{"cells":[{"cell_type":"markdown","id":"ee71dbe6","metadata":{"id":"ee71dbe6"},"source":["[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](http://colab.research.google.com/github/ai2es/WAF_ML_Tutorial_Part1/blob/main/colab_notebooks/Notebook06_ComplexMLClassification.ipynb)\n","\n","# Notebook 06: Complex ML Classification [Colab Version]\n","\n","### Goal: Training a ML using all features/predictors/inputs\n","\n","#### Reminder of Problem Statement\n","\n","Reminder of the ML task we want to accomplish in the paper. \n","\n","1. Does this image contain a thunderstorm? <-- Classification\n","2. How many lightning flashes are in this image? <-- Regression\n","\n","#### Background\n","\n","Please make sure you already did Notebook 4, because this notebook extends what we did in Notebook 4 to now include many more input predictors and some additional evaluations/interrogations. \n","\n","#### Step 0: Get the github repo (we need some of the functions there)\n","\n","The first step with all of these Google Colab notebooks will be to grab the github repo and cd into the notebooks directory. \n","\n","To run things from the command line, put a ```!``` before your code"]},{"cell_type":"code","source":["#get the github repo \n","!git clone https://github.com/ai2es/WAF_ML_Tutorial_Part1.git \n","\n","#cd into the repo so the paths work \n","import os \n","os.chdir('/content/WAF_ML_Tutorial_Part1/jupyter_notebooks/')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"bsNTleSWZPhA","executionInfo":{"status":"ok","timestamp":1648653378793,"user_tz":300,"elapsed":24214,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"a7a5397f-f515-4f7d-c6dd-1f4147449444"},"id":"bsNTleSWZPhA","execution_count":1,"outputs":[{"output_type":"stream","name":"stdout","text":["Cloning into 'WAF_ML_Tutorial_Part1'...\n","remote: Enumerating objects: 301, done.\u001b[K\n","remote: Counting objects: 100% (301/301), done.\u001b[K\n","remote: Compressing objects: 100% (197/197), done.\u001b[K\n","remote: Total 301 (delta 139), reused 236 (delta 96), pack-reused 0\u001b[K\n","Receiving objects: 100% (301/301), 195.77 MiB | 17.84 MiB/s, done.\n","Resolving deltas: 100% (139/139), done.\n","Checking out files: 100% (100/100), done.\n"]}]},{"cell_type":"markdown","source":["\n","#### Step 1 & 2: Import packages and load data for Classification \n","In Notebook 4 we only wanted 1 feature, no we want all available inputs (36 total). So all we need to change is the ```features_to_keep``` keyword to include all indices."],"metadata":{"id":"y57rsa_fZO0-"},"id":"y57rsa_fZO0-"},{"cell_type":"code","execution_count":2,"id":"d56211ca","metadata":{"id":"d56211ca","executionInfo":{"status":"ok","timestamp":1648653414808,"user_tz":300,"elapsed":7869,"user":{"displayName":"Randy C","userId":"12301342565919601334"}}},"outputs":[],"source":["#needed packages \n","import xarray as xr\n","import matplotlib.pyplot as plt \n","import numpy as np\n","import pandas as pd\n","\n","#plot parameters that I personally like, feel free to make these your own.\n","import matplotlib\n","matplotlib.rcParams['axes.facecolor'] = [0.9,0.9,0.9] #makes a grey background to the axis face\n","matplotlib.rcParams['axes.labelsize'] = 14 #fontsize in pts\n","matplotlib.rcParams['axes.titlesize'] = 14 \n","matplotlib.rcParams['xtick.labelsize'] = 12 \n","matplotlib.rcParams['ytick.labelsize'] = 12 \n","matplotlib.rcParams['legend.fontsize'] = 12 \n","matplotlib.rcParams['legend.facecolor'] = 'w' \n","matplotlib.rcParams['savefig.transparent'] = False\n","\n","#make default resolution of figures much higher (i.e., High definition)\n","%config InlineBackend.figure_format = 'retina'\n","\n","#import some helper functions for our other directory.\n","import sys\n","sys.path.insert(1, '../scripts/')\n","from aux_functions import load_n_combine_df\n","(X_train,y_train),(X_validate,y_validate),(X_test,y_test) = load_n_combine_df(path_to_data='../datasets/sevir/',features_to_keep=np.arange(0,36,1),class_labels=True)"]},{"cell_type":"markdown","id":"ef7081d7","metadata":{"id":"ef7081d7"},"source":["Let's print out the shapes of things to make sure they match what we expected, ```[n_samples,n_features]``` for ```X``` and ```[n_samples]``` for ```y```"]},{"cell_type":"code","execution_count":3,"id":"c0beaff8","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"c0beaff8","executionInfo":{"status":"ok","timestamp":1648653424378,"user_tz":300,"elapsed":234,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"f1a52393-dfd3-42aa-d697-19d9c6d76c80"},"outputs":[{"output_type":"stream","name":"stdout","text":["X_train, y_train shapes: (446307, 36),(446307,)\n","X_val, y_val shapes: (86291, 36),(86291,)\n","X_test, y_test shapes: (86296, 36),(86296,)\n"]}],"source":["print('X_train, y_train shapes: {},{}'.format(X_train.shape,y_train.shape))\n","print('X_val, y_val shapes: {},{}'.format(X_validate.shape,y_validate.shape))\n","print('X_test, y_test shapes: {},{}'.format(X_test.shape,y_test.shape))"]},{"cell_type":"markdown","id":"eb9ecadf","metadata":{"id":"eb9ecadf"},"source":["#### Change from Notebook 4\n","\n","Since we are using more than 1 input predictor it is important to normalize our predictors. Why is this important? because in reality each one of our inputs have a range of valid values associated with them, and that range could be large (e.g., -100 to 100) or it could be small (e.g., 0,1). The machine learning will weight these inputs quantitatively, so if we use the default scalings, it might be biased to use the larger magnitude predictors more than the small magnitude predictors. To prevent this we will scale the data to have mean 0, and variance 1. You are likely more familiar with the term *standard anomaly* which is the same thing: \n","\n","$$ z = \\frac{x - \\mu}{\\sigma} $$\n","\n","where $\\mu$ is the mean of that specific feature and $\\sigma$ is the standard deviation, both calculated from the training dataset. We could implement this ourselves, but ```sklearn``` has this built for us: ```sklearn.preprocessing.StandardScaler``` and it works alot like how we fit the machine learning model before."]},{"cell_type":"code","execution_count":4,"id":"869d3d8e","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"869d3d8e","executionInfo":{"status":"ok","timestamp":1648653427946,"user_tz":300,"elapsed":908,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"6eb85ffb-73ca-4606-c26a-a3a7d0d38db6"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["(array([ 4.56472368e-16, 3.53052847e-16, 2.49633326e-17, 6.47008825e-17,\n"," 2.03782307e-18, 6.24083316e-17, 2.36132748e-16, -2.13971423e-16,\n"," -1.21250473e-16, -7.13238075e-17, 3.44901555e-16, 2.64916999e-16,\n"," 8.86962492e-16, -2.30834408e-15, -4.03361604e-16, 3.12551114e-16,\n"," 4.87803898e-16, -1.91810097e-16, -3.89733663e-17, -2.39444211e-16,\n"," 2.79181761e-16, 1.12080269e-17, 1.46977989e-16, 6.58471580e-17,\n"," -1.54492462e-16, 1.64936305e-16, -8.58432969e-17, -8.66074806e-18,\n"," -1.29911221e-17, 2.98031624e-17, -1.96140471e-17, -3.13315297e-17,\n"," -4.83982980e-18, 7.20243092e-17, 2.62369721e-17, 1.23288296e-16]),\n"," array([1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.,\n"," 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.,\n"," 1., 1.]))"]},"metadata":{},"execution_count":4}],"source":["from sklearn.preprocessing import StandardScaler\n","\n","#create scaling object \n","scaler = StandardScaler()\n","#fit scaler to training data\n","scaler.fit(X_train)\n","\n","#transform feature data into scaled space \n","X_train = scaler.transform(X_train)\n","X_validate = scaler.transform(X_validate)\n","X_test = scaler.transform(X_test)\n","\n","#double check that mean is 0 and std is 1. \n","np.mean(X_train,axis=0),np.std(X_train,axis=0)"]},{"cell_type":"markdown","id":"bd97d8a0","metadata":{"id":"bd97d8a0"},"source":["Note that e-16 means $\\times 10^{-16}$ which is effectively 0. So it has successfully scaled our data to now have mean 0 and std 1. We are now ready to train our machine learning model again."]},{"cell_type":"markdown","id":"94e0b0e9","metadata":{"id":"94e0b0e9"},"source":["#### Step 3: Initialize model\n","\n","To keep consistency with Notebook 4, we will continue using logistic regression"]},{"cell_type":"code","execution_count":5,"id":"08ffa287","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"08ffa287","executionInfo":{"status":"ok","timestamp":1648653431331,"user_tz":300,"elapsed":173,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"307b3049-e699-4259-92e3-e2aecd27df94"},"outputs":[{"output_type":"stream","name":"stdout","text":["LogisticRegression(max_iter=300)\n"]}],"source":["#load model from sklearn\n","from sklearn.linear_model import LogisticRegression\n","\n","#initialize, we need to expand the number of iterations to not get a warning about not converging\n","model = LogisticRegression(max_iter=300)\n","\n","print(model)"]},{"cell_type":"markdown","id":"1a71f871","metadata":{"id":"1a71f871"},"source":["#### Step 4: Train your ML model! \n","\n","And just like that, we are back to being ready to train the model!"]},{"cell_type":"code","execution_count":6,"id":"c0256ea4","metadata":{"id":"c0256ea4","executionInfo":{"status":"ok","timestamp":1648653457869,"user_tz":300,"elapsed":24567,"user":{"displayName":"Randy C","userId":"12301342565919601334"}}},"outputs":[],"source":["model = model.fit(X_train,y_train)"]},{"cell_type":"markdown","id":"242303f2","metadata":{"id":"242303f2"},"source":["#### Step 5: Evaluate your ML model\n","\n","Let's see how it did!"]},{"cell_type":"code","execution_count":7,"id":"1720c013","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"1720c013","executionInfo":{"status":"ok","timestamp":1648653463534,"user_tz":300,"elapsed":181,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"5904c15c-0d73-4f7a-c835-c750afeb030d"},"outputs":[{"output_type":"stream","name":"stdout","text":["{'num_true_positives': 42505, 'num_false_positives': 4814, 'num_false_negatives': 3294, 'num_true_negatives': 35678}\n"]}],"source":["from gewitter_functions import get_contingency_table\n","\n","#get predictions\n","yhat = model.predict(X_validate)\n","\n","#the contingency table calculator expects y_true,y_pred\n","cont_table = get_contingency_table(y_validate,yhat)\n","\n","\n","#the function returns a dictionary (thats what the curly brackets mean) of each entry in the table. \n","print(cont_table)"]},{"cell_type":"code","execution_count":8,"id":"3c8db1a6","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"3c8db1a6","executionInfo":{"status":"ok","timestamp":1648653470257,"user_tz":300,"elapsed":176,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"0f1eb385-eb46-4392-f8c8-394d9293d6d5"},"outputs":[{"output_type":"stream","name":"stdout","text":["Accuracy = 91.0%\n"]}],"source":["from gewitter_functions import get_acc\n","accuracy = get_acc(cont_table)\n","print('Accuracy = {}%'.format(np.round(accuracy,0)))"]},{"cell_type":"code","execution_count":9,"id":"2f8faf68","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":362},"id":"2f8faf68","executionInfo":{"status":"ok","timestamp":1648653472886,"user_tz":300,"elapsed":897,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"f324416b-b832-4a50-df9d-6a6d79fd619e"},"outputs":[{"output_type":"stream","name":"stdout","text":["POD:0.93, SR:0.9, CSI:0.84\n"]},{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"iVBORw0KGgoAAAANSUhEUgAAAsMAAAKPCAYAAACIFsdVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAWJQAAFiUBSVIk8AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeXBb53U28Ode7CsBguC+iBJFkdplyZIs23Js2ZZtOXHcOOkkaZbpfJ4kX6ZL2iaTZJI0nWkyk07SeDptOonTOs5MuqT+7DiKd1l2bMuSLWuhNkokxZ3gBmLfgXvv9weLG0BcRJAA7oLzm9EMIVLgKwkEHpx73vMygiAIIIQQQgghpAKxUi+AEEIIIYQQqVAYJoQQQgghFYvCMCGEEEIIqVgUhgkhhBBCSMWiMEwIIYQQQioWhWFCCCGEEFKxKAwTQgghhJCKRWGYEEIIIYRULArDhBBCCCGkYlEYJoQQQgghFYvCMCGEEEIIqVgUhgkhhBBCSMWiMEwIIYQQQioWhWFCCCGEEFKxVBmGw+Ewfvvb3+Lb3/42HnzwQdTU1IBhGDAMg6tXr675/kOhEL71rW+hu7sbZrMZLpcLhw4dwjPPPFOE1RNCCCGELET5pjQYQRAEqRdRbL/5zW/w6KOPLvq53t5edHV1rfq+x8fHcfDgQQwNDQEArFYrEokEMpkMAOBLX/oSfvKTn6z6/gkhhBBCFkP5pjRUWRkGgNraWjz00EP427/9W/zsZz8ryn0KgoDHHnsMQ0NDWLduHU6cOIFwOIxwOIx/+Id/AMuy+Nd//Vc8+eSTRfl+hBBCCCG5KN8UnyorwxzHQaPRiLeHh4fR3t4OYG3vnLLvyFiWxZkzZ7Bz5868z3/lK1/BE088gfr6eoyMjECv16/+L0EIIYQQkoPyTWmosjKc+0Appl/96lcAgHvvvXfBAwUA/uZv/gYMw2BqagrHjx8vyRoIIYQQUpko35SGKsNwqbzxxhsAgMOHDy/6+aamJmzZsgUAVPlgIYQQQoj6VHq+oTC8QjMzM5ibmwMA8QGxmM2bNwMArly5UpZ1EUIIIYSsFuUbQCv1ApRicnJS/LixsXHJr8t+Lvfrl7N79+5Ff//SpUuwWCxiLxAhhBCiZkNDQ7Db7RgeHi7L9/v0pz9dlHFkKzU0NLTka/qZM2fKto4blSrfKAmF4RWKRqPixyaTacmvM5vNAIBIJLKm78dxHPx+P4LB4Jrup5wMBoPUSyiY0tas0+mkXkLBlLTRgud5AADDMNBqtWBZZVw8K/e/cSqVQjKZBMMwMBgMq3pcKvGxXKp+zVJRyno5jsPc3BzC4XBZv+/Vq1fXPI6skO8lV+XON3JEYVhiS70b3L17N86fPw+r1VrmFa3eunXrpF7Ciu3YsQNtbW1Ip9OYmpoq+5Pwai33rl2umpubpV7CivA8j1QqJd5et26dogJba2trWb5PKpXCu+++C47jAABdXV1oamoq+H7q6uqKvbSSq66ulnoJBbPZbFIvYVnJZBK//OUv4fV68dOf/hQNDQ1l/f5dXV04depUyb/P/v37wTCMpBVgsjRllD1kwGKxiB/H4/Elvy4WiwGAokJsJbLZbNBqtTCZTGAYRurlEBnIhjtg/uddSUG4nIaGhsR/K4vFsurwMj09XcxllYXP55N6CQWT8xt9QRBw9OhReL1eAKDnYolQvqEwvGK5FTmPx7Pk12U/V+53t6Qwdrtd/DiRSEi4ksIs99gjqycIQl4YrqqqknA18pVKpTAxMSHe7ujoUEwrCZGft99+G/39/eLt3OdlUj6UbygMr5jb7UZNTQ0A4PLly0t+XXaXZXbXJZEfvV4Po9EIYOGlcVJ84+PjUi/hprK9wsB8dSrbG0fy6fV63HrrrXC5XHA4HHC5XGu6P6oOl4dcq8MbN24U2zj27t27bL8qKR3KNxSGC3L33XcDAF577bVFPz8xMSE+kA4dOlS2dZHC5PbQJZNJCVdC5CKTyYgfazQaRQT4G42Ojpbl+9hsNuzcuRM7duygy9oKIsdA3NDQgM9//vPYv3+/+PpKpFHp+YbCcAE+9alPAQBeffVV9PT0LPj8P/7jP0IQBDQ0NNAPtowptUWClAbP88g9lV4pO/ClptVW7v5rJVaH5cpqteLuu++mdhuJVXq+Ue2jz+v1ir/8fr/4+4FAIO9zuZdHgflLpAzD4Lvf/e6C+3zkkUewb98+8DyPRx99VNyBmkwm8aMf/QhPPPEEAODv/u7vFDVOqtLkVoYpDJPcXmGWZanauYjcNwvFpsRWCaWSujqcyWTojUQRUL4pPtW+tXe73Yv+/m233ZZ3e2hoaMUjwRiGwTPPPIODBw9iaGgIt912G6xWKxKJhHiZ9Ytf/CIef/zxNa2dlFZuZVgtbRIGgwENDQ2orq5GPB5HIpFAPB5HNBpFLBbLC3xkaUqvdo6OjhZ9xBrP8zhz5gxqa2vR3NxMlfP/5fP5FDlqLRwOSzJuTRAEvPzyy+jr68MjjzyCDRs2lH0NakH5pviU/cwvgebmZpw/fx4/+MEP8Oyzz2J4eFjsofvSl76Ej3/841IvkdyEmirDLMuiubkZBw8ehNFozLvUmEgk4Pf7MTk5iYGBAQQCAQlXKl86nQ5arRY8z+dVhcfGxtDS0iLhyuTB4/EgFAohFAphamoKe/fuLXr1fHp6WpFzh8nKnT59GhcvXgQA/PrXv8bnP/95VU4lULJKzjeqDcOrvay3kj9nt9vxve99D9/73vdW9T2IdHQ6nbhjmeM4xU+SaG1txe233w6WZdHb2wu/3w+bzQaLxYKqqiq4XC40NDRg586duHbtGnp6ehAKhaRetuwwDEMVz0VkMhkMDg6Kt+vr66mNJAdVh1fm+vXrOH78uHh7+/btqK+vL9v3VxvKN8Wn2jBMyGJMJhMymQy0Wq0qjpTcunUrBEHASy+9hNnZ2bzPWSwW1NTUoKWlBRs3bkR3dzcYhsGJEyeobULFitkqMTIygnQ6DWC+FaeUpwkqtTpMgXh5c3NzeP7558Ug1tTUhMOHD9ObKiIrFIZJRQmFQvjtb38Lk8kEvV6/ZO+VErAsi9raWly5ciXvBKfsi040GkU0GsXIyAjOnDmDgwcPoqurC7Ozs+jt7S3rWsfHxxVzLDOZl0wm88a1bdiwgarnKlPqQJxIJPDMM8+IezNsNhv+6I/+SPG9+UR9VDtNgpDlxONxBINBqZexJg6HQ+xzFQQBLMsueRksHo/j5MmTiMfjaGpqohcjclODg4PibnSr1UqXtZdBExIW4nkev/nNb8R/G61Wi8cee0yVR/kS5aMwTIhCZSu/jY2N0Ov1C8bo3CgUCiEYDKKqqirvkAmytLGxMamXsCprPYAjEonkHcva0dFRlsvaNGat/Eo1bu2NN97A0NCQePvIkSP0horIFoVhQhQqmUxiaGgILpcLDzzwANra2mAwGJb8ervdDq1Wi1gsVsZVEiW6fv26+HF1dfWaj12uBEquDhc7EF+4cAHvv/++ePvAgQOqPMKXqAeFYVIxNBoNmpubUVVVpejTjnIrdhcuXMD4+Djq6+txxx13YO/evWhvb4fD4YDJZBLDMcuy2LZtG6qrqzEwMCDV0okC+P1+sQcdmK8KlxNVh6VRzECc24LW2dmJgwcPFu2+CSkFahwkFcNms2Hv3r0A5lsGjh07JvGK1i6VSuHFF1/Etm3b0NXVhe7ubnR3dyMajcLv9yOVSsFoNMJkMsHpdKK/vx8jIyNSL5uUwWqnSuS2htTX10tyQINSKXWyRLHdeeedcLlceP/99/Hwww/T5AgiexSGScXIPXlO6mNJiyW7ee7KlSuYmZlBY2Mj6uvr4XA44Ha7wTAM9Ho9YrEYzp07h7Nnz9JYtQJV2uEbW7ZswdjYGMbGxiQ7JUypY9YAZQfiYk6X2Lx5szjOkRC5ozBMKkZuGFbLwRPZ6REcx2F6elq8xFxVVQWbzQae5xGNRpFOpyXvFabxasqg0Wiwbt06tLa2KrqdiKzOagKxIAjiRJtcFISJUtAzHakYagzDVqsVOp0ONpsNdrsder0ewHzP3vj4ODweD4LBoORBmEhjLVMlpA7CSu4dVvJmOqDwK2cnT57Er3/9a8Tj8RKtiJDSosowqRi51Q6lt0k0NTWhs7MTzc3NMBgMCIfDiEajCAaD8Hq9mJ2dhc/nu+m4NUKA+coeVfHIavT19eH3v/89AODpp5/Gpz71qbzCAyFKQGGYVASNRgOLxQJgfhi8ksPwzp07cc8998BoNCIcDiOdTkMQBDidTtTX16OzsxOBQAD9/f3o7e0Vj9OtJFqtFiaTCUajEaFQCBzHged5enOwhOvXryMSiaCjo4MORSgSJfcOAytrl5idncXRo0fF2zabTXyeJURJKAyTipD7pB6NRpc8qU3uNBoN7rrrLiSTSbz11lsYGRmBVquF1WqF1WqF0+lEXV0dGhsbsX//fnR0dODYsWOqaQu5GY1Gg9raWnR1dUGr1UKn0wEAAoEA5ubmMDc3h1AohFQqVdD9KnkT3c2mSiQSCYyNjYHneczNzWHPnj2oqqoq4wqXpuSNdIC6A3EsFsMzzzwj/iw5HA48+uijdGQ3USTqGSYVYakWCaXN3M3OEO7p6cHIyAgYhkEmk0EgEMD4+DguXryIt956C8ePH8fIyAhcLhcOHjxYMS9QbW1t2LZtG1iWxdzcHKamphCNRmG1WrFhwwbs3LkTXV1ddBk3R+6xy9nec0KyFruKxnEcnnvuOQQCAQCAXq/HY489BrPZXO7lEVIUVBkmFUEtm+daWloQj8fFDTqLVbhTqRTGx8cxPj6Oe++9F+3t7aiursbs7Gy5l7tAKSdKMAyDjo4OBINB9PT0IJFIAJhvmXA6naipqYHb7UZzczMaGxvR19eHkZGRih41Fw6HMTk5Kd4u17HLhaDqsPwcO3Ysb3Pmhz/8YbjdbglXRMjaUGWYVAS1bJ5LJBIwGo03rd5lJwFcu3YNHMehsbGxHMuTVGNjI1iWxcjIiBiEASCTyWB2dha9vb04d+4choaGwLIsNm3aVDEv4EtNlci9MuJyuVQX2uRCTdMlzp49i7Nnz4q3Dx48iM7OTimWRUjRUBgmFcHv92NmZgaJRELRYfjatWtgGAZbt25FTU3NTb+eZVmwLFsRo9UsFgsEQUAymQSw+IzTcDiM3t5enDp1CplMBps3bxb7im8m92Q2NZibm8sLaeU+drkQSh6zphbhcBgjIyN47bXXxN/r7u7GgQMHJFwVIcVBYZhUhGvXruGdd97Biy++KPa5KZHP58Px48dRW1uLj3zkI9i7dy8aGxthNpvz5sLyPA+GYcSNUxMTE1ItuWxCoRA0Go04DWGpTZIMw8Dn88Hj8UCr1Vbk7ndBEPKqwg0NDTRFosSUXh0WBAFvvvmm2F9eX1+PI0eOyK6thpDVoJ5hQhTm9OnT0Gg02Lt3L3bs2IGOjg5xUkI0GgXP8+A4Dm1tbVi/fj0GBwcrojIcCASQSCSwefNmsCwLj8ez6Fi5bEgOh8NgWRYGg6HcS5VE7lSJqakpRCIRAPNXD9avXy/l0lZE6b3DSscwDB5++GG8+uqrmJ2dxcc+9rEVX1UhRO4oDBOiMIlEAm+++SaCwSA6OjrQ0NCApqYmtLW1gef5vArxwMAATp06JeFqyyeRSGBwcBDd3d3o7OyEyWTC5OQkIpHIgk1yDMOIO9+VXrErFMdxGBwcFG+3trbCaDRKuKLKofTNdHq9Hg899BAikQhNHSGqQmGYEIWanp6G1+uF3W5HdXU1qqqqoNPpwDAMWJbF9evX4fV6K2pawvDwMJLJJDo6OrB+/Xo0NjZiZmYGXq8X8XgcyWQSqVQKbW1taGlpwczMTEGHkih53nCWz+cTNxjqdDq0tbVJvKKVU0N1WOmBmGVZCsJEdSgME9Xr6uqCXq9HKBTC5OSkuMFKDTiOg9/vh9/vBzD/QiX3U9ZKNV6NYRgIgoCpqSmkUik0NTXB7XajtbUVLS0tSCQSYn+j0WiE3+9HX19f0dchZ9lWiT179mBgYAC1tbXQaullQG2uBzR4f0qHaJqBRSdgb30aGxyFvykeGBhAOp1Gd3f3gs+t5IQ6QpSCngWJ6jU3N4uVjDfeeENVYfhGcg/CpZTtBRYEQeyhrqqqgtvtht1uh16vh8ViQSqVwsTEBAYHByvyqGoAqKqqwi233CL1MlaFqsNLe29SiycvmnF2ZmEv7y21aTy+LYZ9DZkV3dfMzAyOHTuGTCYDr9eL22+/Pa8FC6BATNSDwjBRNZZl83bJK/nADbK4bMg1Go2IRCIQBAHpdBrJZBLBYBDBYBBarRYsyyKVSkGj0VRU68hSaAqAtIodiH8zYMDfv2cBL8xXgx/qYFBrBWYiwIsDAs7O6PDl43Z8a18UH+1YviAQi8Xw4osvIpOZD84jIyPYu3fvoptNKRATNaAwTFTNarWK1YxoNFoRIYhlWej1+ryDJ9SIZVnU19dj8+bN0Gg04pHTkUgEwWAQPp8Pfr8f0WhUfFEHsObHgFL7hgVBAMMweVMllEoN1eFiem9SKwbhL+8B/u+tDKz6P3z+ux9i8JPTwL98wODv37OgwcItWSHmOA4vvviiOG1Er9fjyJEjy05doUBMlI7mDBNVU8vJc1kMw9y0otfU1ITPfvazOHz4cJlWJY2WlhZs3boVHMdhYmICY2Nj8Hq9YBgGDQ0N2Lx5MzZv3kyhCfOTNoaGhhAIBJacv0zKr1iTTJ68aBaD8NduR14QBuZvf+124Mt7AF5g8POL5kXvJztLeGpqCsD8883hw4fhdDpvugY1PL+SykWVYaJquWFYDS0SKwky2TFZar8M3tHRgVAohAsXLohzlDUaDWw2GxwOB9xuN1wuF1wuF0ZGRtDf31+RPcKCIIhTRWZnZ5FKpfIOZFEqtVSH19oucT2gwdkZHSw6Af/31uV/5r+0B/hFj4AzMzpcD2gWbKrr6elBb2+vePvAgQMFTRuhCjFRKgrDRNXUUBnWaDRob29HR0cHIpEI+vr6MDMzs+TX9/f3w+PxyLoCuNaJEm63G3q9Hn19fXkHinAch0AggEAggOnpadTX16OjowPr1q0T5xBXmlgshng8Lt52OBwSroYU2/tT85vlHupgFlSEb2QzAA92MHimd/7P5Ybh0dFRnDhxQrzd1dWFnTt3lmTNhMgNtUkQVVN6GDaZTDh06BA+/elPY9++fTh06BDuu+8+mEwmeDweGI3GRY/SjUajqj51zmw2ixvlgMWr4PF4HENDQ3j33XcRDofR2dlZtKOXx8bGinI/pZatCmdVVVVBr79JYlKQ6elpqZdQFGtpl4im5x/7tSs8TbvOkv/nAMDv9+OVV14R30DX1dXhQx/60KquLinxeZYQqgwTVVN6GN6zZw92796Nvr4+XLhwAW1tbbj11luxfft2zMzM4JFHHkFVVRWA+SN2P/jgA3g8HolXXXqBQAAsy8LpdGJqamrJKjjDMIhGo5iYmMCmTZtgsVgQjUbLvFrphEIhpFIpAPP/FrmX49WwkU5NVtsuYdHNP/ZnIiv7+ulo/p/jeR4vvfSSOHLSarXioYceWtP8aWqXIEpDlWGiWhaLRZwwkEgkFNkvumvXLkxNTeGZZ57B5cuXcezYMczOzmLnzp348Ic/jGg0ir6+PgQCAdTX1+PWW2+FyWSSetklF41G4ff70d7ejq6uriX/ztmQHIvFIAhCRR07zPM85ubmxNtOp1OVB2yopToMrK5CvLd+/nntxQEBkdTyXxtOAi8NCHl/jmVZ3HnnnTAYDNBoNHjooYeKcgVFicUHUrnU98xIyP/KPTJUiZvnGhsb4XQ6cerUKTHIp1IpBAIBbNy4EceOHcPQ0JD49XfeeSe6u7tRV1eH4eFhiVZdHplMBoODg9ixYwfWr18Ps9kMj8eDQCCAZDK5oFJss9nAsmxey4DaBQIBcYycRqNZ0UQAojwbHBxuqU3j7IwOPzk9PzViKf/6wXx7xO7a/BPpWlpa8PGPfxx+vx+1tbVFWxtViIlSUGWYqJbf78fp06dx9epVjI+PS72cgq1fvx7RaDRvs5zdbofBYMDs7CzeffddABDnKF++fBkcx6G+vl6S9Zbb9PQ0Tp06hdnZWdTV1WHbtm3YsmUL1q1bh9raWphMJuj1eqxfvx5tbW3wer1F7aOWc99wJpMRj+gGAJfLteD0MGC+VUINKr06/Pi2GFhGwL98APzDifkKcK5wcv73/+UDgGUE/J9tC38OHA4H2tvbV7vsJVGFmCgBVYaJaiUSCVkHlptxOp2IRCJ5Ve2qqirU19fj1KlT4iEK2SOYeZ4Hx3HQ6RYexSpHa50oAcxX/K9evYq6ujrU1tbC7Xajrq4OmUwGWq0WqVQKer0efr8f165dK9LK5c/n84mPC71en3eVhMhfof3D+xoy+Na+KP7+PQv+5QMGv+gR8GAHgzrLfI/wSwMComkGLCPgW/uiaNd4EImYF2y8LRWqEBO5ozBMiAwxDAOdTofa2loEg0Hx9zOZDLxeL65cuQIgf+6wxWIBwzB5X18JwuEwwuEwJicn4XQ6YbPZYDKZYLPZkEgk4Pf7MTg4mHcKnZrxPJ+3SdDlci07FUAtG+nUMnd4tT7akUSDhcPPL5pxZkaHZ3pzPzvfGvF/tsWwxRbAr3/9OwDAQw89VLYrSRSIiZxRGCZEhgRBwIkTJzA9PQ2O48AwDARBwOTkJJ588slF/0xNTQ10Op0iW0KKIRqNIhqNgmVZsSqa+3GlYFkWbW1t8Pv9SCQSRRsnpwRqCsSrmS6xryGDfQ0hXA9o8P6UDtE0A4tOwN76+R7hTCaD//f/XhDbhV566SV85jOfKdvGSgrERK4oDBMiU9PT02IvZLYCnA3FNzIYDGhqakIgECjaEa9KlRt+Sx2Ex8bG0NLSUtLvsRosy8LlcomtNESZVjtubYODW3C6nCAIeP311zE7Owtg/jFy3333lX3CCAViIke0gY4QBVlqnm5nZycaGxsxMDBQ5hXJW6UHwZX+/dWykQ5Q12Y6YG0HcuQ6c+YM+vv7xdt33nnnmnv2V4s21RG5oTBMKp7SA6TBYIDJZEJfX1/eix1Z+s2DGlVaOwhZuaGhIZw6dUq8vWXLFmzbtk3CFVEgJvJCYZgUjdpn28qRx+NBMpnEBx98IB47rCSl6m9ubm7Gvn37VnWJuVBymFgiCAJGR0cxOTkpnjhXKKoOy9daqsNzc3N49dVXxduNjY04ePBgMZa1Zkp7viLqRWGYEBXgeV7V0xJsNhtqa2vR1NQEp9O57OV/hmHQ0NAAl8sFh8NRxlVKJxgMIp1OIxKJYHx8nKrEKrSaQByPx/HCCy+Ih/bYbDY8+OCD4smcckCBmMgBbaAjhMgWy7JoaGjApk2bYDQawfM84vE4BgcHl6zICoKAy5cvo76+HiMjI2VecfnxPJ8XlBwOx6IHbFQaNU2WyCpkQx3HcXj55ZfFOeU6nQ5HjhyR5XHttKmOSI2eMQkhstXY2IitW7eCYRhMT09jcnISOp0OW7ZsWXbzTywWw+DgoHgcsZr5/X7x76nVatdUDVdTqwSgvnaJQlVVVYkf33fffaipqZFwNcujCjGRElWGCSGytXHjRkSjUZw/fx6RSAR6vR7BYBCbN29GW1sbPB7PgpYAjUZT9hAs1Yi1lR67TNRjpdVhjUaDu+++Gy6XC+l0GuvXry/D6taG47iyj3ojBKAwTAiRqdraWhiNRgwMDCASiQAAUqkUhoeHYTKZ0N7ejrq6OkxOTorzlw0GAzo7OxGPxxU/JWQl5ubmxIkZer2+KJea1XIiXVYlt0swDIMdO3aUYUXKxXFcWarSFPTljUoIhBBJLTVRwu12IxaLLXq89NjYGJLJJNrb2wH8YYSaw+FAc3OzuGFIzVKplNgPCsyfQFjpc5UryWIb6hKJREWNEySkWCgME0Jkh2VZGAwGMAyTF/iyIpEIpqam4HA44HQ6AcxfFna73eKYsXIr94g1r9crfmwymWA2m4t239Q7rAy5gTiVSuG5557Dq6++qurJMoSUAoVhomiVUgmz2WyyGodUDoIgLPui7vF4AMxvsgMAu92O2tpaeDwe1VfH4vE4otGoeJuqwpVNEAQcO3YMc3Nz6O/vx/PPP6/6nwFCiokaWIiiCYKAtrY26PV6aDQa6HQ6JBIJhEIhBAIBJJNJqZe4Znq9Hu3t7RAEAZFIBENDQ1IvqeR4nsfIyAgsFsuSG+ICgQC8Xi8aGxtx9epVOJ1OGI3Givj3yQ3CNpsNRqNRwtUogxp7h4H56nB/fz8GBwfF3+vu7qY3R4QUgMIwUayWlhY88MAD2LFjhzhCiOM4xGIx+Hw+jI2N4dy5c7h06RJisZjEq109q9UKAOImsRt5PB6xOqomPp9v2YMGBEHA5OQkampq0NnZCZvNhlAoVBEjmmpqamA2mzE3NweXy1WS76G2jXSAOgPx6OgoTp8+Ld7esWMHNm/eLOGKCFEeCsNEkdra2vCZz3wGmzZtwsjICAYHByEIAjQaDWw2G6qrq9HW1oY77rgD4+Pj+MUvfoFr164tGSjlLHdCgFqD3vj4+LJzg5cyMzODeDyO5uZm6HQ69PT0lGB1K1fOEWtms7mofcJEeXw+H06ePCnebmlpwe233y7highRJgrDRJEOHTqE9evX47//+7/x+uuvIx6PA5g/dMBqtcLlcqGtrQ27du3C9u3b8Wd/9mf4p3/6J/T19Um88sJlK8MAxBFjZF4ymcTk5CTWr18PjuMwMTEh9ZJUharD8hWPx/HWW2+JLURWqxX79u2jOdOErAKFYaJIO3fuRE9PD1544QUIggCWZcHzPDKZDAKBAAKBAK5fv47jx4/j8OHD+JM/+RMcOHAAQ0NDihq7ZTabxY1zqVRKFT3QxTY+Pg6Hw4GZmRmpl0IUQumBmOM4vP3222L7l06nw1133QW9Xi/xyghRJnoLSRRHq9XCZDJhdnYWgiCAYZgFp5AxDCP+euWVV3D58mVs375dcS8WldAisVaRSAQffPABhoeHpV4KgPKPWCs1tY1ZUzpBEHD69Om80Xq33367uG9iuT57QsjiKAwTxTGbzfAedUoAACAASURBVAgEAti4cSMALNoDLAiC+IthGHg8Hthstrxd+EqQG4apRWJpmUxmwRsiQpaj1NnDyWQSU1NT4u1du3Yt2EBLgZiQwlAYJooTCoVw+vRpdHR04JOf/CQaGxuXrfhWVVWhra0Ns7OzZVxlcSQSCaTTaXGsGiFSoOqwfBiNRhw+fBhutxvt7e3o6upa9OsoEBOyctQzTBTp2LFj6O7uxkMPPYSdO3fi9OnTuH79OmZnZ5FIJADMn0hmNBpx//33Y+PGjXj22WclXnXhskcV6/X6RWftqslqJ0rIUTmnSpDVU2rvsMlkwj333ANg+YOHfD4fqqury7UsQhSLwjBRJJ/Ph5/97Gd44IEHsGvXLjzyyCMA5lsJAoEA4vE47Ha7+EL35ptv4pVXXpFyyWuSSqWkXgIhqqTUQLzSEykpEBNycxSGiWJNTk7il7/8JU6fPo0tW7agra0NNTU1sFqtMJlMSCQSGBwcxLFjx/D+++/TJAZC1kCNY9aUgOd5vPvuu2hvb0dTU9Oq7oMCMSHLozBMFCt7TO+lS5dw6dIlAPP9wVVVVdDpdAgEApibm5N4leWh1lPoCCkHOVeHe3p6MDo6itHRUezcuZNOlyOkBGgDHVGsbA+tTqeDVjv/vi4YDGJ0dBTXr1/H3Nzcsv10cuZ0OtHa2orq6mrx71ZpOI4Dz/OKOzEwS20j1gB1b6ST43SJoaEh9Pb2irfX0i5FG+oIWVplvsoSxXM4HLj11lthNpuh1+vB8zyi0Sh8Ph8mJyfh8XjAcdyKg9TAwAA6OjpKvOqVq6qqgt1uh8PhwMTERMVUuHM30Wk0GvA8D57nwbKsYt/YELIaXq8X7733nni7qakJO3bsWNN9UrsEIYujMEwU55577sGRI0dQW1u74HOJRAKzs7O4fPkyTp48ieHhYcXNn2UYBhaLRbxdyYdtVMLRsizLwuFwgGVZsCyLRCKBUCgk9bKWpObeYbm0S8RiMbz99tvic1dVVRUOHDhQlDeEFIgJWYjCMFGUO+64A5/4xCeQyWTwu9/9DoFAAMB8q4TD4UB9fT1aW1vxwAMP4N5778W//du/4Z133pF41YXJPYI5mUzSJAkFu9mINbvdDrfbDbPZnPf7qVQKPp8PwWCQ/v/LTOpAnMlk8NZbbyEejwOYH6t48OBB6HS6on0PCsSE5KMwTBTl0KFDiMVi+Od//mcMDg7mfY5lWVitVtTW1mLr1q24++678YUvfAE+nw9XrlyRaMWFo1PnKgPLsuJ0gKmpKaTTabAsC6fTCbPZjPr6ejgcDszOzopv+uRCzdVhKQmCgPfff1/s72UYBnfccUfec0KxUCAm5A/Ufw2SqAbDMGhvb8d7770nBuHcy4Y8zyMUCmFgYAC/+c1v8NRTTyGVSuHOO+8salWl1KxWq/hxJbZIJJNJ8dQ9NctWHycmJsTA6/P5cP36dQwPDyMcDsNoNKKlpQVNTU0rnitL1k6qzXS9vb0YHh4Wb+/evRv19fUl+360qY6QeRSGiWK4XC5xwgAwv8FqscCUDcjnz59HX18fOjs7kU6ny7rW1dJoNDCZTABQkUcw6/V6CIIAjuOQSqVUEYgXmyqRvYqRTCYX/T8Oh8MYHh7GxMQEOI5DdXX1oj3yUlLzZAmg/IE4FAqhp6dHvL1hwwZs3Lix5N+XAjEhFIaJwszOzmLXrl2wWCxLHk+cDcMGgwGxWExRm7CsVqu4/lgsprjNf2uVfSMAzL8xUOsECUEQIAgCWJZd9v/Y5/NhYGAA8Xgc1dXVeVcNiLrY7XYcOHAAGo0Gbrcbe/bsKdvjnwIxqXTKSQmk4nm9Xpw+fRotLS348pe/jD179qCmpmZBC0Q2XHR0dKCzsxPXr1+XYrmrspZ+YY/HU+zllJVGo8n7v1RzW4AgCEgmkzCZTHA6nct+bSqVwtzcnFhNlhOqDhdXW1sb7rvvPtx5551lf/xTICaVjDbQEUU5evQo3G43br/9dmzduhWXLl3C8PAwpqenEYlEIAgCtFotqqur8eCDD4JhGJw4cULqZa9YJfcLG41G8WO1zRVebKpEIBCAw+GA2+0Gx3HLjlMLBoNobGxU9RsEuSr3dAkpN7XRpjpSqSgME0VJpVL45S9/id7eXhw4cAAbNmzAtm3bxM9xHCdeao9Go3j66adx7tw5KZe8Yrkn6XEch1gsJvGKystgMIgfV0LoC4fDmJycRF1dHZqbm+H1ejE3N7do+0929NpSrUFSqoTJEqUKxP39/Whtbc177EuNAjGpRBSGieLEYjH8/ve/x+XLl9He3o7m5ma4XC5YLBbwPI9YLIZgMIg33nhDUSe3pdNpXL58GVarVVHTL4rBYDCIvd2ZTAZer1c8iU7N/H6/2CNaV1cHh8MBr9eLaDSKTCYDjuOg0+ngdDrBMIyiHs9keQMDAzh9+jR6e3tx1113oaqqSuoliSgQk0pDYZgoltfrFfuIgfnNVwzDKLqiKghCxbVHAPkb5xKJhIQrKZ3FWiU4jsP09DSi0SjcbjesViuampqQSqWQTqfB8zxMJhM0Gg1mZmZkOxWFqsOFmZmZEZ+3IpEIent7sX///qLcd7FQICaVhDbQEdWIx+NiEFZTv6naabVasT0ku7FMjfbv358X+nNFIhEMDQ1hdHQUwWAQHMfBbDbDZrMhnU5jfHwcMzMzZV4xuVExNtRFo1G8/fbb4thAp9OJPXv2rPl+S4E21ZFKQZVhokpqmE9bKXI3ziWTSVX+323atAlOpxPhcFg8ZncxwWAQwWAQOp0OPM9Dq9WKFWK5q4TqMLC2CnEmk8Hvf/978Q2fwWDAwYMHxTeDckQVYlIJqDJMVKm6urokR5iWis1my5sxXEmyvbFAfovE+Pi4VEsqKr1ej/Xr12N8fBx9fX15n2NZFlqtdkGPeDqdBsdxSCaTigjC5OYEQcDJkyfFo7VZlsXBgwdhsVgkXtnNUYWYqJ18344SskoajQZf+tKX0NXVhR/84Ae4dOmS1Eu6qbq6OpjNZvA8j8HBQUX3PRcqkUggkUhAq9Uik8lIvZyi6+7uRjqdxuTkZF7Pb21trbgxjmVZhEIhTE9Py7YveCWoOry0ixcv5p1GeOutt8Ltdhd7aSVDFWKiZlQZJqqj1WrFS9FKqGhoNBpxdBbDMGvaQKbkgzfUGISzG+ImJibEimB2UkZdXR20Wi04joNGo4HT6cTGjRtht9slXjVZiUL6h0dHR/PelHd2dmLDhg2lWFZJKeH5lJDVoMowUY1si0EymcSPf/xjWCyWgk9xk0JuO0clHsGsZl1dXUgkEhgbG0Mmk4FWq0VLSwucTifm5ubg9/uRSqWg0WjESnF1dfWyB3DIXaVUh4GVVYgDgQBOnjwp3q6vr8ctt9xS6qWVDFWIiRpRZZjIns1mg8vlWvRzDMOIIVgQBHHzlSAIigjCQOWeOpedK6xWdrsdtbW10Ol0aGpqAsuy4sfBYBAejwfxeBwcxyGVSom3bTabovrdF6P2Y5oLYbFY0NDQAGD+Z/32229X/GOfKsREbagyTGTvwQcfxMGDB/GDH/wgr+eusbER+/fvx969e8GyLMbGxnD27FmcPXt22R37cpMbfColDLMsC6fTiUwmg0QiseQ4tfHxccUevhGJRDAwMICmpiZs2LABDQ0NCIfDMBqNOHXqFGpqavK+nud5BAKBJcevEXlaqjqcfWOu0+lw55134tKlS7I7bW4tqEJM1ITCMJG95uZm+Hy+vIC7ZcsW/Omf/ilqa2vB8zxYlkVDQwN27tyJ6upqvPDCC4poNzAajeIkgUwmo6gQvxZGoxEMw0Cn06l2tjDP8+jr68PIyAjWr1+PlpYW1NXViRXgxWQrhmqYKkLtEn/4f2QYRjw2Xk0oEBO1oDBMZK+5uRlDQ0OIRqMA5kdV/fEf/zFqa2tx/PhxvPPOO4hGo+ju7sZjjz2GT3ziExgdHUVPT09B32dgYAAdHR2l+CssKbdFQiltHcWQWx1T64lzWclkEr29vRgfH8f69esRj8cXDbsajQZGoxGZTEZ8rBPlyA3EwWBQVscrlxIFYqIGym5cIhXB5XJhZmZGDE3bt29HW1sbjh49iqeeegr9/f3weDx4/fXX8R//8R8AgB07diyY3SpHuS0SlRKG9Xo9NBoNgPnqaSqVknhF5REOh9HT04OhoSEIgpDX8gNA7BUOBALi3GWlq7Te4enpaUxNTeHFF1/EmTNnFHF1qhioh5goHYVhImvV1dVgWRazs7MQBAE6nQ6dnZ3gOA6/+93vFnz922+/jZmZGWzcuFH2L0QMw+QN3K+UfuHcE+dWUhVWy+EbWYvNETaZTHC5XOA4DlNTUxKsihRDLBbDO++8A0EQcO3aNZw/f17qJZUNBWKiZBSGiay1tLQAgHgIRTqdBsuy8Pv9iMViCy43GwwGTExMwGw2y766ZrFYxB7RZDJZtMMW5DxrmGVZ6PV6APMbjNTeIrESGo0Gzc3NMBqNmJqaUt1x1JVSHc5kMujp6RGvdJhMJnR1dUm8qvLy+XwUiokiUc8wkbVsGP7c5z6Hhx9+GF6vF62trZiZmQGABcHBYrHAYrGIBxzIWSKRwPj4OGw2myo3kC0md1JCKpWSffW+1MbGxrBx40ZEIhEkEglFPG7JQoIg4NKlS+Kb9uxRy9nDdCoN9RETpaHKMJE1r9eL3t5e+Hw+uN1u3HLLLXC73eB5Pm+DSrZCXFNTA7fbjYGBAamWvGKZTAY+nw8jIyMVc2m8kjbOrVQikcDU1JSqg7Daq8PXr1/H3NyceLu7u1uVJyoWgirEREmoMkxk7dSpUzh16hQAwGw2w+12o7m5GQaDYdHLybt374bZbMbVq1fLvVRyEwaDQWwL4TiuaG0hSjc2NiZeAVEztY5am5qawsjIiHi7ra0N9fX1AFZ2Qp2aUYWYKAWFYaIYsVgMIyMjeS88WYIgwGAwoK2tDb29vRgcHJRghWQ5hW6cy6XkwzeWYrPZ4HA4FkyVIMoRCoXQ29sr3na5XNiwYUPe11AgpkBM5E/VbRJTU1P4i7/4C2zYsAFGoxF1dXX48Ic/jNdff31N9/vcc8/h4YcfRkNDA3Q6HWw2G3bu3Imvf/3rmJ6eLtLqSaE4jsNTTz2Fp59+GsFgUOrlLEurrbz3oZFIBPF4HDzPq6pFQq/XF3y8LsMw6OzsxLZt2ypqk5Wa2iWSySQuXLgg9r2bzWZs3bpVFQemFBu1TBQf5ZviUu0r8oULF3DPPfeIfVx2ux1erxe/+93v8MILL+D73/8+vv71rxd0nzzP47Of/Sx+9atfib9ns9kQi8XQ09ODnp4ePPnkk3j55Zdx6623FvXvQ24uk8ko5od148aNEAQBkUgEHo+nIjaScRyHaDSqugMltm3bBp7nMTo6mtc3uhxBEDA5OQmr1Yrx8XFEIpGKaJVQk0QiIU6s0Wq12LFjx5Jvciu9Ogz8IRBTlXjtKN8Unyorw/F4HB/5yEcwNzeHXbt24dKlSwgGg/D7/fjrv/5rCIKAb37zm3j11VcLut8nn3xSfKD85V/+JaanpxEKhZBIJPDyyy+jtbUVPp8Pn/zkJysi3JDVyR7BrNfrYbfb6bGiYA0NDairq0NdXR22b9+O7u5u2O32Ff1Zj8eDt99+u2IOW8lSS3W4qqoKe/fuhdVqxbZt2246OUIpb9RLjarEa0P5pjRUGYZ/+tOfYmRkBFarFUePHsWWLVsAzL97+uEPf4iPfvSjEAQB3/jGNwq63+zpZvfccw9+/OMfo7a2FsB8VeDw4cN4+umnAczvLL5w4UIR/0ZETcpxBLOcZw2vlhwP32hubgbP8xgfH4cgCGhvb8fOnTvR0dGRd6AK8IeJJ7ktFbkvKpXUO6yWQGwymbB3794VVzspEM+jQLx6lG9KQ5VhOPvu5lOf+hSampoWfP6rX/0qAODs2bO4du3aiu83+0S2a9euRT+/e/du8WO1XQomxZN7BHMlnDpnNBpV2Udpt9ths9kQDodx6dIlfPDBBxgcHITBYEBnZyd27NiBdevWiePkstNPWltbcdttty0Iy0T+FptgU+hjmwLxPArEq0P5pjRUF4bD4TDOnDkDADh8+PCiX7N//35xRm0hzebr1q0DAJw7d27Rz2e/r8FgwObNm1d8v6Ry3HgEs9ovket0OlitVlRXV6su/BmNRhiNRrHKGYlE0NfXh9OnT2N8fBx2ux3d3d3Yvn27+KJlMBjQ2NgIu92uyheUQiitOhyNRnHy5MmizIOmQDyPAnFhKN+UjurCcG9vr/juPXv54EYsy2LTpk0AgCtXrqz4vh9//HEAwPHjx/GVr3xFPAUtk8nglVdewec+9zkAwHe+8x04nc5V/x2IeuUewZxIJFQ/azc7Tk2NleHZ2VmcP38efr9f/D2e5xEIBHDlyhV88MEHmJmZgdvtxtatW7Fz5050dXXB4XAsWbGppFYJJUmn0+jp6UE8HsfZs2eLEmYpEM+jI5xXjvJN6ahumsTk5KT4cWNj45Jfl/1c7tffzMc+9jF873vfw3e+8x088cQTeOKJJ8TdlhzHYcuWLXjqqafw+c9/fsX3mXvpIVfu7EqiHpXUIsEwDPR6vXi7GOPU5DRvWBCEJXuzM5kMvF4vgsEg3G43WltbxeecWCyG4eHhMq5UvpRwEAfP87h06RLi8TiA+cd1sY5ZpikTf1AJgbi3t3fJ1/xs5XU5Sss3SqK6MJx76dFkMi35ddkns0IvU3/jG99Aa2srvvjFLyIajeYFmmg0Cq/XC57nC547SipDOTbPyUVur3A6nRbHUFWSdDoNj8cDn8+H7du3o6am5qZvdCvlRLosuQfigYGBvKC2ZcuWvDe1a0WB+A84jiv7DHaO48oSxIvx/Ef5pnRUF4ZLKRwO45Of/CReeOEFPPzww/j2t7+N7u5u+Hw+vPzyy/jmN7+Jr371qzhz5gz+8z//c0X3udS7wd27d+P8+fPFXL4qtLS0wO12IxQKYXp6WlHVVa1WKz6B8TxfEWE4S02HbKyGxWKBzWZDIBCgy+MK4vF48lpX2tvbxV32hBSqu7t7RRVgKZQi3yiJ6uJ97iad7GWtxcRiMQD5lbqb+au/+iu88MILuPfee3H06FHs3bsXNpsNbW1t+MIXvoBnn30WDMPgv/7rv/DSSy+t/i9BllRXV4d169aJVbZiGxgYKPp9ZuU+1mKx2KI704tJyvFqer0eGo0GwHzwTyaTkq1FaizLorGxEQaDYcXtT5XWOyzHzXSBQABXr14Vb7vdbrS3t5fke9EbJLISlG9KR3VhOLePZrkwkP1cQ0PDiu43FArhqaeeAjA/kHoxd911F2655RYAwPPPP7+i+yWFyT3QIBQKSbiSwgmCID5JKamivRqlrArLcd7wchiGQSqVwujoaN5mO5JPToE4kUjgwoUL4htWq9WKzZs3l3QjKAVicjOUb0pHdWG4q6tLfMK6fPnyol/D87y4m3ulI0L6+/vFnp/lqgPr168HANogUyJK3oAWDAYxMDCAy5cvq3qzCMuy0Ol0AObfAFR6iwTHcbh27VpBO7uJdDiOQ09PjzjpRafTYfv27WXpZaVATJZD+aZ0VBeGbTYb9uzZAwB47bXXFv2a9957D8FgEABw6NChFd1vbsP4chWMkZERcR2kuCwWi3jpPZFIIJVKSbyi1eE4TtWbyW7cOKfGoztXo9B/h0prlQDkUR2empoS+/kZhsG2bduW3axUbBSIyVIo35SO6sIwMH8yCzB/Ustio0V++MMfApjfpJadx3czmzZtEk+SevLJJxf9mrNnz+Ls2bMAgH379hW8brI8JbdIVJJij1MjlUXqQNzY2IjOzk4wDINNmzZJMlOVAjFZCuWb0lBlGP7CF76AtrY2hMNhPPzww+LlyXA4jK997Wt49tlnAQDf//73F/xZhmHAMAy++93v5v2+2WwWh04/++yzePzxx8XKTSKRwPPPP4+PfvSjyGQysNvtqp3FJyUlt0hUkkAggFAoVNLqvdL6hlerEqvDUmMYBi0tLdi/f/+ix92WCwVishjKN6WhytFqJpMJzz//PA4dOoSzZ89iy5YtsNvtiEQi4HkeDMPg+9//Pu6///6C7vdHP/oRrly5gnfeeQc///nP8fOf/xwWiwXxeFy8BGqz2fA///M/JZl0UOmmpqbAcRzsdrt4Oo4SMAyD9vZ2cW5jdhOdmqVSKcW2sRDpyWH2cLEO1lgLmkFMbkT5pjRUWRkGgB07duDSpUv48z//c6xfvx7JZBIulwtHjhzBa6+9hq9//esF36fVasWbb76Jf//3f8f9998Pt9uNZDIJk8mErVu34itf+QouXrxY8IOQrEwoFML169dx7ty5gk7WkZrVaoXVakVdXZ2klSaiTJVaHS5Xu0QqlUJfX59s+/ipQkxuRPmm+Bih1MNOyapkD90oZE6gHKxbt07qJaxZR0dHUe+voaEBbrcbADA7O1v2IL/csZ1KJ8XRzOl0GizLgmXZko7aylVJJ9LlKnV1mOd5nDt3DoFAADabDdu3b88bCygnlVAhfvTRR2EwGMp2MMXu3buRTCbx3HPPlfx7lfvvRgqj2sowIXJRCb3ODMPAbrfnbZ5TI57nwXEc0uk0UqlUyQ9OqXSlrA4LgoBr164hEAgAmP/ZlPOpkFQhJqR0KAwTUkI6nU6sNPE8n3e2vJoYDAbo9XrY7fa8qR9qk8lkxI+zm1HKoVJbJYDSBeLx8fG8gws2bNgg+15ICsSElAaFYaIIGzZswP3334/bbrtNUZf9c9tcotGoaiuJuZeW1bpxjuO4vFnB5TiEgcwrdiD2+Xzo7+8Xb9fV1aGtra2o36NUKBATUnwUhokiVFVVwWq1oqGhoawD8Ncqt0VCzpdg10Kr1YrBUBAEJJPJsn3vco1YEwQhryqs0WjyBtWXQyVXh4spFovh4sWL4htTm82G7u7uslX5i4ECMSHFRWGYKIJS+25zK8NKWnchcqvCyWRSldVvhmGg0+nEAExV4fIrRnU4k8ngwoUL4hsbvV6P7du3iydbKgkFYkKKh8IwUQQlhmGTySSGpnQ6LdlpbLl9kcXGMIx4chEAxOPxkn0vqbEsC71eD71eL1kVsdKrw2sJxIIg4PLly2LfPsuysp4esRLT09MUigkpAgrDRPaMRqM4pSCdTismcOVWhdXaImE0GsVgmE6nJZnVWu7T6MrdHkHyrTYQj4yMwOv1ire7urpQVVVVrGVJigIxIWtDz+pE9nKnE4RCIQlXUpjcE6yUUs0uVG5VTarKd6Wp9OrwajU2NsLhcACYn1/c0NAg8YqKiwIxIatHYZjIXrlbJAYGBopyPyMjI+jr68Pk5KQqK8M6nU7steR5vqwb58olnU4jk8mosg9ayVZTHdbr9di1axc6OzuLfrCOXFAgJmR1KAwT2VNqZRiYr5bOzs7mTSJQixs3zqlN9oCNTCYju42BVB1eXSBmWRYtLS2KmhxRKArEhBSOwjCRPSWHYTUTBEEMiFL3cZeib/jGAzaI/CwXiDmOg8/nK+Nq5IMCMSGFoTBMimp4eLjo95nbJkFhWD4ikQh8Ph9CoVDeYRRqwPN83t9Jp9PJLhBTdXhpgiDg6tWrOHfuHIaHh2VV1S8XCsSErByFYSJr2WN+AWnHkxVCq9XC6XRCp9NJvRRRqcarCYKguhPnBEFAOp0Wb7MsSxMkZGyx6vDo6CimpqYAANevX8fc3Fy5lyULFIgJWRmaHE9kLZlM4ujRo7BarXnzbOXMbrejubkZADA3N4eJiQmJV1QZxsfHxX/3teB5Pq+SKOcDNsbGxtDS0iL1MiQ3OjqK1tZWAIDX683bBNvQ0ACXyyXV0iSXDcR1dXUSr4QQ+aJyB5G9dDoNv98vVnrkLne+sBo3lqmZHI5dJqszOjqKaDSKS5cuib9XVVWFrq4u2bW4SIGqxIQsjZ7lCSkyNR/BrNFoUF1dDavVqsgjbG+G4zjFVIWzqHd4HsdxOHPmjHjwi8FgwLZt2+jNTA4KxIQsjp4lCCmiG49gVltl2Gg0gmVZGI3GvENF5GItUyVurAprtVqqKCqEIAiYmpoSe71ZlsWOHTsU01pVThSICVmIwjCRterqanEDnRKU+4CQcssNF0rYzFiIG4+SVlLlu9Krw16vF7FYTLy9efPmvJ9Fko8CMSH55H8NkFQsvV6PD33oQwCAWCyGl19+WdoFrUDuC7DcTp3zeDxobGxc9Z83GAziJWeO4/ImLqhBNvxmMhmqCitIMBhEIBAQb1dXV6vuikwp0MY6Qv6AKsNEtnKDpRLGd7Esm9c6oLbKsMlkEj+W+pCNUmAYBlqtFgaDQVFV4axKrQ6zLCu+cbFYLKiurgawuhPqKhFViQmhMExkTGknz1mtVvFFORaLLbjsrmRarVbshRYEQdaVt7WeRscwjGKrwpUYiG02G1paWmCxWFBfX5/3f0eBeGUoEJNKR2GYyJbSTp7LnSIhtxaJtTIajeLHyWSyIk/0IvJlMBjQ2Ni46OQICsQrQ4GYVDIKw0S2civDSmg5UOvmOYZhVLtxjud5pFIpVR0nrfbqsCAIqrrqIifT09MUiklFojBMZEvKNoncE6xWanJyEl6vF/F4PG9nu9IZDAbx0nMmk8kbPyZXK2mVyB67nA3ESvh7EcDv92NsbKygVh2qDheGAjGpNBSGiSzp9Xrx0jzHcYhGoxKv6OZCoRA8Hg/6+/tl20bg8XgK/jO5LRJq2jh347HLajqcQa3V4Ugkgrm5OaTTaQrEJUaBmFQS9Tz7E1VRa8uBEsXjcbGCKueNc4WgY5eVJ5lM5h3JbjQaC55BToG4MNQ2QSoFPfsTWVLa5jk1SyaTCAaD8Pv9Ui+lIMu1Sijx2OVCqak6zHEcPB6P+H+m0+nQ0NCwqqkfFIgLR4GYqB2FYSJLSqoM5845VTO5tn4Uio5dVhZBEDA5OSn+kOaVGQAAIABJREFUn7Esi4aGhjXNgqZAXDgKxETNKAwTWUqlUgiFQuB5XvZh2O12Y8uWLWhvb88br0bk6caNcko8YGOl1FAdnp2dzetVr6ury5tusloUiAtHbRNErSgME1m6du0ajh07hueffz6vT1COrFYrWJaFzWZTVbDS6XRSL2HNbmyVuHEsl06nU31VWMmBOBAIIBgMirddLldR33BSIF4dCsREbSgME1kTBEHWM2Bzj2AWBEERh22sZKKEXq9HVVUVnE5n3jQJpcutCjMMQ5vmZCwWi2F2dla8bbVa4XQ6i/59KBCvDlWJiZrQKwEha5B7BHM8HlfNYQDZAKymKQs3VoUrqVdYidXh3PYog8GAurq6kv1/USBePQrERA3U8SpHiESUtNFvpViWFUdWCYKg+BPnsq0SDMNAr9eLFWE1tbSoUW1tLaqrq6HRaJY8armYKBCvHlWJidJRGCay09TUhM7OTtTX1xc8R7TccvsXldAisRK5bRHZ+cJqkQ36auiHLpTSqsMMw8DlcqGtra1so+8oEK8NBWKiVBSGiey0trZi69atOHDgAGpra6VezpL0er24q53jONUcwazGE+dyN9IxDFMx7RE3UlogBso/7YMC8dpQlZgoEYVhIjt2u138WM4HbuS2SESjUUXN4V1qE53BYBAvR3Mch3Q6Xc5lkQoWDocxMzMji58jCsRrR4GYKAmFYSIrGo0GFosFAMDzvKStBwMDA8t+Xo39wrlVYaX3CmdVVVXBaDTKImTJgRyrw4lEAtPT0wgGg5iYmJDFRlQKxGtHVWKiFBSGiazkVoUjkYis+1WzoR1QRxjWaDRiL60aNs4B8+Fep9PBarUiEolQIP5fcgrEmUwGk5OT4v/NjYeiSIkCcXFQICZyR2GYyIpSqq1arRbRaBQcxyGVSiGVSkm9pDUzmUzix6lUSvHBkWEYcQY0ACSTyYrtFZYrnucXHLXc2Ngoq0kfFIiLg6rERM4oDBNZyQ3Dcu4XzmQyGB4exuXLlzE4OCj1cooid3KHGjbOmUymvP5nNVS6i0nq6rAgCJiZmcn7f5HrBBkKxMVDgZjIEYVhIiu5bRJyrgznUmpV+MZNdH6/H9FoFKlUSlaXqleDZdm8Snc0GgWw8HhmIp1AIJD3M15TU5PXeiQ3FIiLh6rERG4oDBNZUUqbhBoJgoB4PC7rivxKmc1msSUinU4r9g1LqUlVHY5Go/B6veJtu90Oh8MhyVoKQYG4uCgQE7mgMExkg2VZsTIkCAKFYbIqGo1GnP8MYMH8Z6oO5yt3IE6lUpiamhJvG41GuN1uxfRzUyAuLqoSEzmgMExkw2q1ii+IsVhMtpMk6urq0NTUhKqqqpIfEUsKZ7FYxMdRKpWiWckyIggCpqenxZ9trVaLhoYGxf0cUSAuvunpafpZJZIpzxmXhKxAKpXCxYsXYbfbkUwmpV7OkpxOJ/R6PVwuFwYGBhR98lx24kIikZDFbNe10ul04gYsQRDEXmGyvLGxMbS0tJT8+zAMg/r6eng8HqTTaTQ0NJTtqOViGx0dRWtrq9TLIGuUTqfLUplOp9N5V6yIvCjzWYioUiKRQH9/v9TLWJbBYBDDlhqOYE6lUnC5XDCZTIjH44oPj7mb5pLJ5JIBf3x8HM3NzeValiKUKxDrdDq0tLQgmUzmHfKiRBSICVEHZV2bIkRiVqtV/FjK0/GKxeVyiR8rfYIEML/pMttio/Q3KlIoV//wjdM+lIxaJghRPgrDhBRATdMubDabWOXmeV7WrSkrJQgCYrEYfD7fTXvOaSNdecTjcVXMrV4OBWJClI3CMCErxDBM3hxUpVeGc6vCdCAFySpmdTidTmNychLj4+MIBoNFu185Gh0dpVBMiEJRGCayYLPZcN9992Hfvn3YsGGD1MsRDQwMiB+bzWbxmNhkMqno2bU6nS6vyk1hmOQqRiDOHrWc7dv2er2q2KR5MxSICVEeCsNEFux2O2w2G5qamlBXVyf1chalpn5hl8sljh8Lh8OyHWO3EmazGTabbVXjuahVojSyI9RyW28aGhrEN5NqR4GYEGWhMExkIfcYZrmegKaWfmGGYVBdXS3enpubk3A1a5PdiGUwGOB0OismbJXDWqrDfr8/7w2j2+2G2WwuxrIUgwIxIcpBYZjIgtzDsEajEXe/C4Kg6MpwVVWVONs1lUrJ8t97pXKPXc5kMqu6DE/V4aWtJhBHIpG8N1hVVVWKOGq5FCgQE6IMFIaJLMi96pobuuR8Ot5K5FaFfT6fhCtZG61WmzenVukzkuWqkECcTCbzjlo2mUxwu92lWJZi0MY6QuSPDt0gkmMYJq8fV45hOBwO48qVK7BarYoOwgzDiL21giAoOgznXnZPJpOqmJOsZBzHwePxQBAEAH84ajn7JrLS0QEdhMgXVYaJ5CwWixjQYrGYbENNJpNBIBBQdFuBIAgYGBhAf38/PB6P+G/t8XgkXllhbjx2ea0HbFCrxPJuVh0WBAGTk5Pi44lhGDQ2NlIP9w2oQkyIPFEYJpKTe4uEGsXjcUVvnMud95xIJCpiZJfUlgvEN17dqa+vh8FgKMeyFIcCMSHyQ2GYSC538xyFYXIzBoNB3ABYzGOXqTp8c8sFYofDgaamJtTU1OQFY7IQ9RETIi8Uhonk5F4Zdjgc4iQJIr3cXuF4PC72qJLyWC4Qm81mOJ3OMq5G2SgQEyIPFIaJ5HLDsNz6cVmWRXNzMzZu3IjNmzcrtgfSbDajs7MTLpdrVYdTyIVerxf/D3ieRzweL+r9U3V45TKZjKI3k8oFBWJCpEfTJIjkTpw4IZ5AFwwGpV5OntzwyHGcYntTq6urYTQa0dTUBKPRiImJiQVf4/F40NjYKMHqVi6VSiEYDMJsNtMR0hIaHR2FIAhgGAYNDQ3Q6XRSL0nRsoGYpk0QIg3lloiIaqRSKXi9XgwNDclukkTu0dBybOFYCY1Gk3fogZLHqQFAOp1GMBjMO+q3mKg6vDxBEJBOp5FKpZBMJjE+Pk4V4iKhKjEh0qAwTMgyamtrxY+VGoarq6vzRtcVu7WAVJYb2yOU3nojNxSICSk/egYjZAkGg0GsqAqCoNgTznJPnFPyODUivRtbhTQajexam9SApk0QUl4UhomkLBaLOCZLbnKPkY1Go4q8FGy1WsV5r9lDQ5Yjx8M3tFotnE5nWefWUqvEQjzPI51Oi7dZlpXtz65aUCAmpDzomYxIat++fXA4HIjH4zhx4oSspknk9gtHIhEJV7J6LpdL/Njv9ytyDJnFYoFGo4HNZoNGoynaXGGycoIgIJVKibcZhoFOpxOPWh4bG0NLS4tUy1M1OsaZkNKjyjApuuHh4RV/bXY4v8lkkt10AKX3C+t0urwDTZS4cU6v14uTCgRBKOtjhKrD824MwgDygnDWzY5sJqtHbROElBaFYSIZs9ksXmZNJBILXnClZLfbxYM2UqmUIjedVVdXi4ElHA6XbPpCKeUesJFIJBTZqqJkgiAgk8nkXVHQ6XRLbpijQFxaFIgJKQ0Kw0Qycj55LrdFYmZmRsKVrF7uxrlCqsJy6Rs2Go0lOXa5EFQdRt4bEK1We9ODZygQlxZViQkpPgrDRDJyDsOTk5O4dOkSZmdnMTU1JfVyCsayLPx+P9LptDiXV0kYhqFjl2WAYRjx1D+WZVd8AiMF4tKjQExI8dAGOiKZ3H5WuYXhSCSCvr4+9PX1AQA6OjokXlFheJ7H1NQUpqenodfrpV5OwUwmU97Jf1K2qYyPj6O5uVmy7y+17Ga57IlzK0Wb6kqPTq4jpDioMkwkI+fK8I0GBgakXsKqCIKguF5hhmHEfm0AND2izJaqwBcShLOoQlweVCUmZG0oDBPJ5IZhOY1UI9L2DZvNZjF4ZTIZWYT5Sukdzh61fOOmubWgQFwe1EtMyOpRGCaSMBqN4uX7dDotq7Fq5Tzcodiyl7SVimGYvH9/pZ76p1TZo5YzmQzS6TT1aSsQBWJCCkdhmEgit19YTlVhg8GAI0eO4NChQ9iyZYvUyymYw+FAV1cX1q1bB4vFIvVyCiYIAvx+P2KxGFKpVN6JZ1JTe3X4xqOWGeb/s3fnUY6VZf7Avzd7KksllarUvvZSXd0FvbKj7I0DbetwkDniqEdEUUH0DAMiDv7kzNDKyAwzggKKOjNnPDCiCNLA0M2mKA003fTetXal9jWVpLJXlvv7oyavSW2dqtybe2/yfM7pc5Lq1L1vQ1Xy3Oc+7/NwqyqNWAxlh/OLssSErAwFw0QSarUagUAAPM/Lql44NWijtLQ0ozWZUjgcDnAcl9EnWWl4nkcoFJLVRVKhy8eoZQqI84+CYkKyQ90kiCRGR0cxOjoKtVqddbumfFByf2Gj0cjakSWTSXg8npyONzIygpqaGiGWVjAKsbPE2UYtC4k6TEiDRjoTsjzKDBNJJRIJWU2eSx/BPD4+LuFKVs7hcLDHPp8v45Y3EU4hlUtkO2pZSJQhlgZliQlZGgXDhPyf0tJSGAwGAEA0GoXX65V4RdlTq9Ww2WzsudvtlnA1K2e322E2m5cc80uEl+ocke2oZSFRQCwdCooJWYg+eQj5P+lZYaWVSNhsNhbEhMNhRfXmNRgMUKvVMBgMKC0tlXo5WSmE7HAikVjxqGUhUUAsLQqICfkLCoZJ3lksFqxZswZOp5NlYuVAycFweomEkFlhsfsNLzZ2mYiP53nE43H2XK1WC75hLhsUEEuLssSEzKFgmOSd0+nE5s2bcemll2Ljxo1SLwfA3O758vJy9nyxYFiuU+hMJhO7qEgkEooq75g/dllO/abPRsnZYY7jWJ9vjuMkCYRTKCCWHgXFpNhRMEzyTo5jmMvLy9kt4pmZGUVlKNOzwh6PJ+PWt5ypVCoauywhlUoFvV4PnU4n6oa5bFBALA8UFJNiRcEwyTs5BsNKbamm0Wgy6mzF2DgnVqmEHMcur5SSs8OAsIM1ckUBsXxQQEyKDQXDJO/Sg2G5DFaIRqNs9K+SWqrF43GcOXMGXq8Xfr9fMQGlWq0umLHLSgmIU6OW5WxwcJCCYpmgLDEpJjR0g+SVTqdj9a3xeFw2t8a7urrQ1dUFk8mkqBIJYC6QVFowaTKZWEZSbmOXC1E8Hmcb5rRarawG3SyGhnPIRyogpqEdpJBRZpjkVXpWOBAISLiSxQWDQdlnz5ROq9WyzVs8zysukF+MnLPDyWQyo3OEUoaxUIZYXihTTAoZBcMkr6xWK3sslxIJcnZC1g2nAmFgrjxFKcGZEiWTyUVHLSsFBcTyQ0ExKUQUDJO8kuPmOSXS6XQoKyuTzeanlQgGg/D5fIjFYrIpkxGC3LLDqQlz6eTQOWKlKCCWJwqKSSGhYJjkldwyw1arFRdffDHWrFkDs9ks9XKyVlZWhrq6OmzcuBF2u13q5axYLBaDz+cruJIUuQTEi41aVmIgnEIBsXxRUEwKQUEHw2NjY/jGN76BNWvWwGAwoLKyEh//+Mfx+uuv53zs4eFh3HvvvTjnnHNgtVphNpuxbt063HzzzXjhhRcEWH1hklswXFVVhaqqKmzevBltbW1nfb0cBm9wHIeysjIAc10Z0utBxST2NDoinPmdIzQaDRtuolQUEMsbBcX5RfGNsAq2m8SxY8dw5ZVXsr6rVqsVU1NT2Lt3L1566SXs2bMH995776qO/eyzz+KLX/wiu82f6pfa09ODnp4eTExM4BOf+IRg/5ZC0tvby3655LBxKn0Es1JaqpWWlrKJYbOzs1RuIjNDQ0Ooq6uT7PyJRCKjDluqUctioC4T8kfdJ8RH8Y3wlJ0qWEI4HMbu3bvhdruxdetWnDhxAj6fDx6PB3fddRd4nsd9992Hffv2rfjYL7/8Mj796U/D7/fjlltuQUdHB4LBIAKBAKampvDb3/4W1113nQj/qsLQ2dmJgwcP4s0335R6KVCr1RnT25QybCN9zWIM2RBDSUkJ7HZ7xuY5IrxkMplRJ6xSqQomEE6hXsTKQJlicVB8I47Cepf8P08++ST6+/thNpvx4osvora2FsDc1dPDDz+M3t5ePP/88/j2t7+NnTt3Zn3cmZkZ3HrrrUgkErjvvvvw4IMPZvy9w+HADTfcIOi/hYinoqKC9Vv1+XyIRCISr+jsDAYDTCYTgLnAx+PxSLyis0uNXeY4DlarlW2eK2RSZYc5joNGo0E8HmedI5RaJ3w2lCVWBsoUC4viG3EUZGb4V7/6FQDg5ptvZj8o6e6++24AwOHDh9HZ2Zn1cX/5y19idHQUdXV1+N73vifIWol0lFgikZ4V9vl8easXTllN3XD62OVYLFbwgXCKFJvpUsGwVqst6EA4hTLEykGZYmFQfCOOgguG/X4/Dh06BAC49tprF33NhRdeiNLSUgBYUbF56ofwxhtvVFSvTrK49GBYCSUSKpUKNpuNPVdCiYRarWYTBwFlj11WErVarfgNc9migFhZKChePYpvxFNwZRKnT59m7YQ2bdq06GtUKhVaW1vx/vvv49SpU1kdNxKJ4MiRIwCArVu3oqOjAw888ABef/11zMzMoKamBtdeey2+9a1voampKev1bt++fcl/R6G55JJL2Iav7u5uSYctGI1G1tkikUhgampKsrVky263s7KOSCSiiB69qZIOYG7ARr4z2VLLR7lEIpGASqUq+CzwcqhkQnnSA+JiKaE4ffr0kp/5qSD3bN+vpPhGSQoudTA6Osoe19TULPm61N+lv345LpeL3d7t6urCtm3b8MwzzyAYDEKr1aKvrw9PPPEENm/ejLfeemv1/4ACpdFoUFlZifr6emzYsEHy/rLpWeGpqSnJ15ONVDs1QNqscLalEvPHLisheBeDmOUSiUSClZ6k9xQuRrSxTrkoW5wdim/EU3CZ4fTbsEajccnXlZSUAAACgUBWx/V6vezx97//fVRVVeH555/HNddcA47j8M477+ALX/gCurq68KlPfQqdnZ0ZwctSlroa3L59O7tSKwTpk+cCgYDkH9y5lEj09PRg7dq1Qi9pWakaUEA5G+fSs8KRSITGLgssvXNE6jF166