penguin_random_house/preprocessing.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Penguin Random House Dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The data were collected in January 2022 from [Penguin Random House](https://www.penguinrandomhouse.com/) via [API](https://developer.penguinrandomhouse.com/page).\n",
    "\n",
    "The collected data were parsed and saved [in this folder](data_interm) (see details [here](downloading.ipynb))."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import sys\n",
    "\n",
    "# Append sys.path with the project root path\n",
    "sys.path.append(os.path.dirname(os.path.abspath('')))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import warnings\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "from IPython.display import Image\n",
    "from bs4 import BeautifulSoup\n",
    "from treelib import Tree\n",
    "\n",
    "from utils.preprocessing import get_weighted_rating\n",
    "\n",
    "# Suppress warnings\n",
    "warnings.filterwarnings('ignore', category=UserWarning, module='bs4')\n",
    "\n",
    "# To print many rows\n",
    "pd.set_option('display.max_rows', 100)\n",
    "\n",
    "# To wrap long values\n",
    "pd.set_option('display.max_colwidth', 78)\n",
    "\n",
    "# To use seaborn style for plots\n",
    "sns.set_style('darkgrid', {'axes.facecolor': '.9'})\n",
    "sns.set_context('talk')\n",
    "sns_palette = sns.color_palette(plt.cm.get_cmap('tab20c').colors)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Categories"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Books are assigned to one or more categories identified by a numeric `cat_id`. Each category is part of a larger category set and may be arranged in relation to other categories in the set to form a hierarchy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "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>description</th>\n",
       "      <th>menu_text</th>\n",
       "      <th>parent</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cat_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2000000000</th>\n",
       "      <td>Consumer Category</td>\n",
       "      <td>Consumer Category</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000000029</th>\n",
       "      <td>Children's Books</td>\n",
       "      <td>Children's</td>\n",
       "      <td>2000000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000000006</th>\n",
       "      <td>Children's Activity &amp; Novelty Books</td>\n",
       "      <td>Activity &amp; Novelty Books</td>\n",
       "      <td>2000000029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000000022</th>\n",
       "      <td>Children's Board Books</td>\n",
       "      <td>Board Books</td>\n",
       "      <td>2000000029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000000023</th>\n",
       "      <td>Children's Boxed Sets</td>\n",
       "      <td>Boxed Sets</td>\n",
       "      <td>2000000029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000000028</th>\n",
       "      <td>Children's Chapter Books</td>\n",
       "      <td>Chapter Books</td>\n",
       "      <td>2000000029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000000100</th>\n",
       "      <td>Children's Middle Grade Books</td>\n",
       "      <td>Middle Grade Books</td>\n",
       "      <td>2000000029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000000005</th>\n",
       "      <td>Children's Middle Grade Action &amp; Adventure Books</td>\n",
       "      <td>Action &amp; Adventure</td>\n",
       "      <td>2000000100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000000055</th>\n",
       "      <td>Children's Middle Grade Fantasy &amp; Magical Books</td>\n",
       "      <td>Fantasy</td>\n",
       "      <td>2000000100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000000068</th>\n",
       "      <td>Children's Middle Grade Historical Books</td>\n",
       "      <td>Historical</td>\n",
       "      <td>2000000100</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 description  \\\n",
       "cat_id                                                         \n",
       "2000000000                                 Consumer Category   \n",
       "2000000029                                  Children's Books   \n",
       "2000000006               Children's Activity & Novelty Books   \n",
       "2000000022                            Children's Board Books   \n",
       "2000000023                             Children's Boxed Sets   \n",
       "2000000028                          Children's Chapter Books   \n",
       "2000000100                     Children's Middle Grade Books   \n",
       "2000000005  Children's Middle Grade Action & Adventure Books   \n",
       "2000000055   Children's Middle Grade Fantasy & Magical Books   \n",
       "2000000068          Children's Middle Grade Historical Books   \n",
       "\n",
       "                           menu_text      parent  \n",
       "cat_id                                            \n",
       "2000000000         Consumer Category        <NA>  \n",
       "2000000029                Children's  2000000000  \n",
       "2000000006  Activity & Novelty Books  2000000029  \n",
       "2000000022               Board Books  2000000029  \n",
       "2000000023                Boxed Sets  2000000029  \n",
       "2000000028             Chapter Books  2000000029  \n",
       "2000000100        Middle Grade Books  2000000029  \n",
       "2000000005        Action & Adventure  2000000100  \n",
       "2000000055                   Fantasy  2000000100  \n",
       "2000000068                Historical  2000000100  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load categories info\n",
    "path_cats = os.path.join('data_interm', 'categories.txt')\n",
    "\n",
    "# Read data\n",
    "categories = pd.read_json(path_cats, lines=True)\n",
    "\n",
    "# Set the proper types\n",
    "categories['parent'] = categories['parent'].astype('Int64')\n",
    "categories['cat_id'] = categories['cat_id'].astype('Int64')\n",
    "\n",
    "# Set index\n",
    "categories.set_index('cat_id', inplace=True)\n",
    "\n",
    "# Change the name of the top category\n",
    "categories.loc[2000000000, 'menu_text'] = 'Consumer Category'\n",
    "\n",
    "# Show\n",
    "categories.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Each category has the following [properties](https://developer.penguinrandomhouse.com/docs/read/enhanced_prh_api/concepts/Categories):\n",
    "- `cat_id`, a number that uniquely identifies this category\n",
    "- `menu_text`, a brief description of the category, suitable for use in a navigational menu\n",
    "- `description`, the description of the category, suitable for display\n",
    "- `parent`, it is a `cat_id` of the parent category"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "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>description</th>\n",
       "      <th>menu_text</th>\n",
       "      <th>parent</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cat_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2000000000</th>\n",
       "      <td>Consumer Category</td>\n",
       "      <td>Consumer Category</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  description          menu_text  parent\n",
       "cat_id                                                  \n",
       "2000000000  Consumer Category  Consumer Category    <NA>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# If there are NaNs\n",
    "mask = categories.isna().any(axis=1)\n",
    "categories[mask]\n",
    "\n",
    "# This is ok, because the highest level doesn't have a parent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Consumer Category\n",
      "├── Audiobooks\n",
      "├── Children's\n",
      "│   ├── Activity & Novelty Books\n",
      "│   ├── Board Books\n",
      "│   ├── Boxed Sets\n",
      "│   ├── Chapter Books\n",
      "│   ├── Guided Reading Levels\n",
      "│   │   ├── Leveled Readers\n",
      "│   │   │   ├── Level 1 Readers\n",
      "│   │   │   ├── Level 2 Readers\n",
      "│   │   │   ├── Level 3 Readers\n",
      "│   │   │   └── Level 4 Readers\n",
      "│   │   └── Step Into Reading \n",
      "│   ├── Media Tie-In\n",
      "│   ├── Middle Grade Books\n",
      "│   │   ├── Action & Adventure\n",
      "│   │   ├── Fantasy\n",
      "│   │   ├── Historical\n",
      "│   │   ├── Mystery & Detective\n",
      "│   │   └── Sports\n",
      "│   ├── Picture Books\n",
      "│   ├── Spanish Language \n",
      "│   └── Test Preparation\n",
      "├── Classics\n",
      "│   ├── Fiction Classics\n",
      "│   ├── Literary Collections\n",
      "│   ├── Literary Criticism\n",
      "│   └── Nonfiction Classics\n",
      "├── Fiction\n",
      "│   ├── Gothic & Horror\n",
      "│   ├── Graphic Novels & Manga\n",
      "│   ├── Historical \n",
      "│   ├── Literary\n",
      "│   ├── Military \n",
      "│   ├── Mystery & Suspense\n",
      "│   │   ├── Cozy \n",
      "│   │   ├── Crime\n",
      "│   │   ├── Espionage\n",
      "│   │   ├── Noir\n",
      "│   │   └── Suspense & Thriller\n",
      "│   ├── Paranormal\n",
      "│   ├── Romance\n",
      "│   │   ├── Contemporary \n",
      "│   │   ├── Erotica\n",
      "│   │   ├── Historical \n",
      "│   │   ├── New Adult \n",
      "│   │   ├── Paranormal\n",
      "│   │   ├── Regency\n",
      "│   │   ├── Suspense \n",
      "│   │   └── Western \n",
      "│   ├── Sci-Fi & Fantasy\n",
      "│   │   ├── Contemporary Fantasy\n",
      "│   │   ├── Cyber Punk\n",
      "│   │   ├── Epic Fantasy\n",
      "│   │   ├── Fairy Tales\n",
      "│   │   ├── Military Science Fiction\n",
      "│   │   ├── Space Opera\n",
      "│   │   └── Urban Fantasy\n",
      "│   ├── Spanish Language \n",
      "│   ├── Spiritual \n",
      "│   ├── Western \n",
      "│   └── Women's Fiction\n",
      "├── Games\n",
      "├── Humor\n",
      "├── Merchandise\n",
      "├── Nonfiction\n",
      "│   ├── Arts & Entertainment\n",
      "│   │   ├── Art\n",
      "│   │   ├── Design\n",
      "│   │   ├── Film\n",
      "│   │   ├── Music\n",
      "│   │   ├── Performing Arts\n",
      "│   │   ├── Photography\n",
      "│   │   └── Writing\n",
      "│   ├── Biography & Memoir\n",
      "│   │   ├── Arts & Entertainment\n",
      "│   │   ├── Historical\n",
      "│   │   ├── Literary \n",
      "│   │   └── Political\n",
      "│   ├── Business\n",
      "│   │   ├── Economics\n",
      "│   │   ├── Management\n",
      "│   │   └── Marketing\n",
      "│   ├── Cooking\n",
      "│   │   ├── Baking & Desserts\n",
      "│   │   ├── Cooking Methods\n",
      "│   │   ├── Food Memoir & Travel\n",
      "│   │   └── Wine & Beverage\n",
      "│   ├── Crafts, Home & Garden\n",
      "│   │   ├── Crafts & Hobbies\n",
      "│   │   ├── Home & Garden\n",
      "│   │   └── Weddings\n",
      "│   ├── Health & Fitness\n",
      "│   │   ├── Alternative Therapies\n",
      "│   │   ├── Diet & Nutrition\n",
      "│   │   ├── Exercise\n",
      "│   │   └── Health Reference\n",
      "│   ├── History\n",
      "│   │   ├── Military\n",
      "│   │   │   ├── 1950-Present \n",
      "│   │   │   ├── World War I \n",
      "│   │   │   └── World War II\n",
      "│   │   ├── U.S. History\n",
      "│   │   │   ├── 19th Century\n",
      "│   │   │   ├── 20th Century \n",
      "│   │   │   ├── 21st Century \n",
      "│   │   │   ├── Civil War Period\n",
      "│   │   │   ├── Colonial/Revolutionary Period\n",
      "│   │   │   └── Native American\n",
      "│   │   └── World \n",
      "│   │       ├── Africa World History\n",
      "│   │       ├── Ancient World History\n",
      "│   │       ├── Asia World History\n",
      "│   │       ├── Europe World History\n",
      "│   │       ├── Latin America World History\n",
      "│   │       ├── Middle East World History\n",
      "│   │       └── North America World History\n",
      "│   ├── Parenting\n",
      "│   ├── Pets\n",
      "│   ├── Politics\n",
      "│   │   ├── Domestic\n",
      "│   │   └── World \n",
      "│   ├── Popular Science\n",
      "│   │   ├── Science\n",
      "│   │   └── Technology\n",
      "│   ├── Psychology\n",
      "│   ├── Reference\n",
      "│   │   ├── Language \n",
      "│   │   └── Test Preparation\n",
      "│   ├── Religion & Philosophy\n",
      "│   │   ├── Bibles\n",
      "│   │   ├── Philosophy\n",
      "│   │   └── Religion\n",
      "│   ├── Self-Improvement\n",
      "│   │   ├── Beauty\n",
      "│   │   ├── Inspiration & Motivation\n",
      "│   │   ├── Personal Finance\n",
      "│   │   └── Personal Growth\n",
      "│   ├── Spanish Language \n",
      "│   ├── Sports\n",
      "│   └── Travel\n",
      "│       ├── Africa\n",
      "│       ├── Asia\n",
      "│       ├── Australia & Oceania Travel\n",
      "│       ├── Caribbean & Mexico \n",
      "│       ├── Central & South America \n",
      "│       ├── Europe\n",
      "│       ├── Middle East\n",
      "│       ├── Specialty\n",
      "│       ├── Travel Writing\n",
      "│       └── USA & Canada \n",
      "├── Poetry\n",
      "└── Teen & Young Adult\n",
      "    ├── Action & Adventure\n",
      "    ├── Boxed Sets\n",
      "    ├── Fantasy & Magical\n",
      "    ├── Fiction\n",
      "    ├── Historical Fiction\n",
      "    ├── Mystery & Suspense\n",
      "    ├── Nonfiction\n",
      "    ├── Romance\n",
      "    ├── Science Fiction \n",
      "    ├── Social Issues\n",
      "    └── Spanish Language \n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Visualize the hierarchy\n",
    "tree = Tree()\n",
    "for index, row in categories.iterrows():\n",
    "\n",
    "    # If this category is the top\n",
    "    if pd.isna(row.parent):\n",
    "        tree.create_node(row.menu_text, index)\n",
    "        continue\n",
    "\n",
    "    # Child categories\n",
    "    tree.create_node(row.menu_text, index, parent=row.parent)\n",
    "\n",
    "tree.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": []
   },
   "source": [
    "## Contributor Roles"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "In PRH, every ISBN has up to five people mapped to it and each mapping is qualified with a contributor role code that identifies the type of contribution that person made to the ISBN.\n",
    "\n",
    "Beside of the role code and description, there is a sequence value that [is used](https://developer.penguinrandomhouse.com/docs/read/enhanced_prh_api/concepts/Contributor_role) to prioritize certain roles over others. This sequencing is used to order contributors within a work or ISBN."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "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>description</th>\n",
       "      <th>seq</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>role_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>A</th>\n",
       "      <td>Author</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>I</th>\n",
       "      <td>Illustrator</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Read by</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>U</th>\n",
       "      <td>Foreword by</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>V</th>\n",
       "      <td>Introduction by</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Preface by</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>E</th>\n",
       "      <td>Editor</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>D</th>\n",
       "      <td>Afterword by</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Q</th>\n",
       "      <td>Epilogue by</td>\n",
       "      <td>90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>P</th>\n",
       "      <td>Photographer</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             description  seq\n",
       "role_id                      \n",
       "A                 Author   10\n",
       "I            Illustrator   20\n",
       "4                Read by   30\n",
       "U            Foreword by   40\n",
       "V        Introduction by   50\n",
       "1             Preface by   60\n",
       "E                 Editor   70\n",
       "D           Afterword by   80\n",
       "Q            Epilogue by   90\n",
       "P           Photographer  100"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load roles\n",
    "path_roles = os.path.join('data_interm', 'contributor_roles.txt')\n",
    "roles = pd.read_json(path_roles, lines=True)\n",
    "\n",
    "# Rename code to `role_id`\n",
    "roles.rename(columns={'code': 'role_id'}, inplace=True)\n",
    "roles.set_index('role_id', inplace=True)\n",
    "roles.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": [],
    "toc-hr-collapsed": true
   },
   "source": [
    "## Books"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "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>title</th>\n",
       "      <th>onsale</th>\n",
       "      <th>price</th>\n",
       "      <th>language</th>\n",
       "      <th>pages</th>\n",
       "      <th>publisher</th>\n",
       "      <th>cover</th>\n",
       "      <th>format_family</th>\n",
       "      <th>projected_minutes</th>\n",
       "      <th>series_number</th>\n",
       "      <th>authors</th>\n",
       "      <th>categories</th>\n",
       "      <th>series</th>\n",
       "      <th>works</th>\n",
       "      <th>flapcopy</th>\n",
       "      <th>excerpt</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>isbn</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9780375760686</th>\n",
       "      <td>Morningside Heights</td>\n",
       "      <td>2005-07-12</td>\n",
       "      <td>[{'amount': 21.0, 'currencyCode': 'CAD', 'pricingType': None}, {'amount': ...</td>\n",
       "      <td>E</td>\n",
       "      <td>352.0</td>\n",
       "      <td>{'code': '7070', 'description': 'Random House Publishing Group'}</td>\n",
       "      <td>https://images.randomhouse.com/cover/9780375760686</td>\n",
       "      <td>Paperback</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[{'author_id': 35922, 'first_name': 'Cheryl', 'last_name': 'Mendelson', 'c...</td>\n",
       "      <td>[{'category_id': 2000000083, 'seq': 1}, {'category_id': 3000001343, 'seq':...</td>\n",
       "      <td>[]</td>\n",
       "      <td>[{'work_id': 113275, 'title': 'Morningside Heights', 'author': 'Cheryl Men...</td>\n",
       "      <td>Following the tremendous success of her first book, a nonfiction work on h...</td>\n",
       "      <td>&lt;b&gt;1&lt;/b&gt;&lt;br&gt;&lt;br&gt;This neighborhood was always middle-class and urban. The a...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781605295336</th>\n",
       "      <td>Going Long</td>\n",
       "      <td>2010-03-30</td>\n",
       "      <td>[{'amount': 16.99, 'currencyCode': 'USD', 'pricingType': None}, {'amount':...</td>\n",
       "      <td>E</td>\n",
       "      <td>400.0</td>\n",
       "      <td>{'code': '5H5H', 'description': 'Harmony/Rodale'}</td>\n",
       "      <td>https://images.randomhouse.com/cover/9781605295336</td>\n",
       "      <td>Paperback</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[{'author_id': 2061538, 'first_name': 'David', 'last_name': 'Willey', 'com...</td>\n",
       "      <td>[{'category_id': 2000000052, 'seq': 3}, {'category_id': 2000000144, 'seq':...</td>\n",
       "      <td>[{'series_id': 'D22', 'name': 'Runner's World', 'description': None, 'seri...</td>\n",
       "      <td>[{'work_id': 594144, 'title': 'Going Long', 'author': 'Edited by David Wil...</td>\n",
       "      <td>For more than 40 years, &lt;i&gt;Runner's World &lt;/i&gt;magazine has been the world'...</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             title      onsale  \\\n",
       "isbn                                             \n",
       "9780375760686  Morningside Heights  2005-07-12   \n",
       "9781605295336           Going Long  2010-03-30   \n",
       "\n",
       "                                                                                       price  \\\n",
       "isbn                                                                                           \n",
       "9780375760686  [{'amount': 21.0, 'currencyCode': 'CAD', 'pricingType': None}, {'amount': ...   \n",
       "9781605295336  [{'amount': 16.99, 'currencyCode': 'USD', 'pricingType': None}, {'amount':...   \n",
       "\n",
       "              language  pages  \\\n",
       "isbn                            \n",
       "9780375760686        E  352.0   \n",
       "9781605295336        E  400.0   \n",
       "\n",
       "                                                                      publisher  \\\n",
       "isbn                                                                              \n",
       "9780375760686  {'code': '7070', 'description': 'Random House Publishing Group'}   \n",
       "9781605295336                 {'code': '5H5H', 'description': 'Harmony/Rodale'}   \n",
       "\n",
       "                                                            cover  \\\n",
       "isbn                                                                \n",
       "9780375760686  https://images.randomhouse.com/cover/9780375760686   \n",
       "9781605295336  https://images.randomhouse.com/cover/9781605295336   \n",
       "\n",
       "              format_family  projected_minutes  series_number  \\\n",
       "isbn                                                            \n",
       "9780375760686     Paperback                NaN            NaN   \n",
       "9781605295336     Paperback                NaN            NaN   \n",
       "\n",
       "                                                                                     authors  \\\n",
       "isbn                                                                                           \n",
       "9780375760686  [{'author_id': 35922, 'first_name': 'Cheryl', 'last_name': 'Mendelson', 'c...   \n",
       "9781605295336  [{'author_id': 2061538, 'first_name': 'David', 'last_name': 'Willey', 'com...   \n",
       "\n",
       "                                                                                  categories  \\\n",
       "isbn                                                                                           \n",
       "9780375760686  [{'category_id': 2000000083, 'seq': 1}, {'category_id': 3000001343, 'seq':...   \n",
       "9781605295336  [{'category_id': 2000000052, 'seq': 3}, {'category_id': 2000000144, 'seq':...   \n",
       "\n",
       "                                                                                      series  \\\n",
       "isbn                                                                                           \n",
       "9780375760686                                                                             []   \n",
