{ "metadata": { "name": "" }, "nbformat": 3, "nbformat_minor": 0, "worksheets": [ { "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "Calibration of Heston's Model on SPX data\n", "=======================================\n", "\n", "This notebook demonstrates the calibration of Heston's model on SPX data, using the QuantLib *HestonModel* class. \n", "The code is adapted from the test suite written by Klaus Spandersen.\n", "\n", "The calibration function takes as input a `pandas.DataFrame` constructed in notebook OptionQuotes.\n", "\n", "QuantLib dependencies\n", "---------------------" ] }, { "cell_type": "code", "collapsed": false, "input": [ "import numpy as np\n", "import pandas\n", "from pandas import DataFrame\n", "import datetime\n", "\n", "from quantlib.models.equity.heston_model import (\n", " HestonModelHelper, HestonModel)\n", "from quantlib.models.calibration_helper import ImpliedVolError\n", "from quantlib.processes.heston_process import HestonProcess\n", "from quantlib.pricingengines.api import AnalyticHestonEngine\n", "from quantlib.math.optimization import LevenbergMarquardt, EndCriteria\n", "from quantlib.settings import Settings\n", "from quantlib.time.api import Period, Date, Actual365Fixed, TARGET, Days\n", "from quantlib.quotes import SimpleQuote\n", "from quantlib.termstructures.yields.zero_curve import ZeroCurve\n", "\n", "from quantlib.util.converter import pydate_to_qldate, df_to_zero_curve\n", "import quantlib.reference.names as nm" ], "language": "python", "metadata": {}, "outputs": [], "prompt_number": 1 }, { "cell_type": "markdown", "metadata": {}, "source": [ "Utility functions\n", "-----------------\n", "\n", "The calibration process uses some utility functions, defined below.\n", "The main task is to convert the data into a set of helper objects, one per data point (bid and ask prices)." ] }, { "cell_type": "code", "collapsed": false, "input": [ "def heston_helpers(df_option, df_rates):\n", " \"\"\"\n", " Create array of heston options helpers\n", " \"\"\"\n", "\n", " trade_date = df_option[nm.TRADE_DATE].ix[0]\n", " settlement_date = pydate_to_qldate(trade_date)\n", " \n", " settings = Settings()\n", " settings.evaluation_date = settlement_date\n", "\n", " calendar = TARGET()\n", "\n", " # convert data frame (date/value) into zero curve\n", " # expect the index to be a date, and 1 column of values\n", "\n", " risk_free_ts = df_to_zero_curve(df_rates['iRate'], trade_date)\n", " dividend_ts = df_to_zero_curve(df_rates[nm.DIVIDEND_YIELD], trade_date)\n", "\n", " # loop through rows in option data frame, construct\n", " # helpers for bid/ask\n", "\n", " \n", " options = []\n", " for index, row in df_option.T.iteritems():\n", "\n", " strike = row[nm.STRIKE]\n", " spot = row[nm.SPOT]\n", " if (strike/spot > 1.3) | (strike/spot < .7):\n", " continue\n", " \n", " expiry_date = row[nm.EXPIRY_DATE]\n", " \n", " days = (expiry_date - trade_date).days\n", " maturity = Period(days, Days)\n", "\n", " options.append(\n", " HestonModelHelper(\n", " maturity, calendar, spot,\n", " strike, SimpleQuote(row['IVBid']),\n", " risk_free_ts, dividend_ts,\n", " ImpliedVolError))\n", " \n", " options.append(\n", " HestonModelHelper(\n", " maturity, calendar, spot,\n", " strike, SimpleQuote(row['IVAsk']),\n", " risk_free_ts, dividend_ts,\n", " ImpliedVolError))\n", "\n", " return {'options':options, 'spot': spot}" ], "language": "python", "metadata": {}, "outputs": [], "prompt_number": 2 }, { "cell_type": "markdown", "metadata": {}, "source": [ "The function merge_df merges the result of the calibration (fitted option price and fitted implied volatility)\n", "with the input data set. This will facilitate the plotting of actual vs. fitted volatility." ] }, { "cell_type": "code", "collapsed": false, "input": [ "def merge_df(df_option, options, model_name):\n", " df_output = DataFrame.filter(df_option,\n", " items=['dtTrade', 'dtExpiry',\n", " 'Type', 'Strike', 'Mid',\n", " 'QuickDelta', 'IVBid', 'IVAsk',\n", " 'iRate', 'dRate', 'ATMVol', 'Fwd'])\n", "\n", " model_value = np.zeros(len(df_option))\n", " model_iv = np.zeros(len(df_option))\n", " for i, j in zip(range(len(df_option)), range(0, len(options),2)):\n", " model_value[i] = options[j].model_value()\n", " model_iv[i] = options[j].impliedVolatility(model_value[i],\n", " accuracy=1.e-5, maxEvaluations=5000,\n", " minVol=.01, maxVol=10.0)\n", "\n", " df_output[model_name + '-Value'] = model_value\n", " df_output[model_name + '-IV'] = model_iv\n", " \n", " return df_output" ], "language": "python", "metadata": {}, "outputs": [], "prompt_number": 3 }, { "cell_type": "markdown", "metadata": {}, "source": [ "The calibration process\n", "-----------------------\n", "\n", "The calibration process is performed by the following function." ] }, { "cell_type": "code", "collapsed": false, "input": [ "def heston_calibration(df_option, ival=None):\n", " \"\"\"\n", " calibrate heston model\n", " \"\"\"\n", "\n", " # extract rates and div yields from the data set \n", " df_tmp = DataFrame.filter(df_option, items=['dtExpiry', 'iRate', nm.DIVIDEND_YIELD])\n", " \n", " grouped = df_tmp.groupby('dtExpiry')\n", " \n", " def aggregate(serie):\n", " return serie[serie.index[0]]\n", " \n", " df_rates = grouped.agg(aggregate)\n", " \n", " trade_date = df_option[nm.TRADE_DATE][0]\n", " expiry_date = df_option[nm.EXPIRY_DATE][0]\n", " \n", " spot = df_option[nm.SPOT][0]\n", " \n", " # build array of option helpers\n", " hh = heston_helpers(df_option, df_rates)\n", " options = hh['options']\n", " spot = hh['spot']\n", "\n", " risk_free_ts = df_to_zero_curve(df_rates[nm.INTEREST_RATE], trade_date)\n", " dividend_ts = df_to_zero_curve(df_rates[nm.DIVIDEND_YIELD], trade_date)\n", "\n", " # initial values for parameters\n", " if ival is None:\n", " ival = {'v0': 0.1, 'kappa': 1.0, 'theta': 0.1,\n", " 'sigma': 0.5, 'rho': -.5}\n", "\n", " process = HestonProcess(\n", " risk_free_ts, dividend_ts, SimpleQuote(spot), ival['v0'], ival['kappa'],\n", " ival['theta'], ival['sigma'], ival['rho'])\n", "\n", " model = HestonModel(process)\n", " engine = AnalyticHestonEngine(model, 64)\n", "\n", " for option in options:\n", " option.set_pricing_engine(engine)\n", "\n", " om = LevenbergMarquardt(1e-8, 1e-8, 1e-8)\n", " model.calibrate(\n", " options, om, EndCriteria(400, 40, 1.0e-8, 1.0e-8, 1.0e-8)\n", " )\n", "\n", " print('model calibration results:')\n", " print('v0: %f kappa: %f theta: %f sigma: %f rho: %f' %\n", " (model.v0, model.kappa, model.theta, model.sigma,\n", " model.rho))\n", "\n", " calib_error = (1.0/len(options)) * sum(\n", " [pow(o.calibration_error()*100.0,2) for o in options])\n", "\n", " print('SSE: %f' % calib_error)\n", "\n", " # merge the fitted volatility and the input data set\n", " return merge_df(df_option, options, 'Heston')" ], "language": "python", "metadata": {}, "outputs": [], "prompt_number": 4 }, { "cell_type": "markdown", "metadata": {}, "source": [ "Calibration\n", "-----------\n", "\n", "Finally, the calibration is performed by first loading the option data and calling the calibration routine.\n" ] }, { "cell_type": "code", "collapsed": false, "input": [ "df_options = pandas.read_pickle('df_options_SPX_24jan2011.pkl')\n", "df_heston_cal = heston_calibration(df_options)" ], "language": "python", "metadata": {}, "outputs": [ { "output_type": "stream", "stream": "stdout", "text": [ "model calibration results:\n", "v0: 0.051965 kappa: 0.977303 theta: 0.102573 sigma: 0.987787 rho: -0.747035\n", "SSE: 0.935261\n" ] } ], "prompt_number": 5 }, { "cell_type": "markdown", "metadata": {}, "source": [ "Plot Actual vs. Fitted Implied Volatility\n", "-----------------------------------------\n", "\n", "We display 4 graphs in one plot, and show the bid/ask market volatility with the fitted volatility\n", "for selected maturities." ] }, { "cell_type": "code", "collapsed": false, "input": [ "def calibration_subplot(ax, group, i, model_name):\n", " group = group.sort_index(by='Strike')\n", " dtExpiry = group.get_value(group.index[0], 'dtExpiry')\n", " K = group['Strike']\n", " VB = group['IVBid']\n", " VA = group['IVAsk']\n", " VM = group[model_name + '-IV']\n", "\n", " ax.plot(K, VA, 'b.', K,VB,'b.', K,VM,'r-')\n", " if i==3:\n", " ax.set_xlabel('Strike')\n", " if i==0:\n", " ax.set_ylabel('Implied Vol')\n", " ax.text(.6,.8,'%s' % dtExpiry, transform=ax.transAxes)\n", " \n", "def calibration_plot(title, df_calibration, model_name):\n", " df_calibration = DataFrame.filter(df_calibration,\n", " items=['dtExpiry', \n", " 'Strike', 'IVBid', 'IVAsk',\n", " 'TTM', model_name+'-IV'])\n", "\n", " # group by maturity\n", " grouped = df_calibration.groupby('dtExpiry')\n", "\n", " all_groups = [(dt, g) for dt, g in grouped]\n", " \n", " xy = [(0,0), (0,1), (1,0), (1,1)]\n", "\n", " for