{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "7DBH3hTK8MYH" }, "source": [ "# Download Dataset" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "EVmzxH6f47A4", "outputId": "d501ccfb-b478-44ba-95c3-76ee19ca0c0c" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "--2023-06-21 23:28:09-- https://dsserver-prod-resources-1.s3.amazonaws.com/cnn/xray_dataset.tar.gz\n", "Resolving dsserver-prod-resources-1.s3.amazonaws.com (dsserver-prod-resources-1.s3.amazonaws.com)... 52.216.38.145, 3.5.28.100, 54.231.233.153, ...\n", "Connecting to dsserver-prod-resources-1.s3.amazonaws.com (dsserver-prod-resources-1.s3.amazonaws.com)|52.216.38.145|:443... connected.\n", "HTTP request sent, awaiting response... 200 OK\n", "Length: 1223407715 (1.1G) [application/x-gzip]\n", "Saving to: ‘xray_dataset.tar.gz’\n", "\n", "xray_dataset.tar.gz 100%[===================>] 1.14G 13.1MB/s in 91s \n", "\n", "2023-06-21 23:29:41 (12.8 MB/s) - ‘xray_dataset.tar.gz’ saved [1223407715/1223407715]\n", "\n" ] } ], "source": [ "!wget https://dsserver-prod-resources-1.s3.amazonaws.com/cnn/xray_dataset.tar.gz\n", "\n", "import tarfile\n", "\n", "def extract_tar_gz(file_path, output_path):\n", " with tarfile.open(file_path, 'r:gz') as tar:\n", " tar.extractall(path=output_path)\n", "\n", "extract_tar_gz('xray_dataset.tar.gz', '.')" ] }, { "cell_type": "markdown", "metadata": { "id": "1aPY5RjXxBHs" }, "source": [ "# Detect Pneumonia Using X-Ray Images with CNNs and Transfer Learning\n", "\n", "In this project we will build two deep learning models and train them on a dataset containing images of chest X-rays. The aim of this project is to train classifiers on the dataset that can accurately predict whether an X-ray indicates any signs of pneumonia or not.\n", "\n", "The models could then be used for developing tools that could aid hospitals in accurately identifying whether a patient has a particular disease or not." ] }, { "cell_type": "markdown", "metadata": { "id": "SRDzeeTlthL6" }, "source": [ "## 1. Introduction" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "id": "QyEDhskxMM7v" }, "outputs": [], "source": [ "import tensorflow as tf\n", "import numpy as np\n", "import tensorflow as tf\n", "from tensorflow.keras import layers, models, Model, Input, applications, initializers\n", "import matplotlib.pyplot as plt\n", "%matplotlib inline" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "Rk2KZa2lUs_z", "outputId": "63e3b6bd-bf3b-46ce-a16e-889e72f5955b" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "Found 5232 files belonging to 2 classes.\n", "Using 4186 files for training.\n", "Found 5232 files belonging to 2 classes.\n", "Using 1046 files for validation.\n", "Found 624 files belonging to 2 classes.\n" ] } ], "source": [ "img_size = 256\n", "\n", "train_set = tf.keras.utils.image_dataset_from_directory(\n", " directory='chest_xray/train/',\n", " labels='inferred',\n", " label_mode='categorical',\n", " batch_size=128,\n", " image_size=(img_size, img_size),\n", " validation_split=0.20,\n", " subset=\"training\",\n", " seed=417)\n", "\n", "validation_set = tf.keras.utils.image_dataset_from_directory(\n", " directory='chest_xray/train/',\n", " labels='inferred',\n", " label_mode='categorical',\n", " batch_size=128,\n", " image_size=(img_size, img_size),\n", " validation_split=0.20,\n", " subset=\"validation\",\n", " seed=417)\n", "\n", "test_set = tf.keras.utils.image_dataset_from_directory(\n", " directory='chest_xray/test/',\n", " labels='inferred',\n", " label_mode='categorical',\n", " batch_size=128,\n", " image_size=(img_size, img_size))\n", "\n", "normalization_layer = layers.Rescaling(1/255)\n", "\n", "train_set_normalized = train_set.map(lambda x, y: (normalization_layer(x), y))\n", "validation_set_normalized = validation_set.map(lambda x, y: (normalization_layer(x), y))\n", "test_set_normalized = test_set.map(lambda x, y: (normalization_layer(x), y))" ] }, { "cell_type": "markdown", "metadata": { "id": "EhccP55yU4zK" }, "source": [ "## 2. Data Exploration" ] }, { "cell_type": "code", "source": [ "train_set.class_names" ], "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "svYNGzLxRiYk", "outputId": "c363d8c9-5121-4a45-cbe2-4bc88d9d6673" }, "execution_count": 4, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "['NORMAL', 'PNEUMONIA']" ] }, "metadata": {}, "execution_count": 4 } ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "id": "Vnm09QoaVNbe" }, "outputs": [], "source": [ "train_set_elem = train_set.take(1)\n", "for images, labels in train_set_elem:\n", " images = images.numpy()\n", " labels = labels.numpy()" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 470 }, "id": "z_BJZO6Fk5Pl", "outputId": "70ebd3ff-2650-4865-f006-7e08fbf87e0b" }, "outputs": [ { "output_type": "display_data", "data": { "text/plain": [ "
" ], "image/png": "iVBORw0KGgoAAAANSUhEUgAAAa4AAAGiCAYAAAC/NyLhAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOz9e6it7XrfhV9jjHkYY57Weg/73TuvyU43tZoo2MYk3d1QQe2WJJXWQwpN6R+llBbEXdCNiIHamiIERDBUoy0iLQXrCSGgtgGNYETS1KS2HtDalrQ2h/e81ppzjjnGPI3x+2Pxuefn+c57zHV6k/zm6rphMsd4xvPch+u+7ut7ne77Ga3X63W9KW/Km/KmvClvyj0p41/vDrwpb8qb8qa8KW/Ki5Q3wPWmvClvypvyptyr8ga43pQ35U15U96Ue1XeANeb8qa8KW/Km3KvyhvgelPelDflTXlT7lV5A1xvypvyprwpb8q9Km+A6015U96UN+VNuVflDXC9KW/Km/KmvCn3qrwBrjflTXlT3pQ35V6VN8D1prwpb8qb8qbcq/LrBlw//uM/Xr/hN/yGmk6n9dWvfrX+yl/5K79eXXlT3pQ35U15U+5R+XUBrv/iv/gv6pvf/Gb9iT/xJ+qv/tW/Wr/5N//m+r7v+7766KOPfj2686a8KW/Km/Km3KMy+vU4ZPerX/1qfe/3fm/9B//Bf1BVVavVqr7t276t/ugf/aP1b/wb/8avdXfelDflTXlT3pR7VLZ+rRu8uLion//5n68f/uEfbtfG43F9/etfr5/5mZ/pPnN+fl7n5+ft+2q1qs8++6zeeeedGo1Gv+p9flPelDflTXlTPt+yXq/r5OSk3n///RqPX8z592sOXJ988kldX1/XF7/4xcH1L37xi/X//D//T/eZH/3RH60f+ZEf+bXo3pvyprwpb8qb8mtY/t7f+3v1rd/6rS/0zK85cL1M+eEf/uH65je/2b4/efKkvvzlL9d3fMd31M7OTk2n0xqPxzUej2tra6u2t7drPB7XZDKp7e3tGo1G7W8ymdRkMhncV1XtvqqnFuB6va71et2e49pqtaqqqtlsVl/+8pdrd3e3rq6uarVa1dXVVV1fX9fW1lOyrtfrur6+rqpqbdPPyWRSl5eX7fPu7m7r22q1qsvLy7q+vq7xeFyPHz+uq6ur2traqtFoVHh3/Zn+u376e319XVdXV3V1dVX2DNO39XpdFxcXtbW1Vev1ulm4FxcXrY+M++23366jo6Pa39+vo6OjNubz8/NaLBZ1fX1dq9Wqtre323gY23g8rt3d3QFtPAbusRW9Xq/r6uqqzdVoNGr0Xa1WNZlMbs0R416tVjUejwf8MZlM2hxPJpO6vr6u6XRau7u7tbu72/oPTajH/Vmv160+/rif37a3t+vi4mLQpvtJG8w144cmPJe0uL6+rtPT0y6PwcvMyeXlZWuT+c7IAPSnHtPU9/v/er2uy8vLdu3q6qq1wfPmL/pAv/iDtufn53V1dVUXFxd1cnLS+H5ra2uwZvb39wfrl/6t1+tGj6oa/Aa/QR+PCXqtVqvG+173VdXmkLmiP5PJpA4ODlod6/W68fzFxUW7Bu+wBhkzf+7LYrGoxWJRx8fH7V7oZPlF33q8YZ6ibtOb+/jz/efn5+0+ZIX5Iufu6uqqzs/Paz6f1+np6YCvvI7Ng3yHbsvlsg4PD+tFy685cL377rs1mUzqww8/HFz/8MMP60tf+lL3GYRKFsAHgW5BsrW1VTs7O+07Ao/79vb2Br/xH+Zgwra3t28J+9FoVNPptP31mMygYaHJYtza2mqCrqpqOp3W1tZW6/disaiLi4u6vr6uo6OjAeNZOFNfjj/7PJ1OazQa1XK5bOOEudfrde3t7TVmrHoKhDDzaDRqC+3s7KyNGyBiQW9tbdXl5WVdXFzUaDSq3d3d2tnZaQqEx19VdXl52QAXkDGgJB2pk/EiLKpuFu7W1lYbP2NJ4LJbwvOLoOP3xWIxACkWrfsB4Fc9FXIIre3t7YECVVUDfqOMx+M6OzsbCCc/g/Df3d1t/GDwZ8541oKJ8fObBT1zNplMarFYtD6hIFlZqao2//RvsVg0vt/Z2Wn9MYgwbwZ7Pl9cXDQgs6D0HE0mkzo8PGxtm5+Sxz2nVtqYC+YPXqAf9NvFwGVaMx+sUXgFeUPZ2dmp8/PzWq/XtxRIlBTmyQJ8e3u7jXN7e7spm1Zm4Pdev5EnrCn4YGdnZwA+1LW1tTWYI35jng2wVrgYE2sMHmUeErAynJPXXybc82ueVbizs1Pf/d3fXT/1Uz/Vrq1Wq/qpn/qp+trXvvZCdaWGlJaRtTpbKamp+Lf0tSZQsKDQUhAanmT3wVYSCxTh7v5YY2XBcS0XKiAMMLNA3HczZw84bX0yHupFGaANW6iTyaSWy2UdHx/XyclJE2DUY8Y3bemT54rFDgAakFgUCF/GB80NbCkYPU8IrBSStniSX1wfYzdNWcDUmfzT89ezoO+iB/dV1UBoXV1d1XK5HPBbWqnwKjwEfczD29vbtbOz0wT5arVqws71pXZMHeZxj4k54rvXXFqZvm4F4fLysq0Xe0vgYUCC+UCIw5O2XKxA2mLrCUv3pSdw/QwWrK1x3+8CrXd2dgbKlOnrefPzttC5z3znuaBkX1I56oFHz6Izv7hYBtoSOzs7a2DsuntgZF6oqgHgvWj5dXEVfvOb36w/8Af+QH3P93xP/dbf+lvrx37sx2o+n9cf/IN/8IXqsdCq6oPAJsT3/b3fYKacRAMXgmVnZ+eWNeSF5El3O2YoP2PXTwpDT34u0lyU2Yb747r8vepGY2WcpkVVNUE6n8+bFpwWscfVKyls3a8eHdP6op/ZZoKIFZueopO/p1DOMSTN/PumsZoeqTTleE0XK1zp5vUY3HfzyaY+UKxtU3JeXAyMplvSJu/39xwzCqDpDu8h8BPYbcG6rt56vssC8Hq0stHjuxx/VQ28KazBpIUt1956S5r0+tzre29O8r4eD2RbSbuUR70547NDGtnGXTLX7b9s+XUBrt/7e39vffzxx/XH//gfrw8++KB+y2/5LfWTP/mTtxI2nlXSmqrqa1A94Ml4yiaN2PXbqkDr2tnZqb29vUH9+MVZePTBLh0z+2QyaS4aYiNYPF4cXsR2CaL9oH1ak7emSz8MhKkZp7WRArTqxhp49OhRiw8dHR21OqwB9xjY46B+tHb6mEK8t9hXq1Vzr3C/6eGFiOWbfMKc0BZuTtxfuHfskvS4aJ9YDfOVQsDxEdOUPlhZsSVD3wzSjM9uHI8Zqzq1Y+7HTZd9S/CG/z1v1GUXIO5iAAe+cvtpBeJ1II5q9yRWIdY+BYsMJYn+JZCiSCZ/2J1HfdDc/215WwZsbW21OcHVD+/ipegpYBmDtKKZnoAMLSBnHA7wvPU+Z2wx1zfteV5RegwwKRvcZ3sbevLXxdc2yYUXLb9uyRnf+MY36hvf+MYr1dFbEPavJxBVDf39Jh4TZ0uGxAQzTNVtBrcryUI7Bb4DwABNCiWYAhAyAzv2Q/sIbFsiFqZmUhYVY0JI262W/n9iKggY0/3s7KzOz88HiSX8li5KLyALs1xcKfTpgxcVdO0lIKS7MYHYbVhLtsUCfyQ9XZ9dVFYMegkVV1dXA+Hu+QFEGD9zMhqNBmNL/ss5Mn8xl+kOdWDdfQSkTeuqG9eYedNxvgQm1gCft7e3B/zntQmfX11d3fIs5PzDW9PptMVvrNhY0WCc0NPWgMGNOqHn9fX1LberBW5aJFgbzAEx4fV6XTs7O7eEObxqwW1gyD723J0uXLfFnIqm5VAq+eZb8zUllQGvI/q3XC5vKfqWIUk72vX9L1vuRVbhs0oKqSR2glfV3UTzb8mwtGdtqTcxm9rM/qSlY/dF1U1QHo0vE0Bcr/uXtHE72XZV3WI4M2HVTcYbgoB+0hfchmkdZjFAWEi5zaTlXT54B5vT3cMzBqSsOxUNaJHaZmbYWYu2sKBYkNPP1JbJpPO9nlOPLYVAgpefT83cv6HMpJKRa8c0ZDzm0Z6HIkvOv61/a+zpSfA4EdDb29stwcjg3OPtnHvXZes81/mmvrtuK4kJFLkmPNeeq5RLBlu315tf/8/6NvV/kwx0G71Qx6Y/5s9AvGnuveaSxnfxzrPKvQau3sBhqE0I78WwqaQQ3ST0Li8vm4+XRWHgYYEkI/eEsBkfoTUej+vw8LBOTk7avefn57fM/KobF6iFbo9etDMajQaWBHX4sxM3JpNJ7ezs1HK5vGVhXF1d1ZMnT+rg4KCm02nNZrOu4LQwRkumLwk8mRmWAE+96X5NIITedtu6D86QsuvDPHR9fd2yK50RaUsKOmJBkBWGyxh3GMqHk3RMq4wXmAfhJ6fCW1imJcXcUWyF04/Ly8uBe8/gZWUmeZh70oXpz3ZPwkMed7rqXJ/d8tvb2zWbzWpvb6/xP5ly9NNZibZSnKrvvqc1mUDmebYVzj3MrWlqHrdrGjpQTy+py1Z3uvET5Db97rm2jOnJgwR6eN+8nJ9T4Voul62uXukpQHfJpxcp9xq4TIye9gTBmNjpdNqup9YIw+RChEGd5WTtzX5pLBLSTOmLBSaC4Pz8fLDXyYvXgpT09XQDpWXi56AJbWZWnOs3WCA0cXWwSOyq8+JASaCw9+v09LS++MUvDkCbPgMAVU/jSd5L5wSEXp/tcvKCZdHhLoJedrV4T5Xn2jEG6gVYLHSh7+Xl5a30YuYAEFitnsa7+EzMEkWHBe/+43pKJYlYD/1zbNZzlPETflsul22My+WyCVjcvgBYzyqxqxKBbN5KJcfCjnmwQnF+fl7L5bLOz88HoGgeGo1GzXIfjUb17rvv1uHhYc1ms/r444/bXLN+UojDv/TPLmsAL3nN6x2ATaXA/Ut39Gq1qsVi0VzmdlUiZ2iz6gYUSZn3HMIPbocMZNr3WBMUrVTknKbVbGsUJZaxO+aca94xUvfLdLQ8Sxr3jIoXLfcauDwJFshe+L4XhrEGaS2Nkq6btAZcZ9XmjKSsD6YwEKUVxP0GRrQ3kgQcZ8hU+WSMtHjctwR7CxrAyps5q6oJYNPNVgALgkWZbbIY0+rMa/6fLjZobgvFwizrMt178wcdexaK6crYnP7LOK11U6x1O5bTS+WHntmnXpKOAbY3jnTHOZnC9PQzrsdgmgKwRz8+e35ox5r6+fn5YGO96eq+kYCxt7dXb7311kB4W+B6X6b74bXU0/rhCe9zc/s91x732SJ2EkRVtexIt50KZYJg0qDncs51YXq5Lsu/9PzkPPnPdfTm2nOLlZ7AdVfprfOenHyRcq+ByxoxxUxWVbcWKUTLRennk6jW1LOdqrqlMSaj9FwiaH5mdLtsrEmiWZ6dndXu7m6zbCzckpEsaK3F57gMvtyDoAeMHI/Y3t5uQVkLDgCVdrEU0UDph4EFIZsuPrvF6JvdMgmw3IM1l3ESA5fnn+c8Dru1qIN2eM7KBEkrCBMLM2v7duWaF8yLZKpRyK5LK3FnZ2dwXwIX/GuNPt2jpltaS15TKVySjxLY7ao3D7OFAjoCvgl0ZOcdHBzUu+++WwcHB3V8fFynp6eDuTMQe69kru1U3my1p1Vh96KVCc9pWh+mCaCMtd9T/Py8lW1bRFZIEhxdPC767Xp7JZVl151AmPPKX8/iSiDtKYq9PrxsudfAZaLa6sqAPdq0tV5ndtkysFCijp626YklDbrqxsXlieb+XEhOo0X4kT1V9ZQB7f6azWZ1dnbWftvd3e1qYCl08v+mBUP/fGRNTwvzZzZB4w5DWOGaszZroLUbkT74VJHc20Mfnb2IkGAsdv3t7OwM5tCCkvrS8kklyHvTFovFwJ1D+97AizaKADafwlcITKw2lBKeX6+fujxns1njDUB0b29vcNIHfMxn8zh0ZIOotXmDmK0rCxTmzi41lJOqYezHiobddD42DNcpbaZ1SxzrnXfeqdls1jbufvDBB03Rg496G79T8XHfzDsUtrIwN4yBebLFSR14PABi87dBmvVjELQSCX3s6YEvemMyQPMcY7IbPde/eeJ5Cn20kpZJSfCurVX/7xXLi1cBK5d7DVxVt32+VbeD1uyPcnFQ30RPl+BdVpmZnUWRMS5rXDnJCCULZISehaHT4tG4rWGZuZPxU1DZGvHiSQvSNHUyQI4/hTmLBVqgnXENpcAuMPeH/rpOYkepETqeYE1/NBq1OAqnf1iZYRy+35ZQ0qyqWh30zencpkueIcjcWkjZyjQ9AZs8DQLBjTC3FZ4uSgTn5eVlO5HCip35G56y+yvXE/Xakksr2fNHH+wyzhian0FRe/jwYUvAgP98FFVaTuYTA1Vayl6z6abbBBrwcc9ysfLnvuXas6JgCzDXlXko1xftGEj8u+vyfZvG7P5Dq57rskcD2nJSTT6X6zPpk8/cBXbPKvceuKpuZw5ac7S2wu+j0eiWoKP0YlmuOyfNrqMeSKRgsPCwMPZRN45T2OpwphUJBznmtHBMo7T68rsXQWptFsTWYD3WZHa7PHARInwzwO0FYtrk4oTm7mcKtaobi4D2DbTWpD0nXCM5w8X7h6jTWXOmFX00+Kagy/HzLICe9IRm1A9wMU4LMzK+nAGa82V6pAJkmnE9hXFP6DBf1tzhUbvPTI+tra3a29urd955p1m4rCknuKTmn1aG15TX4KZYz13r5C6rwDRJ/ksepI92G6Yb1h6jHFsPNBOYNv3eA5FeP309+560gbfuok9PQcm+mq9ftrwWwGVtsOqG+Gk+m1kNGmSKIbSsvfVSpFlYo9FTv/bJyUm9++67VVXNwkAoZNzFrqEUZAgsn6RARhnj2tvbq4uLi+a6Ysw91yfj9okBjuPkArdg81jt508LiTIajQYn7LPbH+F6fHxci8Wi3n777QEI8N8JDygWFG+QNc2Ye/eHeXZ2FLFBC2zG6U2ye3t7A2FnQGCOJpOnh9Lyu4WXgfb6+uZAV/PoarUa9MXWpC1qTqpnDqtqoJlbMBhAF4tFAy27rKbT6S33Wg+gPV4sZX/GHegDml2Yb94WwLX8v7e3Vw8fPqz9/f1GdzIfvT8yLSE+e11W3QCjlSXoDX3NY7i3zevwRSpDKVvcDr8vFoumdJi+gC91WMHlbQr8eV3aFWilp9cfrqWs81p2gk5acH4GgHUMFXmHQuR7kxcTMN1nnkk348uUew1cqUHw34TlPwLGGVTpo3fMBQsnNaEUHAgLgCzdbMl4yVDUbddlVf+IolysBl9raSxixmiAss89LQJoZ/ck7VKc0rspFogm3XPL8doKXkeDcOd3Fq3BNmlmIE+3lsEJQUDavV22jC9ByjEjCzAsG1y5Tm6wFWFBaivJbypwuwi1dLt6/5EFkoUmwEPMgf1Rni8Xx2FTwCet6Qs8ahezaZt8RMq/3Zk+rJmEk/39/drd3W3zTL/hV/gg3aD0dTqdNiACLFMZzL5tErCuN/+bt4kXmf/SWjJAmO8pTuTx/cyh1637fReQWu75fo8zn3coJevN+5gHx+yz9ADL9dj78SqWFuXeA5f9yL1FaIGak+n77LdPRu/V78mxQLRQuqsPFlJZX2p/tO1NizxjQKKO9F0nzTa1mQukRyu37XHZ8qEPPcBGy7Vl5AQEa8lZUvD0/lLbZH68QdXZfUlvfkMzTOvULlsLd+px3MUarJ/JxWswtZC29m4eNHAh8A1aFuA59+aBTYLQIGXXp8dv8LLSY7qjwBC3w5IkXgctbL3dtc68ftjgXdWPS3sN51owH/XG72tWbDeVBMxc5+6TlU+Dv/+yPj/fG2eOYxOI9MbQA66esp7uWrff47Vef3t9etlyr4GLicbdkporQsMmbYISwsYC0ZPleBMlhbwFH78nCFYNXZoIJQT3bDZrWtd0Oh0wirUyu7Zs5SEkrC3TllPvLYBpv2p40rWfZTxVw7hOauVYKvTLgr03DtMYwXZwcDCYF7dNf1OztqVmGqdwOjs7a/uD7Kr0eZH8ZncYlowtQPq7Wq3a3jHG50SaXNDQA7eS+dLPmHctzLEwqPfq6qrm83mjizP3XB/C2+5Kz29Pu/c41+unQfnpdNpAJ+cJPoJH4X8nq9hCxyVImjx9wL1opcYWXIIl/TSfZVwsrereerV73/KA+qABrvveXkbqos/Jn/Z22DKjHrtu7bnxWDORxoqKFZxUlE0HFwPXJuBkvHYre10agHpKOmPJ9dADvOct9xq4qm4YyxZK/m73FwLGQsELMC0K6k8Qsnae1+xTtk+XBe/MMVsiZkD7uu3/Zm+OXVXJGGmR2bpJLQfGTwbkWfrP2A1IjAcBY0CjOMbg366vn77LhzR6H6BaNbQaLfBcrxd1AlzSkvkj6M8ceK9VJtiYjlZgHIPwi0RxJWX7KVCcZm8g8aKvGma+WnmaTCY1n88Hb6q2QDe/ux3zaWrE7iv0YE6Yc78M1PwCX+AKtEXb42UnXtBHsg+9LlNhxFrHgoX2p6ent9zHpqmBmWu5pg2wOWcUwMXz4GeYPwt4K2hW/OwOhS/dVr7jyvOX/MJ68jxmPJOS4Ya0FDf12+8QNN+mAkPxfPTWAnW8bLnXwJUDT4LkRFiIZvEE9lwPOTlpXvs+X3cfDYLZptvqgS9/vcBqjw5uY5PbMEvSjWfNwI55pKDNMVmj79HPrriqasIstd7UsJMufE4hlfPMOBzwJzHHAJf97NWTizFjV7Z6sEh7Aj/p5n5aMGWSgmNJPL8puSb5IwWzn/Neuul0Oui/3cGuOy1vKzdWshCoVrLM1znPtia2trbaFgcLQ2hjmvcEop+hby5+LhMvbKH26qDuTXze46UE594a5f5NsiS9QP4tgS5lU/7P+vN3J4y9aOnJ1Ozzi5Z7D1zWgntmK4WFlpk/PVDKjDoXWwxp/poB+LMLwhaLBXPVTeYci9ttWcjYMnPGl4VcL87DdQsy12/B7ZhFntlnl6wXay52rAXqMg3SgmDfEQkbuOk4+82JEfSXOq3R49aibt+X6dRo/qPRqA4PD2symTTrz9o49yedzFMJXswNvLazs1MHBwcNuDhVBDqZly3sreRkGrJfhzKbzdrzrAP6x6G0yafeX+jX0kB/YlLMMa5gXL0GTbuJLUy9kRce5fnZbFaLxWKwgdsgVXXjFoZuPucyLRbzBcV8aWBOJSPdfeb7qmE2Z881OB7fbEvge7pj+Zzx0J5ikQC3CQhxyff6Tnt2a7KWzG+2wMzzpidrxQpCT0m3tyPno6cA/n0NXFW3j8+BidK68uTajZICiEn2gkyNKomeqaaj0WgQgK6qwXeYjtePu327Bi0kYCC7tVzsH/dYfbqEsxWdbZiLEfdTArc38zpl2y9avL6+bnE62ndKMn0hwE7bp6enLZZiALB22QMR6O2sLL5XDV2Nubiurp6ebD+fz+vo6GhgIfQWddJkd3d30CdbpGTNZRYidSfvpkbtEy98XBJ0NE0QWHbBObORZ+CPvb29gcuN/pOqn+Mh1gfwEJfCPep9Zefn5402BwcHzYW4XC5rMpk00DJP02d4z1mPHPbrd5p57M4uxN0MXQ0orAeKDyXI66xDK2G5lw/+S8UqrW676h03pc8eg3mTOny9p3RS7PZPwE5FtmdFupheznjM0pOFd/2eMutly2sBXKlJW5vq3WtGsgvB91XdFvpOYKgaMk0G8H2fNTELKsfILKDNVNbKvAApPWDeJNyzWDN0XQa0bCuz/exm2mSlZpumtwGF8V5cXLSjirBY8simTdYQ13x6CSWBx1op+72sXFhIOT6YdLCQ5RkLB4pBlfvStWnBbO3Yz5u+FmaphKA8YWk4sQWa5qky5gl/BmAy3RvBnnFIrsNLST9ojnXgJCF7UXKc/NkT0LOOKf4NcO0ptfkMNMzELfcnlaEErGzfz3nLSm8u0/vhe9zP3vVUgjaNL122WZdp/TKAk3Qzn7+KtVV1z4HLxcLiLoYwIDjjJ9N+uccJA36Fe9VwYghc4q6ytsVCsSuJe6tu3jKMRmtBaPeLLb4EG36vGp5EnlYJ9SJMqD9jCwiiXvvuG4KRvlgQc83C3EyLVWRasVBOT08baGFhMj89QZG0sGBKiyk1T9o+OTmp6XTa2koL25p+ukh7IImik27MTFbxs3bzePP6ppiM+8S9zmwEmKbTabOiAXbfZ16jD3gCsLLsMvY82nrE6qFP0ABegjbef+b5cQwXPjQ9uc/nFZIMxX38tzKAxeR1mFa8+dXKSw88LD9S7vQEs9cE8+IxWNExX9IP/7ap7gSsu8Dca8b1mp88V716e/3IunvAfZdi+7zl3gMXhKU4qykJaFBar282lfrE8jyc1bEsu1GSSRwX4/fVatVcgdbeMt7j7CTqxF2CIHE2UrolegKE3xKwWBD0ldhapuaaAR1XY0xJ4x5YGfwMANauDZ643ZhT/hCE29vbzV3E+PJFjvSLkzuqhsF7viM8uEYd5+fndXX19AWRPOdYAPNA3WS6WUEAaOElC+9N88KYfYiyTx8x/awo4a7tAZjT55fLZeNFXFZuF4Dza1oMLOv1evCqdvM29ZkugCUABQ/M5/M6OzsbgI63eDC28Xhc+/v7A973enS6uMeOEkhhTduiz7XLPJjWftkjz/odYF7TCeQGeObX6wHldNNBtU6CMWh5bSdAe2yUVHwTZHsKH9/dZp5P2PP88NwmsEqg/Pve4jIBPPG9VGALgHRt2KWRgnuTcK667RJM15mfc/2uZzweD/a1mDnTVeh28rQNF2uLBmszLPdV3X5PEG1nQDYXvOunvtRWvXCcWWeBntaktTeEp+Np0D01VltEPdecAc4LlGuMwc9SFwBGOv1oNGrxnFyYpgXzxtgNtAgxJ4t4rO5v8rj52vEvzzMAZP6zsDNYsA5oFw+A+dwWG/W7/cxApD1itY6TUQxOaR34pAnTDIUBmnhc5mVoDl34nOvHa43fiefSJ8fhmN8e71hOWOFLN6j5L12pvbpcXypLzFGO3/W5zvzea4vfN7kK03Jyv72ON7k7XwW87jVwmelMhASHvB/B72IXVGoUKYy8YGxBGJg29TfdJrZqUvjkQqTeXhvJnO6bYwXpe08m5zcWVwKXtUi7PhMUPVZ/h/appVInffQYLNy530kP3ItggEYOVm/qozVVu6XSerEWbQsiz78z3wHMzmSDhnaR5fgM3O6HaeU+mRcsPOwC7Gn4/MZ8mL700fNkS96xX8aaAp7PWF0+qd5xVPpmgQYdAF3apc/EP70mDA7mf48p6zcfpDKZ7kTLDV9P+ePv6RExQPp5t0f7ButcI8kPqST6Ga8nX7fSZvmWxXNredtTZD3+fCb7/yrlXgOXBZGJ6fhDz50F42faO0zpjaieVNdl4etXXvh9RFXV3Is5qQg0Z4IRsD07O2vPk3nHorSgrhqCpReIMyLzfDQWvNO/p9Npo5mPEfLr202D0Wg0sCCyb7hncuzW/k1zCwB+QxOnD7Qxn89rOp22kzYoBuW7LDH320LWdRHP9AK1kMnFDN0d+8AV66w+p7WPRqNBPCgBzXUaONbrdTvsFKFka8BrYHd3t7kN4ZXZbNbcfikgLWhxQ9Iv+J7TI+A7uxOdkEG7vT1nBsHkMRQS+shveCQYdyog9D+teK/HXNc8S+o/IAo92DtmHqXfVnSRNdCI+bcilVmF9jy4TernN0quJdPTMb5NIGGaeF30LC2K6ZXXE+j9Hfp4TUKHu0Dyecu9B66qoYWQprM13WRiuwUstDKbKRm3py2wwZS+eMH5HjMre2xs9STj9jKvvEC5198BCBjKQXYDnWM97LkZj29OEjez9Swj2uYe983KgJ+1NpyZf6lBokDwmQVKogPva7Ill5qrYyj0zwI+QdTf+d3xNltFgIoFpa1h+mzLhu/c4yQF3I6mLzzteBe8xlh8TiW0g2YeMy48W1Pcb2HqhCTqsrLFPrukO204zkzMrKoGxwYxR+PxuA4ODgbCkRiYY4yp2KAg2MqFx62IACoGNFtjvZLKjukKX5GlybituKVg9n/6kSfs21rKNZGKd1pzVhr8OzxqEE1LKK0099UKTdKH/+Zv5tV0op+blIiXLa8FcPn7Ju2hBzY5Ab16vYB7bjXutXty00T799R0ErC8ENOtaGbfNI5N7hL6YWHLnzVlhD0A6Ha8+N2uAcMMn0Iwr/fmJzW4dNGgqdvaWK+HsRe76Ey33nhy3j1X/jOg9+o04JlGFv4WIAhob/DMOXU7FAvnnA+DqOlnoWJtm7m31WCA3cQ77l8qiAkMFm7uY1W1pBD4zfRIwZjKIG1bULoNf7bCkr/nMx6Xx9BTSHsuuk285rXi66Z5T/64+HffbxqlLMx6cy33yiZZuune3nre1O6rlnsNXF6ACClrtdyTQtDulaqb7Da7WwxI1FNVA8FIYWHZjWFh6v7m0T1pATnwvF7fvK+JOnOB2DrAPWiwubq6am47vqcWxb2A1fn5eT18+LC5MOfz+eB+NH4y+jJ+ghVgTdsgYl+/hZ0Xm60kvxEaOjFuB9jZSMu8UTevNTHPeNxcPzs76x5tZDDy/KRFmaBmPjCvpdvM2Xt2Ja5Wq8Gp6rawsbiYz4wreT3wPy13rLzDw8M2b05aSvfj6elpGz/eAngflzOnn3BgMe7w9fqpO86Wnel7eXlZp6en9emnnw7iZ066INMTuiQYbm09fSllvu7Ep/OnizeVCfprRYR1SYZrVTU3u+fD68nAlUqolQbWshUs8xHPGmwTCKyAwKMJWlZUuWYvBP9N06SvafYsBQC6QEPX77pettxr4Kq6bVbz3+aqJwIBS+p1byI8gda8exaVNTlcJV4QCIwELRZE1Q2jGixhcgfY00Kw5mvB1NPmkqFZpH4dvIXJcrmsra2teu+992q5XLYDXfH5p+Ax+CBY7Daiv3mIaoKW67MgdkzHwOL3ZzlYD0C4r55TaGIg9T3uFwUBxucUgNwPAHg8eQyPlR2uO4sveZl60xIcjYZbOFBuDP7eOMw82GWFAgEtfOJGClxAyy9BNHBNp9N67733BoqH+S8F4fX1dZ2entajR4+agkFmLXTkeZ/IzhrxHEFrz7eFuZWFnkVtGeE1z3O491Jouw+07w3SjLOXPdnjj6wv1wf9RLaZz6wU8pl5TJdgjt2/Vd0+wivLJivLdMnCWHJtvWi518CVRK+6je6biLMJ+XuunpxYg4U11LtA0NpL/maBmX5rf+6Nb1N91sKeRaukpy2ZnZ2d2tvbayAEUKRvn+fpPyA1mUxunYVnoW0BYgHqRYPATkFliwcQ4zeDjEG76vaho9DwLuCyoLIQ8nwz5nQvZUYkfYEGri+FgWNHjp/53nTRue0UPgneSQPHjdx/Enaw0jiCCmDDurq+vm6vkLEFbPo57sTeNeqzezEtBFtCXp+9+n1f737GBY2Tf9OVn/W59O7LcTpzMmVBKm8uPfmT3/P+vN573uPr/eX4/fyrWEru49+3FlcyS9Vwf0uvsCj9uxcxn/nvDCEz/mq1qtlsNghkZ5+sJfY0jNT0cBlZmBtEESrWftGO079tAOktKi8WAMOuOQLkq9Wq3nrrrZZ1iIuI+87Pz9v5dD75w6eM84p2MgGhHxtNEWDWZr3Qt7e321umHXdjk6tdrbzmw+f8ZeCYpBRbylxnLgw0vqeq2pmKdlPyt7e31+rA6kDoQ5PsG3X5vXIGpe3t7ZrNZrcEr+ec8XgfYJ52Aa/6ZBXowj37+/ttrMzlxcVFzefz9qznpurGHV31NLHi008/revr6zbXtoQAJvjr/Py8Tk9Pb2Wbei3l2jTPJk97nquGJ3c4ocN8BO2ggcMDKHL8xnq6yxrBAk3FxMBreUIfvN43KRaphPQ+5/q2N8Xj4l7zkoHLSunzlk1gZMD7PMDvXgOXNVWb9Ga4jFmZUVLrqboNho4JGSAs6FkU3O9DSw1MqfX7O/8nk0k76seLzQvRriC3bU2Z5x1j4Lv7zDuX/GZiZ7ddXFzUZ599Vnt7ezWbzZrVhABAOCHgoM/e3l4dHBzUwcFBHR4eDtxjCA33264pC1n6PJvNWh0sqOvrp6ck4L6yK4kMTuoHrDhai3ZIe2du7UKzIKXutJa8EZaxIYSI8yDU2eRLu+afFMwpvIgf2Q1li9JKisdhC5b/WC8UWykWatfX13VycjKIvSJwDw8Pm4Iwm83afGxvb9d8Pq/xeFzz+bz29/fb/FqLp25OyjfIc//W1lZ7eSOvNfE6SXczY+ldS9CjTV/DsuX0EsuABDDGYivYChDKCkqJ1y398/xbdhGX61nPlhcJVvnf/ff1BCq3kbzXU7Zt4bstK1wuvd83GRbPW+41cFE2AdBdZnVPczED96y5/Nwzsc2E7o/bsnDZ5AZIoE1tJa9nMdjZgrObyGPKReDYAtYAm35tCXA/MS36d3h42IALYHQChy1fW5R5ykNP+OCS4nimXvzC/We+EQoGwFxAKDto3amZ9ha1aejNyfBSauYGX+Ywtf2qaq5ZYpHEstwHa9M995b7huC38ITmfK4abpQ1qJrnHK/sCcaLi4t69OjRrWO18CrM5/NGK4Mr8wtvOEbF/Bikc42b9xmT3f/0MeOFaaGkYO61lfElnnNM0XHAHs9wLWVBtuH57IGD68/kjHRdm+coPWvOylGWnrL/vOXve4vLpSfMc9Jt6nvxU5h0CyZbbP6fzGAmgVEtfM0YLHaDVG9x9BYe/bZ1xb3UZcGdY85XiifwMgYLbm9IzlekAEZYFtDmwYMHtbe3V9PpdHDMD3Shz2Zix28Migb59XrdzqIbj8eDc+/SdWTtmO/L5bKNy5Y5NOFVJM4etfvWQrxHK7s3zR8uBlsnQOBOpMxms/Z+LKxcnrc7izpy86/LarWqxWLR5o2+j0ajwZsNXA91+x4La3iht0n98vKyPv7445rNZo0vLi8va7FY1HK5rJOTkwEI8ocSxPzY5eb1YLDGkoU/THsrP15r9B1esUvd4zZ/WNGxNUM/qT+tctOsB0IGZv9mPt2kwCaPwUuuO12fjNfucNdj0Ergehbg9IDZ1z8P0Kq658AFgR2vcIDXxLL7qWq4D4aSmpMFqzVTu2R4jgmm/UxxNwNhrZApx28pTJPREQjUl64OL2aeMSNaiCdtDKTQx6nUPHN6etreseR9VIyJ1O0HDx7U5eVlPXnypE5OTrpvAM60eR8c67chO6jNfSwsXIxeEO5TCgfox/xybblc1nQ6bafSP3jwoMXodnd3az6fN1eS6xmPx4N3l3lzdG78pm+8y8pWBfPrk1b88snFYjFQqNymY3O42kicyTcOeF2MRjfvvmI+fcSSwY9Udvq6s7MzON/QQtFW1y//8i+3caI0uA9Vw1d4MB4LV8fvmG+vWb/B2lauY7deI/TNbnNny6KQ5cZuK6XmN/jIrkCDMPxrb0eua+iQCvL29vYgJmq6pbuXdc/c02fzh+eHubN88Fr5vEAmafV5lHsPXBRrI/yWrhNPpgW9NbhNJrAFPX+p3VO3Ywm5QN239FlTlxnS4GQLLsfgkuCVzMp198nAaWbOglAwiNNngz7WB+68TGJJP3nSir71LMK0DvNZa/Ieo8fg531MF9l7FxcXLSGChdyzen36QWr/tkR41tZfzo8tP4O692pxL/zD+FKYOr5iLbtHK9P15OSkgX4eV4aL1goQ8UcOiU4lzO8US0GbnwH+BDHusTJlyz23c9g6xjqFZqZ1ygvTqremzLfQmGe97lNhTL7sKZA9pdnjSWU6FbVcI24n7/V4cv57db1I2QRML1PXXeVeA1fV5r0DZhx/TuRPwZFCHgZIEEsGMIP2NDIvnJ4/u/c9NR6EgoErx813C2ZbOUkzC2LuuStt124pW2L25fMcsbG7FotjGz2B6jnrzVtav77P2q0VCy9UhO9odLNRGdDYRKsETYrjVghdx7g876kwUB/jMkhAC1uRrst7hkw3094xnXR5mT8Xi0VzlWLdQINUbrjmd0tZwFpwpqKW8Uv+2wpK128qO6aZrTYLd2/g91jNH8wZ8Tfvi0vgSZ5KPk6ltQdcPX5KME5gTdmWJfuz6ZrXs+v2M5vaMO3yeq8/KUvz/k0g9zzlXgNXLqJNMatMNV2v123jZVpKPFd14wIwUOQk2w0Io/vkCltzVTfmvQV2CiJrugi/XCCOA1njo8/p20dbdd8ynsf/TMnOM+96DGdXZE+w8TyuNNxYCBVn2bn0NE5ccL2z/XoART1+5xTFrkoAbDwet7RwEiMODg5qPp/X8fFxG59dQAaDnuUJ6HtOuZbJEdDHewRxXUFT2plMbjZnp3CzoMnTR9xf4pCc2mH3IXRgTD6gmczNxWLR4oPL5bJZnfTP/GC+MhB