       "9781605295336  [{'series_id': 'D22', 'name': 'Runner's World', 'description': None, 'seri...   \n",
       "\n",
       "                                                                                       works  \\\n",
       "isbn                                                                                           \n",
       "9780375760686  [{'work_id': 113275, 'title': 'Morningside Heights', 'author': 'Cheryl Men...   \n",
       "9781605295336  [{'work_id': 594144, 'title': 'Going Long', 'author': 'Edited by David Wil...   \n",
       "\n",
       "                                                                                    flapcopy  \\\n",
       "isbn                                                                                           \n",
       "9780375760686  Following the tremendous success of her first book, a nonfiction work on h...   \n",
       "9781605295336  For more than 40 years, <i>Runner's World </i>magazine has been the world'...   \n",
       "\n",
       "                                                                                     excerpt  \n",
       "isbn                                                                                          \n",
       "9780375760686  <b>1</b><br><br>This neighborhood was always middle-class and urban. The a...  \n",
       "9781605295336                                                                           None  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load books info\n",
    "path_books = os.path.join('data_interm', 'books.txt')\n",
    "books = pd.read_json(path_books, lines=True)\n",
    "\n",
    "# Set `isbn` column as index\n",
    "books.set_index('isbn', inplace=True)\n",
    "books.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "title                 object\n",
       "onsale                object\n",
       "price                 object\n",
       "language              object\n",
       "pages                float64\n",
       "publisher             object\n",
       "cover                 object\n",
       "format_family         object\n",
       "projected_minutes    float64\n",
       "series_number        float64\n",
       "authors               object\n",
       "categories            object\n",
       "series                object\n",
       "works                 object\n",
       "flapcopy              object\n",
       "excerpt               object\n",
       "dtype: object"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "books.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's analyze each column individually."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Onsale"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Values of the column `onsale` are dates when books began to be sold. So, convert them to the appropriate type:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# If there are NaNs\n",
    "books['onsale'].isna().any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert to datetime\n",
    "books['onsale'] = pd.to_datetime(books['onsale'])\n",
    "\n",
    "# Rename\n",
    "books.rename(columns={'onsale': 'onsale_date'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show the distribution\n",
    "_, ax = plt.subplots(figsize=(10, 8))\n",
    "sns.histplot(x=books['onsale_date'],\n",
    "             palette=sns_palette, ax=ax, bins=100)\n",
    "ax.set_ylabel('Number of books')\n",
    "ax.set_xlabel('Year of starting selling')\n",
    "ax.set_title('Distribution of dates when books started to be sold')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "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>title</th>\n",
       "      <th>onsale_date</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>isbn</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9780394404288</th>\n",
       "      <td>The Prophet</td>\n",
       "      <td>1923-09-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780448089010</th>\n",
       "      <td>Hardy Boys 01: the Tower Treasure</td>\n",
       "      <td>1927-06-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780448089027</th>\n",
       "      <td>Hardy Boys 02: the House on the Cliff</td>\n",
       "      <td>1927-06-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780448089034</th>\n",
       "      <td>Hardy Boys 03: the Secret of the Old Mill</td>\n",
       "      <td>1927-06-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780448089058</th>\n",
       "      <td>Hardy Boys 05: Hunting for Hidden Gold</td>\n",
       "      <td>1928-05-01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                   title onsale_date\n",
       "isbn                                                                \n",
       "9780394404288                                The Prophet  1923-09-12\n",
       "9780448089010          Hardy Boys 01: the Tower Treasure  1927-06-01\n",
       "9780448089027      Hardy Boys 02: the House on the Cliff  1927-06-01\n",
       "9780448089034  Hardy Boys 03: the Secret of the Old Mill  1927-06-01\n",
       "9780448089058     Hardy Boys 05: Hunting for Hidden Gold  1928-05-01"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show the newest and the oldest books\n",
    "books_sorted_onsale = books.sort_values('onsale_date')\n",
    "\n",
    "# The oldest books\n",
    "books_sorted_onsale[['title', 'onsale_date']].head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "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>title</th>\n",
       "      <th>onsale_date</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>isbn</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9788499083209</th>\n",
       "      <td>Trilogía de la Fundación / The Foundation Trilogy</td>\n",
       "      <td>2022-03-22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9788415594765</th>\n",
       "      <td>Wonder: Charlotte tiene la palabra / Shingaling. A Wonder Story</td>\n",
       "      <td>2022-04-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780553539233</th>\n",
       "      <td>Super-Pets! (DC Super Friends)</td>\n",
       "      <td>2022-04-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780451419903</th>\n",
       "      <td>Mark of the Hunter</td>\n",
       "      <td>2022-04-26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780375806131</th>\n",
       "      <td>The Name Jar</td>\n",
       "      <td>2022-07-19</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                         title  \\\n",
       "isbn                                                                             \n",
       "9788499083209                Trilogía de la Fundación / The Foundation Trilogy   \n",
       "9788415594765  Wonder: Charlotte tiene la palabra / Shingaling. A Wonder Story   \n",
       "9780553539233                                   Super-Pets! (DC Super Friends)   \n",
       "9780451419903                                               Mark of the Hunter   \n",
       "9780375806131                                                     The Name Jar   \n",
       "\n",
       "              onsale_date  \n",
       "isbn                       \n",
       "9788499083209  2022-03-22  \n",
       "9788415594765  2022-04-05  \n",
       "9780553539233  2022-04-05  \n",
       "9780451419903  2022-04-26  \n",
       "9780375806131  2022-07-19  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# The newest (their selling will start soon)\n",
    "books_sorted_onsale[['title', 'onsale_date']].tail(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Language"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The column `language` contains book languages."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "E     64232\n",
       "SP      691\n",
       "FR        2\n",
       "JA        1\n",
       "CH        1\n",
       "IT        1\n",
       "LA        1\n",
       "Name: language, dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# There are books in different languages:\n",
    "books['language'].value_counts(dropna=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The majority of books, of course, in English. Convert the language codes to their full names:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "languages = {\n",
    "    'E': 'English',\n",
    "    'SP': 'Spanish',\n",
    "    'FR': 'French',\n",
    "    'JA': 'Japanese',\n",
    "    'CH': 'Chinese',\n",
    "    'IT': 'Italian',\n",
    "    'LA': 'Latin',\n",
    "}\n",
    "\n",
    "books['language'].replace(languages, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Cover"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These values are just images of book covers:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"https://images.randomhouse.com/cover/9780375806131\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Look at a random image\n",
    "Image(url=books.loc[9780375806131, 'cover'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Title"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Book titles. Check if there are missing titles:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "books['title'].isna().any()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Price"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The dataset contains `price` information in different currencies:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "isbn\n",
       "9780375760686    [{'amount': 21.0, 'currencyCode': 'CAD', 'pricingType': None}, {'amount': ...\n",
       "9781605295336    [{'amount': 16.99, 'currencyCode': 'USD', 'pricingType': None}, {'amount':...\n",
       "9780771095764    [{'amount': 19.99, 'currencyCode': 'CAD', 'pricingType': None}, {'amount':...\n",
       "9780679751403    [{'amount': 17.95, 'currencyCode': 'CAD', 'pricingType': None}, {'amount':...\n",
       "9780812974874    [{'amount': 18.0, 'currencyCode': 'USD', 'pricingType': None}, {'amount': ...\n",
       "Name: price, dtype: object"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "books['price'].head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "book_price = books[['price']].explode('price')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "However, there are some text NaNs:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "book_price['price'].isna().any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "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>isbn</th>\n",
       "      <th>amount</th>\n",
       "      <th>currencyCode</th>\n",
       "      <th>pricingType</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9780375760686</td>\n",
       "      <td>21.00</td>\n",
       "      <td>CAD</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9780375760686</td>\n",
       "      <td>16.00</td>\n",
       "      <td>USD</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9781605295336</td>\n",
       "      <td>16.99</td>\n",
       "      <td>USD</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9781605295336</td>\n",
       "      <td>19.99</td>\n",
       "      <td>CAD</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9780771095764</td>\n",
       "      <td>19.99</td>\n",
       "      <td>CAD</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>9780771095764</td>\n",
       "      <td>17.95</td>\n",
       "      <td>USD</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>9780679751403</td>\n",
       "      <td>17.95</td>\n",
       "      <td>CAD</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>9780679751403</td>\n",
       "      <td>15.95</td>\n",
       "      <td>USD</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9780812974874</td>\n",
       "      <td>18.00</td>\n",
       "      <td>USD</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9780812974874</td>\n",
       "      <td>24.00</td>\n",
       "      <td>CAD</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            isbn  amount currencyCode pricingType\n",
       "0  9780375760686   21.00          CAD        None\n",
       "1  9780375760686   16.00          USD        None\n",
       "2  9781605295336   16.99          USD        None\n",
       "3  9781605295336   19.99          CAD        None\n",
       "4  9780771095764   19.99          CAD        None\n",
       "5  9780771095764   17.95          USD        None\n",
       "6  9780679751403   17.95          CAD        None\n",
       "7  9780679751403   15.95          USD        None\n",
       "8  9780812974874   18.00          USD        None\n",
       "9  9780812974874   24.00          CAD        None"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Drop nans and convert from dict to data frame\n",
    "book_price = book_price.dropna().reset_index()\n",
    "book_price = book_price.drop('price', axis=1)\\\n",
    "    .join(pd.DataFrame(book_price['price'].tolist()))\n",
    "book_price.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Empty column\n",
    "book_price['pricingType'].isna().all()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "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>currencyCode</th>\n",
       "      <th>CAD</th>\n",
       "      <th>USD</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>isbn</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9780131872783</th>\n",
       "      <td>23.0</td>\n",
       "      <td>17.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780132157575</th>\n",
       "      <td>73.0</td>\n",
       "      <td>55.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780133036039</th>\n",
       "      <td>21.0</td>\n",
       "      <td>20.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780133102109</th>\n",
       "      <td>29.0</td>\n",
       "      <td>25.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780135853986</th>\n",
       "      <td>23.0</td>\n",
       "      <td>17.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "currencyCode    CAD   USD\n",
       "isbn                     \n",
       "9780131872783  23.0  17.0\n",
       "9780132157575  73.0  55.0\n",
       "9780133036039  21.0  20.0\n",
       "9780133102109  29.0  25.0\n",
       "9780135853986  23.0  17.0"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get pivot table with prices in different currencies\n",
    "book_price = book_price.pivot(index='isbn', columns='currencyCode',\n",
    "                              values='amount')\n",
    "book_price.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# All USD values are present\n",
    "book_price['USD'].isna().any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Join prices with the main dataset\n",
    "# Keep only USD\n",
    "books = books.join(book_price[['USD']])\n",
    "\n",
    "# Drop the old column with price\n",
    "books.drop(columns=['price'], inplace=True)\n",
    "\n",
    "# Rename USD to price \n",
    "books.rename(columns={'USD': 'price'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    64920.000000\n",
       "mean        15.293190\n",
       "std          8.766339\n",
       "min          0.990000\n",
       "25%          8.990000\n",
       "50%         15.000000\n",
       "75%         18.000000\n",
       "max        429.790000\n",
       "Name: price, dtype: float64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show the distribution of the prices\n",
    "books['price'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show the distribution of the prices\n",
    "_, ax = plt.subplots(figsize=(10, 8))\n",
    "sns.histplot(x=books['price'], bins=100,\n",
    "             palette=sns_palette, ax=ax)\n",
    "ax.set_xlim(0, 75)\n",
    "ax.set_ylabel('Number of books')\n",
    "ax.set_xlabel('Price')\n",
    "ax.set_title('Distribution of book prices in USD')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "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>title</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>isbn</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9780739352243</th>\n",
       "      <td>Harry Potter 1-7 Audio Collection</td>\n",
       "      <td>429.79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780739352960</th>\n",
       "      <td>George R. R. Martin Song of Ice and Fire Audiobook Bundle</td>\n",
       "      <td>340.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780385537933</th>\n",
       "      <td>Sycamore Row - Limited Edition</td>\n",
       "      <td>250.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780385510448</th>\n",
       "      <td>The Last Juror (Limited Edition)</td>\n",
       "      <td>250.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780385503839</th>\n",
       "      <td>The Summons (Limited Edition)</td>\n",
       "      <td>250.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780385351478</th>\n",
       "      <td>Robert A. Caro's The Years of Lyndon Johnson Set</td>\n",
       "      <td>224.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780307700766</th>\n",
       "      <td>The Decline and Fall of the Roman Empire, Volumes 1 to 6</td>\n",
       "      <td>180.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780723257639</th>\n",
       "      <td>The Original Peter Rabbit Presentation Box 1-23 R/I</td>\n",
       "      <td>170.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781401234591</th>\n",
       "      <td>The Invisibles Omnibus</td>\n",
       "      <td>150.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781401255305</th>\n",
       "      <td>JSA Omnibus Vol. 3</td>\n",
       "      <td>150.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                   title  \\\n",
       "isbn                                                                       \n",
       "9780739352243                          Harry Potter 1-7 Audio Collection   \n",
       "9780739352960  George R. R. Martin Song of Ice and Fire Audiobook Bundle   \n",
       "9780385537933                             Sycamore Row - Limited Edition   \n",
       "9780385510448                           The Last Juror (Limited Edition)   \n",
       "9780385503839                              The Summons (Limited Edition)   \n",
       "9780385351478           Robert A. Caro's The Years of Lyndon Johnson Set   \n",
       "9780307700766   The Decline and Fall of the Roman Empire, Volumes 1 to 6   \n",
       "9780723257639        The Original Peter Rabbit Presentation Box 1-23 R/I   \n",
       "9781401234591                                     The Invisibles Omnibus   \n",
       "9781401255305                                         JSA Omnibus Vol. 3   \n",
       "\n",
       "                price  \n",
       "isbn                   \n",
       "9780739352243  429.79  \n",
       "9780739352960  340.00  \n",
       "9780385537933  250.00  \n",
       "9780385510448  250.00  \n",
       "9780385503839  250.00  \n",
       "9780385351478  224.95  \n",
       "9780307700766  180.00  \n",
       "9780723257639  170.00  \n",
       "9781401234591  150.00  \n",
       "9781401255305  150.00  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show the most expensive books\n",
    "books_price_sorted = books.sort_values(['price'], ascending=False)\n",
    "books_price_sorted[['title', 'price']].head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Format"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here, there are several book formats:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Paperback      41043\n",
       "Ebook          10501\n",
       "Hardcover       9645\n",
       "Audio           2854\n",
       "Board Book       653\n",
       "Merchandise      122\n",
       "Boxed Set        111\n",
       "Name: format_family, dtype: int64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "books['format_family'].value_counts(dropna=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "They are already well written. There is nothing to improve here, just to rename:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "books.rename(columns={'format_family': 'format'}, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Pages and Minutes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are two columns considering length of book: pages and projected_minutes (for audiobooks):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "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>pages</th>\n",
       "      <th>projected_minutes</th>\n",
       "      <th>format</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>isbn</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9780375760686</th>\n",
       "      <td>352.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Paperback</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781605295336</th>\n",
       "      <td>400.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Paperback</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780771095764</th>\n",
       "      <td>336.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Paperback</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780679751403</th>\n",
       "      <td>224.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Paperback</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780812974874</th>\n",
       "      <td>320.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Paperback</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               pages  projected_minutes     format\n",
       "isbn                                              \n",
       "9780375760686  352.0                NaN  Paperback\n",
       "9781605295336  400.0                NaN  Paperback\n",
       "9780771095764  336.0                NaN  Paperback\n",
       "9780679751403  224.0                NaN  Paperback\n",
       "9780812974874  320.0                NaN  Paperback"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "books[['pages', 'projected_minutes', 'format']].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "audio_mask = books['format'] == 'Audio'\n",
    "audio_books = books.loc[audio_mask, ['title', 'pages', 'projected_minutes']]\n",
    "nonaudio_books = books.loc[~audio_mask, ['title', 'pages', 'projected_minutes']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# There are NaNs\n",
    "audio_books['projected_minutes'].isna().any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# There are NaNs\n",
    "nonaudio_books['pages'].isna().any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Prove that audiobooks have no pages\n",
    "audio_books['pages'].isna().all()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Prove that non-audio books have no minutes\n",
    "nonaudio_books['projected_minutes'].isna().all()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "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>title</th>\n",
       "      <th>pages</th>\n",
