Explore.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/nickvazz/Desktop/virtualenvs/3MLenv/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2026: FutureWarning: From version 0.21, test_size will always complement train_size unless both are specified.\n",
      "  FutureWarning)\n",
      "/Users/nickvazz/Desktop/virtualenvs/3MLenv/lib/python3.6/site-packages/sklearn/utils/validation.py:578: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('arizonans', 4.560170771703952)\n",
      "('keep', 3.967180883085389)\n",
      "('move', 3.9371742800610283)\n",
      "('hampshire', 3.3655846525029016)\n",
      "('hispanic', 3.2695014155842164)\n",
      "('obamacare', 3.234926653154982)\n",
      "('young', 3.2127229073386214)\n",
      "('ct', 3.2016865473296137)\n",
      "('right', 3.1972895642009713)\n",
      "('mississippi', 3.15287688730857)\n",
      "----------\n",
      "('progress', -5.997967275863121)\n",
      "('ic', -4.990274660072231)\n",
      "('blue', -4.687610546466059)\n",
      "('environment', -4.401601988537372)\n",
      "('trivedi', -4.276922058781615)\n",
      "('bucket', -4.0609515926608895)\n",
      "('rural', -3.7721783529670465)\n",
      "('iowans', -3.771255146451933)\n",
      "('nurses', -3.764031663665675)\n",
      "('student', -3.7378402720274453)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.preprocessing import label_binarize\n",
    "from sklearn.model_selection import train_test_split\n",
    "import re\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "# https://www.opensecrets.org/pacs/superpacs.php?cycle=2014\n",
    "\n",
    "jointFund = !ls PoliticalFinancialData/J*\n",
    "superPacs = !ls PoliticalFinancialData/S*\n",
    "\n",
    "df_s = []\n",
    "for s in superPacs:\n",
    "    year = s.split('-')[-1].replace('.xlsx','')\n",
    "    df = pd.read_excel(s)\n",
    "    df['year'] = year\n",
    "    df_s.append(df)\n",
    "    \n",
    "df_j = []\n",
    "for j in jointFund:\n",
    "    year = j.split('-')[-1].replace('.xlsx','')\n",
    "    df = pd.read_excel(j)\n",
    "    df['year'] = year\n",
    "    df_j.append(df)\n",
    "    \n",
    "df_s = pd.concat(df_s)\n",
    "df_j = pd.concat(df_j)\n",
    "\n",
    "df_s = df_s.loc[df_s['Viewpoint'].isin(['Conservative','Liberal'])]\n",
    "df_j = df_j.loc[df_j['Party'].isin(['R','D'])]\n",
    "\n",
    "df_s = df_s.reset_index().dropna(subset=['Viewpoint'])\n",
    "df_j = df_j.reset_index().dropna(subset=['Party'])\n",
    "\n",
    "df_s.replace({'Conservative':'R', 'Liberal':'D'}, inplace=True)\n",
    "\n",
    "df_s = df_s.drop('index', axis=1)\n",
    "df_j = df_j.drop('index', axis=1)\n",
    "\n",
    "replace_punc = re.compile(\"(\\.)|(\\;)|(\\:)|(\\!)|(\\')|(\\?)|(\\,)|(\\\")|(\\()|(\\))|(\\[)|(\\])\")\n",
    "\n",
    "names = df_s['Group']\n",
    "names.append(df_j['Committee'])\n",
    "names = names.apply(lambda x: x.lower())\n",
    "names = names.apply(lambda x: replace_punc.sub(\"\", x))\n",
    "\n",
    "terms_to_replace = [\n",
    "    'conservative',\n",
    "    'conservatives',\n",
    "    'republican',\n",
    "    'republicans',\n",
    "    'progressive',\n",
    "    'democrat',\n",
    "    'democrats'\n",
    "    'democratic',\n",
    "]\n",
    "\n",
    "for term in terms_to_replace:\n",
    "    names = names.apply(lambda x: x.replace(term, ''))\n",
    "    \n",
    "cv = CountVectorizer(binary=True)\n",
    "cv.fit(names)\n",
    "\n",
    "X = cv.transform(names)\n",
    "\n",
    "stance = df_s['Viewpoint']\n",
    "stance.append(df_j['Party'])\n",
    "\n",
    "y = label_binarize(stance.values, classes=['D','R'])\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=0.66, random_state=42)\n",
    "\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.metrics import accuracy_score\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "Cs = [0.01, 0.05, 0.25, 0.5, 0.75, 1.0, 2.0, 5.0, 10.0]\n",
    "for c in Cs:\n",
    "    clf = LogisticRegression(C=c)\n",
    "#     clf = DecisionTreeClassifier()\n",
    "    clf.fit(X_train, y_train)\n",
    "    \n",
    "    \n",
    "    test_acc = accuracy_score(y_test, clf.predict(X_test))\n",
    "    train_acc = accuracy_score(y_train, clf.predict(X_train))\n",
    "    plt.scatter(c, test_acc, color='r')\n",
    "    plt.scatter(c, train_acc, color='g')\n",
    "    plt.plot((c, c), (test_acc, train_acc), linestyle='--', alpha=0.5, color='k')\n",
    "    plt.text(c, test_acc + (train_acc - test_acc) / 2, '{:.2f}'.format(train_acc - test_acc))\n",
    "plt.show()\n",
    "\n",
    "\n",
    "\n",
    "feature_to_coef = {\n",
    "    word: coef for word, coef in zip(\n",
    "        cv.get_feature_names(), clf.coef_[0]\n",
    "    )\n",
    "}\n",
    "for best_positive in sorted(\n",
    "    feature_to_coef.items(), \n",
    "    key=lambda x: x[1], \n",
    "    reverse=True)[:10]:\n",
    "    print (best_positive)\n",
    "    \n",
    "print ('-'*10)\n",
    "for best_negative in sorted(\n",
    "    feature_to_coef.items(), \n",
    "    key=lambda x: x[1])[:10]:\n",
    "    print (best_negative)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "from itertools import chain\n",
    "best = sorted(feature_to_coef.items(), key=lambda x: x[1], reverse=True)[:10]\n",
    "worst = sorted(feature_to_coef.items(), key=lambda x: x[1])[:10]\n",
    "data = chain(best, worst)\n",
    "\n",
    "# data = sorted(feature_to_coef.items(), key=lambda x: x[1])\n",
    "\n",
    "df = pd.DataFrame(list(data), columns=['words','score'])\n",
    "\n",
    "x = df.loc[:, ['score']]\n",
    "df['word_z'] = (x - x.mean())/x.std()\n",
    "df['colors'] = ['blue' if x < 0 else 'red' for x in df['word_z']]\n",
    "df.sort_values('score', inplace=True)\n",
    "df.reset_index(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAu0AAAGzCAYAAACb0il9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAMTQAADE0B0s6tTgAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzs3Xl8ZGld7/HPN6kklUolVdn3rTOsOjBzBwSuqKwuXFQQ8LI7COKFK4IjrndAlPGKyKogqMM4yCYCF8QRF5AZhlFQGUSWQaDTnX3fO52t0/27f5xTh0pN0p3q7nTlpH/v16teSZ3lOb/K75yT55x6nufIzHDOOeecc84dXmWlDsA555xzzjl3fl5pd84555xz7pDzSrtzzjnnnHOHnFfanXPOOeecO+S80u6cc84559wh55V255xzzjnnDjmvtDvnnHPOOXfIeaXdOVdykgYkmaRzkpr3WOb54TIm6fl7LNMclmGSBg426guTlAhj2S5yvXvyPmvutSVpQtJfS/oflzHG3LYeWzD9lnD6zRdR5kWvW1DORf39riRJ7z/fPnlYSEpKepWkz0takHRG0qyk+yR9WNIvSmoqdZzOub15pd05V3JmNgiMAgJ+aI/FHp/3++P2WOZxYRmjYZlx9x/Ae8PX3wAbwE8Ad0h6UykCkvSksJL6mVJs/0qS9JLws95a6lguhaR24F7grcBjgG8C/w/4HLANPBN4O/CIUsXonLuwRKkDcM650J3ACwkq5x/dZf7jgFlgk/NX2nNlHQX/z8xuyb2RVA78HvArwC9L+qSZ3X1A234b8H6Cv/mVXDdiZtuSHgIc5kd3/wpwCzBR6kDO44+BhwJfBX7czEbyZ0pqBZ7LJebLOXew/E67c+6wyFW0H184Q1I3cIzgzuDngIFwWqHcukel0r6DmZ0FfhMYDif99AFua87M/svM5q/kuruU9V9m9q1LLeegmNlkGONKqWPZjaQa4MfDt79UWGEHMLNpM3urmd17ZaNzzhXDK+3OucMiV9F+SHjnL9/jwp93EVTa86cB0d3ChxSUlT+/W9I7JB2XtCFpKWzP/XPhHezC5aOmEZKaJP2hpBOSNgubhkj6fkl/H5a5KunfJd1YxGffNzPbBv4zfNtXEIckPU/SZ8N2y5uShsLP8IBitrNbu3RJ9wCfDt8+saDN/fHzrVtQ9oMkvVvStyWtSVqW9I0wPw/NW27XNu157cjP9zpesM4Ph+V/RdJc+LcZlfQhSTfsEuMY8Gfh2xcXlP2ZvOX2bNN+MfmQNBaW1yXpiZI+LWlR0rqkey+i7XwjkNu/Z4pcNxfTIyW9L4x9I/wsX5H0xt0uniU9WtJHJE0q6IsxI+mTkp6wR/nR31DSwyT9laQpSWcL96Fw3/mz8FjMHcefk/Tci/lszsWJN49xzh0KZjYs6STQT1Ah/3De7MeFPz8HbIW/Px543y7LnDSz4bzpSHo08CmgnuAu9SeAbLjO9wNPk/Q0MzuzS2gtwJeAOuDu8Pf1vLKfTdAUpJyg+cE3gB7gNuDafXz0i1EX/tzMi0NhHM8laKf8OYLmDjcALwaeLenpZvZpLt6ngNPADwOTwD/mzZveTwGSXkhQGa4kyMWnCG4gHQNeDkwB912gmLsJPuNuHknQFORswfQ/BdqBrwP3hPMfAjwbeIakZ5nZX+ct/1fAo4D/DnwH+Je8ed+4QHyXIx8vBW4maIv+9wTHxaOA90nKmtk7LhRDaIagL0QSeIWkl5nZuX2ui6TfAH6XoK/It4C/BlLANQRNg74afs7c8i8D3kGQ03sJLqD7CO72/7ikm83sd/fY3A8A7wHGCS7Qa4DVvLKfDdwOVAH/BfwtwTH9KOAHJT3OzF6638/mXOyYmb/85S9/HYoXwT9sA95dMH2QoPKh8P0kcKJgmXeF676nYHo1QSdXI6hMJPLmXUNQcTTgtwvWe0k43YB/AGp3ibeToFJhwCsK5v0wQWXJgO0i/w73hOvdfIFtviZv+i+E02aAh+VNLyNoc23APNC4x7YeWzD9lt1iAJ4UTv/MeeLfa93vA84QVJhfnstn3vx+4L/lvU8U8/cj6Ei5SlBJ/h8F854OZHdZ5xlhTNNAco994NbzbPP94TLPL5h+sfkYC+dtAj+2RzwLQFUR+9M78vblEwSdTp9HcNGi86z3U+E6a8Azd5n/PcCD8t5fF/4tzwHPLVj2qQQX3AY8YY+/oYV/m7JdtnVd+