43hhLuoB9X1oRXsvT6bTNv+M7GRtL3mCDPfNOHc4qdMyNa3ZdGqxYC17/fJ9MJrVYLLpKrO/1+k9lNpUDJ2X0PD49RdhrNmlsfroLnJ5VesD4KqBVdc+Bq6eFbbJcqoZv9uV5/jIOxbMW1D0GQzgheG0R2Sdu5k3AtdZfNcwUSzBObZbfXfcmGvQ0/mRwC1ALD+5j8XLddLNrjHRy7kcAGISdgs4zaWVUDbVPKxi4ldCUfagttOtp+JlgYrpSfw+gt7e3a29vr72Useom0cKg0PszzfN4IIN5ZupZ4bI7iv8kXHi8jMXz6Pn1/bZQcEdyrw/FpZ+Or/KsXZipLPkzY2Ae4U0UEWjo47ko5vMEvBybLSz3P+lg12m+qoZxGDR6rjfqcbYqB/CmJyGt5QRzLGfzBvEqr9FNa9zzmDRx//1M78/tpyLkeTOvuCTY9VzUvedepNxr4Krqp4tnyUlJBq4aahkGi2eZu5RexpHvZdFusg6yv88zbgvLu0qPKbmeC8V1Jxiy2GwZeAy2LGxZerG6DRZ4ClovwqRbD9DQ6i2Ec84S/BJEDAzJAxYyWK5u0/1O4Ostfi/4tBpc0q2TPG7r3gqNv1vQJICanp6f1Pg97qobBZBioEDQJpCmguk5cPF8J43SArAg9zz7ea/FbDOVWu5HqXS/7wKN7IO/5/z3lEhKunvpc45lE9DcVXrP9Prck5UuKRvy+VRe7rrvVcq9Bq4UvL7Of+65urpqadTJQLZc0uXi4vPn0n3nuqxdZlp5xilgBNx2POPUZywLC3KAgnbTvecF14tpUIiDENPASuzRk8/Uj+uI+ljsJAb4iCjmic2qtIVW7PcTpYDzJmracXH8xgDFviWfm5cg6o3NaXGZd/xHppfHbwsWUDavjcfjtpHWNLViZJ7xVgPXi1JgQev0cWvVyUe8RwuXIDTz9/RKcGr7wcHBgN/N83Z77u3tDehhIWiryevNG4uTP7Md7qHY4oFX0hXqOfGaIAPTtKLfjtdSB1a9x2VPBM/a9Uv/aC9BNsHKPLxerweKg+9J0DLYMY4EqR5oWcnJ+qy4UHrglO37mumf5XlBt1fuPXD10NvuKy/CdNfZlbPJish2eN5xirzXTGPNfbVatXP7qm4WHf208KEtB8iJLVjAWdhaQ7cg5hqgkqDNvbl3ifu8LwZBktllXqjES+jXxcXFYD+Sj2/ixAuEAvOWtLN7LwsasueQYg3d6eFcOzs7azSYTIanU2Rsgva9DQCXGYkZbAmYzWaNdqSSO24C/VBYUBwct0te7cUtbOV4jiwU4VcAAsAlTufDiB1ftRJDer/nw+sMPvBLKq1Q0AfolHFnrx+XjO9Y6WOMPuYqlRq7IT0H0Jw27aZzv3ALp7WzKR5rhcDX+Gx5kTFN32N6+EQaeNG8gOXt9gxIuQ7cX+qyXOitn+cpeW/KzFe1slzuNXC5bBJqm+4xM1lA8VvPRcD9VTebFK1N9sCK3/jvmEcCl33e+SzFcQNvosz+cS0Z1cxqkLW12CvpoqGvtla57vPZ0sIBQJbLZR0fH9fZ2Vm7329NZnzpVqMea8OpHNhSygXvPS09QOjFV6pq4JJDSPeUB67RJxJCPP5UKrjfvGFe9b2ej7sEDL8hkHJPkvnYiprH5HYyxslv2a4t6UwSskBPuuf43Ef31d8taHsCmrEwRpRN091C37Qz/TzOVC57NO9d3ySbDD7ue85vD/x69Eq+yWu+nvKtJx97/b0LgHq/9ayxVy33GriSyDBACkCKBbPN+bRKqoYZUW4vrTbatmbH8xZeLFp/zjRdWxXppuD/dDpti5A20ZiwmHppyQYDB5J5Hq0bGtGvvGYhmHEOCyIAgn5xfttsNquzs7N69OhR/dIv/dJA09vf32+WR1XV0dFRy3ijPtxdPicQgDs7O2sWFRaGNXxr2liFJJEAuixMAzjuNSs5/pxWVFUNNgK77hSgvF6He0hYcV9Go1G7z3xKXall2xJCKUj3VW8t0XfvZQN0Teutra1aLBaNzy3caQP6+j1i5hnqR7kwLbnXYDYa3WRj+rt5ezy+OYYsXe7r9dMjrywbRqNRs/h5Bp71+mPDOO8DM62stLivSVf6aXnibQyWLVbWDBSpyFiBdOKM+w7/9gAt5Zifex4Qzt+eBaj5+VXKvQauqtvZQVU3bjxrf1XD9HVP1Hq9HjCv3REmdGbBcX8mI2SWGczI/hD6YvdgatzeDMnvCJHxeNxcb9SVWnFq3LRrK8SH0u7t7d2K9bhvPs2Ad2zZPTEajQabR3FLQT+fTP53/+7frY8//rgJBMr5+XlLP3/rrbcGrsTUHhGCl5eX7dw7AG61WrWFjFsMoEOgU5y67VMKoK1/M53NK06NRsDjcmN8KBlVNeCDPMHB4JZxQOqy8MsMULtwcUMaJKmvd5I/f6YbY9jf32/rB56EX6w08RwKgXnQ6we+hW88hhRs0B3g4A0FDgNAf+piLHaZGhytmPo9W9DYJ9SwFohRe53Tdrrp/ec4oy1NK86paJr3nFnoZzfRyaCU/cm9lozNoJPeBhev1x4A3WWNvcq9We41cPW0CxcLfU9U1gFz+rlnaRbW2lIjMliaIZOxUlP1MzC7n0+myefNhD3wTGbPvrluxpFjMi0yLmjBQJ2OXaEgLJfL9v4l98UWCkICAezYoBciQqnnGvTYfF+O24BAnzJl3X3xHJpuVnqs8dt1ZkHmZy30e3Nr3nX/k/eyTvcFIKy6EaKbUtmzDsZqoAH4oBcgnpmXyUtW+tx/t2G+Sr60UIY2jA9w5HvPpef5op/EIb0/0qCHnOm5mrnXxTxqWtqq41reZxqkfMh59rPZVtZLucuy6tX7PKU3f7+a5bUBrh7ImPHsyspFivZOffzmtF+7MKrqlrZn0HGWl/vlvqJFE+B3Wa1WTTtmc2dqw9yHYLTLk9+4b5N7KN2LOW7TyaDCnwHJz2RswRYx78vy/hpbB7jyCPDjttrb2xskBCRQWVP1ArJ1YAvBykPOcdXNqyAALSsStGFLlO+mgV/+mECZVhv0MSiY/qavecrfrU2nQjObzQb87teHmOddd9JwvV7XdDqt2WzWXg9DPfkaGuYTgPTWh5673col8wFf8Az95F7mC14kixVa0WYKYytatka5h2u4nJljPpuvne1n+ZBWF+uQz7Y6e8AF7VNBTjmSciaBrCfzLOdSmd7Ud/OOnzHPuP+em3zO971suffAlVp1j9hcM5OZ6HYJZDA560/ts2p4mrQ1TVxwFpAWJhak1tRJ0QbA0jJxHQYK37O7u1ur1aq928sCm8/pKsU95lTiqmqL2hYBn/f29m6lKa9Wq3Yqhul5eXlZn376aXPPeAHs7u62E9ARkAiO2Ww2SG3HXXR6ejqIIW3aqMl3aG7rKYuFmelrwEoXLyCcp5Izr6473bG5jwzaXV5eNnpg5UBHNmqbd0i3pm3GwDvU1uubA3hns1mdnp62jELqtLLBOG2RufhNAIAYFovXSAJaKpkGItYo82JrxrTjGqDIemEOfLqFM3a9znB1UwfjPz8/bwca2+qGF3M+q262rNBvXkHDukpLKDet25o2rXtgk8qJ1535PkME3Gfe496cE7tBs/QMhPw95aPr6o3tZcq9Bi5PACU1kdQsUlugpDZnoHKxtpgairV47kVopFabrgIvai9UhI21Oj/nPnssCBwWX4KfM75yAWVbXgTEahB6uPS2traaoHW9CFRervno0aMWa/J4sTz9PHMLMNIPDu6F1ukqtAbf4wtbUvb3O/ZkIdcTtjyXJ54guD2enjKVe4rcb/6oCzecLUfGlFY8dEMYm+eg68XFRfujDb8k0gqcAduxGqwbx1/oK/0AgBP8XbLv8C5jS/ew1515xMBrmlpBIrEH+pO8YwvKCU/QjLUMX9hFapC2kpbyx2uvx0fwMnQ2b1heWT6l4p5gYLnjMVGyr5YD/r03Tq716tsESD2Afdlyr4Gr6rZWYqLYBZXAlSWFtifT7Xix5MQ7lmFtEYHaiwlRP+4VW1RVN0LCTGJBkNYbzJ7CucfQvc2UPU3IQpVFgNWDkMS1mZtsk5a8UdfzRD9x7yC4TEPGRYwMeiJ8/efF5TF77j3XpqPpk/Pb4zcrJ37ViYU497pgEXm/nOcXQEghYXB0v6zEuH3mD37DCndGn+c0ec28agvD42YM6a72sUe+biFI4Xf3EzoZrFNjpx5AlPugk9vgnEpow1YMrxloZmuM/jFfWF9eL/AjCocBMfdoJgDz3ZZu8maO3aVHE9fd45V8NpWhnKven+u+a257fX4V0Kp6DYCLYqFtoeWJSY3SMRUmiywvTz7PjEajgVsIIeX9VAYo+lV1cwJEarHuW2raqdFbq3P/WAzeXGtBloyLYExrwPEYNFXvnF+v1+3FjVhcCG9oi+vQYPTgwYMajUbNUvLYaZcNu7ZWuceAZG2XeTw/P6+zs7OBQPApBWj/CAXGZqGwWq3q008/rb29vfY251RmLLxMRwszgJf7DMTUk6nlgJRfkQKfwAvX19fNal0sFk0wprLjTMSTk5OW/g39Li4uaj6fDyyTdDPnu6mc+YYwn06nrY+j0c05hsvlsr2Mk0N3sczTleskCK7bVZjrmKxPxmzXsi0TrEH6RP89h6xl2krXGOtuPp8PElCoC0UTusNv3JsWOS5g/2Zl0++Jg8YW7r4/Ldesq6d4WqnhWipjqVD6nmyP59LSow/pFk4DwNdeptx74MpJSt9tVQ0WeFonXow8ZzcA//mcVhP1ZDyM+9kfBWNYC0YQ8d1j4nkLTrv37NpyH6puAvssMAtE6rFQTvcnp0Hw3eBloZLxN+91Ih0cYblYLGo+nw/S/N1HxxNSwdje3q6Li4u6vLxsp3cjQPLVHYCj++/5GY/HA/dm+t9NO/MIcSXiGaYj85QxIdPRNMYqtaYObXHZoam7/7YK7Ba1QDYf2uXEdgAsCSszVtScbML4nI3IXNmrwEslrcSYj6HN/v5+W4t2Fxt4nQQDbaBTT/GiD944nrE6u95TEUxvhOdqNBrV/v7+YG9g0q1Xh9e/58Z7KJE78K6VX8CbvtudaPAyT1s58fpKQLnLi2QFFDokgKa8sUJmHvecJpi6jpctrwVwJeNRLKy9yLNscpH16vNk+PkELq5ndlj22QvhrnHkeCxcLTwROG4nNR//pbCzwKu62buWY023alU1q4HgNvXxqvbT09NbQo2TJdKdy3i8oGzl9cbWo3EKEBa/XVcU0zOVG7fnPpkPANpUOvzZQoK++LpdXKmZWhAnL+QfbZoP88WSPcHmdn3aBgk68EcqWv4tz7fjOuNDMKcFZhr01kIK0N5Y83rvvhxzFj/r/XwGdX5PmqU10rv2IgI7Zc1d9HjW/Xc90+Mf35N1ZDuuq6p/PuGz5OuLlHsNXD2mwSrh813EsssQYWZfud0IaJiZQZfClussSm/E9CkJbHqk/35NhLVZXJeOKfiz3YwwU7oLs28ICD4nA/vzaDRqWWN2ReIKs1BFaHEdWlZVzefzwaveqZvTMnrzZO1+sVi058kwJNhusN3b22sWBjS9urqq5XI5AI31et2sYcZrIcWpGtTL5ui9vb2BZZnzD//ZKplMbs5A9PgZI9aH36+G8Lc7EaB0nxlPWiC2fgB8khHQ/rkXlxlWhONeuP3gg93d3eaCc3yI37DqHGuqqpbFh7Kyv78/OKcyXbI867rMk94KQWyOvzxEmD6SOdtLPNnk8eiBPHSHH1OJsoLheJf7jMeC67ybC96zBXkXcJn3Mtmoqlo/N8kpj3GTe9JjvktW9Pq36bfsw4uW1wK4ENwWomlG222CkE13khnc/mBKEtpCwpqGtXKfg5ZavFOcHfBmkfdcEM7s8hgt7DiBnTarbtxQ1JM0TCBLpaA3fp+E4KQEnjfYOUkG2iDMrXDY9eN6bIUA2svlcuBisaVp0ERYpoIAfZgXa9QASi813kLAxxBBa/qJME+Xqvej+T+xJcZt/r2+vh5Ypq7Dc2kBDGB5/xH0AzRw43o8Vqam02nt7u7WYrGot956q60dn/YBn0MLH0s2n89bf1HCnDC0vb1dy+Vy0D/6klmvji0CNsyVlRHzKSDNfb3tB9fXw9e0wKt2j/Kdvub5hiicOSfwV8aLAP5Nm+bNS9SZ7k/qsZLu9WzAdF1cp07T19ezzykPPP92EUN396U3xhexPLPca+Cq6qfJ3nWdkqBF8WT13EmUtGSybWv2LA5bUr6H7/k/XZgwiZmNa5nllNqQrbkeXXoau7VOt891+7hTWFjjtoXRsxKqhlZYarzpejJ4I5QAs+yv+2Qh5nnkmXzO1/ybrxksHZPJ+jx+P98TEp5bxkasbzS62TxsALUQcv8RkHbLGdD85uOqunXKOHNI+7SxKbmJ+YHXTE/zYtVNAoSTZdLCSEFumpgPrYj0FM5sMwHOJenK8z6ZI5OWsi7Pf/JPDzR6fbjrew9AetaZ5Qj3eWy9+lMGbRpjlrt+T/n4quXeA1fV7RiVF66trBRMtnY2nQqeApd6AYuqm8WQAtGLqeqpUCBpwXVa0KfQoqSbMrVyM5nrNYASHM76qmpwGGoKTbt4vEDsYrU15eA4AtMWSwpjhGnGH/jvA2JxX+E+tAaK6xBhnVofrxjJhA6ns9tNhXBm07N5gOKTFcbjYbYdwACdod9yuWxj4z7AgDjU1dVVm5Otra06PDxs7dm6pd/Q2wkhBgNr37a8U8PuKTo88+mnn9aTJ09aNh+bfTMRZ39/v6pu3KYGu+R95pX1kdmg5jcnvqQl3BPMuY5Ho9FgM715ILMKLTeo23sVR6PhRmJ7eZwh6nWRSkZv7bI2vZaTFpYrPaUUl31aTIzBFpeLvVa0lRYX9/WUdf73FLbeOF+l3GvgSnBIzaL3uzdzch3GtTBnEfbqMxDwm/tkwe30ZL/HiTbt86Ye4jEJltaAr6+vW0wm63D8gjqJszAm+kjfaBPfO/X4cFgvUD5b6AP+AI0BMLVz2qEYBE0LrAYWf+4zAli2trba8Vh27VRVE+a28AxouCKtVKRVdXBw0J7d3d1tB70645F+Zgo8QoW6UR56gA3tOFaJ1G7eczUaPY0LLhaLwaHCGUcx3VarVXMH+jdACRdlZszCY+4nCttHH33UnnnnnXeaa3Zra6vm83nb4nB6enorvR3XnJW02Ww2OOGCseBO9nqtuhHAVTWIH1Icu95kpTHf3r/G+M1jjhlzDR7Lw539vJ+1sgltLQNsVV5eXnb3AKYl6nZyXdp6tRXq62ll0Rf/3rOOLDdyLblYZtqL9TzW27PKvQauqs1+V3/uWWJmaANeMrzbcR29yUtG6/XVLpueay+1sWRImCMBracBJWj7Xsbq9tOSS1eBtTT3I5k2x5N1cY/dnhYUtu5YTLbQci9Yb4FlcNtgboHRW3BZENBWBuzudCEdH9A3TU3HFA7mRcbpk0SsANlKwUozH1uQAACOedoK8wZo87T7lgJutVoNLMmzs7OmPGQSCn1gXKafhbqteSzStABSq6ce87F5zu5B7nHslG0FtNUDG35L3ufZXDNu3/3s0TH/fJ9lQJYev2fdLqadx7WpZD+e97nePT1Z+Cw5+TzlXgPXJo0g70mhbGuhauge5H4EVS6Yngadz9p9xnXaoW27lxBMLPB0cVljqbqx6tLy4DcDHH1FSFm44aarGm4+TnpYS1yvb1w2CfimHe0yBmv/9Ncumlxc1r7X63XLGMN1ST8MBK4DdyJ1WDBa+6Rt6J9ChN88n563qmqnMZD0wDN38SbuvXSfQS9AAODijznMPWNct3WCdeR9fOYDJxlsUqQssNNi8CZd4j/T6bS5S/0WbHsfDEjOUjUg0J5T8j0/JKt4U7CBhzkCrMiE9KtdsJZpyzzv8do1aIUlldsE/Z4yeReAWXnJZCcreD0lNBWjTYDosbEurZyaFl5XbsfrpweszwNerwJaVfccuBzL6DE3izMXf2rqvcQE6vJCzgQC38titAYHY7hvBiovQoOUmdZMgrBwXMbabM8C5Pp6vW5C1dbOJg2bzdHX109PbCD2QrtJoxTUTu22i87/sSh4fj6fD+iDO+7o6KgBy8nJSXN7eZE7M8yCnLHY3bNYLAa0NQ1MQ4SbtzSMx0/fTcVZfycnJzWdTmtra2tw6G8mlFTdgJU1ePMLGXk7Ozs1nU5rb2+v/ZZWmrcJ0Genp19dXdXJyUmbN7881FY/c9Fz8zhelmDgOI6/QxO2DrAOSKunb6Yp9Vc9VVgePnxYi8WiuRlT2NvitGKSXgh41xu+LS/efffd2tp6+iYCUtDzhYxVdSuWTd3punbGnt1utp4Acza0UwB27jN4G5T4b7BJ4EuAch2WaaZpKquZHZkg7Huzn72SsvVVy70GrqrbVldqiGll2apIhnB9Pe0orRq34edtMbkkYJrZUnPLfrl+7rEG7AQRWwuuE5Cwxmgwdd+cAWmB29O+3B50zeOgDAimh8dqbTZfBMnc9hasLYxNbr/UHCmZBs09SSsrN3ZXWhFCGNk67fXZLqZ0g3qeoBFKh4+WMtB6HxZ0dzYhn+nDJo2c3w1QOU98T6Hsa4yRtnZ2dloMy/X24iQAC89wDSXU8wmtbEXl+ss2ve0EpQpFBB6wRehxm27QwSCaJS1Xz6f51vSGjl7bPeD2fRTf3+uLwSvr7PFBuvGhl4vlTMqcbDfLq4DYvQeuqrq1EHvFmlnVTcaTmdyTlkBXdTt2w7X8ntcoXgTJiHlfugM2aVKuI/uZv1ujc4zBbZqmXHcmosHK2mtqj3aXZhZcb9xpjXIuHXOUWVLMjwWD+9yzIg1IFAN4Pu/MNYOqrdtMEsjzInt8YEWD9hh3egV4vgdc1s6hj8/Dy3iq6eZ+WUi6TZdUCFg/ueYYO4B5fn7erHdcn+Yj2rflDIgzDvacuW8GFrJJczxuI4ERnh2Px23fI+CY3gP6S/8SOData/piuicgUVIW5Vh74JW06H02mFiO9ACup4Rv4uEeSCUN3O7nXe41cNk6qLohfApkTxjMSrYc2XBe4BYoZmJrRvYPuw0zdDKMmdMWjTXiqhpsnuV5ihe442T2N/u0jXRvIqhxWTgZhTGn69VuIuheNTy3Ebpb4+V5L5YEdS96XGS43jghA8GVWXR+XUpaMowdOqG9c7J8ghvtA2C8DwvakJ6OiwvBivvUY0OgO7kirWLGzVYMwJpM0RR0HFJrdx/15Env0Mrz5bEyV6PRqIEJ7aXVadqyTuCdtPxSMYHey+Wyjo+Pa7lc1v7+fu3t7Q22JTiWt7OzU2dnZy0rEV5iM3jyldc2tPd2BxSs1epp0gfJGWRlOsbmNcWz9NOgnsraaDRqVjj9SkXIQJpWCtd9n12FPS+Swd7z5j5aIetZ/j1AsQLTU5bNT/lc75rX+ecJYHenU71E+bf+rX9rQNzRaFTf8R3f0X5fLpf1L//L/3K98847dXBwUD/4gz9YH3744Uu1ZSKn66+npfSssU3ENHP4+Z6WBFMgKCn0yYxj4VN1A472KeeEW0OiDoS6+0KffSKHLYWdnZ12kgN+/zxNYL1eD6wQFpWtD55h8WaAHYFqK8KC2KcnoDHv7+/X4eFhHRwctGOgoCFAe35+3kBsNBo1Aeg2kq64nLBK7E7suUoR2IAJ9/p0BscyPDbThrZ7f3Y94uZar9eDlzEyz3n6yHo9fGOAgcZrImMtKZBSMcn9Yb7Pgpr201UNfbGYSLwxTywWizo5OaknT560w5Lt5mPe4F/GNpvN6vDw8JZccWyObQOj0ai9pZkTPJbLZXurAUBK3M/rlTWSLkzLAq/nVCgNDjkfaS2ldZUyq9e+2+lZZa7/LrDryUaXBCxfT3m5Cazust421fUi5VfF4vpH/9F/tP6H/+F/uGlEcYR/9V/9V+u/++/+u/qv/qv/qh48eFDf+MY36l/8F//F+l/+l//lhdvJye6Zv5smadP1XBzc23sm/d/WPrOtbNdtWehSh4VHjhNhhPaf2owtQAQyAiT92xnbcpt8NhiZKVPDzsXYc9khmBAyCHifg8d7kxgrQjHf7+R9bNY6ewuW35yN2RMABl0LRlvLFlI97ZprmRCUVk8KsHSxeq5S2fG4s+SaSNfgJr6yMO4JFvqQAimFuIWsrzv+xqZq+CDjSlaaABNex5LeAZQtxuGTOFB4AFT6BLBuEqZJ9wSZtEpy3D0a99pJvtgEdO5TyoheyT49C6xc8pn87a46ejJk033P05dN5VcFuLa2tupLX/rSretPnjyp/+Q/+U/qL/yFv1D/9D/9T1dV1Z/9s3+2vvM7v7P+8l/+y/Xbfttve6F2epqjFz6Mv14PXR0UGAR3oLU4C+w0t734Ult321XDVxTgTuhpJT3XYE6uA/65qCk8c3BwMMhcpHBILJscLSzpm/38CAUDgH37OfbsK7+ReLC1tdU21+7t7TWwwk1oDR2XHVlfuNucAp3vBzNtaNvPWMEwD3hOcVNB27QCbAERezIwp1Wdc5Px1dx4i8XnuU6w2LQWUqmANvZMXF/fpIUnuPh9dOad6+vr5h7FVei53t3dbbzg8w49BtL5r66eHip7fn5eR0dH9fbbb9fOzk67bsEP7Xd3d5u7MQEdN67jo/DOo0ePbsUq1+ubDFtcznYtw3umN+vEdXuds5l+PB7X6enpgPfOz8+r6saF7DXO3MO79mB4TXlOU6G2IsO9lic+WcV82FNczE9uo6fE94Czx6epELyKpUX5VQGuv/k3/2a9//77NZ1O62tf+1r96I/+aH35y1+un//5n6/Ly8v6+te/3u79ju/4jvryl79cP/MzP/PCwJVgwkK1EDGhq27vwYF58+gha3JmBBa63XFul/qou6oGfbKbpzehySDE4Oizs8N8Vh9124JZr5+++NGnAgCkfnFdClAvEAs2a9cIFO8fcsnUYbdPijL3YWVB3+vr6wZI9I0xjEajOjs7q7Ozs1tHF7ER10pH1U0cCkCwggKA213HeLyB2HuhHI/sxZF4lvZ8Ysl4PB6cwuEEFNqxaxb34aYjyWxtIIThN1uZCUK9dH36zh8uWdYXdOSYJws9wCrXnGnttTUajVq6PjTZ3t6uw8PD9vJJEjIYz1tvvVWPHj0arCW7Lenb1dVVnZ2dtZdA2tUJz1jhc3o+4AQdZ7PZxjCE12muj00WjufPYYasK+cl68u1mX1Iz0pawZ7XHE/PHen6XR/lecHIAPwq5XMHrq9+9av15/7cn6t/+B/+h+tXfuVX6kd+5Efqn/gn/on6P//P/7M++OCD2tnZqYcPHw6e+eIXv1gffPDBxjqJbVCOj4/b503E7QHWptJjDCbQAPesunp9MWPa5+9y10TaBeb7e247x5SSMXnOgt1MbAFsrcgg3hOaBgmPId0h9C03jNqVhjAFhOzWoT3e98U+qgQo0yLHlDGtTJV2vMgxH8DedfUEV9LXcTJndDqmaCsdrd9CKTXttMaz/YyVJp9ZafP1HIP5wN+xWp35aN5wHzz/vo/vzPHJyUnr02w2a0kZVviqnlo1xK0M+B4r4N97XYr7aWWrt/428XvvLy2VXn2bLJHe7wn+CY5ZegDZ+60Hcr1xbwLd7Gsq4K8KRi9SPnfg+oEf+IH2+R/7x/6x+upXv1rf/u3fXv/lf/lf1mw2e6k6f/RHf7R+5Ed+pPubF76TJSycEY6U1IxYJK6P+3IR2/LKoDLFcQsvYJ6jTQu5FDhuN0/sRvj5M4FlrnNYqYPk3qCbcTE0VYDSgO3fqHM0GrVML1tWPGPBayvLbkfAA2HHhmOAi/p4z9NyuaxPP/20zs7OWj/ou60OaJiASht2HcGTaOOcX4eLsqoG5zV6fLj27M5jTMwFfbHVS1tYDbm3CPeXwQ4hz7g9Hz1rpwfaPSUhFYPUwllDdhEzLixBsjGpg/ik58V98Zq4vr6ujz/+uK6vr+vo6Kh2dnaamxvXoa20w8PD2tnZqcePHw94mHqvr6/bm7YTXM0z8Djjd0Zsxit7cUjo7rqtaLiYB12H5ZSBgq0grpuSiornK9vmXq9FP5v3eU5yrnJsCVabPruN3jMvW37V0+EfPnxY/9A/9A/V3/pbf6v+mX/mn6mLi4t6/PjxwOr68MMPuzExyg//8A/XN7/5zfb9+Pi4vu3bvm2w0FhcVcMAuQluHzKZfCzEBBFr3lU3Apw2nCZvgVE1fC+TGc/gkRmIXmB2S1GstfPnEwFwP63XT1OQWZS4fOgTQpu0YbdtMDQIeczExapqcMqAgQLBSvt+GeHV1VUdHR21lPflcjkYmzVY6H18fFyPHz8eWGEIDiwACyXqoE5cXp5T7iVGZVcdzyJcfH9V3QIRQAm+Yg5w8fZikC4I3HT92sXnLRt+QSYH5AIsPp3CPD2bzVp9dgFWVdsqgCXjzeMoPd50TdsGFuaFDE7m3QkqXq9W3K6ururJkyctrf9bvuVbamdnp95+++36+OOPB8k5Ozs7tbe3NziuCfAka9CWOuO1i7+qBmvCAtnzate8AcZeEOjueuwJgH4GKivOXPea772XDZ6BL01L8zx983Xih/b89OKw5s1nWU95r71CCa5+5lXAyuVzT4fPcnp6Wn/7b//t+pZv+Zb67u/+7tre3q6f+qmfar//jb/xN+r/+//+v/ra1762sY7d3d06Ojoa/FUNtclNJRmSa5tQ/64Jy8nyX88nnFqKn+9dtyaW1g335XPU5czN3NyKy4p77f7r1ZcZdTm2TVqlF5sD5db+/SJHt2uXKtfOz89brIIEDQOlF6oTV3r84LZ8f95jfjLvGMitdTpl3nNpZcUAZJrbRbhJS7eQ4pleDMJWrOcepWA2m93aL0YKefJFehJSy6YdezLoI/OEhew/gClpD9Cen5/X6elpPXr0qM7Ozlrsi1eJeKyAl5NuqCO3Z/Tc856rHp8YqHJOkp9SQbZl0bOWvFa4nmsi+9ebg169/t4bb8+t2XvurjqeF9yyuM89GfIi5XO3uP61f+1fq9/1u35Xffu3f3v98i//cv2JP/EnajKZ1O/7fb+vHjx4UH/oD/2h+uY3v1lvv/12HR0d1R/9o3+0vva1r71wYgYFQWRB4QlJl0nVDQF7rgb/vkkbqbp95leP+fy7haDdml4MdjnRPx99ZCuCcVugsWBns9nAOvEbaakXYeoFyvOAC0kcjkvkXp+kO+1YUFpoe1Mpz1GPQWe1WtV8Pq8nT560IH5qqRn/MyA5gxRBmy4a5sLCuscHnhcD73p94+qzBtsT7gZDKw0Jap4TAxRJG2jx1qATJD3PtHFwcFDL5XKQFm4ri5ICMd3mFMdSse7gF/rjN4nDy4zNipatjpOTkzave3t79fDhw3bNHgIsO8YDaOHe9RzAG7kx2es610KuzR7A9dx08IgVqQwdULcTZGyR2l3psZhnewp7DzzN06lU5Xz7Wg/UXcwPzwKglIefR/ncgesXf/EX6/f9vt9Xn376aX3hC1+o3/7bf3v95b/8l+sLX/hCVVX9e//ev1fj8bh+8Ad/sM7Pz+v7vu/76j/8D//DV243LRQm1plz+Z9iLd7MwWLsWUlo2z1gIp5gVwv3WfsDiLzZ1sKCa1gvVTeuguvr68E+GOJNaNTEiyaTm9deZNwqQcPgY026ZyG5nryvqgbxzMlkMnAXkopNSjL9uL5+emjsfD6vzz77rGnpzE1aSggJ6OQTQyxMLEgMPrhR4RvTnrbIviTOBW3W6/VAw4cW9NfAw7xBGxIL1ut1c7XhcrTLyPElPqfA4ndbYAYG1gNxJ8+x3VxOxTc/4AbFLWi6YrGxB485gvfoC/3EhQ1P+mQRu+Q4qPejjz6qb/3Wb23jIZuRjMC9vb02T6enpy1D1GvElhouTOiEouasx9VqNcgwdL961k4qdNA9gQ8+z9hV768n3NMbkS7B7CP39EA577FiSul5JDzelAfPA0g5xv+/srj+8//8P7/z9+l0Wj/+4z9eP/7jP/65tWki5ATzOyU19R4BWYAWzFzn/gyo5j1mnJ6V1esnfbWGzrX0eWfw3wLPFpu1t9T4LKR61kbPfeHrPevLwo5rtrpyGwG0GI1G9eTJk1osFs21lL5699VWki0Tj7E3X/AAc+ykjl6SQlUN7qm6OUXDVqv75v7xXGrxKVCsMCWtc/w9PuV5+kYMyLyVggrwtbvVSR+Ow2aGH3QkMcTFcRgnM5mPfLQSz9C/i4uLWiwWtb293U7yn06nLfbDWOAlFJysxzxhgGZd0weA1PfmWPOz+56xrywph7Jf7m+6hg1+Wby2bQFusnB69fRk2IuUlJ09MOqBa9XtrR0vUu71WYUUE6OXncdn/lvgpmsxBQtn4WUx0xjENmk8Vbff1JpgwTULUDMhCRVonmYAJyDwOV1gCYZbW1u3XjHuRbZJw0SLz6QHNHafH2jQcPKCQYu+cRQQqcybFAMDb29+kydSABiUrERYEXAf082I9cR1eM6ZbDnunvCCd6wI9FLVNwkvW2PuqxM5nHFo4e2+89/uRwoWxybg8n/GY29DWuQGQ/OkBTkAfnZ2VovFom0Gz/lmLyB9x+OAdWnaQx/oRcKCeSCVpJQfBvzkNYP/puL13gNWrwUrYLkGN8mdVJDcLv3N6zkWSgJ1Put6e+Ps1fEqFlaWew9c9gnbpWKNuOpuv62Z0qCWAh93Efe4Xi9QMrvSInH2GvdSv5+3FlV1IzSwtKbTae3v7w9Oiri8vGwnUdgPfn193a4Z0Oi3QciuT8dt7Dqhj1XDo7yIo9jtglXodx2hRUOn+XxeZ2dndXJyMtD2vXgtzJ3NSbs9hcOg6TiYn6MuhLutQVsiy+Vy4ApzTCiFBOCNkrG9vd02sVoj9ykUFrLm5eQRaIv7MsEPF6j3MUEb86wVlQRVW22TyWTgYuQ/FgqKGEkXVTXgPys0gCVgc3BwMBib41VWTNj6sL29Xfv7++03b1pHkcNaM0AalMxX0Ia2rcgyTmhicLfM6Cmc6Xq1R2ATUGU9Ozs7g32r/G73cypfCbr+PYuVLWiVCkleo04r3tm/VLB6APZ5lXsPXDChCQqRE4SqhsHx7//+76/f+Bt/4+C6tderq6v6i3/xL9ZyuRwwDwvemvUmdxvX3AcL1tSGvGDc16qbo3UcL0laVNUAaOw+yAy4vb29WiwWA02femnX7kQDYrZpTRtBQ1u2NM/Ozlqfzs/PW+KFg+e5CBEKOSZbG9bo0mq1IOI+gByapOvPsdKqoZZrHkurnecB5kzOcGFO3G7yQ1po/g3ryvf4tfc5P4BxT5Byj08osRLgthmblR//htCzper4ZFrCgBnfUSocq7Kyxeezs7M2h6apT9I3DTweW1s+AcZ8XlWDeojF9epLZcn9yTmn9Nz4tJnyjPl9lhuzxytJCyt0llWuKxUjz4/Hn8/2wMvz/XmVew9cVUOfMUKp52KxMFiv1/Wd3/md9T3f8z11enp6a7LZZ/Q//U//0y0t221kDMJtPq9p7b753h7g9pIPKLaQ7ALk+RTGJHG4PffTQJ6A4n4i9P28wTX3QKHF+rT31PJ6tHP9AIeLQTUXYoKA3X5O/Xd7vb70tMjkgVRMUqFxWwhp+pOlBx4eqxN7bFkZ7K2F23XWEzApMM1rvm7LyP1JHukBl3kSHvN3HwNmZcV8tF7fvOolrV6DS08WWOmBd7Gmcq56tMjfPE85vwCEadxbaz3FJOvepEjk/7S2NvFQ77t5Jfn/LgB6Fih9nqBV9ZoAV04yZZNZ68k5PT2tv/JX/kp9/PHHLcC7Wq3qH/lH/pH6ru/6rvrSl75UV1dX9fjx41sanIUYFpiFBX3ge4KI+5NnGDpgj2bNq9yxFtDmsQ5xk7z33ns1n8+bNpsuCruy0tKgTxbuWGTuI7+xn8ZZeXbjJL0mk0mdnp621GVvLjXdMimGOEZqowa73DRMHJD4B3PhzdFkP67XNwevMlYseSebcK9dS1U1yLojixI6Onbk35IvPH6EKfShT7ZiTAPcdVgvVVX7+/vteX6nf+mORojPZrMmvHF1ZtIGbZJVuFgsBvMEvez29VzBg6lIEh/laCdcrMwDrzZZr2/eq8bvuNdQxEzTPIvSlh/jd4ahDwm2cuc1CSCaJ/hDcYS+qRB47nogNhrdxI8z5pnKca5Tyyh7W7i3p4RaNrh4bp6ldLtfzGUq1tzb+/yi5bUALpi3qgbuEyYjXUEOoj969Kj+6//6v66//tf/ejsDcTKZ1Pd+7/fW93//99fv+B2/o376p3+6fu7nfm6w4KnT2l5Pq08zu2qojVJ6GXoWYha8PGsXlZMeHKdAC7fGzaLzAvWpINDJB872xrher9vp7rgx03rxArm8vKzj4+OWQu42AGMLB7T0u6wEFrj7Zrelx+O5g094DqFNKjft+nPP7Ucb0IlnLOwca2Wc5k8DgkEagY1i4NR0eAHet7CATrY8qIe+oyzgruUe3LpbW1stZmla0Vf4hfZtPXn+rKDB5ykIHYu6uLiog4OD1kcrTF5bmXXqtt2mM0LhN5QFwIUjuhJUrZDyrLc79Arrjf72gAT65j4uA73H2rPyXbd52GvEvGA6OaxBScut5168y4vBd1+3Yub6Pw/r67UArtQgepqQvxtQLi4u6hd/8Rfrs88+q7Ozs8b4H374Yf2dv/N36p/9Z//Zur6+rg8++KC+9Vu/dVC/J3LTxPZ+67kM/FzVDcPC5I4pUEcyLnWkkPW9tub8/a5F0aMd1hzC0oKjNy7ei8Q7lTwGa4JOpMDC8iLiGdpMAOdZ08vuKAsOW4Xj8bi9moI+GJRSCKRmmXPQ4z/3ORUcz3u6Ta0kZWwi60/Fx/VbmJLAAXhxD8qDzyDE8jCdHXMy7a2o0Xdr42mFIegBG/NmL3EEHjGvMp/eYGylLzMi3b4tOlvFtqpyrtKC6s25P/t59ztl06ay6d4EgJ6F06vreYCjt/6rNltJuQZ8/6ZxvgqAvRbAZUGQAtNCx5pQEvO7v/u7B0fg7O7u1uPHj2u9Xtf/+//+v/U//8//c/2e3/N7bsWJaMf7VWDWFOrcy29YBX4262CDpy0qj9v7Ulj0l5eXtb+/X6PRqG3KdZyOviCgxuNxLRaLwdh8SG8K66qnwPDWW2+1xewU7ATm5XJZJycndXp6OjjDz3MEACFMsSxsdVU9jctxviGC1TT0uJKOjAst3jEUgMvZg6ZVAoKt2HwZogU/gpm++3xGCvOCkHXbGQ/DleUDiX16BwIea8tuWBSAy8vLms/ndX5+3ubZr7mBTqvVqvEF50uyRpwQ5fVgnrfrFncm9zk5Amsddzh0TIC2u3M6nQ7OKqyqwZFg8J5fV2SFFWsO+vj1Jbu7u4MTY6C162WcjM2yIAUyoOhMUc8R/TPYOy5mUDX492Tdi1o0PUDprfe8NxVxPsOzqYxl/c8LoJvKawFcm2JHmP0UBLyFbLpp8Jeb2TkpPCfZbsjMUuq1b+3PjApzcq9PPbDbJIHBiQV2YRFHyfiaj4+CboAdJ33z2c9aAK1WqxaD4Aw5hExaMIzz448/bu/PWq/Xg4SQqptT26+urmq5XA4OGWXsGUNkvhynu7i4aFsCDD6OwfEf2pkOtGvrq7cQ3XefAwmdGSMCmVibBRP3+xmDjJURaOB+WkhATysvPOfUeOjorQ7mBXiaP+/hevLkSXu32+7ubp2fnzdPgNefU+nZCgHvWRACgt4Tt729Pdg3OZlM2uHAxJmdhenEFBS86XQ6sLw8Vs8daxFaOauQtYT7HtrRDqUn4FGo4D17DMzTtjz5bZNlku3ASwZMz6XljMeMYug5thIPn6fVmsWW96b+0mbe599epbwWwJXlecxvSlpgLMIvfelL7VT7Tz75pGv+m9FycjaZ9T2XkzVBgxJCwK6OBGkvBq5bKFuA99xp1Otgql2HqdWNx+OaTqctUaTqRuilNXp+ft5OwnCCggP9jI/4AX8+C6/nauG7QX86nQ5AynQy/fxn7dkCzlY8wG2BYas158SaL/S3BWnFwAITIcd4bG2lEOpp9VgApg/3Uz9C2EDW42HqtxDnO78Dij7myDxofkKZgo7E0szHjNNeCoSpLRQ+e+6YJxS2Hkh7Dd5lrZg/matUmHqAZWDJ+0zTBCBf9xznnPi3TTJmE/j1Sq/unqWVz9xV7gKyz7O8FsBlLcHxkp5mkEyDIMa1g7b+Hd/xHfW7f/fvrj/zZ/5M/d2/+3dvTZjBpmq4GDI420thzzhHail2D1qbskDworX2iFBggdta8bhZ4BYkVcNXjNuNica5t7dX+/v7g7f62vWJe+Xs7Ky9jsTg60w3W2xOj7flgAbJeKDL9fV1A3i09d7cTyaTW8kSBi4LfejtzdpYDxTPY8a4/Gp3LFjGk1qxLTPHWhw7NL8yJ5km7n7iGcAa9brAokM5wBLzmkmBapeYLcHRaDRw/XI/bWKZTKfTAQ3Nx/CBFQM8Hm6HPkMDJ2JZWRiPn55fSDzVZ1163pgr1qSVEqwS7mce6YsVANdjq9wWTXo4MjuzB17+3lOQ/T95kvY3AaLLJuBy/16kbLKmUlZ+HuW1AK7U2kwoL2yuWWP+yle+Uv/xf/wf33qWNN8f+7Efq6qqf/Af/AcH9Vm7TRM7U1h5zm9tTbearQJ/ZmxYIHaNZfaUhTdC6uTkpLnx8OM7TmH64CIl1Zi+Vd1kXnGwL9YBtPDZiavVqn7lV36lvY4EkMEVSDzF7k+EBP3BLWWQ4hncScReqqqWy+WtlzUibPhMPcvlcuCqSrozbltMgOFkMhlYm5vmPc9rPDs7G/Cr964xZwAfPGIXrTcaZ4xwZ2ennT6SQgcBjguW140YwBi73cn0Ex5xdh9CGwB1TG4ymbRtAdfXTw9NRrGwKzPPMyQjlfrh5+Pj44ELz67JxWJRq9Wqjo6OWvyLNWbrm032Bs6cW647OYh+OpaGAuTkJwNUKsgGR1tvo9Fo8MYDZ22mRej5oMAvtibtOsziPrLePXb6a+Uqy12WYa/fvufztrpeC+Dqme49M7pnmT1+/Lj+0l/6S/XLv/zL7RibP/gH/2D9H//H/1H/zX/z39Qnn3xSX/nKV+rbv/3bB8KCOtJ6Sq3VC/uu+JuBoudqSNdUphnbgkA4Y025fTK4fEwOcSszMePC8mBvFq4YriH8EWpkDhLTYsEsFotmUaWFkfOTi9KLDIFBmwiVXBw8w2/QNF2yOWc5PygBCKDr6+vB3jiEgN1p1GMLdzqdbtS2bYUxr27TYOwx0B9bilbKzFu+L60OW8PQ2xaE77fVY8Cjftr3uBh3Cm6UsnRrW6EwzWy98jyJKvn6HcYGeK1Wq2a10c+eZZPrsreeexaQ3evpWTFPJi+mopHCP3+jvp58o3/Zz2zD/c9nN9W9qfg5r5+7Yna9sb1ouffAlaB1F7FSI6qq+vTTT+s/+8/+s/q5n/u5ms/ndXBwUL/7d//u+t//9/+9/vyf//M1Go3qnXfeqffff3+wyK1JpUsyJxJmY3NoPld1ex9Xjo1iayvb4lmECovTe1hwXaLF+02wi8Vi4AJxHbx4EKusqgYZj9Dm/Py8Hj161LTh9frmhZCLxaIBDXUDeKkUuN9O6/fLA8fj8eANyj3gWq/Xg9/SPZgL2EqJ3a24AC2UoaVjZG7fngCy4LA23Z61YNdtLdb3kayAFeo6XZfHZt7PmJw1b6zzdG37OXjMngy70OBRWwKOSUEvW1nww/X1dXtnm8HQypvd4peXlwO+NeDCF1gzT548aXRy/+mn15uF8CbQyvVnIOpZH5u+31VPb333ZF0CcNbb+/y8wLHJkurdZ/q5vTeuwk7xRkv/xz9ddTspIrP9dnZ2aj6f1/Hxce3s7NQf/sN/uH7oh36o/tgf+2PNpZIxlLvAEkEwnU7bYsGiy2A0E80iY9FiHcHMuAoRpqldUjgBgvuqbt51RbxnvV4P4g8ISi9ahBKCdzabtRf4sdid5fXpp5/WkydP6tGjRwPX3gcffNCNx+GuZAyZ3GE3DNbd0dFRW9g+6cIaLs/jrluv17diT9TR0/gtLO0+xMVqCwfXa9VT4MciYAsC1iGWBC5o3KpYtFgEab1bSXGGmrVpuwjhHbtjDWbm++vr66ZIMAdeT/AHc2HeTmUgY6/wuDN0uQ6/YSmZTj4Bwy7StJRZ35eXl3V6eto8ArSHkuXDat9777366KOP6urqauBhoE4DL3PLOluv14P3kfVcbYy96iZFP+fM67Q3J57/XnGyTAJsKiteV1aE6LPp6fv93Z6cVGB6YO5x3AWQzwuGm8prAVwWupu0Jvuue+U3/abf1Jj0J37iJ+q3/bbfVt/1Xd81OF3Ck53atxnSSQpOb3aigzXbNKutfSPwCOya2e3Wcd2j0fBUDCc38JlYh90yXmww1mg0qgcPHrSzG53K7q0Fjx8/ruPj42Zp4cIhXub5gOF9ckBv7vi8v78/OFyYsTAe6MmLCk1T5sOf3aZ9+haQpoGVDMbiOU5LBn5jjAh682iOHavW7kBb4KSV06fcB8UYUgGxW5Xrtk6cJej1skmAWqHieUCYvph/KVm31w8WLfxnz0Rq7bj+SKDAymf89AUez+O6Hj58WIvFok5PT2+5bumn58nWTyqr5hvHI712vO5TBpnmuTbMV6ZFyolNCnNe61nhud4S1Fx6/cz/qdxsetZtv2y598DVY6ScDK7dVd577722KP63/+1/qy984Qv1Xd/1XbW/vz84iy0Zr6dVWMD04hJVm3em85yFtEHMe4a4P6036s94hX9br5+68Nj46awva4aj0aj29/cHgXm0ZoQCSSC4AgESwNHChzpNA3/vgTd7oRxnsnDheZ/v5v7bfWQQyzZtueXCc52pNTO2dHu5v7RLP8wXqcUnP3A/85LjMfimAufEHrvvKL2x5GfT2MLZSgIgxrhTeci1mHNAyeOdbNmllWCLPL0U3Os9dOv1um1wJlkl16FpmwDizwlS5qsEfINGyqPks03fE8TvKj1wymIa9vq8qS3TJAGp1/fko039fNFy74Gr6ib4W9UPLLKI+D21paobRnT2WVXV7/ydv7P++l//6/W3//bfbkI7hYk18Kob5sdCSm0TzRHtF7++M6+wxtC0PTasGQ5ErapbQWsLM1sGjIH4gIUQ9LB7kjgWz5Oxtbu7W4eHh/XRRx/V48eP6+OPP25jvrq6GsS0zODpFrGbF4sJ99DBwUGz6nwIb1o4gCj9w2XU86tDG/qEuxHBkskR3IeLcr2+OUQVfqEvs9msZbddXV0N5scbiB1XshZvHrZVias258sWIQKc8dnVyYZhg2c+a/DOGFfyFSWFKf31qSEJNtkmzwE+q9WqPv3009aXhw8fNtDF3Z1r2dmsl5eXg5escowb84OHYXd3t/7e3/t7bauG3dfevpHx5N5bpdfrG1e0ZcR6fbNB3jIoFY1U2Jhz0y4BLOmeFln20UDLNbtEM7EolXLzTNbjteh57gH+JuXoRctrAVw9sNpkRm/SQOyaOTo6qr/xN/5G/Zk/82fq9/ye31OffPJJ/dIv/VKzHjy5Jr7ddNSZDMpz/N6b5NFoNMiWy1PLYboeIFgTZ0wITcASWgAw1pKx7Eajp64rXt7nwPn+/n7t7OzU48eP69NPP20vgWRvli0vM3JVtXRl04t2811QxIRGo9EgS9J0fOutt9pYvRdvPB63NGhcirTJ/jG7aqtqABYsZp57/PhxO5kBRQnhBj171rUVm3Rx9ha563ASivcWcU+2Cd+YR+kL4JUZhPBGbvjG0vV6mU6njX4WdLQ9Ht+cOAKgmJb0KzMfGeNqtWpvNYA/T09Pm6tzb2+vtQvA8Bla4DJ27NW8R3u7u7v1hS98oZ48eVKfffbZLYB2sevVgO75zblnTaJ0cY37kw4JYOYtnuW/Y2vuoxX4rC/rSIUln9lkDaUFZXp5Tp9lTfXo/CLltQGuTVpGTxNJEzbdBaPRqMVsDg8P6/z8vH7lV36lHj58eKvtnomd7hi+Z//8fLryss5sy8BlwdzTumx52RWFhjoajVomF/3E0kIQ0T7gMJlM6smTJzWfz1tSgje45sLL8cC4mRqdNOOzNcZNWjwghwDjGKmqYbDcsQ2DpQEBYQQgQz9r99SR/3tjsGbaUzws/BhPxtAMchaCabX5PltS1vxNW/rlvnDNvMhzKE8WPnzOw6B7gjOL77cVW1UDSxvF0bEx992uWSsk6S5HkeAt4uwV6wlTA8RdgjZ5Mml9lxwyGKaiZ95JOvbWl3+zUvW8ZRNoPQ+Qbbrf8rYny16mvBbAVbU5Ld4LouomMQJBbU0NwXl2dlbj8bgx9l/7a3+t/tJf+kv1e3/v720L021SNi1YTP9kRtp2faPRqB2jgzC1NozrkPpYjN5ICyBZs2OBWCNfr9dNa8Uth1VzcHAwOIcPocT7ty4vL+uDDz64tT+LDEwnDFgQ7uzsNPcgmV/cY8HN2KjPLkXXSX/pGxYfz0J7W1W5r8eZlxYcZLytVquWTblcLgf733oKCp9zkQL4djc59uRTI3Ap4koGNJNO1GEg5fnMegPY0xKwBen2DWKOX2HR+h7mwDFYJ5T4GLJcAwkMWOVkY3Lv8fFxo8ODBw9ugarn2G5Z2oaWlL29vVqtnh4k/OTJk9aOrSD/93rrWdXQoZc4xXOOz/qeXlLLJjeiacXYbUU5ROG13vMo5HzYM3RXSaWF571+3PcMd7wImPbKawFcPbCAYVwyY+vP/tk/W/P5vH7u536ufutv/a1NIFHncrms/+g/+o/qr/7Vv9oWbJrFqdlg/XjCAB0zi08/qBqmfnMvgj7dUrRhre7y8rIJHUCEe/OIJC8W+kyqPO6gBw8eDNw3xAUODw9ruVzWkydP6urqauAWtODE55/gRRspHLCYUCKsWedxO15YuAa3trbq008/be1gLVE3bi9bUKYzSgz3s43BIMvzJLM4uw36wnMpONlKYYtive6/j4x+ECPz0Uc5nwCSgRga+mioqpszF6uqud1wqyH43AefoALYwT+pKMHjjkdaaUoXli1dW7y4cdNipB3c0Y4dbW1t1f7+fuMJQPb6+rrefvvtBlxbW1uDPW8A9TvvvNPc3dCROXQGJZZaT/g6scnzz5gNdo6jen4cc0pw4Xuem2neSyUgLVOKQS2vWzHy9Sy2opIWCWj8vskKe5nyWgBXr1gD6pXr6+v68MMP6/j4uAmEfGa1WtUHH3xQ5+fng2N5rMnwOQW0+2EmTg2t6nZmoOtM9xP3mzmS4dMKSGZ2PdxjgUzqO9erqh0ZZeFvt4jpmr5ug6XTp/P9SXZ5WaDZdWTBjzCvqoFlknOee6/cZs5PaovURdCd56CpF7mVGoQde5OYo9RIPUd2G2WdaS2ksONZrJTU1J2+jkLk46V6WrPnoseDeb9jRH4u6c3nXj1OfjJYASJpcQB4phFHgvEbdcLTtjxGo6enmhweHrYjyuwi9RoxvbP/FF/3H7+Zz9IVm8/4WdfvNnqfE+x9LevJ/ua9bn+TLHV7ySu95z8Pq+u1Ay4DQMYYTCheYVFV7d1V1moonI334MGDJty51xadTf2q24HYFOYWsqPR8LRs3+e9W/y3ALL/3m4C9wvXC31Py5B7cPXg0uJUgp2dndrb22vxPlyBWEnUiQaJALEw9cZrx45Md4O1z3LE2kHjdVyMMwBHo1HTuqEPfZzNZrde2W5rIa0SxkC9aOoIHgtQ1+nNslg7xAp7Wi8gBZ8yR1VD9/JqtRrsn8v+5oHA0NaZkXZJwt+Al/vBfQnkdg2mcOqlgvs505r5sVLguvzySkAGGjNO/pgTsmzhU7dJpipuafrg17KQqDEaPXXFYzUCet4T6TnxXKZC6vE7MccZiimrTDfLDCufVrBNOys8pne62P079WW7vfsTtHr3JkhtUnZc78uW1wa4ehOY+0ry/tVqVXt7e/WP/+P/+OA9QFl+42/8jfWVr3zllgbEAsvjhux+MoN74cP8fPZmSQtD3JcIe2fDcW+6Y3IhIajy9AIz9enpaXO57e3tDYTegwcP2mIejZ5m+AFoo9GouQzn83ldXV01gWBgBcy88Mbj2++SspuIeUqXVdXT1Hn67H1ezA/xFWJK0J82LcjpkwHRCQh2ha3X6yb0aC8tG9qHjk+ePBnEhfxiQp5lHs0TxP+yPz1ryMLKrinThPnd2tpqhxhPp9O2Ed0uUCxjFDxOTWGj8MnJyYDnad8xKU4pSUWtqgauPq/ZdPEawL2lAECCrwGy+XxeJycntb+/XwcHB21c8MhkMmlvLSBeub+/38aFzGAMOS+ECzx/KAP0E3cn2Ytp4TlTtGcVwYvpfjdoprzz2rLywHhSdtnis9ywl8XFQJfeI571tZSJPSXlVcprA1xVt60qxxzyvqqbybN7woIIhsxNkGlWe3FZU7JgQfh5oaaZnlrpXdcYG3X4xYc++d39TgFrYMXCAtyw0vb39wcCggQF4gz8AQ4+vdvzQNs9vz5WmWnqgLaVAE7joK97e3u3rDArBbTheU9wTEGQz5nmVTcZitznTDqeAahToWHs1OWEGPeLe6jP8al07TkNOjVn8ypKEO5Evyk71wL19gQt7fpwZcZHvQCQ67LChRLkdqBxWi0elxWh1Pp5/vLysm3zcMLQ4eFhAxLWD9mnWGsPHjyojz/++BYte0pKz42f31NZtRVFPT33Y08pSF7NdhNEuJ6Wju9PKy1lqOuAv1Nu5bh71tZdNHqZ8toAlycgBb7v6U3Mponlt5zgquGL4wxcBiaEJ5/NRAmAvesZ88l0ccc8vL8LLdcWmdu3Rejx2H2HYPJG0uvrm1Pe/dJHwAvB5/071uB6gsnBfgt678GBFqTbz+fzpkXaQkilgL90yaTGaFDy/BsgGL/5KoEr4122Wj0XOX6DLPeZf6BrKlz0I4ErhQvjzQQVwMVxH/MrLluukchhhSLfAG3rJ2mffee/lTAD2nq9Huw9hC/Sysy5XK1W7VUuKFlXV1f13nvvtePDbL1cXFzU7u5uzWazqqr65JNPbimg5p+eXIBH0oVoJa4nG9ISgmbJj9l+rw/ZH/OZx+HfXXoWm59NOWkgS7AyTycNP4/y2gBX1W2tA8bkO/fArGjOFpC92JUtAgSthYXdCznpZC45MFxVzVqgff+3RukMPGu/aOrcj/AlU8xjdbzG8RC7Cpzt55jYbDarJ0+eNHBaLBYNvCy8rP0iwCyoLSydKGCNncA4dLNbCE3aG5HtjuQ7bfq6LZcU+swxbmPoaLBwvI566OvBwUGbt62trVosFjUajQZp/uYH93u9vnE7WrmxCy2FW4LvarVqp2qwbw1+3t/fb7Qi/sezu7u7zZWX719zjMtvD0jAhY8BhtHo6TaK0ejGxQ2dOIfR/OY1lIkn0I1+YtEbXB3P9fO+vlwu2/1/5+/8nZrNZrW3t1ff+q3fWp999lmjFxmZ+dJIt+tED+YB/oZXElgMvKmo2t1oIW9XXXovElhMK7wUVo5S+UsFwn22sum2qLdXUgHvuTN7wPwsq+xZ5bUBrtSyTaCexpEWVTKVJ4L6rFluqt++bDMHwo/6rWkjpBAYPpaJ0rO4WBReeNTjDcEen5kosylp7/T0tMW1EDK2sOyGdJ22FJyA4TH43LwEfGftMQbA0laRF1xq2q7LY7VgMXhX3SR/9LRErhvEqM9JPNxnHkla0ybAuMlqoFgYkz1ZdRO/ox8+EcJzcn1984ZogIM+YD2hVNmFm2vDygiWJAqZla3RaDSI/1XdPmvQCoyTXKiD8TAGJ9Ukz/WAAqFpxYN1cH19XY8ePWrHQkETTqaBFkdHR+3tzj0B35MP5oteQkSPnrn2TEvolVZNj2+oO4HQfJDXk99ShuU119cDpLx/U3mee56nvHbA5c8JVD3Nh++bPpvZkujWLnrMkG0bJM2oPcawYM2+wVwsPBZnxhsAkGS4pFECsTeOWsO1a2nTGK219UA8/1t4pfuVV7CzYdd0SbeM5+UuBST7ntZDatS9+qkvXUlVN8kjCYK2NJPHkne5bmUHEHLdGS9BUDuO5IxO8zDA6xiuBZyL24EHnGxh69CvjUm+Zr6T7yhWEGgXEDBv4XFI15bp6TkzYJ6eng5AyooiStX+/n4DLfOS+5w85nBBzuMmCyl5x7RKgMh+uKSy5ev2DsEbtqjyuRzfppLt+HpP3jyrry9aXmvgyoXoBWEBhSk8Go3aJljuscWAILBrCtDw6xPMvBY+TovNUwbSNWRh5JR46k8Ny6dreANyHsFjLR43DEIV9+DOzk5Lec/sJtNvMpkM3FP8Nh6P25lzaNBbW1vt1ed+/Tl0u7i4qNls1jTx4+Pj1gdboHZxOnnAwrKq2pxk9t9qtWrjxjowD3nRAeC4IG0p5sLuLUJbkBZCOzs7zf2JBcM4EN6TyaTF8uwSrbo5gQNrwe5s88Th4eGAb+jPaDRqKfpk06VitVo93Xg+nU7beKEVc+W+G8Tgud5mbwvQVLKguZUdxz6vr69b5p9palcWngiSMnyY9dnZ2eDgY96MYLfo1dXTw5F3d3ebC9XWFOOwleuxMd8eJ+NnLq3UwbNONmEdJdBw3Yqk2+sBEm1DG9PbvMI4e8raJpCyy7pnhXk86VLs9fNFymsFXKmFpwZj4ZWbWnvaM3X0PntBJTimlcF1LwDXZ2GRYGugoW67Fs0YCLQEEltfduNkLIh60Jxx2TiGhUDCEkI4EfOiTe/Dgl49QYUAwH1Fu15QSSvHJg0mHrv325kGCAMEoOfcmYkAA4LYyScWOM5czbRj9zmtLbedLmGPO/nGlhNz51ipXZIpOFKDt6sQBcQuaruved71OH0emsEfgFBmJlKn+cCp/uPxzRsRACCPwxuxuQbY0WevHzInifednJwMrEW73GnL8UnPjV3bqURyPzSCFqw5yx9bNVYWzIc9yykVcMsjW7Neq/Blb/7SKvJ9Lq7TJWm9qSSNfP1ly2sFXDmxLon0Nofv+ryppJa1qU85OWnyJ/DlcwbBnHQztOsgm833ZfvcjzAx8Kc7gUWYLiM/Y9eYrUyDsoWttUn/ofW6j7TpMdkSvWtRmaamR9LUAsvPpHsz6We6GpgYc/7OmD13PS3XCSXJR6ahQSABzWOzYPM1LH+DNuNw+nrypC1Q8xXCGvqYF3pzwlz6M3Vb+QAQTU/qSP5zoU5bqOPxuL2LrupmS4LlB895/XmLxia+s/JG8fh6PMln85RpZn5JnvLn5wGCjIM9T0kAz/Zepr5XAa2q1wy4rFX0wMfahDV5a5ZVN4xmzc4LODUXL0gLDf6cVGGLJ/33rot6fAYhfbMARPg46I2LxBqlXWaM3+PySQsIMgqncvvkgRwnWYns87JAgZbr9XqQiozr8OLios7OzgYJFKZ11fDNuz5bz/OYny0s0qqy9UWds9msuXIWi8XAakyt3FYLdWUGGQkQXKcdTvvIviD0+c65ka4by9pzz6HCjrNgfcEvuL5MR/rDuYXQCfpmNhleisyyTMvVr+Sxq81ua1vavN8s1yMbhx1XM+197mDOMXVhtbFXC/5kkzJWp/mVzc27u7st6Sljl+ZR6G5rzECORWm6mF9TCcJic3YjxYkrnjMDrLM33V621QPTu4oVmJ4C7nG79BRznn3Z8loBVxIwhZY1yqob8LI7o6oGQiI1bQt+f/d9LBY0RBjMmi0uktSQr66uBvttUnOz1unxkgHI4vaYveC9CJxOz5i4l5OzOSUDAejXy+MGgVZk/5lWCBjTHcC5uLiok5OTFvcCYD1OQNX9R8j5hA5r7naLIvw4axJh3tOYXQ/nMlKX367s+KizOa1wGJBok5idBV7GGfwd8Nze3m4nu9CmXZ2z2axWq1UTyswxxxlBT2jFmAA+jsQiDmRAs8AmbuiYGLziuCq08FsOiKdRj7cucIqHaQvf+ii0vb29AW8B4FaOnMgzHo9bnA9hzu+8zoQTQXieurDGHj9+PIipmUd8yr+tXcbV469UELyO+S0VX36Dbl6raRHnZ/MTdOu5HTcVK7fPuqdnSfWuPa91eFd57YCLkmBjAPDEMrkZt3A9Zo6s2/dlf9xWlt79KcRgYu5PbSfrM9i6D+7zJsa2hWUXFosw3X884z6z8B3ktWXk+zI+iECzBu+2epqdr9Ee4GdLkzhOKjeOYwH+SQP3l3s8B1ZATFdAgXH7uCLzmGnjOeQ3+mlaZ/9sYZg3bJ2n8HK8xjQkzojF4TiW67GbMV1tLlZWaCvbt8KRawElLseXvMs9uHttSdsTA73xMDDWtF54Nq0n06BnxXCf54B2UyEz//YUKdMk606Z5nXmenNOeu24vd4cPgtkNsnCTTT5PMprCVw9RrKlgyZnzdmM7UwpM+5dwEVJrdn3bwIfL9QU/FU3i55UY9fDMz4BPRNHrA25bbs3/MJIFrKzwjghw++iyqA61pAtt9HoqasENxXuQSwiu2hJDnDyheeT+fOCpS8sWM7jYxwGG54l0QKawgts5EUYQie715yF5jp9jT9cU1haftElPMbxXDzr8QK4tpQ8Ls8DzxkoMqPSYOkNwQjwy8vLOjk5GVhss9msnZ6OJ+Hy8rK9QQDa2RIw30LL8XjcXMjUzTxMp9M6Pz9vVi31YBl5ns2/ZKdicXp+sFQNNOYdu9UztGDgMm25biUrATfXmxUO3Lpe06SqpxWUCllaPpsUEocRqMtj2yS3LIOyuD6vvU0lrbm7gPJly2sFXPan26JAGE2n07aInQbuGJI1egskC07qt0D0dWufdokgzFlgFsjOunMf0NJsidDnHH9q6xbq9MfWDOnguKJoYzQatdMKOBDW2YMIPmehTSaTRmcnAazXN+60qqonT54MFhXC7+rq6UG9CHXcjr36nDRwcHBQ+/v7NZ1OWzo99TrLrermpHxoggD2Aau2iGzdHB4eDvpsjZf55tw/2uJEcrL2rq+v28kazMnZ2VnrT7pufZyS3bCksiNwEeAAi3nBgg0hirA+Ozurk5OTwSb1vb29Oj09bddoH6UFIMYNCajZCvbacb9RDBgnqelYd9DA8Te7/jwv1GHFx7QDdFHo+M1JRWdnZw2cHj582GQDG5KtkNFWZgYz9wYRlJIEFPOX+2ulwkqpQQg6+X1lCcp3gQ8lMwdT4c+SVp8tuvRibQK1VDY/j/LaABfFyI6AsGXSc0P5syevN6kZp3Fbtqp6/fJ1uwLNiNakXK+L/du+Rh0AhfvvuIv7ioAAhKuqxUoAK7/RuOe68KJCMAC4PJ8WZW++bOFaSGANssDtQptOp+3kb56zll91k0zA894KAWCldgzgoXAYrDwnnkMncFgpsdswFQsLUkDZQsmuRvpo12Bq0ekaM/87XdrzZHp6bE4eQtmzggYQQwMLZz6bnhb69gr48N60FKmDetIb4Wv0ezabNUswBXm6dQFyxwBTBrjkb7Z+03qyByIV2bSUch57igf0eVa//J06e3LkeS2e7Ft6bp6nnrusvZcprzVwVfWP4U8GSUvFYJUmt7VXZ0al9mEQS6ZBEHlxWitzvxhTujJ8Ugb30gcnDPhZhIcB0nEK6HVxcdGsH+IejLOqBtoh/wEax4OoCw3Wc2Jam6b02TEpv9qF5xB+ABf3m6Y553z2b24zXSI8A2hktiV1QEfcfhY6Bq7cX1c1BAcDEy5rQNeWCPzC/Bis+Izlah42cKXb0e3y34BCXz3f8CH7rsxPeAUMfLZiEOD00QqQXfjcZ95xUhDz43WNlWtrm/r86iCPIfewpeJnXnJbppX5c5OiB/9sSuDwONPaMd+4LwlSvWseg4vHmePtFYNWWpXZTg/YEqxftrw2wGXCW8j1MshS0OMGsfZkrbdqmNaa+4esXfoUDRZObnalb7z0jj7QP1x4WDH0jbRh+s87vOwioq9kY1XdDvJzD5YKMSVcMycnJw1weO8WxSC9Wq2aCwz3HouV14/ganTiBoIF68iCzMLYZ+FRz3Q6rel0Wjs7O02zXq/X7R1btlxtdTBm3Iy9Pwt6FB7HSQAzb0Ymsw3LAUCdzWaDw45tKaxWqwFPWLN2ViCA7NgNCgJC3i89pM+415z0AN/aVYTrl/jjZDIZpMa7pDDmJHXzs8fpfnJKid8HZlrBE1XV4m2eR4qVE69tg53Xk+OIBlTmFfqjcPAsBxLjLmfe3QcUKvcXejvb0GCIEpiWcE+AW6Gy4sc80T8AAsWIZ3tzaF5jTq1M9dyMm4DP9xmErJDRfyd2uZ43wKWSDG+ms4BgkfveTYTsue8QDMl8/u6FZMHHPQbIXBzWfi34DZgO9gO4tiKSBrY4EK5OUgFoiAVlvYwRl4oTOPgdIVVVA4FmyxKhWTU8rJb6aDctEbs0DTbQzfueEKJo7Tn/jm16fr3YfL//bJVYM+blkavVqgXhrSx5r1zOEXOcdHIdXvzp8uL72dnZIP6Vc2PaUA91Uw80QChTMjZiy8vPZEwDoLBigFfAfOw1AL2cAERfmSvH4Zw2n+vZbaJ0eKsA683gZgA2/b1me5m28LQTvJIOtqS8pgxoabG7DorlWs8y64FMfubZnltyU3G9m1yrKS+zeBwvU15L4DKhkrE2aQyUuyYORjIjp8nr7yyqvL/XJy82W4k94OA+WxV2F3mxJU3s1kIDps/sp0JgeRGlEKdf1vZ5jgQOLya7YBxgt2uQz26PZ9CQHYtDeNiirLpxieUiYux5vwEkQaLHM1ZAfA/0ur6+bm/XNfjaE+DrridjsNknj506/R1Lg2e99cA0tavQ4OMsRp639yE9DVU18EB4bLbux+Nx23jN3JuP6IszSg3myUdW4gAdz2GO1de4z0k+nlP4quoG2D1f9CPdkJ6j7AvXbEl5PVuQJwhtkmf+3HumBxiUnjzJOu561m33Sirwm+552fJaAReapAO1XrAI6R7w2AKDuc1gpDNzvfcuL/7Sd21Gttsq+2dXkjX/2Ww2EGi2SnoW4yYNZzwet9e1425DsOEWYcOx42Te9Mw7sQAXgAtt1nGLdMXR957G7Nem4+7CDeqDXrEMATNbT6a1BSdzh2aOZXBxcTHY2DsajQZvcs49N54/u6t41vvALABx72TczRq23ZK48MyvBhYOQU5AtKXHXDkJw/xhNxunoeOSsuWKBW4Ag96M1zEoMkSrqvHt9fV1c+XaimP8y+WyZXfmwbc8Nx6P233MDVmDe3t7gzgXihO0YeypzHBCxvX1dR0cHDTef+eddxptjo6OBsqcASYVI9zGZCtatngePf/MRyqFrH3a7CkezHsqAJYBPWXb3gR7NnqWW5YeiLqd5y0vApS98loBF8VaT4KKtbTUCJJxuNbTbHraCgLaQXDHMbAs7ArqaWNmtoyN2e3lfpLRx/i9KCx0Hzx40ISDAdNuwtFoNAACu/DyfVx8xuKydutEjdTcASzuS83bliFuHSwK+uUYHfcxHlsBfpHn6enpLTrRLx9Wy1iqbmvDCGzAyiDmBI7MyuOZnrXP/NpagVbESWkfJSGtbsbt5A2P09a/wQY68xnaWcBlxqHB2G466IOCBE8ZZLHmvQ7sxrRV5HePcQ+giXWYa9s85vXjI7m8JxKa8tvp6Wmj/8HBQQPGxWJRZ2dnAzdiz4pNt5+tNmKS9K0XV8o4lOcrXYXmHyvHm+JV/uO5rLdXEvx6v7s+F8vQV7GyXF474EoTuOr25j2ucX+PmFlPahfP0jA8UQaidE34/k2Tu2lM1JuWZlp1tGfLBsFEvQYjLyj/d0zL19NP7j4n8Kd16D6zyKnLcb6kSQKbg+CpfSL8eSklQGn3ki1Lj8Fz4++2JPndQjQ3Fqei0+Mf12XgymdRIiiei7QIzFceU/KS28VyyP1iVras6fs0EehuQHcb9LfXtpUOj5X7zGcu7ktatY7hmZ6e43RJGlhwXzJG71V0nekW3yQ7UgakUgMdetc3WTu933t84+s9/uvJmV7pPZtr5HnKq4DYawVc1lrT7YYw9p+ZnP/JhKmdmOGcZdizhFhsCDjHcaiHNnjWY6Fux4G8kH2IqgUZmYheIHt7e/XgwYOB+8xt20LEVee3DwNY/Bn0oDv9sBVWNczEGo1GA+DzH+cJooHTN8ZtYM1sO+71ySL08/z8vL3OgvtsJTCGxWIxiH+l8DE4OiGAeXWcMYVOuogsbKtu4jsGf/OyQSlpj3uzqtpBx3bJQjO7fD33Vn4AXdN+uVy2MwztIsa95+tWNuAJx7r4b7BzospoNGquOTY6w8M+SQR+cLILbzW24obrMj0lmYDB/Sg9PINrGSvy+vq6JcBYWOchwT0QYH3gQXDGYsby7uIjK38OZ+Q47wLprMdj7pUeMLl9XzNf9+pJxfBlymsFXD2tw6Bg7ciaDW6UdC1ac7KmlAAHiFAMliwcFjEl94BZw/MJ0NxTdZMmT3Gcbb1eN0Ga437w4EETMkmbxWLRUs0dE/HmW3ztWC25ZyUVBo+VRe1x2jo0UEyn0wHw8TxWUtVNxqX3nxnEbHmcnZ3VcrlsMTHmyuPiJIvxeNyEvtuBhoASLiZcS/v7+20ufWKCFQy3a1cjCoVdkrZuDFB285nuCAnGA4iNx+NmMXGP+dX9I6bFGKAf77DyfNrShDY+4NkgyiZ285z/7Dbnd9YGgtzrFppAe58+49ilBaLduvSfzz5JxwqVtx4AMFVPwe7hw4e1v79fe3t79dlnnw0yV60keC0ZhFBA1+v1YLuEFe20pgw+Ga/rWWEJCAlgqQDl73eVtO7y/6a+9Ky0VymvFXDlJPTMbt/bI266WzaZ7q4jXQO9Pvm729pk4hskabPnMqNYUzczeZMuwsv0MShxHfAgyxBtLGM2XkQ51rQ0KBkYzj04/Pm0DQtauw7TCsp4m12bFrbQy1anlZGeRugYjBMBrGRYITINcEMyp07oSMvbbVpgZ989z26vp+W7LoOkhaE3SPsen6pvUKRdx3KYF19L/scytZXFvi63Tf1phTjmBQ/aE+G+28o0QLnkWrYnw/1wQck6ODhor+Wx4uq6enRImZH/ExgMMK7LyssmAOnJpZQ53JdzlXIpr/V+v6u4/69aXivgqhpqEZvM7R7zwmBOgEiBnN/tBuR6Tky65Ci4V7yPKQVnWl35GgcvCoQ0dfOfc/y8V8rjQbD79AMyBxeLRQtUo0E6acOBZrsrqSfdMwgeg1a6+ohbHR8fD7IODVwW9HZ9UretR1svVUMr1cLGLiJoA01oH5Dl3U6j0ailvQNGtiQNOriTHT+kX3bD2AVlK8s8lsLFigDfoYlddzxHlmW62jINHquLfnz22WeNrxgr/GTXtDMefS9zSVvQBq8EyoaFvpU1v5XbbjfG5U3tzAmvy7GyYDoAfqzl7e3tOjk5GeyRhP6sA/jo3Xffrfl8PjjvsaeoQddUCnI+0yq0G9Hypmehp8vOdffAxoDdU9Ty/k3KUQK/r22q5/MAsNcKuFj4jjP5t02E8nVrWlV1SxBxf/pwacuaPALLcQC7JGkvMw/5XFW3QNTjtHsCJkYoVT2N43zhC18YZNt5jJyQgbDkFIXHjx83S8sCnlMOECx+v5QXOlYcbVloPXz4cACwuPEmk0mdnp62sTteUlXt2CcEEWB9enraXJjX19cNpIlPICDPz8/bM0kP5sOuLWfxmaZYf9DAJ2dAI+p2/8lYrBoqD1dXV4NTTphfCyz4j/pQGHoxRNMP4UD/oEFahT6lfblctjFdXV21w5fPz8+bW9PAvl6v6/DwsF0H8HyPLUtb11grWNfQ/fT0tPWf2NLW1tOT1VEsAF7m3SDt2PN6/dSdC1iyNtPlZovYoLxcLtsLRtM6Bth3d3fbWPiNce/s7NyiO/PDc3aHMv8GfOq0gp0WYco283YqAwmu1NG7bjDqAZKfTR7edO+mel6kvHbA5YV+lxbiv0R/M0vP+qGeTa671IC8cHmW627H96VlwXPWrg10dp+t1099/gSrHTjOFGA0XQLwuAe9j8vuMDOehZhPXU/ty3EoB/K9PyiFrK01CxkW+NXVVUtN9pxxn+mBKwo6eZ5sNdt16zFQpy0yrhPzssXZm28/4/lPHk3XXo+/zCumCaCRfJo8l2nYpm+e7IDitLu7O4hvOibjP//m8Vro0j97Cbhu2vle12NesZvY2ypsIcPLPnnD8sFAZWCBP9PCgNbQbjKZ1IMHD+rk5GRw6LDnNGNX3GP3vefW64j53VRPr2R2pZWZXtlkdVk+9qw10+XXsrxWwFV1+5X0/vPkpCbDde5LYKOkYLPwqBqeEJHAlQyXk+37eqCYVkIKI8r19XXb4zSZTFoCAq/ZsCXF39nZWZ2enjbQckZYxoIylpFAYzpUVXMjeU6sETNvBkLqtJDz/NBn+sMfzxm43Ha6Hc0H1to3AZfHWnVzxiRgYN6z5QENc7EnuLm/OeYUbhZktqYRuMk3jrH5AF0Dl19jX3WTFELiiq0J6gasLPCp1+3bE+H78UpwPdeQ+cL0svvXSSAGcI40A5Q3uV/phy0v5sAgAp94viaTSb311lvN8odOKRPMzwauBK+UN15Ldyky5hHGnGBMHQmWftb1us0E5JSDz1ssi1+2vJbAZe2CRZFamoUUn6ueam5sdu1piBaOaR5X3Q6wZuwJRuSz/fI8xzj4nfZsOVUNNUcvuNFoVG+//XbLeLPbajqd1mKxqOVyWRcXF3V6elrz+by5B7GArGlaEAA49JH/Fszr9boODg7aGC24UgjYJZcpxtSPCxOt+uDgoLldrUW7T3bJkqBiF6T77Aw2vjs2kbEEioVbz8LvCQPzlfkowdH3U5djcgZm87z5iLHmM9DL4+FZQMDgTx1bW1vt8GXceev1uubzeVOSXCfPEn8ki466yLodj8ft1TWcZgJ96E++h4oXklrRY97S+mXdbm9vN7cfrs901VtZ4A+gJt6Ht8JK197eXr3zzju1vb1dv/RLvzSYMwAEtzCFdWtZZbq5fssNeIUxu6/2SFjB6SnJ5lMrOlkArDQAkj/vAq8eKL9Kea2AKxd/ugJ6mktqsga9nmsvNXgzl+ulZJp9anxpGWZdbsM+et/n/o/H4yYMVqtVO7Edpt/Z2WmvZz8+Pq7j4+PmIrSbwoV4Qq8909CC3sqCLQAfJ2SLxMLANLm+vq75fN76BwBTtwWvF2iCI8IMgGfBe44s7DJpxPVW3WQZ7u7uNmHpOAh9Y6y2Piyw/Jt50MIreSsB0cpAxsXMV/6fwq7qBsCxumxN2oJBsUOR8BuRARlbAcmv1GkgMg/5nnQ/9wSl18RoNMz0zGQt5mRnZ6cWi8VgX5itd2gCOOJChb8nk0mdnJy0/uMyJp4GLQE684fnAsXU8TFbRJ7rlAdJDwNwT2lyu+a13lwln/D9Lksp+9YDuN69L1NeK+CqGpq9CQabUN8EZDIzFuLPqXnzXI+pUhPns625ZMC7GDU/23VIe2isCJGqm1dksFDYXIz1tSk+4sVu8E8as+B9Oofpzr253wWBYYHr8V1fXw9ibrY0enPohc93A2pv/nuA5/nP3xkX1ny6YZJXqCOBfpNC5WfS6s/F3hPorif5i9/ML5kR6c302RcnOfjoKR+/ZMvWVnS2bxAwSFqBSAs05xse8dqwm9vKljOGrcgwHoMp/QWQPMcAmxU83POz2ax2d3fbeZ7psk5ZQz8Zb29+kze4L+c2XdzPKs9zX08++XoC5CYA/LzLawdc6XrpCZ2qGjCxhaWFCJYDxSZ1HiVj4e7JhDFT+PcEjheCn7H2Z6vDvnQEzmw2q/fee68JFrvOHjx4UOv109fFP3r06Ja2ljEIQMPuCfeH13iQ+cW9FnwIP2INCIHRaNSy2PyySU6kOD09bb/lZl4LGMernJ2HayldZePxeHBigV01FnIsQE45wM2ZCRibrKG0WnmWeYEvcIUxj04rxx1KP+1ytpXCJtmLi4uBkIUGnlO3Q7+pz5ZG1U0mJ2N33GR/f79Wq1Wz3quqHXhLXbPZrI6Pj9t4T05O2ppKhc5vufZ6yOw/g8BodLMh2RYtipPpyFiJ3/rwWuQERztxALU3Cq/X68bD3sxPfZzNOJvN6tu+7dvql3/5l2s+nw94ya5+K6/eJmELyJa7eTjpwNq17NmUlcj8mRc2WT8GRM9NT3HMYqUn67N782XLawdcVUMgcnaRCWXh6meqhu+v4gRpfkeQpAvEi7rqdjaPFyzCwozKPQgYWwmpsdmNkwIHYWuB7JTiX/mVX6n5fN7GiZ+fBA4EAXVVDQ+LdSIEIOV+ca/3B3lhESBH4/UJA1XVXl6JsIBWCHesyV5MzMCbwtGaOxo27bKQ0o/PnBN0h08sFD3XVpT8u4UEgpW60mo13cybdm2aXwEoWx3mCwsft5OCzff2sgapA7BlW8J0Om0KEpuUrTRMJpPB63NM356gxu3GXGG5pDLg1+7A71Y8PSfU5ZeCEjelTlygVlJ2d3drf3+/9XmxWLTY7d7eXu3v79d4PG5bOug3yhxACc3db+hpQLY8YRzpVnYxPdMC7ClQlou0k9aSC/0yCPUAx/LQciKLlXl/f5nyWgJX1W0N2AS3Jtt7JhdXCjXq9XOeuJ5LahPTbfq96kZQpHsh27aQSqvRgna1WtV8Ph+4MQgW+2SIXPy9wK375gWSGp2FuQGNeUCjR+j5PETTMcfmdtNtkuCRLtCk/6YFZH7pnb7BPe5LKhib3CZuOwW6x2W+ukuwpALWq9+uuaSDrYBNwtTPGMQAXB8JZmWMuvf29tqzZN+lCx9+7QlUW8Zec3bB2QWJcpTzDdizDnIdwrM8jxLh02OwKO2OxyKezWbNmwCoGVg8XtOTe/L3Hv/kmEyPHri43fze49PeM7leekper64e7/b68yLltQQuM2gvs61q+FqGXrF23hNY3GOmhKHNiLgbbHHkorc7IWNfPp3CVgZapvdHUacFnxcgCwmLBtBar9fNkrGrj/1dFigeO4kedkdCY2cLmg7ecMk8nJ+f15MnT+rJkye3gDOtBAvW1WrV3JOmp4HcFq61+KrqAhqWi8/Bq6rBq+bpmwWpzwR0hhvPAy4JMmnl+UxExwNpp2r4avbR6HZ6NnV5Lx4uLoQ84zG/GEj8HL95fc1ms8bX0+l0sBHc5xXCe9TJ3sKzs7OBe9BuVK8fEkXW6/XgZH+7nJlD+MwxVjwutrzh65wfXu1DewYiPBbcs1qt6vDwsGXmQm/cqG+//XaNRqM6Pj4erG3oZ/nifqZrL3kRBcrz3VOm0lNjnu0pwsy9yybrqaeM5TWv8SxWkF62vHbAZZCw3zsLxAbArDEACJlGb8FYVYMFQZ2pydtqMIheXl621FyC09zndlKbyYVqjfvg4KAODw+r6sbds7W11WJen3zyyeDAWh+zg0uHjb2A4+Hh4a2TJlK77S0IL0CEcIIIm54BUysc1tbt6ycOQSAcYeVYjl/dQh+ok+0OjNvvaEpFh98y7Zo6UVgAAI6BYsGbbrY8bfHbBeh4UtLAloABdTJ5eqxRVd1q7/DwsD3DSx2twDGGnuVB3bbEfBbkaDRqL/i8vr5uLx/1iSsoWaYXaw6lB4FN/KiqmjWWc2UeMm1Wq1Xri92B3ENfeMEkSRm0l25n+HI8Hjf3IC51n/ZxcHBQi8WiHfvEPM3n82aNeb3k/NhVCr1ZuwamPADA68wyAn68ay0mOLp/LinTbGXZY5LWqp/ZVHfe8zLl7reHdcpP//RP1+/6Xb+r3n///RqNRvUTP/ETg9/X63X98T/+x+tbvuVbajab1de//vX6m3/zbw7u+eyzz+r3//7fX0dHR/Xw4cP6Q3/oD7Xjfj6vkhpOEp97es/5mbxOseZ/l2mfLh/3JYUSxfek3zr7Zu0KP33VDfP7aBzvhbJwTFOeOg0QaJx+eSP38GI/W5JppSaI4GLhdHof8IpgdUIEfbRF5cNaKXa5+Ls10E1auj+nktITHMlHAFFazX7edElrLHmox2P02XVhcXqcORa705KvHRvsaedo3VbSqB+ey9ebMH+OtTL3jmu6vuwzQEXGYh6XZuXTtDZfJ0/ZRWmr0jSGP4m3zufzms/ndXp62vgVy29nZ6dZnxS/+cHJNrl+k59MA48zx+M11ePp5Em+pwzrWU53fc/fngU8m0DrVSwtygsD13w+r9/8m39z/fiP/3j393/n3/l36k/9qT9Vf/pP/+n62Z/92drf36/v+77vG2w+/P2///fX//V//V/13//3/339t//tf1s//dM/XX/kj/yRlx9FFAhq/3aarT0gsjbZs56sCeX+kGyb6wkUvbp7TEmx+8YWV1pygIoP4q16mq11enpap6enLf5An+wesFXCggSk/LoUMq64Np1Oa29vrwWkATs06Owr8Q3cllhcCDr6Ze3emylzX5ytarsInVFpoZ7xnapqGWFo/gTl7Ya1kERQct2CmhTx5J3sM241071qmIVqnrBWnS/PTHobyHxaipUB87TduU4+cvKM+2WeQ3CTgGG+T5CAn51F6hdOGiTt8YC37QExbfMElnw9j9d2unCTp9yP1eppLGw+n9eTJ0/qk08+qSdPntR8Pm971abTaR0eHg6UBvhgvX56TmIe2GveSf7wPOQ9tqRSebFikkpayr10WXst+D5fz7qsMJtmz1teFbxe2FX4Az/wA/UDP/ADGzvzYz/2Y/XH/tgfq3/un/vnqqrqz//5P19f/OIX6yd+4ifqh37oh+r//r//7/rJn/zJ+l//1/+1vud7vqeqqv79f//fr9/5O39n/bv/7r9b77///gsPwtpM1Y214eNrLADyz0KCGIMPWzUTpCVBnbZs8j4vemc/2YqwCY87Y7Va1Ww2q6qhWU5fcDVSaNMbKwEKrJv5fN7oY39+VTUhZ7pMJpPB+4/QNrkfN5PHiEvFWXuMFUGQsQpeGmj3G2PlmoF2vb55zxeCgvP0UnM2n1jDtyUxHo9bBhn9gg6OM6bQ7PEiPGit3s9bWcAtXVWDhBSe8UZW189nxxWrhie3p+vRrrHt7e0GTKS2M49k1kFfnvFJJXaDTSZPjzyaTqf14YcfDlLNoUVV3TqRBsB5+PDhAJirqhaLRTvSy+cQOqvU1uJqtWouU6+FnlcDPiD70S/OrHoaA+U5ewLW63XzWhwcHNTBwUFtb283VyIZg9fX183d/v7779cv/MIv1NXV1SADk/XnrFxbpal4eVyeW8u15MOUWT0L7S6luWeh21LsuZezbPr9Wc89q3yuMa5f+IVfqA8++KC+/vWvt2sPHjyor371q/UzP/Mz9UM/9EP1Mz/zM/Xw4cMGWlVVX//612s8HtfP/uzP1r/wL/wLL9xuz3rqmcU9K6r359/8rAHOdVhrTjD0JKcrx0DkfvI8TOrfvRB7zyGscd2kJcM9XvCAvPuD0MxDZV0saJzifn5+PjiM1XUghBxjYUykqTN+C/ikuWkL+Ps6QtLzzX2cck5sr7eo0XyhuV9jn2Pi/tSQ+W8ha6s7+cwC1nPk+i2sE4iZC79qxjzq+bOLOBUxbypOAWeBTj1eb0dHRw1wzOe5ngymnrv1et0sMis//Pn0f/rId+YlY0u2NE2DHj+5z+YL6D6ZPH2jwWeffVZHR0d1dHTU3s3lPWv0jYQY5oikKidZVFWzwj3f7mtPziRvey31Siq+m6yyLClrenVuus986Dl51fK5AtcHH3xQVVVf/OIXB9e/+MUvtt8++OCDeu+994ad2Nqqt99+u92TBSamHB8f39kPCwm7ksyYvWdSiPi5quHL+LIOL7Cq23t3Erwc+O71g/vQNCkeTw+wPW4sPAAI37uFNkLQlhvaKG4dXibYc3XiHsK9gwuQxeU+0Xa6Pu3bZ+4sMOhTb/FAK4QX9/qtvix45nA2m7V+GSwTeOwKpK9owk6LTmAxHe2aos202s0rVjRsqXp+nFCUVj20o28p/Nwmz9FXlAXmC3cwNLFlbevNVu7Dhw+b4pPrz3PvV+V4XZH9SrJHuhlx6Zp+zJFddC64LF0MTOaTBCx/ZjzX19f16NGjZqFvb2/X4eFhzefz1l/6kDQ06FIXtIZOlhvpLbKLL9dDz7K6qzzrvrtA5q7nrDD9apV7kVX4oz/6o/UjP/Ijz31/CjsI7KwdExdB4GcoBilr6JQEQ1sBMKdjJPzmzLfU3GkrF1SCn8cA49ulg6A9Oztrf7SFAEfAWGj5jch+8SLFPnVcmXYBbm9vt1eOINjINvM7neivtTLGasEIAFdVc49aYLs/ZIul+ye3CXCPXz5Je1hstgQQQNk272MyaJCh6br97GKxGGxG5x7czGjkPvPPB8JakZpOpwMXFcIeDT4te8ZjxYnN3qvVqr3zykLaY0BBQeGB5liltshYZ4wlLUi/wcBjcObe0dFRUxRsbeEOz7mGVvQDC4g+GLAtHwAE/hx7M7/u7++3l7MeHx+3dfP+++8PTggxbVIumD9QbhaLRaMvbmLq8IZ5PvfANufaciNBDxlDGz1FuAdsm8AoFX4Xt2nZ6OsvWj5X4PrSl75UVVUffvhhfcu3fEu7/uGHH9Zv+S2/pd3z0UcfDZ67urqqzz77rD2f5Yd/+Ifrm9/8Zvt+fHxc3/Zt37axHywaGNIMeJdJnS4DFni6DhIYexqGtSQWsIPAaT7bysLyoA2DHdfS4jPjeX+MtXh+JxPQh6kirACuNPs9drt3EDh2Q9qNZGBEE+8pFglcqXl6f44zz7wwDw4OGv1IwU9aIVStbRtg7aK1xWOryhmUFtYWKj1XKP1w7NVC0aDlhBSDuAUT1xzXgl7plrQ7jb7BIz7B3LyCwKePxMUM8jzPPHAEFEBhUDCPE9P1xnPz3tbWVktsMC2hFa9Z8TqHbkl3QNqxSp5Jq4Zx2aLOtQjQrFZPN/SPxzf7zWazWTs5P2OIxNRSztgDY8USPvS17IvlQVqIzG9PcbF88trYBEzJw71+uPRkqfm2J8NepHyuwPWVr3ylvvSlL9VP/dRPNaA6Pj6un/3Zn61/6V/6l6qq6mtf+1o9fvy4fv7nf76++7u/u6qq/sf/8X+s1WpVX/3qV7v1opltKkk8axZ3TRL/U6NIger/vs/P99qgL9aQeSYZwP5+wMaWSNXN24Szf1mX3REIwwRtxy9Ms9TKTC8WjkEMTdr0dkIAf9bEe1qh+5KMjgDJ9GiAnLqJ07mdPKLLgtSaH8LK+6vSLWMh4z74N9MhBShtcj6h6cYzvm4lgd99PbVk5qbH+ynMPO6cL/NyCna+m45OwuFFilh/trLsDXBbWOMGVTIWLWwtuHd2dgZKQi9jMtee3ZJVNwpu0jCfSyXK4Iil9OTJk0EmKnWyX9ObsnsKm5+xV6W3FiyHenIsZVp+z9KztrLkGvWzr1Lfy5YXBq7T09P6W3/rb7Xvv/ALv1B/7a/9tXr77bfry1/+cv0r/8q/Uv/2v/1v12/6Tb+pvvKVr9S/+W/+m/X+++/XP//P//NVVfWd3/md9f3f//31h//wH64//af/dF1eXtY3vvGN+qEf+qGXyiikJHjwZ23VQowEhhRIjvug7SUDJaC4TT4bEJKZ+d3/HXh3OjSuN8CMPtodmNojLgu76dAGbW3M5/NmHe3s7DQNeXt7u21UtkDnz/tYSGWvugHMq6urppEyZvd7U3zItLS2b2HDc6Tde7/PcrlstL6+vm6CjxifrRdoxRwxBn6jj55Xa9EISva3MQd+aSaWiM/Sy2xJWx646SgkmHC/+22Livgl/fHJEQZDn/rBdScyeM1At3RnAgC493KuGCP89/jx43adTbpsO6Ce5XLZrGV4EfqcnZ0NlDD65oxHjwNr21mctvRwqV9dXQ3iSnZtM9dVNy77nZ2dOjg4GMy7rWPiutvb2/XOO++07Ser1aqOjo5aXIx+4io16Boc6Dftwys57pR9FNeV7mXzunnqWWDzrMIaoJ1N93we5YWB6+d+7ufqn/qn/qn2HRfeH/gDf6D+3J/7c/Wv/+v/es3n8/ojf+SP1OPHj+u3//bfXj/5kz9Z0+m0PfOf/qf/aX3jG9+o3/E7fkeNx+P6wR/8wfpTf+pPvdJAelpTarN2G+Z9Bgq7Fj3p+T999tSZIGXrybEFpzIjVBCyboeSFlK6oyyAEYg+CZ0xstfICSUe92g0aovKbqvMSuQ4qKrhKRybst4oPS0TULNLybTy3jFAi0C959bjhE70i/kwCBhQzEPpvvH8ej4tGP2uMfrjmGHynTVp9io57T01fVsj9MdgxVitrNkFRz3wBULRCoetiaSJgc8WvPs8Ho8HbzFgCwUKjpVHYqLE46