{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Uncertainties and Relationships in Complex Models\n", "\n", "We have a complex box-model with six inputs and 22 outputs. Each input has an uncertainty associated with it. To calculate the uncertainties associated with the outputs, we use a Monte-Carlo approach. \n", "\n", "The model is run 100,000 times. In each iteration, all six input parameters are varied within their uncertainty (mean +/- 2 s.d., normal distribution).\n", "\n", "There are multiple model runs representing different time periods. Some of the input parameters (and their uncertainties) are defined by time.\n", "\n", "Each model output is a text file containing 28 columns and 100000 rows of tab-separated numbers. The first six columns are the input parameters, the remaining 22 are the model outputs.\n", "\n", "**NOTE**: Because of file size limitations, only 5000 data points from each Monte-Carlo run are provided.\n", "\n", "### Questions:\n", "\n", "1. What is the uncertainty associated with each output parameter?\n", "2. How to work out which parameters have most affect on the outputs?" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import zipfile, os, re\n", "import numpy as np\n", "import pandas as pd\n", "from tqdm import tqdm # progress bars\n", "\n", "import matplotlib.pyplot as plt\n", "%matplotlib inline\n", "\n", "plt.rcParams['figure.dpi'] = 120" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Read Data" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "# check whether the data are uncompressed\n", "if not os.path.exists('ModelData/'):\n", " # if not, uncompress it.\n", " cdata = zipfile.ZipFile('ModelData.zip')\n", " cdata.extractall('./')" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "# column names\n", "inputs = ['i00','i01','i02','i03','i04','i05']\n", "outputs = ['o00','o01','o02','o03','o04','o05','o06',\n", " 'o07','o08','o09','o10','o11','o12','o13',\n", " 'o14','o15','o16','o17','o18','o19','o20',\n", " 'o21']\n", "\n", "cols = inputs + outputs" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "# read data\n", "dat = pd.read_table('ModelData/58cm.txt', sep='\\s+', header=None)\n", "dat.columns = cols" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(5000, 28)" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# check shape is as expected\n", "dat.shape" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", " | i00 | \n", "i01 | \n", "i02 | \n", "i03 | \n", "i04 | \n", "i05 | \n", "o00 | \n", "o01 | \n", "o02 | \n", "o03 | \n", "... | \n", "o12 | \n", "o13 | \n", "o14 | \n", "o15 | \n", "o16 | \n", "o17 | \n", "o18 | \n", "o19 | \n", "o20 | \n", "o21 | \n", "
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0 | \n", "5.513469 | \n", "22.542214 | \n", "16.375471 | \n", "27.948968 | \n", "1.134050e+12 | \n", "-9.221467 | \n", "37.032760 | \n", "37.895427 | \n", "37.624753 | \n", "37.769966 | \n", "... | \n", "1.395776 | \n", "0.969037 | \n", "36.154054 | \n", "0.950379 | \n", "31.339841 | \n", "59.710687 | \n", "8.732933e+11 | \n", "1.568549e+12 | \n", "1.347658e+12 | \n", "91.050528 | \n", "
1 | \n", "8.564659 | \n", "22.761417 | \n", "14.423597 | \n", "27.328296 | \n", "9.999169e+11 | \n", "-10.164063 | \n", "37.614166 | \n", "38.061156 | \n", "37.755283 | \n", "37.916002 | \n", "... | \n", "1.650937 | \n", "1.307719 | \n", "36.128628 | \n", "0.922918 | \n", "28.851489 | \n", "68.004597 | \n", "1.024173e+12 | \n", "1.521369e+12 | \n", "1.440711e+12 | \n", "96.856086 | \n", "
2 | \n", "3.546380 | \n", "23.892493 | \n", "18.753798 | \n", "28.859740 | \n", "8.640674e+11 | \n", "-9.967286 | \n", "38.492943 | \n", "38.427891 | \n", "38.106705 | \n", "38.313568 | \n", "... | \n", "0.714762 | \n", "0.362571 | \n", "36.170447 | \n", "0.968083 | \n", "30.291214 | \n", "73.597780 | \n", "1.400406e+12 | \n", "1.849608e+12 | \n", "1.523907e+12 | \n", "103.888990 | \n", "
3 | \n", "5.686242 | \n", "23.733772 | \n", "18.243887 | \n", "26.500796 | \n", "9.809068e+11 | \n", "-10.099581 | \n", "37.463987 | \n", "38.006094 | \n", "37.722967 | \n", "37.890512 | \n", "... | \n", "0.829963 | \n", "0.512122 | \n", "36.152615 | \n", "0.948824 | \n", "31.191693 | \n", "65.656657 | \n", "1.068848e+12 | \n", "1.677187e+12 | \n", "1.423724e+12 | \n", "96.848350 | \n", "
4 | \n", "6.034157 | \n", "23.952596 | \n", "15.686177 | \n", "28.549757 | \n", "9.885885e+11 | \n", "-10.234018 | \n", "38.294067 | \n", "38.403793 | \n", "38.031375 | \n", "38.232365 | \n", "... | \n", "1.972363 | \n", "1.236332 | \n", "36.149715 | \n", "0.945693 | \n", "29.870851 | \n", "74.497722 | \n", "1.368435e+12 | \n", "1.731893e+12 | \n", "1.748829e+12 | \n", "104.368570 | \n", "
5 rows × 28 columns
\n", "Failed to display Jupyter Widget of type HBox
.
\n", " If you're reading this message in Jupyter Notebook or JupyterLab, it may mean\n", " that the widgets JavaScript is still loading. If this message persists, it\n", " likely means that the widgets JavaScript library is either not installed or\n", " not enabled. See the Jupyter\n", " Widgets Documentation for setup instructions.\n", "
\n", "\n", " If you're reading this message in another notebook frontend (for example, a static\n", " rendering on GitHub or NBViewer),\n", " it may mean that your frontend doesn't currently support widgets.\n", "
\n" ], "text/plain": [ "HBox(children=(IntProgress(value=0, max=5), HTML(value='')))" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "585a8e0ff85842fea181dab4bb64a922", "version_major": 2, "version_minor": 0 }, "text/html": [ "Failed to display Jupyter Widget of type HBox
.
\n", " If you're reading this message in Jupyter Notebook or JupyterLab, it may mean\n", " that the widgets JavaScript is still loading. If this message persists, it\n", " likely means that the widgets JavaScript library is either not installed or\n", " not enabled. See the Jupyter\n", " Widgets Documentation for setup instructions.\n", "
\n", "\n", " If you're reading this message in another notebook frontend (for example, a static\n", " rendering on GitHub or NBViewer),\n", " it may mean that your frontend doesn't currently support widgets.\n", "
\n" ], "text/plain": [ "HBox(children=(IntProgress(value=0, max=6), HTML(value='')))" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "032d20e928f84bfc87ff49e75614050d", "version_major": 2, "version_minor": 0 }, "text/html": [ "Failed to display Jupyter Widget of type HBox
.
\n", " If you're reading this message in Jupyter Notebook or JupyterLab, it may mean\n", " that the widgets JavaScript is still loading. If this message persists, it\n", " likely means that the widgets JavaScript library is either not installed or\n", " not enabled. See the Jupyter\n", " Widgets Documentation for setup instructions.\n", "
\n", "\n", " If you're reading this message in another notebook frontend (for example, a static\n", " rendering on GitHub or NBViewer),\n", " it may mean that your frontend doesn't currently support widgets.\n", "
\n" ], "text/plain": [ "HBox(children=(IntProgress(value=0, max=6), HTML(value='')))" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "fe2b4e06c8f44182bbad8e7a15bf3a9a", "version_major": 2, "version_minor": 0 }, "text/html": [ "Failed to display Jupyter Widget of type HBox
.
\n", " If you're reading this message in Jupyter Notebook or JupyterLab, it may mean\n", " that the widgets JavaScript is still loading. If this message persists, it\n", " likely means that the widgets JavaScript library is either not installed or\n", " not enabled. See the Jupyter\n", " Widgets Documentation for setup instructions.\n", "
\n", "\n", " If you're reading this message in another notebook frontend (for example, a static\n", " rendering on GitHub or NBViewer),\n", " it may mean that your frontend doesn't currently support widgets.\n", "
\n" ], "text/plain": [ "HBox(children=(IntProgress(value=0, max=6), HTML(value='')))" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "bb0b74722bfe4268b7ea5e54c6a4fd50", "version_major": 2, "version_minor": 0 }, "text/html": [ "Failed to display Jupyter Widget of type HBox
.
\n", " If you're reading this message in Jupyter Notebook or JupyterLab, it may mean\n", " that the widgets JavaScript is still loading. If this message persists, it\n", " likely means that the widgets JavaScript library is either not installed or\n", " not enabled. See the Jupyter\n", " Widgets Documentation for setup instructions.\n", "
\n", "\n", " If you're reading this message in another notebook frontend (for example, a static\n", " rendering on GitHub or NBViewer),\n", " it may mean that your frontend doesn't currently support widgets.\n", "
\n" ], "text/plain": [ "HBox(children=(IntProgress(value=0, max=6), HTML(value='')))" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "15d1f5fe5fce4ae89a01bf7b8c4f32f1", "version_major": 2, "version_minor": 0 }, "text/html": [ "Failed to display Jupyter Widget of type HBox
.
