{"cells":[{"metadata":{},"cell_type":"markdown","source":"# Setup"},{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"import numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\nfrom gplearn.genetic import SymbolicRegressor,SymbolicTransformer\nfrom gplearn.functions import make_function\n\nfrom sklearn.metrics import mean_absolute_error\nfrom sklearn.preprocessing import StandardScaler\n\nfrom sklearn.ensemble import RandomForestRegressor\nfrom sklearn.feature_selection import RFE\nfrom sklearn.feature_selection import RFECV\n\nimport os\n\nprint(os.listdir(\"../input\"))\nprint(os.listdir(\"../input/LANL-Earthquake-Prediction\"))\nprint(os.listdir(\"../input/lanl-features\"))","execution_count":155,"outputs":[{"output_type":"stream","text":"['LANL-Earthquake-Prediction', 'lanl-features']\n['test', 'train.csv', 'sample_submission.csv']\n['test_features_denoised.csv', 'submission_1.csv', 'train_features_denoised.csv', 'train_features.csv', 'y.csv', 'test_features.csv']\n","name":"stdout"}]},{"metadata":{},"cell_type":"markdown","source":"# Read Data"},{"metadata":{"_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","trusted":true},"cell_type":"code","source":"X = pd.read_csv('../input/lanl-features/train_features_denoised.csv')\nX_test = pd.read_csv('../input/lanl-features/test_features_denoised.csv')\ny = pd.read_csv('../input/lanl-features/y.csv')\nsubmission = pd.read_csv('../input/LANL-Earthquake-Prediction/sample_submission.csv',index_col='seg_id')","execution_count":156,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Scaling"},{"metadata":{"trusted":true},"cell_type":"code","source":"X.drop('seg_id',axis=1,inplace=True)\nX_test.drop('seg_id',axis=1,inplace=True)\nX.drop('target',axis=1,inplace=True)\nX_test.drop('target',axis=1,inplace=True)\n\nalldata = pd.concat([X, X_test])\n\nscaler = StandardScaler()\n\nalldata = pd.DataFrame(scaler.fit_transform(alldata), columns=alldata.columns)\n\nX = alldata[:X.shape[0]]\nX_test = alldata[X.shape[0]:]","execution_count":157,"outputs":[{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/sklearn/preprocessing/data.py:645: DataConversionWarning: Data with input dtype bool, int64, float64 were all converted to float64 by StandardScaler.\n  return self.partial_fit(X, y)\n/opt/conda/lib/python3.6/site-packages/sklearn/base.py:464: DataConversionWarning: Data with input dtype bool, int64, float64 were all converted to float64 by StandardScaler.\n  return self.fit(X, **fit_params).transform(X)\n","name":"stderr"}]},{"metadata":{},"cell_type":"markdown","source":"# Feature Selection"},{"metadata":{},"cell_type":"markdown","source":"## Drop highly correlated features"},{"metadata":{"trusted":true},"cell_type":"code","source":"%%time\ncorr_matrix = X.corr()\ncorr_matrix = corr_matrix.abs()\nupper = corr_matrix.where(np.triu(np.ones(corr_matrix.shape), k=1).astype(np.bool))\nto_drop = [column for column in upper.columns if any(upper[column] > 0.95)]\n\nX = X.drop(to_drop, axis=1)\nX_test = X_test.drop(to_drop, axis=1)\nprint(X.shape)\nprint(X_test.shape)","execution_count":158,"outputs":[{"output_type":"stream","text":"(4195, 334)\n(2624, 334)\nCPU times: user 9.1 s, sys: 36 ms, total: 9.14 s\nWall time: 9.14 s\n","name":"stdout"}]},{"metadata":{},"cell_type":"markdown","source":"## Recursive feature elimination with cross validation and random forest regression"},{"metadata":{"trusted":true},"cell_type":"code","source":"%%time\nrf = RandomForestRegressor(n_estimators=10)\nrfecv = RFECV(estimator=rf, step=1, cv=5, scoring='neg_mean_absolute_error', verbose=0, n_jobs=4) #4-fold cross-validation with mae\nrfecv = rfecv.fit(X, y.values)\nprint('Optimal number of features :', rfecv.n_features_)\nprint('Best features :', X.columns[rfecv.support_])\n\nX = X[X.columns[rfecv.support_].values]\nX_test = X_test[X_test.columns[rfecv.support_].values]\nprint(X.shape)\nprint(X_test.shape)","execution_count":159,"outputs":[{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/sklearn/utils/validation.py:761: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n  y = column_or_1d(y, warn=True)\n","name":"stderr"},{"output_type":"stream","text":"Optimal number of features : 13\nBest features : Index(['abs_percentile_80', 'autocorrelation_1000', 'autocorrelation_10000',\n       'autocorrelation_5', 'autocorrelation_5000',\n       'av_change_rate_roll_std_500', 'ffti_av_change_abs_roll_std_100',\n       'ffti_classic_sta_lta8_mean', 'ffti_skew', 'ffti_spkt_welch_density_10',\n       'fftr_mean_change_rate_last_1000', 'num_crossing_0',\n       'percentile_roll_std_10'],\n      dtype='object')\n(4195, 13)\n(2624, 13)\nCPU times: user 5min 43s, sys: 308 ms, total: 5min 43s\nWall time: 12min 2s\n","name":"stdout"}]},{"metadata":{},"cell_type":"markdown","source":"Give training data some insight of the time range of this experiment. Little bit cheating"},{"metadata":{"trusted":true},"cell_type":"code","source":"X[\"mean_y\"] = np.full(len(y), y.values.mean())\nX[\"max_y\"] = np.full(len(y), y.values.max())\nX[\"min_y\"] = np.full(len(y), y.values.min())\nX[\"std_y\"] = np.full(len(y), y.values.std())\n\nX_test[\"mean_y\"] = np.full(len(X_test), y.values.mean())\nX_test[\"max_y\"] = np.full(len(X_test), y.values.max())\nX_test[\"min_y\"] = np.full(len(X_test), y.values.min())\nX_test[\"std_y\"] = np.full(len(X_test), y.values.std())\n\nprint(X.shape)\nprint(X_test.shape)","execution_count":160,"outputs":[{"output_type":"stream","text":"(4195, 17)\n(2624, 17)\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"list(X.columns)","execution_count":184,"outputs":[{"output_type":"execute_result","execution_count":184,"data":{"text/plain":"['abs_percentile_80',\n 'autocorrelation_1000',\n 'autocorrelation_10000',\n 'autocorrelation_5',\n 'autocorrelation_5000',\n 'av_change_rate_roll_std_500',\n 'ffti_av_change_abs_roll_std_100',\n 'ffti_classic_sta_lta8_mean',\n 'ffti_skew',\n 'ffti_spkt_welch_density_10',\n 'fftr_mean_change_rate_last_1000',\n 'num_crossing_0',\n 'percentile_roll_std_10',\n 'mean_y',\n 'max_y',\n 'min_y',\n 'std_y']"},"metadata":{}}]},{"metadata":{},"cell_type":"markdown","source":"# Define More Genetic Functions"},{"metadata":{"trusted":true},"cell_type":"code","source":"def tanh(x):\n    return np.tanh(x)\ndef sinh(x):\n    return np.sinh(x)\ndef cosh(x):\n    return np.cosh(x)\ndef arctan(x):\n    return np.arctan(x)\ndef arcsin(x):\n    return np.arcsin(x)\ndef arccos(x):\n    return np.arccos(x)\ndef arctanh(x):\n    return np.arctan(x)\ndef arcsinh(x):\n    return np.arcsin(x)\ndef arccosh(x):\n    return np.arccos(x)\ndef exp(x):\n    return np.exp(x)\ndef exp2(x):\n    return np.exp2(x)\ndef expm1(x):\n    return np.expm1(x)\ndef log2(x):\n    return np.log2(x)\ndef log1p(x):\n    return np.log1p(x)\n \n\ngp_tanh = make_function(tanh,\"tanh\",1)\ngp_sinh = make_function(sinh,\"sinh\",1)\ngp_cosh = make_function(cosh,\"cosh\",1)\n\ngp_arctan = make_function(arctan,\"arctan\",1)\ngp_arcsin = make_function(arcsin,\"arcsin\",1)\ngp_arccos = make_function(arccos,\"arccos\",1)\n\ngp_arctanh = make_function(arctanh,\"arctanh\",1)\ngp_arcsinh = make_function(arcsinh,\"arcsinh\",1)\ngp_arccosh = make_function(arccosh,\"arccosh\",1)\n\ngp_exp = make_function(exp,\"exp\",1)\ngp_exp2 = make_function(exp2,\"exp2\",1)\ngp_expm1 = make_function(expm1,\"expm1\",1)\n#gp_log2 = make_function(log2,\"log2\",1)\n#gp_log1p = make_function(log1p,\"log1p\",1)","execution_count":161,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Define Symbolic Regressor and Train"},{"metadata":{"trusted":true},"cell_type":"code","source":"%%time\nest_gp = SymbolicRegressor(population_size=X.shape[1]*17*10,\n                           tournament_size=X.shape[1]*17//1,\n                           generations=50, stopping_criteria=1.79,\n                           p_crossover=0.9, p_subtree_mutation=0.0001, p_hoist_mutation=0.0001, p_point_mutation=0.0001,\n                           max_samples=0.8, verbose=1,\n                           function_set = ('add', 'sub', 'mul', 'div', \n                                           'sqrt', 'log', 'abs', 'neg', 'inv','max', 'min', \n                                           'tan', 'cos', 'sin', \n                                           gp_tanh, #gp_sinh, gp_cosh,\n                                           gp_arctan, #gp_arcsin, gp_arccos,\n                                           gp_arctanh, #gp_arcsinh, gp_arccosh,\n                                           #gp_exp,\n                                           #gp_exp2,\n                                           #gp_expm1,\n                                           #gp_log1p,                                           \n                                          ),\n                           #function_set = (gp_tanh, 'add', 'sub', 'mul', 'div'),\n                           metric = 'mean absolute error', warm_start=True,\n                           n_jobs = 4, parsimony_coefficient=0.00001, random_state=11)\n'''\n\nest_gp = SymbolicTransformer(population_size=1000, \n                             hall_of_fame=100, \n                             n_components=10, \n                             generations=20, \n                             tournament_size=20, \n                             stopping_criteria=1.0, \n                             const_range=(-1.0, 1.0), \n                             init_depth=(2, 6), \n                             init_method='half and half', \n                             function_set=('add', 'sub', 'mul', 'div'), \n                             metric='pearson', \n                             parsimony_coefficient=0.001, p_crossover=0.9, p_subtree_mutation=0.01, p_hoist_mutation=0.01, p_point_mutation=0.01, p_point_replace=0.05, max_samples=1.0, \n                             feature_names=None, warm_start=False, low_memory=False, n_jobs=4, verbose=1, random_state=11)\n '''\n\nest_gp.fit(X, y)","execution_count":187,"outputs":[{"output_type":"stream","text":"    |   Population Average    |             Best Individual              |\n---- ------------------------- ------------------------------------------ ----------\n Gen   Length          Fitness   Length          Fitness      OOB Fitness  Time Left\n","name":"stdout"},{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/sklearn/utils/validation.py:761: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n  y = column_or_1d(y, warn=True)\n","name":"stderr"},{"output_type":"stream","text":"   0     7.85          10.3422       12          2.99863          3.00686      4.71m\n   1     6.45          9.62204       13          2.64898          2.65785      4.24m\n   2    11.30          6.87146       14          2.43764          2.45082      6.49m\n   3     8.82          8.22432        8           2.3151          2.39197      5.33m\n   4     5.57          3.20804       13          2.21993          2.34166      4.50m\n   5    11.37          3.60411       13          2.20611          2.39694      4.61m\n   6    12.70           3.7395       13          2.20725          2.39238      5.85m\n   7    13.20          3.72859       13            2.197          2.43336      4.79m\n   8    12.94          3.63324       13          2.20756          2.39114      4.66m\n   9    13.02          3.70504       13          2.20125          2.41636      4.66m\n  10    13.00          3.71016       13          2.20493          2.40164      5.49m\n  11    13.03          3.67089       13           2.2008          2.41815      4.44m\n  12    13.26          3.72003       13          2.20517          2.40069      4.25m\n  13    13.46          3.69442       13          2.19508          2.44104      5.05m\n  14    13.00          3.68617       13          2.20636          2.39592      4.05m\n  15    13.22          3.63944       13          2.19399          2.44542      3.90m\n  16    13.22          3.67819       13          2.19612          2.43689      3.85m\n  17    13.29          3.62714       14          2.20129          2.42217      4.48m\n  18    13.19           3.6484       13          2.19986          2.42192      3.62m\n  19    13.52          3.64707       14          2.19911          2.43089      3.63m\n  20    13.55          3.63566       14          2.17294          2.21594      4.13m\n  21    13.43          3.70717       13          2.14589          2.11637      3.26m\n  22    13.41          3.64599       13          2.09883          2.30463      3.17m\n  23    12.90          3.62167       13          2.10107          2.29564      3.62m\n  24    13.03          3.61258       13          2.09188          2.33242      2.85m\n  25    12.88          3.71397       13          2.08615          2.35534      2.74m\n  26    12.83          3.66457       13          2.06073          2.25495      2.64m\n  27    12.76          3.61216       13          2.05718          2.26915      3.08m\n  28    12.86          3.64938       13          2.05548          2.27595      2.38m\n  29    13.09          3.60942       13          2.05515           2.2773      2.26m\n  30    12.87          3.64634       11          2.05699           2.3007      2.15m\n  31    12.82           3.5953       13          2.04796          2.30603      2.51m\n  32    12.74          3.64886       13          2.05668          2.27115      1.98m\n  33    13.17          3.63978       13          2.05576          2.27484      1.85m\n  34    12.98          3.67578       13          2.05898          2.26199      2.09m\n  35    13.11          3.58816       13          2.05426          2.28087      1.62m\n  36    12.83          3.59027       13          2.06504          2.23771      1.80m\n  37    12.88          3.59726       