{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/nvmescratch/jay/miniconda3/envs/rapids/lib/python3.9/site-packages/tqdm/auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "from glob import glob\n",
    "from os.path import exists, join, basename\n",
    "from tqdm import tqdm\n",
    "from json import load, dump\n",
    "from matplotlib import pyplot as plt\n",
    "from collections import Counter\n",
    "from sentence_transformers import SentenceTransformer\n",
    "from sklearn.feature_extraction.text import CountVectorizer, TfidfTransformer\n",
    "from quadtreed3 import Quadtree, Node\n",
    "from scipy.sparse import csr_matrix\n",
    "from sklearn.neighbors import KernelDensity\n",
    "from scipy.stats import norm\n",
    "from typing import Tuple\n",
    "\n",
    "import re\n",
    "import os\n",
    "import shutil\n",
    "import random\n",
    "import cuml\n",
    "import pickle\n",
    "import torch\n",
    "import ndjson\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "SEED = 20220101\n",
    "\n",
    "# plt.style.use('ggplot')\n",
    "# plt.rcParams['figure.dpi'] = 300"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "WORK_DIR = \"/nvmescratch/jay/wizmap/temp/\"\n",
    "DATA_DIR = \"/nvmescratch/jay/wizmap/data/\""
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Download the ACL Meta Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "PARQUET_PATH = join(DATA_DIR, 'acl.parquet')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(63213, 5)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>abstract</th>\n",
       "      <th>year</th>\n",
       "      <th>title</th>\n",
       "      <th>author</th>\n",
       "      <th>acl_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>There is a need to measure word similarity whe...</td>\n",
       "      <td>2002</td>\n",
       "      <td>A Study on Word Similarity using Context Vecto...</td>\n",
       "      <td>Chen, Keh-Jiann  and\\nYou, Jia-Ming</td>\n",
       "      <td>O02-2002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Thread disentanglement is the task of separati...</td>\n",
       "      <td>2013</td>\n",
       "      <td>Headerless, Quoteless, but not Hopeless? Using...</td>\n",
       "      <td>Jamison, Emily  and\\nGurevych, Iryna</td>\n",
       "      <td>R13-1042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>In this paper, we describe a word alignment al...</td>\n",
       "      <td>2005</td>\n",
       "      <td>Aligning Words in {E}nglish-{H}indi Parallel C...</td>\n",
       "      <td>Aswani, Niraj  and\\nGaizauskas, Robert</td>\n",
       "      <td>W05-0819</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>The paper 1 presents a rule-based approach to ...</td>\n",
       "      <td>2013</td>\n",
       "      <td>Recognizing semantic relations within {P}olish...</td>\n",
       "      <td>K{\\k{e}}dzia, Pawe{\\l}  and\\nMaziarz, Marek</td>\n",
       "      <td>R13-1044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>In this paper we describe LIHLA, a lexical ali...</td>\n",
       "      <td>2005</td>\n",
       "      <td>{LIHLA}: Shared Task System Description</td>\n",
       "      <td>Caseli, Helena M.  and\\nNunes, Maria G. V.  an...</td>\n",
       "      <td>W05-0818</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            abstract  year  \\\n",
       "0  There is a need to measure word similarity whe...  2002   \n",
       "2  Thread disentanglement is the task of separati...  2013   \n",
       "3  In this paper, we describe a word alignment al...  2005   \n",
       "5  The paper 1 presents a rule-based approach to ...  2013   \n",
       "6  In this paper we describe LIHLA, a lexical ali...  2005   \n",
       "\n",
       "                                               title  \\\n",
       "0  A Study on Word Similarity using Context Vecto...   \n",
       "2  Headerless, Quoteless, but not Hopeless? Using...   \n",
       "3  Aligning Words in {E}nglish-{H}indi Parallel C...   \n",
       "5  Recognizing semantic relations within {P}olish...   \n",
       "6            {LIHLA}: Shared Task System Description   \n",
       "\n",
       "                                              author    acl_id  \n",
       "0                Chen, Keh-Jiann  and\\nYou, Jia-Ming  O02-2002  \n",
       "2               Jamison, Emily  and\\nGurevych, Iryna  R13-1042  \n",
       "3             Aswani, Niraj  and\\nGaizauskas, Robert  W05-0819  \n",
       "5        K{\\k{e}}dzia, Pawe{\\l}  and\\nMaziarz, Marek  R13-1044  \n",
       "6  Caseli, Helena M.  and\\nNunes, Maria G. V.  an...  W05-0818  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metadata_df = pd.read_parquet(\n",
    "    PARQUET_PATH,\n",
    "    columns=[\n",
    "        \"abstract\",\n",
    "        'year',\n",
    "        \"title\",\n",
    "        \"author\",\n",
    "        \"acl_id\",\n",
    "    ],\n",
    ")\n",
    "\n",
    "# Only get the papers with\n",
    "metadata_df = metadata_df[metadata_df['abstract'] != '']\n",
    "metadata_df = metadata_df[metadata_df['title'] != '']\n",
    "metadata_df = metadata_df[metadata_df['author'] != '']\n",
    "metadata_df = metadata_df.dropna()\n",
    "print(metadata_df.shape)\n",
    "metadata_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# for row in tqdm(metadata_df.itertuples()):\n",
    "#     abstract = row[1].lower()\n",
    "#     year = row[2]\n",
    "#     title = row[3].lower()\n",
    "#     author = row[4].lower()\n",
    "    \n",
    "#     if 'diyi' in author:\n",
    "#         print(title, year)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Extract Abstract Embedding "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cuda\n"
     ]
    }
   ],
   "source": [
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
    "print(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = SentenceTransformer('all-mpnet-base-v2', device=device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "abstracts = []\n",
    "for sentence in metadata_df['abstract']:\n",
    "    abstracts.append(sentence.lower())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Batches: 100%|██████████| 988/988 [11:07<00:00,  1.48it/s]\n"
     ]
    }
   ],
   "source": [
    "embeddings = model.encode(abstracts, batch_size=64, show_progress_bar=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle.dump(embeddings, open(join(DATA_DIR, 'embeddings.pkl'), 'wb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Data: x, y, abstract, title, year"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create UMAP from Embeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_neighbors = 60\n",
    "min_dist = 0.1\n",
    "reducer_cuml = cuml.UMAP(\n",
    "        n_neighbors=n_neighbors,\n",
    "        min_dist=min_dist,\n",
    "        metric='cosine',\n",
    "        n_components=2,\n",
    "        verbose=False,\n",
    "        random_state=SEED\n",
    "    )\n",
    "\n",
    "# Fit UMAP\n",
    "projected_emb_cuml = reducer_cuml.fit_transform(embeddings)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(63213, 2)"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "projected_emb_cuml.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f31ec20a5e0>"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.title(f'n_neighbors={n_neighbors}, min_dist={min_dist}')\n",
    "plt.scatter(projected_emb_cuml[:, 0], projected_emb_cuml[:, 1], s=0.1, alpha=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Bundle dataset together\n",
    "authors = list(map(lambda x: x.lower(), metadata_df['author']))\n",
    "titles = list(map(lambda x: x.lower(), metadata_df['title']))\n",
    "years = list(map(lambda x: int(x), metadata_df['year']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "umap_df = pd.DataFrame({\n",
    "    'xs': projected_emb_cuml[:, 0],\n",
    "    'ys': projected_emb_cuml[:, 1],\n",
    "    'abstracts': abstracts,\n",
    "    'titles': titles,\n",
    "    'years': years,\n",
    "    'authors': authors\n",
    "})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "umap_df.to_csv(join(DATA_DIR, 'umap.csv'), index=False)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create KDE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "cur_df = pd.read_csv(join(DATA_DIR, 'umap.csv'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "xs = cur_df['xs'].to_numpy()\n",
    "ys = cur_df['ys'].to_numpy()\n",
    "\n",
    "projected_emb = np.stack((xs, ys), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-8.79407134 -9.23095064 7.85951834 7.422639039999999\n"
     ]
    }
   ],
   "source": [
    "xs = projected_emb[:, 0]\n",
    "ys = projected_emb[:, 1]\n",
    "\n",
    "x_min, x_max = np.min(xs), np.max(xs)\n",
    "y_min, y_max = np.min(ys), np.max(ys)\n",
    "\n",
    "x_gap = x_max - x_min\n",
    "y_gap = y_max - y_min\n",
    "\n",
    "if x_gap > y_gap:\n",
    "    # Expand the larger range to leave some padding in the plots\n",
    "    x_min -= x_gap / 50\n",
    "    x_max += x_gap / 50\n",
    "    x_gap = x_max - x_min\n",
    "    \n",
    "    # Regulate the 2D grid to be a square\n",
    "    y_min -= (x_gap - y_gap) / 2\n",
    "    y_max += (x_gap - y_gap) / 2\n",
    "else:\n",
    "    # Expand the larger range to leave some padding in the plots\n",
    "    y_min -= y_gap / 50\n",
    "    y_max += y_gap / 50\n",
    "    y_gap = y_max - y_min\n",
    "    \n",
