpX5tSLLG2V+5+cfjYKuEG0F7Ovza7z6/m0Oj/czlItssso1i/VijaFmNr3jcQx9lhs3CQgYMvYT02zimsmqtUJCSRBT4oRhQFxrNzYNw5Cyu8GaPUtW20dVSChCZt0uE1cM9icMdFkl6EI7DcOAdXsBhKBPvyPD6QcB6+sVUmmTTCaGqilozaCrKFKzriwR+osIeYR7wc9ET1eTZabS0XImCA107RgCHUnc3fn13iEwlWFMZZggdAlDH0lobXeiiFL14U78vUhZbkIREsPxNElN50xuhFdWbuP4PrFtR/9eDC/3uJcsWzc4nv4UcSWLIjRAsNi4RkrNYUgJblbfJqeP0auP0adP4QSXWa9fQJFUal4RVdJwmlb31yqvUfdKDBoHWGpcZ9W6zYHkw1wqvUBCybYaeYpQ0SSF+3J9CASmonGyZ5C0ahBXdXRZIaNFwTCbSzQnh0IavstKo0qvESehaLhhQMGqk1QNYsodwnxM1kgqXYKuHwlK+2HAzXKexwcnuFJY42i2+9hmWo3taF/f8B1uVFYouXUy2u7nqazIPPLoAa5dXeaVl64yPtHLL3zpftKZGM88ewxFkfnWf3ib3/gbT3D6zMSemjZ3g+f73F7MM9qf6VqXNDSFqaEeXj5/i3/2Bz/A0FQ832d+rYTjevyzP/gBfhBSqNSxHI+vfu9NvvXyBQBcL2ClUMFyPF6/NNOijWUSBofH+/ncw0f45On90QdtEytaWi/zZz85i6lHpRAhIvpZpW7zqfsP0JfpPlruhQ7vl17k7Y3vczT9OL36GLP1GYbxWGoskFYz9OuDlN0SCMGIOdp677o9z49X/wRdivGZgQOMx3MIBG7ok1CMNm/CTV58GIbU/Qpv57/P9co7TMZP8OzA32ydY2PjOX7tbzyGospoqkKp1GD61iphGDIwmEHVdMKgBsE6kZrb3q/dn6mIOYDAQJUnP5YMdzt2MpLcDG6bo5/3QkOz/ahjfb24Ts1zOJkb6pq1bo4XCwRxVefxgUleXL7F4wNTLNZL2EH3mtReUG84kfA1USNASCKiKIUhWvMkqVuR5qdi5/jBwr/E9Ac5lf4cpZLN+dpPmC1eIykNElPSnA+epxIsIQuFPn2Ch3q+RK8+znTtLO+sf483+As02SSnj+GHLj9a/n83G2oK5wo/AgLOFn/I8dTTAOhSDFXSyOlJgjCkR4+1uscAE4ls27Le8j2ulVYZjKVYqpdYqBUZiqUQQN5uYMg1ppI54mqUicUVnbTWqVURZboeipD43Nghaq7DQq2MHwZdg2tOi+8YdBUht1yK4c65EunXBq1m2mbPYXSsh//D3/80Bw8OcN+ZSfYdGGhq2hp8+jPHefiRfaTTsa5Z/QdBteHwh99/i3/ya093DbpCCB4+Os73XrvMRrmOKkskTJ1P3LcfIeDS9AqaIvOLnzzFQ0fGePvKHC+/f5tCpY6hqewbzpFNxhjtSzM+kGX/SI7BnhSGrmJqSqt8IEuCdNxAbmaeJ/YN8stP38erF6YpVRvEdJX9I708eGSMJ05OkUm0rzyim22Fl9e+yTuF57CDOgeSDyALmRPpU8hCIS4nImU6ISNLCjktopv6oct8/RrPLX+VpcYt9ifOEOIRU6K+yPY1jhCiNUhhBTVeXP0z3iv+GCdoIEsqZW+DtNrbEtpJNfVQVpdLfONrrxOGEIvrXDg/x2//nacJgzyIONyjx+LPPuj+DILtXuF5AQtLBSbH784Z3sTlwip1z2XATLDSqPL26jxPDk12vG7DqnG5eBVFSCAEFdeiaDf48cI1blc28IIPbob5lz86T7nSoFq1OXpoCFmRuD2zjuv5JOM6xw8P89bZGWRZUKmqfOUL/5gfvnSRF5Q1Fpav8Vt//Ve4/db71FWF6Y0KX/rcfUyN9+L7AbquRMLlTsCB2BMY5WPkMnFSyd2nB7ficOpJ1OZE29Zgu4ntUo5u4FNxbQy7TtmxyOomRaeB5XsIopHesUSm9fq4opNSO/fHCSId57JrE1NU3l1bYN2qMVcpMpJIdwzB9OiJ6PfZBlXIfGboBH/34DP0G9EF7IchS+UysiRRdRxMVWUwkUAEYFkOiaSJrqv84i8/3OEArGoKud7kNv8vmsdHwg4sFhszBPgoQmHJmqXqlYkrOzdV8+U6a8Uqhr7zJXx0YoBDY3185amTPHxsfMfXAZzcP8zf+rlHdn1NN6QTJn//l56iNx2xMvqzSX73S4/xu196rOO1lucRNm1FN9HwKzy3/Ie8X3oBP2xPRHQ5Ws2kpDRBGLBsLZLVcihNy67Z+mV+sPSvWbFmAPBDDy+8+2QZwGztMter77RWcSVnjcXGDdJqZyzoH0zzW3/7kywuFHjr9Ztbfl+TMLQI/XmEsn9Pnwv/CYLufy4Iw0hS7/b0OhNjOfwmkX0nVkJIZBdecW1KtoXludiBz2i8+7x20Wlwu5Kn4TkMxFKMxtJ8eeIEK40KP1m83rVrvleUynV+85ce5drNFVbWyuTzNT7zyWMYusqPXrhIuWpx+sQYD9w3wb/505doWA4jgxl+4XP38W/+3cv4foCmKfzObzzJd547x/JqiWIpEoCOmRqqKlEoNRjqT1EqNyCkFXT9IGChWGa8SdrfhOv71ByHjGli7OLu0A0pzeDJwX0AHM5EBPQgDKi4NmmtM7hqkkJSNZGF1DZG7QQeZcficn6VN1fnUIRETNE4t7GMKiuMxNuDWEzROihHAEnV5PcOPcuwmW35Yq1UqyxWKqiSFJWG1EgD2qu7zM2sc+TEKEvzefqbDgVbPcjCMMSyXVw3IAgCMlsU5ZJqhrSSZb4+zZAxhiQUFhozaJJOn76zRqtlu1iOy3qxxvhAd+rVRrmOqasM5j4+WypNkRnp67wGupXfLudXGUum6TXvnB8Vr8CSdasj4G6HJCSGzXbro3V7gbKbb/09YO9Bt+SuUdny3ppXZMW6zeHkQ1357rWqzdUriyzM5zlyLKrhCimGkJKIeyyT/v9U0K3VbUrlBn25JEsrpYgKo8o4jsetmXX6ckl6czvP64dhyKCZ5P7ekdZSdyccSPXy6ZFOgeSpZI7JZNQgsn2360X/s0a+WOPQ/kESMZ1a3SaViBT+ZVlioC+FtmVJ7AUBV1bWOoKu7fksFitdrby3I+IOh7vq3kpC6hpwIbqQk4qBKatUvTuTWHbgocoSacXgE0NTCBFNN72yPEPJbnQEXYHomjErQmLASFG1bRYrFbwgQABDiSQJXaPuRMaMsiThCYHnBqwslSiXGvT0JZm9tcrBYyOoqkylalGuWiwsFzE0hZ5svC3oJpQUJzMP3fWYbUfM1ChWG7x+aYbh3nQbf9b1fK7Pr/Htly8SN7TWFNhuCMOQm9VV9if6P/Rq1AsCLudX0WWF/emeVvlusVYhZ7aXhXTJQP6AvQxTTqBJOg0/oj36oYe3R80ON7BxwzvnjhvarFnz1LwSSbWTQdQ/mObTnz1JGMD8bJMyKRKEQRmke5MO/ZkEXct32LBrpFWTmKKxVq9TaDTImTF6TPNDKYjdC3wvYG4+TxCEuK7PxkaFMIwUxDY2quiasmPQFUSiPU7osWZV0eTO5erdcL20xuXiCnFFZaVR4eu3zvK7Rx6764z5x42x4R5OHI7Epv0gQNvSYQ6a9jhb4Xg+P712C0NRcP2AkWyK3ngMLwiYK5SYzRc5OTxAyuxsdvlhwPnCLMuNEs8MHvvAN52UamDKWnvQ9aPl61Qyi64o1NxIaaxHN8kZ3fWKuwXdrZCFRG8iRk+su06BAxTzNZymSeTNK0tsrFcYm+ojCFWWVksossxgX4p6wyG5i5vEvSCbMOlJxvjGT8/hOB73HRwmDGF5o8y71+a5cGuZXDrO7335sR3Famzf43pliaRiMBLr4S8X3uUfHv78h943L4ikOLev5tK6gbbtmtEkE/keHL63wpSTqNKd47lZXsiX6yRMre083g4hpKb85Z193HAWyTvLXYMuwMz0BrMz6y0jTSFUJHmc0JtBqIf2vN8fS9CteTYvr17lsb6DvLR6hZdWr+KFPv1Gmi8NP4Dkq5Rtm9wOJ/JWbGYZQsBqvUavGe9aJ9wLYjGdwA/5+jffxDBUnn7qCOWlIt/4j29TKtUZ3cGNAKLywmYzpeg06DcT9xx0e404JdfCkFX6jASneobueRsfB5IJozWiur2zHpVcon2sWDayJKi7LnOFEpoik9Q1ypbFoYFeZElitlDklZszTPRk2oJu5Bri8lfLF/nfrv+ER3r38VT/YXRZxfY8ZElqTYpthR8EWG4UTGOq1hLfTqsxYooO9p3hDst3cQOfWHMVklB1DmX6OJjpxVS6B570LkE3oevEVK2Di7oVsbjOoWPDuI5HpieOkCJLeSOmkS/WKJUbTI7lyGXj+H7YOr5+6HO9cpYjqQfatmf7DW5U3+dI6gFWrFkSSoakkunIPuOmzm9+9gH+pz9/kT/8/ltofyUjgCCMmlsPH5vg1549w77h3I4UNct3uFCcZyyeYyzei7NLc/duE4ybI7vL9Sp5q07eqmMoKu+sLjCaSJO3G/iB3/Eba7KJvMvYO0R8ar857SiLO1OYMTmJJt35/fzAxQtsanWbYqWOIsuUaxa5dJyBXLLtNzQkE00ysIN667GCs0zBWWY8drTrdxwcSpNMmW0DO0KZIPQXdt3/7fhYgq4beMzX81wtL/Lmxk3++uRjfH3mNXJ6gtc2rvFb+z7BhdUVXD+461JmvlpCERI5M8Z8pcx0qcC+TA/9sbvL9m0iDMOohDC9ThAEfPaZE2i6jGFofOoTR1on1O4i2QF5u07NcxiJpdGkez90Gc3kKxMn+Wfv/oCyY/HO+jyH0wNtTaL/FNgMZHfD9dV1UqbB6dGhaLRWwHq1hiwksjGDuuviBz6P758gE9sScMOQdbvCN+fe4t/dfo2qZ2H5Hl4zE7q0uEo2btITj5HUNRzfxw8ih4pS3eKtmXlWKzUenBjhQH+ObMwkrZnElfYVguW7bYFDEgI9lHAsF2Jh17O9Gwti6/sleffjIskSfYPppr6r1Dye0XsScZ39E30tjVxV3UpbDLhQer0j6Fp+nXcLLzAa28/L69/lRPoRjiYf7PhcWRJ88vR+etNxfvT2NWZXCuiqwuRgD0+d3seR8QF0dfcpTl1WSakmm6zTrauOohP5sMXkSBPY8l027ApZPdFx3AFmykXGkxnKjsXz87exPDcaTZZl3lyexw4i0fD9mXadYhmlKYwk2K7hHIQR/a/gVJitL6MImT4jS6+WRpc1TDnVElUC8EKPilUnJwku31wll4pFLCI/IJs0MbawPHQ5ji7HWkFXQqbfGCet9TX3o/O4JVMG9z84xczttTsPigRC2XuWCx9T0I0pOjFFw5Q1DqeGsX0XQ1YZNNK8snYNPwio2DZpffellhcG6JLChlVvjncqLFTrlB17T0G37jhYrkfS0KnVbG7eWmXfVB9DgxnK9caWDO/u5Y0WXaipLtYtK7sbhBCoIpIbTGsGf//YU63asO/5Ozrc/qeEF/jYgUdc0UnoOquVGieG+ls1xL5E1LUWQpCSZU6NtFv2BGHIXH2DP739Kt9fPEetWQ7wmxdU3XE5MTLAaqXGi9duM5ROslqpYTUdeH0/YDyXwQ9C3p5ZIKapZGMmiWZNdyucwMMJvDZVu9JGhaXbq2T6UmQH0hHJfYtQ0W6Z7l6xE+/WNDTMHUSRJCG3Go5u4OCHHppkIISEKcdRhIopxXZcegsh0FSFM4dGOXNotOtr7gZdUtBlFbeZRW7VpX1h5RIZLcbjfYdRhcyGU+FbC+/wZN9hTmXGO87T26U8t8oFwjDk6dEpjmT7kCUJ2/eYq5RwA5+BWIKebWWeSPwmjoQg2BZ0806Zmteg7tsokoKMxLpdJKXEmkE30RZ0/dClatfQPZujk/0k4wapmIEfBG0BF8CQYm3vTau9PN77FSZix7vGA8fxmJ/b4PzZGcqlBmsrZXpyiSaT4d6u2Y8l6EoIjqSGmUr006MnWKwX+PLog8RkjbOFGWzfa/Jjd9/ZmuNSsBvEFJWEqjORynIst7s/URhGMnjDmRTr1ToVy0aTZdKZOJ//7AnKlo3l+qy7Fo4VktkjFUqVZIbMGGXHYufh0TuouDbPzV9l1apgyiqKJGHICpbvMWAmiCkacVXDtaOl8/yNFYan+jDuUYjn40Tds/nJyiW8wOcrYw9yaKCX7ff0u9XjV60Sf3L7Fb6/cI66f6fJ4QU+XuhzdWWNQwO9EEaTi2XLRpEEpqaSNg0c36di2UzkMvzcycPEmvXJhGJgyO0BLSTE9t2Wpm4QRNlntVijvF6huFom2ZNg4uidCaKksvvvHwIlp8F8rcCBZB/GDmWKD4LNgLpszbJizXI4eT+SkIgrKXTJxFQSyNxdc8Rt2hl1k1gs1SzihorrBR18XiEEmiS3uMr6ltWbFwbcrK7yYG4fqiST05MYssat6ipHUsMdtfijPf28s7pIzjQZT2Za54UsJLJ6ZOq4k8CNIcWjQLetBlz3LQJgMj6E2dRXrri1ltayJhnRexGEhPihRyKutBgwu0GTzbagaygJstogUpeAW6k0eOHHl1hcKNA/kKZ/IM3LL1xhbCLHqTMTaPcgfgUfU9BVJJkzPZNAxKkcNqNa6Xw9jxcE6LJCj2m2jfJ1g6Eo9JlxkpqOsUcpRtf3uLCwghcEzBdKpAw90vVdWadQb5Cv1ZnIZZjZKDKcSbG/b+/3qSiT8ml4Ue1wt3qsKsn0GXEMJZJ1lJBQJYm8XSelRTJ8nuuzupBnbaFArVRncPzuFjEAjuPzje+8w+p6mcmxiFf43PMXI2qS5TE20sOb797m2s0V5hYKAFy5sUy5anHh8gJPPXKQpZUSf/rNN7h8fYmRwc5a9nKjyL+bfo2/Wr7AQ7l9fGWsc4m7F6zbVd4vzLUFXGhmukEAAv70jbPEdY2HJkc50LdzDXIrkmpnpgtQ9x28IECWJRqVBkvTa2T60vQMRCUA126vW+7WxAzCkL9ausK3Z9+P/O3M9F3PWYCkYvCViVPc19Oega5ZC2w4KxxMnmqOokfndEJJExqjaJKOH3rElSSyUFqZ791wcWaFVFzHdjwODPe26R68eXWW0/uGmV0r8sDBzozYkLUWm0QRMkuNYhTAJRldVnECn9nSAgNGmsl4L6tWGTvwOoLucCLFcKKTV6xIEjmznT1hNRxuXl1meLSHbG8CQ44huoy/D+o5EKBJaivN8VwJU45svHRFIa6kkISCH7rNRtre2AuaZKDuUQpAVWTGJ/vYf3CQqX19SJLExffneO775xkZyzE4lL6nFerHyl5wfI9LpXliis58Pc/N6gpDsQyW77Fcq+IGAT2muaNgsS7L9MXukQMnyzwyNURAyEAyQU8ixkqpQto0EQLGc2ly8Rgj2TSeH9yDh0V0EWqSTN1z8YJg16BryAqH03cyKrcpclN1bZ5buILlezirDZZn1sn0JtlYKmDVHd574TJ9ozkOnZ7Ycdu/+uUHcV0/8tiKRZbojYZD0BQaicc0Du2L7vafeuIwvT0JfvOXHiHE44mHJxjs7+G//PUnAHjqkQPkeqJSTRiGOIHHWxu3+JPbr3K+MIsVuBBGjR+B6JoJbMX212mS0jXD8cIALww4OthPXyKOIksRQ2CPugSbXN3ty9KG7zSbLgpGXGf8yDCqrkaavCH4frurRrzLOPEmBDBsphmOpdFlhYSik1KNZulMxZAVNElBlWUUSUImKg/pksKA0RmAqn6JS+U3yWp99OkjqEIjCAPiSorFxm1icpK4nCQmp5qWQ1qLM9oyzvQiJ+atJQ3LcdGa9VvPD5Alge1FAi91OzpXa3YUjNzAb8o8Ru/PajEqroUX+IzHc/yPl/8SRcgMmxk+O3QKgD+dfoUNu0pOT/Bzw2cw5b3pSPt+wNULCyRTJiMTPXhedDP0vYDVpRKJpEG2N4Eub2a6d95bqDeoWAFDqSS+cFkqV9BlhbJtUbIsdEVhKJUkrmSQm0HXC90OylgYhk1j0OZvKjYzfBN1F33crY1Dw9Q4dbp9uOTEfeP88LvnaNR3F4/vho816Hqhz6pdJh3EeCi3L8p0Q5+G6yIh6DHNu17E9wpJCNIxgzAMSRsGQkDS0CM91NSd+qOhqvck2+YGPhu2gx8G9BmJPWU8d1xvXf5q8SoAj/dPUXIsGg2b9SvL6IbG8vQa5UINu+Fw9d0ZwpBdg25fLtnc9wAhZMIwIJMyAR+xmT1tsx8ZHszgOWcJgiKq9ijDg5m254MwZMOu8J2F9/jzmTdZtcqtMkpIyK3qVZJqmkFjd3GP29WraLLBeCwadojJWlftgs2arqEqjGa7D5jsBiEEWS2GKsltI9V1z2k16GRFJp7aUkMUoGxrgHbTcNj6GUczgxxM9RGEzQu22XTa/G03/7tdE76b216fPkJKzbFuL9Kvj6BKGtcqZ6l5ZRYaNwkJ2pa8MtEKCaISwtlbi7xzfYGnTkxxdLwf1/NRZBldU5hdLTCzWmR2IMt9U0N8583LrOQrHBrta5OjfHXtGgeTgwzHotVNWo3hNX+Lp/qPcCo7Tsmp44dRUpFQdP7p8S837YAkdEnZcXR6OwI/oJivsbZSagbgeUYne0lnY2iG0hJEN+VEx/EqNBrkax6ykFgslSk0GvTEYsRUlULdoj8RnVNxJd1aMfihixu0B0Gr7lCv2fhNx9+evkTzxtheXtiKMAypViyWFosoiows3xn3liSBYWrEEwZPfeooub57Hzz5WINuTNFbd0uAR3sP8NWbL6CNRL5KS9UKA/HEPbs47AVRcI3+rGz+ocs4KkS0JKeZQexkPaJJMlPJDOtWjaV6mZiiktHMPS0rJCGYTPRgNjM+N/BRdYWHP9MuO1hYK3P/00cZ2X+3unVIGKwRBOso6jECfxlJSuH780jyMGFYiYrbYQNJHgahE4YNQnzCsE43p9+G7/AHN1/gP8y905U6tNiYoT8cvmvQnWvcJqGkWkHXVFTSaidDwAv9D+SEvBUZNY4qKW1BN8p09z7tdzeOtCykj0SgBiAupxjQR5tBQjARP8w7heeJyykGjQlmalcJCTmWehiAjNaL2Wy2rZdrPH/+JpIQvHtjnrrtUKw2yCRM5taKbFTqPHBghNsreQrVBroaqWLV7ch1YTPo5p0q8/UN+owkDd9lqVEkpmhoUmSRU/ds3s7fRkIgC4l+I92VrbAXrK+WkWRBabVOuVSndyDF6nKJlcUCmZ4EZiy6HnQp3lFeSGg6GSNJ3XF5ZHJsx8Z1TN4adD3cMNIcCf2QcqnOxkqZwkYVVVNQVZl0TxxJAlUyds10Z26v87U/eZVkyiAe1zFMFaO5YhqbyHH/Q1M88YnDH+i4fCxBd6a2ztn8NIokRyIVInJn3bCrnMqO4/g+SU3DVLtbLW+F7XmslWsYqkpPorslyYdF3XbZqNbpS8VR7rJ0SmsGK/UKJccirRldDRy3Q5cVTvZEwweLtRKGrHZtQGX7UmT7UgRh2MosusPHc88R+EuAIPBuE0i9+P4MSmjjOm8AAWFQQlaOIaQUYbAOQgK6fz/Ld3h59VrXgCsQ6LKJIhSCMKDuVdFlA1Xq3JYuGWhbTmZD1royBPygWdO9R/hBgNMUKs80M92taHjOPW03puhdyEp3sGpVWKgVcfewTSEiNsREoqflgQebJRdYt5fYcJaZiB9GCMFY7CBjsZ31eifjR1vTWn4Q0p9OMNSTom67LBcq2K6H5XrEdA1TV6NBlVyaEBjMJhnIREIxpqYy3h9ltqOxHD9dvsi7helIWF5P8WjfgVaJb8BI89mhk+iSiiKkPfPIwzCk6hWb8ooDJJUcc9PrmDGN+x6aZGyyt23EPgiCVt/MkGMdtev+ZHzHTHQr4kqqjeHhBjb5jRK1ooNje3ieT09fEjOmR6P+rURMaarudcfho8P8vX/4WUxTJZHQUTWl5dTh74Hquhs+lqCrSTJpLYYuqc1BhsgosE9PMWxmCXyBGwQM6sZdqVcSgo1qncF08gOJMdYdt+UCXLUc0jEDy3VZKVVZLlY4NT6EFwRULJvMHhxBZSExmsggC7GngLsVt8obPLdwhft7R0lt0YP1moaKeaeCIqIGhh8G5PSdli4SsnIIWZ4iCMuAjJBSKNJpwqAAKMjKfoSUJDqCCqGUQggdITS6/eyykMjpCRYahe6fKCSEkPBCl9fzz7M/cYSpeCc/0ZDN9qArqV2nvjaXtPeCDavGXLVEEIbc3zdCulvQ9d29b1dE56q6RU1sO+qew9duv8uV0jIQrVpMOQr4ac0krRokVYOEGtV7X6/dxvI9fnHyNPuTUZPzdu0SF0qvU/erTMaOMKCP7Wn3th5Hy/WQZYnPPRhlV3NrRVzPJ25oDHTReD62za051eRNn8yMEoQBc/U8OT3BkdQwg8ad8k43C5y9oO6XeWHtz5ipXSSt9vJU3y8zsW8Y2/LoG0x3aJpIktSSjtal2D1fS5uIbQu6XujgeDZhCCMTOeLJ7te0EBKapCMhE+Bve06gajIjo50N5kbD4fx7M/T0JDh0dPgD7fPHEnQHjQwDRuaOtUcYsOFUickRf5cQ0rpOxbHpC2O7Gio2mgFyNHtvHcJNXF5YZalYRiDwg4DBTJLZ9WJL5ev68jqGqjLSk95TrUoSgrS291HOrXXjlCbzcH8/fUYKVYKCkyeppAiAoltjrr6BE/iMx3t3bVYIISErUc1XChqEyhRCRAGWsN4KuKLLnXzT66zze0n07hjkiYwvmx33Y6nTJLYoYEXd7wBZyM1JnzvBQhKCpGpiyCqWf0eMxG820rYj2BRKgRYvepPlIAuJIAipuFHdLqPFULedO5bvtG3X9XzmlwssrJYY7E0xNZK7M8QQtfwwml36bhg20/zc6DFOZofIajF69DhxRUOV5KiJJsmoktT8v8xb6zP8385+H11W+AfHngZgQB9DTm/KEg5hyt055rtpNPckY5yaGmpxkPvSCSC859KHIWs8nNvPAz37osThLtfUVknLnYwv7aDOC6t/xvniT3ECizV7lqOpRzkzdAQRSghJ7PrdDDm+p95Ot6m4mByxFzbhBg7JXh0zl7yrjObmCPKmt5/vBXieT6Vskd+okMnGyfXeuSa+9Y23GJ3IsbZaoVa1//MKupvSfWEYUnBqfGfhXa5XlunREtzfM8nTA8eIqRoxVb1roGs4Lqaq7GrzvBtOjg1wZLiPWyt59g/kKDcsBDCWy1CzHRKGRtLQUWQJZQ/Nsb0iYjsE+KFH3skTk2OEwiGuegQ0WLYaLFmLTMankIixapU5khrBlLUOkviunyM0QG8JyHgYSJK5441s+zLODwIqnsXV8hIr2+xvtkKRNo0WQ9zA5Xr1IvdnHickZMVaYLZ+i4d6nkKV9LagK4QgqRrEFb0t6EY13c6ge6uUxwl8DqR7cIOAuWqRny7cQhaCn588ylurcxzvibK4Hi3RMRlYb7EXIrx3eZ7VfIWZhTyNSZeZxTzPPnqnFhdlrhplt93kchOarPB4/34eJWjedHYPVMcyQwRhwHy92HosoaRJKFE2GZGzAkRT9DrSJwiQhELVWyWm9CDTmWlmEyaaHlL2ypiygaaqH7gJHenKtn8HPwjaPN82Ybser16aYf9wjon+9swvJNLCfXHtzzlX/AlOcOcYVr0iru9QWqyRG85SLdRAgJkwUPX2sqIpJ7pSxto+KwyZrq3hhT6T8T7UZqDVpRj6llFgN7BBClD3YKmlyc3rpHm55TeqeEEex/XJr1cYHMq0vX5+Pk9vf4pioY7vBR0ynXvFx8xeCHh57SoXinP8nQPPUHbr/Pnsmzw9cIyx9N461umYiR+GH1h/VlMUVDnkyEhfNKCgJuhLJZCEwA/8lungRw3Htyk4eapelbyzgS7peKGHG7hIQqLfGGTDWUeXdLJamn5j7x38slshoUTZQdWr4wYuQRiQUpOs2msklDhZbXflIz8MWLXKXCzO8/zKZZ5fudzBpd2KmBxntn6TDWeVmdoNEmqa86W3gZAb1cvE5AQXS++Sd9bZl2gvOyRVg4RitJkCRjXdzuwyaxg8N3udot3gZnmDsmNzJNvHpfwK728soUpyG92ps7zQnumODWZwPZ9aw+HIvgHeOD/d9nqBuCsFKiqB7S3AxRUNXVbadHpb1uphQM1bRxIKcSXiZNt+BSsok9FGWW5cZjz+IHIXh9sg9HFCizVrFTd0mYhNklSSH9m5u7hRIpeKU6nb2K6HIkskTQNFjlYXDbtTMtHya7yZ/x7nij/BburSbqLsruP4FhuLeaqFKkKW8BwPp+Fy6MF9qFtHcqXdV7sAVuByvjjDu/nb/NrE4xxNR5xjIQQJJd0akHBDG3+P8o6q0NsUzjRdQXZl/IZLtieB57XHnK/80kO89MIVGnWbkdEsluUQ+wDDTB9r0A3DkH2JflRJZs2uMFtb51ByZ43QbnA8j5rttJadHwRCCFR50479zuP3mtn6TXO77bSjbnACm4pXJqmkGIuN44c+dmBjSEbr/fdlztyzrq4f+iw0FpGFzKAxwIq1St2PXG1zQQ9Fp4R+F/5h1bN4be0GP1m5xJvrNym69R1fv4lhc5yiu0HeWWN/4igNv8Z8/TaKpHAgcZSKW2LFXiChpEgqmbb3JhSjowPu71DTlYVEEIYs1yukVJ0wDFlr1NiXylF1HU73DdNnRB19Q9ZajJBNNDwXPwgoVuosrJRQFInebJzB3hS5TJwnzuxre70kRMdk217gBT55u06vEafuuSzUi/TqCRbqRfwwZDzeftNzA4uF+jmcoMaAcaTp67XEUuMCSXWAjDaKInS6lX5CQmp+DSewiSsJql4F70M4j2zHeqnG+VvLxA2N28sb+EFIujlCm4oZOzgm21wuvcZ7hR/R8DsddkvuOl7okswlKa2VAZBVuStNM6Kitdf9a56NrGrUPZuLpXnqnk3erhGXDW5VV1tBF6ISw6Z2g9ccqd4LVElrs8iKxXWUqkQqbTIwmL4jTuQFlEp1Gg2HTz17jGQ6xsztNWoVC9PU7vnG97EGXU1WOJEZ42ByiLn6BqqQmUr03dM2dFVhOJNCVz66pf8HRcWt0PAtdFlnxVphyBgio3XPUGNKnEFjGE2KloGSkFCl9gwmYnbc2/eKsjxB3W/ghh4BIb16Dj/00SWdfqOXhLJzzfBKeZFvzb/LS6tXWWoU9/y5cSXJQz2faHvMDRzOF98iLidYs5f5ZN8XujIadg66dy5ALwi4WlzlzZV5jvcMcKZ3uHWj3AmSEKS3DUjYzUaaJAkURWqVpWRZQlMVhrYJbgvoaue+G7wg4MdL1/jJ0lV+efIM+xK9PL98ndlqgYprkdFMPjF4YNu7QvzQJm9Ps9K4zOH0p9GkOBISS/UL1L38jhlaGAYUnQKqpDJsDuMFHqq0M/PHCxzW7HkWG9fJO0toksnx9BP06t01GiRJcHCkF8txiRtDGJqCqak0HJcgCOlNx8gm7tD+wjBg2brF6xvfoeSuR9vY1pAquWv4wmH04CgDE304lhPpi0gSSsfSX2Bsq3NfKy9yPD0FwEI9T0DI8fQoTtJjwMi0vTYmp5qZbtRI80OPiv0erp9HlXtw/TyKlCTExw9qxLTDGMoYSrORtglFkTEMFV1X28oGt26u8OMfXkDVZLLZOMdPjTE3s45tu/T0JpHvIoq0HT8TPV1dVjiQHIDk3Weit6Jmu8xvlKIL6GekubsTGr7FteoMJaeEIin4oUfdb3BQHCCldjagZCFjyB+9vbxAMBYbQREKqqSQUOKoQtmxvreZWdR8mx8tXeBbc+9wpby4Y+PoXhCJtpgklBRjsakdbyAJVe8SdMO2DEoSggEzyeNDE4zEUyiS1CUrCnEDC7WZFQkhyGixKPg0X9sIovLCWqHGpRvLpJMmmaTJ7YUN+rIJlG1W71Gme2+XgeW7vL52mx8vXmUoluZMzyifGT7CcwuX+fr0u4wnehjdlulKQiGlDqFJMVYaV7hU/D6KpKFJMdLqCDl9iqKz0LWLL4REr96HJKToRt3lZhSGISEBs/XLnCv8lEXrBlW3gBXUkIXCmj3Hp/p/nR5tuCNY9yRj9CR3VluLjEO3fv86r67/R9bsWQBkoXIk+TBXKm+0ssySs4btNyI7dk2JJgJ3gbnNbcSUdf7d9MtIQtCrp3i4Z4qxeG/ze7YjpqSa5wCtTLfmXsZU91Oo/xRTPYAXlLDcadLmE1Tt8xjKGKrQ2soaQkAm2znI09ef4pPPHkPTFS6en+Pi+/OMT+TQ71FzYRP/2TtH+EFIzXLwgw9eXvgooMs6p9InWoFCNHU91Q8g8fhhkVDunBi7ZcqRy2rAldIif3L7FV5fv0HVszpO2ijb09AkmYpr7bmRJwuZw8lTUY2a4R0Df0qJVMG2IthWXpCEoNeMkzNirNrXUaQhvMDGCeoEoYchJ3EDC8svo8sJNClOQu0lq0XTTJu3EMuL6ttjgxn6sglW8xXeuTjL2SsLPPvIIdgWdAUCXbq3TDemaPzOwccZMFN8aewkiiQzlcjxWwceZTiW4X+5/Dw/WbzKVybua73HD1wafomMNkZGm4goeET/qlKkKZtSh7pqy0YiON3H4YMwwAksVqzbvJn/HtPV92n41bas0w89rpTfIKnm+GTfr97VTikMA0Lc5tFR2waNgtDn7cIPuF55Nxq6QOb+7Kc5lHyY69V3W0HXCS3yzjJD5r5W03A3mNv84A6mBlm1asiSxAM9+1o37e3+egBxJcNmWcZrajCIMECT+/HDGoqUxAvKBKGDJvdjudHNQhHanpqRmWycTDaO6/rM3Frj+tUlBgbT+H6A7wf3zCD50BEjJKTmLrFYe/3Dboq8fQ1/i4VGXFeJayo3Vixc/+6ZmRNUWG2co+Yuf6j9qHtrOEGl7TEJgbbHzNULHQr2dTTpw3tTVb1Fdqbu746G7/DiyhX+5fUfM1/PdzyvCImcnmQy3ssT/YcYNNL8v678kMUduLrdsJf6tqGopNTICnuzNr8TTzfExwts6l6BirsaiZ/7JTQphiLpyEKl4Rcjp2ciBoPUzHIA3DDiPKtK5Iq8uBZRxZLx9a77JoRoG2TYCyQhGI1n+L3Dd+yYhBCYisonBg7w/7z4E97Lz7cFXVnSkJC5VPwuSXWAycRjmPIdcXLX9ZEkFdv1URVaY6c7IQyjptFi4wbvF1/gUvnVrrXVTXihw+XSa0zEjnMo+WBHsAnDgDC0AIEfVnG820gihq4eRmxhUyw1bvJO/octYZnx+FEe6Pk8CTmDKjQc7jTU1u05gtBvyybDMMT1A7Qt5UIBxOT2a0UWEqd7JonLekfdfjvizUYa3JHJ3Msvqkhqa8y6/ViEbKxXKORrSE2tCEmKVOsqlQbxuM7oWA89vUnUD+Du/BGkaSFztReZq7344TfVBZoi05+Kt/1IO6HkTPP66v/9Y9mPe4HtF7hQ+GMuFP74P9k+FJ0a31s4xx/eeol1u/MGMhbPcX/PJA/n9vNQboqsFme2tsGgmb6noLsXCCKnB11SaTQZEtsz3da+CYWR2EmC0EeTTAICDDmJJkXLXy9waPgljOZF2iovbIHluy0Z6sOT/TQsB8f16dakEvCBBOl3ghf6IKIhja2QhcJo/Ayj8TNd31cs15GEoFRuoGoyiZhONrNzRlp0V7lUeoWzxZ+wZs91fY1A0KMN0/DL1P0KRXeFG9V3GYsdJq6017Ydbw7Lu4okTAQ6flBAVzNt5Y66V+Htwg+peNENPKFkOZV+mj59FDewSak5av4d2uGGs9jiwG4iDOHq6jonh9tLjabcmaDsxhvfipiSau2nH7oEoY8flqk5VwhCl4Y3DWGIH1apOVda71OEuqOK2+zMBuffmyGdidxtVE0hmTI4cmyEdDrG1cuL3Ly+wmNPHWpzktgL/rMvL/SlEqRjxo5KZP9/dKLg1Pjm7Ft8bfo18k6t7bmUavLs4HGeGTzGifRom3NCTNHappP2ig27RlbbXbwoo8XQJaUVdP0m6T4MQ6an15mc7G0LnpKQSaidTVdF0khKfW3b3S6WYgVuy+pbCEHM0Mhl4l0zR4H4SIPuqlXF9r22icO7wXV9qjWbjUIVWZIi2YwcXYOuH3rM1i/zTv6HXKu83bIQ3w5TTnI4+RBHUo/yTv4HXK++C8CNyjvcl366pWS2CS9YRRIxfL+IpoygqAeQpU1WQITp2gVuV9/HDz0kZPbFT3Ew+WCk8iU8erQhlqxbrdcXnGUC2m+sASFzheK2oCs6Mt17Qeu7hNHx8UOPpP4AhAG52GcICZCEjqFOIguTmBrZpcuie6YLcPT4COPjOWJxHcPUOs4dSRb8yVdf4vCxYYZHsv/5SDt+VNA+BkGc/29FzbP53sI5/mz69baAKwuJU9lxfmnsQR7K7adXT3ScKDFFY8C8t6DrBj4/XrrKz40e31UYJa3FovHSZgK4men6fsBzz73P3/k7T38gzmm2S6Zr+27UhGs+nE5GwkTdxOcjCcWP7vy6UlrGlFXuz3WO+oZhSL5SR5YkMol2ipQsSwz2pYnFNGzLJRbrzJ7qXoX3Sy/yXuFHrFqzHeOrm8hpwzyS+wWOpB4hJieZrV9qBd2Su8at2nkGzak2PVld2Y8kJQiCGpKIBmy2ouaVuFF9h3KTrZBUeziZ+SSJJj1QQqJHb5/QKrsbrUx3pVzl5dszPDA23HXQqVumC5H1lxNEQv+qULqOKGuSgSI0HCz80CMIfdLGI123txWyULpmukIITFPbNYPN9SapVW3KpQbDI7tz4rfjA51tCXWY/amfa3ssDKOTaq8TGkG4Oeq582v6jJOEYbDjEkAImcnkZ3CDWtfnPyoowmzVEHdCv3EK+GADHPeCpLqzypcX+Ly2doM/uf0KG05U3xMIhsw0Xxq9ny+OnEasOKTRCfyAwA+Q1WgiR0jRkMCQmUEW0p71C2QhcTQ1gBcELNfLfG/hIjFF44sjx0ltGZfOavE2lsAmT1eWJWwnujCDIGydQ3sNwBkt3pnpNssLm1BkacfzMmqk7XwZnM8v8Ic3XmehXowUx8SdT9vcx6jer5BUdRRJ5ncPP8GnBrv7Zt1azhOGIblUnEszK8RNjWPD/UiSYKg/HdUOk2Hb9w9Cn1VrlpfXv8mNyrtYO5zvEjKjscN8euC/YMjcjyppBKFPTrtzzoSEXCi9yJnss20qW7KUAQSK3BlAwjBkvn6Vm9VzBPhIyBxInGEqfurOMRAy/Xq75qwdNLD8OqacxA18UobOj67epD/emcHHtjXSou8dUPMsVq0CspBRpIjnOx4faBtJlpAw5SR1vxyVF7bdjDzfx/aiQShduePEIQvlrpNwHfsUBNiWx9zsBiGQ+ADuzh8o6Ob0Izza/0/bHivUG1Qth7GenTOlMAyxPR9DVVguVYhpGimzW3YU4gcWflDE9ddQ5UjqcPuFKKHyQO8/uMveRsvMILQRQr3ng7yJ7R3YMAxxvUgTIAxC9qW+yL7UF/awpQDPzyNLmZb27b3tx87UsOVGiT+dfqU1zisLiZOZMX7v4DPcn5tECiUWnCWmL84jSYL1xTxjh6LspH+8F0VT6NESZLRY2/QYRIL0UtNRwPJdblc2kIUABC+s3OCRYJKZah4BzFYLrFjltqDbo8XbWAKbQVcIgaE3bWtWSqyulDhwYKDryez7AcVijWw20QqiScVEkxXYwoe/U16IkE3HSCeMrtoBAnZV0jqeGeKXJk6zYlUYMtP0G4nmVFwkfahKMoasokpySy9i87mOzxKCyYEs799e4tLsCp84uY8fn73Bp07tb3vNVt6nH3rcrJ7lr5b/iDV7bkerKAmZ/YnTfH7o79CjDW7R+5XoNyaaQt/RQdqwF5muvc+pzNNbPnfn68IKatyuvU/JXQUgpiQ5k/1MazR883N69GFUoeM2m+GbE2JBGHJjLc/r03NkTIP+ZJegu2XAAaDmWTh+nYXGBg3fJqenWa7n0SSV8Xg/W8seQkjElCQbDs1Mtz1huLGW57mL15nIZfj8iUMtKdko6N795r5VN6JasXn7jZssLRb5uS+dYWAoind+6CKQ7jpZBx8w6G56xm9Fue6yXKoykcu1OH3bz3Hb87i9VuLYcD9z+SoDqQSZWJww9AlxkYTR/JIBflCk7l5BkXpaNRlFaq+dRPSR7l8hCD38oEaIhyIS2N4SipxCl3fXqt0rwhDy5Uib1nF9xrtY3nSD6y+zUfl9csnfQpXHds3obN9GlfY2Yx8Q8sbGTc4XNukwEg/37uf/ePizHE5FZpGVQhXf9SmsFDFiOkZMZ21uA8/zyY30oDZ5rz1aoiPoztYKJFSdQTNF0WnwxzffQCCYTOToNRKcyy+QUDRUSaFXj3coi2W0eNsQgh8GLWraJu+0VrVZWi4xNpYj0WW+w7JcvvkX7/Crv/IwmUxUi5ZF1KTbur+W71Io1ykVLVRFRlWiUdZu2LSm6QYvcLCDOmdyvUhiEEWozUDzwUdv+5ryjJGtu0Q2vrsm82LjBs8tfZV1Z771WFzJtPYNooC7L3Efnxn87baAu/n9kkqWPm2UZXsagACfi6VXOJ56AvkudLlocm6d27XzrcdGzcMMGJNtrxNCYEgmGW2gxd+VEKiSjusH5OImf+uR+xnNdCZlkZNDpNmxqd9QdKpYvs/+xDDplgh+d3W2aJQ7Kk/4od/RvEsZOkeG+jp+NUns7D8XBdoQISTK7gK6nMSQ06TSJs989sSW14AdVFhtXCSrT5JUh7pubys+VDFrIV/C8wNGc2kUWUJvTprMbBRIGjqqLHFxcYXhdIqYrvL6zTkMVeHocOQmGzUNPKrOedygSI/5THPLAkky8YIyqtxPGLo4QQFFSu26y2EYUHdvAxK60kfdvYkkdDwphhsUCXA/sqALIbbjUq3bOzq+doPt3qZi/QRTu490bHeJv5n6LCPmCHFlZ+L6JqquxXcX3iMkCkSP9x3k9w4+y8EtAylhGCKrMmeePdnSB21ULRzLRWvOwmfUGDk9wfV2wgMrjQrrVpWcHierxXhm8BA1z+GzI0d3NBzcipTabiTpN9XnAPRmpjs11ctWepzvBywtFZEkwfBwdFOrVBrML+RbQVc0p9K2wvY9XC+gYbsUKw1ml/Jcub2C73eWTAS06SS0bSeos2rdJq5kmzVFraNu+UEw3p+lUG2wVqxiuTuPrLqBzbni8xSaFEiBxKAxxcHkA1wovYTtREF3JHaIp/p+hV59pGsQ0SWT4djBVtCFiPq1bN1mJLa7fXhIwJo1x7p9J+gfTj3SlR+uSjoZtb8VdONKGkOKoUrRZOBuo/xCyMTkVCvo9uppZKG2DCl3Q+SgnGztb9j8x/UClkplPD9gf29PFJO23GAlIXddOUYZep3F+juMxR9j1bpIVpvEkNtvGFVvhZIzS0IdZNW6gCEn9xR0P9BaOwhDVopVri6scWslH/kyiTsqXQ3HZS5fpOG6LBUrLJYquF5AXzLOzdU8//71syyXoqva9dep2GcJmgfb8dZwg3XC0MHxo5NNV0epOZcJ7jpTLfDDOuAThj5CyARERXhTGUeTOo0fwzDEclwsx6XWcJheyu+YFXl+gOv5zCzlOXttkduLeRxvOyUmwA/rhDvUd6NZb4Es36lhNfwGbtA5AlrxqnueI3+vMMONSrT8O5Ia4W9MPcnBZLu7aSqXZOzwMNoWlSczYZDuTSI1mxtRptu5/BtLZLlcWuF/vvQ8f3zjDQLggd7xPXNcVUlpjexCdGIHQRDdCBSJ9fUqt6fXeOWV66yvV3Bdn9nZDX7840usrVWwbZd8vkquJ0G45feREB1ZtRf49Gbj3Hd4hJihoqlKhwV3+751mfAiRBIyslAJCbD9OlWvQFdthDCk7FhcK61yq7KOd5dpP0NTGMwmyaV394Rbatxirn4FP/RQhcax9GN8fuh36NVHW6yFtNrHA9nPMmIe3HFpq8kmI+ahtpHXul/mQunljqxwO7zQZb5xrXUemnKSYXN/12ClCJ2MdodZMhY7iiJpSEIwlknT26WWuwkJuVliiKDLKjHF2FNtP8p07yyNgtCP+hQCNFmm7roU6g0MVWlbfUvIHeWFirvEtdJ3cfwqQeCx2riA14UhUvPWCEOfIHRZaZwnpvS06TjshnvKdN+4PsuRkT6WChUuzK6gKTIpMyoJSJJger3AazdmqFgOXzlzlJrtcnFxlbenFzA1lYSucXCgl/vGBlmr1KIGhzCQhI7tzTFf+leEodOsW4UYyhiS0CEUSEK/66IumpyRCUIbRYoTVw8T+YipzTJEeyAMw5C51SJf/cs3OT41SKnW4Nz1Rf673/kcuXTnCVIo16k1IsqT5bicPBDNqW9eiFHN+iZr5f+VdOznSRpPwzYLbUkkEUJHkyM93IJT4NX11zmWPgrA2eI5DicPsS8+hSIUQkIuli5xo3qT4+ljjJmjaFK7yIYb+Pxw8TwNP6qlLVtFLpcW2J/oJ63F7sltI6Wa9OrJlmrTJkZiaR7IjWEHHnm7xqN9kySbAw/RMivyZwvDOkFQRpKyiG1uq71GElmSCJpByQujrGRstId/8b/8iFhMI52O8fWvv0nDcshm46RSJs89d4HXXrtBudLg2NER9u+/k70LIdpob9F2fTYz5smRHI7r8/aFWbygU44vkjnsvFjcwCLvzBGTU6S0fugyfjpdzfPC0nXezc8xU83jBB6apDBoJvnrUw/wqaHuWaQkJNJxg7evzbNRqnFzaZ3x/myb1oQXuNyovsuGvUBMTvJo7kucyX6auJLh+sqfYvlRltujDTJsHkAScktrt+PzkBkwJukzxlmxbjePkcut2jnm6lcYjx3bMbi5gc1c/XLr74PGFKbcqW62uRzfbM7pUozj6SdRRHSuJvS72SLJJNQMdFfY3BVbywsAQTPbjZ4UxDWN5VIVx/e3XYtSRw00CF00Kcls7TXG4o8yW30Fyy8iCx3br1DzViPticZ5hmJn6DOOcrv6/I461d2wp6AbhiE126HacPir8zco121kSSCE1nIh7U8l+PyJgyyVqpTqFpIkkTR0PnFoCkNVODLYF9XXZAlFkhjOpFojff2JX45GVoMCblAgDB1i6gGE0AhCG8ubibLcPXynhHqoeazlNkX5CJ0Xl6mpjPal6c8mOH1ohMmhHlI7OEg4ns/7NxdJxHRO7h+mJ9V+sfvBOg37PWQpTc16HRAkjCcRbBH1lmIIoYKQ8EKPZWsFIQRe4GLIJkeSh7lYvsRcfZ60mqbu1UmpSSbjE7xTeJe5+jyP5B4mJt/J7q6WF7lSXmoJyGzYVf7VtR9zubTIb0w9zr5EH6a8NzWkyBcrRULRqXhW2+N24BFTNBQhkdbufH4Q5PG8G2jaQwRBiSDII0Sqo6bfp6dQhITLnaAbhCGPPrqfM2cmWlNYQRBgWS4bG1VUVSEIAgYH0xQKdcIwxDS3NnAEGXV70A1afYW4qTE2lGV8OMv8chGBILeF/yqga01XlUyGzUPNUV2B6JKRVlyLb8+9z5pdIafFeaJ/P0fTA/zz958jq8d2DLpCRJ5nU0M9PHBwlFcvT1OoWjy4xSJ9w1lgrn4ZL3T4VN9v8HDPF1sBrOytt8Rx5upX+NHKH/JQzxcZMQ928G+jzxPk9GH2xU+xtoVqtm7Pc774PL36aMewxCYafpVVa7b19z59rMNbzA89Ss4a7xR+xDv551CExpnss4yah/bUqIJoqZ9Uevb02u0QQmrxfCPreh2BQJEEA8k4xYbFRq3eUcrZ/G03EYQ+69ZNVqzzGHIPS/V3cYIqWW2SDfs609XnafhFBDIhPoPhKZYb51i3rpHWxtCkvTmX7ynoBmHIwkaZ/kyCsd4MCUNjuCdaCkSGjpEmraYoxHUVXYkMHi8trpKv1Xls/zjpWHsgk7cW+9EjFkAYAwkkoSGJyLnX9hYpWi+R1O9H7ODv1XYg74ERIISgN5Pgrz19imuzq6zkK5w+OLKjYHo6YXB4op/eTIJssrMBUrFeIl/7GgPpf0Kp/j3q9lvEtNPN7xSdqJIwW/uoCIU+vRfLt1hoLFJ2K6iSysnUCQ4mD7DUWGK+schSY4mEmuATvU8xYPR3NNbOFmbJb2t8WYHLD5bOc744x5dH7+f+nkkmE71ktPiuwvFCCPqNNEnVaAu6AA/1Rtl5NDLqEYY1gqCE47xN4K8hiTh+UEAIBVnuXLb2Gcm2z/ZCP7K1VxSUbROHlYrF8y9cYWoqcrMdH+/lxo0VypUGo6N3Lk5J0JHpbqW7CSFa/QNNVXC3lYMEosN9InqcLjftdhxLD/Lfnf4Ci40S92VHGDRTvL0xiywk4jvIRQZhGE3Kjd7pLRwY7u14Xd5eYq1ZR9Wb2gyb55si1NZKxAtdrlfeYb5+jSPJRziSeoQ+Y5yU0oOyRfHNlBPsS5zmWuVtNpyF5nHyuFp5k35jgtOZZ9HlTlulgrPUVuJKq30tbzEvcCm568zXr3K2+GOmaxeQhcLh5MM82PP5rjSwbnADhw17gYYflRw3nEVWrGkGjanWjWYnRCYByxScqBSZUnJtNx4hBLqioHa5poVoJxsGoUtcHeB07LeJ7eEGsF6qMRrfx1TyUwDUbQfb9Vq9rZ2wpwglSxKHR/qaYsoeUliDoAL4TT2CO3e+3sSdaP/Q6DBW3SGT7C412LEzchqF9juuqU5iqpNdX1+sNJhdKXB0cgD1A0o/NmyHF969wXdfvUS5bvHkffv427/wKLEuzbGEqXNofOdGXFx/GEnoxLQHUOWRZnAVFOvfoif+qwBI6C3GhRCCXr2XXv3ORVf1quTtPEIIRmIjjMR2d98FWKgXdhQgX2wU+P0bP2U0luNkZpSDqUH2JfoYNnvoMxKYst5Rfug3kiQVEyh2/8DQwg82AB/XvYQQGrIy0fxzDEU7jhCdq4WcnmgLulsZDNvR25vky1++n2wmHmkYKxIPPDCJ63YGze2NND9o32o2FSOdNEjG9a6/6wc1O5UliWOZIcbiWQw5CopXiisoksxEorN/AFCsNnDcyGzV8wJ603EOjnQG3bpfoe5FOrQ1r8TWBuOweYDL5dfatBYafoX3in/F9erbjJqHGTSn6NVGyWqDZLR+YnKKUfMg+xKnKBVW8ZqZctUr8ubGdwnDgMOpR0gomVadNyBgxZppKzMpkkbZXaforrLYuMFc/Qpz9Ss0/Aqq0DmSepTHe79CThuJ3heGkQ51c8kvELiBjR3UqXolSs4qq/YsM/WLzNevAXC1/CYlZ43JxEmGjCnSaj9xJd0yqnQDh7pfouSssWbPM1e/wmz9IgD9xgQZtf0ajesaCV3r+J2jabSt5QYVgb9jP2YTrh/VjG8tbXBkvL8Ve9aKNWKGRl/6wwTd0IGgCFIGQgcRVpBRCYNFQvTm3dYF5RACtcMHqVpqMHN9hSOnx0mkOu+iHxbza0X+5V+8wmcfPswXHz+KobU3S6p1G8fzSceNHZWA1opVFjfK/ObnHqBcs/nGT8/y0NFxHjsxec/7oykjaMoIYeiCACF0LPcSheq/J21+EVlK3JUrrEl6h+tDpPoUtLyggtAjb10lpvYTU/rIajG0LSO22+GFAdO1NaZra5jLGgNGij4jFZkrqlHjLK2aJFQDTVIoOPWursB39sfCdd5DSElkeRRVPQTI+P4akpRAkrpnODltW9ANgh1F3GVZor+vfTvduLtCdGmkhQFbg9Tt+Q1c12dhpdiVaSJvuxjDMGTh9ho3z87T0x9lTasLeZ75xYc6fLcs3+W78xe5WlrheGaI7y9cwpAV7u/tzkypNhzWyzUMTcENfV6+eJtff7pTj2HT1geg6hXaAt+h5ENcKb/Bzep7HbzdqlfkSuUNrlffIS6nSao5kkqWhJolqWRxAxtF0vC2aENsOIu8tP5NblTfJa5kcAOntQ8RN/jOb3S++FOult+k4m1QdFZbnFyBIKsNEFfSXKu8xfvFF7CDBl7oEIR+xCcIQxRJxfYbuKFFw69SdQtUvWLbQIMXOsw3rrLQuE5K7SGuZDHleJT1IvBCF8uvUvWKVLxCq9SiCo0R8yAptfOG1xOPdei3ROXN9tHztDqOLDSKTg07cEmrsRbrpm673F7aIJMw8fwAQ1MJgpD3biyQTZiRi/ce1BB3DbphaIM/gwjyhP4yBAugHI4CcWgRygNsEiB8L2BjtUwQBOT6o4ulVKixPLdB31B6z0E30ssMWyeT2Pyno+AdsrBaIqarfP3H75GM6Tz70KG2u9lrF6e5MbfOrzx7mt4tjbGt5ofVukMYhjxyfIIQuD63yqvv3+bR45O7Tsttwg/K1Ow30ZQxDPVwq5lWrH+7GWjTBGENx5vG1E6AUGCX5b0mqWiS2jR79BEInKCCE9SIKwOtZWXVW45oNkofnx86xbv5ad7JT991kqzhO0zX1pmuReOcspDuGCyKiPQfAmW3+0x/dOBUhBRDiBSKcoDNlY4sD+16zLZnut4ume5e0S3T3RzagGg1pGsKn3r4EOWaxQtvXeeXPnO67f0dPGgB2b4Uh09PcPPiPNVSnfdfu8Env/xAR9A1ZJUHe8epejYvrtzgZmWteRPsvvIayCboSZqoisylmRUW1ktUGzYxoz0Tk5qyjyF+JDCzZaw5qfbwzMBv4oUOM7VLXTMzP/QoexuUvY3mV5JQm0pn2611AGpekRvV9zoaqNux0Lje9fGQkKK7yrniT1v6B0HYmTXebfvt2wwouestofS7IaePMB4/1lZWcT2fd2YXuLK0xkQu07EvHZ8ZGiw0ityoLFH2GhxMDnEyM4HnB0wvR0I/F24vU6w10BSZvkycly/c5vBoHwdH+vYUM3bPdEUc1PsAAcokhD4IlabtIiBDWEcg43k+hqly5dws9arF7PUV1ldKZHoSLQ7oTgjDECd0qXk1ym6FhcYSBadAEAb06FlGzGEyajoS7G6Sua/NrlKuW/zXv/Y033rxfV44e5MT+4YY6r2THfl+yNtX5njyvn3kUtGMfhiGXJ1ZRZYl9g3n0DWZa3Nr/PdffQ5DV5lZKqCpO3eCt6PunGW5+D/Qn/5H6MohwKfhXMD1FpClHhQ5hyoP43rzmNqJ5mTb3Zl6blBlsf46KXUCQ8my2jiLhERM6SehDqNIOptsjNF4jn96/Ev8i2vP8cb6TWqevfvGt8APAxq+Q2PLqj2rxek3kiw3Sl3NI4VIoOufaH6PnQnm25FSTWKKDk3Vs82a7oeBAJJqJLazmTXn9GQre63WbSQhyKRi1Bo2dcvp2MD2oCsQxBI6y7eii/2JL54mmYl1XS0pksShVD+TiR5s3+PllZv8D+ef4wcLlzmY6ixF6aqCrip88+XzKLJE3XZ54f1bHB7ta6vtykJBESpO6FNy1toClUAwZOzjKyP/gHcLP+JS6TXK3kaT49r9eIYELQ6sQDBsHmDVmm1JNN553Qf/PaLt704/2H37Al0ykISC7dc6xHJ2gyI09iVOM2q2Ny8VWWJ/Xy7qHXSIv3eetwGRQeejvYfbJgsvTi8ThiGzq0U+ed/+Vh+rXLc4MNzL/uFehnIp5D3IIOwadKPRQK25e2p39kCzqF7Kl5m/tUo2l6R/OIuiyBw6NcbwRGe9aivcwGW2Ps+l8hXeLZ5nujaLE7SfCLqkcSCxjweypzmUPMCg0c9r70/zwJExRvrTPP3AAb736iVKtUZb0D02NUA6YTC9tMHRyX4UWSZfrvOD1y/z6PFJDoz0MtqX4UtPnuDlczfJJEwe/+wU33/tUuSO2sxWouzbQqB3GZcM8YMiNesVZJEkCC1K9e/ieHMU63+BrhxAkfujDLd1TO/+w1h+CVno1P01IEQSCqoUp+6vUXEXMJQsSrNbKgnBeDzH//XkV/iL2Xf4q+UL3Kqutrnv7hWykPjUwFESisE3597C82xqvk3JqZNSzSYtT8AemppbEYYhFddqE8XJ21WKTp2E0n1Edy8IwpCS28CQFOq+gyQEB5IDLRrY6EAm+qxSnbV8lc88dqTt/aL5nTv3F/qGs/QOphHA45+/b0dL78h9QsWQVU71jOCFPnO13eUxj4z1U2nYPHXCIG7oFCrt2aciaSiShuNbkbNu4LRlcEJIZLQBnu7/DU5lnuZa5S1uVy+w4SxQdQs44c7Bz5ST/Pzw3+N88Xkull6h4hX4oJrNHxV0KcagMcXh5MMk1CwXSy8zU7u4o87EVshCYTJ+ktOZZ9qOEUQlhL5knL4uo8fNV7T+tOlenlJjxBW9LZmo2w6LGxUePDRKegu7ydRVvvBw+zl1N3xk8kqGqaIoMrGkgRnTGJnqu6v4Tckt8+bGO7yw/goztbnWnU1Cok/vZdDoJ62maPgNLpQvc6Vynan4BF8e/iK26/HK+duUahbTixv0ZhL0ZdobdqN9GR47McnZa4ukEyZxU+PdK/Ocv7lITyqGoSk4no+uyjxyfJL1UpXXL0xz+uBIW2MuxKHS+CkJ45PIov3H0+RxTO0+8rWvUWp8DyF0VKkfTZ2gUPtz/CCPKg8zkPpHzXeIZl1392MT4pMzjqFLSQrOTQbMM8SUO8TzzTrvJqLaZozfnHqMh3L7eH7lEq+t3+BaealrtroThs0MT/UfwQ18/nLhPWrYXCwu8K9vPM9DuX0cS4+Q66JQthOcwGO5UeJaeYn/T3v/HSRZep53or/jz0nvyvuu9mbaTY/DeGAwIDxA0YBeFCmRe2MlrXS1EXsV/EMR2tDelbQKrbSkKInkSgJJEKRAgPDADDAYg/E97b2p7vI2febx3/3jZGV1dZmumgFAae88ETPdnZnf8ef93u81z3NyaWyVLtubizfR5Rd5omsvB9N95Izklo1vM3CZaCxxpjjOi7OXcFox6G4zw2A8v2Y7uXSMB+4bXmdL0vqcDBLkOlMIIfjWn77Kiaf2k+9O3/O8ncAnFIKi02CqUSarxzAVdc24A0MrAq2l2trlvirpUWlWEIUKqv4i1jrad7IkUzD6KRj9HM58kMnmFWbtMYruLBVvgUZQoeFXsMMGQRipKvRaO8lqXTze8bN0mcNcrr7JTPMmdb9ESLhlNd13A1XSCESAKmkYSoyEmiGrdzMQ28OuxP2trjqZ/tgeLldeZ6x+jqI7Q80v4YYRi5gsKeiyiaUkSGl5esxRDmYeo9McvPcB3IGoBVwlrRXoNAYwFIucvn4CdFdfBz25FP2F7VOfrrkGW/1h1D3EhgxQqWyc+x4axQ4cQkmgSptLm081Z3hu9vv8cPENqn6UhVUllaFYP0ezhxmJD9FpdJDSklysXOZa7SbNwGa6OcOSW+T43h380dde5+XTN+jMJfjppw6TvUvnSZYlhrqyvHTqBn/w1dcwdY103GT3QCcnL0/w4qkb2I6HLEvomkI2GePI7j4+eP/qJYoQDpXmt1vNDquhqb0kzEcIRZWE+QSa0o2ujqApHTj+GH4Q8ZSqyspLlk/8Koq8eUmKqeRQJBVFNiiY+9d8fzf/hRCCiu0Q1zV2JroYSXTwaOce3li8wctzl7lanbmn56tJCifyOziSHWTWLrdp9GbtMl+49RovzV1md6qbvaleRpOd9FhZ8kacmGKgKyqBCLEDj6rXZNauMF5f5Ep1hrHaAjdqc8zblVUx3AWnylcm3ubk0k32pHrYnepmR6KTvliODiNJTDXQZQUvDLEDl5LXYKpRYryxwLXqHDdr81ytzqyKPx/NDW1rUoD1Pd1lTN6YY2G6xLVz42Q7U+uKEBadBu8sTXCjusD50jTNwONSeZb/cOUV8kacA5kePtA1umGc926aRwBNNtvZ+lAElL0FOs2hTc8jrqbYnbyfXYnjTNYXcMIKzaBKxS2z6JRJaAqqDAo5FMlAV3TuyzzJUPwgC84EDb9MKEJu1E9zvvzKurHirNbNaPIIWb0bGZlmUKPkzjLZvMaSO7UmDNJv7WEgvhdLSaBKOoYcQxC26mljxNUUaa2DlJZb9Uzn9G4ezH+cPakHKbvz1IMyXmgTiABZktElE1OJk1Cz5IyeVTSVEZabFVZIdNZbYT6U/ySB8MnqXSTV9Q0uQCEdX5UXei/YktFdWKzyla++w+RkEdPUUFQZw9AwdBVdU/jIhw/R2ZmiFjQouWViikVWT7da8dZmhieaU3x58mu8UzqLE0bxx26zkw92PsHB1D66zM5215VAsOQW2yGHuBojo6U5tLOXv//zT1CpO2STFn0d6VWetRCCIAi5dHuOx4+OMtydRZZlskmLdMJisVynXGsSiijuY+gqqbhJRya+ps5OEOKHC4TCReYOVVQEEjq6OkLSfJps/GdRlZVwiq4OrRJXXP570nzintfc2ITUeT0Z64bnMVOtRjWJskJvJsV9mQF2J7t5umsft+oLnCre5nTxNlcrM9jrtB13W2me6TlISrPQZZWUZrU900CE3G4scruxyKsL18i0lmCWoqPKCqokEyLwwwA39Gn4LhW/Sdlt3NPTHm8sMdFY4ofzV0nrMZKqiaXqaJLSlmT3RYATetQ8h4rXpOrba5KGccXgRH4HGS3WvkYTpQqqLNOTXrmed1bZrKTc1seVU7fZfd8gV87c5vjj+zasgjlbnOTPx96h5jkczfXzdM9u7sv2UXKb/PnYSSYaRX5xx4l2/FgIQanaJG4ZuJ5P3LozdCChyyZ6i9M2JKDszW96De9EKARzNQ+BxWzdJ6FlUH0XUzEpGHGm61VC0VJJllRyejc5vbt9XIYSY6x+lppfWrPtpJbjwdzHyRk9rUoCFztoUPEWuVE7zcnidyi12MgEgmZQo8sYZn/6kTaV4lqbEFKpfx7LeBhd29X+XJaU9rEt3zMhGjTsF5Ekh7j12LrnL0RAuf7HpGI/A4TUmt9EUwcx9ROryYCQ2JU8vuXrend11rvFloxuOhXjwROj1PbbdBSSpJImiqogSdGBJxIGS26JM6WLNAOb/lgPiqRgKSbGHUXikcGd5AvjX+Js+QKBCLAUixPZo3y058N0mZ1o0uqlWChCJpqTNIMoRpVUkxSMHIamsntgdaJCCEG14RCGAsfzefXsGOdvzPA3P/4AewY7kSSJasPmB+/cYKQnx7E9/chbUBmWkBDCoea8QMr8CJKkAwGBaGB7l6g0v0XSfApFzkbGvmUMZClStF1wq8QUnZrvkNcTERXhOgjDEM8N0I21y9E7z7HRcLlRKlJxHGqOix8G5GNxFEWiZlcZymUiQyJJWKrOaLKL4UQHDxV2Ygc+dd9hsrnETLPMvF1hwalRcuvsTnVzLDcSMUYpGk927eN6da7VVruCuu9sK1m3FQigEbg0mi7T73Ibh7ODHEz3o0gytu9TtR1s30dTZJqehxAQ0zWqTjSBJ43lZ3Pjl2jX4UE0Q+XkS5cIgpD1UsJp3eLXdj7ER/r2o8sqaT2SnNdlJVLIQPBvLv6AlGbx8YGDtMpXWSrV8f2AYrVJMm5SrdkM9+Uirggl1haQDETQLv7fCmRJ4kChE0HEPWAoCgUrhh+GuGFAfzK1obq2JEnk9B52xA9jKUlUWeOHC19pe71z9i184Ua8BZKELlnoskVSzdFlDrEreZw3Fr/Gxcqr2GGDRXeSF+e/iEBwMP3YBoxugob9IrKcWWV0298Kr9Xt+SrVxtdQ5DTpxC9vcgUk6vbzJGOfRIRNGs7LmOIIpn4U1r2DW4MX3Mb3b2EZj70nw7slo+u0luCGoeE4PmVho+tKO9SQTJqYisHh7H5yeiYSdnNL1Pz6KqNb8sp8efIbnCqdBSClJnmm+yk+3vMsurx+YmbRWWLWnm/He/NGji5z/QYFzw/48otnuTQ2y8R8Gc8P+MVnj7OjN9++SDenlvj2axe5f98AO/ryWMZWjK6Oqe1nrvyvcf3bGOoOgrBM3Xkd27tMOvZJEuZjSJKCG/rcbBmqvJGk6jWZbpbostJUvCYS0G1lVh+357M0X2Xs2hyzUyU+8pljG1Z8NBsu3/nLt3nm08fwFUibxpYeAEWSiakGMdUgZ8QZiG8e3pAkiU/2H+XF2UtcrEzdc/t/3UhrFo937mE4USAUgpMT09xcWOLYQC+WpvL9qzdouB6fOLiXCzNzaIrMfb3dqPLdFOirEU+azIwv0aw7TFyfpWsgTzJzVxhLkkjr1qrW6GWoksQT3bv40q1TfOHG2xzK9hJ3DTw/wAsCTl+ZwtAUfD+k3nTadZ6GHG+TuIQiYMmd2XJFzXIXFsChjq57/HotOs1BPtn3d1EkhWn7Bq8vfr1d5WCHdS5WXos4eu8wH5IkoUkGPdYOnu35W+SNPl5d+Ar1oMyiO8Vri39FVu9mMLZv3XNQlALKBkKujneeUu0PkSWTbPI3MfUHNiWXkSS5tS0JSbZQlU4UuRC1378HNO2X8fwxLONh3ovx3pLRrVZtzpwbp1qzSSUtkgmTRCKKoZw5O86v/vKjxOOR4YVoWeGGLl3mylJbCMHri2/xZvFtAOJKjPsyB+k0Onhl4TUEYMoGGT1Dr9VNuqVlP96cZMGJ6uN0WafX7MZqtSvadQdJljBaSzNdU3ni6CgSEn2dGQ7v7OXYnn70O8IFe4e6+O3PfoCOTKJFVnNvSJJBLv45vGCWYv2LhGEDWU5garspJH+TlPVMS08qWtpV/CbNwEUQMV7pioomq3QYKaw7JiHP9ZkYW+D8qducfuMGs5NFjj68k3WiB22Ul+q89fIVPvTJo+vGAn+U6DBS/Oaup/g3l7/DWG3+R5bfjik6vbEsRafeVrh4L5CQ2Jfu5YHCaKt0TJAydHZ3FvBbTGalZrRScnwfVZbJxlau3WaG7NrZCYrzFTRNZfr2IjO3Fnj042ubGTaDIkk81jXK7158iVOLEzyR2UkoJEb7O+jrzJCMm5RrTXw/bKvL3unpCkJqfhFPOOiSiRCCqWaJHiu9Ja7lrSIUgpJbp+rbdJtpVFkipeXJ6J2rqB0vVV7jcOYp8hvQXJpKnBO5j9IIqry++FUC4TNn32bWHqM/tnuVsY4goShd+MHiutvz/AmEsLHMZzC0Q2yl5FKRcoCMJGnI0tY6YpcRhCVs5y38cB5dHcHQDiPLFsnYZ/GD8W1RDayHLY3u6cnw8z8TaQ45js+p07e4eWuBkeECM7NlfD9YdfNlSabXWj3DTjan+f78S+2ltyRJTDamuFC5RMWrIABLMclqGYbiAxzLHGF/ag9j9dssuVH5TVyJsSs5ShiG1CtN5iaKpLIxdFNHUWV0U2OwK8sv/9T9G56LrinsH+ne8Pv1IEkyhrabnszvYHuXCEWDUEDcOIKu9iMh44eNVoJAY0ci4kfQZRVLWSEhv5Pyb266xAvfPMOZN2/iOj6j+3r44MePsO/IIIa58SyaLSQwLX2LNCLvDaqs8GBhFCGe4Yu3Xued4q1NO9XuhWXj+Ez3QfZn+rhZm+dLt9/kWnX2PTVJ9FoZPtV/PwPxKBEiyxKHelfuse35PDjUj6VppEyT+wfvkK9pcSFshJ2H+qlVmvTt6KSjN8ONC5PbPr5lsvey12SqWSY/vGIElo1s+q4JVJV0TDmOjEzYqrGtekvkjV5CEfKNqXf4lZHH0GSJmm+jycqqCX27qHhNXp2/ytnSOHvTvXR0RU6ELpvsTBxbZXQX3WlOFr/DE52fQ5fXZw8zFIsTuY8wVj/HVPMqIQGecFZp1606X6WXIFw/bm1oBwjCOZrOq/jBBOn4ryC1qoiECKg1v4EfTJGwPoqmRp2AipJtEdroSNLGfNRC+NjuW9juWSzjBKZ+hEr9i9jOGyhKF653GVCwjBMIHKrNr5FRfgNFTrX2/XX8YJaE9TE0dWtcy9s22UvFGt9/8RK7Rrt46ZWrxGLGPXkP/NDnB/MvM2vPtT+r+XVq/uoavOXPJppTXKleZ2diB7P2HF6rhCWrZ9iVGGVmbIHzb1ynUXcY3NWN5/qRxtRQga7Bwob1lOuhWG8yW66yu7uDiaUynekE5l2JtFD41L1x4toAcbkQyToHsyhyB3X3FiAhCJCQCfHJG2vjUrCazeqLf/AixaU6Dz6xlz2H+ukdyBFP3ps/1DA1rJi+VRa59wxL0Xm0czeD8QKvzl/lm1OnuVyZ3paRNGWNPakenujay4n8KKPJTkxF41BmgP3pXp6bOc+3p862ZYa2CkNWuT+/g78x+AAP5HdsWIVgaiqjhY0z05sh15Um1xWVCXmuTzx9b0L59SBa6sd+uLXyvSiuHo9qdUMbL7Sp+IvkjV4kSWbRqREimLPLfGPqFMdyIxzODt6zFHE9nC7e4rvTZ0lpMR4q7GRnsqut8qFKOqOJI5wqPt+umQ2Ex9nSi2T1Ho5knlpTG7uMlFbgYPpRpppXMeQYSTW3AYmQhKb00vCvrPo0CMsEYQldG0FT+/H8W1Qaf0a1+XXSLS4T230b2z2FEE0q9S+STf4PyLKFLKeJyjNlZElfNxzhBwtU6n9K03kVVe0nDIto6ki7siiT/M2WoZ8Fotiy613G9ydQ9P0tY/02QvhUG/+VTPLvIEv3nvi2bXR9PyDwQ1Ipi2KxTl9fdsOMbnSggqu1G5wpX8C7g61IRqLDKPB05xPsS+0GJGbtWU6Xz3O2fJ45Z55Fd4nlcg9VUjmaOURMsahpTfY/MEoiE0PXVYQQ1MrNlQqRbeDWQpHvn79O4kGDf/vtH/JbH3qIkc7VskChcKh5NxEEeEEZWdYJQ4859yaBaKBIJoFwMZQ8hrI1erpm0+VTn3uIvff1t1UctgJJkpBVZdWJel5Asx55EVbMQNNXusTurHTYaB9BGHXhrPe9EAJFktmR6KAvluVDPQe4XV/kVPEWY7UFpu0SZa9J03cJhUCTZeKqQU5P0BfLsifVw950L91mmqRmYcgr56oJmd1mF4PDeT7df5xr1VkuVqa4Xp1j0alS8WzswEOWQJdVEqpJ3oi2uyPRyb50Lz1W5p7MafdCWo/xaEc0sQzG8+yIdXDt9ATXz4yTzidI5xJcfucWn/7NJzEtHcfxUFpil5IkUS43WJiv0dubwYrpqwqUAhFS9Ry+MXGegplgJLl14x9TUmiSiYuNG9qU3XmIRzHkmBpxn+SNBE93HSCpWctygPeEEIJaKxEaV3UulCcxFI2nuw8wHC+scg4kSaLHGuVA+lFOFr/bTqhV/CVemPtTSt4sRzMfJKnlW+xnMgJBIDwafpVFJ1oZdJpDFIz+DZ9BVenDD5Yi6gEiRZlK/fOY+oMIpRcIUZVeEtYnqTa+3B7XdF4jYX0YWc7QsL+H61/C0A5GIYXlfbX5tFcQhGXqza9Ta36dTOLXUZU+Ko0vYjtvE7c+hB9MsVT5P1CUThLmhxDCBSQM7SBBGIU7G85rJKyPI8sJGvYP8PxrGNra8s4153rvW7QaiYRFLhfn0pVpPv2Jo3z7uXM4rr+hXLEnPN4pnWbWnl31+VB8kF8a/Fl2JUbby++R+CAP5O7nVOkMfzL+F6s846ye4QOFhyJu0O4MkgTKHR629S5UOSFa+iktOZGNSpsUKUZ37Al80USXU+hKpnUTRfs5X/7TF6sL3YUQ1Ko2c1MlTEujqy/q1rv/A7v42hdfJ9+ZpHdwe16Yqq4kf1zX5/UfXOZrX3iNIBR85DPH+cCH9mPFjCih2Wxyo7iELiso8opooixJJA2DjnicuXqNpB4tEzVFwWwlYeq+jSLJ1DybtB7Da5F0H8+PcDw3wiqWjDsd39bBeUERU80jhNO6zqvDJkszJW5fmqJzIM/Q7h4G4wWe7Nq/slWxvNmQuj+LLicwlGSbi2P5KryXTLIkSRxI9/Mvj/9iNGe3tivygtH9fSzNlhEhnH/rBgDlcoP5+SoI6OhKRl2Oi1EVgqopCGC8XmSiUURGYqZZ5bnpy1woTfNs3z6e6Nq55WOz1ES7QcIN7VVlY8sSQ5qscrM2j66oPJAfJQhDxhuLJFSDDiONKss4gc+iU6W7FQNecut8/ubL2IHHzw49xMOFXby2cI2/uP06H+s7yoF0X/udlJCIK2mOZJ9mzrnNRONyy/AKan6RV+b/kovlV9mVPE5/bA+WksQNmsw4Y1wsv8qCMxGpFFt7KBhRWCeqRiijKnlo30OLIJynXP8TZDlNEExTt1+IvEv/CkI0I6FafwJLP3bH/TOp1L+Iru3CC6YQ9ku43g08/3abryJ6V1d7up5/E03diWWcoG4/j67uJgwrlBt/gutfJgxLOO55NLVMU87jeJcIwhJC2GjqSGvfGrXm11CVAYJwCT+Y+/EY3XTK4qc/cz++HxKzdL7ytVOr5FPuxqw9x836rVVeLsB96QOMxIdQ7pxVkZAlOJTez6Hy/lVGdzQxQlqLlnnrhQ/ufPEcz2eqWGGwkEGRZapNh1hLoma+UuPqzCINx2Vvb2db/SIdM0lbJoqy1uOL/q2gSQk0eXVQXrrrT+2OoL3vB9y8MsPLz53n6oUpevpzfPLnH2RoZxcnHt3NuZO3+MJ/zImV0AAAV1NJREFU/AE/9xtP0DeYX9d4hKGgWm5gmBrmcsJQV9s7HLs6y9XzExz/wC7S2YgGsVZpYsUiI3qjuMQfnjxJ1jJJGQaWpmGpkXrtjmyWjGlSdV0mqhWavkfaMNlfiDhszxRvM5zoZKZZYrJZpObZGIrGkezQhmKO7eMWPlPNFxlO/TRF5xqSpJExVtolwzBEhIJY0mwnDmVpbXeYECFL7nXOLP7fZPQR9mV+BlNdzcL2XhHtdy0DlQh9zr95k9nxRZKZiLtDNzRkWaLZdCkVG3hugON4FDqSbbKmebvK81OXKbs2vghJaya/vecxnu3b165w2ErNpyHH2oq7XuhS9ZcIRYgsyRiy1n7myl6DhVqV/ek+blRneX72PDHV4BN9xxiKF1hwKvznmy/xazueoMfKcLp0i0c6dnGrvshbizf47OAJhuIFzpcneGH2IgUjSc8dFTaSJNFn7eKRwqd5YfZPmHNut5sgBCGL7hSLi1Ow+NV1zyOn9zAcP9Bu9gjCIrXmN8gkfrW9fUnSUOQCrncJSTJRlW5yyb9P030D2z3TmrTzJK1PEjMfbW87Gfs0tebXsd3TeP51JP04oWhgaAfb1Qq6trMdA77jDhOKMpnEb2C7J3HccxjaAZAU/GAOXR2hK/cJHO9yFFIQDqrSRyL2LJoaTR5J65NUGl/EDyaQJBNZ2ri2/k5syejOL1Q5dfoWTdtD11VURUbXVWp1B9NUNw8vAAU9T1pLUfYq7c/rfgM39DDWEZ6baEYJtjsxa89R82vtConNUGo0+eJrZ/iNpx4gn4zx1ZMXeerAKDFd49tnrnB9domEoXP69jSfOn6AVMzA0FTMbSzz74Vmw+WNly7z7S+9jef6fOoXHubS2XFOvX6DnsE8sYTJM588yn/4l9/kz/7jD/jZv/k4fcOFNa3TdtPle18/zVM/dV/b6JqxlTKxG5en6ezJ8KFPHm1/fyeOdHfzjx9/gpimkdB1dGUtQU3OspCAjGm2mhEiv6DDjJQeLEUjEIIeK4Mmq9vgSYhWDn5YXxPrchouM2PzTN+Y5cGPrl8NEBncq5xe/COmGq8x3XgLSZLZn/l5dOXeGWnH9/HDkLi+9ro4vh+tcFr1qk3Pazvrphqdo2HpnHh6P2OXpjBMHVmWsCydru6o8ce2PTKZGKEQq1Z69xeGOJofwAl8fBFiKdqaDs2xYoma65AxLQbWUcgF0GULtWU4BCENv4oTNrCUSAa+6NZJqCYx1cDwbdzA52Jlimd77uOd4i1u1Obot3KYio4iKZwp3qbHyjBnV+hIJ9mX6uFKdYYlp4YqKa1ktsaSU1tldCFqVNiVOA5C8PLCl1psY/eO6yuSykBsL/2xPe3PhHDw/XGE8JGkZUpYgakfJh3/RTx/DD9YQJIUNHWQXPJ/XKdiQBCGdWQpTjL2WRS5gBfsJB3/RRR59aQc1eeuhizFCMIykmSQjH2GZOwz6x7/emOjfddQ5CyZxN+kbr+A599qe8D3wpaMrqGrdBSS+IFA06L6XE1ViMVsmk1vUw7JbrOLT/V9lAfz9/NO8QzPz/0AgeBM+TxHMoe4L3NglbJoza/zjZnvMnNXOGKyOcU7xTM80/3UPY9XliS8IGSyWCafjDFTrvLO2CRP7R+lI5mgbrscGOjmW6cvoykyKTMiXDE17V0Tr9yNsWuzzE2VqJQafPATR3j4qb0Ul2qcev06xz+wi/7hAj0DOUb39vLWK1f5Y+/7fPoXH2HPgb5V0jCXzo4zeWuBeHIlfJLKWO3jVFWFF799jlQ2xgOP7sHzfK6en2JupsT9j+7mZGWO79y4RsYwyVgmWcMibZogwUgmy4HOTjpicTpia1scdyajCoAOc4VEyAv9NcmaknOJlD6KLGk4/iKSpKDKCVbI2rX2SxOENjXvNjFthHg6Rt+uHmItBri7EQiPueZpZppvEQiXEJ8F++Ka392JpWaTmuuQ1A1qrstio8GRnkihVbR0zmRJ4o3JSYYzGQbSkcG7UYxY7Wqux6GuLhItQ53Kxtl7bDhqBpIlVEki3UqmbdaFH9VFrzX2XhDghyEztSrT1Rr9qRS9qchDanoelqahtCYCXTZR7qgttYMaDb+CpSTosTL84fUXSOtxGr7D4517SbQ06/7s1msA5PQ4buiT1iwezI+2ZeqHYwX+8vZbLLpVSm6T781eICZrpPU4R3NDDMbXJ6lSZY09qYiQ5mTxOS5XXqcRVNb9LUSGek/yAR4ufGqVioSEQhAu0bBfxjSO4Pnj1O3vYOknkCQd1x/D92+ja7sIgzLrBaojQdl3qDb+iiBcQJZTZBK/giJnNrkrK5AkkyAoEoaNduSh7JVo+A3yRqHdN+CFUennnX0E0b5PUm18tbXvJOn4L6MoW1uBbcnoJpMmh+8bbB2sRBCGNBoub7x5o/XZxmN1WaPb7KLL6KTf6qXklTlZPM28s8AXxr+EIinsSe5Ck1XqfoO/mv4mp4pnCRFoLYkSN3TxQp83iid5ovNRlNZVWn44G47LTKkKkkTc0Li9WCaXsNqkxZ88tp8/e/U0+/s66c+nefvmJL/33df4qSN7WKw32t6loSrrGt3ll9X1fZqeT7HRpNhsktB18vEYMV1vj102HlO3Fyl0pTh8YoTnv3aKaxenUVWJ29fn+c//13N0dGeYmSxSWqrx2V/5AAszZf7d//frDIwUGBjpwLR0Sos1Tr52Hbvp8vnf+x7xpMn8bIUL79zi6vlJ+oYKlIt1pseX+A//4lv8xR+9jNKSuj760CiyLHOsp4fBTBTb02QFTY7iua9OjPPi2BgHOrcnSa/Jax+Z6frzxLV+ZEljwX4LkOmKrbRoSpLSzh57YY2Zxgvszuyif3cPpfkKSKxb+K9IGn2xRyg5Y4zVnsNS8xzK/QqavH4FQSgEJdvGCwJMRWWiUmnfTy8IcIKAH96+jabI+GFEqu36PmHr+1AI6q5LeFeFwb2oSbeKpucxU61hBz5BKEgZBrIscW1pCQmYrlbZ39lJIdYKZdzh6UKkV1b3y+SNXh7p2M3uVKT2rCCTMyLv96d6j3A8N8JrC9cwlDsNuIoqR5SlR3PD9MSynCneZqZZ4vGuvSiSjKno5PQEMVVvywrdfU9kKYrP5vQejqQ/yKR9mYnGFZbcaZpBhVAExJQMXeYwe1InGIjtJanmVk3UspLDMh5kqfp/QjVAljOkYp/F1E9EyWDjYUL9KIqUQFX3YweVlvz5naE/CUO/D0XO4IcL6OoIqtLPsoEOQxF1zG5gnCTJRIg6Qtg4gUMjaDDZHKcRNJmyJ4nELmOEIqDL7KZgdNw5GkM/jCJn8cMldHUYVenfcuXIlozu3Qc+Pr7EN799BkmSKJUbW2JLlySJvJ7j5wY+iyEbnC6fZao5ze9d/wMe73iE+9IHOF0+xwtzL2GHNkk1waOFh3BDj+/PvURIyJw9z1j9Fl45SUzXGOmMKgWmihV+77nXub1Y4uhwL34Q8sS+EUa7ogRVXy6F4/v8u+dexwsC9vZ2MtKZ49z4DDfmlvj40SjWmEvE1tVHq7se56dn+dbFq5ybnsULQxbqdRK6TiERZyib4cmdIxwb6CUbi7zQRt1B0xQ+93eeZHRfL6ffuMHiXIUTj+2mWm5w/dI0w7s6+fQvPMSeQ/0IASO7u3j5uQu8/cpVDEsnW0jw8Z99AFWTeeX5i5SXapgxg8efPcjSQo1zJ8cQAvpHCuQKCXoHC+zc28PgaCepTAxNVynaTVRJbideAiHwwpCKY286Wd6NUHhEjFwqQejQ8KeJaCoFmpxsE5Uk9Z2UnAt4YbkdUlAkAy8sU3Vv4ocNDCWKXxdny1x64yq7j4/Ss2Ot8ZckmaTWx7HCb5E1dpA1dtFhHmQtvWaE+XqdlKGTNS1Kts35uTmO90a1k6dmpkkZJg/29/PG5ATj5TIHOjv49vXr1F2X+7q6UGSZkm1v7kXchWhCDlqcAptXUNRdj2tLS5GhFyLKNzgOE+UyKcOk6XmRFEwLkXz5yiqwGVSpByUA4qpBXO1Ytf3xxiIN3+FmbZ4Fu8qORCdjtQVu1uY4Vxrnp/qOsOBUKXkNFuwqtxuLPFLYxe5kayVAxJJm+z4V18ZQVBKa3nYmlrmPXc9HBCZKo4cD8UEOJZ5BVcD2PDRVodpw8D1BVzyDtg7DmoRBMvZZYuYTrWV6HklO4IcuYVgCCcruFEmtFy9s0AyKNP04KX0AtSUDJUkSipRG1g4S2i6qorf4VgJkWWZuoUIqaaEqMlJrZX4nZDlJMvYZFKWTieY0vggYjA0RU2LtoIkfelT8KjFl9Sow2ncGWUujt1Lo2wlLvqvWinrdIZEw6e/LslSsbxrTvftge61ufmnoZ9lb3MXbxVOMNyb4xvR3+dbMc4CEIinsiA/zgfyDPN75AV5deANLsagHdezQZqI5jSiDH4R0Z5KYmspAPsMnju3jyvQ8nzi+n2rTaSUcJJquxw8u3uDRvSM8c3DnphfnI4d3EtNXx4xLTZs/O3mG//zmKbqScZ7ePcrjo8P85zfe4XaxzC8eP8zrt8b5359/iQeG+vnbj5xgKJeJYt7VJoqi8PTHDvP0xw6vu8+q57DoNckZMZ74yH088ZH7AAiCkOJClUTKYn6mzIGjw3R0RwvaIAy5MbNEuRF1WQ13Zmm6HvPlGqEAR43qeUMhODs7yw/Hb5M2TTRZRlMULDXyArsTWwv8A9S9cVQ5gaV24oZFrpT+AxIyltpNSt+NLGkIESJLGpocR0YlYxwAwFDylJwLzDffQEalYD0ACAxLI5VPkuncWMBQkiQMJcXezN+45zGOlYpMV2t0xuN4YUhC06g4DienplhoNNBkBUWW8cOQxWYTxw9ouC5Nz+OtqSmO9/ZuK7wkhMAOlphuvk1WHyWjj2xqeAvxGM/sHF31Wc1xmKhU6EwkyFl3N0hodxndGnV/4+X8bLPMd6bPEoiQD/ccwlJ0vjzxFnbg8lO9h9mX6uVadZZvT59hzinzdNeBqLZXisj9b5aXuFEuoskyMVUjBLKGiaVquGGkYmKhsrBQQ1VkbNenUldxXJ9M0mpphGmYukax0iQbT6HH1qPNlAAdVelpL+1D4WMHRRrBAm5QRZJkqu5kNNlL4IhKW7vt7ntw9tI0u3d0UqnZqKqMoWtcuTFHIh69y52FJP09q5f+smSiazsRQlAwFGRJbne6LkOTNSx147rs6Dy2H458V0a3tyfDG2/d4IevXuOBEyMYhkqtZtO0PZIJE3OTjiqAlJbkqY7HuC99gKu1G0w1p6n7dWRJIatn2J0YZUdiGEVSiKsxTMWgHtRxQ48FZ4GDmd186/QVLk3NkzD0qMxJV9nX10nc0LkwMctircHVmUVuzC5StV1+/cmNu9TCVl+5okzihQaK3N9+eZ6/cp3/9MY7DGTT/M8ffIz7ervRFIWj/b28NjbOXK3O3338YfLxGH/y1mkkSeIfPf0oVkxnbrqE3XSjZoa7sGDXias6c40ql8sLPNA5gHEHEc7NKzN84y/e5LFnDvL8106xa38vn/j5h5BlCc8P+M47VxjqzHJ9ehE/CLg8OU8qZrJYbVBIxukvROTbjw8P89jQUHuZCeCHIV+/cpkfjI1xemaakWyWlLF5yV3Tn0OWFjGULLqcpSf2NL6ok9b34osGblBivvkGTX+GrHkfmpLCaFHlWWonw6mfpunPU/NuEAqPUAQkswm6hjpYnC7Sv6tnw30LEVL356n7M6iSSVLrRV+Hha03mWK2VmeyUsHUVPZ3duKHIdeLS9Rdjz2FDrwgQAgYSKWJaRoZy+RGschwJoOpqoxksxhrFAbWhxfWuFj6C65Wvkq3dYz7cr96T8N7N+K6zu5CYV1nQJW0VfkON2y26BeDNdUWAMdzIxzNDREIgd4KA+1PR5n25WqTXaluRpOd+CJs/2YZ49UypqKiyBJxzcAJfJq+z0KzgQSkDRNVSKQTFkII4pZOw/awXZ/5Yo1YK5loaApdueS7EIuVSKjdaPrOdgjJDatocnxVbPtOCAHZTAzb8ZhfrGEaKqapUcglonfFC9ZoJ67aoyQRV989ZWPgB8zcmieRipHu2Nh5uBPvyuhmMjEeemCUV1+/xsRkkS/+1zdRVZl43OTYkUH6+1Y3CFS9Ip5wSak53NAmpiaJlHDz5PQsNb+MKhmosraGZUyX9XYcMRA+Nb/OzoECzxwSXJ1ewAtDkqbBjs4sQ4Uspqby0K5BKk2HG7OLhB05dvcU6EwlWkskl0iJYcWj9YIyDf8mXlBCkWOY6kqb6OnJaZqex9O7Rzk20Nee1wxVoeF5XJ6d51cfOMrPHbuPt8en+N6V6zw8PEBn3KBes2k2HLL5BEt2gyWnQX88zUyzysXiHA93DaErakSLeBfhgqarJFMWsYTB0Yd20tmzkrYJhSAIQz5+Yh/ffPsyAJWGw2ceOsiNmSXG5qK26WXqQiQJEdaBAElKtLcxUanwg7Ex6q7H/X1968iZrCCm9TLbeIn55pvocgpL7Sbm76VRkxGooHlISoqYnyNj7ENuvSRO08VtusTTMRRJx1Q6kCQFEQjmJ4s4TRfN2Pgx9EOHycar3Kh8i5o/gyLppPVhRhIfoss6gnJHgmMgnW4nxu5ERzxOw/MwFIWsZfHR3RFf8s1ikfl6nQf6+3l8aAjtrvNvBg6BCIgp5rocB4HwmG2ewg6WmLPPUHFvY6kFDDm5ZcPbvkfrQJGU1ZzJCBpBFS90MJS1HpgkSSgo3En5u15pX8RHu/b4DhS6KJgr210OKdQ9F1NR11wfgKbj4RZ8ZFlG15Q2LarnB6sm+ntBllRSev+az00ls+k4RZHZNdJJEIQk4gaeF2IYKpapbWvJ/27hewHXT91ix6HBH6/RlSSJnaNdFApJ6nWHMAxRVQXDUEmnopsmRMSl2QxqOGETTTYICakHZXzhIRCtDhaJsjtHTE2RV9d6O6qktKSSI0PhhC66qmCoCk3Xw9I1XM9nb+9KTDBu6MR0ja5UolUcTXsJZftThMIhpg6DBA3vFqGw0eQsjj+DJFZfkv3dnXz/6g2mShVc30dTomTEzcUiQSjaSbiORJwHh/o5MzXDa2Pj/EznKPWqg9P0EEJQ913enB/n3NIMXhjghgFfvnkOXVFI6mu9zP7hAn/j1x7FihuM7umJzmPLD1HEHSjaWnaAqCKCSVB3AknSpsnjQ0M8MTzMC2M32ZXPkzHNiHVrnf3E1T6yxqG2V5u3jhNgcPPyBG5TIp52iCV20qg2mV4s0zmg0qw7BH7I3K0Fukc6yHVn0JV0+/nIdqZwbZfiTJm+0W6QV59jKAKmG2/yzuK/p+LebndDLdqXmG+e5VDuVxlKPIUi6Ztem454S8b9LiPQEYvxgcEhcpaFIkvU/Wa7A2+sMY2pGDR9m7SeQAjotQotus6QRecSTlDhYO6XuFJOkDf2UPNnuDn3zzmS+1uk9eH3/tJLK2q1umwyHD/E/tQjUcPEjwH5lsG987hlSWo3zqwHy9Cw1kk0bt/LfW9QFJlE/N01SG0VQghEKJi6PoPvhzSrTeyGS3mhytJMif7dG6/W7sS7psvRNIWOQpKOwsZxQTuoU/LmseQEQoTYfo2yu0AjqKHLBm7ooMsGcTW9oWcQ3sFFL5aJSIF8Ms7BgS66M0mCViY6DEUU9Jcl7KaH3XRIpmKEoUDTFbywRM29ghcWKdlvIcsmQdggpR/CDiZxg3nid7XxfvLgXhbrDb5z6Rr/+/Mv8diOIaYrNb518SpdyTiP7RgGIsM+kEkjSxJO4FPoSvHkRw5R6EpT8Rwul+bZl+kkoRnkzRimohGIEInImzBVFcf20FptzVKrTnS58aTZcEmmttj3L1yEKCKCWURwC1CQ5AyRrpmMIkk8OjjEw/0D6IrCH77zDjeKRd6amuQXD91HxlrLXiZJCqFw0eQUimRgKFnmS0ssThaxGw6LU0V6d3YjgpDZWwssTBUpzVXIdKSIp6w1vMWSJGHGTZK5BE7DZX5ikc7B1ax0VW+Si6UvUvOmiakdKJKOF9axgzJl7xbvLP47YmqBbusYm8XWljvs7kbCMEgYBnW/yeXqLbwwIKenqPh15u0iaS2BpRhUvQZO6NJjFQhFwIJ9gZOL/46qN8GR3G/yga5/zGT9Nc4s/RFVbxovqPFo9+9gbaJEsBVISMTUFMPxgxzNfohdieOYSnzd0MJ7xWYe9/uIIIRgcbpIeaFG93AHty5MMHZunI6BPOY2OmJ/ZBppd0OSJHJGNzljhZG+7C3QYQ6QvkuXPhQBvlifvcoJXbyWyoEiKcRage18IkY+sWKEHNtjZrKIJEloukKz4VJcrJNImTTrDvsOD2AYWTrjH97kqJ8kDF3upI6L6Tq/9YEHeGCon+9cusbn3zqN6wfs7SrwsQN7eWbvSlunF4ZYmsb+rk56B/Or2ns/1L+rTXqyHs9BGIacOT3Ojl2dNBou8bjBzFQJ1/HIdySxbY/UFslWhCgj/BtIcgHJeKZdriWEDUThG11RoOW1O76PEIKOeHzTzH3BWh0X7+jP0dH/4D2Px3W8DV/oXFeGXEs8ctU5EDBvn2XOPsdg4nEOZn+ZpNrDnH2Wi6UvMtM8Sd2f43zpC+SM3RjK1pZ26yGuWgzHexFCsORWGI73sC8ZeapO4FHyquiyioyEG9YYqz3PvH0OIQJu1V5gIPEBBJHcuCRJWGoeZY18zF3nFwrefu4MTtPd9Heh08uQdhi5UMD8wIqGnOd4vP3cWQJ/bXJpPeS6M+w5sbO9MmvWbE4+f3ZLYwG6BguMHh5u15BXlqqcfenSPUatoG9nN0P7V7gXFqeLXHrj2pbHD+3vp29nd3v87K15rp0a2/L40cPDdA2txM4nr04zdmHiHqNWsPfETuLpGJfevB41xjQd8j1ZBvf2MbS/b0tiCMv4sRnduxGRYARktbWEMLKkoK8zewsEFa9CI4j4DAxZp9PoWPM7iEhfXNfHbrhREbuqYMV0wjBEN7beRSWvw5ikyDInBvs5PtBHvZXtThoGpro6/jxbrdGRiPP46PC6245KTdY/jjCEZNrEsX3KxTqO7WFaOpVyA9cNSCa3zp0rSWkkdSdIMe5kV5KktbPxdK2KFwRYmortea0uqR/tMu3d1Ln6ocN0423S+gB70p8hZ0STW2/sASwlx1uLv8t0403mmqcpOlfpjm0uuyKEwA8bzNinqHnTqJJJwdxHWh9GlhSsVqdjTF197oai0XXH6keXEwwlnqLiTmAHi+zP/gymkmUo8TQgMW+fY3/mZ9sdc0KEVLzbJLW+dpwbokn2X/3Wv2d+fH0O2btx4JE9/PPnfge5ld9oVJr8s1/6P2nWNpc8X8bDn7yf3/nC/9RWuF6cLvJP/sa/3NJYgGf/5lP8/d/9jXbb/vilqW2N/5l/8Al+/X/9HIoaPf+XXr/GP/mZrY//W//r5/iZf/gJpFbA+p3vneP/+Nu/v+Xxf//3fpOP/PpT7ff15S+/yR/8f/5ky+N/58/+J+7/8GHsukM8HWPXkRFS+e3x9C7jJ2Z0ZUkmb2yPx9YNXOadBeyW0Y0pMUYT67faJZImu/Ztjc/y3WKZJCZprO/FxDSNz9y3j+H89rkBVFVm995egiDEMDVUVSaRMOnqTmNa20wKSAaSvLnnV7ZtTk5PcXF+nhN9fQymM+SsGIV1OtP+OhDiUXJvkNQGyOorZVaSJJMxdjCa/AgVd4ymv8R4/Yf3NLqBcDhf+lNuVL9D3Z9FlUxyxm4O5X6FHuv+LV9fSZLpNA9xrPC3cYManVZU4qfJFiOJD9EXe7hVyB+t4Cbrr3Gt8jUe7PwHq9Scl9Gzo4sP/uJj6BtU/Piez5f/zTc3PJ7hgwM89tkH0TaY2Oy6zZf+9Tc2HL/7/lEe+tgx1A0I/csLFf5yk/GHHtvHsQ8eQtmATnXu9gJf+/3vbjj+/g8f5uCje5HXKzsVgtuXpnju8y9uOP4DnzrB7vt3IK3jaQohuHryJi9/6fUNxz/9uUcZOrA++5kIBed/eIk3vnkKAFVXGTk0SBgKGjUbI25syn29EbZtdNcTRfxxYcFdYrwxSYhARmZHYpih+MBPbP/bxU/t342pqswvVimWG+wd3d4kA1FCIJtbMXzrlZstY7lnfdVnW9yPqaqMZnP0JpN0xOOkTZPsOrHcjeAHATfHF0nGTfLZOFOzZQZ7V1Nirj6+1SGViJls9dGuHhti+0VS2mBU7YBoJ5UkFHpjJxirPU/d/yEL9nlC4SNvwOgvRMjV8l9xqfwlnFZzgSs8ZppvEyw6xLs6W/vZuuHNGbvWXGtF1rHuWCktOVe4UPpT5u0LlN3xdY1u50CeT/7WMxvy9NoNh+f+y8ZGp29nN5/87WeJJddfnZTmK3z93z+34fiRAwN88n94Fiu+viMxcWWaL//bb204fvfxHXzq//UsijZNGC6i38EAJgRcfP3qpkb3wCN7OPzkAW6cuY3vrYQYH/jIEQp9OV7/xslNje6Rpw/y4V95Yl0SrCAIef6PX9rU6D740aM88sn71zX6vh8gybSNrixLJLJxqsU6t6/OICsynf1bo3K9E+/K052xZzlVOosTbh6Peq+YsWe5VruBKqn0x3r5dO/HMO5a/i8zO20kWxKIIKLqgzYl4I8LHYnIWM7NVphfqrF3dPPfh6HA9fyWBPzWb4UkSTiez7/+6stMF6t8/P59WLrGf/zOG5QbNvsG7q2LZagqg5nMlvcJ0UNsux6WqdNoenzv1SvsHe0iuxjji988yT/5ux9bJVHuhGXeXvhd4mon9+V+rc3FsNxWvWQ3CITAUlWS2tqXXpJklpwrTNRfods6hiypUZgqdCh7twnCiDLSDorUvGlS+toJWQjBTPMklytfQYiQvthDWGoHRecqJfcmi/Ylzhb/Cw92/AO0TRQG1mJ14ikQXquoX2mXjC3Y51mwLxAKj0X7Ej2xY2u3Ikuourqhp+p7waZxdlmW0TYZfy9Cf0mRNh2/kQfb3r8S7V/RVTx3ZtV2hBD33L+iyNw8N86BR3Zjtgz/6R9cxHd9NENFvUcVhKIoaLq6rqcu+8H6HvSd41UF1dDWbfCSZAnljjI5SZaJJS2unx3nxvkJ8t3pn4zRjXrCdabsGV5eeK0tjX7Pca1HVEaOeqKRW5R6MrKkoCCjSAqKpKDJKpqsoUkaI/FhBmJ9fLDzcXrMtZ6jG7osuQv0WH1rvgOYaU6T0lJU/So95trwgxCCpudHHAySRCER2xYhtiSxph5RluV2yYwQAsf1Mdd5qF3P59yVaVIJk90jG3MgLE8sgohIxdI1/tFnn1z1m8cObI3h6L1goVTnL799iice2MXoUAcd+QSaqmDoakRdepdx8EObefs8Zjyzygu1A5/JeoWJWglVVtBlhROdq2s0ZUkja4wyUf8hP5z930jrw8TUPH7oUPHGafhzKK0240C4NPz5dY2uLxrcrH4XO1jiSO7X2ZP5aWRJoekvca74ea6Uv8xM4yST9dcYTj7Fu+kwEiJkuvEmpxb/A13WEY7m/w6qZKJIJopk4AuHBefCtre7HQRBSMN2iVvGGqa6e6HRcFkqNwnuaEHOZOKkUltf+YRhjc1kcTZDMhvn+T95mcCPSgJr5Qa7jg6/q239OCEBibTF3uMjVJZqNKo2nuujamuZ+zbDu/J080aOZ7qeoubXOVu+gC5rJJQ4cTVGXI0TU2NYstEytNKGRlaVlKghQlbRJR1d1jAUA0sxiSkxEmqcpJbAlDeWsWkGTa7Wrm5odGedGRRJYaI5vsbohkJwc3GJb1y4wvmZOTRZYV93x6ZNAndCkiR2FXI8tnNkw1dVCHjlres8/cgeJEmi3og4GeoNl4vXIlntVMJsGa3VY/0wZLpW4XqpyFStws5sjoOF7g3ls7cLSWIV5aDnBTieTyK2/lIzEdOxTJ1v/OA8Tz64i2TcxDI18pk4hq6uedlD4WH7iwgRIkTYLgt0goCFZh1NVomr2rp1oKpsMpR4irnmOdywwqJzkcWW6ruETELrYSD+KLdqLxCKAC+sr9kGQNm9TdG9QW/sQUZTH22XW1lqjoPZX6Ls3mon5PrjH0B9FzWwdlDiVu0Fiu4NDmR/sZ0wyxqjJPV+HPs8Ffc2fuisu307aFJqzOOEK0mxlJahYxs5kCAMuXBthqP7+9d0mt0L167NUm24bb22xcUaO3Z0cOTI0Ja3IUlqu4763WD38R1tr3T6xtyGnvd6mF2oslRurAp97hzpxNiGbNdWIcsy+e4Muw4PsThTYvLGLAM7e9oJwq3gXRndUIT0WT18uu9jHE4fBCI2sYyeJq0mSWlpEmqcQPg4YQNZUjFks/3SvZeKwLpfo+yVyet5DMVEkRRUSeF8+Sy+8Nmd3IMiqVyvXSWmxIgpsXan2xoIqNou70xMM12uoigyM9UqMU1DV1U0RUZXFFRZRpFlFFlqGylJihJrWcu6hzS24Ob4AovFftIpi4vXZ+jvzqJrCp4fMLtQ4fLNWXYMFHjwyDCmoSGA03PTPHfrOpcX57laXGSmXmN3rsCBQuePzOj2JlJ8dMduhlIZAKbmypy+NMlHHt+HrqnYjscLr19F1xQePrqDuGXwC5+8n9MXJzlzeQrPC+jvymBZWvuFXX3mIV7YwA5KuGEVo5VgyhgmD3UPtn/nBGvLBWVJpT/2CPXMPDdr36XqTaKgEtd66LQO0Rd7iC7rCCV3jJJ7k40kuaveJA1/ngOZz6HKqz03U8myN/PTzDTfoezdpuHPrestg+B27SWq3iTDiaeJa6vDNxVvnAX7Akm1l6y+o91FltaHyOqjLNqXcMMadX+WtD64ZusLzgwNuUpSja5Pza9Q8UrbMrq6FvFbN21vW6EqiDzdMAxRWrpovh/guNsTIFXVEYJg6yVYd6JebpDIxNtLfKfprorv3gunz0/Q1ZlqT/qXrs3S1ZHCyP74ksL5ngyZQpKLb99EURQGdm39Xm3r7oStYv55Z56qX6HP6mPenqEZNOgwOglEgB02iYkYRXcWWVKo+ksYsoWqd4MIKHvzZLTONiP+dqFKKjE1xpQ9SWfrobRDmzRQD+o0/CYxNUbRLeKpHtodbcR3Q5Yl9nV38I8//CRVx0GWJDRFQVMiVq5lQ7usaLBcX7tsYCWISMHv2q5laqRbJV6SJHFwTx+/+/kX6SokqdUdOgtJiuUGiizTkU+wc6iD3q70qrjSVK3Kl69eZKJabn92Zn6GM/Mz7+q6rYfhdIa9uULb6Lqez5WxOe7b08twf57XT49xa3IJ2/HIpGIcOzCApioc3ttHudrklZM3kBUJWZbX7UqKVjlKO37aZR1Z9zju5Jy4YySGkmFv5qfpjZ/ACcpIKBhKkrjajalkkKTI4637syS19Vc6TlDGDarocnxVS+0y8sY+Os1D1LwZqt7UukZ30b7CmaX/hCoZjCY/suq7UPiU3VtUvHF2JJ/FVFaSiZoco9O6j4n6y4TCo7GB0Q1ESMUr4bSqdOpBjfgWSNrvhmXp1OpO+9nbKjKZGKfOjtNoRKFC3w/YMbo9yk8QrTrw7SPXk+Xcy5ei+DVRDfJ26l4d12NyptS+7nMLFYLg3XvdW0E8ZTF9cx6n6dKoNu894A7c0+iGIsQNXWRJ5vtzz3Msez8Vr8ySu0jNr5HSUuxO7qbqVSl5JQgF8/Y4mhwJ1NlBnSpFKl6RRlCFVr1uSstjyOa2umu80KUeNMhqWcbrtxmrv4oTuIw3b1HzahE1mxrDD33GG7doBHUMxSSrPYN5F4PQrfp1us0+DNVktLD9YPhm6MglyLVIOCo1m1uTS/zCp+4HIbFUrtORT2LqUe2wZWrEYsYqSkkJeKx/iN86fII/v3KW2XodS9XIGCZpI5LdSRsmCU0nrunENA1LVbFUDUNV0WUFXVEjVrFlDl050kZTlicOIj20grUSh+vtSrOjP8+bZ28zNVfmWy9e4OGjI8wuVLk5scCxAwP4fsCrp25y6uIEfV1pGs3IS1pvKadIGnG1i6J7g7PFzyNLOhl9BFUytsBNEC0VNTkWVQqIgFD47fjteP1FZpunmGm8Q1fs6IZGNxQ+ofAYq32PtLEDS8lG/L6tpKouxxmIP8rppT/EDpZWH4EIafjznCv+cav77Dcw7uICcIIqc83ThMInqfWDJBEID7nVhNITu5+Y2kXdn8ELG+seo6lYLZWMxfYxZ7X1icTXP0eBH4RYhkYus33vrlisc+zYMIVWd+n167Pb4k3wgwnUMI0sv7v3aHGqyNM//whmq433zIsXtxWXzqZj3BhfbFPMyrK0ZebDdwtVU+keKlDozW7KHbLu2Hv9oOKVOVk6SU7LcTx7givVywQiIKkmSapJSl6JXqkPUzGJiRhu6JDXe8nqq29A3S8hSyqWEscLHep+BVmVMZStz8pO6HCzdp262YlAcCh9mIwW1cRKrT51CYlqWOW+zBFCETCa2IUhm3SaXQgETmAjI+MLn2bQwA6bxJUk6iZxsOVkW9VxaLgR56mhquTiFgl9bd+/osgI4M0zN3n77G0evX8nvZ0ZrtycZalUp6czTVchtU4MNMQJHQzZQFfg03t288hQjm6jC0VuGQpo8UmseNvc9dkdH22Ou1o/TV2jkE3wvdcuMzFT5OmH93Dx+gxLpTqH9vRFHBJNl2+8cB7H9anUbI7tH0CRZTrza9vBdTlJf/wRLpb+nMn6qxSdq4wkn6HHuh9TyaJIBrKkIElyK1kYRp1dIiQQLn7YbHGpLlD3Z6m445TcGzT8hXb8MKn1siv1iQ3LxUwlg64kuVb5BnV/jh3JZ0nrQxH9ZGvCN5RURCbjL9DwFwiFRyAcqt40V8t/xWTjVXQ5RdoYouLdRiaqogjxKDk3mW6+BQjOFv8z8/ZZBuKPktF3oMkxZEkjrnVhB0VMZf224IZfI5/o5IAZTRyRQ7MxhePdaNoeizWbucUK+3duv0wxFtN5++0x3FZIodFwePiRXVseL0tJfP8aIqyuit1vFfGUxfe/+CphyzutFuvs3EYibbFY58lH9qCq0X5fe/vGj620VYgofm47HqoioxjqPSsk7samRlcIEXm5SFyrX0VXdBpBnZJXIqkl8YTHeOM2TuAQU2JYqtVKmMlrSriSrU40IQRe6KLKeluobqswZJO4GudS5QKD8RHSWmaNsRRCsODMs+Qusju5t+3hSkg0gwbXa5eIKREXxPXaJRpBgx3xXXSavWjrhDyuLyxxaXaesaUi1xeWmCpXsT2PvkyanR15ntg5zLH+lQRdGIY4LYalB44M84HjK3VjR/ZvXmPshi5Xq9fpNruYtedIagmmmpP0Wd3rLsF/VAiCECQYm1xivljl1z77EMP9eZZKDR49voO5xRpxS8f1AhpNl3/y9z6GIks0HQ9DU9E0hWcfW6uCqslxRpIfYsG+yLx9joY/z/nin3Cx+EUsNYepZFFbhkm0vNhAuAShgxOWcYLKuskZWdJJqn3kjd0MJZ6mq9WgsB6yxk4y+igzzbeZarzOdONNLDVHXO1BlSNtODesEQiHW7XvU3Jv4oZVmv4iVW8SN6wC0AwWeXnmnxJXO1HlGBDihjVq3hRuGMng+GGDiforTNRfQZPjxNVOdDlFM1jEVNJk9PUrTEzF4lrzPNdqF1vb8RmMbb0aJW7pSH5ALhPfduUCRLmJ/v5suzxraamOtY2if1lOo+uH8f0bhOEiirJ+1+iG41WZkYOD7RbjuVsL20qkJeIGX3/uTLvsu95wePSBrasubxfFSoNbU0vYrk8+HaczFyef2Xo4aNM3OSSk7tcICRmOjaDLOjk9z2hiFyCoeGWOZ++n6ldxAoe4EieuJtaQAd8JSZJIaJn19ydCSl6ZJbcECAp6npSWbBtwTdbYm9rP3tTGMseSJLEjMcqOxIqxE0JQ864RYpLWsuhy5GGpsorpNwDpDlqd1Zit1vi9V97g2vwiuZjFwZ4uHh4ZYKZS4/9+/W0WanV2FfLRyxtERM81xyVh6BTrTXoySaxN+DxXH7tMVo+uTVyNkdfzrRrkHy8VyZWxOd44Pcb8Uo0Du3ro7UoTBCEXr03T153h1uQS/T1ZGrbL3EIVTVWYW6yyWKqze6SLrkKyTRi9+nwkssYujuR/g6vlrzDZeB03rBLiU/fnqPtz6xzN+lClGEmth7Q+Qs7YSc7YRd7Yi6FkNi3XSetD7El/Gi+ss+Rcbgk8Rh7t3Vh0LrHorOYTUCULWVJxwxrNYIFmsHacqeQomPto+POUnBuE+HhhnZJ7M7oOyOxKfwpdWX/pH4iATrOHuBqtFup+dcO6841QyCW4Mjb3rjy8esNBlmVirUacpWK97fVuHQqKMoAQ24tvAtSKddKFZLvW1veDbSXSwlDwwNGR9oRz9cbcjzW8oKsKPYUU33j5Avft6SNuaeQzWx9/T/dJk3V2J/bQH/vxdIJF7Ps2l6pXOV+5yERzippfBwFpLcVQfJCj2UMMx4bWJMSEEASiiSCg6U1gqt1ocoaad6O1rFtJWtS9MUy1i4HYileU1ztoBg0M2dwwvHCkr5v/91OPcqtYYiiXYSiboSeV5PNvnUKWZOqOy9XZSNJ9tlJjpJDF8QNqjoMAsnFry0bXkHUGYlG9aieRt5DU3l1/93aQSVkM9ubYO9rNvtFudE0lDEN2DBYwDY2RgTyBHxKGgp7ONAvFOqmESTJukFzH2IqwGrUiSzqKpNFlHSGh9TDoPMVs4x2WnMvU/TncoIovojowRTJQZQNF0lFlC0vJEVe7iKmdxNQOLDWPpeSJqx1Yam4Vj8FmkCWV/vgHiGtdrdKws9T8GZyghBc2CIQHd3nTy+oXWX0H/fFHSWi9LNjnmG2eouKO44Y1JElGl5Ok9AEG40/QH38EN6yy5FxlvnmWknuTRrCAF9SQkBlKPLnhMQbCpx5U8UVE7NTwa5jr8OXeC7br393ktyVkMvE1ibThoa3HlAEkSSeSPN+6Gskyct0Zzr1yuZ1Ic7eZSLNtj8WlWttTXizWtpxIE0JwoTTOrebKZCpJEk91HUBnbZ5CkmCp0uD05UnmijUKmTjZ1Pbi6JsaXRkZU7HwNmmAWO4uCsIQJ/BbyYlIznmjelchBH6LkPzt4ileXXyzLbHutdjGFElBtRUu167yysJrHM4c5OcHPttmGQNo+hMsNF+jI/YoFfcSgpBQ8Wl4YzhhkZp3HV3JkdBGUeQ44i65Dwkp0j+6w1ESRCGC28Uytu+zs5DjsdFhHgx8dCWSexFCcGuphCpL7Cjk2N2VJwgFe1rEMUKAF0YhhrihI0QIOHddBYmI8evHEzYQohUhbdFeQtTEodxRfbGMno40XfkUdwr5ybJMT2erxCtl4XoBshQlKHKuh2mszwchgnlC7wyS0g/qTiRJQZYUklovCbWbHus4XliP1H1bvMrR1ZDbOmMSMoqkR//J0Z/L3WyBEDR8j7pXI2tYqLLMeK1E1XXoS6TJGmtXWapsUDD2k9F3sDP1MfzQIcRDiICGv8gb8/8KQ0lxMPtLJLRuoidfQZNjGEoaRTLosY6xJ/0Z/NAmxG8dsYoqmxhyCgUTVImMspPB+BO4Xo1Go04ofHRTI2sOb3ivTCWGF3qU3Yh8PiQgtcFqcCOEoaCQieN5PsYGPAoboVSKEmn5FoHLjXftKb67FdnSTImnfu6Rdkfa2ZcuISvbSKRlYlwdm28n0sJQbNloCwSXK1M80LkSw35x7gJeGKxLwgUws1BhbGqJmKmRiq3veGyGze+OBDk9F7V/tti7dGM1s5YAputVbpSX0GQZQ1GZbUSKp58eXR0GCERA1asx7yxwunyON5beZt5ZwAu99uLekHXyeo6D6f30WT28vPAaV2vX+eHi6+xP7eGh/In29pxgCUW2qDiXsIN5AsclCJuEwkGR4wSSTS24Ttk+j65kMbTVRDTrLkuF4FaxxD/77g+Yq9b5xx9+kqP9PVyanafcdBjJZ/GCgDNTM1iaxrGBXhLmvWj8ajjNrwI+tD00FVU7iKrt3XTs6kMTVFwHS1XR5NVdMKEQ2L5H3fNoeB41z2G6XuNWucRCs44sSYxmcuzNddCbSJI2VjecbBYLlCRp1YtsmRvzQUS/15HkPNxVoiVJMoaSihJXwsX3riIrXchy/i7eBQ9YPr+Vz8uOzVh1CV1RafgRXWRKN5hr1ghCsWncW5IkNMlCu6NWNxAeU423qPuz9Mcfpi/2IMoGzRG6kkCT4zRrdiRVk4rhez4iFHi2z8TYBMl8gpmbc+w5sZOZ61W+/G9fwnU8jj51kEOPZkh3pNBNDfWuOtqGX6Mj0U13S1qn6C5sK5EG0Gi65LNxJmbL7N2xvVyJaWq8884tvJanWa87PPjQPXrYf4QwEyYv/eUb7URaeaG67UTahx7fd0ci7eY29h5RBHx14q32J/NOlXCTRo/BnhySJHHhxsy2REyXsanRXS6rKZcb3L4xTyYXp2cghyxLCBERmc836kzVKygt8u6EZjBVX/3AhCJkxp7jRn2MU6UznK9couJVV/3GkA36rB4OpvbxgcJD9Fk91Pw6Y/XbXK1dxw4cXl18kxO5Y23dqKx5mCyR4GMXT2H7s7hhmYQ2smb5GXnk3j0vyM3FIs9fuU7GsrBUjXgrNDBVrvIHr72NoSoYisp4scxwPsuhnnvzHEhyAjP+udXHEzbvMMD3hhCCsXKJP7t8lp2ZPM+O7Gx3cjU8jx9O3ub8wizXSkvcKC1xq1Ki5q2sUBRJwlI1sqbFh4ZG+dy++9id294ScgtHCXICSekBUQWysM4SDSAIpqlV/yWm9TFM61MsP4oibOA4LyDLOTT9gVXGeNFpoEgyQRiiywrTjSoN3yWpmSiSRBBuvTbTC5tM1l/lfPFPiKtd9MYeWtfgeq6P53goioxje7z+jZOk8ymOPn2Ac69cptCfp16qc/vSJNmuDJPXZthx3xCJTJwP/dJjVBarPPcnL3P25YucePYIu4/voHt4daIppWVZFNOM1a8C4AufnL69e1Ot20zOlujtWCtXdC+oikJ3d7o98ZZKjW0l0t4rNF1tceVG/56fWNpWIi0e0/nad1cSabW6zQdObH3SiKsGw/pKXbIqr6jVrId0wsQvpAiCcN32/nth8+qFUOA4HrVKJEuhdCRp1ByKi1VicZOu3gznFmfwwhBFknlyYAQ/jFR449od3