mlp.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Importing packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import metrics\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sb\n",
    "%store -r"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Fetching data and setting up datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(112800, 785)\n",
      "(18800, 785)\n",
      "(112800, 784)\n",
      "(112800,)\n",
      "(18800, 784)\n",
      "(18800,)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0    41\n",
       "1    39\n",
       "2     9\n",
       "3    26\n",
       "4    44\n",
       "Name: CHAR, dtype: int64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_train = pd.read_csv(\"./EMNIST/emnist-balanced-train.csv\", header=None)\n",
    "data_test = pd.read_csv(\"./EMNIST/emnist-balanced-test.csv\", header=None)\n",
    "\n",
    "cols = ['CHAR']\n",
    "for i in range(1, 785):\n",
    "    cols.append(str(i))\n",
    "data_train.columns = cols\n",
    "data_test.columns = cols\n",
    "print(data_train.shape)\n",
    "print(data_test.shape)\n",
    "\n",
    "X_train = data_train.iloc[:,1:]\n",
    "y_train = data_train[\"CHAR\"]\n",
    "X_test = data_test.iloc[:,1:]\n",
    "y_test = data_test[\"CHAR\"]\n",
    "print(X_train.shape)\n",
    "print(y_train.shape)\n",
    "print(X_test.shape)\n",
    "print(y_test.shape)\n",
    "y_test.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. Modelling\n",
    "\n",
    "In order to optimize some parameters for the model, we will extract a very small sample from the full testing dataset first and experiment with parameters on it before working with the full testing set."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3.1. Attempt 1\n",
    "\n",
    "1. Find the optimal number of neurons for one layer, in increment size of 100\n",
    "2. Find the optimal number of hidden layers, using the optimal number of neurons for one layer as the number of neurons for every layer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#num_neurons = [100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 1100, 1200]\n",
    "#num_layers = 1\n",
    "num_neurons = 900\n",
    "num_layers = [1,2,3,4,5]\n",
    "\n",
    "accuracies = []\n",
    "\n",
    "# for num in num_neurons:\n",
    "for num in num_layers:\n",
    "    print(\"Now starting: {} layers\".format(num))\n",
    "    # mlp = MLPClassifier(max_iter=200, hidden_layer_sizes=(num,), random_state=0)\n",
    "    mlp = MLPClassifier(max_iter=200, hidden_layer_sizes=([num_neurons]*num_layers), random_state=0)\n",
    "    X_train_small = X_train[:10000]\n",
    "    y_train_small = y_train[:10000]\n",
    "    mlp.fit(X_train_small, y_train_small)\n",
    "    y_pred_tr = mlp.predict(X_train_small)\n",
    "    y_pred_test = mlp.predict(X_test)\n",
    "    accuracies.append(metrics.accuracy_score(y_test, y_pred_test))\n",
    "\n",
    "# evaluation = pd.DataFrame({\n",
    "#     \"num_neurons\": num_neurons,\n",
    "#     \"accuracy\": accuracies\n",
    "# })\n",
    "# evaluation\n",
    "evaluation = pd.DataFrame({\n",
    "    \"num_layers\": num_layers,\n",
    "    \"accuracy\": accuracies\n",
    "})\n",
    "evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_neurons = [100, 200, 300, 400, 500]\n",
    "#num_layers = [1,2,3,4,5]\n",
    "\n",
    "accuracies = []\n",
    "\n",
    "for num in num_neurons:\n",
    "    print(\"Now starting: {} neurons\".format(num))\n",
    "    #mlp = MLPClassifier(max_iter=200, hidden_layer_sizes=(num,), random_state=0)\n",
    "    mlp = MLPClassifier(max_iter=200, hidden_layer_sizes=(900, num,), random_state=0)\n",
    "#mlp.fit(X_train, y_train)\n",
    "    X_train_small = X_train[:1000]\n",
    "    y_train_small = y_train[:1000]\n",
    "    mlp.fit(X_train_small, y_train_small)\n",
    "    y_pred_tr = mlp.predict(X_train_small)\n",
    "    y_pred_test = mlp.predict(X_test)\n",
    "    accuracies.append(metrics.accuracy_score(y_test, y_pred_test))\n",
    "\n",
    "# evaluation = pd.DataFrame({\n",
    "#     \"num_neurons\": num_neurons,\n",
    "#     \"accuracy\": accuracies\n",
    "# })\n",
    "# evaluation\n",
    "evaluation = pd.DataFrame({\n",
    "    \"num_neurons\": num_neurons,\n",
    "    \"accuracy\": accuracies\n",
    "})\n",
    "evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Ideal layer size analysis - Number of layers is kept constant as 1.\n",
    "#       num_neurons accuracy\n",
    "# 0\t    100\t        0.543564\n",
    "# 1\t    200\t        0.613085\n",
    "# 2\t    300\t        0.595851\n",
    "# 3\t    400\t        0.600904\n",
    "# 4\t    500\t        0.604681\n",
    "# 5\t    600\t        0.624894\n",
    "# 6\t    700\t        0.659894\n",
    "# 7\t    800\t        0.646330\n",
    "# 8\t    900\t        0.689734 <== Local maximum\n",
    "# 9\t    1000\t    0.656011\n",
    "# 10\t1100\t    0.672606\n",
    "# 11\t1200\t    0.658936\n",
    "\n",
    "# Ideal number of layers analysis - Number of neurons per layer is kept constant at 900 (as determined by the previous table)\n",
    "# \t    num_layers  accuracy\n",
    "# 0\t    1\t        0.689734\n",
    "# 1\t    2\t        0.652181\n",
    "# 2\t    3\t        0.700106 <== Local maximum\n",
    "# 3\t    4\t        0.677021\n",
    "# 4\t    5\t        0.710691 <== would take far too long to train\n",
    "\n",
    "# Therefore, a MLP with 3 hidden layers and 900 neurons per layer is ideal.\n",
    "# Of course, it's not actually ideal as varying layer sizes may lead to better results,\n",
    "# however, I do not have a significant amount of computational power or time to test all possibilities."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# PLEASE DO NOT OVERWRITE THE MODEL (run if you wish but I shall not do it again)\n",
    "mlp_trained = MLPClassifier(random_state=0, max_iter=2, hidden_layer_sizes=(900,900,900,), verbose=True)\n",
    "print(\"Model created\")\n",
    "mlp_trained.fit(X_train, y_train)\n",
    "# On a Ryzen 7 5700U, it took 78 minutes to train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'mlp_trained' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn [1], line 3\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;66;03m# mlp_trained is a saved variable. Layers: (900,900,900), max_iter: 100\u001b[39;00m\n\u001b[1;32m      2\u001b[0m \u001b[38;5;66;03m# mlp_test is the model generated above\u001b[39;00m\n\u001b[0;32m----> 3\u001b[0m to_test \u001b[38;5;241m=\u001b[39m \u001b[43mmlp_trained\u001b[49m\n\u001b[1;32m      4\u001b[0m \u001b[38;5;66;03m# On training dataset\u001b[39;00m\n\u001b[1;32m      5\u001b[0m y_pred_train \u001b[38;5;241m=\u001b[39m to_test\u001b[38;5;241m.\u001b[39mpredict(X_train)\n",