    "    # Regulate the 2D grid to be a square\n",
    "    x_min -= (y_gap - x_gap) / 2\n",
    "    x_max += (y_gap - x_gap) / 2\n",
    "\n",
    "# Estimate on a 2D grid\n",
    "grid_size = 200\n",
    "grid_xs = np.linspace(x_min, x_max, grid_size)\n",
    "grid_ys = np.linspace(y_min, y_max, grid_size)\n",
    "xx, yy = np.meshgrid(grid_xs, grid_ys)\n",
    "\n",
    "grid = np.vstack([xx.ravel(), yy.ravel()]).transpose()\n",
    "print(x_min, y_min, x_max, y_max)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>KernelDensity(bandwidth=0.14677992676220697)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">KernelDensity</label><div class=\"sk-toggleable__content\"><pre>KernelDensity(bandwidth=0.14677992676220697)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "KernelDensity(bandwidth=0.14677992676220697)"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# # Compute the bandwidth using silverman's rule\n",
    "sample_size = 100000\n",
    "n = sample_size\n",
    "d = projected_emb.shape[1]\n",
    "bw = (n * (d + 2) / 4.)**(-1. / (d + 4))\n",
    "\n",
    "# We use a random sample to fit the KDE for faster run time\n",
    "rng = np.random.RandomState(SEED)\n",
    "random_indexes = rng.choice(range(projected_emb.shape[0]),\n",
    "                            min(projected_emb.shape[0], sample_size),\n",
    "                            replace=False)\n",
    "\n",
    "kde = KernelDensity(kernel='gaussian', bandwidth=bw)\n",
    "kde.fit(projected_emb[random_indexes, :])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(200, 200)"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Sklearn\n",
    "log_density = kde.score_samples(grid)\n",
    "log_density = np.exp(log_density)\n",
    "grid_density = np.reshape(log_density, xx.shape)\n",
    "grid_density.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiIAAAGzCAYAAAASZnxRAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAACFZElEQVR4nO29eZwU1bn//xmG6R5mHxiGRRBFUUQUFAU3FFwQA7jE4DUa44rRYNSYa4JLfu6iwe1qXIh+xcSVq0aN+x7FJVFxRcUlLpdFVmF6mNEeZqjfH031nDp9zqlzaumq7n7er1e/YKqrq05t53zqOc9SZlmWBYIgCIIgiAjoEXUDCIIgCIIoXUiIEARBEAQRGSRECIIgCIKIDBIiBEEQBEFEBgkRgiAIgiAig4QIQRAEQRCRQUKEIAiCIIjIICFCEARBEERkkBAhCIIgCCIySIgQRMBcfPHFKCsr01q3rKwMF198cWD73rBhA0455RT0798fZWVlOPvsswPbNhFfvvnmG5SVleGuu+7K+77/93//F71798aGDRuyy8rKynDGGWfkvS2Fxtq1a1FdXY2nnnoq6qZECgmRPHLXXXehrKwM77zzjmN5S0sLxo4di8rKSjzzzDMAugcz+1NVVYUtt9wS06ZNw7x585BOp3O2f8IJJzh+w34qKyvzcoxuLFmyBJdccgnGjh2LxsZGNDU1YcKECXjhhReE669fvx6nnnoq+vbti+rqakycOBHvvvuucN1//OMf2HXXXVFZWYktt9wSF110ETo7O7XbtmrVKsyaNQs77bQTampqUFlZiW233RYnnngiXnvtNU/Hm2+uvPJK3HXXXTj99NNx991347jjjpOu+9xzz+Hkk0/GyJEjUV5ejq222kq67qZNm/CnP/0JW2+9NSorK7Hzzjvj/vvvF6776aefYvLkyaipqUHv3r1x3HHHYfXq1bHbpojXXnsNhxxyCLbYYovsfTRt2jTcd9992tsoJbq6unDRRRfhN7/5DWpqaqJujhZvvPEG9tlnH1RVVaF///4488wzHSJKxa233orp06djyy23RFlZGU444QThenZfL/qsWLEiu16fPn1wyimn4I9//GMQh1aw9Iy6AaVOKpXCpEmT8OGHH+KRRx7B5MmTHd/feuutqKmpQTqdxrJly/Dss8/ipJNOwg033IAnnngCgwcPdqyfTCZxxx135OynvLw81OPQ5bHHHsPVV1+Nww8/HMcffzw6Ozvxt7/9DQcddBDuvPNOnHjiidl1N23ahClTpuCDDz7Aueeei6amJtxyyy2YMGECFi5ciGHDhmXXffrpp3H44YdjwoQJuOmmm/DRRx/h8ssvx6pVq3Drrbe6tuutt97ClClT0NraiqOPPhqnnXYakskkvv76azz66KO466678Morr2Dfffd13daFF16IWbNmeTtBPnnppZewxx574KKLLnJd97777sP8+fOx6667YuDAgcp1L7jgAlx11VWYMWMGdt99dzz22GM45phjUFZWhqOPPjq73tKlS7Hvvvuivr4eV155JTZs2IBrrrkGH330Ed566y0kEolYbFPEgw8+iP/6r//C6NGjcdZZZ6GxsRFff/01Xn31Vdx+++045phjXM9pqfH444/js88+w6mnnhp1U7R4//33ccABB2CHHXbAddddh6VLl+Kaa67BF198gaefftr191dffTVaW1sxduxYfPfdd67rX3rppdh6660dyxoaGhx/n3baabjxxhvx0ksvYf/99zc6nqLBIvLGvHnzLADW22+/bVmWZaVSKWuPPfawEomE9cQTTzjWveiiiywA1urVq3O2c88991g9evSwxo0b51h+/PHHW9XV1eEdQAAsWrQo55h+/PFHa/jw4dagQYMcy+fPn28BsB588MHsslWrVlkNDQ3Wz3/+c8e6I0aMsEaNGmVt3Lgxu+yCCy6wysrKrE8//VTZpu+//94aMGCA1b9/f+G6mzZtsu677z7rrbfeUm5nw4YNyu9FALAuuugi49/J2Hrrra0pU6Zorbts2TKro6PDsizLmjJlijVkyBDhekuXLrUqKiqsmTNnZpdt2rTJGj9+vDVo0CCrs7Mzu/z000+3evXqZX377bfZZc8//7wFwJo7d25stilixIgR1o477mil0+mc71auXKn8bdR8/fXXFgBr3rx5ed3voYceau2zzz45ywE4rkNcOOSQQ6wBAwZYLS0t2WW33367BcB69tlnXX//zTffWJs2bbIsy7Kqq6ut448/Xrge39e7MXLkSOu4447TWrcYISGSR9ibs7W11dprr72sRCJhPf744znrqoSIZVnWqaeeagGwnnvuuewyv0Jkw4YN1jnnnGMNGjTISiQS1nbbbWfNmTMn++DZ2J3MI488Yu24445WIpGwRowYYT399NOe933OOedYAKxUKpVdNn36dKtfv35WV1eXY91TTz3Vqqqqsn788UfLsizr448/tgBYN998s2O9ZcuWWQCsyy67TLnvK6+80gJgPfDAA9rtta/Pxx9/bP385z+3GhoarNGjRzu+Y/nxxx+ts88+22pqarJqamqsadOmWUuWLNEWIitXrrROOukkq7m52Uomk9bOO+9s3XXXXdnvX375ZQtAzufrr7/WOh6VELn55puzx8py3333WQCsBQsWZJc1Nzdb06dPz9nGdtttZx1wwAGx2aaIZDJpnXDCCcp1bObMmWPtueeeVu/eva3Kykpr1113dQhmG/tZ+d///V9rhx12sCorK6099tjD+vDDDy3LsqzbbrvN2mabbaxkMmntt99+Oddrv/32s3bccUfrnXfesfbcc0+rsrLS2mqrraxbb73VsZ5MiHz66afWkUceaTU2NlrJZNIaM2aM9dhjjznW6ejosC6++GJr2223tZLJpNW7d29r7733dvQtIn744QcrkUhYF198sfS477nnHmu77bazksmkteuuu1qvvPJKdp0PPvjAAuBozzvvvGMBsHbZZRfH9iZPnmyNHTtW2R43WlparJ49e1rnnnuuY3k6nbZqamqsk08+2Wh7ukIklUq5iuDf/va3VkNDQ05fWyqQj0gEtLW14ZBDDsHbb7+NBx98EFOnTjXehj33/9xzz+V8t2bNmpxPKpVSbs+yLBx66KG4/vrrMXnyZFx33XXYfvvtce655+Kcc87JWf+1117Dr3/9axx99NH405/+hB9//BFHHnkk1q5da3wsALBixQpUVVWhqqoqu+y9997Drrvuih49nLfp2LFj0d7ejs8//zy7HgDstttujvUGDhyIQYMGZb+X8fjjj6NXr1746U9/atzu6dOno729HVdeeSVmzJghXe+UU07BDTfcgEmTJuGqq65CRUUFpkyZorWPH374ARMmTMDdd9+NY489FnPmzEF9fT1OOOEE/M///A8AYIcddsDdd9+NpqYmjB49GnfffTfuvvtu9O3b1/iYeN577z1UV1djhx12cCwfO3Zs9nsAWLZsGVatWpVzHex12esQ5TZlDBkyBC+++CKWLl2qXA8A/ud//ge77LILLr30Ulx55ZXo2bMnpk+fjieffDJn3QULFuB3v/sdjj/+eFx88cX49NNPMXXqVNx888248cYb8etf/xrnnnsu3nzzTZx00kk5v1+3bh1+8pOfYMyYMfjTn/6EQYMG4fTTT8edd96pbOPHH3+MPfbYA59++ilmzZqFa6+9FtXV1Tj88MPxyCOPZNe7+OKLcckll2DixIn485//jAsuuABbbrml1BfLZuHChejo6MCuu+4q/P6VV17B2WefjV/84he49NJLsXbtWkyePBmLFi0CAIwcORINDQ149dVXHeeqR48e+OCDD7J91qZNm/DGG284pkU3bNgg7Of4T0tLS/Y3H330ETo7O3PupUQigdGjR7veH16YOHEi6urqUFVVhUMPPRRffPGFcL0xY8Zg/fr1+PjjjwNvQ0EQtRIqJWyVPGTIEKuiosJ69NFHpeu6WUTWrVtnAbCOOOKI7LLjjz9e+FYMwDr44IOVbXv00UctANbll1/uWP6zn/3MKisrs7788svsMgBWIpFwLLPfbm666SblfkR88cUXVmVlZY5psrq62jrppJNy1n/yySctANYzzzxjWVbm7RSA9X//93856+6+++7WHnvsodx/Y2Nj1prBkkqlrNWrV2c/7NSLfX34KSL2O5v333/fAmD9+te/dqx3zDHHaFlEbrjhBguAdc8992SXdXR0WHvuuadVU1PjsCINGTJEe2qGRWURmTJlijV06NCc5W1tbRYAa9asWZZlWdbbb79tAbD+9re/5ax77rnnWgCyVqwotynj//2//5e9tydOnGj98Y9/tBYsWJBjkbMsy2pvb3f83dHRYY0cOdLaf//9HcsBWMlk0mHpmDt3rgXA6t+/v+PanXfeeTlWrP32288CYF177bXZZel02ho9erTV3NycnVoTWUQOOOAAa6eddsqeH8vKTFXttdde1rBhw7LLRo0a5emeueOOOywA1kcffZTznd3vvPPOO9ll3377rVVZWenos6ZMmeKwdPz0pz+1fvrTn1rl5eVZC+u7776bYzlR9XXsZ7/99sv+5sEHH7QAWK+++mpOe6dPn27179/f6PhVFpH58+dbJ5xwgvXXv/7VeuSRR6wLL7zQqqqqspqamoT91BtvvGEBsObPn2/UhmKBLCIRsHLlSlRWVuY4mppge6i3trY6lldWVuL555/P+Vx11VXK7T311FMoLy/HmWee6Vj+u9/9DpZl5ThyHXjggdhmm22yf++8886oq6vDV199ZXQc7e3tmD59Onr16pXTxh9++AHJZDLnN3YE0A8//OD4V7au/b2MVCol9Pg/7rjj0Ldv3+znD3/4Q846p512mnLbALKhefy51Q2tfeqpp9C/f3/8/Oc/zy6rqKjIevu/8sorWtvxSlDXgV83qm3KOOmkk/DMM89gwoQJeO2113DZZZdh/PjxGDZsGN544w3Hur169cr+f926dWhpacH48eOFVoQDDjjAEZE0btw4AMCRRx6J2tranOX8M9SzZ0/86le/yv6dSCTwq1/9CqtWrcLChQuFx/L999/jpZdewlFHHYXW1tashWDt2rU4+OCD8cUXX2DZsmUAMs6TH3/8sfRtXYZt/WxsbBR+v+eee2LMmDHZv7fcckscdthhePbZZ9HV1QUA2XPW1tYGIGNp/clPfoLRo0djwYIFADJWkrKyMuyzzz7Zbf3+978X9nP859prr83+xm8/YcJRRx2FefPm4Ze//CUOP/xwXHbZZXj22Wexdu1aXHHFFTnr2+dwzZo1gbWhkKComQiYO3cuzjnnHEyePBkLFizA9ttvb7wNO9yM7ciATHTMgQceaLy9b7/9FgMHDszZnm3m/vbbbx3Lt9xyy5xtNDY2Yt26ddr77OrqwtFHH41PPvkETz/9dE7kRq9evYRhyj/++GP2e/Zf2brsoCGitrZWGL536aWXZnMhHHTQQcLf8h7xIr799lv06NHDIdwAaF/3b7/9FsOGDcuZopJdm6AJ6jrw60a1TRUHH3wwDj74YLS3t2PhwoWYP38+brvtNkydOhWLFy9Gc3MzAOCJJ57A5Zdfjvfff9+xT1H+GP5Zqa+vB4CcFxF7Of8MDRw4ENXV1Y5l2223HYBM/pA99tgjZ59ffvklLMvCH//4R2lo6KpVq7DFFlvg0ksvxWGHHYbtttsOI0eOxOTJk3Hcccdh5513Fv6Ox7Is4XI2qo1td3t7O1avXo3+/ftj/Pjx6OzsxJtvvonBgwdj1apVGD9+PD7++GOHEBkxYgR69+6d3c6IESMwYsQIrfbZ+O0n/LLPPvtg3LhxwlQF9jnUzT9UbJAQiYARI0bgqaeewgEHHICDDjoIr7/+urF1xJ5n3XbbbcNooiuycGBZpyRixowZeOKJJ3DvvfcKw9YGDBggDJGzl9nCZcCAAdnl/Hn87rvvsj4CMoYPH44PPvgAGzduREVFRXa5TkccducVBwYMGICXX34ZlmU5OkrVdeD57rvv0Lt37+zbaJTb1KGqqgrjx4/H+PHj0dTUhEsuuQRPP/00jj/+eCxYsACHHnoo9t13X9xyyy0YMGAAKioqMG/ePGG+EdmzEsQzJGPTpk0AgP/+7//GwQcfLFzH7jv23Xdf/Oc//8Fjjz2G5557DnfccQeuv/563HbbbTjllFOk++jTpw+AjHAaNGiQp3butttuqKysxKuvvoott9wSzc3N2G677TB+/HjccsstSKfTWLBgAY444gjH71paWrQsGIlEIitg3O4lk/vDK4MHD8Znn32Ws9wWn01NTaG3IY7Q1ExEjB07Fo8++ihWrVqFgw46SJicScXdd98NANJOxpQhQ4Zg+fLlOVM9ixcvzn4fJOeeey7mzZuH66+/3jHlwDJ69Gi8++672U7V5t///jeqqqqyb4WjR48GgJxEccuXL8fSpUuz38uYOnUqfvjhB4cDX5AMGTIEmzZtwn/+8x/HclGHJPv9F198kXMewro2PKNHj0Z7ezs+/fRTx/J///vf2e8BYIsttkDfvn1zrgOQydPCXocot2mK7dxoD2APP/wwKisrszl9DjnkEE9WSF2WL1+enbqwsR21ZUnohg4dCiAzhXfggQcKP6z1s3fv3jjxxBNx//33Y8mSJdh5551dM/4OHz4cAPD1118LvxdN9Xz++eeoqqrKOlEnEgmMHTsWCxYswIIFCzB+/HgAmSmbdDqNe++9FytXrszJ33PWWWdhwIABrh/WAX3kyJHo2bNnzr3U0dGB999/3/P9YcJXX30ldCC3zyHvaF0qkBCJkAMOOAD3338/vvzyS0yePNk1ssXmvvvuwx133IE999wTBxxwQCBt+clPfoKuri78+c9/diy//vrrUVZWhkMOOSSQ/QDAnDlzcM011+D888/HWWedJV3vZz/7GVauXIm///3v2WVr1qzBgw8+iGnTpmXfhHfccUcMHz4cf/nLX7Jzz0AmGVxZWRl+9rOfKdtz+umno1+/fvjtb3+b7eBZ/L6h2ufuxhtvdCy/4YYbtH7/k5/8BCtWrMD8+fOzyzo7O3HTTTehpqYG++23n6/2uXHYYYehoqICt9xyS3aZZVm47bbbsMUWW2CvvfbKLj/yyCPxxBNPYMmSJdllL774Ij7//HNMnz49NtsU8eKLLwqX2z4+9lRaeXk5ysrKHPfaN998g0cffVS5fa90dnZi7ty52b87Ojowd+5c9O3b1+GDwdLc3IwJEyZg7ty5QgsA++LDR7rV1NRg2223FU5hsIwZMwaJREIoEgHgzTffdPjMLFmyBI899hgmTZrksAaNHz8e//73v/Hyyy9nhUhTUxN22GEHXH311dl1WLz4iNTX1+PAAw/EPffc43jhuvvuu7FhwwbHvdTe3o7Fixd79tkQvVg+9dRTWLhwYU7SSiATgVRfX48dd9zR0/4KHZqaiZgjjjgCt99+O0466SQceuiheOaZZxzp2B966CHU1NSgo6Mjm1n19ddfx6hRo/Dggw/mbK+zsxP33HOPdF/8XLPNtGnTMHHiRFxwwQX45ptvMGrUKDz33HN47LHHcPbZZ+f4N3jlkUcewe9//3sMGzYMO+ywQ05bDzroIPTr1w9ARojsscceOPHEE/HJJ59kM6t2dXXhkksucfxuzpw5OPTQQzFp0iQcffTRWLRoEf785z/jlFNOcX3L6N27Nx555