bTacvepF340vE22nHyBifZQ+eMFcGj0M4xydFoNJg7knf4w7LugToxOehr4T0ajdqJ/bme0yrxCzx9ig5gazdq1uXvPcvLbme36ft6wOf1wn9bo1lHr1+uI9vqeXaet7wwcP2T/+Q/eScyj0aj+pN/8k/Wn/yTf3LjPW+//Xb9hb/wF1606TvbdJ9gwOzns0xYfrfG02vLbaam0tOQ+UMj2+SHTqGXll2a62mNGLgsVAEx2rIP324yC+q0CCh2K/GMBYaBl/ZSWfBY+Gya5G/Ekgjke4Om4wB3uTPsGuwtsnwu++b60xVnOtv1lG5Hz5N5hnH47MqqaiBifkzXtZUr82xPkaNvVkbMAwnqCeRug9/cjxSqtqQz1RsBjcB3Vqjp6/nIsbm9TK6yy91jSr6wkuR5tuvSVnVmmsLv7ENkPXiukzY9kED5tAzy2DZZRLlmen+9cb9IMb1T/vma+/S8db5suRdZhZvKJoZ0DOOuCTSDZ52bgMe+Z9fr9tAsYThbRJxysF6vbwWKadf1OzmE3y1MWXR2W1RVc1mhKZKlZWvBmyGzL7gUDVyj0c3+Lq7xncAzi5jnoEWCl7V1aGeLgvtWq1U7neDw8HBwEoHBpGq4VcDp4w7KMw7mFE09gc3a5dbW1iDDz1Y4Y7RAQ3MmgSAViwQI3Ki2muxOBRB7fGoXOHw2n8+b68vuVuolLgMNFotFq9OHI/voIvrm4v54nPQD4AUsDRDO8KuqOjk5aWf+HR0dDdzIPO+CJ6Gn8DhWjKIzGt1s8Ujl4+LiorkNzT/U/+DBgxqNRs0VnXN/eXlZp6enLWHImZgGOfqWsU3TeDQa7pFjXtJKNB/7np4SaPnhMVpR7pVN7aTs6ynutLEJxJ4H3O4q9xq4NqG2tcN0W20iupkxQY/i4D1CxYsntdKsx9ZRPsPC7lk59C/fxJouJIQfFsl6fXPaBe3aIkjXjf+8p4oF45c6VlV7UR732RpjzGTapevP7cPgaLbEKUh/5lghjmNCCVgul+2cOPpsevcSLKqqZRyyP8185IC/+Qzt3UkfjAllpMeHmVxhgWFtmmu4QzPNG3oACB6v3aDwqd1+WHGAmGO3yROmhzX9dIvbzVZV7cgpTp2wxXN5eTk4iixjnO4DCR5c9/uwDAhXV1ftfVhWDhmr58bjZg2RTs88pXCHZ46Ojtp8v/POO41GuFuhCUBPoonliQHTSTXch8Ln/nsOiJXZIk25lQCUXh3PSco1805+T4U518vLWE6vam1V3XPgovSsqaq7XYa4WHpuh7y39/smIPQ9aVoniLnvNrvves5M19NaxuNxC+pbu3cfrZ0nc2c/LLjsZ7ewtKA3cCHcscByzH5de9IwQRgLNoHWoM2YLIS4L+NfdnH1gvLuS2qUqYhY+EJHu66sKSNosz6XHg9a6FhTRqHws+k6NR16bdjdmvfcJdAsnHu/G4BJ/rFwdqq9aW3Bz+tXrBTSZwNZxtnol/ndJZWo3hqh71kvv/sgZoCrZ/XQjx7g99aw+559s1Loecr57Mmangzr8fiLlE2yN4tlyudRXgvgqro9Celb9vVc3FXD0whceAa3lpnBWp61VmtNPaZCC7cAoo2qYRDZfymAe4JpNHp6hiFWlgWmGZZFaQGIhm83CoKRPjpbkGw595+66DPava0ua6yArF174/F4cOIA9dj9WnWTfEDiBsBmy5ZiYMuMO79+3sI2LSQ/a4uC+3xwrF1rzpjsAdUmLZgxjkY3G6LzGZIDmFPow3fP713rIYUi31NQUpddrPBYClvotbOzU/v7+4Ps1KoaHA3GvFlxGI2evnXY82tB7LcgYz1Tly1O8wTKBf1PxYbivYLQziBKMg399IbxPD7NHhInSmTxGrZL33+OpTEf3Nfjp+SDTfLtWYDS40/Lt+SrTbxMXdz3suW1AK5NSO4JNTCkxoH1ZW0r/7I9FgIMzULwadqpvc1ms1vgwG8GQ1sQbosMJZ53bMPvDFosFnV8fNz64dMUOOgTy4nFj3sO7Z7TCxDq9uezAFjUpGOjgeLSY3E5LkDcxxtPqQtXGFajM+J4nlfGcx/v5LKAHo1GdXBwMFiUFhinp6cDN2JqxMwNbp+qaqcxpEJB332yBjzDSSXQ3WnpyVs8my5K6O09Tdy/Wt28SJP7vTfJFrSTMVIDt8JmlzWAyb3ePuK0btqHD8/OzgbbLHielyriGkNBMBhwmgvz6HeiIbQTNEaj0aBu5hk3tLP1AEArAaS28/xyuayHDx82NzT3en2wBqHV3t5eo6/3mLG2rHz25IoVMp8mglJiVx/yjPaYR483lVqUOsfIrJSktd0r5j2+b7IcrTwmqH0e5d4D113mZxIsJ7qXzp4uhZ7l1HOppOZjIKL0NOvUhFJ7cmJDD1BzMy6apuMJmV4M2NhKtEYKsPBsnoxubc/Wm/czcY9T/U0HhAexq+l02sZQdXNEE/11vxE+qVlW3aTVIxyhKQCAApABcubLgXHoxnXqzTlk/xuvpjdtbLWZdp7PHt/xLMII2kM76rMQuSt2ZiUK0OO/+5RatJ+xoOq5+JJn+R1lxkqXn7ei1vOU2MJy1qqVP2ct5mtf7Aq18KeNjEn1jjlLyywtQz7DH9Dfaw/+9LhTJpmf0z266T/t2dNh2WaeSh57HmsrS4Kvx+N7so/Zj7+vLa5kIooXsono645JsICtKaQm0gMuJs8uOV+3QErGsZVgAeX7LVQcN7FLi424aFs+gsdC2lZT7+ilqhq4BvPoHmviVTfaOb85mcBg4wQM939ra6sODg4GyQ301cIFADVYAVA5Bq4nffNYIbs3E/zswqXf3t+U/MFZf5zDSB3UafepNetN7VJwX2MBe3y4NtPljHC1+zoFTY4jFaVcL14jfKY9aOi1ZfrjXsMq7Vm6FtLwpRUvg48zFc17PlzXdMQSpt9psfDZdPTLWHO+bG15nab8sAeHsZiOSWN7adISS7mVJdcm1prpmvXyHOV5rK2eVZW/3wVKXv8p61603HvgSqazEOhlFFJgLjObF3rGarjHi8i+Y1sLVdUEJJsqsRLs009NzSBmwZzaNf3nXUD7+/uDU8vJygJ8OIZnPB43oe16cDkdHx83wWFBnxmPBgn331ZSpjubjtCOA2oRdH7JJe4mYlDQZG9vbzC/uf8HoIb+CErq8nu/eCZpkUqP68Xi471QPEc/U1hRhzMYrW3TnwQWaIb1zPeqaiDmP5QRDt119udodPusQgOSN+naXerPWJymDevn7OxsY/KI140tGRIZcM1ZCbrLCsFNnO+SI/HDVp2VM9zPHvv+/v5g/uDPo6OjgYKWsW+7sm3Jel/a0dFRy4glLpwbi9MSc7KK15tdku5XzyKEp9PC8TMJaFlXPpe/uY5UjPI+6GWgzvteptxr4OoRIJk9zdielpla46aySXNNN0rvt8w4S4ZIDTLbdD+p5/DwsAWiueYTOAjcs8itMdotc3l580ZiWyP8GaAdTzGz+qWOFkR2C+LSNMhSl92SeWSPrS9nEWJVuJ8+CgtB4P1M5otUdry4/MLM5KtMi6e+8/PzQSzKJ6Oklsl/W9TUz+9WWFJo9frlNtLlxrV8LnnMwtOACRiYP9P9Zp42fVEiEM5pdcE38IPplgWLzrznRAzX6dgbik+P/qwF+Cpp1XuW9ugTsVd40vsi9/b2BmvUfG2QZm3auu9ZSbbkE+TNQ1aAenyTf5vKJhDc5Ona9Oyme16mvBbAtYkQdyF7T0DcVdcmDbCntSSwuc0eE/FsWjQ9TQcGZ28TgsFuEJ7FwrCfPcfNIiP47cA4hYVtTTyZFVfe1dVVS2pwrMF7tDK+xhgMXu6zNd703fMMfUbL5Qgfg1a6IK3dptC1pZP0T2scIXJ1ddVNn04Bw/3Q1vVnWwhTa8gJthZUBmML/qSbx5qgnFo0ffAWiJ7AzZL0QvjbirOV4PmAL7MuPmN5JUibR6EtGZ7Jt3bvOTbFbxTTgX5k7NZ7PFl3tE1WZO67S0Dw502WUcqdBKxNivUmi+h5gKtXUqnM33JMvfIqlte9Bq4ECl+zVmTty9qCmcEuhly4FDOu3R5pmptRVqun76DqbVLNALqvW1AbzLC03nvvvUFcjdMdWFAIIAuM9Xrd3BZYRj4w9vz8fJCQkNYh1pk1UAsP3Hv0gzr8Di1r7rhqAE5cbcwVfbGrFYFHn1PA0lcEB3xA27gxfa/pbw3cNHT8gBiIkwQYH3PSAxHmyXXv7OwMEkcssBKgEHgphE0zu8ychJPWOmPkeW/SdfyMYq8Ec0A9CHArB4yZvvlEfLJVrbzQD7uzbdnSd7uw3S70pu/Q8/z8vMWAz87OWn/JrqUdZxtCG/gb68kp9F7nfHZ8GZDa2tqqxWJRk8mk/oF/4B+oTz/9tN2zXC4HcoUxer68rSLdnymbWK8GRnshoBlrwy68VNLguWwnf08e3VR6ivirlHsNXFW3N7Z5cWU8JzXT1BaYfP92l5ls154Xuev18UR+R1QCJ/+djcd3C7O33nqrHQKaGUrUw6vPvcGXBXp2dtYsGk6AX69vTmFIN2Zqxl7UeTYbgL63tzfYOJwCBgBHmDk2wB+CLS2t8/PzJnxIiwY0zs7ObmW0Ifx8EkjSy0pBWloWTD2rz8LBSgwCj/EYaB3f7PWFephnhBp1+zQIgw+Cl+dtXRpIzF/QinsRplU3GZo5btPf7jErFfACp6vAQ4AGWwVs8dKOgdxrzApFHmqLMKYvVh7wJozH48F+r+vr65bc5GxDn2LhpKlU1HLeoHnOr131R0dHdXZ2Vqenp4NkFStQ3prCvDpmad702vZ1K2Y9BZ++pdL8PCXv71mnd93vtl+23Hvgqrrbz58uBv/n8/PEt7Kt1I57E2Z3Ra9+T1xq0D0zfmtrq/b29loShIsBNa0RFiqAxrmFPn8vQdQ0yViJtawULAiz3GfTs2TR9BHCdokZQNC6GZNPH3cMzcLC7SUf9OjmOUnaVdWgb6lleu5smWSbOad5jzVg81YucPNHT+um/+k6Nm2yrl58BUuJ+ct2rJj4HreHpm+LNZWZXD9pZaWCkfQEwHMu6UfGgKGNrVr41XNhpSPr7bn8vRZMa4Mw64MjsDwfOac9K6VHi55F1FOKe/c/j9zr1e/ne/dvqvOuul6kvDbAZUZKawHB5t9ywZvJLAR72oM12KrbRyVxLfvjbDb6VzXUbDP2QRtYMvaVW6NngTq4Tt2LxaJOT0/ryZMnzZXn43c8FupBM3XmFJot9CbVeDx++gp7Ym52oZkmq9XT8wWJT7AJOa1X+uMNwCz46XQ60FI9h94EbqUhhb/Hk/MB6OfiQ9iaZjyTIAkPZUzHlpg3RCPYORaJPvuVNFjrVU+tTu4BGExr8yzPcOwWPJWWGe34ZZz2JGCF2CrwHJv/cW+x4ReecJKONxqn8ATgUukZjW5ccKa/EyVwh1kBMmiTocpbiX1ggOOx8KhBBw8B68rjzwxl+MCekdFo1BJ4ptNpffTRR4N3mlEX/XEdqfB5vAY+g7gVgp5i3QtTPKukTONa1pEATMGrYRq+TLn3wPUsBLc74S5CmfCpQWfqurVU6nTdrouFakZnYXpx2r2BkKCuyeTpyQvvvvvuIO3erhoAhbRsFvRyuaxPP/10kM5NnY4JsKj4jRR6C6mq4d4ib/KsujlZwH+r1WoANjyPsPTCJa0adw9uTU7iID5k5YOFnnPDcxbuac1QoImPeYIOvfutlcNbdt35qCADaAom+urTIazpG2ztcnaGW9K8lyoPKPrkB++1spCzheoYbGa6WVm0oHSWK7S165C6eZ+a6Q8/sgYAO1smVmzYkrC1tdV43iDIOuvFpRzTwwpDaeA3b4x3PYAY8+U1mcCB/DBIIhc4tLcXy0ulyclH9qj0vCFeCz1eME/mcy49MPL6fZ6SCn0vMeVlyr0HrqrN2YWbzOyc4BRMWV9qF+kquEtjSVdWtuX23G8LFBYQ2isM63d7ORCP4L64uKjFYjGIZaWG7IXm63aP2AWENmrrkfEbHHLxugBg1qZpk3Y3xSd7bl3GYa0yrQDXk/NqgZ2lp6l6/g1KWdzHXlZfCiqPIceX/31P1pl99xjcJiUVKM+hecZCLzMCUxj26Jhrgftxm8G/KCkGrrSE7Vr2s7bgbKVU1cASM88RZzUNnRWYloPXcU+h8Xwg6Plua51En8vLyzo5ObklZ3qeguTxpPEmQEge8P9eXX6mN9YcZ+9Zlxex6p6nvBbAVdVf3D0NzL/5WprWWbcnD02Ognuip/HYheBFlyDm88woPM8mYywUFijn56Fpk84+Gj09J2+xWNR8Ph+43KputEJruhY+BhZr1JPJpI6Ojpolw/MsUFtvy+Wybb62G9KBfPcFejj+VlW3Yiw+GNglLWMvegv5BD4LUMe2HG/KZyx40xol0J9xC2djOo5GFp8zyTJxgrpdDDZuz65D08OKBHMAGGXmKH3Gvch1H3RrJcnW96Z4EDQmbrZe37ihvZl3NHp6SDR9hs98BFla7Sh35keDBn1gi4T7Bt8amEmGYJy4nnmZKcC2u7t7ixe95queunXhERQ/K437+/u1tbVVx8fHgzrMx+lRsDuyR+e7lJdNf73S++15LS2v2VQW72rzectrA1wGoJ6QhpGtRfBnjahnNveI3IuPGJxYzCRF4DtPgPAztOUFOR4/zSQ8ODgY+L1Z4MSCyMy7uLioTz75pE5OTgavTKfgxjG9nD3FQoZOZIH5MFTHuaAtcRcWFBZib2y4Ypw5VfX0/V4+rcMaPXUDALhBDw4O2j1WKGxhVg0Bz/Npt5ZP/mAMtj75TFr1ZDIZHK5KP9LSI3vNQGQAsPvKFhgH6KbmbXqmEoTAnUwmTWHxmFFY6ENmYY5GNwfbcgILdHMsDXenM1sd06E9XHjmOUARHrIwc2yJ9eGTSuBzshITQPf29gZvMKCuFOasLfrOBmLH5wBKAJu1bHAj5ux4FCeYAHAkY8xmswFfsyl7Op3WW2+9Vaenp7dcv9lneN9xL3gslW/GaPdtxvo973dZRVagngf0PN88D1/eZak9b7n3wAVzupgwmxD+LjPboJQT5EVqRuE5uwMMPskYaSGmNWhLjMXh7Cg0QKyT1epp2u98Pm9WVm9/UGo+mwQiL3F0BiP3pyJAn/32XLtl/FwecOtFyHX77XNOvc8sadVbVJ47gNinzaeG2tMoPX6nW/eUjvytZ2nbfQQAVFXT1FMo+TPtpAUIsNh6SsXKY94US3NcK4HWNPJ5mOllsBDvAbktO7wHTgpJfks6w/uAJiACDfwqHfOvrSoUtd6ce+4AR697eBtQzoQWKwd8dpwbQDRtDw8P22ktjBWFaVMyEmPyWmaMqZzTN8//s0rOqz/fBVibZOurWlku9x64XEzcJF7PjDYz+pqZxMLX929yJVnjRNBZu+sxXF6vutnVD2jZZTAejweHlnL/YrFor0BnISc4ITipn2dTuM9ms7bx9+TkpN1nIWAAWK/Xg7hXL4OSUzVMa58t2BNyObeAsTP27OpjvuyehXYILLte3E9iKhZayQsIPLuAoa/dp24/FQaDm4WNXUwGN/MY9XmTsJ+xy9pjZY4Zc7oHLdgQrFbe7O6ylUXdniMAxJufXay9c6/dru6PY1beIuEzLrFy2NjM5uLR6GbPIPzqFH5nZ1qYW1bQBwMx4yR2jNJmq5r5AeT9RoPk9/H46dmGT548ac9lpmSuYf7nmvTv0CYBCF7zfNxl/fRAK+WoP6csM20/L/C618CVFlJq4Nxj7cWC1felMLP2mBOHn55rFkAWDnYhYMXYZLYwQSuz62Zra6vee++9Go1Gze2Hzxytcz6f18nJSX388cf10Ucf1fHxcRM87qsTHyxUoYm1fqwt+uE9WVU3r5eouolB0X/TgsW2Wq1qPp+3U9RpCwsMBice5tTr8XjcYh6eo8vLy3r8+HGL/VXVwBXrJADaQsCZd3hRIePwnCH8MnvSi9eC1bGXq6urlp4PoOTxVOmusdVZdQOkmRnHeDy2nhKVf4wRxQeesOBOEPFc8aznhHtsRTBmTmLprUnozEZ4+odCZrpAY1yXVgbMx7jjAAtiU/TBfGwBizsys+UYLy5Xb0c4OztriRXz+bzFv5yBiJzhzxmzfgPD9fV1HR4e1jvvvFM7Ozv12WefDZQhKzZ+U4MVTysavc3Kniv+UgG5q/RAMX+/y6r6PK2tqnsOXBQTzJZQVd/MzgVt5nddPU2iNwGpidlN5j76fnzf1srT3B+Pxy2G46wq3ig7Gj1Ngnjy5El99tlnLX3cAi1LCkv65hiWg9NeyLZALNA9ZopPNmBRsQ/G7j4vnlyQdmVRR9KWfTFObrHgTdqnuyVjPL4f5YT6mB8Dfbpj/DnBibr9rHnKgJZ1eC5tDTg13oqWQZbx+c280N9bDOBj0960s5Vpt6zHY3elkzvyZPStrdtHDwGameHHmKy02Z3ptqzEoMhA24yvwGcoeFx3bJY+en6ycC88X1XtPXO4puFTYtXmCeYbIKQtr/ke/6YXJueDvuWf+fd5S2/cPc9I75lcW+bxly2vBXBRUvhb0N71l8/nZPSEWgoq15Gp7z0hdnV1cyBrVd0ShAil2WzWND7q297erv39/Vqvn26oPDk5qZOTk8FmXoqtuhwL7WKVGXwsVCwc3Y90qVhpwFpAyEAXJ14YNNKl0RPcufhs6bkPm2JFnj8LfupKd4ytagNsugB72ujzaphpgTqRJq0n6nXdnmu7E62pMy4EsmMwBhBbbfzGdfN1gmImN/l5gNF9t8JQNUwYyPgQtGbNGCSsROQzPY9H9tPuMs+l16IVQcDPlrFpnUqMY1rOlIVmTsQgHuesVJQ898/9TYBKvvI6yeeeB7gsr8x32Vb2Lfv0qiDVK/cauNIi4r8Xf2orMJuZzMKe7wi01GZZhLaKUqCauatq4CZwGxasLCQyvXyUju/H4jo6OqrT09M6Pj6uk5OTwSKydmsgrxq620i1393dveW+47PTtHMh4NZxTM/uJ8YKwJAwwnh8moPHSMzAgjTdf/TfVha0Js5Df30gMoBpa9EF8IdmxBnNax4DLreq28dCGQjRvHGNbtr+4H77QFXimQhwMg63trbaCyydoWgehYZObQfQUqPnAGH6zO9O24dGnkfzB2OB7rgA2eCOqxFXotfW2dnZIAPU6eNOQiHrDxDE3Qt/cJ+BAAsFN6b7bM/M6elpo0nGkHxvKkFWyB49elTz+bwWi0V7v5fflL2zs1Mffvhhm9P5fF5HR0c1m83qwYMHdXJyMshmtRwyj1n2OAaY93oNeY/c8ypYd5VNoJQKi3nvVdu918CVoGRtx9e4t2dym/nSMum52qpuH+dkIZouMwNCMlrPgqBOQMGLGiHMPqqPPvqolsvlrcM6DcxmfsbpWILB0UFjW1AGTbvgDGzch1vEi509Zk6htiVA/SxiADDpimWYILq1tdXiC07DTw3bAsZ8YNAlBuV+og17rtIFbUvCY0przm4uW0TQ3wqTrbu0uh0PTQA23/OM45IIOfeH+WYeHGukDYCGNG4Ld6ek0/7V1VXN5/NBfa6XmO+TJ09af9PFS1sZj7EwtgKA0mnlLwUo11Ei+POLTCl3ueo28a4VYpQbYsccp+Y3M3MvoPfWW2/VBx980PqHp4Kxmbegiy2znnWUVo+vP09Jq4s6Uo726s7fsp6XKa8FcCXxe6ZrgtKLtpMARz0JPtY4s8B4rqcHXFXV3AY8x+KbzWYtVsS+D2tRLghD992gheBw/6ylU4fH6d8QSHadkNll0EuhayFkgEsLOulpMOEPFyd/jA8hYOBHaFlb9lwlEHlODKLcm/yR9+dc+H76YA+AgTz75vRpAzl0cr97XgLmqGflmZ8NLn6bAVYev7lYcNtqhh+cPejP9NMJHE6Mom5om8ppKos8Y8siAZY2+T150POSQJnzyHcsuKQl410sFs1ivby8bBm7ftVL1Y03BFBLmdOTW16Tm+7pPeP/z7rPxsCrlLsUgBct9xq4KGZqZ/RUDU9DSABKKyetLN/Ts8isxfdcjY6zeFH0BDIbH3n24OCgDg4OmhWDT/+dd96p8Xhcn332WQOEyWRSb7/9dnNP+CQGC7vJZNKSOqqqnTbg9GzucwowJwSku8bH7GBlIeDsllmtVoNTI5zeb0vEdEdT9TvMRqOb9GZceLwivWqYIIIGDW2daGHgTFels+ly7hF28JU1+ZxTx4dcv910Hhd12AqwMmLgZzzQyW7S1N7hkbRS7Tasqq61gaUALXkO/rBAZcNtCmJvKHZq/vHxcXs79+npaQPgra2t5lqz1Ul/eP76+rp5DNLlRT99ZuR6/TQr0bE9Kz3mla2tp+/QyhdZQid7I2ytQQ+vp93d3eZ63NnZqeVyWYeHhy1R4+TkpLmEF4vFwCWOd8CKjb0bbsd8aWW1p4zl5xexurxGes+mAkqx7IPnXgW87j1wWRvoaRyeZFs7tkbMGDbFM16TmnUCobPyqoYvuGMBWlDbHebP6/V68BZft4nmfXZ21lxwpF5zmG26wng/loEp3Wi5KCiOkYxGNycX4BJkfBn3mM/nAyHp+hwjojAH0NRaNP1G2AAgjseky4TU56q6BcxWEgAHWyXUZ2FgJcenJTBuCwrzQPIHfbTiY8GSVi0F/oUGpqcVCKfnm7b0rermhBBibrZULOiqhu/GqrqJIdInx5vcJ4O0BR5tcVzZePz0ZJgnT560I8tsUZp/fJIIRyVBP1vigAexWvgVMIcWCdIoRcwffIGL1PE46EHSBUqDx89aPD09bYrO3t5eHRwc1Ntvv11vvfVWS9zgrQn8HRwc3IonWt448cM84jVlnqoaJmzAS89jdbkO18PvmwDoLiv1Vcu9Bq50j/RMWmtFKSj4/Vl+bJesPyeV53sM4XotlF0X3wEXu1G8z2m5XN46HsmuMrvT7E+nbcCyN44UpP7NGVJOW/YrKqpuLN1evCbnZdPisTZJerOBy997wt58kdoldduy5HqWTc8yT1zvuSA9jz1eSe0zAc/zQF1OUEng21RS07U17DqgCWAD/zkOlt4FK3a+L12rniNcZ7aM8C7gkgQMvI/O/1MpMF08Tngs42T+3XSnT1ZomfNU9DJ+5sLzeCHOzs7q8ePH9e6779b19c3p9laQAZO9vb06Pz+v4+PjLhhtKsmjaXG5r5635ymbZGTK1J48vKuulyn3Griqbqel+3PVMA7Tc0vBWHdpH2beBL2q4bt5qurWES9VQ4vCwWPqscbo7wS3t7a26vDwsJ34znViEQS8DW4sLJ/BZ6b1m1n5c/zB2i7XLy8va7FYDDYOI+DyteLQ1q5T6GUL0lqg59Ggtb+/3zZ5VlV71cnBwcGtubWQtrUKDSyI8jw/z1laTxbUduv6vDxvnq16KqBns1kdHx8PQNR1WEAaFKzc2LVFfwwiWJK2Gl2XQcSaPPwKrVBoqMPKkdumvwBLrsHJZNI2YLtd+B4FCOVnNpvVZDIZHFeGqxNrCCvNsUvo4xdBen17Xu2KNT/kGkDx88nwdqn6PXJ4B+yONXjb+huNRvXRRx/VO++8U8vlsh48eFDr9bpZdFaCWOvppYH2aVW7+H7XmcqQ18VdJRXuu777mU3K+4u6KHvlXgNXmsC9ybTm7+d6xE/tKQlsF0D6ol2/tUEL955/OP3lCENrtywkXoKXmXzWIHGjkFln6wxt2sKIRba3tzfY5GthShKIrarRaNTcSPzZBZbvtrKrxUkndsMZNACl2WzWNnMieBjfpn1WjM0uUsd08t1kzLX3nqUGiUWA+8nPYPX5fVe4bS20EXK0aVdyKlc8YyuLOU6AYN5toTAf8KLjnrRrhQEQIUMV4OLAWrcNAJn34FXzB8DDc47BwdfX19e1XC7bXAGQ9AFevL6+bini0+m08UBVDeKCgATHlVmZWCwWgxhmWttO2OC+i4uL9t4wAy88wmk2vbgffAO/XF5e1nK5rA8//LBms1l97/d+b+NjFCjo5zYsL7wufd2yz54a82DOVVqgd5Wewt4DvRfxArxKudfARfGEGOk9oUn43rUsaVr3XAF3uZZgmmQY//G8wcuuPmveCH4sH7vmqmogFL1AfW6ax2+ty25D7qEdTpinLZ9BZ8vO9VKfQZWx2BdvsDad7R6ywLOyUDXcpJpCiDkyEOe8ptuppzXTjq0j19WLLaQW70QYW0DQw+2bv+hP8rQFhHnAe6AsCFPZoV14lP4b1Nwvz6MBxfRkXGndZtwMXkUZ8ItGd3d32xFQvfYZk/eEmcfNl7RDnxyH8ljNh+ZjeN2A4XnBG2Aw6a0tz8XFxUWdnZ3V8fFxnZ6etgQjXIKOvVGH58R9y/6kRyXXtIt56kXAxeN73rLp3lcBtdcGuJIIuVBSgPbAq/ecSwqQXh/MbOnXt5WWcRlrSV5sZGZV3WQ04ZYzg1cNj5qyAM+UYLu56Lv35vA7bsHlcnkL5LC2DFwW1N4n5LnB+sJacD9ow6AOzez2TLeXac68YNVZ08+FzncUhQRbW1/OJoOOCGpbKDyPtURb7o8ThagvLc60Il03dLDA5H6DfCaduC67d3uuNfiM8ft5+NpCzO4pW5GZMEL7Fvqr1apZdtPptJ1p6XbStYV7DRciVhe09fqwq9Xjt2vbNLM16axbAwdAOB4/jSGTJJWuRysIo9GoJWLM5/P67LPPamdnp9566636xV/8xQHNeL5nyZkXkpcZr2mVCrb7luuzV3q/9yyqu2Tjs+p70XLvgatnmppQXjgwgid1EwBZ000NnYXgRWohhYuGheI6LPwtJNwuzIr7xprl2dlZe0kk6bMcLmoNF6HDomZR7e/vD7LAvNh8ujVnH56dnbX+AlhkMJJejFbqOANg49MNcAElLdmz5tdD+Blr+biIrLFbANvdizDCvcR4R6NR9w3RjJP+e9Nxvs8LOnn8tMFc+sBV5iBdeGRmwhtVw4xGBGgKH8bIPRS7iI6PjwcWn+M75gnvG/IhrrThvtI/EgtYQ4Aup6HgioUPfXIGQDUej9u73lhTk8mktWOXJBt34QX650269BtesPsWHvC85bq3BQpdiTNxT55KY5ceysnp6WkDX1vj0Jo5/eyzz+rtt9+ura2tevjwYVtP8/m8pczv7u4OYtFWGKys2So0j1pJTpDpXd9ULCdT3j7vc59nuffAlaWnofLfVkFaQ54AM6pBy1pgut0skK25Wxty4N6Tn89b80pLkRR43Crusxm26iYIn9qq/dsOuKOFohHm+4GqbqyQyeTpPiq7aKyJ9mJvdgcCQnaZoQA4ixCXFAIuFZNeRiF1AXwGUZ/+AHihnWMpQhcfSeVjo+zCog8u1twZO/22tYWyYA2Y+nqbY9O1bI3aPG6LOi3/XB8G61TSACDup03ve7JlaP7NZ/hMO94bWFUtvgXIwbfMr13C3OO6bZ1ZIAOCnlsLfq9H999ehHSLMxbWn61+xkKs0HSyO5bM3OVyWbu7u/X222/XJ598MnCBcibpycnJQG557lK55vrzFNZkrp9nleQVA+eL9uFly2sHXBQTzu4BL4JcwPnnCfH3BBM+JzBWDRerswhTAGdswRs5072BMCFrK60AxmMN0WDmz+4TbkFeQbLp7DPut1XiLENrrKa3tb/t7e3a29trzxOkx1LhZAE0Wr9/KfvjBWT6e24AqOVyOXi7rTfeWrjx2e6m5BVKTwA6vmSLBLCivXw9TtUNcNn9CBCbnwyi/u42e2661JZNK4OQXcx8xzJKRcVrJQV/AmxVtVMj0s2JQsTeKMabbq387jGZ1j41Hvd2zmfPu5IxTSuBPaXTfWEMTkgyfzjeBc9zIo63mEwmk/aaFmhgnjDYbgIfz33OPzR8UeBy+z2L6i5r7HkttWeVew9cSYB0F3nhGnyS6VPgwyA9Uxym8/uNvFhhNgMYn72oqI8NxBQ0NSwruzVYTN4wiesFYLi4uGguCbeP5cKJBU5aODk5qUePHjXXCG4/AuVuHy12vV63jD8EfdWNOw53ngXT4eHh4B1eaTlYGPuV7YwN90lmXaXQIZuSZy2c+d2C04IN96t5zHxksDFAjUajOjo6ai5C/5k+uBRTa8WFCA2pG4WFcUMbhI6tIluDtuJ7IGQhjoXv1Piqm/gf7dvVTVsJCMwhR38BHJnIYyXDfMK2h8lk0tydCFfmDVp5IzpWvK1pAyDgb+WQsXqvoy0oxpcyIxVMv0LFBwMfHh621w1BQ+h4fHzcFLTj4+Pa2np6YDKn34xGo/YaFMcLk6e9zlNeWObx3UAFr24qqYwmSPfu9XfT6/Ms9xq40qrqaV9Vw13yTJy1/x5wpTbPdbu2UpNK0xtmdzqzf08LJoGUaxlktiCCeZ3GnNlcFiYsdhYk6bzz+bz1z21Np9Nm5bFQzcxmVoOyraO9vb2uxUQMAjqYVgBHJjmYXigCjr9Z0UAYOiUbIelFb5pbmLJIrQylhe1Txquqvc4d12bVcJ8ULjFr7pkiT/3W8J3IAE/4foqVgQRF8wb09BFaFmhW8HzNAtK0MD8yVujsZx3zgX8ck6JNu8vpe1pGtsjN8z4DEAAlvss9Z2dnjdd8tBXP21WfVglryOPIefD36XTa9lfu7e3VaDRqHgC/7sQ8AW/7HWpuK8E/aeN12btOnbnf1MV1uC3/1pPBd4FUj2dfptxr4Kp6vuCfgel57rEQq7qdOprA5eveHFl12xw3A/Um2Yxh4ZoM2nNFWSi4rnQZUQ+WIxs+LfR5nmxH0yUtMIqBzGDDoaIAFXVZAOfiS8sUGiZ9q6q5/Lyfx7S3pu/YT9LLWrWFtBWEquELH1O4IYQstO1y4vlUWHpKS85D8od5g99ssfbu9TgM0j2h5PFbwUvwMt1cl92d5meX5CfawZJylmGuI69Bz68TWzxf9I3NzwatnqIHLa1QmRdNe8sO/4cXuMZGa6xogCtdl1YuErh6oNTzPDwLRKr6MeIXKT25mjR5kWeft9xr4PICSx9zLnz/h2koeT9MfH19k4pe1d+HdZd/nGcSjNwXj8UM6BiMQdGaKy4MAIB4DQKjJzxYjFdXV+3lk9SDZlp1I3gR7rRjF4V97l5AAOlsNmvWFteJX1iYW0Cktkg7aJ/e83N19fQcOJ/OQV8MDFgvSat8a+5qtWrvFzMoYZ1S1+Xl5eDdT9CTOfVmb67hSqLvmYyTSslqNUy+gabwXPIrhcQJNkYnHeBJz6mL+5wgZ/5D2cHNDS0MxFhzPtfSFhJzS2yHZ+1SxcJnbtwfryXoj2vQ6zVjXDyXngTunc1mgzWPZefMUmc82or1fKTFNB4/PZcR3uUYqPH4afYvZXd3t43DLmHknPmFtrx1wXPQs75SIXoe4DLvpLzrydq7yvPc86xyr4HLBDc4pWaXft9e4Xm7pvxb/lmwVQ21KhhiNBoNssYyq9CfU6hYGJIdR7YVC8UaKK8Gpx/sxGfhozE+fvy4CR3G6/1jVTdHVkFHx5Ny87FT4HE7EH9zKnhP0/epE04KABx8L+CTbj+DH4Bi4cgCT7cVffXJ5BwpxFyRVYmwswBx26nZpkVI3+AXW5MJ2LZYLBBsCVjpygQbeCj7ZCsvXTUek+fJ2Y0oUi55BFQvA5HPtm4mk0nbgrBe36SRW9mjzv39/aa0nZyctN8NyvCFT4JBkXHiCO+uM21QDpfLZZu709PTdn4iKfhe57bkoBP8bKsengWAcBMaTFlHo9Go8Rv07ilxXhPmEcu9BHX+wz85x3fFuDwf9MF8ZAt4U8nf/r4HrqrNprEnnUWVllm6PjbVm8xjZqmqW0KKNhJYc8I88RY2FhR5Pl32padRufAbAMU5g7mPi7YNIBTHL5KJadNglVZMuq5MH/9lBhx/tmA8ZguIdJE6JdruFazt3tzmvJmWFl6eTys69CeBK907LglQyTMJXvxRZ1pU5qGexu26N5UeANrdnEKwqgZzzDWfVNHzUCS4MT/+TNuAHhZSuk5pFyswXbvwod1+tp6soABwxI0NvF6vVoQ8BpQT04I6XB9jY6xOvvD3Hg94HlPW+c/96xUrv3fxwyZ+6dWd117EGnvecq+BqwdYfPcCQ6txsL3qdsA4Cd7zofMdrZ56egIQBkzT2nXaOjMDoanhahuNbtxVlPTL4woxI9MWJ01zOC804RlvQgUkbHGldpnuCw7BNehaI3YmIP+9QTvH4zGyMdPPu9jVw5hx06Q1cX19c1wQ82gecgac63NcB0WC+B/3oWhsElief7tkral7r5pd2Ba0KfQt3MyTmzT2Ht+mAMzvKewtuKEX/TQfW/N3u1U32Xenp6eDhKFUemwRn5yctIxC+uHN6E7SoW3vR8SatiXFc1hu5u+tra32bryDg4NBhiFJHLTlMfgcRcfHzs/P22Z7A7K3nmSfvCb8eRMgZT0JZmmpPQtQrJD12t10/S7ASpn5ouVeA1ePgJ7UtKSsqXtxebGl1s9zCLXMCsxJg9l53sIZV1Zqpj2NG+HPa75ZkPQRq4n2fYhq1Y1L4/z8vJ48eTJ4J5EXN4uUkwz4nbqdnu2TD0aj0SAVHsCiWHDA9NRHTIR+O3UdOnCixWQyqYcPHw4yv5yFZ+BnLwz7wCzcTQ/Pl/vmY69sZfAMzwE86/V6cIKEY1A9dx0lBTjXDAwW+LSXChB0TkvIvJxxD9qywKXfTnwZj2/eNMCznDwxGo0Gr9UhFtyzqNlryH0uPhXFVpItVMeRPDYUGdpxLHpra6uWy2UtFot2L2OazWa1Wq3q0aNHAxrDuyQQWQacnJzU2dlZffLJJ3V0dNRitzyHAmietNsOpc1jgxasJWK3ViTpO7xiZdKKFGvCChnjsksylScr1c8qqUym56p3H+08T50vWu41cPVKD8mxLpLAdxG8V2fPqnMbPddBjzF6baVFNp1O2wbdzHrinp5v2guH7CkfN+TxmJFtQbHAbAm5Tf47JsQi4PmM+QEoDuInuCHc6JfT4inQwuOlD6lY0H7Pqjat0017l4ZrIEgeyvlwGrwXefKdAclC23zj++jvszRvP8dJIOblnivb1kEv4YI/p+qbf6ADz7mvpg/WJcVZoeZhC122Z2C5JMA7duT59GeAGLD02BP4oSv3MgZ4dDqdNkXM8wsoUZeTV3yPLZ/VatWULQDXc+h+9uY9+SO3reSauMsaetnSk5G9e+76/XnLawFcifw9k9iCO0GnZ2Fx3XXkZz9bNTzh3Zp3T9Cl+Z3+/v39/bZZl3ZS8GJFpOY0Hj99BQo7871BNBcy3y3wLWR8D0Fm4m5o0xRnL9oS47tBODcAs7/JG6+n0+kgtZk+OiOP/xk7cd99xFMKOgsrCxG7WGzx0BfHeFyP+RCr1XPb4zdb8I6huKRg2iR4NilnefahBbwtb/Mu80I7m8Ag++74Dv0jUcLKAa5csgoBo6S5rQV41cBFH1B07EJ0//J+eyiwJg2W0MKuXDYTn5+f19HRUe3v77c9e5kJzJgBJEAzraX1+mnCzd7e3iDrln6nopEAwVhoc70eZtHai0C7PSXuecuz7u/9/rzXnre8FsBVNbQKEsjSz2vmTO2fBcPCToHe08h6Wrq1xuyn4x4IZIBie3u73n///eYD92GofLdVwikYjkHwplWDFv3zWXxekLh70Aapi8xBDrc1DXKR5Xl+ZEN6oZLRaHCk7+fn5y15BP//dDodnPzNK82Zy/39/UZ/xydYtKvVqtEBwCNGYq3e/RuNbk5S8BylcPYpDBaGthahAQKwp0ClcIUe5ikUAQDFWwp8ggquLniE+4nFeUzUjduLuqx8WbjYbdWz1uhDKjBVN65Qru/v77dnDGi077knI89HdB0eHtYnn3zSvrM/yvzLwdMGDW9sh2/Shek1a7ehn7u+vq75fF57e3t1eHhY7777bvMqMNeTyaQODg7aWufkDK93rgPg29vb9fbbb9fjx48HL4hFXvkP2nteKay7lD/0xfPIWnO26vOWtAZ7Bb6kpHx+mXLvgcsLLK2h3r2biOWFZ5DLjDjKanWTPp6g6PqyjZ7GTx0sVjQ0NCcKi5EF7KAz+3cuLy/r7OysAZ5jB1g6pltPQ2cR4A5EANsasEbc07xxnzBmx0OsKCB8OcGDOvNAXZ7J13BQAFL61XOL+jp9AqyhC/3xd1sOaW0zZoN1uqp8lNN4PG4aueuyRbMpfuD2Eyz4zVYlsVErUmmV0id/Ns/4GXsIcJ+hCDmGlePw51yDXjcAi8+97Ak8+nJ0dNRoScKMt0IQC7bFwnzafQ5fOR7kNm2RppwgrnR1dVUHBwfNlcmLMZkDFDXzk+VHei5YP97H5nGYR03TlIH+ztz2+A2L8XnA5EUB5y6vwMuWew1cJkBPmOQ9+ZsXv4WsgSiF06Z2q4ZCrgd2Xri2OLgOcDm5A42R+yzQ5/P5gEnZj4I7wwF4W112gzAejx1Am0wmg1PZKU7Y6NET4PKCZCymr4PyCA/utxZuwXp9ffMaDtMV2nh8qYD0AM3jNs848SV5qKeUpMBIa763eBP0q25v8Ey+s6KWwikVpzwM1/RyZqZBNMfMdVucfLcg7yWI8JutLfcvFar1et0UMNPD/fTzBwcHgyQlrNvRaDQ4ocV10ZbXh+cu58Pz575YwUFZxBIkExg+ZY+Ys0RZX1bszKv2biTP9LxFyVv5uScfPT/b29u3YmvPKj3Z2FsbPZr27nuRcq+BywsuhU/VTfYUDGAB7gywFAw869hFarf5mefTd5zabAbwfW02m9XDhw8H48Glh+Bl46ZjLtfX14PMQcBrtVoN9p7QHpoy94xGo+aWqapm8bE3K4Uo9PDJ82iWVXXLxZnp2lgduPA8V9CU+3BjICwBU/fF7lvmwSeCsHHbi302mzVXEnOJ2xC6OknB8RLcgJPJpKVn0xfGwCHMVjr4zQeyklJN3VZ+9vf3G/18gkpP6fj/tff2QbplVX3/err73n67b3NnmLkzCATQqIQXDSJOrPAjQg0gZanwhy+UgrGgJIMVxRgKyqhoIolJJSktgv+kwD9EE6tEKsQQQQQKHVGJFAJKCUWcAHNnmPvWt99v93N+f9z6nP6cb++nb997B8aG3lVd/TznOWe/rL3W+q619tr7QF8UHn3I0CJ9cUgSJcp47bWY163o6bNDUhnGwtNkHldXV/u59dYJr20RybDsMocAFPxrA86JPAAYz/GKHhsto9FocM3ACn0ypItXBi1S+T744IN17ty5OnPmTL9xGv5hfh0qzBA6hiuyhFdJJiTyZD1lg8y6h3ptGLbAmbVAr0lfq8CDqW8ngZfbRf/crNd1oIGrqp327t9MsJw8K0mE1FabGSs9qbQgUCRWvtTRsmT5TP9vv/32On78eA9UtnRJtMiTHEiBXV1d7UHApwAwLof5zPiZgYVQsfnSYA19/CK8XHA2IxvEEuwzld90aRkPhPQASIedXPDavKm6qgYn56Ns05DxXKTRQmmFu7xmYKCjLXsn5ievW6QVbB6jD7bWHWpknCgwzxuKlrrxwHieuu3ht/541n3jUObsN3NIf/yyT/Mba4kpkx6n14J9Ygv1k/hw7Nix/rgyQsnwLH2wwuR35IptKtyfb2poeZYAv68zpxsbG/XFL36x1tbW6tSpU4PoA/zk55l/DA36C29yigtr2QleqU88J+ZlzxF9gPYZVcniejOUyn/LsH9zHWkM3Gg50MDVGriVZ1oEtmz53ZZua4JbLnarpOB6ErOu1rPHjh3r49qZQeXwhhUs1irrQ1iiaSk7LGSF5PHiVeDhkGiBAki6QS8rXoN0huvcH3uMBoL0lq0wrMA9Hx4T9aLgfIQVXpq97TR4kh4W/vS8/QxzZN6xZ22vKwW81QfubSkI0xRaWsn5/pYHCT2s1Lz2Ytqm4dUCc3/Pucw+JgAYnF1PK3zXogOAyluCkQkAumUc0Qe/Mge62+t2G563Fl/bC6ae5eXl3pBArpLfDdI2PixP4/HVNHleEcN1z5PH19KJk3SafzcI71VaenBSvZP0cz53I+VAA1dav2bCqmF6dtd1g/dUpauezIrnYWXmsGIKL6neMC5WLe9eqhpa51YCXtui36SPZ5KCPa7l5eVaWVnpPS73zYvSWJpptRLuwup2Ukgu4to6TMXOd4de7P3Y4sx7nHFFmIvx47kQsvR7vDzfFnoUbabOp2eJYvM46KctUFvAGSL1vfSbYm/JdINvHC5xptvCwsIgLGQeMwCtrq4OFv0TLGzZmtbT09O9R2Eeb8mVDRD3xYaHPbGq6r1pPFvmCjr42fSs/N3GCHPK4bPQj7qRE7Z/GMBnZmb6zexXrlzpt5jMzs72iU6E1qG5s2UBHfSHle2VK1d6791GAJEQoiR33HFHHT9+vI4fPz4Ac5/HmYle8Ofm5mafhEJo3wYPtLIXnQDmzykT6L5rAde1DHfzThoL/nwIXNXehGuBddiN39KaqxoSNy1gW8d+b1CGCgx+XN/e3u5f2Ng6Imo0urqOcdttt/VH0BCq8PurEKTl5eUelC5dulQrKyuDdST32UodJYcCQMicLeiMLCsPAKjqKqOzwx+hQqhXVlb6ZzhVgbZMczbCIrhOD+eECgN81VBRpufjOfTceV3BYSBCLjwHYKPMKIwlrVHGwFwBpOnpjUY7m3S9DsR9BkyPbW1tbRAiMr9Yqfn4L5QtHoQ9bK/dGOCSHx3edJvcQ6gNeszNzQ0A0h66kyUso4zf3jr8zbjdP9ZK7WF4bD4NxUrbkQnmmzbx/qgHIPf2g/n5+Z5Hq6rnZ3uWjMe0pc/O9ltfX6+HH364VldXa21trc6cOdMbVTbA8qBk6kEm5ufn64477qgvfOELPU2d0IHMWOZSVlw//WYMnvtrFYPYpGc8vwlW+21nr3LggWsSAFF8LRcx/VzL2rMQT2rfFk5a2Tybayk8yyI1+2jcLwQhw3RXrlyptbW1/hgaipV29r3VPzYRt4CLvvi0APrsxfy0ElljQ5E7tDEp3GJDw/chhA4NorRNj0yXbll4GdbgmkEcWnhe7b25Pn9v8VAaSvb8rfRMk+yf2+F+rqfHmfPHvShI+Npg77lJpdkCVB+zhadmEKKvPj3da1qei7S0k3ftffm6eTBDsT5hnuzUVjgSHrKitndO+3jZXdcNQLnFW/bwHZUxeNL348eP90CZ98JzrE36N8up18BbJa9PAhrTOL3uSSXbdd2+1gpz+/mbBbADDVxZMjQDASGUlWoqGgtbggzFwuIQQoajUrFZILiPBAheRY+g5foDJ0ng7V26dKmWlpbq0qVLfdJBHjpbNUyO8EZZLEy8DJQAB5WilEjwIMae4D4aXQ2REb6YmZmphx9+eAA4Vh700S+sNE1Z1DZwArAJED66yOfnWWkzXs+/E298zcYKv8NP5ilo6sNVExQ9d/a6HYb162kYK/32WPhjPuFrklX4zubb0WjUzx2bXRMA6KeTDAjBETplTuwdwu+uw0YZ9XF+JbycyRnMtQHK+7YsJxnqN11mZ2d7ACf1nLDnpUuX+lCizzOEn6t23orA2LiPCADzQKjbfOHxAiZeVqDPjNWh+q7r6rbbbqvTp08PgMtrc6urq4MTQAwOs7Ozg20jng/Typ8ngZU/73eNy2Pjs3nWPNEyXB6pcqCBC2F3qnKGpnIBNJ/P77zXiu9eWCb7KNcp0orI3e5M5vr6+sCyv/XWW+vUqVO1sLBQS0tLvfJwiAUBI8b98MMP1/Ly8uDdPQZdhM2M7PAOigm6GEQcToKWVvJVVzP0UGKsa62urvb9QWF54dsGBAegorhZA9za2uqFf2Zmpk9V5xXrzCV/1J0KBPCdnZ2tlZWVPmxEn5nHquqNGQTNAmhgoj8Ox3gd0xa3+2VvzsqC9RYrY/iGBB0re/iK+tnrBO8tLi72bWJQ0C8bapmNRvt+F5VfIw/NxuNxLS0t9XV4UzzX8CLsDfgsQsbkdWAbCz7CCbpbhk1DwqjQzin19NvrX5Z7QpLwCP33WhzhbIcHMTYZM/+JXHjvmZ8bj8eDtdmLFy/2W1HuuuuuflwcMu2+bm1t9cYpbT72sY+tL3zhC/0ZpAZlG2CeH8+Bt9f49/2AFjzo+yeBdavOHN9+gbJVDjRwudhK9rUWyhvEWsRL4c42qnaswdZzk1zzqh0wRLm3XqIHuNn7u3LlSi0tLfUAwb0oPK+bWGmyLoVHxXOTYtqAN+OZnp4eKGjqR6gQ8HwPkpWyLeVMOADUoIf7xm/2klJQfN0K221Ax1TGKeD2du31mVfw0ADZfLdXjrfFXxkqScOqdd19tReZ/fR9Xj9sHQEEfU0P97frdl6ASZuZ4GOgAZDsZVsmvMbm8XkeATbzqcfj+w3qeCueI/MwfcODTyC3YeR5gramryMOrTUk+sFcpS5hvfrSpUt1/PjxPvHIhk4CuPkP7xKwa/FMi9983ySenKQzPb5rhRVTBrmWuvRmPLCvGOCiQFgrPk+8rWuupQAl+LjeqtpVB7+3njPje8HS76jKRVIYk9/IfLpw4cJgIy3WJuFEW18IIRmLfu+PQSSZ1WtbAKgPDZ2amuqzpTi5AwUA4HmtyF4Cz/Ib43OGXwI2Y7GX5+L1He8x8/NkllEc4vX84n0xB8yd13cASEJgpD1TRwu0Ujkzbv/GWHyfeSvXAr3m4fCs+dBJNd7AS3+Za3sepu/29tUN7/C+w1bMG3UR5oW3q4brXeZ9j5Uxedy25u1p5WZagAsPDIPN2YeewytXrvRJJazHWlYNGjbSaMeGFwYh/XHmLLxs2WAekdmVlZV6+OGHe15juYC5Iks4eWI02knlv3Tp0i6AtM5LoIBWk4Btv16XP+ezaTS63UeyHGjgMvi0LMaqnTWrFFafREBdTIotNFt1MJJj9bTnvRq23jO0iAJAMdIvr3nccccdvVCsra3V3/7t39bDDz9cFy5c6K1dwodmXCdRVFV/YCcZixkWMxgbUFCGhIUMbmQzcj8WudcGoKFpQmjLAGuFu7i4uCthxN6UQ4J+15aVjr3BvH7y5MldIV7Csvyh+Ogv9KAPKCKs9qrqQ53QAf5w6BBl477RP6/9MI8eM999gLEPczVwA96EvVgXTeuWEC/KH2OIP3tVmTptBeW55jseOjJhIHU4zOAAkHCfT3GwccWaWVXV8vJy/yw0PXr0aJ86jgJnfyN1eyO7w5Nsv6CgI0xHn3iSIIA3NzMz079w0kkW9hCh4aVLl6rrrh6K/YQnPKG/npvobRBtbW31yVznzp3r58mySx32Yg1kXoM3vzG+SYftWg/YKE0DxJ/tDLQMshstBxq4klD+74nKMIuVoCcjvaYMA/g+M7kBzV7YpD45fMfivTPxvHF2ZWWlVlZW+nWRVojL/cEaJ/xQVQOPyYLm9G+u2Yqn71NTU/2L+ehHJrmg2LEgEeL0Uu1FZViNtnnW7/Cyx5XrFlaM7lcaJsxfi4YOzeAVpdXtuaTgVfg5+sm+NNrHi7VFPolPoWWLv+3R0XauD3kdMz0+g4GfY27sSdEe61Ce8/SgoDWAlWFJFxuZ9NdgYQMQ/uFUGHuH1OOMYfpgj8j3uW0bGhh3mXjlcdiASFn3Xk7q9vYPxuL28GgffvjhOnnyZB05cqQ3KqjHoVlohYwkjVteVouHko/p67XCgC4GpNZv7leW7Ov1lgMNXNeaKO7xb7Yifd315URnm/YsEuRaLjPfrUxZ7B+Px/1xNV5cr7oKXJcvX+6TGdICttXnUAzhBhTI5uZmDwQ8U1U9cDqWbgVnkFtfX+/P5bNlnlYdIGOPwHSif/5vRctfAhcWsS1X+kb99hyc5JK8YSXgufd/W8ep7FwytGTPI9cq/D1Dci0l32oPulmpeW3JFjQGyKR5zXEbkMzXjjJ43l1HemCtkJRLeoCMxeHbqp3sPyf+OPyHQWDwtQFFRu729vZA0Xt+nbQD37SSl5hv+u85NAByXxoOHh/3YbR+8YtfrPn5+T6k6fVa+p+8hXftcGTSljnLawlepsPNlGvxLr/dDGhVVV13Lz/4wQ/Wd33Xd9Vdd91Vo9Gofvd3f3fw+yte8YqBEI9Go3rhC184uOf8+fP1spe9rE6cOFGnTp2qH/3RH63l5eXr7rwtEYq/pzuL0BkEqMPKIBVcMrAtcrdb1X4te1rCs7Ozdfz48YHwOGR1/PjxmpqaqpWVlfqbv/mbwUGurf4gFGS/seA7Gu28r2ttbW2QZef1sUz5JexRdTUkc//999e5c+f60wA4YiotTtKS7bHZCmfdjlefk+V45MiRmp+f70OAhG788kmfdmAPzeFIW7w+xJVQj3mBZw02aXH6GXuI5jtvVM3QWCtlmrCt6ZFrmrTtjbwO79iL9WZXwsfwM+9RW1xc7NOo4X2/WNNKDI/dRg5RAfguDT+vt9FW0tZGiPtrL6+qBm8Trto5FQPaMKfnz5+vpaWlunz58mAdDv5Dlnj/23g87pN/mGtvPGZcTjCiPtaIHRHJ01QMnowPvpmfn69jx47ViRMnBjLP/K2vr9dDDz1UDzzwQJ07d65fl+aP+SIjEl577GMf26+N2XhIYyxBfZLB7/lplUleUhoObsv1uy9cv9Fy3R7XyspKPeMZz6h/+k//ab3kJS9p3vPCF76w3vrWt/bfMyT1spe9rB544IF6z3veU1euXKkf+ZEfqVe96lX19re//br64oVLij2KdJPT8slQIff5j2u0lUrJbZrhJ4VK6C8CSl0+lWB6erpWVlbq4sWLTXA2o9hDIEvR+4y4P8HcYRbGCVhNT0/X8vJyra2t9e/2shK2NZiemkGFsQFGDkXynE96R9k4zJjgVLUT+kSx+IQB6veanefZnglt8oyNC4+RsXge+ezEjDR07JkDCB6/2+OzE0ycXcr8ZDiO6y1jy3/2XOwdWVZYVwMM+R1wMn3TamauWoki1GNP256VszlNyzz1Bi8HOrXm0+A6Ho/7/V7QwGMl4uB1VzycnDf6atl2uj+p8jaMuq4b8OrU1FSdPHlyYDB7rs+fP9+fRrOwsNCvpZrmhEhnZmbq1KlTdfbs2f43QNc8ldnPmbrvYhmbVJL/08PK/3t5YF9W4HrRi15UL3rRi/a8Z3Z2ts6cOdP87a/+6q/q3e9+d/3Zn/1Zfcu3fEtVVf3qr/5qfed3fmf9h//wH+quu+7ad1/2SwwT08ztMsmamHRPWjn87nsn1YF1w3cYzAoaDymZIxUU17yuxHqBlYUZKa2uFEqsQLw1K2KA1YoXeiSzct2gn8CGUui6naQDn6HosIrpRUkFYBoZeNyvFoibTrZaTZeW0LrOSZ+hFdcMhG7TihnFiBVs4LNiTG8sgdPGUWvek5/SIHKxwm/Nd0uuWkZhGkwJpmmduw7mKuUHcEJRp3EKiADA2Ud7uu5PeuLuiw0ie4TmEWSGMTl5Jg2/5eXl2tq6+kLKO+64ox+rowAeF9GHNK5ct+np6y09mW1NKnvpyf2CURo+11u+JGtc73//++v222+vW265pb7jO76j/vW//td16623VlXVfffdV6dOnepBq6rq+c9/fk1NTdWHP/zh+t7v/d5d9aFAKUtLS/1nT3xVNa3Bqnb6ra1RK7cW81XVLia0NWqLPvtm4eF6WoyE63j3zurqai0vLw/6iefi+lH0x44dayoQx92raiDgHDbrMWxubtb58+drPB4PTuyg7dwHYyF0CAoFYWAycOHV8J+sO4epsCyhtefXvIAiay3Qo/zdT2cF+vQQEi26rutP4+Y3W/QGMi++p6LiObLhMCYcwvKGXZSmw4p40NCAfvhdYwkMrpN5sfxgcUNfn49oC928a/qzh4i+wAcAr98unOHB9HSRPxsZVriWW/oLn1LnzMxMv05s+vidb/D5sWPHen70i1hzAzDzt7i4OFjfS+ML/nLGLgk4NvQYJ/0llE1CEcbi5uZmnT17tu68886an5+vxcXFgWzZkyL5Jw0lyws83FpHRX9ZJq8FPC3g83fouZfh437eaHnEgeuFL3xhveQlL6knPvGJ9ZnPfKbe8IY31Ite9KK67777anp6us6ePVu33377sBMzM3X69Ok6e/Zss843velN9cY3vrH5W1pvVZOTKloegX9v/VGfrSuuGSQdWjBju730ejL77dixY3Xq1KlaWlrqhTm9C69RsV7E+gUWGgyZnl3VMKRhi5TkC59qbdpaCSI4KDkrFNpaXV0deIBWuniHrMFwn9cevTHbhgOKge95SkEaIDkG5gXFxiHIXpdjbp0UYj6wYWDQX1tbG6yzGSyd4k2fWNsz3TIzLbPbrHC6ruv3A/K8Q0V+BlCkUK8TWng+1+zM09A5PVLa994xeDaTB7L4EGTqYV3Nc8kaaFX1RhcK3R6RFb0/kxUL3/nMQL/FIUGUEK09uzR+0+D1+pe3F9AeoEha/he/+MWepy5fvlwXL16sxcXFuvXWW3v5dop91VUD5NixY7W9vV3nz58fhOft/RnEHGXIJJr9eFwJkK3lFmjoZ1L37dczm1QeceD6/u///v7z0572tHr6059eT37yk+v9739/Pe95z7uhOl//+tfXa1/72v770tJSPe5xj9t1317ARIFoLWHK756MSRbEJDc8lZzDCdlXLFUUKNYjCign21lF3ljskkBCsUWLMtve3u6Bi3PjfLK5n3VWocdhkLUVnfTKtPZU0g6FUFdmy1kpprDYqmOsSQtv2q4ahjPtHcMnuQaTJ2yksWCl5kQBh75sWNiwsfduhZjgYtoayKxUfX+LPzMZwbze4ntfa4FYjg2eSKXO3NAng+q1wpSem7Ty0zhJTwIwtIHjsDX3ei0yExYycpN9Tq/C3rll2BnEeR9vfzh16lQP2HiEGXrnJPvWPE36bpp5vvcDJqb/tdpolZv1tChf8nT4Jz3pSXXbbbfVpz/96Xre855XZ86cqYceemhwD1lCk9bFOBR2UkkhTsbisxUr73+CUdPDagHPJGC01WFLy0rH3g/PUBchlarh5sOqq68zWFpa6j0N/vu4qK7rBtbxaDSq+fn56rpucA4dAmyBXl5ersuXL9fy8vLg/DyswwxNuj5CelboVmo+/w06Hz9+vK+bkIo9NwNF1Y6XaUsxhZ/ihA4scrwmW/GcrQgPoAySxxiHrVy+2/qHH+yB+0gvXseOQZLZgzZS0kvymlf+xvO5ZsI9mYlqsOG/EyBagGBDx4aCQ32WrTy9xP9tCGYEA/D3+Lzh13Sm31xrySN/TmIhLIfB4vAp2X8LCwu7Mj5pj5AvtPV+S8ZnIGROEtTgacvFwsJCv8bVdV09+OCDdezYsdra2qr5+fn+nWGmX1X1Hpfp7JKGD3NlWiMb+yk5hy1d6+8JcumJ3Wj5kgPX5z73uTp37lzdeeedVVV1991318WLF+sjH/lIPfOZz6yqqve97301Ho/r2c9+9k21ZevXFnPVjhAiCJubm71yn/SH4nP9Vg5cS+BIBWDhs5uOkt7e3q7bbrutj9WTJbe9ffXFgmRGkXHEq1C8ITkXq903n6BOH7quq8997nN91qDf7+Q4+ubm5uBMNh80SpiqRQ/qm56ermPHjg0sV6die43J4cmVlZWBd2ZjwN4XC9S+p2pHyV65cqUefvjhnoZsu0hL3MYDStNA6Iw+eIO+WAl7YZ80asbMM35/lNceoLmBBpA370Fn+pIegQEOj92nrKA4HRK0MrOX7W0NHr/74TCrvU/6Ar0AEujq8Fl62QnseBdeO7Jxyfjou/f9MUfwLvWORqPBpuf19fVaWlqq48eP14kTJ2pxcXFwMDVjrarBEVpV1YeJzQvj8XjXkVme3/F43Mv7kSNH6vjx4/1a1/b21QOmz507V7fccsvgdUb0gTVcvzbFBjQ0dNTGhrXn0nK2V7G36s+t+1oglZ7ejZbrBq7l5eX69Kc/3X//7Gc/Wx/96Efr9OnTdfr06XrjG99YL33pS+vMmTP1mc98pv7lv/yX9bVf+7X1ghe8oKqqvvEbv7Fe+MIX1itf+cr6tV/7tbpy5Uq95jWvqe///u+/roxCyiT3NpW0LTMspvQmbE1ci6gpvPkMjAnTcJ+tRoOcmQEgQYnwt7W11QukwTEtOode/OZg6t/Y2OgtOO9DSuvIY4CW9k69X8cKlH6xjmWwdiIDdSbze33CFmxa7VW1C9hQ/nkYLMoTBcnYABrqzf7wWxow2V97AVaKnlf6jUHEPW6/xYdWEOkRmSYOGbofWbIuy4eVnO9J69r94Jr7kH3MZ03HnE/LDXOSYOb1GfeHsDl9xmAZj6/uDXQShZN5bHitrq72c9N1XS0sLAxCiJSUdwNSa8weA3yAJ4dsmY/X1tbqwoUL9ZjHPKb/DfDFkMVwtKftvrXondcnzemNlFZbSa9HBbj+/M//vP7JP/kn/XfWnl7+8pfXW97ylvrYxz5Wv/7rv14XL16su+66q+655576xV/8xUEY5jd+4zfqNa95TT3vec+rqampeulLX1q/8iu/ct2dT+K0ACITA/hvBZoe0l7utttFOAxO+YzXMPybhcWKuar6MBYKzplxnEoOs1sorMBZoyJMRZt4HRcvXhzQIK01exQeN3tfAC7GwzFGXhtgQzLPGljoiy1jhw69X2VqampXGJNxZmjWm6r9ShPCLJ5LABuvz0ZEy6ip2r2Ow1yQ6IGHxRxQv0HRitnn79mAafEdJbMcfW8aBZT8bCB2AgXZnWmhY3TQR9MglXErO20vS5v20qMbjUa7khEwVPxcApfDXoSM8cpWVlYG3/0MvMaxZhh2t956ay9HGeKE5vBfZu95TAkqNqy2t7cH20F4Y/L58+cHY15YWOj1A68YgncmlUkglfOT+uRaxcZWXs/Squ9mQPK6geu5z33unoP63//7f1+zjtOnT1/3ZuNWaQFVWg5mJJgIxkZgPfmppHw9PRun+cJcKBW3bwuccuTIkQG4eJ2CzD4sPt6aynmBKAfaQeCcHYdnRIx8Y2OjLl26VA888EBfN/VAFwOPFaMz2/CkKH61OiFMYvPQnIw6nvfaRdKNDEPWn6ATY6WPjBXA6LqufwEffw6prK6uDsCYfnijMTQjq5E2pqam+rU92mberTBRbPYS7dkTdnI4zYo1i72cVjozvxlQEkQ8VgyVfAWNlS5egBWZAdEgA30dGkyPD95hQHC4yAAAqSRJREFULA4p0raz9og4TE9P9yegp+FphQfd4Sn/3nVdf8ILMga9eVsAskOGYRoCMzMz9cUvfrGn22Me85haWFjosxJpb2FhYbA+ZxpgZE5NTfXRFPiW4jdKHz16tM6cOdNfMz+k92ZvMueTP6/BMbaWd0Ybjti0iu/3XLi+vby8m/G0KAf6rMKq3dZDuu++JxE+1xRaBG25tzlBacnbG3AoymG3FpjayrQSWVhY6BmTDCILPsIPk1o5HDlypD+o9+LFi31WFf3PtZCWd+nkgVSetJNKbn19vfeYUgi8hpIbaCm5JmYa+iRxv9TQaxGTQnMIO/21pZ0etfvtMJzXA80TBivGYFB0CJRnXJct2JaA59rEJLBDgeOlp6LyWOwpM2buyzB3eoQuGfrMNrNdj8X0YGyeZwAQmaDeVuZsyrPDgsxD8rENjVzDyWQpMm/X19f7rRxEGBx2zGUAg7YBOI1aG0Y28rhuveEju3iVi42Vlm5Mfea5naQnW6V1317P34x31SoHGrhMJAt6/jfz+FmHwczsLbe+xQAGKVt6trqtZF0s1CgbhIu4N4w+Nzc3AAOE2QxKbN+L1/Th0qVLdfny5VpaWhq86oJ+GHRy/A5tGYihg0G6qvqEDsJmVmbcm0cXJVOjqFqgDq0z0cFW836MiZZlyXigi5WL73FoLQXeoVB727kmY8VuPuYZh9sYRysElwDnsaPo0lvjPj9r2vAdj9beHNczZNZaT7MCZxxuPz00F3t+Nm7seeHhGhBaf5Zzh93Mb9407NC9k1gII5KZeuXKlVpYWKj5+fnBG5k9txnGTtkx4CePVtUgjGwjEKPLwMWBBTbAWvOSPJCAczMgY3kxf7fqvRnP60ADV5aW55XWLJ+tPBBGK3MYNyc0FQyKONs0UxkI7OFsbm72oTR7SwBQ13V9RtFodPWdUmQhdd3OHg6Ho1AwgN3DDz9cDz74YG+NOdTjv6oaeDiOzVdVf4LD5uZmH++fnp6utbW1AWCm0kbQyI5EWPnzSR65x8tWKvc5xEOohnETrjGoQNfWoahd1w1O2OZUharq58aKznWnB+N5tfIC9A2+KCBobL5dX1/vk3H8m2maYTrf59CtT+bwGXWMOZUkdHTojnlmo6y9CwpzzrMOc2d42eFXK3H4dW1tbXAdj4a+OlEIZY9XYoBjXAbtruv6EyuIPPj+1qtCGHPKNPsAL1682B8EwMHRPu3C/UCOHbpjThYWFgY87vXh2dnZPhzu7QuMCd4/f/78gBdMf/Oa/zLRC36fdJ6hi2WM59Ff3obgPmX/brR8xQCXgSavUQw8TKrBwpPg5w1k9qxgTi+wOzyAAjMI4GH4pAsmvGpnj0rLqh2Nrh5Zs76+3h8PgzD5kNn5+fl+cfnChQsDzwHBdLwfRrPl6gynqh2Fw3vEoFV6B9Tt45bYy0Toin74KCPPiQGLcI1DWLZmfQQVYVQUhunuo6sMKraK/S4m5iQVphNH8CgNlMwTCrcVmvJneIi5y7R738ezgLy9vDS0TIdUdPy3J2JAp317HhmCc71+7xh8y5aO7FtLTlHu3neGEmXsHkNa9LwahHttiNpjsefKaR2Weeq1IejQoenn+XAyEGd8OhvY9ZIejxxiIPg9ZM6U7bquj7ScPHmyf3kkfWbcbO3x+r35wcCVRoOL6X6t0nIUzJOte28GrFwONHCZ6VqCmwBkxWLgshKzYJvIts4oFuRW39x+uuP5ZmGHQFptUY9f12GwcjhvZmZm8AJKM1VrLafFvOkZTQolpJJDKBBarHXviXL4zPUxdodJmScLor2wfFtuCkzOU+t3ir0ohI8/6NYK75lupkt6224n+YPrDnVmfebR/J2+OXFmr34mv1uh2XvMpAxoaJ5I8KW9BBMbYtmnVHymZ4KW+824HQExiNqj4n4y+NILSCPSuiDnK3UFY8WzchYifWitGVLcf9MJI8BGow1o+uY9aq35TsBtGfhppE0qLb5KfvxSlwMNXFXDBAMIn16OmQymwEKC2BniaVmpFlasaMfOWxaI10rm5+eraufNxlhJrIOxN6Prul6w+A2vCg8mmXFzc7OOHTtWMzMztby8XF/84hdrdXW17wt9drgHLxOAsSfF/UePHu03RnLYMWN1SC+fISyIIEAnb4tAyKt2XmrJ+hjx+qmpqVpcXBzMuYGLOU0PK4EDpYGhAh2dYcd82MrmeeYFemElTzpai/7hYfpwV/fLm3/Z7A3/eSsBazCMjfEQIoMuNk5saee6R9IFcCL7EBo59Et729vb/YtQp6amBq+QSQOMefEpK/CaZZM+o6jdX3u5POd5w9ixXFquMW42Nzf7TFJ7hNYLHgcAnfNOn50tSP3w89LSUs3MzPTvQ6NdshszacZ1e/7m5uYG3hjh8ZwH5gs+aRkIaUQl0GQkZFJx3VmfdaXbfKSB7cADV7rBXjep2p0FmMBki95uvS1JM6qvW1m6fltEKCJPGiEDg2XVjnt//Pjxvg36VTV8XxbghbAcP368rly5UsvLy/X5z3++P/E7rXIrWlvK9tgQCE4ot8eDIkCg/fJHLNmFhYUBkFkJQU/CKm4f5c0eFvoPQFq5sLbBs14zoi1AktMyUD4GPkK2Bm0DYtXwVIRcE8pTxZlPUrptFJjvqNvJMpmo4nUoe6dWol7wT973ifT24r03yqDkMdC+DbMM6Tppgd+8ydcGha34pIP373kOCDnbMMXYWVtb6/vpuadvpptPZ+H+o0ePDk61cH+ZL2cwmqYGS9pLb5XnL168WJcvX+4B4fTp0/3LRJeWlgbGQIIBgA9ATk9P1+nTp+vixYv99g4bop6TlneY4JHGeXqDaYj53tQtWUfWR3mkwOtAA1e68pOu8b01EQay/Gs92/qcE5l9aYUFVldXew/MVjQCl2sKVbvDJH79R9d1/UG5PtMsx2IL0YrCv1vh0l+DiBW4+804rAjwbm0UEKK10qY/eHVWfhnCcx9ac8BvXiBu/U8vpBW6aQn+XiES6reRYAPBdOa+vN7iQfOVFaSBlM+tNs2fGaqcpMxMn1Q+no9J0Q2Dh/vjufLzjMFrnznPPI+nj9J33yx/joQA+PylsZG8DCilTrGxmUZZzhXr0IyLF0Q64YW23X//eeM15yey7ut9XNfSOy3QaHlhno9JJXmgVe+1rk0Cxv2UAw1cFl4rxRaBkqnMnCngdsVTOCxwuRZgxmcyneZKX65cuVIXLlyoU6dODX5jg+rKykqv2J0ZRb1Yo5xvNjU1VQ888EBduHChLl261Ft8tEW7PE9IYXV1dWCFc8/i4uIgU3DSIbctReV54RkyxVAWhGl49QlAtL6+3q/NWQFnCALwannEuRk7Q1dsbOZahjRQmg5/ebxpaXoN0BY64+U3J+NUDV8xYZ5ziLOVEcZ/PA5vZEehOoTncwEdFqSeBAZvknc2Zkt2nAUKbXxSObxGG0lTg7fDoAYu05/xwL9k2I5Go35bxPT08Fgxbwmgfq+/cg+RDYxBaMH1NHwy0uIN3blswHu/oAUZiKdPn67Lly/3maSul3lDVhwmJYtydXW11y88a6PRPAvd0/hqAeW1gCuNvElOQsqOeXiSI7HfcqCByxNGaaF/Cknet7Gx0We9WcnlhNBWtmcLrGq3C+6QoIXCip/DYl0vIUGsMurybv+qqwL9xS9+cdeROITTUGy5zmOBOHHiRC0sLPTC4uNw7AGmIuVkAmdvjUY7GydJG2b8nArCGgUngaBAvNkYxeEX6K2vrw/CpF5kN10NWp4nQi8Ur1Wi2FyfDSFniTGPGxsbg9CpQ2vmtyx5QDLhHq+5eAxWpOY5Z+7xHO3bMudz1VUDaXl5uQdabwi312avJBMLAJuqnSOteMYg4bCaaT0ej/vTSWz8+R1c8KJlifByJlUtLi7283Xrrbf2fWWM8DvgubCwUJcuXRrs3YJf7REx7+YvjEnoZC8NsF1ZWdnl8YzH4/4QAAwoH5jtdTpO5kAWnf1Mn06dOlUrKyv9+Kw7XLyO2gIp67Ts87WK69pPSe/uRsuBBi5Kuq2tzwah/N2bhFuWA/e22kWILeRmEFv0rseghRVVNVxXoC63ZTBBQHgBJArA581ZsBBOp/9zNpoVbyozP+v+p4KnT/ZAACnoNDU1NUgftgK3kjEIGOTtfSaNHPpkDa3FEwaHFPiWRdriB19Lz8C/0T8/m+PI+QaUfKpGenPJy63+ZZ9sYPk32rRH2FJylhGHeFt0aMmO+24Ac//xmim5ZmPvLPuTMpyev9ebMfwccTDv+hmPBwCxsdDiR655bxhzyr2sZY9Go74vNnwmefLMJaHG7BP0bc1Baz5d9uNxJY/tpSNbXpUN+BstBxq49hLiqt0JFpMKytPPWFG2wmNVw+OacrHbVgzPJENYOZNt5AVW6qZfVhaMa2lpqR566KFeINg7hSdTtfPWY8YDOGLt+bUpjIvsJYciWorQ/WTMCODW1s7ryPnO5lc8MSuu/A6dNzY2amFhoW8nAdRCigIjXMR9to79vEEs96fZi0h+sJKkDm8aTkXusGR6eFZ0/r66utpcM7QSypBVApn7Q78d/vK82Fv3PJuG9Nl/BmL3J0OfaTA4CYbfnarOPQaSFji7z3huztwDkB0i7rqu93RGo1GfBch2E9r0GYiM02Plmvtoz9iJHni/6+vrdeXKlTpx4kQ/Ds73xNgzfb3Nw/JPBMB9cpkETJOMZ49hr9JyDPJ63p86utW36ykHGrhyfcIFhqraiclmmDCtehSqLX4LGGEiLC8YC+axguA67dmao27CbKS0rq2t1crKSh8ao88WAJ6Znp6u+++/v3/5nIXSpy/kWKp2TsHgpXn0L0M/U1NTfTouawKsV9Enp9QvLi72CtF7saqGlrIFzV4Fnt709PSulzd6nwzzl+FMezAkvzh13d6EwRiFyTiskAxeCLUByIojedDGC0YB9zBfZB9CM6/vOEyWwGr+yL7TDvNDe7mxFhly+NNjBwS4F++BLMCc1zSuTMvxeDzIyEvDAbljz5+NrgQL3kGVRqa9Rv5YB/I8wl8LCwt9OydOnOj7RoYeWbqOUlimnfVJPV3XDdYfE2gY8/b2dp09e7ZOnDhRx44dq6rq31tHH0niOnbsWM/D1k+j0dUDCWZnZ+vChQs9ryRwEOXI7EuKASeXQiaVlr691nM24tIIvt5yoIGrZU34/6SwSbqqtjqu9X1Sm63+2HK3J0ZZX1/fdTJ2hk/cflX1a1YXLlzoX7+QjJdp1a5vcXFxkKFkmjnxgTqdoGLLyeNDSG25ofy8xpUZW1a69AnF47Ru6qmqXcKfHhdjSc+tZcjYK6J/Vv42jKxsedZhG/4brL3m6P7hfVmIkz/MBy1+dl3+nIkN7r+f91xknba6GWcaXfzm+TOv074BjbpR+jaqACZokONPhZdrbjlP7qs9PfrrEF4efNziAcbA0UwO+ZnfnQ3sNSnT0m2R1k6bvF0h11LtiXne0yjmN/fZPJnPtsp+PaGWV57/W7+3wPN6y1cEcCV6t1xVT9Sk3/N763qrvmSivCfr4R6nyjocYeXggkW6urpaly5d6t+SnAyQSpE2WfS1EnE/ATIvolN/rjUwRhgxT8dGiAEuZ3BZsLGibS2bjtSHovA7wKz4TNtJwMVfApJTpO0lWLG1jBbzREtJALZ7hfMMAqZzi2fdVsuwoq00fpKXmevkuUlj5H57X5n84iSeVoGmGT73Z/NN9jfXI/09jSE8ppyjqp2Qtl+fYxDyfBmM4YfRaNQ8yxPP032HdyZlc1btvHuPKAjp7pmZiCzhqSePtzbCGySTTglcLeNov2Uv/dvSuzfTFuVAAxel5U0l8SaBCAUrys+3AI72Jl1vLVrbcjLTcnoDcfWqnX0sVsiEDqampmppaakuX748OAnaaw8pxPzNzs723lZa93hpDiPZC0mL3ULi+D7KbWlpqX/1A68hx7Mk1Z/+tlKqHbZibjIrjfbyUFrG4/VHnz6yvb3dr5ehjObm5mpubq4P2XZd12di8hxrHVi+CayMJUNlmRhD1ho0QHkmcJnPuE7d/t0ZpLQBn3nNClrwnNeX7P103U6YN9er0jP1XKDMOR2CPjiRwpulGRd/Hhv1egPyaLSz79Bzx7gwnPBe8Go5iBpAhB4YUxn6T8AwH1YNPTA256M7shCyhvc9F9QBj128eLGfQ7Kc4Xf+k0WbOgyDNEOzDuM6kpE83tJ1+ympKycZPdyb3w+BK4jessD9x+Sa2LkAej3t2CprWZGt2C/rR1haXqegLa+hoQQuXLhQKysrfRv8jiKCSVE8MLMXqHNRnHat3H3MjMfl42BQjljanAq+ubnZ95E1PCxEAwnWJf1D2Dj2CWXqvTiT4vTMh8Mm/u61OOr0W5qxdm3Jk6mZQpmebFrrAI3XDxNI7Lkk79losUWfSsKK3nNMPfYCUJ72HvK0ivTADDROvnEGmz1m12Hes2fdAi7La0u52wMyj9pDNx877OzkKu/HqqrBuk/LSPA6MeN1iBL5Yy2LNHqPC9CENzJyYV42329ubvZGrUE1ozLWaQm6bsdGrH8zPdKw3m9JZyG9udb97tuNlK8I4HKxom25r62JzRCFyyRAav1lyfCAPzskkPf5vxl2fX29VldXB/u6UIgZooEBWexORWtlYzCv2n3UT/bHfeWPDELoSDw+hd/C47YZB3MBuLYEz8p8Umzfac25HoLHwJtz7dFa+OhDep0JIp7v1meKFX/Ok/tnT8Z0yvn1Z9PR89TiLfffJQETBZ4A5bq5N/9ybc/A1WqXuuw1us/IAbSzAke26JsVuRU6XltLbm0QuC/psZrW9gBTlhwJsVHptjw+aIQcpSHVMixMP4DTZ5C2+NPPTfK09gsoN3Jftn+j5UADV7qdtqrSfeX7JIuD/UbJHPkdZs3wga3eVCT2SojBz8/P9yHC5eXlwStG8BCwjGdmrp5ywcG5VuppudlaY4MjG3YBP8ac3h2gQ+Zi1VVh9cG4BgsDyMbGRl2+fHlgeVKXFQJjSuUyGo16egB6Fj6HYqwgfNSN6UeoBYDyloe0yt1Wy1BJj8gndFixtAwVX3OWol+JQn/wVu21MM489d3Fp4/Yw0ceDD60n2tVprPBgO9OwvCGXObTXgVKu2rHCHLI2skPBkr4wadv2CN1CJ2+Mgds4J+amtr1ehr2OkEbxmWP0Ju/ARsA2HOB/ODBEsaem5vrZdM8hT4i0uBQX/5x7xe+8IV6/OMf39OaVHnWiz23PLu1tdVvvF5eXh7wtA0I86UzMO0Jpu6cVJizvNd6ufU95/5GyoEGLhPCisaTY2b0onxLqUzaXDip3ard1gSM699SEcOMKC7AAeY2MPJmU9a2Wjv3E0Q5kRqAQFioD7qYcaend1Ld6UsqFdpBCfBSyfF43B/TZGVgJWmwQok5VJMeQoZQ2LTMobb0fX19fdA3CoqJNSzGBDAgqBZq09UKnHuyn/ZCMAoA+VT0GSZDqToxwDxi4ea+9AqtHPNZFDjg7HZs2DEPgCOhKoOZ+cXyYRmxwgN4MnyJkm/Nt8PI9nDtJWY4Er4m/bylRDFuAMTV1dUBEE1PTw+2CqThZCPAxgZ0sFcHkMF/9IMxHT16tF9z47gq6kZnYLycP3++HyPzwyZl6k5jlfB3eo7Wk+aR9C6zL9cq6cG5jrxvr+83Ug40cFHMrGnx+lqCVT4PM6YFnfVk8e8Im61fC76FpQWwae1MTU0NTsdIQKnaseR9dJDj7Q5LOAkDpWalQ/8mreP4ZY3e5JkL57bwrEwZUypH/5YvaPR4sVp5zuFdh7Lyz4qaUCu0o6RRQP9MR4cdoY090LRuc/wOKbUs1bw+SQnY4HAbbot9YclnGBLMX7aTPJ8y5XVGaGpDw4aJN/QaoNKI9Fz7pawYN63wrPmM8FgrpEk/LYumWa5b2aPOtjKUmTRHLsy39InwH/WZf1nrpR3egryxsTHYwpLzyDM2lPKYOBfTr8VXSeObKdeq42ba+IoArvS2JrmuZjiuw+xmBu51jJ/fJ4Egvzm9vWroDVoR+nUgCGjVzmsT3ObDDz9cFy5cGJwGYcFESSwsLPTnDZoeLlhweEeZ4usNxLZK6f+lS5d6UAQEqoaL6ng1Fm5CN77GMw4/EUrCKiXzyVZzvtWWQj0cY+U1Ap6ZmZnpX5nCmKxQErjSe/S+O4AKGhOezDU6xurNr56XNFqYO9PLyTMYINSJQnV2pWUBmnhsZGRiqcO7LvCBQdvGnftqb4d59WZZ15ltwX/QjVA1IXIntTAeb3CHpjw/MzPTJz057Jj8bbA3WDpygCzZQKKdlkGCTIxGO2dt2vPGwDx69Gi/Zj0ej2thYaGvDx6dnr6aFXnLLbf0vzlUaAMCL+zo0aN16tSpevDBB/c01NNQyXVFg+Qkgz2jHNTFb5OcCOvdGy0HGrgsOHy3QO1VUkipx4e8Wrm1vDALji1gK/oU2oWFhcELC/kjBFZ1dWMy55h9/vOf78EBb4m2HTZ4zGMeM1BmeQgp/TFAuj4sdBQ5u/m77mrMnFeSb25uDhIvbDkaMNOIsIVrpY5HgALFYAD8DBZuk7F4Hc+vjLAFnorW8wZdrKzor8GMt89S6DeKFqWBx5ueBryFF4LH7XVVJ4Pk2tNoNOpB3et60Jt+O8XZYTp7NABWZhVallJOWMfCiLDitoEE0AN4gBAehz2SlrdimmFMOXRrHvP5nMgEYyJk77HZ+wZcmBMDE/xgfvFJIfSHMDmySHYqY0R/+MBp+o+cnjx5sqamrmYYO6KCoXHx4sW69dZbe94/ceJE/+ofaIu80I8TJ07Uww8/vGtuLY820inMgeekpSeT71zsJPi7eWsvMNxvOdDARdnL9W3dY0HPZ/zXCp+0iA4j2OpsKe2q6sN4rQlFeFDOGxsbde7cuT4ebuFyfcS/J9GmpYwpZmCUAGe4sRi8trbWAxbPVO0o2FR67iMCZxrSdytl+uHsThsBppENhEyyQIF5HSvDthkOccjN3gXg4XBo1XBfXStERt0oEhSAeaLl3RDCa43bXpYtfNfL/f5uDwma4HHQ11yYn7RW5CiC++37CInbiqedBCae9+kStvbTq3PI0GFqnuO/n3PExJ60DUAbVMyZgcv1eAyp5GnHPJ9js2dsbx/Aox/0ja0luXUE3jSvQzuHS1Pe3Jf87nIt43+Szs1685mbBSzKgQeuBBgEcZKCTGWWgpQucwqlv1uoMozYWs/AWs11B+4HLE6dOlWbm5t1+fLleuihh3pryoqDutk8i4WWwJsxfDaHYkVaoVVdBcITJ07UhQsX6sqVK/2LKVvrDFj+VogJWliiKKj0zFz43V6hwcOekd9qawBxaMtAQXsoJgMmoSU/j3JB2ZlWtOVwJzxgRWXg2tjYGCisPLsQ/iKRxKeFTzKaDHZ8T0XupANvBrbx4SiDgcP7uKyozfeWBe7BmIJ23Ov9SH42kxQ87ixW6M40JeTd2tDNZ2jhtUxoAGA4KYf2AHLPG3204eQxes3KOsceOeu1gJc9cMvopUuXamZmpn+zuCMPBmNnntqLtefkeTbYej4tN5OAxvoyjfTUyeZHf78ZEDvwwIX11VJU/F41TLvlu5UC17xwbyFPcMrfW0CYDOG4+fT09MCTAsy4Z2lpqc6fP9/Xb0FDcZ44caJX4MmIPrOO32kDgOR3vBsf9nv+/Pm6dOlSXbp0aZeVz/oBAkEbtv5oC7o4Gw1lT7gKGlkhOdOLMCCAVbVzIsbc3Fxfl9dB8hDh9J4IXTFu1v4cPswTPpg3Aymvf6cYvFAu7MuBL7y2iLJE2XHdazPmc55vrc1U7WRJ8rxT7e3JEE72K2ccSobPAG/G0HU7KfKpsLe3t/uwGL+nUYiMbmxs9ODs8DEbbjGMeJu3DSF+Q37IOPXvyL3XWzEOx+Odo9PQC2TzOZxnBY/8GbisBzAc7f2kBwtNofPi4mIfXmYejh49WnNzc3X58uWerqwrLyws1C233NLzIPMBOBM6nJubqxMnTvTHwrUMjJZnBJ0JYxJC3avYcJvkhX0pyoEGrtaEZJkEZCa0QSe9lgQvnrVVn4u41GmFYOXHdRcUFsJx8eLFOnfu3K5XfXAvay4ZrrKFaDql8gBoYDyUx/r6+uCU+qqdcCR15j4ch59MI5SSrUiUKAo+kyJQRAjP7OxsHTt2rG/fRgHPpMJKAyaFi+sO6VjZANAohgxDUifzxTPQg/5AXys5nqVdaNziNfN0Jgcwz6nUPcfmU/rDuFF+8JY9kVyfsPLlPo+ZuuibQ4mmVxpgLdlKr859Ma9bJvainXmyFSq0YchzCV4YSXmSikOF6c06EtHycra2tgaGlXmragfAmZO1tbW6dOlS3XXXXYNwIwcSQHfaWFhYGBhVLe+pBWDmmf2UScB0LY/qZgHtQAMXJV37/K1qh5Atr8z3TgKuloClZVa1W3Ddlg/PTO/NXg1MeuHChX4RmQLTzs3N7QIo35MhnuxzK4y5trZWq6urdeHChV7IUbI8h6BZsXjh3+2nUDNOgMyCCu2tUACuxcXF/jeHr9ymgSuB3HOQn712kQaI55e68Ax8hJJp7TkwwFipeA0pvTX6kinZVoBemzJotRI6zF8OF7kfnsf0NJNu9M9g3AJuz6m/m0e45v8O5XpeGIMNiDQmM4LiYsPKIMFn86j3zBlsHdlxX6nbfNkCUegEcDrJBkOGZ8g6hofW19draWmptra2Bpu4yb5Nw4vlA9OWzy0jP69NcgRadE3+3+u5vXT19ZQDDVwWJjPwXp6XhZ/7bSGRcUWMOpUACsJ12CKtGmYlwdhTU1N18uTJfk+Gs7pgZNZZ7r///jp//nwtLS31WWgwKgxpYRmNRoPji2ZnZ/swiENuGQoC/MbjcZ07d64uXLjQv+Suauc8RYc3/VJKMg27ruvbN50ohJsIxVpBUwhfIcCcLEIGHtanT9TnFROO9zPHbDhmDux1MQbmmXvsKY3H4/5AYNODEGMCJvNvL9PvVuKlgfAq89MCO9OKsCxH+QAa9qSYV2/utfEEP2YIk0L4K0/4TwOLz85eJaRkw8uenE/zN52saFHSzC/rsAmC6UXSDn2ADga9BLj0SpF5ezeEoDGy8L7n5+f7hCVOdUcGSZ5wuNqRAGiTEQLG5q0uzA8yhUxevny5Ll26VKdPn+7lEk8wE1WYDxtn1nctALNRP8konlTSa2s9m97cVy1wJTNW7XZRPWn5XD7TsvrzXjOqLWcXmIa66QdWOsVhExiRo5MQGFtyCwsLgzRqBD6teScIeI3Lyo6/1dXVPjxYVYMQFkyOoneIhXUPnrFX1nU7Kcrcn1lg0Il1Jl4YmF6u5yy9gBwbSsHK04YJSslrKwZ0h06t+JxtBgjbeGF+fJxPhhZJzmCOGUN6dQ4J0mcLfHoi+ZsLiSBWVIwtlYfXgVyPjQFoQ4KMf2+FF1tenmnL/PG2bfjGRqUVLsDs+ltgmd6U+ck8CE9iJNjQJGxuz4i6mEfkl3YMPPRhdna2ByXo6+0S1jcZBoVm3rZy+fLlOnHixC7wSwPoWl54zlnyj/XUXmUSSOU9bodyPcCY5UADV06aJ9/E8ndfT2VQNVTOFDO+lWMLtPZywxNgrVDwELa3twdvbTVzezHcYSJ7hozBfTewGbSuXLnShwfTqzCYoxisCBB6rEyeS6uKPjjUYu/N2VdeC6PdDLs6tIeic3gv223No+lmD4O9ZA5JGcDcDyc7JBC0rFvG7s9pCRvoPJctb8F9c5uTFEPOSUsuMvPMfJ9y4j5bCZrHUe5u0/dbBqy8U1aTRtRlzzPHmu0ZXPgNj4l1Za89OosVsLL3hLFm+bQ3BS9yTiZjRPa8ZzGNEPM/dKTPZPgauHK+mAcnWUwqk7yulh67kdLiu/2A3bXKgQYuW4cJIJOK7zPwmXHMTH4urWGXtP4T6HiG+5wNOD19de8UgOXjnRxmc6jBgGHLyoLM/ViqhBxR0p/73OcG6xl+3ictpPLmu8fg6xZ4AIkxsD43Pz9fx48fr7m5udra2qqVlZU+Vl91NWNyYWFhEPJLpQDg48kuLy/3fOFwE3TyqRdY0wYfgBP6MtcoKuaTkz22t3eyJwEzaL6xsTEIq1YNEzHgGfMbnp8VKMX9dFJLGlq2zmmLZ1jzcLr4JOBEHljT5B48ijQC0khIeUu5oA3mw3XYOLOsOqGEZ8bjnQxP09WhW3s15nH4dGNjo1ZWVnr+suHGhu4jR47U4uJiHT9+vAct5tfzgFxQ8qQV09LhWzYgWxbTm5+ZmemXDzD8qN/ZoayD8cfBwNk+dG0ZKY6g7FWSN1pG+6RnbqYcaOBqWZr5Py10SstKSYXidmCilkXtMIUBLq1VmIHQBIeDOgSDFWhlSogQ6xCGJPsplQhhN/4saIQil5eXByGMqtplwcG4WKNOdUbA7GnxPxeqWUMh0cJnKaalbCA24KN8mB8rIh9MbO/T6yl5eCwl58+ngqBAvc+JZwyeXPOp5FZWVUPQQhHSnnkvQ9CMw4aRx2nvGvCkXx4bfOn6XOyVG8DN56nIaAf+gNbud4Z2actGp8NsZNNRh/d4OdtvbW2t5ubmBsag+c0y73a98doAalp5rriXkyq2t7d7wLAcOOrgNzDAt/bU4AdHUHzAAEaWdRT1kKjkk+ZTNwFeR48erZMnT9aFCxd26b6cA/Mg99xoSXBMB6D1+XrLVwRw5TpBFisxX0vLlJLrAgahFiDxTH5OQOEPxjPoWunbY8E781FHttAtuNmfquHbgLe2tmptba3/c10ZHrDybwE+45sE8laMCLUXsF2v2zJoZJ2t9cc8Xd3CS/HCe1rx5gO8W543UGRIzjxi+lKv66/ayVJMHjVAUVfOAzSwxZy83FJCngu323rWPOPfqCuvtdqpGh4NlRa4PaE0Ah1G8xxZDjOE3eqTwSHvs4cHGEziKY+PNgAtRzFMXx8O7bE7pIknn0rd97do5Gdo30ehuS+swQKAk2j1pQCtpNn1/HY95SsCuKqGHpSF1ACTceQML/DZ2X62ap1wgEXTsi5aAmNLfGtr5yBRh/vcNgB28uTJwTFAnBdYVf2bhauGe6sIzY1Go94qvXLlSi0vL9e5c+cGVrEZHW8Dz8pg4mwrnnE/u64bhMac4Xb8+PHBHNFnQhqj0c7Bww6jmpZ4ornZnA2qVVeTV3jOr9ZgLCSU2CNGySUQb25u9huyjxw50gM9xoTBlLkjU7Trun4sgB9JIc4OHI2Gb56GZpMAgz5bMSUP2aswX1sm6C/P+DDaVJIt485zCD0Yt+tKI8djYf6zbx5/Giv20K3w2TxOXfPz8z2/Qgf64TMm2XCcQGWAtQc6Go1qeXm5T66A1nhPZMIyr4R8fXD2wsLCQG68VmYDyoZtGuPIwfz8/K61NurFo88EjpbX7M+tv708I/NEy6CfBFA3421VHXDgankIKKiq3ZuAXVJAbYmhgJLo6YlRDwyRWTy07zRvCy2ABnOjWMfjcX94LKEBwMCZiSgwFD3hMC/Mzs/P18MPP1zLy8u1vLzcK0gDQAK617LSw2AM9N0LzxsbG3XkyJH+2KKWZUyIEw/Eh6eaFoClPSXu89rG4uLirtMT+A0wR+nZ0rb16pAigM98sL3Ah+Ga33KDK0r04sWLdfz48V1n0RFGSkVq3oD25iPmoGrngF9oZt7H68sQOTzR4tmVlZUB0NEX38faWfazaieUTD/xujY2NgZhXx9A68zXquH7ybJdb1g3yOBtex2OuuCVo0ePDo5MY44zEShfXslcO6TtEHHKxcbGRq2urvbPHzt2rBYWFurEiRODdV3kOEF7rwxQz/H29natrKzU0tJSnT59uucveI/2LafJUwabDJU/0sVjcfmq9riqdi8O+lqCmktaetwHY+U6VctCSI8v+9HyBs0oaaUgnHgvGf6rql3XrIj589oCgMVBuVU1sGKzfr7DyPQ56WjFxv3sufLZiV6bSqC019oCf3vGFCtY02s0GvX7cQAsklE816a3Pd3kB4cVM7Tp/wYK04R5cdinBVCZvGOe8thRxvm8+5z/7Xm1fm8ZLjbgWla3n0leYRyZWehwtfvitrnmzFB7PozH9fpeGzdpnLifk+S5pdAnGcWpF3gWOjjjdGpq5+3MmVBlHeAwsw0th/HNM4C218lSTvjdvALtMsTf0oU3Cy5ZbtbLcjnwwFVVTUZPRZGTi8CklVs1DPlkO64zQ4kIjF3/VA72ItwnrDna5RUdfv+OFS0F5QoAYKVOTV3dL8RG5hTsqp23BDt0lKDo/1YijNtKb3Z2tubn52txcXFXqMreL9Yl+8CsCL3O4fatvOxNrK+v98rAmWG0QdgzTybAC0IZmXeo21Y5ijKVlEOFLavf6x0OHXr+rMRTodCW10ZafJvXx+OdA2EBDd/j573IT9iP58y3KU8GdPOGx2mjJb0J84TnmvmHt7mfkBxzxm8OkSU4WQb3MkST7tSdYybE72cYv8ewtra2ywAejUb98WXeBN51w7cuQwPGbHl0JAI+Nr24x6FK09CylsZJjn0/JXmRay0+eyTLgQYuC0paTFW74/PpLbj4Xk6EyA2PtqJdJgkJwm7FiCCzf4k/M+Ltt98+CFfaK3Bh31FV1crKSh07dqzPPjx79mwtLy/X0tLS4MBdW2UGQR+rxFjsLaHoSdt1qHF2drZOnjzZKxMWsaG31+KqdjwB39MSIJQaxgDf19bWajS6GiIihXg8HvfrXVNTV08pWVtb6/uUoV9b5lXVhy1R2g4bGqhGo9Eg1EkfWONAITE3ViwoDRsIBpU8/cDhVoMDhonXH3nOafOj0WgQWqYw5jSW4CnowT30xwDtP07H8Pjs9XCdUDihRfp55MiR/p1z9A8QdX/dn0yySS/R4ciW52KgswFq4zN503TJSEICn9PcGfOlS5fqxIkT/Rsd5ubmejnPtUrCm1U74AXvcZboaLRzKG+eUOIkI/OYaej5xCBIcL8eL6kF/K7jkQSxAw1clEmEScL53paFRbHl5ucMUBasXPuwUOXCuU9ctqCjyImNo1ByLKmEaJ9sp/X19bp8+XItLS3162UZ9uC/lacZ3GGLljXl8bgAxjkH0AqFYnD0/dzLgrlDT6a7FQz7bOgbCoVwjZNdmC/3PQ0bew5cT48tf3NizSSapVdk8PK6jv/Sk/F1z2srM85eAJ/tPVpxZ18dlbDnlHMFPTy/LeXH9UxM8dgMBJ6bNAKTBtlWevruY+oJ6vU6bctjc8jYfc8xeq7xOu3Bdl3Xb0PhpA7TPw3ulu6yfGDE8EZlJ6Okp2/auJ8u5tuUg71Ky+tKA3TSbzdaDjxwJeNwbZIC8TNV7diuhSLBzwxtq9Hx6kyHtZVtQLIgsrh6/PjxQSx8UqzdwlNVfaLExsZGPfzww7WystJ7BRmyTKZCgJzMwG+O1Tu86v5VXRU2Mumyv1ZsLevZc4Dy8D6XBCArakIsmT7v8/PsSXEt+YViT8bKKEN/LV6D3l7TaSkf0wfrmjY8v9DNRwhBo+RJA5mVp8dnYMhnPFbPizdCt4AVmjgyYWvf7TtJhXFhzDjUldY/ffe8WEZN2wyHJ32sL7iOxwlImT7wXMpb0sA8gSGJPDlDcG1trd/DaTn3855r2nLd1iHQlCSSpEHy4qSyF4jtpyRfWzfsBWI3Wg40cKVVYsK1GCotLbv9WVCepJPzHPXAQK2JsCdly2w8Hvfp7BlW3Nraqrm5ubrlllt6xq+qPlMp60e4SYRYW1urhx56qC5cuFDnz5/vU6+xxHjGSR94NWk9WkGjXBYWFvp0c0748Bi8YEyqsJUn9ONcOrLOCKkdOXKkX/Pquq5WVlb6ZI/5+fm6fPlyrays1KVLlwYvsLTSNZiurKwMlBsK2ELcUvTetuD78OzYi+YwGPd5nc4nntBX84HDad7MmvzsBBf/5uxHgwRhSXiNw2KhAXzt+511x5gIgxocGCOek/nRIWWHmR1yA7AIgTJ3rEMa/ODV9fX1PgzmA3RJxKFNsjntuaU8+jBaZCa3vnjdjixGh4VtVCSIUebn5/s1avMHMgdYP/jgg3328OnTpweywvw6eSn1F/fwHK8yWV1dHawzQpuc371Ky/CfdN9e31t17QWg+y0HGrhsqbfAxcotLVTf7+/2IHKXfSqPrMPeESWVkZWBY/gwIGCDQOWRO/ZCHNq7cOFCXbp0qVZXV/txey8VzO51MwOnvTOfHkFmnl8i6DPevAnSC8oGCiuGqurT9s3QKC7qt8XLhmlOLiDsh5KGNgCxPUsbESjDXDBPi9b9QcmkEvJ8un7uscVpC98Az3Osi2UWYIb6Uikzzx6zAdWncDgs6OQXz4v7mb+l0rTsmdaWDYcnDWbmedef8gr/A74+EWNqaqrfW8e4qd9rQqYh/U9PMPuX6eQYmqxHtdbectxux56HdQlyRNjw5MmTuwwR8zKyg+Hg94MlsHn8GQW5Vsk293N/tu22Ut8+EuVAA1dVG833Ingy8SRCJlC1LOJWvZN+S2vH/30d5Ur2UnpEZkifOwhosbiOArCCQBkk8EFH99/eh89bs4XqeL/XQlIpWQlaAduCtSfhI3cQUDaLWnG1rFD67jmxYO9lSVqp5PcMH1ImXU++aM1J8qEz6ajbRk6uA0Fvbz7POTSQOhFlUl9zzhKUzEPcNyksylgprTWWNARTlm0Q0Sf6BS9iXLTo3vIIDJAYca0wYIYmbYC15D3l1UZV0opxIFPLy8t95mgr3N4yMHPPW4ZgPb5r8b7pw1i+1OVmQOxAA5dDCxRPTioIl2S6FqPbMs867FWkpZlWMS97I1RhBWXFD2POzc31nlOmTzscsrCwUJcvX64LFy7UQw89NLCUER6Ymww3AwObd7vuavo9mV28UtxjhxZ+dbnDfNArvUmegSbeW5UeKu86IjzEWAkL2VIGIDOu7+Od4AXOv8NjW1xc7H9jzGmdWwGhYJxm737Ah/b4zIN4sG7HYMdYHM6hfu7FkLFS5R7qyNCceZHTJDLph2eraqBwW4aJFS3PdV03CB06u83KFZ5IkLCHyO/wZ2aNUgfelmXDdbPem6F8Z6Zub28PTmoxrfD04K0ERvgMeiDLeIaeG2/fyIQJg9jy8nJtbGzU/Px8nTlzpj+BI40bz4lDvsjg/Px8HTlypJaWlnYBV0vfJfBOMiImlUmGUpZW+PZmyoEGLhPZisCgMAnV0523pQmT5R4KmJQ4v60mM0daN8TK06uBqRw+HI/HfVo17SDcjrUfPXq0Ll68WBcuXKgLFy7047JSg8ltmfk1Iu6HLb30OPxHGHM0GvWvHqctlAdz4DlBkTh8OT09Xevr6z2oZgow9PapBnh/phvhxe3t7V5wedGfF/wBAq/tGWAdmgX0oQN0IVTkEGGuZWEMpFfk5JfWGpD37UFTSnodzKfP5TMfw+O+5ue8nuKSWwJsrWNwAAzUnSFp8yGAZ7ngXq6nJ2lvyEaO1zB5fxxzZWOF0ySgqemB8bi1tdW/9y4Pq7aHjQxmFmbXXd2rhWzNzc0N9kTyH6PFusZykrTvuq4efPDBuuuuu3atF0J7EjEyNE6xR+lUe2c4+t4sk4z1SaUFcnsB0814WpSvCOBqCS3fUzn7N/4bnCgoWU+gGTsV+qT+ES5xXVznvxfnq3bCdAloVTUQDk5593l9aZEjzF4boo0MI/J8WmhplbkOrwfY4vKJFgheWsCEexifN2yiKAwKHlt6Kw4TAgAoGt4gi7GRa1yeWwOOQ2tuH/A07TIM5pBSei5W1OZDK+bkqeQzA4GvUVrz6t+YN5S012NcDzRMHnBdGVLMtieNIenlNi0fOVbmKMdMH8yTnis8OK8P4eWbJ1ohUs8Z3zGISKxq0T51RXpdOXes6a6srPRgaOAyCBH1SfqkQWN5mQQyLR14I16RZWovx+Fmy4EHrmRivkOwFoOn8sRqtmA5REUooQVi/jPDu82tra1+YdcKC2vIJ03Q97m5uT7zDkDY2tqqhYWFXhmfP3++VlZWBqdrIBAkIbCYi3Lf2NjoMwTJmuQ8OISKd1kZRAxSGfIxmFFIwDAIWVhRGraW7ZUZ0Cx8VgiAXSaKMH+0wxoZiS+ZLEJxske+XgNPl2QBAydzk6ecZMiY6+bTPL0bbwav1qBOPU5RNy/SH4wbt49HC4/hiXi/m5Us/c3ohRVnAk4aTY5ItJSr+TXnND1DRw/G43HvVQMeeK5ek2XO7JXMz8/X1tZWra6u9huAMWzsAWb0wTxIH3mXHNEF9IU9rDS8PGc5bvq6vb1dZ8+erePHj9cdd9yxS88wxvX19Z4XbUx03dXQ/+bmZs/3lom8v1X2WrdtlZYRmOWRBLQDDVxVk93aSRbfpGczPIaS39ra2pXZl6CI8MPULaXoZx3OAADsGQFm09PTfezbYaX19fV66KGHanl5uRcUhJm6UEQIB+2zy55w5Hg8HrwU0fe7byir5eXlXiH45XczMzP9OgqKwMoK2li5HzlypJaXlwfggPKxx1RVgxf62ZBIyxY6OozSdVdfed51XX/QaaY9c83gyHfuRWHQf0BkfX19Fyh7TcYvcGRthefhL/eDYj5KxeB23B/zJ0DrMKVD2wYv6k8Pi2soSa/P0R8n7LB/zjRImUzDwd4IzzN+e6CtUNtoNNp1rqcTQsx/pknV1a0mtLG4uDgAG4/NRgiGRnpoW1tbg7R+DCqfXD8e7xzdhYwauG0M+Zlbbrml5/mZmZnBIQUOUTNeRwxSD9qYgJb0eRLY7AUye3lvWaz/brYceOBqCXbrt/2UnOB83UHLasrrMJutbitQ982JCra8YCY8J9bHyDbk4FyHEHIcDm1QpqameuBCGfu1Dj4aClBGCFuhJASWa+w98zpRWvEGdsfcvVZmK7elnCcJRyqttPxTMfp8x67remMCYDY4oSztWfKsLXrT3okjGCQUg5zvox68v+QZA5T50nPfsqjNp1mnx+JxpKdgL9bjMZ9BD9ftzwm6yfsOE/rECRtTgC5glS9dtEzl2HNOGbM9XwMXHjB9Y2xpqHrNFZlxYoojCB5/ywuhj2QLz8zM1OLiYs3Pz/c0NlCbrswfxqZpkbySfJTtXy/AtHThJD7w7zdSvmKAq2p3ONCC1XXDI3ByjSHrQMm1XOpUwFZePJtWo60i2ufNxs6Ow3PyWWfO9vv85z9fly5d6j0IlIhDQy3rHWXMeYa0R4zf48y1HeozGAHszhgk5k4CiEM/9rpsadqbM8B4blifog6Hfy1kCCx1ttZi8BCrdtYLASkvsNsLRqk5M9P1MhY8PebFySPQxd4cY7PF7bqs2O3lZGg7w0S5lmalaxnwn8diALYs+dUx8IzrhFa0xbjSm6A9v+fKfYUHzcPQBePCRoe9MtPddUJzxmgjJX9LBU9EwuF0JzuQ1ON2kC3LvZOPaIfv8BWgCo2Xlpb65+fn5/vPSXtnZxq4/I68FoCnZ9wCr+sFGNpxXWlQme9upBxo4LJXkBZCWokQj3tb3lQKLAobRW/FaiszJwELzABHCA2BQwGYsVFmgNaRI0fq5MmT/frJ5uZmXb58uT9hwgoCxoUJzXgWdhQrY2DtjRAcoS8EzwCd1hQvZeQaa0mMj1epsF4H/a1MEe4rV67UwsLCQHny2VlrVoiAFMLuucRSthBtbm7WxYsX67bbbut/y3dmcRKJEzw8v8wlvOfNrg65QD8DmhWvsyPhTzxq6sGAQWkyRtr3WhU043QTywV0MX+af01P6mfcaXTYWKJdGxXwb4KnATaNOIyJllHJWKenp+vEiRO9p2VDLT0prwcy99QLPatqYKR43TB1CfLM60mShl13NQTNdguMUXiBeV1dXe2ByICLHDIueJ5xXL58ue/rHXfc0R8nlSFFJ4AZXO3dmx/S400jcD8eV8vLsiFgmiYf3kw50MA1yQWtmkyg/bqq9iysOFv1OQaOMqEdCw/7npzc4PsoDqF13dWF1q7r6ty5c/0mYyselIe/G7TcXo6P+61kMoXXjF+1o9ic9u14+ng87kHL+9asDF2X+2PLD9qipHK+rby8ZmU6GmhQmAYb18UxWemVJF+kF+D+pjI13SaFePzdFmmG3bwGiLIl3My9hJMMFg6HAQY2dLjfc5j9Yr6TFg5tT7KoWwZVRjx8D312uNaGTHoBac23vArq9dxZ/hzxSH4nAcIepNeX7IlzX4If9eXJMwnqyAcyTh/X19frwoUL/ZFwphl8Y88GHjHAphdk3pikF/cDXq2SPJQe3c3UXXXAgYviCWu5pyZiSxFxv7/zGSHHK+F6TkArzp8TwwbD/D0FjAwwvpP9xInvhBzMcBnLpg2HVDzOquHGU0JBVcPwFf2xYkGJ2cOwQrpy5Uq/8F2186pxh264P9dNco4QQm+2zFCUrUsrSHu9fMaAsLeCkvTp9rlvx6DMd1vqtmb9u8dDn1O5p9KBpgY6K0Pq4BXynA05NTXVZ7o54YJ+AVwoT4ORQ7QOLZovfeYldDU/VFUfmkrv0wq6pcjSIHBCSUvB2ZCxhW9ae9747DC463YI2ECM/EHz3NyeW1Y8xzaWzO9ra2sDWU0Fz/oYfSFsSNQmDQzGY7D2d48/+TcBPPtzPeBifdZyClKub8b7+ooALoqJXrXDqExOhn24x16DAahquN5jqxQwsxKCkZOxeXZ1dbVOnjxZR48erePHjw+E3y+oc4gL78tWGr+RxeT+I8T2tKw8vM6SXg39YZys/xA6tIBa4fCOIAuLD2716dj0B9o78YMQiENXtJfHGvE8ioOx2apG6I8dOzaYbytZ2kSh056TOLquG2SDocisoLxWRn8cLnHYZnFxsU+dNg94k/poNOoz3Tyn8B60gEbwo+lLKrSfGY12ThJhbPzmze4oYR8u7BCXvQVnim5sbPQ8Bt+QoWtjzOFtvkNHQvNeo7GMGRygL7R0+BUa+F6vcSLX8A20sqeSB1TzDN95RQlhZoCHcD4hRmcRw3eE/+k788kB0Y6QwCdf/OIX6zGPeUx1Xdd719AlX89Ce/Pz8wO+sOwzD7RhoLqecCHPu7SeTTC90XKggasFQlkmhffSG+O6w3RVV70kW7WOGdOHqt2vy0hFj6UFk1hJdF3XH5WDdYbSm5qa6sODVi7eG5VKjMzBtPbS0vVhr1XDV7079Gagyncq+XQK6vfpC2nxQR+A3IkOXnfMxXqDhNdUfOxWniLhdSQnhBw9erQ/vduhRLef1qst283NzV2eLL9ZucIzGBl5r0OsfnaSZWqjKY0w055+p/djI4z+MIfQzX/2agyKPlGGuXH4N+fb3gd1Zzq5DS0bBeZxr0mlN5e8ZW/Wns/s7GwPpBgDeQK7w+b2xEejneQT98vbSDBYWNclyctjhz7mV0DS+8I8HuuV5eXlunjxYj3mMY8ZHH6dBiVzMAmAsi8+DcX8d70FQyRLenFftR6XJ8me0yTC74dQqeDzdQAtwtvLy88WLr/OxMJuBrIlhKLb2NjoLUWfDE1/7U1W7Zy8gVKyAqGNDN1UXfsFmikE9i4oCGALMF0v93kMeXSQwYLQEZYrdTrcZ2Doup0swMwmm5+f77cEJJC3Qi62Sqt23o6cwpn84bUn+ubwLCGpXMimJPjzbMtCNrjRnhMV+M1A4n6lgUJfPXeOOmQyQ4uHGWdLeaY8JWAZbOhjrtPSR69L+To84GfgHWdF5ukTDkHzu0OPNkidCOMwHHXkHKZH5KiEjy6zDLhvVVff6bW0tFSnTp1qvuHa82IjPY1I841l2vMzqe79lGsB4I2AIuVAAxcTkeGnvQTEytCfudcKiu9Yh1U7Xg1hOpiVxAu3byXMcyhTPsOYvNqgaufYImLavNF4bW2t5ufne6vbIUkOpvX+lqmpqcHOehQlJ607JRgB2t7eHiSRkGGHQPkMQBRAVfWZVJubm33Y0N6SLWx7RM4KTLBCKAnD8Jpy2t/e3h54NemBAHQLCwsD3vAxQH75JeFBW/epPFE4KKVcs/K+NAOxjxVyphzvHKPf3hdnXkz6GIAc9kMZtqx28zBrkPYKCOVZYVG3PQoDVK63EJrjNwNcAiN0hkdbcko7lnMDREvRpk4AdDln0x6BDRODiqMrmWzBZ/gdXmbu0A1ERmh3Y2OjLl++vMt4mZ2draWlpbp8+XKfmWggNvhPT0/X2tpanT9/vt/aQoJOAoX7iTz4jeE5zy292fKcJhXozGfa8Hd//qr1uFolvZ0UgrR8KWndUFKRZDjQirZlQVjI/Izj6Fhu6TWMx+P+LMIrV670IQ3WyC5dutT3jaQPKxHGYqZ2m05ld/afQQbFaMXqlGJ+cyiJcIpBsWU9Ioj+PDc3V/Pz8/0JF1U7G4XpvwXOQkgCDRs2qTPPgkzjwvPisCfXMxTlz7mtIg0Vh+don1Cjw572gM03pjv9ymOibJ17Xc7p+CRQeO7tYXtM8IDH5RMkxuNxb2hBd+jieXZ4kpCcTw/ByPL6mTNyCd/SvmltebPxwDjNY/Qr9x0SyQBcGKuzYJ0FzFg9Ph815Tn3mp3f0IChBJAxfozMruv6tcE09Jh7+O78+fN18uTJgbHl+bOhQdu5fcNzbPC3YbSf0vKubgaYrlUONHB5UlsFYbBApZV2rTq8hpP3tpRx9s8lBTv/HKqwgqB9x/zxJlAMaXG5PV9vhQJRZO6LAc73OTtwkoVGPxH43GTrTDmHrACuubm5/kxGgND9smWd4Rk8xFb2pj2ODJNZyO0pp0Dm2opDi77fip9nXRhvPt9SVi2+y/nLtrC0udfrPi2en8SP5hvTPGnqrD6DtY0jQJV1Rnu1pollwYanPRXznO9p8Uh6nfzPtVjfZ5p4zjwv7m8rxOx5AKC8noY82cCy0eC2KA5Brqys9EdxZeTB47C+yLHuVfZ7n+/PficfPVLlQAMXE5OhHAg2KaTANb7b0nXpuq5PjID5UrnnQnn+5RoGYJB7YhyimJmZ6UNyZBtOT08PXqZYVXXs2LH+mvtvj8QhRb5XDV9ESEjD57DZ8uVAT7/6nOsWBj/j7EXTmXYsyAiVgcxvKrYSSkDwuk3rbD7zB/3xwjxWNH8OE1kp4snNzMzU6upqD+C55paJO1ZuphOWd3pznl/zrsNHZAtCD/fZSjRD3DYqnGBgXjbNWyE0xk3Idmtrqz9xhL4RLiRERiEsxjiswL0emOujlte9DE8+O+uS0B3fqd8H69rQanni8FjL+LKhw7gyc9fjNagcO3asDzcSDqeelZWVwXjNX4Tst7e3a2lpqebm5urUqVP93knoj1wjT3l+qOu3UZL03E9prTPSVguAb7ZcF3C96U1vqt/5nd+pv/7rv675+fn6R//oH9W/+3f/rr7+67++v2d9fb1+6qd+qn7rt36rNjY26gUveEH9l//yX+qOO+7o77n//vvr1a9+df3hH/5hHTt2rF7+8pfXm970pkE8fD/FyjotI6yM1u9ppdnCyYmatGCaFrIVaMtSdSydUIZBwKGFI0eO9K8sWVpa6hUqyhUlvbi42Ic/VlZWBkrfXg39S68pExf8jAXGcfmjR4/2IJNH4Fj5oFjH4/HE17Zw0K9BzKFTMzxC4aOnPIZUhnmixOrqalXt7NexN+gwaQoyY1lZWemv+wDg5eXlPkMRJcz8ey2MFHrG4hRyxpKKBV6x1wmf2vOmzkz28LpYK2RLfU5isUI3qG1tbQ0OB/a6SatuG2c2LAB682Su9RIyS1A3IOVJIuZZy71l1nPaMkK5n35bj3C/Q7GOhIxGo/7VJtTRAmDTEJkFsDBYAVIvUzDP/EYfVlZWBvVwYo29TkArIxfWh8gl35Gt69XJk0rLcbgZILuuXn3gAx+oe++9t571rGfV1tZWveENb6h77rmnPvnJT9bi4mJVVf3kT/5k/c//+T/rt3/7t+vkyZP1mte8pl7ykpfUH/3RH1XVVSZ78YtfXGfOnKk//uM/rgceeKB++Id/uI4cOVK/9Eu/dEODmEQAEyutf35PEON6Wh17WSItyy/7YaCrGmbwWVnBPFhi+fZfA5yZ2da16wQsadMKxkrOY7YisTeLAHlvmpWF++OTPGZnZwfhGTw1n/qQwGU6GpgYv8eYxoKtSXtPlFaY0PfbKqe9TPe2lwKYesz29rN/XqtBMdjjTh6yomnxsMdty9lrstdSWNyXewW9BpbrYemd2IurGoZkbdikoqROriU/pnx7fPAU90GDFg0BG69hpTFgjz7nLr399L4oyCOfc3ze7Ox1Mn6nf3hPpntGeFZXV2tlZWVXRq4LbZmu6bV6zCkP+y0t5+BLVa4LuN797ncPvr/tbW+r22+/vT7ykY/Uc57znLp06VL91//6X+vtb397fcd3fEdVVb31rW+tb/zGb6w/+ZM/qW/7tm+r3//9369PfvKT9d73vrfuuOOO+qZv+qb6xV/8xXrd615XP//zPz/YcHitYsW5lzuaRExrLr0ngwf3W9HTthWoSwoMxZk7MB9Mamsca4tQIQKKAve6kL00H7OEwNlCs2By3XSzde8DQdn7ghXmulMxMT5vIgW4trevbq5kA2ZmPKYSp9/OAvT/6enpXefcQcdMpIGuznRrGTDeRkACgr0KwoOed9ry4cVeL3MIZTzeeXsvSsGvaXFI1P1yuLdlWJkveJ6kA4cHE6hN067rBskLeNWAsjMMvYZjmqfMEEbE0+I3A/36+nrf9/SW0uuyXOEZ4qnSLxtKKX9EKJzlSZtEVpwd7IiFAYS5IyTK+A3sDse3eNdGFTTyG8bNc8yXN3GPRlejIUePHq3V1dWeT5M3eMZzlnovPSIbXvstueY7qVgH3mi5KT/w0qVLVVV1+vTpqqr6yEc+UleuXKnnP//5/T3f8A3fUI9//OPrvvvuq2/7tm+r++67r572tKcNQocveMEL6tWvfnV94hOfqG/+5m/e1Q7ZP5SlpaWqGh6y27IcXGBsKxXut4BYyCmEcdKa9UkNtuhS4JhQr4dw/BOAYUa5cuVKrays1PLy8mA8KJKpqak+zJZrClNTU/3O+83NzVpdXa3Z2dm+fwgG40FYCXMwdm/edYo1Ase9JEGwlgBYeUMp6d4oslRSDiM5BGaPi/HmnNKP9ASYJ+aK8x5b61jc67UusiQTBEejUf/+LdpmPLZ6GQtz5z+n3dubY8wGDhdoY5A3jeAtZ/IxZvNoemlW6C0vjeJQUgIM/+2FHT16dHAwbXqe1O3tAAZoy6w9jqqr5weyod9APzMzU2trawNwIUuv6io4ce6fU8NzrMwNEQL4xaFx/3cmYx63RT/sMVknmP9mZq6+wYH2pqen69y5c/0amF8GCu+sr6/3h0ezuZ51WNOGrSHe+G++dNkPALVK8hcleS6vXW+5YeAaj8f1Ez/xE/Xt3/7t9dSnPrWqqs6ePVtHjx6tU6dODe6944476uzZs/09Bi1+57dWedOb3lRvfOMbd13PgV8L5c2cZizfY2vWY3UmXVo0ef+kPjm2D9g4tGJlisVPO1U7+7syQygtuVwjgIEdirLyMz0c13e/3JZDPBm6NHBBOxsFVti2bFtC7nsdjvH9GVr0HNoQ8b0GXvfRYTL/jrXqftsAsiGU6x70wW06fOtjw1xPrvN5Hrx+RFvuCzSaxJfmX/rr9VPayfCo++cQUhpopLpjvPh5ezJ+3jRPpep+Mz6vS06KHEAz/44RbM/Snm7OSfbbvEBJz8G6glR/GzrwBJ6heQ/5dXifNxYwluwb55jeeuutTWPTSwI21kxj0zp/229JwzHnLr/vpTevVW4YuO699976+Mc/Xh/60IduuPH9lte//vX12te+tv++tLRUj3vc46pqiN5JiJyIBJr9AF3VjjB5cdgWfWvdwIV6vFhvpeYxoKy8ITQ9D55x6AbFwT0kHTiUkl6mzzrMkAxHUGU2GkJrULOw2HsiXOUQo61Xh8isPOiHN0hb8NNb8ZyY5tRnj8OAAP0Qbis82jLgM7bMeLRVm16WAc91WUniERtIW4JtgDDApodpb8EK1uNi3DZqHGI2qLlN/mcyjJU6m6qtJFseTfJjymkCnOcw++3CuIgcmFat7FiiJ+6PadmaN9PJwOSxWp7dnoHL4GLPlDGNx+P+VTU2wmjLwAVvI4ueM/prAzD14PU4Aq3SMjTye9LoRssNAddrXvOaete73lUf/OAH62u+5mv662fOnOnfeWSv68EHH6wzZ8709/zpn/7poL4HH3yw/61VOCuwVVpeUlqFyVxVw8NeHf+nTv/ntPNbbrmlqnaUdWvdwH1ID82nNHTdzmtOEHI8lrW1tVpZWendfZT4sWPHeqEbjUb9QZ4wpNffiLvnuXJWrOPxeBBesrAxvqqrafcoMo8X4BuNds7iA5QQIkJm9sQMWF6Mt7fFc103PG2EhAnGRFtd1w3qdWjXIEW41YoJb8Neha3+qanhfqT19fV+PYL7DY78ZuCERnhbKJIMCxokuA4I2UPFGFhYWBicNp7ea4ayqZOwoLMpWdvMrEKKPXsrWnsL3hBrOYAHTFNo5/acnemwGnS1hwiv2sDY3r562oyNScs2Yf/V1dV+jbWqBttOpqam+rViQszj8XiQ8eeTZ+Ap+GpxcbHPYuXUGfrmscJ7lg/6Q6Yq1+2ZZdna2qqlpaUBUEBX1g+ZS28NSH5hbqC/deh+yiQgSqPEXncmvOy3XBdwdV1XP/7jP17veMc76v3vf3898YlPHPz+zGc+s44cOVJ/8Ad/UC996UurqupTn/pU3X///XX33XdXVdXdd99d/+bf/Jt66KGH6vbbb6+qqve85z114sSJespTnnJdnbfFQ3Gig4EoY+WeHO5hjJ5EmMVei5/zd0823wnHETagXz5xAmZPRYBit+J21lHLujFwWBkwBis/J8J435OVnhd8GZuVatXOGhWf7X1RMnmDNqiTLMOW9+e5swcGwLh9j9t0sQKcmZnpacyLG6084CPzVY6HujOshdeSngqfp6Z2TkjIOWZuzaMGQ883/cmMOvOXF/bdZ+p3HS2vNS1o87vlC+OrZdWn5Z/AnEZKGk8GXScWua8Oy+YYoCdz6TMBq3b2Ipo/XM/GxsaukDbjIxxuegJqlmUbujZ44W3uw+tLzxcemZqa6pdTHHXBkEmwSd1UNTyFpuVhcS0jK/st1qctp8L3fdk8rnvvvbfe/va31zvf+c46fvx4T8STJ0/W/Px8nTx5sn70R3+0Xvva19bp06frxIkT9eM//uN1991317d927dVVdU999xTT3nKU+qHfuiH6pd/+Zfr7Nmz9TM/8zN17733TvSqrlVgnCSGFUDr/takJOBV7awpWIj2+mu54mZ4GMkJAK21Ba4TXkuF1rKmq2rXvVbGmTFlAWpZYil0ppPT3608EqSo3wqNfvLbzMxMH/PPsIrXAlOhUQ9K355Ui0/SykyAsgdopT9JiAEl98mKycoDJQNAw1uAbyZjuLh+z2nycnorppHpkOBqnjXtJ4VhKfBBnpFpBcg47e1nGzYu/LzlzmCX/O7rqQsMSnhGDh/aa6Me+pVrlk7YSVrxR9JH0pa+pvEGTd2P5CX2Tz700EP9mMxfAGbLsEtjZT9g5PHst5j21/vs9ZbrAq63vOUtVVX13Oc+d3D9rW99a73iFa+oqqr/9J/+U01NTdVLX/rSwQZkyvT0dL3rXe+qV7/61XX33XfX4uJivfzlL69f+IVfuO7OtywGW0Et69HAYaHyfWml4q1MssrsIdjTygVou+mEDaz0uI8sKDNg1U7sm9+8AEuhfRQXSsUhOoeQEESsPYCjBVQ8Q2Yg9dNPDI+u6wb1kVloAKI+3hVkwMhQpTcpz87O1srKSr+G59OxUzE648uKnsws6rR34n54QzVKzJtTCRF5bQIakZ7MdXhzamqqt+ITgKt2th4wPurMk/ShPSdYWPHbs3SIyvvvuE4mm8O8GTa2F0DBcmdNNRNp+O83+Rpg4F17u6ksmSvLnA0ty5vp6Dn1b+vr6/2LWBk/iVLIE31mbM76o13rBdOI3xwCm5mZGYQuuTeNowwVQxPL7szMTN1yyy11+fLl2tzc7DczV131JC9evFhd1w1eaWTerxpuvUiPpwXejxT4pBdo0L6Rct2hwmuVubm5evOb31xvfvObJ97zhCc8oX7v937vepreV39MZAOJLQ6e8T4Zh79cD5OP4Jrgbis9rgxXcQ9MawXId794cW1trd+/5SNyDFrT09O9wrIXZauyqnogBzwM7PY2TIsM6TlUZ+/Anpb7AH3tLfIfQaI/LQuU/86ESjAzsJjuVcMT3KG736CcFnALNGmLa96y4LlGEdjShIYJ0uYNgwPAQn+o28oeBcq4zROpZJgvK/20ug2K3rNnIyLDVclHKNPce+k5TkPEYOL9U1a0KYMJeAbS9C6go+XWxhzAlWur0JR+Uh9bH+Bh6ndYcDweD7Zy5D6y1D/02d56a72H306ePNn35/jx4/0WFr+GZ2trqy5fvlyzs7N14sSJgYdrXWc+SWAyz6ecXE+x4cF/X8vfb6Qc6LMKbR1kaRHHhDQB9zNBtuxafdirDSuinEwY0gLtUKH367jOBGknm7g4HIfCsSdkQM5QXtadVpoVSoYLAYwMgTllvuXRmY4ZnqnaWbPwM47tJ/1df8uCpO6WUeJxZfiVNmwp+/dJc2WQ5XlA3sDYoq/HTT32mFxanojpMKneBIoWn9kLyLMaHb3wOOxp+LfWnLjPrT605jINx1ZJo9Lz28pehb4ed9bF5wR7A7Xb8fqgDSnPv+d+PL6aDj8eX80e5M0JABd04neMNtMklwhocz/lRsHLc8331j03Wg40cFFSaTscZwImw7aUtJVChhAdfqhqL2SnFZiWvxUcffK5dfxOskKGB73YbKVpELKiHY/H/dFK3gtSVX3YDYvRa2lskkUxZUIIfaQ9EkjSO2KsKIBjx471vzNW5szXCd9QH2PJJBksVydpdN3VUF2GtvBsDYoJfPSJ0JqTY1jU9/43p8ZbiWxvbw823zqTMBMw0kO3pwWtzYuMgfFwDiJjT+vW9SaoZngPPrVMtYD82LFj/XWfirG+vt5/tvdkI8SeFdcZH5mD5h2HS23QQRdHGJAXezItELShwvw65OvMU4dSAQA8Prwk5MEyzvyYj3KLC/fxmfGZDgCm25udna0HHnhgEE5njE7k8boh4ftJYOQw815OwV7FPE1bLaN6LwNjP+XAA1eLsCaS1zbsObRAy8zkuiGyQy9YulU7LjaK3W1k6jPK3hllCYoGVQs/QOg+ZyiT/lpRcAK7T6OgvsXFxYHwuF/8GQxzvcKvhzejmoFHo1HfB8CYflopoPRIJzeIJFh5XcFeAmFYwkFWXqZfGinQIMOTBlOKw3pOiSecZF6ib17PJDkDung+ne7sz57jNJhWV1d7xejjsczL7FXiN0LRzDs8aXmxcseI4aguGyoGWYe2M9kEXnK/fN3zgBLNbRY+3sl0Mdg53EtY1DxftfMaIPoOEHldq6oGJ7bneirbUeAF6GcA8DzRng0Xrzs6sgAP0j78hQxtb2/3J2nY+6Yegw99whh0/ZkVmx4n/bwe7ygdg1w28T03Cl4HHriytMIKLevhet1UBMpWtYEOhW6GafXDwpIAl6GCqampntlsNdvSNMCZ6TPjL+uw9Va1s7/JzJ7hDXspDkECUFYwjC9BwOP3M5PmpmWYmJYWNodc8zf33/wAnVDcGSp0n+wxeO5tJOwVAjWYt+rP9YX0nigen69hTCXIuw/ul5WHPd9UJswRxke27fpav+XYvf7i0FnOewtEbQy5rrT03XcUefLnXvIBoNgI8zyk7Oc6amvDsT06G7U5p+6HPWH4FQNlfn5+AJJJ2/TsbBSnnmzN182UL0WdLl8RwGUlZmFvKUPu84LlXm5rWtq56Gpmyd+qhumvqUiswMz0VTuvmlhdXR0kAAA429vbtba21ocn8DAc2jI96JPDb1aiVuAWWPfbAp6KmjZawJWv5EgQpB2HWWyB2or1XGZh0RrP1srGysoAAf2wlq1IbKR4XkajUTOUlDRxwoN/s5Fh/nS4y7zKNfeH8TBW1jwy1d4Ghg0l+pQhYPqdIIr1X7XjladXZtq5//CEvSI8Y/77GfOG35mV3jLGhtdw3H7XdYM1uNFoNIgQQJfcw5jGhz1Bzxtj4rozXa0rUh5aYNXaWMyYaQfgImpz/Pjx2tra6o+Goy/mTUeBKKkrW8aGZf16jXyP1e217rnRcuCBqyUkEMkWTRLPXg/f0y22wh+Px7WystLvWct323jToBnSoRiUEyElrDIzm2PRrJO4HcJU09M7p647BGDQRAi96dCfraynpqb6RWAzlNvmef+OYGD9ZTsIpa1dK7wMxWXYw2Cd4Szqpj7ACgF2eIX+M2/0xUf9OEQF4BrczS+24DkA2GCLkum6q5tYCSn6bdCMwVa/6WK+MQ9nKMip8hmGslftOUC5Qk8Uog+kNdjQH79Pizad3blX6MkKk2e8PtWicQsgu66rpaWlnucdXXD43iE22ubcP/iWsdswaoXa6DdvI+A6xcsIfoHm1NTVQ6+hr+niTfM2ajBCHRb0+KH/wsJCH67klIy1tbV++05uk6EtG5cuk4zq6wWvNFJ97ZHyug48cCUgJdNTrBRseaYFzr0ZuqmqwRpDKvRJE2vFlF5dKiT64vRkt5+KwIKZ9RrEbeW5HitMA4otS8fMW/3gv1OgM0PL4JjhRnt2WX/OU86PBQrr3Eo/hRWgQ9EZAFH6plWO0cUerb1YjyGF1V6J59vXHPbxfLpeh6XwKgz4eNxux2Ox1+frDmUxxgwJew6slDLMnQovjSlAy2CT91rxVQ3PlJwEkBkG9D05tvTkHDXxfNj4TB7I9bWqndM5uAd60z73ef0sjRau21vMuZqdne2NTdMbz8+y04p4XCts6Hb3C1w5Z3vddzPlQAPXXoqFYss41yY8KQkI3OtJc4gEYYURrWSzj45pu1++x8rW7+DKceY40p1PUHJJoTagAE5WxtyLJe02KXxGYSJsmRTTAi4r+zQSUAqtcEZ6l2nVmyYonjz70Ip3krC25onrhBgBbMaT+5D8mfGbD2wI+VpL4djSz4253GOP1HOddE9aml9a9yZNnCRkIy/bM9+4HYO2w3zOsEv5cijMv+dYUuG2gNtzBz/YU3cfLN/mLWfb2kPJVHXa9hshCDGTfOLElQQu09ZjOHr0aB91YfwYL45YJC8nPySf+t79gFCWvcDLvHAz4HWggWsScblOCnPV7vUT39fyWLhuZY4lBXOjvMzYydxY8emtZWo5yoa6efsxC+IUh7GcRp9AZI+AsRD6IVSZi7W2HC2IphGvXEDwUAK86wfBIQPLioWSL4akJNgkUCOw9Ncn2NMf7puZ2XkNur0PTvCgAGrpEVu5ouBsKTNuwsaj0aifj+RN/gyiXv/ymFmDSR7EQ+q6rg/nUW9up6DP0CFDorTLM9DOIT94LfmJ8JWVYlrt8LdL/u7TaHLfkc9yJPRFJAIeWFhYGBiSBk36idLnPh98zDzQL4ADHvGbx+EznwLCHE5NTfUvckzaQkMiAaPRaOApj8c7RzktLCz0z3iLg/fpoVt4dnl5eVeWMnXbELFBZ95uGRWWwaTt9XhdnnN47JEsBxq4skyykhMoDEgW9gxV2Dq2MOcibU5qKi0vIqclCLMYWO2FEFO3IvV6Ge1YWOzdVO2chjApfJQWv591f3jGmU7U6zUtt+d1EvpCe67HBsIkq4zfrAg8777PRxRhOFig3QcE1Ws30MFC54QWrwv53Wg+Tspp/F5PpF3/GciZY4O6w0leE/K6Wiqe5G+3R58c4mXdFJ7PVHK/ZsYyZA8uZc9p5p4fwKU1V1aa5hd7yqwV+hk+54ZsXhFkfnF7Vuj25KC3t4pgWOSJHMwF1+iPQ7KWa3tqACAGAiHAjK7QXxvIgCr0SA8NgEpgT4PR8s41G8Q3AzyW7UeqfMUAl5kyAaiqdimCFvD4uZZSRNhb4R3fk2352exrejTpsXGsC0xpRTiJEZJJrZjcN1uZOV4rdDO7mdwKPJWFQ4EOA/J9P5ZcC7SS7pOep29ZlxWSaQHd3TeDDL8DxE5L9jqEPSkf69UCIhR+bn6ddH/OkQHfYOLxpmJOhez9XXhm9lDSYzcPWIGaJ3I+JilG8wLXLIuu2/xkwKQ/rc3OFBuNbss8aNqnx5D9sR6wIWnQSiDPKEyuMXL2IGBkQHKEKPvC+qZfsZJhd/fDfJD0yHsNfvsFHuvOLJPau5FyoIHLStwhLzOfFUGLWPsRMhcvDrc29Jp5aRum4r1J7gf32ponPIjHR9/Zhd+yfmwdUc9oNOpDEQg5B3CORldDbViO9owsoA4z2LJHyfngYIDCihEr0l5LWqbQxO1nCJZ7t7e3+1CSM9Lsrbj/KDgfMJzKM5UfPOXwGOOCBj73EXp63YuDdCkozytXrtTi4uJgnPambN066YL58foJ4/Pbpj02bzx3X3yvDQpnZdI+cwdNoY+PDrNBZT4AFNNrnZqa6jfO0lefduF5tHFEmH40GtXi4uJgPOY98wSfobPffGzAMtBYvtKoM1itra0NeI/nCZWjE/jNWwCSbgDU7OxsnT59ejDf1EsdzBveMbqBjEmvkacswNf5Khnzgo1nj38/pWVctQytmy0HGrgmAZFdct/XsuApZi7Xn8BEbN6WLWGgnCCeA+zIFkyrESUAwy8uLvbHNCFs1Et6PExtywkB8gkKWGXpDdGvBIgWcOd6lT05nvOxNP4d8DVAeC3KIZfWXhwrUn4jvDU7OztIUkA5Ufyc2/aeGHtfDqVacCkAdfKUgc/Gk9OsvbfObfhAVmhkoLcBZv52BiX0yNP/CeEyrnw9jueEujEGmBOHDU1PjDDmw/XY2HF7+buVMWM0UJof4SP4zG1C65wX+NJZiJbzVKI2bLx+m0YCn50ab7mycZY6xzKS2YjI5PLycv/SWhtkAFuOB1mwwev1boOf+dpRleRLhxet2/ZbErAS0G62HGjgqhpaRFVDItm6vt66DEyuw1a+73H7+ayVQLrKCKvvyeQFJ3bQtj2x1p9DLBkWaHmZybz+LcM19BslAAjYuk1mbf2ZPm7P82ilYe8n683583yk8mwJYgsgMjTKkUemf8570p/f0rBhLEePHu0TiDyuVMSpiKjb+8cylIfy9TOeP9M/56KldH0t36ztsZsGjKHFe5YBvqOkkx7MG+Bvoy3vQ5YctsuowF4GavaduTYQQ297Nn6+BVyMy4ajPTzrCvpNGzzXMqjgF4fsW7Lgtlu6Ludl0vcbKdejh/dTDjxwpTVXtfsY/6qdfT6tSU2Po+XOWmDtAeU6icNffq6qBuEQGJKQGwLrRWeUp+Pqzh6ywjLDeixeP0kFNQnwU4lX7YTVyCpEeKt23oCbYY1sJxe9fWiqs7QyGww64FUA5E6XtzKgHWdWmg84XYKwG2C0srIyUJ60aaXgvTv2OGytj0Y7r7ZI7xT+5BBlrymNRjvJDITGsg+j0agWFhYGhwX7eCHTH6/d18x/Dr9meJs+MG+Z3GT6mL+cDZpA3grpT01d3cCdIV/m3QDj9095I/nMzMwgs9LvNGNzviMC0Nnfua/V7wQW2mE+xuNxn8HKcz69grCcs20NloAUemN2drY/zHl+fr6WlpYGIAa9vfHciUKtvV+W4ZSVlgHp72kgXKtkm5N+u5lyoIHLhLUC82dn/HgCLLitxdS0oPi+sbFRa2truxaht7e3e+HJ7D3qs4Jy6MoK1tYcQsP1ra2tWlpa2mVdG0Cc5m4GhQaplOjb1NRUH/oAiADI+fn5gaJxWjHttA4ptVKAHngeHi+C7/WdFsCm50p9jMtpwy2Pj3Eb/P3WbdZj0ltJr9UeVHrf7qdByBlwnjd7QSgqew3O9qNuewFeC0FZem0q13t4xmn3Vnw2ihziNX1Yu7KC9zYRe+fmxdzWkZt34dU8XDgVPPJr0PB75tAJnA5inbC+vr6Lh23Qmo/4npEEyzr9cjg+DQHonyF3QruWCehLGjyHGntvFvQ6evRobWxs1Gh09eQayy+najAnDotDI8tOerAJXNdTLHeek9SlN1MONHC57IXkJhzFllQ+m+sL3GOFlMBlwMvJt4K2JWmlgKJwqMeg4PYQeF6bYc8qn22N3/9TKfA8NLJi9TWE3ovQLU+21W4rjOK+mFY5P54zg2cKWnqUKIc0TFy/1zUmxfXT6k6PzvWlgHpOeNbKmHuoM/uQdEOp+qw/Azp0dFv+HaByxp0V617FfUjw4Vl74PYUGIuNRvNG1tXqU8soMb15Bk+eseFZcQRXZufaEKQeGzsGpZZXAvBYhtzPlLGWvKWRS7ZhZgwmPbzXD91i3dPiy5TZ1jxf655rPZt9fSTKVwxwVe1WkHm9dZ9DVwk0VbuBCKZtgZetp6q9T/He3NyshYWF/jcvNmd4aGNjo1dMJGusr68P3m3lhXeet6KmHSuPHE/u0cl67S3QF4DLax4IrmP6fs7CD33S2/I6h72VDL/lW2AZn0Nzo9Go9xCc7WevZWrqagjRoZxUUO5T8pStWHubnndntnFfJnTY2LA3ZZ6Dntvb2z1veN7sPbm/DkmxEXtm5upBzuZXe16WB4wtEkLSCGBOrHwZVwIxXkXV8NU4DsNikHlty/zhe/yb5XBubq73BtO7XVlZGQCaQ+7U72tk/vkvwZK5bEVdUu+YRzECvd6dkQT6ZSPaPIYswIfmE2jqDFTzbssgy8/7La4nwSt1642WAw1cVqYQxN8dl08AsmKykjFDmUlMbLfFtVRwfsbuOnVyMCtKZmNjo8+QYy3pxIkTdeXKlUEYhNM7Njc3a2VlpQ/7rK+v9+9CMsPzzh7G5jBMWrTQi3oMpEn3qmFKdQKi/zJdmvWZzFTzs67T5ws6PMbp+NyXoNPyPlBUufY3NTXVh2SsZB0S9LgNNukZofxtODB34/G4FhcXB2fZ+XgePISNjY1+3cRJABToxqkNpgvjMY86bLW9vd2/w4s/Mlm3trZqbW2t5xnTkXHknFnR22OxtZ7eiuvjbD2HKClOpx+PdzJPs09W0DYCeH5zc7NPX2fMeYA1IGUj0uuX/s2Rkq7repmxtwRPAsh+FnCzx7awsDDYNuO65ubmes+LttkeY77gua2trZqdna1bbrmlHnzwwUG7lnnzLP/TS9yPB5710BdfS+/tqxa4knla1kFrUn3dQpCWQT7jelBkLRDLZxz+cTaa+28Gx0PwnisKezQAM+LtfjmhPQo8obRSW95pyxNNy8uAZlD3d9eX9G0ZEK25crjGKcmMC+Vn2rqODMXwvMHYiQuMNy3lVsjS9EtQZz4zcYQ2nNXHvdDV1rb7blrkmAzMWNRpybtfDjlBRwORE2Ym8ULOW853K4zu/tiz5ru9LCtM0yjXjdJAIkIwKcTaCpfRll+yCZiaL1rzMhrtbNVImW951KYp4Mn3jAJkyBBDh2QeRzUs29Rn3si5S7qYDqkPWtf3Ki3e8Jy57LfOVjnQwAVBnUVopvZ/Sira1j3XKjBkritMupd+IaAwnQWFtGiyiY4cOdK//8jCiRU+NzfXW4RsPrQQVu2ck9ba+5IgkvQzwxoMWum2GaqygDM/Dl1yH/fkGKtqYEk6NJVeRYIadTlcmmHLVMamgYGLdlA+lPRqW0rD+6x4FnBKjw26ek+XQSvplZYwXtr29nZvlVNH0sXP28PhP96wgRm6TDLUKFzL7FJ4JOfZkQlk2J6uDQvuz3pcfyaxZFvJb6ZnnhGJd+zruS8w6ZDh7fF4J3HEssV8O5zn8CyGqceIV8rrgzw3VcMXV5rfDZDQyWDnkoZJ6oT9FPfZ0Q///kiUAw1c6cKmVVU1TKrIkwxaFiKKxG04Rsw9jnfb82CDsq0pW4RWeLaGnJG4vr7enw7hNSiH/AyEznqamprqw4YOlTgU5Ri94+e8kt3MhhJh3Ka76TYej3sPEZrxDHThPnsRznrzGkpaevSRdx3RjpW2+9ZKdOGz06oNfOYbe0nZb8azvb3dhxczDMZ1nvXmUIMOoV7GPWl/HsoMXvH8eG3ExoOVuMPm5l0yBOEjH9pqmbHnTjtW4OkheE4BPoM5Lz+E582T0Kyqan5+fvA9owbIhg2jBP88EQSAXlhYGBh3m5ubtbGx0fMHxqm3BbAeyBjMT3hpjAeZcnYlKev8MQ7CtFU1CFknXS9dutTziw9/9qbk5LPFxcXa3Nwc7AHN/aCML+lrXr3R4uetd2+mHGjgqqrB5FYNDx1NC6JVEuRgWJ5JS402fGLDfixRrm9ubtaVK1dqYWHhmi8qzP/r6+uDg07xzMiQ8hqOMxYdhnE7tsjzANWkl4G2NV6Do6+5PVvXCLKfcbueU757LDkO35/zYA83N+ymJ2E+qqoeWPybBRtDxYrXfTItbEAxRw5tZbqz10wMSOYXgMD9Y2w24jwv0BBecUKET83IcJb7YBoAuPCJ6WTDpWVQZvE8prHoMTrTz5EAeMoGDOPL8LG9Wss8/bBHxHXadTjPQJ28SV38YVDCY5bTlr6CL7juJCrf7z1dlrGu63pPPg2i5AuXHNf1lJb85283Ww48cGVpMc5+Qatqt8JsCVfXdQPgyrbymtvwW2etmGzNuh0Yvqp6wHOmlffstKylquGROi26YBHbO8l+oxxyXGkgJJ38W4JCPtPqn8eTnkNrLP7ua4zL2YsOb3KfAYj+pEfgNphPrlsB2rNLHnR/WspmPB4PwkeZYNKiscfqcKbnKgHAhlrVECwSeJNWHkuCLX8eTyro1hy6LhuGGd5K73iSwZWp+jmWNEY8RocDAWiA2Z5DS8mb9zzHhO2rhsCVoJmRI4/H0QQbRClr/OFNJv1TT7TmpDVfe5VJ/H6jIDipHGjgmiSIzkryvS1i5vVUvjBEChRJElXDk9S3trZ2xcVtVeFxAT62oB2SJLR04sSJ2t7e7jcTopRai65YXQ738FsrvEffyW6kL3l4bK612Psx2DmsYWXoUMteAuE+YFlmYgIei+fN9bpvhGkQXviF9cBUXk4/byk6foPOXdf1G0A9Jjxy05bx2YqvGr4eI2mfJ4h7Thwi8ikQnle8qVSeCQSZCGPwIhMPz9CZdw7T0T6hW/M1/fXpFFa+ecAutE36OHwGLRPYfdaivSTkgfuZQ4dI8WTYpFxVg7As7aJjfF6jQ24Gp67rBun69v65l9NgbCDiBaaB7LW3W265ZbCh2rJpUKJeCjRqGXfwCPrOde2ntAxn6vC83mw50MB1LRQ3w6Ub74IgpfXKby1PKNcZDCIIiReYsThJe6ceW0sGOH5bWFioL3zhC33qN33MEzcMLuvr64Mwj8Hd4yOTys+6Pr67Xy5p7fHZoSgUeYJWVe2yHC1gpgOKp6oGQp7ZY54zp3p7Hqp2e1AJ6rZ8q3bW4egPypY5djo7dEkvGGW5vb3d94V+5YkrrZJeTYbUCDta4eY4mBdnpGF8ZCiJ++yRZDYn172XznIE/1keqIdnoYOjAjaIWvwD/V3S46X+tbW1QWiO64AdMuRsPY6Maq3HmD8BU59IAu2YywQIA7RT6PkNeiYv5ljZDgPwmo+hsb0zEle83zKNzlaZ1IebLXvp4/2UAw1cLi0XPcMAZpBJrmwSMy0FW062sia1WzU8qsbxdlva9mzcJxZt8dIoCTbpgWS4q+WZGpysCNNaS8/WIOPvrdAp/aHeFjDk3O0lSJOAyh6XaeNwnMfsfhpgc/wet+maoJTrBy0gTMXpNk2rpLP7kXRoWdYJyB5D8vckxUQf3VZrjYR73f9UjB7LpH60LHLm0DQxvydfZZ0ApuluPWAe87gcok3v2KDt9dFWCNv9NOBRxyRDpcUDyUdd1/UJLi2jI/sCQOaBzq2SntLNApc9Qc/BVzVw7cXQ9qSqhq9TsEC4WNhSybs9LB0vqHNv9o92CbHYUgeUcPcBLizQxcXFvk2yxMz49iootGOr196Q39oLnVqbgfkt96IYZK0ECJuRQUYhbMd16MEWAOiX2VC5xlE1TO+1Zcv8Ohxn2pDpaa+VvtpCtQUKvRwiJHOMvvnMulS+zgT0vOXYvIfHHovDcxn+SpB1+Ms8TRvOvHP75m3mwWP21grmOcOJno8EVMuNPSl75aaT+Zb5ykxLvAdODbGswscOh9nIoK+ZmOPzNLnOZ8ZvT9Uh1qmpqcGrjjLMhnz7z9ECg+QkgG8ZUY7CMDbuY94sF1NTU7W6urrLAGoZDSnr+y055+bDvQye6y0HGrjMjAawqp2JttWZXoknbC+CJsNwbZI7n1ae/6drnqECCxprWzMzM31au9/T5WcyHXg02smosmJh7cxK3cLoPras+jy5AJAHGIj9J5N63xu/OTyUoVbqTsGyArY1izJA2VMv85EKn/F6LQ2QtqXOHDuDzdYw64lVw1fdV9UgDZvMzzQoTAP3O0HeyiO9Ia+Lmc8BCGhkHk7gSAufvjnUZF5PD91zRThqdna21tfXB0ZHhuT57A39lhEAwUk18JhpAV94zNk/5Gk0urpdwfNm3uDaaDTq1ypTFiwrzDX3se5kY5VxwqM+5BdapSHQ8qQwwqAj4+FNA85axNgyUDrLMkvKm8d7M16XATjB8UbLgQauFjHsivKf0vKG9lN3C7SqhvseWlbuJEC0YjGAZv+3trb6FHj2drC2ZUXvOlPR25rzX+v+rM/hDa4ZNEwD+uZxe7weX8uYyPG05qI1XvqH8rFR4PrTs/BvBnue87za6EBpWckmX2T/PA8ZNkprNkFg0tzYO96vRezxpDJJHs8Qa47J/fO4U9FCK/O6f/c9rX7SVibpZHq9+550nERbP9uiE3SgOMxsPqjaiQTQhtu23GEkJW9k2NFAlyBswwJPEYO2xU/Jl0mrnE9f8//9lpa8tspXLXC5JAB4wkzIa7mnKUCTfqsarnPZAuXeVJwGuqrhq1VSQAj9LC8v95swl5eXd2WmWQAsHB4v9dsqtOLAknPJxIcEidHo6htgnV1G8WbTFORcl8h2W4CeSRzZJyzYPBmEPtkjqdo5QcQhUsIpaVHnetFotLPRNI2ZPAfS9IVOeKeEmrCQadOeWGZPUq/Duhm285zbK0chModpfLifgJZf+cIYrfz5bxAhDA64ttZeCH0m3QxQDmvnfLjPo9HOmpSzK90Hz5U3yduYsK6wx+awqreDkCXpfiPDZPhRfHJGHoZt/kIfODEkw+meAzZMMwcOQ1vuMixNG3sVh3OvB2AmgX6rjq9a4JpkTVsoHYqzAnJIwSUnmnb4b4b0mkRaiVU1UBY8j5LP0zFoG6UEc58/f76OHz9e09PT/a59CzifDWhZnJZt78uCgAIcjXZexEd/0jOCbl6zqNo5XcEeVdXwBYAGe5ROi4E9nzyHINOGj0+qGp4YwDhQaCgKFFqGRcfjcX/cEUYDdSW42mskRGXjpWWMpAK3dZ08xrz41AhfZ8vFeDw8UgiwalnYnrv19fX+gNnxeDx4OSehz5bRx397HvYWGHcaJ+4LdXrdND3MBNLW817HtbfCfQYnp8jnWPiDjkeOHOnDb9A+PSGDKr8jM3w2GJtf/ToV3r1mHQJPQjsAyTTgN9Lxx+NxnTt3rm699dZBuNpgyPfFxcWBYZt87Xlv8f61yiQwsg55JMpXBHC5tFxlhMGMn4zeEjZbHS1PzULq9lpt+DeUJIo1rXs+W8Fzll0eFZVM3fL8bH1yLftqK83CYQs6FXOGyNKQaHmvFiSuuz1f85z6Ga/PmQZe63T9Bhr3mTG3DJCWV5J9oh5AlTrTUPLnpC1KGCUzae9WJsL4usNXBhR7eQmYeT/1YTjBKwYH94cxJZ+novT4Hfby0WOt8K7nblJ7CaA5zhb9ci3YfOtrft7ejj0xnrVn4j5V1SCE7TXMqampAcCYvy1nafxMkt2qne0OrIPZEKZkJqN5Leu+0ZLPu/5D4KrdwJUK0N9bytHPUVKJ21LL0loEpo4WOLr+3HiKUoVZWUSl39PT04MXR1btPqvRwOFi4LKCsGWXnqotPocmPS4LebZt4TX9DC4GkkmWXYJz1Y5nlUCA15jKh99s6Zv2zkBzPymeIwMz99l7tTJMgLZyo04rG+Y9s0RzrnNPmsND5jNnulmhWyampnYSIKzgAVbLQhp/GWYbjUa998Z3g5WThGZnZ3sP1+DrsHDygQECILSSN63zWcCXebMyN00cwvTvrShMGi/wpBNZ/NYG+uqNzq0oScpQS46yb+iq8Xg8AEvC9tQzaQlgkj40/W6mpFF6s/UdaOCq2n02YcvavlZJK6+VZVQ1DCNWXT1VwNlJMLCVjuP0njCAqWonBu61mosXL9a5c+f6/vDb5cuXByGLSdYiAuTQWitU5u8IFIkWjNnCTso14TSvw/iNuklTjxv6kW1nAbWSrNoRGjZrEmrxfSj6ltcMOPjAWxsDaWQ47OkQmE8/IINyenq63+Dq0K35hb44bdkeEvQBQDO9H35zZqTPpaMP1OdU9o2NjV3ecnoGCeL0A2XoU2C43547gMA6jHkg5cdbJAAtwBp6VQ2z/DgA16HspLHfcky/0jABWLzZ27SyV7e5uTkI+7WMEmjo9VGKaZybzc1rGd6nDW+xSFqZZ8x3V65cGbyfj7E4W5M+uY7WPFGss1r6dT9lL+/reutyOfDARUmvyaVFpGuhPvfbosv72deT1li2Z7Dju1NSURLc4wVUmHA0Gg2UvMdgcKHPDqVRf4ab0tqiP46B24vJcdmCTqs36cyfQ49+1kKVnl/L42rVb2s9k0O8lmTl0fLQsv/Qi3vSe0uFYA/N9aa36/nK0C88kv0ynT2fNprsKdmbSMvfnr69I/fNz7RoYyVMvyfNW4Zp02OGxyfNk+kJEGQ436HRDFfby7K3mzJlXrHsEsr1ve6rx+PsQcCVpA2v85pf0vhMfrSXbqCGh6quhnYxWGxEuo7kV49lUtv7Lem5mddv1styOfDAlcSfZD20nklCtoDHIRIrgaodr6kFlK7Hf5RUmBYGKxYnL/h8Nt9rpYcC4rqVx17WjxWY6WMr3kq7avheqhxfC4Ao7nMqGo+Lz60x51wnDXOcttat4FKwsw0rJNrwuFNBpyJ2/7g3wSHDnu5jjg/F5TrtXbifSYuWQjegGYSrdoOQv/Ms/50NmGP3HPg584eTgTzuDC27r+YD82EaRzYo+M4z0D+NA/ctw9kGHI/Dfw5nI7eZFJHRh0nGbwJorskyBn5z8k1GAqDzpJDhI1nMB6Ynv91MOfDAZcsRRkExeGESq9AM0AKVSWszVUMGteWUgl+1+3y2lrByuOf8/Hz/bh/CCZyowWkaCMLc3Fz/bp1c6zANyCJkw3HSYDweD6xbK097ej4tomqYwu/Nt601KgTESsKK1rSy5WvAnpqaGqwTODGFRWjPhRWv5za9oewDygE6O2NxNBpNPD/OQGA6QRtokKFHK00fAOy5SD5sGQwcxMpzmU4O/6yvrw/4OQ+2NSAyLr9N2Ws49modNagavkTSAJ1GGnRFgZp/4MGtra1B9qNDbeaZo0eP9uE0xmZv1UCTih4aGswI1yePOsORMKZp6IL8EbpLr85JIlxzMa/5dB48ag4sdlSFezc2NurYsWN9/XNzcwNdxUZlNnGb7gn2CerXUxKsHslyoIHLgp7eAwxpxjDTO+Zf1d5zACPk8wl8MFVrklsgyDOcKpDCfPny5V55+yDcqampWlhYqKmpqZ7pkjlGo1EtLi4OlGaGgrAI8+y1llWUgGCatzLgoKWFfHt7e7Aw3vJQXL+VuesHHNhflB6JQTHXgeytuL8GAcAab9ihUPfbYS4r+1bIyQCVgOQxTU9P7zpVHEAxj/Fsgi6FcQP4BiSDm+eUsU5NTQ0W8lvedCus5TUm79Fy+DC9NUcSMgxIWwAV47GHYoXu5yzf5kP4ID1sjxN65Nurc12Y4nZMwzx2zMXj5NBb+ui1PhvHvsYxV9TFWEjHdygROTf/YAzzdvUEqzTOu67rDRiH3PcqCcY2AN3nmyn7P4Tq73BpCXBVO97q3/L3Fghcq6TXloLufuSE+axDW25Yx96fRd/Ya5IWqj/nfqYWnUyP9Bjt+eSYMtySJYEmjYMUDrdJ39mkaes+x5melK+3+oXgWtFlv1uemu+dZEFOsrrznqR1tpe0y74kX7X60FrjmRQ+d79btJ1kqCUftMaZ392HNDRb9Vp5Z32pWFtyvdfctXhmEj+4r/tdhrC8Znvu/7W2teT4MYbM93z2wcDj8c6eRCcV4TU6bLnXOFr9+btSDjRwtSzyScqxpUApGe/NUFIL1GgPC6clPBlq8H0ZQsM6M3DNzMz0h9PC2Hhh+c4sLD6u51gNDFVDLwgPx/dsbGwM3jnWGmvS18o7Q3qmsUGEJBX6f/To0d6rTGBzKJU/KxfXB42wErmeAOG1Ap8NaC+NsaZCtDXstHjXSV05bp7xemTVjpfi/tJ+8n3SOzPbvE8o55D2MYygnS399EaSX6E7kYNUwtzvrE6eg9YeG31mvyI05HnCb45E8Bz9Zv48BpeUY9pw6M5hPry0K1euDE7K8DhbXrx/M3+abjkHvBPM9Wfo11Ek09Th5vF4XEtLS7W1tdXrD9Mx++hifkudOWnLyqSSBtkkw+9GyoEOFRoYWhZBWiUuKD0+2xK3pZMhCuq1AvOp3BaYtJ498dyH0jYzszbm0+I95pmZmTp+/PhAkDLe7dMULAgeV9KD/wCnmddK0uOxkvJxQe5X9jEVFr+RdbW5uTkIE6FYTDeec7qvaeR5o/+2SK2wkg6M3WEmg0LyjUN4hGxYgzAPoKDML6nE+GxLuaoGQIJyp4/Ms/nB/z2XqbSZC3sJ1IvSZcxWQg41ZpaoSyp5Gx0OVfPf78JK+too81gMrnge5gXLpQ0t/nsfXxpHzNX09M4+SvNlq26XNMAcUnV4kYN5E+yd0WtZ9pwAXB7r+vr64IBixuqT6vcyhpChVrh4P2VS9KE1jhspBxq4UkCztAg9ScB8La2NfD69KyuutP5SiCyoPi4o6zQTmfn5Lc+x48y7DDnw356Z+5PeUHoZVtoteue1bLNqd9ZcAih/LVpZ6bXatGC5XSs8z6fH15pbF/qUCt88YNoalD02Pns8CSr+S1oYeP0cCjOzLvmcBoI9PBt9NkxybZTn+c1jy7nJsfhajslzaAA3j2J8eK5bijvr9LyZVq2wH0CVoMZzTvLIOW/xgAs80dJPyVctYPeY8rn8LWnhZQgiK+aXFrC0xtGS55stj0Q9Bxq4KC0FkwztiUUIWnFkfm8pRQumr9k6tGDk/hKypGBmrJ8ECsbE/xZweUPj1NTVpA3vS6FvGSp02d7eeTeXPSd7mLbmbIGbxi3lbA/JYUMEp6p2vVKE39jQ6qOQTEtAG3ozTwYKhzkd9ktaWxkngDPuPHfO/JLeJWEzA37Oib3JpKlBxmEhMucoDju6Hf6nF+DsNnvB0IB109XV1QG4ZYGObGy3kQYPWLYYg+eQ/tAPXh3CuKhnamqq/w1wsVfqsbrASw4bmu42agjROSvRwO/wqZORLIvme89jK7xG35h398PjMeh54/aktUUbNsw9mcmEXqtqED7ci4ZZ0iieVKz/JoHdIwGCBxq4WkRKS86fEaiqnUyo1rMGwLRwc4LH46svdDNocT3DSK7H7j9MDMPRv1QSeEjUf+zYsUGattvmmUxtd+YaguGU5tFo50Rq9wdBalmerZJg5bWUqh1AzFAKAkIadNXuo7VsUFiZkJ1HPQbytLh53nSmbvrP2K3wUDJ89iGogIiVZfbF4SErcCt6ez+MmWwy+MUAx72tsDMlw7ic+MIzAMyRI0f6NHfTwN6PjaaWx+s1Fbw86G8jg5Jp9wY8b9qlfs9n8oD5xTxtA8c09NzDO7numsar+YfPBvLWGmeCncOeaXhNTU3VyspKvxUl7/MfHiNGlg2Vra2tWl1drePHj/fX2T6BXuDdeSnPlkfL5bVKOgypRx8pr+1AA1dVO+RHsUWdntheBGx5b5MIb6FvTVq2a6Ghf2tra72iNpPw3ULmPnAMkuuDkVshivS6LJT2KlrPWPBa3pvHmZZxy1Ox0k5hsXKyR0txSm7LkMg5yM9WVnvV1Zrz9Nxa4SqDhsfSssrdRq6BtviH+zLUR5/MPxkWNs8ZFA28qbAMwi3vPefVfZkUkuV+e6tWuB6rvQKeyTBm/nnOUsm7Lv9mcHEo1UCUHorlx/Nl3kbeWyn1pp33mCLDPJuJMh6f59o8AEitra3VLbfc0sx0dKQl+dBnd94o6KTjMEln3Eg58MBFaQFYS9lzbyqQDHUkWKVycl0+hTuFLoUl09+3t6++zdUZQlZ2GUp0RhiWqj2B0Wg0sP4YWwpTjhWL3GNLC8vKw8JN36wUXY+9Gt/femXKJLAw7U2r/M1zN8nwsBFhD8ftJo94vJ7rDOnZeEgLOOfDpaW0cw7tGbT2F/Ib88MY0vJP48ntmgfz/lR+7ic0sxfXUlbpiUxNDQ+y9v7CFhDZo3V/sy8uBgTT1wDufWUZ+psUziNq40SWBHd7kXkAgunpcKXvJaMxDQEbFZ4jj5OoibfH+NncK2mddbOgBZ1c3OfW79dTDjRwWQgS3SkGjhQy7ue+BB6Yju8O+fD7eDzus/CsnA1oaalZuHOj4Obm5oCBKWYu6ubNyKxNMA6PE8XA9fRAHW+nj34XlRW2QxlWAlbUCLeTR6wk8mDRBIcW3XxfAhVrH4wx57BqJ3Q6Hu9k3k1PT/cncngcDo3RXlq6trQZuy1jlAJt8xugQl+sAJlLTgz3obz8eSsDxknV8AQWe20O+S4sLOw6VZ/i97zxPiofSpwyZlCEPxgvLzZ02I26mVOecbh4c3Nzl2LO+U/5So+Sup1qb35nLqqqH+ckw8386nbt/TP/RD6cvUdWKZnBlkcDKff71BXTGZ50+NP3kLrvsKZDlWlUs+HZtDdoeRuGx8349lOoK/nFsuYx3Eg50MBVtRuoqmoXA/i6iWprIsMa1yKqhQdmsLLNybNlnCBq5ZZ7Qkhg8L0wkvebZLv+zHevUcCkZmoLl4XQIb0EMbfjfluR2IpsrbEZHNx/K51JSnOSp5pA5/UqrwOZVxw2tPBSf84//1seOnW25tt1uM+c2MCzPqrHdR89enTQ15YRYAWWvMeaYxon7jP0zzF43lvhQNPVgI8SpY58Gzc08NzYUDMNk589N7TRSmyARvQlQcjrfG7bvO+wXCuxhu+mA0CRUZlJ/JGg7fbzGfZ+JSi4OIxoOudBBck/1l0t/t2rpIHseh+JcqCB61qgBdPlb5Pc3/Ry9lNQOLaIsk8tLy+9PwuSM7BQMlW7M64cFsowk9tOa5XrBi4DDfXRZnpZBirX53EbxLEGqTfX5Ey3VFZp7XtMSRM/b2PE48t1FX5P8LZSzvYnCV+GLtNT89g8fofk3Cd7VS2lZlB1qIjf0ghyn9IjbvF9y0pOg8zhUPqSSrZFt0lJLJbZFtCngeTreN45ppQ/eyL+3SFRFyv7Sfzp+UxjMQGZNtPDdHu+5ueS9zBe09ByOw7d2tudFPrNdm4EuNJ4bM3ZzZQDD1xZkuAuKHdCYZRJ1kFVm2nSarLH1LLGWsKf711aWVmpY8eO1cLCQh/iSYs5hcqe2OLiYu/5+ZUeznrDimdthzoNLPQPJufAUWjBob1WwIzBffN1FCUhIeZhc3Nz1/l8LSve31v7atxOAnUaMHx3ncxd9iXHQjF9R6PRIIPTCg1r3/vg7OXw5y0S6V0lfQ0a5gmuEaJySBAPxwqX52nT/bNys4ft51IBesuB5cXv5yI0WzU8ad1ektc9vZbn/jIHOTf26vIcQ+61rDIO+gh9CVtiPMAjGfZPjxFaeLsGa1SeB+aFsKr7gKwYFC2DgJSByOFR047Pq6uru7ITbTQ6/DrJ6LZ+uFZJYy3n7pEoBxq4XKyw+J6/V7WtSLvx3ON77ZV4wlvWZAoSQlu1c8yTmdkAY+/BAs2fhZo3zfqYmEn9QRk6Bm4ASIXqRWhi7yiuXMg3SLge18uzVrLUmdYrdWJBt6xIh4jSsk/AzPGZNjZYCGMC6AZxJ8Ck1+tipba1tbXLY8o6TfPkz0nzSIjHPGEapjdm2nq8/ssXMdpzScNpktdK3dQFIHu91rKRxpJp5DZNa4NYZmCmwYOM8XyOuaoGAMMcZzjOL05t6ZD0XJh7z5vvb53AUlWDZyxL7GlkrtfX1wc857l1+zbCAGLqML8xB46C+C9l8pEqLZC8nvIVAVw3guap6FrKeJIy8T38nl6bGcqC3hKyqh1l4T1dXnBFeGEwW8AGHBSMlTb3IKR7WVapNOlD0tfKJf+sHBKEUGyt+WrNY0uJp4WYY81xtcLF7n8KadXu0KOLvZlsCxonsOU8pHK41jiy764zlSTXDGQJWi6tU9xdR0s5tgDRdaXRkuOfxH/ZP9fj32xMmJ/TmPF19z35wrLZ4unUD8nf5gdHSiZ5+i2jzYaD6+b+NEr2MnYSxFu0NU1b89EC+5ste+nU6ykHGrhSSKt2M1VO6LVAyHVg5eb3FsFbMWQLu8M8kxT0xsZGra6uDix1P+PQTdXOxk36jAWZp0bYQneIJF9r4jU2Cx7n4iX9vBZXNdzk6jBmzo0ztSbNqa1OKyMrTzw4K1wraATTYSnPZwIzGX3OiPOYHdJxf62EmDOuYZDYQ/cG3VQe8IkVvvct4Ymbnx2ysyfPHCV40LbnA+OC98LhfXphP5OEnFAEza20GavrZk3GfO35sqJOxWkvgpBeGhtVO6FcxkB/CT16vcm0YkymbYYrPT9OjLIs4hkn/47H48G705xdmvR2cejXdHN41nphNNp5A7j1gw1b+t/SR3uB2H5Ky6BrXb+Z8hUBXD5Cx8WLqQlCLaBCCZl5fc+kxXra8nqRhYG/DLFYUGHq9fX1mpub26XQUBAc35KCb0EhROPwDeNxLDyzooi/5xgMDj5eyEf1kESSgOswJXNgQDEI8vuVK1dqYWGhB+BUWLRvMLBnCt34ncxM6J2H0doK5numpFt5cS9t8BxnTxrkraBaaxAcwcPYqdttt0LUaX1Tv1P8AbNWSJO2oS9KeGNjY9A2xgkgYWCCbm6TZ0wD+ua9UoBs1+2Ez1CmGbKzDLfaqaoBX5uXcq68B9E8ylhyj1yG6w0qBhPmoOuuHnALv2XWq/9sONq4oX8u8Mz8/Hytr6/3ugDZs3Fsftja2qqHH364PyC4ZeS5jQSpNCxuttjZuJlyoF9rUrV3huBeHlmuK7QsCgRl0m8uBijqzL7sNVle+5p09hzA4TUoj8l9S2C0orcyzGdMT+qY5KXm8y1aJa1drNRMw2zP9GvRcpI1mNZ0qz8ed4YU/XyGqlqWarZtz8se7KS+5rPJt/kb363AczxpACWYwHcGVICMcTt71fNGe56LDLWlp5z0s7GR85b8b6Dw+JJGOT/74Q3znuvKfvPZRih0SYA1baFlCxRMtyweO7Lv16DsRUMKx4Ul8OwFUq35uNGSYNWar+stB9rjMnOlBcrvFpBUHK21IFvfDqe1ANLClXuiWso+BTHBlROdvXk1J9cbAT0m15vCwjVnU1EXn/PkAMaHBUo7DrOQ/eh2qoYeMHXYmoRueXgtFjif/axpnkrTYbkWoFmBeMNmeuSpSAwe9vrMJ+5fy6pGgXFOpMM7PG/e8thy3JQjR47UxsbGQKFhXU+yos1rngt7GNPT0/2bcVsniyd9KF6zdDiW7+k9uQ+eRxtn9MEehPvi31JG0ytrAYT1QAKMae0wKP+hmz1hyyFZgJ5L3wdNkq4JQgZ+h4IBLs4ahDY2RNxmvtGZv0lHw7V450ZK8p7n52ZAq+qAA1fVMEMLBc11lJmts1axl2QQTOBwqCYFeWNjow/PJVO2LN1U2jA0qfpeazFDE+7wkTBksNFHmJK+Or3XhTBRWo+5RpWhQitdr59wSoABn7pNe2dJOS0/Y/aM36m8rss0qar+QFYUW+v0Dr/ri+c91+6nx0AI2cBtcKE+rzFlKJN2qANaW7D9DjbX37L8zbO0Q/jIvN0KYcNXePl4Wd53NGlja3oy3l6SBqT53jSxceSQnj07zyf86vaZB45LIiyb/bPMpYFqY84GLX2bm5vb5THxPJmoeD4OVzIGZz/6uDdKK5GDrTDwHglN29vbNTc3NzgZxHxFnzMLNjOWPRZ7sDZifD8GgMey39ICLc/hjZYDD1y2XimTPCTu539afPmsBTYteRjQHlduAG2BRQoywlo1FDD3y5a1x+VXE6BY8SpQ6JlS7j5Yebh+eyNJKyvQSWEUC6EZt2W9uQ+mr2nOczmXe82VPUCPtWV0tLyc9J7cr5bA8ayVQO6dadWbtMKzm3QW4V40wJu0ok/+AaxsWNlKd+KH20eZZ59a85/947/rYD3Qa5O02/J+LDN53X/Q0/zq8eSzVbuTjDyGFu2RWR/nZP6GvvawKDZavB/LJdfcaJtxpYdHPycBgevzPHn+Wrxu+bmeMgmcWt9vtBxo4ErmpHhiUoGl9daaxKw/BaNVMgzXAi7/RjID3/E88LqqdsDIlqUZzHtCGMN4vPNuJSvkpBsWI7/hZdGXzDi0NWpgzPUi/re8J9q21ZpCn/RLwLvWuGwdGzCc8j3JmLD3lutDpnMqbs8tHoM94jSuPG+0hYLzuYepmFpRAytT/+7IAP1yVhnPWrHjvadlzW+mRWbbmk9yfrjHe8N4T1S2lSHEluKz4eRQpOlk3TBJQVKfoxM5p6knbOQ5+9fzmHOXhpHnu6VbbDgmP/rN2q3lEb6bhglc3JOhTBfLTxp611NaIJayfiPlQAOXmSMtMJcW2NjKSgZx/cmYMFWWzc3NQTprK4Zu5Ul4xsKQyrLrdk6XQFAyhOjQhNcW6Ovm5mb/OvSNjY3+N3tktoYRSgOP+wRDsx7FywkdnjDtWoJjYbWAOCTJ/PDZgpjepNcEvF/M8+/QSa7b+LNp69AamYgtD9GhSnvNtDkaXQ054ZXb23C24/T09CCsy/2EUWkrLWN+H42GYdWuGx7tBA0z+SctbkcD1tbWdq2LOjxsPrSxaNoQwnPY0vdYFu1VMM/pbTDWlZWVwYZxG6WWU2jN2iC/ceYjKepWqDybr/6wrHpdiXuZQ6fEpyfGM94WkuuTKT8UsgkJTybomqfhtTTK7fVZ55gfsr4bDRVOKpMcgP2W68oqfNOb3lTPetaz6vjx43X77bfX93zP99SnPvWpwT3Pfe5zdwnCj/3Yjw3uuf/+++vFL35xLSws1O23314//dM/vSvd+HpKWnYtEMvffM3C0SJoyyPw9arhWoLva1mE9qqqhrF76kkBT08i92n5M0yb6cAOK1ph0s/sd0s5uW9O6DD9DKhYtGmNpgfmtbB8/UMqLLfnfTU878/OiGt5LK05os+0h5JxaNZ0cT+YmzztO9dxrAS5f2NjY7B+AV28dyx5EcMDY4L2DOgep5VZeoQGH4cZs53Wa3wyLG2gs/FiwwQa+zfohDJPIxRatUCBsRnMUPLwl0Gs5dlM0iV8dr8p0ARDw/czFsshzwBwVcOIgBOp0rOl//Db7OzsYD8YJeVwksFNX2i75QXlnO63XOv+mwGv6/K4PvCBD9S9995bz3rWs2pra6ve8IY31D333FOf/OQna3Fxsb/vla98Zf3CL/xC/31hYaH/vL29XS9+8YvrzJkz9cd//Mf1wAMP1A//8A/XkSNH6pd+6Zeuq/NWrPY0/HvV7uwYPlsBuJ4WA6QVkmVSvDotF4QkTwJvAWDVcHE3hcHKL61BC7ef9T1pibkYzFtj8ZoJZVLoIsMOrfayP61srpyfluXf+pzGigHFFvZe3ra9lBxDetf2uimpgFwyISKNjKpqWs6mg9u0lew+Z98MFPYykl+Tdlb4SVNoYF5LUKvaARDmmrYcNWAMWbdpbuCyksZAIPPVZxvSLpmZrTFOMnJbWwOSdpZJg0XSiTd250G5GcIzPWyAdV3XP+difeU1y71Cg6mn3OebLTnuSdeup1wXcL373e8efH/b295Wt99+e33kIx+p5zznOf31hYWFOnPmTLOO3//9369PfvKT9d73vrfuuOOO+qZv+qb6xV/8xXrd615XP//zP99nx+2nWImmorfweuITvKqGiQ18zntabbtgCRK7d1up8Luu61+ZjeWXa1wAA0KAd8Om3wwzzM/PDzwhF2cz8VuCgwXRYRd+s1XmtO4ETeqA5ljBFDKkoHuGe63AWxY/3hht+jNZklNTU7WxsTHwmBz6tBFiw8CekPvUWoDnP3M1Hl99L5tBOsOjVjpY0+vr64OQcStLzDwEPX2v6/Yzbj8VaIKfPQ/LEjQnc8/tMq+crMKckqrOZ67Dzw5Ptzwdr/E6o87zkt6/vxtcPNeWc/M1fGOA5XnzF3yfeymtZwzMBm7zAcVyYdrvBSrJh9CVMKhDjltbW7W+vt40mBh3yhd9To+S+bgewLEO4fsjAYRVN7kB+dKlS1VVdfr06cH13/iN36jbbrutnvrUp9brX//6/kV1VVX33XdfPe1pT6s77rijv/aCF7yglpaW6hOf+ESznY2NjVpaWhr8Ve3OHmyF1fi+VyjABLUyblkeDj+5LZguj+Px+gDXUVjpgVgAWdNCGSJ0tvYcn/YeKIOoBdqprpPeU5RKd9LaFcVWfNbXqttg7lCJvQZ7Z625SoXk+SVkkwAwiV+wzh0G9L1pgWLFchI/Kf2sgfkeKwHPsbPOCC8xXvpvmgIerVC2+TY9JpcEZfqfITzuxWvnfoMRY3IquOnpcHWrblLWKRm+5Tt8bLBx/12/lbjngRM/DGQpyz6oGH7gb25urn8jgkPOThJKfrRRmbxseQTwkU3zgftJ29DcmYa0nwa/5dhj87gtV9Yl9NNgmbTbT3FdLeD8snlcLuPxuH7iJ36ivv3bv72e+tSn9td/8Ad/sJ7whCfUXXfdVR/72Mfqda97XX3qU5+q3/md36mqqrNnzw5Aq6r672fPnm229aY3vane+MY37rqeYMNnlxSQvVA/rem0GPy5VUdLOae1ZK8qwyJVO4yNZ+RXIiRD049cd0lL1Mrd3lNLESbN7KX4GdrNMEXS0kBjAU7aMPasIz0O6vEc5DzZw5rUN4oVVcvSze8pcLaUW8DvevyftjIsnEZUegemSyqCSWNtKQyHpFNhZSKIeafFi9nnFr1aspTzkf8dysx5SP6mpIdjwIUXrUhbXlEq8/R0kuZp5LbqS9r4PgyWNGRNQ4NYyiBRkKRfS++k4T5pLm4WWCaV5OEbLTcMXPfee299/OMfrw996EOD66961av6z0972tPqzjvvrOc973n1mc98pp785CffUFuvf/3r67WvfW3/fWlpqR73uMdV1e4U08zAahVbH5SWovX9vi+vZd1uw4rN99gqz7UGNhTb6m2FPG1JOhSG94WAwyTcA338mpSq4Vl9BkALSmbBGQRbwpt09nen32N1ui0zN+OYnp7usyMJkThExDy630l3h/PS2rSH4WOPPKe2qNfW1vrn8bhdb46L4vWVliIdjXYO+W3xsfnBfeM3ruU6ksNhKErXlSc+2AhKpQc9CSOa35krCt+pL7dyQGfLcQJrGjK5sTYVtUHP7Ztm5lfXb3q4fkqCehqo7hfJE/CUIyvU6YSSNFIIe1v20mBFHp3cAu+1Du3O+YEeaVjmbwnK1yqer5bRfaPlhoDrNa95Tb3rXe+qD37wg/U1X/M1e9777Gc/u6qqPv3pT9eTn/zkOnPmTP3pn/7p4J4HH3ywqmriutjs7OyuEwGqdhPaQl+1O1HDzGBrxxaTmQFGuJbFYo/BryaxMCQguR0vsFqpmqk56oV+2dsZjYZp0A43tSx42kmwtuLLV114MZ26clHd4RgDTQvcsl0A2ko/Q2Uoc4dK/DK+VESeb9eV4UGvhRmgcwzu++bmZu81t05BgXZ+Hnrlfrx855kzRin0Lw2mrtt52aENICtewMmK3rTxNoTWgczQw96aZcKb791ne74OW5lP7I1Tj9dPW2MyvyAHLV0AH5rfLFtsUWBvFGE4ywX9TUPBKfxZEmztGea6smUUGTbQGAhZPzcdfR+nfHAMFDTMY7wSgCd5m77Xc2V9uVeZ5Ai47Rst17XG1XVdveY1r6l3vOMd9b73va+e+MQnXvOZj370o1VVdeedd1ZV1d13311/+Zd/WQ899FB/z3ve8546ceJEPeUpT7me7lRVG0Babm4ifQv5J01Ky/rIOrme1tek8JjracWXec5KaZJiRiAyROgYe4s5J1lBab3aw0rPoHXemQU///ain9tKWlrBeQz+c72tuUl6WyHZgm0JpeczF/hb40jll31LC99zRhtWyKlE3C/+5/Otelpz0KJjzn3L+KEY8PcqLeMwn0k6eAxOd881oVY7ppXHbx5PAyYTO7JM0jcGMd+Tc9+6x/zd0hncb0PcY+Mz2Yk2ciy3k9reixda7Vxv2Q/IXW+5Lo/r3nvvrbe//e31zne+s44fP96vSZ08ebLm5+frM5/5TL397W+v7/zO76xbb721Pvaxj9VP/uRP1nOe85x6+tOfXlVV99xzTz3lKU+pH/qhH6pf/uVfrrNnz9bP/MzP1L333tv0qvYqewFN675kuEl1TlqfsBXSEmArD373oblV1XtjLWsGQXSCwYkTJ6pq9+G/6U1mzN/K1VloBissYvpjgfV5gk7ZxZr0ojb9ayk0f4aGrfnJ/VaEA3nGY7XyZ6E/QS49Srfp16enkOf+q/wPjbHQTXPml3lKWtPnTHgxr3iseOi8q4m6HApkfhOkzG/QuGVJZ7jJdbivfobx81zXDVPvue72W0aS7zcdGAueUMtLILvW4GrvrKWk8aABK7z1ubm5/gABiqMa9Cn5yrxDGBC+8LmFll2Dgefa35l38zV8QLvmR655r2Fm7tJn2khvL40pg3tGFK632GDweG+mXBdwveUtb6mqq5uMXd761rfWK17xijp69Gi9973vrf/8n/9zrays1OMe97h66UtfWj/zMz/T3zs9PV3vete76tWvfnXdfffdtbi4WC9/+csH+76up0CASVlkk0AqgaNlxVrpZHutNmzppkAmkwAOPsUhLf/p6ek6fvx4Xbhwoba2dt7b5HHyHKGg1uZMC24qd/rlcBbjaq0t0LcM4RgkvN5ESS8OgG6Fc63kuddhO4Oy58pn8ZlGHgNZgN78mu12Xdcfxmvl5PUqK/dJgmm+yVepJ22qdvb1GGDYlIxR5zBZSyGjaMbjcb9Fgr6nhW9+dviVtu2pcz80APw9/uRv85/7auBj3OYvAAXDIGWbOr1+63Xd5ImUva7rBlm1rXFazmwoGljNXwYaG4Hm01xntLx7Dglfpj5w/2x40A7ANTMzM9i7xvv9ct4TWCktOlp+bNztp6QOcRs3Wq4LuK7V0OMe97j6wAc+cM16nvCEJ9Tv/d7vXU/Te/YpBWSSB7YXkJkRrITNKPlseg+2TigwSAvM8Kxs2VYNQ4WtM/Ys5KkQWmP2eCaBdNKoZXmR7djyYvycPQrG0LrfHmDSPRUgfYQ2vjbJekvrNutqCWjVcG3Jc5K0arWfnoyf2UtwrSgNXu5HAmzOZRpYNqISYLjHxpnrT35lvqqGBpDHknOf48rxWxlm2zwzSf4w1PKZltFpWU6wvxbPZRJJa675zX3MOUoaXUsPMXbutxHnz54TaJDHQNnYNC0n9WOSPN1ImVTXlxW4/q6UZA4r/bwvBSiFrap2WU9+Pj2TtC5dtre3+7f2Yu3YWswz4paXl2tra6tPPkEIvU9odna2Py8urSbvLVldXW2uZeSakZU5/1GSea4d16xYFhYWeovei+goEvaa8JbW8Xhcc3NzgxAb7U5PT/eWZe59SWuT9u3pOCsylYqNlAzR+Jrby2QIL2abvxgn4GDFQEjPr3/BSPFcOKRFX9JyNh96PK25cx+tEGgH78U0dqaajyuyV2Q+t7GVe4mgKWGylB17fB438+hEF7+927xjpY1nxjVnCzo8RtvQlTHhUVOc7OGwZ25X8H693LOZYJ5G36QkE3gm16jN4/7NiUreNsPZjRwfBn3QSWxItt4kyQj60x+vIeYRX9cDOC1jj+8tft1vOZDAdfny5cH/vysFZjl//vyj3ZXDclgOy2H5O1NajgXl8uXLdfLkyeuqb9TdjL/2KJXxeFyf+tSn6ilPeUr9v//3//oEhsOyU9jrdkifdjmkz97lkD7XLoc02rtciz5d19Xly5frrrvuGoTF91MOpMc1NTVVj33sY6uq6sSJE4dMs0c5pM/e5ZA+e5dD+ly7HNJo77IXfa7X06Lc1FmFh+WwHJbDclgOy5e7HALXYTksh+WwHJYDVQ4scM3OztbP/dzPXfem5a+WckifvcshffYuh/S5djmk0d7lS0mfA5mccVgOy2E5LIflq7ccWI/rsByWw3JYDstXZzkErsNyWA7LYTksB6ocAtdhOSyH5bAclgNVDoHrsByWw3JYDsuBKgcSuN785jfX3/t7f6/m5ubq2c9+9q4XU361lJ//+Z/fdTDoN3zDN/S/r6+v17333lu33nprHTt2rF760pf2L+38Si0f/OAH67u+67vqrrvuqtFoVL/7u787+L3ruvrZn/3ZuvPOO2t+fr6e//zn19/8zd8M7jl//ny97GUvqxMnTtSpU6fqR3/0R2t5efnLOIovXbkWfV7xilfs4qkXvvCFg3u+Uunzpje9qZ71rGfV8ePH6/bbb6/v+Z7vqU996lODe/YjU/fff3+9+MUvroWFhbr99tvrp3/6p3cdvn1Qy35o9NznPncXD/3Yj/3Y4J6bpdGBA67/9t/+W732ta+tn/u5n6v/83/+Tz3jGc+oF7zgBYMXU341lX/wD/5BPfDAA/3fhz70of63n/zJn6z/8T/+R/32b/92feADH6gvfOEL9ZKXvORR7O2XvqysrNQznvGMevOb39z8/Zd/+ZfrV37lV+rXfu3X6sMf/nAtLi7WC17wglpfX+/vednLXlaf+MQn6j3veU//pu9XvepVX64hfEnLtehTVfXCF75wwFO/+Zu/Ofj9K5U+H/jAB+ree++tP/mTP6n3vOc9deXKlbrnnntqZWWlv+daMrW9vV0vfvGLa3Nzs/74j/+4fv3Xf73e9ra31c/+7M8+GkN6xMt+aFRV9cpXvnLAQ7/8y7/c//aI0Kg7YOVbv/Vbu3vvvbf/vr293d11113dm970pkexV49O+bmf+7nuGc94RvO3ixcvdkeOHOl++7d/u7/2V3/1V11Vdffdd9+XqYePbqmq7h3veEf/fTwed2fOnOn+/b//9/21ixcvdrOzs91v/uZvdl3XdZ/85Ce7qur+7M/+rL/nf/2v/9WNRqPu85///Jet71+OkvTpuq57+ctf3n33d3/3xGe+mujz0EMPdVXVfeADH+i6bn8y9Xu/93vd1NRUd/bs2f6et7zlLd2JEye6jY2NL+8AvgwladR1Xff//X//X/fP//k/n/jMI0GjA+VxbW5u1kc+8pF6/vOf31+bmpqq5z//+XXfffc9ij179Mrf/M3f1F133VVPetKT6mUve1ndf//9VVX1kY98pK5cuTKg1Td8wzfU4x//+K9aWn32s5+ts2fPDmhy8uTJevazn93T5L777qtTp07Vt3zLt/T3PP/5z6+pqan68Ic//GXv86NR3v/+99ftt99eX//1X1+vfvWr69y5c/1vX030uXTpUlVVnT59uqr2J1P33XdfPe1pT6s77rijv+cFL3hBLS0t1Sc+8YkvY++/PCVpRPmN3/iNuu222+qpT31qvf71r6/V1dX+t0eCRgfqkN2HH364tre3BwOuqrrjjjvqr//6rx+lXj165dnPfna97W1vq6//+q+vBx54oN74xjfWP/7H/7g+/vGP19mzZ+vo0aN16tSpwTN33HFHnT179tHp8KNcGHeLf/jt7Nmzdfvttw9+n5mZqdOnT39V0O2FL3xhveQlL6knPvGJ9ZnPfKbe8IY31Ite9KK67777anp6+quGPuPxuH7iJ36ivv3bv72e+tSnVlXtS6bOnj3b5C9++0oqLRpVVf3gD/5gPeEJT6i77rqrPvaxj9XrXve6+tSnPlW/8zu/U1WPDI0OFHAdlmF50Yte1H9++tOfXs9+9rPrCU94Qv33//7fa35+/lHs2WE5qOX7v//7+89Pe9rT6ulPf3o9+clPrve///31vOc971Hs2Ze33HvvvfXxj398sGZ8WIZlEo283vm0pz2t7rzzznre855Xn/nMZ+rJT37yI9L2gQoV3nbbbTU9Pb0ri+fBBx+sM2fOPEq9+rtTTp06VX//7//9+vSnP11nzpypzc3Nunjx4uCer2ZaMe69+OfMmTO7En22trbq/PnzX5V0e9KTnlS33XZbffrTn66qrw76vOY1r6l3vetd9Yd/+If1NV/zNf31/cjUmTNnmvzFb18pZRKNWuXZz352VdWAh26WRgcKuI4ePVrPfOYz6w/+4A/6a+PxuP7gD/6g7r777kexZ383yvLycn3mM5+pO++8s575zGfWkSNHBrT61Kc+Vffff/9XLa2e+MQn1pkzZwY0WVpaqg9/+MM9Te6+++66ePFifeQjH+nved/73lfj8bgXwK+m8rnPfa7OnTtXd955Z1V9ZdOn67p6zWteU+94xzvqfe97Xz3xiU8c/L4fmbr77rvrL//yLwfg/p73vKdOnDhRT3nKU748A/kSlmvRqFU++tGPVlUNeOimaXSDySSPWvmt3/qtbnZ2tnvb297WffKTn+xe9apXdadOnRpkqHy1lJ/6qZ/q3v/+93ef/exnuz/6oz/qnv/853e33XZb99BDD3Vd13U/9mM/1j3+8Y/v3ve+93V//ud/3t19993d3Xff/Sj3+ktbLl++3P3FX/xF9xd/8RddVXX/8T/+x+4v/uIvur/927/tuq7r/u2//bfdqVOnune+853dxz72se67v/u7uyc+8Ynd2tpaX8cLX/jC7pu/+Zu7D3/4w92HPvSh7uu+7uu6H/iBH3i0hvSIlr3oc/ny5e5f/It/0d13333dZz/72e69731v9w//4T/svu7rvq5bX1/v6/hKpc+rX/3q7uTJk9373//+7oEHHuj/VldX+3uuJVNbW1vdU5/61O6ee+7pPvrRj3bvfve7u8c85jHd61//+kdjSI94uRaNPv3pT3e/8Au/0P35n/9599nPfrZ75zvf2T3pSU/qnvOc5/R1PBI0OnDA1XVd96u/+qvd4x//+O7o0aPdt37rt3Z/8id/8mh36VEp3/d939fdeeed3dGjR7vHPvax3fd93/d1n/70p/vf19bWun/2z/5Zd8stt3QLCwvd937v93YPPPDAo9jjL335wz/8w66qdv29/OUv77ruakr8v/pX/6q74447utnZ2e55z3te96lPfWpQx7lz57of+IEf6I4dO9adOHGi+5Ef+ZHu8uXLj8JoHvmyF31WV1e7e+65p3vMYx7THTlypHvCE57QvfKVr9xlFH6l0qdFl6rq3vrWt/b37Eem/u///b/di170om5+fr677bbbup/6qZ/qrly58mUezZemXItG999/f/ec5zynO336dDc7O9t97dd+bffTP/3T3aVLlwb13CyNDl9rclgOy2E5LIflQJUDtcZ1WA7LYTksh+WwHALXYTksh+WwHJYDVQ6B67AclsNyWA7LgSqHwHVYDsthOSyH5UCVQ+A6LIflsByWw3KgyiFwHZbDclgOy2E5UOUQuA7LYTksh+WwHKhyCFyH5bAclsNyWA5UOQSuw3JYDsthOSwHqhwC12E5LIflsByWA1UOgeuwHJbDclgOy4Eqh8B1WA7LYTksh+VAlf8fufCVQWXK6UIAAAAASUVORK5CYII=\n" }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "Training Set Image Label: [1. 