14          2.06166          2.27205      1.39m\n  38    13.11          3.60225       13           2.0579          2.26631      1.56m\n  39    12.95          3.72879       13          2.05039          2.29633      1.16m\n  40    12.91          3.62552       13          2.05545          2.27607      1.04m\n  41    13.06          3.61749       13          2.05488          2.27839      1.14m\n  42    12.90          3.66952       13          2.05607          2.27361     48.58s\n  43    12.86          3.63806       14          2.05743          2.28896     43.21s\n  44    12.99          3.56615       13          2.04449          2.31994     35.64s\n  45    12.94          3.64746       13          2.06048          2.25597     33.80s\n  46    12.85          3.63296       13          2.05705          2.26968     20.71s\n  47    12.93            3.581       13          2.06393          2.24216     13.63s\n  48    12.93          3.56496       13          2.05521          2.27704      8.37s\n  49    12.75          3.60406       13          2.05252          2.28782      0.00s\nCPU times: user 4min 46s, sys: 1.45 s, total: 4min 47s\nWall time: 6min 4s\n","name":"stdout"}]},{"metadata":{},"cell_type":"markdown","source":"## Formula"},{"metadata":{"trusted":true},"cell_type":"code","source":"#print(\"gpLearn Program:\", est_gp._program)\ngenetic_formula = str(est_gp._program)\nfor i in range(len(X.columns)):\n    genetic_formula = genetic_formula.replace(f'X{i}', X.columns[i])\n    \nprint(\"Genetic Formula: \", genetic_formula)","execution_count":188,"outputs":[{"output_type":"stream","text":"Genetic Formula:  sub(add(sub(add(sub(autocorrelation_10003, autocorrelation_10001), -0.299), sin(autocorrelation_10001)), -0.299), sin(autocorrelation_10001))\n","name":"stdout"}]},{"metadata":{},"cell_type":"markdown","source":"## Prediction"},{"metadata":{"trusted":true},"cell_type":"code","source":"y_gp = est_gp.predict(X)\ngpLearn_MAE = mean_absolute_error(y, y_gp)\nprint(\"gpLearn MAE:\", gpLearn_MAE)","execution_count":189,"outputs":[{"output_type":"stream","text":"gpLearn MAE: 2.099577617011157\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_result = pd.DataFrame()\ndf_result[\"predict\"] = y_gp\ndf_result[\"real\"] = y","execution_count":190,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_result[:1500].plot()","execution_count":191,"outputs":[{"output_type":"execute_result","execution_count":191,"data":{"text/plain":"<matplotlib.axes._subplots.AxesSubplot at 0x7f36ef7fa4e0>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 432x288 with 1 Axes>","image/png":"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\n"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_result[-1500:].plot()","execution_count":192,"outputs":[{"output_type":"execute_result","execution_count":192,"data":{"text/plain":"<matplotlib.axes._subplots.AxesSubplot at 0x7f36f00e40b8>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 432x288 with 1 Axes>","image/png":"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\n"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_result.plot()","execution_count":193,"outputs":[{"output_type":"execute_result","execution_count":193,"data":{"text/plain":"<matplotlib.axes._subplots.AxesSubplot at 0x7f36ec17f940>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 432x288 with 1 Axes>","image/png":"iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzsfXmcHEXZ/7dmZ2fvIzuzx2yu3WSGkBMI4YjcIAivIOqLP0EOr1cQ71tUzld8QVQEARU0GJQbVFCQU4LhhgAJBBIIO7tJNptkryR7n1O/P7p7po+q7upjdmY2/f18YDYz3dXV3VVPPfUc34dQSuHDhw8fPvIfgWx3wIcPHz58eANfoPvw4cPHNIEv0H348OFjmsAX6D58+PAxTeALdB8+fPiYJvAFug8fPnxME/gC3YcPHz6mCXyB7sOHDx/TBL5A9+HDh49pguBUXiwSidCmpqapvKQPHz585D1ef/31bkpprdVxUyrQm5qasG7duqm8pA8fPnzkPQghW0WO800uPnz48DFN4At0Hz58+Jgm8AW6Dx8+fEwTTKkNnYXx8XG0t7djZGQk213JKRQXF2PWrFkoLCzMdld8+PCRJ8i6QG9vb0dFRQWamppACMl2d3IClFL09PSgvb0dzc3N2e6ODx8+8gRZN7mMjIwgHA77wlwFQgjC4bC/a/Hhw4ctWAp0QsjthJBOQshG3fdfJ4RsJoS8Qwi5zk0nfGFuhP9MfPjwYRciGvpqAKeqvyCEnADgTAAHUUoXA/il913zsV9g5wZg2yvZ7oUPH9MClgKdUroWQK/u64sBXEspHZWP6cxA3/IW5eXlAICOjg6cddZZpsfecMMNGBoamopu5SZuPRa4/ZRs98KHj2kBpzb0AwAcQwh5hRDyH0LIYV52KhcxOTlp+5zGxkY8+OCDpsfs9wJdweREtnvgw0few6lADwKoAXAkgO8DuJ9wjL6EkAsJIesIIeu6urocXi6zaGtrw4EHHohzzz0XCxcuxFlnnYWhoSE0NTXhhz/8IZYvX44HHngALS0tOPXUU3HooYfimGOOwebNmwEAra2tWLlyJZYuXYpLL71U0+6SJUsASAvC9773PSxZsgTLli3DTTfdhN/85jfo6OjACSecgBNOOCEr954z2CuU2ezDhw8TOA1bbAfwN0opBfAqISQJIALAILEppbcBuA0AVqxYQc0aveqf7+Ddjj6HXWJjUWMlrjhjseVx7733HlatWoWjjjoKX/jCF/Db3/4WABAOh/HGG28AAE466ST8/ve/RzwexyuvvIKvfOUreOaZZ/DNb34TF198MS644ALccsstzPZvu+02tLW1Yf369QgGg+jt7UVNTQ2uv/56rFmzBpFIxLubzieEKoCxfqDnAyA8P9u98eEjr+FUQ38IwAkAQAg5AEAIQLdXncoGZs+ejaOOOgoAcN555+H5558HAHz6058GAAwMDODFF1/Epz71KRx88MG46KKLsHPnTgDACy+8gHPOOQcAcP755zPbf/rpp3HRRRchGJTW0JqamozeT94gEpM+u7dktx8+fEwDWGrohJB7ABwPIEIIaQdwBYDbAdwuhzKOAfisrK27gogmnSnoLUbKv8vKygAAyWQS1dXVWL9+vdD5PgRRGpY+ez7Ibj98+JgGEIlyOYdSGqWUFlJKZ1FKV1FKxyil51FKl1BKl1NKn5mKzmYS27Ztw0svvQQAuPvuu3H00Udrfq+srERzczMeeOABAFI254YNGwAARx11FO69914AwF133cVs/+STT8att96KiQnJ+dfbKwUOVVRUoL+/3/sbyjf4At2HD9fIeqZormDBggW45ZZbsHDhQuzZswcXX3yx4Zi77roLq1atwkEHHYTFixfj4YcfBgDceOONuOWWW7B06VLs2LGD2f7//M//YM6cOVi2bBkOOugg3H333QCACy+8EKeeeqrvFPVNLj58uAbxwFIijBUrVlB9gYtNmzZh4cKFU9YHFtra2nD66adj48aN1gdPIXLh2WQcd/438MHT0t+XbAeKK7PbHx8+chCEkNcppSusjvM1dB+5A9/s4sOHK/gCHVJpvFzTzvcrFEqOZ1+g+/DhDr5A95FdUArUzANIwBfoPny4hC/QfWQfwSKgeo7vGPXhwyV8ge4jNxCOAz2+QPfhww18ge4jNxCOAT0tQDKZ7Z748JG38AW6B/jc5z5nyarowwKRGDA+BPR3ZLsnPnzkLXyBrgOlFElfS5x6hOPSp+8Y9aHH9leBHa9nuxd5AV+gQ0osWrBgAS644AIsWbIEf/nLX7By5UosX74cn/rUpzAwMAAA+N///V8cdthhWLJkCS688EJMZVLWtEdEFui+Y9SHHqtOBv5wYrZ7kRdwSp+bGTx2CbDrbW/bbFgKnHat5WFbtmzBHXfcgVgshk9+8pN4+umnUVZWhp///Oe4/vrrcfnll+NrX/saLr/8cgASq+IjjzyCM844w9v+7q+oiErx6L6G7oMHSgGfBM8UuSXQs4i5c+fiyCOPxCOPPIJ33303RaU7NjaGlStXAgDWrFmD6667DkNDQ+jt7cXixYt9ge4VCJH40H0N3QcPg91AeW22e5HTyC2BLqBJZwoKTS6lFCeffDLuueceze8jIyP4yle+gnXr1mH27Nm48sorMTIyko2uTl9E4kD7a9nuhY9cRc8WX6BbwLeh63DkkUfihRdewAcfSFv/wcFBvP/++ynhHYlEMDAw4Ee1ZALhOLB3OzDuL5Q+VCirkz793ZslLAU6IeR2QkinXMxC/9t3CSGUEDJt6qfV1tZi9erVOOecc7Bs2TKsXLkSmzdvRnV1Nb70pS9hyZIl+MhHPoLDDpv2dbGnCCrHciQu/bs3kbXe+MhBVM+WPv3EM0uImFxWA7gZwJ/VXxJCZgM4BcA277s1tWhqrMXGZ/8mJbUEAjjxxBPx2mvGrf/VV1+Nq6++2vD96tWrp6CX0xiKoyssl6Pr2QLUL8pef3zkFkiB9NnTkt1+5AFEKhatBdDL+OnXAH4AjYqVp+jrAMYGgbGBbPdk/4ZSJNrfWvtgwR8XlnBkQyeEnAlgB6V0g8f9yQ6CRdLn5Gh2+7G/o6hCCl/0Qxd9sLCnFZgcz3Yvchq2BTohpBTAjwFcLnj8hYSQdYSQdV1dXcxjsp6gUyAL9IncEehZfybZQjjma2I+2EhOAHu2ZrsXOQ0nGvp8AM0ANhBC2gDMAvAGIaSBdTCl9DZK6QpK6YraWmPIUXFxMXp6erIrwALyY5jIjegKSil6enpQXFyc7a5MPSIy6+L+uqD5YIACoQrpT98xagrbceiU0rcB1Cn/loX6Ckppt5MOzJo1C+3t7eBp71OC0X5geA8Q6AU6c2NLV1xcjFmzZmW7G1OPcBwY2QcM9QBl0yZ4yodbRGJAx5vS7m3BadnuTc7CUqATQu4BcDyACCGkHcAVlNJVXnWgsLAQzc3NXjXnDK/+AXjie9LfV+z104uzCTWniy/QfSgomQGUhn3/igUsBTql9ByL35s8600uYKATqKjPdi/2XyiRLj1bgLkrs9sXH7mFcNwX6BbwM0X18G102UX1XKAg5DtGfRjhO8wt4Qt0QOuA8wdMdhEokIpG+0kkPvSIxIDBTsnH4oMJX6Dr4W/pphasaJZwzN8p+TBCKYLS7c9RHnyBrkZ5g6+hZwU6J3QkDvS2ApMT2emOj9xExK9qZQVfoKsR8Z0uOYFwDEiOA3v9JBIfKsxolnhd/N0bF75AVyMcA/a0ARNj2e7J/o2wX47OBwPBEDBjrj8uTOALdDXCMYBOSkLdR/aQ2lr7E9cHtH6WcMzfRZvAF+hq+IIkN1BaA5TU+BPXhwoKxXJcioBKJrPbnRyFL9DVUPi4/S1d9hGJ+9EMPoyIxICJYaBvR7Z7kpPwBboaxdVAWa2voecCwnH/PfgwIuzvos3gC3Q9lC1drmJyXCrIMd0Rng8M7AZG+rLdEx+5hIgfi24GX6AD0Na1zPH04qevBK5fCAyxikhNI/j+DB8slNdLVLr+uGDCF+hqECLZ0Ye6JTrdXES7XOu0893s9iPTSG2tc3i35GPqQYi0e8tlpSuL8AW6HrmeXjztYrQ5hSxqmgESmEb36cMz+AmAXPgCXY9c3+rPaJI+p9OAZvHPB4sk5sVcfQ8+sodwHNi3HRgfznZPcg6+QNdjRhMQCOauZhgMSZ+52j8vEY7l7k7JR/YQkcOLfXOcAZYCnRByOyGkkxCyUfXdLwghmwkhbxFC/k4Iqc5sN6cQBYWSUM91DXh/0FyVrbWfRLKfQ2eW80MXuRDR0FcDOFX33VMAllBKlwF4H8CPPO5XdpEPlVF6E9Nf0IX9JBIfMtRmOaWqlb97M8BSoFNK1wLo1X33JKVU4TZ9GcD0qmYcni+nF09muyfm6J/m8eg+XaoPFkJlQOVMX0NnwAsb+hcAPOZBO7mDSByYHJUcL7mM6W5HD/sC3QcHPkkXE64EOiHkJwAmANxlcsyFhJB1hJB1XV1dbi6XOeir5uRy6KK6r9N9QFc0AKHy6b9w+bAPheuHVfFqP4ZjgU4I+RyA0wGcSyn/qVJKb6OUrqCUrqitrXV6ualFrocuAvtHjLaSRJLL78FHdhCOA6P7gMEcVRKzBEcCnRByKoAfAPgYpXTI2y7lAMpqgaKq3NaA9xfyqrDPuuiDgYjPjMqCSNjiPQBeArCAENJOCPkigJsBVAB4ihCynhDy+wz3c2pBSO5zuuwv2XIRP4nEBwO5Hro40pcVvqWg1QGU0nMYX6/KQF/E0L0FKA1LRRAyiXAMaHs+s9dwg3AM2PyoJOgKS7LdG+ewsoGGYwCoFKZZv3hKuuQjD1A1Cygoyl2l5uYVElvolfum9LL5lyl68wrghqWZv044LsU/jw1m/lpOoBZ0eQ9G6r+CyHTjrvHhCQIFMklXjgr0gd3S5xTniuSfQAeAsYHMXyOVXpyjA2Z/EXQ1chJJrm6tnYJS4N5zgfemV8RvRsDbxYVjuT8upjhXJD8FOpD5cKVcZzVUyuXl+oB2i6JyoKIxdzUxp6BJYPMjwD1nZ7sneQLGLi4Slwq6T45PeW8sEZTNoFMsP/JXoA/1ZLb98HwAJAcJgOSFTMmWm26CjoWIn0Tig4FwHEhOSEI915ClHX7+CnRPVz6Gtl9YAlTNzm0NeH/JllNCNP0kEh9q5LLZsbxB+ux+f0ovm78CPROCTM/LneuVURQb4nQXdJE4MLIPGOzOdk8yg9Ep8AlNRygkXbmo1CiyxDe5WKA0In1OheasxHrnqsCc7oJOwXT3F3gpkFafDvzuaO/ay2WUzJDkQS6PC1+gWyBQIH1Ohe04HJciavp3Zf5aTpDryRVeITzNswK9FOhtzwG73/auvVxHJEcziRUlsK99SkOf80+gKw9qSjT0HNcM95f05+o5chLJNL3PTLy/XM2f8Br5ELo4hYEV+SfQFfS2ApMT1se5Qa7Tt1bNnt6CTkGgAKiZl4MRRy6gYc3MwPubTs/KDJG4RNA1vDfbPTEiWCx9TqFjND8FekEISI4De7dm9jqVM6V40lzc0gHpbLm8nryC/olc59ZxAy/vq7hK+pzui7yCcA7XF62ZByn0eerkR34K9KnSnAOB3KdvzfVIHBHoo4tYCMeBPa25mUTiFj0t3jnea6ZjeTaTZ5PLfqRgsWQu9DV0C0yl7TicY5phauLLQnA6Czo1wjE5iSTDu7KpRkUUGB8E+nd6015hqfSZq2ZCp+At+jOaAFKQW3NUjUh8SvuWnwK9pEb6z6tV2Uw7isQl087EqDfX8hqR+PQUdHrkQ9ERJ8hUBM90e048BEOSUM/V+40cIC2uU0TSlZ8CHchQuBJDCwjHJd6NXEwvBnJ7y+klpmvooucx9rJysj+VZ8vV0EVAer/jQ1NG0pWHAl0epFMVrpTroYFKtlyu9s8rlNZIPPjTzZRQ2QgUlnkvkMb60xSu0x3hGNDbMuVUteaQ5dQU0xOIVCy6nRDSSQjZqPquhhDyFCFki/w5I7PdZCAckwbsSF/mrwPkrgY8XQUdC+HpVKVJ5QvJlON9ui/yCsIxYGJESuLJJRAimVyA3BHoAFYDOFX33SUA/k0pjQP4t/zvqQMhKptqhid4cRVQVpe7Wzpgmgk6E0zH0EUC7x1n1XOlz/1hTAA5TtJVD4QqpkwhtBTolNK1APTF8c4EcIf89x0APu5xv6wxlUk/kRwvyDwdBR0L4Rgw2Cnx10wnhGPA3m3eOd6rZkn5E/uLQM/lBEBF+Zyi0EWnNvR6SqkSZ7ULQD3vQELIhYSQdYSQdV1dXQ4vx0BNM0AC+2fooh7h+PQUdHqkio7k4MR1g3AcnpYTJIHpkZ8givI6oKgyd+93Cp22rp2ilFIKk8h/SultlNIVlNIVtbW1bi+XRrBI2lpOFevicG9WqngbIT9qdVxuZJoKOj2ma+hiJhzvuZ4Q5yUIyW1Ol0h8yki6nAr03YSQKADIn53edckGpqrAQy5v6YDc758ZKIVpkWg1ZjRLSST5eJ9m8NLxroQqhuNSbsLEmPs2sw2R8MtIPDfT/4EpnZ9OBfo/AHxW/vuzAB72pjs2obxE1+FKFgMml50uQDpbLlc1FK8QDAEz5ubue3CKogopY9TLHVYkDtBJKYt4WsBi0Q/HgH3bgbGhqemOFdSLUOQAKbBiCnb4ImGL9wB4CcACQkg7IeSLAK4FcDIhZAuAD8v/nhqoH5TXQfu89OLqOUAgmLsCc7oKOhZyIaLn+RuArS+6a0OvdXptMsjnXZsTKLuc3hzU0usXAd/fAsw/IeOXClodQCk9h/PTSR73xQZkwavWnKtmZe5yBYXSdj+XBWYuCLqpQDgGtK6VdmWBLOXFPX2F9HmlF05ohZMnBrz7kAftydhfEs4UqGVBw9Ls9iUFQVOih8jDTFEVpjx0MQdXfwWemZ9yHJEYMDGce0kkbhGJA8N7gMEeb9orqQbKanN3V+k1ajyqL9rbCvz547nJry6A/BboFQ1AqHzqQhd7E0ByMvPXcoLwNBV0eoRzyJ/hJVdKJjh5wjnMceI1QqVSwRe3Av2FG4HEGuDtB7zp1xQjvwU6UdKmp0hDnxyVEkByEalIiWk+gVOhizmwW/KSKyUToYuRHA7lywS8yBfJc1NVfgt0QLYdT4WGnmtOJp19bn+JRZ/iVGpTeJ2uHyj04L7UQQNxYKgnR/InpgBKGLObnVNFVPrMhfHlAPkv0CNxYO92YHw4s9fJFfpW3mDNJUGXSRCSO1QHXj5rpW6q16GLQG7sZqYCkTgw2gcMuEiLUSLdckZxs4f8F+jhGDxNm+ahLCIRdeWqwMwlQZdpZDuiJ1QhfXq9G4p4dF9EVc0KyN0x6zW8TNBybVrNDhd9Hgp03YPyIulH1npHJ5IYm+BEiRDimSAZGptAMunuhQ+PT2J0Queg9VDQdQ/kaIUmwLMkkvHJJPpGxkHtbtFTWpwbwcG4puJ4n5xw0S4wSYH+kXEpNyEQ9GSRH52YxPBYtgICRAuJe+ww94gsbXhsEpt39WFg1N17FUHeCfTRiSTWt+/DF1a/honJpCer8oQc6nfo1U/jgEsfMwpKBR6Q7Jz742tx9OUPYN6P/4UrHt6I79y33lE7y656EgsufVwrjCJxTwRd0yWPYsXVT+O/f2cveabpkkfRdMmj+MY9bwoLyeHxSWze3Y+/vSFF54xOTGLt+xYkbhFvkkjiP3kMy658Eqfd+Jyt8yblxXigY7Or6wPAs+9346+vy5FJkTiQHJdKHrrAyy09WHrlk3L+RJNrjfX5Ld1YcOnjWHj5447OX/t+V2ps2F48AYxNJPH05k40XfIo9gyaUBlUumeZVPfv5MtWO17Eduwdxhvb96Lpkkfx7s4+nHrDc1jXlgOZormGgdEJvNW+D89s7sRrbXuAUBlQOdOVnfD+ddpQvztebGMfGI5JWamjA84uRCnuCl2Dh0OXSdd5aSv+9uYOZ23JaOlS9UXx0LswP723qz/19+tb9zhq4x8bOoS1kY079mHf8Di+c/8GHHTVk/jlE+/hgttfxX/MhLoHoYsj4+mJunlXv6TRCmJwTLq34oF27BtwR7j0amsvvvvABukfHjjekyqB1Ds45j50sacFR981H0cH3gYAvNNhP5nqgttfTf393u5+rN++F6fesBZDY2Jj5L3d6TH566dNaGgDAddRb6tfaEv9PY90YOHljzvaTbd0puelohiFCjIvbvNKoOsf7Dl/eBmPb9zpmipUIxQB/N+/OJqXy6Iayuo/O+CcRnizSuACwIevX4uNO+RJ5oHN9E8vaLk/OvtGHLUjOgUmJtNH7hsexx+ek67/pTvW8U9SFi4Xi/iozrR29m0v2zq/h1YgSJK4cvUj+Ordb2DQi+20B473dapFePlPn5J2My7yJ2jnJgDAFwv+BQD46G+ed2V6ebt9H65+5F1s3tWPjTvsVxv780sWuxeXsuCNbennN59IDOEvtHQ7bk+NQCDzmaN5JdDf3G7UGL985xvp0EUPEz2YE9RlrPcrre63XP/YYOStOf2m56U/UjG0zjWUe1/brvn31l5n5hu1oHaCsUmTjNdQmbS9drFwXfe4dtF+p6NPszvhQVmUE1QKbxvs2IRH39qJU3691nFfAODUG9Zi6XXrgJIZ3icXTY5KpjgHmCyrAwDMIztT3zk1vQDA9x98K7XYv70jA/z94bhU0N0Dlslm+Z7PX/UqtvW4J/3qGcg882VeCfR3OjgreiQuFXcY9GYlBYBTfr3WuNWqmQeAONYA1OtNEOkFw8mWjnmGB4JOj6v++Y6j857ZnGFGZZcRPXe9YoxieHqTdaLQZJKCAEgkGwGkBd2OvcN46l3xRKOOvdow2827+tE/OuHaREL0IyOl9Ttr850OaZGbG3D+Po8LbMDJgfSO63154fzpI+86bpOLFMtkm6tmdtNqzAukF7Fb17oP/SzwNXQtuAp4BsKzduwdxs1rdJOgsASodp5eXFtRlPp7NkmbXSZt7Cwsh4THoYsbd/Th0ofetn1eY1Wx62v//U0TGgMvkkh0EJlwyrvqRwk6aTWaya7Ub7v2iedC8HYgrWh0HamkfiI9JXPkP5yNieFxtnnFjnPzjtDP8YfQ9al/97s0T5n6Z1zKAsUskkg2anYlLAXALgoLfIGuQVGQ091IZtLeX2N5pV1lpqYnQbNqsCQ95QTxXtDd+bL1YNaHe47wIoVsYPWLJvbSsAdJJDrsG7Z2jKofa4JGMS+QNoH1jYgLKt5Qvr+1GBjYBYzYty+z8MgHE0BRleNF/h/r0/dXhLTJ4JrH7Ef4BOANcdySK57g/xh2R9K1fE41AOndziADmAFv3gMAHNZc41lbPOSVQF/cWGXcUgIYL58JFBR5nlSzaSfDpqqwGjoSmOlz1Kv/ZtZ1bCLlqFIE3aCH9VsFoI9b/9rdb+Ib97yJyx7a6LjNCTM7esTDJBIZv3vWels9qTKPJZJRzXv8xRPvCV9LIcWkuj2XYptPepS0dMU/3wUiMYx3vo9b/9NiO2zw4FlVqb/Vu5Hb1tqPpJpFpmBMKiyTDmVBZZHEKK68B/X7dRogoKCiyJKt3DVcCXRCyLcJIe8QQjYSQu4hhLjfZwtAPyQ/cuMLkn3b4ap8QF0583tmck04BowNAP07jb9ZQD2X5pG05nPmLS/YbkuPpVfKWksmSJ4gxZibOQ2Duu3k0Ngk/rGhA3952XlM9Tsdfdg3xNGaM8St853712NbzxA33lltHkvQKMKkH1VIR0mJOs/uW8d2UiqCxMuFqpU2om/HJlzz2Ga81GKPnre2Mm0mVI9ZOxilhfL59ueMI7hIsFNMLi2KQFfZ0dduseej0y/WhFdAx0M4FuiEkJkAvgFgBaV0CYACAGd71TEWqCzKGyq160aiexDDVfMcCzF1nYT/Xm5RKMNFaJl6Ifp/Tc5W+9JQAfP7iSTFpp19rmyIVs7Zv7zcxv8xQ5nO23hRNlWzgWCx5wvX397YgWN/sQZH/N+/mb9T1aZBEb7zVYLu2F+swU4BW/oQxw68ldZjkjp3vLPw4NYShCe7UYIRjJrtehjQKiFagXzNvzYJtcHSdu2AJQb/uaGDny/iwo+kuFHaaS1oQUjT5xGOPyGX4NbkEgRQQggJAigF4FEtODaUwbWosRIzq0s0v63rD0v1EyfFE0QUEHnIUBAEdY4xQ9aoi1h09eQI7NFu7+97zb3TZXffiCtBN55M4lMFz2JlwH5ki96xe9Fx81J/O8kOtEQgIBU1cKmhf+PEGPN7ntNSr6EDWi0OEAtPo5wVcAyFaKe1nu48WmkDAGCeymTiBPr7vHVtQigmfQ8tl69vFA+sMFwW9Nru1+95E1f84x08+Q7jnsJxYKhbKhhiE8pb+fThTUDNPI1A/9mjYgtYRsa7IBwLdErpDgC/BLANwE4A+yilT3rVMTMQYiz/+dD2UiA5IVU6d4HvnnKA5t8de3WadEUjUFjqTKDLw2WIFiEw2IkKpLXPH/5VLJJEaYNljxubSLoSdBOTFL8ovA33hH7GvrbJOJ3UaffqrDj9b2oQQg2TVRguk0gA4Mj5YVvHS2GLUp+30zqM0QKNg1sUZs8yQaMgDgU660kmqBJi2WF7J6V22LM0bDsOfdb537jnTfTZyNLV46I7Xzd+6YJKWrmdLx0zDyQc0/SZF/HDayMbcGNymQHgTADNABoBlBFCzmMcdyEhZB0hZF1XlzuniPo5HdhQqfktbXu0/xKVQVldUoiq0kLNb798UufoUtKLnZhcZMG2PSBNMDeCgCUEb3h6C57f0u24f+MW23GzcaqvfDehEuKxnzyGr9zFmHhuEXGeRDKf7MBy8j6OaLYn0NXa1yQKsI3WGwSVyIQOlxdxf0vQRomnxqNygq20AUlKMI/stB9RJR++NVkn36f2fDtmYb2Gr+D8P75ifW6kjN091u24CV3UFaGfS3ahAM5MLeGykKPz3MCNyeXDAFoppV2U0nEAfwPwIf1BlNLbKKUrKKUramtrXVxODYIbzj5Y840bZ5LyDmsrilLmFwXMSAuXRTUqZy0E4NJJxJhI7+7sw3mrXkkLOpvmJ7UQLoS9WGG9oNDbiP/1trvtPhNh50kk/y76Pv6mBRDuAAAgAElEQVRWdKXtZA+9aamVRh29xyWNVdzfWmkDyPiQxBvkAPrFfhQhdCCM5sBOB9qjNP4TNIpKMoSIizC+BrIHZTD6Fza0e5wxqrBMOtpFKyBAJI4QmbQdnaO0EQqy/V2ZhBuBvg3AkYSQUiK5b08CIGZkcghK00O1XGdy2IdyjBfVYNABA57iaPzdeYdCP7+feGc33tyms8VF4hJfsk16TWUyDZXNASUBTQyzjVY0H0w4FHRqoTyHGLMeX0nwIyT03bnDinPDC0Q8SCizyUypmI+Okk01LTSKJrLbdoy1mZ21RTaReOkY1YdYikLppsZso4JeAbJCk007viN7tMIy6ej56So+wdlOmjg3JLqCGxv6KwAeBPAGgLfltm7zqF+OsH44go1vmZA6WSBaVYwAYw/5id/qaGTDMSncwSaroWL/TgZCoFVzDROMmcikb0NkfDvlhTaJaACAli53zIJO8NhGk8nkRR1VmxS8yvNXQtASNIoiMo6ZKi2O5/DUtCMfwzoykXRmPjQTfgl5J2FXQCpH8xzACy9/HC+bLPQKFMfofJvCMf28gfpKtpnq/d39xggth6GLmscjzyN1n7cLcBspz5gS4JPLZ9rugxu4inKhlF5BKT2QUrqEUno+pTSjVRE02yEG9Om69tsVZERzKEhSg4UApDZuGNwpXmwBFBeabOdS2XL2BDoFsI+WAgA+3Wx395EZT9CTZvwoLpJItkGK/LB7rqKhK+u+InztRpCYPa3dmAEaKnfcNxYSNIpyMoLCYXuFrZXXuoOGMUILmfPrTotcA0KA4fK5kh3f5q5UxOZ/yq/X4hY9TUckJiUA2mSZ1FyttAa9tFxzz8dct8ZWG5d+dJGt67tFXmWK6vHPrx2t+XeCRlFL9klEXXagGzTP/eAEwyGaRBOXNKcEAAnHMa9gN4hqqy7iYFIEZ01ZCOccPod5zEObh4DSiKP+9UMS6A3j9sIoM+XYt3wkDjWxbXCmBU9SKcolraEbTRHu1zaCzsJZtvum9oE8/q1jNL8p/Szt09IjWyG9kwiglTYwzQ8iCTOTBSG004hthUv0Ua7fvlf7hUOWSf31nPpIAGns1kyxYzSvBLp+olTrIlKUbeGN9z+GHzy4wfF1ZteUGr475KdPpf9RXAmUN7jYEkv1P4voKKJQm1msJ4ayg6gsLsQhs6uZx3zrvvWO6lOqn2/1sPu4eC+wpdO8mEh/+VxQBwvXCOSJ5vAdKkKsFxXYS8vsCyoLSfVyXw3GO02KOXDaJKAoDRUYo8DknURZvz2BrmmDI9wsRy2VjklQ+zvolMnF1llwzjKpezGJZNT2riIvwxazCUUh0CsGikBv3bxBU4Wob2QcD6/nVwayMuUwEYm70tAVh4vaJjkpEKamrPi/+czB5gcqJF02oLb91tgU6F4O4pMOrBM6bnvvEG7aABAHSSSp7to2a0ifacscMWhxYo8ifdQ3Toobfm2lDSjoawfGxRkclfdXX2lk4NiJGgzTELZteUu4PUAdJguUNS7AHNKpoX4GGMl3HCRoVNbwxQeLFPFP7cVHAo4TAKnaaA9pEaone1EOced5eh5NvVs0zwS6diDoHZjbaD0mqDF65Lv3b8A3712PLbs5XCROpFE45sBGrRossgah3qrboUWvqygxHy+RuETQNbzX5CBd/1TXL53Yg0oYtWPTmo4OURbSRiyxdkgsdA+MpkwJjjnEez7AXV88nPkTKy4/ZUNXPXyJddG5hl5ZbEwSSyQbESDUluPdLEdBMplEUbTPLqlWuqOHrzgShWQSc4iW4XLIJFtU7VuZGVuGMjKKeogvvunTzYWjQd6X1Uosky45cRQl0W50TraQVwJdL3f1L3EcQWyntYZtncKtweV2ttGHE3/1LK59bLMkMIf3AEPiVYg0/a9owGigRNNXofVc1cZRsQj/OAfkVUrTo4WSKYcVkfDTR3lFCZyp6ARAVYnWdKZ/r2bl3dwklAEARvtwVEMS82qNiSs3/dsoDLbvkTQ1NTVASzKKKOlFKWzw86gGA0ufUMih7JiTrPTCBG1wZcOmKSVEvA1KpWxgAGhecJB0vs3FTwy6uybEEaeL/l38+IIzADi4Z918mIp6okCeCXTAOr4zQRsNgshurKwZEl2D+P1/Whw5RlMTjkj/6y3Whi7u2Gu9vVYPFD2fjQYOQhcVbaq/vAkAexAPcDi/KQUa0KNhHhS8KkCAxY1pm6/+ff3MhARqG63DBAocRfSk0LMFvzn7EMMxv3nGuEj8VqbYTahCOJVdgkIve/nD1pTByvU/dvBMjTNTQatCo9tlz45uhgSNYjbptJc/oeoa5cRlP7elG2coZRC5pxOMVDUD0JKZWV5edf2Lj5svfB4ARw5z/ZuYG1uCSUowX7UI2Y3oevo7x+GFS060dY5T5JVA1z9GvWYHSIO2iezSRI94haZLHk3/I2yfj1s/EGqbl2i0lRcFqE2Fh1L1XIDYF3QAMFg6E5OkgOkM+s/7/Ky5l4u/jteKLrZ9PYDg3CPmpv+lW39bOgcMBTQUTCCI3QUNtjUxQilGA7Jpp+cDlAlyVRcQSRdUv8s0m6D0vN4SyHxMZSeXFzF9J0Moxk5aY6sQtpWgSSQbUUAo0CvuGNW0WVyFLlrJXOhF6oOOlTRgkBbZ0naTKSc08LmjmoXPAyBp6H07gDHx/AnDEwwWoZ3Wakyj1zy2WbBspDSQY3XlmmplmUReCfQU5BlfGmLYHmkUJWRMFz1iDtarOXVxg/lJ1XOBQKFNDV17pWDtAWhEj6YSjFf4+K2vSdlydkwucveSpBD7imdpChooGOUIVuXOQsQZ74VaiOv3U6+09uLb961nnCMduSMw05HJZV+oPsVMqWfZ5GHWDGkRKC9OKxNbaX2KK0WBqKMQhO3EBKQIi/4d4snXVuZmZeEZ2y1eiEMR6D86bSEI5DA+GyYTqgpTSTIcyJbnCx63kbWgOMkXYSyK+uie29YmjHHv5k1MGfJKoIs8qFZORpvp+U5eQEHQcVGN1HwLz0eAUKbgdIv12/ditHq+Y2eh3hxkB052R2oZxApoePTtndzEmR3BmY6SSEBIipnSLKGsd3As5SA9fZk0vhapTESjCGEHjWjG3IJLHze9tHpxP+tQNgd/K20A6dmCPQOjQklnVvNDEeh2wiGVJufUlIEQhULAhslE0z8qC0c7JhftTX3uQ03M43b1jRgrCoWdmx3VSNBGNOt2/a+aZHWngx+EL+sZ8kqgi6Alqd3+isHhkmozdDEVhaBILNnOre7r61vtczjzcMf7hbZY+xRnDiHAYPlcNJNdTI6S9xnRQup5MJPYq4ojuxRS+Cxn0n7xjteY3+8omCUnkYhn2qa6KzNTFpiExS3/6VOpHULaD2IsHWdrzKnmPC8xJ0EbUUWG8JO71uC7D2xAW7e16cDMxzSAUnTSagR67ezatMJJSt7rQyXEzBjq0ytLCpGgUcwi3QhBjDhO7+hlUXMoMBSPDs+XzrRhtmIhQaMo1UXnTEX1ISfIK4Eu4ozoQjX6aQk7AcLGO7jqzMXWB4VjUliZsGaoC3ljRA389+9eNJzlFAkaBSZGhLPl1PGzi5etQBEZx4cbjeagU3691uRcZ6XK1FQGillDj2ff09rvFUbHHQWShjtqw5SQgsxMGUiyzV7KmHvkLfPdihRjvQu2lQOTMalo1EqoIa/oRgpCO9gGbN74pmjv0iDSuE1nxort3lJx5CBYGK1EIhlFgFDM1ZG/8ea2K/NFYYlU8MWOn4vxHY/HhtuGb3KxB4uIVG5G2+4+C+++SuLz7JoahGNAchzYK8gsqH/RoTJ00BqbNknhQ9GatEcprG47WCcV+vjBCvEh0k+lqBueCcmMrVHReBSThgg+I/Nobx6vBwBc85d/Cp8LWcgozJSFfexEKr3PgJe5eNihR6CcjKAOe1XH8l+W3p+y+vOHGY5RQhdnJflJcaw2FR39wwvrjW0mo5gLsfaktiQo96sIN1EGQv0j0DuQFbAifTQdSFUVsykt7YYuMppP8/XYCzHOK7bFbID1Ko+JG2OxeYkeX/rzOiPnAxwMEgU2K6NowhZlmNGavtraa8LuZj1cUjHaNu3ohCBlf4wFxOz7lALdVLIr8zT0Pk7IIwAcOa8GAPB5u5EMAN7sLUQfZ1fGhfIy5HdYuJedcPPDv1pnVoYKAuivaAIAzA+IJYqlhb00BY9fYMyO3UFrMUoLxQW6NskRpyw2CvQEbUQNGbCVPwFI2jkhcpgoDdiPJZf7pPi4VpT3oFlVtIJHwqWvZ2Wm0DDNIErGtKAmxHpnuzEDA7RYONxSVhWygrwS6IAxYP8ThxjpKVuTUTSih5k2zbL/OmaWUlVGeX5Lt6X9mzJGS5rfwvjb/7v1JQa7m3hnu1AFFFUKO241LZdFgOIq7rlrGeGLytTjCVb9IFdrsHUVxWi79qM4dO4M8z4yJyaxb8NWIJu9Cjh25YfX69tMmxAUzKstw0BZk/S36t5FmALNJn4SAbTResyWBboXW3m7iVg0lRkrYQJBbKN1Dp611MIgSrCLzsCXFk7iqo+lzZq8e9MvUtZX0CEcB8YGgH7RwAOWWYug1UZSVl7WFM0GRJ9Tgkp2OsUZkjH/RVkYKJkBdG/BeateEbZ/q7vTShsMlWA08e6uQGxRFFBK0wumQk/A2a5ecPurqb83bN+rMU3wtDf9e1Bzlemx4fJTmG08y4mDT9BGZxmIMgVvwR5zAaf0nSVgCCEYLqnHkC7G+ikz6l8GiguN07GVRlMC3Qoi0yO9a7OZn6DqGs+kyewTz4TRs0Wz4BkcmjKSukFSwaBJUMCc5xH7+SL6xhZFK+VIF7vjy+dyEQMx77bdcnSuFtRwHGOd4gITgGaw8CrBeIZI3LZJKDUQBTLttvcO4cxbXsBlD21MLQYzSQ9KGGnw/SMT+MPahJAGUxhkTwYeDUAiGUUj6bWVRJK+zxiCe8wjISiVFi4WAkTS2yUtLv0ezarEsx5BhFFnNEGjiCZ3GQix2G1qGz1+QS0qioO44wtprprttA7jVDzhLC1OtWNWH8bHP5+TBdvzAV5tTZt9LuEUSdfH1n/1hBj3Wpt29uPqR97VPgebFBi8oZlIStE5IjkjWfSJuhPohJBqQsiDhJDNhJBNhJCVXnWMBdbgYNWEbKVK8QIXlWxEEYljsEMs+YP1oltsetD1+PMXDse3Pmxk60shHAf62nHLE2/htBufM++fvoNyph1LOCvYNyyFnymZguNUilZhOUa/dd96/Oxfm/CSnBFrWEBUMAtPA2DI1Esv4qIhajR92XAMgZ4P8OCXzYfvNY9tYr7DVLGLVKSLhAmBcFH1bTK1WRpFEJOYLVDXUjEEKU3WVRTj7Ss/gqUz0/VLpcLWdcIaujLn1G8jQaMoJuOSWdPqfM49YXgPQmNpE+W2XvZCrHdCmxV2+fKdr+OPz7emxiQAoHImECxxoNSkES4PpXb9SnSO2fDM5yiXGwE8Tik9EMBByHBNUdbT/q+lxqiIYRRjBw3b1kIcITwfMyZ7mMVvDddhXCgcnYdRWug4auDYA2rxZTOOC7l60aPPPodNO60K/Opmj0LxayPxKR0FwT9HMc+YaepWAv1XT2lDFFOsi07Y9SJxYKgbK+qtt8isKBcCgolJigRtxGzSmYqxnpgUj3LhQYmwcFLXUoE+xj5BGzG86z089rZ1m1yTCcSUEJZVTVFiakbS0WG8pDEnAQsaLqBAwB4zKuNyN559iCo6x8Z7yKfEIkJIFYBjAawCAErpGKVUnKvVBdTPqZDDYtaaTBegsHquekerLdgoJMvSdu798tG2HC62kUpeEp+8qf4xEp/UeKvd+Lrb5N2RiAnJ7KlbZeLfskaribdROaLDya4stS031+41gkL152WnL0IgACSSDSggNFVgm0W/q0BUi2tRxX1bCTcefW5AN0VaaBSBPa346l026u+qmmy1Idwow1GiLL5hFec+NyLIwdQ0PCc5eUwEScYFa8pCuOP75wBIj2vT95enGnozgC4AfyKEvEkI+SMhxMBBSgi5kBCyjhCyrqvLettoBjvPKUEbpcktMHNcPX8bApOl3pWEClwLdFNltkbS0B1peDXzABCutn3d41otmQAYRgg9wTpT7U1E67LS0PUYQRHaacSZ38QGM6U6UUbB4c01OGNZY0pQKWyfZmGaqSQzi9vch3K5rqXIAskOktc/S6mw9YSmsDW/USq3kRYVXaiSw0