BFMmzYNo0aNwtFHH43dd98dFRUVWLJkSfY8i/xidBg9ejR+/vOf45ZbbkFLSwv22msvvPjii/jyyy+1fn/qqadi7ty5OOGEE7Bw4UJstdVWeOihh/D666/jhhtuyPHr0eXDDz/EP/7xDwAZn4KWlhZcfvnlAIBRo0Zh2rRpAIBBgwbh7LPPxpw5c7Bx40bsvvvuePTRR7FgwQLce++9jkHl/PPPx4MPPoiJEyfirLPOwoYNGzBnzhzstNNOjqy5UW9TxGGHHYatt94a06ZNwzbbbIO2tja88MILePzxx7H77rtnz8eUKVNw3XXXYfLkyTjmmGOwatUq3Hzzzdh2223x4YcferoWKgYOHIirr74a33zzDbbbbjvMnz8f77//Pv7yl784phJ5br75Zuyzzz7YaaedMGPGDAwdOhQrV67Em2++iaVLl+KDDz4AkJkqnjBhAsaMGYPevXvjnXfewUMPPeRaK6ayshKTJk3CCy+8gEsvvTTn+5EjR+Lggw/GmWeeiWQymRWI/LM7fvx4XHHFFViyZIlDcOy7776YO3cuttpqq5ypHy8+IgBwxRVXYK+99sJ+++2HU089FUuXLsW1116LSZMmOQTCW2+9hYkTJ+Kiiy5yWIYef/zx7HnbuHEjPvzww+wzc+ihh2anc/faay/ssssu2G233VBfX493330Xd955JwYPHozzzz8/p13PP/88pk2bVrI+IhS+m0dU2fauueYaC4A1depUa+PGjdkQUPtTWVlpDRo0yJo6dap15513OkLybNxC2tySW7W2tlq//e1vrYEDB1oVFRXWsGHDlAnNeIYMGSINZ7Phj4v/vPzyy471v//+e+vkk0+2+vTpY1VVVVn77befNFvhI488Yo0ePdpKJpPWoEGDrAsvvDAb3qjDd999Z5177rnWiBEjrF69elnJZNIaOnSo9ctf/jIn5E8VXi1KaPbDDz9YZ555ptWnTx+rurraU0KzE0880WpqarISiYS10047CbNomoTv2vej6MNfx66uLuvKK6+0hgwZYiUSCWvHHXd0hBOzLFq0yJo0aZJVVVVlNTQ0WMcee6y1YsWKnPWi3ibP/fffbx199NHWNttsY/Xq1cuqrKy0RowYYV1wwQWOMFvLyoT6Dhs2zEomk9bw4cOtefPmCa+76FmxQ23nzJnjWG4npWMTo4kSmg0ZMsT685//LNwmf0/85z//sX75y19a/fv3tyoqKqwtttjCmjp1qvXQQw9l17n88sutsWPHWg0NDVavXr2s4cOHW1dccYXWs/P3v//dKisrywlJtY/7nnvuyZ6nXXbZJef5tqxMmHx5eblVW1vrSPx1zz33WAACzzi6YMECa6+99rIqKyutvn37WjNnzsy5vva14J9NVR/LnvsLLrjAGj16tFVfX29VVFRYW265pXX66acL79lPP/3UAmC98MILgR5nIVFmWQF4RhEEQRCBM2HCBKxZsybrnB43urq6MGLECBx11FG47LLLom5OQXL22Wfj1VdfxcKFC0vWIkI+IgRBEIQnysvLcemll+Lmm2/WrmBLdLN27VrccccduPzyy0tWhAAAWUQIgiBiStwtIgQRBGQRIQiCIAgiMsgiQhAEQRBEZJBFhCAIgiCIyCAhQhAEQRBEZMQ6odmmTZuwfPly1NbWlrRHMUEQBEEUEpZlobW1FQMHDswp2MkTayGyfPly42JwBEEQBEHEgyVLlrgWRYy1ELFTV3/6n29RW1sXcWsIgiAIgtChtTWFHbYZolWCItZCxJ6Oqa2tQ10dCRGCIAiCKCR03CrIWZUgCIIgiMggIUIQBEEQRGSQECEIgiAIIjJIiBAEQRAEERkkRAiCIAiCiAwSIgRBEARBRAYJEYIgCIIgIoOECEEQBEEQkUFChCAIgiCIyCAhQhAEQRBEZJAQIQiCIAgiMkiIEARBEAQRGSRECIIgCIKIDBIiBEEQBEFEBgkRgiAIgiAig4QIQRAEQRCRQUKEIAiCIIjIICFCEARBEERk9Iy6AQRBEERp05bucl2nOlmeh5YQUUBChCAIgogMVoS0CwRJ1WYBYq9HgqT4ICFCEARB5B2ZAGlPd0p/U5UsR1u6i8RIkUE+IgRBEERk2CKkPd0pFCGiZTpTOUThQBYRgiAIIq+0ZcVHtwgBgPYOucCoSvZEe7orO1VDFA9kESEIgiAiRyVCAPWUDVHYhC5Eli1bhl/84hfo06cPevXqhZ122gnvvPNO2LslCIIgCgBeYKxPd2B9uiOi1hBREOrUzLp167D33ntj4sSJePrpp9G3b1988cUXaGxsDHO3BEEQRIGQmXLpFiMNyUSErSGiIFQhcvXVV2Pw4MGYN29edtnWW28d5i6JAkLkcEbe8ARRmlQlyoXTM1WJTJ9QlSSXxmIl1KmZf/zjH9htt90wffp0NDc3Y5dddsHtt98uXT+dTiOVSjk+RHEi83onb3iCKH7sF46qpFNkVCXKcz6O7zevTy8sxUWoQuSrr77CrbfeimHDhuHZZ5/F6aefjjPPPBN//etfhevPnj0b9fX12c/gwYPDbB5BEAQRMawYkX3Y9UiEFB9llmVZYW08kUhgt912wxtvvJFdduaZZ+Ltt9/Gm2++mbN+Op1GOp3O/p1KpTB48GAsXbUOdXV1YTWTiACV5YM6GoIoHfi+wA7pFYXpUt9QOKRSKQxqbkRLS4vr+B3qpNuAAQMwYsQIx7IddtgBDz/8sHD9ZDKJZDIZZpOImFC9OUOiaDlBEKVDNZfCnQRI6RGqENl7773x2WefOZZ9/vnnGDJkSJi7JQoE6lwIgrCh/qB0CdVH5Le//S3+9a9/4corr8SXX36J++67D3/5y18wc+bMMHdLEARBEESBEKoQ2X333fHII4/g/vvvx8iRI3HZZZfhhhtuwLHHHhvmbgmCIAiCKBBCdVb1SyqVQn19PTmrEgRBEEQBYeKsSrVmCIIgCIKIDEpVRxAlAoVMEwQRR0iIEESRI8vTAHSHStrrkCAhCCLfkBAhiCLGFhjtOWKkc3OxMacoIUFCEES+ISFCEEUKL0L4cuv83za2ICExQhBEPiAhQhBFiEyEiKqbsthWEhIjBEHkCxIiBFHk8CJkfbojZ52GZCL7PYkRgiDyCQkRgigBeBGyPr0xZx0SIwRBRAHlESGIEkMkQjLLuy0lMv8RgiCIoCEhQhAlQFVCbdFgxYnIj0SVg4QgCMIPJEQIosipSjpnYBuSFRG1hCAIIhfyESGIIqR6s29HVbI8GzmTsYoksD7doRQjbtYTgiCIICEhQhBFjp09tT3dmRUjgNMnpCGZiKJpBEEQJEQIolipZjKlVgmjXpzig7WE8NM5FDVDEERYkBAhiCKGFyOZkFz5Y89+JxYvBEEQwUJChCCKnGqmsF33NI1ckPAChKwhpYEoMoquPZEPSIgQRIkgEiS6vyGKE7ewbCqCSOQDEiIEUWKwgwq9BZcm/HXnqzOzUFVmImxIiBBECUMDS+nBihBZZWYbNtW//Vu6Z4igISFCEARRIphUZa5KlGdCvkmMECFDQoQgCKIEUIkQUUVmPrwboEgqIhwoxTtBEESRk+sT0on2ji6HCFmf3pj9ZP7OLBdZSwgiSEiIEARBlAjt6S6HP8j6dIdDfHQvd4oR9vcAFUEkgoWECEEQRBHDT8kAqukYgsg/JEQIgiBKBD46hreEuEE+IkQYkBAhCIIoUmTWEF0akglhNWaKmiGChIQIQRBEEaLjx9GQrJAuZysyq+oTEYRf6O4iCIIoMkySlrHwwsRZkTnzf7KGEEFDQoQgCCJP8FaKoAd1k9TtNrz4sC0htgipSvYk3xAiVEiIEARBhIhqiiTIGi4qESLKoNqQTDgiZxxTMYwIYSFrCBEGJEQIgiBCQjffhh9BoiNAALGTKis+AH4qpufmf2lKhggXEiIEQRAhoDtNUiWohqwz6Mu2z/uC6EbJiKwgNCVD5AMSIgRBECGj8tVgC8rZyASJyMIiEiC8+JBNwdi4iRCyhhBhQkKEIAgiYExToIvEiNt2VD4gqqyp9nciQWLj1xIStlMuUVzkLY/IVVddhbKyMpx99tn52iVBEETBoBPhkqkV0+WwgvAF7ERF7PhP5nu9FO+6IqIt3ZX9mHxHEHmxiLz99tuYO3cudt5553zsjiAIoiCxBQZvkeBFisoCwqZtX9PuFBtNVYnsOg3JCqxPdygtI0BGRLiJEROBEWSkEFEchG4R2bBhA4499ljcfvvtaGxsDHt3BEEQBQ9r+eCnYFgLCIAcCwiQESC8COGXs4LF3lZW4HgQFm7HwG+TrCOETehCZObMmZgyZQoOPPBA13XT6TRSqZTjQxClCmvOJtN2aWMLEMBpBcl85AJkTVtn9sPCihFbyPAOrqxwkN13MudZmZDhv6P7mQBCnpp54IEH8O677+Ltt9/WWn/27Nm45JJLwmwSQcQanY6ZXYfM2/FDdA2rkuVGVgYbWSQMawUBcqdgeOFhL2uqFnf57BRNe7pTWFuGn6JRRfC4wTrn6kz9EMVNaBaRJUuW4KyzzsK9996LyspKrd+cd955aGlpyX6WLFkSVvMIInao3i5lb5n0RhkvVNfDJBKFtYAAuVaQzP/1RYjoO36KZn26w3WKxtQyl51K4o6HIFjKLMuywtjwo48+iiOOOALl5d0PX1dXF8rKytCjRw+k02nHdyJSqRTq6+uxdNU61NXVhdFMgogFpjVC+EGN3iijwYsQVOcUyR2sVZYQIFeIAGoxwltFbAdWtuIun1dEV0TJsrqyUK6S0iCVSmFQcyNaWlpcx+/QpmYOOOAAfPTRR45lJ554IoYPH44//OEPriKEIEoFL4XK+LwTZN7OH2FZoYISIUBGbKjEiAx7iqa9owtVifLsNI0szwlBBEFoQqS2thYjR450LKuurkafPn1ylhNEqeJnUCMxkl/8ChA/lhAvyPxBZLiF9JIYIcKCMqsSRIzw4tDIEkcxUgzOtW4ixO91MyEjFjJWEXtaRWYZ8QNvFTElY0npzFlGEDx5vSv++c9/5nN3BBF7qpPljkHOa3RFHJFl2LQpFFGiypMRBF6cOFkxAnQLEhnSKRyX3/G4WUX4+1clPLxYV9wEYaHcU4QTkqcEETN0xUiczeS6YchxHzjCFiEmNCQTXPG6Csf3rDDh0REc/PZYTKwiYYhp3WkxytpamJAQKQBkDyE9bMUBbxUB4i0y3CiWkOKoRYgducL6irC+G3ytGJWQ6P5NrljR+R2Ljq+I/b3oXIl+K+vL3Ir+ydpBgqSwICESc1QPYiG8URJ62NexWAbxQief10HkS+H4XiBIAHX1XBaVFUWE7nZ18COodYSgTBRWcc8T9ZPxhoRIjAnTvE0PaDzhr4fpgEjX0z9RiEF22kOafyPhfm1FkTYiYeEUJ2LhobO/IBDds36tUXzxQOrv4g0JkRKiWJwHSwm6JqWHyhfDzbFVJB50xYlqO2FFu5iIEBWy6ByRIKFnKn6QECkC3B4uE0cvekgJv4h8XmTrxYlCmBZzEwQiocKLCl6YhGX5kN0HXvxBVAhzsDDL+IRs1M/FDxIiRQA9VETccBMjcbtnTQZBlSOm229M8OIUG9QUj2ybutjX1+06+08S5zxGkcgSZYclMRIvSIjEGN03SxVuc618B0kPKBEUpm/E+cbvsxV2ZJPO9lVipXt6Qj9PSVBTMG4+GV7PvSo0WDQFpUpVT31dfCAhEnNU0RReHiL+Iaa0zUSYxK2jL4TpFxN0rDNRZjMN43zr5CmxnXGpbk5hQEKkQMhnh656U/DSscRtMCJKj2ITIDz5zMgbp0FcFPrM51cBui0jRDwhIVLEeKnqyv5WJ7eFTsIiCp0jbKIwh0clQkyP028BxFLCq/DyWjeHCBe6IkWKiQiRmSpFHaNuiXoWiuUngO7rny8xkg8BEuRxeE1qF4QIiZOVwy98KnwV5CcSD0iIFCFeLCFu86bObIZ6zm8Uy0/Y8Pdk2PdAmCIk7HvXxEm9lEWIbRWxp2eqEuVZh1VpkjbOGkL9UDwgIVJkuIkQW0SIzJMiMSITICIPdRUij3WAOoJSQDaohiVGvE5J6gzIcbLkqK2czjwahYpppmFWjPDLifhSuHcokYOqA86Jt5d0VNLQuM3r2w+5nukz81bCe6xn9kmChAieNsc9L5paFGfgZNcvBAuB23MqWpZvQeJWB0aFrD/grUW8VQSQiw77+Avh+pYaJESKBBMRwqN6exJZQVgRIis93pCswPp0hzB8zm4f2yG4vemQUCHckIkQ1f0vuvdVgiRfVhwVJiKE/17XEuoXN580QC4IvJ5fVfFA0XFTnxIfSIgUOCqHUpOplGyFT9EbFfNbW4SwAmRNu9M60lSVEAgU55ytyDqigmriECpEIsT0/uctB2Em/Qvah8UkaVnYmBSmC0IAsRE0fBI3XoCQNSSekBApMHRDafmpFMBtOsW99Lfo97wIsZc1VSU2/2ajlnUE0O8kyNmVYFFNx+hPJTqnEQFIrXfsPoMq2pYv4ha+aiJGTCryio6R3Q/1H/EiPnckIcXkAVRNpcimUezvGpIV2b95r3Od6RgWVozI4DtF9njozSWD10EtTh2tKCw17Pa1pzuN7v9ucv2aMtvTD3GXt8nFKuMzpbsfgrJOBNE+0YuGSR8ogvqTeENCJMaYOt4B6mkU1nqxpi3z26Zq81sgY+HQ6dTd8TpvXexWEd2OV3aO+N/H4VyF7VshmpJRPQM29lSizHIHOKcSM3+Lj8WL/wZvGYwKlWUy32nR7efbrwABco8nDs8C4YSESExpy3aqZuGzvOXC7nht4bFyg1NA2H/3q6nAmrbOHGHCWklE2FYPUQcvQzfdcinWhDANPdXJbMtvt1g6YuU0pUOQy0WIbLktRoBc6wigHzorag8LK3Ycyw2nT1SOmqbblYspdyHmJeOpqbVJlpKguw00rBUadMViCC9CdENneQdSVnys35AGAKxry/1tY3UC6zek0VCTxMoNG9GvJiM+2DdFHnuZvU+3aRiWIGL6i2UwtfES8cEjerNWdfCFfA5lgi0Ip032ORJZR9zQ98sC1qed06C807jJoOql2m4c0H3ZcBMg/HLRtFoh3/PFDAmRmKESIbK5bv6Njhcg69o6sD71IwCgtTWds8/1tUk01FUK28MLDLvT7K5u2S1SnB14hfB3PKX+9uImQHQSx6kGryhCUcPGTw0lFt4C6OZkzcKKaf47kT+VKLKMXaf7eXGKEj+CBAgvh4jKWpmPAnyFJrYINaU9CsQcXoSofD1s7KkWVoC0tqaRaskIkfbWdsf6VbVVADYLlC3q0Vgtt2ywYkLkzOomPnhLiFvnWKzTMjo5X7wkjgPkg1eYoaj5wqvjru3T1FSVcDw7Iv8oXozIcXfm5qdFHftx8dFiLTGA92iXqIS+fZ8F4aDLb8dtitrvOSPyD12lmMM/aKLOjfX7EAkQW3y0pdqk+6mrr8xaTQBkp2dkZB92l9oO7LrZvzU6B1UHVSgDp43OXLeO5UuE/dae+X/u4AXkCpJCq45sGrLuF4dYEeTEYc+5jSivjswvyw2RJUbkp1II6AoSGarpGNMyE0R8Kay7mgAg7uBYHxBehLSl2mC1rs2s2LIq8299MwCAlSa1tUnpPm2hkSMqDPw9/AoQIL6DJYvuwJn5Wy5AeMuXKsJJNnjZ+9A597wgMbVABH1t3PZvMrjJrCI8oukaXUSWStZCmdOmzT5ZNk3VPZWOs16matzEmVdhY2qtlAkSt0gd5996IkTXGZ6IDyREigCZI6pUhNj/3yxGvOL37ayYpl68iA/AOQXjZtLn/7YHLlH4qQqd0GgvqH6nK1K8FnwzeVPmxYjbeXb8ViIWRL9XiRB2OStIVNfUzdrFou1Ym+c6NDL/EffosGAsIYU0FVlKkBCJMVmvb+bBy4bLtnWiX00FVm7YiIaaZE5nZ/uESFGIkIYasWXE7gTz9RbFE8cORJVi37mMKzooSLZlYtJnw611By63KZowCSrbqG4+HV1kPlYi7HNuI/QxUYgYGWzEmr1d1TUF1CUZvJAvQWIe2it+blSQNaTwICESM+wkPo76CYly2M5xtnkZcIoRIBOGKwrPzVLf7LSKbKaqtgp19ZXZbQCZzrCpKoGGZIUzvNAldXJYyESIaICLqlS7W3ihTrI5Xb8CO8y6kMSIH3QThclCZ1U+HICeHwebc8f+vZeEgOvaOpRO4W7XFAhv+kE3J4eXe8fE6iH8nnxCihYSIjGEFSNOnJ2Xar67tT5jIamqrUJ7a3W3o2ptn+791FVnRUjt5hDehpqk0FFV1OlFPYjJ3rLDNr+6hY+qzMg60RVufgU8bgOimxgBor+WMozK3RuIEBaRr5UI+9yzuXb4c99U3TN7HWUWS5EI4fP4uFlWRNaRMFBX5vYnZHWsOX7Eh0xE0fRM/CAhEnNsy4j9UFUlegkfTrYDBABsUY/1tUlh3hCgO2yXFSGN1Qn0q6nIsYaIpmTyPXDxnYebqT+MzsbNCmIqQAD/ERYsum/QIufVoOr8+Em9rbuNIESIrh+HF2RiRHc/omzHbtcW0BMkfnxKTPPUOH+r6fujEB6iUHY3fyiiMCAhElPYyAVejGRgc3rIa7801FVuFiS5CctYAQJ0v/G5ZUmVdTpBOySKfqeqPxEmOrk/ALUAyfytniJg4d+kZb47IryIkewx5On8etlP3EWIjW0lsadP7f2ZChI3MQJAa7rGPWePXop4wFyQ6Bbn5NHLoeOE/EMKExIiMYefprEFid1x8BaSpqoEmqo7sIbxHVlfncC6OnFqd6B7gLOtIQC0fEPYNno5Ji8UighRCRBAPyzURHyYks+8FH73JU3n7VOEhA0vSOz96l5X1VRNd5kFdaRUENdYJHJUgkS5LQPh7pUorbeEOaH2QrNnz8bf//53LF68GL169cJee+2Fq6++Gttvv32Yuy06eOsIAKGFpCGZyD7IWUHCdYA82boymztMe0rGJjcZWfQPtY6QCWpaRqcGjIkAAdQiRHWt2HVY3BwmdUJ6g8StBgiP20CpI0JkuIkQL9YQ9vw7/EMElkR7//Z0DWsd0UU2VQO4F6X0Jf648ytLc28ScePnuQHUx6trDSH/kPgRqhB55ZVXMHPmTOy+++7o7OzE+eefj0mTJuGTTz5BdXV1mLsuSkwFSUOyAk1Vdkig+5sGK0LCsoYEhSrpVpgiRJYDhEcnK6oMt6y2NroRG/kSIV5DSb38zrS+i8iiYCJCRBYMNxHCLl/T3iGdrjGFdZBVRUmFHoorCSUW7ddLdJMJrAgha0jhUWZZlpWvna1evRrNzc145ZVXsO+++7qun0qlUF9fj6Wr1qGuri4PLSwsdAum6aYM50UI36HxD7XfAd/LNEu+hI9KhIgEiEkHKrKI6OSfUAmP3OKEFcz/c6u7Zv8OaLDyarXwMqcv2qZurRcv1hBWhPAWRMCs8rRMHOkKEpUFU/bc2rDPr1e/DRkm11