{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.11.11","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":96416,"databundleVersionId":11456366,"sourceType":"competition"}],"dockerImageVersionId":31040,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nimport matplotlib as mpl\nimport matplotlib.pyplot as plt\nimport seaborn as sns","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:42.718822Z","iopub.execute_input":"2025-06-18T07:54:42.719123Z","iopub.status.idle":"2025-06-18T07:54:44.021146Z","shell.execute_reply.started":"2025-06-18T07:54:42.719100Z","shell.execute_reply":"2025-06-18T07:54:44.020111Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# **Import Data** ","metadata":{}},{"cell_type":"code","source":"train_data = pd.read_csv('/kaggle/input/digit-recognizer-challenge/train.csv')\ntest_data = pd.read_csv('/kaggle/input/digit-recognizer-challenge/test.csv')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:44.022388Z","iopub.execute_input":"2025-06-18T07:54:44.022809Z","iopub.status.idle":"2025-06-18T07:54:50.317193Z","shell.execute_reply.started":"2025-06-18T07:54:44.022776Z","shell.execute_reply":"2025-06-18T07:54:50.316274Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Separate features and labels\ny_train = train_data['label']\nx_train = train_data.drop('label', axis=1)  # Still a DataFrame here","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:50.318097Z","iopub.execute_input":"2025-06-18T07:54:50.318691Z","iopub.status.idle":"2025-06-18T07:54:50.440013Z","shell.execute_reply.started":"2025-06-18T07:54:50.318659Z","shell.execute_reply":"2025-06-18T07:54:50.438703Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"fig, axis = plt.subplots(1, 10, figsize=(10, 5))\nfor i, ax in enumerate(axis):\n    img = train_data.iloc[i, 1:].values.reshape(28, 28)\n    ax.imshow(img, cmap=\"gray\")\n    ax.set_title(f\"Label: {train_data.iloc[i, 0]}\")\n    ax.axis(\"off\")\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:50.442003Z","iopub.execute_input":"2025-06-18T07:54:50.442323Z","iopub.status.idle":"2025-06-18T07:54:51.066980Z","shell.execute_reply.started":"2025-06-18T07:54:50.442301Z","shell.execute_reply":"2025-06-18T07:54:51.066058Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"x = train_data.values[:7000, 1:]  # Use .values to get the numpy array, skip label column\ny = train_data.values[:7000, 0]\n\nsome_digit = x[0]\nsome_digit_image = some_digit.reshape(28, 28)\n\nplt.imshow(some_digit_image, cmap=\"gray\")\nplt.axis(\"off\")\nplt.show()\n\ny[0]\ny=y.astype(np.uint8)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:51.067787Z","iopub.execute_input":"2025-06-18T07:54:51.068120Z","iopub.status.idle":"2025-06-18T07:54:51.155053Z","shell.execute_reply.started":"2025-06-18T07:54:51.068088Z","shell.execute_reply":"2025-06-18T07:54:51.154044Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"x_train, x_test, y_train, y_test = x[:6000],x[6000:],y[:6000],y[6000:] ","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:51.155950Z","iopub.execute_input":"2025-06-18T07:54:51.156239Z","iopub.status.idle":"2025-06-18T07:54:51.161988Z","shell.execute_reply.started":"2025-06-18T07:54:51.156219Z","shell.execute_reply":"2025-06-18T07:54:51.161077Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# Training binary classification","metadata":{}},{"cell_type":"code","source":"y_train_1=(y_train==1)\ny_test_1=(y_test==1)\nprint(y_test_1)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:51.163141Z","iopub.execute_input":"2025-06-18T07:54:51.163451Z","iopub.status.idle":"2025-06-18T07:54:51.184485Z","shell.execute_reply.started":"2025-06-18T07:54:51.163424Z","shell.execute_reply":"2025-06-18T07:54:51.183562Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from sklearn.linear_model import SGDClassifier","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:51.185360Z","iopub.execute_input":"2025-06-18T07:54:51.185656Z","iopub.status.idle":"2025-06-18T07:54:51.540375Z","shell.execute_reply.started":"2025-06-18T07:54:51.185625Z","shell.execute_reply":"2025-06-18T07:54:51.539657Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"sgd_clf = SGDClassifier(random_state =42)\nsgd_clf.fit(x_train,y_train_1)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:51.541337Z","iopub.execute_input":"2025-06-18T07:54:51.541656Z","iopub.status.idle":"2025-06-18T07:54:51.969341Z","shell.execute_reply.started":"2025-06-18T07:54:51.541625Z","shell.execute_reply":"2025-06-18T07:54:51.968608Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"sgd_clf.predict([some_digit])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:51.972300Z","iopub.execute_input":"2025-06-18T07:54:51.972529Z","iopub.status.idle":"2025-06-18T07:54:51.978477Z","shell.execute_reply.started":"2025-06-18T07:54:51.972513Z","shell.execute_reply":"2025-06-18T07:54:51.977792Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"classifier guess: image 1 is (true)","metadata":{}},{"cell_type":"markdown","source":"# Performance Measures","metadata":{}},{"cell_type":"markdown","source":"### 1. Cross-Validation: Measuring Accuracy \n#### Implementing Cross-Validation","metadata":{}},{"cell_type":"code","source":"from sklearn.model_selection import StratifiedKFold\nfrom sklearn.base import clone","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:51.979266Z","iopub.execute_input":"2025-06-18T07:54:51.979476Z","iopub.status.idle":"2025-06-18T07:54:51.997289Z","shell.execute_reply.started":"2025-06-18T07:54:51.979459Z","shell.execute_reply":"2025-06-18T07:54:51.996445Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"skfolds = StratifiedKFold(n_splits=3, shuffle=True, random_state=42)\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:51.998143Z","iopub.execute_input":"2025-06-18T07:54:51.998475Z","iopub.status.idle":"2025-06-18T07:54:52.016785Z","shell.execute_reply.started":"2025-06-18T07:54:51.998444Z","shell.execute_reply":"2025-06-18T07:54:52.015813Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"for train_index, test_index in skfolds.split(x_train, y_train_1):\n    clone_clf = clone(sgd_clf)\n    \n    x_train_folds = x_train[train_index]\n    y_train_folds = y_train_1[train_index]\n    x_test_fold = x_train[test_index]\n    y_test_fold = y_train_1[test_index]\n\n    clone_clf.fit(x_train_folds, y_train_folds)\n    y_pred=clone_clf.predict(x_test_fold)\n    n_correct=sum(y_pred == y_test_fold)\n    \n    print(n_correct/len(y_pred))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:52.017854Z","iopub.execute_input":"2025-06-18T07:54:52.018180Z","iopub.status.idle":"2025-06-18T07:54:52.775203Z","shell.execute_reply.started":"2025-06-18T07:54:52.018149Z","shell.execute_reply":"2025-06-18T07:54:52.774409Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from sklearn.model_selection import cross_val_score\ncross_val_score(sgd_clf,x_train,y_train_1,cv=3, scoring=\"accuracy\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:52.776047Z","iopub.execute_input":"2025-06-18T07:54:52.776360Z","iopub.status.idle":"2025-06-18T07:54:53.416198Z","shell.execute_reply.started":"2025-06-18T07:54:52.776339Z","shell.execute_reply":"2025-06-18T07:54:53.415315Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"98% accuracy ","metadata":{}},{"cell_type":"code","source":"from sklearn.base import BaseEstimator\n\nclass Never1Classifier(BaseEstimator):\n    def fit(self,x,y=None):\n        pass\n    def predict(self,x):\n        return np.zeros((len(x),1),dtype=bool)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:53.417394Z","iopub.execute_input":"2025-06-18T07:54:53.417606Z","iopub.status.idle":"2025-06-18T07:54:53.422205Z","shell.execute_reply.started":"2025-06-18T07:54:53.417590Z","shell.execute_reply":"2025-06-18T07:54:53.421441Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"never_1_clf=Never1Classifier()\ncross_val_score(never_1_clf,x_train,y_train_1,cv=3,scoring=\"accuracy\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:53.423022Z","iopub.execute_input":"2025-06-18T07:54:53.423319Z","iopub.status.idle":"2025-06-18T07:54:53.487155Z","shell.execute_reply.started":"2025-06-18T07:54:53.423294Z","shell.execute_reply":"2025-06-18T07:54:53.486318Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Accuracy level is 89%","metadata":{}},{"cell_type":"markdown","source":"### 2. Confusion Matrix: much better way to evaluate the performance of classifier\n The general idea is to count the number of times instances of class A are\n classified as class B.","metadata":{}},{"cell_type":"code","source":"#cross_val_predict() performs K-fold cross-validation, \n#but instead of returning the evaluation scores,\n# it returns the predictions made on each test fold\nfrom sklearn.model_selection import cross_val_predict\ny_train_pred = cross_val_predict(sgd_clf, x_train, y_train_1, cv=3)\n\nfrom sklearn.metrics import confusion_matrix\nconfusion_matrix(y_train_1, y_train_pred)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:53.488024Z","iopub.execute_input":"2025-06-18T07:54:53.488340Z","iopub.status.idle":"2025-06-18T07:54:54.122760Z","shell.execute_reply.started":"2025-06-18T07:54:53.488319Z","shell.execute_reply":"2025-06-18T07:54:54.121989Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"* The first row of this matrix considers non-1 images (the negative class): 5298 of them were correctly classified as non-1s (they are called true negatives)\n* 48 were wrongly classified as 1s(fasle positive)\n* The second row considers the images of 1s (the positive class): 29 were wrongly classified as non-1s (false negatives)\n* 625 were correctly classified as 1s (true positives).\n  \n\n","metadata":{}},{"cell_type":"code","source":"y_train_perfect_predictions = y_train_1\nconfusion_matrix(y_train_1, y_train_perfect_predictions)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:54.123811Z","iopub.execute_input":"2025-06-18T07:54:54.124089Z","iopub.status.idle":"2025-06-18T07:54:54.131097Z","shell.execute_reply.started":"2025-06-18T07:54:54.124071Z","shell.execute_reply":"2025-06-18T07:54:54.130229Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Accuracy of the positive predictions; this is called the precision of the classifier\nEquation for Precision\n **precision = TP/TP+FP**\n * TP is the number of true positives,\n * and FP is the number of false positives.\n * ensure it is correct (precision = 1/1 = 100%).\n *  So precision is typically used along with another metric named recall, also called sensitivity or true positive rate (TPR): this is the ratio of positive instances that are correctly detected by the classifier\n1.  Equation Recall\n **recall = TP / TP+FN.**\n* FN the number of false negatives.\n   confusion matrix,figure example\n * ![image.png](attachment:32a0696d-60d1-46c5-a2a7-8c97e2ec1685.png)","metadata":{},"attachments":{"32a0696d-60d1-46c5-a2a7-8c97e2ec1685.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"### 3.Precision and Recall","metadata":{}},{"cell_type":"code","source":"from sklearn.metrics import precision_score, recall_score\n\nprecision_score(y_train_1, y_train_pred)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:54.132023Z","iopub.execute_input":"2025-06-18T07:54:54.132353Z","iopub.status.idle":"2025-06-18T07:54:54.156858Z","shell.execute_reply.started":"2025-06-18T07:54:54.132327Z","shell.execute_reply":"2025-06-18T07:54:54.155971Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"recall_score(y_train_1, y_train_pred)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:54.157774Z","iopub.execute_input":"2025-06-18T07:54:54.158087Z","iopub.status.idle":"2025-06-18T07:54:54.176499Z","shell.execute_reply.started":"2025-06-18T07:54:54.158060Z","shell.execute_reply":"2025-06-18T07:54:54.175437Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Image represents a 1, it is correct only 92.8% of the time. More\nover, it only detects 95.5% of the 1s.\n","metadata":{}},{"cell_type":"markdown","source":"####  F1 score:  to combine precision and recall into a single metric called the F1 score\n\n Equation ![image.png](attachment:907fb4e0-79f5-48ca-95ec-e85e34b69ceb.png)","metadata":{},"attachments":{"907fb4e0-79f5-48ca-95ec-e85e34b69ceb.png":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAdYAAABDCAYAAAAhxiQeAAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAAAJcEhZcwAAEnQAABJ0Ad5mH3gAACW/SURBVHhe7d17XI73/wfwV0dJWWtpxHIqh1XSjPHVMMkwGqNQSrUxGwpTxjaHzXHlUBaSTlrLoZAyzWEiRtPJKRFF6yQqUXd1d9+9f3/87j6P7quDbKH0eT4eHruvz/tz3dd9Xbvv3vf9uT4HBSIicBzHcRzXLBSFBRzHcRzH/Xs8sXIcx3FcM+KJleM4juOaEU+sHMdxHNeMeGLlOI7juGbEEyvHcRzHNSOeWDmO4ziuGfHEynEcx3HNiCdWjuM4jmtGPLFyHMdxXDPiiZXjuFYpPT0dnp6emDRpEqRSKUpKSrBr1y7Mnj0b8fHxwuptxtOnT1FaWiosZqRSKQoKCoTFXDPiifUV+f333/Hdd99h9+7dqKysFIa5Niw8PBz79u0TFtfx+PFjLFu2DIWFhcJQm6Cjo4OCggJER0cDANTU1KCjo4O9e/c2mlhedyEhITAxMcHcuXOxaNEifPjhh7CyssKiRYvg7OyMnj174uzZswCAyMhIjB07FmZmZrCzs8OSJUtgb2+PU6dOCZ+Wex7EvXRffvklTZkyhWbPnk3q6uo0ePBgEovFwmpcG2VjY0NTpkwRFteRkpJCHTp0oISEBGGozTh69CgBIIlEQkRET548IQB06tQpYdU2w8bGhpKSkti2trY2+fj4sG07Ozu594yXlxdpa2uza3jmzBlSVFSkqKgoVod7PvwX60sWGRkJCwsLHDp0CEFBQYiIiMDly5dx/PhxYVWujdq3bx8iIiKExXWYmpqipKQEgwYNEobaDEVF+T9hwu226H//+x/MzMwAAFlZWSgqKsLAgQNZfNy4cejZsyfbvnv3LgYNGgQlJSUAwKhRo2BsbIzQ0FBWh3s+/F34kk2aNAnW1tZs29LSEkpKSsjJyZGrx716ubm58PPzw5UrV1BQUABfX19ERUVBKpWyOiKRCFFRUcjMzERqaiqCgoLk4qmpqdixYwdOnz7NympUVFQgPDwc27ZtQ3JyMitPS0tDZGSkXN0rV67Ay8sLAQEBePLkCQCgtLQUx48fx7179+Tq5uXlITAwEAEBAcjMzJSLiUQiREdHIzMzEyUlJQgODq5zrNomT56MK1euCIuxceNGbNu2TVgMqVSKCxcu4Pz58ygsLISvry8ePXrE4vn5+QgKCkJYWBjEYrHcvgDw559/Ytu2bXW+aGZlZcHPzw+3b9+WK+fqcnV1ZY+TkpKgoKAAExMTVjZr1ixoa2vL1amdeCF7b7Vr106ujHsOwp+w3MtVUFBAAOjvv/8WhrhXSCQSkYODAwGgmTNn0ldffUX29vakoqJCH330EUkkEqqoqKCpU6cSAPL29qbevXsTALp69SoREW3cuJE2btxI8fHxZGZmRsuXL2fPHxsbS/PmzaOAgACaN28eaWtrk1QqpUOHDpGOjg6Zm5uzur/99htZW1tTeno62dvbU0BAAIlEIpo4cWKdZs/9+/fT6NGj6erVq3T8+HHq1q0b7dy5k4iIqqqq6NNPPyUA9NVXX9GcOXPoyy+/JBUVFfrpp5/Yc9SWnJxMffr0oWvXrrGyzZs30+jRo0kkEsnVJdk5Kykp0ZIlS2jkyJGkrKxM27dvJyKiEydO0Jw5cyghIYEcHBzI3NycNT/m5OSQk5MT+fr60o8//khaWlqUmJhIJSUl9MUXX9D69etp2bJlpKioSH/99Rc7XnR0tFxTcGlpaZ1r0patXLmSDA0NhcWMVColDQ0NCg0NZWUpKSkEgMLDw+Xqck3HE+sr5unpSRMmTBAWcy2ARCIhAHJ/dCIiIggAhYSEEBFRUVERASAbGxuSSqWUnZ1NRESnT5+mAQMGsP12795NioqK9OTJE7p16xaNGjWKqqurWbykpIQ9dnV1lUusFhYW9MMPPxDJ7iEePHiQiIiKi4vlksidO3dIXV2dEhMT2b6hoaGkrKzMEmPN6/31119ZndmzZ5OxsTHbFkpISKA+ffpQamoqeXt704cffkilpaXCasyAAQPIwMCAiouLKT8/n6qqqujhw4fUpUsXysvLIyKi27dvEwA6duwYiUQiMjc3p5ycHPYcNdcjICCAxowZw8r79OlDbm5ubJsn1sZNnDiRrK2thcVMamoqAaDo6GgqLCykw4cPU69evei7774TVuWeA28KfoXu3LmDkJAQBAYGCkNcC1Bzz6lDhw6s7LPPPkPnzp2RmJgIAFBVVQUAODs7Q1FREV27dgUA+Pj4QCwWw87ODnPmzEFsbCycnJwAAN7e3hgyZAgUFBTY83bs2JE9VlFRYY8BYODAgfDw8MDKlSuhpKSEadOm1VsvKioKEomE3V8DACsrK0gkEtbcW/N6O3fuzOoMGjRIrrlWaNCgQdi7dy8sLS3x66+/4tixY3LXREhVVRXTpk2DlpYW3n77bSgrK7Om38WLF+Pzzz/H+vXrYWtri65du+LYsWMQi8XQ09Njz1FzPZycnHDixAlA9nlRVVVFcXExq8c1Ljk5uU4zb22JiYlQV1dHeno6vvnmG9y8eRPHjh3D2rVrhVW558AT6ytSWVmJRYsW4eDBg9DV1RWGuRasU6dO6N69u1yZsrKy3HZqaiomT56M0NBQ+Pn5ITQ0FHv27IGmpibi4+NRVVUlV78x69atg7u7Ozw8PNCvXz/89ddfwioAgFu3bgmLoKGhAQ0NDTx8+FAYYmqSbWNu3ryJTp06obi4GA8ePBCG66jvehgaGiIsLAz+/v4IDAxEaGgoTE1Nn3k9jhw5AkdHR5w7d06u0w3XuIKCAuTk5DwzsZqammLRokUIDAzE8uXL0a9fP2E17jnxxPoKVFVVYd68edi4cSMMDQ0BABkZGaiurhZW5VoYsViM7OxsWFpaCkNydHV1cebMGWExAEBbWxuxsbFN+v8tkUiQlpaGNWvW4OrVq2jfvj3c3NyE1QAAvXr1glgsluuwVFVVhbKyMpiamsrVfR5hYWHYuXMnzpw5Az8/P0ycOLFOh6ln0dXVxbVr1+r9ZaytrY3U1NR6E3ZISAicnJywfft2ODs78w41z6GmQ1xj/+8TExPlWji45sET60smlUrh4OCA999/H2VlZYiPj0dUVBScnJz4UIEWqqYXLgBs2bIFS5cuhZGRESBLXJAlwNrs7e0RHx+P7777DuXl5SguLsbPP/8MIoKDgwOSk5Ph4uKC0tJSiEQieHp6Ij8/nz1nzfMpKirC3t4eUqkUhoaGmD59Ot544w1WD7WO7ejoCG1tbezcuZO9jnPnzuGdd97B1KlT5fap3XNZLBbLbdd26NAhbNmyBcePH4eWlhZGjhwJHx8fTJgwAf/884+wOiB4/TVmzpyJyspKODg44MGDB6isrERAQADu37+P6dOnAwBsbGxw//59SCQSHDp0CLGxsUhLS0NFRQVEIhHEYjHy8vKQnZ3NXm/N+TT037bs4sWLUFdXZ7cnhMrLy5GQkIC+ffsKQ9x/Jbzp2hBPT08aMmQIDRo0iL7++mtydXUlZ2dnMjMz4x0FnsPixYsJQJ1/Dg4OwqpcLcnJyfTTTz/Rpk2bKDc3Vxh+YQBQhw4dyMrKipydnWn//v0slp+fT3Z2dgSAzM3NKTo6msWkUiktXbqUlJWVSUlJifr27Uvx8fEsvmbNGlJVVSVFRUUyMDCg06dPE8k6G3Xt2pXU1NRo9erVREQ0fvx4mjRpEnl5edHo0aMpNTWVcnNz2bFHjBjBnjshIYGMjIxo/vz55OPjQ5MnT2a9lO/fv0+2trYEgMaMGUM3b96kuLg4MjY2JgByvZZruLu706NHj4TF9Mcff8h1gKqxatUqUlFRIX19ffL29paLhYWFUceOHQkA6ejoUGBgIItFRkaSjo4OASBdXV3avHkzERFlZGSQnp4eaWpqkqOjIy1fvpwUFBTIxcWF4uLiaOjQoQSAZs+eTVlZWTRz5kx2TS5evFjr6G3HgwcPyM3NjbS0tAgAzZs3j1JTU+Xq/P777zRmzBgCQEZGRnITSHD/nQL9/x+PJvnkk09QVVXFOhNA1mlj7ty5GDp0qFxdjmsuwcHBCAgIQL9+/XD8+HGUlpYiMTHxpdxvU1BQQHh4ON57771/dTyRSIT8/Hz06tVLGEJlZSVycnLqjQnl5ORAJBKhd+/eTWrZyMvLg1QqRbdu3YShV0oqlSIjIwM9e/ascx+WiJCRkYHu3bvLxSoqKvDw4UO88847gOzeIe+XwLVkz/6E1pKRkYHBgwfLlY0dO7ZJfxg47t/Izc3FzZs3cfbsWfj6+iIhIQESiQTbt28XVn1hlJWV/1VSBQB1dfUGPx/t2rVrMCbUtWtXGBoaNimpAkCXLl1aXFKFrKe1oaFhnaQK2ZeY3r1714mpqamxpArZ/VqOa8ma9imVffO+fft2nRvdM2bM4G907oXR09PDhg0b2Lauri7ee++9Bmeqevz4MSwsLPD48WO58qqqKkyePBlJSUly5Y2puU9XUVEhDHEcxzWoyU3Bly5dwrBhw3D37l32Ldvd3R0rVqyAlpaWsHqjbt26VadzQ239+/dv8jdzrnUoLi5Gbm6usJh566235MZWNsbIyAiOjo4N9o4NDg7Grl278Mcff6Bjx46QSCSYPn06+vbti/Xr1wur1ys3Nxfu7u4IDQ2FiYkJ3NzcYG9vL6zGvSaa8/3JcU1OrDt37oSLiws+++wzAMD9+/dRWFiI9PR0YdVnmj9/PoqKioTFjL+/P9TV1YXFXCt25swZ7N69W1jMjB8/Hg4ODsLiOhITE/Hpp5/i+vXrjX6h8/PzQ3BwMKKjo/Hll1+iW7du2Lx5s7AaxwHN+P7kODxPYp07dy7u3r3LJhMvKyvD3LlzX+oKCLGxsYiNjRUWcy2Io6MjevToISxuFpWVlRg1ahQ8PDxgbm4uDNfxyy+/YOXKlZg+fbrcEJRn4e+z10ePHj3g6OgoLG5Wq1evFhZxzay1XeMmJ9bBgwfjww8/xJYtW1hZTk5Og2OkGlN7HFp99PX15aZ7q3Hv3r3nHpjOvVwDBw6s95dkaWlpowtyd+zYEW+++aawWM6iRYswcuRITJkyRRiqg4jwxRdfIC0tDSoqKvj999+b3ArC32evDy0trUZnHqrxX96f/EvYizdq1ChhUYvWpMQqkUigqamJXbt2Yfbs2cIwU15ejj///BOjR49G+/bthWHG0dGx0Tfxvn37Gp2LlGt9Tpw40WhPXisrK8yZM0dYzHh4eEBPTw92dnYAwG4l1F7+qrb58+dDLBZj9+7d8PT0xIkTJxAVFQU1NTVhVa6VSk9PR2RkJM6ePYsjR46gtLQUYWFhuHjxIr7++mt88MEHwl0a9F/fnxxXW5MS6+XLlzFkyBCcP38ew4cPF4YBWWePVatWYc+ePXj06BHeeustYZVWJT09HYWFhW1mfG5LPt9ffvkFV65cwRdffAHIfl1s3boV27Ztg4GBgbA6lixZguLiYgQEBLCWj/Xr1+P8+fM4cuRIk+bGfZXCw8MhkUgwY8YMYUjO48ePsWHDBri7u7f6z9u/UVxcjA0bNsDDwwMSiQQSiQRRUVGwtrbGqVOnYGFhIdylzcrOzsbWrVuRn5+PAQMGgIhw9+5dWFlZ4aOPPoKnpyfu3buHXbt2sS+fmZmZ2LFjBx48eIAZM2ZgwoQJwqdtEqlUig0bNuDGjRswNDSEpqYmsrOzoaenh2XLliE+Ph5+fn6orq5Gjx49UF5eDicnJ/Tp00f4VK1H7dki6hMQEEAmJiYEgD766CM6cuSIsApTWVlJAOqdqaU1+e2336h37960YMECYei11JLP98iRI6SgoEAQzFSlr68vrEokW0rNzc2NpFKpMEQbN26k69evC4tbHBsbG5oyZYqwuI6UlBTq0KEDJSQkCENtxtGjRwm1lo178uQJgS8bV6/AwEAaMWIE2z558iQFBwcTyZZI1NTUpHXr1tXagygmJqbeGbmeV2ZmJqmqqlJ5eTmRbL3juXPnsvjnn3/O1gSOjo4mLS0tys/PZ/HW5pmJ9XnUrF/Z2hMrEdHy5ctbZKJ5Udra+bZk1dXVcmu1NqYmobwMRUVF9S5u/irx9Vib7ttvv6XFixez7eLiYsrKyiKSJTpnZ2fS0tKSWxc3NDSULly4wLaFqqurWXJuTExMDJmZmcmV1Uy1SUT0v//9j44fP862tbW16dixY2y7teGDRRtQsxZnW9HWzrcpcnNz4efnhytXrqCgoAC+vr6IioqS63gnEokQFRWFzMxMpKamIigoSC6empqKHTt2sN70tVVUVCA8PBzbtm1jK5EAQFpaGls/tcaVK1fg5eWFgIAAtihAaWkpjh8/XqejVV5eHgIDAxEQECC30g1krzc6OhqZmZkoKSlBcHBwnWM1xMXFBb///ruwWI5UKsWFCxdw/vx5FBYWwtfXV25Fm/z8fAQFBbH1WYX+/PNPbNu2DcePH5crz8rKgp+fH27fvi1XzjVdcnIym+CnsrISKSkpbEar5ORk2NjYwM7ODsuWLWP73Lx5s9HOX9XV1fDz8xMW11H72AAQExMDExMTQPaeSUlJYccpKCjA48eP8e6777L6rQ1PrBxXj/Lycixfvhxz587Fpk2bsHr1aly4cAFTp06FpaUlpFIpW63FysoK0dHRsLKygpOTE1JTUwEAmzZtQlRUFN5//324ublhxYoV7PnPnj2LxYsX4+nTp7h16xbGjBmD6upqHD58GCNGjJAbcxsWFoZ169bhk08+QWxsLCIiIlBeXo6ZM2di0qRJuHv3Lqt74MABzJo1C++//z709PQwYsQI7Nq1C5B1QrS1tcWkSZPg4eEBNzc3XLx4EdbW1s22sLWnpydGjhyJw4cPY+rUqViwYAH27dsHADh58iRWrlwJExMTxMTEwMLCgn0Jyc3NhbOzM+7cuYOnT5/C1tYWSUlJePLkCebMmYPQ0FDcvXsX/fv3x8WLFwVH5ZoiOTkZsbGxWLFiBcaOHSs3O9nly5dhZmaGVatWISoqil3jioqKJvemb0xycjLu37+P77//HnZ2doiJiWGxtLQ0KCsrQ0FBAdevX4etrS02bNjwwobtvRTCn7D/xevUFPz999+3qabRtna+TVHzfg4NDWVlERERBIBCQkKIZM2jAMjGxoakUillZ2cTEdHp06dpwIABbL/du3eToqIiPXnyhG7dukWjRo2Sa+4tKSlhj11dXcnc3JxtW1hY0A8//EAku4d48OBBIllTXu1mzzt37pC6ujolJiayfUNDQ0lZWZmuXbtGVOv11l6ZZvbs2WRsbMy2GzJr1iwKDw8XFtcxYMAAMjAwoOLiYsrPz6eqqip6+PAhdenShfLy8oiI6Pbt2wSAjh07RiKRiMzNzeWaIGuuR0BAAI0ZM4aV9+nTh9zc3Ng2bwpumpycHFJTU2PXyd/fn+7fv8/i7u7u7PHGjRtp8ODBVF1dLXeta1u2bBnNmDGDbGxsSEdHh2bMmEEzZsyod8UjIiJDQ0PWpJyens4+P0REwcHBNHz4cNq8eTN5eXnRrVu3au3ZOjXrL9aab59NWcCZ41q6mubx2kO/PvvsM3Tu3BmJiYkAwHoYOzs7Q1FRkY3r9vHxgVgshp2dHebMmYPY2Fg4OTkBALy9vTFkyBC5sdodO3Zkj1VUVNhjyMYGe3h4YOXKlVBSUsK0adPqrRcVFQWJRCLX5GZlZQWJRMKae2teb+3p+QYNGlTvAuQPHjyAvb09+3fhwgV4eXmx7e+++064CyA7xrRp06ClpYW3334bysrKrOl38eLF+Pzzz7F+/XrY2tqia9euOHbsGMRiMfT09Nhz1FwPJycntprWnTt3oKqqiuLiYlavpWlsWkQ0If6iJCcnw8TEhL2nnZ2doa+vz+Ll5eXssaurKwoKCuDh4YG3336blde2ceNGhIWF4bfffkO/fv0QFhaGsLAwNhyutqdPnyIjIwMDBgwAABgYGGDWrFksnpiYiNGjR2PJkiVwcXFp3b2BZZotsWZnZ7MP2urVq3Hnzh1hlVZFKpW2qS8Ibe18/4tOnTqhe/fucmXCFVlSU1MxefJkhIaGws/PD6GhodizZw80NTURHx//XAtxr1u3Du7u7vDw8EC/fv3w119/CasAsjm4hTQ0NKChoYGHDx8KQ0xDw4/efPNNfPvtt+yfkZERpkyZwradnZ2FuzD1XQ9DQ0OEhYXB398fgYGBCA0Nhamp6TOvx5EjR+Do6Ihz587961WGXgYiwpAhQzBp0iS4uLjA3t4eZmZmcHFxwcKFC2FhYYGJEycCsmSzYMECmJmZwdzcHEuWLIGzszNWr179Qr44JCQkwNjYWFgMyP5216ampob169djxYoVdRZd+TeSkpKgr68PDQ0NYQiQNUObmpoKi1u1Zkus3bp1g4eHB4gIPj4+9Y4vbC0iIiJw9OhRxMTEICQkRBh+7bS18/0vxGIxsrOzYWlpKQzJ0dXVxZkzZ4TFgGxSi9jY2CZ9kZFIJEhLS8OaNWtw9epVtG/fvsHFB3r16gWxWCzXYamqqgplZWX/6g+XqqoqjIyM2D8tLS3o6+uz7d69ewt3aZCuri6uXbtW7y9jbW1tpKam4sGDB8IQQkJC4OTkhO3bt8PZ2Rnt2rUTVmkxrl69CnNzc0RGRrJWiezsbHh7e2P79u3w9fVlLRqampr45ZdfcP/+fUydOhVbtmzBzp07ER8fj3HjxjXpvdFUBw4cgL+/PxISEup0Crtx4wacnJxw8uRJnDp1ipXPnDkTgwYNarTjUlNcunQJq1atQmFhIXbs2CEXKysrw9q1axEfH4+DBw8iLS1NLt6qCduGuZanrKyMiouLhcXcSwCA9u7dy7Y3bNggN9av5j5nTEwMKyMi8vPzIwC0YsUKEolEVFRURJs2baLq6mr69ddfCQDNnz+fnj59SmVlZeTh4cHuP7q6utLQoUOJiEgqlZKJiQm7N/bDDz/Q+PHjieo5dkFBAWlra9PSpUvZ6zh16hTp6+uze5Y1+/zxxx+sjre3N3Xq1IltN6Sp91hNTU3p22+/lSu7efMmKSsr0/jx4yk/P58qKirI39+f7t27R3fv3qV27drRiBEj6N69e1RVVUURERF05swZWrFiBbVr147y8/OpsrKShg8fTuPGjWPX4/DhwwSAjY+sOb/aQzdeliNHjtC9e/fYtpOTE1laWrJtqVRK27dvZ9s197trD2c5dOgQAaC0tDRWVp+8vLx6x2o3p6dPnwqLuCbiibUV2LNnD+9Y9IoAoA4dOpCVlRU5OzvT/v37WSw/P5/s7OwIAJmbm1N0dDSLSaVSWrp0KSkrK5OSkhL17duX4uPjWXzNmjWkqqpKioqKZGBgQKdPnyaSdTbq2rUrqamp0erVq4mIaPz48TRp0iTy8vKi0aNHU2pqKuXm5rJjjxgxgj13QkICGRkZ0fz588nHx4cmT57Mxgvev3+fbG1tCQCNGTOGbt68SXFxcWRsbEwAnjkRQFMS66pVq0hFRYX09fXJ29tbLhYWFkYdO3YkAKSjo0OBgYEsFhkZSTo6OgSAdHV1afPmzURElJGRQXp6eqSpqUmOjo60fPlyUlBQIBcXF4qLi6OhQ4cSAJo9ezZlZWXRzJkz2TW5ePFiraO/fKampg12/iHZlx4FBQW5BLZ3714CQJmZmXJ1hbp16/ZadBJ9XTVpSkPu1fL390dKSkqjc5lyL4aCggLCw8Px3nvv/av7eyKRCPn5+WwN49oqKyuRk5NTb0woJycHIpEIvXv3btJaxXl5eZBKpejWrZsw9EpJpVJkZGSgZ8+ede7DEhEyMjLQvXt3uVhFRQUePnzIxlwWFBRAV1e31p4tT2VlJTQ1NREUFARbW1thGADw888/Y8+ePXJjcz/77DPcvn0b169fl6sr9M477yAlJaVNTmXZGjz7E8pxbZyysvK/SqoAoK6u3mDibNeuXYMxoa5du8LQ0LBJSRUAunTp0uKSKmQ9rQ0NDeskVci+xPTu3btOTE1NjSVVyO7XtnTXr19HVVVVo/coExMT0bVrVxQXFyM9PR3z5s3D1atXceTIEWFVrpVplb9Yg4KC6sw201o1tF6ki4sL6x14584dFBUVYciQISzu6+tbZ+D2q1xHdNSoUa1uaadnqaqqgqqqKvbt24fp06cLwxzXoD179mDhwoUoLS1tcFYzAwMDDBs2DFpaWlBQUMCwYcMwbdq0OsOoIPts+/v7s+1Dhw5hwoQJbML8jz/+WG4IS43Wto6p0Itc3/lF4on1FWsosd66dQsSiQSQ9dpNS0uTGzfYv3//Or9eeGJtPrm5uXB3d0doaChMTEzg5uYGe3t7YTWOq9fXX3+NhIQE/P3338IQIFuZSFtbGzExMRg7dqwwXEdxcbHcGFgLCwuEh4ezNWLfeustubHJNXhifTVaZWJta1rCPdbq6mqkpaUhOzsbY8eORXFxMeLj42FqaoouXboIq3NcmzZs2DAYGxs3OI/un3/+CQsLCxQUFKBTp07C8DPxe6wtW9Nu2LRBhYWFCA0NFRbLOXr06Gvzy/lZHj16hDVr1mDr1q0AgAsXLsDa2hpZWVnCqhzXplVUVCAlJaXRGYTi4uLw5ptv/quk2hI8evQIgYGBOHjwoNyiE9z/44m1AdnZ2YiIiBAWy4mJial3tpvXka6uLqZMmcK2J06cyAa7cxz3/w4ePAgrKytUVFQgKCgIv/zyi1xcJBLB3d0dW7duxePHj/Hll1+2ur8hWVlZcHJyQnp6OpYtW4aZM2cKq7R5rb4pWCwWNzglG9e89u/fj6CgIDZ7S//+/REUFIQPPvhAWJXjuNfUiRMnMHr0aCgrKyMuLg4jR47E06dP5ebUbuta7S/WpKQkvPfee1iwYAHc3d3Rq1cvODg4QCQS4dGjR7C2tsYXX3yBlStXwsjICNXV1SgpKcHKlSsxf/58TJ8+nU08XVxcDE9PTzg4OMDIyAi5ublwdXXFoEGD2PH27duHdevWYcaMGTh//jwuXryIkSNHsp56UqkUP/30EzZv3oyvvvqKNZkCgJubG8zMzHDgwAGMHTsWxsbG+Oeff1i8uX3zzTcYOHAgDh8+jF69euHGjRuorq6Gt7c3li9fDktLSzZ2rr5zr6yshI+PD5ydnTFu3Dg8ffpUeAiO49qosWPHsiFR7777LnR0dOqMUGjz5OeLaF0WLlxIzs7OVF1dTUVFRdS/f3/68ccfiWTTtPXo0YMuX77MZmCZPHkymyps0qRJtGXLFqqsrKRZs2ZRaWkpERGlpaWRWCymv//+mzp37syO1blzZ3r8+DGVlJRQZGQkERFNmDCBdu3aRUREq1evZjPlSCQSMjAwYEsjpaSkkKamJmVmZlJ1dTVZW1uTh4cHe+7mdvXqVWrfvj3t37+f4uLiSCqV0tatW2n37t1ERLR582aysrJq8NxXrVpFe/bsISKikSNHUnBwMBER7du3j8aNG8eO069fP7p06RLb5jiubQkODqZFixYJi9u8VvuLFbIB9vr6+lBQUMCbb74JJycnJCUlAbKB+YMHD8b777+PoUOHIjc3F+fOncPevXuxdu1aGBsbY/jw4Th58iSUlZVZM0bfvn2hoqICbW1tuWMNHDgQo0ePxvXr12FlZQXIJg+v8euvv2LYsGGAbBD8iBEj2ALPGhoa6NChA3r06AEFBQUMHz4cN27cYPs2N3V1dWhoaMDGxgbm5uZQVFTEjh07cO3aNfz4448oKiqCra1tg+e+atUqfP7557h8+TIkEskLWW2D47jWraCgABEREVi/fr0w1Oa16sQqpKqqKtcTr/bA7NzcXCgoKGDdunX4/vvvsX79egwZMgRpaWmNLqlVIzIyElOnToWlpWW9vYUfPnwIkUjEtg0MDOqMM60hnFnmRRAOSs/NzcWcOXOwcuVKrF27FtOnT2/w3C9fvgxra2uIxeJmWTaK47jXi1gsxtq1a+Hv74/27dsLw21e/X/5W5Hafa/i4uKwYMECQDbusnasX79+qKqqgq+vLyCby/Pu3bsYOnQoTp48ifj4eEDWjfzOnTuorq5mSzfdvn0bBw4cwIoVK7B161ZER0ezY9TU+fjjj3Hy5El2vPT0dIwbN47Vq/1ahNvNrb7nHzJkCDZt2sRe740bNxo8dy8vL/Tt2xfDhw9HVVUVuxctfF7hNsdxrz+JRIIVK1Zg+fLl0NHRQVZWFgICAoTV2jZh23BrsnTpUtLV1SVXV1f6/vvv5VbxGDNmDPXo0UNuxZHw8HDq2LEjDRgwgGxtbamwsJBIdn+0ffv2NHToUFq8eDFVVFTQ119/TUpKSuTr60tlZWVkbGxM27Zto4ULF9L58+fp7Nmz1LNnT/roo48oPT2d8vPzacyYMbRu3TratWsXLVmyhKRSKYlEIlq0aBEpKytTSEgIZWVl0YgRI6hPnz6UlJTEXltzEYvF7LV7eXmx8tTUVDIwMKAePXrQ+PHjKTExkaiecxeLxRQdHU16enpkb29PTk5O1L17d4qNjaUJEyZQ586d6dixYxQSEkLt27cne3t7vqQdx7UhTk5OpKysTGpqatSuXTsCQCdOnBBWa9Na9XAbNzc3qKurY+HChdDR0RGG6yWRSFBUVFRnIu/y8nKIxWK88cYbcuU1pFIpHjx4AD09PWFITlFREVRUVKCpqSkMtQh5eXno3LkzFBQUWFl9515WVgY1NTUoKSmhtLQUGhoaLMZxHMc1rFU3BVdXV0NBQaHJSRWy+5vCpAoA7du3bzCpQnbP8llJFbIOTS01qUK26kntpIoGzr1Dhw7sPi1PqhzHcU3XahPrP//8g0uXLuHvv/9uM9MKchzHcS1fq20Kzs3NRV5eHiD7lfhv18vkOI7juObUahMrx3Ecx7VErbYpmOM4juNaIp5YOY7jOK4Z8cTKcRzHcc2IJ1aO4ziOa0Y8sXIcx3FcM+KJleM4juOa0f8BZ35jtpSThNAAAAAASUVORK5CYII="}}},{"cell_type":"code","source":"from sklearn.metrics import f1_score\nf1_score(y_train_1, y_train_pred)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:54.177639Z","iopub.execute_input":"2025-06-18T07:54:54.178001Z","iopub.status.idle":"2025-06-18T07:54:54.195022Z","shell.execute_reply.started":"2025-06-18T07:54:54.177972Z","shell.execute_reply":"2025-06-18T07:54:54.194092Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Precision/Recall Tradeoff\nFor each instance, it computes a score based on a decision function, \nand if that score is greater than a threshold, it assigns the instance to the positive class, or else it assigns it to the negative class\n![image.png](attachment:84b4f709-4b09-4456-b80c-2a444bebb810.png)","metadata":{},"attachments":{"84b4f709-4b09-4456-b80c-2a444bebb810.png":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAn0AAADZCAYAAABVYYYcAAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAAAJcEhZcwAAEnQAABJ0Ad5mH3gAAP+lSURBVHhe7L13fFzF1f//3iZp1XvvkiXLlrvcu3HDFUxLQgoEQggJSQihOE9CICGFh4Q8JJBAEpoJYLCNC+7dsmy5d0mWrC5ZvdfV1t8f3zvzu9oYcJEJse/n9dJLu3fvnTtzZubMZ86cOaNzuVwuNGjQoEGDBg0aNNzQ0Ltf0KBBgwYNGjRo0HDjQSN9GjRo0KBBgwYNNwE00qdBgwYNGjRo0HATQCN9GjRo0KBBgwYNNwE00qdBgwYNGjRo0HATQCN9GjRo0KBBgwYNNwE00qdBgwYNGjRo0HATQCN9GjRo0KBBgwYNNwE00qdBgwYNGjRo0HATQCN9GjRo0KBBgwYNNwE00qdBgwYNGjRo0HAT4JpIn9PpdL+kQYMGDRo0aNCg4RrgcrncLw0IdK5rSHnRokVkZWW5X9agQYMGDRo0aNBwlbBarfz2t791v3zNuCbS9+yzz/Lss8+6X9agQYMGDRo0aNBwldi7dy8zZsxwv3zNuKblXQ0aNGjQoEGDBg3/HdBInwYNGjRo0KBBw00AjfRp0KBBgwYNGjTcBNBInwYNGjRo0KBBw00AjfRp0KBBgwYNGjTcBNBInwYNGjRo0KBBw00Aw7PXEHPlSrcUX0N0mP8q6HQ69uzZw29/+1vWrVvHjh072LdvHwUFBcTHx+Pn54fL5cLlcuF0OnE4HOh0OlwuFzabDb1eT19fH3a7XcpM3Od0Ounr68NoNPZ737Zt21ixYgUZGRn4+vr2S8dqtWI0GtHpdGzatIlPPvmEsWPH4nA45Lt7enrkfb29vej1/28+4HQ6sdvt6HQ6nE6n/N/X14der5ffRVl0Op3M17XC5XLR29sLgMFgoK+vj1WrVvHWW29RXV1NWloaHh4e2O12AIqKili5ciWpqamYzWZ27tzJq6++ysmTJ4mLiyMwMBC73Y7D4UCv1+NwOHC5XDLPdrsdp9MpfxPpirwAdHZ2sm3bNsrKyiguLqampoZTp07R3t4OgI+PD/X19Rw+fJjS0lIKCgooLy/HbrcTGhpKZ2cnr776Ko899hgAo0ePlvnIyclh27ZtJCYm4u3tfdP0FwG73Y7BYKClpYXVq1fjcDiIioqSbbmjo4Onn36ajIwMnE4nx48fp6ysjMLCQkpLSzEajXh4eIDSXoT8XC6XbMPqd4m6FkHmXS4XFosFk8lEcXExL774Irt27SI6OpqwsDDZxl0uF/v27ePFF1+kpqaG9PR0jEajTO/NN9/ku9/9LjqdjhEjRoDSR3Nzc9m4cSMpKSmYzWZcLhd9fX3odLoB7TefBqE/uru7QZGR6PMnT57k97//PYcOHSI6Oprg4GAsFgt6vR69Xs+f//xnzp49y+jRo2ltbWXlypW89dZb1NfXM3jwYIxGIzabDRSdoe5XKPJ2v+ZwOLDZbLhcLvlZ6BCXy8XOnTvR6/UEBQVJneRyuXj77bd55ZVXsNlspKamYrfbpS7Kz8/n5Zdf5sMPPyQtLY3Q0FBsNhsGg4FDhw7xwAMPUFRUxKRJk6QOPXPmDL/+9a+JjY0lIiKCtrY22Y40aBgoXGkfLy8vJzEx0f3yNUOz9A0wRMVWVVWRm5vLyZMnOXToENnZ2Xz88ce8/PLLVFdXS8UmlKoYoARR0+l0kqjpdDp5n16vx2QyodPpJGkDqK+vp7S0VBIVMUB5eHhgMpnkwNbW1kZpaSk2mw2TySTf53Q68fT0lEpXTTj1CgEU+bTb7RiNRqxWK1arVebxShv150GUu6enB4CtW7fy0UcfUVpaSm5uLrm5udhsNoxGI3q9nsrKSux2O56enhQWFrJhwwbOnDnDpk2bWL16NV1dXfJeVOWyWq1ywBblQxkUhZwFabDZbOzdu5fXX3+dd999l7fffps1a9awcuVK2traMBgMnD9/njfeeIO3336bd955h3/84x8cP34cgIaGBiorKxkxYgTnzp2joqICg8GAwWDgwoULeHl5ERAQACpicqNDDPoGgwGA6upqDhw4IOtdDOhNTU0YjUZ8fHzIz8/nL3/5C6+99hpvvPEGf/vb39i9ezculwuj0YhLmUCJulTXo0hT1Le4brfbMZvN1NbW8tZbb3HmzBkKCgp4//33qa2tRafTYTAYKCgo4O233+bUqVPs3LmTDRs20NfXh8FgoLy8nJMnT5Kens6JEycoKiqS7ay8vByHw4HZbAaF9IiyfVHQ6XR4enrKPu7p6UlxcTErVqygvLycc+fO8eGHH9LU1CTrQ8gsMDAQgEOHDrFnzx5KS0vZuXMnubm5oNSj0GmiTHa7vR+xdTgc9PX1ybIbDAZ5v07ReXq9nnPnzrFjxw5JUD08PLBarWzevJnDhw/T1NTE+vXr2bdvHx4eHuj1ehobG1m3bh3Hjh2jsLCQjz76iLq6OkwmE3a7nby8PKKiomhsbJR5RtGJer2ekJCQfnpQ6FYNGm4kaKTvOqGqqgqz2cyjjz7K8uXLefzxx5k3bx7vv/8+f/rTn+js7JTKTgz6YpDy8vLC09MTo9GI0Wikr69PkjhAKjHxO8Dw4cO5++678fb2BkVJCoUqyJ3D4WD48OEsWLAAk8kkByqz2Yyfnx8mkwmr1SqVvciXUMSChNkVi4zZbMbT05Oenh7sdjsWiwWLxaKSwrXD5XLh7e3N0aNH+fvf/86wYcN48skniYqK4m9/+xvl5eXy3s2bN8u8/+Mf/6ClpYWf/OQnfOMb32D9+vWsWrUKlHIJWer1eikrg8EgSbLRaJRpobL0eXh4kJSUxJAhQ5g0aRIZGRnk5+eTk5MjrQO5ubmUlZUxYsQIJkyYwNixYwkPDwdgw4YNmEwmfvnLX2K329m6dSsAlZWVHDt2jKSkJFnndsWCcTNAtK3a2lr+53/+h1WrVtHZ2Sl/Azhy5Ai+vr74+PiQm5tLQUEBqampTJgwgfT0dKKiojCZTHKwFmnq9XqMRiMmk0lOmER9i9/Fn8Ph4M033yQ7O5vvfe973HfffRw8eJC33noLgPb2dl5++WVcLhfPPfcco0eP5vXXX2f37t0A7N+/n7a2Nn75y19isVj48MMPQZmU7d+/n/j4eMxmM06nE6fTiUGZWHwR0CskVy0Dq9XKK6+8wunTp3n88cf5yle+wv79+/nggw9kHygvL6empoa0tDRpEUxJSeFnP/sZgwcP5g9/+ANnz56V7V8QSvEuLy8vmZbRaJTfBdkW+TIoOnDTpk089thj7Nmzp58uKiws5KWXXiIlJYWnnnqKqKgo/vGPf1BSUgLAu+++y4kTJ3jooYd4/PHHOX36NOvWrQPg8OHDZGdn89BDD5Gens4//vEPenp6aG9vZ/fu3WRlZRETE4PNZutXDg0abjR8ocu7NwOEosjOzsZisfDMM88wfPhwhgwZwtSpU2lqamLfvn0sW7aM3t5eamtrKSoq4ujRowQGBhIcHMz58+fZsWMHhw4dwmazERcXh91ux2Qy0dPTw969e9m3bx81NTUEBgZiNptxOBx4enoSFhaG0WikoqKCtWvXcujQIRwOByEhIZL8eXl5ERgYiJeXF42Njf3Si46Oxmw2o9PpKCkpob29nYqKCvbt20d9fT3BwcFy1l1aWkp7ezuBgYHSuqJXZu8DBTEY//3vf8fb25v//d//JS4ujgkTJmAwGIiIiCA0NJSWlhbWr1/P9OnT8fT05N133+Vb3/oWixYtYsyYMRQUFFBaWsqcOXNkXgWRbmtro6mpiZaWFrq6umhvb6elpUUORGLAcjgcmEwmRo4cyezZsxk/fjy+vr4cOnSIxx57jEmTJtHe3s6mTZuYOHEiP/3pT5k4cSKTJ08mOTkZi8XC+vXrGT58OFOmTOHChQsADBkyhFdeeYWAgAC++tWvgrJENtCy/LJCbQV65ZVXyM3NJTY2ltmzZ5OUlCQnGW+//TaxsbFkZWWxevVqxo0bx7PPPsvEiROZPXs2qampctIi0gRobW2lqamJtrY22tra6Orqoq2tDYvFgoeHh5zUGAwGmpubefXVV5kzZw73338/gwcPxuFwsHXrVhYvXsyJEydYv349zz77LJMnT2b8+PHs37+fEydOMHfuXA4dOkRCQgLz5s2joaGB2tpa5s6dyyeffEJ7ezvf+ta30KsmUDrF4v9FEAwhYyEbg8FAe3s7VVVV3H333UyfPp2hQ4dSV1fH/v37WbhwIZ6enhw4cIADBw5wzz33UFtbi8Fg4IknniApKYmMjAzWrFlDREQEw4cPl+/R6XR0dXXR2NhIW1sbLS0ttLa20tHRQWdnJwZlgiUg+vnbb7/NihUrCA8Px8PDg2nTphEZGQnApk2baGpqYvny5QwePJhhw4axYsUKAgICSEhI4OWXX+aee+7hrrvuIj09ne7ublauXMmcOXMoKSmhvLycBx98kK6uLrZt28bixYvJz8+nqKiIBx54AC8vL5l3u2LZ198kky4N1x9X2se15d3/EgiLkENZrhLfBfr6+vDw8MDT05Pdu3fzox/9iBdffJFXXnmFwsJCzp49y89//nNWrVrF9u3bWb58OZs3b8bLy4u+vj7eeOMNfvvb37J582b++Mc/8swzz2C1Wtm2bRs//OEPaW1tpaysjKeffpoPPviAbdu28bOf/YyPP/4YvV7Pli1beOaZZ+ju7qa5uZlXXnmFF198kY0bN/KHP/yBP/zhD7S1tWE0Gvn73//OI488wjvvvMO7777L8uXLefnll9EpyzAvvfQS//jHP+SAaTKZ+slgIGAwGGhtbeXixYsMHTqUCxcu8O6773Lw4EEWLFhAeno6KJay0NBQUlNTCQgIICIigsrKSvr6+iguLqahoYH4+HhpGRV57u7uZs+ePfz+97/nt7/9LS+99BLPP/88zz//PPv27ZPLUGp/I4NiJerp6eH1118nNjaWuXPnotPp6Ovro6GhgdbWVt5//33eeOMNTp48icvlwsPDg9DQUKqqqiSZ9vb25ty5cxQUFLBgwQKMyrK5yN9AyvLLCkF6tm/fzrlz51i+fDkxMTG0tbWBYtmuqKigvLyc1NRUent7KSsro6WlhTVr1vDaa69RWFgo/b70KsuR1Wpl586dvPDCC/zud7/j97//Pc8//zy//e1v+eijj6ipqZHEC8Db25vBgwdTU1NDQ0MD7e3tFBUVERcXh8vloqioCG9vb2JiYqQla8yYMTQ0NNDb24uHhweHDx+msbGRoqIiIiIiqK2t5dSpU0yZMgVfX18ciluGsHJ9UcTC3bJos9nw8/PjW9/6FpMmTcJut1NVVUV1dTXJycn4+/uDMvhER0cTEBDAmDFj+MlPfoKHhwc2m43Tp0+j1+ulJRuVb2ZFRQV///vf+c1vfsOf//xn/vznP/Ob3/yGl19+Wbo7eHh4SKu7y+XCy8uLhx9+mK997Wty9UHgzJkzJCYm4ufnh8PhwNfXl8DAQLq6ujh79iwul4tBgwbJ+zMyMrBarbS1tREREYFer6ewsJDKykr8/f2x2Wzs3LmTjIwMAgICZBtwKj6JgiBr0HAj4YvRNjcZHIpjfllZGStXrmTNmjVs3ryZF154gY8//pgFCxYQHBxMeXk5FRUV3HvvvfzrX/8iKiqK5557jvj4eF577TVWrVrF/fffz4svvsi5c+c4dOgQn3zyCT//+c9Zu3Ytf/3rX3E4HBQUFNDe3i5npjt27MBms7F69WrWrFnDI488In1jqqqq6O7uxs/Pj5deeomzZ8/yyiuvsHbtWl544QWOHj3K6tWrAejo6KCkpIR7772XtWvXcu+997J582bq6urw9PTkvvvuY9myZf3IiVg6Gki0tLTQ3NzM+++/zyuvvMKhQ4f49a9/zd/+9je5nFxWVkZISAiRkZGEhoaydOlS3nzzTW6//XbuvfdeGhoa+NrXvoZB2RAiZvJeXl4MGjSIJUuWsHTpUhYtWsSdd97JnXfeSXp6OjrVBhZhmRHEccWKFRw4cIAHHnhALqu3tLRw+vRpKioqKCkpITc3l5///OeSdN955510dnayfPlygoKCmDx5Mvv372fixImMGTNGEm5hlR1oWX4ZodfrKS8vZ8OGDXz7299m3LhxdHR09LMEnTlzBn9/f0aNGkVtbS01NTWUlJRQVFTEkSNH+Pa3v82aNWtk3YjlUw8PD0aMGCHr9a677mLp0qUsWLCAMWPGSP9JFLLi4+PDN77xDc6ePctXvvIVli1bRk5ODvfddx8BAQHU1NRgsVjw8/OTdSM2eXR0dLB06VJCQkJ47rnncLlc3HrrrXz44Yf09PRw66230tvbK/uKXtlE4viCfMfUZE89efHz88NgMPD666+zePFiTp8+zZ133gnKxqX29nbGjRsnLd5ms5nTp0/zwx/+kEcffZTMzEzmzJmDXfGLFO+JjIxk3rx53HXXXSxYsIBbb72VO++8kyVLlpCUlASqybHoZ1/72tdYsmQJwcHB0mVEoK2tDR8fH7y8vDAolsKoqChaW1tpbm7Gy8sLLy8veb9d8Ymtr69n/PjxDB8+nD/96U9UVFTw9NNPU1paSmNjI7Nnz8Zut0vfPqvVKvvfF1U3GjR8UdBI3wBDKFOTyURlZSVr165l/fr1fPzxx5w4cYLvf//73H///aDMckeNGsWcOXOIioqip6eHoqIiPDw8OHLkCJs2baKzs5OWlhYKCwvZvXs3UVFRTJkyBZSZ7LPPPsugQYPQ6XR0dHTgcDjIyMigra2NZ555hg8//JDU1FS+8pWvgLK71Gg00tDQQFFREcOHDyczMxO9Xs+ECROYNWuW9E/y8PBg6tSpDB8+HJPJxOTJkwkJCZGKeNKkSYwbNw694ufnUu1uHGh0dHRIovnqq6/y6KOPsnbtWrKzswEoLS0lMjISX19fiouL2bJlC+PGjeOOO+7g3nvvJTAwkN27d/+bQjcajYSEhJCYmMjQoUOJiYkhLi6OpKQkAgMDpRVGkFlhnRHLuKNHj2b06NGgWCrCw8N55plneOqpp/jFL37BX//6V+Lj41mxYgUNDQ2kpqbyzW9+k8WLF/PQQw/hcDi4ePEid955J6tXr+b555/nl7/8JadOnbopCJ9Op6O7u5u//e1vhIaGMmfOHDo6OgDw9PSU9509e5bIyEjCw8MJDg5m+fLl/PKXv2T58uW89tprDBs2jD//+c9y44TBYJAkJDw8nJSUFGJjY4mOjiYpKYm0tDTi4+Px9fWVbdZoNNLe3s7GjRuJjo7mrrvu4mtf+xrp6ens27eP3t5e/P398fHxwa7a2W1S/AhtNhtRUVE8/vjjLFmyhO9///uYTCaOHj3K4sWL2bZtG8uXL+eJJ55gx44dGJUdv18kxKTF6XRiU3bbijaemZnJo48+Snp6Op988on0eausrGTUqFGgWN5RiO6MGTO47777qKio4MMPP5Sk0KD45pnNZhISEoiKiiIuLo7ExEQSEhKIiYnB399fvlen2qwmSFZHR4fsqyh9S9zvUlwzDAYDnp6e9PX1YTabsVqtskwAZrMZg8p/95577uHrX/869957LxMmTGD79u3MnDmT0tJSnnjiCX7961/z3nvvyeV+MdnToOFGgkb6BhiC8Oh0OoYOHcrzzz/Piy++yC9/+UteeeUVHn/8cSIiIgDw8vKS/ngos2qdTkdNTQ0HDx5k7969nD9/nq985SskJSXR3Nwsd94JghUbGyutDkJpTp8+nZ/85CfSP+n+++/nnXfewel04u3tjcFgoKuri66uLoKDg2XenU4nvr6+NDc3S2IXEBAgla1eWebs6+sDVSgMp7JsJBTlQMPLywun08miRYvk4DNnzhzCwsIoKiqiurqa9vZ2hg4dCsCWLVsoKCjgRz/6EQ888AA//OEPWbp0Ke+99x75+fnoVct/fX19/Otf/+KRRx7h4Ycf5nvf+x6PPfYY3/72t1m1atWnzvTPnj1LbW0ty5Ytw2QyyYEoICCAu+++m8zMTFCI87Rp02hoaKClpQUUH75bb72V0NBQVq9ezahRo+js7GTr1q3MnDkTT09P/vSnP9Hb23tTDDrHjx/n7bff5sSJEzzxxBP86U9/4tSpUzLcTmNjI01NTWRlZYHSJu+44w5GjRolHe8XLVpESUkJxcXFoPQ/sXT48ccf893vfpcf/OAHfP/73+eHP/whDz30EC+99BINDQ3oVRbVgoICPvnkE77xjW/wve99jwceeIAHH3yQbdu2kZOTQ1BQkNx9KtDc3AwKyQCIiYlh7ty5JCUlsXfvXsaPH09ycjIrVqxg8uTJjB49mnfeeYeGhgZJLL6IehYuCsKyZlD8VHt6enA6nUyfPp0HHniA22+/nY0bN3L48GGqqqrw9/cnPj5eptHX10dMTAz33HMPTzzxBAEBAaxbt46enh4MitXfoOxy/t3vfsdDDz3Ed7/7Xb773e9y//338+Mf/5iDBw/2y5sgdYJUCl2ltvaaTCaam5ulNdGlhLwJCAjA19cXl7IkKyDuEWkEBQUxc+ZMMjIyWL9+PXa7neHDh7NhwwYSExNZsGABH3/8MWfOnEGn012XVQsNGv7T0EjfdYLdbsff35/U1FQiIiKIi4sjLCwMlCUNFEuGGChQSJXRaGTZsmU8++yzPPPMM/ziF79g/PjxJCQkSCubUHo6nY6dO3dSXFyMw+HA399fWo7GjRvHG2+8wYoVK5gxYwbvvPMOx44dk4owODgYs9ksd76J9zc1NWFSdtwJ64WAsIioia1YTtOprGADqShdLhd+fn7Sp0pcE0TU19eXyspK6urqiIqKAiWfAKGhoTKdqKgoXMoSnBqenp7MnTuXJ554gieffJLly5fz1FNPsXz5cubPn4+Xl5css0uxBgEUFhYSERFBRkaGTEuv11NQUMDKlSv7LUvZlLAyYvAR9X/q1CkKCgqYNWuW3IV8yy23MGHCBMrKyqirq5Np3Mjw9/dn6tSp+Pj4UFlZSXt7u5yYOJ1OamtrARg2bBgAJSUlfPTRR3R1dUlLkAjXIyYxDsVvztPTk8mTJ/P444/z9NNP89RTT/HUU0/x5JNPsmzZMoKCgnAqsShRTdr8/PyU3P0/q5Zer6e7u5uQkBCsVmu/uikpKSEsLIyAgACcKgtadXU1586dY968ebS3t0tr+ZQpU3C5XFRUVOChhBsR772eELISRNNoNNLc3My7777bTw94e3vLWKJnz54lNTVVkrGDBw/KHckobT48PJze3l7Z5p3KJqSEhATuvPNOnn76aX72s5+xfPlyfvGLX/D973+fYcOGSZ3hUnw6xX+Ufmmz2WTYHp1Oh5+fH5WVlVitVlAmyVVVVYSHh5OQkIBer+fixYsybwUFBXh5eUk9IOqlubmZTZs2MWLECHx8fOjq6mLmzJnMmDEDDw8Pjh07hkPx5f0i6kWDhi8SGukbYAil1dPTQ19fH+1K0F4xCNmUQLMoSkuEY0EJu5KYmMiHH34o47e99957PP/887S1tXHrrbdy4cIF6SO0Y8cOli9fTllZGV5eXnR3d2O1Wtm1axc/+tGPKCgoICUlhdGjRxMWFiaXQPr6+ggODmbu3LlyGbm9vZ0NGzawd+9eli1bBkp4CjV5sdvtcpMBwPvvv8+WLVtwqgJMDzScTieBgYHMmTOH9evXs3nzZlpbW/nXv/5FR0cHmZmZ7Nu3D51OR0xMDChy9PDw4IMPPqC6upri4mLee+89hg4dypAhQ0BlpdTpdIwePZpbb72VW265hblz5zJz5kwWL15MWlpav4FIrwrrkZ+fL5erUBG5np4eXnjhBf72t7/R2trK8ePHWbFiBWPGjCEyMlISUqvVyqpVq5gxYwZxcXGYzWa8vb2pqamhvr4eT09POdDeyHA6nYwcOZLVq1fzr3/9iw8//JBf/vKXjBgxgqeeeooxY8Zw9uxZdDqddNJvbW3l1Vdf5ZVXXqG5uZmTJ0+ye/duHnroIWlh1StWaZfLRUZGBosWLWL27NnMmzePW265hUWLFjFlyhQZPkXUcVhYGKGhoaxbt47a2lo6Ozt55513CAoKYurUqYwdOxaTycTKlStpampi165d5ObmMnHiREk4TSYTNpuN1157jbCwMNLT0zEpIZIuXrxIS0sLer1eTvgE8bne0Cm7d1EIkEuxgmVnZ/PHP/6R+vp62traeO+99/Dz8yMyMpKjR48ydOhQKZ+2tjZee+01Pv74Y7q6uti9ezcHDhxgwoQJBAcHy/Zts9kIDw9n1qxZLFmyhPnz5zNv3jxuvfVW5s+fT1xcHFarVfZBdR5R2kVvb28/y92iRYtkLD6r1crq1avp6Ohg3LhxxMXFMWjQID744AMKCgo4f/48mzZtYt68eaSkpOB0OiXp3bBhAwBz587F09MTh8NBY2MjVVVVtLS0EB0d3W9ZWIOGGwlayJYBhlCsDQ0NBAcHM3r0aOk3JGazgjQ1NjYSEBDAqFGjMJlMmM1m0tLSKCkpYceOHaxbt47i4mLuv/9+xo8fT3h4OOHh4WzZsoX333+foqIiFixYwN13301tbS19fX3MmTOH8PBwzpw5w7Zt21i7di11dXXceeedTJs2jcbGRsxmM+PGjWP48OH09vayb98+1q1bx549e1iyZAnf+c53MChBhkV4FJ3ie9XZ2cnkyZPx8fFh5cqV9Pb2MnbsWHRuwaQHAkKWNpuNQYMG0dHRwc6dO9m+fTtVVVU8+uijDB48mK1btzJ9+nQyMzNxOBzExcURFRXFzp075f1eXl488sgjJCYmyoHEoPgeiTwLa56wWIrvKCRC3NfV1UVxcTHp6emMGjVKLnHr9XppQT1+/DibNm1i9+7djBs3jkceeYTg4GAcih/h2bNnKS4ultamkJAQamtr2bBhAyUlJdx1112MGTMG3Q0eNkK0FbtyUorJZKKjo4OKigpGjBhBXFwceXl5hIWFMXLkSADCw8MJDAxk7969MlhvcnIyjzzyCOHh4bIuRPqiHYk+KCZgesUdQfx3KKGNkpKSOHDgAB988AGbNm1Cr9fzne98h8GDB+Pr60tMTAwHDhxg9erVnD17lhkzZvCNb3xDWoX1ej0VFRUcOXKEpUuXEhMTg4+PDzU1Nezdu5fDhw8zdOhQFi9eLNvfF0HwRVnVsvH29iY8PJy8vDzWrVvHunXr8Pf359FHH8XhcJCXl8eiRYsIDAzE5XIRGxuLQwlhs2XLFo4dO8a4ceN46KGH8PDwkO1byFqv8vcVEP1KyErdt0S+uru7cTqdjB49mqCgIOx2O3FxcRgMBt5//33Wr1/P+fPn+da3vsX06dPR6XQkJSVRUlLC2rVr2bNnD4mJiXzta18jPDxcWnJbWlrYtWsX8+fPJzMzE7PZTGtrKxs2bJBWzbvvvrvfZEDkS4OGa8GVtqPrFbJF57qGKeazzz7LlXDGa3jVfx3a29txOp0yFIBa0boU3xOLxYLD4cDHx0cOPCaTiba2NioqKrDZbISFhcmYVVarFQ8PD6qqquTpBKmpqfj6+tLV1SXjjnl5edHT00N1dTXd3d2EhoaSmJiIwWCgs7MTlxJqQiyhlJWV0dHRQUBAAMnJyZhMJiwWCx0dHXh5eeHr64te2dVWV1cn81NXV4eHhwdBQUHolZM9jErA6IGqa6vViksJ5eBwOCgpKaGzs5Pg4GCSkpKw2WzU1NQQHByMt7c3DmU3oMFgoLq6mra2Nux2O5GRkURGRuJQxSlzKX6R6lm9XuVMLgZiURZRjzabjfr6evz8/AgICJBpWpQjvEwmE+Xl5TQ3N+Pr60tKSoq0/rgU60p5eTm9vb0kJSXh6emJy+WSp3X4+fmRlpYmZSDazo0Om+oEja6uLvz9/TGZTHK519/fH5fK8lpdXU1zczMeHh7ExMRgNpuxK6dM6FVLc6Ku3SGui/4oJgFeXl5UVFRQW1uL0WgkPj6e4OBg+W69Xk9VVRUNDQ2YzWaSkpLk5g6dQuAaGhro6OggNjZWtkcRpsdut8s0RR6+CNIn2rVdCewuZGRQ4hOKE0Pi4uKIiIigs7MTq9Uqd/eK+x0OBzU1NdTW1uLn58egQYPwUEK4iLIKCPmq+xxuA6CapNuVcC+9vb10d3f3iwFqUY6Fq6qqorm5meDgYFJTU2UaKJbIuro6rFYrCQkJhIaGSkIv0i0uLiYlJQVfX1/pn1lQUCAnl2LJX9S1Bg0DgSslfdfLqKaRvgGGSyESwvdLKByh9MQ9QqHoFL8Wq3IGpkshBeoG4lKdyYvqeDABi8UiQxUIxS7SVkNYpMSMXCh89X3i+b6+Pry8vNDr9TL+mFDmDsVqIpZLxC47MWCLsg8ERD7tykkl7nkV+RCEQW1tcYcYGEQaol7U9SGuiefFIGdQEUO1zIRsxOAmBkd3WJXj6sRv7m1BDHw6VTtxKNaomwFCN4j+41AsRkIeesViJIgHKlIunhN9RK9MoESdiXoRabv3DdFuDaqzaN3bsFUJ+izagGj7KPUkiJRoK6hOfhGTIXXbcDqdst98UXUs2rZN8TFFkZu6LAJClqIMotxcQv/Y7XZ6e3vx9vaW8hX1JNq36Gfid1STKKdqeV39vEhb9AOnckSaO4TLiSD86ryJ3+3KEXvinUIXulyufmnalR3fQher86JBw7XAvV1+Hq4X6dNa8wBDDFBqRSlm1bgRPrWCEQOFmDGL38QAJxSzmgQJ0iOUqxi8xDPq3wWZM7qFiVCnIwYF8T5BRsR/8W4Uxe9Qll7FAKK2HgwUdMpynJeXFxaLBatyTq5NWRYT+dLr9dI/x6k45ov7hdIXA48oh/vgxadMTMSgJcomZCrKrpYTqvNG7QqZF3lV51fAoRAGkW+bzSavie/iXeq6vxEh6s2ghDwS5RX9Qa8ia0I2Qs4OZWlY9CNxn3jGpep37tCrdo16KDt+1X1CkDN1GiJfwidXPKdOX9SzmDyJNPuUs67VpPSLgNA5Qk7qNinyZbfbsak2b6nlIGQp+pPdbqejowOnsutfp1g51X1KvENAXR+i3Gqd4VR0k2gL4j6HoptE3xDX1P1G6CTR/0RfNSpHv4m01PXo6ekpyyz6qahrdR41aLhR8O8aUMM1Qyg6tcITSkY92xdkTwwagtgJpWxQYo0JRSgUkiATIn0xQKIoaTGAqQdJ8W7xjPguiJq4Jv4LIie+i0FLrfzFfeId7u8cCIj3WJVlXiE7nSocg04ZDISCFgOCkKEguuJ5kYZIx71OLvW7+Kwur8ibIGFquetV1iaXy4VV2XGoVxEM9XM61dFP7nApBFWUQacM1DcahDwF0RX9Q1jDRL0KWeqUti8+izYg0hFQt0ndJXy09Kq2jYq0i/uNigVP3OtSQoXolLAwoj7U7U/Upfis/k20N5Evkfb1hii7S+VPJ66pdYxOaYvqMhuUpVGnysKv0+nkZhQhL3FdnZb6uviu7gcoJM+pkGZRz6IOhazEfer/guwZFcucTqWXbDYbFosFmxKqRvRH9ftFfRgV6654l1EJkK7OowYNNwI00vclh1o5qZWwUGqC3IiBRSjNgYJLtfyjU4VoGch3fBZcqiUhQW7FAOBQZvWoCLRQ3CaTCafKL8dqtdLb2wsqEiAGv8uB+73is12x1ApiKX7TKWRRkAOj0YhdsVbZbDa5XCgGK1HPomxqciksEIJgqAfSGwXCmqdTCIeQi5o8obJUi+9CduKakN/lQqfa2IGKlOmUZVBRl+I3QW4EGRDtU9StqB9hdRZ5Er+L9EX+DW4+cNcTgrAJeaEisaJ9Wa1Wurq6+rUx0X6Fe4UgxS5FN6jb99USWJEntQyF7HSqSaro7+KaXvElFnVmUCZHQi+KfqRXdKaQuwYNNyuuTENq+I9ADPqofF48lPN7UREP3KwaAwG7aukDlX/MF6E8XcqgLwZJk7JJoqOjQw5SYqlMr9djsVhoa2ujr69PyshiseB0OvHx8ZE78tSDvBhcPg8iL+I59aBiUcVORBnIjUYjPT09ckAUllyXsilFpCkGOEEQDIrDf1lZGSjLjTrFEtPe3i53NeqvkNx82SHq1qCcstDZ2YlDWboTdWa322lubsagHMElZN/Z2SknJuqlycuBkLloY6I+BJmxWCz9iJrZbMaoWIGMqgmGp6cnRmXi4VBiZaKQVNF2LRYL7e3t2JWlUTV5/SKgbvcinw5lEiXanV6vJzAwUE6aRLvu7OyUwcJNJhM+Pj709PRI2avLczVlUpM7QdBEvehVS+PCsiiOlfTw8JAbuFB253Z1deGhxD8UfbKtrU0Gv9eg4WaGtpHjSwydTkdlZSWvvvoqFy9e7KcEPZWgwvPnz5cDlFDQhgGyHOh0OvLy8li/fj33338/3d3drFu3jnvvvVcGO77eED5xvr6+HDp0iLVr11JYWEh8fDz33XcfI0aMkATo+PHjlJeXs2jRIux2OytWrCAnJwen08mECRN4+OGH5S5gp+Lr41IsqII0u0OQO0H6xICEEi9ux44dREREMGnSJEk8Wlpa+Mc//kFOTg6hoaF89atfZc6cOTKtvXv38t5779HZ2cngwYP7hRo5d+4cH3/8MTU1NWRlZfHNb35TOqC/9dZbZGRkMHHiREmGbwQI2Yr2XVRUxLvvvsuyZctkSBydclLNxo0b+epXv4qPjw+vvPIKhw8fxtfXl9mzZ7NgwQK5E/5K4FIseKJv6XQ61qxZw8cff4zT6WTo0KE88MADxMbGkp2dzerVq+XEQtRBeno699xzD57KaSoXL14kKiqKe++9l/T0dADWrl2L0+lk6dKlGBSLlF21k/Z6w6WySgpfNpNyysU///lPDh06hL+/P4sWLWLJkiVyd7/JZGLNmjX4+voyb948ioqKWLFiBQUFBXh7e/Ptb3+b6dOnY1cmg1daHkEWdUqw+R07dtDZ2YlTmax9/etfZ+TIkRiNRo4cOcI///lPuru7iY+P58EHHyQlJQWAnJwcVq1aBcD06dO57bbb0CsBm999911mz55NVlbWF6K3NGhwx5VOOLSNHDcZRAOpqalh5cqVtLS0MHjwYFJTU4mJieH8+fM8/vjj7NmzRypYQV50bv6E6jTd/z7tuvjt4sWLbN68mba2NlpaWti4caM81eJS7xho6BS/odLSUp577jmam5u57bbbqKio4IUXXqC+vl7mY+/eveTn52M0Glm3bh0rV65k2LBhJCcn8/bbb/PRRx+BaklNDDTuMnOXgYB4zul00tHRwQcffMBzzz1HYWGhtOI5nU7+9a9/sXbtWmbMmEFQUBB/+ctf2Lt3L0ajkYsXL/LXv/4Vh8PBqFGjOHPmDC+//LIcjN955x10Oh233norOTk55OXlgXLs2/Hjx28YoucOQaDsdjvvvvsur7/+ujyJQ/yWl5dHeXk5fX19/POf/2TlypUMHTqUxMREnn/+edatW3dJsuFep+5/DtUGJqPRSE5ODi+88AJRUVHMmDGDHTt28H//93+g7JSvqamhuLiYmpoazp8/z6uvvsru3btxuVzs2rWL7Oxs7rzzTsrKyti3b598Ljc3l5aWFjlpEBMI93Z2vaBu005lqdfpdPL3v/+dNWvWMH36dGJiYvjnP//JunXr5D319fUcPHgQp2Jx/cMf/sCRI0eYP38+BoOB3/zmN5SUlKBXrHSijbrL2f1PnS8hk1WrVrF161aqq6uprq6W4aSMRiNNTU289tprWCwWZs+ezdmzZ/nTn/4krafvvvsu0dHRREVFsWLFCurr60EZPM+cOUNcXJx8pwYNNyv+XUNq+FJAzEYNBgNhYWF897vfZfny5SxfvpznnnuOVatWkZyczKpVq7BYLOiUZZfCwkJ27twpyYJ6UOvp6eHAgQPs3buXmpqafu9rbGwkOzub/fv3c/78+X6/iRhlZrNZLq+IZ1pbWyXZuR4wKUt+q1evJjQ0lD/+8Y/cd999/O///i8pKSlSsTscDgoLC8nIyKC+vp5PPvmEb3/72yxfvpzf/va3/OAHP6CtrY2enh4cylKqsPh1d3fT29tLX18fVuXEkt7eXmnZE5YIYQUqKSnhj3/8I9u3b6erq0sSDZ1OR1VVFVu2bOEHP/gBjz/+OC+88AKJiYmsWLECl8vF1q1b8fLy4k9/+hNPP/00v/jFLzh69CjHjx+np6eH/Px8Jk2axLJly/D09KSiogKA9957j+TkZEaPHn1DWflQDfpWq5UPP/yQTZs24e/v30+uAEePHiUpKYn6+nrWrl3LY489xi9+8Qt+/vOf881vfpPOzk7ZF0T/0Sm+eb29vdJq3NvbS1dXF93d3XKiJJZnxWkvS5Ys4U9/+hOPPPIIy5cvx2w209LSwtSpU3nzzTdZs2YNq1at4lvf+hazZ8/m3nvvJSIiguLiYm655RbmzZvHiBEjKC0txel0cuDAAYxGI7feeitOZaesyN8XZXkS7xFkDyWeaEtLC48++ig//vGP+e1vf0tSUhIff/wxdsV/7tSpU/T09DBq1Ci2b99OU1MTf/7zn/nOd77Dyy+/zNixYyktLZXETSy12u12enp66OnpkT6b6r4l6lXkq6uri6amJn70ox+xadMmGYR+4sSJoJykUVJSwq9+9Svuv/9+fv3rX3PixAkOHTokJwN33HEH9957LzqdjnPnztHT08P58+f52te+RkREhHyX0y1MjAYNNws00vclh07xc/Hx8ZHfAQIDAwkNDZXBlnt6enjrrbf49a9/zcsvv8yTTz7Ja6+9Rk9PD0ajkdLSUl544QWef/55XnzxRZ544gn27NkDwJEjR/j1r3/NX/7yF373u9/x05/+lDVr1sj3WSwWPJUAwg4lNIbT6eTll1/mzTfflJsVXFfpz/NZ0Ol0dHR0UFVVxYwZM+jt7WXr1q3YbDb+53/+h6FDh4JyFq5er2fkyJGcO3eOrq4usrKyOHToEPv372fZsmU8/PDD0v9IkLi+vj42bdrE008/zZNPPslvfvMb/ud//ocnnniCTz75RPp06VW7dGtqaujt7eVnP/sZmZmZWFRH1RUUFOBwOJg+fTooA2xmZiY1NTWUl5dz4MAB0tPTCQwMBCAlJYXg4GB27NiBp6cnoaGhdHV1yUHS39+fnJwcGhoauO222+TAeqW+a19miDZz7NgxuXw7bNgwWUadTkdDQwNnzpwhMTGRyspKQkNDmTFjBkePHmXPnj08+OCDfP/735dtUwzqPT09rF+/nl/84hc88cQTPP300zzzzDP8/ve/Z+XKlTQ0NKBTxUmsrKykurqaCRMmUFhYyIEDB8jIyODpp5/G19cXs9mMv78/gYGBMu3Fixdz5513gnJmbENDA6jiVzY3N5Odnc2oUaOIjY2VFjTRXy5lnbxeEPoEpQ15e3vz5JNPyvyLvu7j4yPr5cKFC+h0OiIiItizZw/jx48nKCiInJwcLly4wJNPPsn06dOxWq04FD9BnU5HcXExL730Er/4xS/42c9+xmOPPcZPfvITXnrpJU6dOiXrSLynsrJSTiK3bNnCtm3b6OnpwUM5s/rAgQMMGzZMnlKQmZlJfHw8hw8fxsvLCx8fHzo7O2lsbMRiseDj48OuXbswm83MmTMHlDoRm1Dsio+umARq0HAz4IvTNhquCILcORwOOjo62LRpEx988AFvv/02b775Jj/4wQ84duwYixcvxmw2s2rVKv785z+zcOFC3n33XR5++GE+/PBD1q5di9Vq5f/+7//IycnhV7/6Fa+//jq+vr784Q9/4NSpU7z22mv4+fnx17/+lTfeeINx48bx9ttvY1Ei4LtUOxjVGDt2LMOHD5fLOoIUDrQCraiowGq1sm/fPp5//nlWrlzJo48+ysqVK+U9xcXFJCQkEBcXR1lZGbW1tfz973/nzTff5A9/+AM//OEPuXDhAiZlw4BYhtLpdMTHxzN48GCSkpKIjIwkMTGRjIwMQkJCZNl1ql2LEydO5LnnnmPUqFGgDCQCNTU1mEwmGdUfwM/PD6vVSktLC7W1tXIjBwpJcDqdFBYW4u3tzb333sv69eu599575Xmi+/bt45ZbbiElJQWrEoj2RrL06fV6GhoaWLFiBYsXL2bx4sX9llxRrE12u52UlBTq6uq4ePEif/zjH/n73//OSy+9xFNPPUVVVRVG1S5pFEIZHh7OoEGDSExMJDExkdTUVOkm4aEEKheEvrOzk4sXL/LGG2/w17/+lbfeeovly5dz8OBBea9TWYp/44036OjoYOnSpTKfs2bNorq6mvvuu49Tp04xe/Zsdu7cSX19PQsXLgSlb4u2p/sCQ7YIqNurTqcjLCwMvV7PihUruPvuu6mqquKhhx7C09NTWsBHjBiBSzk1Ztu2bTz33HOsXLmSn/70p/zlL3+RdaVXrOEOhwNfX18GDRpEeno6CQkJJCQkkJqaSnJyMn5+ftKCLtpyU1MTDQ0NZGdns2vXLl5++WUee+wxysvLQSGk0dHRMu96vZ7o6Ghqa2uJjY1l8ODBvPDCC/zf//0fEydOJDw8nJUrVzJx4kQ8PT3p6uoCt1BZdsWn8lL6TYOGGxEa6fuSQxCULVu28MYbb/C73/2On//85zQ1NfHcc89xxx130NPTQ05ODunp6YwZM4auri7GjRtHcHAwW7dupaioiNLSUu68807Gjh1LbGwsv/rVr3jqqafw9fXl0Ucf5eGHH0an0+Hl5YW/vz89PT10dHRc0gohrCJLly6VGxSEtU/8DSQ8PDwoKyvj3LlzzJs3j+eff54ZM2bw1ltvcfr0aQBKSkqIiIjAy8uL3t5eSaKeeuopfvGLX9DY2MiKFSskaUIh1Hq9njFjxvDd736XH/7wh3z729/me9/7Hg8//DCTJk2SQV0dDofcSavX6zGbzdL539vbW+ZVWBzU71Fb57y9vaXlAoWUBAQEyGvTp0/n7rvvZs6cOTz88MM0NDTQ2trKnDlzOHjwIM8//zzvv//+pxLx/zaIMrz//vsA3H333fj4+PQjxiibdGJjY0lISKCzs5OioiIAfvKTn/Dkk09SWlrKa6+9hlU5UQMldpzJZGLatGl85zvf4Uc/+hGPPvooDz74IF/72teYPn06AQEB/ZYaARoaGmhvb+e+++7jhz/8Id7e3rzzzjty57Ber6e0tJTVq1czbdo0EhISQKnLIUOGsHz5csaNG8djjz1Gamoqq1atYvr06VRVVfHcc8/x9ttvy4mMO7m93nCqAi0L/1S7spvY19dXnl979OhRUIhYeXk5M2fOxOl0UlNTQ1FREWPGjOGpp57i61//Ort372bnzp14KKf2uJTA1dHR0dxxxx088MADfP/73+enP/0pjz32GLfffjspKSkYlY1UYpLo4+PDbbfdxn333cczzzzDz3/+c7mpB2WCJOoWZfnY5XIREREBwLJly5g/fz6LFi3i8ccf58SJEyQmJpKQkCB1Z05ODn19fdhUJ45o0HAz4d9HdA1fKlgsFry9vXniiSd47733eO+997j99tuxWq1kZWXh7e1NV1cXbW1t5Ofny+XZH/3oR1gsFkaOHElnZydeXl5SOdpsNiIiIpg2bRqJiYn09fXx97//nd/85jc8+eST7NmzB6MqIO2nwaE6rcCkOkFhoCGsN3fccQeLFy8mNjaWRx55BF9fX06fPi39ksSyj4eHB/Hx8XJwycrK4oEHHiAvL0/6yAnC5nQ6WbVqFT/4wQ949NFH+dWvfsVPfvITvv/977Nx40ZJImxKgFdBwlH5S6rLbFR809QDuVjG8vLywqac4iDgUvwhxfK9h4cHCxcu5Dvf+Q6hoaGsWbOGyZMnU1BQwLvvvktCQgIHDx6UA+GNgLy8PN5++22cTieffPIJK1eupLCwkNzcXBoaGuju7qauro4xY8bgoZx8kZqaysMPP0xGRgZTp05l7ty57Ny5k+rqavRKiA9hmVu1ahVPPvkkP/nJT3j88cd54okneOqpp/jHP/5BdXU1np6ecvDX6/VERkZyzz33MGrUKIYPH853vvMdampqOHXqlMzzwYMH6ezsZP78+aAiIC6Xi6ysLB555BEmTpxITk4OQ4cOZdKkSfzpT3/C29ubmpoafv/739PR0SEtjF8ERDv0VI4qE33XqISgWbZsGS+//DIPPfQQb7/9Nvn5+dTV1dHV1SXP7nY6ncyZM4f777+fuLg47rvvPuLj41m9ejWolo+9vLw4d+4cv/zlL3n88cd58skn+elPf8qPf/xjXnjhBU6ePAmKvPsUX9rRo0fzq1/9itmzZ+Pr68ukSZNYvHgxBw8exKWEvWlra5Pl0amOdmtvbycmJob77ruPu+++m87OTrKzs5k7dy7Z2dns3r2bjIwMVqxYwcGDBzEq4X5EH9ag4WaBRvq+pBBEwsvLC5PJRFpaGhEREWRlZfHEE0/Q3t7OI488wsWLF/Hx8cFoNHLLLbfw8ssv8+qrr/L666/zxz/+kSVLlhAYGIjFYpHBiU0mExcvXmT79u1kZ2fz5JNP0tbWxoMPPsiLL77Ifffd96kkTv3doDotQVj/Bnrm7FSOeIqLi5PECNW77XY7p0+fprW1lYyMDFA2noSHh+Pn5yfv1ytx+4SfmCAEer2exMRERo8ezahRo0hNTSUzM5OsrCyio6PlgCDifqkhyt3T0yOvRUdHY7PZaG1tlddqamoICgoiKiqK6Oho2trapBy7urro7OyUeRcDGcCuXbtob2/nlltuIScnB5vNxre//W0SExPZv3+/TP+/HTqdjpSUFHp7e/nkk0/Yt28fjY2N7Nu3j+LiYrlsO27cOABiY2MJDAzsZ2E1mUxYlSP3cDv/OSoqisGDBzN8+HCGDx/O0KFDGTlyJBkZGfj7+2NVYk+iWJu8vb2JjIyUz5vNZnSKbytAT08P+/fvZ+zYsQwZMkS2I0GqnMqyZX19Pfv27WPRokW0tLQA8NWvfpVFixbR1tZGVVWVTPuLgHiPU/F5FP6Ha9eulTulAcaNG4dOp6OoqIjz588zbtw42a/T09OJjIyUbVSnxMsU8hbvsNvtBAUFMWrUKIYNG8agQYMYPHgwmZmZZGRkEBAQIP1Wvby8MBqNnDx5ktzcXFwqP0ez2UxISAgAQUFB1NTUyHqw2+0UFxfLOhMTM4fDwZo1awgODiYxMZGioiIWL17M17/+dQIDAykoKMBms0nyq+5zGjTc6NBI35cUQnmKWbDYsGG320lMTOSHP/wheXl5vPnmm3h6ejJmzBh2795Nfn4+ISEhtLa28qtf/YqcnBzi4uKIiIhgw4YN0sn87bff5sUXX6S0tJTe3l4WL17M0KFD8fT0ZOvWrbS2tmJUTj9QWzHUlon9+/dz8uRJXKqjwQZaedrtdkJDQxkzZgyffPIJZ8+eBcWpu7m5mcGDB3P27Fm6urqIiooCYMSIEfj4+LBx40bsdjvd3d3s3r2blJQUkpKSpHXAoezcnDx5Mg8++CAPPvgg999/P9/5znd48MEHGTt2LGazGZtyMoMYiIQsBNFVlzktLQ2j0cibb74JQG1tLTk5OYwZM4aIiAimT5/OyZMnZTl2795NV1cXkyZN6pd2e3s7+/btY+HChXInq6+vr2wL6iXi/3YMGjSIv/3tb7z00kv8+c9/5sEHHyQjI4Mf//jHTJo0iaKiIux2O2lpaQAMHTqU5uZmPv74Y1CWY7Ozs5k4cSKxsbEyXbsSj27KlCk8+OCDfPvb3+Zb3/oWDzzwgNx1GxgYKNs3QFxcHKmpqezYsYOenh5cLhebN28mKCiICRMmAFBXV0dZWRkjRozA29u732RA9BGdTse//vUvQkNDGT9+PE6nE4sSPFxYlwSR+qJIn4Be8bvTKTubX3/9df75z39iVTY5bNu2jeDgYBISEsjNzSUmJkb6IE6bNo2zZ89y8OBBUJbdy8rKmDNnjiS9LiXuYVxcHHfeeScPPPAADz30kOxjd999N4MGDUKn+MmKvpWXl8czzzwjw9zk5eWxc+dOxo8fj06nY/LkyZSXl0sr4fbt22lvb2fChAlyNUCv13Py5En27dvHzJkziY6Oxul0SguhWJ53XeI4PA0abgYYnr2S6MpuuF7BAzX8/wNBbW0thw4dYuLEiQwaNEj+lpaWRnt7O7t372bKlCmMGTOG48ePs3XrVs6dO8dbb72F1Wrlq1/9KgkJCQQFBbFjxw727NlDdnY2e/fu5Vvf+haLFi3i8OHD5OTkUF5ezpYtWygrK5O7Izs7O8nLy2PBggX09PRQWFjIzJkz8fX15c9//jO1tbVMnjxZEkQxmAwERDo2m43Y2FjOnTvHqlWrOHLkCNu2bWPZsmXMmTOHjRs3kpGRwfjx43Epjvv+/v6sWLGC3Nxctm/fTn19Pffddx8pKSlygBfK3qkK2SDeKa6JAV1dLvHZ4XBw8OBB0tLS5KYOHx8fdMqAf/LkSdatW4fZbOa73/0uoaGhREdHc+zYMdasWSPrYeHChSxZskQSaoPBwNatW6mqquL222/H398fo9HInj17aGxspLGxkcmTJ5OZmank+r8bBoNB+vGJ4Nlnzpxh8uTJJCYmkp2djb+/vwzdERQURF9fHx9//DHHjh3jww8/xGKx8LOf/UzujkXlsynq2KXalKMm7YLQ6HQ6aeVbvXo1+/btY9++fRw9epR7771XWhrLy8s5ffo0M2bMIC0trZ+rgChPbW0ta9euZdasWaSlpWE2m8nOzqayspITJ05gNBq56667MH6BmwhEGdXvE21r5cqVnDx5kh07dnDw4EG+853vyIni7bffLl1DEhISyMvL46OPPuLMmTN88MEHZGZm8t3vflfqAKNyLKFLFRZJvFPI3T0ver2ekJAQzp49y+rVqyksLOTDDz/E39+fH/zgBzJawcWLF1m1ahW7d+8mJyeHRYsWMX/+fLki4nQ6ZRDsBx98EKPRSEVFBZs3b6a7u5vjx49z++23k5ycLK3+/AeIt4abD1faxsrLy6XL0kBCO5HjSwydTkdraytnzpwhLS2NqKgonE6nXJqoq6sjPz+f5ORkEhMTaWho4NChQ9TV1REUFMTIkSOJj4/HoAQVLikpITs7G71ez9ChQxkyZAg+Pj5cuHCBY8eOYbfbiYiIIDU1lYaGBpKTkzEoJySMGjWK7u5uysvLycjIwGw2c/ToUcxmM0OGDMGgOmfWpJxMMRBwqk4QuHDhAkePHsVms+Hn58e8efMwGo0cOHCgn3xQBpETJ06Ql5eH0+lkzJgxcgeiSNdms2FUdvJdKXSKleLkyZOEhoaSlJQkrTcWi4Xjx49TUlKC1WplypQpDB48GKvVioeHB3V1dezfv5/6+nri4uK45ZZb8PX17Uf6Dh8+TE9PD9OmTZPWv9zcXIqKioiLi2PmzJl4KmF0/tuhLoNOp6Orq4uioiJiY2MJCwvjzJkz+Pj4MGjQIGzKCRF2u52DBw/KDTujR48mIyOD7u5uPD09+1nfxDvcrwmI99uVo8S8vLw4efIkp0+fxuVykZSUxOTJkzEpYVaampq4cOECiYmJREdHyw0jgkTqlJN0iouLGTlyJCEhIdhsNi5cuEBOTg6+vr6MHTuW6OhoSZA+LW/XE6Jd2Ww2Tpw4QX5+Pna7nYyMDKZPn05ZWRnFxcVMnDhRxufU6/U0NTWxf/9+Wlpa8Pf3Z8yYMXJ5XpTHoWxQEVY10V9cKgubeL/6nrq6OnJzc6mvrycoKIhJkyYRHx8v+2pzczO7du2iubmZxMRExo0bR2hoqHy+u7ubnJwckpKSSEtLk/W1b98+enp6CAoKYtasWXLXvJD71egADRquBFdK+q6XUU0jfV9iqAcqMWPW6/X09vZiNpulohIz1kuRLZfKonGpRqdeZlLPyHVuAW7tSkBgnbIkJBQ3yhK0h4cHPT09mEwmPDw8BqyuxcDgUo53UsOpst7olbN3xT26SxxHJwYiAZ0qPtuVQshVvM+uxPwyKOfuoorTppaZ1Wr9t52p4je1vPuUY76Er5pLCY5tsVj6veNGgbqeRTtD1SaFDMXv7nWLIm91f1G3w0u1fQGn4uNmMBiwKbs6RV8SJFP0MWHNEu9XtylhWRR9VE2AXC6X7CPCj0+0GZfLhadyjvZ/Cmp561Vn3Yr82+12PDw8Ltl+UeSk7g+o0hT9DJXlXPQfoZfE76Ivi36N0rcMyqRSyFQNp2pHsmgr4pqoL6PRKPWmS1l+FpuCxDs1aLie+CwddClcL9KntfQvMdTKC0VhqgdFoTgFAROKVyhQlyoAqVPlbyc+2+12+Yywcjidzn6O7XYlnIOAzWaTA6DNZpOKVeRL5HEgIeQg8iHKI/KBanAW73ap/O5cl1hmcihhZoRsrxTqdKxWqxxY1O8VhE/IVshI/U5xv0u19Iiygceo+CkZld2VKBtKUOrlRoKQjyizILooZRWET8jYYrHI9upQAuyi9BG9KhyJUyEW4r5LQfQlh4ooig0hok05FHIm6kmdnnheEHeXy4W3tzcmk4nu7m50yokjNpsNLy8v2adcSh+81GTtekPkE5U1XZTNpbRfdZsVVmVBokT51c8JWQmZo+qrahkLCF0hdBBKP7YqJ6eIfiPyqJ4MiLRFvajJJEqdiL4n/pvNZpyKfhOWWZEPDRpuFmit/UsOtRIVytPT01MqO3EPyhKFWrkKhSeUqzj+yOFwyHAFQokLJalTLGRCEapnzw7FkuXh4SFJh/o3sSNW5GcgIBSzUVkCEwORuCber86n+CyIgciTIBTiWU9PT0zKLuWrgXiPyIfIl3oQExY7NYGzK0GenQpxNSlHzYm8Coh6EWmJQVC880aCIGvq+nYq5EI9OIu6FG0ON0sdSn8Qn3UKSdB/xs5yIU+Rvvgu0rUr7gXq/Il365WTWsRzot+I+4RF3qm0W0FQxOTpP7lEr9Ytomx2u10GZRcEW92e1fXhVE2iRLlFGQXUfdKoWOlEm1bLUPQfkxJCRqRrVcIfiToV+RPvFmmLP5S6EXkWsnapiLu6nah/16DhZoBG+r7kEIpLKFqhVIVSVF9zqJaAdarlTXG/CFEgiJv4LJQ/ipK1KEePCaWoUwZYoWjF/WJgRDVrF4p3oCDy51BZFcQgqh6AUMlK5F2UUSh0lzKAoQSdLSgooL29vV8alwvxDvV3h0KKhcycChER8jcoVj6DEm5GyFOUQwxiOpX/k7oe1fep330jQJRb3YZEvQiZqtuiIMpCPuoBX91WAc6fP09hYWG/a2qIuhT9TJADl8riZVOWfUX7EvlFlXeRX/VzIk2XG9kwmUzSr+yLqkv1e9R5F78JoiQsj4KUqScYoj2L8qhlJn5Xl1PUWVdXF4WFhVRWVkorrqgP8YxLWXYVctO5+QeL6+Kaug7E+5zKLmmRpkjDqRBacd2hLAeLfF+qXWjQcCNiYEdoDQMOobiEkhWKTU2u1MrZXZnrlRm2GLDEb+KzUMxqS5SPj48ctAQMbkRF+CCJgUSnsgQOtAIVA4Moh8iXKK8og7pMahmJwcmlEEG9Xk9ZWRkrV66ksrLyqsKfiDK6VGRPDI5CvkJW6ny4VIRByF3IWq+yRgm5CnmKe9XvvVHgUhFovYrYiTpVl9n9Goq81c8blEmBSOfYsWMcPHhQxql0h5CvS7EGi8mR+M3d+iT6oviurntRFvGb6Jsiz+I38V393BcB9XvUctQrcSw9lJh7alkL2Yg26l5WVO1ULUs1urq62LlzJzk5OXJ5VUzKBIxGI15eXuiUSQ+qyaToO7jVt4DIn07Zga13mziYlADy6nLoFKIuyqNBw80AjfR9ySEUqVCqRtUSifjzUI4/Eve7P6uGUIbq38U1dTpG1a5W9T16t2UydV5Mig/bQENdVtzyoB64xD0GhaCqfxdlEM+ZTKZ+QZWvBmJANKpItVpW4jqqevm0a2qZohqkxH3iT/2uGwXquhXfRTmFPNXyVctFwOBmDVT/ORS/S3cZu0OdD/V73a9/WjrifnVe3NuG+E2dP9GHrjfU5bgUxG+CZIlyiu9qcqQuk/jvDndZdnd3Y7PZPneSpVcIqF41WULpJ0JW4jcBIVuDinyKPLvXoagjnaI7L5V3DRpuVNw4I4cGDVcAMShoCv/GhLpeL0USNHzx+DRyqEGDhi8OmibUoEGDBg3XHRrh06DhPw+N9GnQoEGDBg0aNNwE0EifBg0abji4O/pr+M9DqxMNGv7z0EifBg0abjhoS4lfPmh1okHDfx4a6dOgQYMGDRo0aLgJoJE+DRo0aNCgQYOGmwAa6dOgQcMNB81/7MsHrU40aPjPQyN9GjRouOGg+Y99+aDViQYN/3lopE+DBg0aNGjQoOEmgEb6NGjQoEGDBg0abgJopE+DBg0aNGjQoOEmgEb6NGjQoEGDBg0abgJopE+DBg0aNGjQoOEmgEb6NGjQoEGDBg0abgJopE+DBg0aNGjQoOEmgEb6NGjQoEHDdYWI0afF6tOg4T8LjfRp0KBBg4brCnEah3YqhwYN/1lopE+DBg03HDRy8eWDVicaNPznoZE+DRo03HDQlhG/fNDqRIOG/zw00qdBgwYNGq4rNJ8+DRq+HNBInwYNGjRouK7QfPo0aPhyQCN9GjRouOGgkYsvH7Q60aDhPw+N9GnQoOGGg7aMeOW43qRMq5MbE9e73WgYWHwpSJ/L5cLhcPRrPOKz0+nE6XT+22c11EsHl/r9WnA16V2qE1yqbOLz1b5DyA0lnw6HQypW8Q6dTnfJP/H8leJqnrkWiHx+Vn6FDMQ9Tqez373u3wcC6vyo079UPi+Vp2uF6Avq+nZ/t3iv+ruGmxeXah/q7+76dSDby0D79F2qbX9WH1Nf/7Q8fNbzV4JLpeEuazWE7kb17KXKd73hnkchD/WYgduYIvL+ReRPw8DgS0H6uERHFN/1+v+XRfcOYLfbZaMUfzqdTt4/UHDP12fhUp1GrUjcfxfXrgXuaYoOqNfr++Xd5aZE9Ho9er3+krL8LKhJxhcB8R53xaOWq3tZ3b9fSR1eDYQsRfsTn9V5HmjFKMrkcDj6pe1ebtEfHA5Hv/rXoEHoJ9GXDAaD1AkD3V5FWgOZphqirat1hIBLRQjVZfs8XabWMVcK8U53/e8+jol71HrCXY/wBegwNYScxPudCrFzOBzY7Xbsdjt9fX1YrVYMBgNGo1E+p+HLj4FlSFcJ9eB0qYFJ3UndO4Jer8dgMFy3TiE646UatM7NeobS8MUzqAiI+po6r6JzXync3ytk0dvbS3t7O83NzeTn55OdnU1OTg45OTkcOHCA/fv3k52dzfHjxyktLaW+vp6Wlhba2tqwWCz95OuusNSfvyio8yLahk6nkwOUKLe4x/0ZUR9qmQ8E1DKwWCy0tLTQ0tJCa2srzc3NNDU10djYSEtLi1SOV1PPl4K6Lwilq05flFcoay5BhG90fJFt9L8F6n7gUkieuu2I3weqnbpjIOtE5FPoJfe01fpCr9djNBqlrr1UXxH3qfuIe5pXA5E3974n8i/yA2Cz2bBarXR3d9Pe3k53dzd2u73fc9cTQhYuFRm1Wq20tbVJXVZfX09RURFnz56lrKyM7u5uuASh1fDlhc51DS372Wef5dlnn3W//Kn4vFepO6mAaExiNuEOl9vs6HpAnb7Ij91ux2azgZJfk8mEyWTqp1QvpUDEtYEiJE5llij+qqqqKC8vp6amhpycHI4cOUJvby8GgwGn04nNZsPT05PU1FRGjhxJYmIi/v7++Pj4EBYWRkREBL6+vphMJqmUdDqdJN5CQX3RcCkzT5fLRW9vL21tbbS0tNDc3IzFYsFgMGA2m4mPjyc8PByTydTvOTV0Oh2nTp3igw8+4Otf/zrDhg373LbpDtFWu7u7OXPmDMeOHaOhoQGU9mC1WtHpdCQnJzN58mQGDRqEp6fnFb/nUnAp1tmOjg7a29ux2+14enri5eWFl5eX/KzT6bDb7de1b3wZ4VIGWoPBgM1m44MPPsBisfCNb3wDs9k8IHXw3wqXy4XVapXtorOzk7a2NtmejUYjQUFBBAUFXVNfV8tYr9fT0NDAu+++S1hYGF/5ylfw8PAYkHpw16vis0sZF2w2G729vXR3d2OxWHA6nf3K5ePjQ0BAAJ6enhgMBtl2RBruaQt81m+o9L/QnU6nsx+pFL+Je8VkvbGxkdbWVqxWK/Hx8cTGxuLl5SX77+e992oh0lXrCbvdTn19PYWFhdTW1sr+dPHiRVpaWoiLi2PKlClkZGTg5eUlxyINl8aVymbv3r3MmDHD/fI140tH+kRncCkDW1dXF21tbVitVpqamqitrcVms+FwODAYDISHh5OUlERMTMx1a3h2ux2DYk1sa2ujvr6e9vZ2urq6APD09CQwMJDQ0FD8/PzQ6/U4HA5JPGw2mySFQhnpFGvVteZVraT0ej1Hjhxh9erVFBYWytnYpRASEkJiYiLBwcGSJERERJCQkMCgQYNITU0lNjYWs9ksyax7/Vxr3i8Xot1YrVYaGhooKyujoKCA06dPc/bsWSorK/Hw8GD48OHcfvvtzJgxg/DwcDw9PWWe1cpSNwCkT6cQqvz8fN5//31WrlzJxYsXMZlMOJVZsq+vL3PnzuWhhx5i4sSJA0o46uvryc7OJjs7m7KyMoKCghg6dCjp6enExsYSExNDaGhov4nIzQLRJzTS1x86hYA0NzfT0NBAY2MjeXl5HD16lMbGRnQ6HeHh4cycOZMZM2YQHR191cRPLWO9Xk9jYyMrVqwgPDyce+65Z0BJn1onicm01Wqlo6OD+vp6Lly4wIULFygsLKSlpUVOHr28vBg6dCiTJk1i+PDhhISEoNfrJUH7PLjrQPFul8tFT08PHh4emM3mfs9YLBba2tpoa2ujq6sLp9OJxWKhoaGBixcvUlFRQWVlJQaDgfnz5zNv3jzCw8NlPbi/cyBht9tlnVgsFjo6Ojhy5Ahbtmzh9OnToEwK2tvb6ejoIDU1la985SssWLCA6Ohoma/rlb//dlypXG4q0qfX67Hb7fT09FBaWsrZs2eprq7mzJkznDp1is7OTmw2G0ajUQ70s2fPJj4+Xj4/kHAp1iKLxcKJEyfYuXMnxcXF9Pb2gjJbjIiIIDY2Vg60drtdkjyXy0VERARRUVEyb/7+/gQEBEhieDVwJzR6vZ6NGzfyu9/9jmPHjklLJEqDi46OlgrYYrHI3/SKRc/Hx4eQkBDS0tIYM2YMQ4cOZciQISQnJ+Pr6yvvF2UaaDlfCkKBNzc3c/78eY4fP865c+eora2lvr6e6upqOjo6cDgcREZG8o1vfIO7776b9PR0/Pz8JMFWy0h3jaRPDJ7V1dVs376dN998kzNnzhASEkJYWBgGgwEPDw/8/f0ZP348S5cuZfDgwXh6eronddWoqalh69atbNiwgSNHjmCxWOQEKCEhgeTkZDIzM8nIyCA+Ph6j0fhvpP3LCveB9FLX3b+ryyPap0b6/n8IWVksFs6ePcuePXs4ffo0xcXF5Ofnywmsn58fX/3qV3nggQcYNWqU1GFXCvUz+i/I0me32+ns7KSyspLq6mqKi4spKSmhqKiI8vJyqqqqpM4WSE1NZcGCBXzlK19h2LBh+Pr6XpXhwKVM5u12O3V1ddTU1GAwGAgJCcHT01OOHw0NDRQXF1NUVERdXR1WqxWHw0FnZyddXV24XC46OztxuVwsXLiQe++9l8TExOs6eROya29vp7a2lpKSEkpLS6mpqSE/P5/Dhw/T3Nzs/hhRUVF87Wtf4+tf/zpDhgzBaDR+YePCfyOutP5ueNInlLtaMN3d3Rw9epTVq1dz6NAh6uvrqa2t7ZeO2WxmxowZfPOb32TOnDkEBwfL3wYCLtXyckdHBytXruQvf/kLRUVF8h6TyURYWBgxMTH4+PigVyx9wvLodDqJi4sjNTUVT09PdMqy34QJEwgMDLyqTiIGNr0ya9XpdNhsNnJzc3nzzTfZt28fra2t+Pn5YbPZsNlsZGZmMnjwYOrr6zl37pyc9YrlHUEIPD09CQ8PJzk5mXvuuYdly5YRGhoq36O+dyAgZOTuEKxXfBRLS0s5fvw4+/fv5+TJk1gsFmJiYoiPj8dkMnHhwgUOHz6My+Xitttu47777mPs2LH4+/v3Iw0o7Ut3BaTP4XDI+lGXt62tjR07dvDPf/6TM2fOEBcXx+LFixk1ahQhISHSVyowMJDIyEi8vb1VqV47Ojo6yMvL49SpU5SUlFBXV0dVVRUtLS24FMKTnp7OrFmzmDJlCklJSTIPNpsNg8Fw1Vacy4GQu5CdkK/uM6zEom2J/Ksh3ClEmzcYDP2W8N0JrUjjP0H6LlU+dfnV19zvGwiI8usVq5VBce1obm6mqKiIXbt2sXv3bs6cOUNXVxd6xR1B5PGee+7hBz/4AePHj//SkD71+KCuY/G5vb2dyspKKisrOXHiBMXFxRQUFFBRUSFdIASEzhTvnzRpEk899RQzZ87Ez8/vikifug6dTiddXV1s3bqVbdu20draSkRERD/9UV9fT0NDg7TyBQQEEBUVRUREBNHR0SQmJuLj44PD4SAiIoKUlBT8/f0HjPS5tzmdMm40NjZy6tQpTpw4wcmTJzlz5gw1NTU4HA76+vrQ6/V4eHhgsVjw8PAgJiaGwYMHs2jRIm699VYSEhKkXDVcGldaf9eL9BmevRLW5oaBzJRT5fPgUhR2X18f586dY/PmzRw6dEj6oQjrDcpg0NLSQkhICCkpKQQFBX2q/9/VQihOi8XC4cOH2bFjh7QuORS/kfb2dpqamvqZ6Kurq6moqKCsrIyysjJOnz7NkSNHOH/+PADDhw/H399fKrIrgfszgmiaTCbMZjO+vr6EhIQQFxdHVFQUgYGBjB07ljvuuINx48YRGBiIh4cHoaGh9PX10dnZCYpSELO+rq4u0tLSyMrKwsfHRypDh7K0PlBQK3C1Au3o6KCgoIBt27axceNGioqK8PHxISsri1tvvZUFCxYwbNgwuWnF4XAwbNgwRo8eTUxMDB4eHqDqbOr/dXV1nDt3juHDhxMREaHKzaXhUg2iXV1d5Ofn88knn7Blyxb0ej2zZs3i7rvvZsKECSQnJxMTE0N0dDTBwcEyHwMFnU6Hh4cHISEhJCcnM2LECMaOHUtGRoZcZmlsbKSiooLGxkb0ej3h4eFygiGIrLr9XGn7+yyoFb9a5uqBVv2bU9lwIvLW0dFBY2Mj9fX1cmPMxYsXKSkp4cKFC5SVlVFXV0drayudnZ10d3fjcDiks76AqK9z585ht9sZMWLENVnWrxSfJtPrJXcBl2J1UpON1tZWTpw4wYYNG1i9erV0wlcT6ZCQEIYMGcK0adPk5EUtz6uFTqeTvq8+Pj5kZmZetf5wb7c9PT3U1dVx+vRptm7dypYtW8jJyeHcuXNUVlZKcqV+RrRBg8GAp6cnYWFhzJw5k5SUFGkRv5xyq9uzSN9qtXLu3Dn27t3LgQMHKCwspLKykqamJkl0Q0JCGD58OKNGjWLkyJGMHDmSESNGkJWVRWZmJikpKaSkpBAREYGPjw8GxQ1oINqK0LPiv8PhoLa2ltzcXFatWsX27dspLCykpqYGq9Xaj2yL1YvBgwcza9YsFixYwLhx44iPj8fDw6PfvRr+HVcql/LychITE90vXzO+NKRPCEQ0bpfLRXNzM8eOHSM7O5vGxkapsIODg4mNjcXf318q/KCgIBISEoiOjsbb2/uKBfxpEPnRKSbwlpYWuru7sVqt0jFYNHhhqrdardhsNvr6+ujr68Nms9Hd3U1bWxutra10dHQQExPD5MmTL4twfB7USszLy4vw8HDi4uLIyMhg2LBhDB48mMzMTMaOHcuoUaNISUkhKSmJuLg46b8XERGBv78/DoeDnp4eXC4XNpuNUaNGMWbMGOmrKOrmcpTi5UKnLMk7lJ2mOoW0HDx4UCqihoYGRo4cye23387SpUsZP348gYGBNDU1cfz4cU6fPo1er2f69OlMnjyZ4ODgfxsg1G3sckmfKC+KfHt6eigsLGTDhg18/PHH9PT0MGnSJG677TbGjx8/4Ba9T4NOp8NoNEpyHxoaSmhoqKx3oSzKyspobGzE29uboKAg/P398fDwkIRItO2Bhlr27gONXmVpEf2qq6uL+vp6Ll68SF5eHtnZ2eTm5nLq1CmOHTvG/v372bp1K5s3b2bbtm3SmpOXl0draytms1n6p4r6+iJJn/vg71L5lwkZu8vaXT4DDdFHGxoaOHToEGvWrGHLli2UlpZitVrRKxZTHx8fUlNTWbp0KXfccQdTpkwhPj5eLkteK3Q6HT09PZw+fRpfX9+rJn0iLw6Hg97eXioqKjh48CBbt25lx44dHDx4kPPnz1NVVSWXSvWK64pe2Wzn5eVFWFgYgwYNYsyYMZJ4jR49WuoBd73xWXCvXxRDhMPhoKGhgYaGBuLi4pgwYQKzZs1i0qRJzJo1i/nz5zNr1iyysrLIyMggISGBsLAwvL29MSn+36Kfureba4W63bW3t3P06FE++ugjduzYIZfABYHz8/MjNDSUmJgYhg8fzoIFC7jtttuYOXMmo0aNIjo6Gi8vr39LW8O/40plc8OTPtw6kF3ZOXT27Flyc3Npa2vD4XAQEBDA7NmzmTZtGhEREbS2ttLT00NISAipqakkJibi5+fXr6MI5ctVCB434ufr60tcXJzcOBIQEEB0dDRms1nm8dPg4eGBTlk+TU9PZ+rUqYSFhbnfdllQDygCOoUI+Pj4EB4eTkpKCqmpqSQlJUnrk1h+Dg0NJTk5mdTUVFJTU0lJScFkMtHU1ERDQwMOZcdZUlISQ4cO7bcjdqCVkBouhezn5ubyr3/9i02bNqHX65kxYwZ33HEHM2bMICkpCQ8PDyoqKtizZw/btm2jvLycyMhIZs6cyejRo/+NpKplpbsC0udOIKqqqtiyZQtr166lvLycsWPHcv/99zN16lQCAgL+zdr0RUGv12M2mwkJCSE+Pl5aClpbWzl79iy1tbX4+/sTHx8viakgfQOdX3XbEIOH+28uxZqv0+no6OggPz+fjRs38sknn7B9+3Z2797N2bNnpQN+QUGBtN7U19dTV1dHdXU1tbW1mEwm4uLi5E5H9zr7Ikife58QchV/4nfRHt0J90D2J3W9OhwOzpw5w/r161m3bh0XL17E29sblzKpi4iIYPr06cyePZsFCxZIwmc2mwcsTzqVpc/b2/uKSZ9Tcf/Q6/XYbDa5o/TAgQOyzRw7doza2lp6enrQ6XRyxSMoKIi0tDRGjx7NpEmTGDduHOPGjWPKlClMmTKFsWPHMnToUBnFgCsgfeoxRdyv1+vx8/MjODgYm81GU1MT8fHxzJ49m6lTp0qCJ/y51X9qi971aBcC6nbY1NREdnY269evp6amRrYbgNjYWEaNGsX06dOZMWMGkydPZsaMGUyYMIH4+Hh8fHz69afrkdcbCVcqn5uC9KFqkA5ll1lBQQEHDhyQ8YDS0tK4//77mT17NgEBAZSWllJVVUV4eDijRo0iLS1Ndl6RnnvaVwJBFsQg5efnR2RkJImJiSQmJhIfH09ycjJxcXF4eXlhMpkICgoiMjKS4OBg+VnkqaenB39/f0aPHs348eOvmvS5KxyRR5cyuxVKRCxhiB26YiAUg5HIs91up6Kigvz8fKqqqmT63t7eJCcn9yMLVyrDy4Ve8S88dOgQH3/8MRs3bsTlcjF79mxp3QsPD8disVBVVcX+/fvZsGEDR48exeFw4OfnR1paGmlpaQQHB/+bQhL51l0B6UNFHqqrq8nNzWX16tWcO3eOiRMncu+99zJlyhRiYmIwKL5T10s+argTWfVnp9OJt7c3vr6+9PX1ScfxkJAQRo4cSVBQkGwvapkMNET67uQHlTuCmNht27ZNbkppbW3FqVjQxfKjmGCFhYXJ3fphYWGkp6czcuRIhgwZQnh4eD/XDlFvXwTpU0OUWV0fdrtdbp4yKKFBxL0DKXvRLkRbbGxsJCcnhy1btlBRUSFJuLe3NykpKdxyyy3MnTuXiRMnMmjQIIKDg9G5udBcK3QDsLwrCFFTUxOHDx9m48aNbNiwgZMnT9LQ0NBvcmE0GomNjWXIkCGMGzeOefPmsWDBAhYuXMjMmTMZN24cI0aMYNCgQcTHx0v9bDQaZX2o6+ez4F5/Op0OLy8vgoKCsNlsXLhwgZaWFoKDg6U1T/h6iyV4dV9W/6n750BBpCfe0drayuHDh8nOzsZisaBT/GrDwsJYuHAht99+O/Pnz2fixImkpaURERGB2WyW/VnD5eNK6/KmJH1tbW3k5eWxf/9+qTDT09P5xje+wfDhw/Hz88NkMhETE8OoUaMYPnw4oaGh2O12uYzqcDjw8PC4aj8/9w6t1+vx9PTE39+fyMhIEhISSElJIT09XfpWTZ8+nTlz5jB58mRmzZrFjBkzSElJkSFHRo8ezbx58+RusauFkJU7hMJwqIJaqwmJkIVd2SEtNkIcP36c8+fP09raKtOKiopi2LBhpKen4+vrK9O7HgrJ5XJx+vRpPvzwQ/bs2YOnpydz585lwYIFZGVlERYWhsViobS0lP3797N+/XqOHDkiJwTd3d3o9XpiY2OltcI9n0Jml0v6xP2dnZ3s3r2bjRs3cvr0aZKSkvjqV7/K/PnzCQsLk0thrgFe+v4suJdL/Ncry3ZeSqy+pqYmiouL8fPzY+zYsYSHh/drO2ISMJAQg7Bodz09PTLgrF4JlltVVcXevXtZv349e/fuxeFwkJaWxrhx4xg9ejTp6ekMGzaMrKwsxowZQ1ZWFuPGjWPChAlkZWUxceJEZs6cydixY/tNSgSuJ+m7FClwKtZ3Icu+vj66urpob2+nvb2dixcvYrVa8fPzk6RnoPuQGIzb29tlOJaNGzeSk5ODxWLB5XIRGRnJwoULefDBB1m6dCkjRowgOjoaX19fKTMGMG+6ayB96v7b2trK8ePHWb9+PTt37uT8+fPSsqdTdFxgYCBZWVksXrxY9s+RI0cSr8TvFC4OPj4+eCkxLcXmOnWfuFT9usP9fvW9wuWnvLycU6dOUV5eLl0QxEYvu91OU1MT58+fp7y8nLa2Njo7O+XyqtgM9ll5uFa0t7fLDRw9PT3yenx8PEuXLmXy5MlER0fj5+eHj48Pnm7xDMUYM9D640bEldbjTUH61J3I6XTS29tLcXExx44dkxsNYmJimD17NrGxsfj6+pKSksKECRMYNWoUAQEBtLS0kJ+fT15eHqWlpXR1deHh4YGXl9dlK5pLQd25RQf38PAgICCAwMBAQkJCiImJITk5mYyMDFJTU0lOTpYhT/R6PaWlpVRXVzNmzBhmz54tt+JfDYSs1MpGLT8BQUREQFa9skRiUHY25ufns3btWtasWcPhw4dpaGjAYDDg7e1NeHg4kyZNYtq0aQwaNEhuSFDLYCCgU0h+a2srOTk5fPTRR7S1tTF79mzuuecexo8fT1RUFHa7ncLCQj755BPWrl0r24VOtZTl6elJUlISQ4YMkT6K6nwKGV0u6ROyKy8vZ/369eTm5hIQEMCCBQuYNWuWXOYQ9xkG8OSNz4J7XYvv6usGgwGHw0FVVRV5eXkYDAaGDRtGZGSkJP7i3oHMsxgwRdtra2ujrKyMkydPUlpaitFoxGQycebMGTZs2MDp06eJiopiwYIFLFmyhJkzZzJhwgSGDx/O6NGjGT16NMOHDycjI4OMjAyGDBki/VKTkpIICgrCpOxu1KmsNJ9F+sQ9aghZiLb9WX+XuseqnF4gwl6cOnWKI0eOcOrUKYqLi2lra8Pf35+wsLB+k9CB7EtOp5OmpiaOHTvGhg0byM7O5tixY3IiFxsby8KFC7nrrruYPn06cXFx+Pj49GsPIi8DlSfdNfj0ifz09vZy4sQJNm7cyO7du6XVUhCO+Ph4pkyZwvz585k/fz7Tpk1jxIgRxMbG4ufnJ4mdGp/XBtTfLxeizYt89fb2yl3FTqeT0NBQoqOj8fT05OLFixw4cIBPPvmEffv2UVRURElJCZWVlXR0dKBTJuhisj6QfdSpWBo7Ojo4ffo0hw4dore3V+bfx8eHpKQkjEYjxcXF0mLZ09NDZ2cnHR0d8iQnu7LBUhBU3RVYSm8mXKksbmjSp+4kqGbMJpOJ1tZWiouLuXjxIk5l2SolJUVu2PDx8cHPzw+j0UhpaSm7d+/mk08+IScnh6NHj1JVVYWnp6ecyV4r1EpR5FmvOg7Lx8cHDw8PDIqDtEE5KaKtrY0TJ05QVVVFfHw8kydP7mdxcVf8n6WQPu8abs/39vbS3NxMS0sLFouF9vZ2Ojs7qaioYMOGDWzfvp1z587R0dEhZ/lhYWHMmjWLhQsXMmHCBAICAmSaIq+f9u4rhU6nw6LEQNyxYweHDh0iNTWVu+66i5kzZxIWFobJZKK5uZl9+/axcuVKDh8+TF9fHyghc+x2O15eXowcOZLJkyeTkpKCn5+fzLNetdymu0zSJ8pYU1PDrl272L59Oy6XSw4s6enpmM1mjEp8Kt0AK+bPgrrPqNuj+EPJf3t7OydOnODMmTM4HA4Z6iYoKEgqa3UaAwWRngi5s2/fPo4ePUpfXx/R0dHY7Xa2bdvG7t27CQoKYunSpcybN08O1JGRkYSFhRESEkJISAiBgYEEBQURGBiIn58ffn5+8uSYT2uLn0b61DISUFu3xIYsm82GXXUutfpecc2uxIbr6emhvr6eY8eOsWPHDrZt28amTZvIzs6mvLwcl8tFcnIy6enpcnldDffvlwt1Xet0Ovr6+jh//jxbtmyRVmkRmikuLo6ZM2eydOlSxo4dS0hIiExDpxqo1dfcP18NdNfo06dT9MPOnTv54IMPKC4uxq4EEjaZTCQlJTFz5kxuu+025s+fz6hRo4iLi5MEX+TdvZ27f/6s7+5wufllut8vJqDCxaK2tpby8nL0ej3e3t709PRIXdfU1ER3dzd1dXUUFhZy7Ngxzp8/T1NTEz09PfT09GA2m/Hx8ZHv5hrajIBLCSJdUlLCuXPnaG9vl2n39fXR0dHB+fPnOXDgAMePH6eyslJGBRAuI+Xl5TKuYEBAgDQMqPuTeNe15ve/HVda/hua9AnoFMUjSJROWZoS8Zb6+vrw9PQkNjaWqKgojEYj9fX1lJSUcPLkSXbt2iUVrfD16+3tlZafgIAA91deNUQnFwOAUChqxalTNqT09fVRWlrKwYMHqaioICkpiXHjxhEREdHPfO/eKMRsDDdLjOtzLG1Chihp1NTUyIHo1KlTHDx4kJycHPbu3cvu3bspKSmRy26iTOnp6Xzzm99kypQphIeHy7S4DIV4pdApzvyrV69mzZo1uFwuli5dKgNuC1JXVFTE1q1b2b9/f78gq8IyOXLkSObMmcO0adOIioqSA7x7fnWXSfpQQjCcOXOGtWvXcvr0aYYNG8Y999xDVlYW/v7+/QYvnWpnqvp9A6Wk1XAvk/qaeL/L5aKqqooDBw5I4pOSkkJGRgahoaEyr8IC7J7etcBqtdLe3k5ZWRm5ubkcOXIEl8vFiBEjGDp0qLT8GI1GbrnlFhYsWEBKSoqcKHGJMorP6vav7m/ucCd9w4cP70cE1H/qZ7q6uqioqJDWDZvNJvu5TYl5KQbFiooKDh06xLZt29i8eTP79u3j8OHDXLhwAYAhQ4Ywc+ZMpk2bJv0OPZWTYi7VNq8U6uU1h8NBY2Mj2dnZHDp0iLy8PDo7OzGZTKSkpDBz5kwWLlzIxIkTZZ8WcKk21og/MSm4lvyhyPpql3dRnu/r6+PgwYPs2bNHrvqgWoZcuHAhI0aMIDw8XE4GBER7UevQgYC67bm3U73iwuCjxNyrra2loKCArq4uent7qa6upqSkRLpciPBdIqxLe3s7paWlVFRU0NbWRlBQECEhIZ85Xnwe1HpI5NHpdNLe3k5LSws1NTX0Kkd22mw22trauHjxIlVVVVRVVVFRUUFFRQUFBQXSgn3u3DkKCgqw2+1ERUVJ1wW1Prke+u+/EVda/hua9KkVi7oj2Ww2ampqOHnyJBcuXJCkTzjpNzQ0sHfvXtatWyf9VoqLi7FarTLdpKQkJk6ceNX+c0Ixfx5EGdQNXK2Ijx49yt69e6mvr2fo0KHSR819Wcq9Y4p03OWjltmlIH5zKjtO33nnHf71r39x7Ngxjh8/zqlTp8jPz6empgan0ymJlcvlIioqijlz5rBkyRISEhLQKURQkATHAMfp0ykWqe3bt3Pw4EECAgKYOXMmQ4cOxdfXF5vNRnNzMydPnmTnzp1cuHBBKixPT08GDRrE9OnTWbp0KbfccguDBg3CbDbLgdVd2esug/QJOdfU1Mg21tLSwoQJExg/fjz+/v5YLBa5fKyuD1GX4vPn1dW1wl2pivK2tLRw7Ngxtm/fTn5+PgkJCdxyyy1kZGRIK6hoX+4yulqIsnd0dFBYWMi6devYs2cP0dHRzJs3j/HjxxMTE4PJZCI8PFyGEYqJicFoNEorm5CZIFtqvXC5cCd97su77nWiU6x8Fy9e5Pz58xw6dIijR49SXFxMeXk558+f58yZM5w+fVoeOr9371527NjB9u3bOXXqFBaLhfj4eEaMGMH8+fNZtGgRU6dOZfDgwdL3U5RL9OtL5eVK0dPTQ0FBATt37pTHZtXX1wOQkZHBnDlzWLx4sdwMhSrwuLrdCqjb7LXmTXcNpE/Ixmq1cvLkSQ4fPtyP9AkXH2EE6O3txdPTU242EHm/lB64VqhlJL67Q7S35uZmGa+1oqICh8Mh62XWrFlkZmYycuRIpk2bxpQpU4iLi8Nms1FSUkJ+fj5Wq5WQkBD8/Pyueme1uq0JeRgMBjo7OykrK+PChQtykuBQThcRbkEOJSatl5cXfX19VFRUUFVVRU1NjfRH9Pb2Jjg4WIbLUuvfq8nvjYYrlcENTfpwU8Liv02Jb1dVVUVtbS2tra309vbS1dVFcXExhw4d4vjx41y4cIHKykpaWlokgfHw8CAhIYFZs2ZJ3yther4auFeY6BAOJWZUT08Pvb29Mn8WiwW7En8sPz+f7du3s2/fPtrb26VfktgY0dvbK3dOmUwm2Ul0lyB8l9OJxG9OxTInnHWrq6vlTNOqxBIU9xsMBuzKGcNJSUmMGjWKoUOHEhwc3M/yItK9XKV9OdApPju5ubny+Dg/Pz+6urpkhP1Dhw6RnZ3N8ePH5QYdT09PxowZw+LFi1m0aBGTJk0iMTHxko7ZannpLoP0CTQ2NnL69GmOHz9OS0sLZrOZ5uZmSktLqa+vp7m5WVrKTEroBZ0bgVfjs+rtanGpctrtdmpqajh48CA7duygoaGBzMxMli1bRnJyslTsKJbSgYJOGaAvXLjA7t272bBhA01NTdICGxAQgFWJcent7Y23t7dcRjp9+jQFBQV0dHTg6emJt7c3RmVHpUjbvZyfhU8jfaJuRHp2JRh5hRJIvauri87OTrZt2yZ9R48dO8a+ffvYvXu3tKTt2bOH3NxcWltbCQwMZPDgwUycOJGFCxcye/ZsJk2aRGpqKkFBQXiqYt6p363+fiUQA6pB2ShTVVXF5s2bWbFiBSdOnKC5uRmXsqQ8f/58Fi5cyOjRowkNDZXWIoPBIK1ODQ0NtLa2yv8OJdC72rLEVeZVd40+fSL0yb59+8jNzaVPOSECZYzo6emhoaGBqqoq6uvrZZsRk1N3/TUQEPWn/nOHU1n+9VQiJ+h0Oi5evMiFCxcwmUzMnj2bsWPHEhUVhZeXF76+vvj7+8u4m7Gxsfj4+Mhz1A0GAzExMdI94GpJrDqvBiX2aG1trQx67nA4MJvNpKenyx3OCQkJjB07Vk5gzGYzPT09dCsBvuvr69Hr9YwePZqEhARQdNC15PNGw6XayGfhhiZ97kpYp8yEe3t7aWpqks6tTU1NctmopqaGmpoampubJYnR6XRERESQlZUlg2DOmDGD4cOHS2f7q2mA6soSebVarTQ1NclYYRcvXqS6upr6+nqamppobGyktraW4uJi9u7dy8aNG6mvr0en08mO26v42l28eJGmpiZJHByqAM/uA4R7fj4NIp+ivJ7Krqv29nb6+vowKDuwhKOwIIAeHh44lNMNQkJCCAgI+Lclt88jnVcKnaLYz549y7lz52hSYgWeOnWK48ePs3PnTvbu3cvp06dpbW1Fpzg4Dx06lLvuuovFixczYsQIQkNDZdsR+XOXnbj2eaRP/ZzwqxEkXpwBLDYLNTQ00Nvbi8lkwtPTE6MSq8/93e75GCiIdF2Kn5Fer6evr4+ysjIOHjzIiRMnsFgsTJw4UR7grruEZXkgoFMG+NzcXLZs2UJhYSHe3t5ERkbiVDYZVFdXc/78efLy8jh//jynTp1i9+7d7N27l6NHj9Lc3IzJZMLb2xuDslQkBvIrye9nkT7xu8PhoK6ujsOHD/PJJ59QUlIil6nEBoyuri4ZDaCrq4vu7m5aW1vp6uoiJCSEGTNmsGTJEhYvXsyMGTPIzMwkMTGRwMDAfyM3YiKmJoBXq5f0iuVdnEixe/duGd7K5XIRHR3NtGnTuPXWWxk3bhxhYWHolXPNLRaLbMvieMOCggK5maC5uRkPDw+5ycNd3u7fPwu6q/TpE+/o6uqirKxMuqeo262asF+4cIHa2lr0ej0dHR00K+fFms1mGeh4oCH6nNDTqMieyKeHhwdBQUH4+vrS09NDc3MzoaGhzJgxQ65KiPKgEDE/Pz8ilID5DQ0NHDt2jK6uLjIzM0lNTb1mA4a6H7mU1ZuWlhZaW1vx9PRk+PDhzJ8/X4bzycrKYsqUKTLWoJeXl9y0ZFWO8AsKCmLGjBlydUik7/6+mxVXWvYbmvSJhiHgUqxf9coZsYcOHeLcuXO0trb2GwCEUkfp2IMGDWLSpElMnTqVefPmMW3aNJKSkvBXHXV2pR3fvaJE4+3s7OTw4cNs2rRJWhyFJero0aMcPnyYgwcPcuzYMXJzc6msrJSKQTj2CgvSgQMHyMvLk7NV4SDb2traj0i4y+mzIN6lV0z4oaGhhIeHExwczODBg0lRznSMi4uTVjVB/Do7O2lsbKRX2c0VHBwswzmo33+5efk8CKUgBgJh2e3o6KBDOZKrt7dXWqZ0Oh1paWksXbqURYsWMWTIkH5BeUV64rM7dJdB+lDuE87YUVFRpKSkyHN0dQq5qamp4cyZM5SUlMiQGAZlU4/7UuKl8nI1uFQduFwu+vr65ERCnLG6detWent75eCflZWFh4cHdrtd5s89zWuBTmnfBQUF5ObmUlxcTGtrq9xBLE5POX36NCdPnuTcuXPyqKquri46Ojqoq6uTzu1i0FRvmBHv+Sy4FCL1aaRPPN/U1MSJEyf46KOP2LNnD1FRUYwbN46oqCji4+NJT08nIyODwYMHM2LECMaPH8+YMWPIzMxk4sSJchf3+PHjGTZsGDExMZjNZkwmkyQDIh/qdiDaqPj988rjDp1iyRLuBxs2bODw4cMyQLzBYGDw4MHMmDFDLhfqlclAXl4eGzduZPPmzezatYvs7GxOnjzJyZMn5UaCkpISvL29iYuLw9fXt187u9K2rFNI3+nTp69oeVfISBClM2fOcOHCBem+I+BwOLBYLHR2dtLQ0CDDoOTn58tVA7HB7kry/XlwXYLwoeRbfBd1q1OiPXh5eREdHU1KSgojRowgMjJSEjhRp6JNGI1G+Vt9fT2NjY2kpaWRkZFxTWVxz6tROdknMjKSyZMnM3/+fGYoQZgzMzNJS0sjISFB+tL7+vrS0dFBTU0NpaWl9PT0yBiwkZGRREZGSnmL/nqpd99suNKy3zCkz6Hy1xEQys+qHA1ktVrlWbX79+8nJyeHqqoqOYMSncOpxDKKjo5m5MiRTJ8+nUmTJsmjbdRHsumv0W9JdHBUxxqJaPDHjx+nqKiIvLw8CgsLOX/+PIWFhfKc0MbGRplXp3KmrAjrIM4TLS8vp6qqigsXLnDmzBkZJNnpdEpiJt6vu0w/IFFeg7LRIUQ5n3j06NEMGjSIlJQUMjMziYqKQq/X09LSQm9vLzplubWyshKDwUBUVBRRUVF4u8VB+7z3Xy5EWYRy9vHxkaeHdHV10aOKxaXX60lPT2fBggUsXryYtLQ0vJUTBgTclZo7dJdJ+gTMZjNRUVGkpqbKwX/MmDEMGTKE0NBQWltbuXDhAhcuXKC0tBSbzYaPj4+Muv95+RG4VJ1e6pr43tPTIy1nZWVl5OXlyYnEkSNH2LVrFydPniQoKIi5c+cydepUYmNjZV9Qy3SgIPJmMpmoq6sjLy+P7u5uOjo66FX8rYKDgwkLC5PHO8XHxzNo0CCysrKIi4tDp/h4FhYWUldXh145F1aQj0vJ5FIQZMud9Ak9YrFYyM/PZ9euXWzYsIGWlhZ5LFdQUBCDBg2S8T9HjhzJpEmTmDBhAiNHjmTq1KlyCTc9PZ2QkBA8lSPg1HlTy1bIW/zu/v9K4FI24Jw7d47333+f/fv3c/HiRVwul1ztmDFjBhMnTiQlJQWA2tpazp49y9atW/n444/ZsWMH586do6amhsbGRurq6mhubqa+vp6amhp5IoM4sxlVGa4Eusv06RN9WJ2+aJ+i3kNDQ0lMTCQiIoLIyEipm6KiomSEAbFT9uLFi7S0tMj7xMRhoKBThZoSLiculwtPT0/5u8g3gJeXF5GRkQwePJihQ4fK5VvDJc7WFe3IZDLh6+tLTU0NxcXF+Pv7ExMTI3X2lULd5sRng8GAv3Jaj5jgiAgZ/v7+eCtRMjw9PeVEsbGxkfz8fEpKSmTZnUpQZ+FjKUifex+4WXGlZb9hSJ+7UhSfRcfo6Ojg5MmTbN68mc2bN3Ps2DHKysokaRLPO51OfH19GTZsGHPmzGHBggVMnz6dYcOGyWCSQrGIZ65U6GqId3Z1dXHq1CnWrVvHrl27qKyspK+vT1p53P/URE2viuFkV5xj7UpICIvFIpc1q6urKS4upri4GJ1Ox4gRIyQxudyyuMvYoATr9fX1JTg4WCrPjIwM4uLiCAkJwWKxyKVKFH8Zh8NBXFwcaWlpcvfz5ebhciFk4+XlRWBgIMnJyWRmZhIYGEhvby8dHR309PRgMpkYPHgwy5YtY8mSJWRmZkorrkhHnbdPy5/uCkifU/FfNJvNUnZRUVHExsaSkJAgTwAJCwujoaGBI0eOUFpaSm9vryR