RtLNqqLrTTWqAgJKihe4BIXMoVEWKZZF8xUhPGTlqTYl00l+f5GeUSBLAcwO8opYcAGARwif4gSultlNIVlNIVtbW1ji+2tWcQ45NJaVwIzPUEjQKj+4DBLmywYr9jjTpR1MzDJAIaek0r6K/StOAgQyUYnnanKZKRas+k36FSoGq28PaYqHsoZ9rxSnDxCKi6QnOc7QhU4Ak6s/VZip5woKHPaIIIM6WZ8A0WBFK+Gzv3LpZP0Cg2vjjPU+9nUkwmLL57PZKqNtMEZuzkPdFwvSQCQM08jYbOFehONHT9OeE4QJNiLJMm7yWRjKYUo3ZufohydnaEuhuB3g6gnVL6ivzvByEJeM/RPTCK437xLK76J6+4ghGOw7NsomcEaE9GBE0uCrSDJTpvqaESzMucrEoeH7kpwjHnQiYidm6qDRB0hmZxY+vV1zEjMeJxZXSYcMYnaFR4V6bhPCkolIS6iIZu0vQASrGbVts0NwkIdJPkMxEUBbXT3Fb6fspiQhAsCGDDFadgTk0pM0yUJZNZxR4AAOEYakbSAp1XC4AK7mRMEXHAushAehGjSHQPYp2OoEtJAsxLpyildBeA7YSQBfJXJwEQl7g2oHitW2VyIv27PW2JkeqW5yQzc1Q5wdWPbhKOy6WqyaFGYb30CK3a2LF3GCPjDnjeUwJde++vb92Do659Bv0j0vNlTp5wnCuceTuDztAcVJBh1MKaI9surnlsM/e3BI0CY/146tUNeFuEk1x9o4JFtfWv8ORF2mzMRLIRzQLZtax49oYqiW7i80c1adukUdSSfQiMmju1eSNbvzj2ogL7aKktM47SQlVJIYIBgkQyipmkB8VImzGYGrruq1RlqHAMM0bbUQBzHqQkb0Ew7bMOqgRAK5hl9yZoFNVkEDWQkhPP+v1Lqd8e37gLx1y3Bs9s3m1rsfYabqNcvg7gLkLIWwAOBvB/7rtkDb3T5+dnLTMcs4OGMRkIYUeLNnX7x39/25DE4EbEJylFq2DxW5ZTFADKGg8EYG3n7uwbcbaVi8RlAat1Yv7qyfewY+9wqiADU7OIxA0cJWqwTEOdRbMB8BeoJAU+6Oz3XJNRnHWrHn4KZ9zMrqADcARPOAb0tDDZJRVo5SLBm5edjFs+o92UOs5YBXDr+Yfihk8fjMtPX2RoEwBCFvVA09qwdRhAOkPZHBwdhBnNxHqdenvyU98+Dlt+dhoQiaOATmCWhR2f5VRVEKsrZ55jyNAtrgTK6wUd5vxBmXYGG9+vwsf+zo70opt3XC6U0vWyfXwZpfTjlFLvuF9NoH9QlapiA6m+IYAtE/V4920js5y+0LEbuUKpNLjLyCgaLIpqcAVYyQx06yrB2CmbV1hgcayc3m5lh2X2L3WucRC/2taLT/w2nR2rLDZdIVmgc2zvf3qhFR++fi3e3OZtUJSeRMnKLqt5apE4MDmKRhPqXwKiWQxmlIUQMpgzGlBDBlANTkFyE0TKi/DxQ2YaNGol0iXE4ZtRwFMYWFD4jqyKfvPEKSvShUV1oN+VFgSIFJkWNgYTWNEy63HGskbm90zKBcHQRbPrmeWMpDKJIe5LyATyIlPU6fMRTlF2+QJSoWVWAlP+ZNkDW2mD4+QiQgh+/t9L+QdwInFYt23YAcgCXbQIBwXB3mAdhmmI++zf2Cat+1t72CY0p+hAGCMoSmlQnf3shCjmljhsHqKphuSYN/a6pLBA3D5tY8htp3WYpAQtm97Efa+xWSHtoiUZRQPZg5/+nc0xr4BQeceiKynVyghPFbC4pMEIiT3l12u1SUHqfpj2UndNnmIi4k8zkQUKWZr63d6/brumf5rT8ykOPasQfFAJGjVEj7DgRpzvHR7XaIa9FhoPD4lko2Zws4S+Gan+J5ezq94AACpnYZiG0Ex24o/PJVKTRq/RsSJoUDlTFs42qsyQANpMQjEVP8CkxUL6oflh4WsC0q6sNVmfWnz4hE+MIhAC4admZfIAYNVnV6RMEcqOhl97UtzZN4ZCbKd12Ne+CT/kpMgDKmezUBSYtPBUDW41PY7XeyVMVK2EsChwFW3V0KXSGoyHqg2RNvrIKZ7fifMVAI6GHokDw73iLJOMxhWyNPUY+ekj72qOp1kNWsxXgS6IRDJqiB4BvPFCd/VLzqC173el6DXnkZ2mySTKhVk1ZfSVYHhkRTzwEqwA4FdPb0kJ2Ksf3YRLH9JVpVe2iyxTRCBguw4kIG1PrRaBZFKK6ebV2HHyntTXNRelOpTVAkWV+Ngscy4Ysz6FggG001qM0YLUopLgVBmiXGMGGwkatQwztNOiomE3jJsXP2Hd7zExibJaHcYHAGfc9DzGJpIaM45k0Wc/tGRNzLDz08fMm4lH3n0y11BBTherIaff9fePTGAySdEzIMkDSvM0yiV7EH9avO2vcZCw2ywL8XkjDvvZ07jkr2/JC3OaXtOU40H+ZB2iLxxw0V9e5zcE2IrjuumZD5BQEUfxtrU8iJquSErrlIoBzyZdKDTZHbV2D8nneQf1dW1VwZHZJWdOihd/YGESBdiq2p3c8WKbeB9MkKBRNFkQYpk5EPW7nTbagCQliE5YCHTG3Lj09EWoKA5qwvgAKQrtq3e/gUN++lTq2FcSfI04GY4Z/Cx6mgKzXIVyDvMi870LJo9Z2b9baRRzyG5NdM4jb3XgrlckU1i/HFacDYcokJcCXfxhJThp6Nx3phtML//4JKy//GRu+/e+tj3Vl1ZZMzTlYzIZnI64InT4wakLuL9JPCPmApb3XFpoVMNRIoJEMoogSabS4Fm4/QXzRA+edma2exG5LtO0BACROOrGzG3UZhpjbYXElqhmXWzpGjBtT3RdTtBGlJAxRNGLoTH7uQh6vh+lsDV6PrDYVRq/KiwIYPaMUrTSKCp14al6yuCv3v2G/BfD59CwAPVkL8qRTtIxauh8nHfkXHaXWSdVzwECQUvHqIiGHiKTmK3a9Y+Mp4X77S+05m2m6BRC+4Co4CzoQzm6aaVhW2fQzznPv6K4ENWlIdNrKANQKX573DWPcSMHKMvBJGMbrbesBGN11185nk8tyhJ0+vtW0toNEQ3JKAoI1QxiHpSgOWWBeuxcgZJyNtWZs297iftbyoZNOiyLJhgQjoP0dXDZJZ/b0m2qMc4Nl8l9aMRcshsBJE0q8diDuq7luX98hXmMmXJ5VCyCLx7dbGhzHtmJ/7zHDx0066ctJYT1wCIKD1J6fv7iyc3oHkjHtivPjzXlCwsCmF9rYBthFIWBnDzWLKChm/7MKUen6xxN/zLVyBOBrgXrMTWF2YWFWeXo7n11G5Ze8QS+98AGoeu9+uOT+H2RO5NINqboNV9rY28zzYTBOILYTmst7c6EOFv9zSafYUDqZs/VX/w4AAiX4JKup4TaidDZ2hv4G3fwE2yU61pH5bC25dYRPe175IxGE6WiReZKmUW6MJmk+NfbOw3OUTup/4BakHRYhHuyx0dBgOAyRnx7M9mJiUmT5B6TfqZZFx1y+TMii+58eRsuUZX9sxKwYQaHvBqTSYpfPvEe9g6NCSaPmV+whWXGzZZ9hYG8FOgsxOoqmN9LadPaAXfr2gT6Ryfw4OvtQm2XFfGrpChoUdnAn9vSjb+/adI2ZwBYJXu4SX9m+RNEdyrlMxcazjUFAfpRii5a5YzO1gWU65r1lStM5RBNs3Ovf+p9WE16tfB9rW0PvnLXG7j3NbatWvSVdqI65XjnwUxhYKGFNqKMjKJgUMA8xUAHwqbhqQq4/alpxiQlBoXr6U2d+NTvpfyGb9zzJsye9zWfNAnXBfD0pt24ec0HEm1IOAb0JoCkeXaqWa/3oRw9tEJzz0qAhILBMZH2M4M8Feish81+6froESYstAAz5j9F8WqlacKjv7y8Fd++z6j9W2llSoEEkUowovjjc1IyipCg4/0gZ9rZnbgJGrVRcILRHzf5BwETkwvvTmVmyniBOP877/qAlkdeHxOfMr8Jg02IxT5UTKQrkS5lA3xfBneRLw6CQgpPZWU43/3KNlz6UDrEktmjYBHaaS3znl5rk3IV3k1x+LPvyawkHZDOZB6dmJSTx8aAvUY/ycj4JCaTVGjM6Xf9v3hCyzqqtqlPNfJUoBvBLR0lkOhh5cTQ80mroWQjDqEYO2mNBW2sOexUghHF1aoyaIqg4yEdM8yYPOE4DqsQ6Vf63EQyCnRvwUs/OlG4v5r+ODorTWbFe6/cSSszU37hQAcEaCrsQSX26ChvRR3xZmCZD6W2Kf69abdlXL+hvaQ0N8r7TQQ65xnefM4h+NFpB8phosY+/fjvb+POl62ToMQXKetDLGESunjgZY/ja3e/AZFR1ypIlpaxWsYmyAuBrh+n4vo5O6PNLvShVDy0JhtMr2OVmq1MMG6RZZejOiXodA9UnbZMeB2MxNAE8Wf4u3OXY+lBhwJD3YiG+CXsMoEEjSJM+kGG2UwUpvHa4RgqTDRWBVZvIhNFSxLJRjSix1DX8vGNu/DFO9bhD8+ZUwPosQszMESLUD7QinVtvWi65FG806EjNeNMrLrKYlx03Hw5jK/TNHpKAseEUdbE3ZWe8uv/qM5mn8/7nsnKKZvUeI7RxzbuMiXnUpCgUdTponPU8OPQbYCfTMB+ikr0iAgDHg+ixRb0cbl6WMW4ehG6aNV+mPSjbFLexuqjXMy6F46BDPcKcZQQEJy2NIoly1ZIX1iQInmtyCjPkVON0gQAACAASURBVFf42fQ9ROKymcjdrGTRy2r64KDNVtqAAKEGp60SFbKtx5yj29gHicO9fKANX75Tynv46G+e1x3DCfGUIRImaobihgUoJaNogHHxfX+3ecinGT507TOafxMQoCwCFFv5dazfjMg8dRq84BZ5J9ABMCUAb2UdRxDbaJ25hm5lQ7eqhyYjQRs19JqGy5jPDXShCv2ClWCcQBmIDRO85BmTDjLIlPQwDGJB2tJBB3HVZlDMbEFOhE0qe5F3n2P9BmZK232QuVKUWrOGIcZLiTdrM2U+1I4PZXyKFKU2thlF2UAbugc4lBVWYXwulZDe4jnS+Q4jZax0LS23CpHer1noooAcbklFUrHvee0Wd5XZ3CA/BToDZlqXlZ3Obhq22XUAkXJvvOvYcHw5gCIQrLIDmb2TY4YXFVnHoqcwY66UzGEygQioRak2+9hOazFOCzCwg82dbr4TkRyjZsyU3MVABSXqqckqfNLGkEtwKiIpJsFk0v4ITtBGFA+0c5PGrP0+5sIN4Kf+A0BviZQcZGfMn3vEHFx07Dzh4zVI7cDYMOOOUWCVM3JtirPfj0MXgl1bcoI2eh49woIZvaYE64WD5/gC3DtZFEEXHdeGVKbIucxmb/VcIFCIBUFzAUVB0v1UkjksYn+579PhrnVC3pW9984b7APM2hWsEWtGlAakBR2Pf8XJrQ2jGB0Mx/sT70jvRNHQ7YwTpbD1XI7JxMoeLBI9ZYaBwlrLcExAe08/+8RS/Oi/Flr278t/eT1lin307Z2YmEwiWRMD+juAUbY5RyTLU8kZESnhN9VwLdAJIQVykehHvOiQU/Bs6IBkeywm45hpwnXtBSR6zSB/EgtsBFiVYADgxZZunH7T867WfEXQRSckga4MXsVclTJFsFAQBGqaMSdpk+skHLMW6Fx57twO6XinUzkLCJbg/zW5c+Ruo3VIUpI2JejGp92YcQX6cnQvtfRgjZzpOe6gGhfPjCOKFKeLQ+pnEOLKgczjvAeAx9/ZhQfWpZWX2E8ew2UvyPOKMyZF2VKdENZNBbzQ0L8JYJPlURnGDJMU/VT0CGfQeuW+SCKArbTe8jpmk5iV6UgpxRUPv+NJHxM0iqjMsNcjD95fPvmefB3lKE4PwzHMoTYnfsS6ElAmIJFZ7WYmkZguFIEAEJ6Pg0u7XV1/FCGJXpa3uCvatIPd5jyyM/Wy1ERrb+9QIlTE21SiwNRKyN6hMdwvJ0JZOfJf/tFJGKlsthRuPEZNQqwT6sxg5Td4/gPte1zXXyP9wRHofSNifEVKlm2md/124UqgE0JmAfgogD960x02RIpD/+wT/IwxVrksBdt7hzA+4d1LMR2cJo4wpY4ky8mk1jLc9y+KuokOIDmZ4px5feseuXt83gwAEhsh3WlPOIeVSkB8Acm73oIGdvavCBK0EUVkHCdfeRfe7dBSBUil2kwgWN1GpA/Ke3ytbQ/WbDb6H0R5iRS00gZUkiFg0BvHW6qwtUrD/tZ96/GDv76FLbv7TRkcASmLunrWIkRIHyphPyqFQNmVdhvCMdXH8J6SmYbOQhttkBYXl8Xj1WRpuQS3GvoNAH4AZHaZ0q+yLFSVGMvQKehCFfo40SPHXLcGu/tGGWc5Q4JBr6nArMp9hVxGr41KRYfVfX3+g27PkhQStBEhjOPR515l9M8CkThCmMBMkzqQBmEp26S/czD/Bnha6mWnL8L9F6007dK8iJGcCUin3zdOtOPPL7Vp+6i6MhOROLBnq2lstcjrULQ4gOKlRA8+vzpdHcipGz5V/Lx7C559rxO/e9ZdJXtALmzNSGW/59XtQhFI+8qaAPD9BVZ1cFtpFAFCzR3InAlQGrKm5VBjFCEMlzYKFQQ3Q0pJ5OaM5FnYIiHkdACdlFJT4m5CyIWEkHWEkHVdXe61CinCwP5ZmYweUYNFr2nsDR/DKMYOGtZoTGve69TE5CZNStJsvOoj5v2TBd39TzAY6awghyHacgbJ58wY4WcN8haromABDm+uMW1+UWMl8/v0TsdIAUCtDNjhGEAnubHVxKzTuj6UkVHU62Kst/cOYVhOD7c7lBXH+3jne/jcn17DhvZ9FmdYQz83lLyL219oxYst1n4nRaBbR3ex0eIi9LG2oghPfvtY3PU/RwifM