       "      <th>projected_minutes</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>isbn</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9780345544742</th>\n",
       "      <td>The Complete Vampire Chronicles 12-Book Bundle</td>\n",
       "      <td>5392.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781101965481</th>\n",
       "      <td>George R. R. Martin's A Game of Thrones Leather-Cloth Boxed Set (Song of I...</td>\n",
       "      <td>5280.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780345535535</th>\n",
       "      <td>George R. R. Martin's A Game of Thrones 5-Book Boxed Set (Song of Ice and ...</td>\n",
       "      <td>5216.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780141398877</th>\n",
       "      <td>Little Black Classics Box Set</td>\n",
       "      <td>5120.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780345546005</th>\n",
       "      <td>The Earth's Children Series 6-Book Bundle</td>\n",
       "      <td>4528.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780307700766</th>\n",
       "      <td>The Decline and Fall of the Roman Empire, Volumes 1 to 6</td>\n",
       "      <td>3980.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780804180740</th>\n",
       "      <td>The Will Trent Series 7-Book Bundle</td>\n",
       "      <td>3824.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780141198415</th>\n",
       "      <td>Major Works of Charles Dickens (Penguin Classics hardcover boxed set)</td>\n",
       "      <td>3488.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780307930675</th>\n",
       "      <td>The Inheritance Cycle 4-Book Hard Cover Boxed Set</td>\n",
       "      <td>2912.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780345538260</th>\n",
       "      <td>The Highlander Series 7-Book Bundle</td>\n",
       "      <td>2816.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                       title  \\\n",
       "isbn                                                                                           \n",
       "9780345544742                                 The Complete Vampire Chronicles 12-Book Bundle   \n",
       "9781101965481  George R. R. Martin's A Game of Thrones Leather-Cloth Boxed Set (Song of I...   \n",
       "9780345535535  George R. R. Martin's A Game of Thrones 5-Book Boxed Set (Song of Ice and ...   \n",
       "9780141398877                                                  Little Black Classics Box Set   \n",
       "9780345546005                                      The Earth's Children Series 6-Book Bundle   \n",
       "9780307700766                       The Decline and Fall of the Roman Empire, Volumes 1 to 6   \n",
       "9780804180740                                            The Will Trent Series 7-Book Bundle   \n",
       "9780141198415          Major Works of Charles Dickens (Penguin Classics hardcover boxed set)   \n",
       "9780307930675                              The Inheritance Cycle 4-Book Hard Cover Boxed Set   \n",
       "9780345538260                                            The Highlander Series 7-Book Bundle   \n",
       "\n",
       "                pages  projected_minutes  \n",
       "isbn                                      \n",
       "9780345544742  5392.0                NaN  \n",
       "9781101965481  5280.0                NaN  \n",
       "9780345535535  5216.0                NaN  \n",
       "9780141398877  5120.0                NaN  \n",
       "9780345546005  4528.0                NaN  \n",
       "9780307700766  3980.0                NaN  \n",
       "9780804180740  3824.0                NaN  \n",
       "9780141198415  3488.0                NaN  \n",
       "9780307930675  2912.0                NaN  \n",
       "9780345538260  2816.0                NaN  "
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show the longest non-audio books\n",
    "nonaudio_books.sort_values('pages', ascending=False).head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "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>title</th>\n",
       "      <th>pages</th>\n",
       "      <th>projected_minutes</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>isbn</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9780307914187</th>\n",
       "      <td>The Power Broker</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3967.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780739375976</th>\n",
       "      <td>A Dance with Dragons</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2940.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780739375983</th>\n",
       "      <td>A Dance with Dragons</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2935.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780739308738</th>\n",
       "      <td>A Storm of Swords</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2854.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780449011904</th>\n",
       "      <td>A Storm of Swords</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2850.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780804164818</th>\n",
       "      <td>Tune In</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2625.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781101436875</th>\n",
       "      <td>Washington</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2510.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781415936245</th>\n",
       "      <td>The Autobiography of Henry VIII</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2478.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780307576606</th>\n",
       "      <td>Defend the Realm</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2378.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781101912812</th>\n",
       "      <td>Assassin's Fate</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2356.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         title  pages  projected_minutes\n",
       "isbn                                                                    \n",
       "9780307914187                 The Power Broker    NaN             3967.0\n",
       "9780739375976             A Dance with Dragons    NaN             2940.0\n",
       "9780739375983             A Dance with Dragons    NaN             2935.0\n",
       "9780739308738                A Storm of Swords    NaN             2854.0\n",
       "9780449011904                A Storm of Swords    NaN             2850.0\n",
       "9780804164818                          Tune In    NaN             2625.0\n",
       "9781101436875                       Washington    NaN             2510.0\n",
       "9781415936245  The Autobiography of Henry VIII    NaN             2478.0\n",
       "9780307576606                 Defend the Realm    NaN             2378.0\n",
       "9781101912812                  Assassin's Fate    NaN             2356.0"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show the longest audiobooks\n",
    "audio_books.sort_values('projected_minutes', ascending=False).head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Flapcopy and Excerpt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are several columns with text description:\n",
    "\n",
    "- `flapcopy`, the brief summary that often appears on the inside of a hardcover book's dust jacket.\n",
    "- `excerpt`, a fragment from the book."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "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>flapcopy</th>\n",
       "      <th>excerpt</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>isbn</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9780375760686</th>\n",
       "      <td>Following the tremendous success of her first book, a nonfiction work on h...</td>\n",
       "      <td>&lt;b&gt;1&lt;/b&gt;&lt;br&gt;&lt;br&gt;This neighborhood was always middle-class and urban. The a...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781605295336</th>\n",
       "      <td>For more than 40 years, &lt;i&gt;Runner's World &lt;/i&gt;magazine has been the world'...</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780771095764</th>\n",
       "      <td>&lt;b&gt;Named NOW Magazine&amp;rsquo;s Best Emerging Local Author &lt;br&gt;&lt;br&gt;&lt;br&gt;Where...</td>\n",
       "      <td>&lt;b&gt;Excerpt from Chapter 3:&lt;br&gt;&lt;/b&gt;&lt;br&gt;Wednesday after school, Mom and I to...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780679751403</th>\n",
       "      <td>In an astonishing feat of literary detection, one of the most provocative ...</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780812974874</th>\n",
       "      <td>&lt;b&gt;&lt;i&gt;New Yorker &lt;/i&gt;writer and author of &lt;i&gt;The Library Book&lt;/i&gt; takes re...</td>\n",
       "      <td>Lifelike&lt;br&gt;&lt;br&gt; As soon as the 2003 World Taxidermy Championships opened,...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                    flapcopy  \\\n",
       "isbn                                                                                           \n",
       "9780375760686  Following the tremendous success of her first book, a nonfiction work on h...   \n",
       "9781605295336  For more than 40 years, <i>Runner's World </i>magazine has been the world'...   \n",
       "9780771095764  <b>Named NOW Magazine&rsquo;s Best Emerging Local Author <br><br><br>Where...   \n",
       "9780679751403  In an astonishing feat of literary detection, one of the most provocative ...   \n",
       "9780812974874  <b><i>New Yorker </i>writer and author of <i>The Library Book</i> takes re...   \n",
       "\n",
       "                                                                                     excerpt  \n",
       "isbn                                                                                          \n",
       "9780375760686  <b>1</b><br><br>This neighborhood was always middle-class and urban. The a...  \n",
       "9781605295336                                                                           None  \n",
       "9780771095764  <b>Excerpt from Chapter 3:<br></b><br>Wednesday after school, Mom and I to...  \n",
       "9780679751403                                                                           None  \n",
       "9780812974874  Lifelike<br><br> As soon as the 2003 World Taxidermy Championships opened,...  "
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "books[['flapcopy', 'excerpt']].head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The values contain a lot of HTML tags. Let's remove them:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "def clean_text(text: str) -> str:\n",
    "    \"\"\"Clean text from HTML tags.\"\"\"\n",
    "    return BeautifulSoup(text, 'lxml').getText()\\\n",
    "        .replace(u'\\xa0', ' ').strip()\n",
    "\n",
    "\n",
    "for col in ['flapcopy', 'excerpt']:\n",
    "    nans = books[col].isna()\n",
    "    books.loc[~nans, col] = books.loc[~nans, col].apply(clean_text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0% of column flapcopy are absent.\n",
      "56% of column excerpt are absent.\n"
     ]
    }
   ],
   "source": [
    "# Show how many descriptions are absent\n",
    "for col in ['flapcopy', 'excerpt']:\n",
    "    nans_count = books[col].isna().sum()\n",
    "    print(f'{round(100 * nans_count / len(books))}% of column {col} are absent.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Publisher"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The column contains book publishers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "isbn\n",
       "9780375760686       {'code': '7070', 'description': 'Random House Publishing Group'}\n",
       "9781605295336                      {'code': '5H5H', 'description': 'Harmony/Rodale'}\n",
       "9780771095764                {'code': '4747', 'description': 'McClelland & Stewart'}\n",
       "9780679751403    {'code': '7373', 'description': 'Knopf Doubleday Publishing Group'}\n",
       "9780812974874       {'code': '7070', 'description': 'Random House Publishing Group'}\n",
       "Name: publisher, dtype: object"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "books['publisher'].head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "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>code</th>\n",
       "      <th>description</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>isbn</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9780375760686</th>\n",
       "      <td>7070</td>\n",
       "      <td>Random House Publishing Group</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781605295336</th>\n",
       "      <td>5H5H</td>\n",
       "      <td>Harmony/Rodale</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780771095764</th>\n",
       "      <td>4747</td>\n",
       "      <td>McClelland &amp; Stewart</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780679751403</th>\n",
       "      <td>7373</td>\n",
       "      <td>Knopf Doubleday Publishing Group</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780812974874</th>\n",
       "      <td>7070</td>\n",
       "      <td>Random House Publishing Group</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               code                       description\n",
       "isbn                                                 \n",
       "9780375760686  7070     Random House Publishing Group\n",
       "9781605295336  5H5H                    Harmony/Rodale\n",
       "9780771095764  4747              McClelland & Stewart\n",
       "9780679751403  7373  Knopf Doubleday Publishing Group\n",
       "9780812974874  7070     Random House Publishing Group"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Parse dicts\n",
    "book_publisher = books[['publisher']].dropna().reset_index()\n",
    "book_publisher = book_publisher.drop('publisher', axis=1)\\\n",
    "    .join(pd.DataFrame(book_publisher['publisher'].tolist()))\n",
    "book_publisher.set_index('isbn', inplace=True)\n",
    "book_publisher.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "code           False\n",
       "description    False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# If there are any NaNs\n",
    "book_publisher.isna().any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Penguin Publishing Group                       16007\n",
       "Random House Publishing Group                  10563\n",
       "Knopf Doubleday Publishing Group                6970\n",
       "Penguin Young Readers Group                     3860\n",
       "Random House Children's Books                   3676\n",
       "Penguin Random House Audio Publishing Group     2826\n",
       "Crown                                           1810\n",
       "Candlewick Press                                1526\n",
       "Clarkson Potter/Ten Speed                       1177\n",
       "The Crown Publishing Group                      1060\n",
       "Name: description, dtype: int64"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# The most frequent publishers\n",
    "book_publisher['description'].value_counts().head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Join parsed publishers with the main dataframe\n",
    "# Keep only publisher's code\n",
    "books = books.join(book_publisher[['code']])\n",
    "\n",
    "# Drop the old column with publishers\n",
    "books.drop(columns=['publisher'], inplace=True)\n",
    "\n",
    "# Rename publisher code to publisher_id \n",
    "books.rename(columns={'code': 'publisher_id'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "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>name</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>publisher_id</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7070</th>\n",
       "      <td>Random House Publishing Group</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5H5H</th>\n",
       "      <td>Harmony/Rodale</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4747</th>\n",
       "      <td>McClelland &amp; Stewart</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7373</th>\n",
       "      <td>Knopf Doubleday Publishing Group</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6262</th>\n",
       "      <td>Penguin Publishing Group</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                          name\n",
       "publisher_id                                  \n",
       "7070             Random House Publishing Group\n",
       "5H5H                            Harmony/Rodale\n",
       "4747                      McClelland & Stewart\n",
       "7373          Knopf Doubleday Publishing Group\n",
       "6262                  Penguin Publishing Group"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Move publisher data to a separate dataframe\n",
    "publishers = book_publisher.drop_duplicates('code')\\\n",
    "    .rename(columns={'code': 'publisher_id', 'description': 'name'})\\\n",
    "    .set_index('publisher_id')\n",
    "publishers.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Series"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's extract the series information. Each row contains a `series_number` and info about `series` itself."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "48726 of 64929 books are not part of any series.\n"
     ]
    }
   ],
   "source": [
    "# Most of the books are not in any series\n",
    "no_series = books['series_number'].isna().sum()\n",
    "print(f'{no_series} of {len(books)} books are not part of any series.')\n",
    "\n",
    "# Convert series_number to int data type\n",
    "books['series_number'] = books['series_number'].astype('Int32')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "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>series</th>\n",
       "      <th>series_number</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>isbn</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9780375760686</th>\n",
       "      <td>[]</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781605295336</th>\n",
       "      <td>[{'series_id': 'D22', 'name': 'Runner's World', 'description': None, 'seri...</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780771095764</th>\n",
       "      <td>[]</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780679751403</th>\n",
       "      <td>[]</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780812974874</th>\n",
       "      <td>[]</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                      series  \\\n",
       "isbn                                                                                           \n",
       "9780375760686                                                                             []   \n",
       "9781605295336  [{'series_id': 'D22', 'name': 'Runner's World', 'description': None, 'seri...   \n",
       "9780771095764                                                                             []   \n",
       "9780679751403                                                                             []   \n",
       "9780812974874                                                                             []   \n",
       "\n",
       "               series_number  \n",
       "isbn                          \n",
       "9780375760686           <NA>  \n",
       "9781605295336           <NA>  \n",
       "9780771095764           <NA>  \n",
       "9780679751403           <NA>  \n",
       "9780812974874           <NA>  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "books[['series', 'series_number']].head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "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>series_id</th>\n",
       "      <th>name</th>\n",
       "      <th>description</th>\n",
       "      <th>series_count</th>\n",
       "      <th>is_numbered</th>\n",
       "      <th>is_kids</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>isbn</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9781605295336</th>\n",
       "      <td>D22</td>\n",
       "      <td>Runner's World</td>\n",
       "      <td>None</td>\n",
       "      <td>37</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781601427762</th>\n",
       "      <td>DXR</td>\n",
       "      <td>Double S Ranch</td>\n",
       "      <td>Come Home to the Double S Ranch...&lt;br&gt;Where the Cowboys are Loving, Strong...</td>\n",
       "      <td>3</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781601427809</th>\n",
       "      <td>DXR</td>\n",
       "      <td>Double S Ranch</td>\n",
       "      <td>Come Home to the Double S Ranch...&lt;br&gt;Where the Cowboys are Loving, Strong...</td>\n",
       "      <td>3</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780451467935</th>\n",
       "      <td>BVP</td>\n",
       "      <td>Sugarland Blue Novel</td>\n",
       "      <td>The sexy officers of the Sugarland Police Department struggle to keep both...</td>\n",
       "      <td>6</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780451476999</th>\n",
       "      <td>BVP</td>\n",
       "      <td>Sugarland Blue Novel</td>\n",
       "      <td>The sexy officers of the Sugarland Police Department struggle to keep both...</td>\n",
       "      <td>6</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              series_id                  name  \\\n",
       "isbn                                            \n",
       "9781605295336       D22        Runner's World   \n",
       "9781601427762       DXR        Double S Ranch   \n",
       "9781601427809       DXR        Double S Ranch   \n",
       "9780451467935       BVP  Sugarland Blue Novel   \n",
       "9780451476999       BVP  Sugarland Blue Novel   \n",
       "\n",
       "                                                                                 description  \\\n",
       "isbn                                                                                           \n",
       "9781605295336                                                                           None   \n",
       "9781601427762  Come Home to the Double S Ranch...<br>Where the Cowboys are Loving, Strong...   \n",
       "9781601427809  Come Home to the Double S Ranch...<br>Where the Cowboys are Loving, Strong...   \n",
       "9780451467935  The sexy officers of the Sugarland Police Department struggle to keep both...   \n",
       "9780451476999  The sexy officers of the Sugarland Police Department struggle to keep both...   \n",
       "\n",
       "               series_count  is_numbered  is_kids  \n",
       "isbn                                               \n",
       "9781605295336            37        False    False  \n",
       "9781601427762             3         True    False  \n",
       "9781601427809             3         True    False  \n",
       "9780451467935             6         True    False  \n",
