{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":8338236,"sourceType":"datasetVersion","datasetId":4952065}],"dockerImageVersionId":30698,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"A **classification decision tree** is a powerful machine learning model that is particularly suitable for **tasks** involving **classification** and **concept description**. Conceptually, it operates akin to a series of **if-then rules**, making it highly **interpretable** and intuitive for humans to understand. At each node of the tree, the algorithm evaluates a feature and makes a decision based on a set of rules, **splitting the data into smaller subsets**. This process continues recursively until a stopping criterion is met, resulting in a tree structure where each leaf node represents a class label or outcome. By **following the path from the root to a leaf node**, one can effectively describe the **decision-making process** of the model in terms of if-then rules, providing insight into how it classifies instances. This transparency and simplicity make classification decision trees invaluable for tasks such as concept description, where understanding the underlying logic behind classification decisions is crucial.","metadata":{}},{"cell_type":"markdown","source":"### 1.Read the Data  \nWe read the discretized version of the bank loan data that has been preprocessed in a pipeline in the linked notebook.We chose discretized features as they are more understandable compared to transformed ones in if-then rules.  \n\nhttps://www.kaggle.com/code/zahrazolghadr/bankloan-pipeline","metadata":{}},{"cell_type":"code","source":"import pandas as pd \n\nX_data = pd.read_csv('/kaggle/input/d/zahrazolghadr/bankloan-ready-to-dt/X_data_discretized_no_scaling')\nX_data = X_data.set_index(['Unnamed: 0', 'Unnamed: 1'])\n\nX_train = X_data.loc['train']\nX_test = X_data.loc['test']\n\ny_data = pd.read_csv('/kaggle/input/d/zahrazolghadr/bankloan-ready-to-dt/y_data_discretized_no_scaling (1)')\ny_data = y_data.set_index(['Unnamed: 0', 'Unnamed: 1'])\n\ny_train = y_data.loc['train'].astype('str')\ny_test = y_data.loc['test'].astype('str')\n\nX_train.info()\ny_train.info()","metadata":{"execution":{"iopub.status.busy":"2024-05-30T15:04:22.026042Z","iopub.execute_input":"2024-05-30T15:04:22.026537Z","iopub.status.idle":"2024-05-30T15:04:22.070249Z","shell.execute_reply.started":"2024-05-30T15:04:22.026501Z","shell.execute_reply":"2024-05-30T15:04:22.068797Z"},"trusted":true},"execution_count":62,"outputs":[{"name":"stdout","text":"<class 'pandas.core.frame.DataFrame'>\nIndex: 485 entries, 286 to 37\nData columns (total 8 columns):\n #   Column                 Non-Null Count  Dtype  \n---  ------                 --------------  -----  \n 0   ed                     485 non-null    float64\n 1   age_cat_cm             485 non-null    int64  \n 2   employ_cat_cm          485 non-null    int64  \n 3   address_cat_cm         485 non-null    int64  \n 4   income_ratio_cat_cm    485 non-null    int64  \n 5   debtinc_cat_cm         485 non-null    int64  \n 6   creddebt_ratio_cat_cm  485 non-null    int64  \n 7   othdebt_ratio_cat_cm   485 non-null    int64  \ndtypes: float64(1), int64(7)\nmemory usage: 34.1 KB\n<class 'pandas.core.frame.DataFrame'>\nIndex: 485 entries, 286 to 37\nData columns (total 1 columns):\n #   Column   Non-Null Count  Dtype \n---  ------   --------------  ----- \n 0   default  485 non-null    object\ndtypes: object(1)\nmemory usage: 7.6+ KB\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### 2.Fit the Model and Configure Parameters, then Predict Test Set\nHere you can see how to fit the **DecisionTreeClassifier** model and configure its parameters.\n\n\n#### Parameters Explanation:\n\n* #### criterion:\nThe criterion parameter defines the function to measure the quality of a split in a decision tree. It can be either **\"gini\"** for the Gini impurity or **\"entropy\"** for the information gain.\n\n* #### max_depth:\nThe max_depth parameter specifies the maximum depth of the decision tree. It limits the number of nodes in the tree, helping to prevent overfitting.\n\n* #### min_samples_split:\nThe min_samples_split parameter sets the minimum number of samples required to split an internal node. If the number of samples at a node is less than this value, the node will not be split.\n\n* #### min_samples_leaf:\nThe min_samples_leaf parameter specifies the minimum number of samples required to be at a leaf node. It prevents the creation of nodes that represent very few samples.\n\n* #### min_weight_fraction_leaf:\nThe min_weight_fraction_leaf parameter is similar to min_samples_leaf but instead represents the minimum weighted fraction of the sum total of weights required to be at a leaf node.\n\n* #### random_state:\nThe random_state parameter sets the seed for random number generation. It ensures reproducibility of results when the algorithm involves randomization, such as random splits.\n\n* #### min_impurity_decrease:\nThe min_impurity_decrease parameter specifies the minimum impurity decrease required for a split to happen. It helps control the growth of the tree by only allowing splits that reduce impurity by a certain amount.\n\n* #### class_weight:\nThe class_weight parameter is used to handle imbalanced classes by assigning weights to different classes. The \"balanced\" mode automatically adjusts weights inversely proportional to class frequencies.\n\n* #### ccp_alpha:\nThe ccp_alpha parameter is the complexity parameter used for Minimal Cost-Complexity Pruning. It represents the complexity threshold. Higher values lead to more pruning, resulting in simpler trees.\n","metadata":{}},{"cell_type":"code","source":"from sklearn.tree import DecisionTreeClassifier\nfrom sklearn.metrics import accuracy_score\n\n# Create a decision tree classifier\nclf = DecisionTreeClassifier(criterion='gini', max_depth=3, min_samples_split=20,min_samples_leaf=10,\n                             min_weight_fraction_leaf=0.0,random_state=111, min_impurity_decrease=0.0,\n                             class_weight=\"balanced\",ccp_alpha=0.0) \n                             \n\n# Train the classifier on the training data\nclf.fit(X_train, y_train)\n\n# Predict the labels for the test set\ny_pred = clf.predict(X_test)\ny_prob = clf.predict_proba(X_test)\n\n\n# Evaluate the model\naccuracy = accuracy_score(y_test, y_pred)\nprint(\"Accuracy for test set:\", accuracy)","metadata":{"execution":{"iopub.status.busy":"2024-05-30T15:38:24.439059Z","iopub.execute_input":"2024-05-30T15:38:24.439639Z","iopub.status.idle":"2024-05-30T15:38:24.464405Z","shell.execute_reply.started":"2024-05-30T15:38:24.439594Z","shell.execute_reply":"2024-05-30T15:38:24.463063Z"},"trusted":true},"execution_count":85,"outputs":[{"name":"stdout","text":"Accuracy for test set: 0.6666666666666666\n","output_type":"stream"}]},{"cell_type":"markdown","source":"The DecisionTreeClassifier model **attribues**:","metadata":{}},{"cell_type":"code","source":"clf.classes_\nclf.feature_importances_\nclf.feature_names_in_\nclf.max_features_\nclf.n_classes_\nclf.n_features_in_\nclf.n_outputs_\nclf.tree_","metadata":{"trusted":true},"execution_count":77,"outputs":[{"execution_count":77,"output_type":"execute_result","data":{"text/plain":"1"},"metadata":{}}]},{"cell_type":"markdown","source":"### 3.Check the Overfitting\nTo investigate whether the model is **overfit** or not, you can compare the evaluation metrics between the training and test sets. If the difference is noteworthy (typically more than 5%), the model may be overfitted.","metadata":{}},{"cell_type":"code","source":"# Predict on the training set\ny_train_pred = clf.predict(X_train)\n\n# Predict on the test set\ny_test_pred = clf.predict(X_test)\n\n# Calculate accuracy for the