{
 "metadata": {
  "name": "",
  "signature": "sha256:1961b76026afdba9792fc68df011f7f3de9cffe388949aae54dc5d41a8ba4438"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "heading",
     "level": 1,
     "metadata": {},
     "source": [
      "5. Learning, badly"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The time has arrived, the time to learn. Will we succeed? Let's see it"
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "5.1. Preparing the notebook"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%matplotlib inline\n",
      "%config InlineBackend.figure_format='retina'"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import matplotlib.pyplot as plt\n",
      "import numpy as np\n",
      "import pandas as pd\n",
      "import seaborn as sns\n",
      "\n",
      "from matplotlib import cm as cmap\n",
      "\n",
      "from sklearn.cross_validation import StratifiedShuffleSplit\n",
      "from sklearn.ensemble import RandomForestClassifier\n",
      "from sklearn.metrics import confusion_matrix\n",
      "from sklearn.preprocessing import LabelEncoder\n",
      "\n",
      "sns.set(font='sans')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "5.2. Learning with `random forest`"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "When I needed to select an algorithm for training I didn't know which of them I studied about was to work correctly. So, I asked to my thesis's supervisor, [Fernando Sancho Ph.D.](http://www.cs.us.es/~fsancho/), and his recommendation, according to his experience with other related projects, was to use `random forests`.\n",
      "\n",
      "In the `feature_columns` variable showed in the next piece of code, we can see which attributes are going to be used for predicting."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "labelize_columns = ['medallion', 'hack_license', 'vendor_id']\n",
      "\n",
      "interize_columns = ['pickup_month', 'pickup_weekday', 'pickup_non_working_today', 'pickup_non_working_tomorrow']\n",
      "\n",
      "feature_columns = ['medallion', 'hack_license', 'vendor_id', 'pickup_month', 'pickup_weekday', 'pickup_day',\n",
      "                   'pickup_time_in_mins', 'pickup_non_working_today', 'pickup_non_working_tomorrow', 'fare_amount',\n",
      "                   'surcharge', 'tolls_amount', 'passenger_count', 'trip_time_in_secs', 'trip_distance', 'pickup_longitude',\n",
      "                   'pickup_latitude', 'dropoff_longitude', 'dropoff_latitude']\n",
      "\n",
      "class_column = 'tip_label'"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = pd.read_csv('../data/dataset/dataset.csv')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 4
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Before starting the training, we need to transform the no numeric attributes to numeric ones, so that they can be used with `scitkit-learn`."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "for column in labelize_columns:\n",
      "    real_column = data[column].values\n",
      "    \n",
      "    le = LabelEncoder()\n",
      "    le.fit(real_column)\n",
      "    labelized_column = le.transform(real_column)\n",
      "    \n",
      "    data[column] = labelized_column\n",
      "    \n",
      "    le = None\n",
      "    real_column = None\n",
      "    labelized_column = None"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "for column in interize_columns:\n",
      "    data[column] = data[column].astype(int)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 6
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Let's start the training! We are going to use 10-fold stratified **cross-validation ** for training a random forest model with 256 trees."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data_features = data[feature_columns].values\n",
      "data_classes = data[class_column].values"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 7
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "cross_validation = StratifiedShuffleSplit(data_classes, n_iter=10, test_size=0.1, random_state=0)\n",
      "\n",
      "scores = []\n",
      "confusion_matrices = []\n",
      "\n",
      "for train_index, test_index in cross_validation:\n",
      "    data_features_train, data_classes_train = data_features[train_index], data_classes[train_index]\n",
      "    data_features_test, data_classes_test = data_features[test_index], data_classes[test_index]\n",
      "    \n",
      "    '''\n",
      "    You need at least 16GB RAM for predicting 6 classes with 256 trees.\n",
      "    Of course, you can use a lower number, but gradually you'll notice worse performance.\n",
      "    '''\n",
      "    clf = RandomForestClassifier(n_estimators=256, n_jobs=-1)\n",
      "    clf.fit(data_features_train, data_classes_train)\n",
      "    \n",
      "    # Saving the scores.\n",
      "    test_score = clf.score(data_features_test, data_classes_test)\n",
      "    scores.append(test_score)\n",
      "    \n",
      "    # Saving the confusion matrices.\n",
      "    data_classes_pred = clf.predict(data_features_test)\n",
      "    cm = confusion_matrix(data_classes_test, data_classes_pred)\n",
      "    confusion_matrices.append(cm)\n",
      "    \n",
      "    clf = None\n",
      "\n",
      "print 'Accuracy mean: ' + str(np.mean(scores))\n",
      "print 'Accuracy std: ' + str(np.std(scores))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Accuracy mean: 0.529469\n",
        "Accuracy std: 0.000902712024956\n"
       ]
      }
     ],
     "prompt_number": 8
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "A prediction with an accuracy of `52.95%`. What happened?"
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "5.3. What happened?"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "As I am not a machine learning expert, I'm not `100%` sure of what were the problems for this bad result. This is an indicator that I have yet to study more machine learning theory. A thing I'm willing to do, **spoiler**, specially after the results we will obtain in the [next notebook](6. Learning, a better way.ipynb).\n",
      "\n",
      "For trying to know the reason of the bad accuracy, let's use another tool for measuring the performance, a confusion matrix."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "classes = [' ', '[0-10)', '[10-15)', '[15-20)', '[20-25)', '[25-30)', '[30-inf)']\n",
      "\n",
      "first = True\n",
      "cm = None\n",
      "\n",
      "for cm_iter in confusion_matrices:\n",
      "    if first:\n",
      "        cm = cm_iter.copy()\n",
      "        first = False\n",
      "    else:\n",
      "        cm = cm + cm_iter\n",
      "\n",
      "fig, axes = plt.subplots()\n",
      "\n",
      "colorbar = axes.matshow(cm, cmap=cmap.Blues)\n",
      "fig.colorbar(colorbar, ticks=[0, 25000, 50000, 75000, 100000, 125000, 150000, 175000, 200000, 225000, 250000])\n",
      "\n",
      "axes.set_xlabel('Predicted class', fontsize=15)\n",
      "axes.set_ylabel('True class', fontsize=15)\n",
      "\n",
      "axes.set_xticklabels(classes)\n",
      "axes.set_yticklabels(classes)\n",
      "\n",
      "axes.tick_params(labelsize=12)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {
        "png": {
         "height": 359,
         "width": 471
        }
       },
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAA64AAALPCAYAAABi2Z05AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAWJQAAFiUBSVIk8AAAIABJREFUeJzs3XmYJEWZ+PFvjxzDjSiKqAgoL4IiuoKccoiseKArh4CK\neCyrrAfgyYqKoKgo4on3TzxWAVEUFW8OERHxAEQOXwUWFBBQ5BqGa+jfH5HFJDVV1d3V1V3ZNd/P\n89ST05kRmZEVlVMV+UZGgCRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJ\nkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkqQJjA27AFIvEXFf+7rMnDeM\nssy2iFgH+AtwQWY+bQjHX2rf++mIiDHgYuDRwOMy8/oB79966cNM10t1DOumD14zzWS99BYROwCn\nA1dl5nozsP/XAf8JBDC/Wr0u8Ejgl8DJmbnHoI8rNdkywy6ANEk3Aot6JYiIeZT/5F8BbAQ8CPgz\n8HXgY5l5Tz8Hrr68dwA2Bzarlo+pNh+QmZ+d5H72BA4ANgVWAK4CvgkclZm3t6fPzKsj4jhg/4jY\nKzNP7Kf8A9DzvY+ILYEtWPz+RLXpqMz8n8kcICKeCRxY7WdV4Brge8D7MvOGfgseEU8Ctq7Ktjmw\nMTAPODEz95kg75eAl01wiFMzc9f6iswcj4j3AF8D3k2p85kwY/USEWcC201w/GMz8/WTLWxt38sA\nOwPPAbYCNqBcD/8EzgO+mJmnTGI/mwFvrsq5BnAD8GPgA5l5eXv6WawX6FE3ETEfeC7wbOBpwPrA\nssD1wDnApzPz5xMdYCaumenWzVy+ZqZbLzN8zSwLvKoq15OBtYCHAncCVwI/BT6RmVdNsJ+mXjO9\n6mU1YD/KuT8JeDil7Aso3+8/oJz7P3odYKa+Yyrj08y/hIg4FHhP9edC4O/Vvxdl5q8i4qfAbhGx\nZWaeO+jjS001Z+5saak2DmyemWtn5tqdElRf7N8DPkP5Ylqe0qPgycAHgbMjYqU+j78acBrwAWAP\nFjdaW2WbUER8DjiR0gBeGbgH2BA4FLggIh7RJev7gHuBI6sflbNtwvce+BHwEeDFLG4ctfJOqPqC\n/gnlR+ODKV/S6wFvAC6KiCf0WXaArwCfotzMeCKL/8+byg+N2yk/Gjq9buqS5wQggVdFRHRJMx0z\nXi+VW+h+7rdMtdCVTwOnAq8F/o1yrd4BPAzYFfh2RHyj1+c9IvYDfgW8qMp3ByUK8SrK9bRjl6wz\nXS8wcd18DzgJeCXwBMpn8i5K+V8EnBERH+l1gBm8ZqZdN5W5eM1Mu14qM3HNPITy/9jLKTc+Hwrc\nCqwIbAK8EbgkIl7QbQcNvmYmqpeNgI9S/h97AuX7+FZgFeCpwDuBy6obdR3N4PWyAPgTsESjfwAO\nrJYHZ+ZKrfcnM6+p1h9RLT84A8eWGsuGq0bFeyl3yhdS7s6umJkrU35s3USJOk0qMtrFAuAs4BjK\nF+jfeydfLCIOoESCF1Hudq+cmasB21CirusD3+iUt7qDflqV5j+mUf6ZdAfwa+CTlB9WF0w2Y0Q8\nh3JXeRw4Glg9Mx9MaWReAKwJnBIRy/VZtruB84HPA/9FiS5M1dG1Hw3tr/06ZcjMceA4Sq+WN/RZ\n9unqu15qDuxx7u/os1zLUKIdhwNPzsz5mbk68Cjg2CrNHsCRnTJXUfTPU3pU/C/w8Mxcg/JD9KfA\nSsC3IuKh7XkbUi/LUBoCbwE2yswVM3NVSnTzpCrNgdX/G0uY4WtmWnVTMxevmWnVS81MXDMLKTeh\nXgg8MjOXy8yHUrqPPge4jBIZ/9+IWKs98xy/Zv4BvJ/y/f7w6jP5EEqZ9wauo0RgvxURy7dnnsnr\nJTN/k5kbZebO/Z1aZxHxMMrNiXFKvXU69i8p9b5tFUmXlgo2XDXnVV/UrbuTb8vMr1ZftmTmqZQ7\n6AD7RMQmU91/Zt6cmatk5g6Z+ebMPIHSIJpM2ZandLGC0l35mKy6LGfmryg/RMaBbSLieV1287Vq\nuf9Uyz5LHpmZW2XmGzLzK5S74ZP1vmr57cx8a2YuAMjMSyg3HW6nNNr/q8+ybZmZT83MV2fmFyjd\n/mbL16vliyNihVk8bst06mUmfQpYPzMPz8w/tFZm5nVZulF+qVr12qr7ZrsjKD+ifwPsl5n/rPJf\nDewG/BVYHTiky/GHXS9vpzSMPpyZ2VqZmVdk5l6UZ+ag3OTqZCavmenWzXQNs26mWy8zJjNvycw3\nZeYpmfn32vp7M/NHQOu7YyXKZ6DdnL1mMvMvmXloZv44a92BM/POzPwGsG+1ai06d9We6e+YmXD/\ne5yZd/RI16qXpv42kAbOhqtGwe7AcsDNwOfaN2bmdyl30sco0dLZ9EzKHd37gA+3b8zMC4CfVX++\npMs+vk3pLrxTRDx6Jgo5Ha2bBFNVdc96EqXh/qEO+70GOL76s9t7M1HZlhj8Y7Zk5l+Bcyk/CHcb\nwvEH/tzVIFRRil7Pm3+pWq5A6SZ4v4hYnRJhAjim/RyrH6Wfqf7s+AxzA+rlVxPUzVeq5brV+d5v\npq+Z6dTNIAyzbqZTL8OWmVdQvv+gPIpyv1G4Zibw22o5xpLnPqPXS0TsEBH3RcSVHbadWW17WUSs\nEBHvjog/RcTCiLghIo6PiMd12h/luWWAsWofrddhbYdp9dTae4ZuJEmNY8NVo6D1bM5ZmdktEvqT\navmMWShPXatsf8zM67qk+XFb2geoflhcRLleZ7v8M6l1vrdk5q+7pGm9N5tP4xnl6ZrO6Ou/qpbP\nHERBhmAYI8/Xn4Fs/47alhI5GmfxNd2u9ZlZKyK6Na6aXC/1839Q27ZhXzO96qZuFK+ZXvVSN+vX\nTEQ8ntKoHKdEVetG/ZrZulreC/y+bdtsXS+9bnisRhkB+F2U0ZkXUboB7wWcGxHr19LeRekVdGNt\nXf0Z6dvqO656BvyL0mDfos+yS3OKowprFGxcLS/ukeaSavn4GS5Lu6mUbc2IWCMzOw1ech7wFEpX\nqC8PsHzD1HpvLu2RpvXejFHq7nczWqLOXhIRrwQeQelWdglwCvCZzLytZ85SbwBPn8HyzaQ3R8T7\nKT+0bgEupIyEfVxm3jVDx9y+Wt5D6SlR1/rM/D0z/9Ulf/3ztDGdP19NrpfW+V/f6tJZM+xrplfd\n1I3iNdOrXupm5ZqpRtF/OOU74UhK4+lrmXl2W9KRu2aqwcHWAnahdAUep4yM3D6q8rCvFyjPi/8L\neBbwsywjNW9L6eb7KMrzu3vB/Y8PPSIiHkOJuo5n98H3Ws6r9r0dMOFo5NJcZ8RVo6A1Iu+1PdK0\nop0rz3Lkbiplq6dv1/pyfeq0S9QcU31vlhh0ZJY8lvID8VbKNApbA0dRRqN80gR5Wzcs1mta98JJ\n2pgSybmNMgDKMyjPQZ43E93WI2JlFj9nd3KHRs6En5nMXEhpMIzR/TPTyHqJiEcCr6n+/FKHJEO7\nZiZRN3Ujdc1Mol7qZvSaiYgvVN1J76UMpHU8JVL36szsNBXRyFwzEfGz6tzvBq6mPBp0OfCizHxX\nhyxN+I5ZDtg5M39aG3vjbOCgavuu1awIdVOJ2o/ibwOpKxuuGgWthujCHmnqAxys3DXV4A2qbK0o\n7MOmXaLmmMp7s8TzS7Pgd5Qfq+tUI1k+lDItxWsoz5OtA/wwItbosY96hGMu1d0ZlEFPHpFldNWH\nUH7UvZ3yI3kT4AcdfnBN12coU3TcQueBYibzmYHFn5tun5nG1UsVRfoa5RyvokRi2g3zmpmobmAE\nr5lJ1gvM3jVzM4unFWp1UV0f2KHL+zpK18w/Ked+M4vPfWPKuXd6xrMJ3zHfrJ5Bbvfdark88LgO\n2ydrFH8bSF3ZVViaG1pfTg8ZaimWIpn5iQ7rbgE+FxHnUQYreQTwJsp8vJ3Uu32vSe/ulY2RmYd3\nWHcj8IGI+APwfcqcii+ny3QNUxURh1AGT7sP2L8a8XSmNLFePkHp7ncX8OJJdKmdNZOtmxG9ZiZV\nL7N1zWTmm6lGNo6IVSgDL32AUj+bRcS/TTASbb+GXi/V6M4ARMSDKXPSvocy7/CTImKHBg5K1/7M\nMVBGhI6IGyjv5XQi2K16WXMa+5DmDBuuGgULKAMg9Bqmf8Xav28HiIitgZM7pG1NiP63AZWNqZZt\nKTGV92ac2nsTEZ3m0R2nzKHYcU7cQcrMCyLiBOBllCkVuv0IHzmZ+YOIOIvyY35Xaj/C+62XiHg1\ni59Ve1NmfrNL0sl8ZmDx52ZOXE8R8T7g1ZTuny+pnnXrZNavmSnUTU9z8ZqZQr30NBPXTLXf24AT\nI+IXlC6jQemC+r5aspG8ZqrndT8bEb+mNA6fThkV+eu1ZE34jul1A+pOSqR30D1XpJFlV2GNgtbz\nK70GMWhtu701jxvly+JhlDuV9dfDGNy1MZWywQOft6lrdQHrNSjIXDOd96Zbvc3mlACtUSrX65Gm\n3nXvxq6p5p5u5z7leomIfSnPAI4D787Mj/U47jXVsutnpppncrXqz4muJxhyvUTEoZSut61oZqeb\naS2zes1MsW4mY85cM1Osl8kY2DXTLjOvpUybBkuOPD9y10xdNaVca0Cqndo2z/XvmMlo1Utj6kSa\nSUZcNQoupswp+MQeaZYYXTAzf87M37y5GHg2kyvbjV1GFIbFX043DKpgDdAa7GOjiBjr0sWr9d6M\n88C6a8JNt8kMoFH/sTdKddfx3KdaLxGxJ3Bctb+jM/M9E2RpDUSyVo8RuOufmUs6bIeG1EtEHEzp\n6tiK5Ew0YvisXTN91M1kzIlrpo96mYyBXDM9dGukjdQ100Xr3NsHN5zr3zGTMYq/DaSu5sqFKfVy\nRrXcNiKW75Jm52p52iyUp65Vto0jotuIhf9eLXuVrdXw/W2PNHNN671ZDdi8S5rWe/PrauTLJmnN\nm7fE5PM1rXq7MjNvnuHyzKbJnHtPEbErZdCbecCnM/Otk8h2NqXb5hjd55NsfWauzczLuqQZer1E\nxAHAhyk/mA/JzGMnkW1Wrpk+62YyGn/N9FkvkzHta2YCrUhuewNmZK6ZHrqd+1z/jpmMUfxtIHVl\nw1Wj4GTKwBkPBv6zfWP1IywoXb6On92icRrly/RBlAFJHiAiNqX8mBin/FDs5mnV8qxBF3BYMvNS\nyhyHY8Bb2rdHxNqUZ5ag93sz66p627v689QeSUeu3iJiFxbP5djr3HvtY2fgJMp18aXMfO1k8mXm\nrbVjvjEiHhDFqqa6ak1b0utaH2q9RMR+wLGU6/6IzPzQZPLNxjXTb91MYr+Nv2b6rZdJ7Hda10xE\nPGiC7RsA/1H9+YAboHP9mpnEuT8d2LL6s/3c5+x3zBS0GuQj8x0j9WJXYc15mXl9RHwMeCvwwYi4\nBfh6Zt4XEc+hdHcDOD4z/9jPMSJiNcozseOUL8HWTZ+VI+IhLO4Gdmtm3l0r290R8W7Kc2IHR8R1\nwCer9VtRvizHgLMz8wddjr0K5a7qfcDp/ZR/JlU/fFZg8XvTGmhixbb3ZkGHO9pvp/yo2j0ijgLe\nk5m3R8TGwFcp0xNcTp+jcFbPbtXn7W1F5JdvK9vC2rPPrWf7ngd8GfhlNTJq63PwIsoonssC1wNH\n9yjC1tXyZ/2Ufzr6rZdqBNnHUj6bv2m9LxGxJvAK4N1V0suAL/ZRrm2A71DmNzweeNUUd3EYZSTV\npwFfiog3ZuY/I2Idyufk0ZSpO47qsY9h1svuwP+r/vxQp9FoJzBj18x06mauXzPTqZeZvmaAT0TE\nPcAJwPmZeWe1/9UpDdb3UZ69vJ4yCnK7uXzNfDMiLga+BfwxM+8BiIiHURqdR1TpLuGBAzO1zOh3\nzAD0PQpyRGzE4jmDfz1Bcmkk2HDVqHgHpXH3HOArwBciYhGLRxM8j8V3lftxCmVEyHZH88AfYa+g\n/HC7X2Z+JiKeAuxfpf1ARNzF4gbV5ZQfdt3sRol+/GRAIx0P2rGUkULbvaF6tRxeve6XmT+MiHdS\nnid7C/CmiFgArFIluRF4QevHSh/eBnSamP6F1avly5S6a5kH7Fm9iIjbKN3tHlxLcxXwwszsOGBW\nRKxLuRt+M51Hr55p/dbL8pQGy6sAIuJWyo+r1Wpp/gA8v896eQ+Lr8udgesiolO6jiN4ZuYfImJ/\n4AuUeTP3rcq4apXkdmD3BtfLh1h842u/iHh5l3TjwG7tI9nO8DUznbqZ69fMdOplpq+ZFSjfX68H\nxqubs2Nt+7+S8t4u0Y13jl8zq1Man28H7qvOfRkWf96hRFV3zcxF7Zln4Ttmuibz3Hc3rd8Nx2fm\nXYMojNR0dhXWSMjMeynTDLyGMlfgQmARcD4lErttPaLWh/EpvDqV79XAXpRnbm6lXHuXAu8FnpyZ\nnYbeb3lxtRzWHeGJTPe9OZLyI/lUyqjJy1Ia8x8DnpiZ3QYLmcmynQG8E/ghcEW1fWVKt+/TgAOr\nsl3Q49j3d0FrRUhmWb/nfhJwJCW6fxXlszqfMgDKD4BXUqaL6nee1bHacR/CkiN3TjiCZ2Z+BdgK\n+AZlJNDlgaspEbMnZ+aZPY4/7Hqpn