KEi5KWFH/Ld2RdoBqsD7TIyg7F+jmYPczRzH4OxvnYVQVyNkTeyKJJCIAKmmtMsOkU6zfUfSEPpwFA61vVgo5KyzT20IAz54qlzPH/5Or/5yP0c7e9lJJeJFH137yAfj3FmaobLcwvEDR0vCLi5VORI39ZkOoDIw3PfJgyXkNDQzCeRpM2PSwjBWKXE7556nW/cuExM1SnaDX5h32Hiuo4fBrwwfoM/u3QW7y7D02HFOdjRxZ5cgbRh8sVLZ/n8hVPUPZffeeSpTaVY1j+WkCCsIrcy+HdqzSFCCBYiT1VUkZRuYIOkqnARwkGIBnfGVMOwRLPxRRR1B5p+nDu95R2p1WWI8806oQgZTGZXJRsD4eGFdQwlvaaMLhQ+FW+C8frLXC3/FW5YYSjxVLsCYvrGLPm+XJsD+PqpMd767hkCL0CSJUrzFeKpGFffucmNM7fpHumgUW1i1x3MuIkZ0/nOf/oBS7NRU4thadz/zH0kc4lIrmgdY5Y3OhjKDG9+4e+BTCqG4/o4rs9CsUYhu/VcQK1mYxgqiZb6Qb3u4LpbI0f/UaCyWCPfk2kb2vmJJQJvO/uXOH7fUPu9v3FrfhvhEUHFazKcXJkIb9bmNkysA1yfWGBscpFK3WHPSBfJDdrmN8KmRjcIQ2qVJpIkMbSzE8PUWJitoOlKe1aKaTpLdoMHuvtbhfgK+/Oru1kUSWE0PsLLymurjG5Oz/Jw/gT3Z48yGBvAVFYfvCzJpLU0lmJS8+s0gibjzYkNje526wPXjAfuH+il6XkEoeAvTp3jYwf2cKC7E0vTeGh4gPsH+1isN/j9V97kq+cu8dL1sW0ZXc99Hafx58hSCkGAZj59zzF24PP92zf45o0r1L0ohPDHF05zoNDFI32DpAyT37jvfmRJ4us3rhCEIbtzBe7v6uNIZw87szl6Esmo9K+4yHilzBvTE7w1M8lTgzu2dY2E8CjZr5LQDxKIOnF9z+orKKchFEhyGvzboO1ZdzuSFEeW0yhKL7BiiGQ5Syz+N5GVTjbykpfRYa2fwJhvnuVK5StYSo6Y2oEmJ1pSPw0q3m2WnKsUnWu4YZWCsY89mc+0W4TPvHyJhz56rG104+kYlcUazVqTBz5yhFx3lupSDbths/vYCPVKA0mWW6E3DStp0qzaJLIJdFMj151mcG8fyVzix8pwZ5kauUyMUqVJudrcltFNZ2KcPjfRTqR5XsDAwHuTGtoO7ntsL42a3ab6PPiBPeT7ts5J3Wy6SIrUnnhLlcY2OtIkkprBDxeutD+xg81XxDFDJ5WwqNadNfSkW8GmRtf3Auamy6Rz8ZYarYRhqKtm68OFbvww3NRjkiSJ+zIH+W311zlXvogbenSZHQzHBum1uokpsQ0fyLgax2wZXTd0mbXnN9xPlLDamL7xTn7X9Qy0LMs8sXOE4VyWPz15hr88c4GUaTKazxE3Im9UlWW6kgn2dRf48pmQqXJ1zXY2Pj4P33kLVTuMEOVW+ObeSQ9dVnigu5+Hewf4/ngUrzrU0b0qPDCUyvD3jj/Cz+w5iCCKdeatGAltdfPGie4+vn3zKpO1Ki9N3OLh3kEMRcYN5tGVzrvixF4rgRVdM0FAiIsf1giFQ3B3h5UkI0kpWGabklaMohA+dvOrGOYHkeUUkmy1WKlW5wgk2UI3HsNz30SIWrS99jYEEAA+kQesron7trbCon2JeouJTJYUJBRC4eOJBqHwAImE2sOR/G+S1laIXeqlBm8/d4aHPnoMRVOIpSxGDg6ABIce3ctLX36TzoE8T/zMwz9WI/puEI8ZGLq27bKxSrnJ0aPD5PPR/bp5c74dH/1JoGuoY90VQLjFcNGjD+6MVnit096zs5t8bquTjsAOPH56cEV26qW5S+vywyyU6jz/xhXilo7j+lwam+XpB3ZvcT8r2PSNN0yNXft7kWW5nQ2M3SXAFtM2XxovQ5c19iZ3szOxo6W4GtE4ChGJLloxHd8PkGV51dLAkPU2WY0beiy5S+tuXwhB4F9AlnuQNuj8AQ/ffQtJslD1o+v+QlMUduRz/MyRgwxlMzw0PLBuyZcqywRCUHMcbC+qbLh3YbqKEftZfO9twsAhDKbv8Xva12RfoYP/5aEnSOg6jh/wPz/4KLk7HlRJkshbMfLW5qVGR7t6sFSVmudyYWGOm+UF+hNzzNe/yUj2HyKEhCBAQmWh/jUy1uM0vTFAsNT8PmHokDbvbyUQovskhADRRJKX1WRbhvoOg+n7V2nU/xBFHULXjyFhIG2gIyZJErXqvyEW/2VM69n250JUaNT+E479HVRtD7H4r6JqB9aQ3XRYB3mi559yufSXTDffwglK0URBgCJpmEqWDvMgB7O/SMFcnex97DMP8IV//hW+9UcvoBoqhd4cxz94kKMfPISVMLjvsb3o90gk/nVBkWUUffvGUjdUzpy5jb9MrVizOXFieyugv05k07F1+XS3ph8XCcN+c/Kd9iezdplQhGvm8mzK4hNPHECWojHL8IMAdYvMhHCvRJoUERz/qCBLMvpd8ct63eZbXzvNxz99jJNv3qS3L8vQyIqA3HJrL0TyJ81gI1KNALf5DXTzw8h3GF0hbMJgFlnpgVY8VZJzGxpdiDL5e7s62Nu1EucJwpCy7VBu2pRtm3cmpvHDgBuLRf707TP0Z1Ps7ijQn0mt6VsXwicMxlvxSxVV3Y9QBnGaX2r/xnX96D8vIB7TMVrL26mZMvGYTj6XYEcmxz9+6ElMVcNUA5r+FKbSTYiLRMR1ocmbz/CDqQyWpkETJmoVbpVnyRs3MdU+au4FAtEkCCuocppQBMiSQSBqNNwraHIOOxxHCK9V3rX8VLrY9jexYj99x55WP7Gy3IEkxfC985G6gKRH/60LCUXpwHVexDA/1Daqjv0CQXAL0/p4RGYe3EJRR5Gk1WEGRdLIGbt4sPMfUvMmmW68zZJ7DTeoElM76bYO02UdQVfW1pR2DOT5uX/0KRYmlxBCkO/Jrkp8De3rXzPmR4Fm4FL3HGKqjqVsLin/o4auKW3eBYBKRceyfnKJtL9uxFUD/S7F6PWaUxRZJm5FjsJSuUGj6XJ7poiiSHTlty6M+mMpEp2aWGJxscahw2sZle6G6/icOzvOkx/az6uvXOWBh0YZGllZNvvCw2/V14aItjLw3ZAkNYoj3lUREIYV7PofouqH0c2PIqtb5wi9E44f8MLVm3zzwhUmyxUmylF4wFBVbi4VuTg3z3cuXeNzx+7j2MDdZOkBYTATJc8kkyiGGSDJK3GrpWKdsdsL1OoOHYUkhqGh6yr1uk2j6bSXS4VYvNUU4hOETerhTXxRR5ZMhPDJmoc3PQ9TUUkbJlCmZDcpOxJx/QBuMEvDu04oHELhgLiNpe1CQmHZozWULjQ5gyLHUeU0shTFQYXwcewfYFof39B7VZQCmn6MMIwSTBHpjMpGtZ2G+WHqtd8nCG6hqpHX5TqvEE/8Fqq2NV4AWVJI6YNYSh89vk9C17fU3Td9Y5aJq9Mk0jEURV5TbfBuEIQhjVbb9MoaYQVVr8mt2gI5I8GOxHYZvt4bKlWbeNxok5Z7XtAuH/vvBevxUW9xJBWvyc70CmfFZHNp00QagKZGK3LL0PCD7YVztmR0G3UHSZK2LJBYLjWZnixtyejKskQ6bWEaGum0ha7fXcPYxA3vbCxYraMV+JeRpBSK2gfoLD/OYTBLGMyiqKNoxiN4zqsgPCQ5x7uZazRFpisZR1cVDFXhidERjg/0cri3m9GOPKVmk3/xvZf5vVde5x89/Ri7O1cmDkky0IyHV20vDBbxvZPtf7uuj6YpjAwVUBQZ3w9RFYmujtQ6dimk6U/hixrRWkBFCB9li1wWkdGNeGydMCCu7yLOLoKwgeNPRVLiyAh85upfQZETdCd/FlXeuBxJiCKO8xKG8eSaOHUYNgj8a4TBArqxbDAlkFaMrhAeQnitqgjQ9PuR5S/SbHyZeOK3os+FS6P+n4gl/haK0kcYLuK5Z5AkHcN8at3jCoXgZrnIXKPOiZ4+9FYt72ZP8a6jIwzt78dKmO0up/eCuu1St10uT84zUEijyDK92dVeti6r5I0kaX17tIw/Cuzc2UWt6babCwYH83R3b730LAgFt+dKaIpMNhnD/BGujreCetOlOFvGMNR2Q8/WIZFQDV6cu9j+pBms3wxWbdjIskwiZpCMm5i6SiZlkU5sry76nldHCMHFC1O88uJlPvHpYwzviEqyyqUGsiKTSBhIksTMVIm5uQr7D/YxPNqBqiltcu+NAvuSJCErMulMDNPSSCRNZGUlESaEoOiVqflRwkaRFGKqecf2BGEwRRheRFE/2yq9ioxu4N/Edb6DFf9tNOPDKOpOXPs5Qu8cuvnRbV0kiGK4Jwb7GcplqDsuCcOgIxEZYYCUafBrDxzj733p6/zx26f5R08/SmKNarDAtX+A576Mogwjybn2NersSNFRSGKaqyc2IcQ6mVi5VR6XR5K0Fk9s2GrDvDf81vZE+3+trUoGhtpDdA0FAp8UMqqcRt1EXViSNFT1ALXyP8XRv4WijiBJJkI0CIJpAv82IlxEVXeh6w+sjEOG1jLO9y5Tr/1fqOpuJDlJGMzh+zfwvHOIcAFFHUXgYze+hOe+jSSnoj5/AbHErzFWLvKtm1fpT6Z4YmCERCvXIITgcnGBK8UFdufy3K6U2JUttCee9RBLWcRapW7jM0XilkE2tcKPvNnz3LBdTENre9RCCObKNRqOS6Vp03Biq2g8l1HxmpS9Bn2xrWftf1To6EgyuE3DcSdKNZtivYGmKnRsg/jlRwVdU9B1hVKlSWc+iSxvTwvRDT0+O7CSSHtlfv1EWr3pcfbqFM88vIelcgPb9Ymb+oZKKxvhnka3UXdp1B1MS+PPv/A6v/0/fohkyuLVV64SBiEffPYghqExPV3ihy9dpqc3Qz6f4Mv/9U1+/W8/Rb6Q4NWXr/L9586jGyp79/Wyc3c3hc4kmXQMWZbIZiPWeMvSVyXRFtwlbtZv4bdagw1Zp8voxHdfB0lH1Q4iK90o6p4oztgyutHfY0gYhOECilxAlntRlGEC72rrNxurPSy/VHd+X2k6vHNjkj19HezpWn+5ub+7k8O93bx8/RYfGBnkmT071+xD048TeJeQpARusIjvl4mpSTRDXvdlliRpjTifJEnoyvaL4AF8IZhvRtI2KX21nLwkKSh3xUdVfSuxKo1Y4teRlTyu/SKed7a1vTiK0otuPIiuP4iqjq4KqUhSHKlVLibJSZBMbPtbSOio2k4Syb+HEB528yu47ttRXbN+AEXpj+p4tYOo6k5kpUBelXm4d4Cv3bjEV65d4G8ffoCMYTFVq0Qt2ZJM2bFp+D6LzcamRvdOXB9foKcjTa3hsFRpcGRPH1enFnjl4hjPHttDwtS5ODHHrt4C2XiM756+yocO78LS1agFG8EbV25TrDW5b6R3FQn+MkpugzI2441F+mM5Csb2+Qv+OtFwXLqyCaaXKutOKD92tCpY4pbO5EyJnq50W6NwK/BFyHenz7T/PW0X102kxS2dSt3m9nSR18/eQpJo5162g3sa3YnxRYQQfO6XH+EPfu/7lEsNkimL3r4sb7x6jdnpMp1dKQxDI5WyqFVt8oUkQ8MdvPbDqzzz7CEOHxtix84uLp6f4Plvn+MvvvA6g8MFfu03nqDQmSSRiFpSdV1t3zQ/9LlUucylykr9XEZLsze1G7iN57xK6I8jyUkEE+CdJgimEMIj8M5G8VO5E0nOAGA3/xzP/i6KthcRlojKjlZfMM8PUBQZ1/OpNB060wmKtQaOF1Bp2rheQHMT9iVNkXly1wjPX7nOW7cneXx0BFNTVxlxSU5ixH4a1/4esvEEi+4cFb9EKHx8ETAU2/We5IzuhZvlJSpOFK4ZTGUYSW/uWXmhjx14JLXNmeMUpZN44jeJJ34TgDBYAklCljfevml9PIrDA6o6RCb7r9b9XSz+c5seI4AbNPDCgN+47wTvzE7x8sQYhqqhyTI112XJbvDHF06zK5vnvsLmnLPLqwtVVdA0BVWRiVs6tUZ03WzXxzI03ro6Tl8+je36zJXqLFYaFGtNTt+cIm4alOpNNFXB9gIcL2hN9Gtjjxk9Rj6ZpsNIMtFY+u/O6GYSFuNLFcp1p8V78JMtpdOUSIGl3nBZKtUolhp0FrZ+DZOqtUr11xfBuom0uKVzZE8fZ69OcWhXDy+8dY1a3fnRNkdAVLt69coMt27Ok87EyOQiT2j33h5OnRzj+89fwLI0XDcg35Ekk40jSfDgw6P8lz96Ccf2cF0fx/ZoNFyOPzDCEx/cz/RUkbffuoEiy8RahBGKslKa5oQu885i5OlJKik1wSOFBxmM9SOLAkI08N2T0RJTiiPLHajaAQL/FmEwjqx0oWrHkeVOQCDCeRR1BxJ6q6Rs7UW9NhOxz7t+QBCGdKYTXBif4/rMIg3HY0f35sz4siQxmM0QCMFstU6padOtJfC9gOJchc7+HEuzZax4Eivxc7iBgx7M0gjq2GGdRMt7jRJlYlVZyo8Kr0zcouG7WKrK/d19jGY2P6e67zDZWORA5t7x+WWIsEmj/l9ACokn/u6akq5lKOrWtrneyuNueGFA0W5Sdhx6EymeHhylaDeoex47MjlCIbhSXECRJHyxtv5zOZ4ZBCHVhs3NySWO7OlrLVXBMjQujc2yf7QbTVOYLdXQZJnpYpXBjiwguDQxTyZucXs+ko5xfZ9sIoYqy6RiJrIkkU/G1k3mRTHdn/zS/EeBpKWzL9nJ7bkSS9UGhfSPT5tsPYRCMDNbodZw0HUVZRtkOSAoew32ZlaUR+bt8rqJNFmS2DnYQTphcW1igaVK454Jt/VwT6Pb15/lviODzEyX2L23B8uKYmW+H5BKt5IemkxPKsbQcIFUOvKIglDQ1ZPBNDVkWaJQSDIy2sngcAG76XLxwiQz02V0XWXP3qija3hHRzuDaioGD+dP0GN1UfPrpNQkB9L7It4FKY5uPotuPrvmeIUIcZ1vI6Hj+xeR5AyKOkAs+Q/veTFMTeXWXJF8KkZ/PsPEQhnXD9jT14EsS6Qsg0xs82VpQo+uT9PzcHwf1/ZwGg6ztxbo7M8xe3uBQm8WK2GiKwY91iBu6BCKEEM22ioKZ+Zn+O7YtVX8CT8KvDE9QRAKPjQ8yk/vPrBhU0sgQs6Xb5PVEzQDl/Ol2+SNJJ1mmqrfZLy+yM5kN6ayftmXEE4r1vzeJo561WZ2fJFCT4ZkOkYoxLotnlnTojueZKpWbU9WMU1Ha3kwElCwYuiKinkXMY7nB9ycXETXVKp1G0mSuDa+wHBvDl1VUZSoJnup3KDSsJkv1+nLpXhk3zCVho2hqeQSFnv7o6oDx/OZWCwjhKAjFScdX7tKuLuGNJJSkum2Mu/pev11QZFlevMp7G3w4P7IIKJJU5Fl0gmTmLX1OmpJkjicHaRxB5PiwcwgpqLdzfjZxtxSlZMXxkHwrkRi72l0Y3GDEw9G5BFCCDwvwHV8XvnBZXRN4dEn9vLyDy7hOB6v/fAaDz+6C9PU+N63z/HMTx1ieGQl/nnq5C2+9uWTfPpv3M8DD61mdhdhg+EdMZRW7aQiKfRY3RhuEs1USMa3HuiXMCJjKyWQ5PVn3XIjYovyg5BM3EJVZEa6cmTjFglLR5Flyg2bAwNdxHQt0jRT5E1tiACmKlGHmqVphLbHYqWI2/TQTQ3XjohT7maZ0u8mWpEkClaMsXKJ7966hr9JZ44sRaSIiiyhLmuiKQqGomAqGnFNI6nrJHWDjGHxzNBORjM5jnb1MNgSpITIW3BDj6pnc7sxT9GtMdlYZCTRhR14VL1mqw5YIRAhbhiQ0iwG42vj25JsEUv8WqvcDIIgKpVbrkwACMMqECBJKUQYVcgkUhb1ahNNU7lydgKn6bLzYD+qplIpNmjUHAI/oG9k7T4NReVQRzeHOrpbtJY+yTvayt2wiq6UMZQ0euvzuaUquVQMSYLLY3O89Pb1SNlYUbBMHVmWiJkaz79+hUbTZXSggK4q9OdTDBTSFJIxkqYehSEUhfnFKqmEiaopdGUSyJKErqpUqk0sU6NUbmKaGsn3kLT6bxWa2grD/DU0jsiyxGBfFiHYNiWlhMTuVO+64/xw/QlksCfLYE+WUAi+/eolZEnimYf30JnbWkhjW7UdgR/yjb86xauvXGFwqMAnPnOMeMJgx85OkkmLW2ML/NnnX2V+rsIHnz3IwODqzrDevkw7niWEAyJAEBD4Y0ATWUogQkEgcshKF5IkU607WKaG74ck4wZeELJYqjGzUGH3UBeapmBo6h1xJAnNeJQofBBVRPpBwLnbs+zu7UBTZYJAML5QwtQ1qk2bmu0y3BnFHtNxk1AISpUGY5NLFEsNYqZGb3eGZNxEVWUsU0eRJYLlZS9RHeZio8mfvn2auK6xt6uDYMnm8qUpwjAk8EOun7lFGAhy3ZlNr7MEDCTT/C8PPU5c02n6Hv3JFH3JFL2JFL3xJIaqror8LotNLv+5rFysSCuClIoso7boN+/m1fVCn6vVaTqMNL4I2ZXs5f7cTjQ56hoUCAIRtlqDfWRRx1SliLBcVEBKRckwWly+wTSquoswLBEGk0iShRAdSJLA8y5G3BOijqrtwbFjvPKts+iGSrPu0DWQY+LGPJVinZmW4KDr+KQyMfYfHyYIQwSw1Gxg+z5d8QSqLFNxHPwwIGuZlN2raHICSVIwlRxF5xoJrZeqN4EQvZhqhos3Zjm2r5+YpdPfmeETTxxk11Bk0BUl8prilk5HLkEQhCRjJqausaN75bnWVIXFYp2m7eI4PhNTRbo6UsQsHV1XmZ4p0bQ9ujpSzC1UyWXjq42uiIilwg24AkQoNu3vF0IQhuGG45fDJhtvgE3H36ulONr/6uNfXnIvf3ev8WKD/YfhvUNsy/uQRfTerDqOO7ikNxwfCkSwPpFjuMH4RMzggw/uxvWCqBJIiG0pMG/L6CqqzMc+dZQPf/QQmqa2+7P37IuaAXr6Mhw7MQICNF1ZM3t0dKYodERk2WHYIPQnCMIpEBKykicMK4TBBIq6o0V40pLaaTiMTS1RyCQ4d20KSZIIgxBZkrk2vsCjx3a0Ke0iQ7J6tl02Gi+cv86e3g7GF0rUHZfuTArL0FYp2UqShOt4zM5X6S6kOLS7l/mlGtW6jSLLSFJk0ExT4+3xKRZqdXRFYaJU5luXrnJ7qcSH9+zkEwf30JtKMrK3DxA0qjZTN2bJ92bJ38PoLh/HQDLNP3vimaij9i6junKuPxoIBBONJaabRfal++m1cih3JRNEWEf4t6MGFOFAMIUITSIffw6p3eXnE4oKQTBJENxGkuIRjwJhJF4YlhGy2uokUzFjOh/67HGKCzVEKMh3pTjy8E4EKwlkAUzeXKCjN8tUtUrNc1EliZJjM1WtkjFNdEVhplbjQGcHsqTRDBZbRDd1/LBBzZvACSpYShTH9oOQc9emMQ2NpUqDvSNddBdWV2voskxhkzIoISLlglK5SbPpomkKcwvRaicRN7Adj8WlOo2m20rOrb6m5cUqb37nNOYGyRjP8bAbdxPgr2BppsSb3z6NscGSul5p4G2S/J0bX+DNb51el/sAYGFqKTL8G2D6xixvfOvUBp2rgvHLUxuOBbh9eYrXv/nOmtUfRO/+lbdvbDr+5rnbvP7Nd9b1VMMw5ObZ25uOv/rOTVRdQVonTBAGIbcvTq75XJIkYqbOPSKNG2JbRleSJDQtyuhu9L1pbuzeR+UyACFC2MhqH4p0cFPjEbN05hZrWIbGYqlGRzaB7frILdmMnkLqnt0zqiKTiVsMdWQxtYjtPR0z0VWVIAjX0CFaps7eO1RVezrTdIsU5WoT1/WxTJ1QCJbqDV65eZuKbeOHIX3pFB/Zu4uPHthDV7L1orYuRyIdY+eR4W3VEEbE21svfXkvUCWFfek+dFml28yu37klyq3W3RAkA0Q9+ruc4s5KEEnSMIzHEMJHkjPIUhKBRxjMRFwJrTCDEDbLpDWKqpDriDzljdRV+4YLyKqM5ES1soamkZEsmp7HUrNJIRajEIuhyyqK3I0e1hEEKJJFVt+5huToxMFBvvXKRRDw4KEh+rZdWB8tbft7s/T3rlRpeH7A3HwFTVPYMVTYVMXg5tnb/G+//G/uuZ+ekfV5my+8eoULr15Z97ut4ORzZzn53Nl3Pf6VL7/JK19+812Pf/6PX+L5P37pXY//2u9/l6/9/nff9fgv/ou/etdj3y2ke7jfPx4d420gCMNoufwuAtY/bggisoua4+KFIXFNI6ZvrWvv/0mIniGn1eL81ws3CNqhlP+WEYYhX/2979CobE3IsdCf54O/8Gh7QrIbDl/9d9/Bd7aWuOrb3cOjn36gPb6yVOPr2zBWOw4PceIjR9rv4fzEIs/9lxe3PH7vg7s4/OSBdhhw/MoUL//X17c8/vCTB9j34K52h+D102O88Y137jFqBSd+6iijh1c4dy+8doXT3z+/5fGPfvZB+nf3bOfd3vCH/80b3ffxPt7H+/jvEBsa3f+23YH38T7ex/v4fxjuFdP9/6918vt4H+/jffyY8b6n+z7ex/t4Hz9BvG9038f7eB/v4yeI943u+3gf7+N9/ATxvtF9H+/jfbyPnyDeN7rv4328j/fxE8T7Rvd9vI/38T5+gvj/AXn0G1xowU6bAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# handling arabic text\n",
    "comments = arabic_reshaper.reshape(comments[0:10000])\n",
    "comments = get_display(comments)\n",
    "\n",
    "wordcloud = WordCloud(font_path='assets/NotoNaskhArabic-Regular.ttf',\n",
    "                      background_color=\"white\",\n",
    "                      max_words=1000,\n",
    "                      mode='RGBA').generate(comments)\n",
    "\n",
    "plt.imshow(wordcloud, interpolation='bilinear')\n",
    "plt.axis(\"off\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## splitting the data into target and feature"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature = data.Comment\n",
    "target = data.is_off\n",
    "# splitting into train and tests\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(feature, target, test_size =.2, random_state=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(11200, 47531)\n"
     ]
    }
   ],
   "source": [
    "count_vect = CountVectorizer()\n",
    "X_train_counts = count_vect.fit_transform(X_train)\n",
    "print(X_train_counts.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(11200, 47531)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tfidf_transformer = TfidfTransformer()\n",
    "X_train_tfidf = tfidf_transformer.fit_transform(X_train_counts)\n",
    "X_train_tfidf.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Logistic Regression Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score is 0.77\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.79      0.80      0.79      1406\n",
      "           1       0.79      0.78      0.79      1394\n",
      "\n",
      "    accuracy                           0.79      2800\n",
      "   macro avg       0.79      0.79      0.79      2800\n",
      "weighted avg       0.79      0.79      0.79      2800\n",
      "\n"
     ]
    }
   ],
   "source": [
    "lr_clf = LogisticRegression(C=1).fit(X_train_tfidf, Y_train)\n",
    "# make prediction and print accuracy\n",
    "lr_prediction = lr_clf.predict(count_vect.transform(X_test))\n",
    "print(f\"Accuracy score is {accuracy_score(Y_test, lr_prediction):.2f}\")\n",
    "print(classification_report(Y_test, prediction))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1])"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr_clf.predict(count_vect.transform([take_a_shower('الزمر فالطوندونس هو باطمة تلاقيتو تواتيتو')]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score is 0.79\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.79      0.80      0.79      1406\n",
      "           1       0.79      0.78      0.79      1394\n",
      "\n",
      "    accuracy                           0.79      2800\n",
