Classification_Methods_Final.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"Classification_Methods_Final.ipynb","provenance":[],"collapsed_sections":[]},"kernelspec":{"display_name":"Python [conda env:tensorflow]","language":"python","name":"conda-env-tensorflow-py"}},"cells":[{"cell_type":"markdown","metadata":{"id":"LLs1IHDvQjz1","colab_type":"text"},"source":["# Classification Methods\n","**Using Classification Methods to predict essay grades using vectorized essays** "]},{"cell_type":"code","metadata":{"id":"A_px2GhwQjz2","colab_type":"code","colab":{}},"source":["%%capture\n","from essay2vec import *"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"vbV2jPbWQjz6","colab_type":"code","colab":{}},"source":["%%capture\n","from sklearn.linear_model import LogisticRegression\n","from sklearn.discriminant_analysis import LinearDiscriminantAnalysis as LDA\n","from sklearn.discriminant_analysis import QuadraticDiscriminantAnalysis as QDA\n","from sklearn.neighbors import KNeighborsClassifier as KNN\n","from sklearn.tree import DecisionTreeClassifier as DecisionTree\n","from sklearn.ensemble import RandomForestClassifier as RandomForest\n","from sklearn import grid_search\n","from sklearn.svm import SVC\n","from time import time\n","import warnings\n","warnings.filterwarnings(\"ignore\")"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"uT7HREyMQjz9","colab_type":"code","colab":{}},"source":["#initialize arrays to store results\n","times = []\n","scores = []\n","labels = []"],"execution_count":0,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"-xHrxDVqQjz_","colab_type":"text"},"source":["### Logistic Regression"]},{"cell_type":"code","metadata":{"id":"JoUWYeusQj0B","colab_type":"code","outputId":"3a19e451-1160-42d6-af43-880edde904f7","colab":{}},"source":["#parameter combinations that will be tested\n","parameters = {'class_weight': ['balanced', None], 'penalty': ['l1','l2']}\n","logreg = LogisticRegression()\n","log_grid = grid_search.GridSearchCV(logreg, parameters)\n","log_grid.fit(trainDataVecs, y_train[:,0])\n","print (log_grid.best_params_)"],"execution_count":0,"outputs":[{"output_type":"stream","text":["{'penalty': 'l1', 'class_weight': None}\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"id":"wjw93kbyQj0I","colab_type":"text"},"source":["Utilizing the calculated parameters, we will conduct our analysis."]},{"cell_type":"code","metadata":{"id":"Umc-DO3wQj0J","colab_type":"code","colab":{}},"source":["start_lr = time()\n","logreg = LogisticRegression(penalty = 'l1', class_weight = None)\n","logreg.fit(trainDataVecs, y_train[:,0])\n","preds_lr  = logreg.predict(testDataVecs)\n","tot_lr = time() - start_lr\n","\n","scores.append(spearmanr(y_test, preds_lr)[0])\n","times.append(tot_lr)\n","labels.append('Log Reg')"],"execution_count":0,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"qJ7HpXOKQj0L","colab_type":"text"},"source":["Utilizing the same approach above, we will tune the other three classification methods. "]},{"cell_type":"code","metadata":{"id":"4_dbkr9SQj0M","colab_type":"code","colab":{}},"source":["#K Nearest Neighbors\n","start_knn = time()\n","knn = KNN(n_neighbors = 16)\n","knn.fit(trainDataVecs, y_train[:,0])\n","predsknn = knn.predict(testDataVecs)\n","tot_knn = time() - start_knn\n","\n","scores.append(spearmanr(y_test, predsknn)[0])\n","times.append(tot_knn)\n","labels.append('KNN')"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"NamKfrqrQj0O","colab_type":"code","colab":{}},"source":["#Support Vector Machining\n","start = time()\n","svm = SVC(C = 50, gamma = 0.9)\n","svm.fit(trainDataVecs, y_train[:,0])\n","predictions = svm.predict(testDataVecs)\n","total = time() - start\n","\n","scores.append(spearmanr(y_test, predictions)[0])\n","times.append(total)\n","labels.append('SVM')"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"amrw3o6EQj0R","colab_type":"code","colab":{}},"source":["#Random Forest\n","start_rf = time()\n","rf = RandomForest(n_estimators = 51, max_depth = 14)\n","rf.fit(trainDataVecs, y_train[:,0])\n","preds_rf = rf.predict(testDataVecs)\n","tot_rf = time() - start_rf \n","\n","scores.append(spearmanr(y_test, preds_rf)[0])\n","times.append(tot_rf)\n","labels.append('Random Forest')"],"execution_count":0,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"ZhpD5RTNQj0Z","colab_type":"text"},"source":["### Comparison\n"]},{"cell_type":"code","metadata":{"id":"TmzUY_4cQj0a","colab_type":"code","outputId":"deb2116b-e4cb-494d-f46b-4f04e0de4bc3","colab":{}},"source":["import matplotlib.cm as cm\n","#plot\n","fig = plt.figure(figsize=(7,7))\n","\n","color=iter(cm.rainbow(np.linspace(0,1,4)))\n","for label, x, y, col  in zip(labels, scores, times, color):\n","    plt.scatter(x, y, color=col, s = 50)\n","    plt.grid(color='k')\n","    plt.annotate(\n","    label,\n","    xy = (x, y), xytext = (40, 20),\n","    textcoords = 'offset points', ha = 'center', va = 'center',\n","    bbox = dict(boxstyle = 'round,pad=0.5', fc = 'white', alpha = 0.5),\n","    arrowprops = dict(arrowstyle = '->', connectionstyle = 'arc3,rad=0'))  \n","plt.title('Computation Time Vs. Spearmans Correlation')\n","plt.ylabel('Computation Time (Seconds)')\n","plt.xlabel('Spearmans Correlation')\n","plt.show()"],"execution_count":0,"outputs":[{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAm4AAAJvCAYAAADC9F22AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzs3Xu8VHW9//HXZ6Ns2V5AQ800vOEFjFLRlC7eDwYdp7yk\nlqaA9kvTMiw4WXkgO3YOdPISlp5zQkSq7aUMtat3FC0tKC8g3nVrGrglAd1cBD6/P75rO/OdPRv2\nnj0za23m/Xw85gF7zZo1n/We2TOfvdb6rmXujoiIiIhkX0PaBYiIiIhI16hxExEREekl1LiJiIiI\n9BJq3ERERER6CTVuIiIiIr2EGjcRERGRXkKNm4iIiEgvocZNREREpJdQ4yYiIiLSS6hxE6kBMxtj\nZuvNbFDatXTGzK4zsxfSrkNkU2BmL5rZtRVepn5HRY2bdI2Z7WFm/2Nmz5nZSjNbZmZzzewrZrZF\n2vVVgpmNMrNJPVzGRWb2qRJ3eXKrGTOblDSLG7vdU1Dj+lrW2Bkz+2FS2x4bmOfSZJ79qlTDcWZ2\nn5ktNrO3k/f+jWZ2bDWeTzbMzPY3s5+aWYuZrTKzN8zszuSPoix+l5X1+25mOyW/ux/sZJmZ+B2V\n9JiuVSobY2ajgZuBVcD1wBNAX+BjwInAde5+TnoVVoaZTQO+5O59erCMFcDN7j6uaLoBm7v7mh6W\n2Z1aPgAUfvhvBVwD3JLc2i1297vNrA/Q4O7v1KrGzpjZh4E/Af/u7v/RyTzPASvcff8qPP/XganA\nfcCtQBswGDgGeLT49ZXqMrOzgauBfwCzgGeArYGjgH8Fvu3u/5VehR0lW8bu7e57xcyGA38Gxrj7\n9UX3ZeZ3VNKzWdoFSLaZ2W7ADcALwFHuvqTg7qvN7GLgkymUVg1WrQV7+AupZk1b8pxPEJpsAMzs\nPYTG7TF3/3mJ+dcB62pXYefc/REzexb4HNChcTOzEcDuwMRKP3fy5fht4A/uPqrE/QMr/ZyVYmZN\n7t6Wdh2VZGaHEpq2B4HRRev3QzM7EPhABZ7HgL7uvrrEfbVsmDr9HMrS76ikyN11063TG+EDcx1w\nSBfn7wNcDDxL2EL3AuGLt2/RfC8CtwGHE/66bAMeAw5P7j8h+Xkl8Bdg/6LHXwesIHx5/wF4C/g7\ncHHRfIcTdi0cVjR912T6GcnPM5