3+nkUvr599xmowZvGamUzdz/ZqZTr3M9DXzAcoIxz+mvLfL\nVPu/rlr335T39g/ddjCHr5k3U26snUkp7/KU8/8r5YbyvsBmvW7qzuD10ita2vU7b4ppetmnyt/U\n3wbSwE3nTo804yLiPsp/zOtN40t/zqrudP+Fcjf98Z3uKM/gsZfq9346IqJ1Y2Jdyg+jPw9w39ZL\nn2ayXqr9Wzd98pppJuulmSJiW8pzrb/IzO2HXR5pthhx1VyxtN5keTvlOj10NhutbZbW93469gY2\nAP7foBtHNdbL1M1GvYB10w+vmWayXpqp9QjMoEb8luYE/7NQo1V3ZB9gDs2vNi3VwBl/Bi7MzKdN\nlH4Gjr/UvvfTUY3a+UfKgCePy8yBzq9nvfRnpuulOoZ10wevmWayXpopIrYEzgG+lZl7Drs80mxy\ncCY13fVM7xmQOavqOtVtXtrZsNS+99NRTXL/hBk8hPXSh1moF7Bu+uI100zWSzNl5rnYY1KSJEmS\nJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmS\nJEmSJEmSJEmSJEmSJEmSpCEaG3YBNNoW3jM+PuwyDML8ZcryznuHWw490CjWy6XX3DrsIkzbUx6z\nKgDnXzX3z6Vl3thofF1uus4qAFx49W1DLslgbLj2KsMuwkCM4v9lo2AU62WFZefWf2bzn/zaof2O\nvPOCY+fUe7U0mDfsAkiSJEmS1IsNV0mSJElSoy0z7AJIkiRJ0hLGjLFpMT8NkiRJkqRGM+IqSZIk\nqXnm1lhSmmFGXCVJkiRJjWbEVZIkSVLz+Iyravw0SJIkSZIazYarJEmSJKnR7CosSZIkqXkcnEk1\nRlwlSZIkSY1mxFWSJElS8zg4k2r8NEiSJEmSGs2GqyRJkiSp0ewqLEmSJKl5HJxJNUZcJUmSJEmN\nZsRVkiRJUvM4OJNq/DRIkiRJkhrNiKskSZKk5vEZV9UYcZUkSZIkNZoNV0mSJElSo9lVWJIkSVLz\nODiTavw0SJIkSZIazYirJEmSpOaZg4MzRcQawG7Ac4FNgLWBu4GLgOOA4zJzvJZ+XeCKHrs8MTP3\n6XKs/YDXAhsBi4DzgaMz89Qu6VcADgH2BtYBbgXOBA7LzMu65HkUcASwC7AGcB3wHeDwzLy5S56t\ngXcAWwLzgT8DXwQ+kZn39TjXnmy4SpIkSdJgvAj4FHAtcAZwNbAWpTH7BeDZwJ4d8l1AaRC2+2On\ng0TE0cAbgb8CnwOWpzRIvxcRr8/MY9vSLw/8FNga+A1wMqXxuifw3Ih4Rmae15bnscA5wJpV2S4D\ntgAOBHaJiG0y86a2PC8AvgXcAZwI3AQ8H/gIsE31/vTFhqskSZIkDcafgF3bo54R8XbgPGD3iNgt\nM09uy3dBZh4xmQNUEc03An8BNs/MW6r1HwJ+BxwdEd/PzKtq2d5IabSelJl71fZ1IqVR+sWI2KQe\nDaY0wNcEHtAQjogPAwcDRwIH1NavCnweuAfYITN/X61/F3A6sEdE7JWZJ07mPNv5jKskSZKk5hmb\nN7xXnzLzjE5ddTPzeuAz1Z/b932A4jXV8shWo7U6xlXAsZTo6yta6yNirMozDry1rVzfBX4BbFwv\nVxVt3Rm4sj16CxxGiai+NCJWrK3fA3gocEKr0Vod4y5K12GoNXSnyoarJEmSJM28e9uWdY+MiFdH\nxNur5SY99vMMSiP0Rx22/bBa7lhb91jg0UC2RWHb8zyjtq6V/yftiTPzduCXwEqU51jr5aJLuc4C\nFgJbRcSyHbZPyIarJEmSpOYZGxvea8AiYhngZdWfnRp2OwOfBt5bLS+MiNMj4tFt+1mJMuDT7VUU\nt91fWklr6zasltmleK08G0whz5+nkiczFwFXUh5VXb/LPnuy4SpJkiRJM+sDwBOAUzPzp7X1Cyij\n9v4bsHr12p4ysNMOwGlt3XFXq5a30Flr/epDyjM+QZ6xtjyT5uBMkiRJkppnGs+aNklEvIEyONKl\nwL71bZl5I/Dutiy/iIh/B86mjOL7n8DHZ76kzTYanwZJkiRJapiIeB3wUeBiYMduc5+2q7rWfqH6\n8+m1Ta1o5mp01lpfP85s5hmbYp5Js+EqSZIkSQMWEQdRIqUXURqtN0xxF/+oliu1VmTmAsocsStH\nxFod8rSeOa0/Z3pZq0hdjtMrz4Z01inPn7rlqZ7xXY8yVc4VXfbZkw1XSZIkSc0zB6fDaYmItwHH\nAOdTGq3/mCBLJ60Re9sbeqdRIpu7dMjz7Gp5emtFZl4OXA1sGBHrTiYP5RlbgJ2r6XTuFxGrANtQ\nns89t61cdCnXdsAKwDmZeU+H7ROy4SpJkiRJAxIR7wTeD/wW2Ckzb+qR9t/aG4bV+p2AgymDHf1v\n2+bWfLCHRsTqtTzrAq8F7gSO65Lng/XjRcQLgG2BizPz5631mXkFZSqc9ap91h0OrAh8NTMX1tZ/\nkxIl3jsinlo7xnzKaMlQRkzui4MzSZIkSWqeeYOflmamRcR+lIbdIsrgSgdFLNFD98rM/HL172OA\nx0XEOcA11bonUeZRHQfemZn1qCaZ+auIOIYy4NMfIuJbwHLAXpQRe1+fmVe3HfMY4HnAHsCvI+J0\nYB1gT0rk9JUdTue/gXOAj1cN6csog0XtQOkWfGhbuW6LiP0pDdgzI+IE4F/A8yndlE/KzG90et8m\nw4arJEmSJA3GutVyHnBQlzRnAq2G61eAFwKbU7rsLgv8HTgR+GRm/rLTDjLzzRFxESUauj+lofx7\n4EOZ+YMO6e+OiJ2BQ4B9qrLdApwMHJaZl3XIc0VEbEaZrmcX4DmU52s/ChyemUtMe5OZp0TE9pRG\n7e7AfMqcrwczzZGR595tDM0pC+8ZHx92GQZhfnWL5857h1sOPdAo1sul19w67CJM21MesyoA5181\n98+lZd4MTEY/DJuuswoAF15925BLMhgbrr3KsIswEKP4f9koGMV6WWHZufWf2fwdjhja78g7z3zX\nnHqvlgZGXCVJkiQ1z4jM46rB8NMgSZIkSWo0I64dRMR97esyc6lo5EfEOsBfgAsy82nDLo8kSZKW\nUnOrZ7NmmA3X3m6kPOjcVUTMA/4TeAWwEfAgygPIXwc+1u88RdUw1TtQHtTerFo+ptp8QGZ+dpL7\n2RM4ANiUMnfSVZSRvo7KzNvb02fm1RFxHLB/ROyVmSf2U35JkiRJGpSlIorYp3Fg88xcOzPX7pQg\nIpYFvkeZF2kLYHnKgFdPBj4InB0RK/V5/NUok/h+gDJs9WNq2yb1oHpEfI4yItkOwMrAPcCGlFG+\nLoiIR3TJ+j7gXuDIiPDmhiRJkmbf2LzhvdQ41sr0vJcybPVCYD9gxcxcGdgVuIkSJZ1UZLSLBcBZ\nlHmXXkwZGntSIuIASiR4EfBmYOXMXA3YhhJ1XR/oOI9SZl5FaTSvD/zHNMovSZIkSdNmw7VPEbEW\ncGD159sy86uZOQ6QmaeyeBLffSJik6nuPzNvzsxVMnOHzHxzZp4A3D3Jsi0PvLv682OZeUyry3Jm\n/ooyV9Q4sE1EPK/Lbr5WLfefatklSZIkaZBsuPZvd2A54Gbgc+0bM/O7QFK6Dr94dovGM4E1gfuA\nD7dvzMwLgJ9Vf76kyz6+TekuvFNEPHomCilJkiR1NTY2vJcax4Zr/3aslmdlZrdI6E+q5TNmoTx1\nrbL9MTOv65Lmx21pHyAzFwAXUT4js11+SZIkSbqfDdf+bVwtL+6R5pJq+fgZLku7qZRtzYhYo0ua\n86rldgMplSRJkjRZDs6kGmulf60Rea/tkaYV7Vx5GqML92MqZaunb9dq3D512iWSJEmSpD451Un/\nWg3RhT3S3FH798qUUYJnQz9l6+Smavmwfgsyf8Q+YaN2PqNilOrlKY9ZddhFGJhROpdRs+k6qwy7\nCOpglP4vGyWjUi933jvsEkjTMyKXomZIq+H6kKGWQpIkSUsfB0lSjQ3X/i0AVgNW6JFmxdq/bweI\niK2BkzukHQc2z8y/DahsTLVsM2FU7u617raOyvmMilGsl0uvuXXYRZi2VqT1/Kvm/rm0zBuRH0+t\nSOuFV9825JIMxoZrj0bkeBT/LxsF1ovULDZc+3ctpeG6do80rW23V6P0AixL6Xo73iH9oJ45vhbY\ndJJlgwc+71rXGrTpn4MolCRJkjRpDpKkGhuu/bsY2Ah4Yo80rdF9L22tyMyfM/ODYl0MPJvJle3G\nzLypS5pWw/WGQRVMkiRJkqbK2xj9O6NabhsRy3dJs3O1PG0WylPXKtvGEbFWlzT/Xi17la3V8P3t\nQEolSZIkTdbY2PBeahwbrv07GbgLeDDwn+0bI2JXIID7gONnt2icRomSPgh4U/vGiNgUeCalu/LX\neuznadXyrEEXUJIkSZImy67CfcrM6yPiY8BbgQ9GxC3A1zPzvoh4DnBclfT4zPxjP8eIiNUoz8SO\nA2MsvtFTE/2vAAAgAElEQVSwckQ8pFoHcGtm3l0r290R8W7gU8DBEXEd8Mlq/VaUxuoYcHZm/qDL\nsVehRFzvA07vp/ySJEmSNAhGXKfnHcAPKKP3fgW4IyIWAN+nPB96HvCaaez/FErk9MZq+ahq/dG1\ndTcA+7RnzMzPAJ+n1PHRwO0RcRvwS2Bd4HLgRT2OvRslYvuzAY10LEmSJE3e2LzhvdQ41so0ZOa9\nwK6Uxum5wEJgEXA+JRK7bW004X6MT+HVqXyvBvaiPPN6K6W+LwXeCzw5M//e49gvrpafn0b5JUmS\nJGna7Co8TZk5Dnyueg163zsOYB8nASdNJU9ErAvsRInKfme6ZZAkSZKmzEGSVGPEtbel9Wp5O+Wz\ncWhmLhp2YSRJkiQt3Yy4djcGXBkRAGTmUtHIj4h1gP2A32bmN4ZdHkmSJEmy4drZ9XR5bnTUZebV\nQLd5aSVJkqTZ4SBJqrHh2kFmPmLYZZAkSZIkFTZcJUmSJDWPEVfV+GmQJEmSJDWaEVdJkiRJzeN0\nOKox4ipJkiRJajQbrpIkSZKkRrOrsCRJkqTmcXAm1fhpkCRJkiQ1mhFXSZIkSc3j4EyqMeIqSZIk\nSWo0G66SJEmSpEazq7AkSZKk5nFwJtX4aZAkSZIkNZoRV0mSJEnN4+BMqjHiKkmSJElqNCOukiRJ\nkhpnzIiraoy4SpIkSZIazYarJEmSJKnR7CosSZIkqXHmYlfhiFgD2A14LrAJsDZwN3ARcBxwXGaO\n19JvUKV/FrAB8DDgX8C5wEcz88wOx3g58MUexTggMz/bId8KwCHA3sA6wK3AmcBhmXlZl/N5FHAE\nsAuwBnAd8B3g8My8uUuerYF3AFsC84E/V+X9RGbe16PcPdlwlSRJkqTBeBHwKeBa4AzgamAtSuP0\nC8CzgT1r6d9T5bkY+D5wE/B44PnA8yPiwMz8RJdjfQe4oMP637SviIjlgZ8CW1fbT6Y0XvcEnhsR\nz8jM89ryPBY4B1izOtZlwBbAgcAuEbFNZt7UlucFwLeAO4ATq/N5PvARYJvqXPtiw1WSJElS88y9\ngCvAn4BdM/PU+sqIeDtwHrB7ROyWmSdXm34IvD8zL2xLvx2lofmhiDgpM//e4VjfycyvTLJcb6Q0\nWk/KzL1qxzmR0ij9YkRsUo8GUxrgawKvz8xja3k+DBwMHAkcUFu/KvB54B5gh8z8fbX+XcDpwB4R\nsVdmnjjJMj+Az7hKkiRJ0gBk5hntjdZq/fXAZ6o/t6+t/3J7o7Vafxbwc2A5SoOzbxExBrwGGAfe\n2nac7wK/ADaul6uKtu4MXFlvtFYOo0RUXxoRK9bW7wE8FDih1WitjnEXpesw1Bq6U2XDVZIkSZJm\n3r1ty4nc07Zs95SIOCgiDomIfSPikV3SPRZ4NJCZeVWH7T+sls+orduxWv6kPXFm3g78EliJ8hxr\nSyv/jzoc4yxgIbBVRCzbpZw92VVYkiRJUuPMxcGZuomIZYCXVX92ati1p38MsBOwgNLo6+TAtr8X\nRcQXgIOqKGfLhtUyu+znL9Vygynk+TMlIrsBpRtwzzyZuSgirgQ2AtandKmeEiOukiRJkjSzPgA8\nATg1M3/aK2E1kNLXKN2E352Zt7QluQJ4HRDAisAjKIMe/R/wapYccXi1atm+H9rWrz6APOMT5Blr\nyzNpRlwlSZIkNc6oRFwj4g2UwZEuBfadIO2DgK9Snms9ITM/3J6mev61HoW9E/hmRJwLXAjsExFH\nZeYfBnQKjWDEVZIkSZJmQES8DvgoZbqbHbvNfVqlfRDwv5RBjk4EXjqVY2Xm34AfVH9uV9vUioCu\nRmet9fWy9ZtnbIp5Js2GqyRJkqTGGRsbG9prECLiIODjwEWURusNPdIuCxwP7EXpJvzizLyvj8P+\no1rWR/u9rHWYLnlaz7bWn01t5dmQzjrlaT23ukSe6hnf9SgDTV3RZZ892XCVJEmSpAGKiLcBxwDn\nUxqt/+iRdjngJEqk9cuZuW/bfKpTsUW1vL9xmJmXA1cDG0bEuh3yPLtanl5bd0a13LmaTqde3lWA\nbSgDR51b23RatdylwzG2A1YAzsnMbqMk92TDVZIkSZIGJCLeCbwf+C2wU2be1CPt8sC3gecDXwBe\nOYn9b9Zh3byI+B/K9DQ3suTIxa05ZD9Yb4hGxAuAbYGLM/PnrfWZeQVlKpz1gNe27etwSkT3q5m5\nsLb+m5SI794R8dTaMeYD763+/PRE59eNgzNJkiRJapy5ODhTROxHadgtAs4GDopYoofulZn55erf\nn6FEPP8BXAsc1iH9GfVGJXBeRPwR+ANwDeXZ0W0ooxYvAF5SzbVadwzwPEpU99cRcTqwDrBnladT\ng/m/gXOAj0fETpTuw1sAO1C6BR9aT5yZt0XE/pQG7JkRcQLwL0qjPICTMvMbHY4zKTZcJUmSJGkw\n1q2W84CDuqQ5E2g1XNelTCHzEOBdHdKOA/cB9Ybr0cDTgB2BNSiN5KuBTwLHZOb/te8kM++OiJ2B\nQ4B9qrLdApwMHJaZl3XIc0UV3T2C0v33OZTG9UeBwztM00NmnhIR21MatbsD8ylzvh5Med63b3Pv\nNobmlIX3jPfbP79R5le3eO68d7jl0AONYr1ces2twy7CtD3lMasCcP5Vc/9cWubNwbv+nWy6zioA\nXHj1bUMuyWBsuPYqwy7CQIzi/2WjYBTrZYVl59Z/Zqvu85Wh/Y689fiXzan3amngM66SJEmSpEaz\n4SpJkiRJajSfcZUkSZLUOHNxcCbNHCOukiRJkqRGM+IqSZIkqXGMuKrOhqu0FBofjcGeaQ2MPjrn\nA7fdNTrDV47Sudy1aNGwizAgZRTeGxbcOeRyDMaGjMaowpKkidlwlSRJktQ4RlxV5zOukiRJkqRG\ns+EqSZIkSWo0uwpLkiRJahy7CqvOiKskSZIkqdGMuEqSJElqHgOuqjHiKkmSJElqNBuukiRJkqRG\ns6uwJEmSpMZxcCbVGXGVJEmSJDWaEVdJkiRJjWPEVXVGXCVJkiRJjWbEVZIkSVLjGHFVnRFXSZIk\nSVKj2XCVJEmSJDWaXYUlSZIkNY89hVVjxFWSJEmS1GhGXCVJkiQ1joMzqc6IqyRJkiSp0Wy4SpIk\nSZIaza7CkiRJkhrHrsKqM+IqSZIkSWo0I66SJEmSGseIq+qMuEqSJEmSGs2IqyRJkqTGMeKqOiOu\nkiRJkqRGs+EqSZIkSWo0uwpLkiRJah57CqvGiKskSZIkqdGMuEqSJElqHAdnUp0RV0mSJElSo9lw\nlSRJkiQ1ml2FJUmSJDWOXYVVZ8RVkiRJktRoRlwlSZIkNY4RV9UZcZUkSZIkNZoRV0mSJEnNY8BV\nNTZcJUmSJGkAImINYDfgucAmwNrA3cBFwHHAcZk53iHf1sA7gC2B+cCfgS8Cn8jM+7ocaz/gtcBG\nwCLgfODozDy1S/oVgEOAvYF1gFuBM4HDMvOyLnkeBRwB7AKsAVwHfAc4PDNv7pJnyucyGXYVliRJ\nkqTBeBHwOWBz4FfAR4BvAU8EvgB8oz1DRLwAOAvYtkr7CWC5Ku8JnQ4SEUdTGsIPr473v5SG8vci\n4rUd0i8P/BR4J3Az8FHgZ8ALgd9GxNM65Hks8Dvg5cC5wDHAFcCBwK+qRvq0z2WyjLhKkiRJapw5\nOjjTn4Bd26OeEfF24Dxg94jYLTNPrtavCnweuAfYITN/X61/F3A6sEdE7JWZJ9b2tTXwRuAvwOaZ\neUu1/kOUhubREfH9zLyqVoQ3AlsDJ2XmXrV9nUiJoH4xIjZpiwZ/ClgTeH1mHlvL82HgYOBI4IDa\n+imfy1SMZMQ1Iu5rfw27THNBRIxFxCURcVtEPHzY5ZEkSZLmksw8o1NX3cy8HvhM9ef2tU17AA8F\nTmg19Kr0d1G620KtcVh5TbU8stVorfJcBRwLLA+8orU+IsaqPOPAW9vK9V3gF8DG9XJV0dadgSvr\njdbKYcAdwEsjYsVpnsukjXrE9UZKf++OImJLYAtKKH8zIKpNR2Xm/0zmABHxTEq4fAtgVeAa4HvA\n+zLzhn4LHhFPotwV2bx6bUy50XBiZu4zQd4vAS+b4BCnZuau9RWZOR4R7wG+BrybaXywJEmSpOmY\noxHXXu5tWwI8o1r+qEP6s4CFwFYRsVxm3l3LM94lzw8p3YF3pPyeB3gs8GjgT21R2Hqep1f7PbNa\nt2O1/El74sy8PSJ+SWnYbkmJpk7lXJbNzHs6pOlpJCOulXFK6HztzFy7S5ofUfpbv5jFjdZW3glF\nxKGUynwu8GBKZawHvAG4KCKe0GfZAb5CCc+/gtInvlVXkypb5Xbg711eN3XJcwKQwKsiIrqkkSRJ\nkjRJEbEMiwNL9YbdhtUy2/Nk5iLgSkqwcf1qPytRBny6vYritvtL65CTOUZbng2mkOfPU8nT6Vym\natQjrhO5A7gU+A3wW+Ag4MmTyRgRzwHeQ2lIfpgystaCiNiY8nD0k4FTImLj2t2RqbibMjLYb6vy\n7Q48a4r7ODozj5hKhirqehzwfkoD/HVTPKYkSZKkB/oA8ARKr8ef1tavRmlP3NIxV1k/VqWjtuyV\nHmD1tmPMVp7JnMvqXbb3tLQ3XB9ZfwA5Il45hbzvq5bfzsz7+4pn5iURsSulQbw+8F/AJ/so25b1\n4aIjYts+9tGvr1Mari+OiLdk5sJZPLYkSZI0Ml2FI+INlMGRLgX2HXJx5qxR7io8oU5zKE1G1QX4\nSZQ7Ch/qsN9rgOOrP1/SZ9mGNqBUZv6VMuT16pR5qCRJkiRNUUS8jjL1zMXAjh3mPm2PqLZrrW/l\nu6Vt/UTpZzvPVM5lSpbqhus0tB5WviUzf90lzY+r5eZVX/RhmM5tql9Vy2cOoiCSJEnSVIyNjQ3t\nNQgRcRDwceAiSqO108Ctf6qWG7ZvqJ6LXY8yvcwVAJm5ALgWWDki1uqwv9Yzp/XnTC9r7bJLUXvl\nWaJcPfJM6VymyoZrfzaulpf2SHNJtRwDHj+zxenqJRFxVUTcHRE3RcTZEfGWiFhlEnnPq5ZPn8kC\nSpIkSaMmIt4GHEMZs2bHzPxHl6SnVctdOmzbDlgBOKdtFN7TKG2MTnmeXS1bI/2SmZcDVwMbRsS6\nk8kDnFEtd66m07lf1ZbYBlhA6aU5nXOZNBuu/XlEtby2R5rrav/udDdkNjwWeDhwK2Wqnq2Boygj\nHj9pgrwXV8v1IqKvB6glSZKkvo0N8TUNEfFOyngxvwV2ysxus3kAfBP4B7B3RDy1to/5wHurPz/d\nlqc1H+yh9d/pVaP0tcCdwHFd8nyw3hCNiBcA2wIXZ+bPW+sz8wrK7CnrVfusOxxYEfhq21g4/ZzL\npC3tgzP1q9X1t9egRXdUyzFg5ZktzhJ+B5xDGbXsGoCIWA3YizKi2TrADyNikx4X0r9q/34YffZF\nnz9in7DROZ/RGOygZYVlR+d8tos1hl2EgRmlcxk1O2+05rCLoA5G5ztmtIxKvdx578RpNH0RsR+l\nYbcIOBs4qMMMk1dm5pcBMvO2iNif0ug7MyJOoPwOfz6la+9JmfmNeubM/FVEHEMZ8OkPEfEtYDnK\nb/3Vgddn5tVtxzwGeB6wB/DriDid0ibYkxI57TRI7X9T2hQfj4idKN2HtwB2oHQLPrStXFM+l6kY\nkUtRdZn5iQ7rbgE+FxHnUUL6jwDeRNsHrqbeoF2T7nM4SZIkSSrWrZbzKFNtdnIm8OXWH5l5SkRs\nT/ldvjswnzJP6sGUZ2SXkJlvjoiLKNHQ/SkN5d8DH8rMH3RIf3dE7AwcAuxTle0W4GTgsMy8rEOe\nKyJiM+AISvff51B6nH6UMhXoEtPe9HMuk2XDtT8LquUKPdKsWC3HgdtbKyPi7x3SjgMHTucOxGRl\n5gXV3Y+XAbvSveE6EKNyd691t3VUzmd8vK8BtRunFWldeM9onA/Ab67818SJGq4VaT2rZ8+oueWu\nRYuGXYSBaEVaf3rpjUMuyWA8fYPRiByP2nfMqLBehm8uToeTmYdTIq5TzXcO8Nwp5vkytQbwJNIv\nBA6rXpPN8zc6R2N75ZnyuUyGDdf+tJ5tXbtHmvq2+vOuD6M0VNvNn26hpuDXlIbrej3S1Pv4jcYv\nHEmSJElzkg3X/rQGLtooIsa6zAfbGnl4nNrow5nZhAGxJnP7qt5w7TR0tyRJkjRj5mLEVTOnCY2o\nuag1PPRqwOZd0vx7tfx122hbTbBFtbyyR5onttJ0mChZkiRJkmaNDdc+ZOalwIWUyOVb2rdHxNqU\nh54BvjaLRZtQRGwK7F39eWqPpE+rlmfNbIkkSZIkqbeluqtwRKxEGWBpnNIIXbbatGJEPITFXWoX\ndIiavp3S8Ns9Io4C3pOZt0fExsBXKVPgXA58vs+yrcDiaXcAlm8t28q2MDMX1PLtSxnq+svAL1uj\nfVXT4byIMh3OssD1wNE9irB1tfxZP+WXJEmSpsOewqpb2iOux1Ke37yxWm5VrX9Dbd0NwFvbM2bm\nD4F3Vn++Bbg5Im4B/gg8pcr/gsy8p8+yva12/Bso8zIBvLCtbJ9syzePMh/T94F/RcQtEfFPyhxK\nnwUeDFwFPDsz/9npwNXkxZtT5m49uc/yS5IkSdJALO0N1/EpvJaQmUcCO1Mir/+kRDIvBz4GPDEz\nLxlC2c6gNKh/CFxRbV+Z0sg9DTiwKtsFPY59fzfnzLxzGucgSZIk9WVsbGxoLzXPUt1VODNfAbxi\nmvs4jdIgHKhpzAF1NXBkv8eNiHnAy4F7mOYkwZIkSZI0CKPecPV2ydTtDWwAfCYz/zzswkiSJGnp\nZOBTdaPccB0DrowIoDHzpzZaRIwBhwK3A+8ebmkkSZIkqRjVhuv1dHkuVd1l5jjwhGGXQ5IkSZLq\nRrLhmpmPGHYZJEmSJPXPQZJUZ/dZSZIkSVKjjWTEVZIkSdLcZsBVdUZcJUmSJEmNZsNVkiRJktRo\ndhWWJEmS1Djz5tlXWIsZcZUkSZIkNZoRV0mSJEmN4+BMqjPiKkmSJElqNCOukiRJkhpnzJCraoy4\nSpIkSZIazYarJEmSJKnR7CosSZIkqXHsKaw6I66SJEmSpEYz4ipJkiSpcRycSXVGXCVJkiRJjWbD\nVZIkSZLUaHYVliRJktQ4dhVWnRFXSZIkSVKjGXGVJEmS1DgGXFVnxFWSJEmS1GhGXCVJkiQ1js+4\nqs6IqyRJkiSp0Wy4SpIkSZIaza7CkiRJkhrHnsKqM+IqSZIkSWo0I66SJEmSGsfBmVRnxFWSJEmS\n1GhGXCVJkiQ1jgFX1RlxlSRJkiQ1mg1XSZIkSVKj2VVYkiRJUuM4OJPqjLhKkiRJkhrNiKskSZKk\nxpmLAdeI2APYHngysCmwMvC1zNy3Q9ovAS+bYJenZ+Yza3leDnyxR/oDMvOzHY61AnAIsDewDnAr\ncCZwWGZe1uVcHgUcAewCrAFcB3wHODwzb+6SZ2vgHcCWwHzgz1V5P5GZ9/U60YnYcJUkSZKkwXgH\n8CTgNuBvwOOB8S5pvw1c0WXby4D1gR902f4d4IIO63/TviIilgd+CmxdbT+Z0njdE3huRDwjM89r\ny/NY4BxgzepYlwFbAAcCu0TENpl5U1ueFwDfAu4ATgRuAp4PfATYBnhRl3OZFBuukiRJkjQYBwF/\nzczLI2J74IxuCTPzFOCU9vURsTolOnoX8KUu2b+TmV+ZZJneSGm0npSZe9WOcyKlUfrFiNgkM+sN\n7E9RGq2vz8xja3k+DBwMHAkcUFu/KvB54B5gh8z8fbX+XcDpwB4RsVdmnjjJMi/BZ1wlSZIkNc7Y\n2NjQXv3KzDMz8/LWKfS5m30p3WxPbo9qTlVEjAGvoUR931rflpnfBX4BbEzp3tzK81hgZ+DKeqO1\nchglovrSiFixtn4P4KHACa1Ga3WMuyhRaKg1dPthw1WSJEmSmmP/avm5HmmeEhEHRcQhEbFvRDyy\nS7rHAo8GMjOv6rD9h9XyGbV1O1bLn7QnzszbgV8CK1GeY21p5f9Rh2OcBSwEtoqIZbuUc0J2FZaW\nQqM2vPwonc+z9nrXsIswbQvP/yQwGufS8rtTjxp2EQbqkautOHEiSRqyEfp6n7SI2Ap4IvCnzPx5\nj6QHtv29KCK+ABxURTlbNqyW2WU/f6mWG0whz58pEdkNKN2Ae+bJzEURcSWwEeW53T912W9PRlwl\nSZIkqRn+q1p+vsv2K4DXAQGsCDyCMujR/wGvZskRh1erlrd02V9r/eoDyDM+QZ6xtjxTYsRVkiRJ\nUuOMUo+qyYiI1SiN0K6DMmXmWZSuty13At+MiHOBC4F9IuKozPzDDBd31hlxlSRJkqTheymwAn0M\nypSZf2Px1Dnb1Ta1IqCr0VlrfX1e1n7zjE0xz5TYcJUkSZKk4WsNyvTZPvP/o1rWBzK4rFpGlzyt\nZ1vrz6a28mxIZ53ytJ5bXSJPRCwDrEeZKqfbvLUTsuEqSZIkqXHGxob3mm0RsQXwJMqgTGdNlL6L\nLarl/Y3Damqeq4ENI2LdDnmeXS1Pr61rzT27czWdTr2cqwDbAAuAc2ubTquWu3Q4xnaUSPI5mXnP\nxKfRmQ1XSZIkSRqu1qBMvabAISI267BuXkT8D2V6mhtZckqaz1TLD9YbohHxAmBb4OL6CMaZeQVl\nKpz1gNe27etwSkT3q5m5sLb+m5SI794R8dTaMeYD763+/HSvc5uIgzNJkiRJapy5ODhTRPwH8B/V\nn2tVy60j4kvVv2/MzLe05VkV2Isy0NKXJzjEeRHxR+APwDWUZ0e3AZ5AiYK+pJprte4Y4HnAHsCv\nI+J0YB1gzyrPKzsc57+Bc4CPR8ROlO7DWwA7ULoFH1pPnJm3RcT+lAbsmRFxAvAv4PmUbsonZeY3\nJji3noy4SpIkSdJgbAq8DNiXMtfpOCVy+bLqtXuHPC+hRDG/PYlBmY4GbgJ2BN5AGdDpQcAngU0y\n82ftGTLz7qos76FMR3MQsBNwMrB5Zv6mQ54rgM0ooxtvAbyxOo+PAltm5r865DkF2J4y6vHulGl7\n7gIOBvae4LwmNPduY2hOWXjP+PiwyzAI86u+CXfeO9xy6IFGsV4evPnrhl2EaVt4/icBWOEpc/9c\nWn536lHDLsJAbLz2SgBccu2CIZdkMNZ/2ErDLsJAjOL/ZaNgFOtlhWXnVghz6w+eNbTfkee8dbs5\n9V4tDewqLEmSJKlx5lYzWzPNrsKSJEmSpEYz4ipJkiSpcebi4EyaOUZcJUmSJEmNZsRVkiRJUuMY\ncVWdEVdJkiRJUqPZcJUkSZIkNZpdhSVJkiQ1jj2FVWfEVZIkSZLUaEZcJUmSJDWOgzOpzoirJEmS\nJKnRbLhKkiRJkhrNrsKSJEmSGseewqoz4ipJkiRJajQjrpIkSZIax8GZVGfEVZIkSZLUaEZcJUmS\nJDWOAVfVGXGVJEmSJDWaDVdJkiRJUqPZVViSJElS48yzr7BqjLhKkiRJkhptKBHXiHgusC3wIOBC\n4JuZedcwyiJJkiSpeQy4qm6gDdeIeBzwIWAceF9m/rZt+7LAd4FntWV9R0Q8KzOvHmR5JEmSJElz\n36C7Cu8NvADYDvhDh+3vYMlGK8CGwCkR4X0VSZIkSdIDDLrhuk21/Elm3l3fEBHzgYOqP28CXgfs\nSonAAmwK7DXg8kiSJEmag8bGxob2UvMMuuH6mGr5uw7bngWsUv37VZn5qcw8FdgD+Eu1fo8Bl0eS\nJEmSNMcNuuH60Gp5bYdtO1bLG1kcZSUz7wVOqP58yoDLI0mSJGkOmjc2vJeaZ9AN1wdXyzs7bGt1\nIz49M8fbtl1RLR8+4PJIkiRJkua4QU+HsxBYGVijvjIiVmFxNPXsDvkWVMvlBlweSZIkSXOQz5qq\nbtAR16uq5ZZt659bHWsc+GWHfK1I7a0DLo8kSZIkaY4bdMO11SjdJyK2BIiIVYG3VetvAC7skO+J\n1fLKAZdHkiRJkjTHDbqr8OeA/wJWBM6KiIuBR7O46/AXOjzfCrBDtbxoEIWIiPva12XmoBvpI6ea\nR7dVZ4/LzOuHXCRJkiQtpewprLqBNlwz8/cRcQTwrmrfm9Y2XwIc1Z4nIjaiRFzHgV8MsjyUEYwX\nddtYRYW3ADYHNgOi2nRUZv5Prx1HxJnAdhMc/9jMfP1kC1vb9zLAzsBzgK2ADYAVgH8C5wFfzMxT\nJrGfzYA3V+VcgxLx/jHwgcy8vD19Zo5HxHv+P3t3Hi5JUSVs/L2tQDeN7CCgMixy2EZwVARBQVDm\nQ2VRFkEUUREVERFcPj7RYRtcAR2RZdQBRBlBEHdldFhEZR1FYdD2II2C7LLTNNjA/f6ILDspqu6a\nt6tu9fvzqSduZUZERt5qmz51IiOAs4AjgQPGO3ZJkiRJalrTGVcy88iIuAbYH3g+ZeGln1KCpYc7\nNDmkKoeACxocyjCwWWbePEKdC4Blu7Qdqwcoi1J1OzcRpwD71d7/DXgEWBXYCdgpIs4D9q62E3qa\niNgX+ArwDOBJyvPDz6n63Ssids7Mizs0PRs4AtgvIj6XmTnBe5AkSZImbAhTrlqo8cAVoMoGjpoR\nrOq+izK9uBceAX4PXA38D/AB4IXj7OPgzDyz4XE9E7iVEnh+OzOvBYiI1YGPAgcCu1OeCf6/7Y0j\nYhPgy5Sg9evAIZl5T0SsWR3fHvhWRERm/rXetsq6ng58Eng/8L6G702SJEmSxmVxf+7zOZn5ssx8\nfxV89suqxicD62TmUa2gFSAzb6+mHp9RHTowImZ2aH80Jfi9Gtg3M++p2t8M7ArcAiwPHNbl+v9Z\nlXtHxKzJ3owkSZIkTcZiHbh2WSiq5zLz6sxcMEKVM6pyFrBh/URELE95NhbghPZ7zMx5wKnV2zd1\nuf4twBWU4HbXcQ1ekiRJasCMod691H8aD1wjYtnqtWSX8xtExLkRcVtE3BURP42I7ZoexyLUiz/a\n99Z+bv8MX07Jtg4DP+nS/r+qcrVqcaxOLq/KV09ohJIkSZLUkEafcY2IHYAfUYKmzSnPjdbPr0cJ\niBclWekAACAASURBVJarHX4VsG1E7JuZZzU5nkXkQxHxSWBlymJMvwXOA07PzMem6JrbVOUCoH3x\npI2q8o7MvK9L+9+31f99hzpXVeUrJjRCSZIkaRKG3A9HNU1nXHesyt9n5v90OP95FgatT1C2d2mN\n45SIWK3h8SwKG1Gm1D5E2XJmO8ozqldFxPOavlhELMPCZ1PPz8yH2qqsXpW3desjM+dTguwhoNvv\n/PqqXLuafixJkiRJPdF04Priqvzv9hPVirY71M6vlJmrAHtSMrTLAO9seDxT6WJgH2D1zFw6M1ei\nBIEfBR4DXgD8KCKWaPi6p1K2tXmAzosrza7Kblv0tDxSlct0OV/P1q465tFJkiRJDRga6t1L/afp\n7XCeXZXXdzj3OkqGbxh4b2Y+CJCZ50bE2ylB7fbAvzY8pimRmUd1OHY38KmIuBb4AbAx8DbKFjST\nFhGHAXtT9mXdf5Q9aier/hztKjx9SvKYzJySDZd6Z9DuZ1AM0ucy/5ov9noIjRmkexk0G60xe/RK\nWuQG6e+yQTIon8ujj/d6BNLkNP1/xZWr8t4O57auyusy849t535CCVw3aHg8PZGZP4qISyn3vBO1\nwDUi7ujQZJiyH+w3u/UZEe8GPlHV/WBmntel6ryqHG0bm6Wr8uFR6kmSJElSTzUduLb2FO00BXnL\nqrykw7lWMDdIz1JeSQlc1247viol+GzXaT9WACJiH8pzs8PAkZn5byNc99aqXGOE/max8Fnj27tU\nW7H2890jXG9Eg/LtXuvb1kG5n0ExiJ/LCpu9r9dDmLRWpnXWP03/e2n51Q8/3eshNKKVaf3dbfNG\nqTk9rLPqYGSOB/HvskHg59J7M6bhnN2I2J2ykOoLgU0pj+WdlZn7dKi7FjB3hO7OycyO21dGxL7A\ngZStMZ8ArgGOy8wfdqk/i/KY4V7AmsCDlLjsiMyc06XNc4GjKQnGFSlxw3eAozLz/i5ttgQ+BmxB\niW9uAE4DTszMJ0e411E1HbjeT8m6rlk/GBHPB1oLFV3WoV0r0H2i4fH0Usf/p2XmuJ4rjog9gNOr\n/o7LzGNGafK7qlwtIlbMzE7Z79bKw8O1+u3qgetdYx2vJEmStBj7GLAJZeHWv1BmlHZKWtX9hhIQ\ntvvfTpUj4jjgUOAW4EvAUpSA9PsRcVBmntRWfyngp5RE4tXA+ZR4bQ/gdRGxXWZe1dZmXUrctko1\ntjmUXWMOBnaIiK3a44yI2AX4FmUtnXMos3B3Bj4HbAW8cZTfw4iaDlx/T9k+ZUfg+NrxvatyGPh5\nh3at7OBfGx5PL21elTdNtIOI2Ak4i2rV5cz8yBia/QJ4nPLZvhroNP34n6vytm7fsAD/WJU3dftG\nRZIkSZoq0zDhCvAB4JbMvDEitqEs6Dqa32Tm0WPpvMpoHgr8EdgsMx+ojn8W+BVwXET8IDP/XGt2\nKCVoPTcz96z1dQ4lKD0tIl6QmfUA+2RK0PqUQDgijgcOAY4FDqgdX5byeOQC4JWZ+evq+L8AFwG7\nR8SemXnOWO6zk6ZXFf5RVW4TEV+MiE0j4s3Ah6vjV2Vmp6mpL6rKGxoeT09U+9m29j/tmK4fQx/b\nA+cCzwDOyMwDx9KuWvSqdc1DI+Ip/5ePiNnAe6q33xihq5dW5aVjHrQkSZK0GMvMSzLzxurtVITe\nrX/HH9sKWqvr/hk4iZJ9fXvreBULvIeSQHxKEiwzv0dJKm5Emd7carMuZdHcm9qzt8ARlIzqWyJi\n6drx3Skzb89uBa3VNR6jZKGhFuhORNMZ13+nRPSrAO+lDK7+gT3tIaGIeAYLM4BXNzyeEVVB3CzK\nBzkEtLauWToiVmLh2OdVe5+22h0GrEvJhl6dmfOq46tQ/qAcWVWdQ5nTPd5xbUX59mNJSnC53zi7\nOAJ4LSX4PCMiDs3Me6otib5MmbZ9Hx0+j5rWM8lP29pIkiRJUmOeUy3EuhJwD3BZZl7Xpe52lNjl\ngg7nfgx8HNiWhfHIupR/+/+hLQtbb/OKqt9LqmPbVuVP2itn5sMR8UtKYLsFJZvaGhddxnUpZavO\nl0XEEpm5oMu9jajRwDUz74+IHYFvU6b/tgK/YeBTmdlp7vZOLHyeciyp9CadBLy1w/H3V6+Wo6pX\ny1KUYHI/gIh4kHKPy9XqXAvsPMEP5hgWrgq8PXB7RHSq13E14sy8NiL2B75C2Wt2n2qMy1ZVHgZ2\ny8x7OnVaPSi+GeWZ5fMnMH5JkiRpUoam6VzhCdi+ev1dRFwC7JuZt9SOzabEWA9l5p0d+mnt3FIP\nHNavym5bW7barDeONjdU412PhYFr1zaZ+URE3ERZSGod4A9d+h1R4ztTZebVUaKs1wLPp2zP8t+Z\n+fsuTZ4HnEnZm/SSpsczimFGf1i6Va/uXMrv7mWUbzFWomRrb6M8XH0eZfWwCX2bwML9bqn6HknH\n1Ygz88yIuJ4yTfsVlC8HbqY8mP3JzBxpBbPW6mVnZeajYx61JEmSpLGaR1m19zssXF14U0q2dFvg\nwoh4YWY+Up1rJckeoLPW8fpOLYuyzfAobYaYxC4yU7KlcvXL7bbPaHvdE6diDGO89tupzQEfR7vf\nUdLwUyIztx291pj6+RVlhbExi4gZwNsoD1Z/oYlxSJIkSeM16AnXzLybhVN6W34eEf9MWXB1c+Cd\n+G9yoPnFmfrNgP9xnxJ7UdL+/5GZA7FYliRJkjRdZOYTlEf+YOGCr7Awm7kcnbWO13cEWZRthsbZ\nZlwGOXAdAm6KiCcjYlKb3S4uqlXHDqc8A3tkb0cjSZKkxdmMoaGevfpAa5vQ2a0D1YKwtwHLRMRq\nHdq0nlOtP2fa2vqy44I5o7RZn846tWk9t/q0NhHxTGBtyozOkR5XHNGgBq53Ane0vTSKzBzOzI0z\nc9nMvKvX45EkSZIWU1tUZXugdyElQbdDhzavqcrWgklUW/PcDKxfLcA6ahsWLpi7fYetNZ8FbEV5\nPveKtnHRZVxbUxaevWwSawBNzTOuABGxNWVF25cCz6WsaDtSoDwEDGfmMyZ77cxcfbJ9SJIkSdJU\niYgXAddk5nDb8VcBh1AWO/p6W7NTKTHW4RHxncy8v2qzFnAg8Chweoc2nwA+ExF7tq4XEbsALweu\nz8yftSpn5tyI+Ally9IDgS/W+joKWBo4tb5dKGV9o08De0XEidVaO0TETOBfqzqnjOkX00XjgWsV\nhZ8J7DKB5n2Rl5ckSZLUW9MxMIiI1wOvr962pvNuGRFnVD/fnZkfrn4+AXh+RFwG3Fod24SyovAw\n8PHMrGc1yczLI+IE4FDg2oj4FrAksCdlxd6DMvPmtmGdAOwI7A5cGREXAWsCe1Ayp+/ocCvvBS4D\nvlAF0nMoi0W9kjIt+PC2cT1Ubcd5HnBJRJwN3AfsTJmmfG77Fp7jNRVThc9hYdA6D7iydu564Fcs\nnLPd8ivKxrQ/Q5IkSZKmp02Bt1KyottTAtC1q2NvBXar1T0TuAbYjLJ68AGUrTbPAbbOzE90ukBm\nfoiyM8odwP7AW4DrgJ0y8+QO9f9WjeUYSnD7AeBVwPnAZpl5dYc2c4GXAGdQAtZDq/v4PLBFZt7X\noc13gW0ocd1uwPuAxyjZ43HtdNJJo19kRMSOwPeqt+cC+2Xmw9XiSMPArtUNERH/BPwLJci9Hthl\nlL1FNQ3NXzA8ln1y+97Mam7Co4/3dhx6qkH8XFbY7H29HsKkzb+mzCia9U/T/15afvXDT/d6CI3Y\naI2yxsfvbpvX45E0Y51VZ49eaRoYxL/LBsEgfi6zluiPVYfGaq+vXtOzf0eeve8/Tavf1eKg6Yzr\nm6vyXuDtmflw2/m//+HLzGsy8w3AscDGwPeqOdCSJEmSJP1d04Fra/Wrr2XmIx3Od/rm4l+A3wAb\nAe9ueDySJEmSpGmu6cB11ar8Q4dzQ8DTMqrVqlb/Wb3dveHxSJIkSZqGZgz17qX+03Tg2lql+O62\n460pw6t0aXdLVT6/4fFIkiRJkqa5prfDuYey7HP7agl3AssAG3Rp9+yqXKHh8UiSJEmahoam11pS\nmmJNZ1znVOW6bcd/W5U7RkSna+5alfc0PB5JkiRJ0jTXdOB6eVW+tO34d6tyTeArEbEsQEQsExGf\nB7auzv+i4fFIkiRJmoaGhnr3Uv9pOnC9oCq3aQWnlW8Cf6x+fhtwd0TcBtwPvL86/iRwQsPjkSRJ\nkiRNc40Grpn5c+CrlAzrJrXjjwG7sXDRpiUoz8K2rv8E8L7MvLLJ8UiSJEmSpr+mF2ciM9/e5fh1\nEbEh8D7g1ZQFmeYBVwMnZ+ZvO7WTJEmStPhxcSbVNR64jiQz7wWOrl6SJEmSJI1qkQaukiRJkjQW\nM0y4qqbpxZkkSZIkSWqUgaskSZIkqa9NaKpwRBwBDDc8FgAy0+dfJUmSpMWcizOpbqLPuB7R6CgW\nGsaFmyRJkiRJNf22OJNfq0iSJEkyMNBTTDRw3a7RUSw0JdOPJUmSJEnT14QC18y8pOFxSJIkSdLf\nzfAZV9W4qrAkSZIkqa8ZuEqSJEmS+lq/Lc4kSZIkSThTWHWNBq4RsSFwPWWRpTdk5vfG0GZn4DvA\nk8D6mXljk2OSJEmSJE1vTU8V3rsqbx1L0Fr5PvDnaix7j1JXkiRJ0mJgaGioZy/1n6YD122q8gdj\nbZCZw7X62zY8HkmSJEnSNNd04LphVV4zznbXtrWXJEmSJAlofnGm5avynnG2u68qV2hwLJIkSZKm\nKWfsqq7pjOu8qlx2nO2eVZV/a3AskiRJkqQB0HTG9XZK0LoZcMY42m1WlXc2PB5JkiRJ09AMU66q\naTrj+vOqfFNELD9izUpErADsVb29rOHxSJIkSZKmuaYD17OrcnngmxExa6TKEbE0cA4Ln439RsPj\nkSRJkjQNDQ317qX+02jgmpkXARdWb18N/CYi9mnPvkbEChHxVsrqw6+uDv8sMy9ocjySJEmSpOmv\n6WdcAfYGrgTWAtYDvgo8GRF3AQ8DywCr8tSg+SZgzykYiyRJkiRpmmt6qjCZeTewOVDPns4AVgOe\nX5X16/4IeGlm3tX0WCRJkiRNT0NDQz17qf9MRca1Fby+NiK2BN4MvBx4LmXF4QeBWygLOX09M6+c\nijFIkiRJkgbDlASuLZl5Ga4ULPWd4eHhXg+hIeUb0cG5H/jRN47q9RAaM0j38uIdD+v1EBox/9cn\nAoNzP/dddWKvhyBpCjU+NVTTmn8eJEmSJEl9zcBVkiRJktTXpnSqsCRJkiRNhIskqc6MqyRJkiSp\nr5lxlSRJktR3ZphwVY0ZV0mSJElSXzPjKkmSJKnvmHFVnYGrJEmSJDUgInYHtgFeCGwKLAOclZn7\ndKi7HrAr8H+A9YBVgfuAK4DPZ+YlHdq8DThthCEckJn/3qHdLOAwYC9gTeBB4BLgiMyc0+Vengsc\nDewArAjcDnwHOCoz7+/SZkvgY8AWwEzghmq8J2bmkyOMe1QGrpIkSZLUjI8BmwAPAX8BNgCGu9Q9\nBngjcD3wA+Deqv7OwM4RcXBmntil7XeA33Q4fnX7gYhYCvgpsGV1/nxK8LoH8LqI2C4zr2prsy5w\nGbBKda05wObAwcAOEbFVZt7b1mYX4FvAI8A51f3sDHwO2Kq61wkzcJUkSZLUd6bpdjgfAG7JzBsj\nYhvg4hHq/hj4ZGb+tn4wIramBJqfjYhzM/OODm2/k5lnjnFMh1KC1nMzc8/adc6hBKWnRcQLMrMe\nYJ9MCVoPysyTam2OBw4BjgUOqB1fFvgysAB4ZWb+ujr+L8BFwO4RsWdmnjPGMT+NizNJkiRJUgMy\n85LMvLF6O2LknZlfbQ9aq+OXAj8DlqQEnBMWEUPAeyhZ34+0Xed7wM+BjSjTm1tt1gW2B26qB62V\nIygZ1bdExNK147sDKwNnt4LW6hqPUbLQUAt0J2LKM64RsSSwNrACsGT1QUiSJElSV4v54kwL2sp2\n/xQRK1KeI70VuCgzb+1Qb13gecAfMvPPHc7/GHgFsB3lmVeAbavyJ+2VM/PhiPglJbDdgpJNpWoP\ncEGHa1wKzAdeFhFLZGa3exrRlAWuEfF/KGnpVwBLUb5xGAae0VbvvZSHl2/LzCOnajySJEmS1O8i\n4h+AVwHzKEFfJwe3vX8iIr4CfKDKcrasX5XZpZ8/VuV642hzAyVwXY+FgWvXNpn5RETcBGwIrAP8\noUu/I2p8qnBEPDMiTqdE79tTvgUY6fuSO4F3Ah+PiLWaHo8kSZIkTQfVQkpnUaYJH5mZD7RVmQu8\nDwhgaWB1yqJHfwLezdNXHF6uKtv7oe348g20GR6lzVBbm3GZiozrKcC+1c8PAD+iBK9v6FL/e1W9\n5YBdgH+bgjFJkiRJmkam59pMExcRzwC+Rnmu9ezMPL69TvXYZT0L+yhwXkRcAfwWeFNEfDozr10U\nY16UGs24RsRWwH7V258Aa2fmmykfQEfVHOfW/OmtmxyPJEmSJPW7Kmj9OmWRo3OAt4ynfWb+hZIw\nhKfGVK0M6HJ01jpe35d1om2GxtlmXJqeKvzOqrwV2K3bxrQdXFOVGzc8HkmSJEnT0IyhoZ69FqWI\nWAL4BrAnZZrw3pn55AS6+mtV1lf7ndO6TJc2rWdb68+mttqsT2ed2rSeW31am4h4JmWx3gWUqc4T\n0nTg2oruz8jMeeNo11oBa/WGxyNJkiRJfanageVcSqb1q5m5T9t+quOxeVX+PTistua5GVi/y3pC\nr6nKi2rHWnvPbl9tp1Mf77OArSgLR11RO3VhVe7Q4RpbA7OAyya6ojA0H7i2As/rxtluflXObHAs\nkiRJkqapGT18LQrVQkzfBnYGvgK8YwxtXtLh2IyI+H+U7Wnu5ulb0pxalZ+pB6IRsQvwcuD6zPxZ\n63hmzqV67BM4sK2voygZ3a9l5vza8fMoGd+9IuLFtWvMBP61envKaPc3kqYXZ3qiKsf7ebdWl+q2\nCpUkSZIk9bWIeD3w+urtalW5ZUScUf18d2Z+uPr5VErG86/AbcAREU+b0XtxPagEroqI/wWupcxa\nXY6SAd2YkgV9c2Y+3NbHCcCOlKzulRFxEbAmsEfVplPA/F7gMuALEfEqyvThzYFXUqYFH16vnJkP\nRcT+lAD2kog4G7iPEpQHcG5mfrPDdcas6cD1Dsomt88fZ7vWNwc3NzscSZIkSVpkNgXeStkahqpc\nm7J/KZRta1qB61rV+ZWAf+nQ1zDwJFAPXI8DXgpsC6xISRzeDHwROCEz/9TeSWb+LSK2Bw4D3gR8\ngJIwPB84IjPndGgzt8ruHk2Z/vtaSnD9eeCoDtv0kJnfjYhtKEHtbpTZtDcAhwBf6HB/49J04PpL\nSuC6KwtTwiOq5km/sXrbbYNdSZIkSYuR6bgdTmYeRZlOO5a6206g/4+Me1Cl3XzgiOo11jZ/YQzT\nl9vaXAa8bnyjG5ump3CfXZUvjIiDRqscETOAL7FwqvCZDY9HkiRJkjTNNZpxzcwLIuJiSur6cxGx\nJnA8C1PlwN+XRN6aEvG/ojp8Tmb+psnxSJIkSZqeFvW2NOpvTU8VBtgLuJwyj/uDlDnNrTnQQxFx\nE7AqZUnkluuAd03BWCRJkiRJ01zjqz1n5t2UB4a/W7vGCrUq/8BTg9ZvAS/PzIeaHoskSZIkafqb\niowrmXkv8IaIeBGwD2U68FqU5ZofBv5CWR3rzMy8airGIEmSJGn6cqaw6qYkcG3JzF8Dv57Ka0iS\nJEmSBtuUBq6SJEmSNBEzzLiqpvFnXCVJkiRJapIZV0mSJEl9x+1wVNdo4BoRp9O2Z+t4ZeY7GhqO\nJEmSJGkANJ1x3XeS7YcBA1dJkiRJ0t/121Rh5wNIkiRJcjscPUXTges6Y6gzA1gZ2Bw4ANgA+Abw\nMSY5zViSJEmSNHgaDVwz809jrDoXuCoiTgX+HXgb8HBmvrvJ8UiSJEmantwOR3U93Q4nMxcA7wIS\neGdEvKaX45EkSZIk9Z+e7+OamY8DZ1Kebz2gx8ORJEmSJPWZflmc6Y9V+eKejkKSJElSXxhy3VbV\n9EvgukxVrtREZxHxZPuxzOx5drnfRcQQcD3wPOD5mXlnj4ckSZIkSX0TuO5alX9tuN+7gSc6nYiI\nmcDrgNcAL6WsiLwEcCdwGXBKZv5stAtExKuBgymrJC8L3Ap8H/hEZt41kUFHxDOB7YHXAi8D1gNm\nAfcAVwGnZeZ3R2h/BvDWUS7zw8zcqX4gM4cj4hjgLOBInLotSZKkHnFxJtX1NAsZEbMj4jhKgAbw\niwa7HwY2y8w1MnONDue/D5wLvAPYmPK7eAx4DvBG4OKI+NxIF4iIw4GfUALgFYD5wNrA+4HrImLj\nCY79FOCHwIHAi4ClgEeAVYGdgG9HxDerAHckDwN3dHnd26XN2ZTFsvaLiJjg+CVJkiSpMY1mXCPi\ndMa2F+uSlABxM2Dp6tiTwPFNjmcUz6QEaF8Gvp+ZCRAR6wCfBPYADo6IzMxT2htHxGuBYyj3ezxw\nVGbOi4iNgK8DLwS+GxEbZebfJjC2W4GvAN/OzGura64OfJQS0O4O3AT83xH6OS4zjx7Phaus6+mU\n38H7gfeNc+ySJEnSpJlxVV3TU4X3nWC7vwHvzcyrmxzMKD4KXJGZTwm0M3MusGdErARsB3yIkgFt\n94mq/HZmfqTW/ncRsRPwe8r043cBXxzn2E4G3lVtF1Qf2+3AQRExm7L37YERcURmPjrO/kfzn5TA\nde+I+HBmzm+4f0mSJEkas14vWPQn4FTghZl52qK8cGZe3h60tjmzKteKiOXrJ6opwJtQsq2f7dD3\nrcA3qrdvnsDYrm4PWtucUZWzgA3H2/8Yrn8LcAWwPAufP5YkSZKknmg647rOGOs9Btzf55m8+jOg\nz2g7t21VPpCZV3Zp/1/A/sBmETE7M+dN0dhG+vJhMhMsLge2AF5NWaxJkiRJWmSGhpwrrIUaDVwz\n809N9tdj21TlnZl5T9u5jary9yO0/11VDgEbAL+agrEtoDyn282bI+IdwOqUhZp+B3wXODUzHxrl\nGldV5SsmM1BJkiRJmqxGpwpHxDYRsXVEbNpkv4taRDwHeE/19owOVVavyttG6Ob22s+rNTAsACJi\nGeCw6u35owSg6wLPBh6kbNWzJfBpyorHm4xyqeurcu32qdKSJEnSVJsx1LuX+k/Tz7heXL3e1HC/\ni0y1xcxZwGzgz5RFitrNrsqRpjo/UpVDwDKNDbA8E/wc4AEWBrDtfkUJvNfMzJmZuTKwUnXsfmBN\n4McRseII17mv9vOqkx61JEmSJE1Q08+4zqcsGHRtw/0uSicCW1Oew917DFNqF5mIOAzYm7J10P6Z\neXOnepl5YodjDwBfioirKAsvrQ58EDi8y+Xqz9GuwshTkrua2fSfsB4bnPsZrK8SZy0xOPez7QYr\n9XoIjRmke5n/66f9tTqtDdr9DIrB+W/MYBmUz+XRx3s9Amlymv6/4u2UBZp6vVrxhETEJ4B3A48D\nb87My7tUbS20NGuE7lr70w5Tni9tXeOODnWHgYMz85sjjO3dlC14hoEPZuZ5I1y7q8z8TUScDbwV\n2InugaskSZLUM67NpLqmA9dLKIHri4GvN9z3lIqIwylTb1vZzPNHqN56tnWNEerUz9Wfd12VEny2\nmznC2Pah7O06DByZmf82wnXH4kpK4Lr2CHXq04jvnuiFBuXbvda3rYNyP8PDI+0ENX20Mq3zFwzG\n/QBcceO9o1fqc61M68Vz2te1m75eu/eRvR5CI1qZ1lkvOqjHI2nGfVcNRuZ40P4bMyj8XKT+0nTg\neirwNmDfiPh0ZnbKLvadiDgEOIaFmc+vjtKktXDRhhEx1GU/2NbKw8PUVh/OzHFloyNiD+B0ytzO\n4zLzmPG072Is31/VA9e7GrimJEmSNGYzTLmqptEpvZn5P8DHgOWBCyPiBU32PxUi4gDgeEqAeVhm\nnjSGZhdX5XLAZl3q/HNVXjnR/WojYifKQlEzgFMy8yMT6aeDzavyphHq/GOrTmbe39B1JUmSJGnc\nJpxxjYh9KcHexZl5S+3Y7cAFwA7ANRHxc+DnwK2MvAovAJl55kTHNF7VeE+i3MfRmfnZsbTLzN9H\nxG+BTYEPA3u09bsGC1dWPmuCY9seOBd4BnBGZh44kX469LspsFf19ocjVH1pVV7axHUlSZKk8XBb\nGtVNZqrw6ZSA7w3ALW3HWn/MZgDbVK+xGAYWSeAaEbsB/1G9/WxmHjXOLj5KCfx2i4hPA8dk5sMR\nsRHwNcoWODcCX57A2LYCvgMsCXwD2G8cbfcBdgS+CvyyWk2YiFgOeCPwKWAJ4E7guBG62rIq/3u8\n45ckSZKkJk32GddO34NM5ruRRfm9ymdZOFV634h4W5d6w8Cu7SsMZ+aPI+LjlGdjPwx8MCLmAc+q\nqtwN7JKZCyYwtmNYuGLx9sDtEdFtbO2rEc+gZID3AIiIhyirJK9Qq/Nn4A2Z2XHllIhYizIF+n5g\npEWqJEmSJGnKTTZwbV+U6B0N9zeVhmrXW3WUukt0OpiZx0bEFcAHKFNrn0XJsv4A+ERmTnQ13vrY\nRtsIsX014ouBj1MypusDK1Oyv3cB1wHfA07LzHl09/dpzpn56DjGLUmSJDXCtZlU1+iqwpl5RpP9\nTaXMHGkrmPH0cyFwYRN91frcdhJtbwaOnWj7iJhBWRl6AfCFifYjSZIkSU1pejucfuP3NOO3F7Ae\ncGpm3tDrwUiSJGnxNMN/yqtmkAPXIeCm1rOh490/dXEUEUPA4cDDwJG9HY0kSZIkFYMauN7Jon1e\ndiBk5jCwca/HIUmSJEl1Tawq/P6IeH0TgwHIzMku8ERmrt7EWCRJkiT1hoszqa6JjOt2DfTRMszk\nVyaWJEmSJA2Qfpsq7PcqkiRJkphhZKCaJgLXE4FrGugHfC5VkiRJ0jQVEbsD2wAvBDYFlgHOysx9\nRmizJfAxYAtgJnADcBpwYmY+2aXNvsCBwIbAE5R47LjM/GGX+rOAwyg7iKwJPAhcAhyRmXO6s21C\n5QAAIABJREFUtHkucDSwA7AicDvwHeCozLy/qXsZq8kGrsPAhZn5vUn2I0mSJEl/N2N6PuT6MWAT\n4CHgL8AGjJCci4hdgG8BjwDnAPcCOwOfA7YC3tihzXHAocAtwJeApSgB6fcj4qDMPKmt/lLAT4Et\ngauB8ynB6x7A6yJiu8y8qq3NusBlwCqUYHUOsDlwMLBDRGyVmfdO9l7Gwy1iJEmSJKkZHwDWy8zl\ngANGqhgRywJfBhYAr8zM/TPz/1KytZcDu0fEnm1ttqQErX8ENsnMD2bm+4AXUwLF4yLiH9oudSgl\naD03MzfPzP+XmW8GdgeWBk6rtsWsO5kStB6Umbtm5kcz81WUIHR94NjJ3st4GbhKkiRJUgMy85LM\nvLF6O1rKeHdgZeDszPx1rY/HKJlbeHrw+56qPDYzH6i1+TNwEiX7+vbW8SogfQ8l6/uRtrF+D/g5\nsBFlenOrzbrA9sBN7dlb4AhKRvUtEbH0JO9lXAxcJUmSJPWdoaHevRaR1u4sF3Q4dykwH3hZRCzZ\n1ma4S5sfV+W2tWPrAs8Dsgpuu7Wp7xTTav+T9sqZ+TDwS2A25TnW+rjoMq76vSzR4fyYGLhKkiRJ\n0qK3flVm+4nMfAK4ibIm0ToAETEbWAN4ODPv7NDfH6syxnKNtjbrjaPNDeNp0+leJqLftsORJEmS\npOm6ONN4LEfJnj7Q5fwDlOnGy9Xqt453qw+wfNs1FlWbsdzL8l3Oj2qyGdch3HtVkiRJkjSFJpNx\nbaV5O6WpJUmSJEndtWdU27WOt/ZMfaDt+Gj1F3Wb8dzLuE0445qZf6pe8yfahyRJkiR1shgszvSH\nqly//UREPBNYm7K9zFyAzJwH3AYsExGrdeiv9cxp/TnTOa0uu4xhpDZPG9cIbcZ1LxPh4kySJEmS\ntOhdWJU7dDi3NTALuCwzF7S1GerS5jVVeVHrQLU1z83A+hGx1ljaABdX5fbt+7tGxLOArYB5wBWT\nvJdxMXCVJEmS1Hdm9PC1iJwH/BXYKyJe3DoYETOBf63entLW5tSqPDwilq+1WQs4EHgUOL1Lm8/U\nA9GI2AV4OXB9Zv6sdTwz51K2wlm76rPuKGBp4GttM28nci/j4qrCkiRJktSAiHg98PrqbWs675YR\ncUb1892Z+WGAzHwoIvanBH2XRMTZwH3AzpSpvedm5jfr/Wfm5RFxAnAocG1EfAtYEtiTsmLvQZl5\nc9uwTgB2BHYHroyIi4A1gT0omdN3dLiV