       "9780451476999             6         True    False  "
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Move information about series in a separate dataframe\n",
    "book_series = books[['series']].explode('series').dropna().reset_index()\n",
    "book_series = book_series.drop('series', axis=1)\\\n",
    "    .join(pd.DataFrame(book_series['series'].tolist()))\n",
    "book_series.set_index('isbn', inplace=True)\n",
    "book_series.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Drop HTML tags from description\n",
    "description_nans = book_series['description'].isna()\n",
    "book_series.loc[~description_nans, 'description'] = \\\n",
    "    book_series.loc[~description_nans, 'description'].apply(clean_text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save only `series_id` in the main dataset\n",
    "books = books.join(book_series[['series_id']])\n",
    "\n",
    "# Drop series column\n",
    "books.drop(columns=['series'], inplace=True)\n",
    "\n",
    "# Rename\n",
    "books.rename(columns={'series_number': 'number_in_series'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "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>name</th>\n",
       "      <th>description</th>\n",
       "      <th>series_count</th>\n",
       "      <th>is_numbered</th>\n",
       "      <th>is_kids</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>series_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>D22</th>\n",
       "      <td>Runner's World</td>\n",
       "      <td>None</td>\n",
       "      <td>37</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DXR</th>\n",
       "      <td>Double S Ranch</td>\n",
       "      <td>Come Home to the Double S Ranch...Where the Cowboys are Loving, Strong, an...</td>\n",
       "      <td>3</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BVP</th>\n",
       "      <td>Sugarland Blue Novel</td>\n",
       "      <td>The sexy officers of the Sugarland Police Department struggle to keep both...</td>\n",
       "      <td>6</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NSW</th>\n",
       "      <td>Star Wars</td>\n",
       "      <td>None</td>\n",
       "      <td>6</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NYC</th>\n",
       "      <td>NYRB Classics</td>\n",
       "      <td>None</td>\n",
       "      <td>31</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           name  \\\n",
       "series_id                         \n",
       "D22              Runner's World   \n",
       "DXR              Double S Ranch   \n",
       "BVP        Sugarland Blue Novel   \n",
       "NSW                   Star Wars   \n",
       "NYC               NYRB Classics   \n",
       "\n",
       "                                                                             description  \\\n",
       "series_id                                                                                  \n",
       "D22                                                                                 None   \n",
       "DXR        Come Home to the Double S Ranch...Where the Cowboys are Loving, Strong, an...   \n",
       "BVP        The sexy officers of the Sugarland Police Department struggle to keep both...   \n",
       "NSW                                                                                 None   \n",
       "NYC                                                                                 None   \n",
       "\n",
       "           series_count  is_numbered  is_kids  \n",
       "series_id                                      \n",
       "D22                  37        False    False  \n",
       "DXR                   3         True    False  \n",
       "BVP                   6         True    False  \n",
       "NSW                   6         True    False  \n",
       "NYC                  31        False    False  "
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Move series data to a separate dataframe\n",
    "series = book_series.drop_duplicates('series_id').set_index('series_id')\n",
    "\n",
    "series.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4743"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# How many unique series exists\n",
    "len(series)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "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>name</th>\n",
       "      <th>description</th>\n",
       "      <th>series_count</th>\n",
       "      <th>is_numbered</th>\n",
       "      <th>is_kids</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>series_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>VCO</th>\n",
       "      <td>Vintage Contemporaries</td>\n",
       "      <td>Vintage Contemporaries offers readers stylish trade paperback editions of ...</td>\n",
       "      <td>694</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LGB</th>\n",
       "      <td>Little Golden Book</td>\n",
       "      <td>LITTLE GOLDEN BOOKS have been loved by children for over 75 years. When th...</td>\n",
       "      <td>681</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PBK</th>\n",
       "      <td>Pictureback(R)</td>\n",
       "      <td>None</td>\n",
       "      <td>273</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GNO</th>\n",
       "      <td>Good Night Our World</td>\n",
       "      <td>The Good Night Books series is designed to celebrate special places and su...</td>\n",
       "      <td>213</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DWY</th>\n",
       "      <td>Who Was?</td>\n",
       "      <td>A series of illustrated biographies for young readers featuring significan...</td>\n",
       "      <td>210</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BFV</th>\n",
       "      <td>Longarm</td>\n",
       "      <td>None</td>\n",
       "      <td>185</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BT5</th>\n",
       "      <td>Slocum</td>\n",
       "      <td>The longest-running Western series in America today, the Slocum adventures...</td>\n",
       "      <td>176</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VTS</th>\n",
       "      <td>A Vintage Short</td>\n",
       "      <td>The Vintage Shorts series presents timely reading from enduring classics a...</td>\n",
       "      <td>176</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ADI</th>\n",
       "      <td>Travel Guide</td>\n",
       "      <td>For more than two decades, DK Eyewitness Travel Guides have helped travele...</td>\n",
       "      <td>172</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DQ3</th>\n",
       "      <td>Mad Libs</td>\n",
       "      <td>Go for Mad Libs—Everyone’s favorite funny fill-in books!</td>\n",
       "      <td>167</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             name  \\\n",
       "series_id                           \n",
       "VCO        Vintage Contemporaries   \n",
       "LGB            Little Golden Book   \n",
       "PBK                Pictureback(R)   \n",
       "GNO          Good Night Our World   \n",
       "DWY                      Who Was?   \n",
       "BFV                       Longarm   \n",
       "BT5                        Slocum   \n",
       "VTS               A Vintage Short   \n",
       "ADI                  Travel Guide   \n",
       "DQ3                      Mad Libs   \n",
       "\n",
       "                                                                             description  \\\n",
       "series_id                                                                                  \n",
       "VCO        Vintage Contemporaries offers readers stylish trade paperback editions of ...   \n",
       "LGB        LITTLE GOLDEN BOOKS have been loved by children for over 75 years. When th...   \n",
       "PBK                                                                                 None   \n",
       "GNO        The Good Night Books series is designed to celebrate special places and su...   \n",
       "DWY        A series of illustrated biographies for young readers featuring significan...   \n",
       "BFV                                                                                 None   \n",
       "BT5        The longest-running Western series in America today, the Slocum adventures...   \n",
       "VTS        The Vintage Shorts series presents timely reading from enduring classics a...   \n",
       "ADI        For more than two decades, DK Eyewitness Travel Guides have helped travele...   \n",
       "DQ3                             Go for Mad Libs—Everyone’s favorite funny fill-in books!   \n",
       "\n",
       "           series_count  is_numbered  is_kids  \n",
       "series_id                                      \n",
       "VCO                 694        False    False  \n",
       "LGB                 681        False     True  \n",
       "PBK                 273        False     True  \n",
       "GNO                 213        False     True  \n",
       "DWY                 210        False     True  \n",
       "BFV                 185         True    False  \n",
       "BT5                 176         True    False  \n",
       "VTS                 176        False    False  \n",
       "ADI                 172        False    False  \n",
       "DQ3                 167        False     True  "
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show series with the largest number of books\n",
    "series.sort_values('series_count', ascending=False).head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": [],
    "toc-hr-collapsed": true
   },
   "source": [
    "### Contributors"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The column `authors` may contain illustrators, photographer in addition to the main author."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "isbn\n",
       "9780375760686    [{'author_id': 35922, 'first_name': 'Cheryl', 'last_name': 'Mendelson', 'c...\n",
       "9781605295336    [{'author_id': 2061538, 'first_name': 'David', 'last_name': 'Willey', 'com...\n",
       "9780771095764    [{'author_id': 83047, 'first_name': 'Lauren', 'last_name': 'Kirshner', 'co...\n",
       "9780679751403    [{'author_id': 18770, 'first_name': 'Janet', 'last_name': 'Malcolm', 'comp...\n",
       "9780812974874    [{'author_id': 22849, 'first_name': 'Susan', 'last_name': 'Orlean', 'compa...\n",
       "Name: authors, dtype: object"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "books['authors'].head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "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>author_id</th>\n",
       "      <th>first_name</th>\n",
       "      <th>last_name</th>\n",
       "      <th>company</th>\n",
       "      <th>client_source_id</th>\n",
       "      <th>role</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>isbn</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9780375760686</th>\n",
       "      <td>35922</td>\n",
       "      <td>Cheryl</td>\n",
       "      <td>Mendelson</td>\n",
       "      <td>{'key': 'R_H', 'value': None}</td>\n",
       "      <td>0</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781605295336</th>\n",
       "      <td>2061538</td>\n",
       "      <td>David</td>\n",
       "      <td>Willey</td>\n",
       "      <td>{'key': 'R_H', 'value': None}</td>\n",
       "      <td>0</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781605295336</th>\n",
       "      <td>2213465</td>\n",
       "      <td>None</td>\n",
       "      <td>Editors of Runner's World Maga</td>\n",
       "      <td>{'key': 'R_H', 'value': None}</td>\n",
       "      <td>0</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780771095764</th>\n",
       "      <td>83047</td>\n",
       "      <td>Lauren</td>\n",
       "      <td>Kirshner</td>\n",
       "      <td>{'key': 'R_H', 'value': None}</td>\n",
       "      <td>0</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780679751403</th>\n",
       "      <td>18770</td>\n",
       "      <td>Janet</td>\n",
       "      <td>Malcolm</td>\n",
       "      <td>{'key': 'R_H', 'value': None}</td>\n",
       "      <td>0</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               author_id first_name                       last_name  \\\n",
       "isbn                                                                  \n",
       "9780375760686      35922     Cheryl                       Mendelson   \n",
       "9781605295336    2061538      David                          Willey   \n",
       "9781605295336    2213465       None  Editors of Runner's World Maga   \n",
       "9780771095764      83047     Lauren                        Kirshner   \n",
       "9780679751403      18770      Janet                         Malcolm   \n",
       "\n",
       "                                     company  client_source_id role  \n",
       "isbn                                                                 \n",
       "9780375760686  {'key': 'R_H', 'value': None}                 0    A  \n",
       "9781605295336  {'key': 'R_H', 'value': None}                 0    E  \n",
       "9781605295336  {'key': 'R_H', 'value': None}                 0    E  \n",
       "9780771095764  {'key': 'R_H', 'value': None}                 0    A  \n",
       "9780679751403  {'key': 'R_H', 'value': None}                 0    A  "
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Move information about authors in a separate dataframe\n",
    "book_authors = books[['authors']].explode('authors').dropna().reset_index()\n",
    "book_authors = book_authors.drop('authors', axis=1)\\\n",
    "    .join(pd.DataFrame(book_authors['authors'].tolist()))\n",
    "book_authors.set_index('isbn', inplace=True)\n",
    "book_authors.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since a book may have several authors, we need a dataframe for mapping book-authors and another one for author information."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "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>first_name</th>\n",
       "      <th>last_name</th>\n",
       "      <th>company</th>\n",
       "      <th>client_source_id</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>author_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>35922</th>\n",
       "      <td>Cheryl</td>\n",
       "      <td>Mendelson</td>\n",
       "      <td>{'key': 'R_H', 'value': None}</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2061538</th>\n",
       "      <td>David</td>\n",
       "      <td>Willey</td>\n",
       "      <td>{'key': 'R_H', 'value': None}</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2213465</th>\n",
       "      <td>None</td>\n",
       "      <td>Editors of Runner's World Maga</td>\n",
       "      <td>{'key': 'R_H', 'value': None}</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83047</th>\n",
       "      <td>Lauren</td>\n",
       "      <td>Kirshner</td>\n",
       "      <td>{'key': 'R_H', 'value': None}</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18770</th>\n",
       "      <td>Janet</td>\n",
       "      <td>Malcolm</td>\n",
       "      <td>{'key': 'R_H', 'value': None}</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          first_name                       last_name  \\\n",
       "author_id                                              \n",
       "35922         Cheryl                       Mendelson   \n",
       "2061538        David                          Willey   \n",
       "2213465         None  Editors of Runner's World Maga   \n",
       "83047         Lauren                        Kirshner   \n",
       "18770          Janet                         Malcolm   \n",
       "\n",
       "                                 company  client_source_id  \n",
       "author_id                                                   \n",
       "35922      {'key': 'R_H', 'value': None}                 0  \n",
       "2061538    {'key': 'R_H', 'value': None}                 0  \n",
       "2213465    {'key': 'R_H', 'value': None}                 0  \n",
       "83047      {'key': 'R_H', 'value': None}                 0  \n",
       "18770      {'key': 'R_H', 'value': None}                 0  "
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Dataframe for authors info\n",
    "authors = book_authors[['author_id', 'first_name', 'last_name', \n",
    "                        'company', 'client_source_id']]\\\n",
    "    .drop_duplicates('author_id').set_index('author_id')\n",
    "authors.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "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>isbn</th>\n",
       "      <th>author_id</th>\n",
       "      <th>role_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9780375760686</td>\n",
       "      <td>35922</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9781605295336</td>\n",
       "      <td>2061538</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9781605295336</td>\n",
       "      <td>2213465</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9780771095764</td>\n",
       "      <td>83047</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9780679751403</td>\n",
       "      <td>18770</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            isbn  author_id role_id\n",
       "0  9780375760686      35922       A\n",
       "1  9781605295336    2061538       E\n",
       "2  9781605295336    2213465       E\n",
       "3  9780771095764      83047       A\n",
       "4  9780679751403      18770       A"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Dataframe for mapping books and authors\n",
    "book_authors = book_authors[['author_id', 'role']].reset_index()\n",
    "book_authors.rename(columns={'role': 'role_id'}, inplace=True)\n",
    "book_authors.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Drop authors from books\n",
    "books.drop(columns='authors', inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "In some cases, we can see the same information about authors with different ids. For example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "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>first_name</th>\n",
       "      <th>last_name</th>\n",
       "      <th>company</th>\n",
       "      <th>client_source_id</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>author_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>159086</th>\n",
       "      <td>Daniel</td>\n",
       "      <td>Mendelsohn</td>\n",
       "      <td>{'key': 'DL', 'value': 'New York Review Books'}</td>\n",
       "      <td>20334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20334</th>\n",
       "      <td>Daniel</td>\n",
       "      <td>Mendelsohn</td>\n",
       "      <td>{'key': 'R_H', 'value': None}</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          first_name   last_name  \\\n",
       "author_id                          \n",
       "159086        Daniel  Mendelsohn   \n",
       "20334         Daniel  Mendelsohn   \n",
       "\n",
       "                                                   company  client_source_id  \n",
       "author_id                                                                     \n",
       "159086     {'key': 'DL', 'value': 'New York Review Books'}             20334  \n",
       "20334                        {'key': 'R_H', 'value': None}                 0  "
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "authors.loc[[159086, 20334]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Here is the rephrased explanation from [the PRH site](https://developer.penguinrandomhouse.com/docs/read/enhanced_prh_api/concepts/Multiple_authors):\n",
    "In cases where an author contributes to books by both a core PRH division and by one of our distribution clients, MyHouse (content management system) creates a separate author ID so that our distribution clients have the opportunity to attach their own content to that ID. This is why there are two \n",
    "author records in our system.\n",
    "\n",
    "In this way, distribution authors, whose IDs were created within the content management system as copies of core authors, have `client_source_id` as a reference to the `author_id` of the core author. We can see from the values on the 159086 Daniel Mendelsohn record that it has a reference to the 20334 record."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here, we can assume that the persons themselves are more important than the company where they work. So, we can use only one ID per author and skip the company info."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "external_authors = authors[authors['client_source_id'] > 0]\n",
    "rh_authors = authors[authors['client_source_id'] == 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "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>first_name</th>\n",
       "      <th>last_name</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>author_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>35922</th>\n",
       "      <td>Cheryl</td>\n",
       "      <td>Mendelson</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2061538</th>\n",
       "      <td>David</td>\n",
       "      <td>Willey</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2213465</th>\n",
       "      <td>None</td>\n",
       "      <td>Editors of Runner's World Maga</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83047</th>\n",
       "      <td>Lauren</td>\n",
       "      <td>Kirshner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18770</th>\n",
       "      <td>Janet</td>\n",
       "      <td>Malcolm</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          first_name                       last_name\n",
       "author_id                                           \n",
       "35922         Cheryl                       Mendelson\n",
       "2061538        David                          Willey\n",
       "2213465         None  Editors of Runner's World Maga\n",
       "83047         Lauren                        Kirshner\n",
       "18770          Janet                         Malcolm"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "works_rh = external_authors['client_source_id'].isin(rh_authors.index)\n",
    "\n",
    "# If author works in RH, change to the main `source_id`\n",
    "rh_converted = external_authors.loc[works_rh, 'client_source_id'].to_dict()\n",
    "authors.rename(index=rh_converted, inplace=True)\n",
    "book_authors['author_id'].replace(rh_converted, inplace=True)\n",
    "\n",
    "# If author does not work in RH, set `client_source_id` as main\n",
    "nrh_converted = external_authors.loc[~works_rh, 'client_source_id'].to_dict()\n",
    "authors.rename(index=nrh_converted, inplace=True)\n",
    "book_authors['author_id'].replace(nrh_converted, inplace=True)\n",
    "\n",
    "# Drop duplicates and unused columns\n",
