Week_4.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#Table of Contents\n",
    "* [1. Week 4 Assignment: Generating a Correlation Coefficient](#1.-Week-4-Assignment:-Generating-a-Correlation-Coefficient)\n",
    "\t* [1.1 Subset the dataset into the moderate variable levels](#1.1-Subset-the-dataset-into-the-moderate-variable-levels)\n",
    "\t* [1.2 Pearson correlation $r$](#1.2-Pearson-correlation-$r$)\n",
    "\t* [1.3 Conclusion](#1.3-Conclusion)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Week 4 Assignment: Generating a Correlation Coefficient"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this [assignment](https://www.coursera.org/learn/data-analysis-tools/peer/3bv5A/testing-a-potential-moderator) I've chosen the [Gapminder](./data/gapminder.csv) dataset. Looking through its [codebook](./data/GapMinderCodebook.pdf) we've decided to study the relationship of the numeric variables `incomeperperson` and `lifeexpectancy` taking into account the numeric variable `urbanrate` as a potential moderator:\n",
    "\n",
    "\n",
    "- incomeperperson\n",
    "\n",
    "> 2010 Gross Domestic Product per capita in constant 2000 US$. The World Bank Work Development inflation but not the differences in the cost of living between countries Indicators\n",
    "has been taken into account.\n",
    "\n",
    "\n",
    "\n",
    "- lifeexpectancy\n",
    "\n",
    "> 2011 life expectancy at birth (years). The average number of years a newborn child would live if current mortality patterns were to stay the same.\n",
    "\n",
    "\n",
    "- urbanrate (potential moderator)\n",
    "\n",
    ">2008 urban population (% of total). Urban population refers to people living in urban areas as defined by\n",
    "national statistical offices (calculated using World Bank population estimates and urban ratios from the United Nations World Urbanization Prospects)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "# Import all ploting and scientific library,\n",
    "# and embed figures in this file.\n",
    "%pylab inline\n",
    "\n",
    "# Package to manipulate dataframes.\n",
    "import pandas as pd\n",
    "\n",
    "# Nice looking plot functions.\n",
    "import seaborn as sn\n",
    "\n",
    "# The Pearson correlation function.\n",
    "from scipy.stats import pearsonr\n",
    "\n",
    "# Read the dataset.\n",
    "df = pd.read_csv('data/gapminder.csv')\n",
    "\n",
    "# Set the country name as the index of the dataframe.\n",
    "df.index = df.country\n",
    "\n",
    "# This column is no longer needed.\n",
    "del df['country']\n",
    "\n",
    "# Select only the variables we're interested.\n",
    "df = df[['lifeexpectancy','incomeperperson', 'urbanrate']]\n",
    "\n",
    "# Convert the types.\n",
    "df.lifeexpectancy = pd.to_numeric(df.lifeexpectancy, errors='coerce')\n",
    "df.incomeperperson = pd.to_numeric(df.incomeperperson, errors='coerce')\n",
    "df.urbanrate = pd.to_numeric(df.urbanrate, errors='coerce')\n",
    "\n",
    "# Remove missing values.\n",
    "df = df.dropna()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.1 Subset the dataset into the moderate variable levels"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to verifify whether the moderator variabel, `urbanrate`, plays a role into the interaction between `incomeperperon` and `lifeexpectancy`, we'll subset our dataset into two groups: onde group for countries below 50% of urbanrate population, and the other group with countries equal or above 50% of urbanrate population."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Dataset with low urban rate.\n",
    "df_low = df[df.urbanrate < 50]\n",
    "\n",
    "# Dataset with high urban rate.\n",
    "df_high = df[df.urbanrate >= 50]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.2 Pearson correlation $r$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For each subset, we'll conduct the Pearson correlation analysis and verify the results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "r_low = pearsonr(df_low.incomeperperson, df_low.lifeexpectancy)\n",
    "r_high = pearsonr(df_high.incomeperperson, df_high.lifeexpectancy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correlation in LOW urban rate: (0.46386851310251936, 4.061864446508275e-05)\n"
     ]
    }
   ],
   "source": [
    "print('Correlation in LOW urban rate: {}'.format(r_low))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correlation in HIGH urban rate: (0.62279808470824838, 1.6773499774746408e-12)\n"
     ]
    }
   ],
   "source": [
    "print('Correlation in HIGH urban rate: {}'.format(r_high))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Percentage of variability LOW urban rate: 21.52%\n"
     ]
    }
   ],
   "source": [
    "print('Percentage of variability LOW urban rate: {:2}%'.\n",
    "      format(round(r_low[0]**2*100,2)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Percentage of variability HIGH urban rate: 38.79%\n"
     ]
    }
   ],
   "source": [
    "print('Percentage of variability HIGH urban rate: {:2}%'.\n",