      "   macro avg       0.79      0.79      0.79      2800\n",
      "weighted avg       0.79      0.79      0.79      2800\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# make pipeline\n",
    "pipe = make_pipeline(CountVectorizer(),\n",
    "                     TfidfTransformer(),\n",
    "                     LogisticRegression())\n",
    "# make param grid\n",
    "param_grid = {'logisticregression__C': [0.01, 0.1, 1, 10, 100]}\n",
    "\n",
    "# create and fit the model\n",
    "model = GridSearchCV(pipe, param_grid, cv=5, verbose=10)\n",
    "model.fit(X_train,Y_train)\n",
    "\n",
    "# make prediction and print accuracy\n",
    "prediction = model.predict(X_test)\n",
    "print(f\"Accuracy score is {accuracy_score(Y_test, prediction):.2f}\")\n",
    "print(classification_report(Y_test, prediction))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'logisticregression__C': 1}"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1])"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.predict([take_a_shower('الزمر فالطوندونس هو باطمة تلاقيتو تواتيتو')])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle.dump(model, open('./models/count_vect/lr_model.model', 'wb'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## SVM Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 48 candidates, totalling 240 fits\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=rbf .....................\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=rbf, score=0.539, total=  10.5s\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=rbf .....................\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:   10.5s remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=rbf, score=0.533, total=  10.9s\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=rbf .....................\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:   21.4s remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=rbf, score=0.547, total=  12.8s\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=rbf .....................\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:   34.2s remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=rbf, score=0.549, total=  14.3s\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=rbf .....................\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   4 out of   4 | elapsed:   48.5s remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=rbf, score=0.553, total=  12.2s\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=linear ..................\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:  1.0min remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=linear, score=0.719, total=   9.2s\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=linear ..................\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   6 out of   6 | elapsed:  1.2min remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=linear, score=0.725, total=   9.0s\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=linear ..................\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   7 out of   7 | elapsed:  1.3min remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=linear, score=0.729, total=   9.0s\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=linear ..................\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   8 out of   8 | elapsed:  1.5min remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=linear, score=0.723, total=   9.0s\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=linear ..................\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   9 out of   9 | elapsed:  1.6min remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=linear, score=0.718, total=   9.0s\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=poly ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=poly, score=0.500, total=   9.2s\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=poly ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=poly, score=0.500, total=   9.1s\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=poly ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=poly, score=0.500, total=   9.1s\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=poly ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=poly, score=0.500, total=   9.3s\n",
      "[CV] svc__C=0.1, svc__gamma=0.1, svc__kernel=poly ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=0.1, svc__kernel=poly, score=0.501, total=   9.1s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=rbf, score=0.568, total=  10.2s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=rbf, score=0.566, total=  10.3s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=rbf, score=0.574, total=  10.0s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=rbf, score=0.594, total=  10.0s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=rbf, score=0.589, total=  10.1s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=linear, score=0.719, total=   9.0s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=linear, score=0.725, total=   9.0s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=linear, score=0.729, total=   8.9s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=linear, score=0.723, total=   9.0s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=linear, score=0.718, total=  20.6s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=poly, score=0.516, total=  17.4s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=poly, score=0.513, total=  11.9s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=poly, score=0.519, total=  13.0s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=poly, score=0.523, total=  12.6s\n",
      "[CV] svc__C=0.1, svc__gamma=1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=0.1, svc__gamma=1, svc__kernel=poly, score=0.526, total=  13.0s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=rbf, score=0.515, total=  13.0s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=rbf, score=0.520, total=  13.2s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=rbf, score=0.512, total=  11.7s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=rbf, score=0.524, total=  12.4s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=rbf, score=0.524, total=  13.8s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=linear ...................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=linear, score=0.719, total=  12.5s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=linear ...................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=linear, score=0.725, total=  13.5s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=linear ...................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=linear, score=0.729, total=  11.3s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=linear ...................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=linear, score=0.723, total=  11.6s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=linear ...................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=linear, score=0.718, total=  14.5s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=poly .....................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=poly, score=0.550, total=  35.3s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=poly .....................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=poly, score=0.553, total=  32.3s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=poly .....................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=poly, score=0.556, total=  33.4s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=poly .....................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=poly, score=0.553, total=  28.4s\n",
      "[CV] svc__C=0.1, svc__gamma=10, svc__kernel=poly .....................\n",
      "[CV]  svc__C=0.1, svc__gamma=10, svc__kernel=poly, score=0.550, total=  38.7s\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=rbf, score=0.515, total=  12.8s\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=rbf, score=0.520, total=  11.9s\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=rbf, score=0.512, total=  11.4s\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=rbf, score=0.524, total=  14.4s\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=rbf, score=0.524, total=  14.3s\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=linear ..................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=linear, score=0.719, total=  12.2s\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=linear ..................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=linear, score=0.725, total=  10.7s\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=linear ..................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=linear, score=0.729, total=  11.2s\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=linear ..................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=linear, score=0.723, total=  10.8s\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=linear ..................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=linear, score=0.718, total=  12.1s\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=poly ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=poly, score=0.544, total= 1.7min\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=poly ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=poly, score=0.547, total= 3.7min\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=poly ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=poly, score=0.550, total= 5.3min\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=poly ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=poly, score=0.546, total= 3.7min\n",
      "[CV] svc__C=0.1, svc__gamma=100, svc__kernel=poly ....................\n",
      "[CV]  svc__C=0.1, svc__gamma=100, svc__kernel=poly, score=0.544, total= 1.4min\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=rbf, score=0.732, total=  11.2s\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=rbf, score=0.743, total=  11.1s\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=rbf, score=0.742, total=  11.7s\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=rbf, score=0.720, total=  10.8s\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=rbf, score=0.717, total=  12.3s\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=linear, score=0.776, total=  10.1s\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=linear, score=0.780, total=  11.0s\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=linear, score=0.793, total=  10.9s\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=linear, score=0.776, total=  10.3s\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=linear, score=0.775, total=  10.9s\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=poly, score=0.500, total=  10.3s\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=poly, score=0.500, total=  10.2s\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=poly, score=0.500, total=  11.2s\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=poly, score=0.500, total=  10.5s\n",
      "[CV] svc__C=1, svc__gamma=0.1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=1, svc__gamma=0.1, svc__kernel=poly, score=0.501, total=  10.4s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=rbf .........................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=rbf, score=0.762, total=  14.8s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=rbf .........................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=rbf, score=0.770, total=  15.9s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=rbf .........................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=rbf, score=0.783, total=  16.3s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=rbf .........................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=rbf, score=0.768, total=  15.0s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=rbf .........................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=rbf, score=0.764, total=  14.8s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=linear ......................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=linear, score=0.776, total=   9.5s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=linear ......................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=linear, score=0.780, total=  10.4s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=linear ......................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=linear, score=0.793, total=  11.7s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=linear ......................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=linear, score=0.776, total=  10.3s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=linear ......................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=linear, score=0.775, total=  10.4s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=poly ........................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=poly, score=0.697, total=  15.2s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=poly ........................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=poly, score=0.690, total=  15.9s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=poly ........................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=poly, score=0.710, total=  18.1s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=poly ........................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=poly, score=0.693, total=  15.9s\n",