k in range(0, len(all_groups),4):\n", " if (k+4) >= len(all_groups):\n", " break\n", " plt.figure()\n", " fig, axs = plt.subplots(2, 2, sharex=True, sharey=True)\n", "\n", " for i in range(4):\n", " x,y = xy[i]\n", " calibration_subplot(axs[x,y], all_groups[i+k][1],i, model_name)" ], "language": "python", "metadata": {}, "outputs": [], "prompt_number": 6 }, { "cell_type": "code", "collapsed": false, "input": [ "dtTrade = df_options['dtTrade'][0]\n", "title = 'Heston Model (%s)' % dtTrade\n", "calibration_plot(title, df_heston_cal, 'Heston')" ], "language": "python", "metadata": {}, "outputs": [ { "metadata": {}, "output_type": "display_data", "text": [ "" ] }, { "metadata": {}, "output_type": "display_data", "png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAEKCAYAAADTgGjXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XlYlOX6B/DvIJiaIi6oxeCGigy7gKSm4oLbcTuCuZS4\noR4N0zK18vhLO6lZWiFWtqklorlkbkCFgksoiqCYSy6BCy65gIKobPfvj1fGGWFglvedjftzXVzJ\nzLs8Y/frM892PzIiIjDGGGN6sDF1ARhjjFkurkQYY4zpjSsRxhhjeuNKhDHGmN64EmGMMaY3rkQY\nY4zpTdJKJD4+Hu3bt0fbtm2xdOnScu+vX78e3t7e8PLyQpcuXZCRkaF8b8mSJXB3d4enpydGjx6N\nx48fS1lUxhhjepCsEikpKUFERATi4+Nx+vRpbNiwAWfOnFE7pnXr1ti/fz8yMjIwf/58TJ48GQCQ\nlZWFb7/9FmlpaTh58iRKSkqwceNGqYrKGGNMT5JVIkeOHEGbNm3QsmVL2NnZYeTIkdi+fbvaMZ06\ndUL9+vUBAIGBgbh69SoAwN7eHnZ2digoKEBxcTEKCgrg5OQkVVEZY4zpyVaqC2dnZ8PZ2Vn5u1wu\nR0pKisbjv//+ewwYMAAA0LBhQ8yaNQvNmzdH7dq10bdvX/Tu3VvteJlMJk3BGVNhioQOHNtMamLG\ntWQtEV0ehMTERKxevVo5bnLx4kV8/vnnyMrKwrVr15Cfn4/169eXO4+ITPLz/vvvV6v7VtfPbErV\n7e+6OsaXqe4tNskqEScnJ1y5ckX5+5UrVyCXy8sdl5GRgUmTJmHHjh1o0KABACA1NRWdO3dGo0aN\nYGtri2HDhiE5OVmqojLGGNOTZJWIv78/zp8/j6ysLBQWFuKnn37C4MGD1Y65fPkyhg0bhujoaLRp\n00b5evv27XH48GE8fPgQRISEhAQoFAqpimoSkycDQUHAgAFAbq6pS8MYY/qRbEzE1tYWK1euRN++\nfVFSUoKJEyfCzc0NX3/9NQBgypQp+OCDD5CTk4OpU6cCAOzs7HDkyBF4e3sjLCwM/v7+sLGxQYcO\nHZQzt8xBUFCQwdc4dw7Yt0/48+TJwKZNxrmvvkx1b1N+5uqG46v63FtMMpKik8wIZDKZJP17xjJg\nABAXBwQEAL/9Bjg4mLpE7FmmijFLj21m3sSOL65ETCQ3V2iBfPMNMGeO0DKpUweIieEKxVxwJcKs\nEVciT1jTgxYU9LRra/hw7bq2mPS4EmHWSOz44txZZqBOHeG/AQFCy4QxxiwFt0TMAHdtmSduiTBr\nxN1ZT1jrg8ZdW+aDKxFmjSyqO8uQLL65ubkIDQ2Fm5sbFAoFDh8+LGVRzQZ3bTHGLIlkLZGSkhK4\nuroiISEBTk5OCAgIwIYNG+Dm5qY85tChQ1AoFKhfvz7i4+OxYMECZWUxduxYdO/eHRMmTEBxcTEe\nPHigTNYIWO+3NdWuLe7KMi1uiTBrZDEtEUOy+N67dw8HDhzAhAkTAAgLF1UrEGvm4CB0YTk4mNeq\n9itXrqBHjx5wd3eHh4cHVqxYAQC4e/cugoOD0a5dO/Tp0we5Twp69+5d9OjRA/Xq1cP06dPVrjVv\n3jw0b94c9erVq/Sex44dg6enJ9q2bYsZM2YoX//000/h7u4Ob29v9O7dG5cvX67w/P3796NDhw6w\ns7PD1q1b1d6bO3cuPD094enpiU3cZ1itmVNsr1q1Cl5eXvD19UWnTp1w4sSJCs+vLLYvX76MPn36\nQKFQwN3dHZcuXdL570QnJJHNmzdTeHi48vd169ZRRESExuM/+eQTmjRpEhERpaenU8eOHWncuHHk\n6+tL4eHh9ODBA7XjAdD777+v/ElMTJTkc5hS9+5EgPAzfLhpy3L9+nVKT08nIqK8vDxq164dnT59\nmmbPnk1Lly4lIqKPPvqI5s6dS0REDx48oIMHD9KqVavK/X9PSUmh69evU926dSu9Z0BAAKWkpBAR\nUf/+/SkuLo6IiBITE+nhw4dERPTVV1/RiBEjKjw/KyuLMjIyKCwsjLZs2aJ8fdeuXRQcHEwlJSX0\n4MEDCggIoPv371NiYqJaTEn4eFSqOsS2OTGn2L5//77ymB07dlCvXr0qPF9TbBMRde/enRISEpRl\njY+PlzSuJXtKtmzZonUlsnfvXnJzc6O7d+8SEdHRo0fJ1taWjhw5QkREM2bMoPnz56udY6oH3Jj6\n9xcqkIAAopwcU5dG3ZAhQ+j3338nV1dXunHjBhEJD6Orq6vacWvWrNH4/72yB+3atWvUvn175e8b\nNmygKVOmlDsuLS2NunTpUmlZx40bp/agffLJJ/S///1P+fvEiRNp06ZN5c4zZSViLiZNEr7M9O9v\nfjEoFXOJ7ZiYGI1fkMo8G9unTp2il19+udJzxI4vs8ziK5fLIZfLERAQAAAIDQ1FWlqaVEU1WzEx\nwgwtc0uLkpWVhfT0dAQGBuLmzZto2rQpAKBp06a4efOm2rH67o2RnZ2tFi9OTk7Izs4ud5zqPjTa\n8vb2Rnx8PB4+fIjbt28jMTFR2ZXK1JXleIuLE7pXrZ05xPaXX36JNm3a4K233sKSJUt0uva5c+fg\n4OCAkJAQdOjQAXPmzEFpaale5dSWWWbxbdasGZydnXHu3DkAQEJCAtzd3aUqqtlSHR8xF/n5+QgJ\nCUFkZGS5fl+ZTGbUDZWio6ORlpaG2bNn63RecHAwBgwYgM6dO2P06NHo1KkTbGx43W1FqtNsQXOJ\n7WnTpuHChQv49NNPlePC2iouLsaBAwewfPlyHD16FH///TfWrl0rTUGfkOzJUc3iq1AoMGLECGUW\n37JMvqpZfH19fdGxY0fl+VFRUXj11Vfh7e2NjIwMvPfee1IVlWmpqKgIISEhGDNmDIYOHQpA+IZ2\n48YNAMD169fRpEkTva5dUlICHx8f+Pr6YsGCBZDL5Wqtg6tXr6ptkZyQkIDFixdjx44dsLOzAwD8\n97//ha+vLzp06FDu+s/+A/Dee+8hPT0dv/32G4gIrq6uepXb2plra1hs5hTbZUaMGKHsgZk3b55W\nse3s7AwfHx+0bNkSNWrUwNChQ6XvxRG1c8yILLjoelHtmx471vj91KWlpTRmzBiaOXOm2uuzZ8+m\njz76iIiIlixZohx8LKNvvzERUceOHenw4cNUWlqqNviYlpZGLi4udOHCBa3KPnbsWLV+45KSErp9\n+zYREZ04cYI8PDyopKSk3HmmirHqFtumZk6xff78eeUxO3bsqHJ849nYLi4uJm9vb7p16xYRCWMm\nX375pdo5YseXxUarVTxopaVaH6o6U8vR0fiztg4cOEAymYy8vb3Jx8eHfHx8KC4uju7cuUO9evWi\ntm3bUnBwMOWo1GotWrSghg0bUt26dUkul9OZM2eISHg45XI51ahRg+RyOS1cuLDCe6amppKHhwe5\nuLjQ9OnTla/37t2bmjVrpizHkCFDKjz/yJEjJJfL6fnnn6dGjRqRh4cHERE9fPiQFAoFKRQK6tSp\nE504caLC87kSqZy1DLqbU2zPmDGD3N3dycfHh4KDg9UqFVWaYpuI6PfffycvLy/y9PSk8ePHU1FR\nkdq5YscXpz0xpagoID0d+OwzoIp1MKr7j9SvDyQk8F4kUuPFhk/cuweEhgKLFwtB9wSn6LFMFrPY\nkFVu8mSg/0/jEJtQE6UeXsCePZUer9o3vXnz0z/PmWM+CxKZlbK3B8aNAwYOFALu4UMA1WvQnWnG\nLRETUf0W92HXeMzLnAQMGQJ8/PHTp1PH6/C3QXFxS+QZ//wDvPEGcOwY8N13yPXuztmnLRC3RKyE\n6re413f0AzIyhKaEry+QkqLXdfjbIJNUkybAxo3AsmXAq6/C4d2p2PTdfTg4VL/1JOwpjS2RY8eO\nVTovuqKpZsZktt/WtKQx0eLmzcD06cKb8+cDT6av6nwdZjBuiVQiNxeYPRv49Vfgm28wYEU/5Zid\nQgFkZXGrxFwZbT+RoKCgSiuRxMTEKi8eHx+PmTNnoqSkBOHh4Zg7d67a++vXr8fHH38MIkK9evXw\n1VdfwcvLS/l+SUkJ/P39IZfLsXPnTvWCW8KDpqXJk5/pCnh4HZgwAbh9G4iOBrRcw6B6nSZN+EE2\nFFciWtizBwgPx+POQfjPg0/x2doGGDqUu1jNmejxJepcLxXFxcXk4uJCmZmZVFhYSN7e3nT69Gm1\nY5KTkyk3N5eIiOLi4igwMFDt/eXLl9Po0aNp0KBB5a4vYdGNrsJEi6WlRF98QdSokfBfLaYDm3oa\nsLUxVYxZXGzfv0/0+utEL75ItGOHxpxv1jIl2NKJHV9VjokUFhYiMjISISEhCAkJQVRUFIqKiqqs\nnAxJBQ8IqzhjY2MRHh5uOd/K9KQ6rlGnzpPZVv+SIXf0NOCPP4DVq5HRYhD+3flmpbOwVK/j7f30\nzzxWwiRVrx6wcqXQ5J05E7/Yh2HckJxy08953MQ62VZ1wNSpU1FcXIzXX38dRIR169Zh6tSp+O67\n7yo9Lzs7G87Ozsrf5XI5UioZMH42kd6bb76JTz75BPfv39d4zoIFC5R/DgoKQlBQUFUfxyzFxDwd\n11DtCpg8Gdi0yRVITkZG2wX48pAPJuFbTJ48sMIuAtXrlJ3PYyXaS0pKQlJSkqmLAcBCY7t7dyAj\nAzXffRdrtnoAB78WpgU/wZNATEPyuK6qqeLp6anVa88yJBX8zp07adq0aUQk7B0xcODAcudoUXSL\npKkroH9/oq7YR9dqNqdH4a8TFRRofU3uRtCPqWLMnGJb79hJSiJq1UrI0fPkxJwcoWuVY9C0xI6v\nKruzbG1tceHCBeXvFy9ehK1tlQ0Yg1LBJycnY8eOHWjVqhVGjRqFvXv3IiwsrMp7WgNNCe9iYoBm\nw7uhzrkTeC7vtvB1TmVP+spwNwLTl96x86RVgjp1AC8v4NdfzTIrNRNBVbVMQkICOTs7U7du3ahb\nt27UvHlz2rNnT5W1U1FREbVu3ZoyMzPp8ePHFQ6sX7p0iVxcXOjQoUMar5OUlFStWiKqNH4LLC0l\nWruWqHFjoqioKgfdVVs3pkjeaKlMFWPmFNuibIz2++9EzZsLAa2yc18Zbikbl9jxpfFq/fv3p3Xr\n1lFeXh49evSIjh8/TsePH1duS6qN2NhYateuHbm4uNDixYuJiGjVqlW0atUqIhJ2lGvYsKEy6VlA\nQEC5ayQlJVn97CxNqtwe99w5Ij8/okGDiJ5k7ayIajeCOW25a+64EhGxC+rePaLwcKIWLYj27lV7\ni2PSuIxWiWzbto1GjBhBjRs3puHDh9PPP/9Mjx8/FvXmhjCnB00qWn0LfPyYaPZsIrlc6IcW45qM\niLgSqYpeLYjYWCInJ6IZM5TjehyTxmW0SqRMfn4+bdiwgYYMGUJNmjShcePG0a+//ipqIfRhKQ+a\nIXT6FhgfT/TCC0Tvv0/0TOrnyq7JXQmacSVSOb1bEHfuEI0cSdS+PdGRIzzgbmRix5dOCRhPnDiB\nsWPH4uTJkygpKRFzaEZnFrWqV2TlVriXDVRevw6MGQMUFgpvVDCR4VmcwFEzXrFeOdXtCfRKdfLT\nT0JCxylTtErxw8Rh9ASMN27cwIoVK9C5c2cMHToU/fr1k367RVYpjTNmXnhBmNbVty/g7w/ExlZ5\nLZ67z/SlOpMwK0uPWVwjRgDHjwtZgV96CTh9Wu3tyZN5mwOLoKmJ8vXXX1OPHj3ohRdeoIiICPrj\njz+oVIed+KRWSdGtnlZ9yPv3Ezk7E739NlFhocZrqXYlcNeWOlPFmCXGtkHjGqWlRN98I8w2/Pxz\noidbFfOAuzTEji+NVxs3bhz99ttvVFxcLOoNxWKJD5pYtP6H//ZtogEDiDp1Irp0qcrr8kOrjisR\n7YnyZeT8eSFWe/UiunKFB9wlYrRKRAxxcXHk6upKbdq0UW54ryo6Olq5F3Dnzp2Ve11fvnyZgoKC\nSKFQkLu7O0VGRpYvuAU+aFKo8h/+khKipUuJmjQh2rWr0mvxQ6uOKxH9GPRlpKiI6MMPiRwdKf/b\nGB5wl4DFVCKGZPG9fv06paenExFRXl4etWvXrty5lv6giUXrf/gPHBCmAb/zjsbZW9y1pY4rEf2I\n8mUkNVWYvTVihDCbSwXHpmEsphJJTk6mvn37Kn9fsmQJLVmyROPxd+/eJScnpwrfGzJkCCUkJKi9\nZukPmlh0mh75zz9EffoQdetGdO1apYdy1xZXIvoSbcpuQQHR9OnC2J5KlgzV2GzViisUXYkdXxqT\nYN29e7fSAfmGDRtW+r6hWXzLZGVlIT09HYGBgeXes8hMpyIry0cEVDL1t4yjozBja/FiwM8PWL8e\n6NGjwutWx1lbnMVXHKoxqarK+HxW7drAihXAv/4lTF0fNQpYtAh16jwHQIjNmjWfzXot7mexBibL\n4tuiRQtq2bIltWjRgmQyGTVs2JAaNmxIMpmMWrZsWWXtZEgW3zJ5eXnk5+dH27ZtK3dOJUWvtnRq\nPfz+O1GzZkSLFytnw6jiBWDcEhGbQa3bW7eIhg4l8vKie3+cVMYmj+PpTuz40rhOJCsrC5mZmQgO\nDsauXbtw584d3LlzB7t370ZwcHCVlZMhWXwBoKioCCEhIXjttdcwdOhQXerFakun1kPv3sDRo8Cu\nXcDgwcAzLU/OuMrEZlDrtnFj4OefgTfegP3gIGzqEgkH+1K1tSpz5vC6EpOoqpZxd3fX6rVnGZLF\nt7S0lMaMGUMzZ87UeH0til7t6NV6KCwkevNNoXP52DHJymaJTBVj1hrborVuz58nCgwUxvdUxvZ4\nHE87YsdXlVcLDg6m//3vf5SZmUl///03ffjhh9SnTx+tLq5vFt8DBw6QTCYjb29v5XtxcXHqBbfS\nB00sOs9g2bRJWOz13XfiXteCcSViHHrFVGEh0f/9H1HTpkS//EJE3LWlLaNXIrdv36bp06cr/zF/\n44036M4zU+5Mobo9aLrSawbLmTNEbm5EEyZo3DmxOn3b40rEOAyKqT/+EAJ80iTKuZpf7cfxtCF2\nfGmdgPHBgwd4/vnnpehR04ulJKkzFdXkeDVrAn/8IbxeZZLF/Hxg4kTgwgVgyxagVSuN19Ur6Z4F\n4QSMxqEaU8/u6KmV+/eBiAjgyBEhEDt0UHtb51lhVk70+Kqqlvnjjz/Izc2N5HI5EREdP36cpk6d\nKmpNpg8til6tqfY/69zMLy0l+uwzYZV7XJxad0NWVvXZ4MpUMVbdYlu0sZKYGCJHRyFDg8qMQ2uP\nU12JHV9VXi0gIIAuXbpEPj4+ytcUCoWohdBHdXvQDKH3Q7p/P9GLL9J3LT8gGUrKPYTW3gfNlYgF\nysoi6tqVqEcPoitXiMj641RXYsdXlangAaB58+Zqv9vaalyjyMyQ3tN1u3YFjh7FS/d+xXYMQU/f\nHLWpmTy9kklJr1TwLVoAiYlAr17CgtqtW9XitLp3ZUmhykqkefPm+ONJh3phYSGWLVsGNzc3yQvG\npKHzg/nii3A6uxc12rTCb7kBcLicoXxLtXLSuMcJY3rSO6Zq1ADmzQN27ADmzoXD2+HYtDq/XAXC\n+5WIo8pK5KuvvsIXX3yB7OxsODk5IT09HV988YUxysYkoM+D6dCkJgacX4EaH34gfMOLji53jOpC\nsjp1+OFkhjM49U5gIJCeDpSUCIPtqalqb/MXH5GI2jn2DH1TwWtzrsRFt1rP9g/rPEc/I4OoTRui\niAiix4+VL6uOu1jLQKapYoxjW6BpLE+vdSU//SQMui9eTPRkj6TqOlYidnxpvFrZP9wRERHlfqZP\nn17lhQ1JBa/Nufyg6efZB1Ovf/BzcogGDxY2ELp6tdzb1vJwciVinvT+knL5snByt25Ely5V2/xw\nYseXxhFyhUIBAPDz8yv3nkwmq7KFc+TIEbRp0wYtW7YEAIwcORLbt29XG0/p1KmT8s+BgYG4evWq\n1ucy/TybYVWvLgMHB2DbNmDpUuHEmBih/+qJmBihe+Cbb3ggk4mvoq5TrdaAODsDe/YAy5YB/v5w\nWLECmzaNLHcYryvRjcZKZNCgQQCAcePG6XVhQ1LBa3uuJafLNheq/+DPmaPDw2NjA7z7rvAkjxoF\nvPkmMHs2IJNpTAVu7jgVvGVQjdmhQ3VMBV+jBjB3rpCA9NVXha0RVq4E7O2Vh5SNlWh9TTMndVxX\nWYlURCaTYceOHZVeWJvWSpnExESsXr1aOQtM23NVHzSmH9V/8PV6eHr3FlYKDx8OHDoErFljsV/d\nnv3HeuHChSYrC8e2Zqoxq3erxM8POHYMmDUL8PYGfvxRmNIO69tLR+q41liJzJo1S+NJ2vwjr2sq\n+Pj4eGUqeG3PZeIyqJtg/37grbcAf38hXYqPT7nDuJuAic2gVsnzzwOrVgE7dwKvvAKMGwcsXIiY\nmJrcHasLbQZOHj16RMePH6eMjAx6rDIjpzKGpILX5lwti850IMoMqw0bhGzA33wjpE9RYWnbmpoq\nxji29WPQhI6bN4kGDSLy8SH6809JymcuxI6vKq+2a9cuksvl1K1bN+rWrRvJ5XLavXu3VhfXNxW8\npnPVCs4PmqQMeiDPnCFydyd67TWivLwKr9mli/lPA+ZKxLIYPNuqtJTo22+FL0Gff15ux09r2QZB\n7PiqMouvq6srdu/ejTZt2gAALl68iAEDBuCvv/6StIVUleqW6dTYcnMrHnBv0kTLzL0PHgDTp+P6\nz8mY12YTbjTxwldfCWPv33wDjB5tYOZWI+AsvtZB527UCxeEPd3r1gW+/x54kvYpKOhpd1mrVsLL\nltg1a/Qsvv7+/mq/l5aWlnvNFLQoOhOJajeUo6NuLYhFrj/QLTSiyVhFw0Ofdm+JupBMIqaKMY5t\ncenVNVtURLRokdAqWb2aqLTU4lrSmogdX1WmPfHz88OAAQOwdu1arF27FgMHDoS/vz9+/vln/Pzz\nz+LVZsxsqQ64e3s//bM2M1cOtg7DyziIWbW/QvTjUOVe7pqSQnIqCiY2vWZb2doC770nrCtZsQIY\nPBgbPr2uTORYNiO47JrVOQ9XlZXIo0eP0KRJE+zbtw/79u2Do6MjHj16hJ07d2Lnzp3GKCMzMdUs\nqJs365a5NyYG8BreHk3+PoyarZ0BX19hJpcG1ja9kpmeQVl8vbyAlBTA1xf1u/tg09AYONSnctes\nzl9+tN7Z0Nxwv7HpqfYRV7ljYpnYWCA8XOhz/uAD4Lnn1N5WHYsxdT8zj4lYN53HSlJTgbFjgfbt\ngS++AJo1U76laXdGc5zWLnZ8VVmJ/P3334iKikJWVhaKi4uVhahqsaHU+EEzPb23Nb11C5g0SRih\nj44GPDykLKbeuBKxbnp9CXr0CFi4EFi9GvjkE+HLkEym8cuPXveQmNEH1j09PSkyMpL27NlDiYmJ\nlJiYSElJSVoNuFSViffMmTP00ksv0XPPPUfLli1Te2/x4sWkUCjIw8ODRo0aRY8ePVJ7X4uiM4mp\nDo7rPCBeWkr03XdEjRvTz/4fUs+uhWYxmK7KVDHGsW0cBk1jP3aMyNubqF8/osxMae4hEbHjS6vt\ncfWhTSbef/75h44ePUrz5s1Tq0QyMzOpVatWyorjlVdeobVr16oXnB80s6L34sRLlyilQR9KRQfy\nxAm1c009U4srEetm8LqSwkJhBlejRkTLlwszusS+hwTEjq8qB9anT5+OBQsW4NChQ0hLS1P+VEU1\nE6+dnZ0yE68qR0dH+Pv7w87OTu11e3t72NnZoaCgAMXFxSgoKICTk5OWbStmCnoPiDdvjgWB8fgS\n05Bk2xs/Or0DFBQAqN6DlUx6em8bXcbOTpjBdegQsHu3sAnWM4liDb6HBahys/RTp05h3bp1SExM\nhI3N0zonMTGx0vN0zeKrqmHDhpg1axaaN2+O2rVro2/fvujdu3e54zjTqfnQJhuwpkHGmA0yTJ48\nETU+HIhaC98SxkiionDx4r8AAPXrC93PUuMsvgzQYzC8bVsgIQFYvx4YNgzo1w9YskRYmWsGJI/r\nqpoqrVu31jpflqotW7ZQeHi48vd169ZRREREhccuWLBArTvrwoUL5ObmRrdv36aioiIaOnQoRUdH\nq52jRdGZiWjKkaX1Aq1ffyVq146SG/SndjhrsgVdpooxjm3TMmhnznv3iGbNEhYpRkZW2MVlamLH\nV5XdWZ6ensjJydG5cjIkE29qaio6d+6MRo0awdbWFsOGDUNycrLOZWCmodq19eKLT7ukLl58+nql\n+7D36QOcPIk/m/bCH+iCj5uv5DUjzGg0dc1qtaDQ3l7Y9GrfPmD7dmFv97LpWVqwyEWLVdUy3bp1\nIwcHBwoODqaBAwfSwIEDadCgQVXWTtpk4i3z/vvvq7VEjh8/Tu7u7lRQUEClpaUUFhZGK1euVDtH\ni6IzE1EdTFSdnZKVpVuW4JwcovBBN+jeoVNGLX8ZU8UYx7ZpaRoM1zkLdWkp0ebNRM2bC9tJnzyp\nfEvTpBGDWkFaEju+qrxa2bTeZ3+0UVUW3+vXr5NcLid7e3tycHAgZ2dnynuS9XXp0qXKKb5hYWFU\nWFioXnB+0CyCpgdStXIZO9Z88mWp4kqEqdI7d9bDh8LsLUdHorAwosxMjZWFMaYEG70SMVf8oFk2\nUfYukRhXIkyVpha21uukcnOJ5s8natiQfmnxBjniZrnKwhiJScWOL40r1uvWratxB0OZTIb79++L\n3rWmC17Vaz30XvkuMV6xzjR5doW6TivTb97E4/cXo3BNNOwmjUOtuTOF3UErIebKd7HjS+PAen5+\nPvLy8ir8MXUFwqyLajI7bZI6MmZqz67/0GmdVNOmeG5VJOqdT0etmiSkxh4zBjhxQuMp5pyYlBMw\nMrNiTrmGuCXCtKUpd5ZWa05yc4W93lesADw9gbffBnr3BlR6gsRMTGr0BIzmih8062ROXVtciTBD\n6fSl6PFjYcHi8uVAUREwbhwQFgZouTRCW0brzmLMFAza+4ExM6PTmpPnngMmTAD+/BP44Qfg0iVh\nP5O+fYGNG4GHD41dfK1wS4QxDbglwgxlcIr4hw+BbduAtWuBY8eElkq/fgaVyaJaIvHx8Wjfvj3a\ntm2LpUtnQYyHAAAgAElEQVSXlnv/7Nmz6NSpE2rVqoXly5ervZebm4vQ0FC4ublBoVDg8OHDUhaV\nMcZEpykBo9YtlNq1gdGjhab58ePCCWZGspZISUkJXF1dkZCQACcnJwQEBGDDhg1wc3NTHnPr1i1c\nunQJv/zyCxo0aIBZs2Yp3xs7diy6d++OCRMmoLi4GA8ePED9+vWfFpy/rTGJcUuEScWUm1hZTEvE\nkFTw9+7dw4EDBzBhwgQAgK2trVoFwhhjlkzXFoo5qzIVvL4MSQWfmZkJR0dHjB8/HidOnICfnx8i\nIyNRp+xv+AlOl83ExKngmampbqkg1sQSqeNasu6srVu3Ij4+Ht9++y0AIDo6GikpKYiKiip37MKF\nC1G3bl1ld1Zqaio6deqE5ORkBAQEYObMmbC3t8cHH3zwtODc5GcS4+4sZi503uOkEhbTnWVIKni5\nXA65XI6AJ4NIoaGhWu2myBhj1sicd/mUrBLx9/fH+fPnkZWVhcLCQvz0008YPHhwhcc+Wys2a9YM\nzs7OOHfuHAAgISEB7u7uUhWVMcbMmjmPlUi6TiQuLg4zZ85ESUkJJk6ciHfffRdff/01AGDKlCm4\nceMGAgICcP/+fdjY2KBevXo4ffo06tatixMnTiA8PByFhYVwcXHBmjVreHYWMyruzmLmgtOeSIAf\nNCY1rkSYNbKYMRHGGGPWjysRxhhjeuNKhDHGmN64EmGMMaY3rkT0YKpVzaZcTV0dP3N1w/FVfe4t\nJrPN4gsISRx9fX0xaNAgKYupMw54679vdcTxVX3uLSbJcmeVlJQgIiJCLYvv4MGD1bL4NmrUCFFR\nUfjll18qvEZkZCQUCgXy8vKkKiZjjDEDmGUWXwC4evUqYmNjER4eznPmGWPMXJFENm/eTOHh4crf\n161bRxERERUeu2DBAlq2bJnaa6GhoZSWlkZJSUk0cODAcucA4B/+kfzHFEz9mfnH+n/EJFl3lkwm\n0/vcXbt2oUmTJvD19dXYb0jcOmFWimObWRKzzOKbnJyMHTt2oFWrVhg1ahT27t2LsLAwqYrKGGNM\nT2aZxXfx4sW4cuUKMjMzsXHjRvTs2RM