notebooks/mnist-inr-classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "91a5ed35",
   "metadata": {
    "id": "91a5ed35",
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# INR Classification with Neural Graphs\n",
    "\n",
    "A quick start notebook for performing INR classification with Neural Graphs.\n",
    "\n",
    "**Make sure you change the runtime type to GPU before starting!**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "17d0a94e",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "17d0a94e",
    "outputId": "f4711831-0ac0-4f0f-a4fc-f3f300389cba",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# if not installed already, please install neural-graphs and all its dependencies.\n",
    "! git clone https://github.com/mkofinas/neural-graphs.git"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "clioeDU4RHME",
   "metadata": {
    "id": "clioeDU4RHME",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "cd neural-graphs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "AqLR4erlS8XK",
   "metadata": {
    "id": "AqLR4erlS8XK",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "!pip install -e ."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1EcAWmJBaeMY",
   "metadata": {
    "id": "1EcAWmJBaeMY",
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Import Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "62a35107",
   "metadata": {
    "id": "62a35107",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import logging\n",
    "from pathlib import Path\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "from omegaconf import OmegaConf\n",
    "import torch\n",
    "import torch.nn.functional as F\n",
    "import torch.nn as nn\n",
    "from torch.utils.data import DataLoader\n",
    "import torch_geometric.utils\n",
    "import torchvision\n",
    "from tqdm import trange\n",
    "import wandb\n",
    "\n",
    "from experiments.data import INRDataset\n",
    "from experiments.utils import count_parameters, set_logger, set_seed\n",
    "from experiments.lr_scheduler import WarmupLRScheduler\n",
    "from nn.relational_transformer import RelationalTransformer\n",
    "\n",
    "from nn.gnn import GNNForClassification\n",
    "from nn.dws.models import MLPModelForClassification\n",
    "from nn.inr import INRPerLayer\n",
    "\n",
    "set_logger()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "tZX7UueTUREC",
   "metadata": {
    "id": "tZX7UueTUREC",
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Get Data\n",
    "Next, we download the MNIST INR dataset and place it in `../experiments/inr-classification/dataset/mnist-inrs`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "QKYprAdjU3bi",
   "metadata": {
    "id": "QKYprAdjU3bi",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "!mkdir -p ../experiments/inr_classification/dataset/mnist-inrs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "QHJ2ZlzibwOo",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "QHJ2ZlzibwOo",
    "outputId": "d502ae9a-ddda-48f4-d167-5df0ffb12e97",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "!wget \"https://www.dropbox.com/sh/56pakaxe58z29mq/AABrctdu2U65jGYr2WQRzmMna/mnist-inrs.zip?dl=0\" -O mnist-inrs.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "vWNSiPEqb9un",
   "metadata": {
    "id": "vWNSiPEqb9un",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "!unzip -q mnist-inrs.zip -d ../experiments/inr_classification/dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "65d3a0fc-d98d-45fa-80c2-6dc3bb6122fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "!rm mnist-inrs.zip"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bcc2b61d",
   "metadata": {
    "id": "bcc2b61d",
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Data preprocessing\n",
    "\n",
    "We have already performed the data preprocessing required for MNIST and provide the files with the repository. The preprocessing includes the data splits and the dataset statistics, corresponding to the files `../experiments/inr_classification/dataset/mnist_splits.json` and `../experiments/inr_classification/dataset/mnist_statistics.pth`, respectively. However, if you want to use different directories for your experiments, you have to run the scripts that follow."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "49c99c4a-c7d8-473c-9871-ba2dd80acf9e",
   "metadata": {
    "id": "bcc2b61d",
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "### Preprocess\n",
    "We create the data split using:\n",
    "```shell\n",
    "python ../experiments/inr_classification/dataset/generate_mnist_data_splits.py \\\n",
    "  --data-path mnist-inrs --save-path . --name mnist_splits.json\n",
    "```\n",
    "This will create a json file `mnist_splits.json` in `../experiments/inr_classification/dataset/`.\n",
    "\n",
    "Next, we compute the dataset (INRs) statistics:\n",
    "```shell\n",
    "python ../experiments/inr_classification/dataset/compute_mnist_statistics.py \\\n",
    "  --splits-path mnist_splits.json --statistics-path mnist_statistics.pth\n",
    "```\n",
    "This will create `mnist_statistics.pth` in `../experiments/inr_classification/dataset/`. Note that this script will take a few minutes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dbd035dc",
   "metadata": {
    "id": "dbd035dc",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# Create dataset splits (train/val/test)\n",
    "%run ../experiments/inr_classification/dataset/generate_mnist_data_splits.py \\\n",
    "  --data-path mnist-inrs --save-path . --name mnist_splits.json\n",
    "\n",
    "# compute dataset statistics - could take few minutes\n",
    "%run ../experiments/inr_classification/dataset/compute_mnist_statistics.py \\\n",
    "  --splits-path mnist_splits.json --statistics-path mnist_statistics.pth"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb3a1c45",
   "metadata": {
    "id": "cb3a1c45",
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## INR Dataset\n",
    "\n",
    "We create INR Datasets and Dataloaders, and visualize some INRs (by reconstruction the images)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "08ac6eef-bdc8-4fc5-a975-fbcf77470e62",
   "metadata": {
    "id": "cb3a1c45",
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "### INR Reconstruction\n",
    "\n",
    "First let's take a look at some INRs and INR augmentations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b6d1d32a-c3ab-4efd-9ce9-2296ad9aaadf",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset_dir = \"../experiments/inr_classification/dataset\"\n",
    "splits_path = \"mnist_splits.json\"\n",
    "statistics_path = \"mnist_statistics.pth\"\n",
    "img_shape = (28, 28)\n",
    "batch_size = 64"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5dd2648a-0ff3-445d-a6f3-069d24f73567",
   "metadata": {},
   "outputs": [],
   "source": [
    "orig_mnist = torchvision.datasets.MNIST(\n",
    "    Path(dataset_dir) / \"mnist\",\n",
    "    train=True,\n",
    "    download=True,\n",
    "    transform=torchvision.transforms.ToTensor(),\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "bb00e965",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 435
    },
    "id": "bb00e965",
    "outputId": "f40d5b17-8af4-4301-d204-79d1ce3ef5ed",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAy0AAAEOCAYAAACEt/i3AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOxdd3hUxfp+t9fsbnrvISEkIUBC6CWhhCZFkCZNQRTFju1ee+9yr14FQRAVBUWKIFKk91BCEkpCOpBeNptNNpvdzZ7fH/nNePbsJgQIqLjv8/hIzp4yM2fOzFfe7/t4DMMwcMIJJ5xwwgknnHDCCSec+IuC/2c3wAknnHDCCSeccMIJJ5xwoj04lRYnnHDCCSeccMIJJ5xw4i8Np9LihBNOOOGEE0444YQTTvyl4VRanHDCCSeccMIJJ5xwwom/NJxKixNOOOGEE0444YQTTjjxl4ZTaXHCCSeccMIJJ5xwwgkn/tJwKi1OOOGEE0444YQTTjjhxF8aTqXFCSeccMIJJ5xwwgknnPhLw6m0OOGEE0444YQTTjjhhBN/aTiVlpvEq6++Ch6Pd0PXfv311+DxeCgqKurcRrFQVFQEHo+Hr7/++pY9wwknnHDi74TbsfY64cSfDec8d6KjuBlZ9nbiH6u0nD9/HrNmzYK/vz8kEgn8/Pxw77334vz583920/4U7N+/HzweDxs2bPizm+KEE7cUZCM/deoUPUYWbG9vbxgMBrtrQkJCMG7cOJtjPB7P5j+VSoUhQ4bg119/veV9+Dvg7bffxubNm//xbXDCMT7//HPweDz06dPnz27KnwqDwYBXX30V+/fv/9PaQNa/6upqemzevHng8Xjo3r07GIaxu4bH42Hx4sX0b2IgJf/x+Xy4ublh9OjROHbs2G3px18Zf4X3/Fdow83iH6m0bNy4Eb169cKePXtw33334fPPP8f8+fOxb98+9OrVC5s2berwvV588UU0NTXdUDtmz56NpqYmBAcH39D1TjjhROeisrISX3zxRYfPHzFiBL799lt88803ePbZZ5GXl4e77roLO3fuvIWt/Hvgr6AwtNUG59r752Pt2rUICQlBWloa8vLy/uzm/GkwGAx47bXX/rKCZFZWFjZu3Njh82fMmIFvv/0Wq1evxqJFi3D8+HEkJycjKyvrFrbyr4+/wnturw03I8veTvzjlJb8/HzMnj0bYWFhyMzMxJtvvon58+fjjTfeQGZmJsLCwjB79mwUFBS0e5/GxkYAgFAohFQqvaG2CAQCSKXSv4VLzgkn/gno0aMHPvjggw4v3pGRkZg1axZmz56NF198Eb///jsYhsF//vOfG3q+0WiE1Wq9oWv/ziDr6e2Cc+39c1FYWIijR4/i448/hqenJ9auXftnN8kJB5DJZIiMjMTrr7/u0NviCL169cKsWbMwd+5cvPXWW/jhhx/Q3Nx8XcYgNm732vBXwe3u983IsrcT/zil5YMPPoDBYMCXX34JT09Pm988PDywfPlyNDY24v3336fHiev0woULmDlzJlxdXTFw4ECb39hoamrCY489Bg8PD7i4uGD8+PEoKSkBj8fDq6++Ss9zxDclNJTDhw8jKSkJUqkUYWFh+Oabb2yeUVtbiyVLliAuLg5KpRIqlQqjR49GRkZGJ43UH327dOkSZs2aBbVaDU9PT7z00ktgGAZXrlzBhAkToFKp4OPjg48++sjmepPJhJdffhkJCQlQq9VQKBQYNGgQ9u3bZ/esmpoazJ49GyqVChqNBnPnzkVGRobDeJzs7GxMmTIFbm5ukEqlSExMxC+//GJzjtlsxmuvvYYuXbpAKpXC3d0dAwcOxO7duzttfJy48/Dyyy+joqLihjfY6OhoeHh4ID8//5rnEkrmunXr8OKLL8Lf3x9yuRz19fUAgBMnTmDUqFFQq9WQy+UYMmQIjhw5YnefkpISzJ8/H35+fpBIJAgNDcWiRYtgMpnoOQUFBbjnnnvg5uYGuVyOvn372tHYSHt+/PFHvPXWWwgICIBUKsWwYcPsLOG5ubmYPHkyfHx8IJVKERAQgOnTp0On0wFopY40NjZizZo1lC4yb948AO2vp0OHDsXQoUPt+jhv3jyEhITYHLNarfjPf/6DuLg4SKVSeHp6YtSoUZT2114b2uL6f/7554iJiaGU4UceeQR1dXU25wwdOhSxsbG4cOECkpOTIZfL4e/vb7NnEHz66aeIiYmBXC6Hq6srEhMT8f3339ud90/D2rVr4erqirFjx2LKlCkOlRYyH7lW4bbiNH/66Sd069YNUqkUsbGx2LRpk928Idd++OGH+N///oewsDDI5XKMHDkSV65cAcMweOONNxAQEACZTIYJEyagtrbWrm2//fYbBg0aBIVCARcXF4wdO9aOWj5v3jwolUqUlJRg4sSJUCqV8PT0xJIlS9DS0kLbQ+SQ1157jc5TtpzQkf0OaKW8p6SkQCaTISAgAG+++eZNG0D4fD5efPFFZGZmXhcDhY1BgwYBQIfWRPJdHjhwAA8//DC8vLwQEBBAf+/IuAOtYzZ16lR4enpCJpMhKioK//73v23OSU9Px+jRo6FSqaBUKjFs2DAcP37cYXuOHDmCp556Cp6enlAoFJg0aRKqqqpszj116hRSU1Ph4eEBmUyG0NBQ3H///QCu/Z7JXMnPz8eYMWPg4uKCe++9F0CrTEjWLTYcrZVGoxGvvvoqIiMjIZVK4evri7vvvhv5+fnXbIMjWdZiseCNN95AeHg4JBIJQkJC8K9//QvNzc0253VUbu0MuUzY4TPvEGzduhUhISH0Q+Ji8ODBCAkJcchLv+eee9ClSxe8/fbb7Vod5s2bhx9//BGzZ89G3759ceDAAYwdO7bDbczLy8OUKVMwf/58zJ07F6tWrcK8efOQkJCAmJgYAK1CyObNm3HPPfcgNDQUFRUVWL58OYYMGYILFy7Az8+vw8+7FqZNm4bo6Gi8++67+PXXX/Hmm2/Czc0Ny5cvR0pKCt577z2sXbsWS5YsQe/evTF48GAAQH19PVauXIkZM2bggQcegF6vx1dffYXU1FSkpaWhR48eAFqFj7vuugtpaWlYtGgRunbtii1btmDu3Ll2bTl//jwGDBgAf39/PP/881AoFPjxxx8xceJE/Pzzz5g0aRKA1g/wnXfewYIFC5CUlIT6+nqcOnUKZ86cwYgRIzptbJy4szBo0CCkpKTg/fffx6JFiyCTya7rep1OB61Wi/Dw8A5f88Ybb0AsFmPJkiVobm6GWCzG3r17MXr0aCQkJOCVV14Bn8/H6tWrkZKSgkOHDiEpKQkAUFpaiqSkJNTV1WHhwoXo2rUrSkpKsGHDBhgMBojFYlRUVKB///4wGAx47LHH4O7ujjVr1mD8+PHYsGED/WYI3n33XfD5fCxZsgQ6nQ7vv/8+7r33Xpw4cQJAqzEiNTUVzc3NePTRR+Hj44OSkhJs27YNdXV1UKvV+Pbbb+m3t3DhQgCwG5OOrqdtYf78+fj6668xevRoLFiwABaLBYcOHcLx48eRmJjYoTaw8eqrr+K1117D8OHDsWjRIuTk5OCLL77AyZMnceTIEYhEInquVqvFqFGjcPfdd2Pq1KnYsGEDnnvuOcTFxWH06NEAgBUrVuCxxx7DlClT8Pjjj8NoNCIzMxMnTpzAzJkzr7u/dxLWrl2Lu+++G2KxGDNmzKDj3Lt37xu636+//opp06YhLi4O77zzDrRaLebPnw9/f/82n28ymfDoo4+itrYW77//PqZOnYqUlBTs378fzz33HPLy8vDpp59iyZIlWLVqFb3222+/xdy5c5Gamor33nsPBoMBX3zxBQYOHIj09HQbJamlpQWpqano06cPPvzwQ/z+++/46KOPEB4ejkWLFsHT0xNffPEFFi1ahEmTJuHuu+8GAHTv3h1Ax/e78vJyJCcnw2Kx0PO+/PLL616/HGHmzJl444038Prrr2PSpEnX7Z0khgFXV9cOX/Pwww/D09MTL7/8MvU4dHTcMzMzMWjQIIhEIixcuBAhISHIz8/H1q1b8dZbbwFoHddBgwZBpVLh2WefhUgkwvLlyzF06FAcOHDALs7q0UcfhaurK1555RUUFRVh6dKlWLx4MdavXw+glVY8cuRIeHp64vnnn4dGo0FRURGl1V3rPQOtCkJqaioGDhyIDz/8EHK5/LrGuaWlBePGjcOePXswffp0PP7449Dr9di9ezfOnTuH4cOHX7MNXCxYsABr1qzBlClT8PTTT+PEiRN45513cPHiRTsltiNya6fIZcw/CHV1dQwAZsKECe2eN378eAYAU19fzzAMw7zyyisMAGbGjBl255LfCE6fPs0AYJ544gmb8+bNm8cAYF555RV6bPXq1QwAprCwkB4LDg5mADAHDx6kxyorKxmJRMI8/fTT9JjRaGRaWlpsnlFYWMhIJBLm9ddftzkGgFm9enW7fd63bx8DgPnpp5/s+rZw4UJ6zGKxMAEBAQyPx2Peffddelyr1TIymYyZO3euzbnNzc02z9FqtYy3tzdz//3302M///wzA4BZunQpPdbS0sKkpKTYtX3YsGFMXFwcYzQa6TGr1cr079+f6dKlCz0WHx/PjB07tt0+O/HPBPnuTp48SY+RuV5VVcUcOHCAAcB8/PHH9Pfg4GC7+QSAmT9/PlNVVcVUVlYyp06dYkaNGsUAYD744INrtoN8c2FhYYzBYKDHrVYr06VLFyY1NZWxWq30uMFgYEJDQ5kRI0bQY3PmzGH4fL5NX9j3YRiGeeKJJxgAzKFDh+hver2eCQ0NZUJCQug6QtoTHR1t893+5z//YQAwWVlZDMMwTHp6ut1a4QgKhcJmPSBobz0dMmQIM2TIELvjc+fOZYKDg+nfe/fuZQAwjz32WJv9bq8N3LW3srKSEYvFzMiRI23W1c8++4wBwKxatcqmjQCYb775hh5rbm5mfHx8mMmTJ9NjEyZMYGJiYuye/U/HqVOnGADM7t27GYZpfV8BAQHM448/bnMemY/79u2zOe5oT4uLi2MCAgIYvV5Pj+3fv58BYDNvyLWenp5MXV0dPf7CCy8wAJj4+HjGbDbT4zNmzGDEYjHdb/R6PaPRaJgHHnjApk3l5eWMWq22OT537lwGgM1+zDAM07NnTyYhIYH+XVVVZScbEHR0vyPf+IkTJ+ixyspKRq1W28kYjsBe/9jtVygUDMMwzJo1axgAzMaNG+nvAJhHHnmE/k3G9rXXXmOqqqqY8vJy5tChQ0zv3r07tF4wzB/f5cCBAxmLxUKPX8+4Dx48mHFxcWGKi4ttzmWvCxMnTmTEYjGTn59Pj5WWljIuLi7M4MGD7dozfPhwm+uffPJJRiAQ0Dm0adMmuz2Fi/beM5krzz//vN1vwcHBDtcw7lq5atUqu32L2/f22sCVZc+ePcsAYBYsWGBz3pIlSxgAzN69e23a2BG5tTPksn8UPUyv1wMAXFxc2j2P/E5oGgQPPfTQNZ+xY8cOAK2WAjYeffTRDrezW7duNp4gT09PREVF2cTZSCQS8Pmtr6+lpQU1NTVQKpWIiorCmTNnOvysjmDBggX03wKBAImJiWAYBvPnz6fHNRqNXRsFAgHEYjGAVm9KbW0tLBYLEhMTbdq4Y8cOiEQiPPDAA/QYn8/HI488YtOO2tpa7N27F1OnToVer0d1dTWqq6tRU1OD1NRU5ObmoqSkhLbn/PnzyM3N7dSxcOLOx+DBg5GcnIz333//mrEtX331FTw9PeHl5YXExETs2bMHzz77LJ566qkOP2/u3Lk2FtGzZ88iNzcXM2fORE1NDZ3njY2NGDZsGA4ePAir1Qqr1YrNmzfjrrvuQmJiot19iUV0+/btSEpKohQsAFAqlVi4cCGKiopw4cIFm+vuu+8++t0Cf9A7yLetVqsBADt37nSYaa2j6Mh62hZ+/vln8Hg8vPLKK3a/3Uicyu+//w6TyYQnnniCrqsA8MADD0ClUtl53pVKJWbNmkX/FovFSEpKsln/NBoNrl69ipMnT153e+5krF27Ft7e3khOTgbQ+r6mTZuGdevWUdrU9aC0tBRZWVmYM2cOlEolPT5kyBDExcU5vOaee+6h8xgAtazPmjULQqHQ5rjJZKL7yu7du1FXV4cZM2bQ77K6uhoCgQB9+vRxSH3mzvNBgwZdM2YWuL79bvv27ejbty/1wAKtcgOhGN0s7r33XnTp0qVDsS2vvPIKPD094ePjg0GDBuHixYv46KOPMGXKlA4/74EHHoBAIKB/d3Tcq6qqcPDgQdx///0ICgqyuSdZF1paWrBr1y5MnDgRYWFh9HdfX1/MnDkThw8ftpP9Fi5caLOuDBo0CC0tLSguLgbQ+q0DwLZt22A2mzvcTy4WLVp0w9f+/PPP8PDwcChr3siauH37dgCw28uefvppALBbEzsit3aGXPaPUlqIMkKUl7bQlnITGhp6zWcUFxeDz+fbnRsREdHhdnI/NqDVtarVaunfVqsVn3zyCbp06QKJRAIPDw94enoiMzOT8so7C9z2qNVqSKVSeHh42B1ntxEA1qxZg+7du1P+oqenJ3799VebNhYXF8PX19fOHcods7y8PDAMg5deegmenp42/xHhpbKyEgDw+uuvo66uDpGRkYiLi8MzzzyDzMzMmxsIJ/4xePXVV1FeXo5ly5a1e96ECROwe/du/Prrr5QTbDAYbATfa4G7VpAFfe7cuXbzfOXKlWhuboZOp0NVVRXq6+sRGxvb7v2Li4sRFRVldzw6Opr+zgb3eye0DvJth4aG4qmnnsLKlSvh4eGB1NRU/O9//7vudacj62lbyM/Ph5+fH9zc3G74HmyQMeCOk1gsRlhYmN0YBQQE2AkC3DX6ueeeg1KpRFJSErp06YJHHnnEYUzSPwktLS1Yt24dkpOTUVhYiLy8POTl5aFPnz6oqKjAnj17rvue5N042mPb2ncd7WkAEBgY6PA4ea/k20xJSbH7Nnft2kX3HwISa8UGd560hevZ74qLi9GlSxe7ezj67m8EAoEAL774Is6ePXvNjIALFy7E7t27sXXrVjz55JNoamq6bmW0rTXxWuNOBOT21sSqqioYDIY210Sr1YorV67YHL/WmjhkyBBMnjwZr732Gjw8PDBhwgSsXr3aLvajPQiFQpv4netFfn4+oqKibJTumwGRZbnfkI+PDzQazTX3DcB+rneGXPaPimlRq9Xw9fW95iBlZmbC398fKpXK5nhn8EM7AraFgQ22hePtt9/GSy+9hPvvvx9vvPEG3NzcwOfz8cQTT3R69iFH7elIG7/77jvMmzcPEydOxDPPPAMvLy8IBAK88847HQrK44L0a8mSJUhNTXV4DvnABg8ejPz8fGzZsgW7du3CypUr8cknn2DZsmU2niMnnHCEwYMHY+jQoXj//ffb9QgEBARg+PDhAIAxY8bAw8MDixcvRnJyMuUMXwvcdYXM8w8++IDGfXGhVCodBgh3BjrybX/00UeYN28e/b4ee+wxvPPOOzh+/HiHN15H6ymPx3Noyb0RC/ytREfGKDo6Gjk5Odi2bRt27NiBn3/+GZ9//jlefvllvPbaa7erqX8p7N27F2VlZVi3bh3WrVtn9/vatWsxcuRIAG1bhztjLrT1/q71Xsm3+e2338LHx8fuPK7A2Nb9OoLr2e9uB+69914a2zJx4sQ2z+vSpQtdE8eNGweBQIDnn38eycnJDj3CjtDWmtjRce9sXGtekBp3x48fx9atW7Fz507cf//9+Oijj3D8+HEbD2BbYLNn2GjvO7iZ+dVRdNRL05E1sTPksn+U0gK0fkQrVqzA4cOHbegSBIcOHUJRUREefPDBG7p/cHAwrFYrCgsLbSwfnZ2HfsOGDUhOTsZXX31lc7yurs7OA/JnYcOGDQgLC8PGjRttJj6X0hEcHIx9+/bBYDDYeFu4Y0ZcuSKRiC6K7cHNzQ333Xcf7rvvPjQ0NGDw4MF49dVXnUqLEx3Cq6++iqFDh2L58uUdvubBBx/EJ598ghdffPGGglaBP4LFVSpVu/Pc09MTKpUK586da/d+wcHByMnJsTuenZ1Nf78RxMXFIS4uDi+++CKOHj2KAQMGYNmyZXjzzTcB3BglwdXV1SF1hmvVCw8Px86dO1FbW9uut6WjbSBjkJOTY0MZMZlMKCws7NB64wgKhQLTpk3DtGnTYDKZcPfdd+Ott97CCy+88LdIL9rZWLt2Lby8vPC///3P7reNGzdi06ZNWLZsGWQyGbVmc7O3cecCeXeO9tjO3nfJt+nl5XXDc4KLtubo9ex3wcHBDik3jr77GwXxthBjRUfx73//GytWrMCLL75I6fPXi46OOxmz9tZET09PyOXyNtdEPp9v53HrKPr27Yu+ffvirbfewvfff497770X69atw4IFC244vbqrq6vdNwC0fgfstSo8PBwnTpyA2Wy2SRrCxvW0gciyubm51CsPABUVFairq7vhfeNm5bJ/FD0MAJ555hnIZDI8+OCDqKmpsfmttrYWDz30EORyOZ555pkbuj+xiHz++ec2xz/99NMba3AbEAgEdhbJn376iXJc/wogmje7nSdOnLCrjpuamgqz2YwVK1bQY1ar1W5j8/LyokJkWVmZ3fPYKQi571apVCIiIuK63LVO/LMxZMgQDB06FO+99x6MRmOHrhEKhXj66adx8eLF69rY2UhISEB4eDg+/PBDNDQ02P1O5jmfz8fEiROxdetWmuaXDfLdjRkzBmlpaTbfXWNjI7788kuEhISgW7du19W++vp6WCwWm2NxcXHg8/k235dCoXC42baH8PBwZGdn23zLGRkZdrSqyZMng2EYhx4L9nrT0TYMHz4cYrEY//3vf22u/+qrr6DT6a4r+yMBdw0Si8Xo1q0bGIa5Kd773xVNTU3YuHEjxo0bhylTptj9t3jxYuj1eprONzg4GAKBAAcPHrS5D3dv9fPzQ2xsLL755hub7+XAgQOdXtAwNTUVKpUKb7/9tsN3yE2D2xEQQx13nl7PfjdmzBgcP34caWlpNr93dv2bWbNmISIi4ro8hRqNBg8++CB27tyJs2fP3tBzOzrunp6eGDx4MFatWoXLly/bnEO+a4FAgJEjR2LLli02Kc8rKirw/fffY+DAgXYsm2tBq9XayWPES07WxLbe87UQHh6O48eP26Sw37Ztmx2FbfLkyaiursZnn31mdw/Stutpw5gxYwAAS5cutTn+8ccfA0CnrIk3Ipf94zwtXbp0wZo1a3DvvfciLi4O8+fPR2hoKIqKivDVV1+huroaP/zww3WlLGUjISEBkydPxtKlS1FTU0NTHl+6dAnAjVkfHWHcuHF4/fXXcd9996F///7IysrC2rVrbTTvPxvjxo3Dxo0bMWnSJIwdOxaFhYVYtmwZunXrZrO5TJw4EUlJSXj66aeRl5eHrl274pdffqH0F/aY/e9//8PAgQMRFxeHBx54AGFhYaioqMCxY8dw9epVWqemW7duGDp0KBISEuDm5oZTp05hw4YNWLx48e0dBCf+1njllVdowHBHMW/ePLz88st477332qVRtAU+n4+VK1di9OjRiImJwX333Qd/f3+UlJRg3759UKlU2Lp1K4BWmuiuXbswZMgQLFy4ENHR0SgrK8NPP/2Ew4cPQ6PR4Pnnn8cPP/yA0aNH47HHHoObmxvWrFmDwsJC/Pzzz9cVfwO0UnwWL16Me+65B5GRkbBYLPj2228hEAgwefJkel5CQgJ+//13fPzxx/Dz80NoaKhdKlEu7r//fnz88cdITU3F/PnzUVlZiWXLliEmJsYmODY5ORmzZ8/Gf//7X+Tm5mLUqFGwWq04dOgQkpOT6Xfe0TZ4enrihRdewGuvvYZRo0Zh/PjxyMnJweeff47evXvbBN13FCNHjoSPjw8GDBgAb29vXLx4EZ999hnGjh17zWQwdyJ++eUX6PV6jB8/3uHvffv2pYUmp02bBrVajXvuuQeffvopeDwewsPDsW3bNru4EaD1O5gwYQIGDBiA++67D1qtFp999hliY2MdKv43CpVKhS+++AKzZ89Gr169MH36dHh6euLy5cv49ddfMWDAAIdCY3uQyWTo1q0b1q9fj8jISLi5uSE2NhaxsbEd3u+effZZfPvttxg1ahQef/xxmvI4ODi4U2M5BQIB/v3vf+O+++67rusef/xxLF26FO+++65DWuC1cD3j/t///hcDBw5Er169sHDhQirf/frrr1RpevPNN7F7924MHDgQDz/8MIRCIZYvX47m5maH9ZauhTVr1uDzzz/HpEmTEB4eDr1ejxUrVkClUlHhv7333B4WLFiADRs2YNSoUZg6dSry8/Px3Xff2cmoc+bMwTfffIOnnnoKaWlpGDRoEBobG/H777/j4YcfxoQJE66rDfHx8Zg7dy6+/PJL1NXVYciQIUhLS8OaNWswceLE694XgU6Sy24q99jfGJmZmcyMGTMYX19fRiQSMT4+PsyMGTNoWk82HKUD5P7GRmNjI/PII48wbm5ujFKpZCZOnMjk5OQwAGzSBLeV8thRSjhuejuj0cg8/fTTjK+vLyOTyZgBAwYwx44dszuvM1Iec/vNToXIbSM7xafVamXefvttJjg4mJFIJEzPnj2Zbdu22aUvZZjWVHwzZ85kXFxcGLVazcybN485cuQIA4BZt26dzbn5+fnMnDlzGB8fH0YkEjH+/v7MuHHjmA0bNtBz3nzzTSYpKYnRaDSMTCZjunbtyrz11luMyWRqdxycuPNxrZTHXJAUt45SHrNTfrLx6quvOkzXyoajb46N9PR05u6772bc3d0ZiUTCBAcHM1OnTmX27Nljc15xcTEzZ84cxtPTk5FIJExYWBjzyCOP2KQtzs/PZ6ZMmcJoNBpGKpUySUlJzLZt2zrUHu4aUlBQwNx///1MeHg4I5VKGTc3NyY5OZn5/fffba7Lzs5mBg8ezMhkMgYATdvZ3lgzDMN89913TFhYGCMWi5kePXowO3fudLhmWCwW5oMPPmC6du3KiMVixtPTkxk9ejRz+vTpa7bB0drLMK0pjrt27cqIRCLG29ubWbRoEaPVam3O4a5zBNw2Ll++nBk8eDB9f+Hh4cwzzzzD6HQ6h/2+03HXXXcxUqmUaWxsbPOcefPmMSKRiKmurmYYpnVfmDx5MiOXyxlXV1fmwQcfZM6dO+dwT1u3bh3TtWtXRiKRMLGxscwvv/zCTJ48menatSs9h8xlbkrytua+o7WCnJ+amsqo1WpGKpUy4eHhzLx585hTp07Rc9raJx3JDEePHmUSEhIYsVhsl5K2I/sdw7TKNEOGDGGkUinj7+/PvPHGG8xXX33VKSmP2TCbzUx4eHibKY/bSvc+b948RiAQMHl5eW22o63xJujIuDMMw5w7d46ZNGkSXe+ioqKYl156yeacM2fOMKmpqYxSqWTkcjmTnJzMHD16tEPt4abjPnPmDDNjxgwmKCiIkUgkjJeXFzNu3Di7drX1ntsaa4KPPvqI8ff3ZyQSCTNgwADm1KlTDtPDGwwG5t///jcTGhpK5dopU6bYpHZuqw2O5qXZbGZee+01er/AwEDmhRdesEnBzTAdl1s7Qy7jMcwNVPVy4rpx9uxZ9OzZE999912npSG807F582ZMmjQJhw8fxoABA/7s5jjhhBNOOPE3Qo8ePeDp6XldFbedcMKJvy7+cTEttwOOajssXboUfD6fVot3whbcMWtpacGnn34KlUqFXr16/UmtcsIJJ5xw4q8Os9lsF2e1f/9+ZGRkYOjQoX9Oo5xwwolOxz8upuV24P3338fp06eRnJwMoVCI3377Db/99hsWLlx4w1kp7nQ8+uijaGpqQr9+/dDc3IyNGzfi6NGjePvtt29bqmknnHDCCSf+figpKcHw4cMxa9Ys+Pn5ITs7G8uWLYOPj89NFTF1wgkn/lpw0sNuAXbv3o3XXnsNFy5cQENDA4KCgjB79mz8+9//vuX5xP+u+P777/HRRx8hLy8PRqMRERERWLRokTNw3gknnHDCiXah0+mwcOFCHDlyBFVVVVAoFBg2bBjefffdG06q44QTTvz14FRanHDCCSeccMIJJ5xwwom/NJwxLU444YQTTjjhhBNOOOHEXxpOpcUJJ5xwwgknnHDCCSec+EvDqbQ44YQTTjjhhBNOOOGEE39pdDgqvLMquTvhhBM3j79rKJpzHXHCib8O/o7riHMNccKJvw5u9xriTGXlhBNOOOGEE07cMeAqNh0RrHg8Hvh8PgQCAb2eYRiYzWYwDAM+n0/vQ/62Wq1t3k8oFIJhGDAMAx6Ph5aWljafKxAI6H0ZhoHVaoVYLAafz0dLSwutQSMSiWC1Wu1q0vB4PJs2O+oz6R9pC3luS0sLrFYreDweRCIRWlpa2mxrexCJRODzW8k75J6OxofH49G2cbOpkjYSWK1Wei/ueY76eC3I5XI6BgRkzLl/k2e3tLS0+xwy9uy+k/PZ74Xb9o4o39x3yW2rI5B5ym4DaRu3Xez5KRAIIBAI7M4l9+S2V6/XX7P9twJOpcUJJ5y47SALaXsbOXtRdQQiXJDNkbuRsTdH9r/ZIAs02RTJcx1tkuz2iEQiWCwWMAxDBQ7SBqvVSgWWlpYWCIVC8Pl8mEwmALC5lvSDCBHcPguFQvB4PJjNZtreawlAZBNi94PdHnKeQCCwEX6448e+pr3ntiW88fl88Pl8WCwWu/En17DnAff53M0XAMRiMR1Hbh8dnc8dFwIiLHGFP0fntjV32vqtLYGKK9gQiEQi+n7JeUKhkApNpL3kXZB576jt5Dwytuzx4M71Ow0SiQRisRgMw0Aul8PFxQUMw0Cn06G2trbddygSiaBWq9G1a1ckJSXBz88PjY2NaGhowOrVq1FVVYVu3brBaDRCr9ejuroawcHBqKqqosIbEfjMZjM0Gg1Gjx4NnU6HgoICeHt74+TJkzAYDDbP5vP5CAkJQVJSEqxWK3Q6HfLy8pCfn49HHnkEEREROHXqFA4cOIDGxkbce++9yM3NxdatW+k9BAIBvLy8EBgYCJPJhKqqKphMJphMJhgMBpjNZggEAgQHB8PLywuurq64dOkSXF1dERcXh7NnzyI9PR0hISG4++67cejQIaSlpV3X2Lu5uWHy5Mno2rUrdDodsrOzcfHiReTn59v0mcfjwdfXF6WlpeDxeBg7diytwyYUCuHq6gpPT09IpVI0Njbi8uXLOHv2LC5cuIDm5mb6bfj7+4PH46GiosJuTLlQqVRoamqC2WzGpk2bkJiYCFdXVwCAyWRCUVERXd90Oh1qampQXV2NgoICnD59GqdOnUJZWZnDtU+lUkEqlcLd3R1dunQBAJw8eRJlZWUAgKCgIHh7e8NsNtN9QSKRQCaT0fnKVRTI92uxWGAymdDU1ISGhgbU19dDr9fDYDDAZDJRxVkkEkEgEKCurg4SiQSBgYGoqamBVquFSCSCr68vAgICIJFIoNfr0dDQADc3NwQEBKChoQFVVVVwc3NDQkICEhMTERYWBplMBr1eD51OB6lUisDAQHh4eFBFmsfjQalUorGx8brmSWfAqbQ44YQTtx1kYQYcC3ltCRhkwWQLdOxruBY07m9tWSq592E/TyAQQCgU0o1CKBTaWNOsVqudRYx9f+6z2OdzFQGhUAihUAiz2UwFWbaQyR4zIvg7supxFS+utZAoEwTs68i/2b87EnTZVlrufcjmS+7BFqYFAgFV9ogyx+PxIJFI6HtlC95sEIXFUR/bE8a5CjJRpNi/t6XwcBUY9rG2zmcroOzjjgQf8p6JQEYEHPKOyXXkXux7cBUcRxZ48uw7VVkBWhUWNzc3qNVqqNVqiEQiNDY2Qq/Xt2vR5vP5kMvlEIlEMBqNuHDhAgwGA2JjYxEYGAgXFxcqUA4cOBBCoRC5ubnYvXs3CgoKAPwxt8naJBaLERMTg+joaBQUFECv17er2KvVaqqwVFZWUiWopKQESqUSdXV1qKurg9FopEIkgUAgoII+WUusVis1GJBntrS0oLGxEfX19ZDJZFCpVPD29oZMJoO7uzsAoLGxEefPn0dZWRlEIhFV7IlRiG1oYUMoFMLHxwdNTU3Iy8tDbW0tLl++TJUnNtRqNf03j8dDYGAg5HI5TCYTFeq1Wi1V9jIyMqDVagG0Kveurq6IjIyk49MRT0tzczP9ZsrLy2EwGODq6gqj0YgjR46gvLwcYWFhUCqVqKmpwZUrV1BaWorLly9TZbetOcQwDEwmExoaGlBXVwcAMBqN9HeiOJL+sq/jftfscTabzWhubobRaERTUxOamppgNBphsVjA5/MhkUjo3kTWU9LXvLw8+hyz2YzGxkYwDAOZTAaLxYKCggIUFxdDIpEgJiYGYrEYer0e5eXlqKqqQkREBHx8fKBSqWAwGJCXl4ecnBx4eXkhPDwc/v7+NnPwduMfo7R06dIFbm5uuHr1KkpKSv7s5jjhhBP/D0fWZ8CxQEiE3ras3+1tYkSIbkvQ5VqiicBIFAi5XI6goCD4+vqirKwMeXl5VPjlCtBtWeCJoA+0ChxsJYQ8iwieXGs6WwhxJARdizpAnsGmuZDj5Le2lBPymyOvAHk2t99sTxJbySTPIJ4ech57syf3Il4pbru4QpkjTw57PLiCB2kTUULZY309lBMydo7oSNd6F9zz2f/nfgtchZRNoeFSPNh9IcfY4899h393EIVXrVYjMDAQGo2GWqOlUinkcnmbnhYyx2QyGUwmE3Q6HU6fPo2ioiLExcUhMTGRnpuSkoKGhgYUFRXZeDCJh0MqlVLFJSkpCaNHj0ZzczOys7Nx+vRp5Ofn4+rVqw6/serqahQXF+PKlStU0C8qKgIAlJaWorGxEc3NzdBqtdRCz+PxqMXcxcUFPB4Pzc3NNusj6TP5jhoaGiCTySASiaBQKAAA/v7+AIDKykrs2LEDIpEIISEhCA0NBQAqkFdXV0Ov19usUUKhEGq1Gq6urmhqakJzczOqqqpQUlKCmpoaOyWaCOIERqMRGo0GSqUSJpMJ9fX1KCsrQ05ODvLy8mAwGMDj8aBQKBAWFobQ0FAqSF+5csXu/sQQwlYI2IpBTU0NKioqoFarcfz4cWzYsIEqfXw+H3q9nva1rKyM9sGR0iIQCGAymSAQCOi7sVgsNp4fs9lMjTLE2NXS0kK/P6KEsI08JpOJKisGgwFNTU0wmUx237RQKGzX60reT2hoKPr37w8vLy9UV1ejrq4O58+fx759+2j7ampqYDAYYDQaUVhYCE9PT4jFYtoGs9mMwsJCXLp0CeHh4ejWrdufZgS545UWlUqFwMBAfP755+jXrx9WrlyJp556ym6DdMKJPwN8Ph/e3t4IDAwEwzC4ePEiGhoa/uxm3XZwBU5yrC1hmL3Qt0fp4goqXBoYuR+b2gH84fEg1mu1Wo0BAwbgvvvuQ2RkJPbt24cPPvgAV69epW1mU3LItew2kWewNx9CNRMKhWhubqbWM9Iuwqdne1ZIX0m72ZSv9jwA3DEg13BpduR55P7Ey8S9llCVuMoBVyEi9DKusEzuzePxYDKZIBKJqMWbKDNsgYOrpHAVTnbbHXlEyLPYFAzSp7aUYEfeEjbYc7A9+hH7344UHfKeiYDE9aKxBVFyHZsqRjwNAQEB4PF4KCwsREVFBZ0b7LbdSLzCXxlkbhGqjclkgl6vh9FohEAggIuLC2QymUMqC3sdEQqFdI7W1NRg//790Ov19DqpVIpjx47h1KlTNvfg8XiQSqVUcCVeE4lEAj8/P2i1WvB4PKhUKptvCWide9XV1dDpdCgpKbH5raysDGazGVVVVXTN0Gq18PX1RXh4OCQSCTw8PODi4kJpPuT57FgRsVgMpVIJiUSC5uZm1NbWQqVSoaWlBU1NTQgICLDpD7Hyy+Vy6oUSCoXU8l5fXw+dToeWlha4uLjAz88PHh4eUCgUaG5uRn19PWprax0qxk1NTfTfDMPQvU4sFqOlpQVGoxHV1dWorq4GALi6usLd3R2xsbHw8/NDTU0N8vPzUVFRYadIiMVi+Pv7w8PDAwaDAaWlpdDpdDYCfUtLCyorK1FdXY2ffvoJJ0+eRJ8+feh5xCvBHoe2hHPyjUokEggEAtTX11Plkv1+iWJCvmu2B50YX8h4sL0s3P9IP8g1RGkhbXUEV1dXxMbGYsCAAYiMjASfz0e3bt3w8ccf49y5czhx4gQCAgLo2l9TU4OysjJotVqYTCZoNBr4+PhAKpXCarVCJpOhqamJ0jD/DNzRSotGo8HcuXNx1113YcCAAWAYBsnJyYiKikJGRsaf3bw7AiqVComJibBYLDhx4oTNB+uEY3h4eKBXr15QqVSwWCwYMWIEBg8eDKvVitdffx0///zzn93E2wJH9Cq21Z8tsAkEAirYkfPZwuy1+Prs+7Kt9Gyhjh0/wOfzqRLh5+eHYcOG4d5770VycjIEAgF0Oh2Cg4Op0tJenwjYgalcAZyteACwoQhxlTL2v9nxIOz+sRUbdru4wrVIJLLpO1dBaovqRMBVmLgCPPk/sQyzN1fiuSE874SEBEgkEpw8eZJSQrhUKPazHCkt5HdH8VLkGu5xtveOO76OxswRHHmuuFQ5rpLNnt9ckHfIFriIUAO00qHCw8Ph5+cHFxcXeHl5oWfPnoiMjERjYyN++uknbNu2DXV1dW0q7ncKyLvWarV0fjU0NMBiscDb2xuhoaHw9PTEpUuXUFpaand9S0sLjZfgvpvMzEw65qtWrcKxY8dQUVFBrxUKhZS6Q8ZaJpPRGIPs7Gxs3rwZ586ds2kfgdVqRWlpKUQikU0brFYrtFotmpubodfr6Zytra2Fu7s7oqOj4eXlBa1Wi7y8PBQUFMBgMEClUkEikVDLPlE81Go1xGIxdDodpTGR9nbv3t1uTMrLy3Hy5EnqgeLz+XBzc4OHhwe8vb1hMBjQ2NgIHq81toGsI4TCdq04EwJCCSWKt1KphIeHB/1OAgMD0bVrV7i7u6Oqqgp1dXVQKpUICgqisUXNzc0Qi8UICQlBv3794O7ujoKCAjQ0NECv19sZjsrLy5Gbm4vjx49TL0JJSQnkcrlNnCLZJ8xmM1paWigliyhehFpIYqfI87hzi8QUAX+sYVar1W4tJAYy4v0gHhej0Ujpc1y0te/x+XxIpVK4uLjAYDCgsLAQarUaUVFRmDRpEiQSCd58802UlpZCIpFAKBTCxcUFSqUSOp0OjY2NqKqqop4uo9GIqqoqyGQyGI1GKJVKp6els5GYmIhevXrhueeeg6enJwCgqqoKP//8c6dZmlxcXBAWFobg4GAEBQVBIBAgNzcXp06dQnV19S15qTweD4mJiejduzflFWq1Whw7dgyFhYW3VWkQCASIi4vD22+/jf3799tZoJywhUKhQFJSEkaNGoVJkybB09MT9fX1CAgIgMViQWlpqZ3V658AriWb/X0KBAIa/MeORSGbgMVisRP423oG1w3PFuTZnguyUbm4uCAmJgbJyckYPXo0+vbtCwDUtd7Q0GAj4LD/TdrE3lS4QjaxqstkMoSFhSEkJAS+vr7g8/koKirC+fPnUVJSQhUYAnY/uZZ0LnWI/Ty2IiKXyxETE4MePXrAw8MDZrMZlZWVSE9PR25urg0X3mw2OxxbruWf2zb2eJD7aTQa8Hg8SqUgiQoGDx6Mxx9/HAcPHsTRo0epJY89F9hj0JaSwPaicOeVxWKBWCyGRqOBSCRCfX09Fa6Ihb2tOKm2FBdHWXXYfWcrfkRx4wogjvYJrseKeJzEYjE8PT0RExODcePGISEhgcZxkIDf8+fPQ6FQUGs7d2w6ooT9ncAwDBobG2EwGFBZWQmgdbzEYjGCg4MxaNAgxMTEoLi4GOvWrcPJkycp08JisVChkHwjEomEviO2YLlt2zYbWpSbmxs8PT0hkUhw9epVOqaBgYGIiIhAQ0MD0tLScOjQIRpH4QjsbF3Eak/6YzKZbNpQXV0Nd3d3BAYGAgBOnz6NwsJChIeHY8yYMfD390dBQQGOHj2K5uZmeHt7Q6FQQC6X02+KyCYuLi4AQGlibJhMJhsavVQqpYoFycBFDDs1NTVUgdHpdKiqqurQ/CIeKnJfmUxGFSzi3SZKeW1tLQwGAxiGgYeHB9zc3NDU1ASNRoOGhgZ4enoiLi4Orq6uyMvLQ3p6OrRaLY0Xqq+vp56l6upqpKWloaSkBG5ubqivr0dhYSFUKhWN+zAYDNRDTLxfPXv2hEqlorQquVwOV1dXu/fFHUeDwUC9yGzvNFtJIWsPUV7IfnGt+EtH7AKlUgkvLy9K8bp69SrKy8tx6tQp9OzZE3379kVoaChGjhyJvXv3QiwWo6GhATU1NdQbTZTfwMBABAQE4PLly7hy5QpVZoqKiu4MT0u3bt3g4uKCqqoqeHp6wsXFhWrAx44dQ11dHcLDw1FZWYnMzMzOfDSFSCTC6NGjsWjRIgQHB8Pd3R08Hg9lZWV46623sHbt2k5J1ebr64sFCxZg4MCBCAoKQnBwMIRCIS5duoT9+/fjv//9L3Jzczv1xYrFYkyaNAkPPfSQndKya9cuLF++HAcPHuy05xEkJSUhLi4Ou3fvxuXLl+lxNzc3zJw5E2KxGJs2beqwdaUjINYiV1dXREREwM/PDz179kRtbS3ef//9TntOexAKhXShYhgGBw8ebHcBaQ9SqRRPPvkkxo8fj5iYGEgkEjAMA4PBgF27duHixYs4derULXl/f1Vw4zIEAgGUSiVcXV2hUqloVhSymQGw4aobjUYUFBRQGgzxyDiixAD2Aj4ROIkQRwRDorCMHj0aM2bMQLdu3RAQEACBQIDjx49j1apV2LNnj43FlbS/rUBoLo2L0LK6dOmCsWPHYuTIkQgNDaVBsaWlpdi1axe+/PJLm3WEK6BzBXk2zYs9zmy6mqurK8aMGYP77rsPMTExcHd3h8VigU6nw759+7B8+XKkpaXZjAs7Qxq3f1zPBNvbQ/psNpsRFhaGvn37orKyEkePHgXQymn38fHB3Llz0aVLF3z++efQ6XQ0uNmR18aRosSmkRHhXCwWUyFLrVZT+hThxe/atQuXLl2i15A+OFKIHPXbUewA16MilUrh4eEBHx8fuLq6orm5GVevXkVFRQWlxrDvzQ6qZ3tiyDmhoaGYMGECRowYgfj4eHh4eMBiseDSpUs4deoU8vPzcfr0aZw4cQI6nY7ev6PUwb8j2kpJyzAMtFotWlpaMGTIEEilUvTq1QvLli3Djh07UFFRQb93oHWcpVIplEolWlpaYDKZqBeGYRhKkZHL5fDy8oKPjw9cXFxQWVlJ36VUKkVkZCTUajUqKioo9ctkMtE5fa2+EFitVjshuKamBsXFxbBYLCgrK0NhYSF69eqFf//73xg2bBgkEgmKioogFovx22+/UdoWwzCQSCRUOCZxKqSvbUEoFNLvyGKxoLa2lnovCG2J7RngKlntpYUm+19DQwNcXFygUCig0WjofkuSUuTn56OgoIAKysHBwfDz86MehMbGRnh5eUGlUiE3NxfHjh2DxWJBv3794OfnhytXriAtLQ0NDQ00bigrKwt6vR4ymYzS8ry8vKjCROhRBJGRkVi0aBECAgJQUVGBzMxMSKVSGidkNBodGoxJ8hZCXySGBGIMYseqEI+zXC6HRCKh+x9Zg8i7IOtBY2MjamtrUVNTQ9+hVCpFTEwMQkJC4O3tDZPJhIsXL+L8+fPIzMxEZmYmLl68iNjYWAgEAvj7+yMqKgpyuRw1NTU0iUVVVRXq6+vh7e0NHx8f8Pl8+hwiU/9ZuGmlRSQSwcfHBxEREZg5cyaioqJQX18PjUYDmUwGoVAIb29vytUnHM+dO3figw8+6Iw+UEgkEowZMwbPPPMMevXqRSdIWVkZnn/+eWzYsOGmPRF8Ph99+/bF3LlzMWnSJKjVapSXlyMjIwMajQbBwcGYN28efvvtNxQUFNywkMuFWCzGvffei+eeew6hoaE4c+YMDh8+DLlcjoSEBIwePRpCoRB6vR7p6emd8kygdSGfOXMmJkyYQDNqAK1jPWXKFNxzzz3IyspCdnb2TT9Lo9GgZ8+e6NGjBwIDA8Hj8eDv74/AwEC4u7sjLCwMBQUFt0xpEYlEiIiIQEhICIKDgxEbG4vu3btDLpeDYRj8+OOP+Oijj27IgzZgwAA8/PDD8Pb2pscaGhqwdOlS7Nq1C6Wlpaitrb3j+Obtge35kEgk8Pf3R9++fTFo0CCEh4fTDU8ul9MgV4vFgpqaGpq+cfPmzdi0aROlOrADrLnPIh4INjeYnEsEe4vFAnd3d6SkpGDRokVITEykrvbi4mKsWLECv/zyC2pqamzoWOyUvQBsBH7i/SCKC1HEoqOj8cADD2DkyJFwcXFBYWEhioqKIBQKERMTg2nTpuHw4cMoLCy0yTBF2u5ornAt9OQaYs318PDA3XffjcWLFyMyMhLp6en46aef0NzcjN69e2Po0KHg8/moq6vDuXPn7DYnrvfCkcLiqB2enp6YO3cuevfujS+//JIqVmKxGOPGjcOwYcNw6tQpXLx40Y7a5ghc7xY7MNrb2xtdunShnitfX1/4+/vDy8sLbm5uUKlUyMjIsDEQEOoh+56O+swGOyYGaH0nLi4u0Gg00Gg08PDwQGhoKLp3746oqCgolUqUlZXh6NGj+P3336nhjpvymvtM8p3weDxERERg0qRJ6NGjByQSCZqamlBeXo7t27fj119/RXZ2Nurr62lAdluJDO4kcJVMqVRKLfg6nQ67du2Cp6cnxo0bhz59+sDd3R1+fn7YuHEjCgoKbIRsiURCPXHNzc1oamqiMSUSiQQKhQLe3t7w8PBAS0sL6urqkJubSw12wcHBGDVqFMLCwtDY2IjBgwfD19cXjY2NOHfuHKqqqtp9F+ysgY5gNBpx9epVmmqZMB1SUlKg1WpRXFwMHx8fjBw5EsXFxWhsbKTzhCjyxJNMvAntxVCKRCK4u7tT6hQRzrleVPIbe02SSCTw8vJCY2Mj6urqHCYOycrKgpeXF6xWK6RSKc0uZjAYUFtbi4qKChQWFqKgoABKpRKpqakYOnQoTCYT9u3bh8uXL9PxqqqqQllZGcLCwjBy5Ej06tULWVlZNobWzMxMmuHMarWitrYWPB6PBrsTbw5RNIFWhWXGjBkYNWoUXF1dkZqaiuzsbLrnEK99W++VbYQQCoWQSqUQi8VUGSTKmqurK82Ap1AoqJJDriHXEc9ieXk5srOzcfbsWVy6dAl1dXWwWCwoLy+nSg6JlSRpky0WC1UGfXx8kJycjAkTJsDFxQX19fUwm80Qi8U0vsdkMqG0tBTV1dVQKpXUQykSidqcM7caN620CIVCjBkzBg899BCCg4PtUtoBf1ixFAoFLBYL4uPjoVKp8M0339hZK28Gw4YNw1NPPYUePXrYBKLpdDrs3r27U6hTPXr0wDPPPINhw4aBYRhs3LgRa9asQWVlJVQqFaZNm4YpU6Z0+kt94IEH8OSTTyIoKAhnzpzBkiVLcOnSJWg0GsybNw8xMTFITU3Fnj17OlVpmT59OsaNG4fi4mK6yfJ4PMTFxeGJJ54An8/HgQMHbtrL4ubmhjlz5mDu3Lnw9fWFQqGgbmNHdJfOgEgkwqBBg+Dv749BgwbB3d0dnp6ecHd3h1qthoeHB0QiEZ3HQUFB+M9//mNn/eoIysrKUFxcTLOUAK2K6IgRIyCRSLBy5cp/lMIC2NKLXFxckJCQgBkzZqB///50HSHUJEJFIAGYUVFRVBE5deoU3UDaor+Q4+xAeCLQmc1mqlTw+XwkJCTg/vvvR+/evSESiWyy+WRlZVGFRalU2nCm2xJuifIikUiocBoTE4OFCxdi0qRJaG5uxoYNG3Dw4EFcuXIFIpEI06dPR2pqKry8vKj1ne0NInOIPYaOgreBP+JrvLy8cPfdd+Oxxx5DREQE9u/fj08//RSnTp2Ct7c3lEoloqOjMXToUJw8eRJFRUVUoCEKIWBrDXakIJHj7ADTu+66C7Nnz8aZM2dw9uxZNDU1gcfjoW/fvnjggQcgFAqxd+9elJeX23jA2DEobaUlJn+LRCKEhYXhrrvuwtixYxEcHAy5XE4pH2azmX5/aWlpqKmpsWkz23tyrexapB0ymQze3t4ICQlBeHg4wsLC4OfnB29vb3h7e8PNzQ0ajQYKhQJ8Pp9mAzp8+DC9F6GNsAUgtqeKHOPxeKiurkZNTQ19FyQNbHBwMMLCwijvnwRjE4WIqwTdiUqMUCiEQqGgQedkPAsLC7Fy5Urk5eVh8uTJSExMxMMPP4yIiAisXLkSp06dgtlspsHmZN0ngh6ZJ/X19ZBKpWhpaUFtbS10Oh2MRiP9XSwWo2vXrujfvz/kcjkaGhrQp08fjBs3DhaLBZWVlfjxxx+xcuVKSmO7ERAPh4eHB/71r39h8ODByM/Px7Zt25CVlYXo6GgMHDgQQ4YMwbZt26DVamE0GumYkPlO1lbikXPkVSS0OblcjqamJhp4TrwMJOGB0Wi0m1MajQbx8fHQarU4c+aMTRA+QU5Ojo2Q3NLSAoPBgLKyMhQVFSEvLw/l5eWQyWSYNm0a7r33Xuj1eqxbtw4//fQTqqur4ebmBqVSicDAQCQmJqJ///7w9vbGrl27sGnTJhQVFdFnX7x4EfX19fR7INnOSB/q6+upwtrc3Aw/Pz9MmjQJqamplLrXtWtXAHDYHy5EIhHNYOfj44OQkBCEhYUhMDAQPj4+1JAiFoupUsL1sLD3JrLem81mGI1G5ObmQi6Xo6SkBHV1dTCbzSguLqZxOySzmVAoRFhYGJWrqquroVKp4OPjgy5duoDH48Hd3R2ZmZm4cOECNBoNIiMj4evrC71ej5iYGPj7+2Pfvn3Q6XTQ6XR/X3pYc3MzCgsLaUEdg8GAoqIiaLVaXLx4ET169EB6ejpKS0uhUCjw3HPPYfjw4XBxcYGvr+91Ky1isRipqam4dOkScnJybH6Li4tDdHQ0xGKxzfHg4GB88sknePjhh+kHeiPg8/mYPn06hg8fDh6Ph59//hkvvfQSSkpKqDCiVCoxaNAgREZGQiqVdkomqH/961946KGH4Ovri5MnT+L+++9Hbm4uNBoN7r//fkybNg18Ph86nQ59+vTBwYMHO8Xz0bVrVyxatAiBgYH47LPPaMEkqVSKxx57DEFBQdi3bx+WLVt20yk0g4ODMXbsWISGhlK3+O7du7Fjxw5MnToVw4YNg1AoxI4dO266X2PGjEH37t2RnJxMBRu2gtKWkrRnz54b3uwvXbqEBx54APfddx+GDx+Obt26QSwWY+jQoYiPj0dQUBA+/PBDSlf5J4At4BKBPC8vDzqdDmazGVqtFqWlpairq6OCZEVFBVxcXDB//nwMHz4c4eHhlCvOFWi5lBh2HQs3Nzd0794dRqMRmZmZ1Oro4uKC+Ph4xMXFQSwWUwsg0GqwmDlzJoxGI86dOwej0UgFaUeZvRyl3QVaLZDjx4/HXXfdBYFAgJ9//hn/+9//UFxcTNNjuri4IDY2Fmq12i6tJVvw4PaRS30jAqu3tzemTp2KRYsWoUuXLvj999/x8ssv4/z581AqlUhOTkafPn0gk8lw9epV+Pv7w9vbG/X19QBss64RcIVfttDD5/MhFothNpvRrVs3LFq0CG5ubti5cyfKy8sBtCbxuOeee2g2tt9++40GrVosFjvDD9frw1VOhUIhvLy8EBoaSoup5efnIzMzE0VFRYiKisKMGTPA4/Fw4sQJOq/YxT7JO2wPhB4UHh6O+Ph4JCYmUj69TCajgodIJIJOp0NZWRnc3d3h7u4OhUKB8vJyFBcX0/ux1072++PWtmlpaUFmZiY++eQTXLhwAbGxsUhKSoKbmxtSU1PRq1cv7Nu3D59//jnOnj1LqUBteaX+7mDPP6JwEBoTsfqTb6W+vh6XL19GZmYmHnzwQYwfPx6TJ0+Gt7c3PvroI5w6dQoCgQAajQZubm42VFOyH5jNZhrEzjCtQecymQxqtZpmGCO0KZlMRmu8qNVqXL58GVarFU888QSUSiX+9a9/0X4IhUJK4zIajdBqtdc0AvJ4PIwYMQLjxo2DwWDAunXrsHXrVly9ehUFBQWIi4tDTEwM1q9fT+camZvECEAyUpG2CwQCSCQSmyxrbJoc+T/pG4kFUqvV8PHxgbu7OyQSCUpKSlBcXExjgoRCIWQymY03kVjsSexFc3MzSktLwTAMKisraUpjnU4HHx8fPPzww5g5cybKy8vx3Xff4aeffkJFRQXkcjkiIiIwcuRIpKSkIDg4GCaTCV9++SVWrFiByspKGyMLW2EhIO+VKC2EIRQYGIiZM2di3LhxyMzMRHp6OsaMGWNHv20PJMA9KCgI/fr1w/DhwxETEwOFQkEpYB0xcBuNRpSXl8NoNCIoKAguLi40FKOiogJVVVX0XDI3yRgTD22PHj3AMAzOnz+PwsJCiMViHD16FJ6enhg1ahQ1DC5btgyXL1+GRqNBjx498OCDDyIxMRE+Pj4ICAjA6dOn6T71Z+CmlRar1YqDBw/izJkzdJMnH4PRaMSePXuoBk00TiIYsr0hHUF0dDT69euH+fPnQ6vVYty4cTa/szmDbIFBJpNh0qRJ0Ol0ePjhh2+4r3w+H/Pnz4dQKMSePXvw73//2yZ7kIuLCw1+IlzGm4FEIsEjjzyCxYsXw9PTE4cOHcLixYuRk5MDb29vTJgwAXPmzKHCxdq1a1FdXQ0/Pz9cunTppjYnNzc3vPrqq+jRowcOHz6MzZs30/vFx8djwoQJqK+vx/fff39TViOC7OxsPPHEE5DJZHRhqa+vh1KpxPz588Hn85GVlYX33nvvhu4fFhaGWbNmITk5GeHh4VAqlXRTYX98paWlSEtLw7Zt29C9e3fMmjULbm5uuHDhAt56660bpvtZLBacO3cOb775Jj799FMMHjwYsbGxSE1NRWxsLO655x5YLBY8++yz/4iUx9wFT6/X4+DBg0hLS6MWwfr6eip8EEG2paUFcrkcgwYNQkpKCurr6+kCSgSytr47siYEBARg5syZGDNmDE6cOEEpCOR6qVRKrWo8Ho8GUqrVatx///2oqalBdnY23ZDJvR1lcGEHYpO5o9FokJKSAqVSiV27duGrr75CdnY23VxlMhkVntjeI/bYsZUWwF5hY2+sSqUSkyZNwrPPPgtfX1+kpaXh/fffx9mzZxEWFoaHHnoIY8aMQUhICLRaLU6ePIm6ujqEhobi6tWrNgIU9xlcsKlSJPB9wYIFiIqKwuHDh5Genk4tvEOGDMHdd9+Nuro6/PDDD8jPz7eLMWI/gy18sK2OBCaTCbm5uVi7di22b9+OhoYGlJeXo6amBhERERg0aBAEAgH27t2LvXv30n6xn9PWOwRaraZdunTBwIEDkZKSgoSEBLi6utIgZyJwEj54QUEB5b5Pnz4d7u7uOHPmDPbv309TurLjfoinxxHdhGT40ev1OHToEH13vXr1QlBQEAYOHIiBAwdCJpOhqqoKOp0OhYWFdtnXHMXn/F3BVsaIQE5ofiSdNokbIPPp5MmTtNL9zJkzMWTIEJjNZnz33XfIzc2ltByitHDXKYPBQL8topCIxWI0NzfT6uUtLS1wc3ODTCaDVCpFTk4OnnjiCWg0Grz66qvo378/rfUkEAgQEBAALy8vaDQaSCQS5OXl4eLFi+32nXw/UqkUmZmZ2LBhA7KzsyEUCmlcIPE0Ee808ZiQdYjURSExviR1tCOYzWZKmSLeg6CgIIwaNQp9+vRB165daRB3Y2MjsrOzsXfvXuzZsweXL1+GTCZDaGgoXdONRiN4vNY6M3V1dTAYDBAIBGhoaKBJDcLCwtClSxfcdddd6Nu3LzIyMrBixQocOnSIppTu3bs3pk+fjkmTJsHV1RW1tbX43//+R2Pj2pozIpGICulA63dPPElGoxGRkZGYPXs2Jk2ahKKiIqxYsQIhISEYMWIEysrKOizbkbXOz88PCQkJ6NOnD9Rqtc26XVJSQmOFDAYDTbNMlOaamhoUFRVRjwcpCtrY2IhffvkFW7dubZf9wTAMSktLERgYCLVaTePqSAxLWloarFYrRo8ejdDQUAQHB2Pfvn0AWpPO7N69GxMmTMC4cePg4uKC8PBw6HS6v6/SAsBGq+OCPXEIp89isdxQkcfk5GS88cYb0Ol0+OKLL+x+Z2/c3IWZVKl1dXWl6TRvBMSqsn//frt0p6NGjaIULnaRqBuFh4cHHnjgAXh6eqKpqQkffvghLl68CG9vb3z22WdISUmhGUAYhkFZWRlcXV0RFhaGI0eO3BQdrlevXoiPj4dYLMaPP/5IU0UKBAIsW7YMAoEA33//PTZs2HBTfSRoamrChQsXaF+A1v6/8847NAB/yZIlDlNWdgSrV69Gz549IZPJqMDT3NyMrVu3YvXq1aitrQXQGmdSUVEBPp+PUaNG0fElMUo3C61WSz0IYrEY33zzDfbt2wc3Nzfcc889qKqqwgcffHDHKy5cIZsUuCICB9vCTJQIIhCEh4ejS5cuNMiwtLT0mgIZyQ4lk8kwePBgLFq0CFKpFCtWrMCVK1ds+MMkoJRY3YlLvb6+HiqVCmPHjsX27duRnp5OM4AR5aQtxYHdNrFYjNDQUBiNRqxevRrnzp2jNQ7kcjlGjBiBBQsW0NSdRFAgAhk7nsXR5smNi4iMjMS8efPg6+uLEydO4M0338Thw4cRHByMF154AWPHjqXWxZqaGmRkZNDYHqlUCoPBAJlMRhU4R+AqEBaLBQKBAKmpqRg9ejSqqqqwYcMGZGVlwWKxICAgAAsXLoSXlxe+/vpr7NmzhyqnxPPQHl3SkUekpaWFBj+T8bZarYiIiMDcuXMxcOBA5ObmYtWqVcjIyKDtbes5XLpd9+7dMX/+fEybNg0KhQIikQh1dXU4fPgwcnJyUFRURIvq1dTUoKqqCj4+PnjooYfg5eWFkpIS/PDDDzh58iTtK7tuAwAb5ZzdrqamJhvqWk1NDerq6pCVlQWRSIQ+ffrgsccew4gRIzB9+nQYjUasW7cOBQUFdums7xSwvVLslOgkjTbbu0q8H3K5HAUFBVi6dCkKCgowffp0pKSkQCQSYcOGDTRGgh0ozX4OYQFYrVYolUr6XRD6VHV1NXJzc2mGr7Nnz+Ltt9/GkSNH0L9/f2pZHz16NFatWkVTCZP7EMMuAEptJBXR2d+XXC5HQEAATCYTjh8/TotWCgQCREdHIzw8HJcuXbJRfltaWmw8DVarFZWVldRKbzabHSYpIhSjhoYGem1ERASmTZuGIUOGUCNTbm4u0tPT4eXlhWHDhmHgwIFoaGhAXl4epbiyi0syDIMrV67YJAoxGo3o0qUL7rnnHvTu3RshISHw9PREWloaPvvsMxw+fJjSsrp3746pU6fSTJzl5eV488038fXXX7frqRKJRHBxcaFeJZLWVyQSUVplXFwckpKSoNVq8emnn2Lv3r14+OGHIRAIcOXKlQ4rLSSehCh9+fn5kEqlNB4lNzcXV69eRW1tLRoaGtDY2GgTG0TmrkajQXJyMiIjI6FQKFBbW4t169bhq6++6pCsoNFoMGDAAAQEBMDV1RXbtm1DVVUVjemZP38+nnvuOSxevBgPPfQQKisrkZGRgebmZpSXl2PFihW4evUqJk+eDE9Pz06L1b4R3LaUxzweD/fccw/i4+NRXV2Nd999l9IEOgqSxzsjIwPr1q2z+33Dhg3IyMiwKSJGsH79eiQlJeGDDz7AggULbqovzc3NVBMlGDJkCJ544gl07doVGRkZyM3NvWnKVEpKCs3U8OKLL2LXrl0QCoVYs2YNhgwZYuOpKi4uxsWLF6FUKnHlypUbVlikUikGDRqEN998E+Hh4dQVSxSw8ePHo0uXLjRjTWcW6eTyrp9//nmMGDECBoMBr776Kg4ePHhD3qvnn38ePXv2pKkdt2/fjo8//hharZbmc2dTdwjnecSIESgqKsKaNWvw/fffd2pfiWCcn5+PBx98EN26dcOrr76KBx98ELW1tfj888/vqKrVjsDOSkWoA+wsV8QKTayb5Jzk5GQkJSWhubkZaWlpuHLlCoC2q4yz/93S0gJ/f3+4u7vj8OHD2LVrF+VpE57wli1bkJOTQ2sPkI0+LCwMjz/+OBISEjB58mSUlpZSeqsj6hLb60AssyT9JkkYUF5eTvnTIpEICQkJePTRR5GUlIQDBw7g3LlzVGnhBr+y+8W20rPpZGKxmAoxNTU1WLt2LS5cuIDAwEC88sormDZtGlVGiBBz+fJlCAQC6PV6uzoh3D6yY4jYiiPZJBcvXgx3d3f8+uuvyMjIoOvIiBEjMHLkSFRUVODnn3+m48gwDI39IVSxtrxn7NgdEhdCqnWT9+Hl5YXZs2dj8uTJ0Gq1WLp0KXbs2GHzbbFjWQiliOvpcHd3R0JCAvr16weNRoOamhrs3LkTu3fvRmZmJuWUs6+NiorC/fffT73SK1aswKZNm6iXhbxHMn4Mwzhct8nv7KKkZC4Ty/CRI0do8HlqaipmzpxJFThS3JA8406LaSEcf7KekAKHpKgfiXUh6WtJcH11dTXkcjmeeuop9OrVC9nZ2fR7ZJjWzFbkuwD+qN9Egqd9fX2hVqtpUDfQGui9YsUK1NbWwmKxoKmpCb169UJZWRlMJhP8/f1pqQACkviivr6epp8FQDPPAa3xdOy5weO1pqU1mUw4efIkpbBaLBaoVCqajSssLAwXL16kQjz3O2poaKCB6kTJY4MEhRPjc0xMDPr3749+/frBw8MDGzZswLZt23D48GE0NDSgpaUFUVFReOONNzBu3DgEBQXZGDzIeyEgBgY+nw+ZTIYBAwbgySefxNChQwH8ocQTgw4JRO/evTseeeQRTJ8+nRoP3nrrLaxYseKa+yYxZpDCo2Sd5/F4NMubwWDAjh07sH//fuzduxcA6Bp8vfG7JJ38sWPHkJGRgdraWppcQKfT2axv3P8DQHh4OGbMmIE5c+YgODgYRUVF2Lx5M00k0RYUCgXUajVCQkJw3333YcKECcjIyMDhw4fh6upqkxSivLwcb731FuRyOSZPnoxRo0ZRj09jYyNKSkqgUCgwYMAA+Pv748SJE7cs1vhauG1Ki1QqxZIlS2yoHzcigPJ4PPj4+CAuLs4u4Dw7O9suzgUAJkyYAABUg75RMAyD48ePo3fv3nj33XexYMECXLlyBbGxsXjkkUfQs2dPCAQCbNiwAYWFhTf8HILevXtDJpMBaN1809LSEB0dDaFQiLy8POTm5mLs2LFoamrCiRMnkJ6eDl9f3xv2JAmFQsyZMwcvvPACraw8cOBA/P777zRt3pIlSyAUCnHo0CGsXr36pvvYVjsWLlyIadOmQalUYvHixVizZs0Ne642b96Mbt264dKlS/jtt99w/vx5ujFxoVQqMWfOHEydOhUWi4VyYwm3/0ahUCjs0kECrRbU7du3Q6FQwGq1wtPTEwkJCfD29rbz5N1pYFvLCZ2DZDpiC1YMw1DrmlqtpoH6RUVFyMzMpBsOW0ngej3Y6YjJxuTn54eIiAhUVFTY8Lrz8/Nx6dIliEQiumHz+XyEhoZSj0RYWBgkEgkAe7oWd16RvhC6ik6nQ2ZmJvr27Yv58+dDIBAgLS0NkZGRmDVrFrp3747GxkacPXuWcujJBsKmE5Gx4YLw8Bnmj7oeQqEQDQ0NiI6OxrRp0+Dr6wtPT0/k5OTg6tWr6NGjB6XPHD16lFb4JuPiqDI0d4MlCoZGo8H06dPx6KOPIjIyEhaLBYmJiXj//fdRWVmJ4uJi9OvXD3w+H2fOnAHDMPDx8aFGLHbsETdmha2QsJ9PriHWavK+Bw0ahOnTp0OhUOC9997DunXrqDWZLcg76hf3HRIFa+3atcjMzMTZs2cd1qQQiUTo3r07Hn/8cUycOBF1dXX47LPP8P3336O8vNzG8t0e2IqwUCiERqOBp6cntFotHSsSB9HU1IQDBw7A1dUVfn5+6N69O4YOHYq9