{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# COMPSCI 389: Introduction to Machine Learning\n", "# Topic 5.3 Evaluation Re-Visited... Again\n", "\n", "**Note:** This notebook is described in the slides, `5.3 Evaluation Part 3.pdf`. All of the important content within this notebook is in those slides, so you are not responsible for this notebook. However, you may reference this notebook to run the examples from the slides.\n", "\n", "The code below should be review. It:\n", "1. Imports the libraries we use\n", "2. Defines the evaluation metrics we use\n", "3. Defines the KNearestNeighbors model\n", "4. Defines the WeightedKNearestNeighbors model" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "from sklearn.neighbors import KDTree\n", "from sklearn.base import BaseEstimator\n", "from sklearn.model_selection import train_test_split\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "\n", "def mean_squared_error(predictions, labels):\n", " return np.mean((predictions - labels) ** 2)\n", "\n", "def root_mean_squared_error(predictions, labels):\n", " return np.sqrt(mean_squared_error(predictions, labels))\n", "\n", "def mean_absolute_error(predictions, labels):\n", " return np.mean(np.abs(predictions - labels))\n", "\n", "def r_squared(predictions, labels):\n", " ss_res = np.sum((labels - predictions) ** 2) # ss_res is the \"Sum of Squares of Residuals\"\n", " ss_tot = np.sum((labels - np.mean(labels)) ** 2) # ss_tot is the \"Total Sum of Squares\"\n", " return 1 - (ss_res / ss_tot)\n", "\n", "class KNearestNeighbors(BaseEstimator):\n", " # Add a constructor that stores the value of k (a hyperparameter)\n", " def __init__(self, k=3):\n", " self.k = k\n", "\n", " def fit(self, X, y):\n", " # Convert X and y to NumPy arrays if they are DataFrames\n", " if isinstance(X, pd.DataFrame):\n", " X = X.values\n", " if isinstance(y, pd.Series):\n", " y = y.values\n", "\n", " # Store the training data and labels\n", " self.X_data = X\n", " self.y_data = y\n", " \n", " # Create a KDTree for efficient nearest neighbor search\n", " self.tree = KDTree(X)\n", "\n", " return self\n", "\n", " def predict(self, X):\n", " # Convert X to a NumPy array if it's a DataFrame\n", " if isinstance(X, pd.DataFrame):\n", " X = X.values\n", "\n", " # Query the tree for the k nearest neighbors for all points in X\n", " dist, ind = self.tree.query(X, k=self.k)\n", "\n", " # Return the average label for the nearest neighbors of each query\n", " return np.mean(self.y_data[ind], axis=1)\n", " \n", "class WeightedKNearestNeighbors(BaseEstimator):\n", " # Add a constructor that stores the value of k and sigma (hyperparameters)\n", " def __init__(self, k=3, sigma=1.0):\n", " self.k = k\n", " self.sigma = sigma\n", "\n", " def fit(self, X, y):\n", " # Convert X and y to NumPy arrays if they are DataFrames\n", " if isinstance(X, pd.DataFrame):\n", " X = X.values\n", " if isinstance(y, pd.Series):\n", " y = y.values\n", "\n", " # Store the training data and labels\n", " self.X_data = X\n", " self.y_data = y\n", " \n", " # Create a KDTree for efficient nearest neighbor search\n", " self.tree = KDTree(X)\n", "\n", " return self\n", "\n", " def gaussian_kernel(self, distance):\n", " # Gaussian kernel function\n", " return np.exp(- (distance ** 2) / (2 * self.sigma ** 2))\n", "\n", " def predict(self, X):\n", " # Convert X to a NumPy array if it's a DataFrame\n", " if isinstance(X, pd.DataFrame):\n", " X = X.values\n", "\n", " # We will iteratively load predictions, so it starts empty\n", " predictions = []\n", " \n", " # Loop over rows in the query\n", " for x in X:\n", " # Query the tree for the k nearest neighbors\n", " dist, ind = self.tree.query([x], k=self.k)\n", "\n", " # Calculate weights using the Gaussian kernel\n", " weights = self.gaussian_kernel(dist[0])\n", "\n", " # Check if weights sum to zero. This happens when all points are very far, giving weights that round to zero, causing divison by zero later. In this case, revert to un-weighted (all weights are one).\n", " if np.sum(weights) == 0:\n", " # If weights sum to zero, assign equal weight to all neighbors\n", " weights = np.ones_like(weights)\n", "\n", " # Weighted average of the labels of the k nearest neighbors\n", " weighted_avg_label = np.average(self.y_data[ind[0]], weights=weights)\n", " predictions.append(weighted_avg_label)\n", "\n", " # Return the array of predictions we have created\n", " return np.array(predictions)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's run NearestNeighbor, evaluating using different amounts of test data (from 5% of the test data to 100% of the test data). We will run this several times to visualize how the sample MSE varies across runs.\n", "\n", "First load the data set, split it (with most points for evaluation this time), and train a model to get a single model that we will evaluate. Then get the predictions and errors on the testing set." ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "'Squared errors: '" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/plain": [ "4235 0.024545\n", "24036 3.712057\n", "2760 0.000278\n", "17666 0.017777\n", "3382 1.496536\n", " ... \n", "34459 0.170842\n", "14526 2.667767\n", "20521 1.102500\n", "40100 0.041343\n", "28412 0.384400\n", "Name: gpa, Length: 34643, dtype: float64" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Load the data set\n", "df = pd.read_csv(\"data/GPA.csv\", delimiter=',')\n", "\n", "# We already loaded X and y, but do it again as a reminder\n", "X = df.iloc[:, :-1]\n", "y = df.iloc[:, -1]\n", "\n", "# Split the data into training and testing sets (80% train, 20% test)\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.8, shuffle=True) # Let's use most of the data for evalution in this case\n", "\n", "# Train the NearestNeighbor model\n", "model = KNearestNeighbors(k=1)\n", "model.fit(X_train, y_train)\n", "\n", "# Compute predictions for X_test\n", "predictions = model.predict(X_test)\n", "\n", "# Compute the sample squared errors\n", "squared_errors = (predictions - y_test) ** 2\n", "display(\"Squared errors: \", squared_errors)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Model Evaluation\n", "\n", "Let's consider the problem of evaluating the mean squared error (MSE) of this one model that we have learned. This is independent of the method used to train this model (it could have come from anywhere)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's create a function that takes in a variable `percentage`, and which then computes the sample MSE from only `percentage`% of the provided squared errors." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "# Function to compute average MSE for a given percentage of squared errors\n", "def compute_average_mse(percentage, squared_errors):\n", " # Get the number of samples that we should use from squared_errors\n", " subset_size = int(percentage * len(squared_errors))\n", "\n", " # Randomly select that many indices (without replacement)\n", " indices = np.random.choice(len(squared_errors), subset_size, replace=False)\n", "\n", " # Get the average of the squared errors at the selected indices\n", " average_mse = squared_errors.iloc[indices].mean()\n", " \n", " return average_mse" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now let's use this to plot the sample MSE using different amounts of the test data. This simulates what we would see if we had different amounts of data in the test set." ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Percentages to test, from 5% to 100% in increments of 5%\n", "percentages = np.arange(0.05, 1.05, 0.05)\n", "\n", "# Compute average MSE values for each percentage\n", "average_mse_values = [compute_average_mse(p, squared_errors) for p in percentages]\n", "\n", "# Plotting\n", "plt.plot(percentages * 100, average_mse_values, marker='o')\n", "plt.ylim(1.05, 1.25)\n", "plt.xlabel('Percentage of Test Set Used (%)')\n", "plt.ylabel('Sample MSE')\n", "plt.title('Sample MSE vs. Percentage of Test Set Used')\n", "plt.grid(True)\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "If we run the block above a few times, we see significant variation in the sample MSEs.\n", "\n", "Let's overlay many runs of this plot:" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Plotting 100 times\n", "for _ in range(100):\n", " average_mse_values = [compute_average_mse(p, squared_errors) for p in percentages]\n", " plt.plot(percentages * 100, average_mse_values, marker='o', alpha=0.2) # alpha for transparency\n", "\n", "plt.ylim(1.05, 1.25) # Setting y-axis limits\n", "plt.xlabel('Percentage of Test Set Used (%)')\n", "plt.ylabel('Sample MSE')\n", "plt.title('Sample MSE vs. Percentage of Test Set Used')\n", "plt.grid(True)\n", "plt.show()\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The tables that we presented before were showing single points from these plots! The plot above gives us an idea how much the predictions vary, and hence how much we should trust them.\n", "\n", "Let's capture this using error bars that show the standard error for each point. Notice that we can do this from a single run (we don't require 100 runs with different test sets)." ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkgAAAHHCAYAAABEEKc/AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAABup0lEQVR4nO3deVhU1f8H8PeAw74oyqps4oK4r0hqQoqIRpqmuSVuleZupdHXBNQ0NVMrcykVLbco1zQRF9xXFJcyU0JJBXfZFBzg/P7gx80ZQAaYYRh4v55nHmfOPffeMx+uM58599xzZUIIASIiIiKSGOi6AUREREQVDRMkIiIiIhVMkIiIiIhUMEEiIiIiUsEEiYiIiEgFEyQiIiIiFUyQiIiIiFQwQSIiIiJSwQSJiIiISAUTJKqyZDIZwsLCdN0MogrnzJkzeOWVV2Bubg6ZTIa4uDhdN4k0ICwsDDKZTNfN0BtMkKhMLl26hLfeeguurq4wMTFB7dq14e/vj2+++UbXTSt3bm5ukMlk6Nq1a6HLv//+e8hkMshkMpw9e1Zp2dGjRxEYGIjatWvDxMQELi4uCAoKwoYNG5Tq5a9f2GP06NFae2+lFRMTo9RGuVyOunXrYujQofjnn3903bwyu3PnDsLCwipVAqFQKNCvXz88evQIixYtwo8//ghXV9cC9fKP9+IeERERGmnXnDlzsG3bNrXr379/HxMnToSnpydMTU1hZ2eHdu3aYdq0aUhPTy/x/o8fP46wsDA8efJErfq+vr5o0qRJocsePHjAH2h6oJquG0D66/jx4/Dz84OLiwveffddODg44N9//8XJkyexZMkSjB8/XtdNLHcmJiY4ePAgkpOT4eDgoLRs/fr1MDExQWZmplJ5ZGQk3n77bbRo0QITJ05EjRo1kJCQgMOHD+P777/HoEGDlOr7+/tj6NChBfbdoEEDzb8hDZkwYQLatm0LhUKBc+fOYeXKldi1axcuXboEJycnXTev1O7cuYPw8HC4ubmhRYsWum6ORsTHx+PmzZv4/vvvMWrUqCLrLV68WCnR2L17NzZu3IhFixahVq1aUvkrr7yikXbNmTMHb731Fnr37l1s3UePHqFNmzZITU3FiBEj4OnpiYcPH+LixYtYtmwZxowZAwsLixLt//jx4wgPD8ewYcNQvXr10r0J0itMkKjUPv/8c1hbW+PMmTMFPjDu3bunm0bpWIcOHXDmzBls3rwZEydOlMpv3bqFI0eO4M0338Svv/6qtE5YWBi8vLxw8uRJGBkZKS0rLI4NGjTAkCFDtPMGtKRTp0546623AADDhw9HgwYNMGHCBKxduxYhISFl2nZGRgbMzc010UzCf8dccUmAaqKSnJyMjRs3onfv3nBzc9NO49S0atUqJCYm4tixYwUStNTU1AL/z4gKw1NsVGrx8fFo3LhxoR+kdnZ2Sq/XrFmD1157DXZ2djA2NoaXlxeWLVtWYD03Nze8/vrriImJQZs2bWBqaoqmTZsiJiYGALBlyxY0bdoUJiYmaN26Nc6fP6+0/rBhw2BhYYF//vkHAQEBMDc3h5OTE2bOnAkhRLHv6fbt2xgxYgTs7e1hbGyMxo0bY/Xq1WrHxMTEBH369Clwamzjxo2oUaMGAgICCqwTHx+Ptm3bFvqhrRrH0vrll18gk8lw6NChAstWrFgBmUyGy5cvA8j7ohs+fDjq1KkDY2NjODo6olevXrhx44ZG2gIAr732GgAgISFBKvv999/RqVMnmJubw9LSEj179sQff/yhtF7+3zc+Ph49evSApaUlBg8eDADIzc3FkiVLpOPD1tYW3bt3L3A686effkLr1q1hamoKGxsbDBgwAP/++69SnfzTI3/++Sf8/PxgZmaG2rVrY/78+VKdmJgYtG3bFkBe0qd6SunIkSPo168fXFxcYGxsDGdnZ0yePBnPnj0rEI/IyEh4eXnBxMQETZo0wdatWzFs2LACiUZubi4WL16Mxo0bw8TEBPb29nj//ffx+PFjteJ+4MABKcbVq1dHr169cOXKFaX4du7cGQDQr18/yGQy+Pr6qrXtoqgT72vXrqFv375wcHCAiYkJ6tSpgwEDBiAlJQVA3qnljIwMrF27VorzsGHDitxnfHw8DA0N0b59+wLLrKysYGJiolR26tQpdO/eHdbW1jAzM0Pnzp1x7NgxaXlYWBg+/vhjAIC7u7vUBk3+n0hLS8OkSZPg5uYGY2Nj2NnZwd/fH+fOnStRW/MdPXoUbdu2hYmJCTw8PLBixQqNtbWqYA8SlZqrqytOnDiBy5cvF3muPd+yZcvQuHFjvPHGG6hWrRp27tyJDz74ALm5uRg7dqxS3evXr2PQoEF4//33MWTIEHz55ZcICgrC8uXL8emnn+KDDz4AAMydOxf9+/fH1atXYWDwX66fk5OD7t27o3379pg/fz727NmD0NBQZGdnY+bMmUW28e7du2jfvj1kMhnGjRsHW1tb/P777xg5ciRSU1MxadIkteIyaNAgdOvWDfHx8fDw8AAAbNiwAW+99Rbkcnmhcdy/fz9u3bqFOnXqFLv9zMxMPHjwoEC5lZVVkb+Me/bsCQsLC/z888/SF2C+zZs3o3HjxtLfsG/fvvjjjz8wfvx4uLm54d69e4iOjkZiYqLGegbi4+MBADVr1gQA/PjjjwgODkZAQADmzZuHp0+fYtmyZejYsSPOnz+vtN/s7GwEBASgY8eO+PLLL2FmZgYAGDlyJCIiIhAYGIhRo0YhOzsbR44cwcmTJ9GmTRsAeb2en332Gfr3749Ro0bh/v37+Oabb/Dqq6/i/PnzSsn+48eP0b17d/Tp0wf9+/fHL7/8gmnTpqFp06YIDAxEo0aNMHPmTMyYMQPvvfceOnXqBOC/U0qRkZF4+vQpxowZg5o1a+L06dP45ptvcOvWLURGRkr72bVrF95++200bdoUc+fOxePHjzFy5EjUrl27QNzef/99REREYPjw4ZgwYQISEhLw7bff4vz58zh27Fihx1e+ffv2ITAwEHXr1kVYWBiePXuGb775Bh06dMC5c+fg5uaG999/H7Vr18acOXOk06L29val+AtD7Xg/f/4cAQEByMrKwvjx4+Hg4IDbt2/jt99+w5MnT2BtbY0ff/wRo0aNQrt27fDee+8BgPR/qzCurq7IycmRjquXOXDgAAIDA9G6dWuEhobCwMBA+kF35MgRtGvXDn369MHff/9d4BSira1tqWOjavTo0fjll18wbtw4eHl54eHDhzh69CiuXLmCVq1aqd1WIG9saLdu3WBra4uwsDBkZ2cjNDS0TH/LKkkQldLevXuFoaGhMDQ0FD4+PmLq1KkiKipKPH/+vEDdp0+fFigLCAgQdevWVSpzdXUVAMTx48elsqioKAFAmJqaips3b0rlK1asEADEwYMHpbLg4GABQIwfP14qy83NFT179hRGRkbi/v37UjkAERoaKr0eOXKkcHR0FA8ePFBq04ABA4S1tXWh70G17T179hTZ2dnCwcFBzJo1SwghxJ9//ikAiEOHDok1a9YIAOLMmTPSeqtWrRIAhJGRkfDz8xOfffaZOHLkiMjJySmwDwBFPjZu3PjS9g0cOFDY2dmJ7OxsqSwpKUkYGBiImTNnCiGEePz4sQAgFixY8NJtqevgwYMCgFi9erW4f/++uHPnjti1a5dwc3MTMplMnDlzRqSlpYnq1auLd999V2nd5ORkYW1trVSe//f95JNPlOoeOHBAABATJkwo0Ibc3FwhhBA3btwQhoaG4vPPP1dafunSJVGtWjWl8s6dOwsAYt26dVJZVlaWcHBwEH379pXKzpw5IwCINWvWFNhvYcfL3LlzhUwmUzqOmzZtKurUqSPS0tKkspiYGAFAuLq6SmVHjhwRAMT69euVtrlnz55Cy1W1aNFC2NnZiYcPH0plFy5cEAYGBmLo0KFSWf7fLDIy8qXbU7VgwQIBQCQkJAgh1I/3+fPn1dqfubm5CA4OVqstycnJwtbWVgAQnp6eYvTo0WLDhg3iyZMnSvVyc3NF/fr1RUBAgHScCJH3t3N3dxf+/v5Fvr/idO7cWTRu3LjQZffv3y/w+WNtbS3Gjh1b5PZK0tbevXsLExMTpePszz//FIaGhoJf++rjKTYqNX9/f5w4cQJvvPEGLly4gPnz5yMgIAC1a9fGjh07lOqamppKz1NSUvDgwQN07twZ//zzj9SNns/Lyws+Pj7Sa29vbwB5p2VcXFwKlBd2NdS4ceOk5/k9Qs+fP8e+ffsKfS9CCPz6668ICgqCEAIPHjyQHgEBAUhJSSnQ1V0UQ0ND9O/fHxs3bgSQNzjb2dlZ6mFQNWLECOzZswe+vr44evQoZs2ahU6dOqF+/fo4fvx4gfq9evVCdHR0gYefn99L2/X222/j3r170ulKIO/UW25uLt5++20AeX8nIyMjxMTEqH3aRh0jRoyAra0tnJyc0LNnT+l0SZs2bRAdHY0nT55g4MCBSnE3NDSEt7c3Dh48WGB7Y8aMUXr966+/QiaTITQ0tEDd/Muat2zZgtzcXPTv319pPw4ODqhfv36B/VhYWCiN9TIyMkK7du3UvvruxWM+IyMDDx48wCuvvAIhhHRq+M6dO7h06RKGDh2qNGi4c+fOaNq0qdL2IiMjYW1tDX9/f6X2t27dGhYWFoXGKV9SUhLi4uIwbNgw2NjYSOXNmjWDv78/du/erdZ7Kgl1421tbQ0AiIqKwtOnTzWyb3t7e1y4cAGjR4/G48ePsXz5cgwaNAh2dnaYNWuWdLo9Li4O165dw6BBg/Dw4UOpjRkZGejSpQsOHz6M3NxcjbSpONWrV8epU6dw586dQper29acnBxERUWhd+/eSp+XjRo1KvQUPxWNp9ioTNq2bYstW7bg+fPnuHDhArZu3YpFixbhrbfeQlxcHLy8vAAAx44dQ2hoKE6cOFHgQzAlJUX6kASg9J8a+O8D1NnZudBy1S9yAwMD1K1bV6ks/wqvosYM3L9/H0+ePMHKlSuxcuXKQuuUZOD5oEGD8PXXX+PChQvYsGEDBgwY8NL5RwICAhAQEICnT58iNjYWmzdvxvLly/H666/jr7/+UhqLVKdOnSKnEniZ/HELmzdvRpcuXQDknV5r0aKFFB9jY2PMmzcPH374Iezt7dG+fXu8/vrrGDp0aIGr8kpixowZ6NSpEwwNDVGrVi00atQI1arlffxcu3YNwH/jklRZWVkpva5WrVqBU5Hx8fFwcnJS+vJXde3aNQghUL9+/UKXq56eqlOnToG/WY0aNXDx4sUi9/GixMREzJgxAzt27ChwjOb/KLh58yYAoF69egXWr1evnlJSfu3aNaSkpBQ5Lu1lx2f+fho2bFhgWaNGjRAVFaXxwe7qxtvd3R1TpkzBV199hfXr16NTp0544403MGTIEKXPhZJydHTEsmXL8N133+HatWuIiorCvHnzMGPGDDg6OmLUqFHSsfey03ApKSmoUaNGqdvxMi8eX/Pnz0dwcDCcnZ3RunVr9OjRA0OHDpU+y9Rta1ZWFp49e1Zo3Bs2bKiVZLiyYoJEGmFkZIS2bduibdu2aNCgAYYPH47IyEiEhoYiPj4eXbp0gaenJ7766is4OzvDyMgIu3fvxqJFiwr8QjM0NCx0H0WVCzUGXxcnvw1Dhgwp8gOoWbNmam/P29sbHh4emDRpEhISEgpcql8UMzMzdOrUCZ06dUKtWrUQHh6O33//vdhxFOowNjZG7969sXXrVnz33Xe4e/cujh07hjlz5ijVmzRpEoKCgrBt2zZERUXhs88+w9y5c3HgwAG0bNmyVPtu2rRpkUldfux//PHHQpOw/ETqxffx4pgzdeXm5kImk+H3338v9FhSvey7LMdbTk4O/P398ejRI0ybNg2enp4wNzfH7du3MWzYsFL1SuTm5sLOzg7r168vdLkmx8NoQknivXDhQgwbNgzbt2/H3r17MWHCBMydOxcnT55Ua1zey8hkMjRo0AANGjRAz549Ub9+faxfvx6jRo2S/g4LFiwocpqGkk4HkM/ExKTQAfkApB+JLw4W79+/Pzp16oStW7di7969WLBgAebNm4ctW7YgMDBQ7bZmZWWVqr1UEBMk0rj8AbFJSUkAgJ07dyIrKws7duxQ6h162SmBssjNzcU///yjNC/Q33//DQBFDjK2tbWFpaUlcnJyStU7U5iBAwdi9uzZaNSoUanmyFGNoya8/fbbWLt2Lfbv348rV65ACCGdXnuRh4cHPvzwQ3z44Ye4du0aWrRogYULF+Knn37SWFte3BeQd8VeaWPv4eGBqKgoPHr0qMheJA8PDwgh4O7urrE5o4rqFbx06RL+/vtvrF27VmnOqujoaKV6+RMwXr9+vcA2VMs8PDywb98+dOjQQen0nTry93P16tUCy/766y/UqlVL41MllDTeTZs2RdOmTTF9+nQcP34cHTp0wPLlyzF79mwARce6JOrWrYsaNWpI/6fyjz0rK6tij72S7t/V1RUHDhzAs2fPCvy98v8OqhNwOjo64oMPPsAHH3yAe/fuoVWrVvj8888RGBiodlttbW1hamoq9TgVtl9SD8cgUakdPHiw0F/T+V24+d35+b8eX6ybkpKCNWvWaK1t3377rfRcCIFvv/0WcrlcOrWkytDQEH379sWvv/4qXe7+ovv375e4DaNGjUJoaCgWLlz40nr79+8vtFw1jprQtWtX2NjYYPPmzdi8eTPatWsHd3d3afnTp08LTGTp4eEBS0tLpV+mSUlJ+Ouvv6BQKMrcpoCAAFhZWWHOnDmFbk+d2Pft2xdCCISHhxdYln/c9enTB4aGhggPDy9w3Aoh8PDhwxK3PT+pUJ1dubBjXgiBJUuWKNVzcnJCkyZNsG7dOqVJFw8dOoRLly4p1e3fvz9ycnIwa9asAu3Izs5+6QzPjo6OaNGiBdauXatU7/Lly9i7dy969Ojx0vdZGurGOzU1FdnZ2UrLmzZtCgMDA6VjztzcXO1ZrE+dOoWMjIwC5adPn8bDhw+l/1OtW7eGh4cHvvzyy0Jn137x2Cvqb12UHj16QKFQFLi8Pjc3F8uWLYORkZH0eZSTk1NgLKadnR2cnJykGKjbVkNDQwQEBGDbtm1ITEyUll+5cgVRUVFqtZ3ysAeJSm38+PF4+vQp3nzzTXh6euL58+c4fvw4Nm/eDDc3NwwfPhwA0K1bNxgZGSEoKAjvv/8+0tPT8f3338POzk6jvSP5TExMsGfPHgQHB8Pb2xu///47du3ahU8//fSlpyG++OILHDx4EN7e3nj33Xfh5eWFR48e4dy5c9i3bx8ePXpUona4urqqdSuBXr16wd3dHUFBQfDw8EBGRgb27duHnTt3om3btggKClKq//fffxfak2Nvbw9/f/+X7ksul6NPnz7YtGkTMjIy8OWXXxbYdpcuXdC/f394eXmhWrVq2Lp1K+7evYsBAwZI9UJCQrB27VokJCSU+dJ/KysrLFu2DO+88w5atWqFAQMGwNbWFomJidi1axc6dOiglPAWxs/PD++88w6+/vprXLt2Dd27d0dubi6OHDkCPz8/jBs3Dh4eHpg9ezZCQkJw48YN9O7dG5aWlkhISMDWrVvx3nvv4aOPPipR2z08PFC9enUsX74clpaWMDc3h7e3Nzw9PeHh4YGPPvoIt2/fhpWVFX799ddCB77PmTMHvXr1QocOHTB8+HA8fvwY3377LZo0aaL0Rdi5c2e8//77mDt3LuLi4tCtWzfI5XJcu3YNkZGRWLJkiTQZZ2EWLFiAwMBA+Pj4YOTIkdJl/tbW1lq55YW68T5w4ADGjRuHfv36oUGDBsjOzsaPP/4o/WjJ17p1a+zbtw9fffUVnJyc4O7uLl2ooerHH3/E+vXr8eabb6J169YwMjLClStXsHr1apiYmODTTz8FkDde8YcffkBgYCAaN26M4cOHo3bt2rh9+zYOHjwIKysr7Ny5U9o/APzvf//DgAEDIJfLERQUVGTPW1BQELp164bJkyfj9OnTeOWVV/D06VPs2LEDx44dw+zZs6XPo7S0NNSpUwdvvfUWmjdvDgsLC+zbtw9nzpyRfmCVpK3h4eHYs2cPOnXqhA8++ADZ2dn45ptv0LhxY7XH0BF4vR+V3u+//y5GjBghPD09hYWFhTAyMhL16tUT48ePF3fv3lWqu2PHDtGsWTNhYmIi3NzcxLx588Tq1asLXDabf6m8KgAFLoFNSEgocEl6cHCwMDc3F/Hx8aJbt27CzMxM2Nvbi9DQ0AKXzUPlMlshhLh7964YO3ascHZ2FnK5XDg4OIguXbqIlStXFhuPotr+osIu89+4caMYMGCA8PDwEKampsLExER4eXmJ//3vfyI1NbVAm4t6dO7cudg2CiFEdHS0ACBkMpn4999/lZY9ePBAjB07Vnh6egpzc3NhbW0tvL29xc8//6xUL/9y++IueS7JJeMHDx4UAQEBwtraWpiYmAgPDw8xbNgwcfbsWaX9mpubF7p+dna2WLBggfD09BRGRkbC1tZWBAYGitjYWKV6v/76q+jYsaMwNzcX5ubmwtPTU4wdO1ZcvXpVqlPUJdrBwcFKl94LIcT27duFl5eXqFatmtIl/3/++afo2rWrsLCwELVq1RLvvvuuuHDhQqHTAmzatEl4enoKY2Nj0aRJE7Fjxw7Rt29f4enpWaANK1euFK1btxampqbC0tJSNG3aVEydOlXcuXPnZeEVQgixb98+0aFDB2FqaiqsrKxEUFCQ+PPPP5XqaOoy/3zFxfuff/4RI0aMEB4eHsLExETY2NgIPz8/sW/fPqXt/PXXX+LVV18VpqamAsBLL/m/ePGi+Pjjj0WrVq2EjY2NqFatmnB0dBT9+vUT586dK1D//Pnzok+fPqJmzZrC2NhYuLq6iv79+4v9+/cr1Zs1a5aoXbu2MDAwUOv4z8zMFGFhYdLf1tzcXLRv31789NNPSvWysrLExx9/LJo3by4sLS2Fubm5aN68ufjuu+9K3dZDhw6J1q1bCyMjI1G3bl2xfPlyERoaysv8S0AmhAZGuBJVEMOGDcMvv/xSqptRElUkLVq0gK2tbYFxS0RUPjgGiYhIhxQKRYExODExMbhw4UKZb/NBRKXHMUhERDp0+/ZtdO3aFUOGDIGTkxP++usvLF++HA4ODhg9erSum0dUZTFBIiLSoRo1aqB169b44YcfcP/+fZibm6Nnz5744osvpHvVEVH50+kptsOHDyMoKAhOTk6QyWTYtm3bS+tv2bIF/v7+sLW1hZWVFXx8fApcthgWFibdaTn/4enpqVQnMzMTY8eORc2aNWFhYYG+ffvi7t27mn57pAMREREcf0R6JX9281u3biErKwuPHj1CZGTkS2/GSkTap9MEKSMjA82bN8fSpUvVqn/48GHpvkGxsbHw8/NDUFCQdF+jfI0bN0ZSUpL0OHr0qNLyyZMnY+fOnYiMjMShQ4dw584d9OnTR2Pvi4iIiPRbhbmKTSaTYevWrejdu3eJ1mvcuDHefvttzJgxA0BeD9K2bdsQFxdXaP2UlBTY2tpiw4YN0pwhf/31Fxo1aoQTJ06gffv2ZXkbREREVAno9Rik3NxcpKWlFbi1wLVr1+Dk5AQTExP4+Phg7ty50i0uYmNjoVAolKZq9/T0hIuLy0sTpKysLKVZXXNzc/Ho0SPUrFlTI1PgExERkfYJIZCWlgYnJ6eX3tdRrxOk/CnX+/fvL5V5e3sjIiICDRs2RFJSEsLDw9GpUydcvnwZlpaWSE5OhpGREapXr660LXt7eyQnJxe5r7lz5xZ6GwMiIiLSP//+++9Lb4astwnShg0bEB4eju3bt8POzk4qDwwMlJ43a9YM3t7ecHV1xc8//4yRI0eWen8hISGYMmWK9DolJQUuLi5ISEiApaVlqbdLJaNQKHDw4EH4+flBLpfrujlVCmOvO4y9bjDuuqPN2KelpcHd3b3Y7269TJA2bdqEUaNGITIystg7MFevXh0NGjSQ7ozt4OCA58+f48mTJ0q9SHfv3oWDg0OR2zE2NoaxsXGBchsbG1hZWZXujVCJKRQKmJmZoWbNmvzAKmeMve4w9rrBuOuONmOfv73ihsfo3UzaGzduxPDhw7Fx40b07Nmz2Prp6emIj4+Ho6MjgLwbDsrlcqU7qF+9ehWJiYnw8fHRWruJiIhIf+i0Byk9PV3q2QGAhIQExMXFwcbGBi4uLggJCcHt27exbt06AHmn1YKDg7FkyRJ4e3tLY4ZMTU1hbW0NAPjoo48QFBQEV1dX3LlzB6GhoTA0NMTAgQMB5M05MnLkSEyZMkXq/Rk/fjx8fHx4BRsREREB0HGCdPbsWfj5+Umv88f4BAcHIyIiAklJSUhMTJSWr1y5EtnZ2Rg7dizGjh0rlefXB4Bbt25h4MCBePjwIWxtbdGxY0ecPHkStra2Uv1FixbBwMAAffv2RVZWFgICAvDdd99p+d0SERGRvtBpguTr64uXTcOUn/Tki4mJKXabmzZtKraOiYkJli5dqvYElURERFS16N0YJCIiIiJtY4JEREREpIIJEhEREZEKJkhEREREKpggEREREalggkRERESkggkSERERkQomSEREREQqmCARERERqWCCRERERKSCCRIRERGRCiZIRERERCqYIBERERGpYIJEREREpIIJEhEREZEKJkhEREREKpggEREREalggkRERESkggkSERERkQomSEREREQqmCARERERqWCCRERERKSCCRIRERGRCiZIRERERCqYIBERERGpYIJEREREpIIJEhEREZEKJkhEREREKpggEREREalggkRERESkggkSERERkQomSEREREQqmCARERERqWCCRERERKSCCRIRERGRCp0mSIcPH0ZQUBCcnJwgk8mwbdu2l9bfsmUL/P39YWtrCysrK/j4+CAqKkqpzty5c9G2bVtYWlrCzs4OvXv3xtWrV5Xq+Pr6QiaTKT1Gjx6t6bdHREREekqnCVJGRgaaN2+OpUuXqlX/8OHD8Pf3x+7duxEbGws/Pz8EBQXh/PnzUp1Dhw5h7NixOHnyJKKjo6FQKNCtWzdkZGQobevdd99FUlKS9Jg/f75G3xsRERHpr2q63HlgYCACAwPVrr948WKl13PmzMH27duxc+dOtGzZEgCwZ88epToRERGws7NDbGwsXn31VanczMwMDg4OpW88ERERVVo6TZDKKjc3F2lpabCxsSmyTkpKCgAUqLN+/Xr89NNPcHBwQFBQED777DOYmZkVuZ2srCxkZWVJr1NTUwEACoUCCoWiLG+DSiA/1ox5+WPsdYex1w3GXXe0GXt1t6nXCdKXX36J9PR09O/fv9Dlubm5mDRpEjp06IAmTZpI5YMGDYKrqyucnJxw8eJFTJs2DVevXsWWLVuK3NfcuXMRHh5eoHzv3r0vTaxIO6Kjo3XdhCqLsdcdxl43GHfd0Ubsnz59qlY9mRBCaHzvpSCTybB161b07t1brfobNmzAu+++i+3bt6Nr166F1hkzZgx+//13HD16FHXq1ClyWwcOHECXLl1w/fp1eHh4FFqnsB4kZ2dnPHjwAFZWVmq1mcpOoVAgOjoa/v7+kMvlum5OlcLY6w5jrxuMu+5oM/apqamoVasWUlJSXvr9rZc9SJs2bcKoUaMQGRlZZHI0btw4/Pbbbzh8+PBLkyMA8Pb2BoCXJkjGxsYwNjYuUC6Xy/kfRwcYd91h7HWHsdcNxl13tBF7dbendwnSxo0bMWLECGzatAk9e/YssFwIgfHjx2Pr1q2IiYmBu7t7sduMi4sDADg6Omq6uURERKSHdJogpaen4/r169LrhIQExMXFwcbGBi4uLggJCcHt27exbt06AHmn1YKDg7FkyRJ4e3sjOTkZAGBqagpra2sAwNixY7FhwwZs374dlpaWUh1ra2uYmpoiPj4eGzZsQI8ePVCzZk1cvHgRkydPxquvvopmzZqVcwSIiIioItLpPEhnz55Fy5YtpUv0p0yZgpYtW2LGjBkAgKSkJCQmJkr1V65ciezsbIwdOxaOjo7SY+LEiVKdZcuWISUlBb6+vkp1Nm/eDAAwMjLCvn370K1bN3h6euLDDz9E3759sXPnznJ850RERFSR6bQHydfXFy8bIx4REaH0OiYmpthtFjfm3NnZGYcOHVKneURERFRF8V5sRERERCqYIBERERGpYIJEREREpIIJEhEREZEKJkhEREREKpggEREREalggkRERESkggkSERERkQomSEREREQqmCARERERqWCCRERERKSCCRIRERGRCiZIRERERCqYIBERERGpYIJEREREpIIJEhEREZEKJkhEREREKpggEREREalggkRERESkggkSERERkQomSEREREQqmCARERERqWCCRERERKSCCRIRERGRCiZIRERERCqYIBERERGpYIJEREREpIIJEhEREZEKJkhEREREKpggEREREalggkRERESkggkSERERkQomSEREREQqmCARERERqWCCRERERKRCpwnS4cOHERQUBCcnJ8hkMmzbtu2l9bds2QJ/f3/Y2trCysoKPj4+iIqKKlBv6dKlcHNzg4mJCby9vXH69Gml5ZmZmRg7dixq1qwJCwsL9O3bF3fv3tXkWyMiIiI9ptMEKSMjA82bN8fSpUvVqn/48GH4+/tj9+7diI2NhZ+fH4KCgnD+/HmpzubNmzFlyhSEhobi3LlzaN68OQICAnDv3j2pzuTJk7Fz505ERkbi0KFDuHPnDvr06aPx90dERET6qZoudx4YGIjAwEC16y9evFjp9Zw5c7B9+3bs3LkTLVu2BAB89dVXePfddzF8+HAAwPLly7Fr1y6sXr0an3zyCVJSUrBq1Sps2LABr732GgBgzZo1aNSoEU6ePIn27dtr5s0RERGR3tLrMUi5ublIS0uDjY0NAOD58+eIjY1F165dpToGBgbo2rUrTpw4AQCIjY2FQqFQquPp6QkXFxepDhEREVVtOu1BKqsvv/wS6enp6N+/PwDgwYMHyMnJgb29vVI9e3t7/PXXXwCA5ORkGBkZoXr16gXqJCcnF7mvrKwsZGVlSa9TU1MBAAqFAgqFQhNvh9SQH2vGvPwx9rrD2OsG46472oy9utvU2wRpw4YNCA8Px/bt22FnZ6f1/c2dOxfh4eEFyvfu3QszMzOt75+URUdH67oJVRZjrzuMvW4w7rqjjdg/ffpUrXp6mSBt2rQJo0aNQmRkpNKpslq1asHQ0LDAFWl3796Fg4MDAMDBwQHPnz/HkydPlHqRXqxTmJCQEEyZMkV6nZqaCmdnZ3Tr1g1WVlYaemdUHIVCgejoaPj7+0Mul+u6OVUKY687jL1uMO66o83Y558BKo7eJUgbN27EiBEjsGnTJvTs2VNpmZGREVq3bo39+/ejd+/eAPLGKe3fvx/jxo0DALRu3RpyuRz79+9H3759AQBXr15FYmIifHx8ityvsbExjI2NC5TL5XL+x9EBxl13GHvdYex1g3HXHW3EXt3t6TRBSk9Px/Xr16XXCQkJiIuLg42NDVxcXBASEoLbt29j3bp1APJOqwUHB2PJkiXw9vaWxgyZmprC2toaADBlyhQEBwejTZs2aNeuHRYvXoyMjAzpqjZra2uMHDkSU6ZMgY2NDaysrDB+/Hj4+PjwCjYiIiICoOME6ezZs/Dz85Ne55/CCg4ORkREBJKSkpCYmCgtX7lyJbKzszF27FiMHTtWKs+vDwBvv/027t+/jxkzZiA5ORktWrTAnj17lAZuL1q0CAYGBujbty+ysrIQEBCA7777TsvvloiIiPSFThMkX19fCCGKXJ6f9OSLiYlRa7vjxo2TTqkVxsTEBEuXLlV7gkoiIiKqWvR6HiQiIiIibWCCRERERKSCCRIRERGRCiZIRERERCqYIBERERGpYIJEREREpIIJEhEREZEKJkhEREREKpggEREREalggkRERESkggkSERERkQomSEREREQqmCARERERqWCCRERERKSCCRIRERGRCiZIRERERCqYIBERERGpYIJEREREpIIJEhEREZGKarpuAGlJUlLeo6QcHfMeREREVRgTpMpqxQogPLzk64WGAmFhGm8OERGRPmGCVFm9/z7wxhvKZc+eAR075j0/ehQwNS24HnuPiIiImCBVWoWdKsvI+O95ixaAuXm5NomIiEhfcJA2ERERkQomSEREREQqmCARERERqWCCRERERKSCCRIRERGRCiZIRERERCqYIBERERGpYIJEREREpIIJEhEREZEKJkhEREREKpggEREREalggkRERESkQqcJ0uHDhxEUFAQnJyfIZDJs27btpfWTkpIwaNAgNGjQAAYGBpg0aVKBOr6+vpDJZAUePXv2lOoMGzaswPLu3btr+N0RERGRvtJpgpSRkYHmzZtj6dKlatXPysqCra0tpk+fjubNmxdaZ8uWLUhKSpIely9fhqGhIfr166dUr3v37kr1Nm7cWOb3Q0RERJVDNV3uPDAwEIGBgWrXd3Nzw5IlSwAAq1evLrSOjY2N0utNmzbBzMysQIJkbGwMBweHEraYiIiIqoJKPwZp1apVGDBgAMzNzZXKY2JiYGdnh4YNG2LMmDF4+PChjlpIREREFY1Oe5C07fTp07h8+TJWrVqlVN69e3f06dMH7u7uiI+Px6efforAwECcOHEChoaGhW4rKysLWVlZ0uvU1FQAgEKhgEKh0N6b0CSFAnLpqQLQl3a/ID/WehPzSoSx1x3GXjcYd93RZuzV3WalTpBWrVqFpk2bol27dkrlAwYMkJ43bdoUzZo1g4eHB2JiYtClS5dCtzV37lyEh4cXKN+7dy/MzMw023AtMczMxOv//zwqKgo5JiY6bU9ZREdH67oJVRZjrzuMvW4w7rqjjdg/ffpUrXqVNkHKyMjApk2bMHPmzGLr1q1bF7Vq1cL169eLTJBCQkIwZcoU6XVqaiqcnZ3RrVs3WFlZaazdWpWRIT0NCAgAVE476gOFQoHo6Gj4+/tDLpcXvwJpDGOvO4y9bjDuuqPN2OefASpOpU2QIiMjkZWVhSFDhhRb99atW3j48CEcHR2LrGNsbAxjY+MC5XK5XH/+47zQTrlcrvRa3+hV3CsZxl53GHvdYNx1RxuxV3d7Ok2Q0tPTcf36del1QkIC4uLiYGNjAxcXF4SEhOD27dtYt26dVCcuLk5a9/79+4iLi4ORkRG8vLyUtr1q1Sr07t0bNWvWLLDP8PBw9O3bFw4ODoiPj8fUqVNRr169vF4VIiIiqvI0liAJIXD//n3Y2dmpvc7Zs2fh5+cnvc4/hRUcHIyIiAgkJSUhMTFRaZ2WLVtKz2NjY7Fhwwa4urrixo0bUvnVq1dx9OhR7N27t8A+DQ0NcfHiRaxduxZPnjyBk5MTunXrhlmzZhXaQ0RERERVj9oJkpmZGW7evAlbW1sAQM+ePfHDDz9Ip6Xu3bsHJycn5OTkqL1zX19fCCGKXB4REVGg7GX18zVs2LDIeqampoiKilK7jUQllpSU9ygpR8e8BxER6ZzaCVJmZqZS0nH48GE8e/ZMqY46yQtRpbdiBVDIFY/FCg0FwsI03hwiIio5jY5BkslkmtwckX56/33gjTeUy549Azp2zHt+9ChgalpwPfYeERFVGJX2KjYinSnsVNkLUyygRQu9nGKBiKgqUftWI/l3vS/qNREREVFloXYPkhACDRo0kJKi9PR0tGzZEgYGBtJyIiIiospA7QRpzZo12mwHERERUYWhdoIUHByszXYQERERVRhlGqSdmZmJzZs3IyMjA/7+/qhfv76m2kVERESkM2onSFOmTIFCocA333wDAHj+/Dl8fHzwxx9/wMzMDFOnTkV0dDR8fHy01lgiIiKi8qD2VWx79+6Fv7+/9Hr9+vW4efMmrl27hsePH6Nfv36YPXu2VhpJREREVJ7UTpASExOVbgi7d+9evPXWW3B1dYVMJsPEiRNx/vx5rTSSiIiIqDypnSAZGBgoXcp/8uRJtG/fXnpdvXp1PH78WLOtIyIiItIBtROkRo0aYefOnQCAP/74A4mJifDz85OW37x5E/b29ppvIREREVE5U3uQ9tSpUzFgwADs2rULf/zxB3r06AF3d3dp+e7du9GuXTutNJKIiIioPKndg/Tmm29i9+7daNasGSZPnozNmzcrLTczM8MHH3yg8QYSERERlbcSzYPUpUsXdOnSpdBloaGhGmkQERERka6pnSAlJiaqVc/FxaXUjSEiIiKqCNROkF4cb5R/NVv+jWvzy2QyGXJycjTYPCKqVJKS8h4l5eiY9yAiKidqJ0gymQx16tTBsGHDEBQUhGrVynSXEiKqilasAMLDS75eaCgQFqbx5hARFUXtLOfWrVtYu3Yt1qxZg+XLl2PIkCEYOXIkGjVqpM32EVFl8v77wBtvKJc9ewZ07Jj3/OhRwNS04HrsPSJtKqpnMzsb1vHxwPnzQGGdAuzZrNTUTpAcHBwwbdo0TJs2DUePHsWaNWvg7e0NLy8vjBw5EiNHjoSBgdoXxRG9HD+wKqfC/j4ZGf89b9ECMDcv1yYRFdWzKQfg+7L12LNZqZXqPFnHjh3RsWNHzJkzBwMHDsTo0aPRt29f2NjYaLp9VFXxA4uIyksxPZuKmBjILS0LrscfY5VaqRKk48ePY/Xq1YiMjETDhg2xdOlSVK9eXcNNoyqNH1hEVF6K69ls3hzgd1yVo3aClJSUhHXr1mHNmjV4/PgxBg8ejGPHjqFJkybabB9VVfzAIiIiHVI7QXJxcUHt2rURHByMN954A3K5HLm5ubh48aJSvWbNmmm8kURERHqL01voJbUTpJycHCQmJmLWrFmYPXs2gP/mQ8rHeZCIiIhUcHoLvaR2gpSQkKDNdhAREVVOnN5CL6mdILm6umqzHVQOcnIFTjs3