{ "cells": [ { "attachments": {}, "cell_type": "markdown", "metadata": { "collapsed": false }, "source": [ "# Python: Conditional Value at Risk of potential outcomes\n", "In this example, we illustrate how the [DoubleML](https://docs.doubleml.org/stable/index.html) package can be used to estimate the conditional Value at Risk of potential outcomes. The estimation is based on [Kallus et al. (2019)](https://arxiv.org/abs/1912.12945)." ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "collapsed": false }, "source": [ "## Data\n", "We define a data generating process to create synthetic data to compare the estimates to the true effect." ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "collapsed": false }, "outputs": [], "source": [ "import numpy as np\n", "import pandas as pd\n", "import doubleml as dml\n", "import multiprocessing\n", "\n", "from lightgbm import LGBMClassifier, LGBMRegressor" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "collapsed": false }, "source": [ "The data is generated as a location-scale model with\n", "\n", "$$Y_i = \\text{loc}(D_i,X_i) + \\text{scale}(D_i,X_i)\\cdot\\varepsilon_i,$$\n", "\n", "where $X_i\\sim\\mathcal{U}[-1,1]^{p}$ and $\\varepsilon_i \\sim \\mathcal{N}(0,1)$.\n", "Further, the location and scale are determined according to the following functions\n", "\n", "\\begin{aligned}\n", "\\text{loc}(d,x) &:= 0.5d + 2dx_5 + 2\\cdot 1\\{x_2 > 0.1\\} - 1.7\\cdot 1\\{x_1x_3 > 0\\} - 3x_4 \\\\\n", "\\text{scale}(d,x) &:= \\sqrt{0.5d + 0.3dx_1 + 2},\n", "\\end{aligned}\n", "\n", "and the treatment takes the following form\n", "\n", "$$D_i = 1_{\\{(X_2 - X_4 + 1.5\\cdot 1\\{x_1 > 0\\} + \\epsilon_i > 0)\\}}$$\n", "\n", "with $\\epsilon_i \\sim \\mathcal{N}(0,1)$." ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "collapsed": false }, "outputs": [], "source": [ "def f_loc(D, X):\n", " loc = 0.5*D + 2*D*X[:,4] + 2.0*(X[:,1] > 0.1) - 1.7*(X[:,0] * X[:,2] > 0) - 3*X[:,3]\n", " return loc\n", "\n", "def f_scale(D, X):\n", " scale = np.sqrt(0.5*D + 0.3*D*X[:,1] + 2)\n", " return scale\n", "\n", "def dgp(n=200, p=5):\n", " X = np.random.uniform(-1,1,size=[n,p])\n", " D = ((X[:,1 ] - X[:,3] + 1.5*(X[:,0] > 0) + np.random.normal(size=n)) > 0)*1.0\n", " epsilon = np.random.normal(size=n)\n", "\n", " Y = f_loc(D, X) + f_scale(D, X)*epsilon\n", "\n", " return Y, X, D, epsilon" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "collapsed": false }, "source": [ "We can calculate the true conditional value at risk through simulations. Here, we will just approximate the true conditional value at risk for the potential outcomes for a range of quantiles." ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Conditional Value at Risk Y(0): [0.5467606094959261, 0.7559417564883749, 0.9567242535070148, 1.1530959776797396, 1.34769649731686, 1.5425843074324594, 1.7395359436844482, 1.940354721701296, 2.1469734445741286, 2.361518457569366, 2.586719493648897, 2.8259803249536914, 3.0841842065698133, 3.3684990272106954, 3.6903344145051182, 4.0701961897676835, 4.551586928482123]\n", "Conditional Value at Risk Y(1): [1.110902411746278, 1.3453813031813522, 1.5700384030890744, 1.789671060840732, 2.007332393760465, 2.225459760731946, 2.4461928741399595, 2.6716717587835648, 2.9041560442482157, 3.146142808990006, 3.400855956463958, 3.6723684718264447, 3.9666592590622916, 4.292302995303554, 4.663081975281988, 5.103951906910721, 5.667614205604159]\n" ] } ], "source": [ "tau_vec = np.arange(0.1,0.95,0.05)\n", "p = 5\n", "n_true = int(10e+6)\n", "\n", "_, X_true, _, epsilon_true = dgp(n=n_true, p = p)\n", "D1 = np.ones(n_true)\n", "D0 = np.zeros(n_true)\n", "\n", "Y1 = f_loc(D1, X_true) + f_scale(D1, X_true)*epsilon_true\n", "Y0 = f_loc(D0, X_true) + f_scale(D0, X_true)*epsilon_true\n", "\n", "Y1_quant = np.quantile(Y1, q=tau_vec)\n", "Y0_quant = np.quantile(Y0, q=tau_vec)\n", "\n", "Y1_cvar = [Y1[Y1 >= quant].mean() for quant in Y1_quant]\n", "Y0_cvar = [Y0[Y0 >= quant].mean() for quant in Y0_quant]\n", "\n", "print(f'Conditional Value at Risk Y(0): {Y0_cvar}')\n", "print(f'Conditional Value at Risk Y(1): {Y1_cvar}')" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "collapsed": false }, "source": [ "Let us generate $n=5000$ observations and convert them to a [DoubleMLData](https://docs.doubleml.org/stable/api/generated/doubleml.DoubleMLData.html) object." ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "collapsed": false }, "outputs": [], "source": [ "n = 5000\n", "np.random.seed(42)\n", "Y, X, D, _ = dgp(n=n,p=p)\n", "obj_dml_data = dml.DoubleMLData.from_arrays(X, Y, D)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "collapsed": false }, "source": [ "## Conditional Value at Risk (CVaR)\n", "Next, we can initialize our two machine learning algorithms to train the different nuisance elements (remark that in contrast to potential quantile estimation ml_g is a regressor). Then we can initialize the DoubleMLCVAR objects and call fit() to estimate the relevant parameters. To obtain confidence intervals, we can use the confint() method." ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Quantile: 0.1\n", "Quantile: 0.15000000000000002\n", "Quantile: 0.20000000000000004\n", "Quantile: 0.25000000000000006\n", "Quantile: 0.30000000000000004\n", "Quantile: 0.3500000000000001\n", "Quantile: 0.40000000000000013\n", "Quantile: 0.45000000000000007\n", "Quantile: 0.5000000000000001\n", "Quantile: 0.5500000000000002\n", "Quantile: 0.6000000000000002\n", "Quantile: 0.6500000000000001\n", "Quantile: 0.7000000000000002\n", "Quantile: 0.7500000000000002\n", "Quantile: 0.8000000000000002\n", "Quantile: 0.8500000000000002\n", "Quantile: 0.9000000000000002\n" ] } ], "source": [ "ml_g = LGBMRegressor(n_estimators=300, learning_rate=0.05, num_leaves=10)\n", "ml_m = LGBMClassifier(n_estimators=300, learning_rate=0.05, num_leaves=10)\n", "\n", "CVAR_0 = np.full((len(tau_vec)), np.nan)\n", "CVAR_1 = np.full((len(tau_vec)), np.nan)\n", "\n", "ci_CVAR_0 = np.full((len(tau_vec),2), np.nan)\n", "ci_CVAR_1 = np.full((len(tau_vec),2), np.nan)\n", "\n", "for idx_tau, tau in enumerate(tau_vec):\n", " print(f'Quantile: {tau}')\n", " dml_CVAR_0 = dml.DoubleMLCVAR(obj_dml_data,\n", " ml_g, ml_m,\n", " quantile=tau,\n", " treatment=0,\n", " n_folds=5)\n", " dml_CVAR_1 = dml.DoubleMLCVAR(obj_dml_data,\n", " ml_g, ml_m,\n", " quantile=tau,\n", " treatment=1,\n", " n_folds=5)\n", "\n", " dml_CVAR_0.fit()\n", " dml_CVAR_1.fit()\n", "\n", " ci_CVAR_0[idx_tau, :] = dml_CVAR_0.confint(level=0.95).to_numpy()\n", " ci_CVAR_1[idx_tau, :] = dml_CVAR_1.confint(level=0.95).to_numpy()\n", "\n", " CVAR_0[idx_tau] = dml_CVAR_0.coef.squeeze()\n", " CVAR_1[idx_tau] = dml_CVAR_1.coef.squeeze()" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "collapsed": false }, "source": [ "Finally, let us take a look at the estimated values." ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " Quantile CVaR Y(0) CVaR Y(1) DML CVaR Y(0) DML CVaR Y(1) \\\n", "0 0.10 0.546761 1.110902 0.360683 1.057962 \n", "1 0.15 0.755942 1.345381 0.590911 1.273356 \n", "2 0.20 0.956724 1.570038 0.829543 1.489699 \n", "3 0.25 1.153096 1.789671 1.015038 1.697000 \n", "4 0.30 1.347696 2.007332 1.203284 1.925736 \n", "5 0.35 1.542584 2.225460 1.502494 2.144084 \n", "6 0.40 1.739536 2.446193 1.678826 2.338775 \n", "7 0.45 1.940355 2.671672 1.822482 2.559144 \n", "8 0.50 2.146973 2.904156 2.153119 2.824701 \n", "9 0.55 2.361518 3.146143 2.156969 3.041831 \n", "10 0.60 2.586719 3.400856 2.495657 3.298120 \n", "11 0.65 2.825980 3.672368 2.653846 3.582761 \n", "12 0.70 3.084184 3.966659 2.847948 3.842405 \n", "13 0.75 3.368499 4.292303 3.076347 4.163895 \n", "14 0.80 3.690334 4.663082 3.523163 4.543075 \n", "15 0.85 4.070196 5.103952 3.869020 4.913774 \n", "16 0.90 4.551587 5.667614 4.372097 5.482038 \n", "\n", " DML CVaR Y(0) lower DML CVaR Y(0) upper DML CVaR Y(1) lower \\\n", "0 0.162710 0.558655 0.957745 \n", "1 0.360801 0.821021 1.175284 \n", "2 0.606342 1.052745 1.393604 \n", "3 0.824889 1.205187 1.601061 \n", "4 1.009428 1.397140 1.824750 \n", "5 1.292028 1.712960 2.041147 \n", "6 1.455078 1.902573 2.234534 \n", "7 1.579238 2.065725 2.455107 \n", "8 1.883914 2.422325 2.715407 \n", "9 1.907491 2.406446 2.932027 \n", "10 2.250210 2.741104 3.183526 \n", "11 2.382872 2.924821 3.466440 \n", "12 2.554076 3.141820 3.722848 \n", "13 2.727976 3.424717 4.041284 \n", "14 3.140833 3.905494 4.409746 \n", "15 3.483717 4.254324 4.773177 \n", "16 3.921372 4.822822 5.313209 \n", "\n", " DML CVaR Y(1) upper \n", "0 1.158178 \n", "1 1.371429 \n", "2 1.585793 \n", "3 1.792939 \n", "4 2.026723 \n", "5 2.247020 \n", "6 2.443016 \n", "7 2.663182 \n", "8 2.933996 \n", "9 3.151636 \n", "10 3.412714 \n", "11 3.699082 \n", "12 3.961962 \n", "13 4.286507 \n", "14 4.676405 \n", "15 5.054370 \n", "16 5.650867 \n" ] } ], "source": [ "data = {\"Quantile\": tau_vec, \"CVaR Y(0)\": Y0_cvar, \"CVaR Y(1)\": Y1_cvar,\n", " \"DML CVaR Y(0)\": CVAR_0, \"DML CVaR Y(1)\": CVAR_1,\n", " \"DML CVaR Y(0) lower\": ci_CVAR_0[:, 0], \"DML CVaR Y(0) upper\": ci_CVAR_0[:, 1],\n", " \"DML CVaR Y(1) lower\": ci_CVAR_1[:, 0], \"DML CVaR Y(1) upper\": ci_CVAR_1[:, 1]}\n", "df = pd.DataFrame(data)\n", "print(df)" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "collapsed": false }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4MAAALrCAYAAABER1J5AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdd3RU1d7G8e+ZkkkCCaEXKaFX6SqgiKgIItgRO1ivV68Ve/fae++FIioWRL2CIPKCoiBKFaVI770khCTTzn7/mGRkyATSSJvns1aWctrsszPZ+/zObpYxxiAiIiIiIiIxxVHWCRAREREREZHSp2BQREREREQkBikYFBERERERiUEKBkVERERERGKQgkEREREREZEYpGBQREREREQkBikYFBERERERiUEKBkVERERERGKQgkEREREREZEYpGBQRKScmDp1KldccQWtWrUiOTkZj8dD/fr16devHy+++CI7duwo6yQe0vDhw7Esi1GjRkVsf/jhh7Esi4cffrjQ1yzOueVBamoqlmWxdu3awx77f//3f1iWRUJCAnv37j3s8du3bycuLg7Lsvjtt9+KlL5Ro0ZhWRbDhw8v0vmV2UknnYRlWRE/brebunXr0r9/f8aOHYsxJuq5JZWv+f1NiYiUFAWDIiJlbOfOnfTr14/TTjuNUaNG4ff76du3L+eddx5t27Zl1qxZ3HbbbTRr1ow5c+aUdXJLzIwZM7Asi5NOOqmsk1Iu9O3bl6ZNm5Kdnc3HH3982OM//PBD/H4/HTp04Nhjjy2FFFYsJRVIderUiWHDhjFs2DDOPvtsqlevzvfff89ll13GkCFD8g0IRUQqAldZJ0BEJJalpaVxwgknsHz5ctq0acM777xD7969I47xer2MHj2ahx56iC1btpRRSovuP//5DxdeeCG1atUq1XMrGsuyuPLKK3nggQf44IMPuP766w95/MiRIwG46qqrSiN5Mevss8+OaJk2xvDss89y1113MX78eL744guGDBkScc4555xDjx49qFatWimnVkSkcNQyKCJShm688UaWL19Oamoqv/zyS55AEMDj8XDttdeycOFC2rZtWwapLJ5atWrRpk2bIgV0xTm3Iho+fDhOp5N58+axePHifI/77bff+Ouvv4iLi+PSSy8txRSKZVnccccdtG7dGoD//e9/eY6pVq0abdq0oX79+qWdPBGRQlEwKCJSRlavXh3uDvjCCy9Qo0aNQx5ft27d8APogcaNG8cpp5xCjRo18Hg8NGnShCuvvJK///476nUOHMc2ffp0TjvtNKpXr05CQgJdu3ZlzJgx+aZh9+7d3HLLLTRp0gSPx0Pjxo35z3/+w+7du/M9J9q4v5NOOom+ffsC8OOPP0aMy0pNTT3kuQeaMmUKgwYNok6dOsTFxdGgQQOGDh3K3Llzox6fOw5sxowZLFy4kHPPPZdatWrh8Xho164dzz//fNRufzt27OCVV15h4MCBNG3alISEBJKTk+nevTtPP/002dnZ+d5/YTRs2JD+/fsD8MEHH+R7XO6+M888Mxwo//DDD9x444107tw5fE8NGzZk6NCh/P7774VKx+HGvK1duzbP7+pAe/bs4aGHHqJz584kJSWRmJjI0UcfzWOPPUZmZmah0uL3+xk7diyXXHIJbdq0ITk5mYSEBFq3bs1NN93E5s2bo6Zt9OjRAFxxxRUR36+SGH9qWRYdOnQAYNu2bXn2Hyr/fvjhBwYPHkzdunVxu91Ur16dli1bcumll/LTTz8VOA2TJ08mOTmZ+Ph4xo0bV+R7EZHYpm6iIiJl5NtvvyUYDJKSksKZZ55Z6PONMQwfPpwxY8bgcrk48cQTqVOnDvPnz2fkyJF8+umnjB8/ngEDBkQ9/4MPPuCxxx6ja9euDBgwgLVr1/Lrr78ybNiwcNB3oG3bttG7d29WrFhB9erVGTRoELZt89FHHzF58mTat29f4LQPGDCA+Ph4pkyZQt26dSPSWNBWwAceeIDHHnsMy7Lo1asXjRs3ZunSpXz22WeMHz+ed955hyuvvDLquVOmTOGFF16gefPm9OvXjy1btvDzzz9z++23s2HDBl566aU8x998880cddRRtGjRgh49erBjxw7mzJnD3Xffzddff8306dPxeDwFzoP8XHXVVUyaNImxY8fyzDPP4Ha7I/ZnZWWFH/4P7CJ63XXXsWHDBtq3b8/xxx+Py+Vi2bJlfPbZZ3z55ZeMGzeO8847r9jpO5wlS5YwYMAANmzYQP369TnhhBNwu9389ttvPPDAA4wfP54ZM2YUuAvltm3buOyyy6hWrRpt27alY8eO7N+/n4ULF/Lqq68ybtw4Zs2aRYsWLQCoWrUqw4YN4+eff2bVqlUcf/zx4X0AnTt3LpH7TE9PB0IvaQpq9OjRXHHFFQAce+yx9O3bl6ysLDZu3Mi4ceOoVasWJ5544mGv8/bbb3PDDTdQrVo1Jk2axAknnFC0mxARMSIiUiYuu+wyA5iTTz65SOe/+eabBjC1atUyCxYsCG+3bds89NBDBjApKSlm+/btEec1adLEAMbtdpv//e9/EftGjhxpAFOtWjWTmZkZse/88883gOndu7fZu3dvePuuXbvMcccdZwADmJEjR0acl5uWhx56KGL79OnTDWD69OmT7z3md+53331nABMfH2++//77iH3vvfde+P7+/PPPiH19+vQJp/Ott96K2Ddt2jRjWZZxOp1mw4YNEfuWLFliZs+enSd9u3fvNqeddpoBzDPPPJNnf25er1mzJt97PJjP5zO1a9c2gBk/fnye/WPHjjWAadSokQkGg+HtEyZMMLt3785z/IQJE4zL5TI1a9bM8zvN/X0PGzasQNtzrVmzxgCmSZMmEdszMzNN8+bNDWDuv/9+4/V6w/v2799vLrroIgOYK6644jC58I/09HTz9ddfR1zLmFA+3XPPPQYwAwcOzHPesGHDon4fCyr3u3Lwd88YY7Zv326qVatmAPP555/n2Z9f/jVt2tQAZubMmXnO2bZtm5k/f/4h78G2bXPnnXcawDRv3twsX768SPcmIpJL3URFRMpI7lIRderUKdL5zz33HAAPPvhgRGuHZVk89NBDdOzYkb179/Luu+9GPf/GG29k0KBBEduGDx9OmzZtSEtLi+hquWHDBr788kssy+Ktt96KaNWpUaMGb731VpHuoahy7/3666+nX79+EfuuuuoqBg0ahN/v5+WXX456/rnnnsu//vWviG0nn3wy/fv3JxgMMn369Ih9bdu2pUePHnmuU716dV599VUAPv/88yLfz4HcbjeXX345EL2raO62YcOG4XD8U43nznR5sLPPPpshQ4awa9euPPdV0kaPHs2qVasYNGgQjz76KHFxceF9iYmJvPPOO9SpU4cPP/yQPXv2FOiaSUlJnHnmmRHXglA+PfHEEzRo0IDJkyezb9++Er2XaPbv38+sWbM488wzSUtL4/LLL+fcc88t8Pnbtm2jWrVqUVvy6tSpQ5cuXfI9Nzs7mwsvvJBnnnmGHj16MHv2bFq1alWk+xARyaVuoiIiFdDGjRtZtWoVEAoKDmZZFldccQW33nor06dP5957781zzODBg6Neu23btixbtoxNmzaFt/3000/Ytk23bt1o165dnnM6d+5Mx44d+eOPP4p6SwUWCAT45ZdfAPId03bVVVfx7bff5hv8HOreJ0+eHHHvuYLBIDNmzGDWrFls2bKFrKwsjDHhMYbLly8vwt1Ed/XVV/P8888zefJktmzZEp6IJHecZ+7v92CbN29m4sSJLFu2jLS0NAKBAAB//fVXOI0DBw4ssXQebOLEiQAMHTo06v6qVavSvXt3Jk2axO+//85pp51W4GsvWrSIadOmsWbNGvbv349t20Do+2DbNitXrjxkMFVUjzzyCI888kie7U8++SR33313oa517LHHMmPGDC6//HJuvvlmunTpEhHQ52fnzp2ccsopzJo1i3PPPZexY8eSkJBQqM8WEYlGwaCISBmpXbs2EFo8vLByg5WaNWuSnJwc9ZjmzZtHHHuwxo0bR92ee70DJ0XZuHEjAE2bNs03TU2bNi2VYHDXrl3htOWXnpK8d4AVK1ZwzjnnhIOqaHLHkJWENm3a0KtXL2bNmsXo0aPDQcfIkSMxxnDyySfTrFmziHMeeeQRHn/8cfx+f6mkMZrVq1cDcNlll3HZZZcd8tjclvHD2b9/P5dddhkTJkw45HFH6t46deoUbnnfvXs3v/76Kzt27ODBBx+kXbt2hRrv+8YbbzBo0CA+/PBDPvzwQ5KSkjjmmGM4+eSTueyyy/L9Xt5zzz0EAgFOO+00Pv/88wIFkCIiBaHSRESkjHTr1g2A+fPnEwwGS/3zY/mBsrD3fv755/PXX38xaNAgfvrpJ3bu3InP58MYg9frPSJpzJ0cJnfRdGNMeIbMg9cW/PLLL3n44YfxeDy8/fbbrFixItx6ZozhnnvuCV+jJOS2yuW3fcCAAeGF2vP7adKkSYE+65577mHChAm0adOGr776ik2bNuH1esOtsj179izRezvY2WefzahRoxg1ahTffPMNGzZs4OKLL8bv93P55ZcXau3Ptm3bsnz5ciZOnMiIESPo0KEDM2fO5P7776dly5aMHTs26nlDhgwhISGBH374Ifx9EBEpCWoZFBEpI4MGDeK2225j7969fPPNN5xzzjkFPveoo44CQq1k6enpUVsHc1tpco8tjtxrrF27Nt9jDrWvJNWsWROPx4PX62X16tV07NgxzzElee/Lli3jjz/+oE6dOkyYMAGXK7LqXLFiRbE/I5oLLriAm2++meXLl/PLL7+QlZXFunXrSElJyTNO7bPPPgPg8ccf59prr81zrcKmMXd8Xn7j8NatWxd1e6NGjVi2bBlXXXUV559/fqE+Mz+59/bpp59G/V0fqfzPj8fj4f333+f3339nxYoVPPDAA7z33nsFPt/lcjFw4MBwd9309HReeOEFHnnkEf71r39xzjnnUKVKlYhzTjvtNK677joGDRrE1VdfTUZGBjfddFOJ3peIxKbYfS0sIlLGmjdvzkUXXQTAiBEjDrlWH4S6k+aOS2vYsGG4K2S0lgJjTHh77np+xXHiiSdiWRbz589n2bJlefYvWrSo0F1EcwOO3HFtBeVyucITcOTXSpI7yUpJ3Hvu76VBgwZ5AkEg39ac4qpatSoXXnghELqf3Hu6+OKLiY+Pj5rGaK1t27dvZ+rUqYX67NwgOtrvGv4ZG3iw008/HfgngCsJh7q3KVOmsHPnzqjnFfX7VRDx8fE8/fTTQOg7uHLlyiJfKzk5mYcffpiUlBQyMzPzXR/0xBNPZNq0aVSvXp2bb76ZJ554osifKSKSS8GgiEgZevXVV2nRogVr1qzhhBNO4Oeff85zjM/n44MPPqBLly4sXbo0vP32228H4NFHH2XRokXh7cYYHnvsMRYuXEhKSgrXXHNNsdPZuHFjzjnnHGzb5t///nfE+Kw9e/Zw/fXXF7qbXsOGDYFQy86hxrlFM2LECADefPNNpk2bFrEvtzuf2+3m5ptvLtR1o2nVqhVOp5PFixczY8aMiH3/+9//ePHFF4v9GfnJ7Q762WefhcfMHdxFFELdDwHeeecdfD5feHtaWhrDhg0jLS2tUJ977LHHkpyczJIlS/jwww8j9n3++ee88sorUc+79tpradKkCZ9//jl33XVX1JbFrVu35jvDbTS595Y7a2uu5cuXc9111+V7Xu7361DjPIvjnHPO4bjjjiMYDEadYOZgmZmZvPDCC1HHSs6cOZO9e/fidDrD6Y7mmGOOYcaMGdSrV4/77ruv0BPYiIgcTMGgiEgZql69Or/88gsnnXQSS5cupXfv3jRr1oyzzz6biy++mFNOOYWaNWty1VVXkZGRQYMGDcLn/utf/+Kyyy5j586ddO/enVNPPZWLL76Ytm3b8uCDD5KQkMDHH38cnqimuF5//XWaN2/OjBkzaNq0Keeddx7nnnsuzZo1Y9u2bYWaSANCAWb37t3Zvn07Rx99NJdeeilXX311gR5wTz/9dO6//36ys7Pp168fvXv35pJLLqFbt25cccUVOJ1O3nrrLdq3b1/U2w2rVasW//nPfwgGg5xyyimcdNJJXHzxxXTr1o0zzzyTO+64o9ifkZ8ePXrQrl07MjIyyM7OpnPnznTt2jXPcbfccgspKSlMmjSJZs2acf7553PWWWfRpEkTFi1axJVXXlmoz01ISAgHOJdffjm9evViyJAhdOjQgaFDh+b7O6pSpQoTJ04kNTWVZ555hsaNG9OnTx8uueQSzjnnHNq3b0+DBg144IEHCpyWhx56CMuyeOCBB+jYsSMXXXQRp5xyCkcffTTNmjWjV69eUc87++yzcTgcvPLKK/Tr148rr7ySq6++mm+++aZQeXEoTz75JACffPLJYWeT9fl8jBgxgnr16tG5c2eGDBnCxRdfTK9evejTpw8A991332H/Xo8++mhmzpxJ48aNefrpp7nhhhuO2HhJEan8FAyKiJSxOnXqMH36dL777jsuv/xynE4n06ZN44svvmDJkiX07NmTl156iTVr1nDssceGz7MsizFjxvDxxx9zwgknMG/ePL744gsyMzMZPnw4CxYsCHfbKwn16tVjzpw53HjjjSQmJvLtt9/y+++/c+GFF/Lrr79GXePucMaPH8/FF19Meno6n376Ke+//z7jxo0r0LmPPvoo3333HaeffjpLly7ls88+Y/PmzQwZMoRZs2YVOgA6lBdffJH333+fLl26MG/ePCZNmkRiYiLjxo3j0UcfLbHPiebAlsD87qlp06YsWLCASy65BKfTybfffsuiRYu46KKLWLBgAY0aNSr0595yyy2MHj2arl27smDBAr7//nvq1q3L999/f8i8bd++PX/88QfPPPMMbdu25Y8//uDzzz9nzpw5VKlShdtvv/2wM4Me6Nxzz+XHH3/klFNOYcuWLXzzzTds376dhx9+mO+++w632x31vI4dOzJ+/Hh69uzJnDlzGDVqFO+//z7z588vdF7kp2/fvuG1KR9++OFDHlu1alXeeusthg4ditfrZerUqXz11Vds376dc889l2nTphWohRGgRYsW/Pzzz7Rq1Yo33niD4cOHl8kkVCJS8VlGr5NERERERERijloGRUREREREYpCCQRERERERkRikYFBERERERCQGKRgUERERERGJQQoGRUREREREYpCCQRERERERkRikYFBERERERCQGKRgUERERERGJQQoGRUREREREYpCCQRERERERkRikYFBERERERCQGKRgUERERERGJQQoGRUREREREYpCCQRERERERkRikYFBERERERCQGKRgUERERERGJQQoGRUREREREYpCCQRERERERkRikYFBERERERCQGKRgUERERERGJQQoGRUREREREYpCCQRERERERkRikYFBERERERCQGKRgUERERERGJQQoGRUREREREYpCCQRERERERkRikYFBERERERCQGKRgUERERERGJQQoGRUREREREYpCCQRERERERkRikYFBERERERCQGKRgUERERERGJQQoGRUREREREYpCCQRERERERkRikYFBERERERCQGKRgUERERERGJQQoGRUREREREYpCCQRERERERkRikYFBERERERCQGKRgUERERERGJQQoGRUREREREYpCCQRERERERkRikYFBERERERCQGKRgUERERERGJQQoGRUREREREYpCCQRERERERkRikYFBERERERCQGKRgUERERERGJQQoGRUREREREYpCCQRERERERkRikYFBERERERCQGKRgUERERERGJQQoGRUREREREYpCCwXJi06ZNXHrppdSsWZOEhASOPvpo5s6dW9bJEhGRGKa6SUSkcnOVdQIE9uzZw/HHH0/fvn357rvvqF27NitWrKB69eplnTQREYlRqptERCo/yxhjyjoRse7uu+/ml19+YebMmWWdFBEREUB1k4hILFA30XLgm2++oXv37gwZMoQ6derQpUsX3n333bJOloiIxDDVTSIilZ9aBsuB+Ph4AG677TaGDBnC77//zs0338xbb73FsGHDop7j9Xrxer3hf9u2ze7du6lZsyaWZZVKukVEBIwx7Nu3jwYNGuBwVJ53rKqbREQqrgLXTUbKnNvtNj179ozYduONN5oePXrke85DDz1kAP3oRz/60U85+dmwYcORri5Kleom/ehHP/qp+D+Hq5s0gUw5UL9+fdq1axexrW3btowfPz7fc+655x5uu+228L/T0tJo3Lgxa9asISkp6Yiltaz4/X6mT59O3759cbvdZZ2cCkf5VzzKv+Kp7Pm3b98+mjZtWunKXtVNh1fZv9tHmvKveJR/xVPZ86+gdZOCwXLg+OOPZ/ny5RHb/v77b5o0aZLvOR6PB4/Hk2d7jRo1SE5OLvE0ljW/309iYiI1a9aslH+wR5ryr3iUf8VT2fMv954qWzdI1U2HV9m/20ea8q94lH/FU9nzr6B1U+UZ3FCB3Xrrrfz666888cQTrFy5ko8//ph33nmHG264oayTJiIiMUp1k4hI5adgsBw45phjmDBhAp988gkdOnTg0Ucf5aWXXuKSSy4p66SJiEiMUt0kIlL5qZtoOTFo0CAGDRpU1skQOaxgMIjf7y/rZJQqv9+Py+UiOzubYDBY1smpcCpD/sXFxVWqmUILqjTqJtu28fl8R/QzjpTK8N0uS7Gcf263G6fTWdbJEFEwKCIFY4xh69at7N27t6yTUuqMMdSrV48NGzZUunFhpaEy5J/D4aBp06bExcWVdVIqFZ/Px5o1a7Btu6yTUiSV4btdlmI9/1JSUqhXr15M3ruUHwoGRaRAcgPBOnXqkJiYGFOVl23bZGRkULVq1ZhsHSquip5/tm2zefNmtmzZQuPGjWPqu38kGWPYsmULTqeTRo0aVdjvRkX+bpe1WM0/YwyZmZls374dCM3cK1JWFAyKyGEFg8FwIFizZs2yTk6py+3GFh8fH1MPLCWlMuRf7dq12bx5M4FAoFLOOlcWAoEAmZmZNGjQgMTExLJOTpFUhu92WYrl/EtISABg+/bt1KlTR11GpczE1l+eiBRJ7hjBivrAJlJcud1DY21c05GUm5fqeiuxKrdOjbVx+FK+KBgUkQJT9ziJVfruHznKW4lV+u5LeaBgUEREREREJAYpGBQREREREYlBCgZFRAph1KhRpKSklHUyCqUiprmopk2bRtu2bQs1tu/CCy/k+eefP4KpEjm0ivg3WhHTXFRFKVfuvvtubrzxxiOYKpGSoWBQRCqt4cOHY1lWnp8BAwYU6PzU1FReeumliG1Dhw7l77//PgKpjVQWD1rTp09n4MCB1KxZk8TERNq1a8eIESPYtGkT48ePx+l0smnTpqjntmzZkttuu+2wnzFjxoyI30Xt2rUZOHAgixcvzvecDz/8kCpVqrBy5cqI7Zs3b6Z69eq89tpr4W133nkn999/f8TMfDNmzKBr1654PB5atGjBqFGjIq5z//338/jjj5OWlnbY9IscWK44nU6qV6+O0+lUuZKPw5UrNWvWrHDlypYtW7j44otp1aoVDoeDW265Jc/1b7/9dkaPHs3q1asPm36RsqRgUEQqtQEDBrBly5aIn08++aTI10tISKBOnTolmMLy4e233+bUU0+lXr16jB8/niVLlvDWW2+RlpbG888/z5lnnknNmjUZPXp0nnN/+uknVq5cyVVXXVXgz1u+fDlbtmxhypQpeL1ezjjjDHw+X9RjL7vsMvr378/w4cMjFie/5ppr6NatGzfccAMAP//8M6tWreK8884LH7NmzRrOOOMM+vbty8KFC7nlllu4+uqrmTJlSviYDh060Lx5c8aOHVvg9Etsyy1XNm3axLJly9i0aZPKlSgKUq7UqFGDMWPG5Dm3PJcrXq+X2rVrc//999OpU6eo169Vqxb9+/fnzTffLHD6RcqEkUohLS3NACYtLa2sk3JE+Hw+89VXXxmfz1fWSamQipt/WVlZZsmSJSYrKyu8zbZtY3vL4Me2C5zuYcOGmbPOOivf/bZtm4ceesg0atTIxMXFmfr165sbb7zRGGNMnz59DBDxEwwGzciRI021atXC13jooYdMp06dzPvvv28aNWpkqlSpYv7973+bQCBgnn76aVO3bl1Tu3Zt89hjj0V89vPPP286dOhgEhMTTcOGDc2///1vs2/fPmOMMdOnT8/z2Q899JAxxpjs7GwzYsQI06BBA5OYmGiOPfZYM3369Ihrjxw50jRq1MgkJCSYs88+2zz33HMRaT7Yhg0bTFxcnLnlllui7t+zZ48xxpjbbrvNtGzZMmo+H3fccfneV1pamtmzZ48JBoPhe8u9pjHGfPPNNwYwixYtyjeN27dvN7Vr1zbPPvts+B6rVatm1q9fHz7mhhtuMOeff37EeXfeeadp3759xLahQ4ea/v37R2x75JFHzAknnJDv50f7G8hV2cvf4jhU3hycp2VWphSjXAkGg+Hvdq7ClivGmJgtV4LBoLnhhhuKVK7k3teB91Za5cqB+vTpY26++eao+0aPHm0aNmyY77mHKlcKQs9GxVPZ86+gdZMWnReRovHD3qf3lvrHptyVAiW0LNn48eN58cUXGTduHO3bt2fr1q0sWrQIgC+//JJOnTpx7bXXctVVV7Fv3758r7Nq1Sq+++47Jk+ezKpVqzj//PNZvXo1rVq14scff2TWrFlceeWVnHrqqRx33HEAOBwOXnnlFZo2bcrq1au5/vrrufPOO3njjTfo1asXL730Eg8++CDLly8HoGrVqgD85z//YcmSJYwbN44GDRowYcIEBgwYwOLFi2nZsiVz5szhqquu4sknn+Tss89m8uTJPPTQQ4fMh88//xyfz8edd94ZdX9ut7KrrrqKF154gZ9++okTTzwRgIyMDL744gtefPHFfO/rrrvu4sknn4x67bS0NMaNGwccer252rVr884773DRRRfRqVMnbr31Vl5++WUaNWoUPmbmzJlcfPHFEefNnj2bU089NWJb//7983TrOvbYY3n88cfxer14PJ580yFHUBmVKVA25co111xzyOvEQrli2zaXXnopr7/+eqHLldz7iuZIlysFdeyxx7Jx40bWrl1Lampqka4hcqQpGBSRSu3bb78NP/Dkuvfee7n33ntZv3499erV49RTT8XtdtO4cWOOPfZYAGrUqIHT6SQpKYl69eqFFweOxrZtPvjgA5KSkmjXrh19+/Zl+fLlTJo0CYfDQevWrXn66aeZPn16+KHtwGAkNTWVxx57jOuuu4433niDuLg4qlWrhmVZ1KtXL3zc+vXrGTlyJOvXr6dBgwZAaFzK5MmTGTlyJE888QQvv/wyAwYMCD+AtWrVilmzZjF58uR8079ixQqSk5OpX7/+IfOyXbt29OjRgw8++CD80PbZZ59hjOHCCy885H0dHAw2bNgQgP379wNw5pln0qZNm0N+/tlnn80FF1zAgAEDGDx4MMOGDYvYv27dunC+5Nq6dSt169aN2Fa3bl3S09PJysoiISEBgAYNGuDz+di6dStNmjQ5ZDpESqpcOZRYKVfatGlT5HLl4GCwtMqVgso9b926dQoGpdxSMCgiRePOeZteBp9bGH379s0zZqNGjRoADBkyhJdeeolmzZoxYMAABg4cyODBg3G5Clc0pqamkpSUFP533bp1cTqdOByOiG3bt28P//uHH37gySefZNmyZaSnpxMIBMjOziYzMzPfwHPx4sUEg0FatWoVsd3r9VKzZk0Ali5dyjnnnBOxv2fPnod8aDPGFHjx4yuvvJJbb72VV199laSkJD744AOGDBkSvv9D3VdycnL4OjNnziQxMZFff/2VJ554grfeeqtAn//AAw8wZswY7r///jz7srKyiI+PL9B1DpYbFGZmZhbpfCkBZVWm5Hx2YeSWK7Ztk5GRQdWqValVqxagciVXYcqV4cOHM2LEiCKVKwfel8oVkcLTBDIiUiSWZWHFlcFPAR8uclWpUoUWLVpE/OQGg40aNWL58uW88cYbJCQkcP3113PiiSfi9/sL9Rlud+STpGVZUbflTlKwdu1aBg0aRMeOHRk/fjzz5s3j9ddfB8h3sgMIdZ1yOp3MmzePhQsXhn+WLl3Kyy+/XKg0H6hVq1akpaWxZcuWwx6b+6b+s88+Y8WKFfzyyy/hCR4OdV8H52nTpk1p3bo1w4YN4+qrr2bo0KEFSmvuA3W0B+tatWqxZ8+eiG316tVj27ZtEdu2bdtGcnJy+EENYPfu3UCo25iUjTIrU4pZrjRr1kzlShSlUa4cfF+lVa4UlMoVqQgUDIpITEtISGDw4MG88sorzJgxg9mzZ4enI4+LiyvUulIFNW/ePGzb5vnnn6dHjx60atWKzZs3RxwT7bO7dOlCMBhk+/bteQLc3G5fbdu2Zc6cORHn/frrr4dMz/nnn09cXBzPPPNM1P179+4N/39SUhJDhgzhgw8+YOTIkbRq1YrevXsX+L6iueGGG/jzzz+ZMGHCYY89lC5durBkyZKIbT179mTatGkR26ZOnUrPnj0jtv355580bNgw3LojUhwqVyp3uVJQf/75J263m/bt2xcrDSJHkrqJikil5vV62bp1a8Q2l8tFrVq1GDVqFMFgkOOOO47ExETGjh1LQkJCeMxYamoqP/30ExdccAE+ny+im2NxtGjRAr/fz6uvvsrgwYP55Zdf8nRnSk1NJSMjg2nTptGpUycSExNp1aoVl1xyCZdffjnPP/88Xbp0YceOHUybNo2OHTtyxhlncNNNN3H88cfz3HPPcdZZZzFlypRDduWCUEvGiy++yH/+8x/S09O5/PLLSU1NZePGjYwZM4aqVatGLMp+1VVX0bt3b5YuXcpdd91VqPuKJjExkWuuuYaHHnqIs88+u9CtNLn69++fZ+mL6667jtdee40777yTK6+8kv/7v//js88+Y+LEiRHHzZw5k9NOO61InyuxJ7dcsW2bffv2kZmZSVxcXKHKlQsvvBCPx1NiLyAqYrny7LPPho+vSOUKwMKFC4FQy+qOHTtYuHAhcXFxtGvXLnzMzJkz6d27d0QvBJGCumHiDbSv054ru1xJvKtoXZULpBRmNpVSUNmnNq/s0/8eaUdiaYmKYNiwYXmmUgdM69atjTHGTJgwwRx33HEmOTnZVKlSxfTo0cP88MMP4fNnz55tOnbsaDwez2GXljj4cw9e0uLg6cdfeOEFU79+fZOQkGD69+9vxowZk2dq9Ouuu87UrFkzYgp4n89nHnzwQZOammrcbrepX7++Oeecc8wff/wRPu/99983DRs2NAkJCWbw4MGHnQI+19SpU03//v1N9erVTXx8vGnTpo25/fbbzebNm/Mc27p1a+N0OvPsy+++1q5dm+/SEsYYs379euNyucynn356yDSuWbPGAGbBggV59u3atcvEx8ebZcuWRWyfPn266dy5s4mLizPNmjUzI0eOjNiflZVlqlWrZmbPnp3v52ppiaIpzNISFUVJlyvG5L+0xMGfW9nKlYOX5ihsuZJ7X2VRrkT7DjRp0iTimNatW5tPPvkk38/V0hJlqzzn34ItCwwPY6yHLbN0x9IiXaOgdZNljDFHLtSU0pKenk61atVIS0srsdaL8sTv9zNp0iQGDhyYZ8yEHF5x8y87O5s1a9bQtGnTIg+kr8hs2yY9PZ3k5OSIyRukYEoz/+644w7S09N5++23C3zOm2++yYQJE/j+++/zPeZQfwOVvfwtjkPlTWUoV1Q2FE9Fyb+ilCvfffcdI0aM4I8//sh38qDi/g3o2ah4ynP+nf/Z+YxfOp6hrYbyyYWfFKllu6B1U/n9yxMRESmk++67jyZNmoQn1SgIt9vNq6++egRTJSIVWVHKlf379zNy5MhCzyIr8uf2Pxm/dDwAd6feDfnP/1Qi9A0VEZFKIyUlhXvvvbdQ51x99dVHKDUiUhkUpVw5//zzj1BqpLJ7fObjAJzb6FzaxB16ncySoJZBERERERGRMrZs5zI+/fNTAO5qfNdhji4ZCgZFRERERETK2OMzH8dgOLPRmRwdd3SpfKaCQRERERERkTK0YtcKPl78MQB3N7kbq0rRlkMpLAWDIiIiIiIiZeiJn5/ANjYDGw2kS0IXLI+CQRERERERkUpt9Z7VfLjoQyDUKkiV0vtsBYMiIiIiIiJl5MmZTxI0QU5reBrdE7pjxZdOqyAoGBQRERERESkT6/auY9SiUQDc0/QerASrSIvMF5WCQRERKZBdu3ZRp04d1q5dW+Bz3nrrLc4888wjlygRqdB2795NvXr1ClWuTJ48mc6dOxdqEXiR8uqpn58iYAc4+aiTOS7+OKzE0gsEQcGgiFRSlmUd8ufhhx8u1fSsXLmSK664goYNG+LxeGjatCkXXXQRc+fOZdu2bbjdbsaNGxf13KuuuoquXbsW6HMOvMfk5GSOOeYYvv7663yP//vvv0lMTOTjjz+O2G7bNr169YpYOPnxxx/nrLPOIjU1Nbxt/fr1nHHGGSQmJlKnTh3uuOMOAoFAeP+VV17JggULmDVrVoHSL1KeHVyOOJ1OqlevjtPpVLlygMKUK88//zxnnnlmRLly00030a1bNzweD507d85z/QEDBuB2u/noo48KlH6R8mpD2gbeX/A+APc0uwfL80+rYHBHEOMzRzwNCgZFpFLasmVL+Oell14iOTk5Ytvtt98ePtYYExHAlLS5c+fSrVs3/v77b95++22WLFnChAkTaNOmDSNGjKBu3bqcccYZfPDBB3nO3b9/P5999hlXXXVVgT9v5MiRbNmyhblz53L88cdz/vnns3jx4qjHtmrViqeeeoobb7yRLVu2hLc///zzrF69mrfeeguAzMxM3n///Yh0BINBzjjjDHw+H7NmzWL06NGMGjWKBx98MHxMXFwcF110Ee+8806B0y9SXkUrV5YtW8amTZtUrhygMOXKhx9+yJVXXpnnGldeeSVDhw7NNz3Dhw/nlVdeKXD6RcqjZ355Br/t58QGJ3JCwgnh5SSM3+Cd4yV7ejbBHcEjmwgjlUJaWpoBTFpaWlkn5Yjw+Xzmq6++Mj6fr6yTUiEVN/+ysrLMkiVLTFZWVnibbdsmw5tR6j+2bRc6/SNHjjTVqlUL/3v69OkGMJMmTTJdu3Y1brfbTJ8+3QwbNsycddZZEefefPPNpk+fPmbPnj0mGAyaYDBonnjiCZOammri4+NNx44dzeeff57vZ9u2bdq3b2+6detmgsFgnv179uwxxhjzzTffGIfDYdatW5cn7fHx8WbPnj3mu+++M8cff7ypVq2aqVGjhjnjjDPMypUrI44HzIQJE8L/Tk9PN4B5+eWXD5nGvn37mjPOOMMYY8zSpUtNfHy8+frrr8PHfP7556Z27doR502aNMk4HA6zdevW8LY333zTJCcnG6/XG942ffp0ExcXZzIyMvJNQ3kX7W8gV2Uvf4vjUHlzcJ6WVZlS3HIlt2woSrmSK1bLlU8//dTUqlUr6j0YY8xDDz1kOnXqFHXfunXrDJAnrRXJocqVgtCzUfGUdf5tTt9sPI96DA9jvvu/70zW71nGu9xrvMu9Zt+EfWb3f3ebPc/vMcHM6H8fh1PQusl1ZENNEamsMv2ZVH2yaql/bsY9GVSJK5k5l++++26ee+45mjVrRvXq1Qt0zpNPPsnYsWN56623aNmyJT/99BOXXnoptWvXpk+fPnmOX7hwIX/99Rcff/wxDkfezhgpKSkADBw4kLp16+ZpWRs5ciTnnnsuKSkp7N+/n9tuu42OHTuSkZHBgw8+yDnnnMPChQujXjsQCPD++6HuJ3Fxcfnek2VZjBw5ko4dO/Luu+/y/vvvc+GFF0aM9Zs5cybdunWLOG/27NkcffTR1K1bN7ytf//+/Pvf/+avv/6iS5cuAHTv3p1AIMCcOXM4+eST802HxLayKlNA5UpZlSs///wznTp1KlAeHaxx48bUrVuXmTNn0rx58yJdQ6QsPTvrWbxBL73q96JPfJ9/WgUDBv/ffgBcLVxYjiM7hlDBoIjErP/+97/069evwMd7vV6eeOIJfvjhB3r27AlAs2bN+Pnnn3n77bejPrStWLECgDZt2hzy2k6nk2HDhjFq1CgeeOABLMti1apVzJw5k6lTpwJw3nnnRZzzwQcfULt2bZYsWUKHDh3C2y+66CKcTidZWVnYtk1qaioXXHDBIT+/SZMmvPTSS1x99dU0bNiQ77//PmL/unXraNCgQcS2rVu3RgSCQPjfW7duDW9LTEwkOTmZdevWHTINIpWBypV/FKRcqV+//iGvcSgNGjRQuSIV0raMbbw1N9Rd+t5W94IDLGco6AusDIAXrCoWzvrOI54WBYMiUiSJ7kQy7skok88tKd27dy/U8StXriQzMzPPg57P5wu3gh3MmIIP/r7yyit56qmnmD59OieffDIjR44kNTU13Jq2YsUKHnzwQebMmcPOnTvDM+mtX78+4qHtxRdf5NRTT2X16tXceuutvPLKK9SoUeOwn3/FFVfwwAMPcOONN5KcnByxLysri/j4+ALfy8ESEhLIzMws8vlS+ZVVmZL72SVF5Uqkw5UrtWvXLvC9HEzlilRUz89+nqxAFsfVO46T40/GkRRqhTcBg395qFXQ3cp9xFsFQcGgiBSRZVkl1q2qrFSpEpl+h8OR5yHL7/eH/z8jI/SgOnHiRI466qiI4zweT9TPaNWqFQDLli3L98EuV8uWLenduzcjR47kpJNOYsyYMVxzzTXhmcUGDx5MkyZNePfdd2nQoAG2bdOhQwd8Pl/EderVq0eLFi1o0aIFI0eOZODAgSxZsoQ6deoc8vMBXC4XLlfeqqFWrVrs2bMnz+f89ttvEdu2bdsW3negPXv2FOuhTyq/ylCmgMqVaA5Vruzdu/ew5+dn9+7dKlekwtmxfwev//46APe0uSc0U68rp1VwdQCTbbASLZyNneA/1JVKhmYTFRHJUbt27YiZ7yA0NidXu3bt8Hg8rF+/PvxQlPvTqFGjqNfs3Lkz7dq14/nnn4+6JtbBD0JXXXUV48ePZ/z48WzatInhw4cDoTX+li9fzv33388pp5xC27Zt8wRn0Rx77LF069aNxx9//LDHHkqXLl1YsmRJxLaePXuyePFitm/fHt42depUkpOTadeuXXjbqlWryM7OPuxDq0hlpHIlf126dGH58uVFOjc7O5tVq1apXJEK58VfXyTTn0m3ut04Lf60f8YKBg3+ZTmtgm1Lp1UQFAyKiISdfPLJzJ07lzFjxrBixQoeeugh/vzzz/D+pKQkbr/9dm699VZGjx7NqlWrmD9/Pq+++iqjR4+Oes3cSRT+/vtvevfuzaRJk1i9ejV//PFHeN2+Aw0ZMgS3282//vUvTjvttPDDYPXq1alZsybvvPMOK1eu5P/+7/+47bbbCnRft9xyC2+//TabNm0qYs6EJob566+/Ih4UTzvtNNq1a8dll13GokWLmDJlCvfffz833HBDRIvGzJkzSU1N1SQPEpNUruTvtNNOY9myZXkC0JUrV7Jw4UK2bt1KVlYWCxcuZOHChRGtlb/++isejyc8zlKkItidtZtXf3sVgHvb3YtlW1hxOa2CawKYLIOVYOFqWnqdNxUMiojk6N+/Pw888AB33nknxxxzDPv27ePyyy+POObRRx/lgQce4Mknn6Rt27YMGDCAiRMn0rRp03yve+yxxzJ37lxatGjBNddcQ9u2bTnzzDP566+/eOmllyKOTUxM5MILL2TPnj0Ra285HA7GjRvHvHnz6NChA7feeivPPvtsge5rwIABNG3atFhv8Y8++mi6du3KZ599Ft7mdDr59ttvcTqd9OzZk0svvZTLL7+c//73vxHnjhs3Lk8+isQKlSv5O/roo+nUqVNEuQJw9dVX06VLF95++23+/vtvunTpQpcuXdi8eXP4mE8++YRLLrmExMSSG+8pcqS99OtLZPgy6FS7E6cnnv5Pq6B9QKtgG3d4MpnSYJnCjEKWcis9PZ1q1aqRlpaWZ4B2ZeD3+5k0aRIDBw7E7XaXdXIqnOLmX3Z2NmvWrKFp06bFmkSkorJtm/T0dJKTk6NOtR4rJk6cyB133MGff/5Z4Hz466+/OPnkk/ntt99o1KhRhc2/Q/0NVPbytzgOlTeVoVxR2VA8tm3z+eef88gjjxSqXNm5cyetW7dm7ty5hwyYy7vi/g3o2ah4Sjv/9mbvpclLTUj3pvPpKZ9ypuNMHLVC33n/aj++333ggcRBiVguC+MzGK/B08GD5Sl8cFjQukkTyIiISIGcccYZrFixgk2bNuU7lulgW7ZsYdSoUVSrVu0Ip05EKqL+/fuzefPmQpUra9eu5Y033qjQgaDEnlfmvEK6N50OtTswKHFQuH+msQ3+pQe0CrpKr1UQFAyKiEgh3HLLLYU6/tRTTw23noiIRHPzzTcXqmW1e/fuhV7CQ6QspXvTefHXFwG4++i7sbItrJqhoC+4IYjJMBAH7ual38KrYFBEREREROQIee2319ibvZc2NdtwTtI5oYljLAtjDL4loYmR3K3dWO7SbRUETSAjIiIiIiJyRGT4Mnhh9gsA3Nv5XqxMCysxp1VwYxCTbsAN7hYHtQragJXzcwQpGBQRERERETkC3vz9TXZl7aJljZacm3QulsfCcoRaBf1LcsYKtnKHl5jIZfYbnNWdebaXNAWDIiIiIiIiJSzTn8mzs0LLtdzT5R6cmc5/WgU3B7H32uACd8vIVkGTHWotdNZxHvE0KhgUEREREREpYW/PfZsdmTtoVr0ZF1S/AFxgOXNaBf/KaRVs6c6zdITJMDhrO3FUOfKhmoJBERERERGREpTlz+KZWc8AcHfXu3FmOMOLzAe3BrH32OAMdRE9kJ1lgwectY98qyAoGBSRYjJ+g8kupR+/KevbzZcxhmuvvZYaNWpgWRYLFy7kpJNOOuxSDKmpqbz00kulksZYp7yuGEq1TFG5IsWkvJb8vDf/PbZmbKVJtSZcVPMiDCa0mPwBrYKu5i6s+H9aBY0xsD8UCDoSSydM09ISIlJkxm/wLfeF+raXAiveIq51XKGmXt66dSuPP/44EydOZNOmTdSpU4fOnTtzyy23cMopp5RY2iZPnsyoUaOYMWMGzZo1o1atWnz55Ze43aW/ZlBJW7t2LU2bNmXBggV07ty5QOc8/PDDfPXVVyxcuPCIpk0ql9IuU0DlSlnJLVd++uknjj/++AKdo3JFKorsQDZP/fIUAHd1vwvXPhdW1VAZY2+3sXfZ4AgtMn8gk20gvvRaBUHBoIgURzCn4HKB5Tqys12ZQOgtPkGggM9Ba9eu5fjjjyclJYVnn32Wo48+Gr/fz5QpU7jhhhtYtmxZiaVv1apV1K9fn169eoW31ahRo8SuH6t8Ph9xcXFlnQwpLaVYpkDxy5Wnn36apk2b4vF4mDp1qsqVCkLlihxpIxeMZPO+zTRMbsildS6F3YRfOOWuK+hq5sKR8E/rnzEGs9/gahy5/UhTN1ERKTbLZWHFHeGfIjwYXn/99ViWxW+//cZ5551Hq1ataN++Pbfddhu//vpr+Lj169dz1llnUbVqVZKTk7ngggvYtm1beP8jjzxC7969+fDDD0lNTaVatWpceOGF7Nu3D4Dhw4dz4403sn79eizLIjU1FSBPd67t27czePBgEhISaNq0KR999FGeNO/du5err76a2rVrk5yczMknn8yiRYvC+x9++GE6d+6cb1oAbNvmmWeeoUWLFng8Hho3bszjjz8e3r9hwwYuuOACUlJSqFGjBmeddRZr164tcL7OmDEDy7KYNm0a3bt3JzExkV69erF8+XIARo0axSOPPMKiRYuwLAun08nHH39cqPt77733aNq0KfHx8bzzzjs0aNAA27Yj0nHWWWdx5ZVXAqGH5rPOOou6detStWpVjjnmGH744YcC35OUL6VSppRQudKiRYsilSuH+1tWuXLocsWyLEaNGlWo+1O5IqXBF/Tx5M9PAqFWQXe6+5+xgjuC2NtzWgXbHtQqmGWwEiyctUqvVRAUDIpIJbV7924mT57MDTfcQJUqVfLsT0lJAUIPOGeddRa7d+/mxx9/ZOrUqaxevZqhQ4dGHL927Vq+/vprvv32W7799lt+/PFHnnoq1AXk5Zdf5r///S8NGzZky5Yt/P7771HTNHz4cDZs2MD06dP54osveOONN9i+fXvEMUOGDGH79u189913zJs3j65du3LKKaewe/fu8DGrVq3iq6++ipoWgHvuuYennnqKBx54gCVLlvDxxx9Tt25dAPx+P/379ycpKYmZM2fyyy+/ULVqVQYMGIDP5ytUHt933308//zzzJ07F5fLFX6AGjp0KCNGjKB9+/Zs2bKFTZs2cc455xT4/lauXMn48eP58ssvWbhwIUOGDGHXrl1Mnz49fEzu7/eSSy4BICMjg4EDBzJt2jQWLFjAgAEDGDx4MOvXry/UPYkcSkmXK4f6W1a5cuhyZcuWLeH8VLki5cnohaPZkL6B+lXrM6zBMPATni00d11BV6orYkygMQaTaXDWceKIL93wTN1ERaRSWrlyJcYY2rRpc8jjpk2bxuLFi1mzZg2NGjUCYMyYMbRv357ff/+dY445Bgg93I0cOZJq1aoBcNlllzFt2jQef/xxqlWrRlJSEk6nk3r16kX9nL///pvvvvuO3377LXzN999/n7Zt24aP+fnnn/ntt9/Yvn07Ho8HgOeee46vvvqKL774gmuvvTacllGjRpGUlJQnLfv27ePll1/mtddeY9iwYQA0b96cE044AYBPP/0U27Z57733sKxQ5TRy5EhSUlKYMWMGp512WoHz+PHHH6dPnz4A3H333ZxxxhlkZ2eTkJBA1apVcblc1KtXD9u2SU9PL/D9+Xw+xowZQ+3atcOfdfrpp/Pxxx+Hx2N98cUX1KpVi759+wLQqVMnOnXqFD7+0UcfZcKECXzzzTf85z//KfA9iRzKkShX8vtbVrly6HKlsPenckVKgz/o54mfnwDgzmPvxL3XDYmhfcFdQYJbg2BFaRXMNDgSHLhqlX5oppZBEamUjCnYBBRLly6lUaNG4Qc2gHbt2pGSksLSpUvD2xo3bhx+SAKoX79+nrfvh/scl8tFt27dwtvatGkTbkkAWLRoERkZGdSsWZOqVauGf9asWcOqVavCx6WmpuablqVLl+L1evOdxGLRokWsXLmSpKSk8PVr1KhBdnZ2xGcURMeOHSPSABwyT/74448C3V+TJk0iHtgALrnkEsaPH4/X6wXgo48+4sILL8ThCFVjGRkZ3H777bRt25aUlBSqVq3K0qVL9QZfSlRJlyuH+lsu6OfEerlS0PtTuSKlYewfY1m7dy11q9TlisZXYLwmPFtouFWwiQtH1YNaBbMMjrqOPOsNlga1DIpIpdSyZUssyyqxyRxcrsji0rKsPGNNiisjI4P69eszY8aMPPsOfLg7eCbBA9OSkJBw2M/o1q1b1HFFBz8oHc6B6chtDThUnhT0/qJ1vxs8eDDGGCZOnMgxxxzDzJkzefHFF8P7b7/9dqZOncpzzz1HixYtSEhI4Pzzzy90FzWRQynpcuVQf8slReVKiMoVOdICdoDHZ4bG0Y44dgTxu+MxiQbLsgjuCRLcHATA3S5Kq2Bi2bQKgoJBEamkatSoQf/+/Xn99de56aab8jwI7N27l5SUFNq2bcuGDRvYsGFD+C3+kiVL2Lt3L+3atSux9LRp04ZAIMC8efPC3bmWL1/O3r17w8d07dqVrVu34nK5wpNFFFbLli1JSEhg2rRpXH311Xn2d+3alU8//ZQ6deqQnJxcpM8oiLi4OILBYMS2Ll26FPn+4uPjOffcc/noo49YuXIlrVu3pmvXruH9v/zyC8OHDw+PTczIyCjU5BUiBXFwuXJwkKRypfTLleLcn8oVKUnj/hzHqj2rqJVYi2uaXoO9wcZRM9QCmNsq6GzsxJF0QKugHWoVdDZ1YsWVfqsgqJuoiFRir7/+OsFgkGOPPZbx48ezYsUKli5dyiuvvELPnj0BOPXUUzn66KO55JJLmD9/Pr/99huXX345ffr0oXv37iWWltatWzNgwAD+9a9/MWfOHObNm8fVV18d8TB56qmn0rNnT84++2y+//571q5dy6xZs7jvvvuYO3dugT4nPj6eu+66izvvvJMxY8awatUqfv31V95//30g1C2qVq1anHXWWcycOZM1a9YwY8YMbrrpJjZu3Fhi95uamsqaNWtYuHAhO3fuxOv1Fvv+LrnkEiZOnMgHH3wQnuAhV8uWLcMTQyxatIiLL764xFtYRCBvubJq1SqVKypXJMb5g34e++kxAG477jYS9iRgxYdmvbXTbIIbQy8x4tpGLmliMg2OKg5cNcuufU7BoIgUmwkYjO8I/wQKvwh1s2bNmD9/Pn379mXEiBF06NCBfv36MW3aNN58800g1A3p66+/pnr16px44omceuqpNGvWjE8//bSks4mRI0fSoEED+vTpw7nnnsu1115LnTp1wvsty2LSpEmceOKJXHHFFbRq1YoLL7yQdevWhWftK4gHHniAESNG8OCDD9K2bVuGDh0aHnOTmJjITz/9ROPGjTn33HNp27YtV111FdnZ2SX6Rv+8885jwIAB9O3bl7p16zJ+/Phi39/JJ59MjRo1WL58ORdffHHEvhdeeIHq1avTq1cvBg8eTP/+/SPe8EvFUiplSgmUK3fccQe9evWif//+KldKuVypXbs2n3zyicoVKRde/PVFlu9aTs2Emvy72b8x+w1WYuS6gs6GThwpB7UKZhucdcuuVRDAMgUdDS3lWnp6OtWqVSMtLe2IdtEoK36/n0mTJjFw4MA84xrk8Iqbf9nZ2axZsya8PlMu4zf4lvtCizaXAiveIq51XHjh1tKSOxtmcnJyeGIBKbjKkH/5/Q1A5S9/i+NQeRMtT0u7TIHilSuV4btdlmI9/w5VrhSEno2Kp6Tyb82eNbR/oz1ZgSw+GPwBF3suDk0Ik+zA3meT9V0WGIg/LR5n9X/WELT32Vhui7i2R+a5pqB1k8YMikiRWe7QQxTBwx9bIpyUeiAoIqWn1MsUULkiIkVmjOHfE/9NViCLvql9uazxZQRWBrCqHTCDqAFnA2dEIGjsUO8Ed0N3mZc/CgZFpFgstwV6ISkiJURliohUFOP+HMeUVVPwOD28OfBN7J02OMByWtgZNoF1ASDKDKIZBkdVR3iCmbJU9ikQAB5++GEsy4r4OdyitiIiIkeK6iURkfztztrNLVNuAeC+3vfRwrTA3mtjVc1pFVya0ypYz4mzZmSrIH5w1XVhucq+V4JaBsuR9u3b88MPP4T/ffC6ZiIiIqVJ9ZKISHR3Tb2L7fu307ZWW+7ofAf+VX6IA8tlYWfaBNbm3ypoJVk4apSPNjmV6uWIy+WiXr16ZZ0MERERQPWSiEg0M9fN5L0F7wHw9sC3cWx2YPvscIDnX+oHGxx1HDhrH9AqGMxpFUwtH62CoGCwXFmxYgUNGjQgPj6enj178uSTT9K4ceOox3q9Xrxeb/jf6enpQGhmJL/fXyrpLU2591QZ7600FDf/AoEAxhgCgUBMrrGUO+myMSYm77+4KkP+BYPB8N/AwX9HlblcKky9BIWrm3LLlWAwWGG/F5Xhu12WYj3/Dqxbi1KO6NmoeIqaf96Al2v+dw0AV3a+km6mG95dXhzVHaG6IssQWB1qFXS0duAP/nN9k26wqlo4qjqw/Uf2O1/Q+9LSEuXEd999R0ZGBq1bt2bLli088sgjbNq0iT///JOkpKQ8xz/88MM88sgjebZ//PHHJCYmlkaSJcbUrVuXqlWrUqNGDXUVk5hijCE9PZ3du3ezbds2Dq42MzMzufjiiyvd0hKFrZegcHWTZVnUrVuXGjVqkJycjGWVj7fkIqUhEAiwe/duMjIy2LZtW1knRwrh062f8snWT0hxpfBam9eo6qoasb/Z5mY03NmQtMQ0FjVfBGVUtBW0blIwWE7t3buXJk2a8MILL3DVVVfl2R/t7WujRo3YuXNnpXoYyeX3+5k6dSr9+vXTWjpFUBL55/f72bZtG1lZWSWcuvLPGEN2djbx8fF6YC2CypB/lmVRv359qlSpkmdfeno6tWrVqnTB4MEOVy9B4eum/fv3s2XLljwBdkVRGb7bZSnW8y8hIYG6desWq17Ws1HRFSX/lu9aTrf3uuEL+hgzcAznOs/FBE140hjjNfgn+SEIruNdOOofsMh8Wug4dws3lvPIf98LWjfp9X45lZKSQqtWrVi5cmXU/R6PB4/Hk2e72+2u1AVCZb+/I604+ed2u0lNTSUQCBAMluYiYGXP7/fz008/ceKJJ+r7VwSVIf/cbjdOpzPffbHgcPUSFL5uSklJISkpqcJ2c6sM3+2yFMv553Q6cblcJRIE69moeAqaf8YYbpxyI76gjwHNB3BB0gWYPQarhhX+PfpW+iAIjuoO4o6KC283AYPB4K7vxhkfvS4paQX9TigYLKcyMjJYtWoVl112WVknRSTMsqyYrHScTieBQID4+PiYu/eSoPyrHI5UveR0OvMNtMs7fbeLR/knFcmohaOYsXYGCa4EXu72MvZuG0d1xz8Bn9fgXxF6seVu744I9O19Ns4UJ47q5WMG0QOVvxTFqNtvv50ff/yRtWvXMmvWLM455xycTicXXXRRWSdNRERikOolEZGQHft3cPvU2wF46LiHaLyvMY6qjojunv4VfgiAo5oDZ4MDZhANGCwsnHWdWI7y1x1aLYPlxMaNG7nooovYtWsXtWvX5oQTTuDXX3+ldu3aZZ00ERGJQaqXRERCbvv+NnZn7aZTnU7cUOsG8IMVf0DLX6aN/+/DtApWK59tcAoGy4lx48aVdRJERETCVC+JiMDUVVMZ+8dYLCxe7/I6rkwXVs1/gj1jG7yzveAPjRV0NjygVdBfvlsFQd1ERURERERE8sjyZ/Hvif8G4Pr219PNdMNKsSJa/vx/+rF32uACT09PnlZBR3VHuW0VBAWDIiIiIiIieTz606Os2rOKo6oexYNHPQgJYLn+CfYCWwL4l4a6h3qO8eBIOmApCb/BsixcdUtm1tgjRcGgiIiIiIjIARZvW8yzs54F4MUOL5LkSMKR+E/oZGfaeOeE1lV1NXfhahw5+s7sMzhqOLCSy28gCAoGRUREREREwmxj869v/0XADnBmozMZlDQooqunsQ3eX73gBUeKg7gucRHnG58BB+W+VRAUDIqIiIiIiIS9PfdtZm+cTZI7ieebPY+jmiNiAhj/X37sHTnjBHt5IpaYADAZBkdNB1ZS+Q4EQcGgiIgIwR1BTLYp62SIiEgZ27xvM3dPuxuAh1s/TMNqDbHc/wR1wa1B/EuijxOE0OLzOMBVp/y3CoKCQRERiXH2fpvA1gB2pl3WSRERkTJ28+SbSfemc0yNY/hXg3/hqHLAOMEsm+xfs4Ho4wQh1CrorOXEqlr+A0FQMCgiIjEuuDOI2a9WQRGRWPft39/yxZIvcFpOXm31Kq7q/wR7EeMEqzmI6xyX53zjNeACZx1nhWgVBAWDIiISw+xMm+DOIKhRUEQkpmX4Mrhh0g0A3JR6E50bdI4cJ7jEj739gHGCrrzBXm6roKNqxQmxKk5KRURESlhwZxC8QMV4gSsiIkfIg9MfZH3aepokNuG+1vdheQ4aJ/hXzjjB7h4cyXlDKDvThrhQq2BFomBQRERikp2V0ypYpaxTIiIiZWne5nm8POdlAF5p8wpVU6qG99lZB6wn2MyFq0mUcYJ+A1ngbOCMGGNYEVSs1IqIiJSQ4K7QDKJWvJoFRURiVcAOcO2312IbmyF1h9C/Rf/weL/ccYIm22BVs/KsJ5h7jJ1m46jjwFU3b6BY3ikYFBGRmGNn2wS3B7ESrQozyF9EREre63NfZ/6W+aS4Uniu63MRawYeOE4wvld81HGC9l4bRzUH7kbuiDGGFYWCQRERiTnhVsGEildxi4hIydjh28HDPz4MwONtHqdeSr3wvuC2A8YJdstnnOB+G8tthQLBuIpZnygYFBGRmGK8Bnu7rVZBEZEYZozh7Y1vs9+/n14pvbii/RXhfXaWHVpGAnA1deFKjTJO0GfAC+6G7qiBYkVRcVMuIiJSBIFdAewsW62CIiIx7MtlXzI3fS5uy80bx72B0xGaBdTYBu+cnHGCyRZxXfMZJ5hu46zrxFGnYodTFTv1IiIihWB8BntbKBBUq6CISGzambmT276/DYDbWtxG2xptw/v8S/3Y22xwQvzxeccJGmMwew2O6g5cDV0Vvi5RMCgiIjEj3CqYWLErbxERKRrb2Fz+5eVs2b+Fhp6G3NHxjvC+4PZ/xgnGdYuL2v3TZBjwEBon6K74dYmCQRERiQnhVsF4tQqKiMSqZ395lu9WfUe8I57bU28n3hkPgMk2eGd7wYAr1YW7qTvPucZrIJAzTrBq5QijKsddiIiIHEZwdxA7U62CIiKx6pf1v3Df/90HwLMdnyU1IRUIdf3M/jX7n3GC3aKMEwwa7H02zvpOHLUqTwhVee5EREQkH8ZvCG4LhloFK+A6UCIiUjy7Mndx4RcXEjRBhjYYyvCWw8P7IsYJRllPMHecoLOGE1eDij9O8EAKBkVEpNKzd9uh9aDUKigiEnNsYzPsq2Fs3LeRloktea3na+GAzt5h4//zgHGC1aKME9wXWpfW1dgVdeH5ikzBoIiIVGomYAhsC2B51CooIhKLXpj9AhNXTMTj8PDR8R+RHJcMgDvgJjAncOhxgtkGbHA1cuFIrHyhU+W7IxERkQOEWwWrRA8Ed2TvYKdvZymnSkRESsPsDbO5+4e7AXju6OfoVKsTEOr62Xp9a8gGKymf9QQDBjvDxtnAiaNG5QybKuddiYiIcECroDv/VsF7F97L0T8fzfiV40s5dSIiciTtztrN0C+GEjRBhtQbwjVtrwnvs5fb1Mio8c84QXeUcYJpBmdtJ676lWuc4IFcZZ0AERGRI8XeY2MyDFb16JX43F1zGbNmDAANqzYszaSJiMgRZIxh+FfD2ZC+geYJzXm91+vhgC64I0jwryAAzs5OHClRxgmmGawqFq5GLixn5QwEQS2DIiJSSZlgqFUQN1FbBY0x3D7/dgAurn8xx9U7rrSTKCIiR8iLv77I//7+H3FWHB/1/IhqnmpAaM1Z76+h9QS3p2zHkZo3HLKzbLByxgnGV+5wqXLfnYiIxCx7jx2aAS6fsYKfrfuM2TtnU8VVhf+2/G8pp05ERI6UORvncNcPdwHwbPtn6VK3CxB6Cej93YvJNFAFVhy1Ik/3TxMwkAnOo5w4qztLPe2lTcGgiIhUOuFWQRdRu/fsD+zn3kX3AnBnuzs5Kv6o0k6iiIgcAXuy9jD0i6EE7ADn1j2Xa9tfG94XWBUguDEIDnAd5yLoDEaca+zQeoKO2g5c9WJjNJ2CQRERqXTsvTmtglWjtwq+sPQFNmZupEmVJtzc+uZSTp2IiBwJxhiu+PoK1qWto1lCM97s+SYORyjcsffa+Bb4AIg7Oi7q7KB2mo2VZOFu5I6ZpYgUDIqISKVibENgewAc0VsF1+9fz3NLnwPgqc5PkeBKKO0kiojIEfDynJf5evnXxDniGHvMWFISUoBQ18/s2dlgg7O+E1frvK1+dqaN5bJwN3ZjeWIjEAQFgyIiUsnYe+3QLHBJ0Svz+xbeR3Ywm961e3NOo3NKOXUiInIk/LbpN+6ceicAT7d+mm5HdQvv8y3wYdINVryF51hP3nGCfgNZ4DrKhaNabIVHsXW3IiJSqRnbENwezLdV8Jcdv/DZ+s+wsHi+2/OVdt0oEZFYsjd7L0O/GIrf9nNOnXO47ujrwvsCGwIEVgcA8PTwYMUfVO7boe6hzrpOnHUq/4QxB1MwKCIilYadZofGfEQZK2gbmxHzRgBwRfMr6FS9U2knT0RESpgxhiu/vpK1e9fSNKEpbx33Fg5nzjjBDBvv714A3G3dOOvmDfZMusFRzYGroStmxgkeSMGgiIhUCsaEWgUNBsuVt0Ifs3oMC/YsINmdzCMdHymDFIqISEl79bdXmbBsAm7LzYddPiSlagoQ6inine0FPzhqOnB3cEe/gJvQOMG42AsEQcGgiIhUEnaajb3XxpGUt2pL96fz4B8PAnBv+3upE1+ntJMnIiIlbO7mudz+/e0APNnqSbo37h7e5//Tj73bBjd4enrytvr5Q/9xNXBFrTdiRezeuYiIVBrhVkETvVXwqb+eYlv2NlokteCGVjdE7LOzbIzXlFZSRUSkBOzN3ssFn1+A3/ZzZp0zuaHDDeFx4MGtQfxLQ9Ge5xgPjiqRIY+xDXa6DYCjVmyHQ7F99yIiUimYdIO9J3qr4Mp9K3l1+asAPNPlGeKccRH7fQt9ZM/Kxv+3v1TSKiIixWOM4epvrmbN3jU0SWjC293exhEXKv9NtsE7JzRO0NXMhatR3mUkzF6DIyV0fKxPJKZgUEREKjRjctYVBCx33kr97gV347N99KvXj4ENBkbsC+4OElwfhCA4qqtKFBGpCN74/Q3GLx2P2+Hmw44fUiOlBhCqD7xzvJhsg5VsEdclLs+5doYNHnA1zBskxiLVfCIiUqGZfaFWwWgziE7bOo3/bfofTsvJs12fjXgDbIzBt9AHgLOeE2ft2JtSXESkopm/ZT63fX8bAI+3eJxjmxwb3hdYHiC4NQhOiO8Zn2fYgPEZ8IO7oTtP19FYpVwQEZEKK9wqaOdtFQzYAW6fH5pY4LqW19G2WtuI/cHNQewdNjjA1UJviEVEyru07DQu+PwCfEEfg+oM4sZ2N4bXlA3uCuL7I/SCL65zXLgbaC4TDI0TdNZzxvw4wQMpJ0REpMIy+wz27uitgu+teo8laUuoEVeD+zvcH3mebfAtCj00uFu7ccSrOhQRKc+MMVzzv2tYtWcVjRMb806nd3Ak5IwT9OcsI2HA2dCJq7krz7kmzeCs7sTVwBXz4wQPpNpPREQqJGMMgR0BCJJnfajd3t3894//AvDg0Q9Sw1MjYn9gdQCzz4AH3G3yWXtKRETKjbfmvsXnSz7HZbkY02EMNWoeME5wrhez32AlWniO8eQJ9kxGqLx3NXZFHVseyxQMiohIhWT2598q+Nifj7HLt4t21dpxTYtrIs/zG3x/5nQlahcXswsNi4hUFL+s/4VbptwCwGOtHqNH4x7hgC+wNhCaCMzKWU/woDLdeA0ENE4wP8oRERGpkII7QrOAWp7Iin9p2lLeWvEWAM91fQ6XI7K7kH+ZH7xgVbXydCUSEZHyZfWe1Zz96dn4gj7OrHcmN7W8Kdy6Z6fb+ObldPnv4MZZK3IiMBM02Ps0TvBQlCsiIlLh2Bk2wV1BrCoHvQE2hjvm30HQBBl81GBOqXdK5HlZNv7lofUE4zrGhSceEBGR8mdv9l7O+PgMdmbupGtKV95v/z7OpFDAZ4I54wSD4KjjyNPlPzxOsIYT11EaJ5gfBYMiIlKhGGNCrYL+vK2C323+jqlbp+J2uHmqy1N5zvUv9oceHGo6cDbUUhIiIuWVP+jn/M/OZ9nOZTRMaMjnHT8nqVZSeL9vkQ97b2jNQE8PD5Yjn3GCjVx5lpiQfygYFBGRCsXeaRPcHswzVtAX9HHHgjsAuKn1TbRIahF53l6bwNrQ4vRxneP0llhEpJwyxnD9xOuZtmYaVV1V+aLTFxxV76hwwBfYFCCwIlSee471hGcVDZ+fO06wkcYJHo5yR0REKgx7n41/gx/ceVsF31jxBiv3raRufF3ubn93nnN9f/jC044fPK5ERETKj+dnP897C97DYTkYffRoOjfsHO7Wb2faeH/zAuBq5cLV4KBlJHLHCdZ34qipUOdwlEMiIlIhGK/Bv86P8RkcVSOrr+3Z23n8z8cB+G/H/5LsTo7YH9waJLglNNtcXMe4UkuziIgUzoSlE7hz6p0APN36ac5IPSM8YYyxDd5fveADR3VHnvLcGIPZmzNOUOsJFoiCQRERKfdM0ODf4MekGxwpeauuh/94mHR/Ol2qd+HyZpdHnmv+WWDe1cKFI0lVn4hIeTRv8zwu+fISDIZ/NfkXN7S8ASv+n4DOv9SPvcMGV84yEs684wSteEvjBAtBNaKIiJRrxhgCWwIEdwSxUqw8kwQs2rOID1Z9AMDz3Z7HYUVWbYG1gdAkA26Ia69WQRGR8mhD2gYGfzKYrEAWp9U5jefaPoez6j9d+oM7gvj/Cs0G7enmyfNiL3ecoKuRS+MEC0E5JSIi5Zq9yya4KYijiiPPm15jDLfPvx2DYUjjIRxf+/jI/QGD/8/Qw4O7rTvPOEMRESl7+7z7GPzJYLZkbKF9cns+7Pgh7pR/loqws+zQMhIGXKkuXKkaJ1hSlFsiIlJu2Rk5E8a4iOgqlGvChgn8tP0n4p3xPNH5iTz7/Sv8mEyDlWjhbunOs19ERMpW0A5y0fiLWLRtEXU8dRjfcTzValYLj/czAYN3pheTZbCSLOK6apxgSVIwKCIi5ZLxhSaMwUvUcX7ZwWzuXhiaNXRE2xE0rtI48vxsg39pTqvg0W6NHxERKYdGfD+CiSsmEu+M57NOn5HaIDU8HCB3whh7jw1xEN87PjyZTK7wOMHGGidYFAoGRUSk3DF2aMIYO83GSoleub+87GXW7V/HUQlHMaLtiDz7fUt84AdHigNXE1eUK4iISFl6/bfXeXnOywC81/49ejTuETEpjG+Rj+CmIDgg/oT4/McJNnbhSFRYUxTKNRERKVdyJ4yxt9s4qjnyTBgDsDlzM08veRqAJzo/QRVXlYj99j6bwEotMC8iUl5NXjmZmybfBMB/W/6X85ufH9Hq51/hJ/B3zsLyx3lw1o5cHzZinGANhTRFpZwTEZFyxd4dmjCGRPJ0B8r1wKIH2B/YT49aPRjaZGie/eEF5us7cdbVAvMiIuXJ4m2LueDzC7CNzWUNL+P2NrdHjAsPbA7gWxBaEsh9tBtX44MmjNE4wRKjYFBERMoNe79NYH0AHOBIiF5F/bbzN8auHQvA812fz/MQENwZJLhRC8yLiJRHWzO2MuiTQezz7aN3zd681uG1yCUkdgf/mTm0qQt327yTf5l9BitB4wRLgoJBEREpF4zP4F/vx3hDM8ZF4w16uXHujQBc1vQyutfsHnkNY/AtzFlgvqkr6gL1IiJSNjL9mZw17izWp62nRdUWjOs4Dk+KJ7zfzrTxzvRCABx1HcR1z9vN33gNBHPWE9Q4wWJTDoqISJkztiGwMYC9JzRhTH5dfu5fdD8L9yykZlxNHu30aJ79wY1B7F02OMHdQUtJiIiUF7axGfbVMH7b9Bs14mowodMEatap+c8SEn5D9k/ZmGyDlWwR3ys+z5jx8DjBBhonWFKUi+XQU089hWVZ3HLLLWWdFBGRUhHYGiC4LZjvhDEA323+jleWvwLAOz3eoX5C/Yj9JmhCYwUBdxt3vt1MpWhUN4lIcdz/f/fzxZIvcDvcjOs4jpZHtYxcQmKWF5MWWiYi/sR4rLiDAsHccYI1nbjqa5xgSVFNWc78/vvvvP3223Ts2LGskyIiUiqCu3PG+B1iwpgtWVu4+terAbi+1fUMOmpQnmMCqwLh9abcrdUqWJJUN4lIcYxcMJInf34SgDfbvcmJqSeGl5AwxuCb7yO4NQhO8PT24KiSN0QxaTnjBBtpnGBJUjBYjmRkZHDJJZfw7rvvUr169bJOjojIEVeQCWNsY3PF7CvY6d1Jx5SOPNn5yTzHGJ/B91dOq2B7d75BpRSe6iYRKY7pa6Zz7bfXAnB387u5tOWlkUtILPcTWJWzhERPD84aeWeAtjNC3f9dqRonWNK0Cm85csMNN3DGGWdw6qmn8thjjx3yWK/Xi9frDf87PT0dAL/fj9/vP6LpLAu591QZ7600KP+KR/lXPPnln/Eb/Gv8mP0Gq7pFMBiMev5zS59j+rbpJDoTGdVjFE6c+IOR1wosCYAPSALTxOTZfzjGDi1cbPvtQp0Hlf97obopfyobikf5VzwVIf+W71rOeZ+dR8AOcF7987in9T0E3AHIKe7tjTaBRaFA0NnJiamXt/w22Qa8oUDQrmIXqZyOpiLkX3EU9L4UDJYT48aNY/78+fz+++8FOv7JJ5/kkUceybP9+++/JzExsaSTV25MnTq1rJNQoSn/ikf5VzyHzL+d0Tcv27+M/674LwBXNbiK9VvXs37r+ohjPD4Px/x9DA4c/FnzT3av3F20BOaThsPJzMws2okVgOqmglHZUDzKv+Ipr/mXHkjnzr/vZI9vD60TW3NB7QuYvnl6eH/S/iQ6ru6IEyebam5ilb0K/j7EBbcfmXSW1/wrroLWTZYxxhzhtMhhbNiwge7duzN16tTweIyTTjqJzp0789JLL0U9J9rb10aNGrFz506Sk5NLI9mlyu/3M3XqVPr164fbrbFAhaX8Kx7lX/FEyz//Nj/2Ohsr2cr3teRe3156fd+L9ZnrGdJ4CB8c90HUCQMCvwWw19tYtSxcfYo2qYDZbXC1dOGsXvgF6tPT06lVqxZpaWmVqvxV3XR4KhuKR/lXPOU5/7ID2Qz8ZCA/b/iZJolN+L/u/0fdunUhp3g2GQb/dD94wapv4eoVpewOgJ0WmjnU1bDkJ4wpz/lXEgpaN6llsByYN28e27dvp2vXruFtwWCQn376iddeew2v14vTGfmA4vF48Hg8B18Kt9tdKb/QuSr7/R1pyr/iUf4VT27+BfcEcWxx4KjiwOGJPvbDGMPN825mfeZ6Uquk8saxbxDnyruAfHBPEHt9qMuQp4sHp6vwwRyA7bBxu9w43YU/v7J+J1Q3FVxlv78jTflXPOUt//xBP5d8dQk/b/iZZHcyX3b6kqPqH/XPzKE+Q9YvWeAFR4qD+J7xeSaEyV1Cwl3PjbuJOzzZzJFQ3vKvpBT0nhQMlgOnnHIKixcvjth2xRVX0KZNG+666648la2ISEVlZ+ZMGAOHnARg5OqRjN8wHpflYuzxY0l2532raYzBtyg0aYyzsTPqpANSdKqbRKSwgnaQYV8N439//494Zzyfd/yc9g3b/xMIBg3Zv2Rj9oVmBvWc6Mkz4Vd4CYkaziMeCIqCwXIhKSmJDh06RGyrUqUKNWvWzLNdRKSiMgFDYH0Ak2mwauRfuS9NW8pt824D4JGOj3BMzWOiHhfcGsTeZoMD4jrmbTWU4lHdJCKFYYzh3xP/zSd/foLLcvHx0R/Tp1mfcKufMQbf7z7s7Ta4IP7E+DyzSBtjsPfYOKo6cDVxaWboUqBgUERESkVgUwBrt4WjuiPfsR9ZgSwu/eVSsoJZnFrvVG5re1vU44z9T6ugu6U76ppUIiJSOowx3Dn1Tt6d/y4Oy8HIDiMZ2GpgRKue/y8/gXUBsMDTy4MjJcpagukGh8eBu6k73+WGpGQpGCynZsyYUdZJEBEpUfZ2G3fyobv83LXwLv5M+5M6njq83+N9HFb0h4HA2gAmzUAcuNsVb6yHnWlDHFgevYE+HNVNIhLN4zMf57nZzwHwetvXGdJ2SGQguNaP/6/QUgdx3eJw1c8bgtj7bbBy1hKsqkCwtCinRUTkiLKzctaE8oAVl3/A9fWGr3l7xdsAfNDzA+ol1It6nAkY/ItzHiraxR3ymodjggaTaXDWd6p1UUSkCF6Z8woPTH8AgKdbP80VR18REQgGtwfx/Z7Tk6ONG3fzvC/wTLYBH7gbu4s0q7MUnWo+ERE5YowxBLeFVhe2EvMP2tbvX8+/fvsXALe1uY1+9fvle6x/mR+TbbCqWLhaFK+Di51u40hx4KqjjjIiIoU1auEobp58MwD3N7+fmzvfHJ4sBkJlbPYv2WCDs6ETd8cogaDPYPYbnA2dOGorNCltynERETli7DQbe6d9yGMCdoArZl/BHt8eutfoziMd8y5aniu4PYh/SU6rYMe4Ys0yZ7wGy7JwH+XOM625iIgc2vgl47nqm6sAuLHJjdzX/b6IQNBkG7J/ygYfOGo68BznyTNe3AQMdrqNs74TV/2SX0tQDk/BoIiIHBEmaAhuDWIwhzzuyb+e5OcdP5PkSmJMrzHEOaPPDGrvt8melQ0GnE2cOBsVvSuRsUNrWDnrOkML34uISIFNXjmZi8ZfhG1shjUcxjM9nsHh+CesMH5D9s/ZmP2hXhzxJ0RZS9A22HttnLVzFpV3qCwuCwoGRUTkiLB32aEpwpPyr2p+2v4TT/z1BACvHfMazZOaRz3OBAzeX7zhRYo93fO+YS4Mk2FCU5frTbSISKHMXDeTcz89F7/t57x65/FGrzciA0GvIXt6NvYuG9yhJSSs+ChrCe4xOKvnrCWo3hllRoMkRESkxBmfIbAlAHFAPg14u7y7GD5rOLaxubzp5VyYemH0axmDd64Xe48NHvCc4CnWg4PxGwiAq6mrWJPPiIjEmvlb5jPo40FkBbIYUHsAo04chcv5TzhhZ9pk/5iNSTfgyVlLMDnvC0F7r42jigNXqsrhsqZgUERESlxgewB7v42jpgOiDBk0xnDtnGvZlLWJVkmteLHbi/lf6+8AwXVBsCC+Z3yxZv00xoTWsarjwFFDnWNERApq6Y6l9B/Tn3RfOr1r9GbcSeMiuvXbGTbZM3K6hiZYxJ+UTyCYbmPFWbhTtZZgeaBgUERESpS93ya4LYhVxcq3C+abK97k203fEueI48NeH1LVXTXqccFtwfDi8nGd43DWLd6U4ybTQDy4Gqh7qIhIQa3evZpTR53KzuyddEvpxpcnf0mCKyG8396b0yKYbbCq5gSCUV7c2fttMOBu4o4aKErp029BRERKjDGGwNZAaPa4fN74/rHnD+5ecDcAT3R+gs41Okc9zt5vkz07NGGMK9WFq2Xx3l+aoIHsUCCot9EiIgWzKX0T/Ub1Y3PmZtolteN/J/+PZHdyeH9wV5Cs6VmhQLCaRfzJ0QPB3LUEXY1dOGtoLcHyQi2DIiJSYuy9oaUk8puhc39gP5fOuhSv7WVgg4H8p9V/oh5nAgbvzzkTxlR3ENctrtgteSbN4KjuwFlLDyEiIgWxY/8O+o3sx+p9q2lWpRmTTplETU/N8P7gtiDZP2dDILR8RPyJ8VHHABp/zlqCjZw466gMLk8UDIqISIkwwZxJYyyw3NEDtxHzRrA8fTkNEhrwbo93owZ4xhi8v3ux9+ZMGHN88SaMAbCzbHCB6yhXsdYmFBGJFWlZaQwYNYCle5dyVPxRfHfyd9RPqB/eH9gUwDvLCzY46jqIPz4+atlvggY7zcZV36Uu+uWQgkERESkRwZ1B7DQbR0r0LphfrP+CkatHYmExsudIanlqRT0u8HeA4PqcCWN6FW/CGAitZWX2G1yNXYdc5kJEREL2e/czePRg5u+cT624Wkw6ZRKpVVPD+/1r/fh+84XWfT3KiaenJ+qLNmPnLCFR04mrkdYSLI8UDIqISLEZryG4JYjlsaI+EGzzbuOOv+4A4O72d3NS3ZOiXifPhDEl0J3IpBsc1Ry46qrKExE5HK/Py/kfns/MbTNJdiXzbd9vaZPcJrzfv8KPb36onHaluog7Ji5qkGeMwewNjSN0p2otwfJKNaOIiBRbYFsAOzNnKYmD+G0/z697nnR/Oj1r9eT+DvdHvYadYZM9q+QmjIFQkIrJmT00n66rIiISEvAFuOSjS5i8aTKJzkS+PulrutToAoSCO/9SP/7FfgBcLV3EdYk+ntvYOYFgYk4g6FH5W14pGBQRkWKx99kEtwdxVHVEfSh4ePHD/J35NynuFEb3Go3LkbfqMQGD9xdvaBbS6g7iupfAhDHGYO8LjVPJr+uqiIiE2D6baz+7lvHrxxPniOOz3p/Rq3YvIFSe+hb5CCwPAOBu78bd3p1/ILjHYCVZuJu6i93VX44sBYMiIlJk4aUkAkSdQXTUqlG8vPxlAF475jWaVGkS9RoRE8acEH3sSaHTlmFwVHFowgIRkcOw/Ta3TriVkatG4sDBh70+pF/9fkAouPPN9RFYEwoE4zrH4W7tjnodE8xpEUzRovIVhYJBEREpMnu3jb3LxkrKG2xN2zqNG36/AYAhdYdwdsOzo14jsPyACWOOj8eRWPyHB+M34A+tZ6XuSSIi+bP9Nrd+eSuvLHsFgHeOe4ezG50NhII77xwvwQ2hMjquexzuZvkEggGDvdfGWdOprqEViIJBEREpEhPIWUrCmXcpiSVpS7jw5wsJmAAXNL6Ai6pfFPUawa1BfH/kTBjTJQ5n7RKYMMaY0KQxtRxRxzCKiEiIHbC5cfyNvLH8DQBe7f4qlzW7DPin+35waxAc4OnhwdUoeuhg/KHlI5y1cwJBjdGuMFRLiohIkQR2BLDT87YKbs3aytk/nk26P53jax/Pm8e8GbWbpp1hkz07Z8KYpi5cLUrm/aTJNODJWVNQ05iLiEQVDAS57vPreGP5G1hYvHXsW1zb8loAjM+Q/WN2KBB0hrrv5xsI+nLWEaznwt1UgWBFo5ZBEREpNDvLJrg1iJVgRQRcmYFMzvvpPNbtX0eLpBZ83vtzPE5PnvNNwJD9c3ZowpgaDuK6FX/CGMgZr5JtcKW6SqS7qYhIZRQMBrnms2sYuSK09uu7x737T4tgdigQtPfa4Ib43vH59tow2TnruB7lwtXQVSLjvaV0KRgUEZFCMcYQ3BaELLBq/lPxB+0gw2cPZ+7uudSMq8nXfb6mpqcm/qA/z/ne37yYNIMVb+E5vmQmjAFC3ZSqO3HVUfUmIhJNIBDginFXMHbVWBw4GNlzJBemXgiAnWmTPSMbsy/UwyK+TzzO6tEDQTvLhixwNnKGJupST4wKSbWliIgUitlnCG4PYiVZEa159yy8h683fk2cI47PT/ycFkktop7vX+YPT0bg6eUpsRY8k22wnFbooURvp0VE8vAH/Vz2yWV8uvpTnJaTD3t9yHmNzwNCywRlz8jGZIbWB4w/KR5HUvTy2d5vgw+cjZ246mvG5opMwaCIiBSYsXMmjTFEzBT31oq3wktIvN/jfY6vfXzU8wNbAuEFi+O6lsyEMbnpsjNsXI1cOJLVPVRE5GC+gI+LP76Y8WvG47JcfHT8R+FZQ+10m+zp2aGXakkW8X3i810f0N5nQ5DQ0hF1oq8vKxWHgkERESkwe7eNvceOWFPwu83fceu8WwH4b8f/ckGTC6KeazIMvtm+fyaMaV5yVZDZZ3AkOXDVU7UmInIwb8DL0LFD+XpdqPfGJyd8wqCjBgE5LYI5gaAjxUF8n3is+OgBnp1mgwPczdw4a5XMyzwpW6o1RUSkQIw/p1XQBZYr9KCwaM8iLv3lUmxjM7zZcO5sd2fUcx1BB4FZAfCDo2bJTRgDoZnssHNmD9UsdiIiEbL92Zw39jwmrZ+Ex+Hhs96fMaDBACBnVufcFsFqoa6h0dYHNCa0hqAVF1pMPr9xhFLxKBgUEZECCWwPYO+zcdQIdR3amLmRc348h4xABn3r9uW1Y16LGuAZY2i9sTUmveQnjMldU9BZz4mjurqHiogcKNOXybljz2XKhinEO+L54sQv6Fe/HxAa95c9PRuTZbCSLRJOSsg3EDR7DY54B+6mbnXFr2QUDIqIyGHZmTbBbUGsxNBSEvv8+zjnx3PYlLWJtsltGXfCONwOd/Rz/7apnVY7NGHM8R4cCSX3IGH2hyY6cDZwatyKiMgBMv2ZnDnmTKZtmkaiM5EvT/ySvvX6AgfMGpppsKrmtAhG6RpqbIPZEzrG3dSNo6oCwcpGv1ERETkkY0xo4WEvWAkWATvApbMu5Y+9f1A3vi5f9fmKlLiUqOfa6TbBP4MAOLs4S3SMiQkY8IKrgQtHvKozEZFcGb4MBo4ayLRN06jirMI3J33zTyCYlRMIZhisKhbxfeOjvqQzQYPZHWo1dDdXIFhZ6bcqIiKHZKfZBHeGlpIAuG3+bUzePJkEZwLjTxxPatXUqOcZY/DO9YKBXUm7cDQt2SrHpBkcNR04aqkqExHJtc+7jwGjBvDj5h9JciUxse9EetfpDeQsKJ+zjqCVmBMIRlnexwQM9p7QsIC45nEltgSQlD/6zYqISL5MMNQqaIzBirN4ZfkrvL3ibSwsRvcczTE1j8n33MDaAPYOG5yw6qhVJdqN0860IS5n0hgtdCwiAkBadhr9Rvbjly2/UM1VjUl9J9Gzdk8AjDcnEEw3WAk5gWCU5SOMPzRZjLO2E3czd74zi0rloDGDIiKSL3tXaCkJR4qDrzd8zV0L7gLgqS5PcVajs/I9z3gNvkU+AJxtnWQ7skssTSZgMJkGV6or33WwRERizZ6sPZw26jTmbp9LdXd1Jp08ia41ugKhWZezf8zGTrOx4nMWlI/S7dP4DHa6jbOuE3cTd3jmaKm8VIuKiEhUxpezlEQczNs7j2Gzh2EwXNviWm5uffMhz/Ut8oXGGCZbOFqV4IQxtsGkGZy1nbjq6n2miAjArsxdnDLyFOZun0tNd02mnDLln0DQnxMI7rHBQygQjDIjqPEazD6Dq74Ld6oCwVihYFBERKIKbA9g77dZxzrO/elcsoJZDGgwgBe7vXjILp/BHUECawIAeLp7SrQbp73XxkqycDd2l9jyFCIiFdmO/Ts4eeTJLNixgNpxtfn+1O/pVL0TkBMI/pSNvTvUtT7hpAQc1aK3CJoMg/MoJ64mLpWvMUSvVUVEJA97f2gpiTR3Guf8dA7bsrfRMaUjY3uNxeXIv+owwZxJYwBXMxfO2k7soF0yadqXs+BxE3fUtbBERGLNXv9e+n3YjyW7l1A3ri5TTp1C22ptgVCX+uyfs7F32uCG+D7xOFLyCQTTcwLBhhqHHWsUDIqISAQTDHUP9Wf7uXjxxSxNX0qDhAZM6DOBJHfSIc/1L/dj0g14IK5jXMmlKdtAgND05knq1CIisiVjC/evvJ+N3o3U99RnyqlTaJ3cGgiV49k/Z2Nvt8EVCgSdNfIu7WN8BnufjesoF65GCgRjkWpUEREJM0FDYH2AwLYAN668kenbplPFVYWv+nxFw8SGhzzXzrDxL/EDENc5rsRa70zAYO+3cR7lxFFT1ZaIyLq96zhlzCls9G6kYXxDfjj1h4hA0PuLF3tbTiB4YjzOmlECQX9oshhXPZdaBGOYWgZFRAQITc4S2BAgsDXA81ueZ/Ta0TgsBx8d/1F4/Em+5xqDb54PguCo48DVpGSqF2PnTHFe14mrvqtEl6cQEamI/tr+F/0/7M+mjE3Udtdm8smTaZHcAgiVmd7ZXoJbguCE+N7xOGvnEwim2bjqu3A11hjBWKZgUEREwoFgcHOQL9K+4MG/HgTgxW4vcnqD0w97fnBjkODWIDjA081TIkGbMQaz1+Co5sDdSBPGiIj8uvFXBo4dyB7vHtpUacPtTW4ntWoqkBMI/uoluClUFsefEI+zziECwXoKBEXdREVEYp6xDYGNoUBwYsZErvz9SgBubn0z17W87vDn+wy++aE1Bd1t3VGnLC9SuvYZLE/OhDFxelgRkdg2eeVkThl9Cnu8ezg25VimnDqFWnG1gJxA8DcvwQ05L+VO8OCsFyUQDIQCQWddzRoqIQoGRURiWDgQ3BTk+4zvuWjORQRMgAubXMiTnZ8s0DV8i32YbINV1cLd1l0i6bKzbLDB1dgVdWFkEZFY8vHijxn8yWAyA5n0q92Pyf0mU9NTE8jppv+7j+C6IFjg6eXBVT9v5z8T+KfbvbuJeltIiGpYEZEYZYwhsCkUCP5f5v9xwa8X4Lf9nNfoPN7v8T5OR963ygcL7goSWJmzpmA3T4k8XBi/gUxwNnRGnfRARCSWvPbba1z65aUE7ABDjhrCl32/pIqrSmingeCCIIG1gVAg2NOD66hDBIJ1cgJBLSgvORQMiojEoHAguDHIT1k/cf7s8/HaXs5seCaje40+5FqC4WvYBt/cUPdQZxNn1C5JhU6XfUAXpnoa1i4iscsYw0PTH+LG727EYPhX038xpvcY4pxx4f3NNzfHXm2HAsHjPLgaHSIQrO3EnapAUCKpphURiTHGGAKbQ4Hg7OzZnDv7XLKCWZze4HTG9hqL21Gwrp6BFQHsvaHFjD2dPSWSLrPH4Kzu1HpXIhLTbGPzn0n/4c25bwJwf5v7ub/z/eHJuYwxBP8IctSuowCIOyYu6izOJqhAUA5NwaCISAwJB4Ibgvye/TtnzTqL/YH9nFrvVMadMA6Ps2BBnZ1p4/sz1CoY1ykOK74EuoemG6wEKzSpgVsPLCISm3xBH5d/eTmfLvkUC4sXO73Iv9v9O7zf2KFJu+xVNgDOrk7cTfO+xDNBg73HxlkrJxBUuSpRqJuoiEiMMMYQ2BIKBBd4FzB41mD2BfZxUp2T+KL3F8Q74wt8Ld8CHwTAUdOBq1nx3yvamaGHGlcTF45EVU0iEpsyfBkM/ngwny75FLflZvSxoyMDwWBo+YjAqtBY7b+P+htnsyizhuYGgjUVCMqhqcYVEYkBBwaCf/j+4IxZZ5DmT+OE2ifwZZ8vSXAlFPhauV1MscDTvfhrChqfgeycCWOqa8IYEYlNuzJ3ceroU/l+9fckOhL58vgvGdp8aHi/CRi8P/+zfITzOCdba27Nc52IQLCpluaRQ1M3URGRSs4YQ2BrKBD8y/cXZ8w6gz2+PfSo1YOv+nz1z6x0BblWwOCbl7OmYGs3jpTivVM0QYOdbuOq79KEMSISszamb+S0MaexdNdSarhrMKH3BHrU7RHeb7yG7JnZ2LtscILneA+mjoG/I69j7Jyx1zUUCErBqOYVEanEjDEEtwUJrg+y3LecM2adwU7vTrrV6MY3fb4hyZ1UqOv5//JjMg1WooW7ffHWFDTGYPaGHlpcjVzFbmEUEamIlu9czmkfnsb69PU08DRg4kkTaVejXXi/nWWT/WM2Js1AHMT3jsdZy4k/6I+4jrENZrfBUcOhQFAKTMGgiEglZm+3CawLsNK/ktNnn8627G10SunEtyd9S7W4aoW71l4b//LQw0dct7hiz0pn0gxWlZwJYzTDnYjEoLmb53L62NPZmbWTlokt+bbvt6Qmp4b32/tyAsH9BiveIr5PfNQeGeFAsHpOIOhRmSoFo2BQRKSSCm4P4l/nZ61/LafPPp0tWVtoX609k06eRA1PjUJdyxiDd64XTGhsn6tB8aoPe3+oq5O7iRtHgoavi0jsmbZ6GmePO5sMfwZdk7vydd+vqZNYJ7w/uCeI9ycvJttgVc0JBKtGKS/tnNmYq1sKBKXQVAOLiFRCwe1B/Gv9rPevZ8CvA9iYuZHWya35ru931PLUKvT1AqsCobEqLojrElestBmvAS+4G7pxVFM1JCKxZ/yS8Qz8eCAZ/gxOqnESU06dEhkI7giSPT0bk21wpDhIOCUheiAImL0GKzknECyBZX4ktqgWFhGpZII7QoHgJv8mTv/1dNbtX0eLpBZMOXkKdRPqFvp6Jtvg+yNnTcGj44q19IMJGOx9Ns4GThx1VAWJSOx5d967XPDFBfiCPs6uezZfn/I1yZ7k8P7A5gDZP2aDHxy1HMT3jY8e5IVW5AkFgs3cOOJVpkrh6VsjIlKJBHeGAsGtga2cPud0VmesJrVKKlNOnkL9hPpFuqZ3oTf0UJLiwNWiGN1D7dAbbGetUDdTTRgjIrHEGMMTM5/g2m+vxTY2Vza8ko/6fES86581XgNrA3h/9kIQnPWdxPeJjzoRjAmGJuACcKW61N1eikzfHBGRSiK4KxQI7vDvYOCcgazYt4LGiY2ZcvIUGiY2LNo1twYJrgsCENc9DstR9ADOTrexkizcjd2aMEZEYoptbEZ8P4L7/u8+AO5qfhevH/86Luc/L9j8K/x45+SMzW7ixHOCJ2pZaQKhdQStlNA+tQhKcWgCGRGRSiA3ENzl38UZv5/B0vSlHJVwFJNPnkxq1dQiXdMEDd55XgBcLVw4axZvQXjLZeFuojEtIhJbsgPZXD7hcj5f8jkAz7R7hps73Rzeb4zB/5cf/1+h2ZpdLV3EdYmL2nvC+A12mo2zthNnAyesLJ17kMpLwaCISAVmbENgW4DgxiB7AnsY9PsgFu9dTL34ekw+eTLNk5oX+dr+pX5MRmg687ijizFpTM5SWK6GLhzJeoMtIrFjV+YuzvrkLH7Z+Atuy83bXd7mktaXhPcbY/DN9xFYGQDA3cGNu507eiDoNdgZNq56LlyNXQRMoNTuQyovBYMiIhWU8RsCGwIEtwVJd6Vz1ryzWLhnIbU9tZl88mRaJbcq8rXtdBv/0pw1BbvGFXnxYmMMdnpolgNHTQWCIhI7Vu1exeljT2fFnhVUc1Xj056f0rdh3/B+Yxu8c7wE1+d0xe8ah7ulO+q1TLbB7De4jnLhOsqF5bTCL9pEikPBoIhIBWTvt/Gv92PvscmsksnZP5/N77t+p2ZcTSafPJm21doW+drG5HQPtUMTGDgbFr17qNlnQtOh70ITxohIzJizcQ6DPx7MjqwdNI5vzNd9vqZdjXbh/SZg8M7yEtwSBAs8x3lwNYn+WG5n2uAFZ2NNviUlT8GgiEgFYozB3m0TWB/AeA37qu7j3J/PZfbO2aS4U5jUdxIdUjoU6zMCawPY20OLwsd1jT5upUBp9ZnQjHiNnLCuWEkSEakwvlzyJZd+eSlZwSy6JHdhQt8J1E/8ZzZn4zNkz8zG3hkqZz3He3DVzycQ3GdDMDRjqLOOU4GglDgFgyIiFYQJGgJbAgQ3BcEFOxJ2MHj6YBbtXUSyO5lv+35L5xqdi/UZ/lV+fPNDawq627nzXeT4sGnN6R7qqu/CVDfFSpOISEXx0i8vcdsPt2EwnF73dMb2HktVd9XwfjvLxvujFzvNBjfE947HWTtv7wtjDCbdgAPczdw4axVvAi+R/CgYFBGpAIzX4F/vJ7gziKOKgw3BDQycFlo+oo6nDt/2/ZZO1TsV/fq2wbfQR2BFaEICZyMn7jbRx64U6Hrpoe6hrgYuApYmORCRyi1oBxkxcQQvz38ZgGuaXcNLx7yEy/HPo7adYZP9Y3Z4Yq74PvE4UvK+cDMmtHSEw+PA1dSFM0WBoBw5Gs1fTrz55pt07NiR5ORkkpOT6dmzJ999911ZJ0tEygE73cb3tw97p42jmoPlvuX0/aFveB3BaadOK14g6DVk/5QdDgTdHdx4enqKvKag8Ya6h7oaurA86tJUUaleEimY/b79nP/R+eFA8PGOj/Pqsa9GBILBnUGyp+UEglUs4k/JJxC0DWa3wZHowN3CrUBQjji1DJYTDRs25KmnnqJly5YYYxg9ejRnnXUWCxYsoH379mWdPBEpA8YY7O02/o1+TMDgqO5g4d6FDJoxiJ3enbRObs3EkybSqEqjIn+GnW6TPTP0gIIrZxKDhkWvGoxtsPfZuBq4cFTX+8aKTPWSyOFtT9/O4LGD+W3Hb8Q54vigxwcMaTIkvN8YQ2BlAN8CHxiwquW0CCZECQSDBrPHYFWzcDd140hUGSpHnoLBcmLw4MER/3788cd58803+fXXX1XpisQgEzAENgYIbg2CB5zVnczcPpNzfjyHfYF9dK3RlW/6fEPt+NpF/ozAlgDe2V7wg5Vo4TnBg7N68d5Cm30GR5JDM95VAqqXRA5t+eblDPxkIKszVlPdXZ0vTvyCE+qcEN5vAgbvXC/BdaGlI5yNnHiO8WC5o6whGDDYe22cNZy4U91Y8So/pXQUOBicP39+oS/etWvXQp8jEAwG+fzzz9m/fz89e/aMeozX68Xr9Yb/nZ6eDoDf78fvr3wLz+TeU2W8t9Kg/Cue0s4/O8smsCEQGjOS7AA3TNwwkUtnX0p2MJsTap/AZyd8RrI7GX+w8GkyxmCvsAn+EXpAsWpZuHq4sONt7KBd9IT7Qg80rro54wRzklbZv3+V9b4OVJB6CVQ3SeFU5Pz7efnPnP+/89nt201qYipfnvglrZJbhctkk2EIzA5g0gxY4DzaiaOlI1Q2Bg+6WADsNBtHLQfORk4CzkCB1hCsyPlXHlT2/CvofVnGmAJN8+ZwOAr8ltcYg2VZBIMHf9vlUBYvXkzPnj3Jzs6matWqfPzxxwwcODDqsQ8//DCPPPJInu0ff/wxiYmJRzqpIlKKftzzI6+se4UgQY5JPobbU2/H4/AU6VqWbdFyU0vq7akHwJYaW1jZYCXGoRk/iyozM5OLL76YtLQ0kpOTyzo5Jaow9RKobpLY8MveX3hp3Uv4jZ+WiS25r+l9pLhTwvtrpNegzfo2uGwXPpePpY2XklY1rewSLDGpoHVTgYPB0aNHFzoRw4YNK/Q5sczn87F+/XrS0tL44osveO+99/jxxx9p165dnmOjvX1t1KgRO3furHQPIxB6uzF16lT69euH2130GQ5jlfKveEoj/4xtCGwPYG+ywQFWVQsseGflO4yYPwKD4cImF/LmMW/idhQtDSbbEJgVwOzOeVPdyYmjecFf9B3y2mkGK8HC3cKNFRd5vcr+/UtPT6dWrVqVMhgsTL0EqpukcCpa/tl+mxemvcC98+8FYFCDQXzQ4wMSXaEXHcYYgkuC2EtDPSysGhauni6shOhlrMk2kAWO+jld6ws5aVdFy7/yprLnX0HrpgJ3E1Vgd+TFxcXRokULALp168bvv//Oyy+/zNtvv53nWI/Hg8eTt2XA7XZXyi90rsp+f0ea8q94jlT+GZ/Bv9GPY7sDR6IDR4IDYwxP/fUUDy9+GIB/t/w3L3R7AYdVtAkFgruDeH/2YrJMaG2rXvE465XMLHXGazAOg7uxG2eV/K9ZWb9/lfGechWmXgLVTVI0FSH//Bl+bv36Vt5c+SYA17e6nue6PIfTESrzjNfg/dWLvTUUCLpauIjrHIfljB7g2ftt8IEz1YmrfvHGWFeE/CvPKmv+FfSeCjWBzObNmwFo0KDBIY+xLIv69esX5tIShW3bEW9YRaTysffZ+Nf7Q+NFqjmw3BbGGO5acBcvLw9NU35fh/t4oMMDRX5YCKwP4P3NC0Gwkizie8fjSCqZWerCs4ce5Yo6TbpULqqXJBalb0vnkq8u4dut32Jh8XSXp7mp9U3hMjm4J4j3Fy9mvwEneLp7cKXm/4ht77PBJjRjaO2S6Z0hUlQFrrnnzZtH48aNGTdu3CGPGzduHI0bN2bx4sXFTlwsueeee/jpp59Yu3Ytixcv5p577mHGjBlccsklZZ00ETkCjDEEdwTxrfBh77Nx1AgFggE7wHW/XRcOBJ/t8iwPHv1gkR4WjDH4FvtCM4YGwVnPScKpCSUWCELO4vLJjmK/2ZbyR/WSxDpjGzau3EjfT/ry7dZviXfG88kJn3Bzm5vD5Z1/jT+0fuD+f9YPzC8QNCY0YyiAu5kbZx2nyk0pcwVuGXz99ddp1aoVt9566yGPu/XWW/nggw945ZVXePfdd4udwFixfft2Lr/8crZs2UK1atXo2LEjU6ZMoV+/fmWdNBEpYcZvCGwJENwSBDc4a4S6GXmDXi6fdTlfbfwKh+Xg7WPf5vJmlxf5M7xzvAQ3hSbycrV2EdcxrsgLyUf9DG9oyLmroSvPOEGp+FQvSSwztuHPxX8yePJg1mWvo2ZcTb7s8yU9avUI7Q8afAt8BFYFAHDWd+Lp4cm3LDRBg9mbM7a6qRtHNfWkkPKhwMHg9OnTGTZs2GHfYFiWxZAhQ4o04Uwse//998s6CSJSCoJ7gwQ2BULdQpMcWJ5QmZrhz2DIzCH837b/I84Rx9jjx3JWw7OK9Bn2fhvvTC92Wmgymrjucbiblux4iHD30IYuPdRUUqqXJFaZoOH/2bvv+Krq+4/jrzPuSCAEwt5TRVSQIcgUKWq1te5RC+Kmrv6Uuts62jrqHlXcolZF69Zat7hQURQnIHvvEUKSu875/v44IZqSQJKbkHHfz8cjD03uuSffex7hnvO+5/v9fN6e8TYnvHsCm1Kb6Nm0Jy+NfondcnYDwC/yiX8cx99Ycpdv7xChPqEKr5FNwuBvCXoIup1d7CZ6z5T6o9JhcNWqVXTr1q1S23bp0qV0faGIiPzsbuBqD2NMMC205C7dxvhGjnj/CGZsmEETtwnPjnyWMe3GVOv3eOs8Yh/HIA5W1CIyPILTqmYKxfycpoeKSGNkPMO9b9/LHz77AymTYv9W+/PcqOdoFWkFgLfGI/ZJ8B5LGCL7R3Db72B9YKEPcXDbu8EsinIazovUpUqHwSZNmrBx48ZKbbtp0yb1ExIRoWSNSL5PankKf4uP3dTGjv70qfCq4lX86r1f8X3+97QIt+DlA15mcKvB1fpdyQVJEl8mwAe7uU1kZAQ7u+Y/gTaxn00P1YWNiDQSqWSKSc9P4q45dwHw266/5d4h9xJ1ohhjSM5NkvwmCabkPXZ4BLtp+e+x29YHWq6F293V+kCptyp9ldC3b19eeeWVSm376quv0rdv32oPSkSkMTBJQ2pZiuSPSUxRyd3A6E8XAwu3LuTAtw/k+/zvaZ/Vnnd+8U61gqDxDfEv4yS+CIKg09kh+oto7QRB3+Bv9XHaOzjNa/6Oo4hIXcgvzOfwKYeXBsFr+l7DI0MfCYJg0hCfHif5dRAE3W5u8B5bURBMGcwGg51tE+4Vxm2rGRRSf1X6SuHkk0/m/fff56677trhdv/85z95//331ZdQRDKWMQZvk0diboLU8hRWloXd3C5TvOX7zd8z5u0xLNq6iO5Nu/Pe2PfYq/leVf5dfpFP7P0YqXlBEYPQ3iEiQyNYbu1ceGxrgeG2q1JnIhGRemvh+oUMfWAor698nSwni6eGP8Vle12GZVn4W3yK3yrGW+4Fa7AHhgkPDlf4HmtiwR1Bu7VNeLew1lRLvVelpvPPPPMMF1xwAa+99hrjxo1jn332IScnh4KCAr799lv+9a9/8eabb3LQQQdxyimn1OKwRUTqJ5MwpFYHawOBMmsDt5mxfga/ef83bEpsYu/cvXn1wFdpn1X13qypJSniM+OQBFyIDIngdqq9kGZiBsuyCHUKaXqoiDQKHy76kKOfPpr18fV0iHbguQOeY0DeAABSy0p6tKbAyipZg92y/BkRxhjMVgMpcLu4wXrqChrOi9Qnlb5qsG2bF154gYsuuoj777+fN998s8zjxhgcx2HixInccsstuh0uIhll2/qQ1PJU0DfwZ5VCf+6l5S8xYfoEir1ihrQcwosHvEheJK9qvythiM+M4y39KXBG9o/UaP/A7X5nyfRQt7Oqh4pI4zBl5hQmvjaRhJ9gQIsBPDvqWTpmdwQg+WOSxFcJAOw2NtGh0TLT/H/O+EHbCCIQ6hnCbqlG8tJwVOkj5Gg0yj//+U8uv/xy/vvf/zJ79my2bNlCs2bN6N27N4ceeiidOnWqrbGKiNRLJvFTpVCs8u8GGmO4c+6dXPrVpRgMh7Q/hKdGPEUTt0mVfpe3xiP+WRxTbMCCUJ+SkuY12D+wPH6+j93c3mHVPBGRhsA3Ppe/dTk3fnIjAEd3OpqHhj5EthsUP0zMTgSFYgB3N5fwvhX3aDXJoEiY3cIm1DlU4TpCkfqqWmf1jh07csYZZ9T0WEREGhRjDP4mP+gbuIO7gSk/xR+//CP3zrsXgLN6ncVtA2/DtSv/Fmw8Q+KbBKkfg7WBVlOLyP4VT1mqSX6xj2VbhDqGam0toojIrrA1sZXxz43nxR9fBOCyPS/jqn5XYVt2UDH0uyTJH4IgGNorRGivivsH+kU+FIPbrqRtRAUN50Xqsxr5+GLTpk2MGTOGr776qiZ2JyJS75mEIbUkRXJeEhMrqRRaThDcmtzKsR8ey73z7sXC4oZ9b+DOQXdWKQj6m4MCBtuCoNvTJeuQrF0SBI1vMIUGp72j6aEi0qAty1/GyIdH8uKPLxKxIzwy5BGu2fea0iCYmJX4KQj2DRHeO1xuENy2LIAkON0c3G4KgtJw1ch8n0QiwbRp09i0aVNN7E5EpN4qvRu4PIW/teK7gQAri1Zy1AdHMWvTLKJOlClDp3BU56Oq9LtSc1Mkvg1aRhCByH4R3I67bqqmvzmY/qTqoSLSkM1YMYMjnjqC1YWraRNuwzMjnmFo26FA8F6bmJkgtSD4wC08IExot1C5+zFesD7QamLhdnHVYkcaPJ3dRUQqySQMqeUpvLVBifHy1gZu882mbzjqg6NYXrScNpE2PDfquSr1EPSLfOKfxfHX+gA4HRwi+0UqLGBQG/xiH8vR9FARadimfjeVU186lVgqxt5N9+a5A56jW7NuQDD7IfF5gtTikiC4X5hQjwqCYNzgF/g4LR3cLi52lmZLSMOnMCgiUknJeUnsIhu7mb3DKUFvrHyDkz4+ia2prezRbA9eOuAlujftXunfU6ZlhAPh/mHcHru2abHxgumhblcXu5kueESk4THGcPW0q/nrB38F4LBWh/HoqEdpFmkWPO4Z4p/F8ZYFxb8iQyK4Xcu/NPa3BtNC3U4ubgdXH5BJo1EjYTArK4sJEybQoUOHmtidiEi94m0OWjiYuNlpyfAH5j/A/33xf3jGY3Sb0UwdOZUW4RaV+j110TKi3HGUrIdxWji4bfWZoYg0PMXJYk556RSe+f4ZAC7oegHXDr4W1w3e04xniE+P460MZnpEhpbfp9X4BpNvIAShHiHsVmobIY1LjZzlmzVrxiOPPFITuxIRqVf8Qj/41BiwmlkVV5UzPn/++s/cMvsWAMZ1G8fkwZMJO+FK/Z66ahnxv7ZNg7Kb2kF1PH36LSINzKqCVRwx9Qg+X/k5ISvEXX3u4pS9Tyl9PzUpQ+yjGP4aHxyIDI+U2zbHpIIPxuxcm1CX0C7/YE5kV6iRMPj+++/zxBNPsGLFCtq1a8cJJ5zAwQcfXBO7FhGpMyZhSC4OqoXuSHGqmNM+PY3nlz0PwJX7XMkVe11RqU+P67JlRJlx+EGvLMuycDu4uO3dCgvjiIjUV1+t+orDnzqcFQUryAvnMbXvVEb1HPVTEEwaYh/E8Nf74EJ0ZBSnzfbvt9v6BzptHEJdQqoWKo1W2h9x3HffffzqV78ikUjQr18/Nm7cyK9+9StuueWWmhifiEidMJ4huSQZBKTcii8C1sbWcsi7h/D8sucJ22Ee2f8R/rT3nyoVBLdrGdHDJevgXdMyosw4inz8jUFl1NDuIdyuCoIi0vC8MPsFRjwyghUFK9gjZw8+HPxh2SAYN8TeKwmCIYgeUEEQTARB0G3nEuqmICiNW6XvDBYUFJCTk7Pdz2+//XaeeuopDj/88NKfXXHFFdx222388Y9/rJlRiojsQsYYUitSeOs87OY2vuWXu92cLXM4YtoRLC5cTItwC54Z+Qyj2oyq3P7ruGUEBFOgTL6BMLjdXNw2mhYqIg2Pb3yu/eBarpx2JQBj24zl8b0ep0WbFj8FwZghNi2Gn+9DuCQI5pUTBOMGs9XgdixpJO/oPVEat0rfGezVqxcPPfRQpba1rIrX1YiI1Hf+Wh9vpRf0EKwgHH2w9gNGvzWaxYWL6d60O+8f9H6lgqC30SP2bozE10EQdDo4ZP8ye5cGQWOCdYEm32C3sgn3DhPqoPYRItLwbE1s5fh/H18aBM/ucTYv9HuBFm1/CoJ+kU/xu8XBTI+oRdaYrHKDoF/sYwoNTicHt7OCoGSGSl99TJ48mYsvvph77rmHO++8k+HDhwPwhz/8gZNOOonjjjuODh06MGfOHF588UWuu+66Whu0iEht8TZ7JJclIUKFUyWfWPQEE2dMJOkn2b/V/jw78llaR1vvcL9+sU/ym2RpL6s6axmxrUBMto3byw2qo+7iIjUiIjVh0aZFHPn0kXyz5htCdog79r6DU9qfgt3ip4qf/laf2LQYptBgZVtER0fLLQTjF/kQB6ezE7SO0E0NyRCVvjN49NFH88MPP3DUUUfxy1/+khNPPJHly5dz9tln88ILL2DbNl9++SU5OTm89NJLXHLJJbU5bhGRGucX+UFYM2A32f7t0RjD37/9O6d9ehpJP8kxnY/h9QNf32EQNClD4ocExa8VlwZBp6tD1mFZhHqGdtkFh/GDqnimyOC2dwnvGcZp7SgIikiD9N6i99jvgf34Zs03tM1qyxv7vcGpXU/FyXN+CoIFPrF3S4JgU4vomAqCYGFJEOyiICiZp0rzkiKRCH/+85859dRTueSSS+jduzcXXXQRl156KWPHjq2tMYqI1DqTCArGmGKDlbf9hUDSTzJxxkSeXPIkABfteRF/6/c3bKv8z9SMMXjLPBJfJzBFQTVSu6VNuH941xeIKZn6ZOfauB1c7ObqkyUiDZMxhrs/v5sLXr8Az3gMyBvA1L2m0qV1lzKzOfzNPrH3Y5iYwWpWckcwq5wgWOCDD6HuIezWem+UzFOtaqIdO3bkiSee4M033+Q///kPe+yxB0899VRNj01EZJcwniG5LIm/ycdqsf2a502JTVyz8BqeXPIkjuVwz373cO2+11YYBL0NwbrA+CdxTJHBygraRUR/Ed2lQdCkDP5GH1LgdnUJ7xHGaeHoYkdEGqR4Ks6Zr5zJ+f89H894nNj5RN7e9226tCsbBL2NHsXvFWNiBru5TdaBWeUHwS1BcbBQjxBOG703Smaq0p3BpUuX8sYbb1BUVMSQIUMYNmwYM2bM4OGHH2bSpEn885//5M4772TgwIG1NV4RkRpljCG1MoW3Nqgc+r/TJudsmcOxHxzLvK3zyHFzeHLEkxzcvvw+qn6RT/LbsusCQ3uGCO2xa4uzGGMwhQbiYOfZuB1d7KZqliwiDdfqras55pljmL5sOrZl8/fef+f/uvzfdh9wees8Yh/GIBm8/0UPiG7XGsKYkkrKbnBH0Gmxa2driNQnlb46ePXVV+nduzc33HADjz32GCNGjOCiiy7CsixOP/10fvzxR4YOHcqIESM47bTTWLt2bW2OW0SkRvjrSiqHNtm+cuhrK15j5JsjmVcwj1ahVrw15q1yg6BJGRLfl10X6HZzyTosi/Be4V0bBBMGs8Fg2RahniFCu4UUBEWkQfti5RcMun8Q05dNJzeUywv9X+DC3S7EzSu7vs9b4xF7vyQItraJji4/CPqbfKywRbhnWEFQMl6lrxAuvfRSjj32WBYsWMDMmTN59NFHue2221ixYgUAOTk53HzzzXz99desWbOG3XbbrdYGLSJSE/x8P6gcGgYr+tMFgzGGm364iaM/OJotyS0MazWMm3e/mb2b713m+cYYUktTFP+3mOR3SfCCdYHRsVEiQyLY2bsuhJUWiCk0OO0dQr1Lpj2pQIyINGD/+uZfjHxkZNBIvtkefLDfBxzS7ZDtinwlFyWJfRADD5x2DtFRUaxQOXcENxnsbJtQzxB2rj4oE6n0v4Lly5eXtpMAGD58OMYYVq5cWWa73Xffnf/85z9MnTq15kYpIlLD/CKf5JKSAPezi4qiVBHjp4/nz1//GYPhzF5n8uoBr9I81LzM870NHrF3frYuMNsiMnTXrwuEoECMv9HHyrYI7R7C7eZiR3WRIyINl+d7XPzmxYx/YTyxVIzD2h3GB4M+YI9Oe5RZH2g8Q3xmnMSMkt6tHR0iIyLbzcgwvsFsNFhNglkT5VUVFclElV4zOGLECO6880723ntvmjdvzrXXXkteXh577bVXudsfeuihNTZIEZGaZJLBHT1TWLZy6JLCJRz3wXF8vflrXMvltoG3cdZuZ5H0kqXb+EUl/QKX1O26QAimp5otBkJBgRi3jbvdJ+EiIg3NpuJNnPT8Sbw+/3UALul5CVftdhVO87LrA/1in/j0OP76kkIwe4cI9dm+ZU9pEMy1CPUIlVtMRiRTVToMPvDAA5x88smMGjUKYww9e/bk3//+N9nZ2bU5PhGRGmV8Q3JpEm+jV6Yx8YdrP+TEj05kfXw9rSOtmTpiKiPajCh9nu3beD94JOYmwAt+5nZzCfXd9RcWxhjMVgOJYFqq20EFYkSkcZi9bjZHTD2CeRvnkeVkcd9e93Fc9+O2m3bvrfeIfxzHxIIPxCL7R3A7bH9Za7xgjaCT5xDqFiqzJEBEqhAGO3TowNtvv00sFiMWi9G8efNaHJaISM3bVjnUX+tj59pYTnBRcN+8+5g0cxIpk2LfFvvy75H/pkuTLqXP81f5DJo7CC8ZpEC7VUm/wLxdX3jAxA1+gY+dbeP2crFbbl8BVUSkIXr1x1c56bmTKEgU0CW7C0/3fZr+HfuXKQJjjCG1IEXiq2BaqNXMIjqi/GbyJhWspXZaOoS6h7YrJiMiVWwtARCNRolGo7UxFhGRWuWv9/FWeFhNLKyQRcJLcMHMC3howUMAHN/leO4bch/Z7k8zHrx1HqnpKaImCtkQ6RfB6bzr+1EZz+Bv8bEsC7eDi9veLbNuRkSkoTLGcN