      "[CV] svc__C=1, svc__gamma=1, svc__kernel=poly ........................\n",
      "[CV]  svc__C=1, svc__gamma=1, svc__kernel=poly, score=0.693, total=  15.2s\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=rbf ........................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=rbf, score=0.536, total=  18.1s\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=rbf ........................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=rbf, score=0.536, total=  17.1s\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=rbf ........................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=rbf, score=0.543, total=  16.8s\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=rbf ........................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=rbf, score=0.534, total=  17.3s\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=rbf ........................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=rbf, score=0.533, total=  18.3s\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=linear .....................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=linear, score=0.776, total=  10.6s\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=linear .....................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=linear, score=0.780, total=  10.1s\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=linear .....................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=linear, score=0.793, total=  10.2s\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=linear .....................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=linear, score=0.776, total=  10.0s\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=linear .....................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=linear, score=0.775, total=  10.1s\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=poly .......................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=poly, score=0.544, total= 1.3min\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=poly .......................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=poly, score=0.547, total= 1.4min\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=poly .......................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=poly, score=0.550, total= 1.4min\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=poly .......................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=poly, score=0.546, total= 1.4min\n",
      "[CV] svc__C=1, svc__gamma=10, svc__kernel=poly .......................\n",
      "[CV]  svc__C=1, svc__gamma=10, svc__kernel=poly, score=0.544, total= 1.4min\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=rbf, score=0.528, total=  15.5s\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=rbf, score=0.531, total=  15.6s\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=rbf, score=0.537, total=  13.7s\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=rbf, score=0.527, total=  15.9s\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=rbf, score=0.528, total=  15.6s\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=linear ....................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=linear, score=0.776, total=  10.1s\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=linear ....................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=linear, score=0.780, total=  10.2s\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=linear ....................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=linear, score=0.793, total=  10.3s\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=linear ....................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=linear, score=0.776, total=  10.1s\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=linear ....................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=linear, score=0.775, total=  10.2s\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=poly ......................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=poly, score=0.544, total= 1.4min\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=poly ......................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=poly, score=0.547, total= 1.3min\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=poly ......................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=poly, score=0.550, total= 1.4min\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=poly ......................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=poly, score=0.546, total= 1.4min\n",
      "[CV] svc__C=1, svc__gamma=100, svc__kernel=poly ......................\n",
      "[CV]  svc__C=1, svc__gamma=100, svc__kernel=poly, score=0.544, total= 1.4min\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=rbf, score=0.772, total=  19.2s\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=rbf, score=0.771, total=  19.0s\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=rbf, score=0.788, total=  20.5s\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=rbf, score=0.772, total=  19.6s\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=rbf, score=0.771, total=  18.8s\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=linear ...................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=linear, score=0.774, total=  16.1s\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=linear ...................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=linear, score=0.771, total=  17.1s\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=linear ...................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=linear, score=0.787, total=  15.8s\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=linear ...................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=linear, score=0.776, total=  17.2s\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=linear ...................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=linear, score=0.770, total=  17.0s\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=poly .....................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=poly, score=0.500, total=  10.9s\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=poly .....................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=poly, score=0.500, total=  11.0s\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=poly .....................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=poly, score=0.500, total=  10.4s\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=poly .....................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=poly, score=0.500, total=  10.8s\n",
      "[CV] svc__C=10, svc__gamma=0.1, svc__kernel=poly .....................\n",
      "[CV]  svc__C=10, svc__gamma=0.1, svc__kernel=poly, score=0.501, total=  12.7s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=rbf ........................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=rbf, score=0.763, total=  26.5s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=rbf ........................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=rbf, score=0.775, total=  24.1s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=rbf ........................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=rbf, score=0.788, total=  26.1s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=rbf ........................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=rbf, score=0.774, total=  22.9s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=rbf ........................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=rbf, score=0.771, total=  25.3s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=linear .....................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=linear, score=0.774, total=  15.2s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=linear .....................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=linear, score=0.771, total=  17.3s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=linear .....................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=linear, score=0.787, total=  15.0s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=linear .....................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=linear, score=0.776, total=  15.8s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=linear .....................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=linear, score=0.770, total=  16.9s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=poly .......................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=poly, score=0.634, total=  26.2s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=poly .......................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=poly, score=0.659, total=  21.6s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=poly .......................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=poly, score=0.659, total=  28.4s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=poly .......................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=poly, score=0.642, total=  23.6s\n",
      "[CV] svc__C=10, svc__gamma=1, svc__kernel=poly .......................\n",
      "[CV]  svc__C=10, svc__gamma=1, svc__kernel=poly, score=0.645, total=  25.9s\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=rbf, score=0.538, total=  22.6s\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=rbf, score=0.539, total=  20.6s\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=rbf, score=0.545, total=  18.7s\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=rbf, score=0.535, total=  21.4s\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=rbf, score=0.533, total=  25.6s\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=linear ....................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=linear, score=0.774, total=  14.6s\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=linear ....................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=linear, score=0.771, total=  16.3s\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=linear ....................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=linear, score=0.787, total=  14.7s\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=linear ....................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=linear, score=0.776, total=  15.0s\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=linear ....................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=linear, score=0.770, total=  17.4s\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=poly ......................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=poly, score=0.544, total= 1.4min\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=poly ......................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=poly, score=0.547, total= 1.4min\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=poly ......................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=poly, score=0.550, total= 1.3min\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=poly ......................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=poly, score=0.546, total= 1.3min\n",
      "[CV] svc__C=10, svc__gamma=10, svc__kernel=poly ......................\n",
      "[CV]  svc__C=10, svc__gamma=10, svc__kernel=poly, score=0.544, total= 1.2min\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=rbf, score=0.529, total=  19.5s\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=rbf, score=0.531, total=  20.0s\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=rbf, score=0.537, total=  13.0s\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=rbf, score=0.527, total=  21.6s\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=rbf, score=0.528, total=  19.2s\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=linear ...................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=linear, score=0.774, total=  14.4s\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=linear ...................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=linear, score=0.771, total=  15.2s\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=linear ...................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=linear, score=0.787, total=  13.9s\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=linear ...................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=linear, score=0.776, total=  15.0s\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=linear ...................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=linear, score=0.770, total=  14.7s\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=poly .....................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=poly, score=0.544, total= 1.2min\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=poly .....................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=poly, score=0.547, total= 1.2min\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=poly .....................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=poly, score=0.550, total= 1.3min\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=poly .....................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=poly, score=0.546, total= 1.2min\n",