    "authors.drop(columns=['client_source_id', 'company'], inplace=True)\n",
    "authors = authors[~authors.index.duplicated(keep='first')]\n",
    "\n",
    "# Show\n",
    "authors.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Works"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here is the explanation of work from the PRH: Work ID is a Penguin Random House internally managed number that serves to group different formats of a title together. The object identified by the work ID is called the Work, and it represents the collection of all the different ISBNs together as a single object. Even though a Work is just a collection of ISBNs and does not have data of its own, certain attributes are computed at load time based on rules derived from the available ISBNs. The onsale date attribute is taken from the ISBN in the work with the oldest on-sale date. The title, subtitle and author (the display field) are derived a little differently. For each field, the ISBN values are sorted alphabetically and the first value is used as the work attribute. All other data is derived for a Work based on the ISBNs it has (more details [here](https://developer.penguinrandomhouse.com/docs/read/enhanced_prh_api/concepts/Works_and_ISBNs))."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "isbn\n",
       "9780375760686    [{'work_id': 113275, 'title': 'Morningside Heights', 'author': 'Cheryl Men...\n",
       "9781605295336    [{'work_id': 594144, 'title': 'Going Long', 'author': 'Edited by David Wil...\n",
       "9780771095764    [{'work_id': 93322, 'title': 'Where We Have to Go', 'author': 'Lauren Kirs...\n",
       "9780679751403    [{'work_id': 106482, 'title': 'The Silent Woman', 'author': 'Janet Malcolm...\n",
       "9780812974874    [{'work_id': 124840, 'title': 'My Kind of Place', 'author': 'Susan Orlean'...\n",
       "Name: works, dtype: object"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "books['works'].head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "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>work_id</th>\n",
       "      <th>title</th>\n",
       "      <th>author</th>\n",
       "      <th>onsale</th>\n",
       "      <th>language</th>\n",
       "      <th>series_number</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>isbn</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9780375760686</th>\n",
       "      <td>113275</td>\n",
       "      <td>Morningside Heights</td>\n",
       "      <td>Cheryl Mendelson</td>\n",
       "      <td>2005-07-12</td>\n",
       "      <td>E</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9781605295336</th>\n",
       "      <td>594144</td>\n",
       "      <td>Going Long</td>\n",
       "      <td>Edited by David Willey</td>\n",
       "      <td>2010-03-30</td>\n",
       "      <td>E</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780771095764</th>\n",
       "      <td>93322</td>\n",
       "      <td>Where We Have to Go</td>\n",
       "      <td>Lauren Kirshner</td>\n",
       "      <td>2012-03-06</td>\n",
       "      <td>E</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780679751403</th>\n",
       "      <td>106482</td>\n",
       "      <td>The Silent Woman</td>\n",
       "      <td>Janet Malcolm</td>\n",
       "      <td>1995-03-28</td>\n",
       "      <td>E</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9780812974874</th>\n",
       "      <td>124840</td>\n",
       "      <td>My Kind of Place</td>\n",
       "      <td>Susan Orlean</td>\n",
       "      <td>2005-10-11</td>\n",
       "      <td>E</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               work_id                title                  author  \\\n",
       "isbn                                                                  \n",
       "9780375760686   113275  Morningside Heights        Cheryl Mendelson   \n",
       "9781605295336   594144           Going Long  Edited by David Willey   \n",
       "9780771095764    93322  Where We Have to Go         Lauren Kirshner   \n",
       "9780679751403   106482     The Silent Woman           Janet Malcolm   \n",
       "9780812974874   124840     My Kind of Place            Susan Orlean   \n",
       "\n",
       "                   onsale language series_number  \n",
       "isbn                                              \n",
       "9780375760686  2005-07-12        E          None  \n",
       "9781605295336  2010-03-30        E          None  \n",
       "9780771095764  2012-03-06        E          None  \n",
       "9780679751403  1995-03-28        E          None  \n",
       "9780812974874  2005-10-11        E          None  "
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Parse dicts\n",
    "book_work = books[['works']].explode('works').dropna().reset_index()\n",
    "book_work = book_work.drop('works', axis=1)\\\n",
    "    .join(pd.DataFrame(book_work['works'].tolist()))\n",
    "book_work.set_index('isbn', inplace=True)\n",
    "book_work.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Each book have only one `work_id`\n",
    "book_work.index.duplicated().any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save only `work_id` in the main dataset\n",
    "books = books.join(book_work[['work_id']])\n",
    "\n",
    "# Drop works column\n",
    "books.drop(columns=['works'], inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To avoid the duplication between book data and work data, we can find and save only one isbn per work that will present its information. Let's use the same steps to find the front ISBN as it is written in the description from PRH."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "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>front_isbn</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>work_id</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>299263</th>\n",
       "      <td>9780131872783</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>352442</th>\n",
       "      <td>9780132157575</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>350692</th>\n",
       "      <td>9780133036039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>353576</th>\n",
       "      <td>9780133102109</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>304057</th>\n",
       "      <td>9780135853986</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            front_isbn\n",
       "work_id               \n",
       "299263   9780131872783\n",
       "352442   9780132157575\n",
       "350692   9780133036039\n",
       "353576   9780133102109\n",
       "304057   9780135853986"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Use only `work_id`\n",
    "works = book_work.loc[:, ['work_id']]\n",
    "works.reset_index(inplace=True)\n",
    "\n",
    "# Determine front ISBN\n",
    "works.sort_values('isbn', inplace=True)\n",
    "works.drop_duplicates('work_id', keep='first', inplace=True)\n",
    "works.set_index('work_id', inplace=True)\n",
    "works.rename(columns={'isbn': 'front_isbn'}, inplace=True)\n",
    "\n",
    "# Show\n",
    "works.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Categories"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Categories are sequenced such that the first category should be considered more relevant or important than the last. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "isbn\n",
       "9780375760686    [{'category_id': 2000000083, 'seq': 1}, {'category_id': 3000001343, 'seq':...\n",
       "9781605295336    [{'category_id': 2000000052, 'seq': 3}, {'category_id': 2000000144, 'seq':...\n",
       "9780771095764    [{'category_id': 2000000083, 'seq': 1}, {'category_id': 3000001343, 'seq':...\n",
       "9780679751403    [{'category_id': 2000000013, 'seq': 1}, {'category_id': 2000000082, 'seq':...\n",
       "9780812974874    [{'category_id': 2000000021, 'seq': 2}, {'category_id': 2000000154, 'seq':...\n",
       "Name: categories, dtype: object"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "books['categories'].head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "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>isbn</th>\n",
       "      <th>category_id</th>\n",
       "      <th>seq</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9780375760686</td>\n",
       "      <td>2000000083</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9780375760686</td>\n",
       "      <td>3000001343</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9780375760686</td>\n",
       "      <td>3000001361</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9780375760686</td>\n",
       "      <td>3000001404</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9781605295336</td>\n",
       "      <td>2000000052</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            isbn  category_id  seq\n",
       "0  9780375760686   2000000083    1\n",
       "1  9780375760686   3000001343    3\n",
       "2  9780375760686   3000001361    2\n",
       "3  9780375760686   3000001404    1\n",
       "4  9781605295336   2000000052    3"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Move information about categories in a separate dataframe\n",
    "book_cats = books[['categories']].explode('categories').dropna().reset_index()\n",
    "book_cats = book_cats.drop('categories', axis=1)\\\n",
    "    .join(pd.DataFrame(book_cats['categories'].tolist()))\n",
    "book_cats.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Drop category column from the main dataset\n",
    "books.drop(columns='categories', inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are a lot of category sets [in the PRH](https://developer.penguinrandomhouse.com/docs/read/enhanced_prh_api/concepts/Categories). However, we use only \"CN\" categories here. Thus, we need to remove unnecessary ones."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "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>isbn</th>\n",
       "      <th>category_id</th>\n",
       "      <th>seq</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9780375760686</td>\n",
       "      <td>2000000083</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9781605295336</td>\n",
       "      <td>2000000052</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>9781605295336</td>\n",
       "      <td>2000000144</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9780771095764</td>\n",
       "      <td>2000000083</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>9780679751403</td>\n",
       "      <td>2000000013</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             isbn  category_id  seq\n",
       "0   9780375760686   2000000083    1\n",
       "4   9781605295336   2000000052    3\n",
       "5   9781605295336   2000000144    1\n",
       "9   9780771095764   2000000083    1\n",
       "13  9780679751403   2000000013    1"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "book_cats = book_cats[book_cats['category_id'].isin(categories.index)]\n",
    "book_cats.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ratings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load all ratings\n",
    "path_ratings = os.path.join('data_interm', 'ratings_joined.csv')\n",
    "ratings = pd.read_csv(path_ratings)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of all ratings: 90008708\n"
     ]
    }
   ],
   "source": [
    "init_rating_count = len(ratings)\n",
    "print(f'Number of all ratings: {init_rating_count}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we need to remove the ratings of books that were not downloaded from PRH."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of book ratings that were downloaded from PRH: 16013867 that is 17.79% from all ratings.\n"
     ]
    }
   ],
   "source": [
    "ratings = ratings[ratings['isbn13'].isin(books.index)]\n",
    "print(f'Number of book ratings that were downloaded from PRH: '\n",
    "      f'{len(ratings)} that is {len(ratings) * 100 / init_rating_count:.2f}% '\n",
    "      f'from all ratings.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "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>rating</th>\n",
       "      <th>user_id</th>\n",
       "      <th>isbn13</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10</td>\n",
       "      <td>494492</td>\n",
       "      <td>9780767908184</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10</td>\n",
       "      <td>494492</td>\n",
       "      <td>9780345538376</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>4</td>\n",
       "      <td>494492</td>\n",
       "      <td>9780553383430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>8</td>\n",
       "      <td>494492</td>\n",
       "      <td>9780553380958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>10</td>\n",
       "      <td>494492</td>\n",
       "      <td>9780451191151</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    rating  user_id         isbn13\n",
       "1       10   494492  9780767908184\n",
       "2       10   494492  9780345538376\n",
       "18       4   494492  9780553383430\n",
       "19       8   494492  9780553380958\n",
       "29      10   494492  9780451191151"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ratings.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since people most often rate the book content rather than a particular edition of the book, we will build recommendations using `work_id` instead of `isbn13`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "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>rating</th>\n",
       "      <th>user_id</th>\n",
       "      <th>isbn13</th>\n",
       "      <th>work_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10</td>\n",
       "      <td>494492</td>\n",
       "      <td>9780767908184</td>\n",
       "      <td>20549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>558</th>\n",
       "      <td>10</td>\n",
       "      <td>607953</td>\n",
       "      <td>9780767908184</td>\n",
       "      <td>20549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6545</th>\n",
       "      <td>6</td>\n",
       "      <td>532610</td>\n",
       "      <td>9780767908184</td>\n",
       "      <td>20549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13192</th>\n",
       "      <td>10</td>\n",
       "      <td>215425</td>\n",
       "      <td>9780767908184</td>\n",
       "      <td>20549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36522</th>\n",
       "      <td>8</td>\n",
       "      <td>100677</td>\n",
       "      <td>9780767908184</td>\n",
       "      <td>20549</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       rating  user_id         isbn13  work_id\n",
       "1          10   494492  9780767908184    20549\n",
       "558        10   607953  9780767908184    20549\n",
       "6545        6   532610  9780767908184    20549\n",
       "13192      10   215425  9780767908184    20549\n",
       "36522       8   100677  9780767908184    20549"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ratings = pd.merge(ratings, books[['work_id']], \n",
    "                   left_on='isbn13', \n",
    "                   right_index=True)\n",
    "ratings.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Some users may have rated different editions (ISBNs) of the same work. Let's average their ratings so that there is only one rating per work from each user:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "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>user_id</th>\n",
       "      <th>work_id</th>\n",
       "      <th>rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>494492</td>\n",
       "      <td>20549</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>607953</td>\n",
       "      <td>20549</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>532610</td>\n",
       "      <td>20549</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>215425</td>\n",
       "      <td>20549</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>100677</td>\n",
       "      <td>20549</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id  work_id  rating\n",
       "0   494492    20549    10.0\n",
       "1   607953    20549    10.0\n",
       "2   532610    20549     6.0\n",
       "3   215425    20549    10.0\n",
       "4   100677    20549     8.0"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ratings_per_work = ratings[['work_id', 'user_id', 'rating']]\\\n",
    "    .groupby(['work_id', 'user_id'], observed=True).mean()\n",
    "\n",
    "# Drop duplicated ratings\n",
    "work_ratings = ratings.drop_duplicates(['user_id', 'work_id'], keep='first')\n",
    "work_ratings = work_ratings[['user_id', 'work_id']]\\\n",
    "    .merge(ratings_per_work, left_on=['user_id', 'work_id'],\n",
    "           right_on=['user_id', 'work_id'], how='left')\n",
    "work_ratings.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show the distribution of ratings per works\n",
    "_, ax = plt.subplots(figsize=(10, 8))\n",
    "sns.histplot(work_ratings['rating'], ax=ax, bins=10)\n",
    "ax.set_ylabel('Counts')\n",
    "ax.set_xlabel('Ratings')\n",
    "ax.set_title('Work rating distribution');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "by_work = work_ratings[['work_id', 'rating']]\\\n",
    "    .groupby('work_id', observed=True)\n",
    "by_work_count = by_work[['rating']].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of works which have ratings: 53192 that is 100.0000%\n",
      "Number of works which have at least 5 ratings: 42841 that is 80.5403%\n"
     ]
    }
   ],
   "source": [
    "work_rated_count = len(by_work_count)\n",
    "work_5times_rated_count = (by_work_count['rating'] >= 5).sum()\n",
    "\n",
    "print(f'Number of works which have ratings: {work_rated_count} '\n",
    "      f'that is {work_rated_count * 100 / len(works):.4f}%')\n",
    "print(f'Number of works which have at least 5 ratings: '\n",
    "      f'{work_5times_rated_count} that is '\n",
    "      f'{work_5times_rated_count * 100 / len(works):.4f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's calculate weighted ratings of books. To use usual rating is not the best idea, because a book with a rating of 9 from 10 voters will be considered \"better\" than a book with a rating of 8.9 from 10,000 voters. Thus, it is better to use \"weighted rating\". See more details [in the Datacamp article](https://www.datacamp.com/community/tutorials/recommender-systems-python)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate weighted rating\n",
    "min_work_rate_count = by_work_count['rating'].quantile(0.99)\n",
    "mean_work_rate = work_ratings['rating'].mean()\n",
    "work_ids = by_work_count[by_work_count['rating'] >= min_work_rate_count].index\n",
    "by_work_weighted = work_ratings\\\n",
    "    .loc[work_ratings['work_id'].isin(work_ids), ['work_id', 'rating']]\\\n",
    "    .groupby('work_id', observed=True).apply(get_weighted_rating,\n",
    "                                             min_rate_count=min_work_rate_count,\n",
    "                                             mean_rate=mean_work_rate)\\\n",
    "    .to_frame(name='weighted_rating')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get work info\n",
    "works_data = pd.merge(works[['front_isbn']], books[['title']],\n",
    "                      left_on='front_isbn', right_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABUEAAAJzCAYAAAAhjdKIAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8/fFQqAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOzdeVxN+f8H8FelIomSZF/CRWlVaUHIMHayKwoTkn0f2yy2VEarXRFDM0pkmTFjDGOrZrKMdWY0lmwlWmm/vz/63fPtdm91S0Tzej4ePcY553M+533O/dzb3HefRUksFotBREREREREREREVEMpV3cARERERERERERERO8Sk6BERERERERERERUozEJSkRERERERERERDUak6BERERERERERERUozEJSkRERERERERERDUak6BERERERERERERUozEJSkRERERERESlioyMhEgkQkxMTLllXVxc0Lt37/cQVeliYmIgEokQGRlZrXEQ0YelVnUHQERERERERGUTiUQKlz19+jSaN2/+DqMpkpOTgyNHjuDMmTO4e/cuXrx4gUaNGsHExAQzZ86EgYGBzDm5ubnYsmULjhw5gqSkJOjr62PEiBH47LPPoKqq+s5jfldu376Nn3/+GcOHD38vz76kgIAAdOrUCY6Oju/92kREHwsmQYmIiIiIiD5wGzdulNr+448/EB4ejjFjxsDCwkLqmI6OznuJKTExEStXroSFhQWcnJygp6eHxMREHDhwAKdOncLOnTvRrVs3qXPmzp2L06dPw8nJCWZmZrhy5Qr8/Pzw8OFDbNiw4b3E/S7cvn0bgYGBsLKyqpYkaGBgIIYPH84kKBFRGZgEJSIiIiIi+sANHTpUarugoADh4eEwNTWVOfa+6OjoICoqCp06dZLaP3jwYAwbNgwbN26UGo589uxZnD59Gm5ubli6dCkAYNSoUdDS0kJISAhGjx4Nc3Pz93oPH5rs7GzUqlULtWrxqzoRUVXjnKBEREREREQ1xOvXr+Hr6wtHR0cYGRnBzs4OixcvxuPHj6XKFZ8zMSwsDP369UOXLl3Qr18/hIWFKXQtbW1tmQQoALRr1w4dOnTA33//LbU/OjoaADBp0iSp/ZLto0ePlnvN4nHv379fiHvw4MH49ddfAQB3797FlClTYG5uDmtra6xZswZ5eXkydcXFxcHNzQ0WFhYwNjbG8OHD8f3338uU+/vvvzF79mx0795deKYuLi7C9QICArBs2TIAwMSJEyESiSASiYREb2mWLl0KkUiEly9fYtmyZbC1tYWpqSmePXsGANi/fz8mT54sXNfe3h4LFy5EYmKiUEdiYqIwVcLhw4eFa5ecPuHixYuYPHkyunbtKjyvAwcOyI3r+++/R//+/WFkZIS+fftiz549EIvFZd6LPI8ePcKMGTNgYWEBc3NzzJw5E48ePZIpp2ibrWhZeQ4fPgxDQ0PMnj0bubm5AID4+HhMnToVdnZ26NKlC7p3747PPvsMV69erfA9E9GHjX9eIiIiIiIiqgHy8/MxZcoUxMfHo1+/fnBzc8ODBw9w4MABXLhwAREREdDX15c6Z9++fUhOTsaYMWOgqamJY8eOYc2aNUhLS4Onp2el4igsLERSUhJ0dXWl9v/5559o3LgxmjRpIrW/SZMm0NPTw59//qnwNfbv34/09HSMGjUKampqCAsLw8yZM+Hn54cVK1Zg0KBBcHR0xIULFxAWFgYdHR14eHgI5//yyy/w9PSErq4u3NzcoKmpiePHj2PFihVITEzEvHnzAACvXr0SkrRjx45F06ZN8erVK9y4cQPXrl2Dg4MD+vbti+TkZISHh2P69Olo27YtAKBly5YK3Yubmxt0dXXh4eGB169fQ0NDAwCwe/dumJqawsXFBQ0aNMBff/2FQ4cO4fLly4iOjoa2tjZ0dHSwceNGLF68GF27dsXo0aNl6g8PD8fq1athamqK6dOno06dOrh48SK++OILPHz4EEuWLBHKhoaGYv369ejYsSPmz5+PN2/eYNeuXWjYsKHCrw1QlKycOHEiunTpgvnz5+PBgwf49ttvce3aNRw+fBiNGjUCULE2W5n2XdzWrVvxzTffYMKECVixYgWUlZWRkJCAyZMnQ1dXFxMnTkTDhg3x4sULxMfH486dOzA1Na3QfRPRB05MREREREREH5WIiAhxhw4dxBEREcK+8PBwcYcOHcReXl5SZc+cOSPu0KGDeOHChcK+y5cvizt06CA2NTUVP336VNifk5MjdnJyEnfu3Flqf0Xs27dP3KFDB/HmzZul9puamopHjhwp9xwnJyexnZ1duXVL4ra3txenp6cL+2/fvi3u0KGDWCQSiX/88Uepc4YPHy5Vd35+vtjBwUFsYWEhfvbsmbA/JydHPGbMGHHHjh3F//77r1gsFot//vlncYcOHcTHjx8vMy7J63H58uVy70FiyZIl4g4dOogXLFgg93hWVpbMvosXL4o7dOgg3r59u9T+Dh06iJcsWSJT/vnz52IjIyPx/PnzZY59/fXX4o4dO4ofPHggFovF4rS0NLGJiYn4008/Fb9+/Voo9/TpU7GpqanC9+fs7Czu0KGDeM2aNVL7T506Je7QoYN45cqVwr6KtNnKtO+IiAhxQUGB+IsvvhB36NBBvGXLFqlz9+zZI+7QoYP42rVr5d4XEX38OByeiIiIiIioBvjpp5+grKyMadOmSe13cHBAp06dcPr0aRQWFkodGzx4sFTvOTU1Nbi6uiI/Px+//PJLhWOIj4/Hhg0bIBKJMH36dKlj2dnZUFNTk3ueuro6srOzFb7OiBEjUK9ePWG7Y8eO0NTUhJ6eHj755BOpsubm5khOTkZWVhYA4ObNm3jy5AmcnJzQuHFjoZyamhqmTp2KwsJCnD59GgCEa/z222/IzMxUOL6KmDJlitz9kh6hhYWFyMjIwMuXLyESiVCvXj1cv35dobp//PFH5ObmYuTIkXj58qXUT+/evVFYWIhLly4BAM6fP483b95gwoQJqFOnjlCHvr4+Bg8eXOH7cnd3l9ru27cv2rRpIzxboGJttjLtOycnB7Nnz8Z3332HDRs2yLRJyet7+vRp5OTkVPgeiejjwuHwRERERERENUBiYiL09PRQv359mWPt2rXD7du38erVK6mhzQYGBnLLApA7f2NZbty4gWnTpkFPTw87duyAurq61PHatWsL8zCWlJOTg9q1ayt8LXkrsNevX1/ucGjJ80hNTUXdunWFOTUl91lc+/btAfzv3q2srDBs2DBERkYiOjoaRkZGsLW1xYABA+SeXxmtW7eWu//SpUsIDg7GtWvXZBJ0aWlpCtV97949AICrq2upZV68eAEAwnORDOcvTl47KYuWlpYw5L1kPT///LMw7L8ibbYy7dvb2xtZWVnw8fGRm8gdOHAgjh49iq1btyI0NBQmJiawt7fHwIED0axZswrdMxF9+JgEJSIiIiIiqgHElVi8RklJqUrquXnzJiZPnox69eph7969Uj0sJfT09PD8+XO55z9//