\n", " If you're reading this message in Jupyter Notebook or JupyterLab, it may mean\n", " that the widgets JavaScript is still loading. If this message persists, it\n", " likely means that the widgets JavaScript library is either not installed or\n", " not enabled. See the Jupyter\n", " Widgets Documentation for setup instructions.\n", "
\n", "\n", " If you're reading this message in another notebook frontend (for example, a static\n", " rendering on GitHub or NBViewer),\n", " it may mean that your frontend doesn't currently support widgets.\n", "
\n" ], "text/plain": [ "HBox(children=(IntProgress(value=0, max=6), HTML(value='')))" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "\n" ] } ], "source": [ "# calculate mics for all depth slices\n", "\n", "outpar = 'o08'\n", "\n", "depths = {}\n", "for f in tqdm(fs):\n", " d = float(dex.match(f).group())\n", " \n", " depths[d] = mine_ouput(outpar, 'ModelData/' + f, subset=1000)\n", "\n", "# convert outputs into a DataFrame\n", "depths = pd.DataFrame(depths).T" ] }, { "cell_type": "code", "execution_count": 33, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdQAAALDCAYAAACsHPHSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAASdAAAEnQB3mYfeAAAIABJREFUeJzs3Xt4XHd97/v3d0ZXW7asWHZ8iW0FOcYhIZcmIThNG+Ps\nFEqhcGjOQ9hsmwS6UweawiENbZ9ySills0lDoC3kAgWCgQNp0pYAO4WmDYaSuA6B3IPt2FiOE1/k\nm2TLus98zx9rjT0ez0gzozWaNdLn9Tx6xvqti75j6TeftX7rZu6OiIiITEyi2gWIiIhMBQpUERGR\nCChQRUREIqBAFRERiYACVUREJAIKVBERkQgoUEVERCKgQBUREYmAAlVERCQCClQREZEIKFBFREQi\noEAtgpl1mJmb2b3VrkUkztRXZDpToE6Qmb3HzB43sz4z6zWzjWb2ljHmT5rZ/2Nmz5jZgJkdNrOH\nzOyKyaxbZLKV0lfM7Fwz+7iZPWhmL4Uh7WZWN9l1ixTL9LSZ8ZlZPdAJ9Lr73qz224FbgJeBB4AG\n4DrgDOBmd/98znoM+EfgWmAr8L1w3ncCTcDvufuDFX9DIhUSYV/5EPBZIAW8CHQQ9JF6dx+t/DsR\nKZ0CtUzhHuWjwA7gMnc/ErZ3AD8HZgIr3b0ra5l3Af8f8BhwtbsPhu2XAT8FeoFOdz82aW9EpMLK\n7CuvBuYAz7j7gJl1ActQoEqMaci3CAWOC60PXz+Z+YAACD8UvgA0AjfkrOqm8PWjmTANl/kZcB8w\nj2DvVaQmRdVX3H2ru29294EKlywSGQVq+daErz/IM+1fc+bBzJqAK4B+4D+LWUZkiiipr4jUKgVq\nGcxsJrAY6Ms+TpTlxfB1RVZbJ5AEflVgyCrfMiI1rcy+IlKTFKjlaQ1fewtMz7TPmeAyIrVOf/cy\nbShQRUREIqBALU9mq7q1wPRMe88ElxGpdfq7l2lDgVoGdz8OvAK0mNnCPLOcE75uy2rbQXBN3asK\nXJyebxmRmlZmXxGpSQrU8j0Svr4pz7TfzpmH8DKZx4AZwG8Us4zIFFFSXxGpVQrU8t0dvv65mbVl\nGsOL1T8ADAFfzVnmrvD1r8PLaDLLXEZwt6QDwD9VqF6Raimnr4jUHN0Xs0zu/piZ3QF8GHjGzDK3\nU3snJ2+n1pWz2LeBdxDcvOFJM/seMDdcJgn8T3c/OklvQWRSlNNXzKwduD2rqT18/bKZZW7v9r/d\nfUtFixcpgQJ1Atz9FjN7lmAr+0YgDfwC+Bt3/36e+T28/eBjwHuBm4FB4CfAX7v7Y5NWvMgkKrWv\nAC3Ae/K0r8v6972AAlViQ/fyFRERiYCOoYqIiERAgSoiIhIBBaqIiEgEFKgiIiIRUKCKiIhEQIEq\nIiISAQWqiIhIBHRjhwLMrBW4CtgNDFe5HJFyNQBLgB+7e6Fnkk6I+opMIRPqLwrUwq4CHqx2ESIR\neRvw3QqtW31Fppqy+osCtbDdAN/5zndYvnx5tWsRKcv27dt5+9vfDuHfc4Wor8iUMNH+EqtANbNG\n4K+AtUAb8AzwUXd/uMjl3wl8CLgAGAFeCJcv59FQwwDLly/nvPPOK2NxkVip5FCs+opMNWX1l7id\nlHQvwRMpvgl8kOCB3A+Z2ZXjLWhmfwl8i2DL4sPARwkCeXGFahURETkhNnuoZvY64DrgVne/PWzb\nADwH3AZcMcayrwf+ArjF3T87CeWKiIicIk57qNcS7JF+MdPg7oPAl4FVZrZkjGU/BOwD/tYCLRWt\nVEREJEds9lCBi4FteR6w/Xj4ehGFDxRfTfCM0T8iGOqda2b7gE+6++fH+8FmNh+Yl9PcWWzhItOF\n+opIYXEK1IXA3jztmbZF+RYyszagHfh1YA3wceAl4Abg781sxN3vGednvx/4WDlFi0wz6isiBcQp\nUJuBoTztg1nT88kM784FrnP3+wDM7AHgWYI91vEC9U7g/py2TnRtnUgu9RWRAuIUqANAY572pqzp\nhZaD4DKZBzKN7p42s/uAj5vZUnd/qdAPdvduoDu7zcyKrVtk2lBfESksTicl7SUY9s2VadtTYLnD\nBHuxh9w9lTMt0/HbJl6eiIhIYXEK1KeAFWY2O6f98qzpp3H3dDhtnpk15EzOHHc9EFmVIiIiecQp\nUB8AksCNmYbwzkk3AJvdfXfYttTMVuYse1+47Huylm0C3g284O6F9m5FREQiEZtjqO6+2czuBz4V\nnpq/nSAgO4D3Zc26geBm3NkHbu4Bfh/4gpmtIDjLdy2wDHhr5asXEZHpLjaBGloHfIJT7+X7Fnf/\nyVgLufuAma0huKPSe4GZBMPAv+PuP6xsySIiIjEL1PDOSLeGX4XmWV2gvRu4viKFiYiIjCNOx1BF\nRERqlgJVREQkAgpUERGRCChQRUREIqBAFRERiYACVUREJAIKVBERkQgoUEVERCKgQBUREYmAAlVE\nRCQCClQREZEIKFBFREQioEAVERGJgAJVREQkAgpUERGRCChQRUREIqBAFRERiYACVUREJAIKVBER\nkQgoUEVERCKgQBUREYmAAlVERCQCClQREZEIKFBFREQioEAVERGJgAJVREQkAgpUERGRCChQRURE\nIqBAFRERiYACVUREJAIKVBERkQgoUEVERCKgQBUREYmAAlVERCQCClQREZEIKFBFREQioEAVERGJ\ngAJVREQkAgpUERGRCChQRUREIqBAFRERiYACVUREJAIKVBERkQgoUEVERCKgQBUREYmAAlVERCQC\nClQREZEIKFBFREQioEAVERGJgAJVREQkAgpUERGRCMQqUM2s0cw+bWZ7zGzAzDab2TVlrOdhM3Mz\n+3wl6hQREckVq0AF7gU+DHwT+CCQAh4ysyuLXYGZvQNYVZHqRERECohNoJrZ64DrgD9z91vd/YvA\nGmAXcFuR62gCPgN8umKFioiI5BGbQAWuJdgj/WKmwd0HgS8Dq8xsSRHr+AjBe7q9IhWKiIgUUFft\nArJcDGxz96M57Y+HrxcBuwstbGZLgT8F3uvuA2ZW9A82s/nAvJzmzqJXIDJNqK+IFBanQF0I7M3T\nnmlbNM7ynwGedPdvl/Gz3w98rIzlRKYb9RWRAuIUqM3AUJ72wazpeZnZG4DfAy4v82ffCdyf09YJ\nPFjm+kSmKvUVkQLiFKgDQGOe9qas6acxszrg74Cvu/vPyvnB7t4NdOest5xViUxp6isihcUpUPcC\ni/O0Lwxf9xRYbh3wauAPzKwjZ9qssK3b3fsnXqKIiEh+cTrL9ylghZnNzmm/PGt6PkuBeuBRYGfW\nFwRhuxP4rWhLFREROVWc9lAfAP4YuJHwshczawRuADa7++6wbSkww923hMt9m/xh+y/AQ8CXgM2V\nLV1ERKa7sgLVzNYAnwden3uZi5m1Ao8B6939P4tdp7tvNrP7gU+Fp+ZvB94DdADvy5p1A3AVYOFy\nW4At5AiP6+x09+8U/85ERETKU+6Q74eAL+W5ZhR37wXuIbiFYKnWAZ8D1hKcaFQPvMXdf1JmnSIi\nIpOi3CHfC4E/GWP6vxEM35YkvDPSreFXoXlWF7kunXooIiKTptw91DOBkTGmj3L63VRERESmrHID\n9RXg/DGmX0D+ux6JiIhMSeUG6kPAJ8Knu5zCzJqBjwPfn0hhIiIitaTcY6h/DbwD2BY+xHtr2L4S\n+ACQBD458fJERERqQ1mB6u77zewK4C7gU4SXsAAO/BD4gLvvj6ZEERGR+Cv7xg7uvgt4s5m1AcsJ\nQvVFdz8SVXEiIiK1YsJ3SgoDtKyb0ouIiEwV5d4p6Z+Lmc/d31HO+kVERGpNuXuovZFWISIiUuPK\nPSnphqgLERERqWXlDvl+pYjZ3N3fN/5sIiIita/cId/rgV3Ak5y8ZEZERGTaKjdQ7wLeBZwNfBX4\nhrsfjqwqERGRGlPWrQfd/QPAQuA24K3AbjP7RzN7o4UPIhUREZlOyr2XL+4+5O7fcvdrgNcAzwN3\nAl1m1hJVgSIiIrWg7EDNkSa47aAR3MdXRERkWik7UM2s0czeZWYPA9uA1wJ/CCx1976oChQREakF\n5V42cydwHbAb+ArwLnc/GGVhIiIitaTcs3zXAy8BvwKuAq7Kdy6Sbj0oIiLTRbmBuoHgmKmIiIhQ\n/q0Hr4+4DhERkZoW1Vm+IiIi05oCVUREJAIKVBERkQgoUEVERCKgQBUREYmAAlVERCQCClQREZEI\nKFBFREQioEAVERGJgAJVREQkAgpUERGRCChQRUREIqBAFRERiYACVUREJAIKVBERkQgoUEVERCKg\nQBUREYmAAlVERCQCClQREZEIKFBFREQioEAVERGJgAJVREQkAgpUERGRCChQRUREIqBAFRERiYAC\nVUREJAIKVBERkQgoUEVERCKgQBUREYmAAlVERCQCsQpUM2s0s0+b2R4zGzCzzWZ2TRHLvcPM7jOz\nX5lZv5ltNbPPmNmcyahbREQkVoEK3At8GPgm8EEgBTxkZleOs9wXgXOBbwB/BPwA+ENgk5k1V6xa\nERGRUF21C8gws9cB1wG3uvvtYdsG4DngNuCKMRa/1t035qzv58DXgHcD/1CJmkVERDLitId6LcEe\n6RczDe4+CHwZWGVmSwotmBumoX8JX8+NsEYREZG8YrOHClwMbHP3ozntj4evFwG7S1jfgvD14Hgz\nmtl8YF5Oc2cJP0tkWlBfESksToG6ENibpz3TtqjE9f0JwR7vA0XM+37gYyWuX2Q6Ul8RKSBOgdoM\nDOVpH8yaXhQz++/A+4Db3P3FIha5E7g/p60TeLDYnykyTaiviBQQp0AdABrztDdlTR+Xmf0GwXHX\nHwJ/Xswy7t4NdOesp5hFRaYV9RWRwuJ0UtJegmHfXJm2PeOtwMwuBL5LcGbwte4+Gl15IiIihcUp\nUJ8CVpjZ7Jz2y7OmF2RmnQTXn3YDb3b3vuhLFBERyS9OgfoAkARuzDSYWSNwA7DZ3XeHbUvNbGX2\ngma2APg3IA280d0PTFrVIiIixOgYqrtvNrP7gU+Fp+ZvB94DdBCcYJSxAbgKyD5w8wPgVQQ3gLgy\n585K+9394UrWLiIiEptADa0DPgGsBdqAZ4C3uPtPxlnuwvD1I3mm/RhQoIqISEXFKlDDOyPdGn4V\nmmd1njadZigiIlUVp2OoIiIiNUuBKiIiEoFYDfnWopFUmie6jtA7MExrcwOXdrRRn9R2iojIdKNA\nLdNIKs1dG3ewYVMXB/uGT7TPa2lk7apl3LS6U8EqIjKNKFDLMJJKc+OGJ/jR1gPkng11sG+IOx7e\nxlO7e7hn7SVTJlS1Jy4iMjYFahnu2riDH20N7h3hOdMy3z+ypZu7N+7g5qvPmdTaoha3PfHpHuy1\n/P5ruXaRYihQSzSSSrNhUxfG6WGazYANm3axvoaHfuO0Jx63YJ9stfz+a7l2qV3V2IBToJboia4j\np3woFOLAgb4hNmzaxWsXt5Kw4KkcCYOEGQkzLPPvRKYtM8/J+Sxr/rzryFq20PzlisueeJyCvRpq\n+f3Xcu3litOeeJxqmSzV3IBToJaod2D8MM12qG+QowMzKlTN+Gy8YE7kD/JU2vnKT39V1M+497Eu\n1l2xjMa6JHUJI5mwSB/pNVnBHtcPn7hs2JSjlmsvVZz2xCejljj2l2pvwClQS9Ta3FDS/IvmzGBu\nSwNpB3cn7ZB2J+2On/g34fdOOh1tve6Q8sxH11iD1Kfasu8YPQPFPf3u0PFhvr7pJVYumHWiLZm0\nIFwtCNi6pJFMJKhLBMEdfG8nArgukSCRgLpE4pR2d+drj+2s6BB7nD4I89VWq4cYarn2UlX7g3wy\na4lzf6n2BpwCtUSXdrTR3tLAob7hcT8k2lsaeedlS0r+40qnPSdoT4bwuMGcLnH+TJB75ucG84+k\nUiXV3D88csr3qZSTShUf4IVs2XeMQ8dHxp0vM8S+cesBrlzeTkNdEMzjmcwPwtFUmtG0B1+pNCMp\nJ5V2RjLt2W3pNKMp5+ndPSUdYnii6wirOudOqM6olHp45B9/tpsLl8w5ZUQleA1GUIxTD5dYziEQ\n4+T3J14JDovkWzaRtc7MuspV7Q/yyaolThsOGemwTw2OpLj3sZ3jzl/JDTgFaonqkwnWrergjoe3\njTmfA+tWLSvrF5ZIGInT/lwnl5nxt/+xvej5z1s0h5ULZ5EK/7hTaWc0FQR48H0QENnTU+nxAzc3\nqMfz5EuHT/zPJZNGYzJBQ13WV/LUf3/5pzuL+vC560fbuekNyxlNOaPhe8mE3om2TDimToZmpi0V\nbuiU6kj/UEnzl3pIopJKreWVnn7OnN1UoWrGlx3SYwZw4tTwTqWdLxd5eOQrP93JG1bOL7ixV+zf\niBfYnB9NlXCo5tEu3nX5Uprqg0M19cnxN0KjDOt0uOF44jMhlfV91mfFaM5nx2gqnfdzZMu+Yxwu\nYeO7EhufCtQy3LS6k6d29/DIlu7ThrMy369ZOZ/1qzurU2AESt0T/63zzix548H9ZCc5EbyprEBO\np+kbKm7YOWNGQ/2Jf6dSTn8qRf9w/r3t0XSaex/rKmq9//DTnaxYMIu6ROW2upOJk8Pj9clgeHzx\nnNKOv5d6SKKSSq2lbUYjzQ3JU0ZR3IPwiPpQSD7umUAr7RDJln3H6C3y8MiRgRE2bj1wyuGRKJV0\nqKZ/mG8/vvuUWsygLpmg/sTfYoL6pIV/985XHx1/DxDgK4/uZPXKeRh2YoN65JQN6nTkv9NSN74r\nsfGpQC1DfTLBPWsv4e6NO9iwaRcH+k7uRbS3NLJu1bKaPh4Ek7MnbuGx1Lpk4XnOnN3EX/+fF8YN\ndoC5Mxv4vUsWk3YYHk2f/EqlGAr/PZo1DL29+zjHBov78Dk6OMr27uNjfhBmjhvXJcIPoTAU65J5\n2rI+rJIJoz6RIJFn7+D8xa189t+3Fb1hc2lHW1HvZzKUulF2/a93jPl35HkOg0DuYREg+/BGOD1o\n42RbOiussw6LeNYy6XQwPRO02evzEz8T0l5aMgyNpqivKzwGVezIs+VZw0QP1bjDyGiafNG0Zd8x\njvQXF1pH+kf48daDE95wMAs2NDN7z5lzK/J9PzhS2nuvxManArVM9ckEN199DutXd8buTLeoxGFP\nvNhgB7j+ig4WtjaPOU867Qyn0gyn0hwucTi1tbmO1yyafSIg65JBENaFQRrlmc0Zk7FhUylR125m\nJA2SVT4ckisYcnyx6Pmv6Gyv2HHuZKK0QzUXnNXG+YtbTwyrjqbzHdsP2gdGShstGhpNMaMhedpJ\niZn+kv195sTEk9PsxAmKxVrQ2lTSBlwlNj4VqBNUn0zE5iSQqMVlTzzKYE8kjKZEkqb6JEvaZpZU\nx/L5s1g0Z+zAroQ4bNiUq5ZrL1ape+KVHEUotZarz51fdP+tSyT4wo92FF1LJTcc