FXhGStSZJ7W3Rfu4kQsobkj1QqhUHNjWhpaXEdv/PqI9LS0gIA6N27t/D7dDqNdLq7c0ilUnlpV6GiZyEBbMdWdupGhiyVeylhIkK8vMGJwq5N81GoBj+ZCOExSbmvmvt3cyBVwa6nc8/pipCgEIkQFpUAFLUpDAuJF4KukOt2HU2uW041cYkg4fdHIqRwyZsQ2bRpE84++2zsvffeGDlypHCd2bNn45JLLslXk4oG1mdCLUgAPheJG2Gad6OKgFHhJkJU9WBU8J0oOyCZ/I5HNWeuqn5sWvnYZOAydUDUDUP1M5j5ReYXwiK7Ft1OpU5BwvuPsPsJWpAENUXBR+qIcBOhOpFkQWMqQnT7JRI3wZA3ITJz5kwsWrQIr732mnSd8847D+ecc07271QqhcGDB+ejeQUP7zMhEiSZKBtn1VwR+bSGmIiRKB96XoSo/A9MrBu6Zn03p8TMOuo3xuzfChES9Jy6ThSEKAwVcLbbTYR4QVQagf1OhGxKRu/6dK/jyJDMWUcAsSDR9R0SYZp8rCpRLjznXvyNZNfJaySZG6b3sJ+XIUqOFgx5ESJnnHEGnnjiCbz66qsYNGiQdL1kMolkMrxwxVLAVJDEgXxGwahwc04FxDVheLym/rZxG9R0BwNxNlxxu3Q6b1Nzvu6UiSwM1SS5ldcpGZ0wWrcpGR0RkrNfxkqiI0iCoJ2bwg0TN5EY1BSajmOuqO8wKVLp2B+JjlAIVYhYloXf/OY3eOSRR/DPf/4TW2+9dZi7IxhEgqQ76VC3IDEhzNok+YiCUaHsmDZbQxwmZA/FzVh0By830aErKPNpBeFxG3TsyI/M/91Djf2IEJOpD1mYtI4IkR0D33Y3QaLCzZrmlkGXh09qJrOKqFAJED/iwz5WkaMqi+heZq0WsufcrU5Ud5+Zm7iNrCL+CVWIzJw5E/fddx8ee+wx1NbWYsWKFQCA+vp69OrVK8xdE5uR+Y9k/vYmSGzCeACjeKBlCctE58WL+Vj37VnpWOrRghW0ABFZRUyycuY46VZ1F3P0YlnQGdxyyh/AzNJgMh3jJqLY79lBWyRIAPn9pjul5zeRnBcxEjQyEWISpqwjQHSjhjL3e+EUjCwEQhUit956KwBgwoQJjuXz5s3DCSecEOauCQZ+6oMfTMzSOxfXA+g2JeS3E1YNXEEIj3xlRo0LXv1CRGLEbf2cZR6vpQyRKJH5kci3kSuG/FTmFfUFJmJEJzJPhk4kmEqEeK0GzEfEAaL7rHu/trgrtr4wSkKfmiHigY4YAQSJh1yyJhaqWdLtDYk9D/y0jBu8CV8nnNZPwTIveO1A81WHJgy8+O2IBscgREjuNt1FiVs7WPi6QoC/+0fHyb27XWIxoqqJJd6OOhdOWInLeN8mtq4PkIiVf12xUFqvUyWOmxjRJcw3gXz4iGjPE3uwhshEiCqUVkVYFo+gBYWuVU321utlWkZnP36n0mx0p2G8JvYSbVc0daNC5OzriDzSnKZRXcucejMGx5BZbnadZcn4/IoQ2bMueumw/7bbno3qKjFrZJjQmSwhVMXbMn/LBxJVJxaUVUQmEOzlYe5DNlfMdqSuOT80RIhqoCqWjs025bOlBoJONKaLn3wtJo6oXt6S3cKTZYO5inyIEcd6LsJE5hOjQibaVZliRcQxTxEhpjh6PsIVPyKEXYf1Hg/KczxfnYWJCHHOF+s5RALujnU2xSI6dJGJEdNpDt7srytyTNKwmzihBmWmD9IpNCgx4gWVMDGZwlIl31MJEL4PkkXLmFqDg7bYEU5KqzcsUbyUsedh55uDtIyYiJAwxI5bjR63DKqiiAqVCIm7AHETpKbtZ4VDUPlRcn+Xm7nUK7rCKGg/gaAjU9zCoHXEiIkTu3QbLlYf1fq6AkSnT3CzjrBC0E3cko9I8MS7VyR8oyNCdDoIkSe+yFdEVyx4tYIEKXa8iBA7WZksCqNQREgQlXLZ41FFWrhFUpgKEJUzJIupMNERIV58e9zOtUnitswy9+Rw9m/ZbLW5z6+eGNHFVcQaDOBh1I2R+cixgoudTmTxEjJM6ENns4hxq84LuIWsZeA7M9UUDb/fMDARI0GJEFZ02P83qcAq6rhUJmaZ2djvG2qQ6Aw8rBgJEr0CjuqoE9m69vZ5vEY2ecnXYypA+HVECeJkYkSFyYDrNy+RaH9BF6+TWUac+9Urrme3rRCjBuMGCZEiRT4A54oQ93TMuZ1ZmPPMOsiOr5rzWxEhEyGygnamWVR1KhibihDdCrdxMxub+D6YFLsDzJw5ZdM3unP/bufV67OgU0NHJkBYSx0vhL0miMtpn4fQX69TOmGLEHYbshIYqvIXIisNiZBgICFSAggHNoEIkXV4dhy97K027MQ+OuJC93sdEaIrQPiaMtqDmoEI0fHhYYvGhZUF04/A8SuO3HJY8PelehooeKfDIAW5mwiR+SqJRAmbA0NlFdHB5MXDVIToWguDHPRV0zQq2HaRCAkOEiIlRFBmfd0pGr+IHnQ/IXleRIioJoksPbjft1CZCNGZPnP7jifoqZJ8oGtd8RIyKvt9PlGJEF6AqCo+r2nvEFhI3Gv4uKEjRnT7GNPSA2EM+rICoSa/JYKBhAihTRCdmS6qB91vfgCVCFEJEF3s8+Qn1bbIb8Vup43KNM8iEkesFcWNOE33mGT4BLzl4sgXaidV8XVWVXy2aarumRUjInHs974UiQiTl5w4iBDR9nX6FBIg4UBCpIQIIhxPhcwqInO89GNBMek82DbI8oSYiJCVGzYGWp6dbZ+ojTphxLJicpnfy/0i3MSlrykZr34TASTS4glTkARhKZBNkYqmCVX3Z7+aiuyUIS9Ggpii0TkWFVFWgXaDREZ0kBApAYQVU7nsl4A4fl6nTgoLK0bcEgblq2iUrB32AM8O4ibl4VU1Zby2zySMmIc3zzu+EwgTVXt1HUdVg5lXh2YvglnXUqKKtvFj7QvCUsCjmiZcvyHtWLehJpldRyRGhG1m69FIzlu+rGF+coQQhQ8JkSJFFaam6hxNB1NRB2ySsVAmRnRCdHWsIW5+ITa6kTE6lpA4+l+oBiRdRIOV25t1PsUIoOdHEtb1MWmv3PFYI0SXEyHssoaaZI4Ysbcrq9Sr084gBAnl3iBk0J1RYPCDr+4bg65VRIZuvZQgkYkRUwEiwyThFStAeEsIIBZwfs+T6prwCdW8VJaVEXjWUD7ZmeaAFKYYiRK+baLrLLOG2IJjXZsiMogRI/a2uiNpvPl5+RUkcZ6SIaKHhEiB4FYQzkbHd8K0gw/7DV81RWPilKoSHyZZZFlMBEjQlUJ5gdhUlXA6qAYoPvKJLDuriLD9mnQIUpj5EUgyEbI+9SMa6iod39lTNXyIuV+8+JaQJYRwg+6QmGMaHRJFtckgkpv58Rdx90XRH8j4qRdVLRkbWbVQwLsIYa1VNuvTGx0iyBYlXqZcZFNwyiRsAkuD30HaJM14vgVJPurJ+HGeXZ/60fF/W4wA3aKFvZ9FTqte0LWOkAAhdKE7pQQJo0MPIlW0FzGim4VUBOucq3prlBWzsxF1yKYly9nf8e0X1b/gRYkJOlYcafs2i5FArQSa2TtNrCNBZHQNksCL2jEihF3WUFeJdW0daKx23htB+AjxBHEfFOu0jJcXwlJ2zCUhEmPCtm6YJofytS9NPwETMeJFhPB+Me4VYXMHbZ0U0N3LdOqTOP13+FTTrGVEt5KtDl79fsIauHUESVBTNfnMjeKWETdIWMsI69TKO60GlRNIJkaCsob4qbgdBaI+281iW8VNpxfS8QYFCZESxqRCatCoBh37wQ2iHosKXoywzqtuAiRI5zvRsbADrmnxOBPHY3v7cUIkWr06rUaNV0uI7Q/EOiU31CS7o2PqKoVWEZt8WUUA75YR3eekEMSITpVzEZl729nfFcLxBg0JkRjjN4OoX9wGPS+drKz6p0yQhGG6lVkaTB1Pu5eF32mwbQgqPDUOA7UOUTuryvDleCrNZcJMFzLOyf1qKrKRM43VCWXUDI/9u7CsIiZ4fVbiPDi7VTlX3SciEVfF9PtxPeagISFSgvh523abe+c7WH6bMkc3eUKoXOuISZ6SzG/5PCfdx96N2krg1efDDdWxyCwBOgKiECrzFhJB+3ioE6s5Ewt6tYoAmekZNoIG8ObcrMLtWcks83/vxW1wlllBZOUZ5ORej1KzjpSkEInbDa3CxCpimvbcRsexT5aB0e0hkwkT0ZuAKoJCxzpikp+C3Seg7/eh26GaVAyW4SuVtidTuViwmf42rhYMr0SRk8QWI+wUDZCxbtjCYl1bR7c/iECQ2NMzfASNbWXJl1UkaIthHPpvExEiStkvFoOJbDFRG9s6UghjlR+KXoioBgGvycHyjZsY4dstz6pazlgYnCGRXgYuWX0MEd1+GN7rXdhiRJSczSsmTnVeRIj9t1k+lHAHci+CDdB3OA7CoVQ1+OfLshO2AJEJAL7kgpsYAeAI3ZURxhSNjjUkLKIQJDrOqKryDLZVKmudyknDz0wJb/YfKQUxUpRCxOsbaByUtgyTNpkcv9ccDW7VYG26Mzpu1BIjbnklvPiN6KyvniIx2x/faUQlQoIeFMxEm3cxks+040Fg6hzshqj+k8hfhBcjIliriO4UjW5BQd3zH3ZNqXwM0rJnWOaUyt4TosKF2e+YmkA2duVuoHTESFEJEZXnsgr+IYnDBRfd+EG0SWZNMCnYxYsQ2YPGPmCykuReEHVszvDX8Dz4dYnSyRgozGRSphYIP4LE7bdubTGJMGPX1RUlMjECZJ4zE+dVv1M0uufX7QXCKzrPZlh9tn51b3lVb0BcuFCFSowA8XxZ9kPh9VYC3LyWTeBjum2iUtyydWTt0dmObnVcHWuJjtrPB4WYGCnHpCs5z14cT4PIdhsH3JyfwyBIEeIHmUOryHlVV4wAuVM0QRDmvaYTzh80QdWzYpGJELYuEMBWx95sNd78LLMvYHF4WQ6Sgu+pRCLEzWtZO7xRYl4P6gbw89YcxBu3yqogglf7NjK1LypHrmMVCWsQ9SPeosAt7A+Iz/QETxBTTKJBWPTm7qn+ScwL4/G4+YvoiBHAGUWTXebTUpkvwRv2FA9gYgEJ/t7hp8wy93+3A6tIjADFYR0paCHSlhUeuQJE5MNgwyeuWp92FycyUQIE778RxRsAuz/ZQ+bWebPmYhlBTM2Ynhe362MSbeR3CihoVIOwjhgoBqsJ4F2MxB23RIP8M2eaY8Qv+b5/whAjfgtrhuFczopD+94WiRGgOKwjxdELIXMz8AJENW0AQPimLhIloux3LH7fqGUDf1SDnizlOAuf68CGLxpno+sYZ+8vCIJ+OHOnUsSRSV62lfO9wdu6n5ofunVevGzTFF7o5jvbr1/45yEI0c1vj7eKiPKL+EX3ftK9ZzzfDwb3pJdnPYjK3uyxiZ5ZWVI6FWwUjW0BFIkRoHjSwxesEGGtIfyNvj69EYtXtwPImC5lD+jKzSbKfjWy6QMzQWJCGKa9INBpV055eqZD5GHnofmicV5EiM559/owBpHJNuzr2u3A6H96wg+6fkZ+4WvsqAjqHIQ1bROkf4ssikZ3isYvekUs45tPJggfEJPjs/tIXTEC5E6Z8WKEbWOh+44UrBDhsa0htiXEvtjrN6SzDyWb9IetUrl+Qxora5IOnwYnCeENEERNEZPf5iObpxvOTKzub3yyyrVe8g+4HX8hPoC68ANxFOm5bcK0zskS5+kea1C+M6ZixC2Pjh9EETQ2/ADHihEgt+YM7x8iw6+oEw3ScfFrUokQnb7RS3VvFqnVWOI8LEp3wFsyC913pGiECIv9dm6LEFuAtLaamy7t0Da/YkQ3QgIIZ97V75srnxZelJZYBC9Aun/fvV29/cvPc5APnMoqEmQiNTe8vJWHbRXRy8cSzFuwX6uEbqbgQkQ2RcOLEQA5he9UBFaR12W6gl1mUslaRhA+en6tH05HfudLA+9gLLIesyKEf3kT7Ssr2IvEOlJ0QkSUVAvIFSHs37Z1BEBO0h8AOU5DppiIEBl+3kL97J/tEEQ1atw6Lr8CJLNufkSIV3TPp/u0kzwJWNT+EqFaQSRh4jLrSL7wI4ZEg4ifgZ6//iZiJPub7FS0nu+KFwuGTITIcqmIi74FV9laBz8OqLr3BzuVpgqb1q0DxIsRQG4diUMf6UbRCZGmqoRQcdbWJnPESG2tnpkySKIUIfy+dTIm8qG07KChKhon2hb7exVeU6kHgamPiMn11HEMlRUkBDQiuzy+5ee745ftWyVIghIjYVhCZJF5OmJEJS51RQs7yDl+r5iGMckholsXSiRCZCHYgLrulAi/96JWevaABQiL7DrZ37HoTHvzQlFmrS8EMZIXIXLzzTdjzpw5WLFiBUaNGoWbbroJY8eODWz73Q9ExtFU9JCtT/0oFB52jQYTE2bY6ObR4G8u9dyn2lzqWFcj7bppum/x8ng9HG4iRNVpBVkbRSVGgtqHXjvckqb5s/SZ1Krxah0JewrGi2+IrmWLFy0yq4gNaxUBxAm03KwhMquNTIyIEIkQUUSRaF+mfYWfAdbt/vUiNnJzVnHXyMXiIRIgJla0QhUjoQuR+fPn45xzzsFtt92GcePG4YYbbsDBBx+Mzz77DM3NzZ63a8/l8/P2MiXJOm/xywCn2dIWMk1VCcf2ct/uo7mwujeU0OSomMsEzN9SbMIoHudGUA+XFw96UUpn6W8Vby4yTMSIn8E2KpHodZpKdS7y4feR7+kxXTEiCullIzREAkQ2KJqIEdE69jYy/4qFmm7BvXxNy5q8qIUFO9YYiQ/Fi2OhiJHQhch1112HGTNm4MQTTwQA3HbbbXjyySdx5513YtasWY510+k00unu6ZNUKqXcNitGukmgIZkREE1VG9FU3YE1zEMoyv7JIhIhDclE9kLbF1j0gMgSY/lxcuT3E5cbSdVB8McadsErwPt58eJBLxMhKmHnhVwH4fzgt0ig+/b1q//KxEgpoetbIhMjOetpTMuEEZXlrDLrTpD9hldrsSxcXoR4GsrdYsa/PEcVDRcloQqRjo4OLFy4EOedd152WY8ePXDggQfizTffzFl/9uzZuOSSS4z2wYuR7s7LKUhshvd1OrSKHgxZqKlMhIjKvgPim1/UuaqzY+pH5vARH24CKPftyjz1vZfvw8BU7fv1nnezLLHf+RUj7H6DRCQIvF67OOeMKAbUviS5YaKy3D4mviGiQVjZVymsIdnkkppiJAoHddUz7aWf1MFNhDgsHQVUksCUUIXImjVr0NXVhX79+jmW9+vXD4sXL85Z/7zzzsM555yT/TuVSmHw4MGu+xFZRuwBvyrRK+fiblVXrdV+XoBk/t+9D53U4faA5yYKwg+9tM+HmQOk/dvu/8f3bVS3HpAXEeJWidgvbnV/ghYkQYoQv/uNK346/qCzquoSdHRGnMiXNTiq6DRZhCG/rBgFSax6hWQyiWTSWyQLa4XgO1SvnbhqAA7qoZDdVF4Lv6msIqwYke1fFeUSZxHCI5uyMREhJtVwZYiEXq7jr6DTkQhX9rdBps42ubZBhKOHgei4griGKkTWCBMRojLdhyVmwhYhtqWVPTeyfToGX41p70IirCR3xTg1GaoQaWpqQnl5OVauXOlYvnLlSvTv3z+UfaoFie0sqNcJ5fuBCMoq4iZGALj6HsT9zVU31T4rSHRFSBiDl/w8y9vvWoTQg8COyhISFqb3aTip270JBreBqrscvNn2VWGiflH1TyZh1qogAJ58ixBVzp5S9N/IB6GONolEAmPGjMGLL76Iww8/HACwadMmvPjiizjjjDPC3LXyTdhLxxsXRa7rCyESIzYZvxP/GU112gD4LwrIoio8p2qrfnlveWpqHjcTriidfRgWJlUiNNF+ncvN2uA3s6xXS5+IKMRyUInlTN6WvQoSESqfOBFu0wWq39lWEa1tS65lPvtdZ/kKZ40jHQHi5d6Q5ZvxQ9xfIkWE3uJzzjkHxx9/PHbbbTeMHTsWN9xwA9ra2rJRNPkiaiERRZE7VRRPUNs2XdePKHF3JA2+RLjqDU/W8ehkkw26ndJkYCEnK/MyLaNb8Ve2ba8drVuUU77edmUixM2J3mTQEllFTKdkVCJElsxMNZjrbTf//bSfJIJBIqopUyqELkT+67/+C6tXr8b/9//9f1ixYgVGjx6NZ555JseBlQhv7s8k8ZnuNvKNrpALtDighpm5+81P4AviU4R4Ea9BZq4NG1ZoBOH7otq+TpSTboRT0On2RdMoWYfTgPw55D4a7gm0dEoziELNeUEi2mY+/PDsbanSKqjaL2q3jelUnyhqKEiriNvzH9dcImWWZVlRN0JGKpVCfX09lq5ah7q6uqibYwx747ulWjd1Eo3jzaSDV4uIyaDsdbrBZOAySbeu45zKtiNMgqxZlFmWX0dVE2uIbspxHpO3YC+CRGQREYkRk2kUU8dIUwEC6J97L7WX8pEvScdHzJOFT9JX8NdEZfGSJTPz8nLq5geWr7EjlUphUHMjWlpaXMfvwptMKkK8PvDFhNwh0/yh8VMVGdAbtFRvz25vkFFM0wHRWkLcorPyuW83THK/8OtFWZhQFzcR4tYfie4jp6+W2ok6qpcsmcO6yJkfMBBUGk66vNDkLSMyq4iXKRo3P6w4WkVKb8TLI/IbX+5cqPv2rJszoxBQ5+/IiAqT7LTexIszUZnozUbHfFpK87ompdFV34VxzkRtc6t9EgQ6Be2ECcg0pmCCaKNJAi3dvoj9TiRI3H7DEnZfphIjgF77g7ICypK78UI4CDHCv5zFTYyQEMkT/EAa5Fx+MYkSEawYsf8OarsiZHUy3ASJlwFV1akVsmUsjDDZoM6HTrXcIFKcy5w1M99VSNsi3pZahIjEjer3JtMwJsJe5xkNo4CdCapQfr32yy0+Iv+hIEKpvYoRgG1v8M78QVG4PV0BovNWH3Sp67gJE1UnoAsfiiz7ThfWGuJWrEvaJmmKfu9vT7rRJfrb894RyZKr5ctHxG+kjM51DUOMAGorSZChm7rbCsoKosLkd1H0UW79kNeXnjhOzRVCETwSIiGjyudhglcfCtMMozxxuFF1TMKmmHQwQUcwAHoWA7ZaL2Ae7lrIFhWWMI5D5jgYlhixEW0rH4OXW1QXoCdCgojAU20vn+jkOVK/9MiFuJesquy9F/R9F3cxUhw9Vczxm9jLzYcC0BckpvitbquD6u0jDFOiyEve6fme6USCzE6pmxRNNTcsEiQqi4R4/fyYZ3XrYuQjsy9vDRE5DgLdgiRsMcKTz1wVuudb13/Dj4XTa36hoPuiIKy0YRCErwhLnMUICZE8onOx+QdClX48nwNMkDerm8NYmIQRsaLqHEyzsvKZHO3fiwSJLm6Oa+6/V8+XK/1cDDvOsK04KnHJOg8GLUZ0z0PQvjVu+9VN+a969r2+aPl9OdLFrd/S3Z6Jw7xXeD+0MMVInIhfiwghwtTjATgjmfis5