+alL6aXnC7Td3GbtUFqOGhgYqKytlDMv9+/ezd+9eNm3axPr169mzZw+nTp2ioqICm80mCUxSUhIREREyZqUgP+7vvlboFAIfEBBAYmIiEydO5JZbbmHq1KnMnDmTGTNmMG7cOHmus4jDN3z4cIYMGSJjCba3t3P+/HkqKyuxWq14eXkRHBx82QO3O+kbNmyYtJDo9XpaW1vZs2cPa9as4fz581itVoxGo7RaqMPFiNNAQkND5XVfX1+8lbNS3dsel+iDl5Lxpa59HnSKLrLb7Zw5c4bVq1dz4cIFnE4nPj4+TJo0iUWLFrFgwQKGDRuGwWCgsLCQrVu3yt3vBQUFdHZ2YlPC0qCa2BqNRkJDQ5kyZQqjR48mICDgkiTtcqG7TJ8+9/Yu/oxGI35+fsTGxspzcpOTk0lOTmbEiBGMHDmSkSNHkpqaSmhoKBaLhcbGRvr6+uRpGFFRUYSEhOCjhD25Voi81tbWsn37dtavX0+VEh7MTzk0QJAd0feNRiPe3t4ytqvoh+r2Igiu+I6ig8QyfFNTE0lJSZ8qw6uFkLNB8YF0X1nSq4wmDmU38qlTpygqKpIba7y8vBgyZAgjR46UG7MuRx/fLLjS8t8wpA+VL4oYyETD6Ovrk0ps3bp1HDt2jPb2dukQisonxqXsMp07dy6LFi1iwoQJJCUlERgY+JlO01cLnaLU29raOH36NDt37iQvLw+d6oQLnU6H2WzGoeyKE2UTcCr+VqKzig6uV0z5DmWHlEOJEdbb20tiYiKTJk0iJiZG5kGkc7XQ6/V4eHjg6+uLj48P/v7+0jm4vb2d+vp6udTrcrlITExk0KBBhIaG9htwr1Wm7tDr9fgo5waLOHI1NTWUlZXR2dkpA3MvXryYUaNGScKnJtaivj8Luisgfe5tR6cQaJPJhJ+fHyEhISQmJhIbG0tvb6+09FZXV2M2m+XRS+p2fimof3OolrHVcKhCcohdsUePHiU/P1/6lgpHfLE0ajAYpA9aaGgo6enp+KuOKeQSE7FrhUuJnxYdHS3PPJ04cSKjR49m6NChpKSkEB8fT1xcHFFRUURGRkrXgyhlGV1YQRoaGjh9+jQVFRUEBgYyaNCgf7M4fxr0lyB9astrQ0MDO3fuZPv27fQpMR+bmpqor6/H4XAQHh5OYGAgZrNZWjlE3RuNRjkwOpUdxu5t5XpCr/jm5efns3PnTum/FhYWRmZmprRIent7c/78ebkp5fDhw1RXV9PX19ePmOhUhCMiIoIJEyYwbdo0UlNT+y3voiIxl1tW3ef49Imx4FLvcKpitorVitDQUCIiIkhOTmbQoEFkZGSQmZnJ4MGDSU5Opre3l9raWnqVDXIGZbUiJSWFwMBA+Y5rgaj3xsZG9u3bx6pVqygrKyMgIIAg5dQPYelSy9Y9jUt9dv/e09PD+fPn2bFjB83NzdLHbqBJn5C1TjXOiHyo68jlclFdXc2JEyekzyuK33hGRgYjR44kOjq6X/k1/Hsdfx5uKNInOoxOUZLCD0lE6BcO1eoGI2ZP6llpZmYmt99+OxMmTCA8PFxaLy5F+K5UUamhftahRDHv7OzE19eX+Ph44uPjiYmJkdaf6Oho4uPjCQkJwaX4J+p0Ory9vRkyZAjp6enExsaSnJwsnxHpJCYmEhcXR0pKCllZWcycOZMxY8b0izGoLteVQq/X09nZSXFxMefOnaO6uhqbzUZISAhRigO7cILu6+uTM38xg/X19b0mf5JPg1A4epVz/bFjxzh48CAlJSWEhISwdOlSbrvtNikPpxLHUNS5yNPn5U13FaRPKD3RFlDVgxg8q6ur5a7elpYWIiIiGKEccfVZ1imXisir25r63a2trRQUFLBnzx5WrVrFoUOHMBqNcjATUf57ld3jOp1OLkX7+vpSV1eH0WiUS9JqC4N430DBqVihPDw88FSOufJUwmh4eXlJwmRQxfMSMtQrkyJfX1/Cw8MxmUyUlJRQVFREWFgYw4cPJygoyP2V/wbRltSkb8SIEXgox2PpFOuTenNVWloasbGxdHV1UVtbS0hIiFziMrgdPXUpHTOQMrwc2JSQHsePH6dBOX/W39+f1NRURowYQVxcHA1KWKtt27Zx7Ngx6fMn5CzankuZILhcLhISErj11luZPHkyERERl+zvotyXA91l+PS5y098N6h239qVkFIeHh74+fkRGBiIv7+//O/n54dOp+Ps2bOcPXtWuoXEx8czceJEBg8ejO9VhO36LBgMBjw8POhV7YIWlkk/5dAAMS6pxzxUROqzZOlUjjAUrk7JycksWLCAtLS0z9QpVwP3Ni36pmgfKPUiNgCJww8E6dPr9QwePJjhw4cTFRWFpxJFQf3szYwrLf8NR/rEf5cyazhw4ACbNm1i+/btlJWVyYbiUnahZWVlERgYSGtrq5yVDxo0iHnz5jFo0CBMKgdtdUdyF7T798uButEK8hMeHk5KSgqDBw8mMzOTMWPGMGbMGIYNGyb9gGJjYzEYDLS2ttLZ2Ul8fDy33Xab3FE8atQoeX9mZiajRo1ixIgRZGRkMGrUKKZOncqYMWOIjo6WHVKU6WrKoVOsqSdOnGDlypW88cYb5ObmotPpJPkUOzt7enqoq6ujt7eXxsZGWlpaZLkvtSvxWqFTlIzD4aC8vJzc3FzWrl1LTk4OBoNBHsk2YcIEgoKCpBLFbelBpPVZ0F0B6UOlnFGlLb7rFR/UkpIS9u3bR3Z2Nm1tbcTFxTF16lSysrKIjIzslz93qNMS/3XKBMNisVBVVcWRI0fYsGEDW7dupaWlhSFDhrBo0SIWL17MpEmTiIqKwmq1Ul9fT1tbGyNGjGDRokXccsstxMfHU11dTU9PD5mZmURHR8sQEur3DhTc68LlZjUQ9YaK8LpUvj9CHj4+PnR3d0u/V6PRKJfSvby8ZBqfBv0lSJ9J8ekTpMdsNstQKzNmzGDmzJl4enpSVlZGUFAQaWlphChnpbrrFHX7U7eLz2t/AwEht5aWFsrLy2lpaaFTOZLQ09OTyMhITCYT+fn5bNmyhZycHNrb2/s9LyZMou+JPhUeHs7cuXPJysrCV9nApYYo3+WWU/c5Pn1CppfqY52dnVgsFmllRdW+RL5tShxPQUS2b9/O2bNnsdlseHl5MXPmzH6B3gcSJpMJf39//P39aW1t5dy5c5Qowe69vLzw9/fHrIqrp24/l7qmhkvxcxe7y6urq2U4IHUorWuFun+KtuyeJ6MSkcBisXDhwgVycnI4cOAAtbW1clXIz89PnosdHh4uN6GId1yqjDcTrrT8NxTpQzVQOxwOzp8/z/r169m6dWs/wuft7U1WVhZLlixh9uzZ6PV68vLypM9ZbGysdE5XL1VwiY7k/v1KoB6Q9Ho9ZrOZoKAg4uPjSUlJISUlRcY0EksOaWlpREZG4nA4aGpqoqamhuTkZO699165bCKWuQYPHkxaWhqpqanyempqKklJSYSFhfWzTrgrjCuBTpmlHThwgG3btnHkyBHKysrw8PDAWwmsGaQcdWUwGCgvL6ehoQGbzUZnZyehoaHExcUREREhnZUHCk6nE6vVSmVlpVxyy8nJwW63M2XKFBYvXixPEtApA4RoQ+I7lykX3RWSPiF39z+9ajfkpk2b2LFjB/X19QwfPpxly5Yxf/58uTz2efkTaQrl6HK56Orq4sKFC2zZsoVNmzZx6tQprFYrc+fO5e6772bChAnEx8dLX8Lc3Fzy8vJISEjgrrvuYuHChQwbNgwfHx+5DDNkyBASEhKk07t490BCreBFu3UqVllRToFLfRdp6PV62tra+llQIyMjSU5OJjg4+HPzLYiMO+kT77RYLNL3Mi4uTk7Szp49S3FxMYMGDWKEcsqDmmio/8Q1Ndy/Xw/oFOtoX18fFy9epKamhoaGBlCOvrMofm35+fny+EdRL+JPr7iVCIuRyWTCx8eH+Ph4xo0bR0JCAmazWZILdb1eSRl1l+nTJ9IX9Sbaf01NDUbFH06n9A3hbywmBYcPH2b79u1s3bqV/Px8enp6cLlcBAYGsmTJEiZPniwttgMFIUe1Zc9kMtHY2Mj58+dpaGjA29tbboYQclb3O3U7UqfrVKzljY2NHDhwgF27dqHX67n99tvJysrqF63gWqFTYlV2d3fT3d0NigVT9FmbcgpNR0cH5eXl8ojB8+fPSyufyWRiyJAhLF269N9Wpq6mzdyIuNLy3zCkz13x9PT0cObMGXngtNiObzabGTduHHfeeSe33347I0aMoL6+npycHLq6utDr9SQkJDB58mTi4uIkCVGnL75zFQJ3h/vz6hleQECA9I/zVQJsiuUHk8lEeXk5jY2NjBs3jvnz5zNo0CD5rK+vL4Gq80T9FAd4MUMUisJdQbjn53KgU447KywspLCwkKqqKuxKzK62tja6lUCbwcHBeHh4cObMGcrLy0GZ6fn7+8ula3c/n8uB6zOWMiwWC8XFxezYsYOPPvqIw4cPY7VamThxIkuXLmXq1Kn9nIP5FBlc6po7dFdI+gSE/IUy7O3t5cKFC6xfv57Vq1fT3t7OpEmTuOuuu5g3bx5Dhw6VS07qNNzhfk2nDGxVVVVs27aNNWvWcODAAYKDg1m4cCF33nknWVlZeHt7U11dzbZt2/j4448pKioiOjqa2267jSVLljB06FD0ej01NTUcOXIEh8PB+PHjL5uIXi3UdSzqS4TfMaicxNXvF21D/AmS1dvbS11dndwZHR4ezujRo4mMjOz3zkvBnfQNVx3DBtDa2ipDIwm3hVOnTrFhwwYAFixYIJfHBdzr0l127t+vF4SMW1tb5W7/uro6SZaqlGOzSktLqampwWazSTmLyZJTsVSGh4eTnp5OVlYWt956K7fccgtjxowhIiJCkuRraSu6z/HpQ6WnxSTOYrFQUlLCjh07ZCgaixJbsL6+nurqagoKCmTA7O3bt7Nr1y7y8/MlEfHx8WHUqFEsWrSIYcOG4XmJ3dVXC7U8jErIk/DwcJKTkwkNDaW+vp6ioiJaWlqw2+39NvyY3GJFutzO8XUqm8d6e3s5ffo0GzZsIDc3l5iYGGbOnElycvK/ye9qId5ZX1/Pjh07OHLkCC0tLXLHfXt7O42NjZSWlnLo0CF27NjBhg0bOHnyJDYljixAYGAgs2bNkhNdQRpF+gMh8/92XKkMrhfpG9h1ncuAugE4FMf0vLw88vPzpQXP29ubMWPGcOedd8pGbjQa+zUy0THE90/DQDQ49+d1iq+J+t1qJSAGLrPZ/P+1d97xURXd///c3c1usum9kQqBECD00HuRJvAIiGJDfBTsBRvqYxcLggWwIFhAQVFARJAWIAEDCRASIJCE9JCE9LrZvvP74zszv7uXUA0Q4b5fr30le+/de+/MnJk5c+bMGURGRsLT0xPt2rVDXFwcXF1d7X7D7sM+YsTHxOmQvs+V4OzsjJiYGHTv3h2hoaFQqVQoKSnBvn378N133+Hbb7/F3r17UVhYiNraWv58FnstNTUVRUVFvKyuFBudyhOnzWAwoKSkBPHx8di8eTPf5aBr164YO3YsBg0ahIiICDvrojSvWqOcL4WVBs3W6/WoqKjgq8z37dsHg8GAIUOG4J577sGYMWMQGRkJFxeXSyqpDJYfrDOuq6vDiRMnkJCQgOrqarRr1w5Dhw7F9OnT0blzZ9TX1+PIkSPYtGkTH3X7+Phg3LhxfErXwcGBO7UDQEREBEJDQ7kzO+tgWxumsIE6oefm5iIxMRFHjhxBdXU1lwFBpEBDsjOIQDv+yspKFBUV8cUVhK4yvBRS+YCkbjJ5aW5uxsmTJ3kA8B07duDEiRN8dbvY8f9a5NXVQujUX2VlJQoLC1FVVcXlh1ArcWFhIUpLS9Hc3Mx/A9G0tEajQXR0NCZNmoT77rsP9913H2bNmoVp06ahc+fOdpa11qhfLZWJGHZ/Qv2gc3JykJSUhE2bNmH9+vX45ZdfsGnTJmzYsAEbNmzAt99+i6VLl2LNmjVITEzEuXPn+DPUajW6d++OyZMno2PHjlC2EBalNREEAW5ubujYsSPGjBmDu+66C71790ZxcTE2bdqENWvWYN++fSil+9ua6X7OLcFklK2Qzc3Nhbu7O/r163dZg50rgeVXc3MzkpKSsG7dOvz000/46aef8Ouvv+KXX37B2rVr8d1332HlypVYs2YN0tPTYaULDgVBgIeHB3r06IFBgwYhKCiI1+Nr1b7I/DOuu6VPLATMP2D37t04fPgwTHRXDU9PT74CLSAgAKV0O6m9e/ciMzMTFosFTjQY85AhQxASEsJH8G1ByFjjazQaUVRUhMOHDwMABg0ahPbt20NLQ3iI3/V6vbeSOsl7e3tDq9Wirq4OpaWlMJlMqKurQ3FxMV/kkZOTw/1pLBYLmunOFM7OzlyZvRJYGqV/z507h/j4ePzyyy84cuQIDAYDoqKiMGPGDD5yZB1QS7+/UoQrtPSZ6C4w5eXlOHPmDNLS0pCUlMSD25pMJowaNQrTpk1D//797aa/WaMqVoRagqVFQX2U0tLSsG3bNiQlJSEoKAjTp0/H1KlTERoaiuLiYuzfvx9//PEHfvvtN+Tk5KBLly6YOXMmbr/9dh4xX6VSoaGhgYfz6NOnD49Bd61ljxACnU7H48IlJSVBo9GgU6dO8PT0tMsP1jkIdGpPSTd8P3v2LHbv3o0///wTp06dQlBQEEaOHIm4uLjLWoXZkqWP+RkJdBeYwsJCJCYmcivH8ePH4eXlhdGjR2PAgAHw8PC4JvnTGhiNRhQXF+P06dOoqKhAXV0dHxgLolXFzMLFrPUBAQEIDQ1F165dMW7cOEyZMoW7nPj7+8OJ7lvLlJJ/mn7hEj597BoxOp0OZ86cwZEjR3D8+HGUlJTw7chSU1ORlpaGrKwsVFVVobGxETZRHFS1Wo2+ffvijjvuwOjRo9GhQweeB62NuC0SqDGAWf08PDxACEFNTQ3y8/ORl5eHoqIiVFdXw0Rj0qro9DqrDwqFAnq9Hrm5udizZw/27NmDsrIyxMXFYcaMGejUqRNfINEasPs0NDQgNzcXx48f54G6Dx8+zHdFOn78OA+gLk5zu3btMHr0aPznP/+xU/quRV7/27nSMrtWlr7rrvSJTb5WqxVnz57FoUOHuDKnpM7VrnRj6ZqaGiQnJ2PPnj1ITU1FVVUVLBYLVCoVunXrhkGDBiE0NPS8RuRGQUQrMaurq5GSkoLExEQ4Ojqif//+CAkJgZOTE9AKysvVINBVxF5eXnB2doaNxghkjTyLal9WVgYj3eOSKX0ajQYNDQ1wdXVFXFzcVY06WUfCOnqTyYSTJ09i06ZN2LdvH/c7GzduHCZPnoyYmBgeW0usKP+TPBOuQOkzm80oKirijWFSUhL++usv7Ny5E5mZmfDw8MDo0aMxYcIE9OjRA97e3lCIQhLhMhQ+di2rFyaTCQcPHsT27dtRWlqKmJgY3qAyRWjXrl34+++/UVpaiq50T91x48ahc+fOcHZ25vlrMBhgNBoRHh6O2NhYvpK4Jf+61sRId535+++/sW3bNpSWliIsLAydOnWCl5cXbwfYezL5Y/Xn7NmzSE5Oxr59+5CUlASTycQXN8XExNgNAi6EVOnr2rWrndInCAIqKyuRkpKC/fv3o6CggE+hjxkzBhEREXbO6G0JlmcONHivu7s7zGYzampq+FQu63wFQYCvry9iYmLQs2dPDBs2DGPHjsXw4cPRr18/XiZsoCC0YBH9JwiX4dPHnqegU8/MdeLkyZPIz8+H2WyGwWDgkRN0Oh0POwNRu8IiJIwdOxZjx47l+8S2lgLLaClf2DEFDT3F3Hzc3d0hCAIKCgqQnp6O/Px81NXVobGxEU1NTTCZTGhubkZjYyMaGxuRnZ2N3bt3c4WvQ4cOmDBhAgYOHMjv1Vqwsmby0tzcjKKiIh6vtYkGx7fQqBkCnc729vZGdHQ0+vfvj3HjxmHw4MEIpfvbywpfy1xpud0USh8TBlY5bHT1WVZWFvLpRufsfH19PfLy8vhI48yZM6iurobZbIZSqYS7uzu6devG/XsuZ8rnesAqvUD9bQ4fPozTp0/Dx8cHPXv2REBAAF/NxTq4lhqQa4lA/UqY8ufv7w9XV1f4+PjA19cX3t7efGrQTB2m2e8UdEeMoUOHXpXSx+4B2lBXVlZi37592LRpE8rLyxEREYHJkyfzwLLMn0qcT2I5uhqEy1D62DMqKiqwZ88e/Pjjj9i+fTuOHz+O8vJyODg4cBeEiRMnIiYmBm40biArV1zBFAe7RqAhjLKyspCRkYGSkhIe7qS4uBg7d+7En3/+iTNnznC/17vuugvjxo1DZGQk9xcSK5qenp4ICwvj5SruYK8VAlVei4uL+Y41er0eWq2W+6qyIMiEBitvaGjgW7YdPHgQGzduxMGDB1FZWYnQ0FAMHToUcXFxCA8PvyzLjVTpY5Y+QleuOtB9rtle2cHBwRgzZgymTp2K7t27w9nZ+bLK7kahUCjgRndoCAoK4vWY+WKxtKpUKgwePBjTpk3juxYNGDCAr+RmeSnNz9Zql4TL8Olj1zGZZIs40tPTkZeXx88zJY/5x1lokHsACA0NxfDhwzFhwgQMHToUMTExcHFxsVP4WiM9LcEUJ1b3FTQWqo+PDyIiIuwWT1VUVKCwsJDvoJOZmYn8/HxkZmbi0KFDiI+P57NDPXv2xMSJE9G/f3+7Fbvsea2Fo6Mjt05qNBr4+voiIiICfn5+sNlscHJy4jH4unXrhn79+mHUqFEYOnQo+vTpg6CgIDiIdruROZ8rzZebQuljiCuHgfpzFRYW8lGPyWRCU1MTKunWQLW1tTAYDHy04e3tjR49emDw4MGIiYnh00VXmqnXCtZ4GgwG6HQ6aLVaHt7Fz8/PTkFlDcSNeHeNRgNPT09ERESge/fuiImJQVRUFKKiouDn5wcLjYuloivntDQ4ateuXXkAziuFpdNGV6fV1NTg6NGjOHjwIDQaDUaPHo2pU6eiW7ducHd3550Aa7DFndPV5Bn7TVlZ2UWVPtByrKysxOHDh7F7925kZmbCYDCgW7dumDBhAp/SYKscWTmyz+WWK7uG/TXRDcwtFguKioqQnZ2NrKwsnD59Gnl5ebBarYiOjsbo0aP59moRERFwohu3E5FCp6Rb8ElDnLBrWrvzYDCLvZIGh2YrcJmCxawcbOVpQUEBH+AlJydj//792L9/P8rLyxEYGIgJEybgtttuQ+fOnS9r5S4uofQR6k/LBj/hdOeQoUOHolu3bnB1df1HcnY9sNCg9Rq6P3RoaCgiIiJgs9mg0+m41a99+/aYOHEiJk+ejG7dusHPz49bg23Up1Ip8nmzicIhtUbaBVGcvgtZ+tizQdsGo9GIqqoqFBYWoqCgAAaDgZ9jfYejoyNfMNexY0eMGzcOEydOxNChQ9Ge7i3e0v2vNSzvFDTSg6urK3x9fREcHMxjs/r4+HALpNVqhUajgcFgQHl5OdRqNaKjozFmzBgMGzYMsbGxCAgIsLPctmZaWH3QaDRwd3dHcHAwYmJiMGDAAERFRcHFxYUvIhkzZgz69++P3r17o2fPnujYsSMP2n+t2pKbhSvNm2ul9AmEtWxXwZtvvokr0RnFj2IVo76+HocOHUJiYiJOnjyJY8eOobS0VPQre7y9vTFs2DCMHDkSAwcORGRkJG+gBdHqtBsJazRtdLqU+UF4eXlBK9pNgDW61+t9L9SJsXxrbGxEXV0df++srCzU19fDZDLxkbVCoUBISAg6d+6M4OBguzK9FOLns4ampqYGO3fuxLp162CxWHDnnXdiypQpfBpDmjcXSsPlIG4w09LSsG7dOsyaNQvdu3cHRDLJ3s9qtaK6uhqpqak4dOgQcnNz4eLign79+iEuLu48RUspWbF2JVgsFijo1JbFYoHJZMLp06d5uJYTJ07wwU6/fv3QvXt3tG/fHv7+/pcMRcHej8maWNm72ve9FISG1WB76CYnJyMtLY3HEXR3d+dT4Wq1GpWVlcjNzUV1dTUEqsgQQuDi4oIBAwbg7rvvRv/+/bnll1nqWoKllU1j//TTT2hubsZ9993HXRpAy5nJemNjI59BcHZ25nnW2vnSmjC/MKVSCYvFAgcHBzQ3NyMjIwN//fUXNmzYgNraWowdOxYzZ87EgAED+EIyZh1jiOsZq2O4ynoGkWWe1YuKigqsXr0avr6+mDVrFrc4twSTnby8POzZswebN2/GsWPHUF1dza3DWq0WnTp1QnR0NLy8vBAVFYW+ffsiODiYGwHE5SdWxK41LckNy29CCJqbm9HU1ITa2lqUl5ejoaEBbm5usFqtaGpq4lZbT09Pbm1W0AHktUqHWN4tNKIDIQQGujWnzWaDp6cnHB0doVar7YKss9/iMuSFyQQukE+My73fxRDfQ/zcG8WVpuVqjWqX4oYpfey7hcYHKikpQWpqKnbu3ImDBw/CYDDAiW5pplQqYaWBanv16oV7770XgwYNgq+vLxe8tgYTaGlBX47A30iISGkWT+uySnOhdF0NhK5ALCsrw5kzZ6BQKBAaGoqwsDBoNBrecbQGgmg1aGBgIDIyMrBu3Trceeed6NGjBxoaGrivJVMmrHSFGrPYsk5Wq9XC1dWVW8/E+XK15Spt5GyiOGUZGRkoLCzkCyE6duyIoKAgO/8ead2SIr3/9YCVLxtM1NTUoLi4GCdPnkRGRgaKi4u5v1BtbS3q6+ths9m4pSE8PBwRERHo1q0bYmNj4efnx++LS6SFLWrw9PTEr7/+iurqajz00EOw0R0O2FSZVbS7Bi5xz7aGOB/E9baurg5ZWVlIS0tDc3MzOnbsiG7duiEwMPCiylZrIFClgSndFRUVcHNzQ2NjI9asWYOAgADceeedaGxshEKhaHFBDqEDLpPJhPz8fGRkZCA3NxeNjY1c6VOr1QgICOBT2myKm/kksvtI68eNLl9WVla677qZxsBj5wRBgFqt5ulgfZs0LdciHey+7B1xGfnFrrucd2IKq020w4eS7sojlUn2/WqUW9ZvsOeI64d0cCNO87XmSp9x0yp9TBCsNHxLSUkJqqurYaGLNZhJ20w3Bvf29kb79u35fqZCC9agtoa4srZ1rkQwWyM9NmoRFVc+ZrVilfdK3ulCCHS15qFDh+Dq6gqz2YzffvsNs2fPhslkQk1NDQYMGACNRsMVgYs9uzXSfjEI7RgIITAajVzhdHJysrO6Xuv3+CewvLPSnR+YJaOsrAwFBQU4e/Ys6urqYDKZuBw4ODjAz88PwcHB8KP78Xp7e8PNzc3OqnChcgHNu4qKCqSmpqJdu3bIyMiAyWTCyJEjkZmZiZiYGAQGBvLrxeXclvPzcrFR1wkLdc9g09gXswa3JuK6XFBQgJycHISEhGD79u0IDQ1FTEwM0tPTMWTIEAQGBvJ6LobJvoKuZmfTu8yabrVa7SxNarWa+2i29f6A0ZIctzU5FL8fe9+rfTc2GGDlIy6nlpQxSOrm5SLuT5j8WOnMgHiK/Hpzpc+8KZQ+awsrBlkBswISIy4csRCQG7QAQqb1YeUpLevWLld2v9OnT2Pfvn3QarU4efIkoqOjYTKZEBcXh549e0KgI0Qmi2xQwpRASGT2WtPSM5gS1RYt3AxxHrEP66yZUsKsHIT6jbI6rlKpoFaroaZ7vjIZEd/7YnWf0Dhv27ZtQ01NDfR6PWw2GwICAqDT6TBjxgx4eHi0qGzcLEjTJc6/a41ArX1KpRINDQ345ZdfoKC7qzg6OvLB+8yZM7mi1pJCeqlyZojTJq67bQ3xe14qTbhG7eCNhsmFuH2QtmXsHK5C6SMiNxbWdqtUKu4nLV2NL27frzVXkg5cQ6Xv2qdUREuJFjfqF2uYxIUvFgiZfy+szC/0aU3Y/UJDQ2EwGHDkyBE4Ojri0KFDUCgUiI2N5Q0Fs6CJ3+FavpsU6XNtImsooQ1lS51kW0NcP8UNsEajgbOzM9zd3eHp6QlPT0+7/11dXXksMmljzI5dqu6zIORFRUUoLCzE2bNncfDgQYSEhMBdtEXUtS7LG4FYZqSycz1gz7darXB3d0dISAiOHj2KsrIyZGRk4Ny5cxg6dCgU1H/1QrLM+gWWBjYlKk2TOG2Xkou2Antnlo6W+Lek5Upg7QCDWaHFZQ3RIPtq8oANLAXR9LhareZtiliPuNpn/Ju57kpfSxksbtzFDRUkna24MKWdgcy/C3Gjx8qTycfldOpXi1arRZcuXaDX69HQ0ABHR0dERERwXyfW+IinIpisiTuWa/V+EHWaYtmXnr8QFzt3vRA3puIGlnXaLG9JC3427FqxfIjvJz5/KSIiIuDh4QGj0QjQhr9Dhw68jG9WxPnH0nk900sIgYNoL/QOHTrwcCUAEBAQgHbt2vFyZdbelhCXP2sXpHVDLBvXsl7+U8R1Qvqe0vK6GSGiHahYHSwoKEB1dTUIbQuYj5+NKvlXmh8CVfTYvYxGI/Ly8vg2hczqL85/aVnc7LQJzYm00Li3VBAtHZP59yBu1KTl3FKDx2SitWCVPTw8HO3atUNzczNcXFzQqVMnfg1raJQ0hMX1QJp2cb4Ios6OXXeh92rNvPonSNMDUTqkSjS7TnpMOvhjtHRvKSyvNBoN+vbtC0dHR+j1enTv3p378rFnXM79/k20lB52THr8WiGIFDObzQZ/f3/uRuHk5IRu3bpBSfdgFrf7DPF7suPiQQKrm+J0Xc/0XS3SdxXX7QtddzPB0szSZTAYsG3bNqxZswZbtmxBeno66urqAJEFUDogvByYYnngwAEsWLAADz/8MO68807cfffd+Oyzz3DmzBk730/2XuJ2V4z4nPS8+Jj0/IV+c6O58hz9B7DClmaCONMuRlvMQJkro6VKIf2/pfOtia+vLyIjI2G1WtGuXTu+KhSiZ7IRp/T4tUB6X/Ys6XHGhc5d6Pj1pqWOTJwm9pEqgewYUwjYQi7xvdg10vu3hFKpRHh4OFQqFYxGI9/dhSn/l3uffxPi9LC8Zceuh3wIElcdge5JGxYWBr1eDycnJ4SFhfFzNrp4R1rGDFYPpe/Oyq+lc22Vi5UDOyb93Eyw8mJlqtFoUFtbi/T0dPz5559Yv349fv75Z74TiViGhMtwARNTWFiIt99+G5988gkEQcDQoUNRXFyMV199FcuWLeM+fuL2hyHQ3cIsNA6mFEEUOgciY4L4vPR30u83kuu6kENGpq2wf/9+bN++HePHj8fgwYOlp2VuEvR6PVatWgWz2YwHH3ywxRAhMtee0tJSfPPNNwgODsZ9990HDd2bWubWxWazYenSpTh16hTfatNkMkGlUiEgIAABAQHo1KkT2rdvD29vb27BY0pWS4oUU+S2b9+OBx54AB06dMAPP/yADh06ICEhAU8++ST8/f3x/fffIzg4GFarFbm5uUhJSYHBYECnTp3Qs2dPaLVaKBQKGI1GJCcnIzMzE2q1Gt26dUOXLl3g6OiIqqoqZGRk8CDgubm56N69O6KiopCXl4eDBw/CZDIhMjISvXv35jFGL5drtZDjuip9hYWFdsvvVTQWkVRTlpG5FrCRolqtxpkzZ3Dq1Cl06dIFnTp14mFDZDn898OmhR0dHWE2m3Hw4EGoVCr07NmT78ghc+1g+atSqaDT6eDo6AiTyYTdu3fD09MTQ4cOtesD5Hp3a8AsaALdTq+hoQF79+7F2bNn+WDMbDZDr9fzkG0KhQJ+fn7o2LEjOnbsCG9vb7i7u5+3CpfBlL5jx47hvvvuQ1FREV555RWMGDECQUFBsNAg5h4eHtBqtdi0aRM+/PBD1NTUQKD+pXPmzMHTTz8NtVqNzz77DEuXLoVWq4WV7pzy6quv4t5770VGRgaee+45NDQ0oKmpCXq9Hu+//z46dOiAN954gyuEOp0OTz75JJ588knp616Um0Lpe+utt7gDPeh+f6AFLSNzrWGOvRqNBiqVCnq9niuC4pABMv9OWKditVrh6urKp3DYOXZe5trBBvGsg9Tr9VDRfZYNBgMvB6aUCzRouqz03RqYzWbur0fogjkXFxe4ubmhpqYGVqsVTk5OPO5iY2MjDAYDj1MaGRmJuLg4dOnSBV5eXi0O4AS6f/maNWuwcOFCFBYWQq1Wo3v37nyrvn79+iE9PR1z585FU1MT3nvvPTg6OmLp0qXw8/PD+++/j99//x3PPPMMBg8ejBdffBF6vR4LFy5EQ0MDVq1aBV9fX4wdOxbFxcV4/PHHMXz4cHTt2hWvvPIKkpOT8d5776F79+54++23kZubi9TUVOmrXpSbQul7//33YTQaodPpALqaDtR5XjqvLiPT2jALkIrGgbPZbDCZTLDSQK9yx3NzYROtvlaIVgTK5XxtITSYvlKphJIGU1YoFNzyqtPp4OrqCoPBALPZzP03ZW5umMKvojE5TSYT6urqeIgmtiuSi4sLzGYzTCYT9Ho9DAYDlEolXxDUsWNHhIaGtqj0MWsd8xvcv38/tm7diiNHjiAxMRFmsxm33XYbfv75Z+zevRtz5szB9OnT8e233wJ0T3az2YyAgADcc889+O233/DHH3/g9ttvBwB8//33mDNnDl599VU89NBDGD58OJydnXHgwAF4enoiPz8f/fv3h6+vL5577jm4ublh+/btWLNmDY8icLncFEpfUVERn5NHG3NulLk1UNKYYFZJLCfbP1gpLNzkIUD+TQh0EQ5TJgRB4FYCuZyuHyyf2WCe0JA9rOPXaDRobm7m5SVz80NEq68JIdDpdNi4cSPq6urg4uICk8kEpVKJpqYmNDQ0QKvVwtvbG+Hh4ejYsSPfV5n5xl2oLguCgOzsbBw7dgx9+vRB+/btUVtbi+PHj+PNN9/EqVOnsGrVKhQVFWH+/PmYN28ePvnkE/57ZpS67777sGnTJuzZswcjRowAAPz+++/4z3/+g4cffhgvv/wyRo0aheDgYOzcuRNarRb5+fno3bs31Go1br/9dthsNh6SaM2aNfwZl8NNofTJyMjIyMjIyADAkiVLkJ2dDY1GA4PBAEdHR3h4eCAoKAidO3dGp06d4OnpaefDxwboLc0OsuO//PILHnzwQUyaNAmfffYZAgMD0djYiFmzZiE+Ph6rV6+Gn58f7r//fkRGRmLDhg1Qq9X49NNPUVBQgMceewzfffcdvvzyS7zzzjt49tlnQQjBggULsHLlSnz88ccYP348Ro4ciYCAAMTHx0Oj0aC0tBSDBw+Gv78/Vq9ejbCwMKSkpKChoQETJkyQvu5FuVZKn/LNf6C1XelL/QP9UkbmmsGsz+QKYuBZaTw/gfqJsf+l18ncGMSWW3lK9+oh1OeVzc4w2FS5+DskFj1xnbpU/ZK5NWAuNSq6NVpycjKqqqrg6uqK9u3bo3///hg1ahQGDhyIqKgouLi4cEswq8/MWtwS7JxKpcLx48exdetW5OTkIC0tDV9//TV27tyJAQMG4NFHH0VkZCTOnj2L+Ph4ZGRkYM+ePVi2bBm8vLxw9913IywsDEePHsWOHTuQl5eHTZs24ZdffsGUKVPw7LPPQqfTYcWKFXBwcMDcuXOhVCqhVqtRUVGBAwcOoLm5GXl5eViyZAkyMjIwa9Ys6etelIKCAoSHh0sP/2Ouq9InI9MWYQ0FU9rEDQqbAoZEeWDHBRovzEb9Bdk0osyNg5Wh+APZneSqEeehuD6wfBbLPKsDUth1MrcuTF6Yn6fBYEBlZSVCQ0MxcuRIjBo1CjExMfD09AT+QX0VBAEeHh6IjIyE0WhEfHw8du/ejfz8fEyfPh2vvvoqoqOjodVqER0dDUII9u3bh7y8PIwePRovvvgiunTpgnbt2qFDhw5obGxEUlISzp07h0mTJuG5555DeHg46urqUFhYiC5dumD06NFQ0viiUVFRqKiowNdff42tW7fCx8cHTz/9NGJiYqSvelGuldJ3Xad3/8GjZGSuOeKRJONyGh6xNURW+m48rBzEViabZFN3mStDbMET56N0kCSG1SdmEZR9925txP0/q5N1dXXQarVwcXHhbSgbZJjNZu4DeCWI631VVRUKCgpQV1cHR0dHdOzYEYGBgXZy29DQgMLCQlitVgQFBcHHx4e/AwDU1tYiPz8fSqUSoaGhfO9us9mMyspKKBQKeHt7AwA0Gg1/7pkzZ6DX69GuXTuEh4fzaCWXy7UyqslKn4xMC9joNERTUxNMJhO3VNioY67BYAAhBM7OznB3d4ezszNvaGTF78YitkYpFAo0NzdDr9fDy8tLeqnMJRB3oKwj1Ov1MJlM3M9KoVDASncwIHRVpkajgVarhUCt58rruK2hTNuGDZIhGlSLLcTiwQJbhXslMIVSkAxQGEyZZO/h4OBgd1684Mhms52ndNqoK4P4vcxmM+8jmLz/U25ZpY81Gm2JtvhOMq2LxWKBwWDA0aNHcezYMTQ2NkKj0cBoNMJkMsHBwQFBQUHo3r07wsLC4OnpCQcHB7vOUebGwRpmpVKJ4uJiFBQUoHfv3tBqtXLdvQLE8kxoKJaysjLk5eWhtLQUBoPBbvcCs9kMBwcH9OrVCx07duSKYUsd5ZUgrU+tWYYXa88vdu7fQltJg1TRgyiKAlPubNRXlMnJlQ6gCbVIK2iYJiLy12ZKJFPIzDQ+sJKuJiaS1cXsI76GKYRE5EJCqJInvp4hiAZMV6oIXiul7+pq4HVGnLnSY1ciEFeL+PnseZf7XKkQtCQYjAsdbw3EedVSfl7ovdhvxEJ+MS50H4b099J3aOn/6wEbHbLnKpVKaLVaxMTEwM/PDxUVFcjNzUVZWRk0Gg0iIiLQu3dv9O7dG/7+/iCX2B5I5voilp/CwkIcOHAANTU1dtfIXBqmrDH5VqlU8PHxQXBwMBQKBTIzM5GQkICkpCT8/fffOHHiBFxdXeHv7w+1Ws079X9CS/WJ1bOWzl0N4vZQek9pW4kW2ifpd8al2sMr4XLuQy7iz3qp9pVdL73HhWjpftK/LcFkiSlnYiuwWFm72D0uhNDCvukMBwcHHsaJKZps32eFQgEV3esbLby/WDll17D8EfcbLB3iYyw/2wptZiFHSxljMpmQkZEBnU4HLy8vHu9GqVSioKAApaWlcHd35yMElsmskWEdOYM9g/2VFgwhhAeCZL9lIw2LxQKFQoHy8nJkZGTwrWDEwmulqzjFz7RYLPx/0NEFEx6GjTq1sndhwSXZu7J7EjoSYT4y7Dx7BjvP3p0JpclkQn5+PpqamuDi4oKcnByUl5fDx8eHp9FgMPAtapj5G7QMWEU8ceIEampq4O7uDiWNhcaewd7JYrHAarXCwcGBvxerxOw6lk+CIECn0/G0snuw+7DfXi9YWtj/CoUCzs7