9wKXAV+IiFdRpg9vDrySMi348LZxjftexsuMqyRJkqS+\nMzQ01LPXJGwKvBXYB9ieskXM2tWxtwK71Stn5neBbYBLq3PvAx4DDgH26nSBzPwQ8HbgDmB/4C3A\ndcBOmXlyh/p/q8ZyDCW4/QDwKuB8YLPMvLpDm7nAS4AzKAHrodV9fB7YIjPv69Bm3PcyHm5loyk1\nf8HwcK/H0ISZ1dyERx/v7TiaMjwYHwuzlih/hc1fMBj3A3DFjff2egiTtu0GKwFw8Zx7ejyS5rx2\n7yN7PYRGzP/1iQDMetFBPR5JM+676sReD6ERg/bfmEExiJ/LrCWm18aoZ1x9c8/+A/+2zdacVr+r\nxYEZV0mSJElSX/MZV0mSJEl9x5Sn6sy4SpIkSZL6mhlXSZIkSX1nxvR6JFdTzIyrJEmSJKmvmXGV\nJEmS1HfMt6rOjKskSZIkqa8ZuEqSJEmS+ppThSVJkiT1HddmUp0ZV0mSJElSXzPjKkmSJKnvDJly\nVY0ZV0mSJElSXzNwlSRJkiT1NacKS5IkSeo7ZthU558HSZIkSVJfM+MqSZIkqe+4OJPqzLhKkiRJ\nkvqaGVdJkiRJfcd8q+rMuEqSJEmS+pqBqyRJkiSprzlVWJIkSVLfcXEm1Rm4SouhQfsPwSDdz7qr\nLtPrITRmkO7l+v/6TK+H0KhBux9J0uAzcJUkSZLUd3ymUXX+eZAkSZIk9TUDV0mSJElSX3OqsCRJ\nkqS+M0hrWGjyzLhKkiRJkvqaGVdJkiRJfcd8q+rMuEqSJEmS+poZV0mSJEl9x0dcVWfGVZIkSZLU\n1wxcJUmSJEl9zanCkiRJkvrODJdnUo0ZV0mSJElSXzPjKkmSJKnvuDiT6sy4SpIkSZL6moGrJEmS\nJKmvOVVYkiRJUt8ZcnEm1ZhxlSRJkiT1NTOukiRJkvqOizOpzoyrJEmSJKmvmXGVJEmS1Hdm+Iyr\nasy4SpIkSZL6mhlXSZIkSWpARLwNOG2Uak9m5jOr+msBc0eoe05mvqnLtfYFDgQ2BJ4ArgGOy8wf\ndqk/CzgM2AtYE3gQuAQ4IjPndGnzXOBoYAdgReB24DvAUZl5/4h32TADV0mSJEl9Z5ouznQNcGSX\nc1sD2wE/6nDuN5SAsN3/duooIo4DDgVuAb4ELEUJSL8fEQdl5klt9ZcCfgpsCVwNnE8JXvcAXhcR\n22XmVW1t1gUuA1apxjYH2Bw4GNghIrbKzHu73GvjDFwlSZIkqQGZ+Vvgt53ORcTl1Y9f6nD6N5l5\n9FiuERFbUoLWPwKbZeYD1fHPAr8CjouIH2Tmn2vNDqUEredm5p61vs6hBKWnRcQLMnO41uZkStD6\nlEA4Io4HDgGOBQ4Yy5ib4DOukiRJkvrO0FDvXk2LiBdQspV/ATpO5R2H91Tlsa2gFaAKVE+iZF/f\nXrv2UNVmGPhIvaPM/B7wc2AjYJtam3WB7YGb2rO3wBHAI8BbImLpSd7LmBm4SpIkSdLUeldV/kdb\nVrPlORHx7oj4aFW+YIS+tqMEoRd0OPfjqty2dmxd4HlAtmVh29tsVzvWav+T9sqZ+TDwS2A2sMUI\n42yUU4UlSZIkaYpUiyK9BXgc+EqXattXr3q7S4B9M/OW2rHZwBrAQ5l5Z4d+/tiqWju2flVml2u3\n2qw3jjY3VONdD7ioS51GmXGVJEmS1HeGevi/hr0RWA64IDNvbTs3j7Jq74uA5avXNsDFwCuBC9um\n4y5XlQ/QWev48j1oM6XMuEqSJEnS1GlNE/739hOZeTdPX4X45xHxz8AvKM/FvhP4wlQOcDow4ypJ\nkiSp78wY6t2rKRGxMfAyyrY1nbbB6Sgzn2DhtOJX1E61Mp3L0VnreH2P1UXVZkoZuEqSJEnS1Bht\nUaaR/LUqZ7cOZOY84DZgmYhYrUOb1nOq9WdT51Rl0NlIbdans05tppSBqyRJkqS+M92fcY2ImcA+\nlEWZ/mMCXbRW7J3bdvxCYAjYoUOb11Tl3xdMyswbgZuB9SNirbG0oTxjC7B9tZ3O30XEs4CtKM/n\nXjHyLTTHwFWSJEmSmrcHZfGiH3dYlAmAiHhRe2BYHX8VcAhl25uvt50+tSoPj4jla23WAg4EHgVO\n79LmM/XrRcQuwMuB6zPzZ63jmTmXshXO2lWfdUcBSwNfy8z5ne5rKrg4kyRJkiQ1rzVN+Esj1DkB\neH5EXAa0gttNKPuoDgMfz8ynZDUz8/KIOAE4FLg2Ir4FLAnsSQmUD8rMmztcZ0dgd+DKiLgIWJMS\nXM8D3tFhbO8FLgO+UAXScyiLRb0S+ANw+Ih33zAzrpIkSZL6ztBQ716TFREbUqbTjrYo05nANcBm\nlNWDDwDWBc4Bts7MT3RqlJkfAt4O3AHsT9kn9jpgp8w8uUP9v1H2XT2GEtx+AHgVcD6wWWZe3aHN\nXOAlwBmUgPVQSgb288AWmXnfSL+DpjW+SZFUN3/B8HgfQu9LM6u5CY8+3ttx6KkG8XO568HHej2E\nSVtzxaUAuPne6X8vLY8/8WSvh9CIdVaZBcDcuxfZzK4ptcYKs3o9hEYM4t9lg2AQP5dZSzQRki06\nF835a8/+HbndBitPq9/V4sCpwpIkSZL6TlOLJGkwOFVYkiRJktTXDFwlSZIkSX3NqcKSJEmS+s4M\nZwqrZiAD14h42ioamWl2eRTVnk7XA88Dnp+Zd/Z4SJIkSZI0mIFrzd3AE51ORMRM4HXAa4CXAusA\nSwB3UvYrOqW+CW+H9pcAW49y/ZMy86DxDjoilgD2q8b1QmA1YGXKZsI3AT8FTszMP4/Sz0uAD1Xj\nXBG4C/gv4FOZeWN7/cwcjohjgLOAIynLcUuSJEmLnIszqW6Qs5DDlD2J1sjMNTqc/z5wLmWz3Y0p\nv4vHgOcAbwQujojPjeE6D1D2T+r0emCCY18JOBl4G7ApJWh9EFgaeAFlD6XfRcQu3TqIiH2By6t7\nWRV4hHJv+wG/iYhtuzQ9G0hgv4iICY5fkiRJkhozyIHraJ5JCdA+DGyYmUtn5rLAepSAFuDgiBgt\n63hwKzju8PrYBMc2H/gc8AbgOZm5ZGauDMwEXgvMAWYBX4+I1dobR8QmwJeBZwBfB56dmStSNgz+\nKTAb+FZErNzeNjOHgdMpv5/3T3D8kiRJ0qQMDfXupf6zOAeuH6UErMdnZrYOZubczNwTuKg69KFF\nPbDMfCAzP5iZ383MO2rHH8/MC4Adq0OzgZ06dHE0JfC8Gtg3M++p2t8M7ArcAiwPHNZlCP9ZlXtH\nxGDs7i5JkiRp2lpsA9fMvLzKLnZzZlWuFRHLL4oxjVVmzgXur94uUz9XjfW11dsT2u8xM+cBp1Zv\n39Sl/1uAKyjB7a4NDVuSJEmSJmSxDVzH4N7az88Yod4in0wQERtQgsphSla17uWUbOsw8JMuXfxX\nVa4WERt2qXN5Vb56EkOVJEmSJmSohy/1n0FfVXgytqnKO1tTbbv4UER8krKA0gPAb4HzgNMz87Gm\nBhMRM4BnU1YIPpYSmJ6Vmb9oq7pRVd6Rmfd16e73bfV/36HOVVX5iomNWJIkSZKaYca1g4h4DvCe\n6u0Zo1TfiJL9fIiy5cx2lBWBr4qI5zUwlq9U+9I+DtwKfIOy+vG7M/OtHZqsXpW3deszM+dTguwh\nylY7nVxflWv321RpSZIkDb4ZQ0M9e6n/mHFtExHPpOxjOhv4M/DJLlUvpqzc+9PMvKtquwplu5kj\nKNvW/CgiXpSZCyYxpPspW+ssCaxACTbXAV4ZEedn5r1t9WdX5fxR+n0EWI62Z2Rr6tnaVVn4TO24\nzBywP2GDdj+DYpA+lzVXXKrXQ2jMIN3LoFlnFdfd60eD9HfZIBmUz+XRx3s9AmlyzLg+3YmU6biP\nAXtn5kOdKmXmUZl5VitorY7dnZmfAnarDm1M2Yt1wjLzQ9XWOitTMrtvogSyewOXRcTSk+l/BPWA\neJUpuoYkSZIkjWpAvkNqRkR8Ang3ZVrumzPz8lGadJSZP4qISykB8E6UzGzrGnd0aDJM2Q/2m6P0\n+xBwTkT8HPgdEMAHgE/Uqs2rytG+Tm8FvA+PUm9SBuXbvda3rYNyP4NiED+Xux5s7NH4nmllWm++\nd/rfS8vjTzzZ6yE0opVpnXv3aJNypoc1VhiMzPEg/l02CPxces8Ju6oz41qJiMMp+5o+CeyfmedP\nsssrq3LttuOrUjKY9deqwMyxdpyZtwHfrt5u13b61qpco1v7am/W5aq3t3eptmLt57vHOjZJkiRJ\napoZVyAiDgGOYWHm86sNdNvxS6LMbOrLgtbiS+0B6u+qcrWIWLHDM7CwcOXh4Vr9dvXA9a4udSRJ\nkqSpYcpVNYt9xjUiDgCOpwRxh2XmSQ11vXlV3tRQf+1amdz2oPIXlKnOQ3Tfg/Wfq/K2zJzTpc4/\nVuVNmTmhhZkkSZIkqQmLdeAaEfsCJ1GC1qMz87MN9bsDC/c//eEE2j9jlPPrAa+v3l5YP5eZD9au\neWhEDLW1nc3CrX6+McJlXlqVl45lzJIkSVKThnr4P/WfxXaqcETsBvxH9fazmXnUONoeBqxL2Tbn\n6sycVx1fBXg7cGRVdQ5w2gSGd2JELADOBq7JzEer/penBKyfoDwTeydlFeR2RwCvpQSfZ0TEoZl5\nT0SsSVko6nmU7W4+PcIYtqzK/57A+CVJkiSpMYtt4Ap8loUZ530j4m1d6g0Du7atMLwUZb/W/QAi\n4sGq3nK1OtcCO09wD9dZlKzoQcBwRDxAmfpb7/8m4A2dpvFm5rURsT/wFWAfYJ9qjMtWVR4GdsvM\nezpdPCLWAjaj7N062UWqJEmSJGlSFufAdYgSbEJZ1XckS7S9P5fyu3sZJfO6UlXnNuA3wHnAWRMM\nWgE+Bfwe2BZ4PvDsqv/bKQHxd4GvZmbX/Qwy88yIuB74MGXa8orAzcBPgU9m5twRrv+mqjyrle2V\nJEmSFqUhZ+yqZiD/OETEk5SgdO3MvLnX45lOImIGJWheC/jHzLxhMv3NXzA8PHqt/udebv1pED8X\n93HtT+7j2p/cx1VTaRA/l1lLTK9Q8Mob7+/ZvyM3X3f5afW7WhwMesbVP3DjtxewHnDqZINWSZIk\naaL8h7zqBjlwHQJuigig0f1TB1a1AvHhlGdgj+ztaCRJkiSpGNTA9U4WPr+qMcrMYWDjXo9DkiRJ\nkuoGMnDNzNV7PQZJkiRJk+BcYdU4fVaSJEmS1NcGMuMqSZIkaXobMuWqGjOukiRJkqS+ZsZVkiRJ\nUt+ZXrvOaqqZcZUkSZIk9TUDV0mSJElSX3OqsCRJkqS+40xh1ZlxlSRJkiT1NTOukiRJkvqPKVfV\nmHGVJEmSJPU1A1dJkiRJUl9zqrAkSZKkvjPkXGHVmHGVJEmSJPU1M66SJEmS+s6QCVfVmHGVJEmS\nJPU1M66SJEmS+s50TbhGxJ+ANbucvjMzV+/QZkvgY8AWwEzgBuA04MTMfLLLdfYFDgQ2BJ4ArgGO\ny8wfdqk/CzgM2Ksa34PAJcARmTmnS5vnAkcDOwArArcD3wGOysz7u9zjlDBwlSRJkqRm3Q98vsPx\nh9sPRMQuwLeAR4BzgHuBnYHPAVsBb+zQ5jjgUOAW4EvAUpSA9PsRcVBmntRWfyngp8CWwNXA+ZTg\ndQ/gdRGxXWZe1dZmXeAyYBVKsDoH2Bw4GNghIrbKzHvH8stogoGrJEmSJDXr/sw8erRKEbEs8GVg\nAfDKzPx1dfxfgIuA3SNiz8w8p9ZmS0rQ+kdgs8x8oDr+WeBXwHER8YPM/HPtUodSgtZzM3PPWl/n\nUILS0yLiBZk5XGtzMiVofUogHBHHA4cAxwIHjPk3Mkk+4ypJkiSp/wz18LXo7A6sDJzdCloBMvMx\nytRheHpw+J6qPLYVtFZt/gycRMm+vr11PCKGqjbDwEfqHWXm94CfAxsB29TarAtsD9zUnr0FjqBk\nh98SEUuP52Ynw8BVkiRJkpo1MyLeEhEfjYiDI+KVEdEp9tquKi/ocO5SYD7wsohYsq3NcJc2P67K\nbWvH1gWeB2RbFra9zXa1Y632P2mvnJkPA78EZlOeyV0kDFwlSZIk9Z2hHv5vkoaB1YAzgX+lPKt6\nEXBDRGzdVnf9qsz2TjLzCeAmyuOd6wBExGxgDeDhzLyzw7X/WJUxlmu0tVlvHG1u6NBmShm4SpIk\nSVJzTqdkL58NLA28APh3YC3gxxGxSa3ucpRA9wE6e4AyeXm5Wv3W8W71AZZvu8aiaDOlXJxJkiRJ\nkhrSYVGm64EDIuJh4IPAkcCui3pc050ZV0mSJEl9Z2iod68pcmpVvqJ2rD2j2q51vLVn6gNtx0er\nvyjbTCkDV0mSJEmaen+tytm1Y3+oyvXb6hIRzwTWpmyVMxcgM+cBtwHLRMRqHa7Reua0/mzqnFaX\nXcY1UpunjWuENlPKwFWSJElS3xnA3XBaK/DOrR27sCp36FB/a2AWcFlmLmhrM9SlzWuq8qLWgcy8\nEbgZWD8i1hpLG+Diqty+2k7n7yLiWcBWwDzgig79TQkDV0mSJElqQERsUK382358LeCL1duv106d\nR8nE7hURL67Vn0lZkRjglLbuWlOOD4+I5Wtt1gIOBB6lLBDVqc1n6oFoROwCvBy4PjN/1jqemXMp\nW+GsXfVZdxRl0amvZeb89nudKi7OJEmSJKn/TGHqcwrtBXwwIn5GyXI+RNlH9XXAUsAPgeNalTPz\noYjYnxLAXhIRZwP3ATtTpvaem5nfrF8gMy+PiBOAQ4FrI+JbwJLAnpRVfg/KzJvbxnUCsCOwO3Bl\nRFwErAnsQcmcvqPDvbwXuAz4QkS8ijJ9eHPglZQpzoeP+7czCWZcJUmSJKkZFwHfpwSrbwIOoSzG\ndCnw1szcKTMfrzfIzO8C21R1dgPeBzxWtd2r00Uy80PA24E7gP2BtwDXATtl5skd6v8N2B44hhLc\nfgB4FXA+sFlmXt2hzVzgJcAZlID1UEoG9vPAFpl53xh/J42Ynt9jaNqYv2B4uNdjaMLMam7Co4+P\nXE+L1iB+Lnc9+FivhzBpa664FAA33zv976Xl8See7PUQGrHOKrMAmHv3IpvZNaXWWGFWr4fQiEH8\nu2wQDOLnMmuJKVwvdwpc95eHevbvyBc891nT6ne1OHCqsCRJkqS+M2SOTTVOFZYkSZIk9TUzrpIk\nSZL6zvSa2KypZsZVkiRJktTXDFwlSZIkSX3NqcKSJEmS+o4zhVVnxlWSJEmS1NfMuEqSJEnqP6Zc\nVWPgKkl9ZMXZS/R6CI0ZpHu5b96CXg+hUUs8wwlXkqTpxcBVkiRJUt8ZMuWqGr9ylSRJkiT1NQNX\nSZIkSVJfc6qwJEmSpL4z5Exh1ZhxlSRJkiT1NTOukiRJkvqOCVfVmXGVJEmSJPU1A1dJkiRJUl9z\nqrAkSZKk/uNcYdWYcZUkSdL/b+++wySrysSPf1sYclaiiAPCa0YFEVdAAR2CgqgrmMWAuugCgll+\nSNhlTchifNaEovsDUUAQQVGRICCYRkTSK0FABYkiDHGY3j/OKfpOTVV193T3dHXN9+NTz62bzj33\nVjH2W+8JktTXzLhKkiRJ6jtDplzVYMZVkiRJktTXzLhKkiRJ6jtDJlzVYMZVkiRJktTXDFwlSZIk\nSX3NpsKSJEmS+o4thdVkxlWSJEmS1NfMuEqSJEnqP6Zc1WDGVZIkSZLU1wxcJUmSJEl9zabCkiRJ\nkvrOkG2F1WDGVZIkSZLU18y4SpIkSeo7QyZc1WDGVZIkSZLU18y4SpIkSeo7JlzVZMZVkiRJktTX\nDFwlSZIkSX3NpsKSJEmS+o9thdVgxlWSJEmS1NfMuEqSJEnqO0OmXNVgxlWSJEmS1NcMXCVJkiRJ\nfc2mwpIkSZL6zpAthdVgxlWSJEmS1NfMuEqSJEnqOyZc1WTgKkmSJEmTICLWAl4FvAx4JrAB8BBw\nGfAN4BuZOdw4fjZwXY8iT8zM13W51t7Ae4CnAo8Ac4GjMvOMLsevCHwYeC2wEfBP4Fzg0My8qss5\nGwJHALsAawE3A6cCh2fmP3rUe9IZuEqSJEnqOzO0j+tewJeAvwHnADcC61GC2a8BuwJ7djjv95SA\nsN0fO10kIo4CDgJuAr4CLE8JSE+PiP0y84ttxy8P/BR4AfBr4BRK8Lon8LKI2DEzf9V2zpOAi4C1\na92uArYGDgB2iYhtMvPOXg9jMhm4SpIkSdLkuBrYvT3rGREfBX4F/GtEvCozT2k77/eZecRYLhAR\nL6AErdcAW2Xm3XX7p4HfAkdFxA8z84bGaQdRgtbvZeZrGmWdSAlKj42IZzazwZQAfG1goUA4Ij4D\nHAgcCew7ljpPBgdnkiRJkqRJkJnndGqqm5l/B/6nrr5ogpf5t7o8shW01mvcAHyRkn19a2t7RAzV\nc4aBD7bV6wfAL4CnNetVs61zgOvbs7fAocB9wBsjYqUJ3suYDWzGNSIWtG/LzBkTqEfE9sDPgRsy\nc+MpKP/fgX2AAFaom2cDjwcuBE7JzFdP9nUlSZKksZmZbYV7mN+2bHp8RLwLeCxwB3BRZl7WpZwd\nKUHojzvs+xFwCLADcFjd9iTgCcDVbVnY5jnb1XLPrdt2qMuftB+cmfdGxIWUwPb5lJhlyg1s4Npw\nG6Wz8iIiYnVgb+B5wObAupROx/OAPwFnAp/PzNt7XSAiXkJp6701sBrwV+B04L8y89YJ1n949EPG\nJyIOBv6jrt4P3FLfP5KZv4yInwKviojnZ+bFk319SZIkaWkSEcsCb66rnQLOOfXVPOdcYO/MvKmx\nbWXKgE/31Cxuu2tahza2Pbkus0v1WudsNo5z/lTruxlLKHCdMRnIxTRMafe9QWZu0GH/U4FjgNcD\nTwdWp4yutSqwJeXXiqsi4vndLlCDwJ9QRg5bkxIIbgzsD1wWEU9fzLrPo7SRv3Yxz+/lgLo8MDNX\nbj2fzPxr3d5qX/+pKbi2JEmSNKqhoel7TYFPUOKNMzLzp43t8yh/e28BrFFfL6IM7LQ9cHZbc9zV\n6/JuOmttX2MazplSgx64juZ24OOU0b3WzcwVMvOxwMqUUblupmRgT64jcS0kIl5KyVwOA0cBa2Tm\nmsAzKCODrQ2cFhHLjbdimfnrzHxqZs4Z/eixi4h1gMfVOn+1y7UvpIwatm1EPHcyry9JkiQtTSJi\nf8rgSFcCb2ruy8zbMvOwzPx9Zv6zvn4B7ARcAmxK6d631FuqA9fMvCYzD87Ms5rNgTPzgcz8LiNf\nrPWAF3Yo4r/q8vuZ+cHMnFfPvwLYHbgX2AR455TdxPit2HqTmff1OO74unzH1FZHkiRJGkx1XJlj\ngMuBHcY692lmPkKZPgdK/9OWVqZzdTprbW9eZ0mdM6WW6sB1DH5Tl0PAKs0dtQnw5pTM5afbT6zN\nbk+oq28Y74UjYvuIWBAR13fYd27d9+aIWDEiDouIqyPi/oi4NSJOiIhNO5UHtMobqmW0Xoe2Xea7\ndfnaiFgBSZIkaQkamsbXZIiI9wKfAy6jBK3jHfumlVhbubWhJsr+BqwSEet1OKfVT7XZN/WqVpW6\nXKfXOU+ms07nTCkD195eUJfzgd+17WuNtHV3Zl7S5fyz6nKr2pF6cfQanGl1ygjAH6OMFPYIpRnw\na4CLI2KTxrEPAn+nDFbVckvjdU+z4MxM4C5KwL71YtZdkiRJWupExIeAo4G5lKC152CvXbTG2bmu\nbfvZlPh6lw7n7FqXjw6YlJnXAjcCT46I2WM5h9LHFmBOnU7nURGxKrANpX/uEhvI1cCXkqKPAAAW\n10lEQVS1TUQsGxEbRsQ+wLcogeMnOgwd/bS6vLJHcVfU5RDwlMmtKQCHU4LXnYGVM3MVSpPmv1D6\n5n68dWBm/jIz16eMoAww3BiUaYPMPLpD+b+qde/UTFqSJEmaMjN1cKaIOITyd/hvgBdn5p09jt2i\nPTCs218MHEiJRf63bXdrPtiDI2KNxjmzgfcADwDf6HLOp5rXi4g9gG2ByzPzvNb2zLyOMgDtxrXM\npsOBlYBvZ+b93e5tsi0N0+GMSUT8jDJ3UdMlwL6ZeXKHU9avy7/1KPbmxvtOqfyJWg6YU79YAGTm\nBbVZwknA7hExKzMfbpwznv8Ur6AExVtOSm0lSZKkARYRe1MCu0eAC4D3RizSQvf6zDyuvj8a2DQi\nLqJMqQmlO+IOlKD1kPbpKev0lUdTBnz6Q0ScTIkLXkMZ5Xe/zLyx7ZpHA7sBrwYuiYifAxsBe1Iy\np2/rcDvvBi4CPlcD6asoLTG3p8x+cvBYnslkMXAdcQelyewKlCzmECWrun1EnJGZD7Qd32r62+tX\nhtbgR4v0kZ0kJzWD1oYf1OXylJHIemWFe2n9OrTOYp7PCgP2DRu0+xkUA/W5LDs4DWFWWX6Q7mWR\ngeVntCesNVj3MygG6t+yATIon8sD86e7BuM3NGm9TZeo2XX5GOC9XY45F2gFrt8CXglsRWmyO4sS\nk5wIfKHO9rGIzHx/RFxGyYa+gxIo/w74dGae2eH4hyJiDvBh4HW1bncDpwCHZuZVHc65rs4wcgSl\nWfJLKUm7Y4DDM7PbVDlTYkD+U5y4zHxN631ErAnsRZnq5j3A5hGxfWb26m86HX7daWNmzo+IWynT\n8UxkbqVW4Lr2BMqQJEmSlgqZeTgl4zrW448Fjl3Max3HSAA8luPvBw6tr7Ge8xc6Z2OXOAPXDjLz\nLuDLEXEJJTjcjvLLxPGNw+bV5Yp015oseJgyNQ4AEXFLh2OHgQPqNDxjdU+PfQ9QMr2zxlHepJuJ\nv+510vq1dVDuZ1AM4ucy/5EF012FCWtlWu99cObfS8td8x4e/aAZoJVpvenOB6e5JpNj7dUGI3M8\niP+WDQI/F6m/DE47rimQmb+ntE0HeHHb7lbf1g16FNHc1+zvug4li9l8rUNpptxP1qrL23oeJUmS\nJE22mT4fjiaVGdfRtQLU9du2X16XT42IoS7NiFsjDw/T6GeamTPlB4NW4DreOackSZIkadLMlABq\nOm1cl+3BW2tuo9Upnak72akuL1mSQ0VPomfU5W+mtRaSJEla6phwVdNSHbhGxDKj7N+OkYl/z27u\ny8wrgUsp3+0PdDh3A0q/WID/P+HKTo9WQH7+tNZCkiRJ0lJtqQ5cgZMi4j8j4jkR8eggRhGxTkQc\nAPywbrqChQdmavloXf5rRHwyIlap5z8NOJ0yBc61wFen7A56W+xRkCPiqZQRie+hzGcrSZIkSdNi\nae/jugYl+PwosCAi7qY8k1Ubx1wK7J6Zj7SfnJk/iohDKNPmfAB4X0TMa5x/G7BHZk7XcJQTaemw\nV12ekJmDMfykJEmSZowh2+yqYWnPuL6fMs/SucCNwPKUwPUm4DTgTcBz6/xFHWXmkcAc4AzgDsr0\nM9cCnwWekZlXLGbdemVLh0fZP9ZjenldPX+6ssWSJEmSBAxw3+OIWEAJvDbOzBunuz4zSURsS+nX\n+ovMfNFEyrr/4eGJBM99w7nc+tMgfi7O49qfnMe1PzmPq6bSIH4uK86aWTnMW//58LT9HbnOarNm\n1LNaGiwNGVe/dOP3sbr84LTWQpIkSZIY/D6uQ8D1EQHMqPlTp01EPB94CXByZjookyRJkqaH6Sc1\nDHLg+ncm1sdzqZSZF7N0ZOIlSZIkzRADG7hm5vrTXQdJkiRJ0sQNbOAqSZIkaeaypbCabBIqSZIk\nSeprZlwlSZIk9Z2ZNXmPppoZV0mSJElSXzNwlSRJkiT1NZsKS5IkSeo7Qw7PpAYzrpIkSZKkvmbG\nVZIkSVLfcXAmNZlxlSRJkiT1NQNXSZIkSVJfM3CVJEmSJPU1A1dJkiRJUl9zcCZJkiRJfcfBmdRk\nxlWSJEmS1NfMuEqSJEnqO0OYctUIM66SJEmSpL5mxlWSJElS37GPq5rMuEqSJEmS+pqBqyRJkiSp\nr9lUWJIkSVLfsaWwmsy4SpIkSZL6mhlXSZIkSf3HlKsazLhKkiRJkvqagaskSZIkqa/ZVFiSJElS\n3xmyrbAazLhKkiRJkvqaGVdJkiRJfWfIhKsazLhKkiRJkvqaGVdJkiRJfceEq5rMuEqSJEmS+poZ\nV0mSJEmaRBGxIXAEsAuwFnAzcCpweGb+YzrrNlOZcZUkSZLUf4am8TUBEfEk4LfAW4CLgaOB64AD\ngF9GxFoTu8LSyYyrJEmSJE2eLwFrA/tl5hdbGyPiM8CBwJHAvtNUtxnLjKskSZKkvjM0jf9bXDXb\nOge4vhm0VocC9wFvjIiVFv/JLJ0MXCVJkiRpcuxQlz9p35GZ9wIXAisDz1+SlRoEBq6SJEmSNDme\nXJfZZf+f6nKzJVCXgWIfV0mSJEl9Z2hmTuS6el3e3WV/a/saS6AuA8WMqyRJkiSpr5lx1ZRacdYM\n/a1MkiRJ02qG/h3Zyqiu3mV/a7tzuY6TGVdJkiRJmhxX1eWTu+xv9W3t1gdWXRi4SpIkSdLkOKcu\n50TEQhnjiFgV2AaYB1y8pCs20xm4SpIkSdIkyMzrKFPhbAy8p2334cBKwLcz8/4lXbeZbia2G5ck\nSZKkvhQRmwAXAesAp1GaD28NbA9cDbwgM++atgrOUMtMdwUkSZIkaVDccccddz32sY/9DrAmJVh9\nMWVQ3G8Cb8pMB2aSJEmSJEmSJEmSJEmSJEmSJEmSJEmSZgBHFZYkTbqIeAtwbF19a2Ye1+GYBfXt\neZm5w5KqW7/pl+cQEd8E3lxXZ2fmjdNVF0mS2i073RWQpKVJI0jpZB7wd+C3wCnAyZk5f4lUbOoM\n19dox0y6iFgdOLCuzs3M06biOpNoSp7DYuiXekiS9KjHTHcFJGkpNtz2WhnYBNgTOAH4bURsNn3V\nmxSjteyZyiBpTeBj9fWKKbyOJEmaYmZcJWn6vLJtfS3gBcAbgBWBZwI/jYjnDOJE5Zm5JH88NYso\nSdIMZuAqSdNjODN/0GH7NyPiGOBc4HHARsCHgA8vwboNCoNVSZIGhE2FJanPZOYVwEcbm149XXWZ\n4Ya6vJckSTOMGVdJ6k9nNN5vHBErZOYD0Hn014h4FfAW4NnAesCynZriRsS/1HNfBGwArADcClwE\nfCszzxxL5SJiN+BdwFbAGsAtwAXAFzLz4jGWMebRdCMigH2AHYDZ9Zr3AdfWup+SmT+vx84Grmsr\nYu+I2LtD0R1Hz42Ix9X72wXYjNKM+27gSuB04EuZOW8M9/hE4P3ArsDjgXuAq4D/Bb6emY+MVsZ4\nRcRKlO/CS4FnUTL3ADcDc4EfA9/JzHsWo+zVgN2BHYEtgI0pfbPvAa4HzqE8m/bn36msLYB3AtsA\nT6Q0j78LuB24Abiw1vOaDucuA7we2IvynV+bkmG/HbgNuBz4KXBSZt433vuUJPUfA1dJ6k+3t623\ngsOmYWCFiDiFRQcfWqiZbESsDHyd8od+uw3r9r0i4gzgdZl5b6dK1YDh64wEzi0bUQKJ10bERyjB\nw1h1bdIbEcsCnwb2Y9FWQqtSgpZnA++OiO0z8/y2cofa1luGul23TuXzOWCVtl2PBbatr4Mi4pW9\ngvSIeAUlQF2psXn5RhlvjIiXdzt/cUTELsA3gXU67J5dX68E/gV42zjLXo7yI8dyHXavATynvvaP\niPdm5pd6lHUYZdCsdo+rr6cAOwNb0tYXvP6ocCbw3A7nP76+nk3pK3430O+jSUuSxsDAVZL609pt\n6//scMwQcAwlK3gN8G3gakoG7IWtgyJieeBnwNZ10zXAiZTs4cOUjOKbgQBeBpwaEXMys1Ng9zlG\ngtYHgeMomdYFtfy3A58ETh37rXYWEUPAyZQMH8D8Wu45lABqJeBplADnWY1T/04J5NcFvly3/bzW\nvd1CAXZEHAD8d12dB5xEyejeQflMdgFeXsv+WURslZlXdqj7NsB3Gfn/2Qsoz/