train and test sets\naccuracy_train = accuracy_score(y_train, y_train_pred)\naccuracy_test = accuracy_score(y_test, y_test_pred)\n\n# Calculate the difference in accuracy between train and test sets\naccuracy_diff = accuracy_train - accuracy_test\n\nprint(\"Accuracy for train set:\", accuracy_train)\nprint(\"Accuracy for test set:\", accuracy_test)\nprint(\"Difference in accuracy between train and test sets:\", accuracy_diff)","metadata":{"execution":{"iopub.status.busy":"2024-05-30T15:39:18.528034Z","iopub.execute_input":"2024-05-30T15:39:18.528444Z","iopub.status.idle":"2024-05-30T15:39:18.548773Z","shell.execute_reply.started":"2024-05-30T15:39:18.528415Z","shell.execute_reply":"2024-05-30T15:39:18.547355Z"},"trusted":true},"execution_count":86,"outputs":[{"name":"stdout","text":"Accuracy for train set: 0.6536082474226804\nAccuracy for test set: 0.6666666666666666\nDifference in accuracy between train and test sets: -0.013058419243986208\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### 4.Evaluate the Model\nTo assess classification performance beyond accuracy, you can utilize metrics **based on the confusion matrix**. Evaluating the model using metrics such as **recall**, **precision**, and **F1-score** provides insight into its performance across different aspects of classification. Additionally, you can examine the **Area Under the ROC Curve** (AUC) measure, which assesses the model's ability to distinguish between classes across various threshold values.","metadata":{}},{"cell_type":"code","source":"from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, \\\n                            confusion_matrix, classification_report, ConfusionMatrixDisplay\nimport matplotlib.pyplot as plt\n\n# Confusion Matrix\nconf_matrix = confusion_matrix(y_test, y_pred)\nprint(\"Confusion Matrix:\")\nprint(conf_matrix)\n\ndisp = ConfusionMatrixDisplay(confusion_matrix=conf_matrix,display_labels=clf.classes_)\ndisp.plot()\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-05-30T15:39:43.602291Z","iopub.execute_input":"2024-05-30T15:39:43.602720Z","iopub.status.idle":"2024-05-30T15:39:44.026311Z","shell.execute_reply.started":"2024-05-30T15:39:43.602687Z","shell.execute_reply":"2024-05-30T15:39:44.024935Z"},"trusted":true},"execution_count":87,"outputs":[{"name":"stdout","text":"Confusion Matrix:\n[[90 63]\n [ 7 50]]\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 2 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"report = classification_report(y_test, y_pred, labels=None, zero_division='warn')\nprint(\"Classification Report:\")\nprint(report)","metadata":{"execution":{"iopub.status.busy":"2024-05-30T15:40:15.497876Z","iopub.execute_input":"2024-05-30T15:40:15.498300Z","iopub.status.idle":"2024-05-30T15:40:15.528402Z","shell.execute_reply.started":"2024-05-30T15:40:15.498267Z","shell.execute_reply":"2024-05-30T15:40:15.527151Z"},"trusted":true},"execution_count":88,"outputs":[{"name":"stdout","text":"Classification Report:\n              precision    recall  f1-score   support\n\n           0       0.93      0.59      0.72       153\n           1       0.44      0.88      0.59        57\n\n    accuracy                           0.67       210\n   macro avg       0.69      0.73      0.65       210\nweighted avg       0.80      0.67      0.68       210\n\n","output_type":"stream"}]},{"cell_type":"code","source":"from sklearn.metrics import roc_curve, auc, roc_auc_score, RocCurveDisplay\nimport matplotlib.pyplot as plt\n\nfpr, tpr, thresholds = roc_curve(y_test,y_prob[:,1], pos_label=\"1\")\nroc_auc = auc(fpr, tpr)\n\ndisplay = RocCurveDisplay(fpr=fpr, tpr=tpr, roc_auc=roc_auc, estimator_name='Classification Tree')\ndisplay.plot()\nplt.show()\n\nroc_auc = roc_auc_score(y_test, y_prob[:,1])\nprint(\"ROC AUC Score:\", roc_auc)","metadata":{"execution":{"iopub.status.busy":"2024-05-30T15:42:53.273130Z","iopub.execute_input":"2024-05-30T15:42:53.273618Z","iopub.status.idle":"2024-05-30T15:42:53.528648Z","shell.execute_reply.started":"2024-05-30T15:42:53.273579Z","shell.execute_reply":"2024-05-30T15:42:53.527490Z"},"trusted":true},"execution_count":89,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"iVBORw0KGgoAAAANSUhEUgAAAjcAAAGwCAYAAABVdURTAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuNSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/xnp5ZAAAACXBIWXMAAA9hAAAPYQGoP6dpAABUqUlEQVR4nO3deXhM1/8H8PfMJDPZExFZDbEvtQTBD1WlIUqVrtQWUboQa2ntsdTSxVa0ShGUUopqaRRFUa1IRK1BJGJJQizZ15nz+8PX1EhCJmZyM5P363nmqbnL3M9cY+bdc869RyaEECAiIiKyEHKpCyAiIiIyJoYbIiIisigMN0RERGRRGG6IiIjIojDcEBERkUVhuCEiIiKLwnBDREREFsVK6gLKmlarxc2bN+Ho6AiZTCZ1OURERFQCQgikp6fD29sbcvmT22YqXLi5efMm1Gq11GUQERFRKVy7dg1Vq1Z94jYVLtw4OjoCeHBynJycJK6GiIiISiItLQ1qtVr3O/4kFS7cPOyKcnJyYrghIiIyMyUZUsIBxURERGRRGG6IiIjIojDcEBERkUVhuCEiIiKLwnBDREREFoXhhoiIiCwKww0RERFZFIYbIiIisigMN0RERGRRGG6IiIjIokgabv7880/06NED3t7ekMlk2LFjx1P3OXjwIJo3bw6VSoXatWsjLCzM5HUSERGR+ZA03GRmZqJp06ZYtmxZibaPi4tD9+7d0bFjR0RHR2P06NEYMmQI9uzZY+JKiYiIyFxIOnHmyy+/jJdffrnE2y9fvhw1atTA/PnzAQANGjTAkSNHsHDhQgQGBpqqTCIiIioBrVbgdkYucvO1qFbZTrI6zGpW8GPHjiEgIEBvWWBgIEaPHl3sPrm5ucjNzdU9T0tLM1V5REREFi23QIPE+zm4cT8bN+5lP/jv//58MzUbifdzkKfR4vnabvh+SGvJ6jSrcJOUlAQPDw+9ZR4eHkhLS0N2djZsbW0L7TN37lzMmDGjrEokIiIyW6nZ+bj5eHB55Pnt9NynvoZCLkOBVlsG1RbPrMJNaUycOBFjx47VPU9LS4NarZawIiIiorL3sMvo+r3sBwHmkdDyMNCk5xY89XVsrOXwcbGFTyU7+LjYomolW3i72MDHxQ4+lWzh4aiClULai7HNKtx4enoiOTlZb1lycjKcnJyKbLUBAJVKBZVKVRblERERSeZJXUY37mcjMTUb+Rrx1NdxtVfCx0U/sPi4/O9RyRaV7Kwhk8nK4B2VnlmFmzZt2mD37t16y/bu3Ys2bdpIVBEREVHZMFaXkaeTjS6o6AcYG3i72MJOaVbRoEiSvoOMjAxcvnxZ9zwuLg7R0dFwdXVFtWrVMHHiRNy4cQPr1q0DAHzwwQdYunQpPv74YwwePBh//PEHfvzxR+zatUuqt0BERPTMTNVl5ONi87/gUn66jMqCpOHmxIkT6Nixo+75w7ExQUFBCAsLQ2JiIhISEnTra9SogV27dmHMmDFYvHgxqlatiu+++46XgRMRUblm2i6j//5sDl1GZUEmhHj62bQgaWlpcHZ2RmpqKpycnKQuh4iILEBqdv6Dy6HvG6fLyPuxFhdL6jIqLUN+vyvuWSIiIiqBR7uMHu0mYpdR+cVwQ0REFZoxu4wedBfZsstIYgw3RERk0UzZZeTtYvO/+7xU7C6j8oZ/E0REZLbKosvI28UGnk427DIyIww3RERUbj3eZXT9sQBT2i6jhy0u7DKyTAw3REQkmSK7jB4Z91KSLiO5DPBytv0vvLDLqMLj3zYREZnEk7qMHgYaw7uMCgcYdhnR4xhuiIioVHILNLh5P0cXWkrbZVTJzvqR+Yv0u4y8XWzgaq9klxEZhOGGiIiK9LDLSNfqwi4jMhP8RBERVWBCCJy+kYpjsXd03UfG6DLydn7wX3YZkRQYboiIKpiHgWbX6UTsPp2Ia3ezi9320S4jb5cH/2WXEZV3DDdERBXAkwKNrbUCL9R1Q213B70Awy4jMlf81BIRWainBZpO9d3RvYkXXqxXhSGGLAo/zUREFoSBhojhhojI7DHQEOnjp5yIyIwUaLRITM1Bwt0sJNzNwqXkDOw9n8RAQ/QIfuKJiMqZ1Ox8XPtfeHn4ePj8xr1sFGgL3xiPgYboP/z0ExGVscdbXx4PMPez8p+4v1IhR1VXW1RztUM1Vzv8X83KDDREj+C/BCIiE0jLyUfCHcNaXx7l5qCE+n/hpZqrne7P1SvbwcPRBnI57y1DVByGGyKiUjB268vDAFO9sh3Ulexgr+LXM1Fp8V8PEVEx2PpCZJ4YboiowmLrC5Fl4r88IrJobH0hqngYbojIrD1sfbl2NwtX2fpCRGC4ISIz8LD15dEAw9YXIioOww0RSY6tL0RkTPwXT0Rlgq0vRFRWGG6IyCTScvKx9mg89p5PZusLEZUpflsQkVE9DDXfHYlDarZ+oGHrCxGVBYYbIjKKokJNbXcHvPdCTTSp6szWFyIqM/ymIaJnUlyoGflSHXRv7AUFW2OIqIwx3BBRqTDUEFF5xXBDRAZhqCGi8o7hhohK7MCFWxi9OZqhhojKNYYbIiqRU9fu48MNkcjJ1zLUEFG5xnBDRE917W4W3l17Ajn5WnSoWwXfBfnDWiGXuiwioiLx24mInig1Ox+DwyKQkpGL+p6OWNq3GYMNEZVr/IYiomLlFWjx4feRuHQrAx5OKqwJbglHG2upyyIieiKGGyIqkhACk7afxl+xd2CvVGD1oJbwcraVuiwioqdiuCGiIi354zK2Rl6HQi7D0n7N8Zy3s9QlERGVCMMNERWy/eR1LNh7EQAw49Xn0LGeu8QVERGVHMMNEen5+8odfLz1XwDA+y/URP//qy5xRUREhmG4ISKdy7cy8N66E8jXCHRr7IlPutaXuiQiIoMx3BARACAlIxfBYceRllOAZtVcsOBtP8h5gz4iMkMMN0SEnHwNhqw9gWt3s1HN1Q7fDfSHjbVC6rKIiEqF4YaogtNqBUZvikb0tftwtrXGmuCWqOygkrosIqJSY7ghquDm/nYe4WeToFTIsWJAC9Sq4iB1SUREz4ThhqgCW38sHisPxwEAvnirCVrXrCxxRUREz47hhqiC+uNCMkJ3ngUAjOtSFz39fCSuiIjIOBhuiCqgMzdSEbLxJLQCeNu/KoZ3rC11SURERsNwQ1TB3LyfjcFhEcjK0+D52m6Y/VpjyGS85JuILAfDDVEFkp6Tj8FhEbiVnou6Hg74un9zWCv4NUBEloXfakQVRL5Gi2EbonAhKR1VHFVYPaglnGyspS6LiMjoGG6IKgAhBKbuOIPDl1Jga63A6qCWqFrJTuqyiIhMguGGqAL45lAsNkVcg1wGLHmnGRpXdZa6JCIik2G4IbJwv5y6ic/DYwAA015piICGHhJXRERkWgw3RBbsRPxdfLTlFABgcLsaGNSuhsQVERGZnuThZtmyZfD19YWNjQ1at26N48ePP3H7RYsWoV69erC1tYVarcaYMWOQk5NTRtUSmY+4lEwMXXcCeQVadG7ogcndG0hdEhFRmZA03GzevBljx45FaGgooqKi0LRpUwQGBuLWrVtFbr9x40ZMmDABoaGhOH/+PFatWoXNmzdj0qRJZVw5Ufl2NzMPwWuO415WPppWdcbiPn5QyHkvGyKqGCQNNwsWLMDQoUMRHByMhg0bYvny5bCzs8Pq1auL3P6vv/5Cu3bt0LdvX/j6+qJLly545513ntjak5ubi7S0NL0HkSXLydfgvXUnEH8nCz4utvguqCXslFZSl0VEVGYkCzd5eXmIjIxEQEDAf8XI5QgICMCxY8eK3Kdt27aIjIzUhZkrV65g9+7d6NatW7HHmTt3LpydnXUPtVpt3DdCVI5otQLjtpzCiav34GhjhbDglqjiqJK6LCKiMiXZ/86lpKRAo9HAw0P/yg0PDw9cuHChyH369u2LlJQUPP/88xBCoKCgAB988METu6UmTpyIsWPH6p6npaUx4JDF+uL3GPz6byKs5DJ8278F6ng4Sl0SEVGZk3xAsSEOHjyIOXPm4Ouvv0ZUVBS2bduGXbt2YdasWcXuo1Kp4OTkpPcgskQ/HE/ANwdjAQDz3miCtrXdJK6IiEgakrXcuLm5QaFQIDk5WW95cnIyPD09i9xn6tSpGDBgAIYMGQIAaNy4MTIzM/Hee+9h8uTJkMvNKqsRGc2hi7cxZccZAMDIl+rgzRZVJa6IiEg6kqUBpVKJFi1aYP/+/bplWq0W+/fvR5s2bYrcJysrq1CAUSgUAB7cXp6oIjqfmIbhG6Kg0Qq83swHYwLqSF0SEZGkJL2EYuzYsQgKCoK/vz9atWqFRYsWITMzE8HBwQCAgQMHwsfHB3PnzgUA9OjRAwsWLECzZs3QunVrXL58GVOnTkWPHj10IYeoIklOy8HgsAhk5BagdQ1XzH2jMWQyXvJNRBWbpOGmd+/euH37NqZNm4akpCT4+fkhPDxcN8g4ISFBr6VmypQpkMlkmDJlCm7cuIEqVaqgR48emD17tlRvgUgyGbkFCF4TgcTUHNSqYo8VA/yhsmLIJyKSiQrWn5OWlgZnZ2ekpqZycDGZrQKNFkPXncCBmNuobK/E9mHtUK0yZ/kmIstlyO83R+ASmRkhBKb/chYHYm5DZSXHd0H+DDZERI9guCEyM98djsP3fydAJgMW9/FDs2qVpC6JiKhcYbghMiO/nU7EnN/OAwAmd2uAro28JK6IiKj8YbghMhNRCfcwenM0hAAGtqmOd5+vIXVJRETlEsMNkRlIuJOFoWtPILdAi0713THtlYa85JuIqBgMN0Tl3P2sPAwKO447mXl4ztsJS95pBisF/+kSERWH35BE5VhugQbvr4/ElduZ8Ha2wepBLWGvkvT2VERE5R7DDVE5JYTAhJ9O45+4u3BQWWF1cEt4ONlIXRYRUbnHcENUTi3cdwnbT96AQi7D1/2ao74nbzpJRFQSDDdE5dCWE9fw1f5LAIDZvRrhhbpVJK6IiMh8MNwQlTNHL6dg4rbTAIBhL9ZCn1bVJK6IiMi8MNwQlSOXktPxwfeRKNAK9GjqjXFd6kldEhGR2WG4ISonbqXnYNCaCKTnFMC/eiV88WYTyOW8lw0RkaEYbojKgay8AgxZewI37mejhps9Vgz0h421QuqyiIjMEsMNkcQ0WoGRP0Tj3+upqGRnjTWDWsLVXil1WUREZovhhkhin+46h33nk6G0kmPlQH/4utlLXRIRkVljuCGS0JqjcVhzNB4AsODtpvD3dZW2ICIiC8BwQySR388mYeav5wAAn3Stj1eaeEtcERGRZWC4IZLAv9fvY9SmaAgBvNNKjQ861JS6JCIii8FwQ1TGrt/LwuCwE8jO1+CFulUws2cjyGS85JuIyFgYbojKUGp2PoLXRCAlIxf1PR2xrG8zWCv4z5CIyJj4rUpURvIKtPjw+0hcupUBDycVVg9qCUcba6nLIiKyOAw3RGVACIFJ20/jr9g7sFMqsCqoJbxdbKUui4jIIjHcEJWBpX9cxtbI65DLgGV9m6ORj7PUJRERWSyGGyIT23HyBubvvQgAmNmzETrWd5e4IiIiy8ZwQ2RCf1+5g4+3/gsAeO+Fmuj/f9UlroiIyPIx3BCZSOztDLy/PhJ5Gi1ebuSJCV3rS10SEVGFwHBDZAIpGbkIXhOB1Ox8NKvmgoW9/SCX8142RERlgeGGyMhy8jUYuu4EEu5moZqrHVYO9IeNtULqsoiIKgyGGyIj0moFxmyOxsmE+3C2tcaa4JZwc1BJXRYRUYXCcENkRPPCL+C3M0lQKuRYMaAFalVxkLokIqIKh+GGyEjW/30VK/68AgD4/M0maF2zssQVERFVTAw3REZw4MIthP58BgDwUee66NXMR+KKiIgqrmcKNzk5Ocaqg8hsnbmRiuEbo6AVwJstqiKkU22pSyIiqtAMDjdarRazZs2Cj48PHBwccOXKg2b4qVOnYtWqVUYvkKg8S0zNxrtrI5CVp0G72pUx57XGkMl4yTcRkZQMDjeffvopwsLC8Pnnn0OpVOqWN2rUCN99951RiyMqz9Jz8hG8JgLJabmo4+6Ar/u1gNKKPb1ERFIz+Jt43bp1WLFiBfr16weF4r97dzRt2hQXLlwwanFE5VW+RovhG0/iQlI6qjiqsCa4JZxtraUui4iIUIpwc+PGDdSuXXhMgVarRX5+vlGKIirPhBCY9vMZ/HnxNmytFVgV5I+qleykLouIiP7H4HDTsGFDHD58uNDyrVu3olmzZkYpiqg8W37oCn44fg0yGfDVO83QpKqL1CUREdEjrAzdYdq0aQgKCsKNGzeg1Wqxbds2xMTEYN26dfj1119NUSNRufHLqZv4LPxB9+u0Vxqic0MPiSsiIqLHGdxy07NnT/zyyy/Yt28f7O3tMW3aNJw/fx6//PILOnfubIoaicqFE/F38dGWUwCA4Ha+CG5XQ+KKiIioKDIhhJC6iLKUlpYGZ2dnpKamwsnJSepyyEzEp2Tita+P4l5WPjo39MDy/i2g4CzfRERlxpDfb4NbbmrWrIk7d+4UWn7//n3UrFnT0JcjKvfuZuZh0JrjuJeVjyZVnbG4jx+DDRFROWZwuImPj4dGoym0PDc3Fzdu3DBKUUTlRU6+Bu+tO4H4O1nwcbHFd0H+sFMaPFSNiIjKUIm/pXfu3Kn78549e+Ds7Kx7rtFosH//fvj6+hq1OCIpabUC47f+ixNX78HRxgprglvC3dFG6rKIiOgpShxuevXqBQCQyWQICgrSW2dtbQ1fX1/Mnz/fqMURSenL32Pwy6mbsJLLsLx/C9T1cJS6JCIiKoEShxutVgsAqFGjBiIiIuDm5mayooiktul4Ar4+GAsAmPt6Y7Srzc87EZG5MHjwQFxcnCnqICo3/rx4G5N3nAEAjOxUG2/5qyWuiIiIDFGqkZGZmZk4dOgQEhISkJeXp7du5MiRRimMSAoXktIwbEMUNFqB15r5YEznulKXREREBjI43Jw8eRLdunVDVlYWMjMz4erqipSUFNjZ2cHd3Z3hhsxWcloOgtdEICO3AK1ruGLeG40hk/GSbyIic2PwpeBjxoxBjx49cO/ePdja2uLvv//G1atX0aJFC3z55ZemqJHI5DJzCzA4LAKJqTmoWcUeKwb4Q2WlePqORERU7hgcbqKjo/HRRx9BLpdDoVAgNzcXarUan3/+OSZNmmSKGolMqkCjxYgfTuLszTRUtlcibFArONtZS10WERGVksHhxtraGnL5g93c3d2RkJAAAHB2dsa1a9eMWx2RiQkhMOOXc/jjwi2orOT4Lsgf1SrbSV0WERE9A4PH3DRr1gwRERGoU6cOOnTogGnTpiElJQXr169Ho0aNTFEjkcmsOhKH9X9fhUwGLOrth2bVKkldEhERPSODW27mzJkDLy8vAMDs2bNRqVIlfPjhh7h9+za+/fZboxdIZCrhZxIxe/d5AMCklxvg5cZeEldERETGwFnBqUI6mXAPfVb8jdwCLQb8X3XM7Pkcr4wiIirHTDoreHGioqLwyiuvGLzfsmXL4OvrCxsbG7Ru3RrHjx9/4vb379/H8OHD4eXlBZVKhbp162L37t2lLZsqoIQ7WRiy9gRyC7ToWK8KQns0ZLAhIrIgBoWbPXv2YNy4cZg0aRKuXLkCALhw4QJ69eqFli1b6qZoKKnNmzdj7NixCA0NRVRUFJo2bYrAwEDcunWryO3z8vLQuXNnxMfHY+vWrYiJicHKlSvh4+Nj0HGp4rqflYdBYcdxJzMPz3k7YWnf5rBSGC3jExFROVDibqlVq1Zh6NChcHV1xb1791C5cmUsWLAAI0aMQO/evTFq1Cg0aNDAoIO3bt0aLVu2xNKlSwE8mL9KrVZjxIgRmDBhQqHtly9fji+++AIXLlyAtXXJLtXNzc1Fbm6u7nlaWhrUajW7pSqg3AINBq46jn/i7sLL2QY7hreDhxNn+SYiMgcm6ZZavHgxPvvsM6SkpODHH39ESkoKvv76a5w+fRrLly83ONjk5eUhMjISAQEB/xUjlyMgIADHjh0rcp+dO3eiTZs2GD58ODw8PNCoUSPMmTMHGo2m2OPMnTsXzs7OuodazXmCKiIhBCb8dBr/xN2Fg8oKqwe1ZLAhIrJQJQ43sbGxeOuttwAAr7/+OqysrPDFF1+gatWqpTpwSkoKNBoNPDw89JZ7eHggKSmpyH2uXLmCrVu3QqPRYPfu3Zg6dSrmz5+PTz/9tNjjTJw4EampqboH78VTMS3cdwnbT96AQi7D1/2ao4EXW+2IiCxVie9zk52dDTu7Bzc3k8lkUKlUukvCy4pWq4W7uztWrFgBhUKBFi1a4MaNG/jiiy8QGhpa5D4qlQoqlapM66TyZWvkdXy1/xIA4NNejfBC3SoSV0RERKZk0E38vvvuOzg4OAAACgoKEBYWBjc3N71tSjpxppubGxQKBZKTk/WWJycnw9PTs8h9vLy8YG1tDYXivzl/GjRogKSkJOTl5UGpVBrydqgC+OtyCib89C8A4MMXa+GdVtUkroiIiEytxOGmWrVqWLlype65p6cn1q9fr7eNTCYrcbhRKpVo0aIF9u/fj169egF40DKzf/9+hISEFLlPu3btsHHjRmi1Wt0UEBcvXoSXlxeDDRVyKTkd738fiQKtwCtNvDC+Sz2pSyIiojJQ4nATHx9v9IOPHTsWQUFB8Pf3R6tWrbBo0SJkZmYiODgYADBw4ED4+Phg7ty5AIAPP/wQS5cuxahRozBixAhcunQJc+bMKXGgoorjVnoOBq2JQHpOAfyrV8KXbzWFXM572RARVQQGzy1lTL1798bt27cxbdo0JCUlwc/PD+Hh4bpBxgkJCboWGgBQq9XYs2cPxowZgyZNmsDHxwejRo3CJ598ItVboHIoK68AQ9aewI372fCtbIcVA/1hY614+o5ERGQROP0CWRSNVuCD7yOx91wyKtlZY9uwdqjhZi91WURE9IwkmX6BqDyYves89p5LhtJKjpUD/RlsiIgqIIYbshhhR+Ow+mgcAGD+W03h7+sqcUVERCQFhhuyCHvPJWPmr+cAAB93rYceTb0lroiIiKRSqnATGxuLKVOm4J133tFNcvnbb7/h7NmzRi2OqCROX0/FyB9OQiuAPi3V+LBDLalLIiIiCRkcbg4dOoTGjRvjn3/+wbZt25CRkQEAOHXqVLF3CSYylev3sjB4bQSy8zVoX8cNs3o1gkzGS76JiCoyg8PNhAkT8Omnn2Lv3r16N87r1KkT/v77b6MWR/QkaTn5GBwWgdvpuajv6Yiv+zWHtYI9rUREFZ3BvwSnT5/Ga6+9Vmi5u7s7UlJSjFIU0dPkFWjx4feRuJicAQ8nFVYPaglHG2upyyIionLA4HDj4uKCxMTEQstPnjwJHx8foxRF9CRCCEzefhpHL9+BnVKBVUEt4e1iK3VZRERUThgcbvr06YNPPvkESUlJkMlk0Gq1OHr0KMaNG4eBAweaokYiPcsOXMaWyOuQy4BlfZujkY+z1CUREVE5YnC4mTNnDurXrw+1Wo2MjAw0bNgQL7zwAtq2bYspU6aYokYinZ+jb+DL3y8CAGa8+hw61neXuCIiIipvSj39QkJCAs6cOYOMjAw0a9YMderUMXZtJsHpF8zXP1fuYMCq48jTaDG0fQ1M7t5Q6pKIiKiMGPL7bfDEmUeOHMHzzz+PatWqoVq1aqUuksgQsbcz8N76SORptHi5kScmvtxA6pKIiKicMrhbqlOnTqhRowYmTZqEc+fOmaImIj13MnIRvCYCqdn58FO7YGFvP8jlvJcNEREVzeBwc/PmTXz00Uc4dOgQGjVqBD8/P3zxxRe4fv26KeqjCi4nX4Mh604g4W4W1K62+C7IHzbWCqnLIiKicszgcOPm5oaQkBAcPXoUsbGxeOutt7B27Vr4+vqiU6dOpqiRKiitVmDM5micTLgPZ1trrBnUCm4OKqnLIiKicu6Zbudao0YNTJgwAfPmzUPjxo1x6NAhY9VFhM/CL+C3M0mwVsjw7YAWqO3uIHVJRERkBkodbo4ePYphw4bBy8sLffv2RaNGjbBr1y5j1kYV2Pd/X8W3f14BAHz+ZhP8X83KEldERETmwuCrpSZOnIhNmzbh5s2b6Ny5MxYvXoyePXvCzs7OFPVRBXTgwi1M+/kMAGBs57p4rVlViSsiIiJzYnC4+fPPPzF+/Hi8/fbbcHNzM0VNVIGdvZmKkI1R0ArgzRZVMaJTbalLIiIiM2NwuDl69Kgp6iBCYmo2BodFIDNPg7a1KmPOa40hk/GSbyIiMkyJws3OnTvx8ssvw9raGjt37nzitq+++qpRCqOKJT0nH8FrIpCclos67g74pn8LKK2eabw7ERFVUCWafkEulyMpKQnu7u6Qy4v/wZHJZNBoNEYt0Ng4/UL5k6/R4t21J/Dnxdtwc1Bh+7C2ULtyDBcREf3H6NMvaLXaIv9M9KyEEJj281n8efE2bKzlWBXkz2BDRETPxOB2/3Xr1iE3N7fQ8ry8PKxbt84oRVHF8e2fV/DD8QTIZMBXfZqhqdpF6pKIiMjMGRxugoODkZqaWmh5eno6goODjVIUVQy//nsT8367AACY2r0hujznKXFFRERkCQwON0KIIq9guX79OpydnY1SFFm+yKt3MfbHUwCAQW19Mfj5GhJXRERElqLEl4I3a9YMMpkMMpkML730Eqys/ttVo9EgLi4OXbt2NUmRZFniUzIxZO0J5BVoEdDAA1NfaSh1SUREZEFKHG569eoFAIiOjkZgYCAcHP6b50epVMLX1xdvvPGG0Qsky3IvMw/BYRG4l5WPxj7O+OodPyjkvJcNEREZT4nDTWhoKADA19cXvXv3ho2NjcmKIsuUk6/Be+tPIC4lEz4utlg1yB92SoPvI0lERPREBv+yBAUFmaIOsnBarcD4rf8iIv4eHFVWWBPcEu6ODMhERGR8JQo3rq6uuHjxItzc3FCpUqUn3hL/7t27RiuOLMf8vTH45dRNWMllWD6gBep6OEpdEhERWagShZuFCxfC0dFR92fO90OG2ByRgGUHYgEAc19vjHa1OeEqERGZTommX7AknH6hbB2+dBuD1kRAoxUY2ak2xnapJ3VJRERkhgz5/Tb4PjdRUVE4ffq07vnPP/+MXr16YdKkScjLyzO8WrJYF5LSMOz7KGi0Ar38vDGmc12pSyIiogrA4HDz/vvv4+LFiwCAK1euoHfv3rCzs8OWLVvw8ccfG71AMk/JaTkYvCYC6bkFaFXDFZ+92YTdmUREVCYMDjcXL16En58fAGDLli3o0KEDNm7ciLCwMPz000/Gro/MUGZuAQaHReBmag5qVrHHigEtoLJSSF0WERFVEKWafuHhzOD79u1Dt27dAABqtRopKSnGrY7MToFGixE/nMTZm2mobK9E2KBWcLFTSl0