//vijVEVljDGmJ8m6s2xtbbFy5Ur07dtXmcXXzc2t0iy+\nkZGRyiy+qgzpGmOMMSYdi83iyxhjzPR4xTpjjDG9cSXCGGNMb1yJMMYY0xtXIowxxvTGlQhjjDG9\ncSXCGGNMb1yJMMYY0xtXIowxxvTGlQhjjDG9cSXCGGNMb1yJMMYY05tJ91hfv349vL294eXlhS5d\nuiAjI0P53pIlS+Du7g5PT0+MHj0ajx8/lrKojDHG9CBZJVK2x3p8fDxOnz6NDRs24MyZM2rHtG7d\nGvv370dGRgbmz5+PyZMnAwCysrLw7bffIi0tDSdPnkRJSQk2btwoVVEZY4zpyaR7rHfq1An169cH\nAAQGBuLq1asAAHt7e9jZ2aGgoADFxcUoKCiAk5OTVEVljDGmJ8n2E8nOzoazs7Pyd7lcjpSUFI3H\nf//99xgwYAAAoGHDhpg1axaaN2+O2rVro2/fvujdu7fa8bzHCDMGU+yUwLHNpCZmXEvWEtHlQUhM\nTMTq1auV4yYXL17E559/jqysLFy7dg35+flYv359ufOIyCQ/77//frW6b3X9zKZU3f6uq2N8mere\nYjP5HusZGRmYNGkSduzYgQYNGgAAUlNT0blzZzRq1Ai2trYYNmwYkpOTpSoqY4wxPZl0j/XLly9j\n2LBhiI6ORps2bZSvt2/fHocPH8bDhw9BREhISIBCoZCqqIwxxvRk0j3WP/jgA+Tk5GDq1KkAADs7\nOxw5cgTe3t4ICwuDv78/bGxs0KFDB+XMLXMQFBRUre5rynub8jNXNxxf1efeYrLYPdZlMpkk/XuM\nlTFVjHFsMymJHV+8Yp0xxpjeuBJhjDGmN65EGGOM6Y0rEcYYY3rjSoQxxpjeuBJhjDGmN7NNBZ+b\nm4vQ0FC4ublBoVDg8OHDUhaVMcaYHiRbJ1JSUgJXV1ckJCTAyckJAQEB2LBhA9zc3JTHHDp0CAqF\nAvXr10d8fDwWLFigrCzGjh2L7t27Y8KECSguLsaDBw+UGX8BnkvPpMfrRJg1sph1Ioakgr937x4O\nHDiACRMmABBWv6tWIMz4rly5gh49esDd3R0eHh5YsWIFAODu3bsIDg5Gu3bt0KdPH+Tm5ipf79Gj\nB+rVq4fp06erXWvevHlo3rw56tWrV+k9jx07Bk9PT7Rt2xYzZswo9/7WrVthY2ODtLS0Cs//9NNP\n4e7uDm9vb/Tu3RuXL19WvtevXz80aNAAgwYN0unvgVkfc4rttWvXwtHREb6+vvD19cXq1asrPH//\n/v3o0KED7OzssHXrVrX33njjDbi7u0OhUFT43IjNLFPBZ2ZmwtHREePHj8eJEyfg5+eHyMhI1KlT\nR+2cBQsWKP8cFBRkNWkEzJGdnR0+++wz+Pj4ID8/H35+fggODsaaNWsQHByMOXPmYOnSpfjoo4/w\n0UcfoVatWvjwww/x559/4s8//1S71pAhQzB9+nS0bdu20ntOnToV33//PTp27IgBAwYgPj4e/fr1\nAwDk5eUhMjISL730ksbzO3TogGnTpqFWrVpYtWoV5syZo9zcbM6cOSgoKFCm4QGApKQkJCUl6fk3\nJC6ObeMxp9iWyWQYNWqUsiLTpEWLFvjhhx+wbNkytdeTkpKQlpaGP//8E0SEl19+GZ9//rmyApQE\nSWTLli0UHh6u/H3dunUUERFR4bF79+4lNzc3unv3LhERHT16lGxtbenIkSNERDRjxgyaP3++2jkS\nFp1pYciQIfT777+Tq6sr3bhxg4iIrl+/Tq6urmrHrVmzRuP/97p162q8/rVr16h9+/bK3zds2EBT\npkxR/j5jxgzavXs3BQUFUWpqapXlTUtLoy5duqi9lpiYSAMHDtR4jqlijGPbtEwZ25VdsyLjxo2j\nLVu2KH8/ffo0BQQE0MOHDyk/P5/8/f3p7NmzaueIHV9mmQpeLpdDLpcjICAAABAaGqqxy4IZX1ZW\nFtLT0xEYGIibN2+iadOmAICmTZvi5s2basfqu8FSdna2Wrw4OTkhOzsbAJCWlobs7Gxly1Wbe6i2\ndBnTxNSxLZPJsHXrVnh5eWH48OHKLn5tubm5oU+fPnjhhRfg5OSEfv36wdXVVa9yasssU8E3a9YM\nzs7OOHfuHAAgISEB7u7uUhWV6SA/Px8hISGIjIws1+8rk8kk35WPiPDWW2+pNeOpikHC6OhopKWl\nYfbs2ZKWjVk2U8c2AAwaNAiXLl1CRkYGgoODMXbsWJ3O379/PxITE5GdnY3s7Gzs2bMHBw8elKi0\nArNMBQ8AUVFRePXVV1FYWAgXFxesWbNGqqIyLRUVFSEkJARjxozB0KFDAQjf0G7cuIFmzZrh+vXr\naNKkiV7XLikpgZ+fH2QyGYYMGYL//Oc/at/Crl69Crlcjry8PJw6dUo5RnDjxg0MGTIE27dvx7Zt\n27B7927IZDJlyzUhIQGLFy/G/v37YWdnp3ZP3oaWlTF1bDs5OQEQtgYvM3HiRMyZMweAMGAfGxur\nFttlVOP48OHD6N+/v3L8uH///jh06BBefvllvcquFVE7x4zIgotukUpLS2nMmDE0c+ZMtddnz55N\nH330ERERLVmyhObOnav2vr79xkREHTt2pMOHD1NpaSn179+f4uLiyh0TFBREx44dq/D8tLQ0cnFx\noQsXLlT4Po+JMCLziu3r168rj/n555+pU6dOlV5n7NixamMi27dvp969e1NxcTEVFhZSr169aNeu\nXWrniB1fFhut/KAZ14EDB0gmk5G3tzf5+PiQj48PxcXF0Z07d6hXr17Utm1bCg4OppycHOU5LVq0\noIYNG1LdunVJLpfTmTNniEh4OOVyOdWoUYPkcjktXLiwwnumpqaSh4cHubi40PTp0ys8prJKpHfv\n3tSsWTNleYcMGaJ87+WXXyZHR0eqXbs2yeVy+u2338qdb2mVyKRJRN27E/XvT6Tyv4FVwZxi+913\n3yV3d3fy9vamnj170l9//VXh+UeOHCG5XE7PP/88NWrUiDw8PJTvzZw5k9zd3UmhUNCsWbPKnSt2\nXPOmVIxpYGmLDYOCgH37hD8PHw5s2iRuuZh1sJjFhowx4ypbRhUQAHzzjWnLwqoPrkQYsxIxMcAH\n3RJwsM04ONy/XPUJjImAKxEdTZ4sdBsMGABIuQiUMV05OADzd3ZEzdZywNcXmDsXyMkxdbGYlTPb\nLL6AMDXO19fXrPIbnTsn9DvHxQkVCmNmxd4e+PBD4ORJoQJxdQWWLwcePQLAX4KY+CSrREpKShAR\nEYH4+HicPn0aGzZswJkzZ9SOad26Nfbv34+MjAzMnz8fk5/5VzkyMhIKhcKs5vPXqQME4jC6dcjn\nfmdmvl58URgYSUoC9u8H2rcHoqNx/q9S/hLERGWWWXwBYQFObGwswsPDzWoWVkwMML/tRiTedIPD\nrz8BZlQ2xspRKIDt24F164CVK/Fduh9643cefGeiMcssvgDw5ptv4pNPPsH9+/c1nmOKTKcODsC/\nzn0OHAwFXn9deBKjooSHlVk0q87i27UrcOgQmv24FT9GTEPjuq1gl/Ux4ONj2HWZ2ZM8rkVddaLC\nkCy+O3fupGnTphGR5lXFEhZde0VFRFFRRI0bE82aRXT/vqlLxERkqhiT/L6FhUQrVxI1bUoUFkZ0\n6ZK092NmRez4MsssvsnJydixYwdatWqFUaNGYe/evQgLC5OqqPqztQUiIoBTp4C7d4V+55gY7uJi\n5s3OTmhFnzsHODsDvr6I83kHA1/O5QF3pjtRqyQVRUVF1Lp1a8rMzKTHjx+Tt7c3nT59Wu2YS5cu\nkYuLCx06dEjjdZKSksy3JfKs5GQiX18h98TJk6YuDTOQqWLM6Pe9epV2NZtIN+FIM/AZjQx5bNz7\nM6MSO74ka4moZvFVKBQYMWKEMotvWSZf1Sy+vr6+6NixY4XXMqfZWZXq1Ak4ehR45RWgZ0/gzTeB\ne/dMXSrGKufkhC98v0NP7MXw+r8hOk0BbNnCLWqmFc6dJZVbt4B33wViY4GlS4HXXsPkKTKcOydM\nE46JEQbpmfmytNxZkydD7/jKzRXO/+YbwCE1AZg9G6hdG1i2DOjcWeeyMPMldlxzJSK1lBSh/7l2\nbUx8uBKrj3kD4AR5lsDSKhFREzCWlgLR0cC8eUBgIPDRR4DKxnHMcnECRksTGChUJK+9hmUn+yAS\nb6CHby7P0WeiEzUBo40NEBYmNG38/ICXXgJmzgTu3DG4nMy6cCViDDVqAFOmwObMabi1foyE6wo4\n7F7Pfc5MVDExQgvkt99E7CqtXVvolj19GiguFmYgfvKJMo1KGU6