diff --git a/.github/workflows/autoupdate.yml b/.github/workflows/autoupdate.yml deleted file mode 100644 index 6393c396..00000000 --- a/.github/workflows/autoupdate.yml +++ /dev/null @@ -1,33 +0,0 @@ -# more information to be found here: https://github.com/marketplace/actions/auto-update -name: autoupdate -on: - push: {} -jobs: - autoupdate: - name: autoupdate - runs-on: ubuntu-latest - steps: - - uses: docker://chinthakagodawita/autoupdate-action:v1 - env: - GITHUB_TOKEN: "${{ secrets.GITHUB_TOKEN }}" - - # if true potential actions are only logged. Good for testing purpose - DRY_RUN: "false" - - # if only specific pull request should be updated. Values are: ["all", "labelled", "protected", "auto_merge"]. - # "labelled" requires the PR_LABELS option, in which the labels are defined which should trigger autoupdates - PR_FILTER: "all" - - # PR with thoose labels are not getting updated - EXCLUDED_LABELS: "dependencies,wontfix" - - # which commit message should be displayed - MERGE_MSG: "Branch was auto-updated." - - RETRY_COUNT: "1" - - # time in ms - RETRY_SLEEP: "5000" - - # what to do on merge conflict. if fail it is also reported. options: ["fail", "ignore"] - MERGE_CONFLICT_ACTION: "fail" diff --git a/docs/source/index.rst b/docs/source/index.rst index 68ade3ad..f437ef7f 100644 --- a/docs/source/index.rst +++ b/docs/source/index.rst @@ -30,13 +30,16 @@ Contents :maxdepth: 1 :caption: TUTORIALS + notebooks/sv_calculation notebooks/shapiq_scikit_learn notebooks/treeshapiq_lightgbm + notebooks/visualizing_shapley_interactions notebooks/language_model_game notebooks/vision_transformer notebooks/conditional_imputer notebooks/parallel_computation notebooks/benchmark_approximators + notebooks/data_valuation notebooks/core .. toctree:: diff --git a/docs/source/notebooks/2-SII_network.pdf b/docs/source/notebooks/2-SII_network.pdf new file mode 100644 index 00000000..c87dc9ca Binary files /dev/null and b/docs/source/notebooks/2-SII_network.pdf differ diff --git a/docs/source/notebooks/2-SII_si_graph.pdf b/docs/source/notebooks/2-SII_si_graph.pdf new file mode 100644 index 00000000..04afab8b Binary files /dev/null and b/docs/source/notebooks/2-SII_si_graph.pdf differ diff --git a/docs/source/notebooks/Moebius_network.pdf b/docs/source/notebooks/Moebius_network.pdf new file mode 100644 index 00000000..f3d0364e Binary files /dev/null and b/docs/source/notebooks/Moebius_network.pdf differ diff --git a/docs/source/notebooks/Moebius_si_graph.pdf b/docs/source/notebooks/Moebius_si_graph.pdf new file mode 100644 index 00000000..68c9cdc6 Binary files /dev/null and b/docs/source/notebooks/Moebius_si_graph.pdf differ diff --git a/docs/source/notebooks/SV_si_graph.pdf b/docs/source/notebooks/SV_si_graph.pdf new file mode 100644 index 00000000..a585aa47 Binary files /dev/null and b/docs/source/notebooks/SV_si_graph.pdf differ diff --git a/docs/source/notebooks/data_valuation.ipynb b/docs/source/notebooks/data_valuation.ipynb new file mode 100644 index 00000000..e7b41a59 --- /dev/null +++ b/docs/source/notebooks/data_valuation.ipynb @@ -0,0 +1,792 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "source": [ + "# Shapiq for Data Valuation\n", + "On this page we demonstrate two examples for using Shapiq for Data valuation.\n", + "The first example demonstrates this for a synthetic dataset, and the second for a real dataset.\n", + "In data valuation we are interested given a training and testing dataset to evaluate the contribution of each training point to the model's performance on the test data." + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "code", + "execution_count": 1, + "outputs": [], + "source": [ + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "import shapiq\n", + "from sklearn.inspection import DecisionBoundaryDisplay\n", + "\n", + "# Vector Graphics\n", + "%matplotlib inline\n", + "import matplotlib_inline\n", + "from shapiq.plot._config import COLORS_K_SII, RED\n", + "\n", + "matplotlib_inline.backend_inline.set_matplotlib_formats(\"svg\")" + ], + "metadata": { + "collapsed": false, + "ExecuteTime": { + "end_time": "2024-10-22T16:04:37.434604Z", + "start_time": "2024-10-22T16:04:35.731310Z" + } + } + }, + { + "cell_type": "markdown", + "source": [ + "## Synthetic Data\n", + "In this example we generate a synthetic classification dataset with 2 features, and 22 samples.\n", + "The dataset consists of two classes, each with 11 samples.\n", + "The data is generated from two multivariate normal distributions with different means and covariances.\n", + "This is done in such a way that the two classes are linearly separable.\n" + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "collapsed": true, + "ExecuteTime": { + "end_time": "2024-10-22T16:04:37.540707Z", + "start_time": "2024-10-22T16:04:37.438398Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": "
", + "image/svg+xml": "\n\n\n \n \n \n \n 2024-10-22T18:04:37.516569\n image/svg+xml\n \n \n Matplotlib v3.8.0, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n" + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "def plot_synthetic_data(ax, X_train, y_train, X_test, y_test, title):\n", + " ax.set_title(title)\n", + " ax.scatter(\n", + " X_train[:, 0],\n", + " X_train[:, 1],\n", + " c=[COLORS_K_SII[i] for i in y_train],\n", + " label=\"Training Points\",\n", + " marker=\"o\",\n", + " )\n", + " ax.scatter(\n", + " X_test[:, 0],\n", + " X_test[:, 1],\n", + " c=[COLORS_K_SII[i] for i in y_test],\n", + " label=\"Test Points\",\n", + " marker=\"x\",\n", + " )\n", + " # Manually create legend entries\n", + " handles = [\n", + " plt.Line2D(\n", + " [0],\n", + " [0],\n", + " marker=\"o\",\n", + " color=\"w\",\n", + " markerfacecolor=COLORS_K_SII[i],\n", + " markersize=10,\n", + " label=f\"Class {i} (Train)\",\n", + " )\n", + " for i in [1, 2]\n", + " ]\n", + " handles += [\n", + " plt.Line2D(\n", + " [0],\n", + " [0],\n", + " marker=\"x\",\n", + " linewidth=0,\n", + " color=COLORS_K_SII[i],\n", + " markerfacecolor=COLORS_K_SII[i],\n", + " markersize=10,\n", + " label=f\"Class {i} (Test)\",\n", + " )\n", + " for i in [1, 2]\n", + " ]\n", + "\n", + " ax.legend(handles=handles, loc=\"upper right\", title=\"Data Points\")\n", + "\n", + " ax.set_xlabel(\"Feature 1\")\n", + " ax.set_ylabel(\"Feature 2\")\n", + "\n", + "\n", + "# Meta information\n", + "n_samples = 11\n", + "n_classes = 2\n", + "classes = list(range(1, n_classes + 1))\n", + "random_state = 1337\n", + "np.random.seed(random_state)\n", + "\n", + "# parameters for toy data\n", + "means = [(3, 0), (-3, 0)]\n", + "covs = [np.diag([3, 2]), np.diag([3, 3.5])]\n", + "\n", + "# Construct the dataset\n", + "X = np.vstack(\n", + " [np.random.multivariate_normal(mean, cov, n_samples) for mean, cov in zip(means, covs)]\n", + ")\n", + "y = np.hstack([np.full(n_samples, i) for i in classes])\n", + "\n", + "# Build training and test set\n", + "n_samples_to_select = 10\n", + "random_indices = np.random.choice(X.shape[0], n_samples_to_select, replace=False)\n", + "X_test, y_test = X[random_indices], y[random_indices]\n", + "X_train, y_train = np.delete(X, random_indices, axis=0), np.delete(y, random_indices, axis=0)\n", + "fig, ax = plt.subplots()\n", + "\n", + "plot_synthetic_data(ax, X_train, y_train, X_test, y_test, \"Synthetic Classification Data\")" + ] + }, + { + "cell_type": "markdown", + "source": [ + "To apply `shapiq` approximators we need to reformulate the task of data valuation into a cooperative game $(N,\\nu)$.\n", + "We define $N$ as the set of training points $N = \\{1, \\ldots, n\\}$ and the characteristic function $$\\nu: 2^N \\rightarrow \\mathbb{R}$$ is then the accuracy the model achieves on the test points (cross) given the training points in $S$.\n" + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "code", + "execution_count": 3, + "outputs": [], + "source": [ + "class SyntheticDataValuation(shapiq.Game):\n", + " \"\"\"The synthetic data valuation tasked modeled as a cooperative game.\n", + " Args:\n", + " classifier: A classifier object that has the methods fit and score.\n", + " n_players: The number of players in the game.\n", + " X_test: The test data.\n", + " y_test: The test labels.\n", + " \"\"\"\n", + "\n", + " def __init__(self, classifier, n_players, X_train, y_train, X_test, y_test):\n", + " self.classifier = classifier\n", + " self.X_train = X_train\n", + " self.y_train = y_train\n", + " self.X_test = X_test\n", + " self.y_test = y_test\n", + "\n", + " empty_coalition_value = np.zeros((1, n_players), dtype=bool)\n", + " self.normalization_value = float(self.value_function(empty_coalition_value)[0])\n", + " super().__init__(n_players, normalization_value=self.normalization_value)\n", + "\n", + " def value_function(self, coalitions: np.ndarray) -> np.ndarray:\n", + " \"\"\"Compute the value of the coalitions.\n", + " Args:\n", + " coalitions: A numpy matrix of shape (n_coalitions, n_players)\n", + "\n", + " Returns:\n", + " A vector of the value of the coalition\n", + " \"\"\"\n", + " values = []\n", + " for coalition in coalitions:\n", + " tmp_X_train = self.X_train[coalition]\n", + " tmp_y_train = self.y_train[coalition]\n", + " if len(tmp_X_train) == 0:\n", + " # If the coalition is empty, the value is zero\n", + " value = 0\n", + " else:\n", + " unique_targets = np.unique(tmp_y_train)\n", + " if len(unique_targets) == 1:\n", + " # If we only have one class present in training data, we predict this class\n", + " value = np.mean((self.y_test == unique_targets[0]))\n", + " else:\n", + " # We have at least two classes, we fit the classifier\n", + " self.classifier.fit(tmp_X_train, tmp_y_train)\n", + " value = self.classifier.score(self.X_test, self.y_test)\n", + "\n", + " values.append(value)\n", + "\n", + " return np.array(values, dtype=float)" + ], + "metadata": { + "collapsed": false, + "ExecuteTime": { + "end_time": "2024-10-22T16:04:37.551682Z", + "start_time": "2024-10-22T16:04:37.549258Z" + } + } + }, + { + "cell_type": "markdown", + "source": [ + "As our model we choose the `LinearSVC()`from `sklearn`.\n", + "To get first insights into the data valuation we can compute the value of the full and empty coalition." + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "code", + "execution_count": 4, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Full coalition value: 1.0\n", + "Empty coalition value: 0.0\n" + ] + } + ], + "source": [ + "from sklearn.svm import LinearSVC\n", + "\n", + "classifier = LinearSVC()\n", + "n_players = X_train.shape[0]\n", + "data_valuation_game = SyntheticDataValuation(\n", + " classifier=classifier,\n", + " n_players=n_players,\n", + " X_train=X_train,\n", + " y_train=y_train,\n", + " X_test=X_test,\n", + " y_test=y_test,\n", + ")\n", + "\n", + "full_coalition = np.ones((1, n_players), dtype=bool)\n", + "empty_coalition = np.zeros((1, n_players), dtype=bool)\n", + "print(\"Full coalition value: \", data_valuation_game(full_coalition)[0])\n", + "print(\"Empty coalition value: \", data_valuation_game(empty_coalition)[0])" + ], + "metadata": { + "collapsed": false, + "ExecuteTime": { + "end_time": "2024-10-22T16:04:37.577065Z", + "start_time": "2024-10-22T16:04:37.554955Z" + } + } + }, + { + "cell_type": "markdown", + "source": [ + "The empty coalition value is $0.0$ as the model has no information about the data.\n", + "The full coalition value is $1.0$ as the model is trained on all data points, and they are linearly seperable.\n", + "For this we plot the decision boundary of the `LinearSVM` classifier for the training data and the corresponding test data." + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "code", + "execution_count": 5, + "outputs": [ + { + "data": { + "text/plain": "
", + "image/svg+xml": "\n\n\n \n \n \n \n 2024-10-22T18:04:37.637652\n image/svg+xml\n \n \n Matplotlib v3.8.0, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n" + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "fig, ax = plt.subplots()\n", + "\n", + "classifier.fit(X_train, y_train)\n", + "plot_synthetic_data(ax, X_train, y_train, X_test, y_test, \"Synthetic Classification Data\")\n", + "\n", + "DecisionBoundaryDisplay.from_estimator(\n", + " classifier,\n", + " X_train,\n", + " plot_method=\"contour\",\n", + " ax=ax,\n", + " levels=[-1, 0, 1],\n", + " linestyles=[\"--\", \"-\", \"--\"],\n", + " colors=[COLORS_K_SII[1], RED.hex, COLORS_K_SII[2]],\n", + " alpha=0.5,\n", + ")\n", + "ax.set_xlabel(\"Feature 1\")\n", + "ax.set_ylabel(\"Feature 2\")\n", + "ax.set_title(\"Decision Boundary\")\n", + "plt.show()" + ], + "metadata": { + "collapsed": false, + "ExecuteTime": { + "end_time": "2024-10-22T16:04:37.657117Z", + "start_time": "2024-10-22T16:04:37.567430Z" + } + } + }, + { + "cell_type": "markdown", + "source": [ + "### Computing Shapley Values\n", + "Now we can compute the Shapley values for the data valuation game.\n", + "Intuitively, the Shapley values should all be positive as each training point makes the model more aware of the natural boundarie between the two classes." + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "code", + "execution_count": 6, + "outputs": [ + { + "data": { + "text/plain": "
", + "image/svg+xml": "\n\n\n \n \n \n \n 2024-10-22T18:04:38.986904\n image/svg+xml\n \n \n Matplotlib v3.8.0, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n" + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Compute Shapley values with the ShapIQ approximator for the game function\n", + "exactComputer = shapiq.ExactComputer(n_players=n_players, game_fun=data_valuation_game)\n", + "sv_values = exactComputer(\"SV\")\n", + "sv_values.plot_stacked_bar(\n", + " title=\"Shapley Values for Synthetic (Training) Data\",\n", + " xlabel=\"Data Point\",\n", + " ylabel=\"Shapley Value\",\n", + ")\n", + "plt.show()" + ], + "metadata": { + "collapsed": false, + "ExecuteTime": { + "end_time": "2024-10-22T16:04:39.005886Z", + "start_time": "2024-10-22T16:04:37.657523Z" + } + } + }, + { + "cell_type": "markdown", + "source": [ + "The Shapley values are all positive indicating that all data points have a positive impact on the model's performance.\n", + "Interestingly the Shapley values indicate the first five data points to be more important.\n", + "\n", + "To understand this better notice that we have *four* blue test points and *six* orange test points.\n", + "If we are provided with a training set that contains only orange points, the model will have an accuracy of $0.6$.\n", + "On the other side, if we are provided with a training set that contains only blue points, the model will have an accuracy of $0.4$.\n", + "Thus having orange points in the training set is more important for the model's performance.\n", + "Meaning that the orange points are more important regarding accuracy.\n", + "These are exactly the first five data points, which are all orange.\n", + "\n" + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "code", + "execution_count": 7, + "outputs": [ + { + "data": { + "text/plain": "
", + "image/svg+xml": "\n\n\n \n \n \n \n 2024-10-22T18:04:39.045802\n image/svg+xml\n \n \n Matplotlib v3.8.0, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n" + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "fig, ax = plt.subplots()\n", + "\n", + "plot_synthetic_data(ax, X_train, y_train, X_test, y_test, \"Synthetic Classification Data\")\n", + "for i in range(5):\n", + " ax.annotate(\n", + " f\"Point {i+1}\",\n", + " (X_train[i, 0], X_train[i, 1]),\n", + " textcoords=\"offset points\",\n", + " xytext=(0, -10),\n", + " ha=\"center\",\n", + " )\n", + "\n", + "plt.show()" + ], + "metadata": { + "collapsed": false, + "ExecuteTime": { + "end_time": "2024-10-22T16:04:39.066692Z", + "start_time": "2024-10-22T16:04:39.013827Z" + } + } + }, + { + "cell_type": "markdown", + "source": [ + "### Corrupting the Data\n", + "We can now investigate the impact of corrupting the data on the Shapley values.\n", + "Currently the Shapley values are less interesting as we have a clear boundary between the two classes.\n", + "If we now corrupt the data by adding noise to the labels, the Shapley values should change and identify the corrupted samples." + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "code", + "execution_count": 8, + "outputs": [ + { + "data": { + "text/plain": "
", + "image/svg+xml": "\n\n\n \n \n \n \n 2024-10-22T18:04:39.183419\n image/svg+xml\n \n \n Matplotlib v3.8.0, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n" + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from matplotlib import patches\n", + "\n", + "fig, ax = plt.subplots(nrows=1, ncols=2, figsize=(12, 8))\n", + "\n", + "plot_synthetic_data(ax[0], X_train, y_train, X_test, y_test, \"Synthetic Classification Data\")\n", + "\n", + "corrupted_X_train = X_train.copy()\n", + "corruped_y_train = y_train.copy()\n", + "\n", + "corruped_y_train[5] = 1\n", + "corruped_y_train[2] = 2\n", + "plot_synthetic_data(\n", + " ax[1],\n", + " corrupted_X_train,\n", + " corruped_y_train,\n", + " X_test,\n", + " y_test,\n", + " \"Corrupted Synthetic Classification Data\",\n", + ")" + ], + "metadata": { + "collapsed": false, + "ExecuteTime": { + "end_time": "2024-10-22T16:04:39.216073Z", + "start_time": "2024-10-22T16:04:39.071637Z" + } + } + }, + { + "cell_type": "markdown", + "source": [ + "Let us now look at the Shapley values for the corrupted data.\n", + "The Shapley values should now identify the corrupted samples." + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "code", + "execution_count": 9, + "outputs": [ + { + "data": { + "text/plain": "
", + "image/svg+xml": "\n\n\n \n \n \n \n 2024-10-22T18:04:40.617300\n image/svg+xml\n \n \n Matplotlib v3.8.0, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n" + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "data_valuation_game = SyntheticDataValuation(\n", + " classifier=classifier,\n", + " n_players=n_players,\n", + " X_train=corrupted_X_train,\n", + " y_train=corruped_y_train,\n", + " X_test=X_test,\n", + " y_test=y_test,\n", + ")\n", + "\n", + "# Compute Shapley values with the shapiq ExactComputer for the game function\n", + "exactComputer = shapiq.ExactComputer(n_players=n_players, game_fun=data_valuation_game)\n", + "sv_values = exactComputer(\"SV\")\n", + "sv_values.plot_stacked_bar()\n", + "plt.show()" + ], + "metadata": { + "collapsed": false, + "ExecuteTime": { + "end_time": "2024-10-22T16:04:40.635663Z", + "start_time": "2024-10-22T16:04:39.216847Z" + } + } + }, + { + "cell_type": "markdown", + "source": [ + "With both corrupted samples identified by the Shapley values, we can now remove them from the training data and our model should perform better on the test data." + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "code", + "execution_count": 10, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Accuracy on test data before removing corrupted samples: 0.5\n", + "Accuracy on test data after removing corrupted samples: 1.0\n" + ] + } + ], + "source": [ + "classifier.fit(corrupted_X_train, corruped_y_train)\n", + "print(\"Accuracy on test data before removing corrupted samples: \", classifier.score(X_test, y_test))\n", + "\n", + "cleaned_X_train = np.delete(corrupted_X_train, [5, 2], axis=0)\n", + "cleaned_y_train = np.delete(corruped_y_train, [5, 2], axis=0)\n", + "classifier.fit(cleaned_X_train, cleaned_y_train)\n", + "print(\"Accuracy on test data after removing corrupted samples: \", classifier.score(X_test, y_test))" + ], + "metadata": { + "collapsed": false, + "ExecuteTime": { + "end_time": "2024-10-22T16:04:40.640022Z", + "start_time": "2024-10-22T16:04:40.636571Z" + } + } + }, + { + "cell_type": "markdown", + "source": [ + "To verify this as sensible we plot the decision boundary of the `LinearSVM` classifier for the corrupted and cleaned training data and the corresponding test data." + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "code", + "execution_count": 11, + "outputs": [ + { + "data": { + "text/plain": "
", + "image/svg+xml": "\n\n\n \n \n \n \n 2024-10-22T18:04:40.745038\n image/svg+xml\n \n \n Matplotlib v3.8.0, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n" + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "def plot_decision_boundary(ax, classifier, X_train, y_train, X_test, y_test):\n", + " classifier.fit(X_train, y_train)\n", + " plot_synthetic_data(ax, X_train, y_train, X_test, y_test, \"Synthetic Classification Data\")\n", + " DecisionBoundaryDisplay.from_estimator(\n", + " classifier,\n", + " X_train,\n", + " plot_method=\"contour\",\n", + " ax=ax,\n", + " levels=[-1, 0, 1],\n", + " linestyles=[\"--\", \"-\", \"--\"],\n", + " alpha=0.5,\n", + " )\n", + " ax.set_xlabel(\"Feature 1\")\n", + " ax.set_ylabel(\"Feature 2\")\n", + " ax.set_title(\"Decision Boundary\")\n", + "\n", + "\n", + "fig, ax = plt.subplots(nrows=1, ncols=2, figsize=(12, 7))\n", + "fig.suptitle(\"Decision Boundary of Linear SVM\")\n", + "# Plot the decision boundary of the model with corrupted samples\n", + "plot_decision_boundary(ax[0], classifier, corrupted_X_train, corruped_y_train, X_test, y_test)\n", + "\n", + "# Plot the decision boundary of the model with removed corrupted samples\n", + "plot_decision_boundary(ax[1], classifier, cleaned_X_train, cleaned_y_train, X_test, y_test)\n", + "plt.show()" + ], + "metadata": { + "collapsed": false, + "ExecuteTime": { + "end_time": "2024-10-22T16:04:40.881221Z", + "start_time": "2024-10-22T16:04:40.641575Z" + } + } + }, + { + "cell_type": "markdown", + "source": [ + "## Real Data\n", + "We now demonstrate the data valuation for the [AdultCensus](../api/shapiq.datasets.rst) dataset.\n", + "Due to increasing runtime we choose a subset of the data, consisting of 200 samples.\n", + "Then we divide the data into training and test data at an 80/20 ratio." + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "code", + "execution_count": 12, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Players: 160\n" + ] + } + ], + "source": [ + "from sklearn.tree import DecisionTreeClassifier\n", + "from sklearn.model_selection import train_test_split\n", + "\n", + "X, y = shapiq.load_adult_census(to_numpy=True)\n", + "\n", + "X, y = X[:200], y[:200]\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=random_state)\n", + "classifier = DecisionTreeClassifier(random_state=random_state)\n", + "n_players = X_train.shape[0]\n", + "print(\"Players: \", n_players)" + ], + "metadata": { + "collapsed": false, + "ExecuteTime": { + "end_time": "2024-10-22T16:04:41.055176Z", + "start_time": "2024-10-22T16:04:40.879435Z" + } + } + }, + { + "cell_type": "code", + "execution_count": 13, + "outputs": [], + "source": [ + "data_valuation_game = SyntheticDataValuation(\n", + " classifier=classifier,\n", + " n_players=n_players,\n", + " X_train=X_train,\n", + " y_train=y_train,\n", + " X_test=X_test,\n", + " y_test=y_test,\n", + ")" + ], + "metadata": { + "collapsed": false, + "ExecuteTime": { + "end_time": "2024-10-22T16:04:41.056814Z", + "start_time": "2024-10-22T16:04:41.055761Z" + } + } + }, + { + "cell_type": "markdown", + "source": [ + "In the next step we show how different budgets influence the quality of approximation and the corresponding accuracy tradeoff." + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "code", + "execution_count": 14, + "outputs": [ + { + "data": { + "text/plain": "
", + "image/svg+xml": "\n\n\n \n \n \n \n 2024-10-22T18:05:15.036777\n image/svg+xml\n \n \n Matplotlib v3.8.0, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n" + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "budgets = [10, 100, 1000, 5000]\n", + "erg = {}\n", + "for budget in budgets:\n", + " # Compute Shapley interactions with the SVARM approximator for the game function\n", + " approximator = shapiq.SVARM(n=n_players, random_state=random_state)\n", + " shapley_approx = approximator.approximate(budget=budget, game=data_valuation_game)\n", + "\n", + " # Sort the approximated values of each player and get the keys\n", + " players = np.array(range(0, n_players))\n", + " sv_values = shapley_approx.values[1:]\n", + " idx = np.argsort(sv_values)\n", + " sorted_players = players[idx]\n", + "\n", + " # Compute the accuracy of the model for different amount of removed samples\n", + " percent_removal = [0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]\n", + " accuracies = []\n", + " for p in percent_removal:\n", + " n_samples_to_remove = int(p * n_players)\n", + " removed_players = sorted_players[:n_samples_to_remove]\n", + " cleaned_X_train = np.delete(X_train, removed_players, axis=0)\n", + " cleaned_y_train = np.delete(y_train, removed_players, axis=0)\n", + " classifier.fit(cleaned_X_train, cleaned_y_train)\n", + " accuracies.append(classifier.score(X_test, y_test))\n", + " erg[budget] = (percent_removal, accuracies)\n", + "\n", + "# plot the results\n", + "\n", + "fig, ax = plt.subplots()\n", + "fig.suptitle(\"Accuracy of the model on the test data after removing samples\")\n", + "for i, (budget, (percent_removal, accuracies)) in enumerate(erg.items()):\n", + " ax.plot(\n", + " percent_removal,\n", + " accuracies,\n", + " label=f\"Budget: {budget}\",\n", + " marker=\"o\",\n", + " linestyle=\"-\",\n", + " color=COLORS_K_SII[i],\n", + " )\n", + "plt.xlabel(\"Percentage of removed samples\")\n", + "plt.ylabel(\"Accuracy\")\n", + "plt.legend()\n", + "plt.show()" + ], + "metadata": { + "collapsed": false, + "ExecuteTime": { + "end_time": "2024-10-22T16:05:15.069797Z", + "start_time": "2024-10-22T16:04:41.062860Z" + } + } + }, + { + "cell_type": "markdown", + "source": [ + "Intuitively increasing amount of removed data samples with low Shapley values should yield a better model performance.\n", + "Increasing the budget yields a more clear effect for lower percentages.\n", + "For very high percentages the effect is less pronounced as the model is already trained on the most important samples." + ], + "metadata": { + "collapsed": false + } + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 2 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython2", + "version": "2.7.6" + } + }, + "nbformat": 4, + "nbformat_minor": 0 +} diff --git a/docs/source/notebooks/sv_calculation.ipynb b/docs/source/notebooks/sv_calculation.ipynb new file mode 100644 index 00000000..a7969431 --- /dev/null +++ b/docs/source/notebooks/sv_calculation.ipynb @@ -0,0 +1,84 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "source": [ + "# Shapley Value Calculation\n", + "A popular approach to tackle the problem of XAI is to use concepts from game theory in particular cooperative game theory.\n", + "The most popular method is to use the **Shapley Values** named after Lloyd Shapley, who introduced it in 1951 with his work *\"II: The Value of an n-Person Game\"*.\n", + "\n", + "## Cooperative Game Theory\n", + "Cooperative game theory deals with the study of games in which players/participants can form groups to achieve a collective payoff. More formally a cooperative game is defined as a tuple $(N,\\nu)$ where:\n", + "- $N$ is a finite set of players\n", + "- $\\nu$ is a characteristic function that maps every coalition of players to a real number, i.e. $\\nu:2^N \\rightarrow \\mathbb{R}$\n", + "\n", + "Of particular interest is to find a concept that distributes the payoff of $\\nu(N)$ among the players, as it is assumed that the *grand coalition* $N$ is formed.\n", + "The distribution of the payoff among the players is called a *solution concept*.\n", + "\n", + "## Shapley Values: A Unique Solution Concept\n", + "Given a cooperative game $(N,\\nu)$, the Shapley value is a payoff vector dividing the total payoff $\\nu(N)$ among the players. The Shapley value of player $i$ is denoted by $\\phi_i(\\nu)$ and is defined as:\n", + "$$\n", + "\\phi_i(\\nu) := \\sum_{S \\subseteq N \\setminus \\{i\\}} \\frac{|S|!(|N|-|S|-1)!}{|N|!} [\\nu(S \\cup \\{i\\}) - \\nu(S)]\n", + "$$\n", + "and can be interpreted as the average marginal contribution of player $i$ across all possible permutations of the players.\n", + "Its popularity arises from uniquely satisfies the following properties:\n", + "- **Efficiency**: The sum of the Shapley values equals the total payoff, i.e. $\\sum_{i \\in N} \\phi_i(\\nu) = \\nu(N)$\n", + "- **Symmetry**: If two players $i$ and $j$ are such that for all coalitions $S \\subseteq N \\setminus \\{i,j\\}$, $\\nu(S \\cup \\{i\\}) = \\nu(S \\cup \\{j\\})$, then $\\phi_i(\\nu) = \\phi_j(\\nu)$\n", + "- **Additivity**: For a game $(N,\\nu + \\mu)$ based on two games $(N,\\nu)$ and $(N,\\mu)$, the Shapley value of the sum of the games is the sum of the Shapley values, i.e. $\\phi_i(\\nu + \\mu) = \\phi_i(\\nu) + \\phi_i(\\mu)$\n", + "- **Dummy Player**: If for a player $i$ is holds for all coalitions $S \\subseteq N \\setminus \\{i\\}$, $\\nu(S \\cup \\{i\\}) - \\nu(S) = \\nu(\\{i\\})$ then $\\phi_i(\\nu) = \\nu(\\{i\\})$\n", + "\n", + "## Shapley Values: Cooking Game\n", + "To illustrate the concept of Shapley values, we consider a simple example of a cooking game.\n", + " The game consists of three players(cooks), Alice, Bob, and Charlie, who are cooking a meal together.\n", + " The characteristic function $\\nu$ maps each coalition of players to the quality of the meal." + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "markdown", + "source": [], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "markdown", + "source": [], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 2 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython2", + "version": "2.7.6" + } + }, + "nbformat": 4, + "nbformat_minor": 0 +} diff --git a/docs/source/notebooks/visualizing_shapley_interactions.ipynb b/docs/source/notebooks/visualizing_shapley_interactions.ipynb new file mode 100644 index 00000000..b356f014 --- /dev/null +++ b/docs/source/notebooks/visualizing_shapley_interactions.ipynb @@ -0,0 +1,593 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Visualizing Shapley Interactions\n", + "\n", + "This notebook shows how to visualize Shapley interactions in different ways using the `shapiq` package.\n", + "This notebook we will compute Shapley interactions of different orders with the `TreeExplainer` for a `RandomForestRegressor` trained on the `california_housing` dataset.\n", + "We will then visualize the Shapley interactions with different visualization techniques from the `shapiq` package." + ] + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": [ + "## Import Modules\n", + "First, import all necessary modules." + ] + }, + { + "metadata": { + "ExecuteTime": { + "end_time": "2024-10-24T12:22:45.017956Z", + "start_time": "2024-10-24T12:22:43.699244Z" + } + }, + "cell_type": "code", + "source": [ + "from tqdm.asyncio import tqdm\n", + "from sklearn.model_selection import train_test_split\n", + "from sklearn.metrics import mean_squared_error, r2_score\n", + "from sklearn.ensemble import RandomForestRegressor\n", + "\n", + "import shapiq\n", + "\n", + "shapiq.__version__" + ], + "outputs": [ + { + "data": { + "text/plain": [ + "'1.0.1.9001'" + ] + }, + "execution_count": 1, + "metadata": {}, + "output_type": "execute_result" + } + ], + "execution_count": 1 + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": [ + "## Load Data and Train Model\n", + "First, we load the `california_housing` dataset and train a `RandomForest` model." + ] + }, + { + "metadata": { + "ExecuteTime": { + "end_time": "2024-10-24T12:22:46.343969Z", + "start_time": "2024-10-24T12:22:45.018894Z" + } + }, + "cell_type": "code", + "source": [ + "# get the data\n", + "x_data, y_data = shapiq.datasets.load_california_housing(to_numpy=False)\n", + "feature_names = list(x_data.columns) # get the feature names\n", + "n_features = len(feature_names)\n", + "x_data, y_data = x_data.values, y_data.values # transform to numpy arrays\n", + "print(\"Features in the dataset:\", feature_names)\n", + "\n", + "x_train, x_test, y_train, y_test = train_test_split(x_data, y_data, test_size=0.2, random_state=42)\n", + "\n", + "# train an XGBoost model\n", + "model = RandomForestRegressor(random_state=42, max_depth=15, n_estimators=15)\n", + "model.fit(x_train, y_train)\n", + "\n", + "# evaluate the model\n", + "mse = mean_squared_error(y_test, model.predict(x_test))\n", + "r2 = r2_score(y_test, model.predict(x_test))\n", + "print(f\"Mean Squared Error: {mse}\")\n", + "print(f\"R2 Score: {r2}\")" + ], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Features in the dataset: ['MedInc', 'HouseAge', 'AveRooms', 'AveBedrms', 'Population', 'AveOccup', 'Latitude', 'Longitude']\n", + "Mean Squared Error: 0.2731205827334615\n", + "R2 Score: 0.7915760748983114\n" + ] + } + ], + "execution_count": 2 + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": "With these hyperparameters the model achieves a reasonable $R^2$ score of about 80%." + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": [ + "## Compute Shapley Interactions\n", + "Now, let's compute Shapley interactions of different order. \n", + "To do so, we will rely on the `TreeExplainer`. \n", + "Note that any other explainer (also the `ExactComputer`) that supports Shapley interactions can be used as well." + ] + }, + { + "metadata": { + "ExecuteTime": { + "end_time": "2024-10-24T12:22:46.360063Z", + "start_time": "2024-10-24T12:22:46.344975Z" + } + }, + "cell_type": "code", + "source": [ + "# select a local instance to be explained\n", + "instance_id = 7\n", + "x_explain = x_test[instance_id]\n", + "y_true = y_test[instance_id]\n", + "y_pred = model.predict(x_explain.reshape(1, -1))[0]\n", + "print(f\"Instance {instance_id}, True Value: {y_true}, Predicted Value: {y_pred}\")\n", + "for i, feature in enumerate(feature_names):\n", + " print(f\"{feature}: {x_explain[i]}\")" + ], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Instance 7, True Value: 1.575, Predicted Value: 1.6279253275277228\n", + "MedInc: 3.6908\n", + "HouseAge: 38.0\n", + "AveRooms: 4.962825278810409\n", + "AveBedrms: 1.0483271375464684\n", + "Population: 1011.0\n", + "AveOccup: 3.758364312267658\n", + "Latitude: 33.92\n", + "Longitude: -118.08\n" + ] + } + ], + "execution_count": 3 + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": [ + "For the instance with `instance_id=7`, the model predicts a property value of about 1.62 (in 100,000 USD).\n", + "The ground truth value is 1.575 (in 100,000 USD).\n", + "So, the model is off by a bit but not too much." + ] + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": [ + "Now, let's compute $k$-SII scores of **different orders**.\n", + "Here, we will start with Shapley interactions of order 1 (which corresponds to the Shapley values) and increase the order up to the number of features (for $k$-SII, this corresponds to the Moebius transform)." + ] + }, + { + "metadata": { + "ExecuteTime": { + "end_time": "2024-10-24T12:25:24.261892Z", + "start_time": "2024-10-24T12:22:46.362063Z" + } + }, + "cell_type": "code", + "source": [ + "# create explanations for different orders\n", + "si_order: dict[int, shapiq.InteractionValues] = {}\n", + "for order in tqdm([1, 2, n_features]):\n", + " index = \"k-SII\" if order > 1 else \"SV\" # will also be set automatically by the explainer\n", + " explainer = shapiq.TreeExplainer(model=model, max_order=order, index=index)\n", + " si_order[order] = explainer.explain(x=x_explain)\n", + "si_order" + ], + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "100%|██████████| 3/3 [02:37<00:00, 52.63s/it]\n" + ] + }, + { + "data": { + "text/plain": [ + "{1: InteractionValues(\n", + " index=SV, max_order=1, min_order=0, estimated=False, estimation_budget=None,\n", + " n_players=8, baseline_value=2.0718777703488374\n", + " ),\n", + " 2: InteractionValues(\n", + " index=k-SII, max_order=2, min_order=0, estimated=False, estimation_budget=None,\n", + " n_players=8, baseline_value=2.0718777703488374\n", + " ),\n", + " 8: InteractionValues(\n", + " index=k-SII, max_order=8, min_order=0, estimated=False, estimation_budget=None,\n", + " n_players=8, baseline_value=2.0718777703488374\n", + " )}" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "execution_count": 4 + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": [ + "## Visualize the Shapley Interactions\n", + "Now that we have some explanations to visualize, let's use different visualization techniques from the `shapiq` package.\n", + "We will showcase the following visualization techniques:\n", + "- Force Plot\n", + "- Waterfall Plot\n", + "- Network Plot\n", + "- SI Graph Plot\n", + "- Bar Plot (for global explanations)" + ] + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": [ + "### Force Plot\n", + "First, we consider the classical force plot popularized by [`shap`](https://github.com/shap/shap/tree/master).\n", + "Similar to how Shapley values are drawn on the force plot, Shapley interactions can be visualized as well.\n", + "Positive interactions are shown in red, negative interactions in blue. \n", + "All interactions **force** the prediction of the model *away* from the base value and *towards* the predicted value by the model considering all features.\n", + "The following cell plots the force plot for the SV, 2-SII, and Moebius transform." + ] + }, + { + "metadata": { + "ExecuteTime": { + "end_time": "2024-10-24T12:25:27.045889Z", + "start_time": "2024-10-24T12:25:24.265880Z" + } + }, + "cell_type": "code", + "source": [ + "sv = si_order[1] # get the SV\n", + "si = si_order[2] # get the 2-SII\n", + "mi = si_order[n_features] # get the Moebius transform\n", + "\n", + "sv.plot_force(feature_names=feature_names, feature_values=x_explain, show=True)\n", + "si.plot_force(feature_names=feature_names, feature_values=x_explain, show=True)\n", + "mi.plot_force(feature_names=feature_names, feature_values=x_explain, show=True)" + ], + "outputs": [ + { + "data": { + "text/plain": [ + "
" + ], + "image/png": "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" + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "
" + ], + "image/png": "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" + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "
" + ], + "image/png": "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" + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "execution_count": 5 + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": [ + "The first order force plots shows that the prediction is **negatively influenced** by the *AveOccup* (AO), *MedianIncome* (MI), and *AveRooms* (AR) features and **positively influenced** by the *Latitude* (Lat.) and *HouseAge* (HA) features. Ultimately, the prediction is less than the base value of the model (which is about 2.07). The higher-order force plots show also the interactions. This reveals that the exact location of the property, encoded in the interaction between *Latitude* (Lat.) and *Longitude* (Long.), increases the predicted value of the property. Notably, the interaction between the *HouseAge* (HA) and *AvgOccup* (AO) features has a notable negative influence on the prediction.\n", + "\n", + "Notice how for an increasing order, more interaction terms are being displayed. This is to be expected as with an increasing order more interactions are part of each explanation. Formally, the number of interactions scales with $\\mathcal{O}\\binom{n}{k}$, where $n$ is the number of features and $k$ is the order of the Shapley interactions. Plotting this with force plots can become quite messy quite quickly. This problem is exacerbated when the number of features is large In this example, only 8 features are being considered which typically is not a lot. To solve this other plots can be helpful. For example the waterfall plot." + ] + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": [ + "### Waterfall Plot\n", + "Like the force plot, the waterfall plot is also a popular visualization technique for Shapley values.\n", + "It, too, was established with `shap`.\n", + "Similarly to the force plot, the waterfall plot breaks down the explanation into interactions influencing the prediction positively or negatively.\n", + "The benefit for the waterfall plot is the automatic grouping of low-magnitude interactions (interactions with low value) into an *other* group." + ] + }, + { + "metadata": { + "ExecuteTime": { + "end_time": "2024-10-24T12:25:28.367703Z", + "start_time": "2024-10-24T12:25:27.047881Z" + } + }, + "cell_type": "code", + "source": [ + "sv.plot_waterfall(feature_names=feature_names, feature_values=x_explain, show=True)\n", + "si.plot_waterfall(feature_names=feature_names, feature_values=x_explain, show=True)\n", + "mi.plot_waterfall(feature_names=feature_names, feature_values=x_explain, show=True)" + ], + "outputs": [ + { + "data": { + "text/plain": [ + "
" + ], + "image/png": "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" + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "
" + ], + "image/png": "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" + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "
" + ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAzUAAAJYCAYAAABb8cjcAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAADOoElEQVR4nOzdd3xUVfrH8c+d9EAglNBCDaETOiiCsIKuCIqogCiCFBcFC6uiLii7Kgq7gIqADUQRAQtdQEBBAeuPGkIPvSYhCSW9zczvj5iBkDYJSS5Jvu/XKy8y555773MjhnnmnPMcw2632xERERERESmhLGYHICIiIiIiciOU1IiIiIiISImmpEZEREREREo0JTUiIiIiIlKiKakREREREZESTUmNiIiIiIiUaEpqRERERESkRFNSIyIiIiIiJZqSGhERERERKdGU1IiIiIiISImmpEZEREREREo0JTUiZZzNZuOtt96iYcOGuLm50bBhQ6ZOnUrTpk2x2Wz5vt7HH39M3bp1SU5OLoJoRURERLIy7Ha73ewgRMQ8s2fP5rnnnuPFF1+kVatWVKxYkeHDhzN9+nSGDx+e7+slJSVRv359JkyYwHPPPVcEEWeVmprKnj17uHDhApGRkSQnJ9O9e3eaNGmSr+tERUWxY8cOIiIiSEtLo0KFCjRr1oyWLVsCcOHCBUJDQwkLCyM2NhYPDw+qV69Ohw4d8PX1zXK9K1eusH37diIiIkhKSqJ8+fIEBgbSunVrXF1dC+PRRUREBCU1ImVe+/btqVq1Khs2bABgxowZ/Oc//yEiIgJPT88CXfOVV17hm2++4cSJExiGUZjhZis2NpavvvqK8uXL4+PjQ1hYWL6TmrNnz7J+/XqqVq1KQEAAbm5uxMTEYLfbufXWWwH48ccfCQ8PJyAggCpVqpCQkMD+/ftJTU2lX79+VK5c2XG9uLg4li5diru7O82bN8fDw4OIiAhCQ0OpV68ed999d6H/HERERMoqfVQoUoYlJSWxZ88e3njjDUfb559/Tt++fQuc0AAMHDiQqVOn8vPPP9OjR4/CCDVX3t7ePPbYY3h7exMZGcmKFSvydX5KSgo///wzdevW5a677soxEQsKCqJHjx64uLg42ho2bMjSpUsJDg7O9KxHjhwhJSWFvn37OpKdZs2aYbfbOXLkCMnJyXh4eBTgaUVEROR6WlMjUkaNHDkSLy8vrFYrr732GoZhULNmTUJCQrjzzjuz9D937hyenp6MGDEiU/vGjRtxc3Pj+eefd7S1b9+eypUrs2rVqiJ/DgAXFxe8vb0LfP7Ro0dJTEykY8eOGIZBamoq2Q1i16hRI1NCA1CxYkUqVarE5cuXM7WnpKQAZInL29sbwzCwWPTrV0REpLBopEakjBo8eDBubm588sknvP/++1SuXJljx47x+uuv065duyz9/f39eeKJJ5gzZw7/+c9/qFevHocOHWLAgAHcc889vPPOO5n6t2vXjt9++y3POGw2myMByIuHh0eRTGc7d+4cbm5uxMfH88MPP3DlyhVcXV1p1KgRnTt3znX9i91uJzExkUqVKmVqr1WrFnv27GHLli20b98eT09PIiIiOHDgAC1atMDNza3Qn0NERKSsUlIjUkb16NGDTZs2Ua5cOZ555hksFgsTJ04EoEGDBtmeM378eD799FP+97//MWnSJO69917q16/PV199lWXkISAggC+//DLPOMLDw1mzZo1TMT/yyCP4+Pg41Tc/rly5gt1u54cffqBJkyZ06tSJ8+fPs3//flJSUujZs2eO5x49epT4+Hjat2+fqb1OnTp06NCB3bt3c+rUKUd727Zt6dixY6E/g4iISFmmpEakDAsJCaFFixaOhCQ6OhpXV1fKly+fbX9/f3/+8Y9/MHfuXHbt2kViYiJbtmyhXLlyWfpWqlSJxMREEhIScp0aVqVKFXr37u1UvF5eXk71y6/U1FTS0tJo1qwZXbp0AdITO5vNxsGDB+nQoQMVK1bMct7ly5f59ddfqV69Oo0bN85y3MfHh5o1a9KgQQM8PT05ffo0u3fvxsvLy1FRTURERG6ckhqRMmzPnj35rsI1btw4Zs+eTUhICL/88gv+/v7Z9stYk5LXdDEPDw9q166drxgKW8b0ssDAwEztgYGBHDx4kIiIiCxJTUJCAuvWrcPd3Z0777wzy0jV0aNH2bp1Kw8//LAjSWzQoAF2u51t27YRGBh4Q8UYRERE5CqtVBUpoy5fvsyZM2cICgpytFWpUoW0tDRiY2NzPO/tt98GIC0tLVMJ4+tdunQJb2/vPEdXrFYrCQkJTn0VZDNQZ2SMJF0fa8br6zcSTUlJYd26daSkpNC7d+9sR6oOHDhA1apVs4x61atXj7S0NKKiogrzEURERMo0jdSIlFEhISEAtGrVytHWtGlTAE6cOJGpPcO0adP49NNPmT17Ni+99BJvv/02n376abbXP3HiBM2aNcszjoiICNPX1Pj5+XHu3Dni4+MzbaIZHx8PZE520tLSWL9+PVeuXKFPnz5ZCgRkSExMzLZkc0Zipi3CRERECo+SGpEyas+ePUDmpKZz584A7NixI0tSs3LlSv71r38xadIknn76aY4cOcKHH37Iq6++mm1hgV27djF48OA84yjONTVpaWnExcXh6emZaepXQEAAwcHBHD58ONN0ukOHDjlKXUN6QrJp0yYiIiK4++67qV69eo73qlixImfPnuXy5cuZEqVjx45hGEauo1wiIiKSP4ZdHxeKlEn/+Mc/WLduHWfPns3UHhQURFBQEIsXL3a07dy5k27duvHQQw+xYMECAM6fP09AQACPPfZYltGanTt30qFDBzZu3Jhr5bDCtG/fPlJSUkhISODAgQPUr1+fqlWrAtCyZUvc3d05f/48a9asoV27dnTo0CHT+Vu2bOHw4cMEBARQs2ZNwsLCOH78OG3atKFTp04A/P777+zbt4+6devSsGHDLDE0atTI8X1YWBhr1qzB09OTFi1a4OHhwenTpzlz5gxNmzalW7duRfjTEBERKVs0UiNSRoWEhGQ7xWzEiBH8+9//JjExES8vL86ePct9991H27ZtmTt3rqNfrVq1GDFiBJ9++mmW0ZolS5ZQt25devToUSzPAunPExcX53h98uRJTp48CaQnG+7u7rmef/vtt1O+fHkOHz7MyZMnKV++PJ07d8605ig6OhqA06dPc/r06SzXuDapqVmzJvfffz87d+5k//79JCcn4+PjQ8eOHWnduvWNPKqIiIhcRyM1IpLJlStXCAgIYOrUqYwcOTLf5ycnJ1O/fn3+9a9/MXbs2CKIUERERCQzVT8TkUwqVqzIyy+/zLRp0wpUbezzzz/Hzc2Np556qgiiExEREclKIzUiIiIiIlKiaaRGRERERERKNCU1IiIiIiJSoimpERERERGREk1JjYiIiIiIlGhKakREREREpERTUiMieYqJieFvf/sbMTExZociIiIikoWSGhHJU0xMDFu2bFFSIyIiIjclJTUiIiIiIlKiKakREREREZESTUmNiIiIiIiUaEpqRCRPFSpUoHPnzlSoUMHsUERERESyMOx2u93sIERERERERApKIzUi4pS9e/eaHYKIiIhItpTUiIhTrFar2SGIiIiIZEtJjYg4pXLlymaHICIiIpItJTUi4pRKlSqZHYKIiIhItpTUiIhTjh07ZnYIIiIiItlSUiMiIiIiIiWaSjqLiFOuXLlCxYoVzQ5DREREJAuN1IiIU2JjY80OQURERCRbSmpExClRUVFmhyAiIiKSLSU1IuIUwzDMDkFEREQkW1pTIyIiIiIiJZpGakTEKfv37zc7BBEREZFsKakREaekpqaaHYKIiIhItpTUiIhTfH19zQ5BREREJFtKakTEKX5+fmaHICIiIpItJTUi4pQjR46YHYKIiIhItpTUiIiIiIhIiaakRkScUq9ePbNDEBEREcmWkhoRcUpiYqLZIYiIiIhkS0mNiDjlwoULZocgIiIiki1XswMQkZJvboiNBfttZochIlJg3q4wr5cLtX0Ms0MRkQIw7Ha73ewgROTmZ7PZsFiyDu6m2ezU+cRKeLwJQYmIFKJZPeCZdvq8V6Qk0vQzEXHK4cOHs21fccSuhEZESjwDOH/uvNlhiEgBKakREackJydn2/7eDhsWzdYQERERE2mMVUScUqFChSxtuyLs/BFmQjAiIkWgvI+P2SGISAFppEaKVXh4OO7u7hiGwdSpU3PsFxUVxejRowkICMDT0xMPDw9q167N4MGDOXHiRDFGLBlq1qyZpW3mLhuuGqURkVLCzU2f9YqUVEpqpFjNnDmTtLQ0/Pz8WLhwYbZ9duzYQfPmzfnkk0/w9/dn3LhxTJgwgaCgIL755htatWrF2rVrizlyuX5NzYV4O4sO2klTqRERKSUuXbxkdggiUkD6SEKK1VdffUVQUBC9e/fmf//7H3v27KF169aO4zExMfTt25eLFy8yb948hg8fnun8H3/8kfvvv59HHnmEPXv20KBBg+J+BPnLnBA7NiU0IiIichPQSI0Um02bNnHy5EkGDx7M008/jcVi4f3338/UZ8qUKYSFhTFo0KAsCQ3AXXfdxYsvvkhsbCyvvvpqcYUuQJ06dRzfp1rtzNxlU1IjIqVKterVzA5BRApISY0Umw8++AAPDw9GjRpF7dq1ufXWW1m1ahVWq9XRZ/Xq1QC8+OKLOV7npZdewsXFhY0bNxZ5zHJVamqq4/tlR+xEJpoYjIhIEYiJiTU7BBEpICU1Uizi4uLYsGEDf/vb3/D19QVg6NChXLx4MdPampMnT+Lp6Unbtm1zvFaFChXw9/cnMjKSS5c0/7m4hIeHO75/Z7vKOItI6ZOUqE9rREoqJTVSLObMmUNCQgIjR450tD3++OP4+Pgwb948R1tiYiLe3t55Xq9cuXJAepU0KV7bwuzsiEBTz0Sk1HFxcTE7BBEpIBUKkGLxxRdf4OPjQ8OGDdm9e7ejvVOnTmzZsoWzZ89Su3ZtvLy8SHTik7L4+PQt7KtWrVpkMUtmLVu2BK6WcVbVMymoih4wtZuFBxoZeLulJ8ovbrax+0Le5z4RZPBYcwtNK4OvB5yPh81n7Lzxu41TMTmf18Uffn0k/Z+8qh+kEa0P5CUb2ZWuF5GSQUmNFLmQkBD27t2L3W6nffv22faZOXMmU6dOpX79+uzfv5/du3fnOAUtJiaGc+fO4efnR6