EOMhIluJ2I7/qnyOOgMSqJBRdcMryNIZPtgYad3vHRAOucsCF+RfHeOa9qc7bUr1YYhRkwGjnxEXYV1roMa8IMmamuHqvaPPHKmW5CI+gKd+0TaZ2zuC+JkFekR2Z4JIV5uBq+DgM4gE+YbQD5v/LCOI6iBY316Y/bjXN7hEC/tHV2OjxvsOo4EXxrp8s2SyHkb3KqSPQMfGMWhu+o5+zVtnVlxwg4aouthQpTh3Pa55T9REbX53y8mzwPvOKzyL1vT3iH8AHD0CWxf4PbsO557QX+RnZJmjilKwUYWkSLHdB4238lwwiiMx6MztWAnJTIp1hWkCOH/5vfPDoA6Kca1LDYCC5ofwRY3k6+ss+atIfx3TdU9pW+qQRK386UiH2/MXu+9OE3pqp5N06rIonVtC0lDMiHMxaMjUHItJNGH9RbOk0CERk7K+Qjiz0WdXL4VOi9GRIThWMjngHALw9R5K9ctoMVe26gyvkbJyg3d571fTbBVUFXEWYREMTD5ufd0K3D73bYpssR1Oc7SCmHsvg/n1K0MUX8geumMaoomvk9DCROU/4TqhtKblonWsSmo8+DmZCbycTCpJWOCvCJpfgZB0TVVR2UFWwE3n7il3mZFiC75jHCJG2ENUrL7z23K2a18Qj5rOumkeRfBihDV/bhyw0b0q6nIWuxEqBLoscttaworRqK2isS/NyEA6KWFj6KwkRsqIaHTxnw7teqYOHVFiKpzcus0Mt/5K2Qms4q4CUwdn6Ogk60RcvwOEF4G3igHJl2fN7d7MGwLn6ydrEVVlfFWJELWb0g71lnX1oHG6kT2u4aaZPa3rCBhnVpt3PuPhFCMRGEVoV4kptgDsKwgk59tesG0Y9LZTz5ylNi4TT/wQk8kIjxVwhT8ht2ujiDJJ6LK0Dx8KHEhiBG3FOh628ivNSTqefso8FT91qco9pII0GuAgMjvgxch69oYZ+nUj1if+hENdZXdyxhBYrq/pqpE1gct0+ckhNM0+RYj8e9BCOP6DTo3kGlMfNAipBAIK+JBZHXRcUB1Q+bkKrOKAPK6GToF6/yGBqsIcpv8uW2q7ok1bZ1KnxDbUVUWrq1jis9HsrZ8k+8BSnaOhQLf4J7RSQTo2J9LEkHTaRnbGiISIetTP2bXa211Wkh4UWKyLzkJ5lmOxhJWuE9ECRBUevh84paQzSaqSpAqAZbP2ilAriAJQoyIUDmuqgSIqi3sPHNQyJymAbPBW3Ud3SIXeHN35t9uEeKoy8KexwCsZzxhOmHq7j+ufY7Xe09HRMhEjqotLKoqzypYSwgvQOy/a2uTDqECQOg7IhMf7HI2Kiz7chGRSCYhEnOC9pEQbU/Wcfu9Kd1S08e5LHWxoRNFo1sgjg8hlHng6xKmTwpruVAlqrORWUJcM5Tq+g15Pkf5FyVxFSFe0bX6mYhKlQjh0Qnd5UVGquXHnHVUYoRH5AArWteRvTUCAUpCpAAwzbWhO6iL/E/8DChs+7w4ionanfcQXkWJb9m6Nvm0psgQiQdRuugcy4hEhLjNMfv1m/Cbrl4GL67Z/DBuYkQlQnTr/MjaFARRhlkXy8uCjuVPNzRWtg2/Pkkm2FE1/DI/5PPlkIRIkaK6iVirCD9V4Vb8yuuNKQsb5befD+Gh4x/jZdDwMrUTdG0REbKqnvI6OPKif3aiL5E/igl+RJupGJElq5PlbnETIKJlyno7BegTEtYbsdgi63whcnMat5fnLNP1D+FEtww+tbrbet1/5woA3URmDXWVOdYOr/SrqfAtRvJF4T0hJUrQvhe8GNHBrzoulAgLr0QlQnQiQnSEA9uh2h2nrCYLi8l8fVBWIz9iRIbUF0RzP0EThcVF1Rfkz48rV4zo/MYPbpbEzN8JZn0zC4jMN4kVC43ViayfSG1tEq2tadTVmzmmqrZv/x1HindUKGJMfC9U6PqfhNkBReUQF2Q2UZMBQyU+REnOVH4auuuaoLMNWfpzlTiIctqKFSOA0zk4Z11DARIGsnOlE0qdXa4lnvSfu6DLOqisIpm/9ayLQVwj2T0vKnpo8nsVduSWTUNNMhs5Y1tFbDHCUlubzK5j5xcBugUG/6Igiw5rqu4ZegkDE0iIlDhBdTAy/xC39PH8+irCECzsNk1FSRBhfCyiqBmT6Q/RlIPpdnQEjUkHFqYAMbGwsQOb1zfsIOvy6GJS3MzGdh4G5AN1VCKE3aaOGLHRudb5vj5+hT8bRr5yw8ZsbpB1bR0OMcLDhu821CRzRAhrfZFlYWWdsuOQLZiECBEaurH6+rH/4VpPVNvmOzVdD3zdeWYb3RBefnvOonjuBftsZHPwdmcmTyfdvQ/TcMeo8BOe7fW+M52OdFRI1gylZmGLoekIEjfCtIbKxAiQ+7x5EVS6uZJU4ls2penJCrJ58GenaHgxAnRP0cjyhdiWEF6EsC8IOi8LsmeY8ogQBYndmbRn/xV3pjx8ci2gMBz7vAwWou9k4sTLG4pOhlZVPgzemdOOLBFFzciSfAHxESCyQcirGDERwTpWQLffye4rswGw2+mWF0Q6xxNlhExQg6HIEVZU3FIkRsKq/6TyFwEyyc1sscHmFmGnYmQixLTNjj4hwr43/r0+AUAd6cKS73ozIhGi65Vud5SAvihRdaBhVfCVTTWJjlM1UKiynurgNqXg5oyp2o4ssoTHNL9GmAQx7eAV0fPnJnC8OGvLcrrIojDE/jviNN4q8iVC8lFLStRfysRIvmCtI6y/iMqZlE3r3q+mwjEVA4gjv3SqiEf5DNuQECkgvES6hAkrQuxOuL2jy9F5ijpMvqPkzckqRMet6jT95iXRFSHsQCGyIrDr6HZ4qnMhCh3126HwkSUiQRKXDizstzdVuLkXEaLabs46Cisie2/plI9na4rYdOfKEYv6QsgV4tZG2XPO+wqJxLve1KjcH0u0nmxddgqUFSP2NA1fU0bkD8Lvhz0eL9ZV9r7IVy4REiIFhs4bRL4tIbYVZH26wyE+1rR15sSxs+mI+bc3Pm14GOG+psnhROgm/7KRRZmI8FKfxCSxmlvVXef+xW32lFqbGVz9CBhTh8WcdvicOgrLAVLULtX9JSsfL8uwCTAJugosjN5LfyYqjyFM3ihI6qcavINwJs9sRyBKNosLVR0k1mfLTYTooPMs5kOMFM7dSGSJ+o1FJUIWr27Hyg0bhZUkAWD9hsyDspKZ4wT0Huh8v72JrCG6adB1cLMumEw9qJLS6cJnlQ3K4uE1ksg4nXvA94LugJ1vvxi+YBqLaFm3Ra672mocUL0MBPFcy2p1qZKnsYjuW9FA76U+FOuTwjuxyhzEWcIIvWXv93yX4CAhQhghEyFfrmvDmrZOfLZiQ7aCJB8DDwDra5NCb3BRynB2EPAiQnxZPRQhyCK0Mye6FFID9EJHTb/XgX9j1P9dsANxUAJEN2pCd1owKMKySGTzUHDmfPveZMW+3Yao8viELULYbbllcrXh72PZ/cALFJkVwk2g8A6ybgUZ7XX4bYRFPsUICRHCGJEI+XjlD1i/IY2vv0th2dKWzHqt7WhLtaG6rrr7x4N6O7aV9RRXZP40FSF+nd9MzO+qCBMbVUl59zTi+RskvOZUUUWimFalNauyG1BkRZ4tBGGKEB6+KmscyJcI4bepso5klmmm7FcUjGTRCcUXiRFd/IoQWdi9KroqLDESrzuUiDV8B5L1CWnrdIgQW4BYrWsBABta16Kstg8AINVSBQDSGHkbVWcdpgjh0Q1FBtSdSBgiRHYegjgHbtM/nrYZkAjxI0D85BIRkc9pGd6xsZBqicQBnSRq3cvNpuT4rL02XsSIDmGIEBs+3DwfYoSECGGE0xqScUxduWEj1rV1oLU1jfbWdmxY9n9AyyrH7ywAZbV90N7ajrr6TNZANi5ehknRvTBDdXlkYX+yQmr2bwC9qRg3TCKFbPyeH92pDm/b9ueEGgVhixA+rwsgFyP8dIyNzBpin++giloGQT72bSJG3BBV6pYlCHQjiBINbo6qppY/lc9I0JAQIXxhW0PWp35EquVHtKXaPG3HTjWselhMRIhp2ngvHZEoVl8V7gq4W0FYgn74dTp6N7EShhhRRwT5DEn22V5xGK93ASJ3QM4NKwX0xQjQ7ajKR1zYSejcBqpiFyEq/NwnQVnZvIQEu+F32pG1joQZ1huaEPnmm29w2WWX4aWXXsKKFSswcOBA/OIXv8AFF1yARCL63PaEGUFNeVTVdk/N2G9x/JSG7G1Nt20mHYpsXZOOxcubSFzS2vMEEd7M4mc6Jqzj9jJwBDHQuFt93MUIgKxPkgw+xJNP5W36fAVN1KJDlgIhLDHiVrU3d30vUzXyF54gEDk3BylGQhMiixcvxqZNmzB37lxsu+22WLRoEWbMmIG2tjZcc801Ye2WCIkgMyDW1iaz0zK6jnS6N7zu1EpYeRSCdMa0EYmRMD3YVdc5SGuIjkUo9zfmxxyEVUT6ncQM73UgEBXnkyWZkwkSZ26ebt8kr9ZGwgzV9Ixu8jSZGBFZQ9xEiFvuoJxlLjl/wng5Ck2ITJ48GZMnT87+PXToUHz22We49dZbSYgUEU3VPbGyJhOS21qfRntrNdoAoLZP1lm1rLYPquuqUVVbhbr6yqw1pF9NhcNs7KXzlg2aOum2AfFDavL2q9tm3TojUSeaypcI8YozyiEap1UdHwA2MZ+9v9w2iKOUZPlc3FL4qwYpexsiawiJkG7C9IMSoXJm1ZmCMU0D4AZ/b7slmAzqhSivvV5LSwt69+4t/T6dTiOd7g5DS6VS+WgWYUhVsufmTrECa9o70K+mAus3JLC+NgkM6o0VSzevyITt2iJk8Bb1aKxOOGoliHKHeBlk3DoQfnAwhX3T8VpKPqdNggJpuh1IkFYRuaCLxjlVh+5cEP6tD9r79Jiczblf0VSdPJxUlGDOS10hEiHxRKdgpep3gF4EntcikH77TR3yJkS+/PJL3HTTTUpryOzZs3HJJZfkq0mEB+wbOnNjJtBUlTEfrqxJYusBdVhXJ36YajcnMmusTmD7/jXZipG26g9y7jpncFcMHqKBX/VwBiVA7H3rtklmDmUFhNdBRSRCwnwrDNrqky8/Gv4+0ikoZopufgtVFlDVm3Eco4+iQDXV7NWBXebjo4tOqC+7bna/EhHCX2ud4/Ia9eOXMsuyLJMfzJo1C1dffbVynU8//RTDhw/P/r1s2TLst99+mDBhAu644w7p70QWkcGDB2PpqnWoq6szaSYREm3pruyN3J7uxPLWH7JhvHZtmfUb0sLy1dnpGEaE2FMyKhGiGy2j8g9RWTJkA6MX073JIOvFh8VkIPESJcN3Um7nQDcJFL++6bn1m2PEr1My2zm754XIHST8iAGvbc8V2PEJ1Y0DXn3eVAO5yUuQDNX9JfMHcbvWPKrnXJWE0MSqlkqlMKi5ES0tLa7jt7EQWb16NdauXatcZ+jQodnImOXLl2PChAnYY489cNddd6FHjx7a+0qlUqivrychEiNUFXdZQQLkhhPyAgRwt4SYpnFnRZIOur4bQWzLdNtBR5S4RcSYprV3tkeRKttQrJjsS72eer7fiwgBnAOFrKKyzDfDbpcX/FipCrXKbtgEnYXZzfkzKHStICpMXzrYfQUtRIxtpH379kXfvn211l22bBkmTpyIMWPGYN68eUYihIgntkkz94bvTmNuT9cM524TkQe/38RK8lA89zdvnQFNO+1zCE6mKn8RLxElpgJE1IEGOTWl2o9qf7qOvfkSIez/uyNaOnxN0YjwkoJfdk+QCMkQeLh6wJl7hfsIQITY6/NTgKZ9XOzDd5ctW4YJEyZgyJAhuOaaa7B69ersd/379w9rt0QeYMVIZn5+s5hI9HJ1ogtKgIjaw27PawE3FX625TlCQ+arIoz2MY8oMREh9vIgHdfc3hZV+/NaPM5LOC4gFyHOdeQ1k4JEa76fRIg27DkxESWuVrcArSGmvj/8dZblTgHM+ssw/IxCEyLPP/88vvzyS3z55ZcYNGiQ4zvD2SAihsgsIyaqOkjvfbeS3/L2yPcbVu6JIFBNf2S+V4sSP51nPrzodfdnKka8ms51RIhwfyGfK9NBgUSIO15FCaAn5k1x82vTnXZTHZcXYRvkvWTsI5JPyEekMDDJbBq2w1zQRe8Ac0GiI0ICCQP1kbFVBOvzY7JvP0LA5DyoBnRPodKa++adB3VTcIv8oDJtjUYMkAjxh9epTS+RMDZBiRAZXvtLnX2E6iNCEDz8PKtuRxtmyW+2PX7xMjfPIxv0THMHAN0dligSKIjpCrcQPl+5WDzOofvZZxBWkO5l3gqTeb0uQUEiJHzcRIjs3pH5FYUtQvj13frLMO8hEiJEYOjMSeazQxTtK4gKtIB+SKVowDN9w7ZhO6zM392CxK/5nxcIMjESRPbGMJAN9F5EiBdxaEJU9YOI4BH1A7wIYZ/vNe3Oe6upKuFwdFY5Obs9a0FMb0dFPHoRoiiJYwcYlDjRSQWtG3HBd048du2QsCMzRGIkDGT1VKTra7bDr1+OmwDxW6Y9CuL4DBYLoikZmQixn3E7tQGPLUhYMaJ6wSi2xHQkRIiSx2sIn0ldCre3I1EHZafAt9djBUlYkRkmCcfcq8nKrUf5Ej08UWSNFBG2VYQESLCoazA5nxX2WecFiJ1bSQQrRnjCtoZEDQkRgtiMF/8SHTEiezsC5G9I/HdN1T2xpr1DWf49SNzyp5hMyfipcxEUOgLEJMV2EAQtRgp9MIorqhIIvF8IL0J4AbJ+Qxoqul82VFM0xZchl4QIQQjQ8XcRYTK46rwl2Vlp17R1Zi0kbvs3xc0ZNwg/kCjEiBfrh1s11KCnZ4KoJlwMA1FcCUqE2AKELX1h01CTBOB88dC1eBbLtSchQhAaqKZvvJQO1xEh9ve2GAFgZBXxMrB5LfgVB4KadvFaDdUNt8gZL9WEoxqIVMK8WAZHHREi8glh624ByKm9xcIuZ59zGcXmG2JDQoQgDJCnlC8Hm52wPd2ZjTzRNfnzZlv7TckWIzKrSJD1TNjf+xUj7NRDmMImLL+PoAQJ63SoE8arO2UTxYCvYxkMoiJ01KhyI4lECDvtyhb/BDJiY33qR+F+GuoqPbexUM+tCBIiBGGIqoS4Cta031Td02GKFc0d28saapI5lhHWqS3fmU5NCKLei3L7eXY+NZmekc3z69TKcRMjcRAhsmvLttv+TSENmroixG0qxrZ2yESIHwrpfOpAQoQgAiIIq4gbdmeXL6fVsAhThIjOs9r3Q30uTa+dKPJBJBYdUUMxyccio80xGMurtmbu+1y/l7Z0V0EMnm7VvIOwgnilWKdlABIiBBEoKjFi01SVyEbO9KupcPUTWb8hLbWKAExiM4XZXzUIhJEWX0ZQjqk6ieK8fBdUXhY+JwQgzoRrI7p2MqtIvgd0kQiRXUeRKCkUMaIjQnR8QWwB0tqaa+WsrU1m/99QV4nGavf7Le4iNQiK/wgJImDcBm6Zf0W3z0Hu9ExDTVIZ2meLEUBsFXGbnpENAqpjCcJPJNs+wzovqmMxFSEm8Flr/W2rW4yw22Tb7yU9fz4HdJUIcUtCZx9P3MWI21SMjhXETYDwsCKkoSaJfjUVaKruiaaqBBqSFWhIJqTPQNzOXxCQECEIDbwmOxNZRWyzvYlVxEbXKiJ6m+YHAZ1j8itGvBaak1kPwhQh/DZ1nFVlWXFFyedEIocXkLwYkVlF8jGgK6djmOvAnx82K6jqfowDpiKETVDGixBbgKRanFMydfVOh1Q3EcJi3wtxPHdBQkKEIBQEMW3B5spg/Q3swUrXKsKiYxWRiRHz9nuLfFGVRJf5ceRsI8K6MH62zWbDFaXmz/wtrhVkYhkBgn1Ddrs/2GsqOz/dyxMOMcLuIw5v9ab+IDKHVFuE2AKkvbU9u82q2iqkWn5EXX0lamuTWpYQQG4RjMN5CwMSIgTBEKS/BG9NYB1XAecUDZDp3EzECIvDFyGkSq+ytzJh4S+JCHHz1dCdEvEjEkRRL7oJpExT9LsJEp3ChW75RfwIEnfhIQ9b5eGPL864RQCJpmNsZJYQwClCbHiLCABs378m+9zLREipWEMAEiIEEaqzpmyKJshidapQ3ihM4m7l0FW4+WiItqMTTusWdmtav0dUxIydXrPzvgC56fn56RqRGJE5rwL+BYmJ+NBBdV6dxxSP6Rmd8FzAPTKGhZ+OATLWEBvWGmKLEHYqhvUJkQmQYrWGACREiBImX9EiXsWIyG/E9g9xS/cellUkdz/yUE6RCJENWrwI0PHRMEm3HkRqdn4bMhHC/s0movMiRoDcqAm3tPCm97WO8FBFO6lEXlzz24jQccIFxFOodfWVSLX86BAf9nJWhDTUJKVTMaqEhMUsQgASIkQJoq6kqV/szgRWjDjxbxlxe5PP15uoSoR4EQJBCBAdTKcS2CkZVoCwg5MsK65Oin6RzwggDuP0U6vGbUrNDdbfiT+HIpEdd2sIC3vvyRySAXVmVDtUV2YJKXUrCAsJEaIkkIkPrxEhXgZ3t4RnpgSZ1MxLkT+ZSTsIESIiiO149V9QWUOA3My4bLi1apuy9rglQAPklhK3+1JlxdLBGQGWm+xN5utgE8fBVWUNEeX9EV3bhrrKbAivLVBsK4gsPLfUBYgNCRGiqFEVrnIuU3fGsrdRL2Kk+7f+rSNBOAaKOj23NPYyAZfvlOu6BOVAqXo7lsGHXIsQOeq65YYxnbrxawFh4cWI6Hu2bXY74jTAql5CbIFlTzupRL8tStZvSDsiYgAIQ3NVIiRO5yefkBAhihIdAaLKcSF7G9XNfumGzDqSQeao2f1Wzg6s7ECgirxQOTmaWER03qjDDKs1xa8I8WuJkYkQ2TU0RXfqJmjc/D/iHvUhypHDZ0EGIBUjrJ/WmrZOof+WyiGVBEg3JESIokO3LoY6rNTZ6ZhUT9VFZB2RCRL77YyFH7z8OAaaTsV0L5PnCtElzHBP0bZNInK098M5MLpNy+gS54KGMmROl3EebNlcP27Tpew9lQ3BZ0SKSGCqfEHifF7yBQkRomgwKd0N6EVzAM5Mkfa2dLJf6uIuSAC3aRudwcpPO/34FciiKtxEQq7vgX7lWxMBwn5vKkbYrLi8+GAtIbZ5nm+falojH7ADsNt65tsurMGWFyMAlNNPquXsNSxEYZZvSIgQRYFJlkQbNkeAG7wYCQOZ/0iuKNHDSy6K3G3oCZBcc7ZzUHezeog6dFH4bnfIq14YsGr7fmDrBKn8P9ikVXz7dKbUgoC/1jnTEQGHecfZ8ZL3fcpJOujjuZOdxzifj7hAQoQoOtxECF8zQoQdbunMiBm+GAFyU6rnigl9a4SbQ6NutlQvNWPcLAy64kAlSHR/GxRsBAXvI5Bdh/MRyLQjPAGiOwCKlocVrm4Tx0FXJEYA/wJNdY7ieB7iBAkRouDRqRDKZ0kE4KgbwePIjCnwmM9HwjDZgCF2SlSLE11BoutTo4uJEHArcKdThM4