AssZKljw7P5znzEYjPzs6ira0tqwzwUpQW3yhJwWWGxZKeOUwmk/iv//ov7Ny5k92avPjii/Gb3/wGKpUKU1NTeOqpp6RaqizJbQyz0lqqxWIxnD59GoODgxgfH5d9Jluv17OWXsWcFU4ROju82KjlYqzzWwpliAnJHsU3BRgMp2ejwuHwkq9LfThnO4I3/XWbNm3CVVddteA155xzDq6++moAKytcLwbpwbAcJs/t27cPf/zjH3H69Gn4fD6pl5MVnufh8XgwOTkp9VLOWiqRykwAc7+HUt8JkAOhAuJCG7UsFiqbIIWG4hvxFFwwnF5Hs9wHdurvqqurV3zc1tbWJV+X+jt6E86UPoZZDpnhZDKJqampgtyoKDWNRsMCs2QyWfRZYaHNH7Ws1WqLpnPEatB7MSkUFN+Ip+DeQTds2MA+iE+ePLnoa5LJJOu/t3HjxqyOW15ejsrKyqzXQVmaTOnBsBwyw0Q88XgcHo8H4XCYWqnNk2t2eLFRy6kOI2RphfjhTYoPxTfiKbhg2GKxYMeOHQCA/fv3L/qa9957j90aX+x2wFJStwiWa2Td0dEBAAXbfmQ11Go1u22eTCazLuonc7evGhoaMm6PycVymQme5xEMBhGNRvO4ImVYbUBcDKOWxURlE0TpKL4RT8EFw8DcZBZgron0Yq1FHn74YQBzHRuWuyUw3+c+9zkAc1dkr7322oK/P378OKulKdT53athsVjYlWQwGJR085fVakVNTY1iNhuVl5fDZrNhzZo1GR0wiLKtNCDmeT5jp32hj1oWEwXERMkovhFHQQbDt912GxobG+H3+7Fr1y7WeNrv9+Oee+7Bc889BwDYs2fPgu/lOA4cxy06jnDnzp245pprAACf//znsX//fraJ5cCBA7jxxhuRTCbR3NyML3zhCyL965RHTiUSTU1NuPDCC3H99dejpaVF0rWcjVarzdh4mN7+Ro4oMBNX+n9fCoRzQ1liolQU34ijIO+xGY1GvPDCC7jqqqtw+PBhbNq0CVarFYFAAMlkEhzHYc+ePdi5c+eKj/3000/jiiuuwPHjx7Fz506UlJRArVazTWE1NTX4/e9/n7GbvtjJKRhO1UWpVKqcyjXy0Wu4rKyMBTx+v1+SHpxnMzIygpqaGmg0GpSWlrLJeNQ94uxWMp2O4zjodDrEYrGiGrUsNupJTJSG4htxFOw76ubNm3HixAnceeedaGlpQTQahcPhwPXXX4/9+/fj3nvvXdVxHQ4HDh48iIceeghbt26FSqVCPB7Hpk2bcN999+HYsWNob28X+F+jPC6Xiz3u7u7Gn/70Jxw9ejTriThiKCkpYZlWuY9g5jhONhPnsmEymcBxHIxGoyzrm+VqJeUSqYCYNswJizLERGkovhEex1MKR5ZSE+iy7RMoR3Irsm9qasK2bdsAAOPj4/jzn/+c0/HEzAyXlpaisbERwNzkotTGBTlqampi2f9UL2S5DwWRm8UyxDzPUylEnlGWWFrf+c53UFZWltOUtpXYvn07pqen8eCDD4p+rnz/28jKFGxmmJD50lvHjI+PS7iSs0ufkCf30cvpI4AjkQgFwgJIJpOIRqOIx+NSL6WoUJaYkOJEwTApChzHoaKigj2XczCs1+vZHQGe52VdIlFWVgaDwQCABmzkIr1cIn3Ucjwep4A4z2hzHSHFh4JhIiqtViuLGkebzcYymKFQSBZT8JaSnhX2+XyyDYZUKlVGtp3GLudmaGho0VHLtFlOGhQUE1I8CrKbBJGPtWvXYsOGDQiFQujo6EB/f78k60gP2iYmJiRZQzY4joPdbmfPp6enJVzN8srLy1m/5tnZWYTDYYlXpHxerxdGo5E9p1HL0qOOE4QUPnqXJaJKDdwwmUySZg3TB1bIORjWaDQIBoPgeR7RaFS20/o0Go1iyk6UQq/XZwTCGo1GFndVCGWJCSl0FAwTUaX3GJaqNEGj0WS0KZucnBTkuD09PYIcJ10sFoPL5UJHR4esP3wrKytZoBaJRODxeJYdz0yWp9FoMjrHqFQqCoRlSM6/k4SQ1aMyCSIajuMyPuClCoZ5nsfhw4fhdDqh1+sRjUYlWcdKxGIxtolKjlKb5gBI2ju6EKhUKlitVtZGLR6Pw+v1Zj2Qg+RXKiCm0glCCgcFw0Q0ZrOZ1TuGQiHJNoIlEgkMDAxgYGBAkvMXot7eXpSWlsJsNst6M6ISWK1W9nuSTCbZlMaVTKgj+UdBMSGFg8okiGjkNIZZ7lIz45XE5/NheHg442tUKrEy6d1WeJ7HzMxMRp/mlUyoI9Kg0glClI+CYSIaCoaz53A40NbWhqqqKtahgRS+WCwGn8+HZDKJQCCw6N0TCojljzbYEaJsFAwT0cghGFZKYOlwOKDRaOB0OmU7gjvb/5aUHV6ZeDwOj8ezbC07BcTKQEExIcpENcNENBaLhT2Wqq50+/btcDgcmJiYQEdHhyzrW81mM/R6PYC5+mav1yvxihbiOA4tLS3geR6jo6Oy/O+oZNm0HaQaYuWgemJClIUyw0QUKpUqI8MpRWY4NYJZr9eL9qEkRHu19IlzHo9HllPcysrKoNfrYTAYUF9fT4MgVonjuIwNcytFGWJloSwxIcpAn2hEFCUlJey2bygUQiKRyPsa7HY7u7Uv1xHMGo0mo5zE7XZLuJrFzR+7PDExkbHJazFUKrE4q9UKnU4Hm80GjYZuzBUDKp0gRP7o3ZiIIhAI4JVXXoFGo8noSZtPShjB7HA4WBeJQCAgyx7IFRUVLHCbnZ2VZcCuBGazmV2c5dI9hMollIlKJwiRL8oME1HF43HJRgqnj2CW67jg9Ml4cgwy549dHhsby7qMg7LDf2EwGDIuCkOhUE5DVahcQrkoU0yI/FAwTAqSVqtlgSbP84KNYBaS1WplmcJYLCbL9nNVVVWsvjUcDstyc5/cabVamEwm9jwSiSAcDud8XAqIlY2CYkLkg4JhUpCcTie7De3xeDA7OyvxihZK3zg3PT0tu41zer0edrudPV/N2OVizw6rVCpYLBb2sxiLxQS9U0IBsfJRUEyI9CgYJqI499xzUV5eDp1OJ8n581kisZqOEjqdjrWe43ke09PTQi8rZ9XV1SyIm5mZkazcRanmd45IJBKibOKkgLgwUFBMiHRE20Dn8Xiwb98+vP322+jq6sL09DT8fj+sVivKysrQ2tqKj3zkI7jmmmtgs9nEWgaRQGlpKa6++moAc7eEX3755byvIT0YluPmuXg8juHhYTgcDkSj0ZzqR8VgMplYlwue5zE2NibxipTHYrGwjYc8z8Pv95+1C8dq0aa6wkEb7QjJP8GD4UAggAceeABPPvkkgsHgkq977bXX8OMf/xhmsxlf+cpXcP/998t28hZZmfTb/1K0MzOZTKxGMxaLyTLrmkwm4Xa74Xa7ZdmzV6/XI5lMQqVSwePxIBKJrPpYIyMjqKmpEXB18qfX6zPuiiw1allIFBAXFgqKCckfQT+FXS4XLrroIvzrv/4rAoEAeJ4/6x+/34+HH34YF198Mfr7+4VcDpFIejAsxaYwrVaLqakpJJNJTE1Nya4Wdz6xsoW5mJ6eRmdnJ6anp2XbiUPOotEo2yQXCoXy1jKPSiYKD5VPECI+wTLDkUgEu3btwunTp8HzPEpKSnDzzTdj586d2LZtGxwOB8xmM/x+P6ampnD48GHs27cPTz/9NMLhME6cOIGPf/zjOHjwIBtNS5RJ6syw1+vFH//4R2g0GslqlgtBLBYTLLgqxuxwMBjE7Oxs3ktgKENcmChTTIh4BMsMP/DAAzh16hQA4GMf+xj6+vrw05/+FDfeeCNaWlpQWloKtVoNm82GtWvX4qabbsJTTz2Fvr4+XHvttQCAkydP4oEHHhBqSUQi6cGwz+eTbB3xeByhUEhyEBCdAAAgAElEQVSy8y9Gp9OhpKRE6mWQPJGqFpwyxIUrlSmmbDEhwhEkGI5EInjyySfBcRyuuOIKvPTSSxmN+pfjdDrx0ksv4fLLLwfP83jyySdlOYWLZIfjOMkzw1LItqNERUUF1q5di3Xr1mX0npUDlUrF+h6LodDbrMntjhYFxIWPgmJChCFIMPz888/D6/VCrVbj5z//+YrHjKpUKvz85z+HWq2G1+vF888/L8SyiARKS0vZDvpIJCLL/r5SUalUrG+v0WiUXS2z0+lEa2srqquroVarpV6OopjNZlgsFtYuTy6GhoYoKC4CFBQTkhtBguEDBw4AAK677jo0Njau6hjNzc247rrrAADvvPOOEMsiEkgf0iDF5rnt27ejvb09Y+iGXNjt9oxpbnIq4dBqtSgvL4dKpUJFRYVonV0KMTtsNBrZqGW9Xi+7DDFAWeJiQUExIasjyAa6Q4cOgeM4XHbZZTkd5/LLL8eLL76IQ4cOCbEsIoH0Eol8D2nQarVoaGgAx3FYt24d9u7dK6v+ven/bdxut4QrWSh97HIoFJK01ltJtFptRg14JBKRbZkXbawrHukBMW24I3LQ0tIi6PE4jkNvb69gxxMkGB4eHgYAtLe353ScTZs2AaAshpKVlZWxx/nODKdng6enp2UVCJtMJpY9TCQS8Hq9Eq/oLwwGQ8bgm9WMXV6JQuksoVarRR21LAYKiIsPdaEgcuByucBxnGDlgULf+RUkGE5lkXKdJJe6xS7F7XUijNHRUej1epSVlUkSDKfIbepcelbY4/HIqrfw/LHLyw3LIXMWG7WslPctCoiLE2WLiZRSd23lSpBgOPUhkOvu+NTtxmLpQFCIjh8/juPHj7PnTU1NeTu31MFwT08P1q5du+DrGo0GpaWl7LmcJuKlNn4BcyODxc4Kpyg9O2yxWNgmQ57nMTMzI7sNkcuhgLi4UbaY5JvL5ZJ6CcsSZAOd0FkuOWXNiDKYzWbZjmC22+3sijgYDOY02lho1dXV7PH09LRs613lxGQyZQxz8fv9SCQSEq5odagcjVDPYkLmCDqOmRCpVFZWsseTk5OyytLJdeOczWaD0WgEMHcBmu+xy0rsLKHX69l/M+AvU+aUilqvkRQKikkxE2wcMwD85Cc/ybhVvVJyq/MkypEeDOc7qFtOSUkJyyLG43HZdGngOA5VVVXs+eTkJOLxuIQrUoZ4PI5EIgG1Wo1oNIpwOCz1kgRBZRMkhWqLSTESNBh+/PHHhTwcUZjzzjsPZWVlcLvd6O7uzlvgp1KpUF5ezp7LKRgOhULo6OiAw+FAPB6XTcZaq9WyW/uxWAyTk5OSrENptcOpTiAlJSUFt9GQAmIyHwXGRCpPP/00nn32WXR3d4PjOLS0tOCGG27AZz/7WVE24gkWDMvlQ55Ip6WlBbW1tQDmMo35CobLysrY1LtAICDpMIvFNtHNzs7mbWNatmZnZ9Hd3Q273Q6e56lOfwV4ni+4QDiFAmKyFNp0R3IxMzODr3zlKwCAG264ATfeeOOirwuHw/jkJz+J1157LePrJ0+exIsvvognnngCr7zySsamdCEIEgy/+eabQhyGKJxUtbHp56VSm5XxeDxSL0H22WGVSlVUFwupGmIKisliKFtMVuO1117DM888A47jcP/99y/5uttvvx379+9fsifxe++9h8997nN44YUXBF2fIMFwrpPniPKZzWY2VCISieR1+EBnZyeGh4fhdDplNcyCKJ/RaERJSQn8fr+iN8qtBmWJydlQYEyy9ac//QkAsGXLFrS1tS36muPHj+M///M/wXEcVCoV7r77bnzpS19CTU0NDh48iDvvvBNHjhzB3r178c477+Diiy8WbH3UTYIIQuqOCYFAAGfOnJFNSzWLxQKn08nKN+QifXSwnMixs4ROp4PJZGIDNrRardRLyjvqNEGyRW3ayHIOHz4MjuNw/fXXL/maX/ziFywbfPfdd2PPnj1obm6GXq/HpZdeiv3796OiogIA8Mwzzwi6PgqGiSDSxzDLJSCVktPpRFVVFdra2gSvbVoto9GItWvXYu3atTkPyBGDnALi1KjllNnZWVmN984nar9GVooCYzLfmTNnAAA7duxY8jX79u0DMDeo6u///u8X/L3D4cCtt94Knufx/vvvC7o+0dNWwWAQJ06cwNTUFGZmZmC1WlFeXo729nZZfiCT1UkPhuXUS1cKBoOB/WzLabNVasBGSUkJHA6HbNYlN6lMcGrHciKRoKmYoLIJsjpUSkGAv8QFS+0P8Xq9OH36NDiOwwUXXJARU6T7yEc+gu9///ssuBaKKMFwJBLBf/zHf+AXv/gFjhw5suh0JrVaja1bt+KWW27B5z//eVZvSpQp/Qc3n5uympubMTk5mdca5bNJL42YmZmRRf9ei8UCs9kMYC5AHxsbk3hFi5PDZjqr1cpGLSeTScWNWhYTba4juVguMKauNoUtdWdtqdLBDz74gD2+4IILljxOqj/+zMyMgKsToUzijTfewNq1a3H77bfj0KFDrLfq/D/xeBwffPABvva1r2HdunV44403hF4KySMpaoatViu2bt2KnTt34oorrsjLOc9Go9FkvMnLJUuePnbZ7XYX3WawbJlMJlYbzPO8Ykcti43KJkiu0kspQqEQ+vv7i7YUqRhYrVYAS5fDvffee+zx9u3blzxO6mdE6F7DgmaGf/rTn+JrX/saC3gBYN26ddi2bRuqqqpgNpsRCAQwNjaGw4cPo7u7GwAwPDyMnTt34oknnsCtt94q5JJIHhgMBrYxKxaLCX7FtpT0aYdyueVfX1/PgqlIJCKLdZWVlbE7L4lEQlZDSRYjVXbYYDBkjFoOhUL04bwMKpsgueJ5HrFYDMPDw5KtIRaL5aW2udjfS9atW4f3338ff/jDH3Ddddct+Pv9+/ezx5dccsmSx0l9fi1VRrFaggXD//u//4uvfe1rSCaT4DgOX/rSl/DNb35zyRYaANDR0YF/+7d/w1NPPYVkMomvfvWrqK+vx7XXXivUskgeSFUikR4My6W/cHNzM3ssh6wwx3EZo6onJycVkenMd0Cs1Woz9jBEIpGCGbUsJiqbILlIJpNUGlEkLr/8crz33nv46U9/iq9//esZ7xlHjhzB22+/DY7jsHHjxmVry1PlFGvWrBF0fYKUScRiMdxxxx1IJpOwWq14/fXX8eSTTy4bCAPAhg0b8MQTT+CNN96A1WpFIpHAHXfcUfRXUErjdrvxu9/9Dn/84x9x9OjRvJxz/ghmOQTDZWVlsNlsAObe5OUw0KKiooJlqqUcuyx3arWa3XaLxWKyqkFXAiqbIKuhUqnY7136Y1J4br31Vmi1WszMzOCSSy7B448/jn379uHf//3fsXPnTlZNcLbqgH379oHjOGzdulXQ9QkSDD/zzDM4c+YMOI7Ds88+i8svv3xF3//Rj34Uv/nNbwDMtd8Qun8cEVc0GkVfXx8++OADHD9+PC/nlNMI5pSWlhb2uL+/X/KMh0ajYT0ZgbnbS0raCJbPVmuRSAQzMzPUOSIH1IKNrBTHcdBqtdBqtdDpdBQMF7A1a9bgu9/9Lniex9DQEO644w781V/9Ff7u7/6O3UVtbW1lI5sX093djXfffReA8MPeBAmG9+7dCwDYtWsXrrnmmlUd4+qrr8auXbvA8zxefPFFIZZFClj6rX85ZIV1Oh1qa2vZ876+PglXM8fpdLKuCJFIhPo/n8Xs7Cw8Ho/kFzFKRwExWQmVSsXep0hh+853voMf/OAH0Ov1C5oqbNmyBS+99BL0ev2S3//DH/4QwNzn7c6dOwVdmyA1w6nJIp/61KdyOs5NN92EvXv34sMPPxRiWaSApQfDctgQZjKZEA6HYTabMT09Da/Xm1HGIYXUmwzHcRgdHZV0Laslh1ZrZOWolrhwVVdXQ6vVIpFIYHJykjrTkBW55557cOutt+LVV1/FwMAA9Ho9tm3bho9+9KNn/d7zzjsPmzdvhtPpZK1ChSJIMJwKRtauXZvTcVLfL4fghmQnNUM8n5uy9Hp9Rm2uHOpgPR4P9u3bB6fTKZtShNHRUbjdbthsNkXf+hcjINbpdDAajdRDWGQUFBcOh8OB9vZ2lJSUgOd5qFQqBAIBjIyMoKenR+rlEQUpKyvDpz/96RV/35e+9CURVjNHkGA4dVsx11sd6Y3uiTLU1NTgU5/6FLxeL3p7e/H222+Lfs70LhJut1sWQy1S0ks2enp6cr5AzNXs7KwsykjkRKPRwGKxgOM42Gw2+Hw+es8RGbVhU7by8nJs2bIF4XAYp0+fht/vR3l5ORoaGrB+/XoAQG9vL11YkrNKtV7VarUZrSzPJhwOs+YKqZ7FQhKkZjgVnORaJ5n6/vRgh8ibw+GASqVCWVmZ4LctliLHlmpEXEJtpuM4jgXCADJ6ohNx0QY75aqvr4darUZnZydcLhfcbjc6Oztx+PBhxONxrFu3DjU1NVCpBJ/jRQrII488ArvdDrvdjldeeWVF3/vKK6/AZrPBbrfj8ccfF3xtgvzknnPOOQCA559/PqfjpL6/vb095zWR/EifPDc1NZWXc7pcLnR1dcHr9VIwPI/RaFxy3KXSCREQ06hl6VFQrCwajQbl5eXweDwL3uPdbje6urqQTCbR3NwsSsaOFIZIJIIHH3wQAHDbbbfhhhtuWNH333DDDfjqV78Knufxve99T/AWvIIEw9dffz14nsezzz6LQ4cOreoYhw4dwq9//WtwHIfrr79eiGWRPEgfuJGvbgVutxsnTpzAG2+8IXkv37a2Nnz0ox9FfX29LLIiDQ0N2LBhAyorK2WxHjkxm80LRi1TeYR0KCBWBp1Ox1qfLfaeMjQ0hLGxMVitVlRVVVFnCLKo559/HtPT0ygpKcEDDzywqmM88MADMJvNmJqayjn5Op8gn5af+cxnUFlZiWQyid27d6Ojo2NF39/Z2Yndu3cjmUyioqICf/u3fyvEskgepAfDcpi4lk8cx6GpqQnl5eU477zzUFVVJel6HA4H9Ho9G0hSiD07V5sdNhqNbCQ1MDe+m4b7SI+yxPLR0tKS0Ss9JRwOw+v1QqfTobS0dMHfx+NxDA0NIRKJoLa2dtnWWKR4vfzyywCA3bt3Z/S/X4ny8nLs3r0bPM/jpZdeEnJ5wgTDJpMJDz30EHiex9jYGHbs2IEf/OAH8Pl8y36fz+fDQw89hB07dmB0dBQcx+Ghhx7KGItK5Eun08FisQAAEokEK4xP53K58ryq/KmpqWEbAMLh8KLty/K1y1qlUi3ovayEscursdKAWKvVoqSkhD2PRCKIRCJCL4vkgIJi6VitVlxyySXYsGEDbDbbguyvSqWCz+eDwWCAzWZb9CLb7XZjamoKer1e8qQAkacPPvgAHMfhYx/7WE7HSX1/aiyzUAQrLvzc5z6Hrq4u7NmzB+FwGN/5znfwj//4j/jIRz6C7du3o6qqCmazGYFAAGNjYzh06BDefvttRCIRVrP3rW99C5///OeFWhIRmd1uZ4/zMayA4zio1WrZdI9Iz6K4XC5Ja08rKipYrfDs7Gze6relkm27NbVanbFhjkYtyxu1YssfjuPQ3t6O+vp6+P1+nDhxApOTkwvexxOJBDweD+rq6lBbW4uJiQkEg8EFxxscHERdXR3KysrgcrmoBIlkGBsbAwA0NTXldJzGxkYAwk8oFXSnzT/90z+hqakJ3/zmNxEKhRAOh7F//37s379/0denggej0YhHHnkEX/7yl4VcDhFZvuuFKyoqcPHFF2N6ehoDAwOSZp0tFgu71ZNMJiWdODd/7PLY2BhtCvs/JSUlLNO11N0LIj8UFIururqabVR3uVwYHh5e9k7u2NgYqqurUVlZifr6evT09CxISkQiEcRiMSSTSQqEyQKpC6jU3eTVSnWtWuyCLBeC77C59dZb0dnZibvuugvl5eULRu6l/3E4HLjrrrvQ2dlJgbACpXeSyEcwnNoUVl5eLvmu5ebmZvZ4bGxM0tvu6ZvlUvV9xSCbzIDf70c0GgXP89Q5QoGofEJ4drsdW7ZsgVarxdGjR9HR0bEgEJ5fKpFMJuFyuRAOh9HQ0IDq6uoFr9Hr9dBqtRQIk0Wl7iTnetcytTcp/c60EETpwVRbW4sf/vCH+OEPf4hTp07h6NGjmJqagt/vh8ViQXl5OTZv3oyNGzeKcXqSJ+nBcD42z6X3F5ZySqFarUZDQwN7fubMmWVfL+bwDb1en5GhV+rY5dXKplzC7/dDrVYXbA11MaBMsXA8Hg8GBwfR0NAAi8WCiYkJNojGbrejoqICPM/D5/NhfHycXVy73W64XC6sWbMG69evh1arZe99FosFa9euxezsLPr7+6X85xGZqqqqwuTkJA4fPoyrr7561cf58MMPAczd3RCS6A1JN27cSEFvgUrflCR2MGwwGNhO5mQyKWlNbF1dHXQ6HQAgEAhI2uu4urqa1cP6/f6irIfNJiCmQLgwUFAsjL6+PjgcDrS2tmJychKlpaVYv349dDodYrEY6y3c3NyMw4cPs/e4wcFBxONxbNy4ERs2bEBFRQXi8TiMRiOsVit6e3vPunGeFKdLL70Ux44dw29/+1vcc889qz7Ob37zG3Ach0suuUTA1QkcDEejUfz85z/HK6+8gv7+fiQSCdTU1ODKK6/El7/85YxMIlG+p59+GkajEQ6HQ/R+v+k1sW63W9LgJn3j3NmywmIymUysXITn+aLLCi9Go9FAo9FQt4gCR0FxboLBIFwuFzZt2oQLLrgAGo0GExMT6O7uRjweh06nQ1VVFVpaWrBp0yZ4vV7Mzs4iHo9jcHCQtVFLBcORSATvv/9+wW/cJau3a9cu/OQnP8EHH3yAX//617jppptWfIxf//rXOHjwIDiOw65duwRdn2DBcFdXF6677roFG4k6Ojrwxhtv4J//+Z/x3HPP4YorrhDqlEQGwuFwXmr65DKCOTVKEpjLNg4MDEi2FmDuAlSv18Pj8RR1ADgyMoK6ujpYrVaoVCqo1WrBN1gQ+aGgePUGBwdRWVkJs9mMzs7OjPeyUCgEr9cLq9WK8vJyNDQ0ZLSJnJycxOTkJNRqNTQaDaLRqBT/BKIgH/vYx7BlyxYcOXIEX/ziF1FZWYnLLrss6+9/6623cMstt4DjOGzdujXnFm3zCbKBLhqN4hOf+ATOnDmz5GY5n8+HG264gTZDkFWZ30NXKrOzszhz5gxrND87O5vV94nRbzgYDKKzsxNDQ0OS1lDLAcdxMBqNbFNPavgIKQ600W7lkskkent70dvbu+C/Xar0KvW+5XA4Fv19SiQSFAiTrD322GPQ6XQIhUK4+uqr8Y1vfOOsd1d7e3tx55134pprrkEoFIJOp8Ojjz4q+NoEyQz/93//Nzo7O8FxHM477zw8+OCDuPDCC6HVanHs2DE8+OCDeOGFFzAzM4N/+Zd/wSOPPCLEaUmRKC0tZdPDotGopCOYg8Egjhw5ghMnTrC+vlLL1xhsOauvr2cDUFKdI2hXe/Ep1kyxRqNBZWUl4vE44vF41ns43G43pqenF3RZST1PvcdFo1H6fSI5u/jii/Gzn/0Mt9xyC5LJJB577DH8+Mc/xvr167F9+3ZUVlbCZDIhGAxifHwchw4dQldXF0uqqtVq/OxnP8NFF10k+NoE+TRPzYhua2vDH/7wh4xxjDt27MDvfvc77N69G3v37sXvfvc7CoYLQFNTE6LRKNxud9bZ0dWSS1Y4XepDh0jP6XTCZrOx58PDwxmjl0nxKaaguLm5GWvXroVarWbZ28nJSXR1dWW1mW25doOpkjDqz02E8tnPfhb19fX4zGc+g9HRUfA8j87OTnR2di76+tTPZ3V1NX71q1/h8ssvF2VdgtxHPHr0KDiOwze/+c0l55Lfd999AObqlGi3qfJdddVV+PSnP4077rhD8H5/88mlpVquhCiVUKlUkvdYlhOr1Zox/nVqaooy5YQp9PKJ1tZWtLa2wu124+TJkzh+/DjGx8dRUVGBbdu2ZVwkroRGo0FjYyOam5vhdrsxPDws8MpJMbv88svR29uLRx99FNu3b4dKpVq0vFalUmH79u147LHH0NvbK1ogDAiUGU7tIN2yZcuSr0n/O7fbzdpkEeXRaDQsIEsmk6JnDSKRCNsoJlVmuKSkBLFYDLFYTJLzpzidTjidTgQCAYyOjiIcDku6HikZDAbU19ez54FAgA3iyHZcMykO6QFxoWSLDQYD6urq4PV6cerUKbaBdnBwEOeeey7q6urQ2tq66FCNpXAch4qKClRWVqKmpgahUAhdXV1UF0wEZzAYcPvtt+P222+H3+/H8ePH4Xa72TwKh8OBc845J+eJddkSJBiORCLgOC6j7+x86bcti3nXeyGw2+1sg4XP5xO9zdkHH3wAYC4LKNXPTnt7O6qrqzE0NISOjg5JOhVotVqUl5cDmBtJqdPpijYYVqvVaGpqglqtBjBX0zi/2T8FxGQxhVJCYbPZoNfr0d3dzd4XVSoVkskkuru7oVarUV1djbq6OoTD4WXL2VJ9hTdt2gRg7vdraGgIp06doqmNRHQWiwUXX3yxpGug7dZkxdInnuXzlrRUdWsGgwE1NTVQq9VobGyUrEtB+tjlUChU1OVGDQ0NbPBJIpGAy+Va9KIsm5HNpDgpvYQilXxKXRACYJvcwuEwBgYG4Pf7UVVVlfGevZjUHgiPx4PR0VG88847OHnyJAXCpGhQMExWLH14SjHUZzY1NbEgNDVWfLVWWzdsMBgyarOLfcDG+Pg44vE4eJ7HwMDAsrdxKSAmy0kFxUoLjFPJgbKyskU720xPT2NsbAw6nQ4VFRULXmMymTL2+Ljdbhw7dgynTp0qykmWpLgJ2hvqJz/5ScZmp1xe993vfleoZRGBpWcZxB7DLDWO49DU1MSeSzVxLn3s8szMTNEPlAiFQuju7obZbM7q4oRKJkg2lFRCkRq/bjabYbVaFyQmeJ7HxMQEampqUFlZiY6ODvZ3NpsNF198MUKhEN566y32euqQQ4qVoMHw448/vuzfpz7Mz/Y6gIJhOctXmURNTQ1qamowMTGB8fFxSTZxVFdXs9uRkUhEkiyj2Wxmmwho7PJfxGKxFfWcpoCYZEsJG+5SrS0bGxvhdDrh9/sXbPD1+XyYmZlBdXU1nE4n6wqRKrFSq9XQarWSbwwmRGqClUksNXluNX+IfKlUqozb9WJmhmtra9HQ0IAdO3agsbFRtPMsp7m5mT12uVySNJ6vrq5mj6enp4tyZ7dWq4VWq835OFQyQVZKzmUU/f39CIfDqK2tzShfA/6SfEq9R6feu1JtrN588028/vrrFAgTAoEyw2+++aYQhyEKYLPZ2IaNmZkZUd9Ipe4vbDKZ2MAPnufR19cnyHF7enqwdu3arF5rs9nYZLVkMqnoPsurpVKp0NzcDI1Gg/7+/pxLRChDTFZLbmUUgUAAAwMDWL9+PRobG9mETo7jWGLJbDYDANtgmr7JjhAyR5Bg+LLLLhPiMEQB0rMP2WaFXS5XRt1tNux2O9vcEYlEJOmckJ4VHhsby/uHB8dxGQMlJicni7Kmr76+nrVmbGpqQkdHR87t/CggJrmQUxmFy+WC2WxGbW0tNm7ciCNHjrALxlTP4OnpadlM7yREjgStGSaFLxKJoLe3Fw6HQ9QSiYqKCvZYimyoSqXKKM0QKiu80jUEAgHY7XYkEglMTk7mfQ1Sq6yszBjQMzIyIlhfawqIiRCkDowTiQQ6OjqgUqlQXV2N888/H16vF8DcnSWNRgOXy5X3dRGiJBQMkxUZHBzE4OCg6OdJL5GQIgisqalhmelgMIixsTFBj59NqUQikcDQ0BAmJyeh0+kkqVeWUmlpKStTAeZ+DlayYS4bFBATIUkVGEejURw9ehSBQAA1NTWoqKhAPB5fMJ2OELI4CoaJ7KjV6oxyDClu7w0PD+Pdd99Fc3Oz5LcXo9Fo0W2aMxqNGaOWZ2ZmROuiQQExEUO+A+PU5LnUXSydTodQKCT6eQkpBBQME9kpLy9nm/R8Pp8kWQ2e5zEyMkLdBySg0WgyJv1FIhEMDAyIek4KiImYVhIYJ5NJ8DyfMVluJVL7CopxfwEhq0UT6IjspN8alzorK6alptGld+woNhzHobGxkY1ajsfjeWtpRxc+JB+WatXG8zxisRhmZ2cRi8WozSgheUSZYZK12tpatLW1YXp6GsPDw6JtbJO6pZqUjEYjGhoakEgkMDExUXSb5mpra2EymQCAjVqenZ3N2/kpQ0zyKT0grq2tzbjoi8Vi7KKQECIuCoZJ1urq6nDuuecCAA4dOiRKoKrX6zP6Yk5NTQl+juVUVVVBpVJhdHRUksxMasCGWq1mk++KSXowMDo6ikAgkPc1UEBMpDA8PAy1Wg2bzQaVSgWNhj6eCckX+m0jWcvHGOZoNIq9e/fC6XTCaDTmvYNCe3s7rFYrwuEwDhw4wFoUiSW9q4TFYmEXAjzPC97BQglGRkYQiURgNBrzfiE0fx0UEBMx6fX6BRtjE4kEfD7fgnpfqXsZE1LoKBgmWctHMAzM1YlKUb/pcDhgtVoBzG3iyndWMn3sstvtLroOEili/mytBAXERAwajQZmsxkajQYcxy3YILzYxjepexkTUugoGCZZSw+GxRy4IZWWlhb2eGhoKK+7se12O5uylkgkiqZWOn1srBylLsooKCa54jgOJpOJ/Z4DQElJCaLR6Ip+B+ZvvKPgmJDcUTcJkhWr1QqtVgsACIVCBdfEXafToba2lj0/c+ZM3s7d19e3YOyyUFPW5K6pqSmjjZpcUacJkguDwZBxwQvM1ceHQqGcLwaX6k5BCMkeZYZJVtKHYIh1G7u8vBwcx8Htdue9VripqYkFZG63Gz6fL2/nXrt2LbvQiMViRdNBorq6GhaLBcDcxUhPT4/ss8SUISYrodFoYDKZ2O93SjQaRTAYFPx9jnzYNUMAACAASURBVLLGhKwOBcMkK+nBsFgbm9ra2tgY0QMHDuQ1KGxubmaPUxOc8kGn02H9+vXs+fj4uKwDQqHY7XZUVFSw5zMzM4r4d1NATLLBcRxKSkpgMBjAcRz7eiKRQCAQQCwWy8s6KDgmJDsUDJOspAfDYtQLazQadg6NRoOZmRnBz7GUqqoq1ts2Go3m9Xbjhg0bWNYoEonIZvOYmEpKSjJKUnw+n6JqpKmOmJyNwWCA0Whkz3meRygUQjgclnBVFBwTshR5F+rlaGxsDN/4xjewZs0aGAwGVFZW4uMf/zhef/11wc6RSCSwY8cOcBwHjuPwve99T7Bjy4nYZRIVFRWsTMHr9ea1k0J6Vri/vz+vJRrT09MIhUIA5vrqFjqtVptRIxwOhzE4OCjxqlaH6ojJUsLhMKv7n52dhcfjkTwQXkx6vTHVHCsLxTfCKtjM8LFjx3DllVeyLKbVasXU1BT27t2Ll156CXv27MG9996b83keffRRHDp0KOfjyF2uZRIulwtNTU1L/n36COZ8ZgmNRmPG5rV8lkgAcx9GIyMjqK2txeDgIOs5XIhSo5ZTmfB4PJ73iw+hUdkEARbvihIIBMBxXF4nKOaKMsfKQPGN8AoyMxwOh7F792643W5s3boVJ06cgM/ng8fjwV133QWe53Hfffdh3759OZ1naGgI999/PxobGzOCuULDcRxefvll/PnPf8bx48dFyXCk149OTEwIfvyl2O12FoyNj48jGAzm7dwpyWRSsdnRlaivr2dT9XieR39/v6IChaWMjIxQlrhI6XQ62O12NiwnXSwWU/zP9/zMMWWPpUfxjTgKMhh+8skn0d/fD7PZjBdffBGbNm0CMHf19PDDD+OTn/wkeJ7Ht7/97ZzO8/Wvfx2BQAA/+tGPMlrmFBqe59Hb24v33nsP+/fvF/z4RqORdRWIx+N57WE8MjKCV155BcePH0d3d3fezltsKioqYLPZ2PPh4WFJLjzERAFx8VCpVLBarbBarVCr1dDr9Qs6RhQqCo6lRfGNOAoyGP7Vr34FALj55pszNuqk3H333QCAw4cPo7Ozc1Xn+P3vf4/nn38eu3btwu7du1e/WJJx1SlFW7XZ2Vl0d3fnLSNdXl6ekQlP19PTk5c15BPHcRkDW6ampgp2oyAFxIWvpKQEdrsdOp2OfS2ZTGZ0jSgmlD3OL4pvxFFwNcN+v5/VuFx77bWLvubCCy9EaWkpfD4fXn/9dbS2tq7oHMFgEHfccQeMRiMeffTRnNdc7JxOJ3uspK4Cq8FxHLZu3QqLxYLx8XF8+OGHbANdoeJ5Hj09PWhoaADHcQUfMFK3icKk1WphNpuhVqvZ13ieRyQSEWR4RiGh2mNxUHwjnoILhk+fPs3elFK3D+ZTqVRobW3F+++/j1OnTq34HPfffz8GBwfxj//4j8tuCsvG9u3bF/366dOnczqukqQHw/msF5ZCU1MTKwkpKytbdORzT09PwW2kSyQS6Ovrk/2kOSHR5rrCoFKpYDKZoNfrM74ei8UQCASKZlpkLhbLFhdjgHz69OklP/Oz2aimtPhGSQouGE5vT7XcB1Hq71bazurDDz/Ej370I6xfvx733HPP6hapMLfddhtCoRCmp6fx6quvLhrArVb67cZwOJy3/sKbNm2CVqtFX19f3qbNqdVqtLW1seednZ2K32CzUkruHLEalCVWNpVKBbvdnlECkRqjXGgj6fONyilWjuIb8RRcMJy+KSe96fl8qV3tgUAg62Mnk0ncdtttSCQSeOyxxzJqxlZrqavB7du348iRIzkfP1cmk4n9sVqtggbCABAKhXD06FFUVlbmrVxArVajpaUFWq0WLS0tePPNN+HxeEQ/7/r169lGhFAoVJD1wSnV1dUIBoN5HZ4iZ5QlVqZkMonZ2VmWFY5EIggGg1QSIRIpkgOzs7N5CcxnZ2fR1taWU6sypcU3SlJwwbCYfvzjH+PgwYO46aabcM0110i9nLxI3/gkxqanaDSK3t5e9Pb2Cn7spTQ0NLCd336/Py+BsMFgwLp169jzU6dOLZslVXKpRFlZGSoqKlBRUYHx8fGCrwPPFmWJlSkYDEKlUiEYDAqeDCBELooxvklXcAV8qbG6AJbth5vKQi7WH3IxIyMj+Id/+AdYLBY88sgjuS1SQcQewyyF9IlzZ86cycs529raoNHMXXt6vV4MDAzk5bz5ZjKZMnY4z6+zJNRxQq4MBgNsNtuCrhDJZBI+n48CYSI5im/EU3DBcHrWZbkPndTfVVdXZ3Xcb3/725iZmcE999wDq9WKQCCQ8Sd122x2dpZ9rRCIPYY53+x2O+t3G4/H8xKUWiyWjI0IJ06cyOr7lFZGkRq1nAomQqFQUQwTWQ0a1CEfGo0GpaWlMJvN0Gg0y95+JkRKFN+Ip+CC4Q0bNrAP45MnTy76mmQyyfrvbdy4Mavj9vf3A5jbaWmxWBb8SQVV3//+99nXCkGuY5iXI0WT+paWFvZ4aGgIsVhM9HNu2rSJ/UyOj48XZMcMlUqFpqYmlv2OxWLo7++n2sqzoKBYOhzHwWQyobS0NOO9qNhqJYlyUHwjnoILhi0WC3bs2AEAS05Le++991gHgauuuipva1MiMcskrrjiCuzcuRNbtmzJyweQVqvNaOfT19cn+jkdDkfG1fxSb2BLUUp2uL6+nmXUkskk+vv783KhUSgoKM4vvV4Pu90Oo9HIggue5xEMBuH1eiVeHSGLo/hGPAUXDANzk1mAuUkti7UWefjhhwHMdWzItiH1W2+9BZ7nl/zT2NgIAPh//+//sa8pnclkYgFONBqF3+8X7NglJSUwm80wm81oaGjISz1eU1MTa5jv8XjysnEuHA6zUoGBgYGC/KCtrKxEaWkpez48PFzwg0TEQgGxuNRqNUpLS2GxWDJ6Xs/OzsLj8Sxbh0mIHFB8I46CDIZvu+02NDY2wu/3Y9euXazxtN/vxz333IPnnnsOALBnz54F38txHDiOw/e+9718LlmWhM4Ku1wu9jh9BPPU1FRe+s9KsXEuFArh4MGDePPNN1ecFU6Rc3a4tLQ04//l5ORkXi4yChllicVhNBphs9kySiISiQRmZmYwMzNTdD2wiTJRfCOOgmytZjQa8cILL+Cqq67C4cOHsWnTJlYUnpohv2fPHuzcuVPqpcpaeXk5eyx0iUR6AJWP1ltOp5PtrM1XX8l0hRggqtXqjLITv9+/4ibvZGnUik1YPM9nlESEw2G6g0EUh+IbcRRkZhgANm/ejBMnTuDOO+9ES0sLotEoHA4Hrr/+euzfvx/33nuv1EuUvVTXBUDYzXMcx6GiooI9z0cwnEwm2b9hYGCARqgKIJFIYHBwEIlEAtFotGDbxUmNMsXCiEQiiMViiMVi8Hq9FAgTxaL4RngcX4jFHwUgNYEu2z6BYjGbzbDb7fB6vYLUDDc1NcHhcOCyyy4DMNfQ/tVXX835uNmyWq2IxWKi1gZqNBqsW7cO3d3dgtZCy3UIR6qXcDQalXglxYEyxWdnNBoRi8UW/P5xHFeQ9Y6F4n/+539QX1+f05S2ldi+fTsGBwfxN3/zN6KfK9//NrIyBVkmQYQjRk9Bp9PJHue7zVg+xgO3traitbUVLS0tOHr0aN5LMvKNguD8ovKJpWm1WpjNZqjVasRiMbarPoUCYULIYgq2TILIl5TBsNiMRiPL4Ao9fU0OG+lKS0tZRw4iLSqf+AuVSgWLxZLx86nVaqlnMCEkKxQMk7zSaDSw2+0A5rI0k5OTop4v3x+GbW1tGe3bhM4KSxkQp9rgrVu3DgaDQbJ1kEzFHhQbjUbY7faMi89kMolAIIDZ2VkJV0YIUQoqkyCLstls0Ol0mJ6eFrTu1el0sv6eXq9X9A+r888/HyUlJejr64PL5RJ1EITVamX9GIHsxy4rgU6nQ0NDAziOg06nQ3V1dV6GlpDsFVv5hEajYSOU00UiEQSDQSqJIIRkjYJhsqjNmzdj+/btAIA333wTH374oWDH9vl8KC0tFb1Ewmw2s5KM9vZ20ccvt7e3s9ZNY2NjomW9e3p68rqZbv6o5dnZWTZIhMhPoQfFqTHK8+9OxONxBAKBvAzwIYQUFgqGyaLSB24IOXkudUs3H7fZ04dsjI6OitpBoqKiAlVVVQDmyj8KKSvc0NDA/n+lRi1TwCF/6aUThRQYq1SqjJIInucRCoVoehwhZNWoZpgsSujpc/NFIhFEIhHBj5uiVqszShbEvqV/zjnnsMcDAwOid63IV+1wdXU1rFYrez44OEhBhwIVUl1xIpFgP4PRaJTGKBNCckaZYbKAXq+HxWIBMHfr0ev1Sryilaurq2Ob5wKBgKiDPerr69mAkng8zsZjKp3NZlswHGV+qyqiLEoroeA4DhqNZkF5UygUYgM0CCEkV5QZJguUlZWxxx6PR9CNKC6XS7BjLSe9RELMrLBKpcLGjRvZ856enrxlqcTMDpeUlGSMWvb5fHmZFEjyI5UplnO2WK/Xw263w2KxsE236SgQJoQIhYJhskB5eTl7LFSJhEqlwl//9V/jvPPOyxjzLAabzcYC+kQigf7+ftHOxfM8urq6EIlEEI1G0dXVJdq58kWj0aCxsZEFIJFIhDbMFTC5BcVqtRqlpaUsCFapVDCZTFIvixBSwKhMgiyQXi88NTUlyDGrqqrQ3NyM5uZm0Ucwp2eFh4eHRW3fxvM8+vr6MDAwAKvVmveNZWJ0lkgkEggGg7DZbIjH43C5XEgmk4Keg8iPHDbclZSUwGg0sq4swNzPI005JISIiYJhsoAYmeGmpib2WMxBG1qtFvX19ez5mTNnRDtXukQiAY/Hk5dzzSd0QMzzPAYGBhAOhxEKhWhwQRHKd22xTqeDyWTKmG7I8zz7GSSEEDFRMEwWSK8ZFioYTu/sIGbtaUNDA+uH6/V6MT09Ldq5Cp3Y0wGJ/ImdLVapVDCbzQsmRcZiMQQCASQSCcHPSQgh81HNMMmg0+lgNpsBzHVGEKJ7gMFgQGVlJYC5PrViDtsYHh7GqVOnEAqFRM0Kr1+/PuOiQWq5bqbTarUCrYQUKjE23Vmt1oxAOJlMwu/3w+fzUSBMCMkbygyTDHq9Hi6XCw6HA9FoVJBOEg0NDWwz1vj4uKi7wCORCDo6OtDR0ZFRdygkm82G9vZ2AHPB98GDB2VRU7vacgm9Xo+1a9fC6/ViZGSExtiSsxIqYxwMBlFaWgqe5xGJRBAKhejnjxCSdxQMkwx+vx/PPfccACzazmg10uuFxR5+kU6sD9VUIAzM9UGVQyC8Wmq1Gk1NTVCr1XA4HFCpVNQ5gqxItoExx3ELfidjsRirS6ephoQQqVCZBFmSUEFeQ0MDeyxmm7N8qKyshNPpBDD330duY5dXWi7R0NDARtsmEgmqEyY5WaqUwmAwwG63L6gNBuYGaFAgTAiREgXDRFR2u52N841GoxgbGxPlPE6nk9U6iyk9K+xyuRAIBEQ/50plGxDX1NSwSYPA3KhlMUdkk+IyMjICr9cLk8kEs9lM/YIJIbJFZRJEVOldJAYHB8HzPFwuV0bpRK44jsO2bdtQUlKC8fFxHD58WJQpcA0NDSgtLQUwt7nw9OnTgp8jX8rKyjJa6I2NjWFmZkbCFZFColarUV1dvWCTaSwWw9TUVMbPHiGESI2CYcKo1WpcdtllcLvdmJqawvDwcM7HTK8hFKtEoqqqCiUlJQCA0tJSURr0q9VqbNq0iT3v7u6W9SCA5TbTmUwm1NbWsuder1fUDh+kuJSVlaGqqoq1OATmSnAmJiZYGY4cBnwQQkgKBcOEKSsrw5YtWwAAHo8Hv/zlL3M+5iuvvIIPP/wQjY2Nom2eS58419/fL8qGtjVr1sBoNAKY61jR3d0t+DmEtlhArNVq0djYyDpthMNh2jBHBGE0GlFbW8suTFN8Ph9GRkaW7CJDgTEhRGoUDBMmfQyzUMM2eJ7H6OgoRkdHBTnefCUlJayHcWo0stB0Oh1aW1vZ81OnTilmw096QKxSqdDU1MQydrFYDC6Xi1pZkZzZbDbU19dntDOMRqMYGRmB3+/P+jjzN95RcEwIyQcKhgkjxhhmsTU3N7MP4PHxcVFGt27YsIENpfD7/YrtiKHValkgnEwm0d/fL2rPZ1I8AoEAkskk1Go1kskkJicnMTExkfOFFgXHhJB8oGCYMOmZ4ampKQlXkp1UpjNFrIlzfX19MJlMqK6uxokTJxSXSU1lh6PRKHp6etDY2Ai32y3KhQMpTvF4HGNjY7BarRgeHsbs7Kwo56HgmBAiBgqGCSNkZthgMKCxsRGDg4OiBV21tbWsR24wGBStbZvf78eBAwdgt9vh8XhEOYfYUgFxLBbLeXQzKV4qlQpOpxMcxy0ofXK73Xm/o0TBMSFECBQMEwBzt9BTbcOSyWTOQV9DQwOuv/56AMDJkyfx6quv5rzG+dI3zrlcLsGPP59SA+GU1Y5rJgQArFYrampqoNPpwPM8PB6P7PpSU3BMCFkNGrpBAMwNx0jxer1IJBI5HS+9v7DP58vpWIuxWq0sk51MJvMSDCuNXq/HlVdemZHxp6wwWSmdToempiY0NTWxCXIcxy3oISxH6RPx5gfKhBCSQplhAiAzGJ6ens75eOnB8GIbznIdvFFVVcUej4yMCN7zt7KyEjabDT09PTlfGEhBpVLhwgsvhM1mw6WXXoojR47QBQNZEY7jUFFRAafTCZXqL3mTeDyO0dFRRd4pWSwgpuwxIYSCYQIAGVmeXD/kysrKRB/B3NXVhcnJSbS0tAje3YHjOJxzzjmwWq1oaWnBu+++q7gP/i1btrANkRzHZUzko3IJcjZmszmjJh+Ya104PT2NsbExRV4gLoVKKwghFAwTAJnBcK6bYNKzwgMDA6J1X/B4PDh06JDgx21sbGTBvEajQTAYFPwcYlqzZk1G1v3EiRMYHx/PeA0FxGQxKpUKdXV1sNlsGV8PhUIYHh4WZcy53FD2mJDiQ8EwATC3yc3j8aCsrCzn0bzpgZjSbs2r1Wps3LiRPe/q6hKtTZQYnE4nzj33XPa8v79/yWl5FBCT+ZLJJOupDcyNUR4bG1NM33GxUIBMSGGjYJgAmAtahQhcNRoN6uvrM46rJOvWrYPBYAAwN6pYSRvOTCYTzj//fDaEZHp6Gh9++OGy30MBMZlveHgY69atg9frxejoqGKmLebbUhvyKEgmRHkoGCaCqqmpYVPOpqenVzSKNRttbW2IRqMYGBgQ/ENar9dj/fr17PmpU6cUUxup0Whw0UUXsd3+4XAY7777LpLJ5Fm/lwLi4qRWq1FeXr6ghCYSiaCzs1NRd0TkhLLIhCgPBcNEUA0NDezxwMCAoMfW6XRYv3491Go12tvbsX//fkFrGNva2lgg7/P5FDV2+fzzz2d1zolEAgcOHFhRD1gKiItLWVkZqqqqoNFoEI/HF5RBUCAsLMoiEyJvFAwTQZ2tpVoumpqaoFarAQAzMzOCBsJms3nBpjOlqKqqymg1d+jQIXi93hUfhwLiwmcwGFBbWwuTycS+VlVVJUhvcbJyFCQTIg8UDBNcc801qKiogNvtxqFDhzA1NbXqY73//vtoampCfX09BgcHl33tSnsNp0+cO3PmzGqXuKj29nbWS3ViYmLBrWM5Gxsbw6FDh7B161Z0d3djaGho1ceigLgwqVQqVFVVweFwsJpyYK714cjICAXCMkNBMiH5RcEwQXV1NcrLy1FVVYVjx47ldKzu7u4luxfkorKykmWzZmdncwr45isrK8v4kFFSVjilv78fHo8HMzMzOR+LAuLCYrPZUF1dndElIplMYnJyEhMTE6K1PiTCoyCZEHFQMFzkVCpVxvS5XLLCYkrPCvf392e1MSxbXq8Xx44dQ2trK8bHx1dVYiAHQgTCKRQQK59er0dtbS3MZnPG1/1+P4aHh6kuuIAsN2qaAmVCzo6C4SJnt9tZHa7P50MsFpN4RQsZjUZUV1ez5319fYIeP5lMoqenB/39/RljZ+VKpVKhra0NnZ2dora9ooBY2ex2e0YgHIvFMDIyAp/PJ+GqSL5RoEzI2VEwXORSI3uB3CbPabVaJBIJQTO2Kc3NzazOcWJiAoFAQPBzAJDlhcBitm3bhoaGBlRVVeHAgQMIhUKinSvVZ5mCYuUZHx+HzWaDVqvF1NQUxsfHRfn9JMq1XKAMULBMigcFw0UufQzz9PT0qo+zefNmXHjhhRgYGMCRI0cEa6vGcVzGJjuhN84pzfr161n7utLSUlRWVgqeKV8MZYnlLVUPnH5Bx/M8BgcHkUgkVtRmj5AUCpZJsaBguMil1wt7PJ5VH6ehoQE6nQ5r164VNGCtqanJmAg3OjoqyHE1Gg22bduGzs5Oxdw2rqqqwqZNm9jzvr6+vATCKRQQyw/HcaioqIDT6YTf71/QzjAYDEq0MlIMKFgmhYKC4SKXnhlebZmEWq1GXV0de76S/sJna6+m0WgQjUah1+vR19cn2M739evXo66uDnV1dejs7MTJkycFOa5YLBYLzjvvPFYuMjU1hSNHjuR9HRQQy4fZbEZtbS30ej2AuTsFFotF8KmPhKwWBctEKSgYLnLpmeHVlkmkj2D2eDyCfhj39/djcHAQNTU1gnW6MBgMGQGdkF0YxKDVanHRRRexW+HBYBDvvvuuZC2xqI5YWhqNBjU1NbDZbBlfD4fDom6oJERo84NlKfZtpDaW5uM8RL4oGC5iFosFOp0OwNwH6WrrCtMzu2KMME4mk4L2Fd64cSML3r1e71mHg0iJ4zhccMEFrCtAPB7HgQMHZNEWi7LE+VdeXo7KykrWAQaYG789NjaW0wZYQggpZhQMFzGhOkmkj2B2uVy5LEl0FoslY73Hjx+XcDVnd84558DpdLLnH3zwgawy2RQQ50dJSQlqa2thNBozvu7xeDA6OkoZYUIIyQEFw0XM5XLhqaeegsPhWHXLJZPJxIK1RCIhWJaV4zhRygDa29tZ3e3Y2BgmJycFP4dQ6uvrMwLNU6dO5eV23kpR2YS49Hr9gv+2kUgEw8PDtEGOEEIEQMFwkZuZmckp05heIjEyMiJYXdRll12GaDSKM2fOYHx8XJBjVlRUsOEdPM/LftPc2NgYxsfHUVlZiaGhIXR0dEi9pGVRllgc0WgUXq8XNpsNyWQS4+PjmJqaojHKhBAiEAqGSU7SSw5W2+ZrfkcJm83Gulw4nU68/PLLggTZ7e3t7PHAwIDsW6rFYjG88847grerExNliXO32F2RkZER8DyPsbEx2ohDCCECk//sWSJr9fX17LFQm+daWlrY46GhIUE+/Ovq6ljnjEQigVOnTuV8zHzgeR7d3d1IJBJSL2VFUkExyZ5KpUJNTQ1aW1sXjAWPx+MYHBykQJgQQkRAwXCR0ul0KC8vz9iVvlIGgwGTk5OIx+MIh8OC1N/qdLqMAFuojOjGjRvZ456eHoTDYUGOK7T0TY1K19PTQ0Fxlmw2G1pbW1FeXg6dTofKykqpl0QIIUWDyiSKVF1dHT75yU8ikUigo6MDr7766oqPEYlE8Nxzz0GtVi/oebpaDQ0NLED3eDw5TcVL9/7776O9vR2lpaXo7OwU5JhC27BhAzZu3IiOjv/P3n0Ht33f9+N/fjAIgADBTQLcBEiKQ1za1LA1LMWVhxI7XnEbx60zmsu1dXvJpb32rtfrufe9pu3lLsk1wyNprHgksi3LVmRZshZFitTgFiUS4CZAENwLxPz9gR/ewYcgKYIASQB8Pe5ywfzgQ5oin3jj9X692sNm5XolqJZ4aRKJBGlpaYiJifG5nRBCyPqgMLxJJSUlAXBPjwu0Z63D4Qhaj9Pc3Fx2OZijhsfHx3Ht2jXIZLKQbEOVlpbGVq8LCwsxMzOzJj2bNwrVEvNxHIfU1FQkJSXxSiI8AwBCvZ6dEEIiCYXhTSoYY5iDLTk5ma2Q2Wy2NRmGEYrlEbGxsdixYwe7PjQ0hN7e3g08o7VDoRhQKpVIS0tjA28Ad2242WzG0NDQqtscEkIIWR2qGd6kvMcwB6sUIRDd3d28VeHe3t6w2zS2GhKJBFVVVWwi3vT0NOrq6iK+bdZmrSXOyMhATk4OLwjPzMygo6MDBoOBgjAhhGwAWhnepLzD8OjoqN/P37lzJ2JiYtDd3Y3e3t6ASw8UCgXS0tLY9WCUSJSVlWFmZgZ6vT4kw6Vn1HJ0dDQA92p4TU3NpukYsBlXiWdnZ9mnMna7HQaDISTejBJCyGZGYXgTio6OhlQqBeBu6L+aKVbFxcVITExERUUF/vCHPwRc31paWspqJ4eHhwMeORwbGwutVguO46DVavHFF1+EXMisqKhgtdsulwt1dXWYmpra4LNaf5spFI+OjiI+Ph4WiwVGo3FTfPpBCCGhjsokNiHv9l2rWZVSKBTsGHa7HQMDAwGfU09PD/r7++F0OoPSTs177PLk5GTIBWGNRsMrC2lpaQnapL1wFUmt2EQiEbKysiCTyXzu0+v1GBgYoCBMCCEhglaGN6FAN89lZWWxy4ODg0HpzjA4OIi6ujpIJJKAu1ukpKSwPq2hOHY5OTkZZWVl7Hpvby86Ojo28IxCS7ivFCcmJkKlUkEoFCIqKson4IdiyQ4hhGxmtDK8CXmvDK8mDHuPYA5m+6/u7m7Mz88HHBa8xy53d3eHXOmBQCBgq4Kjo6O4ffv2Bp9RaAq3leLo6Gjk5+cjPT2d9cqOjo6GQqHY4DMjhBCyHFoZ3oS8V4ZXs3nOe2U41FqAZWVlsQEgdrsdd+/e3eAz8jU0NIRLly6hoqIC9fX11EHgAbwDcSiuFguFQqjVat6/K8A9lGZgYGBVNfmEEELWD4XhTchisWB2dhbR0dF+rwwnJydDLpcDcO+MD7TOVa1Ws5HOgRIIBLyxyx0dHbBYLAEfdy1MTU3h6tWrG30aYSfUSigSEhKgUqlYazwAcDqdGBoagtlsppIIQggJJ09kpgAAIABJREFUAxSGN6EzZ84AAGQymd9DKHJyctjlQFeFRSIRvvKVrwAAWltbUV1dHdDxtFota1NmsVhCqg5XKBTShqkg2uhQLJFIkJGRwd4YekxMTGBwcDDkNmwSQghZGtUMb2KrmcbmHYa7u7sDev38/HxIpVJIpVJotVrY7fZVHzMqKgpbtmxh19vb20Nm7HJmZiaOHj2K2NjYjT6ViOOpK96I2mLPGy8AsFqt6OrqQk9PDwVhQggJMxSGyYqJxWLeYIxAw3B5eTm73NzcHNCx8vLy2FSvqampoAztCIb4+Hhs27YN0dHRePjhh1lfYRJ86xmK5+fnMTw8zEoi7t27F3IbNQkhhKwMlUmQFfO0iwIAk8mE2dnZVR8rKSmJBWuHwxFw+7N79+7B4XCgoKAALS0tIVGrKZVKsWfPHvY9m52dxfj4+AafVeQL9oY7iUQCqVSKiYkJ3u0mkwljY2OYn58P+DUIIYRsHArDm0xFRQXsdjvGxsb8noDV19eHX/7yl8jKygq4/rWiooJd7ujoCChYA+5Afe/ePXR1dQXcpzgYBAIB9uzZw4YuWK1W1NTUhEzpxmYRSDDmOA4pKSlITk6Gy+XC7OwsrwTC6XRSECaEkAhAYXiTqaqqYgHtF7/4Baanp/16/tTUVMCruBKJBEVFRex6Q0MD7/7u7m5ebbI/QiEIA8C2bdtYqy2n04kbN25Qi60N5s+mO6VSibS0NFZ6A7g7n4RaK0FCCCGBozC8iUilUhaEbTab30E4WIqKiiAWiwEAw8PDGBwc3JDzWCv5+fm8XsxNTU0YHh7ewDMi3pZbLRaLxUhPT4dSqeTdPjMzA5PJtC7nR0io02g0KCsrQ0xMDE6ePBnwJ4UCgQAKhYL9XSBkvVEY3kQCHbYRLN6jiJuamlZ9HJVKhezsbLS2tm5YsF8oNTWVNwGvq6sLer1+A8+ILMcTjDmOQ0FBAbZs2QKB4E/7iu12OwwGA8bGxjbqFAnZcFFRUSgqKkJpaSlycnIgFAoxMTGBzs5OyGSygH7/KpVKZGVlQSAQ8Pp1E7Ke6CdvE/EOw/7+cddqtZicnAx4hTMtLY11VLBarQFNiCspKUFsbCzS0tJQX1+P/v7+gM4tUDExMdi1axc4jgMAmM1mnxIQEnqSk5NRUVGBmJgY3u0jIyN+19UTEmkeffRR7N69GwAwODiI27dvo6+vDyaTCePj4wHXzVssFt4bUEI2AoXhTWS1K8Mcx+HYsWOQyWSYmZnB22+/veqVAO9V0/b29iVrfB9UN5ydnc369trt9pAoQ8jKymIf883OzuLGjRsh0dWCLI3jOFRWVkKhULDbxsfHcefOHd4bxlCZeEfIWuM4Di6XCwKBAE6nk00pbWxsxPnz52Gz2YK6N8NqtcJiscDlclGPbrJhKAxvIt5h2J8xzCqVitUau1yugD4Su3jxIgYHB1FRUYHGxsZVHUMoFPqMXQ6FXf2tra2w2WzYsmULampqQuKcyPJcLhcaGxuxb98+2Gw2tLW1Qa/X+7yJWdi/mMIxiQQCgYAtLJhMJpjNZlitVnAcB6fTCQBoa2vD8ePHIZfL2SZgoVAIjuOC1h2ns7MTTqeTuu2QDUNheBNZ7cqw92awQAdt2O12tLS0oKWlZdXHyM/PZ+F8bm4upMYu379/Hz09PRSEQ1RMTIzPcIyhoSE0NjZiYGAAFotlRccJdi9jQtaLQCBAVlYWduzYgS1btrA6XavVivHxcXzwwQcwGo0A3KvEns2j6enpyMnJAcdxyM3NhUKhgFQqxcDAAFpbWwPqoe4J3oRsFArDm4RIJEJcXBwA9y8ef35xeYfhvr6+oJ+bPyQSCfLz89n1tra2kKvppCAceqKiolBSUoLc3FzU1tb6dDDR6XSrPjatGpNwIRAIsHPnTlRVVcHpdOL27dswmUxwOBwoLy9HTk4Ojh8/jtOnT8NsNkMgEMDhcKCtrQ0HDx7EE088AaVSCaFQiMnJSSiVShQVFWH37t34/e9/T60HSdiiMLxJJCYmsstjY2MrDpAikYg3grmnpyfo57aUxeqGCwsLWV3uxMTEhv7yTUxMRGZmJpqammhlI4RlZ2dj69atkEgkANzdTIaGhtbsTdRiI6EpIJNQEBsbiz179mB6ehrnz59Hf38/+3fQ09ODo0ePoqioCBqNBmazmZULecLw2NgYrly5gvb2dgDu34H79u1DYWEhjh8/jnfeeYembJKwRGF4k/AOw/7UC2dkZLBxwmazedWT4nbu3AmDwRBQxweFQoHc3Fx2vbW1dcM2qMlkMuzevRtSqRRKpRI3btygFeEQExsbi4qKCt7PPuB+EyUSidb1EwVaPSahYH5+HufOnUNvb6/P7/LJyUn09/ejqKgIUqkUwJ/KF4aHh3Hx4kU0NTWxseQcx2FwcBDnzp2DSCRCQUEB8vPzUV9fzzbhERIuKAxvEmazGXV1dUhKSvIrkHqvzK52VVipVGL//v3gOA5msxknT55c1UaJkpIS1oJneHiY1bWtN6FQiKqqKvYHQ6FQUGugECISiVBcXAytVsva3AHuwRmNjY0b9nPjjcIxCQalUonExEQMDQ2taKFidnaWreouDKwOh4O1vVzs38jVq1d51z3PnZycxP3791FQUICMjAzU19dTECZhh8LwJmEymVY1QSs7O5tdXu3mufLychZKpqam/ArCnlKJmJgYpKens9ubm5tXdS7BsH37dl799Y0bNzA3N7dh50P+JCMjA6WlpWyDJeD+b9TR0YH29vaQqy/3oNIKslJisRgVFRXYtWsXEhMTYbPZ2Ebizz//fMWfUC0MrBKJBMnJybBYLDCbzX6dk+dYVquVtWQjJJxQGCZLiomJYR8x2+12DAwM+H0MgUCAkpISdn217dSmpqZw+fJlbN26FbOzsxtWl1ZYWIiMjAx2vaGhwa+yE7J2CgoKeH2sAfebwIaGhpCZUOgPCshkMSUlJXjkkUcwPDyMS5cuQSgUYsuWLdixYwfi4uLwzjvvrOpNn0ajQUZGBm7evLlst6HFSiAKCgoAAAMDAxSESViiMEyW5F2f29fXt6rShry8PERHRwNwB9qurq5Vn8/IyAguX77MapjXW1paGq+/cWdnZ8Ct5kjw9PT0oKCgAFFRUbBYLGhqatrwqYTBtlhABigkbxYxMTF44oknMDo6ig8//JCt4NbW1uL5559HXl4etm/fjrq6Or+OGxUVhV27dsHhcODy5cvLPtYThDmOQ3x8PLZt24YtW7agp6dnXTdYExJMFIbJkgYHB1FTU4PMzMxVt54qLy9nl5ubm1dVS7awq8RGfNStVCqxY8cOdt1kMm1oqQbxXaGan59Ha2srYmJi0NbWtqka+NMq8ubg2Tdx584dmM1mcBwHgUCAubk5XLt2DU8//TS2bduG3t5ev2rj8/PzkZOTgwsXLmB6enrR1V+JRIKUlBTExcUhOjoa8fHxyMzMhFqtRnd3N86ePcub2khIOKEwvAnk5+dj27ZtGB0dhU6ng16vX9HzzGYzzGYzampqVvW6ntZjgLtuM5BBGxtJIpGgqqqKNaefnp6mUcsbKDo6GmVlZZidnUVTUxPvvkA+eYg0tIoceaKiogDAZ+MbAPT29qKjowMlJSXIyclZcRiWyWTYv38/pqencefOHZ/jezidTlRVVaGwsJDVBg8PD+Ps2bNobW1ddachQkIBheFNICUlBenp6UhPT8fMzMyKw3CgvFeFOzs7/a7bFIvFePbZZ9HU1ITp6ekNC5/l5eWQy+UAAJvNhpqaGthstg05l82M4zjk5+ejsLAQIpEILpcLvb291NfUT0uFZICCcqjz/N7xlIot/GSku7sbhYWFyMzMRHNzMxufvJyysjKoVCqcPn2aPV4ulyMhIQEJCQloaWmBw+GAzWbDtWvXcPfuXczMzKC/vx9Wq3UNvkpC1h+F4U3Aewzzen2MJRaLefW1DQ0Nfh9j586dSE1NxdGjR9Hd3Y3bt28H8xRXrLm5GXK5HHFxcaivr/cZ50vWXnJyMsrLy6FUKtltHMchOTmZwnAQUVAObXNzc3C5XIiJiUFUVJRPGDUajZicnERycjIUCsUDw3BcXBz27t2L8fFx9PX1QaPRID09HVlZWUhLS0N0dDSmpqbYAsrg4KDP9EZCIgGF4U3AOwwvt0vY22K/aP1RWFjIPtIbGRnxeyOTXC7H9u3b2XVPb8yNMDc3h8uXLyM1NTUketRuJlKpFFu3buWNBAeA8fFxNDQ0rPjnmQSOgvLGGxwchMViQWpqKuRyuc/v6PHxcUxMTCAtLY1N6lyO5w1mT08PnnjiCaSnp0MoFGJkZAS3b99GS0sLhoaG1urLISRkUBiOcAKBgPXEBVYWhqOiovCd73wHIyMj6O7uRnV1td+vGx8fD5fLBY7jfOo6V6Kqqor9MjeZTGhrawMAn/HM68XpdMJgMGzIa29GHMchNzcXJSUlvD/qNpsNbW1t0Ov1VLMdQigor4/R0VGMjIwgNTUVycnJPp/0zczMwGazQSKR8FqccRzHehJ7JshJJBK2pyMzMxODg4P47LPP0NrauqLyCkIiCYXhCBcXF8fqyyYnJ1dU65qZmQmRSITU1FRwHLeqMHzlyhU0NDRg69atuHv3rl/PTUhI4PWLvXLlit+vHwi5XI7o6GgMDw+v6+sSN4lEgn379vHexAHu9n7Nzc2wWCwbdGZkNZYLyh4UmFfGbrdDp9MhIyMDeXl56Onpwfz8POsq4antBcD2OQDu4UkvvfQS+vv78frrr7Nj1dXV4caNG9DpdNQfmGxqFIYjnGdoBoAVD4fw7i8cSB/dyclJXL9+3e/nHThwgI037unpQW9v76rPwV8ikQhVVVVQKBRoampat82G5E/m5+d5bdGmpqbQ0NBAb04iGAXmlWtpaUFxcTEqKirQ0dGBjo4OuFwuOBwOyGQypKSkYHx8HP39/axFmuffjkQiYcdxOBy4f//+Rn0ZhIQUwUafwFoyGo3427/9W2i1WkilUqSmpuKJJ57AhQsXVnW84eFh/PznP8czzzzDjimXy1FUVITvfe97K/qFvt48s+aBlYdh7/rM9W5VlZGRAa1WC8C9U3rhqvBaD7nYuXMnlEolBAIBSktL2cAQsr4aGhpgs9nQ2tqKCxcuUBAm6OzsfOD/woFAIIBSqURGRgbUarXfzzebzbhx4wbEYjEOHz6M7OxsyGQyqNVqHDt2DMnJyWhvb2eb7QB3+cR//dd/4Wc/+1mwvxyyQSjfBFfErgw3NTXh8OHDLAAqlUqYzWacOXMGn3zyCV577TX88Ic/9OuYaWlpvBUrhUIBq9WK9vZ2tLe34/XXX8cbb7yBF154IahfSyC8w/BK5s0rlUr28bTVal33OtmHHnqIXW5ra1vXEFRSUsL743T79m3qnbnGYmNjodVqcefOHV4N8OTkJM6ePbupBmeQwPnzB3s9V5pFIhGUSiWUSiUUCgX75Mtms63qd+zt27eRkJCAvXv34hvf+AbMZjPEYjGUSiWam5tx8eJFn+eE40hysjjKN8EXkSvDc3NzePLJJzEyMoLKykq0tLRgYmICY2Nj+Id/+Ae4XC780z/9Ez777DO/jmu32/HQQw/h17/+NQwGA6ampjA7O4tr166hoqICFosFX//611e1YWyt+BuGs7Oz2eX+/n6/68ieffZZ7Nu3j9cCa6UKCgqgUqkAuL/XS5VYrMXqcEZGBrZs2cKu379/H319fUF/HeImEolQWlqKQ4cOIScnBxqNxucxFITJWlrJSnMgK85SqRQpKSnIy8tDcXExMjIy2KdOHmKxeFWfPrlcLpw/fx5vv/02rl+/jsnJSXR2duI3v/kNTp06RX3QIxjlm7URkSvDP//5z9HT0wOFQoGPP/4Y6enpANzvnn70ox9Bp9Phww8/xD/+4z/i2LFjKz7u5cuXeSuXgLv5+b59+/DZZ59h69atMJlM+J//+R+8+eabQf2aVkMkErFVXqfTuaIyCe8SCX9rdTMzM5GRkYGMjAyUl5fjf//3f/0K05WVlezynTt31q2fb3x8PK+Nm9FoDNtpeeEgPT0dZWVlkMlk7LbCwkJ0d3dvyKhtQh7En0AsEAhw/Phx1lpyMXNzc5icnMTk5CTm5uYCOq9I//ia8FG+WRsRuTL89ttvAwC+9rWvsR8Ub9///vcBuD9qunfv3oqPu/AHxVtycjKOHz8OALh165Y/p7tm7HY73njjDZw+fRqXL19+YNDgOC6gMFxRUcEu37171+9V5Q8++AB1dXWYnp5GXV3dso8N1uqwVCrFnj17eB03HvTaZHUUCgX27duH3bt384KwyWTClStXKAiTsCMWi31Cr9Pp9Hkj73Q6YTKZ0NjYiD/+8Y84e/Ysqqur0dzcHHY1z2RjUb5ZGxG3Mjw1NcX+Y33pS19a9DF79uxBbGwsJiYmcOHCBd7H44HwdG4IpT/qExMTrK/kg6SmprKQMj09vaKyCo+YmBi28Q0AGhsb/TtRuGuUr127htra2nX5iFwgEGDPnj3sa7ZaraipqaGP54NMIBBgy5YtKCgoYG86AMBisaC5uZnKUUhYkcvlUKvVUKvVSExMxP3791kfdA+DwYCYmBgMDQ3BYDBgaGhoRaULqw3E1Gljc6B8s3YiLgzfvXuXbcQpKSlZ9DGeP851dXU+v8QCcfnyZQDg9cgNJ94DLXp6evx6bmlpKa8dWiCTwVYaRru7uwMawlFZWcmm87lcLtTV1VGz+SBLTU1FRUUFr+epy+WCTqdDW1sbvfEgYSEhIYEF4IX7IVQqlc/fEZ1Ox1qerYdAV5UpTIcHyjdrJ+LCsPfO3LS0tCUf57kvWN0SPvroI9y8eRMA8PLLL6/4ed61qt78HVQRDHa7HVNTU4iJifGrDEEgEPD+gfhTYO/pg7kRhoeHkZGRAaFQiKamJphMpg05j0iWmJjIC8IjIyNoaGhY8acVhGwEoVCI1NRUqFQqqFQqSKXSRR/n6e8rFAp5K2bhtnoWrBINCtXLu3v37pJ/81dSfhBu+SacRFwY9l7Z865LXMizgzcY7WYGBgbwrW99CwDw5JNP4tFHHw34mMGgUCgwMzOz4rB58+ZN3Lx5E0lJSX6FFY1GA4VCAcD9/dTpdCt+7rFjxyAUCnHt2jVMTk6u+HkegawO9/b2Ynp6Gmq12q9zJit37949NtGwtbV1zftEExIMcrkce/bsWfQ+u92O4eFhGAwGGAwGzM/Pr/PZha5AQ/VGfC/n5+fXpV47GF8b5Zu1E3FheL1NT0/jy1/+MkwmE7Kzs9moy5Va6t3g9u3b0dDQENC5vfzyy+A4DuPj4zh58uSKP5L2p1YYAMrKytjl1tbWFW+cS0pKQnFxMTiOQ15eHt566611D8Sjo6MBlXSQP0lKSsLs7CyvN7PD4cCNGzcwOzsLq9W6gWdHiK/Y2Fio1Wr09PTwujpMTk5iZmaGfaphsVhgMBhgNBphMpnCbuWXhIaioqKw2oAWaL4JJxEXhr0/kp2bm0NMTMyij/P8wfasaK6GxWLBiRMncPPmTSQnJ+PcuXO8vr4bSalUQiwWA3C/S1yr2sz4+HgWRJ1Op18lEgcOHADHcQDcdcarCcL+EIvFcLlcVKcaZBKJBKWlpcjKysLg4CBqa2t594+Pj2/QmRHCx3EckpOTWf2vZwXNarX6jF7X6XSIioqCwWDA2NjYRpwuITyUb9ZOxIVh7zqawcHBJXdSDg4OAsCqxmEC7l+eX/3qV3Hx4kXExcXhs88+C9quzWDw7PwEVj6GeTXKy8vZZb1ev+LewJmZmcjNzQXgDtHXrl0L6DwetDrMcRx27doFqVSK2tpa2igXBBzHITc3FyUlJeyNV1paGtRq9bpPLiRkKWKxGCqVCmq1Gqmpqexn1ZtarfYJw9TqjIQayjdrJ+LCcGFhIduU1drauuh/QKfTyfrvFRcX+/0adrsdL7zwAj755BMoFAp8+umnvB67ocA7DD+oDCAmJgaHDh1Cd3c3uru7/VqhNZvNGBkZQWJi4orLOjiOw8MPP8yut7W1rWlgB9w7YFNTUwEABw8exPnz5+lj+wDEx8ejsrKSDXXx6O/vp1U0EhKSk5NRWFiIxMRE3tQ3b1arFUNDQxgYGFjnsyPEf5Rv1k7EheGYmBjs2LED9fX1OH/+PJ566imfx9y4cYNtEDty5Ihfx3c6nXjppZdw6tQpyGQynD59GlVVVUE592DyDsMPqgHOyclBXl4e8vLy0NXVhQ8++GDFr9PS0oKWlha/VgMLCwuRkpICALDZbEuOXfbXUqvD2dnZyM/PZ9f1ej0F4VUSi8XYunUrcnJyWIkL4O5/2djYSB05SMjwlEQsNDMzwza/mc3mDetmQ4i/KN+snYicQPe1r30NgHtSy2IB7Uc/+hEA9yY1f5b+XS4XvvWtb+HkyZOIiorCqVOncOjQoeCcdJD5UybhHSBXu9t/pUFYKBRi79697PqtW7eCsuPVY+H5JyQk8MY8DwwMbEjbukiQnZ2NY8eOITc3lwVhh8OB1tZWXLhwgYIwWVdCoRBqtRrbtm3Do48+yhvoArgXATyDLkZHR9Ha2orPP/8c586dQ1NTE4aHhykIk7BD+WZtRGQY/va3v43s7GxMTU3h8ccfZ42np6am8IMf/ACnTp0CALz22ms+z+U4DhzH4V//9V997nv11Vfx+uuvQyQS4b333gvZFiMcx/EK3ZdbGRYKhcjOzmbX/R224a+KigrExsYCcG8A8PQuXAsymQx79uxhH5FOTEys6etFsvj4eGzfvh0SiYTdZjAYcP78edy7d8/v0duErIZUKkVOTg6qqqrw+OOPo6qqCjk5OYiOjmZlUB5OpxO1tbX45JNPcOnSJdy7d2/NN+kSstY2e75ZKxFXJgG4Q9BHH32EI0eO4Pbt2ygpKYFSqcT09DScTic4jsNrr72GY8eOrfiYvb29+PGPfwzA/QP17W9/G9/+9reXfLzRaAz461ituLg4iETu/7TT09OwWCxLPjYtLQ1RUVEA3Lv+V9pmTKFQ+L2iK5VKsXv3bna9pqZmTcoVuru7odVqUVVVxZrlz8/Po6amhloirdLY2Bj6+vqQmZmJ2dlZNDY20iY5si6USiXr/uCZGLmY5ORktnHIY3h4eK1Pj5B1tdnzzVqJyDAMuLsctLS04D/+4z9w5swZDAwMIDExEbt27cKrr766qloaD5vNhqGhoWCfctB418k96I9BZmYmu7zSVWGJRIKXX34ZIyMjaGlpQXNz84o+biwoKGDhdGxszK82bP4qLCxkm7s8K0Te/W/J8mQyGa/vKgA0NzdjZmYG9+7dozcVZF1UVlayrjOLmZiYgNFoxODgIG3cJJvGZs43ayViwzDgnhn/4x//mL3jWYmlQl1OTk7Y1Jf5s3kuKyuLXe7t7V3R8YuKili7IoFAsOJQ29TUhPHxcTz00EOora1ds4/Wd+/ezauVamhoWPNuFZFCLpejoqICSqUS58+f5/VktlgsQZ11T4iHWCyGQqHwCbQjIyO8MOx0OmE2m9kGOHqDSzarzZpv1kpEh+HNSiAQYH5+HhKJZNmyh6ioKKhUKgDufyR9fX0rOr73xDl/V3d7e3vx29/+1q/n+IPjOF4vxjt37tAI4BUQCATYsmULCgoK2EakoqIiNDc3b/CZkUgll8tZ+UNiYiKsVis+/fRT3mOMRiNrf2YwGDA0NMQ2xRFCSLBQGI5A1dXVqK6uhkwmW/bj7MzMTLa5zGQyLVtb7JGens4251mtVrS3twfnpIPE5XLhww8/xIEDB5CSkoLLly/D6XSuelzzZqBSqVBeXs6bbrTZVwnI2khISGABWKlU8u6TSqWIj4/nrQ5brVacOXNmvU+TELLJUBiOYAtrPhdaTRcJ7zZld+/eXdEGOIFAsK7dBlwuF65cucJ73QdNqNuMZDIZysvLeSvpgLsNVUNDA41RJkGhUqmQlpYGlUrF9gwsZnR01Kc9GiGErAcKw5tYRkYGu7ySUgK5XI68vDx2fSUT57Kzs3HkyBFcv3593VeRqd3X4jiOQ35+PgoLC1nXEcC9CtfS0kJlJSSotmzZwtvH4OFwOGAymWAwGGA0Glf0yRQhhKwFCsOb2MmTJ5Geno7s7GyflkSLKSkpYWUV/f39D9yUxnEcHnroIcTFxeH48eOIi4tDbW1tUM7d25EjR2CxWFBdXb3s42h12O3AgQO8PtSA+3vT0tJCk/nIqnjan83Pz/u8mTIYDCwMWywWGI1GGAwGmEwm6kpCCAkJFIYjjFqtRlxcHEZHRzEyMsLrBrCQ3W5HT0/PikokOI5DaWkpu76SjXMlJSWszZvVal2TzVhlZWUoLy8HACQlJeHMmTPL/oH1/KHezKG4t7eXheGJiQncuXNnxf2lCQH+NNjHU//rqTcfHx/3CcODg4MQi8XU/owQErIoDEeYoqIiVFRUAACuXr2K+vr6oBw3JyeHNzmuo6Nj2ceLRCLe2OWbN29iZmYmKOfikZmZicOHD7PrNpttxStNm3mVuLu7G+np6TAajdDr9bRZjqyIWCxGamoq1Go1VCoVxGKxz2Pi4uJ8elRPT0+jtbV1PU+VEEL8QmE4wnjX5gWzt64nYANAa2vrA0NnZWUlFAoFAPcfw2CPQY6NjcXjjz/OyjaMRiM+++wzv44R6YE4Pj4e5eXlaGxs9FmRe1BJCSEeYrEYu3fvRlJSEvv3tpB3+zMqtSGEhBsKwxHGuxZ0qYEbsbGxSEhIQH9//4p7dn7xxRcYGxtDcXExGhsbl32sVCrFrl272PWampplyzX8JRaLceLECchkMgDusH369OlVvUYkBmKxWIySkhLk5uaC4zhUVlbi4sWLG31aJEws7P5is9kgl8t9gvDMzAwbfmE2m+kTBkJI2KIwHEGio6NZQLRarZicnFz0cUVFRdi7dy8cDgdqampQV1f3wGOPj49FDWwUAAAgAElEQVTj0qVLuHr16gNXhauqqiCRSACAjWwOpj/7sz9jod9ut+P06dOYnp5e9fEiKRBnZWWhtLSUff8BICYmxqd/KyEeQqEQKSkprPyhvb0der2e9xiDwYC8vDyMjo6yALzU7xdCCAk3FIYjiGezGrD8GGZPf2GhUIiJiQm/XuNBQTguLo5taAOAa9euBXXFaN++fbz2bufPn4fRaAz4uOG+sU6pVKKiosKnS4TRaERDQwONrSU8UqkUKpUKarUaKSkpvP6+arXaJwx3dHTg/v371P6MEBKRKAxHEO8wPDw8vOhjJBIJ1Go1APdwipUO21ipvXv38tqv6XS6oB17y5Yt2L17N7t+8+ZN3L17N2jHB8JvlVgkEqGwsBB5eXm8j7FnZ2fR1NS0opZ5ZHPwtD9Tq9VISEhY8nExMTHgOI73JvZBA3wIISScURiOICupF87OzmahaWho6IErPaWlpdDr9SvqBCGVSnlB8urVqys465URiUR4+OGH2fWurq6gHt9buKwSJyUlYefOnaw0BnAPGuno6EB7ezv1cCVMcnIyDhw4sOT9k5OTrPyB2uwRQjYbCsMRxHu1Z6kwrNFo2OWFH4UulJycjKNHj8LhcKC9vR3nzp1b9vEWiwVvvPEG9uzZg+joaBgMBj/Ofnl2ux2///3vceLECbhcLnz66adrvmEn1FeJ5+fnebXBw8PDaGhowNTU1AaeFdlInvZnBoOB92ZoZGQENpuNtUNzOp0YGRnB4OAgjEZj0NseEkJIOKEwHEHi4uLY5fHx8UUfk5mZyS53dXUte7yysjIA7tpi75rC5VgsFly6dGlFj/XX6OgoTp48CYlEgvn5+TV5jYVCeZV4amoKHR0dyM7ORnNzM/r6+jb6lMgGiI6OZuUPnvZnNTU1vDejTqcTAwMDEAqFMBgMGBoaWnEnGUIIiXQUhiOETCaDVCoF4O4ksdhKT3x8PGJiYgC4VxVNJtOSxxOLxSgsLGTXH9RObb3Mz8+vWxD2ttGhODU1FVKp1KfGu729Hffv36dgs8nEx8cjLS0NKpWKDcPxlpaW5vPJzO3bt9fr9AghJKxQGI4QHMfh9u3biI+PX7JWNCsri13u6+tbtsygsLCQ1x5tYGBgyceKRKKg9hH22Lp1K0wm07Khfb2td+mETCZDWVkZ0tPTYbfbYTKZeJuZHA4H1QZvEvHx8cjNzYVKpWJvfBczOjpKbfQIIcQPFIYjxOzs7APLE7zDcG9v77KP9ZRIAEBTU9OSjxOLxXjppZfQ09ODmpqagPr9esvOzsYjjzwCp9OJc+fO4d69e0E5bjCsxyoxx3HIy8tDUVERRCL3P1ORSITi4mLcunVrzV6XhK6YmJhFf+YcDgdMJhMMBgOMRiO1PyOEED9RGN4kBALBisNwWloaUlNTAbg3ri3XvmzHjh1QKpUoLS1FRkYG3nzzzYDPNT4+Ho899hgEAgEEAgEqKipw//79kJtw5QnFQHCDcVJSEioqKqBUKn1eL9gDTEho8bQ/S0xMxPXr13n3GY1GuFwucBwHi8UCo9EIg8EAk8lEnw4QQkgAKAxvEhzH4fLly8jJyUF8fPyy7ZMqKirY5bt37y650iSTybB9+3Z2/caNGwGfp0QiwYkTJ9jHwFNTU/j4449DLggvFIzVYolEgq1bt7KhKB4TExNoaGjAyMhIAGdIQhHHcUhKSmIb4ORyObtv4dRAq9WKxsZGjI+PU/szQggJIgrDm4TD4UBLS8sDVxalUilvwltDQ8OSj62qqkJUVBQAdyu3QAdgcByHxx57jLWIs9ls+Oijj8JqetpqQ7FGo0FxcTH7fgLuVfm2tjbodLqQfzNAVs7T/kytViM1NZX339xbWlqaT+3vg9ohEkII8R+F4QggEonw1FNPwWw2Y3h4GM3Nzas+VklJCatRNRgMS06yS0hI4NUVX716NeDAduDAAV6IPHfuXEhtnvOHPyUUAoEAeXl5vFDU39+P5uZmmvwVYcrKyqDRaHjTAr3ZbDYMDQ1hcHAQQ0ND63x2hBCyOVEYjgBJSUnIyMhARkYGRkZGAgrD3sFtuY1zBw4cYH/Qe3t7H9iz+EGKi4uxY8cOdr22thb3798P6Jih4kHB2Ol0oqGhAfv378f09DQaGhrC9k0A+ROpVOpTYjQ/P+8ThGdmZlj97/DwMH0KQAgh64zCcAR40Bhmf1qfnTp1CtnZ2SguLl6yg0NaWhq0Wi0AwOVy4fLly6s46z9Rq9V45JFH2PXOzk6fzUORwtOabWGLNpPJhBs3bsBgMMDpdG7cCZJVEwqFSElJgVqthkqlgsViwcWLF3mPGRwcRElJCcbGxtj0t4mJiQ06Y0IIIQCF4YiQnJzMLi9W1nDw4EFoNBr09fXh5s2bS5Y+AO5w293dzVvNXOihhx5il9vb25c93oNIpVI8+eSTrDRjeHgYZ8+eXfXxQlliYiKOHDmCjIwMfPzxx+jo6GD35eTkLNvLmYQmiUTCNr+lpKTwJjVKpVLIZDJeqcvU1BQ+/fRTan9GCCEhhMJwBHhQGM7KyoJCoUBRUdGypQ8rkZeXh7S0NADuDV7V1dUBHc9iseDWrVs4cOAALBYLPvroo4ibpiYWi1FVVYXKykoWlg4ePIju7m72tS588xGK45+Jm6f9mVqtZps9F2OxWKBQKHzqvikIE0JIaKEwHAG8yyQWhuG4uDjExcUBcLdmWjii1V8FBQXscmNjIyYnJwM6HgDcvHkTIyMjsNlsQTleKMnPz8fBgwfZGGzA3dmjvb192dpQCseha+/evYiOjl70vsnJSRgMBhgMBmp/RgghYYLCcJiTyWSsJ6/VavWZAOcdovr6+hatR/W0NOvs7ERHR8eyDfw//fRT3Lt3D9u3bw9KX2GPQDfghZq4uDgcOnQIubm5vNv7+/tx4cIFv3sGUzheX572Z/Pz8z5vMA0GA6uZdzqdGBkZYQF4ZmZmI06XEEJIACgMhznPqi8AjI+P+9zvHZqWqgPWarUoKChAQUEBzGYzfvOb3yz7mjqdDjqdblXnCwCpqakwm80ROTVLJBJh586d2LlzJ6uDBtwdA65cuRJwL2aPxf5bUkAOTHR0NCt/SEpKgkAggNFo9AnDAwMDkEgkrP1ZpJX1EELIZkNhOMwtF4ZFIhFvBPNSYdh74lwgIXclEhMT8cwzz2B4eBgff/xxWA3UWIm9e/fyWsS5XC40Njaiuroa8/Pza/raFJD9Fx8fj7S0NKhUKsTGxvrcn5yc7NONxWw2L9q1hRBCSHiiMBzm4uPj2eWF06rS09PZ6uTIyMiiLZwSExNZYHY6nWhsbPR5jFgsht1uD7j/qVQqxYkTJxAVFYX09HQ89thjeP/99wM6Zqi5efMmtm7dCqlUCqPRiAsXLmzo8ISl3gBt5pAcFRWFkpISqNVqVmK0mLGxMRgMBnAct45nRwghZL1RGA5z3mF44cpwZmYmu9zb27vo80tLS9nlzs5On5pjADh06BBUKhVqamp47cD8IRAI8MQTT/A28y3swRpuBAIBBAIBb9VwdnYWly5dgkgkQnNzc8gOUNjMIdlutyMjIwNisZh3u8PhgMlkYgMwqOsDIYRsDhSGw1x1dTU6OjqQkJCAwcFB3n3eYbivr8/nuSKRCMXFxez6YpPrEhMTUVxczMLs7373u1V1pDh48CDvfM6ePev3JrJQkp6ejiNHjqC/v98n1Le1tW3QWQVuuf7S4RSUPe3PVCoV7t+/z/uZdTqdMJlMSE9Ph8ViYeHXZDJFZB07IYSQ5VEYDnPj4+OLbpyLiopCamoqAHfd6mJhOD8/n31MPD4+jp6eHp/H7N+/n42P7e7uXlUQLi0t5dUlX7t2bc1rk9dKdHQ0Dhw4gJKSEgDuNwstLS2bYnzyckEZ2NiwzHEckpKS2AY4uVzO7lOr1T4/t/fu3UNHRwe1PyOEEEJhOFLJ5XIYjUaoVCqYTKZFN29VVlayy4utCmdlZfHGLl+5csXv88jIyMDhw4fZ9fb2dtTV1fl9nI3GcRxKS0uxf/9+Xp2pzWZDfHz8pgjDD/KgsOwRrNDsaX+mVquRmpqKqKioRR/neVPobbE3kIQQQjYnCsMRamxsDO+88w6kUilvlczD8xEy4K6hbGlp8XnM/v372eW2tja/d9DHxsbiiSeeYFPXjEYjPvvsM7+OEQpSUlLwyCOPsO+Xx/3793Hp0qVF66zJ0oIRmlNTU1FVVcU+tVjIZrNhaGgIBoMBRqNxFWdJCCFks6AwHMakUinm5+eX3aRlsVgW3QjkvXGuvb3dZ2Rsfn4+Lyz7O3ZZLBbjxIkTkMlkANx9dk+fPs3bbBbqJBIJ9u3bh/Lycl5HgbGxMVy8eHHRshISPJ7QrFKpfALt0NAQ9u7dy7ttcnISw8PDMBgMGB4eDtnNi4QQQkILheEwdvToUeTm5sJsNuOLL77wq57XarXCarUiKioKTU1NvPuEQiEOHDjArjc0NPi9+ulpxwa4w/Tp06fDagU1NjYWL7zwAm/srt1uR11dHerr62mj1Rry9MfWarXQaDSQy+U4efIkLxDPzc1hcHAQQqEQer0enZ2dAff+DacNgoQQQoKHwnAYS0lJgUgkgkql8nvF9dKlS6iuroZGo/FZdauoqGAt0Obm5lY1dnl2dhbvvfcejh49ip6enlVtvNtIExMTGB0dZWG4q6sLX3zxBdWarpHo6GhoNBpoNBpkZ2f7tD3TarU+P6e///3vg/qmZKXlG8uhQE0IIeGHwnCYkkgkbGKWw+HgtSnbv38/nE4n+vv7MTAwsGRgsNlsuHfvns9xd+/eza7X1NSsenKa3W7H2bNnV/XcUHDhwgWcOHECV65cQWdn50afTsSJi4tDfn4+tFot1Gr1ksMtZmdnF/0ZDsXV+WAE6sVQyCaEkLVDYThMJSYmsssjIyNwOp0A3IMgKioq2M76X/3qV5icnFzxcVUqFdvwNj4+7lNCEYny8vKwdetWnD59mn0fAff39c0336Ta0zVSUFDA26TpbWRkBDqdDjqdDkajcdP/N1irkL0YCt6EkM2GwnCY8qwKA/w2Ud4tpsbHx32CsEAg4AW+hXp6evDGG2+gqqoKPT09yz423MXFxeHQoUPIzc0F4C4PuX37Nu8xmz2EBSoqKgq5ublISUnB1atXeffpdDoWhp1OJwYGBqDX66HT6agcZQOtZ/BeDoVyQsh6oTAcprzD8MTEBLucnZ3NLvf39/OeI5fL8dJLL0Gn06GtrW3RQRyAu/PD559/HuQzDh1CoRC7du3Czp07IRL96Z9ARUUF7ty5QwE4QEqlEhqNBlqtFhkZGeyThoaGBkxNTbHHjYyMoKGhAQaDAV1dXTT+mPCESij3BwV4QsITheEw5dngBvBXhj2rnIDvH5Pi4mJIpVKUlJQgNjZ2yTAcybKzs3HkyBHe98/lcqGxsRHV1dUUhFdJpVKx7g/JycmLPkaj0aCxsZF328JR1oSEs3AM8MFCbwRIOKMwHKYWC8NSqZT1BnY6nejt7eU9x7u38MKJcwqFIqxan/lLoVDg4MGDKCgo4N1uNBpx4cIFDA0NbdCZhbe9e/di69atUCgUSz7GaDRCp9NRX2ZCIlgw3gisdrN2oK+5Hm9iNuJrIytHYThMJSQksMtjY2MA3KvCnh35BoOB97GzRqNhAdpisaCjo4Pdl5WVha985StoampCTU1NxH1cXVlZiX379vHG9VosFlRXV6OpqYlWg1eI4zif75VEIvEJwna7HX19fejs7ERXV1dEv8kihBAS/igMhyGZTAapVArAPTzDEzY0Gg17jF6v5z2nsrKSXW5ubmZ9iTmOw8MPPwyhUIjKykpwHBdxH13LZDJeEG5ra8OVK1cwOzu7gWcVHhITE1n9r8ViwYcffsi7X6fTobKyErOzs+jq6mIrwDabbYPOmBBCCPEPheEwpFAoMD8/D4lEgtHRUXZ7RkYGu9zV1cUux8bGso11TqcTDQ0N7L6ioiJW42m1Wlc1YCPU1dXVobCwEHa7HRcuXMDAwMBGn1LIEggESE9PZwHYuxzHbrdDLBbzgm5/fz/eeecdGAwGWmEnhBASligMh6Hh4WH89Kc/RUxMDFshjo+Ph1wuB+AuAfAewuFdK9zd3c129AuFQuzdu5fdd+vWLczMzAAAysvLsXXrVsTExMBsNuPWrVu8gB2KOI5DaWkpuru7eS3l7HY7Tp06hcnJyYhuFbdaUVFRyMnJgVarRW5uLvuZWkggEEClUvE2XjqdTgwODq7XqRJCCCFBR2E4jE1NTbFg670q3N/fz1bpBAIBSkpK2H3eG+fKy8uhVCoBuKd83bx5EwDw3e9+F8ePH+e1b3O5XPi///s/nDt3Dv39/SEXKlNSUnDkyBGo1Wro9Xqfj/Opb+3iOI7DK6+8smQAnp+fR09PD3Q6HbU/I4QQEpEoDEeI5uZmGAwGZGZm8oJfXl4eWzGenp5mtcRSqRR79uxhj6utrYXNZsNzzz2HF154AZcvX8Ynn3wCp9OJyspKPP300/j617+OsrIynDx5Erdu3YLVal3fL3IREokE+/btQ3l5Ods8qNFokJubG/Ir2etNpVLBarXySmtcLhf6+vqQn5/PbpucnGTDL/r7+0Ny7DEhhBASLBSGI4jZbIbZbObdtm3bNna5ubmZrRjv3r2brQZ6xi5nZmbi+eefx9WrV/GTn/wERqMRAHDjxg3U1tbiueeew759+5CUlIRf/OIX+OKLL9bpK1tcUVERHnroIRb2AXdJRH19/absobyQSCRCZmYm6/+rUCjQ0tKCzz77jPe4zs5OKJVKNv54eHh4g86YEEIIWX8UhsOMQqFAdnY2hoaGMDo6+sByhRs3bmDbtm1IT09nAw8UCgXKy8vZYy5fvgyn0wmFQoGkpCS0tLTAaDSC4zhwHMc23RkMBgwNDeHJJ5/E3/3d32F0dNRniMJ6SExMxOHDh5GZmcm7vbu7GxcvXtzUJRHR0dHIzc2FVqtFdnY2xGIx736NRuPTIu3u3bu4e/fuep8qIYQQEhIoDIeZzMxMfOlLXwLgXtE7ffr0so/v6upCV1cX5HI5ayW2f/9+NobYYDBAp9MBAGJiYgCABWyXy8ULTUNDQ/jFL34Bu92O5557Dv/yL/+CV199dd1WYcViMfbs2YNt27axEb+Au3b60qVLvN7Jm4lUKkVpaSm0Wi3UajUrF1nI0/4sKiqKGsATQggh/z8Kw2EmPj6eXfaURJSWlqK3txcTExNLPs/TJUIoFPKOcfXqVXZ5cHAQLpcLTz/9NG7duuUTLjmOw+zsLH7yk58gNjYWjz76KCoqKtYtDKempmLnzp3sutPpxO3bt1FTU7Op+9pyHIf9+/cvGoJHR0dZ+QO1PyOEEEJ8URgOM57VW8C9IiqXy3H06FEA7rG3J0+eXPb5DocDv/vd71BYWIi0tDT09/ez+/r7+/HRRx/hy1/+Mp588kn89re/5Y0pdrlcEAgEcDqdeP/99/HII4/g0KFDOHPmDCunWEv9/f24e/cuioqK0N/fjwsXLvBayEUy7/Znt27dgslkYvfNzc3BYDAgLS2NtTrzBODNXDJCCCGErASF4TDj3e5scnIS6enp7LpnqhwAaLVadHV1LRlQ29vb0d7ezq4LhUI4HA58+OGHyM/Px4kTJ2C1WvHuu+/ygpdnZdHTr3hyctKnnCIYPCvYCzcEXr58Gd3d3ZuixlWpVEKj0UCj0SAzM5OVhkxMTPD+mwBAfX09oqKiqP0ZIYQQ4icKw2HGOwxPTEzwukV4VnkzMjJw4sQJTE9Po76+Hnfu3HngcT3ts3Q6Hd577z1885vfxLPPPovk5GS8++670Ol0sFgsLPTu3LkTcrl8TVZms7OzcfjwYYjFYrz11lu8Fm6zs7MRHYRTU1Oh1Wqh1WrZZMCFtFotrl+/zrvNU/dNCCGEEP9QGA4jQqGQt8ltcnKS11HBU7vrmTinUCiQkJAAAEhISMDY2BhcLhdkMhnKysqwb98+zMzMYHZ2Fh9//DGmpqbgcDhw8eJFWCwW/OVf/iUOHTqEnJwcXLlyBdeuXUNHRwd27dqFZ555Bg6HA6dOnQra16dQKHDw4EEUFBSw2/bu3YtLly4F7TVCVW5uLo4ePQqFQrHkY4xGI+v/SwghhJDgoDAcRrw3vk1OTkKhUCAxMRGAu0TCYDBAIpHwBig0NTVBJBLhmWeewdzcHC5evIjvfe972L59O6RSKex2O7v/D3/4A65fv46Ojg5cv34dJpMJTz31FPbs2YOXXnoJf/EXfwGz2YyEhARYLBb827/9GwYGBgL+ujiOQ2VlJfbu3YuoqCh2+/z8PG9ARKSQSqU+pQxTU1M+Qdhut6Ovrw86nQ56vR7T09PreZqEEELIpkBhOIx4h+HR0VHk5uay6319fbDb7SgpKWFt04aGhjA8PIwdO3ZALpdDLpfjtddeg1qtxscff4zLly/D4XDg8ccfR1VVFf7qr/4KpaWl+OCDD1BbW4vOzk789Kc/xblz57Br1y7WyeHcuXO4desWbt26FfDXlJaWhiNHjviUBLS1teHKlSusHVy4S0hIYOUPKpUKr7/+OhulDbg7g0xMTCAqKgp6vR56vR7d3d2buksGIYQQsh4oDIcRT8kDAIyNjUGj0bDrntHDFRUV7LaWlhZIpVLs2rULgLseNT09HadOncKbb77JgmZLSwv279/PQnF8fDyEQiFqamowMzODxsZGNDY24q233oLD4QhK1wiZTIYDBw5g69atvNvNZjMuXrzI63IRjjiOQ3p6OgvAcXFxvPs1Go3PwJL3338fU1NT1P6MEEIIWUcUhsNIUlISuzw+Ps4Lvnq9Hjk5OaxsYn5+Hnfv3sX+/fvZ2GWHw4GoqCg0NjZidnaWN2Hu2rVrGBwcxPDwMB577DF84xvfQE9PD3p6eliniWCtUnIch+eff5630m2z2VBTU4Pbt2+veYu2teLd/iw3N5d93xdyOp28jZAek5OTa32KhBBCCFmAwnAYMRqNkEgkSE5OhkAgYK22TCYTJicncfjwYfbYlpYWyOVylJWVsdsGBwcBgJVRAOAFT71ej9/+9rcQCAR48skn8f3vfx8/+MEPgl6q4HK5UF9fj2PHjgEA7t+/j0uXLoV9Tey+fftQWVm56H1WqxXd3d3Q6XTU/owQQggJIRSGw4h3ne7evXvZ7QMDA5DL5cjJyWG3NTQ0YP/+/RAIBACA3t5edHR04LnnnsOLL76IlpYWnx6+gLvO+Fe/+hXy8/NRXl6O0tJS3LhxI6Dz9qwse2tpaUFmZibu3r2L7u7ugI6/3lJTU5GYmIi2tjbe7Xq9nheGp6am2PCL/v5+n+8BIYQQQjYeheEw1dPTA7FYjPT0dPT19aG4uJgF376+PkgkEmzZsoU9/tq1axgZGcGtW7ewfft2PPvss3jvvfd8AjHHcRgbG8Nbb72F//f//h92794dUBguLCzEQw89hI8//hgGg4F339mzZ1d93PUkFAqRlZUFrVYLjUYDhUIBu92Ojo4OXulIf38/+vv70dfXh87OTgwPD2/gWRNCCCFkJSgMh6mBgQFeW7MXX3yRXW5pacFDDz3Ernd0dGBoaAgulws/+9nP8M///M/48pe/DJvNhjNnzvBCqmfzltlshs1mW3Uda0JCAo4cOcL6IB85cgRvv/122GwOi46ORm5uLrRaLbKzsyEWi3n3i0QiZGdno7Ozk93mcDjw3nvvrfepEkIIISQAFIYjxPvvv4/S0lIUFxfD6XQiKysLgLsmuLq6moVQvV6Pd955By+99BJefPFFJCYm4tNPP0VLSwuvfrisrAwcx2F8fNyv8xCJRNizZw+2b9/OapoBd7iMjY31+3jrraKiAoWFhVCr1eA4btHHzM3NQa/X81qjEUIIISQ8URgOE56BFDMzM2htbfXZ1Ga1WllN8c6dO+FwOCAUCtHQ0MAbXGG32/Hpp5/CZrPhueeew2OPPYb8/HzU19fj9OnTsFgsqKqqwtNPP42+vj6cPn16xeeo1Wpx6NAhKJVKdpvT6cTt27dRW1vLG6scqjIzM5GWluZz++joKJv+Njg4GDYr3IQQQghZHoXhMFFYWMh61Q4MDCzb4aG+vh7t7e3YsWMHamtrefdxHAeXy4Xz58/DZDLh2LFjePjhh/Hiiy/iqaeegsPhgFQqxdDQEH74wx+uqM1ZbGwsDh06xOt7DLhraC9evLjoRr2N4ml/ptFoMDs7iytXrvDu1+l0yM/Ph9PpxODgIAvAY2NjG3TGhBBCCFlLFIbDRExMDLv83HPPYWBgAC0tLT4dDTympqbwxRdf+NzuvaLZ2NiIrq4unDp1CidOnEBKSgpiYmJw6dIl1NfXs1ZsyyksLMSxY8d47drm5uZw+fLlJc9tvcXExLDNb5mZmax8Y2ZmBlevXuV9T/R6Pc6ePUvtzwghhJBNgsJwGIiOjmYBzmazQSwWIyMjAxMTE0hISMC9e/dW3blgcnISk5OT+O///m8AgEAg8GvohclkYrW1LpcLTU1NqK6u3vAgmZqaygJwSkrKoo+Ry+VQqVS8DYQWiwV3795dr9MkhBBCyAajMBwGFtbgetjtduzatQu7du1Cb28v/vjHP65qcIV3APZ3+tvo6Chu3ryJnJwcXLhwAUaj0e/XDyaRSISXX36Zt5K+0NDQECt/MJlM63h2hBBCCAk1FIbDgPfYYu8ODZ6NXiqVChqNBjk5Oairq0NdXZ1fK7NOp3PRwRjeOI5DRUUFXC4XGhoaePfV1tbi+vXr676pTCaTAXCXZXjY7XZMTU3xwrDD4WC9f/V6fdhPuiOEEEJI8FAYDgOejXPAn0Ypz8/PIzk5GbGxsdi7dy8cDgdiY2MRHR2N/v5+6PX6Bx43JiYGR44cwblz53iBcqG0tDQcOXIEycnJsNls0Ol0vLZi6zlZLSEhgZU/pKWl4caNG5XBVowAAB4YSURBVLh+/TrvMTqdDvHx8dDr9dDr9eju7uYNxyCEEEII8aAwHAYSExN9brNYLJDJZCgtLWXXc3NzIRAIeKuinu4Ri3nqqafwyiuv4PDhw3j11Vd9Qq1MJsOBAwewdetWdptYLMaOHTsW3Zy3FjiOQ3p6OjQaDbRaLW+VHHC3c1sYhu/cuYObN29S+zNCCCGEPBCF4TCQnJzsc5tMJkNGRgZSUlIwODiI5uZmFBcXIysrC3/zN3+DkydP4sKFC3C5XEsG4hs3buCxxx7DmTNnfILw1q1bceDAAVaKALg379XW1uLWrVvB/yK9iMVi5OTksBVgqVS66OOcTiesVitEIhHsdju73fsyIYQQQshyKAyHOLFYzMokPMEWAKRSKVJSUmCz2dDU1ITZ2Vl89NFHiI+PxwsvvIDvfve7SEpKwgcffMCGXSzsFNHe3o6XXnqJVyKRkpKCI0eOQK1W886jo6MDly5dWpepa7GxsXjiiScWvc9qtaKnpwc6nQ5dXV3LlncQQgghhDwIheEQ5z0W2PP/nvZqra2tGBkZwezsLFwuFz7//HPMzc3BaDTim9/8Jl555RUkJSXh3XffhdlsZhvl9uzZg5s3b2J+fp6FSZFIhAMHDqC8vBwCgYC9/sTEBC5evIiurq6gf20pKSnIy8tDU1MTb1Ob2WzGxMQEYmNjAbh7Jut0Ouj1evT19a1rjTIhhBBCIhuF4RBnMBjwwQcf4OGHH0ZCQgIA92oxAMzOzrKQ2tTUxCa9nTlzBiaTCX/913+N559/HsnJyfjNb34DvV6Po0eP4gc/+AEuXryIf//3f2ev43K5kJOTw4Kw3W7HzZs3UVdXF7SyA6FQiKysLGg0Gmg0GlbbPDMzg8bGRt5j79y5A4lEQu3PCCGEELKmKAyHOJvNhq6uLmi1WshkMl4Nr2eleG5uDtXV1bznXL9+HSMjI/jOd76DI0eOIC4uDrdu3cLx48dhtVrxy1/+kvc6DocDFy5cwFe/+lV0d3fj4sWLGB8fD/j8ZTIZcnNzodVqkZ2djaioKJ/HaLVanzB8+/btgF+bEEIIIeRBKAyHic8//xyff/45nn32WWRkZPDuq6+v9+kr7HK50N7ejv/8z//E888/jxMnTqCyshICgQB///d/j6SkJAwNDfGe09vbi9/97ne8iWyrlZWVhaqqKqSlpbHQvtDc3Bz0ej06OzsDfj1CCCGEkNWgMBxm3nvvPWRlZeHRRx+FQqHAzMyMzxAMbwMDA/jwww9RVlYGjUaDX//61ygrK0NsbCysVis6Ojp4jw9GEAb+1BJtobGxMVb/OzAwQO3PCCGEELKhKAyHod7eXvzyl7/Eli1bwHHcsjW9UVFRePTRR6HRaNDQ0MAb7Xzo0CH09PSwbhP+iIqKQnZ2NrRaLbKysvDmm2/yBlv09/djfn4eYrEYBoMBOp0OOp0OY2Njfr8WIYQQQshaoTAcwjiOwyuvvAKbzQaLxYJ3332X3ecpg3iQhx9+GM8//zzMZjOvI8Tc3ByuX7/uVxBWKBTQaDTIy8tDZmYmbzR0dnY2r9zB4XDg9OnTMJvN1P6MEEIIISGLwnAI4ziOreRarVbI5XLMzMys+PlZWVmorKxEX18f7ty5A8AdopuamlBdXe1TZ7yYlJQUaLVaaLVapKSkLPm4zMxMn9rfvr6+FZ8rIYQQQshGoDAcwrw3nkVFReHP//zPcerUKQwPDy/7PLlcjoMHD2LLli2w2+24desWXC4XTCYTPv/8cxiNxhW9/rFjx3ijmBcaGhqCXq+n9meEEEIICVsUhkOY9/ALwB1yX3jhBbz33nvLBtqq/6+9ew+K6rrjAP7dFeS1u5YCImii+AJFqgKJMdPoCAaTotHx0RjTTExTNU5SMTFaxWeno05SrZnWBmKtJpnGOtaqaNQgo8ZWpcaggA9ApaA8VHxEluUl7J7+QfYWwoKAe2B37/czw8wu9+zvnvNz2fvb673njB6N0NBQ5XlNTQ3S09Nx/vx5mzeseXl5wdfXF6WlpU1+X1xc3KQYNpvNKCoqUgrgzliNjoiIiEgmFsMO7IfFMNCwUlxkZCQOHTrU4utOnTqFQYMGwcvLC5cvX8a//vUvVFVVNWnj6+urXP4QHByM6upqJCcnN2lTUFCAqqoqFBYWIj8/v8M32xERERE5KhbDDqzxDWpWFosF//73v5XnHh4esFgsTWZyqK6uRlpaGmpra5XrdjUaDYKDg5UC2NfXt0lcb29v9OrVq8kZZ1sFMhEREZEraX7q0YXcunULCQkJGDBgADw9PREYGIhJkybh6NGjjxXXaDRixYoVGDJkCLy9veHn54fY2Fjs3r3bTj1v4ObW/LtKeXm5chPdsGHDMHv2bDz33HPN2l27dg1FRUUYOHAgJkyYgHnz5uHll19GdHR0s0JYCIGSkhKbxTcRERE5FmevbxyNRrjoqgfZ2dmIiYnBvXv3AAAGgwEmkwkWiwUajQbr1q3D0qVL2x23uLgYY8aMUaYp0+l0qKmpUeb6nT9/Pj7++OPH7n9UVBSKiorw9ttvN9uWl5cHg8GAoKAgAA3F7M6dO20umDFz5kwEBwc3+/3Dhw9x/fp15Ofno6CggNOfERFRlzKZTBgxYgQyMjI6ZX9RUVHIzMyETqeTvi97js3Z6xtH5JJnhqurq/HSSy/h3r17GDlyJC5evIjy8nJ89913WLRoEYQQSExMxJEjR9oVVwiB6dOno6CgAP369cOpU6dQUVGBiooKfPjhh9BqtUhKSsJf/vIXu4zD1pnhuro6DBo0SCmEgYbxDh8+HBEREc3a5+fnK49NJhOysrKwZ88eJCUl4cCBA7h8+TILYSIiIifgKvWNo3HJYviTTz7B9evXodPpcODAAYSHhwNo+Pa0YcMGTJkyBUIILFu2rF1xU1JScObMGWi1WuzduxfPPvssAMDT0xOLFy/GggULAACrVq2ScqOZEALu7u7KjXUWiwW1tbXw9vbG0KFDMWLEiGavuXr1Kv7zn/9gx44d2LJlC44ePYrCwkKYzWa794+IiIjkcdX6pqu5ZDH8xRdfAABmzZqF3r17N9u+ePFiAMC5c+eQl5fX7rjjx4+3WXi+//770Gg0uHXrFo4dO9aRrreq8bzDQMNsEx4eHsrzgICAJsstA8CDBw9w+vTpNs8tTERERI7JVeubruZyxXBFRYVyTc6ECRNstnnmmWfQo0cPAGjXxebHjx9vNW7v3r2Vb2md+Wapra1Fbm4uDhw40GwKNSIiInJ+aqxvOovLTa2Wk5OjLCxh/Yf7Ia1Wi9DQUHzzzTe4fPlym+KWlZUpF6u3FBcAhg4diosXL7Y5bkd99913yM/Px3//+1+UlpbCYrFI3R8RERF1HbXUN13B5YrhxjMq2JpF4YfbbM3A0Jlxo6KibP4+KysLFosFSUlJMJvN6NatG8xmM2pra1FbW6vc3UlEROQKLBaLMpNBZ+7TZDJ1yn5ycnJaPOa3ZZYJZ6tvnInLFcPWOXiBhmWGW+Lt7Q0Abf4jkBW3JRaLBVqt1uY1QWQfOTk5AIAhQ4Z0cU9cF3PcOZhn+Zhj+bKyspoca2ULCwvrtH0BeOxC31XqG0fkcsWws2np26D122NnzbeoRsyxfMxx52Ce5WOO5WvprKks1pvGiFzuBjofHx/lcWvz51pvNGvrZNuy4hIRERE9CusbeVyuGG58vUtpaWmL7azbGi9e0RVxiYiIiB6F9Y08LlcMh4WFKfPxXrp0yWYbi8WizL83dOjQNsUNCAiAv79/q3EBKHdZtjUuERER0aOwvpHH5YphvV6P6OhoAEBaWprNNmfOnEF5eTkAIDY2ts2xx40b12rckpIS5Y3UnrhERERErWF9I4/LFcNAw8osQMPF8bamANmwYQOAhov1Q0ND2x33yJEjyMrKarb9D3/4A4QQCAoKUt5YRERERPbA+kYS4YKqqqpE3759BQARGRkpLl26JIQQwmg0isWLFwsAAoBITU1t9lrrttWrVzfbZrFYxKhRowQAERISItLT04UQQtTU1IgNGzYIrVYrAIgtW7ZIHR8RERGpD+sbOVxyajUvLy+kpKQgNjYW586dQ3h4OAwGA0wmEywWCzQaDdatW4e4uLh2xdVoNNi9ezfGjBmDgoICjB49GjqdDjU1NcoiGG+99RbmzJkjY1hERESkYqxv5HDJyyQAYPjw4bh48SIWLFiA/v37o7a2Fn5+foiPj0daWhqWLl3aobh9+vRBZmYmEhMTERYWhvr6euj1eowbNw67du1CUlKSnUdCRERE1ID1jf1phPh+oWsiIiIiIpVx2TPDRERERESPwmKYiIiIiFSLxTARERERqRaLYSIiIiJSLRbDkt26dQsJCQkYMGAAPD09ERgYiEmTJuHo0aOPFddoNGLFihUYMmQIvL294efnh9jYWOzevdtOPXcu9s7znTt38Mknn2DGjBlKTB8fHwwZMgTvvPMOrl27ZucROD5Z7+XGzGYzoqOjodFooNFosGbNGrvFdgYyc1xSUoKlS5ciIiICBoMBOp0OgwYNwqxZs5CSkmKH3jsHWTneu3cvJk6ciKCgILi7u0Ov12PEiBFYunQpbt++bafeO76Kigrs378fK1euxIsvvgh/f3/l7zk3N/ex4/PYR1J09UTHriwrK0v4+fkpE10bDAZl4mqNRiPWr1/fobhFRUUiJCREiavT6YSbm5vyfP78+XYeiWOTkefG+bTmuHv37spzT09PsWPHDgmjcUyy3ss/tGnTpiZ5tzU5vKuSmeNdu3YJvV6vxPb29hY+Pj7K89jYWDuOxHHJyLHZbBavvvpqk/etXq8X3bp1U57/+Mc/Ft98842EETmevXv3NslF45+cnJzHis1jH8nCYliSxqvEjBw5Uly8eFEIIUR5eblYtGiR8uFra5WY1jReJaZfv37i1KlTQgghqqurxYcffujyq8T8kKw8AxBjxowRn332mbh586YQQoj6+npx8uRJMWLECAFAuLm5iaysLLuPydHIyvEPFRUVCZ1OJ/r27SsCAwNVVQzLzPHBgweVwuyXv/ylyM3NVbbdvXtX/POf/xQbN26021gclawcJycnK8XYwoULxe3bt4UQQtTV1YmvvvpKPPnkkwKAGDBggDCbzXYfl6PZu3ev6Nmzp/jZz34mVq9eLbZs2WKXYpjHPpKJxbAk1jNcOp1OFBcXN9s+ZcoUZTnF9rB+69ZqteL8+fPNti9cuFAAEL169RK1tbUd7r+zkJXnEydOtLitrKxM9OzZUwAQs2fPbnefnY2sHLcUJyUlRSla1FIMy8pxeXm5CAoKEgBEYmKivbrrlGTleMyYMQKAiImJsbn9+PHjSjFo6zPb1dTX1zd5XlBQYJdimMc+konFsCTR0dECgJg7d67N7adOnVI+IBqfqXmU6dOnCwAiLi7O5vbi4mKh0WgEAHH48OEO9d2ZyMrzo8yePVsAEBEREXaL6ag6I8cpKSkCgJg4caIQQqiuGJaV448++kgAEH369BEPHz60V3edkqwch4aGCgBi0aJFNrcbjUYl7smTJzvUd2dmr2KYxz6SiTfQSVBRUYGMjAwAwIQJE2y2eeaZZ9CjRw8AaNeNG8ePH281bu/evREeHg4AOHbsWJvjOiOZeX4UPz8/AA03fLmyzshxZWUl3nnnHXh5eeFPf/pTxzvrpGTm+IsvvgAATJ8+He7u7o/ZU+clM8f9+vUDAJw/f97mdut+PTw8MHTo0DbHpaZ47COZWAxLkJOTA/H9KtfWP84f0mq1CA0NBQBcvny5TXHLyspw7969VuMCUD5w2xrXWcnKc1ucOHECADBs2DC7xXREnZHjlStXoqioCImJiUphoSayclxTU4PMzEwAwMiRI5Gbm4tXXnkFPXv2hKenJ/r374/58+ejsLDw8Qfh4GS+j+fMmQOgoQB79913UVZWBgCor69HamoqXn/9dQDAqlWr4Ovr2+ExqBmPfSQbi2EJbt68qTwODg5usZ11W+P2XRHXWXVVPlJSUvDtt98CAN544w27xHRUsnN8/vx5/PGPf8TgwYOxZMmSjnXSycnKcWFhIerq6gAAV65cQWRkJHbu3InKykq4u7ujoKAAycnJGD58OL7++uuOD8AJyHwfT5s2DWvXrkW3bt3w0UcfITAwEAaDAZ6ennjhhReg1+uxfft2JCYmdnwAKsdjH8nGYliCyspK5bGXl1eL7by9vQEAJpOpS+M6q67IR0lJCebOnQsAeOmll/DCCy88dkxHJjPHFosF8+bNg9lsxubNm9G9e/eOd9SJycrxgwcPlMfr16+Hr68vUlNTYTKZUFFRgVOnTmHw4MEwGo2YMWMG7t+/38EROD7ZnxXLli3Dp59+Ch8fHwANl2VYL6GqrKzE3bt3YbFY2ttt+h6PfSQbi2GiNjKZTJgyZQrKysrQt29f/PWvf+3qLjm1P//5zzh79ix+/vOf4/nnn+/q7ricxsWXxWLB559/jri4OGg0GgDAs88+i927d0Or1eLu3bvYunVrV3XVqVVUVGDSpEl47bXXMG7cOJw5cwZGoxGFhYVITk6G0WjE4sWL8eqrr3Z1V4moBSyGJbCeHQCA6urqFttVVVUBAHQ6XZfGdVadmY+amhpMnjwZ3377LQICApCamgp/f/8Ox3MWsnJcWlqKFStWQK/XY9OmTY/XSScnK8eN24WHhyM2NrZZm4iICIwfPx6AfW8wdTQyPyvee+89HDx4EOPHj8eBAwfw9NNPQ6/Xo2/fvpg3bx727NkDjUaDnTt34vDhwx0fhIrx2EeysRiWoPE1TaWlpS22s24LCgrq0rjOqrPy8fDhQ0yfPh3Hjh3Dj370Ixw5ckS50cbVycrxsmXLYDQasWTJEhgMBphMpiY/1pudHj58qPzOVXXG50Vr71frtqKiojbFdUaycmw0GrF9+3YAwMKFC222GTt2LCIjIwFAVcte2xOPfSQbi2EJwsLClP+KvHTpks02FosFeXl5ANDm6XYCAgKUs5EtxQX+fyetq0/jIyvPjdXX1+OVV17BwYMHodPpcOjQIYwYMaLjnXYysnJ8/fp1AA0zSej1+mY/N27cANBwrav1d65KVo79/f0RGBjY5n5Y++CKZOX46tWryrXBISEhLbbr378/AKhi5g4ZeOwj2VgMS6DX6xEdHQ0ASEtLs9nmzJkzKC8vBwCb/33ZknHjxrUat6SkRPmwaE9cZyQzz0DDwfH111/Hnj174OXlhf3792P06NGP12knIzvHJDfH1ksgrEWeLbm5uQDg0tPaycqxVvv/Q6j1C5wt1i9/rvylTjYe+0iqrlzxw5VZl/7U6/WitLS02fapU6cKACIqKqpdcRsvSZmZmdls+3vvvScAiKCgIFUsSSkrzxaLRbz55psCgOjevbuqVzSSlePWqG0FOlk5Tk1NVVb/SktLa7Y9OztbaLVaAUBs3ry5w/13BjJyXFlZKTw8PAQAMXXqVJttMjIyhJubmwAgfv/733e4/85KxnLMPPaRvbEYlqSqqko5oEdGRopLly4JIRqW5ly8eLHy4ZCamtrstdZttgoBi8UiRo0aJQCIkJAQkZ6eLoQQoqamRmzYsEE5sG3ZskXq+ByFrDwnJCQIAMLNzU3s27dP9jAcmqwct0ZtxbDMHD///PMCgAgODhZHjhwRFotFCCHE6dOnxeDBg5XPksrKSmnjcwSycjx37lxl+69+9Stx48YNIYQQ1dXVYt++feKJJ54QAITBYBB37tyROkZHcefOHeXn3LlzSn7S09ObbDObzU1ex2MfdRUWwxJlZmYKPz8/5Q/cYDAof7AajUasX7/e5usedXArKioSISEhSjudTqeceQAg3nrrLYmjcjz2zvP169eVbe7u7iIwMLDVHzWQ9V5uidqKYSHk5fju3bsiIiJCaeft7S30er3yPDg4WFy4cEHiyByHjBxXVFSIn/70p0obAMLHx0eJaz0bbavIdlWNc9HaT0FBgc3X8dhHnY3XDEs0fPhwXLx4EQsWLED//v1RW1sLPz8/xMfHIy0tDUuXLu1Q3D59+iAzMxOJiYkICwtDfX099Ho9xo0bh127diEpKcnOI3Fs9s5z4/lZ6+rqcPv27VZ/1EDWe5n+T1aO/fz8cPbsWXzwwQcYOXIktFot6uvrER4ejsTERGRnZ7v8suJWMnKs0+nw9ddfY9u2bYiLi0NAQABqa2vh5eWFYcOG4d1338WFCxcQFxcnYUTqwmMfyaIR4vs5jIiIiIiIVIZnhomIiIhItVgMExEREZFqsRgmIiIiItViMUxEREREqsVimIiIiIhUi8UwEREREakWi2EiIiIiUi0Ww0RERESkWiyGiYiIiEi1WAwTERERkWqxGCYiIiIi1WIxTERERESqxWKYiIiIiFSLxTARuayqqips3boVEydOxBNPPAEvLy90794d/v7+iIyMxGuvvYbNmzfjypUrzV67Zs0aaDQamz96vR4DBw7Eyy+/jH379kEI0QWjIyIie9AIfooTkQs6c+YMZs6cicLCwja1r66uhqenp/J8zZo1+O1vf9um144dOxYpKSno0aNHR7pKRERdyK2rO0BEZG9Xr15FXFwcjEYjACA+Ph4zZsxAaGgoPD09cf/+fWRnZ+P48eNIS0tDdXV1q/G2bduGp556CgAghEBxcTEyMjKwceNGPHjwACdOnMAvfvELHDhwQPrYiIjIvlgME5HLWb58uVIIb926FW+++WazNjExMVi4cCGMRiM+/fRTdOvWrcV4ISEhGDZsmPI8IiICL774It544w1ERUXh9u3b+PLLL5GRkYGoqCj7D4iIiKThNcNE5FLMZjO+/PJLAEB0dLTNQrgxg8GABQsWwN3dvd376t27N95++23leVpaWrtjEBFR12IxTEQu5c6dO8plDwMGDJC+v8jISOVxUVGR9P0REZF9sRgmIpfi4eGhPM7JyZG+v8aXV3Tk7DIREXUtFsNE5FJ8fX3Rr18/AEB2djbWrVsHi8UibX+XL19WHlv3S0REzoPFMBG5nISEBOXx8uXLERISgl//+tf4+9//jmvXrtltXuDKykokJSUBANzc3DB16lS7xCUios7DYpiIXE5CQgLmzp2rPL9x4wY2b96MWbNmYdCgQQgICMC0adOwa9cu1NfXtyu2dWq1ffv24amnnsK1a9cAAO+//z6efPJJu46DiIjk46IbROSyjh07hk2bNiE1NRV1dXU224SFhWHHjh0YOXJkk9+3ddENPz8/LFmyBEuWLLFLn4mIqHNxnmEiclkxMTGIiYmByWRCeno6zp49i4yMDJw4cQL37t0DAOTm5mLs2LFIT09HeHh4u/cxduxYzJkzx95dJyKiTsIzw0SkOvX19di/fz8SEhJQXFwMABg/fnyTeYIbnxluvAJdTU0NCgoKsG3bNnz11VcAgKioKJw8ebLJcs5EROQceM0wEamO9Wa3Q4cOoXv37gAaLqm4f/++zfbWFeiGDRuG6OhozJgxA4cPH1YujcjIyMBvfvObTus/ERHZD4thIlKtiIgIjBo1CgBgsViQn5/frtevXbsWP/nJTwAAH3/8Ma5cuWL3PhIRkVwsholI1YKDg5XHGo2mXa91c3PDunXrADRcerFq1Sq79o2IiORjMUxEqiWEQEZGBoCGQrgji2bEx8crSzL/4x//QF5enj27SEREkrEYJiKXYjKZ8PTTT2P//v0wm82ttl29erUyT/Bzzz0Hf3//Du1z5cqVABoutVi7dm2HYhARUdfg1GpE5HLOnj2LyZMnIygoCJMnT8bo0aMREhICg8GAiooKZGdn429/+xvS09MBAB4eHti4cWOH9zd58mRERETgwoUL2LFjB1avXo0BAwbYazhERCQRi2Eicilubm7o1asXbt26hZs3byI5ORnJyckttu/Tpw8+++wzREdHd3ifGo0Gy5cvx8yZM2E2m7F+/Xps3bq1w/GIiKjz8DIJInIpnp6eKC0txenTp/G73/0O8fHxGDhwIHx8fNCtWzfo9XoMGjQI06ZNw/bt25GXl4eYmJjH3u+MGTMQGhoKAPj8889x48aNx45JRETycdENIiIiIlItnhkmIiIiItViMUxEREREqsVimIiIiIhUi8UwEREREakWi2EiIiIiUi0Ww0RERESkWiyGiYiIiEi1WAwTERERkWqxGCYiIiIi1WIxTERERESqxWKYiIiIiFSLxTARERERqRaLYSIiIiJSLRbDRERERKRaLIaJiIiISLVYDBMRERGRarEYJiIiIiLVYjFMRERERKrFYpiIiIiIVOt/OXfovblxMnoAAAAASUVORK5CYII=\n"},"metadata":{"image/png":{"width":353,"height":327}}}],"source":["from gewitter_functions import get_pod,get_sr,csi_from_sr_and_pod\n","pod = get_pod(cont_table)\n","sr = get_sr(cont_table)\n","csi = csi_from_sr_and_pod(sr,pod)\n","\n","print('POD:{}, SR:{}, CSI:{}'.format(np.round(pod,2),np.round(sr,2),np.round(csi,2)))\n","\n","from gewitter_functions import make_performance_diagram_axis\n","ax = make_performance_diagram_axis()\n","ax.plot(sr,pod,'o',color='dodgerblue',markerfacecolor='w');"]},{"cell_type":"markdown","id":"7ef18845","metadata":{"id":"7ef18845"},"source":["Amazing, we have achieved 91% accuracy and a CSI > 0.8. These are really good results. \n","\n","One thing you might encounter while reading about other machine learning results is a *line* on this performance diagram. For example see this figure from [Lagerquist et al. 2020](https://journals.ametsoc.org/view/journals/mwre/148/7/mwrD190372.xml)\n","\n","\n","\n","\n","\n","Well, what we have been showing so far is the ```model.predict``` output, which is actually doing something for us under the hood. The ```model.predict``` line of code is actually running the ```model.predict_proba``` which outputs a probability of each class (i.e., does not contain thunderstorm, contains thunderstorm) and then ```model.predict``` is taking the max of the two columns of probabilities to assign the class. In other words, if the probability of class 0 (no thunderstorm) is less than 0.5, then the label would be 1, and if it was greater than 0.5 it would be 0. This is not a bad place to start, but sometimes we can get better performance by using a different *threshold*.\n","\n","So the line you see above comes from varying the threshold value for the label. \n","\n","\n","We can do this ourselves. Let's first look at ```model.predict_proba```"]},{"cell_type":"code","execution_count":10,"id":"e7dff173","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"e7dff173","executionInfo":{"status":"ok","timestamp":1648653632572,"user_tz":300,"elapsed":176,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"c17b01a4-6dc2-4c8a-fa5b-de95ff3981b6"},"outputs":[{"output_type":"stream","name":"stdout","text":["proba:[0.8269921 0.1730079],label:0\n"]}],"source":["#get predictions\n","yhat_proba = model.predict_proba(X_validate)\n","\n","print('proba:{},label:{}'.format(yhat_proba[0],yhat[0]))"]},{"cell_type":"markdown","id":"b82ce138","metadata":{"id":"b82ce138"},"source":["As you can see, the predicted probabilities for the first example shows a 90% prediction for no lightning while about 10% for the lightning. So a threshold of 0.5, which is assumed by the ```model.predict()```, shows that the label is indeed a 0 (or no lightning). But, say if our threshold was 95%, then the label would actually be a 1. \n","\n","So to create the line we will systematically change the threshold value between 0 and 1. Along the way, we will calculate the contingency table at each threshold value, and then calculate the POD and SR. Doing this results in the line.\n","\n","To actually implement this, we will use a ```for``` loop."]},{"cell_type":"code","execution_count":11,"id":"0a7682ed","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"0a7682ed","executionInfo":{"status":"ok","timestamp":1648653635713,"user_tz":300,"elapsed":175,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"fb735ec7-6497-44d4-cc36-3419bfc79a3e"},"outputs":[{"output_type":"stream","name":"stderr","text":["100%|██████████| 50/50 [00:00<00:00, 474.76it/s]\n"]}],"source":["#a fun progress bar to tack progress\n","import tqdm\n","\n","#lets just focus on the output from class 1 (note, the sum of these two columns should be 1)\n","y_preds = yhat_proba[:,1]\n"," \n","#chose you thresholds, this will make 50 evenly spaced points between 0 and 1. \n","threshs = np.linspace(0,1,50)\n","\n","#pre-allocate a vector full of 0s to fill with our results \n","pods = np.zeros(len(threshs))\n","srs = np.zeros(len(threshs))\n","csis = np.zeros(len(threshs))\n","\n","#for each threshold!\n","for i,t in enumerate(tqdm.tqdm(threshs)):\n"," #make a dummy binary array full of 0s\n"," y_preds_bi = np.zeros(y_preds.shape,dtype=int)\n"," \n"," #find where the prediction is greater than or equal to the threshold\n"," idx = np.where(y_preds >= t)\n"," #set those indices to 1\n"," y_preds_bi[idx] = 1\n"," #get the contingency table again \n"," table = get_contingency_table(y_validate,y_preds_bi)\n"," #calculate pod, sr and csi \n"," pods[i] = get_pod(table)\n"," srs[i] = get_sr(table)\n"," csis[i] = csi_from_sr_and_pod(sr,pod)"]},{"cell_type":"code","execution_count":12,"id":"41af202a","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":344},"id":"41af202a","executionInfo":{"status":"ok","timestamp":1648653639532,"user_tz":300,"elapsed":722,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"7a77c3f4-2a9d-430b-b3b4-c1aafba42fcf"},"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{"image/png":{"width":353,"height":327}}}],"source":["ax = make_performance_diagram_axis()\n","ax.plot(srs,pods,'-',color='dodgerblue',markerfacecolor='w',lw=2);"]},{"cell_type":"markdown","id":"baffbe4d","metadata":{"id":"baffbe4d"},"source":["Now you can see where the line comes from. You might be asking what the purpose of the line is? Well for some instances, your trained ML model might be biased (above or below the diagonal). Thus, this could be a way to achieve better performance without having to adjust the training of the model. All you would have to change (i.e., *calibrate*) is the threshold probability to something other than 0.5 to get better results.\n","\n","##### Additional Metric: AUC of the ROC curve\n","\n","Along the same lines as we just did, where we varied the threshold of prediction, there is another metric and diagram that is often used in ML for classification: the Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) curve. The ROC curve is plotted by calculating one of the same axes as the performance diagram, POD\n","\n","$$ \\mathrm{POD} = \\frac{TruePositives}{TruePositives + FalseNegatives} $$\n","\n","but now the x-axis is the probability of false detection\n","\n","$$ \\mathrm{POFD} = \\frac{FalsePositives}{FalsePositives + TrueNegatives} $$\n","\n","Since we need to get the area under the ROC curve (i.e., integrate), we have built a couple functions that handle this for you named ```get_points_in_roc_curve``` and ```get_area_under_roc_curve``` located in the ```gewitter_functions.py``` script."]},{"cell_type":"code","execution_count":13,"id":"b1904a5b","metadata":{"id":"b1904a5b","executionInfo":{"status":"ok","timestamp":1648653649654,"user_tz":300,"elapsed":497,"user":{"displayName":"Randy C","userId":"12301342565919601334"}}},"outputs":[],"source":["from gewitter_functions import get_points_in_roc_curve,get_area_under_roc_curve\n","\n","#lets just focus on the output from class 1 (note, the sum of these two columns should be 1)\n","y_preds = yhat_proba[:,1]\n"," \n","pofds, pods = get_points_in_roc_curve(forecast_probabilities=y_preds, observed_labels=y_validate, threshold_arg=np.linspace(0,1,100))"]},{"cell_type":"code","execution_count":15,"id":"62c6a279","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":386},"id":"62c6a279","executionInfo":{"status":"ok","timestamp":1648653668796,"user_tz":300,"elapsed":1454,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"1a36494f-03d5-47f7-97bc-e11c64702829"},"outputs":[{"output_type":"stream","name":"stdout","text":["AUC: 0.97\n"]},{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{"image/png":{"width":288,"height":351}}}],"source":["#something to help with annotating the figure\n","import matplotlib.patheffects as path_effects\n","pe = [path_effects.withStroke(linewidth=2,\n"," foreground=\"k\")]\n","pe2 = [path_effects.withStroke(linewidth=2,\n"," foreground=\"w\")]\n"," \n","#make figure\n","fig = plt.figure(figsize=(4.1,5))\n","#set facecolor to white so you can copy/paste the image somewhere \n","fig.set_facecolor('w')\n","\n","#plot the ROC curve\n","plt.plot(pofds,pods,color='dodgerblue',markerfacecolor='w',lw=2);\n","\n","#some annotations\n","plt.arrow(0.5, 0.55, -0.1, 0.1,facecolor='r',zorder=5,width=0.01,edgecolor='w')\n","plt.text(0.43,0.65,'Better',rotation=47.5,color='r',path_effects=pe2)\n","plt.text(0.02,0.95,'Best',color='w',path_effects=pe,zorder=5)\n","\n","#set some limits\n","plt.xlim([0,1])\n","plt.ylim([0,1])\n","\n","#set the no-skill line\n","plt.plot([0,1],[0,1],'--',color='Grey')\n","\n","#label things\n","plt.title(\"AUC of ROC Curve\")\n","plt.xlabel('POFD')\n","plt.ylabel('POD')\n","\n","plt.tight_layout()\n","\n","#print out the AUC value\n","print(\"AUC: {}\".format(np.round(get_area_under_roc_curve(pofds,pods),2)))"]},{"cell_type":"markdown","id":"88c0c95f","metadata":{"id":"88c0c95f"},"source":["A perfect performing model will have an AUC of 1. Since the AUC with this model is 0.97, that is really good. Visually to inspect this diagram, the closer the line is found in the top left corner, the better the model is performing. Meanwhile if your line is near the diagonal, then you model has little to no skill (AUC 0.5)."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"W2Mzj7LNZIXE"},"outputs":[],"source":[""],"id":"W2Mzj7LNZIXE"}],"metadata":{"kernelspec":{"display_name":"Python 3 (ipykernel)","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.9.7"},"colab":{"name":"Notebook06_ComplexMLClassification.ipynb","provenance":[]}},"nbformat":4,"nbformat_minor":5}