      "[CV] svc__C=10, svc__gamma=100, svc__kernel=poly .....................\n",
      "[CV]  svc__C=10, svc__gamma=100, svc__kernel=poly, score=0.544, total= 1.2min\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=rbf, score=0.769, total=  17.4s\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=rbf, score=0.773, total=  17.7s\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=rbf, score=0.781, total=  17.7s\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=rbf, score=0.767, total=  17.6s\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=rbf, score=0.771, total=  17.5s\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=linear ..................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=linear, score=0.782, total=  13.8s\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=linear ..................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=linear, score=0.778, total=  14.4s\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=linear ..................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=linear, score=0.787, total=  14.5s\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=linear ..................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=linear, score=0.783, total=  14.5s\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=linear ..................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=linear, score=0.776, total=  15.0s\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=poly ....................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=poly, score=0.516, total=  10.0s\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=poly ....................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=poly, score=0.513, total=   9.5s\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=poly ....................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=poly, score=0.519, total=   9.3s\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=poly ....................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=poly, score=0.523, total=   9.3s\n",
      "[CV] svc__C=100, svc__gamma=0.1, svc__kernel=poly ....................\n",
      "[CV]  svc__C=100, svc__gamma=0.1, svc__kernel=poly, score=0.526, total=   9.8s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=rbf, score=0.763, total=  20.9s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=rbf, score=0.775, total=  20.8s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=rbf, score=0.789, total=  21.1s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=rbf, score=0.774, total=  20.2s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=rbf .......................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=rbf, score=0.771, total=  21.4s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=linear, score=0.782, total=  14.0s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=linear, score=0.778, total=  14.6s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=linear, score=0.787, total=  14.7s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=linear, score=0.783, total=  14.1s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=linear ....................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=linear, score=0.776, total=  14.2s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=poly, score=0.550, total=  27.3s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=poly, score=0.553, total=  24.5s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=poly, score=0.556, total=  26.2s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=poly, score=0.553, total=  22.8s\n",
      "[CV] svc__C=100, svc__gamma=1, svc__kernel=poly ......................\n",
      "[CV]  svc__C=100, svc__gamma=1, svc__kernel=poly, score=0.550, total=  31.0s\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=rbf, score=0.538, total=  19.7s\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=rbf, score=0.539, total=  18.5s\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=rbf, score=0.545, total=  17.4s\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=rbf, score=0.535, total=  18.2s\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=rbf ......................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=rbf, score=0.533, total=  23.4s\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=linear ...................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=linear, score=0.782, total=  14.0s\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=linear ...................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=linear, score=0.778, total=  14.6s\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=linear ...................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=linear, score=0.787, total=  14.9s\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=linear ...................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=linear, score=0.783, total=  14.9s\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=linear ...................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=linear, score=0.776, total=  16.2s\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=poly .....................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=poly, score=0.544, total= 1.2min\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=poly .....................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=poly, score=0.547, total= 1.2min\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=poly .....................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=poly, score=0.550, total= 1.2min\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=poly .....................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=poly, score=0.546, total= 1.2min\n",
      "[CV] svc__C=100, svc__gamma=10, svc__kernel=poly .....................\n",
      "[CV]  svc__C=100, svc__gamma=10, svc__kernel=poly, score=0.544, total= 1.2min\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=rbf, score=0.529, total=  19.6s\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=rbf, score=0.531, total=  20.1s\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=rbf, score=0.537, total=  12.2s\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=rbf, score=0.527, total=  19.7s\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=rbf .....................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=rbf, score=0.528, total=  18.7s\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=linear ..................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=linear, score=0.782, total=  13.8s\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=linear ..................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=linear, score=0.778, total=  14.4s\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=linear ..................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=linear, score=0.787, total=  14.5s\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=linear ..................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=linear, score=0.783, total=  14.0s\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=linear ..................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=linear, score=0.776, total=  14.2s\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=poly ....................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=poly, score=0.544, total= 1.2min\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=poly ....................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=poly, score=0.547, total= 1.1min\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=poly ....................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=poly, score=0.550, total= 1.2min\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=poly ....................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=poly, score=0.546, total= 1.2min\n",
      "[CV] svc__C=100, svc__gamma=100, svc__kernel=poly ....................\n",
      "[CV]  svc__C=100, svc__gamma=100, svc__kernel=poly, score=0.544, total= 1.2min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done 240 out of 240 | elapsed: 106.6min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score is 0.79\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.79      0.79      0.79      1406\n",
      "           1       0.79      0.79      0.79      1394\n",
      "\n",
      "    accuracy                           0.79      2800\n",
      "   macro avg       0.79      0.79      0.79      2800\n",
      "weighted avg       0.79      0.79      0.79      2800\n",
      "\n"
     ]
    }
   ],
   "source": [
    "pipe = make_pipeline(CountVectorizer(),\n",
    "                     TfidfTransformer(),\n",
    "                     SVC())\n",
    "param_grid = {'svc__kernel': ['rbf', 'linear', 'poly'],\n",
    "             'svc__gamma': [0.1, 1, 10, 100],\n",
    "             'svc__C': [0.1, 1, 10, 100]}\n",
    "\n",
    "svc_model = GridSearchCV(pipe, param_grid, cv=5, verbose=10)\n",
    "svc_model.fit(X_train, Y_train)\n",
    "\n",
    "prediction = svc_model.predict(X_test)\n",
    "print(f\"Accuracy score is {accuracy_score(Y_test, prediction):.2f}\")\n",
    "print(classification_report(Y_test, prediction))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'svc__C': 100, 'svc__gamma': 0.1, 'svc__kernel': 'linear'}"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc_model.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1])"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc_model.predict([take_a_shower('الزمر فالطوندونس هو باطمة تلاقيتو تواتيتو')])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle.dump(svc_model, open('./models/count_vect/svm_model.model', 'wb'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random Forest Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score is 0.81\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.78      0.87      0.82      1406\n",
      "           1       0.85      0.75      0.80      1394\n",
      "\n",
      "    accuracy                           0.81      2800\n",
      "   macro avg       0.82      0.81      0.81      2800\n",
      "weighted avg       0.82      0.81      0.81      2800\n",
      "\n"
     ]
    }
   ],
   "source": [
    "pipe = make_pipeline(CountVectorizer(),\n",
    "                     TfidfTransformer(),\n",
    "                     RandomForestClassifier())\n",
    "\n",
    "param_grid = {'randomforestclassifier__n_estimators':[10, 100, 1000],\n",
    "             'randomforestclassifier__max_features':['sqrt', 'log2']}\n",
    "\n",
    "rf_model = GridSearchCV(pipe, param_grid, cv=5)\n",
    "rf_model.fit(X_train,Y_train)\n",
    "\n",
    "prediction = rf_model.predict(X_test)\n",
    "print(f\"Accuracy score is {accuracy_score(Y_test, prediction):.2f}\")\n",
    "print(classification_report(Y_test, prediction))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'randomforestclassifier__max_features': 'sqrt',\n",
       " 'randomforestclassifier__n_estimators': 100}"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf_model.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1])"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf_model.predict([take_a_shower('الزمر فالطوندونس هو باطمة تلاقيتو تواتيتو')])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle.dump(rf_model, open('./models/count_vect/randomforest_model.model', 'wb'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Naive Bayes Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score is 0.72\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.84      0.55      0.66      1406\n",
      "           1       0.66      0.89      0.76      1394\n",
      "\n",
      "    accuracy                           0.72      2800\n",
      "   macro avg       0.75      0.72      0.71      2800\n",
      "weighted avg       0.75      0.72      0.71      2800\n",
      "\n"
     ]
    }
   ],
   "source": [
    "pipe = make_pipeline(CountVectorizer(),\n",
    "                     TfidfTransformer(),\n",
    "                     MultinomialNB())\n",
    "pipe.fit(X_train,Y_train)\n",
    "prediction = pipe.predict(X_test)\n",
    "print(f\"Accuracy score is {accuracy_score(Y_test, prediction):.2f}\")\n",
    "print(classification_report(Y_test, prediction))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe.predict([take_a_shower('الزمر فالطوندونس هو باطمة تلاقيتو تواتيتو')])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle.dump(pipe, open('./models/count_vect/naivebayes_model.model', 'wb'))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "env",
   "language": "python",
   "name": "env"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}