nUn1xd5aODOl3VkpJAaZOBerX\nB778EnBzE72czHCW1p2lDYPj96+/hMSOx48DixcDI0cCNja8l4kF4e4sExMlAWNgoDCLKyQE6NZN\n+Kb34IGo5WSsIgbHr6sr8MsvwI8/Ap9/LsTyvn28l0k1xpWIjkR7WGrUEBYqnjwJXLkCuLsLOY4s\ns2HILIRo8dutG3D4MPDWW8C4cdiOIZjR96y4XWnMIphlKvgrV66gR48ecHd3h4eHB1asWCFlMXUi\ner9zs2bCLJg1a4B33gEGDwYyM0W4MGPlaYpfvcY0bGyAUaOAM2dg17MrPj/WFQ7vTQP++UeKojNz\nJerSRRXFxcXk4uJCmZmZVFhYWOGK9eTkZMrNzSUiori4OAoMDCQiouvXr1N6ejoREeXl5VG7du3K\nnSth0U3n8WOiJUuIGjUi+vBDokePTF2ias1UMWaK+3bvTiQ0g4mGD9fzIrdvE82YIcTvkiVEBQVi\nFpGJROz4MstU8M2aNYPPk+yidevWhZubG65duyZVUc1HzZpCayQ1VRgz8fICEhIA8OwXJi1Rurka\nNRLGSQ4dEuK3fXtg/XphzQmzWmabCr5MVlYW0tPTERgYWO49U6SCN4qWLYXBy507gUmTgJdewp2s\n5dh3+EUAQoXCs1/EZ9Wp4KsQE6OyYt3Qbtq2bYGtW4EDB4BZs4DISGHle7duopSV6aZapoIvk5eX\nR35+frRt27Zy50hYdPPy4AHRvHmUa9eY3sDn9JJ/EeXkmLpQ1YOpYsyqYrukhCgmhqhFC6KhQ4n+\n+qvcIZMmCd1p/fsTx7YRiB1fZpkKHgCKiooQEhKC1157DUOHDpWqmEalV5dUnTrAhx9CdvAAwpvs\nwIHHAXA4e1jKYjImnrLB97NnhQSlXboAM2YAt28rDxFl2jwzHVGrJBWGpIIvLS2lMWPG0MyZMzVe\nX8KiS8bgwcvSUuFb3QsvEE2eTHTnjthFZCpMFWOWGNtatyb++Yfo9deFjdw++YTo0SPq3194JgIC\nuCViDGLHl6TRGhsbS+3atSMXFxdavHgxERGtWrWKVq1aRUREEydOpIYNG5KPjw/5+PhQQEAAEREd\nOHCAZDIZeXt7K9+Li4tTL7gFPmiiPSw5OcKD2KwZ0Q8/CJULEx1XItrT+QvS2bNEgwcTtWpF+d9v\npOGhpVyBGInY8cVpT4xILd22GGtMUlOB//wHqFtXSJ/C+7yLyhrTnkhlwAChOyogQMc1VElJwuB7\nzZrA8uWcdt4IOBX8E5b4oEmipARYtQpYsAAIDwfmz386X5MZhCsR7Rn0Bam0VJgKPG8e0LEjp52X\nGOfOYupq1BD2K8nIAC5dElojO3eaulSsmnFwEKad69XCtrEBxowRkjv6+3PaeQvDlYi1eOEFYbL/\nd98Bb78NDB0KXL4MgBcqMgtRu7aw2FY17fyyZeXSzqvi2DY9rkSsTe/eQqvEzw/o0AH4+GNcPFvE\nUyiZ5WjSBFi5Ejh4UPhxcwM2bqwwOSlPDzY9HhOxZhcvAhERyDp4Fa/lf4XCgJc5y6oOeEzETOzb\nJ7SubWyElknXrsq39B7Qr8YsakxE3yy+2pxb3ejVbHdxAWJj4Rj1f9heeyQOtJsAh6JbUhaTMZ1V\nGdvduwsbuc2YIYyd/PvfQhMEEu3myHQj6oRhFYZk8dXmXAmLbpYMXqh47x7Rm28SOToSffklUXGx\n2EW0OqaKMY7tSjx8SLR0qbBYMSJCWLzIdCJ2fJllFl9tzq1uDM6yam8PfPqpkBU4JkaYAXP0qKhl\nZEwfOsV2rVrAnDnAmTNC95ZCASxdCjx8KHk5WcXMMouvtudabRbfCoiWZdXLC9i/H1i3TtgAa8gQ\nYNEiIY13NVeds/iakl6x3bixkB04IkLY893VFfjgA6G7q0YNSctraaplFl9tzpWw6NVHTo7QJdC0\nKdF33wkZVytQXbOsmirGOLb18McfRF26EHl4EO3axamAKiF2fJllFl9tz2UGcnAAoqKA2FhhfUmX\nLkB6ernDeBolM3udOwv7lyxaBMyeDfToIQzGM+mJWiWpMCSLrzbnSlj06qmkRGiNNGkitE5UmhzV\nNcuqqWKMY1t7FbaSi4qEWHZyIgoJqXAPk+pM7Pgyyyy+ms5VKzg/aNK4c4doyhS1DME5OcKsmepU\ngRBxJWIJKp3Z9eCBsNd748ZE//kP0bVrpiii2RE7vnixYTUwebLQJVWnjjCIqdXg5dGjwNSpQiqK\nL78EPD0lL6e54cWG5k+rxYZ37gBLlgBr1gh55t5+G7C31++5sAIWtdiQmQe9xjQCAoQ+5VdfBXr1\nAt56C7h3T9JyMqYrrRYbNmokrHRPSxOSlLZrB6xYgb/PFvJYnwi4EqkG9F5jUqOGsF/JqVPA/fvC\nNMpVq4TkeIyZAZ2yB7doAfzwA/Drr0B8PNalumEkNqCjf6nOa6848eNT3J1VDYi2Gdbx48CbbwK3\nbgGffQYEB4tWRnPE3VnWLX9nIm6MnYOWziWwXfQB8K9/ATKZVucGBQmte0BoCW3aJF05xcabUj3B\nD5qJEAHbtwv9ymWputu313i4Jfc7cyVSDRAB27YJm7rVqiUsWOzbt8rKxJITP/KYCDMtmUzYq+TU\nKWEufteuQmK8u3crPJzXmDBzNXkyENRDhgHfDUNu0nFhfcmsWcJ6qd9/rzD1fBlO/PgUVyJMP889\nJzxwp08DRUVCa2TFCuHPKgzO+cWYRNS+4PzHRqgVMjKA6dOFdCrdugGJiRWea9BOjlbGpKngz549\ni06dOqFWrVpYvny52ntLliyBu7s7PD09MXr0aDx+/FjKorJKVDqI6OgoTAHeuxfYvVuYCrx7t/Jb\nHH9jY+aqwi84NWoAo0YJLe3Jk4FJk4CePYXV8CKyqoF5UVedqNAmnfs///xDR48epXnz5tGyZcuU\nr2dmZlKrVq3o0aNHRET0yiuv0Nq1a9XOlbDo7Blap+ouLSXavZvI1ZUoOJjo5EljFVESpooxjm3j\n0GoRbVER0erVRC1bCjGdnCzKvQ3e2sEAYseXSVPBOzo6wt/fH3Z2dmqv29vbw87ODgUFBSguLkZB\nQQGcnJykKiqrgtZdUjKZ8NXq5Elg0CDhG9zUqcJsLsbMjFZdUra2wPjxwF9/CU3qkSOB/v2BI0cM\nurc1dfOaTSp4VQ0bNsSsWbPQvHlz1K5dG3379kXv3r3LHVed0mWbks6puu3shH7lV18VZru4uQEz\nZwqvPdk/RhvGntnFqeCZRjVrCl1bYWHA6tXAsGGAry+wcCHQoYPOlxNtawctVItU8AsWLFDrzrpw\n4QK5ubnR7du3qaioiIYOHUrR0dFq50hYdCa2v/4iCgsjatSI6P/+T8jPpQVTNvmJuDuLVeLhQ6IV\nK4hefJGod2+i2FiNWynoS6otGMSOL5Ongq9IamoqOnfujEaNGsHW1hbDhg1DcnKyVEVlUmvXTlgp\nnJICZGcDbdsC770H3L5d6WnW1ORn1mXyG7UQtHU6Bntm4kHIGODddwEPD2FLBZF2WbSU6fGSVSL+\n/v44f/48srKyUFhYiJ9++gmDBw+u8Fh6Zj52+/btcfjwYTx8+BBEhISEBCgUCqmKyozFxUV4yI4d\nA3JyhMpl9mzg5s0KD+eZXcxclf0Dv/PXmhi/N0zYh2flSuCXX4CWLYXFi//8Y9A9NH2JMruZXaK2\na55RVSr469evk1wuJ3t7e3JwcCBnZ2fKy8sjIqKlS5eSQqEgDw8PCgsLo8LCQrVrS1x0ZgyXLwt7\nlzRoQDRjBlF2tqlLpMZUMcaxbf4q3WPn9GmiyZOJHByIwsOJTp3S6x6aZo8Z2s0rdnxx2hMmKa0G\nx69fBz75BFi7Fhg9WtgzW2VShmT3rQKnPWGaaJWP7tYt4KuvhHVUvr7C4txevbTOz6WJoSlXOHfW\nE/ygWQadEtX98w+wfDnw7bfCwe+8A7RqJf19NeBKhIni0SPhm8ynnwqLGd96S5gq/Nxzel3O0ISq\nnDuLWRSdBsebNAGWLhWaEI0bA/7+wIQJwJkz0t6XMSnVqiXE8cmTwMcfCxVKq1bCfvA3buh8OXNL\nucItESYpg7415eQAkZHCyc7Owhz9kSOFTYakvO8T3BJhUlkYehL+f0Si2+2tqNWzC+wmjwcGDtS7\ndaIL7s56gh+0aqS4GEhIEKYJx8YK/cpjxworh2vWlOy2XIkwqZR1tz6PfHwU8DMinl8D/Pmn8CVp\n3DhhAaOBYyeacCXyBD9o1q/CwfF794DNm4EffwTOnhUeurAwwM9P9IeOKxEmlQoHxzMzhbheuxao\nVw+bnh+HDbJX8dihqagZGyxqTMSQLL65ubkIDQ2Fm5sbFAoFDh8+LGVRmRmqcLFV/fpAeDiwfz9w\n+LDQtTVihLDQa+lSYTEjY2auwjVQrVoB778PXLwIREWhXuZJrD7UHjPi+mJTr681rqfShSSLFkWd\nMKzCkCy+RERhYWH0/fffExFRUVER5ebmqr0vYdGZmah0Lr6q0lKiAweEOfkNGgjZVqOjifLzDbq/\nqWKMY5sRCfH/PPJorstmehwykqh+faKuXYk++4zo0iW9rimsMbGQtCeGZPG9d+8eDhw4gAkTJgAA\nbG1tUV+HxH3MOmi9Yl0mA15+WZganJ0tzIRZvx6Qy4VtfBmzQDExwIDhdfFOaihqbtkgzOSaMwd/\nfJWBey4dkFXPA48nTgM2bgSuXdPqmmWzFsVklll8MzMz4ejoiPHjx+PEiRPw8/NDZGQk6jzzN8CZ\nTq1b2VRGndSuLYyTjBwpLGI8fVrrUzmLLzMn5eK/Vi1g4EDMWzYQB88Vwyf/OKYf24+xdzYCr78O\nNGgg7MZY9tOqFSCTqcW1l5fQPSwmyQbWt27divj4eHz77bcAgOjoaKSkpCAqKqrcsQsXLkTdunUx\na9YsAEICxk6dOiE5ORkBAQGYOXMm7O3t8cEHHzwtOA8+MonxwDozRxUOypeWCl+Y9u9/+mNj87RC\n6d5d2MJaJrOcgXVDsvjK5XLI5XIEBAQAAEJDQ5GWliZJOZl1MrskdYyJpMJuXhsbYXLJtCfdW9nZ\nQFIS0Lu3MAHlX/8SFvOGhIheHrPM4tusWTM4Ozvj3LlzAICEhAS4u7tLVVRmhSwljTZjutJqxbpM\nBrRpI4wPrl0L/P03kJaG73LEr0QkGxOxtbXFypUr0bdvX5SUlGDixIlwc3PD119/DQCYMmUKbty4\ngYCAANy/fx82NjaIjIzE6dOnUbduXURFReHVV19FYWEhXFxcsGbNGqmKyqwQpz1h7BnOzoguHQ3g\nVVEvy4sNmVXitCeMlSeMp/CKdQD8oDHpcSXCrE1uLtCgAVciAPhBY9LjSoRZI4uZncUYY8z6cSXC\nGGNMb1yJMMYY0xtXInowVWoMU6bkqI6fubrh+Ko+9xaT2aaCB4CSkhL4+vpi0KBBUhZTZxzw1n/f\n6ojjq/rcW0ySLTYsKSlBREQEEhIS4OTkhICAAAwePBhubm7KYxo1aoSoqCj88ssvFV4jMjISCoUC\neXl5UhWTMcaYAcwyFTwAXL16FbGxsQgPD+fpjowxZq5E3Z1ExebNmyk8PFz5+7p16ygiIqLCYxcs\nWMHIhpwAAAeQSURBVFBuU6rQ0FBKS0ujpKQkGjhwYLlzAPAP/0j+Ywqm/sz8Y/0/YpKsO0tmwH7X\nu3btQpMmTeDr66ux35C4dcKsFMc2syRmmQo+OTkZO3bsQKtWrTBq1Cjs3bsXYWFhUhWVMcaYnswy\nFfzixYtx5coVZGZmYuPGjejZsyd+/PFHqYrKGGNMT2abCl6VIV1jjDHGJCTqCIuBxo8fT02aNCEP\nDw/la5s2bSKFQkE2NjZ07Ngx5euZmZlUq1Yt8vHxIR8fH5o6daryvdTUVPLw8KA2bdrQG2+8odd9\n3377bWrfvj15eXnRv//9b8rNzVW+t3jxYmrTpg25urrSr7/+qvd9db231J/5v//9L3l5eZG3tzf1\n7NmTLl++bLTPrOneUn/mMsuWLSOZTEZ37twx2mfm2Lau2DZVXGu6dxmpY9usKpH9+/dTWlqa2l/E\nmTNn6K+//qKgoKByD1pFf2FERAEBAZSSkkJERP3796e4uDid7/vbb79RSUkJERHNnTuX5s6dS0RE\np06dIm9vbyosLKTMzExycXGh0tJSve6r672l/sz3799X/nnFihU0ceJEo31mTfeW+jMTEV2+fJn6\n9u1LLVu2VD5oxvjMHNvWFdumimtN9yYyTmybVdqTrl27okGDBmqvtW/fHu3atdP6GtevX0deXh46\nduwIAAgLC9O4mLGy+wYHB8PGRvjrCQwMxNWrVwEA27dvx6hRo2BnZ4eWLVuiTZs2SElJ0eu+ut5b\n6s9cr1495Z/z8/PRuHFjo31mTffWRKzPDABvvfUWPv74Y7XXjPGZObatK7ZNFdea7g0YJ7bNqhLR\nVWZmJnx9fREUFISDBw8CALKzs9VmgTk5OSE7O9ug+6xevRoDBgwAAFy7dk3t+nK5HNnZ2eVeF+O+\nz94bkP4zz5s3D82bN8fatWvx7rvvAjDeZy679w8//IB33nlH+bqUn3n79u2Qy+Xw8vJSe93Y/5+f\nxbFtPbFtirgGjBfbFluJvPjii7hy5QrS09Px6aefYvTo0ZKkR1m0aBFq1qyJ0aNHi35tXe9tjM+8\naNEiXL58GePHj8fMmTNFvba29x43bhzefPNNANJ+5oKCAixevBgLFy5UvkZmsEaDY9u6YtvYcQ0Y\nN7Ylm50ltZo1a6JmzZoAgA4dOsDFxQXnz5+Hk5OTWhP56tWrcHJy0usea9euRWxsLPbs2aN87dn1\nL1evXoVcLhf1vprubYzPXGb06NHKb4nG+swV3VvKz3zx4kVkZWXB29tbeQ0/Pz+kpKQY/TOr4ti2\nztg2VlwDRo7tKkdNjEzTgFNQUBClpqYqf7916xYVFxcTEdHFixfJycmJcnJyiIioY8eOdPjwYSot\nLdV6cOjZ+8bFxZFCoaBbt26pHVc2KPX48WP6+++/qXXr1spBKX3uq8u9pf7M586dU/55xYoV9Npr\nrxntM2u6t9SfWVVFg49SfuYyHNvWE9umiuuK7q1Kytg2q0pk5MiR9MILL5CdnR3J5XL6/vvvadu2\nbSSXy6lWrVrUtGlT6tevHxERbdmyhdzd3cnHx4c6dOhAu3btUl6nbJqai4sLTZ8+Xa/7tmnThpo3\nb17hFLxFixaRi4sLubq6Unx8vN731fXeUn/mkJAQ8vDwIG9vbxo2bBjdvHnTaJ9Z0723bt0q+meu\nWbMmyeVyWr16tdr7rVq1UpsGKfVn5ti2rtg2VVyr3tsUsS0jMoNOYMYYYxbJYgfWGWOMmR5XIowx\nxvTGlQhjjDG9cSXCGGNMb1yJMMbMyqJFi+Dh4QFvb2/4+vriyJEjiIyMxMOHDzWeM2nSJJw9exYA\nymUBZ9Li2VmMMbNx6NAhzJo1C/v27YOdnR3u3r2LR48eoUuXLkhNTUWjRo3KnVNaWqrMxwUI+aqk\nWOHPKsYtEcaY2bhx4wYaN24MOzs7AEDDhg2xZcsWXLt2DT169ECvXr0ACK2Nt99+Gz4+Pjh06BCC\ngoKQlpamdq3bt2+jc+fOiIuLw61btxAaGoqOHTuiY8eOSE5ONvpns1parSZhjDEjyM/PJx8fH2rX\nrh1NmzaN9u3bR0TqK66JiGQyGW3evFn5u2o6/bp169LNmzcpMDCQEhISiIho1KhRdPDgQSIiunTp\nErm5uRnrI1k9i82dxRizPs8//zyOHTuGAwcOIDExESNGjMCSJUsAqCcQrFGjBkJCQiq8RmFhIXr1\n6oUvv/wSXbt2BQAkJCTgzJkzymPy8vJQUFCAOnXqSPhpqgeuRBhjZsXGxgbdu3dH9+7d4enpibVr\n1wJQ3ya7Vq1aGrfNtrOzg7+/P+Lj45WVCBEhJSVFmfSQiYfHRBhjZuPcuXM4f/688vf09HS0bNkS\n9erVw/3797W6hkwmw+rVq3H27Fnlhkx9+vTBihUrlMccP35c3IJXY9wSYYyZjfz8fEyfPh25ubmw\ntbVF27Zt8c033yAmJgb9+vWDk5MT9uzZo7EVAgiViEwmw4YNGzB48GDY29tjxYoVeP311+Ht7Y3i\n4mJ0794dX375pRE/mfXiKb6MMcb0xt1ZjDHG9MaVCGOMMb1xJcIYY0xvXIkwxhjTG1cijDHG9MaV\nCGOMMb39P5uQ/i1woEtjAAAAAElFTkSuQmCC\n", "text": [ "" ] }, { "metadata": {}, "output_type": "display_data", "text": [ "" ] }, { "metadata": {}, "output_type": "display_data", "png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAEKCAYAAADTgGjXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYRiG0Zn/BxjrdpbjWBkUAAAAAElFTkSuQmCC\n", "text": [ "" ] } ], "prompt_number": 7 } ], "metadata": {} } ] }