      "\u001b[0;31mNameError\u001b[0m: name 'mlp_trained' is not defined"
     ]
    }
   ],
   "source": [
    "# mlp_trained is a saved variable. Layers: (900,900,900), max_iter: 100\n",
    "# mlp_test is the model generated above\n",
    "to_test = mlp_trained\n",
    "# On training dataset\n",
    "y_pred_train = to_test.predict(X_train)\n",
    "y_pred_proba_train = to_test.predict_proba(X_train)\n",
    "accuracy_train = metrics.accuracy_score(y_train, y_pred_train)\n",
    "loss_train = metrics.log_loss(y_train, y_pred_proba_train)\n",
    "print(\"Training Set:\")\n",
    "print(\" - Loss: \" + str(loss_train))\n",
    "print(\" - Accuracy: \" + str(accuracy_train))\n",
    "# On testing dataset\n",
    "y_pred = to_test.predict(X_test)\n",
    "y_pred_proba_test = to_test.predict_proba(X_test)\n",
    "accuracy_test = metrics.accuracy_score(y_test, y_pred)\n",
    "loss_test = metrics.log_loss(y_test, y_pred_proba_test)\n",
    "print(\"Testing Set\")\n",
    "print(\" - Loss: \" + str(loss_test))\n",
    "print(\" - Accuracy: \" + str(accuracy_test))\n",
    "# Final result - Training - Accuracy: 0.9301152482269504\n",
    "# Final result - Training - Accuracy: 0.8181914893617022"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9999999999999999\n",
      "(41, 0.9999990017750039)\n"
     ]
    }
   ],
   "source": [
    "print(sum(y_pred_proba_test[0]))\n",
    "print(max(enumerate(y_pred_proba_test[0]), key=lambda x: x[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(68, 0.23139621742202607)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iters = enumerate(mlp_trained.loss_curve_)\n",
    "min(iters, key=lambda x: x[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_confusion_matrix(y_true, y_pred):\n",
    "    # visualize the confusion matrix\n",
    "    ax = plt.subplot()\n",
    "    c_mat = metrics.confusion_matrix(y_true, y_pred)\n",
    "    sb.heatmap(c_mat, annot=False, fmt='g', ax=ax)\n",
    "\n",
    "    ax.set_xlabel('Predicted labels', fontsize=15)\n",
    "    ax.set_ylabel('True labels', fontsize=15)\n",
    "    ax.set_title('Confusion Matrix', fontsize=15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "generate_confusion_matrix(y_test, y_pred_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAi0AAAHSCAYAAAAgxfsjAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy89olMNAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB+hElEQVR4nO3dd1iTV/8/8HcCEjYoCogIqCCIihMVtIiK4Kij2lqtA0ddBa3aOuhw1Fq09XHWYtuvsw6q1m0dONDWLYpirXsPUB8VECQCOb8//JGnKSEJECCB96vXfV3lPuc+9ychmo9n3RIhhAARERGRgZOWdQBEREREumDSQkREREaBSQsREREZBSYtREREZBSYtBAREZFRYNJCRERERoFJCxERERkFJi1ERERkFJi0EBERkVFg0kLlQkZGBubNm4d27drByckJZmZmqFy5MgICAjB16lTcvXu3zGJLSEhAaGgo7O3tIZFIIJFIcPv27RK/b3x8PCQSCQYPHlzi9yosDw8P5Xuxfv36AuudOnVKWU8ikZRihAWbPn06JBIJVq5cWdahEFU4TFrI6B07dgyenp745JNPcOrUKTRo0ADvvvsuAgMDcePGDcycORN169bF/v37Sz229PR0dO/eHfv370fTpk0xaNAghIeHw9rautRjMVRr164tsGzNmjV6vdfKlSshkUgwffp0vbZLRKXDtKwDICqOxMREdOjQAVlZWZg8eTK+/PJLWFlZKcsVCgW2bt2KSZMm4f79+6Ue3+nTp/Hw4UMMHDgQq1evLtV7t2jRAn///Tfs7OxK9b6F0aRJE+zduxdPnz5F1apVVcpycnLw66+/wtfXFzdu3IBcLi+jKFVFRkaib9++qF69elmHQlThsKeFjJYQAgMHDkRWVhamT5+O2bNnqyQsACCVStGrVy8kJCSgefPmpR5jXqJUu3btUr+3paUlfHx8DPrLtX///srk5N/27duHx48fY8CAAWUQWcGqVq0KHx8fg04GicotQWSkfv/9dwFAuLq6iuzs7EJfn5GRIb766itRv359YW5uLmxtbcVbb70l1q9fr7a+u7u7yPsj8/PPP4uGDRsKc3Nz4eTkJEaMGCGeP3+urHvr1i0BQO0RHh4uhBBi2rRpAoBYsWKF1vv909GjR0WPHj2Em5ubMDMzE05OTsLf319MnjxZpKenK+sdOnRI5X7/lJ2dLRYtWiSaNm0qrKyshJWVlfD39xc//PCDyMnJyVe/bdu2AoC4deuW2LJli2jZsqWwtLQUlStXFn379hX37t0r4F1WL++1Xb16VdjY2IhWrVrlq9OvXz8hkUjEnTt3hEwmU/te7Ny5UwwZMkT4+PgIGxsbYWlpKfz8/MSsWbNEVlaW2teg7sj7HfzzPXv06JEYNmyYqFGjhjAxMRHz588XQqj/ve3atUsAELVr1xZpaWkq91UoFCI0NFQAEN98802h3iciUsXhITJau3btAgC89957MDUt3Ec5PT0d7dq1Q0JCAqpVq4a3334bGRkZOHjwIP744w8cP34cCxcuVHvtpEmTsHDhQgQHB8PT0xNHjx7FTz/9hL///huHDx+GRCKBtbU1wsPDcf36dRw9ehSNGjVC48aNAQBt2rQp8mvesWMHevbsCSEEWrRogcDAQLx48QLXrl3DnDlzMGrUKK3zZXJzc9GjRw/8/vvvsLW1RceOHSGEwMGDB/HRRx8hLi4OmzZtglSavyP2hx9+wLx58/DWW2+hS5cuOHnyJGJjY5GQkIDz58/DwsKiUK/HwsICvXr1wqpVq3Djxg3UqVMHAPDy5Uts27YNbdq0gZubW4HXDxs2DK9evUKDBg3g5+eH1NRUnDp1Cp9//jkOHDiAffv2wcTEBADQqVMn5OTk5Pt9AICnp6dKu0+ePIG/vz9ycnLQpk0bZGVlwdLSssA4unTpgoiICCxZsgRjx47FihUrlGULFy7Evn37EBQUhMmTJxfq/SGifynrrImoqFq3bi0AiF9++aXQ10ZGRgoAol27dir/Mv7777+Fo6OjACB27Nihck1e74Czs7O4fPmy8vyTJ0+Ep6enACAOHDigcs2KFSsEADFt2rR8MRSlpyUoKEgAEJs2bcpX/9SpUyqvpaCelrlz5woAon79+iI5OVl5/uHDh8Lb21sAEIsXL1a5Jq+XwtLSUhw7dkx5PiMjQwQGBgoAYtmyZWpfh6bXdu/ePREXFycAiBkzZijLV69eLQCIH3/8UQghCuxp2bp1q8jMzFQ5l5aWJt5++20BQKxatUqlTNPvQ4j/vWcAxDvvvCNevXqVr05Bv7fMzExRr149AUBs3LhRCCHEhQsXhEwmE3Z2duL27dta3xci0oxzWsho/fe//wUAVKtWrVDXZWRkYNmyZZBKpfjhhx9gY2OjLPPx8cEXX3wBAAX2tMycORPe3t7Kn6tWrYpRo0YBAI4cOVKoWArryZMnAICQkJB8Zf7+/iqvpSCLFi0CAMybNw9OTk7K89WrV8d3330HoODXPn78eAQEBCh/trS0xIQJEwAU/bW3b98e1atXV1lFtGbNGshkMrz33nsar+3Ro0e+3h0bGxvMnz8fALBt27YixSSTybB48WKYm5vrfI2FhQXWrl0LMzMzjBw5Ejdu3ED//v0hl8uxZMkSuLu7FykWIvofJi1U4SQkJODVq1do2rQpfHx88pUPHDgQAHD06FEoFIp85aGhofnO1a1bFwDw6NEjPUerqlmzZgDexHj69Gm18Wly9+5d3L17F9WqVVP7Ot5++23Y29vj+vXrSE5OzldeEq9dKpWiX79+uHr1Kk6fPo3k5GQcOHAAXbp0QeXKlbVef+3aNSxcuBBjxozB0KFDMXjwYMycOVNZVhRNmzZFjRo1Cn1dkyZN8PXXX+PZs2do2rQpkpKS0K9fP/Tv379IcRCRKs5pIaPl4OAA4H+9D7p6+PAhgDcbnKljb28POzs7pKam4vnz58r75HF1dc13TV4PR0kvy/3mm2+QlJSEHTt2YMeOHahcuTLatGmD7t27Y8CAAVp7BvJee0H/6pdIJHB3d8eLFy/w4MEDODs7q5SX1GsfMGAA5s2bh7Vr18Ld3R25ublaVw0JIfDpp59i/vz5EEKorZOenl6keDTNo9Hmk08+wa+//oqEhARUr14dP/zwQ5HbIiJV7Gkho5U3kfLs2bN6b1vT7qvqJqiWBHW9KDVr1sSZM2ewd+9ejBkzBjVr1sSOHTswfPhw+Pn5KYfMiqMsXnuTJk1Qr149xMbGYvXq1bC3t0fXrl01XvPrr79i3rx5cHV1xaZNm/DgwQO8fv0aQghlAlVQMqNNYYaF/u3SpUu4ePEigDdDmKWx+zFRRcGkhYxW3pfaxo0bkZOTo/N1Li4uAIA7d+6oLU9NTcWLFy9gYWGh0/BEUZmZmQF4s1Lm33Jzc9UOzwCAqakpQkNDsWjRIpw/fx63b99G+/btlSuINNH22v9ZVpThkeIYMGAAUlJSkJiYiPfeew8ymUxj/S1btgAAYmJi0Lt3b7i4uKBSpUoAgJs3b5Z4vOrI5XLlPJYBAwbg9evX6N+/P7KyssokHqLyhkkLGa1OnTqhfv36uH//PmbNmqWxblpaGv766y8Ab+aFWFhYICEhQe2ch7yt41u3bl2ivSp5m75dvXo1X9mhQ4eQnZ2tUzvu7u7KpbR5/8IviJubG9zc3PDkyRMcOHAgX/muXbvw/PlzeHp65hsaKmkffPABqlatCgcHBwwaNEhr/efPnwNQP2S1YcMGtdfkJYqFSXILIyoqChcuXEDfvn3xyy+/YNCgQbh06RImTpxYIvcjqmiYtJDRkkgkWLNmDczNzTF9+nRERUUhIyNDpY4QAtu3b0fz5s1x+vRpAICVlRWGDh0KhUKBiIgIlWuuXr2Kr7/+GgAwduzYEo0/KCgIwJsk6Z9DCLdu3Srw3vPnz1fbA/P7778DeDN8pM2YMWMAABMmTFCZD5ScnKz8cv344491exF65OHhgSdPnuDp06c67WWTNwH4p59+UhkG+uOPP5SroP4tr6fpypUreohY1YEDB7BgwQLUrFkTMTExAIDFixfDw8MD33//Pfbs2aP3exJVNJyIS0atcePG2L9/P3r37o3Zs2dj0aJFCAgIgJOTE1JTU3HmzBmkpKTA3Nxc5Qs9OjoaJ06cQFxcHGrXro22bdsqN5fLysrC2LFj0a1btxKNvU6dOhg0aBBWr16Nxo0bIygoCJmZmThx4gS6dOmCzMzMfMM4M2bMwKeffopGjRrBy8sLQgicP38eV69eRZUqVfDpp59qve/48eNx8OBB7N69G15eXmjfvj2EEDhw4ADS09PRs2dPfPTRRyX1svVm7NixWLlyJX744QfEx8fDz88PDx48wJ9//olPPvkEc+fOzXdNq1at4OjoiE2bNiE4OBi1a9eGVCrF0KFDERgYWORYnj9/jvDwcEgkEuWcHACwtbXFL7/8grZt22LIkCFISkrK94wlItIde1rI6LVu3RrXr1/H3Llz4e/vjwsXLmDDhg04evQoPDw8MG3aNFy7dg0dOnRQXmNjY4PDhw9jxowZqFq1KrZv344//vgDzZs3x7p16wrcp0Tffv75Z0yZMgW2trbYu3cvbt++jaioKKxfv15t/cWLF6Nv377IzMzE7t27sWfPHpiammLChAm4cOECvLy8tN7TxMQE27dvx8KFC1G7dm3s3bsX+/btg7e3N5YsWVLgbriGpm7dujhz5gy6deuGp0+fYvv27Xj58iV+/PHHAntazM3NsWvXLnTs2BGJiYlYuXIlli1bpnaIrjBGjRqFBw8e4JNPPkFwcLBKWZs2bTB58mQkJyfjww8/LNZ9iCo6iSjq9HoiIiKiUmT4/5wiIiIiApMWIiIiMhJMWoiIiMgoMGkhIiIio8CkhYiIiIyCUSctS5YsgYeHB8zNzdGyZUucOnWqrEMiIiKiEmK0S55//fVXDBo0CEuXLkXLli2xYMECbNy4EVeuXIGjo6NObaQND9VYXmXVX/oIlfRMquGBfgCgMM6PtEGb4tJWY/nsh4dLKRKispPz+kGJ3yP7qX6em1Wpam29tGNojLanZd68eRg+fDiGDBkCX19fLF26FJaWlli+fHlZh0ZEREQlwCiTltevXyMhIQEhISHKc1KpFCEhITh+/HgZRkZERFQMilz9HOWUUT576OnTp8jNzYWTk5PKeScnJ1y+fFntNXK5HHK5XPVcrgIyE6PM24iIqDwSirKOwKBVmG/s6Oho2NnZqRzzEm+VdVhERESkI6NMWqpWrQoTExOkpKSonE9JSYGzs7Paa6KiopCamqpyTGhcqzTCJSIi0o1CoZ+jnDLKpMXMzAzNmjXDgQMHlOcUCgUOHDiAgIAAtdfIZDLY2tqqHBwaIiIiQyKEQi9HeWWUc1oAYMKECQgPD0fz5s3RokULLFiwABkZGRgyZEhZh0ZEREQlwGiTlvfffx9PnjzB1KlTkZycjMaNG2PPnj35Judqom0flmdDGmhvY8VFne9H+mEs+7B0dPLTWB6XcqHY92jsoH0vhsT/Fn/fB+7DQlRKyvHQjj4YbdICAJGRkYiMjCzrMIiIiPSjHA/t6INRJy1ERETlSjneY0UfOBOViIiIjAJ7WoiIiAwFh4c0YtJCRERkKDgRVyMODxEREZFRMNqkJTo6Gv7+/rCxsYGjoyN69uyJK1eulHVYRERERcbN5TSTCGEkm178S6dOndC3b1/4+/sjJycHn332GS5evIhLly7ByspKpzZMzWoUO46Mc6s1lls1GVTsexBRxeNgYaOx/L+v0kspkuKR6FDHWL6Ecl4/KPF7yK8d00s7Mq9AvbRjaIx2TsuePXtUfl65ciUcHR2RkJCAoKCgMoqKiIiISorRJi3/lpqaCgCoUqVKGUdCRERUROV4aEcfykXSolAoMG7cOLRu3RoNGqjfel8ul0Mul6ucE0JAItGl85KIiKgUcHM5jYx2Iu4/RURE4OLFi4iNjS2wTnR0NOzs7FQOoTCOMWEiIiIqB0lLZGQkdu7ciUOHDsHV1bXAelFRUUhNTVU5JFLNE92IiIhKlVDo5yinjHZ4SAiBMWPGYMuWLYiPj0etWrU01pfJZJDJZCrnODREREQGhZvLaWS0SUtERATWrVuHbdu2wcbGBsnJyQAAOzs7WFhYlHF0RERERVCOe0n0wWj3aSmol2TFihUYPHiwTm3oY58WbV7dj9dax8I1uMTjICKi4imVfVouxumlHVmDjnppx9AYbU+LkeZaREREBePwkEZGm7QQERGVN0JwybMmRr96iIiIiCoG9rQQEREZCk7E1YhJCxERkaHgnBaNODxERERERqFcJC2zZ8+GRCLBuHHjyjoUIiKiouOOuBoZ/fDQ6dOn8eOPP8LPz6+sQ1FLlz1YMi9v0Vhu6fOOnqIhqth02QO7NDZTkOqwG7einGzrUJFeq17wgYkaGXVPy8uXL9G/f3/8/PPPqFy5clmHQ0RERCXIqJOWiIgIdO3aFSEhIWUdChERUfFxeEgjox0eio2NxdmzZ3H69Gmd6svlcsjlcpVzQgg+NJGIiAwHVw9pZJQ9Lffu3cPHH3+MtWvXwtzcXKdroqOjYWdnp3IIRXoJR0pERFQI7GnRyCgfmLh161a88847MDExUZ7Lzc2FRCKBVCqFXC5XKQPU97RUdvAxiJ4WTsQlKh2ciFv6ytNrLY0HJmYdX6+XdswD+umlHUNjlMNDHTp0QFJSksq5IUOGwMfHB5MnT86XsACATCaDTCZTOWcICQsREZESh4c0MsqkxcbGBg0aNFA5Z2VlBQcHh3zniYiIjAaTFo2MMmkpb7QN/2Te3qe9DY9QfYVDpcRQhiqMhT6GGQzl/TSW4RB9qEivlUpeuUla4uPjyzoEIiKiYhGCm8tpUm6SFiIiIqPH4SGNjHLJMxEREVU87GkhIiIyFOV4jxV9YNJCRERkKDg8pJHRDg89ePAAAwYMgIODAywsLNCwYUOcOXOmrMMiIiKiEmKUPS3Pnz9H69at0a5dO+zevRvVqlXDtWvX+KRnIiIybhwe0sgok5Y5c+agZs2aWLFihfJcrVq1yjCikqXLHiwZf23UWG5V/z19hVMhlMYeKty9onC430fhyEwraa0jz8ku8TjK0zb+pYLDQxoZ5fDQ9u3b0bx5c7z33ntwdHREkyZN8PPPP5d1WERERMXDByZqZJRJy82bNxETEwMvLy/s3bsXo0ePxtixY7Fq1aoCr5HL5UhLS1M5jPBZkURERBWWUQ4PKRQKNG/eHN988w0AoEmTJrh48SKWLl2K8PBwtddER0djxowZKuckUmtITGxLPF4iIiKdcHhII6PsaalevTp8fX1VztWrVw93794t8JqoqCikpqaqHBKpTUmHSkREpDuFQj9HOWWUPS2tW7fGlStXVM5dvXoV7u7uBV4jk8kgk8lUzkl0mCBGREREhsEok5bx48cjMDAQ33zzDfr06YNTp07hp59+wk8//VTWoRERERVdOZ5Eqw8SYaSzUXfu3ImoqChcu3YNtWrVwoQJEzB8+PBCtWFqVqOEojM8rx7+obWOhctbpRAJEZFxynn9oMTv8Wr7XL20Y9H9U720Y2iMsqcFAN5++228/fbbZR0GERERlRKjnIhLRERULpXBPi3R0dHw9/eHjY0NHB0d0bNnz3zzRrOyshAREQEHBwdYW1ujd+/eSElJUalz9+5ddO3aFZaWlnB0dMTEiRORk5OjUic+Ph5NmzaFTCaDp6cnVq5cWahYmbQQEREZijJYPXT48GFERETgxIkTiIuLQ3Z2NkJDQ5GRkaGsM378eOzYsQMbN27E4cOH8fDhQ/Tq1UtZnpubi65du+L169c4duwYVq1ahZUrV2Lq1KnKOrdu3ULXrl3Rrl07JCYmYty4cfjwww+xd+9enWM12jkt+sA5Lao4p4WIqGClMqdly2y9tGPxzpQiX/vkyRM4Ojri8OHDCAoKQmpqKqpVq4Z169bh3XffBQBcvnwZ9erVw/Hjx9GqVSvs3r0bb7/9Nh4+fAgnJycAwNKlSzF58mQ8efIEZmZmmDx5Mnbt2oWLFy8q79W3b1+8ePECe/bs0Sk29rQQEREZCj0ND6nbBV4ul+sUQmpqKgCgSpUqAICEhARkZ2cjJCREWcfHxwdubm44fvw4AOD48eNo2LChMmEBgLCwMKSlpeGvv/5S1vlnG3l18trQBZMWIiIiQ6Gn4aHo6GjY2dmpHNHR0TrcXoFx48ahdevWaNCgAQAgOTkZZmZmsLe3V6nr5OSE5ORkZZ1/Jix55XllmuqkpaXh1atXOr09Rpm05Obm4ssvv0StWrVgYWGBOnXqYObMmXyWEBERGTc9JS3qdoGPiorSevuIiAhcvHgRsbGxpfBiC88olzzPmTMHMTExWLVqFerXr48zZ85gyJAhsLOzw9ixY8s6PIOky3yVzBu/ayy3rNOl2HFoe0w9H1FPVL5o+zMP8M99SVC3C7w2kZGR2LlzJ44cOQJXV1fleWdnZ7x+/RovXrxQ6W1JSUmBs7Ozss6pU6dU2stbXfTPOv9ecZSSkgJbW1tYWFjoFKNR9rQcO3YMPXr0QNeuXeHh4YF3330XoaGh+d4wIiIioyKEfo5C3VIgMjISW7ZswcGDB1GrVi2V8mbNmqFSpUo4cOCA8tyVK1dw9+5dBAQEAAACAgKQlJSEx48fK+vExcXB1tZW+azAgIAAlTby6uS1oQujTFoCAwNx4MABXL16FQBw/vx5/Pnnn+jcuXMZR0ZERFQMZbDkOSIiAmvWrMG6detgY2OD5ORkJCcnK+eZ2NnZYdiwYZgwYQIOHTqEhIQEDBkyBAEBAWjVqhUAIDQ0FL6+vhg4cCDOnz+PvXv34osvvkBERISyx2fUqFG4efMmJk2ahMuXL+OHH37Ahg0bMH78eJ1jNcrhoSlTpiAtLQ0+Pj4wMTFBbm4uZs2ahf79+xd4jVwuzzdzWgjBhyYSEVGFFhMTAwAIDg5WOb9ixQoMHjwYADB//nxIpVL07t0bcrkcYWFh+OGHH5R1TUxMsHPnTowePRoBAQGwsrJCeHg4vvrqK2WdWrVqYdeuXRg/fjwWLlwIV1dX/N///R/CwsJ0jtUo92mJjY3FxIkT8d1336F+/frKTWrmzZuH8PBwtddMnz4dM2bMUDknkVpDamJbGiEbBc5pISJ9K09zWkpln5a1X+qlHYv+M/XSjqExyqSlZs2amDJlCiIiIpTnvv76a6xZswaXL19We426npbKDj7safkHJi1EpG9MWgrn1ZrP9dKOxYBZemnH0Bjl8FBmZiakUtXpOCYmJlBoGMdTN5OaCQsREZHxMMqkpVu3bpg1axbc3NxQv359nDt3DvPmzcPQoUPLOjQiIqKiK+Qk2orGKJOWxYsX48svv8RHH32Ex48fw8XFBSNHjlR5MBMVnrbhn/Td0zSW23SeobEcMJ5uYCLSjxrWDlrr3Et/Wuz7WJmZayzPeJ1V7HuUCv4dqZFRJi02NjZYsGABFixYUNahEBERUSkxyqSFiIioXOLwkEZMWoiIiAwFkxaNmLQQEREZCsGkRROj3MafiIiIKh72tBARERkIoeDqIU0MsqflyJEj6NatG1xcXCCRSLB161aVciEEpk6diurVq8PCwgIhISG4du1a2QRLRESkL2XwwERjYpA9LRkZGWjUqBGGDh2KXr165Sv/9ttvsWjRIqxatQq1atXCl19+ibCwMFy6dAnm5prX6lPRaduHJfPaDq1tWHp101c4RBWaiVTzvzlzDeSL68HL/5bKfYxmHxYqFoNMWjp37ozOnTurLRNCYMGCBfjiiy/Qo0cPAMDq1avh5OSErVu3om/fvqUZKhERkf5wIq5GBjk8pMmtW7eQnJyMkJAQ5Tk7Ozu0bNkSx48fL8PIiIiIikkh9HOUUwbZ06JJcnIyAMDJyUnlvJOTk7JMHXVPeRZC8KGJRERERsLoelqKKjo6GnZ2diqHUKSXdVhERET/w4m4Ghld0uLs7AwASElJUTmfkpKiLFMnKioKqampKodEalOisRIRERUKkxaNjC5pqVWrFpydnXHgwAHlubS0NJw8eRIBAQEFXieTyWBra6tycGiIiIjIeBjknJaXL1/i+vXryp9v3bqFxMREVKlSBW5ubhg3bhy+/vpreHl5KZc8u7i4oGfPnmUXNBERUXGJ8juJVh8MMmk5c+YM2rVrp/x5woQJAIDw8HCsXLkSkyZNQkZGBkaMGIEXL16gTZs22LNnj0Hu0aJLX46hfES1xaotTl32YMm8vU9zGx6hWtsgIsPZh0UbBb+EC8dIfq9lRSJExf1EmZrVKPF7VKSkRRdMWojIWOW8flDi98ic+6Fe2rH89P/00o6hMbo5LURERFQxGeTwEBERUYXEHXE1YtJCRERkKMrxbrb6wOEhIiIiMgrsaSEiIjIQgquHNDLInpYjR46gW7ducHFxgUQiwdatW5Vl2dnZmDx5Mho2bAgrKyu4uLhg0KBBePjwYdkFTEREpA98YKJGBtnTkpGRgUaNGmHo0KHo1auXSllmZibOnj2LL7/8Eo0aNcLz58/x8ccfo3v37jhz5kwZRVwwY/rolEas2pY0a1sSrUsbxrTMnMiQSXXYNZz7sFBpMsikpXPnzujcubPaMjs7O8TFxamc+/7779GiRQvcvXsXbm5upREiERGR/nH1kEYGmbQUVmpqKiQSCezt7cs6FCIioqIrx0M7+mD0SUtWVhYmT56Mfv36wdbWtsB6crkccrlc5ZwQgg9NJCIiMhIGORFXV9nZ2ejTpw+EEIiJidFYNzo6GnZ2diqHUKSXUqREREQ6UCj0c5RTRpu05CUsd+7cQVxcnMZeFgCIiopCamqqyiGR2pRStERERDrg6iGNjHJ4KC9huXbtGg4dOgQHBwet18hkMshkMpVzHBoiIiKDwom4Ghlk0vLy5Utcv35d+fOtW7eQmJiIKlWqoHr16nj33Xdx9uxZ7Ny5E7m5uUhOTgYAVKlSBWZmZmUVNhEREZUgiRCGt8g+Pj4e7dq1y3c+PDwc06dPR61atdRed+jQIQQHB+t8n0pmNTSWG9wbQwCAzGs7NJZbenUrpUgqDhOp5pHk3HI8hk7FU572Tcp5/aDE75Hx+Xt6acdq1ka9tGNoDLKnJTg4GJpyKQPMs4iIiIqN2/hrZrQTcYmIiKhiMcieFiIiogqpHK/80QcmLURERIaCSYtGHB4iIiIio2CQScuRI0fQrVs3uLi4QCKRYOvWrQXWHTVqFCQSCRYsWFBq8REREZUIodDPUU4Z5PBQRkYGGjVqhKFDh6JXr14F1tuyZQtOnDgBFxeXIt2HnXDGSduS5lf3Dmptw6Jme32FUyFwSTMVFf+eLSQOD2lkkElL586d0blzZ411Hjx4gDFjxmDv3r3o2rVrKUVGREREZcUgkxZtFAoFBg4ciIkTJ6J+/fplHQ4REZFeCPa0aGSUScucOXNgamqKsWPH6nyNXC6HXC5XOSeE4POHiIjIcDBp0cggJ+JqkpCQgIULF2LlypWFSjiio6NhZ2encghFeglGSkREVEgKhX6OcsrokpY//vgDjx8/hpubG0xNTWFqaoo7d+7gk08+gYeHR4HXRUVFITU1VeWQSG1KL3AiIiIqFqMbHho4cCBCQkJUzoWFhWHgwIEYMmRIgdfJZDLIZDKVcxwaIiIig8LhIY0MMml5+fIlrl+/rvz51q1bSExMRJUqVeDm5gYHBweV+pUqVYKzszO8vb1LO1QiIiL9YdKikUEmLWfOnEG7du2UP0+YMAEAEB4ejpUrV5ZRVGQsdNmD5eXpnzWWW/sP11c4ROWas3VljeXJL5+XUiRUERhk0hIcHAwhdM82b9++XXLBEBERlZLCfPdVRAaZtBAREVVIHB7SyOhWDxEREVHFxJ4WIiIiQ8GeFo2YtBARERkIbuOvGYeHiIiIyCgYZNJy5MgRdOvWDS4uLpBIJNi6dWu+On///Te6d+8OOzs7WFlZwd/fH3fv3i39YImIiPRFIfRzlFMGOTyUkZGBRo0aYejQoejVq1e+8hs3bqBNmzYYNmwYZsyYAVtbW/z1118wNzcvg2gNg7a9fcvvR7hotO3D8urOfq1tWLiHaK1DVN5p24dFl33H+ffTP5TfxwbphUEmLZ07d0bnzp0LLP/888/RpUsXfPvtt8pzderUKY3QiIiISgzntGhmkMNDmigUCuzatQt169ZFWFgYHB0d0bJlS7VDSERERFR+GF3S8vjxY7x8+RKzZ89Gp06dsG/fPrzzzjvo1asXDh8+XOB1crkcaWlpKgd3HiQiIoPCOS0aGeTwkCYKxZsBvx49emD8+PEAgMaNG+PYsWNYunQp2rZtq/a66OhozJgxQ+WcRGoNiYltyQZMRESkK85p0UjvPS179uzB3Llz8euvvyI7O1vfzaNq1aowNTWFr6+vyvl69eppXD0UFRWF1NRUlUMitdF7fERERFQyitTT8sMPP2Du3Ln45Zdf0Lp1a+X5Pn364LffflP+7O/vj/j4eL2u6jEzM4O/vz+uXLmicv7q1atwd3cv8DqZTAaZTKZyTiLRZV47ERFR6eBEXM2KlLRs2bIFmZmZCAgIUJ7bs2cPNm3aBFdXVwwcOBAHDx7EqVOn8PPPP2PMmDGFav/ly5e4fv268udbt24hMTERVapUgZubGyZOnIj3338fQUFBaNeuHfbs2YMdO3YgPj6+KC+HiIjIMHB4SKMiDQ9duXIFDRo0gFT6v8tjY2MhkUiwadMmzJo1C4cOHULVqlWxZs2aQrd/5swZNGnSBE2aNAEATJgwAU2aNMHUqVMBAO+88w6WLl2Kb7/9Fg0bNsT//d//4bfffkObNm2K8nLKBaHlKA1SiUTrYSws3EO0Hpm392k8iIiMgbYNXQcPHgyJRKJydOrUSaXOs2fP0L9/f9ja2sLe3h7Dhg3Dy5cvVepcuHABb731FszNzVGzZk2VbUt0VaSelidPniAoKEjl3OHDh1GzZk20aNECAGBubo7AwEAcPXq00O0HBwdrXdkzdOhQDB06tNBtExERGaqyGB7StqErAHTq1AkrVqxQ/vzv6Rb9+/fHo0ePEBcXh+zsbAwZMgQjRozAunXrAABpaWkIDQ1FSEgIli5diqSkJAwdOhT29vYYMWKEzrEWKWmxs7PD06dPlT/funULd+7cwaBBg1TqWVlZISMjoyi3ICIiqnjKYHhI24auwJskxdnZWW3Z33//jT179uD06dNo3rw5AGDx4sXo0qUL5s6dCxcXF6xduxavX7/G8uXLYWZmhvr16yMxMRHz5s0rVNJSpOEhT09PHDlyRLla56efflLbXXT//v0CXyQRERGVDHV7k8nl8iK3Fx8fD0dHR3h7e2P06NH473//qyw7fvw47O3tlQkLAISEhEAqleLkyZPKOkFBQTAzM1PWCQsLw5UrV/D8ueZHQfxTkZKW0aNHIysrC35+fmjWrBm+/fZbVKtWDW+//bayzqtXr3DmzJl8S5OJiIhIPaHQzxEdHQ07OzuVIzo6ukgxderUCatXr8aBAwcwZ84cHD58GJ07d0Zubi4AIDk5GY6OjirXmJqaokqVKkhOTlbWcXJyUqmT93NeHV0UaXiof//+SExMxPfff49z587B1dUVq1atgrW1tbLOhg0bkJmZiQ4dOhTlFkRERBWPnoaHoqKiMGHCBJVz/56Hoqu+ffsq/79hw4bw8/NDnTp1EB8fX+rf8UXeEfe7777D119/jbS0NFSrVi1fefv27XHu3Dk+yJCIiEhHQk9Ji7q9yfSldu3aqFq1Kq5fv44OHTrA2dkZjx8/VqmTk5ODZ8+eKaeIODs7IyUlRaVO3s+FmUZSrB1xZTKZ2oQFAGrWrIlGjRqp9L7oStvyq5cvXyIyMhKurq6wsLCAr68vli5dWpSXQERERIVw//59/Pe//0X16tUBAAEBAXjx4gUSEhKUdQ4ePAiFQoGWLVsq6xw5ckRlp/y4uDh4e3ujcuXKOt/bIJ89pG351YQJE3Dw4EGsWbMGHh4e2LdvHz766CO4uLige/fueoujaVVPrXXOPr2utU5FoahgD6C09AjVWJ72XTeN5XYTd2i9h7Z31Nla+x/25Je6T3IjojJWBquHNG3oWqVKFcyYMQO9e/eGs7Mzbty4gUmTJsHT0xNhYWEA3jxGp1OnThg+fDiWLl2K7OxsREZGom/fvnBxcQEAfPDBB5gxYwaGDRuGyZMn4+LFi1i4cCHmz59fqFh1SlqKsx+KRCLBsmXLCnWNtuVXx44dQ3h4OIKDgwEAI0aMwI8//ohTp07pNWkhIiIqTfoaHiqMM2fOoF27dsqf8+bChIeHIyYmBhcuXMCqVavw4sULuLi4IDQ0FDNnzlQZflq7di0iIyPRoUMHSKVS9O7dG4sWLVKW29nZYd++fYiIiECzZs1QtWpVTJ06tVDLnQEdk5aVK1cWqtF/KkrSok1gYCC2b9+OoUOHwsXFBfHx8bh69WqhMzYiIqKKTtuGrnv37tXaRpUqVZQbyRXEz88Pf/zxR6Hj+yedkpZDhw4V6yb6tnjxYowYMQKurq4wNTWFVCrFzz//nG+X3n+Sy+X51qgLIfjQRCIiMhhl0dNiTHRKWtq2bVvScRTK4sWLceLECWzfvh3u7u44cuQIIiIi4OLigpCQELXXREdHY8aMGSrnJFJrSExsSyNkIiIirZi0aGaQE3E1efXqFT777DNs2bIFXbt2BfCmyykxMRFz584tMGlRt2a9soNPicdLRERE+lGspCUnJwe7du3CqVOn8PTpU7Rs2VI5affhw4d4+vQpfH19YWqqv9woOzsb2dnZKk+YBgATExMoFAWnqOrWrHNoiIiIDIrg95ImRc4m/vzzTwwYMAD37t1Tzg3Jzs5WJi3Hjx9Hnz59sHHjxgKfGlkQTcuv3Nzc0LZtW0ycOBEWFhZwd3fH4cOHsXr1asybN6+oL4eIiKjMcXhIM4nQNGW4AJcuXUKLFi2QnZ2N0aNHo02bNujTpw8GDx6M5cuXAwBev34NBwcHdO/eHWvXri1U+/Hx8SrLr/KEh4dj5cqVSE5ORlRUFPbt24dnz57B3d0dI0aMwPjx4wvVe2JqVqNQcREVhlTLZ/Hl1e1a27D00rzXC5Gh0+VvZGPZ4Snn9YMSv0dyULBe2nE+Eq+XdgxNkXpaZs6ciaysLPz+++8IDVW/wZaZmRmaNm2Kc+fOFbp9bcuvnJ2dsWLFikK3S0REZMiEgsNDmhRpG/9Dhw6hRYsWBSYseWrUqIGHDx8WKTAiIqKKRl9PeS6vitTT8uLFC9SsWVNrvYyMDJXnDBAREVHBBCfialSknhZHR0eVibIF+fvvv3VKboiIiIi0KVLS0r59eyQmJmrcKXfLli24fv06OnbsWOTgiIiIKhIOD2lWpKRlypQpMDMzQ8+ePRETE4Pk5GRl2fPnz7F8+XIMGzYMVlZW+TZ0IyIiIvWEQqKXo7wqUtLi4+OD9evXQ6FQIDIyEjVq1IBEIsGqVatQtWpVDB8+HHK5HGvXrkWtWrUK3X50dDT8/f1hY2MDR0dH9OzZE1euXFGpk5WVhYiICDg4OMDa2hq9e/dGSkpKUV4OERERGYEi7dOS586dO5g/fz7i4uJw+/ZtKBQKuLq6omPHjvjkk09Qp06dIrXbqVMn9O3bF/7+/sjJycFnn32Gixcv4tKlS7CysgIAjB49Grt27cLKlSthZ2eHyMhISKVSHD16VOf7cJ8WMnSZN37XWG5Zp0spRVI+lKc9Q8oTbb8XQ/mdlMY+LXebd9BLO25nDuilHUNTrKSltDx58gSOjo44fPgwgoKCkJqaimrVqmHdunV49913AQCXL19GvXr1cPz4cbRq1Uqndpm0kKFj0qJfTFoME5OW/7nTVP3z8wrL/ex+vbRjaIo0PFTaUlNTAQBVqlQBACQkJCA7O1vl4Yg+Pj5wc3PD8ePHyyRGIiIiKlnFepKhXC7Hb7/9hj/++EO5iZyLiwvatGmD3r17w9zcvNgBKhQKjBs3Dq1bt0aDBg0AAMnJyTAzM4O9vb1KXScnJ5VJwf+OVS6Xq5zLe2YSERGRISjPk2j1ochJy/79+zF48GA8evQo35b7P/30EyZNmoSVK1cWe8lzREQELl68iD///LNY7URHR2PGjBkq5yRSa0hMbIvVLhERkb4Y/oSNslWkpOXkyZN4++238fr1a7Rs2RL9+vWDh4cHgDeTc9evX48TJ06gW7duOHz4MFq2bFmk4CIjI7Fz504cOXIErq6uyvPOzs54/fo1Xrx4odLbkpKSAmdnZ7VtRUVF5Vt+XdnBp0hxERERUekrUtLy5ZdfIjs7GzExMRg5cmS+8jFjxuCnn37CqFGjMHXqVOzdu7dQ7QshMGbMGGzZsgXx8fH5lk03a9YMlSpVwoEDB9C7d28AwJUrV3D37l0EBASobVMmk0Emk6mc49AQEREZEg4PaVak1UN2dnbw8fHByZMnNdZr2bIlLl++rJxIq6uPPvoI69atw7Zt2+Dt7a1yXwsLCwBvljz//vvvWLlyJWxtbTFmzBgAwLFjx3S+D1cPkaHj6iH94uohw8TVQ/9zo0GYXtqpc7FwnQXGokg9LVKpFJ6enlrreXp65tsUThcxMTEAgODgYJXzK1aswODBgwEA8+fPh1QqRe/evSGXyxEWFoYffvih0PfSpGf1ZlrrbH2UoNd7UvnR3qmhxvJDKUla29CWlNxtXldrG25nrmqtU1EYypdfRcJEsXDK8xb8+lCkpKVFixa4cOGC1noXLlxAixYtCt2+Lp0/5ubmWLJkCZYsWVLo9omIiMj4FGmflpkzZ+LatWuYNm0aFIr8aaEQAtOmTcO1a9cwc+bMYgdJRERUESiERC9HeaVTT8vq1avznQsPD8fXX3+NX375Bb1794a7uzuAN6uHNm/ejNu3b2P48OG4cuVKkVcPERERVSSiHCcc+qDTRFypVKp2pc0/L80r/3dzEokEubm5xY2zRGibiMs5LVQc+pjTou0PJ+e0kKErT3NaSmMi7hWfznppx/vybr20Y2h06mmZOnUqlwcTERGVMC551kynpGX69OklHIaq6OhobN68GZcvX4aFhQUCAwMxZ84c5fLnZ8+eYdq0adi3bx/u3r2LatWqoWfPnpg5cybs7OxKNVYiIiJ94Y64mhXr2UMl5fDhw4iIiIC/vz9ycnLw2WefITQ0FJcuXYKVlRUePnyIhw8fYu7cufD19cWdO3cwatQoPHz4EJs2bdJbHBz6oeI4qMPwT3HpMvTz6n68xnIL12D9BEOkBr+DSZ+KtLlcaXvy5AkcHR1x+PBhBAUFqa2zceNGDBgwABkZGTA11S0X4+ZyVBEwaSHSj9KY03