DdZA36yYF4fwcWYFW7XX/46Si9vHuOcO0zu165dUg/BXdi7zSzXIfFzQL+k3rx192rP/j+BLoKK+00WNC8BwMyOA78avn2lpMpdYtoCXmpmp3aZ9xKCu4H/bGZ/lD/DzP4RuGyjjkhKS/oh4A6+ewfyPXmL/HL487fM7Kt5cZwDXktwd7iB4C5vqbyGoCL+djP747x8AmBmJ83syxdTsKR+vvu3ebmZ/W1B2R83s6XC9czsY8DHCb5R+cGL2fYeLjUfbzOzvyuI9VbgOMHd5f9WRCw3AX9IUKHuB36RoKJ8HzCtoOlXxy7r/Xb489fN7H6dw83sG7azv8GrCPL7ETP7YMGyd/Dd/fXVe8T5TeC1tvs3ATcTfDvzG7bzGxHMbAj4ufDtL+5RtnOx55V259xhkqtwP1DBMHWws2kMBb8/DkBSG/CggjLIXwb4kJltcX9/BZwCMsD1u8z/khU0t9ml7PfvMf+9e0zfr9fn2lETxHgXwR3HKYJRQCYAJPXx3fbt99tmWAm6PXx7v46+V4KkCuCJ4ds/vcxlNwB/B7QCv2tmu5YfthN/qaS3hO3Kb5d0O/DgcJEH7bbeRcTTx6Xn42/2mP7N8GfnfuMxsy0zeyXQC/xv4EMETV0MaAZeAXxVUrT/S+oEvpfgW5Hb9rmpx4U/b99jfq7S/oOSdqt/fHy3CruCPic/Er79qz3K/leCZmuPCPc1544cb9PunDs0zGxc0neABxBUZj7Izk6oOfmV9vfmLfMdMxsvKDZXuTm5xzYtbEv/MHavCA2dJ+Su85V9nun79R8EbYYhuEs6T9Cm/g4z28hbLhf3tJmt7VHWYMGyV1ozQVMlA759uQqVlAQ+SVDhfp+Z7dX59bcJRt453/+9uvPMK8blyMf9RnkJ5UapSRYblJlNEgz/+McQXew+j+DOfyNBZfvh4eK5b7HGzGyV/TnvscZ3P3MNQZ+ShYL5Q3us1wKkw98ngu4C59XAPvtYOBcnXml3zh02d3L/Svs8QZtVAMzsPkmzfPcu5UEO9bh+4UUOzI5x2t1OYWfP9xF0Jv4MezQ1kfTTBBXTFYJmK3cSjKu+EV60vZGgU+UFa4NX0L47i14sM5sC3ixphOAO9sMk9ZvZpV5sXqy9jrXcXXkD/mIf5ez2jZpzseeVdufcYXMnwcgZjw/bs/cTfG1e+ICduwlG/ejj/A9Vyt15P7bbxsKKX3/Bsvs1TtCZtW+P+XtNv9xycbdKqjGz07ssc6xg2Sttlu+OYvJAgtE/LtWbCZ7m+VXgGbb76D/w3fHsfyNsF16oqOEw9yEO+ciXPwpQE8Gd8tyd/k5J6X3ebR8nuEN/jN3zm/vMp4H79S84jxmCTqhVwKt265vg3NXA27Q75w6bu8KfA8DzC6blyzWReR5BJXCv5XLTnrNHR9NnArXAMkFzlGLkx7CbFxZZ3kUJO+INhW9/pnB+eGFyY/j2Ur+NyN3FLOqmT1ih/qfw7c+db9n9kPRK4JcIRlt5ip3/4UYN4c/79U0Im4g8sXB66GI/6xBXLh/npX20JSEYojRnHMDMxgg6yyaAF+1zc3eFP2/cY/7Phj/v3qOz6a4K9p0De6CYc4edV9qdc4dK+JV9rrNdbgSOu3ZZNFdhvin8+c1w3UIfJqjYdRE0BYgepCRpAPiD8O3b9+ioej5/RjAE3Q9Ienn+DElP5DJUTovw5vDnb0mKxocPK22vI+hUuMjOEWcuxlj484GSiv229haCjo2vkvTzhRVKSX2SLjgqiqSfAt5CcKH1Y7v0YyiU259+Pv/CTVKWoLlF7R7r5T7rQ/eYfz5XKh8X0ijpSwoe8lRTOFPSNXkx3J3r3Bx6XfjzDZKetsu63yMpv/Pu2wjy+0xJzylY9scIRlsCeNNFfI7XEfTreHP4EKb71V8kXbtbnM4dFd48xjl3GN1JMIZ0A0Fnta/tsszXwnkNeevcj5mtS3oWwegivwA8VdIXCUaLeTxBc41PEVQoi2JmY5JeSlDxe2f4+zeAbuCxBJWYXyq23Iv0TuAxBA/zuVdS/sN8HkhwcfFsM5u/lI2Y2QlJXyEYxeZrkr5E0HRhxsx+8wLrfjH8G707fP16uH45QdOJhxG0Pb/QsI9vJbjpNAS8eo+bydNm9mt5yz+P4AE/g2H+qwiGeFwF/pzd7yb/C8FIPY+UdC9Bv4ozwH1m9pYLxHhF8rFPNxCMcLQZ5m6EoP1+D8HDqETQJGbH38DMPirptwgqzB+X9E2Cpki5hys9BHgBwUg0mNlXJL2CYFz4D0q6iaDTcR/BQ6ogGLv/s8V+ADP79/DBXLcR9GP4v5LuA+YIOtFeS9AR9gMED09z7sjxO+3OucMovwKePz57JJz2+T3WKVz2iwSjYvwxQWe2nyKoVH8Z+HngJ87THvq8zOwDBG3qP01Q8XwawegYL+W7Y8AfuPDv8fzw9S8EDxp6JsGILbcB14djx18OTwP+kuDC5zkEHUD31WzBzG4jGGM8N4zgTwBPILiJ9A7gfuOB7yL3bcnDCZqf7PZ6Rt42jxMM5/khgvz/OMEFwgfCWPLvLufHugH8KMHwiz0Ef9sXA0/Zx+e8kvk4n3ng0cD/Ifh2qgH4MeAnCSrTdxFcWH6vmZ0oXNnMfofgWPkwweg6P0VQAd8E3kDBt2Bm9q5w+Y8RXLw+i+Ai5Q7gSbb301AvyMz+kuCBTm8j6FT8AwR5fjDBxcGvEVz0OXckaZf/hc4555xzzrlDxO+0O+ecc845d8h5pd0555xzzrlDzivtzjnnnHPOHXJeaXfOOeecc+6Q80q7c84555xzh5yP035EVVZWWktLS6nDcEUwM/b38EJ3WHjO4sXzFT+es/jxnBVnfHx8y8yq9rOsV9qPqJaWFsbGxi68oHPOOeecKwlJs/td1pvHHFFnz54tdQiuSCMjI6UOwRXJcxYvnq/48ZzFj+fs4Hil/Yjyh2bFz9bWVqlDcEXynMWL5yt+PGfx4zk7OF5pd84555xz7pDzSvsRVVbmqY2bxsbGUofgiuQ5ixfPV/x4zuLHc3ZwvGZ3RHmlPX7S6XSpQ3BF8pzFi+crfjxn8eM5Ozheszuitre3Sx2CK9Lw8HCpQ3BF8pzFi+crfjxn8eM5OzheaXfOOeecc+6Q80q7c84555xzh5x8aMCjqaury/zhSvGyublJVdW+HormDgnPWbx4vuLHcxY/nrPiSBo3s679LOt32o8ovxiLn42NjVKH4IrkOYsXz1f8eM7ix3N2cK66Srukb0j6mVLHcdD8iajxMzu77ycZu0PCcxYvnq/48ZzFj+fs4CRKHcCVZmbfU+oYnHPOXXmV990Hn/98qcNwRaidnobW1lKH4YoQ+5xdfz1cd12po9jVVVNpl1RpZlfNs3UllToEV6RE4qo5HI8Mz1m8JCQ4c6bUYbgilJ875zmLmdjn7BC3VIh18xhJ/1vS1yWtSJqS9D5JTeG810m6R9LrJU0AXwmnD0l6Sfj7myWt5r3WJZmkvnD+QyT9naQ5SWOS/kRSJm/7d0l6u6QPSlqWNCrpZXnz2yXdIWla0ilJX5X0rILPYJJeIemfwxi+JumxefMfL+lLYfnz4XL1F/rblJeXX9Lf1l15fX19pQ7BFclzFi8dHR2lDsEVqampqdQhuCJ5zg5OrCvtwBTwU0AWeBTwQOCP8uY/GjgDHAMeUbiymf2ymaXNLB2WcSfwD8CYpFrgM8B9QA/wSODBwHsLirkRuBWoB14FvEPSNeG8cuA2YABoAN4OfFBSYROdlwA/E8bwT8AH8ua9H3hnOK8deDVwwW8Mzp07d6FF3CEzOTlZ6hBckTxn8TI7N1fqEFyRlpaXSx2CK5Ln7ODE+rtdM/tY3tthSW8gqEDnTAOvtwsMpaKgLcltQCPwBDPblvRUoBL4NTPbBtYkvQr4sqQ2M5sKV/+YmX0297ukBeAG4LiZjQH54y6+R9IvAk8AvpE3/c1mdjyM5c+AV0pqNbNpggr6ANBhZuPAF/b4DDcBN+Xep9Npjh8/DkAymaSrq4uxsbGoV3dDQwOZTIaTJ09GZXR1dbG1tcXMzAwQ3K3v7+9namqK1dVVAGpra2ltbeXEiRPRhUFrayuJRILx8fGorP7+fpaWllhcXNwRw+joKJubm1EMdXV1DA0NRet1d3ezubkZxZBIJOjr62NycpLTp08DUFdXR0tLC4ODg9EoOW1tbZSXl++I4dixYywsLLC0tARAdXU1nZ2djIyMsLUVXPc0NjaSTqd3PMGtu7ubjY2NqDPNfmNob29HEhMTE1FZAwMDzM3NsRyexFKpFB0dHQwPD3Mm/PqwqamJVCrF8PBwVH5PTw9ra2vMhZWMiooKent7mZiYYG1tDYBMJkNTUxODg4PR9jo6OjCzqDIpiYGBAWZmZlhZWQGgpqaG9vZ2hoaGoifnNjc3k0wmGR0djcrq7e1ldXWV+fl5ACorK+np6WF8fJz19XUAstksDQ0NnDhxIlqvs7OTs2fPMjU1VVQMLS0tVFVV7Yihr6+PlZUVFhYWAKiqqqK7u3vHvlxfX082m92xL3d2drK9vc309DQAZWVlHDt2jOnpaU6dOgUEx0hbWxsnT56MOm63tLRQWVlJ/nCp/f39LC8vRzHkH09jY2OcPn3aj6dDeDyNjIxE6+WOp6nJSWxmhvLychobG1laWopir66uJp1O7+hEl8lmwSzaHhItzc2srKxE+19lVRXZTIa5ubkoh+naWioSiShfAA2NjWxubESfuTyRoLGhgcXFxSj26lSKmlQqOu4hOMbOnTsXHTv7jaG2tpZEQQyNjY1s5MWQSCRoKIghlUqRukAMkmguiKGqqorMPmJoampifX09iqGiooL6+vodMdTU1FBdXc3c3Byrq6tsbW5SX1/P9vZ2dPyWlZXR1NTE8vJydAwkk0nq6uqYnZ2N9qO6ujrKysqi/TYXw9raWnQuzcWwsLAQnY9qampIJpPR+Q/YM4al5WW2CmKYmZ2FC8Rwem2N9YIY5hcWOJsXQ1UyyUJBDGe2t1ktMoZMJgMSy3kxNDc3s7q6Gp3PKysryWazzM/PR+fEdDpNZWVldP6D4FyztbUVnccKj6fV1VXKy8piezxpbo5muGL/n4oR63HaJf0UwZ3na4AkwTcHNQQXI68BftjM/nvBOkPALWZ2a960NwFPBb7fzObDab8K/LSZPSJvuVpgBXiUmf2bpLuAe8zs5t3KD5uxvBF4EsEFwTkgDfyemb0mXN6AJ5vZZ8L3fcBJ4AFmdlzStcBvEFT0Vwnuwr8+vJDYU1tbm+UqTS4ejh8/zjXXXHPhBd2h4TmLl9FPfILu8ALGxcPMzAwtLS2lDsMVIfY5u+GG4HWFXBXjtEvqAj5C0Bymx8zqgBfkZoc/L9hGRNKvAM8mqODP580aBXok5X8bMRD+HGF/3kDQpOaHgIyZZQnusO+7l6iZfc3MnmtmbcAzgf8FvOhC65WVxTa1V626urpSh+CK5DmLl5qamlKH4IqUTCZLHYIrkufs4MS5ZpcmiH/OzDYkPYDgjvS+SXoh8OvAj5pZYUX8b4Ft4P9KqpbUBrwV+Ju8pjEXkgHWgHmgQtIrgH0POSmpUtKLJDWHk5aBs2Fc5+WV9viJ9Z2Jq5TnLF4aGhpKHYIrkl8Yx4/n7ODEtk27mf2XpN8A/kJSGvgaQafNRxZRzM8CdcAXC4ZIfKiZjUh6MvAWgnbpG8CngF8tovybCdrKTxM0q7kN+Oci1ofg7vrvS6ohqPzfzv07w95Prh2Wi4/BwUEGBgYuvKA7NDxn8TI6MUF3RUWpw3BFiH1Ti6tQ7HN2iEffi3Wbdrc3b9MeP94+On48Z/Hi+Yofz1n8eM6Kc1W0aXfOOeecc+5q4ZX2I8ofrhQ/7e3tpQ7BFclzFi+er/jxnMWP5+zgeKXduUOioF+FiwHPWbx4vuLHcxY/nrOD45X2Iyr3YAQXH/kPkXHx4DmLF89X/HjO4sdzdnC80u6cc84559wh55V255xzzjnnDjkf8vESSLoLuMfMbi51LIW6urpsbGys1GG4IpiZtwWMGc9ZvHi+4sdzFj+es+IUM+RjbB+u5M7v3LlzpQ7BFWlubo7m5uYLL+gODc9ZvCzeeScNw8OlDsMV4dSpU9TV1pY6DFeES87Z9dfDddddvoCOEK+0H1FeaY+f5eVlrwDGjOcsXk6vrNBw5kypw3BF2FhZoS6ZLHUYrgiXnDMfSGNP3qb9MpL0GkknJH2vpKykd0kaljQv6VOSjuUtWy7plyV9U9KypHslPTFv/o2SxiS9Mvw5L+k2SenSfDrnnHPOOVcqXmm/DCQlJX0AeArwaOAbwMeBOuB6oAP4GnCHpIpwtdcALwCeBtQDtwCflDSQV3QbcB3wIOBhwLXAW/cZ0yV+KnelpVKpUofgiuQ5i5ek37GNncrKylKH4IrkOTs43hH1EoQdUb8DfC8wDNxoZhuS/hvwRaDBzFbDZcuBVeDJZnaPpGXgp83sH/LK+zTwOTO7RdKNwHvCMpbD+T8G/DVQbWY7vj+SdBNwU+59XV1d57333gsE/6i6uroYGxtjY2MDgIaGBjKZDCdPnozK6OrqYmtri5mZGSB4qmp/fz9TU1Osrq4CUFtbS2trKydOnIia4LS2tpJIJBgfH4/K6u/vZ2lpicXFxR0xjI6Osrm5GcVQV1fH0NBQtF53dzebm5tRDIlEgr6+PiYnJzl9+nTus9HS0sLg4CC5/betrY3y8vIdMRw7doyFhQWWlpYAqK6uprOzk5GREba2tgBobGwknU4znNfOtbu7m42NDWZnZ4uKob29HUk7xqgdGBhgbm6O5eVlIKjkdXR0MDw8zJnwa/qmpiZSqRQjIyPRej09PaytrTE3NwdARUUFvb29TExMsLa2BkAmk6GpqYnBwcFovY6ODsyMycnJ3H7BwMAAMzMzrKysAFBTU0N7eztDQ0Nsb28D0NzcTDKZZHR0NCqrt7eX1dVV5ufngeBE3NPTw/j4OOvr6wBks1kaGho4ceJEtF5nZydnz55lamqqqBhaWlqoqqraEUNfXx8rKyssLCwAUFVVRXd39459ub6+nmw2u2Nf7uzsZHt7m+npaQDKyso4duwY09PTnDp1CoB0Ok1bWxsnT56MnmvQ0tJCZWUl+Z24+/v7WV5ejmLw4ym+x9Opu+6i6utfp7y8nMbGRpaWlqLYq6urSafTUZwAmWwWzKLtIdHS3MzKykqU+8qqKrKZDHNzc1EO07W1VCQSUb4AGhob2dzYiD5zeSJBY0MDi4uLUezVqRQ1qVR03ENwjJ07dy46dvYbQ21tLYmCGBobG9nIiyGRSNBQEEMqlSJ1gRgk0VwQQ1VVFZl9xNDU1MT6+noUQ0VFBfX19TtiqKmpobq6ekcM9fX1bG9vR8dvWVkZTU1NLC8vR8dAMpmkrq6O2dnZaD+qq6ujrKws2m9zMaytrUXn0lwMCwsL0fmopqaGZDIZnf/OF8PS8jJbBTHMzM7CBWI4vbbGekEM8wsLnM2LoSqZZKEghjPb26wWGUMmkwGJ5bwYmpubWV1djc7nlZWVZLNZ5ufno3NiOp2msrIyOv9BcK7Z2tqKzmOX/Xh6whMYamy8av4/PeABD9h3R1SvtF+CsNL+ICANXGdmg+H0/wl8CFgpWKUK+Fngs8AUcArIb3xeAfyFmb0srLS/ycya8rb3YOCbQKeZnffpBe3t7ZaruLl4GB4epre3t9RhuCJ4zuJl8o47aPcHv8TK/Pw8jY2NpQ7DFeGSc3bDDcHrKuGjx1xZ7wE2gc9L+hEz+xpBhXwLaDaz+/V6klQFbABPNbO7z1N2vaRM7k470AecAaYvFJRfjMXPGe8gFzues3jJ3TVz8eFP944fz9nB8Tbtl4GZvR74HeBOSd8P3AN8HXiXpBYASfWSniEpZWabwLuBN0p6iALVkn5Q0gPziwbeLKlGUgfw28D7CpvGOOecc865o83vtF8mZvZuSfPAHcDzgCcTVLL/VVITsAjcDXwqXOXVwC8AHwG6Ce68fzmcnjNF0IH120A18EngVfuJp7