WERFVIAaHG39/f3z66adYv349Dh06hO7duwMA4uLi4OHhYfQCyXwIITDz13P448ItqKzkWBnkj2qV7aQui4iIKhiDw82iRYsQFRWFkJAQTJ48GbVr1wYAbN26FW3btjV6gWQ+Vh2Jw7pjVyGTAYt6+6F5tUpSl0RERBWQ0S4Fz8nJgUKhgLW1tTFezmR4KbhphJ9JwocbIiEEMKlbfbz3Qi2pSyIiIgtiyO93qSf2iYyMxPnz5wEADRs2RPPmzUv7UmTmTibcw+jNJyEE0P//qmFo+5pSl0RERBWYweHm1q1b6N27Nw4dOgQXFxcAwP3799GxY0ds2rQJVapUMXaNVI5du5uFIWtPICdfi471qmB6j+d4yTcREUnK4DE3I0aMQEZGBs6ePYu7d+/i7t27OHPmDNLS0jBy5EhT1EjlVGpWPgatOY47mXlo6OWEJX2bw0ph8EeKiIjIqAxuuQkPD8e+ffvQoEED3bKGDRti2bJl6NKli1GLo/Irr0CL978/gdjbmfB0ssHqQS3hoCp1LycREZHRGPy/2VqttshBw9bW1rr735BlE0Jgwk//4u8rd+GgssKa4JbwdLaRuiwiIiIApQg3nTp1wqhRo3Dz5k3dshs3bmDMmDF46aWXjFoclU+L9l3CtpM3oJDLsKxfczTw4lVnRERUfhgcbpYuXYq0tDT4+vqiVq1aqFWrFmrUqIG0tDQsWbLEFDVSObI18joW778EAJjVsxE61OUAciIiKl8MHiShVqsRFRWF/fv36y4Fb9CgAQICAoxeHJUvf8WmYOK2fwEAH3Sohb6tq0lcERERUWEGhZvNmzdj586dyMvLw0svvYQRI0aYqi4qZy4lp+P99ZHI1wh0b+KFjwPrSV0SERFRkUocbr755hsMHz4cderUga2tLbZt24bY2Fh88cUXpqyPyoHb6bkIDotAek4BWlSvhPlvNYVcznvZEBFR+VTiMTdLly5FaGgoYmJiEB0djbVr1+Lrr782ZW1UDmTnaTBkbQSu38uGb2U7rBzoDxtrhdRlERERFavE4ebKlSsICgrSPe/bty8KCgqQmJhoksJIekIIjN58Eqeup8LFzhprglvB1V4pdVlERERPVOJwk5ubC3t7+/92lMuhVCqRnZ1tksJIejHJ6dhzNhnWChlWDvRHDTf7p+9EREQkMYMGFE+dOhV2dna653l5eZg9ezacnZ11yxYsWGC86khSl29lAAAa+zijpa+rxNUQERGVTInDzQsvvICYmBi9ZW3btsWVK1d0zzlhomW5eicLAODLFhsiIjIjJQ43Bw8eNGEZVB7FpWQCAHwrM9wQEZH5KBdTOC9btgy+vr6wsbFB69atcfz48RLtt2nTJshkMvTq1cu0BVZQV+/8L9yw5YaIiMyI5OFm8+bNGDt2LEJDQxEVFYWmTZsiMDAQt27deuJ+8fHxGDduHNq3b19GlVY8cSn/65aqbPeULYmIiMoPycPNggULMHToUAQHB6Nhw4ZYvnw57OzssHr16mL30Wg06NevH2bMmIGaNWuWYbUVR0ZuAVIycgGw5YaIiMyLpOEmLy8PkZGRevNSyeVyBAQE4NixY8XuN3PmTLi7u+Pdd9996jFyc3ORlpam96Cni//feJvK9ko42VhLXA0REVHJSRpuUlJSoNFo4OHhobfcw8MDSUlJRe5z5MgRrFq1CitXrizRMebOnQtnZ2fdQ61WP3PdFcHDK6Wqs0uKiIjMTKnCzeHDh9G/f3+0adMGN27cAACsX78eR44cMWpxj0tPT8eAAQOwcuVKuLm5lWifiRMnIjU1Vfe4du2aSWu0FPEcTExERGbKoJv4AcBPP/2EAQMGoF+/fjh58iRycx+My0hNTcWcOXOwe/fuEr+Wm5sbFAoFkpOT9ZYnJyfD09Oz0PaxsbGIj49Hjx49dMu0Wu2DN2JlhZiYGNSqVUtvH5VKBZVKVeKa6AFeBk5ERObK4JabTz/9FMuXL8fKlSthbf3fWIx27dohKirKoNdSKpVo0aIF9u/fr1um1Wqxf/9+tGnTptD29evXx+nTpxEdHa17vPrqq+jYsSOio6PZ5WREvAyciIjMlcEtNzExMXjhhRcKLXd2dsb9+/cNLmDs2LEICgqCv78/WrVqhUWLFiEzMxPBwcEAgIEDB8LHxwdz586FjY0NGjVqpLe/i4sLABRaTs+Gl4ETEZG5MjjceHp64vLly/D19dVbfuTIkVJdlt27d2/cvn0b06ZNQ1JSEvz8/BAeHq4bZJyQkAC5XPIr1isUXgZORETmzOBwM3ToUIwaNQqrV6+GTCbDzZs3cezYMYwbNw5Tp04tVREhISEICQkpct3Tpn0ICwsr1TGpeLwMnIiIzJnB4WbChAnQarV46aWXkJWVhRdeeAEqlQrjxo3DiBEjTFEjlTFeBk5ERObM4HAjk8kwefJkjB8/HpcvX0ZGRgYaNmwIBwcHU9RHEuBl4EREZM4MDjcPKZVKNGzY0Ji1UDkRz8vAiYjIjBkcbjp27AiZTFbs+j/++OOZCiLpseWGiIjMmcHhxs/PT+95fn4+oqOjcebMGQQFBRmrLpIQLwMnIiJzZnC4WbhwYZHLp0+fjoyMjGcuiKTFy8CJiMjcGe0GMv3798fq1auN9XIkEV4GTkRE5s5o4ebYsWOwsbEx1suRRHgZOBERmTuDu6Vef/11vedCCCQmJuLEiROlvokflR8cTExERObO4HDj7Oys91wul6NevXqYOXMmunTpYrTCSBq8DJyIiMydQeFGo9EgODgYjRs3RqVKlUxVE0mILTdERGTuDBpzo1Ao0KVLl1LN/k3mIf4OLwMnIiLzZvCA4kaNGuHKlSumqIUklpFbgNvpvAyciIjMm8Hh5tNPP8W4cePw66+/IjExEWlpaXoPMl+8DJyIiCxBicfczJw5Ex999BG6desGAHj11Vf1pmEQQkAmk0Gj0Ri/SioTvAyciIgsQYnDzYwZM/DBBx/gwIEDpqyHJMTBxEREZAlKHG6EEACADh06mKwYkhYvAyciIktg0JibJ80GTuaPLTdERGQJDLrPTd26dZ8acO7evftMBZF0eBk4ERFZAoPCzYwZMwrdoZgsAy8DJyIiS2FQuOnTpw/c3d1NVQtJ6OodXgZORESWocRjbjjexrLFp/AycCIisgwlDjcPr5Yiy8TBxEREZClK3C2l1WpNWQdJjJeBExGRpTB4+gWyTGy5ISIiS8FwQwB4GTgREVkOhhvC1TuZusvAq7NbioiIzBzDDWHWr+cBAC/UrQJnW14GTkRE5o3hpoL78+Jt7DufDCu5DNNeaSB1OURERM+M4aYCy9doMeOXswCAoLa+qO3uKHFFREREz47hpgJb+1c8Ym9norK9EiNfqiN1OUREREbBcFNBpWTkYvG+SwCAj7vW41gbIiKyGAw3FdSXe2KQnluAxj7OeKuFWupyiIiIjIbhpgI6fT0Vm09cAwBMf7Uh5HLOG0ZERJaD4aaCEUJg+i9nIQTQy88bLaq7Sl0SERGRUTHcVDA/R99E5NV7sFMqMOFlXvpNRESWh+GmAsnMLcDc3x7csG94x9rwdLaRuCIiIiLjY7ipQJYduIzktFxUc7XDu8/XkLocIiIik2C4qSCu3snEd4fjAABTX2kIG2uFxBURERGZBsNNBTHr1/PI02jRvo4bAhq4S10OERGRyTDcVACPzh8V2qMhZDJe+k1ERJaL4cbCcf4oIiKqaBhuLBznjyIiooqG4caCPTp/1PhAzh9FREQVA8ONBdObP8qf80cREVHFwHBjoR6fP0rB+aOIiKiCYLixQJw/ioiIKjKGGwvE+aOIiKgiY7ixMJw/ioiIKjqGGwvD+aOIiKiiY7ixIJw/ioiIiOHGonD+KCIiIoYbi8H5o4iIiB5guLEAnD+KiIjoPww3FoDzRxEREf2H4cbMcf4oIiIifeUi3Cxbtgy+vr6wsbFB69atcfz48WK3XblyJdq3b49KlSqhUqVKCAgIeOL2lo7zRxEREemTPNxs3rwZY8eORWhoKKKiotC0aVMEBgbi1q1bRW5/8OBBvPPOOzhw4ACOHTsGtVqNLl264MaNG2VcufQ4fxQREVFhMiGEkLKA1q1bo2XLlli6dCkAQKvVQq1WY8SIEZgwYcJT99doNKhUqRKWLl2KgQMHPnX7tLQ0ODs7IzU1FU5OTs9cv1SEEHhz+TFEXr2HXn7eWNSnmdQlERERmYwhv9+Sttzk5eUhMjISAQEBumVyuRwBAQE4duxYiV4jKysL+fn5cHUtenLI3NxcpKWl6T0sAeePIiIiKpqk4SYlJQUajQYeHh56yz08PJCUlFSi1/jkk0/g7e2tF5AeNXfuXDg7O+searX5j0vh/FFERETFk3zMzbOYN28eNm3ahO3bt8PGpugf+IkTJyI1NVX3uHbtWhlXaXycP4qIiKh4VlIe3M3NDQqFAsnJyXrLk5OT4enp+cR9v/zyS8ybNw/79u1DkyZNit1OpVJBpVIZpd7y4NH5o6Z0b8D5o4iIiB4jacuNUqlEixYtsH//ft0yrVaL/fv3o02bNsXu9/nnn2PWrFkIDw+Hv79/WZRabjw6f1Tnhh5P34GIiKiCkbTlBgDGjh2LoKAg+Pv7o1WrVli0aBEyMzMRHBwMABg4cCB8fHwwd+5cAMBnn32GadOmYePGjfD19dWNzXFwcICDg4Nk76MscP4oIiKip5M83PTu3Ru3b9/GtGnTkJSUBD8/P4SHh+sGGSckJEAu/6+B6ZtvvkFeXh7efPNNvdcJDQ3F9OnTy7L0MvXo/FED23D+KCIiouJIfp+bsmau97n57vAVfLrrPCrbK/HHuBc5zQIREVUoZnOfGyoZzh9FRERUcgw3ZuDh/FGNfJw4fxQREdFTMNyUc3rzR/V4jvNHERERPQXDTTkmhMD0X85CCKCXnzf8fYueYoKIiIj+w3BTjnH+KCIiIsMx3JRTnD+KiIiodBhuyinOH0VERFQ6DDflEOePIiIiKj2Gm3KI80cRERGVHsNNOcP5o4iIiJ4Nw005wvmjiIiInh3DTTmy9q94xN7ORGV7JUYF1JG6HCIiIrPEcFNOcP4oIiIi42C4KSc4fxQREZFxMNyUA5w/ioiIyHgYbiTG+aOIiIiMi+FGYpw/ioiIyLgYbiTE+aOIiIiMj+FGQpw/ioiIyPgYbiTC+aOIiIhMg+FGIpw/ioiIyDQYbiTA+aOIiIhMh+GmjHH+KCIiItNiuCljnD+KiIjItBhuyhDnjyIiIjI9hpsyxPmjiIiITI/hpoxw/igiIqKywXBTBjh/FBERUdlhuCkDnD+KiIio7DDcmBjnjyIiIipbDDcmtu98MpLTclG1ki3njyIiIioDDDcmdupaKgAgoIEH548iIiIqAww3JnbmxoNw09jHWeJKiIiIKgaGGxPSaAXO3vxfuKnKcENERFQWGG5MKC4lA5l5GthaK1CrioPU5RAREVUIDDcmdPp/XVINvZ140z4iIqIywnBjQqevpwHgeBsiIqKyxHBjQhxMTEREVPYYbkyEg4mJiIikwXBjIhxMTEREJA2GGxPhYGIiIiJpMNyYCAcTExERSYPhxkQ4mJiIiEgaDDcmwMHERERE0rGSugBLxMHERA8IIVBQUACNRiN1KURkBqytraFQPPsk0ww3JsDBxERAXl4eEhMTkZWVJXUpRGQmZDIZqlatCgeHZ2sYYLgxAQ4mpopOq9UiLi4OCoUC3t7eUCqVkMkY9ImoeEII3L59G9evX0edOnWeqQWH4cYEOJiYKrq8vDxotVqo1WrY2dlJXQ4RmYkqVaogPj4e+fn5zxRuOKDYyDiYmOg/cjm/Yoio5IzVwstvHiPjYGIiIiJpMdwYGQcTExERSYvhxsg4mJiIiEhaDDdG9nAwcSOGGyKLJpPJsGPHDpMf5+DBg5DJZLh//75u2Y4dO1C7dm0oFAqMHj0aYWFhcHFxMXktL774IkaPHm3y45QXMTEx8PT0RHp6utSlWIS8vDz4+vrixIkTJj8Ww40RPTqYuAkHExOZraSkJIwYMQI1a9aESqWCWq1Gjx49sH///jKvpW3btkhMTISz83/fKe+//z7efPNNXLt2DbNmzULv3r1x8eJFox2zqEAFANu2bcOsWbOMdpzHTZ8+HTKZ7ImPsjRx4kSMGDECjo6OhdbVr18fKpUKSUlJhdb5+vpi0aJFhZZPnz4dfn5+esuk+qxt2bIF9evXh42NDRo3bozdu3c/cftBgwYV+ffx3HPP6bbx9fUtcpvhw4cDAJRKJcaNG4dPPvnEpO8NYLgxKg4mJiqeEAJZeQWSPIQQJa4zPj4eLVq0wB9//IEvvvgCp0+fRnh4ODp27Kj7ki5LSqUSnp6euh/2jIwM3Lp1C4GBgfD29oajoyNsbW3h7u5u8lpcXV2L/KE3lnHjxiExMVH3qFq1KmbOnKm37FF5eXkmqyUhIQG//vorBg0aVGjdkSNHkJ2djTfffBNr164t9TGk+qz99ddfeOedd/Duu+/i5MmT6NWrF3r16oUzZ84Uu8/ixYv1/h6uXbsGV1dXvPXWW7ptIiIi9LbZu3cvAOht069fPxw5cgRnz5412fsDeJ8bo+JgYqLiZedr0HDaHkmOfW5mIOyUJfu6GzZsGGQyGY4fPw57e3vd8ueeew6DBw8udr9PPvkE27dvx/Xr1+Hp6Yl+/fph2rRpsLa2BgCcOnUKo0ePxokTJyCTyVCnTh18++238Pf3x9WrVxESEoIjR47omu6/+OILdOvWDQcPHkTHjh1x7949REdHo2PHjgCATp06AQAOHDiA+Ph4jB49Wq+l5ZdffsHMmTNx+vRpODg4oH379ti+fTsAYP369Vi8eDFiYmJgb2+PTp06YdGiRXB3d0d8fLzuGJUqVQIABAUFISwsDC+++CL8/Px0rRL37t3DqFGj8MsvvyA3NxcdOnTAV199hTp16gAAwsLCMHr0aGzevBmjR4/GtWvX8Pzzz2PNmjXw8vIqdA4dHBz07kyrUCjg6OgIT09PAA+6xRo1agQrKyt8//33aNy4MQ4cOIAzZ85g/PjxOHz4MOzt7dGlSxcsXLgQbm5uAB7cVPKzzz7DihUrkJSUhLp162Lq1Kl48803i/37/PHHH9G0aVP4+PgUWrdq1Sr07dsXHTp0wKhRo0rdElHaz9qzWrx4Mbp27Yrx48cDAGbNmoW9e/di6dKlWL58eZH7ODs767Ue7tixA/fu3UNwcLBuWZUqVfT2mTdvHmrVqoUOHTrollWqVAnt2rXDpk2bTNoKWC5abpYtWwZfX1/Y2NigdevWOH78+BO3N7Q5raxwMDGRebt79y7Cw8MxfPhwvR+bh540rsXR0RFhYWE4d+4cFi9ejJUrV2LhwoW69f369UPVqlURERGByMhITJgwQRd8hg8fjtzcXPz55584ffo0PvvssyJvP9+2bVvExMQAAH766SckJiaibdu2hbbbtWsXXnvtNXTr1g0nT57E/v370apVK936/Px8zJo1C6dOncKOHTsQHx+va6FQq9X46aefADwYc5KYmIjFixcX+Z4HDRqEEydOYOfOnTh27BiEEOjWrRvy8/N122RlZeHLL7/E+vXr8eeffyIhIQHjxo0r9jw+zdq1a6FUKnH06FEsX74c9+/fR6dOndCsWTOcOHEC4eHhSE5Oxttvv63bZ+7cuVi3bh2WL1+Os2fPYsyYMejfvz8OHTpU7HEOHz4Mf3//QsvT09OxZcsW9O/fH507d0ZqaioOHz5s8Pt4ls/ahg0bdEGwuMeTajp27BgCAgL0lgUGBuLYsWMlrn/VqlUICAhA9erVi1yfl5eH77//HoMHDy7UndiqVatSnTNDSN5ys3nzZowdOxbLly9H69atsWjRIgQGBiImJqbIZtaHzWlz587FK6+8go0bN6JXr16IiopCo0aNJHgH/+FgYqLi2VorcG5moGTHLonLly9DCIH69esbfIwpU6bo/uzr64tx48Zh06ZN+PjjjwE86OYYP3687rUftm48XPfGG2+gcePGAICaNWsWeQylUqn7XnR1ddW1aDxu9uzZ6NOnD2bMmKFb1rRpU92fH20VqFmzJr766iu0bNkSGRkZcHBwgKurKwDA3d292B/ZS5cuYefOnTh69KguYG3YsAFqtRo7duzQdUXk5+dj+fLlqFWrFgAgJCQEM2fOLPI1S6JOnTr4/PPPdc8//fRTNGvWDHPmzNEtW716NdRqNS5evIjq1atjzpw52LdvH9q0aaN7z0eOHMG3336r16rwqKtXrxYZbjZt2oQ6deroxpr06dMHq1atQvv27Q16H8/yWXv11VfRunXrJ25TVIvTQ0lJSfDw8NBb5uHhUeT4oaLcvHkTv/32GzZu3FjsNjt27MD9+/eL7Nbz9vbG1atXS3Ss0pI83CxYsABDhw7VNW0tX74cu3btwurVqzFhwoRC25emOa0scDAx0ZPJZLISdw1JxZCxOY/bvHkzvvrqK8TGxiIjIwMFBQVwcnLSrR87diyGDBmC9evXIyAgAG+99ZbuB3/kyJH48MMP8fvvvyMgIABvvPEGmjRpUupaoqOjMXTo0GLXR0ZGYvr06Th16hTu3bsHrVYL4EHIatiwYYmOcf78eVhZWen9yFauXBn16tXD+fPndcvs7Ox07xMAvLy8cOvWLUPfkk6LFi30np86dQoHDhwosqUrNjYW+fn5yMrKQufOnfXW5eXloVmzZsUeJzs7GzY2NoWWr169Gv3799c979+/Pzp06IAlS5YYNB7pWT5rjo6OJh379DRr166Fi4sLevXqVew2q1atwssvvwxvb+9C62xtbU0+oa6k3VJ5eXmIjIzUax6Ty+UICAgotnnM0Oa03NxcpKWl6T1MgYOJicxfnTp1IJPJcOHCBYP2O3bsGPr164du3brh119/xcmTJzF58mS9Aa/Tp0/H2bNn0b17d/zxxx9o2LChbgzMkCFDcOXKFQwYMACnT5+Gv78/lixZUur3YWtrW+y6zMxMBAYGwsnJCRs2bEBERISuDlMM0H3Y9faQTCZ7ph/2x7twMjIy0KNHD0RHR+s9Ll26hBdeeAEZGRkAHnTVPbr+3Llz2Lp1a7HHcXNzw7179/SWnTt3Dn///Tc+/vhjWFlZwcrKCv/3f/+HrKwsbNq0Sbedk5MTUlNTC73m/fv3deNWSvtZA569W8rT0xPJycl6y5KTk4ttCXyUEAKrV6/GgAEDoFQqi9zm6tWr2LdvH4YMGVLk+rt37xYan2NskoablJQUaDQag5rHDG1Omzt3rm4glLOzM9RqtXGKf0xyWi5c7ZUcTExkxlxdXREYGIhly5YhMzOz0PrHL41+6K+//kL16tUxefJk+Pv7o06dOkU2u9etWxdjxozB77//jtdffx1r1qzRrVOr1fjggw+wbds2fPTRR1i5cmWp30eTJk2KvZT4woULuHPnDubNm4f27dujfv36hVpSHv5oaTSaYo/RoEEDFBQU4J9//tEtu3PnDmJiYkrc+mMMzZs3x9mzZ+Hr64vatWvrPezt7dGwYUOoVCokJCQUWv+k34NmzZrh3LlzestWrVqFF154AadOndILSmPHjsWqVat029WrVw+RkZGFXjMqKgp169YFUPrPGvCgW+rxMPf4o6gutYfatGlT6POxd+9eXbfdkxw6dAiXL1/Gu+++W+w2a9asgbu7O7p3717k+jNnzjyx1cwYysWAYlOaOHEiUlNTdY9r166Z5DjtarshckoA1gS3NMnrE1HZWLZsGTQaDVq1aoWffvoJly5dwvnz5/HVV18V++Vfp04dJCQkYNOmTYiNjcVXX32law0BHnRxhISE4ODBg7h69SqOHj2KiIgINGjQAAAwevRo7NmzB3FxcYiKisKBAwd060ojNDQUP/zwA0JDQ3H+/HndIGUAqFatGpRKJZYsWYIrV65g586dha5aqV69OmQyGX799Vfcvn1b1/rx+Hvu2bMnhg4diiNHjuDUqVPo378/fHx80LNnz1LXbqjhw4fj7t27eOeddxAREYHY2Fjs2bMHwcHB0Gg0cHR0xLhx4zBmzBisXbsWsbGxiIqKwpIlS554GffDHoGHAS8/Px/r16/HO++8g0aNGuk9hgwZgn/++Ud3efOYMWOwa9cuzJ49G+fPn8eZM2cwefJkHDt2DKNGjdIdozSfNeBBt9TjQe3xx5Na70aNGoXw8HDMnz8fFy5cwPTp03HixAmEhITotpk4cSIGDhxYaN9Vq1ahdevWxY5x1Wq1WLNmDYKCgmBlVXQ39OHDh9GlS5di6zMKIaHc3FyhUCjE9u3b9ZYPHDhQvPrqq0Xuo1arxcKFC/WWTZs2TTRp0qREx0xNTRUARGpqamlKJqISyM7OFufOnRPZ2dlSl1IqN2/eFMOHDxfVq1cXSqVS+Pj4iFdffVUcOHBAtw0Ave+u8ePHi8qVKwsHBwfRu3dvsXDhQuHs7CyEePBd16dPH6FWq4VSqRTe3t4iJCREd35CQkJErVq1hEqlElWqVBEDBgwQKSkpQgghDhw4IACIe/fuCSGEuHfvngCgV8uaNWt0x3rop59+En5+fkKpVAo3Nzfx+uuv69Zt3LhR+Pr6CpVKJdq0aSN27twpAIiTJ0/qtpk5c6bw9PQUMplMBAUFCSGE6NChgxg1apRum7t374oBAwYIZ2dnYWtrKwIDA8XFixefWNf27dtFSX96qlevrvd9//jxH7p48aJ47bXXhIuLi7C1tRX169cXo0ePFlqtVgghhFarFYsWLRL16tUT1tbWokqVKiIwMFAcOnSo2GPn5+cLb29vER4eLoQQYuvWrUIul4ukpKQit2/QoIEYM2aM7vmePXtEu3btRKVKlUTlypXFiy++WOTxSvJZM4Uff/xR1K1bVyiVSvHcc8+JXbt26a0PCgoSHTp00Ft2//59YWtrK1asWFHs6+7Zs0cAEDExMUWu/+uvv4SLi4vIysoqcv2TvjsM+f2WCfEMnZ9G0Lp1a7Rq1UrXv6zValGtWjWEhIQUOaC4d+/eyMrKwi+//KJb1rZtWzRp0qREA4rT0tLg7OyM1NRUvcF+RGQ8OTk5iIuLQ40aNYoclElkDpYtW4adO3dizx5p7s9kiXr37o2mTZti0qRJRa5/0neHIb/fkl+6MHbsWAQFBcHf3x+tWrXCokWLkJmZqbt6auDAgfDx8cHcuXMBPGhO69ChA+bPn4/u3btj06ZNOHHiBFasWCHl2yAiIgvz/vvv4/79+0hPT5f06iRLkZeXh8aNG2PMmDEmP5bk4aZ37964ffs2pk2bhqSkJPj5+SE8PFw3aDghIQFy+X9Dg9q2bYuNGzdiypQpmDRpEurUqYMdO3ZIfo8bIiKyLFZWVpg8ebLUZVgMpVKpdz8oU5K8W6qssVuKyPTYLUVEpWGsbimLv1qKiKRTwf7fiYiekbG+MxhuiMjoHt64zdR3ISUiy/LwRpIKRcmmTCmO5GNuiMjyKBQKuLi46G4OZ2dnV2jyPCKiR2m1Wty+fRt2dnbF3iOnpBhuiMgkHt7K/VnmESKiikUul6NatWrP/D9DDDdEZBIymQxeXl5wd3dHfn6+1OUQkRlQKpV6V0iXFsMNEZmUQqF45v5zIiJDcEAxERERWRSGGyIiIrIoDDdERERkUSrcmJuHNwhKS0uTuBIiIiIqqYe/2yW50V+FCzfp6ekAALVaLXElREREZKj09HQ4Ozs/cZsKN7eUVqvFzZs34ejoaPSbiqWlpUGtVuPatWuct8qEeJ7LBs9z2eB5Ljs812XDVOdZCIH09HR4e3s/9XLxCtdyI5fLUbVqVZMew8nJif9wygDPc9ngeS4bPM9lh+e6bJjiPD+txeYhDigmIiIii8JwQ0RERBaF4caIVCoVQkNDoVKppC7FovE8lw2e57LB81x2eK7LRnk4zxVuQDERERFZNrbcEBERkUVhuCEiIiKLwnBDREREFoXhhoiIiCwKw42Bli1bBl9fX9jY2KB169Y4fvz4E7ffsmUL6tevDxsbGzRu3Bi7d+8uo0rNmyHneeXKlWjfvj0qVaqESpUqISAg4Kl/L/SAoZ/nhzZt2gSZTIZevXqZtkALYeh5vn//PoYPHw4vLy+oVCrUrVuX3x0lYOh5XrRoEerVqwdbW1uo1WqMGTMGOTk5ZVStefrzzz/Ro0cPeHt7QyaTYceOHU/d5+DBg2jevDlUKhVq166NsLAwk9cJQSW2adMmoVQqxerVq8XZs2fF0KFDhYuLi0hOTi5y+6NHjwqFQiE+//xzce7cOTFlyhRhbW0tTp8+XcaVmxdDz3Pfvn3FsmXLxMmTJ8X58+fFoEGDhLOzs7h+/XoZV25eDD3PD8XFxQkfHx/Rvn170bNnz7Ip1owZep5zc3OFv7+/6Natmzhy5IiIi4sTBw8eFNHR0WVcuXkx9Dxv2LBBqFQqsWHDBhEXFyf27NkjvLy8xJgxY8q4cvOye/duMXnyZLFt2zYBQGzfvv2J21+5ckXY2dmJsWPHinPnzoklS5YIhUIhwsPDTVonw40BWrVqJYYPH657rtFohLe3t5g7d26R27/99tuie/fuestat24t3n//fZPWae4MPc+PKygoEI6OjmLt2rWmKtEilOY8FxQUiLZt24rvvvtOBAUFMdyUgKHn+ZtvvhE1a9YUeXl5ZVWiRTD0PA8fPlx06tRJb9nYsWNFu3btTFqnJSlJuPn444/Fc889p7esd+/eIjAw0ISVCcFuqRLKy8tDZGQkAgICdMvkcjkCAgJw7NixIvc5duyY3vYAEBgYWOz2VLrz/LisrCzk5+fD1dXVVGWavdKe55kzZ8Ld3R3vvvtuWZRp9kpznnfu3Ik2bdpg+PDh8PDwQKNGjTBnzhxoNJqyKtvslOY8t23bFpGRkbquqytXrmD37t3o1q1bmdRcUUj1O1jhJs4srZSUFGg0Gnh4eOgt9/DwwIULF4rcJykpqcjtk5KSTFanuSvNeX7cJ598Am9v70L/oOg/pTnPR44cwapVqxAdHV0GFVqG0pznK1eu4I8//kC/fv2we/duXL58GcOGDUN+fj5CQ0PLomyzU5rz3LdvX6SkpOD555+HEAIFBQX44IMPMGnSpLIoucIo7ncwLS0N2dnZsLW1Nclx2XJDFmXevHnYtGkTtm/fDhsbG6nLsRjp6ekYMGAAVq5cCTc3N6nLsWharRbu7u5YsWIFWrRogd69e2Py5MlYvny51KVZlIMHD2LOnDn4+uuvERUVhW3btmHXrl2YNWuW1KWREbDlpoTc3NygUCiQnJystzw5ORmenp5F7uPp6WnQ9lS68/zQl19+iXnz5mHfvn1o0qSJKcs0e4ae59jYWMTHx6NHjx66ZVqtFgBgZWWFmJgY1KpVy7RFm6HSfJ69vLxgbW0NhUKhW9agQQMkJSUhLy8PSqXSpDWbo9Kc56lTp2LAgAEYMmQIAKBx48bIzMzEe++9h8mTJ0Mu5//7G0Nxv4NOTk4ma7UB2HJTYkqlEi1atMD+/ft1y7RaLfbv3482bdoUuU+bNm30tgeAvXv3Frs9le48A8Dnn3+OWbNmITw8HP7+/mVRqlkz9DzXr18fp0+fRnR0tO7x6quvomPHjoiOjoZarS7L8s1GaT7P7dq1w+XLl3XhEQAuXrwILy8vBptilOY8Z2VlFQowDwOl4JSLRiPZ76BJhytbmE2bNgmVSiXCwsLEuXPnxHvvvSdcXFxEUlKSEEKIAQMGiAkTJui2P3r0qLCyshJffvmlOH/+vAgNDeWl4CVg6HmeN2+eUCqVYuvWrSIxMVH3SE9Pl+otmAVDz/PjeLVUyRh6nhMSEoSjo6MICQkRMTEx4tdffxXu7u7i008/leotmAVDz3NoaKhwdHQUP/zwg7hy5Yr4/fffRa1atcTbb78t1VswC+np6eLkyZPi5MmTAoBYsGCBOHnypLh69aoQQogJEyaIAQMG6LZ/eCn4+PHjxfnz58WyZct4KXh5tGTJElGtWjWhVCpFq1atxN9//61b16FDBxEUFKS3/Y8//ijq1q0rlEqleO6558SuXbvKuGLzZMh5rl69ugBQ6BEaGlr2hZsZQz/Pj2K4KTlDz/Nff/0lWrduLVQqlahZs6aYPXu2KCgoKOOqzY8h5zk/P19Mnz5d1KpVS9jY2Ai1Wi2GDRsm7t27V/aFm5EDBw4U+X378NwGBQWJDh06FNrHz89PKJVKUbNmTbFmzRqT1ykTgu1vREREZDk45oaIiIgsCsMNERERWRSGGyIiIrIoDDdERERkURhuiIiIyKIw3BAREZFFYbghIiIii8JwQ0RERBaF4YaI9ISFhcHFxUXqMkpNJpNhx44dT9xm0KBB6NWrV5nUQ0Rlj+GGyAINGjQIMpms0OPy5ctSl4awsDBdPXK5HFWrVkVwcDBu3bpllNdPTEzEyy+/DACIj4+HTCZDdHS03jaLFy9GWFiYUY5XnOnTp+vep0KhgFqtxnvvvYe7d+8a9DoMYkSGs5K6ACIyja5du2LNmjV6y6pUqSJRNfqcnJwQExMDrVaLU6dOITg4GDdv3sSePXue+bU9PT2fuo2zs/MzH6cknnvuOezbtw8ajQbnz5/H4MGDkZqais2bN5fJ8YkqKrbcEFkolUoFT09PvYdCocCCBQvQuHFj2NvbQ61WY9iwYcjIyCj2dU6dOoWOHTvC0dERTk5OaNGiBU6cOKFbf+TIEbRv3x62trZQq9UYOXIkMjMzn1ibTCaDp6cnvL298fLLL2PkyJHYt28fsrOzodVqMXPmTFStWhUqlQp+fn4IDw/X7ZuXl4eQkBB4eXnBxsYG1atXx9y5c/Ve+2G3VI0aNQAAzZo1g0wmw4svvghAvzVkxYoV8Pb2hlar1auxZ8+eGDx4sO75zz//jObNm8PGxgY1a9bEjBkzUFBQ8MT3aWVlBU9PT/j4+CAgIABvvfUW9u7dq1uv0Wjw7rvvokaNGrC1tUW9evWwePFi3frp06dj7dq1+Pnnn3WtQAcPHgQAXLt2DW+//TZcXFzg6uqKnj17Ij4+/on1EFUUDDdEFYxcLsdXX32Fs2fPYu3atfjjjz/w8ccfF7t9v379ULVqVURERCAyMhITJkyAtbU1ACA2NhZdu3bFG2+8gX///RebN2/GkSNHEBISYlBNtra20Gq1KCgowOLFizF//nx8+eWX+PfffxEYGIhXX30Vly5dAgB89dVX2LlzJ3788UfExMRgw4YN8PX1LfJ1jx8/DgDYt28fEhMTsW3btkLbvPXWW7hz5w4OHDigW3b37l2Eh4ejX79+AIDDhw9j4MCBGDVqFM6dO4dvv/0WYWFhmD17donfY3x8PPbs2QOlUqlbptVqUbVqVWzZsgXnzp3DtGnTMGnSJPz4448AgHHjxuHtt99G165dkZiYiMTERLRt2xb5+fkIDAyEo6MjDh8+jKNHj8LBwQFdu3ZFXl5eiWsislgmn3eciMpcUFCQUCgUwt7eXvd48803i9x2y5YtonLlyrrna9asEc7Ozrrnjo6OIiwsrMh93333XfHee+/pLTt8+LCQy+UiOzu7yH0ef/2LFy+KunXrCn9/fyGEEN7e3mL27Nl6+7Rs2VIMGzZMCCHEiBEjRKdOnYRWqy3y9QGI7du3CyGEiIuLEwDEyZMn9bYJCgoSPXv21D3v2bOnGDx4sO75t99+K7y9vYVGoxFCCPHSSy+JOXPm6L3G+vXrhZeXV5E1CCFEaGiokMvlwt7eXtjY2AgAAoBYsGBBsfsIIcTw4cPFG2+8UWytD49dr149vXOQm5srbG1txZ49e574+kQVAcfcEFmojh074ptvvtE9t7e3B/CgFWPu3Lm4cOEC0tLSUFBQgJycHGRlZcHOzq7Q64wdOxZDhgzB+vXrdV0rtWrVAvCgy+rff//Fhg0bdNsLIaDVahEXF4cGDRoUWVtqaiocHByg1WqRk5OD559/Ht999x3S0tJw8+ZNtGvXTm/7du3a4dSpUwAedCl17twZ9erVQ9euXfHKK6+gS5cuz3Su+vXrh6FDh+Lrr7+GSqXChg0b0KdPH8jlct37PHr0qF5LjUajeeJ5A4B69eph586dyMnJwffff4/o6GiMGDFCb5tly5Zh9erVSEhIQHZ2NvLy8uDn5/fEek+dOoXLly/D0dFRb3lOTg5iY2NLcQaILAvDDZGFsre3R+3atfWWxcfH45VXXsGHH36I2bNnw9XVFUeOHMG7776LvLy8In+kp0+fjr59+2LXrl347bffEBoaik2bNuG1115DRkYG3n//fYwcObLQftWqVSu2NkdHR0RFRUEul8PLywu2trYAgLS0tKe+r+bNmyMuLg6//fYb9u3bh7fffhsBAQHYunXrU/ctTo8ePSCEwK5du9CyZUscPnwYCxcu1K3PyMjAjBkz8Prrrxfa18bGptjXVSqVur+DefPmoXv37pgxYwZmzZoFANi0aRPGjRuH+fPno02bNnB0dMQXX3yBf/7554n1ZmRkoEWLFnqh8qHyMmicSEoMN0QVSGRkJLRaLebPn69rlXg4vuNJ6tati7p162LMmDF45513sGbNGrz22mto3rw5zp07VyhEPY1cLi9yHycnJ3h7e+Po0aPo0KGDbvnRo0fRqlUrve169+6N3r17480330TXrl1x9+5duLq66r3ew/EtGo3mifXY2Njg9ddfx4YNG3D58mXUq1cPzZs3161v3rw5YmJiDH6fj5syZQo6deqEDz/8UPc+27Zti2HDhum2ebzlRalUFqq/efPm2Lx5M9zd3eHk5PRMNRFZIg4oJqpAateujfz8fCxZsgRXrlzB+vXrsXz58mK3z87ORkhICA4ePIirV6/i6NGjiIiI0HU3ffLJJ/jrr78QEhKC6OhoXLp0CT///LPBA4ofNX78eHz22WfYvHkzYmJiMGHCBERHR2PUqFEAgAULFuCHH37AhQsXcPHiRWzZsgWenp5F3njQ3d0dtra2CA8PR3JyMlJTU4s9br9+/bBr1y6sXr1aN5D4oWnTpmHdunWYMWMGzp49i/Pnz2PTpk2YMmWKQe+tTZs2aNKkCebMmQMAqFOnDk6cOIE9e/bg4sWLmDp1KiIiIvT28fX1xb///ouYmBikpKQgPz8f/fr1g5ubG3r27InDhw8jLi4OBw8exMiRI3H9+nWDaiKySFIP+iEi4ytqEOpDCxYsEF5eXsLW1lYEBgaKdevWCQDi3r17Qgj9Ab+5ubmiT58+Qq1WC6VSKby9vUVISIjeYOHjx4+Lzp07CwcHB2Fvby+aNGlSaEDwox4fUPw4jUYjpk+fLnx8fIS1tbVo2rSp+O2333TrV6xYIfz8/IS9vb1wcnISL730koiKitKtxyMDioUQYuXKlUKtVgu5XC46dOhQ7PnRaDTCy8tLABCxsbGF6goPDxdt27YVtra2wsnJSbRq1UqsWLGi2PcRGhoqmjZtWmj5Dz/8IFQqlUhISBA5OTli0KBBwtnZWbi4uIgPP/xQTJgwQW+/W7du6c4vAHHgwAEhhBCJiYli4MCBws3NTahUKlGzZk0xdOhQkZqaWmxNRBWFTAghpI1XRERERMbDbikiIiKyKAw3REREZFEYboiIiMiiMNwQERGRRWG4ISIiIovCcENEREQWheGGiIiILArDDREREVkUhhsiIiKyKAw3REREZFEYboiIiMii/D+/WXx7vO+DGwAAAABJRU5ErkJggg=="},"metadata":{}},{"name":"stdout","text":"ROC AUC Score: 0.7708978328173375\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### 5. Hyperparameters Tuning \nTo tune the hyperparameters, we'll set specific values for each parameter and specify an evaluation metric to find the best parameter set for fitting the final model. We can use techniques like grid search along with cross-validation to achieve this. Once we find the best parameter set, we'll evaluate the final model using various metrics.","metadata":{}},{"cell_type":"code","source":"from sklearn.model_selection import GridSearchCV\nfrom sklearn.metrics import f1_score, make_scorer\n\n# Define a custom scorer for class '1' F1-score\ncustom_scorer = make_scorer(f1_score, labels=['1'], average=None)\n\n# Define the parameter grid\nparam_grid = {\n    'criterion':['gini', 'entropy'],\n    'max_depth': [2, 3, 4],\n    'min_samples_split': [10, 15, 25],\n    'min_samples_leaf': [5, 8, 12],\n    'class_weight': [None, 'balanced']\n}\n\n\n# Instantiate the DecisionTreeClassifier\nclf = DecisionTreeClassifier(random_state=111)\n\n# Instantiate GridSearchCV\ngrid_search = GridSearchCV(clf, param_grid, cv=5, scoring=custom_scorer)\n\n# Perform GridSearchCV\ngrid_search.fit(X_train, y_train)\n\n# Get the best parameters\nbest_params = grid_search.best_params_\nprint(\"Best Parameters:\", best_params)\n\n# Get the best estimator\nbest_clf = grid_search.best_estimator_\n\n# Evaluate the best estimator on the test set\naccuracy = best_clf.score(X_test, y_test)\nprint(\"Test Accuracy:\", accuracy)","metadata":{"execution":{"iopub.status.busy":"2024-05-30T15:49:21.293517Z","iopub.execute_input":"2024-05-30T15:49:21.294013Z","iopub.status.idle":"2024-05-30T15:49:28.135007Z","shell.execute_reply.started":"2024-05-30T15:49:21.293977Z","shell.execute_reply":"2024-05-30T15:49:28.133659Z"},"trusted":true},"execution_count":90,"outputs":[{"name":"stdout","text":"Best Parameters: {'class_weight': 'balanced', 'criterion': 'gini', 'max_depth': 2, 'min_samples_leaf': 5, 'min_samples_split': 10}\nTest Accuracy: 0.6285714285714286\n","output_type":"stream"}]},{"cell_type":"markdown","source":"**Evalute the Best Model on Training and Test sets**","metadata":{}},{"cell_type":"code","source":"# Train the classifier with the full training data\nbest_clf.fit(X_train, y_train)\n\n# Predict the labels for the train set\ny_pred_train = best_clf.predict(X_train)\ny_prob_train = best_clf.predict_proba(X_train)\n\n# Confusion Matrix\nconf_matrix_train = confusion_matrix(y_train, y_pred_train)\nprint(\"Confusion Matrix for Training Set:\")\nprint(conf_matrix_train)\n\nreport_train = classification_report(y_train, y_pred_train, labels=None, zero_division='warn')\nprint(\"Classification Report fot Training Set:\")\nprint(report_train)\n\nroc_auc_train = roc_auc_score(y_train, y_prob_train[:,1])\nprint(\"ROC AUC Score for Training Set:\", roc_auc_train)\nprint(\"#\"*60)\n#______________________________________________________________________________________________________\n\n# Predict the labels for the test set\ny_pred = best_clf.predict(X_test)\ny_prob = best_clf.predict_proba(X_test)\n\n# Confusion Matrix\nconf_matrix = confusion_matrix(y_test, y_pred)\nprint(\"Confusion Matrix for Test Set:\")\nprint(conf_matrix)\n\nreport = classification_report(y_test, y_pred, labels=None, zero_division='warn')\nprint(\"Classification Report for Test Set:\")\nprint(report)\n\nroc_auc = roc_auc_score(y_test, y_prob[:,1])\nprint(\"ROC AUC Score for Test Set:\", roc_auc)","metadata":{"execution":{"iopub.status.busy":"2024-05-30T15:51:35.992960Z","iopub.execute_input":"2024-05-30T15:51:35.993448Z","iopub.status.idle":"2024-05-30T15:51:36.077083Z","shell.execute_reply.started":"2024-05-30T15:51:35.993413Z","shell.execute_reply":"2024-05-30T15:51:36.075764Z"},"trusted":true},"execution_count":91,"outputs":[{"name":"stdout","text":"Confusion Matrix for Training Set:\n[[194 170]\n [ 17 104]]\nClassification Report fot Training Set:\n              precision    recall  f1-score   support\n\n           0       0.92      0.53      0.67       364\n           1       0.38      0.86      0.53       121\n\n    accuracy                           0.61       485\n   macro avg       0.65      0.70      0.60       485\nweighted avg       0.78      0.61      0.64       485\n\nROC AUC Score for Training Set: 0.7595472709109072\n############################################################\nConfusion Matrix for Test Set:\n[[80 73]\n [ 5 52]]\nClassification Report for Test Set:\n              precision    recall  f1-score   support\n\n           0       0.94      0.52      0.67       153\n           1       0.42      0.91      0.57        57\n\n    accuracy                           0.63       210\n   macro avg       0.68      0.72      0.62       210\nweighted avg       0.80      0.63      0.64       210\n\nROC AUC Score for Test Set: 0.7269235179451897\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### 5.Ckeck Feature Importances  \nTo identify features that have a noteworthy effect on the target field in the model, we can assess the feature importances.","metadata":{}},{"cell_type":"code","source":"best_clf.feature_importances_\nbest_clf.feature_names_in_","metadata":{"trusted":true},"execution_count":93,"outputs":[{"execution_count":93,"output_type":"execute_result","data":{"text/plain":"array(['ed', 'age_cat_cm', 'employ_cat_cm', 'address_cat_cm',\n       'income_ratio_cat_cm', 'debtinc_cat_cm', 'creddebt_ratio_cat_cm',\n       'othdebt_ratio_cat_cm'], dtype=object)"},"metadata":{}}]},{"cell_type":"code","source":"import numpy as np\nimport matplotlib.pyplot as plt\n\n# Extract feature importances and feature names\nimportances = best_clf.feature_importances_\nfeature_names = best_clf.feature_names_in_\n\n# Sort feature importances in descending order\nindices = np.argsort(importances)\n\n# Plot\nplt.figure(figsize=(8, 4))\nplt.title(\"Feature Importances\")\nplt.barh(range(len(feature_names)), importances[indices], color=\"skyblue\", align=\"center\")\nplt.yticks(range(len(feature_names)), feature_names[indices])\nplt.xlabel(\"Importance\")\nplt.ylabel(\"Features\")\nplt.tight_layout()\nplt.show()\n","metadata":{"execution":{"iopub.status.busy":"2024-05-30T15:58:34.128682Z","iopub.execute_input":"2024-05-30T15:58:34.129291Z","iopub.status.idle":"2024-05-30T15:58:34.438652Z","shell.execute_reply.started":"2024-05-30T15:58:34.129248Z","shell.execute_reply":"2024-05-30T15:58:34.437355Z"},"trusted":true},"execution_count":94,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 800x400 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"markdown","source":"### 6.Visualize the Decision Tree and If-Then Rules\nTo comprehend the structure of the model, we can utilize the decision tree plot or represent its logic through if-then rules in textual format for **concept description** purposes.","metadata":{}},{"cell_type":"code","source":"from sklearn.tree import plot_tree\nimport matplotlib.pyplot as plt\n\n\n# Assuming clf is your decision tree classifier\nplt.figure(figsize=(20,10))  # Adjust the figure size as needed\nplot_tree(best_clf, feature_names=best_clf.feature_names_in_,  \n                           class_names=best_clf.classes_,\n                           proportion=True,\n                           filled=True, rounded=True)\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-05-30T16:00:35.118358Z","iopub.execute_input":"2024-05-30T16:00:35.118840Z","iopub.status.idle":"2024-05-30T16:00:35.652001Z","shell.execute_reply.started":"2024-05-30T16:00:35.118807Z","shell.execute_reply":"2024-05-30T16:00:35.650737Z"},"trusted":true},"execution_count":95,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 2000x1000 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"from sklearn.tree import export_text\n\ntree_rules = export_text(best_clf,feature_names=list(best_clf.feature_names_in_),\n                        spacing=3,\n                        decimals=2,\n                        show_weights=True)\n\nprint(tree_rules)","metadata":{"execution":{"iopub.status.busy":"2024-05-30T16:05:43.302857Z","iopub.execute_input":"2024-05-30T16:05:43.303982Z","iopub.status.idle":"2024-05-30T16:05:43.311437Z","shell.execute_reply.started":"2024-05-30T16:05:43.303940Z","shell.execute_reply":"2024-05-30T16:05:43.310117Z"},"trusted":true},"execution_count":96,"outputs":[{"name":"stdout","text":"|--- debtinc_cat_cm <= 1.50\n|   |--- employ_cat_cm <= 1.50\n|   |   |--- weights: [80.61, 92.19] class: 1\n|   |--- employ_cat_cm >  1.50\n|   |   |--- weights: [129.24, 34.07] class: 0\n|--- debtinc_cat_cm >  1.50\n|   |--- income_ratio_cat_cm <= 2.50\n|   |   |--- weights: [19.99, 102.21] class: 1\n|   |--- income_ratio_cat_cm >  2.50\n|   |   |--- weights: [12.66, 14.03] class: 1\n\n","output_type":"stream"}]},{"cell_type":"markdown","source":"#### Function: Extract Rules from Decision Tree Model","metadata":{}},{"cell_type":"code","source":"from sklearn.tree import _tree\n\ndef get_rules(tree, feature_names, task):  # task='reg' or 'class'\n    tree_ = tree.tree_\n    if task == 'class':\n        class_name = tree.classes_\n    feature_name = [\n        feature_names[i] if i != _tree.TREE_UNDEFINED else \"undefined!\"\n        for i in tree_.feature\n    ]\n\n    paths = []\n    path = []\n    \n    def recurse(node, path, paths):\n        \n        if tree_.feature[node] != _tree.TREE_UNDEFINED:\n            name = feature_name[node]\n            threshold = tree_.threshold[node]\n            p1, p2 = list(path), list(path)\n            p1 += [f\"({name} <= {np.round(threshold, 3)})\"]\n            recurse(tree_.children_left[node], p1, paths)\n            p2 += [f\"({name} > {np.round(threshold, 3)})\"]\n            recurse(tree_.children_right[node], p2, paths)\n        else:\n            path += [(tree_.value[node], tree_.n_node_samples[node])]\n            paths += [path]\n            \n    recurse(0, path, paths)\n\n    # sort by samples count\n    samples_count = [p[-1][1] for p in paths]\n    ii = list(np.argsort(samples_count))\n    paths = [paths[i] for i in reversed(ii)]\n    \n    antecedents = []\n    consequents = []\n    \n    for path in paths:\n        antecedent = \"\"\n        consequent = \"\"\n        \n        for p in path[:-1]:\n            if antecedent != \"\":\n                antecedent += \" and \"\n            antecedent += str(p)\n\n        if task == 'reg':\n            consequent += str(np.round(path[-1][0][0][0], 3))\n        elif task == 'class':\n            classes = path[-1][0][0]\n            l = np.argmax(classes)\n            \n            # Get class name for antecedent\n            consequent += class_name[l]\n\n        antecedents.append(antecedent)\n        consequents.append(consequent)\n    \n    rules = pd.DataFrame({'Antecedent':antecedents, 'Consequent':consequents})\n        \n    return rules","metadata":{"execution":{"iopub.status.busy":"2024-05-30T16:07:35.154544Z","iopub.execute_input":"2024-05-30T16:07:35.155825Z","iopub.status.idle":"2024-05-30T16:07:35.169100Z","shell.execute_reply.started":"2024-05-30T16:07:35.155778Z","shell.execute_reply":"2024-05-30T16:07:35.167828Z"},"trusted":true},"execution_count":97,"outputs":[]},{"cell_type":"markdown","source":"#### Function: Create Evaluation Report for Rules Extracted","metadata":{}},{"cell_type":"code","source":"def calculate_rule_metrics(rules_df, input_data, target_variable, task):\n\n    import re\n\n    # Add columns for metrics\n    rules_df['Sample'] = None\n    rules_df['Support'] = None\n    rules_df['Confidence'] = None\n    rules_df['Lift'] = None\n    \n    # Calculate metrics for each rule\n    for i, row in rules_df.iterrows():\n        antecedent = row['Antecedent']\n        consequent = row['Consequent']\n        \n        def filter_by_antecedent(data, antecedent):\n\n            conditions = antecedent.split(\" and \")\n            filtered_data = data.copy()  # Avoid modifying original data\n            pattern = r\"(?P<col_name>\\w+)\\s*(?P<operator>[<>=]+)\\s*(?P<threshold>[-.\\d]+)\"\n\n            for condition in conditions:\n                condition = condition.strip('()')\n                match = re.match(pattern, condition)\n                if not match:\n                    raise ValueError(f\"Invalid condition format: {condition}\")\n\n                col_name, operator, threshold = match.groups()\n                # Access the appropriate comparison function directly using operator\n                comparison_func = {\n                    \"<\": pd.Series.__lt__,\n                    \">\": pd.Series.__gt__,\n                    \"<=\": pd.Series.__le__,\n                    \">=\": pd.Series.__ge__,\n                    \"==\": pd.Series.