VKlYoydLnG0aNH8a3ThG8OqYyzFJwBrH3QhdZ+MG27jahEGNPGwuaHXWj/pZWjl3M/v211gxNX7Hx3zM6lJGhQ0eAfrQzuDXCh9RdWwuKzv+esHi7Epdgp7655k5Kzs2fPQtu6ZochIgWgpEaK3MyZM7Hb7bz55ptUrlw5y/HJkyfzzTffMHXqVPr06cP+/ft57733WLBgQbbXe+edd7BarfTs2bOoQ5drJCUl8XGwDZvZgUiJ1r+JQRd/g/7fWVkWmp4df3vYSuhIF97oYmHw2tz/hj298frjdlYehZ1DXBnawuB/27Jm3KNaG9TxgU/32vlneyU1IiKlkZIaKVJWq5WVK1dSt25dJk6cmG2fkJAQ5syZw4YNGxg/fjwLFixg8eLF3HXXXQwZMiRT302bNjF9+nR8fHx4++23i+MR5C9uXuWZ/av2ppEb07+xQXi8neWhV/8iRSXCt4ftPNbcwN0FUqy5XCAbJ6+k/+nrYQCZ/4JW8oS3ulj49+82qnkroZHclStfzuwQRKSAVChAitTixYuJjo6mV69eOfYZOnQoAB9++CG+vr6sWLECX19fHn/8cbp3787EiRN544036NOnD7169cIwDBYtWkRAQEBxPYYA25LqEp1kdhRS0rWtZrArws71ufG2MDvl3AwaOzmjtLIn+HlD++rwea/0f8o2nc6acU/qYiE8AT7Zo2xc8ubh4WF2CCJSQBqpkSI1d+5cgCwjLtfq0qULNWvW5McffyQ2NpZbb72VgwcPMnHiRDZs2MC0adOw2Wz4+fkxYMAA3nrrLSU0xcxut/O/P5KxGF4aqZEbUrMcbD2btT1jLUyt8gb7ovL+S3buKRc8XdNHXqIS7Dy7ycrGU5nPC6oKT7Y26L1MG8WKcy5GXwTKmx2GiBSAkhopUlu3bnWq3/nz5zO99vPz4+OPPy6KkKQA/gyDw3FeZochNxkDcHeyAm7yX1PKvFyvfn+tpLSrx51xzzIbnq7QrDI81txCObesfWb2dGHdCTs/nlJGIyJS2impEZE8LT1sw4IdG1qTIFd1q2Ow+WHnspqmn6Vx+CIkpoFHNqd4/vWvUWKac/fefCY9UVl/AlYdtbJvmAtxqTY+2J3ePrCJwW21oOV8lbYQ5/n5+ZkdgogUkNbUSJly8eJFkpOTHa/j4uKIjY11vE5JSSE6OjrTOWFhYbm+Dg8Px26/+klwabzHA40sSmgki0PRdoatszr1FRaXfk5YfPoUtOtltJ2Py/+oyvErsPsCDG529Z+0ad0tLDlsJ8UK9Sqkf/n+tVyijk/2MUhZZ3fsgQbm/97VPXSPsn6P/DLs10YkIpINu91O40/iOR7nqZLOckO+vc/C7bUNan1kzVQs4JO7LAxublB5tjXf1c8Adg1xwcMVWnyefrJ9XO4TEYIv2Gm7oAA3klLLAP7V5DyT79M+NSIlkaafiUieDMNgSN0o/nOgttmhSAm3NNTOgCYWHmxsOPapqeIFA5oYrD5mz5TQBFRM//P4XyWbXQzwcYfLyZmv2bEGBPnB4oNX06R+K7MmLIOaGgxqamHI91bOxmY5LIJh0QQWkZJKSY2IOOWff6vFjGNwKTnvviI5WRpq54/zdj7vZaF5FRtRiTCmjQUXA/7zW+ZxwE0D0xffNJibnqCUd4czT7rwzWE7+6PsxKdCkJ/B8JYGV5Jh0h9Xz191NOskhDbV0qdQrjthJzqxqJ5QSjL/WrXMDkFECkhJjYg4JeLcKZ5uG8CU/7Nj1aRVKSCbHXovszKtu4Xn2lrwcoPt4TBsnZXQS7mfm5AKn+61c0cdg/6NDbxc4XwcfHXQzlt/2jgVUzzPIKXXuXPnoG0ds8MQkQLQmhoRcUpwcDB+DVtTb45VSY2IlDpaUyNSsmnyqDht+/bt3Hnnnfj7++Pt7Y2Hhwf+/v4MGjSIY8eO5Xn+0qVLMQwj16/vvvvOqf633HJLUT6qZMPb2xt/n/RPyF1VCE1ESiHvct5mhyAiBaTpZ+K048ePExkZyV133UWdOnVwdXVl7969rFy5kg0bNhAcHEy9evVyPL99+/ZMmzYtS3tSUhL//ve/8fHx4e67785y/L777qNbt26Z2urXr3/DzyP506BBAwD+2d7CN4dVNUpESh8vLyU1IiWVkhpx2sMPP8zDDz+cpX327Nk8++yzvPfee8yYMSPH8xs0aMC4ceOytL/33nvY7XbuvfdePDw8shzv3LlztudJ8dq/fz9t2rThlprQthrsiUxfHyEiUlpER0UBmn4mUhJp+pncsMDAQAAuXcpjlW8O5s+fD8A///nPHPtcvnyZuLi4Al1fCpdhGLzQwaKERkRERG4aSmok3+Li4jh79iyHDx9m0aJFjB49GoAHHngg39cKCQlh7969NG/enI4dO2bbZ9KkSVSqVAkfHx9q1qzJK6+8gs2mLSCLW82aNR3fD2hsUMXTxGBERIpAlapVzA5BRApI088k3yZPnsyUKVMcr/38/Jg+fTr9+vXL97Xef/997HY7Q4cOzXLM3d2dW2+9lbvvvps6depw5swZFi1axNSpUwkJCWHdunU38hiSTy4uLo7vPVwNnm1n4c0/bBqxEZFSIykxCShndhgiUgAq6Sz5dvjwYXbt2kVsbCzbt29n48aNDBw4kP/973/5uk5qaio1atQgMTGR8PBwKlSokOc5VquVTp06sWvXLlatWkXfvn0L+hiST8HBwbRp08bxOjzeTp2PraTpN4iIlAIq6SxSsmmkRvKtSZMmNGnSBIBRo0axdetWevbsSUJCArNmzXL6OosWLeLixYv07dvXqYQG0kcLJkyYQP/+/Vm+fLmSGhPVKGfwcFODbw7ZldiISOmgcvUiJZbW1MgN69atGwEBASxevDhf53366acAPPPMM/k6r2nTpgBER0fn6zy5MRk/92uNbWdRQiMipUZt/9pmhyAiBaSkRgpFcnJyvqqTnTp1ij///JP69etz11135eteISEhAFSrVi1f58mNOXfuXJa2jjUNOtYAiz7dFJFS4HzYebNDEJECUlIjTjt+/Hi27UuWLOH06dOZPsk/duwY27ZtIyYmJttz3n//faxWK4MGDcrxftm9iU5ISGDSpEkADBgwID/hyw2KjY3Ntl3lnUWktLBZVVlTpKTSmhpx2pAhQ4iKiqJz587Ur1+fxMREdu/ezU8//YSnp2emjTeffvppNmzYwJIlS+jfv3+Wa33zzTe4ubkxduzYHO/XvXt3/Pz8aN26Nf7+/pw7d45Vq1YRHh5O//796dWrV1E8puTA0zP7Gs4PNTKo5g0XEoo5IBGRQmQHvLy9zA5DRApISY047eGHH2bRokWsXr2amJgYDMPAz8+Pfv36MWnSJJo1a+bUdVavXs358+fp2bMnNWrUyLFfnz59WL9+PYsXLyY+Ph4PDw8CAwN5+eWXef755wvrscRJjRo1yrbdzcXgP50tfLbPhmopikhJVc4NHmzmbXYYIlJAKuksIk65vqSziEhpo99zIiWX1tSIiIiIiEiJpqRGRJxSvXp1s0MQESlS9evXNzsEESkgJTUi4hQPDw+zQxARKVLx8fFmhyAiBaSkRkSccvr0abNDEBEpUpGRkWaHICIFpKRGRERERERKNFU/ExGnJCQk4O2tcqciUnrZ7XYMwzA7DBEpAO1TIyJOCQ8PJyAgwOwwRMxx/iL8cdjsKORGdWsOfhVzPHzw4EGaN29ejAGJSGFRUiMiTomJiTE7BBHz3D8FdhwzOwq5Ua88AP8dkuPhlJSUYgxGRAqT1tSIiFPc3d3NDkHEPBfjzI5ACsNHGyAhOcfDFSpUKMZgRKQwKakREac0a9bM7BBERG5MTAIs3prj4Ro1ahRjMCJSmJTUiIhT9uzZY3YIIiI3xjDg3dWQQ42k0NDQYg5IRAqLkpoisnTpUgzD4IUXXii0a+7bt49Ro0bx888/F9o1C6oonk9ERKRI2e1w8Cxs3md2JCJSyJTUlCCHDh1i7ty5/Pnnn2aHImWQn5+f2SGIiNw4FwvMWJPtobp16xZzMCJSWJTUiIhTypUrZ3YIIiI3zmqD1TvgRESWQ8nJORcREJGbm5Iak0VHRzNixAgaN26Mj48Pbm5u1KhRg6FDh2YqoTtlyhQGDBgAwIQJEzAMA8MwCAoKyvc9f/rpJ9zc3Gjbti02m83RnpqaSosWLfDw8OC333678Yf7y4oVK2jbti3e3t64u7sTEBDAW2+9laVfUFAQfn5+HD16lB49elCuXDk8PDxo3749O3bsKLR4pGBOnjxpdggiIoXDYsAH67I0R0RkTXREpGRQUmOyY8eOsXz5clq2bMmYMWN45ZVXaNKkCQsXLuSOO+5w9OvduzfDhg0D4L777mPatGlMmzaN8ePH5/uePXr0YNy4cQQHBzN27FhH++jRozlw4ADjx4+nS5cuN/xsAPPmzaN///6cOnWKoUOHMnbsWFxdXZk4cSLDhw/P0j85OZmuXbtisVh48cUXGTBgACEhIdx///2kpqYWSkwiIlLGWW0w5weITzI7EhEpJNp802RBQUFERETg4eGRqX3kyJF89tlnrF+/nl69etG6dWv69OnD/Pnz6dy5M+PGjbuh+06ZMoXNmzfz4Ycfcs8995CQkMBnn31G165def3112/o2hlSU1N55ZVX8PDwYNu2bQQGBgLw5ptv0q5dO7744gueffZZ2rVr5zgnNjaWESNGMGPGDEebn58fM2bM4JtvvuGxxx4rlNgk/zL++4mIlApxSfDlFnjqbkdTQWY/iMjNQSM1JvPy8nIkNCkpKYSHh3P27Fl69+4NwJYtW4rs3suWLcPX15dhw4YxatQoqlSpwtKlSwvt+ps2bSI6Opq+fftmekPs5eXF888/j91uZ+HChZnOMQwjy9S0Pn36AHDgwIFCi03yLyoqyuwQREQK17vfZSrvfOTIERODEZEboZGam8DEiRNZsGABZ86cwX5d7fxLly4V2X1r1arF7NmzefTRRwFYtWoV1atXL7TrZ9T7b9myZZZjHTt2BODEiROZ2itVqkT58uUztWVshhYdHV1osUn+Xb582ewQRKSwVPSGqUPhgVvA2wO2HYEXv4Ddx/M+t2MgDOsBtzSCVvXAzRWMB7P283SH2U/ALY2hTpX0qmPHIuCzTfDhekizFv5z5YcdOBIGG/fAXW0ASErSdDSRkkpJjcnGjRvHO++8Q7t27XjiiSeoU6cOHh4enD59mn/961+ZFvIXheXLlzu+37ZtG3379i3S++XFYsl58PD6hE+Kl6urfl2IlAqGAWtfg9b1YNoqiIqBMb1g85vQ/iU4Gpb7+b3bwxM9IeQUHI+AJv7Z9/NyhxZ14fudcDISbDa4rSm8Nzw9IRo8o9AfLd8yyjv/ldT4+PiYG4+IFJjepZhs2bJl+Pn5sW3bNlxcXBztX375ZZa+hmEU6r3nzp3L0qVL6dWrF6dPn2bq1Kn06tWLrl27Fsr1mzRpAqRvGnq9jGpmDRo0KJR7SdHLbsRNRG5CP78JJy/A8NnZH+/fGbo0hf7TYNkf6W3f/g6hs+GNh/NONj5aD/9bAUkpMOuJnJOaS3HQ+V+Z2z75Aa4kwLO94YX5EHE5Hw9WBKw2WLcrPZELrIm/fw7PIiI3Pa2pMVlGInPtiExKSgpTp07N0rdixYpAztOwtm3bRnBwsFP3PXLkCC+88AL+/v588803LFu2DHd3dx555BFiY2Pz+RTZ69GjB1WqVGH16tUcP351SkNycjIzZszAMIwCL/xPTk5m27ZtHDx4sFBilbzt2bPH7BBEpDD07wzhl2D5NRs5R8WkJzb3dwL3PD7vvHAlPaEpqJMX0v/0vUn2vrJYYHZ6eedDhw6ZHIyIFJRGaorY1q1bGTNmTJZ2Pz8/3njjDXr37s2sWbO45ZZbuO+++7hy5QqrVq3KNGqT4ZZbbsHT05OFCxdSrlw5KlWqRK1atRg4cKDjeNWqVYmMjMw1JqvVyoMPPkhycjJff/01FSpUoEKFCkyfPp3Ro0czePBgvvvuu0J5vv/973+MGjWKjh07MnDgQHx8fFi1ahWhoaEMGzYsU+Wz/Dhy5Ai33HILLVu2ZO/evQW6huSPpv+JlBJtG8Cu45kWyAPp62qe/Ds0rgX7Thfe/dxcoYJX+nS0DoEw7v70xCavaW7FxWqDT3+ESY+YHYmI3AAlNUVs586d7Ny5M0t7rVq1eOONN3jnnXew2+0sWbKEyZMnU7FiRXr16sUzzzxD586dM53j4+PDRx99xJtvvsnkyZNJS0ujZcuWjqTGWU8//TT79u1j4sSJmaaaPfXUU/zwww+sWLGCDz/8MNtkJb/PN3LkSHx9fXnrrbeYP38+aWlp1K5dm0mTJvHaa6/lK24xV5UqVcwOQUQKQ81KsDWbapJhfxWmqVW5cJOaB2+Br1+8+nr7URgxOz2ZuFkkJMMXP1N7UCezIxGRAjLs+vhVRJxw5coVxxRIkTKn4ej0RfE3G1eX9Epm11o1Hk5HwrOfZm6/GJc+OpO2BD75EZ6ek/n4HS3hpzeh339h1Tbn7j/rCXimd/bVzzJUqwhB9dKnm/UMgtb109fT/F+oc/coDgbQoDoRv/6b6jVrmh2NiBSARmpExCknTpygTZs2ZochItfq0hQ2T8q+/ZHbM7fVfxJORUJiCnhk88+/p3v6n4k3sF4mOxeuwKaQ9O+X/QHjH4If/wONnja/UEAGO3A8grhlv1L9mQFmRyMiBaBCAVKmXLx4keTkZMfruLi4TIURUlJSshRiCAsLy/V1eHh4pvUmpfUeVqu1VDyH7qF7FOQeaVaT91TJyZ6TcOfrmb/2nIQNu7O2h19OPyfsUvoUtOtltJ2/WLQxL/0dfLzSixLcZGzlPBzfl5a/u7qH7lFS75Ffmn4mIk6JiYmhQoUKZochYo6bdfpZdvIq6fztOLi9GdR6InOxgE+egsHdoPJQSElz7l7OTD+7Xqv6sOddeHkBTFvp/HlFyWJAizqk7piKm7u72dGISAFopEZEnHL58mWzQxCRwrD0D6hRCR689WpbFR8YcBus3pE5oQmonv5VEFVy2MjyiTvT/9xxtGDXLQo2O7zQlxMnT5odiYgUkNbUiIhTLl68SN26dc0OQ0Ru1NI/4I/D8Pkz0Lw2RMXCmF7gYoH/fJ2576Y30v9s8NTVtrp+MKR7+vcdAtP/fLV/+p+nImHhlvTvH+sOT/0dVm5LH+Xy8YK728Df28B32+HnrBszm6ZSeRjUlYRD2VSFE5ESQUmNiDjFYtHArkipYLNB77dg2uPwXJ/0/WO2H4VhsyD0fN7nN6gGbz2auS3j9eZ9V5OaXw/CbU3SCxZUrwhpVjh8Hp7/DGZ9X7jPdCNcLPB0L/B0p1y5m2RDUBHJN62pERERyUtJWlMj+eNigVOfgH8VUlJScNeaGpESSR+95tP27du588478ff3x9vbGw8PD/z9/Rk0aBDHjh3L0v/FF18kKCgIX19f3Nzc8PX1pVWrVnz88cf5uu+mTZvo3Lkz5cuXx83NjTp16vDyyy+TmppaWI9WYJcuXeLJJ5/k1ltvpXLlyhiGQVBQUI79Z86cyT333EP9+vVxcXHBMAz27ct5GsLvv/9Ojx498PPzw83NjapVq9KjRw+2bNlSFI8jOdi7d6/ZIYiIFC5XC/TvDP7pmwsfOKDpZyIllaaf5dPx48eJjIzkrrvuok6dOri6urJ3715WrlzJhg0bCA4Opl69eo7+O3fuxN/fn549e+Ln50d0dDRr1qxh9OjR7Nmzh48++ijPe65YsYKBAwfi7e3N4MGDqVatGj/99BPTpk3j4MGDrF69uigfOU/nzp1jzpw5VKxYkcaNG7Nz585c+8+dO5fDhw/ToEEDatSowfnzOU932LJlC3//+98pV64cAwYMoE6dOhw9epSlS5dy5513smHDBnr06FHYjyTZsN6sJW1FRAoqzQb/vNfsKESkEGj6WSGZPXs2zz77LGPHjmXGjBm59k1JSaFx48aEh4cTGxuLm5tbrv0bNGjA+fPn2bZtG61bt3a09+3bl9WrV7Ny5Uruv//+wniMAklISODMmTM0adIEAE9PTxo1apTjJ/sHDx6kYcOGuLu789BDD7F8+XL27t1Ly5Yts/TNeMbNmzfTvXt3R/tXX33Fo48+yoABA/j222+L5sEkk1OnTmVK2EXKFE0/K30sBrSuDzung2EAcOHCBapVq2ZuXCJSIJp+VkgCA9MrwFy6dCnPvu7u7vj5+ZGSkpJpo6LshIeHc/LkSVq2bJkpoQF48sknAZgzZ06u10hJSaFZs2Z4eHiwbdu2TMfeeOMNDMPgiSeeyDPunHh7ezsSGmc0a9bM6TnLcXFxQHpid6369esDaFFnMapSpYrZIYiIFJ6/yjhnJDQAxjXfi0jJoqSmgOLi4jh79iyHDx9m0aJFjB49GoAHHngg2/7nz5/n9OnT/Pnnnzz11FPs3LmToKAgypcvn+t9EhISAPDy8spyLOPckJCQXK/h7u7O0qVLcXV15eGHH3Zc8/fff2fy5Mk0b97cqWlwZrjrrruA9J/runXrOHLkCCtXrmT48OH4+vryyiuvmBxh2XH06E20p4SIyI2q6gMDb8vUdO7cOZOCEZEbpTU1BTR58mSmTJnieO3n58f06dPp169ftv2bNGniGHWwWCx07dqVhQsX5nmf+vXr4+Pjw/79+4mNjcXH5+pmZmvXrgUgKioqz+u0aNGC6dOnM2bMGIYMGcL8+fMZNGgQbm5uLF26NM8pcGZ55ZVXOH78OF9++SW9e/d2tDdq1Ijt27c7RshEREScZjHgmd7gfnP+2yci+aekpoAef/xxgoKCiI2NZfv27WzcuJELFy7k2H/BggUkJCRw6tQpVqxYQXJyMpcuXcpzM0OLxcLQoUP54IMPuOOOO3j77bepWbMmK1euZPbs2VgsFlJSUpyKefTo0fzwww8sX76cAwcOcObMGT788EOaNWuWr2cvThaLhZo1a9KiRQt69+5NnTp12LlzJwsWLKBXr1788ccf+Pn5mR1mmZAx5U9EpMSzGPDk37M038z/HopI7pTUFFCTJk0c60hGjRrF1q1b6dmzJwkJCcyaNStL/2unpU2YMIEePXrQvXt3Dh06RI0aNXK91/vvv09CQgILFy6kV69eAHh4eDBhwgSmT5+OzWZzOu4vv/ySwMBADh06xL333uuYNpeX8+fPk5SUlKktICDA6fsW1IgRI/j666/Zt29fpvt17dqVoUOH8sorr/DZZ58VeRySPuXS19fX7DBERG6MqwUe7gI1KmU5dObMGc0AECmhtKamkHTr1o2AgAAWL17sVP/hw4dz5coVPv300zz7uri48Nlnn3HhwgXWrl3LmjVruHDhAs8++yxxcXHUrl3b6Th/+eUXx3S1I0eOOD3Kc/fdd9OwYcNMX0UtOTmZL7/8kqCgoCwJ1JAhQ/D09OT//u//ijwOSefMNEcRkZtemg3GZl/GOWOauIiUPBqpKUTJyclO/0KMj48HIDo62unr+/r6ZlpX8sEHH2C3253epyUqKorHH38cHx8fBg8ezAcffMCTTz7J559/nue57733HpGRkU7HWhjOnj1LWlpatiNRNpsNm81GWlpascYkIiIlmMWA9g2hY6NsD3t6ehZzQCJSWJTU5NPx48eznXa1ZMkSTp8+TVBQkKPt8uXL2O12KlXKPMSdmprKxx9/DJBp75Xk5GT27NmDj49PnvN6z507x1tvvYWPjw8TJkxwKvYBAwYQFRXF119/zcCBAwkJCeGLL76gd+/eDBgwINdz77zzTqfuUZjq169P+fLl2bt3LyEhIbRq1cpxbPbs2aSkpGRqk6LVpk0bs0MQEbkxGWWcc6CpZyIll5KafBoyZAhRUVF07tyZ+vXrk5iYyO7du/npp5/w9PTMtPHm7t276dOnD3fccQeNGjWiSpUqnD17ljVr1nD+/HnuvvvuTNXSjhw5wi233ELLli0zbVw5f/583n33Xbp160bNmjU5efIky5YtIz4+nvnz5zs1/eytt95i8+bNDBs2jIEDBwLpiVjLli158sknue222/D39y/wz2XChAlcvnwZSN95PiIigjFjxgDQvn17Ro4c6ei7YsUKfvzxRwD27dsHpFeTy1iv8eGHHwLp0+6effZZpkyZQteuXRk4cCB16tQhODiY1atX4+Pjw5tvvlngmCV/9u/fT4sWLcwOQ0Sk4KpVhIduzfHwvn379AGOSAll2O12u9lBlCQzZ85k0aJFHD16lJiYGAzDwM/Pj86dOzNp0qRMIyxnz55l7Nix7Ny5k6ioKBITE/H29iYwMJDHHnuM559/Hovl6rKmffv2ERQUlCWp+eOPP3j22Wc5evQocXFx+Pj40KFDByZPnkzHjh3zjHnbtm3cfvvtBAQEsGfPnkwbX65du5a+ffvSsWNH/vzzzwL/XPz8/HJcc3H33Xezfv16x+tRo0Yxd+7cHK91/V/JDz/8kI8++ohjx46RnJxMhQoV6NixI9OnT9dITTEKDg7WP/ZSdt3xb9i8z+wo5Ea99Si82j/Hw/o9J1JyKakREaecOHGCBg0amB2GiDliEgj97mcaN25sdiRyI5rXhvJZN7POEB4enmdFUhG5OWn6mYg4pXr16maHIGKeCt4ktPSHNtkvMJfSQQmNSMmlks4i4pTQ0FCzQxAxlTagFRG5eSmpERERcUJGKX4REbn5KKkREafUrVvX7BBETFXce3WJiIjzlNSIiFOSkpLMDkFEREQkW0pqRMQpFy5cMDsEEVO1bt3a7BBERCQHSmpERESccPDgQbNDEBGRHKiks4g4JSgoyOwQREyVkpKS47G/L0kjWIOZN61utQ2W3u9idhgiUoSU1IiIU0JDQ2nWrJnZYYiYpkKFCtm2bw+z8+OpYg5G8mX5ERugpEakNNP0MxFxSnJystkhiJgqp40Z399lw9Uo5mBERCQTJTUi4hQfHx+zQxAxVXYb0IbH2/nmkJ00uwkBiYiIg5IaEXGKv7+/2SGI3HQ+2WPHZnYQIiKipEbMs3TpUgzDwDAMHnrooWz7nDp1CldXVwzDyLRQPSgoCE9Pz+IKVYBDhw6ZHYKIqa7fgDY5zc6sXTZsGqURETGdkhoxnZubGxs2bCAhISHLsZkzZ2K327FY9FdVRMx1/bqyJaF2orUnrYjITUHvFMV0Xbt2JT4+nk8//TTLsSVLltCpUyfc3NxMiEyuVbt2bbNDEDFVRESE43u73c47O2xYVCBAROSmoKRGTNemTRvq1avHggULMrWvX7+eM2fOMHz4cJMik2ulpaWZHYLITePPMAi+gKaeiYjcJJTUyE3h4YcfZteuXRw5csTR9tFHH1GhQgWGDRtmXmDiEB4ebnYIIqa6dl3f+ztVxllE5GaipEZuCs899xwuLi68//77AMTGxvLjjz/Sp08f3N3dTY5ORATHhy7nYu0sDVUZZxGRm4mSGrkp+Pv707lzZ5YtWwbAnDlzSExM5NlnnzU5MsnQsmVLs0MQMVVSUnpVgI/3qIhzyaNhNZHSztXsAEQyjBgxguHDh7Nq1SoWLFhAo0aN6Ny5s9lhyV+OHTtGkyZNzA5DxDQ+Pj4kpdn5YLcdawkapanoAVO7WXigkYG3G2wLs/PiZhu7L+R9bscaMKylhVtqGLTyAzcXA2N61vV1tX1gREuDPgEWGlUCqw32RcNbf9jYdPpm+GHdDDGISFHSSI3cNAYPHkylSpV4/fXX2bt3L48++qjZIck1EhMTzQ5BxFT+/v58fcjOpeS8+94sDGDtgy482sxg9m4bL2+xUc3bYPPDLgT65n1+7wALTwQZ2IHjV3Lud3+gwSudLBy9bOe1X21M+tOGjxtsHOjCsJYaJRGRoqekRm4abm5u9O3bl+DgYNzc3HjmmWfMDkmuUa5cObNDEDHVwYOHeHeHrUT9w9m/iUEXf4Nh6228+YedD4Pt/O0bK1Y7vNEl7yf5KNhGxVlWOi608uOpnEc7fj5tp+4cK4PX2vgw2M7MXXZu+8rKwWg7b95Wkn5iIlJSafqZ3FReeeUV3N3dCQwMpGrVqmaHI9e4fjd1kbJm92Vv9kaZHUX+9G9sEB5vZ3no1YQkKhG+PWznseYG7i6QYs35/AtZ90TO1oHorG0pVvj+hJ0XO1go7wZxqfkMXkQkH5TUyE2lWbNmzJkzx+wwJBsHDx6kTZs2ZochYpoVUbVxNShRVc/aVjPYFWHPsqJkW5idJ1tbaFwJ9hVholbDG+JT7SRomysRKWIaExYREcnD6Rg7359yK1EJDUDNchAWn7U9o61W+aJb79LQFx5sZLAs1K5NSkWkyGmkRkzTv39/7Hbn/qXLKKWaYe/evUURkuSiVq1aZocgYpoPg80v42wA7i7O9U3+a0qZl+vV76+VlHb1eFHwcoUl97mQmAb/2mr+z05ESj+N1EiZcvHiRZKTr5YuiouLIzY21vE6JSWF6OjMk8PDwsJyfR0eHp4pOSut90hKSioVz6F76B4Fuce6o6mY/da8Wx2DpOddnfpqUjn9nMQ08MgmEfJ0vXq8sFkM+PpeC82rQP/vbNmOFJnhZvx7pXvoHrpHzq/zy7A7+1G5iJRpwcHBWlMjZdYHu208u8mK3cRNHKt7Q68Gzt1/xRE7MSkQOtKFI5fs9FmeOSUb0dJgXi8XguanOb2mZlZPC8+0tWS7T8215t1tYVhLg8FrbXx96OZ4i2FgxzbOzewwRKQIafqZiIhIHoa2MHhps51Eq3lJTUQCfLE/f0lC8AU7t9c2MMi8/eQtNQ3iU+2EXirUEJna3cKIIAtjf7LeNAmNiJQNmn4mIk5p1qyZ2SGImMbH3eCBWhdxKWH7SC4NtVOjnMGDja8GXsULBjQxWH3Mnqmcc0DF9K+CGtfR4KWOFt7+08bMXTdbQlPC/sOJSL5ppEZEnHL69GkaNWpkdhgiphnRJJGvzpgdRf4sDbXzx3k7n/ey0LyKjahEGNPGgosB//kt85S0TQPTF980mHs106lbAYY0T08IOlRP//PVW9P/PBUDCw+kJy/9Ag2mdXch9KKdg9F2BjfLnET8eMru9J43ReNmS7JEpLApqRERp8TH3ySrfUVMcntzf3ofh/UnwFpC3iPb7NB7mZVp3S0819aClxtsD4dh66xOTT1rUNHgra6ZKw1kvN58xs7CA+kJUOtq6UlM48oGC/tkrUzwt2+sXEgoIT80ESmRVChARJxy+PBhmjRpYnYYIqYJDg4mqlIr7lpidh00yS8VChAp/bSmRvJl+/bt3Hnnnfj7++Pt7Y2Hhwf+/v4MGjSIY8eOOX2dlJQUxo8fT0BAAB4eHnh7exMQEMDrr7+epe+3335LmzZtHPcLDAxk6tSphfhU4oyGDRuaHYKI6XrWNWhcSSs0RERuNpp+Jvly/PhxIiMjueuuu6hTpw6urq7s3buXlStXsmHDBoKDg6lXr16u10hMTKRr166EhIRw55138vjjj5OWlkZoaCgnT57M1Pe9997jxRdfxM/PjyeeeIJy5cqxdu1aXnnlFU6fPs3s2bOL8GnlWvv27VNJZynTatWqhWEYvNDBwugfNVojInIz0fQzKRSzZ8/m2WefZezYscyYMSPXviNHjmT+/Pl8/fXXDBgwIMd+ycnJ+Pn5YbFYOHz4MNWrVwfAZrPRqVMngoOD2blzJ61bty7MR5EcaJ8aKesiIyPx8/MjPsVOzY+txKaYHZE4S9PPREo/TT+TQhEYGAjApUu5rzy9fPkyixcvpnPnzgwYMACbzZZlh9kMv/32G7GxsXTv3t2R0ABYLBYGDx6M1Wrl448/LryHkFzVqFHD7BBETHXu3DkAyrkbPNXaKHHlnUVESjMlNVIgcXFxnD17lsOHD7No0SJGjx4NwAMPPJDreatXryYpKYk2bdowcOBAvL29qVq1KhUqVGDYsGGkpFz96DMpKQkALy+vLNcpV64cADt37iysR5I8uLpqtqpIhjFtLNg0z0FE5KahdylSIJMnT2bKlCmO135+fkyfPp1+/frlet6+ffsAWLBgAa6urrz88stUrVqVr776ii+++IKwsDA2bNgAQLt27bBYLGzbtg2bzYbFcjUH37hxIwARERGF/GSSk7Nnz1K1alWzwxAxzbUb0NavaNA3ENYegzQlNyIiplNSIwXy+OOPExQURGxsLNu3b2fjxo1cuHAhz/NiY2OB9D1P/vzzTzp27AjAc889R6tWrfjhhx/4888/ufXWW6lRowa9e/dmzZo13HPPPUyYMIEKFSrwxRdfsGrVKiB93Y2ISHE4c+aMY6otwD/bWVh1VAUDRERuBkpqpECaNGni2LNk1KhRbN26lZ49e5KQkMCsWbNyPC9jKlmTJk0cCU2GQYMGsXfvXr7//ntuvfVWAL766iseffRR1q5dyw8//ACAj48P//3vf3nhhRfw9vYuiseTbDRt2tTsEERMFRcXl+l19zoGzavAwWjtV3/z0wIokdJOa2qkUHTr1o2AgAAWL16ca7+Mcs9VqlTJcqxOnToAXLx40dFWvnx5vvvuO86dO8eqVatYv349ERERjqSnfv36hfQEkpeMRdIiZZWnp2em1xnlnZXQlAT6ryRS2mmkRgpNcnJylk8yr/e3v/0NINupahl71Fxb6SxDjRo16Nu3r+P1kiVLAOjdu3cBo5X8ypg6KFJWXTv1LMOjTQ1e3AxXNBNWRMRUGqmRfDl+/Hi27UuWLOH06dOZpigdO3aMbdu2ERMT42hr1aoVzZo148iRI2zatMnRnpqaypdffonFYqF///65xhASEsKnn35KzZo1GTNmzA0+kTjLw8PD7BBETJVR6ORaXm4GY9qovLOIiNm0+abkS5cuXYiKiqJz587Ur1+fxMREdu/ezU8//YS7uztr167ljjvuAKBXr15s2LCBJUuWZEpUfvrpJ3r37o2rqyuPPvooVapUYdWqVRw8eJAnnniCuXPnOvq+/vrrrFu3js6dO+Pn58fBgwdZsWIFLi4urFmzhm7duhX7z6CsslqtuLi4mB2GiGly2oD2TIyd+nOtgFZu3IzSS29r802R0k7TzyRfHn74YRYtWsTq1auJiYnBMAz8/Pzo168fkyZNylTyNCc9evTghx9+4OWXX2bhwoWkpqZSu3ZtJk+ezPjx4zP1bdOmDcuXL+ezzz4jMTGRihUr8ve//51p06ZlOxVEis7evXuzfUMnUlbktAFtnQoGb3W1cCBanxHerBrYzwH1zA5DRIqQRmpExCk5fUotInKz0+8vkdJPa2pExCnVqlUzOwQRkQJRpUyR0k9JjYg45fpytiIiJUV8fLzZIYhIEVNSIyJOOX36tNkhiIgUSGRkpNkhiEgRU1IjIiIiIiIlmgoFiIhTEhIS8Pb2NjsMEZF8s9vtGIYKbouUZhqpERGnREREmB2CiEiBHDx40OwQRKSIaZ8aEXHKlStXzA5BpHT69SD8pjfdN6RSeRj19xwPp6SkFGMwImIGJTUi4hQ3N+3GLVIk7pmEPSEZw6LJEwVit4PVBrc1gZbZb7BZoUKFYg5KRIqb1tSIiIiYyXMgJKeZHUXJ5mqB4T1hzuhsD2tNoEjpp4+FRMQpwcHBZocgIpK9NBss+BkuxmZ7ODQ0tJgDEpHipqRGRERESr5UK3y60ewoRMQkSmqkUCxduhTDMHjhhRcK5Xr79u1j1KhR/Pzzz4VyPblxVatWNTsEEZGc2ezw/hpIs2Y5VLduXRMCEpHipKRGbkqHDh1i7ty5/Pnnn2aHIn8pX7682SGIiOTu/CVYtS1Lc3JysgnBiEhxUlIjIk45efKk2SGIiOTOxQLvrc7SrH22REo/JTVSbKKjoxkxYgSNGzfGx8cHNzc3atSowdChQ4mJiXH0mzJlCgMGDABgwoQJGIaBYRgEBQWZFbqIiJQEVhv8dgiCT5gdiYgUM+1TI8Xm2LFjLF++nB49evDAAw/g5ubGL7/8wsKFC9m/fz87d+4EoHfv3oSGhjJ//nzuu+8+unXrBkCtWrXMDL/MCwwMNDsEEZG8uVpg5lr47BlHkz4UEyn9lNRIsQkKCiIiIgIPD49M7SNHjuSzzz5j/fr19OrVi9atW9OnTx/mz59P586dGTdunEkRy7Wio6O1rkZEbn5pNli4Bf43BPwqAnDkyBGaNm1qcmAiUpQ0/UyKjZeXlyOhSUlJITw8nLNnz9K7d28AtmzZYmZ4kodLly6ZHYKIiHOsNpj7o+NlUlKSicGISHHQSI0Uq4kTJ7JgwQLOnDmD3W7PdExvmm9uLi4uZocgImap6A1Th8IDt4C3B2w7Ai9+AbuP531ux0AY1gNuaQSt6oGbKxgPFm28Nnv6FLSX+oGbKz4+PkV7PxExnZIaKTbjxo3jnXfeoV27djzxxBPUqVMHDw8PTp8+zb/+9S9sNpvZIUouNCddpIwyDFj7GrSuB9NWQVQMjOkFm9+E9i/B0bDcz+/dHp7oCSGn4HgENPEvnrgjrsDyP+Hhrvj7F9M9RcQ0Smqk2Cxbtgw/Pz+2bduW6VP/L7/8MktfwzCKMzRxQkhICK1atTI7DBEpbD+/CScvwPDZ2R/v3xm6NIX+02DZH+lt3/4OobPhjYdh8Izcr//RevjfCkhKgVlPFF9SY7HAu6vh4a4cOnSINm3aFM99RcQUWlMjxSYjkbl2RCYlJYWpU6dm6VuxYvrizujo6GyvtW3bNoKDgws/SMmRRtJEyqj+nSH8UvqoR4aomPTE5v5O4J7H56MXrqQnNMXNZkufJrfjaPHfW0SKnUZqpFBt3bqVMWPGZGn38/Ojd+/ezJo1i1tuuYX77ruPK1eusGrVqmzXatxyyy14enqycOFCypUrR6VKlahVqxYDBw50HK9atSqRkZFF/kySrnLlymaHICJmaNsAdh2H69ZBsu0IPPl3aFwL9p02J7a8uFrg/TXUfu8xsyMRkSKmpEYK1c6dOx37zVyrVq1anDx5ErvdzpIlS5g8eTIVK1akV69ePPPMM3Tu3DlTfx8fHz766CPefPNNJk+eTFpaGi1btnQkNVL8fH19zQ5BRMxQsxJsPZC1Peyv4i61Kt+8SU2aDb7+FfvLvaFqVbOjEZEipKRGCkX//v2zVDPLzqxZs5g1a1aW9uzOHTZsGMOGDcv2Os7cSwrX8ePHNSddpKRzdUmvZHYtN1fwcIMq11UIuxiXPjrj5Q7JaVmvlTGlzMu9aGItLDY7aR+ug48amx2JiBQhJTUiIiJlRZemsHlS9u2P3J65rf6TcCoSElPAI5u3C55/JTOJJqyXyQ+bHb+vt8MHtvTiASJSKun/bilTLl68SHJysuN1XFwcsbGxjtcpKSlZihOEhYXl+jo8PDzTyFFpvUeNGjVKxXPoHrrHzXqPYrHnJNz5euavPSdhw+6s7eGX/wr0UvoUtOtltJ2/WMRB3yAXC0ZTf6Kv2wvtZvhvrnvoHrpHzq/zy7BrHo+IOOHMmTPUqVPH7DBESh/PgdlP7youeZV0/nYc3N4Maj2RuVjAJ0/B4G5QeSikOBn/rCfgmd5Fv/nmdc7PeoxazxTvPUWkeGmkRkScklN5bREp5Zb+ATUqwYO3Xm2r4gMDboPVOzInNAHV079uJrUqceHW+mZHISJFTGtqRMQp2hBVpIxa+gf8cRg+fwaa14aoWBjTC1ws8J+vM/fd9Eb6nw2eutpW1w+GdE//vkNg+p+v9k//81QkLNxSdLFbDBh7L+Uq+uTdV0RKNCU1IuKU1q1bmx2CiJjBZoPeb8G0x+G5PunVzrYfhWGzIPR83uc3qAZvPZq5LeP15n1Fm9S4ucATd1KvvEfR3UNEbgpaUyMiTtm3bx8tW7Y0OwyR0sfsNTWllasFhveEOaMJDg5WSXqRUk5rasRp27dv584778Tf3x9vb288PDzw9/dn0KBBHDt2LFPfF198kaCgIHx9fXFzc8PX15dWrVrx8ccf5+uemzZtonPnzpQvXx43Nzfq1KnDyy+/TGpqamE+mjghLU1vukSkBEmzwXO9zY5CRIqJpp+J044fP05kZCR33XUXderUwdXVlb1797Jy5Uo2bNhAcHAw9erVA2Dnzp34+/vTs2dP/Pz8iI6OZs2aNYwePZo9e/bw0Ucf5Xm/FStWMHDgQLy9vRk8eDDVqlXjp59+Ytq0aRw8eJDVq1cX9SPLNXx9fc0OQUTEOS6W9IptLdP/TapVq5bJAYlIUdP0M7lhs2fP5tlnn2Xs2LHMmDEjx34pKSk0btyY8PBwYmNjcXNzy/W6DRo04Pz582zbti3Teo6+ffuyevVqVq5cyf33319YjyF5iIuLo3z58maHIVL6aPpZ0Vj1L+jbCYDIyEj8/PxMDkhEipKmn8kNCwxMr2Zz6bqNza7n7u6On58fKSkpmTZoyk54eDgnT56kZcuWWRaoP/nkkwDMmTPnBqKW/Dp69KjZIYiIOKdOFejT3vHy3LlzJgYjIsVB088k3+Li4rh8+TLx8fHs2LGD1157DYAHHnggS9/z58+TlpbG+fPnmT9/Pjt37iQoKCjPT/wTEhIA8PLyynIs49yQkJAbfRQRESltDAP+eR+4uJgdiYgUIyU1km+TJ09mypQpjtd+fn5Mnz6dfv36ZenbpEkT4uLiALBYLHTt2pWFCxfmeY/69evj4+PD/v37iY2Nxcfn6h4Da9euBSAqKuoGn0Tyo379+maHICKSNw83GNEzU1OzZs1MCkZEiouSGsm3xx9/nKCgIGJjY9m+fTsbN27kwoUL2fZdsGABCQkJnDp1ihUrVpCcnMylS5eoW7durvewWCwMHTqUDz74gDvuuIO3336bmjVrsnLlSmbPno3FYiElJaUoHk9yEB8fr2IBInJzc7HA8B7gWy5T85kzZxxTpUWkdFKhALlhW7dupWfPnjz11FPMmjUr1749evRg165dHDp0iBo1auTa12q18o9//IOFCxc6Sjh7eHgwYcIEpk+fjs1mc4wCSdHTPg8iRUSFAgrXwZnQtHamJv3+Ein9VChAbli3bt0ICAhg8eLFefYdPnw4V65c4dNPP82zr4uLC5999hkXLlxg7dq1rFmzhgsXLvDss88SFxdH7dq187yGiIiUES4WuLNVloQGwNPT04SARKQ4afqZFIrk5GSnRk3i4+MBiI6Odvravr6+9O59dQO1Dz74ALvdTo8ePfIfqBTY9VXoRERuKlYbPH9ftoc09Uyk9NNIjTjt+PHj2bYvWbKE06dP07RpUwAuX76cbXnn1NRUPv74YwC6d+/uaE9OTmbbtm0cPHgwzxjOnTvHW2+9hY+PDxMmTCjIY0gBOfPfR0TENA2qQa+22R7at29fMQcjIsVNIzXitCFDhhAVFUXnzp2pX78+iYmJ7N69m59++glPT0/Hxpu7d++mT58+3HHHHTRq1IgqVapw9uxZ1qxZw/nz57n77rszVUo7cuQIt9xyCy1btmTv3r2O9vnz5/Puu+/SrVs3atasycmTJ1m2bBnx8fHMnz9f08+KmQoziMhNyyB9lMaiz2pFyiolNeK0hx9+mEWLFrF69WpiYmIwDAM/Pz/69evHpEmTHCUzGzVqxD333MPOnTvZsmULiYmJeHt7ExgYyAsvvMDzzz/v1P2aNGmCu7s7CxcuJC4uDh8fHzp06MDkyZPp2LFjUT6qZKNChQpmhyBSOrWqD9u1ue0N8faEx+/I8XBehWlEpORT9TMRcUpCQgLe3t5mhyFS+qSksvfPHQQFBZkdScnlYoEK+v0kUpZppEZEnBIaGqqSqCJFwd0NawUvqFTe7EhEREosTT4VERExWf369c0OQUSkRFNSIyJOqVu3rtkhiJRaGeXuRUSkYJTUiIhTkpOTzQ5BpNSKjIw0OwQRkRJNSY2IOCUiIsLsEERERESypaRGRETEZK1btzY7BBGREk0lnUXEKVarFRcXF7PDECmVDhw4QPPmzbM9lpxmx6Z/qfPF3QVcLIbZYYhIMVJJZxFxypEjR2jatKnZYYiUSikpKdm2J6ba8f/YyiUtacuXrv7wyyN6iyNSlmj6mYg4JSkpyewQREqtChUqZNu++JBdCU0BbA/X0JZIWaOkRkSc4uPjY3YIIqVWjRo1srTZ7Xbe3WFDk6gKQDPrRcocJTUi4hR/f3+zQxAptUJDQ7O0bT0LB6JBb88LwFAqKFLWKKkREaccOnTI7BBEypT3dlpx1XvzAtGPTaTsUVJTAixduhTDMHjhhRdy7GMYBrfcckuOx4cMGYJhGNSsWbMoQnRaeHg47u7uGIbB1KlTc+0bFRXF6NGjCQgIwNPTEw8PD2rXrs3gwYM5ceJEMUUsIlL06tatm+n1ySt2vjsKaRqmKRAVdhUpe5TUlAEpKSmsXr2a6tWrEx4ezrJly0yLZebMmaSlpeHn58fChQtz7Ldjxw6aN2/OJ598gr+/P+PGjWPChAkEBQXxzTff0KpVK9auXVuMkUvt2rXNDkGk1EpOzlwN4MNgG6pILCLiPCU1ZcD8+fO5cuUKM2fOpEKFCsyZM8e0WL766iuCgoIYOXIk+/btY8+ePVn6xMTE0LdvXy5evMi8efP45ZdfeOutt/jPf/7DunXrWLduHVarlUceeUQjNsXIarWaHYJIqRUREeH4Pj7Fzsd77Fg12FBghtbUiJQ5SmrKgM8++4xq1arRv39/7rnnHjZv3kxUVJRT5/7000+4ubnRtm1bbDaboz01NZUWLVrg4eHBb7/95tS1Nm3axMmTJxk8eDBPP/00FouF999/P0u/KVOmEBYWxqBBgxg+fHiW43fddRcvvvgisbGxvPrqq07dW25cWFiY2SGIlAkLD9qJy37bGnGSpp+JlD1KakqQhIQEzp49m+1XTo4fP8727dvp168fFouFMWPGkJKSwgcffODUPXv06MG4ceMIDg5m7NixjvbRo0dz4MABxo8fT5cuXZy61gcffICHhwejRo2idu3a3HrrraxatSrLCMDq1asBePHFF3O81ksvvYSLiwsbN2506t4iIjezoKAg4GoZZxERyR/Dro8zbnpLly5lwIABefbr1KkT//d//5epbezYscyaNYvg4GBatWoFQIMGDXBzc8u2hGhOOnfuzLZt21i9ejUJCQkMHDiQLl268Msvvzh1flxcHNWrV+f2229n/fr1AMyZM4cnn3yS+fPn8/jjjzv6li9fHqvVSmJiYq7XrFevHqdPn+bixYtUqlTJ6WeRgklNTcXNzc3sMERKpUOHDtG0aVM2nbJx5xIlNTfKw2Ij6QV3s8MQkWLkanYA4rz77ruPRx55JNtjjz76aLbtS5YsoXnz5o6EBmDgwIFMnTqV33//ndtuu82pey9btoygoCCGDRtGWloaVapUYenSpU7HPmfOHBISEhg5cqSj7fHHH2fcuHHMmzcvU1KTmJiIr69vntcsV64ckF4lTUlN0Ttx4gSNGzc2OwyRUikpKQmAGTttuBhoPc0NMiyaiCJS1iipKUECAwPzldSsXLmSsLAw7r//fnbv3u1o7969O9OmTWPmzJlOJzW1atVi9uzZjvusWrWK6tWrOx37F198gY+PDw0bNswUS6dOndiyZQtnz551VNfy8vLKc5QGID4+HoCqVas6HYcUXEJCgtkhiJRaPj4+HLtsZ+1xczbbrOgBU7tZeKCRgbcbbAuz8+JmG7svOHd+08rw3h0WuvobpFhh7XE7L2y2EXXdr/KGvvDfbhZ61jXwcIFdETDxNxubzxTuU9ttGu0SKWuU1JRiH3/8sePPjO+v9f3335OcnIyHh4dT11u+fLnj+23bttG3b1+nzgsJCWHv3r3Y7Xbat2+fbZ+ZM2c69q2pX78++/fvZ/fu3bRt2zbb/jExMZw7dw4/Pz+N0hSTjJExESl8/v7+jP8jvYxzcY/SGMDaB11o7QfTtqcnImPaWNj8sAvtv7Ry9HLu5/uXh62DXLiSDBN+sVHeHcZ1sBDk50KnhVZS/8ovavvAH4+6YLWn3yc+FYa3tPBDfws9l1j5JefloSIieVJSU0pFR0ezefNm2rZtm2nKV4bdu3czb9485s2bx5gxY/K83ty5c1m6dCm9evXi9OnTTJ06lV69etG1a9c8z505cyZ2u50333yTypUrZzk+efJkvvnmG0dS06dPH/bv3897773HggULsr3mO++8g9VqpWfPnnneXwpHvXr1zA5BpNTatS+UuSHNTZl21r+JQRd/g/7fWVkWmh7At4ethI504Y0uFgavzX3UY8KtFsq5QfsvrZyJTW/bFmZj40AXhrU0mBuSfs1/dbLg6wEt51sJvZTeb26IlUMjXHjvby50WFiIZeNV0lmkzFFSU0p98MEHJCcn849//IPRo0dnOR4TE8OXX37J/Pnz80xqjhw5wgsvvIC/vz/ffPMN58+fp0OHDjzyyCMcOHAAHx+fHM+1Wq2sXLmSunXrMnHixGz7hISEMGfOHDZs2MDdd9/N+PHjWbBgAYsXL+auu+5iyJAhmfpv2rSJ6dOn4+Pjw9tvv+3ET0MKw4EDB2jTpo3ZYYiUSmvCfElINefe/RsbhMfbWR56NaOKSoRvD9t5rLmBuwuk5JJvPNTIYM1xuyOhAdh02s7hi3YGNrma1Nxe22D3BRwJDUBiGnx3zM4zbS0E+pLnqJCzlNKIlD1aSVdKLVq0CHd3dwYPHpzt8QoVKtCpUyd27tzJkSNHcryO1WrlwQcfJDk5ma+//poKFSrQtGlTpk+fztmzZ3O8fobFixcTHR1Nr169cuwzdOhQAD788EMAfH19WbFiBb6+vjz++ON0796diRMn8sYbb9CnTx969eqFYRgsWrSIgICAvH4UIiI3NZvdzlfnq5uylgagbTWDXRH2LPffFmannJtB41xm+NYqD9XLGewIzxr9tjA7batdTS88XNKTmOtlJHPtqxdeKqLCriJlj5KaUuj3338nNDSUjh07UqFChRz7Pfjgg9hstmw3wMzw9NNPs2/fPv71r39lmmr21FNP8cADD7B69WpHMpKduXPnAmQZbblWly5dqFmzJj/++COxsekf9d16660cPHiQUaNGcfr0aaZNm8bbb79NcHAwAwYMICQkhPvuuy/Ha0rhq1WrltkhiJRKP560czLWxbT71