KpTle9tON7Y5de2jE0xepp+fPPP7Ft2zZcu3YN6enpavdXkUgkOHDgQHFuo9xRt0qVKhrr2Nra6pywEBERGZryvFxZH4r0DS+EwLBhw7Bq1SploiKRSPKtJhJCFHsCr0KhwLhx49C6dWs0aNBAbZ2nT59i5syZGDFiRIHtyOVyyOXyfK+DE4yJiIiMQ5E2l1u6dClWrlyJZs2aIS4uDr169QLw5qGFu3fvxuDBgyGVSjFx4kTcvHmzWAFGRETg4sWLiI2NVVuelpaGrl27wtfXV+OE4ejoaNjZ2akcQpFerNiIiIj0SQiJXo7yqkhzWlq2bIlLly7h9u3bcHBwwJAhQ7B69WqV/Vg2bdqE999/H5s3b0aPHj2KFFxkZCS2bduGI0eO5HvSMwCkp6cjLCwMlpaW2LlzJ8zNzQtsS11PS2UHH/a0ULnHOS1E+lEac1oueHTTSzt+t3fopR1DU6Selr///huBgYFwcHAA8L+N5f6ZtLz77rto1qwZ5s6dW+j2hRCIjIzEli1bcPDgQbUJS1paGkJDQ2FmZobt27drTFgAQCaTwdbWVuVgwkJERIaE2/hrVqSkRaFQKBMWALC0tAQAPH/+XKWel5cXkpIKv+wzIiICa9aswbp162BjY4Pk5GQkJyfj1atXAP6XsGRkZGDZsmVIS0tT1jHU3XeJiIioeIo0EbdGjRp4+PCh8ue85w6dO3cOHTt2VJ6/evVqkVbzxMTEAACCg4NVzq9YsQKDBw/G2bNncfLkSQCAp6enSp1bt27Bw8Oj0PckKq+0Df9k3tyjtQ3L2p30FE3ZKk9byhsLvueFU57no+hDkXpamjZtikuXLil7NUJDQyGEwKRJk3D58mWkp6fju+++Q0JCApo0aVLo9oUQao/BgwcDeJPMFFSHCQsRERkrIfRzlFdFSlq6d++Op0+fYteuN5vXNGrUCH379sX58+dRv3592NvbY8qUKTA1NcWsWbP0GjARERFVTEUaHurXrx969eqlMvSzatUq+Pn5YevWrXj+/Dnq1q2LSZMmoUWLFnoLloiIqDwrz5No9cEotvEvKdzGn4hzWv6twv6FWELK03teGkueT9d4Ry/t+D/Yopd2DE2RhoeIiIiIShsf1ENERGQgODykmU5JS+3atYt8A4lEghs3bhTqmujoaGzevBmXL1+GhYUFAgMDMWfOHHh7e+erK4RAly5dsGfPHmzZsgU9e/YscqxERERlyViGysqKTknL7du3SzgMVYcPH0ZERAT8/f2Rk5ODzz77DKGhobh06RKsrKxU6i5YsKDIO9u2d2qosfxgSuE3xiMyNrrMV0n/v0Eay20+XK2vcEpUefpC0Pa3nqG8VkOJg8oHnZIWhUJR0nGo2LNHdWLgypUr4ejoiISEBAQFBSnPJyYm4j//+Q/OnDmD6tWrl2qMRERE+sbhIc2MYk5LamoqAKBKlSrKc5mZmfjggw+wZMkSODs7l1VoREREesMdcTUz+KRFoVBg3LhxaN26NRo0aKA8P378eAQGBur8BGl1T3lWCAWkEi6gIiIiw1C64xrGx+C/sSMiInDx4kXExsYqz23fvh0HDx7EggULdG4nOjoadnZ2KsettJslEDERERGVBINOWiIjI7Fz504cOnQIrq6uyvMHDx7EjRs3YG9vD1NTU+XOvL179873kMU8UVFRSE1NVTlq2RZ9VRQREZG+CUj0cpRXBjk8JITAmDFjsGXLFsTHx6NWrVoq5VOmTMGHH36ocq5hw4aYP38+unXrprZNmUwGmUymco5DQ0REZEgUXG6lkUEmLREREVi3bh22bdsGGxsbJCcnAwDs7OxgYWEBZ2dntZNv3dzc8iU4REREVD4YZNISExMDAPmGelasWIHBgwfr7T7ch4VIN9r2YXl1Z7/WNizcQ/QVDsFw9j+RatknS1FxH29XJIpyPLSjDwaZtBTlGY4V+LmPRERUTpTn+Sj6oJek5dq1a3j69CkcHBxQt25dfTRJREREpKLIM1Hlcjk+++wzVK1aFT4+PmjTpg1mz56tLF+zZg2aNm2KxMREfcRJRERU7in0dJRXRUpaXr16heDgYMyZMwdmZmbo0qVLvuGZ9u3b4/z589iwYYNeAiUiIirvuORZsyIlLd9++y1OnjyJoUOH4ubNm9ixY0e+Oi4uLvD19cX+/don6BERERF7WrQpUtLy66+/ws3NDTExMTA3Ny+wnre3N+7du1fk4IiIiIjyFClpuXXrFpo3b67cibYgZmZmeP78eaHbj46Ohr+/P2xsbODo6IiePXviypUr+eodP34c7du3h5WVFWxtbREUFIRXr14V+n5ERESGgD0tmhVp9ZCFhYVOycitW7dQuXLlQrd/+PBhREREwN/fHzk5Ofjss88QGhqKS5cuwcrKCsCbhKVTp06IiorC4sWLYWpqivPnz0Mq5S63VHza9p4AuP/EP+myB0vmtfzDyP9k6aV+N2tSz1D2R+GfA/0qz/NR9KFISUvjxo1x5swZPHnyBNWqVVNb59atWzh37hxCQ0ML3f6ePXtUfl65ciUcHR2RkJCAoKAgAG+e8jx27FhMmTJFWc/b27vQ9yIiIiLjUKRuieHDhyM9PR39+vXD06dP85W/ePECQ4cORXZ2NkaMGFHsIFNTUwEAVapUAQA8fvwYJ0+ehKOjIwIDA+Hk5IS2bdvizz//LPa9iIiIyopCop+jvCpST0u/fv2wY8cOxMbGonbt2ggMDAQAHD16FD169MDhw4eRlpaGQYMG4e233y5WgAqFAuPGjUPr1q3RoEEDAMDNmzcBANOnT8fcuXPRuHFjrF69Gh06dMDFixfh5eWVrx25XA65XK5yTggBiQ7DAERERKWB2/hrVuQJIGvXrsWcOXNgbm6Offv2AXizM+6OHTsgkUgwa9YsrFixotgBRkRE4OLFi4iNjVWeUyjeTDMaOXIkhgwZgiZNmmD+/Pnw9vbG8uXL1bYTHR0NOzs7lUMo0osdHxEREZWOIm/jL5FIMHHiREyYMAFnz57F7du3oVAo4OrqCn9/f5iZmRU7uMjISOzcuRNHjhyBq6ur8nz16tUBAL6+vir169Wrh7t376ptKyoqChMmTFA5V9nBp9gxEhER6QunNWtW7GcPmZiYwN/fH/7+/vqIB8CbYZsxY8Zgy5YtiI+PR61atVTKPTw84OLikm8Z9NWrV9G5c2e1bcpkMshkMpVzHBoiIiJDUp6XK+uDQa4PjoiIwJo1a7Bu3TrY2NggOTkZycnJyj1Y8np5Fi1ahE2bNuH69ev48ssvcfnyZQwbNqyMoyciIjIeR44cQbdu3eDi4gKJRIKtW7eqlAshMHXqVFSvXh0WFhYICQnBtWvXVOo8e/YM/fv3h62tLezt7TFs2DC8fPlSpc6FCxfw1ltvwdzcHDVr1sS3335b6FiL1NMydOhQnetKJBIsW7asUO3HxMQAAIKDg1XOr1ixAoMHDwYAjBs3DllZWRg/fjyePXuGRo0aIS4uDnXq1CnUvUh/dOm3MpauT+49oX/a9mHJ+Ps3jeVW9XrrMxzSgYkO+17lKoyjb0Db30+G8ideUQYjABkZGWjUqBGGDh2KXr165Sv/9ttvsWjRIqxatQq1atXCl19+ibCwMFy6dEm5K37//v3x6NEjxMXFITs7G0OGDMGIESOwbt06AEBaWhpCQ0MREhKCpUuXIikpCUOHDoW9vX2hVhlLxL+fdKgDXTZwk0gkytU5ubm5hb1FqTA1q1HWIZQr5SlpodLHpKVwSmNzOSYtqnJeP9BHKBptrN5fL+10v70834pZddMk/k0ikWDLli3o2bMngDe9LC4uLvjkk0/w6aefAnizDYmTkxNWrlyJvn374u+//4avry9Onz6N5s2bA3iz31qXLl1w//59uLi4ICYmBp9//jmSk5OVc16nTJmCrVu34vLlyzq/riINDx06dEjtceDAAaxcuRIffPABJBIJJkyYgIMHDxblFkRERBWOvrbxV7diNjo6utDx3Lp1C8nJyQgJ+d+u13Z2dmjZsiWOHz8O4M0O9fb29sqEBQBCQkIglUpx8uRJZZ2goCCVRTphYWG4cuVKoR73U6ThobZt22osHzRoELp27Yrw8HB07969KLcgIiKiIlK3YlZbL4s6ycnJAAAnJyeV805OTsqy5ORkODo6qpSbmpqiSpUqKnX+vagmr83k5GSdH/lTYhNx+/Xrh/r162P69OkldQsiIqJyRV874spkMtja2qocRUlaDE2Jrh7y8vLCmTNnSvIWRERE5YYCEr0c+uLs7AwASElJUTmfkpKiLHN2dsbjx49VynNycvDs2TOVOura+Oc9dFFiSYtCocCFCxeK9NTl6Oho+Pv7w8bGBo6OjujZs2e+PVmSk5MxcOBAODs7w8rKCk2bNsVvv2meyEdERES6q1WrFpydnXHgwAHlubS0NJw8eRIBAQEAgICAALx48QIJCQnKOgcPHoRCoUDLli2VdY4cOYLs7Gxlnbi4OHh7e+s8NASUQNKSmZmJxMRE9OvXD9euXdM6/0Wdw4cPIyIiAidOnFAunwoNDUVGRoayzqBBg3DlyhVs374dSUlJ6NWrF/r06YNz587p8+UQERGVGqGnozBevnyJxMREJCYmAngz+TYxMRF3796FRCLBuHHj8PXXXyu/bwcNGgQXFxflCqN69eqhU6dOGD58OE6dOoWjR48iMjISffv2hYuLCwDggw8+gJmZGYYNG4a//voLv/76KxYuXJhv3o02RVrybGJiorWOEALVqlXDH3/8gbp16xb2FiqePHkCR0dHHD58GEFBQQAAa2trxMTEYODAgcp6Dg4OmDNnDj788EOd2jWWJc9WZuZa62S8ziqFSIjKTsZfG7XWsar/XilEQoVRnrZCKI0lz6trDNBLO4MerNG5bnx8PNq1a5fvfHh4OFauXAkhBKZNm4affvoJL168QJs2bfDDDz+ofLc/e/YMkZGR2LFjB6RSKXr37o1FixbB2tpaWefChQuIiIjA6dOnUbVqVYwZMwaTJ08u1OsqUtLi4eFR4Bb4ZmZmqF69Otq2bYuIiIh8M4qL4vr16/Dy8kJSUpLySc+hoaEwMzPD6tWrYW9vjw0bNmDYsGE4f/48PD09dWqXSQuR8WDSYpyYtBROWSQtxqRIS55v376t5zAKplAoMG7cOLRu3VqZsADAhg0b8P7778PBwQGmpqawtLTEli1bCkxY5HJ5vo128ja/IyIiMgTGsVVf2SnSnJbt27dj9+7d+o5FrYiICFy8eBGxsbEq57/88ku8ePEC+/fvx5kzZzBhwgT06dMHSUlJattRt9GOUKSXxksgIiLSSVnMaTEmRZ7TEhoaWuKJS2RkJLZt24YjR46obEpz48YNeHp64uLFi6hfv77yfEhICDw9PbF06dJ8banraans4GMUPS0cHiLi8JCx4vBQ4azQ0/DQEA4P/U+1atUKtUSpsIQQGDNmDLZs2YL4+Ph8u+hlZmYCyP8MJBMTEygKeA6GumcuGEPCQkREFYeCX0saFSlpCQ4OxqlTp0psTkhERATWrVuHbdu2wcbGRrkNsJ2dHSwsLODj4wNPT0+MHDkSc+fOhYODA7Zu3Yq4uDjs3LlT7/EQERGVBs5p0axIw0PXrl2Dv78/Bg8ejNmzZysfTa23oApIhFasWIHBgwcrY5gyZQr+/PNPvHz5Ep6envj0009VlkBrYyyrh6h8Kk/d5obi1f14jeUWrsGlEgeVT6UxPPSjq36Gh0be5/CQ0vr169GlSxcsXrwYsbGxCAkJgZubm9rkRSKR4MsvvyxU+7rkUV5eXtwBl4iIqALRqaeldu3aeO+99zBnzhwAb+aSSCQSnZILiUSC3Nzc4kdaAtjTQmWJPS36x54WKkml0dOytKZ+elpG3avAPS23b9/GkydPlD+vWLGixAIiIiKqqDinRbMiDQ+Fh4frOw4iIiIijYqUtBAREZH+sadFMyYtREREBoLz2DTTOWlJTEzEV199VaSbTJ06tVD1Y2JiEBMTo3zGUf369TF16lR07twZAJCVlYVPPvkEsbGxkMvlCAsLww8//AAnJ6cixUdERESGT6fVQ3mrhQorb/O5wq4e2rFjB0xMTODl5QUhBFatWoXvvvsO586dQ/369TF69Gjs2rULK1euhJ2dHSIjIyGVSnH06NFC3Yerh4gqllcP/9Bax8LlrVKIhIxRaaweWuimn9VDH9+twKuHAKBOnTpo3bp1Scai1K1bN5WfZ82ahZiYGJw4cQKurq5YtmwZ1q1bh/bt2wN4s5qpXr16OHHiBFq1alUqMRIREekb57RopnPS0qZNGyxfvrwkY1ErNzcXGzduREZGBgICApCQkIDs7GyEhIQo6/j4+MDNzQ3Hjx9n0kJERFROGexE3KSkJAQEBCArKwvW1tbYsmULfH19kZiYCDMzM9jb26vUd3JyUj6jSB11T3kuqWcnERERFQV7WjSTaq9SNry9vZGYmIiTJ09i9OjRCA8Px6VLl4rcXnR0NOzs7FQOoUjXY8RERETFI/R0lFcGm7SYmZnB09MTzZo1Q3R0NBo1aoSFCxfC2dkZr1+/xosXL1Tqp6SkwNnZucD2oqKikJqaqnJIpDYl/CqIiIh0p5Do5yivDDZp+TeFQgG5XI5mzZqhUqVKOHDggLLsypUruHv3LgICAgq8XiaTwdbWVuXg0BAREZHx0GlOi0JRuqNsUVFR6Ny5M9zc3JCeno5169YhPj4ee/fuhZ2dHYYNG4YJEyagSpUqsLW1xZgxYxAQEMBJuEREZNQ4p0Uzg5yI+/jxYwwaNAiPHj2CnZ0d/Pz8sHfvXnTs2BEAMH/+fEilUvTu3VtlczkiIk3s3dprrZNxbrXGcqsmg4odh6nURGudHEXh9rcyVFIderQV2rcLqzD4Tmim0+Zy5RU3lyOqWGSmlbTWeXZ6mcZyJi2FU56SltLYXC7aXT+by0XdqeCbyxEREVHJUrCvRSMmLURERAaCc1o0M5rVQ0RERFSxsaeFiIjIQHBwSDMmLURERAaCw0OaGeTwUExMDPz8/JSbwAUEBGD37t0AgGfPnmHMmDHw9vaGhYUF3NzcMHbsWKSmppZx1ERERFSSDLKnxdXVFbNnz4aXlxeEEFi1ahV69OiBc+fOQQiBhw8fYu7cufD19cWdO3cwatQoPHz4EJs2bSrr0InIgMlzsrXW0bak+dX9eK1tWLgGaywvL8uZdVGBd9UokvK8Bb8+GM0+LVWqVMF3332HYcOG5SvbuHEjBgwYgIyMDJia6p6HcZ8WIiosfSQtFYku38FG8SWE0tmn5QuPD/TSzte31+mlHUNjkD0t/5Sbm4uNGzciIyOjwGcLpaamwtbWtlAJCxERkaExlgSurBjst3xSUhICAgKQlZUFa2trbNmyBb6+vvnqPX36FDNnzsSIESM0tieXyyGXy1XOCSH40EQiIiIjYZATcQHA29sbiYmJOHnyJEaPHo3w8HBcunRJpU5aWhq6du0KX19fTJ8+XWN70dHRsLOzUzmEIr0EXwEREVHhKPR0lFdGM6clJCQEderUwY8//ggASE9PR1hYGCwtLbFz506Ym5trvF5dT0tlBx/2tBBRoXBOS+FwTkvhTPbop5d25txer5d2DI3BDg/9m0KhUCYdaWlpCAsLg0wmw/bt27UmLAAgk8kgk8lUzjFhISIiMh4GmbRERUWhc+fOcHNzQ3p6OtatW4f4+Hjs3bsXaWlpCA0NRWZmJtasWYO0tDSkpaUBAKpVqwYTE+1PTyUiIjJExtLrVFYMMml5/PgxBg0ahEePHsHOzg5+fn7Yu3cvOnbsiPj4eJw8eRIA4OnpqXLdrVu34OHhUQYRE1FFocvQj7YhpIo0fMQv4cIpz/NR9MEgk5Zly5YVWBYcHMzNioiIiCogg0xaiIiIKiIF+6Y0YtJCRERkIJiyaGaw+7QQERER/RN7WoiIiAwEJ+JqxqSFiIjIQAgOEGlkkMNDMTEx8PPzg62tLWxtbREQEIDdu3fnqyeEQOfOnSGRSLB169bSD5SIiEiPuI2/ZgbZ0+Lq6orZs2fDy8sLQgisWrUKPXr0wLlz51C/fn1lvQULFnBXWyLSWWltKa9tH5bMm3u0tmFZu5MeIil7Uh3+jlZwGwvSkUEmLd26dVP5edasWYiJicGJEyeUSUtiYiL+85//4MyZM6hevXpZhElERKRXXPKsmUEmLf+Um5uLjRs3IiMjAwEBAQCAzMxMfPDBB1iyZAmcnZ3LOEIiIiL9YMqimcEmLUlJSQgICEBWVhasra2xZcsW+Pr6AgDGjx+PwMBA9OjRQ+f21D3lWQjB4SUiIiIjYbBJi7e3NxITE5GamopNmzYhPDwchw8fxvXr13Hw4EGcO3euUO1FR0djxowZKuckUmtITGz1GTYREVGRcXhIM4kwkgf5hISEoE6dOrCwsMCiRYsglf5v4VNubi6kUineeustxMfHq71eXU9LZQcf9rQQVSClNRFXG07EVWUsE3FzXj8o8XsM93hPL+38fHujXtoxNAbb0/JvCoUCcrkcM2bMwIcffqhS1rBhQ8yfPz/fBN5/kslkkMlkKueYsBARERkPg0xaoqKi0LlzZ7i5uSE9PR3r1q1DfHw89u7dC2dnZ7WTb93c3FCrVq0yiJaIiEg/uLmcZgaZtDx+/BiDBg3Co0ePYGdnBz8/P+zduxcdO3Ys69Dy0dZXw48fkeEwlD+Pugz9vLp3UGO5Rc32+gqnRBnJDASDUZ43htMHg0xali1bVqj6/ENBRERU/hlk0kJERFQRcXhIMyYtREREBoLDQ5oxaSEiIjIQxrL8u6wY5FOeiYiIiP7NIJOWmJgY+Pn5wdbWFra2tggICMDu3btV6hw/fhzt27eHlZUVbG1tERQUhFevXpVRxERERMUn9HSUVwY5POTq6orZs2fDy8sLQgisWrUKPXr0wLlz51C/fn0cP34cnTp1QlRUFBYvXgxTU1OcP39eZZfc0lKePxyFZSi7jRKVB9qWNGfe+F1juWWdLvoMp8h02cSTK0D/h9v4a2Y02/hXqVIF3333HYYNG4ZWrVqhY8eOmDlzZrHaNDWroafoCGDSQlSajCVp4Tb+hfOB+zt6aWfdnS16acfQGOTw0D/l5uYiNjYWGRkZCAgIwOPHj3Hy5Ek4OjoiMDAQTk5OaNu2Lf7888+yDpWIiKhYhJ7+K68McngIAJKSkhAQEICsrCxYW1tjy5Yt8PX1xYkTJwAA06dPx9y5c9G4cWOsXr0aHTp0wMWLF+Hl5aW2PXUPTBRC8PlDRERkMLjkWTOD7Wnx9vZGYmIiTp48idGjRyM8PByXLl2CQvHmVzpy5EgMGTIETZo0wfz58+Ht7Y3ly5cX2F50dDTs7OxUDqFIL62XQ0RERMVksEmLmZkZPD090axZM0RHR6NRo0ZYuHAhqlevDgDw9fVVqV+vXj3cvXu3wPaioqKQmpqqckikNiX6GoiIiApDAaGXo7wy2KTl3xQKBeRyOTw8PODi4oIrV66olF+9ehXu7u4FXi+TyZRLqPMODg0REZEhKYs5LdOnT4dEIlE5fHx8lOVZWVmIiIiAg4MDrK2t0bt3b6SkpKi0cffuXXTt2hWWlpZwdHTExIkTkZOTo5f35J8Mck5LVFQUOnfuDDc3N6Snp2PdunWIj4/H3r17IZFIMHHiREybNg2NGjVC48aNsWrVKly+fBmbNm0q69CJiIiMTv369bF//37lz6am/0sPxo8fj127dmHjxo2ws7NDZGQkevXqhaNHjwJ4s2Cma9eucHZ2xrFjx/Do0SMMGjQIlSpVwjfffKPXOA0yaXn8+DEGDRqER48ewc7ODn5+fti7dy86duwIABg3bhyysrIwfvx4PHv2DI0aNUJcXBzq1KlTxpFXbLrk9tr6tspvpyaRfmlb0px5bYf2Nry66SucAhnJrhoGo6wm4pqamsLZ2Tnf+dTUVCxbtgzr1q1D+/Zv9g5asWIF6tWrhxMnTqBVq1bYt28fLl26hP3798PJyQmNGzfGzJkzMXnyZEyfPh1mZmb6i1NvLenRsmXLtNaZMmUKpkyZUgrREBERlQ59JXnqVszKZDLIZDK19a9duwYXFxeYm5sjICAA0dHRcHNzQ0JCArKzsxESEqKs6+PjAzc3Nxw/fhytWrXC8ePH0bBhQzg5OSnrhIWFYfTo0fjrr7/QpEkTvbwmwIjmtBAREZV3+pqIq27FbHR0tNp7tmzZEitXrsSePXsQExODW7du4a233kJ6ejqSk5NhZmYGe3t7lWucnJyQnJwMAEhOTlZJWPLK88r0ySB7WoiIiKjooqKiMGHCBJVzBfWydO7cWfn/fn5+aNmyJdzd3bFhwwZYWFiUaJyFxZ4WIiIiA6HQ06FuxWxBScu/2dvbo27durh+/TqcnZ3x+vVrvHjxQqVOSkqKcg6Ms7NzvtVEeT+rmydTHExaiIiIDIQhbOP/8uVL3LhxA9WrV0ezZs1QqVIlHDhwQFl+5coV3L17FwEBAQCAgIAAJCUl4fHjx8o6cXFxsLW1zbenWnFxeIiIiKgC+/TTT9GtWze4u7vj4cOHmDZtGkxMTNCvXz/Y2dlh2LBhmDBhAqpUqQJbW1uMGTMGAQEBaNWqFQAgNDQUvr6+GDhwIL799lskJyfjiy++QEREhM69O7oyyJ6WmJgY+Pn5Kbu0AgICsHv3bmV5cnIyBg4cCGdnZ1hZWaFp06b47bffyjBiIiKi4iuLHXHv37+Pfv36wdvbG3369IGDgwNOnDiBatWqAQDmz5+Pt99+G71790ZQUBCcnZ2xefNm5fUmJibYuXMnTExMEBAQgAEDBmDQoEH46quv9PreAIBEGOAi+h07dsDExAReXl4QQmDVqlX47rvvcO7cOdSvXx+hoaF48eIFvv/+e1StWhXr1q3DtGnTcObMmUItrTI1q1GCr4KIyLBl3vhdY7m2vWB0UclEe4d+dq7+d04tCTmvH5T4PTrX7Ky9kg5239utvZIRMsielm7duqFLly7w8vJC3bp1MWvWLFhbWyuf8Hzs2DGMGTMGLVq0QO3atfHFF1/A3t4eCQkJZRw5ERERlRSDTFr+KTc3F7GxscjIyFBO+gkMDMSvv/6KZ8+eQaFQIDY2FllZWQgODi7bYImIiIpBX6uHyiuDnYiblJSEgIAAZGVlwdraGlu2bFHOQt6wYQPef/99ODg4wNTUFJaWltiyZQs8PT0LbE/d7oBCCD40kYiIDEZxV/6Udwbb0+Lt7Y3ExEScPHkSo0ePRnh4OC5dugQA+PLLL/HixQvs378fZ86cwYQJE9CnTx8kJSUV2J663QGFIr20Xg4REREVk0FOxFUnJCQEderUwaRJk+Dp6YmLFy+ifv36KuWenp5YunSp2uvV9bRUdvBhTwsRVViciFs4pTERN6RmmF7a2X9vr17aMTQGOzz0bwqFAnK5HJmZmQAAqVS1k8jExAQKRcEjeeoeFMWEhYiIDImR9COUGYNMWqKiotC5c2e4ubkhPT0d69atQ3x8PPbu3QsfHx94enpi5MiRmDt3LhwcHLB161bExcVh586dZR06ERFRkRV2j5WKxiCTlsePH2PQoEF49OgR7Ozs4Ofnh71796Jjx44AgN9//x1TpkxBt27d8PLlS3h6emLVqlXo0qX4XZlk/Eyk2qdq5WrolSOqKLQN/2Re26G1Deu63TWWG8vQDxkHg0xali1bprHcy8uLO+ASEVG5w9VDmhlk0kJERFQRKTinRSODXfJMRERE9E/saSEiIjIQ7GfRjEkLERGRgeDqIc04PERERERGweCTltmzZ0MikWDcuHHKc1lZWYiIiICDgwOsra3Ru3dvpKSklF2QREREeqCA0MtRXhn08NDp06fx448/ws/PT+X8+PHjsWvXLmzcuBF2dnaIjIxEr169cPTo0TKKlAyJLnuwuNk6aiy/m/ZYX+EQGS1Lr25a67w8Mk9juU3QBK1tlN+v2MLjjriaGWxPy8uXL9G/f3/8/PPPqFy5svJ8amoqli1bhnnz5qF9+/Zo1qwZVqxYgWPHjuHEiRNlGDERERGVJINNWiIiItC1a1eEhISonE9ISEB2drbKeR8fH7i5ueH48eOlHSYREZHecHhIM4McHoqNjcXZs2dx+vTpfGXJyckwMzODvb29ynknJyckJycX2Ka6pzwLIfjQRCIiMhjcEVczg+tpuXfvHj7++GOsXbsW5ubmems3OjoadnZ2KodQpOutfSIiouISQujlKK8MLmlJSEjA48eP0bRpU5iamsLU1BSHDx/GokWLYGpqCicnJ7x+/RovXrxQuS4lJQXOzs4FthsVFYXU1FSVQyK1KeFXQ0RERPpicMNDHTp0QFJSksq5IUOGwMfHB5MnT0bNmjVRqVIlHDhwAL179wYAXLlyBXfv3kVAQECB7cpkMshkMpVzHBoiIiJDUp7no+iDwSUtNjY2aNCggco5KysrODg4KM8PGzYMEyZMQJUqVWBra4sxY8YgICAArVq1KouQiYiI9KI8D+3og8ElLbqYP38+pFIpevfuDblcjrCwMPzwww9lHZZauvTlGMpH1ESqebTQXmalsfy/r4xnjpC2fVikOvTC8WmsVJa0fUJL69NprWUflvQdUVrbsOkWXew4LCvJNJZnZss1lpNxMIqkJT4+XuVnc3NzLFmyBEuWLCmbgIiIiEoAh4c0M4qkhYiIqCLgkmfNDG71EBEREZE67GkhIiIyEJwrpxmTFiIiIgPB4SHNODxERERERsHgk5bZs2dDIpFg3LhxAIBnz55hzJgx8Pb2hoWFBdzc3DB27FikpqaWbaBERETFpBBCL0d5ZdDDQ6dPn8aPP/4IPz8/5bmHDx/i4cOHmDt3Lnx9fXHnzh2MGjUKDx8+xKZNm8owWvWM6aOTq1BoLDemfViKy1D+0BvKXhxkeIzld6/LHiyZN37XWG5Vp4v2NsrJPiwcHtLMYJOWly9fon///vj555/x9ddfK883aNAAv/32m/LnOnXqYNasWRgwYABycnJgamqwL4mIiEgjQ/kHk6Ey2OGhiIgIdO3aFSEhIVrrpqamwtbWlgkLERFROWaQ3/KxsbE4e/YsTp8+rbXu06dPMXPmTIwYMUJjPblcDrlctftQCMGHJhIRkcHg8JBmBtfTcu/ePXz88cdYu3YtzM3NNdZNS0tD165d4evri+nTp2usGx0dDTs7O5VDKCrOHA0iIjJ8nIirmUQY2CMlt27dinfeeQcmJibKc7m5uZBIJJBKpZDL5TAxMUF6ejrCwsJgaWmJnTt3ak1w1PW0VHbwYU8LGTROxKWKQB8TcUvjz0LO6wclfo86VZvqpZ0bT8/qpR1DY3DDQx06dEBSUpLKuSFDhsDHxweTJ0+GiYkJ0tLSEBYWBplMhu3bt2tNWABAJpNBJlN9CigTFiIiMiQcHtLM4JIWGxsbNGjQQOWclZUVHBwc0KBBA6SlpSE0NBSZmZlYs2YN0tLSkJaWBgCoVq2aSg8NERGRMRFC89YTFZ3BJS3anD17FidPngQAeHp6qpTdunULHh4eZRAVUckojX9zVTLR/tdAdm5OKURC5ZGJVPvUSUstwz8Zf/+msRwArOr11jkmMl5GkbTEx8cr/z84OBgGNg2HiIhILxQcHtLIKJIWIiKiioD/KNfM4JY8ExEREanDnhYiIiIDweEhzZi0EBERGQgOD2lm8MNDs2fPhkQiwbhx4/KVCSHQuXNnSCQSbN26tdRjIyIi0ifuiKuZQfe0nD59Gj/++CP8/PzUli9YsIAbxOmRuamZxnJ3G0eN5Vee39dnOFRKSms5s6lU8x5KOYrcUomjvDCW3ZLdbZy01rmZ+khjuS7LmVOnvKWx3G72H1rbIMNnsD0tL1++RP/+/fHzzz+jcuXK+coTExPxn//8B8uXLy+D6IiIiPRP6Om/8spgk5aIiAh07doVISEh+coyMzPxwQcfYMmSJXB2di6D6IiIiPRPCKGXo7wyyOGh2NhYnD17FqdPn1ZbPn78eAQGBqJHjx46t6nugYlCCA4vERERGQmDS1ru3buHjz/+GHFxcWofhLh9+3YcPHgQ586dK1S70dHRmDFjhso5idQaEhPbYsVLRESkL1zyrJnBDQ8lJCTg8ePHaNq0KUxNTWFqaorDhw9j0aJFMDU1RVxcHG7cuAF7e3tlOQD07t0bwcHBBbYbFRWF1NRUlUMitSmlV0VERKQdh4c0M7ielg4dOiApKUnl3JAhQ+Dj44PJkyejatWqGDlypEp5w4YNMX/+fHTr1q3AdmUyGWQymco5Dg0REREZD4NLWmxsbNCgQQOVc1ZWVnBwcFCeVzf51s3NDbVq1SqVGImIiEpCed5jRR8MLmmhspOV81pjOfdhoeLQtg+Lsew7YiiM5f24k55S7DakOvSKa9uHJfPqNo3llnV1X9hRksrz0I4+GEXSEh8fr7Gcv2QiIqLyzyiSFiIiooqAq4c0Y9JCRERkIDhyoBmTFiIiIgPBibiaGdw+LURERETqsKeFiIjIQJTnhx3qg8H3tMyePRsSiQTjxo1TOX/8+HG0b98eVlZWsLW1RVBQEF69elU2QRIREemBQgi9HOWVQfe0nD59Gj/++CP8/PxUzh8/fhydOnVCVFQUFi9eDFNTU5w/fx5SqcHnYERKuuzHXH7/6slPH6/Vyiz/88r+KeN1lh7uUjqMZd8abXuoKBSKYt9DH1/C2vZhyTizvNj3oJJnsEnLy5cv0b9/f/z888/4+uuvVcrGjx+PsWPHYsqUKcpz3t7epR0iERGRXnH1kGYG2zURERGBrl27IiQkROX848ePcfLkSTg6OiIwMBBOTk5o27Yt/vzzzzKKlIiISD+Env4rrwyypyU2NhZnz57F6dOn85XdvHkTADB9+nTMnTsXjRs3xurVq9GhQwdcvHgRXl5eatuUy+WQy+Uq54QQfGgiERGRkTC4npZ79+7h448/xtq1a2Funn98Om98dOTIkRgyZAiaNGmC+fPnw9vbG8uXFzwmGR0dDTs7O5VDKNJL7HUQEREVlhBCL0dRLFmyBB4eHjA3N0fLli1x6tQpPb+64jO4pCUhIQGPHz9G06ZNYWpqClNTUxw+fBiLFi2CqakpnJycAAC+vr4q19WrVw93794tsN2oqCikpqaqHBKpTYm+FiIiosIoq6Tl119/xYQJEzBt2jScPXsWjRo1QlhYGB4/flwCr7LoDC5p6dChA5KSkpCYmKg8mjdvjv79+yMxMRG1a9eGi4sLrly5onLd1atX4e7uXmC7MpkMtra2KgeHhoiIiIB58+Zh+PDhGDJkCHx9fbF06VJYWlpqHMEoCwY3p8XGxgYNGjRQOWdlZQUHBwfl+YkTJ2LatGlo1KgRGjdujFWrVuHy5cvYtGlTWYRMRESkF/qaQqtuHqdMJoNMJstX9/Xr10hISEBUVJTynFQqRUhICI4fP66niPREGIG2bduKjz/+WOVcdHS0cHV1FZaWliIgIED88ccfxbpHVlaWmDZtmsjKymIb5agNQ4iBbZTfNgwhBrZhuG2UpWnTpgm8yYGUx7Rp09TWffDggQAgjh07pnJ+4sSJokWLFqUQre6MImkpDampqQKASE1NZRvlqA1DiIFtlN82DCEGtmG4bZSlrKwskZqaqnIUlIAZU9JicMNDREREVDwFDQWpU7VqVZiYmCAlJUXlfEpKCpydnUsivCIzuIm4REREVHrMzMzQrFkzHDhwQHlOoVDgwIEDCAgIKMPI8mNPCxERUQU3YcIEhIeHo3nz5mjRogUWLFiAjIwMDBkypKxDU8Gk5f+TyWSYNm2azt1pbMM42jCEGNhG+W3DEGJgG4bbhjF5//338eTJE0ydOhXJyclo3Lgx9uzZo9wbzVBIhODTmYiIiMjwcU4LERERGQUmLURERGQUmLQQERGRUWDSQkREREaBSYsB4txoIiKi/CrskuenT59i+fLlOH78OJKTkwEAzs7OCAwMxODBg1GtWrUyi00mk+H8+fOoV69emcVARERkaCrkkufTp08jLCwMlpaWCAkJUa5DT0lJwYEDB5CZmYm9e/eiefPmGtt59eoVEhISUKVKFfj6+qqUZWVlYcOGDRg0aFCB10+YMEHt+YULF2LAgAFwcHAA8OaR4VQxnTp1Kl9iHRAQgBYtWujchkKhgFSav1NVoVDg/v37cHNzK3Rc7du3x4oVK+Du7q61rlwuh1QqRaVKlQAAN27cwPLly3H37l24u7tj2LBhqFWrltZ2zp8/j4SEBAQHB6N27dr466+/sGTJEigUCrzzzjsICwsr9Oug4jPEz2hZfD6pdFTIpKVVq1Zo1KgRli5dColEolImhMCoUaNw4cIFjY/kvnr1KkJDQ3H37l1IJBK0adMGsbGxqF69OoA3CZCLiwtyc3MLbEMqlaJRo0awt7dXOX/48GE0b94cVlZWkEgkOHjwYIFtnD17FpUrV1b+ofrll1+wdOlS5R+4yMhI9O3bV9tbgu+//x6nTp1Cly5d0LdvX/zyyy+Ijo6GQqFAr1698NVXX8HUVHPH3OvXr7F161a1vVc9evSAmZmZ1jgA4P79+7C3t4e1tbXK+ezsbBw/fhxBQUE6tfNPtWvXxt69e+Hl5aXT/c3NzVG1alUAwB9//KHynkZEROi0tfXOnTtx6tQphIWFoXXr1jh48CDmzp2rfE9HjBhR4LWPHz9G7969cfToUbi5uakk1nfv3kXr1q3x22+/wdHRscA20tLS8OGHH2LHjh2wtbXFyJEjMW3aNJiYmCjb0vYZ3b59u9rzvXr1wsKFC1GzZk0AQPfu3QtsIzg4GJGRkXj33Xdx9OhRdOjQAd7e3qhXrx6uXr2KK1euYP/+/Rrf082bN6NPnz6wt7eHXC7Hli1b8N5776F58+YwMTHB/v37sXr1anzwwQcFtpGnuF+yTALfMITPqKF8PqkUld2zGsuOubm5+Pvvvwss//vvv4W5ubnGNnr27Cm6du0qnjx5Iq5duya6du0qatWqJe7cuSOEECI5OVlIpVKNbURHR4tatWqJAwcOqJw3NTUVf/31l06vxc/PT8TFxQkhhPj555+FhYWFGDt2rIiJiRHjxo0T1tbWYtmyZRrbmDlzprCxsRG9e/cWzs7OYvbs2cLBwUF8/fXX4ptvvhHVqlUTU6dO1djGtWvXRO3atYW5ublo27at6NOnj+jTp49o27atMDc3F56enuLatWsa23j48KHw9/cXUqlUmJiYiIEDB4r09HRluS7v6cKFC9UeJiYmIioqSvmzJi1atBA7duwQQgixdetWIZVKRffu3cXkyZPFO++8IypVqqQsL8jSpUuFqampaNasmbC1tRW//PKLsLGxER9++KEYOXKksLCwEAsWLCjw+t69e4uAgABx+fLlfGWXL18WgYGB4t1339UYw9ixY0XdunXFxo0bxc8//yzc3d1F165dhVwuF0K8eT8lEonGNiQSiZBKpUIikRR4aPud2NraiqtXrwohhGjbtq0YP368SvkXX3whWrdurbGNpk2biq+//loIIcT69euFvb29+Oqrr5Tlc+fOFY0bN9bYRkpKimjTpo2QSCTC3d1dtGjRQrRo0UK4u7sLiUQi2rRpI1JSUgq8PjU1Vbz33nvC3NxcODo6ii+//FLk5OQoy3X5fG7btk3tYWJiIr7//nvlz5q0bdtWbNy4UQghxJ9//ilkMpnw8/MT77//vmjSpImwtLTM97Tef/vtt9+EiYmJcHBwENbW1iIuLk7Y29uLkJAQERYWJkxMTMTatWs1tmEIn1FD+XxS6amQSYuHh4dYtWpVgeWrVq0S7u7uGttwdHQUFy5cUP6sUCjEqFGjhJubm7hx44ZOf4EJIcSpU6dE3bp1xSeffCJev34thChc0mJhYSFu374thBCiSZMm4qefflIpX7t2rfD19dXYRp06dcRvv/0mhBAiMTFRmJiYiDVr1ijLN2/eLDw9PTW2ERISInr06KH2Me6pqamiR48eIjQ0VGMbgwYNEi1bthSnT58WcXFxolmzZqJ58+bi2bNnQgjdv2RdXV2Fh4eHyiGRSESNGjWEh4eHqFWrlsY2rKysxM2bN4UQQrRs2VLMnj1bpXzx4sWiSZMmGtvw9fVV/i4OHjwozM3NxZIlS5TlK1asEPXq1Svwemtra3H27NkCy8+cOSOsra01xuDm5iYOHTqk/PnJkyeiRYsWIjQ0VGRlZen0Ge3UqZPo2rVrvi/zwnxGrayslP9IcHJyEomJiSrl169f1/parKysxK1bt4QQb/6sVapUSeXP340bN7S2UdwvWSaBqgzhM2oon08qPRUyafn++++FTCYTY8eOFdu2bRMnTpwQJ06cENu2bRNjx44VFhYWKl8w6tjY2IhLly7lOx8RESFcXV3FkSNHdEpahBAiPT1dDBo0SPj5+YmkpCRRqVIlnf/AOTg4iDNnzggh3iRS6v7AWVhYaGzDwsJC2UMkhBCVKlUSFy9eVP58+/ZtYWlpqbWNpKSkAssvXLigNQ4XFxdx8uRJ5c9ZWVmiW7duonHjxuK///2vTl+yI0eOFI0bN873uynMX2J2dnbi/PnzQog372ne/+e5fv26Tu/Hv9/Tf74/t27d0tiGg4ODiI+PL7D80KFDwsHBQWsMeclXnrS0NBEQECDat28vbt68qdNndN68eaJmzZoqvUuFeT/bt28vvv32WyGEEIGBgfn+wbBp0ybh5uamsQ1nZ2fl5/zZs2dCIpGofNmdOnVKODs7a2yjuF+yTAJVGcpn1BA+n1R6KmTSIoQQsbGxomXLlsLU1FT5LxxTU1PRsmVL8euvv2q93t/fX6xevVptWUREhLC3t9c5acmzfv164eTkJKRSqc5/4AYMGCCGDRsmhBDivffeE1988YVK+TfffCMaNmyosY1atWqJ3bt3CyGEuHr1qpBKpWLDhg3K8l27dgkPDw+NbVSvXl3jkMn27dtF9erVNbZhZWWl/BdknuzsbNGzZ0/h5+cnLly4oNN7unnzZlGzZk2xePFi5bnC/CXWvXt3MWXKFCGEEGFhYfmGk37++Wfh5eWlsY28xFUIIR48eCAkEonYtWuXsjw+Pl64uroWeP1HH30k3N3dxebNm1V6r1JTU8XmzZuFh4eHiIyM1BiDt7e3yj3zpKeni4CAANGoUSOdP6Pnzp0Tvr6+YsSIESIjI6NQ7+exY8eEnZ2dmDZtmli8eLGoWrWq+OKLL8TatWvF1KlThb29vZgzZ47GNgYMGCBatmwp1qxZI7p16ybCwsJEq1atxN9//y0uX74s2rZtq3UoorhfskwCVRnSZ7SsP59Ueips0pLn9evX4uHDh+Lhw4fK4RldfPPNN6Jz584Flo8ePVprV7E69+7dE1u3bhUvX77Uqf6DBw+Eh4eHCAoKEhMmTBAWFhaiTZs2Yvjw4SIoKEiYmZmp/Uvhn7744gtRrVo18eGHH4patWqJKVOmCDc3NxETEyOWLl0qatasma8L+t++/PJLUblyZTFv3jxx/vx5kZycLJKTk8X58+fFvHnzRJUqVcS0adM0ttGwYUOxadOmfOfzEhc3Nzedv2Tv378v2rdvLzp16iQePXpUqL/ELl26JBwcHMSgQYPEzJkzhbW1tRgwYICYNWuWGDRokJDJZGLFihUa24iIiBBeXl7i66+/Fi1atBDh4eHCx8dH7N69W+zZs0c0bNhQDB06tMDrs7KyxKhRo4SZmZmQSqXC3NxcmJubC6lUKszMzMTo0aNFVlaWxhjGjBlT4Bd5WlqaaNmyZaES68zMTDFy5Ejh5eUlTExMdH4/hXjzxdCqVat8QyE1atTQOLcnT3JysujYsaOwtrYWYWFh4sWLFyIyMlI5nOLl5SWuX7+usY3ifskyCVRV0GdUIpGUyWe0LD+fVHoqfNJSHjx//lxMnjxZ+Pr6CnNzc2FmZibc3d3FBx98IE6fPq31+tzcXDFr1izx9ttvi2+++UYoFAqxfv16UbNmTeHg4CAGDx6sUxI1e/ZsUb16deUXSd7YffXq1XX6l8qkSZMKnPeSnZ0tunfvXqhEUKFQiG+++UY4OzsX+i+x69evi759+wobGxvlX2CVKlUSgYGBYsuWLVqvf/nypRg+fLho0KCBGDFihJDL5eK7774TZmZmQiKRiODgYI2TPvOkpqaKgwcPinXr1ol169aJgwcPqp03pM6zZ89Uhvn+LS0tTWPPQ0G2bdsmxo0bp1P8//b48WNx4sQJcezYMeXwRHHcuHFDJCUliezsbK11i5sIVpQkUCKR6JQE5klNTRUHDhxQfkYPHDhQ7M+oQqEQQhTtM5o3zF/cz+e/e9XIMFTIJc9Usm7duqWynFTXPQ5ycnKQmZkJW1vbAssfPHig07LQf0pISMCff/6JQYMGoXLlyoW6VgiBx48fQ6FQoGrVqsplpkWVlZWF7Oxs2NjYFKsdKrq0tDQkJCSofEabNWtW4Ocuz/Pnz/Hw4UPUr19fbXl6ejrOnj2Ltm3bFiqe7du349ChQ4iKitK4PFidJ0+e4ObNm1AoFKhevTo8PDwKdf2/3bx5E5mZmfDx8dG6xUFBzMzMir05ZnHbMIQYqGQwaaFSce/ePUybNg3Lly9nGzq2UdzNC9lGfn///TdOnDiBgIAA+Pj44PLly1i4cCHkcjkGDBiA9u3bl+j1BbWxYMECvH79utBtBAYGwtvbu1hxFLUNfWyOWdw2DCEGKmVl2s9DFUZiYmKhJyZX5DauXLmi3D9EKpWKoKAg8eDBA2W5LitV1LXx8OHDCtvG7t27hZmZmahSpYowNzcXu3fvFtWqVRMhISGiffv2wsTEJN+eSfq8vry1IZFIROPGjUVwcLDKIZFIhL+/vwgODhbt2rUr0TYMIQYqXexpIb0oaGfKPDdv3sQnn3xSpN1XK2Ib77zzDrKzs7Fy5Uq8ePEC48aNw6VLlxAfHw83NzeddrNlG6oCAwPRvn17fP3114iNjcVHH32E0aNHY9asWQCAqKgoJCQkYN++fSVyfXlrY/bs2fjpp5/wf//3fyq9MpUqVcL58+fz9YaVRBuGEAOVsrLOmqh80MemWWzjf/SxeSHbUGVra6vclTk3N1eYmpqq7NuSlJQknJycSuz68taGEMXfHFMfbRhCDFR68j9Ag6gIqlevjs2bN0OhUKg9zp49yzYK0carV69UJkJKJBLExMSgW7duaNu2La5evao1BraRX96zxqRSKczNzWFnZ6css7GxQWpqaoleX97a8Pf3R0JCAp48eYLmzZvj4sWL+Z7nVtJtGEIMVHqYtJBeNGvWDAkJCQWWSyQSCC0jkWzjf3x8fHDmzJl857///nv06NFD4wPg2IZ6Hh4euHbtmvLn48ePqzzc8O7du8oHnpbE9eWtjTzW1tZYtWoVoqKiEBISonGIrqTaMIQYqHQwaSG9mDhxIgIDAwss9/T0xKFDh9iGjm288847WL9+vdqy77//Hv369dOaOLENVaNHj1b5ImrQoIFK783u3bs1rpgp7vXlrY1/69u3L86cOYPNmzcXelsCfbVhCDFQyeJEXCIiIjIK7GkhIiIio8CkhYiIiIwCkxYiIiIyCkxaiIiIyCgwaaFyRSKRqBxSqRT29vZ466238H//939aV5iUtJUrV0IikWD69Okq5wcPHgyJRIL4+PgyiauogoODIZFIcPv2bZ3qF/T6i8LDw6NU9tIw1t8NUXnEpIXKpfDwcISHh6N///7w9fXF0aNHMXz4cHzwwQdlHVqJ0WdCQERkiIr27HEiA7dy5UqVn+Pi4tClSxfExsaif//+ePvtt8smsAJER0djypQpKht8ERGRKva0UIXQsWNHDBw4EACwdevWsg1GjerVq8PHxweWlpZlHQoRkcFi0kIVRpMmTQAA9+7dU56TSCTw8PDA69ev8dVXX8HHxwcymQw9e/ZU1snMzER0dDSaNGkCa2trWFtbo1WrVli1alWB9zp69ChCQkJgY2MDe3t7hIWF4eTJkwXW1zRvIiMjA3PmzEHz5s1ha2sLKysr+Pj4ICIiQvnMneDgYAwZMgQAMGPGDJV5Pf/udfr7778xePBg1KxZEzKZDE5OTujbty/++usvtbHl5uZi7ty58PHxgbm5OWrWrImPP/4YaWlpBb6ewnr06BG+/fZbtG3bFjVq1ICZmRmcnZ3Rq1cvnD59WuO1QggsXLgQvr6+MDc3R40aNTB27Fi8ePGiwPrr169H+/btUblyZZibm6NevXqYPn06MjMzdY75yZMnmDJlCnx9fWFtbQ07OzvUrVsXgwYNwqlTpwrz8olIRxweogojPT0dACCTyVTOKxQK9OzZE0eOHEHbtm3h5+cHBwcHAMDjx4/RsWNHXLhwAc7Ozmjbti2EEDh27BgGDx6MM2fOYPHixSrt7dy5E++88w5ycnLQokUL1K5dG+fPn0dQUBAGDx5cqJgfPXqEjh074q+//kLlypURHBwMmUyGmzdvYunSpfDy8kLdunXRqVMn5OTk4OjRo2jUqBEaN26sbMPT01P5/1u3bkXfvn0hl8vRuHFjtGrVCvfu3cOGDRuwY8cO7N69G0FBQSoxDBgwALGxsbC0tERoaChMTU2xatUqHD16FJUqVSrU6ynItm3bMHnyZHh7e8PPzw+2tra4du0atmzZgp07d2Lnzp0IDQ1Ve+2YMWPw008/ITg4GA0bNsThw4exePFiHD58GH/88QdsbW2VdRUKBQYMGID169fD2toazZs3R+XKlXHmzBnMmDEDu3fvRnx8PCwsLDTGm56ejpYtW+LWrVuoWbMmOnbsCFNTU9y9exexsbGoXbs2WrRooZf3hoj+ofQfLE1UcgAIdR9rhUIhAgICBADx+eef56vv6ekp7t+/n++6Ll26CADi448/FllZWcrzycnJonnz5gKA2L17t/J8WlqaqFatmgAgli9frnL/yZMnK+83bdo0lfuEh4cLAOLQoUMq5zt06CAAiD59+oj09HSVslu3bonz588rf16xYoXatv9Z38rKSlhbW4u4uDiVst27d4tKlSqJmjVrCrlcrjwfGxsrAAg3Nzdx69Yt5fmUlBTRoEED5ev5Z5kmBcV44cIFcfHixXz19+zZI8zMzESdOnWEQqFQKXN3dxcAhK2trThz5ozyfHp6umjfvr3y9/ZP3377rQAggoODxaNHj5Tn5XK5GDZsmAAgJk+erHKNut/N8uXLBQDRvXt3kZubq1L/8ePHIikpSZe3g4gKiUkLlSv/TlpycnLE1atXxeDBgwUAIZPJxPXr1/PV37hxY762zp07JwAIf3//fF9MQghx9uxZ5RdXnrwvs6CgoHz1X79+LVxdXXVOWk6ePCkACEdHR5GWlqb1tWtLWj7++GMBQCxevFht+dixYwUAsXnzZuW5oKCgfAlYnt27d+stadGkf//+AoC4cOGCyvm8pOWzzz7Ld81ff/0lJBKJsLa2Fq9evRJCCJGdnS2qVq0qrKysRHJycr5rMjMzhbOzs6hcubLK71vd72bOnDkCgFiwYIHOr4OIio9zWqhcypvPYWpqirp162LlypWwsbHB+vXrUadOnXx1u3Xrlq+Nffv2AQB69uwJqTT/H5W8OS7/nL/wxx9/AHjzpNh/q1SpEt59912dX8P+/fsBAP369YONjY3O1xUk7/X06tVLbflbb70FAMrXk52djRMnTgAA3n///Xz1O3XqhMqVKxc7rjxyuRzbtm3D559/jhEjRmDw4MEYPHgwkpKSAADXrl1Te52699rX1xeNGjXCy5cvce7cOQDA2bNn8fTpUwQGBsLJySnfNRYWFmjWrBmeP39e4L3yNGvWDADw3XffITY2Vjn0SEQli3NaqFwKDw8HAEilUtja2qJhw4bo1auX2i9ZR0fHfPNcACg3TPv888/x+eefF3ivrKws5f8/fPgQAAp8pL2Hh4euL0E5YfjfSVZR5b2eGjVqaKz39OlTAMB///tfvH79GtWqVStwVZO7uzueP39e7NiSkpLQvXt3jZvUFZQYaHqvExMTlb+TvLbj4uK0bkr39OlTeHt7F1jeoUMHjB8/HgsWLEC/fv1gamqKpk2bomPHjhg6dChq166tsX0iKhomLVQu/XvFjCbm5uZqzysUCgBAmzZt9JY4lKW815OX0BWkZcuWpRGOkhACffr0we3btzFq1CiMGjUKtWvXhrW1NSQSCT777DNER0cXezfjvNfv6emJ1q1ba6ybNxFbk3nz5mHkyJHYtm0b9u/fj6NHj+LUqVP49ttvsX79evTu3btY8RJRfkxaiArg6uoK4M3w0CeffKLTNdWrVwcA3LlzR215QefVqVmzJgDgxo0bOl+jiaurK27cuIH//Oc/On0pOzg4wMzMDE+ePMGrV6/Urqi5e/duseO6fPkyLl++jObNmyMmJiZf+c2bNzVef+fOHTRs2FDteQBwcXEB8L/fp4+PT6GSWk28vb0xadIkTJo0CVlZWfj+++8xceJEjB49mkkLUQngnBaiAnTs2BEAsGXLFp2vyZsXsmHDhnxlOTk5+O2333RuKyQkBACwfv16vHz5Umt9MzMz5X3UKezrqVSpkrLXRd3r2bdvH549e6ZTW5rkDS/lJRX/LouLi9N4vbrYLl++jMTERFhbWyuXf/v7+8POzg6HDx/WS9z/Zm5ujk8//RTVq1fHkydP8PjxY73fg6iiY9JCVICWLVuiY8eOOHr0KCIiItRupnb+/Hns2bNH+fN7770HBwcHxMfHq2w+J4TAtGnTCtUz0aJFC7Rr1w6PHz/GiBEjkJGRoVJ++/Zt5SRV4H89CleuXFHb3ieffAILCwt8+umn2Lx5c75yuVyOTZs24f79+8pzo0ePBoB8sT99+hQTJ07U+bVo4unpCalUioMHD6pMgM3KysKoUaO0JhiLFy9WTrYF3mwGOGbMGAghMGTIEGUPkUwmw6RJk5Ceno5evXqp7cF58OABfvnlF60xb926VTlJ+Z8SEhKQkpICa2tr2Nvba22HiAqpbBcvEekXCtinRVN9d3f3AstTUlJEkyZNBABhb28vgoODxQcffCC6du0qatasqXYvkK1btwoTExMBQLRs2VL069dP+Pr6ikqVKonhw4cXap+W+/fvC29vbwFAVKlSRXTv3l289957omnTpkIqlYr58+cr67569Uo4OjoKAKJt27ZiyJAhYtiwYeLo0aMqsVlaWir3punWrZvo27eveOutt4SVlZUAIM6dO6cSw3vvvScACCsrK9G9e3fRq1cvYW9vL5o2bSpatWqllyXPee+LhYWF6Nq1q3j33XeFk5OTqFq1qnK5+ooVK1SuyVvyHBERISpVqiTCwsJEnz59hLOzswAg6tevL168eKFyTW5urhg4cKAAIMzMzETLli1F3759Ra9evUT9+vWFRCIRjRo10vq7yVs+XqNGDfH222+LDz74QAQHByt/7//5z390ej+IqHCYtFC5ou+kRYg3ycCiRYtEYGCgsLOzE2ZmZqJmzZqibdu24rvvvhP37t3Ld82RI0dEu3bthJWVlbC1tRUdOnQQx44dK/BLu6CkRYg3G9Z99dVXws/PT1hYWAhra2vh4+MjIiMjxbVr11Tqnj59WnTs2FHY2dkJiUSi9sv++vXr4qOPPhJeXl7C3Nxc2NjYCG9vb9G3b1+xYcMGlc3lhHizv8mcOXNE3bp1hZmZmXBxcREfffSRePHihWjbtq1ekpacnBzxn//8R/j6+gpzc3Ph5OQk+vfvL27fvi2mTZumMWnJzc0Vc+fOFT4+PkImk4nq1auLiIgI8ezZswLj2LZtm+jatatwdHQUlSpVEo6OjqJZs2Zi0qRJIiEhQaWuut/NuXPnxCeffCL8/f2Fo6OjkMlkwt3dXXTr1k3s379fp/eCiApPIkQxp+QTERERlQLOaSEiIiKjwKSFiIiIjAKTFiIiIjIKTFqIiIjIKDBpISIiIqPApIWIiIiMApMWIiIiMgpMWoiIiMgoMGkhIiIio8CkhYiIiIwCkxYiIiIyCkxaiIiIyCj8P+QnT2/fmubNAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "generate_confusion_matrix(y_train, y_pred_train)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.8.10 64-bit",
   "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": "3.8.10"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "916dbcbb3f70747c44a77c7bcd40155683ae19c65e1c03b4aa3499c5328201f1"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}