Kf1yX/rgfWFcz/dcIXR2tS61+AE4uWWfz49cC1yX1jkp8HFT3m\nS4TmalVS/1VA/6J57kuyGALcC7wNvAJM6OZr+Z6khn/v5P7rgBdKZHRhUudzSc5/AHZO5rkYeDnJ\nZDYwoMRyRwH3J49dDvwaGNqFeiclee5f4r5pwFpgl4Jp/wI8APwzeW8sAi4t4z2/Y7LeF3dh3vb3\n18nA94DXkvW8tbC2gvkPAX4PvJm8jvcBHymaZxDw46T+tuQ9dxOwa9F8Z7a/t5P5FwNvJPdNTu7b\nC/hp8nxLgEuS+9+fvF7LkpovLFr25sAlhPf5m8k63Q8c0cnv0YXAF8j/3j8CHFQi1xnJ+2UV8GpS\nw6CNZPw7YHWpPDuZvwn4AdCSPM8i4Gsl5lsP/JDwx8ETyXPkitbpgmSd3gE+mDyuL/Adwla/Vcnz\nTKHjZ9wLJL//yc/bAv9N+F1ekWT/2/blFr2fos8h8p9T11HwO1rm+n4KeDyZ9wng2O7+juiW7i31\nAnTL9i35kH2mG/Nfl3xA3ACcQ74h+mXRfC8ATxIakIuBryTPtSz5IH2B0CxNIHwRP1X0+BmEL76n\nkuc8l/BluR6YXDDf4ckH38Yat0MIDck64LNJDZ8rmL+F0Cycm3yY/zGZd1TBPJ8jNJr3tT+epOEl\nfMmuo+BLivyX6+8JjdGVyRfEn4A+BfPdm+T0InAZ8EXgzmR5Xf7QZeON2wzg+RIZzU8+6C8gfGGt\nAh4iNOQPAOcBlyf1/KRomZ9Ppv8mWcevExrAN9j4F/bg5PmnFk1vIOwyu6dg2tCkrj8B5xOaiCmE\nXVXdfc9b8t56BNh2I/O2f9E+Cvw1yehSQsP1JNBYMO9RSY1zga8S3vN/TaYdVDDfiUnmk4CzgO8m\neT0PbFEwX3vj9gRwT5LvhOS+SQWv3U+T98xtyWvx1aS2qwi/o/cn0z9W9F55Bfg+8P+ArwELk1oL\nG43298g8wu/i15N5lwAvEb+PHwSWEt73Y4F/A+4qfN4S+fYjNFR3duP1u5vQ1P8P4fd1dlLjD4rm\nWw8sIDSu306y+GDBOj1BaM4mJK/VLsl74w+Exuu/gbMJv7drgFtKfMYVNm7DgaeT98fZwLcInytL\ngfcm8+yQ1LKe8Edz++fIbqV+R8tY378mr+s3gS8n67cC2K67vye6pXdLvQDdsnsjHEOyvvgDaQPz\nfzCZ/5qi6VOTL4bDC6a9QNGWPMIWk/WEv+4Lt6R8gaLmK/kAWwdcXvRctxOap+2Sn7vUuCXTplGw\nla1o/sain/sQ/nK+s2j6isIP64LpUeMGDCR8Cf62aL4vJfOdWTDt3mRaYSO5OeEL56ZuvJ7lNm7/\nALYqmH4p+aagoWD6z5LsN09+3jL5Urq66Hm2JzTj13Sh5oeBlqJpxybPf1bBtAuSjDbYaHUjq8nJ\n8lYQms6LgANKzNfeuLUATQXTT0qmn18w7SngN8XvK0Ij+/vO3mvJtA8nyzut6D21nvCHghXN3964\n/bhgWkNS51oKtsgA/QmNamGTYcBmRcvcJnnP/V+J98gSYJuC6ccl+Y0ueI71FG3Z68LrMCx53GVd\nnP9TyfzfKJp+Y7LeuxdMW0/4Q2mfonnb1+mfFDU0wOnJY0YUTf9/yfoeWjCtuHHbvES9g5LfmW8V\nTBtO0WfTBn5Hu7u+K0mawKJ8v1SJ3xvdanPL4kgcyY5tkn9XdHH+0YRRT5cXTf8B4Yug+Fi4he7+\ncMHP7f+/291fKZpuQKkRhj8q+vkqwpfhMV2suUu84LgXMxtA2O3xAHBgmYs8htB8XVE0/f8IeRdn\n9bYXHJfm4VibhymdSaXd5O5vFfzc/jrNcvf1RdP7AjsnP48kfGHfYGbvab8R3iMPA0d24bl/Cuxs\nZocVTPscYSvMLwumvZn8e3xyrFKPuPvk5HnmE9bjP4B5ZjbPzPYt8ZCZXnDslbv/gtDkjAYwswMI\nuy2bi7LYmrDF5LCCxxa+1zYzs+0IW9v+Scf3mxMaKS+1GsD0guWuJ+z6NEID0D59GaGp3KNgmrv7\n2qQGM7NtCa/tX0rUAHCDuy8v+PkB4t/ZlYStUkckvz9d1d3PoFGEhmVa0fTLCI1r8TGL97n7U50s\n6xfuvrRo2kmErZVPF72O9xLWt9P3tBccH2dmDcnr2kbIvtzPke6u753u/mJBTY8TDl+oxeeIVIga\nN9mQ9g/irbs4f/tfqs8WTnT3xYQv1l2L5m8pmq/9+V4pmm9Z8u+2RdPXE77QCj1dUEvFmNm/mtkf\nzWwlYSvSEsJuif5lLrK9vqcLJyYf7s/Tsf6XSyzjn3TMpBqKn7v99djY6zSY8GV2L/B6wW0JYevq\n9l147hsIr/PnAMysEfg0YUvlmwXz3UjYFfd/wGIzazazz/SkiXP3G9398GR9RhK2KB4A3GZmfYtm\nf7b48cm09tdxcPLv9XTM4mygr5n1T9ZxCzO7xMxaCA1qazLfAEq/317cwGq0FP28DFhVoiFZRtF7\nyczONLNHCVuG30hq+GQnNUTvkYLXZtvk5zWEXaOjCK/PHDObYGY7bqB2KO8z6FV3f7to+pMF9xd6\ncQPLKnXfXsB+xK/h64Tmywm7OktKGuDxZvY08es6jJ59jnRnfdP8HJEK0ahS6ZS7rzCzVwkfLF3R\n/iVZ6q//UjobHdXZ9K58CRfP01ktXT7lh5l9nHD83H2EZu01wu6ScYTj4crR3YaiJ5n0VLmvUwMh\n/9MJB84XW7uxJ3b3183sTuAkMzufcPD4VoQmqnC+VcBhZnYkobn4BHAKcLeZjexki1SXJFsb706W\ntRY4g3BM5AMbeWjha9P+R/LXCMfDldK+VfMqwm7QywnH7C0j5Hgjpf/YXrmBGkq9Rht9L5nZ6YSt\ncrcQDnVYkjzum5TeOrPRZbr7lWZ2G6HxPpYw+OEiMzvS3TvL5FnC+6S7n0HFOnv9N5RdqfsaCMd7\nju/kuUo1Ru2+RVjn6YTj2JYS/ii5kvI3onR3fdP8HJEKUeMmG/Nr4AtmdkjRbs1SXiR8AO1F+AsU\nADPbgbC14KUK19ZA+BIp3Nqxd/Jv+3P9k/ChVLx7ZrcSy+vsw+4Ewof4se27jwDM7KxuLKPYi8m/\n+xT8HzPbnDBS9s4uLifLniNk/7q737OxmTfgZ4Qv+lGERrl9ZGoH7n4vYQvf183sIsIuziMJB+9X\nwl8IjdtORdP3KjHvnuSbtOeSf1d0IYv2cyO+e6qTZEtjd3Yx9tSJwHPuflLhRDO7pCcLdfcXCA3p\n5WbWns/XCJmWmn9lcoLoI81sZ3f/+0ae4kXgKDPbsmgr1NDk355+Bj1HGJxxbxmPPZEwoOYLhROT\nXcevF0zqzh8ZL1Ld9ZUM0q5S2ZiphOMwfpI0YBEz29PMvpL8+FvCF/VXi2b7GuHD6DdVqO/8Ej+v\nIWwhgfDBtY6CY4gSX6LjB+TbAGa2TdH0dcm87/6hk5zfrtQVEt6ma1+wdxG22n2laPrZhON6SjYm\nvcwfCE3WN82swx+J3Tgf2q8I78FzCc3bL73oRMbJMVjFHiW8HxsL5tvHzN6/oSczs37JucNKaT9m\nqPi4qDPMbKuCZXyG0Nz9Npk0j/Cl/3Uz27LEcxZmsY6On81foRtbiSug/T3/LjM7BBhRzsKSTBuL\nJr9AOHateHqx7xDymNVJdsPNrL3x+y3h97T4c2E8YevW77pbe5GbgF3M7AvFdyS7uJs28Nh1FG3Z\nSt4nOxfN196AdeVzpNrrKxmkLW6yQe7+vJl9jnCs0ZNmVnjlhI8AnyE50NndHzOzmcD/S75I5xB2\nKZ1BGJk6p8LlrQY+kTznnwgHgo8inLvrjaSm5WZ2M/CV5HCn5wgj3ko1DfMIH6zTzOwPhBGmNxKa\nqAuBP5jZzwnno/oSYSh98WVp5gHHmNl4wnmqXnD3R4qfyN1bzew/gX83s98TTtWwL6E5eYSiXYG9\nSOGusRVmdi7huK75ZnYDYcvCIMLuzLl0bFw7cPc2M7uVcJybAx1OHkzI8TDCHwcvEV6jcwnHeM0t\nmO9Jwi7vozbwlE3AQ2b2J8KpWl4mfIl+mnC1kF+V2LW3FJhrZjOA9xJGuT4N/CRZB7dw9v/fAguS\n+f5O+NI+krA7tP0PgV8Dnzez5YRTcIwAjiYcE1WsWru4fg2cYGazCZnuQTilyALCruru2puwq/km\nwjqtJWzJ3gFo3tAD3f2PZnYeYSDSIjMrvHLCEYTd599K5r0t2UJ3qYXLpf2NsLX2OMII9BfKqL3Q\nLMI5+65Odss/SGiohxA+C0cSBrSU8mvgYgvXL32IsPv3NPJbY9s9Rzgm+Bwze4vQyP3J3TtsPavB\n+koWpT2sVbfecSPs9rmG8KGykvDBcj/hy3HzgvkaCMdvtJ+I80XCeag2L1re88CtJZ5nHXBl0bRd\nk+njC6bNIGzN2Y3w5bqC0Ch1OGkq4TQYNyXztBK+AIZQcGLLgtqvIJz+Yi3xCXjHkD8h6gJCMzqJ\notOHEL6g7iUcr7SO/Al4o9OBFMx/brK89hOSTqPgtArJPPcSLrNUvF4zCLuzuvoaviepoeSJZYuX\nVyr3ZPrhyfQTiqa3r+OBRdMPIzQsSwlfQk8TjvPpcHqNDdQ+Kln2y53cfwTheKyXk/fny4Qv2T1L\nvL/u3shz9SEcv/jL5H3alrx3/kJo4DcrmLc9i5MJu2U3dgLeDxIuH7ckqfN5QuNyRME82xAavsWE\nhu43hF2xzwPTN5Z3ct+k5L7i01nMAJaVmL/De4wwmKB9/f+SvAZdeo8UZH1x8v/tCCd/XUD4vV1K\naF5OKH7cBl6X/ZPXtP0Evq3AHRScIiWZr4lwjrX2+RZtoL4rS0zvdJ0K3h9fJ5wOqP0EyY8QmsfC\n0+YUv159CXswXkneI3MIp3m5p/g9SbiM1+OEP04LT8Db4Xe+Ausb1alb9m+6Vqn0SskWixPdvXi3\npkjNmNnhhKbnJHe/ZWPzi4j0VOaOcTOzi8xsvZldVjDtvmRa+22dmf04zTpFREREai1Tx7iZ2cGE\ns+QXHz/iwP8SLo3UfkzHJnUhZREREZGNycwWt2RE1k8Jo+reLDFLm7u/7u5LkttbJeaR+qL9/JIF\neh+KSM1k5hi3ZGTg6+7+dTO7F/iru1+Y3Hcv4bw07ReXvh34rrtv6OSJIiIiIpuUTOwqNbNTCSOG\nDupklp8Rhvi/ShiVNZUweu+kUjMn1447ljCicVWFyxUREREptgXhTAd/8OSUVNWQeuNmZrsQTsHw\nL15wEd5C7v6Tgh8XmNk/gLvMbHcvfZ6aY+m958ESERGR3us0Sp9vsiJSb9yA4YSLTc8ruCB0H8J1\nB88HGr3j/tyHCYMUBhPOvl3sRYCf/vSnDBkypCpF9zbjx4/n8ssvT7uMzFAeMeWRpyxiyiOmPGLK\nI+/JJ5/k9NNPh4LLGFZDFhq3u+h4AeHrCGc4/68STRvAAYQDgl/rZJmrAIYMGcKBBx5YoTJ7t/79\n+yuLAsojpjzylEVMecSUR0x5lFTVQ7RSb9w8XBh3YeE0M3sbeMPdn0wu4/E5wpnX3wA+BFwGzHH3\nJ2pdb2+1cOHCjc9UR5RHTHnkKYuY8ogpj5jyqL3UG7dOFG5lWwMcQ7j235aEy3rcDFyaQl29Vr9+\n/dIuIVOUR0x55CmLmPKIKY+Y8qi9TDZu7n5Uwf9fIVyHUHpg9913T7uETFEeMeWRpyxiyiOmPGLK\no/YycwJeqa7PfvazaZeQKcojpjzylEVMecSUR0x51F5mTsBbSWZ2IDBv3rx5OmhSREREqm7+/PkM\nHz4cYLi7z6/W82iLW51YtGhR2iVkivKIKY88ZRFTHjHlEVMetafGrU5MnDgx7RIyRXnElEeesogp\nj5jyiCmP2tOu0jrR0tLCoEGD0i4jM5RHTHnkKYuY8ogpj5jyyNOuUqko/WLFlEdMeeQpi5jyiCmP\nmPKoPTVuIiIiIr2EGjcRERGRXkKNW52YMmVK2iVkivKIKY88ZRFTHjHlEVMetafGrU60tbWlXUKm\nKI+Y8shTFjHlEVMeMeVRexpVKiIiItJDGlUqIiIiIhE1biIiIiK9hBq3OtHa2pp2CZmiPGLKI09Z\nxJRHTHnElEftqXGrE+PGjUu7hExRHjHlkacsYsojpjxiyqP21LjVicmTJ6ddQqYoj5jyyFMWMeUR\nUx4x5VF7GlUqIiIi0kMaVSoiIiIiETVuIiIiIr2EGrc6MX369LRLyBTlEVMeecoipjxiyiOmPGpP\njVudmD+/arvbeyXlEVMeecoipjxiyiOmPGpPgxNEREREeqhWgxM2q9aCRUREREpZuXIlzzzzDMuW\nLWPNmjWp1dHY2MiAAQMYPHgwW2yxRWp1dIcaNxEREamJl156iQceeIDnn38ed6dfv3707dsXM6t5\nLe7O6tWrWblyJX369GHw4MEcdthh7LzzzjWvpTvUuImIiEjVPf/88/z85z9n++23Z9SoUQwZMoSt\nttoq7bJYtmwZTz75JPPnz+f666/n85//PLvsskvaZXVKgxPqRC6XS7uETFEeMeWRpyxiyiOmPGJd\nzeMf//gHP//5z9ltt90466yzOPjggzPRtAH079+fQw89lLPPPpsdd9yRWbNmsXTp0rTL6pQatzpx\n/vnnp11CpiiPmPLIUxYx5RFTHrGu5vHXv/6Vfv36ceqpp7LZZtnc2de3b19OO+00AB577LGUq+mc\nGrc6MXLkyLRLyBTlEVMeecoipjxiyiPWlTzcnYULFzJ06NDMNm3tGhsb2WeffViwYEHapXRKjZuI\niIhUzZIlS1ixYgX77rtv2qV0yZAhQ3j99ddZtmxZ2qWUpMZNREREqubtt98GwrFkvUF7nW1tbSlX\nUpoatzoxe/bstEvIFOURUx55yiKmPGLKI9aVPNauXQvA5ptvXu1yKqK9znfeeSflSkpT41Ynmpub\n0y4hU5RHTHnkKYuY8ogpj1h38kjjXG3lyHqdatzqxI033ph2CZmiPGLKI09ZxJRHTHnEKplHa2sr\n5557LrvuuitbbLEFO+20E6NGjeL+++9n++23Z+rUqSUf993vfpeddtqJdevWMXPmTBoaGthvv/06\nzHfTTTfR0NDAHnvsUbGa06DGTURERFJ3wgkn8OijjzJr1iyeeeYZbr/9do444giWL1/O5z//eWbM\nmFHycTNnzuTMM8+kT58+AGy55ZYsWbKEhx9+OJpvxowZ7LrrrlVfj2rL9rhcERER2eQtW7aMuXPn\nMmfOHD7+8Y8D8P73v5+DDjoIgN12240rr7yShx56iI985CPvPu6+++7jhRdeYNy4ce9O22yzzfjc\n5z7H9OnTOeSQQwD4+9//zn333ceFF17Y63d3a4ubiIiIpGqrrbZiq622Yvbs2SUvOv+BD3yAgw46\niGuvvTaaPmPGDD7ykY+w9957vzvNzDjrrLO48cYbWbVqFQDXXXcdo0aNYocddqjuitSAGrc6MXbs\n2LRLyBTlEVMeecoipjxiyiNWqTz69OnDzJkzmTlzJgMGDOBjH/sY3/rWt3j88cffnWfcuHHcfPPN\n756m46233uKWW27h7LPP7rC8D37wg+y555784he/AELjVrhVrjdT41YndLbvmPKIKY88ZRFTHjHl\nEatkHscffzyvvvoqt99+O6NGjWLOnDkceOCBXH/99QB89rOfZe3atdx0000A3HDDDTQ0NPCZz3ym\n5PLGjRvHtddey5w5c3j77bcZPXp0xWpNkxq3OvHZz3427RIyRXnElEeesogpj5jyiFU6j759+3L0\n0UfzrW99i7lz5zJmzBgmTZoEwDbbbMNJJ5307iCF6667jpNPPpmmpqaSyzrttNP405/+xOTJkznj\njDNoaNg0Wp5NYy1ERERkkzNkyJB3r7wAcNZZZzF37lx+85vf8NBDD5XcTdpu2223JZfLcf/993PW\nWWfVotyaUOMmIiIiqVq6dClHH300P/vZz3j88cd58cUXufnmm/n+97/Ppz/96XfnO+yww9hzzz05\n44wzGDJkyLujRjszc+ZMWltb2Wuvvaq9CjWjxq1OzJ07N+0SMkV5xJRHnrKIKY+Y8ohVKo+tttqK\nQw89lCuuuILDDz+cYcOGMWnSJL74xS8ybdq0aN5x48bx5ptvdmkrWmNjI9tuu21FaswKNW51orMz\nTtcr5RFTHnnKIqY8YsojVqk8+vbty6WXXsqf//xnli5dyooVK1i4cCGTJ0+msbExmvcb3/gG69at\n48ILL+ywnDPPPJOlS5d2+jwXXHABzz//fEVqTosatzpxww03pF1CpiiPmPLIUxYx5RFTHjHlUXtq\n3OpEZ6Nu6pXyiCmPPGURUx4x5RHrSh7tl6Jau3ZttcupiPY62+vOGjVuIiIiUjVbbLEFQDQ6NMva\n6+zXr1/KlZSmxk1ERESqZscdd6SxsZFnn3027VK65JlnnmGrrbZiwIABaZdSkhq3OjFhwoS0S8gU\n5RFTHnnKIqY8Ysoj1pU8NttsM/bZZx8WLFiAu9egqvK5OwsXLmTo0KGZPWFvNquSihs0aFDaJWSK\n8ogpjzxlEVMeMeUR21Aejz76KHfeeScQrh26ZMkS7r333sw2b+7O7373O5YvX86wYcPSLqdTlrUA\nzewi4FLgCne/MJnWCFwGnAI0An8AvuTuSzpZxoHAvHnz5nHggQfWpnARERHh0Ucf5ZJLLuGWW27h\n6KOP5q677gLgwQcf5M4772T48OEcfPDB7LjjjphZytWGhu3VV1/l4Ycf5rHHHuO4445j+PDh3V7O\n/Pnz2x833N3nV7zQxGbVWnA5zOxg4AvAo0V3XQGMAk4ElgM/An4JfLymBYqIiEhJhQ3bHnvswYwZ\nMzjttNPevf+jH/0offr0Yc6cOcybN4/tttuO7bffnsbGxlQaOHdn1apVLFmyhDfffJMtt9ySXC6X\n+Q0+mWnczGwr4KfA2cDFBdO3AcYBp7r7nGTaWOBJM/uwuz+SRr0iIiLSecO2+eabd5j30EMP5eCD\nD+aFF17gqaeeYtmyZSxbtiyFqsHMaGxsZO+992bIkCHsuuuumT2urVBmGjfCVrTb3f0eM7u4YPpB\nhDrvbp/g7k+ZWQswAlDj1gWLFi1i3333TbuMzFAeMeWRpyxiyiOmPPIeffRRJkyYwJ133rnRhq1Q\nnz59GDx4MIMHD65RpZuWTLSWZnYqsD9wUYm7dwTWuPvyoumLgfdWu7ZNxcSJE9MuIVOUR0x55CmL\nmPKIKY/QsJ144onsv//+PPjgg8yYMYNFixYxZsyYjTZt0nOpN25mtgvhGLbT3f2d7jwU2ODIitGj\nR5PL5aLbiBEjmD17djTfHXfcQS6X6/D48847j+nTp0fT5s+fTy6Xo7W1NZo+adIkpkyZEk1raWkh\nl8uxaNGiaPq0adM6DKFua2sjl8t1uGBvc3MzY8eO7VDbKaec0q31+NjHPrZJrEelXo+rrrpqk1iP\nQj1Zj/Y8evt6tOvJelx11VWbxHpAZV6P4cOHbxLrUanX47jjjtsk1qOc16OwYfv973/PD3/4Qx57\n7LF3G7besh7tevJ6NDc3k8vlGDZsGIMHDyaXyzF+/PgOz1sNqY8qNbNPAbcA6wjNGEAfQlO2DvgE\ncBcwoHCrm5m9CFzu7leWWKZGlYqIiFRA8TFsF198cZd2idabWo0qTX2LG6EpG0bYVfqh5PYXwkCF\n9v+/Axzd/gAz2xsYBPyx1sWKiIjUg7/97W+ccMIJ7L///vztb3/TLtGMSL1xc/e33X1h4Q14G3jD\n3Z9MtrJNBy4zsyPMbDgwA3hQI0pFRETKs3jxYi644ALeeuutaHp7w3bAAQfw2GOPMWPGDJ566ik1\nbBmReuPWieL9t+OBXwO/AO4DXiWc0026qHh/fb1THjHlkacsYsojtqnksW7dOk477TRuvPHGd8+h\nVqpha9/hcm7cAAAgAElEQVTCttlmpU9Csank0Ztk6XQg73L3o4p+Xg18OblJGdra2tIuIVOUR0x5\n5CmLmPKI9bY81q9/h/XrV9PQ0EhDQ35r2aWXXsq9997LXXfdxTPPPMMll1zCr371K/bcc09mzJjB\n6aef3mmzVqi35bEpSH1wQjVocIKIiNSzt1e18Mabf2RF21OEnVjG1k37MHDACP700LMcc8wxfPGL\nX2Tx4sXvNmzf/va3u9ywSUd1eckrERER6Zmly//Ma62/JRwN1b5xxlnR9jQvvPRnTvrMTxg4cCDX\nXHNNt7ewSfqyeoybiIiIdNPbq1qSpg1gfdG96znjlGv559JlLFv2JkOGDGHQoEHMmjWLY489lhtv\nvLHW5UoZ1F7XidbWVgYOHJh2GZmhPGLKI09ZxJRHLOt5vPHmHwnbZIqbtmDIB3ai/4At2XXX97NV\n067Rfdtss023ny/reWyK1LjViXHjxnHbbbelXUZmKI+Y8shTFjHlEctyHuvXv1NwTFtpP/jhycn/\njCG7XRQNWChHlvPYVGlXaZ2YPHly2iVkivKIKY88ZRFTHrEs57F+/Wo2ciXIAp7M3zNZzmNTpcat\nTmh0bUx5xJRHnrKIKY9YlvNoaGgkf+XIjbFk/p7Jch6bKjVuIiIim4CGhs3ZumkfNv7V3sDWTfv2\neDeppEONm4iIyCbiPQNG0NnAhLz1DBxwaC3KkSpQ41Ynpk+fnnYJmaI8YsojT1nElEcs63lsucUg\ndhr4yeSn4q/48PNOAz9J0xaDKvJ8Wc9jU6TGrU7Mn1+1kzj3SsojpjzylEVMecR6Qx7bbXMQu79v\nbLLbtP2Yt3DlhN3fN5bttjmoYs/VG/LY1OiSVyIiIpuozq5VKpWnS16JiIhIjzQ0bK6GbROjXaUi\nIiIivYQaNxEREZFeQo1bncjlcmmXkCnKI6Y88pRFTHnElEdMedSeGrc6cf7556ddQqYoj5jyyFMW\nMeURUx4x5VF7GlUqIiIi0kO1GlWqLW4iIiIivYQaNxEREZFeQo1bnZg9e3baJWSK8ogpjzxlEVMe\nMeURUx61p8atTjQ3N6ddQqYoj5jyyFMWMeURUx4x5VF7GpwgIiIi0kManCAiIiIiETVuIiIiIr2E\nGjcRERGRXkKNW50YO3Zs2iVkivKIKY88ZRFTHjHlEVMetafGrU6MHDky7RIyRXnElEeesogpj5jy\niCmP2tOoUhEREZEe0qhSEREREYmocRMRERHpJdS41Ym5c+emXUKmKI+Y8shTFjHlEVMeMeVRe2rc\n6sTUqVPTLiFTlEdMeeQpi5jyiCmPmPKoPQ1OqBNtbW00NTWlXUZmKI+Y8shTFjHlEVMeMeWRp8EJ\nUlH6xYopj5jyyFMWMeURUx4x5VF7atxEREREegk1biIiIiK9hBq3OjFhwoS0S8gU5RFTHnnKIqY8\nYsojpjxqT41bnRg0aFDaJWSK8ogpjzxlEVMeMeURUx61p1GlIiIiIj2kUaUiIiIiElHjJiIiItJL\nqHGrE4sWLUq7hExRHjHlkacsYsojpjxiyqP21LjViYkTJ6ZdQqYoj5jyyFMWMeURUx4x5VF7GpxQ\nJ1paWjT6p4DyiCmPPGURUx4x5RFTHnkanCAVpV+smPKIKY88ZRFTHjHlEVMetafGTURERKSXUOMm\nIiIi0kuocasTU6ZMSbuETFEeMeWRpyxiyiOmPGLKo/bUuNWJtra2tEvIFOURUx55yiKmPGLKI6Y8\nai/1UaVmdg5wLrBbMmkBcIm7/z65/z7gsIKHOPA/7v6lDSxTo0pFRESkZmo1qnSzai24G14G/g14\nNvl5DHCrme3v7k8SGrX/BS4GLJlHLb6IiIjUndQbN3f/TdGkb5vZucChwJPJtDZ3f722lYmIiIhk\nS6aOcTOzBjM7FWgCHiq46zQze93MHjez75lZv5RK7LVaW1vTLiFTlEdMeeQpi5jyiCmPmPKovUw0\nbmb2ATNbAawGfgwc7+5PJXf/DDgdOAL4HvB5YFYadfZm48aNS7uETFEeMeWRpyxiyiOmPGLKo/Yy\n0bgBi4APAYcAVwPXm9m+AO7+E3e/090XuHszcAZwvJntvrGFjh49mlwuF91GjBjB7Nmzo/nuuOMO\ncrlch8efd955TJ8+PZo2f/58crlch78yJk2a1GFYdEtLC7lcrsNFeKdNm8aECROiaW1tbeRyOebO\nnRtNb25uZuzYsR1qO+WUU7q1Hvvvv/8msR6Vej0mT568SaxHoZ6sR3sevX092vVkPSZPnrxJrAdU\n5vXYd999N4n1qNTrcfjhh28S61Gp16Pws7Q3r0ehrqxHc3MzuVyOYcOGMXjwYHK5HOPHj+/wvNWQ\n+qjSUszsTuBZdz+3xH1NwFvAse5+ZyeP16hSERERqZl6v1ZpA9DYyX0HEEaavla7ckRERETSl/qo\nUjO7FPgd4bQgWwOnAYcDI81sD+BzwG+BNwi7Uy8D5rj7E+lULCIiIpKOLGxx2xG4nnCc213AcGCk\nu98DrAGOAf5AODXI94GbgY47sGWDivft1zvlEVMeecoipjxiyiOmPGov9cbN3c929z3cvZ+7v9fd\n25s23P0Vdz/C3bd39yZ338fdL3L3t9Kuu7eZP79qu9t7JeURUx55yiKmPGLKI6Y8ai+TgxN6SoMT\nREREpJbqfXCCiIiIiBRR4yYiIiLSS6hxExEREekl1LjViVJnkq5nyiOmPPKURUx5xJRHTHnUnhq3\nOnH++eenXUKmKI+Y8shTFjHlEVMeMeVRexpVKiIiItJDGlUqIiIiIhE1biIiIiK9hBq3OjF79uy0