    "      format(round(r_high[0]**2*100,2)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAt4AAAGJCAYAAAC99zSNAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3Xl8XHW9//HXJF3ovtEFKEsR+Ape18JFUPYdHCpiJyjI\nKuKI/ryQqywuoCigXgRUbsAiq1omUGAoyCKLgICyiAIX/AIWhAKF0oVuQJfM74/POZ2T6UwySc5s\nyfv5ePTR5MxkzjeT5Duf+ZzP9/NN5HI5RERERESksppqPQARERERkYFAgbeIiIiISBUo8BYRERER\nqQIF3iIiIiIiVaDAW0RERESkChR4i4iIiIhUwaBaD0B6zznXDPw3cAKwOfA6cAPwQ+/9ygqedwpw\nPnCD9/7WEvd5GVjsvf9EpcZR5JxnA98HPgY8B5wODPLen+Wc2xO4FzjFe39xuY/jvX+qyO3jgXOB\nJDARWABkge96799xzs0FDgLGe++XOedmAhngfWCs9/5959z/Ay4CDi31HJb5PR8LXAF81nt/S28f\npy+cc/ti3/uetTi/SKPQnL3BOc+mQnO2c+6zwI3Asd77a4Lvb4n3/uPB7RsD5wCHAhOAt4C5wJnB\nPL4VMA+42Ht/SuRcS4Envfd79fF7L/v7qxTn3HTg50DSe7+sFmMYiJTxbmyXYgHgrcBMLPj7FnBN\nhc97IHA00NzFfb4IfLXC4yh0NbAf8C9gM+AHwOiC+/Spcb1zbjBwO3AsFvAeDlwJfBm43zk3BJtM\nm4Bdgy87KPh/CLB78PFuwFrg/r6Mp058F/hIrQch0gA0Z3dW8Tm72GM558YAfwa+APwKm8evAU4C\nbu7m/LlKjKlGvoG9FiVqOIYBRxnvBuWcm4ZlTTLe+1ODw7c65+ZhGVicc03AmcCJwHjgH8C3vfcP\nF76bj3x+kff+VOfcn4KvuRc4Hngb+CbwNBZwAtzknDsO2AubuG7HgswZwX2WAB93zo0FfgF8BlgJ\nXItlSDuccz8GjgvO9RLwHe/9jQXf60+BU4CNsazxYuB17/02wTv2x4CvAVOwrMfHsRc0gG8G578q\n+HwH59wjwPbAH7BsyOoePPVfBHYCTvfe/zQ4Ntc5txLLKH0heM4APg3cgb3o3Q3sDewL/BH4FPCE\n9355wfe6J5EsSOTz//Le/8I51wHcA2wFrAN+EnzpAc65C4AxwJXe+9OCx/sOcDKW0XkDy6xdEcmU\nfwd70zAeaPfenxT5Xbgl+F7/DzgyeA53w+aNJ7Cf21EEbyaccx3e+ybn3K7Yz9sBL2C/c3f34DkW\n6Xc0Z1dlzh7qnNso+HhwifuAvcHYDmjx3l8fHLvNOfcaMDxIsIQGRR4TSgSpZfx8XgbexBIwU7Gf\nEcB059w/sCsgNwJf996/55w7AUtqbAosAi7x3v848ppwHnbVdSvgLuCo4OvC14hpWHLn48As4GBg\nOHZl4SRgB+zNGMCSYLwjgf8FpgPzsdeL2V08j9ILyng3runB/3+OHvTe/8p7f0Pw6SnAD7HLZ0dh\nAdNdzrltIl/S1bvt/8D+UE8GJgP/gwVvPwtu/z5wZ/DxCOBd7A/6wYLH/jmW1fhW8BinAmnn3EeA\nM4A5wGHAi8BxwYtP1PVYpuZALMjbCJjmnNsSm3jWYpdro99T+MI2BwtOw8nyM1iJx53AEdhlxp7Y\nMfj/zoLjfwxv997/A3uh2d0593HsxeVa4FFgP+fc1sGxeymtq5/L7ljW7BtAR+TYt7AJ+FvOuS8F\nE+mR2HPwOWAp8NOCxzoS+xk8DZzonPtk5LbdsOfxbCCFvVB9Dft9+BT24nU18BT24ryvc24cFrC/\nAXwF8MCNzrnJXXw/IgOB5uzKz9l/BVYF/zJd3G/X4Jx3gL3hCYLrK7EM+NrIfb8eecxVbJiRL9TV\nz2cn4LfYG7CwtGNv4MfYG43jsfl7BDbXPozN3f8H/NA5NyryWMdiyZ4/Bvf5bOS23YPH/Ab2xupT\nWBB/FLAN8O3ge78ruP8MLLi/CXtjcBKWLLo2+JlLjBR4N67wkmFXP8MvAa9577/uvc9if4TDsQmz\ncHIo9i5+NfBV7/21WOZlY+/9+9g7ZoCnvPcLIvf/off+9977NQWP8xnsRWAWcCGWiTgIywb8Kxjn\nV4H7gvN1RL/Ye/8Y8ApwCPZi8E8s0NsHm8Qf8N6/XXDOvwX/v+q9/2fkeJv3PhOMBSwA7onwKlHh\n81X4c/gTFqR/lvwEfzvwUWySBMtKdKfYz+VR7/2V3vs/Rm7/iff+ZuxnDLCn9/5l7IXvueCcU4Bx\nBY91vve+HXsxAKtZD/3Be5/x3j/svb8Ey8JtTf73Z5z3/iUsoF/jvb8X2AXLhB0C/A5owV7gdyvj\nexXpzzRnV37O/hLwyeDfaV3cD+z5CzPbp5IPrN+l83x1XeQxdwFWdPO40ccv9Jr3/oJgPU54+2Xe\n+/bgKsg72Ny9Epu778V+FuEbr7GRx2rz3v8euCz4PDp3r3+NCK5GzARGYfPxIGzuXkBwpQUredwS\n2Db4Pn+LvXlLAPuX+f1KmRR4N65wkto1etA5d6tz7tLg0w46//GHP+9ojVoYSI4oco5VkQl1Nd3X\ngS0qcXwwVpoQTlwHA6d571cAn8BKHV7AAjsfZEUKzcEm/gOxy41/Ao7BLqNdX+T+pYSlHWFGo6d/\nA48H/x9ccPyggtvvw7I83wD+5r1fiI07gS2ueg94qMjjl/NzKfY8Dy34+jVB9vo5LNP2O4pn2MOs\nS3jpNvozXn8e59zPsbrUt7DylHUU/30IX8h+iP289wH2BB4ocl+RgURzduXn7Ke994967x8NxldK\nmOEPs+e/xb7Pi4rcd0H4mN77v5K/ylioT3O3cy78Wa0JypKexTLRt2BXBxJ0/nmWO3f/P+xqawd2\nJeENup67Z2E/9z2x+burKwfSC6rxblDe+xecc78FvuicexW7fHkoNkH+KrjbDcCPnHO/xLKrp2GT\n2BysFGIdsJdzbjeKL6op/OMMPw//0Hd3zj0Xub3UhHQ78HnsMttIrDbt2865idhlshuwkoVNsXKG\nicC/Cx7jBuwy7AQsO/ES8EtsMp5T5JzhGD9UUD7RE4cUXGb7BxbAngp8P7gc+Cjwn0ArVq4R1sOF\nQe5Y7PsHe+FdCEwC7gsyUYXeipz7MaxGs1Cx5/kM59xyLDsC9rzuAgzDJugtscwTRS4LlxI9z77Y\nc70Muxw6iHwGbzVWF5nC3nCswDJ0T2PlJruTz9iIDEias6syZ5frUuzNw6+cc5tjc/sHsGx8Dii8\nAlCORXT/8yn2fJ/knPNYZ5cx2PP7Mex5W45dpQzLSMqN2Qrn7g6sfv8AYAvg1eC28Dmfib0ZegV7\ns3QP9rt5BPZG8VUkNsp4N7bjscziTKAdWzDzA/LB2vnB54di7+jXAvt7718KMhffwRZ5tLNhd43C\nldvRz+8FngnO/+ki96Xg8zQWkH4Paxf1a2wByp+wGsL/xFaS7wyc6r1/nALe+0eA17BM8QPk6xSj\nlyzXj8N7/zq2WGcXrBauJyvHw/v+GFvpfg32InOY9/49LJC8EsveXB/8/xtgj3DRj/f+OWwhzfo6\nQu99DntBy1Givjv4ul8Gz8U1WK1nd2PNYS/cP8Em1nO893OwuvL7sXq972D1gmB1oN39zAp9F7sM\n+musXMYDHw5uuxq7RPtT7AXrAOxy7dXYAp+jvfevdfN9iAwEmrMrN2cXu3/h8xGObUVwniuw56Q9\n+D6fAT4TjL2rc22gzJ9Pscf6BfYG63jsDcHF2OtEFgu4f4IlbXLk5+6eOA/L/v8Cy6A/BnwwaG15\nHfY6dU4w7gOD+87CXue+GVw9kBglcrladrIRERERERkYKl5q4pzbGVvAtVfw+WHA5733RwaffxKr\nq1oL3OW9/2GlxyQiIhuKztdBJ42rsMvUzwAne+9zzrkTsRKitcCPvPe31WzAIiINpqKlJs65b2OX\nLMLFAxdjbdCidWhtwBe8958GdnbOfaySYxIRkQ0VztdYS7kzvfe7Y3P2DGc7IH4Dq/s8ADgv2DRK\nRETKUOka7xexNmZhoP0QVjuWAHDOjQaGBi3JwGrA9q3wmEREZEOF8/UnvPdhN5rbsbl5J+Ah7/0a\nb1tMv4h2LRURKVtFA++gf+TayOftBXcZTb4lDtgK3jGVHJOIiGyocL6m85XJcG4ejS2yLTwuIiJl\nqHU7wWVYU/fQaGwzjq68R/5SqIhIo+mut3K9iLYkC+fmwjl7FNamrDuat0WkUcU6Z9c08PbeL3PO\nrQ620H4J2yHp7G6+bCj1+cKVQ+PqqXodm8bVMxpX//Skc24P7/395Hv7Pgr82Dk3FNsgants4WV3\n6nXeLqbRfm8aabyNNFZorPE20lih8cYbm2oF3qV6i4I1mf8dthnHncFWsyIiUhvh/NwKzAoWTz4L\n3BB0NfkFtvNfE7b4cnWJxxERkQKN2Me7Xt8laVw9V69j07h6RuOS7jTSz6KRxgqNNd5GGis01ngb\naazQeOONjXauFBERERGpAgXeIiIiIiJVoMBbRERERKQKFHiLiIiIiFSBAm8RERERkSpQ4C0iIiIi\nUgUKvEVEREREqkCBt4iIiIhIFSjwFhERERGpAgXeIiIiIiJVoMBbRERERKQKFHiLiIiIiFSBAm8R\nERERkSpQ4C0iIiIiUgUKvEVEREREqkCBt4iIiIhIFSjwFhERERGpAgXeIiJVkmzNDqn1GEREpDyp\nTHpiKpNujvMxB8X5YCIiUlyyNTsBmAw8W+uxiIhIaalMehiwKbARsBRYF9djK/AWEamwZGt2U2AU\nsLbWYxERkeKC7PYmwEigA8jFfQ4F3iIiFZJszTYBWwBDqMAELiIi8Uhl0uOBidhc3VGp8yjwFhGp\ngKCee0sgUeuxiIhIcUFZySZYTFzxBIkCbxGRmCVbsyOx+kAREalDqUy6CQu4R1HBDHchBd4iIjEK\nFlFOpIoTuYiIlC8oK9k4+LSqc7UCbxGRmEQWUSroFhGpM6lMeiiwGTWMfxV4i4j0UbCIcktgMFpE\nKSJSV2pVVlKMAm8RkT7QIkoRkfpVy7KSYhR4i4j0khZRiojUp1QmPQqYBMS682RfKfAWEekFLaIU\nEak/QR33Jtiuk3U3PyvwFhHpgWRrNoFlucOdzUREpMZSmfQgLOAegc3NdTk/K/AWESmTFlGKiNSX\nVCadwEpKxlLhXSfjoMBbRKQMydbsUGz7dy2iFBGpAwULJxsiGaLAW0SkG8nW7CisvKQhJnYRkf4s\nsnCyKtu8x0mBt4hIF5Kt2YnABOr88qWISH8XLJycAgzD5uSGCrpBgbeISFHBIsrNyC/UERGRGggW\nTk4hv6i9YedkBd4iIgWSrdlmbBFlw13GFBHpLxpt4WQ5FHiLiERoEaWISO014sLJcijwFhEJaBGl\niEhtpTLpEcBk+mmM2i+/KRGRntIiShGR2oksnNyIfpz8UOAtIgNasIhyKjAcBd0iIlWVyqSbsR0n\nRwHrqK+gezvghTgfsKKBt3NuZ+B87/1ezrltgKuwF7ZngJO99znn3InAV4C1wI+897dVckwiIiEt\nohQRqY1g4eREYBw2/66r7YjWSwD7ACcCHwOGxPngTXE+WJRz7tvALGBocOjnwJne+92xb2qGc24K\n8A1gV+AA4DznXKzfoIhIMcEiyq3RlT8Rkapa+f4qgG3IdyupB4OBw4A/AJdgQXfsKvmC8yLwOeDa\n4PNPeO8fCD6+Hdgfe3fzkPd+DbDGOfci8BHg8QqOS0QGuGRrdjR2abNeJnwRkX4vlUkPBja78KCz\noH46Rw0HUsBxWI15qAO4M+6TVSzw9t7f6JzbKnIo+gQvB8YAo4F3ihwXEamIYBHleBR0i4hURWE/\n7kSiLmLuccCXgKPoHHuuBuYAvwHeiPuk1bzEGl20NBpYCizDiulDo4AlZTxWvb5galw9V69j07h6\npu7HlcvlWLj0XdpO25taTfq5jhzUT5ZHRKTigvaAm2DlzfXwWrEpcDwwE+ugEloB/A64GlgUHIs9\nTq5m4P2kc24P7/39wEHAPcCjwI+dc0Oxb357bOFld+rxhSuHxtVT9To2jatn6n5cBYsoa2ruBTNq\nPQQRkYoLupVsBgyjPgLubbEFk58BmiPHF2LNP67Dgu+KqsaLUPhktwKzgsWTzwI3BF1NfgE8iL0T\nOtN7v7oKYxKRAUI7UYqIVFcqk56A7TqZo/ZB93Qs4N6r4Pi/gcuBm7HykqqoaODtvX8Z61iC9/4F\nYM8i97kc+8ZFRGKlRZR9E1yNvBKYhpUGnhzcdBUFrWFrMkARqSupTHojrJSj1i1aE8AeWLvq6QW3\nPYN13buLGuzdUPPLriIilbBk+XugoLuvTgSWee93cc5th7XYeg+7OvmAc64NmIFljERkgEpl0k1Y\nR5DR1HYjskHAIcCXsc1voh4Gfg08Uu1BRSnwFpF+JdiJcvO20/YGBd19tT1wB4D3/nnn3PZAokhr\nWAXeIgNUKpMeA0wOPq1V0D0MOBw4Acu4h3JYS8BZlLeGsOIUeItIvxFdRFkn7aoa3d+xhUg3O+c+\nib2gvRm5fQVqASsyIKUy6WFYlnsItUtyjAWOxNoCjoscXwPchLUEfLn6wypNgbeI9AvJ1uxGwOZo\nEWWcrgC2d849CDwEPIGV74RGYa1hRWSASGXSg7B5YASW4a5F0D0Z2/CmBdsAJ7QSmI21BHyrBuPq\nlgJvEWl4WkRZMTsB93jvT3XO7YhdTVhQpDVsORrpZ9NIY4XGGm8jjRUaa7wVHWsul+Od95bx84PO\noimGK4oXHXy27+nXLFixkHv/9RCPv/4UHbl8VcvIISPYY6tP8qktdxwxfPCwL2M13n22rmMdrXec\nMySOxwop8BaRhpZszU7CdqKs5YKe/uoF4Bzn3Hewzc1OwLLcnVrDlvlYjXIlol770pfSSONtpLFC\nY423omNNZdJjgYlxneOig8/2//WHs10PvuSjWIeSfQuOvwpcvmL1yptue/6e9297vtw8QNkaegMd\nEZHYhIsosc23FHRXgPd+EbBfweEFFGkNKyL9T1DHvQm1ixd3xwLunQqO/xO4DFs4ua7ag+oLBd4i\n0nCSrdlB2KY4msNERGIWtAfcBLvCVe3ERjNWxnYi8MGC2/6KtQT8c5XHFBu9aIlIQ0m2ZodhmW4R\nEYlZUFYyKfi0mkH3UPItAacW3HY3FnD/o4rjqQgF3iLSMJKt2TFY+6pGWvAkIlL3Upn0YGAzLACu\n5hw7GvgicDQwIXJ8DXALtrv5vCqOp6IUeItIQ9AiShGR+KUy6QSW4R6LBdzVCronAccCR2CtCUOr\ngAxwJZ33DaiVWJ8PBd4iUteCRZRbYFkYBd0iIjFJZdKjsKuICaoXcE+77qkswL3A4MjxJcA1wO+A\nd6o0lmKagPeCMSxtb2lT4C0iA0OwiHJLbLGNiIjEINgEZ1Nsq/VqBdwfxhZM7v+X+U9CPuh+Ddth\ncg4W8NZCE7AWWA4sam9pW1upEynwFpG6FCyinErj9NEVEal7qUx6Y6yWulplJbtiLQF3KTj+PLZg\n8nYs6K2FJizYXtLe0raqGidU4C0idSfZmh2LbQmsRZQiIjFIZdIjsLKSZio/tzYBB2AB9w4Ftz1x\n4vQvTJ/1xOxkhcdQSjPwLhUqJemOAm8RqSvJ1uwUYAwKukVE+iyVSQ/BenIPo/LrZIYAn8W2bN+y\n4LZ7gVnA3z402fV4u/gYJIBlWCnJ6hqcH1DgLSJ1omARpYJuEZE+COq4J5PfBKeSQfdI4AtYl5KN\nI8fXArdiLQFfqOD5S6noQsneUOAtIjUXLKLcCpskRUSkl1KZdDP5gDtHZQPuiVj/7S9iwXfoXaAd\nawn4RgXPX0zYoWU58HZ7S9uaKp+/Swq8RaSmkq3Z4Wy4S5mIiPRAsM37RGAcFmxXMru7BbbD5Oew\n8pLQUuC3wLXBx9XUhAX8S4Fl9ZDdLkaBt4jUTLCIcgrqzy0i0iupTDoxa8ZPAbYJDlVyPt0Bawl4\nIJ2vUL4BXAFcjwW/1dIErMay24vbW9rWVfHcvaLAW0RqIrKIUkG3iEgvBBvgTF6xemWlT/VJLOD+\ndMHxF7EFk7dSvZaA0VKSxe0tbe9X6byxUOAtIlWlRZQiIn2TyqQHYxvgbATkEomKbHeQAPYFTsI2\nv4n6GxZw30f15vEmYCXWc3tFlc4ZOwXeIlI1WkQpItJ7qUw6Qb6Ou1Ib4AwGZmAtAacV3HY/cBnw\nRAXOW0y0lGRRe0tbw18hVeAtIlWhRZQiIr0XbICzCRaMViLgHgG0AMcBkyLH15FvCfh8Bc5bTAJY\ngQXbtdpGviIUeItIxWkRpYhI7wTtATfFAuNKzKETyLcEHB05/j5wA/Ab4LUKnLdQ3fXcrgQF3iJS\nUcnW7CbYZK6gW0SkB1KZ9HistKQS/binAscDn8fW3ISWkW8JuDjmc5YS7ihZVz23K0GBt4hURLI1\n2wRsjhZRioj0SCqTHopluQcT//z5Qax++2CgOXL8LawlYDu2iLFiOnIdYG0HG3qhZG8o8BaR2GkR\npYhIzwVlJVOwq4Rx96T+T6wl4O4Fx+dh9du3AJXMOK8vJdls1BTaW9pereC56pYCbxGJlRZRioj0\nTNCtZGNgPJbhjivoTgD7AF8BPlpw21NYh5J7qNxVybAryQqs53a1en3XLQXeIhKbZGt2HDAZ1XOL\niJQlqOPeOPg0rgB4MJDEMtxbF9z2IPBr4NGYzlWooTe4qTQF3iISCy2iFBEpX7Dr5CQsFosr4B4O\npLBFk5MjxzuA27FNb56L6VyFElht+NKBVrfdEwq8RaRPOjpyzPjWLVuiRZQiIt0K+nFPAoYQ0yY4\nK95fCfD/gKOAMZGbVgNzsJaAlaipHhAtAOOkwFtEei3Zmh182Rn7QOdWVCIiUiDoVDIFGIZloOMI\nUjcFjv/BfRcCnBw5vhz4PXA1sCiG80Q1Yz2+w1KSuBeB9msKvEWkV5Kt2RHAZjnlN0RESkpl0oOw\nHSfDDXDiKMfbDmsJ+BmgeU3H+jWLC4ErgeuIvyVgtG57dcyPPWAo8BaRHku2Zsdjl0pVzy0iUkTQ\nGnAyMIr4NsCZji2Y3Ct6cOLw8Sxctfh7wM1YeUlcmoBVWLCtuu0YKPAWkR5JtmY3xV5IFHSLiBQI\nWgNOBMYRTw13AtgTC7inF9z2DPDrM/b4+i9Ovf2H7X08T/R868jvJqlSkhgp8BaRsgQ7UW5BfkGQ\niIhEpDLp0ViWO2yp1xeDgEOwgHvbgtsexloCPgLQlIhlr7ImrDxlwO0mWU0KvEWkW8nW7GBgS7QT\npYjIBlKZ9GBsoeNG9D3gHgZ8HmsJuGnkeA64E2sJ+EwfzxFqxrZuX4Z1JdGVzApT4C0iXQoXUdZ6\nHCIi9SYoK5kEjKXvZSVjgSOBL2FlKqE1wE1YS8CXg2Ph7sDze3GeJmAttlBykXaTrK6qBt7OuaHY\nattp2LursPXNVVi96DPAyd57XcYWqQPJ1uwEbEc1/U2KiESkMumxWC13X8tKpgDHYRvfDI8cXwnM\nxloCvhU5fhKwf/DxXdi27+VIYFu3L2lvaVvVh/FKH1Q7430isMx7v4tzbjvgEqzx+pne+wecc23A\nDGxVrojUUGQRpYJuEZFAsAHOZPoeQ22NxUWHFjzWIizY/j2WlY6aSj7oJvj4ti7O0YSVkiwFlmmD\nm9qrduC9PXAHgPf+eefc9kDCe/9AcPvt2C+RAm+RGtEiShGRDUXquMMNcHrro8A3gU8VHH8VKye5\nEdugpreasJaC2uCmDlU78P471uz9ZufcJ7Ff4Dcjt6+g81anIlJFydbsEGwRZaLWYxERqQepTLoJ\ny3CPoW8b4OwOfAXYqeD4c1iHkjuxNn5dmY+Vl0RLTebnbCezsJRkUXtL23u9HKNUWLUD7yuA7Z1z\nDwIPAU9guzmFRmGXQ0SkypKt2ZF0XkEvIjJg5XI5Wtq/Nol8P+7eBNzNwIFYwP3BgttWYjtNfp2e\nLZK8jHx5yevA6tFDRwI8r1KS+pfIVXG/5yDLPcF7f5tzbkegFVtI8HPv/f3OuUuBe7z313fxMPql\nEonZOyvfZ9mK1TQ1KdFdSbmOHFMnjxqIT3KOxrmK0khjhcYab8OMNZVJj7vwoLMWn3L7D1wvH2Io\ncDhwAvkOJKFlWMD9bvD5NykeeJfqXBIu5lwOvN3e0raGBnpuA4023thUO+P9AnCOc+47wBLsF3IU\nMMs5NwR4FrihjMepxx9Wvf4S1eu4oH7HNqDG1ddFlJeevo//6vn39PbFqWLqdVxzL5hR6yGISAnR\nhZOJRK+m21HAF4FjgfGR42uAW4DLgf0oKBUp8jjFOpdog5t+oKqBt/d+EfYLF7UA2wpVRKooWES5\nJTAYXUkSkRilMulpAO0tbS81wrlSmfRQrK1fbzfAmYQF20cAIyLHVwEZrJVyuKYtWipSKtMd7Vyy\nH/Bb4B/a4KbxaQMdkQFIiyhFpFJSmfQZWJkFqUx6TntL23n1eq5UJj0IC7hHYjXcPQ26p2FX7z+L\nJTFCS7AObVls8WShcmq6O7CNbnJoV8l+Q9s/iwwwwSLKaSjoFpGYBdnnwyOHDg8z0uV+fXj/6Mdx\nnyuVSTelMukpwDbYWrOeBrUfBn6BtUGeST7ofg34IXAt1r3kR1jZSDkSWKD9NLZxTli7PacaVw6k\nOpTxFhlAkq3ZjbGdKJU5EZGSqlkqEjlnNHv9NjZXxZo1D7Z4nxD8602nkl2xDiW7FBx/HmsJeDuW\nQb84clu4yU2pLHexNoDnpjLp2VDdn4FUnjLeIgNAsjWbSLZmp2IvNgq6RaSkIAC+Hrg++LhsQZA4\nJ3KorGxtQfZ6CLBb8D+UyGT39FypTHoMluEeT89KSpqAg4CbsFrtaNC9CguoHwDmYhnrch9zLfA2\n1gbwtcLe2+0tbS+V+9y9teLtMk8rtaaMt0g/l2zNNmM7UWoRpYh0qUT5xnU9ybq2t7Sdl8qkrws+\nrmi2NnKuqZTIKAedSibRuQa7W2vWrQFoAb6MzaFRj2DdT0ZgCY0jg+M/I7/JzSHBsWi2uwkL1t9u\nb2lb1ZNTRTQ1AAAgAElEQVTxFBNeJfj5w7OYt+SVMypZTy/xUMZbpB9LtmaHAlvTwxccEZG+KDdb\nG70/+ez1auDB4H/oPmt+BHAhBRn6VCbtUpn0p7GgvCdz4EjgxB/edxFYvXYYdK/FFkweAnwXq8Ee\nHfm63diwZ3doLbAYy26/ElPQ3ad6eqkNZbxF+qlka3YUthOlstwi0q1I0DaHfEA3p72l7aVq1HwX\nZsrLOWeJ4PN6rNPIYcGx27AWfqU2pAltDByD9eEeuXz1yvD4u1jpzZXYTpGhB8lnupdhgTjk2wGu\nwUr79gR+1d7SpnoQUeAt0h8lW7MTUT23iJQpurARC7xnwvoAuGrtAaNBdg+C/LAWfDUW1+wMHEc+\nG31k8PEng8/DDWlCW2DlJIdFHovhg4exas27v8J6aC8pct6fBf/vhgXZd2FdTSZjGe51ZY6/V4Kf\nzQZvkip5Tuk7Bd4i/UiyNZvAsi29aY8l0olzbjBwNdbzfR1wYvD/Vdjv1zPAyd57XVVpYMWyxsB1\nkaxzn2q+K+wIbLHkOGwL9d9hv5vREpAxwN5YbTXku4yMxn6nD6Rz6e0bwG/O2uuU755217m/7Ob8\nPwOuw64semBxe0ubT2XS7VQhIA6vEpy664nzJo3cWPXdDUCBt0g/ESyi3BL7u1YgJHE4GGj23n/K\nObcvcC72+3Wm9/4B51wbMAOrexWpquBNwUwsGx1uoR7uCLmMfPC9gnwZCFggfhHWizvqBWxL91uB\ntUMHDfluN0NoxoL5x4Dl7S1t6+fdKi8wrZc3QVIGBd4i/UCwiHILtCmOxMsDg5xzCSxYWQ3s7L1/\nILj9dix7qMC7wUTrp7sqWajXcoZUJu2Aj5BfNBkNrOdjme+wq8iD2Ny4H7BJ8DWbR+7/JNaD+z7K\nS1oksMD+7faWtjWl7lQPz1Mjq0Uv+WpQ4C3S4JKt2dHYi4my3BK3lcBWwD+xNQNJYPfI7SuwgFwa\nSLGa7a4ytNXM3nYnlUk3Y7tBfiY4tAQrMwGrsQ4XTl6GZb+PxHpwTyRSvx1YDnwPewPZnWZskeUS\nYFk0uy3xq+a6gmpT4C3SwIJFlD3dDEKkXKcAd3jvv+Ocm4plBKNt2UYBS2syMtlANENYKlvYVc12\n9L6FX18HAXcCC57/g3zQDRZ0nwu8SeduJSOAFHA0G8Y6S7GNa97DtmfvSlnZbYlPA6wr6BMF3iIN\nKLKIchgKuqVyFpO/hL8Ee8140jm3h/f+fiyTeE+Zj9VIv6exjzXcWXDSyI3jfmiA3E3P3sHW46zd\n9I/+9Iv1H9/07B0ctsOB6+/4q0PO4ecPz+r0xafueuK86OfRxyr8+jjG2tMvWPb+Ci466GxIwKKV\nS7jyyfZOtx/38dTvJ4ywpPfy91fwwMt/5c//fox31+Y3gkwAwwcP58OTHfOXLRgLjP3olB3Yb5vd\nNvj9zeVyJBIJRg4eTib1v/9MJBqmgq+R/sagxHjL+R2tslh/ARR4izSYgkWUIpV0IXCFc+4B7DL9\nGcATwCzn3BDgWeCGMh+rUaKXHDGNNdIX+wg610jHedk8l8qkt8b6TIP9nD4A/AtYPW/JK8x+Ojsz\nzBZOGrkx85a80ql1YLQbRjDm8LGYt+SVIbOfzp7c3tL2YBxjpQfPbSqTHoeVODXTOUg7CVtbAHDX\nOfdfHPboPh74PDA0ct/3gGU5WLRyzarb/zL/yfX9vOcve2P+bc93irubgvsvam9pW97T8dZYI40V\nuhhvd7+jjU4v3CINRIsopZq89yuxLbML7VnloTScSI3qEKwcbEFw0/rL5rVaPNaDmu0pWCnHJalM\nena16mxTmfQYrKwkDLgLM6Nh/TbYLpMXYB14oi0B38Q2vGknXwM+v+D/UAJbr7CwvaVtNVJz9bSu\nIG4KvEUaxIp314AtdGu0y4kiA0qRGtVxWNnO6sh9Yls8VtB5JNxuPaxpKdqFpFQwE3msLwTjXhI8\nZsXrbFOZ9DBsoXgYm3Q1120CfIXOi30B5mEtAW8hXya1kg01BbcvBZa0t7QV3fegv3bWaAT99Tlv\n6v4uIlJrydbspCXL3gMF3SKNZjWddz2cE/xfuHhsGn0QBO4zgZntLW0HRj7ucUAffM3JWLnKgm7u\n3mepTHpwKpPegu5L6BLAPkAG200yGnQ/hY35YOw5LrUQsgnrvf1Ke0vbv9pb2haVCrpvevYOsLKb\n64M3SiJ9poy3SB0LFlFuDmzU1KTqEpFGUKT39SXY7oadOo5U4rzFPu7lYz2YyqRnU8H+3alMugkr\nZxmFJRVK7bY7GOtkciJWvx71INaD+9EuTpXAdlx9B6vf7nZX31QmPS1cYBqoamcNZdr7LwXeInVK\niyhFGlc3fbHrclOaQpWqsw1aA07Aat+h9JW84VhLwOOwAD3UgfXengU818Wpwp0lF7W3tK3o4n4V\n0dvguT/3sBa9oIvUpWRrdiNsEaWINKiuAq5GWTxWgSz3WGzhZFeX8MYBXwKOovMGTauxLjpXAK92\nc6rl9KH3dntL20s3PXsH85a8Eh7q0Zuj3gbP/b2HtSjwFqk72omyXxqHtZTbD9sARKSuA+6u9DST\nG2S4x1900NkAkyk9t22KtQScCWwUOb4c+D1wNbCoxNcmsEz4Uizg7vP8edgOBzL76exM6NnPSsGz\ndEWBt0gdSbZmJ2GXX7utQZSGsB22c94MNtyuWqTh9CSTGykpGQckghx3sYB4W+DLQBIrDwktxFoC\nXkfxziRgAfcaYHF7S1vsu6jWoNVjQ5QhSe8p8BapA9FFlCjobnQJYA/gWGCXgtuWbHBvkUC9L6gr\nN5MbZriDf12VlEzHFkzuVXD8ZeA3wM1EWjAWaAbexbLbVa/f7kpfg+dGKUOS3lHgLVJjydbsIGwR\nZXN395W6Nhw4DDgG+3lGeewy+dxqD0oaQ39ZUFdGDXf4xvQrWOAd9QzWoeSPFE9AhOUkK+hD/XY1\n9DV4VsDdfynwFqmhZGt2GJbplsa1GbYIbCbWFi2UA+7DAu6/1GBc0iAapSa4VCY3yHCPC/6VSiAM\neuy1f4BtbLNdwW0PYwH3IyW+NtzKfUl7S9s7ffgWqqrefn5SHxR4i9RIsjU7lq4XGkl9m45lt/ej\n82ZkK4EbgWuBf9dgXNJP1UMpSjSTC/w7lUlPBMZSOsO9EfB54ITf/eMmyAfdOeBOrCXgMyW+NkG+\nO4m2cpd+QYG3SA0kW7NTsBcr1XM3lsHAQVjAXdidZD62m9712KVwkZKiQXQ5NcF1VoryKlZOsi2l\nEwdjgSOxtoDjIsfXADdhNdwvl/jaHNadpKzNbkQaiQJvkSoKFlFuAQxFQXcjGY+1A/wiFnBEPQZc\nBdyLfqZShkgQPSSVSd/R3tL27a5qguulFCWVSTeT32myg+JB9xRsYXELtu4htHLvrXcdce+8h/cG\n3ipxinVYdxItQpZ+S4G3SJVoEWVDclg7wEPp3A5wDXAbVr/9bA3GJQ0qEkRPwTLBJ6cyaYLguy5r\nglOZ9CBgY2wzm1Jbu38A61CSpHNssQj7O/n9oR/c//F75z0cDbqnYuUk/8Ky28srMHyRuqLAW6QK\ngkWU4YuM1Lfu2gHOxjbzWFjdYUk/MoTO5RcHpjLptlKBd616O6cy6eFYH+4RlM5wfwzrULJPwfFX\nsXKSG4H3i3zdScC+wFrghkbt4iLSU03d30VE+iJYRLkFCrrr3QisJvVO4DI6B90eOAPYHbgYBd3S\nS0HA/Aj5+WBZmV93HtY5Z2alg9RUJj0ilUlvhXVcGkbxDPfu2JqGDJ2D7n8CpwAHYG9SC4PuJuxN\nx57YlaMcVjozLb7vQKR+KeMtUkHBIsrw8qzUp6nk2wGOjBzPYXXbVwN/rcG4pMGU03UkqO+ejgXc\nw4HRWG3zEUCXAXU3j9vnjiepTHokVlIyFPv9L5y3moEDsQz3Bwtu+yvWEvDPJR4+sdGgoQAvYIkI\nzYkyICnwFqmAgkWUeoGpT9OxcpJ92bAd4BysHeAr1R+W1KPuAttyuo4ULJJ8C9gG+x1bQR8WTPa1\n40mw6c14rGtPsYB7aPD4J2BvVEM54G4s4H6qi1MsBd4eP2ws7S1t6wBtiy4DlkpNRGIWLKL8APZi\nJfVlMLZQcg5Wp70/+XlwPnA+dgn9xyjolkAQ2F4PZFOZ9E+L3F6s60g5pRM5Sm+JXpa3VrxNb86d\nyqSbU5n05FQmvS22n8AgNgy4RwNfBf4EnEU+6F4D3AAcDHyd4kF3DlsTsQYYUdgWsJqlMyL1pNuM\nt3PuD8CVwM3e+7rdnlWkHiRbs8PpnBGS+jAOu5R/JBu2A3wcm+P6RTtAzdnx6qoLSU8fq2CR5Grg\nQay0A0pkfePeNCeVSQ8GJpFvCQgbBtyTsKtBR2BrH0IrgXbs9+vNEqfoABZjQffpdM7Ed7qjstwy\nEJWT8f4JtmHEC865S5xzO1V4TCINKdmaHYeVl0j9cFj2+gHgv8gH3WuAm4HDsGD8bvpB0B3QnB2/\nYl1I1meVgwByTuT2kqUTBZneA+ki6xvJtF8ffLyBSSM3ppxzpzLpYalMegusvCXsUlJoGvAj7E3o\nCeSD7iXARcBe2FWhYkH3OuDN9pa2F9tb2hYDW1GQiQ+y8yIDWrcZb+/9/cD9zrlh2LavNzrn3gEu\nB9q898XaBIkMKMnW7CbYZdn+Erw1sgSw54Wz/wZwS8Fti7ESk+vop51JNGfHK8hS3wGcHBxaQpHy\nkK42wCn2mMU+jurJpjndbL4zDPgEtnX7q1iAXOjDWA/u/encfek1rCXgHOC9Et/OWqwH9zslbheR\niLIWVzrn9sK2fd0PuB271LQf9qJ2QMVGJ1Lnkq3ZJqzllhZR1t4ILIN9NLCl/3enze/+CVwDzKWP\nNbWNQHN2vNpb2r6dyqTBOnqsxjLCxe5XdulE3CUk0cdJZdIJLBEwDvgGFlAD3IW1ygztinUoKexX\n74FZ2O/O2hKnXAe8XSrgLtZ7fNLIjaeX/Q2J9FPl1Hj/G3gJuAI42Xv/bnD8Pqw2UmRAChZRboUW\nKdfaZliQ2akdYALIWYB0FQOoHaDm7MoIgu82IA3sDezdmw4i0H0XkkgZS486fwT12xOwoBtsvcn+\nkbvsjwXTH8Iy3B8qeIjHsQ4l95c4RRPWl3txORnuIpn4c7v7GpH+rpyM9z7Acu/9m8654c65bbz3\nL3rv1wEf78nJnHODsZ64W2Lvlk8M/r8Ku0T/DPZCocyh1DUtoqwL04FjsEzuBu0Af3DSLkd//7JH\n0jUZWW3FNmdL0cz03ljNNwTlH+F9y8led1dCEg3KscB7ZlePncvlaGn/2nhsv4ChlC53S2AZ8KuB\nTQtuuwfLcD9Z4mubsZaHC9tb2kqVnBSlBZQinZWTqTsYuCP4eBIw1zl3Ui/PdzDQ7L3/FPBD7N3v\nBcCZ3vvdsYlhRi8fW6QqtIiypgZjc0TYDvAAOrcDPI+gHeCkccNrMsA6EOecPaCVWNw4BWsX+oHg\n43SR+5R6vGnATuQD92K3dwrKoXjwmsqkh6Yy6c1eX74ArDPKYDYMuudjrQAnYguNNyUfdK/FFhgf\nDHyN4kF3M1bb/WJ7S9urPQ26RWRD5WS8TwL+E8B7/7JzbjrwKJ3rxMrlgUHOuQT27nw1sLP3/oHg\n9tuxS2E39+KxRSpOiyhrZjzW2uyLbNgO8DHsqlm/aAcYgzjn7AGrIAgeAnwB25VxEOsrmdbv5Biu\nGyi5ADLo/3001sZvLbAcWEAPN49JZdKjsb+HYcC6RKJk/mxicL4v0nlH1nfJtwR8o8TXJrArRwvb\nW9q0GFckRuUE3oPovBhpNb1/cVuJ1cT+E6tDS2LZqdAKLCAXqSvBIsotsBdglUJVz3ZYOcmhdM4S\nrgFuwy6bP1uDcdWzOOfsgW4I+Z7XAP8T+Xg5tiNj9Pey6MLdIOj+RnDfsKvIcuCU9pa2B8P7FVuQ\nGBxrJl+7HUbaxbqTgM1TXwY+h2XBQ0uB3wb/lhT5OoLHfgcLuEstqhSRPign8L4ZuNc5l8HeBX+O\nDVt0lesU4A7v/Xecc1OB++g8MYzCJofu1Gvgo3H1XL2Obf241qxdx6Wn703nLlu1cenp+/haj6GY\nOMfVkcvxzItvc+/jr/LPzp1JGDV8MLt/fCq7f3yzwWNGDv0s8NlqjSsOuY4cVP4XKc45eyA7Assa\nT8SC3EXAfwDLsAB4fHC/YcG/NcCDRdr5TcMW/w7BAtsElvFeg5WCdBJdkAgsTGXSm9O573a4tqTw\na8MFk9HyK7Cs9hVYOcy7Jb7XBPbau7Bwh0kRiVc5gffpWC/Y3bGJ4mLvfW9LQRYHjwH2jnsQ8KRz\nbo+g9+xB2CKP7tQ+AtpQDo2rp+p1bOvHlWzNjsBe6Gr+BuHS0/fxXz3/HlfrcRSKcVyd2gEW3PZP\n4Orlq9bcettDL62+7aHur8z3clylgprYzL2g4stY4pyzB6RImclC7CpsWHoxOji2FPsdXYoljNZh\nAe7GqUx6WkHwPTX4unXB4ySwq7uzS5SkNEfOtRk294TB8EkUtAZ8/u15YIH1pwoe6kVsweRt5F93\nCyWw1+KF7S1tNZ/jRAaCcjbQyTnnnsN2qkoAOOd2j9Rl98SFwBXOuQewd/9nAE8As5xzQ7BLxjf0\n4nFFYpdszY7HLjMrA1RZRdsBYgHHvVg5STXaAW4Q1FThnLGLec4e6FZjQetIbFHifPLlJMuC/wdj\ntd6bYpvUFOuEspx8bfdK4PBoiUnwNcOxcpJodjsaDBe2Bvw8cPD/PnoNdA66/4YF3PdROmGgDLdI\njZTTx/sSrBZ7Hp3/iPfq6cm89yuBliI37dnTxxKppGRrdlO0iLLSpgPHAvtSpB0gcC3wSpXGUqzf\n8W1UMPNdKXHO2QNVpNb6C9jr5FtYpns1VjKZxN4sbon97q4j/1yHfb5JZdJvYx1HcsAqLCs+Owy6\nU5l0E1ayMoZ8V5Ku5pwEMBYrfynsjHI/9mbxiW6+vbCGW3ObSA2UU2qyP+DCTRhE+rNka7bp12fs\nC5bh0gtT/AZj7cuOYcPNO+ZjwfYN2KV46R3N2TEIaq3/DFzChosmjya/SU0H8C8sMz4YC8pXYoHx\nbuRrwhPAI8HjjsR6akez213NNyOwn+tWwcehddM3/XDzE68/fSjWNawUZbhF6kQ5gfc8tDOfDADJ\n1uwQYIuOnEodK6C7doBXY+s7ahUUzMfKS6KlJg2X7Q5ozo5Je0vbg6lMejadN7Q5ivzvcLSzyHjs\nNXUUVje9GAt4x2AZ7xywWyqT3pv871Z3v+8TyLcEHB05vhpbLHnFlz52+D1PvP50NOiOrlNIELQt\nVMAtUh/KCbyXAM865x7GGukD5Lz3x1duWCLVlWzNhjWcEi+HBQ6N0A7wMmxM0LhBN2jOjlVBlxGA\nh8kvkkxgtfRnAd/FAuIpWDZ7Mfamcsfg65ZRugVgoanA8Vgd99DI8WVYO8Brg8cvFK5TSABzge+p\nLaBIfSkn8L6D/C5oIaUEpd9ItmYnYBksZYTikcDWbRwD7FJw22Jsx8nZwNvVHVZZGjngDmnOjllk\nO/fdsMzzWmxBJVirxsfIl6MswN78nIW9qfxvrORkDcWvpEQz1B/EWgIeTOerFm9iG960Y2UsxYTr\nFHLBWPYDfg1oy3aROlJOV5OrnHPhSusENtlMq/TARKohWEQ5CgXdceiuHeBVWEa56CYjEg/N2ZUR\ndirBgupxWPC9DDg/WIx5I5DCAua7gP8L7v8z7I0mbBh0hxnq8Ge1VcHt84DLseC+VEtAgq9dh/1t\n6U2WSB0rp6vJecDXsEUji7DWX48BO1d2aCKVE+xEuSX2e60Xqj54e+m7AKdh7QBHRW4K2wFeA/yl\n+iMbmDRnxy+VSZ+B1XkPwRZSjgs+voX8Jjc3kv89Lwyw52MZ6amR2zbHSkkmYhvwRD2FlT7dQzfz\nU5N1jHy9vaXNpzLpGyjY9bL871JEqqGcUpMvYFvQXgycE3z835UclEglBYsot6Q+N+9pJNOBY793\n2cNg9aihlVgQcg3VawcoeZqzYxTZTCes3QZ4BruScFDw/2XYVbNSpUrRHvF3Y6UjaexnE/UY8EvK\n61vfASycMmoS7S1ty6FzPbqCbpH6VM7K9ze89+8ATwMf897fB0yu7LBEKiNYRLkVCrp7azAwAwus\nfw/sH2kCMx84F9sx8Uco6K4VzdnxG0I+6E4AHyFf470/+TrtYsLa6yasS8nJwHl0DrrfwRZMHkX3\nQXe42+S/2lva3im8sb2l7SUF3SL1q5yM9zvOuS9hu2F9wzn3OtbAX+pYsjU7DaqyPXXD0CLKPinZ\nDnDbzcfywqtLT8bKSvTc1p7m7Hi9CjwAbBN8voLOO6x2ZzS2A+4E8sE6WD32HGzdwxt0v7C3GWsN\n+EZ7S1u53VFEpM6Uk/E+AZgUZE1eAi7F2iZJnUq2Zs/Aerxef/09z9d6OEUlW7PTFiwqtTi/Iueb\niu0gp8BwQ2HtaTHbAT/GdsX7Jvmgew1wM3BY65HTwS6f67mtD5qzY5DKpEelMuktsYD7F8DvsKs4\nbwKPk1/sWKrn+6bY8z4bC7zDoHs1VpqyJ3A2Vl7SVdDdBLwPzGtvaZuvoFuksZWT8d7Xe38BgPe+\nFcA5d3JFRyW9FmS6w8U1PPzU61zzh+emzb1gRt1cegzeGBz+k2se48X575wx94IZ51XwXFpE2bVo\n7eldWEDQXTvA2ViZST22AxTN2b2WyqQH/frQ8/nKLadvS74cLXxDWdidJNoGMGo74MvAZ+ic4V6M\nJUQuo3RLwKgEFqS/2d7Stqon34eI1K+Sgbdz7hTsEtlXnXPhQrQcFsAciW2jK9IjhW8MgMOTrdnr\nKvHGQIsouxXWnoYOxP6+D6V4O8CrgVtRO8C6pDm7d1KZdAKr3x4NDFu19j0oPWfML/Ex2GLjE4G9\nCo6/DPwGu0JU7t/OWmx79+Vl3l9EGkRXGe8XsYkk3J2L4P/3sEyY1KG5F8x4KdmanUMQ3O76kU25\n8JQ96ybbXS3BIsrNUJa7HIOx+tPxwA6R42E7wKuAR6s/LOkhzdk9kMqkR2IB93Dy80RPyzgSwB7A\nV7DnPuoZbAObP1J+GVYHFnBvsGhSRPqHkoG3934uMNc5lwGGee//5pwbA+zovX+waiOUHpt7wYzz\nkq3Z6wBm7rPdvFqPJ6rwjQEwJ+5sd7I1OxELJMMt4PvDboSVMAXYiPyisdBKbNHXtagzScPQnN29\nVCYdvskchdVO5+jdm/Mtgb2xeWzbgtsexgLuR3r4mG+3t7QV2wZeRPqRcmq8j8Heye+H7a71Pefc\n7t77syo6MumTYsFspNNJTTPg4RuD047ead6UCSNiq+9OtmYTWJZ7BHbJt7B2WSy7fQj2d71DwW3z\ngd9idagrqjwuiY/m7IiCUpKNyGefexNwDwMuBD6N/S2FcsCdwCws012uBLb75ZvtLW1anCwyAJQT\neCexnqV47193zu0L/B0YkJN4owoXNAYfz6nkgsZyVCDL3YxloQZhwXe0dnl/rGXXQM58l2wHiHVV\nuAq1A+wvNGcDqUx6OBZwjyQfZPf293ssVid/LBbARy0HbgDO78HjNQGrsNaAXW0FLyL9TDmBdzNW\nAxcu8hiKXpwbSjUXNNZCsjU7FNuMQosoN+SAo7EFk0Mix9dgb0auBp6twbikcmKbs51zx2DBJli2\n96NYtvfi4DGfAU723tfFWopUJj2IfCnJIGyMfRnbFOA4IIU9p6Ho8zkYOBi7WtTdm/smrOb+zfaW\ntnf7MC4RaVDlBN6XAU84527BApuDgF9VdFQDRL2UfjSyZGt2FFbLHX1xnY+Vl0RLTQZStruJfDvA\nTxbcthhrBTgbtQPsr2Kbs733V2NvznDO/Qq4HPg+cKb3/gHnXBu2k+nNcQy8N1KZdBOWkR6NvTkI\nF0j2JUH0AaxcLUnn18lFwHNYR6AtgnPlgnNPput5Zh2W4VanEpEBrNvA23t/oXPuIWA3LEt2pPf+\nyYqPrJ/raelHX4L0aixorIVgEeV4ime0LsMyujBwgu4RwGFYwL1FwW1qBzhAVGLOds7tCOzgvf+6\nc+5s7/0DwU23Y29wqx54R7qSjCAfZPd1c5mPXf7EbIA/FBx/FXvTcRO2mc10LOsf7mC5DNtYp5S3\n2lvalvZxbCLSD5ST8QZbtT0eOBcL3hR490FPSz/iqM+Odjpp9KA7WEQ5FctuFQu6S21s0V9NBY4C\nZtJ5K2u1Axy44p6zzwR+EHwcLelaAYzp42OXLdKVZDT5PuVxlD7ugWW4d3rmTR89/hzWoeROOgf1\nT2Bdfw4JPi+2hiQBLMHaA9ZFKY6I1F63gbdz7ifYC/sngJ8CxzvnPua9P7XSg5N467MbPeCG9Yso\nt6BzR4GoYjsx9lfTsfrbfbHyktBKbLHXtVimTgaQuOds59xYYDvv/f3BoWigOwooN5Pbq+Czo6OD\nFatXsmrte/z8wO/R1NTc/ReVYV3HOv7+xv9xz7yHeH1552T1thO2Yu+tP80HN/7A9olE4sJSj7Fo\n5RIAJowY54BTATpyHWzUPJTxw8bS1NRU6kvj1kiBfSONFRprvI00Vmic8ca6fqycjPcB2AT+hPf+\nnWCF/NMEk4z0XH8t/ai0MhZRFu7E2B+7mYQLuY4BPlRw23ws2L4BtQMcyOKes3cH7ol8/qRzbo8g\nED+o4LaulP3iFbQAHI3Vbg8j3gX9Q7G59wTyV8fAgoC7/2vXL+930cOXuxcWvRweL/cKWrjF++vt\nLW3vxzfcbuVonIXljTRWaKzxNtJYofHGG5tyAu/CmrmhRY5JD5Vb+lGLIL0eF30mW7OjgU1onHfI\ncRt/25/nAdyH2gFK1+Kes7cD/hX5vBWY5ZwbgnXEuaEPj91JULc9Fqvb7msLwEKjsXaax2BlOKE1\nQBbb1n3eVmOnRmtNSl1BKwzG12F13Fo4KSJdKifwvh64DhjvnDsF+BLWEUH6qNzAtpr12fXW7zsY\nR4W5QNQAACAASURBVFeLKKP6YzeTDxK0A5z755cgH3SH7QCvwupQRUKxztne+/8p+PwFrGtOLFKZ\n9BDs77uvu0mWMgkryToCC+hDq4AMcCXFF0ZOx2q4wz7b4RW0Q8jPMXcCP9LCSREpVzldTc53zh0I\n/BvYHPi+9/7Wio9MOqlG9rne+n2XsYiymP7QzSSBBTbHsmE7wEVYEKV2gFJUI8zZqUy6mXywHc3I\nxxlwb42Vk8yg85qQJViHn98D75T42pOwAHsLrGNJGFhPJh90r8P+TrOpTHp+e0tb3VwhFJH6VW5X\nk2HYVrtrsVZK0k/UY1kJbLATZU80ckeTLtsBHn3w9h+85g/P7YnaAUr36m7ODuq2x2JdUKJbt8dd\nuvhh4CvAfnSuIX0NKyeZg21iU1SwYHJ/LNO9DHuDMAYL0vcOxh1mwacAlwCrU5n0nPaWtppfIRSR\n+tbtkmvn3AXAfwPPYxmUc5xzZ1Z6YFJcsjU7LQyWy7nvgkUru7r9DOyy9PXJ1uwZQfA9J3KXmiz6\nTLZmN8KyVT0Nuk/CeuteHHzcKKYCpwMPAN8jH3TngLuxUoEZu35kU1DQLd2oxzn77VWLwWrFJ2I7\nqFZiLcKuWOnVDVjgHAbdHns+9gd+RxdBd4HBWIegDmBh8O8TWHkJ5HeCDf8mD09l0mXNzSIycJXT\n6ygJ7Om9/6X3/iLs0trRFR2VFFUYKJdz359c8xjF7luirGRaUNM9E5hZi/ruYBHllvR8tXOxjiZT\nS9y3XuyI7Sj4R2xb6rAH90rsUvh+wMmoB7f0TN3N2avXrYHKLIxuwjqr3IjVau8Sue0JLPN9KDAX\ny/53a8KIcWDlKFtgpTrhNvTjsO+hDZsjTwYWxPA9iMgAUk5G8U3s8uDC4PMhqLa06npSf93XWu0a\n1nRPwi7r9ufOHIOx2tFjgB0KbpsPXINl7EpfqpBGVo32WQNhzh4CfBb4MvZGPeoeYBa93DQoKDUZ\nB7yBdVIahP3cxgG/j9Ryv5TKpDt1nFKdt4h0p5zAewnwD+dcFqvFOwh4yzl3JZDz3h9fyQFKZdRT\nL/FgEeXmdK777Kl672gyHvgC1s5s44LbHsUy3GoH2P80Y/Pme8C7VOcNVX+es0dif0fH0vnvaC1w\nKxZwv9iXE3TkcmB/hyuxDPcabCOq1Vi2e732lrbzUpn0dcHHCrpFpFvlBN43Bv/Cy4SPk298PlB7\nKlddOYFydKFkeN81azsA7i0WVNfDNvLJ1uwg7JJuHIso67GjSdgOMEm+JhTsxfxWLOBWO8D+IYGV\nPrwf/HsXWD73ghlruvyq+PXHOXsi9nf0RfIlWWDPcQar7X6jj+doBlZ8aNK2YO0YD8fexIBtSFU0\no62AW0R6opxg5y3v/R/CT5xzCeC/vfc/q9ywpJiuAuVi/beTrdlxWC/avYPFkxvUbNeym8l7q9eB\nLaLsqa62ha+HgLsJq6s9BrUD7M+i2ez3sAzpqrkXzKh1cNuf5uwtsHKSw+j8xnUJ8NvgX197aDdh\nAfyCcMfJaCY7pABbROJQTuB9vnPuM9huZZtjC1gWA404iTe8HtR0/xnYe/CgpuixmvXkLpRszY75\n32/v1Zsvredt4UcAn8Mycxu0A8SycrehziSNKEF+S/Aw0K5FNrsc/WHO/hBwInAAnZsAvIG1BLwB\nC5b7Ivx5vtXe0rZBCZACbRGphHIC7x2B7wP/h2V4TvXez+n6S0RKCxdRNjWV01SnIUwFjsI6HUQv\ng+ewhV7XAH+twbik98JsdlgysgrLZjdCDX4jz9mfxDqRfKrg+ItY/fatlNmdpBtrgYXa4l1Eqq2c\nwHtrbBL0WPZkN+fcHd57dV2oEyXqvx8Mjk2PHKtpBqcfLqLcEVvktQ+ds3IrsX7o12CLsqT+NdO5\nNnvF3AtmNOqViUabs5uAfbGA+8MFtz0J/Bq4j/jq09+KY4v3sGe3MuMi0hPlBN73A6d5769yzm0E\nnAM8Te9