1jRLLAD47+Z1Z8/DC0M1kVDC1lIFXWjoR8F4GOEkDYA9wI4kRByp/4gSultciHpFbW1tS4u5w5WiR6eXSeZNu/oCbpSJcIYg+W4AyKaNvRVUURRXhTEXA7zJKAVdHpYZrgJRoDMmqG6vpzMMWN4K/d4wTB/JniV33tQiT5ayuyrpe6UWrjsOwrfvvKU1N+sWrObdvbhmOvW4Df/dqYhdtAIRmkh1rzAphCW6I/t2uUbMIMMYAYk05RdOdRf0ohxWuAokao5UpbiQXK6IBxQX4GjYhHrA2VIAr0lNeBFizr//rzleOTrRyMUDKBLgCwtG3As0CmlP6KUzqKUNgE4G8AzlNLzPOuZCZykwLcmGzCT9DC5rkW2R9GqYqHrqAWmnqDLLMT10Lk1aLv2o/jwwjrNVt0JrvrYYu5vPEHXsXdYNgmZxFiXhrEP5VynLwHFAQ0V+OoJqmIKcjJHjYlAdxOvyx8L6V2Z2gn65rY9WHH10+ZXjVizLnL7o3p2rUnF4Zh+Xp39Yua94xfUGuhuAcnx3kobQHrdm1oU6PmOmGnzJkiSQjl5z/7zOuvQWSkeJPPYf+8wVNEMjA2gZ+dWfOveN7HgsseFzjt1SRRLZlaleiSVBTSj+ph65GUcOg93/88R+MZJcfzotAMNvwnxZJuoJp861KQIswp7UYFeWs40S6QZ9vjXoVTqq74SjB0UF/JL5/HMTx+69hmYmPjlHwi2EXPTVWNVCYL6+qbhmKmGnqmdaQuNGmycT8oVdUwX8eIqoKxOgFPdCPWt7EINBmmR5nmN6R3wnHtf/fnDDQUpFFiaD20+z4Quf4KbNcoBFekTN8pF+l6/K2UcaKtPZvjrVkk5+9rN9+Oh9R3Gd2IBRYmwCtfMAjeXNwKdUvospfR0L9oSAudJfSgWwXdOPgANDG3aLNJFBLNr+GYM47XMzTtmL3qSUot0aut132oH2SIPRL2NT0Sj6C9rMufFYU28SAwVY53cSkBWtkbR3ZEeiWQjGkkvgpNpjVN7JZPrRuKm6ehik5UY4pX/uaFD/sV+6r+CVtqAOaQTQUtCLDEoha1ZSkhq4TMtlktT1M9Ow1PTZfsyr9f+q0OKnnJiUlODR5aWQhYk+rTS0M3glnXx6Li4jY6nbYiY6iaTNGVTdFwJxgKJZCOipBclVLu1TgpkPq049HBESS9KOcKZiRQPjHEhkELSzEf+2h+cIH4tFZSFsXZUiql+eP0O/PZZVdq32WXDMRxS2o2HvnoU+3fOuXq5l6DaAivv7+43P0EAiWQjCskk5giUBBSBUtjaqQ37YwfNTIWJNjKS98QighpRRYYQBrsalZeyUWGZZCl3dqJTEiLROVOMvBToTl7uCIqkRI+pcIxa0GuazeEkpdhBwxihhRllXQSA2VTb/vmrXrXsX6heKkRta/KnakdyzrF4oYUFARQW8BNTeFAWxsTm9QCAb9673vxCaoRjCI704uAwT+NkX1i/aCu1ZhUtbvMua7ZKK2SCldNNFFhVaaGmShMb2WUa1fYkoCnkzYRgFBNg1uf8inLJCtzEd0p85ZkfMK0c847inONlzQHARxY3qAZcZgV67agxlNByCJqELrIKTAMAauaBgnDvR2QQOnKE0wYp/Z4bH22CCL8YAkkzx1tCqTXL06ad3FeLZVSIk2cVxVyyi5k/IXo+4Fwgp+5pChQuwN0CpqCVQ5aWTeSdQJdgPmDLOTVArfjKzWCHtD4dV9uBlq4BvCTH8orwd/y/FbNTfXW6BbbiQGmzEHSmPaxpBkD45iBmefYS9BebVEsSkT88E4fJKaMIoQNh5q7MUjFQqH85GYWilpKEBwVW9OhDObppJf/9OVAMW+T8iVkmSWNm6DYJExWBxIPE3pUSUKHnzUsyYyFBGzGLdHFZJs2wMCrZ4NNkaZkxjTpBfgp0i5d74oF1zO8TtBEVZNgR13WhPnJDB3WI2TZanyp+e9Kv/oNz/vAyVj3fiv/7l0R5YzY4lQiVBG3kVoKxXFssDjATdFIHTU4uLAGqZ9ueuHtL5phoytaz1enmNZE0WRjNXoRC/Wszo1D/6NW1Zr0EL8KCu0uyas8yUcaqVWLwF6hh1askAmij9SbPyfqubvrMIZbHKLBimTRDdWkIqz9/mNxONKdYF/NCoM8o1af1m79cXtQEq3ivKGorinCzyYD52xvpJKIxFGI7rdO86J8+8i4GUglCAvY5l5VgRNp3rDWG47Z3D8OV87i7IxHtq7G6hPm91bktOjIrYaSof91xfgyiBLvoDM8pkVtdvL8gI5PLLOFMFC2cRUYUrSYLggjsvGKRWsNmOH5BXaod/q7ft6EzoY9rdqOtAazoEbGRcPqyRu5veo4gMxudyCTOtJOoReaIZwpYq5Mjcea5ZufFFx2CcjKCOsbuSOR93v0l8dRuNVppA8rJCGg/S5u1uHIkblk+TwTsxdNtibsoakkfKjFo67wXLzkRr/3kw4bv04WtnUdsKGGi+vBUUb9XQq5P6jQc04xCWw+v7N8JGkUlJzpnbMKZP8IN8kKge4UOWEWPOF9RkzqJ3uqSXtOtk8kKCRplClihyReOoZyMGDhKzBCQnYzzGfZGkQUuWsXW0D9zxFwcbZL2rWhiwzvZ1YtMEZ4P9CaYsdW8XWB5URDfPCmu64OZ78bZmDNf8PltNlaXYEYZm2raraPQba5HgkZRSCYx26Edf25Y3IaeYpnULbR23kYoGDB9D26LujtBXgj0mZzttl1kMnpk1gxtHxM0ihIyxix+KzJs+lCGLobjS0TgTgpoKspAZAlYSwmrcLqwbLi8c1Ohi6zJ7nzg15SFcKcJMZOyK/vZn/9puKLlQiJT/+rZJa3ewbdPPkDbB5Nas05NLqblDp2acaiJ41rwfMC5EpLmQXIYPmnCuMq+nlkCoPVDrJSLewCcKmV+YhEbt51/qObfTiZBSDbbmDluRFDD0W4u+FCT5t/pF80SmGLXStBGR2FctRVFlse0CrDFcRGWuE5sZdpVNALBzBXA5mEnajBMQxwhIbZwOXF6/eNr6YQkJdLFqaBiwaqupRMkklHUm+RPWMEsTFTEVdvi0q5tF253JAujlSmyNDbZ3dSzueSFQFcLUbeLXoI2YjbpMuW6dgL9y2tJGgen3XRvS34LDg5sqMSz3zsehzXN4B6jCDr2VtECFY0Y0nGUpM/lnB0IAOHYlMfsSrsyhxM3zE+IsnpGy2ZVp/5OMFgX3UKpa8kSJE7nSAvDZCKU+i8jHT1lb9G+5TPLcc+XjsQ+lKOHVkzZoq+wTFYzdk4i+O25y1NkaU520plAXgh0PdwIdbfRI6cvizK/H9UR/HR6QK+ZoFHHlWCaImW46Zzl3N/Tgs7B5AlIg1gv6Cy56iNGgU5AEQqakYm5B1cTsxpIMvWv20Wo3aTWrBvOKVZ1LNHmjmFQWbitLwrwlBBz4fbRZVGsnB8GIJltzFgXvYTbSBelKA1vfPlsi1MAt9EjV5zBpqatr9QTSPHpNcWTUtgxzKKRxiySMn37LAHbJUDxyhvEpkRb4Thmk05DMofe/+A1ErSBeV3LpyhT/9p1PCqYVys56axqzTqFQojlxPHOsgZsTeVPOI90MYueEgGXB8lxj0yuZeZHstkOMzonCxI9LwS6QUgISMRnv3c883u3jpuCAMEFK+cavm9meNhbPYjLBTIZuhhlZsuJVFrhZdpZpdMXEGrYHYlqlQc6rDGaSDYariu8JQ7HHDu87rswXT6PX2vWvCGrYiXFZByNsK4qJIIxFKKdY8YRRYJGHSfvKefXkn2oUNnxCWhGhKPCMskyi9pBIsmOznGW4uUOeSHQnaApUobfn5c2ORw6V7In86JH7OCQOdWG74oKA/jZJ5Zovkskoxx6TbHHvo3WyZVgMsW6GGUK2OYIuwqQGnVNi+1n2qWcqc6ePY/SwQq8nY7QVikSM7BLik5TtXNaX2tWacOqrWe/dzyu+SS7AHoiybHNC9wXjx7Cqb9h1WdXpM4H4NhswoveMam66BgKyyTbHCd+wbTilX0KgLwQ6PpHK76Gps+MVhVj1WdXoDRUIPOVdzCOEkOSscMtLizAyYvqNd+5pdecQBDbqLNCCwqaTDQ8Xlp6gUBNuHNOOxGAdhBbcm7IAt05R40z8ISE0HvnOEZFx8x1Zy0DIEWAMGvNWggOs92SG/PhsXF2OUinZpyTFkpjn0chIPLuHvn60Zi38GDm+aK4/XMrcN+FR+LFS07Erz99kOXxXpB08WqiWnEqZQJuikTPJoSsIYS8Swh5hxDyTS87ZgrB56SfCyctrMcRzTUyv4dzIcm7fF2F1matd7qku2OPc1m7JbSHSDk/hJEnEISUE662bUZUU4lkWR2+uszZQDfLBDTr8wBK0UmrnU1cTh1V0QLAxx8gCU4l6snJuOPlYXShCv3UGAoq0rMLj52HV39ykuH7BI2ilIym8ifSJgjrVsuLgqroKfva6pKZVbjknNNkO77ufMEhc+KB9ThiXhiN1SVCcekKy6TTwhyAAFnaFMKNhj4B4LuU0kUAjgTwVUKIsQhiFhFQTzrVn9zoEcFJKhpf6kV6sdtKMFea1BdNCzoH2XLFlZgsqzdqrhbPMBA5ABUDbSJXMEB57AqnjjqcNWRBntZKGwymCSGhzKH+FV1Yi4IK2Rp7Wy7SzrEH1OJPMhkUAFx07LzU2U41TEKIQQEB1IWt7Qvk/3z/eGaYqC0lJFiE7Q7rkzqBW5ZJBSyytLyKcqGU7qSUviH/3Q9gE4CZXnVMDa1cFt/IEM3f0r8u+FAT36YqCFHOCKX4rX5y2HnR6Uow3TbOSsNKS9EPRCmDUmwKFtQegDNn20xCicS4lLRW+OWnDsKZBzfiIDnGW20ZMq+jCrTo+FSEnV+FJZisnMVYDMROryotxGeOmGNaa1YEK+amcwp+9F8LU3+zBMmknljIBtLZmvZ3EmF5N+iG+lk53xjZlRlYF+YQb2ce2YlPHiKJwEw5cq3giQ2dENIE4BAArzB+u5AQso4Qsq6ry90qmIKDDCxlAp6woM519MjRHPsjC27pNdOV4521EbCwh7PCxEQFOsIxhPa2WB+nOScODPc6SuaI1ZXjxrMPSVEZq/t56uIG03MTtBE1hiQSsfucnDHf1YQ/+zCF49681qwZuAyiyShmkh4UQwo1dZvQshszMEiLXNXbbKFRV8l7CRpFk8NwTCfXAtLzy+nza5Wjc64/s9mzvjmBa4FOCCkH8FcA36KUGijHKKW3UUpXUEpX1NaKC0LdNbRtCp4XUN2dugW30SMzq0tw6UcXWh8INjGTVTjT10+M4S9fPFw+35jsYSccysq/2UKjzrPlwjFgeA9myExzQr3i2KSdQH29//34YlyYMkUYwfIXiK5bRfULcGBhJ/5rSdrpHSDiU0cx/bEXT+FmmFDGh1qjtiOSjl+gn5PGwtZ20eoyeS9BG1FCxhBFr+M+iCLNMmnDF6RCs1xUQxlfanbOvDK5AAAhpBCSML+LUvo3b7pkDVGBprYRqieOEj3iZtCKarFMek2LHcZ3T1mAY+RdgFIJxqkTV+1HKAoaX7er3YosnM+LcaqesyA7UxVNdU4NPwqHB8Xopr23AsTr+OGWpiRKVojEUTg5hAVlg/J1xSroKFCOVdeaTdNACO4SOGYU1kJlZwO7+vOHY8Plp+AuFcEZ0y5v44bTWq8zJcTVu3IANyai+y9amWoDgI4/P4/i0Ikk0VYB2EQpvd67LnmHJTOrUn/rJ47ibHQKnub704/rYtFVkS7OtnPGSjBBTsFk5tnyoaGCAN647GQ89e1jdf1zkS0nC+dvH2Jj4FbPBQJpCuPa8pBtNVWRbQLRlSm001ppV+aCjKxxot3iQDZmy4sWc/EU9VdwbraN1sttqoWnPVSVFuIoFQVxgkYxUy5sbVckVagZCB2aMbyya4sulm4SAJVcg220HklSoPEP5VXYIoCjAJwP4ERCyHr5v//yqF8G3Hj2wam/rRxgLKjnzV8vXonwnMWIFex2HD3CGyrnHzkXYVX0RYsH2oa6EsyiaKWtSaZosYEAUFYURLxem225ndZpsuXsFEBWhHOg14YdvSAI1DS72h0p3PN2THGsXZnwc5R3IqdF01FRwn4GAJXFhXjnqo8IFHfmo7woiCOaa3DDp6V5sChaCUCqa7mDassJOiX5u+/CIwEgVdjaSXm24sICd2GiAF695lxMBMsY4b7eIVKenqOJZBQNZA/KMOyorQCRyNKGymbZLlnoNdxEuTxPKSWU0mWU0oPl//7lZefUUGxVgQBBWcidQD90bg2WLz8MQTqGRtLtSHM+wCQNXV0RhkWvafdq6kowdgd3SqBzBNAkCrCN1jvbrRQEgZp5qUEskXMJ9DAcx3HhvfiGrhCEXdi1P0vp9yqtT7SBikagsBShfQl7F1ShrCiI8mg8VWvWCe67aCU+LkdRTKiy2/S2ead6oVLmj5coYwcJGtVQP9taZAhBQW1cd0/eiPXjDqjFCQtq8Yuz0klHbhZaADhtiXT+YFmTVqDna5RLxrH7HSwbeAG/P2+5ZbyxHieknD66AeEyDf1D8yP468Urmb8FAgTfkYscKMVvXTmZUvHs9kmPlJ16gUXGoePtbSSuDUMUmbmRGEr6t+I7J813dMmQ7Ato1lV5P2S2FM54LSdVvoVGMZekd2VjE4K7s0AAqJmP0B6bET063Hz+kegvmYlz5lmTn1lhfDL9nNM2YOk7pzzcymltqfFmf0zUKqGL+kXGZp9IJI4PVfcipvhFXAhHteP3ji8cjj99/nDUVappGdRmUftQ/BuDlc1AT0sqOifvnKJThtdWAX+/GKcubrCtTSvORYNDMBVtoR609l7nAfV8Lb24MH09