xT2LGrC78RjtGpnB0EBW/IpERFR6nN5CL3E67Cpiz+UkhG//A0mD5uYV/HgBjtZXERrkhe5N+CuFiMqI42yokmGCVAXsuZyEMT+dg1ApT07JxJifzmHZkFZMkrSMvXeVhLaTAH2eIJXjbKiSYYJUyeXkCoTv/LNAcgQAAoAMQPjOP+Hv5cAvbC1h710lou0kQJ8nSOU4m8qrivYOMkGq5E4nPEJSSmaRywWApJRMnE54BB+PmuXXsCqCvXeVjLaTAH2eIJXjbCqvKto7WKoEKTs7GzExMYiPj8egQYNgaWmJO3fuwMrKChYWFppuI5XBvbSik6PS1CP1sfeuEtJ2EsAJUqkiqqK9gyVOkG7evInu3bsjMTERWVlZ8Pf3h6WlJebNm4esrCwsX75cG+2kUrKzNNFoPVIfe++IqFKoor2DBiVdYeLEiWjTpg0eP34M0xcyxjfffBP79+/XaOOo7Nq528DR2gRF9U/IADham6CdO280rGnsvSMi0l8l7kE6cuQIjh8/DiMjI6VyNzc33L59W2MNI80wNJAhNMgLY346BxmgdLonP2kKDfLiKR4tYO8dEVExKvCVmyVOkHJzcwu9ncitW7dgWdjgwZc4fPgwFixYgNjYWCQlJWHr1q3o3bt3kfWTkpLw4Ycf4uzZs7h+/TomTJiAxYsXK9WJiIjA8OHDlcqMjY2Rmfnfr3QhBEJDQ/H999/jyZMn6NChA5YtW4b69euXqP36onsTRywb0irvSqq0LKncwdqkal5JVU5XZOT33iWnZBY6DkmGvL8Be++IKp4Xp+aoeTMFPlbW/CGpDRX4ys0SJ0jdunXD4sWLsXLlSgB5919LT09HaGgoevToUaJtZWRkoHnz5hgxYgT69OlTbP2srCzY2tpi+vTpWLRoUZH1rKyscPXqVem1TKZ8UM+fPx9ff/011q5dC3d3d3z22WcICAjAn3/+CROTyvlrvnsTR/i7WuJ041fy5uKJXI92jWpXzf/w5XRFBnvviPRTgak5Nl6G4+7rVfMHpbZV4Cs3S5wgLVy4EAEBAfDy8kJmZiYGDRqEa9euoVatWti4cWOJthUYGIjAwEC167u5uWHJkiUAgNWrVxdZTyaTwcHBodBlQggsXrwY06dPR69evQAA69atg729PbZt24YBAwaU4B3oF0MDGXz+vZT3wq0GUFW/mMvxigz23hHpF72cmkOf5ymqwFduljhBqlOnDi5cuIBNmzbh4sWLSE9Px8iRIzF48GClQdu6lJ6eDldXV+Tm5qJVq1aYM2cOGjduDCDvprvJycno2rWrVN/a2hre3t44ceJEkQlSVlYWsrL++4JLTU0FACgUCigUCi2+Gw1SKCCXnioAfWl3Pk21v1atvMeLMjL+23bjxkVfkVGKfXZpWAu+Y1rjfIvOed31G9ehjacjDA1k+nPsAFJbNd5mHpe63b426Vnbc3IFwnb8UczUHH/At37Nsvf8ajA2Bt99B8PZs0u8Xs706cidMaN0O9Xm31bLx426n2GlmgepWrVqGDJkSGlW1bqGDRti9erVaNasGVJSUvDll1/ilVdewR9//IE6deogOTkZAGBvb6+0nr29vbSsMHPnzkV4Iadl9u7dCzMzM82+CS0xzMzE6///PCoqCjl6djrxxfYfOHBAo+3XdmwMMzPx+v/33v3210lE3dCv2L8oOjpao9urTMel1o6d/39eluPe+NEjmDx+XOL1MmvUQJZN6cbJaTI25dH+aykyJKcaFrk8b2qOLHy7eQ/qWxeWRqlP9iwTtv8/xin+l31wryUvdae+cd26MFm4UKnM4PlzvBoSAgA4PHcuclUurAL+Pza7d5dqn9o87rX5WQ8AT58+VaueTAhR7F95x44dau/4DdVTF2qSyWTFDtJ+ka+vL1q0aFFgkLYqhUKBRo0aYeDAgZg1axaOHz+ODh064M6dO3B8oVuvf//+kMlk2Lx5c6HbKawHydnZGQ8ePICVlZVabda5jAzIa9QAACgeP9a/eSteaP/Te/cg12S3q7Zjo++xR97/pejoaPj7+0Mulxe/gppy0tIL7V3TG+V47JTluDeYObP8exk0GJvyaP/Oi0mYEnmp2Hpf9WuKoGalPzUV9cddzN51Bclpz6UyBytjTO/hiYDG9i9ZswT0+TNNm5/1yPv+rlWrFlJSUl76/a1WD5K6SYtMJiv0CjddksvlaNmyJa5fvw4A0tiku3fvKiVId+/eRYsWLYrcjrGxMYyNjQvdvia/LLTqhXbK5XKl13pBpf0ajbu2Y6PJ7et4vIEmY1/oYFhrPRsMW87HTqlj/8EHwJtvKpepMfbO0NERhqXdpyZjUw7td6yu3pe8Y3XzUv8d9lxOwvhNFwqcxrubmoXxmy5oboyTPn2mFbNtTX/Hqrs9tRKk3NzcMjVGl3JycnDp0iXpCjt3d3c4ODhg//79UkKUmpqKU6dOYcyYMTpsKZGaKsl9kfRyMKw+0/fZkMuh/dqemoO3H9IvOr1ZbXp6utSzA+QNoI6Li4ONjQ1cXFwQEhKC27dvY926dVKduLg4ad379+8jLi4ORkZG8PLyAgDMnDkT7du3R7169fDkyRMsWLAAN2/exKhRowDk9XJNmjQJs2fPRv369aXL/J2cnNTuKSPSqUpwXyR+UVBFpO2pOXj7If1SqgRp//79WLRoEa5cuQIAaNSoESZNmqR0ZZg6zp49Cz8/P+n1lClTAADBwcGIiIhAUlISEhMTldZp2bKl9Dw2NhYbNmyAq6srbty4AQB4/Pgx3n33XSQnJ6NGjRpo3bo1jh8/LiVQADB16lRkZGTgvffew5MnT9CxY0fs2bOn0s6BRJWMvvcEQItfFPp8uTNVCNqcmqMy3X7oxYk07W48RrtGZpXux0yJE6TvvvsOEydOxFtvvYWJEycCAE6ePIkePXpg0aJFGDt2rNrb8vX1xcvGiEdERBQoK25M+aJFi146iSSQ14s0c+ZMzJw5U612EpFmae2LopKcfqws9PVLVHVi3Zob18GnqUuZ215Zbj9UYOzgjxfgaH1Vv8YOqqHECdKcOXOwaNEijBs3TiqbMGECOnTogDlz5pQoQSKiqklrXxSV4PRjZaHvX6IvTqyrcNXMbUYqw+2HqtLYwRInSE+ePEH37t0LlHfr1g3Tpk3TSKNID/BUBpWB1r4oKsHpR1X6eE+wqvQlWhL6fvuhqjZ2sMQJ0htvvIGtW7fi448/Virfvn07Xn/99SLWokqHpzKoDPT9i6K86OM9waral2hJ6fPth8prkHlF+VFQ4gTJy8sLn3/+OWJiYuDj4wMgbwzSsWPH8OGHH+Lrr7+W6k6YMEFzLaWKhacyqIz0+YviRdoaZ6OvvTC8Uqt4+nrz8PIYZF6RfhSUOEFatWoVatSogT///BN//vmnVF69enWsWrVKei2TyZggVWaV8FQGlT99/aLIp61xNvrcC1OZrtTSJn28ebi2B5lXtB8FJU6QEhIStNEOIqqi9PGLAtDuh7k+98JUliu1qCBtDjKviD8KDMplL0RElUhxH+ZA3od5Tm7pbmiqz70w+V+iRX2FyQA4VvArtSqDnFyBE85Nsb3Rqzhx43Gpj8UX5Y8dBFDg71vWsYMl+VFQXkrcgySEwC+//IKDBw/i3r17BW5DsmXLFo01joioItJ2D48+98JwAL7uaXOKBW2NHayIPwpKnCBNmjQJK1asgJ+fH+zt7SGT8SCn8lFRrmwg0vaHub7Pl1NZBuDro/IYx6ONsYMV8UdBiROkH3/8EVu2bJFu/kpUHirSlQ1E2v4wrwy9MPo+AF8flec4Hk2PHayIPwpKPAbJ2toadevW1UZbiAqV/4voxV+iwH+/iPZcLsWElURlUB7jbPJ7YRwsjZXKHaxNKuwl/qryv0R7XTkMH7caGk+OtDHORp9VxHE86tLm+KbSKnEPUlhYGMLDw7F69WqYFjbPDZEGVcQrGyq9omZJz86GdXw8cP48UK2Qj44qNEt6efXwaOueYPn09V5pgP7fykQbKuI4npKoaKdmS5wg9e/fHxs3boSdnR3c3Nwgl8uVlp87d05jjSPS58ud9VYRs6TLAfi+bL0qNkt6eX2Ya+OeYIB+JxgVbb6ciqIijuMpKW3/KCiJEidIwcHBiI2NxZAhQzhIm7RO338R6aViZklXxMRAbmlZcL0q0nv0In0dZ6PPCYZWepUL6zV99uy/5xcuAEUd8xXouK+I43hKQ1s/CkqqxAnSrl27EBUVhY75t5Qg0qLK8ItI7xQ3S3rz5kD16uXapIpM3ya61PfT1lrpVS7m3pJyX9/CF1SwXtPKMLi/IilxguTs7AwrKytttIWogMryi4iootD309Za6VUurNcUgCI7G8eOHkWHjh0hL2rcXQVT0cbx6LMSJ0gLFy7E1KlTsXz5cri5uWmhSUT/4S8iIs3S99PWWulVLupUmUKBlKQkoGVLQGW8bUWmr6d+K5oSJ0hDhgzB06dP4eHhATMzswKDtB89qniXD5J+4y8iIs3R99PWetmrXNwYp7g4oLCrwsswxknfTv1WRCVOkBYvXqyFZhC9XEW6soFIn+llgvECvexVLmaME4oa01vBxjhVNaW6io1IFyrKlQ2kX/R5rh9t0MsEQ4Xe9SoXMcapWBVwjFNVUuIE6UWZmZl4/vy5UhkHcFOVp4PudCqcPs/1o016l2AUQq/G2fD/tl4qcYKUkZGBadOm4eeff8bDhw8LLM/JydFIw4jKk0Z7GXTQnc5ekoL0ea6f8qBXCUYROM6mnFTRH30lTpCmTp2KgwcPYtmyZXjnnXewdOlS3L59GytWrMAXX3yhjTYSaZXGexnKuTudvSQF6ftcP+WFCQappYqOoSpxgrRz506sW7cOvr6+GD58ODp16oR69erB1dUV69evx+DBg7XRTiKt0EovQzn+amIvSeH0fa4fogqlio6hKnGC9OjRI9StWxdA3nij/Mv6O3bsiDFjxmi2dURapO+9DPrefm3S97l+iCoUbf7oq8C3eSlxglS3bl0kJCTAxcUFnp6e+Pnnn9GuXTvs3LkT1Xn7AdIj+t7LoJftL6exDPo+149eqqLjVKiMKvBtXkqcIA0fPhwXLlxA586d8cknnyAoKAjffvstFAoFvvrqK220kUgr9L2XQS/bX05jGfR9rh+9VEXHqVAZVeDbvJQ4QZo8ebL0vGvXrrhy5QrOnTuHevXqoVmzZhptHJE26Xsvg162v5zGMlSGuX70ThUdp0JlVIFv81KmeZAAwM3NjfdkK43CuqPVwe5ojdH3Xga9bH85Hr+VYa4fvcLPJqpk1E6QTpw4gYcPH+L111+XytatW4fQ0FBkZGSgd+/e+Oabb2BsbKyVhlY6xXVHF4Xd0Rqj770M+t7+8qB3c/1U4AGrVIFx/JdWqJ0gzZw5E76+vlKCdOnSJYwcORLDhg1Do0aNsGDBAjg5OSGMX97qKaw7+tmz/87THz1a9AFNGqPvvQz63v7yoFdz/VTgAatUgXH8l1aonSDFxcVh1qxZ0utNmzbB29sb33//PQDA2dkZoaGhTJDUVVjmnpHx3/MWLQBz83JtUlWld70MKvS9/fSCCjxglSowjv/SCrUTpMePH8Pe3l56fejQIQQGBkqv27Zti3///VezrSMqJ3rVy1AIfW+/3tD2qYwKPGCVKjCeKtMKtRMke3t7JCQkwNnZGc+fP8e5c+cQ/kKXXlpaGuT8j0tElRlPZRBVGWonSD169MAnn3yCefPmYdu2bTAzM0OnTp2k5RcvXoSHh4dWGklEVCHwVAZRlaF2gjRr1iz06dMHnTt3hoWFBdauXQsjIyNp+erVq9GtW7cS7fzw4cNYsGABYmNjkZSUhK1bt6J3795F1k9KSsKHH36Is2fP4vr165gwYQIWL16sVOf777/HunXrcPnyZQBA69atMWfOHLRr106qM2zYMKxdu1ZpvYCAAOzZs6dE7SeqKnJyBU47N8U9ixqoeTMFPlbWVXOME09l6A6v1KJypnaCVKtWLRw+fBgpKSmwsLCAoaGh0vLIyEhYWFiUaOcZGRlo3rw5RowYgT59+hRbPysrC7a2tpg+fToWLVpUaJ2YmBgMHDgQr7zyCkxMTDBv3jx069YNf/zxB2rXri3V6969O9asWSO95vQERIXbczkp7yq5QXPzCjZehuPu67xKjsoXT29SOSvxRJHW1taFltvYlHwyusDAQKWB3sVxc3PDkiVLAOT1WBVm/fr1Sq9/+OEH/Prrr9i/fz+GDh0qlRsbG8PBwaHEbSaqSvZcTsKYn84VmIgyOSUTY346h2VDWjFJ0hf63gPD05tUzso8k3ZF9/TpUygUigIJXExMDOzs7FCjRg289tprmD17NmrWLPqGnllZWcjK+m+emdTUVACAQqGAQqHQTGMVCsilpwpAU9vl9nW77fKghfbn5AqE7fij0Fm6BfImowzf+Qd869es2Kfb9P1vC0ifMWX5rDH47jsYzp5ddIUiemBypk9H7owZpd6vxtSqlfcojVLGTRNxp9LRZuzV3WalT5CmTZsGJycndO3aVSrr3r07+vTpA3d3d8THx+PTTz9FYGAgTpw4UeDUYb65c+cqXbWXb+/evTAzM9NIWw0zM5E/T3lUVBRyTDR7D63KtP0DBw5odPvabru2aaP911JkSE4t/P8DkJckJaVk4dvNe1DfurA0qmLQ97/ti6Kjo0u9rnHdujBZuLDE62XWqIGs3btLvd/KoCxxp7LRRuyfPn2qVj2ZEKJCfLLJZLJiB2m/yNfXFy1atCgwSPtFX3zxBebPn4+YmJiX3kj3n3/+gYeHB/bt24cuXboUWqewHiRnZ2c8ePAAVlZWarW5WBkZkNeoAQBQPH6s+YkiK9H2n967B3n16lrZtlbarm1aaP/Oi0mYEnmp2Hpf9WuKoGYV+DSGvv9tkfeLNzo6Gv7+/pxOpRwx7rqjzdinpqaiVq1aSElJeen3d6XtQfryyy/xxRdfYN++fS9NjgCgbt26qFWrFq5fv15kgmRsbFzoQG65XK65P94L25HL5ZqfEK6SbV+j/2m03XZt00L7Haurl0g4Vjev2F8e+v63fYHGj3tSC+OuO9qIvbrbq5QJ0vz58/H5558jKioKbdq0Kbb+rVu38PDhQzhWpsF8+j4gk3SunbsNHK1NkJySWeg4JBny7vnWzr3kF2gQEVV0Ok2Q0tPTcf36del1QkIC4uLiYGNjAxcXF4SEhOD27dtYt26dVCcuLk5a9/79+4iLi4ORkRG8vLwAAPPmzcOMGTOwYcMGuLm5ITk5GQBgYWEBCwsLpKenIzw8HH379oWDgwPi4+MxdepU1KtXDwEBAeX35rWNl8RSGRkayBAa5IUxP52DDFBKkvKHZIcGeVXsAdpERKWk0wTp7Nmz8PPzk15PmTIFABAcHIyIiAgkJSUhMTFRaZ2WLVtKz2NjY7Fhwwa4urrixo0bAIBly5bh+fPneOutt5TWy7+RrqGhIS5evIi1a9fiyZMncHJyQrdu3TBr1qzKNReSDi6JfXEyQbsbj9GukRm/PPVc9yaOWDakVd48SGn/jcFzsDbhPEhEVKnpNEHy9fXFy8aIR0REFCgrbkx5fqJUFFNTU0RFRanTPP1WzqfKCkwm+OMFOFpf5ZdoJdC9iSP8XS1xuvEreTNpb1wHn6YuTH6JqFKrlGOQqHxxMsHKz9BABp9/865oU7hW0duMEFGVYqDrBpB+y8kVCN/5Z5GTCQJA+M4/kZNbIWaTICIiUgsTJCqT0wmPkJSSWeTyvMkEM3E64VH5NYqIiKiMeIqNyuReWtHJUWnqEZUJp7cgIg1hgkRlYmep3m0b1K1HVCac3oKINIQJEpUJJxOkCoV3fCciDWGCRGWitckEiztVcuECYGlZcD2eKqna+PcnIg1hgkRlppXJBIs5VSL39S18AU+VEBGRBjBBIo1QnUzQLnI92jWqXfr5coo4VaLIzsaxo0fRoWNHyKsVcviy94CIiDSACRJpzIuTCcKtBlCWyQSLOlWiUCAlKQlo2VKv78peJrxSi4hI65ggEekbXqlFRKR1TJCI9A2v1CIi0jomSET6hqfKiIi0jrcaISIiIlLBBImIiIhIBU+xUdXDq8CIiKgYTJCo6uFVYEREVAwmSFT18CowIiIqBhMkqnp4qoyIiIrBQdpEREREKpggEREREalggkRERESkgmOQiEhZcdMgXLgAWFoWXI9ju4ioEmGCRETKipkGQe7rW/gCToNARJUIEyQiUlbENAiK7GwcO3oUHTp2hLxaIR8d7D0iokqECRIRKSvqVJlCgZSkJKBlS0AuL/92ERGVIw7SJiIiIlLBBImIiIhIBRMkIiIiIhVMkIiIiIhUMEEiIiIiUsEEiYiIiEgFEyQiIiIiFUyQiIiIiFQwQSIiIiJSodME6fDhwwgKCoKTkxNkMhm2bdv20vpJSUkYNGgQGjRoAAMDA0yaNKnQepGRkfD09ISJiQmaNm2K3bt3Ky0XQmDGjBlwdHSEqakpunbtimvXrmnoXREREZG+02mClJGRgebNm2Pp0qVq1c/KyoKtrS2mT5+O5s2bF1rn+PHjGDhwIEaOHInz58+jd+/e6N27Ny5fvizVmT9/Pr7++mssX74cp06dgrm5OQICApCZmamR90VERET6Taf3YgsMDERgYKDa9d3c3LBkyRIAwOrVqwuts2TJEnTv3h0ff/wxAGDWrFmIjo7Gt99+i+XLl0MIgcWLF2P69Ono1asXAGDdunWwt7fHtm3bMGDAgDK+KyIiItJ3le5mtSdOnMCUKVOUygICAqTTdwkJCUhOTkbXrl2l5dbW1vD29saJEyeKTJCysrKQlZUlvU5NTQUAKBQKKBQKzTReoYBceqoANLXd8lIO7c+PtcZiTmpj7HWHsdcNxl13tBl7dbdZ6RKk5ORk2NvbK5XZ29sjOTlZWp5fVlSdwsydOxfh4eEFyvfu3QszM7OyNhsAYJiZidf//3lUVBRyTEw0st3yUp7tj46O1tq26eUYe91h7HWDcdcdbcT+6dOnatWrdAmStoSEhCj1TKWmpsLZ2RndunWDlZWVZnaSkSE9DQgIAMzNNbPd8lIO7VcoFIiOjoa/vz/kcnnxK5DGMPa6w9jrBuOuO9qMff4ZoOJUugTJwcEBd+/eVSq7e/cuHBwcpOX5ZY6Ojkp1WrRoUeR2jY2NYWxsXKBcLpdr7o/3wnbkcrnSa71Qju3XaNypRBh73WHsdYNx1x1txF7d7VW6eZB8fHywf/9+pbLo6Gj4+PgAANzd3eHg4KBUJzU1FadOnZLqEBERUdWm0x6k9PR0XL9+XXqdkJCAuLg42NjYwMXFBSEhIbh9+zbWrVsn1YmLi5PWvX//PuLi4mBkZAQvLy8AwMSJE9G5c2csXLgQPXv2xKZNm3D27FmsXLkSACCTyTBp0iTMnj0b9evXh7u7Oz777DM4OTmhd+/e5fbeiYiIqOLSaYJ09uxZ+Pn5Sa/zx/gEBwcjIiICSUlJSExMVFqnZcuW0vPY2Fhs2LABrq6uuHHjBgDglVdewYYNGzB9+nR8+umnqF+/PrZt24YmTZpI602dOhUZGRl477338OTJE3Ts2BF79uyBiZ4NiiYiIiLt0GmC5OvrCyFEkcsjIiIKlL2sfr5+/fqhX79+RS6XyWSYOXMmZs6cqVY7iYiIqGqpdGOQiIiIiMqKCRIRERGRCiZIRERERCoq3TxIVE6SkvIeL3r27L/ncXGAqWnB9Rwd8x5EREQVGBMkKp0VK4BCbr0i6dix8PLQUCAsTCtNIiIi0hQmSFQ6778PvPFGyddj7xEREekBJkhUOjxVRkRElRgTpAokJ1fgtHNT3LOoAbsbj9GukRkMDWS6bhYREVGVwwSpgthzOQnh2/9A0qC5eQU/XoCj9VWEBnmhexP21BAREZUnXuZfAey5nIQxP51DUlqWUnlySibG/HQOey4nFbEmERERaQMTJB3LyRUI3/knCruBSn5Z+M4/kZNb/C1WiIiISDOYIOnY6YRHSErJLHK5AJCUkonTCY/Kr1FERERVHBMkHbuXVnRyVJp6REREVHZMkHTMztJEo/WIiIio7Jgg6Vg7dxs4WpugqIv5ZQAcrU3Qzt2mPJtFRERUpTFB0jFDAxlCg7wAoECSlP86NMiL8yERERGVIyZIFUD3Jo5YNqQVHCyNlcodrE2wbEgrzoNERERUzjhRZAXRvYkj/F0tcbrxK3kzaUeuR7tGtdlzREREpANMkCoQQwMZfP69lPfCrQbA5IiIiEgneIqNiIiISAUTJCIiIiIVTJCIiIiIVDBBIiIiIlLBBImIiIhIBRMkIiIiIhVMkIiIiIhUMEEiIiIiUsEEiYiIiEgFEyQiIiIiFUyQiIiIiFQwQSIiIiJSwQSJiIiISAUTJCIiIiIVTJCIiIiIVDBBIiIiIlKh0wTp8OHDCAoKgpOTE2QyGbZt21bsOjExMWjVqhWMjY1Rr149REREKC13c3ODTCYr8Bg7dqxUx9fXt8Dy0aNHa/jdERERkb7SaYKUkZGB5s2bY+nSpWrVT0hIQM+ePeHn54e4uDhMmjQJo0aNQlRUlFTnzJkzSEpKkh7R0dEAgH79+ilt691331WqN3/+fM29MSIiItJr1XS588DAQAQGBqpdf/ny5XB3d8fChQsBAI0aNcLRo0exaNEiBAQEAABsbW2V1vniiy/g4eGBzp07K5WbmZnBwcGhjO+AiIiIKiO9GoN04sQJdO3aVaksICAAJ06cKLT+8+fP8dNPP2HEiBGQyWRKy9avX49atWqhSZMmCAkJwdOnT7XWbiIiItIvOu1BKqnk5GTY29srldnb2yM1NRXPnj2Dqamp0rJt27bhyZMnGDZsmFL5oEGD4OrqCicnJ1y8eBHTpk3D1atXsWXLliL3nZWVhaysLOl1amoqAEChUEChUJTxnf0/hQJy6akC0NR2K5H8WGss5qQ2xl53GHvdYNx1R5uxV3ebepUgldSqVasQGBgIJycnpfL33ntPet60aVM4OjqiS5cuiI+Ph4eHR6Hbmjt3LsLDwwuU7927F2ZmZhppr2FmJl7//+dRUVHIMTHRyHYro/yxZVT+GHvdYex1g3HXHW3EXt0zRnqVIDk4OODu3btKZXfv3oWVlVWB3qObN29i3759L+0Vyuft7Q0AuH79epEJUkhICKZMmSK9Tk1NhbOzM7p16wYrK6uSvpXCZWRITwMCAgBzc81stxJRKBSIjo6Gv78/5HJ58SuQxjD2usPY6wbjrjvajH3+GaDi6FWC5OPjg927dyuVRUdHw8fHp0DdNWvWwM7ODj179ix2u3FxcQAAR0fHIusYGxvD2Ni4QLlcLtfcH++F7cjlcqXXpEyjcacSYex1h7HXDcZdd7QRe3W3p9NB2unp6YiLi5MSlISEBMTFxSExMRFAXq/N0KFDpfqjR4/GP//8g6lTp+Kvv/7Cd999h59//hmTJ09W2m5ubi7WrFmD4OBgVKumnAPGx8dj1qxZiI2NxY0bN7Bjxw4MHToUr776Kpo1a6bdN0xERER6Qac9SGfPnoWfn5/0Ov8UVnBwMCIiIpCUlCQlSwDg7u6OXbt2YfLkyViyZAnq1KmDH374QbrEP9++ffuQmJiIESNGFNinkZER9u3bh8WLFyMjIwPOzs7o27cvpk+frqV3SURERPpGpwmSr68vhBBFLledJTt/nfPnz790u926dStyu87Ozjh06FCJ2klERERVi17Ng0RERERUHpggEREREalggkRERESkggkSERERkQomSEREREQqmCARERERqWCCRERERKSCCRIRERGRCiZIRERERCqYIBERERGpYIJEREREpIIJEhEREZEKJkhEREREKpggEREREalggkRERESkggkSERERkQomSEREREQqmCARERERqWCCRERERKSCCRIRERGRCiZIRERERCqYIBERERGpYIJEREREpIIJEhEREZEKJkhEREREKpggEREREalggkRERESkggkSERERkQomSEREREQqmCARERERqWCCRERERKSCCRIRERGRCiZIRERERCqYIBERERGpYIJEREREpEKnCdLhw4cRFBQEJycnyGQybNu2rdh1YmJi0KpVKxgbG6NevXqIiIhQWh4WFgaZTKb08PT0VKqTmZmJsWPHombNmrCwsEDfvn1x9+5dDb4zIiIi0mc6TZAyMjLQvHlzLF26VK36CQkJ6NmzJ/z8/BAXF4dJkyZh1KhRiIqKUqrXuHFjJCUlSY+jR48qLZ88eTJ27tyJyMhIHDp0CHfu3EGfPn009r6IiIhIv1XT5c4DAwMRGBiodv3ly5fD3d0dCxcuBAA0atQIR48exaJFixAQECDVq1atGhwcHArdRkpKClatWoUNGzbgtddeAwCsWbMGjRo1wsmTJ9G+ffsyvCMiIiKqDPRqDNKJEyfQtWtXpbKAgACcOHFCqezatWtwcnJC3bp1MXjwYCQmJkrLYmNjoVAolLbj6ekJFxeXAtshIiKiqkmnPUgllZycDHt7e6Uye3t7pKam4tmzZzA1NYW3tzciIiLQsGFDJCUlITw8HJ06dcLly5dhaWmJ5ORkGBkZoXr16gW2k5ycXOS+s7KykJWVJb1OSUkBADx69AgKhUIzbzAjA/L/f6p4+BDIzNTMdisRhUKBp0+f4uHDh5DL5cWvQBrD2OsOY68bjLvuaDP2aWlpAAAhxEvr6VWCpI4XT9k1a9YM3t7ecHV1xc8//4yRI0eWertz585FeHh4gXJ3d/dSb/OlXF21s10iIiJCWloarK2ti1yuVwmSg4NDgavN7t69CysrK5iamha6TvXq1dGgQQNcv35d2sbz58/x5MkTpV6ku3fvFjluCQBCQkIwZcoU6XVubi4ePXqEmjVrQiaTleFdUUmkpqbC2dkZ//77L6ysrHTdnCqFsdcdxl43GHfd0WbshRBIS0uDk5PTS+vpVYLk4+OD3bt3K5VFR0fDx8enyHXS09MRHx+Pd955BwDQunVryOVy7N+/H3379gUAXL16FYmJiS/djrGxMYyNjZXKVE/TUfmxsrLiB5aOMPa6w9jrBuOuO9qK/ct6jvLpNEFKT0+XenaAvMv44+LiYGNjAxcXF4SEhOD27dtYt24dAGD06NH49ttvMXXqVIwYMQIHDhzAzz//jF27dknb+OijjxAUFARXV1fcuXMHoaGhMDQ0xMCBAwHkBWXkyJGYMmUKbGxsYGVlhfHjx8PHx4dXsBEREREAHSdIZ8+ehZ+fn/Q6/xRWcHAwIiIikJSUpHQFmru7O3bt2oXJkydjyZIlqFOnDn744QelS/xv3bqFgQMH4uHDh7C1tUXHjh1x8uRJ2NraSnUWLVoEAwMD9O3bF1lZWQgICMB3331XDu+YiIiI9IFMFDeMm6gCycrKwty5cxESElLglCdpF2OvO4y9bjDuulMRYs8EiYiIiEiFXk0USURERFQemCARERERqWCCRERERKSCCRIRERGRCiZIVOHMnTsXbdu2haWlJezs7NC7d29cvXpVqU5mZibGjh2LmjVrwsLCAn379i0wyzqV3RdffAGZTIZJkyZJZYy99ty+fRtDhgxBzZo1YWpqiqZNm+Ls2bPSciEEZsyYAUdHR5iamqJr1664du2aDltcOeTk5OCzzz6Du7s7TE1N4eHhgVmzZindq4ux14zDhw8jKCgITk5OkMlk2LZtm9JydeL86NEjDB48GFZWVqhevTpGjhyJ9PR0jbeVCRJVOIcOHcLYsWNx8uRJREdHQ6FQoFu3bsjIyJDqTJ48GTt37kRkZCQOHTqEO3fuoE+fPjpsdeVz5swZrFixAs2aNVMqZ+y14/Hjx+jQoQPkcjl+//13/Pnnn1i4cCFq1Kgh1Zk/fz6+/vprLF++HKdOnYK5uTkCAgKQyRtbl8m8efOwbNkyfPvtt7hy5QrmzZuH+fPn45tvvpHqMPaakZGRgebNm2Pp0qWFLlcnzoMHD8Yff/yB6Oho/Pbbbzh8+DDee+89zTdWEFVw9+7dEwDEoUOHhBBCPHnyRMjlchEZGSnVuXLligAgTpw4oatmVippaWmifv36Ijo6WnTu3FlMnDhRCMHYa9O0adNEx44di1yem5srHBwcxIIFC6SyJ0+eCGNjY7Fx48byaGKl1bNnTzFixAilsj59+ojBgwcLIRh7bQEgtm7dKr1WJ85//vmnACDOnDkj1fn999+FTCYTt2/f1mj72INEFV5KSgoAwMbGBgAQGxsLhUKBrl27SnU8PT3h4uKCEydO6KSNlc3YsWPRs2dPpRgDjL027dixA23atEG/fv1gZ2eHli1b4vvvv5eWJyQkIDk5WSn21tbW8Pb2ZuzL6JVXXsH+/fvx999/AwAuXLiAo0ePIjAwEABjX17UifOJEydQvXp1tGnTRqrTtWtXGBgY4NSpUxptj17drJaqntzcXEyaNAkdOnRAkyZNAADJyckwMjIqcLNge3t7JCcn66CVlcumTZtw7tw5nDlzpsAyxl57/vnnHyxbtgxTpkzBp59+ijNnzmDChAkwMjJCcHCwFF97e3ul9Rj7svvkk0+QmpoKT09PGBoaIicnB59//jkGDx4MAIx9OVEnzsnJybCzs1NaXq1aNdjY2Gj8b8EEiSq0sWPH4vLlyzh69Kium1Il/Pvvv5g4cSKio6NhYmKi6+ZUKbm5uWjTpg3mzJkDAGjZsiUuX76M5cuXIzg4WMetq9x+/vlnrF+/Hhs2bEDjxo0RFxeHSZMmwcnJibGvwniKjSqscePG4bfffsPBgwdRp04dqdzBwQHPnz/HkydPlOrfvXsXDg4O5dzKyiU2Nhb37t1Dq1atUK1aNVSrVg2HDh3C119/jWrVqsHe3p6x1xJHR0d4eXkplTVq1Ei6YXd+fFWvGGTsy+7jjz/GJ598ggEDBqBp06Z45513MHnyZMydOxcAY19e1Imzg4MD7t27p7Q8Ozsbjx490vjfggkSVThCCIwbNw5bt27FgQMH4O7urrS8devWkMvl2L9/v1R29epVJCYmwsfHp7ybW6l06dIFly5dQlxcnPRo06YNBg8eLD1n7LWjQ4cOBaaz+Pvvv+Hq6goAcHd3h4ODg1LsU1NTcerUKca+jJ4+fQoDA+WvQ0NDQ+Tm5gJg7MuLOnH28fHBkydPEBsbK9U5cOAAcnNz4e3trdkGaXTIN5EGjBkzRlhbW4uYmBiRlJQkPZ4+fSrVGT16tHBxcREHDhwQZ8+eFT4+PsLHx0eHra68XryKTQjGXltOnz4tqlWrJj7//HNx7do1sX79emFmZiZ++uknqc4XX3whqlevLrZv3y4uXrwoevXqJdzd3cWzZ8902HL9FxwcLGrXri1+++03kZCQILZs2SJq1aolpk6dKtVh7DUjLS1NnD9/Xpw/f14AEF999ZU4f/68uHnzphBCvTh3795dtGzZUpw6dUocPXpU1K9fXwwcOFDjbWWCRBUOgEIfa9askeo8e/ZMfPDBB6JGjRrCzMxMvPnmmyIpKUl3ja7EVBMkxl57du7cKZo0aSKMjY2Fp6enWLlypdLy3Nxc8dlnnwl7e3thbGwsunTpIq5evaqj1lYeqampYuLEicLFxUWYmJiIunXriv/9738iKytLqsPYa8bBgwcL/XwPDg4WQqgX54cPH4qBAwcKCwsLYWVlJYYPHy7S0tI03laZEC9MFUpEREREHINEREREpIoJEhEREZEKJkhEREREKpggEREREalggkRERESkggkSERERkQomSEREREQqmCARUaUSFhYGe3t7yGQybNu2TdfNqbJ8fX0xadKkYuu9+uqr2LBhQ5n21b59e/z6669l2gaRKiZIRFo2bNgwyGQyyGQyGBkZoV69epg5cyays7N13bRi6VuSceXKFYSHh2PFihVISkpCYGCg0vKwsDDpb1HUo7TCwsLQokWLYus9ffoUISEh8PDwgImJCWxtbdG5c2ds375d7X1FRESgevXqpW7TjRs3IJPJEBcXp/Y+tWHHjh24e/cuBgwYIJVNmTIFNjY2cHZ2xvr165XqR0ZGIigoqMB2pk+fjk8++US6dxqRJjBBIioH3bt3R1JSEq5du4YPP/wQYWFhWLBgQam2lZOTwy+CIsTHxwMAevXqBQcHBxgbGyst/+ijj5CUlCQ96tSpg5kzZyqVadvo0aOxZcsWfPPNN/jrr7+wZ88evPXWW3j48KHW913RfP311xg+fLh0o9idO3diw4YN2Lt3L+bPn49Ro0bhwYMHAICUlBT873//w9KlSwtsJzAwEGlpafj999/Ltf1UuTFBIioHxsbGcHBwgKurK8aMGYOuXbtix44dAICsrCx89NFHqF27NszNzeHt7Y2YmBhp3fzegh07dsDLywvGxsZITExEVlYWpk2bBmdnZxgbG6NevXpYtWqVtN7ly5cRGBgICwsL2Nvb45133pG+bIC8UyATJkzA1KlTYWNjAwcHB4SFhUnL3dzcAABvvvkmZDKZ9Do+Ph69evWCvb09LCws0LZtW+zbt0/p/SYlJaFnz54wNTWFu7s7NmzYADc3NyxevFiq8+TJE4waNQq2trawsrLCa6+9hgsXLrw0jpcuXcJrr70GU1NT1KxZE++99x7S09MB5PWW5PcuGBgYFNobZGFhAQcHB+lhaGgIS0tL6bVCoUD//v1RvXp12NjYoFevXrhx44a0fkxMDNq1awdzc3NUr14dHTp0wM2bNxEREYHw8HBcuHBB6omKiIgo9D3s2LEDn376KXr06AE3Nze0bt0a48ePx4gRI6Q6LzsmYmJiMHz4cKSkpEj7evHvVho3b95EUFAQatSoAXNzczRu3Bi7d++Wlhd3LGVkZGDo0KGwsLCAo6MjFi5cWOw+79+/jwMHDij1CF25cgW+vr5o06YNBg4cCCsrKyQkJAAApk6dijFjxsDFxaXAtgwNDdGjRw9s2rSpLGEgUsIEiUgHTE1N8fz5cwDAuHHjcOLECWzatAkXL15Ev3790L17d1y7dk2q//TpU8ybNw8//PAD/vjjD9jZ2WHo0KHYuHEjvv76a1y5cgUrVqyAhYUFgLzk47XXXkPLli1x9uxZ7NmzB3fv3kX//v2V2rF27VqYm5vj1KlTmD9/PmbOnIno6GgAwJkzZwAAa9asQVJSkvQ6PT0dPXr0wP79+3H+/Hl0794dQUFBSExMlLY7dOhQ3LlzBzExMfj111+xcuVK3Lt3T2nf/fr1w7179/D7778jNjYWrVq1QpcuXfDo0aNCY5aRkYGAgADUqFEDZ86cQWRkJPbt24dx48YByOsdWrNmDQCUqjdIoVAgICAAlpaWOHLkCI4dOwYLCwt0794dz58/R3Z2Nnr37o3OnTvj4sWLOHHiBN577z3IZDK8/fbb+PDDD9G4cWNp32+//Xah+3FwcMDu3buRlpZWZFtedky88sorWLx4MaysrKR9ffTRRyV6r6rGjh2LrKwsHD58GJcuXcK8efNKdCx9/PHHOHToELZv3469e/ciJiYG586de+k+jx49CjMzMzRq1Egqa968Oc6ePYvHjx8jNjYWz549Q7169XD06FGcO3cOEyZMKHJ77dq1w5EjR8oUByIlGr/9LREpCQ4OFr169RJC5N2pOjo6WhgbG4uPPvpI3Lx5UxgaGorbt28rrdOlSxcREhIihBBizZo1AoCIi4uTll+9elUAENHR0YXuc9asWaJbt25KZf/++68AIN0Zu3PnzqJjx45Kddq2bSumTZsmvQYgtm7dWux7bNy4sfjmm2+EEEJcuXJFABBnzpyRll+7dk0AEIsWLRJCCHHkyBFhZWUlMjMzlbbj4eEhVqxYUeg+Vq5cKWrUqCHS09Olsl27dgkDAwORnJwshBBi69atoiQfa66urlKbfvzxR9GwYUORm5srLc/KyhKmpqYiKipKPHz4UAAQMTExhW4rNDRUNG/evNh9Hjp0SNSpU0fI5XLRpk0bMWnSJHH06FFpubrHhLW1dbH7KqpNCQkJAoA4f/68EEKIpk2birCwsEK3UdyxlJaWJoyMjMTPP/8sLX/48KEwNTUVEydOLLJtixYtEnXr1i20zR4eHqJJkyZiy5YtIisrSzRp0kScPXtWfPPNN6JBgwbilVdeEZcvX1Zab/v27cLAwEDk5OQUuU+ikqimq8SMqCr57bffYGFhAYVCgdzcXAwaNAhhYWGIiYlBTk4OGjRooFQ/KysLNWvWlF4bGRmhWbNm0uu4uDgYGhqic+fOhe7vwoULOHjwoNQL8KL4+Hhpfy9uEwAcHR0L9PSoSk9PR1hYGHbt2oWkpCRkZ2fj2bNnUg/S1atXUa1aNbRq1Upap169eqhRo4ZS+9LT05XeIwA8e/ZMGkek6sqVK2jevDnMzc2lsg4dOiA3NxdXr16Fvb39S9tdnAsXLuD69euwtLRUKs/MzER8fDy6deuGYcOGISAgAP7+/ujatSv69+8PR0fHEu3n1VdfxT///IOTJ0/i+PHj2L9/P5YsWYLw8HB89tlnuHTpklrHhCZNmDABY8aMwd69e9G1a1f07dtXOjaKO5aePXuG58+fw9vbWyq3sbFBw4YNX7rPZ8+ewcTEpEB5WFiY0inD8PBwdO3aFXK5HLNnz8alS5fw22+/YejQoYiNjZXqmZqaIjc3F1lZWTA1NS1pCIgKYIJEVA78/PywbNkyGBkZwcnJCdWq5f3XS09Ph6GhIWJjY2FoaKi0zotfSKampkpjaor7AkhPT0dQUBDmzZtXYNmLX+hyuVxpmUwmK3YA+EcffYTo6Gh8+eWXqFevHkxNTfHWW29JpwzVkZ6eDkdHR6WxVvnUuTpLG9LT09G6desCV04BgK2tLYC8040TJkzAnj17sHnzZkyfPh3R0dFo3759ifYll8vRqVMndOrUCdOmTcPs2bMxc+ZMTJs2Te1jQh1WVlZISUkpUP7kyRMAgLW1NQBg1KhRCAgIwK5du7B3717MnTsXCxcuxPjx44s9lq5fv16iNuWrVasWHj9+/NI6f/31F3766SecP38eq1evxquvvgpbW1v0798fI0aMQFpampTQPnr0CObm5kyOSGOYIBGVA3Nzc9SrV69AecuWLZGTk4N79+6hU6dOam+vadOmyM3NxaFDh9C1a9cCy1u1aoVff/0Vbm5uUjJWGnK5HDk5OUplx44dw7Bhw/Dmm28CyEssXhzI3LBhQ2RnZ+P8+fNo3bo1AOD69etKX4atWrVCcnIyqlWrJg3+Lk6jRo0QERGBjIwMqRfp2LFjMDAwKLa3Qh2tWrXC5s2bYWdnBysrqyLrtWzZEi1btkRISAh8fHywYcMGtG/fHkZGRgVipS4vLy9kZ2cjMzNTrWNC3X01bNgQt27dwt27d5V62M6dOwcTExOlAc/Ozs4YPXo0Ro8ejZCQEHz//fcYP358sceSh4cH5HI5Tp06JW3v8ePH+Pvvv4vs4QTy4picnIzHjx8r9S7mE0Lg/fffx1dffQULCwvk5ORAoVAAgPTvizG4fPkyWrZsWWxMiNTFQdpEOtSgQQMMHjwYQ4cOxZYtW5CQkIDTp09j7ty52LVrV5Hrubm5ITg4GCNGjMC2bduQkJCAmJgY/PzzzwDyBt0+evQIAwcOxJkzZxAfH4+oqCgMHz68RF/ibm5u2L9/v/RFBgD169fHli1bEBcXhwsXLmDQoEFKvU6enp7o2rUr3nvvPZw+fRrnz5/He++9p9QL1rVrV/j4+KB3797Yu3cvbty4gePHj+N///sfzp49W2hbBg8eDBMTEwQHB+Py5cs4ePAgxo8fj3feeafMp9fyt1+rVi306tULR44ckWI6YcIE3Lp1CwkJCQgJCcGJEydw8+ZN7N27F9euXZMGGbu5uSEhIQFxcXF48OABsrKyCt2Pr68vVqxYgdjYWNy4cQO7d+/Gp59+Cj8/P1hZWal1TLi5uSE9PR379+/HgwcP8PTp00L3FRAQgIYNG2LgwIE4fvw4/vnnH/zyyy+YPn06Jk6cKPVQTZo0CVFRUUhISMC5c+dw8OBB6X0VdyxZWFhg5MiR+Pjjj3HgwAFcvnwZw4YNky7dL0rLli1Rq1YtHDt2rNDlP/zwA2xtbaWr3Dp06IADBw7g5MmTWLRoEby8vJR6G48cOYJu3bq9dJ9EJaLrQVBEld2Lg7QL8/z5czFjxgzh5uYm5HK5cHR0FG+++aa4ePGiEKLoAbnPnj0TkydPFo6OjsLIyEjUq1dPrF69Wlr+999/izfffFNUr15dmJqaCk9PTzFp0iRpEHLnzp0LDKLt1auXCA4Oll7v2LFD1KtXT1SrVk24uroKIfIG+Pr5+QlTU1Ph7Owsvv322wLbunPnjggMDBTGxsbC1dVVbNiwQdjZ2Ynly5dLdVJTU8X48eOFk5OTkMvlwtnZWQwePFgkJiYWGauLFy8KPz8/YWJiImxsbMS7774r0tLSpOVlGaQthBBJSUli6NCholatWsLY2FjUrVtXvPvuuyIlJUUkJyeL3r17S/F2dXUVM2bMkAYFZ2Zmir59+4rq1asLAGLNmjWF7nPOnDnCx8dH2NjYCBMTE1G3bl0xYcIE8eDBA6lOcceEEEKMHj1a1KxZUwAQoaGhRb7H27dvi+DgYOHi4iJMTU2Fl5eX+OKLL8Tz58+lOuPGjRMeHh7C2NhY2NrainfeeUepPcUdS2lpaWLIkCHCzMxM2Nvbi/nz5xd6fKmaOnWqGDBgQIHy5ORk4erqWmCgenh4uLCxsRGenp7i1KlTUvmtW7eEXC4X//7770v3R1QSMiGE0HGORkSV3K1bt+Ds7Ix9+/ahS5cuum4OVRDJyclo3Lgxzp07B1dX11JvZ9q0aXj8+DFWrlypwdZRVccxSESkcQcOHEB6ejqaNm2KpKQkTJ06FW5ubnj11Vd13TSqQBwcHLBq1SokJiaWKUGys7PDlClTNNgyIoA9SESkcVFRUfjwww/xzz//wNLSUprcsCxfgkRE5YkJEhEREZEKXsVGREREpIIJEhEREZEKJkhEREREKpggEREREalggkRERESkggkSERERkQomSEREREQqmCARERERqWCCRERERKTi/wBgAQd2fQsOjQAAAABJRU5ErkJggg==", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Modified function to also compute standard error\n", "def compute_average_mse_and_std_error(percentage, squared_errors):\n", " subset_size = int(percentage * len(squared_errors))\n", " indices = np.random.choice(len(squared_errors), subset_size, replace=False)\n", " subset_errors = squared_errors.iloc[indices]\n", " average_mse = subset_errors.mean()\n", " std_error = subset_errors.std() / np.sqrt(subset_size) # Standard error\n", " return average_mse, std_error\n", "\n", "# Percentages to test\n", "percentages = np.arange(0.05, 1.05, 0.05)\n", "\n", "# Compute average MSE values and standard errors for each percentage\n", "average_mse_values, std_errors = zip(*[compute_average_mse_and_std_error(p, squared_errors) for p in percentages])\n", "\n", "# Plotting with error bars\n", "plt.errorbar(percentages * 100, average_mse_values, yerr=std_errors, fmt='o', ecolor='red', capsize=5)\n", "plt.ylim(1.05, 1.25)\n", "plt.xlabel('Percentage of Test Set Used (%)')\n", "plt.ylabel('Sample MSE')\n", "plt.title('Sample MSE vs. Percentage of Test Set Used')\n", "plt.grid(True)\n", "plt.show()\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Running the code above a few times, we see that it captures how we should trust earlier points less... but it seems to be a bit over-confidenct, suggesting the true values are closer to the points than they really are.\n", "\n", "We can view these error bars as providing a 95% confidence interval, **however**:\n", "- This assumes that the squared errors are normally distributed.\n", "- The confidence interval is 1.96 times the error bar width.\n", "- The confidence interval should only contain the true MSE (**seemingly** just below 1.150) 95% of the time.\n", " - We'll come back to this \"seemingly\" later.\n", "\n", "Let's investigate each of these a little more. First, are the squared errors normally distributed?" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk0AAAHHCAYAAACiOWx7AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAABamUlEQVR4nO3deVxU9d4H8M8AM8MiIGhsCohLroiKaagpprLItVCzckVDTS+kiKFRahjdcMmttMinUm9pplZWroy44IKaCJqm5oLiApgrCjgMzHn+8OE8jiAcYHCG8fN+veZ1m9/5nXO+3xHxc885c45MEAQBRERERFQhM0MXQERERFQXMDQRERERScDQRERERCQBQxMRERGRBAxNRERERBIwNBERERFJwNBEREREJAFDExEREZEEDE1EREREEjA0ERmJJk2aYPTo0YYuw+TNnz8fTZs2hbm5OTp06GDocozWypUrIZPJcPHiRUOXQmQ0GJqIakHpPzhHjhwpd7m/vz/atWtX4/1s2bIFcXFxNd7OsyIpKQnTpk1D9+7dsWLFCnzyyScVzv/999/Rq1cvODk5wdraGk2bNsXrr7+Obdu2PaWKjV9cXBxkMtkTXzk5OYYukUhvLAxdABE9dObMGZiZVe3/x2zZsgXLli1jcJJo586dMDMzwzfffAOFQlHh3E8//RQxMTHo1asXYmNjYW1tjXPnzmHHjh1Yu3YtgoKCnlLVdcOXX36JevXqlRmvX7/+0y+GqJYwNBEZCaVSaegSqiw/Px82NjaGLkOy69evw8rKqtLAVFxcjPj4ePTr1w9JSUnlbsdYFRcXQ6vVVtqjvr322mto2LBhldZ58OABFApFuf9noaY/W1qtFkVFRbC0tKz2Nogex9NzREbi8WuaNBoNZs+ejRYtWsDS0hINGjRAjx49oFKpAACjR4/GsmXLAEDndEip/Px8TJ06Fe7u7lAqlWjZsiU+/fRTCIKgs9/CwkJMmjQJDRs2hK2tLV555RVcvXoVMplM5whW6WmYv/76C8OGDYODgwN69OgBADh+/DhGjx6Npk2bwtLSEi4uLnjrrbdw8+ZNnX2VbuPvv//GiBEjYG9vj+eeew4zZ86EIAi4fPkyXn31VdjZ2cHFxQULFiyQ9NmVhpxmzZpBqVSiSZMmeP/996FWq8U5MpkMK1asQH5+vvhZrVy5stzt3bhxA3l5eejevXu5y52cnHTeX7lyBaGhobCxsYGTkxOmTJmC7du3QyaTYffu3eK8J1235u/vD39/f/F9UVERZs2aBV9fX9jb28PGxgYvvfQSdu3apbPexYsXIZPJ8Omnn2Lx4sVi/3/99RcA4PTp03jttdfg6OgIS0tLdO7cGb/99luZ/Z88eRIvv/wyrKys0LhxY3z88cfQarXl9l5du3fvhkwmw9q1azFjxgw0atQI1tbWyMvLw+jRo1GvXj2cP38e/fv3h62tLYYPHw5A+s+xTCZDZGQkVq9ejbZt20KpVIqnUdeuXQtfX1/Y2trCzs4O3t7eWLJkiV77o2cDjzQR1aK7d+/ixo0bZcY1Gk2l68bFxSEhIQFjx45Fly5dkJeXhyNHjuDo0aPo168f3n77bVy7dg0qlQrfffedzrqCIOCVV17Brl27EB4ejg4dOmD79u2IiYnB1atXsWjRInHu6NGjsW7dOowcORIvvvgi9uzZg5CQkCfWNWTIELRo0QKffPKJ+A+XSqXChQsXMGbMGLi4uODkyZNYvnw5Tp48iYMHD+qEOQB444030Lp1a8yZMwebN2/Gxx9/DEdHR3z11Vd4+eWXMXfuXKxevRrvvvsuXnjhBfTs2bPCz2rs2LFYtWoVXnvtNUydOhWHDh1CQkICTp06hV9++QUA8N1332H58uU4fPgwvv76awBAt27dyt2ek5MTrKys8Pvvv+Odd96Bo6PjE/ddWFiIPn36ICsrC5MmTYKbmxu+++477Ny5s8KaK5KXl4evv/4aQ4cOxbhx43Dv3j188803CAwMxOHDh8tcwL5ixQo8ePAA48ePh1KphKOjI06ePInu3bujUaNGeO+992BjY4N169YhNDQUP/30EwYOHAgAyMnJQe/evVFcXCzOW758OaysrKpU861bt8qMWVhYlDk9Fx8fD4VCgXfffRdqtVo8IlZcXIzAwED06NEDn376Kaytrav0cww8PP26bt06REZGomHDhmjSpAlUKhWGDh2KPn36YO7cuQCAU6dOYf/+/Zg8eXKVeiSCQER6t2LFCgFAha+2bdvqrOPp6SmEhYWJ7318fISQkJAK9xMRESGU99d448aNAgDh448/1hl/7bXXBJlMJpw7d04QBEFIS0sTAAhRUVE680aPHi0AED788ENx7MMPPxQACEOHDi2zv4KCgjJjP/zwgwBASElJKbON8ePHi2PFxcVC48aNBZlMJsyZM0ccv337tmBlZaXzmZQnIyNDACCMHTtWZ/zdd98VAAg7d+4Ux8LCwgQbG5sKt1dq1qxZAgDBxsZGCA4OFv7zn/8IaWlpZeYtXrxYACCsW7dOHMvPzxeaN28uABB27doljj/+Z1yqV69eQq9evcT3xcXFglqt1plz+/ZtwdnZWXjrrbfEsczMTAGAYGdnJ1y/fl1nfp8+fQRvb2/hwYMH4phWqxW6desmtGjRQhyLiooSAAiHDh0Sx65fvy7Y29sLAITMzMwnfkaC8P9/puW9WrZsKc7btWuXAEBo2rRpmZ+XsLAwAYDw3nvv6YxL/TkWBEEAIJiZmQknT57UmTt58mTBzs5OKC4urrAPIil4eo6oFi1btgwqlarMq3379pWuW79+fZw8eRJnz56t8n63bNkCc3NzTJo0SWd86tSpEAQBW7duBQDx9MW///1vnXnvvPPOE7c9YcKEMmOPHpV48OABbty4gRdffBEAcPTo0TLzx44dK/63ubk5OnfuDEEQEB4eLo7Xr18fLVu2xIULF55YC/CwVwCIjo7WGZ86dSoAYPPmzRWu/ySzZ8/GmjVr0LFjR2zfvh0ffPABfH190alTJ5w6dUpn/66urnjttdfEMWtra4wfP75a+wUefialR2C0Wi1u3bqF4uJidO7cudzPc/DgwXjuuefE97du3cLOnTvx+uuv4969e7hx4wZu3LiBmzdvIjAwEGfPnsXVq1fF+l988UV06dJFXP+5554TT49J9dNPP5X5OV+xYkWZeWFhYU88ijVx4kSd91J/jkv16tULbdq00RmrX78+8vPzxdPaRDXB03NEtahLly7o3LlzmXEHB4dyT9s96qOPPsKrr76K559/Hu3atUNQUBBGjhwpKXBdunQJbm5usLW11Rlv3bq1uLz0f83MzODl5aUzr3nz5k/c9uNzgYf/SM+ePRtr164tc5H03bt3y8z38PDQeW9vbw9LS8syFxLb29uXuS7qcaU9PF6zi4sL6tevL/ZaHUOHDsXQoUORl5eHQ4cOYeXKlVizZg0GDBiAEydOwNLSEpcuXULz5s3LnIJs2bJltfcLAKtWrcKCBQtw+vRpndO55X3+j4+dO3cOgiBg5syZmDlzZrnbv379Oho1aoRLly6ha9euZZZXtf6ePXtKuhC8vPqBh6fyGjdurDMm9ee4om3/+9//xrp16xAcHIxGjRohICAAr7/+Or/9SNXC0ERkpHr27Inz58/j119/RVJSEr7++mssWrQIiYmJOkdqnrbyjhK8/vrrOHDgAGJiYtChQwfUq1cPWq0WQUFB5V5QbG5uLmkMQJkLfp/k8dCiT3Z2dujXrx/69esHuVyOVatW4dChQ+jVq1eVtvOkGktKSnT6//777zF69GiEhoYiJiYGTk5OMDc3R0JCAs6fP19m/cf/TEo/83fffReBgYHl7rOiYFybnnSUSalUVvmWG1K27eTkhIyMDGzfvh1bt27F1q1bsWLFCowaNQqrVq2q0f7o2cPQRGTEHB0dMWbMGIwZMwb3799Hz549ERcXJ4amJ/0j7OnpiR07duDevXs6/y/99OnT4vLS/9VqtcjMzESLFi3EeefOnZNc4+3bt5GcnIzZs2dj1qxZ4nh1TitWR2kPZ8+eFY9AAEBubi7u3Lkj9qovnTt3xqpVq5CdnS3u/8SJExAEQefP48yZM2XWdXBwwJ07d8qMX7p0CU2bNhXfb9iwAU2bNsXPP/+ss80PP/xQUo2l25LL5ejbt2+Fcz09Pcv9syqv/qdN6s9xZRQKBQYMGIABAwZAq9Xi3//+N7766ivMnDnTYOGR6iZe00RkpB4/LVWvXj00b95c52v0pfexefwf4v79+6OkpARLly7VGV+0aBFkMhmCg4MBQDwK8cUXX+jM+/zzzyXXWXqE5PEjQosXL5a8jZro379/uftbuHAhAFT4TcAnKSgoQGpqarnLSq+jKT191b9/f1y7dg0bNmzQWX/58uVl1m3WrBkOHjyIoqIicWzTpk24fPmyzrzyPtNDhw49sabHOTk5wd/fH1999ZUY7h71zz//iP/dv39/HDx4EIcPH9ZZvnr1akn7qk1Sf44r8vjfIzMzM/EU96N/l4ik4JEmIiPVpk0b+Pv7w9fXF46Ojjhy5Ag2bNiAyMhIcY6vry8AYNKkSQgMDIS5uTnefPNNDBgwAL1798YHH3yAixcvwsfHB0lJSfj1118RFRWFZs2aiesPHjwYixcvxs2bN8VbDvz9998ApJ3ysrOzQ8+ePTFv3jxoNBo0atQISUlJyMzMrIVPpSwfHx+EhYVh+fLluHPnDnr16oXDhw9j1apVCA0NRe/evau8zYKCAnTr1g0vvvgigoKC4O7ujjt37mDjxo3Yu3cvQkND0bFjRwDAuHHjsHTpUowaNQppaWlwdXXFd999B2tr6zLbHTt2LDZs2ICgoCC8/vrrOH/+PL7//nvxz6PUv/71L/z8888YOHAgQkJCkJmZicTERLRp0wb379+X1MOyZcvQo0cPeHt7Y9y4cWjatClyc3ORmpqKK1eu4NixYwCAadOm4bvvvkNQUBAmT54s3nLA09MTx48fl/yZbdiwodw7gvfr1w/Ozs6St/MoqT/HFRk7dixu3bqFl19+GY0bN8alS5fw+eefo0OHDjpHJokkMdwX94hMV+ktB/74449yl/fq1avSWw58/PHHQpcuXYT69esLVlZWQqtWrYT//Oc/QlFRkTinuLhYeOedd4TnnntOkMlkOrcfuHfvnjBlyhTBzc1NkMvlQosWLYT58+cLWq1WZ7/5+flCRESE4OjoKNSrV08IDQ0Vzpw5IwDQuQVA6VfL//nnnzL9XLlyRRg4cKBQv359wd7eXhgyZIhw7dq1J9624PFtPOlWAOV9TuXRaDTC7NmzBS8vL0Eulwvu7u5CbGysztftK9pPedv7n//5HyE0NFTw9PQUlEqlYG1tLXTs2FGYP39+mdsBXLp0SXjllVcEa2troWHDhsLkyZOFbdu2lbnlgCAIwoIFC4RGjRoJSqVS6N69u3DkyJEytxzQarXCJ598Iu67Y8eOwqZNm4SwsDDB09NTnFd6y4H58+eX28f58+eFUaNGCS4uLoJcLhcaNWok/Otf/xI2bNigM+/48eNCr169BEtLS6FRo0ZCfHy88M0339T4lgOP9l96y4H169eX2UZFfy5Sf44BCBEREWXW37BhgxAQECA4OTkJCoVC8PDwEN5++20hOzu7wr6IyiMTBIlXWRLRMyMjIwMdO3bE999/X+WvntNDu3fvRu/evbFr1y6du30TUd3Fa5qInnGFhYVlxhYvXgwzM7NK78RNRPQs4TVNRM+4efPmIS0tDb1794aFhYX4tezx48fD3d3d0OURERkNhiaiZ1y3bt2gUqkQHx+P+/fvw8PDA3Fxcfjggw8MXRoRkVHhNU1EREREEvCaJiIiIiIJDBqaEhIS8MILL8DW1hZOTk4IDQ0tcxfaBw8eICIiAg0aNEC9evUwePBg5Obm6szJyspCSEgIrK2t4eTkhJiYGBQXF+vM2b17Nzp16gSlUonmzZtj5cqVZepZtmwZmjRpAktLS3Tt2lXnZm9ERET0bDPoNU179uxBREQEXnjhBRQXF+P9999HQEAA/vrrL/FOx1OmTMHmzZuxfv162NvbIzIyEoMGDcL+/fsBPHxmU0hICFxcXHDgwAFkZ2dj1KhRkMvl+OSTTwAAmZmZCAkJwYQJE7B69WokJydj7NixcHV1Fe+I/OOPPyI6OhqJiYno2rUrFi9ejMDAQJw5cwZOTk6V9qLVanHt2jXY2trW6jOwiIiISH8EQcC9e/fg5uZW+fMPDXqXqMdcv35dACDs2bNHEARBuHPnjiCXy3Vuhnbq1CkBgJCamioIgiBs2bJFMDMzE3JycsQ5X375pWBnZyfegG7atGllbpD3xhtvCIGBgeL7Ll266NwYraSkRHBzcxMSEhIk1X758uUKb/DGF1988cUXX3wZ7+vy5cuV/ltvVN+eu3v3LoCHDykFgLS0NGg0Gp0HTrZq1QoeHh5ITU3Fiy++iNTUVHh7e+vcpj8wMBATJ07EyZMn0bFjR6SmppZ5aGVgYCCioqIAAEVFRUhLS0NsbKy43MzMDH379n3is57UarXOc4uE/7uePjMzU+fBkvqg0Wiwa9cu9O7dG3K5XK/bNgbsr+4z9R5NvT/A9Htkf3VfbfV47949eHl5Sfq322hCk1arRVRUFLp374527doBAHJycqBQKFC/fn2duc7OzsjJyRHnPP5co9L3lc3Jy8tDYWEhbt++jZKSknLnlD5N+3EJCQmYPXt2mfHU1NRynzlVU9bW1jh06JDet2ss2F/dZ+o9mnp/gOn3yP7qvtrosaCgAIC0Z20aTWiKiIjAiRMnsG/fPkOXIklsbCyio6PF93l5eXB3d0dAQADs7Oz0ui+NRgOVSoV+/fqZ5P+DYH91n6n3aOr9AabfI/ur+2qrx7y8PMlzjSI0RUZGYtOmTUhJSUHjxo3FcRcXFxQVFeHOnTs6R5tyc3Ph4uIiznn8W26l3657dM7j37jLzc2FnZ0drKysYG5uDnNz83LnlG7jcUqlEkqlssy4XC6vtR/Y2ty2MWB/dZ+p92jq/QGm3yP7q/v03WNVtmXQWw4IgoDIyEj88ssv2LlzJ7y8vHSW+/r6Qi6XIzk5WRw7c+YMsrKy4OfnBwDw8/PDn3/+ievXr4tzVCoV7Ozs0KZNG3HOo9sonVO6DYVCAV9fX505Wq0WycnJ4hwiIiJ6thn0SFNERATWrFmDX3/9Fba2tuI1SPb29rCysoK9vT3Cw8MRHR0NR0dH2NnZ4Z133oGfnx9efPFFAEBAQADatGmDkSNHYt68ecjJycGMGTMQEREhHgmaMGECli5dimnTpuGtt97Czp07sW7dOmzevFmsJTo6GmFhYejcuTO6dOmCxYsXIz8/H2PGjHn6HwwREREZHYOGpi+//BIA4O/vrzO+YsUKjB49GgCwaNEimJmZYfDgwVCr1QgMDMQXX3whzjU3N8emTZswceJE+Pn5wcbGBmFhYfjoo4/EOV5eXti8eTOmTJmCJUuWoHHjxvj666/FezQBwBtvvIF//vkHs2bNQk5ODjp06IBt27aVuTiciIiInk0GDU2ChMfeWVpaYtmyZVi2bNkT53h6emLLli0Vbsff3x/p6ekVzomMjERkZGSlNREREdGzh8+eIyIiIpKAoYmIiIhIAoYmIiIiIgkYmoiIiIgkYGgiIiIikoChiYiIiEgChiYiIiIiCRiaiIiIiCQwigf2kjTHjh2DmVnVc27Dhg3h4eFRCxURERE9Oxia6oArV64AAHr27InCwsIqr29pZY0zp08xOBEREdUAQ1MdcPPmTQCAY9A7KLFzq9K6mpuXcXPTAty4cYOhiYiIqAYYmuoQuWMjWDRsZugyiIiInkm8EJyIiIhIAoYmIiIiIgkYmoiIiIgkYGgiIiIikoChiYiIiEgChiYiIiIiCRiaiIiIiCRgaCIiIiKSgKGJiIiISAKGJiIiIiIJGJqIiIiIJGBoIiIiIpKAoYmIiIhIAoYmIiIiIgkYmoiIiIgkYGgiIiIikoChiYiIiEgChiYiIiIiCRiaiIiIiCRgaCIiIiKSgKGJiIiISAKGJiIiIiIJGJqIiIiIJDBoaEpJScGAAQPg5uYGmUyGjRs36iyXyWTlvubPny/OadKkSZnlc+bM0dnO8ePH8dJLL8HS0hLu7u6YN29emVrWr1+PVq1awdLSEt7e3tiyZUut9ExERER1k0FDU35+Pnx8fLBs2bJyl2dnZ+u8vv32W8hkMgwePFhn3kcffaQz75133hGX5eXlISAgAJ6enkhLS8P8+fMRFxeH5cuXi3MOHDiAoUOHIjw8HOnp6QgNDUVoaChOnDhRO40TERFRnWNhyJ0HBwcjODj4ictdXFx03v/666/o3bs3mjZtqjNua2tbZm6p1atXo6ioCN9++y0UCgXatm2LjIwMLFy4EOPHjwcALFmyBEFBQYiJiQEAxMfHQ6VSYenSpUhMTKxJi0RERGQiDBqaqiI3NxebN2/GqlWryiybM2cO4uPj4eHhgWHDhmHKlCmwsHjYWmpqKnr27AmFQiHODwwMxNy5c3H79m04ODggNTUV0dHROtsMDAwsc7rwUWq1Gmq1Wnyfl5cHANBoNNBoNDVptQytVgsAUFrIIJgLVVpXZiGDlZUVtFqt3uvSl9K6jLW+mjL1/gDT79HU+wNMv0f2V/fVVo9V2V6dCU2rVq2Cra0tBg0apDM+adIkdOrUCY6Ojjhw4ABiY2ORnZ2NhQsXAgBycnLg5eWls46zs7O4zMHBATk5OeLYo3NycnKeWE9CQgJmz55dZjwpKQnW1tbV6rEyc4M9AJRUcS1PYMAPuHr1Kq5evVobZemNSqUydAm1ytT7A0y/R1PvDzD9Htlf3afvHgsKCiTPrTOh6dtvv8Xw4cNhaWmpM/7oEaL27dtDoVDg7bffRkJCApRKZa3VExsbq7PvvLw8uLu7IyAgAHZ2dnrdV3p6OrKzszF9axaEBl6Vr/CIotwLyF3zHlJSUuDj46PXuvRFo9FApVKhX79+kMvlhi5H70y9P8D0ezT1/gDT75H91X211WPpmSIp6kRo2rt3L86cOYMff/yx0rldu3ZFcXExLl68iJYtW8LFxQW5ubk6c0rfl14H9aQ5T7pOCgCUSmW5oUwul+v9B9bM7OH1+upiAUKJrErrqosFFBYWwszMzOj/ItXGZ2dMTL0/wPR7NPX+ANPvkf3VffrusSrbqhP3afrmm2/g6+sr6UhJRkYGzMzM4OTkBADw8/NDSkqKzjlLlUqFli1bwsHBQZyTnJyssx2VSgU/Pz89dkFERER1mUFD0/3795GRkYGMjAwAQGZmJjIyMpCVlSXOycvLw/r16zF27Ngy66empmLx4sU4duwYLly4gNWrV2PKlCkYMWKEGIiGDRsGhUKB8PBwnDx5Ej/++COWLFmic2pt8uTJ2LZtGxYsWIDTp08jLi4OR44cQWRkZO1+AERERFRnGPT03JEjR9C7d2/xfWmQCQsLw8qVKwEAa9euhSAIGDp0aJn1lUol1q5di7i4OKjVanh5eWHKlCk6gcje3h5JSUmIiIiAr68vGjZsiFmzZom3GwCAbt26Yc2aNZgxYwbef/99tGjRAhs3bkS7du1qqXMiIiKqawwamvz9/SEIFX+Ffvz48ToB51GdOnXCwYMHK91P+/btsXfv3grnDBkyBEOGDKl0W0RERPRsqhPXNBEREREZGkMTERERkQQMTUREREQSMDQRERERScDQRERERCQBQxMRERGRBAxNRERERBIwNBERERFJwNBEREREJAFDExEREZEEDE1EREREEjA0EREREUnA0EREREQkAUMTERERkQQMTUREREQSMDQRERERScDQRERERCQBQxMRERGRBAxNRERERBIwNBERERFJwNBEREREJAFDExEREZEEDE1EREREEjA0EREREUnA0EREREQkAUMTERERkQQMTUREREQSMDQRERERScDQRERERCQBQxMRERGRBAxNRERERBIwNBERERFJwNBEREREJAFDExEREZEEDE1EREREEhg0NKWkpGDAgAFwc3ODTCbDxo0bdZaPHj0aMplM5xUUFKQz59atWxg+fDjs7OxQv359hIeH4/79+zpzjh8/jpdeegmWlpZwd3fHvHnzytSyfv16tGrVCpaWlvD29saWLVv03i8RERHVXQYNTfn5+fDx8cGyZcueOCcoKAjZ2dni64cfftBZPnz4cJw8eRIqlQqbNm1CSkoKxo8fLy7Py8tDQEAAPD09kZaWhvnz5yMuLg7Lly8X5xw4cABDhw5FeHg40tPTERoaitDQUJw4cUL/TRMREVGdZGHInQcHByM4OLjCOUqlEi4uLuUuO3XqFLZt24Y//vgDnTt3BgB8/vnn6N+/Pz799FO4ublh9erVKCoqwrfffguFQoG2bdsiIyMDCxcuFMPVkiVLEBQUhJiYGABAfHw8VCoVli5disTERD12TERERHWVQUOTFLt374aTkxMcHBzw8ssv4+OPP0aDBg0AAKmpqahfv74YmACgb9++MDMzw6FDhzBw4ECkpqaiZ8+eUCgU4pzAwEDMnTsXt2/fhoODA1JTUxEdHa2z38DAwDKnCx+lVquhVqvF93l5eQAAjUYDjUajj9ZFWq0WAKC0kEEwF6q0rsxCBisrK2i1Wr3XpS+ldRlrfTVl6v0Bpt+jqfcHmH6P7K/uq60eq7I9ow5NQUFBGDRoELy8vHD+/Hm8//77CA4ORmpqKszNzZGTkwMnJyeddSwsLODo6IicnBwAQE5ODry8vHTmODs7i8scHByQk5Mjjj06p3Qb5UlISMDs2bPLjCclJcHa2rpa/VZmbrAHgJIqruUJDPgBV69exdWrV2ujLL1RqVSGLqFWmXp/gOn3aOr9AabfI/ur+/TdY0FBgeS5Rh2a3nzzTfG/vb290b59ezRr1gy7d+9Gnz59DFgZEBsbq3N0Ki8vD+7u7ggICICdnZ1e95Weno7s7GxM35oFoYFX5Ss8oij3AnLXvIeUlBT4+PjotS590Wg0UKlU6NevH+RyuaHL0TtT7w8w/R5NvT/A9Htkf3VfbfVYeqZICqMOTY9r2rQpGjZsiHPnzqFPnz5wcXHB9evXdeYUFxfj1q1b4nVQLi4uyM3N1ZlT+r6yOU+6lgp4eK2VUqksMy6Xy/X+A2tm9vB6fXWxAKFEVqV11cUCCgsLYWZmZvR/kWrjszMmpt4fYPo9mnp/gOn3yP7qPn33WJVt1an7NF25cgU3b96Eq6srAMDPzw937txBWlqaOGfnzp3QarXo2rWrOCclJUXnnKVKpULLli3h4OAgzklOTtbZl0qlgp+fX223RERERHWEQUPT/fv3kZGRgYyMDABAZmYmMjIykJWVhfv37yMmJgYHDx7ExYsXkZycjFdffRXNmzdHYGAgAKB169YICgrCuHHjcPjwYezfvx+RkZF488034ebmBgAYNmwYFAoFwsPDcfLkSfz4449YsmSJzqm1yZMnY9u2bViwYAFOnz6NuLg4HDlyBJGRkU/9MyEiIiLjZNDQdOTIEXTs2BEdO3YEAERHR6Njx46YNWsWzM3Ncfz4cbzyyit4/vnnER4eDl9fX+zdu1fntNjq1avRqlUr9OnTB/3790ePHj107sFkb2+PpKQkZGZmwtfXF1OnTsWsWbN07uXUrVs3rFmzBsuXL4ePjw82bNiAjRs3ol27dk/vwyAiIiKjZtBrmvz9/SEIT/4K/fbt2yvdhqOjI9asWVPhnPbt22Pv3r0VzhkyZAiGDBlS6f6IiIjo2VSnrmkiIiIiMhSGJiIiIiIJGJqIiIiIJGBoIiIiIpKAoYmIiIhIAoYmIiIiIgkYmoiIiIgkYGgiIiIikoChiYiIiEgChiYiIiIiCRiaiIiIiCRgaCIiIiKSgKGJiIiISAKGJiIiIiIJGJqIiIiIJGBoIiIiIpKAoYmIiIhIAoYmIiIiIgkYmoiIiIgkYGgiIiIikoChiYiIiEgChiYiIiIiCRiaiIiIiCRgaCIiIiKSgKGJiIiISAKGJiIiIiIJGJqIiIiIJGBoIiIiIpKAoYmIiIhIAoYmIiIiIgkYmoiIiIgkYGgiIiIikoChiYiIiEgChiYiIiIiCRiaiIiIiCQwaGhKSUnBgAED4ObmBplMho0bN4rLNBoNpk+fDm9vb9jY2MDNzQ2jRo3CtWvXdLbRpEkTyGQyndecOXN05hw/fhwvvfQSLC0t4e7ujnnz5pWpZf369WjVqhUsLS3h7e2NLVu21ErPREREVDcZNDTl5+fDx8cHy5YtK7OsoKAAR48excyZM3H06FH8/PPPOHPmDF555ZUycz/66CNkZ2eLr3feeUdclpeXh4CAAHh6eiItLQ3z589HXFwcli9fLs45cOAAhg4divDwcKSnpyM0NBShoaE4ceJE7TROREREdY6FIXceHByM4ODgcpfZ29tDpVLpjC1duhRdunRBVlYWPDw8xHFbW1u4uLiUu53Vq1ejqKgI3377LRQKBdq2bYuMjAwsXLgQ48ePBwAsWbIEQUFBiImJAQDEx8dDpVJh6dKlSExM1EerREREVMcZNDRV1d27dyGTyVC/fn2d8Tlz5iA+Ph4eHh4YNmwYpkyZAguLh62lpqaiZ8+eUCgU4vzAwEDMnTsXt2/fhoODA1JTUxEdHa2zzcDAQJ3ThY9Tq9VQq9Xi+7y8PAAPTytqNJoadqpLq9UCAJQWMgjmQpXWlVnIYGVlBa1Wq/e69KW0LmOtr6ZMvT/A9Hs09f4A0++R/dV9tdVjVbZXZ0LTgwcPMH36dAwdOhR2dnbi+KRJk9CpUyc4OjriwIEDiI2NRXZ2NhYuXAgAyMnJgZeXl862nJ2dxWUODg7IyckRxx6dk5OT88R6EhISMHv27DLjSUlJsLa2rnafFZkb7AGgpIpreQIDfsDVq1dx9erV2ihLbx4/smhqTL0/wPR7NPX+ANPvkf3VffrusaCgQPLcOhGaNBoNXn/9dQiCgC+//FJn2aNHiNq3bw+FQoG3334bCQkJUCqVtVZTbGyszr7z8vLg7u6OgIAAnVCnD+np6cjOzsb0rVkQGnhVvsIjinIvIHfNe0hJSYGPj49e69IXjUYDlUqFfv36QS6XG7ocvTP1/gDT79HU+wNMv0f2V/fVVo+lZ4qkMPrQVBqYLl26hJ07d1YaSLp27Yri4mJcvHgRLVu2hIuLC3Jzc3XmlL4vvQ7qSXOedJ0UACiVynJDmVwu1/sPrJnZw+v11cUChBJZldZVFwsoLCyEmZmZ0f9Fqo3PzpiYen+A6fdo6v0Bpt8j+6v79N1jVbZl1PdpKg1MZ8+exY4dO9CgQYNK18nIyICZmRmcnJwAAH5+fkhJSdE5Z6lSqdCyZUs4ODiIc5KTk3W2o1Kp4Ofnp8duiIiIqC4z6JGm+/fv49y5c+L7zMxMZGRkwNHREa6urnjttddw9OhRbNq0CSUlJeI1Ro6OjlAoFEhNTcWhQ4fQu3dv2NraIjU1FVOmTMGIESPEQDRs2DDMnj0b4eHhmD59Ok6cOIElS5Zg0aJF4n4nT56MXr16YcGCBQgJCcHatWtx5MgRndsSEBER0bPNoKHpyJEj6N27t/i+9BqhsLAwxMXF4bfffgMAdOjQQWe9Xbt2wd/fH0qlEmvXrkVcXBzUajW8vLwwZcoUnWuN7O3tkZSUhIiICPj6+qJhw4aYNWuWeLsBAOjWrRvWrFmDGTNm4P3330eLFi2wceNGtGvXrha7JyIiorrEoKHJ398fgvDkr9BXtAwAOnXqhIMHD1a6n/bt22Pv3r0VzhkyZAiGDBlS6baIiIjo2WTU1zQRERERGQuGJiIiIiIJGJqIiIiIJGBoIiIiIpKAoYmIiIhIgmqFpgsXLui7DiIiIiKjVq3Q1Lx5c/Tu3Rvff/89Hjx4oO+aiIiIiIxOtULT0aNH0b59e0RHR8PFxQVvv/02Dh8+rO/aiIiIiIxGtUJThw4dsGTJEly7dg3ffvstsrOz0aNHD7Rr1w4LFy7EP//8o+86iYiIiAyqRheCW1hYYNCgQVi/fj3mzp2Lc+fO4d1334W7uztGjRqF7OxsfdVJREREZFA1Ck1HjhzBv//9b7i6umLhwoV49913cf78eahUKly7dg2vvvqqvuokIiIiMqhqPXtu4cKFWLFiBc6cOYP+/fvjv//9L/r37w8zs4cZzMvLCytXrkSTJk30WSsRERGRwVQrNH355Zd46623MHr0aLi6upY7x8nJCd98802NiiMiIiIyFtUKTWfPnq10jkKhQFhYWHU2T0RERGR0qnVN04oVK7B+/foy4+vXr8eqVatqXBQRERGRsalWaEpISEDDhg3LjDs5OeGTTz6pcVFERERExqZaoSkrKwteXl5lxj09PZGVlVXjooiIiIiMTbVCk5OTE44fP15m/NixY2jQoEGNiyIiIiIyNtUKTUOHDsWkSZOwa9culJSUoKSkBDt37sTkyZPx5ptv6rtGIiIiIoOr1rfn4uPjcfHiRfTp0wcWFg83odVqMWrUKF7TRERERCapWqFJoVDgxx9/RHx8PI4dOwYrKyt4e3vD09NT3/URERERGYVqhaZSzz//PJ5//nl91UJERERktKoVmkpKSrBy5UokJyfj+vXr0Gq1Ost37typl+KIiIiIjEW1QtPkyZOxcuVKhISEoF27dpDJZPqui4iIiMioVCs0rV27FuvWrUP//v31XQ8RERGRUarWLQcUCgWaN2+u71qIiIiIjFa1QtPUqVOxZMkSCIKg73qIiIiIjFK1Ts/t27cPu3btwtatW9G2bVvI5XKd5T///LNeiiMiIiIyFtUKTfXr18fAgQP1XQsRERGR0apWaFqxYoW+6yAiIiIyatW6pgkAiouLsWPHDnz11Ve4d+8eAODatWu4f/++3oojIiIiMhbVOtJ06dIlBAUFISsrC2q1Gv369YOtrS3mzp0LtVqNxMREfddJREREZFDVOtI0efJkdO7cGbdv34aVlZU4PnDgQCQnJ+utOCIiIiJjUa0jTXv37sWBAwegUCh0xps0aYKrV6/qpTAiIiIiY1KtI01arRYlJSVlxq9cuQJbW9saF0VERERkbKoVmgICArB48WLxvUwmw/379/Hhhx9W6dEqKSkpGDBgANzc3CCTybBx40ad5YIgYNasWXB1dYWVlRX69u2Ls2fP6sy5desWhg8fDjs7O9SvXx/h4eFlLkY/fvw4XnrpJVhaWsLd3R3z5s0rU8v69evRqlUrWFpawtvbG1u2bJHcBxEREZm+aoWmBQsWYP/+/WjTpg0ePHiAYcOGiafm5s6dK3k7+fn58PHxwbJly8pdPm/ePHz22WdITEzEoUOHYGNjg8DAQDx48ECcM3z4cJw8eRIqlQqbNm1CSkoKxo8fLy7Py8tDQEAAPD09kZaWhvnz5yMuLg7Lly8X5xw4cABDhw5FeHg40tPTERoaitDQUJw4caIanw4RERGZompd09S4cWMcO3YMa9euxfHjx3H//n2Eh4dj+PDhOheGVyY4OBjBwcHlLhMEAYsXL8aMGTPw6quvAgD++9//wtnZGRs3bsSbb76JU6dOYdu2bfjjjz/QuXNnAMDnn3+O/v3749NPP4WbmxtWr16NoqIifPvtt1AoFGjbti0yMjKwcOFCMVwtWbIEQUFBiImJAQDEx8dDpVJh6dKl/CYgERERAahmaAIACwsLjBgxQp+16MjMzEROTg769u0rjtnb26Nr165ITU3Fm2++idTUVNSvX18MTADQt29fmJmZ4dChQxg4cCBSU1PRs2dPnYvWAwMDMXfuXNy+fRsODg5ITU1FdHS0zv4DAwPLnC58lFqthlqtFt/n5eUBADQaDTQaTU3b16HVagEASgsZBPOqPe9PZiGDlZUVtFqt3uvSl9K6jLW+mjL1/gDT79HU+wNMv0f2V/fVVo9V2V61QtN///vfCpePGjWqOpvVkZOTAwBwdnbWGXd2dhaX5eTkwMnJSWe5hYUFHB0ddeZ4eXmV2UbpMgcHB+Tk5FS4n/IkJCRg9uzZZcaTkpJgbW0