\nS8gU5RFTHnnKIqY8YsojpjxqT41bnWhubk67hExRHjHlkacsYsojpjxiyqP2NDhBREREpIc0OEFE\nREREImrcRERERHoJNW4iIiIivYQatzoxduzYtEvIFOURUx55yiKmPGLKI6Y8ak+NW50YOXJk2iVk\nivKIKY88ZRFTHjHlEVMetadRpSIiIiI9pFGlIiIiIhJR4yYiIiLSS6hxqxNz585Nu4RMUR4x5ZGn\nLGLKI6Y8Ysqj9tS41YmpU6emXUKmKI+Y8shTFjHlEVMeMeVRexqcUCfa2tpoampKu4zMUB4x5ZGn\nLGLKI6Y8YsojT4MTpKL0ixVTHjHlkacsYsojpjxiyqP21LiJiIiI9BJq3ERERER6CTVudWLChAlp\nl5ApyiOmPPKURUx5xJRHTHnUnhq3OjFo0KC0S8gU5RFTHnnKIqY8YsojpjxqT6NKRURERHpIo0pF\nREREJKLGTURERKSXUONWJxYtWpR2CZmiPGLKI09ZxJRHTHnElEftqXGrExMnTky7hExRHjHlkacs\nYsojpjxiyqP2NDihTrS0tGj0TwHlEVMeecoipjxiyiOmPPI0OEEqSr9YMeURUx55yiKmPGLKI6Y8\nak+Nm4iIiEgvocZNREREpJdQ41YnpkyZknYJmaI8YsojT1nElEdMecSUR+2pcasTbW1taZeQKcoj\npjzylEVMecSUR0x51J5GlYqIiIj0kEaVioiIiEhEjZuIiIhIL6HGrU60tramXUKmKI+Y8shTFjHl\nEVMeMeVRe2rc6sS4cePSLiFTlEdMeeQpi5jyiCmPmPKoPTVudWLy5Mlpl5ApyiOmPPKURUx5xJRH\nTHnUXo9GlZpZo7uvrmA9FaFRpSIiIlJLmRxVamajzGymmT1vZu8AbWa2wszmmNm3zOx9VapTRERE\npO51qXEzs0+b2dPAtcBaYApwAnAscBYwBzgGeN7MrjGz7atUr4iIiEjd6uoWt38DxgM7u/tZ7v4/\n7n67u9/l7je5+7+7+5HAnsCbwOnVKljKM3369LRLyBTlEVMeecoipjxiyiOmPGqvS42bu49w99+4\n+/qNzPd3d/+Gu1/e1QLM7Bwze9TMliW3h8zsEwX3N5rZj8ysNdkt+wsz26Gry5dg/vyq7W7vlZRH\nTHnkKYuY8ogpj5jyqL0eX/LKzPoAw4CX3P2fZTz+k8A64Nlk0hhgArC/uz9pZlcDo4AzgeXAj4B1\n7v7xDSxTgxNERESkZjI5OAHAzK4ws7OS//chHN82H3jZzI7o7vKSLXm/d/dnk9u3gbeAQ81sG2Ac\nMN7d57j7X4GxwEfN7MPdfS4RERGR3qyc87idBDya/P84YHdgX+By4NKeFGNmDWZ2KtAE/BEYDmwG\n3N0+j7s/BbQAI3ryXCIiIiK9TTmN20DgH8n/RwM3u3v7iNNh5RRhZh8wsxXAauDHwPHuvgh4L7DG\n3ZcXPWRxcp+IiIhI3SincVsMDE12k34CuCuZ3kQ4Vq0ci4APAYcAVwPXm9m+G5jfgI0enDd69Ghy\nuVx0GzFiBLNnz47mu+OOO8jlch0ef95553UYMTN//nxyuVyH67NNmjSJKVOmRNNaWlrI5XIsWrQo\nmj5t2jQmTJgQTWtrayOXyzF37txoenNzM2PHju1Q2ymnnNKt9dh///03ifWo1OtRuPzevB6FerIe\n7XX29vVo15P1yOVym8R6QGVej2HDhm0S61Gp1+Pggw/eJNajUq9H4WN683oU6sp6NDc3k8vlGDZs\nGIMHDyaXyzF+/PgOz1sN3R6cYGaTga8CrxGatb3dfbWZjQO+4O493oVpZncSBivcRGgMty3c6mZm\nLwKXu/uVnTxegxOK3HHHHYwcOTLtMjJDecSUR56yiCmPmPKIKY+8Wg1OKGtUqZmdBLyfsJv0lWTa\nmcCb7n5rj4syuxt4idAgvg6c6u6/Su7bm7CF7lB3f6STx6txExERkZqpVeO2WTkPcvdflJg2s5xl\nmdmlwO+Al4GtgdOAw4GR7r7czKYDl5nZP4EVwA+BBztr2kREREQ2VV1q3MzsK11doLv/sJs17Ahc\nD+wELAMeIzRt9yT3jyccO/cLoBH4PXBeN59DREREpNfr6uCE8UW37wFXAJOT2xXJtK92twB3P9vd\n93D3fu7+XncvbNpw99Xu/mV3H+juW7v7Z9x9SXefp94VH8BZ75RHTHnkKYuY8ogpj5jyqL2uXvJq\n9/Yb8C3gb8AQd9/O3bcDhhBOwntx9UqVnmhubk67hExRHjHlkacsYsojpjxiyqP2yhlV+hxwUnIV\ng8Lpw4FfJM1dqjQ4QURERGops5e8IhyLVurYuD6E49VEREREpArKadzuBv4n2aoFvLu17WryJ+MV\nERERkQorp3EbR7jk1V/MbLWZrQYeIVxR4exKFiciIiIied1u3Nz9dXcfTbiw/GeAkwkDFUZrtGd2\nlbpcSD1THjHlkacsYsojpjxiyqP2yjoBL0ByYfmnK1iLVJEuSRJTHjHlkacsYsojpjxiyqP2yhlV\n2gcYAxwN7EDRVjt3P6pSxZVLo0pFRESklrJ8yasrCY3bb4AngO5f7FREREREuq2cxu1U4GR3/22l\nixERERGRzpUzqnQN8GylC5Hqmjt3btolZIryiCmPPGURUx4x5RFTHrVXTuP2A+ACM7NKFyPVM3Xq\n1LRLyBTlEVMeecoipjxiyiOmPGqvnMEJvwKOBJYCC4B3Cu939xMqVl2ZNDiho7a2NpqamtIuIzOU\nR0x55CmLmPKIKY+Y8sjL8uCEN4FfVboQqS79YsWUR0x55CmLmPKIKY+Y8qi9bjdu7q6z7YmIiIik\noOwT8JrZ9sA+hNOBPO3ur1esKhERERHpoNuDE8xsSzO7FngNuB94AHjVzKabmbaZZtSECRPSLiFT\nlEdMeeQpi5jyiCmPmPKovXJGlV4GHA4cBwxIbp9Kpv2gcqVJJQ0aNCjtEjJFecSUR56yiCmPmPKI\nKY/aK2dUaStwkrvfVzT9SOAmd9++cuWVR6NKRUREpJZqNaq0nC1uTcDiEtOXJPeJiIiISBWU07j9\nEfiOmW3RPsHM+gGTkvtEREREpArKadwuAD4CvGJmd5vZXcDLybQLKlmcVM6iRYvSLiFTlEdMeeQp\ni5jyiCmPmPKovW43bu7+BLAXcBHwN+Ax4BvAXu6+oLLlSaVMnDgx7RIyRXnElEeesogpj5jyiCmP\n2uv24ITeQIMTOmppadHonwLKI6Y88pRFTHnElEdMeeRldnCCmV1kZuNKTB9nZv9WmbKk0vSLFVMe\nMeWRpyxiyiOmPGLKo/bKOcbti0CpndoLgHN6Vo6IiIiIdKacxu29hKsmFHsd2Kln5YiIiIhIZ8pp\n3F4GPlpi+keBV3tWjlTLlClT0i4hU5RHTHnkKYuY8ogpj5jyqL1yLjL/f8AVZrY5cE8y7WhgKrrk\nVWa1tbWlXUKmKI+Y8shTFjHlEVMeMeVRe+Vc8sqA/wK+AvRNJq8Cprj7JZUtrzwaVSoiIiK1VKtR\npd3e4uah0/s3M/suMARYCTzj7qsrXZyIiIiI5JVzjFu79wLbAc+5++pkS5yIiIiIVEk553F7j5nd\nDTwN/Jb8SNLpZqZj3DKqtbU17RIyRXnElEeesogpj5jyiCmP2itni9vlwDvAIKDwqMQbgU9Uoiip\nvHHjOpwzua4pj5jyyFMWMeURUx4x5VF75YwqHQkc6+6vFO0dfQbYtSJVScVNnjw57RIyRXnElEee\nsogpj5jyiCmP2itni9uWxFva2m0HaIBCRml0bUx5xJRHnrKIKY+Y8ogpj9orp3F7ADij4Gc3swZg\nInBvRaoSERERkQ7K2VU6EbjbzA4inMdtKrAfYYtbqSsqiIiIiEgFdHuLm7s/AewNzAVuJew6vQU4\nwN2fq2x5UinTp09Pu4RMUR4x5ZGnLGLKI6Y8Ysqj9so6j5u7L3P3S939ZHcf7e7fdvdSF56XjJg/\nv2once6VlEdMeeQpi5jyiCmPmPKovS5f8srMBgJbuvtLBdP2A75O2Oo2291/XpUqu0mXvBIREZFa\nqtUlr7qzxW0a4fqkAJjZDoSBCgcDjcB1Zvb5ypYnIiIiIu2607gdCtxW8PMZwFJgf3f/FPBN4LwK\n1iYiIiIiBbrTuL0XeLHg56OAW9x9bfLzbcBeFapLRERERIp0p3FbDgwo+PnDwMMFPzthl6lkUC6X\nS7uETFEeMeWRpyxiyiOmPGLKo/a607j9CfiKmTWY2UnA1sA9BffvDbxcyeKkcs4///y0S8gU5RFT\nHnnKIqY8YsojpjxqrzujSj8I3A1sQzhx7/fc/eKC+2cBb7v7OdUotDs0qlRERERqqVajSrt85QR3\nf8zMhhCujvAPd3+4aJYbgIWVLE5ERERE8rp1ySt3byVcLaHUfb+pSEUiIiIiUlKXjnEzs1O7ukAz\ne7+ZdfmapWZ2kZk9YmbLzWyxmf3KzPYumuc+M1tfcFtnZj/u6nMIzJ49O+0SMkV5xJRHnrKIKY+Y\n8ogpj9rr6uCEL5nZk2Y20cz2Lb7TzPqb2Wgz+zkwH3hPN2r4OOHkvocAxwCbA3eYWb+CeRz4X2BH\nwmlJdiJc7F66qLm5Oe0SMkV5xJRHnrKIKY+Y8ogpj9rrzuCEHPBlwvnb3gYWA6uAbQnNVCtwHXCZ\nuy8pu6Bwaa0lwGHuPjeZdi/wV3e/sIvL0OAEERERqZksDk64Dbgtaaw+BuwK9CM0bH8lNFbrK1DT\nAMIWtqVF009LLqn1D+B24LvuvrICzyciIiLSK3RrcAK8O0ChKju1zcyAK4C57l44QvVnwEvAq8AH\ngamE88adVI06RERERLKoOyfgrYUfA0OBaDCEu//E3e909wXu3ky4TurxZrb7hhY2evRocrlcdBsx\nYkSHgynvuOOOkmd/Pu+885g+fXo0bf78+eRyOVpbW6PpkyZNYsqUKdG0lpYWcrkcixYtiqZPmzaN\nCRMmRNPa2trI5XLMnTs3mt7c3MzYsWM71HbKKadoPbQeWg+th9ZD66H1SGE9mpubyeVyDBs2jMGD\nB5PL5Rg/fnyH560Kd8/EDbiKsFVtUBfmbQLWA//Syf0HAj5v3jyXYMyYMWmXkCnKI6Y88pRFTHnE\nlEdMeeTNmzfPCYd6HehV7Je6vau0GszsKuBTwOHu3tKFhxxACOe1qha2CRk5cmTaJWSK8ogpjzxl\nEVMeMeURUx611+VRpVUrIJyP7bNADni64K5l7r7KzPYAPgf8FngD+BBwGdDi7kd1skyNKhUREZGa\nqdWo0rKPcTOzvma2j5n1dKvdOYTrn95HGHzQfjs5uX8N4fxufwCeBL4P3Exo9ERERETqRrebLjNr\nIpww98xk0t7A82Y2Dfi7u/9Xd5bn7htsHt39FeCI7tYpIiIisqkpZ4vbfxJ2Vx5BOAFvu7uAUypQ\nk1RB8Sideqc8YsojT1nElEdMecSUR+2V07h9Gjjfw1UNCg+QWwDsWZGqpOKmTp2adgmZojxiyiNP\nWcSUR0x5xJRH7XV7cIKZtQEfcPfnzWwF8KHk/x8C7nf3/tUotDs0OKGjtrY2mpqa0i4jM5RHTHnk\nKYuY8ogpj5jyyMvy4IS/AJ8s+Lm98zsb+GOPK5Kq0C9WTHnElEeesogpj5jyiCmP2itnROg3gd+Z\n2dDk8ReY2X7ACODwShYnIiIiInnd3uKWHNu2P6FpexwYCSwGRrj7vMqWJyIiIiLtyjqPm7s/5+5f\ncPcPu/tQdz/d3R+vdHFSOcXXhKt3yiOmPPKURUx5xJRHTHnUXtknzzWzHYAdKGr+3P2xnhYllTdo\n0KC0S8gU5RFTHnnKIqY8Ysojpjxqr5xRpcOBmcAQwIrudnfvU6HayqZRpSIiIlJLtRpVWs4Wt2sJ\n1xQ9i3BsW7oXOxURERGpE+U0bnsAJ7r7s5UuRkREREQ6V87ghLsJl7ySXmTRokVpl5ApyiOmPPKU\nRUx5xJRHTHnUXjmN29nAODObZGYnmlmu8FbpAqUyJk6cmHYJmaI8YsojT1nElEdMecSUR+2VMzjh\nOGAWsE2JuzU4IaNaWlo0+qeA8ogpjzxlEVMeMeURUx55Wb7k1TTgp8BO7t5QdEu9aZPS9IsVUx4x\n5ZGnLGLKI6Y8Ysqj9spp3N4DXO7uiytdjIiIiIh0rpzG7RbgyEoXIiIiIiIbVk7j9jTwn2Z2nZl9\nzcy+UnirdIFSGVOmTEm7hExRHjHlkacsYsojpjxiyqP2yjmP29nAW8Dhya2QAz/saVFSeW1tbWmX\nkCnKI6Y88pRFTHnElEdMedRet0eV9gYaVSoiIiK1lOVRpSIiIiKSgi7tKjWzy4CL3f3t5P+dcvcL\nK1KZiIiIiES6usXtAGDzgv93