y8/BI/kbvSmpqaLryQO1Q8Z/GSbWyEvPbV7vBL19dDRcWFF3SHxiXnzOsve/I27YdU2Kb9lv22cyrkT0SNn62tLe91HzOes3jxfMWP5yx+PGfFKaZNuzePOaK87Wb85I924eLBcxYvnq/48ZzFj+fs4Hil3TnnnHPOuUPOK+2HlJndfrFNY5xzzjnn3NHibdqPKG/THj/eDjB+PGfx4vmKH89Z/HjOiuNt2p2P0x5DuafzufjwnMWL5yt+PGfx4zk7OF5pP6L84Qbxk//IbhcPnrN48XzFj+csfjxnB8cr7c4555xzzh1yXmk/oiSVOgRXpAp/gEjseM7ixfMVP56z+PGcHRzviHpEeUdU55wr8JWvwH/8R6mjcM7t5frr4brrSh3FFVVMR9TEQQfjSsPbtMfPxMQEHR0dpQ7DFcFzFi+zU1M0nzlT6jBcEZaWlshms6UOwxXhknLmdZfz8uYx+yTpLkl/JOnDklYkjUl6tqRrJX1B0ilJ/ybpQeHySUm/L+mkpEVJn5f0qHBeRtKapB8o2MYfSvpk3vsXSvpPScuSviHp2fuN179BiR/vcR8/nrN42djYKHUIrkhbW1ulDsEVyXN2cLzSXpwXAO8EssAfAO8B3gA8B2gERoE/DJf9A+ApwJOBVuATwGckdZnZMvBR4MW5giUlgecDt4bvbwRuCZepB34e+FNJjz3QT+icc8455w4dbx5TnI+Z2d0Akm4H3ga838yGwmkfBG6VVEZQ2X6OmR0P132zpBcSVMzfQFA5/ztJv2hmK8AzgA3gb8PlbwJ+18y+FL6/R9KHgRuBewoDk3RTuA4AtbW1HD8ebDocXvI4AAAgAElEQVSZTNLV1cXY2Fh0p6mhoYFMJsPJkyejMrq6utja2mJmZgaA8vJy+vv7mZqaYnV1NSq3tbWVEydOcO7cOQBaW1tJJBKMj49HZfX397O0tMTi4uKOGEZHR9nc3IxiqKurY2hoKFqvu7ubzc3NKIZEIkFfXx+Tk5OcPn0agLq6OlpaWhgcHIy+UWhra6O8vHxHDMeOHWNhYYGlpSUAqqur6ezsZGRkJLoT0NjYSDqdZnh4eEcMGxsbzM7OFhVDe3s7kpiYmIjKGhgYYG5ujuXlZQBSqRQdHR0MDw9zJvyavqmpiVQqxenTp6Oc9fT0sLa2Fg2dVVFRQW9vLxMTE9Hd3UwmQ1NTE4ODg9H2Ojo6MDMmJydz+wUDAwPMzMywsrICQE1NDe3t7QwNDbG9vQ1Ac3MzyWSS0dHRqKze3l5WV1eZn58HoLKykp6eHsbHx1lfXwcgm83S0NDAiRMnovU6Ozs5e/YsU1NTRcXQ0tJCVVXVjhj6+vpYWVlhYWEBgKqqKrq7u3fsy/X19WSz2R37cmdnJ9vb20xPTwNQVlbGsWPHmJ6e5tSpUwCk02na2to4efJk1JyspaWFyspK8vuD9Pf3s7y8HMWQfzzlcubH0+E7nkZGRqL1csfT+vo6MzMzlJeX09jYyNLSUhR7dXU16XQ6ihMgk82CWbQ9JFqam1lZWYn2v8qqKrKZDHNzc1EO07W1VCQSUb4AGhob2dzYiD5zeSJBY0MDi4uLUezVqRQ1qdSOIfOy2Sznzp2Ljp39xlBbW0uiIIbGxkY28mJIJBI0FMSQSqVIXSAGSTQXxFBVVUVmHzE0NTWxvr4exVBRUUF9ff2OGGpqaqiurmZubi7af+vr69ne3o6O37KyMpqamlheXo6OgWQySV1dHbOzs9F+VFdXR1lZWbTf5mJYW1uLzqW5GBYWFqLzUU1NDclkMjr/AXvGsLS8zFZBDDOzs3CBGE6vrbFeEMP8wgJn82KoSiZZKIjhzPY2q0XGkMlkQGI5L4bm5mZWV1ej83llZSXZbJb5+fnonJhOp6msrIzOfxCca7a2tqLzWOHxtLm5yalTpy7qeNLcHM1wVf1/KoZ3RN0nSXcB95jZzeH7BHAGeLyZ3RVO+1HgDqADmAauNbOv55XxMWDazF4evv8W8BYz+xNJdwJfMLPfDOetA2eB7bwwEsDdZvaUC8XrHVHjx8x81J+Y8ZzFi33pS+jLXy51GK4IfozFzyXl7IYbgtdVxJ+IWnpzBHfNBwqmDwAjee/fA7xE0jXAD4bvc6aAl5tZNu+V3k+FHYiuDl185N8xd/HgOYsXv5ERP/l3al08eM4OjlfaD4CZnQNuA35H0jFJlZJ+CbgG+EDeou8FHg68FficmeXXAN4GvEbSIyWVSaoKf7+6LkGdc84555y3aT9ArwZ+B7gTyABfB55sZlGDKDOblnQH8HTgefkrm9nbJc0B7ya4Q78dlvHaKxO+c84dLVZWBv7gl1ixRMJzFjOXlLPy8ssbzBHjbdqPqM7OTsvvROYOv7W1NVKpVKnDcEXwnMWL5yt+PGfx4zkrjrdpdy6G/AI6fjxn8eL5ih/PWfx4zg6OV9qPKH8iavzkhml08eE5ixfPV/x4zuLHc3ZwvNLunHPOOefcIeeVducOCR+LOH48Z/Hi+Yofz1n8eM4OjndEPaL84UrOOeecc4ebd0R10WOkXXzkHjXv4sNzFi+er/jxnMWP5+zgHPpKu6S7JN1S6jj2S9LrJN1zgWVWJT3uIOPwSnv8rKyslDoEVyTPWbx4vuLHcxY/nrOD4w9XKgEzS5c6Buecu9pU3ncffP7zpQ7DFaF2ehpaW0sdxtXl+uvhuutKHYXbhVfaDxlJ5YCZ2SXdKi8rO/RforgCNTU1pQ7BFclzFi/VVVVw5kypw3BFqCov95xdaZc4ZLSfFw9OXGp2tZI+KGlZ0qiklwFIapd0h6RpSackfVXSs/JXlGSSXinpC5JOS/pPSddKepakb0lakfRRSem8dYYk/bakz4ZNWb4u6Ufy5j9c0uckLUlalHSvpAcVbPe3JE1KWpD0J5ISBTE9Kfz9ceH7Z0v6NrAGtEgql/TLkr4Zfu57JT1xv38wr7THT3t7e6lDcEXynMVLc1NTqUNwRcpmMqUOwRXJz4sHJy41uxuBW4F64FXAOyRdA5QDtwEDQAPwduCDkr5nl/WfG67/LeATwFOARwAPAK4HXlGwzsuB1wBZ4K3AJyX1h/P+GPgnoAloBl4MLOWt+yjgNNALPBp4FvCCC3zG/wk8BqgDZsNtvwB4Whj3LWEMAxcoB/CHK8XR0NBQqUNwRfKcxcvExESpQ3BFmpubK3UIrkh+Xjw4cWke8zEz+2zud0kLwA1m9mEgf1zD90j6ReAJwDfypr/FzE4CSPoAQSX6sWZ2Cjgl6VPA9xVs8y/M7J/zyn0p8Hzg9cAW0AP0mtkg8JWCdUfN7E3h79+W9E9h+X9+ns/462Y2n3sj6ZeAnzazb4WTPi7pX4DnEFTgd5B0E3BT7n06neb48eMAJJNJurq6GBsbY2NjA4CGhgYymQwnT56Myujq6mJrayvq+V1eXk5/fz9TU1Osrq4CUFtbS2trKydOnIg6u7a2tpJIJBgfH4/K6u/vZ2lpicXFxR0xjI6Osrm5GcVQV1e34wDv7u5mc3MziiGRSNDX18fk5CSnT58GoK6ujpaWFgYHB6PHJbe1tVFeXr4jhmPHjrGwsMDSUnA9VV1dTWdnJyMjI2xtbQHQ2NhIOp1meHh4RwwbGxvMzs4WFUN7ezuSdlQMBgYGmJubY3l5GYBUKkVHRwfDw8OcCb/ybWpqIpVKMTU1xfb2NgA9PT2sra1F/7AqKiro7e1lYmKCtbU1ADKZDE1NTQwODkbb6+jowMyiJ9JJYmBggJmZmahzUE1NDe3t7QwNDUXba25uJplMMjo6GpXV29vL6uoq8/PBbllZWUlPTw/j4+Osr68DkM1maWho4MSJE9F6nZ2dnD17lqmpqaJiaGlpoaqqakcMfX19rKyssLCwAEBVVRXd3d079uX6+nqy2eyOfbmzs5Pt7W2mp6eB4JunY8eOMT09zalTp4DgGGlra+PkyZPRRW5LSwuVlZXkD5fa39/P8vJyFEP+8ZTLmR9Ph+94GhkZidbLHU/z8/MkZmYoLy+nsbGRpaWlKPbq6mrS6XQUJ0AmmwWzaHtItDQ3s7KyEu1/lVVVZDMZ5ubmohyma2upSCSifAE0NDayubERfebyRILGhgYWFxej2KtTKWpSqR0V1Ww2y7lz577buW+fMdTW1pIoiKGxsZGNvBgSiQQNBTGkUilSF4hBEs0FMVRVVZHZRwxNTU2sr69HMVRUVFBfX78jhpqaGqqrq5mbm2N1dZVz585RX1/P9vZ2dPyWlZXR1NTE8vJydAwkk0nq6uqYnZ2N9qO6ujrKysqi/TYXw9raWnQuzcWwsLAQnY9qampIJpPR+Q/YM4al5WW2CmKYmZ2FC8Rwem2N9YIY5hcWOJsXQ1UyyUJBDGe2t1ktMoZMJgMSy3kxNDc3s7q6Gp3PKysryWazwbEZ1h/Odzzt9f9pbm6Ompoa///E/v4/FePQj9Mu6S7gHjO7OW/aEEHF9WPAG4EnAY3AOSAN/J6ZvSZc1oAnm9lnwvdPAj5tZsor7w3AI8zsSXnlv8nM3pG3zF8CK2b2Ukm9wM3ADxPc7f8ocLOZrUp6HfAkM3ts3rq3Awkze35hTOEoMncCVWa2Fc5vBaaAU+FnyqkguJh42YX+bm1tbZbbKV08HD9+nGuuuabUYbgieM7iZfQTn6Dbh6OLlZmZGVpaWkodxtXlhhuC10Xy82JxihmnPS532vfyBuDBwA8R3N02Sf8JXI7HcfXt8v5TAGY2DPwcQNhM568JmsP8n0vYXn7lfAnYAJ5qZndfTGHl5eWXEIorhebm5lKH4IrkOYuXbH09eKU9VtK1taUOwRXJz4sHJy5t2veSIei4OQ9USHoFUNie/WK9UNJjJCUk3UjQ7v0DAJJulNSl4Fm9K8B2+LoszGwTeDfwRkkPUaBa0g9KeuB+yvDHCMdPMpksdQiuSJ6zeKmqrCx1CK5IFYm431u8+vh58eDE/Wi4maAj6jRB5fk24J/Pu8b+vRv4XYK26MPA08P26wCPB/4vwUXDCsGd9t+/TNvNeTXwC8BHgG6CO+9fDqdfUK4dlouP0dFR/0oxZjxn8TI1O0tPRUWpw3BFWFhYoNXHab+yLvGbej8vHpxD36a9FHJt5s3s1lLHcrG8TXv8eDvA+PGcxYvnK348Z/HjOStOMW3a4948xjnnnHPOuSPPK+1HVMLbAcZOb29vqUNwRfKcxYvnK348Z/HjOTs4XmnfhZn1xblpDBCNk+viIzd2t4sPz1m8eL7ix3MWP56zg+OV9iPKK+3xk/8QDxcPnrN48XzFj+csfjxnB8cr7c4555xzzh1yXmk/onyc9vip9DGkY8dzFi+er/jxnMWP5+zgHPlKu6S7JN2yx7wfkLQq6VA8PlTS8yR963KU5U9EjZ+enp5Sh+CK5DmLF89X/HjO4sdzdnCu6iFGzOzzQLrUceSY2QcIn7p6qc6ePXs5inFX0Pj4OJ2dnaUOwxXBcxYvM//4j7SMj5c6DFeExcVF6uvrSx3G0XX99XDddZe1SD8vHpyrutJ+lPlDs+JnfX291CG4InnO4mVzbQ3OnCl1GK4IZ9bWIH1o7q0dPQdwg8/PiwfnyDePCdVK+qCkZUmjkl4GIOlxkkxSInz/eElfCpebl/TPkurDebdL+itJt0pakjQi6VdzG5CUlPQRSeOSTkn6lqT/nR+EpCFJr5X0qXCZQUlPz5t/o6SxvPcJSTdJui9cfkzSrx30H8s555xzzh0uV0ul/UbgVqAeeBXwDkm7PWP3/cA7gSzQDrwa2Mqb/3Tg34Fm4KeBX5f0vHCegE8B3wNkgJuAt0j6kYJtvAS4OVzmncB7JdXtEffrgJeH8WeAhwF37+PzUlZ2taT26Mhms6UOwRXJcxYv6draUofgilSdSpU6BFckPy8enKuleczHzOyzud8lLQA3ANMFy20BA0CHmY0DXyiY/1Uz+5Pw9y9K+jPgZ4EPmNk68Od5y/6tpL8Hfhj4h7zpt5rZlwEk/QnwZuAhwL/mb0jB8C+vBF5kZv8WTl7YJabc8jcRXCgAUFdXx/HjxwFIJpN0dXUxNjbGxsYGAA0NDWQyGU6ePBmV0dXVxdbWFjMzM0DQmbW/v5+pqanoYQm1tbW0trZy4sSJaCz41tZWEokE43ltRfv7+1laWmJxcXFHDKOjo2xubkYx1NXVMTQ0FK3X3d3N5uZmFEMikaCvr4/JyUlOnz4dfbaWlhYGBwejZkBtbW2Ul5fviOHYsWMsLCywtLQEQHV1NZ2dnYyMjLC1FVyLNTY2kk6nGR4e3hHDxsYGs7OzRcXQ3t6OJCYmJqKyBgYGmJubY3l5GYBUKkVHRwfDw8OcCb+mb2pqIpVKsbi4GMXa09PD2toac3NzAFRUVNDb28vExARra2sAZDIZmpqaGBwcjLbX0dGBmTE5OZnbLxgYGGBmZoaVlRUAampqaG9vZ2hoiO3tbQCam5tJJpOMjo5GZfX29rK6uhqNuVtZWUlPTw/j4+PR15/ZbJaGhgZOnDgRrdfZ2cnZs2eZmpoqKoaWlhaqqqp2xNDX18fKygoLCwsAVFVV0d3dvWNfrq+vJ5vN7tiXOzs72d7eZno6OMTLyso4duwY09PTnDp1CoB0Ok1bWxsnT56M+oC0tLRQWVnJ2Fj0hRf9/f0sLy9HMeQfT+vr6ywtLfnxdAiPp5GRkWi93PF0amWFMzMzlJeX09jYyNLSUhR7dXU16XQ6ihMgk82CWbQ9JFqam1lZWYn2v8qqKrKZDHNzc1EO07W1VCQSUb4AGhob2dzYiD5zeSJBY0MDi4uLUezVqRQ1qVR03ENwjJ07dy46dvYbQ21tLYmCGBobG9nIiyGRSNBQEEMqlSJ1gRgk0VwQQ1VVFZl9xNDU1MT6+noUQ0VFBfX19TtiqKmpobq6mrm5OcyM9bU16uvr2d7ejo7fsrIympqaWF5ejo6BZDJJXV0ds7Oz0X5UV1dHWVlZtN/mYlhbW4vOpbkYFhYWovNRTU0NyWRyx5jje8WwtLzMVkEMM7OzcIEYTq+tsV4Qw/zCAmfzYqhKJlkoiOHM9jarRcaQyWRAYjkvhubmZpYWF5kP6wrFHk97/X8yM8zM/z+xv/9PxdBRb/ss6S7gHjO7OW/aEHALcBy4E6gws21J1wK/ATwBWCXoFPr6cN7tQNrMnplXzv8CbjKzB0qqAn4X+AmgFTAgBXzYzF6Qv938p61KMuDJZvYZSTeG87skNQMzwMPM7GvFfu62tjbL7ZQuHo4fP8411+z2BZA7rDxn8TL6iU/QHV7AuHiYmZmhpaWl1GEcXTfcELwuIz8vFkfSuJl17WfZq+VO+76ElePnAki6juAO+RjwZ+EifQWr9IXzIbjL/ePh6ztmdk7SXxM0m7kYcwQXDg8Eiq60O+ecc865o8MbPockVUp6UXiHG2AZOAts5y32cEkvCTuIfh/wc3y3SUwG2ARmgTJJzyJoGnNRLPgK5I+A35P0CAUaJD3mYst0zjnnnHPx5Hfad3om8PuSaoB54HbgvXnzPw48GngTcCr8+f5w3puAa4FhYB34f8AnLjGe1wJLBM10OsPf/5A92rXn84crxY+Paxs/nrN4aW5rg7z21e7wyzY3Q0VFqcM4ug6gruDnxYNz5Nu0Xy5hm/aEmT2/1LHsR2dnp437Q0RiZXV1lbSPRxwrnrN48XzFj+csfjxnxSmmTbs3jzmi/Imo8eMdh+PHcxYvnq/48ZzFj+fs4Hil3TnnnHPOuUPO27Tvk5ndWOoY3NEWDM3v4sRzFi+er/jxnMWP5+zgeJv2I6qrq8vyHxDjnHPOOecOF2/T7qIn0rn4mPGHvsSO5yxePF/x4zmLH8/ZwfFK+xHllfb4iR5T7mLDcxYvnq/48ZzFj+fs4Hil3TnnnHPOuUMuVpV2Sa+TdE+p44iDsrJYpdYBNTU1pQ7BFclzFi+er/jxnMWP5+zg+OgxR5RX2uOnvb291CG4InnO4qV9ehr+/u9LHYYrgh9hB+T66+G66w6kaD8vHhyvtB9CkirNbOtSyvCHK8XP0NAQfX19pQ7DFcFzFi8To6N0nDlT6jBcEebm5mhqaip1GEfPAdYR/Lx4cA7d7VhJWUl/KmlM0pykv5P0oIJlfk/SjKQpSX8gqSJv3p9KGpK0KumkpN+WVJY3/y5JfyTpw5JWwu08W9K1kr4g6ZSkf8vfpqRnSbpX0mIY0ycl9RfE9D8kfTFcZl7SR4uM6R2S/lLSIvCH4fQHS7pD0rSkcUl/LGlf3zv5UJ7xs729XeoQXJE8Z/HiNzPixwdViB8/Lx6cQ1dpB94HPAB4BNADfAv4jKR0OP9RwDbQDTwOeAbwq3nr/1u4TC3wHOAXgJ8r2MYLgHcCWeAPgPcAbwiXbwRGCSvOoVPAzwJNwIMBAR/MzZT0ZOCjwJuBVqATeFeRMb0I+Itw+zdJagI+D/xT+Hd4OPBA4G33+4s555xzzrkj7VA9XElSOzABXGdm/xlOqwAmCSq6DwZeBnSY2dlw/suAV5vZwB5lvh3oMrNnhO/vAgbN7MXh+wywBDzXzD4UTnsGcKuZ1e9R5vXAl4E6Mzsl6Q5gyMx+YZ+fc7eYJs3sOXnL3AQ8y8wekzft+4HPAqnc5y9Y/qbc+9ra2s4vf/nLACSTSbq6uhgbG2NjYwOAhoYGMpkMJ0+ejMro6upia2srGmO1vLyc/v5+pqamWF1dzZVLa2srJ06ciO6AtLa2kkgkGB8fj8rq7+9naWmJxcXFHTGMjo6yubkZxVBXV8fQ0FC0Xnd3N5ubm1EMiUSCvr4+JicnOX36NAB1dXW0tLQwODgYfaPQ1tZGeXn5jhiOHTvGwsICS0tLAFRXV9PZ2cnIyAhbW0Hro8bGRtLpNMPDwzti2NjYYHZ2tqgY2tvbkcTExERU1sDAAHNzcywvLwOQSqXo6OhgeHiYM+HX9E1NTaRSKb797W+TTCYB6OnpYW1tjbm5OQAqKiro7e1lYmKCtbU1ADKZDE1NTQwODkbb6+jowMyYnJzM7RcMDAwwMzMTDcNVU1NDe3s7Q0ND0R2R5uZmkskko6OjUVm9vb2srq4yPz8PQGVlJT09PYyPj7O+vg5ANpuloaGBEydOROt1dnZy9uxZpqamioqhpaWFqqqqHTH09fWxsrLCwsICAFVVVXR3d+/Yl+vr68lmszv25c7OTra3t5mengaCPh7Hjh1jenqaU6dOAZBOp2lra+PkyZPRHdiWlhYqKyvJfzBZf38/y8vLUQz5x9PS0hLJZNKPp0N4PI2MjETrRcfTP/wDdd/5DuXl5TQ2NrK0tBTFXl1dTTqdjuIEyGSzYBZtD4mW5mZWVlai/a+yqopsJsPc3FyUw3RtLRWJRJQvgIbGRjY3NqLPXJ5I0NjQwOLiYhR7dSpFTSoVHfcQHGPnzp377jB6+4yhtraWREEMjY2NbOTFkEgkaCiIIZVKkbpADJJoLoihqqqKzD5iaGpqYn19PYqhoqKC+vr6HTHU1NRQXV3N3NwcZ86ciZbZ3t6Ojt+ysjKamppYXl6OjoFkMkldXR2zs7PRflRXV0dZWVm03+ZiWFtbi86lufIXFhai81FNTQ3JZDI6/wF7xrC0vMxWQQwzs7NwgRhOr62xXhDD/MICZ/NiqEomWSiI4cz2NqtFxpDJZEBiOYxh83u/l66f/MlLO572+P+0sbFBa2ur/39if/+fHvCAB+z74UqHrdL+fcC/ArVmtpo3/V7gw0AKeKqZPSJv3o8CnzSzSgXPzv1N4LlAB8Ed8STw72b2A+HydwH3mNnN4fsEcAZ4vJndlVfmHWaWCN//EPBa4KFArnlKLXCNmQ1K+gbwbjP7o10+035j+hcz+8289d4FvBhYyy8OqAIGzGyc8/AnosbP5uYmVVVVpQ7DFcFzFi9bX/gClV/7WqnDcEXIVdrdZXbDDcHrAPh5sThxfiJq7hImumseVqp7gNxlXrek8rx1+oBc7fTZwKuAFwJNZpYF/oSgsntRJFUCdwB/DzzQzOqAH8rNDn8OETRd2c1+YypsuDdFcHGRzXtlzCx5oQo7eJuyOMq/gnfx4DmLl9ydLRcf+XfpXTz4efHgHKpKu5lNAp8C3iypVVI18PvAFvC34WINwGslVYWdRX8F+PNwXoagvfsMYJIeDzz/EsOqBKqBxbApTAdwS8EybwdeLOkZkiolJSU98RJj+nPgekkvl5RSoFvS0y7x8zjnnHPOuZg5jEM+vgB4I0Gb8STw78CTwgozBM1nKgnurp8D3k/QiRTgduCxwNfCeZ8O51/0d0BmtirpJcDrJL0NGATeCjwlb5l/lPQc4GbgNoLmNp8l6ER6UTGZ2YikxwC/B7yGoGnQOPAh4BMX+3mcc+5qZWVl4E0tYsUSCc/ZQSgvv/Ay7tA5VG3a3eXjbdrjZ3t7m0TiMF5Hu714zuLF8xU/nrP48ZwVJ85t2t1l4mPbxk80OoSLDc9ZvHi+4sdzFj+es4PjlfYjyivt8ZMbNsrFh+csXjxf8eM5ix/P2cHxSrtzzjnnnHOHnFfaj6iw066LER/XNn48Z/Hi+Yofz1n8eM4OjndEPaK8I6pzzjnn3OHmHVFd9Fh2Fx9+kRU/nrN48XzFj+csfjxnB8cr7UeUf4MSPxsbG6UOwRXJcxYvnq/48ZzFj+fs4PhAms45564KlffdB5//fKnDcEWonZ6G1tZSh3G0XH89XHddqaNwF8Er7UdUWZl/iRI39fX1pQ7BFclzFi91NTUwNVXqMFwRaior4cyZUodxtBxw81k/Lx4cr9ntQtKLJQ0qbwgWSVWS5iQ9TVKnpL+SNB2+PiypI2/ZuyTdUlDmkKSXhL/3STJJPyPpPyWdkvRFSQ/NW75W0u2S5iWNSXpl+PPG/XwGr7THTzabLXUIrkies3ipra0tdQiuSKlUqtQhuCL5efHgeM1udx8CGoEn5U17JrAB/A1wB3AWeCDwIEDAJyWVF7mdFwA/CjQDY8C78ua9HXgocG24nWuBfX9HuL29XWQortROnjxZ6hBckTxn8TIxMVHqEFyR5ubmSh2CK5KfFw+ON4/ZhZmtSXov8FLg0+HknwduBb4PeDjwBDNbBpD0MmAWeCTwxSI29TtmNhmW8efAR8Lfy4HnAU83s4lw2k3Ai/YqKJx/U+59Op3m+PHjACSTSbq6uhgbG4s6iDQ0NJDJZHYcXF1dXWxtbTEzMwNAeXk5/f39TE1Nsbq6CgR3qlpbWzlx4kT01NXW1lYSiQTj4+NRWf39/SwtLbG4uLgjhtHRUTY3N6MY6urqGBoaitbr7u5mc3MziiGRSNDX18fk5CSnT58GoK6ujpaWFgYHB6MOt21tbZSXl++I4dixYywsLLC0tARAdXU1nZ2djIyMsLW1BUBjYyPpdJrh4eEdMWxsbDA7O1tUDO3t7UjaUTEYGBhgbm6O5eVlILhr1NHRwfDwMGfCr3ybmppIpVI7/jn19PSwtrYWTauoqKC3t5eJiQnW1tYAyGQyNDU1MTg4GK3X0dGBmTE5OZnbLxgYGGBmZiZ6tHRNTQ3t7e0MDQ1FF3fNzc0kk0lGR0ejsnp7e1ldXWV+fh6AyspKenp6GB8fZ319HQjuqDQ0NHDixIlovc7OTs6ePctU2AxhvzG0tLTw/9m78zjH8rre/69PVSpJJakklUpS+97sqCAConB/gIC4gXhBVBAGFUHUnzg/r151WNSRHyKoICAoKsoielU2Aa+IjDLCBWackcVhqb26tlQqlaRSqaVT9bl/JDmTynR3JTOdSU7zeT4e9ehOzpJ38jnLNyffc47P5zuTYWpqilxJKrUAACAASURBVHw+79xhz+fzMT4+fmZZ7u/vJxqNnlmWR0dHKZVKbG1tAeVfnmZmZtja2mJvbw8oryNDQ0MsLi46V1tKJpN4vd4zVz+Ynp4ml8s5GWrXp2p9bH3qvPVpZWXFma66PmWzWXypFN3d3QwMDJDNZp3svb29hEIhJydAJBoFVef1ECGZSJDP553lz+vzEY1ESKfTTg1DfX30eDxOvQBiAwMcHR4677nb42EgFmN3d9fJ3hsIEKzbFkSjUU5PT+++NXyDGfr6+vDUZRgYGOCwJoPH4yFWlyEQCNxje1SfQURI1GXw+XxEGsgQj8c5ODhwMvT09NDf338mQzAYpLe3l3Q67awr/f39lEolZ/3t6uoiHo+Ty+WcdcDv9xMOh9ne3naWo3A4TFdXl7PcVjMUi0VnW1rNkMlknO1RMBjE7/c727+rZcjmchzXZUhtb8M5GfaLRQ7qMuxkMpzUZPD5/WTqMlwqlSg0mSESiYAIuWyWo9VVjiKR+7w+XWn/lE6nbf/UxP6pGXad9isQkQcD/wlMADHgi8AU8B3AW1U1Xjf+DvAzqvo3InILcKuq3lQzfAm4WVXfISJTwCLwAFWdqwx/IvBJoIfyUf5N4OGq+uWaeaSBX1LVd56Xf2hoSDet76arzM3NceHChXbHME2wmrnL6gc+wHjlC4xxh1QqRTKZbHeM68ujHlX+axHbLjbHrtN+DajqV4B/p3x0+yXAR1T1IrAK9IuIc6aFiMSAfqD6VXQPCNYM9wDNbHXSwDEwWTOPcOU1GtLd3WxPHdNuo6Oj7Y5gmmQ1c5dEItHuCKZJ1j/afWy72DrWaL+6t1LuIvMC4O2V5z4HfAl4s4iERSQCvAW4E/h8ZZzbgGeIyIiI9AKvpXwEvSGqegK8F3iliAyLSAD4XeD0Grwn06HsPAT3sZq5i910zn2qXW2Me9h2sXWsT/vVfQB4E5AH/hHKDWoR+X7g94G5ynj/Cjyj0timMuxhwF1ADngNsEZzfgF4M/Bl4AB4HZChfDLsuWzn5D5bW1t2dQuXsZq5y042S7Cn4eMnpgPkMhn8to5dWy3+Jd62i61jjfarUNWSiKwCH1LV05rnVylfTeZK0+WBH657+m01w5coX3Gmdppbap+rzOMF1ceV7ji/DyxjjDGmaccPfShYX1tX2ZubY9BqZgxgjfarEpHvBR4OfF8bXnsCmAZuBaKUj7p/nbu74Jw3fevCmZawa+u7j9XMXaxe7mM1cx+rWevY1WOuoHKEvRf4RVV9Vxte/wLw95Qb7peAzwIvV9WvNjL92NiY1l62zhhjjDHGdBa7esw1oKrjqhpvR4O98vpzqvrNqtqnqjFV/Z5GG+xgJ++40b25ZqtpL6uZu1i93Mdq5j5Ws9axRvt1yhrt7lO9YYdxD6uZu1i93Mdq5j5Ws9axRrsxxhhjjDEdzhrt1yk7EcR9QqFQuyOYJlnN3MXq5T5WM/exmrWOnYh6nbITUY0xxhhjOlszJ6LaJR+vU3ZzJfdZXFxkenq63TFME6xm7rL2kY8wmkq1O4Zpwvb2NolEot0xri+PfCQ84hEtm71tF1vHGu0uICI3ADc3+k0MwH5BcR/7ouU+VjN3Ob10CS5dancM0wQ9PraaXWst3m7ZdrF1rOOzMcYYY4wxHc4a7R1ERHpF5GYR+ZqI7InIgoi8CngbMCIihcrf886bV3d3d+sDm2sqmUy2O4JpktXMXfr7+9sdwTSpr6+v3RFMk2y72DrWPaaz/AnwQOCZwFeAYWAIWOac7jEiciNwY/VxX18fc3NzAPj9fsbGxrh48SKHh4cAxGIxIpEIi4uLzjzGxsY4Pj4mVenz2d3dzfT0NJubmxQKBWe+g4ODLCwsONeCHxwcxOPxsLa25sxrenqabDbL7u7umQyrq6scHR05GcLhMEtLS8504+PjHB0dORk8Hg9TU1NsbGywv78PQDgcJplMMj8/73QDGhoaoru7+0yGmZkZMpkM2WwWgN7eXkZHR1lZWeH4+BiAgYEBQqEQy8vLZzIcHh6yvb3dVIbh4WFEhPX1dWdes7OzpNNpcrkcAIFAgJGREZaXl7lU+ck3Ho8TCARYX1933vfExATFYpF0Og1AT08Pk5OTrK+vUywWAYhEIsTjcebn553XGxkZQVXZ2NioLhfMzs6SSqXI5/MABINBhoeHWVpaolQqAZBIJPD7/ayurjrzmpycpFAosLOzA4DX62ViYoK1tTUODg4AiEajxGIxFhYWnOlGR0c5OTlhc3OzqQzJZBKfz3cmw9TUFPl8nkwmA4DP52N8fPzMstzf3080Gj2zLI+OjlIqlZybfHR1dTEzM8PW1pZzDeFQKMTQ0BCLi4vOz7nJZBKv10vtSdzT09PkcjknQ+36VCgUSKVStj514Pq0srLiTFddn3bTaYqpFN3d3QwMDJDNZp3svb29hEIhJydAJBoFVef1ECGZSJDP553lz+vzEY1ESKfTTg1DfX30eDxOvQBiAwMcHR4677nb42EgFmN3d9fJ3hsIEAwEnPUeyuvY6emps+40mqGvrw9PXYaBgQEOazJ4PB5idRkCgQCBczKICIm6DD6fj0gDGeLxOAcHB06Gnp4e+vv7z2QIBoP09vaSTqc5OT1lb2+P/v5+SqWSs/52dXURj8fJ5XLOOuD3+wmHw2xvbzvLUTgcpqury1luqxmKxaKzLa1myGQyzvYoGAzi9/ud7R9wxQzZXI7jugyp7W04J8N+schBXYadTIaTmgw+v59MXYZLpRKFJjNEIhEQIZfNcrS6ylEkcp/Xpyvtn0qlEkdHR7Z/orH9UzPs6jEdQkTiwDbwaFW9rW7YDTTZp31oaEirC6Vxh7m5OS5cuNDuGKYJVjN3Wf3ABxi3E1FdJZVK2ZHba+1Rjyr/tYhtF5vTzNVjrHtM56ieav3VtqYwxhhjjDEdxxrtnWOp8u8DLzPs9H7MYYwxxhhjOoz1ae8QqrotIn8FvEVEXqiqXxWRYcr92jeBuIgMqOrO1edU5vFYad3GrmvrPlYzdxkZH4ea/tWm88WHh8Hu8H1ttfhCFbZdbB1r2XWWFwOvBD4mIkkgBbwK+Cvgg8DXRKQbeJmqvvdqM6qeCGTcI5fLEYvF2h3DNMFq5i656WliLezLa669bCZj65jL2HaxdezrawdR1X1V/RVVnVHVUOXfd6lqSVWfq6oDqho9r8EO1mh3o+oZ6MY9rGbuYvVyH6uZ+1jNWsca7cYYY4wxxnQ4a7Rfp0Sk3RFMk/x+f7sjmCZZzdzF6uU+VjP3sZq1jl2n/To1NjamtTeIMcYYY4wxncWu026cOzwa97AvWe5jNXMXq5f7WM3cx2rWOtZov07ZLyjuU73tsXEPq5m7WL3cx2rmPlaz1rFGuzHGGGOMMR3OGu3XmIhMiEhBRGaamGZJRH7qWubosptRuI5d19Z9rGbuYvVyH6uZ+1jNWsdurnSNqeoKELqW8xSRG4CbGz1RAazR7kaRSKTdEUyTrGbuEllchA9+sN0xTBOip6d2R9Rr6ZGPhEc8oqUvYdvF1rFG+zUmIl5VPW53jlKp1O4IpkmLi4tcuHCh3TFME6xm7rK+usr4pUvtjmGakE6lSCaT7Y5x/bgfLlJh28XWsa+v95GI3CIibxaR94nILvAhEVERuVAZLiLyP0VkRUSyIvIOEfkbEXln3axGROSjIrInIvMi8qzK9E8A3lYZXqj8Pe9+fZPGGGOMMaatrNF+bbwI+EtgALixbtiPA/8DeA4QBz4DPOsy8/gp4CYgArwF+AsRCavqp4CXAuuqGqr8vac1b8MYY4wxxnQi6x5zbXxIVT8KICLFumEvAP5UVT9befynIvIzl5nHO1T1PyrzeDvwBuAhwGcvM+49iMiN1Hxh6OvrY25uDijfnWxsbIyLFy86l2KKxWJEIhEWFxedeYyNjXF8fEwqlQKgu7ub6elpNjc3KRQKznwHBwdZWFjg9PQUgMHBQTweD2tra868pqenyWaz7O7unsmwurrK0dGRkyEcDrO0tORMNz4+ztHRkZPB4/EwNTXFxsYG+/v7AITDYZLJJPPz886lLYeGhuju7j6TYWZmhkwmQzabBaC3t5fR0VFWVlY4Pi73YBoYGCAUCrG8vHwmw+HhIdvb201lGB4eRkRYX1935jU7O0s6nSaXywEQCAQYGRlheXmZS5Wf6ePxOIFAgFKp5NRsYmKCYrFIOp0GoKenh8nJSdbX1ykWy4tYJBIhHo8zPz/vvN7IyAiqysbGRnW5YHZ2llQqRT6fByAYDDI8PMzS0pLTjSqRSOD3+1ldXXXmNTk5SaFQYGdnBwCv18vExARra2scHBwAEI1GicViLCwsONONjo5ycnLC5uZmUxmSySQ+n+9MhqmpKfL5PJlMBgCfz8f4+PiZZbm/v59oNHpmWR4dHaVUKrG1tQWUz/GYmZlha2uLvb09AEKhEENDQywuLjr3NUgmk3i93jPXGZ6eniaXyzkZatenas1sfeq89WllZcWZrro+lU5OSKVSdHd3MzAwQDabdbL39vYSCoWcnACRaBRUnddDhGQiQT6fd5Y/r89HNBIhnU47NQz19dHj8Tj1AogNDHB0eOi8526Ph4FYjN3dXSd7byBAMBBw1nsor2Onp6fOutNohr6+Pjx1GQYGBjisyeDxeIjVZQgEAgTOySAiJOoy+Hw+Ig1kiMfjHBwcOBl6enro7+8/kyEYDNLb20s6nebk9JRUKkV/fz+lUslZf7u6uojH4+RyOWcd8Pv9hMNhtre3neUoHA7T1dXlLLfVDMVi0dmWVjNkMhlnexQMBvH7/c72D7hihmwux3FdhtT2NpyTYb9Y5KAuw04mw0lNBp/fT6Yuw6VSiUKTGSKRCIiQWl3lqNLn/L6uT1faP5VKJba3t23/RGP7p2bYHVHvIxG5Bfi0qv5a5fEUsAg8QFXnROQu4M2q+paaaf4WKKjqDZXHS5RPNH1HzTgKPFVV//nenIg6OjqqtTtc0/ny+TzhcLjdMUwTrGbuUvjXfyX01a+2O4ZpwsHBAb29ve2Ocf141KPKfy1k28Xm2B1R73+nVxm2BkzWPVf/+L7M/7LsjqjuUz0aatzDauYutUd8jTtUj2ob97DtYutYo7313gX8hIg8WkQ8IvIioNnrLW0CcREZuPbxjDHGGGNMp7M+7a33l8AY8PdAH/B3wD8Azdzn91+ADwJfE5Fu4GWq+t6rTSAi9y6taZvu7u52RzBNspq5S1dPD/T0tDuGaYJ4vVaza+l+2GbZdrF1rE97G4jIncBfq+r/36rXGBsb09qT6YwxxhhjTGexPu0dRkSeKyK9IuIXkV8EHgr8r1a+ZvXsfeMe1bPZjXtYzdzF6uU+VjP3sZq1jjXa7x8vptwvfRt4PvBMVZ1r5Qtao919qpcBNO5hNXMXq5f7WM3cx2rWOtan/X6gqk9pdwZjjDHGGONedqT9OtXVZaV1m76+vnZHME2ymrmL1ct9rGbuYzVrHTsR9TplJ6IaY4wxxnQ2OxHV2M2VXKj2VsvGHaxm7mL1ch+rmftYzVqnoUa7iLjiIqki8jYRecd9nMeEiBREZKbdWS4zzy+LyAsbGdd+QXEfO3nYfaxm7mL1ch+rmftYzVqn0RNR0yLyNeDzwG2Vf7+sqh1VGVV96TWYxwoQ6oQsl5nnw671PI0x5huF97/+Cz71qXbHME3o29qCwcF2x3C/Rz4SHtHszdhNp2m00R4HHgH8APCHwC4Qrdwk6DZVfXmL8pl7ye5I5j6DtmNyHauZuwxEo2DXkHaVSCAAly61O4b73Y9dZm272DoNdY9R1Uuq+nngO4EHVDrMPwzIA7OtCicit4jIH4rIX4tIXkQuisiPiMg3ichnRGRPRD4nIg+qjP9OEXl35f8iIr9ZmWav8u9rKsO8IvJWEdmsDFsSkZ+vDJsSERWRC5XH3yIi/yoiWRHZFZHba17vSSJym4jkRGRHRP5dRPrrs1QeL4nIb4jIv1S633xJRL67ZvgNlYy/UPl3R0T+TERCdfP4qVZ93qa9PB67AqvbWM3cxQ5muI9dCc19bLvYOs2uDUlVXQdQ1WXgh4DoNU911o8Db6m8zu8Cfwq8FvhRYABYBd50memeAvwE8B2q2gd8M/DhyrAXAo8DHl4Z9u3Av1/h9d8KfILyrw0J4CeBbGXYu2uyDQO/BBxf5b28DHhFZfzfBz4kItM1w4co/6LxoEreb6qM1zQ7EdV91tbW2h3BNMlq5i7b29vtjmCalM1mzx/JdBTbLrZOs1+H/lNEXqKqb688vgQkr3Gmen+nqv8G5aPXwB8A71bVpcpz7wUud8LnMeAHHiYi26qaAT5TMywEPFREPqOqm5TvWHo5x8AEMKmq88CddcNmgRFVXauZ/5X8papWvxz8qYj8NOU7pP5W5TkBXq6q+8C+iLwS+KCIvFRVr9oKF5EbgRurj0OhEHNz5Zuu+v1+xsbGuHjxIoeHhwDEYjEikQiLi4vOPMbGxjg+PiaVSgHlo1LT09Nsbm46dzjr6+tjcHCQhYUF52STwcFBPB7PmRV1enqabDbL7u7umQyrq6scHR05GcLhMEtLS8504+PjHB0dORk8Hg9TU1NsbGywv78PQDgcJplMMj8/75xwOzQ0RHd395kMMzMzZDIZZ6Pf29vL6OgoKysrHB+Xv1sNDAwQCoVYXl4+k+Hw8NDZwTeaYXh4GBFhfX3dmdfs7CzpdJpcLgdAIBBgZGSE5eVlLlV+8o3H4wQCAdLptDPdxMQExWLRea6np4fJyUnW19cpFosARCIR4vE48/PzznQjIyOoKhsbG9XlgtnZWVKpFPl8HoBgMMjw8DBLS0uUSiUAEokEfr+f1dVVZ16Tk5MUCgV2dnYA8Hq9TExMsLa2xsHBAQDRaJRYLHbmagGjo6OcnJw4t7JuNEMymcTn853JMDU1RT6fJ5PJAODz+RgfHz+zLPf39xONRs8sy6Ojo5RKJba2toDy0bqZmRm2trbY29sDyuvI0NAQi4uLzpfcZDKJ1+ul9nKp09PT5HI5J0Pt+lStj61Pnbc+raysONNV16dsNosvlaK7u5uBgQGy2ayTvbe3l1AodKZhH4lGQdV5PURIJhLk83ln+fP6fEQjEdLptFPDUF8fPR6PUy+A2MAAR4eHznvu9ngYiMXY3d11svcGAgTrtgXRaJTT01Nn3Wk0Q19fH566DAMDAxzWZPB4PMTqMgQCgXtsj+oziAiJugw+n49IAxni8TgHBwdOhp6eHvr7+89kCAaD9Pb2kk6nnXWlv7+fUqnkrL9dXV3E43FyuZyzDvj9fsLhMNvb285yFA6H6erqOtP4j8fjFItFZ1tazZDJZJztUTAYxO/3O9u/q2XI5nIc12VIbW/DORn2i0UO6jLsZDKc1GTw+f1k6jJcKpUoNJmh7+AALRav6fp0pf1TOp22/VMT+6dmNHWd9krXj7+n3OD9HOWjwTlV/f6mX7mx17sFuFVVb6o89lD+ovAkVb2l8tzTgX9QVU+lUe9R1edXhv0E5aPqj6Tc2P4tVf24iHQDvwg8B3gw8GngJlW9XUSmgEXK3YDmRGQSuAl4GtAN/G1l3IKIfBPwq8CTgQLwnsprlC6TZQl4vaq+ueb9vQ/Iq+pPi8gNleHxmuEPBu4CRlV1vTKPm1X13KvSDA0N6ab13XSVubk5Lly40O4YpglWM3dZ/cAHGK98gTHukEqlSCZbfWzwG8CjHlX+ux/YdrE5rbxO+7GqPgn4H8A88MfAf29yHvcbVf0zVf1/KHdr+QDwYRHpU9UTVX29qj4WGKXcMP7gFeaxrKovVtVJ4InAUyk31FHVL6rqj6nqEPBs4KXAi64Saeoyj2vvgNQvIpG64ZeApr+OWZ8y95menj5/JNNRrGbuMjIy0u4IpknxePz8kUxHse1i6zTbaL8NQFVvUdXfU9V3qepRC3LdZyLyGBH5byLSS7kbyx6gwImIPFlEvk1EvMAh5aPkl+1+UjlBdExEhPKJtyWgVDmZ9UUikqiMmqvMo3SVWC8QkceJiKdyZP2RlI/OVynwBhEJisgI8BvAu87rGnM5dp1U97G+m+5jNXOXarcG4x7VLizGPWy72DrNHo5dFJFgpc91pwsBr6N8Uucp8DXgWapaFJEk5ZNXpygfyf4C5SPll/Mk4DVAhHKj/YPA71SGPRv4HREJAjvAO4G/uEqmtwG/DTwGWK7kma8Zvgl8sZK1F/gQcK8up2mNdvfZ3d1lYGCg3TFME6xm7pLf3yfa44p7BZqK/eNjQlaz++5+vHKSbRdbp9lGewp4v4j8vKp+tRWBaqnqE+selyifrFn73D9SeR+qekPN8/8CfNsV5vs+4H1XGLZU+xqqerU7kH7fVbLfcJmnV1T1yVeZH6r6RuCNVxg2dbVpjTHGXNnxQx8K1tfWVfbm5hi0mhkDNN895iLlLiD/JiKrIvJ+Efn1FuQy91G5N49xE7/f3+4IpklWM3exermP1cx9rGat09SRdlW9SUQeAAwCK5T7ZN8/pyObpthNRNxnbKyhk8dNB7GauYvVy32sZu5jNWudpo60i8gvALdQ7tP9WeC7KJ8sac6hqlNXu1Sjqr6z0Uv+NMJuruQ+tdd/Ne5gNXMXq5f7WM3cx2rWOs12j3kJ8CBV/U7KNxXqAW6+5qnMfdbM9fdNZ6jeIMS4h9XMXaxe7mM1cx+rWevcm+u0FwBUtQj8HPDMa57KGGOMMcYY42i20f4FEXlBzeNTytcWNx2mq6vZ0pp2i8Vi7Y5gmmQ1cxerl/tYzdzHatY6zV7y8eXAh0TkJcB/AA8DPnXNU5n7zBrt7hMOh9sdwTTJauYuVi/3sZq5j9WsdRpqtIvIpKouq2oGeLyIPBn4JuCTwPtbGdCNROQpwMdVtW3XXSyVrnZjVtOJlpaWuNDB1yO+80644452p+gsW1s7DA4OtjuGadDAwDrPeMZEu2OYJnT6dtHck9WsdRo90n6biHwCeL2q3la5cdG/tDBXW4jIEnDz1a7y0i4icgtwq6re1O4s5hvTyQlcutTuFJ2lVBL7TFzk9NTuX2GMca9G+1BMAZ8B3ici/yoiP9C6SMYYY4wxxphaDTXaVXVfVd8IPAB4E/ArIvIVEflpEfG1NGELiMjPici8iOyJyJaIvFNEPgZMAG8WkYKIfLky7jtF5N11098iIjfXPH6UiHy2Mt1twDdf5jVfICL/KSI5EfmyiPxIzbAnioiKyLNF5GuVXB8XkdHK8LcBTwB+ufIahfPeo8fT7OkKpt3Gx8fbHcE0qb+/v90RTBOsK5P72HbRfaxmrdPsHVFVRP4e+ATwNOCVwG8CQy3I1hKVO7q+DniMqn5JRELAt6rqv92b7jEiEgb+EXgb8N8oX7/+Q3Xj3ED5c/ohyifwfgfwURG5qKq31oz6LODRlf9/FHgN8EJVfamIPJirdI8RkRuBG6uP+/r6mJubA8q3FB4bG+PixYscHh4C5bO7I5EIi4uLzjzGxsY4Pj4mlUoB5buqTk9Ps7m5SaFQcOY7ODjIwsICp6enQHlH6PF4WFtbc+Y1PT1NNptld3f3TIbV1VXnGq6xWIxwOMzS0pIz3fj4OEdHR04Gj8fD1NQUGxsb7O/vA+WTXJLJJPPz88716IeGhuju7j6TYWZmhkwmQzabBaC3t5fR0VFWVlY4Pj4GYGBggFAoxPLy8pkMh4eHbG9vN5VheHgYEWF9fd2Z1+zsLOl0mlwuB0AgEGBkZITl5WUuVfpVxONxAoEAX//6153bP09MTFAsFkmn0wD09PQwOTnJ+vo6xWIRgEgkQjweZ35+3nm9kZERVJWNjY3qcsHs7CypVIp8Pg9AMBhkeHiYpaUl59yHRCKB3+8/c1OMyclJCoUCOzs7AGxuBoFhdnd3neyBQC+BQNDJCRCNRjk9PXVeTwQSiST5fN5Z/nw+L5FIlHQ67SxHfX19eDweZ5mp1ufw8ND53D0eD7FYrC5DgEAgcE4GIZFI1GXwEYlEzs0Qj8c5ODhwMvT09NDf38/u7i7FYpGenh6CwSC9vb1nMvT391Mqldjb2wPKJ4fH43FyuZyzDvj9fsLhMNvb285yFA6H6erqcpbbaoZisejUvpohk8k4NQwGg/j9fqdeV8+Q5ejouC5DiurtHa6cYZ9i8aAuww6l0klNBh87O5m6DJfY2ys0lSESiSAC2WzOmVcikaBQKHBwUM7g9XqJRqPs7Ow4N5QLhUJ4vV4ymbszxGIxjo+PWVlZJ5HojPXJ6/UyMTHB2tqa836i0SixWIyFhQVnutHRUU5OTtjc3GwqQzKZxOfznckwNTVFPp93Phufz8f4+PiZfUN/fz/RaPTMvmF0dJRSqcTW1pZTw5mZGba2tpxlKxQKMTQ0xOLiolOLZDKJ1+vl4sWLzrymp6fJ5XJOhvP2T4eHh844tn9q3/5pZWXFme68/dPh4SGDg4O2PtHY+tQMaeQmPCLyX0CIciO/BOwB+cpfTlV/uOlXbhMRmQb+C7gB+Jiq5muGLVHXaBeRdwIeVX1+zXO3UGlAi8jzgN8DRlT1pDL854E3VU9EFZEvAG9R1bfXzONPKH/+PyUiT6R8Uu+kqq5Uhv8s8HOq+pD612zkfQ4NDWl1oTTuMDc319En79x+e/nP3C2VSpFMJtsdwzQomVzlB3/QjgK6SadvF809Wc2aIyJrqjrWyLiN9mm/A7gE/C7wUFV9mKo+TlW/200NdgBVXQR+BHgRsCIinxeRH70PsxwDVqsN9orFunEeALxBRLLVP+BHgZG68dZr/r8P9N2HXMYYY4wx5jrRaJ/251Hu+jEEfFFEXisirukSU09VP6iqTwfilL+IvEdEHkj5ZlH19oBg3XO1je2LwLiIdNc8N1U3/ibwMlWN1vyFVPV7m4h9uWxXJGJXSXAbOw/Bfex+CO7S3d19/kimo9h20X2sZq3T8CerqmuUT0D9q26TkQAAIABJREFUTeAngU+KyGeB31XVL7cq4LUmIg+i3O/831S1ICLVDpMnlBvXD6qb5Dbgxyt9yueBnwWma4b/A/BG4JUi8prKsF+om8cfAK8QkbuA24Eeyiernqpqox0ONoEHNjiu7ZxcaGpqqt0Rrqq7G3p62p2iswwPx9sdwTRhfLz+x03T6Tp9u2juyWrWOo3eXOkhQBiIVP72gT8HXgB8AXBTC9EL/DrwV1I+HL0CvEBV5ytfSN4kIi8GVlT1m4H3AI8DPg0cUz7h9N+rM1PVnIh8L/AW4P8DvgL8EfCGmnHeKCLpyrSzlM8L+BLlE3kb9Qbgz0Rkl3Jf+OjVRq6ehGPcY2Njg+Hh4XbHuKJHPKL8Z+7W6TUzZ5VPgLN6uYmtY+5jNWudRk9EXQV2gRyQrf9T1TdcZXLTBnYiqvvYyTvuYzVzF6uX+1jN3Mdq1pxmTkRt6Ei7qtrp9sYYY4wxxrSJnUV1nbIT5NwnHA63O4JpktXMXaxe7mM1cx+rWes01D3GuM/Y2JjW3tDCGGOMMcZ0llZcp924TPUuXsY9au8cZ9zBauYuVi/3sZq5j9WsdazRbkyHsF+93Mdq5i5WL/exmrmP1ax1rNFujDHGGGNMh7NG+3XKbq7kPkNDrr3J8Dcsq5m7WL3cx2rmPlaz1rF7zRrTITr5i9add8Idd7Q7Rec5Pvbi9bY7hWnUQx7i4du/vd0pTDM6ebtoLs9q1jrWaG8hEfGq6nE7Xvvk5KQdL2vug7W1tY69IcXJCVy61O4UnWd7O0symWx3DNOgzc1twG474iadvF00l2c1ax3rHlNDRG4RkTeKyHtFJCciqyLyM5VhN4jIxbrxXy0it9ZN/2YReZ+I7AJvEpFo5XFaRPIi8jUReXbNNI+tTLcjIssi8lsi4qkM84rIW0VkU0T2RGRJRH7+fvo4jDHGGGNMh7Aj7fd0A/As4PmVf/9GRD7exPQvAp4D/BjgB34d6AOmgQIwAQQBRORBwCeAnwL+FhgFPgQcAr8NvBB4HPBwVU2LyBAwct/enjHGGGOMcRtrtN/T36nqv1T/LyIZ4FFNTP8hVf1o5f9FETkGBoAHA7er6nLNuD8LfFhV31d5vCwirwN+g3Kj/RgIAQ8Vkc+o6iawebkXFZEbgRurj8PhMHNzcwD4/X7Gxsa4ePEih4eHAMRiMSKRCIuLi848xsbGOD4+JpVKAeV+adPT02xublIoFADo6+tjcHCQhYUFTk9PARgcHMTj8bC2tubMa3p6mmw2y+7u7pkMq6urHB0dORnC4TBLS0vOdOPj4xwdHTkZPB4PU1NTbGxssL+/77y3ZDLJ/Py8c2mpoaEhuru7z2SYmZkhk8mQzWYB6O3tZXR0lJWVFY6Py72