Source

ml-workshop / 02-Regression.ipynb

Full commit
{
 "metadata": {
  "name": "02-Regression"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Load data"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.datasets import load_boston\n",
      "boston = load_boston()\n",
      "print(boston.DESCR)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Boston House Prices dataset\n",
        "\n",
        "Notes\n",
        "------\n",
        "Data Set Characteristics:  \n",
        "\n",
        "    :Number of Instances: 506 \n",
        "\n",
        "    :Number of Attributes: 13 numeric/categorical predictive\n",
        "    \n",
        "    :Median Value (attribute 14) is usually the target\n",
        "\n",
        "    :Attribute Information (in order):\n",
        "        - CRIM     per capita crime rate by town\n",
        "        - ZN       proportion of residential land zoned for lots over 25,000 sq.ft.\n",
        "        - INDUS    proportion of non-retail business acres per town\n",
        "        - CHAS     Charles River dummy variable (= 1 if tract bounds river; 0 otherwise)\n",
        "        - NOX      nitric oxides concentration (parts per 10 million)\n",
        "        - RM       average number of rooms per dwelling\n",
        "        - AGE      proportion of owner-occupied units built prior to 1940\n",
        "        - DIS      weighted distances to five Boston employment centres\n",
        "        - RAD      index of accessibility to radial highways\n",
        "        - TAX      full-value property-tax rate per $10,000\n",
        "        - PTRATIO  pupil-teacher ratio by town\n",
        "        - B        1000(Bk - 0.63)^2 where Bk is the proportion of blacks by town\n",
        "        - LSTAT    % lower status of the population\n",
        "        - MEDV     Median value of owner-occupied homes in $1000's\n",
        "\n",
        "    :Missing Attribute Values: None\n",
        "\n",
        "    :Creator: Harrison, D. and Rubinfeld, D.L.\n",
        "\n",
        "This is a copy of UCI ML housing dataset.\n",
        "http://archive.ics.uci.edu/ml/datasets/Housing\n",
        "\n",
        "\n",
        "This dataset was taken from the StatLib library which is maintained at Carnegie Mellon University.\n",
        "\n",
        "The Boston house-price data of Harrison, D. and Rubinfeld, D.L. 'Hedonic\n",
        "prices and the demand for clean air', J. Environ. Economics & Management,\n",
        "vol.5, 81-102, 1978.   Used in Belsley, Kuh & Welsch, 'Regression diagnostics\n",
        "...', Wiley, 1980.   N.B. Various transformations are used in the table on\n",
        "pages 244-261 of the latter.\n",
        "\n",
        "The Boston house-price data has been used in many machine learning papers that address regression\n",
        "problems.   \n",
        "     \n",
        "**References**\n",
        "\n",
        "   - Belsley, Kuh & Welsch, 'Regression diagnostics: Identifying Influential Data and Sources of Collinearity', Wiley, 1980. 244-261.\n",
        "   - Quinlan,R. (1993). Combining Instance-Based and Model-Based Learning. In Proceedings on the Tenth International Conference of Machine Learning, 236-243, University of Massachusetts, Amherst. Morgan Kaufmann.\n",
        "   - many more! (see http://archive.ics.uci.edu/ml/datasets/Housing)\n",
        "\n"
       ]
      }
     ],
     "prompt_number": 1
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "boston.data.shape"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 2,
       "text": [
        "(506, 13)"
       ]
      }
     ],
     "prompt_number": 2
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Some shortcuts"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "indexof = {name: index for (index, name) in enumerate(boston.feature_names)}.get\n",
      "data = boston.data\n",
      "prices = boston.target * 1000"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Have a look at the data"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plt.scatter(data[:,indexof('RM')], prices)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 4,
       "text": [
        "<matplotlib.collections.PathCollection at 0x106eafed0>"
       ]
      },
      {
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAD9CAYAAABTJWtQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3XlcVNX7wPHPrDADgiIKCi6JiIJLlgKWqYm4r7lbpraZ\npl+zLNOvlbaole1ppbm2Ydnugopbfl0wlzTFFJVQFjE3kJ2ZeX5/3HHCn+WCIJrn/Xr5Ku7cc+5z\nL3qeuefce45ORARFURRFAfTlHYCiKIpy41BJQVEURXFRSUFRFEVxUUlBURRFcVFJQVEURXFRSUFR\nFEVxuWxSOHv2LH369KFBgwaEhoYSHx/P6dOniY6Opl69erRv356zZ8+69p82bRrBwcHUr1+fVatW\nubbv2LGDRo0aERwczJgxY1zbCwoK6N+/P8HBwURGRpKcnFzKp6goiqJcqcsmhTFjxtC5c2f279/P\nnj17qF+/PtOnTyc6OpqDBw8SFRXF9OnTAUhISGDx4sUkJCQQGxvLyJEjOf8axIgRI5g7dy6JiYkk\nJiYSGxsLwNy5c6lcuTKJiYmMHTuW8ePHl+HpKoqiKJdyyaSQmZnJxo0beeihhwAwGo14e3vz448/\nMmTIEACGDBnC999/D8APP/zAwIEDMZlM1K5dm7p16xIfH096ejrnzp0jPDwcgAcffNBVpnhdvXv3\nZs2aNWVzpoqiKMplGS/1YVJSElWqVGHYsGHs3r2bO++8k3feeYeMjAz8/PwA8PPzIyMjA4C0tDQi\nIyNd5QMDA0lNTcVkMhEYGOjaHhAQQGpqKgCpqanUqFFDC8aZdE6fPo2Pj49rf51OV0qnqyiKcmu5\n2kkrLnmnYLPZ2LlzJyNHjmTnzp14eHi4uorO0+l016XRFpGb9s+LL75Y7jHcqvHfzLGr+Mv/z80e\nf0lcMikEBgYSGBhI8+bNAejTpw87d+7E39+f48ePA5Cenk7VqlUB7Q7g2LFjrvIpKSkEBgYSEBBA\nSkrKRdvPlzl69CigJaHMzMwL7hIURVGU6+eSScHf358aNWpw8OBBAOLi4ggLC6Nbt24sXLgQgIUL\nF9KzZ08AunfvTkxMDIWFhSQlJZGYmEh4eDj+/v54eXkRHx+PiPDpp5/So0cPV5nzdS1ZsoSoqKgy\nO1lFURTl0i45pgDw/vvvc//991NYWEhQUBDz58/HbrfTr18/5s6dS+3atfnqq68ACA0NpV+/foSG\nhmI0Gpk1a5ara2nWrFkMHTqUvLw8OnfuTMeOHQF4+OGHGTx4MMHBwVSuXJmYmJgyPN3y0aZNm/IO\n4ZrczPHfzLGDir+83ezxl4ROStrxdB3pdLoS948piqLcqkrSdqo3mhVFURQXlRQURVEUF5UUFEVR\nFBeVFBRFURQXlRQURVEUF5UUFEVRFBeVFBRFURQXlRQURVEUF5UUFEVRFBeVFBRFURQXlRQURVEU\nF5UUFEVRFBeVFBRFURQXlRQURVEUF5UUFEVRFBeVFBRFURQXlRQURVEUF5UUFEVRFBeVFBRFURQX\nlRQURVEUF5UUFEVRFBeVFBRFURQXlRQURVEUF5UUFEVRFBeVFBRFURQXlRQURVEUF5UUFEVRFBeV\nFBRFURSXyyaF2rVr07hxY5o2bUp4eDgAp0+fJjo6mnr16tG+fXvOnj3r2n/atGkEBwdTv359Vq1a\n5dq+Y8cOGjVqRHBwMGPGjHFtLygooH///gQHBxMZGUlycnJpnp+iKIpyFS6bFHQ6