dtu/0gVKZbS2tqZdQqYoj5jyyFMWMeURUx4x5VF7XWrc3P1Id3+z\n4P+d3Y6qbrlSrnHjxqVdQqYoj5jyyFMWMeURUx4x5VF7XT7GzczWmdkO1SxGqmfy5Mlpl5ApyiOm\nPPKURUx5xJRHTHnUXpdHlZrZeuC97r6kuiX1nEaVioiISC1pVKmIiIiIRLp7At6zzeytDc3g7joB\nr4iIiEgVdHeL2znA+A3cvlrR6qRipk+fnnYJmaI8YsojT1nElEdMecSUR+11t3E7yN1338Btj6pU\nKT02f37Vdrf3SsojpjzylEVMecSUR0x51F53BiesA3bS4AQRERGRWBYHJ1i1ihARERGRjetO4/Yd\nYIMDE0RERESkero8qtTdv1PNQkRERERkw3QetzqRy+XSLiFTlEdMeeQpi5jyiCmPmPKoPTVudeL8\n889Pu4RMUR4x5ZGnLGLKI6Y8Ysqj9ro8qrQ30ahSERERqaUsjioVERERkRR1u3Ezsx3NbJaZvWpm\na81sXeGtGkWKiIiISHlb3K4DDgS+C5wEnFB0kwyaPXt22iVkivKIKY88ZRFTHjHlEVMetVdO4/Yx\n4DR3v9rdZ7v7rYW3ShcoldHc3Jx2CZmiPGLKI09ZxJRHTHnElEftdXtwgpktJDRuf61OST2nwQki\nIiJSS1kenPBV4L/MbLfKliIiIiIiG9LlKycUuBFoAp4zszbgncI73X27ShQmIiIiIrFyGrevVrwK\nEREREdmobu8qdfeZG7pVo0jpubFjx6ZdQqYoj5jyyFMWMeURUx4x5VF75Wxxw8z6AJ8GhgAOLARu\nc3edxy2jRo4cmXYJmaI8YsojT1nElEdMecSUR+2VM6p0MPBbYGfgKcCAvYGXgU+6+3OVLrK7NKpU\nREREainLo0p/CDwHvN/dD3T3A4BBwAvJfSIiIiJSBeU0bocDE919afsEd38D+EZyX7eY2UVm9oiZ\nLTezxWb2KzPbu2ieRjP7kZm1mtkKM/uFme1QRu0iIiIivVY5jdtqYOsS07cC1pSxvI8D04BDgGOA\nzYE7zKxfwTxXAJ8ETgQOA94H/LKM56pbc+fOTbuETFEeMeWRpyxiyiOmPGLKo/bKadx+DfyvmR1i\neYcC1wC3dXdh7j7a3We5+5Pu/jgwhrDrdTiAmW0DjAPGu/uc5IoNY4GPmtmHy6i/Lk2dOjXtEjJF\necSUR56yiCmPmPKIKY/aK2dwwgBgJnAc+ZPvbkZo2sa4+7IeFRQGPzwFDHP3hWZ2JHAXsK27Ly+Y\n70Xgcne/ssQyNDihSFtbG01NTWmXkRnKI6Y88pRFTHnElEdMeeTVanBCt08H4u5vAp8ys72AfQmj\nShe6+7M9LcbMjLBbdK67L0wmvxdYU9i0JRYn90kX6BcrpjxiyiNPWcSUR0x5xJRH7ZWzqxQAd3/G\n3W9399sq0bQlfgwMBT7bhXmNcA65To0ePZpcLhfdRowYwezZs6P57rjjDnK5XIfHn3feeUyfPj2a\nNn/+fHK5HK2trdH0SZMmMWXKlGhaS0sLuVyORYsWRdOnTZvGhAkTomltbW3kcrkOxws0NzeXPMHh\nKaecovXQemg9tB5aD62H1iOF9WhubiaXyzFs2DAGDx5MLpdj/PjxHZ63Grq0q9TMLgMudve3k/93\nyt0vLKsQs6sIu18/7u4tBdO1q1REREQyLWvncTuAMNqz/f8bunVb0rR9CjiysGlLzAPWAkcXzL83\nYQDDH8t5vnpU/JdNvVMeMeWRpyxiyiOmPGLKo/a6dIybux9Z6v+VYGY/JuwazQFvm9mOyV3L3H2V\nu7nhVlAAACAASURBVC83s+nAZWb2T2AF4US/D7r7I5WsZVM2aNCgtEvIFOURUx55yiKmPGLKI6Y8\naq+cUaXXAhe4+4qi6VsC09x9XDeXt57Sx6qNdffrk3kagf8mNHiNwO+B89x9SSfL1K5SERERqZms\n7SotdCbQr8T0fsAZ3V2Yuze4e58St+sL5lnt7l9294HuvrW7f6azpk1ERERkU9Xl04EkJ8K15La1\nma0quLsPMBpQMyUiIiJSJd3Z4vYmsJSwW/Np4J8Ft1bgWuBHlS5QKqN4KHa9Ux4x5ZGnLGLKI6Y8\nYsqj9rrTuB1JGNlpwEnAUQW3jwGD3P3SilcoFTFx4sS0S8gU5RFTHnnKIqY8Ysojpjxqr5zBCbsC\nL7v7+uqU1HManNBRS0uLRv8UUB4x5ZGnLGLKI6Y8YsojL8uXvHoJwMyaCOdS61t0/2OVKU0qqZa/\nWOvXr+e1116jra2NtWvX1ux5u+vJJ5+syHIaGxsZMGAA2223XUWWlwZ98OYpi5jyiCmPmPKovW43\nbma2PTADGNXJLH16VJH0Wq+88grz589n0aJFtLW1pV1Oze24444MHTqUgw46iC233DLtckREZBPU\n7caNcBH4AcAhwH3A8cCOwLeBr1WsMulVFi1axE033UT//v058MAD2Xfffdlmm23YfPPNN/7gXszd\nWb16NYsXL2bBggU8+OCDPP7445x55plsvfXWaZcnIiKbmHLO43YUcKG7/wVYD7zk7j8FJgIXVbI4\nqZziC+ZW0jPPPMNNN93EkCFD+PKXv8wxxxzDLrvswjbbbEO/fv0yeZs3b15FltPU1MS2227Lvvvu\ny4knnsg555zDmjVrmDlzJitXrqxa5pVWzfdHb6MsYsojpjxiyqP2ymnctiR/vrZ/Atsn/38c0EiA\njKrmrssHHniA97///Zx44ok0NJTzlqq9d955pyrL3W677TjzzDNZunQpTzzxRFWeoxrqcdd2Z5RF\nTHnElEdMedReOaNK/wx8293/YGa3Ec7vdhHwFeAkd9+z8mV2j0aV1s7y5cu57LLLOP744/nQhz6U\ndjmZMWvWLNatW8eYMWPSLkVERGogs6NKCce47ZT8/zuE64aeBqwBxlSmLOktnn76aRoaGthnn33S\nLiVT9ttvP26//XZWrlxJv36lrhAnIrXyxhtvsHDhQl577TVWrVrF+vWZPZuVJBoaGmhsbGSHHXZg\nv/32Y/vtt8fM0i4rE8o5HcjPCv4/Lzmv275Ai7u3VrI4yb7ly5ez9dZbs8UWW6RdSqYMHDgQd+et\nt95S4yaSktbWVn75y1/y2muv0bdvX3bZZRf69etHnz46+UHWrV+/npUrV/Lwww8zZ84cBg4cyPHH\nH8/OO++cdmmpK+d0IP8O/Le7twEk/843s35m9u/ufkmli5Sea21tZeDAgRVf7po1a2hsbKz4cqut\nra2Npqamqi2/b99wesPVq1dX7TkqqVrvj95IWcR6ax6tra1cd9119OvXj5NPPpm99tqrIqPcq/3Z\n0dtUO4+1a9fy/PPPM2fOHK6//nrOOOOMum/eyjmSfBKwVYnpTcl9kkHjxo1Lu4RMufXWW6u6/N62\nSV/vjzxlEeuNeaxfv55Zs2bR1NTEmDFjGDp0aMVOTVTtz47eptp5bLbZZuy9996cccYZ7LDDDsya\nNYs1a9ZU9TmzrpzGzQgXmi/2IcJF6CWDJk+enNpzjx07lhNOOKGmz/nSSy/R0NDw7u0973kPRxxx\nBHPnzgXgiCOOqGk9WZfm+yNrlEWsN+bR0tLCsmXLyOVyFT8Ztj47YrXKo7GxkRNOOIFVq1bx9NNP\n1+Q5s6rLjZuZ/dPMlhKatqfNbGnBbRlwJ3BTtQqVnqnH0bVmxj333MM//vEPHnjgAd73vvfxr//6\nr7z++uvstNNOG19AHanH90dnlEWsN+axYMEC+vfvX5VdavrsiNUyj2233Zb3ve99LFy4sGbPmUXd\n2eL2VeBCwha3ScD4gts5wMfc/byKVyibvJdffplPfepTbL311vTv359TTjmFJUuWRPP8x3/8Bzvu\nuCP9+/fnC1/4AhdddBEHHHDABpfr7my33XbssMMODB06lG9+85ssX76chx9+OJrvJz/5CUOHDqVf\nv34MHTqUq6++Orr/oYce4oADDqBfv358+MMf5tZbb6WhoYHHHtNleUWy6OWXX2bw4MG97pAF2bi9\n9tqLlpaWtMtIVZcHJ7j7TAAzewF4yN2rcwZTqTvtTdsDDzzAO++8w7nnnsupp57KPffcA8DPfvYz\nvve973HNNdfwkY98hObmZn7wgx+wxx57dPk5Vq5cycyZM/n/7d17fBT1vf/x1ydECOGqVRRQRMEf\nFxUFvBy8VEUFirqoteDRioLl1B5RQUWrPQJW1ILXilKtN0Br1GMVvKDgrVREDkoQRY1cBLEWUFAR\nCCCQ7++P2WTz3YSQy94m+34+HvuAzM7OfuadzeSTmfnOmFnZwIHSZY8dO5YHHniAI488koULFzJs\n2DCaNm3KRRddxKZNm4hEIpx55pkUFBTw5ZdfctVVV+kXgkgG27Jli+4XXE81adIkVHelSYbanOO2\nAmhtZu0qeyS6QEmMRx99NN0lVOr1119n8eLFFBQUcOSRR3L00UfzxBNP8I9//IMFCxYAcP/99zNs\n2DAGDx5Mx44duemmmzj88MOrtfzjjjuOZs2a0bRpU+666y6OOuooTj31VAoLg2sjjh07lrvuuosB\nAwZw4IEHcvbZZzNixAgeeughAJ588klycnL461//SufOnenbty+jRo1KThhplKmfj3RQFr4w5rFz\n505yc2tzmdLdK912SCDVeeTm5rJz586UvmemqU3jtpKgedvVQzJQpm5sioqKOOCAA2jTpk3ZtC5d\nutCyZUs+++wzAD7//HOOPvpo73XHHHNMtZb/7LPP8uGHH/L8889zyCGH8Pjjj9OgQQNWr15NcXEx\ny5cv59JLL6VZs2Zlj3HjxvHFF18AwQWGu3Xr5u2lq+57h0mmfj7SQVn4lIdv9erVCVnOKaecwtVX\nX52QZaVTovKQ6qtN49ad4J6kpY9jCc5xWwL8KnGlSSI98MAD6S6hUs65Sg87xk+Pn6e6t2rbf//9\n6dChAwMGDODWW2/l7LPPZvv27Zxxxhls2rQJCM5xW7RoUdnjk08+4b333ttlfTW9TVwYZOrnIx2U\nha++5jFkyBBycnJo0KABDRs25OCDD+b666/f7bUXzzjjjBRVmBxTpkwpW+/SUfcNGjTgscceq9Xy\nEpVHOq4+EFY1btycc4viHh845x4GriW4X6lItXXt2pVVq1bx9ddfl0379NNP2bBhA127dgWgU6dO\nzJ8/33vdBx98sNtlxzdc5513Hrm5uUyaNAmAVq1a0bZtW5YvX87BBx/sPQ488EAAOnfuzEcffeTd\nlP7999+v3cqKSEb5xS9+wZo1a1ixYgX33nsvDz30UCgvf1JTLVq0YM2aNWWP1atXc+GFF9Z6eTt2\n7EhgdbI7tdnjtiufA0fvdi7JSj/88IO3V2vRokX861//4rTTTuPwww/nwgsvZOHChcyfP5+LL76Y\nU045pWzU6BVXXMEjjzzC1KlTWbZsGePGjeOjjz7a7QCByvaMXXnlldx+++1s3boVCM5xu/3225k4\ncSJLly5l8eLFTJ48mXvuuQeACy64gJ07dzJs2DCKioqYOXMmd911FxC+i+yKiK9Ro0bss88+tG3b\nlkgkwmmnncbrr7/uzfP73/+eTp060aRJEzp06MDo0aO9c6xuvvlmunfvzpNPPslBBx1Ey5Yt+c//\n/E82b95cNk9xcTGDBw+mWbNmtG3blrvvvrtCLT/88AODBw9mr732okmTJvTv359ly5aVPT9lyhT2\n3HNPXnnlFTp37kyTJk0YOHBg2cCrgw46iL322ourrrpqt0cFzIx99tmHVq1alT3K3wHnL3/5Cx07\ndqRRo0Z06dKFJ5980nt9Tk4ODz74IAMGDKBp06bcdtttACxevJj+/fvTrFkz9ttvPwYPHsz69evL\nXvfcc8/RrVs38vPz2XvvvenTpw9btmzh5ptvZsqUKWUj9hs0aMA///nPKtchm9W4cTOz5nGPFmbW\nGRgHLE18iVIfzJ49mx49eniPP/4xuDva9OnT2XPPPTnppJPo06cPHTt25Omnny577QUXXMCNN97I\nqFGj6NmzJ19++SWXXHLJbu+PWlljdfHFF7Nz507uv/9+AC699FIeeeQRHn/8cbp168bJJ5/MlClT\nykasNmvWjJdffplFixbRvXt3brrpJsaMCW4QovuzitQfixcvZu7cud75rADNmzdn6tSpfPbZZ9x3\n33088sgjZX/YlVq+fDnTp09nxowZvPLKK8yePZs//elPZc9fe+21vPPOO7z00kvMmjXLG3xV6uKL\nL6awsJCXX36ZefPm4Zyjf//+XpNYXFzMxIkTefbZZ5k5cyZvv/0255xzDq+99hqvvvoqTz75JA89\n9BDPPfdcrXN44YUXGDFiBKNGjeKTTz7hv/7rvxgyZAizZ8/