8onDxqcCVcYU\nhz3xSgV7nD4IxxKXDZty1HLtxYrDB/lk1FIL/aXaG3AKVClKtffEKxHscfogHE8cNmzKVcu1F6va\nH+STUUst9Jdqb8ApUKWmRB3scfogLEa1N2wmopZrH0+1P8gnq5Za6C/V3IAzL+fiuGnAzM4Dnnvu\nuec477zzql2OVNBIKp33w2feFBiSfP755zn//PMBznf35yvxM9RXThWnW/JVohb1l8K0hyrT3nQY\nkpTJE6c98UrUov5SmAJVJBSnD0KRuFN/Od303pwQERGJiAJVREQkAgpUERGRCOgYamENANu3F38b\nL5G4yfr7reRd89VXZEqYaH/RZTMFmNnvAg9Wuw6RiLzN3b9biRWrr8gUVFZ/UaAWYGatwFXAbiDq\n5/x0EnwAvQ0o/uaYEkdx/102AEuAH7t7byV+gPqKFKkWfpcT6i8a8i0g/M+s1BZ95p87KnWxvUyO\nGvldPlnJlauvSDFq6HdZdn/RSUkiIiIRUKCKiIhEQIEqIiISAQVqdRwAPh6+Sm3T77Ky9P87dUz5\n36XO8hUREYmA9lBFREQioEAVERGJgAJVREQkAgpUERGRCChQRUREIqBAFRERiYACVUREJAIKVBER\nkQgoUEVERCKgQBUREYmAAlVERCQCClQREZEIKFBFREQioEAVERGJgAK1CGbWYWZuZvdWuxaROFNf\nkelMgTpBZvYeM3vczPrMrNfMNprZW/LMZ2b2JjP7ezN7ysyOmNmgmW01s8+Z2ZnVqF9kshTbV8J5\n325m95nZlrCvDJjZi2b2LTO7dLJrFymGHjBeBDOrBzqBXnffm9V+O3AL8DLwANAAXAecAdzs7p/P\nmrcJGACGgZ8ATwNJYA1wAbAf+A13f3Ey3pNIJUTRV8L5vwpcBfwM2EPQb5YDvxMue6O7/0PF35BI\nCRSoZTKzK4BHgR3AZe5+JGzvAH4OzARWuntX2F4PfAS4MzNv2J4A7gT+APi+u7918t6FSOWV2lfC\naU3uPphnXa8lCNlBYL67D1e6fpFiaci3CAWOC60PXz+ZHZDhh8IXgEbghqz2EXc/Zd6wPQ38Vfjt\n6uirF5k8UfSVcNppYRq2Pwv8EmgF5kVXucjEKVDLtyZ8/UGeaf+aM894RsLX0QlVJBJPkfUVM1sB\nvBo4COwdZ3aRSVVX7QJqkZnNBBYDfdnHibJkjoOuKHKV7w1f833giNSsifYVM/tvwJUEx03PBjKH\nRH4/HN0RiQ0Fanlaw9feAtMz7XPGW5GZXQZ8DDgGfHTipYnEykT7yn8D/iTr+33A9e7+wwhqE4mU\nhnyrKBy++h5QD/wPd99R5ZJEYsXd/9TdDWgBfg14BPhXM/vz6lYmcjoFankyW9WtBaZn2nsKrSAM\n0x8RXDZwnbt/N7ryRGJjwn0FwN2Pu/uT7v5u4IfAJ8LRHZHYUKCWwd2PA68ALWa2MM8s54Sv2/It\nb2bnAhuBduD/dvd/qkSdItU20b5SwA8AI7hOVSQ2FKjleyR8fVOeab+dM88J4XV0Gwn2TN/h7g9W\npDqR+Cirr4xhcfiqs+IlVhSo5bs7fP1zM2vLNIYXq38AGAK+mr2AmV1EMMw7C3ibu/+fSalUpLpK\n6itm1mhmF+ZbUTjMux5IobPiJWZ0lm+Z3P0xM7sD+DDwjJllbqf2Tk7eTq0rM3/4QfIf4bT/AFaZ\n2ao8q/6cu495PEmklpTaV4Bm4CkzewZ4juB2hTOAczl5veqt7r5lkt6CSFEUqBPg7reY2bMEW9k3\nAmngF8DfuPv3c2ZvJfjwALg6/MrnXsY5QUOk1pTYV44D/y/BMdKrCM41cIJjsd8AvuDumyerdpFi\n6V6+IiIiEdAxVBERkQgoUEVERCKgQBUREYmAAlVERCQCClQREZEIKFBFREQioEAVERGJgG7sUICZ\ntRJcVL4bGK5yOSLlagCWAD9290LPJJ0Q9RWZQibUXxSohV0F6Mb1MlW8DajUIwLVV2SqKau/KFAL\n2w3wne98h+XLl1e7FpGybN++nbe//e0Q/j1XiPqKTAkT7S8K1MKGAZYvX855551X7VpEJqqSQ7Hq\nKzLVlNVfYnVSUvjYpk+b2R4zGzCzzWZ2TQnLv9PMNpnZcTPrMbPHzGzN+EuKiIhMTKwCleBJKx8G\nvgl8kOCZhw+Z2ZXjLWhmfwl8i2BX/cPAR4FnOPkwYhERkYqJzZCvmb0OuI7gOYe3h20bCJ6HeBtw\nxRjLvh74C+AWd//sJJQrIiUaSaV5ousIvQPDtDY3cGlHG/XJuG3Ti5QvNoEKXEuwR/rFTIO7D5rZ\nl4H/ZWZL3L3QgeIPAfuAvzUzA2a6e1/FKxaRcY2k0ty1cQcbNnVxsO/koal5LY2sXbWMm1Z3Klhl\nSohToF4MbHP3ozntj4evF1H4zKurgceAPyIY6p1rZvuAT7r758f7wWY2H5iX09xZbOEi00WpfWUk\nlebGDU/wo60HsJxpB/uGuOPhbTy1u4d71l6iUJWaF6dAXQjszdOeaVuUbyEzawPagV8H1gAfB14C\nbgD+3sxG3P2ecX72+4GPlVO0yDRTUl+5a+MOfrT1AACeMy3z/SNburl74w5uvvqcaCoUqZI4bRI2\nA0N52gezpufTEr7OBX7f3W93938Efgd4gWCPdTx3AufnfL2tyLpFppOi+8pIKs2GTV2n7ZnmMmDD\npl2MpNJR1iky6eK0hzoANOZpb8qaXmg5gBHggUyju6fN7D7g42a21N1fKvSD3b0b6M5uCw7Fiki2\nUvrKE11HTjlmWnCdwIG+IZ7oOsKqzrlRlClSFXHaQ91LMOybK9O2p8Byhwn2Yg+5eypnWqbjt028\nPBEpRe9AadfGlzq/SNzEKVCfAlaY2eyc9suzpp/G3dPhtHlm1pAzOXPc9UBkVYpIUVqbc7vj2PYf\nHWJPzwCpdO7RVpHaEKdAfQBIAjdmGsyskeDkos2ZS2bMbKmZrcxZ9r5w2fdkLdsEvBt4wd0L7d2K\nSIVc2tFGe0vDuMdQAWY31XHm7EZe2HOUn7x4gOf39NLTrz1WqS2xOYbq7pvN7H7gU+Gp+dsJArID\neF/WrBsInm6R3U/vAX4f+IKZrSA4y3ctsAx4a+WrF5Fc9ckE61Z1cMfD28ad9/+6eDEzGuoYHk2T\nSjl7ewbZ2zPIjIYkC+c0s7C1iab65CRULVK+2ARqaB3wCYIwbCO4deBb3P0nYy3k7gPhPXtvA94L\nzCQYBv4dd/9hZUsWkUJuWt3JU7t7eGRLN8apl85kvl+zcj4ffctrSJpx8PgQe3sGOdg3hDv0D6fY\n0d3Hju4+zmhpYFFrM/NmNZJM6KRBiZ9YBaq7DwK3hl+F5lldoL0buL4ihYlIWeqTCe5Zewl3b9zB\nhk27ONB38sq49pZG1q1axvqsOyXNn9XE/FlNDI+m2dc7yJ7eAfoGRwE43DfM4b5hkkljwewmFrU2\n0zqjvirvSySfWAWqiEw99ckEN199DutXdxZ9L9+GugRL585g6dwZHBscYU/PIPuODjISDgm/cmSA\nV44MMKMxyaLWZhZoSFhiQIEqIpOiPpko6zrTWU31vHpBPefMb+Fg3xB7egc5lBkSHkqxvbuPHQf6\nOGNmA4vmNDOvpZGEhoSlChSoIlITEglj/uwm5s9uYmg0FQwJ9wxyfGgUdzjUN8yhvmHqksaC1iYW\ntjbT2qxrWrtiAAAgAElEQVQhYZk8ClQRqTmNdUmWzZ3Jsrkz6R0YYW/vAPt6BxlNOaMp5+XDA7x8\neICZjXUsmtPEgtYmGus0JCyVpUAVkZrW2lxPa3M9K+bP4kBfcHOIw8eHcYfjQ6O8uL+P7d19zG1p\nZFFrE+0aEpYKUaCKyJSQSBhnzm7izNlNDI6kTpwl3D+Uwh0OHhvi4LEh6usSLJjdxMI5Tcxu0pCw\nREeBKiJTTlN9ko72mXS0z6S3f4Q9vQPsOzpIKuWMjKbZfbif3Yf7aWmqO3GWcENdnG4cJ7VIgSoi\nU1rrjHpaZ9Sz4sxZHDg2xJ7eAQ6HT8HpGxxl2+AxXuw+RntLIwvnNNE+U0PCUh4FqohMC8lEcPbv\ngtZgSHhv7yB7ewboHw6GhA8cG+LAsSEa6hIsaG1i0ZxmWhr1ESnF01+LiEw7TfVJzm6fydntM+np\nH2ZPzyD7jwVDwsOjaV461M9Lh/qZ1VTHojnBkHChm1CIZChQRWRamzOjgTkzGnh1ehbdxwbZ0zPA\nkeMjABwbHGXrvmBIeF5LcCLT3JkNBR+qDjCSShd9RyiZWhSoIiIEQ8ILW5tZ2NrMwHAqOJGpd5CB\n4RTpNOw/Osj+o4M01idYGN44YmbWkPBIKs1dG3ewYVMXB/tOPnpuXksja1ct46asexbL1KRAFRHJ\n0dyQpHNeC69qn8mR/hH29Axw4NgQqbQzNJKm62A/XQf7aZ1Rz8LWJs6Y2cAHvvkLfrT1wGnPfz3Y\nN8QdD2/jqd093LP2EoXqFKZAFREpwMw4Y2YDZ8xsYDSVZv+xIfb2DNDTHwwJ9/aP0Ns/wvef2cOP\nth4ATn1EXfb3j2zp5u6NO7j56nMm7w1MY9UYelegiogUoS6ZYPGcZhbPaaZ/eJQ9PYPs7R3g+NAo\n/7Gle9zlDdiwadcpj6uT6FVz6F2/VRGREs1oqGP5/BauXN6OYRwLn9k6FgcO9A3xRNeRyhc4TY2k\n0ty44QnueHgbh7LCFE4Ovf/B13/OSCpdkZ+vQBURKZOZ4acN8o6td2B4/JmkLHdt3FH00HslKFBF\nRCagtbmhovNLcUZSaTZs6jrtpLBcmaH3SuylKlBFRCbg0o422lsaxv0gB5g7Mzg5RqL3RNcRDvYN\njzteUMmhdwWqiMgE1CcTrFvVUdTA71Ur5tE7MFLxmqajUofSKzH0rkAVEZmgm1Z3smblfIDT9lQz\n31+wuJXfOu9Mnnqph12Hjk9qfdNBHIbeFagiIhNUn0xwz9pLuOWaFbS3NJ4yrb2lkVuuWcFd/+PX\naKxLAvDi/j6e39NLOl3aCU1SWGbofTxGcAlNJYbedR2qiEgE6pMJbr76HNav7ix4Q4HZzfU8vbs3\neNpNT3Bbw9ee1XoiaKV89ckE115yFnf/+FdjzufAulXLKnItqvZQRUQiVJ9MsKpzLm86fyGrOuee\n8sE9q6mey85uo3VGPQA9/SP8bOcRjg3quOpEHRsc4dJlZ3DB4ta80zND72tWzmf96s6K1KBAFRGZ\nRI11SS5Z2saC1iYABkdSPNF1hO5jg1WurHYNjqR4ancPAO9/Qyfrf/NVzCsw9F7J+ylryFdEZJIl\nEsb5i1uZ1VTHi/v7SKWdZ3b30jk/xdntM6tdXk0ZTaV5ancPQyPBdaWvWdjKm85fyC1vfLXu5Ssi\nMl0smzuTGQ11PLenl1TK2dHdx/GhUc5dOJtkopgrW6c3d+fZV3rpC2/9eNYZzSydOwM4OfQ+mTTk\nKyJSRfNmNXJZxxk0NwQnJu3rHeTnu44wOJKqcmXxt3X/sRP37J3b0sCrz5xV1XoUqCIiVdbSWMdl\nHWfQNjM4WenowAg/6zqsm0CMYdeh47x8eACAlqY6Xru4FbPq7tUrUEVEYqChLsHFS9pYNKcZgKGR\nND/fdZj9R3WyUq7uo4O8uL8PgMb6BBctmUNdDB6JV/0KREQECE5Wes2i2bx6wSzMIJ2GZ1/uZceB\nPtx1EwgIHur+3J5eAJJJ46Ilc2iqj8d1vApUEZGYWXLGjHCvKxjC3HngOM++0ktqmt9ZaWA4xVMv\n95BOgxm8dnErs5rqq13WCQpUEZEYmtsSnKw0IzxZqfvoED/rOjxtT1YaSaV5cvcRRkaDy2NevWDW\nabd5rDYFqohITM1srOOys8/gjPAetX2Dozy+8zC9/dPrZKV02nnm5V76h4KNiWVzZ3BW24wqV3U6\nBaqISIzVJxNcvGQOS84IAmR4NM3PXzrM3t6BKlc2eX657yhHjgeXx8yf3cjy+S1Vrig/BaqISMyZ\nGa9eMIuVC0+erPT8K0fZ3n1syp+s9KsDfeztCc50bp1Rz3mLqn95TCEKVBGRGnFW2wx+bWnbiZOV\nug728/TLvYym0lWurDL29Q7yqwPBs2ObG5JccFZrrO8gpUAVEakhbTMbuPzsucxsDO4ce/DYED/r\nOsLA8NQ6WenI8WFe2BtcHlMXXh4T98fcKVBFRGpMc0OSyzraaJ8VnOV6fGiUx7sOnzjOWOuOD43y\ndHh5TCIBF54158QGRJwpUEVEalBdMsGFZ7XS0R6crDQymuYXLx3hlZ6TJyuNpNJs2nGIHzy3l007\nDjFSA0PDw6PB02NGU8Gx4XMXzqZtZkOVqypO/CNfRETyMjOWz5/FzMY6frn3KOk0/HLPUXr6h/n3\nF/bz9f/axcG+k3ut81oaWbtqGTet7qz4o8zKkUo7T7/cc2L4+ux5M1nY2lzlqoqnQBURqXELW5uZ\nUV/H0y/30D88yp/+0zM8+8rR0+Y72DfEHQ9v46ndPRV90HY53J0X9hw9cY3tgtYmOufF8/KYQuLz\nvykiImVrnVHP684+g//4ZXfeMAXIXGDzyJZu7t64Y/KKK8KOA30nHgTQNrOe1yycXeWKSqdAFRGZ\nIpIJ45Et+8edz4ANm3bF5pjqKz0DdB3sB2BGY5ILzppDIsaXxxSiQBURmSKe6DrCoePj35bQgQN9\nQzzRdaTyRY3jUN8QW/YGe9T14SPs4jQUXYrarFpERE7TO1DaZTOlzh+1Y4MjPPNKL+7B5TEXnTWH\n5oZ4X2s6FgWqiMgU0dpc2uUlpc4fpcGRFE/v7iUVXh5z/qJWWmfE51Fs5VCgiohMEZd2tNHe0kAx\nRx/bZzZwaUdbxWvKZzSV5undPSceRXfOmS3Mn91UlVqipEAVEZki6pMJ1q3qoJjb5b/p/AUkq3CT\neXfnuT1HOTY4CsDitmaWzZ056XVUggJVRGQKuWl1J2tWzgcouKd6weJWrjynnadf7iGVntyn1Wzb\n38fBY0MAzG1pYOWCWZP68yspVoFqZo1m9mkz22NmA2a22cyuKWM9D5uZm9nnK1GniEhc1ScT3LP2\nEm65ZgXtLY2nTJvX0siHrj6HP33zSuoSCQ71DfPkS0cm7fKZlw71s/twcHlMS1Mdr10c30exlSNu\nd0q6F7gW+BzwInA98JCZvcHdf1rMCszsHcCqShUoIhJ39ckEN199DutXd/JE1xF6B4ZpbQ6OmdYn\nE4ym0jzzSi+H+4bp6R/h57uOcNGSOTTVV+4M2+5jg2zbfwyAxvoEFy2ZQ12NXh5TSGzejZm9DrgO\n+DN3v9XdvwisAXYBtxW5jibgM8CnK1aoiEiNqE8mWNU5lzedv5BVnXNPXN9Zl0xw0VlzmD872IPt\nGxzl57sq9wi43oERng/v3pRMGBdWOLyrJTaBSrBnmgK+mGlw90Hgy8AqM1tSxDo+QvCebq9IhSIi\nU0QiYbx2cSuL24Kbzw8Mp/hZ12GODY5/Y4hSBJfHBMdqzeD8xa3Mbqrty2MKidOQ78XANnfPvQnl\n4+HrRcDuQgub2VLgT4H3uvtAKePyZjYfmJfT3Fn0CkSmCfWVqcXMOHfhbOqTCboOHmd4NH1i+HfO\njIlfozqSSvPkSz0MjwbHaFecOYt5sxrHWap2xSlQFwJ787Rn2haNs/xngCfd/dtl/Oz3Ax8rYzmR\n6UZ9ZQpaPr+FhmSCbfuPMZpynnyph9ee1XraSU2lSKedZ1/p5fhQcHnM0rkzWHLGjKhKjqU4BWoz\nMJSnfTBrel5m9gbg94DLy/zZdwL357R1Ag+WuT6RqUp9ZYpaOncGdUnjl3uPBs8l3d3DeYtaWdBa\n3g0Xtuw7xuHwWazzZjVyzvzaehRbOeIUqANAvs2hpqzppzGzOuDvgK+7+8/K+cHu3g1056y3nFWJ\nTGnqK1PbojnN1CWN517pJZ2G517pZSSVLnnPcufB4+zpCT6yZzfXc/4UuzymkDidlLSXYNg3V6Zt\nT4Hl1gGvBu4xs47MVzhtVvj91B5nEBGJyPxZTVy8pI1kMgjArfuOseNAX9HL7+sdZEd3MH9TfZIL\nl7SSrMFHsZUjToH6FLDCzHKfKnt51vR8lgL1wKPAzqwvCMJ2J/Bb0ZYqIjJ1tc1s4JJlbTTUBRGx\n88Bxtuw7ivvYd1Xq6R/mhb29ANQljYuWzqGxbupdHlNInIZ8HwD+GLiR8LIXM2sEbgA2u/vusG0p\nMMPdt4TLfZv8YfsvwEPAl4DNlS1dRGRqmd1Uz6UdbfxiV3AT+5cPDzCacl6zcDaJhDGSSp9y04jX\nLJrF0y8HQ8VmcMFZc2hpjFPEVF5Z79bM1gCfB16fe5mLmbUCjwHr3f0/i12nu282s/uBT4Wn5m8H\n3gN0AO/LmnUDcBXhbSrDYN1CjnC8fqe7f6f4dyYiIhkzGuq4tKONJ1/q4fjQKPt6BxkYSfHo9oN8\n4792cbDv5PNUW5vqWb1yHr99/gIuOGsOZ8ys3qPhqqXcId8PAV/Kc80o7t4L3AN8uIz1riO47eBa\nghON