WBgYGCIVCLC8vn8lweHjI9vZ2UxmGh4cREdbX1515zc7Okk6nyeVyAAQCAUZGRlheXuZSpb9JPB4nEAigqk7NJiYmKBaLpNNpAHp6epicnGR9fZ1isQhAJBIhHo+fuSbuyMgIqsrGxkZ1uWB2dpZUKkU+nwcgGAwyPDzM0tKScz3/RCKB3+9ndXXVmdfk5CSFQoGdnR1WV31kMgFisQF2d3ed7IFAL4FA0MkJEI1GOT09dV5PBBKJJPl83ln+fD4vkUiUdDrtLEd9fX14PB5nmanW5/Dw0PncPR4PsVisLkOAQCBwTgYhkUjUZfARiUTOzRCPxzk4OHAy9PT00N/fz+7uLqpKKpUiGAzS29t7JkN/fz+lUom9vT2gfJfieDxOLpdz1gG/3084HGZ7e9tZjsLhMF1dXc5yW81QLBad2lczZDIZp4bBYBC/38/Ozk4DGbIcHR3XZUhRvVLblTPsUywe1GXYoVQ6qcngY2cnU5fhEnt7haYyRCIRRCCbzTnzSiQSFAoFDg7KGbxeL9FolJ2dHadLYCgUwuv1ksncnSEWi3F8fOysY+1en6rZJyYmWFtbc95PNBolFouxsLDgTDc6OsrJyQmbm5tNZUgmk/h8vjMZpqamyOfzzmfj8/kYHx8/s2/o7+8nGo2e2TeMjo5SKpXY2tpyajgzM8PW1pazbIVCIYaGhlhcXHRqkUwm8Xq91N7ob3p6mlwu52Q4b/9UrZntn+7O0I7908rKijPdefsnVWV7e9vWJxpbn5phd0StISK3ALeq6k01zy0BNwMl4Obau1aJyKuBp6jq42um/7Sq/lrNOAHgfwLPpHyU/Z+BX1XVORH5GPAkykfWq7qALlUNiUg38IuUj9w/GPg0cJOq3n7eexkZGdHaldN0vnQ6TTweb3eMy7r99vKfOatQ2CMU6mt3DNOg2dldvuu7+tsdwzShk7eL5vKsZs2xO6K2xh6Vbi01LtdV5bT2gaoWVfWVqvotwCzlxv9fVAZvAu9V1WjNX1hVQ5VpT1T19ar6WMpdZ+4CPthI2OpRBuMetUc1jTtUjzobd6ge7TfuYdtF97GatY412ht3B9AnIs8VkS4ReSLlI+BXJSLPEJGHVU4uLQIHlBvuAG8Fni0iz6mcdNotIhdE5OmVaZ8sIt8mIl7KR+MLgF0WxhhjjDHmG4z1aW+Qqi6IyM8Brwf+BPgI8OfAY86ZdBp4AzAMHAH/B/jJyjw/LyJPBX6LcgO+B1gC/qgybRJ4EzAFXAK+ADybBohIY2/MdIze3t52R7ii7m7o6Wl3is4TCPTY5+IigYCv3RFMkzp5u2guz2rWOtan/To1NjamtSf/GGOMMcaYzmJ92o3dXMmFas/ON+5gNXMXq5f7WM3cx2rWOtZov07ZLyjuU73Ml3EPq5m7WL3cx2rmPlaz1rFGuzHGGGOMMR3OGu3Xqa4uK63bDAwMtDuCaZLVzF2sXu5jNXMfq1nrWMvuOmWNdvcJhULtjmCaZDVzF6uX+1jN3Mdq1jrWsrtOVW+9a9yj9lbVxh2sZu5i9XIfq5n7WM1axxrtgIh8TERe0e4cxhhjjDHGXI7dXAlQ1e9pdwZjOtWdd8Idd7Q7RWfa2upjcLDdKUyjBga8XLjQ7hTGGHPvWKP9OuXxWGndZnx8vN0RLuvkBC5daneKztTXF7PPxkUSiaF2RzBN6tTtorkyq1nrWPcYQERuEZGbK/9/SKW7TFpELorI20UkUhn2AyKyXjPdC0VEReSZlcceEcmLyBMqj39WRL5UeW5TRN4lIvGa6V8tIreKyKtEZENEMpXX81SGe0XkrZVp90RkSUR+vpH3ZNdpd5/Dw8N2RzBNKpWsxe4mx8dH7Y5gmmTbRfexmrWONdpriEgf8M/AfwETwKOBBwN/URnlFiAuIt9UefzdwNeBp1Uef3vl389U/t0EfgiIAo8FHgj8Yd3LPhbYByYr0z8H+PHKsBcCjwMerqp9leH/3sh7sTuius/29na7I5gm7e0V2h3BNGF3N9vuCKZJtl10H6tZ61gfirO+H/ACv6KqJaAoIi8H/kNEhlR1U0T+D/A0EfkS8BTgpcDrKtM/DfhkZVpU9e9q5r0sIq8F3lH3mquq+vrK/78mIp8AHgP8OXAMhICHishnVHWT8heBexCRG4Ebq49DoRBzc3MA+P1+xsbGuHjxovMNOBaLEYlEWFxcdOYxNjbG8fExqVQKgO7ubqanp9nc3KRQKDdO+vr6GBwcZGFhgdPTUwAGBwfxeDysra0585qeniabzbK7u3smw+rqKkdHR06GcDjM0tKSM934+DhHR0dOBo/Hw9TUFBsbG+zv7wMQDodJJpPMz887vygMDQ3R3d19JsPMzAyZTIZstryj7u3tZXR0lJWVFeeObQMDA4RCoTNnu4+Pj3N4eOhseBrNMDw8jIiwvu78GMPs7CzpdJpcLgdAIBBgZGSE5eVlLlX6VcTjcQKBAOl02pluYmKCYrHoPNfT08Pk5CTr6+sUi0UAIpEI8Xic+fl5Z7qRkRFUlY2NjepywezsLKlUinw+D0AwGGR4eJilpSXnKkOJRAK/38/q6qozr8nJSQqFAqurBVIpHx5PN7HYALu7u072QKCXQCB4Jns0GuX09NR5PRFIJJLk83ln+fP5vEQiUdLptLMc9fX14fF4nGWmWp/Dw0Pnc/d4PMRisboMgXt8fvfMICQSiboMPiKRyLkZ4vE4BwcHToaenh76+/vZ3d111otgMEhvb++ZDP39/ZRKJfb29oDyZVjj8Ti5XM5ZB/x+P+FwmO3tbWc5CofDdHV1OcttNUOxWHRqX82QyWScGgaDQfx+Pzs7Ow1kyHJ0dFyXIUX1B7orZ9inWDyoy7BDqXRSk8HHzk6mLsMl5wtOoxkikQgikM3mnHklEgkKhQIHB+UMXq+XaDTKzs6Oc6AiFArh9XrJZO7OEIvFOD4+JpvNMjd31Nb1qVofr9fLxMQEa2trzvuJRqPEYjEWFhac6UZHRzk5OWFzc7OpDMlkEp/PdybD1NQU+Xze+Wx8Ph/j4+Nn9g39/f1Eo9Ez+4bR0VFKpRJbW1tODWdmZtja2nKWrVAoxNDQEIuLi04tkskkXq+XixcvOvOanp4ml8s5Gc7bP1XXKds/3Z2hHfunlZUVZ7rz9k/pdNrWpybWp2aIdaMod48BbgXywA+r6rfVDOurPP9YVf2ciNwEPAH4VeB9wIMoN6S/A3g38G5VfUtl2h8Cfgm4APgp/7IRBDyqeiIirwaeoqqPr3m9d1aGP19EuoFfpHz0/cHAp4GbVPX2897T8PCwVlcM4w5LS0tMTU21O8Y93H57+c/cUzqdJh6Pnz+i6QjDw+v8wA+MtDuGaUKnbhfNlVnNmiMia6o61si41j3mrFVgotqnvGK28m/1a+bHKTfanwn8by1/6/k48FzK3Wk+DiAiY8D/otwdZkJVw9zd7UUaCaOqJ6r6elV9LDAK3AV8sJFpu7u7GxnNdBDbyLmPNdjdZWTEGuxuY9tF97GatY412s/6CFACXiMivSIyBPw+8OFK1xSAzwOHwC8A/1R57p+AX6bc1eVrledClD/ftKoeisgDKB+db5iIPFlEvk1EvJXXLAANdVav/jRo3MN+GXGfXM76SLtJOm19bd3GtovuYzVrHevTXkNV8yLyVOD3gIuUG8ofpdwgr45zWul3/kzgk5Wn/wkIA39TM95XRORXgb8UkRDwRcrdZx7dRKQk8CZgCrgEfAF4diMTWqPdfar9ETtNdzf09LQ7RWc6OTmyz8ZFjo4O2h3BNKlTt4vmyqxmrWN92q9TQ0NDWj3RwrjD3NwcF+zOL65iNXMXq5f7WM3cx2rWHOvTbujqstK6TTgcbncE0ySrmbtYvdzHauY+VrPWsSPt16mxsTGtvcyWMcYYY4zpLHak3TjXFjXuUXs9W+MOVjN3sXq5j9XMfaxmrWONdmM6hP3q5T5WM3exermP1cx9rGatY412Y4wxxhhjOpw12q9TdnMl9xkeHm53BNMkq5m7WL3cx2rmPlaz1rFGuzEdQqShG+WaDmI1cxerl/tYzdzHatY6dnMlFxGRW4BbVfWm88Y9OWnoxqmmg6yvr3fktW3vvBPuuKPdKTrT1tYeg4OBdscwDRoYSPOMZ0y0O4ZpQqduF82VWc1axxrtxpirOjmBS5fanaIzlUpin42LnJ7aEUBjjHtZ95hrTES898c0xhhjjDHmG4c12u8jEblFRN4sIu8TkV3gCyKiIuKpGecGEblY8/idIvI3IvJHIrINfLDy/B+LyJKIFERkUUR+Q0TuVY08HvsRxW1mZ2fbHcE0KZFItDuCacLYWEP3LzEdxLaL7mM1ax1r2V0bLwKeA/wY8HTgIw1M8yzgJcD/C/RUnvsc8AogBTy2Mp914O3nzUxEbgRurD7u6+tjbm4OAL/fz9jYGBcvXuTw8BCAWCxGJBJhcXHRmcfY2BjHx8ekUimgfAWa6elpNjc3KRQKznwHBwdZWFjg9PQUgMHBQTweD2tra868pqenyWaz7O7unsmwurrK0dGRkyEcDrO0tORMNz4+ztHRkZPB4/EwNTXFxsYG+/v7QPkWyclkkvn5eed6sENDQ3R3d5/JMDMzQyaTIZvNAtDb28vo6CgrKyscHx8DMDAwQCgUYnl5+UyGw8NDtre3m8owPDyMiLC+vu7Ma3Z2lnQ6TS6XAyAQCDAyMsLy8jKXKv0q4vE4gUCAu+66i2AwCMDExATFYpF0Og1AT08Pk5OTrK+vUywWAYhEIsTj8TM3shgZGUFV2djYqC4XzM7OkkqlyOfzAASDQYaHh1laWnJuwpVIJPD7/ayurjrzmpycpFAosLpaIJXy4fF0E4sNsLu762QPBHoJBIJOToBoNMrp6anzeiKQSCTJ5/PO8ufzeYlEoqTTaWc56uvrw+PxOMtMtT6Hh4fO5+7xeIjFYnUZAgQCgXMyCIlEoi6Dj0gkcm6GeDzOwcGBk6Gnp4f+/n52d3cpFAr4fD6CwSC9vb1nMvT391Mqldjb2wOgq6uLeDxOLpdz1gG/3084HGZ7e9tZjsLhMF1dXc5yW81QLBad2lczZDIZp4bBYBC/38/Ozk4DGbIcHR3XZUhRvbzylTPsUywe1GXYoVQ6qcngY2cnU5fhEnt7haYyRCIRRCCbzTnzSiQSFAoFDg7KGbxeL9FolJ2dHec8nlAohNfrJZO5O0MsFuP4+Jivf/0i0WhXW9enan28Xi8TExOsra057ycajRKLxVhYWHCmGx0d5eTkhM3NzaYyJJNJfD7fmQxTU1Pk83nns/H5fIyPj5/ZN/T39xONRs/sG0ZHRymVSmxtbTk1nJmZYWtry1m2QqEQQ0NDLC4uOrVIJpN4vV5q7849PT1NLpdzMpy3f9rf3ycYDNr+qSZDO/ZPKysrznTn7Z/29/cZGRmx9YnG1qdmiF0E/76pnBy6oao/Wnn8ROCTQI+qlirP3QDcXL1NrYi8E3iQqj7unHm/ERhT1f9e81oNnYg6NDSk1YXSuMPc3FxHnrxz++3lP3NPqVSKZDLZ7himQcnkKj/4g+PtjmGa0KnbRXNlVrPmiMhatX14HjvSfm0snj/K1aeR8jWSfo3y0foRQAA/8Pn7nM4YY4wxxria9Wm/Nk5r/r9X+TdY89zIOdMA/AjwcuAFQFxVo5S7xdyryx3YdVLdJxCwSwe6jddr55C7id/vb3cE0yTbLrqP1ax17Ej7tfc1yg33l4jI64FvBn66gekiQIlyf3YVkScBzwfuujch7I6o7jMycrnvdu3X3Q09PeeP940okYi2O4JpwtCQnTjsNp26XTRXZjVrHWu0X2OquiciLwR+F3gl8Cngj4GXnTPpO4HHA1+kfBT+48C7gUfdmxx2cyX3WV5eZnJyst0x7uERjyj/mXvq1JqZyyuf0Gf1chNbx9zHatY61mi/j1T1iZd57v3A++uefk3N8BsuM80h5SPrTb3WVcZtdFTTIS7ZXXpcx2rmLlYv97GauY/VrHWsT7sxxhhjjDEdzhrt1ynr0+4+8Xi83RFMk6xm7mL1ch+rmftYzVrHGu3XKbt6jPvYGffuYzVzF6uX+1jN3Mdq1jrWaL9OVe/iZdyj9o5zxh2sZu5i9XIfq5n7WM1axxrtxhhjjDHGdDhrtBtjjDHGGNPhrNFeISK3iMjNLZr3DSJysRXzvhKPx67m6TYTExPtjmCaZDVzF6uX+1jN3Mdq1jrWsnMBEXki8EmgR1Ub6qxu12l3n2KxiNfrbcm877wT7rijJbP+hlYslggEWlMzc+1duHDEE55g9XKTVm4XTWtYzVrHGu3XKbsjqvuk02mi0WhL5n1yAna/i2tvd7dAT49dKcEtdnayQF+7Y5gmtHK7aFrDatY61j3mrKiI/L2I7InInIi8AC7fvUVEXi0it9Y87hWRm0Xka5XpF0TkhZd7ERF5jIisisjLa557gYj8p4jkROTLIvIjlecngI9VRsuKSEFEfu1av3FjjDHGGNO57Ej7WT8J/HDl72nAB0RkvsFp/wR4IPBM4CvAMDBUP5KIPBd4E/ATqvqRynM3AL8J/BDwH8B3AB8VkYuqequIfA/l7jHRRrvH2HXa3aenp6fdEUyT7CZm7mLn+riPbRfdx2rWOrYFO+ujqvrh6v9F5P3ATwCfutpEIhIHngc8WlXvqjy9XvmrHe9VwIuAp6rqF2oG3Qj8tqreVnl8q4j8NXADcCsNEJEbK/MBIBwOMzc3B4Df72dsbIyLFy9yeHgIQCwWIxKJsLi46MxjbGyM4+NjUqkUUG6QTE9Ps7m5SaFQAKCvr4/BwUEWFhY4PT0FYHBwEI/Hw9ramjOv6elpstksu7u7ZzKsrq5ydHTkZAiHwywtLTnTjY+Pc3R05GTweDxMTU2xsbHB/v6+896SySTz8/NO3/2hoSG6u7vPZJiZmSGTyZDNZgHo7e1ldHSUlZUVjo+PARgYGCAUCrG8vHwmw+HhIdvb201lGB4eRkRYX7+77LOzs6TTaXK5HFC+6cTIyAjLy8tcqvRXicfjBAIBLl265NRsYmKCYrFIOp0GyhvByclJ1tfXKRaLAEQiEeLxOPPzd3+vHBkZQVXZ2NioLhfMzs6SyWRIpcrf93w+L5FIlHQ67dSwry+Ex9Pj1Kv82cQ4PDxy3rPH000sNsDu7q6TPRDoJRAIOjkBotEop6en5PP5SgZIJJLk83ln+bt8hj48Hk9dhgEODw9rMniIxWJ1GQIEAoFzMgiJRKIug49IJHJuhng8zsHBgZOhp6eH/v5+dnd3OTk5IZVKEQwG6e3tPZOhv7+fUqnE3t4eAF1dXcTjcXK5nLMO+P1+wuEw29vbznIUDofp6upylttqhmKx6NS+miGTyTj3ZAgGg/j9fnZ2dhrIkOXo6LguQ4rqqTBXzrBPsXhQl2GHUumkJoOPnZ1MXYZL7O0VmsoQiUQQgWw258wrkUhQKBQ4OChn8Hq9RKNRdnZ2nC6BoVAIr9dLJnN3hlgsxvHxMaVSibm5ufu8PqVSKWfZCgaDDA8Ps7S05NQikUjg9/tZXV115jU5OUmhUHDq4/V6mZiYYG1tzXk/0WiUWCzGwsKCM93o6CgnJydsbm42lSGZTOLz+c5kmJqaIp/PO5+Nz+djfHz8zL6hv7+faDR6Zt8wOjpKqVRia2vLqeHMzAxbW1vOshUKhRgaGmJxcdGpRTKZxOv1cvHi3T9UT09Pk8vlnAyN7J/m5uZs/1SToR37p9prrzeyf9re3rb1icbWp2aInbBYJiK3ALep6i/VPPda4FuB9wI3q+pYzbBXA09R1ceLyKOBzwFhVd27zLxvAN4ClIDXqOrv1A0/AE4qw6s8wL+p6vfemxNRh4eHtbpiGHdYX19nZGSkJfO+/fbyn7m2stms9d10kYmJbZ7+9ES7Y5gmtHK7aFrDatYcEVmrbV9ejR1pP2vqMo8vAntAsG5Y7RK5VPn3gcCVmka7wDOAj4mIqurraoZtAq9S1b+8wrSnV019GfZlzH2qRyiMe1SPiBl3qB79Mu5h20X3sZq