Puttl7weRKHHzFwByjzXIe8lENITpSBrnwVcUFRbUuYjzcccVEiJE0eImQmwBIqvtwooRmyDfsE0q4bq9xerO9euEfLLrAWqnXr/nQ2UR4Ofp2X3mMypHNR0kEqkqp+KoBYhp9WUvvyuUgdgtRF3n90QwkBAhSg7ZdEy+kOXtYFF1kCpLie60jcqi4+bYyyLKf5EPZ8sgxYjudkyie1TnRCU+grZQqAZLrwNpMQ3A9rEUi7gqVEiIEIQEWSIq20+EHcx1I1J0OzQvmU5FmORGMc2OKqqVErdQU1kRPTfxwR+vF2db0bkQiZB8ig8il3zVnCLkkBAhCE1Yp1URYdZ0kb25iZIymSJzRA0iRbtI5ESBfvVf/WPVCd3MLjN0KNWFRIc/dEWIKOkfERw9om4AQfhF1kHI3s75yrV8/gc2PbMNH+JrD9amA6zft6/qZHnO8bKDGT/gVSXKsx8dwhAh9v+9+EOoHGSD9NdZn96oPNZcC4nYAhKkCLGvtepDeMf0WSTLSXiQRYQoKmQWAjZNM1vACoC0iJUIUSgvP0Vjt0OGbodmMtCIUsbnC9G0jHzwDbZtumJEZRHxm8JdJvK8+oKQwIgGsRM4XYt8QBYRoijItRJ0DwL2YMXO8zdV98xaRvrVVOT4gjisIZudW20Bk3l77sD6dIfDMuLwsQjg7akt3ZX96MBbRkwtEEFZQ9z2a9KuoHxOgs4nYiOzgKisICRCokeVhVlnOREseREi6XQao0ePRllZGd5///187JIoYRwD8uaBghUjTVWJbNglO01jCxKRg6pIjGT+zYgR0VRNe7rL9aMLK0p0I2oyf/dUfnRZ097hqfKsvJ367TGZWlLhR4zoFqjzYwUhEUKUKnmZmvn973+PgQMH4oMPPsjH7ogSJTdjYndVW7uIlVtpb1Vor+o7fqqGbYMKlRjxahb248DKW0PcxEfQ0TKqLLJ+9sHXDNFNG6+bE8RPSC4JkGghq0f0hC5Enn76aTz33HN4+OGH8fTTT4e9O4IA0D0Yq8QIgBy/EbaGiAz1986B2cQfwi3JWGYdvUHLrb4IjzNniFyErGnvEAq4sOBrf3hdV1bEzGs9nCBECAmQeKAS7uQjkh9CFSIrV67EjBkz8Oijj6Kqqsp1/XQ6jXS6O8tlKpUKs3lEEcJaRURiJEP3IMQPPmvaO5RihM0tYtejsQfsjKWlI2sd0UUlWnTESWY9dYepU1/EzkCrqsNjH68tRtjj9WKxcGu3W2p7+XbFlhW3qqpOYeKeG8RPRAwRHbJ6M4A8DJ+uWXiEJkQsy8IJJ5yA0047Dbvtthu++eYb19/Mnj0bl1xySVhNIkoEXoywqAQJax3hxQhbj8b+v50CXiRGbHQGOl60sIOdzArgt6ouu+32ji5HGnwgtw4PL74yx7JRaEVwq8OTj1oosgJmvCABnOc/iNwgKmgwiw+yFO8kQvKP8ZM0a9YsXH311cp1Pv30Uzz33HNobW3Feeedp73t8847D+ecc07271QqhcGDB5s2kSByEoCx1hFALEhY3CwjMvjBWR02KhYsKmuKyHrinJIwFyQsa9o7hMUAcwSJxymafJm6ZVNToukbU0uOjgAJM7mdCtHASoOoHLcU73Tu8kOZZVmWyQ9Wr16NtWvXKtcZOnQojjrqKDz++OMoKyvLLu/q6kJ5eTmOPfZY/PWvf3XdVyqVQn19PZauWoe6ujqTZhJEFlWoHp/WnHXW1K3SC3RH34jKvpuiG92h8lNQDYJshWK+MKAtRFZu2JhTDJBN9NZU3RNNVYnscTYkE65+E0FMV/hNKpVvx0T+mMMc2FTnhgZUIt+kUikMam5ES0uL6/htbBHp27cv+vbt67rejTfeiMsvvzz79/Lly3HwwQdj/vz5GDdunOluCcIzIusIAImFpLuYGuvEKkIU5hsEKisKbzkJIlpFJEKE621Iayd+M8FkkNRdVzYoi8SQH3GicqDN+CY57zlKFU4QuYTmI7Lllls6/q6pqQEAbLPNNhg0aFBYuyUIKW6ChBUjQHf0iD1F06+mImeKwv4eCMYa4oao2m3mGNR+GaWGblVVQG+6iBcrsvo8gHP6TFQUMd9ihIQPEXeo5yJKDpX/CDvA2IKCtYqEZQXJ7K9bYJiWuPcqQljhZSOziOhaQ/JRYVaXoKoYy0I8Rf48omW2dYQsIwSRS96EyFZbbQVDdxSCCBVRqC+AnCkaQJ5fRGUNEVku3MJDdb6zkdc4cQ/lVU1HqMSW7R/CwvuHmLYnn5gM/jLRwkYbAXLRKKpHlM8KzSR0iEKBLCJESSPLxuoF1nFTvo4iPfjmpGs62AO/iZNqzjaYY+XbZU9F8bDCqyFZkfM7L+2Jq3VAFt7JwooQNvyZzykjEmpBHreonXE9rwTBQ0KEIFyQ5RbhrSGAPPumLrq/8SNA7PUdzrrZBF/qqSfW8sOnd2fbZNqeuA6aOmIEyE0Pb4dxsz49YfjxuLWNrCNEIUBChCh57MGGn7LoTgPePcjw0xKqgVkHvwOTbMA39Y3gfUXsBG88vNUnCBHCtzGug6afGj5B4sXPJe7nlihtSIgQBAdvJQC6BUe2Jk2Vc9BWDczBt09PfOhsR1aPR4ZIbNnHGpTvQ9wGTV2rSNiI2qAjjNjrElfLE1HakBAhCLhbRYDuAnnd37lbBzJ/h9vxuw0seiGszno8MkEiS1rmJQQ2tw1iK07cBk4+1FtUs8iGdeT1Ik51xIduLpPM3+WxPa9E6UJChCAEsFYRmYUgH9YBEX6TetmI6rHopj0PWoSw64gESdSDJi9URY6+shwvQPAihM8IrAOFEBNxhYQIQXDwjpyAu4XAi49EVOm+AbEw0IkWCsrao6o0LBIkcXqLF90fGcSOyjJxqjoWnbIEbuHDbJtkIcQkRog4QEKEIDYjyivCT1nwqARIFB28FwGSWR6eCNGdSuAL+ckESVQDp+z+YJFZPthj8NN+mQgRORY7SeQtnwlBmEJChCBc4Cu26mQOjZMIcZsW8ZI3JQhLiGxKgR0wM7/pipUYsdvAtkksSnLb6KXd/HQMX6jQhq+L1FSVcIQR22KEh6wiRNSQECEICaJwzbAGGz94jaZQ1UuxCaKonnDfCr8GPhspkCtGokZWt0i1LkEQYkiIEIRH4jQw2ujmudARIfZyv2LES/4NkRiJI3EVGWx0F6DO6BvXYyBKh/g+4QQRA2SDaFxESJvDiVE/rJOINyp/JRY28RwvPuzvbcKybhGEX0iIEARDXJJXmSKKqggLPyKMT55GmKOTeM5GluPGvoZkDSHiAAkRgnAhLtYPniAFk0wYBPUW7ahurCFG6O1dL0rHTZDkO8cNQXiBhAhBcPixivBOjLrr5xM+Cii7XDH4qwYx0TGIjl8mRnTaWqqDJy9GAHGEjgxZeDlZQog4QUKEIAR4ESNeOvd8hk6Kw47tqBR1mnARqnbLiu7JBlM+h4hz/9GHRkeJSYQOTxyiugjCDRIiBCGB7bBZURJEbRd+fS+Dg8h0n/l/t7jw8uYs/t6fGJANpnybVfv0st9iQnQ/xiGLL0H4hYQIQWgQREcedi4MPsJHd3C30YkO8nsevL7d00DqhM4HUUyQECGIPNCdHdNZCdUmCKsIv01T8pmUSzZ1E+Y+CYKIJyRECCJkTJN5mRKUgyz/+3wKARIdBFG6kBAhiCIhqGkTgiCIfNIj6gYQRLHBD+g60yUkAgiCKFVIiBBEHmDFSFWyvGTzYhAEQfCQECGIEBBZOGQChKwhBEGUMiRECCIkdAQGiRCCIEodclYliBDhk1CR8CAIgnBCFhGCyBMkQgiCIHIhIUIQBEEQRGSQECEIgiAIIjJIiBAEQRAEERkkRAiCIAiCiAwSIgRBEARBRAYJEYIgCIIgIoOECEEQBEEQkRGqEHnyyScxbtw49OrVC42NjTj88MPD3B1BEARBEAVGaJlVH374YcyYMQNXXnkl9t9/f3R2dmLRokVh7Y4gCIIgiAIkFCHS2dmJs846C3PmzMHJJ5+cXT5ixIgwdkcQBEEQRIESytTMu+++i2XLlqFHjx7YZZddMGDAABxyyCGuFpF0Oo1UKuX4EARBEARRvIQiRL766isAwMUXX4wLL7wQTzzxBBobGzFhwgR8//330t/Nnj0b9fX12c/gwYPDaB5BEARBEDHBSIjMmjULZWVlys/ixYuxadMmAMAFF1yAI488EmPGjMG8efNQVlaGBx98ULr98847Dy0tLdnPkiVL/B0dQRAEQRCxxshH5He/+x1OOOEE5TpDhw7Fd999B8DpE5JMJjF06FD83//9n/S3yWQSyWTSpEkEQRAEQRQwRkKkb9++6Nu3r+t6Y8aMQTKZxGeffYZ99tkHALBx40Z88803GDJkiLeWEgRBEARRdIQSNVNXV4fTTjsNF110EQYPHowhQ4Zgzpw5AIDp06eHsUuCIAiCIAqQ0PKIzJkzBz179sRxxx2HH374AePGjcNLL72ExsbGsHZJEARBEESBUWZZlhV1I2SkUinU19dj6ap1qKuri7o5BEEQBEFokEqlMKi5ES0tLa7jN9WaIQiCIAgiMkiIEARBEAQRGaH5iBQKbekux9/VyfKIWkIQBEEQpUdJChFefMi+I1FCEARBEOFSUkKEFRntCjECAFXJ8uz6JEgIgiAIIhxKRoiIREh7ulO4blWyZ3YdW5CQGCEIgiCI4CkJIdKWFR5OAdLe4WYVyQgSEiMEQRAEEQ5FL0REIsQWIOvTHZJfJVCVKEd7uhNVye5TRGKEIAiCIIKlqIWIjghZn96YXb8hWcF8lxEj9u+rSIAQBEEQROAUbR4RUxEi+psgCIIgiHApSiEi8wkB5CJEhsyhlSAIgiAI/xSdEFE5puqKkIZkIvt/1keEIAiCIIhgKapR1o8Isf1DbGz/EADkH0IQBEEQIVFUQoTFiwixLSG2CKlK9nSIEIqYIQiCIIhgKRohwltDAPc8IawVhBUh9nQMiRCCIAiCCJeiESIsIudUHl6EsFaQzL8kQgiCIAgibIpSiLihK0JIgBAEQRBEuBSlEMmkZs9YRRqSiaxVRDYVY/8m8y+JEIIgCILIF0UnRKqS5d0F6xLlaO/ocoTjOtYlp1SCIAiCiJSiESLVmwvT2dhWETYMl4f8QQiCIAgiWopGiADdYsS2irDJyFgHVnY5iRCCIAiCiI6iEiKAU4zY8KKET1BGAoQgCIIgoqHohAjQLSzsqRpZZlQSIARBEAQRLUUpRGxIaBAEQRBEvCm6oncEQRAEQRQOJEQIgiAIgogMEiIEQRAEQUQGCRGCIAiCICKjqJ1VCYIgiHBhE0kCFCRAmENChCAIgjCCFx+y70iUEDqQECEIgiC04UVIu0CUVHG5nEiQECrIR4QgCILQghUh7ekuhwhpT3dmS2nw36ksKARBFhGCIAjCCFtksDW8ur/LLMsUHu2SZrYmCJvQLCKff/45DjvsMDQ1NaGurg777LMPXn755bB2RxAEQYRIW1Z8OEVIe0eX42PDWkfY3xMET2hCZOrUqejs7MRLL72EhQsXYtSoUZg6dSpWrFgR1i4JgiCIiGHFCEHoEIoQWbNmDb744gvMmjULO++8M4YNG4arrroK7e3tWLRoURi7JAiCIAiiAAlFiPTp0wfbb789/va3v6GtrQ2dnZ2YO3cumpubMWbMGOnv0uk0UqmU40MQBEEQRPESihApKyvDCy+8gPfeew+1tbWorKzEddddh2eeeQaNjY3S382ePRv19fXZz+DBg8NoHkEQBGGIHYJblf03E+tQlShHVaLbIZX9P0HoYCREZs2ahbKyMuVn8eLFsCwLM2fORHNzMxYsWIC33noLhx9+OKZNm4bvvvtOuv3zzjsPLS0t2c+SJUt8HyBBEAQRLLwYAQSCxBYqm9elXCKEjDLLsizdlVevXo21a9cq1xk6dCgWLFiASZMmYd26dairq8t+N2zYMJx88smYNWuW1v5SqRTq6+uxdJVzOwRBEEQ08LlEVLChuyRESotUKoVBzY1oaWlxHb+N8oj07dsXffv2dV2vvb0dANCjh9Pg0qNHD2zatMlklwRBEESMqE6WZ8UInyNElDeEBAjhRig+InvuuScaGxtx/PHH44MPPsDnn3+Oc889F19//TWmTJkSxi4JgiCIPFGdLBcKDBIhhBdCyaza1NSEZ555BhdccAH2339/bNy4ETvuuCMee+wxjBo1KoxdEgRBEHmGhAYRBKGleN9tt93w7LPPhrV5giAIgiCKACp6RxAEQRBEZJAQIQiCIAgiMkiIEARBEAQRGSRECIIgCIKIDBIiBEEQBEFEBgkRgiAIgiAig4QIQRAEQRCRQUKEIAiCIIjIICFCEARBEERkkBAhCIIgCCIySIgQBEEQBBEZJEQIgiAIgogMEiIEQRAEQUQGCRGCIAiCICKDhAhBEARBEJFBQoQgCIIgiMggIUIQBEEQRGSQECEIgiAIIjJIiBAEQRAEERk9o26ACsuyAACtramIW0IQBEEQhC72uG2P4ypiLURaW1sBADtsMyTilhAEQRAEYUprayvq6+uV65RZOnIlIjZt2oTly5ejtrYWZWVlgW8/lUph8ODBWLJkCerq6gLffpwopWMF6HiLmVI6VqC0jreUjhUo7uO1LAutra0YOHAgevRQe4HE2iLSo0cPDBo0KPT91NXVFd1NIKOUjhWg4y1mSulYgdI63lI6VqB4j9fNEmJDzqoEQRAEQUQGCRGCIAiCICKjpIVIMpnERRddhGQyGXVTQqeUjhWg4y1mSulYgdI63lI6VqD0jldGrJ1VCYIgCIIobkraIkIQBEEQRLSQECEIgiAIIjJIiBAEQRAEERkkRAiCIAiCiAwSIgRBEARBRAYJkc18/vnnOOyww9DU1IS6ujrss88+ePnll6NuVqg8+eSTGDduHHr16oXGxkYcfvjhUTcpVNLpNEaPHo2ysjK8//77UTcnFL755hucfPLJ2HrrrdGrVy9ss802uOiii9DR0RF10wLj5ptvxlZbbYXKykqMGzcOb731VtRNCpzZs2dj9913R21tLZqbm3H44Yfjs88+i7pZeeOqq65CWVkZzj777KibEgrLli3DL37xC/Tp0we9evXCTjvthHfeeSfqZkUGCZHNTJ06FZ2dnXjppZewcOFCjBo1ClOnTsWKFSuiblooPPzwwzjuuONw4okn4oMPPsDrr7+OY445Jupmhcrvf/97DBw4MOpmhMrixYuxadMmzJ07Fx9//DGuv/563HbbbTj//POjblogzJ8/H+eccw4uuugivPvuuxg1ahQOPvhgrFq1KuqmBcorr7yCmTNn4l//+heef/55bNy4EZMmTUJbW1vUTQudt99+G3PnzsXOO+8cdVNCYd26ddh7771RUVGBp59+Gp988gmuvfZaNDY2Rt206LAIa/Xq1RYA69VXX80uS6VSFgDr+eefj7Bl4bBx40Zriy22sO64446om5I3nnrqKWv48OHWxx9/bAGw3nvvvaiblDf+9Kc/WVtvvXXUzQiEsWPHWjNnzsz+3dXVZQ0cONCaPXt2hK0Kn1WrVlkArFdeeSXqpoRKa2urNWzYMOv555+39ttvP+uss86KukmB84c//MHaZ599om5GrCCLCIA+ffpg++23x9/+9je0tbWhs7MTc+fORXNzM8aMGRN18wLn3XffxbJly9CjRw/ssssuGDBgAA455BAsWrQo6qaFwsqVKzFjxgzcfffdqKqqiro5eaelpQW9e/eOuhm+6ejowMKFC3HggQdml/Xo0QMHHngg3nzzzQhbFj4tLS0AUBTXUcXMmTMxZcoUxzUuNv7xj39gt912w/Tp09Hc3IxddtkFt99+e9TNihQSIgDKysrwwgsv4L333kNtbS0qKytx3XXX4ZlnnilKc9lXX30FALj44otx4YUX4oknnkBjYyMmTJiA77//PuLWBYtlWTjhhBNw2mmnYbfddou6OXnnyy+/xE033YRf/epXUTfFN2vWrEFXVxf69evnWN6vX7+inUIFgE2bNuHss8/G3nvvjZEjR0bdnNB44IEH8O6772L27NlRNyVUvvrqK9x6660YNmwYnn32WZx++uk488wz8de//jXqpkVGUQuRWbNmoaysTPlZvHgxLMvCzJkz0dzcjAULFuCtt97C4YcfjmnTpuG7776L+jC00T3eTZs2AQAuuOACHHnkkRgzZgzmzZuHsrIyPPjggxEfhR66x3rTTTehtbUV5513XtRN9oXu8bIsW7YMkydPxvTp0zFjxoyIWk74ZebMmVi0aBEeeOCBqJsSGkuWLMFZZ52Fe++9F5WVlVE3J1Q2bdqEXXfdFVdeeSV22WUXnHrqqZgxYwZuu+22qJsWGUVda2b16tVYu3atcp2hQ4diwYIFmDRpEtatW4e6urrsd8OGDcPJJ5+MWbNmhd3UQNA93tdffx37778/FixYgH322Sf73bhx43DggQfiiiuuCLupvtE91qOOOgqPP/44ysrKssu7urpQXl6OY489tmDeQnSPN5FIAACWL1+OCRMmYI899sBdd92FHj0K/52jo6MDVVVVeOihhxwRXscffzzWr1+Pxx57LLrGhcQZZ5yBxx57DK+++iq23nrrqJsTGo8++iiOOOIIlJeXZ5d1dXWhrKwMPXr0QDqddnxXyAwZMgQHHXQQ7rjjjuyyW2+9FZdffjmWLVsWYcuio2fUDQiTvn37om/fvq7rtbe3A0BOZ92jR4+s9aAQ0D3eMWPGIJlM4rPPPssKkY0bN+Kbb77BkCFDwm5mIOge64033ojLL788+/fy5ctx8MEHY/78+Rg3blyYTQwU3eMFMpaQiRMnZi1dxSBCACCRSGDMmDF48cUXs0Jk06ZNePHFF3HGGWdE27iAsSwLv/nNb/DII4/gn//8Z1GLEAA44IAD8NFHHzmWnXjiiRg+fDj+8Ic/FI0IAYC99947JxT7888/L5i+NxQidZWNCatXr7b69Olj/fSnP7Xef/9967PPPrP++7//26qoqLDef//9qJsXCmeddZa1xRZbWM8++6y1ePFi6+STT7aam5ut77//PuqmhcrXX39d1FEzS5cutbbddlvrgAMOsJYuXWp999132U8x8MADD1jJZNK66667rE8++cQ69dRTrYaGBmvFihVRNy1QTj/9dKu+vt765z//6biG7e3tUTctbxRr1Mxbb71l9ezZ07riiiusL774wrr33nutqqoq65577om6aZFBQmQzb7/9tjVp0iSrd+/eVm1trbXHHntYTz31VNTNCo2Ojg7rd7/7ndXc3GzV1tZaBx54oLVo0aKomxU6xS5E5s2bZwEQfoqFm266ydpyyy2tRCJhjR071vrXv/4VdZMCR3YN582bF3XT8kaxChHLsqzHH3/cGjlypJVMJq3hw4dbf/nLX6JuUqQUtY8IQRAEQRDxpjgmjwmCIAiCKEhIiBAEQRAEERkkRAiCIAiCiAwSIgRBEARBRAYJEYIgCIIgIoOECEEQBEEQkUFChCAIgiCIyCAhQhAEQRBEZJAQIQiCIAgiMkiIEARBEAQRGSRECIIgCIKIjP8fR1r+R49MWLoAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure()\n",