e/dCq9XaURXvJBALPqE2EaoW0OrVJEXzDAYDVY69vb0REBBAa7iRpAUkExap1cH2cEgkElpgMSAggNKyr169Svdcq9WKkydP4tSpU2hpaYG3tzeNB6irq4NQKIRIJIJSqaTtNxqNKCsrowH9BPX19TQmgmSBIiA0opaWFhQXF9OCgP7+/vD19YVYLEZERAQGDx6MkydPori4uM21ghh9uHB1daX1YPR6PZRKJe655x70798fv//+O5YvX46jR4/SeSUWixEQEIDo6Gi4urqCx+PRpDrkPVRXV9soLSR+jyRieeCBBzB48GDodDpUVlbS5DgkZuvUqVPUS3zvvfdSJfWXX37Bjz/+aLe/EqptbW0tHT9iCCMKKAlrIEwgk8mEI0eOYPfu3dT7T5J5XIuCzAVJY0zW/IaGBkrfvNY9FAoFkpOT8eSTT6J///6QSCQ4fvw41q9fj7S0NBrf6Ug24fF40Gg06NevHxITExEUFIRjx47hv//9Lw4fPky9N2R/AVpZUVu2bEGPHj2gUCig1WppO7VaLbZs2YK+ffti3rx56N+/v8OiuLcDt01p8fPzQ2xsLA20upGK9FlZWWhoaEBkZCTWrVuHadOm2bj5AfsN595778WKFSvohkfqA9wIrFYrNm7ciC5dumDo0KFIS0vD0aNHER4ejujoaBsudmcgPz8fzc3NkEqlePbZZ6n1dOPGjXj33XcxaNAgjBkzhnLUKysrbzi+RCQSYebMmfjXv/5FCxxarVaEhIQAaE1ywK4XkJycjOPHj+PixYuYMWNGp/SXID4+HjNnzoSvry927NiBTZs23ZSXIzs7G3PmzLnmeRqNBvPnz8eECRNgMpnwxRdf4Ntvv71hhUUsFiMlJQXDhw/HqFGjsHDhQhw7dsxujpLYr1WrVuGBBx6gC82drrQAtt8ru9gg2TysViu1cJlMJsTHxyMhIQFWqxV5eXl2cRCAPa0HgA3lKDs7G6WlpYiIiMAzzzwDb29v7N+/n6ZSFgqFaGxspB4QAJg9ezaee+45BAcH04Doq1ev2ng4HIFQ3dj9tVgs+O677+Dj44PZs2ejX79+KCgoQGhoKMLCwiCVSlFRUYGSkhKaXpSdppmdTYut2BFPFHsNIgGuZrMZ/v7+mDp1Kq3t8c0339BNqnfv3igqKsLWrVtRWlpKsyQR6zW5H6GpcUEEYZFIhOHDh2Px4sWIjo6mwbgRERGIjIyEwWCAVCql9J3u3bvjww8/xM6dO7F8+XIUFRXZKEfcQH/Sb/aYcGl4RGBNTEzE448/jrCwMKxcuRI//PADVVjIfsCee229O6vViurqahw8eBBHjhyBXq+3ESBFIhE1xhmNRri7u2P06NFISUlBQ0MDli1bhq+//trGw9IWyHxiC0cCgQBJSUmYM2cONBoN1q5di19//RVWqxUGg4F63BsaGnDgwAH06tUL0dHRtPBxYWGhjfftRoyFf0Ww5yIpJkmoYoR6RwKegdZ54u3tjV69emHo0KHo06cPBgwYAIZpLbhIPBKEOkxoWWRuKRQKBAcHU08twzBQq9UYOHAgmpubUVxcTN8daVdpaSl++eUXxMfHIzg4GHq9HvX19TZt0ul0dgoLAJpxjKxFbFitVmzevBn9+/endMX8/HyaOQ9opZANGDCArm2OqF+k71yQwo4SiYQGgLu6uiIxMREBAQHIy8tDRUUFDVjv3bs3unXrhj59+qBnz550fHx8fGhSIgAoKCigfeHxeOjZsydcXFzg4eGBvn37wsXFBUePHkVGRgaKiorg7u6OpKQkDBgwAF5eXkhMTITBYEBCQgI1zDQ3N2Pv3r02Y6hWqxEdHQ1vb28IhUIcOHDAJlUz8d6Q4sWEnUPqf1VWVlKZQyaTQaFQUFoUMRgQz2Z7WcRUKhX1iOl0Ouj1ejQ2Ntp4yCUSiU3dNuItGzduHJYsWUID6Ddu3IhVq1ahoKAACoUC/v7+UKlUuHLlCqW3s99rSUkJtm3bBr1eDx8fH/B4rfUKydwkqb+Jt6esrAwnT57EoUOHEBQUBA8PD1oUVSAQ4PLly/j222+RnJxMM6j9GbhtSgvJ2tLS0kJTzF0vDhw4gP3792PcuHEICQnBxo0b8c033+C///2vXR5wgUAALy8v3H///ZRPWVJSglWrVt1wHxiGwdKlS5GTk4Ply5fTzYnQEsRiMfR6PaqrqzuF2kOekZqaCpFIhOrqarzzzjvYt28fwsLCMHv2bGqFvtmCmX5+fhg6dCh8fHxw+fJlrFu3Dvv27aMf5NixY/H000/bbHx+fn7Yvn37TfeTjeDgYDz33HNISkrCb7/9hueee65Dm/3NQiQSoX///pg5cyZEIhFeffVVLF++/KayzanVarz44ovo168famtrbbKycFFeXo7NmzdjzJgxcHd3h6+vb5sC4p0CttDNVjKIQEuswiSehc/nY/jw4QgKCoLRaERWVhZdrNmeDkeWZLaX4sSJE/jll1+wePFipKamws/PDyEhIVi3bh3KyspsBGA+n4+oqCjMmjULwcHBAFrTZh44cIAqVIC914FY7UgMAmArGP/0008Qi8V4+OGHERYWhoiICJq9BgC1CpONlet56KiVy2w24+jRo1i6dCkGDRqEhoYGmM1mbNiwARs3boSXlxcWLlwIhUKBS5cuobi4mMYEkAxeQqEQRqPRjjLFVZqsVisCAgIwcOBA+Pr6IiMjA5s3b8aJEyeoJTw5ORmPP/44BAIBKisr4efnB7lcjoqKCpuAYS79jT1+hOLEBVthi4iIwMMPP4z4+Hhs27YNq1evpgYdIuxcC+y+EmsjG0QwBkDjrYRCIcLDw5GSkgK1Wo3169djy5YtdvOUDUe0O3Zq69jYWDz11FMYM2YM9u7da1ONmngGiLBVVVWFw4cPY+jQofD390dMTAxOnjxpQ4G9U6hhhEZIPLPs98oWRK1WK9zc3DB06FAkJydj+PDh8PLyokpmVlYW1qxZg/T0dBiNRkoPIt4LMl7E08Ln81FUVITGxkakpqYiPj4ex48fp9m3SHwE2ztZU1ODMWPG0AQfZC5arVabecX1IgJAY2Oj3XxnGAarV6/GY489hocffhgPPPAAysvLoVKpUFhYiLS0NPTu3Rs9evTA1KlTYTAYsG/fvnblLhKIT7wmpCYWycCl1+tx5MgR+Pj4YPHixZDL5dTrR5KZZGdnIzQ0lMYAlpeX0+xher3eLqFIv379aOKm/Px85OXl0fIKZWVlEAqF6NatG+69917MmTMHHh4eGDFiBE2Jb7FYsGvXLly4cIGuxTKZDLGxsYiLi4NKpYLFYsHx48dpH0n/SPwQMaIQ5b+xsRFms5myMUg8mZ+fH8xmMwoKCujaL5FIaFY1R2BTykhhZO66QmrfsNf2qKgozJ49G/Hx8airq8Pq1auxYsUKVFdX01pyQGttGnKtI4aN0WikCTmGDh2KK1eu4KOPPqKUUVdXV5o0pbq6GgaDAefOnUO3bt0QHR0NAAgMDIRAIMDVq1dpfZzQ0NA259Gtxm1RWtRqNWbOnAmGYaDX6/H222/f0H3MZjOeeuop+Pr6onfv3ggKCqJF0r777jucOnWKaobDhg3DvHnz0LdvX/rBb9myBVlZWTfVF6vViu3bt2Pw4MG45557IJVKceXKFQQFBeHJJ59EVVUV8vPzOyX+wWg04uWXX8bLL79sc1wkEmHo0KHo2bMn6urqkJ6eftOB21qtFpmZmeDxePjss8+Qnp5u8yHGxcVRC9I777yDhoYGZGRk4MiRIzf1XDbY7ucLFy7gww8/RE5Ozm3ZZN3c3JCYmAhfX19qHb6Z52o0Grz99tvo0aMHAGDatGnIyspqU1iyWq1oaGigBViDgoIgl8tvSmn6q4OdJpgInMSTwK5XQTaFrl27YvTo0QBarcrHjx+ngjVbIWjLfU88BrW1tVi/fj169+6NpKQkJCQkIDw8HElJSVi/fj1Onz4No9EIPz8/dOvWDVOnTsXAgQNptrBdu3YhKyuLxuAQxZ5LDeN6XdnKUH19PVatWoX09HRERUVBo9GgubkZ0dHRmDhxIgwGA6qqqmj2GUf3cOQdYFPjyPhdvXoVH330EVatWgWDwQCxWAytVguxWIwJEyZg4MCBqK6uxt69eykXn4w/obSRfl2rRotEIkFhYSE+/fRTHD16FMeOHaMCM6HYkYrTa9asweXLl1FaWkoDbNlVw7keM7bQzRXiiMBiNpvh6uqKKVOmYNiwYTh06BCWLl2Ks2fPOqSckXY7Ok6UZu57JcIN8Qyy4e3tjXHjxiE+Ph65ubk4ePAgVYQdxT1xn8t9np+fHyZPnoxBgwZBLBZj27ZtNMBapVLROkFE4CRjazQaoVQq4eHhAYlEQpXhOymehc/nQ6VSwWg0UsOAyWSC0WikRRG7du2Kfv36YcSIEejXrx+dX83NzaipqcGKFStoGm72WBqNRpt4FqDVag60MiAuX74MV1dX+Pv7w9PTk3omDQYDpXOxFaqYmBhMmjQJcrkcZrMZv/76K4A/qK8k9iYgIAAikQg5OTl2MWwymYwaD4BW+WDOnDl48MEHMWrUKBQWFmLTpk3YtGkTNBoNFi1ahClTpmDKlClwd3dHY2Mjjh492iYbhNDsgNb1VSQSoaGhgXoudTodPvzwQzQ2NmLKlCk041dmZibKy8vh6+sLLy8vXLp0CUVFRRCLxTh48CBOnDhBY3AIbYwYWT08PHD+/Hmkp6cjPz/frk1msxm1tbW0IC9JS028JefPn8fq1atx5swZqkh4e3sjMjISwcHBEAqFKC8vt1m/BAIBzbBGEmqw4/GIEsD+LhsaGmiWOPItmc1mqji3BbPZTIP9ScIqR3sT+5ivry8mTJiA2NhY5Ofn46uvvsJ3332HhoYGxMbGolu3bpDJZDSDGvGEtBUWUFBQgNOnT2P48OFYsGABmpqa8P3331OPMXm/xPAuFovh7u6OkJAQ1NbWIjAwEG5ubggMDERjY6NDCvTtxG1RWp544glERUWBYRhcuHDBYdxJR1FaWoqvvvqKFu/SaDS0jsfp06dRUVGBqKgoxMfHUxcs0LrQfPTRR53RHTAMg4KCApv0u/Pnz7ex+NxKkGB5hmFQU1ODrKysm1aS6uvr8cknnzj8jVSB5fP5+P333/Hee+91ugdAJBJhxIgRmDNnDq5cuYInn3ySWkduNVxcXDB27FjMnz8fLS0t+P777/HLL7/clPdKoVAgMDCQxiT5+vpCo9FAp9M5XOQUCgViY2MREhICnU6HkpKSTg38/6uDLcBxg6RNJhPNRBUfH4+WlhakpaUhLS3NIceYzd1nu+FJkDrDMDh58iSWLl2K5557Dl27doWLiwsmT56MPn364OzZs6iurqZ0BHd3d0o/OX78OJYvX46GhgYIBAK64DsSbh29Z/baQBIJnDlzBi0tLVAoFJg6dSpSU1NtAtAdCdltCZ5shYUNkgWGrdQEBARgyJAhUKvVKCsrQ0NDAy0AzE4Ty7YAcpUWdpYmklhg06ZNtAo28EfcRXBwMFJTUyGXy/Hbb7/hP//5DyoqKiCRSGyoG+wEDW2hrTGRSqVITk7GtGnTcPnyZXzxxRc4fvy4zf3Zc6K95A0AqLWeKDzkvbDTFBPqCykWO378eADAkSNHcObMGRuDUkc2e0JxIqlzY2JioFQqaTCst7c39T6T+xGOvpubG+Li4hAQEACtVova2lqbd0aKpd4JFDGz2Qy1Wg2JRELjdhQKBWQyGbp27Ypx48ZhzJgxtCYW0BqbodVqaTKW3NxcG4W3PYSFhQEAKioqUF9fD4VCgdLSUnTt2hUTJ06ETqeDxWJBTEwMIiMjqRdCpVJh6NChiImJwdWrV5GWlkYT+ZB3ptFo0L17dwwaNAhnzpxBTk4O3NzcYDQaqYeeFFpm14Q5ceIEzpw5A39/f5SWltL1iJQjIAmQUlJSUF1djcLCQiiVSuTl5bXb3+bmZmi1WhuDBdDqhfjwww+xe/duZGRkQCAQoH///hg/fjxmzZqFpqYmfPnllzhy5Ag0Gg3Onj1L99H6+nrExMRgwIABWL9+PQAgLy8PBw8ebJPWLpFI0L9/f4wePZrGARED36lTp/DBBx/Y1G4TCAQ0EYHFYkFpaSkyMzNtKN5k3SJ0QmJ8IOsDqT/DhdVqhV6vp+NBvDHtoampiVLROvLdubm5Yfjw4Rg8eDDKysqwZcsWbNu2DTweD7169UKXLl0gk8mogZNhGCiVSrvahGxYLBZs27YNXbt2xdSpU7FkyRKEhYVh/fr1dG0k65tMJkNISAg0Gg2lxWk0Gvj5+cHT0xM8Hg9eXl7t9uFW45YrLSqVCpMnTwbQurA+//zztEjhjcBkMmHlypXYuXMnPv30UwwbNgwymQweHh5ITU21sdKRDfDq1auUX32rUFtbi+rqaiiVSri7u98yag+Px0P37t0xduxYNDY2YufOndi6dWunP4fA1dUVb775JlJSUmAwGPDxxx93er/4fD569OiBRYsWwcXFBV9++eU189R3FsRiMbp3744ZM2bA3d0dO3fuxJo1axxafa4HJpMJ1dXVNCZp+fLlWLVqFbZs2YLMzEzqQZHL5fD09KQZy8xmM37//XcaLHcng+01IGALlGwBPywsDBMmTIBAIEBVVRU2b95Ms95wsyqROACuGx4ALSBpMBiwefNm1NXVYdGiRRg8eDANPCX8X7KJ1dXVwc3NDRcvXsRTTz2F3Nxc+ju3dogjhYGdIY0oUyTGgAjFJpMJzc3NNCmAl5cXvLy8IBKJaBFL9niRZ7DXGUeB5ez4CBLUSzIoRUVFoUuXLqivr8e2bduwf/9+m2K/7AKV5FlcRYytQDEMQ6sls9vT0NCAwMBAzJ07FyNHjkRNTQ2+/vprSgdjKwBcChM3gJzrdSEgQkfPnj2xcOFCqFQqrFq1CkeOHLEpskau5d6HGJu4CjBbCSR9ZY8BqenR0tKCPn36YM6cOfD398fOnTuxfv165Ofntxnv5EiAIV5e4nHU6/W4fPky9Ho9fH19MWfOHHh6emLPnj3Iz8+ne5xUKoWXlxcGDBiAGTNmQKFQYPPmzTh06BD0ej3tGztb2d8dRqMRCoWCFpf18vJCQkICQkJCqALg7u4Oo9FIef/btm3DN998Y1PEt6P7WUhICEwmE80wVllZiV9++QVisRh33303xo0bh4qKCkRERCA0NBQmkwlpaWmoqqpCYmIiKisrcfz4cezatYvekwSvx8TEICEhAYGBgTTBEPFK5uXloaGhARaLxa74I2l/UVGR3dicPXsWWq0W7u7uuOuuuzB16lS8//77NKNgTk5Ou4yCtrwCAJCRkQGpVIro6Gi88sorSElJQU1NDTZu3Ij169fTAHau4a+oqAjh4eEAWuf67t2721RYFAoFUlNTaWp0rVZLPYk//vgj1q1bZycnkN/z8vJw/vx55ObmoqSkxM4T39jYSNNPk++bUGG59aEI+Hw+DAaDzRp1rT2aJN9hG3/agkqlwqBBg3DXXXdBJBLh6NGjtP5hVFQUvLy8YDQaUVNTg4aGBvpsQnUDQDPlcRkaly9fxieffIKGhgZMnz4d06ZNQ2xsLPbu3YuLFy+ioqICpaWl8PPzg0KhwPHjx5GRkUEVb4ZhIJfL4e7uDg8Pj+uusdiZuOVPnjNnDuW/VVRU4MyZM51y35KSEvzrX//Cs88+i6FDh8LPz48KOWSTNhqNuHTpEl566aVb7hI/f/48srKyMH78eAwcOBBHjx697pTOHYFAIMDEiRPh5uaG0tJS7Nu377pT8F0P7r//fsyaNQs8Hg/fffcdDhw40OnP8Pb2xowZM9CjRw/s2rULe/bsuS20KKlUioEDB2Lu3LmIiorC9u3b8eWXX9plpbsRVFVV4csvv4Sfnx8GDBgAmUyGRx55BJMnT8aGDRtw/Phx8Pl8dOnSBcnJyejSpQs8PDywfft2vPTSSzab6p0KtuDIVlxIJjEiwBEvXM+ePcEwDE6fPo2dO3fa3IsInUSg5NKa2J4Lo9EIsViMpqYm7N27F5WVlViwYAHGjBkDX19fSkkjljG5XI709HS89NJLOH36tE1hQ7YA6Kj6uKNMV0T4ZQvMQKt1s7CwEMXFxQgLC0PPnj1x4MABlJSUUI48+97c/rNrnBCQTGIk9oEEXSqVSiQnJ6Nnz54oLy/HmTNnUF5eTqkSjgQ5R5suV4khY0MUNVIbZ8yYMZg1axYsFguWLVuGw4cPU6WNCOnk39zxamuzZ79Tq9WKsLAwzJkzB3Fxcdi8eTN27txpo0C1h7aUIW7KZ/a7JzVWSKD8woULERISgp9++gkrV65Eeno6tdg6iqNx9B6JEkf2q4qKCuzZswchISG0IK23tzfi4uKQm5uLwsJCNDQ0IDg4GDExMejVqxdVar799lucP3+etrGt9/p3BTEoSKVSeHp6YsCAAXjwwQcRGxtLC3FeunQJJ06cwJYtW7B//367gOXrAZ/PR3R0NDUUnDlzBkePHkV+fj7OnTuHYcOGISAgAAaDAZcvXwaPx4NSqaQFJjMzM3HhwgWb+AtS4JjEj5D0uECr0uLi4gKBQICLFy9SpcURiPBNFFMyNy9evIgVK1bAw8OD3u/ChQsIDg5GUFAQSktL7ZQgMlfIfdgeSULLAoCuXbvi+++/R2RkJGpra7F7924sXbrULoEMMUC1tLTQgogEjhQWkUgEDw8PjBkzBgsWLIBKpcKOHTtw9epVNDQ04OLFizh06JDDay0WCy5fvozy8nIaZ+YIRJkiqbDJOkOyu3GFcpLxjV18k1uY2hE6mpTJ1dUVAwcOxIwZMxAQEICLFy/i9OnTlBrX3NyMK1euoKmpiQbr8/l8SKVSG0+LSqVC3759ceDAAbv3WlhYiFWrVtHC7HFxcejduzd914Sqe/bsWezcuROVlZW0FhBJvU9qBkokkj+NZnrLlZZZs2bRwksbN25sN9PC9cBqtSInJwcvvvgixo8fj0mTJiEqKgre3t6U37d161b8+OOPNpaNW4X8/HycPXsWI0eOxJQpU/DLL7+gqqqq0zKJEfD5fNx11130b6FQCG9vb7i4uKCmpuamCmdy4e3tjZkzZ0Imk+G3337DK6+80mn3JlCpVLj77rsxY8YMWvE2Ozu705/DhZeXF4YOHYpnn30WUqkU77//PjZv3typGbuOHj2K999/H/Pnz8fIkSOhUChoECOpFSSVSmG1ttbH2L17N9atW/ePUFjY4AqmXC+Gn58fBg0aBJVKhaamJuzevdsmTScRkokA25bHgwiJZrPZJuvQhQsX8PHHHyMrK4su5m5ubpBKpaisrMRvv/2GL774AqdPn6ZxA9y0nVzKFmkP1zIP/FEtHoBNsHtLSwvKy8uRnp6Ofv36YejQodi2bRstdsp+HgE7gxbXy0ys8KR97IB+UneEz+ejvr4eVqsV4eHhUCgUaGpqwpUrV+yMIW0pMuzNm3iOiIDMMAyioqJw9913IyAgAFu2bME333wDvV5PK5izx4v7ztraGNlZ5SwWC5RKJUaOHIkRI0YgIyMDP/30E62szW5rWwpQe78RECWLjIXV2pqiPCUlBffddx/c3NywcuVK/PTTTzT4lYxFW3EzXHC9PQaDgaaVbWxspEkORo8ejdTUVOh0OhiNRri5uQFoVXJ27tyJr7/+msZmsZXDO8HDwobFYkFZWRmd3wqFAnq9HoWFhTh69Cj27t2L48ePU6oTAVuQ7ijq6urg5eUFtVqNiooKFBcXo7m5GQaDAWvXrsWPP/6I4OBg+Pn5wd3dHUKhkFaAJ8oG2Z/9/f1t4mfKy8shEAhoxiYACAgIQHh4OBimNcNXZWWlw++B9IWkiyfHyJzLysrCiy++iODgYMoIaWxshFqtdkhBJvOWWPAVCgWkUinEYjFkMhlMJhMsFgu6du0Kq9WK06dP4/fff8eWLVuoB4M91iqVClKpFLW1tQ7XEBKoT54XGBiICRMmYObMmQCAFStW4LfffqNxLST+SKlUQigU2qTtJXOCzHmiyDkCiUkhsSpkXxCLxVAqlZQmSxI5uLm5wdvbm67FJP39tbwt5P04aguPx0NoaChSUlIwfvx4uLm54dSpU9i9ezeysrJs6GgWi4V6bdjjptFoKHWOz+fDx8cH0dHROH/+vJ28XVFRgbNnz1KKWVxcHE26QhJKXb16FQaDAaGhoYiNjYVCoUBZWRkKCwuRn5+PyspKlJeXd7ps21HcUqWlT58+8PLyAo/HQ1VVlV1A+c2CUL++++47HD16FElJSejSpQuEQiGuXr2Kb7/99oZTAN9IW/Ly8lBeXk45gdfSwm8EPB4PkZGRsFqtkMvlGDZsGBITEyGRSLBx40YcPHiwU54jlUoxd+5cBAUFwWw2Y+XKlZ2exYvH4yEsLAyPPfYYamtrsXr1ahqncCshk8kwY8YMPProo5DL5XjrrbewatWqTlOoCSwWC3bs2IHs7GxUVlaiX79+cHV1hcFgQGVlJQ1cNhgM2LhxI37//febpqX9ncClA5G/iVeAYRhK34uLiwOPx8OlS5dw6NAhAK18Z7LYstPekg2c6wUhz2QrFERwKSwsxPfff4/Tp0+jV69eiIyMhEgkQmZmJvbu3Yvi4mKb9nH74Sg2gut1IW1zNAbEM6HX65Gfnw+dToeYmBjq+eHGkrTlGSDpXtmpbcn9pVIptT4SHj2Px4NCocCgQYPQq1cvNDU14eDBgw4LdrZFLWLTt4hgTM6Ty+UYMmQIEhISoNPp8MMPP6CgoMDG20Q2dO4744J7nIwD8VhOmTIFVqsVP/74I86cOWMjTDhSYB0d557DVkbJuyMetYCAAMyePRuzZs2CTCbD0qVLsW7dOlRVVdHMYux55+j5XLDnJHl+bW0t9u3bh9raWuTn56N///60vgipJcLn81FZWYkjR47g0KFDyM3NpbE0bAMA6cOdEowPtL6P0tJS7N27F+7u7nB3d8f27dtpcDYpBqnX62m8FIlnaC+jIxfEOysSieDq6kqzGAKt9HCdTodLly4hOzsbYrEYGo0GKpWK1nQh85x4FQFQA0hLSwvN6ETu6erqitjYWBiNRqSlpaGsrIxa2LnUV/ZcZ1MdgVZl69SpUzh+/DhkMhkY5o/CmSQjWlVVlU1ficEDaE2Q4+fnh8DAQPTp04fKBBUVFVi/fj3S09Nx+PBh1NfXQy6X23hMgdb9VqlUQqfT2SkthE7v4eFBqX6JiYm466674OLigi1btlAvi7u7O5qamtDc3AwPDw+4uLjAaDRSpYXU1RKLxdQzTTzLbUGpVFK6MIk3lUqlcHNzg0gkgl6vp+mehUIhPDw8oFQqUVtbSzOYcZWmtsD95v38/BAfH4+7774b3bt3R1VVFX788UccOHAAeXl5HZqbjY2NaGpqom0ha3R0dDRMJpNdAXCDwUDjuvLy8lBfX0+/A6K0SKVSxMfHw8vLC25ubmhpaYGHhwdd9yorK3Hu3Lk7z9MikUjw5JNPIiAgAAzDYN++fbeMo19fX4+zZ8/i7Nmzt+T+HUVdXd1NZ/G6FhiGwalTp9CrVy+oVCpMnz4dhYWFWLduXafS0UjqTqFQiPXr1+PEiROdrkyQhAIhISHYtGkTdu7c2emKgyM8/fTTmDVrFry9vfHee+/hhx9+uKXPLSoqwltvvYXY2FiaxYVYB0mq0lOnTt1RgkRHwRVSudQpEixO0g3v3bsX2dnZ1DVOvje2UEYUC3YwN7EmEYGabOrs6wwGA86cOYOsrCxaxbmxsREmk8lGeOVaztjKCWC74XM9HGzvBzdTGoltIYXmSJ8B+/S7bYHdJnZ72Iog0Lp55efnIz4+Hj4+Ppg0aRLKysqwfv16WkvBkaWSW5OG3XfyLknlcFJvwMfHByUlJTh8+DBOnz5N28ZWUogwxu6bI4+ZIzqei4sLRo0ahbi4OGzbtg07duywS6/KRVveOEfjyX4HZDwUCgXGjh2LRx99FBqNBl9++SXWrVuHyspKu+B97jy51vO4/eXz+WhqakJ6ejqqqqpw4MABqNVqmM1mWqRZJBKhqqoKxcXFVBBh8/PZc5dQL+8ktLS0ICcnB1988QUUCgUqKirQ1NQEjUYDs9kMg8FABWYyP9ujypGAbqVSifLycpjNZkRHR8NgMODw4cM4d+4cdDodLdpKPFpkzhHFQaFQwMvLC35+frSdKpUKcrmcKpN6vR4lJSW0qC0705uHhwf8/f1tCu9y4/W4kEgkkMvllAJF1i8ANvucVCpFv3790L17d/znP/+xu4/FYoFer6dZ07y9vTFixAhERESAx+Nh/fr12LlzJ9LT02mhS0fJa7RaLfR6vcPgdg8PD3h5eUEsFtO1kNS8uXjxIi2OSmpnmc1mWCwWiMVi1NfXU/onoUnJZDK0tLTQ7G/XyuzV0tJClR1Sl4p4jomCR8auqqoKkZGR6N69O+rr6+Hh4dEhxYK9HwCtylVCQgJmzZqFIUOGwMfHB7/99htWrVpF40g6CpIcgD2XSeFks9kMrVaLkpISOl9IyZHq6mr4+PhALpfTUAqNRoOAgAAEBQXROcMuskr+I0WH/yyv7S1TWry8vNCnTx+aYWfNmjW36lF/GeTn5+Po0aNQKpW37KUSTvgnn3wChUKB2tpafPbZZ9i4cWOnUsMaGxuxbNkyHDp0CEeOHLGzxNwsSPD9gw8+iNLSUmzevNlhga1bAV9fX4SGhiInJ4daMG81SktLUVpaesuf83cDV2khQc3kmLu7O7p16wapVIqqqirs378fDQ0N1KrPFaiJIMAVgNnWf7bwzn4+EQZIUUlileZ6Usg13LgL9r3I+VzaGPtZbXkxSktLceTIERgMBpoul3tf9r+57SD34tY7YdcCKCsrw9GjRxEbG4vAwEAUFxfjxx9/xObNm1FSUmKTyIT9LEdeJq6gzRbudTodFWwyMzPpN8Cl8bHjEa8F9hjw+Xx07doVkydPRmNjI3bv3k2TCbTntWmrH2xw/2Z7TlxcXODj44Pq6mqkpaVh06ZNqKyspMIqN/jeEf2Rm2yAey4ZQxKvZDKZUFhYSIuaEiGW8PDZWYDINVyBicz/OxFWqxVVVVU2NdtIhiX2GBDBty2BVqlUwt/fH0FBQXBzc8POnTtRV1cHT09PZGZm4rfffqPV7dVqNRobG+1oVg0NDVRRcnd3h4+PD/UkECVEr9dTobCqqgpFRUU2gn1jYyOlQrEVYWKMIcHX7GcT7z079qIteHh4YOTIkRg4cCBVWkQiEeRyOaVNEXoYaQsR0okiqFQq0bVrV1RWVuLChQsO5xa7SC8XxLtEUhCTuiHnzp1DUVERLl26hKamJmo8YiddYdMdrdbWQqskYUFH5nhLSwvq6urod0biJy0WCxobG1FfX09pYyQuidS+KSgogFKpvKF44j59+uCFF17A0KFDIRaL8d133+HDDz/E+fPnr/teBOw5IBQKoVKp4OnpCYVCQfcash7U1dWhqKgIarUaYrGYKtBsBa2urg7Nzc20jo3RaIRer6dzgHh2/gzcMqVFq9Vi9erVeOKJJ3DhwoXbEqfwZ4OkVV6/fj3OnTt3Szh/VqsVW7ZsQUVFBcRiMXQ6HU6ePHnTxSW5YBgG6enpnRKU7gg8Xmve9sDAQBw5cgR79+69Jc9xhA8++AC7du2CVqtFbm7ubXuuE7bgCo5k0SQFu0gK3TNnzsDHxweXLl1CQUEBPZed3YYE77M3Nfb9HdUX4SbuIDVRiMDPpVm1RQ/raB+5Sk1bQnJJSQn1bl66dMmm7ez4FC4FjXhGSJvZygARfolC0dDQgG3btqG4uBhKpRKXLl1CXl4e5VATxY9cQ8aC2x9H7Sdjy+O1Vos+duwYeDyeTbE9InixPQDccSVtZr8vrpJKar9IpVLqyWGf68gLxm1rR2JO2IqB1WpFXV0d9u7di8uXL6OkpATZ2dk2fSJzmX1frmLnaNy492AroWTOsGv3MAxDMxSx70M8LOyxI/f7syyktwvk+yCWey4cfb8SiQQajYYqFr6+vvD29oZaraYKwPnz5/HTTz9ROhChaKvValRXV9uk1SW/m81m1NTUoKCgAGq1GhqNBhqNhtLWiAfBYDDAzc0NPB6PFqfW6/VIS0tDRkaGDW2wpaUFEokE7u7u1FvATmBBPA2OPKJsEGWJgMTHEqXKYDBQZcVoNEKn06GwsBDdu3eHQCBASkoK3N3dkZeXh/379+PcuXN2zyACc1tKhF6vR0hICBISEhAZGQkvLy/w+XyUlZVBq9Xa1MBi95MLkjjlelFTU2NDISNKm8ViQX19PV17ZTIZqqqqcOLECWRlZaGsrIwqU9cDuVyOpKQkdOvWDVqtFjt27MDbb79N97WbBdm7GhsbodPp0NDQQONciGeFbRwTCoVwd3cHn89HbW0tNfiUlpaivr4evr6+6NatGy2qWldXB5lMhoCAgOuOCess3DKlpaGhAf/73/+we/du1NfX3zYr+p8JUtX1Vscl1NTU0OJUf1eQyuRjxoyBVqu9LVXvCYqKiuxSRDrx54EtgLGLSlqtVpSVlWHZsmXYuXMn6uvrUVhYCACU+sK+nnheiEBPAkLZgiuX1sW2LgF/CDTc/zui7ZD/uO3gCsbkucAfQa7cTGNsodRgMCArKws5OTm04CLZrNlCPteTROhWbXmC2LBarSgqKkJJSQm9llhzHaUgZhdRY7ebtIddr4UcJ8dIymbyTkkbuQomeRfsmA6uIkSeJ5FIqAB/+PBhPPPMM8jJycHVq1chEons6Fjs8eeiLUGe3Sfyvsj8MRqNOHnyJDIyMmzq2nCVOkeeqPaUBvZ8a6t9pB3sectWThwp1+y232nUMC7YFJ+2xprH40Eul9P/FAoFre8hk8lorCG7cOXq1attyjXweDxoNBq4uLhQqh4bIpEITU1NyMnJoUpLVFQUqqqqoFar0a1bNwBAZmYmAgICEBISgvz8fErzrq2tRVlZGc20p1AoqMIklUohlUodCvGkAOC1BMq6ujps3rwZPXv2BNAqmPv6+tJK9iQwncxtg8GAjIwMmpI3KCgIV69eRW5uLo31IR4Vkm2L0LgcKS08Hg8DBgzAxIkTER8fD6VSCZPJhMrKSqokESMBOwC9M8FOygL84RUiRRZNJhMUCgVCQ0NRU1OD119/HWfOnEF9fb1NkHxHQeJMvvnmG5SXl2Pr1q3XTP7D5/Ntagm19/0yTGua6bq6Opw9exYVFRVQKpXo0aMHUlJS0LVrV0gkEhQXF+PcuXPQarVgGIau78XFxaipqUFtbS30ej08PDxgtVoREhJC1xA/Pz/06tXLLj7zdoHHdHDU2+MHO+GEE7cXf1dr6Y2uI+x6I+Q+1xqDjpzzd8Dt6kd7z+F6OTpyLVeJ68x+cO/tqF5NZzzjTpg/7eHv2L/OkkWIN5Uo1OQ/Lh1Ur9e3OfelUiml7nEVCO4cJXF4QqEQAoGAWr2JpZ8kyiCUI7VaDavVSq3lgG3sHjHc3IwwT4TQq1ev0vZxBXk2PDw84OvrS/9ubGxEY2MjpcOxv3W2l6WteaZSqaDRaCCRSKiiZbFY6DgQj3tH6px0JtgUTvKuiAJ2s2nDCf3ObDZ3OAb6Wp5gAoFAABcXF6pkknTnLi4uNIsbn8+nWe9I7AqbDkwURHbfSQFgkrxBrVbj8uXLf4rn1qm0OOHE3xB/R2EDcK4jTjjxV8LfcR1xriFOOPHXwe1eQ/68spZOOOGEE/8PtiWJzb9n04UcndtRsD0FUqkUDMPYBYcKhUKIxWKaD59hGFpLhE3JaSsTFDt+hGTDYVtGuZQNtnfAEfebWIHJORKJBAKBwGHwJ3ecOpLalvSXBHGSmkEmk8kunoc8QyQS2RRMvBbVqC2KFLuPbdGq2kqxzD3GTmFLLN5isZjSx8g5jt4b917cIHl2Wm0yZuzYFnIN2+LOztTjKKYE+KMYILFuEhALOpefzx570mb2+LNjnUjMC5dmx471+DsqK45AaF4ajQZSqRQmkwnV1dXtZoNUKBSIioqiXH2RSITi4mIcO3YMJSUlbV7n6ekJkUhEqWOO6o6IRCJEREQgKiqKxgmQmBBi5a6qqmozsQ07kcPDDz+M7t2707ozx44dg8FgQI8ePRAbGwsej4crV67g4sWLSE9Pp3GtPj4+6NatG3g8HsrLy5Gfn0+/cR6Ph5SUFCiVSmzduhUxMTGYOXMmXnjhBbu2iEQiWtySeF+IBZ78nx1fRyimKpUKIpEIBoMBPj4+6NOnD5qbm7F9+3a7GJDQ0FBK+XUEjUaD6OhouLi4oKCggNY/6ihIUc1Lly51qGA1yRpHiic2Nzejqamp3XgZtVoNb29vShX09fVFZWUlrQ8oFAoRFBSEfv36wcfHBxkZGdi3b59NPzw9PREcHIzAwEAYjUbU1tbaUGybmppQV1eHkpISWCwWhIaGorKy0m484+LisGvXLly9epXGkl+5cgVarRZarRZ1dXUQiURobGykYykUCik9jMwToVBIa5W5uroiPDwcsbGxCAsLQ1NTE06cOIHvvvvuT0nq4VRanHDCidsOtrDIFbDZwl1bgnF7QhcR/EiaX/I8Am62HTbNgq0QsAVndvu4z3LUbu4mx6W1kYKSbCWNrZyQAGs22EoWj9eakrOpqclO+G3LEs0VyElgPfmb3XfCcybpdskz2GPA/jcRwtl9JMIxVylgv+9rxZhw37OjLGNWq9VOSOUWEnQER0oo93nc5ANtZSZiBwy3FaBKBCESI9BWkDhbKSRKBjlOkk2oVCrU19fDy8sLlZWVtEAeW5AkYPftTsscRqgvxLhQV1fXrsJChNLm5maUl5fTeKSKiooO0X7IcwwGA01pzJ5/ZrMZZWVlkEql8PHxoRndvLy84OrqCqCVHgTATnEhChERRF1cXFBVVUXrJplMJgQFBcHV1ZUWmywtLcXly5dtEvGQrIMkcJpAo9EgMTERPXv2xP79++Hm5obHH38c4eHhbY4VmUdkDpI5yU6bzTbGkG+XpDAn/c7IyGizOG17qKurQ0FBAYKCguyq1HcEdXV1qKqq6jClSyAQ2KzNpK5PW1RRoVBIUwWHhYXBx8cHrq6uNuU3/P39ER0dTWlw4eHhqKmpoUmOiEHCYrHQ+D+Scp8ojFarFUqlEmq1GrW1tSgqKoKrq6ud0kIC60nbJRIJXFxcUFxcDIFAAI1GQ8suWCwWFBUVOVzTSFIGPz8/Ou56vZ7GWJI02H8GnEqLE044cdvBFTod/UY2TK4wTP4mKTlNJhNNUUwEa7LJsp/RlgLEztjEzprFzj7EzmLVlsLF3dS4nhKy4ROrJbGIcwUAdspctqeC7WHiZg/jjp8jQZ30iz3mjlL/ssec1LRgJxBg94/8TZQf9hg5akNHlCtum9ljw24bOcZOVMAeY/J7W/VeuPdit5H777aEFm6KY3b/2M8lwcSOlC52ggKiQJNryLwlzyceMPI8UqSWBAy3tLTYpD7mtpPH4900J/+vBLFYTL8fo9F4zQxSpLq72WxGTk4OcnJy4OLigqampmtm4CQeBOLJIRXaxWIxJBIJVRDq6upQWFhIiwiLxWJ4e3tDp9NBq9WirKzMLmhfKBTC1dWVxrAAwNmzZ1FdXY2srCwIBAIkJCQgNjYWarWatpUYFtiwWCyora2lno/g4GCaZCAoKAharRZXrlzBuHHjMHXq1DY9HSaTCY2NjTbGH5FIROcie21kG26MRiO14PN4rYXFL1++bLOmkfnfkYQ41dXV1BB1vbBYLKisrKQGHxcXF5oGmwu2J5KdgbE9kKQN7JTU9fX1NrEqCoUCHh4eEAgEaGpqglwuR3h4OM6ePWuzzspkMgB/eLjMZjOMRiMaGhpofRSucYW9BgOtMUZpaWloaGiAQCCAm5sbTCYT6urqUF1dDYlEAldXV7i5uaGhoQF1dXV2xTF5PB7EYjFcXFzg7e0Ni8WCmpoaFBYWoqmpCUajEUVFRX9aQo9brrQIhUKb6q8EXLe4EzcOdrrXmx1Tssm3RYG5VWALb8558c8ElyJGNgFCAfH19aUFs5qammC1WummStYTRzQs9j3ZzyLzmwQfkrlPrKns80jbuBSpa3l8uH9zzyc0Nfa5bIoRsVhKpVLweDxq0b1WLQJH7SL3dORhEIvFcHd3R2RkJKqqqpCdnU3b0VZaYjLGjtIvO8psxVaAHGVPI+c6UhrYfzv6XSQSoaWlBX5+fggODqZZmNqaD+zxaA/keWwFmt0GolQQsPtBLLZEiGsraJq93pL/c7017OcB9l43hmHsrK5szxpRiu6kzGE8Ho9SJklRwrb2DmIdJkaOxsZGWiTvypUrHRKKIyMjUV1dTSk6ZM9ydXWFRqMBABvFhfw7KCiIFl7U6XR2Cgufz4dSqYRAILChMBUVFaGyspKuNQ0NDaipqYFer6dZrkifybfl5+eH0NBQ6HQ6lJeXQ61Ww9/fnwZ+19XVQa/XQ6lUIjk5GXw+HydOnHDYX6vV6pCK6kh2I3IeUYqJ0H/lyhXodDqbuRkREUFLDSgUCpsAfjZ4PJ5NjZumpibodLrrztplNBrpvO/Tpw/NMuiIKsqmurG9xY5AsnoRxaGurg5arRaNjY1UGYuIiEB8fDw0Gg0YhqE1xmQyGbp27YqLFy+CYRiUlZXBxcUFarWaKqIGg4EWdiQFJNljJRKJEBwcbJMuub6+Hnv27IFarYaPjw9VtMlaRN6pQCCAVCqFTCazU1rEYjE8PDxoHaCWlhbU19ejsrISFRUVKCsrw+XLl+88pYXH48HHxwdvvPEG5syZAx6Ph9WrV9ONtqysDN9++y2uXLnS6UKqTCajvL3OvjeXYsEFUSCulZqusyCVSjFjxgy8+OKL+OKLL/DZZ5/ZFbrqCMRiMTQaDXr06IHo6GgcP34cubm51GrXVnGozkJ8fDyeffZZ7N+/HytXrrytHwSfz4dcLqcCxo2M3/WCWFYEAgEdW7LY/xPAtXSzPQrkb2Ldj4uLw/z585GUlASj0UgpIDqdDtnZ2Th48CBycnLogt6WVZz7m6NnkirIVqsVUVFRcHV1xYULF+imy/VikDay+0XuRwQb9vPJ+UTIIFQhduxBS0sL5HI5lEoldcmTOeKILsVWRtgWT27/2HQH9r1EIhFkMhmSkpIwfvx4WpW5rKzMxnLd1jfJXg9VKhWio6MRFhYGkUiEyspK5Ofno6ysjK4l3FgVttWQLbg76q+j90oUO6FQiLCwMIwePRqnT5/Gjh072iz+1lGhh63YktgZ8s7YzxcIBPDx8UFkZCStzl5VVUXpJo5SMLMVFQI2zaa9uCQyX0QiEWJjY2G1WnH16lUYDAYbihm7kjq5552iuJD+kwxW7LpN7HM8PT0RFhYGPz8/GudRXFwMrVZL11v2mLWF6OhopKenU4NGS0sLxGIxXF1daW0pvV5v40kkawop4ldfX+8wjspgMNgJjw0NDTZesYyMDBQXF9MsTsAfRU579+6N+Ph4jB07Fu7u7ti+fTt2795NhW4SCyUQCCCRSODv7w+dTofly5fjs88+6/CYt7U/ke+PYf6guZrNZhQXF9O4GKlUCjc3N3h6eiI3Nxc8Hg+JiYmoqanB+fPnbcZEIBAgJiYG/fr1o7EVDMMgLy8PBw4cQGlpaYdlO/Z8JzQtq9XqMH7JkSe0vfsSjz5JeV1fX4/Lly+jsrISABAbG4vk5GTU19ejtLQUOp2OvmeS0YtkJMvJyYHZbEZMTAxV+Kqrq6kHkTAKCCQSCfz8/NDU1ITy8nJquMjMzES3bt0oPZAYswnVt6qqiqaW5s45oJW+GBgYCIVCAb1ej5qaGpSWlqKqqorO1dtp0Obilikt7u7u+OqrrzB8+HBKkbjvvvvoJt7S0oIxY8ZgyZIlOHLkSKc++4UXXsBdd92FefPmITMzs9MGWCKRICIiAq6ursjMzLSzmPB4PCQnJ2PgwIHYsmULMjMzb7nXQKVSoXv37ggODsaIESOwZ8+eGyoImZiYiA8//BB9+vShAs/hw4dRVFSE8+fP47vvvqNBaYTj3pnw8fHBoEGDAAD79u27rUUfQ0NDsWLFCsTHx2P58uVYunQpqqurO31zJ1YZpVKJlJQUPP300wj9P/b+OzzK68z/x18zmq4p6r0DEgKBKEIS1XRMN7bBccGO7Wzikk2cst/NZrO5sptrd+P9JOtsnLKJHcfGBTu2AZteTAeBKBKghhDqvZeRNNJImt8f+p3jZ0YjOnbs9X1dXIA085TznOecu7zf7zs+ng8//BCHw0FWVhZ79uwZMa++jCYcKM+sudJJBIiKiuK73/0uGzdulIGMyHxrNBrsdjtHjhzht7/9LceOHZNZJPAOO1M6w8osvNislLCA+fPnk56ezu9+9zs3jPJocLbRghdh4p4CAgKYNWsWRqORrKwsmekVsCZfX1/mz5/P1KlTOXXqFCdPnnRzvj0DHs9ze7s2b9cBn1ZdxDmXL1+OXq/n7NmztLa20tfXJzN0nvetrO66XC4MBgPTpk3jG9/4BsuWLSMgIICWlhauXLlCbm4u58+f59KlS5SXl2O3292cTCXcZLQxHG3NUVbZtFotM2bMIDIykry8PLemj7drSqiXuH/xHFQqFTExMXzzm9/Ez8+PgwcPsm3bNq5eveq1Z4gS6nUjwaAS9y7gSMHBwcydO5dly5Zhs9moqamhrq6OmpoaDh8+THFxsVuQLapCIvD6otvQ0JDs5eFpIhHl7+9PUlISixYtYsqUKahUKi5cuMDOnTslhOpG9zI/Pz+3NUtUQK1Wq0z6Wa1W2tvb0el0xMTEuDnwdrtdVkWUwchoFbju7m63z7lcLtra2uQ8V6vVMpPu5+fH7NmzWbZsGQ6Hg5KSEs6fP09LSwsOh0M2zzSZTLS1tVFbW8s777wjoVt3wkRgLLg5QjJXVGwjIiLc5JJdLhenT5+WyQphPj4+pKWl8cQTTxAfHy8d94iICJYtW8bcuXN56623yM7Ols9eWQn3rCIrra2tjcDAQJKTk2ltbR3BfxJQOCWc8lrzo6enB7vdjsViQaVSSfln8dw0Go2sWjQ1NdHe3k5BQYF8JkajUe73g4ODdHd3Exsbi8FgoL6+nvr6+lHfVeVecezYMaqrq3E4HJSWlhIUFERzc7OEeokKX09Pj+RuNjU1jbh/o9FIYmIi4eHhuFwuGhoaKCkpobm5+W9mzbhrQYvAzYmy0uXLl4mOjiYgIAB/f38sFgvjx49n7dq1nD179pYy+WKzValUbnjUqKgoEhMT5cZ1pwKHjIwM3njjDSwWCw888ABHjhxx+72fnx+PPfYYjzzyCElJSfzoRz+SWtZ3w9RqNdHR0WRmZt7WcXx8fIiLi2Ps2LHApwvAnDlzmDNnDgDz58+nr6+PrKws9u7dS3l5+TX162/WmpubqaioYNGiRWzfvv0zDVoCAgKYO3cuvb29PProo2g0Gl5++WWqqqru6DnmzZtHcHAwq1atYt68eVgsFgD+7u/+DoCHHnqIF198kf/5n/+5Y+f9otho70h4eDghISHU19dTV1cniYNxcXFMnjwZi8XCjBkzSE9P5+zZs6Nm1r1Bs2B4k9JqtYSHh8tstfhcZmYmq1at4q9//avbd5TO9WiwJsFNEBu3cg1KSUnh2WefxWKxMDg46AZPcTqd+Pv7s2LFClavXk1YWBj19fUUFxfLTUMZQCidfs/A5VpVJ2HCEY6JiSEjIwN/f3+sVqtU21Lye0b7rgj49Ho9iYmJzJgxg4CAAAYHB/H392fmzJlkZmbS3t4uVW1EMFZcXExXV9dNQ5eUEDdhAns9MDBARkYGkydPprS0dESjytE4Lp6cKXBXHVNmkcX9KlXWdDodGRkZEtJitVp59dVXqaysHBFIC76B59wQ1yI+q3yGAv42bdo0xo0bx6RJk1i4cCGRkZHye8JhHRoakuRt5b192Sq5o/kNAjpmtVpJTU1l48aNREREALB48WKmTZvGRx99xPbt20dUFEeztrY2enp6JHxU8BlcLpdM5Pn5+aHRaEhLS2PKlCns2rWL+vp67HY7Q0NDzJo1i9raWrKysq57PiXhXSQ1xD35+/vL9zIoKIiUlBRCQkJoaGiQiUa1Wo3FYnHjQtTX13Pp0iUaGhqorKwkPDyc5ORkr93sb8bEOyAUCcPCwpg0aRL9/f0y6WKz2VCpVG4QOMHbUFp8fDxPPfUU69evx8fHh/r6epqbm3G5XAQGBnL//feTkJDAH/7wB1lNFe+igDyJjvCea8rFixcJDw9Hq9USFxdHcXGxW3VT2aBXVI481wSlOZ1OWlpaZDVVmQADJJzKZrNJyJcQgBAm3k21Wk1MTAyrV68mKiqKGTNm8O6773LgwAHa29tHXENHRwf9/f0kJCTg7+/P73//e5xOJ1VVVZhMJikWoVKp3GCEoumpMjkqEqrBwcFYrVZ5rpsVMvgs7K4FLe3t7fy///f/SE5OZvPmzVRWVqLRaLjnnnv4x3/8R+bOncvAwADd3d237NSbTCYWLVqEn58f7777rlzAIiMjJR7vTpqfnx8xMTF0d3czY8YMjh075vZSKNU2MjIyuOeee3jvvffuGrTKYrGwaNEioqOj6evro7Gx8YabFSnNarUyYcIEAgMD3X4+MDAg4QZLlixBrVazevVqHnvsMd577z22b99OXl7eHQlcmpqaKCwsZNq0aSQmJmI2m2/pXm7Vent7OXr0KH19fSxfvpwLFy7wwQcf3JFnZ7PZ+P73v8/3vvc92aSpra2N+vp6ent7CQkJkRWY9PR0/P39aWtruwN39cW35uZmNm/ezLvvvktFRQW1tbXAcGXw6aefZsmSJRLaIQIEkQ1XQmGEc6k0sRnp9XpSU1OxWCxs376dzs5ONBoNkZGRBAQEyM97k8291mYGI9WkfHx8iImJYd68eQCMGTNGyvMKJ1aovISEhJCens6FCxek3KUnnMjzfrwFKaORzsXmajKZmDNnDhMmTKC3t5eqqirq6uokcdyTfyL+raw4wHCWLiEhgeDgYPr7+yUZVODpNRoN48ePZ+LEiaxevZqCggK2b9/Otm3bJBRVeZ2jcevERux5j0NDQzQ1NXH16lXmzp3LrFmzOHr0KA0NDW6Z2Bvl7IlnoQwQld8RGXIleddsNlNSUkJdXR3r16+nrq6Ot956C4fD4QYFVV7Ltfg2yvGOiorimWeeYd26dSQkJNDd3U17ezsXL14EwGw2YzKZiIiIYO7cuRw/fpzLly/LMbvW3PmymcDhKxMDTqdTdmpfuHAh8fHxTJ48mddff53s7OzrBnTV1dX09vZisVgkh0OQ3kWGHYaf0+rVq4mIiGDnzp0Shvbwww/zne98h3feeYczZ85c93xBQUGykqfT6dBqtdhsNqxWq2w4OXPmTBYsWEB8fDxOp5Ps7GyOHDkiK4xarZa2tjYpaVtVVSWhS8HBwaSmphIYGHjbQYtYb10uFyaTiRUrVrBy5UrOnDnDlStXJA+lubnZLcEcHR1NQ0ODm/peYmIiixcvlkm9MWPGEB4eTktLixQxSEhI4Nlnn0Wr1bJz5076+/tlAAlIDklHR4fbO1tZWUlDQ4NUcPPz85NNGIXUvXjnxL+v95wcDoeEv+r1enkfLpeLmpoaLl++zMSJE2UQ6rkXKdEFLS0tMvAzm8388Ic/JDo6mtdff53W1la384rAQwRNwjo7O8nJyZHBqtgXLRYLFosFo9FIT08PHR0dUm1s7NixaLVa2tvb5fzw9/e/Lnrg87C7FrQ0NTXx4Ycfuv3M6XRSUFBARESEzGqeOHHilstOQ0NDREdH88ADD1BUVDQqoexOm1arJS0tbYSefnNzMy0tLQwNDRETE0Nqaipbt269a0FLWFgYDz74IGFhYRQWFvLXv/71lioUISEhTJ482e1nXV1dnD9/njNnzuBwOMjIyGDSpElYLBYmTpzIv/zLvzB79mx+9atfcf78+dt2squrq9m9ezfz589n3bp1nDhxYoSW+d20oaEhcnNz2b9/P//5n//Jhg0bOHv2rNz0b8fWrFnDN77xDTQaDfn5+Zw7d46zZ89K52bevHn4+/uzevVqySvYvHnz30w59rM2ZSZcaMyLDQiGncmoqChJsKyvryc/P5/u7m4ZtHhyTbwRKpWOeFRUFIsXL6ahoYHDhw9jMpncMmZKrLO3hXs0rptSGUpcl3iuKpVKOprCCdFqtdjtdmpra+nt7SU6Opro6Gi3a1H2dxGOk+e1eLtGT1K6qOwkJyezbNkyzGYzeXl5nDhxgrq6OsAdCud5XHEcMbZWq1UmHHx8fPD396epqYlTp05x8eJFVCoVSUlJpKSkEBkZSVpaGikpKaSnp/PnP/+ZQ4cO0dnZOapSm/i/uC5vwVl7ezs5OTmsXLmSKVOmkJCQQFNTk3wGo30Xrg+l8zSRIRaBTV9fH/39/Rw+fJhDhw7xk5/8hAULFnDo0CFKSkpk1Uqp3OYNuubtvkTCb8OGDQQHB1NUVMTBgwfJzc2lvLwcl8tFQEAAkyZNYsOGDUybNo0ZM2ZQXV0tYUnCvky8lmuZ0+mkvLycnTt3UlZWhslkIjg4mPT0dBYvXsyYMWMICwuTPBclodmbNTU1MTg4iNlsRqPRyN4ZXV1dkijt4+NDYGAgY8aMkSiQ/v5+xo4dyz/8wz8QFxcnnWZvMGAlj0n0yujt7ZXiI6KKYDAYWLt2rQyOWltbyc3NJTc3l8LCQjo7O/Hz86Ozs5Ouri6Sk5NJSUlx26cTEhKIiYm5rmraaOZtHhkMBhYvXszTTz9NREQEBQUFWCwWqYIlRARgeF6npqZSWFhIcXExMJz4mDhxIjExMdKfamxsJCcnh+LiYhwOh+TejhkzRsKGd+7cSWdnp5T+tlgs6PV6N9EUQCpmCQsMDGTKlCloNBrq6+tl75PRIHujWV9fHw0NDRiNRrfvNTQ0cOXKFSIjI1GpVFitVvz8/GhoaPB6nLq6OrZt28by5cv57W9/y8SJE5kxYwb5+fkcPnzYzZcUcMH8/Hy5jogkkeczFXuPgOUODQ1J6OTq1auZMWMGDoeD3NxcKisrsVgssu9MX18fTU1NGI1GHA7HqJDMz8o+U8lj4ZBFR0czMDBAaWnpbUX4vb29VFZWkpSUxMaNG6murqarqwudTgfc+c65SUlJwPCColT1UJoyo9fW1nbXnG5fX1/Wr1/PtGnTGBgYIDc3lzNnztzSsbRarcw81NbW0tnZSXZ2Nq+++irZ2dk4nU5iY2NZsWIFEydOZNWqVURHR7No0SIiIiJ49dVXeeedd0ZkAm7GBHm1qKiIpUuXMnnyZE6dOvWZVVtEpaOyspLi4mIWLFhAeHj4HQlali5disViIScnh3//93/nwIEDbljSkpISVq9ejV6vJzw8nHHjxnltMPdlMk+8vtJRUyoyDQwMyE1GrVYTFBREdHQ0q1atYurUqTgcDvLy8sjPz5dOgycBXRxTmCd5XUhCCriqgAwI+Vghq+oprqEMrsQ5lfcnslwC5y0qF8HBwdLR1Wq1GAwGuRGI4ynL+Q6HQ2YqRSMwb9m/6zniQt1HWR0ICAhg/fr1TJkyhY6ODo4cOUJWVtYIJSpvpnx+AwMDWCwWYmNjcTqdNDU1UVtby8GDB9m+fTs5OTk4HA5iY2PJyMhgxYoVzJkzh+joaJYsWUJkZCRBQUFs2bJFPm/PwFN5T94CUEE0zc3NpaCggGnTpjF79mwKCgrcHJWbDViUAas3crwYS7VaLZV/amtrKS0tlfcl5pTy+56y3GIslQGbmKtms5mpU6cSGhpKXl4ev/jFL8jOzpYSpE6nE7PZLCu50dHRxMXFyUaIntf8f8muXLkik3n+/v7SKVu1ahVGo5E5c+YQHx9/3aCls7NTku+FEy7GVMkvExl6GJ4zNpuNZ555hqSkJHbs2EFZWRn+/v5eVbCMRiNqtZquri7ZCNDhcNDc3MzAwAAGgwGdTsfUqVNZv349QUFB9PX1uQWxjY2Nct1oa2sjPDychx56iOTkZC5duiTnmNFoZGBgYFQH+nrmCbM0mUwsX76c5557jqSkJHJzczl37hzV1dVSkADc4bUOh8Pt/6GhoUydOpXBwUGqq6v58MMP2bt3LwUFBdJR1ul0ZGZmsmLFCpKTk3nsscfQarVs27ZNBkVCAtwzKeC5borxeeKJJ+jv7yc7O5tPPvmEkpISt+d7IybULJXW3d1NaWkpY8eOxWAwYDAYsNlsNDY2el1zurq6OHDgAFOnTiUmJoaLFy8ybtw4VqxYQV1dHfn5+W4+ZUdHh5TPF8/EYDCMCFqEGIQIyAICAli8eDGPP/4406dPlz2BfH19iYuLkwHz5MmTGTt2LE1NTeh0Oqqrqzl37hx1dXVeSfyfhd31oMVsNjN58mQCAwP57ne/S2ZmJmq1msuXL3Po0KFbfmGElZSUcO7cOdasWcPp06e5cOECvr6+ABJHeCcCh5CQEB544AHgUy1yT4uJiSEyMhK1Wk1fXx9nz569K46nRqNhyZIlPPvss/T393P27Fnef/99Wlpabul4gYGBjB8/HoBz586xdetWTp06xdWrV+VLXlFRwR/+8AeCgoI4f/48f/d3f0d6ejoTJ07k2WefpaGhgQ8//PC2NsampiaJ758/fz47duyQGZi7acIRnD17Nu3t7cTExEhM9J3ISooxFE3MlAGLSqUiISGB7373u6SkpHDlyhXOnz8/Kjfjy2Ke2XNv2WXR/0JIlcbGxjJ79mwyMzOZM2cOISEhXLx4kYMHD1JfX+8mrS6IzqNxF5T/Hxwc5NKlS5SUlLBkyRKOHz/O+fPnZWDl6+srM1We1+h5L8I8IUgi6Bk7dizLli1Dr9fT0tJCd3e3vH+xKcfHxzNu3DhMJhPl5eWy54OyG7sIlEZb27w51+KahLPh4+PD8uXLeeCBBzCbzZw+fZqPP/6Y2tpa6YiPxtkRxxaKRE6nE4vFQnBwMH19fVy4cIEtW7Zw4sQJampq6OvrQ6VSUVFRQWlpKSdOnGDZsmV8/etfZ8aMGXId6enp4aOPPnJTa7sZE9W58+fPk56ezpw5c9i7dy+dnZ239B6L+xPvsDdRAgHvi4yMxM/Pj/T0dDo7O4mJiSEmJoZJkyaRm5tLV1eX5CYoIWeeqm/iWSnnqmcvhtraWpl9F9eVlJTEunXriImJobS0lMuXL8vqo0r1aX+WL3vwotfrJY7fc/4IEnpzczN9fX0SPnUjSAHBvRIwP29NLAXxWavV0tnZicPhYPbs2axcuZLi4mL27dtHcHAwMIwuUD5XvV5PQEAAfX19ki8qVCZFhaWnp4fu7m50Op0MAgTs68SJE1RXV0slMwFlW7p0KbNmzUKn00l4WV9fHy0tLej1+ltGSSj7roSFhbFy5UqeeuopJk2aRH5+Pn/5y1/Ys2cPTU1NREVFYTKZ5HXB8FyvrKykublZHjM4OJhJkyYBUFNTwyeffMLFixflWLhcw0poR48elVCs2NhYNm7ciEajYcuWLdKRt9vtUu5XWZlWKp319fVx6dIl6urq2LBhA4sWLWL27Nl88sknHDlyhNLS0tuiMMDwc758+TKRkZFSZvtaVlZWxqVLl3jqqafIzc3FYrGgVqtJSEjgypUrbvOut7eX/v5+6Y+K8fE0wT8UMGqXy8XatWtZsmQJDoeDQ4cOkZWVxdWrV+no6MBkMpGSksLkyZOZN28eNpsNh8PBgQMHJNTSE3r3WdldC1p8fX1Zu3YtqampzJw5Ez8/P+kYixLYhx9+eNs3XVZWxs6dO0lLS2Pu3Lnk5eXJhzpu3Di3RmO3Y2vWrCE5ORkYzqScO3fO7fcajYZFixYxdepUfHx8OHbsGIWFhSMie6HqcKum0WhYvXo1//zP/yyJdx988AG7du26ZZKl2WyWqh4dHR0cPnyYiooKr59tbm7mzTffpKamhp/85CfMmjWLuLg4nn76aerr6zl+/PgtP9POzk7OnDnDsmXLWLhwIQsWLJAynnfLLBYLy5Ytw8fHh3HjxnH//ffLjMhTTz1FcXHxqM23btQ+/PBDFixYwKxZs/j2t79NVlYWtbW1nDhxAqfTyf3338/EiRMpLCzkj3/8I8eOHbtDd/fFNoGPXrZsGZMmTWLixIkkJyfLDbCgoIB33nmHQ4cO0dvbi06nc3MwPWFAnlls4Yir1WpKS0vZtWsXU6ZM4Z577qG8vBy9Xo/BYJBN30Zz2sW1el67OIdwGMUakZmZydDQEOXl5VRVVdHT0+MmRzp9+nQmT56MVqslLy+P3Nxcmb0V1SSj0Sg7oHszz2BbBDhKJ2nWrFl85zvfISEhgcbGRj766CPOnj0r78sbCd+TJySqBaLXhK+vL1qtlvr6erKysrhy5Yp8HsLJGBwcpLKyks2bN1NTU8MLL7zAokWLSE5O5vHHH6esrIysrKxRg9nR1hfh5Hd0dHD+/HlqamqYPHky06ZNo6SkRI7zzRzTM5ss7l84P0L9y2w2c8899+Dj40NycrLM9jocDsaMGUNISIjclzwDydGqSeJvl2u4v8PRo0dZuHAhkyZN4u///u+5ePEiBQUFEh6yevVqlixZQkNDA6+//jrHjh2TRHxxrNEEKb4sJjqUi+fj6+sre5tYrVYiIiKYOXMmKSkpqNVqWlpaOHLkyA3Bqnt7eyVvYrRkpM1mY8KECfj5+XH58mXMZjMPP/wwAwMDbN++HRjmvIj+UsK0Wi0hISGYTCbp1IoO6IL7K2BiHR0d7NmzhzFjxrB48WJCQkJISkoiMTFRSj+3t7cTGxvLmjVruO+++7BarTQ0NBAQECAz5w6Hg97eXjc+xM2aj48PmZmZPPjgg6xcuZLg4GBOnTrFa6+9xt69e2ltbZXJFiH7LAIvQMIXYfhdCAgIIDQ0VCY4qqqqMJvNUore4XDI5E5hYSHjx4/HarWiUql4+OGH6erqYv/+/XR0dEgpcqPRKCubohGkyWSS0LH29nY2bdpEV1cXCxcuZPHixUydOpW5c+eyd+9ejh07Rnl5+S0noLu7u7ly5YpU7Wptbb3mO2i326moqGDs2LH4+fmxe/dujh49Snl5udtaLBIoykrLaNLdvb29bpVdEZgAFBUV8c4773DixAm6urqkuEp1dTXt7e1Mnz6dadOmycaqQv3x81pH7lrQEhQUxPPPP09KSgomk0kunP39/Rw/fpzXX39d4qZvxwQc4MqVK2RkZLBr1y75OyFReCds9erV+Pr6MjQ0RGVlJYcOHXL7vdVqZdq0aURGRuJyudi1a5esfGg0Gmw2G/fddx9qtZpXXnnllq9j3Lhx/Nu//RvJycn09/dz5swZtm7d+pmqwjidTg4ePMjAwAA//elPpcrYsmXLyMrKuuVrcTgcHDt2jLy8POLj4xk/frxc5O6G6XQ61q5dy9NPPw0MB2TvvPMOubm5MruuVDq5Vdu/fz8/+9nPmD9/PitWrGDhwoW0t7dz/vx5urq6WLlyJU6nk+3bt/PWW299bmXXz9KUjqLSIQT3bu2RkZF885vfZPLkyRL+IuATFy9eJCsri+rqagAJ3xIwqNGIx0qOifhMT08P58+f58qVK0ybNo2cnBz5mcTERMaPHz8iy6VctL0FNEpowsDAAEFBQaSlpeHv78/g4CA1NTVumzZ8WpkODQ2lq6uL0tJS6uvrZWCQkZFBWloaNTU17N+/3w0Xr1Sh8Ra0KN/LxMREfvSjHzFlyhScTicnTpxg+/btshqghOiNtjnpdDq5GQKSr2Oz2YDhzLSoKog/ymP19PRw8uRJ7HY7Op2OefPmkZGRwbp16ygpKaGjo+O62HLPSpDIsJeUlFBZWcmiRYuYNm0a+/fvl0GL8p6UKmTeeDSePCUBAxLOq/gza9YsMjMzKSkpwdfXl5ycHE6dOkV4eLiUC1XCBcW5R0MCKO8HhufToUOHMBqNPPzww6Snp7NkyRLJGers7GTRokX09PTw3nvv8eGHH9LW1jaiGvdlDlgACdGLjY1l0aJFZGRkUFVVxblz5xgYGGDChAmsXr2a6OhofHx8KC8vJysr64Z4HcpEhLd3DIZ5pkLRraOjgwULFjB37lyOHj1Kfn4+VqtV9j1TPguDwYDJZKK9vV2+0zqdTp5jcHBQ9utwuVzk5OTwH//xH1y9epUNGzYwceJE/v7v/57du3ezdetWzGYzy5YtY9WqVcTGxlJRUSGdb2UiRaVSycaYN2tarZYnn3ySRx99lMmTJ+Nyufj444955ZVXyMnJkYGJgCbpdDrZ+0OY0vnV6XRSlMZut1NcXExvb6+EKwn5ZgGbBSgtLWXu3LlyjJYsWUJ5eTkXL16kv79fBiriez4+PvKdN5lMUuWwpaWFzZs3k5OTw8yZM8nIyGDBggVMmzaNOXPmsG3bNk6ePClVzG7WamtrpXrYjfBBurq6uHLlClqtlsrKSo4fP05raysqlQqbzYbBYJDNJpVJ+dGCFs9z9vX1cejQIZKSkmRgGB0dTXt7u4REiz3o1KlTMtA6deoUdXV1n6sK4V0LWgQx0Gw209/fT2trK6GhoWg0GlJTU5k+ffptZ7CFdXd309XVxfTp0/nXf/1XoqKiAFi5ciU+Pj4cPnyYLVu23NY5Zs6cKf+t1+uJj4+XUToMO0yCjCc6lY4fP54xY8aQmZkpSaHZ2dm3FbTcd999suIjqh63K81bW1tLdnY26enpN8wDcjqdHD16lBdffFGqe6xYsYKjR4+yb9++W76W6upq6uvrAZg6dSoWi+W2KlPXMpPJxMaNG6mvrycvL4/IyEguXbrEnj177uh5+vr6eO+99zh8+DBHjhxh48aNzJ07l4SEBAYHB9FqtZw4cYL33nvv/0TA4mnenEXhaPn6+mIwGNDr9VKaNDo6Gr1eT1xcHLGxseTl5Ul8uFKycrTzCKdDqfok/i3Wka997WsEBgbS09PD0qVLMRgMvPfee5w+ffqGGxYqpTRhGK+dkJAgs91Wq5WwsDBJthXcF5PJJCE94eHhzJkzB5fLxaRJk7j//vuJiIhg8+bNHDhwQJ5L2edFnFM4J56Osc1m49lnn2X+/Pmo1Wqqqqp45513pEMjPuutUuUpMaz8XFtbG01NTQQHB8uql3DuBWRPudkJ8vrp06f57W9/y4QJE/D392f+/Pns2bOHkydPjnAKPWFq3v7tcg13mC4oKGDWrFlMmDCByMhIGfwpK2DeVMiUv/c8h7geMW/EPvf444/T19fHsWPHmDRpEidPnmTPnj0SOuat0bE3NTpxDcp7EdfU3t7Otm3buHTpkpTjXrRoEbGxsVK56Pjx43z88ceyEZy3Y34ZTXDDAgICCA8PJyYmhtmzZzNnzhwGBweZOHEilZWVxMfHS2n/9vZ29u7dy/Hjx2/oHILYLSrAIvPv2UMpKCiIgIAAEhISmDhxIt3d3Zw8eZKSkhKCg4NRq9Vu67xIKAjxDfGOuFwu2XSysbHRzfF0Op0UFxfzxhtv0N7ezuOPP86UKVPQarVYrVaMRiPx8fH09vaydetWdu3axblz5yQvRzimfn5+BAUF3fR4R0VF8a1vfYv169eTkJBAc3MzmzZt4o9//CNVVVUyWy965ZhMJvkujPb+CnW0oaEh2SrDaDRitVrp7e11C+CHhoYwmUy0trZSWFjIunXrqK6uJjQ0lKioKK5evUp/f78bhE15ThHkiDUXhgMoIRpy4MABZs2axaJFi1i5ciXTp09n165dbNmyhXPnzt005aCvr4+amppr9mUSJt7VM2fOSBlnMdesVivh4eGYzWZaWlq8cn5v9No+/vhjQkJCmDBhAk888QSVlZXs3buXkydP0tfXh9VqJTIykubmZgoKCrDb7ZSXl7txtj4Pu2tBS11dHT//+c8JDg6mra2N3Nxc/vSnP5GWlsbYsWN5+umnZRnvdq2np4fOzk70ej0pKSlyQBMSEvj6179OfHw827Ztuy0sr9AYh2EM5V/+8hfKy8ulPJxarSY5OVlmVx9++GGWLl2KyWTCZrPh6+vLjh07bqsHx7p162RPj87OTn71q1+xd+/eWz6eMNFoCIYlCIODg0eFhynN6XRSVlYms88TJkwgNTX1toIWh8PBiRMnWLJkiey/UVNTc8e5QSqVivj4eOLj4/nlL39JZGQk69atu6PnUFpvby/l5eVs3ryZI0eOsHv3bsaMGSP5T+3t7Xe0L8zfunmDbQmehbJKUlpayh/+8AciIyOprKzEaDTy4IMPSlGIuXPnkp2dLSsEIrPszUFTOm6iEqLMZLa1tdHQ0ICPjw8zZ86UgcDYsWMJCQmhrq6Oq1evUllZOUKJSxzf896UjqnoJSB+Pn36dIxGI0uXLqW/v1+qpKWlpUn53Dlz5pCSkkJAQABWqxWn08np06dHZIeV3BJlkOBJRPXx8WHlypVs3LgRnU5HT08PH3zwAfv37/cqCe1t7ISJAElA8rq7u2loaGDcuHGEhoZK0rM3x1w4fyIbmpeXR3V1NQEBASQlJTFx4kSOHDniNWgZzTwd/AsXLtDS0kJiYiIJCQlcuHCBvr4+twqSMojxdvzRggoxvmq1msTERGbPns2bb76J0+kkJSWF3t5eNxiMuGcx18RxPE1cm2eAKH42ODhIfn4+paWl5Obm4u/vz5w5c2RyUHAVPIMeZfXss1JkvJumVqtJS0uju7ub8vJyHA4HRqORiIgIfHx8qK6u5vTp01I+VijxhYeH09PTQ2FhIZs2beLdd9+9YQEZZV8WQXgWcCzxrJqamjh27BixsbHExcXR3d1NSUmJG5fGZDLJxBx8ypH1nGcOh0OqZ3njQbhcLhobG9myZQslJSU8+uijrFixgnXr1qFSqbBYLLS1tUkye0tLC2azmYCAAPnutbe33zQaZfr06Tz55JPMnDmTwMBASkpK+NOf/sRbb70lE4yin4zVasVkMrnxQ71xgcT9OJ1OCWPSaDQYDAZ6e3vp6+vDaDRis9mklHNQUBDBwcGcOHGCadOmERYWhsPhICgoyG0sBY8DPk06CEn2vr4+tx5KTqeT9vZ2ysvLOXXqFHv27GHZsmXce++9PPnkk8ydO5fNmzfzxhtv3LTqmjcetDfTaDRYLBaqq6spLi5mzpw5MunidDoxGAyYzWa6u7tvq+JRWlrKtm3bMJvNzJ8/n3HjxtHW1kZWVhZ1dXW0tbVJOHNzczPnz5+Xqpqf5xpy14IWp9PJli1b5A329/ezd+9e0tLSUKuHO7neKUJgRUUFr776quwKK4j4KpWKmpoaXnnllds+V3Z2NjNnzpREO7ERKh+e0OGG4WaCAQEBdHd3k5+fz+9+9zsOHjwog5ybNYPBwM9+9jNiY2MZHBwkNzeXt99++4ZUfm7GJk+ezIIFCygpKblu1l8ogoh7FqS327WsrCxKS0uJi4tj4cKF7Nu3744HLYJfoNfrKS8vv6Vs062YcF6VxFuNRsPEiRNZsGABW7du/dITZYUpnSlwb5ooyt9dXV3s2bMHHx8fCTcKCAggMzMTs9lMWFiYnH+CYCmgG+Cu6qOsOnhm/QGqqqr44IMP8PPzY9y4ccTExODv709ra6tUTFFih8V5hXk6+cpgDKCxsZFz586RmJhIaGgoFouF6dOnk5qaCnwqHSyO6+PjQ2xsLMHBwbS3t3Po0CF27NjBqVOnZDDgrTmhZyZTCWsZP348zzzzDL6+vrJZ7J///GfZiMzb3LseTEw4Ef39/TQ3N+Pj48PEiRNZs2YNtbW10kFTEveVxHCj0UhcXBwWi0W+H0Jh7WY3R3GNIhPd0NBAZmYmiYmJGAyGETAJz8yvt9+JNV8EOULZTzivIlF29uxZ0tPTZYM74XCBe0d7b4GuMDFGnhUfJYRGqBT5+flJvlV1dTVGo5EJEyYwY8YMGhsb3SSZlVC2L4PpdDqefvppCgsL2bJlC5WVlTJo9vHxwdfXl4sXL3LlyhW6u7tJSkri3nvvxWQykZuby8svv8z27dtvCnosuI5iPojAVYhjwHDQIqofa9eulbKy06ZNk4IhggulNG/PRQSfnvwXYWazWQarXV1dsjGsIOKLJpharVYGAna7nYCAALcEijKAup6tWrWKdevWER8fj9FoJDs7m9///vccPnxY+iJarRY/Pz/8/PwwGAxu/C8RzHuzvr4+amtr6e7ulpWioqIiBgYGGBgYkFV34XMEBAQQFBTEyZMn2bVrF4888ggBAQHEx8e7HdcTwjo0NCQhemI9Egks4asKdbiWlhYuXLjAvn37WLt2LRs2bOBf//VfmTNnDj/+8Y+vqzh3syZ8Y9HgNzc3l5kzZzJ9+nQuX76M3W7HbrdLWfnR1my9Xk9QUJBMTLW2trrByARcLzc3l+DgYMaMGcOUKVOIjIxk3LhxuFwuoqKimDRpEklJSTQ3N5OXl0dra+vnnvS4q+physhSrVZLrHN/fz9FRUW3rA/uaQMDAxw5coTs7Gzi4+NZsmQJ3//+93E4HKxZs4by8vLbPsdDDz3EihUr2LBhA2az2e13arWa8PBwCUsDOHz4MB9//DHHjx+ntLQUu91+y4IAvr6+/Pd//zfjx4/H5XLR0tLCb37zmzsGm2poaOD06dPMnz9fNkJ0uVy8+eabo0rzaTQali1bxv/3//1/siPznbKysjJycnJIT08nKirqjgRCnqZSqYiNjR2BX7/bZc8xY8Zw6NAh/Pz8cDgc/Nd//RdNTU185zvf4amnnuLMmTNUVlbe1Wv4WzCxWYxWFRH8FJfLJSENLpeL6OhoAgIC0Ol0DA4O0tzcLEvkAs87WtAnnMzRoDJDQ0OyR0lQUJAkqldUVPDSSy9J7sCNmFI9TJyvtraWd955h7a2NtlUUgQewolPTk5m0qRJsg/EiRMnOHz4MDk5ORQWFlJbWyurKWKzFRuxt/VFGdgYjUaeeOIJpkyZIqsAv//976moqECn07n1j/F8J7xtVOI7wjFqamri9OnTzJkzh4iICB5//HEGBwd57bXXqKurQ6PRSFiVMJvNxoIFC3jkkUeIi4sDPq1IiGv3Vu0Q5vkslbCuqqoqzp49S0pKCuPHjycgIIDOzs7rOu6evxfOnZiDomIhHLLExES6urpobm6WyTgBU/RUbPPkl4wGY1RegzLQttvtshfGCy+8wPjx42ltbeW1116jt7eX9evX88gjj1BYWEhBQYEbpEb8/WWAivn4+DBv3jxCQ0O5evUqzc3N9PT00NLSQkhICEajkY6ODkka9vX1xeFw4HA4pHLpzXIllc6tWEeUCRIYHt/6+noKCwtZuXKlbBIoGq7eTP8P0djZ2+dFwhCGYacPP/wwixYtkvNSr9dLdad7772XoaEh3njjDWpqamhubpaNAysrK2+458Y3v/lN7r33XoKCgjCZTOzatYvf/OY31NbWyndatA4wm82SkwPIYN0bTFKY0+mUqnjjxo1j6tSpVFVV0dTURHNzMyrVcJ8Ts9kskwCdnZ3U1taSl5dHW1sbsbGxJCcnS34Q4JbwFGulco8QyWbxnolqj7ju1tZWjhw5QkFBAUVFRfzzP/8z69atY3BwkB/+8Id3hJsNwz5VVFQU8fHx+Pn5ERYWhslkoquriylTprBr1y66urqora3FarVisVgICQmRDZeV9xMWFkZKSgpWq5X+/n4qKiooKCiQcz40NBSbzUZ+fj7Z2dmkpKTQ0dFBV1cXqampJCQkkJCQILm3ly5dYtu2bZKnrVRk+6zttoIWoelsMBjcOq16mtVq5YEHHpDQJm9Zztu1gYEBurq6uHTpEuHh4bLBkJiEt2u1tbX85S9/4c033xzxO41Gw+OPP86//du/ERAQQHt7O7/97W/ZvXu3V9nFm7Xly5fzyCOPyA1w69atfPzxx7d1TKW1t7fz2muvERkZybe+9S2Cg4P5+c9/zoMPPshvf/tbysvL3TY7Hx8f7r//fjZs2EBISIhcmESTrds1kQUZGhoiNDRUqg7daRMbur+/PzNnzrzrmciMjAxefvllSf77l3/5F15++WWJ93322WdZsWIF//u//3vXruFvxZRrgJ+fH5MmTSIgIIDKykqpkS+ehfhcfHw8Dz30EGvXrsVsNtPc3Exubu6oPTi8ObSeAY1wJoUD4nA4yM/Pl3Pj6aefJjQ0lIaGhhtyeIV5yh2rVCopT37lyhUCAwPl+ywy53q9nm984xtER0cTFBREUVER7777Llu3bqWjo2OEw6nRaNy6yHtzBpQO1aJFi3jssceAYT7Xa6+9xo4dO9ySAkp1KTEmnnA3T3iTCBb7+vrYu3cvmZmZUnr3hz/8IVOmTGHHjh3U19dLyc+hoSHZuX3p0qVMmDABGH7WNTU1FBcXMzAwIAMYb89TWZHwBqVqbW3l1KlTrFmzhgkTJhAbG0tlZeVNSVcLZ0ZZsROKXAMDA+h0OillKgRX+vv76ejowOFwuF2fcFqF3ajzKgJ8kdmfMWMG//AP/8DcuXOprq7m1Vdf5bXXXsPhcGAwGCS/obCwUM4Lb0HQF9lUKhW+vr6kpaVx9epVSkpKKCwslF3C9Xo9er1ecjZERaqnp4fm5uZbSvgJ51e8E06n021eKG3q1Kmkp6dz4MABtmzZQnV19Q3Dg5T3ONp6YzKZpNDCzJkzWbdunVt/uvz8fM6ePYvVamXKlCn8+Mc/Ji0tjf/4j/+gqKhI8oxv1Df6wQ9+wLp162TV93e/+x1vvfWW2/dVquE2AQJCpAze+/r6pEy0sjLlaV1dXVL1LzU1FbvdTn19PQ0NDVy+fJn29nY5Js3NzXR2dtLb20twcLBUGfPz88NqtY5w5gGvnBLlvjDaezE4OEh9fb1s+vzAAw+gUqlIT0+XTXFv1QQHKTQ0lNjYWMLDwyWiIDExkcbGRmbMmIHVagWGg9nBwUHCw8Nl8FJeXi7HVK/XExUVRXR0NL6+vlKMoL6+XnInAwMDiY6OlvwU8b50dHSQnJyMXq+X7S/E/lpdXY3LNdx3KCoqisLCws8FFXLLQYtarSYzM5P//d//JSoqiqqqKjZt2sSrr77qVm40m81873vf48c//rFc/D/88EN++MMf3pEb8DRlGVzgwO9UZmm0LImPj4/EcAMS+36zi5Q38/X15bnnnpOOe3t7O3/4wx/u+GRpaGjg0KFDLFmyhPj4eHQ6HWlpabz22mvyM94ysMqfbdu2jc2bN9+R63nrrbdYvHgxaWlpPPXUU7z44ot3ZDyFaTQaFi9eDHzakKmjo+OW4Xs3Yps3byY6OhqXy8U//dM/8Zvf/IbBwUF6e3upqKjAx8fnlpVcvqhmNBrJzMzkmWeeYcqUKVy+fJmdO3dy6NAhLl++TH9/P1qtlrFjx/LAAw+wceNGxo0bR19fH1u3buX48eNSFUpslEpn2/N99ebcikqC0sFUzvPQ0FAiIyO5ePGiG5b9WnwWz2BJZOtdLhft7e1eoZcCoijgYXa7XcJeBPFUpVK59XbxNGVVQBzL6XQSHBzM888/L5MM2dnZvPzyy7ICI3pQKK/dM0vvWRVQZjCFU11WVsZ7773H+PHjmTx5MhaLhfvuu4+1a9fKRoiCXKrT6eQxlbCNY8eOce7cOQkp8eRneLtvbwHNwMAABQUFVFZWkpGRwdSpU8nKyhrx3EYLWpTX5vk7wQkwGo0EBwdLaeywsDD6+vpkkCmqaN4y5t7moshgKlXXlBWZiRMn8r3vfY+5c+fidDr53//9X1555RVaW1sxGo3U1NSg0WgICQmR74LS7kTfqb8FE7CppKQkHn74YRobG3E6nbS2tmIwGGSj06ioKHx9faUTB0hBhps1IREs/IuBgQEZnCpt/PjxLF26lJCQEPbs2cNHH31EYGDgTSdPR4NE+/v7o9fr6e7uxtfXl7lz5xIaGirnaE5ODi+//LLskWI2m5k+fTrd3d2cP39eCo7cqH3ta19j5cqVjB07lpycHH76059y9uzZEe+EWq2WMErleyuqq6LaIhqgintR9olpa2sjOzubqVOnEhwczOTJkxkzZoxsFXHs2DFyc3Nlxt/Pz4/ly5fz4IMPEhUVRWBgoISrejPx3JXv5WjVLG8mxBtqa2vp6uqioqKCgIAAbDYbtbW1N5UM0Gq1GI1GKVgVEBAgE+1VVVUkJCQwc+ZMqqur8ff3d0t0uVwudDqd/K5GoyE/P1/em5B5Fonk9vZ2CavTarUkJyczc+ZM7rnnHhwOB8nJyUycOJEJEyZgtVqpq6vjlVde4ZNPPpGQQzHPfX19ZYD8edhtBS0mkwmTyYRGoyE+Pp5//ud/ZsKECbz99tuy4eNPf/pTQkNDgeEo9/z58/zoRz+6Y9Awb6bEBUdHR3Px4sW7mkFPSEggLS1NbhDnzp27re7wSnvhhRdIT0+X2bGlS5dy6dKlO3JspQ0NDbFnzx40Gg3/+I//yIQJEyQR+Vrmcg0rnBw7dow33njjlhtcelpRURFFRUWkpqaybNkyfv3rX9/RoEWlGpYOdDqd+Pv7k5KSwp49e+5aM8s///nPREZGSuz7pk2b5AIXFBTEAw88wMDAwGfSTPNvwcS7otFoZIYpOjqamJgYKZBw9OhR7HY7cXFx3HPPPdx7772y+ZpQqhHvgmdG2ZtDOBokzNM5F58Xm4NQJhLmzYkeLUsvHHGhpqOES4jziA1Gr9djs9nQarX09vZSVVUlG3mJzUa5bno6NSIjL8ZCWWVZs2YNc+fOlY3x/v3f/53KykoMBgN2u92N76N8Pp6bsDf5XAGvUKlUOBwOjhw5wssvv8wzzzxDSkqKhOz5+vpKvqEYA9FEz2w2MzQ0xLFjx3j11VcpKSlxO4c3u5HKaG1tLSUlJcyYMUPKTdfX17uNFXhvuOgNESC4DOLzoj+NzWYjKSlJ7jcik38j16icQwL6pwyixby22WysW7eOZcuWoVareeutt9i0aZOUQrVYLCQmJtLb24vdbneDQorAWQRzX3QbGhri6tWrJCUlERYWxj/90z+xcOFCCgsL6e/vx2KxEB4eTlhYGP7+/kRERODr68upU6e4cOHCDZ8nLCyMlpYWnE4nvb29sroi0ABKzpJ4f6ZOncqECRO4ePEi+fn5dHV1YbFYbuh8AlI12l4n1ojBwUHMZjOZmZksW7YMgNbWVnbs2MGWLVvIzc2lra2NwcFBOjo6OHjw4Ihj+fv7o9FopAiPNwsJCSE0NJTu7m7++Mc/8t///d+jiieJ+etwOGQGXwnFEuuRSMqoVCrGjh1LWVkZ7e3tDAwM0NnZyYEDBwgJCWHZsmUSRj04OMiMGTOYPXs2KpWKuro66uvr6enpcUsciIr2taB/Su7jjah5KaXJXS4X9fX1GI1G7HY7dXV1WK1WkpOTCQsLo7y8/IZlkUXfFhjumyLggCqVitraWoxGI48++qiERC9cuJCBgQGcTifNzc2cOnXKLYBVHrelpUVW4svLy6mpqcFkMpGamkpGRgbr169nzpw5qNVqKioquHz5MsXFxaSmphISEkJra6tcLwYHB+ns7JRBtxAx+LwgprcctAwNDVFYWMg777zDww8/TGBgIBqNhg0bNrBhw4YRn+/v72f37t2yq/zdtPz8fAnlmDRpErt3776r2aXy8nJycnJYvnw5MJzJuRMOtkqlIiIiQm40v//978nNzb3t445mXV1d/PWvf6Wjo4P7779f4leFQyJMucH29fVJ6eOTJ0/e0evJzs7m3nvvlZr6d9KGhoaoq6sjMTGRp556CpVKxYULF+6Imp03mzhxImq1mpqaGjZu3IjD4cBsNhMTE8M999zDwoULKSoqumPB7t+6iQqAw+EgNzdXNl5LTExkzJgxfPOb32T58uX09/cTExOD1WrF5XLR29vL66+/zh/+8Afy8vKAT4nG13rHvRHjwd2pFEGI6Jbc1NQkCeqiy7Zyo1Eee7Tsv/K9ERuSIMEqIRJicxcZyra2NkpLSyUcQqVSyTVNeT5l53pldl78bmBggODgYBYsWCCz+P/93/8tx1usUyLAUtqNKneJzVxs7q2trbz77ruUlZXxyCOPsHLlSgICAqS6mVDsMRgM0gHr7Oxk3759vPrqq2RlZY36HG9ko1Q+z66uLsrKyujr62P8+PGyIa/yMze6NwiHUswP4ZiJjPfSpUsxm82cOXOGxsZGr9A1zwwvfPoueF67571qNBqmTJmCRqPhzJkz7N+/H6fTSVRUFFFRUSxdupS1a9dSUlLCpUuXpGiA8phfdFiYsIGBAXbs2CEFcWw2G0uXLmXp0qUjPiuecVNTEwcOHLjm/FKawWAgJiZG8lGFMy66uivHUojQDA4OkpCQgL+/Px999JHsz6GsKnoz8e5YLBYGBgZwOBwj+lbBcNDS09ODw+GQSYBz585JbsPRo0e5cOECTqeToKAguru76e/vR61Wj+js7nK5MBqNmM1mr9K5MAxFOn78ONu3b6eiouKa80fA7wQsUqwJIuEi/jYajXKfi4uLw263yz40AMXFxbz55ps0NTWxatUq4uPjqamp4ZNPPiE8PJykpCSCgoIIDAyksbFROtTR0dEMDQ25NaD0ZmINvVEEjqiuq1QqGaR2d3djsVgkOiImJkZyTCorK6mpqbnuXi7WAAEtU8r2i6Bl/fr1LF++HB8fHyZMmCCrqZs2bZIqZ57mcDgoKSmRpPnGxkaZ4FqzZg0vvPCCDKKdTidtbW3s3buXoqIi1q5dy9e//nWCgoJYvHgxzc3NtLa2yj/wadDyedltBS0lJSX853/+J0eOHGHDhg0sWLBghJPb399PZ2cnu3fvlk387rY1NzdTUlJCUlLSHWsueS1T4rS9YYhv1SIjI1m2bBkajYYdO3bwox/96A5c7bVtcHCQXbt2sW/fPtatW8e9997LPffcIxddcV91dXXs3r2blpYWPvjggztGRvO8lrsVzQ8ODnLo0CFSU1OZNm0aFy5c4PTp03flXErz8/OTfXrUarUsfRcVFfE///M/HD58+K5fw9+CCSfN6XSSn58v+QcLFixg5cqVxMbGShUYwfmorKxky5YtbN68mYKCAuDTLL+y6uBJjBV/X2stUEKERAm+pqaGvLw8xo4dS2RkpJROVdq15qcn4ROQm7NSkUtATvr7+yXszGQySYy2IOnDsNOiJLN7JhOUToUIgITCWm9vL/v27ePll18ece+ezrsn1MrbfQo1LPFdZYW7u7ubgwcPUlBQwI4dO1i8eDEZGRmEh4fLa+7o6KC2tpbCwkKys7PJysqirKxMVolElUsJNblRU0JTRCZ3cHDQDdbg7fko71fZeFIEhGKeCXU7p9NJQUEBmZmZzJkzh9zcXE6cOCE5Lp7HVcqrejPPihcMP3MBGRPwPqvVyqOPPsojjzyCRqMhIiKClJQUioqKeOONN8jNzR3hJI/GDfoimtPp5MMPP8Rut8vms1qtVjr/Yk6K9+3y5cvs37+fDz74YFTJXU9zuVzSoQYkDFUJJ1Kr1dLxF3LaHR0dnDhxgtOnT2O32/Hx8aGtrW3UZ+7j4yP3VyHOAcOtFTQaDVVVVTgcDnQ6HUFBQVIZLDo6GpVKxeuvv87+/fuJi4uTfaxERr6jo4O2tjYpp6600WCqSrPb7Zw7d+6GxkuMmTfellardZNBrqqqQqUabmxpsVhoaWmRgYbT6aSkpIS3336bwsJCHnroIWbNmkVKSgpNTU2cOHFCvs+it0tUVBRtbW20trZSXFwskxOjQfLEdd5I1VH4dUIRUKxHGo2GmJgYYmJiSElJkRA2IbksnPybed9cLpesLBUXF3Pw4EHS0tLw9fWlpaXFLZF2retVdrZXrlnz58/HZDLJcd6zZw8ffPABhw4dwm63S+RDeno6gYGBxMTEyGaWws/t7u6WFeHPw25bPUyUJE+dOsXixYt5+OGHiYuLkwtmbm4uf/jDHz5T0o7T6eTQoUPMmzePoKCga2Y47oQJyclLly7R29tLYWHhDStyXMuam5vZu3cvSUlJ/OQnP7njsr/XsoGBAd5//322bNmCVqsdMYZCGvBuWlFREXl5ebckfXo9GxgYYP/+/Tz//PO4XC6Kioq4ePHiHT2H0oQgg6+vL3PmzKGjowO73U5HRweXLl3ilVdeYffu3Xft/H+LJiokAwMDUsijuLiY8+fPM2vWLBISElCpVDQ1NXH16lVycnI4d+6cJFiKLJ5yAfVGyhZ2LciYMmgRm1JnZyeHDx8mPT0dm80mm5zdqCmhVhqNxs3B9ww2BFm1uLiYc+fOYbfbyc3NdctoCUiDWFuU0qmezoKSQN/X10d+fr7kHXrytkQFyZt5OtFKToS3DV9ZcRFKStu2bePjjz9Gr9fLhJbSqVSq9Siv/UagVd6u1/M7ly9fZvv27TidTrceJt72BeXPPB0Dzyw1DDuyhw8fZuHChVitVo4dO0ZRUZH8rjcei2eAqJwnIjhSQrmE6trg4KAkNMfFxREXF4fD4ZCV/Y8++oi33nqL/fv3y6y18hl92Xq1tLe3895775GXl8eMGTMkVMxkMjEwMEBLSwtVVVWUlZVRVlZGRUXFTSXX+vr6qK6ulvtcd3f3CGEI4YQLcnlXVxcnTpzA4XBIWXIRPIxmykRnT0+PfGadnZ1MmDBB9uxISUmhpaWF4uJibDYbsbGxREZGUlFRQX19PZ2dnTKgEeuN3W6nvb3drfr6eTx7nU6Hr68vRqNxRGLFcwzE+y/8nyNHjjB16lQ2bNhARkYGiYmJ6PV6amtrqa6upry8nOzsbDo6OrBYLPj5+ZGSkkJFRQUtLS0j1nmRTLhRf1QkK/R6vQxc9Ho9ZrMZf39/bDYbZrOZyMhIYmJiCAgIkBLBNptNtr7o6Oi4rl/Y19cnK/kOh4P9+/eTmZlJWFgYx48fJzs7G71eLxsfXwvR46lsl5CQ4CacVF1dzbFjx8jPz8dgMGC1WrHb7VRUVEi6QGNjIwaDAbVaLXtrDQ2N3mvns7A7Jnnc3NzMu+++y7vvvnunDnlbdvz4cQ4cOHBTij+3aoODg24N2jo6Ou4IbtjhcPD888/fgSu8dbsZicY7bUeOHOHIkSN35dhDQ0Pk5uayY8cOAP70pz/d8Z43Stu1axeDg4My07pr1y6ysrK4dOmSzLD/XzNlg0bhkJWUlFBRUcGBAwdkQ1eB0e/s7HQj2nuqOt0I6fha3BZv/z5x4gRvvfUWlZWV8nveYBvXM+GMKgnyyn/7+PjQ09PDgQMHyMvLo6Ojg9bWVvr7+90w4GLTUzq4nvcnPi/WoKamJnbv3k19fT3V1dUYDAY3R1k5fp5Bn6d6mLKq5fk9EWR4q2oJ2Ibye+K7ymOI34vPeK6jyoBmNOEA5bFdLhdXr15l06ZN+Pj4yIBXq9WOcOCUx/Y8t1BBUp5DjFVWVhZvv/024eHhbNmyRWavPa/dExamHBvP/yvhsIIT1Nrayocffkh7e7sUNLh48SJZWVmUl5dLPqHoHu7tWX4ZqiwwfC/x8fFUVlZy8eJFLl68iNlsxtfXV1Yju7q65JohKrKianWj765SFWpoaMjN6VRm64WgCsCVK1dkIH4jY62s4CmDG6vVyowZM5g8eTKXLl2SSbzBwUGsVqsUGgoMDJTS4+KaWltbKS8vH8EdvtG9XEg730qV09NEssXhcNDT0+O2figVzERFWRlgGQwGHA4Hx44d49ChQ6hUKiZMmEB8fDwajYbW1lbq6uro6ekhMjKSxYsXEx0dLQUIWlpapNy16PUi1s2bgfA7nU7JE7PZbAQFBREVFUVISIiE19XX1xMUFER4eDjh4eHAsGLogw8+SF9fHzt37uTEiRPymXhyZQYGBmhsbHR7Rnl5ebz88ssEBgZy8eJFysrK0Ol0TJ48GX9/fy5evOj1PgwGA35+frLp79DQkITiCZXB+Ph4Hn/8cSZNmsSFCxcoLS3FarXicDikQq2/vz+hoaF0dnb+zXDh7mqfls/TLly4wCOPPPKZnU9owH9lXxyrr6/nwQcf/EzO9eKLL/Liiy9+Juf6ophn6V5JUm9qahpBEBXBiidEQ+n4KjdYZdCiVBbz5LCM5kTDcM+gd999l6GhIelQeDry3s6t7C8i+A+eDoM4juAdDA0N0dDQ4NYbSWTcxe/Fz5UwEqUpyZPClHhkGHZcxDmVFZTrBXSjBQXKAMLzO8p79TbWnoHKtcbVM0i5UYewo6OD8+fPj+ACeau0eDu/8vpEpUj5LKqrq/nLX/4iOWsiwPF2zZ5BhCckT0A6lNUYIcQwNDTE5s2b2bdvn6zcisBWvDvifCIQFg6ieF+uVYn8IplKpSIsLExWYu12uwxSAJkZNpvNMjMtelbU1tbS1NR0R4K3rq4uCfcUAY3dbqe2tlbyIG7VROPjoaEh2traCAgIICoqirq6OqKioqRKnRAIEJy4xsZGysvLbzkbbjKZCA0NlZXS20GNKJUQ4dNeN8LEczAajXKNFPNTBIJWqxV/f3/Z50VUzXQ6nZQLjo+PJzk5WSqfKkVNzGazFEOxWCySoC7gnTdq/f399Pf3YzabiY2NZezYsQQGBgLD/l9dXR2tra34+vrK927KlCls2LABnU5HeHg4bW1tnD17Vgaevr6+dHd3y2DOc48YGBjg3Llz9Pf34+fnJ5W7wsPD0ev1tLe3U1JSgtVqpb29HZVqWAo8NDRUtt8QUDqXy8XFixfp7++Xsspms5m0tDS5nrS3t0uFNrvdjs1mk/NstCann7V9aYOWr+wr+8r+dk0ZRMCnTqGnc6t0mD2rFeLnnpwM8X9lZkjpRHo6i57BjdKxFXBP5Wc8z+/N2VY62iI4UN6rcDJUKpXc4JSkafG50Soi4nee1QKlcyogIvAp/0Q4Nt4c5tEcdnE+AVfyvPfRvu/5OeX5xLGVzr+373meY7Rji997Vn3AHeal7B9xvfOB98aS4rqFuqLT6aS8vFzOHWUFTFTUPMdHzENxLM+fi3FRBiM+Pj6yAZx4T5xOp4QDid5WQoBBZMmVcD1vgfoX0VyuYey/xWKR6kl2u1062CLTHBoaSkREBKGhoRgMBlpbW2U/lzulYOrp+Io5oVKpMBgMXp3RG7HKykra29tpbGwkNTWV2bNn09TURGtrK0VFRTQ1NREREUFQUJAM0hwOB01NTV4DFqPRSEREBMHBwdjtdq5ever2ObVaTVBQECkpKQQEBFBUVOTWvuJmTKk8Krg+FotF8imEwEhrays2m43g4GA5b5WVF7VaLUUQfH19Jb/CbDYTGhoqqx2i6aToX9Pb24vFYpF9SsQ7JgRwxDtzM6bRaPDz8yMyMpLY2FgJRRRVLoEIaG1tlVXP4OBg2VR9/PjxhIWFyepKQEAAAQEBtLS0yIBImFiLlHteREQEZrOZwcFBYmJicDqdUt579uzZ7Ny5E71ez9ixYwkJCZF7i5+fH2q1mubmZo4ePcrFixcxGo309va6QR6vXr3KlStXZLNQUWURe8dXlZav7Cv7yv7PmrfqiLfMvzAl5tnzGNciUwsTxx0tuPD8nPidtwz5jWb4xX14XrPS+VVWKpTwIc/79xwnZaDj7bziXoTTKhwC8X1PJ2o0iNVox/b2/9FstGDDm/N8vcqLt+d/LROOj2dAMxofx9M8+Uee9ywcVPi0aaNn0O0t6FJWPDyfrWdgKHgRyp8pgx7hUCi5MJ7wudvJ+P8tmsvloqqqSsL8hoaG5LsgAkcB1YbhioiPj490Ku/meAhnUVTcbjVAFBDA0tJSent7qaurIzs7WwZb3d3d1NXVERkZiV6vl5LoozU69PHxwWq1Ehoaik6no6KiYsR1C/ic4MncamNuMb4icBb8RUEyFyYqOQI+pqyGizkuKiyiMaXRaJTcn/7+flpaWqRymwhwWlpaqK2tpbe3101sQAQ0Wq32pjm5ImkjxqW4uFhC6ETQIaCCDQ0NABw6dEjeb0tLC1euXJH/7+rqksqJnigd5TsuxlJUq/r7+ykoKKC3t5eGhgZZiYPhud/X1yfHVUmPqKiooKmpSVZgW1paZBDn4+NDc3OzG2RZrVZLQv9nyae+nqlcN/hGfdkWva/sK/si2xc1U/rVOvKVfWV/O/ZFXEe+WkO+sq/sb8c+6zXkq0rLV/aVfWV/0+ZJkBamzMgpoTQi4+wJSfI8pmc1RHl88XuRyRWYX3HeG+EGePIZhI12baN9/kZMCVXy/Lny3oTk77Waqol7FgTw0ZoSXiuLPFoVxXOMvcGmRjumEu6mPAaMXvVRcot0Op1bhUqj0cgM57WIx55cGMCtIugJbfSEvHkThLiR+/UGCRTHE2pVynfAaDS6QcQ8q3mfl3LUZ2HBwcGsWrWKxMRESktLOXDggFRBAne5aYPBQEBAgIRViQy00jQajVRNstvtDA0N8fOf/5wPP/yQS5cueR1H0c9JVA1udKxFh/LJkydTV1fH2bNn6ejo4J577iE+Pp7Gxkbq6+tpa2ujs7OTvr4+yecwGAz09vZKorR4V/v7+2/qGuBT+JjgxKnVw41TBW/E19dXKqV1d3fT0tJCV1cXbW1t1NTUXBdup9FoCAkJISoqiuzsbAAsFssNwfQ0Gg1RUVFYLBaam5vdVOAEV0m8yyrVcONom82G3W7HZDLR3Nw8KsdHVKBEpUkpM9/T0yPV1/z9/dFqtW5CGAEBATgcDrd7ED17lP28hKrZta5DmOeaIKSiBdytr6+Pmpoauru75bhERERw8uRJ9Ho9EyZMwNfXlwsXLsjr8vHxkf3BTCaTW8PitrY2ea2C59LZ2SnfF+Waq9PpiIyMpK2t7bpy2XfLvgpavrKv7Cv7zE2J1/XE9CsJ2uJnYuH05nh5k3AdDTIG7p3cvfENxDGVkDThlHpzSJXf93QulaTqa3EzrjVO3mBwntwNz98pzyG4MgK6IK5BwGmMRqN0RERw5gnFUwY93qACNxLIeeNWKJ1KTyfdc354CzKvN5bKcVeeHz7lIozGO/G8d+UxxVgAI9TURCClhHYoVe/Es1DyT7yZMhD3PLYyIBGYc8GTEkpZnk3+lMf4WyDU3kkTDvukSZOIi4vD6XRy5MgRGhoasFgs2Gw2BgYG6OnpQafTERISQlxcHEajUfJAhAnVKj8/P9nfaGhoSBKbrxWAChI4IL93PdPpdIwdO5bExEQpWQyQlJTE2LFjUavV9Pb2yv4cImCx2WyyWSuA2WyWfUQEREylUqHVat3m6mhQHwFDFEkVi8XCxIkTycjIYNy4cRgMBrq6umhsbKS2tpbm5mZ6enro6urC39+f0tJSmpqaRtyz4EaIaw4KCuLMmTMykXAjplYPN3gUED+laTQa6XRrNBqCg4Mxm810dXXR09NDYmIivb29owYLQ0PDEr5ibMV6KpTCrFYrJpNJBlj9/f2yMa7BYKCsrMzteKLvjjJoEdyT5ubmUe9R9FsS6ncOh0Oqq/X09BAQEEBoaKiEPQpivL+/PwkJCZw8eZKAgADuu+8+8vLyZGAo7lEktUTvIbVaTUhICP7+/nItFgGXuA8l/FL0KYqKivpySB5/ZV/ZV/aV3aiNVgXxDAA8HWFBMPcmJ6rkgozmzAonUWDglYIAygBDyf+4UWfZE7aiJIB6CwA8A57R+B2jOT6e1yHuSeC1lfhkEYiIzUp0dxf9lvr6+uQ9e+MOiWMpHexrBQzegrjRxkN8TvmsvQVO4jl58nPAXTVNVE3E+CllsYWJ7+v1elld8XbNMFKaW/wM3EnYSu6IZzAurlGQar2JOijP5znnlNUTz3dHeS1iHguegLf58UWEhF3PREPHkJAQYmNjSU5OBpC8DaPRyNDQkKxEiYaQJpMJm82Gr6+vnBNinIxGo5vK3Pbt26murh51/MS6odfrMRgMuFyfNo69lrlcLhoaGjh58iQ1NTUyCGlra5NKWbW1tTQ0NNDT04NWq6Wzs1PK4w4MDKDVaqUyluBWiDljNBoxmUyoVCra29tHDVqEspp4xzQaDb6+vowbN47MzEz0ej2VlZW0trbS3NxMTU0N/f39+Pj44O/vL9W0Ojs7ZXUqMDCQCRMmkJSUhEajobq6msLCQjkmQpb4euZ0Oqmrq5MSvkqz2+0MDAxgMpmYMmUKWq2W4uJiGhsbCQkJISkpiZqammuOv6jSiPdGp9MRFhZGfHw8UVFRBAQE0NnZSX5+PoGBgfj5+WEymairqxsxnt6EPsRcGC1BodVqmT9/PvPnz8fX15fKykr57EtKSujs7KSrq4vu7m76+/tllWNoaIjm5mb5f6PRSGRkJAcPHnS7LpfL5VX9q7a2lsTERPz9/WltbZXHERVD8X64XC45Nz0Dss/avgpavrKv7Cv7XEw4oSIQgZFO4I0EJsoqCDAiAFEeT/xeZLuFKasWwoRSjNjMRaZJWdm5HkH/Wg7Ltaoj1/u+t5979kYR51D2R9Hr9eh0Oikner1zCufFm7Msfn+tKoXnvY0WsHmr0ox2bd4Uxzwb1glHU2SZvVWklH0SxL14Hks8b28iBEqnQEC2PEnzSrsR9R0ljEk55p6wM6FKpeybo9PpcDqd9Pf3jwjulYHjlzFocblcdHV1UV5ejs1mIyIigrCwMMrLyykoKKCpqQmTyURYWBh6vR6XyyUJyAaDQUJmROVCwOmUlp2dfc1nqKxmip4wgsR8rURKb28v58+fR6fT0dfXJ59nbW0t7e3tVFVVUV9fLwMBoVTlOZeVAbG3d0QJobrW9QNSfljApMRaITqid3R00NTUJFWmRLDY19dHa2srjY2NOBwOQkNDmTFjBpMmTaK7u5v6+nqKioqA4bkeFBQ0QhBgtGsbreeeGF8/Pz9CQkK4cuUKjY2NBAQEMG/ePNLS0jh16tR1zwGf9qXz9fUlIiKCxMRExo0bR2BgIKWlpdTV1WG1WgkJCaGrq4vq6mqvoi+ea8DAwIAk3nuaWq1m/vz5PPfccyxYsACA0tJSjh49SnV1NRMnTqSoqIi6ujouX75MT0+Pmwx1XV0dZrNZjkVZWRmVlZUjziPmS2xsLBaLhcHBQfkMDQaDnPsGg0GKHQjJaOX78Hm39/hSBi0PPPAAP/nJT8jOzuY///M/KS8v/7wv6Sv7yr4yhQlnVziN3pxJz6BC4Le9VR7Epq38nXC2xffFpuytz4onTEmlUpGWlsbGjRtpa2vj97//PVeuXJEldKUT7m0jVSoZKe9Bqfjkibf3vH7PipMS3uQJl1Oq9YjfeQZiolpgt9vdfiYa1Pn4+MiMXl9fn9zAtVqtm3PsbdzEM/AcD7HZiXOLAFPZ5d3TIRewBOGgjRagiF414n7VarXMqItKk+c1K/uXKJ00Mb6ezobyupVjqjStVivnq6j0+Pv7Ex4eLhvgtbe309XVJcdLq9XKyojyWYt/K4MMpYmxE/NH3J+4flGFcrlcXp+XqAZ9WQIXJSTLaDTS1tZGU1MTUVFRjB8/npiYGAYGBqitraWxsRGLxSLlj41GIwMDA24KfMpmkZ7zTyg0XetalPA/MZdEb42Ojg6vgYbIwntCbpqamty+J6olXV1dXiuHLS0tMvARTqWAionv3MxzF/Opq6uL4uJi+vr6qKyspLa2Fp1OR2xsLAaDgfDwcOLi4mRvl5KSEnJycqiursZisRAcHExkZCQtLS1uvYMAKQcsxi8oKIiwsDCqqqpGcCauV9ltaWkhKyuL5uZmrFYrs2fPZs6cOURFRd2wvLGQfZ40aRJJSUmEhIQAyGBS3I/VauXKlSvo9Xqio6Ol6lprayvd3d0j1hFRZfHcuwwGAxkZGTzzzDMsX75cBtSJiYn4+PhQU1NDZ2cnoaGhvP/++17hiQMDAxQXFwPQ09PDuXPnqK6uHnFvfn5+zJo1i7Vr15KSksKYMWN44403eOONN2htbZUcObPZjNlspre3F4fD4aaKZrVa3daxz8O+dEGLSqVixowZxMXFkZyczDvvvENFRcWXZpEGCAgIwN/fn2effZYpU6YQGxtLS0sLb775Jnv37qWsrOwLQ7jU6XRkZGSwdu1a/P390Wg0NDU1UVxczPHjx2V34btter2epKQknnvuOb72ta+hUqnYuXMnmzZt4siRI58rhvPLaEqYjRJ2A59m0wW5WJBhR3sGwhHzzIiP1lNEHFt8V/kZETyEhoaycuVKVq9eTU9PDx9++CFXrlyR3xcVH0/zVsW5nnmr+iiDGCUkbbTAxptDIjLvQ0ND6HQ62XdGnGP8+PGkpaUxZcoUkpKSMJvNVFRUcOnSJXJycjh37px0HMS9iu7Pnuf3vAbP6orn+Atn0FsVQzkfRutgrqx4CWdn0aJFzJgxQ/IXOjo6qKmp4eLFi2RnZ0uZVOVYKzk+nuPnGYSJxnjKyoZoiifgEjabjdmzZzN37lwmTJiARqPh6tWrnDt3juPHj1NWVoZWqx31vq41dwRsRVQCdDodiYmJNDc3S4lTcWwRlAl+gggMxdz6LNbUu21qtZrQ0FCMRiP9/f10dXVRUFBAS0sLbW1t9Pf3M3bsWCZPnix7mwh4S2xsLEFBQXR3d5OTk8PAwAAOh0OOiwgIBf4fru00azQaGTy5XC6ZDRdrmHIeRUVFMW/ePLq6ujh48CDd3d0jYI8wDHkTMJ2AgAB8fX1Rq9WyC7zSRgt8xO+ude2jcZzEnGlra6O1tZWamhoaGxsxGo2MHTuW2NhYAgICMJvNWCwWuX/7+/tTU1NDXl4eDodDEtfFtUVGRspKgOjbAsPB3dSpUwkNDcVkMnHp0qVrwpCio6Nl7x2n00lnZycNDQ34+voyceJEEhMTgeHgTxko+fv709nZKZ+rwWCQQU1KSgqLFi0iLS2NmJgYenp6KC0tpaGhgdbWVtRqNX5+fhgMBgwGA8HBwYSHh2M0Guns7JSVbc+EgbcGnWFhYSQmJrJhwwbWrFnjhjYQjR1VKhU1NTX4+/tjMBhGHQtxLw6Hg/Lycq/v9+zZs/n+97+PVquloqKC9PR0vvvd71JSUsKBAwcwGo3ymQQFBcnn3tLSIoP5qKgoenp6vgpa7qTdd999rFu3DovFAniHTNyuqVQqQkJCCA8Pp7Ozk9LS0jt+jtFs6dKlbNq0CZvNJrOsLS0thISE8Mtf/pKNGzfyve99j6ysrM/smm7VtFotq1at4i9/+Yt8YeDTzaG9vZ2tW7fym9/8hqKiorv2ooSGhvLCCy+wZs0axo0bJzPbDz74IPPmzeM73/kOO3bsuCNa5ZmZmfz0pz9l3rx51NTU8N3vfpc9e/bcgbv4YprAXQ8ODmI2mwkMDGRgYIDm5mb6+vro6elxg2kkJSXR0tLiRmhUBjmjZZE9Ky5Kh078TqkgNXnyZGbNmoXFYkGlUkmyoqfz7Wnezi2gOkJQQHzPYrEQGxsr8cTl5eUSr+0tCLue8yGuX5BpBYxOqM8oM5fr16/n0UcfZfLkybKaYDQaGTNmDHPmzKGoqIgXX3yREydOjHpOTzUvZcAiqlxKuIq3cRI/1+v1mM3mEQGqcrNXijC4XJ/i0A0GA+np6Tz++OPMmzfPzUkcGhqipqaGnTt38sEHH3DixAm36ornvFD+WynGAMheEMp77O/vlyTeuLg41q1bxwMPPMDkyZPlmtbb28vEiRPp6OigqqrKjSN0LVOS7sV8cDgcaLVali5dyjPPPENgYCAOh4M//elPvPvuuwwODkoiuCDUKvk/3iB+X0QTgaqfn5/cA0VPFmXgPXv2bOLj47HZbAQGBqLRaAgMDGTatGlMnDiR5uZmsrKy3GCUMDxmogHl9cZLcFjEeHd3d7tBMvv7+zGZTIwdO5b09HQeffRR5s2bx65duygtLSU/P19WNZRVvfb2dlkRMpvNBAcHYzKZCA4OJjAwkJqaGln90Wq1xMTEEB4ejtlslpU+q9VKZ2cnFy5ccEtCKMfRaDSO4IkAst/J1atXaW1tpbq6Gj8/P2bMmMHKlSuJjIxkaGiIyspKqqqq5Ny22+04HA4aGhqora3F19cXPz8/ent7aW5uls63SqWSfppOp2PixIlER0cDw++0Xq/3GrSIyk5ycjJxcXGYzWbKy8vZsmULjY2NREdHExAQwJUrV8jLy0Ov19Pa2goMB0YJCQmoVCquXr2KSqVizpw5JCcn43A48PPzY/r06UybNo2IiAhcLhfBwcFcvnwZgEuXLlFXV0dQUBBms5mIiAi3SrJ494SZzWYZECvN19eXqKgoJkyYQGZmpltlvqenR/o7fn5++Pj4YLFYGDNmDJWVlddMoApRAW+WmZlJSEgIP/7xj9m3bx+//OUv2bFjB+vWraOkpISmpia6urrQaDQEBASQkJAg56PYA0NCQrxWcT5LuyNBy4QJE/iv//ovJkyYwPvvv8/bb78tJwkMdwLdtWsX//RP/8Svf/3ru1YF0Gg0spurJyn2TprZbOall15i/fr1tLS08Ne//pVXXnmF/Pz8u7YhGI1GHnzwQX71q19hs9m4evUqR48eZfPmzVRVVdHb28vPfvYzHnvsMe6//36ZHbgbJmAYDz30EEuXLpULTU9PDydPnqSgoIDjx49fk/wGwxmFkydP8tOf/lQuXsKWLl3KtGnTePrpp1mwYAHr16/n0qVLd7RiptVqWbhwIY8//jgPPfTQiDnj4+NDeHg4MTExI/DNt2L33Xcfv//97+np6eHNN99k7dq1/Pa3v2Xs2LG3fewvminJzYIPMHHiRB577DESEhLIyspi3759NDc3Y7PZpFP4rW99ixMnTvDrX/9aOimAG0QK3KVulaassHhyF8T1iKy9wGir1Wp8fX1lBtubeVZulP8X/1Zen0qlIiYmhieffJIHH3wQp9PJBx98wLvvvkthYeGIzJwy6FHyQpQmziU2QIfDIT8ryKqBgYH83d/9Hf/4j/+I1WqlqamJt99+myNHjtDZ2UlAQAAbNmxgyZIlLF++nMLCQjeirHJD9Fb9UFZXPDk74m9lkBgfH8/cuXPJzMwkKipKwl+amprIz89nx44dVFVVuY2tpwloyIkTJ8jLy5PXa7FYSE5OZu7cuTz33HMkJSXxi1/8gsOHD3sNosQcUga24pwCiiYymMK5EL9bsGABGzduZOnSpQQFBbkd22QykZqaKtcRzy7ko42jp+KXmK+rV6/m17/+NT09PeTl5bFq1SqMRiMHDx6ksbFRXpdyrnyZkAbCrFYrMFyRUK4F8ClPpKenh3HjxuFwOKiursZgMBATE4PVasXHxwdfX19CQ0NHHFtku69XMdVqtVIKV3xH+e4aDAZSUlLIzMxk0aJF0nEUogDCARTBvfI5KcUxRHJHBEZWq1V2Rp82bRrLly9nypQpmM1myUPQ6XTyvcvNzeXXv/41mzdvdus4HxYWRlhYGDk5OSPurauri5KSEtmkMSoqikceeYS1a9e6wVxPnTrF+fPnZWWps7OTiooKWRHcv38/Op2O0NBQSkpK3DgswhHW6/XExsZiNpupr6+noaHBayAVExPD9OnTGT9+POnp6WRmZuJwOPjzn/+MXq8nLCyMqKgoOjs7uXTp0giSv06nw8/Pz63iGBwczIwZM6irqyM3N5eGhgby8/NJTk4mKCgIh8NBSkoK06ZNo6amhsOHD5OamkpoaKhMnvT29uJ0OkdUcpXS454mKuAiuSHWlAsXLpCfn09AQABjxowhJiYGo9FIe3s7BQUF10QOCfK8p4WEhDBmzBhOnjzJ8ePH6evr4/z582zevJmHH36YxMREVKphpTmHw8HAwADjxo3D399f8pgGBgYwm81e99XP0u5I0NLT00NbWxvh4eH84Ac/4Ic//KHbJgDDg/mzn/2MTZs2jdBEvxOmUqm49957Wb9+PQEBAYB3BZc7YXa7nV/84heMGzeOlJQUHn/8cTIzM9m5cyfbt2+nuLh4hCzfrZpKpSI1NZUXXniBRx99VDouly5d4plnnnH7bFZWFqtWrSI5ORk/P787GrQIQq5Op+PHP/4xGzduJCQkxA2XPTQ0RHp6OpWVlTz33HPXDVpguCPu//zP/4z4+S9/+UvWrl3Lv/7rv5KQkMAvfvEL7rvvvjvWmdXHx4fg4GBeeOEFlixZ4ga9UUrVCifmdi0yMpIXX3yR7u5uMjIy6OjooL6+nieffJIxY8Zw9erVO3BXXxxTjqkY97a2NoxGIzNnzmTZsmX827/9G83NzW7ZeJvNRnh4OH/84x/p6OhwCw48oV5KJ1qJH/cGA4JhR0WtVpOamsr8+fMJDg6WTnRdXZ1bEKBUHfPmWCrPoVyHlNdbXFzMgQMHSElJ4Z577uHv//7vWbBgAR9//DFHjx6loKCA9vZ2uY6KwEW8A6ORw5XEeUByO6ZPn85jjz3GU089ha+vL11dXbz//vts2rSJCxcuuB1r6tSppKamSiy6uG5loOeNzyLuTckzERUZ4YhrtVrGjRvHmjVrWL9+PSkpKSNgcG1tbQQFBVFYWEhjY6NUfVLyBsSfvr4+zpw5w+XLl2V1Tpw7PDyclStXSpJrc3MzV65coaqqSkLmPLO54r48Cfre1h69Xs+qVat46qmnWLBggYR9CUfG5XJhtVoJCAggMjJyBLZep9O5ZWFFxl0QoYWJ8Y+JieEHP/gB9fX1PP300xQUFPD888/z7LPPkpiYSGNjo/zOaHA3b7/7oplwsOx2+6j9IgT3zG6309HRQUVFBT09PQQFBVFZWUl8fLyshHjCaVwul1en2fMaLBYLFotFwrbEszSbzURHR7No0SJWrlzJ9OnTCQwMpLm5mdbWVpksUAZbns9E7LkwXL2pra2VlWCn0ymd9MDAQOrq6mhtbZWKX3q9Hr1eT3JyMlFRUUyZMoWMjAz++te/yh4dKSkpTJ06lebmZnJycmQA19jYiE6nIyIiAq1WS2NjIwaDgcWLF7NixQpZoXY6nfz2t7/lpZdeora2lsDAQJYsWSIrKSEhIfT29tLd3c3Ro0eZO3cuPT09bu+RXq+XQZyoJDY1NdHY2Oj1fUtOTiY9PZ1x48Yxbtw4hoaG2LZtG3/+859paGggJCSEvr4++Uc8JyXEU0BHm5ubUavVtLS00N3dTUNDAwcPHqS+vp7w8HCio6PR6XRotVq+973vMW/ePMxmM5cvXyY4OJiQkBAGBgZob2+ns7NT9q1RroujOfjd3d3U1NTQ29srr6+np4d9+/axZ88e2tvbycjIwOVyERgYyOTJk+no6OD06dPU1taO6gfpdDqioqJoamqS78XAwACRkZGEhISwZ88et6SJCOStVqv0gex2O+Xl5eh0OsLDw/Hx8aGtrY2Ojo6/iUrtHQlaysvL+clPfkJZWRkbNmwgICCApqYmCgoKJHln6dKlAHckY+3NoqKiWLt2rVyIVCoVhYWFlJeX3/EF2uVykZeXx9e+9jUefvhhvva1rzFhwgSmTp3K888/z6ZNm/jNb35DVVXVbZ972rRpvPvuu8THx8ufqdVqEhIS3OAZer1eYl7vJF7ZZDIRHR1NWloaS5cu5dFHH2VoaIju7m7279/Pnj17CA4Ols5AVlbWHWk65HA4eO+994iKiuLHP/4xS5cuvWNzR6fTsXLlSj788EO5qLS2tpKXl0dDQwPTp08nISFBOrMzZszgL3/5yy3zWrRaLd/61rcIDAxk/fr1sgr561//moULF/6fDFo8HTKXy0VpaSkvv/wyZWVlLFy4kLi4OPr6+iTMwGQysWrVKvz9/fH396e+vt4rN0Uc3zNw8DwffMpTEIpT8fHxrFmzRkIIYDiwLisrczvejfS6EE6TeEd1Op1UshJZ1BMnTtDQ0MC8efNYvXo106dP5yc/+QnFxcV8/PHHbNu2jUuXLknH6loBtCdJX/nz+fPn8/3vf1+q07S3t1NfXy8zaGJcAgMDiYmJQa/XExkZSXR0NIWFhfI4o60t3qoqwoQMbFJSEjNmzGD69OmyQtvZ2cnWrVvJzs6WjfOampqkYlJbW5vb/Qin3vM+BwcH3YjSer1eQsO2bt2KzWbjxz/+MTNnzpSZSFE9ESacMW9rtvK+xTj7+fnx0EMP8fjjjzNjxgwGBgbIzc3lwoULuFwuUlNTSU5OlvMgISEBi8XiJoRwLfiy53UYDAb+/u//nuTkZB5++GHy8vJwuVzs2bOHDRs2jMC8i2qmWOOU1a4vCufxWtbf309HR4dXvgB82n+nr68PvV5PcHCw7CkSGBgIDM/brq6uW9pbhFiEmIe9vb34+PiQlJTEwoULWbFiBbNnz5YEeiFTbLFY2LFjB8ePH7/m8YODg2V1Vyg2icSFkBUuKipi//79MrgRlQQh4RwXF8c//uM/Mnv2bPr6+mTzwHXr1jF//nzOnTvHmTNngGGY9OOPP+4mJDE0NERBQQGtra2MHz9e3q+Pjw+5ubmSKC6c/6ysLAnTbGtr4/333+fMmTNyjMePH09DQ4MM7lpbW9Hr9fT29lJXV0dLSwsVFRVefQiz2czEiRNJTU1Fo9Fw9uxZsrOz3fbm5uZmVCqVlJw2mUxERUVRW1uL3W5Hr9fT0dEhVb9Ej5fLly9TVVUluVACGtjd3U1vby/5+fnMmDFDVo/sdjttbW3U1dVRV1dHV1cXnZ2ddHZ2uu1JFosFnU7nFmiId76trQ2LxSLHuby8nIMHD3Lw4EF6enpwuVxkZmYSHR2N1WplzJgxxMfHjyqKIp6/QG6UlZVJmehx48YRFBREQUGB2/titVqpq6ujvr6ekpIS+fn6+npaWlqIiYmRwZterx9Vwe2ztDvGaamoqOBnP/sZf/rTn/D19cVut1NXV8fQ0BCTJ09m6dKlqNVqJk2aRH19/Z06LTAMnVq9ejX33nuv3IjUajUdHR13rOLhaUNDQ1y9epVf/OIXbNmyhUcffVQGTc8//zxqtZpf/OIXbpmvmzW9Xs8vfvELYmJi5EYpSJ+vv/66lPLT6XSsW7eO73//+4SEhLBo0SIyMjKoqKi4ZWk6vV5PaGgo3/jGN3jyyScJCwujrq6OtrY2rl69ynvvvcdLL710y/d2I2az2fD39/eaDb0du//++/nVr34lq4FlZWW89NJL7N69m+bmZr71rW/xwx/+UGaAVqxYgclkuuVgbOnSpXz961/nd7/7HceOHQOGHYcJEyYQFRV1zYZT/xdMCTW4cOECRUVFvPHGG1itVnp6emhubsZut7NgwQLS09MJCAhg/Pjx0pn2rLB4/kxp3jgvApoRGBjIihUrWLVqlZsEqiAeCqUiARlTmmcGSqniJK5JyJ8qr6Wrq4vz58+Tn5/P0aNHWbVqFWvXriU5OZlnnnkGq9XKb37zG8rKykZk7USfGSH56TmmwjmNi4vj8ccfZ/HixVLuWHSEDwsLkwRak8nEfffdx7PPPjtiXqpUKhk0KQM+bzAwJfdEr9dLrsejjz5KbGwsvb29DA4OcvLkSV555RU++eQTurq6UKlUkgw92nMUGV7xc0F8Fv0ihFMnKhjKcXC5hsnKIrvtTcHNmzPvGWAPDQ1hNptZv349zz//PCkpKfT19XHhwgU++OADjh07Rl9fH6tXryYoKEiuI6IhncjIi/FTmifHRnxmYGCAVatW8cQTT/Dqq6/yySefYLFY6O7uZtq0abJy5ilQoJwz4t93g+f5WZvLNSxxLPZ3UXkR74PA9ou+Ir6+vjLQTElJISUlRR5LSCFfSwbcmwlVJZHwsNvt+Pv7M3XqVDZu3MikSZPo6+ujvLyc4OBgBgcHSUxMJCsri61bt17XNwkLC8PPz09WW4WqU0dHB6WlpTIIUalU0mlXyl23tbVRWlpKXFwcer2e+vp6ZsyYwde+9jXmzp3L9u3beeedYUonNwABAABJREFUd+S+JuDQ48ePlzB7YcIBPnr0KFarldTUVKZMmUJ0dDRhYWG8/vrrVFdX09DQwKVLl/j617/OwMAAZWVlFBcXy94u8fHxTJ06laysLFwuFyUlJXIca2pqJI/GW2U6MjKSjIwMxo4dy/bt23n77bc5f/6825gJGV8xBmPHjiU8PFyuZYIw39vbS0BAAGPHjsXHx4cLFy7Q1tYm3x+xfhuNRsaNG0dsbKx05sW1VVdXU1xcTHNzs1TXUqlUBAQEyERoSEgIVquVlpYWufZHRkYCwwil+Ph4rFYr/f395OXlcfnyZRoaGujt7ZXiHQkJCZJDHRUVdc0AW4g2CG5ud3c3PT09JCQkYDQa3ea3Wq0mMjKSpqYmKSog+EgOh4O6ujqampqor68nPj4evV4v/cnRuKOfhd1RIv7Q0NA1STo+Pj5kZmayf//+O3ZOlWpYmvTRRx8lLCyMvLw8jEYjcXFxTJgwgYSEBFpbW+/aAA8MDFBQUMDPfvYzdu7cyU9/+lPm//81t3fv3s3BgwdvOatlNBolscvpdHL16lWKi4vp6upi8uTJXLp0CbVaTWJiIvfffz8xMTG0tbVx+PBhCgoKbqniYjAYCA0NZfbs2fzd3/0dycnJ9PT0cOnSJb797W8zY8YMPvzww7tKxlKr1YSFhfHMM8/w9NNPo9Vq2bRp000pMo1moaGhvPPOO3IDLy0t5ZVXXmHPnj1S0aSkpITW1lZCQ0Opra1l+/bttxX8vvDCC/T09PDGG2+4Vca+/e1v09/f/3+uyqI0pYMl4Ct9fX3U1dVRW1vrptLV399PQEAAarWaqKgowF0JDEZySZQcE29ZdKVzPXnyZNatW0dERARnzpwhICCA2NhY2UyssbFxhMMnYGjXwvmKrsJKJ1lALUUmuL+/n0uXLlFRUcGpU6d47rnnWLJkCfPmzePAgQNUVVXJbtnCaVf2I/G8TyWW2mKxSJ6f3W6noKCAxsZGGhsbJTlXo9EQHx/P0qVLJeHz9OnTVFdXSzw9IBNSwrxVWASpPyoqitmzZ7Nu3TqJPd+yZQuHDx+WAibC6VCOo8Dui6SC8r33hKQplbBEUkeQqvV6PePGjWPdunU89thj+Pj4sG/fPolz9wbB8pxLngGx+H9qaiqPPPIIKSkp8tm99dZbHDp0iNLSUkmYX7JkiawKFhcX09bW5na93tY05ZwW59RoNKxbt46enh5+//vfy+dvsVhYuXIlfX19VFRUuMlJe87JLws0DD4NWpT8KJ1O58bjgeE1o7W1VVa3bTab5IUoA+7IyEh0Op2bGIanjdavR/k+OJ1Ourq6qKiooKmpiby8PKZMmUJYWBgWi4WKigp+/etfc/LkyeveY3BwMNHR0fT09GAwGPDz82NgYICqqiocDgc6nU5WA4SpVCoJoxUBlYA+paWlsXLlSoKCgti9eze/+93v6O7ulvflcDioqKggMDAQu90ukRwGg4HY2Fi0Wi2/+93vqKqqIj09nejoaKqrq6moqGD69OksWrRIKp7l5uZiMpkoLCykq6sLgDNnzlBeXi7XErH/Cu6NgPEp/Raj0Sgh8RkZGcybN4+WlhY2b948ImDRarWSHyIC+J6eHgkLFM+wvb0dnU5HcnIyCQkJ1NTUUFBQIKsqTqeT6upqampqmDhxIitWrGDevHn4+PjgcDjkOlVbWyuRPIIs7+fnh7+/Px0dHbS2tmI2m91EhlQqFXFxcQQHB1NRUSEllXt7eyksLKS5uVkKfjQ2NnL06FHGjx/PPffcg9VqJSgoyGvSQSRchMS1VqslKCgIk8mEy+WSDSGV611MTAyzZs3CZrPx9NNPs3LlSiorK7lw4QKHDh2isrJSBpMqlYrQ0FAZFBsMhs+tweRnrh52pwnyWq2WhIQEQkNDyc/P57//+79JT0/nsccew8/PTzbdudvmdDrJyspi7969pKamEhIS4tZN91ZMvEQwTDbcuHGjDFpERD9z5kzWrFnDPffcQ0NDAx999BH/8z//I3W7b8Z8fX1ZvHgx3/zmN5k1axYmk4n9+/ezd+9etm/fTnl5+Q0ttrdjRqORSZMm8e1vf5uVK1ei1+s5d+4c/+///b/bhr3p9Xq++93vSo5BWVkZ//Iv/8IHH3zg9rlTp05RUFBAcnIybW1tvPHGG7cVtNjtdnbt2uUmU5mamsrUqVP56KOPRpBI/y+akj8kssHCeRPwAZFBVavVsjGcUjrZ00nzJIZ7yw6J74kOyHq9nlOnTrFp0yZmzZpFTEwMoaGh2Gw2t74fyuMq+8EIEwIDogqgzOAJZ1WZ8RJNEDs7Ozl8+DBxcXFMnTqVsLAwgoOD5UYlYCDe4AHiHMpgBoahYO3t7bhcLi5cuMA///M/c+7cOcndslgszJkzh3vvvZeUlBSuXLnC3r172bZtm1TYUZ5DeY/K8won0GazsXjxYpnR1Wq1nD17ll27drFr1y5KSkrQaDQYDAY5/kqHUBl8eFaWPKFiop+M8nmr1WpsNhtTp07liSeeYM2aNajVag4ePMh7770nP+upfObp8CoDKVGJEdnX5cuXk5GRwdDQEPn5+fzmN7/hwIEDNDU1yeupr6+ntbUVk8lER0cHV69edZNaVUIHPeejZ3VEqNi99tpr1NfXYzAY6O7uZtKkSaSmprJz507q6+vlPFHOdY1GIwNAz999kU0J1xXP01tfH6W5XC4ZtAjH88qVKzQ1NaHVagkICHAjwCvNYrFcFx5jt9s5efKkRCVERkYyefJkWbHdtGkTO3fuHJFQUfLhhPn5+RETE0N/f79MfDQ2NtLc3ExFRcWo+4bT6aStrU0KRKxZs8ZN9vntt9/m3LlzaDQaIiIiaGpqktVkAZMWks2NjY1MnjyZBQsWEBUVha+vL+fPn+f99993W4Pi4+N57bXXmD9/PqdOneKll16isbGRy5cvy317YGBghFRzS0uL5AR5628yNDSE0Whk/PjxEgL55ptvjghYRIVBPD8hb15fX+/GWWpvb6ejo4OgoCApmFFbWzviumB4rkRHRzNnzhwCAgJobGzEZrNhMBhkcCUCLn9/fxmYDg0NyeqVWP/F2mYymYiMjGTcuHH09fXJ+drU1ER1dbXsBaPRaOjs7KSyspJjx44xZswYxo8fj8lkkuuJMBFEiKrK+fPn8ff3l41mYbhSVlhYKP9vMpnYuHGj9HHE3nbPPffw0EMPsWfPHl588UUKCgoYGBigra0NvV7vtsZ+XvaZBy13erF0Op2cO3eOP/zhDxQWFnLkyBEmTJgAICfFZ2WC0CSwrrersiBKr6LUmZCQQFxcHDAcrE2ePJlly5YRHx9PW1sbf/rTn3jzzTepra29pfNFRETwxBNPsGzZMmB4bH19fZk+fTo+Pj5kZ2dz7ty5UTHEt2tGo5FFixbxox/9SOJHjxw5wh/+8AepIHQ7lpSUxNq1a+UG98orr4wIWGBY110sCqmpqcydO5eCgoJb5rRs3ryZ2tpaORfHjh3LT3/6U2JiYqioqCAxMZHS0tI7Ukn6opm3TDDg9v6IBV9sPlqtVvajENkezwqLOMZo5xLfUQYWxcXF/PWvf6WiooJDhw4RGhoq5YMFxMgbud8bt0U4vMpFXqwLyoqPuD+lKpXFYiEwMFBC0ZQlfREAiXv07O0gPuf5M5Ghs1qtZGZmygxcQEAAEydOZObMmSQnJ1NfX8+f//xn9u3bJxWAlMkX5Xrq6XiJQC06OpoVK1awaNEibDYbDQ0NqFQqUlJSsNls7Nu3j5ycHHlsT86F0sFW/t6zyqKU4xSB49DQEP7+/ixcuJBHHnmEBQsW4HQ6OXLkCH/84x/Jyclx64EjghIY2chROZ7KTTo5OZkVK1ZgMBioqanhzTffZPv27W7CEIDEuItx97YfjAYPU/5eBDd79uyhuLhYBlfjx4/nBz/4AeHh4fT09JCcnExFRYUbuRc+bbIq7u/zJtLeDRO9fcS9efbYEe9fbGwsY8eOlbyEs2fPUlhYSHFxMSaTCYPBgMlk8trA0WQySYL3tfb1pqYmmpqaMBqNZGZmSl7DhQsX2Lp1K4ODgxgMBnkcpUCCpyn7x3R3d3PhwgVycnK8ZrlVqmH5Yh8fHwwGAytWrGDDhg0EBweTk5PDvn37yM/Px+l0YrVa8fX1pby8XK4VRqNREvHLysrIy8ujqqqKdevWSQik4JNcvHhRwqLUajXJyckSchcVFUVYWBhnzpy5brJPvMP9/f1e/QpRsVy+fDmzZs3i448/5o9//KPXOSxUHsVaIbi3ShPCLmL9q6ur89o9HoaTuImJiVgsFmpra8nNzUWn00k4V0REBBaLBZdrmCxvsVjo6emhrq6Ozs5OANnkVhm0CD9Ro9FI+oCo0Ilmp2IdcTgclJSUyES1VqslLCyM1tZWud4KHpMIWrKzs/H395dVJUAGqU6nE5vNxoIFC3jyySelkulf/vIX6urqmDp1Khs2bGDWrFls2LCBX/7yl7JXkUgQ9fT03DLt4E7YF75Pi8vloqCggJKSkhFyc+Xl5dTV1d1W4GCz2XC5XHISKjdH8X/RX2LhwoUsW7YMX19fampq6O/vx2azuUXeN2OigZFKNdy1+pVXXpG/E9G1KLUeP36cbdu23RaHprm5mU8++QSr1UpoaCg+Pj6EhoaSkJDAI488wvnz5/nWt741QmnoTllUVBSPPvooM2bMQKPRcOXKFd5++21OnTp1Rxz6WbNmyQ1r+/btvPvuuzf0vdsNtE+ePElrayvjxo1j6tSprF69mnvuuYfa2loeeugh0tPTefvtt/nkk0++FBnQmzFlVUVpSodfmOBwCF6D0kH09n1xfE9TQsVEICIwxWVlZRIiIGRR8/PzpeKh4EoolRFHg38q4Vk6nU42q/PkYwh+g+jWnZ6ezvLlyzGbzbKhZWhoKJ2dnTgcDjenyfPcymqFgMtERUURFRWFy+UiJSWFf/qnf5KkdaPRiMVikZ3ki4uLycvLo7KyUjoRYnME96BFbIKeAYXdbufChQsEBgYSGRmJ2WwmMjKSuLg4IiMjWbp0KT/84Q85deqUrPaIaotnrwpv96YMNAC3oFLgtJcvX878+fOxWCxkZWVx+vRpysvL3bKeQj1MqUymnDOjVc8yMzNJTk6mt7eXI0eOsGfPHjo6OiTkT7m5t7a2SlijIGR7ZtY9gz/P+xPQmU8++UQ29ktMTOSJJ55g0aJFNDU1MX/+fBISEvjggw/45JNP6O/vd+NBenJlvmzmGdCKBIGYrz4+PlitVlmZcrlcXL16VVa/nE6n5MCKeeG55/j4+GAymRgcHByRDBWVOBgONLRaLbNnz+ahhx5i4sSJqFQq9u/fT2lpqeyhJP5465QOUFhYSENDgxTZqa2tJS8vT1ZePK/Bx8cHPz8/xo0bx/Lly7n33nvp6Ojg9ddf58SJE1J9zGQyUVdXR3NzsxtvTfB9Dh8+THFxMUVFRRLR0drayoULFxgYGGDDhg08/vjjsjLicg1zxcrKyrDb7cTFxfH000+jUqn44IMPJDzMmymFUMQ76Blk+/j4kJaWRmlpKf/5n/8pK2VKc7mG1d5Ekme0wHxoaAitVisDjMrKSq9wQL1ez5QpU5g2bRqdnZ2cOHGC8+fP4+Pjw/Tp02XfnerqaqlE2N/fT11dHVVVVTKo7OnpcdvfRHAiuCOiwiNgv3a7XQaoAiYrIGmCn+fr6yvPJ+a6kP8W5/D0A69evSqTX/PmzeP73/8+sbGxHD16lDfffJOLFy/i6+vLoUOHuHjxIi+88ALTpk1j3Lhx5OTkYLFY0Gq1tLW13VDvortpX/igBZB4TE/bu3ev17LfjdqaNWtYuHAhQ0NDUmHD39+fvr4+OdEDAgJISUkhMDCQjIwM2bNErVbz5JNPys995zvfuSnH28fHh4iICDdlH4PBIDMFlZWVNDU1yQZLKSkpPPXUU7z++uuSoHyz1t7ezltvvcXp06cJCwuTi7C/vz/r169n6dKlPPTQQ3ctaBEvYnd3NzabDYvFwpo1a5g+fTrHjx/n448/vi3Vmw0bNkiH4J133vnMmiRVV1ezYMECnn/+eWbOnIlOp+M3v/kNOTk5GI1G0tPT+c53vkN9fT15eXmfyTV93iacJyV2H9yDDE+VKFFhAdy6V3s6ecrjClNuHEqomDif4EYIGJVWq5Xk3SNHjlBfXz+iiqMkRHoGXuK4wcHBzJ07l6lTp6LRaKivr6ejo0MSx8V3RDOv+Ph4+beQHf3617/OrFmzuHjxIjk5ORQUFMhN0WAwyAqgcPzFvel0OmC4j1ZAQAA1NTUMDQ2h0WhwOBy0tbXR09ODSjVMXI+LiyMiIoLFixfT3t5Ofn6+DOyEg3Wt9088h+rqaj7++GPOnDmDxWKRmUmz2cyTTz7J/PnzWbt2Lbm5uW7OtWfwAyODMm88E6WDIwIHgeVXqVT4+fmxYMECbDYbp06d4sKFC1RXV7sFRzcC43W5XJjNZjIzM9HpdBQVFXH06FHKyspQqVSYTCbUarV8HiqVSkKNTCYTGo1GOmkiyBJzZ7RxFc9zYGAAu93OnDlzeOCBB+QzfemllygoKMBqtXLPPffw2GOP0dLSQnZ2thxXZYB9rWD+i27C4RXJDQGTFHNYdFyPjIykvb2d3NxcysvLpbMsPtfX1yePoZyP/f39XsdNBCGCSwQwZswYKYPtcDg4evQou3fvpru7WzrY3irCSuGIy5cv4+Pjg81mk4qsQtlLJFaEabVakpKSWLx4MQsXLiQ4OJijR4+ya9cuKioq5Fy32+00NDS4Ke2J+/T19SU+Pp6srCyZIElJSWHOnDlYrVby8/NpaWnBz8+PjIwMwsPDsdvt5OXl8fHHH3Pu3DmWL1/O9773PSZPnszUqVM5duwY3d3dslu8pwnhBCHDrNfr8ff3x2g0SvljIRW+a9cuoqOjGTduHDt27BhRpe/t7aW1tfW61R3RGLK7u3vUJK/VaiU9PZ3p06fT29vL5cuXqaurIy0tjdmzZ0s56DfffJPLly/Ld7u+vt7t/AJ2J/YwIScs+p+IiokI3FpaWiSpPyQkhMDAQEJDQ6XiYWtrK/X19fT29spjCsGRa1lfXx+1tbWMHTuWe++9l/T0dGpqajh06BDt7e2YTCYmTJhAc3MzZWVlfPTRR8TFxREVFUVxcTFhYWFSyv3zrtR+4Tkt17Ls7OzbgodNnz5dEsGFg2s0GiV2Vvw/LCxsxHcjIyN57LHHACgtLb3p+w4KCuJb3/oWkZGRXLhwgWPHjtHZ2UlOTg4ul4vGxkba2tpISkpi+fLlrFq1inXr1pGVlXXLQYvL5aKjo4OzZ8+6/VxUXJYsWcLDDz/Mn/70J8rLy2/pHNeyyspKfv/735OTk8P06dMxm81SLeree+/FarWyadOmW95wp0+fjlqt5syZM16baY1md2LOfvvb32bVqlUUFxfz8ssv89e//pX29nZUKhW5ubm89NJLPP300/zsZz/7P8FxUWYyvT1P4SgrnTslLEpUW8TPPe16MByRGVU6rqIsLzK0drtdwsZEkKA857WcTRGQCPWZ1atXM3bsWJnwEJlFAR2wWq34+fmh1+sZGBiQzm58fDxjxoxh0qRJmEwmKisrR+2VIs4rKkeil8PkyZOxWq2cO3eOuro6Ojo6OHfunMzaWq1WCamaP38+S5YsoaSkhCtXrshqi7eKmLIXixKm5XQ6ZYdsUfER/WPi4+NZsGABCxcu5I033qCoqMjtGSnx0p7nHO05KwNPkUF/8803KS0tlco7UVFRbNy4kTVr1vDHP/6RV199VUI3PM0zMFI6BBaLhfHjx6NSqSgoKCA3N1eqyQ0NDblB+Ww2m4ShCKiKt8BstPtRnl9U22fMmMHy5cspLy/n5z//Oe+//74UUigqKuIHP/gBq1evJi8vT0rwKo/7Za20ADLR4OPjQ39/v1uQMTQ0RHBwsOw+XlxcTG5uLqWlpZIjJyoCIogRakvieXV0dIwQTxAy5mItEV3VV6xYwZw5c6iurubUqVPs3r2bvLw8t/kmYH+jmVAvFBK8FouFmJgYnE4nNTU1aDQajEYjiYmJzJ07l/nz5zNhwgTsdjs7duzg448/pq6uTopadHd3jyC6i+tQmrLqGB4eTkJCAlFRUcTHx1NeXs6hQ4dkT5jLly9z6NAhcnJyZJDjcDgoLi5m165dqFQqFi9eTE1NDefOnRvBxTMYDLL/lc1mw2azER0dzZgxY1CpVBQXF5OZmSlVEBcvXsz27dvZuXOn1zVe9EcS897z/TYajQQHB6PRaGRfFW8mSO0imSJgXb29veh0OsaPH49araanp4eqqirUarVXcrqoRonrGBgYoLi4GJvNJukL7e3t8nMi+e7j40NkZCRjx44lNTWVoKAg2tra5Bqu1+tl8snlcklxEZ1OR0BAgFeFXsH1GxgY4Pjx4+Tn51NWViZ71kVERKDRaGSPl46ODuLi4hgzZgwhISGUlpb+TUilf+E5Ldey0ZQWbtTefvtt2tvbpRrHokWLqK2t5dSpUzd1nOzs7JuGN6lUKkJCQjAajVKxx8fHh23btrl97sqVK1y4cIHKykrWrl0ro/JbUXZQq9X4+fm5VZIA2VhKjOXdmrhdXV2cPn2a/Px8tm3bJptnfe9732PZsmU8/fTTvPnmm7c8h8TGXVpaOiqOFYYx6zExMcCwlPfly5dvG5721ltvsWPHDoqKiuQCL66ptLSUHTt2sGjRIiwWy/+JoEWYJyxGyfUQGSixcQhlFCGwMRqMx5tjNhp3QJlhVWZcVSqV3PysVqtbptpbtno0/kxjYyP79u2js7OT2NhYfH19ZUbebrfT1NQk8cp6vV5WfAQERMC2mpubKSgooLi42M3p8JaUUfJfgoODSUxMlM3mhoaGKCoqYvv27bLPjcFgQKPRUFJSgt1u595778VisQCfBiai8aYIQMR5PAMMcbzQ0FCZSRTOo3CAYHitEcqI4hhi/L1VU7w5897uW0BxT58+TUlJCTabTVaRnnvuORYtWsTcuXPZtm0btbW1XrlQnudRVsQMBoOUgq2pqZHJrKGhIdlbQaj/REVFER0djUajoaamRvZq8Jwr1wpilKT9zs5OSfI+ffo0ly5doru7WzrqFy9e5Pjx4yxcuJCYmBgZECrH7nZ5ln+rJmDaokmekkcgLDk5mUmTJuFyuSSPpampCb1eT2BgoFv22ul0ys7pItj05FwI4n54eDhWq1U2GRQQYLVazY4dOzhx4oSU9r2eeZvnQ0PDCmVhYWGEh4fT2dmJRqMhNjaWxYsXk56eLiWbGxoayM3N5dKlS/T09MgATlSGbTYbXV1dbvciztfd3U1ZWRl9fX2yQaUYD4PBwJIlS4iMjCQ7O5t9+/Zx/vx5ampqaG1tRaVSkZGRwbRp0zh//jyvvvoqu3fvJj4+nszMTACpQqp8LkFBQbIHV2RkJIGBgcTFxTFp0iTJT1uxYgU2m4309HRCQkLIyckhPDx8RPNqrVZLcHCwhNUPDQ3JpsDCAgMDMZlMOBwOOjs7R93XBSTMarViMpnIy8vj4sWLnD17litXrpCWlkZ0dDTTp09nz549dHd3y3Mqz9fR0THCX+rq6qKoqIigoCD8/f1l82Llvm+z2Zg4cSLJycmkpaURGBhIVlYWx44dY3BwkNjYWFpaWkYgjIxGI0lJSXR3d48IyAYGBmhqamLfvn2cOXNGrllqtVoS9wcHB91k8ceNG0dlZSUdHR2yCvR522cStNTW1lJcXCwjts/KHnjgAbZu3XrLMKDi4mKJAxWlwN7eXhnV3qh5I/Vdz/R6vVTViIuL47HHHuMvf/nLiM8NDg5SUVFBWVkZgYGBBAYGesV83oilpaXx0EMPsXXrVo4fP45Wq2XhwoU8/PDDZGRk0N/fz69//evbgtzdiNntdlliHRoaIiQkBLVaLXlFt2rCIQoNDSUoKGjU/ijJycnExsYCUFVV5aaAcqu2Y8cOmQH3tO7ubs6ePcuiRYsIDQ39zGBrn6d5LuTCafTmOGq1Wvr7++nq6qK0tJTp06cTERHhlqXz/J63SshonAWl0wzDDlBUVBRBQUFotVruuecejh49KisTSj6Jt0qAcDxE5u/SpUtcvXpVKoSZTCZ8fHzo7u52a+4oqiTK4wmHXfR48VS48rxfQYwVv7NYLJKUGR8fT19fHydPnpQbmnDIu7q6uHr1qpS39FQgE46gpwlIjPJaU1NTWbBgAadPnyYrKwudTkdaWhpf+9rXWLVqFXV1dbz++uuSr+MZrCod6+tBEZTfVfJihIJSe3s7fX19socHICto4tmLexiNS6O0oaEh+V1BelaaIAKr1WpSU1NlwHjhwgXJu/Q0zwBN/O0pENDd3c0nn3wiOS7icwIeIgKXFStWEB4eTkFBwahj9WUyjUYjYYhWq5WBgYERe67BYCA9PR2z2YzT6aSsrIy2tjY0Gg3+/v709/dLlT34lBMmyPee67/ghiQnJ3PfffeRlJREaWkpjY2NTJo0ieTkZK5evcq5c+eoqalBp9NhsViu6TsIP8NT5ltcT0tLCzabDbVazdixY1mzZg3Lly9ncHCQM2fOyOx6W1sbJpPJDVakVqvp6uqisrJyxL0IfkxHRwdXrlwhODiYSZMmSTnl0tJS4uPjCQoKIjMzk8TERN544w22bNlCa2srRqORBx98kEceeYSQkBDeeusttmzZQldXFyEhISQnJxMUFCQTgMpeH2azmf7+fkJCQuS+HBQUhF6vp6ioSCYCBESutraWoKAgVq1axeuvvy6DL6PRyJgxY6Tamq+vL21tbXR3d7slCsxms6yIiqajAo7lOd5Xrlyhs7MTX19fqqurZeuMpKQkgoODpRjShAkTpBy+3W6ntbVVJo1HEy1qaGhArVYTGxuLn58fdrvdbW7YbDapihsYGIjT6eTo0aPU1tYSGRkp4b6eJnhNVqt1RNDidDql/HN0dDQhISFSDU+j0dDR0UF9fT2dnZ0EBgbKvc7hcHDlypU70jT8TthnErS0trZSW1tLYmKiVKa606bX65k2bRpr1qxh3bp1mEwm0tPTeemll9i6dStXr17l9OnTN31cJT7xZoOV2zEfHx8pMafVajGbzbz22msjPqdSDavyPPzwwzidToqKiq5JfLuWpaWl8cQTTxAYGMjixYuZNm0aEydOJCQkBIPBwLJlyzhz5sxnFm0nJyfzq1/9itTUVHp6enjhhRduC08pHMm0tDSmTp3qtV/QvHnzePLJJwkNDcXlGibs3464gbBrBT2idK/X60lMTOTcuXO3fb6/dfPmPCnhQSLgGBwclEF4S0sLFy9eJC0tTXb4FZ2NPTkr3uxamXThLAcHBzN16lSmTJnCjBkzMBgMzJo1i+eff55Dhw6Rm5srN16x2MPIDLYyCBPZWiWHQUCqhMMvMNHwqRKSuLbRZJLFeZX3p4RpiU704ntBQUEyIyw2VZPJxNDQEH19fcTHx7NkyRJ6e3ulAyRIseLZeOOYKFWpjEYjs2bN4rnnnmPWrFnMnTuX2NhY5s6dS3h4OE1NTbz44ots3rzZjaTqLRAb7Rkq4WDi/Mpx0Wq1sveNWq1m5syZ/OAHPyAjI4NLly6xbds2N0y/N6dFacrrE4TjpKQk2TfixIkTNDU10d/fLzlCkyZNYunSpURFRVFWVibV2K63fikrVsp7Fu+D6IouBFqEVKwgGAsnRPCZhImxEnPibilAftbm4+NDQEAARqNRNl8UxGXlPIqNjWXq1KmoVCqqq6tpaWlBp9MRExODv78/LS0tGAwGt2MPDQ03o6yvr5fCH6KBqVqtxmg0kpCQwNq1awkLCyMzM5Pe3l7Jvevs7KSnp0cqEIq1wpuJ4El0bPe2h4t+K4IrtXPnTs6ePSublcbGxsp+M9HR0QQEBMg5U1FRwdWrV2UQIMxkMhEfH09+fj49PT10dnYyZcoUJk6cKBtSNjU1SQdfSMN/4xvfoL29naysLKZOncrDDz9MQEAA77//Plu3bqWnp4eZM2fyjW98g5kzZ2I0GlmxYgU5OTmcPHlSvlNCbldwfESDz6qqKk6fPk1PTw+BgYGkpKQQGxtLbm4uERERrFixgra2Ni5fvozNZqO5uZm4uDiCgoLkc+rt7ZVcMpFoEP8WyoWi75eyEblOpyM4OBhfX18cDgc9PT1uQXBmZqZsUnn06FF6enowGo1u/LPrNSkdHBykvr4ef39/bDYbY8eOJTAwUP5ejLXRaMRkMlFQUMCpU6cYGhoiKCiIrq4ur2gasUaPdm4hBiOSgO3t7RIO3dXVRW1trRSS6OnpoaSkhIKCAremmp+3fSZBy9DQEG+88QYLFizA39+f8ePHu2GZb9d0Oh0PPfQQ//AP/0BsbKzU/9doNKxatYoFCxbQ3d1NRkaGV6zf37pdvXqVvr7/H3vvHd9mdbaPX9rbsuS994ydxCE7ZO9BgIQwwqbs0VIK7dv1vn2/3bSFltlCKRA2JE1ICEnI3sOxMxzvvYcsS7Il2ZJs6feHf/fh6LHsJMQJDW/uz4cPsS094zznOece13XdLmRmZvpl4aVSKSZMmIDnnnsOkyZNwnvvvccWhG9ilF265ZZb4PUOdnC22+0oKirCK6+8ggMHDly2gIUa+TU0NMBsNuOWW27B97//fUyYMAENDQ14+umnUVNTc0nneOaZZ/D6669Dp9Nh3bp12Lt3L959913U1dWhvLwcRqMRy5cvx4IFC+B0OvG///u/eO+994ZtNjZaxmf2v6ms8tVqwuoEWSBytsViQWlpKUQiEdLS0hAVFYWmpqaAji8QuJojPCcA5tTr9XosXboUTz75JNLS0qBSqSCXy6FUKrFy5UqsXLkSZ86cwW9+8xvs3r0bAJgzEygg4s/H95QJ5KR7PJ6AHB8KVoR8Dj4oClSVoPNRptjtdqOlpQUlJSVoaWlh35fJZFAqlZgzZw7uv/9+pKSkYOPGjSgqKvKrYgXaBPlKFkHHfL5BQqxYLMaCBQswY8YMJu+6a9cuvP3229i1axecTueQvivnqwTQcyZnnX/GFAQaDAbWA6GtrQ2pqal48sknsXDhQtTW1uL3v/89du/ezYLF4ZpL8mPJY/y7u7tx8OBBxh949tlncd111+Ho0aM4ceIEmpqakJ6ejrVr1+L666+H0+nEZ599hp07d8Jut7Nz8FW+88Ebaf4Q4Zsy8SQ6QH/3+XysAzdPQudVqoTz62o2kWiwPwbB46iSKRSkAIDMzEwkJCRAJBKhvr4ebrcbcXFxiImJYUGu0Mh5o2ev1WpZUEy/M5lMaGhoQEJCgp8crt1uZw0hqT/GcIGiTCZDUlISwsPD2fUFClpIGKS1tRXNzc0oLS1lFbbs7Gzo9XoYDAaEhIQgKiqK9e4oKirCiRMnYLPZhsyr5ORk6PV6dr/kDGs0GkRHRyM4OBixsbFD0BtGoxGrVq1CbW0tCgsL0djYCJ/Ph6amJoSEhGDx4sVMGZNU8+bOnYuCggIUFhYyh7upqQk+32Cj0Pb2dtjtdlZ9rqmpgd1ux5YtW7Bnzx6EhYUhNjYW48ePR1RUFHsHW1tbcfLkSQQHB7Px6O3tZSR43mitBb5u3k1VWApaxGIxQkJCkJaWBofDgcrKSjZuJJRy9uxZfPLJJ3jnnXfQ09OD+Ph4Bg+TSqUMHTCSUeNGsViMlJQUPPTQQ0x8ICcnB+Hh4UhPT0d4eDi2b9+O5uZmBu+rra0N6OdRZUQIMedVbwktROsIrds07uHh4bj55psRERGB3bt3s88qlcqA8NYrbVeM00JQHIVCgWnTpo1q0KJUKjFu3DhGjuKNiHJ8J9yrzf7617+irq4O+/fvh0gkwowZM/Doo48iODgY6enpGBgYwKuvvoqXX375kiBU69atQ0lJCesFA4BBW6hZ0+Wy7Oxs/OUvf0F4eDjcbjfTPbfb7fjFL36Bffv2XfLLsmvXLuTn52Pq1KkIDQ3FqlWrsGjRItTU1KCoqAhhYWGYNGkSZDIZ3nrrLbz55ptob28fpTsc3oxGI2688UZYrdYhkI7vqvFVAWCoo86/x/Tc+/v7Wb8bvV7PoJB8EDBcNUX4O97hp9+TjGV8fDz0ej1TSqHeDQD8CLv8cXlIFn9/Pp9v2M1LGGzRcajiQptMIJy78Hf8z1Q5IP6FWq1Gf38/jh8/jo8//hilpaXQarVISEjAlClTMG/ePEydOhVOpxMff/wxPvroI9TU1Ax53wgSw2/uvLIbMIhb//zzz1FdXY2wsDCoVCq43W40NjairKwMnZ2drBLAj4vQeQ8UwPCwOX7cKPMNDPZAuueee3DdddexngZGoxHFxcX4y1/+gq1bt7JqhUgkYg6MUPRhOOvt7WUqRmvXrkVubi4yMjKwfPlyFBQUoKurC0lJScjLy8PAwAA++OADvPvuu2hoaBiiODecUUDFz49ATRMp0JXJZOjv74dWq8WKFSvg9Q42CeSDSj77eiHciqvB+Comj0igf9NaEBYWhoyMDCYL29LSwpT54uPj0d3djfLy8iFBhbCFAjmjBAt0uVwoKirCCy+8gLa2NixfvpxVLgnylJGRgRMnTqCjoyPg3JJIJEhPT0dmZiYMBgOAwTkWiHPZ1dUFYGgvJqlUCqvViubmZmi1WkRERCAqKgoymQzl5eWora1FdXX1kP07LCwMIpGI7XEejweVlZUoKSlhal8TJ06E0WgM6Du1t7ejpqYGp06dQnh4ONasWYObbrrJD0LJJ1wUCgUyMzORmZnJmkPSmLhcLpSXl0OlUsFgMKCvrw9msxlerxfl5eXw+QYbPd50002YOHEiXC4Xg02dOHECarUaOp0OPT09GBgYgNlsZv1F+PeG3iM+mCd4IQkvuFwuWK1WxlmihEhwcDDWrFmDzMxMHD9+HJ999hlMJpNfZbSvrw/9/f0X9J4Dg60qTpw4gcmTJ+P222/HlClTUFlZyaCHUVFREIvFKCwsZI1rqVfTcEaSybxptVrI5XLGMWxtbWUcZuKwqFQq5OTksARdUVERLBYLOxfBmoXVuittVyxoEW7Mo2nd3d1499130dLSwl786OhorF69Gps3b0ZDQwP+8Y9/XFVcgfT0dKhUKvh8Przwwgvo7+9HSUkJyzKTKs2JEyfw/e9/H0VFRZdMkHc6nTh48CAOHTrEfhfIYbocVlFRgX379uHJJ59k2cOamhr8/Oc/x9atW0cFztDQ0IC5c+fi+uuvx7p161gmacKECRg/fjxEIhGqq6vx/PPP4+OPP74iAQt1LZ41axZef/11hvP/v2LC6spwMCQykgsWi8WsSzgfKBA/4XxzNtDfLRYLtm7dCpvNxnogZWRkICIiAiaTCYWFhdi6dSvLcJJDHqgPRiCegpB/wn+ejAIV+tvF8hD48xNOvaurCwqFAmvWrEFKSgp8Ph/S0tKYwIDL5cKBAwfwr3/9C7t27RpWCIJgBGT8xulyudh9trW1+b07wiAjUIUjkOzrcHC/QJUJguCQeEJSUhIjzB8+fBivvfYavvzyS1bJpL4IwNfwwOGcSuHzLCoqwl/+8hdUVVXh5ptvxuTJk5GWloaUlBR2fVVVVVi3bh0++ugjNDQ0+FWsLgQKN1z1TmgUFMvlcixYsACzZs3CG2+8gbKyMr/x+65UV3jzer2sfxLNY5LN5blj6enpSE1NhUwmg8lkYmIeCQkJMBqNcDgcsNvtfskIyr7zEq+UAOVhZL29vdizZw+amppgt9uxevVq6HQ6REZGIiIiAgqFAhUVFaitrR1CmhaJRIiJiUFGRgaSkpJYwKPVagPe73DPkKoSvb29kEqlMBqNkEqlsNvtOH36NM6cOTOkgk9rbVlZmV+lpbq62g/RoFKpYLFYAiaGli5dikWLFrFKpMfjQUlJCT755BNs2bIFdrsdkydPxg033IC8vDx0d3fj5MmTAZELVKHt7e2F3W6HQqGA0WhEb28v4/KNGzcOq1atglKpRGNjI1pbW5Gfn4+KigrMmjULarUaFosFNpuNNXolbjAR4imQIR+jt7cXSqUSKpUKISEhrPLT2dmJsrIyVrmcPHkybrvtNtx7772w2+1Yt24d6uvr2XjQXkRVe7lcfkGoiYKCAvz9739Hf38/pk+fjqysLKSkpDBom8vlwu7du1mfnfPZcEkyCrbp2VE10GAwsKAlOjqawa6pX4vD4WAJJh56lpqaiurq6m9lXbliQQtFvsDlkT0+e/Yszp496/e7733vewCuznL4rl27MGnSJHz44YeYOXMma3YEDDoIf/3rX7F9+3YUFBRcMkGdt29rg+vo6MDbb7+N1NRUrFixAp9//jn+93//F6WlpaNajvR4PNi3bx9uvfVWpKWl4cknn2TNxsrLy/Hzn/8cX3zxxaidbySTSqWYO3cunn/+eZw+fRr/+te/rsq5+k2MT2KM9Hfga3iOz+djTnhISAiCg4P9nHzKnPGVGp5cT1lpn89fapS+09fXh+PHj6OgoIAdNyIiAlqtFp2dnYwvQNcXqInZhTy/kYKyizkWwYKGy7r19/fj6NGjePnll+FyuTB9+nRkZ2ez+y0pKcH+/fuxd+9enDhxgglsUK8LggLwzrMwecDD3og3QUEJ398lUNVmuDlwoe87zQnaSPv7+1FWVoa3334bdrsd6enpKC4uxpYtW1BYWAiHw8Eyq7xs/UiKWrQeCgONmpoavPTSSzhw4ACmTZuGyZMnIzs7G1qtFsXFxfjwww9ZACismgiPH+i++DHjuU7kiAFfVwI0Gg2efPJJPPfcc9iyZQvefvttP8eQAibiVFyqqMh/ovl8PkYUDg8Ph1QqZWtEZmYmEhMTIZFIGIciPDwcBoMBAwMD6OrqYt3dATDSvEgk8oMYGY1GNucIAmkwGCCVSlFTU4N//etf6OnpwY033oj4+HgGv8rMzIRMJhvixIpEg6IwVJH0eDxob29ngdjFWH9/P9ra2lBUVORXhWhtbWU8K974gI9QMNSjhp+TFMyRgtnAwACrWPFzyul0oqCgAKdOnUJlZSXMZjO6urrw+eefY/PmzYiIiIBMJkNLS8t51z6xWAyj0QiDwYCmpiYEBwfDaDRCp9OhtrYWjY2N2Lt3L/Lz83H8+HGEhYUhMzMTra2tTKqeoINyuRwajYYpora3t8Pr9TKVOZfLBZVKhdjYWL8GjU6nE5WVlbBYLIiLi8Py5cuxdu1a6HQ67Ny5EwcPHmTjRFwqqrCQBDaf/CGlMAque3t70dPTg87OTmzfvh1dXV2YNGkSoqOj4Xa70dfXh4iICFRVVeHtt99GbW3tBc0Dt9sdUGAo0J7T19cHt9sNtVoNjUYDvV6P3t5ebNy4ETabDSqVCkFBQUhJSUFbWxuraEdEROCpp57Cc889961Uba9Y0FJQUIDKykqkp6ezjNTltqvZAaRGQ5dLuOA/0crKyrB69erLfh6fz4ejR4/i6NGjWLdu3WU/XyDT6XT4r//6L9x///04ePAgfvSjH/3HqHNcaSNHYDhsP+84t7a2oqCgAAsWLEBwcLAfL4IcV77HCxkdmz8W4O88B7L29na/qoFarWaZv0BOJX9cYYXhfGPAXzMfjPGfofuk/wt72BAvTSqVsl4v27Ztw6FDh5gUOmWPh6ssEBSOr0TwUCoyvpke3X8gp5z/Hjk6wt+NVHEQQu4I/gOABVakutTX14f8/HwGlWlpaWENQ/nrCVTtCARJDBRskTrZwMAAc9Q+/PBDGI1GKJVKmEwmmEwmNteElSTh/QW6BhpLCvros8TdoADGYDBg0aJFmD9/Pr766iu8+uqrLFDijwt8rRLFB5rfNevu7oZIJGJOLzUjjIuLg0QiQU9PD+uILpPJ0N7ejsbGRj8lJqrUUMBCzy42Nhbd3d1MHYoaIWq1WrS3t6OwsBBVVVU4duwY7rzzTuTl5aGurg5nzpwJGCjS3CD+VWNjI86cOfON1TmpP5LT6WROuFgsRlhYGCQSCZqbm0eEFREkizeSZa+ursaRI0dw9OhRhISEIDU1lfGHyEnv6+tjHePJ4SZFtY6OjgtOiGq1WoSFhcFqtaKnpwdKpRL9/f347LPPsHnzZowZMwYqlQo9PT3QaDSYNm0axGIxrFYrk60OCgpCWFgYC1aampr8xFDo37QekPPOvxuk7kiiCvn5+aipqcEHH3zg1wRaLpezRIJGo4FWq2WJ5PDwcAQHB0Oj0SAoKAherxcWi4WtzT6fD21tbdi+fTu2b9/O1iaFQoHo6Gg/UZQLMZfLhcbGRr/f8dA1fh3S6/WsIkiVoeLiYpSVlTElNoLNUeDa2dkJjUaDyZMnXzAEbrTtigUtXq8XP/zhD/HUU0+hurr6Sp32ml2z/0jr6enBz3/+c/z85z//ti/lWzHeGeWx9zwkSihZDAz2xnjvvffQ3t6Ojo4OKBQKP9gFD4kRkqmBr7PVwt/TddA10UYpzCQRgZzf3AJtxkJ+Bs95oZ/pGimwOZ8jSd/lx04IryMHntSjBgYGmFIM9YMh+IJCofBzjilzL6zc8kEDX9k5n+qXEOJEDngg+Vj+O3x1Q3hMqVTKxoqfMzKZjPEMSNqTjFeHEx6PniPdm3B+BDKCINH8IA4JcQ7oMxQkBApYeCgkb/Q7Ptjhv0+NR/nznDlzBt/73vcYtIXGgx8jGrOLwdtfjUYVAZL8DQsLQ1xcHKKjoyESidDT0wOJRAKdToeBgQF0dnaivr5+CBS4r6+PwZRo/OPi4lBTU8McTrlczt6F/v5+OJ1OOJ1O1gMoKysLNpsNRUVFw/ZT6urqQnl5Obq7u1FfX3/ebu5k/BpCx6JnLCRLu1wutLa2smsdDnZIyl281dTU4Pjx4wgODkZHRwf7u9frRU1NDQoLC1FbW8uaVmo0GkRGRiI2Nhbp6enIzs5mxPCWlhYUFRWht7d32LWOmj/29fWhra2Nvc9UCXO73Th27Bjje2RlZSE5ORlVVVUwmUwMImcwGFjTb4vFMoSTI1wLXC7XEPU4WotaWlqwb98+HD16FAUFBUM4jbS+iEQiVhWi+4uNjUVWVhZbW9rb2+FwOBg/hUy4z7hcLr/qikajgcFgQEdHx4jVDUIM8CaXyyGTyRhETiQSITQ0FHl5eZgwYQLCw8PZPLVardDpdBg3bhzy8vIAAAaDgfXvoxYizc3N31pR4Io2lzx48CAOHjx4JU95za7ZNfsPNApIyDmjjVTIZeAdScr2b9q0CTt37mQZMjKeC0KcBR6PTl3Zhfhs4cYvJODS98kJJJiZMGgZiYMg3CQDbdrDyRgLHXn+b3Q+PjCiDCEFJpS9IxUfp9PJOobzxj8Deg5C/glf2eLHW3gdgQQEyAEXBm0XmvXnoWDAYEacsNakRhQo4OCPzzvshEHn/81fLx/s8BWwQFh1yrTyc4KCC2EQMdI8oTHkg0MKyKi6x487VXUA+IkLEAGaPxeN1X9Cg7jLaUREJlI1Oe2dnZ2orKxEe3s7ez61tbVDuK4ejwcul4u96/xz4ruW+3w+2O12uFwuP0hOb28vaz1A3IFA5vV60dzcjIaGhouG2VA1guajy+Vi8sperxc2mw0OhwNWqxVdXV1+7wCJEtG8EYpiCK2wsBBBQUFMTdDj8aCurg5Hjhzx478Cg2pqdXV1MBqNyMzMREpKCiIiIhAfH4+cnBzk5uZi586dAeFO9J709fXBYrGwcVMqlYiMjITdbmdcD7lcjoiICERGRqKlpQXnzp1DT08PU3ej95Ya9gqNl0H2+QaVs/gmstTUm9TUSktLWdd6oRF3UCqVQqVS+e0NSqUSYWFhkMvl6O7uRlNTEzo7Oy+60qnRaJCZmQmPx8OqVhdqtOfRe6/T6ZCcnMx6nfHrsFqthlarhdFohNFoZO+C1WpFaWkp9u3b963DS0W+C7z7y8FDuWbX7Jp9M7taoY/X1pFrds3+c+xqXEeurSHX7Jr959iVXkOuaKXlml2za3bNAtlI2ecLUVEaiRfxTa8h0HkDkcpH63z0uwvhd4ymDUcUF57vQp7DhZwrEIxMeNzz3etw18I7tMK/CSFZF3M/wmoSD20MVAHj+TeBKlKXYpcyDy7XHPo2LdA7SVlv4hKEhIQwrpbQvF4vuru70djYiK6urguqRNGxqaLF9+Xgnz0PSxwYGAjYEHA4S0hIgM/3teAHLw8cSA2Mn2fEjyMoK1VMeQlsqvrRtcvlchiNRtbXiSp2PFyRH7PzzSP+/DzHjyBswGAFgTiC/DtF3DGCWcpkMr97oePQOXw+HyOW87xGgsoSfC0lJYUph5E0M6nHUWWUOE00ZlTdpLFTKBTQaDRMYZK/X4Jn8Q1ClUola9ZIx9BqtayvjNPpZJBDmk/CvkwikQgajYZJ/EulUlZxEolE6OvrQ3l5OeRyORITE/0qTHRe6tvT2Ng4BFo7nOIYz98DgJycHAbfO378+KiKJF2oXQtartk1u2ZX3HiIEMHAAPjBdWgTC/Q3oXN6Pqebb0zIO628U8l3NA70M11LIMeX/i10UOn6+GsgEzq8fK+QQARtHrIFBCaJe71eBnPgr5EfH+G48IpafFPCQNcgvHch/0J4LbwNFxwJG0UGejaBzkd/Fz5zHmbHP1u6x+GClpGCCn5M+U2eeCs0h3l418DAgB+sixds4Oe2ECoihAQKn7EQsiccR34OCuc331vnajfhfOeN4EOxsbG47rrrsGLFCuTl5bEGkfz76Xa7UVBQgJdffhnbt2+HxWIZcg7e1Go1cnNzkZaWxhSiJBIJ61ZPkBuCLfb09LBA42KCluXLl8Pn86GhoQE9PT1MzttsNvtJ45PjTlLw5ExTw1elUsmEBQja5nK5GHGe+CohISGYNWsWkxO2WCyMf0LvD60BAwMDDIo23HokkUig1WoRFBTk15meOq8Dg6pa1I+FjkViEyqVikGVwsLCEBYWBoPBwPiGBOkyGo1wu91oaGhAQ0MDmpqaWJf68PBwREZG4sSJEzCbzVi6dCm0Wi2DS3Z2dqKmpoY1tZZIJIiOjkZkZCQkEgkcDgfj1VBgFxMTg7S0NMTExDDYMADWpLSxsRFHjhzB6dOnIZVKER0dDZlMhvr6evbZ1NRUTJo0CT7foBJhe3s7urq6GKxXrVZDr9dDqVSiu7sbMpkMycnJCA4OBjDYYyc1NRXh4eEAgNraWjz77LMIDw/Hs88+y/qv9Pf3o6GhAX19fQgNDcXZs2fxzjvvMOidRqNBaGgoOjs7/ZTyZDIZQkJCoFAomPKvQqHA7373OyxZsgQmkwmpqanfSjPsa0HLNbtm1+yKG21OlMESdkbnHUihk8VnEgOR2ANlv/mgRej888cVOrBC8qoQz8vzOITck0AOJn8Ocgb4Y/EOJh1Do9HA5/OxzB05EMRRIAsUlNG18PcrrBrwHB7eYebHi65XaMIAMtB4Dpeh5R0gkgqlDDDveAe6H/4YdC90fGHwRL/nM+H0O/5ah5s3Qi4N/z1y3MRiMePY8Bltut8LDRaEwaWwMiKct8MFmHQs4Zh9l1TDaKwCYeydTidqa2vR0NCA+vp6iMViaLVa5OTkMP4RGY0LZfX5MQ8KCmINCynANBgMSEpKQmJiInw+H7q7uxkvhaoDIpEIZrMZTU1NsFqtQ8jRwvsICgpiTjJ9tqysDBqNBg0NDUyqV61WDyGVkwNOPVq0Wi0GBgbQ09ODrq4uSCQS6PV6WK3WIepx9F1gcK3r6upCa2sr2tvb0dvbOyRpwicFaJ4PV3Eh0Qha54gQLpVKWdDS1NTkNydp3fF4PHA6nYyTQxwXl8uFiIgIqFQqqNVqBAUFISkpCWKxGKGhofB4PKwvjVqthkqlgtFoZNe4detWVukhfhIvD06y2BqNhom8uFwuyGQyhIWFISIiAgkJCYiKimKKWxSIEpeFr0xQNc5kMrFKDgVlzc3NTOLaZrOxoMHj8bA5RH1rAKC6uho6nQ6hoaEIDQ1la7fNZmPCI7RPdHR0QCKRQKlUwuPxwOPxoKysDGfPnmXPm0QPFAoF1Gq1XwAik8nYd2ndIjU+CsRGu9/ihdq1oOUymFgsRkZGBpYtW4b8/HwcPnz4O7VZXLNrdqnGO5FkfAZU+L7wzmmgQCaQCQME/hzk3AqbPJIDKJfLh3Xy+Ew6HxQE2ryF2UehU8kHU0JyulQqRVZWFiZOnAiz2YxDhw4xUi1dg3AcAlVuhKR4cqLpXMHBwQgLC4PD4QjYuVsIPeGPI7ynQPdJ3+OPQVAPXgmLr6AEqkAEqtwEqmgJr50CBvq+EGJC3x0J6iAMHIT3RM6fz+djUCS3280cViI780FpoPPx5xFWBOm58QEQ/wyE9y9UQqPjfFf2ovOtAeRUNzY24vTp08jNzUV8fDzCwsIAfE3Arq2txcmTJ9He3g61Wo2oqCiIRIOCBiQTTv1JqPpgt9tZ93a5XA6r1cqOCwBWqxUtLS1+/V8CmVgsRmxsLNLS0qBQKFBVVYXq6moMDAzAZDKhtbUV1dXVbK2kykig4wQHBzNyNTBYbVKpVJg6dSq0Wi22bt06hATu8/mYU+x0OlFVVYWWlpaAzZwpWUJS54GSScI5SIpfNJYEq+Of0Ujfp+Crt7cXVqsVJpMJGRkZyMrKQnx8PLKzs5GWlsZkid1uN/bu3cv66NBaTlXO8/U7GRgYgNVqhVqtRlhYGHtfqHKXk5PDHPfu7m5WSSPVRr76R8ezWCxwOBzs3lQqFex2O6uuENmdjIJwEkshAYKuri5WQaFzuFwuNDc3M8nurq4urFu3Dv39/cjOzsb06dMRERGB+vp6lJaWorS0lD2L0NBQDAwMoK2tzU8OmsaMlPho7oWEhAypGn8bdlUGLYTvy87OZuUxhUKBrKwsREZGoqysjOH2jh49yjq6XimLiorCI488gvvvvx8ffvghTpw4MeobhVQqxaRJk3DXXXehuroa77//Pjo6Okb1HP+XLC4uDmvXrkVcXBxqa2vx73//+4IbOl2zb2bDda8Xwn2EzqkQ/kLfEWbCA1UAqIQPDG7qsbGxCAkJgV6vZ1CEyMhIaDQadHV1oampCa2trazLO8kBC6FXlIkVwskCObv83/iNQijxnJ6ejh/84AeYPn06tm/fjrKyMiareyHSvMIqh3BMaTwmTZqEBQsWoLKyElu3bkVra+uQDD0PZxI25hRCxviAiL8OYaAnlUqh0WhYJ3uqJI0UiNBmSYGYsHJDEBYKSIRVEn7shJWMQE5ToGvgYSx88DAwMIC4uDjccMMNSElJQWNjI7Zu3YqKigq/MR+pOiWs9NF98X8L9M4IoXeBKi+B5tj/BRsYGEBpaSk+//xztLS0IDo6mjnDra2tqKioQFVVFbq6uqDX6xEdHc3UnpxOJ5NDHjt2LKqqquB2u+F0OmGz2eD1etHT0wO1Wo3Y2Fi43W6YTCZYLJYL5sfQfKJKBD2zpqYm1hGev5dAqn9arRY6nY5JLyuVSmi1WixduhT33HMPzpw5g0OHDsFsNgcMLIBB1bOmpqZhe8pQNZSHkvLvts/n88vM07GpWkHQOY/H4/euUVWDVN94xUeClgGDzrxCoUBycjLmzZuHtLQ0REVFsb87nU6kpKTAYDDA4XDAaDQiKioKSUlJKC4uPu9zIKMKTFhYGDQaDVpaWtDX18cqE6GhoUzhzO12M/gYBXRUVaJrClS9JQhboHfR6/Wy6g/JUPNKZ93d3ayy293dDbPZ7PcMz5w5g6CgIIwbNw7jxo1DWFgYent70dvbyxokq1Qq+Hw+dHZ2DoF4yWQy1miTervIZDJoNBrU1dUhMzOT8Yi+DbvqghapVIr58+fjkUceQWRkJNRqNZswkZGRCAoKQnt7O2w2GyQSCY4dO4aamhqYTCacPn0aBQUFl/X6UlNT8cgjj+CWW26B0+lEdXX1ZQmYdDodnnrqKdxxxx04ffo0CgsLL3vQIpFIkJubi//6r/+Cz+djmvRbtmy5rOe9EpaSkoLnnnsOwcHBaGhoQEVFxbWg5TJbICdxOC7G+bLS9Hlh9lv4WcoUSaVSLFq0CLfffjtSUlIYN0Gn07FmW3a7HTabDV1dXWhsbERZWRlKSkpQUlKCqqoquFwu5igLIVj8OYdzjClzTv8mPolYLEZcXBweeOABrFy5EiaTCWVlZX7NRy9kwxAGGzS+FFSIRCJER0dj5cqVuPnmm3Hu3DnU1taivb3db9z5sR+ORyMMzgJBrMg0Gg3y8vIwb948JCcnw+FwYN++fdi+fTt6enr8xkI4R8g54isk/Hnp/oSBy3DjIqxEDDe2wmfIZxrpGcrlckyYMAH33nsvkpOTUVZWhrKyMtTW1vrJb/McjEBzWDhX+PEcLuAQBvfCYIWOQ1nSqz1wEYlECAsL89vzpFIp4zvwcCyfz4fW1lbs2rULR48eZVVUkgnu6+uDx+OBVqtFfHw8YmJi0N/fD6lUipCQEFRUVKC3txfR0dFoa2tDQ0MDioqK0NfXh5CQENY5nCdVR0dHIzo6Gq2traipqRlWUtjr9aK1tRV2ux1yudzPaee5NWRKpZJJ8ZLJ5XIYDAZ4PB60tbVBpVJBr9cjOzsbt912G4KCgrBv3z6YTKYhZHY+wz/cOktGsCa5XM4CLH7NIyiY8P5cLhd8Ph/rvN7Z2cnmp06nQ3R0NHw+H5OilkgkMBgMSEhIYJUjEgdISkrCTTfdhIkTJ0Kr1fq9vwqFApmZmcjJyYHD4YBSqYROp2PQuws1n88HpVKJ3NxcZGZmorKyEqWlpSzhRTwXgoJ6vV5G/ucFAuj+eaM5MlIwS0ELSUzzY0oBEl2L3W5n8ELegoKCMGHCBIwdOxYWiwXd3d1+74TdbveDIgqvUSaTweFwsHlLlZfTp09Dp9MxGeVvw666oEWtVuPBBx/E3LlzsXv3bhw+fDhgIBIaGopp06bB5/PhjjvuQHBwME6cOIGXXnrpsvWKiYyMxEMPPYS77roLnZ2d+OUvf4mjR4+Ouq61RCLBxIkTsXz5cgCDXVezsrKwb9++UT0PmUwmQ05ODu68807MnTsX48ePZ3AIvV6PY8eOMc33q9FkMhmioqJY2V1I1Lxml8cCLXpCp54y8jwkjLDnlDUm2A3BjgD4YbfJQec3gKSkJCxfvhxTp05FZWUlvvzySxQXF7MECFlERATDr8+bNw933nknzp07hzfffBM7d+5Ef38/I2ILM2qB4FPny+T7fD5ERkbijjvuwKpVq9DQ0IBXX30VX375pV+3aqEgAD9+fOA23PhS4DZlyhSsWLEC0dHR8Hq9yMzMxMGDB1lAJYSFCY8rdFJ4h5s2NnK6jUYjpkyZglmzZmHRokWIj4+HRqOB3W5Hf38/8vPzWdaPV2XiHX4e0gUMvruE4afjCCEn5NSFhITA5XKhqamJbdb8fBspMB4OukLX4HK5oFAokJGRgczMTOYsKRQKiMXiITC44YIWYTBFc5cw7jyvgDLb5CzRM+MhYETI5rPi3wUTiUSIjIxkuH/KHqvVarhcLpjN5iEVCpvNxojzgYx4C0FBQWhtbWVJCd7BlkqlLGttt9uh1+sRFxeHuLg49Pf3o7q6Gh6PBzNmzMCKFSsQFBSEHTt24NNPP0VjYyPkcvkQ6BVBki7knpOTk/36vlCVha6ZGshGR0fjhhtuQHJyMl588UVs27aNEbqBrx1TAEO4cYGMPk9jzCcFKDARNrUk02g0rE+LXC6HxWJh8zA3Nxc6nQ6NjY3o7e2FSCRiPV6SkpJYFUWpVCImJgbXX389cnNz2e/pfabqeEJCAubOnYuSkhKYzWao1eqADXNHMqlUitjYWMyYMQNTp05FQ0MDjhw5gpaWFjgcDrS0tLCKFgUotBf19vair69vRMGFC62+BRpLqk47nU50dHSgr6/P71mShYWFITs7Gy0tLdi6dSs2btyImpoadu5A8D/+3NTri69i9/b2orS0FG63GyqV6ltLfFx1QUtfXx/Wr1+P/fv3Y+PGjbDZbAGbN8lkMuzfvx8ikQi7d+/GP//5TyxZsgTp6emYM2dOwCzGpdqSJUuwZs0atLe34+WXX8bOnTv9SF6jZQqFAk8++SSLrmUy2bByjpdqCQkJePrpp7F48WJERUUx5Q06b0ZGBqKjoy86aAkODsbDDz+MkJAQnDp1CocOHUJ7ezu0Wi2mTp2KxMREAMCCBQuQkJCAmpoaPPzwwxe0uF+shYaGYsGCBezngoICVFRUXPJx09LSGFaZN8r8HT9+/JLPcbUaBQZCZ5AcK9oEyBlTqVQMw8sTtQH4OYCUIKCMnTAjT5ulw+HA0aNHcfLkSRw7dgx1dXWMe0BOJnW9ViqVUKvVmDNnDv7f//t/WLx4MUQiEbq6ulBQUOCnNsZfz3DGO5aU+ab71Ov1WLFiBe68807Y7Xb89a9/xRdffDFk3gdyQAMRyckCbTAajQbz5s1DXFwceyZERBZWKEaCTgmvg7K5fMCSnp6ONWvW4JZbbkFCQgJ0Oh17Pnq9Hnl5eYiLi0Nzc7Nf88pAHeX5Z5mZmYm8vDwYjUZ0dHSgoqICJpOJOTDh4eHQ6XSIjY1FVFQU6urq8Pnnn+PMmTOsKiOEsZ3PhBBAMrlcjujoaKhUKvT19aGxsRHt7e1DRCZGGlchDDIoKAiZmZmIiIgAycCKxWJ4PB50d3ejs7MTZrMZZrOZvRvCCtVwwdbVbCLRYPdxg8EwpLJH+Hyz2TzkXpVKJXt+PFSRVKOuu+46qFQqmEwmBk/ikxh8xp6aOpKD5/V6GSw9KioKvb29mD17NsaPH4/ly5fjj3/8I06cOAGRSDQiMX84S09Ph16v9/NdRCIRq/QQjyMoKAizZs3CjBkzcODAAWzcuBFms5kRxSkhQNWSC7keUrTSarXQaDQM2kvwKKo6CN+LqKgoTJkyBfPmzcOYMWNQW1uL/Px8FmDPmDEDnZ2dMJlMLCiMjY2FwWCAzWZDY2MjgoODMXfuXNx4442IjIyETCbDpk2bcOTIEZSWlsLpdOLee+/Fvffei+DgYMyYMQMVFRVMKvn06dMjOum8SSQSpKamYsWKFRg7dixCQ0Oh1+sRHByM06dPo6qqir1rfOVUIpHA7Xaju7sbFosloJ8yHI/tfEZoIo/HA4PBAK1Wy4JyCpT4hJZWq0V2djY6OzuxdetWbN++HQ0NDUMapY40BsIGy5Q0oWCJFPK+Dbvqgha3243PP/+cldmGM4/Hwx7k7t27cfr0aSxduhQZGRn4/e9/j8cff3xUy1tLly7F008/jYiICHz88cfYunXrRckbXozJ5XIsWrSI/WyxWC4LlCkiIgIPPPAA7rzzThgMBrYRdnR0oKurC9HR0czRAQY77er1ephMpvNK4RmNRvz0pz+FRCJhmYn+/n6IxWLWVRYA0yYfM2YMFi9ejE8++WTU71MqlUKv17OfhyubXowtX74cP/nJT5CWljZEZYOqVDabDS6XC5s2bcI//vGPq7padbE2HC4f+BrCFBYWxoiOvb29kEqlUCgUDG4gkUig0+kYJIQ2Ego4RsogtrW1YePGjWz+AV93Seaz1uR8t7W1YefOnZgzZw4efvhhXH/99Zg6dSpqa2sZRGW4gEHomJIUJf2NAjSVSoV58+bhwQcfhNFoxCuvvIIdO3awDCkfrAmhaCOdf7jrCQ0NxezZs+HzDSrOmEwmVFZWMgLz+TgfdO10PTyEiZ6vRCJBVlYWbrnlFtx1112s90RXVxe6u7thMpmQnJzMssUejwfh4eGIiopifTN4o3Hw+XzQaDTIyMjA6tWrMWvWLJZNb25uZgEEZYUpkKqrq0NxcTHOnDkzJEgJBEcL5PgPV21SKpVsHent7UVjYyNMJpPf58/nuPDno/nw8MMPIysrCxKJBAqFgl0zzduBgQHU1tZi69at2LRpE6qrq4dwXITH/y4YJRisViuDhNPYklQvmUKhgFKphNvtZuMml8uRkJCAuLg4xMbGYt68eVi4cCHKy8vR2NjI5GBpL2ptbR2SsSdOCyW5aA+rr6/Hhg0b0NjYiFWrVmHatGl499138Zvf/Ab//Oc/2ffFYjHCw8OhUCgYjCeQRUVFwWAwoLu7O6DaFnHyzGYz4uPjsXTpUgwMDLBr4JNA9D3imcjl8hH3O0qKqlQqKJVKKJVKVr0jZTBKMJBJJBJkZmZi5cqVWLFiBbKysqBQKNDQ0IDW1lYAg+9bYmIinE4nIiIimDKaSqWCzWZDZWUlzGYzsrKyGB1ALpfj+eefx4svvoienh54PB7k5eWhra0Nx44dQ2pqKlJSUvDTn/4UOp0OIpEIGzduRGVlZcDkNm9isRipqal46qmncNNNN0GtVqOnpwcajQaJiYlsn7HZbKwKQf1QKNHV1dWFtra2gBW9i/U3jUYjVq1ahUmTJmHLli3YvXs34xSRQhj1puHhggaDAQaDAevWrcPx48f9IIcXYjyqAfg6CUXPWiqVjjp66GLssgQtt956K5YvX45XX30VJSUlSEhIwIoVKwAMTtSIiAgsXboUAGCz2VBTU4OmpiZ88MEHOH369HmPf7EOpVKpRHZ2NoDBBb6kpGTUAhaxWIzZs2fjV7/6FdLS0vCvf/0Lf//73xk+c7RNJBJhypQpTCEEGJRGHG1eiUKhwM9+9jOsXbsWwcHBEIlEaGtrw49//GOUlJSgubkZa9aswXPPPYe0tDQAwPe+9z3MmTMHv/rVr/DZZ5+NOMZWqxXr1q3Dvffei5CQEISEhAz5DO9UEKbychlt8na7HU1NTX6LwDexhQsXYurUqWhsbER3dzc0Gg1sNpufAzhmzBjI5XIkJycjISEBjz322Hemh8L5LFDAwmfnFQoFZs+ejdTUVBw7dgylpaVISEjAuHHjEBERAa1Wi5CQECZ/ScpXtbW1OHbsGI4fPx7wGVLWlIi45yMn82tNVFQUYmJi2HEoULoYo3Pxz5kwyhMnTsSjjz6KzMxMfPDBB/j3v/8Nm802LOlxOGgRWSCoGI3zwMAAMjMzkZyczBIGe/fuRWFhoZ84AA81CnR+Ov5wjnFqairuuOMO3HbbbUhISIBEIsGRI0fw1Vdf4auvvkJzczPuu+8+LF26lMEZHn/8cWRnZ+OVV17Bli1bhnU2nE4nKisrcfbsWaSlpSErKwtGoxGxsbEAvu6hYrVaodFoIJfLmSKQsLpyvirLSGs5wbdIOpTGi/gSQoI8r2QX6Fh0fcHBwRg/fjzGjx+P0NBQNDU1oa2tjQXWEokEQUFBiIqKwqxZs5CTk4P4+Hj87W9/Q01NTcAg7Jtme/8TjRI/tbW1jAsVyMRiMYMJkSyxwWBAeno6Jk+ejJycHGRmZiI3N5c58KTYpFQq2XiVlZWhs7Mz4DnofZbL5YiKikJGRgZCQkJQX1+Pf/zjH1i+fDlmzpyJ3//+90hISMCzzz7Lri06Ohrx8fHo7u5GcXExI/6TKZVKFrAQkZrM5/OxBKFSqYTRaMTChQsxduxYvPrqq9i9e/ew/hIlzwgJQA0LieNDv1OpVFAoFH58GILaEp+Dlw5Wq9WYMWMG7r33XixYsAAhISGQSCTo7Oxk85fOb7fbWaKC+pm4XC4UFRWxiobNZkN4eDhkMhlOnDiBPXv2YMKECbj++usxduxYxMfHo6GhARs2bEBERARuvPFGBAUFMQlhgo6dzzQaDZYsWYJbbrkFSqUSx44dQ1tbG+bOnYu4uDj2XFtaWmCxWFjQR5W7np4emM1mdHZ2XhLChqpnK1aswI9+9CMkJSVhYGAABQUF7Hn19vYygQAeHqbRaJCWloaCggLk5+dfsD8hEokQHx+PiIgItLe3D+FH0xocHR0NkUh02fzbC7HLErTk5eXhxhtvZBN+0aJFfpg7nsjo8/kwYcIE+Hw+lJeXX1DQcqGm0Wgwf/58PPXUU4iPj4fX68V7772Hf/zjH6NyfMKYPvDAAxg7diy+/PJLvPHGG2hoaBiV4wcyCpLIzp49i//5n/8Z1chXLpfj8ccfxy233MIClrfeegsvvfQSSktL4fP5oNPpoNPpEBMTg1dffRUOhwNhYWGoqKiAxWI574Tu6urCc889h1deeYUFsGq1Gkaj0a+b7L333oucnBxIpdLLKqJAGavhyLsXa9u3b4fT6cRnn32GqqoqaDQaWK1Wv6Bl/PjxSE9Px/PPP4/p06fj+uuvv2y8pP8043X/eTgQjY9cLkdWVhbuvfderFixArW1tUhLS0Nubi5z9sjZ5/kdJSUl6OzsxKlTp/zUychRJ4ItQWx4ngs5nW63m3WhBgYzV9OnT8fdd9+NJUuWwOv1Yt26ddi2bRtsNpsffyOQgy/M0tNxeXnM9PR0PProo5gxYwY2btyI1157DZWVlUOy9BfjdAorIfQ7Go+FCxey41ksFuzZswdtbW1DIEr8fQidYP5doXGgTF1QUBDmzp2Lu+++GzExMeju7saGDRvwl7/8hVV0goODodPpkJaWhqeffhoPPPAAUlNTcfLkSdZXgjeqstG1FRUVobW1FTt37mTKQVQhk0qlsNls6O/vx+LFi7F48WI4nU6m6MOP0Td55/kGki6XC3K5nK2XfEdtGs9AlSr6O3891LzS6XTiyJEjaG1tRUNDA86cOYOuri52f/TZ0NBQLFu2DE899RSmTJmCyMhIFrSQ0dz5LkDDeLPb7SMGLADY/Aa+buqXnp6O2NhYRERE+El+HzlyBJ988gkKCwtZI0Gag1VVVcOeRyaTsWMRf4qUq4izcfz4cUybNg3f//730d/fzwRt+vr6oNfrERkZyZx0PtAwGAzw+XywWq0MtsNDX51OJ1pbW6HRaDBhwgTMnTsX+/fvx0cffeQHGxpubOielEolEhISWPaeSN88l4oCHeLyUZNHq9XKBERuvfVWrFmzhlUHeSMOEPlIjY2NsNls0Gq1DIrldrtRVVXFxj0mJgZqtRoikQgtLS145JFHsGTJEr9ml1KpFPv370dhYSFOnz6NEydOoLGxERMmTMCSJUsuyHknaLrBYMDx48fxxhtvwGw2Q6FQYMWKFdBoNEhNTWXVdYvFws5PylxUhfmm7xmNcVxcHFauXIn09HRW1eH9ZkIVyGQy6HQ6qFQqAF9XH48dO3bB61l4eDj+/ve/Y9asWXA4HPjss8/w4osvMogbzTlSgWtvb8f+/fu/W5yW+vp6uN1uTJkyBUVFRSgsLMSYMWMAfC3Z1tvbi8jISACDE3fLli3Yv3//JZ9bpVIhMTERGRkZePjhhzFnzhzIZDKYzWasX78ezzzzzLBKHhdrer0ea9euxR133IH9+/fj97//PYqKikbl2MOZTCbDT37yE/azxWIZ1UAPGOSbLF26FOHh4RCJRDhw4AB++tOfwmq1IjMzE5MnT8aPfvQjZGRkQCQSQafTQSwWY926dXjllVdQXV19QS+M2+1GZWUl6+7Lb7L0/RkzZrAq2eWI7CUSCdLS0jBz5kwAYI6HkIdysbZ9+3bs2LGDXXOgrH9hYSHGjRsHvV6PyspKFBYWXtI5ryYjB4pwx7QBUqnd6/WyzZxkJwmL7fF40NHRwaoc5MR1dHRg165dOHToEKxWa8BFlYc+8RsewcromCEhIUhMTERmZibWrFmDxYsXQy6Xo7S0FG+88Qbee+895gjxpNSRbDj+gsFgwK233oqVK1fizJkzePPNN1lygDch/IsnawcyCsQJ6kbf9/kGSfE33ngjw8CXlJSgqKjIr+LAn284iJgwi8/LAIeHh2PChAnMEdm8eTNeeOEFNDQ0YO7cuRg3bhxmzZqF2bNnQ6fTARh8H1977TWsW7cOFRUVQ4IWCvJ4TlBHRwc6Ojpw8OBBFqwQVt/pdCIkJARJSUmYP3/+iFwduvfh+Du8KRQKv+ctkUgQFRWF2NhYP87SSEHLcNAzgjXabDYcOHAABw4c8AvWtFqtXzac4CHU2ZwCG7oGysTSXKD37Wo2sViMpKQkBumkjt0knUtZeq1Wi6SkJMjlckRGRuLmm2/GokWLoNVqGdRKoVDAarVi/fr1+PDDD1lGm6rjtCYEmhNi8WBTQwpO7HY7WlpaWFPLqVOnYv78+cjLy0NxcTHeeecd3HfffXjmmWdgMpnw4osvoqurC263G6GhoUhISEBjYyNz6qnJH1V+qApHMElgMHlTU1ODhIQEXHfdddDpdPjiiy9QWVkJmUw2hMsg5HHRXFAqlYiPj4fH44HdbkdXVxdbR2n9IMg7zWkKGgmSdscdd2DMmDGw2+2ora1llUCCNtKaROeuq6tDT08PoqKiEBkZyXqB8MiKCRMmIDg4GE1NTUhNTUVmZqZf0pugYUePHsWhQ4f83rO2tja0traed76rVCrMnz8fwcHB+OKLL/DRRx/hq6++gtvtRmJiIqZMmcLUz1JSUpiPSzA7EmY4nzLY+czn88FgMGDRokXIzs5GdXU1ioqKsHfvXnR3dzP5YuJ0qVQqJnUNDCKXDh06xNZ8YGRY2uLFi7F+/XrI5XKUlZUhNzcXS5YswbZt21BSUoKBgQG/AFqhUGDv3r3nhdpdTrssQQvpXAcFBWHDhg348MMPLzu0Jzw8HOPGjcPatWuxcuVKRlIvKirCiy++iMLCQpw7d27Uzknyv3fddRdaW1uxZcsWlJaWjtrxh7PZs2f7yaRejmjXbDbj8OHDyMvLg8FgwIEDB7BkyRI888wziI+PZx1mvV4vrFYrtm3bhi1btmDHjh2XBKsaDt5yOU0ikbDFEvhaJWU0mieN5PjIZDLcc889+OEPfwi32438/PxvdSH4NoyydryRY2Wz2XD27Fm0t7cjMjISn3/+Od5//33Wx8DtdkMkEkGtVkOhULDsFuFthe8Fj72mc1NlgCAJlFnNy8vD3LlzMWfOHKSnp0Mmk2HHjh1Yv349c+7pWQ0Hnzof4ZqcXplMhtzcXCxevBitra147bXXcOTIEYYdJiedMpsjmRAOxgcdQlGC8ePHIz4+Hh0dHVCpVDh69KgfNyfQs6JzDHefgaqUNpsNHo8HYrEY3d3dTDksJycHYWFh0Ov1cLlcOH36NI4fP459+/Zh9+7dw8Jw6Py81DRV0KjCI5S8VSgUiIiIYA6u0KkIJEvKQ+Tofvj5wzfGJGz72LFjmahBf38/ent7/RzG88H5Av1eqGBFBFiaC2FhYbjppptw7733AhhMltTW1vrh0mlO8BXNq914p83r9aKhoQF9fX2My0LrQHZ2Np566ilMnDgRMpkM8fHxzFnevHkz4xE2NTXh+PHjqKmpYWPtcDhYQi2QSSQSZGRksOaGzc3NjIdBx6ReL3PnzkV4eDh2796Nhx9+GC+99BJ+/etfw2q14quvvkJfXx+USiXCwsKY7D4AJnxD90XJFp6HQo5zXFwcxo8fj8LCQjQ0NCAjIwNtbW2swk+BnsfjgcViYXs139E9PDwc/f39kMlksNvtfmtKb28vCz6oeiWRSLB8+XI89thjmDx5MrxeL6tupqSksOfjcDiwc+dO7Ny5E2fPnmXXXVpayhA51MiQyOwikQhZWVlITU1FdXU1enp6kJmZyfyfjo4O7NmzB4cOHcJXX32FqqoqAEB8fDzmzZsHl8uF+vp6nDlzZljhJbFYDIPBgNtvvx2TJ09GSUkJdu/ejaNHj7LxOXXqFAoKChAeHg61Wo3U1FSUl5fDarUyJTkSIwhUjT7fPOY/S/sBcZaLioqwfft2HDlyhK0FpFRIPCaJRMKulZJ9arUaQUFBbP4Jz5mSkoL7778fP/jBD1BeXo4//vGPcLlc+O///m8YDAYkJiaivb3djwOWmZnJhEXWrFmDjRs3fitw9ssStLz11lvIzMzE9773Pbz11lvYvn07urq6LttiOW7cOPzyl7/EokWLhqhoffjhh/jqq6+G4EQv1UJDQ/HEE08gMTER//73v7F+/XqIxWLWPOlylc7mzJnD/t3e3o533nln1M9BihT0vJ577jmmkAGA4Vd37tyJbdu24ZNPPrksk5dkD2kDuhzzJzQ0FA888AD7mTK2TU1No34uMrlcjgcffBB//vOfIZVKceTIEbz55pvfCWfiQi3QfOGDcZ/PhyNHjuC9997Dj370Izz66KPIz89HWVkZtFotgMEMs8Ph8JurBP/SaDRss+c3Bd55J6gWyRzn5ubiiSeewA033ACj0ci+43a7cfr0aRw4cID1XaKMrpAvcj5HlCA95OAnJCTgnnvuQXZ2NjZv3oy6ujpERkaiu7sbDoeDfT7QeF1s5ZGy7P39/Vi4cCGAQUhmd3c3Tpw4wTDafGWGnksgiV4eHsU3cqTAoqWlBeXl5ayfxfTp03H77bcjJCSEVRQqKiqwbds2HD16FMeOHUNDQwNkMhkTXAhUnaAKBo0nVVf4gIkgg16vF7GxsSz4bGtr8yP3f9PqrXB9V6vVTHlqYGAATqeTwXl4CI5wjguNhwB6vV6m2kPH4IN8tVqNhx9+GPfccw8SExOxd+9evPXWWwETAUKY39Vu/f39OHz4MKtseTwedHZ2Ms4KVdoIwhMfH4/8/Hz885//RENDA06ePMngXjKZDAaDARqNBllZWYxQbbfbR6ygxsfHIyMjA8HBwejs7GQdznkuGMEgrVYrcnNz4fV6sXfvXjzzzDN46aWX8PTTT6OtrQ0+36ByYFBQEJPep3lNKlF8wO7z+ZhgjVQqRVxcHO68807k5uYiKSkJU6ZMgUgkQlFREd566y18+eWXbM5pNBp4PJ4hCUapVAqdTgePx8MCD4LQUuJBKpXCYrGgo6MDCQkJeOqpp/DUU08B+HpOG41GGAwG9Pf3o7m5GSdOnMCRI0dw9OhR5Ofn+yVfGhoaEB4eDrPZjIaGBrjdbtTV1aGmpgY5OTlYtGgRRCIR9u3bB5FIhJycHDYOBw8exHvvvYft27ezOW00GjF37lzMnz8fDQ0NOHfuHIqLiwM+Q6lUioULF+Lmm29GXFwczp49i02bNqGwsNCPq9jQ0IBDhw5hxowZTPKZIG4kvEDrEKlcUrXqfCZ8F6OionDLLbcgNTUVbrcbXV1dOHv2LBMEINNoNNDr9Ww95/lxWq0WCQkJGBgY8OPWUJJu5syZ+NnPfoZp06bh/fffx8svvwybzYbx48ezfYZEbux2O7q7u5GUlISsrCwcPnwYq1atwtNPP40vvvjiuxO0AMDPfvYzNDc347e//S22bt2Ku+++e1RkZANZS0sLDh8+jOTkZERERLDfS6VS/Pa3v8Wdd96JlStXjhrXhLgyBK9YuXIlFi5ciLa2NuzevRtbt27FoUOHRl3uWCQSYe7cuQAGJ7vFYrnszTIBMKiUzzeoHPaXv/wF+/fvx+nTpy9bcCaVSpGXl8ekPnfv3j1ESehSjTazxMREP9zu5TSlUonvfe97+NOf/gSpVIqTJ0/i8ccfvyJVuv8ko3HmORDEBaCNsq+vD2+++SaUSiWefvppvP7667jvvvtQV1fHjkP9Wgh7Tc4rz1sgE1YL+J/7+/vR3t6O4uJiJCUlITo6mhFcg4OD8eijjyI3NxfPPPMMKioqmDNKJuSajOQYklMZHByMNWvWYPny5VCpVFiwYAHGjx+Prq4unDp1Cl9++SWOHDnCnAviBQ5XceGhXULjyd9isRhLlixBb28vVCoVioqKUFJSwhw4yhgGqq4Egjfxks3CKgHB/QBg/PjxEIlEcLvdKCgowFdffYUtW7agvr6eqfIIn4uQxB7omfKNKIX4e7VajfT0dCQkJEAsFqOsrAxtbW1D7os/3oUEovQMKDiKjY1l/avIWSBojXCejFQhFwYYfHWMMuxExn3ooYfw05/+FDKZDPn5+fjxj3+M5uZmJhJBTRSFFZfLvcZdCSNY0MDAAMLCwhAXFwexWIzOzk54vV4olUoMDAyguroaxcXFuOGGGzB9+nQcOHAAJpOJVcTsdjuSkpKQnZ3NuBM9PT2ora1FbW0turu7UVFRMURwgyo9brebVVREIhH0ej1CQ0MRFBTECNtNTU1+725/fz+DDy9YsADLli3DkSNHoFKpEBwczIIW/vnxzWFpfpAYj1gsxnXXXYdx48YxTkVvby8T95g5cyZef/11/OY3v4HdbodMJhsWEUFVRoJ9EW+Q1ti2tjbIZDLMmjULP/vZzzB79myWRCBlOwqIDh8+jE8//RQnT55kilqB3iWRSMSCxPLycjQ3NyMiIgK33347oqKiUF5ejmPHjiEjI4M1lPR4PDh06BC2bdvGrluhUCA2NhYikQhVVVWoqKhAa2vrsO9aXl4eZs2aBa1Wi4MHD2Lz5s0B0Tjd3d04c+YMampqEBISAplMhsTERBQVFaGnp4cpnlGDUbFYPKQVx3DrCP/7sLAw3HfffVi0aBE0Gg3cbjcLhIlrSfdJwiI0H+gdVygUSExMhF6vH+IvhYeHY9GiRXjqqaeQnJyM3//+93jnnXdYNUUkErF9j5qzU9+ppKQk7Nq1C9OmTcOf/vQnbNy48bvFaQEGX87169fjhhtuwOzZs3HzzTfjz3/+82W50Y6ODrz00kv47LPP2AsPALGxsXjttdeQk5ODSZMmjUrQotPpsHbtWvzmN7+BTCZDX18fWwCio6Px8MMPY+HChXjwwQdx9OjRUc2eazQaTJ48GcBgtWPPnj0oKSkZteOT6XQ6hISE+OnTA4PjvHr1ahw7dmzUzym0pKQkPPTQQ4iMjITD4cCuXbvOK6N8sSaRSBAXF4fExETmcF3uLOSSJUvwt7/9DSKRCPX19bj77rtRXV19Wc/5n2g8pIiqeISf5SEQZrMZzz//PJKSknDbbbfhwQcfxG9+8xvmRJDzSA4ZHU+hUPg1O+OrAfQzOYJUyWtpacG7776LXbt2sQ0IGMyo/va3v8WyZcuwadMmlj3lg58Lec9pgxKJRAyScPfdd8NoNLISvsFgQEREBCZPnozx48fj97//Pfbv3896MJwPIhboOviN0ePxICYmBmPGjGFZ4cLCQjQ2Ng7BmwNDFa6E/Av6HX9v9PzS09Mxbtw4BhGhzx49ehSvv/46tm7dOiQTSAEkbcLC/WK4zZ8/N73HXq8XISEhGDduHDQaDUwmE06dOsUgQXRvI/F0+L8HuneC8CUnJyMpKYll5p1OJ2w2W8Dr56tEwvPx48ifh4zUgmbPno2nn34acrkcx48fxy9+8QucPXuWzRGe90PznO73u0DGJ0g4OcNBQUHQaDRwOByw2+2s0ajP58O5c+dQWVmJ5ORk/PCHP0RoaCgKCwuhUChQUVGBrq4uyGQy5OXlITo6Gk1NTXA4HKiqqkJPT8+w7zYpOJHwjFgsZn1CVq5cyar2XV1dDLJoMpngcDjg8XhQUlLCVMUKCgpgMplgMBhYktDj8TAFM2HFjf5zu93Izs7GDTfcALlcjk2bNmHv3r0sO//YY4/h1ltvZcIYPAdIaAQtstlsMJvNcDqdjNhtsVgwMDCAmJgY3HfffbjvvvsQFhbGiOrUJ4bkn7dt24YdO3aguLh4RLVXuVwOYDDBQV3be3p6MH36dEyePBn9/f0oKSlBa2srUlJSmEBFQUEB9u7dy6pqRqORyV8fP34cJ0+eRHNz84iJTofDgdOnT+Pw4cM4ffr0sOiKgYEB1NXV4cSJExg7dixkMhkiIyMRHByM1tZWSKVSaLVaBu0jqGB9fT27R4VC4cdDA8DEBRwOB7RaLW699Vbcf//90Ol0kMlkqKmpQWVlJVtjPB4P6xnG91JxOBxMKpvQPi6XC3a7ncGigUFF0//6r/9CREQEfvGLX+DDDz9EeHg4enp6GB8nJiYGVVVVcDgcsNlsyMjIgFwuR35+PtLS0vDCCy/AYrHg/fff/9Z4cZcctKjVaoSFhWFgYIBNPHIompqasGzZMrS3t+PXv/4165dyOUpK/f39aGxsRGNjI/tdUVERGhoaEBsbi8WLF2PDhg2XdA6lUolbb70Vf/rTn6DRaNDV1YXt27dj37598Pl8SExMxJw5czBu3DjceOONKCkpGdUmlqtWrWL/drvdKC4uHrVjkwUFBeH222/H6tWrodfrWZZHJBJh06ZNV6whosFgQHBwMCQSCQ4dOoS33377sswbIXeF1IUul02fPh0+nw+NjY3405/+9H8yYCHjM/RarZb1AqJn4HK5GGn4lVdeYQpee/fuxeHDhxmWmyfV0wIt7NEidDx4iBifze7s7ERnZ6efQ6nRaHDzzTcjOjoaUVFRrBpBsIDh5qWQY0IOqV6vx2233YZnn30W8fHxaGlpwY4dO3DixAlYLBYkJCTgxhtvREZGBhYuXIhz586hubl5yD2NBEfjiff0OQrclixZwuZ9f38/Kisr/foKCMnpw52PfuaDQRp/o9GIJUuWYObMmVAqlTCbzQgKCmJO5IEDB1jAwj87/roDGR+c8E4c/3denS0+Ph55eXlQq9U4dOgQ8vPzmRPFP5uRxnO4MaCKBnElqNu5VCqF3W6HyWTyg2oJg5/h7pGH4lJFj7/W4OBgLFy4EPHx8SgpKcHzzz+PQ4cODTkPP1+EFZyr3Xy+wV4/7e3tjBumVqvhdDr9nCmZTAar1Yrnn38ekydPxv33349HHnkE27dvx+bNm9HV1YWmpibU19ejvr4eTqcTDQ0NaG9vZ5Co4c7f3t7OegsRHIh65ojFYixfvhzjxo1jvsn+/ftRUFDAgiQSsFEoFKiurkZ5eTl6enpYEpTWFYJL8sEMiYZIpVJMmTIFM2fOhMViwYEDB/D555+jp6cHMTEx6Onpwfvvv4+NGzdCJpNBr9f78cWIn0KQ9u7ubnR1dcFms7F77+rqglwux7x58/Dss89ixowZMJvNqK+vR3R0NBQKBZqamnDmzBkcPXoUxcXFqKysZA17R7Le3l4oFArodDoYDAYW+FDVPC4ujjWg1Ol0OHr0KNrb2/Hpp5+irq4OERER0Ol0rG/JxYgsVVdXo6OjA729vedFxZjNZuTn5+OWW26BRqOBSDTYuqOsrIxV9gwGA4DB95eCJRL2of4qfC8hnU6H7u5uiMViXH/99XjyyScRERHBkm0FBQVs3e/u7oZKpWIcFuLxkcwyzRWfb1AsweFwwGw2+4ksZGdnQ6vV4oMPPsBHH33E9l+qqvCKhGq1mokkuFwujBkzBn/4wx8QHh6Ol19++bI0Z79Qu+SgJTMzE48//jikUinMZjNKSkpQW1sLhULB1GRI0u6Xv/wlHnnkEVaev9rMaDTiiSeegEajQX9/P06cOIGf/OQnaGlpAQDWnTc9PZ01YBpN+9GPfjSqxwtk48aNwwMPPICEhARYrVaUlZUhLy8PCoUCmZmZV6SxEDWko6rZ/v37LwtJXSaTIScnh/3s8XhQUFBwWRXglixZAgB444038Oabb16281wNRo6ZTCbD2LFjsXLlSsTFxaGtrQ3l5eUwmUzMSdXr9aiqqsKCBQtw//33o7S01K8ZJ2WYaUMIBKXieRrAUAeYfgd8vZm73W4mJwkMltipdO/1ev3e8UBOPW/0t5iYGFx//fVISkqC3W7Htm3b8Oabb7IGaDExMYiNjUVGRgZUKhWDywlla89H9hcafXfNmjUAwCAHNpuNcfGEju1IAQQFb/zxKZObnJyMBQsWICUlhXWqnzJlCpM21el0bB8g5z+Qshs/bvw18ZLZ/Of4z2o0GowdOxbJycno7e3Fvn37/GRrL5brQddFDoJUKkVfXx+TbKY+IFqtFg0NDWhqahpyzAupchCkjOYWQezo90qlEqmpqQCAzZs348iRIyyjyo8LH9DxASxf7blazefzwWQysfsgSJPwvoxGIzQaDc6dO4cTJ05Ao9Fg9erVyMvLw/bt2wEMvo8KhQJlZWU4c+YMU6OzWq3o6+sbMqf0ej28Xi9zFmNjYxlcu7+/HwUFBfjnP/+J7u5uJCYmIiUlBWlpaWhvb0dRURHcbjfi4uIwZcoUBpU8fvw4BgYGYDab/dYUei+IpwcMQtooa09QpIqKCpbNHxgYQHx8PJYvX47W1lZs2LABZrMZSUlJMBqNzDnV6/VITExEd3c3ampqWO8qm83Gegz19vZCo9FgxYoV+PWvf42EhARWzSI56DNnzuDLL79Efn4+Cxx6enqGqOsFSvB4vV7Y7XaEh4cjISEBdrsdjY2NMJvNaGlpYbwWtVqN7du3449//CNOnTrFKs8hISFwu92Mz3Q+o/WUemwF6rMl5JUBg8FVSUkJCgsLERERwSBix44dY3sB+bzk+NN8oeoLJTPomFRJy8zMxBNPPMFk4dVqNfLz83Ho0CGYTCa0t7fD5XKxRqe0LzkcDtarJSoqir0XtKY7nU4/ieTTp09DLpdjx44d7DrJd6WxoQTi+PHjcebMGQwMDGDq1KmsUXZRURFcLhfCwsJQWVl52ZEpgeySgxaHwwGTyYSlS5ciMzMTwCBBPDg4mMGLiBQ3Y8YMJCcnX7GghUp4ABjc6VKy9VarFXv37kVKSgoA+EENoqKicNttt+GOO+5gsnOjzWkZM2aMnyOkUqmQnp4+alwhUpWIiIiASCTC3r17ceDAAWRmZkIul2PWrFmMc3A5LS4uDsuWLUNERASqqqrw1Vdfjfo5RSIRMjMzcdddd7HftbS0YPfu3X7VutE2ymhHRUUhNzcXp06dAjC4UOp0OigUCtY0kRpVmUym72TTSd7hICWW9PR0LF26lFVtCe4FgOGZ582bh9TUVL8Fl7LR5JgJM/DC4IIcNyJQCuE4/IaVlJTE3omgoCCoVCoGCeE35kALeCCnmKAn7e3tkMlkqKqqQnl5OZxOJ+Li4nDTTTdhzpw5aGxsxKFDh1jmlVfLEprQgafxoN+Rsy0WizFlyhQAYJnpyMhIjBs3jiniDMd5GMkBp3PQ+I8ZM4aRZouLi1FaWoqMjAxoNBrExMT4wXiFx6ZkF0+4H25cR4J1RUREYOrUqTAajcjPz8fhw4fhdDrZMxe+UxeCOxfye6jHztSpUxlxur29HUeOHGH4f57/NNImT9fFixrQePDf5SsAJCnb19fHeFwqlQp6vR56vZ4J01gsFjQ3NzO40dUetIhEIoSEhDBI6XAQTafTySBZHR0d+PjjjxEUFITMzExkZmaisbERXq8XUVFRcDgcqK6uRnd3NyQSCfR6PXQ6HYN3AUBKSgpuuukmhIaG4quvvsK+ffvgdDqRlJSE7u5utLS0MJGalpYWLFiwAAsWLGCBk1arxfz583H//fdj4sSJsFgsWL9+PatI1NXVDYFl+3w+Jn3MN32USCRwOBzYvXs3RCIR1q5di0WLFkGn0yEsLAxer5f1aiHILFWeRKJBUjspc9XU1MDj8aCrqwtms5kpjsXFxWHBggX4/ve/j/j4eDbvYmNjYTabsWfPHmzatAknT56EzWZjXe35ipdEImHCKHa73e8dkEqlUCqVrNcIrXVE9h8YGIBarUZsbCyqq6v9iPxer5f5XxdaYYmOjmb3N5zRe00JGKrGt7W1YefOnZg6dSrCwsIQExODqKgotLS0MGUv4mUS30ihUECr1bJKC1/9cDgcmDJlCp588klMnDgRZrMZKpUKVVVV2L59O0pKSlBXV8cStjxMrK+vD83Nzejs7GT9gPbt28fgYVKplEEKKaG3f/9+VFRUoKenB2q1Gl1dXX6VsIGBATQ2NiIqKgorV66EwWBAX18fcnNzER4eDpPJBKfTiaCgoEtuCXEpdslBS3l5OX73u9/h6NGjmD59OsuQp6enMyeDJlRzczPDMI6GkYNntVoDQnpuv/12JCQksIl0qZUPp9OJv//974iKisKyZcsgFouRm5uLmJgYzJ8/HytXroTZbMZbb73FZAwvlykUCixbtgyxsbF45plnRuWY1LMkIiKCdQWeM2cOlEolRCIRysvLr4jzPGnSJIwbNw4KhQIffPABamtrRx3aIJVKsWjRItY9GwBbvMPCwi6bethbb72FqVOn4tFHH0VKSgo++eQTAIPPMyEhAQaDgXVIJvWlF198Ec3NzZfler5NIwfQ4/GguLgYJpMJJ06cwIwZM5CVlYWoqCgYDAaoVCq2gVRVVaGjo4OVycmh5IMUvnJCxlcn+ACH8MhEaqSGatS8S61WIzc3l4k18A7MSJUN4d942FNLSws+//xzpKenY/78+UhLS2PzfcWKFbj55pthtVrx9ttvY//+/X7SpsM5myM5wz6fj2Vh+UZkdH/Lli1Df38/ampqhr1+/j7oXRQ2RqQNVaPRID09HSEhIUz2l8ZYJBLBbDb7wdGAr6stNEZCArvwHs/n/JO8a3Z2NsRiMfLz81FVVcWgRML5wVd3hOflYV38eT0eD9RqNWbPno2srCx4vV44nU4UFxejsLCQkZh5+GGgcRTycCig4uc2XR9BeA4cOIDp06dj5cqVCAkJQX5+PkwmE5RKJWJiYhAfH4/Q0FCoVCo4nU4UFhZiw4YNDM5ytZtIJIJWq2U8RyF0kszlcmFgYADJycnIyspCQkIC63OxcOFChISEoLy8HDKZjEHCTCYTJBIJgoODYTQa/YIIr9eLzMxM3HbbbZg1axZeeOEFbNy4ERqNBlFRUXC5XLBarXC73Th16hTkcjnGjx8Pp9OJrVu3wmazYeHChdDpdDh8+DB27dqFPXv2sOMLfQbK2tO7Q4pytP51dXWhtbUVmzdvhkqlws0334yVK1eisLAQ//73v9m+odfrER0dzSCwXq+XVQDIP3O5XGhqamLXn5SUhPvuuw+rVq1CSkoKC5Z1Oh1MJhN2796NTz75BIcOHUJ3dzdrFeB0Ov0qCsTDGBgYYNVcOmdcXBzjaNTX17NElN1uh9lsRnNzM7q7u1FaWoq6uroh69/5EBjk+/FV5PMlQGn9IS4eXWtvby+OHTuG/fv3Y9WqVUwAwWazoampifGo5HI5SxYQ7Io4mgqFggVu48aNwx//+EdMnTqVCRlQ5X3//v0oKytjXBVKshGnx2w2o6mpCSqVCikpKUhOTgYwuI6GhobC4/HA7XbD7XazvY2qhwQ15AM92pfef/99zJ8/H/Hx8Zg5cyYTSSGIID3DC2kgfrlsVIj4PT092Lp1K3bu3MmUC1JSUhj2DxgclKamJpw8efKSz6fRaLBgwQLMnDkTYWFhsFgsqKqqYhrpzc3NyMnJwd13383ITh9//PGoNJWsqanBxo0bcdNNN2H16tWYPHkyaypWV1eHl156CZs3bx6yKV+q0aQkUyqVGD9+PLZs2TJq5xCJRFCpVEwDnhZ1KnVu3rx51BpzjmRRUVHQ6/UAgJKSklGvWAFfyx3yZrVaYbFYRjWwFtrGjRuRnp6O+++/nzUdo8yOTqeDUqlEQ0MDqqqq4Ha7h+DivysmhAoQRnzbtm04fPgwIiIiEBERgZCQEBiNRpZl7OvrQ3d3t5/Ki9CZ9Pl8Q4JrypwR3AIYrMROmzYNU6dORUhICCwWC1MNamhoYKX7NWvWICwsDG1tbSguLg6ovBPIyRc6qeR0EB9tz549yMvLw5IlS5CWlgalUom4uDjU1tbijTfewPbt22E2m/3knANl/PnzCRuK8fwLiUSC+Ph4OBwOqFQquFwu6HQ6pKeno6enhykgnS8YC/QzOdRUaaEqCVUisrKyEBISApvNhtOnT7NeFoECzEtxqqkqp1KpkJycjJCQELS0tDBiMn0mEEGev57huC3897xeL7RaLXJzc6FQKFgly2KxwOl0sooG/8xofIYzmss0X/iAhzKnLpeLic6sXLkSN910E+bPn4++vj7mMPGcD2oSSJWY70LQwjvf1LsmENTHYDAgKSkJ1113HTIyMhAbG8skeaVSKTIyMgCAdaEPDg5GbW0tC6xdLhdbS3Q6Herq6vCPf/wDERERmD17Np555hnU1dXh2LFjiIyMRFhYGLRaLctKOxwO6PV6pKamIiIiAh0dHTh37hx+/OMfM2jYcOu7Wq2GWq2GXq9niUOxWAyNRsMy6iKRCK2trbDZbFi/fj1rKFleXs6qSMAgTI6UuLxeL4MCtbS0sMCmr68PHR0d7Dvh4eGYMWMG0tPTIZFI/K6zt7cXZrPZb85Rgpp6ygCD64JWq2WQOgqSiDc4ZswYdqza2lrmN1ksFhQWFkKpVKKvrw81NTWsb8xwRpVKqpLL5XJWFejs7ITL5UJra+sFcVYpaUKcSjpmbW0t3n//fcTHx2Py5MlIS0tjbQtITIECBgBMSY3Ght5BkUiEe++9F7NmzQIwGHQ3NjZiw4YNWL9+Perq6vyuk2Bfvb29cDqdaG5uRl9fH9LT05nsNjC4zhuNRna/VOWldYiScgQj5I/f0tKCDz74AEeOHMH111/PlHjDw8P9+E7t7e1MuOXbsFFTD6PeHiTdejkVpiIiIvC73/0OycnJbJJaLBa0t7ejp6cHJpMJCQkJyM7OhslkwubNm7F169ZROffAwABOnDiBzz77DDk5OZg2bRoLVg4dOoTjx4+PesACAPPnz/d7YZ1OJz766COsW7du1M4xMDCAgoIC1NTUIDk5GZGRkexv7e3tWL9+/RWdqDU1NWhqarosUAZSNxIayVdeLrPZbPjjH/+IY8eOscZZwjE9c+YMjh07xrorX0rDzv9UE5LUybzewYalNpuNNQuTSqVDlJ6EWHP+b8MZn72XSqXIysrCnXfeiYULF0KpVKK3txcWi4U5ef39/QgLC0NycjIqKiqwdetWHD582K/ycaFGnAO6F7fbjRMnTmDXrl1YsGABUlNT0dTUhE8++QRbt25Ffn7+kOd+IecTbuqU0SfVqezsbFZ1lsvlsFqt2LhxI7Zt28bEBfggUMj1EV4DXwWg//f29qKoqAgtLS1ISkpCYmIik02lng20RhJsTQjno78F4tQIr4l+x1d85HI5qzSQjCqtI0K5av4+Rgo86Wf+uoKDgxEaGupHYCWSLHFchDLDwvENVJHjKzH0O8p0i8ViNDQ04JVXXsGJEyeYNDzxKMkppH4XbW1tMJlMMJvN3wk+CzA4hgR7GinJZDAYEBcXh6ysLCQnJ0Mmk8FisaC+vp6pZHV1dcHhcLAgiPpT9PT0+Ck+UU+YiooKvPrqqwgJCUFGRgYeeughlJSUoLm5GRqNhlVmoqKiMHXqVKSlpSE0NBR5eXkoLi5Ge3s7amtrz3uPCQkJiIqKgkgkYtVfWkOAr3tSUY+arq4u7N27N+Cx9Ho9VCoVkx2Oi4uDx+NBR0cHSyBQsoca3zY3N+PkyZMYM2YMIiIi/CBBYWFhyM7ORk1NDRwOB1pbWxk6g59fxGHT6/WsITCfhJTL5YxHw693vb29OHr0KFOTrKqqgtlsZhVbPpkDgPFbqKLs8/mg0+kQFBSEvr4+do/DBSzCtYaCe4VCwcaYgpGDBw/i3XffRUxMDGJiYtj+UFVVhdOnT7P+KcAgBIzWAT6YmT59OmbOnAmfb1C6++jRo/joo49w8OBBWCyWgNUgEimgPnnh4eHIyMhAaGion3oYwbxcLhcrHPDHo2ouBYRkItGgCE5ZWRmkUinS09Oh0+nQ19fHnkNHRwfa2touq490PrtskseX02w2GzZu3IiEhATIZDJkZGRg7NixCA8PZ4vZ0aNH8fe//x2VlZWoqKgYEcN4sdbY2Ij//u//RkhICAwGA7q7u1FdXR1Q4nK0bPfu3eyFcrvdePvtt/H888+PqoqD1+vFvn37sGbNGr/KztGjR/Hqq6+yTraX01QqFVt8qqqqYDKZLktmkCS577jjDoSGhjJtfpKjvJxms9mwefNm7N27F3q9fohjRA2dvq1MxpWw8z1TPiPMN88SQsF4iBiAgJlt+r3w/J2dnSgoKEBHRweCgoKQnJyM+Ph4ZGVlscxTUVER1q9fj/z8fJSUlKCjo4Mdk3csh3NC+fvhKxE+n49B/0jZp6Ojg0mo8vCgQJyK4c4ndORJ2pmOR1XpzMxMtLS0YMuWLfjHP/7B+nsIqwnC4w0HnaJ7o4xkUVERzp07x3Dwzc3NKC4uxvr161FUVORXCePPx1c8eNGBQOck40n7NDeoY3Sgxpx8Xxf+Hod7biOd3+12o7y8HEFBQQgNDYXVasWxY8fQ1taGvr6+885z4T3ywRM5F5SdpowpBYjt7e3YsWMHDh065Idjl0gkjBMm5BYMN6ZXo1F1S9iTifgu6enpmDhxIrKysuB0OnH06FGYTCZYrVaWcSb5X0p6ktQvPQe+utDf34+EhATExMTAYrFg7969GDt2LBYuXIhHH30UL7zwAnPmEhISsHDhQsybNw+tra3Yv38/ampq4PP5LlidMicnB2PGjEF7ezuam5uZohf1b6F7puaWNpuNNdP1+QYb67rdbiiVSiaDq9PpkJWVhdjYWEaCpyCIjk3zqLW1FX//+9/R2dmJWbNmITg4GAqFAiqVCn19fWhtbfWT4SXyPplYLIZer4fRaGTCJnwV3Ofz4fjx40wtUviuWCwWVFdXIyYmBsBgJUihUDCuBR/kSCQShIWFscqIVCpl8K6R/DJSbyPIKPUs4QM4ClqokudyubB//35MmzYNa9euZWIER48ehc1mQ2dnJxtT6tkkrAImJiayQK2mpgYbNmzAkSNHAMCPu8QbEffdbjcT/4iOjkZPTw8rFng8HlZ5S0pKwpgxY1BfX4/Kykr2bPR6PeLi4mCz2Rh/VyT6uu+Q2WxGe3s7k0MmWJnBYGDcU6rUfxt2VQYtFosFf/nLX6BSqRi2VafTsb8T5q6jowMul2vUnT+v18skEq+U1dTUYOrUqQC+htq1t7eP+nnMZjO2bduGyMhI9Pb2Ij8/H++99x4qKyuviC53RkYGJk+eDK1Wi5CQECxfvhwff/wxU5IaLfN4PPjDH/6At99+G3K5nDUXvJIvIi9v+X/R+Ax9IElWvhojzKqRM847nCPND2EAMDAwgIqKCrzzzjvw+XwICgpCUFAQ5HI5g05YLBYGE6GMGTmRvHPPk90pkz1cUMHfR19fH6qrq1mzTF45K1DlAfCHNvFZ/0BOPo0JL5975MgRPPTQQ4iKikJXVxcqKipgNpvZJjvSOy48Pk9o55+DSCRCW1sbtm/fDq1WC7FYjAMHDmDXrl0oLy/3y9IJnys9b7r2QAEKX7UQVuzo88QxUKlUiI6ORkpKCs6ePcsyrsLvD1fRCfQc+c82Nzfj7bffxoYNGxAREYH+/n6cOnXKLxM8XCAbCBoHgDVIpfEJFHDTM6W9TvhMgK+DM8r68g7j1W5SqRSzZs3Cvn37/MZZqVQyhS+Px4Njx45h27ZtLECh+R0REcHUmIiTQGpkwwWx9fX16OjowJgxY6BWq7Fz505ERkbi1ltvxeOPP46YmBhs3rwZMpkMkyZNQmZmJsxmM5v3lZWVfvNdp9MhPDycVXeFNnbsWEyePBmnTp1Cd3c33G43qyx5vV83FBSLxTAajdBqtVCr1QgODoZIJEJnZydaW1thtVpZEmzixImMu9fV1QWtVguZTIbS0lJ4vV709vayYNvn86Gurg5vvvkmvvrqK4SFhbH5pFQq4fV64XA4WNBhsVj87k+pVDLuDald2e12v0pLS0vLiOt2R0cHI71TRaylpWXIO0HBoEajgVKphEKhgM83SPwfKWghIQClUsmCfark9/f3s55HwsonBa1KpRIdHR147bXXWPWY7zMVCIoKADt37oRYLMa0adNQU1ODqqoq6HQ6eDyeYeHwNE5arRYRERGIi4uDy+VCdXU1Tp8+DWAw4VlUVMTU4yQSCdra2vzWmYiICBiNRj8+UHBwMLxeLxobG+F2u1FRUQGZTMYEteRyOSQSCZOopqDn20isXpVBC8HBvk2t6G/DRoMPdD5zuVz44IMPsGXLFraIXcmsf1BQEOvPQrKpn3322aif3+cb5FBcjsDvmp3faBMYKetLsq9CpSciU4/0fSE3gfgsvEMnFovZgj5cSZ7IpQRvomBF6GzTdQ13r+SA089SqZQ5k7Q58se/UBspk887qQQrcjgcOHnyJIPV8tjz81UFhA42/07yajtisRhOpxObNm3CkSNHEBQUBLPZjIaGhiF9S+j/fPAxXHWMP2cgNS4K3IiMGh4eDoVCgejoaKSlpSE8PJx1qRc6psJAiH9edH/8WNI1uFwuFBcX+wULNI/ON6ZCCW76HSVR6Dp54QfeAvFvhME0Bfd8k8lAymlXm0kkEuTk5GDnzp3sd1QhIE4F9ZggR5TGQKfTQSQSQafTQafTQS6Xw+FwsCZ7w+01brebZc5TU1PhdDrx/PPP4/jx4/jBD36AW2+9FTk5OSwJ0NbWhoKCAtTX17P3j0wkEiE5OZk543V1dX6oAp/Ph+joaCQlJaGurg5qtZrJ/NK9UU+8/v5+xuHKyclBTEwMg3cVFhbiyJEj8Hg8GD9+PMaOHYuWlhbU1NQgKCiIBXeUqQ80X202G86dO8ecd4lEAoPBgNDQUCgUCka8D8QTc7vdjNhPXeL5xCC9Z9TMl/pzkcnlcmg0GoSEhKC6uhp2uz0gB6i/vx/Nzc3Q6/UICQlhkCzqQTLcu0h/o/9oXabKptvtZus0GQWMp0+fRk1NDUpLS9HR0RFwzvAJN/5d7ejowPr167Fz507Wn4zGarhKnEg0yDmOiopCdHQ0RCIRSktLUVxczAIdOldqaiq0Wi2qqqpYY1C6X4IoU6JKJBIhJibGryJEZH+qOBI3iAQDKDC+FrRcs2/dKDtxOfqiXIgVFxcjPz8fOTk5sFqteP311/2aYV2z74aNtIHQ3yk4CLToUxaZN6Hjyy+ovEwmLcQEtaFO2pRNokzZwMAAa1450uJ8IVA3IVSL+gTQ34Um5EDw5xou+z9SZp/+RipERC7lYRH8uema+SpGoOvknwP1kyEZ2r6+PnR1dfk557RhkqrWxfKDRqqo0d/EYjH6+vpQXFyMmJgYmEwm1NbW+ikaBfpuoH8DQxvQ8o6OQqFgPSn44ILuj68IBhpP4fWQU8xX7gJV3QIFJoGeU6Dg5NtwNEbbxGIxxo0bh+TkZFRWVrLf05zyeDx+85w3cp4rKiqg0+kwe/ZsTJ8+HYmJiaitrYXJZEJPTw/kcjmMRiNaWlpYbxVKdigUCibA8Prrr2PLli1YsmQJJkyYgODgYPT09KCpqYn112hra/Orpvh8PlbdCAoKQlpaGmJjYxkcx2KxsKqKTCaDRqNhScTm5mbU1NSgq6sLAwMDkMvliI2NhUajQWhoKEJCQgCA8XN0Oh3mzJmDu+66C0ajEf39/SgrK4PZbEZHRweam5vx2WefjTjetG4GBQWxSgYFAm63OyBCweVywWKxsKCROC0SiYStx0ajESqVCjqdjkH2PB4PQkJCkJCQgODgYMTGxrJqDT93qbJJ1VOqOvb39zOJZYlEwvg7gXgYfP8YarBJv/d6vWz+CN+Zzs5O9PT0+FVVAhmp2wFDob1U1YqMjGR8kZGCZolEgtDQUMTHx0MqlaK2thZFRUXo7e1FamoqS+ITpK2wsJBB3mlfpSQcAD9Z5oSEBNTU1LDknkKhgFwuR3t7O3w+H6viUQXv24SvXwtartl/lJnNZjz77LP46U9/ysq03wW1m2s21Mjx4zHOfLWDFlc+g8dDpwItmudbSGkzomyaSCRiSirCvhj8d4RkRt5RFGbrfT7fEH5IIAeVz87z1RueFMo7roB/gCTMmNMmzTvJVCHiieG88eRun8/HKlh8VYGOJySry+Vy5kgQtI2vHgnPzzv2wykm8ZUGchDpnunZ0fn57DX9nkQHzp49i//93//Fiy++yJIwBGPjicz8eYXjyz8nPkjgHTR6bqTmRFLZRPbnAx56NmR0LiLE0r0T6ZUPWEQikd91BwpWhefhg9jhIJhXq0kkEsyePRs333wzPvzwQzQ1NbFggiqqWq2WzW/K5JPDxScOdu/ejTlz5jCuSmdnJ4NZabVa1ldu9uzZAAYdv6CgIEgkEkyZMgVdXV3Ytm0bUxWbNGkS4uPjWe8cUsISznm3241z585BJpMhKiqKqSXSd+gzYrEYYWFhCA4ORl1dHZqamvwal7pcLrS1tbFrBwbnc1dXFzweDyZNmoQf//jHCA8PZ/No3LhxLIFy6NAhdk1BQUFITU1FV1cX6uvrh6ynPp+PSebyVbxAVdH+/n709fUxyWaCYFFSAfi6oSEdNzg4GHq9HgkJCcjKymJBV3t7+5BzJCQkYGBgwA/ySUGrVquFVqtFcHAwkpOT0d7ePiRoIVgY7QH07lIVgtb84Sq+F9vSItC7l5SUhLi4OJhMJgZvJCijcOzVajWio6Oh1WpRV1eH06dPo6enB9nZ2YiKimLVMIlEgoKCAgY9j4yMZAIFcrmcJaSpojN27FhERkbCYrEgODgY3d3dCA0NRW9vL1paWqDRaBisz2QyoampKWCV90qZyHeBZw2Evb1m1+yafTt2tWZLr60j1+ya/efY1biO0BoiDGJH45gjHStQZZe+I4QpCu2bXCMlLIRVwOGOFYgjxQf4w629FwtJvVLGw0dHuu8LOc5/8jy/mOsbrgJ/qfd4Id8f7h250mN7LWi5ZtfsKrT/5EV4JLu2jlyza/afY1fjOnK51xDiRlEQQBVWoWN3sY4iH1RciBPOKyMKOVi8jdYz5Cu/dNxvGsyMFCBdqoNN1/hNAtbhnP5vYoGc+JGEPYDhZdy/qfGBLV9ZFdrlfM+v9BpyDR52za7ZNbviJoT9AF9vQgQhIBgMDxkiiBL/+UDHDrQxnG9D4b8PBHYsCJIkhNoI4WAXujkJeQ6BNmQhiZTPMPMOAP2bh5yRA8ZfF0G4COZE3+PPE2hDHk4JZzhoFX+d9Ezpmgn6FQj6x+PMAxFo6e88oV44pqPpTJIzR1hvGjfChJMUq1gsZt3W6Vj8v4VzgvDjgSwQNJLGiI7HH5eHWfLGE4vpWFqt9lvjLI6WiUQiJCQkoKGhYUTIm0QigUajQV9f37CQRHq+Xu9gl/jrr78et956KxYuXIjQ0FCMHz8eRUVFmDZtGiwWC9xuN9RqNeRyObq7u9Ha2jpkPAleJpFI4HA4IJPJkJOTg9zcXLhcLhQWFuLcuXN+SlFEcifI4YsvvogxY8bg9OnTaG5uhtVqhdVqhdlsRk9PD7tvk8nEGkQOd38X8j7k5eXhr3/9K1trqW1EdXU1pFIpxowZA5/PB5PJhI6OjiHvbnBwMIKCghgHR61W+wmOUO+bjIwMbNq06Rs5u0lJSfjRj36ElJQUvP322/j000/9ruF88Mfly5dDqVSivLwcJSUl3ygo0+v1CA0NRWRkJIxGI+rq6lBaWgqZTIbs7GxMnDgRnZ2d2LBhg9/YR0dH47HHHkN+fj5KS0vR1dU1RM3wQk0ulyM6OhorVqxARkYGGhsb8fzzz0Oj0WDs2LGMD0WiDT09Pejt7WV7KDW6vNDG1UqlEsHBwXA6nejp6flWkx3XgpZrds2u2RU3r9frhxvmjTDSfJAhJK0T7pjngvDBTCC70AzXSBvfcH8TOoxCdTH+3EICPX9NPHeC/uOddDp2oGCK57gMR/YmxSsy/hkQ30MYkJEjRc9EeG4KJIWBDl0vOeeBHASFQsFw+3ROPtgKZMN1s+fHVC6Xs34Wwq7QgaA8gZ4rScsK1cbofoh7Qn0caE7LZDL09fX59WoQjhNdl/A+6BnyhGCS2KZzCANdCvSosSXfJJQnCtP3rvaABRi8l6CgoPM6qYTxHw6DL5VKGT9DJBIhNDQUsbGxiI2NhVqt9vusxWJBT08Pex+sViva29v9xlMikUCn00Gv1zM1rPDwcCxZsgSrV69GeHg4Dhw4gIMHDw6RthW+dzt27MDx48fR3NzM+piQs0n8C57/MpxdqIOp1+sxceJE2O12HD9+HAcPHkR7ezsUCgWWL1+O4OBgnDt3zq/jPX8OIrPTfFQoFEyCuL+/H1qtFjk5OZg7dy4+//zzi3Z85XI5Fi1ahHnz5sFoNKKiogIHDx5EW1tbQG5XIGtsbERzc/M3FvdRKBSMd9LS0gIAGDNmDBISEuB0OhEdHQ2DwcD+Fh4eDpfLBavVCp/Ph+zsbBQVFaGxsdGPE0PB9YU2bXS73ejp6WFy1tRgleYINTkViQYVxyQSCfR6PeP8CVXK+P2G1kh+v6ImodTH5kKDncthox60SCQSBAUFscVAoVCgoaHBjzx4za7Zt206nQ5RUVGwWq2j3gPmYk0qlSI4OJgt8E6nEyaT6aKJflebCfuZAF9np+VyOYKDg6HT6Zjz6nA40NHRwTZG4SYlfIZ8ZSBQBSOQo83/jRwe3vGlY9AxhQ0m+Wu6GCw2n+3lrzHQZizEuPP3xZO3hZ8ROm2klkYQA7fbzaobfK8Q4ZiOpEglhCZQNYXkMvv7+1nDRz4QGq5qJqz0CDdU+gwvAkDH7e3tDSiUIAwayShQojEPtDFTQECfCw8PR05ODkwmE86dO+cXRPPPjuYJVT3od2R0ncIxD0Ry5gMW+pnOq9PpkJycjOjoaFitVtTW1jKHiUjPfNXlajWRSASNRnPez1FVRK/Xw+l0MmU2MolEApVKxVT1ZDIZHA4HGhoaEBQUBKPRyJIq1CtFLpfDZrPBYrH4OX7keJI8scVigVKpxJIlS/C9730PWVlZEIvFKCkpYQELH6gKrbi4GBKJxK+qQnOf1grqDTIa1tnZiY8//hinTp3CuXPnUF1dDZVKhaVLlyIjIwOFhYWorKwM2Kjb5/Ohu7sb3d3dEIvFrNnkwMAAqzhRz6YPP/zwovdbpVKJxYsXY8WKFQgKCkJ3dzeio6Nx44034ssvv0RHR8ew+yX/vp85c+aix0V4HSqViqkGFhQUwGQyITc3FyEhIfB6vWhoaGCVL+p/4nQ60draipaWFta7RiaTQa1Ws+BOo9Hg3LlzF3wtZrMZpaWliIiIQHp6OgCwoIRvQEoS1UqlEgMDA7BYLHA4HGz/k0ql0Gq10Ov1bJ2mBtf9/f0IDg5GfHw8IiMjmVJcU1PTt7aGjGrQotPpkJmZiZtuugnx8fFIS0tDYmIiHnvsMezbt+//XF+Vq914JZHzleEv1dRqNeLi4mAwGNDd3Y36+vphmyxdqolEItxwww3417/+hU8++QTf//73R3XxvxhTKBQYP348HnvsMUydOhXp6ek4efIk/vCHP2DHjh2XbQy+beNlcEk7npR+wsLCkJaWhry8PEycOBHR0dHo7+/HgQMHsGHDBtakjcee830oAgUKwynA8JUb+h39XyhnKYSG8d8NlMEPFDjQvwNB1XhMOZ9N52FGvBPM/8yfl4fQBapM0VjwWXnKKlN/Al61jD/HhQZg/OcMBgOmTp2Kjo4OnDp1Cj6fj0kCC4MKHjImHFO6Lh5eBXxd5SJHIDIyErGxsejt7UVVVRW6uroCPi/htQYKMvkghseOi0QiKBQKzJ8/H4899hiKi4vxq1/9imV96TnSvQpV0wJVtPjxDVQV4M/PQ+h8Ph/0ej0yMzOxdOlSzPn/lbAqKyvx73//G9u2bUNLSwtzsL8L8DCf78LUmyjDrNfrmRNGcxz4Oiik59TX14fS0lL09PTgwIEDMBgMDMJDFQaXy4Xe3l6/CjG9Q6RYZrVaoVQqsWjRIjz44INISUlhndo9Hg+bD2q1OmDVDfh6vtN10rMWypaPlpnNZmzduhU1NTXo7u6G1+tFXl4ekpKScPbsWRw8ePCCmiF7vV6m2Nfc3MzePWr0GhISclFBi0wmw6xZs3DPPfew5pg2mw1BQUFYuXIlJk2ahO3bt2PPnj1+DS5pPeH7rlys8QqFFNQqFAoYDAbo9Xo4HA5UVFSgqqoKYWFhEIvFsNlsbG52dHRAq9Uyv6aurg5erxcRERHQarVYvHgxcnNzsX//fpSUlPidO5DKodB6enoQGhqKKVOmABh8LyhZQ2On1+uhVquhUqngcDjQ29vL5N81Gg3Cw8MRHh7OGklS41HqwUNQSJlMBqlUyhqiXmhVaLRt1IIWtVqN1atXY+3ateju7sbcuXPR19eHxsZGfPLJJ5g5cyaOHz8+Wqe7ZlfA5syZg9deew2FhYV48skn0djYeFkqEiQd+Yc//AE5OTnYsmULfvWrX6GsrGzUz8WbVCrF7NmzkZCQgLNnz17WcwlNqVQiNjYWEyZMwJo1a3DjjTfCbrejsrISbrcbq1atQm1tLU6dOnVFr+tKWSD4lNfrZdmzGTNmIDIyEvHx8VCr1RCLxWhubkZ0dDTa2tqY0xXI8QcuHApGJgwk6Lr4a6VNkCfpDgwM+BF1LwRXzRt//R6Ph8mD8tl4voLCj18g+JSwisQHAMJqt9frZbyM3t5eyGQyyGQyJvMZqBpxIdAs/m9erxeTJ0/Gc889h/r6evz+979HdXU1gzCQE0n3GOg4/L2NdP6BgQFMmjQJP/rRj5CXl4ft27fjjTfeYM5qoECAn3s0JrzxHAP6LMHDXC4XjEYjZsyYgdjYWLz22mtMupQ+G4h3IwyEhPcrDKCF85BkpF0uF6sYL1y4ELfddhuD91gsFqhUKqSmprIeDtSv4WoPWIDBsWltbQ34N7FYzBqLUqNRmmdUue3o6GDBZV9fn18g3dbWhvr6evT39yM+Pp4ljvgu6cJ5rlQqodFoWIXPaDRi+fLleOSRR5Ceno7a2loAQGJiIqxWK+x2O4PuBKpcAGCdyCmQp6QOXQPfA2Q0rL+/n3WQ7+zsRFJSEiIjI1FQUICjR4+yoJd6LQ0ndcwb/V2j0SA2NhZKpXLY+x3OEhMTsWzZMkybNg3h4eGw2+0IDQ2FRqOBRCJBVlYWsrOzkZKSgo8//hh1dXUABt9dpVJ50cEKv06oVCqoVCoA8IO76XQ6FgTYbDZ4vV7WqJqv4IvFYlbBaGlpYXLCkyZNglarxbx58zBx4kRUVVWhsLDQ7zoCXbeQv9bd3Q2z2YzGxkb2ne7ubrbuaDQatp+43W60tLSgoaEBFosFPp8PSqUSUqmUJUEIXUDn6O3tZa0BiNPn8XgQFBT0rXFbRi1oiYuLw6233oqJEyfid7/7HSwWC0pLS3Hw4EH88Ic/9GusdCVNp9NBJpOxEldcXBzcbjcOHz6MpqamUTmHSqXCuHHjEBkZiRMnTjA84+UyhUKB66+/HpGRkWhqakJ+fv6wXVQvxSZOnAi1Wo1ly5bhgQcewIkTJ+BwOFBcXDyqDR8NBgOWLFmCSZMmoby8HJs2bWILz+U0IZTlSlloaCiWLl2KFStWYM6cOZDJZCgrK8OXX36J48ePo6ysDJGRkTCZTN/K9V0J4ysJ1AgOGGwYNn/+fEyePBn79+/Hp59+CrFYDKPRiO7ubuZgBIJNAf7BSiAHFPB3vPmggZ8PtIATn4BgJAaDAXFxcYiJiYHH40FJSQlKS0vZwh7oPgOZsMrj8w028Bo/fjwiIyNx5swZ1NTUDNsNnY7BQ+CE1Rw+EKD/+CqDXq/H9OnTERsbi9LSUpw9e5bBBi50MxrucwTlksvlmDFjBiZMmIDx48eju7sbe/fuRV1dHRobG5nzRvfIc3p4GEygoFSYgQwLC8Odd96JpUuXorKyEgcPHkRZWRmD9NDGyz9/njsVyHgBA766R8YHRDycjL82YSNUcgh44rXw/objaZFz4Xa7kZSUhJkzZ2LBggWYOnUqgoKCcPLkSRw4cACFhYUwm81wu91ob29nvUt4ftLVbD6fb9g9KCUlBffccw/kcjlOnDiB2tpaNDU1wWaz4brrrkN0dDS++uor2O12qFQqVjXxeDx+/BadTocxY8agoaGB8UiGg0zKZDJWiYmJicHq1atx9913w2g04tChQygpKWEk8MbGRr8+LMNl0p1OJ/r6+vy6yFNy45vA7VUqFUQi0bC9z7q6unDo0CF4PB7ExsYiODiYQcIGBgagUqkglUpZ40ar1XreeSSVSqFWq5GWlgaNRsMc5kAWqPoslUqRkZGB5ORk1mfHYrGwxphmsxl6vR65ubm444470NPTg3fffZdVEmisLiRwUSgUiI2NRVdXFywWix/3jN55WteoYWYghAafvAoNDWUCBm1tbWhtbUV8fDxiYmLQ3d3NGmr29fUN2e8DXTPBt8hMJhM+//xz1muHuFZ8ZZcCt+7ubr+KK/XFiY2NZf1igoODAQzyt1paWliVUavVQqVSwW63o76+/lvz54FRDFq6u7tRVVWFzMxMZGVl4Q9/+APrCPvYY4/5ZXdoQVAoFKPumGm1WkRHRyMtLQ16vR4ZGRnQ6XQIDw/H2LFjkZ2dDavVinvvvXfUgpaEhAT85Cc/wbRp0/DLX/4S77333mXjI4jFYixYsADPP/88UlNTUVxcjJdeegmffvrpqAYuY8aMwbx58xhu+IknnsDatWvhdDqxfft2vPjiiyyzcCkmEomQmJiIxYsXY2BgAOfOnUNhYeF3ks8hkUgwZ84crFixAjfffDPCwsJw6NAhHDlyBMXFxdi2bRvL6pWWln7LV3t5jXeeyUQiEaxWK+rq6jBu3Dh4vV4cOXIEJSUlMBqNCA4OZmVpkUjEYB/Uefl8EADh+YTld3LqCeoRERGBpKQkxMbGIiIiApGRkYiOjkZSUhKioqLQ2tqK559/HmVlZcMqOAW6R75qwnNZkpOT8fDDD2Ps2LF45ZVX0NrayhRfKADhnfhA1aSRKhV0z4TzX7hwIX74wx8iKysLZ86cwTvvvINt27b5ZQxp8xVWQujvwgoBTxQXiUQYO3Yspk+fDrlcjt7eXqxevRrLly9HTU0Ndu3ahU2bNqG8vHxIwDhcxUoIi6LzSiQS5OTkYPr06ejv70dlZSXKyspY00YKKPjz8Ip0NGY8RI3/LD+ufJXL5XKxtUoYdPE8Kv6YvEDBhXCzAgWjkydPxt133405c+YgOjoapaWl+OCDD5iDTOpOfJBHFaLhVOCuNgvk7Gu1Wtx88824//77ceDAAXz55Zeorq5mHbznzZuHxMREnDx5EgMDA9BqtWzdcLlcEIvFLLseFxeH7Oxs7N27F0BgiCkAphrn9XqRmJiIO++8E3fffTfkcjk2bdqEjRs3sjVDJBKhoqKCfXekPdvj8UCj0cBgMMDr9bLqHgUwQgGT81lwcDBCQkJgt9vR1tY2ZI+lgJ4qoKWlpayaRx3SRaLB5qp2u/28ZGxqmkl8CLvdjvDwcISFhSE/P3/I54m8T/dJv3M4HExBraOjAxUVFejs7ERXVxccDgf0ej1uuOEG3HTTTZg5cyb27duH4uLiIaIjw5lINKhEt3TpUqjVaqxbt479noJKmUzGVLiSkpIYBDCQ/8q/0z6fj3F6+vv7UV1djfHjxyMvLw+NjY2sehEZGXlBPg9V2ej6BgYGUFlZicrKSnY+/r3o6+uD2WxmFSGXy4WgoCDExMQgOzsbkydPRm5uLmJjY2EwGFgTypaWFjQ2NqK7u5u9C3q9HlarFbt378YXX3zBqjtX2kYtaGltbcXmzZsxfvx4LFmyBIWFhXjnnXf8yGIikQhxcXGYP38+UlNTcebMGT/Jum9iEokEsbGxyMzMhMFgYIFJWloagoOD2YOw2+3Q6XTo7+9HY2PjqGXyVSoVpk6diuuuuw5BQUGs8+7lsjvuuAM/+MEPkJiYiIMHD8LlcuHBBx/EwMAA3nvvvVE5R1BQEB544AGMGTOGbbwGgwEGgwHAYCbL6XTiT3/60yWXqPV6PZYuXYq0tDR0dHSwLOx3zVQqFRYtWoQHHngAy5YtQ39/Pz766CO8/PLLKC8vvyyVsv9k45074Osgprm5GXv27MGUKVMwa9YslJeXo6OjAx0dHYybIJfLkZWVhRkzZkCn0+HkyZM4efKk3yYaCHYjdA55/oRCoUB8fDyDRCQnJyMnJwcJCQmIiIhgwZHL5UJISAjkcjkqKipQX18Pl8vFSupkgYIV+r9QWnhgYABqtRrjx4/HpEmTEB0dzTJqxEPhA5RAKmF0rEDBSqAAYP78+XjuueeQm5uLw4cPo6+vDw899BAkEgk++ugjOJ1OP67O+aqSfKBCsKjIyEisWrUKubm5bB0JDQ2F3W7H1KlTkZ2dDZ1OhxdffBFdXV0BM61CR5H4HPwYEG57wYIFiIiIQHt7O3bu3InS0lIWRPHBDjntSqWSqeFcSEWMbGBggN0PQWWIhM1DLPhqIJ85JqeDl/AeLtgUPlOVSoW5c+fi8ccfx9SpUyGVSrFhwwa88847KC4uhs1mYzAiXv2MzhfoHN8lCw8PR0REBA4fPowPPvgAhYWFjIeh0+kwe/ZsGAwGaDQaloUmaVhKWvT398Nms0EqleLUqVPnXZvpu/Hx8Vi9ejXuuOMOyOVyrF+/Hm+//Tbq6+sxYcIEKJVK2Gw2v6BlJL5Ceno64uPjYbPZ0NbWBpPJhN7e3ouGvpJ1d3cjKioKWVlZ0Gq1aGpqYmR5/j0jpbLw8HDExcUxro/NZoPZbGaQITKNRgOtVgu5XA6pVMqqVzabjfElfD4fU9Sj4CcQbJKkwwky6nK5cObMGdaF3WKxoLOzc0iyxOPxYPz48cjKykJaWhqKi4vPOx5UBTIYDHjooYdw9913D5Fi5rlE/f39kMvlyMnJgVKpxKFDhyCXy0cMjEwmE9ra2qDX69Hf3w+TyQSbzcY60ysUCuh0OsTGxl6QD0CoBK1WC5fLdd7A0ev1MuhvZGQkQkJCkJaWhilTpmDq1KlISUlhfrHL5YLD4UBnZydaW1vR1NSErq4uqFQqBmfPyMiAWCxGVVUVTp8+/a0kP0aViF9aWoqSkhLk5ubi6aefhtlsxvr161l5/KabbsLq1auRl5eHQ4cOjYpzmpaWhh/+8IcMI0gEp9LSUnz11VeMQD5lyhQsXrwYbrcbf//731lkeikmlUqRnp6O+fPns0i5o6Nj1LGmZNHR0fjZz36GjIwMFBUV4Sc/+QlCQkLwi1/8AgsXLsSmTZsuiCh3Pps6dSrmzZsHnU4HYPDF3bNnDz7//HMoFAr84Q9/wAMPPICamhq8//77l3SuiIgI3H///VCpVDhw4AD27t37rcrpXQ4LCgrCsmXL8NxzzyE2NhY1NTXYsmULXn/9dVRXV3/bl/etGp/BJ3JpSUkJCgoKkJOTg9tuuw0NDQ344osv0NPTg6ioKEyYMAF33XUXxo8fjyNHjmD37t3o7e31cxSF5PhAZHlyGmUyGcaMGYO1a9di1qxZMBqNLAPb1NSEAwcOoLm5GS6XCxMnTsS8efPQ1NSEjRs3ory8HIC/EtpI90qfpWui/i5jxozB0qVLERsbi4GBATgcDkaSBwLDhPhjDWd8dYbGNyUlBT/5yU8wfvx4bN++HX/5y1+gUCjw6KOPIi0tDQaDgWX9hCT84RxsGnf+d9nZ2ZgzZw4MBgN7Nrt378aBAweg1WqxcuVK3HHHHSgqKmLJq5GqOcKqD/+Z8PBwzJw5EzqdDmfOnGEJHb7CwTunAFh2nP5G2exAcDt6VuRM0TEJD97X18cgI/R3+oxOp2NcB95obCm7KTQhtyU0NBTLli3DE088gXHjxqGxsRH/+te/8NFHH6G+vj7g9VJmnH8+FwP/u5rMaDQiMzMTxcXF2Lx5M4qKivwQHjk5ORg3bhw6OjrQ3d0Np9MJpVLJ1E6pt0VXVxdsNhscDgfMZvN5CcdKpRIJCQlYsWIF7rzzTmg0GmzevBnr1q1DcXExgoKCEBcXB4lEgqqqKtTX1494PHo+oaGhkMvlaGpqQnNzMxwOxwVDnQKZw+FAfX09oqOjkZiYCLlcjr6+PgwMDLB1jLg9FLAQGbunpwcdHR1+wjBSqRQhISGYMGECZs+ejeTkZOYkFxQU4PPPP0djYyOsVitbkwm5Mdz8E4vFDM5E72dXV9cQOBK9e1QNqaurQ11dHXJzcxEdHX3eOS6TyaDVaqFUKhEdHY358+cjLCwMCQkJiIqKgsVigVwuZ/Ayj8cDiUSCVatW4f7774fD4UBhYSEjpw8XfFqtVrS2tjKJarVazdS3zpw5g4SEBLaW8euDQqFAcnIy4uLiUFVVxQj8FBjK5fILTnSqVCpkZWVhzJgxCAkJQUhICOLj42G323H48GFYLBamFma1WtHR0YHm5ma0tbWx5FVERASmTZuG66+/HnK5HGFhYd8avH5Ug5b29na88cYb0Gg0uPHGG/GTn/yElehiYmLwxBNPICkpCV1dXfjoo4+GEI8u1iQSCTIyMrB8+XJERUVh586d+Oyzz1BSUoLq6mpGfHrooYeQkZEBi8WCP/3pT/jkk08u2TGWyWSYOnUqnn32WUyZMoVtyhfivHxTu+uuu5CSkoIjR47gxz/+MQoKCqDRaPDcc89BLpdDqVSOStCSmZmJ8PBwiEQiNDQ04L333sNnn32G0tJSSKVSRERE4JFHHsHUqVPx4YcffuNFVCQSwWg0Ii4uDgDY4nilbGBgAM3NzUMInWFhYcjLy0Nubi4jIV5ImXk4e/DBB3HfffchMzMTVVVVePzxx1FaWjoq8Lqr1fiFnndKRSIRWlpasHHjRoSGhmLJkiV44oknoNPpUF1djezsbKxevRpjx45FVVUVNm7ciJKSkiGNCIUBCjlrfBNI+qxarUZmZiYmT56M5ORklJWV4dixYygqKkJ5eTna29vhcDhw++23Y+zYsejp6cHf/vY3fPzxx7DZbH7ONA+R4nkKvKNPPJmBgQGIxWJMmzYNTzzxBGbOnMm+T9dHymqEmafj8BUH3skeDtLE//+OO+7AxIkTsXv3bvztb3/D0aNHoVQqWaYzUD+T4da04YIYtVqNxMREGI1Glplbt24dNm3ahObmZkilUlitVvzmN7/BihUrsH37dgbhEVal6Hr4oEF4bSEhIRgzZgyDHHd0dLDKCql3UYWDnjsPr6ExH87ZoWcrrAQRLLCpqYkRjOm5xsbGYuzYsYiLi0NLSwv279/PoI38ePHXIRxP+ndERATWrFmDRx99FElJSTh8+DBee+017Nq1Cz09PX4VHLoGCnopcCUC+bflbFxOUyqVmDZtGlatWoW6ujqIxWIkJiayvhwqlQo33ngjjEYjjh49ylTewsLCEB4ejpCQEIhEIrS2tjKJYbfbfV7sPiUub7vtNqxZswZBQUHYsWMH3nvvPRQXF8Pj8SA0NBRxcXEQi8UoLS0dUamSrza2tLTAZDKhpaWFyVZfqhFHRq1WIzw8HEajEQqFggUtwODccTqd6OzshEgkQk9PD8xms5+TTAliEpFJS0tjfXNcLheysrLQ3NyMQ4cOwWq1ore3F1qtFiKRaNj7p/tTKpUMjtbT0xMwIBCLxUw6uLOzEy6XCy0tLSyo4qvoQhOJRKx6YLfb4XA40NfXx/qsAIPVI51OB7PZzM6fk5ODhx9+GOPHj0dvby/i4+Nx7ty58/o/JDNMQZvL5UJlZSXKy8uZNHFTUxO7f4VCgRkzZuDBBx9Eeno6Nm/ejH/84x/o7OyEVquF0+lkZHthwER7HL8WpKSk4IYbboDRaERRURGqqqpw+PBhxl+UyWRwuVys2WRvby+6urrQ0dHB/J7a2lqUlZVhz549SEtLGyIdfiVtVIOW/v5+nD17Fj//+c8hk8mwbNkyPP/884zEFRYWho6ODjz66KM4ceLEiKXRCzGv14tTp07hd7/7HcLDw7Fjxw6Ul5ejp6eHPbTbbrsNd911F+Lj4/HLX/4Sb7/99kWrVwhNJBIhLS0Nr7zyClJSUtDW1gaz2YzY2NhLOu757J577kF9fT3++te/4tixYwAGsyfHjx9n8IfRsIKCAhw4cAAejwfr16/HkSNHGOmxv78fL7/8MubOnYtp06ZBq9V+Y+k7pVKJ6667jmUUe3t7h5X4pWZJMTExKC8vv+SyJKkkHTt2zK8rbWRkJO655x7cdtttiI2NRVVVFR555BEUFxd/o5f0qaeewiOPPIKkpCSsW7cOb731Fo4fP35Z5aOvBhNCKHkojcfjwYkTJ5hK2B133IFf/OIXqKmpQUhICJKTk9HY2IiXXnoJu3fvZrwFYcIgEBeADzDI3G43amtrsWnTJmzYsAHHjx9HdXU128y8Xi9WrVqFH/zgBwgPD8f//M//4O2334bNZmPNEQOdc6T7Jsd27NixLPHR1NQEk8mEqKgov2sjI+iE0+lkQc/5TIhxDg4Oxp133onKykr885//xOHDh1l1Yf/+/QwKMhxMSVgF4Z0D+tnn8zFngiorX331Ffbu3cs2XovFgk8//RT33HMPIiMjWSDHPyvhmBI0ItAYa7VaaLVaJrfa09Pjxx2hIDAhIYEpD1VXV7N+CjxOPNA5fD6fXwNOGlODwYCBgQGcOnUKVquVBZEJCQm47777sHz5coSHh+Ps2bPo6OhAfn4+41KRQxBIGIK/b61Wi6VLl+Kxxx5DfHw8vvzyS/zpT3/CmTNn4HK5WPaVVzmjqhH/XGh/+C7wWYYzuVyOCRMmIC4uDlqtFq2trWhuboZcLseYMWOYc0vCEFFRUQwO1dzcPKRL/flMJpNhypQpWLVqFeLi4pCfn4/3338fBQUFDA4VExODhIQESCQSxrUYzvi5X1JS4lcpuxATiQaVzJRKJavWkgUHBzNCe3t7O0sshIaG+h1jYGAAZrOZBRcEkQIGE8XBwcGYPHky7rrrLoSEhKCurg41NTVQq9Xo6+uDQqHAlClTsHbtWigUCpw7dw5VVVUwm83DkvcJYsnDOamaHMhonpOyFyVcWlpaRuwhEhoaioGBAfbOuN1umEwmmEwmBoUyGo2wWCzQaDQsUWQ0GrFo0SKkpKSwyur48eOxZ8+eEROtlASiQJggYqT6RX1PduzYwb4THx+P733ve5g0aRKCgoIwefJkfPbZZ+jo6IDdbmfrEK09vP8ll8sZRJagzPPnz8f06dOxb98+fPHFF+jo6AAwiFKiiktbWxsTLKHgijik1NeJnl9FRYUfN+9K26g3l+zv70dDQwN++ctfYuzYsazpjc/nw9mzZ/HXv/4VO3bsuGgSWSCjCPXdd9+FRCKB0+n026Dvvfde/M///A9CQkLwyiuv4LXXXrvkTL5YLMbixYvx5z//GampqcjPz8cLL7yARYsW4d5770ViYiKCgoLgcDgYuSsiIgI1NTV45513vvF58/LykJKSgldffRXbtm0b8vfRnED5+fn44Q9/CK/XC6vVOqQq1dXVxfCYl+J8ezweBmsQiURMP1xo4eHh+OCDDxAbGwu5XI4//vGPeOONN77xecViMW6++Wb09vZiz5497B7kcjluuukmPP300wgNDWUa//Hx8SgrK7voIDsnJwerV69GYmIitm3bhr/+9a8MZ/9/3YQ9Doi8SJuP2+1GZWUlNm7ciLy8PFx33XUIDQ2FWCxGcXExXnzxRaxfv57JmBLfhDY84SZPcqbCrBRhrU+fPo2qqir09/ezjD813Vq0aBF+/etfIzExES+88AL++c9/wmq1ssBXyFsQ/pvuEfiapKnRaLBs2TI8+eSTjFeyYcMGLFiwAJGRkYiKikJsbCxMJhOio6MxZcoUpKeno7GxEZs3bx62IepwmXQa44kTJyIhIQG//e1vGZeFHNre3l4/iALvwA/nyPNGzrHH44HdbsfBgwdRXl6O3t5emM1mJkFK1tHRgdbWVjgcDgZDIK4KX1ESno+veNB4koNDxHjimvAE67S0NDzxxBPIzMyEUqnEO++8g08//ZRJ2QrnhbAiIlxjZTIZkpOT0dfXx9SofD4fdDodlixZgjvvvBPR0dEMehQfH4/CwkI2rmRKpXIIHI8f7/T0dKxduxbJycn44osv8Oc//xmFhYXss3xljJ/7CoWCjQnxcGjOftfM5XLh4MGDaG5uxqJFixAbG4vQ0FDEx8f7VVnFYjHmzZuHqKgo5OXlQavVorOzE/v378eJEyfQ1NTEOogT32Mkk0qliI6ORmhoKNra2vDvf/8bJ06cYM9TpVIhMTERERER8Hq9KCsru+AAZDiYeaBnKJfLERkZyRqI2u12OJ1OyOVyTJo0CampqbBarejr60NlZSWam5uh1+thNBqZ48/bcApzERERuO+++7Bq1Sps2bIFu3fvhsVigclkYgmLzMxM/OpXv0JkZCQMhv+PvfcOb6s838dvWXtblmTZlvd2bGfZ2dNZkAVJWGEkjAKlfGjZbRmFltFCyyx7hJFAGQkkITsh09lO7HjFe+9tWda2pd8f/j1vjxQ5AeJA4ctzXVzEtnTec97znvc8477vRwOLxcJk1H3foTKZjD2/5AhTUoaaJGo0GvB4PHR1dXlVlSlAMhgMGD9+PMaMGYP6+vrzQq71ej0sFgsGBweZepbZbEZ3dzdEIhHUajVmzZqF1tZWpphltVoRHBzMgt6uri60trYiIyMDixYtwu7du4dVROPz+azxpccz1Ni1oaGBBUtWqxU9PT0sCa1SqXD55ZcjNTUVn332GVJTUxlfiL5Pn9NoNOeMSxLfNpsNHR0diI+Px6RJk2AymXD48GE0NjbCbrcjOjoaoaGhMJlMKC4uZmppNBbt5UFBQRgYGIDZbGb7yE8tmT7iQQvZ6tWrWamNLCYmBidOnBjRTdPtdvvNjKxcuRJPPvkkIiIi8M033+DJJ5+86GY4PB4P99xzDx544AGEh4ejtrYW77zzDqRSKUJDQyEUCnHrrbdi2rRpcLvdSEtLg0ajQUBAAD788MOLCloefPBB9PX14b333mObmUgkQlJSEoRC4UVD7bjmdDrR2to67N91Oh10Oh0OHDhwUc0PSRSBNnGtVouQkBAvVYq0tDT8+c9/xrRp09hL2Gg0/uAxyajZ2J49ewAMvQgWLFiA//u//4PD4cALL7wAuVyOlStXsszo97VZs2YhPj4ehYWFWLduHcrLy8/7EhQKhUhKSkJWVhYaGhqwd+/eHxUu92PacC9u2piJCDpu3DgG3aCN22w2o7e3l70IaRMlqA5wLune90VPmH/CiBO5lEuiVqlUWLBgAf785z8jPj4eb775Jl5//XXWi+O77mO+vUBUKhVuvvlmPPjggzAYDCgtLcXevXthsVhYD4Drr78e06dPBwCkp6ez9frvf/+bOUTcfi7A8ER8ACyLOWvWLHR0dGD37t2swiiVSpGUlASFQoG8vDwGYfKV4/VnwylfERadYC1cThGds9FoRGxsLI4dO+Z1Hb7z5WvknHODxYGBAbYXabVaGI1G1NTUsOrD1KlTceedd+LKK69kTtqmTZvYPR8YGGDQQe44tKa410XPsEgkQkJCArq6unDq1CnWk2Hs2LG49tpr4XQ6sWbNGkRERGDcuHFekrPcLLJvUogbLMrlckyYMAFJSUnYv38/1qxZwwiw3GCKKnfUPyIxMREpKSno7OzEsWPH0NLSwjLMv5SgxRcOajabUVBQgLa2NpSUlLBkxtSpU3H11VcjPDwcPB4PZ8+exY4dO3D8+HGkp6cjMTER4eHhCAkJgUKhgF6vh0qlQkVFxQV9BlKE2rp1K+rq6s5xYIVCIeOGmEymEVEsdblcrFJAkC2n04menh4GmRKJRMjIyMD06dOxaNEinD59Gq+//jpKS0uZNDP9R/4EPfPDJecUCgXuvPNO3HvvvXjnnXfwwgsvMGec+6yKxWJs27YNIpEIhYWF6OjoGPa9R71tqApB/USAIY7SmDFjkJaWBj6fj46ODrS1taGurg4VFRUsqBw7dixWrlyJxMREfPTRR+dVtSLHm5JSwcHBqKmpQU5ODm688UZIpVLMmzcPx48fR19fH5RKJSwWC8RiMcaNGwe3241PPvkEcXFxmD59OoxGI5KTk/H55597QezIrFarF6zX5XKhu7ubcWW6u7shl8sZN2bs2LG45ZZbUFtbi23btjGlMpvNxpJ5Ho8HYWFhrCpis9nYO4Hgx3q9Hnq9HlOmTIHT6cSJEyfQ1tbGxJRUKhU6OjpQV1fHehbROqIAX6PRQKVSsb1qpJuZ/lC7JEHLwoULcd1117Euy7t378bo0aPR1taGXbt24ejRo/j9739/SbSeFQoFJk+ejCeeeAKRkZF466238Nhjj110wCKXy/Hiiy/i2muvhUqlAo/Hg9FoxL/+9S+26Hi8Ifleo9EIm82G9vZ2fPTRR3j33XcZDOGHWmRkJOrr6xlMi8fjITExEevWrUNBQQHuvffeYaP9kTZqWLZnz56LxjVyH4K4uDikp6czfGdKSgrefPNNTJgwgUX/brcbu3fvvthLYGPTBhkSEoLly5cjKSkJX3zxBfbs2YMHHniA6ZZ/X4uLi8OoUaPQ1dWFRx99FAcPHhyWR6VWq3HDDTdgxowZSE9PR2xsLDo6OvDII4/gq6+++sUJE3DNl1BNjohAIMDMmTNx0003ISoqChaLBcePH8fg4CDi4+Px6quvYv78+XjllVdQWVnJjsFdT+SgeTyec0iLFJxQpp42atL2FwgEmDx5Mh544AEkJyfjo48+wgsvvID6+noA3k0c6Wd/PBauRDGPx8O4ceOwbNkyrFy5EuHh4azZ2JNPPsmuXygUMgXEzs5OtLW1YcOGDfj222+91JB8uRD+IFzcKhYw1HuJiJZisZgFyk888QSamppQV1fHMp3cY3GrHNw55sLD6DrJaSJYAeBdpaC5SE5OhsFgYNKf3PtHWXF/FRAKUrhrx+l0wmKxQKfTIT09HWPGjEFtbS08Hg8yMzNx7733YsGCBVAqleDxeOjr62P9dbiQPd9KFfdafeEqQqEQUVFRbGw+n4/o6Ghce+21mDBhAt59911s2rQJ99xzDxuXOwf+xuBeJzAEP0tNTQWPx8NXX33F+HW+1TTi0EyePBlZWVmYN28egoODUVZWhnXr1uGzzz5DZ2fn/4TTMRLG4/Gg1WrZWqVn3W63o6WlBdu3b2fzWF1djYkTJ0KlUmH//v147rnnkJeXBx6Px5zRe++9Fy+//DI+/fRTfPDBB+jq6mJZdlqD/lT47HY7tm7divz8fPT09HhxE4ChexsUFAQ+nw+TyTRiIj1OpxNarRbBwcFM0MjpdEIsFmPGjBlITU1FZmYmEhISIBKJ8NBDDyE/P5/tGVFRUUhJSUFMTAyrfnITOf4sJSUFd911F3p6evDMM88Mm7Ds6enBt99+i4CAAK+eIf6MEkQDAwOMd8bj8aDX6zFnzhxcf/31SEtLY8mplpYW5OTkYP/+/SgpKWH8j8mTJ6OzsxO5ubnn9fXa2togFApZw2KSlN6zZw9Onz6NCRMmIC4uDosXL8bGjRvR1NQEm82GGTNmwGg0Ijs7G08//TR+85vfYOHChRgzZgySkpKQkJCAN954A8eOHfMaz3c/9p0zCojUajVTx3Q6nfjyyy/B5/OhVCqRnZ2Nrq4uKJVK9Pb2sgqRSCRCZGSk1/uPKlWhoaFITk7G6NGjUVVVhcLCQvD5fOh0OjidTrhcLlitVphMJr+JGpVKxfhOFHj9r+wdIx60yGQy3HzzzQgPD2cEzBUrVsDtduPOO+/E3//+dyxfvhxSqRR33HGHF59gJGzChAl48cUXkZCQgNOnT+Pdd9+96Gy1wWDA4cOHERUV5aXoIhQKz8GDHjp0CO+99x52797NYCYjAd0ih50sKCgI//rXv5CYmIjS0tJLQq7UarVISUmBWq1mePuuri48+OCDI6pAQw92WloaXn75ZVx11VVobm7G4sWLkZqaitraWmi1WlZqvtjGliSRzXVkQ0NDkZGRgdzcXHzwwQeQyWSIiYmB3W7/QaSzgIAAxMbGQi6XnwM/IaPA895770VWVhbi4+PZPQ4PD8ftt9+OgoKC7yTf+HM0XwUs+tnj8TC1ElLd2bFjB1577TXU1NRgyZIlePzxx3HzzTdDKBTimWee8ZtdO18DNnJQuWpcFLAAQ/vIQw89hPT0dBw8eBBvvfWWV1UQ8C+rTEabP/c5GTVqFJ588kksXryY7SOBgYEIDAxkLxvKnu7duxcbNmzA6dOnUVNTw6pPXDWx88HR6DNUJSDSZmRkJFMno74Jt9xyC2bNmoXPPvts2AZ6vtdM//bt+cG9bmDI6U5PT4dGo4FSqUR/fz8qKirQ39+PGTNmsApJVFQU7HY7+vr6YLPZ/O4vvsR8GsPtHmqc1tDQgODgYKSnp+P+++9HcnIy5HI5li1bhujoaBQWFkIgECA1NRXt7e2M2zhctcjXuBUWh8MBuVwOlUrFeqLQPUpLS8PRo0fx+eefIzw8HGlpaWhqavJqFuorJ01jcueTx+NBpVKxoL2oqAhms5kpnRGcxuPxIDQ0FLfccgsL8qnyMm7cONTX12P//v0sw/tLMblczngU1JU8JCQEHs9Q40lK9qSlpSE6OhpmsxktLS0Qi8WQSqXo6+uDy+VCe3s7Q0skJiZicHAQ7e3tkMvlXsIUMpnsHGiMx+NBV1fXsH7MwMAAlEoly4hfLI+XayaTCc888wymTp0KPp+P+vp6tLa2orW1FWfPnmVJg02bNuHEiRMQi8WIiYmB0WhEeHg4wsPDERYWxrhV3P3Pn8XHx4PH42Hjxo2wWCyIj49Hamoq4yw3NTXh5MmTMJvNqKyshNFohEwmOy83hZJKNDbJv6elpeGmm27C7P+/+TIwNNfR0dGIjIxERkYGmpubWfPa4ODg79TY22KxQKFQsO721DC4tLQUL7/8Mh599FF4PB5MnToVYrEYeXl5SEpKwtKlS5GdnY177rkHXV1d2LRpE66++mrWgHvFihUQCATo7+9HYWEhG4/EB+j66HrJlEol+Hw+pk6diubmZiQlJbG5XrRoEYKCglBUVMS4RG63myneRUZGMvQJmdvthtVqhVgsZnL9XV1dUCgUXmIKBOMl5UKPxwOJRAKj0cjum91uZ0kz32Drp1QgHPGghbScaaE999xzbDN/8803Gbdj+vTpuOOOO/D888+P2MWPGjUKf/nLX5CWlga3240PP/wQZWVlF318hUIBo9HI5PX27t2LtrY2VFZWYt++fVCr1bj11ltx3XXXoby8HNnZ2ZekipScnAytVou+vj7MmTMHCxYsQGNjI3bu3Dni4z388MO46aabkJCQwO5lfX096uvrMX36dLjdblx99dUYM2YM2tra8MEHH+DgwYPfe5zGxka8+OKLuO+++yAUChEXF4e4uDjmaB07dgx33nkn3njjDcyePZvBZy7GAgMDodPpUF9fzxyhqKgoxllQKpV47LHHEBkZiccff5xl+L+PEeyI1J/8mUgkwty5c5GVlQU+n49nnnkGTU1NSEpKwg033MA6Mf8SjRugkHGdR6PRyPTwa2trsXXrVhw9ehRutxvvv/8+urq68I9//APLly9HQ0MDPvzwQ7S2tnoRkbnHI5UUgkRwKyxc54/P52PixIl44oknMHnyZHR0dODzzz9HaWmpV9BA536+vcX37yQ3CQBnzpxBcXExLBYLzp49i9zcXCgUCvz2t7/F4sWLGTywubmZzZVEImFOBSnE+BK3/VV7uOfgdDqh1+tZVnbFihW4/vrr0d3djf3795/DlfHltPjOqe89HRwcZC9npVKJZcuWYcWKFUhMTIRGo0F7eztaW1shEokwffp0DA4O4sYbb8ScOXNQXFyMDRs24Pjx417EeN/r4xopgzU1NWH37t3QaDQICwtDZmYmMjIyMDAwALvdjg0bNuCLL77AnXfeiVGjRoHH4yEkJIRVhKhSPlxVk1sxox4oarWadQg3GAzo7OzEuHHjEB4ejk2bNiE6Ohq33norYmJi8Oqrr6KgoOCcninchn3+1o9AIIBKpfL6PVUCgaFnRqvV4s4778Qf/vAHdHZ24p///CfKy8uRlZWFBQsWMDw6nfsvyShgIXhMf38/dDodZDIZcxCDgoJgs9nQ09MDj8eDmJgYtu8nJyfj6quvxpw5c9DR0YGqqipoNBp4PB5GZKf1R2pT38fEYjG0Wi34fD7q6upGVDHS4XBg7969yMzMxNSpUzFmzBicOnUKf/nLX7Bv3z5ER0cjLS0Nx48fh8lkQmpqKgwGAwwGA+u7RpwOAMMmWElti1Smli5divHjxyMtLQ1arRYez1B/rfvuu4991mAwQCKR+G1iyTV69rhCQlKpFKmpqUhMTIRQKERvby9qampgNpshk8kQFhaG+fPne+0FAwMDSElJwYIFCzA4ONRw0V+AyOfzoVKpGMSdz+czLtrBgweRmpqKq666CmFhYcjIyIDZbEZrays2b96Mf/3rX2hra4NUKkVFRQX+8Ic/4E9/+hOuvPJKSCQSzJs3D/n5+Ux1rqOjg4l+EMfIarUymWlgqPIjEAjw/PPPIz4+Hunp6YiMjMQNN9wAPp+Pb7/9FiaTCQaDAS6XCxKJBAqFAkqlEtHR0Thy5Mg51ycQCNDQ0IDKykokJSUhIyMDYrEYNpsN9fX1jAvkdg/J/cfExCAmJgajRo1iSq4tLS2oqqpiBH2uxcbGIiMjA19//fXPv08LAERHR0Mmk7EFtXXrVvaic7vd+Pbbb/H444/j9ddfx+23347a2lp8/vnnFzVmQEAAQkJCcP/992PSpElwOBx48sknsW7duhGB1tTU1GDSpEmYP38+3n//fUbkIiP9fLJLEYFSZ1On04nf//73uOWWW+DxeNDe3n5OSfJiLSMjA8uXL8eoUaMA/PcFGxkZiaioKOYIhYaGIjg4GKdOnTqvjOP5rLu7G3//+99x8uRJLFq0CGKxGDweD7t372aSs9zSpEAgwBVXXIFTp0794OtrbW1FTU0NwsLC8Pzzz+P999/HkiVLMDg4iClTpmDJkiUsQ3no0KEfxNuhOSLyK6n7kInFYvz1r3/F/fffj4aGBrz00kt49913MWrUKEyfPh21tbVMbemXaP42O+6cKZVKqNVq1miMq/g3ODiILVu2IDw8HM8//zyWL1+OY8eOeWU7uc4G99i+43IrmCKRCOPGjcPvf/97TJ48GS0tLXjttdewbds2v9l/f879+a6zpKQEzz33HMLDw3H48GFUVFRAIpEw8rhOp8OcOXPgdruhUCggFAq9nGku9txfZdVXzQvwDgQHBgZw7NgxrFixAitXroTBYMBdd90FkUiE+vp6nDp16pwXvW/Q4htkchWqfM9p/PjxuPXWWzF58mQWSAYGBrIGZXS8+Ph4GAwG1NbWskDM3xxzK0bcxozA0Mt/7dq1KC8vR1JSEoKDg1nWs6ysDLm5ubBarbj11lsBDCECqLEarRtynGhNUMWIAlqaW7oPNpsNvb29CA0NxTXXXIPc3FzMmzcPEokEiYmJmDp1KtLS0pCbm4svv/wSfX1953CufAnP3CqLx+NhAhMERaNAl+ZJqVRi1apVuOOOO9DU1IS//vWv2LhxI0aPHo1FixahubkZR48eZbh1ggBdjIT7/4rxeDyvgAUYyqST8AHxM2pqalBfX4/AwEAYjUakp6cjOjoaKSkpWLZsGSQSCXp7e7Fz507k5uYySVdSEKSKLZ/PZ40Tv6uFhoYiKCgIAJCdnX1RHFBfGxwcxLZt23Dw4EFcddVVmD59OhQKBSZNmoSYmBiEhoaiv78fAwMDSEhIgMvlgkwmY/wth8OB1tbW7/SOsVqtKCgoYL6aRCKBXq+HQqHAwYMH8frrr+P48eMM1kncE3+BQ2BgINvPaW8RCoUQiUQYGBiATCZDaGgoZDIZampqsHnzZuzatYtVUrOysjBz5kwkJiYypUWBQICMjAwkJiZi9uzZ+OSTT7Bv3z50d3ezRK/D4YBOp4PBYGACBCQ+QqJDmzZtQmpqKsLCwth759tvv8VXX30Fh8OBhIQExMTEoKOjA+Xl5fjkk08QHByMCRMmwO12w2AwIDMzE1VVVejo6GDVnO7ubvT09HjNh1gsxtmzZ3HnnXfiueeewyOPPMIqIO3t7di2bRtOnz4NjUYDp9OJuro6JqjR3d2NwsLCc0SZaH8pLCyE1WpFQkICli9fDoPBwAL86upq8Hg8qNVqhIeHIz09HRkZGYiLi4NIJEJfXx9aW1sREhICiUTC5g4YKgxMmzYNUVFR2Lx58y8jaMnNzWUZDR6Ph5dffhn33HMPLBYLe8nt3bsXFRUVbNO5GFMoFJg6dSruuecezJ49Gz09PXjggQewefPmEduY3W43CgsLvcp+P7bNnz8fPT09KCoq8lLFOXPmjF8C2A81rVaL22+/HWlpaUwJjh7YO++8E8HBwejt7cWGDRsYge3UqVMXBcEzm83YuHEjNm7cOOxnuM5LcHDwDx4LGOpSm52djRtvvBF33303Zs2ahdDQUAQGBmLBggUYGBhAS0sLnnzyyR88tzKZjCmfXHHFFdDpdNi9ezdTdFm6dCnuu+8+WCwW/OMf/8DOnTtx11134dFHH0VQUBC2bdvGSu3/LxgXdunxeBgnbGBgAJGRkZg0aRLOnDmD1tZWtv4PHDiAU6dOQa/XMxgeGb0IuRAxXw4It6eKXC5HZmYmbr31VsydOxeVlZV47bXX8PXXX18wu+ovgOASxck6OjqwdetWL7Uo2qPImaRsvsPh8Ootwh2LWxHw5a/4Bk6UrSeHf8OGDRg1ahT++Mc/IiwsDJ2dnSxoqa+v94vbP59xqyBchzs4OJjBOy0WC06fPo0zZ87A4XAgPT0d48ePh8ViwcaNG9Hc3Iyuri7k5OSgoqKCvTuGEwLg/sytXJSVlaG6upoFuyaTiUkCDwwMQKVSsc8HBQUhMjKSVZG5jqlAIDgnO+xvThobG7Flyxbcfffd+OMf/4jCwkIkJCQgKCiIVU/Ly8vx9ttvo7y8nGU2AwIC2H33B4GjpnkEnevp6YFGo8HChQshFAqRk5ODtrY2BAcH46qrrsLvf/97yOVyfPnll+jq6sKqVatw6623YsyYMdi+fTtycnLYGh5OFernZiTM4WsDAwOsJwgFLTk5OTh69Cjmzp2LlJQUpKSkQKVSITQ0lFUPW1tbUVBQAKVSCb1ej+7ubtYjpKqqinGftFotnE4nW1sXsqioKIhEIuTl5WHnzp2XZO6dTifOnDnDiNxLlixBWloagyYBQ/tMT08PGhoaUFJSgqamJpaNv5APRmu0ra0NJ0+ehNFoxG233YbQ0FB88MEHePXVV1FeXv6dzlWlUiErK8ur8/zg4CCEQiHEYjEEAgGio6MRFxcHgUCAuro6FBQU4MyZM6xa0dLSgo6ODhiNRvT09MBms0Emk0EikbDEU0tLC9xuN5qbm6FWq3Hy5EkG6dRoNIyrQT2d6L7U1NTgnXfeYeiKrq4udHZ2QigUIisrC4GBgZDL5UhPT0d1dTXa2trwzjvvYNeuXWhvb0dZWRlUKhUUCgWbd6vVek6fHYFAwKqBH3/8MUpLS7F48WJkZmZi3LhxKC4uxvvvvw+TyQSlUgmTycQqQgBYksnfWiBJ45aWFpw4cQKZmZlITk7GtGnTwOPxUFdXB6lUyni30dHREIvFrLmkWCxGXFwcjEYjQkJCoNPpUFhYCLfbjfj4+O/UvPNS2ogHLS6XC01NTQxPvHz5crhcLibTCQzJzpHyAakZ/BCTyWRYtGgR/vznP2PUqFEoLi7GCy+8gG3btv0iMkm+dt999yEuLg6BgYFwuVy47rrrsHfv3hEdIygoCDExMZDJZPj222/x29/+FiaTifF57rzzTvz2t7/FN998M6LjXsja2trYQzJ27NiLOpbb7cZrr70GnU6HyZMnIz4+nkHObDYbcnNz8cYbb7AM+w+xs2fP4vjx48jMzMRdd92F2267DUVFRTh16hQmTZoEjUbDnoFHHnkE999/PyupHz9+HOvXr79o8Yb/ZfPnHNP9dbvdjHCZkZGBkJAQrFq1ClKpFCdPnoTFYoHNZoPRaASfz4dYLGZNDLlqQv7w2SSrS+TPgIAAlr1bvXo1pkyZgqqqKrz99tv45ptv0N/ff17uCvfcuZ/zfUFxM2xUnqffcSVrZTLZOc3B6JzJ2fWFf5FRwOP7Ny4E7vDhw3jzzTcxb948mEwmDAwM4Oabb2Z9Aqh3AfBfkuyFrt3fuWi1WqSlpUGpVOLrr7/Gq6++itLSUkyaNAkZGRnweDx444038MYbbzD+ha+88XBVH8o2ch1AbgNJ38wxQbAGBwdZNUcsFrP3EHf+uOpgviR87vwKhUI4nU588MEH0Ov1yMrKwoQJEyCRSCAQCNDX18ee4x07dsBms7H7ym0KSnPMneeBgQEGCezv70djYyN4PB6uvvpqTJ06FeXl5WhtbYXBYMDkyZOh1WpRU1ODmJgYPPvssxgzZgwGBwdx8uRJrF+/HlVVVQyq+ksIWMhI4YjrdHIDd3oeW1pasH79evD5fMydOxfx8fEQiUSMB9PT04OjR49CpVIhLCyMqbwZDAYAQw6i0+mEXC5nvTtI6ep8z4VKpWLw6pycHJw9e/Y7VWe/i5Gq3cDAAOvf1NPTw/gLSqXSC0oqFosREhKCkJAQBAUFYdeuXfj666+/M2JBLBZj+fLlePTRR2Gz2fDVV1/h6NGjLAn3XaS0hUIhxo8fj6SkJK9nmZ6zgIAAVsEJDg5GYGAgxo4dC4/Hwxp4R0REMOn7xsZGbN26lcmIJyUlMT9lYGAAaWlpMBqNsFqtyMvLAwD09fXBarWCx+OxSimXuG+1WnHw4EEcOnSI3Zvo6GiMHTsWERERjINIRPWcnBycOXMG+/btQ1dXFzweDzIyMpgIVX9/P+x2OwusbDYbe7bNZjNbo2fOnEFBQQEMBgNWrVqF5ORkxMfHM0lprhHEzF9Cc2BggH3earUiPz8fe/fuBY/HQ2BgIMaNG4fY2FhIpVIG7Tty5AhycnJQWloKs9mM4OBgjB8/HtOmTcPUqVMRExODuLg4RvjnVox/ChvxoMVkMuHhhx/Gb3/7W6xevRp6vR6rV6/G6tWr/X6WiEff12QyGbKysvD4448jJSUFRUVFWLNmDZMO/SXaxx9/DGCIgPj444+joaEBBw4cGPFxuC/Q+fPnMwnPuro6bNiw4UcPWADgwIEDuO666xgm1Vfu9fvamTNncNNNN2Hp0qXQ6/W46aabkJKSgmPHjuFPf/rTiEhIV1dXo66uDsnJyWzDHj9+vNdn+Hw+I9pWV1fj2LFjePvtt1FWVvaLIs36M1++Ateam5uxc+dOBAYGYsWKFdDr9bj//vsZxnZgYAAikQhhYWGor6+HWq1m3YJ9+Sy+YwFg6lNBQUFYsmQJ7rjjDowbNw5nz57Fhx9+iK1bt8JkMjF4li+B2l//EH/8HIJLAP9VLKPz9/0ONXmk44tEIgZZItgAGfFzhoOr+VYj6BikTJOdnQ2JRII777wTlZWV2LVrFzs3eqn6in/Q9XPFC7h9E7jzTC9ju90OHo+HzMxMXHHFFcjKyoJIJMIrr7yCt99+m0GWCI5Fjr2/4IsI8DQ2d064FRKPx8MqKCQXOjg4CIfDgaqqKkZU1Wg0kMvl7Ngej2dYR5TbqJICAJfLhbNnz+Lpp5/Gxo0bMWrUKMyaNQsTJ07EmTNn8OKLLyI7O5tVWOhc6X7QPfcNUimI4fF46O/vx/HjxzF9+nSm9hQfH8/WikQiAY/HQ2xsLMLCwtDf34/y8nLs2rUL33zzDfLz85myGdlPmSUdKaN7FRQUhMHBQRaA0Lz6kt4LCwths9lQXl7OKmzk/MlkMgQGBiIsLAyBgYEMFkhZcjomcSBIFpYLxaS1C4AFnSkpKcjIyIBKpWKZfzIejweZTMYqqmT0vFNCZbiqJzUWNJvNsFgsqKioQG9vL3Q6HVujlZWVcDqdMBqN0Gg0bA3odDrU1NR854BFKpUiKysLd999N86cOYMHH3wQdrsd06ZNg0qlgsPhgNFoRGtr67CNMKlPkc1m86IMkBIeVRiBof5NJSUliI2NhcFgwJw5czBnzhzGMzp58iSys7ORl5eH0tJSptim0+lYI8/Y2FjW34XrJ1APKbPZzKTz/XFu6BoCAgKgUCgYnIwrNW0ymWC1WtHf34/Ozk7WV8bj8TCCPFXrALDGvb4QsaCgIAQFBbEqyD//+U/GQ+HuecB/A3F6L13oOW5qasLevXshFAoxZcoU6PV6qNVqJpZAQQ3JqNM5dXV1IT4+HmPHjmWJgaqqKrS1tTE+3y+m0gIMOWtPPfUU00iXSqUYO3YsK9M3NjbCbDYjLy8PX3311fc+vkKhwJIlS/Dkk08iPj4etbW1+OCDD/Dpp59eNNzs52BGoxELFy7Ejh07zttP5YcYEWndbjfmz5+P+fPnAxh6CezevRuPPfbYiI73fc6LLDIyEhqN5qJJjWazGf/5z39Yn5s//elPEAgETOrvYqt1mzdvRk9PD373u98hIiKCNWsiMrbFYkF9fT0aGhpQXV2NLVu2IC8vj5V3f8nmr8rCVdtyOBzIy8tjza9mz56N8ePHIzAwEEqlkr2wc3JycPLkSRw7doxBYGgz9edUcyswarUa8+bNw2233YYxY8agsLAQa9aswfr169HR0cGy4P4cSrILvTh84Ti+imn0O3KA6NlzuVxMwIRbdeA6S+SkcR18royz71zL5XJ23JaWFiQmJmLSpEk4dOgQuru7WYBFsBou3It7vdy54BoXckdqaEKhEIsWLcK8efMYlv3jjz/GJ598AqfTyZrLUXDD7bzNFVXwDRR9r5FL6CWoF60n+s/j8cBkMrEgkjDbdDy6NvosfZ9bYaFAkQvPq6ioQEVFBXbs2IFTp07h7rvvhkwmY1Uzh8PBhA/IqeNyhHxhi3StVB06evQo3nzzTcyePRujRo2CVquFTCZj99NqtaKrqwslJSU4ffo0Tpw4gYKCAi9xFn9Vol+CqVQqdl02m431v3A4HOfsMZWVlV7ysMCQ8z9nzhxkZmZCKpVCpVIxHoLFYkFBQQE7jkwmg8VigcPh8OpZQeRzUglzuVzQaDSYMmUKxo8fD7VazYIqMuq8Tn07CH5G8sitra1sn6P/uPeMeH8A0Nvbi/7+fgiFQgQGBkKr1aKnpwd5eXmoqKhASEgIZsyYAYPBAB6Ph8OHD2PTpk0XnFuRSASpVIqUlBRkZmbilVdewRdffAG1Wo0xY8awKoXFYoFQKIRGo4HZbGaJI67JZDL09PSgurrab3KHAnCPx4Pa2lps3rwZTqcTWVlZrC9bd3c3cnJykJ2dzSSmqVJusVhgsVjQ0tICoVAIqVTKkh+0jwJDss2hoaHo7e2FxWK5IPxapVLBYDBAoVBgYGAAzc3NqKioQGNjIyorK1FXV+d1T3U6HWJiYtg+NlzzYTKlUskqeLRmbTYbTp48CYlE4vWeIDU7p9MJqVQKiURyQSSI1WpFYWEh2wNnzZqF6Oho6PV6BgkbP348JBIJ40yLxWKm/tvb2wuz2YzGxkY0Nzejvb2d9YL6qeySNZe0WCx49tlnWRbjgQcegFqtRn9/Pw4cOID6+noUFxd/b8y+UCjE8uXL8fDDDyM+Ph51dXV47bXX8NFHH/0/gf+nDVIkEuHDDz8c8eN3dXXhyJEjiI2NRVxcHNxuN+rq6nD06FG8/vrrIz7edzWCOABDG91IqrC4XC7k5OTgn//8JxYvXoyYmBjk5uZedNDS09ODzZs3M9J/QEAA0tLSkJ6eDo9nSHFl3759OHPmDINO/r9k/ojtBHsg/lt5eTnq6+tx4MABxMbGIjU1FeHh4bBarSgrK0NVVRWam5vR1tbGMuB0LN+XBB17cHAQgYGBmDdvHn73u99h7NixyM/Px/vvv4+NGzeis7PTC77ANX8VHBrP37X5VjzI8eB+j0v0JuEGytT5QrzoOP7m0PdcfOfXVyUrMDAQQUFBOHHihBep3heywe2X4i8Y4t4/Gq+7uxtHjx5FeHg4oqOjIRQKcfbsWWzZsgUbNmxAW1sbc/C4FQg6b26gwoUN0t99oWPE8ePONX2XG7hQLw1S06GXPsG2/N1LbuDEHY+bCafKb3Z2NstCkxPc3d3tJRxwvqCXK+lMWfeWlhZs27aNwUeCg4MRHh6O0NBQBAQEoK6uDoWFhaiurkZXVxf6+vq8oGe+9+mXABELCAiA0WhEYGAgU9UjsYrvIytss9lY1lwqlUIul0Ov17NqRXNzM3tmrVYrc9K5FTkKIDQaDasWREdHY8KECTAajXC73az9AX2euBHUn4SUsWJjY2GxWNDa2urVrd7lcqG1tZX5OP39/V7NUKmyc+WVVyIwMBCnTp1CVVUV8vLy0NPTg9zcXGRkZMDtduM///nPOcEb13g8HiQSCYNCCYVCfP755+jo6IBSqYROp0NtbS17fpxOJyQSCYNE2e32c6pKXD6GP+Pz+YyDY7Vacfr0aXR0dKCkpASjR4/G4OAgTp8+jYKCAphMJtbbRCQSMWU4u93OYHJlZWVob29HYGAg44gBwLRp0xAaGoqmpiYmbX0+hU6CcJNEfHV1NZqamrykldVqNbRaLcLCwhAWFga1Ws3mlzrKU0Dgu59TTyuCK3LlpyUSCfh8Pnp6elgDYJKFp+v/LvD1/v5+nD59mjU3nThxIhPMSk1NRUxMDBYsWICenh4mZKFQKOByubB37140NTWhrKwMbW1tDC49Uv2GfohdsqAFAFM+6enpwe9///uLPh6Px8OYMWPwl7/8BXFxcWhqasKrr76KdevW/T8RsABDL9T29na0tbWdt/PrDzWTyYQPPvgA1dXVGD9+PNxuN3Jzc3H48OERDRS+r508eRJvvPHGJcsS2u12HDp0CIcOHRrxYxcUFKCgoAAAsH379hE//i/FyIGnDZ0y3U6nE9XV1aiursaJEyeYAo7ZbPYKArhBiT+jFxcpzdxzzz2YNm0aqqqq8NFHH2Hjxo3o6uo6x9njKl35Gvfl7Ps5fwEF1zGnTDodhwIK6nhNP3OP5fvS8w0ifB1g+g5l9N1uN9RqNes8XVVVxdYmF5bkewzuPfI133E8Hg/rZVBXV4eUlBRIpVKcOHECR44cYWpdvvwZmp/hgi7u/eDe4+HuD/3MFWMoLi7Gu+++izFjxrCqJkFTfJ177njc/3MDITJaf1arFUeOHGHE++7ubhaQEqyIe998O5APt3ZJfYgI6AqFgjnrlN3mzhkFPNyeNsMF4j9H4/P5GDduHIRCIdrb29Hf34+2trbv3QeFeGJ1dXXo7OzE4OAga8BXV1eHxsZGdsyOjg7WfZxb+ZRKpVAoFFCpVBAIBBAIBIiIiEBCQgJkMhna29tZry8KWChLThA1iUSCiIgIJCYmorGxEcCQQpdUKmWKWgMDA8z5p+/RvTUajbjhhhswdepUdHZ2wmKxgMfjobGxESUlJaiqqkJJSQksFgvjd/garS2ZTMYy+gR78ng8kMvlLFkklUpZooMqGyRP7HsPLgQjIl4iEemJE1ZeXo6mpibGL2lvb4fJZIJCoUB4eDiDRtIxAgIC0N/fj66uLohEIoSEhIDP58NgMLD95fLLL0d4eDhaWlrA4/FQX1/PmnPSHHArknRNNpsNra2taGxs9IKTGQwGpKenIyUlBXq9HlKpFA6HA9XV1QDg1cyWu79QM1QK+ggmSkkOqVSK4OBgdHV1scaxAoGA8bcMBsP3EkxxOBwoLi5GU1MTDh8+jODgYGi1WgQGBkIqlSIgIABisRihoaEwGAywWq04evQozpw5wzgydL/dbrff/fLHsksatIy08XhDHecpG7Fp06b/iQqLzWZDcXEx8vLy0Nraesm5CCaTCWfOnLlk0W5HRwfWr1+P9evXX5Lj/xBzuVx4+OGHf+rT+NVG0HwzzvQ7rvNMgQj9jrDIXOM6j75VAq4zy+UlREZGIiIiAvX19fjmm2+wfft2dHZ2niNJS9UPLkF7OPPlufhz/un3lHXnckAcDgdKSkpw6NAhVFVVwWazMYeXO1/capS/OaVx6e/c/jQEQfN4POjr68P+/fvR0tLCxAK4Mr90DH8VAn9j+l5fTU0NmpubsW/fPvB4PMYR4t4rrvPvG+D5/p57bb7n46s05gvtojmsq6vDJ598gq+//ppl2LnyuNygwjcY4xp3/snx5eLyW1pa2Pe5lR6uqpy/KgjXuFwk7lqx2+3MceKuMe5c+a5XcpgvNObPxQICApCamgo+nw+JRMJ6bn0fo4DDbrejvLycwb60Wi3MZjNOnz7tRdLu7u72qpYKBAIIhUIolUoGmeHxhlQLVSoVy/IT4VqtVkMoFEIulzM1KIILarVaREVFITU1lfkPvb29XhUEXw4brdHY2FgsX74cK1asgN1uR29vL9RqNQwGA6RSKTweD1pbW5mAie9aJjVDoVAIvV4PYAhxQRwRWnO1tbWw2+2szwnxN8i57+/v/94BMQUqBIUaHBxkTUNtNhurDlAVjFoieDxDEGJ6dgnO19fXx4ROAgMD4fF4mFABACQkJDDOi1qtRlVVFVMgo+efrkGlUiEkJARWqxU1NTWsYz0wtKdGRkZi7NixTGkrIiICIpHIC5pps9nYPsC9f9QglhscCgQChIWFsd8HBwezZJbRaER1dTW6u7sZr/e7rHfakyjIIIWws2fPsqo+JTgUCgUSExMRFxeHjo4O5s+63W4WmBPcjpANP4X9rIIWt9uNo0eP4ne/+x3D/f3UAQswVDbesmULKisrWTbgUprD4cD7779/Scf41X61H9N8M9fkXNGLhPs5biDDfRH4voy5lQcuSXZwcBDHjx/HU089BafTiZycHKbU5uug0hi+cIcLZQ99z8e3IuNbUQCGOFZ79+5FY2MjmpqavHrT+M6VL6eD5sXffHADEapcAUNO0aFDh+ByuVhVxzcY4goKcCsWdFyaY98AlP5G8C9yBIhDwj32cI60v2DBNxChefD9LJejwj1eQEAA+vr6GFzHHwyN+3kak47PFWLgVjS450BZXaFQyOaAW2mjz/iuK3/Xzx2f+o9Qp3O6r/6qC9wAjxvI/FKMx+Mx8rLH42G9a77vMQIDA5lT6nK5oFAooNVq0d7ejurqaq9jms1mdo/FYjHjLVHAQjBK7h5Fz51Wq0V8fDzsdjva2tq8enYIBAKWsR83bhzjqVLl53z3TaFQMJhraGgompub0drayppHRkdHo729HR0dHQz5wjWpVAq9Xo/6+nqIRCIolUr09fUhOTkZycnJDBJEAYJer2fNO4kDRwEBV8SDjMvronVPDR0BsOAHGPKjLBYLVCoVtFot1Go1410QHMrjGZLDdzgcTA6YYGnkUAPeiR1fuBrw355cV155Jbq6utgcUQ8+4rbq9XpUVFSgp6fHK0mRkJCA6dOns2qaQqFAYGAgHA4HmpubWbVscHAQQUFBsFgsXhwzHo8HrVaLkJAQpnrY19cHtVqNzMxMpk42bdo0iEQiVFdXM9lhYEhKu6Ojg6FflEqlX1+Y9iza3wnuRiIvTqcT/f39rPJeUlKCiooKJjJAZrVa2dx+32rmSNvPKmgBhlSF3nvvvZ/6NLzM4/Ggra3tR4NPDQ4Oori4+EcZ61f71S6VcR1kf04mGRdKRU6Dr6PqD4rl63xzKxalpaWorq5mTqVvRYfMFzI1nCPN/Tz39/4CKn/fp8/W1dWhpaWFVUR8r4O+z32Bch1vrhQlzZlv4EfjEwTD4/mv/DMFNtQM1fe8uSpY/u4XN2AhJ4X7M9dp4lYduJUSbrDFPV9fiJy/sbmVEu6xaR65QRgFcNwXO9dJpN/7Vsy4XBlat75cLAomqLriK2rA5edw4WLcdeb7WeI/+FYj/R3H39zQvP0SpI89Hg9TbTpz5gyqq6vP23ndn1FDw7a2Nibj2t3djYqKCnR2dp6DZOA6a8TtkEqlzHl3Op0sy0/OOTDUTDEiIgI1NTUoLS1FZ2en17EkEgnGjRuHKVOmID09HSdPngSA8wZh9HyqVCpkZGQgJiaGiZZYLBYmvR0REYHOzk7mTHO/r9PpoFarIZPJUF9fD4/Hw0QE0tLSGNmceBk6nQ5BQUFMXpyCFIlEArFYzAjlZrOZjT1u3DhERUWxhoktLS0wm804efIkg7UJhULYbDZ0dXWxXixyuRwqlQoajYYlOmQyGYRCISQSCauO0LomqB33+RaJRLDb7WhoaGD3sre3l4niAMCiRYtw9uxZ1NbWorq6mn2W+vWQND0dV6lUIj4+HvHx8QgODoZGo2Hn39rayriIdH8zMjKQmpqK9evXnxO0GI1GpualUCiwf/9+mEwm3HjjjZg4cSJ6enowadIkiMViPP3002x9u91uREZGMggaIZCG6yNIQgRSqRQajQbBwcHsPlosFqagRiIA7e3tfvm8P3WwQsbzfJeUIfy/oH61X+1X+2nsOz62/3P2Q/YR34zxpbAfY4wfa3x/AYpvUOdvnPN9hgtju9A5+gsgv+94F2P+zt33HPxB0s7n9A937PONcaHv/FQ2XDD9czEKEuPj4xlntr+//3tXkqgvCHEKeDwe41ZQk8rhgjsiT5PzS3NKiQONRoPQ0FDm9FJ1hcvDI+Pz+QgJCUFwcDAUCgWampqYQ3q+OfB4PJBIJIiNjYVer4fL5WLVWYIdEYGc+oVwv0/NHHk8HsxmM8vAExHbbDaz5qwkckLwTm7gThDawcFB2O12xt2Sy+VMcIAgZlarFS6Xi/EkqBEsJWnoPhDkDPhvooirLknVW26ygCslT5AxEisg6Nr48eOZ6hodmwIV4itRpYUCP5oHYCjQVSgUEIlEbK1QhY1kr00mEzo6OuB0OhESEsLuKTcIpuaSJLoiFArR0dEBl8uF6OhoaDQaOJ1OBgOrqKhgUC0AiImJQU9PD7vfKpXKC8roz4j0T+PRfkdBGc3rcLLVw9mPvYf8GrT8ar/az9B+js4G8Os+8qv9av9L9nPcR37dQ361X+1/x37sPeRnBw/71X61X+2XZb5KR8B/4TtciAThcEnNBhjKfBG8yde4neh9u9LTuFx+BDUM5Bplwkj9x9+xCVrE1eQH/GfvqQmib/mdspwEV+NeH5eHcj7zHY97fUqlkikPcfkv9DlqfMY9H8pwUubZ3xz7Qqq4UDf6+UKQLt8KCLfXjL8qha9Ygr9j+utX43sskUjk1UfHt2HlcN8jI1Iw9ZjxbQR3PvOVkv6hdj5lu+GqR78EeNhwJhAIoNfroVKpYLVa0d3dDavVel7HSiaTITo6GgkJCZg2bRpSUlLQ3NyMXbt2oaCgAHV1dSwDnpmZiQ0bNrDvSqVSXHXVVYiNjcXJkycRFBSE1tZWlJWVeUGXKIvP5/Oh0+mQkJAAvV7PoG3+jDgmJpPJ7zrh8XiIioqCXq+HzWaDxWKBUqlEbGwskpOT2XMrEAhQW1uLqqoqJr1LlZP09HSEhobijTfewKRJk/DII4/gjTfewN69e73WTWJiIpKTk3Hs2DEvef6QkBAkJydDpVLBbDaju7sbYrEYBoMBOp0OSqUSdXV12LdvH+NcTJw4kUHgqHfTxRg979/lWfroo4+wa9cu7N+/H3a7HdHR0awiRsGwxWJhSn0qlYpxZogXFBQUxAQCBgYGEBQUhJCQEFYtmjhxIiwWCz744AMkJCQgMTERcrkcUVFRSEhIQEBAAEpLS1FQUACr1QqNRsP2joGBASZ5TaIEvb29aGtrY++A+fPnY9q0aWhra8ORI0eY+iMwtJbj4+NhMpmg0WgYBE0mk2FwcBDNzc04c+YMa8h5IaPKksPh8Cvi8FPYr0HLr/ar/Wo/unEdJy6HgIwbbJBz6c8h5DqYRFQm5STuS4z7by4vgbqRc51krqNNkAUKarhETy6ZerhO6v4cbO45cx1fcn65Drdv40HfeTrf/JI8OMEj6Hsko8ptvEkKRfTyl8lkjJzJDWZ8gz9ffos/ThB9j7D/RM4nGAiXy+IbgPq7Vn+BIXdc7jEIZsLlQHGbzpH5U2ei31NATYElgHNIzXQ8Wq80H06nk/WA4QZEXF4S91po/glqQnNM/CCuNLRvsOgbqHHnhMut+aUGLMDQc07NGolXdL7nRaPRIDk5GWKxGB7PUGPD/Px8HDt2jDmlNJ99fX1eDrtWq4VKpUJXVxcjs+fm5oLH46Gjo4NxPaiDuNvtZs0py8rK0NDQgJaWlmHPzel0Mslsf0bQpv7+fhgMBmg0Gibf3NvbCx6Ph4iICCQlJZ0DaRIKhdBqtcjMzGROOcGrqOkql4hdVVWFsLAwTJ48Gdu2bWNz0tnZCavVinHjxiEgIAAlJSVoampCV1cXxGIx3G432tvbvfaxs2fPsn8bDIZzghaZTAaNRsMU1i5k1Gy1q6sLFovlvJ8VCARMlUytViM5ORmRkZEwm80sOJTJZLDb7ex3wBD8Sq1WM8nhkJAQBAQEMLnt+vp65uBzEzwymQwSiQR9fX1obm5m5PuYmBgolUrw+XzIZDL09fWhoaEB3d3dbL8wm81wuVzo7+/32hMPHDiAI0eOnLOHAUOE+eLiYng8HjQ2NqK0tJTdB3rPXeiZIKNzE4vFTAhBLpdDq9VCLBbj8OHDP4m4x69By6/2q/1qP6n5405w/+bb/MtgMDBVGZPJ5EWKHi7bRoEKFwNN2UZfR5yboSbHjxp5kfPHrYbQeXKrQ5QB9z0fIrpzGxmS+kxfX58XMdw3APBV9jofV8LXOeWSsGm+uMpXdDwalyR8uUGXb0aTOrzT8bk9XWguPB4PCyC5c0F4dW6fHa7sMPecucZ12mlOuCR3CnC5ZPjvoirlKx7AnTuuWIQ/srfvdVNAyO1lQOIG3KqZ7/Vx1wsFwtx77C/g4AZtwwVdXIEAAF5iCr9Eo4ByOFlwMuKe2O12lJSUsIaNvk4nkZQlEgmrisyePRt6vR7d3d0oLy9HY2Mj+vr60N7ezo5P/AiPxwOVSgW5XA6RSIT29navz3GNSPy0513IuaREitPpRFRUFOOi1NXVoba2FqWlpUzWe2BgAAqFAgqFAmKxmEn/Er+ktbUVx48fh9FohNFoREVFBRtncHAQeXl5CAsLOye5VFhYiNDQUCQlJcFoNKKnp4epjgmFQlRUVHg9NxSI8Hg85ghzg42AgAAEBQWxYIKCkYCAACYvTYECMCRFTc/+hYKWqqoq1NXVwW63IzIyEqNGjWIBj8lk8kpgBQQEsJ48VN3v6elBU1MTQkNDWfBnNpvR29vLpIirq6u9km5KpRJWqxV1dXVsXYSFhWH8+PHQ6XQs0Kirq0NfXx/MZjMGBgZgtVrhcDjOUZDjNuT1Z7RHyOVyxMfHIzY2FgqFgokddHZ2oru7mwVF9D7gynkTZ4v2LVJSMxgM0Ov10Gg0OHbs2C83aOHxeJgzZw4+/vhjBAcHo6SkBO+//z5ee+21H2P4X+0HGFeVhux8MI9LOf6POe6v9uOYv83OX7aY66QlJSXh8ssvx2WXXYaUlBQ0NDTg/fffx1dffcWaqZGj4uvccpWYuLKcBJ/gngPp+nOJuKT8wnXGybgVF+443GNygwPKspOj6XQ6WWad64T7mr/nwPdcfCs7NKc01uDgICOEcueLjk8vLW4lhHu+vtUF3/Oj8+E236MmeeQQEGHX97vDKWj5BrVc5TRugOH7M1VZJBIJFAoFnE4nOjo6vGB9/mBtvqpz5zOC9NExudfFvY++QdtwgTo3uCZHwt9n6NxoTfjyPChY4QboZFRd/KXbhe4dOaFcQjP3HgUGBmLatGk4cuQIent7ER8fj8OHD2PcuHEYPXo0mpub0dnZibq6OvD5fEbw5vYOITI3BaAOh+O8bRqoIunPfGVttVotIiMjIZfLAfzXSSVYWWdnJ1paWnDs2DFWadHr9ax7fWhoKKRSKQvE2tvbsWPHDqxYsQLJycksaJHJZMy5bmpqOue5dzgcOHnyJPr7+5kaVlNTE0pKSs67zgg6p9VqIZVKMTg4CIlEgrCwMIwaNQqjR49GREQEBAIBI/ArlUpoNBr09/ejoKAAhw8fRllZGSwWy3eCh509exZlZWUM9trf34/8/Hw0NzcjMDAQAQEBaGpqQktLC4RCIQIDA9n7gMfjQalUwu12sz4tUVFRaGlpQV1dHdra2tDc3MwCXACor69HYGAg62nT09PDghetVguXy4Xm5mZUVFSgqamJBU7UQBYY2mOCgoKgVCpht9vR09MDqVQKg8GAgYEBVFdXo7Ozk302IiICDocDvb29KCgoQH5+PkQiEbRaLbs/ISEhTKSAzOl0oq+vD/39/ejt7UV3dzdEIhF0Oh37HkGl6f79FHZJgxYej4dp06bhwQcfxNKlS9nvRo0aheuvvx7r169nmuQjNR5BRMjIkfDFm4/UWBfSUf85GY/Hg0wmQ3p6OpYvX45Vq1bBYDCwv23atAmvvfYaDh06dEkgBnw+H2PGjMGMGTMwbdo0zJs3D4GBgWhqasI///lPvPPOO98J2/9DxuU2vSPz5Rf8aiNn54M8+P4tMDAQkydPxpVXXompU6ciNjYWcrkcYWFhsFgsKCkpQU5ODoD/ZuK5Dhs57GTkFHLlMQGwPYKgUhqNBlqtFh6PByaTCU6nk2WnyMixp6y+r/kGBtzrFAqF0Gg0GBgYYMc/n/nbZ3yrPVTxoM+S4+6vQkDz4i/Q4v59OKPzpWeH9kI+n4+goCCkpaVh/vz5WLRoEVJSUiAQCNDa2opvv/0Wn332GU6dOoWenp7zwgR9zbeBpD8HSiAQYMyYMZg5cybGjx+PiRMnQq/Xo6ysDJ999hk+++wzdHR0eFWz6H4MB/MD/su9ovsskUi8AhapVAqj0cjkb6lxnslkQmtrq18ek785p/9z1ZK4FSCukhMXbsY130CFa5eqKfGPbVKp9KKuhZxI36QFqSwRz+XUqVMAgIqKCrjdbshkMpw6dQqnTp3yqppR4kQgEDBIDTl9crkcra2tzKkdzghS5tvrTSgUwmAweME8u7q6wOPxEBoaCqfTiebmZpjNZqSkpCAwMJDBvtrb29HV1QWFQsH8I3KKSeIZGHq2qqqqUFxcjPj4eISFhaG5uRkRERGIjY1FX18furu7odVqUVtby87D7Xajo6MDJSUlaG9vh0ql+s4VzoaGBhgMBoSHh0Mmk0EqlUKtViM8PBwhISEICQmB0+lEV1cXamtrWfUmMDAQBoMBK1euRE1NDU6dOoWysjI4HI7z7qMFBQXsObTb7di5cydaWlpwyy234Oabb4ZEIsHhw4fxySef4Pjx47BYLAwuR/eTx+MxNbLo6GgEBweze2a329Hc3Mz21M7OTlRXVyMuLg5xcXGQyWRobGxEQ0MD6uvrIZVKUV5ejurqalZ1J5iZXC5niTODwYDU1FQkJSVBLBaz3ioWiwV1dXXIzc1FcXExnE4nOxZZQEAApFIpVCoVgoODERISApVKxd5bpDrX3NzM4HFKpRJxcXEQiURs/ru7u6FWqxEUFASNRvOTCWJcsqCFz+dj5syZ2LhxIxQKhVcJmyT+MjIysG3btosei6QLY2Nj8fe//x2XXXYZiyKPHDmCUaNG4f7778cXX3wxIhkmPp+P6dOn44EHHsC6deuwY8eOC5L9RtKoFEqa4Dab7aKvSyKRYOLEiVi9ejUWLVoErVYLu93ONk+JRIKlS5dizpw5eOmll/D000+P6PUGBARg/PjxeOaZZzBv3jzmPJrNZuj1evzzn/8EALz++usjOq5arcbtt9+OF1544Zy/lZSU4M4770Rubu7/DAntl2a+XAhfJ9lgMGDWrFlYtWoVFi1axLLPtN6VSiUr1XObiA1XyaExhEIhI0KT48Pn86HRaDBx4kSsWLECM2bMgEKhYM3VmpubsWHDBvznP//xgm+cL7D17TXCrQTIZDLMmzcP48ePR01NDXbv3o3KykqveRnuWrjH5Bo3CKHPcPkrXMEBwiyHhYVBJBKhtraWvSS5Dsl3eTlRhUImkyEzM5KeN20AAQAASURBVBPXX389lixZAp1Oh4GBAfT29jJc9PXXX4958+bhpZdewptvvgmbzeZF6P+u3B1ugMuVnp06dSoefvhhTJkyBXK5nMmjjh07FsnJyXC73XjrrbdYlY2qILQmhptvrmAErSXKoIeHh2PlypVYuXIlg9CQU1FWVoatW7fi4MGDaGho8OJFUVDte0/9QQu5ASlVzWg9+atSUuWOMOwU0P+v9Fu4GCPJY1/o0Xcx30oel/OUmJiIzMxM9Pb2MpghzSfxT4qLi/2SxylBERsbi/nz52PUqFFobW3FyZMnWaX2Qsk+alTJ7akCgHGpfK2zs5Nl2YGhirDdbmeVRe65kfPp8XgQEhICHo+H8vJyRuSWyWTQ6/Vobm5GcHAw60UTFxeHyMhIWCwW6HQ6Vm2urq72qugNDg6ipaUFfX1950B6hzOaR6lUCqlUCrfbjd7eXvT396OmpgY8Hg89PT1oaGhAc3Mzent7WYIhPT0dCxYswLhx4xAeHo4TJ04gPz8fdXV1MJlMfgP3srIy9rvu7m6YTCZcd911eP7551nV+5prrsGVV16Jw4cP46OPPkJ+fj46OjoQGBjImkA2NzezHic6ne6c5DU9x1qtFuHh4QgODsa0adMQFBSEdevWIS8vDwUFBfB4PIwnJZfLYTQa2foMCgpCcHCwVzCk0+kQHx8PvV4Pt3uow71EIoHL5WK9+/wlMMiPI04T+Y9UBbRYLHA6ndDpdBg1ahQyMjIQERGBsrIybNy4ERUVFZDJZFCr1XA6nThx4sRPxo27JEELVViefPJJFu3TDW5paUFmZiY0Gg2mTZt20UGLUCjEuHHj8Nvf/hYzZsyA0Wj0alQ2depUAMBbb70Fj8eDTz/99KKvTyAQIC0tDQsXLsScOXOwZcsWPProo6itrb3oY1/IxGIxEhMTsWjRIixcuBB8Ph9r1qzB1q1b0dvb+4NfSFFRUbjnnnuwePFimEwmbN++He+//z7bPJcuXYorr7wSSUlJ+O1vf4s1a9awLuIjYYmJiXj44Ycxc+ZMDA4Oorq6Gi+//DLq6+uxcuVKLFy4EH/84x+xdu3ac7JQP9TEYjFuueUWvPDCC3C5XF54WKlUipSUFGRnZ2PdunV47LHH0NDQMCLj/mr/NV/+CNfkcjmmTJnCnFy3e6j5VU5ODnp6erBw4ULEx8cjMjKSOWTnc3i5Xd0HBgYgEokY90GhUGD06NG45pprsHz5coSHhzOnWKfTAQBSUlKQkZGB/v5+fPHFF0zxhbt5kzNLWfzhnFJgqFKh1WqxYsUKCAQCGI1GvPfee6ipqfH7+e9q3O9wkxnUTM9isUAul2PChAmYOXMmwsPDIRQKUVlZiZycHJw6dYp1nfaFFXEdAQoUaQyBQIDk5GT85je/waJFi2C1WrF27Vps27YNNTU1UCgUSElJwfLly5GVlYUbb7wRu3fvRlFREQsIuBwX32vxhUxxoVy058fGxuL222/HggUL4Ha7cfr0aaxduxYVFRWYM2cO7rjjDtx1113Ys2cPU3jiQtoo2UXjcs+D68ByoXJKpRKrV6/Gn/70JwZlCQgIYLCapKQkpKenIzU1FV999RVycnKGTTJx+U6+HBVf6B9lSulzJBZA50yOFDeQ5zYf/TlbQEAAq7S2tbV5JRH8iTNwTa/XAxhSg6I1FBkZiYGBAaSmpmLixIloaGhAYWGhV1BEa8xfwELzTMnMSZMmsYTHwoUL8e677+KLL7447/M8XMWTgoSenp7zVgG58EhyQslUKhXbP10uFxQKBUwmE/Ly8lBWVgZgaL8dPXo0JBIJc6SNRiPrRG+xWNDZ2Qm5XI6xY8eiv7+fCRkQH6Ojo+OcavSFzGQyseoEiYRYLBZYLBZ2HHo/0/W7XC7k5uZicHAQERERmDBhAsRiMWQyGRISEtDd3Y2GhgY0NDR4vdcpKUJwtGnTpuFPf/oTBgYGkJOTgy+//BIAMH36dEyfPh1Tp07FN998gw8++AC1tbXg8Xiw2WxwOp2or69He3s71Go1BgcH/fZJSUpKwlVXXQWRSITExEQYjUbs27cPEonknHtE8DuVSsUqdRqNBsAQ36iyspJxbNLS0qDX69k7KCIi4oLzbDab0dzcDLvdjpaWFsjlcpb45vF4iI2NxcyZM7FgwQLo9XoUFhaiqKgIhYWFjIPV0NAArVaLtLS0X1alJTExEU899RSmTJkCPp+P7u5urFmzBmvWrEFoaCj2798/IuMQ5nT16tVYtmwZXC4XrFYrOjo6GBFMq9VCoVBAq9XiqquuwoYNG76zLOVwNjg4iMbGRiYrN378eEyePBn19fWXFCqmVqtx+eWX49FHH4XZbMarr76KhQsX4o033sD8+fPxyCOPMHLa9zWHw4GKigp8/vnnWL9+PbKzs70e9ry8PLzwwgtYu3YtFi9ejOuuuw4vvfTSiFwXn89HSkoKZsyYwTK+TzzxBDZv3gyHw4E9e/bg/vvvxxNPPIF77rkHzz777IiMq1Qq8dhjj8HlcuHEiRNeAe2sWbOQlZUFnU6HVatWobOzE48++uj3zup9F6NKITW0os3A5XIxqcxfapXHF49P/AuPx4O0tDTceOONWLhwIYChytf27duxa9cuGI1GTJo0CTKZDFqtln2fm/XzB73hOjNE1tXr9ZgzZw6uuuoqTJ06FRqNBl1dXejo6EB7ezvMZjP4fD4MBgPi4+Nx9dVXY/v27bDb7edIH/vjeZDR2FwnuaWlBQEBATAajbjyyitx9uxZtLe3X5BQ6s/OF7ARTMxisSAoKAirVq3C0qVL0draiqqqKiQnJ+OGG26AUqlESUkJqyD4Vm24Xey53BJyBgYGBlBaWoquri4cP34c2dnZaGtrg9vthlwuR1FREY4cOYLHH38cy5cvx+WXX46ioiIGyaNxuEEDt/rA5ehwP0PV54kTJ2LGjBkAgMrKSqxduxafffYZent7GZznoYcewhVXXIEXX3wRbrebvfjpeAQT851T7nriVjhCQ0MxYcIEBqH9+uuvGfE6Pj4eCxYsQFxcHG644Qa0t7ejuLiYZYJ9jRt8cO+rvyoJrXO6flIuk8lkXjBJUiAiYvgvxQIDA1mWu7+/nwlgEIzHarWyigkZEYzFYjGsVqsXzCYxMRF2ux1lZWWsM3hbWxurxMbExKC6unrYd3xiYiIefPBBXHbZZWhubsbBgwdRUVGBhIQEPPzww+DxeFi/fj2DFvmax+NBV1fXOVUVUjYjwjwXzkbPhMFgYPMxZswYtLW1ecHkFQoFXC4XnE4nFAoFhEIhqqurcfbsWS+FLoLElpWVseC7ubkZjY2NqKmpQVNTE6KiovD0009j5syZKCkpwc0334wzZ87g008/RUtLi1//6nyBJMFvae+22+3svkgkEoSGhkKn00Emk7Fu7TR/hYWF2L59O9LS0pCSkgKn04n29nY0NDSgt7eXVRu594ygV2FhYbj11luRmpqKrq4uvP322zh+/Dh4PB6ys7ORmJiIVatW4ZprroFGo8Gzzz6L/Px8r4CMVB+Hs4SEBCxevBgWiwXt7e3o6OiAWq1GSkoKHA4HTCYTiouLWeLBbrezzvVUdSJkQXt7O6vYqVQqrwCPyyHkCoD47lukyEZKiFqtFrGxsRgzZgzmzp2LuXPnoqWlBZ988gleeOEFlJeXszVHiYLx48ezppc/hV2SoOXBBx9EZmYmwzC/9tpreP755xEQEIApU6YAgBch9IeYQqHAtddeiyeeeAJSqRRmsxnZ2dk4ceIECgsLkZubi7a2NsTHxyMkJAT33nsvw2lyM5k/xAYGBnD69