1jp2Ietb3isj3iUi3iDwdeBbw58AdQJ+IPFdEuipHvp9TnUhVt4G/At4iIg8CEJFhEfnW2pmr6n8AjwdeJiK1jfY/AF4hIo+uzN9X+f+jKsM3K/8+6Nq/ZWOMMcYY0+nsSPtZf0b5ZNT3ASngpar6KQAR+Tng9ZRPOP0I5cb8Y2qmfTHwSspH0pOV6V9F+cRSh6p+XUS+E/jfIjIA/LSqvlFE0sDbgFnK3WS+VJkfqvo1EflD4JMi4gVeq6qvvdob6eqy72NuE4lEWjbv7m6wc4OuvXDYb5+ri4TD9T+Ymk7Xyu2iaQ2rWetYn/br1NjYmNae/GM6n6raVX9cxmrmLlYv97GauY/VrDnN9Gm3w7HXqeoZz8Y9as+yN+5gNXMXq5f7WM3cx2rWOtZoN8YYY4wxpsNZo90YY4wxxpgOZ43265TdqdF97Lq27mM1cxerl/tYzdzHatY61mg3pkPYSeHuYzVzF6uX+1jN3Mdq1jrWaL9OVW8jbdzD7mDrPlYzd7F6uY/VzH2sZq1jjfb7QEQ+JiKvaOH8L4iIishU5fHzROSrrXo9Y4wxxhjTmezmSpchIrcAt6rqTVcbT1W/5/5J5Lzee4D33J+vae4/zVzX9s474Y47WhjGNCSVCpNMtjuFaVQ87uPChXanMM2w6327j9WsdazRfi+IiFdVj9ud42o8Hiut28zOzjY87skJXLrUwjCmIf39SauDi4yMjLc7gmlSM9tF0xmsZq1j3WPqiMjbgCcAvywihcrfDSJyUUR+VkSWgJ3KuLeIyM2V/79XRP60bl7fKiLHIjJYefxgEfkHEdkSkTUReauIBGvGnxWRT4hIXkTuAp5UN78bRKSh25yenp7eh0/BtEMqlWp3BNOkfD7f7gimCZlMpt0RTJNsu+g+VrPWsUZ7HVV9KfAp4HWqGlLVUGXQEPAtwMOBwctM+g7gh0UkVPPci4F/UNUtEYlX5vsJYKIyrwcCfwAgIt3Ah4EVYBh4amX6e8Ua7e5jDUD3OTw8bHcE04T9/f12RzBN+r/t3Xmca2ld5/HPr5JKJalUkkolqX2/LAJKg4zYjjPTNKIjLS2MC6BsI+C0CIKtKPJiRpFFUWEUF0BbaGVrBNmGARwUerAFGejuiw1oQ93a9zW13Npu1X3mjySnU9X3dlW6K5Wc6u/79arXvZWz5Jv8zpPz5NRzztHnov+oZpWjMRTleZVzbvMq074AzAHPAW4xsyjwXOBnCtNfAAw55/5n4fcdM/tN4PNmdhPw/cCjgGudcxeBi2b2BuCTxwlmZjcDNxd/j8ViDA0NARAOh+nq6mJyctLrZKRSKRKJBCMjI946urq62N3d9b4lBwIB+vv7mZ2dZWNjA4CmpiZaW1sZHh72vhi0trYSDAaZmpry1tXf308ul2NlZeVAhomJCXZ2drwM8Xic0dFRb7nu7m52dna8DMFgkL6+PmZmZrwdbjweJ5vNcuHCBe/SUm1tbQQCgQMZBgYGWF5eJpfLARCJROjs7GR8fJzd3fxcNTVWAAAgAElEQVToppaWFmKxGGNjYwcybG9vs7CwUFaG9vZ2zIzp6WlvXYODgywuLrK6ugpANBqlo6ODsbExLhXGVaTTaaLRKIuLi95yPT09bG5ueo/V19fT29vL9PQ0m5ubTEw0sL6eJBaLeTkBkskEzuE9nxlkMlnW1ta82jc0hEgkkiwuLno1bGqKEQzWe/XKvzcptrd3vNccDAZIpVpYWVnxskejEaLRxgPZk8kkly9f9j64j5+hiWAweChDC9vb2yUZgqRSqUMZovd7/+6fwchkMocyNJBIJI7MkE6n2dra8jLU19fT3NzMysqK1y4aGxuJRCIHMjQ3N7O3t8f6+joAdXV1pNNpVldXvTYQDoeJx+MsLCx421E8Hqeurs7bbosZNjc32dzcPJBheXmZvb09L0M4HGZpaekYGXLs7OweyjBP8UptV89wkc3NrUMZltjb2y/J0MDS0vKhDJdYX98oK0MikcAMcrlVb12ZTIaNjQ22tvIZQqEQyWSSpaUl74pZsViMUCh04Kh6KpVid3eXXC7H0NDO/dpT8fnS6fSBW7B3dHTgnPOuhmFmDA4OMj8/721bjY2NtLe3Mzo66tUik8kQDoeZmJjw1tXb28vGxoZXn1AoRE9PD1NTU97rSSaTpFIphoeHveU6OzvZ399ndna2rAzZbJaGhoYDGfr6+lhbW/Pem4aGBrq7uw/sG5qbm0kmkwf2DZ2dnezt7TE3N+fVcGBggLm5OW/bisVitLW1MTIy4tUim80SCoWYnLzvD8T9/f2srq56GY7aPxXblPZP92Woxv5pfHzcW+6o/dPi4qLaUxntqRym62ne3+ETUc3sRcCbnXMdR8z3WuAZzrlrzeyFwBuBXufcZTN7B/BioLTTb0ADMAj8IPCnzrl0yfofB9wD9DvnRgs53uic6zrqNXR0dLjSxim1b2Zmhvb29mPNe+ed+R+prtXVHIlEstox5Jh6exf4kR/JVDuGlKGcz0WpDapZecxs6jj9OtDwmKu50tiS44w3uRV4kpk9FngJ8B7nXHG5WfId/GTJT8I5F3bOTQGTQLOZJUrW1/dgX0BdnUrrN/qQ8x912P0lnVaH3W/0ueg/qlnlqGd3ZbPkx5uXxTk3DXwGeAvwA8C7Sya/B3iCmb3MzKKW121mzyxM/wrwHeBtZtZoZp3AA15y8oHo5kr+U/pnWPGH0uEwUvv010f/0eei/6hmlaMx7Vf2VuDdZrZCfgjLq8pY9hbgE8DnnHOjxQedc+Nmdi3wO8B/B6LAFPBB4OPOuT0zewbwLmCmMO1twJMfzAvQsCf/KY6dO45AAOrrKxhGjqWubl918BHnjt/GpDaU87kotUE1qxyNaT+j2traXPFEC/GHoaEhzunOL76imvmL6uU/qpn/qGbl0Zh2IRAIVDuClCmT0Xhbv1HN/EX18h/VzH9Us8pRp/2M0m2E/SccDlc7gpRJNfMX1ct/VDP/Uc0qR532M0pjyvyn9Pqv4g+qmb+oXv6jmvmPalY56rSLiIiIiNQ4ddpFRERERGqcrh5zRnV1dbnSW0dL7bt06RL1un6gr6hm/qJ6+Y9q5j+qWXl09Rjh8uXj3MBVasnGxka1I0iZVDN/Ub38RzXzH9Wscs7EzZXMrAf4FvA9zrnhauepBeq0+8/S0hLNzc0HHjt/Hu6+u0qB5Ehzc7u0tlY7hRxXS8s6N97YfPSMUjOu9LkotU01q5wz0Wl3zo0DsWrnqBQz6wNGgEc454aqm0ZO0/4+XLpU7RRyNXt7pvr4yOXLuhSuiPjXw2J4jJmFqp3htOk67f4TCj3sNlPfCwZ1EzM/qa8/E8epHlb0ueg/qlnlnFqn3czCZvZmM7tgZitm9kUze0Jh2m+Z2R1m9ptmNmNmy2b2LjMLFqZ/wMz+8tD6nmhmu2bWamZ9ZubM7Fxh2ovMbNLMftHMRoGlwuNJM/vzwrRFM/uMmT2qZJ0PmKMwjzOzV5rZl83sopl93cy+28x+yszuNbM1M/uImcVKlkma2TvMbMzMlszs02Y2UDL9VjO7zcz+pDB9zszeUPJyv1n49+tmtmFm7zzq/dYdUf2np6en2hGkTKlUS7UjSBna2tqrHUHKpM9F/1HNKuc0Dzu8E+gC/hMwC9wE/J2ZPbIw/cnAx4FeoA/458LPe4BbgE+Y2Sudc8UzHF4KfMo5N1cYPnJYG/B44HFAcYD3e8kPo3kSsAa8Gfh7M/uukvU+UI6iFwH/BZgC3leY/4uF9UaBLwGvAH7H8oe8PwZMA08ALgK/DXzKzB7vnCv+cf1ZwAuBVwLfB3zRzD7vnPsC8Fjyw2Mef7XhMWZ2M3Bz8fdYLMbQUH7WcDhMV1cXk5OTbG9vA5BKpUgkEoyMjHjr6OrqYnd3l/n5eSDf8e/v72d2dtY7saSpqYnW1laGh4e9cfOtra0Eg0Gmpqa8dfX395PL5VhZWTmQYWJigp2dHS9DPB5ndHTUW667u5udnR0vQzAYpK+vj5mZGS5evAhAPB4nm81y4cIFilc/amtrIxAIHMgwMDDA8vIyuVwOgEgkQmdnJ+Pj4+zu7gLQ0tJCLBZjbGzsQIbt7W0WFhbKytDe3o6ZMT097a1rcHCQxcVFVldXAYhGo3R0dDA2NsalwriKdDpNNBrlnnvuIZFIAPkPvc3NTSYm1pmfbyAQCNDS0kIul/OyRyIRYrGYlxMgmUzgHN7zmUEmk2Vtbc2rfUNDiEQiyeLiolfDpqYYwWC9V6/8e5Nie3vHe83BYIBUqoWVlRUvezQaIRptZHFxsSRDksuXL7O2tlZmhiaCweChDC1sb2+XZAiSSqUOZYgSjUaPyGBkMplDGRpIJBJHZkin02xtbXkZ6uvraW5uZmVlhbW1NSKRCI2NjUQikQMZmpub2dvbY319HYC6ujrS6TSrq6teGwiHw8TjcRYWFrztKB6PU1dX5223xQybm5tsbm4eyLC8vOzdSK2xsZFwOMzS0tIxMuTY2dk9lGGe4sXErp7hIpubW4cyLLG3t1+SoYGlpeVDGS6xvr5RVoZEIoEZ5HKr3roymQwbGxtsbeUzhEIhkskkS0tL7O/nM8RiMUKhEMvL92VIpVLs7u7yr/86RiJRT319Pb29vUxPT3vvaSKRIJ1Oc+HCBW+5jo4OnHPMzMx429Hg4CDz8/PettXY2Eh7ezujo6NeLTKZDOFw+MBNZnp7e9nY2PDqEwqF6OnpYWpqyns9yWSSVCrF8PB9p2Z1dnayv7/P7OxsWRmy2SwNDQ0HMvT19bG2tua9Nw0NDXR3dx/YNzQ3N5NMJg/sGzo7O9nb22Nubs6r4cDAAHNzc962FYvFaGtrY2RkxKtFNpslFApReiWz/v5+VldXvQxH7Z9WV1dJJBLaP5VkqMb+aXx83FuuuH8qft4dbk+rq6v09PSoPXG89lSOU7nko5m1AIvAo51z95Y8/h3g9cA54AXOudKjzx8GFp1zv1Do+H4H+F3n3C1mFiXfCf4Z59ynD4/5NrMXke/ox51zm4X1tReWucY59/XCY/XADPBy59xtZvZbD5Sj8LsrzPPewu8/Tr7T3uGcmyk89sdAl3PuWWb2RPKd/lTxi4GZBYAN4GnOuTvM7Fagxzl3fcnzfhX4iHPuLQ9mTHtbW5srbpTiD0NDQ5w7d+7AY3femf+R2jQ/P082m612DDmmbHaCZz6zu9oxpAxX+lyU2qaalaecSz6e1pH2YvW+cmisdYj80XfId6hLXQSaAJxzzszeDbyYfGf8p4B14LMP8JzzxQ57QfGT2vva55y7ZGZjQOnfcq6ao8TMoekUO+xXWOYR5N/nySuMMy/dexzneUVERETkYei0Ou3FQ77fU7jSywGFI9xHuRV4vZk9FngJ8B7n3ANd1/DwtOLfNAaB4pH2IPkO+/0ynaBZYBfIlAyFKVfZ12+sq3tYnGN8piSTyWpHkDJFo5FqR5AyNDWd2YuMnVn6XPQf1axyTqXT7pwbM7OPA39qZi8v/N4E/CBw/pjrmDazzwBvAX4AeH6ZGWbM7NPAW83sZ8mPaX8j+Q71/y5nXWW6A/gG8A4ze61zbt7MmoHrgc8c+mvA1SyQ77g/CjjW8Bh12v0nlUrd77FAAHRjudqVSDSipuYfzc2JakeQMl3pc1Fqm2pWOad5IurPAL8OfK4wvnyd/Fjvl5exjluATwCfc86NPogMzwd+D7gLCANfBX7IObf+INZ1LM65fTN7Gvmx+18xszSwQv7E1U8fcx1bZvZa4BYziwAfcM697IGWKZ48If4xPDx8v3GA11yT/5HaNDR0/5pJ7RoaGua+0ZriB1f6XJTapppVzqmciCqnTyei+o9O3vEf1cxfVC//Uc38RzUrTzknouoPuyIiIiIiNU6d9jNKN1fyn87OzmpHkDKpZv6ievmPauY/qlnlqNMuUiOKNyQR/1DN/EX18h/VzH9Us8pRp/2MUqPxH52D4D+qmb+oXv6jmvmPalY56rSLiIiIiNQ4ddpFasQV7pgrNU418xfVy39UM/9RzSpHl3w8QWZ2O3CHc+51V5h2HfAFoN45V/GLqHd1dbnJyclKP42IiIiIPEjlXPLxNG+uJKfo8uXL1Y4gwPnzcPfdx5t3bW2NeDxe2UByolQzf+ntXeb663W3Rj+Zn58nm81WO4aUQTWrHHXazyh12mvD/j5cunS8edfXt4lE1AH0E9XMX9bWLgLqtPvJ2tqaOoA+o5pVjsa0n7ykmX3UzNbNbMjMXnClmczsVjN736HHbjezN5b83mlmHzCzKTObN7MPmlmm0i9ARERERGqLOu0n78XAe4Bm4JeAW8zs35e7EjNrAP4BmAYeCQwAe8AHjrN8XZ1K6zcNDaFqR5AyqWb+EomEqx1BytTY2FjtCFIm1axyNDzm5H3aOfe/iv83s48BPwe8t8z13AA0Aa92hbOFzew1wKSZdTnnDpxlamY3AzcXf4/H4wwNDQEQDofp6upicnKS7e1tAFKpFIlEgpGREW8dXV1d7O7uMj8/D+Tvqtrf38/s7CwbGxsANDU10drayvDwsDcEp7W1lWAwyNTUlLeu/v5+crkcKysrBzJMTEyws7PjZYjH44yOjnrLdXd3s7Oz42UIBoP09fUxMzPDxYsXvdeWzWa5cOECxROp29raCAQCBzIMDAywvLxMLpcDIBKJ0NnZyfj4OLu7uwC0tLQQi8UYGxs7kGF7e5uFhYWyMrS3t2NmTE9Pe+tybpD19Q22trYACIVCJJNJlpaWvGvpx2IxQqEQOzv3vfepVIrd3V3vfQ8EArS0tJDL5bzskUiEWCzm5QRIJhM4B6urq4XtAjKZLGtra17tGxpCJBJJFhcXvRo2NcUIBuu9euXfmxTb2zveaw4GA6RSLaysrHCpMOYnGo0QjTayuLhYkiHJ5cuXWVtbKzNDE8Fg8FCGFra3t0syBEmlUocyRIlGo0dkMDKZzKEMDSQSiSMzpNNptra2vAz19fU0Nzd7Gebn52lsbCQSiRzI0NzczN7eHuvr60D+i3Q6nWZ1ddVrA+FwmHg8zsLCgrcdxeNx6urqvO22mGFzc5PNzc0DGZaXl9nby5/X3tjYSDgcZmlp6RgZcuzs7B7KME/xugRXz3CRzc2tQxmW2NvbL8nQwNLS8qEMl1hf3ygrQyKRwAxyuVVvXZlMho2N47Wn5eX7MhTb09bWNkNDQ9TX19Pb28v09LT3niYSCdLpNBcuXPCW6+jowDnHzMyMtx0NDg4yPz/vbVuNjY20t7czOjrq1SKTyRAOh5mYmPDW1dvby8bGhlefUChET08PU1NT3utJJpOkUimGh4e95To7O9nf3/eufX3cDNlsloaGhgMZ+vr6WFtb896bhoYGuru7D+wbmpubSSaTB/YNnZ2d7O3tMTc359VwYGCAubk5b9uKxWK0tbUxMjLi1SKbzRIKhSi9KEJ/fz+rq6tehuPsn4aGhrR/KslwUvunwcFBFhcXvf1FNBqlo6ODsbEx7/M1nU4TjUYZHx/3luvp6WFzc9P7vLtSe1pYWFB74njtqRy6eswJKlw95mvOuV8teex3gScCb6bk6jFmdisQdM4979DydzjnXmdmvw68Cdg49DRh4Hrn3JceKEt7e7srNgypnjvvzP8cx+LiIul0urKB5ESpZv7S3j7NM57RUe0YUobR0VH6+vqqHUPKoJqVR1ePqa6+K/x+pWsvrgOHi1S6N5kFxpxzgw8mhL6M+Y9OHvYf1cxfdKdo/yke6RT/UM0qRwOfT97TzewGMwuY2X8GnkV+jPthXwOeYmaPNrN6M3sV0F8y/aNAvZm9wcwSAGaWNbNnV/wViIiIiEhN0ZH2k/du8iej3gbMAzc55/6xcHOlUu8HrgW+BOwC7wT+qTjRObduZteSH1Zzj5klC+v7O+BDR4UIBAIP/ZXIQxYIQH398eZNpZqOPa/UBtXMX9Lp5mpHkDLp0oH+o5pVjsa0n1G6I6r/7Ozs0NDQUO0YUgbVzF9UL/9RzfxHNStPOWPaNTzmjNKYMv8pPStd/EE18