__eq__,\n                }[operator]\n\n                filtered_data = filtered_data[comparison_func(filtered_data[col_name], float(threshold))]\n\n            return filtered_data\n\n\n\n        # Filter train_data based on antecedent conditions (boolean mask)\n        filtered_data = filter_by_antecedent(input_data, antecedent)\n        index = filtered_data.index.tolist()\n\n        # Calculate support (proportion of samples satisfying the rule)\n        rule_support = len(filtered_data) / len(input_data)\n        rules_df.at[i, 'Sample'] = len(filtered_data)\n        rules_df.at[i, 'Support'] = rule_support\n\n        # Calculate confidence (conditional probability of consequent given antecedent) and lift index\n        if task == 'class':\n            \n            if len(filtered_data) == 0:\n                confidence = None\n                lift = None\n            else:\n                class_probabilities = target_variable.value_counts(normalize=True)\n                if target_variable.index.name:\n                    class_counts = target_variable.loc[index].value_counts()\n                else:\n                    class_counts = target_variable[index].value_counts()\n                if consequent in class_counts:\n                    correct_predictions = class_counts[consequent]\n                else:\n                    correct_predictions = 0\n                confidence = (correct_predictions / len(filtered_data)) \n                lift = (confidence) / class_probabilities[consequent]       \n\n        elif task == 'reg':\n\n            # Confidence and Lift calculation for regression is not typically used\n            confidence = None\n            lift = None\n\n        rules_df.at[i, 'Confidence'] = confidence\n        rules_df.at[i, 'Lift'] = lift\n        \n    return rules_df\n","metadata":{"execution":{"iopub.status.busy":"2024-05-30T16:08:13.754022Z","iopub.execute_input":"2024-05-30T16:08:13.754425Z","iopub.status.idle":"2024-05-30T16:08:13.769185Z","shell.execute_reply.started":"2024-05-30T16:08:13.754397Z","shell.execute_reply":"2024-05-30T16:08:13.767719Z"},"trusted":true},"execution_count":98,"outputs":[]},{"cell_type":"markdown","source":"#### Get Rules from Bankloan DT Classifier","metadata":{}},{"cell_type":"code","source":"feature_names = X_train.columns.tolist()\n\n# Set the maximum column width\npd.set_option('display.max_colwidth', None)\n\nrules = get_rules(best_clf, feature_names, task='class')\nrules","metadata":{"execution":{"iopub.status.busy":"2024-05-30T16:09:40.548561Z","iopub.execute_input":"2024-05-30T16:09:40.549679Z","iopub.status.idle":"2024-05-30T16:09:40.561695Z","shell.execute_reply.started":"2024-05-30T16:09:40.549636Z","shell.execute_reply":"2024-05-30T16:09:40.560322Z"},"trusted":true},"execution_count":101,"outputs":[{"execution_count":101,"output_type":"execute_result","data":{"text/plain":"                                                Antecedent Consequent\n0        (debtinc_cat_cm <= 1.5) and (employ_cat_cm > 1.5)          0\n1       (debtinc_cat_cm <= 1.5) and (employ_cat_cm <= 1.5)          1\n2  (debtinc_cat_cm > 1.5) and (income_ratio_cat_cm <= 2.5)          1\n3   (debtinc_cat_cm > 1.5) and (income_ratio_cat_cm > 2.5)          1","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Antecedent</th>\n      <th>Consequent</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>(debtinc_cat_cm &lt;= 1.5) and (employ_cat_cm &gt; 1.5)</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>(debtinc_cat_cm &lt;= 1.5) and (employ_cat_cm &lt;= 1.5)</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>(debtinc_cat_cm &gt; 1.5) and (income_ratio_cat_cm &lt;= 2.5)</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>(debtinc_cat_cm &gt; 1.5) and (income_ratio_cat_cm &gt; 2.5)</td>\n      <td>1</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"markdown","source":"#### Report Table for Rules Extracted on Train Dataset","metadata":{}},{"cell_type":"code","source":"import pandas as pd\n\ncalculate_rule_metrics(rules, X_train, y_train, task='class')","metadata":{"execution":{"iopub.status.busy":"2024-05-30T16:10:55.618511Z","iopub.execute_input":"2024-05-30T16:10:55.618957Z","iopub.status.idle":"2024-05-30T16:10:55.658099Z","shell.execute_reply.started":"2024-05-30T16:10:55.618927Z","shell.execute_reply":"2024-05-30T16:10:55.656962Z"},"trusted":true},"execution_count":102,"outputs":[{"execution_count":102,"output_type":"execute_result","data":{"text/plain":"                                                Antecedent Consequent Sample  \\\n0        (debtinc_cat_cm <= 1.5) and (employ_cat_cm > 1.5)          0    211   \n1       (debtinc_cat_cm <= 1.5) and (employ_cat_cm <= 1.5)          1    167   \n2  (debtinc_cat_cm > 1.5) and (income_ratio_cat_cm <= 2.5)          1     81   \n3   (debtinc_cat_cm > 1.5) and (income_ratio_cat_cm > 2.5)          1     26   \n\n    Support Confidence      Lift  \n0  0.435052   0.919431  1.225066  \n1   0.34433   0.275449  1.104073  \n2   0.16701    0.62963  2.523722  \n3  0.053608   0.269231  1.079148  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Antecedent</th>\n      <th>Consequent</th>\n      <th>Sample</th>\n      <th>Support</th>\n      <th>Confidence</th>\n      <th>Lift</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>(debtinc_cat_cm &lt;= 1.5) and (employ_cat_cm &gt; 1.5)</td>\n      <td>0</td>\n      <td>211</td>\n      <td>0.435052</td>\n      <td>0.919431</td>\n      <td>1.225066</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>(debtinc_cat_cm &lt;= 1.5) and (employ_cat_cm &lt;= 1.5)</td>\n      <td>1</td>\n      <td>167</td>\n      <td>0.34433</td>\n      <td>0.275449</td>\n      <td>1.104073</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>(debtinc_cat_cm &gt; 1.5) and (income_ratio_cat_cm &lt;= 2.5)</td>\n      <td>1</td>\n      <td>81</td>\n      <td>0.16701</td>\n      <td>0.62963</td>\n      <td>2.523722</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>(debtinc_cat_cm &gt; 1.5) and (income_ratio_cat_cm &gt; 2.5)</td>\n      <td>1</td>\n      <td>26</td>\n      <td>0.053608</td>\n      <td>0.269231</td>\n      <td>1.079148</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"markdown","source":"#### Report Table for Rules Extracted on Test Dataset","metadata":{}},{"cell_type":"code","source":"calculate_rule_metrics(rules, X_test, y_test, task='class')","metadata":{"execution":{"iopub.status.busy":"2024-05-30T16:13:19.283832Z","iopub.execute_input":"2024-05-30T16:13:19.284309Z","iopub.status.idle":"2024-05-30T16:13:19.320858Z","shell.execute_reply.started":"2024-05-30T16:13:19.284276Z","shell.execute_reply":"2024-05-30T16:13:19.319627Z"},"trusted":true},"execution_count":103,"outputs":[{"execution_count":103,"output_type":"execute_result","data":{"text/plain":"                                                Antecedent Consequent Sample  \\\n0        (debtinc_cat_cm <= 1.5) and (employ_cat_cm > 1.5)          0     85   \n1       (debtinc_cat_cm <= 1.5) and (employ_cat_cm <= 1.5)          1     78   \n2  (debtinc_cat_cm > 1.5) and (income_ratio_cat_cm <= 2.5)          1     33   \n3   (debtinc_cat_cm > 1.5) and (income_ratio_cat_cm > 2.5)          1     14   \n\n    Support Confidence      Lift  \n0  0.404762   0.941176  1.291811  \n1  0.371429   0.346154  1.275304  \n2  0.157143   0.515152  1.897927  \n3  0.066667   0.571429  2.105263  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Antecedent</th>\n      <th>Consequent</th>\n      <th>Sample</th>\n      <th>Support</th>\n      <th>Confidence</th>\n      <th>Lift</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>(debtinc_cat_cm &lt;= 1.5) and (employ_cat_cm &gt; 1.5)</td>\n      <td>0</td>\n      <td>85</td>\n      <td>0.404762</td>\n      <td>0.941176</td>\n      <td>1.291811</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>(debtinc_cat_cm &lt;= 1.5) and (employ_cat_cm &lt;= 1.5)</td>\n      <td>1</td>\n      <td>78</td>\n      <td>0.371429</td>\n      <td>0.346154</td>\n      <td>1.275304</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>(debtinc_cat_cm &gt; 1.5) and (income_ratio_cat_cm &lt;= 2.5)</td>\n      <td>1</td>\n      <td>33</td>\n      <td>0.157143</td>\n      <td>0.515152</td>\n      <td>1.897927</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>(debtinc_cat_cm &gt; 1.5) and (income_ratio_cat_cm &gt; 2.5)</td>\n      <td>1</td>\n      <td>14</td>\n      <td>0.066667</td>\n      <td>0.571429</td>\n      <td>2.105263</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]}]}