6oG3uPtPyqxTREQmoKk+yaUdbbTOqGc0neavvvc8n/v3F08JU4DewREefGoPX320a0pfazqW\ncgP1QuAHY0z/N+CSUlfq7oPhbQcXunuTu7/O3X+YM89qdx/3CLe7m7v/Yak1iIjIqeqTCS5eMoeN\nWw/w7Cun7UedYvPOw9y9ccckVRYv5QbqmcBY96ca5fS7qYiISI1y4OEX9o07nwEbNu2atCfYxEm5\ngfoKcP4Y0y8g/12PRESkBj3RdYRDx8e/z68DB/qGeKLrSOWLiplyA/Uh4BPh011OYWbNwMeB70+k\nMBERiY/egeHxZ5rA/FNBuec0/zXwDmBb+BDvrWH7SuADQBL45MTLExGROGhtLu2s3VLnnwrKClR3\n329mVwB3AZ8ivISFYG//h8AH3H1/NCWKiEi1XdrRRntLA4f6hhnr9g4GtLc0cmlH22SVFhtl3ynJ\n3Xe5+5uBdoK7Gb0eaHf3N7v7zrGXFhGRWlKfTLBuVceYYQrBXtW6VctOPMx8OpnwO3b3I+7+M3d/\n3N2n31FoEZFp4qbVnaxZOR84OSyZkfl+zcr5rF89PR+RW+6dkv65mPnc/R3lrF9EROKnPpngnrWX\ncPfGHWzYtIsDfSefuNne0si6VctYv7pzWu6dQvknJfVGWoWIiNSE+mSCm68+h/WrO0+5l++lHW3T\nNkgzyj0p6YaoCxERkdpRn0ywqnNutcuIlXKHfL9SxGzu7u8bfzYREZHaV+6Q7/XALuBJTj82LSIi\nMu2UG6h3Ae8Czga+CnzD3Q9HVpWIiEiNKesIsrt/AFgI3Aa8FdhtZv9oZm+08EGkIiIi08lEbuww\n5O7fcvdrgNcAzwN3Al1m1hJVgSIiIrUgqnOc0wQ3yDCC+/iKiIhMK2UHqpk1mtm7zOxhYBvwWuAP\ngaXu3hdVgSIiIrWg3Mtm7gSuA3YDXwHe5e4HoyxMRESklpR7lu964CXgV8BVwFX5zkXSrQdFRGS6\nKDdQN8C4Dx0QERGZNsq99eD1EdchIiJS06b3nYxFREQiokAVERGJgAJVREQkAgpUERGRCChQRURE\nIqBAFRERiYACVUREJAIKVBERkQgoUEVERCKgQBUREYmAAlVERCQCClQREZEIKFBFREQioEAVERGJ\ngAJVREQkAgpUERGRCChQRUREIqBAFRERiYACVUREJAIKVBERkQgoUEVERCKgQBUREYmAAlVERCQC\nClQREZEIKFBFREQioEAVERGJgAJVREQkAgpUERGRCChQRUREIqBAFRERiUCsAtXMGs3s02a2x8wG\nzGyzmV1TxHLvMLP7zOxXZtZvZlvN7DNmNmcy6hYREYlVoAL3Ah8Gvgl8EEgBD5nZleMs90XgXOAb\nwB8BPwD+ENhkZs0Vq1ZERCRUV+0CMszsdcB1wK3ufnvYtgF4DrgNuGKMxa9194056/s58DXg3cA/\nVKJmERGRjDjtoV5LsEf6xUyDuw8CXwZWmdmSQgvmhmnoX8LXcyOsUUREJK/Y7KECFwPb3P1oTvvj\n4etFwO4S1rcgfD043oxmNh+Yl9PcWcLPEpkW1FdECotToC4E9uZpz7QtKnF9f0Kwx/tAEfO+H/hY\niesXmY7UV0QKiFOgNgNDedoHs6YXxcz+O/A+4DZ3f7GIRe4E7s9p6wQeLPZnikwT6isiBcQpUAeA\nxjztTVnTx2Vmv0Fw3PWHwJ8Xs4y7dwPdOespZlGRaUV9RaSwOJ2UtJdg2DdXpm3PeCswswuB7xKc\nGXytu49GV56IiEhhcQrUp4AVZjY7p/3yrOkFmVknwfWn3cCb3b0v+hJFRETyi1OgPgAkgRszDWbW\nCNwAbHb33WHbUjNbmb2gmS0A/g1IA2909wOTVrWIiAgxOobq7pvN7H7gU+Gp+duB9wAdBCcYZWwA\nrgKyD9z8AHgVwQ0grsy5s9J+d3+4krWLiIjEJlBD64BPAGuBNuAZ4C3u/pNxlrswfP1Inmk/BhSo\nIiJSUbEK1PDOSLeGX4XmWZ2nTacZiohIVcXpGKqIiEjNUqCKiIhEQIEqIiISAQWqiIhIBGJ1UpJI\nNY2k0jzRdYTegWFamxu4tKON+qS2OUWkOApUqSmVCL2RVJq7Nu5gw6YuDvYNn2if19LI2lXLuGl1\np4JVRMalQJ2g6bJXU+33WanQG0mluXHDE/xo6wFyr7062DfEHQ9v46ndPdyz9pIp+XsVkegoUMs0\nXfZq4vA+Kxl6d23cwY+2Bneq9Jxpme8f2dLN3Rt3cPPV55RVf5SqvWEjIoUpUMswXfZq4vI+ow69\ndNoZSacZGE5x72M7i6rha491ceNvvorG+mQJlUcnDhs2IjI2BWoZam2vplxxeJ8jqTQbNnVheWrI\nde+jXfzuRYtOLDeS8lNeR1NBkKZSwZq27DvG4eMjRdVx8PgwX/rPnZy/eDYNyQQNdSe/GuuSNNQl\nqE8ajcnkifZkIpobeMVlw0ZExqZALVEpH/CZvZr6ZCLvvO6ntxZaZ55Z8Txz55tvLIXWO5JK87Ui\n997ufbSL37vkLJIJwx3S7qTdceeU79MevGcnbEsHr2S+D+f1cNm0w7Ov9J6yRzaWQ/3DfO/pvaxc\nMKuo+fuHiwvT7PlHU85oKkX/cGrc+ZNJozEnfLPDuNjwjcOGjYiMT4Faoie6jhT9AZ/Zqyn2Az5O\ntuw7xqEi994O9Q/zz794pSLv81DfYEnzZ0LSDOqTCeqSRkMyQV0y2IOsTyaC9oQxNJouad3L589i\nyRkzGB5NM5xKMTSaZng02PPNJ5Vy+icYvgkz7n10/A0bAzZs2sV6Df2KVI0CtUS9A8WFaUape0Fx\nUc7e23jMIGGGGZgZicz3ZH2fCF4tbF80Z0ZJdVx+dju/vnwudUWEyoLWJtpbGjjUNzzmaIMB7S2N\nvPXCRXnDKp12hlPpEwE7nApfT3xffvhu2XeMw/3j/986cKBviCe6jrCqc+6484tI9BSoJWptbihp\n/uXzZ7F8fstp7ZZnhM9OO0I29vzFzmcFFs7Xmpl1cGT8vapslyw7g8tfNTcIQ+xEeCZyQrRU5y9u\n5XP/vq3o0LuiyDCFYA923aoO7nh425jzObBu1bKCe36JhNGUSNJUxAlLpYZvqRs2pW7wiUh0FKgl\nurSjLZK9mrh74/mNJb3P31wxryLvM6rQK+Sm1Z08tbuHR7Z0n3ZcPPP9mpXzWb+6s8TK8ys1fOuT\nCe7c+Kui11/qBp+IRKf2PumrLPMBP965P+V+wMdFnN7nTas7WbNyPnD6XnXm+3JDrz6Z4J61l3DL\nNStob2k8ZVp7SyO3XLOiamfPJhLGFcvn0t7SMMbYRcAILqG5tKNtMkoTkTy0h1qGyd6rqZa4vM9M\n6N29cQcbNu3iQN/QiWntLY2sW7VsQifj1CcT3Hz1Oaxf3Rm7myZUeg9dRKKjQC1DpT/g4yJO73My\nQq8+mYjlCT1x2bARkbEpUMsU572aKMXtfcY19CopThs2IlKYAnWCpssH/HR5n3EVtw0bETmdAlWk\nhmjDRiS+tGkrIiISAQWqiIhIBBSoIiIiEVCgioiIREAnJRXWALB9+/Zq1yFStqy/30rek1B9RaaE\nifYXy/dMTgEz+13gwWrXIRKRt7n7dyuxYvUVmYLK6i8K1ALMrBW4CtgNRP0Ij06CD6C3ATsiXrdM\nrrj/LhuAJcCP3b23Ej9AfUWKVAu/ywn1Fw35FhD+Z1Zqiz7zzx3u/nwlfoZMjhr5XT5ZyZWrr0gx\nauh3WXZ/0UlJIiIiEVCgioiIRECBKiIiEgEFanUcAD4evkpt0++ysvT/O3VM+d+lzvIVERGJgPZQ\nRUREIqBAFRERiYACVUREJAIKVBERkQgoUEVERCKgQBUREYmAAlVERCQCClQREZEIKFBFREQioEAV\nERGJgAJVREQkAgpUERGRCChQRUREIqBAFRERiYACVUREJAIK1CKYWYeZuZndW+1aROJMfUWmMwXq\nBJnZe8zscTPrM7NeM9toZm8pctl2M9sbfgD9tNK1ilRTKX3FzP4y7BeFvt402fWLjKeu2gXUiFeA\nc4He7EYzux24BXgZ+BLQAFwHfM/Mbnb3z4+z3nuAlujLFamaqPvK14CuPO3boypYJCrm7tWuoSaZ\n2RXAo8AO4DJ3PxK2dwA/B2YCK929q8Dy6wg+LN4P3Ak86u5XVrxwkUlWTl8xs78EPga8wd03TmrB\nImXSkG8RChwXWh++fjLzAQEQfih8AWgEbiiwvqXA3wFfBv61EjWLVEPUfUWklihQy7cmfP1Bnmn/\nmjPPCWZmwL0EQ2IfrkhlIvFSVl8JXWlmf2xmf2Jm7zSz9ujLE4mGjqGWwcxmAouBPnffm2eWF8PX\nFXmmfQhYDfyWux81szMqU6VI9U2wrwB8Iuf7ITP7G+AvXMerJGa0h1qe1vC1t8D0TPuc7EYzew3w\nv4C73f3fK1SbSJyU1VeAp4H3Aq8CmoFlwP8EeoCPAp+MtkyRidMe6iQxs3rg68Be4CNVLkck1tz9\nX3KaXgL+wcx+AfwX8Mdmdoe7H5z86kTy0x5qeTJb1a0Fpmfae7La/gy4GLjB3fsqVZhIzJTTVwpy\n918AjwP1wKqJlSYSLQVqGdz9OMH1di1mtjDPLOeEr9uy2n4NMGBj9gXqwM5w+q+HbUV9sIjUgjL7\nyngOhK8zJ1KbSNQ05Fu+R4C1wJuAr+ZM++2seTIeBvINT7UA7wT2A98H+qMtU6TqSu0rBYWHTn4t\n/PZXkVQnEhHd2KEI4QXoO4Gvufv1YduEbuyQZ926sYPUvCj6ipnNAha5+9acdTcAnyW4GcoW4Dx3\nT1f4LYkUTXuoZXL3x8zsDoJrSZ8xswcIbqf2TuAM4ObxwlRkOiijr8wFfmlmTwC/JDiRbx7wBuBs\ngpGedylMJW4UqBPg7reY2bPAB4AbgTTwC+Bv3P37VS1OJEZK7CuHgc8DrwPeSBC6wwR7uJ8G7nD3\n7smqXaRYGvIVERGJgM7yFRERiYACVUREJAIKVBERkQgoUEVERCKgQBUREYmAAlVERCQCClQREZEI\n6MYOBZhZK3AVsJvgonKRWtQALAF+7O6Fnkk6IeorMoVMqL8oUAu7Cniw2kWIRORtwHcrtG71FZlq\nyuovCtTCdgN85zvfYfny5dWuRaQs27dv5+1vfzuEf88Vor4iU8JE+0usAtXMGoG/InjUUxvwDPBR\nd3+4yOXfCXwIuAAYAV4Ily/q0VA5hgGWL1/OeeedV8biIrFSyaFY9RWZasrqL3E7KelegidSfBP4\nIJACHjKzcR9rZmZ/CXyLYMviw8BHCQJ5cYVqFREROSE2e6hm9jrgOuBWd789bNsAPAfcBlwxxrKv\nB/4CuMXdPzsJ5YqIiJwiTnuo1xLskX4x0+Dug8CXgVVmtmSMZT8E7AP+1gItFa1UREQkR2z2UIGL\ngW3ufjSn/fHw9SIKHyi+GngM+COCod65ZrYP+KS7f368H2xm8wkeYJyts9jCRaYL9RWRwuIUqAuB\nvXnaM22L8i1kZm1AO/DrwBrg48BLwA3A35vZiLvfM87Pfj/wsXKKFplm1FdECohToDYDQ3naB7Om\n55MZ3p0LXOfu9wGY2QPAswR7rOMF6p3A/TltnejaOpFc6isiBcQpUAeAxjztTVnTCy0HwWUyD2Qa\n3T1tZvcBHzezpe7+UqEf7O7dQHd2m5kVW7fItKG+IlJYnE5K2ksw7Jsr07anwHKHCfZiD7l7Kmda\npuO3Tbw8ERGRwuIUqE8BK8xsdk775VnTT+Pu6XDaPDNryJmcOe56ILIqRURE8ohToD4AJIEbMw3h\nnZNuADa7++6wbamZrcxZ9r5w2fdkLdsEvBt4wd0L7d2KiIhEIjbHUN19s5ndD3wqPDV/O0FAdgDv\ny5p1A8HNuLMP3NwD/D7wBTNbQXCW71pgGfDWylcvIiLTXWwCNbQO+ASn3sv3Le7+k7EWcvcBM1tD\ncEel9wIzCYaBf8fdf1jZkkVERGIWqOGdkW4NvwrNs7pAezdwfUUKExERGUecjqGKiIjULAWqiIhI\nBBSoIiIiEVCgioiIRECBKiIiEgEFqoiISAQUqCIiIhFQoIqIiERAgSoiIhIBBaqIiEgEFKgiIiIR\nUKCKiIhEQIEqIiISAQWqiIhIBBSoIiIiEVCgioiIRECBKiIiEgEFqoiISAQUqCIiIhH4/9u7+2i5\nqpDf1VQAABLwSURBVPKO49+HEEJsBJGABBSiEQQVCcgSQrFEKIJWpUV0gRYUcQHia1W0ausbIguq\nLkstKIhihErBCqhl8aIQaTULjYCICpFIQpYgSZAXU5IQkqd/nDMymcx9ydw999658/2sNWvm7vO2\nc0/2/ObsOXtfA1WSpAIMVEmSCjBQJUkqwECVJKkAA1WSpAIMVEmSCjBQJUkqwECVJKkAA1WSpAIM\nVEmSCjBQJUkqwECVJKkAA1WSpAIMVEmSCjBQJUkqwECVJKkAA1WSpAIMVEmSCjBQJUkqwECVJKkA\nA1WSpAIMVEmSCjBQJUkqwECVJKkAA1WSpAIMVEmSCjBQJUkqwECVJKkAA1WSpAIMVEmSCjBQJUkq\nwECVJKmAcRWoETElIs6OiPsjYnVE3BIRh3ewnxsiIiPiS92opyRJrcZVoAIXA+8HLgXeC6wHromI\ng4e7g4g4GpjTldpJkjSAcROoEfEy4FjgI5l5emZeABwKLAXOGeY+tgY+D5zdtYpKktTGuAlU4Biq\nK9ILGgWZuQa4CJgTEc8Zxj4+RPVv+lxXaihJ0gC2HOsKNNkXWJSZj7WU/7R+ng0sG2jjiNgV+Efg\nbZm5OiKGfeCI2BHYoaV41rB3IPUJ24o0sPEUqDOAB9qUN8p2HmL7zwO3ZeZlHRz7NOATHWwn9Rvb\nijSA8RSoU4G1bcrXNC1vKyJeAbweOKDDY58HXNFSNgu4usP9SROVbUUawHgK1NXAlDblWzct30RE\nbAmcC3wzM3/WyYEzczmwvGW/nexKmtBsK9LAxlOgPgDs0qZ8Rv18/wDbnQC8ADglIma2LHt6XbY8\nMx8feRUlSWpvPN3lezuwR0Rs01J+QNPydnYFJgM/Bu5tekAVtvcCryxbVUmSNjaerlC/DXwQOJl6\n2EtETAFOBG7JzGV12a7A0zLzrnq7y2gftlcC1wAXArd0t+qSpH7XUaBGxKHAl4ADW4e5RMS2wE+A\nUzPzf4a7z8y8JSKuAM6qb82/B3gLMBM4qWnVecAhQNTb3QXcRYv6e517M/Oq4f/LJEnqTKddvu8D\nLmwzZpTMfBT4CtUUgpvrBOCLwPFUNxpNBl6TmTd3WE9JkkZFp12++wAfHmT59VTdt5ulnhnp9Pox\n0Dpzh7kvbz2UJI2aTq9QnwWsG2T5k2w6m4okSRNWp4H6e+DFgyx/Ce1nPZIkaULqNFCvAc6o/7rL\nRiJiKvAp4PsjqZgkSb2k0+9QPwMcDSyq/4j33XX5nsA7gUnAmSOvniRJvaGjQM3MByPiIOB84Czq\nISxAAtcB78zMB8tUUZKk8a/jiR0ycynw6ojYDng+Vaj+NjMfLlU5SZJ6xYhnSqoDtKNJ6SVJmig6\nnSnpO8NZLzOP7mT/kiT1mk6vUB8tWgtJknpcpzclnVi6IpIk9bJOu3y/NozVMjNPGno1SZJ6X6dd\nvm8FlgK38dSQGUmS+langXo+cBzwXODrwCWZ+cditZIkqcd0NPVgZr4TmAGcA7wWWBYRl0fEEVH/\nIVJJkvpJp3P5kplrM/NbmXk48ELgV8B5wJKImFaqgpIk9YKOA7XFBqppB4NqHl9JkvpKx4EaEVMi\n4riIuAFYBOwNvAvYNTNXlaqgJEm9oNNhM+cBxwLLgK8Bx2XmypIVkySpl3R6l++pwH3A74BDgEPa\n3Yvk1IOSpH7RaaDOo/rOVJIk0fnUg28tXA9Jknpaqbt8JUnqawaqJEkFGKiSJBVgoEqSVICBKklS\nAQaqJEkFGKiSJBVgoEqSVICBKklSAQaqJEkFGKiSJBVgoEqSVICBKklSAQaqJEkFGKiSJBVgoEqS\nVICBKklSAQaqJEkFGKiSJBVgoEqSVICBKklSAQaqJEkFGKiSJBVgoEqSVICBKklSAQaqJEkFGKiS\nJBVgoEqSVICBKklSAQaqJEkFjKtAjYgpEXF2RNwfEasj4paIOHwY2x0dEf8ZEb+LiMcj4u6I+HxE\nPGM06i1J0rgKVOBi4P3ApcB7gfXANRFx8BDbXQDsBVwCvAe4FngXsCAipnattpIk1bYc6wo0RMTL\ngGOB0zPzc3XZPOBO4BzgoEE2PyYz57fs7+fAN4A3A1/tRp0lSWoYT1eox1BdkV7QKMjMNcBFwJyI\neM5AG7aGae3K+nmvgnWUJKmtcXOFCuwLLMrMx1rKf1o/zwaWbcb+dqqfVw61YkTsCOzQUjxrM44l\n9QXbijSw8RSoM4AH2pQ3ynbezP19mOqK99vDWPc04BObuX9NMOvWb2Dhkod5dPUTbDt1K/afuR2T\nJ42nTpxxwbYiDWA8BepUYG2b8jVNy4clIt4EnASck5m/HcYm5wFXtJTNAq4e7jHVu9at38D58xcz\nb8ESVq564s/lO0ybwvFzduMdc2cZrE+xrUgDGE+BuhqY0qZ866blQ4qIl1N973od8LHhbJOZy4Hl\nLfsZzqbqcevWb+DkeQu56e4VtJ7xlavW8oUbFnH7skf4yvEvNVSxrUiDGU/vEA9Qdfu2apTdP9QO\nImIf4LtUdwYfk5lPlqueJqLz5y/mprtXAJAtyxo/33jXcr48f/Go1ktS7xlPgXo7sEdEbNNSfkDT\n8gFFxCyq8afLgVdn5qryVdREsm79BuYtWLLJlWmrAOYtWMq69RtGoVaSetV4CtRvA5OAkxsFETEF\nOBG4JTOX1WW7RsSezRtGxE7A9cAG4IjMXDFqtVbPWrjkYVauemKTK9NWCaxYtZaFSx4ejWpJ6lHj\n5jvUzLwlIq4Azqpvzb8HeAswk+oGo4Z5wCGw0YXFtcDzqCaAOLhlZqUHM/OGbtZdvenR1U8MvdII\n1pfUX8ZNoNZOAM4Ajge2A+4AXpOZNw+x3T7184faLPsRYKBqE9tO3aqr60vqL+MqUOuZkU6vHwOt\nM7dNmbcZdtlEHKO5/8ztmD5tKx4aots3gOnTprD/zO1Gq2qSetC4ClSNPxN5jObkSVtwwpyZfOGG\nRYOul8AJc3br2X+npNFhoGpA/TBG8x1zZ3H7ske48a7lBBsPnWn8fOieO3LqXGfXk5pNxF6rkTJQ\nNaDNGaP57sN2H9W6lTJ50hZ85fiX8uX5i5m3YCkrVj01Wdf0aVM4Yc5unNrDV+FSaRO512qkDFS1\n1TxGc6jvF+ctWNrToTN50ha8+7DdOXXuLD9xS4Poh16rkei/f7GGpR/HaE6etAVzZm3PkS+ewZxZ\n2/flG4I0GGcWG5zvGGrLMZqSmjmz2NAMVLXlGE1Jzfqx12pzGahqqzFGczifRndwjKY04dlrNTQD\nVW01xmgO59OoYzSlic9eq6H5LqgBvWPuLA7dc0eATa5UGz87RlPa2Lr1G1iw+CGuvfMBFix+aMJ8\nl2iv1dAcNqMBOUZTGr6JPj7TmcWGZqBqUI7RlIbWL+MznVlscL17ZjWqHKMpDaxfxmc2eq0+cPge\nTJ82ZaNl06dN4QOH79HzHxpGwitUSRqBfppVDOy1GoyBKkkj0BifOZTm8ZlzZm3f/Yp1WaPXSk/p\n748TkjRCjs9Ug4EqSSPg+Ew1GKiSNAKOz1SDgSpJI+CsYmrwzErSCDmrmMBAlaQRc3ymwGEzklSE\n4zNloEpSQY7P7F9+bJIkqQADVZKkAuzyHdhWAPfcc89Y10PqWNP/327OJmBb0YQw0vYSmUONnupP\nEfE64OqxrodUyFGZ+d1u7Ni2ogmoo/ZioA4gIrYFDgGWAaUn35xF9QZ0FNDbf89J4/1cbgU8B/hR\nZj7ajQPYVjRMvXAuR9Re7PIdQP3L7NYn+sbLxZn5q24cQ6OjR87lbd3cuW1Fw9FD57Lj9uJNSZIk\nFWCgSpJUgIEqSVIBBurYWAF8qn5Wb/Ncdpe/34ljwp9L7/KVJKkAr1AlSSrAQJUkqQADVZKkAgxU\nSZIKMFAlSSrAQB0FEfGxiMiIuLPNsoMi4n8j4vGI+ENEnBsR08ainhpYROwXEd+NiD/W5+rOiHhP\nyzqeyxGyrfS+fm4rzuXbZRHxbOCjwP+1WTYb+CHwG+D9wLOBDwK7A68axWpqEBHxSuB7VHN8ngGs\nopro+9lN63guR8i20vv6va04DrXLIuIyYAdgEjA9M1/ctOwaYDawZ2Y+Vpe9HbgQOCIzrx+DKqtJ\nRGwDLAJ+AhyTmRsGWM9zOUK2ld5mW7HLt6si4q+AY4D3tVm2DXA4cEnjP1VtHtWnujeOSiU1lDcB\nzwI+lpkbIuIvImKjduO5HDnbyoTQ923FQO2SiJgE/Bvw1cz8ZZtV9qbqcl/YXJiZTwC3A/t2vZIa\njr8GHgN2iYi7qRr9YxFxfkRsXa/juRwB28qE0fdtxUDtnlOB3YB/HmD5jPr5gTbLHgB27kaltNl2\np3oDuBq4Dng98DWq8/v1eh3P5cjYViaGvm8r3pTUBRGxPfBp4IzMHGgi6Kn189o2y9Y0LdfYmgY8\nDfhyZjbuVPxORGwFnBIRH8dz2THbyoTS923FK9Tu+AzwR6purIGsrp+ntFm2ddNyja3GefhWS/l/\n1M9z8FyOhG1l4uj7tuIVamERsTtwMtXNFTtHRGPR1sDkiJhJ9T1Do8tjBpuaAdzf1YpquO4HXgQ8\n2FK+vH7eDlhcv/ZcbgbbyoTT923FK9TydqH6vZ4L3Nv0OADYo379ceBO4Elg/+aN6+6R2VRf0Gvs\n/bx+3qWlvPFdzwo8l52yrUwsfd9WDNTy7gT+rs3jV8B99euLMvNR4AfA30fE05u2P57qu4grRrPS\nGtDl9fNJLeVvp3pjmO+57JhtZWLp+7bixA6jJCLms+lg9f2oBkH/GriAasaQDwA3Z+YRY1FPbSoi\nLgLeRvWG8SNgLvAG4KzM/Gi9jueyENtK7+r3tmKgjpJ2bxJ1+cHA2cB+wJ+o/iN+JDP/NOqVVFsR\nMZlqSrwTqbqvlgL/nplfbFnPc1mAbaV39XtbMVAlSSrA71AlSSrAQJUkqQADVZKkAgxUSZIKMFAl\nSSrAQJUkqQADVZKkAgxUSZIKMFAlSSrAQJUkqQADVaMmImZGREbE7A62PSwifhMRk7pRt/oYR0bE\n7RFhu9CYsq30Jn8ZfSIiLq4baEbEuoh4MCJuiIi3daNR1Me7quAuzwE+k5nrC+5zI5l5LbAOeHO3\njqHxz7YyNNtKewZqf7kWmAHMBF4F3AT8K/D9iNhyDOs1qPovU8wC/msUDncx8J5ROI7GN9vK0C7G\ntrIRA7W/rM3MP2Tm7zPz1sz8LHAU1RvGWxsrRcQzIuKrEbEiIh6LiBsjYp+m5Z+su3tOiYhlEfF4\nRFweEds2lgNvAY5q+qQ/t6kez4uIm+rtfhERc4ao97HADZm5prkwIl4bET+LiDURsTIirmxatiQi\n/iki5kXEqohYGhGvi4gdIuLquuyOiNi/5VjfA/aPiFnD/J1qYrKt2FY2m4Ha5zLzRuAXwNFNxVcA\nO1K9ebwUuBX4YUQ8s2md5wNvBF4LHAnsC5xXL/sc1d83bHzKn0H1B4UbzqzXmQ0sAr41xKf+lwML\nmwsi4m+AK4Fr6mMfBvy0Zbt/AH5cL/9v4JvAPOASqr/DuBiYFxHR9Pu4D3iwPqb0Z7YV28qQMtNH\nHzyoumeuGmDZZcCv69cHA48CU1rWuQc4uX79SeBJYJem5UcC64GdBjoeVfdZAic1lb2wLttzkLo/\nAhzfUvYT4JJBtlkCfLPp553q43y6qezAumynlm1vBT4x1ufMx9g8bCu2lU4fXqEKIKgaC8A+wDTg\nobqrZ1VErAKeS/XdTMN9mfn7pp8XUPV4vGAYx7uj6fUD9fOOg6w/FVjTUjYb+OFmHOfB+vmXbcpa\nj70aeNoQ+1Z/sq1szLbSZNx+ua5RtRdwb/16GlXDndtmvUcKHW9d0+vGm9NgH+5WAtu1lK3enONk\nZta9VcM59jOBFcPYv/qPbWVjtpUmXqH2uYg4FNibp+4KvJWqy+fJzLyn5bGyadNdI2Lnpp8PBDYA\nd9c/PwGUGgd3G1V3V7M7qL4LKioitqa6urit9L7V22wrG7OtbMpA7S9TImKniNglIvaLiI8CVwPf\np7oBAeAHVF1SV0XEK6MaYH5QRJzZcpffGuAbEbFPRLwcOBe4PDP/UC9fArwkIl4QEdMjYvII6n0d\n1fdVzT4FHBcRn4qIvSJi74j48AiO0XAgsJbqd6D+ZVsZmm2lhYHaX46k6qJaQnVX4SuoxpEdlfUg\n8KzuNHg1cDPwdao7Cy8DduOp71GguvHiO1R3Dl5P9Sn4tKblF1J9Al9I1SX0lyOo96XAiyLiz985\nZeZ84A3A64DbgRuBl43gGA3HAZdm5uMF9qXeZVsZmm2lRdR3aknDVo+d+9vM3Oxp0UZwzH8BtsnM\nU7p4jOlUb2z7Z+a9Q60vDcW20l+8QlWvOBNYGt2dO3QmcJpvEOpxtpUx4l2+6gmZ+Qjw2S4fYyEt\ng+KlXmNbGTt2+UqSVIBdvpIkFWCgSpJUgIEqSVIBBqokSQUYqJIkFWCgSpJUgIEqSVIBBqokSQUY\nqJIkFWCgSpJUwP8Do8LfeE1gzJQAAAAASUVORK5CYII=\n", "text/plain": [ "