xfVy39UM/9RzSpHnXYRERERkRqnTruIiIiISI3TmPYzSmPa/Wdvb49gUOeG+4lq5i+ql/+oZv6jmpVHY9pF14/2oeLd2MQ/VDN/Ub38RzXzH9WsctRpP6PUafef0tuuiz+oZv6ievmPauY/qlnlqNNeg8zsM2b236udQ0RERERqgwYd1SDn3I8+1HWY2UlEedg7fx7uvvt0nmt5OUUqdTrPJSdDNfOX9vYY585VO4WUQ9f79h/VrHLUaT+jdEfUk7G/D5cunc5zNTWlTu255GSoZv6SybRVO4KUqbu7u9oRpEyqWeVoeEwNMrPbzeyNhf93mdn7zWzSzNbM7LyZPfGodezv71c+qJyolZWVakeQMqlm/jI3N1ftCFImXQXNf1SzytGR9hpmZhHg88A/AdcAy8CjgYtHLatLefrPJR2y9R3VzF92d3erHUHKtL29Xe0IUibVrHLUaa9tNwAtwE3OuZ3CY9+60oxmdjNwc/H3WCzG0NAQAOFwmK6uLiYnJ73GlEqlSCQSjIyMeOvo6upid3eX+fl5ID/Epr+/n9nZWTY2NgBoamqitbWV4eFh7wo1ra2tBINBpqamvHX19/eTy+W8I5HFDBMTE+zs7HgZ4vE4o6Oj3nLd3d3s7Ox4GYLBIH19fczMzHDxYv67SjweJ5vNcuHCBe/LSVtbG4FA4ECGgYEBlpeXyeVyAEQiETo7OxkfH/d23i0tLcRiMcbGxg5k2N7eZmFhgYmJBhYXI6TTaVZXc+zs7HqvJx6Ps7AwT/H7USKRwAxyuVVvXZlMho2NDba2tgAIhUIkk0mWlpa8v4bEYjFCoZD3Hhffm93dXe+xQCBAS0sLuVzOyx6JRIjFYiwsLHjLJZMJnIPV1dXCdgGZTJa1tTWv9g0NIRKJJIuLi14Nm5piBIP1B44ct7Sk2N7e8d73YDBAKtXCysqK11mNRiNEo40sLi6WZEhy+fJl77Jfx8/QRDAYPJShhe3t7ZIMQVKp1KEMUaLR6BEZjEwmcyhDA4lE4sgM6XSara0tL0N9fT3Nzc2srKx49WlsbCQSiRzI0NzczN7eHuvr6wDU1dUVtqNVrw3ctx0teNtyPB6nrq7O226LGTY3N9nc3DyQYXl5mb29PS9DOBxmaWnpGBkeeFu+eoaLbG5uHcqwxN7efkmGBpaWlg9luMT6+kZZGR5qeyq9gkWxPeVyOYaGdqivr6e3t5fp6WnvPU0kEqTTaS5cuOAt19HRgXOOmZkZbzsaHBxkfn7e27YaGxtpb29ndHTUq0UmkyEcDh+4nXtvby8bGxtefUKhED09PUxNTXmvJ5lMkkqlGB4e9pbr7Oxkf3+f2dnZsjJks1kaGhoOZOjr62Ntbc17bxoaGuju7j6wb2hubiaZTB7YN3R2drK3t+f9paKuro6BgQHm5ua8bSsWi9HW1sbIyIhXi2w2SygUOnDktb+/n9XVVS/DUfunYpvS/um+DMX9UzkZ2tvbMTOmp6e9dQ0ODrK4uOjtL6LRKB0dHYyNjXmfr+l0mmg0yvj4uLdcT08Pm5ubXm0Ot6fFxUW1pzLaUzl0c6UaZGa3A3cAq8BznHPfW+46Ojo6XGnjlAfnzjvzP6dhY2ODWCx2Ok8mJ0I185fBwRxPfWqy2jGkDEtLS7S0tFQ7hpRBNSuPbq50dowCfWYWKnfBujqV1m+i0Wi1I0iZVDN/aWpqqnYEKVMyqS9ZfqOaVY56drXtU8AK8Gdmlra8x5hZ71ELFv+kI/5ROrRC/EE18xf99dF/SocViD+oZpWjMe01zDm3ZWbXA78H3AM0AheAnwPGHmhZORmBANTXn85zBYPu1J5LToZq5i91dRoOKiL+pU57DXLOXVfy/3HgOdVL8/B2zTX5n9MwNLTOuXOtp/NkciJUM38ZGtLVY0TEvzQ85ozSzZX8p7Ozs9oRpEyqmb+oXv6jmvmPalY56rSL1Aidh+A/qpm/qF7+o5r5j2pWOeq0n1G6I6r/6G6N/qOa+Yvq5T+qmf+oZpWjTruIiIiISI1Tp/2MMrNqR5Ay6dr6/qOa+Yvq5T+qmf+oZpWjO6KeUV1dXa701tEiIiIiUlt0R1Th8uXL1Y4gZdI4QP9RzfxF9fIf1cx/VLPK0XXaz6jT7rSfPw93332qT3nmzM1Bqy757Suqmb+0tOxw443VTiHlWF9fp1WNzFdUs8pRp11OxP4+XLpU7RT+trdneg99RjXzl8uXda6PiPjXw354jJkFzKwi74OZVe0G5zoRxH8aGhqqHUHKpJr5SyQSqXYEKVMsFqt2BCmTalY5Z7JnZ2a3m9mfmNlHzWzdzIbM7AWFadeZmTOz55jZt4FNIGtmYTN7i5mNmNmKmf2jmT25ZJ1mZq8xs3Ezy5nZLWb2N2Z2a8k8o2b2ejP7rJmtA79SePzJhUxLZjZmZm8ws2BhWsjM/szMZgtZR83sFYVpSTO7zcwWzWzNzL5tZj95nPdAnXb/SSQS1Y4gZVLN/CWdTlc7gpSpra2t2hGkTKpZ5Zzl4TEvBn668PPDwMfN7ELJ9GcD1wIbwB7wh8B1wNOAceAVwN+b2Xc55yaB5wOvBp4O3Am8EHgn8P5Dz/vfgGcB/wxEzOxRwD8ALwE+AnQCnwS2gTcV1nMt8Djn3KKZtQEdhXW9GmgC+gs5e4DGK71YM7sZuLn4eywWY2hoCIBwOExXVxeTk5Nsb28DkEqlSCQSjIyMeOvo6upid3eX+fl5AAKBAP39/czOzrKxsQFAU1MTra2tDA8Pe+PmW1tb2d6uZ35+zVtXOp1mc3OTzc1NAOrr62lubmZ5edm7W1pjYyPhcJilpSVvuebmZvb29lhfXwfyXz7S6TSrqzl2dna91xOPx1lYmKd48aN4PE5dXR25XO5Qhotsbm4dyrDE3t5+SYYGlpaWD2W4xPr6RlkZEokEZpDLrXrrymQybGxssLWVzxAKhUgmkywtLXk3wIrFYoRCISYmJmhsbPTqs7u7673vgUCAlpYWcrkcu7v5DJFIhFgsxsLCgvd8yWQC52B1dbWwXUAmk2Vtbc2rfUNDiEQiyeLiolfDpqYYwWA9Kysr3rpaWlJsb+9w8eJFAILBAKlUCysrK1wqjAmJRiNEo40sLi6WZEhy+fJl1tbWyszQRDAYPJShhe3t7ZIMQVKp1KEMUaLR6BEZjEwmcyhDA4lE4sgM6XSara0tL0NxO1pZWSGXy9HY2EhjYyORSORAhqtvy6vs7Owc2o4WKF7J6+rbstpTOe1pefm+DMX29I1vDJNIhKmvr6e3t5fp6WnvPU0kEqTTaS5cuG830dHRgXOOmZkZbzsaHBxkfn7e27YaGxtpb29ndHTUq0UmkyEcDjMxMeGtq7e3l42NDa8+oVCInp4epqamvNeTTCZJpVIMDw97y3V2drK/v8/s7GxZGbLZLA0NDQcy9PX1sba25r03DQ0NdHd3H9g3NDc3k0wmD+wbOjs72dvb804wrKurY2BggLm5OW/bisVitLW1MTIy4tUim80SCoUovZJZf38/q6urXoaj9k/Ly8ukUqmHtH8KBoNMTU0dyJDL5bx2XswwMTHhtc1UKkU8Hmd0dNRbrru7m52dHS9DMBikr6+PmZkZ7/MhHo+TzWa5cOGC16bb2toIBAIHMgwMDLC8vOy1sUgkQmdnJ+Pj495nfEtLC7FYjLGxsQMZtre3vc/942Zob2/HzJienvbWNTg4yOLiore/iEajdHR0MDY25n2+ptNpotEo4+Pj3nI9PT1sbm56n3eH29Py8jL9/f1qTxyvPZXjTF7y0cxuB5accz9R8tiHyHd83wt8AXi0c+7ewrS6wrTnOuc+UbLM14EPOud+18z+HrjLOfdrJdO/BnzDOfeiwu+jhfl/o2SetwMZ59xzSx77WeD1zrlzZvZC4HXkv2R82Tl3qWS+3wR+lPwXiDudc8c+u7Strc0VN8rTcOed+R958Obn58lms9WOIWVQzfwlm53gmc/srnYMKcPQ0BDnzp2rdgwpg2pWnnIu+XiWj7SPXOH3J15lehqIABc4aIj80W3IHyH/2KHpo8d43kcATzGzXMljddw3NOl9QAb4feDRZvYl4HXOuTsLjwWAW4CewheH33DODV3heUVERETkjDrLA5/7rvB76d2GSo9aL5IfrjJ4aJlB8kNlAKaA3kPTD/9+eL0As8AHnHPJkp+4cy4G4Jzbd879gXPuyeS/GPwr8InCtE3n3P9wzj2+kGUP+KsrvdjDAoHAcWaTGtLU1FTtCFIm1cxfmpubqx1ByqS/ZPmPalY5Z/lI+9PN7Abgs+THqT8L+CHyR64PcM5dNrN3A79tZveQ79z/InCO+8asvxd4q5l9GLib/Bj3a4BvHpHjz4B/MLPPkO+M75Mfo37OOfdZM7seWAP+hfwXh43CPJjZjeSP/t9L/oTZLfId95oTCEB91YMTqcMAAAjWSURBVK6Vc1YE9R76jmrmJ+GwiuU3oVCo2hGkTKpZ5ZzlTvu7yY8Tvw2YB25yzv2jmV13lfl/Ffht8uPdE8A3gKc554pnHfw10AV8lPzJoX8LfIp8R/uqnHNfNbOnAW8g34GvJz+s5h2FWbLA28n/JeAS+c578Qox/cBbgXZgh/zJrS8+xmv3TgQ6Lddck/+RB29oaEzjAH1GNfOXoaEJ8sdixC8mJyfVxnxGNaucs9xpzznnXn74Qefc7cD97rDhnNsif7WWV19pZS5/xu6bCj8AmNl54P+VzNN3lWW/Qv4KNleadhv5LxZXmvZHwB9daZqIiIiIPHyc5THtJ87Mnm1mkcI13X8ZeAzw4WrnEhEREZGzTZ328ryU/ImlC8DzgB+v1Su5BINn+Y8oZ1N/f3+1I0iZVDN/Ub38RzXzH9Wscs5kz845d12F1vtDlVhvJRRvLCH+sbq6SiqVqnYMKYNq5i+ql/+oZv6jmlWOjrSfUeq0+0/pHRzFH1Qzf1G9/Ec18x/VrHLUaRcRERERqXHqtJ9RZve7QI7UuHA4XO0IUibVzF9UL/9RzfxHNascy1/JUM6arq4uNzk5efSMIiIiIlIVZjblnOs6zrw60n5GnfbNleSh05cs/1HN/EX18h/VzH9Us8pRp/2M0l9Q/Gd7+wFvris1SDXzF9XLf1Qz/1HNKkeddhERERGRGqcx7WeUme2RvxGU+EcM2Kh2CCmLauYvqpf/qGb+o5qVJ+OcazjOjGfy5koCwOxxT2yQ2mBmk6qZv6hm/qJ6+Y9q5j+qWeVoeIyIiIiISI1Tp11EREREpMap0352va3aAaRsqpn/qGb+onr5j2rmP6pZhehEVBERERGRGqcj7SIiIiIiNU6ddhERERGRGqdO+xlhZjeY2Z1mtmNmf3jEvKNmdq+ZnS/8PPu0ckpemfV6hJl9ycy+bWZfNbPHnlZOuY+Z1ZnZH5vZBTMbMrOXP8C8amNVctz2YmYvNrPvFOr5F2ZWf9pZJe84NTOz68xsq6RNnTezSDXyPtyZ2dsLn3HOzK55gPnUxk6YOu1nx3eAnwN+/5jzP9s5d03h50MVzCVXVk693gX8uXPukcBbgFsrmEuu7nnAY4BHAt8HvPqIL1BqY9VxZHsxs37gDcB/AM4BrcDPn2JGOei4n3H3lrSpa5xzW6eWUEp9BPhBYOxqM6iNVYY67WeEc+7bzrmvA3vVziJHO269zCwLPAl4X+GhvwW6zexchSPK/T0b+Avn3L5zbhn4EPDcKmeSEmW0l58EPumcm3X5qzG8E9WyKvQZ5z/OuS865yaPmE1trALUaX/4+mszu8fM/tLMMtUOI1fVDcw45/YACh9+40BPVVM9PPVw8MjSKA9cB7Wx03fc9lJuLaVyyvmMGzSzuwpDaF52miGlbGpjFRCsdgA5HjP7MvCIq0x+gnNuoozV/Ufn3HhhfNkbgb8Cnv5QM8p9TrhecgqOqlmZq1MbEzlZdwFdzrlVM+sCPm1mi865v6l2MJHTok67Tzjnrj3BdY0X/r1UOAny2ye1bsk7wXpNAO1mFnTO7ZmZkT9aMX5C65eCo2pmZuNAL/DlwkN9XKUOamNVc9z2Mg4Mlvzed4V55HQcq2bOubWS/0+a2QfJj5dWp702qY1VgIbHPMyYWaOZJUseei5wd7XyyANzzs2TP8L0vMJDPwFMOueGqpfqYevDwEvNLGBmKfJj3O93gqnaWPWU0V7+FrjRzNoKncSbgNtOL6kUHbdmZtZuZnWF/zcBP4baVS1TG6sA3RH1jDCzp5L/E3wcMGAVeJlz7pNmdiNwo3PuJWY2QL4xBQrzDQOvdM6NVif5w9Nx61WY91Hkr6bQAqwB/9U5d09Vgj+MmVkAeDvwo4AD3u6c+6PCNLWxGnG19mJmt5A/Me6ThfleCrymsNjtwE3OuUunn1iOU7PCJVZ/gfzJ+0HyX6Jf79SJOXVm9i7gBqANWALWnXPn1MYqT512EREREZEap+ExIiIiIiI1Tp12EREREZEap067iIiIiEiNU6ddRERERKTGqdMuIiIiIlLj1GkXEREREalx6rSLiIiIiNQ4ddpFROQhM7M3mNn5Qz+7ZvbZh7DOsJn9jpndZWb/YmZDZvbqk8wtIuIXurmSiIicODN7LfDzwHUP9m6wZvYO8neVfYVz7pKZNQJdzrl7TyCfAeiOmiLiFzrSLiIiJ8rMXgPcBDzlwXbYC24A/k/x1ufOuYvFDruZNZrZH5jZ3Wb2DTP7TOHxFjN7r5l93cz+zcx+uSTXm8zsrwrz/hsQM7Okmb3DzL5UmP/txQ69iEgtUaddREROjJn9GvCL5DvsIw9xdZ8E3m9mnzKzV5hZqmTaR4DLwJOcc48DXlyyzOedc48H/h3wajN7VGHa9wKdwE875x7lnFsHbgM+7pz7AeAxwOOA6x5ibhGRE6dOu4iInIjCePNfIt9hv3CVeb5sZotX+ekundc593LgGuBzwPOBuwrj3J8CtAGvcc7tF+adNrOnAnXOufcUHlsHxoBsYZVPBH6p8Dhmdj3w/cBbzOw8cBfQj/aNIlKDgtUOICIi/mdmvwK8ivwY9qGrzeecu7ac9RaGw9xrZh8E5oA0+c73Hc65y4dm/27yHe9ipigwAHyz8IXgknPuWyXzPwH4c+fcr5WTSUSkGnQ0QUREHpLCuPFfAa53zn3nhNZ5g5nFCv+vIz/85S7n3CQwDTzRzOoL01sL49AngMeaWZ2ZBYC3AR92zi2THxrztUNPMwn8cMnzRMzssSeRX0TkpKnTLiIiD5qZPZJ85xjgQ4cu+fjCh7DqHwPOm9k95I+enyN/YirAh4FvkT+Cfh64pXAVmI8B9wLfAL4JrAHFE1Gv1Gn/CHAH8PXCev4v0PcQMouIVIwu+SgiIiIiUuN0pF1EREREpMap0y4iIiIiUuPUaRcRERERqXHqtIuIiIiI1Dh12kVEREREapw67SIiIiIiNU6ddhERERGRGqdOu4iIiIhIjVOnXURERESkxv1/U7jTC8fPdI8AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 800x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Draw plot\n",
    "plt.figure(figsize=(10,6), dpi= 80)\n",
    "plt.hlines(y=df.words, xmin=0, xmax=df.word_z, color=df.colors, alpha=0.4, linewidth=8)\n",
    "\n",
    "# Decorations\n",
    "plt.gca().set(ylabel='$Word$', xlabel='$Z-Score$')\n",
    "plt.yticks(df.words, df.words, fontsize=12)\n",
    "plt.title('Word Politicization Score', fontdict={'size':20})\n",
    "plt.grid(linestyle='--', alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}