25rv55ps599xzWbx4MUOHDmXDhg2c\neuqp9OzZk8LCQmbOnMk333zDwIEDAVizZg0XXHABv/nNbygqKmL27Nmce+65OOe49tprGThwIP36\n9WPt2rWsXr2a4447rtbrUO8552r0AEqAnXGPEuBLoFdNl5eMB8G5d27BggVOAmeddVZSlvvqq6+6\nBx54ICnLrsrpp5/uBg8eXOvXP/XUU7V+7ZNPPukaNWrktm7dust51qxZ48aMGeO++uqrWr9PKiXr\n8xFGysIXxjzuuOMON3v27CrnueSSS1xubq5r2rSpy8vLc2bmcnNz3QsvvFDl6y688EJ39NFHl309\nduxY17RpU7d58+ayadddd53r1auXc865TZs2uUaNGrm///3vZc9/9913Lj8/340cOdI559ySJUuc\nmbl58+aVzbN+/XqXn5/vnnvuOeecc5MnT3Y5OTluxYoVZfNcdtllrmnTpq64uLhsWr9+/dzvfve7\nXdY/efJkZ2auWbNmrmnTpq5p06audevWZc8ff/zx7rLLLvNeM3DgQHfmmWeWfW1m7pprrnHOxbal\n48aNc/369fNe99VXXzkzc0uXLnWFhYUuJyfHrVq1qtK6LrnkEnfOOefssu5SCxYscGPGjNntfOmw\nYMECR3CTgh4uiT1ObcZLnxL3dQnwLbDMOacD3Rlq+PDh6S6h1rZs2cKDDz5I3759ycnJoaCggDff\nfJM33nij1susycjQJ554goMPPpi2bdvy4Ycf8vvf/55BgwZ5hxbCLsyfj0RTFr76nEfv3r158MEH\n2bRpE/fccw977LEHZ599tjfPM888w8SJE1m+fDmbNm1i+/bttGzZ0punffv23o3WW7duXXYR8eXL\nl7N9+3Zvm7PnnnvSqVOnsq+LiorYY489vHn22msvOnXqVDa6HiA/P5/27duXfb3vvvvSvn17Gjdu\n7E2Lv4B5vObNm7Nw4cKyQ6o5ObGDb5999hm//e1vvfmPP/547rvvPm9az549gdi2dNGiRbz11ls0\na9bMm8/MWL58Oaeffjq9e/fmsMMOo2/fvvTp04fzzjuvQpayezVu3Jxzs3c/l2SaPn36JGW5ZkZJ\nSUlSll3+PWbMmMGtt97Ktm3b6NSpE88//zynnBL/N0T1dejQodrzrlmzhtGjR7N27Vpat27NoEGD\nGDduXJWvKc2k/AYxkyXr8xFGysJXn/No0qQJBx10EBBcr+6II47g8ccfZ8iQIQDMmzePX//619xy\nyy306dOHFi1aUFBQUOEctfgb2JffLpY2R1WdE1s6T2XTy7+usvep6r13JScnp2y9K1PZSPr4aaUX\nOC7dlpZerHzChAkV1qd169bk5OTw+uuv89577zFr1iwmTpzIH/7wB+bPn182GEyqp1a/Vcysk5nd\nb2Zvmtkb0f93TnRxkvny8vKSfhXrvLw8Xn/9ddatW8fGjRv54IMPGDBgQFLfs7xRo0axYsWKsuu+\n3Xnnnbs9v600E50HJxIOZsaNN97IH/7wh7LBS3PnzqV9+/b8/ve/p0ePHnTo0IGVK1fWaLkdO3Yk\nNzeXefPmlU37/vvvvRuld+3alR07dni341u/fj1LliwpG12fKl26dGHOnDnetLlz59KlS5cqX9ej\nRw8++eQTDjzwwAqj9MvvEezVqxdjxoxh4cKFNGzYkBdeeAGAhg0bZv2FdaurNoMTfgksBnoCi4CP\nCM4p+zj6nGSRfffdl82bN7Nu3bp0l5JRVq5cSV5eHi1atEh3KSJSTb/61a9o0KBB2bXrSu+L+cwz\nz/DFF19w3333MW3atBots0mTJlx66aWMGjWKt99+m8WLFzNkyBAaNGhQNk/Hjh2JRCIMGzaMd999\nl0WLFvHrX/+aAw44gEgkktB13J1Ro0YxefJkHnroIZYtW8bdd9/NCy+8sNs7xlx++eV89913nH/+\n+XzwwQd88cUXzJw5k6FDh+KcY/78+dx+++0sWLCAr776ir///e+sW7eurDFt3749H330EUuWLGH9\n+vW6xEgVarPHbQJwu3Oul3Pu6ujjOOC26HOSgWq6samuDh060LBhQz755JOkLD9ZioqKkrZs5xyf\nfvopnTt39jbOmSxZn48wUha+bMqjQYMGDB8+nDvuuIPi4mLOOussRo4cyRVXXEH37t2ZN29ehfO/\nquOOO+7gxBNPJBKJ0KdPH0488cSyc8RKTZ48mZ49e3LWWWdx/PHHk5OTwyuvvJLybciAAQP485//\nzJ133slhhx3Gww8/zOTJkznxxBPL5il/2LR0W9q6dWveffddSkpK6Nu3L926dePqq69mzz33xMxo\n3rw5//znPznjjDPo1KkTo0eP5u677y47FD9s2DA6derEUUcdRatWrZg7d25K1ztUajqaASgGOlYy\n/RCgOJkjKWpQo0aVxhk4cGDSlv3cc8+5O+64w61bty5p75Fozz77bNKW/f7777sxY8a4JUuWJO09\nEi2Zn4+wURa+MOZRnVGltZXMbUcYpToPjSp1tdrj9g/gxEqmnwC8U4vlSQo888wzSVt23759ady4\nMZMnT2blypW7PNE2k/zqV4m/O9u2bduYO3cuL7/8MscccwwdO3ZM+HskSzI/H2GjLHxhzSNZ26Fk\nbDvCLNV5hOH3S7LV5nIgLwLjzawnUHq25X8Q3Kd0jJmVHZB3zr1Y9xIl0zVt2pSLL76YqVOnMnny\nZJo3b06nTp1o3rx5hRFP9Y1zjm3btrF27VqWLl3Kjh07OPbYY+nXr5/urCCSJrm5ud5t6qT+2L59\ne73/vbI7tWncJkX//e/oo7LnINhdGI4TfKTOmjZtyu9+9ztWrVrFp59+yvLlyykuLq73G08zo1Gj\nRrRs2ZLevXvTtWtXXZdIJM2aNGnChg0b0l2GJMGGDRu8a+Zlo9pcxy0cF6aSlPvmm2948MEH6dat\nW72+aKeIZLYOHTowf/58du7cGZoBQrJ7zjmKiopCdRpKMqgJyxKlF5RMhrVr13Lttddy0EEHMWnS\npFBcdDaZeYSR8ohRFr4w5nHooYeydetWli5N/O2zs2mUbXWkMo+vv/6a77//nkMPPTRl75mJanOo\nFDM7muDWV62Ia/6cc1cnoC5JsGRc/Xzt2rXccccdTJo0iT322INRo0YxYsQI9txzz4S/V6LV56vB\n14byiFEWvjDm0apVKw488ECmT59OixYtaN26dcKWXZO7rmSDVOWxfv16nn32WfbZZx/vtl/ZyGo6\nQsPMbgTGAZ8DawnOZSvlnHO9E1de7ZhZD2DBggUL6NGjR7rLqXfiG7YRI0aEpmETkeywdetWnnji\nCdavX0/Pnj3p2rUrbdq00aChEHHO8c033/DJJ59QWFhIXl4eF198cYX7oWaKwsLC0uvz9XTOFSbr\nfWqzx+0qYKhzbnKCa5EMF+Y9bCKSXfLy8rjooot44403+PDDD3n33Xdp2LAhjRs31nlvIVBSUsLW\nrVvZunUreXl5dO7cmVNPPTVjm7ZUqk3jVgK8m+hCJP2+++47Zs6cyfnnn+/9VaqGTUTCKC8vjzPP\nPJP+/fvz5Zdfsnr1arZu3brbm7DXVElJCc45NYQJZGbk5eXRqlUrDj74YGVbXk2v2AtcB9ybzKsC\n1/WB7pxQwTvvvFPl8zt27HCnnXaaa9u2bdm0NWvWuGuuucY1btzYNW/e3I0ePdp99913yS41JXaX\nR7ZRHjHKwqc8fOXz2L59u5s6dao75JBD3Omnn57GqtJHn4+YTL5zwp1AJzNbbmYvmdnz5R+Jaigl\nsSZMCG4ju30LbFob/Fve7bffzptvvsnkyZO9UaIPP/ww1157LStWrODmm2+uN3vZSvOQgPKIURY+\n5eGbMGECO3bs4IknnqBr164MHjyYrl27cs8996S7tLTQ5yP1ajM44X7gN8DbVBycgHMu7WPHNTih\nos/fKObDSfl8Ph1cCVgOdBoAva6BL7b/g1NPPZURI0ZgZmWHRK+66ipGjBjBXnvtle7yE664uDjr\nL+JYnvKIURY+5RGzY8cOJk+ezIQJE1i6dCkDBgxg9OjRWf17Rp+PmEwenHAx8Evn3CuJLkaS4/2/\nwIzL88lpEDRtEPy75CX44IW1PNpsEG3btmXSpEk0bNiQa6+9tt42bKW0ofEpjxhl4VMeQcNWUFDA\nLbfcUtawPf3001ndsJXS5yP1anOo9DtgeaILkeRYNQdmXA44KNnhP1eyAx7jBL7f+A2rV6+he/fu\nXHzxxWzbto0JEyawZMmStNQsIpIJKjskumDBAqZNm6amTdKmNo3bWOBmM1ObHQLv3Q05uxiMU0IJ\n2/iRxuzFXo0O4Ntvv+XVV1/l+eef5/nnn0/KVcdFRDLJ0qVLOeaYY/jyyy/Lpqlhk0xWm8btSuAX\nwFoz+9jMCss/Elyf1MH2LfD59GDP2ixGVXg+hxxGsZbrWc/lW5bz6UdLWbo0eCxZsoQzzjgjDVWn\nxqhRFfPIZsojRln46nMeW7duZeDAgXz//ffsvffeFRq2Ll26VGjY6nMetaE8Uq8257jpRm0hse3H\n2DltLWhX5byuJJh/j8YpKCwDtGtXdR7ZRnnEKAtfvchjyxb48Udo3hwaxzZyV199NZ999hlz5szh\n+eefr9Y5bPUijwRSHqlX41GlYaBRpYHtW+D2prHmrSqWAzdsyp7GTUSywJw5cPfdMH06lJRATg4M\nGADXXMMz//oX559/PpdccgnvvvuuRolKnWXyqFIAzKwn0IXgciCfOucWJqwqSYg9GgeX/FjyUsWB\nCeXl5AbzqWkTkXrjL3+Byy+HBg2Cpg2Cf196ic9eeIFL9tiDpk2bMnnyZI0SlVCp8TluZtbKzN4C\n3gfuA+4HFpjZm2a2T22KMLMTzexFM/vazErMLBL3/OPR6eUfM2rzXtmm19VQsrPqeUp2wn+MTE09\nIiJJN2dO0LQ5Bzvi/mrdsYM+wNbt2ynevJlGjRoxa9YsTjjhBPLz87n22mvTUrJIddVmj9tEoDlw\nqHPuMwAz6wpMIWjk/rMWy2wCfAg8Bvx9F/O8ClwClN5Ec1st3ifrtDsBzpgEk39XxL65nb09bzm5\nQdN2xiRod3z6akyHoqIiOnfunO4yMobyiFEWvlDmcffdwZ62+KYt6lZgsRn7d+sGQ4d6z/385z+v\nctGhzCOJlEfq1ebOCRuA05xz78dNPwaY5ZxrWaeCzEqAs51zL5ab9jjQwjl3bjWXoXPc4px+QoRh\n+71I0QuxOyd0PifY05ZtTRtAJBLhxRdf3P2MWUJ5xCgLX+jy2LIFmjaNHR6tSk4ObNrkDVjYndDl\nkWTKIyaTz3HLAbZXMn07tbu8SHWdbGZrge+Bt4D/cc59l8T3q1cefep+2rULBixs+xEaNc/uc9ru\nv//+dJeQUZRHjLLwhS6PH3+sXtMGwXw//lijxi10eSSZ8ki92jRabwF/NrM2pRPMrC1wD/BmogqL\n8yowGOgNXAecBMwwM6vyVVKmdMj2Ho2h6b7Z3bSBhrDHUx4xysIXujyaNw/2pFVHTk4wfw2ELo8k\nUx6pV5vGbTjQDFhpZsvNbBmwIjrtikQWV8o596xz7mXn3CfRQ6hnAscAJ1f1uv79+xOJRLxHr169\nmDbNvxTdrFmziEQiFV5/+eWX8+ijj3rTCgsLiUQirFu3zps+ZswYxo8f701btWoVkUiEoqIib/rE\niRMrXLSwuLiYSCTCnDlzvOkFBQUMGTKkQm2DBg3Semg9tB5aD61H/Hp8+y2RVq0oauDfMmYi+Jch\nz82l+KyziAwalJnrUV++H/V0PQoKCohEIhx++OF07NiRSCTCyJGpGeVX6+u4mdnpQGeCwQKfOufe\nSEhBlZzjtov5vgH+4Jx7uJLndI6biEi2mjMHfv7zYFTprpjBO+/A8Vl4kq8kRarOcav1OWnOuded\ncxOdc/clqmmrLjPbH/gZsDqV7xtm8X89ZDvl4VMeMcrCF8o8TjgBJk0KmrPcuFO5c3OD6ZMm1app\nC2UeSaQ8Uq/ajZuZ9TazT82swgkBZtbCzD4xsxNrU4SZNTGzI8zsyOikg6NfHxB9boKZHWtmB5rZ\nqQS33VoCzKzN+2Wj4uLidJeQUZSHT3nEKAtfaPO47LJgj9qAAbFz3krvnPDOO8HztRDaPJJEeaRe\ntQ+VmtmLwNvOuXt28fyVwCnOuXNqXITZScDbBHdhKG8K8N8EjdqRQEvg3wQN22jn3Le7WJ4OlYqI\nSGAX9yoVSaRMvBzIEcD1VTw/C6jVJaedc7Opeu9fv9osV0REhMaN1bBJvVGTc9z2pfLrt5XaAdTq\nllciIiIisns1ady+Bg6v4vluaLBAxoof7pztlIdPecQoC5/y8CkPn/JIvZo0bjOAP5pZXvwTZtYY\nuBl4OVGFSWINjbsfX7ZTHj7lEaMsfMrDpzx8yiP1anKO2zjgXGCJmd0PfE4wmKALcDnQgODeQbO3\n9AAAIABJREFUvZKBxo4dm+4SMory8CmPGGXhUx4+5eFTHqlXowvwmtmBwF+AvgQX3oWgeZsJ/Ldz\nbmWiC6wNjSoVERGRVMrEUaU4574E+pvZnkBHguZtqXPu+2QUJyIiIiIxNWrcSkUbtfcTXIuIiIiI\nVKHWt7yScIm/uW62Ux4+5RGjLHzKw6c8fMoj9dS4ZYnCwqQdbg8l5eFTHjHKwqc8fMrDpzxSr0aD\nE8JCgxNEREQklVI1OEF73ERERERCQo2biIiISEiocRMREREJCTVuWSISiaS7hIyiPHzKI0ZZ+JSH\nT3n4lEfqqXHLEsOHD093CRlFefiUR4yy8CkPn/LwKY/U06hSERERkTrSqFIRERER8ahxExEREQkJ\nNW5ZYtq0aekuIaMoD5/yiFEWPuXhUx4+5ZF6atyyREFBQbpLyCjKw6c8YpSFT3n4lIdPeaSeBieI\niIiI1JEGJ4iIiIiIR42biIiISEiocRMREREJCTVuWWLIkCHpLiGjKA+f8ohRFj7l4VMePuWRemrc\nskSfPn3SXUJGUR4+5RGjLHzKw6c8fMoj9TSqVERERKSONKpURERERDxq3ERERERCQo1blpgzZ066\nS8goysOnPGKUhU95+JSHT3mknhq3LDFhwoR0l5BRlIdPecQoC5/y8CkPn/JIPQ1OyBLFxcXk5+en\nu4yMoTx8yiNGWfiUh095+JRHjAYnSELpB8unPHzKI0ZZ+JSHT3n4lEfqqXETERERCQk1biIiIiIh\nocYtS4waNSrdJWQU5eFTHjHKwqc8fMrDpzxST41blmjXrl26S8goysOnPGKUhU95+JSHT3mknkaV\nioiIiNSRRpWKiIiIiEeNm4iIiEhIqHHLEkVFRekuIaMoD5/yiFEWPuXhUx4+5ZF6atyyxHXXXZfu\nEjKK8vApjxhl4VMePuXhUx6pp8EJWWLVqlUa/VOO8vApjxhl4VMePuXhUx4xGpwgCaUfLJ/y8CmP\nGGXhUx4+5eFTHqmnxk1EREQkJNS4iYiIiISEGrcsMX78+HSXkFGUh095xCgLn/LwKQ+f8kg9NW5Z\nori4ON0lZBTl4VMeMcrCpzx8ysOnPFJPo0pFRERE6iirRpWa2Ylm9qKZfW1mJWYWqWSeP5rZv82s\n2MxeN7OO6ahVREREJF0yonEDmgAfApcDFXYBmtn1wHDgt8AxwGZgppk1TGWRIiIiIumUEY2bc+41\n59xo59w0wCqZ5SrgFufcS865xcBgoA1wdirrDLN169alu4SMojx8yiNGWfiUh095+JRH6mVE41YV\nMzsI2A94s3Sac+5H4P+AXumqK2yGDh2a7hIyivLwKY8YZeFTHj7l4VMeqZfxjRtB0+aAtXHT10af\nk2oYO3ZsukvIKMrDpzxilIVPefiUh095pF4YGrddMSo5H668/v37E4lEvEevXr2YNm2aN9+sWbOI\nRCqMh+Dyyy/n0Ucf9aYVFhYSiUQq7B4eM2ZMhevZrFq1ikgkQlFRkTd94sSJjBo1yptWXFxMJBJh\nzpw53vSCggKGDBlSobZBgwbVaD0WLlxYL9YjUd+P8qONw7we5dVlPUrzCPt6lKrLevTo0aNerAck\n5vvx7rvv1ov1SNT3Y9WqVfViPRL1/Si/LQ3zepRXnfUoKCggEolw+OGH07FjRyKRCCNHjqzwvsmQ\ncZcDMbMS4Gzn3IvRrw8ClgNHOuc+KjffP4CFzrkKSelyICIiIpJKWXU5kKo451YAa4BTS6eZWXPg\nWGBuuuoSERERSbWMaNzMrImZHWFmR0YnHRz9+oDo1/cC/2NmZ5nZ4cBU4F/A9HTUG0bxu4iznfLw\nKY8YZeFTHj7l4VMeqZcRjRtwFLAQWEBw3tpdQCFwM4BzbgIwEXiIYDRpY+AXzrmf0lJtCBUWJm2v\nbSgpD5/yiFEWPuXhUx4+5ZF6GXeOWyLoHDcRERFJJZ3jJiIiIiIeNW4iIiIiIaHGTURERCQk1Lhl\nicouSJjNlIdPecQoC5/y8CkPn/JIPTVuWWL48OHpLiGjKA+f8ohRFj7l4VMePuWRehpVKiIiIlJH\nGlUqIiIiIh41biIiIiIhocYtS0ybNi3dJWQU5eFTHjHKwqc8fMrDpzxST41bligoKEh3CRlFefiU\nR4yy8CkPn/LwKY/U0+AEERERkTrS4AQRERER8ahxExEREQkJNW4iIiIiIaHGLUsMGTIk3SVkFOXh\nUx4xysKnPHzKw6c8Uk+NW5bo06dPukvIKMrDpzxilIVPefiUh095pJ5GlYqIiIjUkUaVioiIiIhH\njZuIiIhISKhxyxJz5sxJdwkZRXn4lEeMsvApD5/y8CmP1FPjliUmTJiQ7hIyivLwKY8YZeFTHj7l\n4VMeqafBCVmiuLiY/Pz8dJeRMZSHT3nEKAuf8vApD5/yiNHgBEko/WD5lIdPecQoC5/y8CkPn/JI\nPTVuIiIiIiGhxk1EREQkJNS4ZYlRo0alu4SMojx8yiNGWfiUh095+JRH6qlxyxLt2rVLdwkZRXn4\nlEeMsvApD5/y8CmP1NOoUhEREZE60qhSEREREfGocRMREREJCTVuWaKoqCjdJWQU5eFTHjHKwqc8\nfMrDpzxST41blrjuuuvSXUJGUR4+5RGjLHzKw6c8fMoj9TQ4IUusWrVKo3/KUR4+5RGjLHxhzWPb\ntm18/vnnLFmyhI0bN/LTTz8lZLnfffcde+21V7Xnz83NpXHjxhx00EF07dqVFi1aJKSOTBHWz0cy\npGpwgho3ERGpN5xzzJo1i/nz57Nz507atm3Lz372Mxo2bIiZpbye7du3s3HjRr788kt27NjBwQcf\nzMCBA8nLy0t5LZJcqWrccpO1YBERkVRyzjFjxgw++OADTjrpJLp3754xe7i2bdtGUVERr732GlOn\nTmXw4MFq3qRWdI6biIjUCwsXLuT9998nEolw8sknZ0zTBtCoUSOOOOIIBg8ezPfff8/LL7+c7pIk\npNS4ZYnx48enu4SMojx8yiNGWfjClMfHH39Mhw4d6N69e9LeY86cOXV6fevWrTnhhBP4/PPPE3be\nXTqF6fNRX6hxyxLFxcXpLiGjKA+f8ohRFr6w5LFp0yZWrlzJoYcemtT32b59e52X0bVrV7Zv386S\nJUsSUFF6heXzUZ9ocIKIiITekiVLeOqppxg5cmRGHSLdlYkTJ3LIIYfQr1+/dJciCaJbXomIiFTT\nli1bAMjPz09zJdWTn5/P1q1b012GhJAaNxERCb2SkhIguG5aGOTm5rJz5850lyEhpMYtS6xbty7d\nJWQU5eFTHjHKwqc8fDqny6fPR+qpccsSQ4cOTXcJGUV5+JRHjLLw1bc8hgwZwrnnnutNe+6552jc\nuDH33nsvQ4YMIScnhwkTJnjzTJ8+nZycHKZPnw7A7NmzycnJoVu3bsSfK77nnnsyderU5K5Ihqhv\nn48wUOOWJcaOHZvuEjKK8vApjxhl4avveTzyyCNcdNFFPPTQQ4wYMQKAxo0bM378eDZs2ODNa2ac\nfPLJ3rTly5dnTZNWmfr++chEatyyhEbX+pSHT3nEKAtffc5jwoQJXHnllTzzzDMMHjy4bPppp53G\nfvvtx2233VbhNa1bt/a+vuKKKxg9enS9uCZbbdTnz0emUuMmIiJZ54YbbuDWW29lxowZRCIR77kG\nDRpw2223MXHiRP7973/vchlmxogRI9ixYwf3339/sksWAdS4iYhIlpkxYwYTJkxg+vTpFQ59lhow\nYABHHnkkY8aMqXJZ+fn5jBkzhttuu42NGzcmoVoRnxq3LPHoo4+mu4SMojx8yiNGWfjqYx5HHHEE\n7du3Z/To0WzevHmX840fP54pU6ZQVFRUNq2wsOJ1VS+99FL23nvvrLz9U338fGQ6NW5ZorKNTTZT\nHj7lEaMsfPUxj7Zt2zJ79my+/vpr+vbty6ZNmyqd78QTT6Rv377ccMMNZdNWr15dYb4GDRowbtw4\n/vznP1f6fH1WHz8fmS4UjZuZjTGzkrjHp+muK0weeOCBdJeQUZSHT3nEKAtffc1j//33Z/bs2axd\nu5Z+/frtcs/b7bffzksvvcTcuXMBOOOMMyqd77zzzuPQQw/l5ptvxsySVnemqa+fj0wWisYtajGw\nL7Bf9HFCessREZEwK23evvnmG/r27VvpOWqHHXYYF154IRMnTqzwXPz1226//XYee+yxKg+/itRV\nmBq3Hc65b51z30Qf36W7IBERCbc2bdrwj3/8g3Xr1tGvX79Km7dbbrkF51yFPWnxX59yyin07t2b\nHTt2JLVmyW7huKlb4BAz+xrYCrwH3OCc+yrNNYmISIg8/vjjFaa1adPGG4AQr127dmU3sS910kkn\nVXqv0ddee63uRYpUISx73OYBlwB9gcuAg4B/mlmTdBYVJvHXKcp2ysOnPGKUhS9secQfvky0goKC\nhCwn2XWmStg+H/VBKBo359xM59zfnXOLnXOvA/2BPYGBVb2uf//+RCIR79GrVy+mTZvmzTdr1qxK\nP3yXX355haHOhYWFRCKRCjfWHTNmTIWh4KtWrSISiVT4S27ixImMGjXKm1ZcXEwkEmHOnDne9IKC\nAoYMGVKhtkGDBtVoPTp27Fgv1iNR34/hw4fXi/Uory7rUZpH2NejVF3WY/jw4fViPSAx348DDjgg\nFOuxxx57AMFNzwsKCirM+3//93/MmjXLm7Z9+3YKCgpYtWqVN/3jjz+uUC/A//7v/1a4c8Ly5csr\nbeZeeeWVCiMuV69eTUFBAcXFxfz00080bNiwwnqUCsvnqvy2NMzrUV511qOgoIBIJMLhhx9Ox44d\niUQijBw5ssL7JoOFtes3s/nA6865P1TyXA9gwYIFC3Q7DhGRLLBixQqmTJnCZZddxn777Zfucqrk\nnOPOO++kZ8+e9O7dO93lSIIUFhbSs2dPgJ7OuaRdJyUUe9zimVlToAOQXRfMERGRSh1wwAE0atSo\nynPVMsWqVavYvHlzhSMhItURisbNzO4ws5+b2YFmdhzwArADSMzJBiIiEmq5ubl07tyZxYsXVzpo\nIJN89NFHNG/enAMOOCDdpUgIhaJxA/YHngKKgKeBb4H/cM6tT2tVIVLZ+RrZTHn4lEeMsvBlch7L\nli1j6NChZfcTPeqoo/j+++957rnnkta81XWP3nvvvceCBQs49thj68WFejP581FfhaJxc879p3Nu\nf+dcY+dcO+fcBc65FemuK0wSNRKqvlAePuURoyx8mZjHsmXLuOSSS+jcuTOvvfYaRx55JBAcLh00\naBBLlizhkUceYd68eWzYsCGh7/3xxx/X+DXbtm3j448/5umnn2bmzJmceOKJHHfccQmtK10y8fNR\n34V2cEJVNDhBRKT+WbZsGePGjePJJ5+kVatW3HDDDQwbNoy8vDxvvpUrV/Lee++xbNkydu7cSW5u\nLg0bNkz5Hi7nHDt27OCnn34Cgjs1dO/enR49etSLvW3iS9XghDBdgFdERLJQfMN2zz33VNqwlWrf\nvj3t27dn69atfPHFF2zcuLGseUq13NxcGjduTPv27WnZsmVaapD6RY2biIhkpJo2bPHy8vLo2rVr\nkqsUSS01biIiklHq2rCJ1GehGJwgdVfZVaezmfLwKY8YZeFLZR7lBx3MmjWLe+65hy+++IIrrrgi\nY5o2fT58yiP1tMctS/Tp0yfdJWQU5eFTHjHKwpeKPMK0h02fD5/ySD2NKhURkbSo7ihRkTDQLa9E\nRCT0brnlFi644AJvWhgOiYpkKjVuIiKSFK+88gqjR4/mqKOOAtSwiSSCGrcsMWfOnHSXkFGUh095\nxCgLX3Xy2FpSwrodO9haUlI27auvvmLw4MGcddZZnHXWWfWmYdPnw6c8Uk+NW5aYMGFCukvIKMrD\npzxilIWvqjwWbN3KVWvWcPTKlZy0ahVHr1zJVWvW8H8bN3L++eeTl5dHkyZN6NKlS+gbtlL6fPiU\nR+ppcEKWKC4uJj8/P91lZAzl4VMeMcrCt6s8nv7xR8atW0cOUP527g2Ar268kR+efZYcM1q1asX1\n11/Pb37zGxo3bkyDBg1SVXpS6PPhUx4xGpwgCaUfLJ/y8CmPGGXhqyyPBVu3Mm7dOhx+0wawcdEi\nfnjmGXCOkpIS1qxZw8iRI2nWrBn5+fkUFibt91lK6PPhUx6pp+u4iYhIjUz94YcKe9pK5R16KHtf\ndRWN9t2Xrk2acGGLFrHn8vI49NBDU1anSH2kxk1ERKpta0kJbxUXU7KL53Nyc2l15ZUArAYGtW9P\nXo4O7ogkin6assSoUaPSXUJGUR4+5RGjLHzxeWwqKdll0xavJDp/faLPh095pJ4atyzRrl27dJeQ\nUZSHT3nEKAtffB5Nc3Kq/YsjJzp/faLPh095pJ5GlYqISI1ctWYNbxcXV3qOW6kGQO/8fO7db79U\nlSWSVhpVKiIiGWlwy5a7PVxaEp1PRBJLjZuIiNRIz7w8btp7b4xgz1p5DQADbtp7b3qE9CK7IplM\njVuWKCoqSncJGUV5+JRHjLLw7SqPQc2bM7VNG3rn55f9IskhODw6tU0bBjVvnrIaU0mfD5/ySD01\nblniuuuuS3cJGUV5+JRHjLLwVZVHj7w87t1vP95v357Z7drxfvv23LvffvV6T5s+Hz7lkXoanJAl\nVq1apdE/5SgPn/KIURY+5eFTHj7lEaPBCZJQ+sHyKQ+f8ohRFj7l4VMePuWRemrcREREREJCjZuI\niIhISKhxyxLjx49PdwkZRXn4lEeMsvApD5/y8CmP1FPjliWKi4vTXUJGUR4+5RGjLHzKw6c8fMoj\n9TSqVERERKSONKpURERERDxq3ERERERCQo1blli3bl26S8goysOnPGKUhU95+JSHT3mknhq3LDF0\n6NB0l5BRlIdPecQoC5/y8CkPn/JIPTVuWWLs2LHpLiGjKA+f8ohRFj7l4VMePuWRehpVKiIiIlJH\nGlUqIiIiIh41biIiIiIhocYtSzz66KPpLiGjKA+f8ohRFj7l4VMePuWRemrcskRhYdIOt4eS8vAp\njxhl4VMePuXhUx6pp8EJIiIiInWkwQkiIiIi4lHjJiIiIhISatxEREREQkKNW5aIRCLpLiGjKA+f\n8ohRFj7l4VMePuWRemrcssTw4cPTXUJGUR4+5RGjLHzKw6c8fMoj9TSqVERERKSONKpURERERDxq\n3ERERERCQo1blpg2bVq6S8goysOnPGKUhU95+JSHT3mkXqgaNzO73MxWmNkWM5tnZkenu6awGD9+\nfLpLyCjKw6c8YpSFT3n4lIdPeaReaBo3MxsE3AWMAboDi4CZZrZ3WgsLiX322SfdJWQU5eFTHjHK\nwqc8fMrDpzxSLzSNGzASeMg5N9U5VwRcBhQDQ9NbloiIiEhqhKJxM7M9gJ7Am6XTXHAdkzeAXumq\nS0RERCSVQtG4AXsDDYC1cdPXAvulvhwRERGR1MtNdwF1ZEBlVxDOA/jss89SW00Gmz9/PoWFSbse\nYOgoD5/yiFEWPuXhUx4+5RFTrufIS+b7hOLOCdFDpcXAL51zL5abPhlo4Zw7J27+C4C/pbRIERER\nEbjQOfdUshYeij1uzrntZrYAOBV4EcDMLPr1fZW8ZCZwIbAS2JqiMkVERCR75QHtCXqQpAnFHjcA\nMxsITAF+C8wnGGV6HtDZOfdtOmsTERERSYVQ7HEDcM49G71m2x+BfYEPgb5q2kRERCRbhGaPm4iI\niEi2C8vlQERERESynho3ERERkZAITeNWkxvMm9nbZlZSyeOlcvM0MbP7zewrMys2s0/M7LepWZu6\nqUkW0flHmFlRdD1XmdndZtaoLsvMJInOw8xuMLP5Zvajma01sxfM7P8lf00SIxmfj3Lz3hD9Wbo7\nOdUnXpJ+XtqY2RNmti463yIz65HcNUmMJPy85JjZLWb2RXSeZWb2P8lfk7qr4e+VXDMbHV2/LWa2\n0Mz61mWZmSbReWTTtrS6n49y89d+W+qcy/gHMIjgsh6Dgc7AQ8B3wN67mL8l0KrcoyuwHbio3Dx/\nBZYAJwLtgGHRec5M9/omOIsLgC3R17UDTgO+Bu6s7TIz6ZGkPGYAFwFdgMOBlwkuLdM43eubjjzK\nzXs08AWwELg73euaxs9HS2AF8AjBrfgOjM53ULrXN0153Ah8A/SLznMu8CMwPN3rm+AsxgNfAX0J\nLvlQer/sI2q7zEx6JCmPbNqW7jaPcvPWaVua9nCqGeA84M/lvjbgX8B11Xz9COCH8h8W4GPgD3Hz\nfQD8Md3rm8gsgInA63HT7gT+mah861selbxmb6AEOCHd65uuPICmwOdAb+Dt2mxs6ksewJ+A2ele\ntwzK4yXg4bh5ngOmpnt9E5zF18BlVa1nlm1Ld5tHJa+pz9vSauWRiG1pxh8qtcTcYH4oUOCc21Ju\n2lwgYmZtou9zCnAISb5wXl3UMou5QM/SXbxmdjDQH3ilDsvMCMnIYxdaEtxa7bsElJ00Sc7jAeAl\n59xbia47WZKYx1nAB2b2bPTwT6GZ/SYZ65BIScxjLnCqmR0SnecI4HiCvS0ZqZZZNAK2xU3bApxQ\nh2VmhGTksQv1eVta3TzqvC0Nw3XcqrrBfKfdvdjMjgEOBYbEPXUFweHSf5nZDmAnMMw5926dK06e\nGmfhnCuw4Pp3c8zMoq9/0Dk3vrbLzCDJyMMTnedeYI5z7tOEVZ4cScnDzM4HjgSOSkrVyZOsz8fB\nwO+Au4BbgWOB+8xsq3PuyQSvQyIlK48/Ac2BIjPbSXDu9B+cc08negUSqDbbvZnA1Wb2DrCc4LDx\nucTOFc+qbSm7z8NT37elVCOPRG1LM36PWxV2dYP5eJcCi51zC+KmX0mwwT0T6AFcA0wys94JrTI1\ndpmFmZ1McA7KZUB3gg/SmdU4ebi6+WaiROYxieAcyfMTX2bK1DoPM9ufYGP7a+fc9pRUm3x1/Xzk\nAAucczc55xY55/4KPEzQzIVRXfMYRHAu3PnReS4GRpnZRUmsOVmq2u5dBSwFigj2rNwHPEbwR39t\nl5npEplHvd6Wsps8zOwAErQtDcMet3UEK75v3PRWVOyGPWbWmGCj8j9x0/MI/lIe4Jx7LTp5sZl1\nB64FMvVwUG2y+CPBMfbHo19/YmZNCfY2jqvlMjNFIvN4iCCPMmZ2P8FhoROdc6sTVnXyJCOPnsA+\nwILoX8wQ/CX6czMbDjSKHkLIRMn4eQFYDXwW97rPCJqaTJasPCYAtznn/rfcPO2BG4AnElN6wtU4\nC+fcOuBcM2sI/Mw5t9rM/kQwUKVWy8wgycijTDZsS6uRRw8StC3N+D1u0c609AbzgHeD+bm7efkg\noCHwt7jpe0Qf8SGV7ubPSLXMIp/gZNDySkpfW8d80yrBeVi5H6bSDc0A4BTn3KpE1p0sScrjTYLR\nYEcCR0QfHwBPEoyWytSmLSk/L9Gv36Xi4ZJOwJd1LDmpkphHPhW3pSXUv21p6Wt/iv5S3gP4JTCt\nrstMt2TkUW452bItLX3trvJ4g0RtS+sy6iJVD2AgwUl+5Yflrgf2iT4/leAvvvjXvQM8tYtlvg18\nBJxEMHT3EoKhu/+V7vVNZBbAGIIRtYOi63k6we7cp6q7zEx+JCmPScD3BJeK2bfcIy/d65uOPCp5\njzCNKk3G5+MogkMhNwAdCA4TbgTOT/f6pimPx4FVBHtUDgTOIbg8SIVtciY9apHFMdF1Oyi6bXgD\nWAY0r+4yM/mRpDyyaVu62zwqeY9abUvTHk4NQvxvguu/bAHeA44q99xbwGNx8x9CsAet9y6W1wp4\nlOC6K5uBT4Gr0r2eic6C4K/emwiuWbc5+rr74j9MVS0z0x+JzoNgb8HOSh6D072u6fp8xC3/rdps\nbOpTHgRNykcEf+x9AgxN93qmKw+gCXA3wSGhzQSN3c1AbrrXNcFZ/Dz6vS4maEwfB/aryTIz/ZHo\nPLJsW1qtz0fc8mu1LdVN5kVERERCImPPQRARERERnxo3ERERkZBQ4yYiIiISEmrcREREREJCjZuI\niIhISKhxExEREQkJNW4iIiIiIaHGTURERCQk1LiJiIiIhIQaNxGREDKzx83s+UxZjoikhho3kXrI\nzPY2s7+Y2ZdmttXMVpvZq2bWK921hZGZdYg2OF9F81xuZk+ZWc9011ZdZnagmZWYWbe4p64ELklD\nSSJSC7npLkBEkuJ5gp/viwhu/r0vcCrws3QWBWBmuc65Hemuo7rM7CjgDeBjYBjwOdAMGADcCZxS\ny+Xu4ZzbXt3pCWBAhZtTO+c2JuG9RCRJtMdNpJ4xsxbACcD1zrl/Oue+cs594Jwb75x7udx8JWZ2\nmZnNMLPi6F6kX8Yta38ze8bMvjezdWY2zcwOLPf8UWY2y8y+NbMfzOwfZtY9bhml7zPdzDYCN5rZ\nSdHpfcysMPr+b5jZPmb2CzP71Mw2mNnfzCyv3LL6mtk75ep5ycwOLvd86V6lc8zsLTPbbGYfmtl/\nlJunnZm9aGbfmdkmM/vYzPpVEelk4HPn3InOudeccyuccx85524haN5Kl3uYmb0ZXZd1ZvaQmTUp\n9/zjZvaCmd1oZl8DRdHpK8zsf8xsipn9ADxUnewr+b5XmQ3wRfTfD6MZvRV93eTyh0rNrKGZ3Wdm\na81sS3SZR5V7vvR719vM3o9m/K6ZHVJFhiKSIGrcROqfTdHH2WbWcDfz/hH4X6Ab8DfgaTPrBMGe\nMWAmsAE4PvrYCLwWfQ6CPU+To88dCywBZpRvWKLGEOwFPBx4LG76fwO9gHbAswSH7s4H+gN9gCvK\nzd8EuAvoCfQGdgIvVLJe44AJwBHRmp4ys9Lt3SSgIUFzexhwPUFeFZjZkUDX6HtW4Jz7MTpfY+A1\nYH20tvOA04CJcS85Ffh/0efOLDf9GuBDoDtwSzWzj7e7bI4h2OvWG9gPOLd0NeKWcwdwDsHe2u7A\nMmCmmbWMm28cMDL6fjvwv68ikizOOT300KOePQh+8a4DioE5wK3A4XHzlAD3x017r3RTEhXkAAAE\nc0lEQVQa8Gvg07jnGwKbgdN28b45BM1G/7j3uTNuvpMIGouTy027PjrtwHLT/gLMqGI994kuv2v0\n6wOjX19Sbp4u0eX+v+jXi4Cbqpnjr6KvPWI38w2L5p1XbtovCBqafaJfPw78G8iNe+0K4Lm4aRfu\nLvvo8p6vRTbd4uYrWw6QD2wDBpV7Phf4F3BNFd+7X0SnNUz3Z18PPer7Q3vcROoh59wLQBvgLOBV\ngl+2hWY2OG7WeXFfv0fQ6ECwF+4QM9tY+iDYo9QI6ABgZq3M7GEzWxI9zLeBYM9Pu7jlLthFqR+X\n+/9aoNg592XctFalX5hZx+iggOVmtoHg8J+r5P3KL3c1wZ6m0uXcB9xkZnPMbKyZHb6L2oi+Dio5\nNyxOZ2CRc25ruWnvEjSyncrX5So/vy8+nyPYTfYVCq1+NlXpQNCozS2dEK13PrHPRdm6lPv/6ui/\nrRCRpNLgBJF6yjn3E/Bm9HGrmT0M3AxM3d1Lo/82BT4ALiDWwJT6NvrvVGBPgsOZqwj21swj2DtU\n3uZdvFf5k/Bd3Nel08r/gfkywR6q3xDsvcoBPqnk/eKXS+lynHOPmtlrwBkEh2J/b2bXOOceqKS+\nJQTr3gX4aBfrALs48T/u/WHXOcRPr0728aqbTVV21ahWtn67zFhEkkc/ZCLZ4zOCvWHl/UclXxdF\n/18IHAJ865z7Iu5ROhLxOOA+59xM59xnBL/M905G8Wa2F8H5YeOcc2875z6n8lGyu9s7hnPua+fc\nX51z5wF3ExzqrGy+D4FPgWvMLL6BKh0IQnSeI6PnupU6geDw4ZLd1VOJ6mRfvo7qZPNT9N8GVbzv\nMoLv4Qnllp0LHEWwjiKSZmrcROoZM9srOrrxQjM73Mzam9mvgFHAtLjZf2VmQ8zsEDO7GTgauD/6\n3N8IztuabmYnRJdzspn92czaROdZClxkZp3N7FjgSYLz6qpVag1X7XuCw4X/ZcF11XoTnIxf2d6h\nXb+p2T0WjGZtb2Y9CC7nUVVTMoSgKfqnBSNeD4rmeiOxPP8GbAWmmNmhZnYKwSHZqc65Xe0hq0p1\nsi+vOtl8A2wB+kUPcTePX4hzrpjgvMI7oqNUuwKPAI3xBx9UlnFNv58iUgtq3ETqn00EhytHALMJ\nzkW6meAyE1fEzTuGYATnIoLBCOc754oAnHNbgJ8THAL9O0Fz8zDBeVY/Rl8/lOBQaSEwBfgzQYNQ\nXnUOIe6Wc84BgwhGMX5M0JhcW83llp/WgKA5/RSYQbCH8fIq3vd9gj1Oy4C/Rl83nWC06YjoPFuA\nvsBeBOeDPQu8TsW8K32LSt6zOtmXn3+32Tjndkbr+S3wNRWb+FK/j77nVILDtQcDfZxzG6qqeRfT\nRCTBLPh5F5FsY2YlwNnOuRfTXYuIiFSP9riJiIiIhIQaN5Hspd3tIiIho0OlIiIiIiGhPW4iIiIi\nIaHGTURERCQk1LiJiIiIhIQaNxEREZGQUOMmIiIiEhJq3ERERERCQo2biIiISEiocRMREREJif8P\nIIgVdaMPmaIAAAAASUVORK5CYII=\n","text/plain":["<matplotlib.figure.Figure at 0x7f3c80b00e80>"]},"metadata":{"tags":[]}}]}]}