Gk0NTVBo9EgLCwMc+bMwbFjxxjJaqRNKpXi7rvvxurVq/Hxxx/jvffeQ09PD7q7u3HVVVexzsA/1Orr6/HUU0+dIxXLNavViuPHj2Pp0qWIjY39wWP5WlhYGJYsWQKtVovGxkZ89dVXOH78ONv8BgYGsH79eqxYsQLLli3DK6+88oOcOl+z2Wz4/PPPcc0112DdunV499132d/efvtt/PnPf8ZDDz3k5RSPlFGJl8/nQ6FQICYmBmq1GnFxccxh6ejoQE5ODvbs2XPBbrc/V/PdVOnFJpfLMWvWLMydOxcikQgnTpzAhx9+iF27dqG5uRlXXXUVc8i4vAguxMhXFcz3Zcnj8Vil47bbbkNycjJMJhMOHTqEw4cPIzc3F9XV1TCbzdBqtRg7dizmzp2LuLg4hISEoLy8nHXF9q0Wnc/p5mbCa2pq0NjYiMjISISHh+Pyyy9Hfn4+iouLv3cChOApVD3iktZJkSgwMBD33XcfFi1ahH379uGdd95BQ0MDli5dihkzZkCtVqO3t5fBXbnn7xsIUmBDxG6Xy4XS0lI0NzczR5DuL61ps9mMiooKFBUV4ZprrmHQPpvNdg7XyJeEz4UXc6+N/h0aGop58+bBYDCgoaEB69atw+bNm1ll1mw245133sGcOXOwbNkyrF27Fm1tbQwmQTAx30DN9+VM10xVjsHBQbS1tcFiseDEiRP45ptvWEVPp9OhtrYWDz30EEJDQxESEuIFaTsfT4krw0znxe29Q1lm4gMQnEen00GtVkMqlTKVvcOHDyM7O5s5PT/3JAhVR1UqFRwOB5NmVavVrKrQ1NQEq9XK4IcymQxBQUEQiUReEt8CgQB2ux2VlZUQiUQoLS2FXC5Hf38/LBYLu0chISFoaGjw65QHBgZi+fLluOyyy9DT04MdO3bgiy++QHd3NzIzMxEVFYU333wTbrcbX3/99bC9SfyR9IlXQATw4OBgtg5cLhfMZjN4vKEO9GlpaZg1axby8vKQk5MDHm9IWESlUsFut0MkEkEul8NkMqGhocGLU+x0OtHZ2YmAgAB0dnay47e3t8NqtaKlpQUejwdjx45FQkICSktLkZubi8svvxzXXXcdI/7X1taeI5ZBgSRJPnPvIyWQ/PGb7XY7g7/JZDIYjUaYzWb2WbfbjezsbPzf//0fkpOTUVNTg5ycHBQUFKC8vNwvKiMoKIglxbKysthxTCYT5HI5BAIB1Go1Ojo68O6772JgYAALFiyAy+XCU089hfLy8nMgtMM9T2q1GjKZDDabDVVVVeju7kZHRwe0Wi14PB5LJPB4PAQHByMuLg6TJ09GSEgI2tvbcfLkSVRVVcFms6Gvrw8DAwNs321paYHT6URNTQ1bNwEBAVAoFF7qlcTzoWogceIMBgOysrKQlZWFmTNnQqFQ4ODBg/jwww+xefNmFsyKxWIEBQWxQImEB34qLvclCVqCgoIgEAjQ3t6O559/Hm+//TaDTIwfPx4ejwfd3d2M4PZ9TSAQYMKECXj44YehVCpx9uxZ7Nu3D+vWrUNpaanXZ8vLy1FeXo7JkycjPDycKW1crHHLxDKZDAsXLkROTg7Wrl170ZUcX+Pz+Vi2bBluu+02vP3223jllVeYk2a1WmG1Wlll6Yea2+2+4Hmnp6dj7ty56OjowNdff/2Dx+IaQe2WLFmCtrY2vPHGG1izZg26urq8PldbW4uNGzfij3/8I2bNmoXt27df9NgDAwOorKyESqXCjBkzsHXrVjQ3N0MulyM4OJhhbIEh0uNIPaQBAQG4/fbbceutt0Imk7HmhG63m2X/yKKjo5GTk/OLDVoA/5AOtVqNhIQE8Pl81NXV4eOPP8bGjRvR0dEBpVKJUaNGQaPRoKqqCuXl5V5ldjqGb5XDtzIgEAgwbtw4rF69GqNGjUJeXh6++uorbN++HdXV1cxBHhgYQGNjI9ra2iCXyxEXF4fRo0ezcf1VWLjVB19FMC7PoK2tDWVlZZg0aRKEQiEmTJjAOC7E0aHj+Au66Hq5v6PzoX8T7EooFGLVqlW44oorsHbtWqxfvx5tbW1wuVzo7OxkmTtfWJjvveIaF4pLlRZyMogjQuR2u90OuVyOxMRELF68GM3Nzdi3bx8GBgb8NoM833rxnYeAgAAm4tHW1oa1a9eyoIRMLpejvb0dmzdvxl//+ldkZGRg+/btLEPvLxPs73q5fBO6hydPnsSqVaswZcoU7N69G83NzYyETRBpQgFQppbLw/ENcsmx4AY2XHluamp3ww034NZbb2XBEN0Pkv0l6VSxWIzq6mr09PT8YoRjiGAcGBgIHo8HjUbDKg4kh05KWYmJiRg7diwiIyPh8Qz1sDh79izrnREaGorKykoUFxeDx+MhKCgICoUCQUFBsFqtjGw+3NpLTEzEwoUL4XK5sHfvXqxZs4Zl0CMjIxnc8rrrrsP69etZkEFZdK4SnK+RjDfxb0g62eMZUt/q7u5Ge3s75HI5YmJikJmZiZ6eHuh0OjQ1NUGv1yM4OBi9vb2w2WwYGBhAXV0dmpqavPZNt9vNjtXU1OTV94qeD+Iqv/322/jmm2/Q19eH1157DSkpKZg+fTpKS0vhcDhYcCyXy6FUKlm1yLffFiVJzmdms5lxaxQKBdrb2+F2D/V/IVhTVFQUrFYrcnJysHXr1mEFniiAcjqdCAoKQlhYGLt2h8MBhUIBq9WKzs5Otmd+8sknkEqlWLJkCfh8Pp577jnU19ezfY0QEf6qZ93d3cjLy0N7eztKSkrQ0NDAAkMAaGtrg9FoREJCAhYsWIDJkycjNjYWwcHBaG1thcvlQmFhoVdCuqOjA4cOHUJBQQGTMeaaUCiEVCplFRfqS0OIjcHBQej1eqxevRq/+93vIBKJUFVVhZ07d+Ldd989BwbP5/MhkUggkUjg8XjQ29uLvr6+nyzxcUmClmPHjiEoKAj79u3D22+/7UU8jImJwcDAAEpKSpCdnf2Dji+RSDB79mzExMSgsLAQTzzxBPbs2TPs5ynLQNyXP/3pT9+ZKDac9ff3Y9++fcjMzIRUKkVYWBjGjBkDlUrl1YBqJGzq1Kl45JFH8PXXX+Ott946xwGi0uNILyKNRoOYmBjIZDKEhITglltuwbx58/Dpp5/i4MGDIzJGYGAgFi1aBJlMhr1792LLli3nPIRkZWVlUKvVmDJlyogELQ6HAzt27MCrr76KBQsWoL6+HkePHsW4ceOQnJyMuXPnQq/Xo6ioCHv37r1oAiufz8fEiRMhEAiwd+9ehl0VCoUsWKFMDFl/f/95JTJ/rubbh4RLkgeGHLyioiKEhIQgOzsbW7duZfMSFhbGMLXFxcWoqKiAx+NhfAQu54Frvj+LxWJkZmYiLS0N+fn5eOutt7Br1y4vWAFBLMix0Gg0iIqKwsKFC7Fz506vlxU38CIj55PMF3JlNptRUFDAGkuGhIQgJSUFcrn8nJ4tvuZbBfAlsZOzQUHTlClTcNddd+GLL77AmjVrYLfbIZPJ0NvbC4fDgba2tnPgpf7gbb5/J4eKYCvkFNO8hISEwGAwQKVSITIyEsuWLcPkyZPxn//8hz1XEokETqeTQaO4FSPfNcLltVAQoVarsXDhQuj1enz99df47LPP0Nra6nVPaI6PHz8OqVSKadOmsaBFKBR6BWD+rteXU0Pjk/wqAGRmZmLp0qVob29HVFQUUlNTGYfz2LFjyM3NZfuIPzljbuBE95MLJyQoYUREBEQiEQoKCvDVV18xDhYFRr29vSyrTXwEcjR8+9D8HM3j8aC9vZ1V6Khfic1mY9Avu90OrVaLiRMn4uqrr0Zqaip0Oh14PB66urpQXFyMwsJCqFQqTJ06Ffn5+XjxxRcZx4OCcLLm5mYv/pZUKmV/S09PR0REBIqKirBhwwbU1dVBIBBAoVAgKSkJ8fHxaG1tRXBwMFNuMhqN6OrqQk9PD9sjfJ1fgrtxx+VKhnN7dgwMDCAqKgoGg4E1vhQIBAgNDWXyuFyHnOBf9LxqNBqWvAGGAg5SqCOFwJSUFPD5fHz22WcMXnvixAmsWbMGs2bNQkREBCZNmoTGxkYMDg4yYQFKyHHXIY/HQ1RU1AWDFr1ej7lz5+Kyyy6DTqdDX18fSkpKsHfvXlgsFlxzzTUAgF27dp03YAHAZIQJikd718DAAPr7+71gdB0dHTCZTKitrWU+yeLFi1FTU4M1a9YwWBYdg6vsRlZYWMgaFLtcLphMJtYnhcfjITMzE1dccQUmT56Mnp4eHDt2DAcOHMD48eORlZWFOXPm4MCBA+zYtB77+/v9JqlJ9Y7WJjUo9UWm6HQ6xMTEIDs7GyUlJTh06BAOHDhwTr8yoVDIgldKaplMJibo8FPYJQla1qxZg+zsbFRUVHhhgxcvXoyZM2fCarXiyJEjP7gyIJPJMGrUKPT19eHQoUM4c+bMOZ8hWT+lUok77rgDq1atAo/Hw7Jly/DCCy9cNIzLYrFg9+7duPbaa5GcnAzg0pTcDQYDHnvsMXR3d+O1117zCrYEAgGCg4PR09ODwsLCER03NDQUt99+O2bMmAGlUonIyEgYDAZ4PB6WKRiJF59Op8OyZcvQ2NiITZs2MQylPyO1jbS0tIsel8zlcqGlpQWhoaG47bbbcO211yI+Ph7AkFTkpk2bsHbtWuTl5V1UhjI8PBzXXnstVqxYgaNHj+KPf/wjDh06xKApNpvtFyNF+l2MnKfhAsGenh58/fXXOHbsGJqbm5noQ0BAANLT05GRkQG73Y4TJ06wv3ErC74kZn//DgwMxJgxY2A2m7FlyxYcPHiQqVwR4dLhcCAyMhJarRYzZszAokWLEBoaitGjRyM0NBTNzc0MSsUNIs4HGeCaw+FgRMewsDAmg0wiAcM50P6ui8bl/o1+joiIwD333IOenh68//77DDbB4w31UTEYDGhpacHp06dZhYn7/eF4FdwAkVtxoXMJCwvDqlWrMG3aNKjVami1WkRFRWFwcBCdnZ0MOuErSe8LoaKg1LebMwU3RqMRs2fPRktLC3bu3Ml6T9C1UMUH+G/VdOzYscwp447nC1nkmm/fHzrXtrY2NDY2suv1eIYUmkQiEVpaWrB9+3Z88sknOHnyJHMy/QXXvgEv10iEZcmSJZgxYwZycnLwt7/9DVu3boXRaITFYoHdbmf7CDma5HQSvI/+/XM2t9vNAgMKSOvq6sDj8Vj1ure3F2q1Gunp6ZgxYwa0Wi2Di8nlci9p4LS0NERERKCmpgZvvvkmC5h9+SfUf4NbhXE6nYiNjYXb7ca+ffuQl5cHp9MJl8uFkJAQZGVlITg4GHl5eVCr1QgLC2OqhZQwkMvl0Ol0cLlc6OrqYveQehxRY1uhUOildkf3mJvwEAqFrJpAiRbqk0TNC6niIBaLWQZeq9Vi7ty5CAgIQH19PZPWJX6I2+3GihUrvJ4lt3tI5fPQoUOYNWsWVCoVYmJiEBISAofDAZFIxJ5bcrxNJhNLxBER3NeoqiiVSjFr1izccMMNmDBhAuRyOXg8HubOnYtFixahqKgIcrkcBw4cwGeffYaqqqrzrhuZTMbgUuXl5YyjY7fb0dHRAZFIhI6ODgYLpOaR+/btg0ajQUhICObPn88gfrQvk3PvK0BQXl6Ouro61tdFpVIhICAAZrMZISEheOSRRzBx4kTs3r0b7733Hvbv3w+r1YorrriCiSKo1WpER0dDIpEwTorb7WZz6mv0LiJxAFpP3L1FIpHg2LFjOHLkCDo6OiCRSLzexxQsU6WShAvMZjNTqPup7JIELb29vTh58qTX7/R6PR599FGoVCo0NTVh9+7dP+jYfD6fEd8bGhrw+eef+61sJCYm4rbbbkNISAgWLlyIwMBAFl2PhPoUlYu5coOXwq644gqkpaXhN7/5DRobG73+plAoMGHCBPT19Z3X2f8hFhkZiZUrVyIpKekcTPfChQtRUlKCNWvWXNQYAoEAMTExUCgUOHHiBE6ePHnelyltWOPGjbuocbnGdcZCQkLQ09PDCMRnzpxBbm4uKisrL7rKsmLFCjz++ONobm5mVSKCcPy/aP7gMIS/ps2/sbGREdXpv5SUFFx//fUICwvD6dOncezYMYZPp0w0F+rDNcpSAkMZqYSEBEyYMAGtra3Yu3cvWltbGUxgcHCoozlBmaZNm4a0tDSEhoYCAGpqapizzQ3gufwV7rX6y6DTvxsbG1FfX4/w8HAvTs4P4bT4q4QEBARgxowZGD9+PJ566im0tbUxZSPCQE+bNg0OhwPl5eV+eRy+QRkZN2vPDXDIoqOjsWLFCowbN47xXOh7WVlZKC0txbZt29DS0sKCTl9Oy3DYcRpHIBAgNjYWoaGh2LlzJwu8yLjVLj6fzyDCoaGhkMlk7H1AGVN/UsTc66MxSU2Mz+cjKiqK9TgYNWoU+vv7UVpaiqNHj+LEiRMoKipi6j/c++TLa/F3ffS74OBgzJw5E1dccQVEIhHq6+vZ3+x2O3OifJud8ng8L2fqlwIPc7lcCAoKglwuh9VqZYRx6uXB4/GYwykSiRAWFob+/n709PSwKpRKpcLEiRNZsBcaGoqoqCjU1NQwCXSz2Qy3283kfmkNSSQStLe3QyaTISkpCd3d3Th9+jTjKbhcLsTExDCZbWAokIyKikJjYyNKS0tZcELNEyMjIwGAvetlMhnrDULQMEooyGQyVt2USqVIS0tDbGwsc4rb29vZc9DT04OOjg50dXUxMj4JvwBDFR5qTDh27FiEhoaCz+ejpaUF7e3tTDgiLi4OwcHBXveBOF0E/ydYHCld0X/EuxKLxdi3bx88Ho9f343ml5LOISEhcLlcqKyshFgsRmBgIEJCQpCWlsZ4gfn5+SgpKblgME5zQ0nCY8eOITIyEm63GxaLhTUa5fIiJRIJHA4HcnJycPz4ccyfPx8RERE4e/YsJBKJXy4cGcH3AbDeOKQKNnHiREyZMgU2mw3PPPMMcnJywOfzMWHCBCxcuJBVhRsbGxEbG4uwsDDWwNJqtcJsNp8TtAwODqKrq4sFxTQ+d1+Ji4vDNddcg/DwcJw9e5ZJV1OgS8JAYrGYrTUKVqj55Pmk4S+1XdLmkmQymQyPPfYY0tLSYLfb8dVXX6GkpOQHHYvP5yMmJgahoaFM0o5rCQkJCAwMxG233YYbbrjBi8PS3NyMZ5555oLlyO9qVqsVlZWVmD59+ogcz5+lpKSgsLAQhw4dOudvWq0WWVlZyMvLYxryI2WdnZ0oKSlBfn4+jh07BpfLhfT0dGRlZSEpKQmPPPIIlEolXnnllR88hkAgQHR0NCwWCw4dOnROUOZrqampXkHGxRhtsE8++aTXJlxfX4+33noLW7ZsuShhA18LCgpiuNCJEycyicX/V4MW4FxJV/od8F8HlT5DmetrrrkG8+bNg9PpxJYtW1BSUsKwu9zPD2f0N5FI5NVAbXBwEGq1mm3eGo0G8fHxmDt3LhYsWMDgOJTR3bx5M+ODcGVmL1QNoXMkx5KyZbW1tZg2bRpcLpdXvxl/NtxY3OCIPufxeCASiZCeno7u7m7s2LEDHo8HKpWKKS6pVCpMmTIFVVVVTNrSn9Pu2wPGX5DE5fIAQ850bm4u8vPzUVNTg76+PsTGxiIrKwujR4/GvffeCz6fj88//xx9fX1e3/eFMfk6JBQsUqfv/v5+ZGdno6GhgWVJ6XsUvJFaHEkSczPV3ODId665P3NhZBqNBmlpafjDH/6AiIgIuN1DPVRKS0vx9ttvY8+ePejs7GQVIlKGokw1V42Njs3NdnL5T4Sxj4mJQUBAAAwGA1JSUlBbW8sy18NV/Ljcqp8K1jHSplQqER8fj/j4eKjVaka+P3LkCE6dOgWz2QyHw4FDhw6hqKgIMpmMkcupd49Op0NgYCDGjh2L6upq1pOL+jKp1Wo2HkEIqZJL3C9q2me1WlkAbLPZoNPpcNlll0Gv1zM1KgAYO3YssrOzvSrrlMXmckwEAgELUgYHByGVSqFUKgEMVTsVCgVbKxqNBjNmzGAZ8bq6OuYT9Pb2oqmpCc3Nzex9Q3sWVTPoXKhBKlWbhUIhdDodqqqqWC+QyMhICIVCSCQSxrUkov6CBQsYRJGqFH19fawdQ1BQEGbMmIH9+/fD4/F4tU3gClwIhULG2SktLUVHR4fXtU6ZMgVZWVmIiYlh3/8uSAWCyFGQunPnTlx33XWM5N/a2gqbzQalUgmBQMBUzYgX2NHRge7ubuj1erhcLkilUvZcXShgonGpakyVOj6fj9jYWNbU8+6778aSJUuQn5+PXbt2wWQysQaQpKrW39/vN/nucrnQ1NTkVbnmqjJKpVKMHTuWccuDgoKg1+sRGhoKl8uF6upqNh4lAWmPdDqdl6S59/e1Sx60KJVK3H777Vi9ejX4fD5OnTqFt9566wc7a7SQ3W43kpOT8cknn3jJVQYGBkIsFiMyMpIRIIGhByI7OxtbtmwZkesCwGQ0b731VjbGpTCr1XrOAyGRSLBgwQIYjUZ89tlnI9I0k2sNDQ14/PHHYbfb0dbWhsHBQWi1WmzYsAGvvfYakpKS8Lvf/Q7l5eXMEfq+xuMNyZR2dnbi6NGjF6xmDFdK/iHj3nLLLfjDH/6AMWPGwGq1orW1FbGxsbBarYywOpL2n//8B2q1GllZWXjggQdw44034quvvsK//vWv/6dgYeczwuwODAwwLDQ5EARduOqqqyCXy3Ho0CFs3rwZZrOZORO0p/irCADesCbCdFssFsTHx+Oxxx5DV1cXIx1qNBoYjUbodDqm9EJO++7du3HgwAH2zPlyD8h8+2r4GjnoNpuNJV+4cwDALySKHHD6NxnXMeVWDAhqwc2YEkZbLBZjypQpiIyMxKZNm5hAwXD9UnzPwx/nA/gv16SsrAxvvfUWenp6YDKZMDg4CIVCgX379uHhhx/G1KlT8dvf/pbBulwul9+59J03cr7p+kQiEXp6epCfn+/lmHGlP2kuqe/OcMpd/oJB7hxTYEFw49WrVyMjIwNmsxllZWUYO3Ys7HY7KioqGKGZgg/fMX1hZty/cwn4JHCwbds26HQ6XH/99bj11luxYMEC5OTk4B//+AccDge7xoCAAK++Otxn4lK9p35sGxgYgF6vR3p6OqKjo5nK47x58/Cvf/0Le/bsgdPphMPh8JvU4/P5SEpKQkpKCpRKJZxOJ2tO6HA4mJwtPYsdHR3M1+Di/cViMXp7e6HVahmvxmAw4Oabb8by5cvh8Xiwc+dOTJgwgTXB9ldhcLlcaG1tZSR2qVQKuVzOAi2ZTMa4brQ2TCYTeDwewsLCMGvWLABD7+68vDyYzWao1Wr09PSw7DwZ8Ti4iQiqdCuVSpbQIBgvWV1dHfMHCCLE4w0puQ0ODjJZ+N7eXqbu2N3djfr6era/jho1ij1TtM/QPNLzL5fLGVSztbUVZWVl6O7uZgmfU6dOwWAwYNGiRYiOjoZCofhOPggFq2T79+9HeXk5QkJCMG3aNHz55ZcICAhg4hXEs3O73TAajZBKpeju7mb7DsGqhiPik4nFYkilUjafJBqwdu1aLFiwAE8++SQmTZoEl8uFyy+/HG63G/v370dpaSmkUimr7FNfLO47gmsej8dr/6PEDPWWIdGJoKAgGI1G/PWvf4XJZIJarYbD4cDBgwexfft2Rv6nij9Va7n76U9lIxK08Hg8REREYObMmViyZAkMBgN6enpw+PBhjB07FvPnz2ebwksvvYSqqiqIxWJWNv0+fIyBgQHs27cPX375JVauXOm3ymG327Fr1y6EhoZi/PjxEAgEOHnyJJ599tkRzWyTPry/F/pImcViYY2EKCshEAiwYsUKPPbYY7DZbAy/PZLmdDrPUWJraWlBR0cH/vKXv2DdunWIjo7G73//e+zZs+eigibC2l7IDAbDDx6Da3w+H88++yxCQkLQ1taG++67D3K5HO+///6IHN+fVVRU4O9//zvee+89zJo1CytXrsQ999wDrVaL++677xeT/fyu5g+GIxaLER8fj+nTp2PChAkICgpCd3c3ent7ERwcjOnTpyM8PBzt7e34z3/+g7KyMkgkEowZMwZisRilpaVobW1lGWp/DiiN6XK5WCbryiuvxMKFC716XvD5fJhMJpw8eRIKhYKJbBw/fhwff/wxc4IuRGoe7jyIJA/811nhZrbo8wRp862i+P7sO680JgVBjY2NiIuLQ1ZWFnJzc9HV1YWYmBhcffXVWLx4MaxWK06dOgU+n89IzXQ8enENt0Z9YVx0fgQrKC0tZc083W43WlpasGfPHigUCgQHByM5ORnLly9HdnY2y2xyoX6+xGTuHNLvyNFoaWnx6uNCjju3OhIbGwseb6hZH13ncNfm777Rd5KTk3H33XcjNTUVZWVleP3116FQKJCcnMx4BdzKiT+ujO/a4cJSKHsrEAggk8kwODjUdPell17CoUOHkJycjOnTp+O2224DADzyyCMAhpANvllRmhN/Y/4czePxoK+vj0GOqEdbSEgIpkyZgquuugrFxcV+Gz4rlUrExcVhypQpWLp0KSZNmoT+/n7k5uZ6NZMEvJuz+vbWISe/q6sLZ86cwbJly6BWq6FWq3HrrbfilltuQUREBNatW4eioiLYbDao1Wq0tLTA7XZDLBYzng1BkiwWCwsuSMaZuEpKpRIBAQFwOBws0VVfXw+hUIjo6GgGXaWGgCEhIdDpdDCbzV7XRPuD3W5ngS4wlBytqqpiEDri9RGHAhiCKI4ePRoikYidAyWUoqOjmVywWq1mEKXW1la0tLTAbDajp6cH1dXVbC3SZ0k8wWw2w2AwYPTo0azZotlsZrLOtDe1t7ejoqKCJT4VCgXjx3HvD60VMofD4VVdbW1txYcffohnn30WK1aswP79+zEwMMD2BgqiwsPDMW3aNGi1WvT19aG9vR1KpRJqtfq8SloikQgqlYr1bRIKhVAoFNDpdCgqKsJzzz2HnTt34qWXXsItt9wCl8sFjUaDvLw8HDx4EHa7HXw+n1UNv6ufQNXnwMBAuN1uBnGkChE11pXL5bBYLDh79iwGBweRlJQEg8GAo0eP4sCBA6itrWVNdyUSCVNWDQkJwf79+3+SvWREgpZJkybhsccew8SJE5lixeDgIGbNmsWwcfRis9vtrDz1hz/8AcXFxfjzn//8nbvJU0nx5ZdfhlarxZQpU5jiA0Wvb7zxBsRiMf75z3+yDOxTTz3FVEF+TvbWW29h0aJF+OKLL/DUU0+ho6MDd9xxB5YtW8bwmZciaBnOBgYGsGvXLjzxxBP417/+xdRYLrVpNBpcd9114PF4Fy06cM899yAkJAQmkwl33nkndu7ciYyMDFRWVo7Q2Z5rhN/t6OhATU0Ntm/fjnvvvRdXX301Pv/8cxw7duySjf2/aFQtoIyfQqHAlClTsGrVKsyePRs6nY4FGAR90el08Hg8qK2thUAgwMyZMzFu3DhMnjwZTU1NWLduHdra2pijzTVuLxX6f2VlJdPhT0lJYS/Pnp4eFBUV4ZtvvkFAQAAefvhhyOVyVFVV4R//+AdOnTrFKhLc6/F3jcNBtrifJ0x4f38/e7nRsbn8HN+ghXs8umZS3+I6yDabDWfPnkVSUhKWL1+OyZMno6qqijUUMxqNKCsrQ1lZ2TmONfc+0b/9VQu410rjy+Vyr+Zr3Iy/0+nE/v37odfr8fLLLyM9PZ0R5rlSwNyqC91T7ouSrpkboHEV33wrDpGRkZgwYQKAIZIsV5nP3zyTcZXp6G8zZ87E6NGj0dDQgOeffx6bN2/GpEmTGLna4XAMG1gOB0WjIMfj8XgFH1zOVHd3N/bt24fDhw/jyy+/xKOPPsoaw+3atWtY9bxfCpcFGJorpVKJ1NRUjB07Fl1dXWhoaEB/fz9iYmIwadIkXH755diwYQP4fD7ri6XVapGYmIiJEydi9OjRkMvlGBgYwLfffotNmzax9cDj8RgJuqur6xyhCGBofilzX1VVBYPBgFtuuQX5+fm49tprERwcjI0bN+Ldd99lXImQkBAcOXIEPB4PqampzG+hQJWbPZdIJEwdlNYdyddSpdjlckGpVDJVr4GBAdTX14PHG+J+ut1u1NTUeFVTyD8jSWyy+vp6vPrqq+xZioqKgkajYc8/JVOio6Px5ptv4m9/+xvj7eh0OhQWFsJgMCA0NBRhYWGsUTQJqahUKigUCsYD5vF4mDNnDnPKW1tbodPpMGvWLEybNg0nT55EXV0dU4lTKpUMZu12u1FaWoqmpiYmFJSWlsbeo8HBwawfDT2PBLEknhDxUbZs2YKMjAzMnj0b8+bNw+effw6n0wmVSoWIiAgEBwdj/PjxmDBhAhQKBWuyqdVqwefzvcQvfI32IOqbJJVKoVKpWPPT+vp6tLa2oqamBl988QVrOtna2or+/n62j3zX4IASf4GBgVAqlfB4hoQPHA4Hent7UVNTg7a2NmRnZ0OtVrN9jWBqWVlZuPLKK6FSqVBfX88EGcLDwxEdHY3ExESkpKQgLi4Ohw8f/nkGLaNHj8ajjz6K+fPng8/n48iRIwgNDUV0dDQCAwO9PisWi/Hxxx8zUhbJbWq12u8ctABDL628vDxce+21iIqKwoQJE5iEYUdHBwYHB/H6668zrsL//d//4dtvv/1ZbtrNzc1YvXo1nnzySaxduxZ8Ph8nT57EQw89hDvuuAMNDQ0jGozp9XrMmzePNTXyZ4QF/eMf/zhi417I7r//flZWfumlly7qWIsWLWJVv+3bt2NgYAAqlQrR0dEjJledlJQEoVDIyLdcs9lsqKurw3PPPYeJEydi7ty5/88FLYB3pjwmJgY33XQTrrzySqZNT2pJ1PWaHNiMjAwEBQUxB0UsFuPw4cOsdwE53lynjUt+pzHtdjuOHz+O1tZW6PV6hIWFQa1Wo7KyEkVFRXA4HHjxxReRlJQEq9WKd999F99++62XxCg38+8PVuTLA+FWl+jzAwMDaG9vR0tLC3Q6HTQaDZPm9HXQfY/ve2x/ECSPx4OioiLo9XqMHj0aSUlJ0Gq1sFgs2Lp1K+bPn88yrP6gaFyOilAoPG+1mj5HGe/q6moUFRWxhmjcylRbWxu2bduGRx99lHF5uPPi+0L0V7GiANRqtUIgEECpVKKurs6v0plKpcINN9yAyZMng8fj4ejRo8PC3nzHoHHI4uLiMGfOHPB4PGzZsgXffvstc4h8e85wHT6q+pDD6xuI0t9JIjUqKopBZIgvQ+vP4XCgvb0dhw8fRlxcHFJSUnDy5En09PScAx/kqp75Uzn6uRmPx4PRaERaWhpkMhnLAre1taG1tRUKhQJ33HEH5s6dC5vNhqioKMTFxTFHjZICBQUFOHz4ML755hvk5ubC4xlS0UpISEB8fDyio6Px3nvvsX4/XBU2AKyy0NXVhby8PMyePRtTpkyBxWLBW2+9hfXr1yM/Px/h4eGsKpOfnw9gCMpeVFQEk8nEKpxc55fg0KRgRo0kqepAAZZEIkFqaiqAIYGQvLw8REZGIioqCs3NzVAqldDr9V6cGaoicK23t5e1oSASOAV2VDXs7OzEqVOnMHfuXNa6or+/H3V1dfjkk09QXFyMt99+G7GxseDz+QgLC8Pg4CCsVitkMhlUKhWTy+XxeIiOjkZPTw+D+RoMBsycORPz589HX18f9u7dC2CoOmYwGGAwGCCTydDV1YWqqirs3r0bmZmZGD9+POMomUwm5vvV1tYySBs9U8DQXkkVmr6+Prz77rtQKpW4+eabUVxczCo8NpuNKR6Gh4ejubkZJ06cYD5sT08P27f8mcfjYU12qXJGldOIiAj09vbCZDKhqqoKGzduhFgshslkgtFoxNKlS9HW1oaGhgaWfCFO1XAtO6RSKUaNGgUeb0j5rr29Hc3Nzejs7PTi4FAfI6fTCb1ej+joaERFRWH27NlISkrCqVOnkJeXB6FQiMTERERFRbHGtRR8/lR20UGL0WhEamoqhEIhKioqcODAAUyZMgURERFemUgyjUYD4L/ZJpvN9oP6UFAGtri4GCUlJV4ZwhtvvBFz5sxhFR7Syh5pGxgYQF5eHr788ksmfRwRETHifVrOnj2Lm266CUFBQWzj0Ov1+Nvf/sbKeiNlEyZMwIMPPohjx47h2WefHVbzXKvVQiwWo76+/gePb7PZsH//fqxevRpJSUl+FdD0ej3uu+8+3HPPPRCLxaiqqvrBTUkB4M9//jPmzJnDCHnk1NDLfKRs9uzZeOihh6BSqdDY2HiOAhmPx8OkSZMQFBSEjRs3jti4Pycjh5605dPS0qBUKpGTk4Pc3FzExsYiISGBVWNMJhNTX0lISAAAtoH39fVdsPcSvViI40G49bKyMpSXl0MsFiMgIAA2mw0BAQG44447sGDBAshkMhQUFDAnl2Ry6ZjkiPuDI/j+zM3kcx3K9vZ2FBYWYvTo0axBHgAvOV663uHGGC7DDgBNTU0MI20wGMDn81FRUQG324358+czR4gbSHIDwPM9G77BIeHW77rrLhQXF+OFF15g5HjuXiESiTBq1CgoFAoUFxezjOJwFR2uceFOAwMDaGlpgVAoREpKCoqKijA4OMg4B6QUd/XVV+PGG29EUFAQiouLceTIEa/zPt8Y3CBEpVJh+vTpyMrKwuDgIIqKitDb2wuJRILo6GjWh4OymFQB4gYxw10b9zMKhQKzZ8/Gfffdh/DwcFRUVKCxsREdHR2w2WwMZmI0GpkjRPeJnisSe+AGRherhPi/YBTIHT16FHFxceDz+Yw073a7oVQqYTQaMWbMGK9qIdmpU6ewd+9eHDlyBKWlpWhoaMDg4CAkEglLuqalpSEpKQmffPIJ+x5lzolHpVarwefzUV1djX//+9+YPXs2QkJCUF9fj82bN7MeTOHh4VCr1UyFyePxoKCggFUdBgYGGISHjAIFLneOKga0tgCwni8AWA8Qo9HI4F0qlYqtP64IgT/z/f3p06eRlpbG+ry4XC6cPHkSJpOJOfN1dXVobGxEa2srenp6IBAIoNVqYTKZYDAYkJCQgLq6OiYnz72HRLLv7u5m2X6qHiUnJyMtLQ0Oh4N1cCfYmdvtRm1tLTZv3gyRSISVK1fCaDRiYGAATU1NMBgMjAcUEBDApIK5SSO73c5ko5ubm/Hvf/8bd9xxB5544gm4XC58++23MBqNmDt3LrKyshjnifrcNDc3w2azQSaTwWAwwGazQSAQQKfTweFwsES8xWJBf38/U2SjKhmfz0dwcDCcTidmz56NzMxM9Pf3Y926dZgxYwauv/56dHV14csvv2SiL1wFOX9+s8PhQGlpKTo7O9HT0+MFKRMKhUxspq+vD3a7HSEhIRgzZgzmzZuHxYsXQ6FQYPPmzXj66afhcDiYsBVxrXp7e9Hb24uOjo6ft3oYvYSJ0EqbQ2FhIb766iusX78e06ZNw5/+9CeEhIQwwpXJZMLBgwcvSq6X+7IGgGXLluHJJ59k8qHPPPMMa/w10uZ2u1FcXIwdO3bguuuuQ0ZGBubMmYPKysoR7WBOZU0KIAQCAaZMmQKpVIrDhw+P2DjAfzMQd911F0QiEf7+978zh4P+vnLlSjz11FNwOBz49NNPLypoGhwcREREBG6//XamXhQZGYmMjAxcc801SE5ORnBwMOtlUlZWdlHqb9Slm2thYWG48cYbf/Ax/dm6detQXl6O+++/HxMnTkRaWhrD51NG3Gw2Y+3atSMqDvFzMV/YDMGKgCEp2lmzZiEqKgoCgQD79u3DN998g8bGRsTHx2PKlClISUlBUFAQXC4XioqK8O233zLlK19Yke949DeCL9H6pSynWq3GokWLcNNNNzEe1d69e1FcXOx3o+Y6Q8ONSUZcFl/IU2dnJwoKCrBy5UoEBQV5OWBk3CrRhXp8AGBKPOS0NjU1oampiWHxnU4npk6dirCwMFRVVZ3DgeBCobj8G64NJ9krkUhY12Wr1YpNmzahqqoKXV1dEAqFSE9Px6RJk3DjjTfCarVi7969XiRPLpzFd51wAz5ycqqqquBwOHDjjTey5mcymQyjR49GZmYmMjMzkZiYyCpFn3zyiVcl2bdqxZ17CkzJSHmOnBNqMieXy6HVaiGVSs85T+41cIMG3/nj/tzd3Y0PP/wQVVVV+P3vf4+pU6ciOjoafD6fqUiR47djxw5kZ2ez9w5VA7jjUpb2lxC0uN1unDlzBr/73e/w0UcfYezYsdDr9RAIBBgzZgwmT57MkpbAkIRwQ0MD6urqkJOTg8OHD3uJNpB5PEMdvxsbGxESEoLAwMBz5osCc71ezxoD2+12tLe3o7KykmXwqbGjyWTCxIkTIZVKvXqw+LZL8N1bqE8GBaKUqae1RWIxEokECoUCTU1NOHjwIBu/tbUVZrMZWq2WOa/fhxdB89HY2IiEhATGbTCZTCguLmaCHXw+nxG7uc+l2z3USyQ9PR0lJSVoaWnx4n94PB7s27fP6x7Y7XYcOXIEkydPxqhRo3DddddBoVDg7NmzsFqtcLvdTAqZ+Lzd3d0skCQoVmxsLCP3y2QyVFZWor+/HzqdjlV9Cb4pEokQEBDA1GUffPBBfPnll8jNzUVvby9SUlKg0WiQk5OD2tpanD59GiaTyStpEhISAqPRiJiYGMTHx6O0tNQLPdTV1YXIyEgEBAQwzhCpROr1esyZMwdZWVmsuWRzczPGjBmDu+++GyaTCRs2bEB/fz/rHzOc2Wy2YeHzLpeLIZGEQiEmTZqEm266CcuWLYNOp4PFYsHLL7+MF154AW63G0lJSRCLxYyPBAxBCM+cOcMC4J/CLjpoqampwenTpxEeHs42RcJRPvXUU9i8eTOTgFyzZg2EQiEmTJiAsrIyBAQEjFhVgsfjYfz48bjnnnsQHR0Nt9uN06dP45VXXrnkk0uE3qSkJJSWlo5owOLPBAIBJk6cCAAjLplLpdSUlBT85je/wapVq3DgwAE0NjbC7Xbj8ssvR2hoKAYHB7F3715s2rTposZrbm7GBx98gIceeghLlizx66DQ5vfCCy/gb3/720WNx3XEKONOZVfAO6N6MWa1WrF//37s378fAJCWlgapVIr+/n7I5XJ0d3f/qFyk/0XjQlVqa2uRn5/PyvDk8J06dQqffPIJvvzyS1ZJCQkJYQ3TSImlra2Nqe5woTB0r7mQGF/lFS6XQKlUMqGE0aNHQyAQYMeOHfj3v//tV2KSW+H9rsaFW9ExzGYzcnJymPY/9TPgPt/cxMFwiQJuJYZw+aQ4wx2Psnbz5s2Dx+NBXl4eG4te/v7G8L1WrpNN/3a7hxr/VVZWIiEhAQ8//DCWLVuGM2fOoKGhgeGux4wZg6CgIHz88cfYvXu3VyDGvW/+mmzSmESq7erqwunTp3HzzTcjNTWVKc5Ro046Zm1tLT777DN8+umnrKP3cPfPNyCkCltvby/q6+tZZjU1NRWhoaGMW+B2u70cDOoZ4k8u1JfHwB1TKBTC7R5qWLhv3z4EBQUhODjYS2a1t7cXzc3NXsemBACX70P3zjdh83M3i8XC5ods9OjReO655zBv3jy2Pr744gv84x//YKp5w5nD4UBjYyOsVisjm/vyFXg8HlQqFdvP6X0vEonQ19eHgYEBBAYGIioqCnq9HjqdDgsXLgSfz8eHH374nd/ZpE4VGBjI+sxQoEDQH4PBgNTUVMhkMrS3t+Ps2bNQqVQAhp4JInITtOz7BCxkBDcMCgpiSeL+/n7Y7XZIpVLo9Xro9XqYTCZGtI+Li2OqbQKBAIsXL0ZTUxPy8/O9iP2+c2EymXDgwAHWRHXhwoWIi4vD3r17UVVVBblcjtjYWMhkMia0EB8fj8HBQdTW1iI3Nxfh4eGQy+WQy+WIiYmBWq3Gzp070draigkTJqC8vJzxebh8O4VCgaqqKjzxxBPweDy4/vrr2X5aW1uLLVu2YOvWrXA4HMjIyEB4eDg6OjpQXl7OIF1TpkzBqFGj2LuIO4fd3d3QarWsZ47ZbIZIJIJGo2F7LgC0trZCLBajqKgIS5YswdVXX42TJ0/i7NmzAL6btDPXSBEuICAASqUSs2fPxh/+8AekpaWxd8F7772HNWvWsPVD8DqTyQS73c7kvP8XEh4XHbSUlpbi73//OyQSCaZOnQoA2LdvH958801kZ2ef8+JzuVw4evToxQ7rZfTyfeKJJ5CRkYGAgACUlpbiN7/5zSUPIID/6qBTJ/Ufwwj7yN2sR8LOnj2Ll156CTKZjKkyLViwgL0ABwcH0dPTgw8++AD/+te/LtrB7+vrw+7du3HdddexEjfBSIi4fvToUWzevBmHDx/+QZsu1/7+979j5syZuOyyy/DUU09h7ty5aG5uxvXXXw+Xy4VDhw6N+PoEhhRdfrX/GmWvSRazqqoKa9euhd1ux+TJk+FwOHDkyBHs3r0bp06d8oJ+tba2DgtbpGMD3hUQgnUA3lKy5NQNDAxAKpVi4sSJuOmmmzBz5kzIZDJkZ2fjH//4h1cPIcrqcWEbNO5wUDA6H99/088ulwulpaU4fPgwZsyY4RXQ+eOo+HO2icROGXZu52waAwDr5UBiBna7/Zzr872W75P4IZGDDz74AG63G3PmzEF8fDzrU0GOj8Viwbp16/DCCy8wZ9K3suI7rr/nXyAQoK+vD1u3bj2nqknOTElJCQoLC3H06FEcOnSIVWuHExXwd02kfkYKllu3bsWKFStYjxapVIpFixahpaUFx44dQ2trq981wL0O37G4948avREUqLu7+xxniGvUA+Z88J+L3T//l4wcKJFIxGA3brcbBQUFeOGFF2AwGDB27FgEBARg5syZrCLV19fH5NSHg6YTSZpkZoGhe6NSqRASEoLQ0FAMDAzA6XSyefd4hgjN1L+Fnt/k5GSIRCK8++67ePfdd78zp4iuR6FQQK1WIywsjMGaPB4PIiMjERoairFjxzKhEFJatNvtrAJoNpvR1NT0vbqY83hDilkUDLlcLtYLh5p39vb2wul0wmAwIDIyEk6nE7t27UJeXh5iYmJgNpvR2NgIi8WC0aNHY+bMmYiNjcX69etRU1Pjd1yPx4P29nbs3bsXAoEAS5cuRXJyMlJSUs75XF5eHk6dOsWgk6+++iry8/NZUtDtdkMmk0Gr1bJ+fXPnzkVQUBBKS0vhdrsZ1JIqk0qlEtXV1Xj66afR3t6OadOmoaWlBV9++SW2b9+O2NhY3HnnnaySV1FRgfXr1+PAgQNwOp2IjY2FTqfz2zKhu7ub9dahtUsBBQnIBAUFQaVSobu7G/v370dISAgSExMxd+5ctLS0sAayCoXC67y5900oFLJqkkajQVRUFOtqP3v2bNx+++2Qy+VobW3F2rVr8c0336CyshJut5sleQi+RrBsUrej5C6pvf0UNiLwsLy8PCxdunQkDvWDbPTo0bj//vuZKkxvby9ef/111NTU/CibdH19PbKzszF69OgfZTyqtJAU4kja4OAgduzYgZaWFixdupRJWAcEBKCoqAglJSUoKCjAxx9/PGLjHT9+HA8//DBuuOEGqFQqtLS0ID8/Hzt27EBZWdmIR/dPP/00409MmTIFMpkMFosF27ZtwzvvvDOiY/1q5zd6eVssFhw8eBBFRUVQq9WwWCyMPOgLcaFyPgW3wzX28ufskwNI1TXK2stkMkyZMgUPPPAAU7Sprq7G+vXrkZeXB7fbDalUyjqa+0KJhjMu7Oh8FRKPx4O2tjbs3LkTGo2GZQK5LyC6Hl9yv+9xuP8mfg5VFElqtbe3F1KpFBEREairq0N/f7+XAhl3vnyrk9xr8Kc0Rs7bkSNH4HA4cObMGUyaNIlVwFtbW1FYWIjs7GwcPXqUwWS45PHhpKK580lZQpKBzc7Oxt/+9jcsWLAABoMBdXV1yMvLw9mzZ9HY2Mj6PADDBydc4yo6URBI119VVYUPP/wQOp0O48ePxzXXXMOw4nv27MGWLVsYeZu7hn2DTV/nlfpQUODCrfxykQy+jgq3yuXvunwrcL+E4IXmwOVysUoEAKau9uKLL+Kvf/0r4uLiMGHCBEyYMAFWqxWlpaXYunUrNm7ciDNnzvg9NqFAuM5+ZmYmoqKiEBYWxuBYNpvNC3ZHsB+n0wmJRMKSfP/+97/R0NCAu+++G2VlZVi/fv0Fr48cR4fDwZSgnE4nc3qDg4NZpYUglsSlok7p1EmehBy+qwUGBiIpKQl6vR51dXUMTguAVW57e3vR3d3N9hZS9youLkZSUhLreUNVyoyMDMb5euKJJ/wmXUhQobGxEdnZ2VAoFJg3bx70er2XuAY1/7Rardi1axe+/fZb7Nq1C4GBgaisrIRYLIZer4fBYEBMTAxb//Hx8Sxp09zcDJfLBZvNxipkVD1tbGzE559/jtzcXNZEPCUlBddffz2mTJnCgoMxY8agp6eHOf0k00/iSNyqNVWVlEolCzbUajVLeLS0tECtVmPixIk4ceIEamtrceLECcyaNQsTJkxAUVERKisrodFoEBwcDD6fj8bGRpSXl3txLIlwHxYWhrCwMPT29qK1tRUGgwGnTp3C4cOHGTKB23fI4/EwWW86Dle4gIJ1SijbbLafZB+55M0lL7UJhUJERUVBKpWy8tk333yDDRs2/KhN+44dO4agoKARJ+H7M5lMhqioKHz00UcjHrSQnTlzBmfOnMHTTz99SY7PNavVig0bNmDDhg2XfCwAOHLkCGJiYnDnnXdCrVZj2bJl2LFjBz7++GPU1dX9KOfwq3krU5HT1dHRgfb2dvaSIVUdIhQPDg56Pdfk6HJ5IgC8nG+ukQPquyHL5XJoNBqYzWacOXMGnZ2dOHz4MHbt2gWbzQapVAqn08mqBFzCtD9ita9zyg1c6KXg6zxSllin06Gzs9PLAeV24vbnzNP3uWNxf88N7OjlSQmCQ4cOwWKx+D0OBXZcZ9/fmMC5Wfze3l4cPnyYEd7JGedK+HLncLjr8meUBeTOkdlsxo4dO7B3714GobLZbOfgwM/H7fBV8+Kej+962r9/PxobG5Geno7x48cjJCQEpaWl2L59OxMD4M6LL8/K35zRPfatmvkbnz4/3N98+/X80ozgk8TVGBwchF6vh1wuR3NzM7Zt24awsDDce++9CAsLY06x0WiE2+32C/cEwAQNSLWLLDU1lalykWxtTU0N+vv7/cp1E9+puLgY+fn5mDlzJjQajVdV83xGwXhvby/LatvtdhYch4eHs34czc3NyM3NhdFohM1mQ29vL4RCIfr6+phylu+x/e2RtD4dDgc6OztZ416RSMQ4NVzIdktLC0tCSCQS1naioqICFRUV6O3tRVRUFON8hYeHQ6FQ4Mknn/QaVyAQQKPRQK/XQ6VSwel04uzZs3A6neju7kZGRgYLSqurq3HkyBHk5+ejqakJRUVFbH76+/tx6tQp1NXVQa/XIyoqCmPHjmX7W1NTE6xWK0tAUfNQX79NIBAwGKjH44FIJGINywsLC9Hc3IyQkBCMHz8ecXFxCA0NRW9vL5RKJQoKChj0W6/XIzg4GCaTiQV2BN+Ty+XQ6XSIjo7G6NGjYTQaYbFYsHz5cmi1WkRHRyMlJQXBwcGIiopCQkICPB4PE5nQ6XRISkqCSqXC6dOnGb/IaDQiOjoaZrMZBQUFUCqVCA8PB4/HQ0lJCbq6ulgVl/ZRaj7pz3g8HmuyTGuL25/wx7affdDicrnw9ddf4+uvv/5Jz6OiogKvvPLKjzKWQqHAwYMH8emnn/4o4/0SzeVy4Y033gAwBBn71X5co6w/8N+SNjn/RBQktRpy4si4zt9wJGZ/QQxVWbhSxeQUdnZ24ptvvsGBAwdYUOJ0OhkUg5wj3woANxih4/meo7/PcysG3MCnsbERBw4cYD0YyGhsbvAwnAoVfdb3fHznLyAgAHv27MHevXthsVjOCUhorrhjXqjqSddF94Lbq4QggXSvgf8GMr4VNe55+1YpuMEuHZeuiYJRm83mtUa49+N8wiG+L2LfoIz7XYfDgaKiIhQXF2Pr1q2QyWSMy0LXxg00ueILNA4XKsSFdpxP8MEfBHG4z/r+zp8U9M/R+Hw+UlNTmXom9YPzeDxQq9WQSCTo7u7GsWPHEB0djSuuuILJ7b788svYtGkTGhsbvaqmVBGQy+V+O5wnJCRALBZDo9EgISEBEonEq1O70+lkKm42m41VEG02GyQSCRobG/HQQw+x+yeVSiEQCIYlV5NaaHNzs1cFR6FQwGw2IzAwkPVt6ejoQGdnJyIjI9HV1YWuri4IBAIvhUwArHltYGAgAgICWJNcmlPi8VitVpw5cwYlJSVQq9VITk5mUCZ6vqjJo91uZ0GNVqtFYGAgOjs7UVFRAZPJxO5Ff38/goKCkJiY6HevkUgkLCj0eDzo6upCf38/du/ejf3798NqtaK7uxutra1+AzFgaF0TfFgkEqGqqgpms5nRBPbs2YP29nZ0d3cz9S9/PC96TigpLBAIIJfLWU+t4uJidHZ2IiIigkGqNBoNlEolysvLWaVLrVbDaDRCLpfD4/EgKCgI4eHh0Ol0SEhIQExMDEJDQ6HX6wEMBTKXXXYZpk+fDr1eD6vViubmZuTn57MeK319fRCJRNBqtYiIiIDD4YDZbEZRURGCgoIwc+ZMHDx4EG1tbYiIiPj/2Hvv8Cav83381rK2LFnee2ODDTbTjLA3hDBDdklDyGyaNGm6Pm2SNqtN2oS2KWkmpIFMIGEHMHtjDDZeeBtvS7Zlbcuy9PvDv+fk1WsZCJD55bkuLsCW3nHe857zjPu+H6hUKtTX18NisUAqlUKr1TJBkq6urn6qdRTM0DMiuXaCvLnd7itGGnwb9qMPWv5ftOrqatx2223f92XcsBt21catNgDwqRiS08t1lrmwJApAAF/ozaUy2PwKCdfZpWDG5XIxbgU52FwjNS6qWnCdaG5mnO7Pnw0E3SGCrdfbp9bDleIG+kPCKLPLJ2Bz740rLgH0r/J0dHRg8+bNDF/O3aj8Zee53d3pfP6CMRoLcgLp3ugeuByBy2H7B3LQ/UH/uMELnZdbUfN3T/zKij/jB6DcLDXNVavVCovF4nPv/GNyn8NA93epaoq/oMffdfMDdq792IMVMqFQiGHDhjEVUhItaGtrg0QiYRlsjUaD8+fPw263Izw8HA0NDTh58iRz1v0F0QONHfWG6uzshM1mQ2JiIjIyMlBYWMikfsPDw1nfEu6zdDqdPiqmAoEAqampiIiIwNmzZxmUkGsajQYKhQKBgYEMFgZ8XWGSSqWw2Ww4e/YsrFYrpFIpC2La2toglUp9nrdIJIJcLodWq+0n8gGgXxUb+LoXUEJCApu3TqeTQY6am5sREBCAQYMGYejQoazhIq3fx48fh9FoRGNjo4/iFt9ovYqIiEBCQgKD1gF9701raytOnTqFpqYmxhW7nNF63tnZycbh8OHDMBqNft8Zuj+xWMzU3xwOB+vJEhAQwBx+CngJkulwOJCcnMyCTOo0HxQUhPDwcLZf0fepYkFqciRQQj1QvF4vTp48idOnT+PgwYM4ePAgu26BQMD+PWTIEKbcVlRUBKvViuPHj6OhoYH1XmloaEBXVxd6e3uhUChY9YySShTIEqqBEgC051qtVh9o7fdtN4KWG3bDbth3bgNVAID+vSS4yir0XS6x3p9dyjnlBkV0Lm7FhOuc0r8H4s1wz+cPKuUvO36p7/IrSPx+Kdzv8K+HxoPui8tF4VaryDGh7Bx3TPjjNVBVi39PfOecWxEhaU/6Hfd8/qpQA40X/YzPq+EaF97Cdcq4fJnLGXeuXI4HwK9qUQDMJcMPlJXkzj3uc+LPmUsFVf4c7IGqNDTe10Md8fs2r9eL5uZmFrQT8Zwa+ZHDaLPZYDabUVpayuak2WxmXC2CGBKZnxoh+gumz549y5zYw4cPQ6fToa2tDS0tLWhvb2f9echBvNz12+12OBwOxhXhW0NDAxQKBZvH3GorqSZSECUWi2GxWFBWVgaHw8GCIG4QQkmM9vZ2n7XhSsxgMKCoqIidp6urC/X19Ux0gARDnE4nOjo64Ha7ERAQwKpWxcXF+OKLL5CXl+e3QkzjUVNTw2B7pHal1+sREBCApKQkWCwWtLa2XvG1U8WG5slA8H2v18uSUsSZpGac3EqSXC6HVCpFZ2cnjEYjDhw4AKAvkaxSqfDxxx+jqqoKMpkMNpsNFosFTU1N7J5oXpnNZkZsp7mi1WoRFRWFiIgIOBwO5Obm+uVc0dw/d+4cTCYTFAqFT1WvqqqKcaEMBgO6u7uZWAT93O12w2azMVRDQEAAC1QA+KAcBqr8fl8m8F7h0/8+y0E37IbdMF/7JhvOD8lurCM37Ib9cOzHuI7cWENu2A374dh3vYbcqLTcsBt2w75z88dV4JND/WWRB4Ju0O8A/xl5/u+upAJyueNfrfH5JvQzf/Clb3JuGsdLffZy3JtvYvyKB5cfwK+k8ZXTKKN3uYz0pZ4395xcGB2/qSb3OHy5av5x/GUUubA/fyR+f8+TjkkwDH8cHfo/cOnKIP8a/FUN/amG+atk/RiDFH8mkUiwfPly2Gw2dHZ2MsiWUNjXHLmnpwdms5kRtD0eTz/uSEBAgA8Uhvhb9EwsFovPPKFmt2Kx2EdGnDLRAQEBDKbW29vLmjz29PSgqanpG1UIuPdJx+MbF8rGrZRy5ypBf1QqFZRKJVQqFcLCwhAZGQmNRsPI9Bs3boRGo8GwYcNYM+f6+nq0trZCoVAgJCQEVVVVuHjxIgAgNTWVyRlTZWIgUSCCHVGn+5iYGBQXF0MoFOKxxx6DzWZj1S2n0wmLxQKTyYSuri7250phjRKJBCEhIZDL5axvDHfsqG8V0McNjoiIgNPpRGNjI3uG4eHhUKvVsNvtaGhoYOp0AFjFJi0tDZmZmTAYDJDL5QD65kt1dTWMRiOkUinjkFBFj4ya0NJ6Q6T/zs5OKJVKhISEwOvtk37mw/XEYjE0Gg1rQCyXy6FQKAAAu3fvhkAggFwuZ9XmS1VIpFIp5HI5VCoVVCoVpFIpg7pRFdBms/lAkOl9EQqFfmWdvwu7EbTcsBt2w75z84fPv5xzeiUk4oGOx3cAhULhgOpR/kiilzqfv89dykH8pj/nn4/7OW7g4M855d7PQE45N3jk8iG4Dpw/h5l/vf4cb+7vSImMODzkzPs7JjeY8nfv/ECLeCzcn3Hvi4Jhf469v2DjUs/oUtA1rvNIUCz+POTO/YF6dgwEEeNfH10PN1DnJgG4NpDz+2M1lUoFoA8aZrVaWbBKz7q7u9unl5K/Zy+TyRAYGMicS+JeEcyMa6RARn96e3thNpvhcDiY4IRcLodEIoFQ2NeZXa/Xw2q1XrFiGNdofviDpopEIgQGBsLr7ZM1p/skfg6XoyGXyxEaGsq4LEFBQYiNjUV8fDykUimqqqqwceNGqFQqDBo0CCKRiPXlUKlUiI+PR3R0NOvwrlAoYLFY2Pp5JdAhgkINGjQIw4cPR0lJCYRCIZ588kkYjUYYjUa0t7ejubkZFy9eRH19PS5evOj3OVzKRCIR45/4e1e54xgZGYmoqCjU19f7rBUUfHLXS/49jhs3DoMHD0ZtbS10Oh16e3sZUd7j6ZPHH6iZp0ajYT3paE2y2+1snfKnsEjm9fZxL2luCwQCNofp/pVKJePyXerZkIBFREQEk/F2Op1ob2+HyWRifYyowaRYLIZer0dgYCBrbPt9JEFuBC037IbdsO/FuP0ogEsTpf1xOPjGzzYOhO0fyHkfKOt9OWece+6BMtv8a+M6meTM0/G4wQNJiw7ktPrj0VwJt2agyg5fqIBP5ueS8S8VAPErAF6vlynCeb1eJqAw0PVxeUdk3P/zv0ecnYCAALbhU5aaX+ng3j/3vP4CJe598ucN9xnxg2IAPs+NP9fJ/Dmj9DO6XxqLgfhF/OsgR4gCQn6156dgbrcbn332GXs/iJtCxs9uD2QBAQHQaDRMhtZiscBoNMJqtfb7PjmM9H6SpLHD4WDXQWqDxE+IiYlBW1ubDwn/So2y5f4cQwpKyUEntTyZTMYc9p6eHpYxdzgcrB9ZQ0MDqqurkZCQgOjoaFaN6u7uxvnz51FdXc2Uyihr39jYiKamJnZNTU1N7FpIdYz/DMjo3YyMjERGRgaSk5PZe0dVGupx09TUhObmZqaGZjabv1GgTfLFJOPLvx4K7hQKBWtCSUELjR/1aeFyk7hzYdSoUUhMTERrayva2tpgMBjQ2dnJBAJiYmJYhcSfjDCdm9TdqIcUVQPpuv09dyLG2+121mBZIpEgIiICwNeJkks1mOWOFVXigL5eaVarFUKhEIMGDUJUVBTkcjkMBgPKy8thMBhYtY7UJm8ELTfsht2w/yfM4/GwbBZ/cb3aSgQ/y81tDAj4700gEolY1v9SwQH/OvjwJi5Egytxy/8edxMlJ5SrbkVjQw6nUCj0qUoQvImy5nxnmQIBrjPOD064ZFOuuhY5H5Rx5n6fjj8Q+d9fAMT/P5FztVotlEolXC7XgJr/3GCHzB9skII+yjxyFeK41yGVSlnWmjKQ/rgRl3r+/s7LfQb0e3Jw6Oc01lxCtb8x41bF+EHZQAEHFwpHz5WgLZSl5gY0A0HZfmzm9XqZ0t/VmlqtRmhoKCIjI5lCl8fT1/zUX/VLIpFAJpOxfiUEz5LL5QgODkZQUBACAgIYhIec4srKSpjN5ity8PjJhEupEJpMJnR3d0Mul0Mmk7E1gmBBDoeDBVH0/CUSCVODslgssFgsTAiAVKy4ZHWXy4Xy8nL2jgG+DrxAIEBISAgjrg90LwqFAhEREdDpdNDpdOwzhYWFaGpqYtUVg8HAYF1ms3nA3iEDGYkBkBIj36ZMmcLmjVgsRnV1NbsviUSCxMRExMfHo7y8nI0NBYFqtRqJiYkYPXo0Czi6urrQ0NDAIGHBwcEIDAxkYxQQEMD2FrKAgAAEBgZCrVajoaEBJpPJZw+4kiCNm4SgRp/0/e7ubvbcL2USiQQSiQQ9PT0wGo0QCASIjIzEhAkTcNNNNyEsLAxisRhmsxl79+7Fnj174HQ6oVQqLymC823bjaDlR2oSiQQZGRlIS0vD/PnzIRaLsW3bNmzdupVN4Otlo0ePxoMPPoicnBwAfS/Mtm3b8J///IdhXG/Y1ZlUKkVWVhYeeeQRDB48GOvXr8eaNWu+taahPyTzx0Hhwnj8VTj4jh0/u0qbM3W853MP+EELV/mGLCAggAUTXq8XcrmclftpcyHuBDeg4F/z5TYNLkRNLBZDq9UiICAAFosFVqvVZ2Oia6drHqhSRE4qN7PPry7RxkbnzczMxLhx45CVlQWRSIQjR45g165daGtrY+ej4/Mz/PQ33Tv3HHQd9DzHjRuHxYsXIzMzEwqFAp2dnTh58iR27NiBwsJCv5ntgcaNzkWwEzKxWMycR5vNxlR7yInhZoP582cg3gn/99yx4M4lbh8ggvbQ//3NZzLuGHGPd7nviEQi5tTJZDJMnToVM2fORGRkJM6dO4fPP/8clZWVrJJHwfZPCSJ2tabRaBAREYGIiAgGrSEsP1ftjmshISFQqVTQarVITU1FVlYWEhIS2FpD6l3nz5/HiRMnUFlZiaqqKnR0dLDu9mKxuJ/UMNe+STDpdrtZ1p2M5otMJmNVEoL5uFwuBAYGsmSR2WxGY2MjtFot+66/gL+np4fxeLhBhE6ng1arhUwmg06nw6RJk+D1enHkyBHU1tYyrgfdu8fjQXt7O6sQeDweHDhwgHVnpwoDwZ8GCjzofecH5MTRoPfSX6XtzjvvhFAoREdHB8rKylBZWcl+x63CBgcHs+qURqOBQCCATqdDeHg4hEIhurq6WGWXKhQU2HF5ICqVCi6Xi/VtAb5eizweD0wmU79g75uaWCz2aZR6JVUW+h7tBXK5HEOHDsWCBQswZswYKJVKNs56vR4JCQlQq9UseONL8H+X9pMNWvR6PR577DH09vbixRdfvO769AKBAHFxcbjnnnswdepUaDQa7N+/H6dPn8b+/fv96q5fDxOJRHjooYcwePBg3HTTTYiOjmaLg8FgwPbt26/r+e6++24899xzCAsLg1QqZT+Pjo5GVlYWnnzySRQXF1/TOZRKJaZNm4aVK1di+PDhKC0txdq1a7Ft27YBuxb/FEyv1+N///sfRo4cyRaJu+66C2VlZdi5c+f3fXnfuvHhBANlv7nNrvzhu/1xIggywjVycLmEdcKf00JPDh7hhUeOHImJEyfC7XZj3759TCKYzsHFj1+Jw8GFg/X09EAqlWL06NGYPn06Bg0aBI/Hg/r6ehQVFbGGYhTAcO/Pn3ErEf4qADSWGo2GdYrOyclBdnY2EhISoNFoYLfb4fF4cOzYMbS1tQ1I8uaOKf8c/GtSKBS444478NRTT7Eu3nQvEyZMQHp6Ov7xj3/gzJkz/WBlXONWzriQJ5fLBaVSifj4eIwaNQo33XQT4uLi0NTUhEOHDuHo0aOorKz0m7UdKDD2VwWhz/vjx9DPhEIha3LoDy4jkUhYcM3lDvm7Hn9VGG7Q7nK5IBAIkJ6ejl/96ldYvHgxZDIZhEIhIiIicPToUZSXl/t85/uCdHwbplAofBx2vl3qXqkzuVQqZeTvzs5O1jfD37scFhYGvV6PrKwszJkzByKRCNXV1dixYwcOHTqE4uJidHZ2sp4q9EelUiE5OZlVXq5nos9fZl6pVGL48OGYMGECVCoVqqurkZeXh8rKSrS3t0OtVkOpVEIgEKCrq4sFUVSlo6CCaySTC4D1E3G73WhoaEBqaiqeeeYZjBkzBk1NTXj55ZdRV1fHqlAkitDQ0MAcfnouhYWFsNvtEAgEkEql/Tgd/p4fQbjoHaP3gPhhQqEQQUFBrOki91nedNNNiI2NhcfjQUFBAerq6rBnzx4Afe8TBTFxcXFITExER0cHKioqWE8VCqj0ej2D5pHgQ3d3NxwOB8xms0/1hv98TCYTWltb0dDQgLKyssv2qrqUES+LgrRLJfv4c1qpVCIsLAyDBw9GTk4OJk6ciMjISBQWFuLQoUOQy+WYPn06IiMjYTabIRQKGTyMu6Z81/aTClqGDBnCnPlJkyYhOTkZx44d65dxvVYTiUSYOHEiPvroI2g0GlaiT09Ph8vlgt1ux4cffoiXXnoJHR0d1+WcDz/8MH7zm99AKBSyjCy/fwU1MLoeplar8e9//5t1EiajTVulUmHatGn4z3/+g0mTJl31eUJDQ/HHP/4Ry5cvR2BgIEQiEUJCQjB8+HDcc889eOutt7B9+/afXOUhIyMDn332GWJiYli2OTU1FQkJCWhsbPy+L+9bN3JAucZdcPm8jUsFBDKZDFqtFgKBAJ2dnf0UV8i4EBlyermOLC3K4eHhSE1NxbRp0zB79mzI5XJs2rTJ53q5zic5q/7gRNxMIDdwoOzl5MmT8fTTT2PChAk+0J3e3l4UFhZi48aN2LZtG8rLy9nmxz23v8y5P6gV0Od0zZw5E4sXL8bIkSMRHBzMIF8E3TKZTDAYDGwM+ffFDc64f/vj73g8HoSEhOC3v/0t7rnnHhaYU9dqoVCImJgY3HzzzbBYLDAYDKivr/fpN8M1mjPc4JPGV6VSYdasWbj77rsxZMgQdv4pU6YgPz8fH374IXbu3MnUjvyRVP0FK9zfUYWHf11qtRoAmIPEbQwaFhaG4OBgdHR0wGAw+A3M/TnX3P9zVaL4558+fTr+8pe/YMiQIcjNzcWJEyeQkZHBnDruc6J//1SClksFLJQV7+7uxsWLF/utNe3t7VCpVHC73azycDkoHlVdR40ahYCAAPztb3/DO++8w+BklIUOCQlh1VqRSASpVMqaOdpstiu+P51OxyBelzLyPyjREhwcjBEjRmDZsmVIS0uDQCDAgQMH8Oabb2Lv3r3o6upiTQSpjw0AppxlMBhgs9mYE8xNFMnlcggEAtYXRK/X449//COmTZvGkjEjR45Ebm4uWltbWSDi9X6thmW325kf0dXVBYFAwGBYTqeTOf/cSkt0dDTGjRuHcePGITExEWq1GgKBACaTCQ0NDSzRc+LECbS3t8Pr9UKpVDKoFFlvby9Lvg4ePBjz5s1DSUkJGhsbWSIpMDAQt9xyC8aOHYvS0lLs2rULJSUlqK+vR0dHB+sb093djfLycubjud1upnZG1+10OvsFJTabjZH2r6bqSf1rqFGoXC5HUFAQcnNz+32W5iD/nddqtcjMzMS8efMwf/58JCQkwG6348svv8R///tf5Ofns6AtNTUVzc3NLFiRSCRszfs+7HsLWsRiMbKzs/HII4+grq4O//znP78RRlUoFEKv1yMnJwc///nPMX78eCiVStZttKamBo8++ig2btx4yXLs1VpAQAACAgKwadMmFBcXY9CgQRgyZAgyMjKgUCjw8MMPo7S0FOvXr7+m80+bNg1Tp07F+PHjER0d3e/39PLX19fj3Llzl1zIr9TUajWef/553HLLLaw7q9frRUdHB+69914MHjwY6enpuOmmmxAfH4/ly5fjk08+uapzUZASFBSEjo4O1NbWMiWWoUOH4s9//jOsVityc3Ova+CZlJSE2bNnIzU1FR9//DG6uroQGRmJ0NBQjBo1CnPmzIFWq8Vtt92GQ4cOXTcMuEQiwQMPPIDHH38cer0eL730Et5++20YjUZMnz4dq1atQlBQEOuo+1M1ckwpw8atjHC5IQCYxCI50mq1mmXsxGIxkpKSMG/ePIjFYuzfvx9nz55lDbWIyMrN0vM5LVFRUcjJycHs2bMxceJEREVFQSwWw2q1oqioCFu3bsXGjRtRXV0NoG/tomPz4VD+MuX8n3PJoLGxsbBardi5cyeKi4vR09ODESNGYNy4cRg9ejRCQkIgEAjw/vvvM6w5OSkDNf3iw7KUSiUyMzOxZMkSLFiwADExMT7OhNvtZse0Wq1oampiY83Ndl4KrsQ9NxHgg4OD8eyzz+LOO+9Ed3c3g9GUlJTg008/RW1tLQYNGoR7770XN998M06ePIm1a9f6OPbcIIm7jlLQRs6gTCbD8OHDMWzYMDQ3NyM/Px9WqxVqtRrZ2dmIjo6GWCzG1q1bfeA/3OD5UvfmD/4XERGB9PR0pKSkQCgUor6+HiaTCTqdDmFhYUhISMDo0aNhMpnw73//G0ajEW632wfeQtdAc4JfsRuo6hQcHIxly5bhd7/7HbRaLf7yl7/g9ddfh1KpxOzZszFy5EgkJyejpqaGBYh8bP1P0YRCIVQqFSIjI5GYmMgIy/z7JrljqtpFRERAoVCgpqZmwMRHfX09FAoFurq68Pbbb+PFF1/s9/6RVK9cLodSqWRrFzWt/CYysSEhIZBIJGhubmZqXv5MLBazRAxBVuvq6nDo0CEEBwcjOjoaM2fOhEgkQktLC06dOgWTyQStVouenh4GXQoICEBcXBzcbjdqa2sZoZ2MJHbb29shlUpxxx13YMWKFRg+fDgL6CUSCcLCwpCUlAS3282qjpQksNlsLOAB+t4BSsASeZ4gbwEBARg6dChmz56NmTNnYvDgwZBKpSzooYqLQqGAx+NBc3MzXn/9dbz77rvo6uryqzy2e/dubN++HUajEampqYiNjcXSpUtx4sQJ1NbWIi4uDnfddRfmzp3LSOpJSUmIi4tDSUkJ60RvMplQWFiI8+fPs2NTlZU7d/hcJoLZNTc3++wPoaGhTGnuUpU48osIlqdWq6FWqxEfH98vaKF9wB9vNCEhAUuWLMEDDzzA5Ki3b9+ODRs24PTp03C73aiqqsKHH36I6OhoxMXFYdCgQeju7kZtbS2DVH4f9q0FLXq9Hr/73e9QXFyMdevW+bxwYrEYgwcPxhNPPIFbb70VlZWVKCgowKZNmy57XJLNu+WWW3DfffchPj4eXm9fB9y8vDx8+OGHOHXqFC5cuPCtZed7e3tx5MgRjBw5Em1tbeylp0367rvvhkKhwP3334+tW7cO2IX1cjZkyBC88cYbPmobRqORlTQPHTqELVu2QCKRoLW1FaWlpdd8zyqVCs899xxWrFjBJCUBYMuWLbj11lvhdruxY8cOiMVizJ07F5s2bcIvf/lLnDx5kkkifhOjTrMAcPDgQTzxxBNMHvLDDz/E4sWL8fbbb2PYsGHXRRdcq9Viw4YNLMgFgPvvvx9utxsymcwHRlFcXMxUPa6HhYWF4fXXX8fSpUthtVrx1FNP4d1332W/v3jxIvbt24elS5fijjvuwMsvv8wc5Z+acTdDLnma22cBAIYNG4YpU6bAZDJh+/btMBgMsFgsLNjJzs7Gk08+iYULFyI3Nxe5ubks48hV3iHnlJw3sViMUaNGYfny5Zg3bx4SEhLYZpSbm4vdu3fj9OnTDFLBvT4KoLjzghxaCgS46lRcTgGRpoG+LBxBSltaWtDS0sI4JitWrMDy5cuRkJCAjIwMBAcHsw2TK4fJr25Q1Yfkf3t7e5GVlYU//elPmDJlCgv2zGYzzp49i9zcXEgkEixduhTZ2dmMWErrCPc58d+DgeSWe3t7oVQq8eijj2LZsmVwu93Q6XQQCAT473//izVr1uD8+fOQSCTQarVoaWnBSy+9hPnz52PHjh3sPmm8aez4PwO+hkgRZMVut+PAgQP4xz/+gYaGBoSHh+Ppp5/GLbfcgmXLlmH//v1svb4cmZTPueLe44gRI7B48WLMmzcPGRkZ7LmazWaWOaYMfl5e3iUrHNxMLL9S52/tSUxMxKpVq/Dkk0/CbDbj//7v/7B+/XpWNTx79iwAICcnB/Hx8Vi7di1qa2sHDKJ/SsaFfTqdTphMJgwdOhRZWVmwWCzYv38/3G43e6fMZjOkUimGDBmC1NRUBhnyt9fU19cjLi4OFosFdXV1CA4Ohsvl6sdJ8Hg8sNvtsNvtrIM6VQ++idE8Cg0NhVQqZTK0fOPu+RKJBBaLBWfPnkVJSQkOHDiARx99FBMmTMD06dNx+vRpFBYWsuCAmxwKCAjw4S4Qt46M+Hs5OTl49tlnMWXKFJ81Mzs7GzKZDE1NTSxpIZFIoNfr4Xa7YbFYmK9CRtwOgtR1dXXBZDJBLBZj0qRJWLlyJXJyciAQCFjViaB9XKPk08MPP4y2tjasX7/e7xx/5513UFpait7eXiQnJ2PlypWYMGECxo0bh9bWViQnJ2P27Nm4cOEC1q1bh/3796Onpwfz5s3DtGnTUFhYiCNHjqCtrc3H3yEeDFcMge4P6INiabVaqFQqpgBGY3rnnXdi4cKFSE9PR11dHTZs2IAdO3ago6MDSqUSaWlpkMvlqKmpYckQl8sFsVjMghYudJ+qaANVrJVKJcaOHYu77rqL8eII0XLmzBm25rtcLtTW1qK2thZOpxOTJ09GQ0MDTpw4wfbL78O+laAlKCgI//d//4f7778fdrsdxcXFKCgoYFHi9OnTcffddyM5ORnV1dV4+eWXsWXLlsseVyAQYPjw4fjss88QHByM6upqrF69Gl988QXOnz8Pu93+jdUmrtZsNhtzKAUCAQIDA3HTTTchKyuLKXns2rXrmiofpO7A3TTXrVuHp59+GkAfxIg284HkBr+JCQQCTJ8+HbNmzYJSqWTnLS0txapVq3wcGLfbjYKCAuzZs4fBuq7mfGq1msn1DYSTvB5Bg1AoxNChQ/HZZ58hISEBwNc43crKShw8eBBWqxVKpRJTpkzBsGHDWLOx62EymQwPPPAAFi1ahNraWrz88ss+AQsAlJWVoaysDI8++igWL16MiIiIn2zQwn2m/CaAxC8JCQnB3XffjYceeghlZWUwGAzIzc1lTuO0adNw8803IysrCzU1Nfjiiy9QVFTkN5vMzaZLJBJMmjQJTz/9NLKzs5Gfn481a9bg1KlTKCsr8xuoUiBA0C5/hHSuY8p1tP05vRR8EIaagg6tVouEhAQkJCSwfghNTU0+zgO/UsRvoMkXF6CqCgU0KpUKb7/9Nv7yl7/AaDRi3LhxWLJkCRwOB6qqqlBbW3tFDpY/vDbd7+TJk7FkyRJW6RGJRPjwww/x17/+lW32Ho8HZrMZn376KcaOHYthw4YhKSmJCQBcyigzSwEiZbQpm9ve3o7W1laYTCZcuHAB3d3diIyMZIEEd93k2qW4SQJBnyTqpEmT8Itf/AJz5sxh/UCMRiOqq6tRWVmJtrY2KBQKjBgxAllZWYz7xD3OQDYQp4UsLCwMK1asYMmdJ554Avv27UN3dzeTlKZAOyoqCosXL8aRI0dQXV3N5oo/Fbgfo2m1WtaXwul0oquri8Gza2trUVdXB5lMhlWrVuGpp56CSqXC66+/jt27dzNuWVJSEoYOHYpZs2ZBq9Wio6NjwOdDiY6cnBxERUXh+eefR1VVFUwmE+Lj4yGXy1FZWYnt27dj165daG5uZhyIq9nDqNJOyAOZTOaTsPC3X9L61NHRAZvNhoqKClitVpb8XLp0KbZu3YoTJ07A6XQyuVugb4+Kj4+HTqdDW1sbKisrfdad7u5upKam4vHHH8eUKVPQ09ODc+fO4amnnoJEIsFrr72G8PBwBvUkGWaSf6ZkCdfIZyPnmeSmhw0bhvvuuw/Tp09Hb28v5HJ5Px+DKzRC7210dDSWLVuGs2fP+uXacjmJlZWV2LJlC+68807Mnz8f0dHRbC2/ePEitmzZgpKSEgQEBGDMmDFYuHAhWlpacP78eZ/1USgUMkigw+GAXC73qQrT3kGkfhJEiImJwdNPP40777yTQfAiIyMRGxuL2267DUKhEIMHD0ZUVBS2bNmCZ555Bh0dHVCpVBAKhQgLC4NWq4VGo2HPkZJk3D4zfKRPcnIyFi5cCJVKha6uLvz1r3/F1q1bUVNTw4JCsVjs04vH4XAgOjqacXtMJtP3toZc96BFKBQiOzsbS5YsYRN2x44d2L59O0QiEWbPng2dTgeXy4Xq6mq8+uqrWLt27RUdm0joWq0WL7zwAt54441+L8F3aUT4SkpKwvDhw3HvvfciKysLTqcTx48fx8aNG79xdoVrFy5cwM0334wXX3wRsbGx6OzsxOuvvw4ArKQtEAhgsVhw6NAh5OfnX9P9REREYOnSpUhLS/P5+euvv+5XkaypqQlvvfUW/vOf/2D+/PlYvXr1Nzqf19snW1lRUYERI0YgJiYGKSkpaGhogFQqZXKBBw8evOZgNDs7G2+99RYSExPhcrlQV1eHP/zhD6ivr0d+fj57OZOSkhAbG4thw4ZdERzmSkwkEmHlypV46qmnYDab8fOf/xyHDx8e8PN5eXlYvHgxUlJScOLEiZ+Eg8E3bqNB7ibMleacMmUK40sNGTIE//rXv3D+/Hl0d3ezCkRPTw+OHj2KtWvX4ssvv2TZQzo2OQvcoEUqlWL27NnIycnBhx9+iFdffRWNjY1Mlpaw6wKBgEGJaPHn44O5cCjAt88JGX8OEfeMMuwymQxRUVFMTnPevHnIzMyEx+PBpk2bsHnzZh84AV9IgK88xXXGKdP//PPP44477kB6ejqOHz+OTz75hHW7zs7ORlxcHFwuF1pbWxkmm+sQ+eO0+JMg9nq9iIyMxMKFC5GQkMAIsiaTCc3NzT5ZVuoabrFYsHPnTkyaNAmDBg3C8ePH+wV6NGe4XB7utTmdTnR2dkIul2PYsGFIS0tDc3MzEhISMGjQIIjFYpw5cwbt7e1QKpU+zhj3WXKN/9wkEgmysrLwq1/9CtOmTYPdbsfu3buxdetWFBYWMgcW6OuY/dRTTyEzM5P1yiCYIhcaeSV8SxpzmUyGBQsW4JFHHkFLSwtWrFiBgwcP+owPzYuuri4cPnwYU6dOZVUuPqTxx2yU8AoPD0dgYKBPzxQi1MvlcsycOROjR49m0t733nsvEhMTYTKZkJSUhBEjRrBxKSgowPbt29HU1AS1Wo2AgABYrVbm9CkUCoSHh8Nms6G1tRUVFRX49NNPYTQaERISgtDQUGRlZeH+++/HE088gVOnTmHr1q04evQoWltbvzE0j2S8bTYbenp6+iXRYmNjERMTA4VC4VPdtVqtaG1tZYFAXV0dNm3ahIcffhgpKSmYOHEi8vLy4Ha7fap8JK/ucrngcDj6BUdUyViwYAG7vrVr16KkpARz586FQqFgfAfqMt/b24uGhgYEBwczmXOCbdH+TwGk0+mE0WiERqPBnDlzGM+P+EE2m41VEZxOJ44ePYrm5mYkJiYiOTkZIpEIOp2ONcO8EoGg48ePQygUQi6XY9q0aQgPD0dAQADkcjnjzURERGDo0KEICQnB6dOnfcZMJBIhOjoaSqUS9fX1PlwyMlqb6VlarVaEhITgoYcewrJly2AymbB//34cOXIE5eXlqK6uhtVqRVJSEh544AHcfPPNqKurQ3Nzs4/8ckJCApKSkvpJEBOPitQpub6TSCTCjBkzMH78eFitVvz973/HJ598wnrVhIWFISwsjCmKkfKZ1+tFUFAQg05zIX7ftV33oCU5ORnPPvsstFot1qxZw3DZ8+bNg06nY9m1rVu34s0338Tp06e/8TksFgs+/fRTNrA0sYnP4k/3/3qbXC7HkCFD8Itf/AKzZs2CXq+Hy+VCSUkJPvnkE3z66aeoqam5pg3C4XCgpKQECxcu7Pc7rVaLX/3qV0x3nTCc12KDBg1Campqv7FTKBR+s089PT24ePEiy3RejdlsNhQUFGDOnDlISUnB7NmzUVNTw6T3qqqq8Oyzz17zvQ0dOhRJSUno6OhAbm4u7r//fr9Sg5S1AYC2trZr5kMJBH0KVD//+c/h8Xjw29/+9pIBCwCcOHECR48e/Unjz/kkYa7z73a7MWbMGDz44IMICwvDxo0b4fV6kZWVhbFjx0KlUkEmk6G9vR2bNm3CunXrUFBQALvd7tNQkD+PyfmjjFxjYyM+/fRTNDc3M76FTCZj1U2z2eyXrM0NGqivib+MPbc8z62yEJRIIBBAo9FgzJgxzLlKT0+HQqFAaWkpPvnkE3zxxReoqanxOxfIWeePKRkFWGazGYcOHUJ5eTnL1lqtVggEAuj1eiQnJ0OlUjFVG+67xpWJ5gfxA3VyHzJkCIYNG+bjjJBjRU67TCZjEBqBQIC6ujpIpVKkpaWx3gEEHaFzch02blDm9faRfM+cOYPx48cjMTERixcvhsViwejRozFp0iQ0NjZi165dDFZB90aVB/589GderxdxcXFsHXnvvffw5ptvor6+vl9vHplMhtDQUCiVSjQ3N8NoNPYTTuBDCrn3wx1/+nvMmDF45JFHEBAQgJdffrlfwELEasq0lpeXo7S0lM1pf7ycH6t5vV4YDAao1WqoVCoolUro9XpIpVLWDG/06NH4/e9/D5PJhBdeeAFTp05FTk4Opk6dCpfLBa1WC4fDgfLycmzduhVfffUVysvLERcXB5VKxRKiVLWsqKjAzp07kZubi5MnT6KmpsZvp/ugoCDccsst+NnPfoY1a9agqKgIL730Eg4dOuQXeTEQXI+y79wAgkwo7GsAuHjxYhaU03vT0NCA0tJSNDc3M16JRCJBZ2cnVCoVsrOzkZiYiIqKChbseDweWCwWnDlzhlVZaH/UaDRQq9UsOGxuboZAIMDHH3+M/fv3Q6PRMNgpBW+BgYGoq6tj99Ha2srWbbVazcaWFAopeHc6nRg+fDimTp0KtVrNuI6dnZ2Mx9Hd3Y3jx4/jiy++QElJCVwuF2JjYzF27FjcfPPNAyI2aNz4sN7jx4/DarXi0KFDWLx4MW6++WYkJSVh4sSJqKurw+jRozFjxgx4vV6cOnXKZy2Ojo5GWFgY6uvr4XQ6ERkZCavV6sOVo2CQusi3tbUhNDQUTU1N+OSTT7Br1y7s3r0bHk9fjxnqsSKTyRhPqa2tDS6XCwEBAUxuGfi6is5fs4nbxO/7k5GRgenTp8Nms+Gzzz7D9u3bmWCCSqVCYGAg4x05HA6oVComJEGVSVJ//L7sugct2dnZUKvV2LNnD55//nmGT3zttdfw8MMPw2g0Ys+ePXj44Yf9ZruuxFQqFe655x6Ul5dDqVRCrVYjNjYWkZGROHv2LPbt24e2tjbY7XZGfryeptFocMstt+CRRx7BiBEj4HK5UF9fj9OnT+Pf//43jh079q1mxwWCvoZOpK9OL/u1mEgkQlpamg9/Buh7qSdOnIh33nlnQGk+UiuRy+XfuLJkNpuxbt06qNVq3HHHHbj77ruRlpaGlJQUREVFoaqq6rqIC5w6dQobN27EsWPHsH79+gHHKyYmBrGxsQD65BivteeNQqHACy+8gMGDByMvLw/r16+/7HdIb/5SMJIfu/lTUaKAQiKRIDQ0FBKJBHl5eXjllVdQVFSEQYMG4aGHHsIdd9yBrq4ufPzxx1izZg3Ky8sBfO2k07vHV9bSaDSskReRQ4cPH47U1FS2jpCiVkNDAwoLC1FRUcGIuVR1oYDkUn0XuNAFvkNCXbUjIiIwY8YM3HXXXRg/fjx6e3vR1NSEL774Ap999hmOHz8Om83WL4vG7c/BNy6nhq6DlLa4UqZ0D6mpqUhPT2dZ5YaGBjgcDh+lLO75BwoEydRqNYYNG4aYmBiWwaZqC0En+BljcjICAgIQGxvLYHEUtPCNe04KPGw2G7766ivo9XrccsstGD9+PAIDA5GYmAidTofdu3ejsLDQb0POS6lGcXktPT09aG1txfr169HQ0IBt27axzuBEeKVeDaTsA/RVpOl++Oe/3D5B1cKgoCAsXboUycnJ+Oqrr7B9+3YfZ4+UsOhauecgovO1VP1/iEYONNBHEhcKhQyCExERAb1ej/b2dhw9ehRvv/02tmzZgpUrV2L06NGIj49Hb28vtm/fjnXr1qGhoQE6nQ7Dhw+HSqVCZ2cn60VB1ZbW1lbk5uZCLBajpaXFJ+FC7ws1Hfzggw/w0UcfYeXKlbj//vvx7rvv4v/+7/+wbds2xnHlriP+1OxonfBnHo8HnZ2diIyMxMSJE1kCQygUYty4cTCbzSwbTgmZ4OBgVjFJSUlBU1MTUw00m81oa2vDtm3bfLrKA2DCBm63G6tXr8bGjRvhdDpx7Ngx5iB3dHSw84SFhbE2DPRMRCIRq4TFxsayoIU7J4nDM3r0aGRkZDDnnK5dKBTCZDLh5MmT2LJlC86fP4/Ozk50dHSgvLwcjY2NiI2NZVwl7nHpfdBqtejs7ITX2yfAQLCzixcvoqamBmq1GjNmzEBkZCTuvPNOKBQKREdHQyKRoKCgwMdvoHfv4sWLaG1tRUhICHQ6Hex2u4/PScIt1NPF4/GgqakJa9euZeeWSCQM5kXQ4fj4eCQmJqKpqQkFBQWwWq2sikbBOjWuJR+J5iI9Fy70WiQSYezYsQgNDUVubi62bduGqqoqiEQipgbW1tYGs9nMrr+7uxtCoRCBgYGsMfDVJqivl133oOX48eP4xz/+gYKCAlgsFmi1WowdOxZpaWm4ePEi/vvf/2Lt2rVXFbAIBH0NflQqFR588EF0dXXBaDQy1TEqn8+ZMwcmkwlNTU3Yu3cv6urqUF5efl2aLopEItxyyy3429/+hpCQEFgsFhw5cgQbNmzAnj17vpOymUwmw89+9jOMGTMGQN+mSOTLq7WgoCCkp6dDo9GwBZCyyCUlJX43V1qggD6HMDw8HDU1Nd/43C0tLfj73/+OyspKPProo5g7dy4AMAm+a22+BADFxcW47777Lvs5yhS5XC5UVVVdc4UnNDQUERERMJlMWL169RVpsg8dOhQZGRkoKCi4pnP/0I3ffI/mWG9vL4qKirB69Wq0trbiwoULCA4ORlpaGrKysmA0GvHRRx9h7dq1TGmFHF+uI8EPighSJBKJEBQUhMGDB+M3v/kNLBYLampq4HA4WIYqOzsbCxYsQFtbG6qrq1FRUYGzZ8/6SFxyoW1c54OMC6mi/3M3z7lz52LVqlUYPnw4DAYDDh48iE2bNuHw4cNobW2Fx+Ppl9XiVgXoGrhZevo9l7fAVROTyWQMKhMWFoY5c+YgKysLHo8HBoOBQTqIH8HvMM8PwgiSRL9TKBSIiYlhqoMWiwUajQYOhwONjY1M0pQbdIhEIoSGhrImnqRk5s+oIkeBIrfydeHCBfz3v/9FfX09Vq5ciXnz5jEZ57y8PBgMBpbBtNvtPmsaBS/++B7cez537hxKSkpgtVphs9nYddJ3yAmMjo5mUsd1dXXM2eFXVPjjyQ+oaM6kpKRg8uTJsFqt2LBhAxobG9l9AGDjStfi9XoxePBgJCYmoqCgYEBlvh+zeb19JHAKbmmOm0wmNDY2sopLe3u7zzwbM2YMenp6sHnzZvz3v//FxYsXERERgdDQUHg8HlRXV6OlpQV2u91HjEYikUAsFjNSNZHLaU6Sf6JQKNDR0YGOjg785z//QUlJCZ588kn84Q9/QHJyMt58800mO9zb2zsgxJ324IGe14ULF1BZWQmvt6+B5OnTp2GxWBAZGcn2MW6TQACM8K7X6xEVFcWk181mM5xOp18OpcfjgVKpRG9vL4xGI4qKitDW1sYy8AAQHx+PkJAQGI1GFBYWorKyEoGBgUhKSkJaWhprWltXVzegHyYSidj6QdBcquK0t7ejqakJBw8exOnTpxn/gvYNWsNbW1tZFQ4Ag5nT+67T6ZiyZGxsLFJTUxlVoby8HBcvXkRhYSHS0tKQkZGBqKgoNDY24ty5c0zEQaVSwWq1ore3F42NjSwQHD58OM6dO9evzUVAQAACAwNZ0ozmIlXphEIhQyHRehAUFMTaduzduxcXLlzwCYQ6OjpQUlLCAh0+zI/WDZfLxX4XERGBzMxM1NXVYevWrcjLy4PZbEZMTAyAvj5//GQuvVuJiYlszoeFhcHpdKKlpeV7qdpe96Dl4sWL+OCDDwD0Paw777wTTz/9NNxuN1555RWsW7fuqh1BoVCI9PR0tLe346uvvsLevXtx4sQJVFRUAAB7SeLi4jBp0iRkZGRg6tSpcDgc2Lx5M5MnvpbKi0QiwYIFCxASEgKgD9504cIFVr4kyMO3ZYMGDcKkSZOwfPlyeDx9HWZ37dqFkydPXtNxaSGmTdjhcEAkEqGyshL/+te//GZ2lUolxo0bB6BP1exqAhayzs5ObNq0icEvNBoNrFYrjEbjd7bByuVyJCYmIiQkhDV+utYKFmV8Dxw4gB07dlwWLigWizFmzBgEBQUxSMlP1biBCtf59Hg8qKqqQl1dHVwuF3Q6HVasWIEHHngAXq8X//znP/HJJ5+gra3Nx/mj79Kx+e85vZfkiFRWVuL48eM4cuQIw0cTvIYaFY4cORJpaWmYNWsWzGYzcnNz8cUXX6C8vNxHztKfk83N0vODjdDQUGRkZCA9PR0ej4cRuZ1OJ9tYzWZzvz4y3MwZnYPv9HK5QcDXjTYJQuJ2u5GQkIDbbrsNM2bMgEqlQmtrK/bt24eSkhLWoI0bWNAmyOdg8K+FC9Wl4wgEAlRUVODcuXM+jTKpMqZQKJCamgqRSASz2TxgYO/P4aZsOPV4oApIYmIiEzNwu91MzQvo31eGnEd/gS6NJ32erypFikrcY2o0GgwePBghISGor69HZWUly1hy732gdY0f0AgEAgwePBjBwcHYvn07jh492q9iQxUtGh+JRIKcnBwEBwezjDoFrz8lI8cM6HPyOzo60NjYyN7N2tpaxMfHY8aMGVi8eDEmT56M06dP4+zZs9i9ezeAvj0VABobG9HY2IjOzk42x4lPAnzNIREKhXC5XHA6nSzBRTAd6kQfEREBjUYDm82GEydO4KGHHsLvfvc7LFmyBFqtFgcOHMCBAwfYfBIK+3pEUcNMIkFzM+f8vcBms+H06dNobm6GXq+H0+nEBx98gJaWFiQnJ2PUqFHIzMxEaGgoEwdqaWnByZMnGecnMDCQQQkHCmiputvb24vAwED2jmi1Wuh0OohEIkydOhUAUFFRgaamJiQmJmLKlCmYMmUKdDodSkpK0NLSgvr6er/KbNx3zGazoa6ujo3zmTNnUFxcjJMnT6KqqgoWiwU9PT2Mz2ez2VjAHhgYiNbWVrS2tkIsFkOj0UCpVDJxl/DwcOj1eiQlJWHkyJFwOp3Iy8tDZ2cnPB4PGhsbcfjwYSY6oNfrYbfbcfDgQbS3t7MeLeXl5ez5REREYOrUqWhtbe0nVyyRSKDT6RASEsLmEnWU5zaZpb44BDVOSkpivJN9+/bBaDT2mwMkoe12u/tVWnp7e/txkzIzMxESEoJjx47h8OHDMBgMkEgkCAwMRHNzM/N1+HuKWq3GmDFjEBwcjHPnzmHUqFHIzs7G22+//b1A2L81yWOpVIqHH34YDz/8MCQSCV544QV8+OGH1+TQu91urF27Fna7HR999FE/KeGuri7k5+cjPz8fe/fuRVxcHDIzMzFlyhSsXLkSY8aMwfr163Hw4MGr7krb09ODDRs2IC4uDsOHD0d4eDieeOIJLF68GCdOnMCuXbuQl5fHHJDraaNHj8bdd9/NtLW7u7tx8OBBrFmzBhKJBCkpKTAYDKxk/k3MaDQyMhZljAQCAXJzc/3eh0DQ18Rq2LBhbLO4VjObzSyLo9FoEBoaivvuuw8GgwH79u371l+Q2NhYjBw5ElKpFJWVldd8TyKRCBkZGZDJZDh48OAVVVkoY1NaWooLFy78JPDnAxk3+8t1priqJ2q1GkuXLsX999+P3t5eRhykTCWAftUAMv4iT1AEkUiEHTt2MEgRQSEoq+R2u1FTU4PGxkZs374dOp0Oo0ePxoIFC3Drrbdi/Pjx+Pjjj7F9+3a0t7eju7u7X5dzfgWE4EN0z+3t7Thz5gxSUlJYYBQWFoZ58+bh/PnzOHToEE6dOoXq6moGLeLDfvw5MvyqCPc6aCNLS0vDgw8+iKVLlyI4OJjhuT/55BO4XC5kZGSgu7ubqZbRtfvrC8OtNAmFQqbiRPhsqVQKl8uFEydOoKqqilWPuNAvChCtViurePnjXvDhDnxRAHKuDAYD8vLycPPNNyMoKAhBQUGYP38+zp8/j+PHj6Orq4uND4l9XEpVi8tV4cuIcgNlsqCgIKSkpECj0aCpqQmNjY1Mbpt/XH/GnzsymQzJyckQi8WseR4A1vOFG7jRu0PN+FpaWlh3chrTn1oixOv1oru7m8HBuT6GRqNhEEwAePHFF3H48GFUVlZCLpcjNDSUwW5IuYpr3P8nJSVBrVaju7ubEZ2VSiVMJhOcTicbe3LuIyIiYLPZUF5eDoPBgH/961+QSCS4++67IZfLsXHjRgB9EGKdTseqjRcvXmToAgr8/ZnH48GJEyewb98+3HrrrRg5ciTOnz+P//73v8jLy8P+/fsxZMgQliXv7e1lleP6+nq2vspkMgBfN+lta2vzkUEmeFdvby+cTidT97NYLLDb7Rg5ciRCQ0NRWVnJ3rW5c+ciOzubtRXo7e1lxPaBniGJDXR2dsJkMkEqlSIvLw9r165FaWlpP6g/P5kSFxeH6OhonD59GmazGYGBgdBqtUz1EegLULVaLVJSUtDb24vc3Fzk5eWxBFhjYyOqq6sRGhoKoM/RDwoKYmMzePBgAGDSzGq1GkOGDIHVasWZM2f63ZdOp4NWq0VXVxesViu0Wi0LXmj9Iglk4rBIpVKMGDEC0dHR+Oqrr5Cbmwuv1+tTYQbA5mJLSwtD+FDQQgR8+jzxBbu7u3HixAkGa6WglQJ/f1DStLQ0ZGZmQi6Xw2QyITs7G2KxGO+9995PJ2iRyWS488478ac//Qnd3d144YUX8L///e+aMbW9vb3461//CuDSjcCAvheqqKgIRUVF2LlzJ06dOoVVq1bhL3/5C7Zv346//e1vjCj2Ta9h+/btsNlsmD17NtLS0pCamsoUK7Kzs3Hw4EG89NJLV3V8fyYWizF16lT8+te/ZqQ7AGxiTp48GRkZGUhLS0NTUxMqKirw+eeff6PzW61WVkoPDg5mPy8sLPTrbAcEBGD27NmYOnUqGhsbfUih12q0EFmtVuTk5OCPf/wjhEIhTp06dV2Co4EsKCiINfC8ePHiNcMJIyMjMWbMGEgkEtTW1l62YiSVSjF+/HikpaVh3bp1aGxsvKbz/5CN65jR5sPNQns8Huh0Otx22214+OGHIRKJ8Nprr7GAhWwgJ53raJLRcTs7O/HZZ5+xbtC0qZHzR8fq6elhqlTV1dU4fvw4xo8fj1/84hd46qmnEBoayrDVdHyqHnGrLPQ3d80yGAz46quv0NbWhokTJ2Lo0KEIDw9HZGQkZs+ejSFDhiAhIQEbNmxAaWkpq5RwA7SBqjt0HsJH08ai0+kwbtw43H333Zg1axYLDCm7N27cOCQkJCA9PR2dnZ3Yt28f9u/fj+bmZlYx4Tu8/EqLxWJhAT81fzMYDDhy5AhsNhvEYjHkcjmDa6hUKkyfPh05OTkwmUwoLS0dEMvPHUP+7ynzKpVKYbFYWP8WSqhkZGTgkUcegVarxalTp3yqwhRQXCpoIceI2/+GCw+ksSGoR2RkJAQCAVpaWhiEiGuXklXmzkGv1wuVSoXw8HAAfRVpbuCkVqths9lY5YWI19RQk6s8x61s/pSM+CYEeyQLDAzE0qVL8cADD0CpVOJPf/oTPv/8czaH6P2ncbtcMMft2E48LeJuNDQ0sL4s9Du3242Ojg44nU5WATh37hzuueceGAwGeDweaDQaxMTEsOx7S0sLC0pJrczj8aChocHvu9fQ0ICNGzciKSkJOTk5WLZsGZxOJ3bu3Imuri7W14qSCCTaY7FYGJSO/Am6bjLiMmi1WsZX4c55agA5bNgwdHR0QKPRICkpyWcONzU1QaVSITQ0FGFhYdDpdOjo6BjQ2SUlxYiICJjNZjQ2NqK8vBwtLS0ICAiAVCpl/B/ueGi1WsyfPx8CgQD5+fkICAiARqPpB68lCHx3dzd2796NAwcOsCCXqjwGg4GpHMbHxzN+ntvtxuDBg6HVamE2m5GXlwehUIiamhq0t7f7rSARtM1utzOSOwU73DEg4Q6bzQatVovhw4fDaDRix44dqKurY73suKZUKuFwONDS0uKTiBEK+/qIcaFhYWFhiImJQUdHB2pqatgzpPWMYNFUiSfTaDSYPXs2UlJS0NjYCJfLxdai74tze92DlsjISNx+++144IEH0NjYiNWrV+PDDz+8biTAq8k8m0wmvPfee6wnzO233462tjY899xzV3UNLpcLubm5OHXqFMNFTpo0CTNmzEBKSko/LOy12vz58/F///d/yMrK8tmsZTIZJk+ejJtuuol1z6Vyqdlsxvr1679RZYv6hEyYMAFAXza4tLS0X9AiFAqRk5ODhx9+GAEBAWhubr5mTo0/++yzzxAQEIAFCxbgL3/5C44fP47f//73Vy3gcDkjmUPaPK6Vz+JwOHDhwgVWor8cljw0NBSTJ0+G3W5HQUHBNSuX/RiMHDMuFIlI6rfffjtWrlwJg8GA119/HV988UW/gMUfqZrLjfFnLpcLTU1N7NwUCBCMizJV/Gtqbm7Gxx9/DIPBgGeffRYrVqxAb28vCyq4BHmug01ZRHIKqOLS1taGQ4cO4fz589Dr9YiPj8fkyZNxyy23YMiQIWhra8OuXbt8svzcjXog4jif6O319vVKmT59Oh588EGMHj0aCoWCkfzlcjkmTpyISZMmQa1WMzjFqFGjEBkZiXXr1jGVn4HGlO7X7Xbj/PnzqKiowKhRo9hG6HK52GccDgd6enogl8sxdepULFu2DHq9HocOHUJZWRk7Hv9++LwM+j91lKeeKWKxmGXC7XY79u7di/DwcOTk5CAyMhJbtmzBW2+9hebmZvaOcY83kFFWlMaA28me4HBisRhhYWEICQmB1WpFXV0dW0e48+lSz4wf8EokEshkMsYh4DtuNK49PT0QiUSIjIzEjBkz0NbWxoJFf9Wpn4p5PB6ffU4gECA+Ph4LFizAokWL0NzcjE2bNmHHjh39oJTfZI3t6uryUSekP4GBgejp6UFtbS2MRiPMZjMTv6DPuN1uREdHIysrC8eOHcPWrVsRFBSEmJgYaLVaVukxm81sfoWHh+P2229nMuU7d+5ETU0NvF4vg2D29vaiuLgYubm5SExMRHR0NO644w6Eh4cjLy8PTU1NMBgMMBgMaG9vZ9UMEjzxer0+vKiWlhYfdTsyoVDI1KWowi0QCDBu3DjMmjULiYmJiIyMhFAoRGdnJ3OoaZ2l80VFRSEsLMynkzzXAgMDkZqaivDwcIjFYiQmJiI+Ph5tbW39YJ1ker0eCxYswKxZs3D48GEUFxdDpVIx0Q/uPiEUChlHxZ+am9vtRnt7O5qbmxlXBuiDFLe3tyM6OhoJCQkwmUywWq2w2+3wePqaFUdHR8NoNPogUyg5QwG13W6HUqn0Oa9UKmWNJz0eDzIyMhAREYHc3FymrjtQwshoNPZLrgqFfb16aM2lMVKr1Whra/Pxa5xOJ5Oep8CWW2UbOXIkpk+fziTjnU4nm9vfFwrkugYtI0eOxD333IPp06cjLy8P77zzDo4fP/6DUC3xePr6fbz++uv44IMPMHv27KsOWgCwLIXJZEJ1dTV6e3sxcuTI63jFfZn/yZMn4/e//32/gAXo2zQjIyMB9GUaqbtvYGAgHn74YTQ2NmLv3r1X3OOkpqYGu3btQkJCAiIjIyGXy/Hwww8jMzMTPT09WLduHcaMGYNx48Zh8eLFSE9Ph9lsxqlTp64KknY5O3z4MJqamjBt2jRkZ2cjNTUVL7/88rcStCiVSgwdOhSxsbEoLS3FqVOnrvk8HR0dDIsaFhY2YJkf6BMAuPfee5GcnIwPP/wQpaWl13TuH7pxnVJuxsbr9SI2Nhb33XcfFi9ejPz8fLzxxhs4e/Ys7Hb7JZ0uCn4GWkz5ilUBAQHM2eVCuPgZaVKYIefvwP/fcf2TTz7B3Llz8c9//rPf5sjnunChOdyN1+FwwOFwoLm5GW1tbdDr9Zg8eTLLeFLfGLp+vlwndzzo/wTlICc2KioKM2fOxH333Yfs7GzmYMtkMtYALioqiqnzNDQ0QK/XIy0tDfPmzUNBQQFyc3PZ8bjVBe690c/Ly8uxadMmhISEICoqCnK5HEuXLoVSqURjYyNsNhskEgkGDRqE5cuXY9iwYbBardi/fz9qa2vZnPCnMMeFBAJfy3vSGJAsqNPphN1uh06nw5EjR9DU1IS//OUvGDp0KCQSCbZt28akrvlVvksFZnw1Hm4w6na7IZfLkZSUhODgYLS0tKC0tJStI/Q9qphxx5B/Hu6/e3p6mEQ1Nf0lUi+pESkUCvT09CAoKAg///nPkZGRgbVr1zJZWxorrnDCj9kEAgGCgoJ8+qgAfet4Tk4OFi1ahLCwMGzevBn79+9nAhPXYtTXiMaQVDvVajVCQ0Oh1WpZlp7k0rnB/KxZszBixAisXr0aNTU1GDx4MFN941Z9VSoVOjo6WJA0ffp0lszYsmUL6uvrIZFIIBKJ0NnZyYJ4ahYZHx+P4cOHo76+HhUVFaitrUVXV1e/AE2r1SIkJAQCgQClpaVMNpsCcao8dHR0MOUoImMDffN/yZIlGDlyJDQaDYN20RgZjUYUFxfDYDAw2ePk5GQkJSWhqKio3zpNggYk1SsSiZCeno5FixbB4XDg/PnzPgnUoKAgpKWlYcSIEVi+fDmsViu2b98Ou90OrVbLAgWuw3/kyBHU19ejubkZKpUK8fHx6OnpQVNTE7q6uphyWmtrK+txk5CQgJCQENbDxmq1oqqqCsDX/Lj4+HiMGTOGqTCePXsWlZWVbH12uVysgtzd3c18MnreFGBFRUVh6tSpsFgs2LNnD+sZ5A+G2tnZyfr/UEWJ1hMKTsmIz8dtTkpJHZPJ5AP9pYat0dHRWLRoEdLT01FWVoaGhgYmDEEVxu/DrlvQotVqsXDhQsyZMwd5eXl46aWXUFxc/INaIHt7e3Hw4EG0tLQwibdrNbVajfnz5+MXv/gF0tPT4XQ6cfjw4WtWEUtLS8Nzzz2HkSNHMjUNf2axWJCfn4+zZ8/i1KlTSExMREZGhk9G8Eqtvb0d7777LqxWKx599FEkJiZi0aJFmDp1Knp7ezF9+nQMGzaMSft5PB6UlJTg9ddfv278ndbWVrS3tyMlJQUulwt5eXk4ePAgFi9eDLVa/a0RSXU6HQYPHgyFQoGWlhaGQ78WEwqF0Ov1Pl2H/X0mMzMTf/zjH5GcnIzNmzdj69atPhWFn6LRYk4wLnIWExIS8NBDD2HOnDk4dOgQ3njjDRQVFfk4AFynlmt8CVJ/85+rEsU1/ne4wQBt9h6Ph23qpBJDTiTQvwrMrazQJkWbHBFjXS4XJBIJ4uLiMHr0aMycORMxMTEwGAw4e/YsmpqafPgT/Cw8/3zA1zA4uVyOUaNGYdGiRZg7dy7i4uJYlYOw5nQdbW1tKC4uxrlz59DQ0MCgkgQBoSoSdX3mnpOeBUGobDYbPv74Y0ilUqxYsYJB3saNG8ccAoFAgOTkZAbHPHz4MHJzc316EAz07LgVF6qscSFTbrebNbJMTk6GUChEXl4ezp49i4yMDKYSRWM1ULadX+HiQ/z4AaTX62W4d7Vajfz8fNTU1PSDu3Grgf6EDei49LfFYkFtbS0cDgeCgoKYMxYQEMDmj8fjwaBBg/DUU08hKysLn3zyCXbs2DGg9OtPwQICAqDT6eBwONDV1YWIiAjcddddWLJkCQwGAzZs2IBDhw4x/sCVcAr5plAomNR5U1MTtFot5HI5e3d7enrQ1tbGIFDx8fGIj49HZ2cnWlpa0NbWBplMhjlz5uDee+9FTU0NTp48iaioKKSmpjIIGb3b3G7qnZ2d+OSTT6DRaHDXXXdhypQpyMzM9Hn/SPI3IiKCJTCJG1ZYWIji4mK0t7f79cO8Xi9TGistLYVcLkdwcHA/tU6Xy4WOjg44HA50d3ez+Tp79mwsX74cGo0GDQ0NEIvFjAtSV1eHw4cPY//+/YwAHxUVhfT0dCQnJ+OVV17ptz57vX2qX4cPH0ZwcDCr0k6ePBlKpRIHDx5EXV0d5HI5UlNTkZiYiODgYCQnJ0MqleKDDz5g0DCCl/f29vpI9B4+fBherxfh4eHIyMhAQkICLBYLTpw4gTNnzrDn2dLSgo6ODuzevRuBgYEQi8WsUiuRSJgqH/C1qm1mZiZGjhwJoVCI9evXo7Kysl/PQJo3ZBSo0Ro2atQoTJw4Edu2bUNhYSGrGhOcj/u9pqYmdiyuQqU/38/tdrO1n66HIGqUVFGr1ZDL5ZDJZFAoFJg7dy7mzp2L9vZ2lJWVMQgunf9HDQ8bOnQonnjiCaSnp+OLL77A+++/jwsXLvygAhbg60Z/arUa+/fvv+bjhYeH47HHHsM999zD5I/Xr1+PN99885qCloSEBPzf//0f6zLLdVSoo+zRo0exceNG2Gw2lJWVoaWlBVarlfX4AMBUNq7UiKi3du1aZGRksGqLXC6H1+vFokWLGNmzt7cXlZWV+MMf/nBduRdFRUU4ceIEMjIycPfddzOyP78T+fU2Og/Qpyp1PSpHbrcbu3fvxty5czFo0KB+AVdMTAxWrlyJJUuWQCqV4h//+Ac+/fTTb5W380MxvjoVOV2/+tWvMH78eGzbtg3vv/++T0bL33cHsoGcenIK/QU8fM6EVCqFx+Nh2TqZTMYy7VlZWTCZTAxiwF/E6d+kMMMnV5NSjFAoRGxsLG6//XYsWbKENS7ctGkTPv30UybhSdfId+j5gRr3XsLDw7F48WIsX74cISEhDCZGQUZZWRnOnTuHgoIClJaWoqamhjVUFYvFrA8A11HhO9hU3eEKEAiFQhgMBnz++edIS0vD4sWLodFooNFoEBkZ6QOp6u3txZ49e/Duu+8yJ4BfPeKOKZ/4z92gacxJJamiogI5OTmYMWMG3G434uLiIBQKmQMGwGfTJ/jXpYwbpHGhgPRvnU6HhIQEBkMxGo2sauYPduPP+L93Op0oLCxES0sLcnJyEBYW5iP1HRMTg6VLl+K+++6Dx+PB3//+d2zcuJEJIpD9lPgsXm9fN3WdTgegj2D99NNP45ZbbkFBQQHWr1+P4uJiBAUFwW63o6Gh4Rvfv0qlQlRUFJOsJoWnoKAgn4w5wdNsNhtsNhvi4uIwYsQIhIWFQavVIioqCklJSejq6sK2bdsgFAoRFxfH+Cp2ux1SqRQ6nQ5yuZy9Hx6PB+Xl5XjppZdw8OBBBgUfPnw4e99iY2MZNFUsFqOzsxN79+7FunXrcPToUXR1dQ043yigpUBDo9EgKyvLp8cLVaSlUik6OzvR1NQEr9eL7Oxs/OMf/0BgYCC+/PJLeL1eTJkyBSKRCHV1dThy5Ah2796N2tpajB49mnFBwsLC+u2D3OqpwWDAp59+CofDgbFjxyI2NhZ6vR4zZ85ETk4O7HY7g0parVb09PTA4XDgiy++wKZNm2A2m6FSqRing9uLBejb5zUaDQugYmNj0dLSwjgjXm+fTHt1dTVcLhcqKiogkUiwcOFCLF++HAqFAiaTCffeey9OnjyJ8vJy5Ofnw2KxQKFQICUlBaGhoSgtLcW777572cQnXT/QVznKysqCTqdDfX09477QHsQ1Lnke+JpvRZVZ/ly32WxM0IXWSafTiaCgIBYsU6N2hUKB6dOn495774VCocDJkyfhcDigVCpZo1ou9Oy7tmsOWhITE/H4449j+fLlePXVV/Gf//yH4Q9/aJaYmIhnnnkG9fX1eOWVV77x92UyGQIDAyGTybBo0SKMHTsWs2fPhlKpRE1NDf72t79h48aNMJlM13T/BAPg9isoLS1FYWEhDh8+jM2bN6O7u5tF8VzIAkmlXosRmdbhcPgQwKRSKbxeL+rr67F//3789re/ZaTB62VdXV04efIk5s2bhylTpmDs2LHM0dq2bdt1Px+ZWCxmKipcAtu12s6dOzF69GisWLEC48ePx65du5iyTGpqKhQKBTZt2oS33noL5eXl36pc9g/JuM62x+NBQkICfve732HmzJlYt24d6zTOD1C4AfxAVQfu7/jmrypCECHuQk8kdXLCCf/u9XqRlpaGBx54AE6nE1u2bGGOO/f7VDkix56yfwSxCgoKQlRUFIYMGYJZs2ZhxowZ0Ov1qKqqwhtvvIFNmzahubnZBxtP1+2P78Hn0AB9UJnw8HCo1WoWSLS0tKCkpITJPNfU1LCghB/QEVaaW9Xy917QePPHsLa2ls3/QYMGQSAQsADP7XajtLQUubm52LRpE86cOcPI89xqA9f42HT+c+T+u6OjA+fOncPcuXMxf/58TJo0CRqNBiKRCPn5+WyNpEoHOTX86gn9jDvWXIgVzQ1SLwsMDIROp4PX2yePzOezcCsr/qBa/vgtXq8X+fn52L9/P+644w589NFHKCoqQkBAAOLj4zFo0CAIhUJs374db775JgoLC9nc9Wc/lYqLy+WCwWBAeHg4fvnLX2Lx4sXYt28f1qxZg7a2NqjVavT09KClpeWKIdJk1ByUpIyBvvngdDqZAwd8XU2l991gMMBoNOL06dMQCPr6Fun1egQFBcHj8TBHVCqVsky31+tlvoVUKvXZA+iYO3bswL59+xAWFoZhw4Yx3kdQUBD0ej0CAwNht9tx5swZbNu2Dfn5+ZfdS3p7e5nUL9C3v4eHh7OkJ70TLpeL9QMhcZ877rgD8fHx+OCDD/Dyyy9jzJgxjBtRW1uLI0eO4PTp0xgyZAiDttfV1TH4J3f+8SXMq6ursWHDBjQ1NWHOnDkYPHgw9Ho9goOD2RpqsVhgtVpRX1+PAwcOYOfOnTAajcyhpnVKpVIhOjoaJpMJdrsdM2fORHt7O4Nrmc1mNDc3s54pQB8P5eLFi/B6+0QwDAYDhEIhZDIZOjo6UF1djbCwMCxduhSFhYW4cOECmpubmaIlBVWXMhoDr9fLeuSQCltVVRWqqqqYb0fBMnducvmTIpEIOp2OiUr4m+vEvwkPD2fVku7ubhgMBlY9pKBp/vz5yMnJgVgsZpU6tVrNkC4Oh4P1vPo+7JqCFqVSiZkzZ2LKlClYt24d3nvvPZ+uyz8ky8rKwosvvoj09HRs2LABRUVF3/gYN998M958802WfSDy5rFjx/DCCy/gwIED14U8XVtbi1OnTuGmm27Cjh078OWXX6KkpIT1ULheDvWl7J///Cfa2tpw2223QaVSobKyEkajEVu3bkVdXR3a2tq+Fa4SvcSdnZ2Ij4+HRqMB0JcVevXVV7+Vc4rFYiQlJWHw4MEMP369xtjpdOLVV1+F1+vFY489hgceeIAtWJ9//jn27t2Lzz77jMm8/r9iXKiTUqnEvHnzMGHCBGzevBlvvfUW2zT4jhcXDsRv5sj/HD/gGQhaxv05/Zsw41xssFwux4QJE/D73/8eOTk52LRpE1OQEYvFPu8+F/7D3UgIRrtq1SpkZGQwHk1LSwu++uorfPTRRzh8+DCsVivbXPw5sfRz7j1SJYbGrKOjA8ePH4dSqYTFYmEyytQ0lRwmytLSMUlJhjJ8l0vA8DlJXPjWsWPHsHr1akyaNAkSiQStra3o6OhAU1MTioqKUFhYyHgZ/O8PdB7utXKfGfc52u125Ofn4/jx4z59tfbu3YstW7agpaUFAoEAUqnUJ4gg4wcqA0EAqRIF9Dl90dHRkMvlcDqdMJlMbP5wgy2as/w5428sKbBpbm7GG2+8AYfDgQULFuCee+6BUChEW1sbduzYgc8//xwnTpxgsFI+94gbVHMFBH7sJpVKcdNNNyE8PBzPP/88NmzYAJfLhYiICOaU8iE6ZNTIlMsxIKOAyGAw9PsdNQi02+0DwqK5iYWGhgakpKQgJiYGISEhLEi22Wysrxv18wgMDPSbmKN9n6BflMygZ6tWqxEREQG73Y7GxsYrSn4RHIsCa6fTifr6epjNZuh0OsTHx8NkMqG9vZ31dqqtrUVbWxskEgny8/PxwgsvoLa2FhcvXsS+ffsYBLWmpgZKpRKDBw+G2WzG+fPnmaN7ORK3x+NBa2srNm/ejNraWkybNg0ZGRkICwtjlaiWlhYUFRXh6NGjOHnyJLuH3t5extPTaDRMKKm6uhodHR2Ii4tDZ2cnC3Cam5tRXFyM2tpan2ugdyUmJgYymQwff/wxGhoa0NXVBaVSidjYWOTk5LBn2tLSgubmZr9rF73DAyVYKEkjEAigUqmQn5+Pc+fOAfhapp8+r9VqERoaygIUguRptVrU1dX5rcoAXyvlTZ06FY899hjeeOMNtLW1ITk5GdnZ2Rg7dixmzJiBuLg4pqZZU1OD8vJyJsIgl8vhcDjQ2tqKkpKS760wcU1BS3h4OOLj4/Haa6/h448/Zvrd37fFxMRgwoQJUKlUUKlUuPXWW5GRkQG5XA6z2Yw1a9Zc1YCPHTuWdYynjGhubi4qKyv7NX27FvN4PPjb3/6Gv//97/0UYr4r6+7uxv/+9z9s2LABgC+R+Nu+jt27d6Oqqgpr1qzBTTfdhK1bt+Jf//oXjh079q2cW6lUIiUlhckTUm+M62Xt7e344x//iGeffdbHKSFi9Q+xKvldGDmFOp0OQqEQr7/+OrZu3coCFsC3qzw3A0/EZ/6xyPjvCx++RE6dP0gQfS8kJARJSUlITU1FcnIy0tLSkJWVhYiICBw9ehRr165lHAx+5Yd7Lu7x9Xo9kpOTERoaiu7ubuTl5WHv3r04ffo0Lly4wJqpUvWHuwHRPXA3QX+KMnT9jY2NWL9+PXbt2gWTycSOTZUfrhABHy8NfO24c4M87rjz+6bwzevtk2Rdt24dvvrqKwQEBKC9vZ1BVrjXSrwMrqPNPy5VV7i9SS6laEYZW7vdjuTkZFRUVGDXrl04duxYP97IQOe81M/5nwkODsbQoUMRGhoKo9HIeHH8ecgdH+790PVwP0OBaG9vL8rLy/GHP/wBL730ElP7oYwxvwpEz5l+xrWfSnJEJBIhJiYGDocDzz//PAoKClh23GQyQSQSwW63+808R0REMD6Xw+GAxWLxSYhdihNK0LRLGTVMJQEZrVaL7u5udHR0wGw2M8eeuArEJ6AGkwMZVfX41+Z0OpkQhFQqZf/nGrdVApnZbGZVR4KskhBPSEgI8vLy0NDQAIfDgeDgYEyePBnNzc3Iz8/H66+/zpx9l8vFxEQoMB8xYgRCQkJQXl6OhoYGJo7BVbO7lLlcLpw6dQrFxcUYPHgwhgwZgpCQEDidTlRXV6OkpISpqZGRNHxgYCDi4+NZo2p6pwwGAxwOBzweD0uOlpSU+L2esLAwhIeHY9euXayHHe1XmZmZbJwSExPR1NTkk3whI2ROZ2cnezf9JbZ7enpw5swZlJaWorq6ms0vWu8VCgUiIiKQkJAAl8sFmUwGjUbDGplyE1z+5q3dbkdFRQVcLhceeOABzJkzB11dXUhKSkJ0dDSD2wJ9c6m4uBhFRUWwWCyMt0k9ei5cuIDCwsIfZ9BSVVWF3/72t9frWq6b1dfX46OPPmL/f/31131+f7WL9q9+9Ss8+eST13ycK7GrIdJfb/u+HGqPx4OKigrMmDGD/ezbHGsqi3u9XlRXV6OysvK6n+OH8Dx/SEaOmkgkgsViwYYNG2AymfrBWigzB/g6mFxCITnWXMI0fZd/PprTAxEJyUF1u92sMeCBAwd8Ng8qkdPmQxsFf45y4WBkVVVVePXVV/H2228zeAZlCYlHQ0pAXP4GP8giBTC+8a+BssV0HK4AwKWCDa5jM5Cqmj+IFl92GuhLgPjrGUUVHvq8P0UtfrWDnp/H4/GBV3E/QwGjxWJBbm4uysrKoFKpmMgH9zzcYIGfAeYfk3tN9DyICE/dpyUSCdRqNS5cuMDI89z7pevlPj9+4MUNgPkJK4FAwLKs/OMO9Dz5c/P7aAh3vU0oFCIpKQlKpRLnz59n6mDENaH+I/4qhdHR0VAqlUzyl/ssr2aNJkgQV0rWbrfj3LlzOHfuHDQaDdLS0pCSkoLg4GB4PB7WL8PpdCIqKgpxcXFQqVQQCoWXFG7xZ4T8IH6Mv+eblJSE5ORkqFQqWK1WVFdXo7q6ml0L0LcPDhs2DBMmTIDFYsGFCxcY5IkqrzqdDkajEZ9++qnfKhONRXJyMiZOnAiBQMC4E1arlVVZySIiIi7JH/V6+wj1p06dQmlpKQIDA+FyufyqoQF971VoaCiioqIQEhKC7u5uVFVVscrTxYsXERAQgOTkZEgkEnR1dQ0IozSZTDh8+LBP0EsQ097eXtZHJzMzk/WR4QfIkZGRDLrl8XjQ2Ng4IBqnpaUFzz77bL+gVaVSIS0tDdnZ2ZDL5excFosFAQEBCAsLG7CKQ0psXq8XJSUl+Pzzz2Gz2ZCYmIiUlBSEh4ez9aO2thaVlZWor69n0shcOGtPTw+qq6tRVlbWL1j8Lu1baS75Q7PrObg/lSzVj8G+q7E2GAx4//33kZGRgba2tm9FvvmG+Rplynt7e32EBwaCDgxUDeF+nv7t7/v8jZz7OS6kCoCPY8y/BqrAcQMJOhYXqkWbPDmT3N+RVDrfKJvlL/jhyx2Tw+uPA8GtJtFn6f748EqCwVFAwD0POee0qXEDRRonf/fAPzZxgUjOFPgaNsfvJ+DvOGR8IvuVOJgul8unkSTwNYmfX/UcyPjXQio6NFb0rIloHRMTg87Ozn6qTcTlofvnEq4HOidfxIECfW5Adblx4D+TnwI0TCgUYvjw4ew9ogw+N8gjSWLu/UdGRjJ8Pkl+e71eBiX7pkFLQEAAUlNT4XQ6fZJdXHEKagtQVFQEvV4PlUoFgUDAEjTEGRkyZAjrSA98/f4QDJ3//pF5vV4mnQ58nWTgVrKHDh2K8ePHIzIyEk6nE6WlpThy5Ahqa2vZGAYHB2PBggVoamrCxx9/zDrDR0dHs+aY7e3tKCgo8BuwaLVajBo1Cnq9HlFRUYiJicGFCxfY70gJi3sP34R/SzwWf+s98YciIiKg1+vh9XpRWVnJZPJJIctisSA2NhaJiYlMsCM2Nha9vX397err69n5uMGFWCzGhAkTIBAIUFVVBbFYzOZRQkIChg0bBoVCAaPRiNTUVPa9kJAQZGRkwOl0oq6ujvWt4T47Mo/HwxJMXNPr9Rg+fDgyMzNZwoKqWSQ0QPdH0GZ6/hKJhP2/t7cX586dY0EP0AeR1Gg0kMlkrKJOvC2Px8N6djkcDpjNZlRUVKChoeHHTcS/YTfsx25erxdnzpzB5MmTv+9L+X/KiA/AzVzz4TFcp4wykFxOADnb3MABGLjxIjnN/mBXXHI1FxJF1wd8HfzQdZJDzz8fN0iiLD6R8fl8F/osP3tOji333vgB0kCwOI/HM2BGnTYsIqxSUMN1lOn4brd7QBge1wHnZrW5DjYfpnSpbDY30OPeD//aqRIVEBDAzkFQNXomtEnTWNFcocy6UCj0cfK4z4SCCy40jsuhovGlMXI4HOzeDhw4gFOnTsHr7etxQXOZX33iqsBxf84fA39VHv5z5VYN+ZA+7rt1JQHOj8VIgSsgIIA1c6YxdrvdcLlcrPEfGWH/dTodQkJC2Nwh5++bogoEgr7+RtHR0cw5B/rmaGZmJpKSkmA0GpGfn8+I4JTxV6vViI6ORkhICNRqNSNKZ2RkIC4uDgCYrHVISAiUSiVMJhNT77qUyeVyhIWFwW63w2AwQK1WIywsDAkJCUhISGAywJ2dnbBarSwAIfWoQ4cOoba2FsnJyYiKioLNZmNVZ5vNBrlc3i/potVqMXPmTIwZM4YFWZ2dnbh48SIjexPHRalU4tSpUwD6J1EuZ9y1hPtuEKdHqVTCYDCgubmZCSYQLxbogz5ZrVZ0dXUxtbfAwEAoFAq0t7f7BC1cGzVqFDIyMlBaWorW1lZWudVoNBCLxYiNjYXL5UJVVRX0ej1aW1sB9FVmMjMzYbFYYDKZoNFo4PF4EBYWxkQHBjKRSASpVAqNRsNEG9xuNwtEKPliMplYolUgEPgIKdBnSBWMOHxWqxWdnZ1wuVxQqVRMdII4QVRB1ul0LEnT0dGBxsZGuFwuBAUFob29/XsJXATeKzzrQCW0G3bDbth3bz/Wit+NdeSG3bAfjv0Y15Eba8gNu2E/HPuu15AblZYbdsNu2A/erlWm9Zt+n0uq93eMS1V0vqtrvB7nA/rfI/9nl/r+j8np9QelA659072acfiu5/NPzUj4gLLI1Kn8UuqdEokEAQEBDOdPUBj6Dr8iR/wSgsr4M65M/kCfoUokd80g0jvXAgICmEAHFyZEfVK4qnlUTaIqKVeo5HrOC5FIxCrcVMXiH18oFCIiIoIJ2TgcDjaWVLmk7/NhZRkZGejo6GANWAkia7fb+0nISyQSKBQKVikmyBxVBrjjpdPpIBaLfXoyAX39fOhagD64WEdHB6u2EXzPbDaz1hVKpZLxjRwOB1MXJZlqvV7PromOTQ0q+UYCKMTlc7vdbJ7RmAkEAvT09LDKHH9eqdVqqFQqBmuk+Wuz2Rgv8mr5WfSsCBkgk8mYuANVualqW1lZeaPScsNu2A27MvuxOixcuAwXgkXmLxjgw334DhuXL8JVCeMTErmQH+5n+bwDcgII5kC/J8lH4GtnhUvG9+cEc8nq/rgn3Ovmq3HxSe1XqmLlzzGnn/sjadI9EqGXC7PhXiOfM0T8CD5UgzsGXIUvPlSJe338Yw90n1zSOX+8+XsUf4wlEonP8+Q2ffQXnNLf/DnJnRsCgYBxI+h3dB4upI8LYQP65g85cP7ms7+x5P6e6zxwP3+poMzfu/FjXEdonj766KMIDAyERCJBRUUFDh06hOrqar/3JBKJEBERgaioKOh0OkgkEtjtdpSXlzNIkFQqhVarZd9XKpUIDAzEhQsX2PNVqVQM4uP1ehEWFgaLxYLm5uYBielKpRJqtRparZbNlc7OTnR2drLviEQiZGZm4qmnnsL58+fx17/+lR0jKioKKSkp0Ov1TATEaDSioaEBnZ2drMdYd3c3c+AvlUzhBmd8rhzfQkJCEBkZCaBPBdNgMPgEhkKhEEOHDsVjjz0Go9GI9evXo7q6GkqlEnq9nsGnPJ6+HjUXLlzwUR88c+YMNm7cCKPRCLVazQQMCgsLGdGcgrawsDAMGTIECQkJEIlEDH529uxZ1uBaIpEgMzMTCxcuRHd3Nz799FNUVFSw66UGtna7HSKRCAcPHsT777+P2tpaKJVKLF68GCNGjMD+/fuxY8cO9PT0YMyYMZg6dSrkcjnOnj2L/fv3w2AwICEhAcuWLcOjjz6K8PBwdo6uri689tpreO655/qNZ2RkJGJjYxEeHg6FQoGuri5IJBJERERArVYjKCgIcrkcRqMRp06dwsmTJ/tB8aZPn44RI0agu7sb9fX1DBp27NgxKBQKJCcno76+Hp2dnQM+V+48oCCJ+HlCoZCpkg0aNAipqalISEhAVFQUVCoVC46WLFnC+Irfpd2otPwETCqVIiYmBnK5HHV1ddfcXPKG3bBv27hOJ//n/EXQnxwv/TsgIIAFE9yM6UDH4DtrXP4G9+dc+V0KmrgKU1xng9t7g+8YXir7Sueh83L7m/C/yyX28u+N7+DyxQno2vmOPmX5uFLCoaGhCA0N9cEvc6+ZOy5cLg5xRbj/p3PS5wMCAiCRSOB0OiEWi6FSqeB2u30I0MTl4QYX/p4j/xnSOSkwonHgBmWkOEafH4irw8fN86+Dz/+hgIzOT31uuGpW3GdN5Gt/z4/PReGenzs3uQEUBUmXyqz6C4p/7EY8ho6ODkYSH+jeKMBuampiTp7L5WKd37n/JwUsgaCvfxIFLImJiRg5ciRGjhyJjo4O1NXVobi4GA0NDT4cOVLpo8CAGj5brVYEBwdDKpVCKpVCJpOxAEOhUGD06NEYP368T6NDoK/fUm1tLaxWKwICAljQ43A4oFaroVQqWaXpSgIWjUYDlUoFj8cDi8UyoLy/SCRiZHOqNvHX38GDB+OZZ56BQqHAl19+CYvFgujoaEgkEtZlndYA4lUQ6d3r9eLo0aMwmUyMoG80GtHc3AyPx4Pw8HDo9Xr2Dnk8HkZkJ7K9x+NhJHShUIioqChMmTIFYrEYX331VT/VQuppExQUBKVSCZvNhoKCAtaItLy8nD1X6nnU1NSEhoYGBAcHw2azMV5bR0cH7HY7tFqtzzm4jar51tbWBoFA4BPIUnVErVYjJCQEISEhCA4OhtFoRElJiU/QIhAI0NnZiTNnzqCtrQ1msxkqlQrBwcEA+tZk+jw3ecJ9prQn0DwUCASsZxAlYgCgubkZVqsVlZWV0Gq1CAwMhFwuR0BAAAIDA783ftx3HrQIhUKkpKRgzJgxKC8vx4kTJ77rS/hJmVAoZB26jUYj3nvvvW81aJk1axYSExPR1dWFHTt2+FVBumE37HJ2uarBQNAlcpwFAgHi4uIwYsQIWCwWHD9+3Cfr46/i4PV6+5H56Xd8R5RboaCMPJ2XSOdcGAn/Wi93L/Q9vrPLdai5n+c6DJfKynOdCn9wEW71xuv1ssyaQCBAVlYWbr75ZnR1dWHLli3M0aDP8ysSXCea/r7UMyVJVrlcjpycHAwfPhwdHR3YvXs3mpqafJTYBnqOAxk3cKGMIf2Ry+Xo7u72CTT5crhceA03aOQHQ9ygjcaAG4B4PF/3YSD4B3d86DsU5JEamL9qGr8i5S+Iou/yv0/BOPc4/Grkj9kEAgG0Wi26urpQXFyMiooKH8lhvjmdzgEbQXKPSdlngijp9XoGQ1q2bBkyMjIwbNgwVFRUoLOzExaLpZ8am0wmQ0xMDIKDg+F0OtHc3Iz29naYzWZ0dXWxgIgbYNB5/Sm7ORwO1NbWorW1lfUzogaV5OxzFZ8uZVKplDmfROx2Op1+zyuXyyGTydDV1QWj0QiTyeQz1+Pi4vDss88iOzsbf//739HU1ITU1FQmjiAWi1kvKLfbDYVCgbCwMMhkMpw/fx5A3zocHx+Pjo4O1NfXo7y8HAaDAcHBwUhISIBYLIbFYmFrkMViQWtrK8RiMetTIpfL0dXVBalUiuTkZIhEIuzZswdnz57tJ0Hc2toKl8uFgIAAmM1mNDU1QaFQYMSIEXC73aipqUFZWRmTZRYIBGhqakJubi50Oh0UCgViYmKgUqmg0+nY/ZDRsxjIL3K73WhpaWEVO6oUSqVS6HQ6REdHQ6/Xs8oR/z0Vi8VoaGhAVVUVO0diYiICAwMBfK1wSapiDoeDPVsKxrlQNBKHcblcPiIg1P+pq6sLzc3NbJ+g4yqVyv83ghaBQICMjAw8/fTTGDp0KP74xz9+l6f/yRlp1f/85z9HSEgIvvzyS784yuthEokEs2bNwssvv4y0tDT28n/++efXfGyRSISgoCAkJiZi2LBhiIiIYL8rKirCjh07vrHKyA374Zs/iM9AsBhy6skBGzZsGJ566ilERUXhpZde8skWcqsMfIeR6xhw5XxpEecGNdR4ke8o0vxsa2uDy+ViFYuBnHZuNYbv6HPvm/tvLiSJ+3MuxIrGhTtG/j7rT42K6xh7PB5kZGRg1apVUCgUePfdd1FXV+fT5JJ/jWT+xph7Pdx77OnpgUKhwJw5c7By5UqkpaWxjPHGjRt9JKL598qHNXGNfqdSqRAUFITo6GgkJSUxeVuHw4HS0lIcPXoUbW1tzOHjXzu/8sdXMfNXGeRmLCnopHu12WxsbnAhdANB+Lhjxf23v6Ca5gg3iOLDxPjz0d/nfqzm9XrR0tKCmpoaXLhwATabjT1rf7LhV2o6nQ6xsbFwu92w2+1QKpUwm83o6emBXC5nTm5TUxNMJlM/JbegoCBMnDgRkyZNQnh4OOx2O06ePInc3FzU1tb65XUAffOlrKwMO3fuRFVVld9r48oaA2By4gTvuZJ7Jvgn8HUgPFCwRA42BUT8Ofvb3/4W8+fPx9tvv41z584hKioKoaGhcDgckEqlrKpCMr3kqMtkMhQVFQHo45h0dXWhpqYGpaWlqKiogMfjQXBwMFNV6+7uZgGWy+WCVCplymCUmKDx6O3txbFjx3D27Fl0d3dDqVRCKpWy4LKqqgoajYbxhoqKiiAUCjFy5EhWrbpw4YLP+mq1Wtn+EhISgrCwMAQGBiIuLs5H3hjog53l5+dfstdbb28vLl68CIVCgdjYWGg0GraPdHR0oK2tDdXV1aivr/fZWwAw3hB33vX29voopFGlms9r4T5vqkJTFYwq3/6Ce65C4+UQBN+FfadBS3x8PP70pz9h0qRJWLt2Lfbs2XPVx6KoeuzYscjIyGATVyAQwGQyoaioCHa7nTXMycjIwJAhQzBkyBD89a9/ZVroP2YLDw/HL3/5SyQnJ+O1117DwYMHL5tNuhrTarWYPXs2fve73yEtLQ1An3b4ihUrrjlokUqlmDBhAm677TakpqZi0KBBCA4OZi/HhQsX4HK5sHPnzuvaFE0kEmHGjBlQq9UYNGgQc0StViu2bduGo0ePfq8v5k/d/DltAwUs9BlysFNTU/HHP/4R06dPx5tvvon8/Px+jr6/Y3Ald2nRFovFDPaRkpLCFn+73Y6Ojg4YjUbWoNFkMkEmk2H06NGQSqXYt28fioqKWLacu7lcifEzVXy4Em0wXNw5PyC7FHTJn+PKhR7RmMbFxWHFihUIDQ3Ff//7X5w4cQIul6vfBkU8Cv75/Bm3MkDfCQkJwYIFC/DII49gyJAhrDHauHHjsHHjRr/Xz58T/EoCHVur1eKmm27CjBkzMHToUMZdCAgIYD00ZDIZPv/8c+Yg+JNW5gcU/B42fNPr9UhLS0NSUhLCwsKg1Wohl8vR3t6OI0eO4PTp0wwKQ9UVIo7T+QdqLMkNVvi8K24VBQALuCmrzb0nrowzzYEfu3k8HpSVlaG1tRUCgYA1ErRYLHA6nQw2dCU9aUQiEaKjozFhwgRMmTIFdrsdu3btQllZGerr61m2vqCgAAKBAO3t7Sybzk2GpKWl4fbbb8fNN9+MpKQkKBQKuN1uTJw4EWPGjMGHH36IQ4cO+Z1HBH36+OOPffpWXcpInpYy9kFBQTAajQzWxDdunxRulXGgd/hS1ak77rgDd955J44dO4Zdu3axY9vtdni9Xga/IggTdXGPjIxkpHV6bpWVlTh37hxKS0thsVig0Wjg9XqZTLDFYmHvLFXBqLmk0WiE2WyGTCZDeHg4rFYrSktL4XK5EBwczDrSV1VVsa7zQUFBcDgcMBgMcDqdGDZsGMLDw+FwODB06FCYzWZWReO/KwaDgcH0BAIBk7Cm59HS0oIzZ84MCLkjczgcqKiogN1uR3R0NOM7UeAVFhaGkSNHQiqVIj8/Hw0NDX5hrQBYpQb4umcPtzkxGe0XtJ9QMEJJO9oPiNNC8DKuWABBfLnCA9+1fWdBS2RkJJ588klMnjwZ69atw5tvvunzwL+JSSQSTJ06Fb/5zW+QkJCAiIgIn67TDocDTU1N6OnpQXt7O4xGIyIiIhAREYGQkBBcuHABb7311vW6te/F9Ho9HnnkEYwZMwarV6/+1gIWAPjDH/6AuXPn+mQVxGIxUlJSrvnYYWFh+PWvf42JEydCIpGgu7sb586dQ09PDwYNGoTk5GSsWLEChw4dGnAxvhpLT0/Hn//8Z2g0GoSFhTFntbu7GwqFAlVVVT6de2/Y9TVuZeJSsCp+ZSI8PByPPPIIxo0bh3/961/43//+57PJDxT48PtfkOM3duxYrFq1CqNHj0ZQUBBbjCkjarPZWKNAg8EAmUyG9PR01t+hqqqKZSEHypj7g1aRyWQyln3kO13EleBXV2iT8Xe/fI6OPweVnGeBQIDQ0FA88MADGDNmDP79739j3759TMmH7/zynRvu//mOMTcQJRL0smXL8OCDDyIuLg5VVVUIDAxEaGgo0tPToVAoGKmfe15/QRK3YgH0zaWwsDDccsstWLx4MQICAtDQ0ID8/Hz09PQgJSUFQ4cOxdy5c3HkyBFcvHixXwaSjsmvQvBVmaiLOmG/MzIycMcdd2Ds2LEsYCGSd3R0NBoaGlBRUeFzPn4g6W9ecAMWfoDP/Rm3kkaOKJeDRX9oXtCY/RT6tYhEIoSFhcHlcsFkMqG7uxsqlQpRUVGQSqVoa2tDTU1NP4gQGTnNOTk5WLBgASZPngyXy4W3334bhYWFLLFJ71BhYSEjSrvdbrS3t8NqtUKlUmHmzJm47777MGbMGOj1enYOsVjM+p1kZWXhiy++wPr161FbW+sjEBEbGwutVguj0Yj29vYrHoOenh5ERUUhJycHISEhqK2txYEDB1BeXt7PuQ0ICIBCoWCVFS6s8JvYsGHD8Nxzz6G9vR3btm1jXeVpfQwJCYFer4fH09e8saKiAj09PdDr9dDpdKyPiNfrRVFREbZu3YqzZ88ybgr1eDGbzbBYLAz2S++ExWJBY2MjhEIh8w/S0tIgEAiwe/dunD59GqGhoUhOToZQKPSBDZ49exYRERFQKBRQKBRISkpCQkICwsLCEBkZidGjR2Ps2LEIDg7GZ5995hdySBWv0tJS7Nq1C4MGDcKIESPgdDrR0tKC1NRUBnu9lFmtVtTW1sLr9WLMmDEYOnQo4uLi2NpusVhw9uxZyOVyHD58mCmK8atq1MOFxnQgfpvH42HcT1pDaA5SgozgXxTs8YMagi3/PxG03H333Vi2bBlee+01vPvuu9cEY9Lr9Vi1ahVuuukmv7+Xy+VISkry+7vu7m4kJCRc9bmBvgeckZGBFStWICMjA0ajEefOnUNVVRXy8vJw8eLFbzWbJZVK8eSTT2LBggV4/fXXsWXLlm8lYElOTsaCBQtw1113QS6X4ze/+Q0KCgrw5JNPYubMmdflHAKBgGF8jx49ivfffx9nz56Fx+PBgw8+iDvuuANTpkxBQkICCgoKrgusgZyNrKwsFuz29vaitbUVNpsNkyZNQn5+Pnbu3Amj0XhdKzx8CwwMRGRkJMaMGQO5XI74+Hh4PB588sknKCws/ElkRf3ZQI4a99/8z2g0Gvz85z/H8uXLsXr1arzzzjvo7OxkThn9TVAyLi+Efg/0vT+EpX7wwQcxd+5cKBQKAGAd2sViMXQ6HYKCghAVFeWjpkUEWhK/8Be0XO5evV4vlEolsrOzMXHiRMTFxcHpdKKpqQm1tbWoqKhAZWUly/aR3CR3wxoILnW5caT7lMvluOOOO7BkyRK8//772Lp1K1OG8ScCQFk4buWFy/Phn4cc5ujoaCxatAiPPPIIgoOD8dZbb+H48eO45ZZbcMsttyA0NBSpqak4ffo0u96BSPCXGleSOC0oKMDOnTvZOrJ06VKsWrUKw4cPR0xMDOrr633GkB8ocY0CR38mFouRlZWFBQsWIDg4GL29vTAYDGhpaYFAIEBaWhqmT5/OcOhWq9Un4Bzo+fgLVPiVFQBM2hTogzWlpKQgISEB0dHR0Gg06OzsxNmzZ3HmzBnGc+TK1/6YTSAQQKfTobu7mzXJ6+npgVKpRFRUFNRqNWugaDAY/FZuExMTcfPNN+P2229HXFwcLly4gJ07d+L48ePo7e2FWq2GWCxmcJzu7m7WENdkMqG9vR0ymQxz587F73//e8TGxvo4cx0dHTCbzazBYXZ2NuLj4zF58mT86U9/wtGjRxk3JyUlBXK5HGKxGC6Xy2+1hWCGfD5eUlISlixZgqSkJFitVkydOhU7d+7EkSNHUF1dzT7vdruZfC1VKoCvVfSuxFQqFf7whz8gLCwMX375JWpqahgcj65PoVBApVLh4sWLqKysREtLC7xeL2voKJPJ2Lu9Zs0anD9/3ic46OnpYftub28vJBIJent7WRd2CvpvvfVW/PrXv0ZUVBQCAwPR1taGvLw8iEQiREZGIiEhAS0tLTAajT7QSLlcjqCgIAZXGzRoEAYPHgyVSgWgT61NLBajsbHRBw1EUDSn08nWvpMnT2LdunUoLy9n1fvs7Gy/ne39WXd3N9rb2xESEoK0tDSmXtfZ2Qmj0Qi5XI5JkyZBKpXCYDAwX5Ob8Fer1SzQFYvF0Ov16OnpgdVq7QerHyhQ5SpIUkKOm8CRSCRQqVSMD0OVpu8DZvqdBC3Tpk3DE088gdOnT+PNN9/8RpkEvonFYowZMwbTp09nP2ttbcWRI0eQl5cHt9uNm266iakpjBw5kmW+CgsLkZqayrqVXq0pFAo8/vjjWLx4MZRKJXp6ejBt2jQ4HA5UVVXhs88+w5dffonm5ubr/lAFAgGmTp2KO+64A0ePHsXGjRu/FeJ9QEAAli1bhgceeAB6vR6HDx/Gu+++yxbh0aNHQ6fTYdWqVXj//fevqAzvz5qamvDII48gISEBdXV1qKioYAFYfn4+Fi5cyOQBr5cJBALMnz+fbQCFhYXYvHkzzp07h7CwMDz88MP485//jGXLluG5555DXl7edd/kRSIRnn32WYwZMwYxMTHQaDSMKOf1epGRkYE//elPOHv27HU97w/FLgfn4sOgJBIJpkyZgl/+8pc4dOgQ3nzzTSbp6E/ylf/ekVMql8tZN/X58+dj3LhxDCNdUlKCU6dOoaWlBTKZjGXfqCM1l2Td2dnJMOpAf2lhMgqiuH/oXdFqtZg7dy7uv/9+yGQyli0nx3vPnj0MC9/T08OCLRozOh43k87N4nOdbW5lhn4+c+ZM/OIXv8Du3bvx/vvvs3WEewxulm2grD8fwkTHp4Bl4cKFePTRRxEfH4+1a9fi1VdfhdFohFgsxqxZsxAYGIicnBycOnXKJ0FwKSll4OuAwuPxoLW1Fdu2bcOJEydw4cIFFBcXw2g0QiqV4uLFi+jp6UFwcDBLkPAhaP7ujzvO3HlEn5FKpYiMjGSE63379uHgwYM4c+YMEhISsHLlSjzyyCMYPXo0vvjiC+zevRtOp5MlSvgOqL/5Q8+OnAkADDLY09ODmJgY3HrrrZg2bRri4+Oh0WgYv8Zut+Po0aN48803cfToUTidzp+Melhvby/Onj3Leneo1WrmWFF3b1KJ6uzsZH016F2PjY3FokWLsHLlSuh0Onz00UfYvHkzU7ZLSkpCd3c37HY7465QDw+Hw4Guri4IhUKMGzcOv/vd75CcnMyuq6SkBDt27EBhYSHsdjtSUlIwbdo0ZGdnIyQkBBMnTsRrr72GRx99FKdPn2ZqU16vFzKZbMC9jgJOrgmFQoSGhiIqKgrBwcFQqVQYNWoUEhISMH/+fJw6dQq5ubkMvSAQCKBWq+FyuVg/FZlMdsU+xK9//WtMmTIFRUVFyM/PZ2gEqkzL5XI2r6urq9HW1sbmbXd3NxoaGnzWojNnzvR7BwjaR8kLCl5oz5gyZQoWLVqEWbNmITk52aeqW1tbC41Gg8TERKjVatTU1MBisbBrCA4ORlhYGAQCAdra2hAdHY3MzEzI5XKYTCYcPHgQycnJGDFiBJYsWYLi4mJ2j2q1GoGBgbDZbFAoFBCJRDCbzdizZw+Ki4uh1WoxZMgQxMbGfqNkI3FvRCIR2tvbUVtbi3379uHo0aNMvjkpKQklJSUoLS1Fb28vysvL0d3djYiICHg8HibbLRAIWIKLIHVX4ptpNBqkp6cjKiqKQSAbGhrQ2NgIh8MBgaBPYSwoKAgJCQmIiYlhYgrftX3rQUt0dDTeeOMNmEwm/PrXv76mgAXoG9wHH3wQCoUC3d3dyM3NxV//+leUlZWxjOcHH3zANobMzEzWIKqsrAzJyckoLS29pmughUKpVALoW0yCgoIA9MGdhg4diszMTDzzzDNXHHFfqUVHR+Pll19GZ2cnXnnllctqcV/tOVatWoWf/exniIyMxIkTJ/Dkk08yeNaBAwdQVFSEcePGYcWKFfjyyy+vOhDs6elBSUkJLly44APfCQ8Px/Tp06HT6dDY2Og3W3YtVl5eDgAwm814++23sWHDBrhcLowdOxY9PT3MWQWA+++/n6l5XA+bO3cu/vWvfyEkJAQBAQHo7OxERUUFSktLUVtbi3HjxmH06NG49dZb0dTUdM1B9g/ZLrW4E6QL6HuvHnzwQXR3d+PPf/6zzybLhwvRz7jGzRg5HA5ERERg+vTp0Gq1cLlcOHz4MN544w2cPHmS8TnoGqi/g0gkQkdHB9xuNwICAlhljnsNXNgO97wE2yFohkAggEKhQEJCAvR6Pex2OxobGyGXyxETE4OYmBhkZ2cjKSkJ77zzDqqqqtDd3e1DTOc785cK/rjwJ4/Hg2HDhuHZZ5/FhQsXsGbNGphMJhYkUDaN+D9Ewhzo2REkgR9IxcbG4vbbb8f999+P+Ph47N27F+vWrUNDQwM8Hg+OHz8Og8GAiIgIjBs3Dl988QWTKPVHjPf3XMnJN5lM2LZtG4MwUEUqMTER6enpkMlkLFPp9Xp9eqYMlJDgBsP+ghqhUMgSLB0dHTh+/Di2bdvGej7IZDKkpaUhISEBnZ2dyMvLY1lnbuWPG6hwK0z0M+71cWVMFy1ahF/96lfIyMiATCZjwS45y1lZWRg8eDDmzJmDuro6VFdX/2Qqtx6PB+fPn4darUZsbCwiIyMRGhrKsudebx/xm6pOSqUSEokE9fX1sFqtGDlyJJYuXYqIiAj873//w3PPPQe3243k5GRERERAIOiTO+bKcYeGhrL+OgKBAGFhYZgxYwZDdbS0tGD9+vVYv349e1/pfdq5cycWL16MGTNmYPz48cjOzsarr76Kl156iWXuKysrGf/Dn/FhQQKBALGxsYiPj0djYyMOHjyIc+fOISAgAOPGjcPYsWMxceJE5OTk4L333sORI0fgcrmYHLFEImFV0stxfLVaLX7zm9/g/vvvR1NTEzZv3owdO3awCjSNk0AggNPpRGtrK+rq6vpRAPg8PH7ViOCX3OQMAKSkpGDOnDlYunQpUlJS0NjYiI0bN6KjowPz5s3D2LFjUVhYiPPnzyMwMBAxMTGMWM9t0GgymVBbW8uqoQ888ADri7Jq1SocOXIEkydPxiuvvIKcnBxkZWWhtbWV3ScFKyRZbbFYUF1djcbGRoSGhrKqeG1tLbtum812Wag5vfNNTU04c+YMg7GOHTsWkZGR0Ov1aGtrQ1tbG6RSKRISEti5SBQC6Fsf9Ho96urqYLFYBvSZ9Ho9kpOTMWzYMAwfPhwZGRlITExkZHtKirS3t+PMmTPYvHkzDh48yKqZ1Ez0+6jYfutBy7hx45CQkIAHH3wQZWVl13QsoVCI8ePHY8qUKQCAxsZGrF27FseOHfN5EbgvCumw06Roa2u7poVbIBAgOjoa48aNAwAcPXoUjz76KOLi4vDLX/4SU6ZMgVarxe23345Tp07ho48+umSX3m9iI0aMwOrVqyEWi3HnnXeisLDwuhyXb5MmTcLy5cvZ4v3ss88ytQ+gLzNAuPfIyEiMGzcOmzdvvurzkdNDptfr8ac//Qlz5sxhL+X1DP68Xi/Twler1Vi+fDnTIn/mmWd8uubu2bPnunFpNBoNfv3rX+Ohhx6CTqdDb28vDh48iD//+c8oKChgGaWCggJERERg7ty5+Pzzz3+yQYs/cjE5a/w+JrNnz8bkyZPx0EMPobi42O87zMXd+oP1kCMuFAoxYsQIZGZmQiqVory8HO+99x527NgBt9vtw1sA+qoqBNdoamryW10YCOLD/TkXa0x47OTkZHR1deHjjz/Gtm3bEBkZiYULF2L8+PFISEjAPffcA4PBgP/85z9s8x2I8D9QoMbnuMTGxmL16tVQKBR44YUXUFJS4gMv5fa7ofvk9iLx9/y41Z7e3l5ERETgnnvuwcqVK1lW7pVXXsHJkydZx+mmpiZUVFQgMTERmZmZGD58OJqbmwfkAHF5PdyfyeVy2Gw2HyiESqXCoEGDsGTJEsydOxdyuRy1tbWs+k0OG99p4gd8FOhxf048ErvdjrKyMhiNRgQGBiIpKQkpKSmYNGkSlixZgoiICEZiNZvN6O7u9rvJ87lb3GCdawQPjIuLwxNPPIHbbrsNer0eFosF27Ztw+bNm3H48GEYDAZIpVIsXLgQv/3tb5GcnMwaKpLa1I+92kJ8lM7OTtTV1bEmjiTp6vV6WfWFflZZWYmuri6oVCqMGTMGWVlZyM/PxyuvvIKOjg4mBez1etHW1oaKigp0dHSwZxYUFASPxwObzQan0wmNRoOoqCgWvJaVleHo0aMs2y+RSNDc3AybzYbi4mJUVlaioKAADocD06dPR05ODl599VUcPXoUhw4dYgTygYw/J3Q6HX7xi19gwYIFOHfuHN5//30cOXIECoUC5eXlLHiZM2cObDYbWltbUV9fD7vdzvgtFCiQkaADV4EtPDwcjz/+OO6//37U1tbixRdfxMGDBzFs2DCEhYUxf4s4RVqtFqWlpd9o3xIKhdDpdADgs99mZmbi5ptvxpQpUzBixAhIpVJ8/PHH+Oc//4ny8nLodDokJiZi+PDhKCsrg9vtZiIcra2tjBtI91JfX8/ko6nq6nQ68d5772Hr1q1M9riurg5DhgzBqlWrYLVacfz4cVitVnR1dTGIGK13JH5hMplYbyuiP0RHR0MqlbI+Pf6sp6cHXV1d0Gg07PORkZGYO3cuFi5ciNDQUAgEAhw/fhy7d++GzWZjvXZIMIab4HC5XAwixvddJk6ciHvuuQczZ85k7wfB3oRCIRwOB8xmM4RCIWssmp6ejmXLlmHdunVYs2YNSkpK+imTfZf2rQYtpGp14sQJfPDBB9e8UFJpUCKRwGKxYMuWLfjyyy/ZhhkYGIhZs2YxzsqQIUMwZcoUHD16lG3O18pPkMlkeP/996FSqVBcXIzf/e53KCoqQnFxMYqKivDnP/8ZS5YsgVqtxpo1a7Bnz57rkqUXi8V44YUXMHToUNx7770oKChgQQORyDs6Oq65IvH888/jySefZBH3b3/7Wxw/frzfuFFGNTAwEMOGDbumoAXoyxAkJCRg1qxZWL58ObKzs5kiBmFlr5d5vV5s3rwZzz77LCIjI5m6i8fjYQu5x+PB6dOnsXPnzgEXm29iTz31FBYvXozs7GxIJBIcPHgQ//73v1FWVobS0lKfSkB6ejpiYmKY9OxP0bjQIvo//3e0KEZFReE3v/kN657MJaZLJBKIRCK2yXIz41znjCAJBM9JT09HREQEHA4Hjh8/jsOHD8PtdiM0NBQpKSlM2jImJoZp8be1teH06dM4c+YMGhsb2WZN18O95oESI3Q9ERERuP3225GdnY3a2lps2bIFX331FTQaDaqqqlBQUICFCxciOTkZN998M44cOYJTp071g2vxx5R7DsAXr04V4ldffRVjxozBggULcOjQIUilUgwdOpTxUiigcLlcDJ7h9XoZPG2ge6NrUqlUWLBgAe655x7ExMSgtbUV7777Lo4fP856JNCxKioqMGfOHISEhCA1NRV79uzxCT642TyCRXF5KAQdBL7mMoaHhyMtLQ05OTmYMmUKwsPDWWUrLCwMBoPBb18UPtyNgmc+hIz+drvdyM/Px7lz5zBjxgwsWrQIWVlZiI2NhU6ng1AoRFNTE7Zv346tW7deUU8rbiDM7Q/k8Xggk8mwbNkypr4ml8uRl5eH1atXY+/evQzGRE4aZYQrKipQV1f3rXLzvmsTiUQYNmwYzpw5A6PRiLq6OkYyJ96ZVCplcKWGhgaUlJSgubkZo0aNwty5c+F0OvHXv/4VtbW1bKwcDgdaW1thMBjQ09OD0NBQGI1G5piazWZcvHgRVVVV0Ol0zDmVSqXIzs7Ggw8+iKlTp7LGhVR9M5vNcDqdzDHu7e3FrFmzkJKSwnyZb8JLFQgEuPvuu/HAAw+gu7sbzc3NDDLU3d2NgwcPsnsfO3YssrKykJiYiPLycnR2dkKv1zOkCBnts8QdIUtKSsLUqVPR2NiIF198EfX19Xj33XcxefJkJkSi0WiYqhZB8a+0h5tQKGRNNy0WC5unc+fOxWOPPcb4PjU1Nfj000/xzjvvMEedJI1pvg8ePBiJiYmQyWSQy+XQarVob2+HzWaDx+NBR0cHk1C22WxYt24d0tLS8M4778DtdkMqlTI1L6lUytQI165di08++YSJDRgMBia/TNVPALDZbLDZbCzJSiic8PBwpnTGXz8JUtjd3Q21Wo0RI0YgODgY48ePh1arhUAgQHV1NQ4fPsyCkI6ODnR2dvpUaKk60tzczBIrZBTgPvzww9Dr9T7+VE1NDU6ePInTp0+jrKwMdXV1MBgMTGDikUceQWZmJpYvX46SkhJ88MEH/QoF36V9q0HLokWLoNPp8Oyzz163qIycD3JKpFIphgwZgnvuuQe33347g/TQZwBgyZIlWLRoEWpra7Fs2TImU3o1JhKJMHz4cFgsFmzduhXHjh0D8DWe8v7770dpaSkef/xx6PV6TJ8+HevXr7/mDWPVqlWYNm0a1qxZg61bt+Luu+/GL37xCyQkJLCgpaysDM8///xVE/NzcnJw0003sc1y7dq1WLduXb8Sb09PD15//XVMmTKlHyznaiw+Ph7PPvssFixYAJVKxZ4t2fDhw/HMM8/g73//+3Vz4ru6uvCzn/0Mn3/+OcuwkblcLrz66qtYvXr1NcMZdTodHnvsMTzxxBNQqVSwWCzYt28fFi1a5DerJpfLmXTjwYMHfTo2/5SMHD9ySv1VCWghvu+++xAdHY1HH32USUkSOZP7XlHywl/vAe4CK5VKodfrGU63ubkZKpUKs2bNwoIFCzB48GCEh4cjKioKSqXSJxioqqrC0aNHsXfvXhw+fBgXL14EAObw89cV6nZMGTl65omJiZgwYQJzMIg0bjKZsG/fPhQXF8NsNuPxxx9HVlYWsrOzWZaWCPH+khP8d4c/PnPnzsWiRYvwj3/8AwUFBVi5ciWefvpp5pwHBQWhoqICf//737F161bmRHOlMi9nI0aMwJ133onExES0t7dj165d2LdvH1uTrFYry4a3tbUxuBZt0DSGNAf4alskv0nOH5FGk5KS8PTTTyMrK4s5HlwbNWoUfvazn+Ff//oXLl68eEVrFzcQpooVd0xra2vx6aefQqfTYeTIkcjOzmaVIpvNhm3btuHNN99EcXExE3LgEv+5MDF/z4zGPzg4mFVpFQoFrFYr3nrrLbz44oswGAw+WHaPx8OSSU6nE3v27EFbWxtb13/sVRagb9zCw8OZrK7VaoXBYIBCoWCVFQpy29raWMNAr9eLiRMnIjExEWvXrsWOHTuYH0GKYFSxiYmJQWBgIOvTUlRUBIPBwKquZrMZ+fn5OHr0KMaNGweNRsNaMKjVatb/Iz4+Hnv27EF1dTVUKhVOnDiBQ4cOYdGiRViwYAGqq6tx5MiRb3T/MTExGD58OHp7e1FdXY1jx475ZNUNBgO2bduGgIAAhIaGQqvVIigoCE6nEx0dHVCr1QgICIDH42FrkkQigVKp7CcCMHbsWISFheGTTz6ByWTC448/jokTJ+Lzzz/H3r17mYJZcnIybrnlFqhUKqSnp/dTz1IoFAy9we1j4vX2yRtzJYanTp2KP//5zxg+fDgMBgMOHz6MrVu34vjx4z7Out1uZxXT+Ph4jB49GgEBARCJRKxyYbVa0dnZCbfb7ZOA9Hq9eOedd5Ceno6KigqIRCIMGTIEK1euREJCAuOPDB48GM888wzmzp2LNWvWsHYa7e3tLKlFDR1dLhe6u7sZfJmUIENDQyESiVjwxDW32439+/dj69atrIFpRkYG+31TUxM++ugjhg7hXj/Q9y5ERESgoaEBvb29/XyGkSNHYvXq1cjJyYFAIEB3dzeOHj2KU6dO4dy5c6wKSHLV3Dn06aefQiaT4YknnkBsbCymTp3KIHEnTpz4XgKXby1o0Wq1+NWvfoWSkhKcPHkS0dHRiIqKwuDBg/H5559fVsf6cqZSqXDXXXdh3LhxGDZsmE+Jk4vjBr7OhCYmJuLQoUOYOnUqzpw5c9XndrlcsNvtqKmp8fs7o9HIFoLY2FgftZirsdDQUNx9992oqqrCv//9byxcuBCvvfYavvzySzz00EOQyWRITU3FI488gjvvvBPFxcUoLi7+RufQ6/WYPXs2MjMz0d3djbKyMmzYsMEvZ0YkEmHevHkArs8GGB8fj7vuussHIsF1ZCUSCR566CHU1tbitddeuy7n9Hg8OHz4MO655x7873//Yx1lnU4n3nrrLfznP/+55oBFJBLhH//4B2699VZIJBIUFBTgD3/4A3bt2jXgd7KzszF58mTk5ub6QPJ+ajYQrIqMHI7k5GTce++9+Oqrr1BRUcEkKjUaDYqKilBRUcH0+7kchUslJWitEAgEUCqVGDNmDOLi4jB8+HCkp6ez66I/XK5GcnIykpKSEBwczDYuaiLIr2pwJWdJWYbM7XYz/PnHH3+MlpYWxiUhdaLOzk7WA4YqSlzjw8T4lSY+xC4lJQWrVq1CWVkZPvvsM0ycOBErVqzAvn37sP3/a+/Lo6Ms7/0/syaZNTPJJJnsC9kJJAGEyCarF4O4a6tWWqm49ihWe+u9/dUeT2uvXrWu1VqwipZFUEE2gbCELYTsJCRkT0gmyWSSWTL7/vsj53l8ZzIBWVrR+37O4QCZybs87/M+z3f5fD/fPXvg8/kwc+ZM3HTTTXjggQfQ1dWFqqoqerxLBV78fj8SEhJQWlqK/Px8WK1WnDlzBl999RU6OzsnbHCET0+cEmaPEXI8psNCPmNK9wKgGaGUlBRKq/P7/ZSy09/fj8jISMyYMQNr1qzB8PAwPvjgA9r0LtgxCkWBC5XFAsblTzdu3AiLxYJXXnkFarUaNpsNNpsN27dvx9tvvx3QLJDQ0iYbS2bmh3yHOCxr164FAFRWVuLdd9/Fli1b4Pf7ERERQSPswPi6k5WVhcLCQlRWVl41Lft6RajsO1MEwul0wmQyYWBggBbYp6Wl4fnnn4fD4cAHH3wALpcLpVKJ8PBwGkAhDAKSxSPHbG1tDTiXx+NBeXk5PB4P7HY7oqKi0NHRQWWO09LSaANVUrvo8/mwe/dudHd347PPPsOpU6cgl8u/c1aCIC4uDm1tbXjrrbfQ3NyMs2fPQiwW06aYwDittbGxkWahiIOv1+uhVCppsJMEd5hrBsHixYvx8MMPY2xsDHv37oXRaERdXR22b9+O06dPY3R0lDrMmZmZKCwsxNy5c5Geng6JRBLgJNhsNnR2dtL6XwK/3x+QXU1OTsbPfvYzFBcXQ6/X49NPP8V7772H3t7eCWuI2+2mgZzExEQkJibSukOBQEAL77u6ukI2qCY1tcA4fXvlypVYvnw5Ojo68OGHH6KmpgY5OTl44IEHsGjRIrzwwgv4y1/+Ah6Ph3PnzsHpdNJaJ6fTSdcwMmeIPDKzFi0U+vv78eqrr6KnpwcPPPAAEhISAIzvZfv27cPHH38MvV4/gT1AHC2JRDLBqREKhZg1axZefvlllJSUwOPxYP369di8eTO6u7shEAggl8spZZocVygUIjIyEj6fDykpKUhISIDJZMKBAwfA5XLxhz/8AXa7HTfccMP30vT7X+K0cDgcvP7664iMjIRIJEJZWRlycnLQ3t6OnJwcPPjgg/j5z39OFQ8uB8xJq1QqqR446UA8MjKC/v7+gAX8xhtvREREBI2oPP3003j44YevKPthtVpRVFQEqVSK+vr6gM+4XC5mzZqFhx56CDExMbDZbNiyZctFeaqXAo/Hw5o1azBt2jSsXbsWbW1tcLlc+M1vfoPdu3dT3igpsnv++eeRm5uLlpaWy/KCZ8yYgXvvvRcymQx1dXV4/PHHUV1dHfK7