2H1/Hnd4O12iNajuDJfZ6kTZs2pR/aQfA+mJiZILUomKbvdHKIDI6UOz3eJA1+vo/TuuSOoKbQi5SrymHwf82YMYO3336bjRs30rp1aw455BD23XffGhiaiEjN8bf4JJcmIRRUDl1TvIYTPzqR6eunY2Hx935/5497/rHsepRCn9jHMTCwLncdHcZ0wA2n/bZZJcYYTJHBFBvsFjahDiGsZpaaI4tIoxDzYox7cRz/nv1vACZ2mchNfW4i3Dxc9v24yCf+cRx/ow8WhPYJEeq9/fpACFrs+Ft83HYubhd3l6/nFmlIKn1Vc9hhh3HJJZcwevRoAFKpFOPHj+eZZ57BGFO63RVXXMHEiRO55557anywIiLV4RcHlUNNyuC0cPhy45cc9+FxLC9aTm4ol0eHPcqhHcoWvTIpQ/yjOMTBam4xt9NcOjodd+m4TbKkQEwE3O4ubmtd1IhI47EkfwmXz7+cRcWLCNkhbut9G6fvdvr26wPXesSmxyAOhCEyNILbrvxLWBMP1lS7HVzczm6ZO4sisr1KVxx4/fXXy7SR+Otf/8rTTz/NGWecwZw5cygsLOSbb77huOOO47777uPBBx+slQGLiFSFSRpSS1KYrQa7uc3UxVM58O0DWV60nN1zdufDgz/cPggaQ3xGHH+zHwSxYS6+7e+6MfvBp9pmq8FuYxPuHSbUftdXKxURqS3TFk9j2CPDWFS8iFaRVvx3wH85o88ZZYKgMYbkj0li04IgaDe3yTooq8Ig6BcH75tORye4I6ggKLJT1Z7v9MADD3DEEUdw3333lf5s7733ZurUqSxfvpz777+fM844o0YGKSJSHcY3pJan8DZ6mFzDX2b9hVvn3ArALzv8kseGPkZuOHe75yVnJ/GWeWBDdHgUP3sXBsGYwd/qY+eUVAnNszUlVEQaDWMMd3x2Bxe9eRGe8eiR1YNXB75Kj449yhTDMilD4otEaRsfp4tDZL/t+wdu26cpMEGfwa5OsKZa75silVKtMFhYWMiaNWs49thjy3386KOP5qqrrkprYCKSWfwCH3yC+QolX5Zj/fT/VayYaYwhtSqFt9ojPyufCR9P4M1VbwJwSZ9LuHqfq3Hs7auBplakSH4b9BUMDwjjtHbwvdoPg6U9A11wO7u4bV1VvhORRqUoWcRZr5zFE98+AcBv2/+WI1sfSZf2Xcq8x/uFPvGPSmZnWBDuF8bdvfyAZ3yD2WSwsizcLm6dVHkWaciqFAa3/SOMRqOEw2GaNGlS7nZZWVn4/q77JF1EGjZvk0dyYRK29XbfFgit4L+WZYEDOGC5VhASHcANAqNll4RGh9LgaGIGb4XHHG8Ox713HPML5pPlZHH/kPs5vuvx5Y7D3+wT/zQOgNvLJdQzVNsvPfhEu9AEU6DySu4GllMiXUSkIVu8eTFHPX0Us1bPwrEc/rH7P5i450TeWfxO8F5fwlvtEfskBgkgAtGhUZy25Qe80oqhLYJpoXYTvXeKVFWVwuBll13G9ddfX/r9t99+y5FHHrnddvPnz6ddu3ZpD05EGj+/2Ce1NBX0i2phgSG4Q/iz/xrfQApIBOEJ/6f/WiVXEQYThEZD6YXFaxtf45SvTqEgFfSs+vfIf7Nv3r7ljsPEDbGPYpAiWKfXP1zrr90kgruBVraF29PFbqWegSLS+Ly98G1OePYENhZvpHW4Nf/a918c0P0AUiZVuo0xhuTcJMlvkmDAbmETGR6pMOD5RT4UE1QM7eyqdYRINVU6DI4aNarM7fmhQ4eydOnS7bZLJBI89dRTjBo1qmZGKCKNlkkZUktTmCKDlVfSLqHkbuDPWVTtJO97PjfOvpGrv7066FnVegRPjXiKNtE25Y/DN8SmxzCFBquJRXRYtFZDmUkFjeMtLJz2Dk57BzuqT7RFpHExxnDz9Ju57J3L8I3PwGYDeWr/p+jasmuwgVeyXcoQnxkP1moDbjeX8MBwxesDtwRV7J1uTjClXh+iiVRbpcPgtGnTKrVdPB7nqaeeomvXrtUdk4hkAGMMqZUpvA0edouaK5JSkCzgrM/O4vllzwNwVq+zuGXALYSdiu/0Jb5K4K/1wYXoiGitNXM3yaA4jGUsnFwHp52D3VwFYkSk8SlMFHLaS6fxzA/PAHByh5O5c+idZIWzymwXjUdJvZsKAp4VrNV2e1awPtAzmM0l6wO7ujgttD5QJF013j05JyeHAw44oKZ3KyKNjL/ex1vlYefYNVb+e3b+bI7/8Hh+LPgR13K5fdDtnNnrzB0+J7kgSWp+MFUpsn8Eu3nN36EzyZ/dCWzh4LRxsHM1JVREGqcFGxdw1NSj+Hbdt7iWyy1738JZfc7Ctsu+v/qrfPrP74/xDFbUIjIsgtO6gvWBJY3knbyS9YHZmk0hUhOqFQaLior47rvvWLFiBcXFxTRt2pTdd9+d3r171/T4RKQR8gt8ksuSEKLG7sI9vfhpzv78bApThXTM6siTI55k/1b77/A53jqPxJcJAEL7hHA71uznYyZeUhzGBqelg9O6JATqTqCINFKvz3+d3z77WzbHN9M23JYnhz3JiPYjymxjPEPi2wSpuSlChLDyLKIjothZO1kf2N7F7aT1gSI1qUpXPvPmzeOKK67g1VdfJZFIbPd4p06duPDCC/nDH/6w3ac/IiIQfLqbXJrEJEyNlABPeAkunXUp9/x4DwCj247m8WGPV7g+cBu/0Cf2cSzoS9XZIbRnzVUONbGSEOiC3drGbe1i5VgKgSLSaBljuO6D6/jLtL9gMAxuPpipo6bSsUnHMtttq9rs5wdV51e2XEnXUV2xw9tfNxpjMPklH6h1c3DbqX+gSE2rdBj8+uuvGT16NJ7nMXbsWLKzs/nss89YvXo1l1xyCZ7n8dZbbzFp0iTefvttXnzxRVy3xmehikgDZnxDclkSP9/Hzkv/A6PlRcv53ce/49P1nwJwaZ9LuWqfq8rtH1hmHClD/KN40M6huU1kcCTtCwxjDCZmoBAIg9MuuBNoNVEIFJHGrSBewITnJvDCvBcAOKPbGdw6+FYiTqR0G2MMqR9TJL5JBJWiI+AOdJm/dT7dnG7b7bN0fWCTkv6BzbU+UKQ2VDqtXXLJJeTl5fHxxx+Xto1IJpOMHz+eN954g88++4xrr72Wp59+mnHjxnHbbbdx8cUX19rARaThSa1O4a/1a2S93Lur3+Xk6SezLr6O3FAuDw99mF93/PVOn2eMIf5ZSTPjCERGRMqtWFdZxhhMsQkqokYsnE4OTktH/a5EJCPMWTuHo586mtmbZxO2wtw24DbO2P2MMtv4RT7xGXH8NcHdQKe9Q2RwhFQoBT9uv88y6wO7uhVOHxWR9FX6X9cnn3zCOeecU6Z/YCgU4s9//jNffPEF33//PQAnnHACp5xyClOmTKnxwYpIw+Vt8vBWeJBNWus9fONz4/c38qtpv2JdfB39mvfjk0M+qVQQBEj+kMRb7oEN0eHRaoc2Ywx+oY+/wQcDbheXcJ8woS4hBUERyQgvffcSQx4cwuzNs2kfbc/bY9/eLgimlqYofqM4CIIOhAeGiYyMYEXLPw/4hT5mq8Ht4BLqFVIQFKlllb4zaFkWjrP9LXrHcTDGkJ+fX/qzoUOH8sQTT9TMCEWkwSttLA9pndg3JzZz2qen8Z8V/wFgQvcJ3DHoDrLcrJ08M5BaniL5XRIILkgqqlq3I8YP7gKamMHOsoN1LHluhRc2IiKNjed7/PWNv/LXGX8FYFirYTw14inaZf10w8AkDIkvE6SWlLz3t7CDis3Nyj8H/Hx9oNvdxWnjaIq9yC5Q6TA4dOhQ7r33Xk455RSaN28OBP9wb7zxRsLhMHvttVfpths2bCAnJ6fGBysiDc//Npavrq83fc0JH53Aoq2LiNgR7hh0B6f2PLXSz/c3+8Q/iwPg7uYS6lH1gjFmq8FP+thNbJzuDm5LFyusixURyRybCzYz7t/j+M+y4EO53+/2e27qf1OZXq7eOo/4p3FMUdA7MLRniNBeoYqXB3iUrg8MdQ1h5+puoMiuUukweO211zJy5Eh69erF2LFjycrK4tNPP+XHH3/kiiuuIDc3t3TbN998kwEDBtTKgEWk4aipxvKPLXyM8784n5gXo2uTrjw94mn65/Wv/DjihthHMUiB3dYmvG/FDejLFXywjRW2CHcJY+fZKm0uIhnnu8XfcfRzRzNv6zwidoR/7vdPTu5xcunjxjMkv0+SnB3MwLCaWET2j+C02vEsDH+zT6h1KOgfqGmhIrtUpcPgwIED+eijj/jLX/7Cf//7X+LxOHvssQf33HMPEydOLLPtlVdeSceOHSvYk4hkinQby8e8GBfOvJCHFzwMwKEdDuXh/R8mL5JX6X0Y3xCbHsMUBp86R4dGq1S8xvimtAR6aPcQTpYq2olIZjHG8Oxnz3La26ex1dtKp+xOPDPiGQa2HFi6jb/FJ/5JSXEugqme4f7hCj84M6akBQ/gdHAIdQmlVcxLRKqnSr0fBgwYwH/+85+dbjdy5MhqD0hEGod0G8sv2rqI3370W77a9BUWFlf3vZpL+lyCbVXtU+PEVwn8tT64EB0ZrdJYjAlKm9u5NmxAFyoiknGSySRXvHIFN397MwCj2oziieFPlPZyNcaQmp8i8XUCPCAMkf0iuJ0qvsQ0qZL1gSWTNNxOrt5fReqIGgGKSI3b1lieZFA0oKr+u/K/nPrJqWxKbKJVpBWPDn2Use3HVnk/yflJUvODOZ6R/SNVXodiCoJ2EW5nFxZW+deLiDRoazet5cSnT+S9Ne8B8H97/B/X7Xsdrh1cPvrFPokZCbzVHhD0Vw0PDu9wqqdf5EMx2C1tQu1CsBwVihGpQwqDIlKj0mks7/kef//u71z3/XUADG45mCeHP0nnJp2rPA5vrUfiywQAoX1CuB2r9nbnF/vgB1Od/CZ+lX+/iEhDNuPHGRzz4jEsL15OE6cJ9w25j+O6Hlf6eGp5ivjncUgQtIzoG8bdza0w2Bnf4G/2sUJWUIW5jUvKT+2iVyMiFVEYFJEaVd3G8uvj6zl5+sm8s/odAM7e7Wxu7H9jmQp12xhjwAA+wX9/9mV8ExSMmR4DA05nh9CeVascapIGisDpGjSQ95MKgyKSGYwxPPjBg5z/wfnE/Ti9mvbimZHPsFfzoGq8SRoSXyVILSppGdHc3unMCxMz+Ft97BY2oU4h7JySbfXWKlLnFAZFpMZUp7G8MYZPZnzCuEXjWGFWkE02t0Vu4/gVx5NcliRpkmUDXxUuHuwWNpHBkSpNQdpWMMZt5+K201ukiGSOWCzG+S+cz4M/PgjArzv+mof3f5jccFAx3ltf0jKipPBLqHeI0N6hCguEGd9gtgTbup1d3PZaGyhS3+hKR0RqRHUayxtjuO2D27hy5ZUkSdKLXkyxptAn2Se4O1cdVvBl59pERkSqdOFhjMFsMjgtHNzObpXubIqINGRL1i7huGeO4/MNn29XtMsYQ+rHkiIxBqxsi8iQCE6biqsrm4TB3+Jj59i4nVzs5tVvLyQitUdhUETSVp3G8hvjGzl92um8tvE1AI7MPZJ7B9xLs3AzsEoKCliATWnA2/Zl2dZ2Pyt9LI2LDbPFYGVZuF1d9REUkYzxznfv8NtXfsu6xDpahFvw6NBHOaTDIUAwLTT+eRxvWUmRmC4OkYERrPAOWkZsNZAEt72L28GtVkVpEdk1KhUGP/jgg2rtfNSoUdV6nog0HNVpLP/Juk8Y99E4lseWEyHC9W2u55wx59Tpp8Z+UUlvrK4udraaHotI4+f7Pje/dTOXf3o5Pj79mvfj6ZFP071p9+DxAp/YR7FgqqcF4f5h3F47KBJT0jJi24dqdkvdDRSp7yoVBkePHl3mH7MxplL/uD3Pq/7IRKRBqEpjed/43Dr7Vq785ko849GTnkxpPYX9DtyvTi8YTMJArKRgTAs1lReRxm9L4RZOe/Y0nlv8HADjuo3jrv3uItvNBiC1IkX8szgkwYpaRIZFcFpX/P7oF/oQA7tVybTQSi4XEJG6Vakw+N5775X5Ph6Pc8kll1BUVMRZZ53FHnvsAcCcOXN44IEHaNKkCTfeeGPNj1ZE6pWqNJZfF1vH6Z+ezhur3gDgGI7httzbaDOqTZ2uzTNesK7Fba+CMSKSGWYvn80x/z6G2Vtm41outwy8hYm9JmJZVtAe6PskyR+SQBDuIsMiFYY745U0kA8FrXicNo7WW4s0IJW68jnggAPKfD9p0iTC4TCffvop0Wi09OeHH3445557LgcccACvv/46Bx10UM2OVkTqjao0lv9w7YecPP1kVhavJGpFuYEbGB8ZT/ao7ArXnewKxhjMZoOTV1IwRtOZRKSRe37m85zy+ikUpApoH23PUyOeYmjroQCYuCH+WRxvVTCzy93NJdwvXHG10JKWEU6eE9wNbKq7gSINTbX+1T7xxBOMHz++TBDcJjs7m/Hjx/Ovf/0r7cGJSP1k4j81lrdyKw5QvvG54fsbOPjdg1lZvJLdI7vzNm9zsnMyWSOy6vzCweQbrCYlBWNU7lxEGrGUl+Lyly/nmFePoSBVwPDWw/n0l5+WBkFvk0fxW8VBEHQgMiRCZECk3CC4rYG8iRncLi6hXqE6fz8Xkeqp1r/cwsJCVq1aVeHjq1atoqioqNqDykTXX389++23Hzk5ObRp04YjjzySuXPn1vWwRMownsFb65GYk8Bb4+2wsfya4jX8etqvueqbq/CNz0mtT+LtxNv0sfoQHhTe4dqTXcEv9MGBUNeQ1raIlEPnpcZj/Zb1HPbQYdzw1Q0AnLf7ebwx5g3aZbUDILU4ReydGKYw+IAs+osobrfyJ4+ZhMHf6GNlW4R2DxHqFNKHaSINWLWugMaOHcsdd9zB888/v91jzz33HHfccQdjx45Ne3CZ5P333+fcc8/l008/5a233iKZTHLwwQdTWFhY10MTwZigEXtyXpLkgiQmZYIqcRW0X3hv9XsMfn0w76x+h2wnm/v2uo9/bvonTa2mQZPi7qFd/ArKMnEDcQh1DmHnKgiKlEfnpcbhi0VfMOj+Qby16i2ynCweG/oYtwy8hZAdwviG+JfxoFCMB047h6yDsiospOUX+JhCg9veJbx7GKe5Cm6JNHTVqpZw9913M2bMGI477jjat29Pr169AFiwYAErV66kZ8+e3HXXXTU60Mbu9ddfL/P9lClTaNOmDTNnzlSLDqlTfiyoFuqt8zC+CRoHV7B+xPM9rv3+Wq777joMhj65ffjXgH/R7dNu4IPT0SHUt46DYMrgF/jB+pbWCoIiFdF5qeF7ZPojnPPOOcT8GD2a9uCZEc+wT4t9APCLfeLT4/jrg7Y6oT4hQnuFyp3tYfxgfTURCPUMqWWESCNSrTDYsWNHvv76a+677z7++9//smTJEgD22msvLr74Ys4880yysrJqdKCZJj8/H4C8vLxyH4/H48Tj8dLvt2zZAkAymSSZTNb+AHexba+pMb62XaE6x8+kDKkNKcwagyk2WDkWhMHHh3K6xqwqXsVpn57Gh+s+BGBC9wncuPeNhD8KY+IGK9fC3s8m5adq5DVViw9mk8FqZWHaGFKpyo1Ff3/paezHr7G+rv+1s/MS6NxUX8RTcf74yh+5f/b9APyy/S95cMiDNA83J+kl8df7pD5NQQxwwR3sYnWwSJnU9u/vKfDzfexcG7ezi9/Ex0/5NTLO+nr8Ggodv/Q09uNX2ddlGWNMLY9Fqsj3fX7zm9+wefNmPvroo3K3ufrqq7nmmmu2+/mTTz5JdnZ2bQ9RpIyvtnzF7UtvJz+VT9SOcnanszmgxQHstXgvWha0JO7GmdVrFvFwfOc7E2lgioqKOOmkk8jPz6dZs2Z1PZxaUZnzEujcVB+sT6znpsU3MbdoLhYWJ7Q7gePbHo9t2WCg/Yb29FzZExubwkghP3T7geJIcV0PW0RqWGXPTQqD9dDZZ5/Nf//7Xz766CM6depU7jblffrauXNn1q9f3ygvRpLJJG+99RYHHXQQoVDdTjNsiCp7/Pxin9TqFGa9ARusphbsYElIyk/x9+/+zs1zbgZgn+b78NjQx9gtZzdS36Twf/TBBne0i51Xx5VDC4O3ulDPqle9099fehr78duyZQutWrVq1GGwMucl0Lmprk1bMI3fvfg71sXX0TzUnIf2f4hD2h8ClBQA+9LDXxLc1bM72TiDnPKLvxgwBcF7pt3Bxm3j1krvwPp2/BoaHb/0NPbjV9lzU7U7LL/xxhs89NBDLFy4kE2bNvG/mdKyLBYsWFDd3Wes8847j1dffZUPPvhghyfcSCRCJBLZ7uehUKhR/kFv09hfX22r6PiZpCG1NoVZbXASDlYza6f9/5YXLefk6Sfz8bqPATiz15nc1P8mstwskguTQRAkKE/utq7bZu4mZjCeIdQjVGFhhMrQ3196Guvxa4yv6ecqe14CnZvqijGGWz64hcvevwzPePTN7cvUkVPpmdMTAH+rT/zjOP5mHywI9w3j7lF+b9VtTeSt7KDtzq4oElPXx6+h0/FLT2M9fpV9TdW6Qrvpppu47LLLaNu2LYMHD2afffapzm7kZ4wxnH/++bzwwgtMmzaN7t271/WQJAMY3+Bv8kmtTOEXBKXC7ZY7v2v22orXOOPTM9iQ2ECOm8O9Q+7l2C7HAgStJ75IABDaK4TbpY6DYMrgF/q4nV3sVioYI1JZOi81DFsTWzntudP494//BuC3XX/LPYPvIdsNpuWmVqWIfxqHBBCB6NAoTtvyA56JBwW2nJYObhdXbXdEMkC1rtLuuOMOxowZw2uvvdYok3RdOPfcc3nyySd56aWXyMnJYfXq1QDk5uaqGI/UCr8gmBLqb/DBBTuv4p6B28S9OFfMuoJ//vhPAPq36M+/hv+LXjlBRWF/q0/s4xgYcDo7hPaq48qhJY2RnTYObvvyPwUXkfLpvFT/zVk3h2OeOoYfNv2Aa7nc1P8mzt79bCzLwhhDcnaS5LdBEQk7zyYyPIKdXX7A87f6kAS3k4vbwVXvQJEMUa0wuGnTJo499lgFwRo0efJkAEaPHl3m54888ginnHLKrh+QNFomHkwJ9dZ4kCKYElqJk/6cLXMY//F4vtn8DQDn7n4u1+97PREnmBJmEobYBzFIlFx0DI7UafgyJiiFbufahLqEKmyHISLl03mpfnvhhxeY8OIECpIFtIu0Y+rIqQxtPRQIZkTEZ8TxlgWlQd0eLuEB4XLfB7d9aGaFLUI9Qtit1DZCJJNUKwwOHjyYuXPn1vRYMprq+Miu4K3z8Nf5+IU+dlMbq9nOT/jGGKYsnMKkmZMo8opoFWnFA0Me4LCOh/20jW+IfxLHFBisLIvIiEidfKpsjAGfoIVEUUlPrK6hna5/FJHt6bxUP3m+x5/f+TM3TL8BgOEth/PkyCdpl9UOAL/IJ/5hyfpAG8IDwoR6lv/hvUma0rYRoS4h7BxNCxXJNNUKg/fccw+HHnoogwYN4qSTTqrpMYlIDfMLg2IuqUUp3Khb6YbBmxObOWfGOTy37DkAxrQdw8NDH6Z9Vvsy2yW+SuCt9sCByMhIja4zMb4J+l75Jf9fEvbwgu8tY2EwWJS8HhtwwHKtYM1LFSuHiojUV+uL1nPSv0/ircVvAXB+r/O5fuD1hOwg7HnrPeIfxzGx4MOw6PAoTuvy1wf6xT4UgdPWIdRZH5qJZKpqhcETTjiBVCrF+PHjOfvss+nUqROOU/bNxrIsvv766xoZpIhUj/GCKaHJ5cGaESvXwg5XLhxNXzedCdMnsLRoKa7lck3fa5i056SgV9XPJOclSc0PmrdH9o9Uq1qnMQaz1WCSQbgDfgp4FkF7C5tgTaNLsMYxbEMIrJAVTH0qCYA4lP6/FdLFjYg0DjNXzuToqUeztGAp2XY2kwdP5sTuJ5Y+nlyUDIp3+WDn2sEHc022f783xmC2mGBtd1cHt13ttI0QkYahWmEwLy+Pli1bsttuu9X0eESkhvhFPqnlJQVitlV6r0RO83yPf/zwD/723d/wjU+Ppj14bNhj7Ndyv+23Xe2R+KqkcmjfEG6nqr+lGBNUNLWzbJzWQc+rbeEOl+2Dno3Ws4hIRnn4q4c55z/nEPfi9MzuydMjn2afvKCSu/ENiW8SpOYGH8o5HR0iQyLlfhhmvGAttZVV0jYijVY7ItI4VCsMTps2rYaHISI1xfgGf71PckUSYsHdwMqGp2WFyzj1k1P5cN2HAJzU7STuGHQHzULbNyv1t/jEpgeVQ91uLqHe1SsoZbYY7IgdFC7QehURkVLxVJzz/3s+D3z5AAC/avMrHhrxEC0iLYCgcFf80zjeqqBQTKhPiNDeofL7ByYM/hYfJ6+kbUQFVUVFJLPUbQMwEalRfrFPakUKf11wN9DKKwmC3s6f++KyF/n9jN+zKbGJpm5T7hp0Fyd1L39NsIkbYh/GIAl2K5vwoHC17tb5BUGBA7e7qyAoIvIzy/KXcewzxzJj5QwsLK7sfSWX7XtZ6VR9v8An9lEsmPLpQGRwpMK+rv5WHxLgdnRxO6pthIj8JK0wmEwmmTNnDvn5+fi+v93jo0aNSmf3IlJJxhj8DcG0UFNkgruBlVwvV5Qq4uKvLubB+Q8CMChvEI8Oe7S0d+D//h5vjUdiVgKz1WA1sYgOj1arbYNf5IMHoR4hnOaaqiQiss27i97lxGdPZF3ROlq4LXh08KMc0vWQ0se9NV4wMyNBaQVnJ2/791HjB9VCLbekbURrtY0QkbKqFQZ93+fyyy/nnnvuoaioqMLtPK8StyNEJC0mbkitSOGt9YKCKi0rPy30203fMn76eGZvmQ3ARXtexFX7XEXYCW+3rbfOI/FtIrjrCBCC6IgoVrTqFxYmZiAeTC91WikIiohA8IHbzdNv5rJ3LsM3Pv1y+jF15FR65PYofTw1PxWs1TYlPV1HlF/B2aRKeq02K2kb0UyzL0Rke9UKg9dddx033XQTEydOZMSIEYwfP55//OMfNG/enHvuuQfLsrjxxhtreqwi8jPbCq+klqfwt/rYzexKlwY3xnDvvHu59KtLiftx2kXb8fDQh/lFu19st623wSP5XTJoHQHBtM5eLuE9w9ULgnGDKTQ4XRycNgqCIiIABfECTnv5NJ794VkAxncYz51D7yQ7nA0ExV8SXyVILQgKxbhdXcL7VdBIPm7wC3yc1g6hLiGsiO4Gikj5qhUGp0yZwvHHH8/kyZPZsGEDAAMHDmTMmDFMmDCBoUOH8u677zJ27NgaHayIBEzCkFqVKu3tV9m+gQDr4+uZ+NlEXl3xKgCHdjiUB4Y8QOto6zLb+Zt9Et8l8FaUhEAL3B4uoT6hahceMEmDv9UP1q10cDVdSUQEmLt+Lkc9fRSz188mZIW4pc8tnLn3mdh28F5r4obYx7HSmRmhfiFCe5RfKMbf6kMS3E4l6wOrMY1fRDJHta7oli9fzpgxYwCIRIKa9bFYDIBwOMy4ceN4/PHHa2iIIvJz3maPxI8JUitSWNlWcEewkqFq2ppp7Pff/Xh1xauE7TC3DLiFF0a9UCYI+gU+sU9iFL9RXBoE3a4uWYdlERkUqX4QTAVrV9y2JRcoCoIiIrw892UGPziY2etn0z7SnjeHv8nEvhNLg6C/2af4reIgCLoQGRkh3Hv7ol3GGLxNHhgIdQ/hdlYQFJGdq9adwZYtW7J161YAmjZtSrNmzVi4cGGZbTZt2pT+6ESklEkaUqtTpSXE7Ty70o2Ck36Sf636F8/Neg6DYY9me/CvYf+ib4u+pdv4hT7J75OkFqfABD9zOjmE9w5j56a31sR4Bn9zMGXJ7aILFBER3/hcPe1q/vbB3wAY3nw4T4x4gvY57Uu3Sa1IEf80DimwmlpER0TLfT82nsFsMtg5NqGuWh8oIpVXrTDYv39/Pv/889LvDzzwQG6//Xb69++P7/vceeed9OvXr8YGKZLp/C0+yeVJ/M0+dlO7Smv1FhQsYML0CXy+Mfg3e1rP07h5wM00cZsE+y72Sf6QJLUwBSW1YZz2DqF9QjXSkNj4wUWKk+cQ6hpSSXMRyXibijcx7vlxvDb/NQDO7nI2Nw65kbAbFO8yxpCcnST5bRIAu41NdFi03LV/pesDWzq4XV3sqIKgiFRetcLgWWedxZQpU4jH40QiEa699lpGjRrFqFGjMMbQokULnnrqqZoeq0jGMd5PdwONZ6p0N9AYw+OLHufCmReyNbWVJk4T7h18L8d3Oz54PG5IzkmSnJcs7UNot7EJ7xOusQqfxgRB0Mq1CHULVbrAjYhIY/Xtmm85aupRLNi8gKgd5e5+dzOu97jSx03KEP88jre0ZJp+L5dw/3C57/1+oR9UZlb/QBGppmqFwd/85jf85je/Kf2+T58+LFiwgGnTpuE4DsOGDSMvL6/GBimSifytQaVQb6OH1cTCaVb5gLYxvpHzPj+P55Y9B8CI1iOY0HoCR3U+CpMwJH9MkpybhKAoHXbLkhDYtuaqe26rdmo3sQl1D1Wr8qiISGPy9HdPc9pLp1GUKqJLtAtPD3+aAW0GlD7uF/vEP4rjb/TBgvCAMKFeoe32Y4zB5AfN5kPdQ9ht1D9QRKonrabzP5ebm8sRRxxRU7sTyWh+vk9yYRITN9gt7CqtsZu2ZhqnfXIaK4pX4FouV/e9mvN3O5/3f3wfb05QfIZEsK3d3A6mg7Z3avxCwmwx2JEgCFa36IyISGOQ8lNc+tal3PrprQCMaTmGx0c+TqusVqXbeOs84p/EMcUGwhAdFi33AzrjBf0DrSZWsD4wzTXdIpLZaiwMikjN8At9kouT+AkfJ6/yd+oSXoKrv72aW2ffisGwW85uPDbsMfo37098fpzBcwbjpYJpR1Yzi/DeYZxONR8CIahIig1udxc7RxcqIpK51hWu48R/n8i7S94F4I+9/sjfBv4Nxw7e30vXB36XBBO8P0dHRrGbllMoJmHwtwTnBrerW26zeRGRqlAYFKlHTNyQXJzEFAXrAytrzpY5nDL9FL7a9BUAp/c8nZsG3ETW1ixi78bwN/iECUMTCO8Vxu3qVnrtYVX5hT54EOoZwmmupvIikrlmrpzJ0VOPZmnBUpo4TXhwvwc5uvvRpY+bmCH+WTzoGQs4XR0iAyNYoXLWBxb5EAO3vRu0jdD6QBGpAQqDIvWESZUEwS0Gq4VVqTt2xhgemP8Al3x1CcVeMS3DLZk8ZDK/af8bkj8kKZ5THFQIdWF+m/nsuf+ehELbrz+pKX6xDwkIdQvhtFQQFJHM9chXj3D2f84m7sXpld2Lf4/6N31a9Cl93FtbMi00Fqz9Cw8I43bfvgerMQazJej343R1cNupT6uI1ByFQZF6wHiG1NIU3gYvWCNYibt262LrmDhjIv9Z8R8AftHuFzw45EHabm1L8RvFmIKSi4cODva+NiuXr6SP3WdHu0zvNcQNFIPT2cFuo6lLIpKZEl6CC167gMlfTgbgV+1+xSPDHyE3nAsE7XaSs5Mkv//ZtNChUezmFfQP3Gywsi3cLm6NtPsREfk5hUGROmaMIbUiRWpNCju3csVi3lj5Bmd+diZrYmsI22H+3u/vnNf9PFLfpIgtjAFgRS3CA4J1gSk/VbuvIWkwWw1ORwe3gz61FpHMtKpgFcdNPY6PV36MhcWf+/yZK/pegW0FQc/EDLFPY/hrgqaubjeX8MBwuVM+TdLg55esD+ziqhCXiNSKWgmDc+fO5R//+AcPP/xwbexepFFJrU7hrfSCZvLlrBP5uZgX44pZV3D3j3cD0Ce3D1P2n8JehXsRf71kuhHg9nAJ9wvvkr5+JhVcsLjtXdxOCoIikpk+XvIxxz1zHKuKVpHr5jJl6BQO63RY6ePeGo/4pz+bFjowTKh7+dP2/WIfisBtV7I+cCfnBhGR6qpyGFy3bh0LFy6kRYsW7L777mUemzFjBjfccAMvv/wytm0rDIrshLfew1vmQRSsyI5P9t9t/o6Tp5/M9/nfA3DO7ufw993/jjPLIb4yDoCVYxEZFMFps2umEhnP4G/2cVo7wQVLLRWlERGpr4wxTP5sMhe9dxFJP0mfnD48c8Az7JazW/B4edNCh0XLbQlhjAmm+Ps/Wx+o91URqUWVDoPxeJwzzjiDp556CmOCuw/77LMPL774IllZWUycOJFXXnmFrKwszj77bCZNmlRrgxZpDPx8n+SSJDjscPqPb3zu/vFu/jTrT8T9OG0ibbh/8P2MLRpL4s1E0C7ChtCeIUJ7hqrUkzAdxjeYTQYnzyHULaTKdiKScYqTxdy17C7e/TpoG3Fsp2O5b//7aBpqCpQ0kf80jr+2ZFpod5fwgAqmhW5bH5hVsj6wCq2FRESqq9Jh8LrrruOJJ55g//33Z8SIESxatIjnn3+eCRMmsHbtWlatWsWVV17J+eefT15eXm2OWaTB8wt9kouS4FFu0YBtVhWv4sxPz+St1W8BcFiHw5jcezK53+aS2BB0jrdb2kT2i+zSxsMmZTD5BivXItQ9pClMIpJxlm5aypFPHslXG7/Cxubaftdy4Z4Xlk6V325a6KAwoW7lTws1cYNfULI+sLOL3UTrA0Vk16h0GJw6dSoHH3wwr7/+eunPbrnlFi6++GL69OnDnDlzaNeuXa0MUqQxMTFDclESEwtaSFTk5eUvc/aMs1kfX0/UiXJjvxs5JX4KqfdT+MYHF8L9wrg9d906PRM3mEIDFtgtbNwu7k6nt4qINDYfLPiA4549jrWxteQ4OTw54kkO7nAwUDIt9IeSaaGAlVsyLbRZBdNCtxpIgdvJDQpwaZaFiOxClf7oacmSJRxxxBFlfnbUUUcBcMkllygIilSCSRqSS0p6CTYvv5dgYaqQcz8/l+M+PI718fX0a96Pjwd9zPgF40nNToEBp6ND1qFZhHqFaj0IGmPwi3z89T4mbrBb24R7hwntEVJ1OxHJOPd8dA+/eOIXrI2tpW9uX27Z4xYObHsgEEwLjb0fKw2Cbg+XrLFZ5QdBz2A2GizbItQzpEbyIlInKn1nMJFIkJubW+Zn277v1KlTzY5KpBEyniG5NLnDXoJfbPiCCZ9MYH7BfAAu3O1CrvCuwPncwWCCdhEDw7idar8rjPENpsgEdzCzLJzODk6eo+lLIpKR4qk4f3jhD9z/w/0AHNf5OO7e726mL5wOgLfaI/ZpDOKAC5GBEdxu5b9Xl5kWqrYRIlKHqnRFWdEdCJWSF9mxbb0EvTVeub0EU36Km364ib999zc849ExqyP39biP4YuG/9QuoqdLuG/tt4swqZKpoCmwmlg43R3cPE0HFZHMtXrzao6deiwfrwn6B/6t39+4aM+Lgh6uBlLfpfDnBEVi7FybyLCIpoWKSINQpTB4+umnM3HixO1+/utf/xrHKVv1yrIs8vPz0xudSCNgjCG1qqSXYM72vQQXbl3IaZ+cxifrPwHg2M7HcpO5idzZucHdwByLyH4RnNa1W1nOJAz+Vh/LsrCaWbitXezmti5URCSjzVg4g2OePYblxcvJDeXy6LBHObTDoQCYYkPfhX3xC0uqhfZ0Ce9biWqh3VzsPFsfpotInat0GJwwYUJtjkOk0fLX+0EvwayyvQSNMTy+6HEunHkhW1NbaRZqxu0Db+fo1UfjL/V3SbsIY4JpoBQBLjgtHZzWDnaz8qexiohkksc/eZyz3j6LmB9j95zdeXbUs+zRbA8gqBaa/CRJ83jzYFrofhHcLhVMC40FH7Y5LUuqhWpaqIjUE5UOg4888khtjkOkUfI2eySXJiEEdtZPJ/8N8Q2c+/m5vLDsBQBGtB7Bg0MepMOPHUgtTYEFkeER3A61szawzHrAqIXTwcFp6WA1Kb+ojYhIJkmlUlz68qXc+u2tABza4VAeHfooueFcjDEk5yRJfhs0kd8a3UqLA1rgNt/+/bq0ibwHbmcXt72mhYpI/VL7VShEMpS/1Se1OAU+ZXoAvrXqLc787ExWFa/CtVyu7ns1k3pPwvvBIzk/qEAXGVJLQdALehyapMHOtnG6lRSFiepTahERgI35Gznh6RN4e9XbAFza51Ku2ucqHNvBJAzxGXG8FR4AdjebWU1nMSZnzHb7KTMttLumhYpI/ZTW1WYymWTOnDnk5+fj+/52j48aNSqd3Ys0WH7MJ7m4bC/B4lQxf/r6T9z9490A7NFsDx4d+ij98/qT/PGnnlThAWHcrjUcBFPBf8yWkkbxXUJBRVM1ixcRKfXtwm858vkjWVi4kGwnmwf3f5BjuhwDgL/ZJ/ZxLCgAYwfv1aarwZ+3/fWPiRn8wp81kde0UBGpp6p1xen7Ppdffjn33HMPRUVFFW7neV61BybSUJmkIbU4hSkIgqBlWXy96WsmTJ/A7C2zAfj9br/n+n2vJ9vNJrU4ReKrBAChvUOEdgvV3FhKKteVViTdzSXcMqz1gCIiP2OM4cXPXuTkt09mq7eVLtldeG7Uc/Rt0ReA1JIU8c/j4IGVbREZFsFp6ZD0ktvtp3RaaCdNCxWR+q9aH1Vdd9113HTTTYwbN47HHnsMYww33HAD9957L3379qVfv3688cYbNT1WkXrPeEFTeW+jh9XcwliGW2bfwvA3hzN7y2zaRtvy4gEvcsegO4IguCJFfEYcCIJaqE/NBUG/2Mff4GOFLNwewec+TnNHQVBE5Ge8hMfVL13N0W8czVZvK6PbjOaTX35C3xZ9MZ4h/mWc+KdBELTb2mQdnIXTcvvqzqVN5F2LUK8QbicFQRGp/6p1Z3DKlCkcf/zxTJ48mQ0bNgAwcOBAxowZw4QJExg6dCjvvvsuY8eOrdHBitRnxjeklqfw1nrYzW2Wx5Zz+qen8/7a9wE4vOPhTB48mdbR1gB4az3in8TBgNvVJdw/XCPrSUzSYLYYCJcULGjrkrJSae9XRKSxyd+Uz4TnJvDSipcAOGf3c7ix/42E7BB+sU98ehx/fTANNNQnRGivUPkfqMXBL1K1UBFpeKr1brV8+XLGjAkWS0ciEQBisRgA4XCYcePG8fjjj9fQEEXqvzK9BJvZPLP8GQb+dyDvr32fJm4T7h18L/8e+e+fguAmj9hHMfDA6eAQHpx+EDS+wc/3MQUGu5VNeI8woc6hWm9SLyLS0Bhj+HHhjwx7ZBgvrXiJsB3m/iH3c9vA2wjZIby1HrE3Y0EQDEFkRITwPuVMsTcl/yk2uJ1dQr1CCoIi0qBU685gy5Yt2bp1KwBNmzalWbNmLFy4sMw2mzZtSn90Ig2AiRtSK4I7gpvdzVz4xYVMXTIVgMEtB/PI0EfoldOrdHu/wCf2fgySYLe2iQyNpDV10xiDKQ5aRdjNbNz2qlonIlIR4xne+OINTnrnJDYlN9E+2p6nRz7NkFZDgg/2fkyR+DoBBqxci+jwKHbO9gHPeAazqWQ9dk8Xt7Wr910RaXCqFQb79+/P559/Xvr9gQceyO23307//v3xfZ8777yTfv361dggReojYwz+Jp/U8hT+Vp8P4x9yxhdnsKxoGY7lcPlel3P5Xpfj2j/9M/OLfGLTYhAHu7lNdEQ0rTUlJlFSrCACbreSixFVCBURKZdX7HHr27dy2ZeX4eMzuOVgnh7xNB2yO2CShvjncbxlQfE7p6tDZFCk3PdokzD4W3ys5hasB6eFoyAoIg1StcLgmWeeyaOPPko8HicSiXDttdcyatQoRo0ahTGGFi1a8NRTT9X0WEXqDZMsmRa6yiNu4vxt2d+4dc6tGAw9mvZgytApDGk1pOxz4obY+zFMkcHKsYgeEK32FE7jlVyIYOG0dXDaOZqaJCKyA4XrC5n48kSeWPYEACd3P5m79ruLqBPF31LSNmKLAQvC/cO4vcq/0+cX+hAHt4OLaWtg/q5+JSIiNafSYfCiiy5i/Pjx9OvXjyOOOIIjjjii9LE+ffqwYMECpk2bhuM4DBs2jLy8vFoZsEhd8/N9ksuT+Pk+c8wcTvniFL7Z/A0Ap/U8jZv630TTUNMyzzHJkiC4JWhAHD0gihWtehA0xmAKTemdRbedi91cU0JFRCpifMPKxSs55pVj+GzzZziWw439b+Tc3c/FsixSy0qqOqfAyippG9GqnGqhxuBv9oNqod1D2G1sUikV5xKRhq3SYfDWW2/ltttuo3fv3owbN46TTjqJrl27lj6em5tbJiCKNDbGM6RWB0ViPM/jvvX3ccXXVxD347QMt2TykMkc0Wn7fwPGM8Q+iuFv8iEM0dFR7CZVv4tn4ga/wMfOsnF6ODitHCxHIVBEpCImYZj17SyOevsolsSW0DzUnKdGPMWYdmMwviHxdYLk3KBXoN3aJjqs/A/qTMpgNhvsHJtQlxB2rmZiiEjjUOl3sx9//JErr7wS3/f505/+RI8ePRgxYgT33ntvaXsJkcbK3+qTnJcktSTFSm8lR3x9BH/86o/E/TgHtz+YmYfNLD8I+ob4J3H8tT64ED0git2sahcRJhWsTTTFBrejS3jPMG5bV0FQRGQH/AKfVz98lQP+ewBLYkvo2bQnHxz8QRAEY8FsjW1BMLRHiOjoCoJgPLgjaLe2Ce8WVhAUkUal0u9ovXr14qqrrmL27Nl88cUXXHDBBSxZsoRzzjmHDh068Jvf/Iann36a4uLi2hyvyC5l/OBuYGJuAn+zz4tbX2S/9/bjndXvEHWi3DHoDl4+4GXaZ7Xf/rnGkPgigbfCAxuiI6I4edtPParwd5vgTqDJN9jNbcK9w4S6hqo1vVREJFMYY/DWetz55p0c+dGRFHgFjGw9kg8P/pA9mu2Bt96j+M3i0g/pIsMihPctp20EwQeBpqikbUQPvf+KSONTrQIyAwYMYMCAAdx8881MmzaNJ598kueff55XX32Vpk2bcuSRR/K73/2OQw45pKbHK7LL+MUllULX+2xxtnDxjxfz2KLHAOjfoj9Thk2hd7Pe5T7XmGD6UWpRCiyIDI3gtK1CEEyVfBLd1Mbt4mK3tNNqPyEikglMyhBbFmPStEncu/ReICgUc/d+dxOyQyTnJUnMSoAPVrOSthHlzNYwvsHkGwhBqGcoeA/W2mwRaYTSmutgWRYHHnggDzzwAKtXr+bFF19k+PDh/Otf/+JXv/pVTY1RZJcyxuCt80jOSeKv95menM7gDwbz2KLHsLC4tM+lfHDQBxUGQYDk7CSpuUFhgfCgMG6nyn/usi0IOm0dwr3DOK0dBUERkZ3wC33W/7CeI/5zRGkQvLbftdw/5H5CJkT8sziJL4Mg6HR2yBqbVX4QTBr8jT5WU4vwbuFgfbaCoIg0UtW6M/i/EokEr776Kk8++STTpk0DoG3btjWxa5Fd6ucN5BNOgmuXX8vNs2/GNz5dm3Tl4f0fZkSbETvcR3J+kuS3wTqU8L5hQj1Clf/924JgGyeYEppGD0IRkUxgTBDe5s+Zz9GfHM0PhT+Q5WQxZegUjux8JH5BSduI/JK2Ef3CuLtX0Dai2IcicNo6hDqHqt3+R0Skoah2GDTG8M477/Dkk0/ywgsvkJ+fT05ODieccALjxo1jzJgxNTlOkVr1vw3k51nzOPXTU/ly45cAjO8+nlsH3kqzULMd7ie1NEViZgKA0J4hQntUIwi2dgh1UxAUEdkZ4wU9Xz/+7mOO/+p41iXW0T6rPc+Pep4BeQNILS9pG5EEK1rSNqJ1+W0jTIEJ7hp2cXDbu5qRISIZocphcMaMGTz55JM888wzrFmzBtd1OeSQQxg3bhy/+c1viEajtTFOkVrz8wbyxjI8uPFBLp11KcVeMS3CLbh7v7s5pssxO91PalWK+GdxANyeLqF9qhAEPQVBEZGq8GM+qWUpnvzmSX7/w++J+3H2bbEvz416jo7RjkHbiDk/tY2IDI1gZ5UzLdQL2kZYWRZuF7dKhb5ERBq6SofBK6+8kqeeeoqFCxdijGHYsGH85S9/4YQTTlCDeWmwft5Afm1oLb+f9XteX/k6AGPajuHB/R+kY3bHne7HW+8R/zheuhYlPCBc6TUmxgvuSjqtSoJgSEFQRGRHvM0eySVJ/v7N37l24bUAHN7xcKYMnUITrwmx92NBtVDA3d0l3K/8aqEmYfC3+Dh5TlCsK1ttI0Qks1Q6DP7973+nd+/e/PWvf+V3v/sd3bp1q8VhidQukzSk1gYN5I1v+E/xfzj7g7NZH19PxI5w7b7Xcu7u52JbO74wMMaQmpci8U0CPHDaOUSGRCo9vag0CLZUEBQR2RnjG1JrUhQuKWTiNxN5ZtUzAEzqPYlr970Ws8FQPL0YU2yCthGDI7idy7/U8Yt8iIHb3sXt5Or9V0QyUqXD4MyZM+nfv39tjkWk1hkvKDSQWhWsDSyMFHLp3Et5eMHDAOzTfB8eHfooezXfa6f78gt94jPipZ8+O+0cIsMjlW4GXyYIdlehAhGRHTEJQ3JZklXLVnH8N8czY9MMXMvln/v9k1N6nBJ8MDcrAWYnbSNMSdsIG9xuLk5bVQsVkcxV6TC455578vvf/5699tqL888/v8Lt7rzzTmbPns2dd95JKFT5NVMitcmYYE2et9rD3+xDCD73P+fUaaeycOtCLCwu6H0B1/S9hogT2em+UgtLLjpSgBNUDXV7ll+drtx9eAazyeDkKQiKiOyMX+CTXJrk2xXfcszXx7C0aCktwi2YOmIqB+QdQPzTON5SDwim6kf2i5R7p690fWCTkvWBzbU+UEQyW6XD4P3338+UKVP44Ycfdrjdr371Ky655BL69u3L2WefnfYARdLlF/ikVqfwN/oYDH4zn3/M+QfXfX8dnvHonN2ZB/d/kNFtR+98X0U+ic8TeKuDiw67lU1kSAS7aeXXmRg/uBix82wFQRGRHTDG4K/zSS5L8vrK1zn5m5MpSBXQK6cXL4x6gV70ovjtYsyWkrYR+4Zxdyv/gzkTN/gFwWwMt4tbbjEZEZFMU+kw+Mwzz3DMMcfQo0ePHW7Xs2dPjjvuOJ566imFQalTfpGPt8bDW++BB1aOxaLYIk5971Q+2/AZACd2PZE7Bt1B83DzHe7LGIO3xCP+ZVCiHBvCfUsuOqpQftz4BrPRYLcoCYIRBUERkYqYLYbk4iT3LLuHi7+/GN/4jGoziqdHPE2ztc0onlEMqR23jQDwt/qQBLeTi9vBVcVmEZESlQ6D3377Lb/73e8qte2wYcN45ZVXqj0okXSYhCG1LoW3xoNYEAIJw+OLHufCmReyNbWVZqFm3DXoLk7sduLO9xczxL+I460ouRvYwiayf6TctSg73I8fTA21WlgKgiIilZBMJrnguwu4b9l9AJzS4xTuHHAn1vcW8blBK58dto3wgyUCVtgi1COE3crW+kARkZ+pdBhMJBKEw+FKbRsOh4nH49UelEh1mJTB2+DhrfLwi3ysbAurpcWmxCbO/fhcnl/2PAAjWo/g4aEP07VJ153uM7U8RfyLOMQBG0J9QoT2DFW5GXFpEMwtCYJRXYyIiOxIfiyf4185njeXvYmFxbX7XssFXS8g8WECf11QuCu0R4hQ3/Lfk00qCIJ2M5tQ1xB2jqaFioj8r0qHwQ4dOvDdd99VatvvvvuODh06VHtQIlVh/KAqZ2pVCr/Ax4pY2C2DT3/fW/0ep396OiuKV+BaLlfucyUX7XkRjr3jogEmYYh/GcdbUnI3MNcmPCSM06LqxQZKg2CzIAjaUV2QiIjszD2f38ObS98k285myvAp/Drya+JvxTGxkrYRQyK4nSpoG1HsQyE4bR1CnTQTQ0SkIpUOg2PHjuWxxx7j8ssvp02bNhVut3btWh577DGOO+64GhmgSEWMMZgtJigOs8kHJ5jCadkWcS/OVd9cxe1zbsdg6JXTi8eGPsbAlgN3ut/UqhSJzxNBnyoLQr1DhPYKVbplRJkx/jwI9gipYIGISCVdPPxi5q2exxk5Z9CvsB+x6bHKtY0oMOCD09XBbV+1dd0iIpmm0leml156KbFYjDFjxvDZZ5+Vu81nn33GL37xC2KxGBdffHGNDVLkf/lbfZILkyTmJvA2eVjNLOzcIAjOzp/NyDdHctuc2zAYTu95OjN+OWOnQdAkDfHP48Q/iGOKDVaORfQXUcJ9w9ULgqYkCOYoCIqIVJVru9w/8n72Xrp3af9Ap4tD1tis8oOgFxToslyLUK8QoY5Vn9IvIpJpKn1nsEePHjzzzDP89re/ZdiwYfTo0YN99tmHnJwcCgoK+O6771iwYAHZ2dlMnTqVnj171ua4JUP5cZ/kqiTeWg+SQXEYOxxcFBhjuHfevVw26zJiXoyW4ZbcO+ReftPpNzvdr7fGIz4jjikyALi7u4T3CVe74lxpEGyqICgiUh3eOo/CZwuDmR87axuRMPhbfJw8B7ezi91E77kiIpVRpXfLX/3qV3zzzTecddZZxGIxXnzxRR5//HFefPFFioqKOPPMM/n66685/PDDa2u8jdYHH3zA4YcfTocOHbAsixdffLGuh1SvGC8Iacm5SVLLU1jhknWBJT361hSv4cgPjuSCmRcQ82Ic1O4gZh42c6dB0KSCtYGxaTFMUdCIOHpglEj/SPpBsElJEMzWRYmINEx1eW5KLS9ZAhCB6Jgood1D5QZBv9DHbDW47V1CPUMKgiIiVVDpO4PbdOvWjcmTJzN58mQKCgrYsmULzZo1IycnpzbGlzEKCwvp168fp512GkcffXRdD6deMQlDcmmy9PttxWG2eW3Fa5z12Vmsi68jYke4bt/rOGf3c7CtHV8QeOs94p/FMVtL7gb2dAn3C2OFqj+tqDQIZpcEQV2UiEgDVpfnpvC+4WAZgG3htNq+eJcxBpNvwAa3m4vT1lHbCBGRKqpyGPy5nJwchcAacuihh3LooYfW9TDqHb/IJ7kkib8hKCNuZVulJ/uiVBGXzbqM++YF/af2zt2bR4c9yt7N997hPo1nSH6XJDk3GRQjyLII7xfGbZ/WPweMCaqa2tm2gqCINAp1eW6yLIvIvpHgvfp/mJTBbC6Zit81hJ2r91sRkepI7+pX6kw8Hi/Ty3HLli1A0KA3mdz+xNkQefke3jIPU2Twc33YACkvBcDXm77m1E9P5ceCHwE4b/fzuHqfq4k6UZJexa/f3+jjfeFhtgR3A+2uNk4/BxM2O3zeThmCflZZNnYXGy/i4SW96u+vhm37m2gsfxu7mo5fehr78Wusr6s6avrc5KU8Un4Ky/vZHb8E+AU+dksbt7Nbp++3jf1vu7bp+KVHxy89jf34VfZ1WcYYU8tjkSqyLIsXXniBI488ssJtrr76aq655prtfv7kk0+SnZ1di6OrW77xeXHtizy5+klSJkULtwV/6PIH+jfrv8PnWb5F1zVd6byuMxYWCTfBvI7z2JC7YReNXEQaq6KiIk466STy8/Np1qxZXQ+n1ujcJCLScFT23KQwWA9V5oRb3qevnTt3Zv369Q36YsR4htSaFP5KH8LBtFAI7gj+e/a/eWzdY3y47kMADu94OHcNuotWkVY73Od2dwM72zj7Ouk3ITYE6w09sFvbQT+rcP1cr5JMJnnrrbc46KCDCIVCdT2cBkfHLz2N/fht2bKFVq1aKQxS8+cmb5NHal4Kq7kVvIeHwO3obrd2vK409r/t2qbjlx4dv/Q09uNX2XOTpok2UJFIhEgkst3PQ6FQg/2DNglDckUSa61FqEkIK/rTif75Zc9zwdwLKPQKyXayuWXgLZza49QdXgwYz5D8PklqTgoMEIHIoAhup/T/7LeVMbeb2sGFSV79uDDZmYb891Ef6Pilp7Eev8b4mqqrps9NtmuDBSbfYOfawfrApvVvfWBj/dveVXT80qPjl57Gevwq+5oUBqVe8It8UktSeJu8oHl8SUXPwlQhf5z5Rx5Z+AgAA/MG8uiwR9ktZ7cd7s/bWNI3MD+4G+h0cYgMiKR9N9AYE3w67YPb3sXt4KZ/h1FERCoWBqeFQ6hzqN7OvhARaagUBuuJrVu3Mn/+/NLvFy1axKxZs8jLy6NLly51OLLa5+cHFUNNoQnusNnByX7WxlmMnz6eHwt+xMLimLbH8MDIB8gOVbzuxHiG5A9JkrOTP90NHBjB7VzDdwM7udgtGsbdQBGR6qrrc5Pd1CbULVTm3CAiIjVHYbCe+OKLLzjwwANLv580aRIAEyZMYMqUKXU0qtpljMFf7wc9BD2w8oK2EcYY7pp7F3/6+k8k/AQdsjrwwJAHSG5OErIrvuW93d3AziV3A6Np3g30DaZAdwNFJPPU9bnJCpffY1BERGqGwmA9MXr0aDKplo/xDamVKbwVHoTAbh6sAVkbW8sZn57BG6veAIIiMfcNuY9mbjPe3vx2+fvyDMnZSZI/6G6giEhNyrRzk4hIplEYlF3OJA3JpUn8NT5WU6v0zt1bq97i9E9PZ01sDRE7wo0DbmRir4lYllVhD0Bvk0diRgJ/c9CU3unkEBmou4EiIiIiIjujMCi7lF9cUihm40+FYhJegqu+uYpb59wKQJ/cPjw+7HH2br53hfsxfsnawG13A8MldwO76G6giIiIiEhlKAzKLlOmUEwLG8uxmFcwjwnTJzBz40wAzup1Fjf2v5EsN6vC/eySu4EdSu4GqnKdiIiIiDRSCoNS635eKMakgoqhAP9a9C/+8MUfKEwVkhfO494h93JEpyMq3I9lLLwfPBKzE2XuBjqdnbTv3Jm4wS/wsXNK+gbqbqCIiIiINHIKg1KrjG9IrUrhLQ8KxTgtHLYkt3D+5+czdclUAEa1GcUjQx+hU3anCvfj5/vsO29fvJgHgNPRITwwjJ2VXvNh3Q0UERERkUylMCi1xiQNqWUpvDUeVpOgUMyM9TMYP308iwsX41gOf9n7L1zS5xIcu+LS4anlKVKfpsjxciBUcjewi+4GioiIiIikQ2FQaoWJGZKLk6WFYnzH55bvb+Gab68hZVJ0bdKVx4Y9xv6t9q94H8aQmpsi8XUCgI1NN9L2gLa4TdP/s/ULfEjqbqCIiIiIZC6FQalxfrFPclESszkoFLMqvopTPzyVaWumAXBcl+P4537/pHm4eYX7ML4h8WWC1IIUAHYPm++afEe7rHZpjc34BrPZQARCvULYLXU3UEREREQyk8Kg1Ci/sCQIFhisPIv/rPoPZ316FhsSG8h2srl90O2c3P3kHQYwkzTEP4njrQrWB4b3DWN6GpiX3thM0uDn+9i5NqGuIeym6a03FBERERFpyBQGpcb4W32SC5OYIkO8WZzLv7ycyfMmA7Bvi315bNhj7NFsjx3vo8gn/kEcP98HByL7R3A7uRU2na/02Ip9KAKnrUOoc0jTQkVEREQk4ykMSo3wt5TcEYwZ5jnz+N3bv+Pbzd8C8H97/B9/6/c3Ik5kh/vwNnnEP4xjig1W1CIyIoLTsuLCMpVhzE/VQp2uDm47F8tWEBQRERERURiUtHmbPVKLUvgJn3/n/5tzPz+XramttI605qH9H+KQDofsdB+plSnin8QhBVYzi+ioKHaTNNtGeMH6QCvLwu3i4uSlFyxFRERERBoThUFJi7fRI7k4SSwe45KFl/DA/AcAOKDNATw67FHaZ7Xf6T6SPyZJzAoaydttbaLDomlP4zQJg7/Fx8lzcLu42NlaHygiIiIi8nMKg1Jt3nqP5JIkC7YuYNzX45i1aRYWFpftdRl/3vvPuPaO/7yMb0jMSpCaF1QMdbu7hAeF057G6Rf6EAe3vYvbycUKaVqoiIiIiMj/UhiUavHWBkHw+dXP8/tvfk9BqoBWkVY8MvQRDm5/8E6fb1IlFUNXBhVDQ31DhHqH0mrzYIzB5Buwg2DptEm/Mb2IiIiISGOlMChVYozBW+NRuKiQy+Zdxr2L7gVgeOvhPD7scTpmd9zpPvxin/iHcfxNPtgQGRLB7ZLen2Lp+sAmVtA2IlfTQkVEREREdkRhUCrNGENqVYoFcxfwu29+x5ebvwTgoj0v4pq+1+x0WiiAv9kn9mEMUxQ0fo+OiOK0SrNiaNzgF/g4LR3cri52VEFQRERERGRnFAalUoxvSK1M8cLMFzjru7PIT+WTF87jof0f4rCOh1VqH6lVKeLTSyqG5pRUDE2z8bu/1YckuJ1c3A4ulqtpoSIiIiIilaEwKDtlfEPRkiIuf/9y7lpyFwD7t9qfx4c9TpcmXSq1j+T8JIkvSyqGtraJDo9iRdJYH+gb/M0+Vtgi1COE3crW+kARERERkSpQGJQdMp5h4eyF/Pbt3/J5/ucAXND7Av7e7++E7NDOn28MyW+SJOckAXC7uoT3C2M5aQTBVBAE7WZ2sD4wR9NCRURERESqSmFQKmRShpemv8RpH57GptQmmoea8+D+D3J4p8Mr/fz4Z3G85SUVQ/cOEeqTXsVQ4mCKDU4bh1DnUFp3F0VEREREMpnCoJQrEUtw+cuXc+vsWwEYlDeIJ4Y/Qbem3Sr1fBMzxD6M4W8sqRi6XwS3Wxp/buan/TpdHNx2blp3F0VEREREMp3CoGxn2fplnPjMiUxfNx2A83Y/j+v3vZ6wE67U8/3NPrGPYphCA2GIDo/itKl+xVBjgmmhEPQPdNu4Wh8oIiIiIpImhUEp47+z/8vJL53M+vh6moWacf+Q+zmq81GVfn5qRYr4pyUVQ5uWVAxNY02f8Q1mk8FuYsM6cPLUSF5EREREpCYoDAoAKT/FlW9fyfWfXA9A/xb9eWL4E/TM6Vmp5xtjSM5JkvwmKBRjt7GJDkuzYui2QjEtbNxOLiyu9q5EREREROR/KAwKqwpW8dtnf8v7S98HYGKvidw44EaiTrRSzzeeIfFFgtTiFABuT5fwgDCWnUYQTBr8fB+ntUOoa4iUlar2vkREREREZHsKg0LKT/Hduu/IcXO4Z8A9HN/z+Eo/18QMsY9i+Bt8sCDcP4zbK701fSZuMAUGt52L26WkkXyy2rsTEREREZFyKAwKnXM78+xRz9JyZUt2b7l7pZ/nbfKIfxTHFBkIQXRYFKdd9QvFAPhFPsTA6ezgdnTTursoIiIiIiIVUxgUAA7ocgDxLfFKb59aXlIoxgMrxyI6Mr1CMQD+Vh9S4HQtaR2hQjEiIiIiIrVGYVCqxBhDcnaS5LclhWLalhSKCacxLdQYTL4BB0I9Qjit0ru7KCIiIiIiO6cwKJVmUob453G8pR4A7m4u4X3TLBTjB0HQili43Vyc5gqCIiIiIiK7gsKgVIpf7BP/KI6/saRQzIAwoV6htPZpPIPZbLCaWoS6h7CbpjfNVEREREREKk9hUHbK21hSKKbYQLikUEzb9O7gbesh6OQ5uF1d7CwFQRERERGRXUlhUHYotSxF/LOSQjHNLKIj0i8UYxIGf8tPPQTTWW8oIiIiIiLVozAo5TLGkPw+SfL7oFCM094hsn8k7eBmYgZTaHDbu7idS3oIioiIiIjILqcwKNsxKUN8RhxvWUmhmN1dwv3SKxQDJT0E4yU9BDuoh6CIiIiISF1SGJQy/GKfxGcJ/E0+2BAeGCbUI71CMQB+gQ8euF1dnLaOegiKiIiIiNQxhUEp5W/2SXyZwMQMRCA6PIrTOs1CMSYoFGOFLEI9Qzgt1TpCRERERKQ+UBgUABI/JIh/GgcfrNySQjFptnowftA6ws6yCXULYeeqYqiIiIiISH2hMCikVqQoerUIAKedQ2RYBCuUZqGYVEkPwVwrCIJNFARFREREROoThUHB7egS3jeMv8Un1DeUdhD0i31MkcFpWdI6Iqr1gSIiIiIi9Y3CoACQdVAWie8TaRV2Mb7Bz/exHAu3q4vb1sVyFARFREREROojhUEBSLu657ZG8nYzm1BnrQ8UEREREanvFAYlLcYYzFYDKYJG8h3dtBvTi4iIiIhI7VMYlGozKYPJN1hRC7eni93SVv9AEREREZEGQmFQqqW0SEyeg9vZxc7WtFARERERkYZEYVCqpEyRmC4ubjsViRERERERaYgUBqXSVCRGRERERKTxUBiUnSotEpNUkRgRERERkcZCYVB2qEyRmF4qEiMiIiIi0lgoDEqFVCRGRERERKTxUhiU7ahIjIiIiIhI46cwKGWYpMFsMSoSIyIiIiLSyCkMSllxFYkREREREckECoMScMBuauPkOditVCRGRERERKSxUxgUACzXIrxHuK6HISIiIiIiu4gWhImIiIiIiGQghUEREREREZEMpDAoIiIiIiKSgRQGRUREREREMpDCoIiIiIiISAZSGBQREREREclACoP1yN133023bt2IRqMMGTKEGTNm1PWQREQkw+ncJCLSeCkM1hNPP/00kyZN4qqrruLLL7+kX79+HHLIIaxdu7auhyYiIhlK5yYRkcZNYbCeuPXWWznzzDM59dRT6dOnD/feey/Z2dk8/PDDdT00ERHJUDo3iYg0bm5dD0AgkUgwc+ZMLr/88tKf2bbN2LFj+eSTT8p9TjweJx6Pl36/ZcsWAJLJJMlksnYHXAe2vabG+Np2BR2/9Oj4paexH7/G+rp0btq5xv63Xdt0/NKj45eexn78Kvu6FAbrgfXr1+N5Hm3bti3z87Zt2zJnzpxyn3P99ddzzTXXbPfzN998k+zs7FoZZ33w1ltv1fUQGjQdv/To+KWnsR6/oqKiuh5CrdC5qfIa69/2rqLjlx4dv/Q01uNX2XOTwmADdfnllzNp0qTS77ds2ULnzp05+OCDadasWR2OrHYkk0neeustDjroIEKhUF0Pp8HR8UuPjl96Gvvx23b3S3RukqrR8UuPjl96Gvvxq+y5SWGwHmjVqhWO47BmzZoyP1+zZg3t2rUr9zmRSIRIJLLdz0OhUKP8g96msb++2qbjlx4dv/Q01uPXGF8T6NxUFY399dU2Hb/06Pilp7Eev8q+JhWQqQfC4TADBw7knXfeKf2Z7/u88847DB06tA5HJiIimUrnJhGRxk93BuuJSZMmMWHCBAYNGsTgwYO5/fbbKSws5NRTT63roYmISIbSuUlEpHFTGKwnTjjhBNatW8eVV17J6tWr2XfffXn99de3W7gvIiKyq+jcJCLSuCkM1iPnnXce5513Xl0PQ0REpJTOTSIijZfWDIqIiIiIiGQghUEREREREZEMpDAoIiIiIiKSgRQGRUREREREMpDCoIiIiIiISAZSGBQREREREclACoMiIiIiIiIZSGFQREREREQkAykMioiIiIiIZCCFQRERERERkQykMCgiIiIiIpKBFAZFREREREQykMKgiIiIiIhIBlIYFBERERERyUAKgyIiIiIiIhlIYVBERERERCQDKQyKiIiIiIhkIIVBERERERGRDKQwKCIiIiIikoEUBkVERERERDKQwqCIiIiIiEgGUhgUERERERHJQAqDIiIiIiIiGUhhUEREREREJAMpDIqIiIiIiGQghUEREREREZEMpDAoIiIiIiKSgRQGRUREREREMpDCoIiIiIiISAZSGBQREREREclACoMiIiIiIiIZSGFQREREREQkAykMioiIiIiIZCCFQRERERERkQykMCgiIiIiIpKBFAZFREREREQykMKgiIiIiIhIBlIYFBERERERyUAKgyIiIiIiIhlIYVBERERERCQDKQyKiIiIiIhkIIVBERERERGRDKQwKCIiIiIikoEUBkVERERERDKQwqCIiIiIiEgGUhgUERERERHJQAqDIiIiIiIiGUhhUEREREREJAMpDIqIiIiIiGQghUEREREREZEMpDAoIiIiIiKSgRQGRUREREREMpDCoIiIiIiISAZSGBQREREREclACoMiIiIiIiIZSGFQREREREQkAykMioiIiIiIZCCFQRERERERkQykMCgiIiIiIpKBFAZFREREREQykMKgiIiIiIhIBlIYrAeuvfZahg0bRnZ2Ns2bN6/r4YiIiOjcJCKSARQG64FEIsFxxx3H2WefXddDERERAXRuEhHJBG5dD0DgmmuuAWDKlCl1OxAREZESOjeJiDR+CoMNVDweJx6Pl36/ZcsWAJLJJMlksq6GVWu2vabG+Np2BR2/9Oj4paexH7/G+rqqQ+cmqQodv/To+KWnsR+/yr4uhcEG6vrrry/91Pbn3nzzTbKzs+tgRLvGW2+9VddDaNB0/NKj45eexnr8ioqK6noI9YbOTVIdOn7p0fFLT2M9fpU9NykM1pLLLruMf/zjHzvcZvbs2fTu3bta+7/88suZNGlS6fdbtmyhc+fOHHzwwTRr1qxa+6zPkskkb731FgcddBChUKiuh9Pg6PilR8cvPY39+G27+9UQ6NxUsxr733Zt0/FLj45fehr78avsuUlhsJb88Y9/5JRTTtnhNj169Kj2/iORCJFIZLufh0KhRvkHvU1jf321TccvPTp+6Wmsx68hvSadm2pHY399tU3HLz06fulprMevsq9JYbCWtG7dmtatW9f1MERERErp3CQiIj+nMFgPLF26lI0bN7J06VI8z2PWrFkA9OrVi6ZNm9bt4EREJCPp3CQi0vgpDNYDV155JY8++mjp9/379wfgvffeY/To0XU0KhERyWQ6N4mINH5qOl8PTJkyBWPMdl862YqISF3RuUlEpPFTGBQREREREclACoMiIiIiIiIZSGFQREREREQkAykMioiIiIiIZCCFQRERERERkQykMCgiIiIiIpKBFAZFREREREQykMKgiIiIiIhIBlIYFBERERERyUAKgyIiIiIiIhlIYVBERERERCQDKQyKiIiIiIhkIIVBERERERGRDKQwKCIiIiIikoEUBkVERERERDKQwqCIiIiIiEgGUhgUERERERHJQAqDIiIiIiIiGUhhUEREREREJAMpDIqIiIiIiGQghUEREREREZEMpDAoIiIiIiKSgRQGRUREREREMpDCoIiIiIiISAZSGBQREREREclACoMiIiIiIiIZSGFQREREREQkAykMioiIiIiIZCCFQRERERERkQykMCgiIiIiIpKBFAZFREREREQykMKgiIiIiIhIBlIYFBERERERyUAKgyIiIiIiIhlIYVBERERERCQDKQyKiIiIiIhkIIVBERERERGRDKQwKCIiIiIikoEUBkVERERERDKQwqCIiIiIiEgGUhgUERERERHJQAqDIiIiIiIiGUhhUEREREREJAMpDIqIiIiIiGQghUEREREREZEMpDAoIiIiIiKSgRQGRUREREREMpDCoIiIiIiISAZSGBQREREREclACoMiIiIiIiIZSGFQREREREQkAykMioiIiIiIZCCFQRERERERkQykMCgiIiIiIpKBFAZFREREREQykMKgiIiIiIhIBlIYFBERERERyUAKgyIiIiIiIhlIYVBERERERCQDKQyKiIiIiIhkIIVBERERERGRDKQwWMcWL17M6aefTvfu3cnKyqJnz55cddVVJBKJuh6aiIhkKJ2bREQyg1vXA8h0c+bMwfd97rvvPnr16sV3333HmWeeSWFhITfffHNdD09ERDKQzk0iIplBYbCO/fKXv+SXv/xl6fc9evRg7ty5TJ48WSdcERGpEzo3iYhkBoXBeig/P5+8vLwdbhOPx4nH42WeA7Bx40aSyWStjq8uJJNJioqK2LBhA6FQqK6H0+Do+KVHxy89jf34FRQUAGCMqeOR1C6dm7bX2P+2a5uOX3p0/NLT2I9fpc9NRuqVefPmmWbNmpn7779/h9tdddVVBtCXvvSlL33Vk69ly5btojPFrqdzk770pS99NcyvnZ2bLGMa+UeZdeSyyy7jH//4xw63mT17Nr179y79fsWKFRxwwAGMHj2aBx98cIfP/d9PX33fZ+PGjbRs2RLLstIbfD20ZcsWOnfuzLJly2jWrFldD6fB0fFLj45fehr78TPGUFBQQIcOHbDt+l2XTeemmtXY/7Zrm45fenT80tPYj19lz00Kg7Vk3bp1bNiwYYfb9OjRg3A4DMDKlSsZPXo0+++/P1OmTKn3FxS72pYtW8jNzSU/P79R/oOtbTp+6dHxS4+OX/2hc1PN0t92enT80qPjlx4dv4DWDNaS1q1b07p160ptu2LFCg488EAGDhzII488opOtiIjUCp2bRETk5xQG69iKFSsYPXo0Xbt25eabb2bdunWlj7Vr164ORyYiIplK5yYRkcygMFjH3nrrLebPn8/8+fPp1KlTmcc0g/cnkUiEq666ikgkUtdDaZB0/NKj45ceHb+GR+emytHfdnp0/NKj45ceHb+A1gyKiIiIiIhkIC0AEBERERERyUAKgyIiIiIiIhlIYVBERERERCQDKQyKiIiIiIhkIIVBqTfuvvtuunXrRjQaZciQIcyYMaPCbR944AFGjhxJixYtaNGiBWPHjt3h9pmgKsfv56ZOnYplWRx55JG1O8B6rqrHb/PmzZx77rm0b9+eSCTC7rvvzmuvvbaLRlv/VPX43X777eyxxx5kZWXRuXNnLrzwQmKx2C4arUjl6dyUHp2b0qNzU3p0bqoEI1IPTJ061YTDYfPwww+b77//3px55pmmefPmZs2aNeVuf9JJJ5m7777bfPXVV2b27NnmlFNOMbm5uWb58uW7eOT1Q1WP3zaLFi0yHTt2NCNHjjRHHHHErhlsPVTV4xePx82gQYPMYYcdZj766COzaNEiM23aNDNr1qxdPPL6oarH74knnjCRSMQ88cQTZtGiReaNN94w7du3NxdeeOEuHrnIjunclB6dm9Kjc1N6dG6qHIVBqRcGDx5szj333NLvPc8zHTp0MNdff32lnp9KpUxOTo559NFHa2uI9Vp1jl8qlTLDhg0zDz74oJkwYUJGn3CrevwmT55sevToYRKJxK4aYr1W1eN37rnnmjFjxpT52aRJk8zw4cNrdZwiVaVzU3p0bkqPzk3p0bmpcjRNVOpcIpFg5syZjB07tvRntm0zduxYPvnkk0rto6ioiGQySV5eXm0Ns96q7vH761//Sps2bTj99NN3xTDrreocv5dffpmhQ4dy7rnn0rZtW/bee2+uu+46PM/bVcOuN6pz/IYNG8bMmTNLp+ssXLiQ1157jcMOO2yXjFmkMnRuSo/OTenRuSk9OjdVnlvXAxBZv349nufRtm3bMj9v27Ytc+bMqdQ+Lr30Ujp06FDmH32mqM7x++ijj3jooYeYNWvWLhhh/Vad47dw4ULeffddfve73/Haa68xf/58zjnnHJLJJFddddWuGHa9UZ3jd9JJJ7F+/XpGjBiBMYZUKsXvf/97rrjiil0xZJFK0bkpPTo3pUfnpvTo3FR5ujMoDd4NN9zA1KlTeeGFF4hGo3U9nHqvoKCA8ePH88ADD9CqVau6Hk6D5Ps+bdq04f7772fgwIGccMIJ/OlPf+Lee++t66E1CNOmTeO6667jnnvu4csvv+T555/nP//5D3/729/qemgiNUbnpqrRuSl9OjelJ1PPTbozKHWuVatWOI7DmjVryvx8zZo1tGvXbofPvfnmm7nhhht4++236du3b20Os96q6vFbsGABixcv5vDDDy/9me/7ALiuy9y5c+nZs2ftDroeqc7fX/v27QmFQjiOU/qzPffck9WrV5NIJAiHw7U65vqkOsfvL3/5C+PHj+eMM84AYJ999qGwsJCzzjqLP/3pT9i2PqeUuqdzU3p0bkqPzk3p0bmp8hrnq5IGJRwOM3DgQN55553Sn/m+zzvvvMPQoUMrfN6NN97I3/72N15//XUGDRq0K4ZaL1X1+PXu3Ztvv/2WWbNmlX795je/4cADD2TWrFl07tx5Vw6/zlXn72/48OHMnz+/9EIF4Mcff6R9+/YZdbKF6h2/oqKi7U6q2y5ejDG1N1iRKtC5KT06N6VH56b06NxUBXVbv0YkMHXqVBOJRMyUKVPMDz/8YM466yzTvHlzs3r1amOMMePHjzeXXXZZ6fY33HCDCYfD5tlnnzWrVq0q/SooKKirl1Cnqnr8/lemV2yr6vFbunSpycnJMeedd56ZO3euefXVV02bNm3M3//+97p6CXWqqsfvqquuMjk5Oeapp54yCxcuNG+++abp2bOnOf744+vqJYiUS+em9OjclB6dm9Kjc1PlKAxKvXHXXXeZLl26mHA4bAYPHmw+/fTT0scOOOAAM2HChNLvu3btaoDtvq666qpdP/B6oirH739l+gnXmKofv+nTp5shQ4aYSCRievToYa699lqTSqV28ajrj6ocv2Qyaa6++mrTs2dPE41GTefOnc0555xjNm3atOsHLrITOjelR+em9OjclB6dm3bOMqYx3/cUERERERGR8mjNoIiIiIiISAZSGBQREREREclACoMiIiIiIiIZSGFQREREREQkAykMioiIiIiIZCCFQRERERERkQykMCgiIiIiIpKBFAZFREREREQykMKgiIiIiIhIBlIYFBERERERyUAKgyIiIiIiIhlIYVBERERERCQDKQyKiIiIiIhkIIVBERERERGRDKQwKCIiIiIikoEUBkVERERERDKQwqCIiIiIiEgGUhgUERERERHJQAqDIiIiIiIiGUhhUEREREREJAMpDIqIiIiIiGQghUEREREREZEMpDAoIiIiIiKSgRQGRUREREREMpDCoIiIiIiISAZSGBQREREREclACoMiIiIiIiIZSGFQREREREQkAykMioiIiIiIZCCFQRERERERkQykMCgiIiIiIpKBFAZFREREREQykMKgiIiIiIhIBlIYFBERERERyUAKgyIiIiIiIhlIYVBERERERCQDKQyKiIg0cqNHj2b06NGl3y9evBjLspgyZUqdjUlEROqewqCIiMhOfP/994wbN46OHTsSiUTo0KED48aN44cffqjroZX64YcfuPrqq1m8eHFdD0VERBoIhUEREZEdeP755xkwYADvvPMOp556Kvfccw+nn3467777LgMGDOCll16q6yECQRi85ppryg2Db775Jm+++eauH5SIiNRrbl0PQEREpL5asGAB48ePp0ePHnzwwQe0bt269LH/+7//Y+TIkYwbN45vvvmG7t271+FIdywcDtf1EEREpB7SnUEREZEK3HTTTRQVFXH//feXCYIArVq14r777mPr1q3cdNNNAJxyyil069Ztu/1cffXVWJZV5mePPPIIY8aMoU2bNkQiEfr06cPkyZO3e263bt349a9/zUcffcTgwYOJRqP06NGDxx57rHSbKVOmcNxxxwFw4IEHYlkWlmUxbdo0YPs1gxWZM2cOxx57LHl5eUSjUQYNGsTLL7+80+eJiEjDpDAoIiJSgVdeeYVu3boxcuTIch8fNWoU3bp145VXXqnyvidPnkzXrl254ooruOWWW+jcuTPnnHMOd99993bbzp8/n2OPPZaDDjqIW265hRYtWnDKKafw/fffl47jD3/4AwBXXHEFjz/+OI8//jh77rlnpcfz/fffs//++zN79mwuu+wybrnlFpo0acKRRx7JCy+8UOXXJyIi9Z+miYqIiJQjPz+flStXcsQRR+xwu759+/Lyyy9TUFBQpf2///77ZGVllX5/3nnn8ctf/pJbb72Vc889niDQHwAABFVJREFUt8y2c+fO5YMPPigNpccffzydO3fmkUce4eabb6ZHjx6MHDmSO++8k4MOOqhSdwH/1//93//RpUsXPv/8cyKRCADnnHMOI0aM4NJLL+Woo46q8j5FRKR+051BERGRcmwLdzk5OTvcbtvjVQ2DPw+C+fn5rF+/ngMOOICFCxeSn59fZts+ffqUuTvZunVr9thjDxYuXFil31mRjRs38u6773L88cdTUFDA+vXrWb9+PRs2bOCQQw5h3rx5rFixokZ+l4iI1B+6MygiIlKOyoa8goICLMuiVatWVdr/xx9/zFVXXcUnn3xCUVFRmcfy8/PJzc0t/b5Lly7bPb9FixZs2rSpSr+zIvPnz8cYw1/+8hf+8pe/lLvN2rVr6dixY438PhERqR8UBkVERMqRm5tLhw4d+Oabb3a43TfffEOnTp0Ih8PbFYnZxvO8Mt8vWLCAX/ziF/Tu3Ztbb72Vzp07Ew6Hee2117jtttvwfb/M9o7jlLtfY0wVXlHFtv2+iy66iEMOOaTcbXr16lUjv0tEROoPhUEREZEKHH744dx333189NFHjBgxYrvHP/zwQxYvXsykSZOA4G7d5s2bt9tuyZIlZb5/5ZVXiMfjvPzyy2Xu+r333nvVHmtFQbQyevToAUAoFGLs2LHV3o+IiDQsWjMoIiJSgYsuuojs7GwmTpzIhg0byjy2ceNGfv/739OsWTPOO+88AHr27El+fn6Zu4mrVq3arhrntjt9P7+zl5+fzyOPPFLtsTZp0gSg3DC6M23atGH06NHcd999rFq1arvH161bV+1xiYhI/aU7gyIiIhXo1asXjz32GL/97W/ZZ599OP300+nevTuLFy/moYceYtOmTUydOrW04fyJJ55YWnnzD3/4A0VFRUyePJndd9+dL7/8snS/Bx98MOFwmMMPP5yJEyeydetWHnjgAdq0aVNuGKuMfffdF8dx+Mc//kF+fj6RSKS0j2Fl3H333YwYMYJ99tmHM888kx49erBmzRo++eQTli9fztdff12tcYmISP2lMCgiIrIDxxxzDF9++SXXX389Dz74IGvXrsX3faLRKDNnzqRPnz6l27Zs2ZIXXniBSZMmcckll9C9e3euv/565s2bVyYM7rHHHjz77LP8+c9/5qKLLqJdu3acffbZtG7dmtNOO61a42zXrh333nsv119/Paeffjqe5/Hee+9VOgz26dOHL774gmuuuYYpU6awYcMG2rRpQ//+/bnyyiurNSYREanfLFNTq89FREQyxGOPPcYpp5zCuHHjeOyxx+p6OCIiItWiO4MiIiJVdPLJJ7Nq1Souu+wyOnXqxHXXXVfXQxIREaky3RkUERERERHJQKomKiIiIiIikoEUBkVERERERDKQwqCIiIiIiEgGUhgUERERERHJQAqDIiL/334dCAAAAAAI8rce5LIIAGBIBgEAAIZkEAAAYEgGAQAAhmQQAABgSAYBAACGAsY765eLtrz0AAAAAElFTkSuQmCC", "text/plain": [ "