    "ax = fig.gca()\n",
    "\n",
    "ax.set_xlim(x_min, x_max)\n",
    "ax.set_ylim(y_min, y_max)\n",
    "\n",
    "# Contourf plot\n",
    "ax.set_title(f'KDE on {grid_density.shape[0]} Grid of {sample_size} Samples (bw={bw:.2f})')\n",
    "cfset = ax.contourf(xx, yy, grid_density.round(4),\n",
    "                    levels=np.linspace(0, np.max(grid_density), 20),\n",
    "                    cmap='Blues',\n",
    "                    alpha=1)\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create KDE for Each Year"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([8.0000e+00, 3.8000e+01, 6.7000e+01, 1.7100e+02, 7.4500e+02,\n",
       "        1.9680e+03, 3.5770e+03, 7.2440e+03, 1.5364e+04, 3.4031e+04]),\n",
       " array([1952., 1959., 1966., 1973., 1980., 1987., 1994., 2001., 2008.,\n",
       "        2015., 2022.]),\n",
       " <BarContainer object of 10 artists>)"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(years)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 42/42 [00:50<00:00,  1.21s/it]\n"
     ]
    }
   ],
   "source": [
    "time_grids = {}\n",
    "for cur_year in tqdm(range(1980, 2022)):\n",
    "    cur_kde = KernelDensity(kernel='gaussian', bandwidth=bw)\n",
    "\n",
    "    cur_indexes = np.where(umap_df['years'] == cur_year)[0]\n",
    "    cur_kde.fit(projected_emb[cur_indexes, :])\n",
    "\n",
    "    # Compute KDE and transform it to likelihood scale\n",
    "    cur_log_density = cur_kde.score_samples(grid)\n",
    "    cur_density = np.exp(cur_log_density)\n",
    "    cur_grid_density = np.reshape(cur_density, xx.shape)\n",
    "    cur_grid = cur_grid_density.astype(float).round(4).tolist()\n",
    "\n",
    "    # Record this year's grid\n",
    "    time_grids[str(cur_year)] = cur_grid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_min, x_max, y_min, y_max = float(x_min), float(x_max), float(y_min), float(y_max)\n",
    "\n",
    "grid_density_json = {\n",
    "    'grid': grid_density.astype(float).round(4).tolist(),\n",
    "    'xRange': [x_min, x_max],\n",
    "    'yRange': [y_min, y_max],\n",
    "    'sampleSize': sample_size,\n",
    "    'totalPointSize': umap_df.shape[0],\n",
    "    'padded': True,\n",
    "    'timeGrids': time_grids\n",
    "}\n",
    "\n",
    "dump(grid_density_json, open(join(DATA_DIR, 'umap-grid.json'), 'w'))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Pre-compute Embedding Summaries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [],
   "source": [
    "umap_df = pd.read_csv(join(DATA_DIR, 'umap.csv'))\n",
    "\n",
    "data = []\n",
    "texts = []\n",
    "\n",
    "text_key = 'abstracts'\n",
    "texts = umap_df[text_key]\n",
    "\n",
    "xs = umap_df['xs']\n",
    "ys = umap_df['ys']\n",
    "\n",
    "# Create data array\n",
    "for i in range(len(xs)):\n",
    "    cur_data = {\n",
    "        'x': xs[i],\n",
    "        'y': ys[i],\n",
    "        'pid': i,\n",
    "    }\n",
    "    data.append(cur_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [],
   "source": [
    "def top_n_idx_sparse(matrix: csr_matrix, n: int) -> np.ndarray:\n",
    "    \"\"\" Return indices of top n values in each row of a sparse matrix\n",
    "    Retrieved from:\n",
    "        https://github.com/MaartenGr/BERTopic/blob/master/bertopic/_bertopic.py#L2801\n",
    "    Arguments:\n",
    "        matrix: The sparse matrix from which to get the top n indices per row\n",
    "        n: The number of highest values to extract from each row\n",
    "    Returns:\n",
    "        indices: The top n indices per row\n",
    "    \"\"\"\n",
    "    indices = []\n",
    "    for le, ri in zip(matrix.indptr[:-1], matrix.indptr[1:]):\n",
    "        n_row_pick = min(n, ri - le)\n",
    "        values = matrix.indices[le + np.argpartition(matrix.data[le:ri], -n_row_pick)[-n_row_pick:]]\n",
    "        values = [values[index] if len(values) >= index + 1 else None for index in range(n)]\n",
    "        indices.append(values)\n",
    "    return np.array(indices)\n",
    "\n",
    "\n",
    "def top_n_values_sparse(matrix: csr_matrix, indices: np.ndarray) -> np.ndarray:\n",
    "    \"\"\" Return the top n values for each row in a sparse matrix\n",
    "    Arguments:\n",
    "        matrix: The sparse matrix from which to get the top n indices per row\n",
    "        indices: The top n indices per row\n",
    "    Returns:\n",
    "        top_values: The top n scores per row\n",
    "    \"\"\"\n",
    "    top_values = []\n",
    "    for row in range(indices.shape[0]):\n",
    "        scores = np.array([matrix[row, c] if c is not None else 0 for c in indices[row, :]])\n",
    "        top_values.append(scores)\n",
    "    return np.array(top_values)\n",
    "\n",
    "def merge_leaves_before_level(root: Node, target_level: int) -> Tuple[list, list, dict]:\n",
    "    \"\"\"\n",
    "    Merge all nodes to their parents until the tree is target_level tall (modify\n",
    "    root in-place) and extract all data from leaf nodes before or at the target_level.\n",
    "\n",
    "    Args:\n",
    "        root (Node): Root node\n",
    "        target_level (int): Target level\n",
    "\n",
    "    Returns:\n",
    "        csr_row_indexes (list): Row indexes for the sparse matrix. Each row is\n",
    "            a leaf node.\n",
    "        csr_column_indexes (list): Column indexes for the sparse matrix. Each column\n",
    "            is a prompt ID.\n",
    "        row_node_map (dict): A dictionary map row index to the leaf node.\n",
    "    \"\"\"\n",
    "    \n",
    "    x0, y0, x1, y1 = root.position\n",
    "    step_size = (x1 - x0) / (2 ** target_level)\n",
    "\n",
    "    # Find all leaves at or before the target level\n",
    "    row_pos_map = {}\n",
    "    stack = [root]\n",
    "\n",
    "    # We create a sparse matrix by (data, (row index, column index))\n",
    "    csr_row_indexes, csr_column_indexes = [], []\n",
    "\n",
    "    # In the multiplication sparse matrix, each row represents a tile / collection,\n",
    "    # and each column represents a prompt ID\n",
    "    cur_r = 0\n",
    "\n",
    "    while len(stack) > 0:\n",
    "        cur_node = stack.pop()\n",
    "        \n",
    "        if cur_node.level >= target_level:        \n",
    "            # A new traverse here to concatenate all the prompts from its subtree,\n",
    "            # and to merge it with its children\n",
    "            local_stack = [cur_node]\n",
    "            subtree_data = []\n",
    "\n",
    "            while len(local_stack) > 0:\n",
    "                local_node = local_stack.pop()\n",
    "                \n",
    "                if len(local_node.children) == 0:\n",
    "                    # Leaf node\n",
    "                    subtree_data.extend(local_node.data)\n",
    "                else:\n",
    "                    for c in local_node.children[::-1]:\n",
    "                        if c is not None:\n",
    "                            local_stack.append(c)\n",
    "                            \n",
    "            # Detach all the children and get their data\n",
    "            cur_node.children = []\n",
    "            cur_node.data = subtree_data\n",
    "            \n",
    "            # Register this node in a dictionary for faster access\n",
    "            row_pos_map[cur_r] =  list(map(lambda x: round(x, 3), cur_node.position))\n",
    "            \n",
    "            # Collect the prompt IDs\n",
    "            for d in cur_node.data:\n",
    "                csr_row_indexes.append(cur_r)\n",
    "                csr_column_indexes.append(d['pid'])\n",
    "                \n",
    "            # Move on to the next tile / collection\n",
    "            cur_r += 1\n",
    "\n",
    "        else:\n",
    "            if len(cur_node.children) == 0:\n",
    "                # Leaf node => it means this leaf is before the target level\n",
    "                # We need to adjust the node's position so that it has the same\n",
    "                # size as leaf nodes at the target_level\n",
    "                x, y = cur_node.data[0]['x'], cur_node.data[0]['y']\n",
    "                xi, yi = int((x - x0) // step_size), int((y - y0) // step_size)\n",
    "\n",
    "                # Find the bounding box of current level of this leaf node\n",
    "                xi0, yi0 = x0 + xi * step_size, y0 + yi * step_size\n",
    "                xi1, yi1 = xi0 + step_size, yi0 + step_size\n",
    "                row_pos_map[cur_r] = list(map(lambda x: round(x, 3), [xi0, yi0, xi1, yi1]))\n",
    "                \n",
    "                # Collect the prompt IDs\n",
    "                for d in cur_node.data:\n",
    "                    csr_row_indexes.append(cur_r)\n",
    "                    csr_column_indexes.append(d['pid'])\n",
    "                \n",
    "                # Move on to the next tile / collection\n",
    "                cur_r += 1\n",
    "\n",
    "            else:\n",
    "                for c in cur_node.children[::-1]:\n",
    "                    if c is not None:\n",
    "                        stack.append((c))\n",
    "    \n",
    "    return csr_row_indexes, csr_column_indexes, row_pos_map\n",
    "\n",
    "\n",