HevXr2fXrl1s\n27YNgOnTpxMdHc3BgweJiopi+vTpACQkJLB48WISEhKIjY1l5MiRiAgAI0aMYO7cuSQmJpKYmEhs\nbCwAc+fOpXLlyiQmJjJ27FjGjx9fVueqXIVPPplHtWrB+PjU4D//eQabzVbeIf2jnTt3Ehoagbe3\nP23adCU9Pf2qyp89e5bu3Qfi7V2NOnWasG7dulKJS0R48813qFq1Dr6+tZg0aQoOh+OKyzscDp5/\n/iV8fWtRtWodZsx42/Xv6VqlpaXRunUXvL39adgwkl9//fWKyx49epS77mqPm1tFjMaKWK2+VKhQ\nHS8vP0wmX3Q6Tzw8qrNy5UpOnjxJx4698fauRr16d7J58+bL1r9s2TJq1QqjYsXq9O8/lPHjJ5X4\nGpw9e5bOnftgMlVEr69ItWpBpfb7/deSy6hdu7acPHnygm0hISFy/PhxERFJT0+XkJAQERGZOnWq\nTJ8+3bVfhw4dZMuWLZKWlib169d3bf/yyy9l+PDhrn22bt0qIiJFRUXi6+t7UQxXEKZSipYuXSpW\nay2BeIFEsVrbyPjxz5d3WH8rIyNDvLz8BD4VSBGj8TkJDW0uDofjiuto27abmM2PCBwT+EmsVl85\nePDgNce2aNGn4uFRX+BXgf1itTaTN954+4rLv/nmu2K13imQILBbrNb6smDBomuOy263S0jIHWIw\nTBJIEVgg3t7+F/07/ztFRUVy221hotN1FGguMEOgrkCsQGWBr53XcZTo9d5y++0txWQa7dy2RDw9\nq8jRo0f/sf5du3aJxVJFYLVAshiNvUWvr17ia9C2bTfR6YY4j/+jQCVxd69UKr/fm0FJ2s4rulNo\n164dzZo1Y86cOQBkZGTg5+cHgJ+fHxkZGYD27SMwMNBVNjAwkNTU1Iu2BwQEkJqaCkBqaio1atQA\nwGg04u3tzenTpy+KY/Lkya4/69evL1ECVK7MN98sIzf3KSAcqEtu7mt8/fXS8g7rb8XHxwNNgQeA\nAGy2qRw+fJgTJ05cUXm73c6GDbEUFr4PBAJdEenG2rVrrzm2mJil5OT8F2gC1Cc392ViYn66ivI/\nkZv7EtAAaExu7qSrKv9P0tPTOXo0Bbv9JSAAGAI0dPUEXEpycjInTmQhkge8CuwEJgGngXuBPmjX\n8V0cjjx2795GUdHbzm290enasHHjxn+sf9WqVRQVPQC0A2pis83C4cimJNfg/O9W5GPn8bsBPXE4\nGpTK7/dGtH79+gvaypIwXm6HTZs2Ua1aNf7880+io6OpX7/+BZ/rdDp0Ol2JDn41SnqCytXz8fHC\nYEjCbj+/JYmKFb3LM6R/5OXlhcNxDLCh/XU+gcORj6en5xWV1+v1uLl5kJubDIQAgl6fhLd3+2uO\nzcfHC70+ib96jJLw8bny61ipkheQVCzWJCpXvvbfg6enJzZbDnASqAIUYbcfw9v78nV7eXlRVJQJ\nWJ2xnY+xOZAMONB6pVMBQafTI5IG1AAciCRf8jheXl6YzVv4q7cyCbC4Pr+aa6DX6zGbreTl/cH5\n3y0kodOdvaJzvRm1adOGNm3auH6eMmXK1VdyNbcVkydPlhkzZkhISIikp6eLiEhaWpqr+2jatGky\nbdo01/7nu4bS09Mv6D764osv5PHHH3fts2XLFhFR3Uc3itTUVPH1rSFm88Oi1z8jVquvrF+/vrzD\n+lt2u13uvberWK1tBJ4XD48QmThx8lXVMWvWR2K11hCdbpJYLN2kUaNIyc/Pv+bYEhMTxdvbT4zG\nEWIwPCkeHr6yc+fOKy6/a9cu8fDwFYPhSTGZRoqXl1+pdXs8++wk8fBoIPCCeHi0lvbte4rdbr+i\nsqNHjxN39yABb4Ehzv8OEwgSaCkwSaC61KrVQKZNe0Os1joCz4vV2kHCw++VwsLCf6z73LlzEhTU\nSNzd+4pO919xd/cTs7mC6PXDRa9vLe7uFa7q7+LMmR+K2VxdYKJAB9HpqklYWLjk5eVdcR03s5K0\nnTpnwb+Vm5uL3W6nQoUK5OTk0L59e1588UXi4uKoXLky48ePZ/r06Zw9e5bp06eTkJDAoEGD2LZt\nG6mpqbRr145Dhw6h0+mIiIjgvffeIzw8nC5duvCf//yHjh07MmvWLH777Tc+/PBDYmJi+P7774mJ\nibkgDp1OV2oDbMqVycjIYNGiReTl5dOrV08aNWpU3iH9I5vNxsKFC0lOPkrz5s3o1q3bVdexdu1a\n1q/fgJ9fVR566CEsFsvlC12Bo0eP8sUXX2Cz2enXry/16tW7qvKJiYksXvwVRqOBQYMGUbNmzVKJ\nC+CHH35g+/Yd1KlzG4MHD8ZovGzHAaANoH/33XfExcWRmpqG0WjAYDBQt24wR48mc/RoCk2bNmHG\njBmYTCZWrlzJpk2bCQiozrBhwzCbzZes/9y5c8yfP59Tp07Tvn00IkLHjvdRWHgPer03ZvMyNm2K\nu+K/k2vXrmX+/AWcOnWS6OhoHn/88VL7/d7oStR2XipjHDlyRJo0aSJNmjSRsLAwmTp1qoiInDp1\nSqKioiQ4OFiio6PlzJkzrjKvvvqqBAUFSUhIiMTGxrq2b9++XRo2bChBQUEyevRo1/b8/Hzp27ev\n1K1bVyIiIiQpKalUsp2iKGXr+PHjUqNGiFSo0FoqVGgtNWvWl4yMjBLXt2bNGnnssdHyzDMTLhiM\nHjHiSTEanxYQARGdbqa0bdu9NE7hX68kbecl7xRuFOpOQVFuPEOHjuDzzz2w2WYAYDQ+zf3357Jg\nwYdXXddXX33NsGFPkps7FoMhHS+vGPbsiScwMJBevQbz/fdtgWHOvTfQsOF/+e23/5XeyfxLlaTt\nVG80K4pSIocOHcVma+X62WZrxeHDR0tU14QJU8nNXQSMw25/k6ys+5gzZy4APXq0w2p9C23Q+QRW\n60t069bu2k9A+VsqKSiKUiKtW4djsXwM5AF5WCwf06pVeInqysvLRXsSSmO3VyEnJw+AIUMe5Jln\nBmC13ombW10GDmzAlCn/vfYTUP6W6j5SFKVECgsL6d9/KEuX/gBAt249iYmZf9mB5L/z3HMv8P77\na8jNfRdIw2p9lLVrfyQiIqKUo761lKTtVElBUZRrkpWVBWjvGJSU3W7nhRde4YsvvsXT05PXX59E\np06dSivEW5ZKCoqilKpt27bxwQdzERFGjhxGixYtyjsk5SqopKAoSqnZsmUL7dp1Jzf3OUCP1TqN\nFSuW0KpVq8uWVW4MKikoilJqunbtz7Jl9wCjnFvm0r79ClauXFKeYSlXQT2SqijKNXM4HIwePY7l\ny78DnkJ7P6AQqEBBQWH5BqeUOZUUFEW5wKxZHzFv3kZE0tFmP00DhmK1PsvIkYPLOTqlrKmkoCjK\nBWJjN5KbOxqoDHgC43F338Ds2VPp168voM3p1LlzX4KDmzF48GNkZmaWZ8hKKVJJQVGUC9Ss6Y/J\ntN31s16/naioltx//yAAsrOziYxsy6pVjTl0aCZff+0gOrqnGvf7l1ADzYqiXODEiRPccUdLMjOD\nEXHHbN7Ktm0bqFu3LgBxcXH07j2ZrKzzcw/ZcXevxqFDuwgICLimY69YsYJly1ZTtaoPo0aNxMfH\n5xrP5tZWkrbzyubKVRTlllG1alUSErazbNky7HY7HTp8RJUqf01B4ebmhsg5/lpQJx+Ho6BEbzIX\n9+GHHzNu3DRyc0diNv/OnDkt2Lt32792QZwblbpTUBTlqhQVFRER0Zb9+6uTn98Oq/VzunW7jZiY\n+ddUb6VK1Tl7diWgrZNgsfTh7bejGT58eClEfWtSdwqKolzWli1bWLQoBjc3M0888RjBwcFXVd5k\nMrFxYyyvv/4W+/dv5u67ezNq1Mhrjis/Pwfwd/1st/uTk5NzzfUqV0cNNCvKLWTlypW0a9eTjz6q\nznvvmbnzzrs5cODAVdeTlpbG2rWb2bRpC2vXbuHMmTPXHFuvXn1wd38E2AsswWhcTOfOna+5XuXq\nqO4jRbmFNGvWlh07HgUOAxuB0wwcGMYXXyy44joyMzOpW7cxp08/icPRHpNpNg0a/MKuXf9Dry/5\n98z8/HxGj36W5ctX4eNTmZkzp6kpNa6RmuZCUZRLCg1twf79VYFcYCywDZPpHZYu/YrWrVvj5uZ2\n2TpWr15Nnz6vkJW1wbnFgcVSnYMHtxMYGFiG0StXS01zoSjKJQ0b1hdYDnwDdAYmU1QUTo8ejxAa\n2oyMjIzL1mG1WnE4TgJ255Zz2O25WCyWa44vNzeXP/74g8JCNZ1GeVFJQVFuIU8+OdrZxWMvttWd\n/PypHDvWiZEjn7lsHZGRkTRuHIDF0gN4Gw+P9gwe/CCVK1e+ptgWL/6KypWrExZ2D35+tdiyZcs1\n1aeUjOo+UpRbzPDh/+Gzz3aRmzsG2AYsAXYC+6hf/2m+/34hx44do2HDhvj7+/9tHQUFBcya9SG/\n/36EFi3uYMiQIeh0uhLHlJycTGhoM3Jz1wCNgZ+oWHE4GRl/XPP7D7cyNaagKMpl2e123nzzXT7+\n+DOSk7Ow21cDtTCbRxEcvJMjR/7AbG6AzbaXr75aeF2eAFq+fDmDBr1LZuZK1zartQb79m2kdu3a\nZX78fyuVFBRFuWKFhYV07z6An3/egl7vRrVqFUlNPUFe3m6gCrAZD49unD2bgdFYtq80JSQk0KxZ\nFHl5vwJ+wB4sllacOJGC2WxWdwslpAaaFUW5YmazmRUrvmHfvs3s2LGSV16ZgMkUiZYQAO7CZhNO\nnz5d5rGEhobyzDOjsFhux8urA1ZrFF26dMbHpyoWiyft2vVwrQWtlC11p6AoNxC73c68efNISDhI\nkyZhPPjgg9f07P/V2L9/P3fe2Ya8vE1AXeAHfHye4M8/j163GPbt28cff/zBsWPHePrpd5xjDFVw\nc3uUXr2MfPnl3OsSx7+F6j5SlJuYiNCjx0DWrEknN7cTHh4/0KNHGJ9//sl1i+Hjjz9hzJinMZl8\nMRrziI39joiIiOt2/PNGj36KDz6oBpx/Gmo//v49SE8/eN1juZmpuY8U5Sa2b98+1qzZQm7uQcCN\nnJzRfPvtbSQnJ1OrVq3rEsPw4Y8wYEBfMjIyqFmzJu7u7tfluP9fYKA/bm7bKSgQQAf8QrVq1col\nlluNSgqKcoPIzs7GaPQFzr9VbMVorEh2dvZ1jcPb27vcp6seNeoJFi5sw7Fj96JNkreWOXNWlGtM\nt4or6ii02+00bdqUbt26AXD69Gmio6OpV68e7du35+zZs659p02bRnBwMPXr12fVqlWu7Tt27KBR\no0YEBwczZswY1/aCggL69+9PcHAwkZGRJCcnl9a5KcpNpXHjxlitZ9HrZwCHMBhepnJl01XPYvpv\n4OHhwc6dG1mwYDQzZ3Zm//6d3HnnneUd1i3hipLCu+++S2hoqOvllOnTpxMdHc3BgweJiopi+vTp\ngPZY2eLFi0lISCA2NpaRI0e6+rNGjBjB3LlzSUxMJDExkdjYWADmzp1L5cqVSUxMZOzYsYwfP74s\nzlNRbnhWq5VNm1bTokUcvr7R3HPPL2zcGHvLPo7p7u5O7969efDBB9WcSteTXMaxY8ckKipK1q5d\nK127dhURkZCQEDl+/LiIiKSnp0tISIiIiEydOlWmT5/uKtuhQwfZsmWLpKWlSf369V3bv/zySxk+\nfLhrn61bt4qISFFRkfj6+l4UwxWEqShKGcvKypIuXfqJ2WyVSpWqy6JFn5Z3SMpllKTtvOyYwtix\nY3njjTcueEY4IyMDPz8/APz8/FyTaKWlpREZGenaLzAwkNTUVEwm0wWZPiAggNTUVABSU1OpUaMG\nAEajEW9vb06fPn3R2qyTJ092/X+bNm1o06bNVSU/RVGuzZAhI4iLc6OwMJXCwsMMH96NOnVu4+67\n7y7v0BSn9evXs379+muq45JJYenSpVStWpWmTZv+44F0Ot01zXlypYonBUVRrr+4uNUUFOwCKgJ3\nUlAwlLi4NSop3ED+/xfmKVOmXHUdl0wKmzdv5scff2T58uXk5+eTlZXF4MGD8fPz4/jx4/j7+5Oe\nnk7VqlUB7Q7g2LFjrvIpKSkEBgYSEBBASkrKRdvPlzl69CjVq1fHZrORmZl50V2Coijlr2LFypw7\nlwBUBwQ3twR8faPLOyyllF1yoHnq1KkcO3aMpKQkYmJiaNu2LZ9++indu3dn4cKFACxcuJCePXsC\n0L17d2JiYigsLCQpKYnExETCw8Px9/fHy8uL+Ph4RIRPP/2UHj16uMqcr2vJkiVERUWV5fkqilJC\nH300A6t1EGbzaKzWLtSqdYyhQ4deUdlt27bRuPHdVK1ahz59HiQzM7Nsg1VK7koHH9avXy/dunUT\nEZFTp05JVFSUBAcHS3R0tJw5c8a136uvvipBQUESEhIisbGxru3bt2+Xhg0bSlBQkIwePdq1PT8/\nX/r27St169aViIgISUpKKpXBEkVRSt+ePXvkrbfekrlz50pOTs4VlTl69Kh4elYR+EzgoJjND0nr\n1p3LOFJFpGRtp5rmQlGUMvXpp58ycuQysrNjnFuKMBg8yc7OLLc3pm8VapZURblFZGRkEB3di0qV\nAmjYsAU7d+4s75D+kaenJ5AKnG+cMtDrDZhMpnKMSvknapoLRbnJiAjR0T35/feWFBW9y9mzG7j3\n3s4kJu5xPfRxIzlx4iQ222F0utsQaYWHRzwTJryIwWAo79CUv6HuFBTlBiQixMfHs2LFCk6cOHHB\nZydPnuTAgQSKivLQZhHNQ+SOG3JN43nzFvDUU6+Tnz8PkbcxGlfw7LP389//qpkLblTqTkFRbjCb\nNm1i6NDHSU5Ow909DDjA6tU/uqawzs/Pp7DQBngCLYC3yM8/ToUKFcox6r83c+YicnPfAToCYLNl\nsGfPjZe8lL+opKAoN5ClS5fSu/dQCgtHAycoKvoReJkBAx4hKek3AFatWoXJFE1R0XRnqTbYbEG0\natWqvML+R2azCSg+y2s27u635lxONwvVfaQoN5Bx416isHAR8CIwE+gJHCEtLcm1j81mw2gsflfg\nBgjdug1i+fLl1zXey3nxxbFYrWOA94HX8PB4jaefHlneYSmXoJKCcks6ffo0U6a8zMiRT7J06dLy\nDsclJycbCCi2JQDYSljYHa4t3bp1w2RaDcwA4oCuiLQkNrYVvXsP4dNPP72uMV9Kx44dWbYshoED\ndzNkSBIbN66iadOm5R2WcgnqPQXllpOVlUXDhuFkZNxNYWEoVusspk4dy5gxo8o7NJ55ZhKzZm0k\nN3cmcBzoi7+/N1u2rKd27doAnDlzhpCQJvz5ZzbauEITYCAwCqiGXp9MTMxCLBY3nntuKrm5uQwZ\n0o/nn3/uuq21rNwYStR2ltqrc2XoJglTuUnMmTNHrNZeAuL8kyBeXlXLOywR0aaPf+qpCVKtWj2p\nXbuxfPTRR2Kz2S7Y5/333xc3t/4CBoEHBaYJ+AjscZ7PTnFzqygWSxWBnwS2itXaXCZPfrWczkop\nLyVpO9XXBuWWk5ubi91e/Hl+PwoKcsstnuKMRiMzZrzK66+/QKtWLUhLy7hoOU4tfj/gTrQ7hdeB\nQKCRc4+m2O1VyMvrD3QFIsjNncmCBYuv45koNyuVFJRbTqdOnTAavwFigN+wWB6iV6++1zWGc+fO\n8eGHHzJt2jR27NhxwWcTJrzI44+/zqJFjXjttcM0a9aa3Ny/klbXrl0xm78ABgOrgHwgEdjn3GMP\nNtsxwFGs1j+xWCxXHefPP/9Mo0Z3ERBQn0cfHU1+fv5V16HcZMrgjqXU3SRhKjeRTZs2SZMmLSUw\nsIE89th/JC8v77odOysrS4KCGonF0kuMxqfFaq0qP/zwg4iI2Gw2MRrdBY47u4IcotdHSp06jWT8\n+OeloKBARLQJKhs1uksqVw4UP7964uVVTcBT4E5nV9Jjotd7il7/jMAbYrH4y3fffXdVce7fv1+s\nVl+BrwX2isXSQwYMeKjUr4dSdkrSdqqBZkW5zj744AOeeWY9+flLnFvWERAwgpSU3yksLMRi8cTh\nOIf2qClAZ8Adi6WAzp19WLJEe7po3bp1tG/fB5vtXcAdbaD5ISAHSMLHZxdDh95PdnYugwb1pnXr\n1lcV59tvv81zzx2isHCmc8tJ3N2DyMtT017fLErSdqqX1xTlOjt16jSFhcHFttTj3LmzAJjNZipV\nCuTUqQeACcA2IB4IJy/vG77/3oe8vNlYLBbGjXsRm+0V4AFnPUXASKAv0JVz5w5hNpv5+OPplITV\nasVozKCw8PyWdNzcrCWqS7l5qDEFRbnOOnRoj7v7AmAzkIGb29N06NDR9XnTprcDZ4FhwPfAWMAK\n2AFcy9+mp/+JlgjO2wo0BmYDj1FUFMebb87A4Sg+tnDlBgwYQOXKezGbHwJex2rtztSpL5aoLuXm\noZKColxnkZGRzJv3LlWrPoDVGkrXrmbmz5/p+nzixNFYrXvQvvX3Bt4CQrFau/HAA0NdaxCEhzcC\nXgBGA22Aeeh0xf9Je+Fw2Evc9ert7c3u3VuYNCmIJ544zpIlsxg58vES1aXcPNSYgqLcgNavX8/b\nb8/h7NlM9PoiDAYPoqPvZty4J11TTqelpVGvXhg5OTpgKtrdxcvAGKAbFss0Onf2do1BgDbD6qlT\np7jtttswm9UcRP92apEdRbnJFBQUYLPZLtp+zz33YDLpiY/fwrZtCWRkpPHII0MvWIOgevXq3HZb\nGLAQeBx4DhiFu/uHGAwdsVr3MGhQL9f+//3vFKpVu40mTdoTEBBMQkJCmZ+fcvNRSUFRykF2djYd\nOvTCw8MLi6UCEye+eME3ukcffYxvvvmGgoJscnN9OXAghMcff4oDBw7wwAOP0rZtd559drzz/YXz\ng7+HgE/Iz78Hu70tp04VcP/9j7NmzRqWLl3KtGmzsNkOUlCQzMmTk+jYsU95nLpyg1PdR4pSDoYM\neZzFi7MoKFgAnMFqjeaTTyYwcOBAfv/9dxo2jMRuXws0RRtTmEtgoJCZeZpz5zoBPwCh6PUJ6PWV\nsNlmAc8CFdEeYR0BTAe2MGBAECdPphIXVx2Y54ygCHDDbrep+ZD+xVT3kaKUkz179tCyZSeCgu5g\n5Min/vHN382bN/POO++wfPlqCgqeBcyAH7m5jxAX9z8AfvnlFwyGdsAdgA54CjiM2Qw5OQOBZcBS\nYBMOxx/o9bn4+o4A/kRLCHuBlkB9IAt3dzN2uw7YCJxzRrIcg8FLJQTlIupvhKJco5SUFFq2jGbT\npl4cOfIxCxYc4YEHHr1ov/fem0V0dH+ee+4Qp0/nor1/ACC4uW2jVq1qAFSrVg2TaT/a9BUAe9Dp\ndERFtcLNt36WAAAgAElEQVTh2ATY0KbMLgS8MZujOHfuDLAObVzhM6Am8CIm0yHGjh3BsGH3YzDk\nACHAPcBgunfvVFaXRLmZlcq71GXsJglTuUV98skn4uFxf7FZV8+JwWC+YHbTgoICMZmsAkec+2wX\nsIrB0EV0uuai03lK5859JCsrSxwOh/TuPVg8PcPEYhkgbm6+MmfOJ1KzZgOBhwW+Eegg0FcgUSwW\nPzEY3AXOFovhAfHzqym7du0SERGHwyEvvPCyGI0W0etN0q1bv+s6tYdSPkrSdqo3mhXlGmnvDZxG\n676pAJzFYDBd0DWTlZWFTmcCaju33Ak0dI4bvADcz5o1ExgyZATffvsZX3+9kLi4ONLS0mje/HnS\n09M5c6YCMAetS6kz4INe/wPdu/cmMzOf9esfJT//JeA3rNZYNm7cTHBwMIWFhZw5cwY3NzM1atTD\nbHZj8OC+rvcdFOUCZZCcSt1NEqZyizpw4IAYjZWcE9K5idFYXZ5//qUL9nE4HFK3bhOBlwRyBVYK\nWAXGFft2nyaenr5/e4zVq1eLl1eLYvsWCFjFbB4g7u7DxMvLT3r2HCR+fnUlLCxSNm7cKCIic+bM\nEzc3TzEavUWnqyiwWCBWrNbqsnr16jK/Nkr5KknbqcYUFOUaDRs2GpGRQBZwGL3eSMuWERfso9Pp\niIv7AaPxHcAbGALcBvyGNi2FH1AHm01HTk7ORce466678PHJxGR6CliOXt8bqEth4Rfk588jO/s/\nmEwmjh9PZO/eLbRs2ZL4+HhGjRpHQcGX2GxnEHkZeAXoQG7ucyxa9HUZXhXlZqWSgqJco127tmG3\nP4nWrRNAUVF/3n//fc6ePXvBfrVq1eLrr+ei07kDVYCH0QabnwfWAMex21vz6KNjXGVEhPT0dAoL\nC4mPX8fAgblERLyLv/8hYKLzmOBwNOD48VOucklJSXTo0IuCgipo02UMRHvBTRvA1ulOUKGCmtxO\nuZhKCopyjapVqwlscP5UhMjPrFz5J40bR3LmzBnXfocOHeLUqVPodAXAerSJ7h4CngAaAt4UFU1n\n9eo4AE6fPk2zZq2pU6cRVasGMmnSKyxY8CFbt65k3LgRWK1vAsnAH1itU7nvvg6uYz344EjOnRsD\n/A4cBNKB/wIewGQ8PWfz5JMjy/KyKDepSyaF/Px8IiIiuP322wkNDWXChAmA9pc1OjqaevXq0b59\n+wu+EU2bNo3g4GDq16/PqlWrXNt37NhBo0aNCA4OZsyYv74JFRQU0L9/f4KDg4mMjCQ5Obm0z1FR\nytRnn32Ep+cIdLq2aLOU+lFU9D8yMiL4+OPZAKxYsYImTVowZswKtElLKzhLVwd+LVbbPipVqgzA\no48+yd69DcnPP0FRUSqff76VmTNnMm7cBDZv3sk99/jj4XEHHh7NeOKJDowZ84SrlgMHDuBw3Of8\nyR3ojMEwi86dW/PUU8KuXdogtKJc5HKDDjk5OSKiLSgeEREhGzdulGeeeUZee+01ERGZPn26jB8/\nXkRE9u3bJ02aNJHCwkJJSkqSoKAgcTgcIiLSvHlziY+PFxGRTp06yYoVK0REZObMmTJixAgREYmJ\niZH+/fuXymCJolxPaWlp4uVVVWCBwGSBugIB0r59ZxERqVDBVyBE4DaBYIE+AmucK6V5ik7XVszm\nx8Rq9ZW1a9eKiEhgYAOBPcUGl98WDw8/MZuHC8wRD4+mMm7cxL+Np02brmIwTBZwCGSLm1u469+s\ncusoSdt5xSVycnKkWbNmsnfvXgkJCZHjx4+LiEh6erqEhISIiMjUqVNl+vTprjIdOnSQLVu2SFpa\nmtSvX9+1/csvv5Thw4e79tm6dauIaInH1/fipy9UUlDKSmFhoeuLy7X4/PMvpHLlugK+Arc7G/PN\n4uZWU1577TWBCs4njg4IdBIIEIPBR+ARgd+c7x+YXA33rl27xNu7lkADgVcECsVk6ipmc0ixJJEh\nRqP7Be9DnHf06FGpVStUPD3ribt7FRkwYJjY7fZrPk/l5lKStvOy7yk4HA7uuOMODh8+zIgRIwgL\nCyMjIwM/Pz8A/Pz8yMjIALSpfCMjI11lAwMDSU1NxWQyERgY6NoeEBBAamoqAKmpqdSoUQMAo9GI\nt7c3p0+fxsfH54I4Jk+e7Pr/Nm3a0KZNm6u/LVIUp5SUFDp37se+fb9gsXgxb95H9OvXt0R1ffll\nDI88MoG8vPfR+u3fAhoBUFAwiQULZqEtldneWeIDoAnam8kfAQbgE+BPXnjhZZKT05g1azbwBtAA\nmIDROIvAQB9OnAgrthKaJyIOHA7HBbOnAtSoUYODB3dx8OBBPD09qVWrlmtxHuXfa/369axfv/6a\n6rhsUtDr9fz6669kZmbSoUMH1q1bd8HnOp3uuvxlK54UFOVade3an4SEDjgcG8nJ2c2wYZ0IDW1A\nw4YNr7iOzMxM0tLSmDBhGnl5bwPdgFnAMdc+ev0xKlSwYjanFGvMj1Gpkg9ZWRnACaAa4ADSEfFk\n9uzP0ZbYPD9G8C1GY0N+/jmehg2bo9O9h0gz3N1fp1OnPphMpr+Nz2w2X9X5KDe///+FecqUKVdd\nxxU/feTt7U2XLl3YsWMHfn5+HD9+HID09HSqVq0KaHcAx4799Q8iJSWFwMBAAgICSElJuWj7+TJH\njx4FwGazkZmZedFdgqKUpqKiIvbsicdun4T2Lf0OoAtbtmy54joWLfoMf/9aNG/eneTkRGCH85Pq\naDOUTgCewGr9iJkz38XTcx063WC0+Yj68OGHb/Dkk2OBCOBVoAegx2Y7h83Wkr/mPQLIpbDQRocO\n/Rg27AFat15NvXpP8vDDdfnii7nXejkU5QKXTAonT550PVmUl5fH6tWradq0Kd27d2fhwoUALFy4\nkJ49ewLQvXt3YmJiKCwsJCkpicTERMLDw/H398fLy4v4+HhEhE8//ZQePXq4ypyva8mSJURFRZXZ\nySoKaN2Unp6VgN3OLUXo9bvx9/e/bNnY2FjatevOQw89QX7+ZnJyEoGf0LqMnkB732AZ2pQXy7Db\ndcycOZPs7HOIOIBDGAyeHD9+ghkzpjFiRHcMhum4uyfi7n6YLl2i0RZEW4eWWBYCHXA4upKQMI33\n319Jenoac+e+xQcfzFBTVSil71IDDnv27JGmTZtKkyZNpFGjRvL666+LiMipU6ckKipKgoODJTo6\nWs6cOeMq8+qrr0pQUJCEhIRIbGysa/v27dulYcOGEhQUJKNHj3Ztz8/Pl759+0rdunUlIiJCkpKS\nSmWwRFEuZfHir8RiqSJW61Dx9LxDOnToddmB2O+//14slmoCTwrcU2zAVwSqCzQSGCHwiXNg+W6B\nigLhAkECPQSKBNZJ/foRIiJy5MgRadSohbi5eUlQ0O2yefNmCQlpKlbrXWI0hopO5+0s53Ae51eB\nQLFaq8jPP/98PS6VchMrSdupFtlRbln79u1jy5Yt+Pv707lz539cW8Bms7FkyRKeffZFjh17GOiJ\nNv30DiAQ2AnchdnsRWGhHm0Bm6fQXiz7Ae2OpDIQDTwIVKFx4zfYvn0Nfn61OHPGA6gENKVSpaXs\n27edjRs3kpOTw6+/7mHmTLDb33ZG8z+07qlRdOu2gR9//KKMro7yb1CStlPNkqrcUIqKinj11ddZ\nvz6eoKAaTJ8+mSpVqpTJscLCwggLC7vkPjabjbZtu7JzZw65uW2AGWgDw88BjTCZ6mIyJTFnznya\nN29Go0YtKSj4CjjfDapDe7JoEloi+Rq9fhvPPvse77zzLmfO6IGPATvwKPn5Xhw4cIB+/foBkJyc\nzMKFEWRmWtDWSJgGvAQ4sNnspXtBFAWVFC6QnZ3N77//TpUqVahVq1Z5h3NLGjToYZYtO0Fe3nA2\nb/6ZuLhWJCRsx8PDo1zi+emnn9i16yw5OZvQBqVHoa1qNhM3NwPTpg1iwIABVKumLZBjNhspKKhW\nrIbqQA7aGMMXQFV0uuZ8//0qjh5NB9517pMHTKagYDwVK1Z0la5VqxY7dmxi5MixrFkzG7v9EcAd\nq3UsY8bML/sLoNx6SrkLq0xcjzB37dollSpVFy+vJuLu7itjxz5X5sdULpSVlSVGo0Ugx9VXX6FC\na/npp5/KLabZs2eL1Tq02NhBkYBB7r67k6xateqi/bt37ysQ4Xx5bYWAt3OKbDcBH+fLaEFStept\nEhXV07lvgHN7DYEKrrf9/79vvvlGWrToKC1bdpalS5eW9akr/wIlaTvVhHhOvXo9wJkzr5OV9Sv5\n+QeYPfsb4uLiyjusW472zkvxv5b6MhtPys/P59lnJ9GiRUcefHA4GRkZbNiwgZiYGA4dOgRAy5Yt\nEVmKtr5xLkbjRJo3v4f//W850dHRABQWFjJx4mRatOhIUZEdgyER6IvWZbQAvd4bCAJSgQTgfvLy\nbISF1UK7+zgM7EObybQBffs+QEFBwUXx3nfffWzevIKNG5fRpUuXMrkmiqKSAtr0xEeP/o72DxnA\nB5utHfv37y/PsG45FSpUoFOn7lgsfYGlGI3j8fJKKbO313v1up8PPtjH1q2j+fJLD+rUaUKXLsN5\n7LFvaNKkBd9//wMNGjTgq6/m4+v7AAZDJZo3/5Uff/zygnoGDBjGO+/8wtato1m9uh56vQOzuRkw\nBTe3n6hY0QQMQ5uYDmAAbm5Gzp0rBAYAbmhjD/cD2Tgc7qxYsYLffvtNPWChXH+lfbtSFq5HmLVr\nhwl85uwiOCUeHsFqZapykJ+fL+PGTZQ77rhXBg16WNLT08vkOH/++aeYzV4C+c7f+WrnZHW5zp+3\niYeHz2XnRfrjjz+c6yP/1eXl4XGvVKpUXUwmP6lWLVief/55sVpbCeQ59xkiBoNVgoJCxN39XmcM\nDoH/CjQX8BKdrpIYjb7Srl13KSoqKpNroPz7laTtVHcKTt999xk+PuPx8rodd/d6PPZYb9q1a1fe\nYd1yli5dxsyZs/j99wMsX76cI0eOlMlxtG4qcf4BOIr2ZrPF+XMz8vOzyc/P/7viFBUV0afPYOrV\na4zdrgfuQxtQhtzcbDIz21BUtIETJ4Ywf/5iIiKsmEz+gBfwA3b7KA4fro7NthuTqRZQC51uJudX\nYhPZhM0Wxdq1vzJ79uwyuQaK8nfUewrFZGdnc+DAAXx9fdXTR+UgJSWFkJDbyc1dhdZAL8Pb+xGO\nH08qkzd3u3cfQFzcOfLyHsZoXILdvgyRzUAYOt171Kkzl0OHdv9t2alTX+fVV+PIzf0B7SG+AYAd\no/E27Pb5iKQAngBYLLUROYfd3oaiot1oy3DGovXe3sXrr9/H7bffzqpVq5gxIw343HmUbKAyw4cP\n56OP3iv181f+/UrSdqo7hWI8PT258847VUIoJwkJCZhMTdASAkAXbDa3C+bNKk1Llizi6acjaNNm\nIcOG+fLBB6/h7t4Ck6kCtWrNZsWKJf9Y9n//20Fu7lC0OwsT8AQeHjvp2fNP3Nzc0AaQAfLJyztJ\nfv4PFBV9g7YcZibwI9o4Qn0sFgvR0dE0btwYgyG92FHSAQN33KEmtVOuH3WnoNwwDhw4QNOmrcnL\n2422kH0C7u538eefKXh6epbZcQsLCzFrEw5ht9s5d+4c3t7el5z9d+zY8cyadZLCwk8AHUbjRHr2\nTOWrrxbQp89gYmOPkpvbDYvlJ/LyNqG9h2B2ln4IqAc0w2zux86dGwkLCyMnJ4ewsHCSkxsAzYF3\nadKkJjt3bv7Ht60V5VJK0naqpKDcUKZMmcZrr72HyXQ7RUXb+fjjdxg8+P4yOdbGjRvp3fsBTp5M\nJSCgLk899Tg//7wDLy8rEyaMpX79+v9YNjMzkxYtokhJ0aHTuVGhwgm2bVtP9erVsdvtvPHGG2za\ntI3w8KZ8++1qfvstCrv9BeB3dLq7MBgKqVTJj4ULZ9KpUydXvVlZWXzwwQckJCTStu09DBs2TK2D\noJRYidrOUhrkLlM3SZhKKdm7d6/8+OOPcvjw4VKv+9tvv5UaNRqIt3c1MRorCjwuUEvAIlBVYK7o\ndC+Lp2cVWblypaSmpoqI9nJjjRqhotN5iMFQSaKiusqRI0ekffuu4uHhI9Wr15N58+bLvn37ZM6c\nuWKx+Iq3dzuxWKrKxIkvSFhYhBgMbmIwWMRiqSQBASGyePFXIiLicDgkKSlJDh069I+T8n3yyTzx\n968rlSoFyBNPPCWFhYWlfm2Uf5+StJ03RWurkoJyNTIzM+WppyZISEi43H13R9m+fbuIiMTHx4vV\n6iewTuAPgTsEAgW2C4QK/FzszeWnRK+vIEajRdq16yQmk5do6y+nCEx0vp3sKW5urQWSBJ4VcBeL\nJciZYBY5H1NNFoulihw6dEgefPAxcXfv4tx/g1gs/rJmzRqJiuomFoufWK3VJTz8XsnKyrrgfFas\nWCFWa02BrQKHxGq9V8aN+295XFrlJqOSgnJLO3PmjLRs2UF0OpOASaC/wBNisVSSX3/9VZo0aS7Q\nWGCMwBKBjgIfO5NAmMAvxZLCCwKVnNs9RZv++vxnDtGmxHZ3TmPR2fnfg87PP3N+7ibgIe7uIRIX\nFyeVK9cUSLzgGHfd1UYslh4ChQI2cXMbLI8//uQF5/Xww08IvFWs3C9Su3bjcrrKys2kJG2nGr1S\n/jWGDRvFtm01EckFkoBNwFby8hxERLRl9+5gtCeb5gPvA/HAYmfpocAQYBXawjZvoj2c9yTwGJCG\nNiU2wEnn/z8BFKItiBPu3H8S8DzQHcgFdpGf/yfZ2dl4eno549KYzUmcOJFFXt4gtCeYDBQUPMAv\nv+y54Lx8fLwwGpOKbUnC29v7mq6VovwTlRSU66qoqIizZ8+W6oMDIsLy5ctZvTqOwsJn0d4bCABG\nAm2B7RQU5KItfRmD9jhoZbTZSbejrY9wDPgDbW3kJ9EeNf0AeAR4A/AF7gZeBFqjrZfQwrnfQ2hr\nKjQHCtDWTFiOtvZBMDrdAB56aARWqxWzuT863VOYzfdTpcpW2rVriZvbMrQ1mgWzeSlhYcHExCwm\nLOwuQkLC8fKqQMWK32M2P4RePx6rdSRvv331a+8qyhUp9fuVMnCThKlcxltvvScmk1VMJk8JDW0u\nKSkppVLviBFjxcOjgUBtgU+LdfH0EHjb+XM1AS/RVkCrK/CSwA7nQHOgwEABX+efDc7unyXFumy6\nCxic9SwSyHB2RTURmO4crH6x2P6LnN1TRc6xi1cF+gkYRa+3iJ9fHdm7d69kZWVJkyZ3iadnqFSo\n0FiCg2+XL774QqzWQIFYgQ2i198mkZGt5OWXX5YpU16S3bt3l8p1U/79StJ23hStrUoKN78NGzaI\n1VrDOcjqEIPhBYmIiLrmeg8dOiTu7lUENgvc7+zb7yja0pi3izaX0UrnGMNtoi2T2fT/jQ8ECMwW\ng8FLoKFAgYDRmSC+Fm35zVqiLbnZzzlW4CbaVNcDnXVWFJhdrN7VAjUFgp1jDmucP//hPP/nJTKy\nnYiIFBUVydatW2Xz5s2Sn58vvXoN/n91LRedro40bBihnjpSrkpJ2k7VfVQKVqxYQWhoJLVqNeK5\n517AZrOVd0g3nPj4eIqK+gC1AR12+1Ps2rXV9bnD4eCVV16jdu3GhIQ05+uvl3Dq1Cl69XqAwMAG\n3HNPJxITEy+q9+TJkxgMldH68IOBR4F1mM2H0N4eDnR+5oY2z5Eb2jKZhc4a8oBzwJPY7ZWAA2jd\nQ2agH9oYwWdo3ULnnJ9XABqgTUMRizb7qR2YDGxA65Iahfbm8mFgLvAL0AeoddH5G41GIiIiaNGi\nBW5ubnh4uKPTnSp2lqcRCeGPP86xa9euq7zyinKVyiA5lbobOcytW7eK1VpV4EeBHWK13iNPPz2x\nvMO64Xz55Zfi4dHC+ZSNCCyTwMAQ1+fjx08UgyFAtMXuh4q7u78EBTUWk2mUwG+i070tvr415MyZ\nMxfUqy3M41Osm0gEPnJ+4+/ovAtoKfCm87Ns52e3CQwV7VFUL+c3+yoCjzrvBqyizVg61nn38aBo\nTyrVFjhdrIuogmiPoD7i7Fq6zbmPl0AbAQ8xm8Ocx7qj2PkvlRo16v/ttdqzZ49YrZVFewLqNWdc\nceLpGSZbt24t09+T8u9Skrbzxm1ti7mRk8IzzzwnMLlYg7RX/P2DyzusG47NZpP27XuKp2cjqVCh\nl3h4VJH169eLiMipU6ecXTcvCCwVaCfQXAwGbwG769p6ed0rsbGxsn37dnn44ZFy330DJDKypWiP\njn5e7HfwjUAzgRBnl1J1geRin08R0DsbdA9nY91TYGaxfcYLdHEmFZMzMbwk8ESxffJE62Y6LlDf\nGb+XsxEPEfhJ4FXR6dxEp6srUEegnuh00eLhUUU2bNjwj9frt99+Ez+/OqLXhwrMELP5MQkNba66\nj5SrUpK2U63RfI08PKwYjcf5q8coA4vFcqkitySDwcCKFd+wbt06Tp06RYsW71KjRg0AVq5ciUg4\ncP6JmlaADyImtC4bb8COw/EnR44coVev+8nLq4G27rE7WlfQ02hPExmAsWgzkZ4GVgKhwFfAOLTp\nrb8F5gANga7AXrQnlkKKRRwKfIzW5XMv2uOr36A9Znoa8EF7nDUUbZ6m/miPptqccZxBe7z1XUR+\nAyKB0cAa3NweZ/78mbRq1eofr1fDhg05fHgPEydO4Zdf4mjYMJjXXluJyWS6iquuKFdPzX10jdLS\n0mjUKJzMzN7Y7QFYLO+wYMG79OvX9/KFb2BJSUls3LiRSpUq0alTJ4zGsvv+EBMTw7BhC8nPX+Hc\nkglUpk+fgSxffoDc3EG4u6+lceNsKlb0YdUqI1pCWIHW9/8KMBuoijZu0AP4CMgH7kTrzwctuZxF\nW/tgLtoT2U+gTVVtA+oD36GNM3RESzZDgGVoj6N+jPZoarbzWNlAHFpiiAJ2OfcZCGQBdwGvAQuA\nlsAYQPDwaMQddwSSnV1A27Z38eqrL5CUlMTTT7/I8eN/0q1bFJMmjS/Ta36rEhFWr15Neno6zZs3\nJzQ0tLxDKlNq7qNykpKSIhMmTJIRI8bI2rVryzuca7Zu3Trx8PAVT8+B4ukZLi1atCvTboszZ86I\nv38dMRieEVgiRmOk9O07WBwOh8yfP186dOgh7u5eAjoxGHycXTZuzv5+u8A+Z5fN+W6dz8Rg8BKT\nqYpzXKCvc1whRLT5jbY497OLNoZhFnCTatXqiMXiI25uFUV7i/mcc79MZxeVu7O7ydPZjeTjrPc2\nqV69njOunGJxjBSdroMYjV7i7t5e4BsxmR4Ug6GCGAxTBFaLxdJVOnbsJd7efqLTvSmwSqzWNvLI\nI6PK7HrfqhwOh9x33wPi6dlQPD3vF4ulisTELC7vsMpUSdrOG7u1dbrRk8K/Tc2aoc6+fRGwidXa\nVubNm1emx0xLS5OhQx+Xe+/tIdOmvSE2m01ERA4fPixWq69o8xIVCUwV7XHTEwJ3iTaA/Iyzgb5X\n4AHRBn59nQ15NdEGiLuJNvD8tfOzh5zjDhWd+/s49zeLwVBZtEdJzzfuJ4slgk+cyWWqaO8qjBe9\n3ltSU1Od4wnnp834UyBAmjQJl99//10mTnxR2rTpLu3bdxFPz07F6s4Vvd4s7u4Di207IWazx2WX\nAlWuzqpVq8TTM0z+Whb1V7FaK/6rr3NJ2k51f6pc5OTJdLRpGwAM5Oc3Iy0trUyPWa1aNebP//Ci\n7du2bcNgaAPc49zyHDAVbVqIp4HhaI+DNkNb6ewIWhfPd0Bv4HW0R0oT0R5F7QF8idbllIXWvTMR\nbUyhCRCJ3T4PrXuqGdoYRZ7z+B84Y7gX7e3lCUATfH39MJvNmEx2ioqmAm8DxzGZvHnzzanUrl2b\nV155EZ1Ox7fffsuWLcXPswgQdLrCYtsK0OnU0+KlLT09He13fH4Vv8YUFOSRl5eH1Wotx8huLOpv\nnnKR8PC7MRqnojW2R3B3j+Huu+8ucX2ZmZmsXr2aTZs2Ybfbr7hcUlISf/zxB3b7XrT3BEBr3AXw\nxGD4lYoV3dHGFaoAW4DdaA3+nWhjAqPQ+vs/QmvIx6ENLv9fe+cdHVXVtfFnek8jlRQCCQRIIAld\nmijNCIJ0BAHpIF2l6atgQYpSRX0VUFExQX0V8FNAUJpUEUSQjqFDgEBCQsokmef749wMiYAkITBE\nzm8t1mLunHvuvnfg7Ht2/Q2irMVwCGUSDpFz8CqAr5Xr9YRQKp4oWLNI5ClkAvgQavUwfPDBdHh7\ne6Nt2/YwGsMB9IVeH4egIBsGDhwFs9kKT88ArFq1Cq1atYKn52nodCMBxMNsbotu3Z6G1forNJqJ\nAL6A2dwOo0ePkn0USpm6desiL28NhO+HUKtnIiysulQIf+cu7FhKnTIi5r+GCxcusG7dZtRo9NTr\nLZw3790Sz3X48GF6e4fQza0prdZqbNy4NbOzs2973qJFn9Bk8qabWwuq1R7U6yNpNA4g4EG9vi4t\nlk60Wn0VG/8RxZzThSJ8VE0RhhpJoHMBs0yy4j/woQgjHqqYhWpQZCbnj/uAokRG/ucUxV8wXzFj\nNSFQkWq1J7/++munzHa7na+/PpWtW3fmiBHPs3z5MKpU71JkTW+i2ezNEydO8MKFCxw6dDQfe6wL\n3357NnNzc3nq1Cn26/cs4+K68r33/vuvNmm4ki+//Ipmswc1Gj2rVIm9Kz077idKsnb+4xknT55k\ns2bNWL16dUZGRnLu3LkkRVx5ixYtWLlyZbZs2bJQQtGbb77J8PBwRkREcPXq1c7jO3fuZFRUFMPD\nwzly5Ejn8aysLHbt2pXh4eGsX78+jx8/Xio3JrlzMjMzb9n0pag0bvwY1er8xLIcmkxxnD17DvPy\n8piWlnbTcy5fvkyj0Z3AAeW849TrbXz99de5Zs0aLly4kMOGDWPt2rUpyk+EE3hU8QtMJtBHWfht\nih8hf3E/SeGgNisK4RWK/AMdr9c6OkGgKUWuRP55Z6lS6RkZ2YAqlRd1Ol+aTB5ctmw5STIjI+MG\nRdFhMKUAACAASURBVHfu3Dkajd4F5iDd3Nry22+/vaPnKblzHA4HMzIyXC3GPaHUlcK5c+e4e/du\nkmRaWhqrVKnC/fv3c+zYsZw+fTpJctq0aRw/fjxJ8s8//2R0dDTtdjsTExMZFhbmfOOpW7cut2/f\nTpKMi4vjypUrSZLvvvsuhw4dSpJMSEhgt27dSuXGJPcH/v6VKaKD8hfHmWzZsg3NZg9qtUZWrBjJ\nw4cPFzpn7969tNmqFjgnhwZDJdaq1Yz9+w9jkyataTLVpkbTjyIiqBtFcbq1Bc7pS6CBspMYTuBj\niuJ1MRRRS/njvibgpziJ3ZWdg6eiVEYp58VQr/ckKf4fHDhwgGlpabx27Rofe6wjNRoDtVoDR44c\n6/z3npmZSYPByuv9E9JpsVTi1q1b7/lvIHlwKXWl8Hfat2/PNWvWMCIigufPnycpFEdEhChX8Oab\nb3LatGnO8a1bt+bWrVt59uxZVq16PaU/Pj6egwcPdo7JT93Pycmht7f3jUJKpVAmuXDhAk0mP2VR\ndhD4g3p9Nep0VgK7CDioUr3DChWqFzKXpKWl0WbzoSgiRwIdKZrcfEmVapiy0HsRaEugF4EXKLKF\n83cWDsUkFEdRWC5KWeTrUoSkFiyJsU1RBl4UhesqKYomkMAgiozoFxgUVLgkxaFDh1izZgNqNB0I\nZBG4RLO5NhcsWOgc89//LqDZHECLpTctlmrs1WuQNAtJ7iklWTuLHH10/Phx7N69G/Xr10dSUhL8\n/PwAAH5+fkhKSgIgErkaNGjgPCcoKAhnzpyBTqdDUFCQ83hgYCDOnDkDADhz5owzs1Wr1cLd3R2X\nL1+Gl5dXoetPnjzZ+fdmzZqhWbNmRRVd4iLGjn0FOTntAeyA6F+QC7vdHYAKov9ALMhhOHt2Aq5e\nvepsHGM0GvHppx+id+/uyMszICPjPEQWsQ1kF4hCd3EA3obIUAaA1hBO4/nK3MchCtPpIXoihENE\nG30MEbEUq8g0CsBQiCY6KyAil3oCcINIOosG8C06dRrovK/9+/ejfv1mSE+3AJgDUWTPgIyMQfjp\npy0YMKA/AGDw4AGoV682du/ejZCQXmjevLl0HkvuKuvXr8f69evvaI4iKYX09HR06tQJc+fOhc1m\nK/SdSqW6J//QCyoFSdngyJETyM0dCqAZRKOaDRCROzMgwkmrQUQRaWC1WgEAn376OQYPfhYOhwp+\nfv4YO3Y4Ro58HoX/qeohyliMBrAcIprkF4gM5vxQWhUKB9dpIMpaPA4RhvoMRAmNoRBKog5EqGt9\n5fsdEA10IgGcw/z5n6NZs2Z48skn8cYbM5Ge3gHAHwC2KOcQev02hIYGFnoGsbGxiI2NLeETlEiK\nx99fmF99tfjNmG4bkpqTk4NOnTqhV69eePLJJwGI3cH58+cBiNhfX19fAGIHcOrUKee5p0+fRlBQ\nEAIDA3H69Okbjuefc/LkSQBAbm4uUlNTb9glSMomTZrUgcm0AMBeiPLV+S8UPQHkQKPpBbO5JT7+\n+ENoNBrs27cPQ4Y8j6ysbbDbU3H69AiMHDkRYnHuBBEy+gqAgxBhpocBnIPolnYe4m0/FiKclBD1\niFZDlLK4CLGAH4Uoj3EeQomEQNQsagyhIGpBtOPcD1EeYz6AX5GXp8Hs2R/C4XDg5583AfgZgBmi\ntPbjUKsbICRkFyZMeOFuPEqJ5J7xj0qBJPr374/q1atj9OjRzuPt2rXD4sWLAQCLFy92Kot27doh\nISEBdrsdiYmJOHLkCOrVqwd/f3+4ublh+/btIInPPvsM7du3v2Gur7/+Gs2bN78rNyq597z66kt4\n5BEDNJo5uF5MDhAF6QIREqLFzp3r0b17NwDAb7/9Bo2mJUQtIYAcBrFwV4N4k58EYB7EzmMQgJ8g\nCs/FQCiBoRCF6s5CFKP7CaIIXRZE/oFdGZsK4DTE7uIlAO9BtOOkIl8KxE7DT/lsAxAIjYb45JNP\nkJJigiii9xOEiWk9+vePxZ49W2TvZEnZ558cDps2baJKpWJ0dDRjYmIYExPDlStXMjk5mc2bN79p\nSOqUKVMYFhbGiIgIrlq1ynk8PyQ1LCyMI0aMcB7Pyspily5dnCGpiYmJpeIskdw/JCUl0WDwJuBP\n0fUslMAw9u07qNC4n376iRZLZYoWmS0J9KAoQVFOcSi35PVcBLMSMVSOoqNZTYoQ1EMFnMhvEBhX\n4HNNxal8tMCxV5XzbIpT+R2Klp0milLaaRRluc0MCKhIrdZMUSKjYPlsDS9duuSipyuR3JqSrJ2y\nSqrknvDee//FmDFvwm4fCsAIk2kKtm9fhxo1ajjH2O12+PqGITW1BUTXsw8hzD8GAA9BvPW/A2HW\nyYaoUuqh/L0/hHnoZYhSFgDQBcKPMAnCyRwCsRv4HMATzjEazf+B1MLh8IfYzaihUtWGVrsJublp\nMJm84OVlw9mz/eBwPKzM+wtEF7k3oVa/i3LldNixYwNCQ0NL/dlJJCWlJGunrH30L2XLli1Yu/Yn\nlCvnhWeeeQYWi+WuXSszMxNJSUkICAiAwWC46Zhnnx0Cm82Kjz76ChaLCZMnryqkEABg3759yMuz\nAvgIwiw0DKJ0dU0AcyHs/PEQ/QrmA+gBUeOoP0SE0ZMQ/Q86QfgMdkD0QbgEUf7aDqEg+kH4G45B\no9mCChWq4K+/RinHCaATyEZQqXZAr1dBp6uE06d/B1BFmRMQZba1AMLhcGzB5cufY9SoF7F8+Rd3\n9CzvFiSRlJQElUoFX19fGQUluTWlu1m5O5QRMe8bPv98Cc3mAKrVE2gyPcmIiFheu3btrlzr66//\nR5PJg2ZzEN3d/bhx48YSz7Vr1y6aTGEEcgm8T9EWM99Mc1DJNdhOILjAcRKoRmCi8vc/KJLRrBSt\nNjsopiAN1WpPinIWG5W59VSpyimmo4MF5ptO0X7Tg8AZAseVnIf8qqublD/BFC05SeAnRkc3LcUn\nW3pkZmayVasnaTB40mDwZKtWTzIzM9PVYknuASVZO8vEaiuVQvHw8goisIP5iVxmcxsuWLCg1K9z\n+vRppZfwb8q1VtLd3a9EJQQuXrzImJjGFHWLzBT1hfoWWKiTlIW+ofL9RV6vS1SOIuHsYUVB+BMI\noihpYSGgpUqlo5dXqDLGXTm+TDnfXfET5FC01gylWl2FanUzin7MFShaca6lSHDrplw7gaIkdyZN\npvYcM2ZCqT/j0mDs2JdoMj1JIJtANk2m9hw37j+uFktyDyjJ2imrpP4LuXYtBUCY8kmF3NwwpKam\nlvp1Dhw4AJ2uBkQYJwA8hrw8S6Gw5KLSrVs//P57FET10R0Q0T1LIUxBWyHMQjoAHlCrQ6HR1IJK\nNRBADQifQgCA3wGchIhA+j+IKq8VAZwBeRFXr4ZAqz0I4H2IcNJqEOGyQQDOQEQZVYBKlY1WrSJg\nMOwH8AVENNTLEGGw/4MwWU2AMHOthUbjgZYtTXjzzUnO+8nKykLfvs/C27sCQkNrYMWKFcV+JqXF\n5s27kJnZDyK/Q4/MzH7YvHmXy+SR3N9IpfAvpFWrNjAYRkGEZv4ErTYeLVq0KPXrVKhQAXb7nxC5\nAgCwH7m5yfD39y/2XJs3r4dIHtNDJIw9A2AIgINQq+OgVl8EMAwq1S7Ur++PxYunon37S1Crz0Ms\n/HaIvgh/Kuc+AZGxPFGRbylyc1vA398HQUGvQiSutQbQBkKRvAvhe9gHgyEbCxe+j1mz3oBONw4i\nmzofO4TfYRGAKjAYjFi9+nssXx4Po9HoHDV48GgkJJxBcvLPOHFiNrp3H4idO3cW+7mUBlWrVoRO\n96MiN6HTrUHVqhVdIoukDHAXdiylThkR877h6tWr7NjxadpsvgwKqsrvvvvurl3rjTem02Tyo7t7\nK5pM3vz0089LNE9AQDiv1zrKo6h6+gmBH6lWl6eoZ0QCx6jTmenu7k+Npj2B5hShqdX+5mcIoQhh\nfUzxMfQjEEkvrwq0WMoRWKeM203ASK3Wg25urWgy+XDOnPlOuU6cOMHy5StT1EFaopi1hijy+RHQ\nUqt15wcffMBVq1Y5Q1Pd3QMUX4SQR62ewMmTXy2VZ15cLl68yLCwGrTZ6tFmq8ewsJoyhPYBoSRr\npwxJlZSYffv24YsvEnDlyhXUrh2L5s2bo2LFkr2Brly5Ek8+2QN2e2uI7mkWiHDUTwFMBXBMGWmH\nqE/0HwDjlGOjILKQz0KYha5B7B4eh4hW2gWx+7ADqA6DIRPZ2Wec17bZGuL994fBw8MDlStXRpUq\nVQrJlpycDF/finA4HoFInOsPEY76AYRZ63sA3WGzxUKjOYb161eiTZtuOHNmAUQo7UQAC2G1GjFj\nxqsYOnRQiZ7RnZCZmYktW7ZApVKhYcOGhXY1kn8vJVo7S1kx3RXKiJj/Kn799Vc2bNialSvX4XPP\nTaTdbi/0/bZt22g2e1OlepFq9fO02Xx44MCBO7rmgQMH2KFDR4pGOAMpqp96Kg7j/ylv3r1oMPiz\ncJnsz5RIoRoEXqNWG02LpTzDw6MpmuM4CoztSLXaQGCf8vkETSbvQkmTV65cYY8eAxgWVouPPdaJ\nJ06coL9/BEVV1mME/qvsRAruTOoR2EzgI0ZGNuDSpV/SZPKn6Bv9kHLebzSbK3LZsmV39JwkkqJS\nkrWzTKy2UincW44dO0aLxZvAIgJbaDK1ZN++QwuNeeSRdgQWOBdFlWoKn3564B1f2+FwEFBRZDP3\nUUw0/hR9EAIJNFCa1zRVIoeSCNSmWm3hjBkzOH78RI4ZM4ZGozeNxsEEvCmylnMJbCXgzpYtW9Nk\nKkd392Y0mXw4e/Y7ha5fp87D1OsHENhOjeY1+vtXYo8e/Si6uXkQqEgRCXVWuf9LFOGyxwicps3m\nS5L85Zdf6OkZqlw3X3nMZ69eg251+xJJqSKVgqRUmDNnDg2GQQUWsvM0mdwLjalTpzmBHwqMWcw2\nbboXaf6cnBwOG/YcrVZvengE8K23Zjm/279/P0VegbeyY9BT9GPIv046RdiqjaLchZFqdU02aNCc\nly9fZlxcZ4pw003K+ERlwVYT8KTBYOOBAwf466+/MjKyPg0GG0NCqjvzK06dOkWj0UdRIvkd0xox\nISGBgIai7AUJTCPgS42mq6K0XiDgoEbzMhs3fqzAc3qUwhch5tJqn+Po0WNL4VeSSG6PVAqSUuH9\n99+nydStwEJ8kCqVha+/Ps3ZnnPOnPk0m6MpmuVsodkcxqVLvyzS/BMmvEK9vgqB/xD4imZzBJcs\n+YJ2u53u7sGKyagCRae0Z5U381SK3IRPKZLSNlCtdmd4eAwHDx7FtLQ0tmrVgXp9f4o8hhSn/BrN\ncJYvX4mPPtqev/76K0kyMrI+1epxylv+Clos1/sn6/XuivIRTm+rtQY3bNhAnc5M4LRyfBUBE3W6\nAKrVJmo0JppM/gwPj+bp06dJktu3b+ewYcOo13tQo3mOen0/ensHO7+XSO42UilISoXk5GT6+YVS\noxlBkVkcTmAkzebafOMN0VnP4XBwypTpLF8+gsHB1fnee/8t0txZWVk0GHwUW/sIik5oQ9ipU2++\n9dZMiuieLAo/wPOKGSk/Q9mk7BxqEuhNna433377befcOp1JUR5tCQwjkEFhx/d3KgNSRBSJea77\nGszmJxkfH0+S7N69L43G+gQGUadrypCQqnzssXYMDIygTleVwDxFnvzdyGEajd7ctGmTU2kuXvwp\nzeYA6vUjaTLVYVBQJQ4YMIALFy7k2bNnS+unuiV79uxh+/Y92KxZOy5YsEh2fHtAkUpBcsfk5OTw\n999/59q1axkVVYcig3ipsvhtZaVKMXc0/+LFi6lWNy6wIG8lUI5Dh47iU0/1p3Di5u9Qdii7BRuB\nDwi8RKA2gW8JTCFg48yZM51ze3qWJ/Cr8vb/OAENTSYvzp07j7VrP0yt1sjAwCocMGCAohTyQ0b3\nEwh2OoA3bNhAg8GDGk1LZZcSSOFYbkWtNoY+PqHU6QIKyEm6uzd3VgV2OBy0WLwI7GF+JVW12p9G\nYxTd3NrRZvMtpKRKm0OHDik+odkEvqbFUp1vvz37rl1Pcv8ilYLkjrh69SpjYhrRaq1Mq7Uyvb2D\nqVZPKLD4fcfq1R8q8fxr167l448/To1maIE5UwnoeOrUKU6f/hZNpjgCdkVpjCXgR5XKqIz1YsHY\nf+Ap2mzebNu2Gy9cuMAlS76gyeRPrfYFms1xjIysx4yMDFauHEO1ehKB55RF3oOihHeIomQ8CYTR\n2zuE+/fvZ2holKJ4qMjShMBCAlUIrKBWa6PJ5EHR35kE/qLJ5MOjR4+SJHNzc6lWaynKSlBRdA/z\nup9iCatVq1daP9sNvPzyK1SrXyjwnHYyIKDKXbue5P6lJGunrJIqcfLii6/iwIFwZGdvBADY7d1A\nvgNRLsIXwBSEh5esCdL06TPx+uvvIjOzKRyOLwD0AlAdGs0LaNiwBYKCgjB69EisXr0B27ZVRlaW\nGg7HBahUOVCpNCBvVpZBi7S0QVi9OhOPPNIWf/yxFWFhlbBu3TqUK9cBXbt2xcSJk3D06EmQKyA6\nuH0LkcHcA0ArAIeUz1YkJ7+Pbt36IynpFIAmyjV0ABoAuADRr3kycnOzAJigUjWH1RqB7Oyj6NKl\nI1JSUgAAGo0G9es3w6+/jkdu7qsANkPkN2iUOZvg7NkXkJubix9//BEpKSlo3LgxQkJCSvRsJZJS\n5S4op1KnjIhZbA4dOsQNGzYwOTnZ1aKQJBs1iiPwtmLSySKwQnk7H00RAfQdDQZ3nj9/vljzZmVl\nKfb+U8rb8xQCVmq1JrZo0b7Q/R8/fpyRkbFUq+tROIuzqNPVpVbrTr0+kkAkgW8oGuj4U1QxddBs\nDuCJEycKXbdr1z40mVpTREm9pJiBLitvz6OoUmkIjC/wRn2BBoM7GzZsRY3mJWW3cpKiKdDHFIX3\nfChCUR1Uq8fSxyeUJlMQrdbuNJsDOXWq8HEkJiYyJCSSKpWeRqMHDYYw5bw8arXPsXnzdmzYsCWt\n1jq0WjvTavXhL7/8cuc/IsW/K6vVx2k+MpurSfPRA0pJ1s4ysdr+G5XC6NHjlfIQDWmz+ZbaglBS\n0tPT6etbiSLqpyaBaGq1najXVyhkO7daK/HgwYPFmjs5OZl6vRtF9FAsRZJZKENDI5mWluYc9/33\n39Nk8qKILvqmkNkqJqYJExISOHLkGEZHN6VG46f4AoQJSq9344ULF0iSqampPHjwoLLov0gRsZRL\nUfIigcBiCkdxOQKVFRMWCcylWu3JM2fOMDKyHlUqM0XYq1X5U4si9PS6WUaEvyYpn0/TYPDgyZMn\nGRPTiEZjVwIf0mxuwho16lGnM1GvtzEmphFnzZpFs7l5AZPSN6xUqWap/Z4FHc0LF34kHc0PKFIp\nlBHWr19PiyWswFvr/9HHp4JLZRo37iUaDN0o6g45CDxLT88Qenj4U9QgukS1eiYDAyvfkN18OxwO\nB6tVq0PR5nKwMn8e9fqefO45UW56+fLl1GotBPoru5IRzHdG63TPsXfvwc75cnNz2aRJa8X/MJMG\nQ3U2bvwIN27cyFmz5lGvt1Kr9VLe8P9DkVHcncJp7kfhuK5DoBGBCAqfQnVlvJ579+6lw+Hghg0b\nFCU1hsBL1OutNJke4nVfwetUqwvXXHJzi+Ls2bOp0ZSn8EUMI3Caer2NP/zwA5s0iWNsbDPGxtb7\n2y7lHK1W71L9TSUSqRTKCAsWLKDZ/EyBBcFBlUrD7Oxsl8kUF9eVBZOsgJ9Yo0YT/v7774yIqE2j\n0Y2xsU2cztRbYbfbOXXqDLZv35MvvTSJ6enpJMkxY8YpC/KPBa6xlC1adCRJBgdXp8hgHqfsKCIJ\nNCAQQ3//MHbp0otdu/bl2rVrSZLZ2dmcPXsO69VrSr3enxZLbxqN5anV+lL0d3AjkKxcJ4tAIHW6\n/F7M0wksV5RBPV5voNODanUwv/32W+f97N+/n2PGjGWjRo+wWbPHGRRUhUZjCE2mh6nR2KjR2Chy\nFkhgBd3d/ennV0lRaj9T9IRoSp3OXek98QGB1cr9eRI4wXyT0qOPtrsrv63kwUUqhTLC1q1baTYH\nU9jDRTRKUJBro0MmTXqdJtMTyltwHvX6fuzb99lizeFwONi2bVeaza0JfEKjsRtr127KnJwc+vmF\nEeilLPy5BOzUatvyxRcnkSQ9PQOVxdWbopbRRALerFYtUnlbn0bgXZpM/s6qr0lJSTQY3JnvVwBe\npmiAc5Qisuh6HoJOV4fe3uUJdFGOvUbg6QJjPiTgT6PRu5B5zOFw8PHHO9NobK0oD2+qVLUoEuQG\nUdRoMlGrNbNcuSB++OGHtFqjC8ybR8CHfn5B1GieK6AQ91PkaBgJ6FirVlMmJSWV2u8pkZBSKZQp\n3nhjOg0Gd9psESxXLoi7d+92qTxZWVls0aIdTSZ/ms3BrFWrCVNSUoo1hygR4U0g07kgWq3VuXXr\nVgYEVKEoV/0wRVc0b4aGRjrbQvbqNYhG45MUhe+CKEphv0aNpg6FqSd/kf2awcE1SJJ79+6lzVaV\novREC+XNO4jCLFedwOsUzt2FdHf3Z+3aDSnyF0iRKT23wCK9m4A7o6IasGrV+hw0aCTT09N54sQJ\npezFagrzV36m9GrlWiQwif36DabD4eD27dtptVZTlAEJZFOj8eLIkSOp0z1b4Hq7KMJj7dRoTExN\nTS3131QikUqhjJGUlMR9+/aVqH3l3cDhcPDYsWM8fPiwMzO3OPz11180mQIKLIikm1ttbtq0ifPn\nv0ezOZzAR1SpRtFqLce//vqLKSkpHD16HFu27Mjo6PpK0pWVwFXmJ36JQnR/ML+8hEpVjnv27GFG\nRga9vAIpMph7UrTTHEoRIRRDkY/gRSCc1avX5sGDB6lW2wiMorD1V1KURhb1+q40GLyp0Uwh8AuN\nxu5s3vwJHj16lGZzoDK+YM9oB0UF1iwCb7NfP7GrysnJYa1aTWg09iCwhCZTW7Zs2Z4nTpygu7u/\nki+xmCJLfC7V6rcZFlZ6DmaJpCBSKZQxzp8/z7179943SuFOycvLY61aTWgwDCDwJVWqJnRz8+GW\nLVtIkvHxCWzXrgf79BnMw4cPMysri9Wq1aFe34/AFzSbW/PRRx+n1RpeYPElhTP4dQIrCVSmwVCX\nn38umvn89ttv1Ol8lTf3/PGtCHTi9SqmZ2gw2Hjq1CkeO3aMbdt2Yv36zdmyZVtqtUZqNHrWqtWE\nVuujBeawU6ezMjk5mbGxjSl8Ed4UIaqkCFH1J5BAs9mXmzZtcj6H9PR0jhv3EuPiunLy5DecvqKj\nR4/ymWeGsE6dR2gw2KhW61ilSiyPHTt2738syQOBVApliNdfn6aYj6rSyyuQu3btcqk8DoeDs2bN\nZUREPUZHN+GKFStKNE9KSgo7dHiKarWVwtn6Cs1mb65bt+6GsT///DNtttoFTEOrqVYbWK5cCDWa\n1wkcp0o1lyLssw5Fuew5NJmCuXXrVuc8Tz89kDrdKGUeB7XaJtRoaii7jEQCQVSpAmkweLF//2GF\nwjNzc3OZkZHBiRMn0mAoKEsqtVoT09PTeeXKFWX3kq8cggiYqVK50ccnjMuXLy/R8843nUkkdwup\nFMoIW7ZsodkcUuBN9guXO5pnzZpLs7kGgfUEltNs9r/pQl4Uhg4dRbX6pQJv3V+wZs3G7Nr1GbZo\n0ZGLFn1Mh8PBNWvW0M2toTJmKkVIaH8ajVXo4xNOT89A1q37CBctWkSLxZvu7g/TZPLn2LH/KXS9\nCxcuMCysJm22WFqtUYyKqs927brRbK5AjSaAItFNLPQWSywTEhKc5+bm5vKRR9rQbK6nmKmeIvAR\nzeZGfOaZISSFSUiUzc4mcEDZMcwmsJ1GY2e2bdu1xM9dIrmbSKVQRrgfQ1KrVq2vKIR8mWazT58h\nJZpLFLZ7r8BcG6hWe1GlmqqYW6pz2rS3mZ6ezuDgCKpUI1m4aU06zeaQQs738+fPc+3atbfs7paV\nlcVNmzZx8+bNtNvtdDgc3L17Ny0WnwImHxKYzBdffMl53ooVK2i11lH8ESkEBlGt9uCcOfOcfhWH\nw0Gj0Y3AIYpEuM4F5sukRqN36W8nkdyKkqyd6ntZUkMiqFKlClSqjQAuK0e+h49PEPR6vctkMhgM\nAFKcn1WqFJjNhhLN1bNnB5jNUwFsBLAHOt1IkNVBTgDQDRkZ8Zg5cz4sFgtmz54ClepTiLpEAcoM\nFuh0Ybh48aJzTj8/PzRv3hxVq1a94Xo5OTlITU1Fw4YN0bBhQ+h0OqhUKsTExKBq1epQqVYoIzNh\nsaxGRMT1HszHjx9HZmZFAFoA7gDeA3ANQ4YMglqtVp6FCrNmvQWzuQVE7aRzAKjMcBUqlQrLly/H\nW2+9hR9//LFEz0wiuW+4C8qp1CkjYhaL556bSJPJl+7uDejm5ufyMhcrVqyg2exPYDZVqkm0Wu+s\n5/LixZ+yQoUaLF8+gs2ataJaPbLA2/Wf9PauQJKcMOFFivyCMALzlWieZbTZfHnx4sXbXufLL7+i\nyeROg8GLfn4V+fvvvxf6/uDBg/TxqUA3tzo0m4PZoUPPQjuAiIhaFIlumxQfxDhWrhx702utW7eO\nL774Er29QxTn+Pu0WGIYERFLiyWGOt0YWizhfPHFySV+bhJJaVKStbNMrLb/RqVAkkeOHOGmTZt4\n+fLlUp87OzubvXsPosFgpdXqzWnT3r7tOevWrWOfPkM4dOioYtc3+juHDh3iiBHPceDA4UxISFAK\ntM0lsIJmcwwnTXqDJPnWW2/RYOihmGbqENBSo/FwRizdirNnzzI6uqHihN6tKJvP6OsbekM4bVpa\nGjdv3uwsX5FPUlKSEqK6XHEe6wj4Mzq67j9e+8qVK5ww4T986qn+fPXV12gyBRO4psiQRL3exkuX\nLpXwyQll9dlnn7N378F85ZVXi50vIpHkU+pKoW/fvvT19WVUVJTzWHJyMlu0aMHKlSuzZcuWnwmD\nLgAAFpRJREFUvHLlivO7N998k+Hh4YyIiODq1audx3fu3MmoqCiGh4dz5MiRzuNZWVns2rUrw8PD\nWb9+fR4/frzUbuxBZ9SocTSZHiNwgcAhms1VmJCw9KZj7XY7R44cy8DAqqxatV6h364kHDhwgFar\nD1Wq/xCYQZPJl++99x7btOnGhx56jPPmvetcnFNSUhgaWp1mc0fqdKNpMnk7m9X8EzExjahWd6TI\nUbgevmo0evPcuXNFkjMlJUXp1XBFOT+XQA16e4cWuYDc2rVr6e7etJAMFksFHjlypEjn34zx41+m\nxVKTwHwaDL0ZHh7Na9eulXg+yYNLqSuFjRs3cteuXYWUwtixYzl9+nSS5LRp0zh+/HiS5J9//sno\n6Gja7XYmJiYyLCzM+R+rbt263L59O0kyLi6OK1euJEm+++67HDp0KEkyISGB3bp1K7Ube9AJC6vF\n601gSOBdPv30wJuOHTJkNE2mlhQJYitoNvvwt99+u2FcURfKgQOHU6WaXODaX7F27UduOT41NZXv\nv/8+Z8yYwT179tx2/oyMDGo0egLbKcpZ5GcZ76XB4FYsp2/jxi0JRFG02OxAwJcaTUCR24teunSJ\nbm5+FNVXU6lWzy5R0cB8cnNzqdUaCZx3BiFYrY/yq6++KtF8kgebu2I+SkxMLKQUIiIinPX0z507\nx4iICJJilzBt2jTnuNatW3Pr1q08e/Ysq1at6jweHx/PwYMHO8ds27aNpAj78/a+eZVIqRSKT4MG\nLQl85FyYdbphfOGFCTcd6+kZRFEvSIxVqydw0qTrdvH4+AS6uflSrdayUaNWzhLVt6JHjwEE3img\nFH6+o45tfycvL496vVUxOb1AUe67BQEbJ0x4sdhz+fgEU+QfmCiynZezbt0WhcadPXuW69at4/Hj\nx5mens5XXnmNXbo8wzlz5nHbtm2sWDGKOp2JNWo8dEe7hOzsbEXhZTmfn9XamZ999lmJ55Q8uJRk\n7Sx257WkpCT4+fkBEBEhSUlJAICzZ8+iQYMGznFBQUE4c+YMdDodgoKCnMcDAwNx5swZAMCZM2cQ\nHBwMANBqtXB3d8fly5fh5eV1w3UnT57s/HuzZs3QrFmz4or+QPHOO2+iWbM45OZuhVp9BR4euzB2\n7NabjjWZzLhyJQlAGABAqz0PqzUSALBr1y707z8KGRkrAURix47x6Ny5DzZs+OGW1+7btxuWLeuD\njIxKANxhNo/CgAEDSu3e1Go1pk59A88/3wDA0xBRQ8kwmXzwxBNtij1X8+YtsXRpBMhxytFFsNks\nzjFLl36Ffv2GQqerhqys/fDx8cWlSzWRldUS33//OZ54Yhf++mtvqdybXq9Hq1ZPYN26PsjKGgeV\nagfU6k1o3nxeqcwv+Xezfv16rF+//s4muZ3W+PtOwcPDo9D3np6eJMnhw4c7Sw+QZP/+/fn1119z\n586dbNHi+lvXxo0b2bZtW5JkVFQUz5w54/wuLCzspl3IiiCm5CYcO3aM8+bN4wcffPCPzuzPP1+i\n1PeZRq12CP38Qp27gdmzZ9NgGFbgrT+dWq3httf+3//+x8jIhxgeXptvvz37rjR5adeuM/X6YAI9\naTS2ZuPGrZmbm1vsefbt26f4QMYzPwN78+bNJIVpS/Rj/p355b5FzaQ85/Mo2OCnNEhPT2ffvs8y\nNLQmGzZszb1795ba3JIHi5KsncXeKfj5+eH8+fPw9/fHuXPn4OvrC0DsAE6dOuUcd/r0aQQFBSEw\nMBCnT5++4Xj+OSdPnkT58uWRm5uL1NTUm+4SHnQuXryI11+fjpMnz+Oxx5pi8OCBUKlUtz2vUqVK\nGDFixG3H9ezZAwEB/li27Ht4evpj2LDt8PHxAQD4+PhAq12G7GwHADWAPXB397ntnB07dkTHjh1v\nO+5OWLbsS3z66afYvn03IiLqYejQIdBoNEV+XufOncPevXsRFBSE3377BYsWfYLc3Cz07r0W0dHR\nAMRuVqv1gejPDABeANwgngUAGKFWG2C320vtviwWCz766N1Sm08iKRa30xp/3ymMHTvW6TuYOnXq\nDY7m7Oxs/vXXX6xUqZLz7bBevXrctm0bHQ7HDY7mIUNE1mx8fLx0NN+E1NRUBgVVoU43nMAnNJvr\ncPTocffs+tnZ2axf/1FaLI1pNA6iyeTDb775plSvsW/fPr788it87bXXbxmBRorQ0lmzZnH8+InO\nZjt/JzU1lQEBlajV9iSw8JbPa+XKlUrpjEdoMgXw+edv7otIS0ujxVKOwEZlZ7CdgJUazWQCW2kw\nDGCdOg+XaCdkt9t56NAh2UdBctcoydr5j2d0796dAQEB1Ol0DAoK4kcffcTk5GQ2b978piGpU6ZM\nYVhYGCMiIgqFFeaHpIaFhXHEiBHO41lZWezSpYszJDUxMbHUbuzfwhdffEGr9fEC5psL1GoNJTKT\nlBS73c6lS5dy/vz5/OOPP2i324scXbNz5042bhzHatUa8MUXJzMnJ6fQ96LhkDfV6vHUaEbQzc3v\npo7aa9eusXLlGBqNXQi8SrM5hO+//6Hz++PHj7NNm660WAIoGtdUUv5sueF55eXl0WbzVhLWSCCZ\nFksFZ9DD31m9ejWtVm/abBE0mTw4d+47fPzxLgwPr82nnx5Y6P9AUUlMTGRISFVaLKE0GNw5atQ4\n2UdZUuqUulK4X3iQlcLixYtpsRSstZNGjUZ/w+J6L8jJyWHPngOo0eip0ejZp8/gf5Tj6NGjSn+E\nBQQ20mx+mEOHji40pmnTNgQWFYh8msS+fYfeMNcnn3xCiyWO16uY7nP2NE5NTaWfX0Wq1X0oGtck\nKWPmE6h9w/O6fPky9XpbodwCm60rlyxZcst7SUtL4759+0otkaxu3UeoVk8roJQiuWzZslKZWyLJ\npyRrp6x9dJ/TunVr6HSboVLNBLABJlN3dOzYHVptsd1Bd8wbb0zHt98mIi/vEvLyLuKrrw5hxoxZ\ntxy/bNky2O1dAQwA0AQZGZ/j008/KzQmNTUNQIjzs8MRgpSU9Bvmunr1KvLyQgDk+wZCkJWVDpLY\nvHkzMjMrwOGoCeAJAL7KmL4A9tzwvDw8PODh4QXgK+XIX8jL24gaNWrc8l6sVisiIyPh7u5+yzHF\n4c8/98Dh6Kt88kJGRnvs2bOnVOaWSO4EqRTuc/z8/LB9+3q0bLkFkZEvYsiQGvjssw/v2fV//vln\nzJkzB99//z3WrNmMjIxRAGwA3JCRMRI//vjLLc/V6XTQaNIKHEmDTle46F+PHu1hNk8EsA/ADpjN\nb+Kpp9rdMFerVq2gVv8PwAoAf8FgGIK4uPZQqVTQ6XQg0yFCatcBuKac9T1sNt8bnpdKpcIPP/wP\nXl5jYLFUhMEQg7fffvUflUJpkZOTg3Hj/oOcHDWA/1OOZsFs/gnh4eF3/foSyW0p/Q1L6VNGxPzX\n8dJLr9JiCaPBMJwWSyQrVapJjeY/TpOLVjuBvXoNKnSOw+HgBx8sZPPmHdiuXTd6epanVvsCgQU0\nmyM4Y8bMQuPz8vL48suv0c8vjOXLR/xjJvHPP//MypVrsVy5EHbr1pdpaWkkhW8qMrIe9fqeBB4l\n4E2drjbd3f25Y8eOW86XlZXFI0eO3NIk5HA4+P77H/LRR59kt259efjwYe7atYsdOjzNFi06Mj4+\n4abn/RODB4+i2fwogc8p+jLUoMkUwk6dni5RC1SJ5J8oydpZJlZbqRRuzeHDh1mzZkMaDFZWrhx7\nQ5XQkpKUlESDwb2Aff4qjUY/enkF0WptS6v1cfr7VyyUZ0KSU6ZMp9kcRWApVaqptFrLsVevAezc\nuQ9feuk/DAqKoMFgY6NGrYtco6goXL16lRMnvsyOHXtx3LgJXLt27U1zXorD5MlTlBpEX1KtnkKL\npRzN5nIE5hBYQrO5Ej/8cGGx5rTZfAicUJ5pCjWajhw1apR0MkvuClIpPGBkZ2czICBMaVl5hcBi\nenqWLxVn6P79+2/olezu3ojfffcd4+PjmZCQcNOom3LlQgjsK1Be41lOmzaNiYmJitP5ewKXqdWO\nZ3R0wzuW827i4VGewMECTvAYijLf+c9kPcPCbl5m+1Z4eQUXSIQjDYZenDNnzl26A8mDTknWznvv\nrZSUGseOHUNamgrkSOVIbzgc7+OPP/5AkyZN7mjuSpUqwWzOxbVrH4DsA+A7AMfQqFEjeHp63vI8\n8e9QU+CzxukMVqkeBfA4ACA39038+acF6enpsFqtJZaTJNavX48LFy