0.]\n", "Training Set Image Shape: (256, 256, 3)\n" ] } ], "source": [ "idx = 0\n", "plt.imshow(images[idx]/255.0)\n", "plt.show()\n", "print(f\"Training Set Image Label: {labels[idx]}\")\n", "print(f\"Training Set Image Shape: {images[idx].shape}\")" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "id": "BLlIqNYkkAxy" }, "outputs": [], "source": [ "test_set_elem = test_set.take(1)\n", "for images, labels in test_set_elem:\n", " images = images.numpy()" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 452 }, "id": "fIqUVucsk_Ne", "outputId": "67d9baf5-3b84-416e-8310-055a68420967" }, "outputs": [ { "output_type": "display_data", "data": { "text/plain": [ "
" ], "image/png": "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\n" }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "Test Set Image Label: [0. 1.]\n" ] } ], "source": [ "idx = 16\n", "plt.imshow(images[idx]/255.0)\n", "plt.show()\n", "print(f\"Test Set Image Label: {labels[idx]}\")" ] }, { "cell_type": "markdown", "metadata": { "id": "11JF7kx1VN0z" }, "source": [ "The X-rays are not centered. The images are all rotated at different angles. Additionally, even though the images look to be in greyscale, they have `3` channels.\n", "\n", "The images don't seem to have a lot of information, but, because of the overlap of ribs and lungs, it could be difficult to extract features specific to the infection. A more complex model could perhaps manage to extract relevant features.\n", "\n", "Given the differences between how the X-rays were taken, the images could be rotated, at the very least, as part of data augmentation." ] }, { "cell_type": "markdown", "metadata": { "id": "2C-1UAgdVPvy" }, "source": [ "## 3. First Model: Simple CNN I" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "FN3Wio55Vc4a", "outputId": "55c64421-2e0f-4941-f07f-c09783c3dc6c" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "Model: \"sequential\"\n", "_________________________________________________________________\n", " Layer (type) Output Shape Param # \n", "=================================================================\n", " conv2d (Conv2D) (None, 127, 127, 16) 448 \n", " \n", " max_pooling2d (MaxPooling2D (None, 63, 63, 16) 0 \n", " ) \n", " \n", " flatten (Flatten) (None, 63504) 0 \n", " \n", " dense (Dense) (None, 16) 1016080 \n", " \n", " dense_1 (Dense) (None, 2) 34 \n", " \n", "=================================================================\n", "Total params: 1,016,562\n", "Trainable params: 1,016,562\n", "Non-trainable params: 0\n", "_________________________________________________________________\n", "Epoch 1/5\n", "33/33 [==============================] - 41s 742ms/step - loss: 1.4654 - accuracy: 0.7389 - val_loss: 0.3292 - val_accuracy: 0.8872\n", "Epoch 2/5\n", "33/33 [==============================] - 35s 880ms/step - loss: 0.2241 - accuracy: 0.9154 - val_loss: 0.2026 - val_accuracy: 0.9226\n", "Epoch 3/5\n", "33/33 [==============================] - 37s 966ms/step - loss: 0.3417 - accuracy: 0.9267 - val_loss: 0.1928 - val_accuracy: 0.9216\n", "Epoch 4/5\n", "33/33 [==============================] - 35s 920ms/step - loss: 0.2012 - accuracy: 0.9324 - val_loss: 0.2039 - val_accuracy: 0.9273\n", "Epoch 5/5\n", "33/33 [==============================] - 35s 883ms/step - loss: 0.0774 - accuracy: 0.9723 - val_loss: 0.1683 - val_accuracy: 0.9474\n" ] }, { "output_type": "execute_result", "data": { "text/plain": [ "" ] }, "metadata": {}, "execution_count": 9 } ], "source": [ "first_model = models.Sequential()\n", "\n", "first_model.add(layers.Conv2D(filters=16, kernel_size=3, strides=2, activation='relu', input_shape=(img_size, img_size, 3)))\n", "first_model.add(layers.MaxPooling2D(pool_size=3, strides=2))\n", "first_model.add(layers.Flatten())\n", "\n", "first_model.add(layers.Dense(16, activation='relu'))\n", "first_model.add(layers.Dense(2))\n", "\n", "opt = tf.keras.optimizers.Adam(learning_rate=0.01)\n", "loss = tf.keras.losses.CategoricalCrossentropy(from_logits=True)\n", "first_model.compile(optimizer=opt, loss=loss, metrics=['accuracy'])\n", "\n", "first_model.summary()\n", "\n", "first_model.fit(train_set_normalized, epochs=5, validation_data=validation_set_normalized)" ] }, { "cell_type": "markdown", "source": [ "Both the validation and training accuracies are quite high even though the model was trained for only `5` epochs. It is possible the model is overfitting. Adding regularization, data augmentation, and more fully-connected layers could help prevent some overfitting while still ensuring good performance." ], "metadata": { "id": "UtEUtLBG3U8b" } }, { "cell_type": "markdown", "metadata": { "id": "GfpRgOS8Vjrj" }, "source": [ "## 4. First Model: Simple CNN II" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "QzpLopPNVsKk", "outputId": "51c8a2ce-3716-45db-b66a-08fd82718fc2" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "Model: \"sequential_1\"\n", "_________________________________________________________________\n", " Layer (type) Output Shape Param # \n", "=================================================================\n", " random_zoom (RandomZoom) (None, 256, 256, 3) 0 \n", " \n", " random_rotation (RandomRota (None, 256, 256, 3) 0 \n", " tion) \n", " \n", " conv2d_1 (Conv2D) (None, 127, 127, 128) 3584 \n", " \n", " max_pooling2d_1 (MaxPooling (None, 63, 63, 128) 0 \n", " 2D) \n", " \n", " conv2d_2 (Conv2D) (None, 31, 31, 256) 295168 \n", " \n", " max_pooling2d_2 (MaxPooling (None, 15, 15, 256) 0 \n", " 2D) \n", " \n", " flatten_1 (Flatten) (None, 57600) 0 \n", " \n", " dense_2 (Dense) (None, 256) 14745856 \n", " \n", " dropout (Dropout) (None, 256) 0 \n", " \n", " dense_3 (Dense) (None, 64) 16448 \n", " \n", " dropout_1 (Dropout) (None, 64) 0 \n", " \n", " dense_4 (Dense) (None, 32) 2080 \n", " \n", " dropout_2 (Dropout) (None, 32) 0 \n", " \n", " dense_5 (Dense) (None, 2) 66 \n", " \n", "=================================================================\n", "Total params: 15,063,202\n", "Trainable params: 15,063,202\n", "Non-trainable params: 0\n", "_________________________________________________________________\n", "Epoch 1/15\n", "33/33 [==============================] - 44s 984ms/step - loss: 0.6785 - accuracy: 0.7090 - val_loss: 0.5788 - val_accuracy: 0.7629\n", "Epoch 2/15\n", "33/33 [==============================] - 34s 842ms/step - loss: 0.5633 - accuracy: 0.7344 - val_loss: 0.4263 - val_accuracy: 0.7629\n", "Epoch 3/15\n", "33/33 [==============================] - 33s 811ms/step - loss: 0.5017 - accuracy: 0.7480 - val_loss: 0.3565 - val_accuracy: 0.7878\n", "Epoch 4/15\n", "33/33 [==============================] - 34s 827ms/step - loss: 0.4485 - accuracy: 0.7802 - val_loss: 0.3006 - val_accuracy: 0.8901\n", "Epoch 5/15\n", "33/33 [==============================] - 33s 828ms/step - loss: 0.3659 - accuracy: 0.8273 - val_loss: 0.2303 - val_accuracy: 0.8967\n", "Epoch 6/15\n", "33/33 [==============================] - 38s 964ms/step - loss: 0.3124 - accuracy: 0.8688 - val_loss: 0.2547 - val_accuracy: 0.8795\n", "Epoch 7/15\n", "33/33 [==============================] - 33s 826ms/step - loss: 0.2994 - accuracy: 0.8779 - val_loss: 0.2138 - val_accuracy: 0.9073\n", "Epoch 8/15\n", "33/33 [==============================] - 33s 833ms/step - loss: 0.2984 - accuracy: 0.8803 - val_loss: 0.2044 - val_accuracy: 0.9140\n", "Epoch 9/15\n", "33/33 [==============================] - 38s 963ms/step - loss: 0.2829 - accuracy: 0.8832 - val_loss: 0.2149 - val_accuracy: 0.9092\n", "Epoch 10/15\n", "33/33 [==============================] - 32s 822ms/step - loss: 0.2731 - accuracy: 0.8937 - val_loss: 0.2375 - val_accuracy: 0.8958\n", "Epoch 11/15\n", "33/33 [==============================] - 37s 934ms/step - loss: 0.2654 - accuracy: 0.8966 - val_loss: 0.2083 - val_accuracy: 0.9092\n" ] }, { "output_type": "execute_result", "data": { "text/plain": [ "" ] }, "metadata": {}, "execution_count": 10 } ], "source": [ "first_model = models.Sequential()\n", "\n", "first_model.add(layers.RandomZoom(height_factor=0.1))\n", "first_model.add(layers.RandomRotation(factor=0.2))\n", "\n", "first_model.add(layers.Conv2D(filters=128, kernel_size=3, strides=2, activation='relu', input_shape=(img_size, img_size, 3)))\n", "first_model.add(layers.MaxPooling2D(pool_size=3, strides=2))\n", "\n", "first_model.add(layers.Conv2D(filters=256, kernel_size=3, strides=2, activation='relu'))\n", "first_model.add(layers.MaxPooling2D(pool_size=3, strides=2))\n", "first_model.add(layers.Flatten())\n", "\n", "first_model.add(layers.Dense(256, activation='relu'))\n", "first_model.add(layers.Dropout(0.5))\n", "\n", "first_model.add(layers.Dense(64, activation='relu'))\n", "first_model.add(layers.Dropout(0.5))\n", "\n", "first_model.add(layers.Dense(32, activation='relu'))\n", "first_model.add(layers.Dropout(0.5))\n", "\n", "first_model.add(layers.Dense(2))\n", "\n", "early_stopping_callback = tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=3, restore_best_weights=True)\n", "opt = tf.keras.optimizers.Adam(learning_rate=0.001)\n", "loss = tf.keras.losses.CategoricalCrossentropy(from_logits=True)\n", "first_model.compile(optimizer=opt, loss=loss, metrics=['accuracy'])\n", "first_model.build((None, img_size, img_size, 3))\n", "\n", "first_model.summary()\n", "first_model.fit(train_set_normalized, epochs=15, validation_data=validation_set_normalized, callbacks=[early_stopping_callback])" ] }, { "cell_type": "markdown", "metadata": { "id": "Hed8-eRliifs" }, "source": [ "The architecture was modified a few times to try and reduce overfitting while still yielding reasonable performance. The experimentation included:\n", "\n", "- Trying different combinations of data augmentation layers.\n", "- Increasing the number of filters, trying different sizes, different strides, and number of convolutional layers.\n", "- Adding more fully-connected layers with a different number of units.\n", "- Removing batch normalization layers after the fully-connected layers and replacing them with dropout layers. Before this change, the model's performance was a bit erratic.\n", "- Reducing the learning rate helped the model to converge.\n", "- Adding an early stopping callback that would stop the training if the validation loss didn't decrease for three consecutive epochs. Using the `restore_best_weights` argument ensured that the model would use weights from the epoch with the lowest validation loss only." ] }, { "cell_type": "markdown", "metadata": { "id": "kd9aQRZaWLZS" }, "source": [ "## 5. Second Model: Transfer Learning" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "_DbaC91yZumN", "outputId": "1970688e-f89e-4283-de4b-4d5f24559861" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50v2_weights_tf_dim_ordering_tf_kernels_notop.h5\n", "94668760/94668760 [==============================] - 5s 0us/step\n", "Model: \"model\"\n", "_________________________________________________________________\n", " Layer (type) Output Shape Param # \n", "=================================================================\n", " input_2 (InputLayer) [(None, 256, 256, 3)] 0 \n", " \n", " tf.math.truediv (TFOpLambda (None, 256, 256, 3) 0 \n", " ) \n", " \n", " tf.math.subtract (TFOpLambd (None, 256, 256, 3) 0 \n", " a) \n", " \n", " random_zoom_1 (RandomZoom) (None, 256, 256, 3) 0 \n", " \n", " random_rotation_1 (RandomRo (None, 256, 256, 3) 0 \n", " tation) \n", " \n", " resnet50v2 (Functional) (None, 8, 8, 2048) 23564800 \n", " \n", " global_average_pooling2d (G (None, 2048) 0 \n", " lobalAveragePooling2D) \n", " \n", " dense_6 (Dense) (None, 512) 1049088 \n", " \n", " re_lu (ReLU) (None, 512) 0 \n", " \n", " dense_7 (Dense) (None, 128) 65664 \n", " \n", " re_lu_1 (ReLU) (None, 128) 0 \n", " \n", " dense_8 (Dense) (None, 32) 4128 \n", " \n", " re_lu_2 (ReLU) (None, 32) 0 \n", " \n", " dense_9 (Dense) (None, 2) 66 \n", " \n", "=================================================================\n", "Total params: 24,683,746\n", "Trainable params: 1,118,946\n", "Non-trainable params: 23,564,800\n", "_________________________________________________________________\n", "Epoch 1/10\n", "33/33 [==============================] - 60s 1s/step - loss: 1.3513 - accuracy: 0.7697 - val_loss: 0.1894 - val_accuracy: 0.9226\n", "Epoch 2/10\n", "33/33 [==============================] - 41s 1s/step - loss: 0.1610 - accuracy: 0.9396 - val_loss: 0.1328 - val_accuracy: 0.9512\n", "Epoch 3/10\n", "33/33 [==============================] - 38s 965ms/step - loss: 0.1564 - accuracy: 0.9360 - val_loss: 0.1453 - val_accuracy: 0.9503\n", "Epoch 4/10\n", "33/33 [==============================] - 38s 957ms/step - loss: 0.1104 - accuracy: 0.9584 - val_loss: 0.2188 - val_accuracy: 0.9216\n" ] }, { "output_type": "execute_result", "data": { "text/plain": [ "" ] }, "metadata": {}, "execution_count": 11 } ], "source": [ "base_model = applications.resnet_v2.ResNet50V2(\n", " include_top=False,\n", " weights='imagenet',\n", " input_shape=(img_size, img_size, 3)\n", ")\n", "\n", "base_model.trainable = False\n", "\n", "input_layer = Input(shape=(img_size, img_size, 3))\n", "\n", "preprocessed_input_layer = applications.resnet_v2.preprocess_input(input_layer)\n", "preprocessed_input_layer = layers.RandomZoom(height_factor=0.1)(preprocessed_input_layer)\n", "preprocessed_input_layer = layers.RandomRotation(factor=0.2)(preprocessed_input_layer)\n", "\n", "features_layer = base_model(preprocessed_input_layer, training=False)\n", "\n", "global_pooling = layers.GlobalAveragePooling2D()(features_layer)\n", "\n", "fc1 = layers.Dense(512)(global_pooling)\n", "fc1 = layers.ReLU()(fc1)\n", "\n", "fc2 = layers.Dense(128)(fc1)\n", "fc2 = layers.ReLU()(fc2)\n", "\n", "fc3 = layers.Dense(32)(fc2)\n", "fc3 = layers.ReLU()(fc3)\n", "\n", "output = layers.Dense(2)(fc3)\n", "\n", "second_model = Model(inputs=input_layer, outputs=output)\n", "\n", "second_model.summary()\n", "\n", "early_stopping_callback = tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=2, restore_best_weights=True)\n", "opt = tf.keras.optimizers.Adam(learning_rate=0.01)\n", "loss=tf.keras.losses.CategoricalCrossentropy(from_logits=True)\n", "second_model.compile(optimizer=opt, loss=loss, metrics=['accuracy'])\n", "\n", "second_model.fit(train_set, epochs=10, validation_data=validation_set, callbacks=[early_stopping_callback])" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "23Jt9ScyBZZt", "outputId": "d36bec78-82df-43cd-d865-97f55b6489a0" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "Epoch 1/5\n", "33/33 [==============================] - 48s 1s/step - loss: 0.1525 - accuracy: 0.9381 - val_loss: 0.1518 - val_accuracy: 0.9426\n", "Epoch 2/5\n", "33/33 [==============================] - 40s 1s/step - loss: 0.1080 - accuracy: 0.9613 - val_loss: 0.1097 - val_accuracy: 0.9570\n", "Epoch 3/5\n", "33/33 [==============================] - 40s 1s/step - loss: 0.0915 - accuracy: 0.9630 - val_loss: 0.0917 - val_accuracy: 0.9665\n", "Epoch 4/5\n", "33/33 [==============================] - 39s 1s/step - loss: 0.0880 - accuracy: 0.9651 - val_loss: 0.1188 - val_accuracy: 0.9551\n", "Epoch 5/5\n", "33/33 [==============================] - 40s 1s/step - loss: 0.0789 - accuracy: 0.9701 - val_loss: 0.0791 - val_accuracy: 0.9656\n" ] }, { "output_type": "execute_result", "data": { "text/plain": [ "" ] }, "metadata": {}, "execution_count": 12 } ], "source": [ "base_model.trainable = True\n", "\n", "for layer in base_model.layers[:-10]:\n", " layer.trainable=False\n", "\n", "opt = tf.keras.optimizers.Adam(learning_rate=0.0001)\n", "loss=tf.keras.losses.CategoricalCrossentropy(from_logits=True)\n", "second_model.compile(optimizer=opt, loss=loss, metrics=['accuracy'])\n", "\n", "second_model.fit(train_set, epochs=5, validation_data=validation_set, callbacks=[early_stopping_callback])" ] }, { "cell_type": "markdown", "metadata": { "id": "K5kcobG3WsDc" }, "source": [ "The ResNet v2 architecture was selected since it was stated to be a better alternative to the original ResNet per the research paper.\n", "\n", "We included the same data augmentation layers as the first model. Instead of flattening the layers, we used a global pooling average layer followed by three fully-connected layers and an output layer. We also included an early stopping callback like with the previous model.\n", "\n", "We experimented with the number of layers to unfreeze for fine-tuning. Unfreezing layers beyond the last `10` either had diminishing returns or it worsened the model's performance.\n", "\n", "The model doesn't seem to be overfitting given the added regularization and data augmentation, and the progression of the accuracy and loss values during training." ] }, { "cell_type": "markdown", "metadata": { "id": "XwIzwzEdWsXH" }, "source": [ "## 6. Evaluating on the Test Set" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "011kZGeBu7tA", "outputId": "cd2dc375-4d77-4d07-9f44-d3eb8d86f196" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "5/5 [==============================] - 4s 280ms/step - loss: 0.3667 - accuracy: 0.8365\n", "Test set accuracy of first model: 0.8365384340286255\n", "5/5 [==============================] - 9s 2s/step - loss: 0.4675 - accuracy: 0.8381\n", "Test set accuracy of second model: 0.8381410241127014\n" ] } ], "source": [ "first_model_test_loss, first_model_test_acc = first_model.evaluate(test_set_normalized)\n", "\n", "print(f\"Test set accuracy of first model: {first_model_test_acc}\")\n", "\n", "second_model_test_loss, second_model_test_acc = second_model.evaluate(test_set)\n", "\n", "print(f\"Test set accuracy of second model: {second_model_test_acc}\")\n" ] }, { "cell_type": "markdown", "metadata": { "id": "qlsDgm2dW1jk" }, "source": [ "The second model, using transfer learning, outperforms the first model but not by a *significant* margin. The test set accuracies of both the models are not much lower than that their respective validation accuracies. That can be considered a good sign that the model probably isn't overfitting.\n", "\n", "The accuracy of either of the models is not high enough to be suitable for a medical diagnosis, in our opinion. However, the second model with some more training and experimentation is likely to yield better performance given the lack of additional data." ] } ], "metadata": { "accelerator": "GPU", "colab": { "provenance": [] }, "gpuClass": "standard", "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.10.6" } }, "nbformat": 4, "nbformat_minor": 0 }