ywHYfFZ2zPaapXPOdmoWS5z3+vPr+KVPtIDcPiinVZ+UP66yvBd/dOv75/zoL+ISJ40/awE6N+/f56fOl17/LbbbnPqU6rnn3+e559/Ptc+ORUZgMyFA3KydevWPPsAnD9/Pkubn59fjveW4mdojrpIkXhvpw0X7FgLYdKUAY4kIi/Jf00p83K9+v21ktKuHs9JxrFsz7/m+ilW+GiPnb6BFr65z8Krv6YXChjTxkKHGnnfJ7/0+0qk7NFIjZQpFy9eJDk52fE6Li7OMToEkJKSQnR0dKZzwsLCcn0dHh6eKYksrfc4ffp0qXgO3UP3uJnuYbfbCYmkUBIagG51DJKed3Xqq8lfdVsS09Knhl3P0/Xq8ZxkHMv2fJfMfdafsPPMJivdahvsHupK6EhX+gQYvPpreiGCuJQCPHAO7Hb7TfvfXPfQPXQP5+6RX4ZdE09FxAnBwcEqFCBSBCb/aWPib1Zs9htPbKp7Q68Gzl1nxRE7MSkQOtKFI5fs9FmeucrZiJYG83q5EDQ/jX1R2V+jVnk495QrL2+xMm175rcTC+6x0DvAoOoHmYdxvN2gVVVIsUHwBRgZZDDn7y70XmZl3YnCeUviYbGR9IJ7oVxLREoGTT8TEac0a9bM7BBESqVRrQz+8xsUxnaREQnwxf78JQbBF+zcXtvAIPPGn7fUNIhPtWeqVna983FwIcFOhxrXnw2dahoEX8gaS0Iq/HnNB7J31jNISLXz2zl9xioiBafpZyLilDNnzpgdgkipVNXboE+Ny7iatAxkaaidGuUMHmx8NYAqXjCgicHqY/ZM5ZwDKqZ/XWtZqJ17AwxqX7PQv0ddgyaVDZaE5p6odK4FDzYymLc3fdSosBgWvb0RKWs0UiMiTomLizM7BJFSa3iTBFaF5VI7uQgtDbXzx3k7n/ey0LyKjajE9AX8Lgb857fM40ebBqYvlGkw92qmM/n/bAxo7MLPA114f5eN8m7wUkcLIZF2Pt93NampWwG+vc+F747aCI+HFlUNnmptEBIJE34pjHGqq+y2wr2eiNz8lNSIiFM8PT3NDkGk1LqnVU26hKZPy7IW8ywsmx16L7MyrbuF59pa8HKD7eEwbJ0116lnGc7GQvdvrLz7Nwv/7WYhxQprj9t5cbMt0yhPTDKExaVvtFnZE87Fwcxddt7+00acSRuPikjpoUIBIuKUtLQ0XF31OYhIUQgODua4dyse+k4jDIXBw2In6QW3vDuKSKmhSacFsH37du688078/f3x9vbGw8MDf39/Bg0axLFjx7I9Z+3atdx6661UrlwZd3d3qlevzr333suePXucvu/Ro0f5+9//ToUKFXB3dycwMJDZs2cX1mPlqSDPfb2lS5diGEauX999951T/W+55ZaielTJxr59+8wOQaRU6xtoUKuc2VGUDtqmRqTs0ceuBXD8+HEiIyO56667qFOnDq6uruzdu5eVK1eyYcMGgoODqVevnqP/l19+yeOPP0716tV57LHH8PPzY9++faxcuZLNmzeze/duGjVqlOs9z58/z2233UZMTAyPPfYYderUYenSpTz77LPExMQwYcKEon7sfD93dtq3b8+0adOytCclJfHvf/8bHx8f7r777izH77vvPrp165aprX79+jf0PCIiN4tatWrhajEY297C+K22QqmEVpZpEopI2aPpZ4Vo9uzZPPvss4wdO5YZM2Y42tu3b8+ePXs4efIktWvXdrRPmjSJf//737z00ktMnTo112s/9thjLFq0iE8//ZSRI0cCkJqaSvPmzQkLC+PMmTNUqmTOItOcnjs/3nvvPV544QUeffRRFi1a5GhfunQpAwYMYPLkyYwfP76QIpaCCA8Pp0aNGmaHIVIqRUZG4ufnx8VEO7U+tpJszfscyZn2qREpezT9rBAFBgYCcOlS5pWVcXFxuLm5Ub169UztderUAaB8+fJ5Xvv777+nevXqjoQGwM3NjX/84x/Ex8fz5Zdf5nr+Tz/9hJubG23btsV2TVWY1NRUWrRogYeHB7/99luecWQnp+fOj/nz5wPwz3/+M8c+ly9fVgUuE7m5aX66SFE5d+4cAJW9DIa2MEwr71xaGJp/JlLmKKm5AXFxcZw9e5bDhw+zaNEiRo8eDcADDzyQqV/37t1JSkrinnvuYcuWLRw5coSFCxfy6quv4u/vz9NPP53rfY4dO8alS5do2bJllmN33nknAP/3f/+X6zV69OjBuHHjCA4OZuzYsY720aNHc+DAAcaPH0+XLl0K9bmdFRISwt69e2nevDkdO3bMts+kSZOoVKkSPj4+1KxZk1deeSVTciZFT/vUiBSP59paSNMcihuiOSgiZY/W1NyAyZMnM2XKFMdrPz8/pk+fTr9+/TL1mzFjBpGRkaxZs4ZNmzY52jt27Mj3339PlSpVcr3PiRMnAKhZs2aWYxmjJGFhYVmOXW/KlCls3ryZDz/8kHvuuYeEhAQ+++wzunbtyuuvv57n+RmcfW5nvf/++9jtdoYOHZrlmLu7O7feeit33303derU4cyZMyxatIipU6cSEhLCunXrCnRPEZGbSbNmzRzft/Qz6F4bfj1X/OWdSw/94ETKGiU1N+Dxxx8nKCiI2NhYtm/fzsaNG7lw4UKWfq6urtSqVYt27dpx3333UbVqVbZu3cqSJUu45557+PXXX/Hw8MjxPhlTrrLrU65ceqmcpKQkp2JetmwZQUFBDBs2jLS0NKpUqcLSpUudOjeDs8/tjNTUVFauXImXl5djxOdaffv2pW/fvpnaJk6cSKdOnVi/fj3fffddluNSNJo0aWJ2CCKl1pkzZxwfUgE838HClrMajRYRcZaSmhvQpEkTxxu9UaNGsXXrVnr27ElCQgKzZs1y9OvduzchISEcO3YMHx8fAJ566ikaNWrEm2++ydSpU5k4cWKO98lYc5OcnJzlWHx8POD8xoi1atVi9uzZPProowCsWrUqy1qfvDj73M5YtGgRFy9epG/fvlSoUMGpc1xcXJgwYQL9+/dn+fLlSmqKSVhYGAEBAWaHIVIqXb9e8N4Agzo+cCbWpIBKOMOi2fUiZY3+ry9E3bp1IyAggMWLFzvaDh48yKZNm+jWrZsjockwatQoALZu3ZrrdRs0aABkP8Xs6NGjQPZT03KyfPlyx/fbtm1z+rycZPfczvr0008BeOaZZ/J1XtOmTQGIjo7O9z2lYGJiYswOQaTUuv6DKReLwT/bW9By94Kxa82lSJmjpKaQJScnZ/rE7fjx4wBYrVnrc6akpOR47FoNGzakUqVK2W5+uHHjRgCnN6KcO3cuS5cupVevXjRv3pypU6fy66+/OnVubq5/bmecOnWKP//8k/r163PXXXfl69yQkBAAqlWrlq/zpOBymyIpIjfm2qlnGUa0NPBwMSEYEZESSElNAWQkKtdbsmQJp0+fdowiALRt2xaLxcLmzZuJiIjI1D9jT5d27do52mJiYti2bRvHjh3L1Peee+4hIiKCefPmOdpSU1OZO3cu3t7eDB48OM+4jxw5wgsvvIC/vz/ffPMNy5Ytw93dnUceeYTY2LznOOTnuSG9atu2bdty/IT//fffx2q1MmjQoBzvmVHm9FoJCQlMmjQJgAEDBuQZtxQOrakRKTrZfWjl62kwrKXKOxeMfmgiZY023yyALl26EBUVRefOnalfvz6JiYns3r2bn376CXd3d9auXcsdd9zh6P/oo4/y1Vdf4efnx8CBA6lSpQq///47mzZtolq1auzbt4+qVasCVzebvPvuu1m/fr3jGufOnaNNmzbExsYyZMgQateuzbJly9i7dy+TJk3itddeyzVmq9VKmzZtOHz4MD/99BNdu3YF4OOPP2b06NHcd999fPfdd4X63L169WLDhg0sWbKE/v37Z7mev78/kZGRnD59OsdNHQMDA/Hz86N169b4+/tz7tw5Vq1aRXh4OP3792fJkiW5xiyFJzg4mDZt2pgdhkiplNP/Xwej7TT/3IqrAdp6xTlWO3i72okdq721RMoSFQoogIcffphFixaxevVqYmJiMAwDPz8/+vXrx6RJkzKV5gRYuHAhTZs2ZeHChXz66aekpqZSuXJl+vXrx4wZMxwJTW78/f35/fffGTNmDF9//TVJSUnUqVOH999/n+eeey7P859++mn27dvHxIkTHQkNpBcs+OGHH1ixYgUffvghY8aMKbTnzs3q1as5f/48PXv2zHWX+j59+rB+/XoWL15MfHw8Hh4eBAYG8vLLL/P88887fT8RkZtZTr8Hm1Ux+KCnhcjEYg6ohKuVdAxoZHYYIlKMNFIjIk45f/48tWrVMjsMEZE8aWRZpOzRmhoRcYqXl5fZIYiIOKV+/fpmhyAixUxJjYg45dSpU2aHICLilIw93ESk7FBSIyIiIqVKZGSk2SGISDFTUiMiTmnUSItuRURE5OakQgEi4pSTJ09qnrqIlAh2ux1DNbBFyhSN1IiIUy5fvmx2CCIiTjl48KDZIYhIMdM+NSLiFDc3bWQnIiVDSkpKjsd2hNv5eI+tGKMpmVpUMfhne0MjXlJiKKkREae0aNHC7BBERJxSoUKFHI+99aeV1cc0VSU3dsBqtzOgiQu1fcyORsQ5SmpExCl79uyhdevWZochIpKnGjVqZNt+8oqd746mv2nXWI1I6aIPKkTEKaopIiIlRWhoaLbtHwbbsGg2lUippKRGRJxStWpVs0MQESmw+BQ7H++xY9XnMyKlkpIaKZClS5diGAYvvPCC2aFIMfHx0cRqESkZ6tatm6Vt4UE7cTnXDxCREk5JjYg45cSJE2aHICLilOTk5Eyv7XY77+7QKhqR0kxJjYiIiJQqERERmV7/dNpO6KX0AgEiUjopqZEitWLFCtq2bYu3tzfu7u4EBATw1ltvZekXFBSEn58fR48epUePHpQrVw4PDw/at2/Pjh07TIhcrtewYUOzQxARKZD3dtpwUYEAkVJNSY0UmXnz5tG/f39OnTrF0KFDGTt2LK6urkycOJHhw4dn6Z+cnEzXrl2xWCy8+OKLDBgwgJCQEO6//35SU1NNeAK51qVLl8wOQUTEKUFBQY7vj1228/1xVCBApJRTUiNFIjU1lVdeeQUPDw+2bdvGxx9/zLRp09izZw9Nmzbliy++YNeuXZnOiY2NZdCgQWzcuJE333yThQsX8swzz3D+/Hm++eYbk55EMly8eNHsEEREnHLkyBHH97N3q4yzSFmgpEaKxKZNm4iOjqZv374EBgY62r28vHj++eex2+0sXLgw0zmGYWSZmtanTx8ADhw4UPRBS65cXFzMDkFExClJSUkAxKXYmRuiMs4iZYGr2QFI6ZSx8VnLli2zHOvYsSOQtZpWpUqVKF++fKa2jF2ho6OjiyJMyYdrp3OIiNzMMkrQf7HfTkIxz16u6AFTu1l4oJGBtxtsC7Pz4mYbuy84d37TyvDeHRa6+hukWGHtcTsvbLYRlZi5X0Nf+G83Cz3rGni4wK4ImPibjc1nlMFJ2aSRGrlpWCw5/3XUbvbm27t3r9khiIg4xd/fH5vdznvFXMbZANY+6MKjzQxm77bx8hYb1bwNNj/sQqBv3uf7l4etg1wI9DWY8IuN6Tts9Akw+HGAC27X/BNZ2wf+eNSFrv4G07bbGP+LjfLu8EN/C7fXLqqnE7m5aaRGikSTJk0A2LdvX5ZjGdXMGjRoUKwxyY2xWq1mhyAi4pRDhw4RUbEVx64U7337NzHo4m/Q/zsry0LTP4z79rCV0JEuvNHFwuC1uSdZE261UM4N2n9p5Uxsetu2MBsbB7owrKXB3JD0a/6rkwVfD2g530roXzVc5oZYOTTChff+5kKHhfp9LWWPRmqkSPTo0YMqVaqwevVqjh8/7mhPTk5mxowZGIbBY489VqBrJycns23bNg4ePFhY4YoTKleubHYIIiJOM6OMc//GBuHxdpaHXp1dEJUI3x62c3+ggXseSxMfamSw5rjdkdAAbDpt5/BFOwObXH2Y22sb7L6AI6EBSEyD747ZaV/DcGpUSKS00UiN3JCtW7cyZsyYLO1+fn7873//Y9SoUXTs2JGBAwfi4+PDqlWrCA0NZdiwYbRr165A9zxy5Ai33HILLVu21JSoYlSpUiWzQxARcUqyT102nCz++7atZrArwp5lk89tYXaebG2hcSXYF5X9ubXKQ/VyBjvCs47mbAuz0zvgalLj4QKXkrJeI2P9UPvqBkcva9q2lC1KauSG7Ny5k507d2Zpr1WrFufOncPX15e33nqL+fPnk5aWRu3atZk0aRKvvfaaCdHKjTh27Bht2rQxOwwRkTzNPeCGi1H8e9PULAdbz2ZtD4tP/7NWeYN9UdkHVbNc5r7Xn1/FK32kJ8UKhy/aub22QXk3iLumEEJX//TEx9/nRp5CpGRSUiMF0r9/f6cW7z/00EM89NBDefbLacSlZcuWWe6TXZuIiAhATLKdhUc8bjihMSDP6WIZkv9awuLlevX7ayWlXT2ek4xj2Z5/zfVTrPDRHjt9Ay18c5+FV3+1EZ8KY9pY6FAj7/uIlFb6ay8iTlFhBxEpCZaG2km23fiS4W510quWOaPpZ2kcvpi+rsUjm1M8/3q3lZiW8zUyjmV7vkvmPutP2Hlmk5X/3m5h99D0ix+5ZOfVX21M6+5CXIpTYYuUKioUIGXKxYsXSU5OdryOi4sjNvbqisyUlJQse+KEhYXl+jo8PDzTyFFpvUdUVFSpeA7dQ/fQPUr3PW6vbWBkWdWSf4ei7QxbZ3XqKyzurzjjr04ju1ZG2/m4nOPKmHaW0/nRiXZSrhnF+WC3neofWem8KI32X6bR9DMrV/76MV9bQKAw3Oz/zXWP0nmP/DLsmscjIk4IDg7WmhoRKRG6L4jht0jvYl9T8+19Fm6vbVDrI2umtOqTuywMbm5QebY1U2JyvYgxLmw+Y+fh1ZmLBRwa4cLZWDt3Lsm9JPQ391m4N8Cg5kdWYgphtObMky7U9inmEnIiBaSRGhFximHoHzYRKRkG+V8o9oQG0qe+1Shn8GDjq78vq3jBgCYGq49lHmkJqJj+da1loXbuDTCofc1C/x51DZpUNlgSmvsDda4FDzYymLfXXigJjUhJo5EaERERKVVCQ4/QZ2sDjl4u3vtaDPj1ERdaVoVp221EJaYv4K/rAx0XWjNNCzvxj/SFMg3mXs10avvA7iEuXE6G93fZKO8GL3W0cDYu/fyMpKhuBfj2Phe+O2ojPB5aVDV4qrXBoYvQ/WtrpopoN0IjNVKSaKRGRJyyf/9+s0MQEXFK/fr1eL69heJ+O26zQ+9lVr45ZOe5thamdbcQlQg9vrU6tc7lbCx0/8bKsct2/tvNwsudLHx/ws5dSzJPW4tJhrA4O8+0tfDhnRb6BRrM3GUv1IRGpKTRSI2IOEVrakSkpAgODiaweWtqfqQ3+TdCIzVSkmikRgrF+vXrGTBgAM2aNcPT0xPDMJgyZUq2fS9dusSTTz7JrbfeSuXKlTEMg6CgoByvbbPZePfdd2natCk+Pj54enpSp04dnnrqKaKictiaWQqdr6+v2SGIiDitvLvBP1oZuOg9uUiZoKRGCsXy5ctZtmwZcXFxee5ncu7cOebMmcOhQ4cIDAzEYsn9r+HIkSN58cUX8fDw4JlnnuGVV14hMDCQTz75hC5dumCz5V4NRgqHn5+f2SGIiDilVq1aADzT1oJN81FEygQlNVIoxo8fz8WLFzlz5gzPPPNMrn0DAgI4dOgQly9fZtu2bbi5ueXYNyUlhcWLFxMQEMCuXbuYMmUKb7zxBj///DM9e/YkNDSULVu2FPbjSDaOHDlidggiIk7JqNYY4GvQJwCN1oiUAUpqpFA0aNDA6elJ3t7eNGnSxKm+ycnJpKamUqVKFVxcMm+zXKNGDQAqVqyY3akiIlJGnTt3zvH98x0sppR3FpHi5Wp2ACK58fHxoUWLFuzYsYOxY8cybNgw3NzcWL16NUuXLuXOO++kXbt2ZodZJtSrV8/sEERE8u2OOgZNKkHoJVBuI1J6aaRGbnrLly+ndevWzJw5k3bt2hEUFMSrr77K4MGD2bBhg9nhlRmJiYlmhyAi4pRmzZo5vjcMgxc66O2OSGmnkRq56Xl5eVG7dm2qVavGPffcA6QnOp999hmenp588MEHJkdYNly4cMGx+FZE5GZ25swZAgMDHa8fa27w0haISTExKBEpUkpqxGmpqamcOXMmU5uPj0+RVsWKiYmhY8eONG3alJ9//tnR/s9//pM77riDjz76iGHDhtGxY8cii0FEREqWuLi4TK+93Qyeam3wzg671teIlFIajxWnHT58mIYNG2b6GjJkSJHe8+OPPyY8PJwBAwZkOTZgwADsdrumoBWTVq1amR2CiIhTPD09s7SNaaPyziKlmUZqxGn169dn8eLFWdqKUsbIUFpaWpZjGW3ZHZPCd/jw4Uzz1EVEblbXTj3LUK+iwf2BsOYYpCm5ESl1lNSI08qXL88jjzxSrPcMCgoCYOHChTz33HOZjn355ZcA3H777cUaU1mVnJxsdggiIk7Zt28fbdq0ydL+z/YurDxqLf6ARKTIKamRQrF//37Hgv1Dhw4BsHbtWsdIy9NPP02LFi0c/SdMmMDly5cBsFqtREREMGbMGADat2/PyJEjARg+fDjTp09n+/bttGjRgj59+jiufeDAAbp160bPnj2L5RnLugoVKpgdgojIDelWG1pUgQPRKu8sUtoYdrtd/1/LDVu6dGm2614yLFmyhP79+zte+/n5ERUVlW3fu+++m/Xr1zteR0dH8+KLL/Ljjz9y4cIFAGrWrMmDDz7I1KlTcXd3L6SnkNwkJibi5eVldhgiInkKDw93bNB8vS/22fjnzzb07id3jX1tbH3UDU9Xw+xQRJyipEZEnBIcHJztdA4RESl99DtfShpVPxMRERERkRJNSY2IOKVOnTpmhyAiIsWkqKubihQ2JTUi4pTU1FSzQxARkWISHx9vdggi+aKkRkScEh4ebnYIIiJSTCIjI80OQSRflNSIiIiIiEiJpupnIuKUtLQ0XF21tZWISFlgt9sxDJVzlpJD71BExClHjx6ladOmZochIiLF4ODBgzRv3jzbY2k2OweiizmgEqiRL3i5KTEsLkpqRMQpSUlJZocgIiLFJCUlJcdjy0LtDFpjK8ZoSqYHGxksu9/F7DDKDK2pERGnlC9f3uwQRESkmFSoUCHHY+/usGHRAESeDkepamhxUlIjIk7RPjUiImVHjRo1sm3fEW5nWzjYtCI7T2naCqFYKakREaccPHjQ7BBERKSYhIaGZtv+/k4brhqlcYqrm1Z5FCclNSIiIiKSp/B4O18fspOmURqnGIbeZhcn/bTFFOHh4bi7u2MYBlOnTs22j5+fH4ZhOL5cXV2pWrUqffr00aiBCfz9/c0OQUREikndunWztM3ZY0flAZyXmkuxBSl8GhcTU8ycOZO0tDT8/PxYuHAhL7/8crb9KleuzPjx4wGIjY3lt99+Y926dfz+++/s27dPb7SLkba0EhEpO5KTkzO9TrHambXbprU0ctPSSI2Y4quvviIoKIiRI0eyb98+9uzZk22/cuXKMW7cOMaNG8cbb7zBxo0bGThwIJcvX+bdd98t5qjLtvPnz5sdgoiIFJOIiIhMr5ccthOVaFIwJZSHh4fZIZQpSmqk2G3atImTJ08yePBgnn76aSwWC++//77T5999991A+maQIiIiUvRUxjn/0tLSzA6hTNH0Myl2H3zwAR4eHowaNQpfX19uvfVWVq1ahdVqxcUl702qMiqyVKpUqahDlWvktLO0iIiUPkFBQY7v/y/Mzq4LJgZTQlmtVrNDKFM0UiPFKi4ujg0bNvC3v/0NX19fAIYOHcrFixdZuHBhlv5Wq5WzZ89y9uxZ9u3bx/Tp05k1axYWi4URI0YUc/Rl26lTp8wOQUREismRI0cc38/YoTLOBWEY+qEVJyU1UqzmzJlDQkICI0eOdLQ9/vjj+Pj4MG/evCz9z58/T506dahTpw5BQUG89NJLjr7dunUrztDLvPj4eLNDEBGRYpKUlATA+Tg7S0JVxrkgPD09zQ6hTNH0MylWX3zxBT4+PjRs2JDdu3c72jt16sSWLVs4e/YstWvXdrT7+fk51tucO3eOefPmceLECdzc3Io99rLO29vb7BBERKSY+Pj4APBxcPEXca7oAVO7WXigkYG3G2wLs/PiZhu7nZwC17QyvHeHha7+BilWWHvczgubbVkKHTT0hf92s9CzroGHC+yKgIm/2dh8pnAyuMTERMCnUK4leTPsqtMqxSQkJIQ2bdrkWhr4pZdecuxb4+fnh5eXF6dPn3Ycj4uLIygoiAsXLhASEkLDhg2LPG5Jl5qaqmRSRKSMSEpKAlcPan1s5VJS8d3XAH55xIXWfjBte3oiMqaNhTo+0P5LK0cv536+f3nYPdSFK8kwc5eN8u4wroOF07HQaaGV1L9ytNo+sGuIC1Z7er/4VBje0kKLKtBziZVfzt74szQpn8ihp5TUFBclNVJsnnjiCebNm8ebb75J5cqVsxyfPHkyrq6ujrUb2SU1AEuWLGHgwIHce++9rF69ulhiFwgODqZNmzZmhyEiIsUgODiYPa6tGLa+eEdqBjQx+PY+F/p/Z2VZaPpb1KpeEDrShXUn7Axem3s8H9xpYVgLg6afWTkTm97Ws67BxoEujPrBytyQ9GvO7mlhVCuDlvOthF5K7+flCodGuBCZAB0W3vgi/+a+qex/wuuGryPO0fQzKRZWq5WVK1dSt25dJk6cmG2fkJAQ5syZw4YNGxxlm7MzYMAAWrRowbp16wgJCaFVq1ZFFbaIiEiZZLfDOztsWIDiTGv6NzYIj7ezPPTqZ+5RifDtYTuPNTdwd4GUXPKNhxoZrDludyQ0AJtO2zl80c7AJoYjqbm9tsHuCzgSGoDENPjumJ1n2loI9CXPUaG8uFi0dL046actxWLx4sVER0fTq1evHPsMHToUgA8//DDP67366qtYrVbGjx9faDFK7mrWrGl2CCIiUkzOu9djb1TxJjQAbasZ7Iqwc/00om1hdsq5GTTOZTeHWuWhejmDHeFZJyFtC7PTttrVamQeLulJzPUSUtP/bF/9xiuXpaSk3PA1xHlKaqRYzJ07F4AhQ4bk2KdLly7UrFmTH3/8kdjY2Bz7ATzyyCM0btyY9evXExwcXJihSg6c2UNIRERKh4/3u5tSxrlmOQjLpthmRlut8jkHVbNc5r7Xn1/FK32kB+DwRTut/KD8dUtFu/qnX99fS2FKHE0/k2KxdetWp/qdP3/e8X1kZGSufQ8fPnxDMUn+nD17lqpVq5odhoiIFLEzMXa+P+1+w6M0BjiSiLwk/zWlzMv16vfXSkq7ejwnGceyPf+a66dY4aM9dvoGWvjmPguv/ppeKGBMGwsdauR9H2d5eHjc+EXEaRqpkTLl4sWLJCcnO17HxcVlGhVKSUkhOjo60zlhYWG5vg4PD89U0a203sNqtZaK59A9dA/dQ/fQPXK/x+kLl7Fx48M03eoYJD3v6tRXk7/qByWmpU8Nu56n69XjOck4lu35Lpn7rD9h55lNVrrVNtg91JXQka70CTB49df0VC6uEGaOpaWllZj/5jfjPfJL1c9ExClJSUnaSExEpIzo/lUqv503sN7Au8Tq3tCrgXPJ0YojdmJS0qucHblkp8/yzONEI1oazOvlQtD8NPZFZX+NWuXh3FOuvLzFyrTtmQNfcI+F3gEGVT/IPIzj7QatqkKKDYIvwMgggzl/d6H3MivrTtzYW2SVdC5emn4mIk45d+6c9gUSESkjBvlHsPVcjRu6RkQCfLE/f4lB8AU7t9c2MCBTsYBbahrEp9ozVSu73vk4uJBgp0ON68+GTjUNgi9kjSUhFf68ZoDgznoGCal2fjtXCJ/5GyYsSirDNP1MRJySV/EGEREpPTr5RFHHhEGGpaF2apQzeLDx1YSgilf6/jWrj9kzlXMOqJj+da1loXbuDTCofU3sPeoaNKlssCQ090Slcy14sJHBvL3po0Y3ykuzG4qVRmpExCmaeiYiUnZUKF+Of7a3MG6zLUt55aK0NNTOH+ftfN7LQvMqNqIS0xfwuxjwn98yT0nbNDB9oUyDuVczncn/Z2NAYxd+HujC+7tslHeDlzpaCIm08/m+q09StwJ8e58L3x21ER4PLaoaPNXaICQSJvxSOIWsE5OSgPKFci3Jm9bUiIhTrFaryjqLiJQRKSkpJNjcqPmx1VF5rLj4esC07hb6BRp4ucH2cBi32crOiMz9Tvwja1ID0LwKvPs3C11rG6RYYe1xOy9utnEhIfM9Pu9l4ZaaBpU94Vxc+gafb/9pIy61cJ5Da2qKl5IauSExMTEEBgYSGRnJgw8+yLJly/I8Jzk5mRdffJEVK1Zw4cIFPDw8aNOmDe+++y6dOnXK9pw//viDV199lV27dhEfH4+Pjw/NmjVj4cKFNGjQoLAfS7IRHBxMmzZtzA5DRESKQcbv/DE/WpkbYidN7xbzrblvCvuf8DY7jDJD08/khjz11FPExMQ43d9ms9G1a1d27NjBbbfdxogRI7hw4QJff/01d9xxBxs3bqRz586Zzlm4cCEjRoygRo0aDBkyhBo1ahAREcH27du5dOmSkhoREZEi8mw7Cx/tyWbjF8mTZjcULyU1UmCbNm3im2++4fnnn+edd95x6pxPP/2UHTt2cN999/Hdd9852seMGUOnTp0YPXo0wcHBjvZTp07x1FNP0apVK3777TdtZGWi6tWrmx2CiIgUk1q1agHQrIpBz7qw+Qw3VN65LEpJTgH0vqW4qPqZFEhqaiqjRo2iQ4cODBs2zOnzNm7cCKSP8FyrdevWtGzZkj179nDw4EFH+5QpU4iPj2fGjBl4eHgQExOTaXMnKT5KKEVEyg7jmnLEL3SwKKGRm56SGimQV155hbNnzzJv3rx8nZeRkJQvn7UaSEZ1rU2bNjnafv75Z7y8vIiKiqJBgwZUrFgRLy8vWrRowYYNG27gCSS/Tp8+bXYIIiJSTM6dO+f4vlcDg/oVTAymhHL3cDc7hDJFSY3kW0hICB988AFPPfUULVu2zNe5zZs3B2DNmjWZ2mNiYjhw4AAAJ0+edLSfPXsWq9XKww8/TJMmTfjggw94/vnnOXnyJH379uX333+/sYcRERGRXFkMg+c7WNBWkvljtRZOaWhxjtbUSL4NHz6c6tWrM3Xq1HyfO3bsWD788ENmzZpF+fLl6devH2FhYbz22muOzR0TEq7WXExKSsJms9GzZ0/Wr1/vaO/cuTMDBgxgwoQJbN68+YafSfLWuHFjs0MQEZFi0qxZs0yvh7UwGL8VEoq5vHNJZk3TD6s4aaRG8mX69Ons3r2bWbNmFWiNRY0aNVi7di01atTgP//5D61bt6ZXr14kJSXx+OOPA1Cx4tXtgd3c3AAYOXJkpuv079+fKlWqsHv37ht4GsmP8PBws0MQEZFicubMmUyvK3gYjAwycNFwjfP0sypWSmrEaQkJCbz55pt07NiRunXrsnv3bnbv3s3hw4eB9Clku3fvJiIiItfrdO3alRMnTrBr1y6WL1/Orl272Lt3r2O9zbVT2qpUqQJA3bp1s1ynSpUqxMfHF9bjSR7yU7pbRERKtri4uCxtz7azYFPBAKd5eXqZHUKZouln4rTY2FhiY2PZtm0b7dq1y3J848aNtGvXjueff5533303z+u1bduWtm3bOl5v3boVLy8v+vTp42hr2bIl58+f5/jx43Tp0iXT+ZGRkVSooJWLxcXdXQseRUTKioziPddqVMmgVwP44aTKOzsjMSkJyFoYSYqGkhpxmq+vL7Nnz87SHhERwaRJk2jfvj3Dhw/nlltuAeDYsWNER0fTtGnTPJOP8ePHc+bMGf7xj3/g6+vraB82bBg//PADH330EUOGDHG0z5s3j0uXLnHvvfcWzsNJnq6fXy0iIqVXYGBgtu3/bG9h3QktgHeKXZlfcTLsdv3E5cbs27ePoKAgHnzwQZYtW+Zo79WrFxs2bGDJkiX079/f0Z4xfa158+YYhsGmTZv4/fff6dSpE1u3bs2yVqdHjx78/PPPdOjQgV69enHy5Em+/fZbvL292b59e46/eKVwBQcH06ZNG7PDEBGRYpDT73y73U7jeVaOXi72kEqc5hVT2P8Pb7PDKDO0pkaKXYcOHdi+fTvTpk1j6tSpRERE8Nprr/Hrr79mW3xg/fr1PPfcc5w7d47//ve/rFq1iu7du/PHH38ooRERESlGhmHwfHu9fXRG5XIuZodQpmikRkSccu7cOfz9/c0OQ0REikF4eDg1atTI9lh8ip1Vx/T2MS9+MQe465b87ecnBac1NSLilHLlypkdgoiIFJOcEhqAcu4GjzZTveK8BAdrn5ripPFDEXHKyZMnzQ5BRESkxKhfv77ZIZQpSmpERERERAqZ9tIrXkpqRMQpKsogIiLivMjISLNDKFOU1IiIU6KioswOQURERCRbSmpExCmXL182OwQREZESo3Xr1maHUKYoqRERp7i6qliiiIiIsw4ePGh2CGWK3qWIiFNatlStfRGRMmPrfnh1kdlR3Nw6N4Gpj+d4OCUlpRiDESU1IuKUPXv2aChdRKSsmPMj9t8OYWiPzZz9fhie6wO1q2Z7uEKFCsUcUNmm6Wci4hS7Xf+yiYiUKYY22MyVYcBHG3I8nNsGplL4lNSIiFOqVKlidggiIiI3D6sNPlgHicnZHg4NDS3mgMo2JTUi4hQNo4uIiFznSgJ8/avZUQhKakq8pUuXYhgGhmHw0EMPZdvn1KlTuLq6YhgGQUFBmY4FBQXh6elZHKFm8ueffzriXrlyZY79MvpkfLm5uVGjRg0GDhzIuXPnii9g4cSJE2aHICIicnMxDHjnO8hminbdunVNCKjsUqGAUsLNzY0NGzaQkJCAt7d3pmMzZ87Ebrdjsdw8OezMmTPx9PTE3d2dTz75hH79+uXYt379+jz99NMAREdH89NPP7FkyRL+/PNPDh8+jJeXVzFFLSIiInINux32n4FfDkC3FpkOJSdnPy1NisbN8y5XbkjXrl2Jj4/n008/zXJsyZIldOrUCTc3NxMiyyo5OZm1a9fSo0cP7rnnHn7++Weio6Nz7F+tWjXGjRvHuHHjmDJlCv/3f//HbbfdxpkzZ/j888+LMfKyLSAgwOwQREREbj4uFpixJktzRESECcGUXUpqSok2bdpQr149FixYkKl9/fr1nDlzhuHDh9/Q9T///HMsFgt9+vTJ1H7p0iVq1apFxYoVOX78uFPX+uyzz4iJieGJJ55gzJgxJCcn8+GHH+YrnjvuuAOAQ4cO5es8KbjLly+bHYKIiBQnFT9zjtUGK7fBqQtmR1KmKakpRR5++GF27drFkSNHHG0fffQRFSpUYNiwYTd07eHDhzNw4EC+//57ZsyYkeme4eHhzJ492+lP8j///HP8/Py4//776datGw0aNGDRovxt8HX06FFAFbmK08WLF80OQUREipMq+TvPYsCH6zM1Xb+OWYqWkppS5LnnnsPFxYX3338fgNjYWH788Uf69OmDu7v7DV//888/p2HDhkyYMIE9e/YwdepUfvzxRwYNGsSQIUOcusbRo0fZuXMn/fr1c6zxGTBgAIcPH+bPP//M9py0tDTOnj3L2bNn2b17N6+99hrLli3Dw8PjhpM1cd7NtCZLRETkpmK1wccbIOHqOpprP2SWoqd3KaWIv78/nTt3ZtmyZQDMmTOHxMREnn322UK5vpeXF0uXLsVut3P//ffz73//m4YNG+ZrXcuMGTOw2+2Ohf8Azz77LC4uLsyaNSvbc3bt2kWdOnWoU6cO7dq14+2336Zu3bqsWLGCevXq3fBziXNatWpldggiIiI3r9hEWLjF8TIpKcnEYMoeJTWlzIgRIwgPD2fVqlUsWLCARo0a0blz50K7fps2bZgwYQKnTp3CbrezZMkSPDw8nDrXZrOxdOlS6tWrh81mY/fu3ezevZvIyEiaN2/O2rVrSUlJyXJe48aNWbx4MYsWLWLSpEnUrl2bqKgop+8rhWPv3r1mhyAiIsXIMExcVFPRGz55Ci58DnGL4ac3oG0+CtY09Yd1EyF2EUR/AQueg6rZ7LfWsAYseQkuLoD4r+CXt+FvLQse97tXyzv7+PgU/DqSbyrpXMoMHjyYF154gddff529e/fy73//u9Dv8f333wOQkpJCcHAwbdu2deq85cuXOyqBtGvXLts+8+fPZ9SoUZnafH19eeSRRxyvR4wYQcuWLRk0aBDHjh3TL41iYrVazQ5BRESKkd1uN6dWgGHA2tegdT2YtgqiYmBML9j8JrR/CY6G5X6+fxXY+lb6xpgTFkF5LxjXF4LqQadXIDUtvV/tKvDHlPSpY9NWQXwSDO8BP/wber6eXqY5P+zA4fPw8z7oEYS/v39Bnl4KSCM1pYybmxt9+/YlODgYNzc3nnnmmUK9/vjx4/nzzz8ZOnQoNWvWZOzYsRw7dsypc+fMmYOrqyvvv/8+s2fPzvJVrlw55s2bl+d1atWqxbhx44iMjGT8+PE3+kjipEqVKpkdgoiIlAY/vwmf5/L+pH9n6NIUhs2GN79NX4D/t3+nJx9vPJz39Sc8BOU8ocd/YNb3MGUZDHwH2jSAYXdc7fevB8G3HHSfmN5n5lq4bTyEXYL3Clg11sUCM1YDqtBa3DRSUwq98soruLu7ExgYSNWqVQvtulu2bGH69Om0atWKzz77jC1btnD33Xfz0EMPsXPnTlxcXHI8NzIyki1bttC+fXuee+65bPusWbOGH374gWPHjtGwYcNcYxk3bhzvv/8+8+fP5/XXXy/U55TsqdKciIgUi/6dIfwSLL+mgFBUDHz7OzzWDdxdISUt5/MfuhXW7IAzUVfbNoXA4XMw8DaY+2N62+3NYPcJCD1/tV9iCny3HZ7pDYE18x4Vup7Vln7v4+H5O09umEZqSqFmzZoxZ84cXn755UK75uXLl3n00Ufx8vJi2bJluLi40KNHD55//nn27NnDP//5z1zPnz17NikpKTzwwAM59hkwYAA2m81RvS037u7uPPPMM8THxzNx4sT8Po4UQEYZbRERKSPMWlLTtgHsOu5Ym+Kw7Uj6CEzjWjmfW6syVPeFHdnMItl2JPO6HA+39CTmehkVzNoXcNNpiwVmr6N27doFO18KREmNOOWRRx4hLCyMmTNnEhgY6Gj/73//S6dOnfjwww9Zt25djucvWrQIi8XC448/nmOfQYMG4enpydKlS52K6eWXX6ZKlSosWLCAyMhI5x9GRERE8mbWPjU1K6VPAbteRlutyrmfe23f68+v4pM+0gPp619a1YPynpn7dW2W/uf/t3ffYU2d7R/Av2EPGbKHIEMFQQVxj4pa62odbR11b9qqrbVqrdo6aqX9Yd/XbbEqah2vdVtH1TrQtu6JuBiCojIUQYYQRs7vjzTRmABhhsD3c11cNs95zjn3ebAmd57lXMYRCoUSYO2fkGRkl+18KhMOP9NyAwYMgPDmNxlFULW0oLorWhWVsOjo6ODChQslnq/Ot/wmJibIyclRKCvu2QwNDfHs2bMij1PFcnNz03QIRESkbfR0pSuZvU5fT9pLYv3GQj/Ps6S9M8YGgFjF8LLcf3tVjIvZe092TJyv4vz8V3XyCoCfjwB9WwG/TQPmbJMuFDCxF9DSs+T7lCQ7F+J1x4DgCWW/BpUKkxoiUktWVhYsLS01HQYREVWVihh+1sEbCF+ounzIW4plbh8DD55Kh4QZqviIavRvkqFqyJiM7Jihvorz9RXrHLkGTF4L/DgcuPYfaVn0E2mCs3gUkFWefWZEqHvsNhBcjktQqXD4GdUqz58/h1j8arffrKwsZGZmyl/n5eUhNTVV4ZzExMRiXyclJSn0KNXUeyQnJ9eI5+A9eA/eg/fgPdS8R0UMP7sRD3Sbr/hzIx44ek25PCn934dIezWM7HWysifPi76fbNhZUeenZiouMrDqD8B+LNDua6DFdMD7c+lS0IDiAgKlJsB4eBft+51Xo3uUlkhQd+wSEdVq169fh7+/v6bDICKiqjB8KYT//QWRpBI+Jp76DohPAcasVH18x3TpymRO4xUXC1jzCTCsE2A1svjVz5I3AOGRwOD/KJbfXQE8SpUmUMX5bRrwXkvAcRyQ8VKtR1JiYoiYv2ahQUCzsp1PpcaeGiJSCxMaIiKqErvOAQ51gQ/aviqzNgMGtgcOXFZMaDzspT+v231OmpTUe22if9emgJczsPNs8fdu5yW97/oTZU9odHWA8d2QpSMp2/lUJpxTQ0RquXXrFnx9fTUdBhERVRGRSASNLIG26xxw7p50g06fesCzTGBiT2myMG+7Yt0TC6R/un/yqix4tzQBOvUdsOyQdHWzGf2AiHhgw8lX9VxtgR3TpPvSJKUDvi7AJz2AiAfA7C1lj18iAT7rDVMhq+zXoFJjUkNEasnPV7GSDBER1ViCIGhmqxqJBOj9vXSy/ufvSlchuxQDjF6h3jyXR6lA4LfAf0dLFwHIKwAOXQGmbVTs5cl4KZ2DM7k3YFUHeJwKLD8ELNpV9kUCdHWA7n5AA0fUzytmQQOqcJxTQ0RqiYuLg7u7u6bDICKiqlCZc2pquqNzge7+nItaxTinpgwuXbqEbt26wdnZGSYmJjA0NISzszM++ugjxMaq2MG2GHl5eZg1axY8PDxgaGgIExMTeHh4YP78+Up1d+zYAX9/f/k9GzRogJCQkAp6qrLp3LkzRCIRXF1dS3XeypUr4e3tDUNDQxgZGcHX1xcbN24ssv6TJ08wdOhQODg4QF9fH+bm5mjWrBn27dtXvgcgtdnb25dciYiIqDZr6Ai846fpKGolDj8rg/v37+Pp06d455134OLiAj09Pdy8eRP79u3D0aNHcf36ddSvX7/E6+Tk5KBjx46IiIhAt27dMGrUKBQUFCAqKgrx8fEKdZcsWYJp06bB1tYW48ePh6mpKQ4dOoSZM2fi4cOHWLmyiBVEKlFYWBjOnDkDA4PSbU41ZcoULF++HO7u7vjss88AAHv27MHYsWORmpqKadOmKdS/desWunTpgtzcXPTv3x9eXl5IT09HZGSkUjtR5YmKiuI3TkREtYlGxp5pMRGAqX0AkbThnJycNBtPLcPhZxVo5cqV+OyzzzBlyhQsXbq0xPrjxo3Dxo0bsX37dgwcOLDIemKxGLa2ttDR0cG9e/fk35hLJBK0bt0a169fx5UrV+DnV3XfDKSlpaFhw4bo0qULwsPDYWxsjIcPH5Z4XlxcHBo1agRHR0dER0fD0NAQgPQZGzZsiOfPnyM+Ph42Njbyc3x9fZGUlISLFy/C09Oz0p6JisdudCKiWoTDz0qvjhGQuB6oYwwAePr0KWxtbTUcVO3B4WcVqEGDBgCkH/hLkp6ejm3btqFdu3YYOHAgJBKJ0qZEMv/88w8yMzMRGBioMARIR0cHw4YNQ2FhIUJDQ4u9X2pqKhwcHGBpaYkHDx4oHBs9ejREIhEWLVpUYtwyH3/8MSQSCX7++We1zwGAo0ePoqCgAP3795cnNABgaGiIfv36ITs7G2FhYfLyvXv34vbt2wgKCoKnpyfEYjEyMjJKdU+qGKUdYkhERFRr6OoAQd3lCQ0APH78WIMB1T5MasohKysLjx49wr1797B161Z8+umnAID333+/xHMPHDiA3Nxc+Pv7Y9CgQTAxMYGNjQ3Mzc0xevRo5L22YkZurnQFDmNjY6XrmJqaAgCuXLlS7P2sra2xefNmZGdn48MPP4REIl07/bfffsOvv/6KLl26YM6cOWo995EjR7Br1y7Mnz9foUdFHbJnkcX9OhMTEwDA+fPn5WV79+4FALi7u6NNmzYwMTGBhYUFHB0d8dNPP5Xq3lQ+st8dERHVEhx+pj6JBJjcS9NR1GpMasohODgYLi4u8Pb2xvDhw5GdnY2ffvoJ/fv3L/HcyMhIAMCvv/6K48eP46uvvsKyZcvg6+uLTZs2oU+fPvK6AQEB0NHRwcWLF+XJiMzx48cBAMnJySXe85133sGUKVNw5coVTJ8+HQ8ePMDHH38MGxsb7NixQ61nzsvLw8cff4yAgAB8/vnnap3zupYtWwIATp8+rXTsr7/+AiBdFEBGtvDC9OnTkZGRgZCQECxatAj6+vqYMWMGgoODSx0DlU1KSoqmQyAioqrEkWfq0dWRbvbprrigTuPGjTUUUO3EhQLKYdSoUWjatCkyMzNx6dIlHD9+XO0PfpmZmQCA7OxsnD9/Hq1atQIAfP7552jWrBmOHTuG8+fPo23btnBwcEDv3r1x8OBB9OrVC7Nnz4a5uTk2bdqE/fv3A5DOSVFHSEgIzpw5g+XLl+PAgQPIysrCwYMH1e5xmTZtGpKSkvD777+rVf9NHTt2RPPmzXHu3DkMHz4ckydPBgCsWrUKly5dAqDYI5CdnQ1A2kt19epVeW/V2LFj4enpiZCQEMycORO6urplioeIiIioXAol0gUC3pCQkCCfmkCVjz015eDl5YUhQ4YgKCgIa9euxaZNm/Df//5XvqJXcWQfzr28vOQJjcxHH30EADh8+LC87H//+x/69OmD48ePo3PnzggICEBYWBh+/PFHAK+GbpVER0cHe/bsgaGhIWJiYvDJJ5+gZ8+eap177do1rFmzBuPHjy/XogSHDx9Ghw4d5HOK2rVrh6NHj2Lu3LkAgDp16sjryubdvPvuuwrD7xwcHNCxY0e8ePFCngxR5WratKmmQyAiIqpeRCLA2xno3ETpUFZWlgYCqr2Y1FSgTp06wcPDA9u2bSuxrmzJZ2tra6VjLi4uAIDnz5/Ly+rUqYPff/8djx8/xv79+3HkyBEkJyejbdu2AAA3Nze14zxw4ABevnwJALh586ba502cOBF16tTB8OHDce3aNflPYWEh8vPzce3aNbX26XFwcMDff/+N2NhY7N27FydOnEBiYqL8uV//VsPR0VF+zptkiyY8ffpU7WegsouKitJ0CEREVIVEIk6qKZEgAF/2lS/j/DojIyMNBFR7cfhZBROLxWpl5p07dwagep6CbO8VVZsdOjg4oG/fvvLXO3fuBAD07t1brfju3LmDGTNmwMXFBR06dMD27dvx3XffyXtJipOUlIS0tDS0b99e5fGAgAC0bt0aFy5cUCsWd3d3hR3qDxw4AEBxoYXWrVtj//79ePTokdL5srk39erVU+t+VD7qDnEkIqKaQRAErhVQEnMTYFgnlYc49KxqcZ+aMrh//z48PDyUynfu3InBgwejadOmuHHjhrw8NjYWqamp8Pb2hrm5ubzcx8cHd+/exZ9//om3334bAJCfnw9fX1/ExsYiMjKy2ElmERER6NixI+rUqYOYmJgSh6Dl5+fDz88PsbGxCA8PR4sWLdCsWTPExcXh9OnT8l6fovzvf/9T6D2SmTVrFgwMDLBgwQK4uLjIk66nT58iLi4OTk5OJSYef/75J3r37g1vb2+F3qOkpCS4u7vD2NgYsbGxqFu3LgBpm/r6+sLKykphYQGqPLGxsdwniIiotuA+NSXT1QGm9wN+HKHyMPd3q1pMasqgQ4cOePbsGdq1awc3Nzfk5OTg2rVrOHnyJAwMDHDo0CF06dJFXr9nz544evQodu7ciQEDBsjLT548id69e0NPTw9Dhw6FtbU19u/fjzt37mD8+PFYu3atvO78+fPxxx9/oF27drC1tcWdO3ewd+9e6Orq4uDBg+jUSfW3BK8bN24cwsLCsGDBAnnPTGRkJFq3bg07Ozvcvn1b7bk5r7O1tVW5+eYPP/yA2bNnY8KECfjll18U4rh//z5atGgBS0tLXL16FYcOHYKlpSVOnz4Nb29vhessWLAA8+fPh4uLCwYNGoS8vDxs27YN6enp2LRpE4YNG1bqmKn0cnNz2ZVORFRbDF8KYdsZiPgpsWgiERAfCriq3mCTSU3V4vCzMhg8eDC2bt2KAwcOICMjAyKRCLa2tujfvz8WLlyo9hJ+Xbt2xbFjx/DVV19hy5YtyM/PR7169RAcHIxZs2Yp1PX398eePXsQFhaGnJwcWFhYoHv37li8eLFa3Zu7du3Chg0bEBgYqDDUrEmTJvjxxx8xZcoUjBw5Ert27SpdY5RBq1at8M8//yA0NBRisRg2NjYYNGgQfvrpJ5VD7ubNmwc7OzssXboUK1asgEgkgo+PD8LCwhSG4lHlunv3Lv9xJiKqLcZ0heTWQ+jqcPp1kTp6F5nQAKrnA1PlYU8NEamF3zgREdUu/HeftAnTbyJSCxdkICIiouqKSQ0RqaWgoEDTIRARURUqzXYRRJrGpIaI1JKUlKTpEIiIqAplZ2drOgQitTGpISIiIiIl3NyatAmTGiJSS5MmTTQdAhEREZFKXP2MiNRy7949eHl5aToMIiKqIoIgQCQSaToMIrVwnxoiUktOTo6mQyAiKjuJBDh4Gcgv1HQk1YeHPdDco8jDd+7cgY+PTxUGRFR2TGqISC2mpqaaDoGIqOxCjwKT1mo6iuqlvi1w/2egiA028/LyqjggorLjnBoiUourq6umQyAiKru0bAi6/Nij4MFT4PDVIg+bm5tXYTBE5cP/u4lILXfu3NF0CEREVJF0dYAlB4o87ODgUIXBEJUPkxoiIiKqHTjnXVGhBDh5E7idoPJwVFRUFQdEVHZMaqhK7Nq1CyKRCF9++WWRdUQiEdq0aaPy2IgRIyASieDo6FhZIVIJnJycNB0CERFVND0dYMVhTUdBVG5Maqjay8vLw4EDB2Bvb4+kpCTs3r1b0yEREZE24iYWygokwMaTQFqW0iHOpSRtwqSGqr2NGzfixYsXWL58OczNzfHLL79oOqRa6cmTJ5oOgYiIKoO4AAg7oVwsFmsgGKKyYVJD1V5YWBjs7OwwYMAA9OrVC+Hh4Xj27JmmwyIiIm3DjSRVEwRg6UGgUHEPn+TkZA0FRFR6TGqoSr18+RKPHj1S+aPK/fv3cenSJfTv3x86OjqYOHEi8vLysGrVqiqOnBo3bqzpEIiIyonjz4r0KBU4eEXTURCVGZMaqlJr1qyBi4uLyh9Vli1bBkEQMGnSJABAp06d4Obmhq1bt1Zl2ATg4cOHmg6BiKh8mNMUTcXyzk2bNtVQMESlp6fpAKh26dOnD4YMGaLy2NChQ5XKdu7cCR8fHzRr1kxeNmjQIISEhODs2bNo3759pcVKirKzszUdAhFR+XD0WdEKJcDpW8DNB0DT+gCA6OhoeHt7azgwIvUwqaEq1aBBA7WTmn379iExMRH9+vXDtWvX5OWBgYFYvHgxli9fzqSmChkbG2s6BCKichFpIquxMAFCRgLvtwFMDIGL0cC0TcC1++qd7+0MLBkLdPQG8gqAQ1eALzcCzzIU6znUBRYMBt7xAxwsgSdpwP6LwKJdwHPllc1U0tMBlh8C1k4EAOTm5qr9mESaxqSGqq3Q0FD5n7L/ft3hw4chFothaGhY1aHVSp6enpoOgYioXARBqNq0RiQCDn0D+NUHFu+XJiITewLh3wEtZgAxicWf72wNnPkeePESmL0VqGMMTO8r7UlpPRPIL5DWMzUCzv0AmBoCq48ACamAnxswuRfQpYn0XoIaY+8KJMDmcODHEYC1GczMzMrZAERVh0kNVUupqakIDw9H8+bNMW7cOKXj165dw/r167F+/XpMnDhRAxHWPpGRkfD399d0GERE1cep74D4FGDMStXHB7QDOngDAxYDu89Jy3acBaJWSntVhi0t/vqzP5QmLC1mAAn/rvp5MRo4Ph8Y3QVY+6e0rG8rwM0OeHcRcPi1yf7PM4F5g6UJzvU49Z4pvxBYfxz46n04Ozurdw5RNcCkhqqlVatWQSwWY8KECfj000+VjmdkZGDz5s3YuHEjkxoiIlJPVY8+G9AOSEoD9px/VfYsQ5rYDO8EGOhJh5QV5cO2wMHLrxIaADgRAdx7DAxq/yqpMf93eHByuuL5iWnSP3Py1I9Z8u/yzl/2xd27d/llFmkNrn5G1dLWrVthYGCAYcOGqTxubm6O1q1b48qVK4iOjq7i6GonBwcHTYdARKRdmrsDV+8rD/26GC3tgWnkVPS5TlaAvSVwOVb52MVooLnHq9dnbkv3mFk2DmjTSDpsrVcAMGcAsPeCNAkqjcR/5+MQaREmNVTtnD17FlFRUWjVqhXMzc2LrPfBBx9AIpFg2bJlVRhd7aWnx45dItJyVb2ks2PdV70lr5OVOVkVf+7rdd8839pM2tMDAHceAUGhgE894PyPwKO1wOFvpL06AxeXPm4dEbDkAOrVq1f6c4k0hJ9SqEoMGDAAQgmTFF8/XlJdAJg6dSqmTp1a7thIPY8ePYKNjY2mwyAi0gw9XelKZq/T1wMM9aUJxuueZ0l7Z4wNALGK4WW5/w4HMzYo+n6yY+J8Fefnv6ojG772+Lm0B+fwVeDBU+AtH+Dz3sCzTGDGppKf73USAfjnLkQ34oG3+e8+aQcmNURERFQ7iMoxqaaDNxC+UHX5kLcUy9w+liYWOXmAoYqPWkb/JizFzXWRHTPUV3G+vmKd9t7AwdlA26+BK/8OV9t/Ech4CcwbBISdkPbmlIZseee3W5buPCIN4fAzqlWeP38OsVgsf52VlYXMzEz567y8PKSmpiqck5iYWOzrpKQkhZ6lmnoPFxeXGvEcvAfvwXvUznu8fPkS5Rp/diMe6DZf8edGPHD0mnJ5Uvq/gae9Gkb2OlnZk+dF30827Kyo81MzX/XSfNxdukjAlTfm3/x+CdDRkSY9pSUAElPplgna+jvnPbT7HqUlEtQZ50NEtV5sbCz3qiEi7bVoF4R52yEqlFTcNUta0nnHdOCtxoDTeMXFAtZ8AgzrBFiNLH71s+QNQHgkMPg/iuV3VwCPUqUJFAAc+Va6bLPjG1sgtGoAXAwBPgkF1hwr7dMh/9ZS6Pu4lvo8Ik1gTw0RqeX1b1yIiEgNu84BDnWBD9q+KrM2Awa2Bw5cVkxoPOylP6/bfQ54ryVQz/pVWdemgJczsPPsq7KoROl9An0Vz5cNi7um5h41Mro6QA9/xOnllu48Ig3inBoiUouhoaGmQyAiKhdReebUlMWuc8C5e8CGydKVyZ5lAhN7SpOGedsV655YIP3T/ZNXZcG7pQnQqe+AZYeAOkbAjH5ARDyw4eSreisPA2O6AAdmAysOS+fzBPoAQzsBx65LFxAojUIJMLXPv0P2iLQDkxoiUkujRo00HQIRUbkIglC1+29KJEDv74HFo4DP35WuVnYpBhi9Aoh6UvL5j1KBwG+B/44Gfhwu7dk5dAWYtlGxlyfqCdBiBvD9UOmmng6WwJM0YPE+5eRJHQ0cgO7+MI2JKf25RBrCOTVEpJbr169zZ2ki0l6VMaemJhIBWDkBmNgLeXl5MDAoZtlpomqEc2qoVC5duoRu3brB2dkZJiYmMDQ0hLOzMz766CPExirvenzo0CG0bdsWVlZWMDAwgL29Pd577z3cuHFD7XvGxMSge/fuMDc3h4GBARo0aICVK4uYlElERFSUKh59ppVMDIGRnQEAt2/f1mwsRKXA4WdUKvfv38fTp0/xzjvvwMXFBXp6erh58yb27duHo0eP4vr166hfvz4AYPPmzRg1ahTs7e0xfPhw2NraIjIyEvv27UN4eDiuXbuGhg0bFnu/J0+eoH379sjIyMDw4cPh4uKCXbt24bPPPkNGRgZmz55dFY9NAOzs7DQdAhFROTGrKZauDjDhHaCOsaYjISo1Dj+jCrFy5Up89tlnmDJlCpYuXQoAaNGiBW7cuIH4+HjUq1dPXnfhwoWYO3cuZsyYgZCQkGKvO3z4cGzduhXr1q3DuHHSpSrz8/Ph4+ODxMREJCQkoG5dFWv4U4V7/vw5rKysNB0GEVHZcPhZyUQAYlYDHg4AgJSUFH6hRVqDw8+oQjRo0AAAkJaWJi/LysqCvr4+7O0Vl6h0cXEBANSpU6fE6x4+fBj29vbyhAYA9PX1MWHCBGRnZ2Pz5s0VET6p4eHDh5oOgYiIKouuDvBuS3lCA2hgtTiicmBSQ2WSlZWFR48e4d69e9i6dSs+/fRTAMD7778vrxMYGIjc3Fz06tULp0+fRnR0NLZs2YI5c+bA2dkZkyZNKvYesbGxSEtLQ5MmTZSOdevWDQBw4cKFCnwqIiKq0fghvWiFEuCL9xSKHj9+rKFgiEqPc2qoTIKDg/HDDz/IX9va2uKnn35C//795WVLly7F06dPcfDgQZw4cUJe3qpVKxw+fBjW1tYoTlycdLMwR0dHpWOynqHExMTyPAaVApd0JiLtxxH3KokANHKSbuxJpKWY1FCZjBo1Ck2bNkVmZiYuXbqE48ePIyUlRaGOnp4enJycEBAQgD59+sDGxgZnzpzBzp070atXL/z999/FbuiYlZUFQPWmj6ampgCA3FzudlxVkpOT4e7urukwiIjKjjlN0b7sq9ST1bhxYw0FQ1R6TGqoTLy8vODl5QUACAoKwpkzZ/D222/j5cuXWLFiBQCgd+/eiIiIQGxsLMzMzAAAn3zyCRo2bIjvvvsOISEh+Pbbb4u8h2zOjVgsVjqWnZ0NADAyMqrQ56KivXjxQtMhEBFRZTAzBoZ1UipOSEiQj4wgqu44p4YqRKdOneDh4YFt27YBAO7cuYMTJ06gU6dO8oRGJigoCABw5syZYq8p6xVQNcQs5t9djlUNTaPKoa+vr+kQiIjKhRPfVdDVAT7uAZgqf0koGzFBpA2Y1FCFEYvF8n8A79+/DwAoLCxUqpeXl1fksdd5enqibt26iIyMVDp2/PhxAECbNm3KFTOpz9fXV9MhEBGVi8DxZ8okAjCxp8pDHA1B2oRJDZWKLFl5086dO/Hw4UN4e3sDAJo3bw4dHR2Eh4cjOTlZoa5sH5uAgAB5WUZGBi5evIjY2FiFur169UJycjLWr18vL8vPz8fatWthYmKCYcOGVcRjkRquX7+u6RCIiMqHOY0iXR2gb0vATfVeNBx6RtqEc2qoVEaMGIFnz56hXbt2cHNzQ05ODq5du4aTJ0/CyMhInrA4OTlh8ODB+N///oemTZti0KBBsLa2xtmzZ3HixAnY29vj66+/ll/32LFjGDhwIHr06IEjR47Iy0NCQnDs2DFMmjQJ58+fR7169bB7927ExMRg4cKFJa6gRkREREUolABT+xZ5ODIyEv7+/lUXD1E5MKmhUhk8eDC2bt2KAwcOICMjAyKRCLa2tujfvz8WLlyosFLKli1b4O3tjS1btmDdunXIz8+HlZUV+vfvj6VLl8LGxqbE+zk7O+Ps2bOYOHEitm/fjtzcXLi4uGDZsmX4/PPPK/NR6Q3q/L6IiKotVxuICiWajqJ68akHdPLRdBREFUIkCAI7Y4moROnp6bC0tNR0GEREZXZv+x/w8uCQKjl7C6C+6qFnAJCUlAQHB4cqDIio7NhTQ0RqiY+P5zAEItJqOd6OgH9DTYehNZjQkDbhQgFERERUK7i5uWk6BCKqJExqiEgtXAWHiLSdbONmIqp5mNQQkVpSU1M1HQIRUbk8ffpU0yEQUSVhUkNEaklLS9N0CEREREQqMakhIrXo6upqOgQionLx8/PTdAhEVEm4pDMRERHVCrdv34aPD/dlIaqJuKQzEaklIiICzZo103QYRERllpeXV+SxdRESzPlbAn7Vq1k+1sCRAbow0hNpOhTSMkxqiEgtEgl34iYi7WZubq6yXCII+L+LEqS8rOKASMnpR8DTl4CL6l8VUZE4p4aI1GJlZaXpEIiIyqWozST/jBcQk161sRBRxWJSQ0RqsbS01HQIRETlEhUVpbJ8yRUJdDnaiUirMakhIrXcv39f0yEQEVW46DQBR+OBQs6lIdJqTGqIiIioVnB1dVUqW3GVvTRENQGTGiJSi7u7u6ZDICIqF7FYrPA6Qyxg/U2BvTRENQCTGiJSS0ZGhqZDICIql+TkZIXXG28JyCnQUDBEVKGY1BCRWlJTUzUdAhFRhZEIApZcloCdNEQ1A5MaIlKLSMRB50Sk3Zo2bSr/7yNxAuLZAU1UYzCpISK1+Pn5aToEIqJyiY6Olv/3fy9zgQCimoRJDRGpJTIyUtMhEBGVS25uLgDgTqqAEw+5jDNRTaKn6QCISDsUFHA2LRFpNzMzMwDSZZz1REBBNU5qLAyBkE46eL+hCCb6wMVEAdPCJbiWUvK5rRyA0U100MZBhGa2gL6uCKKfVP8b/omfCF1dRWjjKIKruQgbIyUYc0RSwU9DVPnYU6PCrl27IBKJ8MMPP2g6FLlTp07Bz88PpqamEIlECAoK0nRIVMtYWlpqOgQionJxdnZGeq6ADbeEap3QiAAc+kAXQxuLsPKaBF+dlsDORITwwbpoYFny+b09dDC+qQgCgPsviq87s7UOurqKcCtVQD67rkiLlSqpuXz5MsaPHw8vLy+Ym5vDyMgI7u7umDRpEtLT00s8f+7cuRCJRBCJRHj06JHKOk+ePMHQoUPh4OAAfX19mJubo1mzZti3b19pQi1RZGQkgoKCcOrUqQq9bmXIy8vDRx99hMePH2PKlClYvHgxhg8fXqn3DAoKwi+//FKp9yDtYmNjo+kQiIjK5e7duwiLFCCu5h3PA7xE6OAswugjEnx3TsDq6wI6/1aIQgFY0KHkj24/X5fAYkUhWm0pxJ8Pik9UAn8rhM2qQvTeLYG4sKKegKjqlWr42fLly7Fjxw507NgRH3zwAQwMDHD69GmsXr0ahw4dws2bN+Vdu2+KiYnBTz/9BENDQ6XNr2Ru3bqFLl26IDc3F/3794eXlxfS09MRGRmJ+Pj4Uj9cce7evYu1a9fC3d0dXbp0qdBrV7SIiAikpKRg2rRpCA4OrpJ7rl27Fg8fPmSPEMnFxMTA399f02EQEZVZoQAsvVL9l3Ee0EiEpGwBe6JeRfosB9hxT8BwHxEMdIG8YhKQlJfq3+shV4CjGqJUSc2IESPwn//8B7a2tgrlY8eOxYYNG/DDDz8U+aF71KhRcHBwgKenJ44fP66yzqBBg1BYWIhr167B09OzNKFptdTUVFhbWxd5PCEhAQCKraNNMjMzYWhoCAMDA02HQkREtcgdwQ0JmZqOomTN7US4miwoJV8XEwV87KeDRnWByGcaCY2o2irV8LN33nlHKaEBpEkNIO1pUSU0NBTnzp3D6tWroaurq7LO3r17cfv2bQQFBcHT0xNisbhMO5inp6dj1KhRsLe3h76+PiwtLdG9e3eF2H744QcMHDgQADB79mz5kLjX16+XCQ4OhouLC/T19WFjY4MvvvhC5X3//PNPdOjQAWZmZtDX14eTkxMmTZqEvLw8hXpNmzaFra0tbty4gbfeegt16tQpdlhP06ZN8cEHHyjFKluJSiKRYO7cufD09IShoSGMjIzQrFkz7Ny5U+la3377LQICAmBlZQU9PT1YWlqiW7duCqtaRUZGyvcjOXr0qPx+sjLZcVU9OEFBQQqxAUDPnj0hEonw8OFD9OrVCxYWFrCwsEBUVBQA4OnTpxgxYoTCcMOuXbvixo0bCtfOyspCUFAQnJycYGhoCBMTE7i6umLYsGFFth1VLDc3N02HQERULj/fMtCKZZwdTYHEbOVyWZlTHS14CKIqViGrn8XGxgIA7OzslI49e/YMs2fPRr9+/dCzZ08sXbpU5TX27t0LAHB3d0ebNm1w+fJlSCQSODg4YNq0aZg+fXqJcYjFYrRv3x537tzBW2+9hcDAQERFRWHPnj3o0KEDLly4AC8vL/Tu3RtRUVHYuHEj+vTpg06dOgEAnJycFK63ceNGpKWlYcCAAbC0tMTu3buxbNkyuLq64ssvv5TX27BhA4KCguDg4IDRo0fD2toa58+fx88//4ybN2/izJkzCtfNzc1F165d0axZM3z55ZdITk4u8plmzZqFP//8UynWevXqAQC6d++OkydPolOnThg8eDDEYjH27t2Ljz76COnp6ZgwYYL8WmvWrEGTJk0wYsQIWFtbIzIyEvv378dbb72FyMhIODs7o169eli8eDFmzJgBHx8fjBkzpsR2V0dgYCCsrKwwadIkZGdnw9LSEk+fPkVAQACePn2Kvn37wtfXF0+ePMH27dvx1ltv4dy5c/D19QUADB48GIcPH0b37t3Rrl07FBQUICoqCufPn6+Q+Khkst8bEZE2inwq4Gxy1Y8QEAEwUP19rhLZnBZjPaic35Jb8Oo4Eb1BKKe8vDzBy8tL0NHRES5evKh0vF+/foKFhYWQkpIiCIIg9OjRQwAgJCQkKNRr3769AEAwMzMTvL29hZ9++klYtGiR4OLiIgAQFi1aVGIsCxYsEAAIw4YNUygPCwsTAAjdunWTl+3cuVMAIAQHBytdR3bM0tJSHrcgCMKLFy8EMzMzoVGjRvKyzMxMwcLCQvDx8RHEYrHCdaZPny4AEHbu3Ckva9KkiQBAGDNmTInPU1KsP//8swBAmD9/vkK5WCwWPD09BVtbW6GwsFBenpaWpnTt7du3CwCEKVOmKJQDEHr06KFU/+bNmwIAYcKECUrHJkyYIAAQbt68KS+T/b7ffvttpfoDBw4U9PX1hfDwcIXyyMhIwcjISOH+pqamQsuWLZWuUVqpqalCbm6u/HVmZqaQkZEhfy0Wi4Vnz54pnPPkyZNiXycmJgoSiaTG3+Py5cs14jl4D96D96id9/jhTIaAxflV/hO4vUBQl9d66TmZYomwLqJQ6Vq9dkmv1X1ngdr3X3FV+jlAnbqZYomw4abyfav65+ELidb8veI9Ku8epSUSBKFc8+UGDBiA3bt3Y/LkyVixYoXCsd9//x39+/fHf/7zH0ydOhWAdDjS0aNHkZCQIO9tAAB/f3/cuHEDdnZ2iI+Ph7GxMQAgKSkJnp6e0NfXR2pqapHD1wCgVatWuHLlCp48eQIHBweFY+7u7khKSkJWVhZ0dXWxa9cuDBw4EMHBwZg1a5ZCXdmx0aNHY8OGDQrH2rRpg9u3byMzUzood/369Rg/fjwWLVqEkSNHKtRNTExE69atMWrUKGzcuBGAdDhZZGQkkpKSYG9vX1LzKsTzZqwdO3bElStXcPfuXaV2WbhwIX755RdcunQJLVu2VDhWWFiI58+fyxdsaNSoEQICAvD333/L64hEIvTo0QNHjhxRODcyMhJNmzbFhAkTlFZHCwoKwtq1a3Hz5k00adIEwKvf98mTJxUWZJBIJLCwsECjRo2wf/9+pWfu27cv4uLikJaWBkDaCyiRSLB//3506NBBrXajinX9+nUuFEBEWislW4DzmgIUSKp26Ja9CdDTXb177o0WkJEHRI3TRXSagHf3KO4XM7aJCOt76qLpxgK159SseFsHk5vrFLlPzesyP9fFrihB4/vUPAzShYs5h9hR6ZSrA3PcuHHYvXs3+vTpo5TQ5OTkYNKkSfD395cnNMUxNDQEALz77rvyhAYAHBwc0LFjRxw7dgyXLl1C27Zti7zGkydPULduXaWEBgA8PT0RHx+PR48eoX79+mo9n6rFCurWrYusrCz565s3bwIA5syZgzlz5qi8ztOnTxVem5mZqZ3QFCc+Ph65ubnFznV49OiRPKnZsWMHFi5ciHv37iE/P1+hnixJqywtWrRQeJ2QkICsrCxcvXoVLi4uKs+RzeMBgO+//x5TpkxBx44dYWdnh5YtW6Jfv34YN25csYkuVRw/Pz9Nh0BEVGZ2piK855SFg4/NqnSPmuSXwKZbpbvh9RQBb9UTQQQoLBbQxlGE7HwBUWkVGiJRjVDmpCYoKAhhYWHo2bOnyj1k5s6di8ePHyM4OBjXrl2Tl8sSglu3biE1NVX+QcnR0REAVCYksgTgzeSgsqnzYVnW0TV16lS0atVKZZ03kw5ZAldegiDAzMwMa9asKbKOLAk8evQohgwZAgcHB0ydOhUNGjRQ2MhTIlHvW5nXE403FbfjvLm5ucJr2f38/f3x1VdflXjfoKAgvPfee9i2bRvCw8Nx8eJFHD58GMuWLcPly5cVEmGqHHfu3IGPj4+mwyAiKrNBTsnY90j11hPVya4oAQO9dPBBIxF2/7uss7UxMNBLhAOxgsJyzh4W0j9L2mSTqKYrU1IjG2bUo0cPHDx4EDo6youoPXjwAIIgKA3JkunZsycMDQ2Rm5sLAGjdujX279+vclPOJ0+eAIDCcDVVnJ2dcfnyZSQnJyv1hNy/fx/GxsbyaxT34bw0vL29AQB16tTBkCFDKuSa6nJxccHFixfRs2dP1K1bt9i6YWFhkEgk+OOPP9CsWTN5eXp6OrKzVSyxUgRZ0ikbFva60uwl5OrqChMTE2RnZ6vdbk5OTpg+fTqmT58OiUSCUaNGYcuWLVi/fj0mT56s9r2pbN5cyY+ISNu0dNRFO0fgQhIgqcab1eyKEnDuiYANPXXgYy3Bsxxgor8OdEXAvH8Uv4Q8MUj6Baz72leZjqs5MMJH+jmnpb30zzltpX8+yAC23H718O95iOD37zpP+jpAM1uRvO7vMQJuculo0hKlWtIZAD755BOsXbsW77zzDg4dOlRkb8akSZOwcuVKpR/ZssmLFi3C8uXL5fXHjh0LIyMjHDx4UOEDc2xsLP7++284OjqiefPmxcb23nvvQRAEzJgxQ6F806ZNiIuLQ4cOHeTxWlhIv9pITU0tbRMoGDJkCMzNzbF69Wp58vW6zMzMct+jKCNGjIAgCAornL0uLi5O/t+y535zCtXUqVOVygBpb1J6erpSubW1NSwsLHDhwgWF3p0bN27gn3/+UTt2XV1d9OzZE9HR0Vi1apXKOg8ePAAA5OfnK60Qp6OjIx/S9uwZ/8WtCm/2thERaZv69etjakudap3QANKEq/fuQvx2V8DnzXWwOFAHz3KArjsK1Rp65m4hwvcddfF9R120dZImKLLX45oqfvT7sNGruoZ6IgTYv3odYM95LaQ9StVT880332DNmjWwtrZGt27dsGTJEoXjTk5OGDp0KADpEr6BgYFK1zhw4ABu3ryJkSNHKvS8ODg44Ouvv8b8+fPh5+eHQYMGIS8vD9u2bUNBQQEWL15cYnwzZ87E9u3bsXnzZiQkJKBjx46Ijo7G7t27YWFhoTDvp02bNjAyMsKWLVtgamqKunXrwsnJCYMGDSpNk8DS0hKrVq3C2LFj4e3tjX79+qFhw4ZIS0tDVFQUwsPDsWnTJgwYMKBU11XHpEmTcPDgQezevRuNGzdGt27dYGtri4SEBFy5cgWPHj1CSkoKAOCjjz7C9u3b0bt3bwwZMgQGBgY4deoUYmNjYWam3BXfuHFjXLt2DVOmTIGbmxtEIpF8j55hw4Zh9erVaNWqFd577z08fvwYO3bsgKurK2JiYtSOPzQ0FFevXsVnn32GXbt2oVWrVjA0NER8fDxOnz4NHx8fHDlyBM+fP4eLiwvat2+PZs2awd7eHvfv38eOHTtgamqK4cOHV0h7UvFUDQ0lItImt2/fRv+mfnAwAZJeajqa4qWLgQnHJJhwrPh6r/fQyJxOENRaGAAAxhyRaHxhAKKKUKqk5vLlywCkvRszZ85UOt6kSRN5UlMW8+bNg52dHZYuXYoVK1ZAJBLBx8cHYWFh6Nu3b4nnGxoa4uzZs5gyZQr++OMP/PXXXzAxMUHnzp2xdOlS+VAxQDpZ/+eff8Z3332H4OBgFBQUoEmTJqVOagBg+PDhcHd3x7x583Do0CFkZGTA1NQUjo6OGDlyJNq3b1/qa6rrjz/+QEhICDZt2oR169ahoKAAlpaW8PLywjfffCOv17dvX4SGhuL//u//sGLFChgYGKBly5Y4ffq0fO+b18lWdfvll1/kQwRlSc2SJUvw4sULHDx4UL456ZIlS3DhwoVSJTWyTUhnzpyJQ4cO4Z9//oGOjg6sra3RvHlzTJo0CYC0h2DIkCE4e/YsLl++DLFYDEtLS3Ts2BGLFi1CgwYNytGCpK6oqCiufkZEWk9fV4TPW+jgm78l1b7HhojUV+4lnYmoduCSzkSk7VJSUmBnZ4dnLwU4hxYijx0U1RKXdKayKPWcGiKqnVxdXTUdAhFRucgWCbIxEWG4jwh6/NxMVGMwqSEitcg2ayUi0laPHz+W//fnATpVul8NEVUuJjVEpJY3V6AjItJmfnYidHQGdNlbQ1QjMKkhIiKiWqFx48YKr79sqYNC9tYQ1QhMaohILbI9poiItFVCQoLC6z6eIjjX0VAwRFShmNQQkVqio6M1HQIRUblkZWUpvNbTEWFKgA4/DBHVAPz/mIjUItuviIhIWxkZGSmVjWsqgr6uBoIhogrFpIaI1GJmZqbpEIiIykXVZs1WxiKM8uXyzkTajkkNEanF2dlZ0yEQEZVLZGSkynIu70yk/fQ0HQARaYe7d+/C399f02EQEVU4XxsROrsAZx4B7LDRHAGAhMkllRGTGiIiIqoVHBwcijw2v70u1t2UVGE0pErdvETUM+PIACo9kSAIzImJqETPnj2DjY2NpsMgIqIajO81VFacU0NEaiksLNR0CEREVMPxvYbKikkNEaklMTFR0yEQEVENx/caKismNUREREREpNU4p4aI1JKfnw99fX1Nh0FERDUY32uorNhTQ0RqiYuL03QIRERUw/G9hsqKSQ0RqeXly5eaDoGIiGo4vtdQWTGpISK1mJqaajoEIiKq4fheQ2XFOTVEpJa8vDwYGBhoOgwiIqrB+F5DZcWeGiJSy+3btzUdAhER1XB8r6GyYlJDRERERERajUkNEanFyclJ0yEQEVENx/caKismNUSkFpFIpOkQiIiohuN7DZUVkxoiUsvjx481HQIREdVwfK+hsmJSQ0REREREWo1LOhORWsRiMQwNDTUdBhER1WB8r6GyYk8NEaklISFB0yEQEVENx/caKismNUSklqysLE2HQERENRzfa6ismNQQkVqMjIw0HQIREdVwfK+hsuKcGiJSS0FBAfT09DQdBhER1WB8r6GyYk8NEaklMjJS0yEQEVENx/caKiumwlRrFBYWIioqStNhaK379+9zRRoiIqpUfK8hmUaNGkFXV1ft+kxqqNaIioqCj4+PpsMgIiIiohLcvn0bjRs3Vrs+59RQrVETe2qysrLQunVrXLx4EXXq1NF0OLUC27zqsc2rFtu76rHNpZKSktC1a1ecPHkSDg4OlXYftnfVK0ubl7anhkkNkRbLyMiAhYUFXrx4AXNzc02HUyuwzase27xqsb2rHttc6tGjR3BxcUFCQgLq1atXafdhe1e9qmhzLhRARERERERajUkNERERERFpNSY1RFrM0NAQ8+bN40oxVYhtXvXY5lWL7V312OZS5ubmCAwMrPQhYWzvqlcVbc45NUREREREpNXYU0NERERERFqNSQ0REREREWk1JjVERERERKTVmNQQEREREZFWY1JDVM388MMPGDhwIDw8PCASieDm5lam67x8+RLfffcdfH19YWxsDCsrK7Rr1w579+6t2IBrgPK2+caNGyESiYr9efz4ceUEr4Uq4u94fn4+Vq9ejRYtWsDS0hKWlpYICAjAsmXLkJeXV/FBa7mKaPOsrCx89dVX8PT0hKGhIRwcHDBmzBj+3VYhKioKc+fORdu2bWFrawszMzP4+/tj0aJFyM7OVvs6hw8fRvv27WFqagorKysMHDgQcXFxlRi59qqINt+xYwfGjBkDPz8/6OvrQyQSIT4+vnID11Llbe+0tDQsW7YM3bt3h4uLC4yNjeHl5YWgoCAkJCSUKSaufkZUzYhEIlhZWSEgIABXrlyBubl5qf9RTUtLw9tvv43o6Gj5P9DZ2dm4c+cOvLy88MUXX1RK7NqqvG1+//59nD17Vqk8MTERX331FZo3b46rV69WYMTarSL+jg8bNgzbtm3Dhx9+iLfffhuFhYU4cOAAjh07hkGDBuG3336rnOC1VHnbPCcnBx07dsS1a9cwcuRItGvXDnFxcVi1ahXq1q2LixcvwsHBofIeQMt8/fXXWLVqFfr27Yu2bdtCX18fp06dwo4dO9CsWTOcP38exsbGxV5jz549GDBgAPz8/DBhwgS8ePECS5cuha6uLi5fvgwnJ6cqehrtUBFt3rlzZ1y4cAF+fn5IT0/HvXv3EBcXV+YvF2uy8rb3kSNH8N577+Htt99G165dYWNjg8jISKxZswYGBgY4e/YsfHx8SheUQETVSmxsrPy/fX19hfr165f6GsOHDxfMzMyEW7duVWBkNVdFtLkqwcHBAgBh5cqVFXK9mqK87f348WMBgNC/f3+FcolEInTs2FEQiUTC8+fPKyLUGqO8bb5kyRIBgBAcHKxQ/s8//wgikUgYN25cRYRZY1y6dElIT09XKp8zZ44AQFixYkWx5+fl5QlOTk6Cq6urkJmZKS+/du2aoKOjI0yYMKHCY9Z25W1zQRCEBw8eCPn5+YIgCMKkSZMEAEJcXFxFh1ojlLe94+LihJiYGKXyP//8UwAgfPjhh6WOicPPiKoZDw+Pcp0fHx+Pbdu2YcKECfDx8UFhYSGysrIqKLqaqbxtroogCAgLC4OxsTGGDRtW4dfXZuVt78zMTABQ+qZaJBLB0dEROjo6MDIyKtc9aprytvmpU6cAAGPGjFEob9++PRo2bIjt27cjNze3XPeoSVq2bAkLCwul8sGDBwMAIiMjiz3/9OnTePLkCcaPH486derIy/39/dG5c2f89ttvyM/Pr9igtVx52xwAXF1doaenV+Gx1UTlbW83Nzd4enoqlXfr1g1WVlZq/b7exKSGqIY5cuQIJBIJfHx8MGLECJiYmMDMzAz16tXDkiVLNB1erXH69GnExMTgww8/hKWlpabDqVE8PT3h6emJsLAwrFu3DvHx8YiNjcV///tf7NmzB7NmzSpxmAmVjlgsBgCYmJgoHTMxMUF2djZu3rxZ1WFpnUePHgEA7O3ti6136dIlAEC7du2UjrVt2xYZGRmIioqq+ABrIHXbnCpGedv7xYsXyMzMLNP5TGqIaph79+4BAGbNmoUrV64gNDQUW7Zsgbu7O7788kvMmzdPwxHWDuvXrwcAjB8/XsOR1Dx6enr4/fff0aRJE0yYMAHu7u5o0KABvv76a6xYsQILFy7UdIg1jq+vLwDg5MmTCuWJiYm4e/cuAJR5cm9tUVhYiIULF0JPTw9Dhw4ttu6TJ08AAM7OzkrHZGVcoKFkpWlzKr+KaO9FixYhPz8fo0aNKvW57GMjqmFkQ3Py8vLw119/wdraGgAwaNAg+Pj4ICQkBF988QXq1q2ryTBrtPT0dOzevRsNGjRAYGCgpsOpkYyNjdGwYUO0atUKXbt2xcuXL7Fp0yZMnjwZpqamGDlypKZDrFE+/fRThIaG4tNPP4VYLEbbtm3x4MEDzJgxA4WFhQCkKy5S0b744gucO3cOwcHB8PLyKraurC0NDQ2VjsmGVrK9S1aaNqfyK29779q1Cz/99BN69uypNNRVHeypIaphZMNu3nvvPXlCAwD6+voYOnQocnNzcf78eU2FVyts27YNOTk5GDdunKZDqZGSkpLQqlUruLq6YvXq1RgwYABGjhyJP//8E61bt8bkyZORlpam6TBrlAYNGuDQoUMwNTXFRx99BDc3NwQGBqJevXry3khzc3MNR1l9ffvtt1i5ciWCgoIwa9asEuvLhvnJhv29TjZ3SdVQQHqltG1O5VPe9j58+DCGDRuGFi1a4LfffoNIJCr1NZjUENUw9erVAwCVy6s6OjoCAD/wVbL169dDT08Po0eP1nQoNdIvv/yC1NRUDBw4UKFcR0cHAwYMQGZmJpfQrgSdO3dGdHQ0bt26hdOnT+Phw4fYvXs3nj17BgDw9vbWcITV0/z58/H9999jzJgxCA0NVesc2SIYqoaYycpUDU0jqbK0OZVdedv7yJEj+OCDD+Dr64tjx46V+QsSDj8jqmFat24N4NVkvdfJyuzs7Ko0ptrk+vXruHr1Kvr168d9OyqJ7EOdbNjT6woKChT+pIolEokU9o4Qi8U4efIkGjRogEaNGmkwsupp/vz5WLBgAUaNGoV169ap/e1zq1atAADnzp1Dt27dFI6dP38e5ubmbO8ilLXNqWzK295HjhxB//794e3tjePHj5draDx7aoi02MOHD3H37l2FpT07deqE+vXr48CBAwrf8mVnZ+PXX3+FpaWlyhV1SD2q2vx169atAwAOPasgqtpb9qF648aNCnXz8/Oxbds26OnpoXnz5lUZZo1S0t/x182ePRupqamYM2dOFUSmXb777jssWLAAI0aMQFhYGHR0VH/kki228PocmcDAQDg6OmLdunUKS/LfuHED4eHhGDhwIPT19Sv9GbRNedqcSq+87X3s2DG8//778PLywokTJ2BlZVWueESCIAjlugIRVajNmzfjwYMHAIAVK1YgLy8P06ZNAwDUr18fI0aMkNft3LkzTp8+rbTj8eHDh9G3b184OTlh4sSJMDAwwIYNGxAZGYn169dj7NixVfpM1V1FtDkgHevu6OgIExMTPHz4ELq6ulX2DNqkvO2dkZEBf39/xMXFoU+fPujRowdevnyJLVu2ICIiAjNmzEBISEiVP1d1VhF/x1u0aIEuXbqgYcOGEIvF2LdvH06dOoWgoCCsWbOmSp+nulu1ahUmT54MV1dXLFy4UOnDnr29Pd555x0AwOjRo7Fp0yacOnUKnTt3ltfZuXMnBg8eDD8/P0yYMAEZGRlYsmQJRCIRrly5wuFnb6iINj9z5gzOnDkDADh48CAuXLiAadOmyZfl/+abb6rkWbRBedv78uXLeOuttyAIAn788UfY2Ngo3WP48OGlC6rU23USUaUKDAwUAKj8CQwMVFlX1Y7H4eHhQpcuXYQ6deoIxsbGQvv27YXff/+9ah5Cy1RUm2/dulUAIMyePbtqAtdSFdHeycnJwqRJkwQ3NzdBX19fMDExEVq1aiX88ssvgkQiqbqH0RIV0eaTJk0SGjZsKBgbGwtmZmbCW2+9JWzbtq3qHkKLjBo1qsj2frPNZXVPnTqldJ0DBw4Ibdq0EYyNjQVLS0vhww8/VLkLO1VMm8+bN6/Ya9Ar5W3vDRs2FHt+WdqbPTVERERERKTVOKeGiIiIiIi0GpMaIiIiIiLSakxqiIiIiIhIqzGpISIiIiIircakhoiIiIiItBqTGiIiIiIi0mpMaoiIiIiISKsxqSEiIiIiIq2mp+kAiIiIiKjmGD16NDZt2gQA8PX1RWRkpMJxiUSC4OBgbNiwAQ8fPoSrqytiY2MREhKCsLAw3L59Gzo6pf/ePTQ0FMHBwYiOjoahoaHCsaVLl2Lq1Kny10+fPoWNjU0Znq70UlJSEBUVhcTERGRmZsLQ0BD29vZo2bIlLC0t1bpGYWEhLl++jOjoaIjFYlhZWaFVq1aoV6+evE54eDiioqKKvMawYcNgampaYTFVN+ypISIiIiK1HD16FCKRqMifX3/9FQBgY2ODzZs348cff1S6xurVqzF37lx88MEHCAsLw5o1a5CRkYH/+7//w8yZM5USmgULFkBHRwd37txRutbYsWOhq6uLQ4cOYfTo0cjLy8OaNWuU6vXs2RObN2/G+++/X0Etob4bN24gLi4OTk5OaN++PRo3bozExETs2bMHz58/V+sa4eHhiIiIQIMGDdC+fXvo6Ojgjz/+QFJSkrxO48aN0aVLF4Wfzp07Q09PD3Xr1pUnNBUVU3UjEgRB0HQQRERERFT9hYSEYObMmVi+fDnq1q2rdLxHjx6YMWMGwsPDER8fr/IaLVq0gI2NDY4ePSovW7p0KebNm4fk5GQYGRkp1E9JSUH9+vUxcuRIhYRlxYoV+Pzzz/H9999jzpw5AICZM2fit99+Q1xcHEQikdK958+fjwULFlRpT01SUhJsbW2hq6srL3vx4gV27doFd3d3dO3atdjzU1JSsG/fPrRp0wZ+fn4AgIKCAuzatQvGxsbo169fsff+/fff0apVKzRv3rzCYqqOOPyMiIiIiNQSEREBCwsLTJ48WWXSUJLc3FzcuHEDCxYsUCjfsGED+vbtq5TQAICdnR2GDRuGzZs3Y9GiRbCxscHp06fx5Zdf4sMPP5QnNAAwaNAghISE4NSpU9Xmg7mDg4NSmYWFBerWrYv09PQSz79//z5EIhEaN24sL9PT04OXlxcuXbqErKws1KlTR+W5MTExAIAGDRpUaEzVEYefEREREZFabty4gebNm5cpoRk3bhyMjY1RWFiIb775BiKRCO3atUNcXBwiIiLQrVu3Is+dOnUqcnJyEBoaioSEBAwaNAje3t7YuHGjQr0WLVrAysoK+/fvL3V8qkgkEuTm5qr1U5rBT4IgICcnR2US96bU1FRYWFjAwMBAodzOzk5+vKjYY2NjYW9vDzMzswqNqTpiTw0RERERlSgvLw/37t1Dx44d8ezZM6XjFhYW0NfXL/L8YcOGQV9fH2vWrMGyZctgZWWF+vXr4+zZswCAgICAIs/19fVF9+7dsWrVKuzbtw/5+fnYt2+fyh6KgIAA/PPPP2V4QmVJSUk4ePCgWnWHDBmiVvIASHtQsrOz0aJFixLrvnz5EiYmJkrlsrLs7GyV5yUkJEAsFqNhw4YVHlN1xKSGiIiIiEp0+/Zt5OfnIzQ0FKGhoUrH7927h0aNGhV5fteuXXHixAmYmppi8uTJ8gUBvv32WwCAu7t7sff/8ssv0bNnT6SkpODw4cPw9PRUWc/DwwObN29W97GKZW1tjd69e6tV19jYWK166enp+Pvvv2Fvb19se8kUFBQozH2RkZUVFhaqPC8mJgY6Ojrw8PCo8JiqIyY1RERERFSiiIgIAMDGjRvh7OysdFydHoGIiAj4+voqrHCWmpoKPT29IueFyMhWP/Pw8ECPHj2KrFe3bl3k5OQU2cNRGoaGhgrLJpfXy5cv8ccff8DAwADdunVTa+lqPT09lYmLrExVwpOfn48HDx6gXr16JQ4nK0tM1RGTGiIiIiIq0Y0bN6Cnp4chQ4Yoze8ozTWKS0iKcvz4cUyfPh0NGzZEdHQ0jh07hu7du6usK5vbUpZ5P28qLCyEWCxWq66RkVGxCUFeXh7++OMP5OXloW/fvgpLLBfHxMRE5RCzly9fAoDK68THx6OgoEBpgYCKiqk6YlJDRERERCWKiIiAu7t7mROa9PR0JCQkoGnTpgrl1tbWKCgoQGZmpso5Kffv38fgwYPRvHlzHD9+HI0aNcKSJUuKTGrS0tJgYmKi9nCw4iQnJ1fInJqCggIcOXIEL168wLvvvqtyOeyiWFtb48mTJ8jLy1No+5SUFPnxN8XExEBfXx9ubm5FXrc8MVVHTGqIiIiIqEQRERFo27Ztuc4HgGbNmimUe3t7AwDi4uKUjmVlZaFfv37Q19fH3r17YWFhgYkTJ2LBggW4c+eOwjLHMnFxcSrLy6Ii5tRIJBKcOHECycnJ6NGjB+zt7VXWKygoQFZWFoyMjBSGjHl4eCAiIgJ37tyR71NTWFiIe/fuwc7OTmnYXk5ODh49eoQGDRpAT0/1R311Y9ImTGqIiIiIqFhJSUlISUmRJyBlcePGDQDKSU27du0AAJcvX1Y4JggCRowYgXv37uHUqVPyuS0TJ07Ejz/+iKVLlypsxilz9epVDBs2rMxxvq4i5tScP38eDx48gKurK8RiMaKjoxWOy+YipaSk4ODBgwgICEDLli3lx+3s7ODh4YGLFy8iJycHFhYWiIqKQmZmJgIDA5XuFxsbC0EQih16pm5M2oRJDREREREVS5aQPH36FFu2bFE67ufnpzSs7E0RERFwdnaGlZWVQrmHhweaNGmC48ePY+zYsfLy+fPnY9++fVizZg06dOggL7e1tcXw4cOxefNmBAcHKwy/unLlCp4/f45+/fqV6Tkrg2wfmYcPH+Lhw4dKx9VJIDp37ow6deogOjoaeXl5sLKyQs+ePeHo6KhUNyYmBsbGxioXc6jImKobJjVEREREVCzZ0LENGzZgw4YNSsd//fVXtZKaN3tpZMaOHYu5c+ciJycHxsbG2Lt3LxYuXIhPPvkEQUFBSvWnTp2K9evXIzQ0FHPmzJGX79y5E66urujatWtpHq9S9enTR616Tk5OKp8VkK6A1rZtW7WG//Xv37/CYtImIqE0258SERERERVj9OjROHnyJK5evQo9PT1YWlqWeM6LFy/g4eGBkJAQjBs3rkz3FYvFcHNzw9dff40pU6YoHMvNzUVWVhZCQkKwePFiPH36FDY2NmW6D1VP2rkQNRERERFVWwkJCbC1tUXHjh3Vqm9hYYGvvvoKixcvhkQiKdM9N2zYAH19fXzyySdKx0JDQ2Fra4vFixeX6dpU/bGnhoiIiIgqzO3bt/HkyRMAQJ06dcq1YlpFSUhIwL179+SvAwMDoa+vr8GIqKIxqSEiIiIiIq3G4WdERERERKTVmNQQEREREZFWY1JDRERERERajUkNERERERFpNSY1RERERESk1ZjUEBERERGRVmNSQ0REREREWo1JDRERERERaTUmNUREREREpNX+H/QICz1XyNH5AAAAAElFTkSuQmCC" + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "execution_count": 6 + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": [ + "With the waterfall plot, similar tendencies can be visualized and interpreted from the Shapley interactions like with the force plot. However, with all remaining features being put into the *other* group, the plot also hides some potentially important insights.\n", + "\n", + "To circumvent this, and to get a more holistic view over the Shapley interactions of potentially high interaction order, we proposed the `network` and `graph` plots." + ] + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": [ + "### Network Plot\n", + "The network plot visualizes all first- and second-order interactions of computed Shapley Interactions. While the `network` plot **does not plot higher-order interactions** it does plot all second order interactions. Similar to the color scheme of `shap`, positive interactions are shown in red, negative interactions in blue. The strength of an interaction is encoded in the width of the edge connecting the two features. The strength of a first-order interaction is encoded in the size of the node for the respective feature. For clarity, second-order interactions are also plotted with a decreasing opacity depending on the strength of the interaction.\n", + "\n", + "The following code will plot the network plots for the 2-SII and Moebius transforms. Note that for the Möbius transform, the network plot will only show the first- and second-order interactions (not the higher-order interactions)." + ] + }, + { + "metadata": { + "ExecuteTime": { + "end_time": "2024-10-24T12:25:28.556951Z", + "start_time": "2024-10-24T12:25:28.368699Z" + } + }, + "cell_type": "code", + "source": [ + "si.plot_network(feature_names=feature_names, show=True, draw_legend=False)\n", + "mi.plot_network(feature_names=feature_names, show=True, draw_legend=False)" + ], + "outputs": [ + { + "data": { + "text/plain": [ + "
" + ], + "image/png": "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" + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "
" + ], + "image/png": "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" + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "execution_count": 7 + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": "The network plots show like the force and waterfall plots what features interact with each other and how strong these interactions are. The network plot is particularly useful when the number of features is larger and the force plot does not contain enough information to be interpretable. Here, the network plot for the Möebius transform is slighlty different to the 2-SII plot. So the true functional decomposition of the game (i.e. explanation) is different than the aggregated representation in the second order Shapley interactions. This indicates that interactions higher than second order are important for the explanation of the model's prediction." + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": [ + "### SI Graph Plot\n", + "The SI graph plot can be considered the more general version of the network plot in that it can **visualize all higher-order interactions**.\n", + "The SI graph plot visualizes all interactions as a graph where the nodes are the features and the edges are the interactions between the features.\n", + "Interactions between **more than two features** are visualized as a **hyper-edge**.\n", + "Again, the color scheme is the same as for the network plot (red for positive interactions, blue for negative interactions).\n", + "The strength of an interaction is encoded in the width of the edge connecting the two features. The strength of a first-order interaction is encoded in the size of the node for the respective feature.\n", + "The transparency of the edge/node is used to encode the strength of the interaction/feature attribution.\n", + "\n", + "The following code will plot the SI graph plots for the SV, 2-SII, and Moebius transforms:" + ] + }, + { + "metadata": { + "ExecuteTime": { + "end_time": "2024-10-24T12:25:34.899598Z", + "start_time": "2024-10-24T12:25:28.557957Z" + } + }, + "cell_type": "code", + "source": [ + "# we abbreviate the feature names since, they are plotted inside the nodes\n", + "abbrev_feature_names = shapiq.plot.utils.abbreviate_feature_names(feature_names)\n", + "sv.plot_si_graph(\n", + " feature_names=abbrev_feature_names, show=True, size_factor=2.5, node_size_scaling=1.5\n", + ")\n", + "si.plot_si_graph(\n", + " feature_names=abbrev_feature_names, show=True, size_factor=2.5, node_size_scaling=1.5\n", + ")\n", + "mi.plot_si_graph(\n", + " feature_names=abbrev_feature_names, show=True, size_factor=2.5, node_size_scaling=1.5\n", + ")" + ], + "outputs": [ + { + "data": { + "text/plain": [ + "
" + ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjIAAAIvCAYAAABuhDEcAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAABlkUlEQVR4nO3dd3gU5d7G8e9sdtNDDb0oHUINAUVEkKIUUVBQigUQUSn2cuwiBzk2RAWCoCCKVFEERRBQpAgIBCQUpfcaWkjPlnn/yIH3YBJIYJPNkPtzXbnU7Mw8v4DJ3nmqYZqmiYiIiIgF2XxdgIiIiMiVUpARERERy1KQEREREctSkBERERHLUpARERERy1KQEREREctSkBERERHLUpARERERy1KQEREREctSkBERERHLUpARERERy1KQEREREctSkBERERHLUpARERERy1KQEREREctSkBERERHLUpARERERy1KQEREREctSkBERERHLUpARERERy1KQEREREctSkBERERHLUpARERERy1KQEREREctSkBERERHLUpARERERy1KQEREREctSkBERERHLUpARERERy1KQEREREcuy+7oAESkc0t0mKU4PTjeYJnhMMIyMDz8DAh0GQXYDwzB8XaqIWIiCjIjkiXNpbpLSTVLSTc6lmSSle3B6wO0xMf9xrc0wsNsgwG5QJMBGiL9BiMMgLNCGv5+CjYhkzzBN858/U0RErojTbXI62cPxRDdnUj2ku0xcHjIFl8vxM8DhB0EOgzKhfoSH2CgS4JcnNYuItSnIiMhVS0r3EJfk5niCh4R0D+lu7z3bZkCww6BEkI0yoX4UD7Zht6mXRkQyKMiIyBVzeUyOnHNz4KyLpHQTdx7/NAm0Q8lgP6qU8FMPjYgACjIicoXOpbnZe9pNXJLbqz0wl2MAoQEG1xWzU76In3pnRAo5BRkRyZXzvTD7z7pITMs8cfdSTNPk7KkTJCcm4ExLxWa34x8QSIlSZfEPCMxVHf5+UCpEvTMihZ2CjIjkWLrbZNdJF0cSXDnqhXGmp7Hx91/Yvnk9u7duZPfWPzl98nim6/zsdq6vEUH1eo2pXjeShs1aU+H66pd9vgEUDTSoEe6gVIjCjEhhpCAjIjmS7jLZcdLJkQQ3Ls+lrz1x+AALZk1k8ewvOXPqBOGlShHVOIomTaKIjIwkPDycwMBA3G43ycnJ7Nixg5iYGNavj2Hr1i243W4ib27LHb0GcMOtnfCzX3qniLAAg5olHZQJU5gRKWwUZETkstLdJttPZISYS03oPXZoHxPffYk1v/xISGgoffv04bHHHiMiIiLHG90lJyfz7bffMnZsNH/8sYZSZStw3+Mv0uG+/ths2W9GHupvUCtcYUaksFGQEZFLcrlNtp90cSjelW2I8Xg8LJj5OV+8/yrh4SV57dVX6d27N6GhoVfV9oYNGxg1ahRff/01jZrdypPDx1Gm4nXZXh8WYFCnlINwDTOJFBoKMiJySXtOO9l9yoUzm+GkE0cO8tHLj7Lpj2U8+uijfPDBB4SFhXm1hiVLlvBw//6cPHmKh18cQccej2Tbw1MiyEaDcg6CHTpKTqQw0He6iGTrTLKbQ/HubEPM/p3beL5nK04f3cvixYsZP36810MMQLt27diyeTMP3N+bsUOf4rN3/oXHk3VRZ1M97DvtRr+jiRQOOmtJRLKU7jbZe8ZNYnrWgWDv9s280qcj11euxKJFP1OmTJk8radIkSJMmDCBhg0bMmTIENJSkhny1uhMPTMeE44luikeZFCuiH7EiVzr9F0uIlk6FO/iZHLWa6yP7N/NG/3vour11/Hrr79QvHjxfKtr8ODBBAcH8/DDDxMSVpSHX3g70zWpLpN9Z10UDbJpiEnkGqcgIyKZnE3JGFLKapm1y+nk3WceoESxIixa9HO+hpjz+vXrR3x8PM888wy1G91A89u6ZLomPtVk32k3EWUUZESuZfoOF5FMjid6SMpmSOmbzz5g7/YtzJgxnVKlSuVzZf/vqaee4s477yJ66FPEnzmZ6XWPCXHJbs6l5eP5CSKS7xRkROQiyU4PJ5LcWR49sHf7ZmaMe4eXXnqJqKiofK/tfxmGwfjxn2J6nIwf/nyW16Q4TU4mXmb3PhGxNAUZEbnIySQPKc7MMcY0TT55dSC1atbi9ddf90FlmZUrV44xo0ezbP4s1i1bmOl1jwnHE92k5/Wx3CLiMwoyInKBy21yPJsjCLau/50dWzYwatSHBAQE5H9x2ejduzfNmt3E95NHZ/l6ktPkZJKGl0SuVQoyInLBmVQPCelZD8XMnz6BmjVr0a5du3yu6tIMw2DIkMH8uXopB/dsz/R6ujtjzo+IXJsUZETkgoQ0D2muzJ8/HXeM3xd9z6BBA3N8ZhLA0KFDMQyDChUqZLmB3c0334xhGPTt2xeAyZMnYxgGJ09mnrx7Kd27d6dkeDgLZnye5euJ6R5SsgloImJtCjIickFCmpnlJN9f507D3+GgT58+uX6mw+Hg5MmTLF++/KLP79+/n9WrV1/1eUwAAQEBDHjkEX6Z8zUupzPT6043JLs0T0bkWqQgIyJAxvyYxGx6Lf7asJrmzW+mWLFiuX6uv78/HTt2ZPr06Rd9fsaMGdStW5dq1apdSbmZ3HHHHSQmxHNg91+ZXnO6zSwnMIuI9SnIiAgAyS4PzmzmxO7etpGmTZtc8bN79erF7Nmzcf5Pb8m0adPo3bv3FT/znxo1aoRhGOzaujHTa26TbPfFERFrU5AREQBS0sHlyfxmf+bkceKOHbmqfWPuvPNO0tLSWLRoEQDbtm0jNjaWnj17XvEz/yk0NJRatWqzc8uGLF8/l6Y5MiLXIgUZEQEg1eXJctn17m1/AlxVkAkODqZLly7MmDEDgOnTp3PTTTdRpUqVK35mVpo0iWLPf+v9pzSXqf1kRK5BCjIiAoAJmFm8z8efzlhBVKFChat6fq9evZg7dy4pKSnMmDGDXr16XdXzslKhQgXOZXFcAWR8fZ4sepxExNoUZEQEAI+HLFcspaemYBgGDofjqp7fvn17HA4Hb7zxBnv37uW+++67qudlJSgoiLTUlCxfM82MnX5F5NqiICMil5aLfWMuxeFw0K1bNz788EPatGlDmTJlvPLcf7LZ9GNNpDCx+7oAESkYbDYwyNwr4x8QiGmaOJ1O/P39r6qNRx55hBMnTjBgwICrek52UlJScAQEZvmaYYDNO5lMRAoQBRkRAcBmGBhG5nkyxUqWBuDAgQNUr179qtq44YYb+P7776/qGZdy8OBBipYoleVrBmBTkhG55qgPVkQACPAzsGfxE6F63UYAxMTE5G9BV2Dd+hiqRURm+Vqg3cDfT0FG5FpjmGZW6xREpLA5l+Zm42EnSVnsgNuvdU36PNCL9957zweV5cy5c+coWrQoT48Yz233PJjp9eol7NQsdXUTlkWk4FGPjIgAEGy3YffL+rVqdSNZX8B7ZDZuzNjRt0bdzD0ydhsE+6s3RuRapCAjIgDY/QzC/LP+kVA36mZW/f47p06dyueqcu6HH36gSLESVKpWO9NrdptBsENBRuRapCAjIheEBRhk9XbfpksvPKbJF198ke815URKSgoTJ02i3T0P4WfPvIbB4YeCjMg1SkFGRC4IDbARYM/8hl+0RCladOhGdPQ4PJ6Cd2bRzJkzOXvmDJ16PpLl62H+NgId+nEnci3Sd7YFuT0ZZ8ZkdcAfZGzDnu42celcGcml4kE2igRk3XNxR69H2bt3DwsXLsznqi7NNE3GjB1Lk1tuo1zlqple9/eDMmH6USdyrdI+MgWc023i+O+S0VSnyfaTTg7GuzmR6OZ4ooeTSW5SXSYeM2PDL7thUDzIRqlQG2VC/SgXZqNmuIPSoRmzOF0eEwPw034akgW7zaB0qB+nUzIfIFm70Q1ENG7G008/Q+vWrQkKCvJNkf8wadIkYtavZ/ikH7N8PdTfRsngbGYxi4jlafl1AeMxTUwzI2gcjHexZGcqm4462XzcyZ7Tris+K6ZksI16ZRw0KOug+XUB3FDJH7vNwOU2sWtvDfkfKU4PMYfTOZeW+X+2g3u28+TdzXhiyBA++OADH1T3j3oOHqRuvXo0a9eFp0eMz/S6zYBa4XaqlNCya5FrlYJMAXE+UGw6ms78v1NZsiuV3addedZekQCDllUCuL1GIO1rBOH/3745m5fO1RFr2xHnZPdpV5aHSH47cRRffPAaK1eupHnz5vle23mmadKhQ0c2xG5m7Lz1hBYplumaUH+DyPIOwgLUIyNyrVKQ8SHTNDGBNJfJ7M0pTNmYxPaTeRdeshPmb9CtXjB9o0KoUsKOy2Ni19BToRaf6ib2qJOE9Mw/HtxuN/+6vy3JZ06watXvlC9f3gcVwogRI3j11Vd5a8IcmrRsn+l1mwHXFbNTp7R6Y0SuZQoyPuL2mCSmm3z8ewIzY5NJzOINwxeaV/bnuVvCaFIxALfH1FyaQmzfGSc7T7pwZrFI6cSRg/zr/raULF6EZb/9RqlSWZ9vlFfGjh3LkCFDuH/Iq/Qe8mqW15QIstGwnIMgrVYSuaYpyOQzl8fE6Tb59I8kPluXSFIBCTD/1KZqAK+0LkKNcAce09SQUyHkdJtsOe7kaII7y9cP7dnByw+1p1TJ4ixZspiKFSvmS13vvfce//rXv7i775P0/9d/MLL4fzPQblC3tIMyYRpSErnWKcjkk/O9G99tSWb40nOcSi54e3H8k82Ae+sH82bbIgTYDQ03FUJnU9xsPpb1EBPAkf27ee3hO8CdzoTx47nrrrvyrJZTp07xxJNPMn3aNHoOeokHnng9yxBzfkipdil7lq+LyLVFQSYfuDwm8akeXvjpLL/sTvN1OblWJtTGex2LcWvVQPXOFEL7zjjZecqFM+uOGU7HHWPMG0P4Y+lP3P/AA3zy8ceUKFHCqzXMnTuXRx97jOSUNB57bSRt7uqV7bUlg2w00JCSSKGhIJMPfvw7hVd+Pkt8qrX/qO+tH8Tw24tht6HemULE5THZedLFwXhXpr1lzjNNk6XzpjP+7ecJCQ7kxRdeoG/fvlcVaEzTZOXKlXw4ahTfz5nDja07MeSt0ZQoXS7be4oGGESUdlBc+8aIFBoKMnnE898/1nd+S2D82kQfV+M9EaXtfNG9JOEhNoWZQsTpNtkR5+TQOTeX2jD61PEjfDHydVYsmI3dz4+ePXsycOBAoqKi8PPLWbg4c+YMM2bMYOzYaLZu3UKF66vTc9DLtL6z5yWHisICDGqXclAqRCFGpDBRkMkDLo9Justk8Nwz/LrHekNJl1My2MZn95QgsrxDw0yFSLo7o2fm8Lnse2bOO3PyOItmf8nCmZ9z4ughQkJCaNQokqZNmxAZGUnJkiUJDAzE5XKRkpLCjh07iImJYf36GPbs2Y3NZqNZ28506vUoDZvdis126WGiIgEGtcIdlApViBEpbBRkvMzlNolP89Bz+il2+GBPmPzisMFHnYvTqXagwkwh4nSb7D7t4nC8i7Rs5sz8L7fbzZZ1K9m5JYZdWzeye+tGjhzYk+m64JBQqtZpSLW6kVSvG0mDG24hvOzlV0EZQLFAg5qlHDqGQKSQUpDxIpfH5Eyyh+7TTrLvTA5+yluczYAPOhXj7rpBCjOFiNtjcizBzb4zLs6lmVnu/nspKUmJpCQlkJ6ehp+fHf+AAEKLlsjx0NN5AX5QJtSP64vbCQ3QxF6RwkpBxkvOr0zq9vVJ9haCEHOezYCPOxejcx2FmcImMc3DvjMujiW6Sc/H/+UNMoaSri9up2yYnzZtFCnkFGS8wGOapLugy5Q4/o67doeTsuNnwJf3leSmyv6aAFzInO+d2X/WRWK6edm5M1cryG5QKsSmXhgRuUBBxksGfHeaRTtTfV2GzxQJMJjftxTlw/x0mnYhlOz0cDLJw/FENwlpHlK9mOf9DAj2NwgP9qN0iI3iwTb1/onIBQoyXvDB8nOMXn3tLLG+UtVK2PmhTzhBDkNvNIWUy2NyJsXDiQQ3p1I8pLszemk8ufwpY7eBw2YQ7DAoG2YjPMSPEH/1wIhIZgoyV8HlMfltTyr9vz3j61IKjLbVApjUvaSvy5ACICndQ1K6SbLTQ2KaSUKaSbrHxO0BE5Pzs4QNw8AwMsJLsMNGWIBBiMMg2N8g1N+GQz18InIJCjJXyGNm/GBu89kJTlrg3KT89EGnYtxTN0iTMOUibo9JijPj0FSPCR4yJu7ajIyPIIdBgF29LiKSOwoyV2HIvNP88FfhnReTnSIBBr8+UpoSwTaFGRERyVMKMlfA5TH5ZVcqj87xzZCSaZq4Tx8g/XAs6cf+wkxLxHSlg82O4QjAXqIy/hUa4F+2DoYj0Cc1tq4awOR7NcQkIiJ5S0HmCjjdJrdOOMGhc/m3eYbHmUJy7A+kbpqD83AszqSzAJQqU47ixUsQEBiAy+kkKTmZwwf343a5MPzsBJarjaNaK0JvfAB7icr5Vi/A5O4luOX6AK1iEhGRPKMgk0suj8mXG5IY9su5fGnPeWofiX9MITVmJs6kM7Ru05bWt7YiKiqKqKgoypQpk+me1NRUYmNjiYmJYd26dcz+bg4J5+IJrt2GkBv7EFizNcZlzq7xhtql7CzoV0ormEREJM8oyORSitND83EnOJ2StxN8TWcq8b+MImF5NGFFijKg/8M8/vjjVK9ePdfPSk5OZvr06Xw8eiybN20k6LooinX7EEfp3D8rtz7qXIw76wRpozwREckTCjK54PaYjF2TyMgVCXnaTtrBjcR/+yyuU3sZ+uabPPvsswQFBV31c03TZNmyZfQf8Bj79u+nyG0vEtZiAIYt7w7bq1TUj+WPlVavjIiI5AmtdcylKRuT8uzZpmkSv/QTToy7i9rlwti4YQOvvvqqV0IMZOzXceutt7Il9k+efmIw8QuGc3LCPbgTT3nl+Vk5GO/ml12puNzKyyIi4n0KMjnkcpss3JHKicS8GVIyTZOzPw4l/ud3eeXll1n7x2rq1auXJ20FBQUxcuRIVqxYQUjSQU59dg+u+CN50hbAlxuSNeFXRETyhIJMDtn9DL7akDe9MaZpcnb+MBJ+/5wxY8YwfPhwHA5HnrT1v26++WZW/b6Ckv5OTk3sgTshLk/aWbkvjYNnXWgUU0REvE1BJocOx7tYczA9T56dsGI8CSsn8MknnzB48OA8aSM7NWrUYPlvv1LUSOHU5PvxOFO83oYJTNuUnOvzdkRERC5HQSYHnG6Tn7bnzQ6+6Ue3cW7ROzz//PM88cQTedLG5VStWpXFixbijttF/OIP8qSNRTtTtcuviIh4nYJMDjj8DJbs8n6QMd1O4r99lhrVazB8+HCvPz83GjRowL+HvUXiygmk7V/v9efvOuXiULzL688VEZHCTUEmBxLSPKw/7P1hpXO/jSXt6DamfDWZgIAArz8/t5577jkaN25C/HfP5skQ08IdWr0kIiLepSBzGW6Pycp9abi8vFjJnXiKhN8+4YXnn6dp06beffgVstvtfPXlF7hOHyBxzRSvP3/53jStXhIREa9SkMmBTUedXn9m0voZ+BnwwgsveP3ZVyMiIoLu3bqTum4Kpse76S32mPf/HEVEpHBTkLkMP5tB7DHvDiuZHjcp66bQq1dPSpYseCdEDx48iNQTe0jdvdKrzz2T4uFYQv4dtCkiItc+BZkc2HLcuz0JqTuWknbqIIMHDcr1vQ0bNsQwDFasWJHl6wsXLqRdu3YUK1aM4OBgIiMj+eSTT3C5cj7R9uabb6ZO3fokr/ky1/VdzsYj6bi1DltERLxEQeYyTia5iU/17htv8ub51I6ol+u5MVu3biU2NhaAadOmZXp95MiRdOzYkSJFijBlyhR+/PFH2rVrx/PPP0/37t3x5HCoyDAMHn/0EZL/Wuz1Sb9/x7m0n4yIiHiNgsxlHM2DoRDP0Vha3XIzRi4PUpw6dSo2m43WrVvzzTff4HT+f0/Rxo0b+de//kWfPn347rvvuPPOO2nTpg3vv/8+48ePZ+7cuYwdOzbHbTVv3hzT48Z59K9c1Xg5xxLd2PV/nYiIeIneUi7BY5ocPefdIONJTyb12A6ioqJydZ9pmkyfPp02bdrw7LPPcurUKRYuXHjh9dGjR2MYBm+99Vame/v06UPNmjX56KOPctxe/fr1sTscpB+OzVWdl3Mi0Z3rACciIpIdBZlLcHvguJcPiXQe3Yrp8eQ6yKxatYp9+/bRu3dv2rdvT8mSJS8aXlq2bBkNGjTguuuuy3SvzWajc+fO7Nmzh8OHD+eovYCAAOpE1CP98OZc1Xk5eXXopoiIFE4KMpdxMtnLQeb4TgzDyPXJ1tOmTSMwMJB77rkHh8NB9+7dmTdvHomJiQAcPnyYypUrZ3v/+dcOHTqU4zajIhtixu3IVZ2XczJZq5ZERMR7FGQuI93LO9F60pMJCAzC398/x/e4XC6++eYbOnXqRNGiRQHo3bs3ycnJzJkzx6v1/a8iRYqAM9mrz0zXKQUiIuJFCjKXYBgZw0te5U7H3z93xxEsWrSIuLg47rzzTs6ePcvZs2epX78+5cqVuzC8VKFCBQ4cOJDtM86/VrFixRy36+/vj+ny7h46blNLlkRExHsUZC7FBD9v/wn5OXA6cxcOzoeVfv36Ubx4cYoXL06JEiU4evQoS5Ys4cSJE7Rq1YrNmzdz8ODBTPebpslPP/1E1apVqVChQo7bdTqdGH6OXNV6OX6a6CsiIl6kIHMJJuDw8+4zDUcgaakpOd6gLjk5mblz59K1a1eWLl160cf06dNxuVzMnDmTJ554Ao/Hw5tvvpnpGVOmTOHvv//mmWeeyVWtiYmJ4AjK1T2XY/fyn6eIiBRudl8XUJCZQIkg72Y9R3g1PB4Pf//9d44m/M6dO5fExESefPJJbr311kyvv/fee0ybNo0nnniCd999l+eff574+Hj69etHcHAwP//8M6NGjaJLly4MyuVOwps2b4USVXN1z+WU9PKfp4hIQkICGzdu5K+//iIpKYn09HT8/f0JDg6mVq1aNG7c+ML8Qrn2KMhcgt2AMqHe7ULwr5ARXmJiYnIUZKZNm0blypWzDDGQsUfM008/ze7du3nuueeIiIhg5MiR3H///aSnp1OrVi0++OADBg0ahM2W8xDhcrmI3fQnwe3a5/ienCgTpi4ZEbk6LpeLH374gW+//ZaYmBi2b9+OaZr4+fkRHByMv78/6enpJCcn43ZnrJSsUaMGUVFRdO3albvvvjtXCy6kYDNMU7MvL+XPI+l0mXLSq8888eEtDOhxB5988olXn+tNsbGxNGzYkNKPfktg1WZee+599YN4v1Nxrz1PRAqPo0eP8tlnnzFhwgQOHz5M48aNuemmm2jSpAlRUVHUqVMHu/3/fz93u91s376d9evXExMTw5o1a1i7di1lypRhwIABPProo1SqVMmHX5F4g3pkLqNsHvQg2MrVZ9XqP7z+XG/6448/wDAu9CB5S5lQP5xuE4efJv2KSM6kpaUxbNgw3nvvPfz9/XnggQcYOHAgjRo1uuR9fn5+REREEBERwUMPPQRknFn36aef8vHHHzNixAieeuophg8fTnBwcD58JZIXNGHhMkqH2gjx9+6bblCd24lZv5a//vLuOUbeNOmLLwmudjO2gFCvPrdmuB0tXBKRnFq/fj1NmjTh/fff59VXX+XIkSOMHz/+siEmO3Xr1mX06NEcPnyYESNGMG7cOBo1asTvv//u3cIl3yjIXIbNMKhb2rtLkIPrdcK/SDjjxo3z6nO9ZdOmTaxZ/TvBzfp4/dmR5f2x25RkROTSTNPk3//+N82aNcPhcLB+/XqGDh3qtUm7YWFh/Otf/+LPP/8kPDycW265hZdffhmPR8eoWI2CzGW4PSb1y3o3yBh2fwKjejNp8pcXjhgoSMaNG4d/sbIE1bndq88N8zeoVEyjmSJyaR6Ph0GDBvHGG2/w6quv8scff9CgQYM8aatWrVqsWLGCd955h3fffZe+ffvmeHsMKRgUZC7DBBp4OcgAhN74AMlJiXz66adef/bVOHToEF9+NYWgJg9g+Hk3dNQt4/0/RxG5tpimycCBA5kwYQKff/45b731Fg5H3v7s8PPz48UXX2T69OlMnz6dhx56SD0zFqIgcxl2m0HLKgF4ezDEXqwCIc368drrb7Bz504vP/3KmKbJIwMexRNQhLDm/bz+/BbXB+DyaJGciGTv9ddfvxBi+vfvn69t9+jRg2nTpjFz5kyefvrpfG1brpyCTA6UCPajQTnv/0ZQrMNLEFqGh/r0u7DXgS9NnjyZnxcuoGjXd7EFF/P68zvUDESLlUQkO8uWLePtt99mxIgR9Ovn/V+mcuLee+/lk08+YfTo0cyfP98nNUjuaB+ZHHB5TMatSeSDFQlef3bqnjWcmNCNkSNH8uyzz3r9+Tl14MAB6tZvgFn9dkre95HXn1+xiB+/Dyzj9eeKyLUhKSmJBg0aUL58eZYtW5arDTy9zTRNOnXqRGxsLFu2bKF4ce19VZCpRyYH/IyM3oS8EFi1GWG3PMYLL7zAd999lydtXM7Jkydpd3sHnPZQincemidttKsRiEeZWUSy8fLLL3P06FEmTZrk0xADYBgGn332GYmJiT79BVNyRkEmBwzDoEa4g3p5NFm1WMdXCarXmZ49ezFv3rw8aSM7cXFxtG13O/uPxFGi3/Q8GVIC6NUwGOUYEcnK1q1bGT16NG+//TY1atTwdTkAVKxYkVGjRjF58mTWrFnj63LkEhRkcsjlNnmocd7s/GjY/Chx38f4127H3ffcw8SJE8mPEb/du3fTvEVL/t57iBL9Z+AoVS1P2mlc3kHtUg78tH+MiGQhOjqasmXLMnjwYF+XcpG+fftStWpVxo4d6+tS5BIUZHLI7mdwd0QwRQLy5s3YsPtTouc4gqN68sgjj9D17ns4duxYnrTl8XgYPXo09eo34ODpVEo+Ogf/snXypC2AhxqH4HKrO0ZEMktISOCrr75iwIABBe4gR5vNxsCBA5k1axZxcXG+LkeyoSCTC3Y/6Nkw787jMPzslLjnPcIf+IyFv66kdkRdpk2b5tXemd27d9Pq1jY8+eST2BvdR6knFuEIr+K15/9TeLCNzrWDsGu5kohk4euvvyYlJYVHH30002tDhw7FMIwLH6VKlaJNmzasWLEi3+rr168fhmEwadKkfGtTckdBJhcMYHCzMK+fvfRPwfU6Uerppbgq38L9999PrTp1GTNmDPHx8Vf0PNM0+eWXX7j7nm7UrFWLddv2UnrALEp0eRtbQIiXq7/YkJtCdbaSiGRr2rRpdOrUiYoVK2b5elBQEKtXr2b16tWMGzeOU6dO0bZtW7Zs2ZIv9ZUsWZLu3bszderUfGlPck9BJhcMwyAswOCRpnn75g/gF1KCkr2iKT3gGw77V+XJp56mXPkKPPLII0yZMoVt27Zdcu+ZEydOsGDBAoYNG0aNWnVo164dC1dvoWjnYZR66hcCq92c519DpaJ+PNg4RGcriUiW3G43GzZsoGXLltleY7PZaNasGc2aNaN79+788MMPuFyufN0VvWXLlmzdupXk5OR8a1NyTgff5JKfzWDgjaFM2ZDM6ZS838I6sFpzAqs1xxV/lKR105n6wxwmTpwIQFBwCPXqNyC8RHGCggJxuVwkJiWz7a+/OXbkEACO4KL4V29J6UffJqBKM4x87B557pawjDMeRESy8Pfff5OcnExUVFSO76lcuTKlSpVi7969eDweRowYweeff87Ro0epUqUKzzzzDI899tiF64cOHcoHH3zA0qVLGTx4MLGxsVSpUoX333+fzp0756jNJk2a4PF42LRpEzfddFOuv07JWwoyV8Dfz+CFlmG8/POVDfVcCXvRchRt9yy0exZPSjzphzeTfngzW4/9hbk/EVynMW12bI5A/GrcRfitDfCv0AC/EpXzNbyc16icgy4RQdg0riQi2Vi/fj0AjRs3zvE9586d49SpU5QvX54XXniBjz/+mNdee43mzZvz448/8vjjj+N0OhkyZMiFe5xOJz169OC5556jSpUqjBs3jrvvvpsNGzZQv379y7ZZr149/P39iYmJUZApgBRkroCfzaB3oxB+/DuF3/en53v7tqCiBFZvQWD1Fvnedk4E+MFHnYvjMUGjSiKSnc2bN1O1alWKFi16yevOn0Z96NAhnnvuOdxuN+3atePBBx/khRdeYOjQoQDcfvvtnDx5kmHDhjFw4ED8/PwASE9P57XXXuPhhx8GoH379tSoUYMRI0Ywffr0y9bp7+9PREQEsbGxV/HVSl7RHJkr5PaYfHhHcULzeOKvFT3TIozrivtpboyIXFJCQgIlSpS45DVJSUk4HA4cDgdVqlRh6dKljBkzhtDQUJxOJ/fee+9F1/fo0YO4uDh27Nhx0efvvvvuC//u5+dH165d+eOPP3Jca/HixUlI8P4xNXL11CNzhfxsBqVCbLzWpggvLcy/IaaCrnF5B4/dGKohJRG5rLS0NAICAi55TVBQEMuXL8cwDMLDw6lUqRI2m42vv/4agDJlLj7D7fx/nz59+sLnHA5HpvOSypQpw9GjR3Nca2BgIKmpqTm+XvKPgsxV8LMZ9GoYwsYjTmbGajZ76VAbn91TQkNKIpIjfn5+F4aNsmOz2WjSpEmmz5/vyTlx4gQVKlS48Pnjx49f9DpkzJE5c+bMRWHm+PHjlCtXLse1ulwuQkNDc3y95B8NLV0l0zQZ0b4oTSoUrB0p81uAH0zsVoJiQTYNKYlIjgQFBZGUlHRF995www04HA6++eabiz4/a9YsSpcuTc2aNS/6/Jw5cy78u9vt5vvvv+fGG2/McXvJyckEBQVdUa2St9Qjc5UMw8AwTT7vVoJOX8RxJCH7vV2uZf/pUIy6pXWekojkXNWqVZk0aRIulwu7PXdvR+Hh4TzxxBO8//77BAYG0qxZM3766SemTZvG6NGjL0z0hYzJusOHDyc1NZUqVaoQHR3NwYMH+f777y9cM2zYMIYNG8bu3bu57rrrLmrLNE3+/vtvbrvttqv6eiVvKMh4gZ/NIMwfpvUsSbepJzmVnPf7yxQkL7UKo1u9vDu6QUSuTVFRUaSkpPDXX3/laBn0P73//vsUK1aMzz//nOHDh3P99dfz6aefXrSPDGTMkZk+fTqDBw9m8+bNVKlShW+//ZYGDRpcuMbj8eB2u7M8EubAgQOcOnUqV/vdSP4xzPw4ZrmQcLlN9p5x0WP6qUITZp67JYwnm4f5ugwRsaBz585RrFgxJk6cSL9+/fKkjfMb4iUmJl7xM7777ju6devGkSNHcjWvRvKH5sh4kd3PoEoJO7PvD6d06LX/R/vyrUUUYkTkihUpUoSaNWuybt06X5dySWvXrqVcuXIKMQXUtf9um8/sNoPKRf2Y36cU9co4fF1Ongi0G4y9qziP36gZ/CJyddq3b8/s2bNJS0vzdSlZcrvdzJgxg/bt2/u6FMmGhpbyiMtj4vbAs/PP8OPf187eA+XCbEzqVpJapeya2CsiV+3vv/+mTp06TJ06ld69e/u6nEx+/PFH7rzzTtauXUvTpk19XY5kQUEmD3lME5thMG5NAiNXJOC0+LSZmyr7E92lOEUCtcRaRLynbdu2pKWlsXLlSl+XkkmnTp2Ii4sr8MNfhZmGlvLQ+d1tH7sxlAX9rDvUFOww+PdtRZnRK5yiCjEi4mWDBg3i999/Z82aNb4u5SJbt25l4cKFDBo0yNelyCWoRyafuDwmBjB2TSJjViWQZpHtZm6+zp8POhWnTKhNQ0kikidcLhc33ngjaWlpxMTEXPbYgvyq6eabbyY+Pp4///yTwMBAX5ck2VCPTD6x2wz8bAZDbgplxeNluLd+UIHexr9WuJ3J3UswrWe4QoyI5Cm73c7kyZPZsWMHw4YN83U5AIwcOZL169czefJkhZgCTj0yPnB+7szuU07eXZbA4l2peArI38J1xfx4+uYwutYNwmOiYSQRyTfDhw9n6NChrF692qcTa7dt20ZkZCRPPfUU7733ns/qkJxRkPEht8fEz2Zw5JyLyTHJzNqczJmU/J8RbDOgddUA+kaF0LJKIC6PqQAjIvnO6XTSvHlzjhw5wsqVK6lSpUq+13D48GFuueUWgoKCiImJUW+MBSjIFACmaeIxwWPCgh0pLNyRyrI9aSSm591fjQE0LOfgtuqBdKsfTLkwPwUYEfG5Y8eO0aJFC0zT5Ndff8107lFeOnr0KG3btiUxMZGVK1dSuXLlfGtbrpyCTAHjcpvY/QxcHpM/DqTzy+5UNh11su2Ek2Tnlf9VGUDVEnbql3VwU2V/bq8RSIngjPDiZ2QcfikiUhDs27ePtm3bkp6ezsKFC6lbt26et7l7927at29Pamoqv/zyC7Vq1crzNsU7FGQKMPd/VzrZbAYe02T/GTexx9I5muDhRKKbE0lu4hI9pLlNXJ6MISKHzaBYkEHpED9Kh/pRJtRGzXA7dcs4CHJkzO12uk0cfgouIlJwHTlyhPbt27Nr1y6GDx/O008/fdGJ1t7i8Xj49NNPefHFFylfvjyLFi3i+uuv93o7kncUZCzG5ckYhvIzuORKItPMCDcAdpt6XETEepKTk3n11Vf5+OOPadasGV988YVXe0r27t1L//79Wbp0KY8//jjvvfceYWE6P85qtPzaYuw2A38/47LLoQ3DwOGX8aEQIyJWFBwczKhRo1i+fDlxcXHUr1+f+++/n99//52r+R183bp19OvXj4iICPbs2cOSJUsYN26cQoxFqUdGREQKvOTkZMaNG8e4cePYvXs3DRo04OGHH6Z58+bUr1//kquL0tPT2bx5M2vWrGHy5MmsX7+e6667jscee4whQ4YowFicgoyIiFiGx+Nh8eLFjBs3jvnz5+NyubDb7dSrV48GDRoQFhZGQEAA6enpJCQksGXLFjZv3kx6ejo2m4327dszaNAgOnbsmCdzbiT/KciIiIglpaamEhsby/r164mJieGvv/4iOTmZtLQ0/P39CQ4OplatWjRp0oSoqCgaNmxIcHCwr8sWL1OQEREREcvSZF8RERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsBRkRERGxLAUZERERsSwFGREREbEsu68L8BbTNDl48CAxMTHExMSwe/dukpOTcTqdBAYGEhYWRv369YmKiqJx48YULVrU1yWLiIjIVTJM0zR9XcSVMk2TP/74g+joaBYuXEhcXBwAZcqUoU6dOoSEhGC320lLS+PMmTNs3ryZ5ORkAGrWrEmvXr0YMGAAFSpU8OWXISIiIlfIkkEmJSWFqVOnEh0dzcaNG6lSpQr3338/N9xwA1FRUZQvXz7L+9xuN9u3bycmJobly5czY8YMUlJS6NKlC4MHD6ZNmzb5/JWIiIjI1bBckFm1ahX9+vVj586ddOrUicGDB9O+fXtsttxP9zl37hxTpkwhOjqabdu20b17d8aOHUvp0qXzoHIRERHxNssEmZSUFF5//XU+/PBDmjZtyqRJk6hbt65Xnm2aJrNmzWLIkCEAjB07lvvuu88rzxYREZG8Y4lVSwcOHKBJkyaMGTOGd955h99//91rIQbAMAx69OjB1q1badWqFT169KBfv344nU6vtSEiIiLeV+B7ZHbs2EG7du2w2+38+OOPRERE5HmbU6ZM4eGHH6ZTp07MmjWLgICAPG9TREREcq9AB5n9+/fTokULwsLCWLx4cb6uLlqwYAH33HMPHTt2ZNasWdjt18xKdRERkWtGgQ0yiYmJREVF4XK5WLFiRbYrkfLSjz/+yN13382jjz7K2LFj8719ERERubQCO0fmpZde4uDBgyxYsMAnIQagc+fOfPTRR0RHR7N48WKf1CAiIiLZK5A9MkuXLqVNmzZ8/PHHPPnkkz6txePxcNttt7Fz5062bNlCkSJFfFqPiIjIP5mmyf69ezN2t18fw5atW0hMTCItLQ2Hw0FQUBDVq1UjKiqKJk2bUDsi4pqZMlHggkxiYiINGjSgUqVKLF269Ir2h/G2ffv2Ub9+fXr16sWECRN8XY6IiAgAW2JjGRc9jpmzv+HUqVMAlC9fnoYNG1KsWDH8/f1xuVwkJSWxbds2duzYAUBQUBDtb7uNQQMH0fb22wrEe+2VKnBBZuTIkbzyyits27aNatWq+bqcC8aMGcMTTzzBjh07qFGjhq/LERGRQsrj8TB75izGRkezfOUKypYtS58+fbjllluIioqibNmy2d577tw5Nm7cyNq1a/nqq6/YsmULNWvUYOBjj/HIo48SGhaWj1+JdxSoIOPxeKhZsybNmjXj66+/9nU5F0lNTaVChQr07duXkSNH+rocEREphHbv3En//o+wbMVyWrVqxaBBg+jatSv+/v65fpZpmvz+++9ER0cze/ZsKpQvz6TPJ9K6Xds8qDzvFKi+pMWLF7N7924GDRrk61IyCQwMpH///nzxxRcXDp4UERHJDx6Ph9EffUyDRo3Yf/AAv/76K7/99hv33XffFYUYyNgMtkWLFkybNo2///6b666/nja3tWPw4wNJTEjw8leQdwpUkImOjqZhw4bcdNNNub63YcOGGIbBihUrLvr8vn37MAzjwofNZqNChQr07t2b/fv356qNxx57jLNnzzJjxoxc1yciInIlkhIT6dzpDp585mn69unD5s2bad26tVfbqFq1Kr/++iujR49m8pSviIyMZM+uXV5tI68UmCCTlpbGggUL6NOnD4Zh5OrerVu3EhsbC8C0adOyvGbEiBGsXr2alStX8s4777B69Wo6deqE2+3OcTvVqlWjZcuWzJs3L1f1iYiIXIlz8fG0v709y1euYMGCBYyNjiY0NDRP2rLZbAwZMoRNmzaBYdCiZUu2bdmSJ215U4EJMlu2bMHpdNKsWbNc3zt16lRsNhutW7fmm2++yfKMpBo1atCsWTOaN2/Ogw8+yEcffcS2bdvYvn17rtpq1qwZMTExua5RREQkN1KSk7mz851s2baVX375hQ4dOuRLu9WrV2flypWEh4fT7vbbC3zPTIEJMjExMfj5+dGwYcNc3WeaJtOnT6dNmzY8++yznDp1ioULF172vrD/zszO7cGQUVFRHDp0iBMnTuTqPhERkZwyTZO+ffqwfkMMCxYs4MYbb8zX9suUKcPixYsJDQ2lfYcOBXrOTIEJMuvXryciIoLg4OBc3bdq1Sr27dtH7969ad++PSVLlsxyeMnj8eByuUhPT+evv/5i6NCh1K5dm3r16uWqvaioKAD1yoiISJ6ZOW06s2bPZvLkyVc0b9QbypQpw4IFCzhy9Cj/euFFn9SQEwUmyGzfvp26devm+r5p06YRGBjIPffcg8PhoHv37sybN4/ExMSLruvRowcOh4OAgAAiIiLYv38/3377LX5+frlqr0qVKgQFBeV6SEpERCQnjh87xpCnnuS+++7j3nvv9Wkt1apV49133yV6/Kf8uniJT2vJToEJMikpKReGe3LK5XLxzTff0KlTJ4oWLQpA7969SU5OZs6cORdd++6777Ju3TrWrl3LnDlzKF++PB06dODw4cO5atMwDEJCQrQEW0RE8sTAxx/HZrMxZswYX5cCwKBBg2jVqhX9BzxCwrlzvi4nkwITZNLT03O9Fn7RokXExcVx5513cvbsWc6ePUv9+vUpV65cpuGlqlWr0qRJE5o2bUrXrl2ZN28ehw8fZtSoUbmu1d/fn/T09FzfJyIicinr165lzty5fPzxx5QqVcrX5QAZq5kmTpzI4SNH+HzCZ74uJ5MCE2QCAgJIS0vL1T3nw0q/fv0oXrw4xYsXp0SJEhw9epQlS5ZcckJuqVKlCA8PZ+vWrbmuNT09nYCAgFzfJyIicinjxo2jcuXK3Hfffb4u5SLVqlWjW7dujBv/KR6Px9flXKTABJng4GDi4+NzfH1ycjJz586la9euLF269KKP6dOn43K5mDlzZrb3Hz9+nJMnTxIeHp6rOj0eD4mJiYSEhOTqPhERkUs5c/o002bM4PHHH8/1/M38MGjQIHbu2lXg5soUmCATERHB5s2bc3z93LlzSUxM5Mknn+TWW2+96KNnz55ERkZeNLy0c+dO1qxZw+rVq5k9ezZ33XUXhmEwYMCAXNW5c+dOUlNTiYiIyNV9IiIilzJ50iTcbjf9+/fP0fVDhw7NdnO87F6bM2cOhmHQtm3uz1Nq0aIF9erVY2z02Fzfm5cKTJBp0qQJ27dvJyGHa9WnTZtG5cqVufXWW7N8vU+fPqxZs+bCzr2vvPIKN910E82bN2fgwIGEhYXx66+/0rJly1zVeX7ZdePGjXN1n4iIyKXM/2kBHTp0oHTp0nnWxtSpUwH47bffOHLkSK7uNQyDhx56iJ8XL8blcuVFeVekwASZqKgoTNNk48aNObr+hx9+YP/+/dkeZ/DUU09hmibVqlXDNM2LPuLi4liyZEmuQwxk7HdTpUoVSpQoket7RUREsmKaJjEbN+Tpxnfnzp1j/vz5tGvXDo/Hc0XnBt5www2kpKTw97ZteVDhlSkwQSYiIoKgoCBWrVrl61IuafXq1Rc2xRMREfGGPbt2cfbs2Tx9f/nuu+9ITU1l6NChREVFXeidyY3IyEgA1q9b7+3yrliBCTJ2u50uXbowadKkAjcj+rytW7eyZs0aunfv7utSRETkGhKzPmPawpUEGZfLlekjq/fRqVOncv3119O8eXN69+7Nhg0bcr25a5EiRahZs2aB2t2+wAQZgMGDB7Nz505+/fVXX5eSpXHjxlG2bFnuvvtuX5ciIiLXkJ07dxIeHp7rvWOSkpJwOByZPv79739fdN2xY8dYunQpPXv2xDAMevbsic1mu6JemYiICHbs3Jnr+/JKgQoyN998M/Xr1yc6OtrXpWSSkJDAV199xYABA3K9cZ+IiMilpKSkXNG2HkFBQaxbty7Txz9X5M6cORO3203v3r0BKF++PK1atcrybMLLCQkJITU1Ndf35ZUCFWQMw2DQoEHMnTuXTZs2+bqci3zyySckJyfnerm2iIjI5TidTux2e67vs9lsNGnSJNNH+fLlL7pu6tSp1KpVi0qVKl3YCf+uu+5i9+7d/PHHH7lq09/fn3RnwdndPvd/anns4YcfJjo6mr59+7J27VocDoevS2LLli0MGzaM559/nkqVKvm6HBERucYEBATk2dE3u3btYt26dQAUL1480+tTp07N1Wqp1NRUAgMCvVbf1SpQPTKQkfQmT57M5s2bGTFihK/LweVy0a9fP6pVq8bQoUN9XY6IiFyDQkNDiY+PxzRNrz972rRpGIbBnDlzMu2E3759+wvDTjkVHx9foHa3L3A9MpCx2dzLL7/M8OHDueOOO2jSpInPahkxYgQbNmxg9erVBAYWnAQqIiLXjnr16nHu3Dn27dtHlSpVvPrsadOmccstt9C1a9dMr507d44uXbqwZMkS2rdvn6Pnbdq0iQd73+/VGq9GgeuROe/1118nMjKSTp06sWPHDp/U8MUXX/Dmm2/y+uuvc8MNN/ikBhERufZFNc34hd3by5pjYmLYvn07Dz30UJavd+zYkVKlSuV49dKxY8c4fPgwUU0Kzn5qhpkX/VhecvLkSVq2bMm5c+dYsmQJtWvXzre2v/zySx5++GEGDBjAuHHjst1BWERExBsqVqjIgw89yH/+8x9fl5Kt+fPn07lzZ/bs2kWVatV8XQ5QgHtkAMLDw/nll18oVqwYt9xyC8uXL8/zNj0eD++++y59+/alf//+jB07ViFGRETyXJOoxqxcudLXZVzSypUrKVGiBNdXrerrUi4o0EEGoFy5cixbtoyIiAhuvfVWnnrqKZKSkvKkrZ07d9KqVStefvllXn31VcaPH18gj1IXEZFrT4/7erBy5Ur++usvX5eSJafTyeTJk+nR/d4C9Qt+gQ8yACVLluTXX39l5MiRTJgwgYYNG/Lrr796bXZ3eno6o0aNomHDhhw9epRly5YxfPjwAvUXJSIi17Z77u1O6dKl+fTTT31dSpa+//57jh07xsBBA31dykUK9ByZrOzYsYN+/fqxatUqIiMjGTRoEL179yY4ODjXzzpw4ADjx4/n888/58SJEzz55JOMGDGiQC0rExGRwuPVl15mzLhojhw5UuDei1q3bo3b6WL5yhW+LuUilgsykDGP5eeffyY6Opr58+dTpEgRunXrRtOmTWncuDENGjTIcqn0qVOn2LBhAzExMaxYsYKFCxcSGhpKnz59GDhwIHXq1PHBVyMiIpJh/759VK1WjbfeeovXXnvN1+Vc8Ntvv9G6dWumfz2Vnvf39nU5F7FkkPlf+/btY/z48SxcuJAtW7bgcrmw2+1UqVKFkJAQHA4HqampnD17loMHDwIQFhZGVFQUvXv3plevXoSGhvr4qxAREcnw8ov/YuRHo9iwYQP16tXzdTkkJibSoEEDKpavwG/Ll2GzFaxZKZYPMv8rNTWV2NhYYmJi2LVrFykpKTidTgIDAwkLC6N+/fpERUVRvXr1AvcXISIiApCWlkbjyEiCgoNZvXq1z4/qeWLIECZOmkTsn39SvWZNn9aSlWsqyIiIiFwL1v3xB82aN+e1117jrbfe8lkdixcv5vbbb+fjD0fx5DNP+6yOS1GQERERKYCGvzWM14e+yfjx43n00Ufzvf21a9fStm1bWjS/mfkLfiqwIxkF8qwlERGRwu7VN17nxIkTPP7445imyWOPPZZvba9evZpOnTrRoF59Zn87u8CGGFCQERERKZAMw+Cj0Z8A8Pjjj3P48GFef/31PJ8zM3v2bPr06UPjRpH8OP9HQgr4gpiCG7FEREQKOZvNxsdjRvPvoW/x9ttv06xZM7Zs2ZInbZ08eZKePXty77330rF9e35e9DNFixXLk7a8SUFGRESkADMMg9fefIM1q1aRmpJC48aNGT58OImJiV55vtvtZubMmdStW5fFixcz9aspfPPttwQXsA35sqPJviIiIhaRmprKsDeH8t7IDwgJCbmqDV1PnDjBpEmT+PTTT9m/fz9d7ryTT8ePp2y5cnlQed5RkBEREbGY/Xv3MmH8BD6bNJG4uDhatGhBixYtiIqKIioqiuuvvz7TeYFxcXHExMQQExPD2rVrWbhwITabjZ733cegQYNoeuONPvpqro6CjIiIiEWlpaXx3TezmTFzBjEbNnL4yGEAihQpQtGiRQkICMDpdJKUlMTJkycBKFq0KFGRjenUsSN9H+5HyfBwX34JV01BRkRE5Bpx/NgxYtatZ8uWLSQmJpKamoq/vz+BgYHUqFGDJk2bULV69Uy9NVamICMiIiKWpVVLIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWXZfFyAiIpLvPB5wmxn/NE2wGWCzgZ8NDMPX1UkuKMhYkNtj4vKA+d//Nsj4vrPbwKZvQBGRi7ndkJwGiamQlAbnkiEhBdz/DTGQ8UPUABx2KBYCoUEQEgDBgRDsr3BTgBmmef5vUQoi0zRJc4PLnRFenG7zwvfd//7Fnf8Ws9nAYTOw28BhN/D30zefiBRCpgmnE+BEPByPh3QnpLsyPnLytudnA397xkdQAJQrDmWKQUhgnpcuuaMgU0C5PSapTpMUl4nbA4kJCWyO3cimjTHEbtzA3r27SE1JxeV04h8QQEhICLUj6tEwMooGjRpTq05dAvz9sftBkN0g0GFg6DcKEbnWpTkhLh6OnobTiRk9MN54m7P7ZYSYMsWgbDEoGZbxm6P4nIJMAeP2mCSmmaS5TNLS0/nph+/58vNxrFm1EtM0CQwMpFGjRtSpU4eQkBAcDgdpaWmcPXuWTZs28ddff+HxeAgICOCOu+6h74DHaXpjc2w2g2CHQYi/Ao2IXINcbjgYB/tPQmJKRqDJCwYQHADFQ6FKGShdLG/akRxTkClAUtI9JKSbxMXFMXHcaL7+ciJxJ47TsmVLHnzwQW644QYiIiKw27Of2pSUlMSff/7JypUr+fzzz9m1axd16tan34CB9HqwH4EBDooE2jTkJCLXjrNJsOsIHD2TMXSUX0ID4brScH0ZCHTkX7tyEQWZAsDtMUn4by/MD99/y8vPDiE1NYWHHnqIgQMHUq9evSt6rsfjYcmSJURHR/PDDz9Qr0EjPh43kYi69QnyNwhV74yIWJnLDQfiYM9xOJd08cTB/GK3QaliUL2semd8REHGx9LdJvEpHk7ExfHK808y77tv6Nq1K+PGjaNs2bJea2f9+vX07duXHTt28Oy/XmPIMy8SHOigWJBNK51ExHqSUuGvg7nuhTFNk8On44jZvZ3dxw+Tkp6G0+Ui0N+fsMBg6l9XjcgqNQgLCsldPaGBcH1pqF4+Y6Kw5BsFGR9Kc5nEp3rYujmWXvfcgTM9jTFjxtCzZ8886SlJS0tj2LBhvPvuuzS98Sa+mjmXEsWLUizIhp9NYUZELCIhGbYcyAgxOXgLi09K5KtlC1iw8Q9i9m7nxNnTAIQEhxAcHHRhruG5hAScTieGYVCzQmWaVKlFrxa30aHRjfj5+V2+Ln97xryZ2hUzJgdLvlCQ8ZF0t8nZFA8x69bS655OVK1Shfnz51OuXLk8b3vVqlV07tyZCpWuY+bchZQpHU5x9cyIiBWcS4bN++H4mcsOJf25dwfjFn3P1ysWke5y0qZ1a2648UaioqJo0qQJFSpUuOiXRpfLxV9//UVMTAzr169n+bJlbN6yhSplK/B4uy483OYOwosUu3SjDr+MnpmIygoz+URBxgfcHpPTyR62btlM146tiahThwULFlC0aNF8q2Hz5s20bduWChWv45sfFlGyeBGKB+ubTkQKsMQUiN0Px05fMsQcP3uawZ+P5Ns1v1G+XDkee/xxHnnkEcqXL5+r5kzTZO3atURHRzNz5kwME/7d4xGe6dzj0j00Dj+oWhbqVNIwUz5QkPGBsykejp04ReubGlG2TGmWLl1KsWLF8r2OjRs30rp1a5o2u5kps+ZSJNBGsL++6USkAEpzwp974PDpbIeTTNNk1qpfGDxxFIbDj1EffUSPHj1wOK5+RdHJkycZMWIEH330ETfWrMsXg16hdoXrsr/B3w41K0CtClfdtlya3rXyWYrTQ5rL5LUXnyY1JZkff/zRJyEGIDIykqlTp7Lk55+YMfVLEtNNXB7lWhEpgPYeh2Nnsw0xyWmp9PjwDXqOepPWt7Vl67ZtPPDAA14JMQDh4eF8+OGHrFy5ktOeNBq90JdPf56T/Q3pLth/ImNnYclTCjL56Pwy64Xz5/HtrGl8/PHHVKjg27R+xx130KdPH9546VkOHzrEuVSPT+sREcnkRHzGMmuXO8uX45MSaT/8WX7a9AczZszgm9mzKV26dJ6U0rx5c/7ctIn+jzzCwM8+4N/ffEG2AxsJKbDrKKTm0eZ8AmhoKV+dS/Vw5PgpWt3YgBuaNmHevHkFYh+Xs2fPUrduXWpF1Gfatz9SNMhGkEMZV0QKgDQnxOzOOHIgC4kpydw+/Fn+OnaQBQsX0KxZs3wpyzRN/vOf//Dqq6/y754DeK1736wvtPtlDC/VrpgvdRVGOv06n3hMk1SXyVeTxnMu/izjx48vECEGoFixYowePZpu3bqxYd0fNLupGUHapFJECoK9xzPOTsqC2+2m28jX2HxoL78u/ZWmTZvmW1mGYfDKK69gmiavvfYapYoU47Hbu2a+8PymfSXCoHT+LegoTPRrdz5JdZq4XG6+mjSBXr165Xr2fF7r2rUrVapU4YvPP8XlzlgeLiLiU2eT4ODJbIeUPv7pGxZvWsuc7+fka4j5X6+88gqDBg3i6cmfsP3w/qwvSkiBvcfAo6H7vKAgk0+SnSZLfv6Jw4cOMnjwYF+Xk4nNZmPgwIHM+24WJ0+dJNWpICMiPnb8bMaS6yxsP7yfV6dP4KmnnqJdu3b5W9f/MAyD999/n0qVK9Ev+j+43VmHLs4kZpzGLV6nIJMP0lwmHg9M/mwcN9xwA1FRUbl+RsOGDTEMgxUrVlz0+X379mEYxoWPwMBAateuzZtvvklKStY/ALLTr18/DMNg+pQvSHWZ2U9gExHJa6nOjHkxWaykdLvdPDzuP1SsVJG3337bB8VdLDg4mC8mT2bNji18NH9W1hclp2XsRCxepyCTD1xuk9S0NFYuX8r999+f6/u3bt1KbGwsANOmTcvymhEjRrB69Wrmz59P+/btGTZsGM8880yu2gkPD6dDhw4sXfIzmOBUL6iI+EpcfMaQTBamrljEqr8388XkyQQHB+dzYVm7+eabefrpp3ltxmecPHc28wUmGT1MSan5XNm1T0EmH7g88Pe2LTidzisax506dSo2m43WrVvzzTff4HRmXspXo0YNmjVrRtu2bfn444+57bbb+Oqrr/Dkcky2adOmbN60EY9p4tY8GRHxBY8nozcmm8Mgx/48h/a3306LFi3yubBLe+WVVzAN+GLp/KwvSErNCDPiVQoy+cDpMYn9cwM2m42GDRvm6l7TNJk+fTpt2rTh2Wef5dSpUyxcuPCy90VGRpKSkkJcXFyu2mvSpAnn4uPZt2e3emRExDfOJWfMKcnC+l1/sXbnVgYVwLmG4eHh3HfffYxbPDfrXyJd7hwfdCk5pyCTxzxmxvyYTRtjqFu3bq67QVetWsW+ffvo3bs37du3p2TJktkOL/2v/fv3ExYWRnh4eK7aOz9/Z9OfG3CqR0ZEfCExNdvemHGLvqdypUrccccdWb4+dOhQQkNDr6r5yZMn5+jnbFYGDRrE3mOH+fnPP7K+ICUt40O8RkEmj7n/G8r379tDrVq1cn3/tGnTCAwM5J577sHhcNC9e3fmzZtHYuLFv614PB5cLhfx8fFMmTKFb7/9ltdeey1nR8//j/DwcEqWLMmBfXtQjhERn0jKOsi43W5mrf6Vh/v3z/XPtty4miBz4403Ur9ePaavXJL1BekuSFKQ8SYFmTx2PgukpaYSEhKSq3tdLhfffPMNnTp1unAydu/evUlOTmbOnIvP+Dh/MFqxYsV46KGH6N69Oy+++OIV1RwcHJzrFU8iIl4Tn5zlp7cfOUBiSjItW7bM54JyzjAMbmnZkvV7t2d9gdOVsYJJvEZBJp+4XC7s9txtpLxo0SLi4uK48847OXv2LGfPnqV+/fqUK1cu028L7777LuvWrWPx4sV06dKFGTNmMH78+Cuq1W6343I5/z+FiYjkl3RnxtBSFtbv/huAxo0bX/HjX3rpJerXr09oaCgVKlSgV69eHD169MLrt956K8uWLWP+/PkXtrUYOnRortqIiori70P7SEzJIpC5PNmuxpIroyMK8klAQABpablL4efDSr9+/ejXr99Fr8XFxXHixIkL/121alWaNGkCQOvWrWnatCmvvfYaDzzwQK57gtLS0ggICISCcYKCiBQmSWnZzo+J2bOd6lWrXeihvhInTpzglVdeoXz58sTFxTFy5EhatWrFtm3bsNvtREdH88ADDxAcHMwHH3wAQMWKuTsnKSoqCtM0+XPfTlrUyWKBRzY9TnJlFGTy2PksEBIaxunTWR96lpXk5GTmzp1L165deeqppy567dixY/Tq1YuZM2dy5513ZrrXz8+P9957j9tuu40JEybkaj8Zj8fD2bNnCQ0LU44RkfzndGe7lf/mA3to1Djyqh4/adKkC//udru56aabqFixIr/++iu33347ERERFClShNDQ0Cs+gDIiIgKHw8HmA3uyDjLpTnC7IQ/n+RQmGlrKY/b//gnXqVuPP//8M8f3zZ07l8TERJ588kluvfXWiz569uxJZGTkJSejtWvXjhYtWjBq1Kgs953Jzs6dO0lOTiaibv0LtYuI5BuPme3y5MS0FIoXL35Vj1+wYAHNmzenaNGi2O32C70tO3bsuKrn/i+Hw0FoSAgJWQ0tQcawvVZTeI3eqvKYYRjY/aBhZBOOHDly0VjspUybNo3KlStz6623Zvl6nz59WLNmTfbnegBvvvkmBw8eZOrUqTmuNyYmBoD6DRvj8FOfjIjkM9OT7fy8VGc6gYGBV/zodevWcdddd1G+fHmmTJnC6tWrWbNmTcazU727425AQACpzmymE5imDpD0Ig0t5QO7zaBBo4zJaTExMXTu3Pmy9/zwww+XfP2pp566MOSU3ZlI7dq1y/V5STExMVx3fRWKlyiB3aYgIyL5zMj+547Dz56rHuZ/mjNnDkWLFmXWrFnYbBm/x+/fn82J1VfJ6XTi8MvmLdYwQD9fvUY9MvnAYYNKla+jZMlwli9f7utyLmnZsmU0iIzCAOwavhWR/GYzsg0zgQ4HSUlJV/zolJQUHA4Hxv88P6sea39//6vqoTFNk+SUFIL8A7K+wABsevv1Fv1J5gOH3cBmGNx9by8mT56c69VL+SUmJoaYmBi63dcbmw31yIhI/vOzZdtbUb1MBf7auu2yj3C73cyePTvTR2RkJMeOHeOJJ57gl19+Yfjw4Xz55ZeZ7q9Tpw7r16/nhx9+YP369Rw5cgSAYcOGYbfbL9uLc/DgQVJSUqhWtkLWF9iy/xol9xRk8oHdZuBvN+jzyGPExcXx7bff+rqkLEVHR1OhYiXate9EkF3fZCLiA0H+4Mi6Oziqam1it2wmPT39ko9ITU3l3nvvzfRhs9l49913mTt3LnfddRfLly/nxx9/zHT/iy++yM0338xDDz1E06ZNmTBhApCxqtPtdl92yH79+vX/rTeb3dyLBKtHxosMM7eTKOSKpLpM4lM8dL/zNkxXOitXrvR1SRc5c+YMFSpU4KnnX+aZF18hPMSG7RJj1SIiecI0YdXfcOxMppd+/zuWFq8NZMOGDURGXt0y7Lz06quvMunTCRydMDfzizYDGlwP1crle13XKkXCfBJoN/CzQb8Bg/j9999zdIJ1fnrnnXdwu930fuhhAv87FCYiku8MA4plvYlno+trYLPZWLt2bT4XlTvr1q6lSZVsemP87RBy5SuvJDMFmXwU7DDodGdXWrVux4ABA4iPj/d1SQCsWbOGDz74gBdeeZMyZcoS5FCIEREfCg0kq42sQgKDaBnRiKlff+2DonLm8OHD/Lp0KR0a3Zj1BQ47hGQzCViuiIJMPgr2t+FvNxg5ZgJn4+NzteNuXklNTaVfv340jIxi4JPPEeRvaP8YEfGtkEDwd2T50qDb72bFypVs3rw5n4vKmc8++4xAhz8PtuqQ9QUBDghWj4w3KcjksyKBNipVrszQt9/niy++yHSKdX578cUX2bNnDx+Nm0igv51Qf4UYEfGxIkHZDr90vaElZUuEM27cuHwu6vKcTicTxo/nwZbtKRKcxfCYYUDpohkrs8Rr9KeZz+w2gxCHwf19+nNHl3vo3bs3v/32m09qeeeddxg9ejRDR3xA7doRhAXaLtpfQUTEJ/wdUK54lm/4DrudR9veyVdffsWxY8d8UFz2pk+fztFjxxh4+91ZXxASAKWL5WtNhYGCjA+EBNgIdNiI/nwKN9x0M507d2bJkiX51r5pmvznP//h5Zdf5rmXXqf/o4MI8Tfw15CSiBQUZYtl2yvzRMfuBDv8Gfj447nevTyvHD9+nGefeYZ7b2pDg+urZ31RyTAoEZq/hRUCCjI+UjTQICw4gMnT53DDTS244447mDhxYp5/UyYnJ/PEE0/wyiuv8MIrb/L8y28Q5G8QEqD/FUSkAAkLzhiGyUJ4kWKMG/Ac38+dy4wZM/K5sMxM02TQwIHY3CZjH3k264sCHFC2+CWPYJAro3cvHzEMg6JBNoqFBfPljDl07/kAjzzyCJ06deLQoUN50ubvv/9Oo0aNmDhxIu+OGsvzL71OSICNMIUYESmIyhbP2CAvC92atabHzW0ZMnhwjg/jzSszZ87kuzlziH7kOUoVzeZ07rCgbIOZXB29g/mQzTAoFmSjSHAAo8ZMYMqsufy5KZa6devy6aefeu001ri4OJ599lluueUWihYvyZLfY+jb/zFC/A2FGBEpuMKLZMwpyaYTY0z/Z/E3/Oh8xx0+285i1apV9H+4P71a3Eb3m1pnfVGAAyqXynYlllwdvYv52PmembBAg/Yd7+C3NZvo0LkrAwcOpFKlSrz00kvs3bs31881TZNVq1bxwAMPULFiRT799FPeGP4uc39eRu1atSgWbNNwkogUbH42qFEOima9QV54kWL8/MpI9uzcxR2dOuV7mPnjjz/o1LETTavWYuLAl7O+yCCjZ6lyeL7WVpjoiIICxO0xSUgzSXeZ7Nq5g68mjWfG1C85Fx/PLbfcQtOmTYmKiiIqKorq1atfOIYeIC0tjS1btlw4+HHlypVs27aNKtWq89DDj9Kjdx9KlixJsL9BiL+h1UkiYh17j8HWg5DmzPLlP3ZspcOI56haozo/zp9PuXJ5v/3/okWLuOfue2h0XTV+evmDrJdbQ8YuxU2qZxvG5OopyBRAqS6TVGdGoElKTub7b2fyy6IFbP5zAwf27wPAbrcTFBR04bj55ORkTNPEz8+PmrUjaBgZRddu99GydTv8/GwE2g2CHNrsTkQsyO2GjXvhwAnI5h0rdt8u2o94jnTTzegxY+jVq1ee/MKWlJTEyy+/zOjRo+kQ2Yxvn3+b4IBsNrgLcEDdSlClrNfrkP+nIFOAuTwZgSbFaWKaGd+/p06dZPOfG9izexdpaak409MJCAgkJDSU2nXqElG/IcHBwRhk9MoGOQwCHTo7SUQsLj4JYnbDmcRsLzl57ixDJn7IzN9/oWuXLoz79FPKlvVeiFi2bBkP9+vH0SNH+U/vx3iiY/eLesYvYjOgUimIrAJ+WZ/mLd6hIGMRLo+Jyw1Oj4nLbeI2Mw6JPc8wMjbbc9jAz++//7QpvIjINeToadhyAM4lX/Kyb9csZeBnI0l2pvPAgw8waNAgGjRocEVNulwufvjhB6LHjmXJL7/Qok5DJg16mRrlKmV/k2FkbOjXoIrOVcoHCjIiImIdh07CtoOQkHLJy06eO8vYhd8x4Zd5HDkVx83Nm/PAgw/StGlT6tevj79/1su6Ac6cOcOGDRtYvnw5Ez//nMNHjnBTrfoMbn83vVrcln0vDGRM7i1THBpcn7HkWvKcgoyIXJW0tDQ2b97M5s2bOXfuHKmpqRfmcFWpUoWoqChKly7t6zLlWnLoJPx16LI9MwBOl4t561cwduEclm3biMfjwd/fn/r16lG7Th2Cg4NxOBykpaVx5swZ/tywkT37MlaKhgWH0LN5WwbefjeRVWtevi7DyFihVL9yxoZ+ki8UZEQkV0zTZMWKFUz9+mvWrV/Pli1bcDqdGIZBUFAQAQEBuN1ukpOTcblcAFSsWJGoqChuv/12HnjgAYoUKeLjr0Is7+jpjDBzNjHbCcD/lJyWyqZ9u4jZ8zcxe7az+/gRUtLTcLpdBDr8CQsMpn7lqkRVrUVUtdrULFfp0r0v/8vulzGcVLdytkcrSN5QkBGRHElISGDKlClER0ezdetWqlWrRsuWLS9sCdCwYUOCgv6/K900Tfbu3XthS4C1a9eyfPlygoKCePCBBxg0eDD16tXz4VcklhefBDuPwrEz2S7NzhdhQXBdKbi+TMZKJclXCjIickmmafL555/z/PPPk5SURJcuXRg0aBBt2rTJ9fLWQ4cO8dlnnzFhwgSOHTtGt27diI6O1tCTXDm3Gw6chD3HMoJNfr6j2f0yjh2oXg5K6fgBX1GQEZFsHThwgAEDBrBo0SIe7tePt4YNo2LFilf9XKfTyYwZM3j22WcxTZOxY8dy3333aaNGuXLne2fi4iE5LW/bstkgNDBjt171wvicgoyIZGnGjBk8+uijFC1alM8++4wOHTp4vY0TJ04wePBgZs+ezb333ssXX3xBSIh2QJUr5PbAyXMZQ00n4iEpNeNz3uJvzxhGKl8SyhTVbr0FhIKMiGQyduxYhgwZQq9evRg3bhxFi+Ztt/msWbN4+OGHadiwIfPnz6dYsWJ52p4UAgnJcDwejpzOCDTpLnC5c/8cf3vGR/FQKF8iYwhJPTAFioKMiFwkOjqawYMH88wzzzBy5Mh8G+5Zu3YtHTt2pHr16ixZsoSwsLB8aVeucemujGXaSamQmApnkyAlHZyujF1Fz2+bbpCxfNrPBo7/9rwUDc5YgRQSmPHvGvoskBRkROSCWbNm0aNHj3wPMedt2LCB1q1b07RpUxYuXIjdbs/X9qWQSE6DlDRwm+AxwfRkzHuxGRkTeEMCwF+9LlahICMiABw8eJB69erRqVMnpk2b5rOJt8uWLaNNmza8/fbbvPTSSz6pQUSsQ0FGRDBNk06dOhEbG8vWrVt9PkflpZdeYtSoUWzYsIG6dev6tBYRKdgUZESESZMm0b9/f+bPn0+nTp18XQ6pqak0btyYkJAQVq9erSEmEcmWgoxIIZeUlETFihXp0qULkydP9nU5F6xdu5abbrqJcePG8eijj/q6HBEpoBRkRAq5zz77jMcff5w9e/Zw3XXX+bqci9x9993s2bOHP//8U5vliUiWFGRECjHTNImMjOS6665j7ty5vi4nk8WLF3P77bezYsUKWrRo4etyRKQAyuGxniJyLVq9ejWbNm1i0KBBmV4bOnQooaGhPqjq/7Vt25YaNWoQHR3t0zpEpOBSkBEpxKZNncp1113Hbbfd5utSsmSz2RgwYADffvstqampvi5HRAogBRmRQmzd+vXccsst2GwF90dBy5YtSU9PZ/Pmzb4uRUQKoIL700tE8pTT6WTTpk1ERUVd0f2bN2+mffv2hISEULRoUbp3786BAwcuusYwDN577z2GDh1KmTJlCA8Pp1+/fiQlJeW4nQYNGmC324mJibmiOkXk2qYgI1JIbdu2jbS0NJo0aZLrew8ePEjLli05deoUX3/9NZ9++ikbNmygVatWJCQkXHTtmDFj2LlzJ19++SVvvPEG06ZN49///neO2woKCqJu3brErF+f6zpF5NqnXaZECqnY2FgAGjVqlOt7R40ahdPpZNGiRZQoUQKAyMhIIiIimDx5Mk888cSFa8uVK8fUqVMB6NChAxs2bGD27Nm88847OW6vUaNGxGpoSUSyoB4ZkUIqISEBh8NxRSuTVqxYQZs2bS6EGIDatWvTsGFDVq5cedG1/5xIHBERwaFDh3LVXvHixTP19IiIgIKMSKGVlpZGQEDAFd175swZypQpk+nzZcqU4fTp0xd97p/nNvn7+5OWlpar9gIDA7VqSUSypCAjUkjZ7XZcLtcV3VuiRAlOnDiR6fPHjx+/qJfGW1wuFw6Hw+vPFRHrU5ARKaTO93K43e5c39uiRQt++eUXzpw5c+Fz27dvJzY2Nk924E1OTiYoKMjrzxUR69NkX5FCqmrVqgDs2LGDOnXqZHmN2+1m9uzZmT7/1FNP8cUXX3D77bfz6quvkpqaymuvvUblypXp27dvrurYv38/1apV44033uCNN97I8pq///77Qr0iIv9LQUakkGrcuDEAMTEx2QaZ1NRU7r333kyfnzJlCsuWLeP555/n/vvvx8/Pj9tuu40PP/yQsLCwXNVhmiZutxuPx5Pt6xs2bOD555/P1XNFpHDQoZEihVj16tW58847GTVqlK9Lydbu3bupXr06CxYsoEOHDr4uR0QKGM2RESnEoqKiWLt2ra/LuKR169YBXPEOxCJybVOQESnE2rdvz+rVq9mzZ4+vS8nW1KlTiYyMpFSpUr4uRUQKIAUZkUKsZ8+eFC1alPHjx/u6lCzt27eP+fPnM2jQIF+XIiIFlIKMSCEWHBxMv759mThxYoHccG78+PEUKVKEXr16+boUESmgFGRECrnHBw7k1KlTfPnll74u5SLx8fF8/vnn9O3Th5CQEF+XIyIFlFYtiQgPPfQQ8+bNY+vWrVSoUMHX5QDQ/+GH+Wb2bLZu3UqlSpV8XY6IFFDqkRERPvroI4KCgnjssccoCL/bLFiwgElffMGHH36oECMil6QeGREBYN68eXTp0oXJkyfTp08fn9Vx9uxZ6tWrR7169ViwYAGGYfisFhEp+BRkROSCPn36MGvWLH7++WdatmyZ7+2npKTQsWNHNm3aRGxsrHpjROSyNLQkIheMHz+e5s2b07lzZ1avXp2vbaekpNCtWzfWrVvHjz/+qBAjIjmiICMiFwQGBvL999/TqFEj2rVrx8KFC/Ol3TNnztChQweWLVvG3Llzufnmm/OlXRGxPgUZEblIWFgYCxcu5NZbb6Vjx448+eSTJCUl5Vl78+fPp169esTGxrJo0SLatWuXZ22JyLVHQUZEMgkODmbevHmMGjWKzz//nIYNG7J8+XKvtnH27Fn69etH586dadCgAbGxseqJEZFcU5ARkSz5+fnx9NNPs2nTJsqWLUurVq3o2LEjP/zwA263+4qfu3v3bl544QWqVavGd999x8SJE/npp580J0ZErohWLYnIZbndbqZOncqYMWNYt24dlStX5pFHHqFVq1ZERkYSFhaW7b0ej4cdO3awbt06pk2bxsKFCylevDgP9+vH0888Q8WKFfPxKxGRa42CjIjkyrp16xgXHc3MWbNITk7GMAxq1apFZGQkxYsXJzAwEJfLRXJyMtu3b2fjxo0kJiYC0LRpUwYNGkSPHj0ICgry8VciItcCBRkRuSIul4tt27YRExNDzPr1xG7eTEJCAikpKTgcDgIDA6latSpRUVFERUXRuHFjihcv7uuyReQaoyAjIiIilqXJviIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYlkKMiIiImJZCjIiIiJiWQoyIiIiYln/Bw2xdOM+3P7CAAAAAElFTkSuQmCC" + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "
" + ], + "image/png": "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" + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "
" + ], + "image/png": "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" + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "execution_count": 8 + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": [ + "**Interpretation of the SI graph plots.**\n", + "The first graph plot shows the SV similarly as the force plots and waterfall plots (AveOccup) is the most influential and negative (large blue node size).\n", + "The *Latitude* (Lat.) feature is positive (red large-ish node). \n", + "In the 2-SII graph plot, again all interactions are plotted. This representation is the same as the network plot. Only the scaling can be a bit different as sizes are differently computed.\n", + "Notably, in the last graph plot for the Möbius transform (the full functional decomposition), all interactions are plotted. \n", + "Next to the first- and second-order interactions hyper-edges connecting various features can be seen.\n", + "For example, there is a sizable positive third order interaction between Longitude (Lon.), Latitude (Lat.), and MedianIncome (MI). \n", + "A positive fourth order interaction including the same three features exists with the HouseAge (HA) feature.\n", + "In summary, the Möbius graph plot shows how many interactions are necessary to fully explain the model's prediction at this point. \n", + "Still, it remains a challenge to interpret the graph plot in its entirety.\n" + ] + }, + { + "metadata": {}, + "cell_type": "markdown", + "source": [ + "### Bar Plot\n", + "While all of the previous plot can theoretically be drawn for global explanations as well, the bar plot is specifically designed for global explanations.\n", + "The bar plot from `shap`, shows the mean absolute value of the Shapley values for each feature.\n", + "In `shapiq` this is similar in that the bar plot shows the mean absolute value of the Shapley interactions for each interaction of features." + ] + }, + { + "metadata": { + "ExecuteTime": { + "end_time": "2024-10-24T12:32:19.966897Z", + "start_time": "2024-10-24T12:25:34.901599Z" + } + }, + "cell_type": "code", + "source": [ + "explanations = []\n", + "explainer = shapiq.TreeExplainer(model=model, max_order=2, index=\"k-SII\")\n", + "for instance_id in tqdm(range(20)):\n", + " x_explain = x_test[instance_id]\n", + " si = explainer.explain(x=x_explain)\n", + " explanations.append(si)" + ], + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "100%|██████████| 20/20 [06:39<00:00, 19.95s/it]\n" + ] + } + ], + "execution_count": 9 + }, + { + "metadata": { + "ExecuteTime": { + "end_time": "2024-10-24T12:32:20.139865Z", + "start_time": "2024-10-24T12:32:19.968887Z" + } + }, + "cell_type": "code", + "source": "shapiq.plot.bar_plot(explanations, feature_names=feature_names, show=True)", + "outputs": [ + { + "data": { + "text/plain": [ + "
" + ], + "image/png": "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" + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "execution_count": 10 + } + ], + "metadata": { + "kernelspec": { + "display_name": "shapiq2", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.9" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/shapiq/games/base.py b/shapiq/games/base.py index 956544ff..f8e01953 100644 --- a/shapiq/games/base.py +++ b/shapiq/games/base.py @@ -4,7 +4,7 @@ import pickle import warnings from abc import ABC -from typing import Optional +from typing import Optional, Union import numpy as np from tqdm.auto import tqdm @@ -94,6 +94,7 @@ def __init__( normalization_value: Optional[float] = None, path_to_values: Optional[str] = None, verbose: bool = False, + player_names: Optional[list[str]] = None, *args, **kwargs, ) -> None: @@ -134,6 +135,13 @@ def __init__( # define some handy coalition variables self.empty_coalition = np.zeros(self.n_players, dtype=bool) self.grand_coalition = np.ones(self.n_players, dtype=bool) + self._empty_coalition_value_property = None + self._grand_coalition_value_property = None + + # define player_names + self.player_name_lookup: dict[str, int] = ( + {name: i for i, name in enumerate(player_names)} if player_names is not None else None + ) self.verbose = verbose @@ -157,7 +165,105 @@ def is_normalized(self) -> bool: """Checks if the game is normalized/centered.""" return self(self.empty_coalition) == 0 - def __call__(self, coalitions: np.ndarray, verbose: bool = False) -> np.ndarray: + def _check_coalitions( + self, + coalitions: Union[np.ndarray, list[Union[tuple[int], tuple[str]]]], + ) -> np.ndarray: + """ + Check if the coalitions are in the correct format and convert them to one-hot encoding. + The format may either be a numpy array containg the coalitions in one-hot encoding or a list of tuples with integers or strings. + Args: + coalitions: The coalitions to convert to one-hot encoding. + Returns: + np.ndarray: The coalitions in the correct format + Raises: + TypeError: If the coalitions are not in the correct format. + Examples: + >>> coalitions = np.asarray([[1, 0, 0, 0], [0, 1, 1, 0]]) + >>> coalitions = [(0, 1), (1, 2)] + >>> coalitions = [()] + >>> coalitions = [(0, 1), (1, 2), (0, 1, 2)] + if player_name_lookup is not None: + >>> coalitions = [("Alice", "Bob"), ("Bob", "Charlie")] + Wrong format: + >>> coalitions = [1, 0, 0, 0] + >>> coalitions = [(1,"Alice")] + >>> coalitions = np.array([1,-1,2]) + + + """ + error_message = ( + "List may only contain tuples of integers or strings." + "The tuples are not allowed to have heterogeneous types." + "Reconcile the docs for correct format of coalitions." + ) + + if isinstance(coalitions, np.ndarray): + + # Check that coalition is contained in array + if len(coalitions) == 0: + raise TypeError("The array of coalitions is empty.") + + # Check if single coalition is correctly given + if coalitions.ndim == 1: + if len(coalitions) < self.n_players or len(coalitions) > self.n_players: + raise TypeError( + "The array of coalitions is not correctly formatted." + f"It should have a length of {self.n_players}" + ) + coalitions = coalitions.reshape((1, self.n_players)) + + # Check that all coalitions have the correct number of players + if coalitions.shape[1] != self.n_players: + raise TypeError( + f"The number of players in the coalitions ({coalitions.shape[1]}) does not match " + f"the number of players in the game ({self.n_players})." + ) + + # Check that values of numpy array are either 0 or 1 + if not np.all(np.logical_or(coalitions == 0, coalitions == 1)): + raise TypeError("The values in the array of coalitions are not binary.") + + return coalitions + + # We now assume to work with list of tuples + if isinstance(coalitions, tuple): + # if by any chance a tuple was given wrap into a list + coalitions = [coalitions] + + try: + # convert list of tuples to one-hot encoding + coalitions = transform_coalitions_to_array(coalitions, self.n_players) + + return coalitions + except Exception as err: + # It may either be the tuples contain strings or wrong format + if self.player_name_lookup is not None: + # We now assume the tuples to contain strings + try: + coalitions = [ + ( + tuple(self.player_name_lookup[player] for player in coalition) + if coalition != tuple() + else tuple() + ) + for coalition in coalitions + ] + coalitions = transform_coalitions_to_array(coalitions, self.n_players) + + return coalitions + except Exception as err: + raise TypeError(error_message) from err + + raise TypeError(error_message) from err + + def __call__( + self, + coalitions: Union[ + np.ndarray, list[Union[tuple[int], tuple[str]]], tuple[Union[int, str]], str + ], + verbose: bool = False, + ) -> np.ndarray: """Calls the game's value function with the given coalitions and returns the output of the value function. @@ -168,9 +274,8 @@ def __call__(self, coalitions: np.ndarray, verbose: bool = False) -> np.ndarray: Returns: The values of the coalitions. """ - # check if coalitions are correct dimensions - if coalitions.ndim == 1: - coalitions = coalitions.reshape((1, self.n_players)) + # check if coalitions are correct format + coalitions = self._check_coalitions(coalitions) verbose = verbose or self.verbose @@ -194,7 +299,13 @@ def _lookup_coalitions(self, coalitions: np.ndarray) -> np.ndarray: for i, coalition in enumerate(coalitions): # convert one-hot vector to tuple coalition_tuple = tuple(np.where(coalition)[0]) - values[i] = self.value_storage[self.coalition_lookup[coalition_tuple]] + try: + values[i] = self.value_storage[self.coalition_lookup[coalition_tuple]] + except KeyError as error: + raise KeyError( + f"The coalition {coalition_tuple} is not stored in the game. " + f"Are all values pre-computed?" + ) from error return values def value_function(self, coalitions: np.ndarray) -> np.ndarray: @@ -404,3 +515,36 @@ def get_game_name(cls) -> str: break parent = parent.__base__ return "_".join(class_names) + + @property + def empty_coalition_value(self) -> float: + """Return the value of the empty coalition.""" + if self._empty_coalition_value_property is None: + self._empty_coalition_value_property = float(self(self.empty_coalition)) + return self._empty_coalition_value_property + + @property + def grand_coalition_value(self) -> float: + """Return the value of the grand coalition.""" + if self._grand_coalition_value_property is None: + self._grand_coalition_value_property = float(self(self.grand_coalition)) + return self._grand_coalition_value_property + + def __getitem__(self, item: tuple[int, ...]): + """Return the value of the given coalition. Only retrieves precomputed/store values. + + Args: + item: The coalition to evaluate. + + Returns: + The value of the coalition + + Raises: + KeyError: If the coalition is not stored in the game. + """ + try: + return self.value_storage[self.coalition_lookup[tuple(sorted(item))]] + except (KeyError, IndexError) as error: + raise KeyError( + f"The coalition {item} is not stored in the game. Is it precomputed?" + ) from error diff --git a/shapiq/games/benchmark/unsupervised_cluster/base.py b/shapiq/games/benchmark/unsupervised_cluster/base.py index c005b9b3..1b597ce9 100644 --- a/shapiq/games/benchmark/unsupervised_cluster/base.py +++ b/shapiq/games/benchmark/unsupervised_cluster/base.py @@ -71,12 +71,11 @@ def __init__( self.data = data self.random_state = random_state - self.empty_cluster_value = empty_cluster_value super().