    "def get_tile_topics(count_mat, row_pos_map, ngrams, top_k=10):\n",
    "    \"\"\"Get the top-k important keywords from all rows in the count_mat.\n",
    "\n",
    "    Args:\n",
    "        count_mat (csr_mat): A count matrix\n",
    "        row_pos_map (dict): A dictionary that maps row index to the corresponding\n",
    "            leaf node's location in the quadtree\n",
    "        ngrams (list[str]): Feature names in the count_mat\n",
    "        top_k (int): Number of keywords to extract\n",
    "    \"\"\"\n",
    "    \n",
    "    \n",
    "    # Compute tf-idf score\n",
    "    t_tf_idf_model = TfidfTransformer()\n",
    "    t_tf_idf = t_tf_idf_model.fit_transform(count_mat)\n",
    "    \n",
    "    # Get words with top scores for each tile\n",
    "    indices = top_n_idx_sparse(t_tf_idf, top_k)\n",
    "    scores = top_n_values_sparse(t_tf_idf, indices)\n",
    "\n",
    "    sorted_indices = np.argsort(scores, 1)\n",
    "    indices = np.take_along_axis(indices, sorted_indices, axis=1)\n",
    "    scores = np.take_along_axis(scores, sorted_indices, axis=1)\n",
    "    \n",
    "    # Store these keywords\n",
    "    tile_topics = []\n",
    "\n",
    "    for r in row_pos_map:\n",
    "        word_scores = [\n",
    "            (ngrams[word_index], round(score, 4))\n",
    "            if word_index is not None and score > 0\n",
    "            else (\"\", 0.00001)\n",
    "            for word_index, score in zip(indices[r][::-1], scores[r][::-1])\n",
    "        ]\n",
    "\n",
    "        tile_topics.append({\n",
    "            'w': word_scores,\n",
    "            'p': row_pos_map[r]\n",
    "        })\n",
    "\n",
    "    return tile_topics\n",
    "\n",
    "def extract_level_topics(\n",
    "    root: Node,\n",
    "    count_mat: csr_matrix,\n",
    "    texts: list[str],\n",
    "    ngrams: list[str],\n",
    "    min_level = None,\n",
    "    max_level = None\n",
    "):\n",
    "    \"\"\"Extract topics for all leaf nodes at all levels of the quadtree.\n",
    "\n",
    "    Args:\n",
    "        root (Noe): Quadtree node\n",
    "        count_mat (csr_matrix): Count vector for the corpus\n",
    "        texts (list[str]): A list of all the embeddings' texts\n",
    "        ngrams (list[str]): n-gram list for the count vectorizer\n",
    "    \"\"\"\n",
    "\n",
    "    level_tile_topics = {}\n",
    "    \n",
    "    if min_level is None:\n",
    "        min_level = 0\n",
    "\n",
    "    if max_level is None:\n",
    "        max_level = root.height\n",
    "\n",
    "    for level in tqdm(list(range(max_level, min_level - 1, -1))):\n",
    "        # Create a sparse matrix\n",
    "        csr_row_indexes, csr_column_indexes, row_node_map = merge_leaves_before_level(\n",
    "            root, level\n",
    "        )\n",
    "\n",
    "        csr_data = [1 for _ in range(len(csr_row_indexes))]\n",
    "        tile_mat = csr_matrix(\n",
    "            (csr_data, (csr_row_indexes, csr_column_indexes)),\n",
    "            shape=(len(texts), len(texts)),\n",
    "        )\n",
    "\n",
    "        # Transform the count matrix\n",
    "        new_count_mat = tile_mat @ count_mat\n",
    "\n",
    "        # Compute t-tf-idf scores and extract keywords\n",
    "        tile_topics = get_tile_topics(new_count_mat, row_node_map, ngrams)\n",
    "        \n",
    "        level_tile_topics[level] = tile_topics\n",
    "        \n",
    "    return level_tile_topics\n",
    "\n",
    "\n",
    "def select_topic_levels(\n",
    "    max_zoom_scale,\n",
    "    svg_width,\n",
    "    svg_height,\n",
    "    x_domain,\n",
    "    y_domain,\n",
    "    tree_extent,\n",
    "    ideal_tile_width=35,\n",
    "):\n",
    "    \"\"\"\n",
    "    Automatically determine the min and max topic levels needed for the visualization.\n",
    "\n",
    "    Args:\n",
    "        max_zoom_scale (float): Max zoom scale level\n",
    "        svg_width (int): SVG width\n",
    "        svg_height (int): SVG height\n",
    "        x_domain ([float, float]): [x min, x max]\n",
    "        y_domain ([float, float]): [y min, y max]\n",
    "        tree_extent ([[float, float], [float, float]]): The extent of the tree\n",
    "        ideal_tile_width (int, optional): Optimal tile width in pixel. Defaults to 35.\n",
    "    \"\"\"\n",
    "\n",
    "    svg_length = max(svg_width, svg_height)\n",
    "    world_length = max(x_domain[1] - x_domain[0], y_domain[1] - y_domain[0])\n",
    "    tree_to_world_scale = (tree_extent[1][0] - tree_extent[0][0]) / world_length\n",
    "\n",
    "    scale = 1\n",
    "    selected_levels = []\n",
    "\n",
    "    while scale <= max_zoom_scale:\n",
    "        best_level = 1\n",
    "        best_tile_width_diff = np.Infinity\n",
    "\n",
    "        for l in range(1, 21):\n",
    "            tile_num = 2**l\n",
    "            svg_scaled_length = scale * svg_length * tree_to_world_scale\n",
    "            tile_width = svg_scaled_length / tile_num\n",
    "\n",
    "            if abs(tile_width - ideal_tile_width) < best_tile_width_diff:\n",
    "                best_tile_width_diff = abs(tile_width - ideal_tile_width)\n",
    "                best_level = l\n",
    "\n",
    "        selected_levels.append(best_level)\n",
    "        scale += 0.5\n",
    "\n",
    "    return np.min(selected_levels), np.max(selected_levels)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "63213it [00:00, 66402.43it/s] \n"
     ]
    }
   ],
   "source": [
    "# Build the quadtree\n",
    "tree = Quadtree()\n",
    "tree.add_all_data(data)\n",
    "\n",
    "# Build the count matrix\n",
    "root = tree.get_node_representation()\n",
    "\n",
    "cv = CountVectorizer(stop_words=\"english\", ngram_range=(1, 1))\n",
    "count_mat = cv.fit_transform(texts)\n",
    "ngrams = cv.get_feature_names_out()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6 10\n"
     ]
    }
   ],
   "source": [
    "max_zoom_scale = 20\n",
    "svg_width = 800\n",
    "svg_height = 800\n",
    "xs = [d['x'] for d in data]\n",
    "ys = [d['y'] for d in data]\n",
    "x_domain = [np.min(xs), np.max(xs)]\n",
    "y_domain = [np.min(ys), np.max(ys)]\n",
    "\n",
    "# Get suggestions of quadtree levels to extract\n",
    "min_level, max_level = select_topic_levels(\n",
    "    max_zoom_scale, svg_width, svg_height, x_domain, y_domain, tree.extent()\n",
    ")\n",
    "print(min_level, max_level)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 5/5 [00:20<00:00,  4.03s/it]\n"
     ]
    }
   ],
   "source": [
    "level_tile_topics = extract_level_topics(\n",
    "    root, count_mat, texts, ngrams, min_level=min_level, max_level=max_level\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_dict = {\n",
    "    'extent': tree.extent(),\n",
    "    'data': {},\n",
    "    'range': [np.min(xs), np.min(ys), np.max(xs), np.max(ys)]\n",
    "}\n",
    "\n",
    "for cur_level in range(min_level, max_level + 1):\n",
    "    cur_topics = level_tile_topics[cur_level]\n",
    "    data_dict['data'][cur_level] = []\n",
    "\n",
    "    for topic in cur_topics:\n",
    "        # Get the topic name\n",
    "        name = '-'.join([p[0] for p in topic['w'][:4]])\n",
    "        x = (topic['p'][0] + topic['p'][2]) / 2\n",
    "        y = (topic['p'][1] + topic['p'][3]) / 2\n",
    "        cur_data = {\n",
    "            'x': round(x, 3),\n",
    "            'y': round(y, 3),\n",
    "            'n': name,\n",
    "            'l': cur_level\n",
    "        }\n",
    "        data_dict['data'][cur_level].append(\n",
    "            [round(x, 3), round(y, 3), name]\n",
    "        )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [],
   "source": [
    "dump(data_dict, open(join(DATA_DIR, 'topic-data.json'), 'w'))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Export NDJSON"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [],
   "source": [
    "umap_df = pd.read_csv(join(DATA_DIR, 'umap.csv'))\n",
    "\n",
    "text_key = 'abstracts'\n",
    "time_key = 'years'\n",
    "label_key = 'titles'\n",
    "\n",
    "texts = umap_df[text_key]\n",
    "\n",
    "if time_key is not None:\n",
    "    times = list(map(str, umap_df[time_key]))\n",
    "else:\n",
    "    times = []\n",
    "    \n",
    "if label_key is not None:\n",
    "    labels = umap_df[label_key]\n",
    "else:\n",
    "    labels = []\n",
    "\n",
    "xs = umap_df['xs']\n",
    "ys = umap_df['ys']\n",
    "\n",
    "umap_data_short = []\n",
    "\n",
    "for i in range(len(xs)):\n",
    "    cur_row = [xs[i], ys[i], texts[i]]\n",
    "    \n",
    "    if time_key is not None:\n",
    "        cur_row.append(times[i])\n",
    "        \n",
    "        if label_key is not None:\n",
    "            cur_row.append(labels[i])\n",
    "    \n",
    "    else:\n",
    "        if label_key is not None:\n",
    "            cur_row.append('')\n",
    "            cur_row.append(labels[i])\n",
    "\n",
    "    umap_data_short.append(cur_row)\n",
    "    \n",
    "with open(join(DATA_DIR, \"./umap.ndjson\"), 'w') as fp:\n",
    "    ndjson.dump(umap_data_short, fp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "rapids",
   "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": "3.9.13"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "c9758838403ef5c64e3388c7f7eb0622abdec8769e35829ff20981977cccefff"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}