tpcUqmxvsAaDsBfgV8wQG/ICrV68a/TcbVSqVoUuoVabeH2D6PZp6f4Dp98j+6j5991hQUCB5brVC0+TJk3XeazQaFBQUQKFQwNraWi+hydjFxsbqHJ3Ky8uDu7s7AgICYGdnp9d9paenIzs7G9O3ZkFo4FX5Co8oyr2A3DXvISUlBT4+PnqtS180Gg1UKhX69etX5uHPpsDU+wNMv0dT7w8w/R7ZX91XWz2WnimSolqh6fbt22XGzp49i4kTJ4rXBdWUi4sLACA3Nxeurq7ieG5uLjp06CDOuX79us56xcXFuHXrlri+i4sLcnNzdeaUvq9sTuny8iiVSiiVyjLjcrlc7z+wZmYPr9dXFwsQSmRVWlddLKCwsBBmZmZG/xepNj47Y2Lq/QGm36Op9weYfo/sr+7Td49V2Va1nz33uBYtWmDOnDlljkJVl5eXF1xcXHTuMJ6Xl4dDhw7Bz88PAODn54c7d+4gLS1NnLNz505otVp07dpVnJOSkqJzzlKlUqFly5ZwcHAQ5zx+J3OVSiXuh4iIiEhvoQl4eD3RtWvXJM+/f/8+MjIykJGRAeDhxd8ZGRnIysqCTCZDVFQUPv74Y/z222/4888/MWrUKLi5uSE0NBQA0Lp1awQFBWHcuHE4fPgw9u/fj8jISLz55ptwc3MDAAwbNgwKhQLh4eE4efIkfvzxRyxZskTn1NrkyZOxbds2LFiwAKdPn0ZcXByOHDmCyMhIvX02REREVLdV6/Tcb7/9pvNeEARkZ2dj6dKl6N69u+TtHDlyBL179xbflwaZsLAwrFy5EtOmTUN+fj7Gjx+PO3fuoEePHti2bRssLS3FdVavXo3IyEj06dMHZmZmGDx4MD777DNxub29PZKSkhAREQFfX180bNgQs2bN0rmXU7du3bBmzRrMmDED77//Plq0aIGNGzeiXbt2Vf5siIiIyDRVKzSVHukpJZPJ8Nxzz+Hll1/GggULJG/H398fgvDkr9DLZDJ89NFH+Oijj544x9HREWvWrKlwP+3bt8fevXsrnDNkyBAMGTKk4oKJiIjomVWt0FR63yAiIiKiZ4Ver2kiIiIiMlXVOtL0+N2zK7Jw4cLq7IKIiIjIqFQrNKWnpyM9PR0ajQYtW7YEAPz9998wNzdHp06dxHkyWdXuKURERERkrKoVmgYMGABbW1usWrVKvNfR7du3MWbMGLz00kuYOnWqXoskIiIiMrRqXdO0YMECJCQkiIEJABwcHPDxxx9X6dtzRERERHVFtUJTXl4e/vnnnzLj//zzD+7du1fjooiIiIiMTbVC08CBAzFmzBj8/PPPuHLlCq5cuYKffvoJ4eHhGDRokL5rJCIiIjK4al3TlJiYiHfffRfDhg0Tn+lmYWGB8PBwzJ8/X68FEhERERmDaoUma2trfPHFF5g/fz7Onz8PAGjWrBlsbGz0WhwRERGRsajRzS2zs7ORnZ2NFi1awMbGpsJHohARERHVZdUKTTdv3kSfPn3w/PPPo3///sjOzgYAhIeH83YDREREZJKqFZqmTJkCuVyOrKwsWFtbi+NvvPEGtm3bprfiiIiIiIxFta5pSkpKwvbt29G4cWOd8RYtWuDSpUt6KYyIiIjImFTrSFN+fr7OEaZSt27dglKprHFRRERERMamWqHppZdewn//+1/xvUwmg1arxbx589C7d2+9FUdERERkLKp1em7evHno06cPjhw5gqKiIkybNg0nT57ErVu3sH//fn3XSERERGRw1TrS1K5dO/z999/o0aMHXn31VeTn52PQoEFIT09Hs2bN9F0jERERkcFV+UiTRqNBUFAQEhMT8cEHH9RGTURERERGp8pHmuRyOY4fP14btRAREREZrWqdnhsxYgS++eYbfddCREREZLSqdSF4cXExvv32W+zYsQO+vr5lnjm3cOFCvRRHREREZCyqFJouXLiAJk2a4MSJE+jUqRMA4O+//9aZI5PJ9FcdERERkZGoUmhq0aIFsrOzsWvXLgAPH5vy2WefwdnZuVaKIyIiIjIWVbqmSRAEnfdbt25Ffn6+XgsiIiIiMkbVuhC81OMhioiIiMhUVSk0yWSyMtcs8RomIiIiehZU6ZomQRAwevRo8aG8Dx48wIQJE8p8e+7nn3/WX4VERERERqBKoSksLEzn/YgRI/RaDBEREZGxqlJoWrFiRW3VQURERGTUanQhOBEREdGzgqGJiIiISAKGJiIiIiIJGJqIiIiIJGBoIiIiIpLA6ENTkyZNxJtqPvqKiIgAAPj7+5dZNmHCBJ1tZGVlISQkBNbW1nByckJMTAyKi4t15uzevRudOnWCUqlE8+bNsXLlyqfVIhEREdUBVbrlgCH88ccfKCkpEd+fOHEC/fr1w5AhQ8SxcePG4aOPPhLfW1tbi/9dUlKCkJAQuLi44MCBA8jOzsaoUaMgl8vxySefAAAyMzMREhKCCRMmYPXq1UhOTsbYsWPh6uqKwMDAp9AlERERGTujD03PPfeczvs5c+agWbNm6NWrlzhmbW0NFxeXctdPSkrCX3/9hR07dsDZ2RkdOnRAfHw8pk+fjri4OCgUCiQmJsLLywsLFiwAALRu3Rr79u3DokWLGJqIiIgIQB0ITY8qKirC999/j+joaJ1n3q1evRrff/89XFxcMGDAAMycOVM82pSamgpvb284OzuL8wMDAzFx4kScPHkSHTt2RGpqKvr27auzr8DAQERFRT2xFrVaDbVaLb7Py8sDAGg0Gmg0Gn20K9JqtQAApYUMgnnVHpIss5DBysoKWq1W73XpS2ldxlpfTZl6f4Dp92jq/QGm3yP7q/tqq8eqbK9OhaaNGzfizp07GD16tDg2bNgweHp6ws3NDcePH8f06dNx5swZ8fl3OTk5OoEJgPg+Jyenwjl5eXkoLCyElZVVmVoSEhIwe/bsMuNJSUk6pwf1aW6wB4CSSufp8gQG/ICrV6/i6tWrtVGW3qhUKkOXUKtMvT/A9Hs09f4A0++R/dV9+u6xoKBA8tw6FZq++eYbBAcHw83NTRwbP368+N/e3t5wdXVFnz59cP78eTRr1qzWaomNjUV0dLT4Pi8vD+7u7ggICICdnZ1e95Weno7s7GxM35oFoYFXldYtyr2A3DXvISUlBT4+PnqtS180Gg1UKhX69esHuVxu6HL0ztT7A0y/R1PvDzD9Htlf3VdbPZaeKZKizoSmS5cuYceOHeIRpCfp2rUrAODcuXNo1qwZXFxccPjwYZ05ubm5ACBeB+Xi4iKOPTrHzs6u3KNMAKBUKqFUKsuMy+Vyvf/Ampk9/JKjuliAUCKrZLYudbGAwsJCmJmZGf1fpNr47IyJqfcHmH6Ppt4fYPo9sr+6T989VmVbRn/LgVIrVqyAk5MTQkJCKpyXkZEBAHB1dQUA+Pn54c8//8T169fFOSqVCnZ2dmjTpo04Jzk5WWc7KpUKfn5+euyAiIiI6rI6EZq0Wi1WrFiBsLAwWFj8/8Gx8+fPIz4+Hmlpabh48SJ+++03jBo1Cj179kT79u0BAAEBAWjTpg1GjhyJY8eOYfv27ZgxYwYiIiLEI0UTJkzAhQsXMG3aNJw+fRpffPEF1q1bhylTphikXyIiIjI+dSI07dixA1lZWXjrrbd0xhUKBXbs2IGAgAC0atUKU6dOxeDBg/H777+Lc8zNzbFp0yaYm5vDz88PI0aMwKhRo3Tu6+Tl5YXNmzdDpVLBx8cHCxYswNdff83bDRAREZGoTlzTFBAQAEEo+1V7d3d37Nmzp9L1PT09sWXLlgrn+Pv7Iz09vdo1EhERkWmrE0eaiIiIiAyNoYmIiIhIAoYmIiIiIgkYmoiIiIgkYGgiIiIikoChiYiIiEgChiYiIiIiCRiaiIiIiCRgaCIiIiKSgKGJiIiISAKGJiIiIiIJGJqIiIiIJGBoIiIiIpKAoYmIiIhIAoYmIiIiIgkYmoiIiIgkYGgiIiIikoChiYiIiEgChiYiIiIiCRiaiIiIiCRgaCIiIiKSgKGJiIiISAKGJiIiIiIJGJqIiIiIJGBoIiIiIpKAoYmIiIhIAoYmIiIiIgkYmoiIiIgkYGgiIiIikoChiYiIiEgChiYiIiIiCRiaiIiIiCRgaCIiIiKSgKGJiIiISAKGJiIiIiIJjDo0xcXFQSaT6bxatWolLn/w4AEiIiLQoEED1KtXD4MHD0Zubq7ONrKyshASEgJra2s4OTkhJiYGxcXFOnN2796NTp06QalUonnz5li5cuXTaI+IiIjqEKMOTQDQtm1bZGdni699+/aJy6ZMmYLff/8d69evx549e3Dt2jUMGjRIXF5SUoKQkBAUFRXhwIEDWLVqFVauXIlZs2aJczIzMxESEoLevXsjIyMDUVFRGDt2LLZv3/5U+yQiIiLjZmHoAipjYWEBFxeXMuN3797FN998gzVr1uDll18GAKxYsQKtW7fGwYMH8eKLLyIpKQl//fUXduzYAWdnZ3To0AHx8fGYPn064uLioFAokJiYCC8vLyxYsAAA0Lp1a+zbtw+LFi1CYGDgU+21Np06dapa6zVs2BAeHh56roaIiKjuMfrQdPbsWbi5ucHS0hJ+fn5ISEiAh4cH0tLSoNFo0LdvX3Fuq1at4OHhgdTUVLz44otITU2Ft7c3nJ2dxTmBgYGYOHEiTp48iY4dOyI1NVVnG6VzoqKiKqxLrVZDrVaL7/Py8gAAGo0GGo1GD53/P61WCwBQWsggmAtVW1d9F1bW1hg3bly19m1pZY20I3+gcePG1VpfitLPS9+fm7Ew9f4A0+/R1PsDTL9H9lf31VaPVdmeUYemrl27YuXKlWjZsiWys7Mxe/ZsvPTSSzhx4gRycnKgUChQv359nXWcnZ2Rk5MDAMjJydEJTKXLS5dVNCcvLw+FhYWwsrIqt7aEhATMnj27zHhSUhKsra2r1W9l5gZ7ACip2kpdfIA319Rov8ePH8fx48drtA0pVCpVre/DkEy9P8D0ezT1/gDT75H91X367rGgoEDyXKMOTcHBweJ/t2/fHl27doWnpyfWrVv3xDDztMTGxiI6Olp8n5eXB3d3dwQEBMDOzk6v+0pPT0d2djamb82C0MCrSuvmn9qLW9s+h/OwOVA4N63SukW5F5C75j2kpKTAx8enSutWhUajgUqlQr9+/SCXy2ttP4Zi6v0Bpt+jqfcHmH6P7K/uq60eS88USWHUoelx9evXx/PPP49z586hX79+KCoqwp07d3SONuXm5orXQLm4uODw4cM62yj9dt2jcx7/xl1ubi7s7OwqDGZKpRJKpbLMuFwu1/sPrJnZw+v11cUChBJZldZ9oClBYWEhHlRjXXWxgMLCQpiZmT2Vv4S18dkZE1PvDzD9Hk29P8D0e2R/dZ++e6zKtoz+23OPun//Ps6fPw9XV1f4+vpCLpcjOTlZXH7mzBlkZWXBz88PAODn54c///wT169fF+eoVCrY2dmhTZs24pxHt1E6p3QbRERERICRh6Z3330Xe/bswcWLF3HgwAEMHDgQ5ubmGDp0KOzt7REeHo7o6Gjs2rULaWlpGDNmDPz8/PDiiy8CAAICAtCmTRuMHDkSx44dw/bt2zFjxgxERESIR4kmTJiACxcuYNq0aTh9+jS++OILrFu3DlOmTDFk60RERGRkjPr03JUrVzB06FDcvHkTzz33HHr06IGDBw/iueeeAwAsWrQIZmZmGDx4MNRqNQIDA/HFF1+I65ubm2PTpk2YOHEi/Pz8YGNjg7CwMHz00UfiHC8vL2zevBlTpkzBkiVL0LhxY3z99dcmdbsBIiIiqjmjDk1r166tcLmlpSWWLVuGZcuWPXGOp6cntmzZUuF2/P39kZ6eXq0aiYiI6Nlg1KfniIiIiIwFQxMRERGRBAxNRERERBIwNBERERFJwNBEREREJAFDExEREZEEDE1EREREEjA0EREREUnA0EREREQkAUMTERERkQQMTUREREQSMDQRERERScDQRERERCQBQxMRERGRBAxNRERERBIwNBERERFJwNBEREREJAFDExEREZEEDE1EREREEjA0EREREUnA0EREREQkAUMTERERkQQMTUREREQSMDQRERERScDQRERERCQBQxMRERGRBAxNRERERBIwNBERERFJwNBEREREJAFDExEREZEEFoYugIzfqVOnqrVew4YN4eHhoedqiIiIDIOhiZ6o5P5tQCbDiBEjqrW+pZU1zpw+xeBEREQmgaGJnkirvg8IAhr8ayrkDdyrtK7m5mXc3LQAN27cYGgiIiKTwNBElZI3cIfSpbmhyyAiIjIoo74QPCEhAS+88AJsbW3h5OSE0NBQnDlzRmeOv78/ZDKZzmvChAk6c7KyshASEgJra2s4OTkhJiYGxcXFOnN2796NTp06QalUonnz5li5cmVtt0dERER1iFGHpj179iAiIgIHDx6ESqWCRqNBQEAA8vPzdeaNGzcO2dnZ4mvevHnispKSEoSEhKCoqAgHDhzAqlWrsHLlSsyaNUuck5mZiZCQEPTu3RsZGRmIiorC2LFjsX379qfWKxERERk3oz49t23bNp33K1euhJOTE9LS0tCzZ09x3NraGi4uLuVuIykpCX/99Rd27NgBZ2dndOjQAfHx8Zg+fTri4uKgUCiQmJgILy8vLFiwAADQunVr7Nu3D4sWLUJgYGDtNUhERER1hlEfaXrc3bt3AQCOjo4646tXr0bDhg3Rrl07xMbGoqCgQFyWmpoKb29vODs7i2OBgYHIy8vDyZMnxTl9+/bV2WZgYCBSU1NrqxUiIiKqY4z6SNOjtFotoqKi0L17d7Rr104cHzZsGDw9PeHm5objx49j+vTpOHPmDH7++WcAQE5Ojk5gAiC+z8nJqXBOXl4eCgsLYWVlVaYetVoNtVotvs/LywMAaDQaaDQaPXT8/7RaLQBAaSGDYC5Uad1iuTmsrKxgaSGD4imuK7OQwcrKClqtttLPo3S5vj83Y2Hq/QGm36Op9weYfo/sr+6rrR6rsj2ZIAhV+9fQQCZOnIitW7di3759aNy48RPn7dy5E3369MG5c+fQrFkzjB8/HpcuXdK5PqmgoAA2NjbYsmULgoOD8fzzz2PMmDGIjY0V52zZsgUhISEoKCgoNzTFxcVh9uzZZcbXrFkDa2vrGnZLRERET0NBQQGGDRuGu3fvws7OrsK5deJIU2RkJDZt2oSUlJQKAxMAdO3aFQDE0OTi4oLDhw/rzMnNzQUA8TooFxcXcezROXZ2duUGJgCIjY1FdHS0+D4vLw/u7u4ICAio9EOvqvT0dGRnZ2P61iwIDbyqtG7+qb24te1zOA+bA4Vz06e2blHuBeSueQ8pKSnw8fGpcK5Go4FKpUK/fv0gl8urtJ+6wNT7A0y/R1PvDzD9Htlf3VdbPZaeKZLCqEOTIAh455138Msvv2D37t3w8qo8MGRkZAAAXF1dAQB+fn74z3/+g+vXr8PJyQkAoFKpYGdnhzZt2ohztmzZorMdlUoFPz+/J+5HqVRCqVSWGZfL5Xr/gTUze3jpmbpYgFAiq9K6DzQlKCwsxIOnvK66WEBhYSHMzMwkfx618dkZE1PvDzD9Hk29P8D0e2R/dZ++e6zKtoz6QvCIiAh8//33WLNmDWxtbZGTk4OcnBwUFhYCAM6fP4/4+HikpaXh4sWL+O233zBq1Cj07NkT7du3BwAEBASgTZs2GDlyJI4dO4bt27djxowZiIiIEEPPhAkTcOHCBUybNg2nT5/GF198gXXr1mHKlCkG652IiIiMi1GHpi+//BJ3796Fv78/XF1dxdePP/4IAFAoFNixYwcCAgLQqlUrTJ06FYMHD8bvv/8ubsPc3BybNm2Cubk5/Pz8MGLECIwaNQofffSROMfLywubN2+GSqWCj48PFixYgK+//pq3GyAiIiKR0Z+eq4i7uzv27NlT6XY8PT3LnH57nL+/P9LT06tUHxERET07jPpIExEREZGxYGgiIiIikoChiYiIiEgChiYiIiIiCYz6QnCq+06dOlXpnNLHxBw7dky8J1XDhg3h4eFRq7URERFVBUMT1YqS+7cBmQwjRoyodK6VlRV++OEH9OzZU7wHl6WVNc6cPsXgRERERoOhiWqFVn0fEAQ0+NdUyBu4VzjX0uLh3cadh83Bg2IBmpuXcXPTAty4cYOhiYiIjAZDE9UqeQN3KF2aVzhHYS4AKIHCuWmVH9dCRET0tPBCcCIiIiIJGJqIiIiIJGBoIiIiIpKAoYmIiIhIAoYmIiIiIgkYmoiIiIgkYGgiIiIikoD3aSKjJeURLOXhI1iIiKg2MDSR0anKI1jKw0ewEBFRbWBoIqNTlUewPI6PYCEiotrC0ERGS8ojWIiIiJ4WXghOREREJAFDExEREZEEPD1HJonfvCMiIn1jaCKTwm/eERFRbWFoIpPCb94REVFtYWgik8Rv3hERkb4xNBE9htdDERFReRiaiP4Pr4ciIqKKMDQR/R9eD0VERBVhaCJ6DK+HIiKi8jA0EelReddDabVaAMCxY8dgZlb+/WR5PRQRkfFjaCLSg4quh7KyssIPP/yAnj17orCwsNz1eT0UEZHxY2gi0oOKroeytJABAJyHzcGDYqHMuqXXQ+3duxetW7eu8r55lIqI6OlgaCLSo/Kuh1KYCwBKoHBuCqFEVmYdfmuPiKhuYGgiMjB9fGuPR6mIiGofQxORkajOt/ZqepRKqbTETz9tgKura5XXZeAiomcNQ9Njli1bhvnz5yMnJwc+Pj74/PPP0aVLF0OXRVSumhylenDlJO7s/Br/+te/qrXvRwOXlG8IPoqBi4jqIoamR/z444+Ijo5GYmIiunbtisWLFyMwMBBnzpyBk5OTocsjeqLqHKXS3Lyst8Al5RuCj6rJES61Wg2lUlnl9QCGNSKqGYamRyxcuBDjxo3DmDFjAACJiYnYvHkzvv32W7z33nsGro6odugjcFX2DcFH1fQIF2RmgKCt1qrVDWtVPZL2uJoEPYZEIuPB0PR/ioqKkJaWhtjYWHHMzMwMffv2RWpqqgErIzJepYGrsm8IPqomR7gKLxzB3b3fP/XTkVU9klZGDYLe0wqJjwdDQwW92lq3suBrjDVXhqH46WNo+j83btxASUkJnJ2ddcadnZ1x+vTpMvPVajXUarX4/u7duwCAW7duQaPR6LW2vLw8FBQUQHbrErRFD6q0rtm9bFhaWkJ2MxOCVl35CgZYV2sBFBS4Q5t9GUJx3ai5Kus+3l9dqLmq61bWY3nrKlAMeRX3W2Kmrf66xfmwVCph6/sKzG0bVGld81uXUFBQgIYvhkKjrF+ldTW555B/am+19lujdW9cRv7JZLz22muS5ltZWWHZsmUICAh4GAwNFPRqa90y/T2l/dbmukpLKyz/KhFOTk7QarUoKCjA3r17JR8NNTMzE8NkVRli3dIeb968CblcXq19l+fevXsAAEGo+Ch56SQSBOHq1asCAOHAgQM64zExMUKXLl3KzP/www8FAHzxxRdffPHFlwm8Ll++XGlW4JGm/9OwYUOYm5sjNzdXZzw3NxcuLi5l5sfGxiI6Olp8r9VqcevWLTRo0AAyWcWnJ6oqLy8P7u7uuHz5Muzs7PS6bWPA/uo+U+/R1PsDTL9H9lf31VaPgiDg3r17cHNzq3QuQ9P/USgU8PX1RXJyMkJDQwE8DELJycmIjIwsM1+pVJY5D12/fv1ardHOzs5k/zIA7M8UmHqPpt4fYPo9sr+6rzZ6tLe3lzSPoekR0dHRCAsLQ+fOndGlSxcsXrwY+fn54rfpiIiI6NnF0PSIN954A//88w9mzZqFnJwcdOjQAdu2bStzcTgRERE9exiaHhMZGVnu6ThDUiqV+PDDD6v9tVRjx/7qPlPv0dT7A0y/R/ZX9xlDjzJBkPIdOyIiIqJnW9VvbUtERET0DGJoIiIiIpKAoYmIiIhIAoYmIiIiIgkYmozcsmXL0KRJE1haWqJr1644fPiwoUvSm4SEBLzwwguwtbWFk5MTQkNDcebMGUOXVWvmzJkDmUyGqKgoQ5eiN1evXsWIESPQoEEDWFlZwdvbG0eOHDF0WXpTUlKCmTNnwsvLC1ZWVmjWrBni4+OlPaPKCKWkpGDAgAFwc3ODTCbDxo0bdZYLgoBZs2bB1dUVVlZW6Nu3L86ePWuYYqupoh41Gg2mT58Ob29v2NjYwM3NDaNGjcK1a9cMV3AVVfZn+KgJEyZAJpNh8eLFT62+mpLS36lTp/DKK6/A3t4eNjY2eOGFF5CVlfVU6mNoMmI//vgjoqOj8eGHH+Lo0aPw8fFBYGAgrl+/bujS9GLPnj2IiIjAwYMHoVKpoNFoEBAQgPz8fEOXpnd//PEHvvrqK7Rv397QpejN7du30b17d8jlcmzduhV//fUXFixYAAcHB0OXpjdz587Fl19+iaVLl+LUqVOYO3cu5s2bh88//9zQpVVLfn4+fHx8sGzZsnKXz5s3D5999hkSExNx6NAh2NjYIDAwEA8eVO1B4YZUUY8FBQU4evQoZs6ciaNHj+Lnn3/GmTNn8Morrxig0uqp7M+w1C+//IKDBw9KejSIMamsv/Pnz6NHjx5o1aoVdu/ejePHj2PmzJmwtLR8OgXq42G3VDu6dOkiREREiO9LSkoENzc3ISEhwYBV1Z7r168LAIQ9e/YYuhS9unfvntCiRQtBpVIJvXr1EiZPnmzokvRi+vTpQo8ePQxdRq0KCQkR3nrrLZ2xQYMGCcOHDzdQRfoDQPjll1/E91qtVnBxcRHmz58vjt25c0dQKpXCDz/8YIAKa+7xHstz+PBhAYBw6dKlp1OUHj2pvytXrgiNGjUSTpw4IXh6egqLFi166rXpQ3n9vfHGG8KIESMMU5AgCDzSZKSKioqQlpaGvn37imNmZmbo27cvUlNTDVhZ7bl79y4AwNHR0cCV6FdERARCQkJ0/ixNwW+//YbOnTtjyJAhcHJyQseOHfE///M/hi5Lr7p164bk5GT8/fffAIBjx45h3759CA4ONnBl+peZmYmcnBydn1N7e3t07drVZH/nAA9/78hkslp/dujTotVqMXLkSMTExKBt27aGLkevtFotNm/ejOeffx6BgYFwcnJC165dKzxFqW8MTUbqxo0bKCkpKfMIF2dnZ+Tk5Bioqtqj1WoRFRWF7t27o127doYuR2/Wrl2Lo0ePIiEhwdCl6N2FCxfw5ZdfokWLFti+fTsmTpyISZMmYdWqVYYuTW/ee+89vPnmm2jVqhXkcjk6duyIqKgoDB8+3NCl6V3p75Vn5XcOADx48ADTp0/H0KFDTeYht3PnzoWFhQUmTZpk6FL07vr167h//z7mzJmDoKAgJCUlYeDAgRg0aBD27NnzVGrgY1TIKERERODEiRPYt2+foUvRm8uXL2Py5MlQqVRP73z7U6TVatG5c2d88sknAICOHTvixIkTSExMRFhYmIGr049169Zh9erVWLNmDdq2bYuMjAxERUXBzc3NZHp8Vmk0Grz++usQBAFffvmlocvRi7S0NCxZsgRHjx6FTCYzdDl6p9VqAQCvvvoqpkyZAgDo0KEDDhw4gMTERPTq1avWa+CRJiPVsGFDmJubIzc3V2c8NzcXLi4uBqqqdkRGRmLTpk3YtWsXGjdubOhy9CYtLQ3Xr19Hp06dYGFhAQsLC+zZswefffYZLCwsUFJSYugSa8TV1RVt2rTRGWvduvVT+xbL0xATEyMebfL29sbIkSMxZcoUkzxyWPp75Vn4nVMamC5dugSVSmUyR5n27t2L69evw8PDQ/ydc+nSJUydOhVNmjQxdHk11rBhQ1hYWBj09w5Dk5FSKBTw9fVFcnKyOKbVapGcnAw/Pz8DVqY/giAgMjISv/zyC3bu3AkvLy9Dl6RXffr0wZ9//omMjAzx1blzZwwfPhwZGRkwNzc3dIk10r179zK3iPj777/h6elpoIr0r6CgAGZmur8mzc3Nxf/Ha0q8vLzg4uKi8zsnLy8Phw4dMpnfOcD/B6azZ89ix44daNCggaFL0puRI0fi+PHjOr9z3NzcEBMTg+3btxu6vBpTKBR44YUXDPp7h6fnjFh0dDTCwsLQuXNndOnSBYsXL0Z+fj7GjBlj6NL0IiIiAmvWrMGvv/4KW1tb8boJe3t7WFlZGbi6mrO1tS1zfZaNjQ0aNGhgEtdtTZkyBd26dcMnn3yC119/HYcPH8by5cuxfPlyQ5emNwMGDMB//vMfeHh4oG3btkhPT8fChQvx1ltvGbq0arl//z7OnTsnvs/MzERGRgYcHR3h4eGBqKgofPzxx2jRogW8vLwwc+ZMuLm5ITQ01HBFV1FFPbq6uuK1117D0aNHsWnTJpSUlIi/dxwdHaFQKAxVtmSV/Rk+HgLlcjlcXFzQsmXLp11qtVTWX0xMDN544w307NkTvXv3xrZt2/D7779j9+7dT6dAg31vjyT5/PPPBQ8PD0GhUAhdunQRDh48aOiS9AZAua8VK1YYurRaY0q3HBAEQfj999+Fdu3aCUqlUmjVqpWwfPlyQ5ekV3l5ecLkyZMFDw8PwdLSUmjatKnwwQcfCGq12tClVcuuXbvK/TsXFhYmCMLD2w7MnDlTcHZ2FpRKpdCnTx/hzJkzhi26iirqMTMz84m/d3bt2mXo0iWp7M/wcXXtlgNS+vvmm2+E5s2bC5aWloKPj4+wcePGp1afTBDq6K1tiYiIiJ4iXtNEREREJAFDExEREZEEDE1EREREEjA0EREREUnA0EREREQkAUMTERERkQQMTUREREQSMDQREdWQv78/oqKiDF0GEdUyhiYiemr++ecfTJw4ER4eHlAqlXBxcUFgYCD2799v6NJq1cqVKyGTycq8LC0tDV0aEVUBnz1HRE/N4MGDUVRUhFWrVqFp06bIzc1FcnIybt68aejSUFRUVKvPHrOzsyvzoFGZTFalegRBQElJCSwsqvaru7rrEZEuHmkioqfizp072Lt3L+bOnYvevXvD09MTXbp0QWxsLF555RVx3tmzZ9GzZ09YWlqiTZs2UKlUkMlk2LhxIwBg9+7dkMlkuHPnjrhORkYGZDIZLl68CAC4efMmhg4dikaNGsHa2hre3t744YcfdOrx9/dHZGQkoqKi0LBhQwQGBgIATpw4geDgYNSrVw/Ozs4YOXIkbty4Ia6Xn5+PUaNGoV69enB1dcWCBQsk9S+TyeDi4qLzcnZ2rrCe0l63bt0KX19fKJVK7Nu3D2q1GpMmTYKTkxMsLS3Ro0cP/PHHH+K2nrQeEdUMQxMRPRX16tVDvXr1sHHjRqjV6nLnaLVaDBo0CAqFAocOHUJiYiKmT59e5X09ePAAvr6+2Lx5M06cOIHx48dj5MiROHz4sM68VatWQaFQYP/+/UhMTMSdO3fw8ssvo2PHjjhy5Ai2bduG3NxcvP766+I6MTEx2LNnD3799VckJSVh9+7dOHr0aJVrLM/j9ZR67733MGfOHJw6dQrt27fHtGnT8NNPP2HVqlU4evQomjdvjsDAQNy6dUtne4+vR0Q19NQeDUxEz7wNGzYIDg4OgqWlpdCtWzchNjZWOHbsmLh8+/btgoWFhXD16lVxbOvWrQIA4ZdffhEE4f+fgn779m1xTnp6ugBAyMzMfOK+Q0JChKlTp4rve/XqJXTs2FFnTnx8vBAQEKAzdvnyZQGAcObMGeHevXuCQqEQ1q1bJy6/efOmYGVlJUyePPmJ+16xYoUAQLCxsdF5BQUFVVhPaa+PPsX9/v37glwuF1avXi2OFRUVCW5ubsK8efOeuB4R1RxPcBPRUzN48GCEhIRg7969OHjwILZu3Yp58+bh66+/xujRo3Hq1Cm4u7vDzc1NXMfPz6/K+ykpKcEnn3yCdevW4erVqygqKoJarYa1tbXOPF9fX533x44dw65du1CvXr0y2zx//jwKCwtRVFSErl27iuOOjo5o2bJlpTXZ2tqWOSJlZWVVYT2lOnfurFOHRqNB9+7dxTG5XI4uXbrg1KlTT1yPiGqOoYmInipLS0v069cP/fr1w8yZMzF27Fh8+OGHGD16tKT1zcweXlUgCII4ptFodObMnz8fS5YsweLFi+Ht7Q0bGxtERUWhqKhIZ56NjY3O+/v372PAgAGYO3dumf26urri3Llzkmp8Ut3NmzevcM7j9VQ2XpnqrkdE5eM1TURkUG3atEF+fj4AoHXr1rh8+TKys7PF5QcPHtSZ/9xzzwGAzpyMjAydOfv378err76KESNGwMfHB02bNsXff/9daS2dOnXCyZMn0aRJEzRv3lznZWNjg2bNmkEul+PQoUPiOrdv35a0bX1p1qyZeN1TKY1Ggz/++ANt2rR5anUQPYsYmojoqbh58yZefvllfP/99zh+/DgyMzOxfv16zJs3D6+++ioAoG/fvnj++ecRFhaGY8eOYe/evfjggw90ttO8eXO4u7sjLi4OZ8+exebNm8t8g61FixZQqVQ4cOAATp06hbfffhu5ubmV1hgREYFbt25h6NCh+OOPP3D+/Hls374dY8aMQUlJCerVq4fw8HDExMRg586dOHHiBEaPHi0e/aqIIAjIyckp89JqtVX4FB8ePZo4cSJiYmKwbds2/PXXXxg3bhwKCgoQHh5epW0RUdXw9BwRPRX16tVD165dsWjRIvG6HHd3d4wbNw7vv/8+gIensH755ReEh4ejS5cuaNKkCT777DMEBQWJ25HL5fjhhx8wceJEtG/fHi+88AI+/vhjDBkyRJwzY8YMXLhwAYGBgbC2tsb48eMRGhqKu3fvVlijm5sb9u/fj+nTpyMgIABqtRqenp4ICgoSg9H8+fPF03i2traYOnVqpdsFgLy8PLi6upYZz87OhouLi6TPsNScOXOg1WoxcuRI3Lt3D507d8b27dvh4OBQpe0QUdXIhEcvDCAiMkIymQy//PILQkNDDV0KET3DeHqOiIiISAKGJiIiIiIJeE0TERk9XkVARMaAR5qIiIiIJGBoIiIiIpKAoYmIiIhIAoYmIiIiIgkYmoiIiIgkYGgiIiIikoChiYiIiEgChiYiIiIiCRiaiIiIiCT4X6PSEho7gXzhAAAAAElFTkSuQmCC", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Creating a histogram of the squared errors\n", "plt.hist(squared_errors, bins=30, edgecolor='black')\n", "plt.xlabel('Squared Error')\n", "plt.ylabel('Frequency')\n", "plt.title('Histogram of Squared Errors')\n", "plt.grid(True)\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Yikes, the squared errors are **not** normally distributed. We can still use error bars based on standard deviation, but we should be cautious how much we trust them.\n", "\n", "**Note (this is for your information, but you will not be tested on it)**: By the [central limit theorem](https://en.wikipedia.org/wiki/Central_limit_theorem), the sample mean of any random variable becomes normally distributed as the nmber of samples increases. Also, the reliability of the confidence interval from standard error is related to how close to normally distributed the sample mean is, not each individual sample. So, as the number of samples grows, these error bars should become more and more reliable. You may see texts saying that for more than ~30 samples the central limit theorem kicks in and these normality assumptions are reasonable. That may be true in some cases where samples are already roughly normally distributed (e.g., measurements of quantities in the real-world), but it is often **not** true in ML applications where distributions can be extremely far from normal.\n", "\n", "We've investigated the first of the three bullets:\n", "- This assumes that the squared errors are normally distributed.\n", "- The confidence interval is 1.96 times the error bar width.\n", "- The confidence interval should only contain the true MSE (**seemingly** just below 1.150) 95% of the time.\n", "\n", "Now, let's investigate the second by scaling the error bars by a factor of 1.96.\n", "\n", "**Note: ML texts usually do NOT do this! You must do this in your head!**" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Inflate the error bars\n", "inflated_std_errors = [1.96 * se for se in std_errors]\n", "\n", "# Replot\n", "plt.errorbar(percentages * 100, average_mse_values, yerr=inflated_std_errors, fmt='o', ecolor='red', capsize=5)\n", "plt.ylim(1.05, 1.25)\n", "plt.xlabel('Percentage of Test Set Used (%)')\n", "plt.ylabel('Sample MSE')\n", "plt.title('Sample MSE vs. Percentage of Test Set Used')\n", "plt.grid(True)\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "If you run the plot above a few times, you'll see that it does tend to include the apparent MSE of just below 1.150 in most cases!\n", "\n", "We can also overlay this on our previous plot to see how it captures the variation. Note, however, that the error bars are meant to include the true MSE with probability 0.95. They aren't necessarily meant to capture the range of curves that we see. That is, if the sample MSE is below the true MSE, the error bar should be big enough to reach up to the true MSE 95% of the time, but need not reach up to include 95% of the sample MSEs that we observe." ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Plotting 100 times\n", "for _ in range(100):\n", " average_mse_values = [compute_average_mse(p, squared_errors) for p in percentages]\n", " plt.plot(percentages * 100, average_mse_values, marker='o', alpha=0.2) # alpha for transparency\n", "\n", "plt.errorbar(percentages * 100, average_mse_values, yerr=inflated_std_errors, fmt='o', ecolor='red', capsize=5)\n", "plt.ylim(1.05, 1.25)\n", "\n", "plt.ylim(1.05, 1.25) # Setting y-axis limits\n", "plt.xlabel('Percentage of Test Set Used (%)')\n", "plt.ylabel('Sample MSE')\n", "plt.title('Sample MSE vs. Percentage of Test Set Used')\n", "plt.grid(True)\n", "plt.show()\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "If you run the code above many times, you'll see that the error bars do tend to cover the values that we sample.\n", "\n", "Notice that the error bars are still fairly wide even at 100% of the data.\n", "\n", "**Question**: Why?\n", "\n", "**Answer**: This relates to the \"seemingly\" statement about the true MSE being between 1.125 and 1.15. Even when using 100% Of the data, we're still computing a **sample** MSE. All of our 100 trials will get the exact same value, since they just use all 34,643 points in the testing set, making it look like there is no variance in the MSE when using 100% of the test set. If we had even more data, then the sample MSEs that we saw with 34,943 points (100% in the plot above) would still have variance. The error bars are better here, showing that even with 100% of the testing data we have significant uncertainty about the actual MSE!\n", "\n", "**Question**: If the squared errors really were normally distributed, is the guarantee that the true MSE is within all of the (inflated by x1.96) confidence intervals with probability at least 0.95? Or is the guarantee that, if we were to pick any one point and its error bar (before looking at the data), that one error bar will hold the true MSE with probability at least 0.95?\n", "\n", "**Answer**: The latter. Each error bar can fail with probability 0.05. The chance that one or more error bar does not include the true MSE can be much larger than 5%." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's revisit the table we made, comparing nearest neighbor variants:" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
 ModelMSERMSEMAER^2
0k-NN k=1 sigma=None1.0661881.0325640.793455-0.635682
1k-NN k=100 sigma=None0.5567960.7461870.5873800.145797
2k-NN k=110 sigma=900.5556010.7453860.5866710.147631
\n" ], "text/plain": [ "" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Load the data set\n", "df = pd.read_csv(\"data/GPA.csv\", delimiter=',')\n", "\n", "# We already loaded X and y, but do it again as a reminder\n", "X = df.iloc[:, :-1]\n", "y = df.iloc[:, -1]\n", "\n", "# Split the data into training and testing sets (60% train, 40% test)\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.05, shuffle=True)\n", "\n", "# Model parameters to test\n", "parameters = [\n", " {\"k\": 1, \"sigma\": None}, # Standard NN\n", " {\"k\": 100, \"sigma\": None}, # Standard k-NN\n", " {\"k\": 110, \"sigma\": 90} # Weighted k-NN\n", "]\n", "\n", "# Dictionary to store results\n", "results = []\n", "\n", "# Training and evaluating each model\n", "for param in parameters:\n", " if param[\"sigma\"] is None:\n", " model = KNearestNeighbors(k=param[\"k\"])\n", " else:\n", " model = WeightedKNearestNeighbors(k=param[\"k\"], sigma=param[\"sigma\"])\n", " model.fit(X_train, y_train)\n", " predictions = model.predict(X_test)\n", "\n", " mse = mean_squared_error(predictions, y_test)\n", " rmse = root_mean_squared_error(predictions, y_test)\n", " mae = mean_absolute_error(predictions, y_test)\n", " r2 = r_squared(predictions, y_test)\n", "\n", " results.append({\"Model\": f\"k-NN k={param['k']} sigma={param['sigma']}\", \n", " \"MSE\": mse, \"RMSE\": rmse, \"MAE\": mae, \"R^2\": r2})\n", "\n", "# Creating DataFrame for results\n", "results_df = pd.DataFrame(results)\n", "\n", "# Finding the best (minimum or maximum) values for each metric\n", "best_metrics = {\n", " \"MSE\": results_df['MSE'].idxmin(),\n", " \"RMSE\": results_df['RMSE'].idxmin(),\n", " \"MAE\": results_df['MAE'].idxmin(),\n", " \"R^2\": results_df['R^2'].idxmax()\n", "}\n", "\n", "# Highlighting the best values in the DataFrame\n", "def highlight_best(row, best_metrics):\n", " return ['font-weight: bold' if (col in best_metrics and row.name == best_metrics[col]) else '' for col in row.index]\n", "\n", "# Apply the highlighting\n", "styled_results = results_df.style.apply(highlight_best, best_metrics=best_metrics, axis=1)\n", "styled_results" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Recall that running this many times resulted in different conclusions about which method was the best.\n", "\n", "This is because we are comparing sample MSE values. ML texts reporting tables of results typically include quantification of uncertainty with $\\pm$ values.\n", "\n", "Example: $0.149729 \\pm 0.0013$.\n", "\n", "This latter value provides a measure of how much the sample estimate can be trusted.\n", "\n", "However, different texts show different $\\pm$ values. Common choices include:\n", "- Standard error\n", "- A confidence interval (usually 95%). This is often 1.96 times the standard error, but it can be computed using other (often more reliable, but looser) methods.\n", "- Standard deviation. Note that standard deviation does not provide a confidence interval (which gets tighter as you use more data), but rather a quantification of how much the samples vary independent of the number of samples used.\n", "\n", "Each text or table should indicate what values it is reporting using $\\pm$.\n", "\n", "Let's add $\\pm$ values showing (1.96 x Standard Error). Let's start by updating our evaluation metrics to also report the standard error.\n", "\n", "**Note**: $R^2$ isn't merely the average of some values, so we can't use this approach to get a reasonable confidence interval for $R^2$, so we leave it out. There exist \"bootstrapping\" methods to measure uncertainty about $R^2$ estimates, which are beyond the scope of this class." ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "\n", "def calculate_se(values):\n", " return np.std(values, ddof=1) / np.sqrt(len(values))\n", "\n", "def mean_squared_error_with_se(predictions, labels):\n", " squared_errors = (predictions - labels) ** 2\n", " mse = np.mean(squared_errors)\n", " mse_se = calculate_se(squared_errors)\n", " return mse, mse_se\n", "\n", "def root_mean_squared_error_with_se(predictions, labels):\n", " squared_errors = (predictions - labels) ** 2\n", " rmse = np.sqrt(np.mean(squared_errors))\n", " rmse_se = calculate_se(np.sqrt(squared_errors))\n", " return rmse, rmse_se\n", "\n", "def mean_absolute_error_with_se(predictions, labels):\n", " absolute_errors = np.abs(predictions - labels)\n", " mae = np.mean(absolute_errors)\n", " mae_se = calculate_se(absolute_errors)\n", " return mae, mae_se" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now let's change our loop to use these functions, and our printing to plot the $\\pm$ values." ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
 ModelMSERMSEMAE
0k-NN k=1 sigma=None1.104 ± 0.0751.051 ± 0.0290.803 ± 0.029
1k-NN k=100 sigma=None0.565 ± 0.0410.752 ± 0.0200.586 ± 0.020
2k-NN k=110 sigma=900.565 ± 0.0410.752 ± 0.0200.586 ± 0.020