xWSpZyb0rwemz7\nuYsrIvYCjmckwL+U8vyuoQTpG1EG/9ppMS/xGErQ+ldKNvMPlGe9oFH2yyn3/IWI+FtmLvI9iIg9\nGAla76OMoH0xcCelBcAGlIB1Jzr/uPBVRoLWP9XzE7gfWI3yfd6OEpzbz1mSBoSBqyT1p5c13v+5\nR3PHXSgB0hvb5nz9RuP9JxgJWj8FfDQzF5pPNiI+BXyNEpTuSGnC+eW2Y7YD9q2rdwEvycy5jUOO\nj4gvUgaWmozpZz7ASNB6A7BbZl7e4biPRMRzKMElmXk/8IPaZLjlxi6DYT0qIp4LHFVX5wJ7ZOZf\n2g77ckS8jDLP7kqUwPNf2sppZaVb/x97RGYe1nbMf1N+aOiUAR+3iNik1uUxwCPAQZn5+S7HrsHC\ngf5YPQTskpk/6VGPZwJnUZqrHxURp3X4AeQddTmf8h3qmLWuGd7N27atA+xRV38NbF8/707nb4h9\nmyVpYDg4kyT1mYh4CnBkY9NJPQ6/CXhrW9DaLGt94N119eTM/HB70ApQz9+Hkb6hB3Uo7n2N9+9t\nC1pb5WQtZ0IiYhXKaMpQMrvdgtbWded26qs6Th8DlqFkt3frELS2rnUG5ccAgK1rv+Gm3SnZa4Bz\n2oPWWsbDwFuBida55cOMNEn+ZLegtV77H5l53ngvkJkLegWt9ZjLGBlYbGNKFrbdpnV5ea+m1pn5\nUGb+pm3zJo33x3cLWuv5f8nMm3rVV5I0cxi4StL0GIqIPSLiFY3XWyLiK8BvGRlQ56+ULGk3x/b6\n452S0ZtFaTJ5VI/jWsHriXV10zqwEPBoc+Nd6+otlL6b3co5k9IMeSJ2Bdas74/vFbROhohYk5Es\n9wmZefMopzTvv73pbbNP5me6FVA/ty+OuZJd1Azva+rqP4GPT7TMCfplXQ4xkulvag1q9YQ62NN4\nNAfEesZ4KyZJmrlsKixJ0+f7o+y/AtgzM+/occwvRilju7ocogQK649y/FqN90+hNNGF0rR0Vn1/\nXpf+r01nA08d5Zhetm2879nEd5Jsw0iz0gW1H2a3ZqbDLDxA0VPa9m9Vl49Q+uP2cvZ4KtnF5pSB\nqqBkeEcd7XgiahPsvYHtKfe+BmXQqU4e32HbTymDOK0FnBcRnwTOGOMox5cDf6P0g3177Qf9VeBX\nY/hOSpJmMANXSZo+7X9oz6MM3jOXEtR+rzYp7XX+X0e5xuzG+++Oo25DjGQ8oQQKLYtMT9LBteO4\nVicb1uUwE8/ejsXsxvt9GenLOxZrta23ntUto2TDYeLPCUaeFUzxs4qI91KaSXcaWbj1fW4G/J0y\nqp8AdqMMrPUsyoBSCyJiLmUKnHOAs1rTPzVl5oKIeBdl0KnlKCMjvw24OyIuogyCdVZm/m4xbk+S\n1McMXCVpegxn5jKTUM5ogdHqzWuOs+xmcNKcGmYs82JONOvXDHg6Ts0zyVZvWx/Ps5rVtt56Vkvi\nOcESelYR8Qbg6Lo6TMn2nwf8mTKP60N1X3M050W+45n5j4h4PqUP8z6UqXseQxlJeEtgf+CeiDgG\n+M/2H28y84yIeB5wGGWu2lmUz2/X+joyIv4IfCAzz5rwjUuS+oKBqyQNtlYgsyAzJ/JvfjMgWqnr\nUSNWnsC1YOHpf9rnU50Kzft7a2YeN8GyVmPJPCdYcs/qP+ryYeDl3YLCiHj6aAXVeYIPiYiPUbKu\n21Cah+9ImXZoVeAQ4HmM9K1unv8H4FV1EK9tKINAvZAywvNylP6vZ0bEmzLz+PHcpCSpPzk4kyQN\nttbIuEN1epDF1WySvNkYjt909EN6ao0GO0RpUjrVmiMIP2GCZbWe1XoRMVrwOtHnBCPPCqboWdXp\ndmbX1VNHyWQ+sce+hWTmcGb+PjO/mJmvo0yj80rKnK4AO9fph7qdf29mnpWZh2bmDpRm2q15eIeA\noyPCv3UkaQD4j7kkDbbWtCdDLDr67XhcykhT0BeNIRh48QSuBQsPOvXyxSyjOe3PaPN5ns9I8+CJ\nPCeAX9XlMpQBjHqZ6HMC+AMjWdcdahZysq3beD9av9ydF/ciNZA9jTI1Ucu23Y7vcP6dmfk+oDWN\nztqM7YcWSVKfM3CVpMH2HUYCzg+NIQPYUWY+BJxZV9cFXt/t2IjYlUVH2h2vHzGSdXt9RCzO1CfN\n5r89m+Rm5m3Aj+vqthExZzGu19IcLfrAbgdFxIqMbxCojuq8vCfU1VWBj0y0zA6afXG7Zokj4gmU\n+Wkn6obG+8XpCz7R8yVJfcbAVZIGWGb+Bfh8Xd0MOD0i1u12fEQ8JiJ2ioiDO+xuzkn62Yh4Vofz\nNwO+PpE6A2TmfZTRZ6H0WTy9V/AaEc+OiI3ayrgTuLuuPmcMl/1/lP6bAN+JiJ6Zw4jYOCI+ExFr\nt+36IXB1ff/iiDikw7mzKM9pzM1qR/FJRgL1D0XE/t0OjIg1IuKF4yz/SkaC1z0iYqv2A+r36lRG\n6WcbEV+JiK5TJUXEssA7GpsubezbOSL2i4hVFz3z0WM2BVo/PNzL5IzcLEmaZg7OJEmD7yPAsynN\nUncArouIk4FfArdTAsP1KIPkzKFkVH8GHNksJDMvjIgvAe+mTJVzcUQcR5mCZJgykM7bgRUpAcwr\nJlLpzDwqIralNBV+IjA3Ir4PnAvcVq/zZErT3i0pzXJvbCvm55Q+k0+KiBOBUxgJZocpc9I+UK83\nNyL2pcwLuibwo4i4kJL9/TMlqF2Lkk3etl4TFg7oW1O2vJ0yrcss4PCawT2x1vuJwFso89yeArxq\nMR9R85p/rtc8gfKj9DER8TbgJOA6SrPpx1MGL9qVMjXS+eMo/+GI+DJwUL2n8yPiWODXwHxgC0qm\ndXXgOMo8r93sA+xTR/49F/gjcBclK74J8FpGsrpX13toWR/4LPDpiDgHuAS4njKC8+Moc+juxcjA\nWMdk5oNjvU9JUv8ycJWk6TFan8tJKyMz50fESykB1r6UgO+N9dVJr/lh96c0R30TsDzwzvpqeQT4\nICUgnlDgWr0aOAb4N0qTz1fXV6c6L+iw/QhgF8o971lfTbNpBLuZeWxE3EoJXteljFi7TY/63Q50\nmm/0ooh4DfBtSkC2LYv21TyfEuhPOHCt1/xeRNwHfIMSxG1eX+2GKZ9TN92+VwdTMtc7UD779vlu\nh4H/AT5N78C1dewz6qubS4E92gLP1me8HKUvbbes+ALgc5l56Cj1kCTNEDYVlqQlb5jxz6nafn5z\nOarMfDgz96dkCz9ByVTdSskizqNk5c6gZGc3z8yO/RQzc0Fm7g3sXo+/lRK43QAcD2ybmc0M5Gh1\n7Lk/M+dn5r9TssGfpQxEdBcly3cXZRCezwEvzMwLOpx/KSUz+jXgqnqvw/T4DDLzh8DGlGD5B5TA\n9j7gwXq/F9Vr7gZsUJskdyrnVODpwBcoz/cBSsb1glr2jpnZzP5OWGaeQclaHgScDdxC6eN8H3AN\n8D1KZnS/Dqf3/F7VAHInSsb9l5QBoR6gZKNPBHbOzHePVg4l87sP8E3gt5S+zPNrHa+nZKFfD2yR\nmQtl0DPzW8DWlCD6dOBPlObA84F/AHMpTeO3zMyDulxfkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJ\nkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJ\nkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJA+7/APikO1dtZ4afAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x7f3ec6b54590>"
       ]
      }
     ],
     "prompt_number": 9
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "This is pretty strange. It looks like all the classes *want* to be only in two of them! Let's check how the tip is distributed in the dataset."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tip = data.groupby('tip_perc').size()\n",
      "tip.index = np.floor(tip.index)\n",
      "\n",
      "ax = tip.groupby(tip.index).sum().plot(kind='bar', figsize=(15, 5))\n",
      "\n",
      "ax.set_xlabel('floor(tip_perc)', fontsize=18)\n",
      "ax.set_ylabel('number of trips', fontsize=18)\n",
      "ax.tick_params(labelsize=12)\n",
      "\n",
      "tip = None"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {
        "png": {
         "height": 354,
         "width": 924
        }
       },
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAABzkAAALECAYAAACbsI3oAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAWJQAAFiUBSVIk8AAAIABJREFUeJzs3Xu4rfd4N/rvkpVEzhLSRF2J0L43STUEdajSIngdSlE9\n8Dp0t6re7rYo26Hdb1UV0ZZqq5c6lGrpWxSbyN7Uue2mmygqCbcLpRVhJULO5DD3H88zX8M0xlyH\nOUbmHGt+Ptc1rt94Dvfvd4+5/vyu3/MkAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAADwXXZsdgPXp6o6McnDktw7yW2THJfk20k+n+SsJC/p7gum1J003rM7\nP9LdZ6+z/iOSPHFc+5AkX0zypiRndPdlu+n9jkmemuQeSY5J8rUk70zygu7+3G5qj0/yzCQPSnKz\nJN9M8i9J/ri737ub2oOSPCnJI5P8YJJrkpyb5NXd/Yr1agEAAAAAAGARtk3IWVUnZAgVV60kuSTJ\nYUl2jucuTvLw7n7/mtqT8p2Q83tC0An36+5/m7H+y5P80nh4dZKrkhwxHn8+yd27+yszah+b5JVJ\nDkhy3dj3URn+/S5P8uDuft+M2lOTvDdDMLr6mw8f51pJ8qzuPmNG7ZFj7e3HU5dn+FsdPB6fmeSh\n3X3ttHoAAAAAAABYhBtsdgPXo9VQ78wkP53kmO4+OsmhSR6Q5AtJjk7y1qo6bsYcK939/et8ZgWc\nT8wQcF6bYTfm4d19VJK7ZQheb5nkDTNqT03yirH/v0lyXHcfk+QWSf4hQ0j791V1kym1hyR5W4aA\n82NJbjP+5mOS/FGGkPR5VXWfGb/3FRkCzouSPKi7jxjXe1yGkPZBSX53Ri0AAAAAAAAsxHbayXlk\nkpuvE0TeKsm/Jrlhkmd393Mmrp2UYbflSncfsJfrHpzkS0mOTfKi7n7qmuu3S3J2hn+LB3f3mWuu\nvzXJg5N8JMlduntl4tphGR4de8KMuZ+U5EVJLk1y67U7RavqzUl+KsnHuvuOa66dNva1kuQhU/r6\n9SR/nOTKJCd19649/qMAAAAAAADABmybnZzdfcmsgHO8/pkkHx4Pbz/rvn1weoaA87oMuyfXrvvx\nJO8eDx81ea2qbpRhl2kyhJgra2ovT/Ky8fDnp6y9Ot/rZzwK9w/G8bSqqjXXHjmOn1kbcI5enuHR\nt4dkeM8pAAAAAAAAXC+2Tci5h74+jnu1W3M37jmOn5r1zs0k71xz76ofy/AOzJUk79pN7fFVdfLq\nyao6Iskd1tyz1oczBJU7ktx7Rt9T1+3uq5J8cDy814z5AQAAAAAAYO6EnKOq2pnhHZlJ8qkZt+2o\nqg9V1SVVdWVVfaGq/rqq7jbj/iQ5ZRzPWeeec8fx2Kq68ZTaC7r74hm15025P0lWA8+VWWuPO0M/\ns+b+VNWOJLfei75PXuceAAAAAAAAmCsh53f8apLjklyb5K9m3LOS5E5Jrsnw+NkTMzwS9h+r6sUz\nam46juevs/bkDs/j96a2u69M8s0MuzGn1e5u7dVrk/cfmeTQDL93T/q+6Tr3AAAAAAAAwFwJOZNU\n1alJnj8e/ll3f3rNLVcmeWmSeyQ5oruP6e7DktwxydvHe36jqp45ZfrDJuaY5YqJ74fvZe1k/bTa\nPV17XrUAAAAAAACwUDs3u4HNVlU3TfLWJDdM8tEkT197T3d/NcmvTTn/r0keUlV/l+QRSZ5VVX/e\n3d9cbNcAAAAAAACwfW3rnZxVdUySdyU5KUkneWB3f3sfploNRg9Ncu811y4fx0PWqT904vtle1k7\nWT+tdk/XnlctAAAAAAAALNS23clZVUcleWeSH0ryxSSnd/eufZmru/+9qi5McpMkt1hz+fwkt03y\n/etMMXlt8v2cX55y/btU1SFJjlqnNkluliHEXW/tydpLMjyK9tD11p5Ru9dWVlZWNlIPAAAAAADA\nctqxY8eOfanbliFnVR2W5Kwkd8gQ0J3e3f+5oOXOSXL/JLdZ555TxnFXd3994vy543h8VR2z5tra\n2pWJ+5Nk9b2iOzIEud8TclbVDZLcas1a6e6Vqjo3wztH96Tvc9e5Z7d27bp0n2uPPfaIDc+xUXrY\nGj1s9vp60MNWWl8PethqPWz2+nrQw1brYbPX14MettL6etDDVuths9fXgx62Wg+bvb4e9LCV1teD\nHrZaD5u9/lboYds9rnbc+fj2JHdNcmGGgPNzG5zzFhl2cSbJF9Zcft84nlJVx8+Y4r7j+J415/8p\nyTUZgsrTd1N7fnevBpvp7suSfGQ8vM+M2jsnOTJDQLp27dW+p9ZW1Q2T3H1G3wAAAAAAALAw2yrk\nrKqDkrw5yU8kuTjJfbv7vDlM/bxxvCLJe9dce0+SryU5IMlvTunpthkCzJUkr5u81t2XJHnHePiU\nqtqxpvawJL8yHv7tlL5eP46PmhGwPnUcz+7uz665tjrfravqgVNqH58hIL0iyVumXAcAAAAAAICF\n2DYhZ1UdkCH0u1+Gd07ev7s/voe176+qp1XVrcdHvKaqdlTVaVX1liQ/O956Rnd/Y7K2u7+d5Nnj\n4ZOr6ilj2JqqumuGgHBHkn/u7rOmLP87Sa5Ocqckr6mqG4+1J2YIbE/IENieMaX2LzK8b/SIJGdW\n1clj7RFV9cIkD80Qrj5rbeH4t3nDePiaqrr/WHtAVT1mYr0Xd/eFU/9wAAAAAAAAsADb6Z2cd0vy\nsPH7QUneVlWz7v1Sd99p4vjmGUK9M5JcU1WXJDk0yQ3H6ytJ/rS7f2/aZN39sqo6LcPuxz9M8oKq\n+laSw8ZbPpfkZ2bUfrKqHp/klUkeneTR4/pHjrdcluTh3X3RlNqrquohGXaT3j7JOVV1aZLDMwSr\n1yV5Vne/e8bf4fFJfiDDu0vfUVVXZtiRetB4/e0ZQlgAAAAAAAC43mybnZwZQr1kCCQPTnLsOp+b\nrKl9WpJXJPl4hvd4Hp7hXZmfTvKqJHfp7iett3h3PyHDjs/3ZdhJeoMk5yV5bpLbdfcF69S+NsM7\nRN+Q5Ctj/18a175dd79/ndpPJrlNkj/JEKYemGRXkjOT3Ke7X7hO7aVJfjTJM5J8Ism1Sa5M8qEk\nv9zdD+nu69b73QAAAAAAADBv22YnZ3d/IPsY6nb3m5K8aQ49vDHJG/ex9uwkP7ePtV9N8qTxs7e1\nVyd54fgBAAAAAACATbeddnICAAAAAAAA+wEhJwAAAAAAALBUhJwAAAAAAADAUhFyAgAAAAAAAEtF\nyAkAAAAAAAAsFSEnAAAAAAAAsFSEnAAAAAAAAMBSEXICAAAAAAAAS0XICQAAAAAAACwVIScAAAAA\nAACwVIScAAAAAAAAwFIRcgIAAAAAAABLRcgJAAAAAAAALBUhJwAAAAAAALBUhJwAAAAAAADAUhFy\nAgAAAAAAAEtFyAkAAAAAAAAsFSEnAAAAAAAAsFSEnAAAAAAAAMBSEXICAAAAAAAAS0XICQAAAAAA\nACwVIScAAAAAAACwVIScAAAAAAAAwFIRcgIAAAAAAABLRcgJAAAAAAAALBUhJwAAAAAAALBUhJwA\nAAAAAADAUhFyAgAAAAAAAEtFyAkAAAAAAAAsFSEnAAAAAAAAsFSEnAAAAAAAAMBSEXICAAAAAAAA\nS0XICQAAAAAAACwVIScAAAAAAACwVIScAAAAAAAAwFIRcgIAAAAAAABLRcgJAAAAAAAALBUhJwAA\nAAAAALBUhJwAAAAAAADAUhFyAgAAAAAAAEtl52Y3AAAAME+/+htPzgW7Lp55feeBByRJrrn62pn3\nHH/s0XnpS148994AAACA+RByAgAA+5ULdl2co059zMbm+ORr59QNAAAAsAgeVwsAAAAAAAAsFSEn\nAAAAAAAAsFSEnAAAAAAAAMBSEXICAAAAAAAAS0XICQAAAAAAACwVIScAAAAAAACwVIScAAAAAAAA\nwFIRcgIAAAAAAABLRcgJAAAAAAAALBUhJwAAAAAAALBUhJwAAAAAAADAUhFyAgAAAAAAAEtFyAkA\nAAAAAAAsFSEnAAAAAAAAsFSEnAAAAAAAAMBSEXICAAAAAAAAS0XICQAAAAAAACwVIScAAAAAAACw\nVIScAAAAAAAAwFIRcgIAAAAAAABLRcgJAAAAAAAALBUhJwAAAAAAALBUhJwAAAAAAADAUhFyAgAA\nAAAAAEtFyAkAAAAAAAAsFSEnAAAAAAAAsFSEnAAAAAAAAMBSEXICAAAAAAAAS0XICQAAAAAAACwV\nIScAAAAAAACwVIScAAAAAAAAwFIRcgIAAAAAAABLRcgJAAAAAAAALBUhJwAAAAAAALBUhJwAAAAA\nAADAUhFyAgAAAAAAAEtFyAkAAAAAAAAsFSEnAAAAAAAAsFSEnAAAAAAAAMBSEXICAAAAAAAAS0XI\nCQAAAAAAACwVIScAAAAAAACwVIScAAAAAAAAwFIRcgIAAAAAAABLRcgJAAAAAAAALBUhJwAAAAAA\nALBUhJwAAAAAAADAUhFyAgAAAAAAAEtFyAkAAAAAAAAsFSEnAAAAAAAAsFSEnAAAAAAAAMBS2bnZ\nDVyfqurEJA9Lcu8kt01yXJJvJ/l8krOSvKS7L1in/qAkT0ryyCQ/mOSaJOcmeXV3v2IP1n9EkieO\nax+S5ItJ3pTkjO6+bDe1d0zy1CT3SHJMkq8leWeSF3T353ZTe3ySZyZ5UJKbJflmkn9J8sfd/d7d\n1G7oNwMAAAAAAMC8bZudnFV1QpJ/T/KiJA/MEPZdkeTgJD+c5OlJzqmqn5hRf2SS/zfJC5KcmmRl\nrL1Lkr+oqrdV1QHrrP/yJH+X5CeSHJ7k6iS3SvJbST5eVTddp/axST6U5GeSfN/Y982S/OJYe891\nak9N8qkkv5bkpCRXZghJH5TkH6rq6evUbug3AwAAAAAAwCJsm5AzyQEZQrozk/x0kmO6++gkhyZ5\nQJIvJDk6yVur6rgp9a9IcvskFyV5UHcfkeSwJI9LclWG0PB3py1cVU9M8ktJrs2wG/Pw7j4qyd0y\n7Oa8ZZI3zKg9dVz7gCR/k+S47j4myS2S/MPYw99X1U2m1B6S5G0ZQs2PJbnN+JuPSfJHSXYkeV5V\n3WfqX2wDvxkAAAAAAAAWZTuFnF9PcrvufnB3v7m7v5kk3X1Nd/8/GYLOq5IcmeQJk4VVdVqSR2QI\nSX+hu88aa6/r7tcmecZ465Or6tg1tQcnefZ4+JLuflF3Xz3WfyjJQ8d571ZVD5rS93MyPFb4I0ke\n290XjbVfyvDo3f9IcqOJHiY9IcmJSS5N8pPdfd5Ye2l3Py3JWzMEnc9fW7iR3wwAAAAAAACLtG1C\nzu6+pLv/bZ3rn0ny4fHw9msuP3IcP9PdZ04pf3mSSzK8Z/Nha66dnuTYJNdl2D25dt2PJ3n3ePio\nyWtVdaMM4WuSvKi7V9bUXp7kZePhz0/pa3W+13f3V6Zc/4NxPK2qas21jfxmAAAAAAAAWJhtE3Lu\noa+P49r3TK6+8/Jd04q6+6okHxwP7zWj9lMzgsYkeeeae1f9WIZdnCuz1p6oPb6qTl49WVVHJLnD\nmnvW+nCGoHJHknvP6HtffjMAAAAAAAAsjJBzVFU7M7wjM0k+NXF+R5Jbj4fnrDPFueN48przp+xF\n7bFVdeMptRd098Uzas+bcv9kHyuz1h53hn5mzf3z+M0AAAAAAACwMELO7/jVJMcluTbJX02cPzLJ\noRnCwvPXqV/dpXnTNedXj/ekNkmO35va7r4yyTcz7MacVru7tVevTd6/0d8MAAAAAAAACyPkTFJV\npyZ5/nj4Z9396YnLh018v3Kdaa4Yx8PXnF+t35PatfV7Ujtr7Y30vdHfDAAAAAAAAAuzc7Mb2GxV\nddMkb01ywyQfTfL0ze1o+zn22CO2xBx62D962Oz19aCHrbS+HvSw1XrY7PX1sH162HngAXOZY9F/\np/3930EPy9XDZq+vBz1stR42e3096GGr9bDZ6+tBD1tpfT3oYav1sNnrb2YP23onZ1Udk+RdSU5K\n0kke2N3fXnPb5RPfD1lnukPH8bIZ9XtSu7Z+T2pnrb2Rvjf6mwEAAAAAAGBhtu1Ozqo6Ksk7k/xQ\nki8mOb27d0259ZIMj2U9NMn3rzPl6rWvrDl/fpLb7mHt2vovT7n+XarqkCRHrVObJDfLEOKut/Zk\n7UZ/817ZtevSfa5d/d8BG5ljo/SwNXrY7PX1oIettL4e9LDVetjs9fWw/Xq45upr5zLHonrcLv8O\neliOHjZ7fT3oYav1sNnr60EPW62HzV5fD3rYSuvrQQ9brYfNXn8r9LAtd3JW1WFJzkpyhwwB3end\n/Z/T7u3ulSTnjoe3WWfaU8bx3DXnz9mL2l3d/fWJ86tzHT/uOl2vdrLPJFl9r+iODEHu96iqGyS5\n1dq+5/CbAQAAAAAAYGG2Xcg57nx8e5K7JrkwQ8D5ud2UvW8c7zNjzhsmuft4+J4ZtadU1fEz5r/v\njNp/SnJNhqDy9N3Unt/dq8FmuvuyJB9Zr+8kd05yZIaAdFbf+/KbAQAAAAAAYGG2VchZVQcleXOS\nn0hycZL7dvd5e1D6t+N466p64JTrj88QFl6R5C1rrr0nydeSHJDkN6f0dNsMAeZKktdNXuvuS5K8\nYzx8SlXtWFN7WJJfWdPjpNeP46NmBKxPHcezu/uza65t5DcDAAAAAADAwmybkLOqDsgQ+t0vwzsn\n79/dH9+T2vG+N4yHr6mq+6/OWVWPSXLGeO3F3X3hmtpvJ3n2ePjkqnrKGLamqu6aISDckeSfu/us\nKcv/TpKrk9xpXPvGY+2JGQLbEzIEtmdMqf2LDO8bPSLJmVV18lh7RFW9MMlDM4Srz5rnbwYAAAAA\nAIBF2rnZDVyP7pbkYeP3g5K8rapm3ful7r7TmnOPT/IDGd7j+Y6qujLD7syDxutvzxBIfo/ufllV\nnTbO8YdJXlBV30py2HjL55L8zIzaT1bV45O8Msmjkzy6qi7JsIsySS5L8vDuvmhK7VVV9ZAMu0lv\nn+Scqro0yeEZgtXrkjyru9894++wz78ZAAAAAAAAFmXb7OTMEOolw87Fg5Mcu87nJmuLu/vSJD+a\n5BlJPpHk2iRXJvlQkl/u7od093WzFu/uJyT52Qzvurwkw9/+vCTPTXK77r5gndrXZniH6BuSfGXs\n/0tJXjXWvn+d2k8muU2SP8kQph6YZFeSM5Pcp7tfuE7thn4zAAAAAAAALMK22cnZ3R/IBkPd7r46\nyQvHz77UvzHJG/ex9uwkP7ePtV9N8qTxs7e1G/rNAAAAAAAAMG/baScnAAAAAAAAsB8QcgIAAAAA\nAABLRcgJAAAAAAAALBUhJwAAAAAAALBUhJwAAAAAAADAUhFyAgAAAAAAAEtFyAkAAAAAAAAsFSEn\nAAAAAAAAsFSEnAAAAAAAAMBSEXICAAAAAAAAS0XICQAAAAAAACwVIScAAAAAAACwVIScAAAAAAAA\nwFIRcgIAAAAAAABLRcgJAAAAAAAALBUhJwAAAAAAALBUhJwAAAAAAADAUhFyAgAAAAAAAEtFyAkA\nAAAAAAAsFSEnAAAAAAAAsFSEnAAAAAAAAMBSEXICAAAAAAAAS2XnZi5eVSck+e0k90tyXJKvJ3lf\nkud39zmb2RsAAAAAAACwNc095Kyq70vyb0l2JDk/ye27+7op952c5ANJbjJx+qZJHpnk4VX18939\n1nn3BwAAAAAAACy3RTyu9qeSHJshvHzLtIBz9Lp8d8A56eAkr6uqWy6gPwAAAAAAAGCJLSLk/NFx\nXEnylmk3VNWDk9xu4tQHkzwlye8k+ep47pAkz1lAfwAAAAAAAMASW8Q7OU8Zx8u7+5Mz7nnMxPd3\nJHlwd68kSVW9LsknkhyW5KFVdWh3X7GAPgEAAAAAAIAltIidnDcfx89Ou1hVN0hy+sSp318NOJOk\nuz+f5K/Hw0OS3HEBPQIAAAAAAABLahEh55HjePGM66dN3HN+d394yj0fnPh+q3k1BgAAAAAAACy/\nRYScB+7m+l0mvr93xj3nT3w/emPtAAAAAAAAAPuTRYScl47jsTOu32Pi+z/uwXwHbKwdAAAAAAAA\nYH+yiJDzP8bxVlV12OSFqjooyf0mTv3TjDkmd29eOuMeAAAAAAAAYBtaRMj5kXE8KMkT11x7TL77\nfZznzZjj5Inv/znH3gAAAAAAAIAlt3MBc/5dkl8Yvz+3qo7K8Fja2yf5nYn7XrfOHHcex5Ukn557\nhwAAAAAAAMDSmnvI2d3vqqoPZnj35kFJfmvKbZclefG0+qo6NMl9xsNLulvICQAAAAAAAPwvi3hc\nbZL8bJJzZ1z7dpLHdPcFM64/Ismh4/d/nHdjAAAAAAAAwHJbSMjZ3V9NcockT07y3iSfTfKJJK9M\ncofufus65U+d+H7WIvoDAAAAAAAAltci3smZJOnubyV5yfjZm7ofXkxHAAAAAAAAwP5gUY+rBQAA\nAAAAAFgIIScAAAAAAACwVIScAAAAAAAAwFJZ2Ds5J1XVaUkenuTOSf5LkqOTHJzkm0m+muTsJB9I\n8obuvuL66AkAAAAAAABYTgsNOcdw86VJ7jLjlmPHz22SPDbJH1fVS5I8t7uvXmRvAAAAAAAAwHJa\n2ONqq+pXk/x/mR1wTnNkkv8zyUer6riFNAYAAAAAAAAstYXs5Kyqxyf50zWnL0vy/iSfSnJRkm9l\nCDVvmeExtj80ce8PJ/lAVd2xuy9bRI8AAAAAAADAcpp7yFlVJyZ50cSpi5P8jyR/2d1XrlN3apIz\nktxv9VSS5yf5tXn3CAAAAAAAACyvRTyu9peTHDZ+/0qSu3b3S9cLOJOkuz/Z3fdP8uKJ079YVYcv\noEcAAAAAAABgSS0i5HzQxPf/3t29l/X/R5J/Hb/fMMnpc+kKAAAAAAAA2C8sIuQ8aRy/luRte1vc\n3dcmedXEqZvPoScAAAAAAABgP7GIkPOAcfx0d6/s4xyfnjIfAAAAAAAAwEJCzi+P40EbmOPAie9f\n2cA8AAAAAAAAwH5mESHnh8fxlKra16DzDuO4kuRfNt4SAAAAAAAAsL9YRMj5l+N4ZJIn7G1xVR0+\nUfdP3f35eTUGAAAAAAAALL+5h5zd/cEkrxoPX1BVD9jT2qo6NMkbkpyQ5PIkvzLv/gAAAAAAAIDl\ntoidnEnyvyf52ySHJHlbVf1lVZ026+aqOqKqfiHJOUn+a5ILk/xUd5+3oP4AAAAAAACAJbVz3hNW\n1aszvEvz20l2JTk2yeOSPLaqvpbkU0kuGq8fkeSWSW6VZPL9nR9P8qiqetR6a3X3/zbv/gEAAAAA\nAICtbe4hZ5LHzji/I8lx42d3Tt+De1aSCDkBAAAAAABgm1nU42qvDzs2uwEAAAAAAADg+reInZyv\nXcCc06xcT+sAAAAAAAAAW8jcQ87ufty85wQAAAAAAABYtcyPqwUAAAAAAAC2ISEnAAAAAAAAsFSE\nnAAAAAAAAMBSEXICAAAAAAAAS0XICQAAAAAAACyVnftSVFXvS7Kyetzd95p1baMm5wYAAAAAAADY\np5AzyY9PfF8baP545mduYSkAAAAAAACwf5jH42p3zGGOzZgbAAAAAAAAWEL7upPzORPf1+62fE7m\nx05OAAAAAAAA4LvsU8jZ3c/el2sAAAAAAAAAGzWPx9UCAAAAAAAAXG+EnAAAAAAAAMBS2dd3cs5U\nVdflO+/S/C/d/fl5rwEAAAAAAABsX4vaybkjSQs4AQAAAAAAgHlbRMh54Tiev4C5AQAAAAAAgG1u\nESHnarh55ALmBgAAAAAAALa5RYSc7x7HH6qqQxYwPwAAAAAAALCNLSLk/JskK0lumOSXFzA/AAAA\nAAAAsI3NPeTs7o8n+cPx8Per6p7zXgMAAAAAAADYvhaxkzNJnpXkz5IcmuRdVfXSqjp1QWsBAAAA\nAAAA28jOeU9YVe/L8LjaHUkuTXJEkicm+ZWquiTJF5JckuS6PZmvu+817x4BAAAAAACA5TX3kDPJ\nj884vyPJUUlutxdzrWy8HQAAAAAAAGB/sqjH1c7Ljs1uAAAAAAAAANhaFrGT8zlznMtOTgAAAAAA\nAOC7zD3k7O5nz3tOAAAAAAAAgFVb/XG1AAAAAAAAAN9FyAkAAAAAAAAslbk/rraqHpvhXZpf7u73\n7OMc90xyQpJ092vn2B4AAAAAAACw5OYeciZ5dYaQ851J9inkTPLrSR4yziPkBAAAAAAAAP6XRT2u\ndscm1wMAAAAAAAD7Ke/kBAAAAAAAAJbKVg05DxrHb29qFwAAAAAAAMCWs1VDzh8cx29uahcAAAAA\nAADAlrNzsxtYq6p+Id8JOT8z57kPT3KvJD+S5I7jeMx4+dbd3TPqTkry+T1Y4ke6++x11n9Ekicm\nuW2SQ5J8McmbkpzR3Zftpvc7JnlqknuMPX8tyTuTvKC7P7eb2uOTPDPJg5LcLEN4/C9J/ri737ub\n2oOSPCnJIzP8u1yT5Nwkr+7uV6xXCwAAAAAAAIuwoZCzqt6XZGXi1I6J73euqnUDtDV1hyT5gSQ3\nnjj/7o30N8XpSd485fzKlHOzXLDOtZmP162qlyf5pfHw6iRXJblVkt9K8vNVdffu/sqM2scmeWWS\nA5Jcl+SSDGHlLyb5uap6cHe/b0btqUnemyEYXRlrj8kQeD6wqp7V3WfMqD1yrL39eOryJAcnuUuS\nu1TVTyZ5aHdfO+t3AwAAAAAAwLxtdCfnj69z7egkP7GBuS9K8hcbqJ9mJcMOyI+Mn/OTvHxv6rv7\n+/d20ap6YoaA89okT0/yp919dVXdNcnrk9wyyRuS3H1K7alJXpEh4PybJE/u7ouq6sTx/H2S/H1V\nVXdfuKb2kCRvyxBqfizJo7v7vKo6Isn/SPKbSZ5XVR/r7n+Y0vorMgScFyV5bHefVVU3SPLfkrws\nQ1D6u0l+e2//JgAAAAAAALCvtuo7OT+a5PTu/tqc5317dx/f3T/Z3c/J/HeKfo+qOjjJs8fDl3T3\ni7r76iQUd0o4AAAgAElEQVTp7g8leWiG8PVuVfWgKVM8J0MY/ZEMQeNFY+2XkjwsyX8kuVGSZ0yp\nfUKSE5NcmuQnu/u8sfbS7n5akrdm2EX7/Cl9n5bkEWNvv9DdZ42113X3ayfWe3JVHbvnfxEAAAAA\nAADYmI3u5HzOlHP/Yxw/l2Hn4Y4p96x1XYZHoZ6f5OxZ78bcqO6+bhHz7sbpSY7N8Bv/aO3F7v54\nVb07w47MRyU5c/VaVd0oyQPGwxd198qa2sur6mVJfj/Jz2d4Z+ekR43j62c8CvcPkvxUktPGnaCT\nf/dHjuNnuvvM7y3NyzP8+x+ZIWyd965bAAAAAAAAmGpDIWd3P3vtuapaDTk/292/u5H59xP3HMdP\nzXrnZpJ3Zgg577nm/I9l+DdaSfKudWp/P8nxVXXy6m7N8ZG0d5i4Z5oPZ3hH55FJ7p1kMuRc7WXq\nut19VVV9MMMja+8VIScAAAAAAADXk0U8rvZLSb6Y4d2X+5sdVfWhqrqkqq6sqi9U1V9X1d3WqTll\nHM9Z555zx/HYqrrxlNoLuvviGbXnTbk/SU4ex5VZa487Qz+z5v5U1Y4kt96Lvk9e5x4AAAAAAACY\nq7mHnN19UnfforsfN++5t4CVJHdKck2Gx8+emOGRsP9YVS+eUXPTcTx/nXknd3gevze13X1lkm9m\neCzwtNrdrb16bfL+I5McmuH37knfN13nHgAAAAAAAJirRezk3B9dmeSlSe6R5IjuPqa7D0tyxyRv\nH+/5jap65pTawybmmOWKie+H72XtZP202j1de161AAAAAAAAsFAbeifndtHdX03ya1PO/2uSh1TV\n3yV5RJJnVdWfd/c3r+8el9mxxx6xJebQw/7Rw2avrwc9bKX19aCHrdbDZq+vh+3Tw84DD5jLHIv+\nO+3v/w56WK4eNnt9Pehhq/Ww2evrQQ9brYfNXl8PethK6+tBD1uth81efzN7sJNzPp4+jocmufea\na5eP4yHr1B868f2yvaydrJ9Wu6drz6sWAAAAAAAAFspOzjno7n+vqguT3CTJLdZcPj/JbZN8/zpT\nTF6bfD/nl6dc/y5VdUiSo9apTZKbJendrD1Ze0mGR9Eeut7aM2r32q5dl+5z7er/DtjIHBulh63R\nw2avrwc9bKX19aCHrdbDZq+vh+3XwzVXXzuXORbV43b5d9DDcvSw2evrQQ9brYfNXl8PethqPWz2\n+nrQw1ZaXw962Go9bPb6W6EHOzkX75xxvM0695wyjru6++sT588dx+Or6pjd1K5M3J8knx7HHUl+\naFphVd0gya3WrJXunpxrT/o+d517AAAAAAAAYK6EnHNQVbfIsIszSb6w5vL7xvGUqjp+xhT3Hcf3\nrDn/T0muyRBUnr6b2vO7ezXYTHdfluQj4+F9ZtTeOcmRGQLStWuv9j21tqpumOTuM/oGAAAAAACA\nhRFyzsfzxvGKJO9dc+09Sb6W5IAkv7m2sKpumyHAXEnyuslr3X1JkneMh0+pqh1rag9L8ivj4d9O\n6ev14/ioGQHrU8fx7O7+7Jprq/PduqoeOKX28RkC0iuSvGXKdQAAAAAAAFiIbRdyVtVNVj9Jjp64\ndPTktclAsareX1VPq6pbj494TVXtqKrTquotSX52vPWM7v7G5Hrd/e0kzx4Pn1xVT6mqg8Y57poh\nINyR5J+7+6wpLf9OkquT3CnJa6rqxmPtiUnenOSEJBcnOWNK7V8k+WKSI5KcWVUnj7VHVNULkzw0\nQ7j6rLWF3f3xJG8YD19TVfcfaw+oqsdMrPfi7r5wytoAAAAAAACwENsu5Mywq3L1c/Z4bkeSD625\nduJEzc0zhHrnJrmqqi5McvlY/5AMQeGfdPfvTVuwu1+W5BUZ/t5/mOSyqro0yT8nOSnJ55L8zIza\nT2bYNXlNkkcn2VVV30jy7xkeJXtZkod390VTaq8a+7soye2TnFNV30zyjQy7OK9L8szufveMv9Xj\nx9944yTvqKrLx9/9miQ3TPL2DCEsAAAAAAAAXG+2Y8iZDKHknnxWPS1DSPnxJBcmOTxD6PjpJK9K\ncpfuftJ6C3b3EzLs+Hxfkksy/O3PS/LcJLfr7gvWqX1tkrtm2Fn5lSQHJ/nSuPbtuvv969R+Mslt\nkvxJhjD1wCS7kpyZ5D7d/cJ1ai9N8qNJnpHkE0muTXJlhkD4l7v7Id193Xq/GwAAAAAAAOZt574U\nVdXq7sWvdve/zLelxeruvQ52u/tNSd40h7XfmOSN+1h7dpKf28faryZ50vjZ29qrk7xw/AAAAAAA\nAMCm26eQM8N7JFeSvDPJAyYvVNVjx2tf7u73bKw9AAAAAAAAgO+2ryFnMrzHcppX5zsBqJATAAAA\nAAAAmKtFvZNzVgAKAAAAAAAAsCH7GnJeOY6HzasRAAAAAAAAgD2xryHnBeP4w1V10LyaAQAAAAAA\nANidfX0n50eT3CLJjZK8uar+IMl/JLl24p5DqurEDfaX7v7SRucAAAAAAAAA9h/7GnL+dZJHjN8f\nMH5W1tzz40m+sI/zJ8N7PVeSHLCBOQAAAAAAAID9zD49rra7z0zyP9ec3jF+pp3blw8AAAAAAADA\n99jXd3ImyX9L8utJPjWnXqYRdgIAAAAAAADfZV8fV5vuvi7JnyX5s6o6JMkxSQ5M8vnxlg8meWwE\nlQAAAAAAAMAc7XPIOam7r0zy5SSpqtXTV3T3F+cxPwAAAAAAAMCqjTyudnfs4AQAAAAAAADmbi47\nOde41zhetIC5AQAAAAAAgG1u7iFnd79/3nMCAAAAAAAArFrk42oBAAAAAAAA5m4Rj6v9HlV1QpIH\nJ7lrkh9IcnSSGyb5RpKvJflokg8keVd3r1wfPQEAAAAAAADLaaEhZ1XdLMmLkzw0yQFTbjlxHE9P\n8owk/15Vv9fdr15kXwAAAAAAAMDyWtjjaqvqvkn+LclPZ3rAOc1JSV5VVX9fVQcuqjcAAAAAAABg\neS1kJ2dV3SnJm5McOnF6JcknknwyyYVJvpXkyAyPr71TkmMm7n1oktcnecQi+gMAAAAAAACW19xD\nzqq6QZK/zHcCzmuT/HmSF3X3F2fU7EzywCTPT3Lr8fTDq+rnuvt/zrtHAAAAAAAAYHkt4nG1j0hy\nyvj9iiT37+7fmBVwJkl3X9Pd/1eS05K8beLS7yygPwAAAAAAAGCJLSLkfPDE96d397v3tLC7v5Xk\nUUlWA9GqqppncwAAAAAAAMByW0TI+SPjeGmSV+5tcXdfnuTl4+GOifkAAAAAAAAAFhJyHj+O54w7\nM/fFxya+f98G+wEAAAAAAAD2I4sIOXeO49UbmGOydufMuwAAAAAAAIBtZxEh565x3Mi7NCdrL9zA\nPAAAAAAAAMB+ZhEh5znjeFxVPWgf53jclPkAAAAAAAAAFhJynjXx/aVVdeLeFFfVbye503i4K8lH\n5tUYAAAAAAAAsPwWEXK+NslF4/cTknyoqn56d0VVdVxVvSbJcyZOv6S7V+bfIgAAAAAAALCsds57\nwu6+pKqekuSvxlM3TfKGqvpiknck+USGHZrfTnJEkh9M8qNJTk9y4MRU5yR50bz7AwAAFutXf+PJ\nuWDXxVOv7TzwgCTJNVdfO7P++GOPzktf8uKF9AYAAADsH+YeciZJd/91VZ2Q5LkTp2+e5L/v4RSf\nT/Jfu/uquTcHAAAs1AW7Ls5Rpz5m3+s/+do5dgMAAADsjxbxuNokSXc/L8kDk/zHXpRdl+TVSU7r\n7i8vpDEAAAAAAABgqS1kJ+eq7v6/q+oHkjx0/Nw1w47OSVckOTvJB5K8qru/uMieAAAAAAAAgOW2\n0JAzSbr7miRvHD+pqp1JbpTk4CTf6O7LF90DAAAAAAAAsP9YeMi51hh6Xnh9rwsAAAAAAADsHxb2\nTk4AAAAAAACARRByAgAAAAAAAEtFyAkAAAAAAAAsFSEnAAAAAAAAsFSEnAAAAAAAAMBSEXICAAAA\nAAAAS0XICQAAAAAAACwVIScAAAAAAACwVIScAAAA/z979x4t2VnWif/b6U4g95ttOjAyiPBAYkzk\nogiIA4HwE4jDxcELDAFUBCcLMQqD4KxRcYZLHIgZB+WiglHiErxgDPjDgCBCQCEOBnJ7EUcYCR2a\n0Ll0LqbTOfNH7QOVyrlX7a6qPp/PWmft2rX3+z7fs6tyUqefs/cGAAAA5oomJwAAAAAAADBXNDkB\nAAAAAACAubJt0hNW1b9LstCtfqa1tnvSNQAAAAAAAIDNa+JNziQf6pa3J7lPD/MDAAAAAAAAm1gf\nl6vd1y2vaa3d0MP8AAAAAAAAwCbWR5Pzum55fQ9zAwAAAAAAAJtcH03OL3TLE3qYGwAAAAAAANjk\n+mhy/lm3PKmqNDoBAAAAAACAieqjyfn7SW7u5v7lHuYHAAAAAAAANrGJNzlba9cm+alu9Ser6r9W\n1ZZJ1wEAAAAAAAA2pz7O5Exr7cIk/yGDMzp/Kck/VNVPV9V3VtVRfdQEAAAAAAAANodtk56wqu5K\nstCtLp7BeUqSX+seL1TVWqbakmShtbZ1sgkBAAAAAACAeTbxJmdnpcvTunQtAAAAAAAAsGG9XK52\ngjREAQAAAAAAgLvp40zO0yc418LquwAAAAAAAACbycSbnK21D096TgAAAAAAAIBFs365WgAAAAAA\nAIC70eQEAAAAAAAA5oomJwAAAAAAADBXJn5PzqVU1RFJHpXkEUm+KckxSba01n5sf9QHAAAAAAAA\nDhy9NjmrakeS/5Lk+UkOG9m8kOQeTc6qen+SB3WrT2yt/VOfGQEAAAAAAID50tvlaqvqCUn+Icl/\nyj0bnCt5T5L7d1/PnXgwAAAAAAAAYK710uSsqu9OclGS7UNP/58kf5bk2lWG/0GSO7vH/37y6QAA\nAAAAAIB5NvEmZ1VtTfJ7SQ7tnroqyfe11r6ttfaMJJ9daXxr7YYkH+5WT6uqYyadEQAAAAAAAJhf\nfZzJ+ex8456aVyV5VGvto+uc4xPd8qAk3zGpYAAAAAAAAMD866PJOXyJ2bNbazdtYI4rhh4/cMw8\nAAAAAAAAwAGkjybnw7vll1trH97gHNcPPT52vDgAAAAAAADAgaSPJucJ3fJzY8xx+9Dje40xDwAA\nAAAAAHCA6aPJuTCBOY4ZenzDBOYDAAAAAAAADhB9NDm/0i3vN8Ycpw493jXGPAAAAAAAAMABpo8m\n55Xd8t9W1QM2OMfTuuVCkk+MHwkAAAAAAAA4UGzrYc7/P8lTkmxJ8qokP7GewVX1lCTf1a1e01r7\nl8nGAwCAA9fZLz0nO3ftXnLbtoO3Jknu3Ltv2fE7th+bN51/Xi/ZAAAAACaljybnO5P8SpKjk/xY\nVX28tfbbaxlYVd+Z5HeHnvq1HvIBAMABa+eu3Tn61LM2Pv7yCyaYBgAAAKAfE79cbWttd5JXDz31\ntqp6R1Wd0q0vjI6pqvtX1X9L8tEkx3dPX5Hk7ZPOBwAAAAAAAMy3Ps7kTGvtvO6szOd2T52V5LlV\n9ZUkh3bPbamq/53khCQ7RqbYneTprbW9feQDAAAAAAAA5tfEz+Qc8uNJzh1a35JBQ/OooedOyz0b\nnJ9L8pjW2ud7zAYAAAAAAADMqd6anK21O1trP5/k0UkuyhKXqR1xfZL/kuThrbWr+8oFAAAAAAAA\nzLdeLlc7rLX2iSRPr6rjkjw2yakZ3Hfz8CQ3JPlykkuTfLK1tq/vPAAAAAAAAMB8673Juai19rUk\nf9Z9AQAAAAAAAGxIn/fkBAAAAAAAAJg4TU4AAAAAAABgruy3y9UmSVXdN8mDkhyb5F5JbkxyXZLP\ntNb27s8sAAAAAAAAwHzqvclZVScleXGSH0xyYpItS+z2r1X1t0l+O8kfttbu6DsXAAAAAAAAMJ96\nu1xtVd27qt6Y5DNJXpLkPlm6wZkMzur8viS/m+QzVfW9feUCAAAAAAAA5lsvTc6qOiLJ+5P8zAZq\nPCjJB6vqWRMPBgAAAAAAAMy9vs7kfEeSxw6t70ny9iT/IclDkmxPclSSf5PBGZwvT/K/h/Y/OMk7\nq+qhPeUDAAAAAAAA5tTE78lZVU9K8syhpy5O8sLW2nVL7L4nybVJPprkDVX1w0l+K8nhXbY3JXn0\npDMCAAAAAAAA86uPMzmfN/T4L5I8fZkG5z201v4wyVOS3NU99ciqetCE8wEAAAAAAABzbOJnciZ5\n1NDjl7TW7lp2zyW01v6mqt6Z5LlJtmRwJufnJhGsu1fo6Um+K8kjuuVx3eaHtNbaKuMPyeA+o89O\n8sAkdya5MsnbW2tvW0P9ZyX5qSSnJTk0yReS/FGS17fW9qwy9hFJXpbB5X2PS/KVDO57+rrW2udX\nGbsjySuTnJnkvkluTPK3SX6ttfZXq4wd63sGAAAAAACASevjTM4Tu+VVrbV/2uAcFw89PmHMPMOe\nmOQ9SX4hyf+XbzQ4F1YbWFVHJbk0yeuSnNqNuVeS70nylqq6qKq2rjD+rUn+MMnjkhyRZG+SB3dZ\nPl1VJ64w9nlJPp7kh5J8c5JbM2hW/ng39vErjD01yWeTvCTJ/ZPclsH3fWaSS6rqFX19zwAAAAAA\nANCHPpqcN3XLL48xx/DlbW8eY55RCxmcAfneJL+U5CfXMfZtSR6W5PokZ7bWjszg3qHPT3J7Bk3D\nX15qYFX9VJKfSLIvg7Mxj2itHZ3kMRmczfmAJO9aZuypXe2tSX4/yQmtteOSfGuSS7oMf1xV37TE\n2EOTXJRBU/Pvk5zSWju2W39DBmfKvqaqzpj09wwAAAAAAAB96aPJ+cVu+c1jzLHYsFtI8n/Hi3M3\nf95a29Fa+4HW2quTfGAtg6rqoUme1eV5QWvtfUnSWrurtXZBkp/vdj2nqraPjL1XBg3VJDm/tfbG\n1trebvzHkzyjm/cxVXXmEuVfncFlhT+Z5Hmtteu7sV9M8swMjs8xQxmGvSjJ/TJoFP9Aa+2qbuzN\nrbWXZ3BW65Ykr53k9wwAAAAAAAB96qPJ+Wfd8uSq+pYNzvHUbnlrkhXvGbke670/6JBnd8trWmsX\nL7H9rRmcwXpoBo3HYU9Msj3JXRmcPTma6dP5RrP1OcPbquqYJE/pVt/YWlsYGXtLkjd3qz+6RK7F\n+S5srS11Zu2vdsuHVlWNbBvnewYAAAAAAIDe9NHkfHsGZw4elOT89Q6uqkckeW63+luttVsnmG2j\nFu95+ZdLbWyt3Z7kI93q6cuM/ewyjcYkef/Ivou+N4OzOBeWqz00dkdVnbT4ZFUdmeThI/uM+kQG\njcotSZ6wTO6NfM8AAAAAAADQm4k3OVtrX0ryY93q06vqwu6MxFVV1VMyaMhtS3JZkldMOt96VdWW\nJA/pVq9YYdcru+VJI8+fvI6x26vq+CXG7myt7V5m7FVL7D+cY2G52t2ZodeM7D+J7xkAAAAAAAB6\ns62PSVtrf1xVT05yQZIfSfKkqnpnkr9I8pkk1ye5I8mRSR6Q5JEZXG71sd0UFyZ5cWvtjj7yrdNR\nSQ7LoFl47Qr7LZ6leeLI84vraxmbJDsyOD5rGttau62qbkxydDd2tO5qtRe3De8/7vcMAAAAAAAA\nvdlQk7Oq7sqgAbaaLd3yuCQv6b5Gx23JPT07yY9WVVprWzeScYIOH3p82wr7LV5W94hlxq9l7Oj4\ntYxdHH/0MmPXWntSYwEAAAAAAKBX45zJuVRzcpLjtmRtjVQAAAAAAABgE+nlcrUTtNFG6iTdMvT4\n0BX2O6xb7llm/FrGjo5fy9jlao/mHl5f79j11F237duPHGf4xOaQ4cDIMO36MsgwS/VlkGHWMky7\n/mbJsO3g8S6Esu3grWNnnHaGcetPIsNabIb3owzzk2Ha9WWQYdYyTLu+DDLMWoZp15dBhlmqL4MM\ns5Zh2vWnmWGjTc7TJ5piebNwJudNGVyW9bAk91lhv8VtXx55/tokp61x7Oj4Ly2x/W6q6tAMLlW7\n3NgkuW+Stkrt4bHjfs8AAAAAAADQmw01OVtrH55wjpnVWluoqiuTPCLJKSvsenK3vHLk+SuSPHmN\nY3e11r429PziXDuq6riRbaNjF0ZqX90ttyT59izR5Kyqg5I8eDT3BL7nddm16+YNj13864Bx5hiX\nDLORYdr1ZZBhlurLIMOsZZh2/c2W4c69+8YeP27GaWcYt/4kMqxkM70fZZj9DNOuL4MMs5Zh2vVl\nkGHWMky7vgwyzFJ9GWSYtQzTrj8LGQ6aStX586FuecZSG6vq3kke261+cJmxJ1fVjmXmf9IyYz+a\n5M4MGpVPXGXsta21xcZmWmt7knxypdxJHpnkqAwapMvl3sj3DAAAAAAAAL3R5FybP+iWD6mqpy6x\n/YUZNAtvTfKnI9s+mOQrSbYm+bnRgVV1WgYNzIUk7xze1lq7Kcl7u9WfraotI2MPT/LikYzDLuyW\nz1mmwfqybnlZa+1zI9vG+Z4BAAAAAACgN5uuyVlV37T4leTYoU3HDm8bbii21j6d5F3d6juq6snd\nXFur6qwkr++2ndda++pwvdbaHUl+qVs9p6p+tqoO6cY/KoMG4ZYkH2utvW+JyL+YZG+S7+5qH9+N\nvV+SP0nyLUl2D2UY9pYkX0hyZJKLq+qkbuyRVXVukmdk0Fx91ejAcb5nAAAAAAAA6NOma3JmcFbl\n4tdl3XNbknx8ZNu3jIx7Ybf/8UneW1W3JLklyTuS3DvJn2fQkLyH1tqbk7wtg+P9P5Lsqaqbk3ws\nyf2TfD7JDy0z9vKu9p1JnptkV1XdkOSfM7iU7J4kP9hau36JsbcneVqS65M8LMkVVXVjkhsyOIvz\nriSvbK19YKna43zPAAAAAAAA0JdtfReoqkOTfGeSkzI4c/KwDJqKa9Jae3UPsRbWO6C1dnNVPTrJ\nOUl+NMm3Jbkjgybg21trv7XK+BdV1QcyuLzsaUkOTXJVkj9O8vrW2i0rjL2gqq5I8vIM7oN5XJIv\nJrkkyWtba/+0wtjLq+qUJK9M8tQk902yK8nfZXAW5odWGDvW9wwAAAAAAAB96K3JWVUnJvmVJD+c\n5PANTrOQZKJNztbahs9eba3tTXJu97WR8e9O8u4Njr0syY9scOx1SX6m+1rv2LG+ZwAAAAAAAJi0\nXpqcVfWYJO9NclQf8wMAAAAAAACb18SbnFV1QpKLc88G578k2Znk1nVMt+7LygIAAAAAAAAHtj7O\n5Hx5kqO7x/syuMzpb7TWvtRDLQAAAAAAAGCT6aPJ+ZShx2e31t7aQw0AAAAAAABgkzqohzn/bbe8\nToMTAAAAAAAAmLQ+mpxbuuVVPcwNAAAAAAAAbHJ9NDm/2C3v3cPcAAAAAAAAwCbXR5Pzw93y5Kra\n2sP8AAAAAAAAwCbWR5PzN7vlUUn+Yw/zAwAAAAAAAJvYxJucrbV/SPKGbvWNVXXapGsAAAAAAAAA\nm1cfZ3ImySuS/E6SY5N8rKp+oaq291QLAAAAAAAA2ES29TFpa+2uJD9RVR9P8uYkv5Lk1VX1f5Ls\nTHLHOuY6vY+MAAAAAAAAwHzqpcmZJFX1zCQvT7K1e2pLkgd0X2u1MOlcAAAAAAAAwHzrpclZVWcn\n+fUJTLVlAnMAAAAAAAAAB5CJNzmr6pFJ/ufI059L8okk1yW5dR3TOZMTAAAAAAAAuJs+zuR8ab5x\nBuZ1SZ7TWvurHuoAAAAAAAAAm1AfTc7vG3r8g621S3uoAQAAAAAAAGxSB/Uw5zd3y3/U4AQAAAAA\nAAAmrY8m5+5u+cUe5gYAAAAAAAA2uT6anP/cLY/pYW4AAAAAAABgk+ujyfmn3fI7quqoHuYHAAAA\nAAAANrE+mpxvz+CStQcneUUP8wMAAAAAAACb2MSbnK2165I8P8ldSV5RVS+cdA0AAAAAAABg8+rj\nTM601v48yfcnuSHJW6rqkqp6VlX9m6ra1kdNAAAAAAAAYHOYeMOxqu5KstCtbumWT+i+Frp91jLV\nliQLrbWtk84IAAAAAAAAzK++zqrcss7nAQAAAAAAANakl8vVTpCmKAAAAAAAAHA3fZzJefoE51pY\nfRcAAAAAAABgM5l4k7O19uFJzwkAAAAAAACwaNYvVwsAAAAAAABwN5qcAAAAAAAAwFzR5AQAAAAA\nAADmiiYnAAAAAAAAMFe2TXrCqvp3SRYmNV9r7SOTmgsAAAAAAACYfxNvcib5UAZNzi0TmGshydYJ\nzAMAAAAAAAAcIPq6XO0kGpwAAAAAAAAA9zDr9+TULAUAAAAAAADuZuKXq22trblxWlVbkxyb5DuS\nPDPJjyU5NMnbk7y4tbZ30vkAAAAAAACA+dbHPTnXrLW2L8lXM7iP54eq6teTXJzkBUl2JHnqFOMB\nAAAAAAAAM2imLlfbWmtJnpLkX5M8uar+85QjAQAAAAAAADNmppqcSdJa+8ckF3SrP1dVUz3bFAAA\nAAAAAJgtM9fk7HygW25P8vhpBgEAAAAAAABmy6w2OXcOPX7A1FIAAAAAAAAAM2dWm5xHDz0+Zmop\nAAAAAAAAgJkzq03O7x96vHtqKQAAAAAAAICZM3NNzqp6XJKfGHrq76cUBQAAAAAAAJhB26YdoKq2\nJjkuyWlJfjjJ8/KNXK219qlpZQMAAAAAAABmz8SbnFV1V5KFDQzdMrK+L8lLxk8EAAAAAAAAHEj6\nOpNztGG5Xjcl+cnW2iWTCAMAAAAAAAAcOKZ+udohC0k+l+SPkvxGa+3aKecBAAAAAAAAZlAfTc7T\n17n/HRmcufnF1tpNPeQBAAAAAAAADiATb3K21j486TkBAAAAAAAAFh007QAAAAAAAAAA66HJCQAA\nAAAAAMwVTU4AAAAAAABgrmhyAgAAAAAAAHNlW98Fqup7kzw2yclJjklyWJItax3fWju9p2gAAAAA\nAADAHOqtyVlVT0ry60keNMY0CxOKAwAAAAAAABwgemlyVtWPJfmtCUy15jM+AQAAAAAAgM1h4k3O\nqvq2JL858vSXkvxNt7xlHdM5kxMAAAAAAAC4mz7O5PzpJAd3j29N8qIkF7bWNCwBAAAAAACAsfXR\n5GjQchoAACAASURBVHzC0OMXtNbe3UMNAAAAAAAAYJM6qIc579ctd2pwAgAAAAAAAJPWR5Nzcc6r\ne5gbAAAAAAAA2OT6aHL+S7c8pIe5AQAAAAAAgE2ujybnpd3ywT3MDQAAAAAAAGxyfTQ5f7tbHldV\nZ/YwPwAAAAAAALCJTbzJ2Vr7WAaNzi1Jzq+q7ZOuAQAAAAAAAGxefZzJmSQ/neR9Sb41ySeq6gk9\n1QEAAAAAAAA2mW2TnrCqfjHJQpLLknx3Bo3OS6qqJflYkp1J/nWt87XWXj3pjAAAAAAAAMD8mniT\nM8kvLvN8dV/rsZBEkxMAAAAAAAD4ur4uVwsAAAAAAADQiz7O5PzIBOdamOBcAAAAAAAAwAFg4k3O\n1trjJj0nAAAAAAAAwCKXqwUAAAAAAADmiiYnAAAAAAAAMFc0OQEAAAAAAIC5oskJAAAAAAAAzBVN\nTgAAAAAAAGCuaHICAAAAAAAAc0WTEwAAAAAAAJgrmpwAAAAAAADAXNHkBAAAAAAAAOaKJicAAAAA\nAAAwVzQ5AQAAAAAAgLmiyQkAAAAAAADMFU1OAAAAAAAAYK5ocgIAAAAAAABzRZMTAAAAAAAAmCua\nnAAAAAAAAMBc0eQEAAAAAAAA5oomJwAAAAAAADBXNDkBAAAAAACAubJt2gHmSVU9P8nvrLLbLa21\nI5cZf0iSn0ny7CQPTHJnkiuTvL219rY11H9Wkp9KclqSQ5N8IckfJXl9a23PKmMfkeRlSb4vyXFJ\nvpLk/Ule11r7/CpjdyR5ZZIzk9w3yY1J/jbJr7XW/mq13AAAAAAAADBJzuTcmL1Jdq7wdQ9VdVSS\nS5O8LsmpSRaS3CvJ9yR5S1VdVFVblytYVW9N8odJHpfkiC7Dg5P8QpJPV9WJK4x9XpKPJ/mhJN+c\n5NYMmpU/3o19/ApjT03y2SQvSXL/JLdl0CQ9M8klVfWK5cYCAAAAAABAHzQ5N+ZjrbX7LPP1oGXG\nvC3Jw5Jcn+TM7mzPw5M8P8ntGTQNf3mpgVX1U0l+Ism+DM7GPKK1dnSSx2RwNucDkrxrmbGndrW3\nJvn9JCe01o5L8q1JLuky/HFVfdMSYw9NclEGTc2/T3JKa+3Ybv0NSbYkeU1VnbH8oQIAAAAAAIDJ\n0uTcD6rqoUmelcHZmy9orb0vSVprd7XWLkjy892u51TV9pGx90ryS93q+a21N7bW9nbjP57kGd28\nj6mqM5co/+oMLkv8ySTPa61d3439YpJnJvm/SY4ZyjDsRUnul+TmJD/QWruqG3tza+3lSd6TQaPz\ntes7IgAAAAAAALBxmpz7x7O75TWttYuX2P7WJDdlcJ/NZ45se2KS7UnuyuDsybtprX06yQe61ecM\nb6uqY5I8pVt9Y2ttYWTsLUne3K3+6BK5Fue7sLX25SW2/2q3fGhV1RLbAQAAAAAAYOI0OfePxXte\n/uVSG1trtyf5SLd6+jJjP7tMozFJ3j+y76LvzeAszoXlag+N3VFVJy0+WVVHJnn4yD6jPpFBc3ZL\nkicssw8AAAAAAABMlCbnxpxSVVdU1W1VdXNVfbaq3lhV9x/dsaq2JHlIt3rFCnNe2S1PGnn+5HWM\n3V5Vxy8xdmdrbfcyY69aYv/hHAvL1e7ODL1mZH8AAAAAAADolSbnxhyf5MFJ9iQ5JIPm4M8kuaKq\nRi/7elSSwzJoFl67wpyLZ2meOPL84vpaxibJjvWMba3dluTGDM7GXGrsarUXt43mBgAAAAAAgF5o\ncq7Pl5L81ySnJLl3a217kiOSPDWDsykPTfK7VfXYoTGHDz2+bYW5b+2WR4w8vzh+LWNHx69l7HK1\nx80NAAAAAAAAvdg27QDzpLV2SZJLRp7bm+QvqupjST6V5IFJXpfkMfs/4Xzavv3ImZhDhgMjw7Tr\nyyDDLNWXQYZZyzDt+pslw7aDt449ftyM084wbv1JZFiLzfB+nIUMz33Bi/Ol67624fH3PeG4/N7b\n3zzBREub9msx7foyyDBrGaZdXwYZZi3DtOvLIMMs1ZdBhlnLMO3608ygyTkhrbWbquo1SX4nySOr\n6vjW2vVJbhna7dAVpjisW+4ZeX5x/FrGjo5fy9jlao/mHl5fbSwAAMBM+NJ1X8vhJz1n4+OveucE\n0wAAADApmpyT9XfdckuSb01yfZKbMrik62FJ7rPC2MVtXx55/tokp61x7Oj4Ly2x/W6q6tAkR68w\nNknum6StUns095rt2nXzRod+/a8DxpljXDLMRoZp15dBhlmqL4MMs5Zh2vU3W4Y79+4be/y4Gaed\nYdz6k8iwks30fpyFDNN+P65m2q/FtOvLIMOsZZh2fRlkmLUM064vgwyzVF8GGWYtw7Trz0IG9+Ts\nz0KStNYWMrhfZzK4l+dyTu6WV448f8U6xu5qrQ1fh2lxrh1VddwqY4dzJsnV3XJLkm9famBVHZTk\nwcvkBgAAAAAAgF5ock7WI4ce//PQ4w91yzOWGlRV907y2G71gyObF8eeXFU7lqn7pGXGfjTJnRk0\nKp+4ythrW2uLjc201vYk+eRKuTP4fo/KoEE6WhsAAAAAAAB6ock5IVV1VJKf71b/rrsf56I/6JYP\nqaqnLjH8hRk0C29N8qcj2z6Y5CtJtib5uSXqnpZBA3Mhyd1uFtNauynJe7vVn62qLSNjD0/y4pGM\nwy7sls9ZpsH6sm55WWvtc0tsBwAAAAAAgInT5Fyjqrp/VV1aVWdV1X2Gnj+kqr4/yceSPCjJviSv\nHB7bWvt0knd1q++oqid3Y7dW1VlJXt9tO6+19tWRsXck+aVu9Zyq+tmqOqQb/6gMmqJbknystfa+\nJaL/YpK9Sb67q318N/Z+Sf4kybck2T2UYdhbknwhyZFJLq6qk7qxR1bVuUmekUFz9VXLHDYAAAAA\nAACYOE3O9fmeJO9I8i9VdWtV7UqyJ8n7Mrhv5S1Jnt9a+/ASY1+Y5LIkxyd5b1Xd0u3/jiT3TvLn\nGTQk76G19uYkb8vg9fofSfZU1c0ZNFbvn+TzSX5ombGXd7XvTPLcJLuq6oYMLqd7Rpf/B0fOPF0c\ne3uSpyW5PsnDklxRVTcmuSGDszjvSvLK1toHlqoNAAAAAAAAfdDkXLudSX46yR8muTqDBuVRGTQJ\nP5nkdUlObq29c6nBrbWbkzw6g0va/kMGZ3zeluTjSX6ytfa01tpdyxVvrb0oyQ9ncI/OmzJ47a5K\n8t+SfGdrbecKYy9I8qgMzib9cpJ7Jflikt/uxn54hbGXJzklyf/MoJl6cJJdSS5OckZr7dzlxgIA\nAAAAAEAftk07wLzozmr8X93XRufYm+Tc7msj49+d5N0bHHtZkh/Z4NjrkvxM9wUAAAAAAABT5UxO\nAAAAAAAAYK5ocgIAAAAAAABzRZMTAAAAAAAAmCuanAAAAAAAAMBc0eQEAAAAAAAA5oomJwAAAAAA\nADBXNDkBAAAAAACAuaLJCQAAAAAAAMwVTU4AAAAAAABgrmhyAgAAAAAAAHNFkxMAAAAAAACYK5qc\nAAAAAAAAwFzR5AQAAAAAAADmiiYnAAAAAAAAMFc0OQEAAAAAAIC5sm3aAQAAADgwnf3Sc7Jz1+4l\nt207eGuS5M69+5Ydv2P7sXnT+ef1kg0AAID5pskJAABAL3bu2p2jTz1r4+Mvv2CCaQAAADiQuFwt\nAAAAAAAAMFc0OQEAAAAAAIC5oskJAAAAAAAAzBVNTgAAAAAAAGCuaHICAAAAAAAAc0WTEwAAAAAA\nAJgrmpwAAAAAAADAXNHkBAAAAAAAAOaKJicAAAAAAAAwVzQ5AQAAAAAAgLmybdoBAID5dfZLz8nO\nXbuX3Lbt4K1Jkjv37lt2/I7tx+ZN55/XSzYAAAAA4MClyQkAbNjOXbtz9KlnbXz85RdMMA0AAAAA\nsFm4XC0AAAAAAAAwVzQ5AQAAAAAAgLmiyQkAAAAAAADMFffkBIA5dfZLz8nOXbuX3Lbt4K1Jkjv3\n7ltxjh3bj82bzj9v4tkAAAAAAPqkyQkAc2rnrt05+tSzxpvj8gsmlAYAAAAAYP9xuVoAAAAAAABg\nrjiTEwAAAHqy0uXlk7VdYt7l5QEAAO5JkxMAAAB64vLyAAAA/dDkZOat9JfP/uoZAAAAAABg89Hk\nZOaN+5fP/uoZAAAAAADgwHLQtAMAAAAAAAAArIcmJwAAAAAAADBXXK4WAGBM7h8NAAAAAPuXJicA\nMNdmocHo/tEAAAAAsH9pcgIAc02DceVGb7J/mr3jNpsnkQEAAACAzUOTEwBgzo3b6E3Gb/bOQgYA\nAAAANo+Dph0AAAAAAAAAYD00OQEAAAAAAIC5oskJAAAAAAAAzBVNTgAAAAAAAGCubJt2AAAAmISz\nX3pOdu7aveS2bQdvTZLcuXffsuN3bD82bzr/vF6yAQAAADBZmpwAABwQdu7anaNPPWvj4y+/YIJp\nAAAAAOiTy9UCAAAAAAAAc0WTEwAAAAAAAJgrLlcLABsw7r3/Evf/AwAAAADYKE1OANiAce/9l7j/\nHwAA+9+4f6znD/UAAJgVmpwAAAAAm8S4f6znD/UAAJgV7skJAAAAAAAAzBVNTgAAAAAAAGCuuFwt\nAHNnpfsIJe4lBAAAAABwoNPkBGDujHsfocS9hAAAAAAA5pnL1QIAAAAAAABzRZMTAAAAAAAAmCua\nnAAAAAAAAMBc0eQEAAAAAAAA5oomJwAAAAAAADBXNDkBAAAAAACAuaLJCQAAAAAAAMwVTU4AAAAA\nAABgrmybdgAAAACAvp390nOyc9fuJbdtO3hrkuTOvfuWHb9j+7F50/nn9ZINAABYP01OAAAA4IC3\nc9fuHH3qWRsff/kFE0wDAACMS5MTAAAmxFlCAAAAAPuHJicAAEyIs4QAAAAA9g9NTgAAgAOQM4sB\nAAA4kGlyArBu/tEUAGafM4sBAAA4kGlyArBu/tEUAAAAAIBpOmjaAQAAAAAAAADWQ5MTAAAAAAAA\nmCuanAAAAAAAAMBc0eQEAAAAAAAA5sq2aQcAYH3Ofuk52blr95Lbth28NUly5959y47fsf3YvOn8\n83rJBgAAAAAA+4MmJ8Cc2blrd44+9ayNj7/8ggmmAQAAAACA/c/lagEAAAAAAIC5oskJAAAAAAAA\nzBVNTgAAAAAAAGCuaHICAAAAAAAAc0WTEwAAAAAAAJgr26YdAAAAADiwnf3Sc7Jz1+4lt207eGuS\n5M69+1acY8f2Y/Om88+beDYAAGA+aXICAAAAvdq5a3eOPvWs8ea4/IIJpZmecZu9Gr0AAPANmpwA\n6+AfJQAAgI0at9l7IDR6AQBgUjQ5AdbBP0oAAAAAAMD0HTTtAAAAAAAAAADr4UxOAACACRv3EveJ\ny9wDAADASjQ5WZOq2pHklUnOTHLfJDcm+dskv9Za+6tpZgMAgFkz7iXuE5e5BwAAgJW4XC2rqqpT\nk3w2yUuS3D/JbUmOy6DheUlVvWJ66QAAAAAAANhsNDlZUVUdmuSiDJqaf5/klNbasd36G5JsSfKa\nqjpjeikBAAAAAADYTFyultW8KMn9ktyc5Adaa19OktbazUleXlXfluTpSV6b5JKppWRTcG8rAABY\nP5+jAQCAA5EmJ6t5Tre8cLHBOeJXM2hyPrSqqrXW9l80Nhv3tgIAgPXzORoAADgQaXKyrKo6MsnD\nu9X3L7PbJ5LclOSoJE9IoskJAAAAsIJxz7B2djUAgCYnKzupWy4kuWKpHVprC1V1TZLvGtofAAAA\ngGWMe4a1s6sBAJKDph2AmXbi0ONrV9hvcduJK+wDAAAAAAAAE+FMTlZy+NDj21bY79ZueUSPWZiy\nlS6lk7icDgAAADB/XDqYRd4LAPNHk5Ope/rTn7Li9iuuvDpbr77m6+uP/qH/vq75L//UR1atMew9\n73nfPZ5b6UPOZ/7+b5IkC3ctLDvnIYdsy4Me+MBl51/JWrIf3H3Q2rt3Xy/zJ93rcMQJ6z7+l77r\nF77++PJPXZenf+GaFfZe//EZnn8tvv0hD17X/qPHZ/T9OGq9x2c9781kMsdn357rlq3b9/Gf1PFZ\n7nWYh+M/bPS1WO/8l3/qIyu+H0f1cXyGX4uN/HxY6f04qq/352KGcX5+rvazIen3/XnFlVfnseu8\n3Nno8VnttZiHnw8rvQ7z8vNhuddhHo7/sFn4/LaStRyf4ddiIz8f1vJzYVFf78/FDBv9/LbWn9F9\nvj/37bluXXMns/f5beth37yu/X1+29j8y5nk8VnqdfD5bWXr/f368k99JFsO2pJkbb9jz+vv18n4\nn9/W8jqsln/0tbj8Ux/5+uO1vA5PevLT1tXQWSrv5/7xH3PHHXcuuf/DHvX4JGtvKm3058NaLx28\n3M+H5f6tY1Lvn+H3ZB/zL2fW5n/c4x53j2Mwyfm//vntiBPWtP9SPz9Xuoz0vB//xfmXez9Oav7V\nLNZ/97v/vJf5F83q8V+02usw6/kPpPmXei3mKf8k5n/c4x6XZPWfCxudfzWanKzklqHHh46sDzus\nW+7ZSJFLL/3o6jvd8NWNTJ0kufGGr66tRmf79iPv8dxXb7hp2Q+8N3zgT9Y0766v7Fx2/pWsJ3uS\n/PDzfnzJ5+97wnH5vbe/ebz5N/A6fO1f7n4719XqrXR8Fv9qbqX5V7PtO05e12uwZN4x3o/bDt56\nt/rrfX23bz9yyeOwnOWOz3J113psFjOs9/gPj19LrRWPzxivw2KGv9nA8R8ev5q1HJ/h73G9Px9u\nvOGrEzkOy9Vd8/tzgxkWj89a6yyXc7nXYj3vz0sv/ej4P58n+LNhyfknbKnjs1LN1Y7P6OuwkZ8P\na/3ZkKyQdUKvw0Z+Pi/OsRYrHZ+laq/n/bnt4K0b/vm8OH61erP++W2112Gtx2cxw4Z/PvT0M3p/\nfn5bS62+P79N5PPzFD+/Pf7Jz1rX/n18fht+HXx+W9lqx2f0+9vfn99Wex3m4fPbSr9f37jG36+T\nwe/Y0/r9Oln6d+z9/fltnN+vk3u+Fus5/ovjx/79egWHn/Sc1TNc9c65//z23Be8OF+67mtjvT+X\n+zef1TKudf612OjxX6u//uu/7nX+SXx+m8jv151pHf/F92Nf8y9n0sdn9L+J9c7/s//55Useh+X0\ncXyGX4uNHJ+1/FxY1Nf7czHDOMdnuffksFl7f05j/ol8PlxhrpVeh75/Pq9my0Rn44BSVY9I8ndJ\nFpKc1Fpry+z3t0m+K8n/aq399H6MCAAAAAAAwCZ00LQDMNOu7pZbknz7UjtU1UFJFq//eeX+CAUA\nAAAAAMDmpsnJslpre5J8sls9Y5ndHpnkqAzO9vzg/sgFAAAAAADA5qbJyWou7JbPqaodS2x/Wbe8\nrLX2uf2UCQAAAAAAgE1Mk5PVvCXJF5IcmeTiqjopSarqyKo6N8kzMjiL81XTiwgAAAAAAMBmsmXa\nAZh9VXVqBpeiPb576uYkR2Tw/rkryataa+dOKR4AAAAAAACbzNZpB2D2XX/99dcdf/zxFyQ5JMk3\nJTkmye4kH0ryn1pr75xmPgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGBytkw7AMyDqqokT0hyUpITkxzRbdqT5MtJrkzywdba52To\nN8O0OQYDjgMAAAAAANOkycnMm2YzparOSPLaJA/N6v+9LCS5LMmrWmsfkGGyGbocm/q9MJTFccj0\nG60zUH97kmcmOX21DEn+tLX2VRlk6CvDtOvLIMOsZZh2fRlkmKX6MsgwaxmmXV8GGWYtw7TryyDD\nLNWXQYZZyzDt+rOSYSWanMysaTdTquoVSV4zVPumJNckuTbJrd1zhyW5T5IHJzlqKMsrW2vnyjCx\nDJv+vdDlcBwyE8dh2vW3JvnlJOckOXSNw25Ncl6SX2yt3SWDDJPKMO36MsgwaxmmXV8GGWapvgwy\nzFqGadeXQYZZyzDt+jLIMEv1ZZBh1jJMu/6sZFiLbfujCKzXBpspj0jy/qoau5lSVacP1X9Pkl9N\n8onW2sIy+x+U5JFJXpbkGUleU1WfbK19SIaxM2z690I3r+OQmTgOU63f+dMkZw7V/0gGfy315SUy\nnJzksV2OX0hyapKnySDDBDNMu74MMsxahmnXl0GGWaovgwyzlmHa9WWQYdYyTLu+DDLMUn0ZZJi1\nDNOuPysZVuVMTmZO10y5JBtvptyV5IwxG2sXZfAf8Btaay9f59hzuywXt9b+vQwbz+C98PV5HIdM\n/zhMu3435wuTvCXJ7Ul+PslbW2u3rzLm3klemOT1Se6d5Cdba78lgwzjZph2fRlkmLUM064vgwyz\nVF8GGWYtw7TryyDDrGWYdn0ZZJil+jLIMGsZpl1/VjLA3Kqqi6rqrqr61Q2MPbcbe9GYGa6rqr1V\ndeQGxh5ZVfuqaqcM42XwXvj6PI5Dpn8cpl2/m+fSbp6zNjD2rG7sx2SQYRIZpl1fBhlmLcO068sg\nwyzVl0GGWcsw7foyyDBrGaZdXwYZZqm+DDLMWoZp15+VDGt10P4oAuv0yCT7krx6A2N/JYN74H33\nmBmOTrKntXbzegd2Y/bkG/cjlGHjvBcGHIeBaR+HaddPBpd++Nckv7+Bse9Mckc3hwwyTCLDtOvL\nIMOsZZh2fRlkmKX6MsgwaxmmXV8GGWYtw7TryyDDLNWXQYZZyzDt+rOSYU00OZlFs9BM+VKSo6rq\npPUO7MYc2c0hw3gZvBcGHIeBaR+HaddPBvfSvrNt4MbdrbV9GTRpt8ogw4QyTLu+DDLMWoZp15dB\nhlmqL4MMs5Zh2vVlkGHWMky7vgwyzFJ9GWSYtQzTrj8rGdZEk5NZNAvNlIszuO/e71XVieuovyPJ\n73Wr75Vh7AzeCwOOw8C0j8O06yfJPyY5vKqesoEMT05yaJLPyyDDhDJMu74MMsxahmnXl0GGWaov\ngwyzlmHa9WWQYdYyTLu+DDLMUn0ZZJi1DNOuPysZ1kSTk1k0C82U1yb5WpKHJbm6qn6zqp5eVQ+q\nqsOr6qDu6/AaeEZV/UaSa7ox13dzyDBeBu+FAcdhYNrHYdr1k+TCbvm7VXXmOjI8NckF3eofyCDD\nhDJMu74MMsxahmnXl0GGWaovgwyzlmHa9WWQYdYyTLu+DDLMUn0ZZJi1DNOuPysZ1mTL/igC69E1\nBD6b5LgkN2fwH9T7k1yR5Nokt3W7Hprkvkm+PckZSZ6TwZlS1yc5pbV23Zg5Tk1yUZL7DT29sMzu\nw/8tfSHJ01prl49TXwbvhaH6jkOmfxymXb/LcEiSSzNoHC9k0ET+y2Uy3CfJKV2Gh3TPX5bk0a21\nvTLIMG6GadeXQYZZyzDt+jLIMEv1ZZBh1jJMu74MMsxahmnXl0GGWaovgwyzlmHa9Wclw1ppcjKT\npt1MGcpxaJIXZdCkePgqu38yg6bHW1prt0+ivgzeC0P1HYdM/zhMu36X4agkb0vyrJFNozlG/x//\nriQvbBu4p6gMMsxqfRlkmLUM064vgwyzVF8GGWYtw7TryyDDrGWYdn0ZZJil+jLIMGsZpl1/VjKs\nhSYnM2vazZQl8hyZ5KQkJyY5vHv6lgzusXd1a21PH3Vl8F4Yqus4ZPrHYdr1h3I8NMmzkzw+g7+S\nOmxkl1uTXJnkQ0n+4P+1d99hklRl+8e/G8hLkkUyLAiPgqIk4VWRoMArOYgCgoIEAUFRAVHwVRF/\ngCAiQRFEEARMoCiIghIVUcAlLuFRYUmSl7BLlp3fH89ppqa2u6e7q6erZvr+XNdc011dVeeenp6n\nq+ecOuXut3azfWVQhiq1rwzKULUMZbevDMpQpfaVQRmqlqHs9pVBGaqWoez2lUEZqtS+MihD1TKU\n3X5VMjSjTk4ZFarQuSfVoNdC0PMQyn4eym4/k2McsFAuw/Pu3ugsU2VQhjHbvjIoQ9UylN2+MihD\nldpXBmWoWoay21cGZahahrLbVwZlqFL7yqAMVctQdvtVySAiIiIiIiIiIiIiIiIiIiIiIiIiIiIi\nIiIiIiIivaLpakW6zMx2J118193PVYbyMpRNz0HQ8yAiIiIiIiIiIiIi3Tax7AAi3VSRzpSz0/cB\nQBlKyqDXQtDzEMp+HspuP2U4O5NhT2VQhjIzlN2+MihD1TKU3b4yKEOV2lcGZahahrLbVwZlqFqG\nsttXBmWoUvvKoAxVy1B2+2VkUCenjDWld6ZkVOFM6X7OoNdC0PMQyn4eym4fYPdMhlIOcpRBGSrU\nvjIoQ9UylN2+MihDldpXBmWoWoay21cGZahahrLbVwZlqFL7yqAMVctQdvs9z6BOThmryuxMWanE\ntmuUYVC/vxZq9DyEsjv+y2z/wfR9QBmUoQIZym5fGZShahnKbl8ZlKFK7SuDMlQtQ9ntK4MyVC1D\n2e0rgzJUqX1lUIaqZSi7/Z5nKPufvSJdZWZTarfdfXp5SaRsei0EPQ+h7Oeh7PZFRERERERERERE\nRERERERERERERERERESkRDqTU6TLzGwcsByAuz84zOpjgplNAtYElgKeAP7m7i+Xm6q3zGwxYEq6\nO93dny4xjoiIiIiIiIiIiIjImKZOTqk8MzPgg8CqRCfapPTQLOBR4C7gSnf/ZzkJhzKzyURH34C7\nT+jC/pYA9gXWAiYAdwLnuPs9w2x3IzDZ3QtdD9HM5geOAnYmnvubgEPd/Zb0+D7AMcCiDNaUF4Bj\n3P3oIm13kHMa8bz37BqQZvZ+4FhgPYbW1BuBL7v7NT3IMAHYgfg7WZaY73w6cJm7/75LbbwfuNPd\nn+nG/ooys4WB+d390TqPrQZsDawAzAT+DvzW3f/b5QwLAOsDbweWBBYEZgPPAv8EbnD3e7vZpkgR\nZrY89M8AHBGRVqk+iojUp/ooIlKf6qNIdaiTUyrLzDYlOs/WZPjX6gDwD+Bwd//TSGdrppudnGa2\nEXAxsFDuodeBU4HD3P3VBts+BizehQx/JDrPsp4D1gHWBn7WZPMT3f3gIu23KnU4zaRLncuZegxT\n9gAAIABJREFU/c4EHnP3Veo8tgtwDjCxweavA3u4+/kFM2wA7Ab81d1/nHtsNeAi4K11Nh0A/gbs\nWK8zsM0Ms4FXgEuJn/kyd59dZJ8d5lgROBPYKC16HDjE3S9Ijx8NfBEYn9v0fmAHd7+tCxlWAr4B\nbA/MS+P6NADcSnT4X1i03To5Fic6tz/AMINAgF+7+1PdztAuM5sb2IX4Oz23xBzLAuO79WHEzFYh\nnvt73P2l3GPvJWpo7Uz3K939z91ot82Mk4Dn6XKNbKHdhYmff8W06H7gKnd/tkftrwJ8jvg7WY7M\nIBDiPeqxLrTxVeIY5Pdl1MVcloWAjYH5gVuyA6LMbCKwN7AtQweBfH+4gVMd5JgAbAhsCqxGvP7n\nGAgCXDhSMx+Mthqp+th/9TG1XVqNVH1UfWSU1EeoRo1Ufeyf+pja75saqfo4JIPqY+c5ulYjVR+H\nbVv1sYeqUCNVH4enTk6pJDM7DDiawdfo88C9wH+AF9Oy+YGlic6dWifgAHH23HG9SztUtzo5zWxJ\n4szERYGXic6qGcD7gCXSajcDW7r7k3W2L9zJaWY7Axeku+el9t4LfBT4EXEW21uB7wCnEL+fFYFD\ngb2IYruuu0/tsP3did9pK+YFfpBu75F9oMiBXurce8zdl84tn0L8fuYD/kW8Xm9MD68HfAlYhTir\ndTV3f6hAhh8BnwQ+4e7nZZYvTnSiLZUWefqC+L3UOmanEb+HIQeHbWbIH1Q8Trw2znH32zvdb5sZ\nJgF3EAcOWbOBbYA3A2elZTOA+9K6i6dljwGrF3nDN7OtiY79+Rqs8l/gGeLvttb5PQD8AtjN3V/v\ntO1MhgnAkcDnm+TIexE4EfhayR+gu32m++ZEvcme6X42cGazn9PMHgXe3IVBIKsQr4c10qJZwMHu\nfmaauvwnwMdymw0AfwI+4u7PF2m/zaxdHwhiZmsDr7r7HQ0e/zJwOPF+nT3mfBH4lrsf1YUMuwNH\nABe4+9dzj+1EvB7mbbD5DODD7n5twQyzid/rE8D5wI/d/c4i++wwxxbEe/XCxPM9APzQ3fdL//z4\nI/D+Opu+Buzt7j/pUo6diRkOlmP4zxqvEMcTX3L3WV1qf1TWSNVHYAzVx7TfUmuk6uOQHKqPjN76\nCF0fSKz62HrWMVkfUxuqkag+ZtpXfRzcX2k1UvVR9THTTun1MeUovUaqPrZGnZxSOWb2AaJIjCPO\nYjyeuMZj3c4uMxtPdCodQpxZNRvY1N2vLpDBOt2W6Ny4geKdnMcAhxEjcrbwNO1lGiVyAHGW67xE\np9Ym7v5wbvtudHJeCmwBHO/uh2WWfw/Yh+jEOcvd966z7fnEaLLT3P2ADtvvRiEs+nto1Mn5HWLk\n0l3Ae/MHU2mkz/XEVKbHuvvhBTLcDRiwtLs/nll+HPG6fwLY090vy223JfHm9mbiDa7jzv/0PLxE\njA56Z+ahAeA24uzO80dypI6ZHQ58kxil9HlgKnE28YnE72EhYCVg31xn8NbAj4m/zW/kD9LaaH8V\nolN5PuLv7hRi8MVEYiTVZ4Hlga8AxxHP0y7AgcA8xAHibp20ncvxW2CrdPd54Dri53+UOQeBrEYc\ncNUGglzi7tsWzdCpLv+D6gvAtxs8/A9gZ3f/d4Ntu1EfFyA+8NXrdN+UGGX31bTsQeAR4rq9tUEJ\nV7n7Jp22nzKcTesDQeYCdk23f5x9wN337LD9uvUxPXYicFBmUa02TM4sO9ndP9dJ25l2LiLe+z/k\n7ldklr+DeB3MRXw4voDBQSBvIz4cz0/Uk9Xd/ZECGfLvVbWzuGt1ccSv0WxxhvmdDH7YrA22ADgY\nWIz4MPwy8Dvi2GIF4EPECNBXgTXd/e6COb6Z2qkZYOjnjddS+28iauQiafk/gY284KwDKcOorJGq\nj2OrPqYMpdZI1cc3Mqg+DmYYlfURujqQWPVR9bHWTt/XSNXHIRn6vj6mfZVWI1Uf38ig+kj59TFl\nKL1Gqj62rtEUiyJl+hzxB3uCux863MppRMANwIcznT6fBzru5ATuofU3tpGyRfp+gGeu6+dxXcGT\nzOxqohPYgD+b2SaNDjYKWIt4Hk7JLT8d2D899q0G2x5PdPCs3+VM7RqpwRybpe+H1hst5u7Pm9mh\nxHQN+el+27U08HK2gzPZJn3fw93/UCfD78xsj5ThI0THWxHPufsaZrYGsDtxALM4McpuDeA4M/s9\ncdBxiXf5GpjE9A8A+7v7z9PtO8zsJeCn6f6Xsx2cAO5+iZkdTJzluSXw9Q7b/yLRwXkRcXCfPSvz\nD2b2feLg4ijgRo+ps6ea2XnAVcAuZnaOu/+xw/axuAbuVsRB1JeAM9z95WG2mZcYlPAtYGsz29vd\nz+w0Q0FdqavpNVirPXcDPydGDG5IHIyvDdxoZlu7+1+70WYd+xMHsLOI963ame4nEANU1iLO5N7F\n3S/NZN+VeC1+wMy2cfffFsiwe4fb7ZG5PQB0/CGsHjN7H4Mfvn5JTK0+PT22InHm+07AZ8zsfHe/\nqUBzaxI/ww255V8kPnz9HdjW3Z/IZfw/4j10PWKkcqEPgsQ07ucSdXGxlGtNhtbFS0egLtZ8jvjw\ndTuwnbtPT8/1xcRAi/mBh4EPeuYa5hbTSv2BwYEa+3cawMw2IT6ADRAf9L9DzHQwnpjO5jDivWge\nYrqf8cQHwOOJD8W/M7N1ioz0HOU1UvWxD+oj9LRGqj4G1UdGfX2ELtRI1cc3qD4G1UjVx1qGvq+P\nUIkaqfrYhOpj/x1Dqj62R52cUkXrEdcy/EYH2x5FjKZYtws5yj7TeSVi6su61xh199vTm9wVxNmC\n16WOzkKj6HImE51rD+eW35e+v5ot5Dm3E6NW8qOw2vEIsAwxVe+n3f3WRita7+fDX4EYMdPsGrBX\nEb/DOa7n2aaJaT/1MrwIXN5k2yuIMzDrXbOzI+n3cGvqxN2cOKDbijjY2SZ9PW1mPyWms/1Hl5p+\nKzGK71e55b9m8MC+0bUvLyQOfFcu0P6mqZ3PeZ1pZ9395fSc3Eycbf2ntPw2i7NQTyOmce64k5OY\nthhgP29xGuZ0AHKKmT1HHBh9kriuaUfMbF86/yA1afhVWvJZYuqcq4kpu2sHWaea2VrEdCZvBS43\ns+19ZK7VXOt0/4K716ZJvt3M5mdw9OsR2Q9gAO5+vpmtDHyNOFgv8iGs5h5iCulm71sTiOnOIUbe\n1YzEgJ790vdL3H2n7APufj/R4T8v8Rx+CijyAWxJ4Hl3n5lbvjHxs+2d//CVcjxuZnsTU2BvTvEP\nYC+5+0FmdggxSGl3YlDF3MTPuS3wVKYudjSNexO1Ucv71T7suvv9ZrYfMasAxD8Ehrxnu/vDZnYg\n8X61ccEMn0nfv+bu38w9NhXYycweJz4Q7u3uPyQ+eP0lZVyD+IdAkQ9ApdZI1cc3qD4216saqfoY\nVB+DjiFVH/P6uT6CaiSoPtaoPoaya6TqY3Oqj/13DKn62AZ1ckoVLQzMqlNMh+XuM81sFoOnRXdq\nBnGq96do3olVz6LE6ftFzQ280GxEirs/amYbEp1cawPXmtlmzToD21T3Go7uPstiRt9nmmSbbWYz\nKfa7WJWYnvRAYsTYacBXGrw2en3m7UTiTf+1Riu4+6upqBd9PT4MrGxmy+Y6nF8EJniDqZxThtlm\n9joxoqer0mvzEuASM3sTcebu7sA6xCirA4EDzOwuYoTVeV7sAuQLUOc5T8/zTGI6iP80yDor1Yb5\nC7S/FPCMN59yo3bdhLfnlv+M6OR8T4H2IUaCvUJcE6Bd5wNnpH0UcVrB7bthQwY7nIeMInP3qWa2\nLjF1ypbE63Nnd/9NlzOsljL8PLf818SHsAFiuuh6ziY+hK1TMMNviUEFSwHfJy5uX3eUXm4gyEYF\n2x3O+sTP//Um63yD+FDyvibrtOIVYuRi3puJs8+nNdrQ3aeZ2fPEtS