fD4/oGfL1WbRnE4ntFotRCIRhoeHcejQIbz99ts4f/48pFIptm/fjptuuglPPfUUPvnkk6t2JgjIYmmz2ajTotVqUVNTc9U1NGKxGP/1X/+FVatWgcPh4J133rlkU1PSfG/69OnYtm3bNRdxuJ4QXNhM+MHMOgIAePTRRxETE4OioiJ8+eWXEIlEMBqNiImJQWtrK/7zP/8TZ8+enSCRG6y4FHxOwlUXi8VYunQpzZKQoskLFy5gZGQERqMRVqsVcrmcFiwKBAIsXLgQjY2NtHlgMJibEzEomYGVjo4OfPTRR7BYLKirq6OOCZElLSkpwYoVKyCTyVBfX4/m5uaADSJUhJ7pxBD6EhlHPp+PVatWITMzE08++SQaGxsRHh6Ol156iVJsAGD37t1YvXo1nnrqKeTn56OpqYmqx1wqWEDWhZKSEigUCjQ0NGDnzp04duxYyMyvWCxGeno6VRvSarUB3wvOXE32b3LfJJDkcrnQ2dmJmpoanDp1CiMjI4iLi6Pyzvfccw/KyspQW1tLxzL4mQUjeLxJY1JmE06TyQSFQoHIyEgYDAbU1NRQQ0IoFAZIzzKPG+wkMTNJMpkMjz76KFavXg2r1YovvvgCGzZsQGNjIwQCAQQCAZ1/5FhRUVG45ZZbkJOTgz179sBgMATQMH8MmRYmSHZarVbTInyj0QiLxYLR0dGA4va1a9dCoVDglVdeQX9/P4RCIRQKBZKTk+F0OmnwgNSkEMpZKPh8PtqAsqurCy6XCyaTidITZ8+ejdTUVGi1Wmi1Wng8HiQmJmLKlCno7u6Gz+dDZ2cnwsPDL+t+iTOwc+dO6PV6OBwOSCQSSCQSGlAFxp91REQEVQfTarX0nTKbzVRljklpZc7N+Ph43H///UhMTMRbb72FhoYGKJVKHD16FBaLhWZvJBIJPB4PUlNTkZCQQOt94uLi0N3dHbAuAwjI5BDDm2nQ5+bmYu7cufD5fNi5cyfefvttmtEOBsmqk2s3mUwYHh6m8rxut5sGgyaDx+NBeHg4SktLcffdd8Pr9aKyshL79u2DRqNBXV0dNBoNcnNzkZ2djeLiYkoPJBlMkp0j6wGRxiYNOonzPJnN6fV6qTT5woULqdPicrngcrkgkUiQnp5OM2Xx8fEoKChAYmIieDxegC3N4/EglUpx11134dlnn0VmZiZaWlrw4osvYvfu3XA6nbRHls1mQ19fH91beDweFAoF8vLyMH36dNx5552YPXs2Tp48iU8//RQ33HADFi5ciKSkpKsOxl8p/iVOS05ODubNmwebzYbm5mZUV1fj97//PRoaGlBaWornnnsOq1atwnvvvXdZx/X5fGhubobZbIZUKgXwbSPJb775Bjt27EBnZycGBwfpIsXlcvHEE09gzZo1VEls0aJFEIlEVyQV7Pf7J0RcgPENu7i4GP/zP/+DG2+8EUajEevWrZv0ZfuuKCkpwbp161BWVka7z/f09GDDhg0B35NIJLTg8nIhEomQnp5Om2KR4tLJkJ2djZSUFHA4HFgsFjQ0NFz2OZmoq6vDkiVLIJVKodFoApQ2zGYzfv/732PPnj1ITk7GnXfeib///e9XdT4ClUqF5557Dmq1mkbnk5KSsHTpUhw/fvyKqWgcDgfz5s1DaWkpBAIBXn/9dbz66qsXzS6KRCLMnTsXt9xyC06dOoWTJ09elbN7vSN4owK+NQzJZ4WFhbjrrrvgcrloNLGyshJNTU2YM2cOnnnmGcybNw9nz56lBcZM9SommJF5t9tNZXGzsrJoVL6pqYlmuIjTYrPZIBAIkJSUhJqaGixfvhy33norIiIikJ+fD6VSSdWwLsd5Hx4eRnl5OY0SEona8PBwLFiwAE8++SRuuukmGI1GfPHFF6ipqZkgrxxsfJKxYxoJJHpKxquqqgpffPEFnE4nzp49C4FAAIvFAj6fT4uQs7KyIBQKYTKZ6DlJw8JgZ4FZ7xEREYHs7GwkJyfD4XDgzJkzOHDgAEwmE8RiMbhcLpUy9Xq9mDZtGmbOnAlgnDJGDJxLgWlcERqQzWZDS0sL3n77bfB4PNojgkmZcblcuOWWWzB9+nRMmTIFtbW1AfcwGUI18iRjy+FwkJKSgvvvvx9paWmIiIiAzWaDSqXCf/zHf6C6uhrt7e30HMG9g0KJKpAsAQmC3XvvvZBKpXj11Vfx7rvvUqOPGDRk7InDS+ZoY2MjTp8+HdBHiPn3Dx2k11B6ejrUajWkUikcDgd1VPR6PfR6Pa2TEIlEWLVqFbq7u6HT6bBs2TJYLBYkJiYiLi6O0mR8Ph+ys7ORmpqK8PBwHDlyJOT5yXtotVoDvtPa2oq6ujqcPn0aU6dOhd/vR3t7OyQSCVJSUjB9+nQcOnSIZvcuJ1odFhYGmUyG/v5+ar8Qw93hcNCMDZfLhVqtxty5cxEfH4/29nb09fXRrDFZr0hDQQLm/JgyZQqKioqoghSpDRscHKTvBLkemUyGBQsWIC4uDvX19aiurkZKSgpUKhXGxsaoalvwXM/NzQWPx8PIyAjMZjNVdoyPj0dnZye+/PLLCRkEAh6PRw13i8WC5uZmNDQ0QK/XQ6FQIDw8HHa7PUDyeDJwOBzIZDKkpaWhtrYWH3/8cUAtstVqpccoKCjA+vXrYbFY4HK5aLd6hUIBmUxGbVOy3zscDvT19V2Usk/mklQqpe06fD4fwsLCsHTpUlgsFhw/fhx2ux1paWmYNWsWoqOjYbFYoNVqaSBXKpUiJycHY2NjKCkpwZQpU3DixAn84Q9/oP2HiHNLFFQJBAIB1Go1iouLMXfuXNx6663Izs6G2+1GT08PRCIR+vr68Mknn0ChUHxvNXLX3Gnhcrl4+umnIZPJ8Oyzz2LLli0B6ll79+7FT3/6UyQlJV32sX0+HzZt2gSFQoHVq1fD6XSiqqoKn332Gaqrq0NOCp/Ph3fffReNjY04dOgQvUbCVb0WEAgEmDlzJn73u9/hxhtvhNVqxaZNm7Bnz56rVg6bM2cOBAIBPv/885DHIgWmN998M1WbutwiKbFYjClTpiAqKgoOhwOVlZWTKn9ERUXhoYceopJ7w8PDqKysvOL7A0BpHJNBJpPRRfJadWIVCAR4+OGHsXjxYng8HnR0dKC1tRWzZs3CPffcg6qqKqxfv/6Knp9SqcS8efOgUqmwc+dO/PWvf53UYREKhUhISMCKFSuwdu1acLlc7Nu370dL6SAIXvCIQUf+zefz8ZOf/ARmsxmvvfYaNm3aBKPRSKNnFosFjz/+OO1LQdRcmMY68//BmY+amhps3rwZhYWFcDqdGBgYQHl5Oaqqqihnmvk7Wq0WbW1t0Gq1KCgoQFpaGpWiBCZ2OSfOAlPiM9hQJHOC1MPw+XxMnz4da9aswaJFi+Dz+bB9+3Z89dVXAcbIxYzOYEU2YHxznzt3LmQyGV577TVKXxgbG6PHIobP0qVLsWzZMpw6dQp1dXXUcQ5Vz8J0nkghc2JiIpRKJRUtIPSR6Oho+P1+qjwYFxeHn/zkJ3RTbGtro5voxRwJpuNEmpwRKdihoSF8/fXX9LuEkkcMMdINWiaT0UZ+hA9Pjh18Th6PBy6XO6F/DFHpEQqFeOCBB7B8+XLweDzU19ejqqoKJSUlWLRoEc6cORMgScocw1D0JpLBcbvdiI2Nxfz58yGXy7FlyxZs3LgRRqORFh4TR530+ImPj8fKlSuxZs0aeDwe2qCWjBezVuiHDq/XizNnzsBqtSIqKor2ThkbG6NKchaLJaCB4q9+9SskJCSgtbUVDz/8MAYHBzE2NoaYmBjEx8dDoVDg2LFjMBgM4HK5SE5OpvVSwRCLxYiPjwcwTu1hOh5EKnZ0dBS9vb2IjY2lQjPp6elUKepyKWHh4eFQq9VwOBwBtovP54PZbKb2T0REBBXwuPPOO8Hj8dDc3EwbO5JeMCTLQQRvmLQjr9cLlUoFiUSCrq4uiMViJCcnUzEDojJGaGckq7R161Zs376dSgBnZ2dDKBRicHAQ9fX11K4gznZWVhZsNhv4fD6kUint40IYEEQeHvi2Bwn5o1KpcNdddyE1NRUdHR04evQoqqqq4PP5oFKpqCM/WS1esBjM0aNH8cYbb6C2thbHjx8P+G58fDwdp+TkZBiNxoDnR7L3VqsVubm5EIvFGBwcRFhYGGw2G6xW60XfO0JPnD9/PiIjIzE2NobR0VFwuVwkJSVh5cqV6OvrQ2trKxITE5GcnAyDwYC+vj5YrVZ6D2FhYUhNTaXUaSLXLBQKqWpjsC3C5/MRFhaG9PR0ZGdno6ioCEuWLEF2djZ1bmQyGWbMmIGmpiZKE7zS3nxXi2vutKSmpiI/Px/79+/HV199FWD08Xg8LFq0CHl5eXjjjTeu6Pg2mw1/+ctfsH79evh8Pjgcju9kWDIjbmFhYZg1a9YV0dNCITMzE+vWrcPixYvhcDjw9ddf45133gnZ1fZyUVRUhIGBARw9enTCZ1wuF4mJibjtttuwcuVKnDt3Dn//+98vqQl+MbS2tqKlpSXkmEZFReE3v/kN5VZbLBaUlZVdUdH/d0VYWBjuvfdehIeH04Zi1+q4jz/+OPh8Pnp7e/GnP/0J5eXleOaZZ7B27VosWLAA33zzTQAf/buisLAQixcvhk6nw9atWyfQ2Xg8HlQqFW1G9eCDD6KwsBDNzc347LPPcOTIkWtGgbteEZwtYGZCfD4f8vLysGzZMhw9ehTbtm2jm7RCocDo6CiWLl1KKWKkSRpzU7hYkbXP50N3dze2bdtGqQ5jY2M0s0KK50lEn9RbjI6OUlUtYiwS2inwrTFLMh3EKCYbJuHLk2sNphvl5eXhl7/8JRYvXgyXy4WvvvqK9jUBvnVuLhaQCHacyPVkZGRgeHgYZ8+eRUREBNxud0Afibi4OCxZsgQ/+9nP0NjYiPfff5+e92JOUnCBPLPjNlFd8vv9GB0dpc3OYmJisHr1atx11100yrp//34MDAzQjA5xAoMzEUyBBVL/Q/jrpFaEUKyIAILX60V0dDSWLl0KmUxGVRbJM2KOV/A9kflCfkauj3xPoVBg+fLlEAqFaG1txVtvvYWmpia43W489NBDmD59+qQGaihZbuZY5ufnY968eejv76eGIJEmtdlsNModFxeHwsJCrFy5kkq/bt26FQcPHoTRaAyo8yDH/r6Uf64VCLWKdDEfHR2lynZhYWFQKBSIiIiAVquF2+1GcXExfv3rX+Po0aM4cOAAenp60NHRQQULfvrTnyI5OZmqBBKD2W63hxyrBQsW4Omnn4bH48Hf/vY3HDlyJKRhqtPp4HQ6kZCQgNjYWMhkMpw7d+6yHBZirMfHx0MgEITcGxwOBwQCAVQqFW688UbcfvvtmDdvHkQiEU6dOoWOjg4olUqIxWK6hhCnnxkEILUZLpcLsbGxkEgkNBtD+hwRVoLX64VcLkdRURHmzZuHiooKbN++He3t7YiJiaEKemKxGAkJCbTHh0AgoI0JORwOVXSUSCRUHnxkZATJyclYtWoVzGYzhoaGIBKJaPaGKB4uXboUHo8HFRUVqKqqoqIqIpGIBowmo/eRjL5AIEBKSgoGBwfxxz/+kTqqJKggl8sxa9YsOtesVuukth3pcJ+cnAxg3M4wGo1wOp30OkJlKNxuNxQKBQoKCqDRaFBbW4vTp09j2rRpuOOOO5CRkQG1Wo26ujo4nU5oNBrU19ejo6ODypmT8w8MDMBkMqGmpgaNjY1YtGgR3nnnHezatQvl5eUwGAyw2+00eBUTE4PU1FSkpKQgIyMD+fn5SE1NhU6nQ3t7O+2hNzo6ivb2dtqo9vvCNXdaCG/65ZdfDuB6czgc5Obm4sknn0RbW9tV9RjxeDyX1VQxIiICmZmZAMY3GmJsXy1I59zVq1dj4cKFsNls2Lt3L/785z9TSsDVgCgOhQJRqyktLcWKFStQVlaG999//4oaZ3q9XthsNho5VKvVUCgUMBgMSEhIgEAggFAoxOrVq/Hcc8/RqPixY8fwwgsvXPH9kegTMXIGBwcD7lcsFmPx4sUoLS2FUChEc3PzNVMPS0pKQkREBFwuFyoqKrBp0yYAwI4dO2h260pllkk3ZpvNhsTERNx0002UdywUCiEWi1FSUoK4uDhkZGSgu7sb7733Hnbu3In29vZrcn/XO5hStsH1LH6/H3PnzkVMTAyVchUKhXA4HBgZGUFubi5++ctfoq2tjRaKhzIsgp0CQo0gm3ZnZye6urpCRteDjUliIBYXFyMmJoby5kkxN5/Ph1AoDIi4hsomke8xpYM5HA7y8vKwZs0alJaWwuFwYMeOHfjHP/6B5uZmWhDMzEaFAvOaSZE74WuTuS6RSDA0NESVppRKJYqLi7Fq1SrMnz8f5eXlePvtt9HR0REgHsCMwjKdMOZYE3Uel8sFPp9Pm2SaTCbweDyEh4fTPjRE6ZGo6e3du5dmDphOQnDWJdi59fnGZcpJdLO/v586n+RZKJVKLFu2DHfffTeEQiHKy8vR1tYWMjt2MYeFzA3mvI2Li0NWVhbsdjsOHTpE5+u+fftQXFxMOf/BY0fGNNQeQX4mEokQGRkJHo+HadOmUUORrMexsbHIzMxERkYGEhMTYTAY8P777+PAgQPo7u6eoAYUXEf2Qwah8zAbfBIuv0KhAJ/Ph8FggE6ng1KpxAsvvIDq6mo8+uijE4KVjY2NNDpPjPLBwUHavyhUAC88PBwxMTGYMmUKzd41NDRgcHAQRqMx4N0gPXoMBgNaW1uxe/fuCffBVM4iINRHUjDN4/FgNpspdZC8C3w+n74Dt956K0pLS5GdnQ2Hw4GKigocO3YMZrMZSUlJNMNqMplozRs5Lylc5/F4tO6BrD0DAwNoaWmB3W4Hn89HVFQU0tLSsHTpUqSmpuLEiRP4+uuvYTAYIJFIEB0dTR0Ht9sNHo9Hs9IcDoeKzAwNDVEnjNSIaTQadHZ2UtplUVER7atCak8FAgESEhLg8/lw+PBh1NTUwGw2Qy6XIzIyEnK5nK6zxNEHxtdx8jzJWurxeKijRq41MjKSBp1mz56Nm2++mQoXnD59+qJUb6vVSmtXPR4PbeipUqng948rJgav42SdTk9Ph1arxeHDh1FVVQWPx4NZs2ZBpVJBKpXSPirnz5/H6dOnMTQ0hJycHFoDQ56lWq3GyMgIvvjiC/rsb7vtNqxduxZ8Ph9DQ0N0zolEIojFYvD5fDo3ysrKUF9fj8bGRni9XloXVVFR8b06LMC/wGkhSgokksXlciESiVBSUoInnngCIpEI//3f/43BwcGrPhdZpEgRPhMCgQASiQRTp05FXl4efvGLXwAAbd53tbQt4qE+8sgjePrpp9HV1YVdu3Zhw4YNV+Q4hAJRGlm+fDl+/etf0xedw+EgPT0dt912G4aGhvDGG2/gyy+/vOLCbavVivb2dmoQrlu3DpmZmWhra8OCBQtokd+KFStoHUtZWRnefPPNq7o/tVqNl19+GTKZDBcuXMCuXbuonCCfz8fs2bPx5z//mRo3H3zwwVVlkQg4HA7S0tIgEolgs9mo4cvlcmEwGKgM65Wip6cHhw8fxi233IJXX30VJpMJRqMRsbGxEIvF6OjooDKV+/fvx9atW68ZVfGHAuamHlxwDQApKSngcrnUiOBwxjtQ5+fn47HHHoNEIsGrr74asmaMadgGK0IFZzqI0SAWi+FwOGCxWCj1h8hmKpVKpKSk0EZnSqWSFtCTzsfkHJMpTzGvgxlhJzSU++67D6tWrYLdbsf27dvx2Wef4fz589RIJZs/8x6D6xMmM7hdLheOHDmC/Px8PP/88zh16hTCwsLA4XCQmpqKkpIS+P1+bNy4EZs2bcLg4OCkBu5kmQFgPNpL6gXUajVKS0sRHR0No9EIqVSKpKQk5OfnY+rUqRCLxRgdHcWRI0fw7rvvorOzkz6Ti717wVLBpEnnz3/+c2RkZKClpQUVFRXo6OiAXq+HQCDArFmz8MADD2DKlCnQarXYtWvXd86yB18LcRzIOIjFYkRERMBut6OlpYX2gBkaGqKBteBMByl+nsxhIRmm7u5ulJeXY/ny5XjiiSfgcDgozQMYl0Lt7OzE2bNnsWnTJtTW1uLChQuTNrAk52dKLv9QQYKPRP6byPVyOBy4XC5aeO5yuXD//fdjzpw5+N3vfgej0Uh7qADfNic1Go3o6+ujjIyenh6YTCY4nc6QQhuVlZUoLy+HWq3GTTfdhJSUFNpfqbq6Gt3d3RgaGqI1XKQvSn9/f0Bgis/nIz4+ntbSMBEVFQWpVAqv10sDNmRdIllXLpeLqKgozJkzB/fccw8WLFgAlUoFnU5HVQE1Gg3kcjnkcjmt3yPF66R+iowpAFrb1tjYiIqKChqB12g0NCsxffp0ZGRkQKvVYvPmzTh06BDCw8ORnJyMlJQUJCcn0+aFZL4JhUJa80XOp9PpKA2KFK+3t7dj+/bt8Hg8SEpKQnR0NNLT0+l12Ww2mM1mdHZ2or6+HpWVlTAajbRLu1gsps4+eZ7EaUlISIBOpwtwOnw+H/r7+2mGJT4+Hvn5+bBYLJDL5bjjjjuQmZkJPp8Po9GIQ4cOXbRGxm63U0fMYDDA4XBQCqLVaoXNZqPUXOZxBAIB8vLy0N/fT1XgNBoNBgcHkZ+fT9ceu92OwcFB9PT00Ewzc13m8XiIiYmBTCaDz+fDjh076M+SkpJoY2uXy0WdcrvdjpGREQwMDECr1WJkZAQ6nQ4ejwdKpRIRERGwWCwT2l98H7jmToter4fL5UJxcTFycnIQHh6OKVOmYP78+eBwOPjjH/+I/fv3X/V5SOObOXPmwGg0ora2Fr29vZDL5YiPj6dRkHnz5iE1NZVyqhsaGvDmm29etdMikUhw33334eabb8bAwADWr1+Pjz766JrTeo4cOYINGzbg3nvvxVNPPQVgfOOsqqrCN998gy+//BJVVVVXxS90Op1obGxEbW0tFAoFpk+fjszMTDgcDhrhAb4VIdi7dy8++OCDq8p6EJ777bffTu9p9uzZaGxshMlkglAoREFBAeLj4+Hz+bBx40Yq8Xm18Pv9qK+vx9jYGF1IhUIhUlNTsWzZMmRlZdEO9VeC8+fP43//939x8uRJpKWlTUhPE230H2vjyO8KZpSPacT7fOP9Sux2O5YtW4bY2FioVCqq3uV0OvGnP/0JBw8epLUrQOgu9ARMuhgql+jnAAAXNUlEQVSJUrpcLkRGRmLJkiWYMWMGhoeH0dTUBJPJRCkmSUlJtHM9SaG7XC6cPn0a+/bto83lgp0KAmaxNdNhIQ5TTEwM7rzzTtx1113weDzYvHkz/vnPf6K9vZ3KHQf3o7kYpYg5DkSG2ev1YteuXUhLS8Pdd9+NO+64A3K5HE6nE/X19bQHQmNjY4CBFmocJ+OGk0LgpqYm1NTUYNmyZZg/fz5yc3PhcDggEomgUCggEongdrvR2dmJAwcO4OOPP0ZdXd2kRkCwcc+kOBGDNS0tDSUlJSgqKsLs2bNxww03oKurizpLU6dORUFBAUZHR/H5559j3759AVkdcg9MUYHvUk9DMtRkvsjlciQkJCA8PBzz589HWloaqqurKef8Ulmd4HFub2/HBx98gIqKCiQmJkIoFCI8PBw8Hg+jo6OUyqvRaCgNjxyb+fxCUSd/DAiO+BLj3uFwwGq1wmw2Q6FQ4L777oPdbkdnZyfNagCgcrEzZ85EUlISdu3aRY9FalImm5c6nQ7Hjh1DRkYG5syZQ5sazp8/nzZ6bGpqgkajgd/vR2FhIXJyclBbWxtguHu9XgwMDIRcOyQSCSIiIqDT6WAwGOicJTVNwLgc9pIlS/DQQw9h5syZiIiIQH9/P7Zt24YvvvgCHR0dkEqlSElJofsQsX0INYsY9qRg3Ocb791C6jrWrVuHNWvWYMaMGXRN6e/vx5dffonjx4/D7XYjPT0dCQkJyM3NpQ4LANqbymQywWQyQa/XUxUzcv/E8Cbr18jICL7++mucO3cOCoUCUVFRSEpKop3hx8bGoNFo0N3djcHBQYhEIkRHR0MsFtNscmxsLHXMSL0jMF6PIhaL0dTUFPAekPEXi8WYP38+VqxYAbPZjNjYWCxYsAAKhQI6nQ579+5FRUXFRemVZC3Q6/UYGRmhWVEyJqR/FLkXr9dLG4GqVCpwuVwUFBSgubmZsjaEQiE0Gg3dtwYHB2E2m+nvkWfqdDrR0dFBRR/i4uJojxi9Xo/W1lbodDrodDra8NTj8VBabTAiIiLg8/lgtVoxOjpK96GoqKjvjcJ+zZ2WsrIylJSU4NZbb0VBQQEGBgaoxvfx48cvu4nSZFAqlXjsscewcOFC2O129PT0oL+/HzKZjHJHmZ0/SefrN954A8eOHbuqc0dGRuKxxx7DI488gqioKGzduhVbtmz5lzzE0dFRvPbaa6iurkZqairlbdfW1qK5ufmyi/kmQ3t7O1577TXodDrcdtttUCqVEIlE9HOLxYJDhw6hvLwcmzdvviZyvAaDAfv376fOSW5uLnJzc6kxwuFw0Nvbi7KyMrz++uvXpEaIwGQywe12IzIyEnfccQdVIpk1axYiIiJQWVl5xffo8/mg0Wiwbdu2a3a9P3YwjUW/349Tp05h6tSpWL58OZWNHhoawpEjR7Bnzx7U1tYGFBaHMuQno/sA3xZYx8fH44477sDNN98Mi8WCnp4eGAwGuhFGRkZCJpNRap/FYsGRI0fw0UcfoaKigm62wbUmzHsJdqTId2NiYnD33XfjkUcegVQqxUcffYSNGzeiq6uLOjdMoz1UBikUmE4SiXTqdDp8+OGHqKurQ3R0NGQyGRwOB2pqatDS0kI7NRMnK/jczGMy75HcJ/m8ra0Nn376KcbGxlBUVISEhAQolUp4vV6MjY2hu7sb7e3tKCsrw4EDB9DR0TGB1sY816UMbFIAW19fTykYxcXFKCoqotFdohb3zTffYNOmTTQ7Fkz1CpVVYjqYTEoj+V2dTofh4WFkZ2djyZIlEIvFUKvVmDdvHng8HiorKzE6OhqQ3Qh2xIKzZczvnT9/ns4HJr2RWRzOfA6h5gazHudSPYx+aCBCNESamCmqwOFwkJGRgaSkJAwMDNBgiNVqRXR0NJYvX47S0lLExcWhrKyMNjwMpiiGgtvtRnl5OXQ6HRYsWICsrCzIZDLEx8djypQpmDZtGm655RYaBBAIBOjs7ERra2tAYMDn8wXU2zJBouAkWg+MP0vivKSkpGDlypX4xS9+gYKCAvh8Ppw7dw6fffYZtm3bhr6+PrjdblozYrFYaJ0XcUxI7RkwPt+Y1+J0OnHixAlkZWVh+fLlkMlktO9eVVUVHA4HoqKiEBsbi5SUFKSkpCA1NRUxMTG0VoY0WhwbG8Pw8DDa29vpPs7hcCgtKZjKSDq+czgcmg2LiIigohhEeIDD4aCwsJDWzZGaRLFYDLFYTJXjCGbMmAGNRoOOjo6Qqm1hYWGIj49HUlISwsLCIJFIMDAwgLKyMhw5cgQnT568qMqhUChEXl4eoqKi0N3dDZPJhMjISISFhVGKbFRUFKXrkpophUJBHY709HQsXryYNq/MzMzE+fPnUVlZSefL0NAQXC4XZRoRW8Xj8WBwcJBmDzs7O2kjZSatV6FQQKlUUkqvwWCgDhQwvt7LZDJae0uyW263GyKRiAoBfB+1cdfcaWlqasJLL72E2NhYREVFYWRkBFqtFt3d3VfdUJIJZiRRKpWioKAABQUFIb/rdDqxc+dOvPbaawFqKlcCojr1zDPPQC6X48SJE3jzzTcnleW7FjCZTNi5c+e/7PjAOEXsxIkT0Gg02L17N8RiccDnTqcTTU1NtIj1auH3+zEwMIDf/va3yMjIQEFBAaZPn46FCxfCaDSit7cXLS0tlNvJlB+8FiCOhVqtxrRp02j6t6+vDxs3bsSOHTuu6XxlMRFMY5y5qAJAbW0tXnnlFSojazab0d/fjwsXLtCuwkyj9rtEj8l3mEXeRFXKYrFAoVDQehVCOXG73bDZbNDr9VQpb+vWraiurqaZOuBbNZzJjHsC4uAolUrcfvvteOaZZ6BSqbBlyxZ8+umnVPiBWdQPgPKrg2WiL0VHI3QSAOjr68Pw8DDlxJPfCTZ4g9XOLpXZIVkmLpcLk8mEw4cPo6WlBSkpKYiPj6fcdtJjoKenB319fVQNjmmoX0z7n4wtqbUhBnhjYyP+9re/Yf/+/UhPT0dhYSGysrLA5XIxNDSElpYWnDlzBpWVlejr66PO3KUMU3Lv5LkynZawsDC4XC5qGBQUFGDOnDnIz8+HUCiETqfDpk2bcODAAVgsFvB4vAnPlIw706Fh0v6Yji9Tqjb4uQc7zMEg52De048h28LhcBAZGUn7hNhsNng8HkgkEigUCmqAEqWjoqIiyOVyFBQU4IYbbkB+fj6ioqJQXl6Oo0eP0vdZLBbDbrdflLtP6jBPnjyJ5uZmKBQKiMViZGRkYNGiRfRcZI709vbi4MGDOHjw4HeuCTAajbTeggmfz4fMzEzcf//9ePDBBzFlyhSq+vnxxx9jz549GBkZoc/YYrFAo9HAbDZDJpPRYKTFYplAlQp1DV9//TV6e3vB4/EwNDQErVYLuVxOC/WlUimt1SPOBZ/Ph9PphNfrhdFoRFdXF5qamqiaFYFUKqUy7cHzMiwsDBEREXA4HDCbzSEp1GQ97erqwuDgIFULy8vLo7U+THWtoqIixMXFoaqqKqTIjs1mw+nTp6HX62n/KELDvHDhwkUZLUTpa8WKFXC5XNi1axelYZH3Ozw8nKquMUUynE4nzGYzNm/ejBdffBH5+fk0WGa1WvHOO++gvr4ecXFx6OjowODgIFWN02g0AY4ZaQ5J1hCDwUD7FpGaHeI8paenIzExESMjI3SOhIWFUUc0OjoaTqeTNigVCoVITExEWlraVdWlXw2uudPi8Xiuum/Hd4FOp8OLL76IrVu3YurUqZg5cyZiY2OhVqvR1dWFrq4udHR0oK6uDg6HA+fOnbsiJahg3HjjjVi3bh1kMhkOHz6M//f//t81q2G5HtDd3Y3u7u5/y7kILa25uRlHjhxBdHQ0EhMTaXSGFFL+K/jXTqcTv/3tb/Hss8+itLSU1uns3LkTBw8e/FE3drweMFl0HQCNvDU1NaG5uXlC5DOY/hJ83FCODDOqTj7zer24cOEC/vGPf6CiogJZWVnIyspCQkIC7WBNMrh6vR59fX2oq6tDX18fjXYyxTJCRc+ZsrvEuOTxeCgsLMQjjzyCpKQknDhxAp988gk6Ojrg9/sn1DxM5rBMBqYBEKp2gnn8UAZsKGcllFHMjPgznQmz2Uw558wiWHL/hINNHBai8EOcgskK8cn1MCmBAGiX7qamJqpUpFar4Xa7MTw8jNHRUZhMJkrJIUZScPAquI7nUpke4vB+/vnnCAsLQ15eHrhcLk6ePIkdO3bg6NGj0Gq1NGvFrKdijjMT5D0gctl+vz9ALQ0AHTdyTHLcyZ4V816DM0Y/ZJCAgV6vp53IiYOmVquRlpaGtLQ0mEwmZGRk4Fe/+hW8Xi+ioqIQGRmJEydO4J///CelUQGgtTGXoluLxWKarRweHqbdynt7e9Hf34+TJ08iPDycOgZ9fX3QarWXVRMwmUyuWq3G/fffj0cffRRqtRp2ux0HDx7Ehg0bqPPFBGmaSzq2EyeISERP5vASBS2dToeGhgbacJMU1DNrLJxOJz0OqZMxGAzo7e3F2bNnUVVVhaGhoQnzzu/308BKdHQ0OBwOnE4nxGIxlU93Op0YGRnB6OgorFYrxGIxoqKiqDKbzWZDb28vlWdnKhfa7fYAYRU+nw+JRAKlUhnSHrTb7VR1i7w3Docj4LoJtT1Y9lcgECAzMxOzZ89GV1cX/TmhkMrlctrQk+xphNZpNBrh9/uxfft2FBcXo6SkBGlpaejt7cWWLVuoeuzw8DClQALj2bgLFy4E0GYjIyNpTx0mjVYmk8Fut0Or1WJsbAx2u53SI8fGxlBXV4eRkREkJiaioKAAcXFx0Ov1aGpqwtjYGKRSKdLT0zFnzhxMnToVe/bs+V7WkX9Jc8l/BxwOB6qrq9HQ0ACpVEonBNHFttvtsFqtVN3nWgwuh8NBUVER1Go1hoeHsWHDhmsmwft/GSQtbTKZrolj+V3g8/lw/PhxdHd346WXXoLb7YZer6cpZhb/WgRnCIIj/cxIcygjjOnohMrUXCyazPy9sbExnDlzBg0NDVR1RiwWQygUwuVyYWRkhFIJSeEiMYzIBhvKCA2m+zA/EwgEKCgoQGZmJnp7e/Hxxx+jpqYGHo9nQm8XZh1FcGaKaVQTGkUwfYk4B0StjPQ9YNLAmOM+WRYnFMgYMr9LpKJD0Z3I+cizYn6PSHYGr9PMLAT5LtMAJ+cnvGy73Q6DwYCWlhb4fD7qnBHZVsLdDkV7myyjRM7JHA9CZ/F6vdizZw8qKyuhVCqpZO7AwACcTiftbE/mADEEmdfAvGfyzMizZRZMM/utTObMTWaAhsrI/dDhdrthMpkm3A8pGE5PT0dBQQEiIyMpzaW/vx/r16/HqVOn0NXVBY1GA4PBQCV3/X4/rR+5GIhSFXFcPB4PoqKiaHPntrY2qsREFL9CgcvlQqVSQaVSUSldu90Ol8s16fs3Z84c3HvvvVCr1XA6ndixYwc+/PBDVFZWTtqokji/hD7l9/tDOixkLhEjVyAQUKfBZrPRRpLh4eGUjkeO5/F4aA8Zi8WCrq4u1NXVobGxEaOjoyHXcb1eT4+bl5cHqVRK61Li4uIgl8upfTA2Nga3203HnWROCdWNwO12o7u7m75/arWa3ldHRwe0Wu2kgj5+/3izz4s1/MzIyEBcXBy6u7sDGCB8Ph+xsbGIiYkJEGjxer0YHh6GRCKhindOp5MquJHvAOM0/b/+9a90DHp6eqDVaqFWq2EwGKDX6yeskcx75/P5dA8jfWpIZtnr9UImkyE3NxcXLlygPYqKiorQ29uLmpoahIWFIS0tDSqVCp2dnThx4gR6enpoXWJMTAwMBkNAXdK/Gz9YpwX41ksnRXP/DnR3d8NqtdJC4B9Dmv3/KrxeL3p7e//PF8R/X2Aab8yof/DmRgw95mehiqiBQGchlDwv2bx4PB4EAgHtKu5yuSiNIjjDEJyxIXUiTMOCXGNwhDZU9gUY32gOHDiAEydOYP/+/dSouZTjQAx18ofp0DAzkkwjm1lkGcpgDZYVDgZxiCbLThCQDfJSAaJQ52FuvJNdD/P/TOeMPGeS7SHXQhrnEcpI8BwhCkyTXRO5P/IZk0pHzk9kWnt6eiiFhpkVC36e5DovVp9Ejs+c9wAopY1ko4J/P3iuMZ1TZqbvx7JnMevTgG/nhM/no31bCLd/7969qKysxOHDh9HX1webzQafzwelUonFixdDJpOhr68PbW1tVPELAJUND4bb7UZfXx+dD1FRUZg+fTri4+PhdrupCiXJak7mhPB4PKSkpCA7Oxsmkwk9PT3QaDST2jN8Ph/JyckQCASUyr1hwwbU19fD7XbTfkVMYQYmhEIhpTYGrwWENuTzjcvgkowKicoTWWCSGSXvHlm7vV4vpXGRrHRLSwvMZvOkc444DyqVCjk5OYiOjqaOEKGZ+f1+yOVyWrcEjK8XZ8+eBY/HC2g4TWiYvb29VJVMLBbTdaGhoQGtra1UQCUYAoEACoWC0vqA8ebWQqGQ0vVITWDwWk9kmIkKKROkcz2hMLpcrpBsDqfTifLycjQ3N0OlUiErK4uqvtlstkuurQKBAFKpFBqNBjKZjKqIkTlDxEIGBgZw6tQpaj/X1dXhzJkzEIvFiIuLw/nz51FfXw+TyRRwTpPJBJ1Oh8jISNZp+SHA7/dj3759yMjIuGj0hAULFpcGczEMVjgCEODEXMzYnizDcqmfMx0ZgUAQYPwSo5KZdidOCTMafrHrZ4Io/rhcLjidTuzduxdlZWUTqGahHDay4RLDOxR9K9S4ABOpQSRyH+zEBRu8zIxPsEPEBKF1EdpXqO8Fq5qR3yOyy8SAJCCGerCzQa6N+WyCzxecyWHeA5FVJc8wVB+dUFk75rGJEct0islxmNcd/BxIx2u73R4wBsQ5CX4XSMGx0+mkxiGzyJwcg+kkkd9lXjuh8DB/Hmqe/dDA5XKRnZ1NDW2Xy4WBgQFaVG6xWDA4OEjVACsrK3Hu3DmaKQXG++DMmjULBQUFlOpHKJoAaH0Ls0CZCeaYh4eHg8PhUNEFiURCn5PFYoHJZJp0fSBGN+lVNllhPjlnW1sbtmzZAq1WiwMHDlChBuKkk4xqMEQiEVUkA74NGgWDHIc4g0KhkM53MtdIEbzH46EMFz6fT43zhoYGnD9/HhaLZdJ7AcYzVnw+HwqFAtHR0dRpIeukx+NBREQEoqOjoVKpaNaFOadJXxWLxYKEhATaVoCMATPTrdfr0dnZGfA85XI57VkSHh6OnJwccLlcNDQ0wGAwwOv10sySx+NBb28vzGYznWvA+LtM5NyJNDITfr+f0lSJ4zfZfPD7/UhMTMSiRYsgl8vR09MzwWHJyMhAWFgYNBrNhMbpaWlpsNls6O/vh1gsRn5+PmJjY6mSb2ZmJiorK7Fjxw5oNBrweDwcP34cdrsdXq8XLS0tMBqNE4SPiGqZz+ejLU2+D3D839FduljqnAULFv9e/FCjpew6Mo6L1eWwYPHvwg9x/jEpTEwEZ+eYmYDJAh/kOwACnPkruSZyHOa7zfwzGYiTOtk1TnbNJCBxOc/wYhm+yb73XT4LRTW9HKOWBIVCXR8za0h+Tu49WKAiVNCBmWkMrm0MheDMHTlucLAoePyIeAsJbl1NGwoul0uzSszgC/N6JguuMGsWg8eVKShCstuTCYRMhuBz/7vXENZpYcHiB4gforEBsOsICxbXE36I6wi7hrBgcf3gunVaWLBgwYIFCxYsWLBgweL7wOSi+CxYsGDBggULFixYsGBxHYB1WliwYMGCBQsWLFiwYHFdg3VaWLBgwYIFCxYsWLBgcV2DdVpYsGDBggULFixYsGBxXYN1WliwYMGCBQsWLFiwYHFdg3VaWLBgwYIFCxYsWLBgcV2DdVpYsGDBggULFixYsGBxXYN1WliwYMGCBQsWLFiwYHFdg3VaWLBgwYIFCxYsWLBgcV3j/wPttjYguVZYggAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1000x2000 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from torchvision.utils import save_image, make_grid\n",
    "\n",
    "from experiments.data import BatchSiren, INRDataset\n",
    "\n",
    "dataset = INRDataset(\n",
    "    dataset_dir=dataset_dir,\n",
    "    split=\"train\",\n",
    "    normalize=False,\n",
    "    augmentation=False,\n",
    "    splits_path=splits_path,\n",
    "    statistics_path=statistics_path,\n",
    ")\n",
    "loader = torch.utils.data.DataLoader(dataset, batch_size=batch_size, shuffle=False)\n",
    "batch = next(iter(loader))\n",
    "\n",
    "dataset_aug = INRDataset(\n",
    "    dataset_dir=dataset_dir,\n",
    "    split=\"train\",\n",
    "    normalize=False,\n",
    "    augmentation=True,\n",
    "    splits_path=splits_path,\n",
    "    statistics_path=statistics_path,\n",
    ")\n",
    "loader_aug = torch.utils.data.DataLoader(\n",
    "    dataset_aug, batch_size=batch_size, shuffle=False\n",
    ")\n",
    "batch_aug = next(iter(loader_aug))\n",
    "\n",
    "inr_model = BatchSiren(2, 1, img_shape=img_shape)\n",
    "out = inr_model(batch.weights, batch.biases)\n",
    "out = out.transpose(1, 2).unflatten(2, img_shape)\n",
    "\n",
    "out_aug = inr_model(batch_aug.weights, batch_aug.biases)\n",
    "out_aug = out_aug.transpose(1, 2).unflatten(2, img_shape)\n",
    "\n",
    "indices = [\n",
    "    int(p.parts[-3].split(\"_\")[-1]) for p in dataset.dataset[\"path\"][:batch_size]\n",
    "]\n",
    "orig_images = [orig_mnist[idx][0] for idx in indices]\n",
    "\n",
    "fig, ax = plt.subplots(1, 3, figsize=(10, 20))\n",
    "\n",
    "ax[0].imshow(make_grid(orig_images).permute(1, 2, 0).clip(0, 1))\n",
    "ax[0].set_title(\"Original Images\")\n",
    "ax[0].set_axis_off()\n",
    "\n",
    "ax[1].imshow(make_grid(out).permute(1, 2, 0).clip(0, 1))\n",
    "ax[1].set_title(\"INR reconstructions\")\n",
    "ax[1].set_axis_off()\n",
    "\n",
    "ax[2].imshow(make_grid(out_aug).permute(1, 2, 0).clip(0, 1))\n",
    "ax[2].set_title(\"Augmented INR reconstructions\")\n",
    "ax[2].set_axis_off()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "95c99851",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_set = INRDataset(\n",
    "    dataset_dir=dataset_dir,\n",
    "    split=\"train\",\n",
    "    normalize=False,\n",
    "    augmentation=True,\n",
    "    splits_path=splits_path,\n",
    "    statistics_path=statistics_path,\n",
    ")\n",
    "val_set = INRDataset(\n",
    "    dataset_dir=dataset_dir,\n",
    "    split=\"val\",\n",
    "    normalize=False,\n",
    "    splits_path=splits_path,\n",
    "    statistics_path=statistics_path,\n",
    ")\n",
    "\n",
    "test_set = INRDataset(\n",
    "    dataset_dir=dataset_dir,\n",
    "    split=\"test\",\n",
    "    normalize=False,\n",
    "    splits_path=splits_path,\n",
    "    statistics_path=statistics_path,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e5ad1393",
   "metadata": {
    "id": "e5ad1393",
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "---\n",
    "\n",
    "### Create Dataloaders"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "6aef0a69",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "6aef0a69",
    "outputId": "fa4aa4ef-e7e0-4739-d7d5-8409fe936b21",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-02-22 20:37:03,779 - root - INFO - train size 55000, val size 5000, test size 10000\n"
     ]
    }
   ],
   "source": [
    "batch_size = 64\n",
    "num_workers = 8\n",
    "\n",
    "train_loader = torch.utils.data.DataLoader(\n",
    "    dataset=train_set,\n",
    "    batch_size=batch_size,\n",
    "    shuffle=True,\n",
    "    num_workers=num_workers,\n",
    "    pin_memory=True,\n",
    ")\n",
    "val_loader = torch.utils.data.DataLoader(\n",
    "    dataset=val_set,\n",
    "    batch_size=batch_size,\n",
    "    num_workers=num_workers,\n",
    "    shuffle=False,\n",
    "    pin_memory=True,\n",
    ")\n",
    "test_loader = torch.utils.data.DataLoader(\n",
    "    dataset=test_set,\n",
    "    batch_size=batch_size,\n",
    "    shuffle=False,\n",
    "    num_workers=num_workers,\n",
    "    pin_memory=True,\n",
    ")\n",
    "\n",
    "logging.info(\n",
    "    f\"train size {len(train_set)}, \"\n",
    "    f\"val size {len(val_set)}, \"\n",
    "    f\"test size {len(test_set)}\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "278aef59",
   "metadata": {
    "id": "278aef59",
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Let's look at the input INRs' shapes: Weight matrices are of shapes (in_dim, out_dim, channels), and biases are of shape (out_dim, channels)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "cc3db169",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cc3db169",
    "outputId": "08848651-0c23-4911-b5eb-b755e74b89c3",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['weights', 'biases', 'label']\n",
      "Weights: [torch.Size([2, 32, 1]), torch.Size([32, 32, 1]), torch.Size([32, 1, 1])]\n",
      "Biases: [torch.Size([32, 1]), torch.Size([32, 1]), torch.Size([1, 1])]\n"
     ]
    }
   ],
   "source": [
    "point = train_set.__getitem__(0)\n",
    "print(list(point.as_dict().keys()))\n",
    "\n",
    "# weights\n",
    "print(\"Weights:\", [w.shape for w in point.weights])\n",
    "\n",
    "# biases\n",
    "print(\"Biases:\", [b.shape for b in point.biases])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d130ca34",
   "metadata": {
    "id": "d130ca34",
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "---\n",
    "\n",
    "## Initialize Neural Graphs Transformer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e47ed3d5",
   "metadata": {
    "id": "e47ed3d5",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-02-22 20:37:06,369 - root - INFO - device = cuda\n"
     ]
    }
   ],
   "source": [
    "# get device\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "logging.info(f\"device = {device}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "cd3e7642-0460-43f5-9234-dd21f2bd87f3",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-02-22 20:37:08,044 - root - INFO - weight shapes: (torch.Size([2, 32]), torch.Size([32, 32]), torch.Size([32, 1])), bias shapes: (torch.Size([32]), torch.Size([32]), torch.Size([1]))\n"
     ]
    }
   ],
   "source": [
    "# Compute the layer layout\n",
    "point = train_set[0]\n",
    "weight_shapes = tuple(w.shape[:2] for w in point.weights)\n",
    "bias_shapes = tuple(b.shape[:1] for b in point.biases)\n",
    "\n",
    "layer_layout = [weight_shapes[0][0]] + [b[0] for b in bias_shapes]\n",
    "\n",
    "logging.info(f\"weight shapes: {weight_shapes}, bias shapes: {bias_shapes}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "de3e700b-63ad-4cc7-aff4-61ba1bf0e440",
   "metadata": {},
   "outputs": [],
   "source": [
    "stats = {\n",
    "    \"weights_mean\": [\n",
    "        -4.215954686515033e-05,\n",
    "        -7.55547659991862e-07,\n",
    "        7.886120874900371e-05,\n",
    "    ],\n",
    "    \"weights_std\": [0.06281130015850067, 0.018268151208758354, 0.11791174858808517],\n",
    "    \"biases_mean\": [\n",
    "        5.419965418695938e-06,\n",
    "        3.7173406326473923e-06,\n",
    "        -0.01239530649036169,\n",
    "    ],\n",
    "    \"biases_std\": [0.021334609016776085, 0.011004417203366756, 0.09989194571971893],\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "716b620c",
   "metadata": {
    "id": "716b620c",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# inr_model = INRPerLayer(n_layers=3, up_scale=16, out_channels=1)\n",
    "\n",
    "inr_model = None\n",
    "d_node = 64\n",
    "d_edge = 32\n",
    "\n",
    "graph_constructor = OmegaConf.create(\n",
    "    {\n",
    "        \"_target_\": \"nn.graph_constructor.GraphConstructor\",\n",
    "        \"_recursive_\": False,\n",
    "        \"_convert_\": \"all\",\n",
    "        \"d_in\": 1,\n",
    "        \"d_edge_in\": 1,\n",
    "        \"zero_out_bias\": False,\n",
    "        \"zero_out_weights\": False,\n",
    "        \"sin_emb\": True,\n",
    "        \"sin_emb_dim\": 128,\n",
    "        \"use_pos_embed\": True,\n",
    "        \"input_layers\": 1,\n",
    "        \"inp_factor\": 1,\n",
    "        \"num_probe_features\": 0,\n",
    "        \"inr_model\": inr_model,\n",
    "        \"stats\": stats,\n",
    "    }\n",
    ")\n",
    "\n",
    "model = RelationalTransformer(\n",
    "    d_node=d_node,\n",
    "    d_edge=d_edge,\n",
    "    d_attn_hid=128,\n",
    "    d_node_hid=128,\n",
    "    d_edge_hid=64,\n",
    "    d_out_hid=128,\n",
    "    d_out=10,\n",
    "    n_layers=4,\n",
    "    n_heads=8,\n",
    "    layer_layout=layer_layout,\n",
    "    graph_constructor=graph_constructor,\n",
    "    dropout=0.0,\n",
    "    use_cls_token=False,\n",
    "    pooling_method=\"cat\",\n",
    "    pooling_layer_idx=\"last\",\n",
    ").to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "cb0530a6",
   "metadata": {
    "id": "cb0530a6",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-02-22 20:31:04,077 - root - INFO - number of parameters: 376586\n"
     ]
    }
   ],
   "source": [
    "logging.info(f\"number of parameters: {count_parameters(model)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e429cc7",
   "metadata": {
    "id": "6e429cc7",
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Eval Function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "caa7571a",
   "metadata": {
    "id": "caa7571a",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "@torch.no_grad()\n",
    "def evaluate(model, loader, device, num_batches=None):\n",
    "    model.eval()\n",
    "    loss = 0.0\n",
    "    correct = 0.0\n",
    "    total = 0.0\n",
    "    predicted, gt = [], []\n",
    "    for i, batch in enumerate(loader):\n",
    "        if num_batches is not None and i >= num_batches:\n",
    "            break\n",
    "        batch = batch.to(device)\n",
    "        inputs = (batch.weights, batch.biases)\n",
    "        out = model(inputs)\n",
    "        loss += F.cross_entropy(out, batch.label, reduction=\"sum\")\n",
    "        total += len(batch.label)\n",
    "        pred = out.argmax(1)\n",
    "        correct += pred.eq(batch.label).sum()\n",
    "        predicted.extend(pred.cpu().numpy().tolist())\n",
    "        gt.extend(batch.label.cpu().numpy().tolist())\n",
    "\n",
    "    model.train()\n",
    "    avg_loss = loss / total\n",
    "    avg_acc = correct / total\n",
    "\n",
    "    return dict(avg_loss=avg_loss, avg_acc=avg_acc, predicted=predicted, gt=gt)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a2e1421",
   "metadata": {
    "id": "5a2e1421",
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Model training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "e0a38d25",
   "metadata": {
    "id": "e0a38d25",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "parameters = [p for p in model.parameters() if p.requires_grad]\n",
    "optimizer = torch.optim.AdamW(\n",
    "    params=parameters, lr=1e-3, amsgrad=True, weight_decay=5e-4, fused=False\n",
    ")\n",
    "scheduler = WarmupLRScheduler(warmup_steps=1000, optimizer=optimizer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "a5af6b6d",
   "metadata": {
    "id": "a5af6b6d",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "epochs = 50  # doing just 50 epochs here\n",
    "eval_every = 500\n",
    "\n",
    "global_step = 0\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "\n",
    "val_acc = -1\n",
    "val_loss = float(\"inf\")\n",
    "best_val_acc = -1\n",
    "best_val_loss = float(\"inf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "ab237a0a-7475-4f1f-95b6-d2a2666a7acc",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[49 860], train loss: 0.046, val loss: 0.268, val acc: 0.914, best val loss: 0.250, best val acc: 0.921: 100%|██████████████████████████████████████████████| 50/50 [2:29:41<00:00, 179.62s/it]\n"
     ]
    }
   ],
   "source": [
    "epoch_iter = trange(epochs)\n",
    "for epoch in epoch_iter:\n",
    "    for i, batch in enumerate(train_loader):\n",
    "        model.train()\n",
    "        optimizer.zero_grad()\n",
    "\n",
    "        batch = batch.to(device)\n",
    "        inputs = (batch.weights, batch.biases)\n",
    "        label = batch.label\n",
    "\n",
    "        out = model(inputs)\n",
    "        loss = criterion(out, label)\n",
    "        loss.backward()\n",
    "\n",
    "        optimizer.step()\n",
    "\n",
    "        if scheduler is not None:\n",
    "            scheduler.step()\n",
    "\n",
    "        epoch_iter.set_description(\n",
    "            f\"[{epoch} {i+1}], train loss: {loss.item():.3f}, val loss: {val_loss:.3f}, val acc: {val_acc:.3f}, \"\n",
    "            f\"best val loss: {best_val_loss:.3f}, best val acc: {best_val_acc:.3f}\"\n",
    "        )\n",
    "\n",
    "        global_step += 1\n",
    "\n",
    "        if (global_step + 1) % eval_every == 0:\n",
    "            val_loss_dict = evaluate(model, val_loader, device)\n",
    "            val_loss = val_loss_dict[\"avg_loss\"]\n",
    "            val_acc = val_loss_dict[\"avg_acc\"]\n",
    "\n",
    "            best_val_criteria = val_acc >= best_val_acc\n",
    "            if best_val_criteria:\n",
    "                best_val_acc = val_acc\n",
    "                best_val_loss = val_loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "f56ce6b6-00e1-41da-b686-796f3bb433d9",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-02-21 10:04:22,778 - root - INFO - [49 860], train loss: 0.046, test loss: 0.266, test acc: 0.913\n"
     ]
    }
   ],
   "source": [
    "test_loss_dict = evaluate(model, test_loader, device)\n",
    "test_loss = test_loss_dict[\"avg_loss\"]\n",
    "test_acc = test_loss_dict[\"avg_acc\"]\n",
    "\n",
    "logging.info(\n",
    "    f\"[{epoch} {i+1}], train loss: {loss.item():.3f}, test loss: {test_loss:.3f}, test acc: {test_acc:.3f}\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7e88482b-5597-4cab-908a-35c00d34ebf9",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "## Initialize Neural Graphs Graph Neural Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "d71d69ee-25a6-400b-ad39-315534c01dfb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# inr_model = INRPerLayer(n_layers=3, up_scale=16, out_channels=1)\n",
    "inr_model = None\n",
    "\n",
    "graph_constructor = OmegaConf.create(\n",
    "    {\n",
    "        \"_target_\": \"nn.graph_constructor.GraphConstructor\",\n",
    "        \"_recursive_\": False,\n",
    "        \"_convert_\": \"all\",\n",
    "        \"d_in\": 1,\n",
    "        \"d_edge_in\": 1,\n",
    "        \"zero_out_bias\": False,\n",
    "        \"zero_out_weights\": False,\n",
    "        \"sin_emb\": True,\n",
    "        \"sin_emb_dim\": 128,\n",
    "        \"use_pos_embed\": True,\n",
    "        \"input_layers\": 1,\n",
    "        \"inp_factor\": 1,\n",
    "        \"num_probe_features\": 0,\n",
    "        \"inr_model\": inr_model,\n",
    "        \"stats\": stats,\n",
    "    }\n",
    ")\n",
    "\n",
    "gnn_backbone = OmegaConf.create(\n",
    "    {\n",
    "        \"_target_\": \"nn.gnn.PNA\",\n",
    "        \"_convert_\": \"all\",\n",
    "        \"in_channels\": 32,\n",
    "        \"hidden_channels\": 32,\n",
    "        \"out_channels\": 32,\n",
    "        \"num_layers\": 4,\n",
    "        \"aggregators\": [\"mean\", \"min\", \"max\", \"std\"],\n",
    "        \"scalers\": [\"identity\", \"amplification\"],\n",
    "        \"edge_dim\": 32,\n",
    "        \"dropout\": 0.1,\n",
    "        \"norm\": \"layernorm\",\n",
    "        \"act\": \"silu\",\n",
    "        \"deg\": None,\n",
    "        \"update_edge_attr\": True,  # Check with False as well\n",
    "        \"modulate_edges\": True,\n",
    "        \"gating_edges\": False,\n",
    "        \"final_edge_update\": False,\n",
    "    }\n",
    ")\n",
    "\n",
    "gnn_model = GNNForClassification(\n",
    "    d_hid=32,\n",
    "    d_out=10,\n",
    "    graph_constructor=graph_constructor,\n",
    "    gnn_backbone=gnn_backbone,\n",
    "    layer_layout=layer_layout,\n",
    "    rev_edge_features=False,\n",
    "    pooling_method=\"cat\",\n",
    "    pooling_layer_idx=\"last\",\n",
    ").to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "be992707-7a6d-4d75-8da9-f91ad33f2087",
   "metadata": {},
   "outputs": [],
   "source": [
    "logging.info(f\"number of parameters: {count_parameters(model)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "79f47955-dc3f-4bf6-9eb3-22931ad2505a",
   "metadata": {},
   "source": [
    "## Model training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "059d50f2-d4f9-4038-b4b7-d67ff85f219f",
   "metadata": {},
   "outputs": [],
   "source": [
    "gnn_parameters = [p for p in gnn_model.parameters() if p.requires_grad]\n",
    "gnn_optimizer = torch.optim.AdamW(\n",
    "    params=gnn_parameters, lr=1e-3, amsgrad=True, weight_decay=5e-4, fused=False\n",
    ")\n",
    "scheduler = WarmupLRScheduler(warmup_steps=1000, optimizer=gnn_optimizer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "4f093516-cd65-4727-9802-063920ee0488",
   "metadata": {},
   "outputs": [],
   "source": [
    "epochs = 50  # doing just 50 epochs here\n",
    "eval_every = 500\n",
    "\n",
    "global_step = 0\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "\n",
    "val_acc = -1\n",
    "val_loss = float(\"inf\")\n",
    "best_val_acc = -1\n",
    "best_val_loss = float(\"inf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "875074ba-74a9-43dd-8a09-a6ff51563e2a",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[49 860], train loss: 0.640, val loss: 0.493, val acc: 0.828, best val loss: 0.488, best val acc: 0.843: 100%|█████████████████████████████████████████████████| 50/50 [39:30<00:00, 47.41s/it]\n"
     ]
    }
   ],
   "source": [
    "epoch_iter = trange(epochs)\n",
    "for epoch in epoch_iter:\n",
    "    for i, batch in enumerate(train_loader):\n",
    "        gnn_model.train()\n",
    "        gnn_optimizer.zero_grad()\n",
    "\n",
    "        batch = batch.to(device)\n",
    "        inputs = (batch.weights, batch.biases)\n",
    "        label = batch.label\n",
    "\n",
    "        out = gnn_model(inputs)\n",
    "        loss = criterion(out, label)\n",
    "        loss.backward()\n",
    "\n",
    "        gnn_optimizer.step()\n",
    "\n",
    "        if scheduler is not None:\n",
    "            scheduler.step()\n",
    "\n",
    "        epoch_iter.set_description(\n",
    "            f\"[{epoch} {i+1}], train loss: {loss.item():.3f}, val loss: {val_loss:.3f}, val acc: {val_acc:.3f}, \"\n",
    "            f\"best val loss: {best_val_loss:.3f}, best val acc: {best_val_acc:.3f}\"\n",
    "        )\n",
    "\n",
    "        global_step += 1\n",
    "\n",
    "        if (global_step + 1) % eval_every == 0:\n",
    "            val_loss_dict = evaluate(gnn_model, val_loader, device)\n",
    "            val_loss = val_loss_dict[\"avg_loss\"]\n",
    "            val_acc = val_loss_dict[\"avg_acc\"]\n",
    "\n",
    "            best_val_criteria = val_acc >= best_val_acc\n",
    "            if best_val_criteria:\n",
    "                best_val_acc = val_acc\n",
    "                best_val_loss = val_loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7f4926b2-13c7-462c-a46a-397cc9119d31",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_loss_dict = evaluate(model, test_loader, device)\n",
    "test_loss = test_loss_dict[\"avg_loss\"]\n",
    "test_acc = test_loss_dict[\"avg_acc\"]\n",
    "\n",
    "logging.info(\n",
    "    f\"[{epoch} {i+1}], train loss: {loss.item():.3f}, test loss: {test_loss:.3f}, test acc: {test_acc:.3f}\"\n",
    ")"
   ]
  }
 ],
 "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": "3.9.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}