\n" ], "text/plain": [ "" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Load the data set\n", "df = pd.read_csv(\"data/GPA.csv\", delimiter=',')\n", "\n", "# We already loaded X and y, but do it again as a reminder\n", "X = df.iloc[:, :-1]\n", "y = df.iloc[:, -1]\n", "\n", "# Split the data into training and testing sets (60% train, 40% test)\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.05, shuffle=True)\n", "\n", "# Model parameters to test\n", "parameters = [\n", " {\"k\": 1, \"sigma\": None}, # Standard NN\n", " {\"k\": 100, \"sigma\": None}, # Standard k-NN\n", " {\"k\": 110, \"sigma\": 90} # Weighted k-NN\n", "]\n", "\n", "# Dictionary to store results\n", "results = []\n", "\n", "# Training and evaluating each model\n", "for param in parameters:\n", " if param[\"sigma\"] is None:\n", " model = KNearestNeighbors(k=param[\"k\"])\n", " else:\n", " model = WeightedKNearestNeighbors(k=param[\"k\"], sigma=param[\"sigma\"])\n", " model.fit(X_train, y_train)\n", " predictions = model.predict(X_test)\n", "\n", " mse, mse_se = mean_squared_error_with_se(predictions, y_test)\n", " rmse, rmse_se = root_mean_squared_error_with_se(predictions, y_test)\n", " mae, mae_se = mean_absolute_error_with_se(predictions, y_test)\n", "\n", " results.append({\n", " \"Model\": f\"k-NN k={param['k']} sigma={param['sigma']}\",\n", " \"MSE\": f\"{mse:.3f} ± {1.96 * mse_se:.3f}\",\n", " \"RMSE\": f\"{rmse:.3f} ± {1.96 * rmse_se:.3f}\",\n", " \"MAE\": f\"{mae:.3f} ± {1.96 * mae_se:.3f}\"\n", " })\n", "\n", "# Creating DataFrame for results\n", "results_df = pd.DataFrame(results)\n", "\n", "# Finding the best (minimum or maximum) values for each metric\n", "best_metrics = {\n", " \"MSE\": results_df['MSE'].idxmin(),\n", " \"RMSE\": results_df['RMSE'].idxmin(),\n", " \"MAE\": results_df['MAE'].idxmin()\n", "}\n", "\n", "# Highlighting the best values in the DataFrame\n", "def highlight_best(row, best_metrics):\n", " return ['font-weight: bold' if (col in best_metrics and row.name == best_metrics[col]) else '' for col in row.index]\n", "\n", "# Apply the highlighting\n", "styled_results = results_df.style.apply(highlight_best, best_metrics=best_metrics, axis=1)\n", "styled_results" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "These $\\pm$ values show that we should not see this table and conclude that thre first row is better than the second, or even that the zero'th row is worse than the other two!\n", "\n", "Let's change the train/test split to be more reasonable and try again." ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
 ModelMSERMSEMAE
0k-NN k=1 sigma=None1.134 ± 0.0241.065 ± 0.0090.817 ± 0.009
1k-NN k=100 sigma=None0.573 ± 0.0130.757 ± 0.0060.593 ± 0.006
2k-NN k=110 sigma=900.573 ± 0.0130.757 ± 0.0060.593 ± 0.006
\n" ], "text/plain": [ "" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Load the data set\n", "df = pd.read_csv(\"data/GPA.csv\", delimiter=',')\n", "\n", "# We already loaded X and y, but do it again as a reminder\n", "X = df.iloc[:, :-1]\n", "y = df.iloc[:, -1]\n", "\n", "# Split the data into training and testing sets (60% train, 40% test)\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.5, shuffle=True)\n", "\n", "# Model parameters to test\n", "parameters = [\n", " {\"k\": 1, \"sigma\": None}, # Standard NN\n", " {\"k\": 100, \"sigma\": None}, # Standard k-NN\n", " {\"k\": 110, \"sigma\": 90} # Weighted k-NN\n", "]\n", "\n", "# Dictionary to store results\n", "results = []\n", "\n", "# Training and evaluating each model\n", "for param in parameters:\n", " if param[\"sigma\"] is None:\n", " model = KNearestNeighbors(k=param[\"k\"])\n", " else:\n", " model = WeightedKNearestNeighbors(k=param[\"k\"], sigma=param[\"sigma\"])\n", " model.fit(X_train, y_train)\n", " predictions = model.predict(X_test)\n", "\n", " mse, mse_se = mean_squared_error_with_se(predictions, y_test)\n", " rmse, rmse_se = root_mean_squared_error_with_se(predictions, y_test)\n", " mae, mae_se = mean_absolute_error_with_se(predictions, y_test)\n", "\n", " results.append({\n", " \"Model\": f\"k-NN k={param['k']} sigma={param['sigma']}\",\n", " \"MSE\": f\"{mse:.3f} ± {1.96 * mse_se:.3f}\",\n", " \"RMSE\": f\"{rmse:.3f} ± {1.96 * rmse_se:.3f}\",\n", " \"MAE\": f\"{mae:.3f} ± {1.96 * mae_se:.3f}\"\n", " })\n", "\n", "# Creating DataFrame for results\n", "results_df = pd.DataFrame(results)\n", "\n", "# Finding the best (minimum or maximum) values for each metric\n", "best_metrics = {\n", " \"MSE\": results_df['MSE'].idxmin(),\n", " \"RMSE\": results_df['RMSE'].idxmin(),\n", " \"MAE\": results_df['MAE'].idxmin()\n", "}\n", "\n", "# Highlighting the best values in the DataFrame\n", "def highlight_best(row, best_metrics):\n", " return ['font-weight: bold' if (col in best_metrics and row.name == best_metrics[col]) else '' for col in row.index]\n", "\n", "# Apply the highlighting\n", "styled_results = results_df.style.apply(highlight_best, best_metrics=best_metrics, axis=1)\n", "styled_results" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can now conclude that the zero'th row is worse than the other two with reasonable confidence, but the other two are still very close in performance. To get more certainty, we would need to either obtain more data or compare methods with a more significant difference in performance.\n", "\n", "**Question**: Can we *really* conclude that NN is less effective than k-NN (with k=100) with resonable confidence for the problem of predicting student GPAs?\n", "\n", "**Answer**: Not quite! This was for *one* model trained using *one* split into training and testing data. If we used different data to train the model, we might end up with better or worse models from each method. One method might be more sensitive to variations in the data than the other. We did *not* account for the variance of the learned model, we only accounted for the variance of the sample MSE on the test set!" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Algorithm Evaluation\n", "\n", "Notice that the discussion so far has focussed on using a test set to evaluate a single model that was trained from data. This captures our uncertainty about the performance of the model that was learned. If we run the algorithm many times on different training sets, we could obtain models of different quality. The true MSE of each model could differ! Our analysis so far did not capture this.\n", "\n", "The analysis above is useful for testing how much you can trust a specific model, but less useful for comparing algorithms in general. To compare algorithms, we can do the following:\n", "- Specify a number of trials, `num_trials`\n", "- For each trial $i$ in $1,...,\\text{num\\_trials}$ do:\n", " - Sample a data set (ideally independent of the data sets for other trials)\n", " - Split the data set into training and testing sets\n", " - Use the ML algorithm to train a model on the training set.\n", " - Use the trained model to make predictions for the testing set.\n", " - Compute the sample performance metric (e.g., sample MSE) for the test set. Call this $Z_i$.\n", "- Compute and report the average sample MSE.\n", "- Compute and report the standard error of $Z_1,\\dotsc,Z_\\text{num\\_trials}$.\n", "\n", "This standard error incorporates uncertainty due to both the sample MSE and the varying MSE of the learned models." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Cross-Validation\n", "\n", "Notice that we can't easily do this using the GPA data set, since we can't generate `num_trials` indepent data sets (unless we consider data sets much smaller than our actual data set).\n", "\n", "Cross-validation is a technique that resolves this, by repeatedly splitting the same data set into different training and testing sets. The most common version is $k$-fold cross-validation, which operates as follows.\n", "\n", "- **Input:** Dataset `D`, Number of folds `k`, Machine Learning Algorithm `ML_Algo`\n", "- **Output:** Cross-validated performance estimate\n", "\n", "Procedure:\n", "\n", "1. Split `D` into `k` equal-sized subsets (folds) `F1, F2, ..., Fk`.\n", "2. For `i` from 1 to `k`:\n", " - Set aside fold `Fi` as the validation set, and combine the remaining `k-1` folds to form a training set.\n", " - Train the model `M` using `ML_Algo` on the `k-1` training folds.\n", " - Evaluate the performance of model `M` on the validation fold `Fi`. Store the performance metric `P_i`.\n", "3. Calculate the average of the performance metrics: `Average_Performance = mean(P_1, P_2, ..., P_k)`.\n", "4. Optionally, calculate other statistics (like standard deviation or standard error) of the performance metrics across the folds.\n", "\n", "One notable variant of k-fold cross-validation is **leave-one-out (LOO) cross-validation**, which sets `k` equal to the size of the data set so that each fold is a single point.\n", "\n", "Scikit-Learn has a useful function [KFold](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.KFold.html), which simplifies creating folds.\n", "\n" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Average MSE: 0.571\n", "MSE Standard Error: ±0.004\n" ] } ], "source": [ "import pandas as pd\n", "from sklearn.model_selection import KFold, cross_val_score\n", "from sklearn.metrics import mean_squared_error\n", "import numpy as np\n", "\n", "# Load the data set\n", "df = pd.read_csv(\"data/GPA.csv\", delimiter=',')\n", "\n", "# We already loaded X and y, but do it again as a reminder\n", "X = df.iloc[:, :-1]\n", "y = df.iloc[:, -1]\n", "\n", "# Define the model\n", "model = WeightedKNearestNeighbors(k=300, sigma=100)\n", "\n", "# Choose number of folds for k-fold Cross-Validation\n", "k = 20\n", "kf = KFold(n_splits=k, shuffle=True, random_state=1)\n", "\n", "# Function to compute MSE for each fold\n", "def mse_for_fold(train_index, test_index, model, X, y):\n", " X_train, X_test = X.iloc[train_index], X.iloc[test_index]\n", " y_train, y_test = y.iloc[train_index], y.iloc[test_index]\n", " model.fit(X_train, y_train)\n", " predictions = model.predict(X_test)\n", " return mean_squared_error(y_test, predictions)\n", "\n", "# Compute MSE for each fold\n", "mse_scores = [mse_for_fold(train_index, test_index, model, X, y) for train_index, test_index in kf.split(X)]\n", "\n", "# Calculate the average MSE and standard error\n", "average_mse = np.mean(mse_scores)\n", "mse_standard_error = np.std(mse_scores, ddof=1) / np.sqrt(k)\n", "\n", "print(f\"Average MSE: {average_mse:.3f}\")\n", "print(f\"MSE Standard Error: ±{mse_standard_error:.3f}\")\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Compare this to what we had done before, where we were only training the model once, and then were estimating the accuracy of that one model." ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
 ModelMSERMSEMAE
0k-NN k=300 sigma=1000.572 ± 0.0130.756 ± 0.0060.592 ± 0.006
\n" ], "text/plain": [ "" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Load the data set\n", "df = pd.read_csv(\"data/GPA.csv\", delimiter=',')\n", "\n", "# We already loaded X and y, but do it again as a reminder\n", "X = df.iloc[:, :-1]\n", "y = df.iloc[:, -1]\n", "\n", "# Split the data into training and testing sets (60% train, 40% test)\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.5, shuffle=True)\n", "\n", "# Model parameters to test\n", "parameters = [\n", " {\"k\": 300, \"sigma\": 100} # Weighted k-NN\n", "]\n", "\n", "# Dictionary to store results\n", "results = []\n", "\n", "# Training and evaluating each model\n", "for param in parameters:\n", " if param[\"sigma\"] is None:\n", " model = KNearestNeighbors(k=param[\"k\"])\n", " else:\n", " model = WeightedKNearestNeighbors(k=param[\"k\"], sigma=param[\"sigma\"])\n", " model.fit(X_train, y_train)\n", " predictions = model.predict(X_test)\n", "\n", " mse, mse_se = mean_squared_error_with_se(predictions, y_test)\n", " rmse, rmse_se = root_mean_squared_error_with_se(predictions, y_test)\n", " mae, mae_se = mean_absolute_error_with_se(predictions, y_test)\n", "\n", " results.append({\n", " \"Model\": f\"k-NN k={param['k']} sigma={param['sigma']}\",\n", " \"MSE\": f\"{mse:.3f} ± {1.96 * mse_se:.3f}\",\n", " \"RMSE\": f\"{rmse:.3f} ± {1.96 * rmse_se:.3f}\",\n", " \"MAE\": f\"{mae:.3f} ± {1.96 * mae_se:.3f}\"\n", " })\n", "\n", "# Creating DataFrame for results\n", "results_df = pd.DataFrame(results)\n", "\n", "# Finding the best (minimum or maximum) values for each metric\n", "best_metrics = {\n", " \"MSE\": results_df['MSE'].idxmin(),\n", " \"RMSE\": results_df['RMSE'].idxmin(),\n", " \"MAE\": results_df['MAE'].idxmin()\n", "}\n", "\n", "# Highlighting the best values in the DataFrame\n", "def highlight_best(row, best_metrics):\n", " return ['font-weight: bold' if (col in best_metrics and row.name == best_metrics[col]) else '' for col in row.index]\n", "\n", "# Apply the highlighting\n", "styled_results = results_df.style.apply(highlight_best, best_metrics=best_metrics, axis=1)\n", "styled_results" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "It's not that one of these is *better* than the other. Rather, they are estimating fundamentally different quantities. The former estimates the MSE we should expect if we use our algorithm on a randomly selected training set from the GPA data set. The latter estimates the MSE that we should expect for one specific model that was trained using one specific train/test split." ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "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": "3.11.7" } }, "nbformat": 4, "nbformat_minor": 2 }