26ItWn3wC/MbPFiA/auxMj\njycTH1QONLNpDNbF/Fn6nViBGHj0t9zyvxMDciYCja6Lcj0xuGzZghnWIwbkfLfJOicQ7w0fA34I\n4O7PmdkXgUuJ0cdFPgCVXSNVH4PqY3O9qpGqj0H1MZRdH6H8Gqn6GFQfg2qk6mON6mMou0aqPjan\n+th/x5Cqj23o+j+9RbrgEWAhM1u13Q3TNgumfRRRG/GytLtPb+cLeKhg2zWPE8/Dws1WcvcZxOiS\nG4jifmU6+OiGJ4FJaeRUPcOd7ToJ6HiedHef5TGX/LrEdR8/A9xrZvkLjbeSpdvuAxa1uEZqMwsQ\nBz5FXEH8fJ/OLf8r8Rp5d6MN02OTiLnhR4y7z3D377n7ukQH3/HE/Ozj0v3jKP638TywsJktmF1o\ncf3TWkfyW+ptaGaLE8/DcwXanwUsaGb1DvZqFq+30N2fI65L8OYC7UM6q7fRgXYzHmefvk6MOOyG\nx4hrYbTzlT8rvFNLE6P67qj3YBoIsT3xIWwe4BcWF2vvpoWoP7qx9v4zq97oxpTvQWKQwpJFArj7\ndsAOwEzgZOAmM1uvweq9HAiyJDELwC1N1rmVmPKk6D9G/g3MazG9UtZTaXnDY10zm0C8PppOt9Ip\nd3/a3U9x93WA1YkP548RdfEdRJ18yMwuNbOPFGxuLmL2hHyG2cQHMBi8Xkbef9M6RWvDosTfxKwm\n69QGokzJLf8jUZ9WL5ihKjVS9VH1sZle1UjVx6D6GKpSH6G8Gqn6iOpjhmqk6mON6mMou0aqPjan\n+kjfHUOqPrZBnZxSRZcSxeknZrbUcCvXmNmSwE/S3d8VzHBz+t6w86gHbiOeh2Gv55g6UDYDriGK\n4BUUP3sQYq7vccR1P/OWJ+ZCr8vMliHe2Apf5NhjyoH1iGutTgLOM7MrzextRffdosXM7KrsF9GZ\nNY4mU9Ga2dLE/O1PFmz/VOLN8VAz2yuz/P8R07eebWZz5LA43fbsdPfighla5u53u/thxGtkc+Is\nxpcp/sZ2B/Gc5+ez35/B6Xy/0GDbz2f20ampxEHOPk3WqU0hck92YTrQW5A4WC7iX8ACZrbFsGvm\nmNnmxDVFi16794H0/QvuPqWdL2KkXTcMMMyHCo8zjT9BjGabi3hP+WSzbdr0HHVGN7p77SD4hWG2\nf5kuHIe5+8XEyLiTgHcB15vZGWa2aG7VXg4EeY0Y7deQxxnoLxKzFhRRGz38tdzy3xH198NNtt2B\n+B12c5r3utx9mrt/kRjRuiXwC+I5mkhMvfOzgk08QdSG5bMLzWwFBs9gf1eDbVehO+9VTwGLmNnk\nJuvU3quGfFBLo3NnAosUzFB2jVR9DKqPzfWqRqo+BtXHUHZ9hPJrpOrjYHv9Xh9BNRJUH2tUH0PZ\nNVL1sTnVx5w+OIZUfWyDOjmlio4hpotdC7jHzE4zs+3MbBUzW8DMxqevBSxsb2bfB+5N2zyd9lFE\n7UzOIp2cRd/wrkzf92hlZXd/gSjqlxMdnPMWbB8Gn4cN6rT3sDefenSz3D4KcffZ7n4SMYXtxcS8\n5rea2TFmNl832mhiLmCj3Ndk4nfc7M29Nn/7bUUad/d7iYt2TwDOMLO/mtlniDfuI4kLSk8zs9+a\n2fHp6xJgGnFw9hAxmqmn3P11d7/c3T9GjDr7VMFd1qYtOdrMfmhmnzazM4kLrF9GzHe/RzoAXdXM\n5jWzt5rZt4kLpEMc9HSqdl2IE8zsm9kDDTNbxsy+CxyRFl2Q23YK8ft7gGJq+z3HzLZqdSMz25K4\nYDvATwtmqP1NdzIVTLdGOz4IzG9mTadASaPN9gNOIZ7/H5pZRxd9r+MJYB4zq3fQeB8wvdGGqdN7\nIYr/UwB446z3zxNnvd8C7E2c9b5nN/bfgWnEWdcNp4dOoz8XIt7viziFeB63NbNzM7+PI4njgTPM\n7JNm9sZ7cjqG2Is0nQsxpU1PpLr4e3ffmaiL+xEzMRQ9Zrgp7ePo2s+anuOjiX8YvAIcY7nZB9I6\nx6W7NxbM8Bfis8XR9R5MuWrXEvlH7rF5iQ9gTxXMUHaNVH0Mqo/N9apGqj4G1cdQdn2E8muk6mNG\nn9dHUI0E1cca1cdQdo1UfWxO9bGBMXwMqfrYBl2TUyrH3R8zsw8Qc6EvD+ybvhq9cWeL1gPAtl58\n7u2riNEnA2Y2vp3Tst39KbozgOBXwHeALc1sdW8wZUSu7ZfMbFuiCO1A8YL+G+KMxWc72PbA9P3K\npmu1yeN6iDuY2dbEGY6HEdeCPKLphp0b7iCm2dTItbP6rm+yTkvc/WQze4V4TfxP+hpg8Hc8Htgq\nfWXdBWzt7kWnzC0ktV/0QtNnEhesXgfIntH6EjHCbBLRIb937vHac/QPBjsqO/Ez4KPENQ4OB75s\nZs8SB/YLZtr5PXNeR2K79P3PBdqHmIv/o8SAjt+Y2b3EmdvTiGkqatfRnY+YbuYdwKZERzjEc3Bi\nwQw3AzvS2Qewbo12vIk4w3xL4AfNVkyjGQ8ys5eIzu7vEWdAFzWNeF7fRe46DO6+8jDbGnEM1q3p\nzWvtTrWYTudA4mD3zPRB49PECLyRsKTFdX+zxhH1aU0a17/ac1DkOr24+7MW09BcBuxGvD9cTnwY\n/QnwWeLaLSeYWW0E4VsYHNF4FYMfxHoqzcJwBvEhseGsAC36IVGbPgasY2a3E6/NVYDTiZGmHwdu\nsRgYdj9xjZE9GRzQ1egaN606EfgIsLeZrUbUq3uJGvl24GCgNuVR/u+2dp0OL5ih7Bqp+hhUHweV\nViNVH9+g+hjKro9Qfo1UfayjH+sjqEYmqo9B9TGUXSNVHwepPnZojB1Dqj62QZ2cUknufrvF9TX3\nBXYF1qb5G/dNRMfe6Z67QHaH7c+ih9N7NsjwoMUp8BNoYxSOu79qZjsRF5oudLDj7jczOHVvy9Ko\nldoZjkWvj1qXu19iMW3sN4CDGJyquNvt/LjA5rVrhxY6wKhx99PTgcVBad91r/9IHPj8lRg1c5bH\nPOjd0OvpOIZw99fMbDPiTO0diAOKm4EvufttAGa2I3GAnb0GwQBxBuf+acqGTtsfSH9bxwIHEGf4\nZqcseY040Dk0HfRnTSUOdAp1cqa/7w8QB1sfIQ4cagcP+Tbzv69fAPsUeQ6S2ijTtTvY9jXgOoqP\nNr2CeG/Ym2E+gNW4+5fM7AVi9OH4LmS4kahzG9L4YvONbJO+F+30nkMalHOymV1EXEdke+J3VqSD\nfzj1asM4YGcafwDbOn0vfLa/u19nZhsQ03OvTvzM2+dWW4Shr9nZxAeOz9b5e+05d/9nwe0vsziz\nfW/iw21tmvn7ga8QHzo2IT4MfY+hg2QAfuTuvy+Y4W9m9mWiRr83fdWe22xbX3f3v+U23yl9v7pg\nhrJrpOpjUH0cqrQaqfqo+pjJUHZ9hPJrpOpjA/1YH0E1UvXxjQyqj6HsGqn6OJTqY0Gj/RhS9bE9\npf7DWqRVZrYgMU3pUsACafELRAfaPd78IrwyxllcDPskYsTMgLuvWHKknjCztxAjiBYhDihnEX8T\nd7t7o4tfj3lmNg8xymgFYg76m7zBBeoLtPFm4AOpjQFi6pRrut3OMBnWJDq7NyYOMvLTlrxInMl7\nNfBTd7+1S+2OAxaGGOHXjX12kGEhYkTYRGA3d2/5bGkz+xxxRjTu3vFZ9xbX3N0AeMDdb2hjuwnE\nh6+lgD3cvd0PcG1JU4p8j7hOBVDs587te8owq7zUaGYFM7uG+Ps51N0v7FKeCcTZ7LsC7wGWya3y\nAnAn8CfgfHe/hy5IP8vT7t5s+vKeSAMxsoNAvptGs2JmKxEfkLZh8Ho3dwGnuPvpXcywDfGPjvz1\nSW4HjnL3i+ps8+aU+YluvX+VUSNVH9/YT9/Xx7T/KcOs0rMaqfqo+pjbZ18eQ6o+tq6f6mPaZ1/X\nSNXHIfvsy/qYMpRaI1Uf39j/lGFWUX3ssbJrZL/Xx1apk1NERGSUSx+KFmLoIJDnqzB6TspnZgsQ\n0wjVBoJ8suRIPWGD16GoDQKZqb+JN+rFEsCskRwkZmZLEZcdGCD+WVH0UgJFsqhGSl2qj6qPWaqP\ngOqjJP1aH0E1sh7VR0D1URLVR9XHvF7USNVHEREREREREREREREREZExRGdyioiI9IE00mo5iGv+\n9nmG5YkRl/2eobTfhZ4DZahaBhERATObBKxJTPv3BPA3d3+5nzKU3X7FMqwBLF1yhio8D32doez2\nq5Ih5ZgfmEZ8hlmp1+0rQzXaTxnmI6bm7NvfgzJUJ0MF2l8MWJF0OS93f7rXGSb2ukEREREpxWLE\ntUMHgAl9nuF+ZSj9d6HnQBlGLIOZLQHsS1yfeQJxnZhzhrtGjJndCEwu+sGw7PaVQRmq1L4yVCdD\n+gfYUcDOwCTgJuK6Xbekx/chriu1KIMD4l8ws2Pc/egibVclQ9ntK4MyVC1D2e1XJUMbxpGmKO1x\nu8pQrfYhpmstO0MVngdlqEaGUto3s/cDxwLrkTmZMh27ftndr+lVFp3JKSIi0gfMbDIx6nXA3Uvp\nyFAGZahK+8owdjOY2UbAxcQ1QrJeB04FDnP3Vxts+xiweJEMZbevDMpQpfaVoXIZ/gh8MLf4OWAd\nYG3gZ002P9HdDy7SfhUylN2+MihD1TKU3X4VMpjZ7rTeMTAv8IN0e4/sA+5+rjJ0nqHs9pVBGaqW\noez2U4aZwGPuvkqdx3YBzqHxSZSvA3u4+/mdtt8OnckpIiIiIiKjnpktCVxEdCK8DPwNmAG8D1gC\nOAh4n5lt6e5PjrX2lUEZqtS+MlQuw84MdiKcB9wMvBf4KPAlYP302HeAU4D/ENOOHQrsBRxkZue7\n+9TRmqHs9pVBGaqWoez2q5IBOLvD7X6cuT0AdNyRoAyVaF8ZlKFqGcpuH2CB9DWEmU0BziT6Fv8F\nHA3cmB5ej6jfqwA/MLPr3P2hAhlaok5OERGRUcLMrMDmiyqDMnQzQ9ntK4My1HFQ2tf9wBbufi+A\nmU0EDiCmOlsH+LOZbeLuD3ep3aq0rwzKUKX2laFaGXZL349398PS7ZPN7GlgH+J/Q2e5+yGZbRzY\nx2IqyV2IDoUiHQllZyi7fWVQhqplKLv9qmTohirMlKgM5bcPylCjDKHsDCPV/meB2nVp3+vuz2ce\nu8vMLgKuB94O7A8cPkI53qBOThERkdHjHsq9zoAyKEOV2lcGZcjbIn0/oNaJAODu/wVOMrOrieki\njcHOhH+PofaVQRmq1L4yVCvDWkSNPiW3/HTin08DwLcabHs80ZGwfoPHR0uGsttXBmWoWoay269K\nhkeAZYiz7D/t7rc2WtHMJgHP0/3LPChD+e0rgzJULUPZ7TezWfp+aK6DEwB3f97MDgUuY87pyEfE\n+F40IiIiIl0zrsCXMihDtzOU3b4yKEPWSsB/gT/Ve9Ddbyemh5wGrABcZ2ardqntKrSvDMpQpfaV\noVoZJgMv1zlL9L70/VV3/2eDbW8HXk3ZRnOGsttXBmWoWoay269KhlWBk4F1gRvN7CQzW7DBuiM1\noE8Zym9fGZShahnKbr+ZFYDXaHBsm1xFHP/OcT3PkaBOThERkdFjRvr+KeIfZu18ra0MytDlDGW3\nrwzKkDc38EI6O6oud38U2BD4B7AUcK2ZrTFG2lcGZahS+8pQrQwvEf9oyrc7K918pkm22cBMYlqy\n0Zyh7PaVQRmqlqHs9iuRwd1nufvniI6E24DPAPea2cfqrN7NwYHKUKH2lUEZqpah7PaHMRF43t1f\na7SCu78KPEeda3qOVCAREREZHW4C/hdY2t2nt7Ohmc0afi1lUIZR1b4yKEPe48AyZrawuz/XaCV3\nn2FmmxDT57wHuNLMNh8D7SuDMlSpfWWoVoYngRXNbH53f7HO48P9c2wS8PQoz1B2+8qgDFXLUHb7\nVckAgLtPNbP1iI6Eo4DzzGwvYqrxe7rRhjJUv31lUIaqZSi7/QbuA95qZhObDeIjOjjnmM52JOhM\nThERkdHj5vT93cqgDBXIUHb7yqAMebcR/wwb9rofqaNhM+AaYFHgCmChUd6+MihDldpXhmpl+FfK\nYHUeWx5Ys9GGZrYMMA/w6CjPUHb7yqAMVctQdvtVyfAGd5/t7icR00ReDGwM3Gpmx5hZ0bNWlWGU\ntK8MylC1DCW3v5iZXZX9AhYnanfDqWjNbGlgXmIwy4hTJ6eIiMjocVP6XqQToeg0FsqgDFVpXxmU\nIe/K9H2PVlZ29xeALYHLiU6EeUd5+8qgDFVqXxmqlaFWozeo097D7v5Yk203y+1jtGYou31lUIaq\nZSi7/apkmIO7P+LuOwDbEmfjHwbcne73hDKU374yKEPVMpTU/lzARrmvycRn9w832W6T9P22EUuW\noU5OERGR0eMqYAdgXzNr6z3c3Z9y9/HuXvS9XxmUoSrtK4My5P0qfd/SzFZvse2XiA+FtW2LdLSW\n3b4yKEOV2leGamX4DXAG8GwH2x6Yvl/ZdK3qZyi7fWVQhqplKLv9qmRoyN0vAVYDTgSWBX4yUm0p\nQ3XbVwZlqFqGHra/5zBff2+y7X7p+/UjlG2IXl+UVEREREREZESY2XLABGCGu7d8/Q8zmwC8Dxjn\n7teO1vaVQRmq1L4yVCtDJ9KgleXT3Ufc/bV+y1B2+8qgDFXLUHb7ZWUwszWAk4AVgAF3X3Gk21SG\n6rWvDMpQtQxlt9+ImU1JNx9z95fLzCIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIi\nIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIi\nIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIn1mXNkBREREREREZHhm9n5gL+B/\ngGWABTIPf97dTzKz6cDyadkn3f2c3qasPjPbHrgo3T3a3b/S5f1PR7+DvmZmuwNnp7tfcvfjyswj\nIiIiIjJWjS87gIiIiIiIiDRnZscC1wKfAIyhHZwAA3U2q7esr5nZfMCJ6e6TwLEj0MwQugfdAAAP\n6ElEQVRA5rt+B/3pXOCOdPsrZrZ0mWFERERERMaqiWUHEBERERERkcbMbCvgi5lFLwC3A89llk3v\nZaZR7AsMnmV5grvParSimc3O3N3Y3a9tsY1xue/SZ9x9wMy+BvwKmAR8E9iz3FQiIiIiImOPOjlF\nRERERESq7VOZ27cAG7n7zLLCjFZmtihwSLr7LPD9Fjdt94xMncEpuPvFZnY3sCrwcTP7lrvfW3Yu\nEREREZGxRJ2cIiIiIiIi1bZu5vbJ6uDs2GeAhdPts5qdxZnT1hmZ7r5iW6lkLDsZOA2YABwO7F5u\nHBERERGRsUXX5BQREREREam2xTK3HyotxShmZvMAB6S7A8AZJcaR/nEBMb00wE5mtmSZYURERERE\nxhp1coqIiIiIiFTbhMzt10tLMbrtCCyebt/o7t7m9rq+prQtnXV9cbo7N7BPiXFERERERMYcTVcr\nIiIiIiJSIWY2BbivwcNXm1l+2ZHufmSB9pYgptH8EGDAZOAl4FHgr8Cv3P2yMvdrZrMzdzd292vN\nbH5gJ+AjwNuAJYF5qf98ZKcJ/WWTdvYAzqrzUL3nHeCB/PS0ZjYdWD7d/aS7n9OgrUY/08eAXYjn\nbHHgKWAa8HPgXHf/b6P83VTndTjF3R80szcBnyCe9xWBNwGPE9eLPd/dGz6/Tdqaj/iZNwfWJH7u\nuYEngDuA3wE/dveXOsy8KLArsAPwFmCJtP9mv5/NgK2B9YGl0s/5KvAf4E7gGuAid39kmB/vwtQ2\nxPN21DDri4iIiIhIi9TJKSIiIiIiMroNdLqhmR0CfA1YIPfQ3MT1K98G7GlmNwB7uPs/y9wv8bMO\nmNlawE+BVRqsk83yJuADmcfa7rAdJk+zZa38bmo/0zuIDrF8b+rS6WtT4LNmtpO739tJ2AIGAMxs\nA+BnRIdy1nLpaxsz+zSwi7s/1sqOzWxX4DiiIzGvtt8tgCPM7FNtdLjXMm8KnFMn8xvr5PKsC3wP\nWLvO+hOBldPXdsAJZva/7n5Vkxx/BF4D5gLeYmZrufvUFn8GERERERFpQtPVioiIiIiIVMuLwOXA\nH9L3rBvT8uzXvzppxMxOJzqXah2RA8Q1P68BbgZezqz+HuB6M1uzrP0m44gz8a4kOjgHgAeAa4nn\n5pk623yIwc++T7j7PU32/zCDz31Wvef9D6ndehnbtSxwFdHBOQBMJ56vfxBnD9a8kzirdOUO2ijq\nXcRzsySR0YGrgdsYOo3yhsCVZjZ5uB2a2dHATxjs4BwgzpS8nvj5H8isvjTwGzP7eBuZ3wtcmsn8\nz5R5KjCrTp6diN9ptoPzdeJM2quBv6RMtc7RCcCkZgHc/UXgpsyiLdvILyIiIiIiTehMThERERER\nkQpx9yeIaTuBOaY1/aK7X1e0DTPbn6HXB5wG7O/uf8mssyBwMHAE0ZkzGbjQzN7l7nN0EI3kfnO+\nCywIXAcc5O63ZfY9F3Oesbdx5vaNzXbs7n8C/pT21fXnvYFxwInE83AnsK+731B70MwWBg4HDknr\nLglcYGb/4+6z6+xvpDKeCcwD/Jn4nd6VybgE8C1iOlaAVYHTgQ832qGZ7Qd8KbPoN8DX3P323Hpr\nAqcSHeITgNPN7BZ3v7OFzKcTZ1D+GjjU3d+YyjZNkbtQ5v57iA7X2v9JZgHHAKe7+4xcpiWIqW/3\nGyZDzd+JDleIs4o1Za2IiIiISBeok1NERERERKSPpOsTHp9ZNA1Y392fy67n7jOBr5vZA8CP0uIV\nga8CX+zVfutYELgC2Cp/fUp3f404azTr3Znbt1NNiwP3ABu4+7PZB9Lzd5iZPUV0JAKsA+wNnNHj\njNcBm6bnOZvxcWAPM3sB2D8t3t7MNnP3K/I7MrMViI7dmm+6+1frNerut5jZRsRZpBsR1109Ftiq\nhcwLAme7+1519vsScY1YzGwicB6D/yOZkX7OWxpkehw4DTgtXUd1OLdlbq/VwvoiIiIiItICTVcr\nIiIiIiLSX/YGah0zA8Ce+Y7ILHc/G7gku32Djp2R2m/eq8Be+Q7OesxsHHH9z5qOpvbtgQFgv3wH\nZ5a7H8/QaU8PGPFUQ70K7J3v4Mw5mKGdzI0yfp44KxTgukYdnDWpzT2B2u98czNbafjIPAF8poX1\ndiE62mv2a9TBWSfbiy2s9u/M7UlmNqWVfYuIiIiISHPq5BQREREREekv22du/8Xdb2q45qBvZ24v\nwtApYEd6v3mXufsjLawHcR3HuTP3W92u1+5scTrcUzO3V+/xtTkvd/emncTu/jIxrW3N5mla2DeY\n2XgGp7WFoa+BZvueTkyVCzEV7Qdb2OyCFjshd83cvsPdL2wlUxseztweB6zQ5f2LiIiIiPQldXKK\niIiIiIj0CTObm6HTZV7SaN0sd/8z8Exm0Xt6sd8G2rk25uK5+zPqrlW+37e43u9y99frdpAmOsk4\nkTmnZ12d6NAGmA1c1UaG7HTD67Sw/rCvlTRV7fsyi37ZRp5WPZO7/+YRaENEREREpO/ompwiIiIi\nIiL9YzmGntl4W6MV67gD2CDdfkuP9lvPfW3se4HM7QHSNRgr6I5WVnL3GWb2KLBUWrTKyEWaQ0sZ\ngbty91cBrs/cf2fm9uvARWla4VZkXx+TW1i/ldfKsgx9ndzcYpZ25M8mXaDuWiIiIiIi0hZ1coqI\niIiIiPSPRXP3n2xj26ea7Gek9lvP823sO2tc+qqip9tYdwaDnZytPF/dMECLGd39ZTN7gcGOvHzG\nxTK35wL+t8NMiwzz+ACtvVbelLv/RGdxmqrq605EREREZFRTJ6eIiIiIiEj/mCd3/9U2tn0lc3ve\nHu23ntlt7PuF3P356q5Vvk6fr/zzPpLazVjr5MxnzJ/FONBhnlY6Dlt5reRfcy93kGU4+ddd/nUp\nIiIiIiIdUCeniIiIiIhI/3g2d3/BNrZdqMl+Rmq/ReXPKM2ftVcVnT5fz3U7SAPj6F7G7O/4OXfv\n1dmojeSvl7nwCLTRi7NFRURERET6zviyA4iIiIiIiEjP5DtXVm5j2+z1EPP7Gan9FvUfhp75uFyX\n998tK7WykplNZOjP8PjIxKmr1YzLMnRAdT7jo5nbC5lZK2fvjqTHcvffNgJtLJu5PQBMH4E2RERE\nRET6jjo5RURERERE+oS7Pwk8lFn03la2M7PFgLdmFt3ci/0W5e4DwN2ZRe10vmaN9DUV12txvXcx\ndPrXf4xAlkZazZhdb4A5M96QuT0O+J8ioYpy92eAezKLNhqBZrKvu5nu/sAItCEiIiIi0nfUySki\nIiIiItJfrs3c3tHMWrmu424MdvQNANf1cL9F3ZS5/c42tsteN3Gkr+W5pZm1Mh3srpnbLzL0Zxtp\nHzWzVv6HkM34UL5Dz90fBW7LLNq7G+EK+kPm9o6p872b3pW53cuOaRERERGRMU2dnCIiIiIiIv3l\nh5nbiwNfbrZy6vA5IrPoL+5+T51VR2q/RV2Tub1uG9v9J3PbuhOloQWA/2u2gpmtCOyXWfRzd39x\nRFMNNSXX/hzM7D3AdplFP2qw6vGZ2zub2ebFohV2CvB6uj0f8IMu7z97duvVXd63iIiIiEjfUien\niIiIiIhIH3H3PzO0o+UrZrZHvXXNbDJwKTA5LZoNHNnL/XbB5Wn/AJPN7B0tbpc9425PM1uku7Hm\ncLCZ7VPvATNbhni+atevfAU4boTz1HOCmW1R7wEzWxW4KLPoaeC0Bvv5KXB9uj0e+GWj10qujfnN\nbFcz6/a0xvcD388s+rCZnWtmk5pkGW9mO5rZWs32bWYLAO9OdweAywoHFhERERERACaWHUBERERE\nRER6bg/gFuBNRCfTWWa2C9H5NB2Yn7iu5qcY7IgEOMndryphvx1z9xlmdhWwSVq0JXBnC5ueB+yc\nbr8TeNjMpgLPEJ1VAI+7+75diPlTYBfgdDPbEbiAeL4WBN5PPF8LZ9Y/0t3v7UK77bgA+BhwqZld\nCPwaeIT4XW8K7Mng9UIHgM+4+1P1duTuA2b2YeDvwArE6+IsMzsUuBCYCswA5gIWBVYlzsL9ICM3\ndfAhRGdk7RqhuwEfMrPzgb8AT6Y8yxFnZm4HLJG+T22y302BCen2fe7ebF0REREREWmDOjlFRERE\nRET6jLs/ZGYbEtciXCYt3jR91TMAnOruB5ex3y44h8FOzg8D3xpuA3e/zMx+BOyVFs0PrJ9b7QGK\nGwDOAGYSnZmtPF/HdqHddgwAXwUWAbYAdkxfjdY91N1/1myH7v6Ema0H/JLoyIXozGw6bW+mjeGM\nG36VIXleM7NNiM7cbdLiycBB6avTHNnn6SftZBIRERERkeY0Xa2IiIiIiEj1DdBah8oALa7r7tOA\n1YETgOeb7O9mYAt3b9TR05P90vpzUM+FwBPp9jpm9rZWNnL3fYCtiDMt7yV+ntk0f55b/h0k44AB\nd98P2Je4Fmi9bR8Edm3j+eq2/wJbE52QM6if8W5gM3f/Tis7dPcngI2IM2ZvZnBa4XoG0v6/Dawx\nzK47eq24+4vuvh3RET61yT4GgMeBU4mzUesyswWBbdPdV4Ez280kIiIiIiKNtTWyUURERERERMYe\nM5sIvA9YmTh77SWiE+d6d3+4avvtMMtXga+nu9919y/0sv08M8t26G3k7tel5eOIMxtXBhYnOhTv\ndPcbepxvCnBfujsArOjuD6bH5iI6J6cQ09U+CUx191sLtjmZeL0sSUxT+1/g2ZTjztQp2jNmtjQx\nvfISxHTBLxFT9E5LnfnDbf8p4Afp7vnu/vGRyioiIiIi0o/UySkiIiIiIiJjnpktAtxPdFY9Byzv\n7jNLzFO3k7MqmnVyyvBSZ/UdwGrA68Dq7n5PualERERERMYWTVcrIiIiIiIiY567Pwscn+4uDOxf\nYhwZ+7YhOjghzuJUB6eIiIiISJepk1NERERERET6xYnAA+n2IemaiSJdZWbjgSPT3VnAESXGERER\nEREZs9TJKSIiIiIiIn3B3V8CPp/uLgYcVmIcGbt2A95JTPP7TXd/pOQ8IiIiIiJj0sSyA4iIiIiI\niIj0irtfzBgb8GtmxxGdakUNuPvmdZaP68K++4a7nwucW3YOEREREZGxTp2cIiIiIiIiIuUY6NJ+\n1gU26NK+8rqVUUREREREpKvUySkiIiIiIiLSY+7ezbNJB+hyZ6S7T2eMnfEqIiIiIiIiIiIiIiIi\nIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIi\nIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIi\nIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIi\nIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiLt+v9QK4ElOtxt6AAAAABJRU5ErkJggg==\n",
       "text": [
        "<matplotlib.figure.Figure at 0x7f3ec54d7190>"
       ]
      }
     ],
     "prompt_number": 10
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "By looking at the previous figure we can say that the *social norm* is to tip the 20% of the charge. Perhaps that is the question, to know if a tip will be above or below that norm.\n",
      "\n",
      "For answering that, let's change the classes for use only two:\n",
      "\n",
      "$$\n",
      "``<\\:20\"\\:\\:and\\:\\:``>=\\:20\"\n",
      "$$"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tip_labels = ['< 20', '>= 20']\n",
      "tip_ranges_by_label = [[0.0, 20.0], [20.0, 51.0]]\n",
      "\n",
      "for i, tip_label in enumerate(tip_labels):\n",
      "    tip_mask = ((data.tip_perc >= tip_ranges_by_label[i][0]) & (data.tip_perc < tip_ranges_by_label[i][1]))\n",
      "    data.tip_label[tip_mask] = tip_label\n",
      "    \n",
      "    tip_mask = None"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 11
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data.to_csv('../data/dataset/dataset.csv', index=False)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 12
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Will this change work? Let's find it out in the [next notebook](6. Learning, a better way.ipynb)."
     ]
    }
   ],
   "metadata": {}
  }
 ]
}