OCAsLg7e3N8rLy6HX69G5c2doNBo0NTUhPDycp5XJChGNNkEbDDbKlJYNJNORMq2DQBv/hoYG7Nu3DxkZGQgKCkJoaCgvH3Hdv9XyX9pOSb8LovaR1XV2HSEEgYGB6Ny5Mzp27Ijy8nLU1NTAxcUFY8eOxejRo+32OGZ16WryWBAEnDt3DikpKTh9+jTy8vKQl5fHY756enrycrRYLPxZbKpZfJylT9yuW61WZGRkoLKyEr6+vvw8k5Fjx46hvr4e7u7uILStFecNa9tYHRbnI7uHjbb37Dc26iLCnsXSya5lx9n17Dfia8XHpfmq0+mQk5MDtVoNQRCQmpoKlUoFV1dX/htpnrC2ymKx8P41MzMTbm5u0Gg0vG1mz4Wob7BarTzwMLsf60PYe7P3ZMdY+8ieK06DQNtG1i9cKdJ7sedIj0NiSZSeb+m37H3ZsZY+LZ2X3v9yuFYLOa5fj3oRmBCJEahgv/7667jvvvv4Virs2p07d+K9995DRUUFr3gMJlSggskqj8Vi4UoXO2+1WnkFtNKdGdg9mEAnJydj//79EAQBmZmZWLp0KcrKyuwqjFiICSFobm626+jNNMK4ki7rtlJFABIzM/OVsVgsvCIJtHKKr2eViNAtxNh7s3MsrYQ22t9//z2WLVsGANizZw9Wr14NBW0czTSQLavcSjpSqqmpwebNm1FXVwcAWLt2LX7++ecWKypEPhgC9fthld1Et8OyWq18z00Frfxs43WWf6wc2TtcL8SVk8EaOHd3dwwbNgwzZ85EcHAwCgsL4ezsjG7duvF31mg0qK+vx6FDh1BSUsLTxxpZcZ6xBoHlj/S5Mv8cK1XqKisruYKQnp7OZZmVAyvjWwWWXpZ+cR1j/7O2SaAdr4Xug6pUKpGUlIQ9e/agubkZKpUKXl5eaGpqgkqlwp133onRo0dDq9Xa3fNq5Zv9LikpCQ899BBeeeUVfPTRR3j33Xfx/PPP4/7778fWrVt5ObL6Jm6H2Ielh33YoNJqteKPP/7AmjVroFAocPLkSfz000+8rV26dCl/BssPm2g60kC3CGPPYOfMZjMsdBAN0SDWaDTy92XtvritZu0lu4a9BysviMqQnWPPZe1MQUEBPvjgA2RlZaG+vh4ff/wxTp8+DdD+EHQ3LNanWKnStmXLFhw9ehQAkJCQgEWLFqG+vp6njdUpQvtS5s6ikEyJ2mw2qNXq89LE/he3eaxcxIjTeq2QPkP6DlKk14uRyrhYzsTf2wotp6KNYLPZUFlZicOHD2PVqlUghNgFaayrq+NCrKTLpZlSpVaroRJtt6VUKqHRaLiJl1UqIprHZ79j35XUUrBixQrk5uYCAKKionDPPffA39+fX8NQ0EZURfeadHBw4Nc4OjryvSaZYgZagTUaDZycnCAIAvePISJrJLs3q0gsHSzdTBhZgwRaqVk6VSoVjEYjysvLYTQaMWzYMEydOpXfl6WdvSu7X1ZWFn788Uc0NjYCAKZPn45JkyYBVJBZ+kA3TmdKuZUqcEyhY88ghPAVTOy9lLTcmJmdpbutVBLWUFmtVsTExOC+++5Du3bt0NTUhJCQEDg6OvJrqqurkZaWhtzcXN64iREfY/+TW0jhuF4w2SGEoKCgANXV1XBycuIWbkFkRRB3QLcSrJ6x+srqHJNJVl+lcmw0GtHU1MQVlr1790IQBMyYMQN9+/aFA91TtTXlms1CPP744/j888+xdOlSLFu2DMHBwfjoo49w7tw53o6I2zCWPnGaWFvE2k+FQoExY8bgrrvuAgCsWLECP//8M2+T/vvf/2LMmDFQ0YG1iu7ew9o4lUoFC7XyOzg48HZMo9FAIZr9EV8jCAKcnJzg5OQEJe2XlHTAr1AoeF/g4OBg1y6zNLI0iJUw9m4QGTIsFgvc3Nzw0EMPISwsjCvvKjrFydpotVqN4uJifPzxx8jKygIADB48GI8//jjc3d15n8P6FJbPFjqro6Bb7In7Ayud7WHpc3Z25mlg/SyTLWk7KXNtaTPTu5Bo2wLdiPuvv/6Ch4cHjh49ioCAAHTp0gWgo7+cnBxMmTKFC+bOnTuxYsUK7NmzByqVCu3atYOVWu/y8/Px1VdfYe3atWhsbERjYyPq6+vh7+8PnU6HzZs34/vvv8dvv/2Gc+fOISwsDM7OzoiPj8fXX38NpVKJ9u3bw2azQa/Xw9XVFSkpKWhsbERgYCAUdNS3Y8cOlJeXIzIyEnl5efjxxx/x7bffIi8vD2FhYXBzc4NVNBIlhECv1+P48eOoqKjAH3/8gT/++ANKpRKRkZFQqVQoLCzE0aNHkZeXh/Xr10NJl46npaVhxYoV/J3btWvHp17PnTuHtWvXYu3ataiqqkJGRga8vb0xatQo1NXVobGxESEhIVCpVMjIyMCqVauwcuVKlJaWIjY2FnV1ddiwYQMOHjwIPz8/+Pv7o76+HkajEcHBwVAqlTh69ChWrlyJjRs3ory8HC4uLvD19UVDQwNSUlJQUlLCo58bDAZERkbCwcEBCoUCaWlp+Pjjj7Fp0yY0NTXBy8vrvECc5AJR1683rGNUKpXw9PREQEAAjEYjLBYLPD09oRQptH5+fggODoabmxsU1Np3+vRppKWlwd3dHW5ubrzDZQ04+y7TeiiVSuj1esTHx6OyspJ/9/LyQnR0NFdwFBJr180MU3BZ2tl3BVWQmIyrVCpUV1cjISEBCoWC+6wSQhAUFISIiAi4uLigsLAQZ86cwbBhw9C9e3eo6J6qrSnLgiDg9OnTOHjwIJ588knExsYiICAAISEh8PX1xaZNm9C/f3+EhYWhoaEBa9aswfLly3HkyBGEhITA3d0dGo0Ger0ef/zxB5YvX469e/fC2dmZD9yLi4sh0NmJr776ChUVFYiOjoaPjw8KCwvh4uKCuro6pKSkwM3NDa6urgAdFO/atQsdO3aEo6Mj9u3bh6VLl+Kvv/6CxWJBeHg4b+fVdHYHAE6fPo3U1FQUFBTgp59+Qm5uLsLCwuDh4QGdToeMjAxkZmbijz/+wPHjxxEbG4va2lqsW7cOq1atwsmTJxEcHAxfX18o6UA/Pj4en3/+OQ4dOgRBEJCTk4Po6Gi0b98excXF8PT0hLe3NwwGAzZt2oSlS5ciNTUVGo0GAQEB2L17N3bs2AFCCMLCwuDg4IDS0lL4+fnB3d0dRUVF+PHHH/H999/j8OHD8Pf3R2BgIMxmM1JTU5GdnY0TJ05g7dq1OHToEAICAnjok8LCQqxduxY//PADMjIy4OHhAT8/PxCRH1xrykxb5UrTeFNP714MrVaLO+64A1OnTsXXX3+N/Px8gJrE2cjIaDTiiy++wNKlS3kIjUWLFuHLL7+ERqNBamoq5s6di8zMTISEhGDv3r2YMWMG9u3bB4vFgqVLl+K3336Dv78/IiIisGPHDnz00UewWCyor6+Hq6sr9+c6ceIEPvvsMzQ2NmLv3r344Ycf+JRlYWEhli9fjpqaGmRnZ+O9997DqVOnEB0djRMnTuDdd99FZmYm1NTXAtRCZjAY8P3332P+/PlobGyEj48PPvvsMyxduhQAkJ+fjwULFuDbb79FaWkpBEHAli1b8MEHH6CxsRG+vr7Yv38/Fi5ciMrKSlRWVuKtt97Ctm3b4OTkhNOnT+PQoUN8pLl9+3asX78eCoUCycnJePXVV1FRUYG4uDgcPnwYL7/8Murr61FTU4O6ujpUVVVBrVbjp59+wl9//QWlUoktW7bgrbfe4spjamoqPvzwQ5w9exbNzc345ptvsHjxYtTV1UGlUuHzzz/H5s2boaCbYH/44Yfw8vJCx44dsXbtWnz66adoaGiAINorlSmANxLWILFRtNFoREBAAOLi4riyzzo6V1dXuzhQVjol0tzcjKqqKt7os980NjaiqamJN34yrQPLz/z8fDQ3N6NTp04IDAxE+/btkZeXh7q6Ogh0wMUU81sBprhBNF3FFD72nVnFmpubUVpainPnzsFMfXcBwMXFBZ6enrxODBo0CLGxsXCgvsBX2rFdDHYvBXVDaW5utjt/8uRJWK1WuLu7o7KyEu+88w6SkpLQsWNHVFVV4YUXXsDBgwdhsViwZMkSfP/99+jUqRMsFgvefPNNJCQkgBCCdevWYe3atXwlssFgQG1tLaxWKz7//HP89ttvaGhowMqVK5GZmQkFtfBt2LABBw4cgM1mw7fffovPP/8crq6u8PLywqpVq/D111/DaDRCSa1iLL8PHDiAhQsXIikpCR4eHti3bx/ee+89VFdXQ6FQYPXq1Xj11VeRkJCAxsZGZGVlYeHChdi3bx/CwsJQVFSE1157DWfOnAEAbN26Fd988w0UdDZo1apVyM3NRbt27VBRUYH33nsP6enpAIDPP/8c33//PTp06ACz2Yw333wT27ZtA6HTzY2NjXB0dMTu3bvx6aefwmQyITs7G2+88QaOHTvG3VsWLlyItLQ0ODk54fDhw3juuedw5MgRODs7Izk5Ge+88w6KiopQVlaG//3vfzh69Ci6deuG0tJSfPbZZzh16hQf9MpcX9p0jhO6KbOPjw+effZZqFQqLF++HKAjeScnJ7i5uSElJQUrVqzAzJkz8eabb2LBggW4++67kZCQgFOnTuGXX36Bi4sLli1bhldffRUPP/wwHB0d0dzcDIPBAKvVivnz5+OFF17AggULcPvttyMrKws6nQ6jRo3CgAEDMGnSJPTt2xcAuBLUp08fnDp1ik/9JiQkQKPRYMCAAfj5559hsVjw8ccf47nnnsMnn3yC0tJSbN26laePNWqEENTU1CAiIgIPPfQQnnrqKUyePBlbt27lo9CGhgbMmDEDixcvRocOHfDDDz8gMDAQCxcuxOuvv47HH38cycnJSEhIQFpaGvLz8/HSSy/hvffew4MPPoioqCiudDC/EgDYuHEj3Nzc8O677+Lxxx/HAw88AIFOLUycOBGxsbGYOHEifHx8+LROSUkJVq5cCR8fH/zvf//DCy+8gLlz5yI3Nxe7d++GzWZDQ0MDBg4ciKeffhqLFi1Cjx49cOjQIQDA4cOHkZ2djRkzZmD+/Pl477330LlzZxCRY7K1FVb9tQZW0RJ9Nj1C6JS8h4cHrKJFKcx6wt6d5WO3bt0wZcoU+Pv782kxnU6H+Ph4JCQk8GegBX8QmStHoLMEBQUF6Nq1KyZMmIDx48dj8uTJCA4ORlFRES+fWwmxXAl0ittsNiM3NxfHjx/nCorNZkNAQADuuOMOPrPCFGQ2kBEEAb6+vggLC+N1ALS+tDaCIKCurg5r167lPn0vv/wyvvrqK4wfPx6xsbFYt24d9u7di6effhovv/wyFi9eDC8vL6xZswanT5/Gvn370L59ezz99NN45513MG/ePHh4eKC5uRkKGsuxS5cuiI6ORvfu3XHbbbdBQ32cbTYbYmNjodVqcfDgQVitVtTX1yM1NRVDhgxBfX09Nm3ahBEjRuDNN9/E66+/jqlTp2LPnj0oKyvjijX7W1paCkII7rjjDjz99NN4+umnceTIEezcuRMODg4oLCxEeHg4vv76azz99NPYu3cv8vLy8Nlnn+HFF1/EBx98AJvNhu+++w61tbXYvn07QkJC8PHHH+PVV1/FrFmzYDAYYDQaYbVa0djYCJvNhpMnT+Lvv//G3Llz8fzzz+OVV17ByJEjUVZWxi2oY8eORceOHWEymbhfNuuHXnnlFbz66qtYuHAhzGYzvvrqK95/KpVKTJkyBQsWLMDzzz+PyspKVFRUoKqqCocOHcKwYcMwb948PPPMMxg0aBAcHBy4rDDZkbk+tAmlT9z4igXAIlp44e/vj8ceeww7duxAUlIS970DgPLycpw7dw7Jycn48MMPsWjRIiQlJSE3NxeFhYWoqalB37594ebmBgDo0qULRo0ahcbGRnh4eOCpp56CyWTCjz/+iG+++Qb79u2DXq+HXq+Hm5sb9Hq9nX+ai4sLzGYzBg4cCFAlhhCCAwcOYNSoUfDw8EB+fj7Ky8uxcuVKvPvuu/jmm2/4qjM26hM3lIQQDBo0iG9BM3jwYJjNZpw9exaETqv07t0bTk5OqKurQ2VlJUpKSrB48WJ8+OGH2LJlC6qrq5GVlYUTJ07Aw8MDvXv3BgC0b98enTt3hl6vB6h10UIdi/Pz8zFixAgeEX3EiBFYvHgxQkNDYbFY0NTUxCOQs6kKnU4HvV6P0aNH8+nqLl26ICQkBMeOHYNer4efnx9iYmK4X0dISAjq6uqg1+sxYsQIeHl54emnn8aHH36I6upq3HHHHfDy8uINCFOgbjRMNhXUSsoabkHkHM7kkEgsKQyNRsOne8XnVdQ3RnxMr9ejrKwMOp3O7h6CaMWeFHbMJvERvND1NzuCIKC5uZnXgQ4dOqB79+7o0qULBgwYAI1Gw+vgzYC03MUQkcO/UuLrxs4XFRUhMTERVVVV/JxKpYK3tze8vb25jBLJAhD2XPY/qx+tDasX2dnZSEtLw8mTJ1FWVsYHjQBw6tQpDBgwAN26dQMhBFqtFhMnTuQLE2bOnIkjR47gsccew+bNm9GhQwf06tULWq0WVhpYmj3HZDLBxcXFrp10dHTE0KFDceDAARgMBiQnJ6OpqQkDBw5EWVkZTp8+jTNnzuCDDz7AokWL8PfffyM9PR3FxcU8HSz/PD090adPH3Tr1g0A0K1bN0RFRSE7OxtmsxkBAQHo06cPnwI9c+YMSkpKsHr1arz77rv44osvkJeXh7S0NGRnZyM/Px+9evXiMyNxcXEIDw+HmcZMdHFxgbu7O3JycuDg4ID+/fsDAJydnfH8889j3rx50Gq10Gq1UIt85i0WCxoaGnDmzBkMGDAAkZGRAIDAwED06NEDR48eRVlZGUB93du3bw8ACAoKgpubGxoaGhAcHIw+ffrwmayUlBSMGDECUVFRgKjvk7l+tH4NvUpYBycWANaosFWsEydOxKRJk/D1118jJycHhE5DNDY2wsnJCZ6ennwxQYcOHfD8888jNDQUTU1NcHZ2Pu9ZCoUCdXV1+PTTT/Hpp58iMzMTSqUSvr6+MBqN0Ov1vHNgU7gq6shLCIG/vz/69OmD+Ph47N27FxaLBQMHDrRTAhQ07IfNZsPs2bNxxx13cIWBpdVms8HJyQkuLi4AXYHLHIHZVKebmxuvkCaTCUajEY6OjvDw8IBCoYCHhwdeffVVTJgwATqdDq6urnB0dOQLXdzd3WGjq3qZxcpms/GVd6CVnL2TiQYg9vLy4vnPGn6TyQStVst3OGDKGWv8W+pQbdQCZjQa0bt3b3zyySeYNGkSUlNT8dhjj2H+/Pm8UWIKn1jpuVFI0wEqP6zjE09Bs2Pij0AtKizvmKXQ2dkZo0ePxvDhw6FSqbg8FBYWIj4+nrsxsN+x+wt05SHrjJgsEjo9w74zCwXrxFkZ3QrYbDY4OjqiZ8+e8PHx4XlLCEG7du0QEREBgSru4vz9tyEuV+n/er0eRqPRLm0Wi4UHVVZQq52DgwNiYmL4QIzQNou1UUyGxAoje4ZWq+VTdA50YcS1qLNmsxn+/v54/vnn8cMPP2DVqlX45ptv8MYbbyAwMBCNjY1Q0cUJRqORt1dOTk5QKBTQarWYO3cuFixYALVajdWrV+ORRx7BTz/9xH1zWT3W0i3BWB1nbSQAjBgxAmazGTt27MDff/+NTp06ISIighsFnJ2duaLYuXNnvP7661y5gaguM2UMdNZFRcOpNDQ0wEBDd7F2x0rdQzw8PEDoYhSlUol7770XL730ErRaLXQ6HTdKgA4y/f39eRkz+bfZbNDpdHwQwO5no/EXbTYbn0IndKcKNuXLQvCwdocZUFh7Y5ZM7Tc3N6O8vBze3t745JNP8Mgjj6Curg5LlizBvHnzsGPHDt4uittQmWvP+T3aDUYsOEq68kf8/ZFHHkFVVRU2bNjAG5vg4GAQQjBs2DA8++yzePHFFzFlyhR4eHjAyckJkZGROH78OA/XkJeXhwMHDsDNzQ2pqalISEjAY489hnfffRdz5sxBhw4doKArm6w0zAhz3rXR8CusUk6aNAklJSVYtGgR3N3d0bFjR1itVri6uiI6OhpPPvkknn32WTz//PN8k2dIplpAlazS0lKABpEsKyuDWq2Gl5cXr1hMOXN3d4efnx+6deuGefPm4YUXXsCjjz4KBVUwu3fvjrKyMuTn50OtVsNoNCIvLw+Ojo5QiFbNKhQKhIWF4cSJE/y52dnZeOedd5Cfnw8bXbTCFGamFHp6esJqteL48eP8dw0NDSgrK0P79u3h5OTElWSGQqGAma4oS0lJQVlZGR5//HH88ssvePnll5GamoqcnBxApCBeiw7kRiIuc4E6dzOLNTvn5uaG0NBQvuADND9ycnJw4sQJ6PV63sGJG3TWMYs7ZamM3SqwDsvJyQnWFsJ3qOmKSHG+/dtgsiGIpg0FkYKvohEEIJpyPXfuHHbt2oWTJ0/yjtpGZ1FiYmLg7OwMC12RCdEg40Kwd7jWWOkKUVdXVzg4OHDlism+s7MzXzhXWlrK052eno6AgADo9Xps2LABsbGx+PTTT/HNN98gLCwMf/zxB2pra+Hi4sLbc5PJBLMohJarqyufBQkLC0OvXr2wdOlSnDhxAgMGDACoFdDX1xfjxo3D888/jwULFmDy5Mlwp7FcGaysLBYLCgsLYaLb1bGwQlFRUdzaxpRCBwcH+Pj4oF27dnjuuef49GlUVBQEQUBgYCBCQkK4fx8AVFdX49y5czAYDBBotAhBEBAcHIyGhgaUl5fz99mwYQO++OILmEwmWK1W/r4CdVfx9PREaGgoUlNTubEAtA8NDg6Gh4cHLBYLbKLYsxqNhs9YFRUVIT4+Hv/5z3+watUqLFmyBMXFxVi3bh1s1DpspqFdZK4PbUbpa6nQFdTXglmrACAiIgKzZs1CQUEBH7X07NkTI0aMwPLly/HXX39h586dmDdvHn799VcEBARg0qRJSEtLw//+9z988803+OCDD1BWVgY3NzdotVo0NDQgLy8PGRkZ+P3337Fu3Tru7+dAQxDs2LEDmZmZaG5uhpFuNwSAm+m3b9+OXr16wcXFBY6OjhgyZAgSExOxePFipKam4uOPP8abb77JwwuIO2SBjlDXr1+PX3/9Fbt378bChQsRGRmJbt268YaIjWCDg4MxZMgQfPPNN1i6dClOnjyJN954Ax9//DF0Oh0GDBgAV1dXvPvuuzh48CBWrlyJLVu2ACKrnE6nAwAMHz4cSUlJWLlyJQ4ePIjFixcjOTkZnp6ecHFxwblz57BhwwY0NzdDrVbDYDDAx8cHI0aMwNq1a/HFF1/g4MGDeOedd2C1WjFixAje0LBnMQRBgKOjI7Kzs/H666/j+++/R3p6OvLy8uDm5oaAgACAljvrsFkHdLPAFFqxUsb+t1qt8PPzw4ABAxASEgIj3fLNbDYjPT0df//9N1/swn7DFBibKBagTbRKk43M2blbBVa3lEolqqqqkJ2dDYPBcJ6C9G9F/O4GGtKEddSgyktZWRnMNNA7AHh4eGDAgAF89bJ48GCjFnoisjK1lfxhlm3WFjAlQ0EtcQqFAkOGDAEAfP311zhw4AC++uor7NixA3feeSf8/Pzw22+/4cknn0RiYiIKCgr4AjTmnsKULK1Wi6ysLCQkJKCurg41NTU8b9VqNSZNmoT9+/ejvr4eEyZMAAB06tQJUVFRWLJkCbZs2YK9e/filVdewaZNm3jem81mWOk0sqOjIxITE/Hpp5/i4MGDWLRoEQwGA8aOHQtQn3G2e4xKpcLEiRORlpaGBQsWIDk5Gd988w3eeecdFBQUwMfHB3fccQf279+PlStX4sCBA/j666+RmpoKQi2LBoMBBoMBsbGx6NWrFz799FPs2bMHv/76KxYvXswVaQDYsWMHzpw5AyudlREEAZMnT0ZWVhZeeeUVHDlyBMuXL0dCQgJmzpwJb29vOyMIqOyxRSmNjY1YsmQJXn/9de7/3tzcjMGDB0NBLZHiwarMtadNhmxhjbLNZkNZWRm6d++ODh068BFfYGAgXFxcEBkZiZ49e3KrV25uLpKSkpCSkgIvLy/cd999iI6ORlhYGGJiYvjS8vbt20Or1cLHxwdTp05FY2MjkpKScObMGZw9exaxsbEICwtDly5dEBQUBKVSid27d8PJyQnt27eHUqnE0KFDodVqoVQq4ezsDA8PD8ycORNeXl4wmUzcn+3AgQM8dtvUqVNxxx13ALQRYI2uTqfD7t274eLigvLycqSkpCA4OBjz5s2Dn58fGhoaAAD9+vWDhsZ0ioqKgkKhQEpKCo4ePYry8nLMmzcPQ4YMgZubG3x9fXH48GGkp6ejqqoK3bt3R2xsLLp27QqdTgcPDw/06tUL7dq1g16vx8GDB3HkyBE0NDTgueeeQ5cuXeDm5obS0lLs378fnTp1gpOTEzw8PNC/f3907doVDQ0NOHHiBI4ePYqSkhI89thjGDp0KEwmE2pqahAdHY2QkBAAQG1tLZycnLiPlU6nw6FDh3D06FEUFRVh5syZGDlypJ0FBpdhbfi3IU4LUzyYksY6WSYXYsVQoVAgNDSUh8tRKpWor69HWloabDYbtwyyTkZFp5ENBgMfnYvz9WaGpdFC3QyOHDmChIQEtGvXjlsgIJpu+zfmiVSOWLkrqMtBamoq/vrrL4SHh8PT05NblQIDA7m7iZKuKmWdLvttS/e/EbDn1tbWAgCGDh0KHx8frgQyeReoxcvPzw/Hjx9HSkoKDh48iEmTJuH++++Hp6cngoODkZWVhSNHjuDQoUPw9/fH3LlzERISgpKSEnh6eqJv375wdnZGVlYWKisr0atXL5jNZkRERKBr165QUH9uo9GI4cOHY9CgQTBR/z+2Mjw5ORnp6elwd3fHPffcg65du8JisfDBmdlsxuHDh1FaWsqjLjQ1NeHxxx9Hr169YDKZ0NTUhM6dOyMqKgoWiwUhISFwdnbGoUOHkJ6ejpycHIwYMQJ33XUXnJycEBISgvLycuzatQu5ublwdnZGXFwchg8fDo1Gg5qaGgwePBjBwcEIDAzE0aNHkZiYiKysLPTv3x+PP/44PD09odfrceTIEQBAdHQ0fH19ERcXh06dOsHV1ZUvjMzMzMS0adNw3333QalUora2Ft7e3ujVqxccHBzQ3NyM+vp69OrVC7169QKo33tKSgrOnj2LYcOG4cEHH4SCWj0daOzCm50rTeO1CtkikH8wnHvzzTdxJTrjhR7Fjos7PoEqfU1NTVzRAR25s47SaDTyyqRUKmE0GnH27FlYaIwkjUaD5uZmbNy4EaDOvA4ODjh8+DBmzZqFJ554Ak8//TTMZjOKi4ths9ng7u4ODw8PPvJ1oAsRKisrIQgCvLy8oNfroaaBKgU6UjabzXCkwZct1LqiVCpRXl6O2tpaaLVahIaGQqBWMNbAKhQKVFZW4vHHH8fQoUNx//33o6qqCkFBQXB0dOQjLr1eb+fXwfKgrKwMFRUVCAoK4lPBTk5OsNlsqKmpQU1NDdzc3PjiE3d3d77iysnJifuglJWVobm5Gd7e3vDx8YGJ+hJaLBaUl5fD3d2dl4+rqytPd2lpKXQ6HQICAuBMA3ACQENDA5ycnLjTvJlO7To6OkKj0XBLhM1m4/6BbNSuFAWZvpmUPrGCwRQ9sbyIlTOTycSn4ZlVWU13arHRiPc5OTnYuHEj+vfvj6FDh4KIthczmUxwpIGjWV0SP+NmhuWDjU5zlpaWora2FuHh4XbT5uK25t8GoQorK2uFyK9OqVQiNzcXOTk56NOnD7y9vbmMMVlgaTZS32CFJHSNoo1YhgldNW+kK4sVIl9aZiVSiHbkqKqqQmVlJTw9PeHv78+ngVUqFV8Ap1Ao0K5dO2g0Gj6DohIFl6+pqUFjYyOCgoKgUqmg1+uh1Wp5vrHnQpSfDg4OaGxsRHV1NYxGI3x9faHRaHgdZHWWEIIPPvgARUVFWLJkCUpLS+Hp6QkPDw9ulWX1nc0yGQwGODk5obq6GrW1tXB2duZhoViZCYKA8vJymEwm+Pr68mcplUoYDAZoNBoQurmB0Wjkob/atWvHrzWbzaipqYGGBmI2Go1wp9ucWuiiDhaPlW1nyHwAbTYbtxYyy7OS7l8uCAJqa2tRXFwMNzc3hIWFQaCzW6wP/TfWwSvlStPY2kY1RptW+oho2os1ShZqamdKC6sohE5LsMZKXDFXrFiBb775Bv/973/h6+uLrVu3oqqqCm+99Ra6d+/OKxURLRpRinbbEHcg7JyJbr1joyNmQeSwD9qYKmnkd1ahWBpYg6Wg/nXnzp3Dc889h+HDh+ORRx4BRI262PzN/hfnAxslMeVQoNHgQS0dKupozX7Ljlnp1CnrKFj+gVZg1kmI7y9Qa4C4PNgxG/X/U4t2NDGLnHtZ2VglPmgMsSJqEe1jyz43AyzfpPnJGm5W5qyMBYm1j50XqNwZjUZUV1dzvyN279OnT6OgoAC9e/dGYGCg3bMgmuIX318M+87e4d+G+L2NRiMc6C48ZmpdbyuI815aDqyMFKKdg1h9IbSzVyqVKCsrw86dO+Hv748RI0ZAJQqQzO5nEW2Zxe7D7sGexc6zZ7Ln4Co6q9bERhd1OTg4cB819v6snpjNZt7uiPPTSqdTCVV22HkbHVya6LaYrD0WaIxQjUYDQVRHxRBRO8naeHYNa+Mg6jMgMlSwvx9++CGKi4t5+DGI4s460F09xP0cazdYu87aUEjaDLF8CHQwyd6PKZKgbTqDtbWszrB7sO8sf1k/wOSRvR/7DUsvS7/4+aB5wGBlw56jvM5bbt4opLJ0Ka6V0tcmpneZ8LT0v7gBU4i2R2NCxCqaOEPFAqtQKNChQwdEREQgj27W3bVrVzz55JOIjo6GVbI7hkISmoO9BxNycQVhCpX4nRlsBMNgz2DXS3/n4OCA6OhoBAcH89hRhC7eEL8H+7A8YJWf3YM9R/wMMeJKKW0kGOxcS79n58X/CzTvWMMqPsfKRwz7jfQ6ccNyoff/NyNON0uf+HhLaZfmg/i4hsYKdKLbNrHyzM3NRUlJCYKCgvh0piAIqK+v5xZEVjdYp8FkgDXorFEWv/O/CXGeCiJLp1QWryfSesbynSk2DIEqHxBFC2CDMNb2gKZFp9OhqKgIrq6uCAsL4+0FawfYddL8EER1U3pe/B7i7zcC8TsoRduQsWMsreJyFUT9hvQ34vaSpV38O1aHxMekHwZ7rjQP2TPEz1XQfoPVu8jISHTo0AE2qogLov4EojJT0IDL0raePUtBDQfi57K/4rQoRDNL7Br2XHYfiBb9sPsJkvwV35ulmx0T5wURGWHEecYQP5fd+2anpXy4GDf19G5rw57DrGtsZGOhU2OschHRPrEtFQgRBeRVtBC5v6XfXA0CjStGaBgEA11AcqH3kpG5EAKdNmFhe5gMmUwmbN26FWq1GmPGjIGabgvFLEPiusmUPuau8G+DpYd1PKDKEkvrjYI9m4g6W9bJQvSOrM1h5xS0w05JScG5c+cwfPhwuLm5cUWQtWVMobhVOtF/G2zAxf5nri9SpfVGIW0HmIzKtA5XmpetZVSTcuMlrZVhipxApz7ZCIk1kMwaZaWm/4spVoJo+y0zXVbOKsKFfnM1EELgRDfgttGpjNZ+hsytAaHhSry8vOAginqvoHuoMr9QJssmkwlFRUUw0nhlTPFgfj7Sgc6/ARtdIQ1ah8+dO4fU1FQ0NTXd0DrF6jRro4hoIQIrN/F1rG1iHXFtbS1KS0vt0gaRki63GW0bB+rqwwZbzP2lLZQZkzE2aPi31n2ZS3PTKX2CyOeDmc4hMpczEztEU67i0bYUcaMLWjnE920NWCdlk/jIteYzZG4tmDyxuqBSqdC/f3/07dsXGhr4W6VS4ezZs9i2bRuys7P5ginSgp/QvwlWh0AVorKyMpw4cYKvAr1RiMuEtT026jNFqN8v+5ucnIycnBzeCVssFgwePBh33303X5QlTqeCWgPlNqPtIu5/WHm1hTJjfZzYraOtWB9lWp+bslTFIyf2P6tY4hEME/aLjbSYIqZqYcus1oI17OJKxpRUGZkrhck1k1WpBU/c+Tg6OiIiIsIu+LiCrig/deoU9Hp9i/WprSOIrGXBwcGIi4vjuwrcKFgdZ7MMoJ2rmkYCYNY6o9GIkydPoqioiCuCVqsVWq3WbvUx66DZjAZTKmXaLkrRhgPM2nujlStW58VuUDI3LzdW2q4RrFFkVgrWwbGOkHUG7PiFOjKBWg1rampQUFCA+vp63nC3duMq7qRZQ8C+y8hcCUS0UhAilwbWmItlLSgoCOPHj0dERAQsdKWjzWZDeno64uPj+RZUECl8F6ovbQlWxwHA398f3bp1g7Oz8w2vU+z5rH4XFRUhJSUF9fX1cHR0hMVigZOTE8aPH4+4uDio6Q4JTKkziFawits31mawYzJtD9bXsDqGFvzorjcCXeCVlZXFp51Z/b7R7yZzbbgplT42yhd/Z4sxxOdYxyAWcra0nfk46fV67Nu3Dzt27IBOp4NArSQXmxK+Gth7iTsrcRraKqwBkxuHtodUfqSy39J5Vk9iY2MxceJEbh0TBAF1dXXYsmULEhIS+DHxvZhFkdCpIrESKVY4Qa0c7Hxry464TjPYAK81YemUtgXi9IjbCjZgVIpWlBYVFeHQoUMoLy+HTWRxCQ4OhpubG1cUBKrosdiaDGmZylyaG9VmicuHWd1ZW38jqaqqwh9//IHk5GRubQZ9X5ZH4nrK8k/m38mNl7hrjLRRFHcI4nOgih6oeVupVEKn0yE+Ph5lZWXw9PTkkewhWQ7fGkjfpa0jbjjF7369G1KZ82FlwZQMdqyl/0HLTHo+ODgYkZGREEThTvR6PfLy8nD27NnzlDdWdxhKGkNNrBCJlSTxu7W2zIg7KKUohJFYWfqnMNlXSoIDs+eK64ZCoYBer8e5c+dQX1/PlWBCCLp27Yrx48fD19eXX8ven1Drnbhs2P3FdU76v8zFEeeXWFauNey54jKSfr+WiNMp/l+pVMLLywsZGRnIzMzkdZfVVeY2wKajr+c7y7Q+bSJOX1vAJtowGgCampqQlJQEZ2dndOvWDTabDV5eXtz3iQn/rQzrkFg+sMZBbhRuHpiyYaU+ZZGRkYiMjISzszMv56ysLCQlJUGr1cLb25srK2wKkilETNETKzBEFHS4tRB3ToQQnD17FqdOnYKbmxscHR2ll/8jpHIuTqONWheVSiWam5vx+++/o6SkBO3bt+c7JDg6OsLb2xtarRagUQIEkfLX0jNk/jksT6+XwtcWaEnpEwQBVVVVUCgU6NatG44fPw69Xg8/Pz+A+otKZZrVb5kr40rr8bWK0yeXnGhUbqWBmq1WK44ePYrm5mZ07doVAQEBaGxshF6vB0S7fbBO71b8sHxjH3FHz87L/Lsh1DLH6gYhBP7+/ggICICNrjoVqGJlMpnsrFtWqxW1tbWora3lssHkwmw282vZpzVhz2JWirKyMqSnp6Ourk566VXD6gGbxoZIQdbpdDh37hxfBCMIAh88durUifvpsXSzfLbRxTbMInkt8uZWRtpeQWJ9uxU+LWGhu4FERUWhR48eOH78OE6fPs0XfbHBPK6zZVLm2iArfaLpLge6f2JCQgKMdFNtZ2dnNDc3w9HRkY/GbSJr1q36EY/+BNoIsHMyNwficmUjfhud5mHnbDYbOnTogAkTJiA8PJwriUajEUlJSdi3b5/dLiDsXswKaG1l31iIlCWB+sFFRUVh3Lhx8Pf3l1561TClQa1W81WPoHmVm5uL9evXo6CgACqVCkajEYQQxMXFITY2FgrJVoZi657UmiJbVFoPlp9sSl7aht3sH3F6WR6AyiwhBI2NjQgODsawYcNw7NgxHDhwgA9E2MCE3Ufm38tNuSPHlSCIzPwGgwFJSUkoLy/HiBEj4OPjA6vVitLSUhw9ehR9+/ZFcHBwmwqqeaNglh+2BRhrOFhnJXPzwPZ0ZhYqZi0R+8ixus2ULYPBgOTkZJjNZgwcOJCvnLVYLKioqICTkxPc3d1hNBrh6OjIp4EvByZrF0L8LjY6rXwt6qvZbLbb75qtfMzPz0dKSgr69++PsLAwvjMQqxcmkwmEblsIUcgMcVuEq5gOkrk4TJlmlivWVl1Knm4WxG0zkzmFQoHs7GwUFRWhf//+fDB2+vRppKeno1evXoiNjYWS7tsu/n1ru2Xc7Fxpfb5W7nO3pNJHRMoJ2w6nvr4e+/fvh1KpRI8ePeDj4wODwQClUona2locOXKEr7jCZXQ8NzuEBvx1cnJC165d4erqajcFIHNzIFWcmIVKDCtvdh37nbh+MHk5deoU4uPjERsbi6FDh9opkOL7sGcJLVgWxO8gVpSkShP7XyEJf9SSfDLZZdeK6ze7Xnqsvr4eCQkJcHBwwJAhQ+Di4sItnTabDWq1GgaDASoa41OcP+L7sU9LaZVpHQSqjBcXF8NKQ98wmW5JHm5WpEofIQQNDQ3w9vZGz549+S4+SqUShYWFOHHiBNq1a4e4uDh+Pfu0pPSJ66GMPVeaL7LS10oQOvXEtsSx2WxoampCQkICBEHAsGHDuOmbNdQmkwm1tbWoq6uD2Wzm4RZYR3GrotfrUVBQgEmTJsHT05OPBK9UuGVuTqRKmCAIKCoqQklJCYKDgxEYGAgFjR9YWFiIhoYGxMTEwNHRkU/7MmWQTS0RaqmRTqmKOxuxIilWWNk58XuJOz9x+yTQVcuCaFqM/YZdZ7FYsH37dqjVaowcORJqtfo8xa4ttHky/1eex44dQ3FxMXx9fUEI4Xvf4hpEY/i3QOj2f97e3nBzc4Nareb+uY6OjigsLMSBAwfQs2dP9O7dG2az2c4qL6570nooY8+V5ous9LUSYqdUhUKB5uZmHDp0CPX19Rg4cCD8/f1RV1cHrVYLq9UKo9EIlUrFfZpYZ3Cro6RBq5OSkjBs2DD4+PjwfLlS4Za5dSCiKSaTycQHVvv27UNhYSGmTJkCX19fGI1GuwUP0raDdTBMuWPnFdRfzmq1Qk33D87Pz0dZWRk6d+4MLy8vu/tJOy+TyQQnJydusWNKZ1paGo4ePYqRI0ciIiKCu3jo9XoolUo+iBQrozJtB0EQkJqaCp1Oh549e3LZYEoM2kj/dL1hcs/+t4m2BXRwcIBKpUJGRgaOHz+O3r17o3PnznzQJW7nWZ2SB/0X5krz5VopfbfUXIJ45K5SqWAwGHDgwAGYzWaMGTMG7u7u0Ov1cHZ2htVqhdls5r46Nhq3yGQywWw23/IfiyiGk7Tyy8hcCDa1ZqGxMAW6sKN79+4YOnQoXFxcAAAajQYNDQ3YtWsX0tPT+ZSuIJqOYx0Wkzm2AlagMfnYuaqqKuTk5PDdRQhVFgXRPt3Msujg4MAVPrH1x2AwoKyszM4yJAgCnJycoNVq+XtZqa+rTNuDlRFT9vR6Pcx0Jbm0fbtVPiaTidcbk8kEk8kEUAs7aJ8ZExODPn36IDk5GWlpaVCr1Vz+IVotL66LMm2XW0rpY9hsNuh0Ohw8eBBKpRK9e/eGRqOxUwrNdBqXdQqsoyAi3xsldXpVilaD3SrfWYep0WigEPmJyJVe5kIwRUocCoIpW35+fujYsSPfigxUQSwqKkJ1dTWUonh/VqsVTU1NsFgs/F4McR01Go2w2WyIjo7G2LFjERAQwOVT/BtxfbZarUhMTER6ejoU1IVDr9ejV69eeO655xAeHm63H7FNtJUjqxcybRMFncY3U0VPJQqPI23fbqXvAnWfYAN45r5ktVqh0+lACEF4eDiGDBmCwsJCHDt2jF/L6iOhVnGF7JPa5rmlSogJucViwZ49e1BXV4fevXvD19cXoI2/wWCA0WiEVqvl17LOSlwZ2OjoVvywSi4+xvJXrvQyF4PJiyAK3WIVraaEaJrJzc0NkyZNQmxsLD+vUqmQl5eHXbt2oaKiAqDKIbsf68Qhssa5urrCx8eHd25MyWOyzDpB0JXKOTk5KCws5HKtpNO3rE1g70xE07msPgi32MKAfxtMxphsmEwmGAwGLl/Stu5W+BA6LcvCD7H6x/LJTGd2QkND0bVrV6SnpyM9PR2gFkF2H5l/B7dUDy0IAmpqanD48GEAQL9+/eDu7o7m5mYu/E5OToBo712myIgVPaYAsgaedSIK0fTTzfxdfJzlK2iHziw3MjJSxDIj7ijEcsOsdOwTEBAALy8vCKKOyGq1oqGhgU9FsY5cp9OhoaGB31MpcjgXyy87VlFRgezsbFRVVQE0BIujoyPGjh2LXr168XdQqVSwUqskUzzZPdmzFC0sBpFpezAZYJYtJZ2tELdlrI27UPt3s32HyCfPRqe6Wd6wqVyVSgWTyYSwsDD06tULOTk5OH78OJ8RU4gWOsm0bW4ZpU+gncbJkydRVlaGuLg4+Pr6QqfTQSFaiSuI/HxYxyRuzFmludVho0PWSUO29MlcBgpqbRd3QMx6Bkn9YnWWnWd+elFRUbjjjjsQHBzM66XVasXOnTuxceNGvkCEySa7RmylEwQBtbW1SE5ORlFREQSq3Dk4OCA8PBwRERF27yPtNJmliKWJ/V6W/7aLtB1nf8WfWxFp2pmsE+rrKogGXIQQREdHw9PTE4cPH0ZVVRWvH+wambbNLdNCMSUuOjoaNpsNx48fB2hHYKWLNqxWK1+oIYh2IYDEUiDzf4iVYoh8o2RkLoW4o5F2Oi1dI8bBwQHu7u7cAgdav728vBAaGsoVPqVSCZ1Oh6SkJGzZsgWnT5+G1WrllrqAgAAMHDgQUVFRdvdn8ixG/K5SWjom0/Zg8sTaKabYiNuwWxWxDEvrHVvwwvatPnfuHJqbm9GpUye+S5VUmZZpu9xSPbRSqYSfnx+GDh2K/Px87N27l/skCDSqvlqt5o0Bc/iVaZmWKvit3njKXHuYxYZZIVR0wdXAgQMxfPhwO2WwsrIS8fHx+PXXX5GQkICamhoI1HLh5eWFqKgoaLXa8+RW+l3m5oLJkMyFEUSGD7Zgqq6uDomJiXB0dESfPn3g5ubGFeeW+gOZtsctpfSB+u20a9cOU6ZMQV1dHVJSUnjHwRQ9UGuCcIvvuiEj05ZhdZMpgKwOQ+Rv5+3tjdGjR+Puu+9G3759uYLHfLpkZGRahtUTpVIJrVaL6upq7NmzB8HBwRg6dCi3/DGFT1b6/h3cUkofG92ZzWYEBgZixIgRKCkpwbFjxyCIfPlYyAjZR0dGpm1CJHH0xP+zDshGVwAPHDgQEyZMQM+ePfkewGLkzkpGpmUIXaGel5eH3bt3w9vbG4MGDYKTk5PdLBirQ9K6JdP2uKU0GiLaDcBqtcLHxweDBg1CVVUVDh8+DL1eD41Gw5U/sQAzHxDm1yf+fit+5JHdrUNbLGdWP5lfKUMh2meXnQdNg3RAx47JFr9bAyYX0nZM2rbdip8L1XGlUonq6mocOXIEXl5efIcINiMmRVb62j633DZsNlF4Baa4nDt3Dtu2bYOPjw/GjBkDhUIBo9HIr1MoFGhqauIdiJLGLrpQRbkVUCgUaGhowIkTJzBy5Eh4enradbAybROm5LDGHnQnC4gUJiWNTwmRrLPFD0q6bRqrFy3B6hYLecT+Z4oWg92HWemIaCsnMUx5Yx00QzytJJU9Fo+P7bBhFW3LJuZissrSIYW9N3sf9mxxuyJFfA2DpZelwSzaEsxGgwez/9nvxP9LuVD+yfxfOZ84cQI1NTUIDw+Ho6Mjl8tbBalssnqjVCr5VK1Go4Fer4eTkxMUCgUqKyuRkJCAoKAgDB061E5e2T3ESJ8h8/+50ny5Vtuw3XJKnxSBjvZLS0uRlJQELy8vDBgwADYaO0wQBDQ0NCAjI8NO6TOZTLxRvhVhHYvVasWQIUPg4uLCR3+3cr60ZZhSYKUxuCCxgDEFTUHjUrIGnIgUM6bYMOWDNWSsbrPrrTTgMjuvoEqmVEFi/4PKEkRTtTZqxRPfQ/o8Mey54vs50D1xWecm/r20EWbKnPg5oPdheSB+LrtW/L/4HuL0sXdj9xUklnKBBgpW0JA27H2JSDFkecHyWKC7mgjUNYW9hzRdMv+Xv6dOnUJ+fr6d7DNakqebDSbD7H+2iLGhoQHBwcHo0KEDHB0doVKpYDQaUVNTg/T0dLi7u6NPnz5wdXWF2WzmdVnmyrjSeikrfdcQ1khXVVVh06ZNCAoKwvDhw3njX1FRgYyMDERERMDLy4tXnFsZ1rkIggA3NzfeAbFOS6btweofk3em6Clp2CKxksQUC2ahYwoZ+z3biQIiRY+dYwqMWq3mOx6wPXXFDR9ThtgzpDClj3UyF1JoCFWMmNM5RAMztvrQSoOrX8xPl6WBPUOcD+JrBKoYs0EOyw92rfh3RGJ5Iy1YWiGyuojfkVkrWZrY78Tv2dIzZM5HEO22ZKUhugS6gI/l/c0Ok0uWXiWNfXn69GmUlZVh5MiRAABHR0eUlZVxC9/AgQOh0WhgNpvtdqCRZe7KaKntuhiy0ncNYaNzpVKJgoICHDhwAKGhoYiLi4ODgwMKCwuRlZWFvn378i3bZP5/eYqnusXWFJm2BSsv1sEp6TRuWVkZXFxc4Obmxo8z9Ho9qqqq4O/vz4/bbDbe+Is7S6aEMOWkuroaKpUKXl5eMJvNIIRArVbz3zE5sYpi5+ECyhd7joIOxKTPZShE1kRCCBoaGuBAt1CTtj/i+7POkMGeKUimlIlIqWNYqVWzqakJzs7OcHR0bPH9pPcUp7OqqgoqlYq7SbDrBKqsNDQ0wNPTEw4ODrzzlb63NN9k7JHzpWXy8vJQUFCAgQMHAgCqqqpw9OhRCIKAESNGwNXVFTqdDiqVCmq1GlbRTIHM5XOl8netlD5ZVaeFwRr2iIgITJw4EWfPnkVSUhIsFgs0Gg0fHUJkgWDWkVvtY6MWCbZVHUS7HVjoXsUybQ+p4iQIApqbm7F27VqsW7eOX2cymbhCsWbNGixbtowfIyKrlvi+4nszC8KPP/6ILVu2wEaVRKbYiRUVgVpbGFLFRaxkKUV75DKs1ALJ6nBeXh727NkDq9UKo9GIb7/9FseOHePvxZQtscIkrv/snEKhsLMKStOooJY4lraioiIsX74c2dnZordrOR0s/9m9DAYD1q1bh7179wL0WcwKq1AocPLkSSxduhTnzp3jx8T3Fr+buFxk7GH5xMqC0AGKhfp83uwfcdrZ/6D+r+xYTU0Nj8M3btw4aLVaWK1WODk5cZ9YsbzJ/PuQlT7aUBIaysViscDd3R1Dhw5FQ0MDDh8+zKeqWCWRNv632od1wCzfWB6yPJE7nraJQKfgmVIDAFqtFnq9Hj/88APy8vL4dJdA3RrWrl0Lo9EIjUYDBwcHO4WPyYFSqeTTqOwYAPTo0QNdunThdYvJjFU0vSmWn5Y+hFrgxRZk9gz2nXVkAJCSkoL169fDwcEBSqUSiYmJqK2tBUOQbJcmiBQ+9j/7zo6x50Ai32q1miusDQ0N+PPPP1FRUcGvY9eyDlaaDvYcm82GtLQ05Obm8uNsoQEAVFdXY/fu3XxfYZbX7MPyW/yeMi0jVlpYPWCDgZv9I80HsdInUN/1xMRE+Pr6YtCgQVCIXCoIIXzgp6DuEjL/TpRvXsn8rIRrZX683gh05RzoIgRCCDw9PeHu7o7jx4/j7NmzMBqNiIyMhIuLC2xyMEqANhzijkxsnZBpewhU4RL/7+DgALVaje+++w5hYWHo3bs3V37i4+Px999/47nnnkNYWBjMZjNSUlKwe/dupKSkwGQywcvLCxqNBjU1NaioqEBRURESExNhtVrh6+sLNzc37geblZWF7du3Izk5GUajEUFBQVAqlWhoaMDp06f5NKzFYkF+fj7Kysrg4eEBlUqF/Px8bNmyBYcPH4bBYEBAQABUKhVXJhUKBWpra7F69WpkZmYiLi4Ozs7O2L59O4KDg1FTU4NDhw6hqakJ7dq1g0KhQE1NDaqqqlBSUoLExESoVCp4e3sjKysLe/bswYEDB6DX69GuXTv+jMbGRmzfvh179+5FeXk53N3d4eLigqqqKmzduhUxMTGorKxESkoKFAoFfHx8eJ0oKirCjh07kJSUhOrqagQHB/P2Jj4+Hj4+PhgwYAAAIDs7Gxs2bEBlZSV0Oh1SUlIwdepU+Pr6Ii8vD3/99ReSkpLQ1NQEf39/Pu0m172Lw2Sb/c+Uv1uhLZemkfVhVVVVyM7ORklJCdzd3TFy5EhoNBpYLBY40NXvrI1n+aeQB/dXzJXmV0FBAcLDw6WH/zGy0kdR0BEfE2pBEODq6gp3d3ecPHkSBoMBnTp1gqurqzyVQmH5xPJBVvjaPkx5YeUmCAK8vLxw5MgR1NbW4rbbbuPWqzVr1sDJyQmzZ8+Gg4MD1q1bh88//xwFBQU4deoU/vzzTwQEBCA6Ohrp6en44IMPsG3bNiQkJMDLywt79uxBUVERBgwYgNOnT+Ptt99GUlISioqKsGfPHhiNRvTo0QMFBQX46KOPEBISgvDwcNhsNixduhQHDx7EbbfdhpycHHz00Uc4cOAAcnJyEB8fD4VCge7du/MBmFKpRHJyMr777jucOHEC/v7+6Nq1K3bv3o1Tp07h+PHjSElJwe+//46oqCiEh4cjOzsbH374IRITE/H777/D29sbTk5OeO+997B//34UFhZi9+7dcHNzQ+fOnWG1WvHll19i9erVKC0txd69e5GZmYkhQ4agqakJ27dvx7lz55Cens4V48jISAQHB6OwsBDvvfcedu3aheLiYmzfvh3Nzc3o0qULFAoFduzYAT8/P/Tr1w979+7Fm2++iaNHj6KgoACZmZnQ6XSYOXMmamtr8frrr+Pw4cMoLCzErl270NDQgB49ekAQhX251dumlhC3VeK/t2JeidNfWVmJgwcPIiIiAgMHDuRTuszqz9oLabshc2VcaZ5dK6VP7qEpUkFmpu/Q0FCMHz8eAwYMgIoucZcb1ZaR8+TfhUCnbVxcXDB58mRkZmYiLy8PAJCfn4+TJ0/itttug1arRWFhIU6ePIlnn30WGzZswMaNG9G1a1ds27aNT5elpKRgyJAhWLNmDWbMmIHq6mrYbDY0Nzdj8eLFcHZ2xo8//oj169fjnnvuwXfffYeTJ08CAJqbm2EymQA6eKitrYVer4cgCNixYwdKSkqwefNmbNy4EXPnzkVDQwOMRiPUajUfhPXs2RMTJ07EyJEjMXv2bHh7e6O0tBQKhQIff/wxvvvuO7Rv3x5ffvklQK36x44dQ0xMDH788UdMnDgRixcvhr+/P9auXYvff/8ds2fPxrJly5CdnY2cnBxs3LgRL730EjZu3IgvvvgC/v7+aGpq4pa20NBQLF++HGvXrkVjYyN+/fVXmEwmfPPNN6itrcWqVauwfv16LFiwAD/99BO2b9/OfYZZPrz77rvw8fHBli1b8Nlnn0GgVlmTyYSEhATk5uZiyZIl2LBhA5577jnY6CIStkiGtV0yMpeDj48Pxo8fj0GDBsHNzY3LIvMDlNv1mwtZ6bsIhDr6hoaGomvXrvDy8oJApwRknwaZm4mxY8fC398fhw4dAgAcPHgQhBD07dsXoB3D3Llz4evri127duH3339HfX09ysvLYTAYAACdO3fGuHHjEB4eDi8vLyjoYofS0lJUVlZi5MiR8Pf3h4ODA6ZPn46IiAjs27cPJpMJjo6O0Gg0gEhpYYOr6OhoVFRU4Pnnn8eXX36JoKAg3H///dBoNCA03hgAuLu7w9vbG+7u7vD19YXFYoGXlxcmT56MwMBA+Pn5oW/fvqirq4PVaoXBYEBQUBAmT56MTp06wWAw4PTp07Barfj777+xbds26HQ6VFZW4vTp0wgODkZERAQ+/fRTLFmyBKWlpbj33nsRGBiI5uZmAMCwYcPg7u6Odu3aoX379tDr9SgoKEBWVhbGjBmD8PBwqFQqjB8/Hh07dkRycjKampqg1Wqh1WqRk5ODpqYm3HPPPXBzc4OPjw+mT5/Oldvo6GjodDr873//w/Lly+Hr64uHHnoIHh4esFgscHR0lK3tMpeN1WqFp6cnevToAVdXV1hFcR/Ffn8yNw9y63AJ2CiH+a4x/w+lvGRd5iZAoL6YAQEBiIuLQ2JiIiorK3Ho0CF06dIFoaGhAI3d9euvv+Lpp5/G+vXrkZqaCqPRCBcXF654OTk5cWsdu7dareb+sixWH1PyXF1dUVVVBRuNxcfqFLMuEBp7b+zYsXjnnXdQV1eHtWvXYu7cuVi+fDlMJhO/jln72P/st1arFR4eHvydtFotbDYbGhsbQQiBk5MT342gqakJLi4uKCwsRGJiIhITE5GTk4PRo0fDy8sLLi4ueO+999CrVy9s3boVL730El566SWcOXMGjo6OfBcDhslkgl6v5wNEV1dXfpx9Ly8v54NIlm6NRgNnZ2d+n+DgYPj7+6O5uRn9+vXD0qVL4erqiu+++w4PPfQQXn31VdTU1HD/QBmZy4X1ZcyarKAhj0wmE+/zZG4uZKXvMrDSJe9s5CM3rDI3C0S0krdPnz7Iz8/Hxo0bUV5ejnHjxvFBz9atW7Fx40Y8+uijWLx4MT744APExcUBAIxGI1QqFRobG7mCB9H0sUajgcFgQGlpKUBXvep0OpSUlCAqKgouLi4wmUxobGwE6GrC5uZmbinMy8tD+/btsXr1avz++++YPHkyNm/ejJSUFP4chl6v5/VTSYNOMyscqMJloy4aDg4OMJlM/J39/PwgCALuvvtufPzxx3jnnXewYMECjBkzBsHBwaivr0djYyNeffVV/Pbbb/j4449RXFyMn376CUajEXV1dXbpZ50mW0lcU1MD0PQ3NzejsbERoaGh0Gg0sNEwLWq1GjU1NSgoKOD3qaioQH19PTQaDXJyctCuXTusXLkSf/31Fx566CHs2rULu3fvBkS7h8jIXA6CaJW82Wzmih9T9mRZuvmQlb5LwEbfgmQbJBmZmwUmzz179kRkZCQ++ugjeHt782CtAFBeXo6GhgZER0fDzc0NGRkZ2Lx5M0pKSuDo6MgtAw6inTpYCKTQ0FB0794dGzZswJkzZ2C1WvHdd9+hoqIC3bp1g6urKwwGA/bt24fMzExs2bIF+/fvh0ajgVarxf79+/Hmm2+ioKAA/v7+6NKlC1xdXe2mg5niqlKpUFJSgvz8fBiNRl5vGWwlooruMtLc3MzT7+3tjY4dO+KPP/5AQUEBVCoVNm7ciEWLFkGn06G6uhpvvfUWtm/fDk9PT3To0AHu7u5wdXWFg4MDfx8GU/Z8fX3Rq1cv7Nq1C4cPHwYAbNmyBSdPnsTAgQOhpPsbg/oQx8bGYu3atSgpKUFRURG+++47mOjuIunp6Xjuuedw9OhR+Pr6okOHDnZ5LrudyFwOYmXORn1yxZZ22UXg5kVevXsJBEkwVab0yYqfzM2EIAh8mjM7OxtTp061Wx0bFBSE0tJSrFixAlu2bEFJSQn8/f3h6OiIkSNHwmg0oqKiAgMGDICXlxcAICMjA4GBgejVqxeioqJw5swZrFq1Cj/88AOKiorw5JNPYvjw4XB2doZWq0V8fDz27NkDAAgPD0enTp3Qp08fuLu7Iy0tDatWrcLq1auRlZWFu+++G7fddhu3TLD6qFAo8NdffyE7Oxu9e/dGWVkZunXrhoiICABAcXExjEYjRo0aBZ1Oh+LiYgwbNgxubm5QKpWIiYnB0aNH8f333+Pbb79FTk4OHnzwQYwaNQrOzs6oqanB+vXrsWLFCvz555/o3r07nnjiCSiVSuTm5iIuLg4hISEAgLS0NLi6umLMmDGIiYlBXl4eVq5cidWrV+PEiROYPXs2pk+fzvMqODgYPXr0QM+ePZGVlYUvvvgCe/fuhbe3NwICAjBo0CB06tQJaWlp+PHHH/HDDz8gLS0N999/P2bOnGkXK1FG5mJIZUQQbUnHvsuy1LpcaV5eq9W713UbNhkZGRkZGRkZmYtzrYxqsg1XRkZGRkZGRuYWQFb6ZGRkZGRkZGRuAWSlT0ZGRkZGRkbmFkBW+mRkZGRkZGRkbgFkpU9GRkZGRkZG5hZAVvpkZGRkZGRkZG4BZKVPRkZGRkZGRuYWQFb6ZGRkZGRkZGRuAf5RcOZJkyahT58+0sMyMjIyMjIyMjJXiclkwsKFC6WH/zH/SOmT9+iTkZGRkZGRkWldCCFXvHXb5fCPlD4ZGRkZGRkZGZl/B7KZTkZGRkZGRkbmFkBW+mRkZGRkZGRkbgFkpU9GRkZGRkZG5hZAVvpkZGRkZGRkZG4BZKVPRkZGRkZGRuYWQFb6ZNoMjY2N2LlzJyorK6WnWh2z2YzDhw/j+PHj0lOXjclkQnJyMo4dOyY9xWlqasKePXuQm5srPXVNMZvNOHr0KJKSkqSnZG4wjY2N2L59O86dOwdcphxdawghyMnJwZYtW6SnWgVxkIiGhgasX78ehw8ftrtGRkbm2tMqSt++ffvwyy+/tPjZvn07v66hoQEWi8Xut22JhoYG6aHLQqfTXVW6du3aZZdXv/76Kw4cOICioiLppf+YK8n7mpoa6aFrzv79+zF8+HDcdttt0lOtTklJCe68807ExcUhOztbevqyKCwsxL333ov+/fsjIyNDehqgaZo0aRJGjRrFO/jrQXFxMR5++GH06dMHe/fulZ6+oVyubF2JvF4N69atuyFKR0JCAkaPHo3x48fDYrFclhxdD1544QX06dMHn3/+ufTUPyI9PR3PPvssJk2ahIiICLzzzjt4+eWX8fPPP+Oll16SXi4jI3OtIa1AdnY2GT9+PBEEgfz8889k3759ZNeuXeSJJ54gw4cPJ4QQYrVaSUBAAHn//felP7/h7Nq1i8TExBClUkmCgoLI33//Lb2kRfLy8sjw4cOJs7MzcXJyIl988YX0kouSkZFBRo4cSQCQr7/+muzatYt8/fXX5LbbbiNxcXFk+/bt0p9cFVeS98uXLydarZZYrVbpqVanoqLC7vvy5ctJZGSk3bFrRUZGBgFAioqKpKcum1FDKY0AABz9SURBVIKCAgKAZGVlSU9xfvvtN6JSqUhzc7P01DWlqqqKACC7du2SnrphXK5sXYm8Xg16vZ5ER0eTxsZG6anrwvfff08CAgL498uRI0IIKS8vlx5qVbp27UpeffVV6eGrZsOGDeShhx4iNpuNEELI+vXrSXBwMDGZTNJLZWRkrhOtovQRQsijjz5KOnToYHfs3Llz5IEHHuDfP/roI5Kfn293zY3m6NGj5IknniAVFRWktLSUhIeHk0GDBkkvO4/6+noya9YskpmZSZqamsh//vMf4uDgQHQ6nfTSizJv3jwSFRUlPUw++OADolKpyP79+6WnrorLzfujR4+SlStXSg+3OqmpqWTkyJF2x+bMmUPuvvtuu2PXitWrV5PAwEDp4Svi999/J+7u7rxTa4nXX3+d9OzZU3r4mrNnzx4iCAKpra2VnrLjUgpYa3IlsnW58no1rFixgrzyyivSw9eNZ555hkyePJl/vxw5+uGHH8jTTz8tPdxqGAwG4uDgQDZv3iw9dVWUl5eTgIAAUlpayo89+eST5LHHHrO7TkZG5vrSakrf4MGDyYwZM6SHybFjx6SH2hSNjY12je29997LrZMXw2az2VkKVq5cSRwdHYler7e77lIMGjSIzJw5U3qYWK1WEhMTQ2JjY6WnbgruvPNOMm/ePLtjXbt2JZ9++qndsWvFE088QaZOnSo9fEW88sorZPTo0dLDdkycOJHMnTtXevias2jRItKpUyfpYTv++usv8s4770gP39TYbDYSGxtrp4xcbwYOHEjeffdd/v1y5Mjf35/8/PPP0sOtxuHDhwmAVsuX9957jwwcOJB/t9lsxM/Pj6xevdruOhkZmetLq/j0EUKQnp6Onj172h07deoUevToAQDYunUr7rvvPtGv/s8X8Nlnn8XYsWOxZMkSfPLJJ1ixYgUAYO3atbjnnnvsrv/iiy+wbNky/n3Dhg38nosXL8Zdd93FzxUXF+P555/HkCFDsHLlSn5ciouLC9/f7uzZs0hLS8PHH38svew8BEGAi4sLQJ3mf/75ZyxfvhyOjo7SSy+IzWZDeno6zyMxCoWC+/kYDAZ+vLm5Ge+//z7GjBmDZ555xs5Burq6Gq+99hoGDx6MDz74gB9vKe8TExNx++23Y/LkySgrKwOov9X8+fOxefNmu2uzs7Px0ksvYezYsZg9ezaqqqr4ub///hvTpk2D0WgEaJlOnjwZ69atE93h/1NWVoYFCxYgPj4eeXl5WLBgAc6ePYumpiacOnUK/fr1A+h9Z8yYgQ8//JD/Ni0tDVOnTkVlZSV27tyJcePGobi4GLhIvhBCsGrVKgwbNgyPP/44P56SksKfVVxcjEcffRQzZ87kz2LEx8fjySefxJAhQ/DGG2/AarXyc8nJyejbt6/d9TabDd999x3uueceDB48GImJiYiLi7O7RgohBD/++CMeeOAB9OnTB9OmTeP+bMXFxZg9ezZSU1MBAAUFBXjooYcwf/58yV2APXv24NFHH8WAAQOwfPnyiz63sbERs2fPxp9//okRI0Zg8ODBWLRoET9fWVmJefPmYceOHSgoKMCECROwe/dufv7gwYOYM2cORo0ahQMHDvDjAPDnn39i5syZGDt2LF/A0pJs6XQ6vPbaaxg0aBC++uorfrwlebXZbPj2228xZ84cDBkyBD///DM/ZzQa8dprr/F63tzcjNdeew2jR48W3eH/+PPPP9GnTx8EBgYCl0gnK5dp06ZhxowZKC0tPe9e0nSKf9dSeZrNZqSmptrJTUtyxPj777/x1FNPoby8HLt378bbb78Ni8Vy0fcGXUD00UcfYerUqejfvz/+97//2d0XAHJycvDCCy9g2LBhePjhhxESEsLzBRepUwBQV1eH999/H9OmTcOECROQlpYGACgtLcWCBQuwatUq6HQ6LFiwAAsWLMATTzyBiooK7Nq1C2+//bZdPZKRkbmOSLXAqyE7O5sAIOvXryd1dXWksrKSfPLJJ2TFihWEEEI+/PBD0qdPH9KtWzf+m5UrV5J7772XEELI5s2bCQCyePFiYjAYyJtvvkmGDRtmNy1mtVqJm5sb+e233wghhHzyySdkwoQJJDY2lnz99ddk8uTJ5LbbbiOEEBIfH0/uuece0tzcTFavXk1cXV0vOo1VV1dHvvjiCxIVFXVJvxoxJpOJbN68mfTs2ZOsW7dOevqSZGZmEgAX9N176aWX7Hx9iouLyfjx40lRUREpLCwkAMjhw4cJoVNns2fPJpWVleSPP/4g//3vfwm5QN7/8ccf5JVXXiHl5eXkgQceIEePHiUZGRlk3rx5BADZsGEDv/b3338nEydOJE1NTaShoYFERETwKftff/2VTJkyhTg4OBCz2UwWLVpE3n33XdKuXTvyxhtv8HuIqaurIzt37iQASFJSEjl37hwhhJC9e/cStVpNDAYDWbhwIfnggw9IXFwcueuuuwghhOzYsYPMnj2bACDx8fFk7NixpE+fPqS4uPii+TJ//nzy119/kVOnTpFJkyYRnU5HjEYjUavVZM+ePWTz5s3kmWeeIXPmzLHzsyLUAvPSSy8RQghJT08nCoWCfPfdd4RQy4W7uzvZtGkTv768vJyMHz+eHDhwgNhsNrJ7924CgJw4cYJfI6W0tJRMnjyZJCQkEKvVShYvXky6dOlCCCHk5MmT5JFHHiEASEJCAvnrr7/IE088QcaNG0eGDRvG72G1WsmcOXPIypUricFgIFVVVcTd3Z0sXbpU9KT/z+bNm3lenj17luj1erJkyRICgGzZsoUUFBSQ//73v8TX15d8//33ZMKECWTIkCHkzz//JIS6Hnz44YfEZrOROXPmkIkTJ/J7P/bYY+T3338nNTU15IEHHiAHDx5sUbZMJhOZMGECOXv2LPnzzz/JI488QsgF5FWv15MxY8aQLVu2EEIIWbZsGVEqleTMmTOkoaGBPPbYYyQkJIS89dZbJDc3l8yZM4c8+eSTRKlUErPZzO9DCCHDhg0jJ0+eJIT60l0onUajkdx1111kz549xGQykU6dOpFFixbx+7SUTnKJ8iS0ngIg1dXVhFxAjsSUlpaSd999l4SGhpKCggJSU1Nz0fcmhJDExEQyc+ZMkp2dTSwWC5kyZQp5/PHH7e67fPly8uyzz5KamhpiMpnIyJEjybRp0/j5i9Wp7Oxs0q9fP5KXl0dsNhuZMWMGiYiIIFarlTQ0NJDMzEyiUqnIJ598QnJyckhOTg55//33iVqtJqWlpcRisYjeREZG5nrSKkrfL7/8QgCQBx98kDz00EPkrrvuIs7OzuTIkSOE0CnUGTNm8Om8s2fPEgcHB1JYWEgIIaSystKuUWlsbCQzZ8608/84ceIEAcCd/2tra8mHH35I+vfvT958801+XVlZGQkKCiJVVVWEEELeeustMnv2bH6+JZYtW0Y++ugjEhsbS/z8/MihQ4ekl7TI5s2byeLFi8ntt99O1Go1+eqrr6SXXJR169YRAFzxkXLHHXcQAHwKul+/flxBTEhIIFFRUcRsNpPm5mbSpUuXFqdmpHlPCCFjx44l//vf/+yua2pq4osbysrKCCGE5ObmEldXV5Kbm8uvmzt3Ll9s0djYSL7++msSFxdHli1bRnbv3k0IIcTHx4esXbuW/0bK+vXriZ+fn92xDz/8kPTt25d8/PHHJCkpiRBCSO/evclbb71FCC3v5ORk4unpSR5//HHemV8sX2w2G3F0dCTx8fGiJxGSnJxMFAoF+e233/jAZP78+WTEiBH8ml9++YW0b9/eroPq1KkTmTNnDiGEkNOnTxMApKSkhJ+fMGGCXef9888/E2dn5wt2ciaTiXTv3p1s27aNH3v44Ye5wt7Y2EgyMzOJUqkkW7duJYsXLyaEEDJ9+nSuJBGad2J/r7KyMgKAJCcn82NibDYbWbduHQkKCuLHjEYjUSqVZOHChaS+vp6YTCbi4uJCpk+fToqLi/l1W7duJf379yeEKpujR48m33//PSGEkFWrVrXos9WSbCUlJRFXV9fzfA5bktd58+bZ+Qaz9K1evZpYrVai0+lIYGAg+fbbb8kzzzxD9Ho9WbZsGenYsSP/DaHlPmHCBP79Yul84YUXyMsvv0wIbZ/CwsLIqVOnCLlIOi9VnoQQ8uWXX9r58LYkR1Iee+wxcuedd/LvF3vvM2fOkKCgILtFUlFRUeTHH3/k3//++2/St29fO7eW6Oho8uGHHxJyiTplsVhI165d+eCHiNoxtiiKGQHS0tL4NS+//PJN66oiI/NvolWUvpdfftmuAyGEkIceeogYDAb+3d/fn1vD1qxZQ7y8vPi5o0ePEhcXF7tVXYGBgXaKw8KFC89b8HDHHXeQ8PBwOz+6+fPnk//85z8kJSWFvPjii2ThwoUX7HSl1NTUEDc3N/Kf//yHEELIqVOnyLJly/jnYishhw4dapemy+HFF18k/v7+0sOcDh06kLi4OEIIIVu2bCHe3t4kNzeXfPDBB+TJJ58kNTU1hBBC3n77bTJmzBjJr/8/4rwnhJD333+fKBQK8uWXX9pd98UXX9j5gT311FPnLWr573//S/r27cu/33PPPaR9+/bcgnPmzBkC4KJO+PPnzyeTJk2yOzZt2jQSEBDAFbTm5mbi4OBgl+eff/45USqV3FJDLpEvhBBy2223EW9vb26JIYSQpUuXEq1Wa2e5GTRokN3KxdjY2PNWMnbo0IG88MILhNAVmGKZ/+abb0hISIjo6v/LP7FFTsry5cvtFj9ZrVYSGRnJFVFC7+vt7W2npAcHB/NONyMjgzg4OJCcnBx+fuPGjdxqeiGeeeYZMmXKFP69pqaGQGSJO378OAFw3kCmd+/e5OOPP+bWuY0bNxJCCNHpdMTV1fWCK9+lstXc3Ew8PT1Jz549zxv0iOX13LlzRK1W28nB2bNnCQCydetWQkQyN2vWLL5K+p577rFTFAkhZMaMGWTv3r12x1pKZ3V1NVGr1SQxMZGsWrWKPPTQQ9xae7F0Xk55zp49m8yaNYt/l8pRS7A8F9PSexOaRrGSyVYGZ2dnE0KV+4iICLsFNdXV1UQQBJ43F6tTTLaampr473/88UeiUChIQ0MDIXR2QKVS2cnflClT7NItIyNzY2gVpe+2226zG0ETQuymVU6dOkUgchLetGkTUSgUfIrj7rvvtpuKYo04GzmeOnWKuLm5kXvuuYdfQwghQUFBZNmyZXbHoqOjyZgxY8iuXbsuOqV7Ifr27ctXkG7bto28/PLL/CPtMMS88MILl2y8pYwePZqMGzdOepgQaiUCwDu/efPmkQ4dOpCffvrpvBXCI0eOvODKPmneE0KIxWIhc+bMIQDI8uXL+fE777zTbtFBt27d7DoQQggZNWqUnZUjJCTE7vu3335L2rVrx7+3xODBg8nbb79tdywkJITMnz+ff09ISCBKpdJuscysWbPI9OnT+XdyiXwhNCxMnz59iFarJSkpKYQQQu677z7St29fLh9Go5FoNBry119/EUI7QQB21hGTyUScnJzI+vXrCaHWF/FCkP79+9tNj+Xm5hIXFxeuJLZEv379+PQ1ocooAHL06FF+7L777iM9evTgA6K8vDwCgCt5L7/8MvH29ubXG41G0qNHDzvFvCWkiwm2b99OnJ2deViQFStWkMDAQLt6zCxsTzzxxHlT1omJiQTAeZY7hlS2CLW8+fj4kA4dOpD6+npCWpDXDRs2ENBpaMb+/fsJAP6uq1atIi4uLiQvL49fExoaaqfY5OXl8QGUmJbS+fPPPxOVSkWWLl16nvX8Yum8nPKMiYmxW6wklSMper2eODg4kMTERLvjLb13fX39eTMO06ZNI25ubtyqt337dgKJy8Gjjz5qp7RdrE49+eST50VpeOedd0hMTAz//t5779lNaRNCSPv27e0GWTIyMjeGVlnIcezYMXTv3t3umEql4v8nJCSgY8eO3El49OjRGDVqFAYOHIhHHnkEkyZNwhNPPMGvT0xMRGhoKEJCQlBRUYFly5ahffv2do7pxcXFKC0txdChQ/kxdnzgwIEYPXo0FIr/Sx5bZHApCCEoKCjA9OnTAQDjx4/H+++/zz/Dhw+X/oSTl5eHadOmSQ9flJbyDXSxw8svv4wHHniAL04pLi5GcHAwZs2aBa1WC4jSdezYsfMWkNhsNqCFvAcApVKJlStX4oEHHrBb5JKQkIBhw4bx76WlpfxZoAFzDx48iHvvvRegiwqKi4vx1FNP8Wv279+PIUOG8O9SrFYrUlNT7cqyrKwMxcXFuPPOO/mxpKQk9OjRgy+WAYBDhw61WN4XyhcA8PX1RXx8PNq3b88XlyQnJ2P69OlcPlJTU2E2mzFgwACAphuAXdr37NkDLy8vjBkzBqALQcRpEC/Iqa2txZIlS2C1Wi+4mIIQYlf+SUlJOHLkCNRqNbp27cqvS0xMxLx58+Dg4ADQ/A0MDET79u0BuriFPZcQgtdeew0ALvhcXGAxweLFi/HOO+/Az88PoHk9YMAAu3p89uxZAMC0adP4OxJCYDKZcOzYMQiCAI1Gw68XO+tLZQv0Hffu3YvS0lLs27ePXyeW15bKYuvWrZgwYQJ/18TERB78FwCKiopQVFRkJ4effPIJnn32Wf6d0VI6i4uLIQgCHnnkEf4e4rrWUjovpzwbGxuRmZlpVzZSOZJy7Ngx2Gw29O7d2+54S+994sQJmEwm/g7Lly+HxWJBz549+WK1tLQ0aDQadO7cGQCwatUq5ObmonPnznB1dQUuUaekbQJoecyePZt/P3nypN3itObmZuTn56NXr178mIyMzI3hHyt9Z86cQUVFBW9wWyIxMRHR0dF47733YDab4eLigsGDB2P69OlYsWIFZs2aZXd9WloagoKCcPjwYbzwwgt4++23cerUKRBCsHz5coB23E5OTrzxYgQEBGD37t2w2WxobGzEa6+9hl27dtldA7r6bO7cuXY7BHz00Ud44okncMcdd9hdK2X+/Pl8tRoAHDhwACaTia9+PHbsGGbMmHHR3QcyMzNRXV0Nf39/fowQgu3bt2PQoEG4//778d133/FzAQEBSE1NRXV1NSwWC1avXs1XGbu7u2PPnj0wGAwoLS3Fa6+9hry8PKCFvL/rrrtw9OhRCIKA4cOHY9SoUQBdoVteXg6dToc//vgDABAZGYmDBw/yd3j33Xfx4osvcuUoMTERPj4+6NSpE78mKSkJvr6+F1wBfebMGTQ3NwMAnn/+eVitVqSkpECtVtt1FAcPHkRERATeffddvloxLy/vvM7vQvlSVFSE6dOno7q6Gm5ubujduzdGjhyJuro6nDlzhq/cZc8KCQnBr7/+iqysLLRr1w5qtZqn3Wq14p133sGaNWvg4eEBq9WKEydOQKVSYcGCBSCEwGw2o7q6GhkZGXjxxRcxe/Zs6PV61NTUYM2aNaI3/j+MRiNMJhNMJhN27tyJ3bt3o1evXggKCsKXX36JEydOoKioCIWFhRg0aBD/XVJSEjp16oS33nqL36OmpgaVlZV46qmncN9996G8vBwuLi5YuHCh3TMZx48fh8Fg4Pm9dOlShISE2CnvycnJLeY1AOzcuROgytVDDz2Es2fPwt3dHYQQbNu2DVarFVu2bOG7O0hla9u2bXjmmWdgs9nQpUsXREVFcQVUKq+RkZEALSPQAcLGjRvt6kZiYuJ5eaTVarF//34cP34cNTU1iI+P54M5MRdKp9ls5opoUlIS5s2bB4vFcsF0Xk55pqenw2azobi4GCtWrDhPjloiPT0dvr6+SE1NtatTLb13Y2MjAECv1+PTTz9FUFAQFAoFvL29sWDBArt3rKysxKJFi+Ds7Axvb2+Ehobi7bffRn19/QXrFGibkJGRwXcv2rZtG7Rard1q8oyMDLt3O3XqFARBuKhyKyMjc52Qmv6uhJUrV5Lu3bsTACQ6Opo7mkvp27cv6dmzp930S69evYi3tzdZtWqVnUMxIYQ8/vjjBABfgVtaWkoEQSD//e9/+XTG/PnzW4ynt3z5ciIIAvH39yf9+vW7YHDj5uZm0r17d+Ls7ExGjhxJHnjgAfL7779LL2uRhx9+mCiVStKvXz8yc+ZM8vnnn9v5ryxatIgoFAo7PysxX331FenWrRsBQNzc3PiilalTp5IlS5a0OHWUkpJCXFxciIuLC+ncubOdnxCbRvLz8yO333673ZSUNO9vv/124ujoSEaOHElefPFFnp9bt24larWaO3MTulOJs7MzmTVrFnniiSfIokWL7Pwj58yZc16MQWdnZzJp0iQ+XSclNTWVACC33XYbn95/5ZVXzvMd7NGjB4mNjSVnzpwhhPoZabXa86abLpQvZWVlJCoqivj4+JCJEyeSVatWEUII2blzJ3FycuJTWYT6t/n6+vLVj4T6kGq1WvLMM8+QBx98kK8cJXQaDQCZNm0af5+xY8cShUJB5s6dS/R6PYmPjydqtZq89tpr/HdS2LQz89d67rnnSHBwMJ/KYwGkxfVj4sSJpFOnTuT06dOE0PcEQIYPH06KioqI2WwmCoWCTJkyxS6NYpYvX04AkL59+5JHHnnETpYITZ9KpTpvAQwhhMycOZMAIGFhYWTKlCl8MVZ9fT2JjIwkgiCQ7t27291TKlvx8fHE3d2dREVFkTvuuMPOP04qr1arlYwfP56Eh4eTl156iTz44IN8QQUhhBQVFRG1Wk3S09P5sUWLFhEPDw/ub/juu++STz75hJ9nXCidDQ0NJCoqiqhUKtKhQwfy7LPPcl/Bi6XzUuW5efNm4uDgwKfVW5IjKUuWLCFKpZLMnz+fuyNc6L3LysqIq6sradeuHfeB7NWrF+nfvz9fKHLgwAEiCAKJiIgg+/btI4QQMmTIEBITE8Nl6kJ1itBVvZGRkWTQoEHkpZdeIk899ZSdD63RaCQODg52i+G++uor0q9fP/5dRkbmxvGPlL7LpaKiws6/7ueffyZz584lzzzzDFGr1WTAgAF222E1Njaetz0WW/XHqKysPE9ZZBQVFZHKykrp4fOwWq0kOzubr/S9Es6ePcs7PClWq/W8920NampqLrhAoqioqMVOXpr3hDbc0mutVut526IR6rienZ193j0ILQMpTJG7GAUFBXbfW3pH6X3q6uqI0Wi0O8a4WL7k5OTYLRCqq6s7L4B2Y2Nji/eurKw8Tw4Z0uM6nc5uBabNZjsvDVJ0Op2dgs9WZTJqamrOW4RUW1trJ/dWq9VudTWhW7BdjPvvv5/cdddd5MyZM+fdn9COu6WBByMzM7NFJUWv15/3LuQCsmU0Gklubu55dbglWSBUZsTKBaO5ufm8d9Hr9VxJYyFFWtpy7WLpNBgMJDMzs8X8uVA6L1WeRqPxvEUrUjmSotPpztt+7WLvXVlZaSfLLclCQUGB3Xu1JKcXq1MWi4VkZWWdV48IIeTIkSPEx8fHrgzvvffeFpVuGRmZ649AxBE3rzE2mw1z5syBv78/D7p78uRJTJkyBb6+vjh06JD0JzIyMq1M586d8cgjj7To4yYj80/45JNPUFtbi7fffhugbgy9evXCwYMH4ebmJr1cRkbmOvOPffquhMOHD+OHH37AyJEj+bGuXbvimWeeOc85WEZGpvWpqalBVlaW3e45MjKtxc8//4zQ0FCADvLfeOMNfPLJJ7LCJyPTRriuSl9sbCz69++P2bNn44033sCff/6J999/H8nJyVi9erX0chkZmVakoaEB77//PgghSExMRElJifQSGZkrZvHixYiMjMTtt9+O8ePH4/XXX0dwcDCmTZuGu+66C2PHjpX+REZG5gZxXad3QUd/u3btwokTJ+Di4oKBAwciNjZWepmMjEwrk5CQgJKSEri4uMDFxQVdu3blYU9kZK6W06dPY9++fZg6dSoPcWM2m3mYIRkZmbbD/wOumo+Z9+lyIgAAAABJRU5ErkJggg=="}}},{"cell_type":"code","source":"y_scores = sgd_clf.decision_function([some_digit])\ny_scores\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:54.195857Z","iopub.execute_input":"2025-06-18T07:54:54.196139Z","iopub.status.idle":"2025-06-18T07:54:54.212127Z","shell.execute_reply.started":"2025-06-18T07:54:54.196115Z","shell.execute_reply":"2025-06-18T07:54:54.211255Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"threshold = 0\ny_some_digit_pred = (y_scores > threshold)\nprint(y_some_digit_pred)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:54.213060Z","iopub.execute_input":"2025-06-18T07:54:54.213458Z","iopub.status.idle":"2025-06-18T07:54:54.230453Z","shell.execute_reply.started":"2025-06-18T07:54:54.213432Z","shell.execute_reply":"2025-06-18T07:54:54.229634Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"threshold = 8000\ny_some_digit_pred = (y_scores > threshold)\ny_some_digit_pred","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:54.231520Z","iopub.execute_input":"2025-06-18T07:54:54.231831Z","iopub.status.idle":"2025-06-18T07:54:54.252378Z","shell.execute_reply.started":"2025-06-18T07:54:54.231802Z","shell.execute_reply":"2025-06-18T07:54:54.251576Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":" y_scores = cross_val_predict(sgd_clf, x_train, y_train_1, cv=3, method=\"decision_function\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:54.253355Z","iopub.execute_input":"2025-06-18T07:54:54.253620Z","iopub.status.idle":"2025-06-18T07:54:54.902701Z","shell.execute_reply.started":"2025-06-18T07:54:54.253602Z","shell.execute_reply":"2025-06-18T07:54:54.901925Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from sklearn.metrics import precision_recall_curve\nprecisions, recalls, thresholds = precision_recall_curve(y_train_1, y_scores)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:54.903659Z","iopub.execute_input":"2025-06-18T07:54:54.904029Z","iopub.status.idle":"2025-06-18T07:54:54.910569Z","shell.execute_reply.started":"2025-06-18T07:54:54.903998Z","shell.execute_reply":"2025-06-18T07:54:54.909939Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def plot_precision_recall_vs_threshold(precisions, recalls, thresholds):\n    plt.plot(thresholds, precisions[:-1], \"b--\", label=\"Precision\")\n    plt.plot(thresholds, recalls[:-1], \"g-\", label=\"Recall\")\n    plt.xlabel(\"Threshold\")\n    plt.ylabel(\"Score\")\n    # Grid and legend\n    plt.grid(True)\n    plt.legend(loc=\"lower right\")\n    plt.title(\"Precision and Recall vs Threshold\")\nplot_precision_recall_vs_threshold(precisions, recalls, thresholds)\nplt.show()\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:54.911489Z","iopub.execute_input":"2025-06-18T07:54:54.911792Z","iopub.status.idle":"2025-06-18T07:54:55.109686Z","shell.execute_reply.started":"2025-06-18T07:54:54.911768Z","shell.execute_reply":"2025-06-18T07:54:55.108887Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"threshold_90_precision = thresholds[np.argmax(precisions >= 0.90)]\ny_train_pred_90 = (y_scores >= threshold_90_precision)\nprecision_score(y_train_1, y_train_pred_90)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:55.113282Z","iopub.execute_input":"2025-06-18T07:54:55.113542Z","iopub.status.idle":"2025-06-18T07:54:55.123738Z","shell.execute_reply.started":"2025-06-18T07:54:55.113522Z","shell.execute_reply":"2025-06-18T07:54:55.122903Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"recall_score(y_train_1, y_train_pred_90)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:55.124735Z","iopub.execute_input":"2025-06-18T07:54:55.125281Z","iopub.status.idle":"2025-06-18T07:54:55.143590Z","shell.execute_reply.started":"2025-06-18T07:54:55.125260Z","shell.execute_reply":"2025-06-18T07:54:55.142811Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from sklearn.metrics import roc_curve\n\nfpr, tpr, thresholds = roc_curve(y_train_1, y_scores)\n\ndef plot_roc_curve(fpr, tpr, label=None):\n    plt.plot(fpr, tpr, linewidth=2, label=label)\n    plt.plot([0, 1], [0, 1], 'k--') # dashed diagonal\n \nplot_roc_curve(fpr, tpr)\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:55.144354Z","iopub.execute_input":"2025-06-18T07:54:55.144571Z","iopub.status.idle":"2025-06-18T07:54:55.281084Z","shell.execute_reply.started":"2025-06-18T07:54:55.144556Z","shell.execute_reply":"2025-06-18T07:54:55.280222Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### ROC curve","metadata":{}},{"cell_type":"code","source":"from sklearn.ensemble import RandomForestClassifier\n\nforest_clf = RandomForestClassifier(random_state=42)\ny_probas_forest = cross_val_predict(forest_clf, x_train, y_train_1, cv=3,method=\"predict_proba\")\n\n ","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:55.281923Z","iopub.execute_input":"2025-06-18T07:54:55.282195Z","iopub.status.idle":"2025-06-18T07:54:58.947485Z","shell.execute_reply.started":"2025-06-18T07:54:55.282176Z","shell.execute_reply":"2025-06-18T07:54:58.946551Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":" But to plot a ROC curve, you need scores, not probabilities. A simple solution is to\n use the positive class’s probability as the score:","metadata":{}},{"cell_type":"code","source":"y_scores_forest = y_probas_forest[:, 1]   # score = proba of positive class\nfpr_forest, tpr_forest, thresholds_forest = roc_curve(y_train_1,y_scores_forest)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:58.948471Z","iopub.execute_input":"2025-06-18T07:54:58.949311Z","iopub.status.idle":"2025-06-18T07:54:58.955974Z","shell.execute_reply.started":"2025-06-18T07:54:58.949280Z","shell.execute_reply":"2025-06-18T07:54:58.954932Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Now you are ready to plot the ROC curve. It is useful to plot the first ROC curve as\n well to see how they compare (Figure 3-7):","metadata":{}},{"cell_type":"code","source":"plt.plot(fpr, tpr, \"b:\", label=\"SGD\")\nplot_roc_curve(fpr_forest, tpr_forest, \"Random Forest\")\nplt.legend(loc=\"lower right\")\nplt.grid(True)\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:58.956887Z","iopub.execute_input":"2025-06-18T07:54:58.957254Z","iopub.status.idle":"2025-06-18T07:54:59.122132Z","shell.execute_reply.started":"2025-06-18T07:54:58.957224Z","shell.execute_reply":"2025-06-18T07:54:59.121189Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"roc_auc_score=(y_train_1, y_scores_forest)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:54:59.123093Z","iopub.execute_input":"2025-06-18T07:54:59.123392Z","iopub.status.idle":"2025-06-18T07:54:59.128010Z","shell.execute_reply.started":"2025-06-18T07:54:59.123361Z","shell.execute_reply":"2025-06-18T07:54:59.127100Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Multiclass Classification\nDistinguish between two classes, multiclass classifiers (also called multinomial classifiers) can distinguish between more than two classes.","metadata":{}},{"cell_type":"code","source":"sgd_clf.fit(x_train, y_train)  # y_train, not y_train_5\nsgd_clf.predict([some_digit])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:55:10.295014Z","iopub.execute_input":"2025-06-18T07:55:10.295400Z","iopub.status.idle":"2025-06-18T07:55:16.511808Z","shell.execute_reply.started":"2025-06-18T07:55:10.295374Z","shell.execute_reply":"2025-06-18T07:55:16.511082Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"some_digit_scores = sgd_clf.decision_function([some_digit])\nsome_digit_scores","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:55:51.886413Z","iopub.execute_input":"2025-06-18T07:55:51.887349Z","iopub.status.idle":"2025-06-18T07:55:51.893484Z","shell.execute_reply.started":"2025-06-18T07:55:51.887323Z","shell.execute_reply":"2025-06-18T07:55:51.892549Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"np.argmax(some_digit_scores)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:56:11.572942Z","iopub.execute_input":"2025-06-18T07:56:11.573255Z","iopub.status.idle":"2025-06-18T07:56:11.579539Z","shell.execute_reply.started":"2025-06-18T07:56:11.573234Z","shell.execute_reply":"2025-06-18T07:56:11.578423Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"sgd_clf.classes_","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:56:20.280432Z","iopub.execute_input":"2025-06-18T07:56:20.280705Z","iopub.status.idle":"2025-06-18T07:56:20.287471Z","shell.execute_reply.started":"2025-06-18T07:56:20.280685Z","shell.execute_reply":"2025-06-18T07:56:20.286617Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"sgd_clf.classes_[1]","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:58:18.685192Z","iopub.execute_input":"2025-06-18T07:58:18.685475Z","iopub.status.idle":"2025-06-18T07:58:18.692120Z","shell.execute_reply.started":"2025-06-18T07:58:18.685457Z","shell.execute_reply":"2025-06-18T07:58:18.691106Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"this code creates a multiclass classifier using the OvO strategy, based on a SGDClassifier:","metadata":{}},{"cell_type":"code","source":"from sklearn.multiclass import OneVsOneClassifier\novo_clf = OneVsOneClassifier(SGDClassifier(random_state=42))\novo_clf.fit(x_train, y_train)\novo_clf.predict([some_digit])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:58:21.123422Z","iopub.execute_input":"2025-06-18T07:58:21.123832Z","iopub.status.idle":"2025-06-18T07:58:22.806080Z","shell.execute_reply.started":"2025-06-18T07:58:21.123708Z","shell.execute_reply":"2025-06-18T07:58:22.805300Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"len(ovo_clf.estimators_)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:58:37.333739Z","iopub.execute_input":"2025-06-18T07:58:37.334027Z","iopub.status.idle":"2025-06-18T07:58:37.339570Z","shell.execute_reply.started":"2025-06-18T07:58:37.334006Z","shell.execute_reply":"2025-06-18T07:58:37.338788Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"forest_clf.fit(x_train, y_train)\nforest_clf.predict([some_digit])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:58:56.930321Z","iopub.execute_input":"2025-06-18T07:58:56.930596Z","iopub.status.idle":"2025-06-18T07:59:00.370269Z","shell.execute_reply.started":"2025-06-18T07:58:56.930577Z","shell.execute_reply":"2025-06-18T07:59:00.369275Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":" forest_clf.predict_proba([some_digit])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:59:12.335410Z","iopub.execute_input":"2025-06-18T07:59:12.335730Z","iopub.status.idle":"2025-06-18T07:59:12.352857Z","shell.execute_reply.started":"2025-06-18T07:59:12.335693Z","shell.execute_reply":"2025-06-18T07:59:12.351973Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"cross_val_score(sgd_clf, x_train, y_train, cv=3, scoring=\"accuracy\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T07:59:38.252809Z","iopub.execute_input":"2025-06-18T07:59:38.253159Z","iopub.status.idle":"2025-06-18T07:59:47.871343Z","shell.execute_reply.started":"2025-06-18T07:59:38.253131Z","shell.execute_reply":"2025-06-18T07:59:47.870401Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from sklearn.preprocessing import StandardScaler\nscaler = StandardScaler()\nx_train_scaled = scaler.fit_transform(x_train.astype(np.float64))\ncross_val_score(sgd_clf, x_train_scaled, y_train, cv=3, scoring=\"accuracy\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T08:00:27.995859Z","iopub.execute_input":"2025-06-18T08:00:27.996184Z","iopub.status.idle":"2025-06-18T08:00:54.401877Z","shell.execute_reply.started":"2025-06-18T08:00:27.996161Z","shell.execute_reply":"2025-06-18T08:00:54.401030Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# Error Analysis","metadata":{}},{"cell_type":"code","source":"y_train_pred = cross_val_predict(sgd_clf, x_train_scaled, y_train, cv=3)\nconf_mx = confusion_matrix(y_train, y_train_pred)\nconf_mx","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T08:03:21.001343Z","iopub.execute_input":"2025-06-18T08:03:21.001674Z","iopub.status.idle":"2025-06-18T08:03:47.196890Z","shell.execute_reply.started":"2025-06-18T08:03:21.001651Z","shell.execute_reply":"2025-06-18T08:03:47.195974Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":" plt.matshow(conf_mx, cmap=plt.cm.gray)\n plt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T08:04:02.031471Z","iopub.execute_input":"2025-06-18T08:04:02.031816Z","iopub.status.idle":"2025-06-18T08:04:02.204536Z","shell.execute_reply.started":"2025-06-18T08:04:02.031791Z","shell.execute_reply":"2025-06-18T08:04:02.203519Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"row_sums = conf_mx.sum(axis=1, keepdims=True)\nnorm_conf_mx = conf_mx / row_sums\n #Now let’s fill the diagonal with zeros to keep only the errors, and let’s plot the result:\nnp.fill_diagonal(norm_conf_mx, 0)\nplt.matshow(norm_conf_mx, cmap=plt.cm.gray)\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T08:05:36.225401Z","iopub.execute_input":"2025-06-18T08:05:36.225751Z","iopub.status.idle":"2025-06-18T08:05:36.412461Z","shell.execute_reply.started":"2025-06-18T08:05:36.225728Z","shell.execute_reply":"2025-06-18T08:05:36.411659Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Define this before calling plot_digits\ndef plot_digits(instances, images_per_row=10, **options):\n    size = 28  # for MNIST: 28x28 images\n    images_per_row = min(len(instances), images_per_row)\n    n_rows = (len(instances) - 1) // images_per_row + 1\n    n_empty = n_rows * images_per_row - len(instances)\n    padded = np.concatenate([instances, np.zeros((n_empty, size * size))], axis=0)\n    image_grid = padded.reshape((n_rows, images_per_row, size, size))\n    big_image = image_grid.swapaxes(1, 2).reshape((n_rows * size, images_per_row * size))\n    plt.imshow(big_image, cmap='binary', **options)\n    plt.axis(\"off\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T08:12:00.093988Z","iopub.execute_input":"2025-06-18T08:12:00.094297Z","iopub.status.idle":"2025-06-18T08:12:00.100599Z","shell.execute_reply.started":"2025-06-18T08:12:00.094277Z","shell.execute_reply":"2025-06-18T08:12:00.099735Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"cl_a, cl_b = 1, 5\nx_aa = x_train[(y_train == cl_a) & (y_train_pred == cl_a)]\nx_ab = x_train[(y_train == cl_a) & (y_train_pred == cl_b)]\nx_ba = x_train[(y_train == cl_b) & (y_train_pred == cl_a)]\nx_bb = x_train[(y_train == cl_b) & (y_train_pred == cl_b)]\n\nplt.figure(figsize=(8,8))\nplt.subplot(221); plot_digits(x_aa[:25], images_per_row=5)\nplt.subplot(222); plot_digits(x_ab[:25], images_per_row=5)\nplt.subplot(223); plot_digits(x_ba[:25], images_per_row=5)\nplt.subplot(224); plot_digits(x_bb[:25], images_per_row=5)\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-18T08:12:39.424158Z","iopub.execute_input":"2025-06-18T08:12:39.424431Z","iopub.status.idle":"2025-06-18T08:12:39.586570Z","shell.execute_reply.started":"2025-06-18T08:12:39.424414Z","shell.execute_reply":"2025-06-18T08:12:39.585786Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# Multilabel Classification","metadata":{}},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}