qc6VCitV/z71gxnkLFq0898Rz755ZB0H3IKztV3MMD1fLRZTltAOcgy3GkvrUjL3xC0tG\nVgErGySbXY5GmbMHYy0BTwAKN565H/s7fyKmcyWwcpu321va+hzApzLp9WtqtGOliPREWaUm3vtX\nAbz37wHfcs7dWNlhSTkKt3sP/48enzJhRM23g0+2Zodh5Rhx9jCudsCtdoCNq5kNa7MbNZtdjnqf\ns0cAR2BXiyZFjq/D3lBfjmXr49CEbfu+oL2lrVd/m4WZ7eDzTmtqUpn0dcp8i0g5ygm8VzrnZgHb\nYDWsPwNOreiopFvFupiUOF67QdoYxgKTadw2ZmE7wEOxDF1oDRYkXIXaAdaTMJv9/rChzWA7C67o\nR9nsctTrnD0B+1s6EhgVOf4+tljyCuJ7Q53A3mS93t7S1usFsMpsi0jcygm8Z2F1tDsDy4HXsRZO\nh1RwXNKFUjtThh9Hjy9YtJIpE0ZUb3ARydbsFGwHvYYKejo6cgB7o3aAjaAwm71i7gUz3g9vnHvB\njGW1GlgN1d2cff0ztwL8ic4tAZdh47oWKwOJyzosw92nmvauMtvasVJEequcwHua9/4y59xXvffv\nA99xzj1V6YFJ4woWUW4BDKWxgu4RwOfO+vUjULBDHWoHWA+asCsn4QLIcBFkI/2OVUPdzdkPvfI4\n5IPuN7EWh+3Eu5NnAljc3tK2MMbHLEo7VopIb5UTeK9xzo0JP3HObYtlFHrMOXcMVtcHMAz4KPBp\n4GIsQHsGONl736hlCVXR1S6WhcenTBgxvdhjVErMiyirpVM7wIVL17cHDtsBXo3VcUt1hdnsaKeR\n97v+EiHGOTtm87Bs/FysVCsu4QY4r7W3tMXRahCwgLqrzLYCbhHpjXIC77OwS4RbOOeywC7A8b05\nmff+aiyIwTn3K2wRzfeBM733Dzjn2oAZwM29efyBpNQulkWOn1utMSVbsyOAzap1vhhMB46joB3g\n0CHNvL963dXYJXC1A+y7qcH/XdXvhs9/YaeReggYG01sc3Zcjv9EC1f8LXMw8a71CDfAebO9pW1V\njI+7njLbIhK3bgNv7/0dzrkngP/EMlBf8d6/2ZeTOud2BHbw3n/dOXe29/6B4Kbbsf7MCrwDhZ1L\nokp1K6lFF5Nka3YC1u2j3q9WdNsO8PyvffrMUy66v2pvWPq5k+jcc/2y4GNlsyukEnN2X31kyvYQ\n79ywDuvHXfENcBRwi0icug28nXNDsY0NHPAN4JvOufO9932pcz0T+EHwcbTF3ApgzIZ3H5hKdS6p\nJ0E991RgOPUddHfVDvCvWP/te4GOYRsNOrPKY+uvpmJBdwL73dgPuA5rFadsdoVUaM6uJ2+3t7TF\nuRhTRKRqyik1uQRYiF2WXwtsC/wG+FJvTuicGwts572/PzgUXRg1CihnR7F6DfBiG9eCRSvZZmqn\n9yDTFyxaeW4vO5RU5Plau3Ydl52+T58e/NLT94mrX29R899czr2Pv8qjzy5g7br8SAc1J9hx+yns\nvePmbDFl1M5YB4iqjau36nlcuVyOjo4czU0JBg1qYuny97nk+r/T1JQgkbD316cdvdOjVe6yU49z\nRZz97IuJdc6uEwngHayspB5/piIiZSkn8J7uvf+4c+5A7/1K59zR2CLI3todW7AWetI5t0cQiB9U\ncFsplX7h6o0cZY6rq/KR0Inn3j0NuL7gWG92oCx7XD2RbM2OAjbpy2Ncevo+/qvn3+NiGlJUE7AX\nVk6yc8Fti4DZa9flZv/lmTfe/sszb1RzXH1Sh+NqBtZe/p19n//yj++egpWMrM9kTx4/gpfeWL7+\nqg222LeaV20q8rvfAOKes2spln7cIiL1opzAu8M5F+29ujF9axG3Hfz/9u49zq6yvvf4ZyZkggkB\nAwQpBtog5YfWWgVvtQUiVESPw9TGJASqKC2nRuhRmeMF1N5rbO1I9RxMLQcFtZCLUcaRQ6CAkgBH\nAbGIgj8EohAkQkIgF5KQy5w/nmdn9uzsuezZa629npnv+/XKK/s2a39n7Zm1f/PsZ/0eHqm63g1c\nEZ/jAcJCCuNWZ3fvPwNnAi8MN31kuM4lY3zeEYv9BrZ1JGFKUNlGnqYR9td7gKNr7lM7wOZMIoye\n7mRgXvbzfT1duwH6errqziEe6iRgyVXWx+xW2UsY4c59HreISFFGU3h/HrgZONLMPg+8k4H52Q1z\n93+puf5zYM5Yt1dmtcVuLLovjHdvIi58M8xJkpkULVnNFY+tAo8h/NyUqege1A6w6na1AxybSYST\n13ZQp8hulAruwmV6zG6BNsLx8WlNKxGR8WbIwtvMFrj7MkKnkR8SPrpvBzrd/b6C8iWrttglnFR2\nZtVDZjCK1dqaLVrWb9wGdVa5bHS7nd29BwFHNZMlB68l9IUf1A6QsCjHN1A7wNGaxECBvY0mimxp\nnXFwzK5MK8m0H7eISJkMN+L9d2a2ErjR3U8EflpQpuTVW9IduJ0wxWEToegGWJXCaGDJln4frh3g\n44Ri+xtkuyLeeNNOGNGuFNqb1WFkXEj5mN0PPKlpJSIy3g1XeN9BGAVrM7Pagqvf3VNambAM1gGV\nOdvPEIruj9Y+KMu52ACxg8SY5op3dve2E6aWdND6ovswQjvAhezfDvAuwvzt79L6nGVUOcFwJ6HQ\nfq6vp0vz3MefFI/ZbYROVk+VaVrJ/GWLZoN6eItI9oYsvN39fOB8M/u2u59VYKbkDXNi5LBztvPq\n2z2WueJxPvdvMXgKRyucQDhZ8izCaHfFLsLS01cTTpyUAW3xX2X6yBbC9JHSFDaSvcSO2ZVl3kvX\nreRbD6yC2FFq/rJFK5cvWFK69RNEJF3DzfE+0d3vBXrM7JTa+6tWm5Q6qotdCCPZfT1da4cqfOtN\nT+ns7r0dWJfF6Hcj2+js7j2QMNLdKu2EE27PA95Yc99G4Nr4b0OxsUqruuPIdsJy6yq0J5iEjtn9\nhIJ7c6uD1Jq/bNHsY2cMOvTNnb9s0VKNfItIVoabavJ+wupnf1Pnvn7gtDwCjUOLiPuq6iTL0RTC\nRxIWwhi27WDW4kmUL6U1XUumAX9CGOGuLfx/Rhjd/g4Tux1gG+EPk52EE9F2AFv6erpKNWooLVH2\nY3Y74RyXUk0rEREp0nCF9/Fm9t0h7puIi1I0JE4bWQi8jPBms57QSnAhdYrpmukplR68lQJzTJ1I\nxpD5UOAIip8nPYuwqt48QvFdoXaAA639dk6dcgDAE4TRbM1ll1qlPWZPbjsA4OGydytZvmDJ2m89\nsIpHNz1WuWmlRrtFJEtt/f31Bx7MbE68WPuANsKJOrfRGmVdjW5frjhtZAWhgH5ZvP8xwijuIwwU\n1PutRBm/dhZwWc3251UuNFiAj2p/tWJRnIvPOdE/d829NxPaAVZn3EbYf1+nBe0AW7xCZKW1X2Xa\nyNaqEyFL/7NfMmXNlYsSH7Mhrdeif/6yRcdCMidXJrVvSScrpJU3payQXt7MDFl4l1hZX6x6hTeE\nKSMzCIX3dMLId8WQS8BXn2hJ6EoCg0/WHO3Uk2H3V2d37yTCHwSTh3pMxirtAN8LvLzmvlK0Ayyw\n8K6cuFoptLcRpo0MNZpd+p/9kilrrokopdcipayQVt6UskJaeVPKCunlzYwK7+wMylW9NDxwK7AE\nOJsGiudKa8FoRc3dQxbtw+Wq2f40wqI4RezPSjvAc+LlancRppPcSgnaAeZYeLcTurHsII5mA7sa\nOAkyiZ/9EilrrokopdcipayQVt6UskJaeVPKCunlzcxoloyXBsXR6sqJTLdW9etuqK1f1VLzs0d6\nbIP52ggj8YeQf6F7AqE7SSc17QDf+MrfmPz9nzzZxfhtB1iZn/08WqhGZMJQH3ARGYpGvLPTD7TV\nTDOpGO3o9JBqp56MdapJZ3fvFMIc8jwX02gnLFd9HvCGmvs2AtcA1/7bx0+/s4VzqYfUxIh3pX/2\nDkKx/WzGC9WU+me/1SHqKGuuiSil16KprPOXLRp0rC6gD/iE2bctkFLelLJCenkzoxHvRIxlEZxq\ncRXKI4GDyW+UexrhDec9wNE19z3IQDvA8dD6rjI/exdhOtFOwtSR7eqfLTIxxZHuQesxqA+4iFRT\n4Z2xYVatzGTbY/m6zu7emcChhL8w8yi6x3s7wEmE72UnA0X288AOtfUTERGR0VLhnYNmR6ez0tnd\nO/lLl5wOoatKHqOwryN0JxmqHeDXgHU5PG+e2vfu7YdQYO8gdhvJeMqIiIxDyxcsWTt/2aJBAy8a\n7RaRaiq8czJcwV3Vq3vdaB4/FnExnJk5TOEfqR3gVwntD1vWDrBBlXnZlaXWtxz9kun09XT9oqWp\nRCRJyxcsWTx/2aKl8bKKbhEZRIV3weJJkhcSRqF3A1uA9VktC9/Z3fsi4DcIr22WZfdw7QB/QCi4\nS9EOcATthP1SWW59K2ElSM3LFpFMqOAWkaGo8C5QHOleSCi624CDCCPIz9DksvBxIZyjCHOsa4vf\nWfH/sUz7GLIdIOFEyasodzvASYQ/cLYzsBLkztZGEhERkYlIhfc40Nnd+2LgJdQ5eXLVnWsBPh+v\n3gR8aRSbHK4d4Abg2vhv45hD5+cAwvzsfVNH+nq6drc2koiIiIgK70LFjifXMjDVZCthqskLjKH7\nSWd37wHAS4EDqT+tZNa9/lT19TOA6xl65HukdoBXxa8vUzvASQxMG3meMKKtRWpERESkdFR4F6yq\n48mYT66s6sk9nVBwNzs/OaV2gO2E1SC3E/5o2aKWfiIiIpICFd4tEIvssSyC0wbMZKA94EgF97oT\n7Qge+/XWyvWbGDzaPVw7wG8QTpgsQzvAdvJbDVJERESkECq8ExHbAx4er456hPvMN83mutWPfjBe\nXcdAO8DzgFfUPLws7QDbCN/jdsJ0nOc0qi0iIiKpU+FdYnGEewZh1cn2ER4+nHWEFoAXEbqqHF5z\n//cJ00m+R+vaAbYTuo9sIxTaz7coh4iIiEguVHiXUJzDfRih6G7K47/eArCY+u0A+wgFd0vaAe7Z\nsxcGiu1NmkIiUj5mdgnh+NEBfBFYTTjRei/wE+BCd1cffBGRUWjrz2Fpw5z1M3g+clk0nauzu3cq\nYXR7Gs2dMDlcO8CNwDW0rh1gpQvJ1uv+ufOpSZPax+VrmRPlakxZcyXDzOYAF7v7WWY2DfifwIlA\nj7uvNrMlwI3uft0Im0rptUgpK6SVN6WskFbelLJCenkzoxHvgsVFdPZ1MamaTvJiwoh0M11KKu0A\n3w0cU3Nfq9oB1i7Jvkl9tUWScQZwv5ldBxwMfAS4wN1Xx/tviI8ZqfAWERFUeBcqLhc/N17uA65k\n8Oj2WAvuIdsB/t5vz2y77+dPnwvcM8Ztj8UkYCeh0N4MbNeS7CJJmkno6f8O4FjC9LTqUaqtwCEt\nyCUikiQV3gWJI93vIuzz9nh5TdVDxtK2b7h2gCuAry2a+6pb3v+ZW/Iuuicx0Fu7cnKkFrERSd8G\n4EF33w08ZGY7CIt2VUwHnh3ltlL64zulrJBW3pSyQlp5U8oK6eTNdEqMCu+cdXb3Hk8YEZpGODmp\n2kLgjfHyaJdz7yC0A3wvcELNfesIJ0s20w5wVtW2hlK9LPsOwiI2OjFSZPy5Hfgg8DkzOwqYCtxi\nZqe6+23A2wiLbI1GKvM5U5t7mlLelLJCWnlTygrp5c2MCu8cxKXcDyWciPR2wg/YTfHfGfFh32eg\n6IaRl3M/jFConxMvV/sBoeD+Ls21A/yLqny1fwhU5mlvAzb39XSVadl4EcmBu19vZqeY2V2ET+o+\nAPwCuMLMOoAHCIttiYjIKKjwzsiu3Xv4k499ZyZwEHAg8BuE0aB+Bhat+QChuK54Y+126jiB0J0k\n73aAsxgouomXbwCcME97s+Zpi0w87v6xOjfPKTqHiMh4oMK7CVX9tqdf/pHTYKDvdvX85pmEbgAQ\nRqw/W3Vf9Qh49XLuI7UDvJbQEjDrdoCVFSP3Er6HR/p6uh7P+DlEREREJiQV3mPQ2d17EKH9376O\nJJMm7TdVaR1hOsm58fpmwgj3LAYK7C8xMAK+Lm7vXYQOJUfXbC+vdoDthHna9xEK+nfG21dWWh6K\niIiISPNUeI9SZ3fvgYQR7ekMjAzXm3pRfXLitcDJ8fouwlSRlzB4Hve6+DWXUKcdIOHEpauAuzP4\nNio9tXcSiu3thCkklXnh/9jZ3XsNDPQZFxGZaOYvWzQbYPmCJToOikimVHgPo7O7t4OBYvsABk5c\nHGquc72TE6+Pt82Mt19add+I7QAZW5vBffbu3Xeu5TbgOWDbcHO1VXCLyET2rQdWQTj+Mn/ZopXL\nFyxZ3NpEIjKeqPCuEudsTyOcIDmNwcX2sN1Cnt70POx/cuL18d9jwCLCqHcbMB84CziuZjOPE4rt\nbzD2doAQpo/sArYdMWMqfT1dDzWxLRGRCWH+skWzj50xaNHfufOXLVqqkW8RycqEL7w7u3unEU5+\nPBCYwuApJM205oOBPt2TCe0F++P/tfv9B4TpJN/L4Dm3ABvVV1tEpGkdhKmAKrxFJBMTsvDu7O6d\nSiiAK/Opmy60Z86YCvX7dB9I6Hwyo+ZLdgHfBr5Kc+0AJxGWZn+mr6drSxPbERGZ0JYvWLL2Ww+s\n4tFNjwEcGW++TFNORCQrE6bwjidHHsLg+dpZ96WuzOluJxTdCxl8siTAJuDrhBMvx9oOsI2QvzK6\nrcVsREQy8M5XnMm19/d+GLicsEIvaMqJiGSk8MLbzC4hLAbTAXwRWE2YZrEX+Alwobs3XRDH+drT\n478XEUaGRzVfuwnTgNOA97B/O8CngB6aawdYWT3y2b6ers1jDSkiIsNax0DRDZpyIiIZaevvL24x\nQjObA1zs7meZ2TTCkuonAj3uvtrMlgA3uvt1w2ymn8EdQADo7O6dTDgpcgqh0J5CfgV2rVmnvfbo\nW2695/Ft7N8O8A5gJfB/x7jtNsIbwDOMbfXIuvurJMqaTbkao1wykpRei36gbf6yRZcAcxmYcrIe\nKOOUk+T2batDNCClvCllhfTyZqbowvvThJ39O4QTGj8C9Lr7rHj/WcAZ7n7RMJvpryqyKydETiGM\naO8Z5uvyMFw7wOWEKSXNtAPcAzzV5NztMv9wlzWbcjVGuWQkKb0W+7LOX7boZAZPOQGYV7IpJ0nu\n20SklDelrJBe3swUPdVkJmEKxjuAY4E+Bu/4rYR52EP69TPbAI4HdtfcVVTR3QH8N0LBfULNfY8T\nTpZcydjbAbYRpqJs7Ovpem6M2xARkTF4ausGLrr+U7NjcV075UREpClFF94bgAfdfTfwkJntAF5a\ndf904NnhNtDfD//28dN/mmPGujZv28lt9z7B6h+tY8vzg6doH3/MiznttcfwquMOP7q9ve0TwCeG\n2k7s913pgrJP/95+Jk1q55CDOph64OSs4xf3sUbjyppNuRqjXKM3IUd5UjB/2aJLYh/vFZVOJvOX\nLVpJmHICYapJmUa7RSQxRRfetwMfBD5nZkcBU4FbzOxUd78NeBthifRhvf8zt1i+MQc5gTC6/Q5C\nP+6Kfe0AH3rs2Z9dfM5JXpOreun4inorW0IYvX86p3aAZf44p6zZlKsxyiXJi8vEz626qdLJZPH8\nZYuWgpaQF5HmFVp4u/v1ZnaKmd1FaLn3AeAXwBVm1gE8QFi1sdXaCd1JzgNeX3PfBuAaYClDtwOs\nV2DPYv+VLXuB+zSlRESkvFRwi0hWCm8n6O4fq3PznKJzDGEa8C5CO8BZNfc9SGh7OFI7wHoF9vV1\nHrcLeFxFt4hI6y1fsGRtnFZyUrxJ00pEJHMTZgGdERwNvJtQdNe2A7wF+ApwT5PPsQ64EXgL4ePv\nm/t6unRQFxEpieULlix+auuGT190/afK1rlERMaJ9lYHaLHXE1pF/SdhWkml6N5GKLb/CLiQxoru\ndYTpJRU3xdvagX8iFN8Ap3V2914y5uQiIpK5Iw46XFNLRCQ3E3HEezID7QBfXnPfY4R2gN9k7O0A\nYWDpeIAnCCdPTiGcqHkaVcsQd3b3LtXIt4iIiMj4N5EK78OAc4CF8XK17xMK7u+S3WqX6whTVZ4i\nnEQ6l9AD/FDCCmgiIiIiMoFMhMJ7uHaAfcDVwM8yfs52YAvwK+C3GGhRVRnp7oiXV2q0W0RERGRi\nGK+Fd7PtAJv1RKUnd2d3b+1964EPA+tUdIuIiIhMHOOt8J5GGF1+D6FTSbUHCKPbI7UDHJP+vf0A\nOwkF9b7l6/t6utZ2dvcOWvmsr6drTdbPLyIiIiLlNl4K7+HaAd5MKLjvzjPAi6dPoa+n65fVt3V2\n984G6OvpWtzZ3Xt7vKyiW0RERGQCSr3wfj1hOsnpDF4aeithBcyvMXjJ9qy1x+daf9DUjkF3xFaB\nc+PlDcDh8fLKvp6uxTlmEhEREZESSq7w3rV7L8Afk287wJG0AduBJ/t6uvabthJHuitTSzqAk4FH\nCCdUqoWgiIiIyASUXOH9sf99O4SFaKp9nzCd5Htk1w5wKG3Aei31LiIiIiKNSG7lyi3PVzrysQtY\nCXQRppvcSv5F9wvAIyMV3XE0e2XV16xhoJWgWgiKiIiITEBt/f39rc7QkO5/va3/ocef/QJwLfBM\nQU/bBjzT19P19DCP6WfwPPN9J1dGs2hNC8H9cpVIWbMpV2OUS0aS0muRUlZIK29KWSGtvCllhfTy\nZia5Ee+PvPu1AJdTXNG9B/jFCEV3XbHIPhtYAVwWL4uIiIjIBJRc4V2gNkLHkkf7erp2jmUDNSdZ\nQjixcvZQjxcRERGR8Su5kysL0gY81dfTtanJ7cwi9BXfxcAc7+r+3prrLSIyzsxftmg2wPIFS3SM\nF5FBNOJd32PNFt2xj/e1wMuAE4AjCSdcVqaerIiPERGRcWL+skWXEI/x8bKIyD4qvAfrB9b29XRt\nb2YjcUR7ITCDMNq9mzBt5XY09UREZFyav2zRyYRjf8Xcyui3iAhoqkm1vYSie08O2+6naqqJeq2o\nWgAAFNVJREFUiIiML3F0eyHhU85NwPrWJhKRMtKId7CLcBJllkX3KsLBl/j/tX09XWsY6O8N6ukt\nIpK8OKo9lzDAsonwaWcHsFLzvEWkmka8YQfweF9PVyYNzeO87cp0kmuAPqr6d/f1dC3u7O5dGi/r\ngCwiMr6sJ7S7vXD5giVrWh1GRMoluQV01m/c1n/Bp2+2DDbVBjzX19OV1ceB/Z3dvccSTqrpiLe9\nAMwjdDchjngXrcxN6suaTbkao1wykpReizFljVNNKoMuK5cvWLI401RDG/f7toVSyptSVkgvb2Ym\n6lSTNuDpDIvuakcS5vi9LF6+ijDtZFVnd++qHJ5PRERaLBba84B5BRbdIpKYiVp4P9HX01XEypcd\nwOuqrp/c2d17cgHPKyIiBVu+YMlazekWkeFMtDne/cAv+3q68uwwUpnfBzAZOCjH5xIRkRxoERwR\nycNEGfFuI5xE+UieRXc8WXIlYW73C8BXgdVVD1nTonneIiIySloER0TyMhFOrsxq+feR7DtRoHZJ\n+Mr0Ep1cuZ+yZlOuxiiXjCSZ1+KprRv6L7r+Uz+suXleiUe+k9m3pJUV0sqbUlZIL29mxvtUkz2E\nVoGFLl5T2yZQo9wiIkmbNX/ZIk07EZGmjdepJu2EJdofLbroFhGRdB1x0OEweKGzDcBlaNqJiGRg\nvE41ebKvp2tLIYEGlPVjk7LmgvJmU67GKJeMJKXXoh9oiydXziIU3dXKNu0kuX3b6hANSClvSlkh\nvbyZGU8j3pUTKB9uQdEtIiLjSCyu17U6h4iML+Ol8G4Dft3X0/VYX0/X3laHERGR9MXiu3raycqS\njXaLSGLGw1STnYQFcXa3KlM0qo9NajueFKDMH+eUNZtyNUa5xjEzuxd4Ll59FFhMWJF3L/AT4EJ3\nH+mNJKXXYr+sJe/pnfS+LbmU8qaUFdLLm5mUC+82YENfT9fGVmeKRvwh6uzuvQSYG6+u7OvpKmJZ\n4TL/cJc1m3I1RrnGKTM7ELjT3U+suu3bwL+4+2ozWwLc6O7XjbCplF6LlLJCWnlTygpp5U0pK6SX\nNzPptRMMfyfsAdb19XTtbG2Y4cXR7VkMzBOcW3X33M7u3qUFjnyLiDTq94CpZnYj4f3iE8CJ7l5Z\nGOwG4AxgpMJbRERIsPCeNnUyhDaBpR6qj6PbFwIzgE3ANa1NJCLSsG3AZ939SjP7bWBVzf1bgUOK\njyUikqbkCu/pUztIoOieDSwkFN3E/88kvGmdFm9bqdFuyUsLziWQ8ekh4GEAd/+5mW0EXlN1/3Tg\n2VFuq9TH7RopZYW08qaUFdLKm1JWSCdvplNixktXk1QsAeYB8wqa3y0TUPy0ZQWwIl4WGavzgR4A\nMzuKUGjfZGanxvvfBqwe4mtrtSXyL6WsqeVNKWtqeVPKmlreTCV3ciXlnZA/KFedqSaXt6jYLuv+\ngvJmSzZXHOleUXPzvJxHvpPdXzI8M5tM6GByDGF/fhTYCFwBdAAPABeM964mJZdS3pSyQlp5U8oK\n6eXNTHJTTcqu6iP+xZ3dvUuJJ1fqI38RSY277wLOrXPXnIKjiIiMC4WPeGfQE7asfyX1d3b3Xkqd\ndoEtnm9b1v0F5c2WdK4WtK1Men9JIVJ6LVLKCmnlTSkrpJU3payQXt7MFFp4Z9QTtpQvVuwv/sOa\nm+cBZ1N87+5qpdxfUVmzJZ+r4D/2kt9fkruUXouUskJaeVPKCmnlTSkrpJc3M0VPNZloPWFnod7d\n0gL6GRMRESmforuaVHrCvhV4P/AfNfcn2xP2yMOmAaysuqlyuaP4NCIiIiJSNkWPeGfVE7aUrVj6\nero+vX7jNgDW/NcTJ93541/xzOYdtNHGjIOn8KZXHcW8049/tAXRSrm/orJmU67GKNfoTciPV0VE\npPjC+3zgd4ELa3vCuvtthJ6wt4xiO2V84+oH2o48bFq9lm4dGzfvuPCyD89Z06pcLXje0ShrNuVq\njHKJiIiMQtFTTa4EXmxma4ClwPuADwF/a2Z3Ev4Q+EbBmYrwAgyc8CYiIiIiE48W0MlOvQV0KidW\nbgAOj5dz7WxSp5tFWfcXlDebcjVGuWQkKb0WKWWFtPKmlBXSyptSVkgvb2ZUeGdnv1yxCJ4FXFbz\n2FxWEhyif3NZ9xeUN5tyNUa5ZCQpvRYpZYW08qaUFdLKm1JWSC9vZoqeajKudXb3zq6MOFdNK1lX\n1HOzf+tCTW0RERERKQktGZ+RFbc8BPGEys7u3m3ADMLc7pXx31xCa8FV6rEsIiIiMvFoxDsDnd29\ns+/88a8qV18K/CFwHHAkoeBeCtwa7z8tTgnJVCzmB/URV4EvIiIiUh4qvLPVweAFgA4FphHmeZ9G\n7G5CTtNA4pzueYQ55EUvTS8iIiIiw9BUkwz09XStXXHLQzy87jkIJwxsBV4ETCJMOeksMktRzyUi\nIiIio6cR74zMO/14gC7gcuCXwG7gacLJlacxMNUENA1EREREZMJRO8Hs7MvV2d17MqEAf6Hq/nmV\nCwUX3WXdX1DebMrVGOWSkaT0WqSUFdLKm1JWSCtvSlkhvbyZUeGdneEW0Ml10ZwRlHV/QXmzKVdj\nlEtGktJrkVJWSCtvSlkhrbwpZYX08mZGhXd2hlpAp9Xzrsu6v6C82ZSrMcolI0nptUgpK6SVN6Ws\nkFbelLJCenkzo8I7O8rVuLJmU67GKJeMJKXXIqWskFbelLJCWnlTygrp5c2MTq4UERERESmACm8R\nERERkQKo8BYRERERKYAKbxERERGRAqjwFhEREREpgApvEREREZECqPAWERERESmACm8RERERkQKo\n8BYRERERKYAKbxERERGRAqjwFhEREREpgApvEREREZECqPAWERERESmACm8RERERkQKo8BYRERER\nKYAKbxERERGRAqjwFhEREREpgApvEREREZECqPAWERERESmACm8RERERkQKo8BYRERERKYAKbxER\nERGRAqjwFhEREREpgApvEREREZECqPAWERERESmACm8RERERkQKo8BYRERERKYAKbxERERGRAqjw\nFhEREREpwAGteFIzOwL4IXA6sBe4Kv7/E+BCd+9vRS4REdmfjtkiItkofMTbzCYDXwK2AW3A54BL\n3f2UeL2r6EwiIlKfjtkiItlpxVSTzwJLgCfj9RPdfXW8fAPwRy3IJCIi9emYLSKSkUILbzN7L/C0\nu98Ub2qL/yq2AocUmUlEROrTMVtEJFtFz/F+H9BvZn8EvBq4GphZdf904NkRttE2wv2tolyNK2s2\n5WqMco1fWRyzIa3XIqWskFbelLJCWnlTygrp5c1MoSPe7n6qu89x9zcD/wW8B1hlZqfGh7wNWD3k\nBkREpDA6ZouIZKslXU2q9APdwBVm1gE8AHyjtZFERGQIOmaLiDShrb9fXaBERERERPKmBXRERERE\nRAqgwltEREREpAAqvEVERERECtDqkyv3Y2aHAF8ntKnqAC529++b2RuBfwV2Aze5+9/Fx/818PZ4\n+4fc/W4zOxy4BjgQ+BXwPnffnlPeduCLwKuAncCfu/sjeTxXzfNOBr4M/CYwBfgH4EHqLOVsZhcA\n/52wj/7B3a83sxcR9vNMYAtwnrtvyDDfiEtMF53LzC4BOgk/V18kdGNoaa74Ol5NeB33ABfE/1uS\ny8zeAHzG3d9sZsc1m2Oo39sMsr0a+AJhX+0E3uPuT7UiW3WuqtvOAS5y9zfF6y3ZZ3kzs3cC73L3\nc+P1po/TZtYJfCo+9svu/n8K+l5aciyvyVDa37+anLm8/+SYdxJwBXA84QTh9xNe41LmjZkzfQ/N\nOeu9wHPx6qPA4rLmzaMOaDRrGUe8Pwz8p7vPAd4LXB5v/zdgobv/IfAGM3u1mZ0InOLubwDOrnrs\nXwFfj0sa/wj4ixzz/jHQEd9gPw705Phc1c4lLGxxCnAm4XvvoWYpZzM7EvhL4E3AW4HFsRvBIuC+\n+NivAp/MKtholpguOpeZzQF+P75OpwJHU4799XZgkrv/AfB3wKdblcvMPkp4c5oSb8riddvv9zaj\nbP9KKGzfDHwT+JiZvaTobHVyYWavAc6vut6SfZY3M/s84ee1uh/vEpo4Tsdjx+eAtxB+T/97LECK\n0KpjOVDu37868nr/ySvvO4C9cbufJLvjbC55c3oPzSvrgQDu/ub478/KmjfHOqChrGUsvC8D/j1e\nngxsN7PphAPi2nj7jYRliv8AuAnA3R8HDoijKH8ArIqPzXtJ433P5e4/AF6b43NVW0F444LwOu6i\n/lLOrwPucPdd7r4ZeJgwolO9j1aR7T4azRLTRec6A7jfzK4D+oDvACeVIJcTfm7bCCsAvtDCXA8D\nf8JAIdXU6zbM720W2c529x/Hy5OB7cDrW5BtUC4zOwz4R+BDVVlbkasIdxDeiCrf+8HAlCaP0ycA\nD7v7c+6+C7gdOKWYb6dlx/KKMv/+1cr8/SfPvO7ey8AA3G8Bm2jyOJvz/s30PTTnrL8HTDWzG83s\nljj6W9a8mdcBY8na0sLbzP7MzO6v/gcc5+474l8cXwMuIRQkm6u+dEu87WAGPt4Y6va8lzQ+uCbb\nnviRZa7cfZu7b40v+grCX17VzzuafbS55ram2chLTLckF+GjoZOAdxE+ZrymJLm2Ed4IfkYY4fhC\nq3K5+zcJH5VVNJuj9ndjzPlqs7n7egAzexNwIeEP9sKzVeeKv/dXAhcTjjsVLdlnWal3nDazk9x9\nec1Dh8reyHF6qMcWoSXH8ooy//7VyZrH+0+uP/vuvsfMrgY+D/wHJd2/Ob2H5rlvtwGfdfe3Et5b\n/6Pm/jLlzaMOaDhrS+d4u/uVhDeqQczsd4FrgW53XxNHUqZXPeRgwjLFL9TcXlm+eHN8zNOMfknj\nsdpck6Hd3ffm+Hz7mNnRhI/ZL3f3a83sn6vuruyj2nzT69ye5T4aaYnpVuXaADzo7ruBh8xsB/DS\nEuT6MLDK3T9hZrOA7xJGcFudC8Kct2Zy1D62so1MmNkC4FLg7e6+0cxane0k4DjCSNWBwCvM7HOE\n17QU+2wshjpO1zFU9kaO0/Vew01jzd6glh3Lh1D237+s339y/9l39/PilLS7CL+jZcybx3tonvv2\nIcKIMO7+czPbCLympHnzqAMazlq6qSZm9grCX9AL3f1GgDjU/4KZHRs/kj+DMCH+DuCtZtZmZscA\nbe6+Md7+9rjJvJc03vdc8SOWHw//8GzEg8dNwEfd/ap4849s/6Wc7wJONrMpFk5cfTnhBIJc9pGP\nfonpQnMRPrI+E8DMjgKmAreUINczDPy1vInwx3DLX8eoqRzuvoX6v7dNM7M/JYx0z3H3X8SbW5rN\n3e9291fGn/2zgQfc/WLg7lbmKkpGx+kHgd82sxlxTuUpwP8r6FtoybF8GGX+/cv8/Sfv44WZfTxe\n3U44KfueMubN4z005+PK+cTzIeJ763TgppLmzbwOGEvW0nU1IZz00AF8wcwAnnX3dzLwEcYk4EZ3\nvxvAzNYQDszthDdiCGdYX23hrNSngXNyzPst4C1mdke8/r4cn6vapYSPM/7KzCpz7T5I2G/7lnL2\ncHbuF4A1hH10qbvvNLMlhH20hnB2d177qO4S00Xn8nBG8ilmdld8vg8Av2h1LsIUiS+b2WrCz/0l\nhDPZW5mrspxtFq9b3d/bZrJZ+Pj/88AvgW/G48T33P1vW5itdgngtspt7r6+xfssT/0M/t6bOk67\n+24zu5gwT7IduNLdn6QYrTqW1yrz719FXu8/eeX9JvAVM7uN8IniBwnT+8q6f6tl9R6aV9Yrgavi\n8/UTfm82ljFvjnVAQ1m1ZLyIiIiISAFKN9VERERERGQ8UuEtIiIiIlIAFd4iIiIiIgVQ4S0iIiIi\nUgAV3iIiIiIiBVDhLSIiIiJSABXeUhgzO8nMrmh1DhERGZmO2SLZUx9vEREREZEClHHlShmnzGwO\n8DeE1a3uAk4GZgJ/6e6rzOw3ga/E254H/tzd7zez9wEXx6/7IXCRu28zs/XAt+N2ngS+CPwPYBbw\nXndfbWbHxdsPi9v8S3f/LzO7CtgLvJKwAtvfu/vXzewg4HLgdwirUP2Tuy81s/cC58Xt9AFHxTxj\n/fqfAh8hLF28FvjTuDLWpcC58fabgI8CxxBW1bsfeA3wa2Ceu29q7hURERmajtk6Zkv2NNVEilS9\nvPRkd38T8GHC0tEQDrYr3P13CQf7T5rZKwnLE5/i7q8CtgF/HR9/BNDn7i+P1//Y3U+JX/uheNvV\nwEfd/STgL4ClVXmOAn4fOA34FzN7CfBJ4B53fy1wKvAJM5sdH/9S4NXu/okMvv7vgbfEx/0MOMHM\n3g50AicSDtbHEZaiBXgV0BP3zbOEA72ISJ50zNYxWzKmEW9plVXx/58Ch8bLpwALANz9BuAGM7sI\n+HbVSMG/E0ZYKm6I//8SWBMvPwbMMLNpwOuAr5hZ5fHTzOxQwpvJV9x9D/CEmd0B/CFwOjDVzM6P\nj59KGAnpB+51971Vz93M1/cBd5rZdcBKd7/PzN4NXOPuOwHM7MuEEZfrgafc/b74tT+p2mciIkXQ\nMVvHbMmARrylVXbG//uBtnh5V9VlzOwV8Xpb1de1U/UHo7vvrrpvT81zTAK2u/trKv+AN7r7M3Ue\n3w7sjl9zbtXjf5+BN5ztNdsf89e7+4eAucAzwNfN7NwRvtcdVbf31zxORCRvOmbrmC0ZUOEtRRrp\nwLMaOBvAzN4CfAn4HnCWmc2Ij7kAuHU0T+bum4GfxwMkZnZGfI5Klvnx9t8E3hDvuxX4QLz9KODH\nhPl6tdnH/PVmdoCZPQRscPfPAF8lfEx5K7DQzA40swOA9432exURyYGO2eiYLdlS4S1F6q/6v7/O\n7RcBc83sR4Q5gRe4+/3AYuA2M3sQOJgwJ4+abQy1/XOBPzez+wjzEudXPWaqmd0DfCc+1ybgb4EX\nmdn9wM3AR9z90SEyj+nr44jPXwE3m9ndhBONetz9+ritewgfTa4F/hfhDWCo71VEJC86ZqNjtmRL\n7QRlQjKzrwDfdfevtuLrRURk9HTMlvFCI94iIiIiIgXQiLeIiIiISAE04i0iIiIiUgAV3iIiIiIi\nBVDhLSIiIiJSABXeIiIiIiIFUOEtIiIiIlIAFd4iIiIiIgX4/yRj9ke2ZqIGAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10b0fc080>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Silent matplotlib warning. \n",
    "import warnings\n",
    "warnings.filterwarnings('ignore',category=FutureWarning)\n",
    "\n",
    "# Setting an apropriate size for the graph.\n",
    "f,a = subplots(1, 2)\n",
    "f.set_size_inches(12,6)\n",
    "\n",
    "# Plot the graph.\n",
    "sn.regplot(df_low.incomeperperson, df_low.lifeexpectancy, ax=a[0]);\n",
    "a[0].set_title('Countries with LOW urbanrate', fontweight='bold');\n",
    "sn.regplot(df_high.incomeperperson, df_high.lifeexpectancy, ax=a[1]);\n",
    "a[1].set_title('Countries with HIGH urbanrate', fontweight='bold');\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.3 Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see above, the correlation in countries with high urban rate, `urbanrate` $\\ge 50\\%$, is higher than in countries with low urban rate, and in both cases the $pvalue$ is significan. So, we can say the variable `urbanrate` moderates the relationship between `lifeexpectancy` and `incomeperperson`. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "End of assignment."
   ]
  }
 ],
 "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": "3.4.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}