__init__( data.shape[1], normalize=normalize, - normalization_value=self.empty_cluster_value, + normalization_value=0, verbose=verbose, ) @@ -93,7 +92,7 @@ def value_function(self, coalitions: np.ndarray) -> np.ndarray: worth = np.zeros(n_coalitions, dtype=float) for i, coalition in enumerate(coalitions): if sum(coalition) == 0: - worth[i] = self.empty_cluster_value + worth[i] = 0.0 continue data_selection = self.data[:, coalition] self.cluster.fit(data_selection) diff --git a/shapiq/games/benchmark/unsupervised_data/base.py b/shapiq/games/benchmark/unsupervised_data/base.py index d3149f4b..f5a7ad3a 100644 --- a/shapiq/games/benchmark/unsupervised_data/base.py +++ b/shapiq/games/benchmark/unsupervised_data/base.py @@ -23,7 +23,6 @@ class UnsupervisedData(Game): def __init__(self, data: np.ndarray, verbose: bool = False, *args, **kwargs) -> None: self.data = data self._n_features = data.shape[1] - self.empty_coalition_value = 0.0 # discretize the data from sklearn.preprocessing import KBinsDiscretizer diff --git a/shapiq/interaction_values.py b/shapiq/interaction_values.py index 33de3498..6ea1b45c 100644 --- a/shapiq/interaction_values.py +++ b/shapiq/interaction_values.py @@ -579,20 +579,21 @@ def to_dict(self) -> dict: "baseline_value": self.baseline_value, } - def plot_network(self, **kwargs) -> tuple[plt.Figure, plt.Axes]: + def plot_network(self, show: bool = True, **kwargs) -> Optional[tuple[plt.Figure, plt.Axes]]: """Visualize InteractionValues on a graph. For arguments, see shapiq.plots.network_plot(). Returns: - matplotlib.pyplot.Figure, matplotlib.pyplot.Axes + """ - from shapiq import network_plot + from shapiq.plot.network import network_plot if self.max_order > 1: return network_plot( first_order_values=self.get_n_order_values(1), second_order_values=self.get_n_order_values(2), + show=show, **kwargs, ) else: @@ -601,22 +602,32 @@ def plot_network(self, **kwargs) -> tuple[plt.Figure, plt.Axes]: "but requires also 2-order values for the network plot." ) - def plot_stacked_bar(self, **kwargs) -> tuple[plt.Figure, plt.Axes]: + def plot_si_graph(self, show: bool = True, **kwargs) -> Optional[tuple[plt.Figure, plt.Axes]]: + """Visualize InteractionValues as a SI graph. + + For arguments, see shapiq.plots.si_graph_plot(). + + Returns: + The SI graph as a tuple containing the figure and the axes. + """ + + from shapiq.plot.si_graph import si_graph_plot + + return si_graph_plot(self, show=show, **kwargs) + + def plot_stacked_bar( + self, show: bool = True, **kwargs + ) -> Optional[tuple[plt.Figure, plt.Axes]]: """Visualize InteractionValues on a graph. For arguments, see shapiq.plots.stacked_bar_plot(). Returns: - matplotlib.pyplot.Figure, matplotlib.pyplot.Axes + The stacked bar plot as a tuple containing the figure and the axes. """ from shapiq import stacked_bar_plot - ret = stacked_bar_plot( - self, - **kwargs, - ) - - return ret + return stacked_bar_plot(self, show=show, **kwargs) def plot_force( self, @@ -639,6 +650,9 @@ def plot_force( matplotlib: Whether to return a ``matplotlib`` figure. Defaults to ``True``. show: Whether to show the plot. Defaults to ``False``. **kwargs: Keyword arguments passed to ``shap.plots.force()``. + + Returns: + The force plot as a matplotlib figure (if show is ``False``). """ from shapiq import force_plot @@ -655,7 +669,7 @@ def plot_waterfall( self, feature_names: Optional[np.ndarray] = None, feature_values: Optional[np.ndarray] = None, - show: bool = False, + show: bool = True, max_display: int = 10, ) -> Optional[plt.Axes]: """Draws interaction values on a waterfall plot. diff --git a/shapiq/plot/__init__.py b/shapiq/plot/__init__.py index 6f10b964..404a8c7f 100644 --- a/shapiq/plot/__init__.py +++ b/shapiq/plot/__init__.py @@ -5,6 +5,7 @@ from .network import network_plot from .si_graph import si_graph_plot from .stacked_bar import stacked_bar_plot +from .utils import abbreviate_feature_names, get_interaction_values_and_feature_names from .watefall import waterfall_plot __all__ = [ @@ -14,4 +15,7 @@ "force_plot", "bar_plot", "waterfall_plot", + # utils + "abbreviate_feature_names", + "get_interaction_values_and_feature_names", ] diff --git a/shapiq/plot/bar.py b/shapiq/plot/bar.py index 281440df..01ec8b78 100644 --- a/shapiq/plot/bar.py +++ b/shapiq/plot/bar.py @@ -15,9 +15,9 @@ def bar_plot( list_of_interaction_values: list[InteractionValues], feature_names: Optional[np.ndarray] = None, - show: bool = True, + show: bool = False, **kwargs, -) -> plt.Axes: +) -> Optional[plt.Axes]: """Draws interaction values on a bar plot. Requires the ``shap`` Python package to be installed. @@ -55,11 +55,8 @@ def bar_plot( feature_names=_labels, ) - if show: - ax = shap.plots.bar(explanation, **kwargs, show=False) - ax.set_xlabel("mean(|Shapley Interaction value|)") - plt.show() - else: - ax = shap.plots.bar(explanation, **kwargs, show=False) - ax.set_xlabel("mean(|Shapley Interaction value|)") + ax = shap.plots.bar(explanation, **kwargs, show=False) + ax.set_xlabel("mean(|Shapley Interaction value|)") + if not show: return ax + plt.show() diff --git a/shapiq/plot/force.py b/shapiq/plot/force.py index afc4e7ef..38f48cf7 100644 --- a/shapiq/plot/force.py +++ b/shapiq/plot/force.py @@ -18,7 +18,7 @@ def force_plot( feature_names: Optional[np.ndarray] = None, feature_values: Optional[np.ndarray] = None, matplotlib: bool = True, - show: bool = True, + show: bool = False, **kwargs, ) -> Optional[plt.Figure]: """Draws interaction values on a force plot. @@ -37,26 +37,15 @@ def force_plot( check_import_module("shap") import shap - if interaction_values.max_order == 1: - return shap.plots.force( - base_value=np.array([interaction_values.baseline_value], dtype=float), # must be array - shap_values=interaction_values.get_n_order_values(1), - features=feature_values, - feature_names=feature_names, - matplotlib=matplotlib, - show=show, - **kwargs, - ) - else: - _shap_values, _labels = get_interaction_values_and_feature_names( - interaction_values, feature_names, feature_values - ) - - return shap.plots.force( - base_value=np.array([interaction_values.baseline_value], dtype=float), # must be array - shap_values=np.array(_shap_values), - feature_names=_labels, - matplotlib=matplotlib, - show=show, - **kwargs, - ) + _shap_values, _labels = get_interaction_values_and_feature_names( + interaction_values, feature_names, feature_values + ) + + return shap.plots.force( + base_value=np.array([interaction_values.baseline_value], dtype=float), # must be array + shap_values=np.array(_shap_values), + feature_names=_labels, + matplotlib=matplotlib, + show=show, + **kwargs, + ) diff --git a/shapiq/plot/network.py b/shapiq/plot/network.py index 7f814c2c..c0e2c246 100644 --- a/shapiq/plot/network.py +++ b/shapiq/plot/network.py @@ -29,7 +29,8 @@ def network_plot( center_image_size: Optional[float] = 0.6, draw_legend: bool = True, center_text: Optional[str] = None, -) -> tuple[plt.Figure, plt.Axes]: + show: bool = False, +) -> Optional[tuple[plt.Figure, plt.Axes]]: """Draws the interaction network. An interaction network is a graph where the nodes represent the features and the edges represent @@ -59,9 +60,11 @@ def network_plot( center_image_size: The size of the center image. Defaults to ``0.6``. draw_legend: Whether to draw the legend. Defaults to ``True``. center_text: The text to be displayed in the center of the network. Defaults to ``None``. + show: Whether to show the plot. Defaults to ``False``. If ``False``, the figure and the axis + containing the plot are returned, otherwise ``None``. Returns: - The figure and the axis containing the plot. + The figure and the axis containing the plot if ``show=False``. """ fig, axis = plt.subplots(figsize=(6, 6)) axis.axis("off") @@ -175,7 +178,9 @@ def network_plot( if draw_legend: _add_legend_to_axis(axis) - return fig, axis + if not show: + return fig, axis + plt.show() def _add_weight_to_edges_in_graph( diff --git a/shapiq/plot/si_graph.py b/shapiq/plot/si_graph.py index daaec3d7..2f2c4489 100644 --- a/shapiq/plot/si_graph.py +++ b/shapiq/plot/si_graph.py @@ -18,6 +18,9 @@ ADJUST_NODE_ALPHA = True +__all__ = ["si_graph_plot"] + + def _normalize_value( value: float, max_value: float, base_value: float, cubic_scaling: bool = False ) -> float: @@ -310,14 +313,14 @@ def _adjust_position( def si_graph_plot( interaction_values: InteractionValues, - graph: Union[list[tuple], nx.Graph], + graph: Optional[Union[list[tuple], nx.Graph]] = None, n_interactions: Optional[int] = None, draw_threshold: float = 0.0, random_seed: int = 42, size_factor: float = 1.0, plot_explanation: bool = True, - compactness: float = 1.0, - label_mapping: Optional[dict] = None, + compactness: float = 1e10, + feature_names: Optional[list] = None, cubic_scaling: bool = False, pos: Optional[dict] = None, node_size_scaling: float = 1.0, @@ -326,7 +329,8 @@ def si_graph_plot( spring_k: Optional[float] = None, interaction_direction: Optional[str] = None, node_area_scaling: bool = False, -) -> tuple[plt.figure, plt.axis]: + show: bool = False, +) -> Optional[tuple[plt.figure, plt.axis]]: """Plots the interaction values as an explanation graph. An explanation graph is an undirected graph where the nodes represent players and the edges @@ -338,7 +342,8 @@ def si_graph_plot( interaction_values: The interaction values to plot. graph: The underlying graph structure as a list of edge tuples or a networkx graph. If a networkx graph is provided, the nodes are used as the players and the edges are used as - the connections between the players. + the connections between the players. Defaults to ``None``, which creates a graph with + all nodes from the interaction values without any edges between them. n_interactions: The number of interactions to plot. If ``None``, all interactions are plotted according to the draw_threshold. draw_threshold: The threshold to draw an edge (i.e. only draw explanations with an @@ -351,8 +356,8 @@ def si_graph_plot( compactness: A scaling factor for the underlying spring layout. A higher compactness value will move the interactions closer to the graph nodes. If your graph looks weird, try adjusting this value, e.g. ``[0.1, 1.0, 10.0, 100.0, 1000.0]``. Defaults to ``1.0``. - label_mapping: A mapping from the player/node indices to the player label. If ``None``, the - player indices are used as labels. Defaults to ``None``. + feature_names: A list of feature names to use for the nodes in the graph. If ``None``, + the feature indices are used instead. Defaults to ``None``. cubic_scaling: Whether to scale the size of explanations cubically (``True``) or linearly (``False``, default). Cubic scaling puts more emphasis on larger interactions in the plot. Defaults to ``False``. @@ -372,14 +377,19 @@ def si_graph_plot( interactions are plotted. Possible values are ``"positive"`` and ``"negative"``. Defaults to ``None``. node_area_scaling: TODO add docstring. + show: Whether to show or return the plot. Defaults to ``False``. Returns: - The figure and axis of the plot. + The figure and axis of the plot if ``show`` is ``True``. Otherwise, ``None``. """ normal_node_size = NORMAL_NODE_SIZE * node_size_scaling base_size = BASE_SIZE * node_size_scaling + label_mapping = None + if feature_names is not None: + label_mapping = {i: feature_names[i] for i in range(len(feature_names))} + # fill the original graph with the edges and nodes if isinstance(graph, nx.Graph): original_graph = graph @@ -389,7 +399,7 @@ def si_graph_plot( for node in graph_nodes: node_label = label_mapping.get(node, node) if label_mapping is not None else node original_graph.nodes[node]["label"] = node_label - else: + elif isinstance(graph, list): original_graph, graph_nodes = nx.Graph(), [] for edge in graph: original_graph.add_edge(*edge) @@ -399,6 +409,12 @@ def si_graph_plot( original_graph.add_node(edge[0], label=nodel_labels[0]) original_graph.add_node(edge[1], label=nodel_labels[1]) graph_nodes.extend([edge[0], edge[1]]) + else: # graph is considered None + original_graph = nx.Graph() + graph_nodes = list(range(interaction_values.n_players)) + for node in graph_nodes: + node_label = label_mapping.get(node, node) if label_mapping is not None else node + original_graph.add_node(node, label=node_label) if n_interactions is not None: # get the top n interactions @@ -500,4 +516,6 @@ def si_graph_plot( ax.set_aspect("equal", adjustable="datalim") # make y- and x-axis scales equal ax.axis("off") # remove axis - return fig, ax + if not show: + return fig, ax + plt.show() diff --git a/shapiq/plot/stacked_bar.py b/shapiq/plot/stacked_bar.py index 374c60f5..a782030d 100644 --- a/shapiq/plot/stacked_bar.py +++ b/shapiq/plot/stacked_bar.py @@ -21,6 +21,7 @@ def stacked_bar_plot( title: Optional[str] = None, xlabel: Optional[str] = None, ylabel: Optional[str] = None, + show: bool = False, ): """Plot the n-SII values for a given instance. @@ -43,6 +44,7 @@ def stacked_bar_plot( title (str): The title of the plot. xlabel (str): The label of the x-axis. ylabel (str): The label of the y-axis. + show (bool): Whether to show the plot. Defaults to ``False``. Returns: tuple[matplotlib.figure.Figure, matplotlib.axes.Axes]: A tuple containing the figure and @@ -147,4 +149,6 @@ def stacked_bar_plot( plt.tight_layout() - return fig, axis + if not show: + return fig, axis + plt.show() diff --git a/shapiq/plot/utils.py b/shapiq/plot/utils.py index aee0f0cd..0c7b10b0 100644 --- a/shapiq/plot/utils.py +++ b/shapiq/plot/utils.py @@ -1,5 +1,7 @@ """This utility module contains helper functions for plotting.""" +import copy +from collections.abc import Iterable from typing import Optional import numpy as np @@ -7,7 +9,7 @@ from shapiq.interaction_values import InteractionValues from shapiq.utils import powerset -__all__ = ["get_interaction_values_and_feature_names"] +__all__ = ["get_interaction_values_and_feature_names", "abbreviate_feature_names"] def get_interaction_values_and_feature_names( @@ -26,6 +28,9 @@ def get_interaction_values_and_feature_names( Returns: A tuple containing the SHAP values and the corresponding labels. """ + feature_names = copy.deepcopy(feature_names) + if feature_names is not None: + feature_names = abbreviate_feature_names(feature_names) _values_dict = {} for i in range(1, interaction_values.max_order + 1): _values_dict[i] = interaction_values.get_n_order_values(i) @@ -39,7 +44,7 @@ def get_interaction_values_and_feature_names( _values = _values_dict[_order] _shap_values.append(_values[interaction]) if feature_names is not None: - _name = " x ".join(f"{feature_names[i]}".strip()[0:4] + "." for i in interaction) + _name = " x ".join(str(feature_names[i]) for i in interaction) else: _name = " x ".join(f"{feature}" for feature in interaction) if feature_values is not None: @@ -49,3 +54,36 @@ def get_interaction_values_and_feature_names( _shap_values = np.array(_shap_values) _labels = np.array(_labels) return _shap_values, _labels + + +def abbreviate_feature_names(feature_names: Iterable[str]) -> list[str]: + """A rudimentary function to abbreviate feature names for plotting. + + Args: + feature_names: The feature names to be abbreviated. + + Returns: + list[str]: The abbreviated feature names. + """ + abbreviated_names = [] + for name in feature_names: + name = str(name) + name = name.strip() + capital_letters = sum(1 for c in name if c.isupper()) + seperator_chars = (" ", "_", "-", ".") + is_seperator_in_name = any([c in seperator_chars for c in name[:-1]]) + if is_seperator_in_name: + for seperator in seperator_chars: + name = name.replace(seperator, ".") + name_parts = name.split(".") + new_name = "" + for part in name_parts: + if part: + new_name += part[0].upper() + abbreviated_names.append(new_name) + elif capital_letters > 1: + new_name = "".join([c for c in name if c.isupper()]) + abbreviated_names.append(new_name[0:3]) + else: + abbreviated_names.append(name.strip()[0:3] + ".") + return abbreviated_names diff --git a/tests/tests_games/test_base_game.py b/tests/tests_games/test_base_game.py index 8d285305..5e3ef654 100644 --- a/tests/tests_games/test_base_game.py +++ b/tests/tests_games/test_base_game.py @@ -5,10 +5,96 @@ import numpy as np import pytest +from shapiq.games.base import Game from shapiq.games.benchmark import DummyGame # used to test the base class from shapiq.utils.sets import powerset, transform_coalitions_to_array +def test_call(): + """This test tests the call function of the base game class.""" + + class TestGame(Game): + """This is a test game class that inherits from the base game class. + Its value function is the amount of players divided by the number of players. + """ + + def __init__(self, n, **kwargs): + super().__init__(n_players=n, normalization_value=0, **kwargs) + + def value_function(self, coalition): + return np.sum(coalition) / self.n_players + + n_players = 6 + test_game = TestGame( + n=n_players, player_names=["Alice", "Bob", "Charlie", "David", "Eve", "Frank"] + ) + + # assert that player names are correctly stored + assert test_game.player_name_lookup == { + "Alice": 0, + "Bob": 1, + "Charlie": 2, + "David": 3, + "Eve": 4, + "Frank": 5, + } + + assert test_game([]) == 0.0 + + # test coalition calls with wrong datatype + with pytest.raises(TypeError): + assert test_game([(0, 1), "Alice", "Charlie"]) + with pytest.raises(TypeError): + assert test_game([(0, 1), ("Alice",), ("Bob",)]) + with pytest.raises(TypeError): + assert test_game(("Alice", 1)) + + # test wrong coalition size in call + with pytest.raises(TypeError): + assert test_game(np.array([True, False, True])) == 0.0 + with pytest.raises(TypeError): + assert test_game(np.array([])) == 0.0 + + # test wrong method for numpy array values + with pytest.raises(TypeError): + assert test_game(np.array([1, 2, 3, 4, 5, 6])) == 0.0 + + # test wrong coalition size in shape[1] + with pytest.raises(TypeError): + assert test_game(np.array([[True, False, True]])) == 0.0 + + # test with empty coalition all call variants + test_coalition = test_game.empty_coalition + assert test_game(test_coalition) == 0.0 + assert test_game(()) == 0.0 + assert test_game([()]) == 0.0 + + # test with grand coalition all call variants + test_coalition = test_game.grand_coalition + assert test_game(test_coalition) == 1.0 + assert test_game(tuple(range(0, test_game.n_players))) == 1.0 + assert test_game([tuple(range(0, test_game.n_players))]) == 1.0 + assert test_game(tuple(test_game.player_name_lookup.values())) == 1.0 + assert test_game([tuple(test_game.player_name_lookup.values())]) == 1.0 + + # test with single player coalition all call variants + test_coalition = np.array([True] + [False for _ in range(test_game.n_players - 1)]) + assert test_game(test_coalition) - 1 / 6 < 10e-7 + assert test_game((0,)) - 1 / 6 < 10e-7 + assert test_game([(0,)]) - 1 / 6 < 10e-7 + assert test_game(("Alice",)) - 1 / 6 < 10e-7 + assert test_game([("Alice",)]) - 1 / 6 < 10e-7 + + # test string calls with missing player names + test_game2 = TestGame(n=n_players) + with pytest.raises(TypeError): + assert test_game2("Alice") == 0.0 + with pytest.raises(TypeError): + assert test_game2(("Bob",)) == 0.0 + with pytest.raises(TypeError): + assert test_game2([("Charlie",)]) == 0.0 + + def test_precompute(): """This test tests the precompute function of the base game class""" n_players = 6 @@ -34,15 +120,25 @@ def test_precompute(): dummy_game.precompute(coalitions=coalitions) assert dummy_game.n_values_stored == 1 + with pytest.raises(KeyError): # test error case where not all values are precomputed + _ = dummy_game(dummy_game.empty_coalition) + # test with large number of players and see if it raises a warning with pytest.warns(UserWarning): n_players_large = 17 dummy_game_large = DummyGame(n=n_players_large) - # call precompute but stop it before it finishes dummy_game_large.precompute() - assert dummy_game_large.n_values_stored == 2**n_players_large + # test empty and grand coalition lookup + dummy_game = DummyGame(n=4, interaction=(0, 1)) + dummy_game.precompute() + assert dummy_game.empty_coalition_value == dummy_game(dummy_game.empty_coalition) + assert dummy_game.grand_coalition_value == dummy_game(dummy_game.grand_coalition) + assert dummy_game[(0, 1)] == dummy_game[(1, 0)] != 0.0 + with pytest.raises(KeyError): + _ = dummy_game[(0, 9)] # only 4 players + def test_core_functions(): """This test tests the core functions of the base game class object.""" diff --git a/tests/tests_plots/test_bar.py b/tests/tests_plots/test_bar.py index 589e0566..4e980045 100644 --- a/tests/tests_plots/test_bar.py +++ b/tests/tests_plots/test_bar.py @@ -22,3 +22,8 @@ def test_bar_plot(interaction_values_list: list[InteractionValues]): axis = bar_plot(interaction_values_list, show=True) assert axis is None plt.close() + + # test show=True + output = bar_plot(interaction_values_list, show=True) + assert output is None + plt.close("all") diff --git a/tests/tests_plots/test_force.py b/tests/tests_plots/test_force.py index b105077c..c5b8dd44 100644 --- a/tests/tests_plots/test_force.py +++ b/tests/tests_plots/test_force.py @@ -36,3 +36,8 @@ def test_force_plot(interaction_values_list: list[InteractionValues]): fp = iv.plot_force(show=False) assert isinstance(fp, plt.Figure) plt.close() + + # test show=True + output = iv.plot_force(show=True) + assert output is None + plt.close("all") diff --git a/tests/tests_plots/test_network_plot.py b/tests/tests_plots/test_network_plot.py index f5eff0b3..7e3e1037 100644 --- a/tests/tests_plots/test_network_plot.py +++ b/tests/tests_plots/test_network_plot.py @@ -53,6 +53,11 @@ def test_network_plot(): with pytest.raises(ValueError): network_plot() + # test show=True + output = network_plot(interaction_values=iv, show=True) + assert output is None + plt.close("all") + assert True diff --git a/tests/tests_plots/test_plot_utils.py b/tests/tests_plots/test_plot_utils.py new file mode 100644 index 00000000..591f28bb --- /dev/null +++ b/tests/tests_plots/test_plot_utils.py @@ -0,0 +1,24 @@ +"""This module contains tests for the plot.utils module.""" + +from shapiq.plot.utils import abbreviate_feature_names + + +def test_abbreviate(): + """Tests the abbreviate_feature_names function.""" + + # test with all cases + feature_names = [ + # seperators + "feature_A", + "feature-B", + "feature.C", + "feature D", + # seperators with extra dot at the end should not be included + "feature E.", + # capital letters + "CapitalLetters", + # normal base case + "longlowercase", + ] + expected = ["FA", "FB", "FC", "FD", "FE", "CL", "lon."] + assert abbreviate_feature_names(feature_names) == expected diff --git a/tests/tests_plots/test_si_graph.py b/tests/tests_plots/test_si_graph.py index 181f742e..6b9baa1e 100644 --- a/tests/tests_plots/test_si_graph.py +++ b/tests/tests_plots/test_si_graph.py @@ -67,6 +67,17 @@ def test_si_graph_plot( ) graph_tuple = [(1, 2), (2, 3), (3, 4), (2, 4), (1, 4)] + # test without graph and from interaction values + fig, ax = example_values.plot_si_graph(show=False) + assert isinstance(fig, plt.Figure) + assert isinstance(ax, plt.Axes) + plt.close(fig) + + # test with show=True + output = example_values.plot_si_graph(show=True) + assert output is None + plt.close("all") + fig, ax = si_graph_plot( example_values, graph_tuple, @@ -106,12 +117,7 @@ def test_si_graph_plot( plot_explanation=True, n_interactions=n_interactions, compactness=compactness, - label_mapping={ - 0: "A", - 1: "B", - 2: "C", - 3: "D", - }, + feature_names=["A", "B", "C", "D"], ) assert isinstance(fig, plt.Figure) @@ -149,12 +155,7 @@ def test_si_graph_plot( adjust_node_pos=True, interaction_direction="negative", min_max_interactions=(-0.5, 0.5), - label_mapping={ - 0: "A", - 1: "B", - 2: "C", - 3: "D", - }, + feature_names=["A", "B", "C", "D"], ) assert isinstance(fig, plt.Figure) diff --git a/tests/tests_plots/test_stacked_bar.py b/tests/tests_plots/test_stacked_bar.py index bfef22c6..77af99af 100644 --- a/tests/tests_plots/test_stacked_bar.py +++ b/tests/tests_plots/test_stacked_bar.py @@ -50,3 +50,12 @@ def test_stacked_bar_plot(): assert isinstance(fig, plt.Figure) assert isinstance(axes, plt.Axes) plt.close() + + # test show=True + output = stacked_bar_plot( + interaction_values=interaction_values, + feature_names=feature_names, + show=True, + ) + assert output is None + plt.close("all") diff --git a/tests/tests_plots/test_waterfall.py b/tests/tests_plots/test_waterfall.py index 2213c4ca..bedc047e 100644 --- a/tests/tests_plots/test_waterfall.py +++ b/tests/tests_plots/test_waterfall.py @@ -36,3 +36,8 @@ def test_waterfall_plot(interaction_values_list: list[InteractionValues]): wp = iv.plot_waterfall(show=False) assert isinstance(wp, plt.Axes) plt.close() + + # test show=True + output = iv.plot_waterfall(show=True) + assert output is None + plt.close("all")