6gXr16qFixYonnutnche9F\njcJuOI0ypuhMnjwBEyZ0REbGc9Bqj8DNbQN69JhZKvJKJKWBdDSXYTw9PZGTcwlAsnLkGnJyTpVK\nVrjBYMD69T+gcuUPoVbbEBIyCWvWrPhHhQAAw4cPgtncA8AKqFRzYDQmoGvXrsp5RwDkKiNPQKVS\nwWQylVhGh8OBDh16ol274Rg48CtERdXDqlWrSjzf3xk2bBDM5u4AvoNKNQsmUyJMpvkA/gvgG5jN\n/fHcc4OKNeeIEc9iyZKZ6NXrD4wYYcCePduc2eMSyX1B6W5W7g5lREyXMGbMBFosVanVPk+LJYZP\nPz2w1O3TxZnP4XBwzpz5bNgwjk880d1Ztyc3N5cPPxxHi6UJtdrnaDYHc968d+9ILtFfObZARdH1\n9PIKvKM5C+JwODhr1jw2bBjHdu2e4p9//slt27bxscc6s3HjNvzoo0+kL0ByX1OStVOlnHhfo1Kp\nir1Nf5D44Ycf8Mcff6By5cro2LFjkeoiuYLc3FzEx8fj7NmzeOihh9C0adM7mu+9997DCy/sQWbm\nB8qRHKhURuTkiDpEM2bMwrff/ghfXy/MmDEJ1atXv8M7kEjKFiVZO6VSkJRZdu7ciYcfboeMjA0A\nwqFWz0C1at9i375tGDVqHBYu3IyMjJehUh2AzTYN+/btdJZql0geBKRSkJQJUlNT8fHHH+PKlRTE\nxT1WqA9Hccd++OEijBgxCoAawcEVsWbNMlSsWBEWixcyMv4AICryGgwDMH16DYwaNeou3plEcn8h\nlYLkvic1NRXR0Q8hKSka2dmVYDJ9hE8+eQddunQu8djc3FykpaXBw8PDaTqz2XyQnr4DgIhGMhr7\n4K236mL48OF3/R4lkvuFkqydMvpIck9ZvHgxkpJqICsrHuQUZGR8idGjX7qjsVqtFp6enoV8KaNH\nj4DZ3AFAPNTqSTCZ1qJz5xsVj0QiKYzMU5DcU1JTr8JuL5hLUBHp6al3PPbvvPbaywgMDMCyZd/A\n378cXn11C/z9/UsuuETygCDNR5J7yq+//oqHH26LzMx4AGEwGp9Dx46eWLJk4Q1jd+7ciaZN2xRp\nrEQiuRFpPpLc99StWxfx8R8iOHg4PDwaonPncliw4OZN6evUqVPksRKJpHSQOwWJRCL5lyJ3ChKJ\nRCK5I6RSkEgkEokTqRQkEolE4kQqBYlEIpE4kUpBIpFIJE6kUpBIJBKJE6kUJBKJROJEKgWJRCKR\nOJFKQSKRSCROpFKQSCQSiROpFO4B69evd7UId0RZlr8syw5I+V1NWZe/JNwXSmHVqlWoWrUqKleu\njOnTp7tanFKnrP/DKsvyl2XZASm/qynr8pcElyuFvLw8DB8+HKtWrcL+/fsRHx+PAwcOuFosiUQi\neSBxuVLYsWMHwsPDERoaCp1Oh+7du2P58uWuFksikUgeSFxeOvvrr7/G6tWrsWDBAgDA559/ju3b\nt+Odd95xjinYZlEikUgkRae4S7zL23EWZcGXvRQkEonk3uBy81FgYCBOnTrl/Hzq1CkEBQW5UCKJ\nRCJ5cHG5UqhTpw6OHDmC48ePw263Y+nSpWjXrp2rxZJIJJIHEpebj7RaLebPn4/WrVsjLy8P/fv3\nR7Vq1VwtlkQikTyQuHynAABxcXE4dOgQjh49iokTJzqPZ2VloX79+oiJiUH16tULfVeWyMvLQ2xs\nLJ544glXi1JsQkNDUbNmTcTGxqJevXquFqdYpKSkoHPnzqhWrRqqV6+Obdu2uVqkInPo0CHExsY6\n/7i7u2PevHmuFqtYTJ06FZGRkahRowZ69OiB7OxsV4tULObOnYsaNWogKioKc+fOdbU4t6Vfv37w\n8/NDjRo1nMcuX76Mli1bokqVKmjVqhVSUlJuPxHvc65du0aSzMnJYf369blp0yYXS1R8Zs6cyR49\nevCJJ55wtSjFJjQ0lMnJya4Wo0T07t2bixYtIin+/aSkpLhYopKRl5dHf39/njx50tWiFJnExERW\nrFiRWVlZJMmuXbvyk08+cbFURWfv3r2MiopiZmYmc3Nz2aJFCx49etTVYv0jGzdu5K5duxgVFeU8\nNnbsWE6fPp0kOW3aNI4fP/6289wXO4V/wmw2AwDsdjvy8vLg5eXlYomKx+nTp/HDDz9gwIABZTaK\nqizKnZqaik2bNqFfv34AhJnS3d3dxVKVjLVr1yIsLAzBwcGuFqXIuLm5QafTISMjA7m5ucjIyEBg\nYKCrxSoyBw8eRP369WE0GqHRaPDwww/jm2++cbVY/0iTJk3g6elZ6NiKFSvQp08fAECfPn2wbNmy\n285z3ysFh8OBmJgY+Pn54ZFHHkH16tVdLVKxGDNmDN566y2o1ff9o74pKpUKLVq0QJ06dZy5JGWB\nxMRE+Pj4oG/fvqhVqxYGDhyIjIwMV4tVIhISEtCjRw9Xi1EsvLy88PzzzyMkJATly5eHh4cHWrRo\n4WqxikxUVBQ2bdqEy5cvIyMjA99//z1Onz7tarGKTVJSEvz8/AAAfn5+SEpKuu059/1KpVar8fvv\nv+P06dPYuHFjmapF8n//93/w9fVFbGxsmXzbBoDNmzdj9+7dWLlyJd59911s2rTJ1SIVidzcXOza\ntQvPPvssdu3aBYvFgmnTprlarGJjt9vx3XffoUuXLq4WpVgcO3YMc+bMwfHjx3H27Fmkp6djyZIl\nrharyFStWhXjx49Hq1atEBcXh9jY2DL7YpePSqUqUl5YmblLd3d3tGnTBjt37nS1KEVmy5YtWLFi\nBSpWrIinnnoKP//8M3r37u1qsYpFQEAAAMDHxwcdOnTAjh07XCxR0QgKCkJQUBDq1q0LAOjcuTN2\n7drlYqmKz8qVK1G7dm34+Pi4WpRisXPnTjRs2BDlypWDVqtFx44dsWXLFleLVSz69euHnTt3YsOG\nDfDw8EBERISrRSo2fn5+OH/+PADg3Llz8PX1ve0597VSuHTpktNbnpmZiTVr1iA2NtbFUhWdN998\nE6dOnUJiYiISEhLw6KOP4tNPP3W1WEUmIyMDaWlpAIBr167hxx9/LBTZcD/j7++P4OBgHD58GICw\ny0dGRrpYquITHx+Pp556ytViFJuqVati27ZtyMzMBEmsXbu2zJl+L1y4AAA4efIkvv322zJnwgOA\ndu3aYfHixQCAxYsX48knn7ztOS7PU/gnzp07hz59+sDhcMDhcKBXr15o3ry5q8UqMWWthlNSUhI6\ndOgAQJhjevbsiVatWrlYqqLzzjvvoGfPnrDb7QgLC8PHH3/sapGKxbVr17B27doy5cvJJzo6Gr17\n90adOnWgVqtRq1YtDBo0yNViFYvOnTsjOTkZOp0O7733Htzc3Fwt0j/y1FNPYcOGDbh06RKCg4Px\n2muvYcKECejatSsWLVqE0NBQfPnll7edx+UF8SQSiURy/3Bfm48kEolEcm+RSkEikUgkTqRSkEgk\nEokTqRQkEolE4kQqBYlEIpE4kUpBIpFIJE7+H39qTtQsTbxeAAAAAElFTkSuQmCC\n"
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Learn (rooms only)"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.linear_model import LinearRegression\n",
      "clf = LinearRegression()\n",
      "X = data[:,indexof('RM')].reshape(data.shape[0], 1) # Make it a list fo 1-length vectors\n",
      "clf.fit(X, prices)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 5,
       "text": [
        "LinearRegression(copy_X=True, fit_intercept=True, normalize=False)"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### What have we learned?"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "clf.coef_"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 6,
       "text": [
        "array([ 9102.10898118])"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "clf.score(X, prices)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 7,
       "text": [
        "0.4835254559913339"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Plot the results"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "xs = X.ravel()\n",
      "fig = plt.figure()\n",
      "ax = fig.add_subplot(111)\n",
      "ax.scatter(xs, prices)\n",
      "ax.plot(xs, clf.predict(X), color='red')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 8,
       "text": [
        "[<matplotlib.lines.Line2D at 0x107f2cfd0>]"
       ]
      },
      {
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAD9CAYAAABA8iukAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n"
      }
     ],
     "prompt_number": 8
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Now learn on all the features"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "clf = LinearRegression()\n",
      "clf.fit(data, prices)\n",
      "clf.coef_"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 9,
       "text": [
        "array([ -1.07170557e+02,   4.63952195e+01,   2.08602395e+01,\n",
        "         2.68856140e+03,  -1.77957587e+04,   3.80475246e+03,\n",
        "         7.51061703e-01,  -1.47575880e+03,   3.05655038e+02,\n",
        "        -1.23293463e+01,  -9.53463555e+02,   9.39251272e+00,\n",
        "        -5.25466633e+02])"
       ]
      }
     ],
     "prompt_number": 9
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "clf.score(data, prices)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 11,
       "text": [
        "0.74060774286494269"
       ]
      }
     ],
     "prompt_number": 11
    }
   ],
   "metadata": {}
  }
 ]
}