vZOprEisssrXT4xhbrjUFa2pMo/U4YsVxVpOlIRO0NmCXKZN4SgR1dAxOQbs3Wr/epCYLf94wQrc/BktPXC8vgJt134UZx8+h3legkZRRCZSFYjG7cRrR2K67bT9qT9rRimqZy9C3eh22+fqMT6p0tBpIyrJMGqxz5Ueq9h8lcLWtpzdMpQEqJmxpQiTflTJyXvlxTZ5eMJxBPt3ID7DGX+PFRY3pv1rCsukhn/GxvZPUTDGquYBE8NoRE82ihUByBeBHokDo/uAQftx7MPjkqApKtTdalktUFSFpcXOY+MVPmQWzjp0NhpkSt00vabUF9FdxndPWYD/fP8EjFU1IUkJmsku00WEBSUJSu1Te+XHJ+GtK09J/fvcj56EIjLhLFsuHAeS46pzRQS6tDtyU4D5w4vqTZOmvnlSHJedrk1cVpxtyq7MbNdiQDgO7N2KX33iQPud1bfT8wECLvW3EpUf6arPnwkgbSKpZWSAiqA0lBaeieT/b+/cY6wqzgD++3b37vt57wUEFtiFvVDB0JZCca3BiuEhGrWRJhhjTbUaa+3bGqipsamtfaQPm5gQU000adX6SEutjdVKozGtioIWa+QhoAiICq5Skef0j5mzXHbv7t4zc869nMv8ks2eM/fMmZnvzHzz+mZmLOMObw+19B90Zbv1p+dxVu8ZgLZ0mTqmhWxzyDhldR65ZV49bQ2pok7hGkjweYeyEArmDIJdJm2HXG6+YAZfn99DbrpuYPygN9xReFGSDIVuNkjq79qHKIhTzJ7Ucyalj/9BBLI99LYXf9BxIW46fzr3XnX6IPd0Uy3/WjGfb5yTo2PCDFJyhIkDN2Yqkl9f2ssOMswf1Wf2Di8+/cEE8tnTRve7NdbW0Goqo3lTR9EyTtvU56p2Fj8OHmB6OtfM0IWmoZj5DeMnzgmkby+YypVndnPjklO55wpt0z/lVL1lwPfn1iCoIY8TLEg2B+ooF3cfGvnZ4chMgcP7uf38MSM/OwwrzbGMP7rotP4K8sfz6ujsaChonloMo1rqePharYgbx30C3ttov1DJfOMvTNxPfY1Fe9WkKfPxG4xuqQtl0RQwpyvNzM42vrdoGnBsf/qAuZMzPHBNL79YOpNdqU5OTb1Nz+jiD5oOSDfV8t2F06gepYdZp9XoyeTmYRp8cRFPfyZqTG1towAWnzaW1dd/ftB4KwCZHGO3Ps31CxfAaruoXXHm0CvDRMxY+pvnwJ238tAXR8Gq8GF8ckI79Myk86PwLeimuhqevuHsAgdwwJZbjVHSR3o/7ZVL2qh5KmQmNJXtJVMOwM7a4mwJGzNQ3z7AZjcerjKLjZ5ZPp9sUwp+2cbU6l1MzWXhUAirhqBX4Rpno+h629wWzUwZ1czmnyzR4lYKaurpqdoF9Slc7EJmTexgwy3nUvv8NnjsYe64uBv+MrK/QRgLqMtyB8Fm6iGCXlxTXQ2rrjsTgHU3LegfGslnTleaOV1peGcuvHgPuTO64K+WATaPhrpWumUn48a3UdcQvlfhSjJa6G0ToLrOujAVVOagK4oP3oJD/3OIXBFk9ORf9V6HFqnpqtvYpE1INxbMzCKize+Mgk3t3Rx+5WJTBho6wu3PIjJ4MjVmxrc3UJeq0XK0CbdfwTjGOTOwd2KvfKurzPczk7ZR9Xhqa6r6K556W8seY55qHafaJmgdH1ml395Ye9yQ0iCyPVoPfOiwmElEl/V3N1r3klxJhkKvqjaCiriLnrFv+YeiMa2Vpks42Rwc3Af77JZTD4uIloWtsgoqm+BdYf2UkqxluPWt0HyKe5xbToHa5ugrM4dNzwpiGiFOCjWTcyuzmYET0TEycFjX5T1ltEVPhkIH83EjLgT9H7EEHyA/c9ts4BEUsLhatbaKLvAbNl6ZKbo1dCD8AdhOZHrsw81XMLabsASVb02n+wAACQtJREFUZxx5ee9WbT0UBe2ToLrWUSFrC6iCJ8IUQzYoMyUwABw0r2P5fbM56HsTDoXchTQikqXQ926Fo3Y7uBV+5xRAdKaLm2wPhFlROZD+rnpMCj1QdGHGlfv9ToF9u8IpyaAA7Xmdku554TIhG1UrOOvYci1EpgfUEWtT0EFUVR+3aMyqAsvm4MgB6HvDshFjb90WmpaxkGqKoIVuev17trjHyYLkKPRsTivzKBV6qkGPzx9xX+gxIpkcHP545OeGonU81DTAfjernCEJFN2Rg4RWsEFlczjMJGMQXglkXyjcj9/HKp37I9gBMNMT/bxN8P1c8thAMj3hvukg/4Gs++zDB53nXbelHIlg/Ns1Pwbf4WCJe56G5Cj0IHNETWBBA/Fmmqxj/KuqjmXwOHCR73FpK1KG6cnFPxslLuG6fsOAOL5jHO90Ta+z/xjze8HwIvi+6aG3cC4FyVHox33cCBVBXBVFHOEEMoij4kl3Yy3X9GQIsT84AKl6aC+8ojNWUvXQPsHOb1RKM6qKIZ+Gdr1YLkpc0xuYp9oSWLeViijKaG0TtHaamwRtn1tyGjqgMTvyc2GJo3AVoqMLJPymYscRZ+WTarBXdDV1dsq5VLIfiK0cjW21e/gx9QqDdEX1Ttf8Fpin2hJYt5WKqPJjqXsWeSRHoUM8CqBUGaamFjommRvLAhfnkAu4FWAbv3GnZyhs81FgWw04tb5qm+JpeUatSGyG0gbi+o1LmUeiqmhL1esvQLIUeixjjyUUvmtYcdf8WYcWns23KZdCzzgMXUWVX5KQlxvT0JAe+bnh6E9nmectiiGqxl25ep64H0G3WEReE5FNIrI8qkgNSRyFILAeKQWuHzruysdFvjbj++XK+C7hRlWpBu+xPY2i4DtjkGe5J0ZLWenXhdv4bkhKOUw0AJcj6KqB24FzgenAJSIyfXhfjsSRYeO2HsnHNZz6Vmh229RpWFziZzXkkrAxdFe/hd7z0bvRvA9OzFZ/uf2Xg4QOuXwW2KSUel0pdRC4D7gwmmgNQSlMF+Mkigopzszi1HK18Ns6Ti/mKDWt+gQiKyKbOItoqXk+HV1QFfF+e66tTVfz1HJNMLr0nNomQI3dFsauuCj08UD+Tv3bjVt8dHRpQUUtrDGnGcUSs5lRdpq2dKl1UGJxZvCWcXrM1Kbr2TIW6tvCfRsRGDPdTR42BItIbL53VBXqaNOZrQ6xhe9IVKcgO1VbLEVFEM+UZZkLzFNtJxnjsm4biummTRrWDDefqiqdT6pLv32u2B5XJSJLgcVKqa+Y+8uAuUqp6wY8dzVwNcDEiRM/s22b49Lkt17Uir3RcbImn0P7oW97acZ0d63XXWPbArLzZdj6NPR+Ldp4Bezdpu2H65rD+92xVlcKLSGGhfq261Zlyynhw3Nh0xO6FZZbEM6fUvDMbTB1MYx2POzilT/B5LO00oqK9zZrhd46Lpr3HT0KW/4Jk8+2V8ob/q4r+4lz7fy/dJ9OT/c8O/9hOHwAdqyzj2vAjrW6sh4zI5JoicgLSqnZIz7noNB7gZuVUovM/QoApdStQ/mZPXu2WrNmjVV4Ho/Hc7JSrEJ3GXJ5HsiJSLeI1ALLsDq+wePxeDxRYD2DopQ6LCLXAY8B1cBdSqlXIouZx+PxeELhNCWulHoUeDSiuHg8Ho/HgWStFPV4PB7PkHiF7vF4PBWCV+gej8dTIXiF7vF4PBWCV+gej8dTIVgvLLIKTOQdwHapaBaIcCejisTLaHi8fEbGy2h4yiWfSUqpEY+kKqlCd0FE1hSzUupkxstoeLx8RsbLaHhOdPn4IRePx+OpELxC93g8ngohSQr9jnJHIAF4GQ2Pl8/IeBkNzwktn8SMoXs8Ho9neJLUQvd4PB7PMCRCoZf8MOoTBBG5S0R2i8j6PLe0iDwuIhvN/w7jLiLyWyOjl0VkVp6fy83zG0Xk8nKkJQ5EZIKIrBaR/4rIKyLyTePuZWQQkXoReU5EXjIy+qFx7xaRZ40s7jdbYCMideZ+k/m9K+9dK4z7ayKyqDwpigcRqRaRtSLyiLlPpnyUUif0H3pr3s3AZKAWeAmYXu54lSjt84BZwPo8t58Dy831cuBn5noJ8Df0uWqnA88a9zTwuvnfYa47yp22iOQzFphlrluADegDy72MjslIgGZznQKeNWn/I7DMuK8EvmqurwVWmutlwP3merope3VAtymT1eVOX4Ry+g7wB+ARc59I+SShhV76w6hPEJRSTwF7BjhfCNxtru8GLspzv0dp/g20i8hYYBHwuFJqj1JqL/A4sDj+2MePUmqnUupFc/0h8Cr6XFsvI4NJ6z5zmzJ/CpgPPGjcB8ookN2DwDkiIsb9PqXUAaXUFmATumwmHhHpBM4DfmfuhYTKJwkKvfSHUZ/YjFFK7TTXu4DgAM+h5HRSyM90fT+NboF6GeVhhhPWAbvRldVm4H2l1GHzSH56+2Vhfu8DMlS2jH4D3AAcNfcZEiqfJCh0zxAo3dc76c2URKQZeAj4llLqg/zfvIxAKXVEKfUpoBPdanQ83bpyEJHzgd1KqRfKHZcoSIJCfwuYkHffadxOVt42wwSY/7uN+1Byqmj5iUgKrcx/r5R62Dh7GRVAKfU+sBroRQ83BSeW5ae3Xxbm9zbgPSpXRp8DLhCRrejh3PnAbSRUPklQ6P4w6uNZBQRWGJcDf85z/5Kx5Dgd6DPDDo8BC0Wkw1h7LDRuiceMXd4JvKqU+lXeT15GBhEZJSLt5roBWICea1gNLDWPDZRRILulwJOml7MKWGasPLqBHPBcaVIRH0qpFUqpTqVUF1q3PKmUupSkyqfcs8vF/KGtEzagx/5uLHd8Spjue4GdwCH0mNyV6PG6fwAbgSeAtHlWgNuNjP4DzM57zxXoSZpNwJfLna4I5XMmejjlZWCd+VviZXScjGYCa42M1gM3GffJaIWzCXgAqDPu9eZ+k/l9ct67bjSyew04t9xpi0FWn+eYlUsi5eNXino8Hk+FkIQhF4/H4/EUgVfoHo/HUyF4he7xeDwVglfoHo/HUyF4he7xeDwVglfoHo/HUyF4he7xeDwVglfoHo/HUyH8H5suu6MX6j7DAAAAAElFTkSuQmCC\n"},"metadata":{}}]},{"metadata":{},"cell_type":"markdown","source":"## Submission"},{"metadata":{"trusted":true},"cell_type":"code","source":"submission.time_to_failure = est_gp.predict(X_test)\nsubmission.to_csv('submission.csv', index=True)","execution_count":194,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"submission.head(10)","execution_count":195,"outputs":[{"output_type":"execute_result","execution_count":195,"data":{"text/plain":"            time_to_failure\nseg_id                     \nseg_00030f         4.279934\nseg_0012b5         4.732133\nseg_00184e         5.449688\nseg_003339         8.283347\nseg_0042cc         5.443126\nseg_004314         1.578514\nseg_004cd2         6.690328\nseg_004ee5         3.231288\nseg_004f1f         3.649851\nseg_00648a         2.784753","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>time_to_failure</th>\n    </tr>\n    <tr>\n      <th>seg_id</th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>seg_00030f</th>\n      <td>4.279934</td>\n    </tr>\n    <tr>\n      <th>seg_0012b5</th>\n      <td>4.732133</td>\n    </tr>\n    <tr>\n      <th>seg_00184e</th>\n      <td>5.449688</td>\n    </tr>\n    <tr>\n      <th>seg_003339</th>\n      <td>8.283347</td>\n    </tr>\n    <tr>\n      <th>seg_0042cc</th>\n      <td>5.443126</td>\n    </tr>\n    <tr>\n      <th>seg_004314</th>\n      <td>1.578514</td>\n    </tr>\n    <tr>\n      <th>seg_004cd2</th>\n      <td>6.690328</td>\n    </tr>\n    <tr>\n      <th>seg_004ee5</th>\n      <td>3.231288</td>\n    </tr>\n    <tr>\n      <th>seg_004f1f</th>\n      <td>3.649851</td>\n    </tr>\n    <tr>\n      <th>seg_00648a</th>\n      <td>2.784753</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"name":"python","version":"3.6.4","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":1}