lzuOaVRUVGp0H7gf/dV0fvz8vLClClTcPbsWfzxxx8ICQnB1q1b8fnnn8PZ2blCdclTvNelxPXr1zFlyhS0bNkSCxYsQPPmzVG7dm0oKSlh3rx5Ct+DpJyXlxf09PTklmnRooVU2bLqUZS8diWvnorUW5l26ejoiFOnTmHXrl2wt7eHtra21HE1NTWEhITg+vXr+O233/D777/D398fgYGB8PX1Rd++fSt8TSL6cDEJSkREREREVAO0bNkSv/32G9LT06GlpSV17N69e9DU1JRJAv3zzz8y9Uh6D0qSY+W5desWJk+ejLp162LPnj2l9qDr0qULoqOj8fTpU6nFkZ4+fYqkpCT07t1boeu9Lcl9ybt3yb6S996hQwd06NABn332mbAgk6+vLyZMmAAlJaVSk36VdezYMRQUFGDHjh1Ssbx+/Rrp6ekK1yPpZaqtrQ1bW9syy0quk5CQABsbG6ljCQkJCl8TKOqpmpycLNMbNCEhAQ0bNhSG+lekzVamfXfr1g0jR47EtGnTMHHiRISGhspd5MnY2BjGxsYAitrjsGHDsHnzZiZBiWoYzglKRERERERUAzg6OqKwsBDbt2+X2n/27FncunULvXv3hrKy9FfA6OhoPHv2TNjOzc1FaGgoVFRU0KtXr3KveevWLbi5uaFOnTrYu3dvmYnTQYMGAQD27NkjtV+yXZl5JyvD0NAQTZs2RWRkJJKTk4X9eXl52LVrF5SUlNCnTx8ARUPoS84zqaWlhebNm+PNmzfCMHVJUk/RYerlKa1H67Zt22TikVw/NTVVZv+nn34KNTU1BAQEyJ1zNSMjQ5iiwM7ODrVr18b+/fvx5s0bocyzZ88QHR1d4Xso2Q5/+ukn/Pvvv3B0dBT2VaTNVqZ9A0VTGuzYsQOPHz+Gi4uL1Gv+8uVLmfL6+vrQ0dGpsteSiD4c7AlKRERERERUAwwfPhyHDx8WEj5du3bFw4cP8e2330JXVxfz58+XOadNmzYYNWoUxo4di7p16+LYsWP4888/4eHhIdVbU57Hjx/Dzc0NaWlpcHFxQXx8POLj46XK9O3bV0gQOjg4oFevXggJCUFGRgZMTU1x9epVHDp0CEOGDEHXrl2r7mGUQUVFBStXroSnpydGjhyJ0aNHo27dujh58iSuXr2K6dOnCz0oo6KisGfPHjg6OqJVq1aoVasW4uLicP78eXz66afCPKZdunSBsrIytm7dirS0NGhoaKB58+YwMTGpVIyOjo4IDQ3FZ599hjFjxkBVVRUXLlzA3bt3ZXo7AoCpqSkuXbqE7du3o2nTplBSUsLAgQOhr6+PL774AitWrMCAAQMwZMgQNGvWDC9fvsRff/2Fn3/+GcePH0fz5s1Rv359zJkzB15eXhg7diyGDRuGN2/e4ODBg2jdujVu3bqlcPza2tr46aefkJSUBCsrKzx48EBoh56enkK5irTZyrRvia5du2L37t2YOnUqXFxcsGfPHjRu3BhbtmzBhQsX4ODggObNm0MsFuPMmTNISEjA1KlTFb5fIvo4MAlKRERERERUA6iqqmLXrl3YsmULTpw4gZ9++gn16tVD//79MXfuXLlJTWdnZ2RmZmLfvn148uQJmjZtis8//xyTJk0q93qJiYlC78OAgAC5ZU6fPi0kQQHAz88PwcHBiI6OxpEjR9C4cWPMnj1bZiXxd613794IDQ3Fli1bsGvXLuTl5cHAwABr1qzBqFGjhHLW1ta4ffs2fv31VyQnJ0NZWRnNmzfHkiVLpOYDbdq0KdatW4cdO3bgyy+/RF5eHoYPH17pJKiFhQUCAgIQHBwMPz8/qKurw9bWFvv27ZM7D+nq1avx1VdfYevWrcjKygJQtOgPADg5OaF169bYvXs3wsPDkZGRgQYNGqBNmzaYM2eO1JD1yZMnQ0NDAyEhIfD19UWTJk2EuV4///xzhePX0NDAnj17sG7dOvj6+kIsFqN79+5YunSp1NykFWmzlWnfxZmamiIkJARTpkwREqGOjo5ITk7GDz/8gBcvXqB27dpo1aoV1qxZg5EjRyp8v0T0cVASV2Z2YSIiIiIiIvpoxcTEYOLEiVi/fj1GjBhR3eEQERG9c5wTlIiIiIiIiIiIiGo0JkGJiIiIiIiIiIioRmMSlIiIiIiIiIiIiGo0zglKRERERERERERENRp7ghIREREREREREVGNVqu6AyAiIiKij4dYLAYHElFxSkpKAMB2QVLYLkgetguSh+2C5FFSUhLaRlVhEpSIiIiIKiQvL6+6Q6APiKqqKgC2C5LGdkHysF2QPDW1XeTn5yM9Pb26w/ho6evrV3mdTIISERERkcIKxWLcfsr/oSciIiIqTTv9+lCtxZTbh4avCBEREREp7E1uPhaEnqvuMIiIiIg+WL6uPdCpiVZ1h0ElcGEkIiIiIiIiIiIiqtGYBCUiIiIiIiIiIqIajUlQIiIiIiIiIiIiqtGYBCUiIiIiIiIiIqIa7YNIgm7duhUikQguLi4VOi8rKwtbtmzB4MGDYWZmBnNzc/Tr1w8LFy7EnTt3hHJXr15FQEAA0tNr3kqmr169wpw5c9CtWzeIRCIsXbq01LK5ubnYv38/RowYga5du8LMzAx9+vTB7NmzERsb+x6jrpiYmBiIRCLExMRUSX25ubn45JNPEBYWJuyLjIyESCTC7du35Z4zdOhQmfYpEomkfszMzDBw4EBs374deXl5MnW8fPkSGzduRP/+/WFsbIyuXbvC2dkZUVFREIvFZcYcEBAgcz15P0uXLhXuJTExsRJPR77evXuX2bYqqrz7ycrKKvN8eW3CxcVF6jVKTk5GQECA3Ne0ZNl3YdmyZZg5c2aV1feuPyc/Zjdu3MD48eNhbm4OkUiEyMjIUsuKRCKsXbv2PUb3/ij63i/r/Xf8+HHheHUpGV/Hjh1hb2+PGTNm4ObNm+/02uX9LiiNIp/PktemstcgIiIiIqK380GsDn/48GEAQFxcHB4+fIiWLVuWe05BQQGmTJmCf/75B1OmTEGXLl1QUFCAf//9F6dOncKdO3fQsWNHAEVJ0MDAQAwfPhxaWjVrda7g4GCcOXMGGzZsQNOmTaGjo1Nq2UWLFuH06dNwc3PDnDlzUKtWLdy/fx9nzpzB77//Disrq/cYefXZt28fcnNzMWbMmLeua8CAAZg0aRKAomTTxYsXsWnTJiQnJ2P58uVCuXv37sHNzQ05OTmYMmUKjI2N8ebNG/zwww9YunQpzp49C19fXygry/+7xKhRo9C9e3dh+/79+1iyZAlmzJgBBwcHYb+Ojg40NTURHh4OPT29t74/icDAQGhqalZZfRIhISHQ0NCQ2V+nTp0K17V69Wqp7RcvXiAwMBDNmjVDp06dyiz7Lnh4eKB///64fPkyunXr9tb1vevPyY/Z8uXLkZeXB39/f2hqair0bEj++69Vq1awsLCQ+rypLpL4CgoK8PjxY2zfvh2TJk3CkSNH0KxZs+oOT0p4eLjUto+PD+7fv4/AwECp/VX5uUxERERERBVT7UnQuLg43L9/H7169cKZM2cQERGBefPmKXTelStXsHHjRgwdOlTY37NnT7i6uqKwsPBdho3CwkIUFBRAVVX1nV6nPP/88w9atWqFAQMGlFnu0aNH+OGHHzBnzhx4eHgI++3s7DBhwoR3/rzKk5ubCzU1tXd+nby8POzevRtjx46tkuvp6urC1NRU2Lazs0NCQgJOnDghJEELCgowe/ZsvHnzBhEREVIJml69ekEkEsHLywudOnWCu7u73Ovo6+tDX19f2FZXVwcAtGzZUur6EmUlwyujc+fOVVqfhJGRUZX9YaJdu3bvpGxltWjRAt27d8eOHTveOgn6sX5Ovi///PMPxo0bB3t7++oO5aNS1vuv+OeNPO/jd2Dx+CwsLNCuXTsMHz4c586dw7hx497ZdSuj5OewlpYW1NTU5H4+v6339fuSiIiIiKimqfbh8BEREVBSUsLKlSvRvn17REVFKfTFPDU1FQDQqFEjucclPeoCAgKwfv16AECfPn2khqQBwOvXr7FhwwY4ODjAyMgIDg4O8PLyQnZ2tlR9kiGUe/bsQd++fWFkZIT4+HhhWNulS5ewePFiWFpawsLCAsuXL0dOTg4ePXqEadOmwczMDL169cLOnTsVei7lxZWYmAiRSISLFy/ir7/+Eu6rtCHjij6vtLQ0dOrUSWqo+KNHjyASidCrVy+pczw9PTFq1ChhOz8/H8HBwcLzsbe3x8qVK/Hq1Sup83r37g0PDw8cO3YMgwYNgpGREY4dOwagqMfklClTYGJiAmtrayxfvhyZmZky8d68eRPu7u6wsbGBkZERevToAQ8PD6SlpZX1WPHLL78gOTlZKiFU1TQ1NaWGw//000/4559/4O7uLreHmpubGwwMDLBr1y65w+grqrQhsVFRUXBycoKJiQksLCzg4eGB+/fvK1RnyeHwkmvExcVh1apVsLa2hrW1NTw9PfH8+fO3vofiFG0TxYe4x8TEYNiwYQCKhqVL3h+SYdLyhsM/evQIc+fOhbW1NYyMjDBo0CCZ3l2SYfgnTpyAj48P7O3tYW5uDldXVyQkJMjENGTIEFy4cAGPHj16q2fwrj8nJS5fvgxnZ2eYmprC1NQULi4uMlNlSIYq37t3D3PnzoW5uTlsbW2xbNkyZGRkSJVNT0/H559/DisrK5iZmWHq1Km4f/8+RCIRAgICyo2/vNdE0g7z8/MRFhYmvM4V9ezZMyxfvhw9evSAkZERunfvjvnz50vdz7Nnz7Bs2TLY29vDyMgI/fr1Q2hoqEL179+/H+PHj0e3bt1gZmaGoUOH4ttvv5V5DSWfjb/++iuGDRsGY2Nj9O/fH4cOHZKp88qVKxg7diy6dOkCe3t7bNy4sUo+PwDIHQ5f2u9AoOg9Onv2bOF1Gjx4MKKjo6skluIkvdFL3uft27fh7u4OCwsLGBsbw8nJCT/99JPM+YqWK+nx48cYMGAABg8ejGfPnlXNzQDIzMws9/OzrN+XirbJtLQ0rF27Fr169YKRkRF69eqFzZs3V1l7ISIiIiL6WFRrT9CsrCz8+OOP6NatG5o1a4YRI0bAy8sL58+fR48ePco819DQEKqqqvj6668xa9YsWFtbo2HDhjLlRo0ahYyMDOzZsweBgYFCMkBPTw+FhYWYPn064uPjMXPmTBgbG+PatWsIDg7G3bt3sWvXLigpKQl1/fDDD2jcuDEWLFgADQ0NtGzZEo8fPwYArFixAgMGDICfnx+uXLkCf39/iMViXLt2DaNGjcKkSZNw7NgxeHt7o3Xr1nB0dCz13hSJS09PD+Hh4fjyyy/x+vVreHl5ASi9h5uBgQHq1asHf39/KCkpwc7ODk2aNJEpV79+fXTu3BkXL14UkkSXLl1C7dq18eTJEzx48ACtWrVCYWEhYmNjMXbsWOHcFStWICoqCm5ubkKPyM2bN+Pq1av4/vvvUbt2baHs9evXce/ePXh4eEBPTw96enpITk6Gs7Mz1NXVsXr1ajRs2BBHjx7F119/LRVjVlYWJk+ejM6dO+Prr7+GlpYWkpKScOHCBeTk5JT6XAHg7Nmz0NfXR4sWLeQeLygoQH5+fpl1FCcWi4XykuHwp06dwsiRI4UyFy9eBFD0ZVYeJSUl9O7dGzt27MDNmzffSc8hf39/bNmyBWPHjsWcOXOQmZmJ4OBgjBs3DkeOHKn0EM1ly5ahV69e8PX1xdOnT+Ht7Y1FixZh7969Cp0v73krKSlBRUUFABRuEyUZGhrCy8tLZsqA0oZJJycnY9y4cVBSUsKiRYvQqFEjnDx5EqtWrUJKSopU72mgaKhr165dsXbtWmRkZMDHxwczZszAiRMnhNgBwNraGmKxGL/99hvGjx8v7Je0hV9++aXcZ/Q+PieBove5JNks+TwJCQmBm5sbdu3aJdOb1dPTEwMGDMDo0aNx9+5dbNq0CQCEPzoVFhZi2rRpuHXrFjw9PdG5c2dcuXKl1N7OJSnymjg4OCA8PBxjxoyRmpqiIp4+fSq8X6dPn4527drh5cuX+PXXX/H69WvUq1cPz58/x8iRI1G3bl3MmzcPTZo0wYULF7Bx40akpqZi7ty5ZV7j0aNHGDp0KJo1awYVFRVcv34dGzduRFJSksy5t27dgre3N9zd3dGwYUN89913WL58OVq1agVLS0sAwF9//QVXV1e0bNkSXl5eUFdXx/79+3HixIkK3XvJ91/x95488n4H3rlzB+PGjUPbtm2xcuVKNGjQAD/88AMWLlyI7OxsqT+UiUQiWFlZSf2RTZH4CgsL8fjxY2zcuBEaGhro06ePUObvv//G+PHj0aRJE3z55ZfQ0NDAd999B09PT3h5eQl/DFG0XEk3b97EtGnT0K5duyqfFkTRz095vy8VbZNZWVmYMGECUlNTMWPGDLRt2xbXr19HUFAQHj9+DG9v7yq7HyIiIiKiD121JkFPnDiB169fY8SIEQCKFp/x9fVFREREuV/uW7RogS+//BJr164VhoW2aNECdnZ2GDdunDDPnb6+Ppo2bQoA6NSpE5o3by7UcfbsWcTExGDlypVwdnYGUDScuW7duli3bh3Onz8vNS9abm4uQkND5X4J6tu3LxYsWAAAsLW1xeXLlxERESE1DNXKygpnzpzBsWPHykyC/vbbbwrFZWpqCk1NTRQWFpabONPQ0ICvry+WLVsmDNPW09ODnZ0dnJychC/XAGBjY4MDBw6goKAAKioquHTpEvr27YuYmBhcunQJrVq1wo0bN5CWlgYbGxsARcNRDx8+jMmTJ2PJkiUAAHt7ezRt2hQzZ85EVFSUVMI0NTUVBw8elHo9vL298erVKxw9ehQdOnQA8L9hu0+fPhXKJSQkIDU1FUuWLJGaz3DQoEFlPgOgaH7YkvNDFufk5FTqMXlzpoaFhcl8of/000+lek0+efIEAEpNvAIQ5rd78uRJlSdBnzx5gm3btsHNzQ2LFy8W9ltYWKBfv34ICQkRXrOKcnBwkJr7NC0tDd7e3khKSlIosSpvmLihoaHQYzM0NFShNlGSpqam0JOttCkDigsNDcWLFy8QFRUltKmePXsiIyMD27Ztg7Ozs9SwYZFIhI0bNwrbKioqmDt3Lq5fvw4zMzNhf8OGDdG4cWPEx8dLJUHLSjSV9D4+JwFg06ZN0NXVRWhoqDDdQs+ePeHo6IhvvvlGplfsmDFj4OrqCqDoM+/hw4eIiIjAunXroKSkhHPnziE+Ph5r1qwREmF2dnZQU1ODr69vufetyGuio6MjTP1QcmoKRfn7+yMtLQ3Hjh1D69athf0DBw4U/h0YGIjs7GxERkYK7drW1hZ5eXnYtWsXXF1d0aBBg1KvUfzzoLCwEJaWligoKEBoaCjmzJkj9ce21NRUhIeHo3HjxgAAS0tLxMTEIDo6WvicDg4OhrKyMvbs2SPcv4ODQ7nTopRU8v3Xvn17oZehPPJ+By5fvhza2toICwsT5he1t7fHq1evsHnzZjg5OQk9jlVUVEqd91iR+LS1tREcHCw1H2hQUBDEYjH27t0LXV1dAEXTjDg5OcHX1xdDhgyBsrKywuWKO3v2LObOnYu+ffti7dq1VT70X9HPT3m/L1euXKlQmwwLC0NCQgIiIyOF95GNjQ1q166NdevWwd3dHe3bt6/S+yIiIiIi+lBV63D4iIgIaGpq4pNPPgFQlDDo2bMnTp8+LTOEWh4nJyf8+uuv8PPzg4uLC7S0tBAeHg4nJ6cyv8hJSIaODxkyRGq/pEdIyaHl3bp1K7UXSM+ePaW2DQwMZPbXqlULrVq1EpJiVRWXoiTPdtu2bZg8eTKaN2+Oo0ePwtnZWWqYvo2NDTIzM3H9+nWIxWJcvnwZtra2sLGxEXo1Xrx4EWpqajA3Ny8zZkdHR9SrV08m5o4dO0p9oQOA2NhYdOzYUUh2SQwePFhqu3Xr1qhfvz6WL1+Ow4cPV2iocVJSUpnzZfr4+ODQoUMyP8WTI8UNGjRIKBMWFoaFCxfi7NmzQkJcUZLV4YsnQ6rKhQsXkJ+fjyFDhiA/P1/4adiwITp37oy4uDgARcmZ4scLCgrKrbt4jywAQuKxvDYusXfvXplnXbxnkqJt4m1JhrmXXCRo2LBhyM7OxtWrV6X2l+zVW9Z9N2zYUGaI608//aTQMFzg/XxOvn79Gn/++Sf69esnJEABoHbt2ujfvz+uX7+ON2/eSNUr7xnk5OTgxYsXACAMo//000+lyhVPLpaloq9JZf3222+wtbUt9T0OFCXDbGxsoKOjI/Ue6dmzJ3Jzc3Ht2rUyr3Hz5k14enrC3t4ehoaGMDQ0hJ+fH9LS0pCSkiJVtnPnzkICFCia/7d169bCqAOg6Nna2tpKfZapqKgo/GwlSr7/vvnmmzLLl/wdmJOTg5iYGHzyySdQU1OTeTYvXryQmibi1q1b2LNnT4Xj+/777xEUFASRSISZM2cKw/CBonZia2srJDaBos/RIUOGICkpCf/++2+FykkcOnQIHh4emDhxIjZu3PhO5j5V9PNT3u9LRdvk2bNn0bFjR7Rr106qnOQPKCWnuyAiIiIiqsmqrSdoQkICrly5giFDhiA3Nxe5ubkAgH79+uH06dOIjo7GxIkTy61HS0sL/fv3R//+/QEA8fHxmDJlCtauXVtuz8C0tDSoqanJLAxRv359qKmpCfPpSZQ2rx4AmV5Aki9M8vanp6dXaVwVUadOHTg4OAjDg+/du4dJkyZh8+bNGD16NLS0tGBhYQE1NTVcvHgRderUwcuXL2FrawslJSVs2LABYrEYFy9ehLm5uZAwkczFWfwLpoSurq5CzzI1NRWtWrWS2V+ybL169RAWFobg4GCsW7cO6enpaN68OSZMmAA3N7cyE4nZ2dlSSZ6S2rVrJ7enaPGh/MXp6OigS5cuwrakt6iPjw/Onz8v9IYFiobESpLjJUm+9MqbouBtSZJSpc2DKvlyHRQUJLWScbNmzcodrl2yfUsW6yhvWgKJTp06lbkwkqJt4m2lpaXJHSovuU7J9qutrS21XdZ9q6urK/w8Snpfn5Pp6ekQi8Vyn6tk6pD09HTUqVNH2F/ea5+WlgZ1dXWZPxzJ+4yQp6KvSWW9evWq3EWAUlJScOrUKRgaGpZaR2kSExMxYcIEGBgYYMmSJWjWrBlUVVXx888/Y+vWrTLzT8vrUaqmpia89kDRvct7jhV9X5T3/iuv/tTUVOTn5yMkJAQhISFyz1EkUa9ofHZ2dnBwcMDGjRtx8OBBAEXtpKxnIWknipaTOH78ODQ1NYUe2O+Cop+f8l5XRdtkSkoKHjx4UKm2S0RERERU01RbElSy0MPRo0dx9OhRmeMREREKfbkvydzcHHZ2dvjpp5+QkpJS6vx3QNEXkNzcXKSnp0t90UpLS0Nubq7MF5R30UuvKuJ6GwYGBhg4cCBCQ0Nx//59GBsbo3bt2jA1NcWlS5egoaGB1q1bQ19fHzY2NkhNTcWVK1dw5coVzJw5UypmoCjhVvIL24sXL2QSi/KeZYMGDYSEXXHJycky+0QiEfz8/CAWi3H37l0cOHAAXl5e0NLSkpqPsyRtbe1yF096W5Jei7dv34a9vT1sbGwQHh6O06dPy02CisVi/PLLL2jQoME7WYVdkrALCgqS6mEmIfniPXr0aCE5Xnx/dapIm3iX13mb91xpiVxFvK/PSS0tLSgpKcl9rklJSVBWVq5QsgwoemY5OTnIzMyUSoTKe86lnf+uXpPitLW1y13sRltbG507d8asWbPkHi/ZS6+406dP482bNwgICBD+ICLZX1nv631RUsnPbS0tLSgrK2PEiBFS050U16ZNmyq7fp06ddCiRQvcvn1b2KdoO6loe9q0aRN8fHwwYcIE7Nmzp9Q/YL0P8n5fKtomtbW1oaGhUeo8ypWdD5qIiIiI6GNULUnQ/Px8HD16FG3btsUXX3whczwqKgqRkZG4efNmmb0XNDU1ZYaoFRQU4MGDB6hduzbq1asHoPTeFd26dcPOnTuFIeESkmSDvPkK34d3EVdmZiaUlZWFOduKkwxXLJ68tLGxQXBwMJSUlGBrawugaH7VNm3aIDAwELm5ucJ8oEDRAjCSGIsnPH/55RdkZGQoFLO1tTV27tyJv//+W2qOsrJWGVZSUkLHjh2xatUqHDp0CHfu3CnzGgYGBnj48GG5sbwNSQySoap9+/ZF27ZtsWPHDvTv31+md1tISAju3buH+fPnv5PEo52dHVRUVJCYmFjmXLSNGzeWmyStTpVpExKSZ1myp5083bp1w44dO3Dnzh2p4ddHjx4V/ihQGfn5+Xj27JnMsFdFz31fn5NqamowMTHBjz/+iIULFwq9pXNycnDq1CmYmJhI9QJVhJWVFXbu3ImTJ09KLY5z/Phxhc5/V69JSd27d0d0dLSw6FtpZS5duoQ2bdpUeGEcSQKr+GuQk5MjN6mtKGtra/z66694+fKl8DlTUFCg8LOtKnXq1IGlpSVu376NTp06oVatd/u/FK9fv8ajR4+kpgHo1q0bfvnlF7x48ULo6SkWixEdHY3GjRujbdu2FSonoa2tjT179uCzzz6Ds7MzQkJCZKZmqE6Ktsnu3btj165d0NXVfScjDYiIiIiIPibVkgQ9e/YskpOTMXXqVCF5Vlzjxo0RGRmJiIiIUr/cx8TEYP369Rg8eDDMzMxQv359JCUl4dChQ/jrr78wY8YMIQki6Zm3b98+DBkyBLVq1YJIJIK9vT1sbW3h5eWFjIwMGBsbC6um2tvbw87O7t09hDK8i7j+/fdfTJ06FYMHD0bXrl2hq6uLV69e4cSJEzh37hyGDBki9QXJxsYGfn5+iI2NleppZmNjg2+//Rb16tWDkZGRsN/AwAAjR44UhkTa29sjISEBfn5+6NixY6kr7xY3adIkRERE4LPPPsOcOXOgo6ODo0eP4v79+1Llzpw5gwMHDsDR0RHNmzdHQUEBoqOjUVBQILWQlTxWVlYIDg5GTk5OmcPiFfXixQthbsI3b97g2rVr2Lp1K5o2bYp+/foBKJoLNiAgAK6urhg9ejSmTJkCY2NjvHnzBj/88AOioqLQr18/TJ069a3jkadFixbw8PCAj48PHj16BFtbW2hqaiI5ORnx8fFo27atVLL9fbpx44bcxLxIJEKdOnUUbhPyNG/eHHXq1MGxY8fQvn171K5dG82bN5cZyg4Arq6uOHz4MNzd3TF79mzo6enh5MmTOHXqFObPn1/hXpASd+/exZs3b2Q+5xRZHf59f07OmzcPU6ZMgaurKyZPngyxWIyQkBC8fPlSoYWMSurevTvMzc2xZs0apKWlCavDHzlyBED5Pevf1WtS0uzZs3Hu3DmMHz8e06dPR/v27ZGamopff/0V8+bNQ+PGjTF37lxcuHAB48aNg4uLC1q1aoXXr1/jwYMHOHPmDHbv3l3qYld2dnZQVVXFggULMHXqVGRlZSEkJOStEoYzZszAL7/8gkmTJmHGjBmoXbs29u3bV+lpF97G559/jgkTJsDFxQVjxoxB06ZNkZ6ejoSEBPz5558ICAgQylZ0dXjJ54NYLEZSUhL27dsnrHIuMXPmTJw5cwaTJk2Ch4cHNDQ0EB4ejhs3bsDHx0doZ4qWK65evXrYtWsXZsyYgYkTJ2Lnzp0wNjZ+yydWNRRtk66urvjxxx8xYcIETJo0CR06dEBubi4eP36Mc+fOYeXKlVILTRERERER1WTVkgSNjIyEmppaqYmx1q1bw8rKCseOHcOSJUvkJqtMTU0xbNgwxMTE4PDhw0hNTUWdOnUgEong5eUlVXfXrl3h7u6Ow4cP4+DBgygsLMTp06fRvHlzBAcHw9/fH+Hh4QgMDESjRo0wceJEzJ49+70Nfy9JSUmpyuNq1aoVJk2ahMuXL+PHH3/Eq1evoKqqCgMDA+FLbHFdunSBpqYmXr9+LZWAkSRBLS0tZb70f/XVV2jRogUiIiIQFhaG+vXrY8CAAZg/f75CCcdGjRph3759WLt2LVavXo06derA0dERK1euhIeHh9S9aGpqYvv27UhKSoK6ujoMDAzg5+cns0BVSYMHD4a/vz/OnTuHvn37KvLoynTixAmcOHECQFHPw6ZNm2LkyJFwd3eX6p3Trl07HDlyBDt37sShQ4fg7+8PVVVVdOzYEevWrcOwYcMqtGpyRXl6esLAwAD79u1DZGQkCgoKoKenB1NT02r9Uu/m5iZ3f1RUFDp16qRwm5BHXV0da9euhb+/P1xdXZGfn4/169fLneOvYcOGOHjwIHx9feHt7Y2srCy0adNGamXzyvj555+hq6sr1WtaUe/7c7Jbt27YvXs3AgICsGjRIgBFnwOhoaHo2rVrheNXVlbG1q1bsWHDBmzbtg15eXkwNzeHt7c3Ro8eLfTUL827ek1KatKkCb7//nv4+/tj69atSEtLg46ODiwtLVG3bl0ARQnniIgIBAUFYcuWLUhOToampiZat26N7t27l/neNTAwwObNm+Hn5wdPT0/o6upi5MiRaNSoEVasWFGpmDt06ICQkBB4eXlhyZIlqF+/PoYMGYL+/ftj5cqVlaqzsjp27IjIyEgEBQXBx8cHqampqF+/Ptq2bSuzKFZFFf980NHRQdu2bREQECAsEgYUPd8DBw5g06ZNWLlyJfLy8iASiRAUFCTV813RciXVrVsXO3bswKxZs+Dm5oZt27ZV6v1Q1RRtk5qamjhw4AC2bt2K/fv348mTJ9DQ0EDz5s1hb29fpdPrEBERERF96JTEkmWpif4jpk2bBmVlZWzZsqW6Q6EarLCwEI6Ojhg6dCjmzJlT3eF8MKKjo7Fw4UJ8++23sLCwqO5wiKgSMrNzMXRD+dOSEBEREf1X+br2QKcmWnj58mV1h/LR0tfXr/LOidW2MBJRdZk/fz5GjBiBu3fvQiQSVXc4VEMdPXoUb968weTJk6s7lGpz9OhRvHz5UpjP9erVq9i1axe6du3KBCgRERERERG9V0yC0n+OSCTC2rVrkZSUxCQovTNisRje3t7lDvuuyTQ0NBAaGooHDx4gOzsbenp6GDFiBObOnVvdoREREREREdF/DIfDExEREZHCCgoL8efDlOoOg4iIiOiD1U6/PlSVxBwO/xY4HJ6IiIiIqpWykhI6NdGq7jDoA6KqqgoAyMvLq+ZI6EPCdkHysF2QPDWzXYiRn59f3UFQCUyCEhEREVGFsFcDFaejowOA7YKksV2QPGwXJA/bBb0vytUdABEREREREREREdG7xCQoERERERERERER1WhMghIREREREREREVGNxtXhiYiIiEhhYrEYubm51R0GfUBq5oIW9LbYLkgetguSR0VFBQUFBZwTlKRwdXgiIiIiqlaFYuBxek51h0EfFLYHkoftguRhuyBpTeprQFVZGQUFBdUdCv0HMAlKRERERArLzS/AjrN3qjsMIiIiqgE+69kRrRtqVncY9B/BOUGJiIiIiIiIiIioRmMSlIiIiIiIiIiIiGo0JkGJiIiIiIiIiIioRmMSlIiIqAwikUihn8TERERGRkIkEuH27dvvNCbJdYyNjfHs2TOZ40OHDoWLi8s7jaG6RUdHY8CAAejSpYvw/OW5d+8eAgIC5B7v3bs3PDw83mmcgwYNwrBhw2T2S17D6dOnyxwLCAiASCRCQkICgKI2GBAQUKVxvYs6iYiIiIg+ZFwYiYiIqAzh4eFS2z4+Prh//z4CAwOl9uvp6b3PsAAAOTk58Pf3x7p16977tatTSkoKli1bhl69euHLL7+Eqqpqqc//33//RWBgIKysrNC8efP3HClgZWWFAwcOIC0tDfXr1xf2x8bGQkNDA7///jsKCwuhrKwsdaxRo0Zo27YtgKI2qK+v/95jJyIiIiKqSdgTlIiIqAympqZSP1paWlBTU5PZr6am9t5j6969O6KionDv3r33fu3qdP/+feTl5WHw4MGwtLSstuevCCsrKxQWFiIuLk5qf2xsLMaOHYuMjAzcunVL2J+bm4tr167B2tpa2GdqasokKBERERHRW2ISlIiIqIplZmZi1apVsLa2hrW1NTw9PfH8+XOZclFRUXBycoKJiQksLCzg4eGB+/fvK3ydzz77DPXq1cOmTZvKLRsQEAAnJydYWlrCwsICI0eOxIkTJ2TKiUQirF27Fvv370ffvn1hbGyMUaNG4datW8jPz4efnx969OgBCwsLzJw5Ey9fvqzS+7p8+TKcnZ2F5LKLiwtiY2OF40uXLsX48eMBALNmzYJIJCp16H9kZCRmzpwJAJg4caIwdUFMTIxUuV9//RXDhg2DsbEx+vfvj0OHDsnU9ezZMyxbtgz29vYwMjJCv379EBoaWu79WFlZQUlJSeoeHj9+jMePH2PQoEFo0aKF1LGrV68iJydHKglacui6ZLj8vXv3MHfuXJibm8PW1hbLli1DRkaG1PUzMjKwfPlyWFtbw8zMDJMnTxaG2RMRERER/ZdwODwREVEVkwzV9vX1xdOnT+Ht7Y1FixZh7969Qhl/f39s2bIFY8eOxZw5c5CZmYng4GCMGzcOR44cUWh4vZaWFtzd3bFx40ZcuXIFZmZmpZZ98uQJJkyYgCZNmiA/Px8xMTFYuHAhsrKyMGrUKKmyP//8M5o3b44lS5YgPz8fXl5emDFjBuzs7JCXl4c1a9bg2bNnWL9+PVauXImgoKAqua9Lly5hypQpMDExgZeXFwAgJCQEbm5u2LVrF7p16wYPDw906dIFX331FRYuXAhLS0toamrKrc/BwQELFy6Ej48PVq1aBUNDQwBAu3bthDK3bt2Ct7c33N3d0bBhQ3z33XdYvnw5WrVqBUtLSwDA8+fPMXLkSNStWxfz5s1DkyZNcOHCBWzcuBGpqamYO3duqfeko6OD9u3bSyU6Y2JiUK9ePXTq1AmWlpaIjY3F5MmTAUAoVzwJWhpPT08MGDAAo0ePxt27d4Vk+Pr16wEAhYWFmDFjBq5du4ZZs2bB0NAQf/zxBz777LNy6yYiIiIiqmmYBCUiIqpiDg4OWL58ubCdlpYGb29vJCUlQU9PD0+ePMG2bdvg5uaGxYsXC+UsLCzQr18/hISEYMmSJQpdy9nZGWFhYfD19cW+fftKLSdJjAFFyTEbGxu8evUK+/fvl0mCFhQUYMeOHahduzYA4M2bN1i6dCkeP36MPXv2COX++ecf7N27F2/evEGdOnXe+r42bdoEXV1dhIaGQl1dHQDQs2dPODo64ptvvkF4eDhatmwpJDHbtGkDU1PTUuvT0dFBmzZtABQlPuWVTU1NRXh4OBo3bgwAsLS0RExMDKKjo4UkaGBgILKzsxEZGSkkcW1tbZGXl4ddu3bB1dUVDRo0KDUOKysrfPvtt8K8oLGxsbCwsICysjK6du2K9evXC/OCxsbGQl9fH61atSq1PokxY8bA1dVViOfhw4eIiIjAunXroKSkhN9++w1xcXFYtWoVJkyYAACws7ODiooK/P39y62fiIiIiKgm4XB4IiKiKtanTx+pbZFIBKCoNyYAXLhwAfn5+RgyZAjy8/OFn4YNG6Jz584y80eWRV1dHbNmzUJcXBx+/fXXUstdvnwZU6ZMga2tLTp37gxDQ0McOnRI7tDobt26CQlQADAwMABQlNwtzsDAAGKxuEru6/Xr1/jzzz/Rr18/IQEKALVr10b//v1x/fp1vHnzRpFHUiGdO3cWEqBA0fNs3bo1Hj9+LOw7e/YsbGxsoKOjI3VfPXv2FObwLIu1tbXUvKCxsbFCgtXKykqYF1RSl5WVlUKx9+7dW2pbJBIhJycHL168EK4DAIMHD5YqN2TIEIXqJyIiIiKqSdgTlIiIqIqV7BUoWbQnJycHAIQk1dChQ+WeX9FVzIcPH46QkBD4+vqiR48eMsevXr2KyZMno1u3bli1ahUaN26MWrVq4cCBA4iIiJApX3wVcwBQVVUtc39ubu5b31d6ejrEYjEaNWokc0xPTw+FhYVIT09HnTp1Sq2jMuT14FRTUxPuCShajf7UqVPCcPqSXr16VeY1LC0thXlBO3XqhMePHwuJzhYtWkBfXx+xsbHIzs5Gdna2QkPh5cVesp2lpqZCXV0dWlpaUuUUmWqBiIiIiKimYRKUiIjoPdPW1gYABAUFSfVClKjoSufKysqYN28ePDw8cPToUZnjJ06cQK1atbB161apuovPUVoV3ua+tLS0oKSkhOTkZJljSUlJUFZWlknmvS/a2tro3LkzZs2aJfd4eUlrbW1tYV7Qjh07QkNDA507dxaOd+3aVUiCAorNB6qIBg0aICcnB+np6VLPLikpqUrqJyIiIiL6mDAJSkRE9J5J5mVMTEyEo6NjldTZp08fmJubw9/fX2ooOwAoKSlBRUUFSkpKwr6UlBScPn26Sq4t8Tb3paGhARMTE/z4449YuHChMCQ+JycHp06dgomJSaV6gUp6q0oSjJXRvXt3XLp0CW3atCl1EabyWFtbY//+/fjpp59gbm6OWrX+979gXbt2ha+vL16/fo2mTZuiRYsWlY615DV37tyJ6OhoYU5QAHIT5URERERENR2ToERERO9ZixYt4OHhAR8fHzx69Ai2trbQ1NREcnIy4uPj0bZtWzg7O1e43oULF2L8+PEAgIYNGwr7HRwcEBoaioULF2LMmDFITk5GcHAwdHV1kZWV9cHc17x58zBlyhS4urpi8uTJEIvFCAkJwcuXL+Hr61upmNq1awclJSUcOnQImpqaUFNTq3Ayc+7cubhw4QLGjRsHFxcXtGrVCq9fv8aDBw9w5swZ7N69GyoqKmXWYW1tjbCwMJw5cwbz5s2TOmZpaYmMjAzExsZi2LBhlblNuezt7WFpaQkvLy9kZWUJq8MfOXKkyq5BRERERPSxYBKUiIioGnh6esLAwAD79u1DZGQkCgoKoKenB1NTUxgbG1eqTgsLC/Tu3Ru//PKL1H4bGxt8/fXX2LlzJ6ZNm4amTZti0qRJSElJQWBgYFXcjuBt7qtbt27YvXs3AgICsGjRIgBAly5dEBoaiq5du1YqnmbNmmHp0qXYu3cvXFxcUFBQgL1791ZoyHnjxo0RERGBoKAgbNmyBcnJydDU1ETr1q3RvXt3KCuXv86kZF5QsVgscy/t2rWDjo4OXr58qfCiSIpQVlbGli1bsH79euzYsQN5eXkwNzfHjh078Omnn1bZdYiIiIiIPgZKYrFYXN1BEBEREdHH4U1uPr46Gl/dYRAREVEN8FnPjmjdUBN5eXl4+fJldYdDHxB9fX2p6byqQvldF4iIiIiIiIiIiIg+YkyCEhERERERERERUY3GOUGJiIiISGFqtVTwWc+O1R0GERER1QBN6mtUdwj0H8IkKBEREREpTFkJaKalXt1h0AdEVVUVAJCXl1fNkdCHhO2C5GG7IBniAhQWVncQ9F/BJCgRERERVQgXLqDidHR0ALBdkDS2C5KH7YLkkbQLoneNc4ISERERERERERFRjcYkKBEREREREREREdVoTIISERERERERERFRjcY5QYmIiIioQjh3FxUnWeiE7YKKY7sgedgu/pvy8/ORnp5e3WEQMQlKRERERIoTA0jNKajuMOhDwvZA8rBdkDxsF/852hrqTDzRB4NtkYiIiIgUll9QiF9uJVZ3GERERPQR6N25ORqoq1R3GEQAOCcoERERERERERER1XBMghIREREREREREVGNxiQoERERERERERER1WhMghIREZVCJBIp9JOYmIjIyEiIRCLcvn37ncYkuY7kp3PnzujRoweWLVuGpKSkStV57tw5BAQEVHGk79+NGzcwfvx4mJubQyQSITIyUm65nJwcBAQEICYmRubY0qVL0bVr13cdaqnXiYuLg4WFBYYOHYqUlBQARe2wJrw+RERERETViQsjERERlSI8PFxq28fHB/fv30dgYKDUfj09vfcZFgDAy8sLrVu3xps3b3Dx4kXs2rULV69exdGjR6Gqqlqhun777Tfs3bsXs2bNekfRvh/Lly9HXl4e/P39oampiZYtW8otl5OTg8DAQHh6esLa2vo9R1m6s2fPYvbs2ejUqRO2b98OLS0tAEXtUF9fv5qjIyIiIiL6uDEJSkREVApTU1OpbS0tLaipqcnsrw4ikQidOnUCANjY2CAlJQURERH4/fffYWNjU83RFcnNzYWamtp7u94///yDcePGwd7e/r1ds6qcOHECixcvhrW1NQIDA1GnTh3h2IfQ3oiIiIiIPnYcDk9ERFSFMjMzsWrVKlhbW8Pa2hqenp54/vy5TLmoqCg4OTnBxMQEFhYW8PDwwP379yt9XUNDQwDAy5cvpfafPXsWzs7OMDMzg6mpKSZNmoTr168Lx5cuXYq9e/cCkB7+DwAxMTEQiUQyw8YTExNlhptLhnffvHkTLi4uMDU1xapVq4R6165di8jISPTv3x8mJiYYMmQIzpw5o9C9PXr0CHPnzoW1tTWMjIwwaNAgqV66kikC8vPzERYWJnUPJSUmJsLS0hIAEBgYKJQtOdw8ISEBU6ZMgZmZGXr27IkNGzYgNzdXqkxOTg78/PzwySefwMjICPb29vjqq6+QmZmp0H1JfP/991iwYAF69eqFLVu2SCVAAdnh8AEBARCJRLh37x7mzp0Lc3Nz2NraYtmyZcjIyJA6Nz09HZ9//jmsrKxgZmaGqVOn4v79+xxiT0RERET/OewJSkREVIWWLVuGXr16wdfXF0+fPoW3tzcWLVokJBoBwN/fH1u2bMHYsWMxZ84cZGZmIjg4GOPGjcORI0cqNbz+8ePHAIDWrVsL+w4dOoQVK1Zg4MCBmDx5MgoLCxESEgIXFxd89913EIlE8PDwQE5ODk6cOCEz/L+icnJyMGvWLLi4uMDDw0NqWP7p06dx7do1zJkzBxoaGti5cyc8PT3xww8/oEWLFqXWmZycjHHjxkFJSQmLFi1Co0aNcPLkSaxatQopKSnw8PCAg4MDwsPDMWbMGAwYMACTJk0qtT49PT2EhITAzc0NI0eOxKhRowBAarh5bm4uPDw8MGbMGEydOhWxsbHYsmULNDU14enpCQAoKCiAu7s7bt68CXd3d3Tp0gUJCQnw8/PDX3/9hb1790JZufy/Ne/evRteXl4YMWIE1qxZAxUVlXLPkfD09MSAAQMwevRo3L17F5s2bQIArF+/HgBQWFiIadOm4datW/D09ETnzp1x5coVuLu7K3wNIiIiIqKagklQIiKiKuTg4IDly5cL22lpafD29kZSUhL09PTw5MkTbNu2DW5ubli8eLFQzsLCAv369UNISAiWLFlS7nUKCgqQn5+P7OxsXL58GQcPHsSgQYOEHqGvX7+Gl5cX+vXrB19fX+E8Ozs79OvXD0FBQfD390fLli2hq6sL4O2HXefm5mLu3LkYMmSIzLG8vDyEhoZCQ0MDQFHP1e7du+PEiROYNm1aqXWGhobixYsXiIqKQseOHQEAPXv2REZGBrZt2wZnZ2fo6OhAR0cHAKCrq1vmfaipqcHIyAhAUeJTXtmcnBwsWLAAffv2BVA03cDNmzcRHR0tJEFPnDiBy5cvY/v27ejZs6dQrnHjxpg5cybOnTsHBweHMp9XRkYGvLy8YGtri3Xr1kFJSanM8iWNGTMGrq6uAABbW1s8fPgQERERQl3nzp1DfHw81qxZIyR77ezsoKamJtUmiIiIiIj+CzgcnoiIqAr16dNHalsyLPvJkycAgAsXLiA/Px9DhgxBfn6+8NOwYUN07twZcXFxCl3HyckJhoaGsLCwwMyZM9GpUyds2LBBOH716lWkp6fLXEdVVRVWVlYKX6eiJInDkrp16yYkQIGiZGXDhg2FHqylkQzJlyRAJYYNG4bs7GxcvXr1rWMuSUVFRUhsSohEIuE1BIqmGWjQoAHs7Oyknq+dnR1UVFQQGxtb7nU0NDRgbW2Ny5cvl7qSfVl69+4tE2NOTg5evHgBAEIMn376qVS5gQMHVvhaREREREQfO/YEJSIiqkINGjSQ2pYsDJSTkwMAQoJq6NChcs9v3ry5Qtfx8fFB69atkZmZicOHD+PIkSNYt24dVq9eLXUdDw8PuedXtNehIjQ1NWXms5Qo+VyAomdTcp7NktLS0uSu8t6oUSMAQGpqaoXjLI+GhobMgk4lY01JSUFqaqrQ87akV69elXsdFRUVbNu2DdOmTcOKFSugpKSEESNGKBxneW0tLS0N6urq0NTUlCon6flLRERERPRfwiQoERHRe6StrQ0ACAoKQuPGjWWOK7qaert27aRWh8/IyMCBAwcwfPhwGBsbC9dZvXo1unTpUqlY1dXVAUAmUVlagu9dJFYbNGggJHSLS05OFo5XB21tbejq6mLr1q2lHldEnTp1hETo8uXLoaSkhOHDh1dJjA0aNEBOTg4yMzOlEqHynicRERERUU3H4fBERETvkWS4dGJiIrp06SLzU9qq5uX5/PPPUatWLfj5+QEAzM3NoampiYSEBLnXKZ4YlSRes7Ozpeps1qwZAODu3btS+3/55ZdKxVgZ3bp1w927d3Hnzh2p/UePHkXt2rUrNY9pafdbET169EBKSgqUlZXlPltFe/QC/0uEWlpa4vPPP8fhw4crHVdxVlZWAICTJ09K7T9+/HiV1E9ERERE9DFhT1AiIqL3qEWLFvDw8ICPjw8ePXoEW1tbaGpqIjk5GfHx8Wjbti2cnZ0rVe+YMWOwb98+XLlyBWZmZvj888+xYsUKpKWloW/fvkKvyhs3bkBNTQ1z584FALRv3x4AsHPnTnTv3l1I7DVq1Ag2NjbYvn076tevj6ZNm+LSpUv46aefqvKRlMnV1RWHDx+Gu7s7Zs+eDT09PZw8eRKnTp3C/PnzoaWlVeE6a9eujRYtWuDMmTOws7NDvXr1oKenJ7dnbmkGDRqEqKgoTJ06Fa6urjA0NISSkhKePn2K8+fPY+LEiTA3N1e4vuI9Qj///HMoKSlh2LBhFb634rp37w5zc3OsWbMGaWlpwurwR44cAfBueu4SEREREX2omAQlIiJ6zzw9PWFgYIB9+/YhMjISBQUF0NPTg6mpKYyNjStdr4eHBw4fPgx/f3+EhITAyckJTZs2xc6dO7F8+XJkZ2dDV1cXRkZGGDt2rHDegAEDEBcXh7CwMAQGBkIsFgu9P729vfH111/Dx8cHBQUF6NWrF3x9feHk5PTWz0ERDRs2xMGDB+Hr6wtvb29kZWWhTZs2UiueV8bXX3+N9evXw93dHXl5efD09MSsWbMUPr9WrVrYvn07QkNDER0djcDAQKipqaFp06bo1q1bhXqCShRPhC5btgwA3ioRqqysjK1bt2LDhg3Ytm0b8vLyYG5uDm9vb4wePRr16tWrdN1ERERERB8bJbFYLK7uIIiIiIjo/YiOjsbChQvx7bffwsLCosLn5+YXIOL3e+8gMiIiIqppendujgbqKnj58mWpZXR0dACgzDL036Ovr1/lI5fYE5SIiIiohjp69ChevnwpTHlw9epV7Nq1C127dq1UApSIiIiI6GPFJCgRERFRDaWhoYHQ0FA8ePAA2dnZ0NPTw4gRI4T5YImIiIiI/iuYBCUiIiKqoRwdHeHo6FjdYRARERERVTsmQYmIiIhIYbVUlNG7c8UXfiIiIqL/Hm0NdYgL8qs7DCIATIISERERUQUoAWigrlLdYdAHRFVVFQCQl5dXzZHQh4TtguRhu/jvERfkIz+fSVD6MDAJSkREREQVwtVbqTiu6kvysF2QPGwXRFSdlKs7ACIiIiIiIiIiIqJ3iUlQIiIiIiIiIiIiqtE4HJ6IiIiIKkQynJEI+N8cf2wXVBzbBcnDdvH28vPzkZ6eXt1hEH2UmAQlIiIiogrJLVSq7hDoA5KbI1nwgu2C/oftguRhu3g7ddSYwiF6G3wHEREREZHCCgrFuPs4rbrDICIi+s8RNasPNU5qSFRpfPsQERERERERERFRjcYkKBEREREREREREdVoTIISERERERERERFRjcYkKBER0TsiEokU+klMTERkZCREIhFu3779zuL54YcfIBKJ8PPPP8sc6927N0QiEW7cuCG1Pzc3F126dMH8+fMBAAEBARCJRO8sRnkksU2dOlXu8W3btkk9y3dJJBIhMjLyreuRPEd5P8ePH6+CSImIiIiIqDgujERERPSOhIeHS237+Pjg/v37CAwMlNqvp6f3XuKxsrKCkpISYmNj4ejoKOx//PgxHj9+DA0NDcTGxsLIyEg4duXKFeTm5sLa2hoAMGrUKHTv3v29xFtc3bp1ceHCBTx//hyNGzeWOnb48GHUrVsXWVlZ7z2utxUSEgINDQ2pfa1ataqmaIiIiIiIai4mQYmIiN4RU1NTqW0tLS2oqanJ7H9fdHR00K5dO8TGxkrtj4mJgZaWFvr27YvY2FhMnjxZOBYXFwcAQhJUX18f+vr67y/o/2dpaYkbN24gKioK06ZNE/bHx8fj33//xciRI3Ho0KH3HtfbMjIygpaWVrVdPzc3F2pqatV2fSIiIiKi94XD4YmIiD4gmZmZWLVqFaytrWFtbQ1PT088f/5cplxUVBScnJxgYmICCwsLeHh44P79++XWb21tjbt37yItLU3YFxsbCwsLC1hZWeH3339HYWGhcCwmJgZ6enpo3bo1APnD4U+ePImRI0fC3NwcZmZm+OSTT7Bu3TqpMs+ePcOyZctgb28PIyMj9OvXD6GhoQo/FxUVFQwZMkRmKHpkZCSMjIzQvn17mXMuXLiA6dOno3v37jA2Nka/fv2wfv16ZGZmSpVbunQpunbtioSEBEyZMgVmZmbo2bMnNmzYgNzc3DLjevnyJVasWIGePXvCyMgINjY2cHZ2xrVr1xS+t9Lk5OTAz88Pn3zyCYyMjGBvb4+vvvpKJv79+/dj/Pjx6NatG8zMzDB06FB8++23Uq8jUDStgIeHB44dO4ZBgwbByMgIx44de+s4iYiIiIg+BuwJSkRE9AFZtmwZevXqBV9fXzx9+hTe3t5YtGgR9u7dK5Tx9/fHli1bMHbsWMyZMweZmZkIDg7GuHHjcOTIkTKH11tZWWHfvn2Ii4sThsTHxsbC2dkZVlZWyMjIwK1bt2BkZITc3Fxcu3YNn3zySan1/fHHH5g3bx7Gjx+PuXPnQkVFBYmJifjzzz+FMs+fP8fIkSNRt25dzJs3D02aNMGFCxewceNGpKamYu7cuQo9mxEjRmD37t34448/YGFhgTdv3uDkyZOYP38+8vLyZMo/fPgQFhYWGDNmDOrWrYsHDx5g27Zt+PPPP/Htt99Klc3NzYWHhwfGjBmDqVOnIjY2Flu2bIGmpiY8PT2Fcnfv3pU6b9GiRXj48CHmzZuHZs2aITU1FdeuXUNqaqpC91RQUID8/HxhW0lJCSoqKigoKIC7uztu3rwJd3d3dOnSBQkJCfDz88Nff/2FvXv3Qlm56G/Zjx49wtChQ9GsWTOoqKjg+vXr2LhxI5KSkmSe7fXr13Hv3j14eHhAT0/vvU3FQERERERU3ZgEJSIi+oA4ODhg+fLlwnZaWhq8vb2RlJQEPT09PHnyBNu2bYObmxsWL14slLOwsEC/fv0QEhKCJUuWlFq/paWl1LygkvlALS0t0bRpUzRt2lSYF/TatWvIyckRhsLLc/XqVdSrVw+rVq2S2j9q1Cjh34GBgcjOzkZkZKSQdLO1tUVeXh527doFV1dXNGjQoNxn0759e3Tp0gWRkZGwsLDAjz/+iNzcXAwaNAiHDx+WKT9u3Djh32KxGObm5mjTpg0mTJiAO3fuoGPHjsLxnJwcLFiwAH379gUA2NjY4ObNm4iOjpZKgpYUHx+PuXPnYtiwYcI+SR2K6Natm8w9Hjt2DCdOnMDly5exfft29OzZU4ipcePGmDlzJs6dOwcHBwcART1ZJQoLC2FpaYmCggKEhoZizpw5UFJSEo6npqbi4MGDaN68ucIxEhERERHVBEyCEhERfUD69OkjtS0Zev7kyRPo6enhwoULyM/Px5AhQ6R6EDZs2BCdO3cW5vAsjY6ODtq3by/MCxoTE4O6deuic+fOAIqSpJJ5QSVlykqCmpiYID09HXPmzMGQIUNgZmYGHR0dqTJnz56FjY0NdHR0pGLu2bMn9uzZg2vXrgmJvvIMHz4cvr6+WLFiBSIjI9GnTx/Ur19fbtnk5GRs27YNv/zyC5KSkqR6iyYkJEglQVVUVGRiEIlEuHTpUpnxGBsbY8eOHSgoKICNjQ06dOgAFRUVhe4FAPbu3Su1MFLt2rUBFD2zBg0awM7OTuqZ2dnZQUVFBbGxsUIS9ObNm9iyZQuuXr2KlJQUqWHwKSkp0NXVFbY7duzIBCgRERER/ScxCUpERPQBKdkjUrJoTU5ODgDgxYsXAIChQ4fKPV+RBJe1tTX279+PtLQ0YT5QSeLO0tISXl5eKCwsRGxsLJo0aYKWLVuWWlfXrl0RFBSEvXv3Ys6cOcjPz4eRkRFmzZolJBVTUlJw6tQpGBoayq3j1atX5cYsMWjQIGzYsAG7du1CbGwstm/fLrdcYWEhJk+ejJSUFMycORPt27dHnTp18OzZM3h6eiI7O1uqvIaGhswCQWpqauXOCfrNN98gODgYe/fuhZeXFxo0aIABAwZg3rx5Ci141KlTJ7nlUlJSkJqaWu4zS0xMxIQJE2BgYIAlS5agWbNmUFVVxc8//4ytW7fK3GejRo3KjYmIiIiIqCZiEpSIiOgjoq2tDQAICgpC48aNZY4rstK3lZUVwsLCEBcXh9jYWIwdO1Y41rVrV2RkZODatWu4evUq+vfvX259jo6OcHR0RG5uLuLj4xEUFAQPDw9ER0ejbdu20NbWRufOnTFr1iy551ekZ2L9+vXh6OiIoKAg6Onpwc7OTm65v/76C3/99Re8vLykhqqXXFTobeno6GDFihVYsWIFnj59ih9//BG+vr7IysrCxo0bK12vtrY2dHV1sXXr1lKPA8Dp06fx5s0bBAQEoGnTpsLx06dPyz2v+NB4IiIiIqL/EiZBiYiIPiKS4dCJiYnCwkYVJZkXNCoqCo8fP4aVlZVwrE2bNmjUqBF27tyJ7OxsqWPlUVNTQ7du3aCsrAwXFxfcu3cPbdu2Rffu3XHp0iW0adMGmpqalYq5uAkTJuD169fo2bNnuUPPVVVVpba/++67t75+aZo0aQJXV1ecPn0ad+7ceau6evTogRMnTkBZWbnU3qDA/5Kaxe8zJycHR48efavrExERERHVNEyCEhERfURatGgBDw8P+Pj44NGjR7C1tYWmpiaSk5MRHx+Ptm3bwtnZucw6tLW10aFDB/z888/Q0NCQSbJJFh0Cyp4PFAD8/Pzw/PlzYdGe1NRU7Nq1C1paWjA1NQUAzJ07FxcuXMC4cePg4uKCVq1a4fXr13jw4AHOnDmD3bt3V2geza5du6Jr165lljEwMECLFi3g6+sLANDU1MTx48dx8+ZNha9TnoyMDEyaNAmDBg1C27ZtUadOHfzxxx+Ij4+Hm5vbW9U9aNAgREVFYerUqXB1dYWhoSGUlJTw9OlTnD9/HhMnToS5uTns7OygqqqKBQsWYOrUqcjKykJISAhq1eL/4hERERERFcf/QyYiIvrIeHp6wsDAAPv27UNkZCQKCgqgp6cHU1NTGBsbK1SHtbU17t69C1NTU5neklZWVvjhhx/QrFmzcoeqm5iYYN++fdi4cSNevXqF+vXrw8TEBKtXrxbmn2zcuDEiIiIQFBSELVu2IDk5GZqammjdujW6d+8OZWXlyj2IMqiqqmLLli1Ys2YNVqxYATU1NfTu3RubNm2Ck5NTlVxDXV0dXbp0weHDh/H48WMUFhaiWbNmmDNnDqZMmfJWddeqVQvbt29HaGgooqOjERgYCDU1NTRt2hTdunUTXhcDAwNs3rwZfn5+8PT0hK6uLkaOHIlGjRphxYoVVXGbREREREQ1gpJYLBZXdxBERERE9HHILyjE1X9TqjsMIiKi/xxRs/pQUxbj5cuX1R1KldLR0QGAGndf9Hb09fWrfD77qu96QURERERERERERPQBYRKUiIiIiIiIiIiIajQmQYmIiIiIiIiIiKhG48JIRERERKQwFWUliJrVr+4wiIiI/nPqqNVCQX5edYdB9NFiEpSIiIiIKkRNmetq0v+oqqoCAPLy+MWc/oftguRhu3g7Bfl5yM/Pr+4wiD5aTIISERERUYVw9VYqjqv6kjxsFyQP2wURVSfOCUpEREREREREREQ1GpOgREREREREREREVKNxODwRERERVYhkOCMR8L85/tguqDi2i/+e/Px8pKenV3cYRESlYhKUiIiIiBRXWIDchGvVHQV9QHKrOwD6ILFd/LfUadkRqKVe3WEQEZWJSVAiIiIiUlhB9mvc3TCpusMgIqIPiGjpHqi1NanuMIiIysQ5QYmIiIiIiIiIiKhGYxKUiIiIiIiIiIiIajQmQYmIiIiIiIiIiKhGYxKUiIiIiIiIiIiIajQmQYmIiD5AIpFIoZ/ExERERkZCJBLh9u3b7zSm93UdiXPnziEgIKBC5+Tl5SEsLAyjRo2Cubk5TExMMHjwYAQFBSErK0umvEgkqvA13qfExESIRCJERkZWdyhERERERB81rg5PRET0AQoPD5fa9vHxwf379xEYGCi1X09P732G9V799ttv2Lt3L2bNmqVQ+devX8Pd3R1XrlzB2LFjMWvWLKiqqiImJgY7d+7EDz/8gJCQEOjq6r7jyKuOnp4ewsPD0bJly+oOhYiIiIjoo8YkKBER0QfI1NRUaltLSwtqamoy++l/NmzYgLi4OPj7+6Nfv37CfhsbGzg4OMDZ2RlLlizBrl27qjHKiuFrTkRERERUNTgcnoiIqIbIzMzEqlWrYG1tDWtra3h6euL58+cy5aKiouDk5AQTExNYWFjAw8MD9+/fr5IYXr58idWrV+PTTz+Fqakp7O3tMXXqVNy8eVOqXEFBAYKDg9GvXz8YGxvD0tISw4YNQ0REBABg6dKl2Lt3LwDpqQFKk5ycjIiICPTo0UMqASphamoKJycnnD9/Hjdu3JBbR3R0NEQiEa5evSpzbP369TAzM0NmZiYAwMXFBUOHDkVsbCycnJxgbGwMR0dHHDt2DABw7NgxDBw4EKamphg1apTMNZcuXYquXbvi1q1bGD9+PIyNjWFvb49NmzYhPz9fKCdvOPyff/6JuXPnolevXjA2NkafPn2wbNkyvHjxQuoaAQEBEIlEuHfvHubOnQtzc3PY2tpi2bJlyMjIKPVZEhERERHVROwJSkREVEMsW7YMvXr1gq+vL54+fQpvb28sWrRISCYCgL+/P7Zs2YKxY8dizpw5yMzMRHBwMMaNG4cjR4689fD61NRUqKioYPbs2WjYsCEyMjIQFRWFsWPH4vDhw2jXrh0AYOfOndi+fTtmz54NQ0NDZGdn459//kFaWhoAwMPDAzk5OThx4oTM1ADyxMTEID8/H7179y61TJ8+fXDw4EFcvHgRRkZGMsf79+8PLy8vfPvtt1K9L7Ozs3H48GEMGTIEmpqawv7nz59j9erVcHd3h66uLnbu3IlFixbhn3/+wfnz5zF79myoqKjA29sb06dPx+nTp6Guri6cn5ubi5kzZ8LZ2RkeHh44f/48tm/fjvT0dHzxxRel3sfjx4/Rrl07DBo0CPXr18eTJ08QGhqKcePG4fjx41BTU5Mq7+npiQEDBmD06NG4e/cuNm3aBKAosUtERERE9F/BJCgREVEN4eDggOXLlwvbaWlp8Pb2RlJSEvT09PDkyRNs27YNbm5uWLx4sVDOwsIC/fr1Q0hICJYsWfJWMbRt2xarVq0StgsKCtCjRw8MGjQI3333HT7//HMAQHx8POzs7DB58mShbM+ePYV/t2zZUpi7U5Hh4E+ePAEANG/evNQyzZo1kypbkqqqKsaOHYtt27Zh6dKl0NHRAVDUQzQtLQ0TJkyQKp+amorQ0FB07NhRuPfevXvjwIED+Pnnn1GvXj2h7MyZM3H58mWpe8zJycG0adMwduxYAIC9vT3evHmD8PBwTJ8+Hfr6+nLj7N+/v9R2fn4+LC0t0atXL5w7dw6Ojo5Sx8eMGQNXV1cAgK2tLR4+fIiIiAisW7cOSkpKpT4vIiIiIqKahMPhiYiIaog+ffpIbUuGj0uSfhcuXEB+fj6GDBmC/Px84adhw4bo3Lkz4uLi3joGsViM77//HiNGjEDXrl3RuXNnGBkZ4f79+0hISBDKGRsb4+zZs9i4cSPi4uLw5s2bt762ospK/I0dOxZisVgYlg8ABw4cgKWlJTp06CBVVl9fX0iAAkVJ1tq1a8PMzEwqAWpgYABAfvJ14MCBMtuFhYX4448/So0xMzMTmzdvFqYSMDQ0RK9evQBA6hlLlOwdKxKJkJOTIzN8noiIiIioJmNPUCIiohqiQYMGUtuSYdE5OTkAICS9hg4dKvf8snpRKmrXrl3w9vbGhAkTMGfOHGhra0NJSQkrVqwQ4gCAadOmoXbt2jh69Ch2794NVVVV2NvbY+HChULSsCKaNm0KoGgOzdI8fvwYANCkSZNSy+jq6uLTTz/FwYMHMWXKFFy/fh03b97EN998I1O2fv36MvtUVVVlXgdVVVUAkLp/AFBXV5dKlgJAw4YNART1Mi3N/PnzERcXh5kzZ8LQ0BB169aFWCzG6NGjZa4BlN8uiIiIiIj+C5gEJSIi+o/Q1tYGAAQFBaFx48Yyx0vOJVkZx44dg7W1tdSQeKAoqaelpSVs16pVC1OmTMGUKVOQnp6O8+fPw8fHB1OnTsUvv/xS4WHaVlZWqFWrFk6fPo1x48bJLfPzzz8DKFotviwuLi44evQofvvtNxw/fhyNGjVC3759KxSPInJycpCRkSGVCE1JSQEgm7iUSE9Px7lz5+Dp6YmpU6cK+x8+fFjl8RERERER1SQcDk9ERPQfYWdnBxUVFSQmJqJLly4yP2Wtvq4oJSUloeejxLlz5/Ds2bNSz9HS0sKAAQMwYsQIPHnyBOnp6QD+l5TNzs4u97p6enoYPnw4fvvtN/z4448yx69evYrIyEjY29ujS5cuZdZlbGwMExMTbN26FSdPnsSYMWNk7qmqHD9+XGZbRUUFFhYWcssrKytDLBbLJKy/++67dxIfEREREVFNwZ6gRERE/xEtWrSAh4cHfHx88OjRI9ja2kJTUxPJycmIj49H27Zt4ezsXG49Fy9exIMHD2T29+3bFz179sTWrVsREBCArl274u7du9i+fbtMz9Pp06ejQ4cOMDIygra2Nh48eICDBw/CxMREGGbevn17AEUryXfv3h3KysplJjCXLVuGhIQELFiwAHFxcejZsydq1aqF2NhYhIaGomXLlgqviO7s7IxFixahVq1aGD16tELnVJS6ujq2bduGrKwsdOzYEefPn0d4eDjGjRtX6qJImpqasLCwwK5du6CtrY2mTZvi7NmzOHv27DuJkYiIiIiopmASlIiI6D/E09MTBgYG2LdvHyIjI1FQUAA9PT2YmprC2NhYoTo2btwod398fDw8PDzw+vVrhIeHY+fOnejUqRM2b96MgIAAqbJWVlY4deoUwsPDkZWVBT09PTg6OmLWrFlCmQEDBiAuLg5hYWEIDAyEWCzG3bt3S42rbt26CA0NxYEDB3D06FFERESgsLAQLVu2xJQpU+Dq6gpNTU2F7rFfv374/PPP0adPH7lTB1QFNTU1BAcH4+uvv4afnx/q1auHzz77DHPmzCnzPF9fX6xZswZeXl4Aiob37969W2YBJCIiIiIi+h8lsVgsru4giIiIiD4kp06dwqxZsxAWFgYrK6sqr3/p0qX4+eef8fvvv1d53e9aflY6rnpYV3cYRET0AREt3QO1tiZ4+fJlmeV0dHQAoNxy9N/CdkHy6OvrV3idgPKwJygRERHR/0tISMCTJ0+wceNGmJqavpMEKBERERERvX9MghIRERH9v9WrV+PKlSswNDQUhpsTEREREdHHj8PhiYiIiEhh4oJ8ZP4VX91hEBHRB6ROy44oqKXO4fBUKWwXJA+HwxMRERFR9VJWgVpbk+qOgj4gqqqqAIC8vLxqjoQ+JGwX/y0FAPLz86s7DCKiMjEJSkREREQVwp4aVBx78JA8bBdERPShUa7uAIiIiIiIiIiIiIjeJSZBiYiIiIiIiIiIqEbjcHgiIiIiqhDJMFci4H9zP7JdUHFsF9UvPz8f6enp1R0GEdEHg0lQIiIiIlKcuBBI/qe6o6APCJe9IXnYLqqXaqPWQC1+3SciKo6fikRERESkMHFeNlIiv6juMIiIqAwNR3wBNGpX3WEQEX1QOCcoERERERERERER1WhMghIREREREREREVGNxiQoERERERERERER1WhMghIREREREREREVGNxiQoERER1UgikUihn8TERERGRkIkEuH27dvvNKbyrjN06FC4uLhUqm6RSISAgIC3CY+IiIiIqMbi6vBERERUI4WHh0tt+/j44P79+wgMDJTar6en9z7DIiIiIiKiasAkKBEREdVIpqamUttaWlpQU1OT2U9ERERERDUfh8MTERER/b/MzEysWrUK1tbWsLa2hqenJ54/fy5TLioqCk5OTjAxMYGFhQU8PDxw//79dxJTTk4O/Pz88Mknn8DIyAj29vb46quvkJmZWeZ5AQEBEIlEuHXrFtzd3WFmZgZra2usXr0aWVlZ7yRWIiIiIqIPFZOgRERERP9v2bJlUFdXh6+vLxYuXIjY2FgsWrRIqoy/vz+WLVsGY2NjBAQE4Ouvv8bDhw8xbtw4JCUlKXSdgoIC5Ofny/zIK+fu7o6wsDCMHDkSO3bswIwZM3Ds2DFMnz4dhYWF5V7L09MTnTp1QmBgICZNmoSIiAjMmzdPsQdCRERERFRDcDg8ERER0f9zcHDA8uXLhe20tDR4e3sjKSkJenp6ePLkCbZt2wY3NzcsXrxYKGdhYYF+/fohJCQES5YsKfc6Tk5OpR6zsrIS/n3ixAlcvnwZ27dvR8+ePQEANjY2aNy4MWbOnIlz587BwcGhzGsNHDhQSHra2dmhVq1a8PX1xZUrV2BmZlZurERERERENQGToERERET/r0+fPlLbIpEIAPDkyRPo6enhwoULyM/Px5AhQ6R6bjZs2BCdO3dGXFycQtfx8fFB69atZfYvXLhQavvs2bNo0KAB7OzspK5nZ2cHFRUVxMbGKpQELbnt6+uLuLg4JkGJiIiI6D+DSVAiIiKi/9egQQOpbTU1NQBF83ICwIsXLwAAQ4cOlXt+8+bNFbpOu3bt0KlTJ5n9tWvXltpOSUlBamoqDA0N5dbz6tWrcq+lq6srd1uRc4mIiIiIagomQYmIiIgUpK2tDQAICgpC48aNZY5LkqZVeT1dXV1s3bq1zHjK8uLFC6lEqCSRq8i5REREREQ1BZOgRERERAqSDENPTEyEo6PjO79ejx49cOLECSgrK5faG7Q8x48fR8eOHaW2AcDS0rJKYiQiIiIi+hgwCUpERESkoBYtWsDDwwM+Pj549OgRbG1toampieTkZMTHx6Nt27ZwdnaususNGjQIUVFRmDp1KlxdXWFoaAglJSU8ffoU58+fx8SJE2Fubl5mHcePH4eKigosLS1x/fp1BAUFoWfPnpwPlIiIiIj+U5gEJSIiIqoAT09PGBgYYN++fYiMjERBQQH09PRgamoKY2PjKr1WrVq1sH37doSGhiI6OhqBgYFQU1ND06ZN0a1bN4XmIA0KCsKmTZuwZ88eqKqqYsSIEQqtYE9EREREVJMoicVicXUHQURERERVKyAgAIGBgYiLi4OWllaV1VuYk4Vn21yrrD4iIqp6DUd8ATRqh5cvX1Z3KFJ0dHQA4IOLi6oX2wXJo6+vDyUlpSqtU7lKayMiIiIiIiIiIiL6wDAJSkRERERERERERDUah8MTERERkcLEhQXIfXKnusMgIqIyqDZqjTzU+uCGF3PYM8nDdkHyvIvh8FwYiYiIiIgUp6QMNGpX3VHQB0RVVRUAkJeXV82R0IeE7aJ65QHIz8+v7jCIiD4oTIISERERUYWwpwYVxx48JA/bBRERfWg4JygRERERERERERHVaEyCEhERERERERERUY3G4fBEREREVCGSYa5EwP/mfmS7oOLYLuTLz89Henp6dYdBRPSfxCQoERERESlOXAjljCfVHQV9QAr+/78cYkbFsV3IUqmvD9TiV3AiourCT2AiIiIiUpi4IBdpF0KqOwwioo9OfTs3oF7T6g6DiOg/i3+YIyIiIiIiIiIiohqNSVAiIiIiIiIiIiKq0ZgEJSIiIiIiIiIiohqNSVAiIiIiIiIiIiKq0ZgELUVkZCREIhFu375dbhnJT+fOndGjRw8sW7YMSUlJ5V4jJiZG6vxOnTrB1tYWs2fPRkJCQoVjDggIkKqvtJ+lS5eWW9fSpUvRu3dvYTsxMREikQihoaEKxVDdEhISYGxsXO5rKFHytRCJRLC2tsbYsWPxww8/vNNYFX22ZZ1b/MfCwgLDhw/HwYMHqz7YKiKJOzIystyyf//9N+bPn4/evXvDyMgINjY2GDVqFDZs2FCpax8/frxSz1pCkfeYSCRCYmKiQp8jVaG86wwdOhQuLi4VrrcibXPp0qXo2rWrwnUWf+3PnTuHgICACsf3rkk+F2JiYkot8zG2h3dNJBJV6et548YNjB8/Hubm5mV+buTk5CAgIEDu66Vo+3xbS5culXrdjYyM0L9/fwQGBiI3N1eqrCLti4iIiIiIqg5Xh68CXl5eaN26Nd68eYOLFy9i165duHr1Ko4ePQpVVdVyz1+4cCEsLS2Rk5ODa9euITg4GHFxcTh58iQaNGigcByjRo1C9+7dhe379+9jyZIlmDFjBhwcHIT9Ojo6Fbm9CikZQ3UQi8VYsWIFtLS0kJycXKFzJa8FALx8+RLfffcd5syZg6CgIDg6Or6LcKvEpEmTMGDAAABAeno6Tp48idWrVyM7Oxuurq7VG9xbkCQ/OnTogNmzZ6Np06Z48eIFbty4gePHjyuU0C/p+PHjuHPnTqWfS3h4uNS2j48P7t+/j8DAQKn9enp6laq/ptPT00N4eDhatmwp7Pvtt9+wd+9ezJo1qxojqxy2h3dv+fLlyMvLg7+/PzQ1NaXaTnE5OTkIDAyEp6cnrK2t33OU/6OhoYGQkKKV01+9eoXIyEgEBATg6dOnWLt2rVDO0NAQ4eHhaNeuXXWFSkRERET0n8IkaBWQ9OIEABsbG6SkpCAiIgK///47bGxsyj2/TZs2MDU1BQBYW1sLX/Z+/vlnjBw5UuE49PX1oa+vL2yrq6sDAFq2bCnU/66VjKE67N+/H48ePYK7u7vUF05FFH8tAKBHjx6ws7PD8ePHP+gkaNOmTWXivnXrFk6cOPFRJ0HDwsKgqqqKvXv3QkNDQ9g/YMAALFy4sFpiKvle0tLSgpqa2nt7j33satqz+i+2h9zcXKipqb236/3zzz8YN24c7O3t39s134aKiorM5/GAAQNw5MgRrFq1SvjdrKmpWaPbCRERERHRh4bD4d8BQ0NDAEU9CSvDyMjorc4vzw8//IARI0bA2NgYFhYWmDZtGu7cuVPhenJycjBr1ixYWloiNjYWgPzh8IWFhQgNDcXgwYPRpUsXWFtbY/HixTK9NHv37g0PDw/8+uuvGDZsGIyNjdG/f38cOnRI4ZgeP34MX19frFy5EpqamhW+p5Jq1aqF2rVrIy8vT2r/y5cvsWLFCtjZ2cHIyAiffPIJtm7dioKCgkqVK0nes62ounXrysT9+vVrbNiwAQ4ODjAyMoKDgwO8vLyQnZ0tXHfo0KH49NNP8fr1a+G8xMREWFpaYubMmcK+tLQ0rF27Fr169YKRkRF69eqFzZs3y1zz+fPnmD17NszMzGBhYYHZs2cr3EM3NTUVmpqaUglQCWVl2Y+vqKgoODk5wcTEBBYWFvDw8MD9+/eF4y4uLjh9+jQeP34sDFctPu3Du5KZmYlVq1bB2toa1tbW8PT0xPPnzyscf1WqSNssLCzE5s2bYW9vjy5dumD8+PH4888/5dZ769YtjB8/HsbGxrC3t8emTZuQn58vHC85HH7p0qXYu3cvAOmh5WU5fvw43NzcYGdnBxMTEwwcOBDBwcEyw41dXFwwdOhQXL16FWPHjoWJiQn69OmD7du3o7CwUKrsvXv3MGXKFJiYmMDa2hrLly9HZmZm+Q+yEj6U9qDI7wLJMPKbN2/CxcUFpqamWLVqFQAgIyMDy5cvh7W1NczMzDB58uQKTeXy6NEjzJ07F9bW1jAyMsKgQYOketZKhvXn5+cjLCyszLYh+YwCgMDAQKFsyWH5CQkJmDJlCszMzNCzZ09s2LBBpt3k5OTAz88Pn3zyCYyMjGBvb4+vvvqq0u1BRUUFHTt2RF5eHtLT04X98obDV6TN/v3335g8ebLQZletWoVffvlFps6bN2/C3d0dNjY2MDIyQo8ePeDh4YG0tLRK3Q8RERER0ceKPUHfgcePHwMAWrduXanzExMT3+r8skRGRmLZsmXo3bs3PD09kZWVhaCgIIwbNw6HDh2CgYGBQvW8fPkSHh4eSEpKwoEDB8oczrds2TKcOHECkydPhpWVFZ4/fw4/Pz+4uLggMjJSKsF169YteHt7w93dHQ0bNsR3332H5cuXo1WrVsIX3LKsWrUKNjY2+OSTTxSab7KkwsJCIWHz8uVLHDhwAM+ePcOyZcuEMm/evIGLiwueP3+OOXPmoE2bNjh//jw2b96MR48eCb1PFS1XUkWerYRYLBbilgyHj4+Px+LFi6Xubfr06YiPj8fMmTNhbGwsTL9w9+5d7Nq1C+rq6ti8eTNGjBiBVatWwcfHB3l5eZg3bx7q1auHdevWAQCysrIwYcIEpKamYsaMGWjbti2uX7+OoKAgPH78GN7e3sIzmDRpElJSUrBw4UK0bNkSv/76K+bPn6/Q62FiYoJff/0VixcvxpgxY9ClS5dSe6D5+/tjy5YtGDt2LObMmYPMzEwEBwdj3LhxOHLkCPT09LB69Wp88cUXUsOVi9cnSYj+8ssvCsWnqGXLlqFXr17w9fXF06dP4e3tjUWLFgnJP0XjL09BQYFUwrE0FW2be/bsQevWrfHll1/izZs3CAwMxKRJkxAVFSU1NDk3NxczZ86Es7MzPDw8cP78eWzfvh3p6en44osv5Mbi4eGBnJwcnDhxQmZoeWkePnyIXr16wdXVFbVr18bdu3exZcsW/Pvvv0Lbk3j+/DmWLFmCyZMnY/bs2Th16hR8fX2hp6eHYcOGAQCSk5Ph7OwMdXV1rF69Gg0bNsTRo0fx9ddfKxRPRb2v9lCWivwukPxRxsXFBR4eHlBVVUVhYSFmzJiBa9euYdasWTA0NMQff/yBzz77TKHrJycnY9y4cVBSUsKiRYvQqFEjnDx5EqtWrUJKSgo8PDzg4OCA8PBwjBkzBgMGDMCkSZNKrU9PTw8hISFwc3PDyJEjMWrUKACQGpmQm5sLDw8PjBkzBlOnTkVsbCy2bNkCTU1NeHp6Aih6D7m7uwuJwy5duiAhIQF+fn7466+/sHfvXrl/gCnP48ePoaWlpdB0NIq02efPn8PZ2Rl169bFl19+CW1tbRw/fhxr1qyRqisrKwuTJ09G586d8fXXX0NLSwtJSUm4cOECcnJyKnwfREREREQfMyZBq4Ak8ZCdnY3Lly/j4MGDGDRokNAjtDySxFtubi6uXbuGbdu2wdLSssp7qBUWFmLTpk0wMjJCcHAwlJSUABQNwe/bty+CgoKwadOmcut5+PAhPvvsM2hoaODgwYNlfhmPj49HVFQUVq1ahQkTJgj7O3XqhOHDh+Pw4cNS+1NTUxEeHo7GjRsDACwtLRETE4Po6Ohyk6CHDx/GlStXcOLEiXLvoTQl5ySsVasWli5div79+0td559//sHWrVvRq1cvAIC9vT0KCwuxZ88eTJ48GQYGBgqXK64iz7a4DRs2yCwUNGnSJEyePFnY/u233xATE4OVK1fC2dkZAGBnZ4e6deti3bp1OH/+PLp37442bdrgyy+/xKJFi2BtbY1//vkHt2/fxv79+1G/fn0ARcPUExISEBkZiY4dOwIomgqidu3aWLduHdzd3dG+fXscPnwY//77L7Zv346ePXsCALp37443b94gIiKi3PuaMmUK7ty5gyNHjuDIkSNQVVUVep1KEgAA8OTJE2zbtg1ubm5SiV8LCwv069cPISEhWLJkCdq1a1fmcGUVFRWFnndFOTg4YPny5cJ2WloavL29kZSUBD09PYXjL4+Tk1Opx6ysrIR/V7RtKisrY+fOncIcxxYWFujbty927tyJr776SiiXk5ODadOmYezYsUKdb968QXh4OKZPny53qoyWLVtCV1cXgOzQ8tLMmDFD+LdYLIaFhQW0tLSwbNkyLF++XGou5dTUVOzatUv4PLa1tUVcXByio6OFhFJoaChevXqFo0ePokOHDgCAnj17wtXVFU+fPlUopop4X+2hNBX9XZCbm4u5c+diyJAhwr6zZ88iLi5O6rPdzs4OKioq8Pf3LzeG0NBQvHjxAlFRUcJnSM+ePZGRkYFt27bB2dkZOjo6QtJQV1e3zPahpqYmjKLQ19eXWzYnJwcLFixA3759ARR9Zt28eRPR0dFCEvTEiRO4fPmy1GeWjY0NGjdujJkzZ+LcuXNSc2yXRvLHiLS0NHz//fe4ceMGvvzyS4U+YxRts+np6Thw4ADatm0LoOj5TZkyRfhDLFDU8zU1NRVLliwRnjMADBo0qNw4iIiIiIhqGg6HrwJOTk4wNDSEhYUFZs6ciU6dOlVo5WpJLxozMzO4urqibt262LJlC2rVqtocdUJCApKTkzF48GDhSy9Q1IPG1tZWoRVq//zzT4wZMwYtWrTAvn37yk3SnT17FsrKyhg4cCDy8/OFn/bt26Nx48YyQ707d+4sJECBonlNW7duLfWlTp4XL15gw4YNmD9/fplzkkp6TBb/KW7JkiU4dOgQDh06hJ07d2L06NHw8vJCWFiYUCYmJgb16tUTkkcSki+nkntStJxERZ9tcW5ubkLcoaGhmD59Ovbt2wcvLy+puAFIJTKKx1P89R8yZAhGjhyJr776CqGhoViwYAFMTEyE42fPnkXHjh3Rrl07qWfZo0cPqXuLjY2FlpaWkEwoXr8i1NXV4e/vj5MnT2L58uXo168fHj16hE2bNmHw4MHClBEXLlxAfn4+hgwZIhVPw4YN0blzZ8TFxSl0vZ9++gk//fSTQmUrok+fPlLbkiG9T548AVB18fv4+AjtoPhPyV7lFW2bffv2lVrkrUmTJjAzM5M7VcPAgQNltgsLC/HHH38odA+KePDgAZYsWSJM62BoaIglS5agsLAQDx48kCrbuHFjmT9IiUQiqc+U2NhYdOzYUUiASgwePLjKYi7ufbWH0lTmd4EkcSghee1LPiNF39uSoeDFE3NAURvMzs7G1atXFaqnIlRUVGQ+i0QikfDcgaLPtgYNGsDOzk7q2UsSvIpMT5KRkQFDQ0MYGhrC1tYW33zzDT777DOMGTNGoTgVabNxcXHo2LGjkACVKJncbN26NerXr4/ly5fj8OHDePTokUIxEBERERHVROwJWgV8fHzQunVrZGZm4vDhwzhy5AjWrVuH1atXK3T+kiVLYGlpiTdv3uD06dNC0mn79u1VGmdqaioAoFGjRjLHGjVqJBwvy4ULF/Dq1SuMHj1a6IVXlpSUFBQWFpa6Uu+rV6+ktov34JJQU1OTmbOtpPXr10NfXx+DBg0S5lx78+YNgKLhgJmZmdDU1MThw4elhrYDwN27d4V/t2zZEl26dBG2u3fvjsTERPj4+GDo0KHQ0tJCWlqa0HOtOMlzlTxHRctJVPTZFqevry8Vt42NDdLS0hASEoKRI0fCwMAAaWlpUFNTg5aWltS59evXh5qamkw8I0aMwKFDh6CpqYnRo0dLHUtJScGDBw9K7e0seV1TU1PLfAaKatu2rfBlv6CgAD4+Pti9ezd27tyJxYsX48WLFwCAoUOHyj2/efPmFbpeVSvZriVD8CXDUasq/nbt2gmLtBVXu3Ztqe2Ktk15ZXV1dfH3339L7VNXV0e9evWk9jVs2FBunZWVmZmJCRMmQENDA56enmjdujXU1dVx/fp1fPXVV8L8thKKfKakpqaiVatWMuUq2k4V9b7aQ2kq+rtAU1MTderUkalDXV1d5vNE0T/epKWlyV3lvbQ2WBU0NDRkptMo2RZSUlKQmppa7mdbedfZu3cvxGIxkpKSsH37dmzfvh2dOnXCgAEDyj1fkTablpYmt81K3m8S9erVQ1hYGIKDg7Fu3Tqkp6ejefPmmDBhAtzc3KSS4ERERERENR2ToFWgeOLBxsYGGRkZOHDgAIYPHw5jY+Nyzy+eeLOyskJBQQHCwsLwww8/SA3Dflva2toAIHdRmuTkZLlfvEqaNm0a/vrrL8yfPx/e3t749NNPy72msrIyvv32W7k9Wyua7CvNP//8gzt37shNtk6YMAG6urq4cOECevXqVaGFlgCgffv2OHfuHO7fvw9jY2M0aNAAN27ckCknea6S56hoOYmKPltF4haLxbh79y4MDAzQoEED5ObmIj09XSpxkZaWhtzcXKl4MjMzsWzZMnTo0AEPHz7E119/LdW7WVtbGxoaGqXOmShJhDRo0AA3b96UOa7owkjyqKioYObMmdi9e7eQhJO07aCgIKmexBLvcyXrynjf8Ve0bUqSciX3lSyXk5ODjIwMqURoSkqK3Dor6/Lly0hOTsa+ffukpsiozOJuEg0aNJB7j2/TTt/Gu24PFf1dIC9R1qBBA+Tk5Mh8niQlJSkUQ3nPvKraS0Vpa2tDV1cXW7duLfV4eVRUVKT+KNWtWzcMGDAA69atQ8+ePavk916DBg2E91Zx8vaJRCL4+fkJvw8OHDgALy8vaGlpYeTIkW8dCxERERHRx4LD4d+Bzz//HLVq1YKfn1+lzp8zZw4aNGgAf39/mdVg30abNm2gp6eH6OhoiMViYf+LFy9w6dIldOvWrdw6lJSUsG7dOowePRoLFixAVFRUmeV79OiBwsJCvHjxAl26dJH5KTmUr7LWrFmDvXv3Sv1IFuhYs2aNsEKwtra2TAzlkfQUlXz57datGzIyMnDmzBmpckePHoWSkpKQiFW0nERFn62icUvm1JO8vkePHpWJp/hxAFixYgVSUlIQHByMFStW4PDhw1LxdO/eHQ8ePICurq7c11WSuLG2tkZ6ejrOnj0r95rlKS2hcu/ePQD/6zUmGaqamJgoN57iK0qrqanJ9BasbhWJvypUtG3+9NNPyMvLE7afPn2KK1euSM0zKnH8+HGZbRUVFVhYWJQajySpp8jrIknIFR+eLxaLFZpjtjTW1ta4c+eOTM/W6OjoStf5Nt51e6iK3wWSNlLyGSn63u7WrRvu3r0rk7w+evQoateurfD8sMVVpB2VpkePHkhJSYGysrLcZ1+ZXriamppYsGABkpOTsX///krHVpylpSXu3LmDhIQEqf3Hjh0r9RwlJSV07NgRq1atQq1atd7qDwdERERERB8j9gQth+QLd0BAAAIDA7F3795Sh3ZLtGjRAmPGjMG+fftw5coVmJmZCSvxrl+/HiNGjCjz/Hr16mHatGnw8vLCiRMnMGjQIMTExGDixInw9PSUWbxHUcrKyliwYAGWLFkCDw8PjB49Gq9fv0ZQUBCUlZUxc+ZMhepRUlLCqlWrUKdOHSxduhQ5OTmlznVmaWmJESNGYPHixXBxcYGFhQXU1NTw/PlzxMTEoGfPnlXS21VeMlMyf5qRkZHcIcLy/Pvvv8JcdGlpaTh9+jTOnz+PPn36oEWLFgCK5qzbv38/Fi1ahDlz5qBt27Y4f/48QkNDMXr0aCGxq2i54irybIt78uSJEHdmZiYuX76MQ4cOwdDQUOgtZ29vD1tbW3h5eSEjIwPGxsbCiu729vaws7MDAHz77bc4efIkNm/ejBYtWqBFixaIjY3Fl19+CWNjY7Rt2xaurq748ccfMWHCBEyaNAkdOnRAbm4uHj9+jHPnzmHlypVo1qwZhg0bhtDQUOEZtGrVCr/++isuXbqk0OuxatUqZGVloW/fvmjfvj0ACCvZa2hoYOLEiQCK3nMeHh7w8fHBo0ePYGtrC01NTSQnJyM+Ph5t27YVFoNq3749Tp48ifDwcHTq1Anq6upCUuldrQ5fnorEXxUq2jbFYjGmTp2KiRMnIjs7GwEBAVBTU5NZCVxdXR3btm1DVlYWOnbsiPPnzyM8PBzjxo0rc65eyWu7c+dOdO/eXUhAyWNmZgYtLS2sXr0as2bNgpKSEg4ePCjMD1sZkyZNQkREBD777DPMmTMHOjo6OHr0KO7fv1/pOt9GVbWHixcvysyRChTN7/m2vwvs7e1haWkJLy8vZGVlCavDHzlyRKF7dHV1xeHDh+Hu7o7Zs2dDT08PJ0+exKlTpzB//nyZYfaKqF27Nlq0aIEzZ87Azs4O9erVg56entzetKUZNGgQoqKiMHXqVLi6usLQ0BBKSkp4+vQpzp8/j4kTJ8Lc3LzCsQ0ePBi7d+/Grl27MH78eGhqala4juIkbXbq1KmYPXs2dHR0cOzYMfz7778AIKxgf+bMGRw4cACOjo5o3rw5CgoKEB0djYKCAnTv3v2tYiAiIiIi+tgwCVoKyXySxXsbVYSHhwcOHz4Mf39/hISEVPh8Z2dnhIWFISgo6K2HRRc3bNgwaGhoYNu2bZg9ezZUVVVhaWmJb775psK9MhctWoQ6depg9erVyM7OxqRJk+SWW7duHUxMTPDdd99hz549UFZWhp6eHqysrKq8h9vb8vHxEf5dt25dNG/eHEuWLJFawb527drYs2cPNm3ahK1btyItLQ1NmzbF/PnzMXXq1AqXk0fRZyuxZ88e7NmzBwBQp04dNG3aFO7u7pg8ebKwGrGSkhKCg4Ph7++P8PBwBAYGolGjRpg4cSJmz54NJSUl3L59Gxs2bMDYsWOl2t0XX3yBP//8E/PmzcN3330HTU1NHDhwAFu3bsX+/fvx5MkTaGhooHnz5rC3txeGstapUwd79+7F2rVr4ePjAyUlJdjb22PTpk3CCuJlcXZ2xvHjx3HgwAEkJSUhOzsburq6sLKywrRp06QWsvH09ISBgQH27duHyMhIFBQUQE9PD6amplLTUri4uODOnTvw9vZGRkYGmjVr9t6TnvIoGn9VqGjbnDhxItLT07F69WqkpaXByMgIGzduFP4wIKGmpobg4GB8/fXX8PPzQ7169YTEYlkGDBiAuLg4hIWFITAwUBi2K4+Ojg62bNmCjRs3YsGCBahXrx4GDhwIZ2dnuLu7V+p5NGrUCPv27cPatWuxevVq1KlTB46Ojli5ciU8PDwqVefbqor2sHHjRrn74+Pj3/p3gbKyMrZs2YL169djx44dyMvLg7m5OXbs2KHQ76yGDRvi4MGD8PX1hbe3N7KystCmTRusWbMGo0aNUuj+5Pn666+xfv16uLu7Iy8vr8J/OKxVqxa2b9+O0NBQREdHIzAwEGpqamjatCm6detW6flYlZWVMX/+fLi7uyMsLAwzZsyoVD0SjRs3xr59+7Bu3TrhD2eOjo6YPXs2lixZIkxJ0apVK2hqamL79u1ISkqCuro6DAwM4OfnJ7NIFBERERFRTackLj4WjgRr1qzBt99+iz/++ENmQQgiIiKiD83KlStx7NgxxMTEvNO5kAvz3uDlifXvrH4iopqqvp0bCus1fasRLB87yVRd/+VnQLLYLkgefX39Kl/Ikz1BS7hx4wb+/PNPREREwNHRkQlQIiIi+uAEBgaiSZMmaN68OTIzM/Hrr7/i+++/x/Tp0z/4xeCIiIiIiKoDk6AlzJkzBwUFBRg8eDD+j737js/p/P84/koiQQRJEGrV7G1ErCwSK3ZrlaJK7EbE7LCKDqpq1ghNNSqKGiUUVbSoVmlorVZLi1KjtiRWdn5/+OV83TLcIYR4Px+PPL4917nOdT5n3Pc3PrnG8OHDszscERERkVSsra2ZN28e58+fJyEhgTJlyjBq1ChjrmQRERERETGn4fAiIiIiYrHkpEQSrvyb3WGIiDxxbAoWIz7J+qke8qthz5IWvReSFg2HFxEREZHsZWVNUv7i2R2FPEZSFhKNj4/P5kjkcaL3IrWkJEhISMjuMEREnlpKgoqIiIhIpqinhtxJPXgkLXovRETkcWOd3QGIiIiIiIiIiIiIPExKgoqIiIiIiIiIiEiOpiSoiIiIiIiIiIiI5GiaE1REREREMiVlrj8R+N8COHov5E56L8wlJCQQHR2d3WGIiDzVlAQVERERkUxIxiZe/5CX/0n6/8W/bbI3DHnM6L34H5u8BdE/vUVEsp++iUVERETEckkJ3Dr2Y3ZHISLyxMhbvh7YFsjuMEREnnqaE1RERERERERERERyNCVBRUREREREREREJEdTElRERERERERERERyNCVBRUREJEcYOXIk7u7u6e43mUxMmDAh0+3e73H341GeS0RERETkaaIkqIiIiIiIiIiIiORoSoKKiIiIPCRxcXHZHYKIiIiIiKAkqIiIiDyFZs+ejclk4tixYwwdOpRatWpRt25dRo0axbVr19I8ZuHChfj5+eHq6kq7du3YsWNHmm0eOnSIwMBAatWqRZ8+fQC4efMmH374IQ0bNsTV1ZWGDRsyadIkYmJiMowzKSmJ8ePHU61aNb755hujLCwsjNatW1OtWjW8vLwYPnw4Fy9eNDvWz8+PoKAgvv/+e9q1a4ebmxstWrRg5cqV93vbRERERESeWLmyOwARERGR7DJw4ECef/55OnXqxJEjR5g+fToAEydONKu3efNmnJycGDZsGLly5WL+/Pn069ePRYsWUatWLbO6gwYN4sUXX6RHjx4kJiaSlJREYGAge/fuZcCAAbi5uXHgwAHmzp3LkSNHmD9/PlZWVqlii4mJ4fXXX+fXX39lwYIFxnyno0aNYsOGDfTu3RtPT0/Onz/PzJkz8ff3Jzw8HHt7e6ONP/74gylTphAQEEChQoVYsWIFo0eP5tlnn8XDwyOrb6eIiIiIyGNLSVARERF5anXu3JmePXsCULduXf79919WrVrFBx98YJaYjI6OZs2aNTg5OQHg4+NDo0aNmDNnDvPnzzdrs2PHjvTv39/Y3r59OxEREYwdO5Zu3boZx+fLl48PPviAHTt2UK9ePbM2rly5QmBgIJcuXeKLL76gfPnyAOzdu5c1a9bw9ttv07VrV6N+5cqVefHFF1m9erVZeWRkJMuXL6do0aIAeHh4EBERwbp165QEFREREZGniobDi4iIyFPLz8/PbNtkMhEbG8ulS5fMyuvWrWskQAHs7e3x8/Njz549JCcnm9Vt2rSp2XZERAQAbdq0MStv166d2f4Up06donPnziQkJLB8+XIjAQq3E6rW1ta88MILJCQkGD8VK1akaNGi7N6926ytKlWqGAlQgNy5c1OmTBnOnDmT7j0REREREcmJ1BNUREREcgQbGxsSExPT3JeQkABArlzmv/o4OjqabdvZ2QEQGxtrVl64cOFUbRYqVIjY2Fhu3bplNgS9SJEiZvWioqKws7OjQIECZuUFCxbEzs6OyMhIs/L9+/dz9epVRo4cmaqty5cvk5SUhJeXV5rXefXq1QyvL+UatWCTiIiIiDxtlAQVERGRHKFw4cLcvHmTa9eukT9/frN958+fN+rcj7t7hsLthGTu3LnJmzevWfnd83s6OjoSFxdHdHS0WSI0KiqKuLi4VInK1q1b4+DgwIcffoiVlZUxXB/AyckJa2trvvjii1QJXYB8+fLdx9WJiIiIiOR8Gg4vIiIiOYK3tzeAsYr6nVLK6tSpc19t79y506yX5c2bN9m6dSseHh5pLmqUVlxr1641K0/ZTtl/pyFDhvDmm28yceJEQkJCjPL69euTlJTEpUuXqFatWqqfcuXK3df1iYiIiIjkdOoJKiIiIjlCnTp1aNiwIePGjePUqVO4u7uTlJTErl27WLx4MW3btqVKlSr31XbBggXp1asXgYGB2NjYMH/+fK5fv86AAQPueayvry9169Zl0qRJXLt2DTc3Nw4ePMicOXPw9fXFx8cnzeNeffVV7O3tGT9+PLdu3eK1117Dw8OD9u3bM3z4cPz9/alduzZ2dnacP3+eiIgIGjRoQIsWLe7rGkVEREREcjIlQUVERCTHmD17NvPnz2f9+vWEhYVhZWVFmTJljKTh/WratCmlSpVi8uTJXLhwgfLlyxMSEkKtWrXueayVlRVz585l1qxZLF++nODgYIoUKUL37t0ZPHhwhj1Ju3btSp48eRgzZgyxsbGMHDmSDz74gOrVq7NixQoWLlyItbU1Li4ueHp6YjKZ7vsaRURERERyMqvku5c0FRERERFJR3JiHNd/X5/dYYiIPDHylq9Hom0Brly5kt2hZDtnZ2cA3Qsxo/dC0lKsWLF7TjuVWZoTVERERERERERERHI0JUFFREREREREREQkR9OcoCIiIiJiOetc5C1fL7ujEBF5YtjkLUhigmahExHJbkqCioiIiEgmWJFoWyC7g5DHiK2tLQDx8fHZHIk8TvRe/E9iQjIJCQnZHYaIyFNPSVARERERyRQtXCB30oIWkha9FyIi8rjRnKAiIiIiIiIiIiKSoykJKiIiIiIiIiIiIjmakqAiIiIiIiIiIiKSo2lOUBERERHJlJS5/kTgfwvg6L2QO+Xk9yIhIYHo6OjsDkNERDJJSVARERERsVhSchJ/XjqW3WGIiGSLCoWfxTaX/hktIvIk0re3iIiIiFjsVnwMQ7+akN1hiIhkixltR1O5cPnsDkNERO6D5gQVERERERERERGRHE1JUBEREREREREREcnRlAQVERERERERERGRHO2Bk6BXrlzhxIkTWRCKiIiISPbYuHEjJpOJ7777LtU+Pz8/TCYTv//+u1l5XFwc1apV4/XXX39UYTJy5Ej8/Pwe2flERERERHIKi5Oga9asYezYsWZl06ZNw8fHh5YtW/Lyyy9z/fr1LA9QRERE5GHz9PTEysqK3bt3m5WfOXOGM2fOYG9vn2rfvn37iIuLw8vL61GGKiIiIiIi98HiJOiyZctISEgwtn/77Tc+/fRT3N3d6dixI7/99hthYWEPI0YRERGRh8rZ2ZkKFSqkSnRGRERQoEABWrZsmWrfnj17AJ7YJGhSUhLx8fHZHYaIiIiIyCNhcRL033//xWQyGdsbN26kYMGCzJ8/n3HjxvHSSy/xzTffPJQgRURERB42Ly8vjhw5QlRUlFG2e/duateujaenJ7/88gtJSUnGvoiICFxcXChTpgxXrlxhzJgx+Pj44OrqSrNmzQgJCSExMdGof/r0aUwmE2FhYYSGhuLn50fNmjXp3Lkz+/fvTxXPypUrad68Oa6urjRv3pyVK1emGXdsbCwzZ86kWbNmuLq64uvry7hx41KN0DGZTEyYMIGFCxfStGlTXF1d2bt37wPeNRERERGRJ0MuSyteu3aN/PnzG9u7du2ibt262NnZAeDq6sratWuzPkIRERGRR8DT05PFixezZ88emjRpAtxOgnbr1g1PT0+uXbvGH3/8gaurK3FxcRw4cIBmzZpx69Yt/P39OX/+PEOGDKFs2bLs2LGDGTNmcOrUKSZMmGB2ns8//5wKFSrw1ltvATBz5kwCAgLYsmWL8bvWl19+yZgxY2jatCmjRo0iKiqK2bNnk5CQgLX1//6GnZiYSEBAAIcOHSIgIIBq1apx/PhxZs6cyV9//cXnn39uVn/jxo0ULVqUN954A3t7e0qXLv2wb6uIiIiIyGPB4iRokSJFOHnyJHB7MaTDhw/ToUMHY//NmzexsbHJ+ghFREREHgEPDw9jXtAmTZoY84F6eHhQvHhxihcvzu7du3F1deXAgQPExsbi5eXF6tWrOXr0KCEhITRq1AgAX19fkpKSWLhwIb1796Z8+fLGeQoUKEBISIiRnHRxcaFjx45s376dVq1akZSUxMyZM3Fzc2P27NlYWVkBUKtWLVq0aEHRokWNtjZs2MDPP//MvHnzaNCgAQB16tShaNGiDBgwgB9++IGGDRsa9ePi4ggLC8PBweFh304RERERkceKxcPhvby8WLJkCfPnz2fkyJFYWVkZv2wD/PPPP2a/lIuIiIg8SZydnalYsaIx92dERAT58uWjSpUqwO0kacq+lP/18vIiIiKC/PnzGwnQFO3atTOrm6Jhw4ZmvTMrVaoEwNmzZ4Hbv1NdvHiRVq1aGQlQgFKlSlGzZk2ztrZv346joyM+Pj4kJCQYPz4+PtjY2KQ6t7e3txKgIiIiIvJUsrgn6JAhQ9i3bx9TpkwBoH///pQsWRKAhIQENm/eTLNmzR5OlCIiIiKPQMoffaOiooz5QFNGunh4eDBp0iSSkpLYvXs3zzzzDKVLlyYqKorChQunaqtIkSIAREZGmpU7OjqabadMLRQbGwvA1atXzY6/u82UZCnA5cuXiYyMpGrVqmleT0pbd8ckIiIiIvK0sTgJWqxYMb7++muOHj1K/vz5KV68uLEvJiaGcePGGT0ZRERERJ5Enp6eLFq0iD179rB7925efvllY5+7uzvXrl3jwIED7N+/nxYtWgC3k5q///57qrYuXrxo7M8MJycns+PTavPOuoULFyYkJCTDtlLc2bNURERERORpYvFweAAbGxtMJpNZAhTAwcGBJk2aGD1DRURERJ5EKfOCrlmzhjNnzuDp6WnsK1u2LEWKFCE0NJSYmBhjn7e3N9euXWPbtm1mba1duxYrKyu8vLwyFUPKedavX29WfurUKfbt22dWVr9+fS5fvoy1tTXVqlVL9aPfzUREREREbrO4JyjcXoF03bp17Nixg8uXLzNs2DCqVKlCVFQU27ZtMybiFxEREXkSOTk58dxzz/Hdd99hb2+faph57dq12bRpE4CR3GzXrh1Llixh2LBhDBkyhHLlyrFjxw7CwsLo1KkT5cqVy1QM1tbWDBkyhDFjxjBw4EBeeukloqOjmTVrVqrh7K1atWLNmjX07duXnj17UrVqVaysrPjvv//YsWMH3bt3p1atWg9wR0REREREcgaLk6C3bt2id+/e7Nu3j7x58xITE0NUVBRwuyfo1KlT6dChA6+99tpDC1ZERETkYfPy8uLIkSPUqFEDW1tbs32enp5s3LiREiVKGL0s8+TJw8KFC5k+fTohISFERUVRvHhxXn/9dfr27XtfMXTs2BGA0NBQBg4cSIkSJejfv78xTD9Frly5mDdvHmFhYaxbt47g4GDs7OwoXrw43t7e6gkqIiIiIvL/rJKTk5MtqTh58mQWLVrERx99RK1atahbty4LFiygTp06AIwbN44DBw6watWqhxqwiIiIiGSf67E3aDU/ILvDEBHJFjPajqZy4fJcuXIlu0N5Ijk7OwPo/okZvReSlmLFimX5fPYWzwm6ceNGOnfuTJMmTdIMonTp0pw5cyZLgxMRERERERERERF5UBYnQS9cuIDJZEp3f968eblx40aWBCUiIiIiIiIiIiKSVSxOgjo6OnL+/Pl09//999+4uLhkSVAiIiIiIiIiIiIiWcXihZHq1KlDeHg4ffr0SbXv1KlTrFq1irZt22ZpcCIiIiLyeMlrm4cZbUdndxgiItmiQuFnszsEERG5TxYnQQcOHEiHDh146aWXeOGFF7CysuLHH39k586dLFu2DDs7O/r16/cwYxURERGRbGZtZU3lwuWzOwx5jNja2gIQHx+fzZHI4yQnvxcJCQnZHYKIiNwHi1eHB/j999956623+Ouvv8zKK1asyJQpU6hUqVKWBygiIiIij4/k5GTOnTuX3WHIY0Sr+kpa9F5IWvReSFr0XkhaHsbq8Bb3BAVwdXVl7dq1/PXXXxw7dozk5GTKlClDlSpVsjQoERERERERERERkaySqSRoiueee47nnnsuq2MRERERERERERERyXL3lQQVERERkadXyrA1Efjf3I96L+ROOe29SEhIIDo6OrvDEBGRB5BuErRx48aZbszKyorvvvvugQISERERkcdXUnIyZ2/8l91hiIg8MsXyFyNXLvUfEhF50qX7TV68ePFHGYeIiIiIPAHiEuMI+3VhdochIvLI9Kzdg+L5nsnuMERE5AGlmwRdtGjRo4xDRERERERERERE5KGwtrTi2bNniYmJSXd/TEwMZ8+ezZKgRERERERERERERLKKxUnQxo0b8+2336a7f+vWrfc1j6iIiIiIiIiIiIjIw2RxEjQ5OTnD/UlJSVhZWT1wQCIiIpL1wsPDMZlM/Pnnn9kdSpYaO3YsJpOJkSNHPvJzP6n3dOTIkfj5+WV3GCIiIiIij5TFSVAgwyTnsWPHyJ8//wMHJCIiImKJW7dusWHDBgA2bdrE9evXszkiERERERF5XKW7MBLA6tWrWb16tbH98ccfs2LFilT1oqKi+Pvvv2nSpEnWRygiIiJPnbi4OOzs7DKsk5L4bNSoEdu2beObb76hY8eOjyhCERERERF5kmTYEzQ6OprTp09z+vRprKysuHLlirGd8nPmzBmSkpLo0KED77777iMKW0RERB6GjRs30r59e9zc3Khduzb9+vXj8OHDxv6wsDAqVarE6dOnUx07evRoPDw8iI2NNcrWrFlDhw4dqF69OrVr1yYoKIgTJ06YHefv70/btm3ZuXMnHTp0oFq1anzyySf3jHXVqlXky5ePSZMm4ejoyKpVq8z2DxgwgPbt25uVde3aFZPJxM8//2yUbdmyBZPJxLFjxwA4efIkI0eOpEmTJri5udGgQQMGDx7MyZMn7xnTmTNneP7552ndujXnzp0D4Ny5c4waNQpfX19cXV1p3rw5YWFhZsdFRERgMpnYsGEDU6dOxdfXl1q1atGzZ0+OHz+e6jzbt2+nW7du1KxZkxo1atCjRw8OHjyYqt7KlStp3ry5cd6VK1fe8xpERERERHKiDJOgPXr0YOvWrWzdupXk5GTeeustYzvlZ8uWLaxbt47x48fj7Oz8qOIWERGRLBYeHs6QIUMoWrQoM2bM4N133+XkyZN06dLFSBC2adOGXLlysWbNGrNjY2Ji2LhxI88//zy5c+cGYNasWYwaNQo3Nzdmz57N+PHj+ffff+nSpQsXLlwwO/7cuXOMHj2aTp068emnn9KwYcMMYz116hR79uyhZcuWFCxYkFatWrFv3z4jToC6devy559/cvXqVQBu3rzJgQMHyJMnD7t27TLq7dq1iyJFilC+fHkALly4gLOzMyNGjGD+/PmMHDmSyMhIOnXqxOXLl9ON6dChQ3Tu3BkXFxeWLl1KsWLFOH/+PC+99BJ79+7ltddeY968eTRp0oTJkyczY8aMVG1MnTqVCxcuMGHCBMaNG8eJEyfo378/iYmJRp2VK1fSr18/ihYtyrRp05g6dSoJCQn4+/tz5MgRo96XX37J6NGjqVixIsHBwQQFBRESEmKWABYREREReVpkOBz+Tnf2AhEREZGcJSkpienTp+Pq6srcuXONecC9vLxo2rQpc+bMYfr06Tg7O1OvXj1Wr17NgAEDjHqbN2/m+vXrdOjQAYCzZ8/yySef0KtXL4YPH26cp3bt2jRv3pwFCxYwYsQIozwyMpK5c+dSu3Zti+JduXIlycnJRk/P9u3bs3jxYlatWmWcz9vbm6SkJH7++WdatmzJL7/8AkCnTp3MEoE7d+6kTp06xraHhwceHh7GdmJiIg0bNqRu3bqsX7+eHj16pIpn+/btDB06lKZNmzJhwgRsbW0BCA4OJiYmhvDwcFxcXIDbydn4+Hjmz59Pz549cXR0NNoxmUxMnjzZ2LaxsWHo0KEcPHiQmjVrcvPmTSZNmkTz5s2ZNm2aUc/Hx4fmzZszZ84cZs2aRVJSEjNnzjQS0CnPqVatWrRo0YKiRYtadJ9FRERERHKKTC2MJCIiIjnT8ePHuXjxIq1btzZbCNHFxYW6desSERFhlLVv357Tp0+zZ88eo2zNmjWUL18eNzc3AH766ScSEhJo06YNCQkJxk+hQoWoUqWK2bEAzs7OFidAk5KS+OqrryhTpoxxTNWqValUqRJfffUVCQkJAJQvXx4XFxej1+euXbuoWbMmfn5+/Pbbb1y7do3z589z7NgxsyRoXFwcoaGhtGrVipo1a1KlShVq1KjBzZs30xyavnLlSoKCgujevTuTJ082EqBwOzlap04dnJ2dze5DgwYNiIuL48CBA2Zt3b1qu8lkAm4nlQH2799PdHR0qvtqa2uLp6encV//+ecfLl68SKtWrcyeZ6lSpahZs6ZF91lEREREJCdJtydo9+7dsbKyYv78+eTKlYvu3bvfszErKysWLlyYpQGKiIjIwxcZGQlAkSJFUu0rUqSIsR+gYcOGODk5sXr1ajw9PTl37hy7du3ijTfeMOpcunQJgLZt26Z5vpIlS6Y6h6V27NjBf//9R2BgINHR0UZ5y5Yt+eijj9i+fTuNGzcGoE6dOuzcuRO4nQRt1qwZtWvXJleuXOzevZtr164Z9VJMnDiR5cuXExAQgIeHB/nz58fKyoqAgACz+U5TfP311zg4OKSafxTg8uXLbN68mapVq6Z5LSlD9VM4OTmZbacsDpVy3pT7GhQUlGZ7KQnPlHbTe54pSVURERERkadFuknQlMWQkpOTjW0RERHJmVKSbxcvXky17+LFi2ZDtm1tbWnVqhWrVq1i7NixfPXVV1hZWdGmTZtU7c2ZMyfNodd3r/x+Z2/Fe0lZACkkJISQkJA099+ZBP3qq684ePAghw8f5p133sHOzo5atWqxa9cuoqOjKVOmDM8884xx/Pr162nbti1Dhw41yuLi4oiKikoznunTpzN16lS6du3KwoULjblF4fZ9qFKlCoMGDUrz2LuTwfeScl/feecdqlWrds966T1PEREREZGnTbpJ0K1bt2a4LSIiIjlH2bJlcXFxYd26dfTo0cNISl66dIldu3YZScUUHTp0YNGiRWzatIk1a9ZQr149Y85LuD1HpY2NDadPn6ZJkyZZFmdkZCRbt27F09OTgQMHptofGhrK9u3buXTpEoULFzZ6eM6YMYN8+fIZw/Xr1q3L2rVriYqKSjUE3crKymxIO9xOrN65ONGdnJycWLhwIa+++irdunVjwYIFVKpUCYB69eqxa9cuypYti4ODwwNff61atXBwcOD48eO88sor6dYrW7YsRYoUSTWH6alTp9i3b5/mBBURERGRp47FCyOJiIjIk2/nzp2cPHkyVXnTpk154403GDFiBEFBQXTq1ImbN28yZ84crK2tGTBggFn9ypUrU6lSJWbMmMG5c+cYMmSI2f5SpUoRFBTE1KlTOXXqFHXr1sXBwYGLFy+yd+9eypUrR7du3TId/9q1a4mLi6NHjx54eXml2p+QkMAPP/zAV199RZ8+fShWrBhlypThp59+ws/PDxsbG+D2okkpCwvdORQeoEGDBqxevZpy5cphMpn49ddfWbZsGQUKFEg3rvz58zN//nz69+9P9+7dCQ0Nxc3NjaFDh/LTTz/RpUsX/P39efbZZ7l58yYnT55k27ZtfPbZZ0ZMlsiXLx9vvfUWY8aMISoqiqZNm+Lo6MilS5f4/fffsbOzY+jQoVhbWzNkyBDGjBnDwIEDeemll4iOjmbWrFmZmnpARERERCSnsDgJWrlyZSZPnkzr1q3T3L9hwwbeeOMN/vzzzywLTkRERLLWnSuP32nv3r20a9cOe3t7PvnkEwYPHoytrS0eHh589NFHlCtXLtUxL774IhMnTsTR0TFVb0qAgQMHUr58eRYvXkx4eDiJiYm4uLhQo0YNo0dmZoWHh1OkSBEaNmyY5v66detSqlQpVq1aRZ8+fYDbSc4TJ05Qt25do56rqysFCxYkOjo6VTJ1zJgx2NjYEBISQmxsLDVq1DASnBnJly8fn376KYMGDaJXr1588sknuLu7s2rVKubMmcPHH3/MxYsXcXBwoEyZMtSrVw9r68yvUdmhQweKFy9OaGgoo0ePJiYmhsKFC+Pq6srLL79s1OvYsSNwu3fswIEDKVGiBP3792fPnj3s3r070+cVEREREXmSWSWnTPp5D5UqVWLKlCnpJkG//vpr3nzzTSVBRURERHKwW/ExfPj9pOwOQ0TkkelZuwfF8z3DlStXsjuUJ56zszOA7qWY0XshaSlWrFim1g2wROa7H6Tj7Nmz5MuXL6uaExEREREREREREckSGQ6H/+6779iyZYuxvWLFCnbu3JmqXlRUFLt27aJWrVpZH6GIiIiIiIiIiIjIA8gwCXr48GFWr14N3F4pdc+ePezZsydVPXt7e2rWrMnbb7/9cKIUERERERERERERuU9ZNieoiIiIiOR8iUlJ/Bv5b3aHISLyyBTLXwzrJCvNV5gFNPejpEXvhaTlYcwJavHq8Fu2bDFeTBERERF5OllbWVE83zPZHYY8RmxtbQGIj4/P5kjkcZKj3oskSEhIyO4oRETkAVmcBC1RosTDjENEREREnhDqqSF3Ug8eSYveCxERedxk2erwIiIiIiIiIiIiIo8jJUFFREREREREREQkR7N4OLyIiIiICKB54sVMytyPei/kTo/7e5GQkEB0dHR2hyEiIo+QkqAiIiIiYrFkkomKj8zuMORxkgPWvZGH4DF+LxzzOpFL/xQWEXnqWPzNf+3aNfLnz59hnb/++ovnnnvugYMSERERkcdTQlICW49/m91hiIjcN79yTSlo65jdYYiIyCNm8Zyg/fv3Jy4uLt39R48epVevXlkSlIiIiIiIiIiIiEhWsTgJ+scff/Dmm2+mue/48eP07NmTPHnyZFlgIiIiIiIiIiIiIlnB4iTozJkz2bp1K+PHjzcrP3nyJD169CBXrlwsXLgwywMUEREREREREREReRAWJ0Hr1avH+PHjWbJkCZ988gkAp06dokePHgB8/vnnlCxZ8uFEKSIiIiIiIiIiInKfLE6CArz44ou8/vrrzJgxg3nz5tGjRw8SEhJYuHAhpUuXflgxikg2M5lMFv2cPn2a8PBwTCYTf/755yOJ7Y8//mD48OE0atQIV1dXatWqRadOnfjkk0+4du3aI4nhfqTcp9OnT2f62K+//pqwsLAsjykuLo4lS5bQvn173N3dqVmzJo0bN2bw4MHs3r3bqPeon7Gfnx8jR458KG0/yLXc69i2bdvi7++f6XZPnz6NyWQiPDw808c+Ln799VdMJhNXrlwBYNKkSWneiy+//JLAwEAaNWqEm5sbzZo14/333zeOuxc/Pz+z7yA3NzeaN2/OtGnTuH79epZeU1ay9J2+cuUKkyZNokWLFlSvXh0PDw9eeOEF3nrrLf77779Mn3f//v3Mnj2b6Ojo+wlbREREROSJZvHq8CkCAgI4d+4c06dPx9nZmUWLFlGuXLmHEZuIPCaWL19utj116lROnDhBcHCwWbmLi8ujDIulS5cyfvx4nnvuOfr160fZsmWJjY1l//79LFq0iMuXL/PWW2890pgeha+//prDhw/Ts2fPLG132LBhbNmyhV69ejFkyBBy5crFiRMn2LZtG7/88guenp5Zej7JuX777TdKliyJs7MzAAcPHsTNzS1VvVmzZuHl5cXrr79O0aJFOXr0KHPmzGHr1q2sWbOGAgUK3PNcHh4expzlt27d4sCBA8yZM4djx44xd+7crL2wR+j69et06tSJ2NhY+vTpg8lk4ubNmxw7doxvvvmGU6dO8cwzz2Sqzf379xMcHMyLL75o0b0VEREREclJ0k2C3p3cuJOTkxP58uXD3d2db775xii3srJiwIABWRuhiGS7GjVqmG0XKFAAOzu7VOWP0t69exk3bhwNGzZk1qxZ2NraGvvq169Pnz592LdvX7bF96Q5deoUGzduZMiQIQQFBRnlPj4+dO3alaSkpEceU1xcHHZ2do/8vPLgfvvtNyPpmZiYyB9//EHXrl1T1VuzZg2FChUytj09PalQoQL+/v589dVXFvWkLVCggNl3UZ06dbh06RJLlizh5s2b2NvbP/gFZYNNmzZx6tQplixZgru7u1HeuHFjAgICsuUzKSIiIiLyJLuvJGiKzZs3s3nzZmNbSVARSXH9+nXefvttNm3aBNzurTV27FiKFi1qVm/NmjUsWrSIo0ePkitXLry8vBg+fDhlypTJsP158+ZhbW3Nu+++a5YATZEvXz58fX2N7Zs3bzJr1iw2btzIpUuXKFy4MC1btmTIkCHkyZPHqGcymejevTtlypQhLCyM8+fPYzKZeO+993juueeYM2cOq1at4saNG3h7ezN+/HijtxvcHuZaqVIlXnjhBYKDgzl9+jQlSpQgKCiINm3a3PO+3et++Pv7G0PTTSYTACVKlGDr1q0AREVFERwczHfffcfFixcpUqQIbdu2ZcCAAWnepxSRkZEAFClSJM391tapZ0/Jymfs7+9PdHQ0I0aMYNq0afz1118EBAQwaNCgNOP5888/+eijj/j111+Jj4+nYsWKBAYG0rRpU7N6586dY/bs2fz4449cuXIFJycnPDw8eO+998ifP3+abZ85c4ZXX30VGxsbPv30U4oVK5Zmvftx5coVpk+fzrZt24iKiqJ48eK0b9/eOF96Ro4cyXfffceSJUt4//33OXjwIAUKFKBPnz707NmTn376ienTp3P06FFKlSrFyJEjzd5/gGPHjjFz5kwiIiK4ceMGZcuWJSAggNatWxt1bt26xcyZM9m8eTMXL17E3t6eZ599lv79+9OoUSOLr/P333/n5ZdfBuDo0aPcvHmTatWqpap3ZwI0RUq9c+fOWXy+uzk4OJCUlJQqUbh06VKWLFnCiRMnyJcvHz4+Prz++uvGnOZvv/02X331FStWrDA+X/Hx8XTp0oUrV64YvVOTkpL4/PPPWbVqFSdOnMDe3p4GDRowbNgws89QXFwcH330EWvXruXatWtUqVLF4qkdMvuZ3LdvH3PmzGH//v3ExcVRqVIlBg8ebLwHs2fPNn63a9y4sXHcli1bNKe7iIiIiDwV0k2Cbtmy5VHGISI5zKhRo2jUqBHTpk3jv//+Y8qUKQwbNozPP//cqDNr1iw+/vhjXn75ZYYMGcL169eZO3cuXbp04auvvkp3eH1iYiI///wzrq6uqRJuaUlKSiIwMJC9e/cyYMAA3NzcOHDgAHPnzuXIkSPMnz8fKysro/53331HyZIlGTFiBAkJCUyaNIn+/fvj4+NDfHw877//PufOnWPixImMHTuWOXPmmJ3v999/58iRIwwaNIiCBQuyfPlyhg0bRq5cuXj++efTjdOS+/HOO+/w7rvvmk1HkNJb8saNG3Tt2pXIyEj69+9PuXLlOHjwIHPmzOHMmTNMmTIl3XOXL1+e/PnzM2vWLKysrPDx8bnnUNusfsbnzp1j9OjRBAYG8uyzz5IvX740z/v333/zyiuv8Mwzz/Dee+9hb2/PihUrGDhwIJMmTaJdu3YA/Pfff7z00ksABAYGUqFCBa5cucL333/PzZs300yCHjp0iH79+lGhQgWCg4NxcHDI8B7A7fcxISHhnvVu3bqFv78/58+fZ8iQIZQtW5YdO3YwY8YMTp06xYQJEzI8Pi4ujtdee42uXbsSGBjIypUrmThxIlevXuXbb78lMDAQR0dH5syZw4ABA/juu++MBNrhw4fp0qUL5cqVY+zYsTg6OrJx40befPNNYmJi6NixIwATJ07k22+/ZciQIZQrV45r165x+PBhIyGXET8/P86cOWNsf/jhh3z44YfGdpMmTQAYOHBgusltgJ9//hmAihUr3vOcAMnJycb9j4mJ4cCBAyxfvpxGjRqZPb+UJGD79u0ZPnw458+fZ+bMmbz88st89dVXFCpUiNGjR7N//36GDBnCqlWryJcvH1OmTOHw4cN88cUXxhDyUaNGsWHDBnr37o2np6fRlr+/P+Hh4Ubv09GjR7N+/Xr69OlDnTp1+Ouvvxg4cCC3bt2653VVr14dgOHDhxMYGEjt2rXTHcK+Y8cOAgMD8fT05IMPPiB37tysWLGCgIAA5s2bh6+vLx07duTatWssXLiQ4OBg49141NOYiIiIiIhkl3SToCVKlDD+OykpiXPnzmFvb4+jo+OjiEtEnnANGzZk9OjRxnZUVBRTpkzhwoULuLi4cPbsWT755BN69erF8OHDjXq1a9emefPmLFiwgBEjRqTZ9tWrV7l165bZ91RGfvzxRyIiIhg7dizdunUDbg/zzpcvHx988AE7duygXr16Rv3ExEQ+/fRTo4forVu3GDlyJGfOnGHhwoVGvaNHj/L5559z69Yt8ubNa5RfuHCBr7/+mvLlywPQoEEDWrduzcyZM9NNglp6PypUqJDudASLFi3i+PHjhIeHU6lSJeD20OA8efLwwQcfEBAQkG5iyd7enmnTpjFq1Cjjubm4uODj40OHDh3w8PBIdUxWP+PIyEjmzp1L7dq104wxxZw5c0hOTubzzz+ncOHCADRq1IgOHTowbdo02rRpg7W1NbNmzSIqKor169eb9Tp94YUX0mx3+/btDB06lKZNmzJhwoQMe87eqUOHDunuu3Me1dWrV3P06FFCQkKMXpW+vr4kJSWxcOFCevfubbwzaYmNjeXNN9/Ez88PgFq1avH9998zf/58vvnmG0qVKgXcfm5t27bl22+/5ZVXXgFg8uTJODk5sWjRIiNB5+vry9WrV5kxYwYdOnTA2tqavXv30rp1a6MXJ5j3GszIvHnziI+P5/vvv2fevHksWbIEKysrPvjgAxwcHBg8eDCA8czSEhkZyfvvv0+ZMmUy/IPBnbZu3UrVqlXNyry8vMyS/lFRUXz66ac0b96ciRMnGuWurq68+OKLhIWF8cYbb5A7d27jfrzzzju0bNmShQsXMmrUKGN4/969e1mzZg1vv/222RD/ypUr8+KLL7J69Wq6du3KsWPHWLt2LX369DHmLPXx8cHJySnd77Y7ubu789prrxEcHExgYCBWVlaUK1eOevXq0a1bN+N5A4wfP54qVaoQGhpq9BCtX78+HTp04KOPPsLX15dixYpRvHhxI1b1/hQRERGRp41Fq8PHx8fTpEkTVq5c+bDjEZEc4u7EScrQ0rNnzwLw008/kZCQQJs2bUhISDB+ChUqRJUqVdizZ0+WxRIREQGQajh6So/BlP0pvL29zYbIpySmGjZsaFavfPnyJCcnG9eUolKlSmbJLGtra1q2bMmJEye4ePFimjFmxf3Yvn07lSpVokKFCmZt1K9fH8Bshfe0NGjQgC1btvDJJ5/Qu3dvSpYsydq1a+nWrRuhoaGp6mf1M3Z2dr5nAhRuP6+6deuaJdOsrKxo06YNFy5c4J9//gFuJ7/r1q17z6kVAFauXElQUBDdu3dn8uTJFidA4fZCYStXrkz1c/d5IyIiyJ8/f6ph5Snv4b2ej42NjVmyPm/evDzzzDOULVvWLCGWslhhynOIjY0lIiKCZs2aYWdnZ/YsGjRowKVLlzh+/DgAbm5uhIeHExwczIEDB4iLi7P4PlSoUIHKlStz4cIFatSoQZUqVahcuTL//PMPDRo0oHLlylSuXDnd4d23bt1iwIABREVFMXPmTIvng/X09DTu+ZIlS3j77bc5evQor776qhH//v37iY2NTfUdULlyZZ577jmz74By5crx7rvvsm7dOoYOHUqjRo3MFiHbvn071tbWvPDCC2b3smLFihQtWtR4jul977Rq1SrDqQ/uFBgYyPbt25k0aRKdO3cGICwsjNatWxvnOXnyJCdOnKBVq1YkJSUZ8SQmJlKvXj0OHTrEjRs3LDqfiIiIiEhOZtHq8Llz58bJycmsp5OISEbu7jWektCIjY0F4NKlSwC0bds2zeMz6qWU8n10d/IxPVFRUdjZ2aUaSlqwYEHs7OxSDfUtWLCg2XZKQiy98rsTRWn1dEuZ+zAyMjLNJNCD3I8Uly9f5uTJk6l6xaW4evXqPdvImzcvDRs2NBK+x44do0ePHsyYMYNOnTqZ3cOsfsbpJcfuFhUVleY9Tjk+5XlevXrV4vk8v/76axwcHGjfvr1F9e+Ukvy7252JdLA87vTY29unSs7a2tqmeq/vfg6RkZEkJCSwYMECFixYkGbbKe/GmDFjcHFxYd26dcyePRt7e3saN27MsGHDMpx6IjExkeTkZAB++eUXmjVrRkJCAqdPn+bixYvUrFmThIQErK2t05xfNiYmhv79+/PHH38wf/58oyezJfLnz28236i7uztOTk689tprrF69ms6dOxv3Nr37f+rUKbOyxo0b4+joSGRkJH369DHbd/nyZZKSkvDy8koznpR7md45c+XKhZOTk8XXV6hQIdq1a2ckyzdv3syQIUOYPHkyK1euND5nEyZMSHdKhaioqHSnlxAREREReVpYlASF28Oqvv/++zRXdxURyayUJMCcOXPSTK5k1AvMxsYGb29vduzYYQy9zoijoyNxcXFER0ebJYyioqKIi4vL8mk+UpISd7p8+bIRS1oe5H7c2Ya9vT3jx49Pc//9zP1Xvnx5XnjhBcLCwjhx4oQxJNgSmb2mO+dlzYijo2Oa9zill23KPXZycrJ4cZ3p06czdepUunbtysKFCzMcln6/HB0d+f3331OV3x13VitQoADW1ta0b9/ebJj7ncqWLQvcTrQOHTqUoUOHcvnyZbZu3crUqVP577//WLJkSbrn6Nmzp1lP1r/++stsgcWU3pAvvvii2TyhcDtZGxQUxP79+5k3bx61atW672tN8dxzzwG3F9CC/93b9N6bu+/96NGjSUxMpEyZMrzzzjusWrXK+EOwk5MT1tbWfPHFF+TKlfrXqJRk453nvDMRmpCQYNEfJNLTrFkzKlWqxN9//23EAxAUFGRMlXC3jKYgEBERERF5WlicBB02bBi9e/dmxIgR9O7dmzJlypA7d+6HGZuI5GA+Pj7Y2Nhw+vRpY8GUzAgICGD79u289957zJw5M1Uy4ubNm+zbtw8fHx+8vb0JDQ01hnanWLt2LXB7+HtWOnLkCMeOHTMSaUlJSXzzzTeUKVMm3d6OmbkfdnZ2xMTEpCqvV68e8+fPp3Dhwvdc1Ohu169fx9ra2pgv8k4pQ6Ut7amZ4kGfcXq8vb3ZunWrWXIpOTmZdevWUbRoUWM4eL169Vi3bh0nT57k2WefzbBNJycnFi5cyKuvvkq3bt1YsGBBpnojWhr3N998w7Zt28yGxK9duxYrK6t0exY+qLx58+Lh4cGff/5J5cqV00zcpaVQoUJ07NiRvXv3snnz5gzrvvfee9y4cYPt27cTFhZm9Dj98MMPKVCgAEFBQQCpekDGxcURFBTEL7/8QkhIiNkcqg/i8OHDwO0pFgBq1qxJ7ty5Wbt2rdm7ePjwYf766y8CAgKMsiVLlrBx40ZmzZpF2bJl6dixI+PGjTPmEq1fvz7z5s3j0qVLNG3aNN0YUp7n2rVrzd6l9evXk5iYeM9ruHTpEoUKFUr1x4GbN2/y33//GZ/HlOkQjhw5wpAhQzJs8+5ewiIiIiIiTxOLk6B169bFysqKw4cPG4mDu1lZWfHHH39kWXAiknOVKlWKoKAgpk6dyqlTp6hbty4ODg5cvHiRvXv3Uq5cObOE5d1q1arF2LFjef/99+nUqROdO3embNmyxMXFceDAAZYtW0bLli3x8fHB19eXunXrMmnSJK5du4abm5uxarqvry8+Pj5Zem1FihShX79+xurwy5Yt4+jRo3z00UdZcj8qVqzIN998w/Lly6lcuTK5c+fGZDLRs2dPNm3aRNeuXenRowfPPfcccXFxnDlzhh9++IGxY8emu5jUP//8Q9++fWndujXu7u4ULlyYq1evsmHDBn744QfatGmT6cTqgz7j9AwYMIBt27bRo0cPgoKCsLe3Z/ny5fz+++9MnTrVSBoNHjyYH374gVdeeYXAwEAqVqxIZGQk33//Pa+99lqq3qn58+dn/vz59O/fn+7duxMaGpqpnq/30q5dO5YsWcKwYcOM1dd37NhBWFgYnTp1MpK3D8Nbb71F165d8ff3p3PnzhQvXpzo6GiOHz/Ob7/9xuzZswHo1KkTDRs25LnnnqNAgQIcPnyYTZs20aBBgwzbT4l94cKF+Pr6Uq1aNeLj4zl8+DDjx483G65+p8GDB7Njxw4GDBiAvb09+/fvN/Y5OztTunTpe15bdHS0cVxcXByHDx9m7ty55M+f31i0qkCBAgQGBjJz5kxGjRpFy5YtuXDhAjNnzsTFxcWY8/PPP//kww8/5JVXXqF58+bA7SkCxowZQ506dWjTpg0eHh7GCvP+/v7Url0bOzs7zp8/T0REBA0aNKBFixaUL1+eNm3asGDBAqytrY3V4efPn2+2an16vvrqK1asWEHr1q2pVq0a9vb2nDlzhsWLF3P16lXeeOMN4PbvXu+++y6BgYH069ePtm3bUqRIEa5evcqRI0e4ePEi48aNA/7XQ3bx4sW0adOGXLlyYTKZLJ5/VURERETkSWZxErRdu3YWD1UUEbHEwIEDKV++PIsXLyY8PJzExERcXFyoUaOGRcmnV155herVqxMWFsbcuXO5fPkydnZ2VKhQge7du9OlSxfgdpJg7ty5zJo1i+XLlxMcHEyRIkXo3r07gwcPzvLvNldXV1544QWCg4M5deoUJUqUYPLkyfdc7drS++Hv78/hw4eZMmUK165do0SJEmzduhUHBweWLl1KSEgIS5Ys4ezZs9jb21OyZEl8fX0zHG797LPP0qNHD37++Wc2bdrE1atXsbW1pXz58kYC7X486DNOS/ny5Vm6dCnTp09n7NixxMfHYzKZmDNnjlkvv2eeeYYvv/ySWbNmERISQlRUFM7Oznh4eKQ7P2K+fPn49NNPGTRoEL169eKTTz7B3d39vuK8W548eVi4cCHTp0834ilevDivv/46ffv2zZJzpKdSpUqEh4czZ84cpk6dSmRkJAULFqRcuXK0bNnSqOfp6cl3333HggULiI2NpVixYrzyyitGT86MJCYm8uOPPzJq1Cjg9tygsbGxZos53W3btm3A7SkT5syZY7YvraHzadmzZ4+xaJCtrS1FixalQYMG9O/f3yzpHxQUhLOzM4sXL2bdunXky5cPHx8f3nzzTZydnbl+/TpDhw6lfPnyxjUAdOzYkd27d/POO+/g5uZGmTJl+OCDD6hevTorVqxg4cKFWFtb4+Ligqenp7FAGNyep7Nw4cKEh4fz+eefU7lyZYKDg3n99dfveV0NGzbk4sWLbN26lcWLF3Pt2jUcHBxwdXUlJCTErDexr68vy5YtIyQkhHHjxnH9+nWcnJyoVKkSL774olHP3d2dgIAAVq9ezbJly0hKSmLLli1aKV5EREREngpWySkrGYiIyAPz8/OjUqVKzJ07N7tDERF5KOIS4wg/tCK7wxARuW9+5ZpS0NaRK1euZHcoT52UqWp07+VOei8kLcWKFcvyDkupl2gVERERERERERERyUEsHg6f4ueff+a7777j1KlTwO0535o0aZLlC4uIiIiIiIiIiIiIZAWLh8MnJSUxYsQI1q9fT3JyMtbW1ka5lZUVrVu3ZtKkSZo3VERERCQHS0pO4tKNi9kdhojIfXPM60RyQrKG3mYDDXuWtOi9kLQ8jOHwFvcE/eyzz1i3bh0tWrQgMDCQ8uXLA3Ds2DHmzZvHunXrqFSpEr17987SAEVERETk8WGFFQVtHbM7DHmM2NraAhAfH5/Nkcjj5HF+L5ITkklISMjuMERE5BGzuCfoCy+8QLFixZg/f36a+/v06cO5c+f4+uuvszRAEREREXl8JCcnc+7cuewOQx4j6sEjadF7IWnReyFp0XshacnWhZFOnTqFn59fuvv9/PyMeUJFREREREREREREHhcWJ0Hz5s3LpUuX0t1/8eJF8ubNmyVBiYiIiIiIiIiIiGQVi+cEdXd3Z8mSJTz//PNUrFjRbN/Ro0f54osv8PT0zPIARUREROTxkjJsTQT+N/ej3gu5053vRUJCAtHR0dkckYiIPO0snhP0yJEjdO7cmYSEBPz8/KhQoQJwOwG6detWbG1tWbZsGSaT6aEGLCIiIiLZJzk5mWtxSmaIiGXsbfORGJ+ouf4E0NyPkja9F5KWhzEnqMVJUIDffvuNCRMmsH//frPymjVrMnr0aFxdXbM0OBERERF5vCQkJbDvv93ZHYaIPCFMhauSmzxKbgigZJekTe+FpOVhJEEzHA4fGRmJo6OjsV2tWjWWLVvGlStXOH36NMnJyZQqVUpDX0REREREREREROSxlWEStE6dOphMJjw8PPDy8sLT05MCBQrg7OysxKeIiIiIiIiIiIg8ETJMgtatW5d9+/Zx+PBhFi9ejJWVFSaTyUiIenh4kD9//kcVq4iIiIiIiIiIiEim3XNO0MTERH777Td2795NREQE+/bt4+bNm1hZWWFtbU2lSpXw9PTE29ub2rVr4+Dg8KhiFxEREZFHTHOCikhmaE5QuZPmfpS06L2QtDyMOUGt71XBxsaGGjVqEBAQwPz589mzZw9Lly5l6NCheHt7c/z4ccLCwggMDMTb2ztLgxMRkadLeHg4JpPJ7KdOnTr07NmTXbt23VebsbGxzJ49m4iIiCyO9vGQ0fWNHDkSd3f3h3r+cePGUblyZaKioszKz5w5g8lkwt3dnaSkJLN9ERERmEwmli1bliUxzJ49G5PJZFFdk8nE7Nmzs+S8WWHkyJH4+fnds56/vz8mk4mWLVumuX/9+vXGZyanvusiIiIiIg/inknQu9nY2FCzZk369etHcHAwH330Ee7u7iQnJ5OYmPgwYhQRkafMpEmTWL58OcuWLePdd98lNjaWV199lYMHD2a6rdjYWIKDg9m9O2f2XMvu6/P09CQpKYk9e/aYlUdERJA3b16uXbvGH3/8YbYvJVYvL69HFmdOkC9fPo4fP87+/ftT7Vu9ejX58uV79EGJiIiIiDwhMpwT9G4xMTHs3buXiIgIIiIi+P3330lMTKRgwYI0adIEDw+PhxWniIg8RUwmE5UrVza2vby88Pb2ZvPmzbi5uWVjZHI3T09PrKys2L17N02aNDHKd+/eTZ06dTh27Bi7d+/G1dXVbJ+Liwtly5Z9oHPHxcVhZ2f3QG08SUqVKoWNjQ3h4eHUqFHDKD937hw7d+6kQ4cOfPnll9kXoIiIiIjIYyzDnqCxsbHs2rWLmTNn8sorr+Dh4UHv3r1ZuXIlRYsWZdSoUXz11Vf8/PPPBAcH06NHj0cVt4iIPEXs7e2xsbEhPj7erDwqKooJEybQqFEjXF1dadSoETNmzDDqnT592vgDXXBwsDFcePbs2Xz33XeYTCaOHDlitLdy5UpMJhMjRowwypKTk/Hy8mLKlClGWWxsLDNnzqRZs2a4urri6+vLuHHjuH79ull8SUlJhIWF0bp1a6pVq4aXlxfDhw/n4sWLZvX8/PwICgri+++/p127dri5udGiRQtWrlyZ4X3J6PrudPz4cfr06UPNmjVp0KABH374IXFxcWZ1LL2muzk7O1OxYsVUPVF3796Np6cn7u7uZvvi4uI4cOAAnp6eRtmff/5JQEAAtWvXxs3NjQ4dOvDtt9+atZcyVcJPP/3Em2++iaenJy1atEg3rmvXrjF69Gi8vLyoWbMmvXv35vjx4xley533YuLEibRu3ZpatWrh5eVF165dU03JcPr0aUwmE2FhYYSGhuLn50fNmjXp3Llzmr01V65cSfPmzXF1daV58+b3fL5pad++PV9//TUxMTFG2erVqylUqBD16tVL85h9+/bRt29f3N3dcXNzo1OnTuzYsSPT5xYREREReZJl2BPUw8OD+Ph4ihQpgru7O61bt8bT05Py5cs/qvhEROQplJiYSEJCAsnJyVy4cIGPP/4YwGw+xBs3btC1a1ciIyPp378/5cqV4+DBg8yZM4czZ84wZcoUXFxcWLBgAb169eKll16iY8eOwO1JtvPly4eNjQ27du0y5pPctWsXefLkMUt2/fHHH0RGRlKnTh0jtoCAAA4dOkRAQADVqlXj+PHjzJw5k7/++ovPP/8ca+vbf2McNWoUGzZsoHfv3nh6enL+/HlmzpyJv78/4eHh2Nvbm51nypQpBAQEUKhQIVasWMHo0aN59tln0x1pkdH1pYiLiyMoKIjOnTvTt29fdu/ezccff4yDgwMDBw7M9DWlxdPTky+++IKoqCgKFizImTNnOHPmDB4eHjg6OjJhwgSSkpKwtrbmwIEDxMbGGvOI//3337zyyis888wzvPfee9jb27NixQoGDhzIpEmTaNeundm5Ro0aRbNmzfjoo4/MEoF3SkpKon///hw4cIBBgwZRtWpVfv31V1599dV0r+FOcXFxREdHExAQgIuLCzExMWzZsoVevXrx2WefUbduXbP6n3/+ORUqVOCtt94CYObMmQQEBLBlyxby588PwJdffsmYMWNo2rQpo0aNIioqitmzZ5OQkJDhvb1bq1atmDRpEps3b6ZNmzbA7SRomzZtsLGxSVV/x44dBAYG4unpyQcffEDu3LlZsWIFAQEBzJs3D19fX4vPLSIiIiLyJMswCRoXF0euXLmoXLkyVapUoWrVqg88dE1EROReOnToYLadN29eJk2aZDYEeNGiRRw/fpzw8HAqVaoEQJ06dciTJw8ffPABAQEBVKxY0RiGXaxYMbPjAVxdXdm5cyc9e/YEbs9j2a1bN0JDQzl+/DjlypVj586d2NraUrt2bQA2bNjAzz//zLx582jQoIFx3qJFizJgwAB++OEHGjZsyN69e1mzZg1vv/02Xbt2Nc5ZuXJlXnzxRVavXm1WHhkZyfLlyylatChw+w+RERERrFu3Lt0kqJ2dXYbXB7d7Nb7xxhs0bdrUiPXQoUOsW7fOSIJaek3p8fT0ZPHixezZs4cmTZoQERGBg4MDlStXpmDBgsa8oK6ursaiPSnzgc6ZM4fk5GQ+//xzChcuDECjRo3o0KED06ZNo02bNmZJwvr16zNmzJh0YwH48ccf2bNnj9m99/HxwcbGhlmzZmV4LED+/PmZOHGisZ2YmIiPjw+nTp1iyZIlqZKgBQoUICQkxIjTxcWFjh07sn37dlq1akVSUhIzZ87Ezc2N2bNnG6tc1qpVixYtWhjP3BKOjo74+fkRHh5OmzZt+OWXXzh58iTt27fnxIkTqeqPHz+eKlWqEBoaasRXv359OnTowEcffaQkqIiIiIg8NTLserBs2TIGDx5MQkICc+fOpXPnzri7u9O3b1/mzZvHvn37SEhIeFSxiojIU2Lq1KmsXLmSlStXGom5kSNHsnnzZqPO9u3bqVSpEhUqVCAhIcH4qV+/PoBFCwXVqVOHPXv2EB8fz19//cXFixfp0qULzzzzjNEbdNeuXdSsWZO8efMa53V0dMTHx8fsvClJtpTzbt++HWtra1544QWzehUrVqRo0aKp4qtSpYpZMix37tyUKVOGM2fOPNC9tLGxMRKbKUwmE2fPnjW2Lb2m9Nw5Lyjcvve1a9fGxsaGUqVK8cwzzxj79uzZQ7FixShdujRwO/Fct25dIwEKYGVlRZs2bbhw4QL//POP2blSkrkZSTlX69atzcpTek5aYtOmTXTp0gUvLy/jD8E7d+5Mc0h9w4YNzRK1KUn5lHv8zz//cPHiRVq1amUkQOH2HJ81a9a0OKYU7du35+eff+bMmTOsWrUKNzc3KlSokKreyZMnOXHihJGITXmuiYmJ1KtXj0OHDnHjxo1Mn19ERERE5EmUYU/QGjVqUKNGDQICAkhMTOTgwYPs3r2b3bt3ExISwvTp08mbNy81atTAw8PDmPtLRETkQVSoUMFsYaR69erRpk0bxo0bR7NmzQC4fPkyJ0+epGrVqmm2cfXq1Xuex9vbm5CQEA4cOMChQ4coXbo0JUuWpE6dOuzatYuOHTvy66+/EhAQYBxz+fJlIiMj73ney5cvk5SUlO4K6HfH5+jomKqOnZ1dqrk7M8ve3j7V4kF3t2vpNaXHycnJbF7Q3bt306VLF2N/yryg3bp1Y//+/TRv3tzYFxUVZZYATVGkSBHgdg/ZtMozEhkZSe7cuSlQoIBZuYuLyz2PBfjmm28YOnQoL7zwAn379qVw4cJYW1szc+bMNJOgdz+7lPsdGxsL/O/+pRV7kSJFzBLSlvD19aVIkSIsWbKEjRs3Mnz48DTrXbp0CYAJEyYwYcKENOtERUVpVXkREREReSpYvDq8jY0NNWvWpGbNmvTr14/ExER+++03du/ezbZt24zhXX/88cfDjFdERJ5C1tbWlC9fno0bN3L58mUKFSqEk5MT9vb2jB8/Ps1jLEl41a5dm9y5c7Nz507++OMPY95Pb29vJkyYwC+//EJMTIxRDrcTfoULFyYkJCTNNp2cnIz/tba25osvviBXrtT/d/s4JZ4svaaMeHl5sWTJEv7880/OnDljtvCRh4cHU6ZMYf/+/cTExJglhh0dHY1k3Z1SFo+6O8F4Z0/K9Dg6OhIbG0t0dLRZIvTChQv3PBZg/fr1lCpVimnTppmd7+bNmxYdf7eU+3f3gljpld2LjY0Nbdu2Zf78+dja2vLCCy9keN6goCD8/PzSrJNWAlpEREREJCeyOAl6p5iYGH799Vd2795NREQEv//+O8nJyVkdm4iICHB7oZu///4bOzs7HBwcgNu9Q+fPn0/hwoV55pln0j02pVdeWovo2NnZUatWLXbs2MGxY8d4//33Aahbty5RUVGEhoaSL18+3NzcjGPq16/Phg0bsLa2TrfnZEq9efPmcenSJYuGcN+vjK7PUpZeU0a8vLxYtGgRwcHB2Nvbm7Xj7u7OtWvXWLhwIYBZgtTb25utW7dy6dIlIyGXnJzMunXrKFq0KOXKlbuvWEJDQ1m3bp3ZvKtr16616HgrKyty5cpllgD9+++/2b9/v9miU5YqW7YsRYoUYf369fTo0cMoP3XqFPv27cvUnKApXnrpJf7++2+qVauWqsfrnectVaoUR44cYciQIZk+h4iIiIhITmJREjQ2Npa9e/cSERFhJD1TVu3NnTs3tWvXxsvLK90hfyIiIplx5MgRYyjx5cuXWb16NceOHaN79+7kzp0bgJ49e7Jp0ya6du1Kjx49eO6554iLi+PMmTP88MMPjB07lhIlSpAnTx5KlSrFtm3b8PHxIX/+/Li4uBiJpzp16jB9+nSsra2NFcuLFClChQoV+Omnn2jYsKFZT85WrVqxZs0a+vbtS8+ePalatSpWVlb8999/7Nixg+7du1OrVi08PDxo3749w4cPx9/fn9q1a2NnZ8f58+eJiIigQYMGtGjR4oHv1b2uzxKWXlNGPDw8sLKyYsuWLfj4+Jjds/Lly1OoUCG2bNlCiRIlKFWqlLFvwIABbNu2jR49ehAUFIS9vT3Lly/n999/Z+rUqRb1/Lybr68vHh4eTJo0iRs3bhirw3/11VcWHd+gQQO+/fZb3nvvPZo1a8apU6eYPXs2xYsXJykpKdPxWFtbM2TIEMaMGcPAgQN56aWXiI6OZtasWRYN709LmTJl+OSTTzKsY2VlxbvvvktgYCD9+vWjbdu2FClShKtXr3LkyBEuXrzIuHHj7uv8IiIiIiJPmgyToLNmzSIiIoKDBw8aSU9bW1uqV69uJD1r1KiRaq4xERGRBzFixAjjvwsWLEjp0qV5//33ad++vVHu4ODA0qVLCQkJYcmSJZw9exZ7e3tKliyJr6+v2TDq8ePHM3HiRAICAoiPj2fgwIEMGjQIwBjqXrlyZbNh33Xq1OHo0aNmQ+EBcuXKxbx58wgLC2PdunUEBwdjZ2dH8eLF8fb2pmTJkkbdDz74gOrVq7NixQoWLlyItbU1Li4ueHp6YjKZsux+ZXR9lsjMNaXH0dERk8nE4cOH05wf3N3dnU2bNpn1AoXbCdKlS5cyffp0xo4dS3x8PCaTiTlz5tCkSROLr+FO1tbWfPzxx0ycOJFPP/2U+Ph4atWqxaeffkrLli3vefxLL73EpUuXWLFiBStXrqRcuXKMGTOGbdu2WbTgVlo6duwIQGhoKAMHDqREiRL079+fPXv23HeblvD19WXZsmWEhIQwbtw4rl+/jpOTE5UqVeLFF198aOcVEREREXncWCVnMI69UqVK5MqVC1dXV7y8vPD29qZWrVpGLxwRERERebokJCWw77+Hl7gVkZzFVLgqucnDlStXsjsUeQw4OzsD6H0QM3ovJC3FihW7r1FhGcmwJ+inn35K7dq1sbe3z9KTioiIiIiIiIiIiDwqGSZB69Wr96jiEBEREREREREREXko7mt1eBERERF5OtlY2WAqXDW7wxCRJ4S9bT4S4xOzOwwRERElQUVEREQkc3KTJ7tDkMeIra0tAPHx8dkciTxO7nwvEhISsjkaERERJUFFREREJJO0cIHcSQtaSFr0XoiIyOPGOrsDEBEREREREREREXmYlAQVERERERERERGRHE1JUBEREREREREREcnRNCeoiIiIiGRKylx/IvC/BXD0XmSPhIQEoqOjszsMERGRx56SoCIiIiJisaSkZM6ejsruMEQEKFbckVy59E86ERERS+j/MUVERETEYnFxCcwP2ZbdYYgI0CewEcVLFszuMERERJ4ImhNUREREREREREREcjQlQUVERERERERERCRHUxJUREREREREREREcjQlQUVERJ5iISEhmEwm/P39M31sbGwsYWFhdO7cGXd3d1xdXWnUqBFvvPEGu3fvNurNnj0bk8mUlWEbIiIiMJlMfPfddw+l/TutW7eO559/nmrVqmEymTh9+nSmYrp+/Tpdu3alWrVqbNmyBYCRI0fi5+f30GMXEREREXnaaWEkERGRp9jq1asB2LNnD//++y+lS5e26LgrV67Qu3dvjh07RqdOnejXrx/58+fnzJkzbNy4EX9/f3755Rfy58//MMN/ZC5fvsyoUaNo1KgR7733Hra2tri4uFh8/JUrV+jbty///PMP8+bNo06dOgAEBQXRvXv3hxW2iIiIiIj8PyVBRUREnlJ79uzhxIkTNGrUiG3btrFq1Spee+01i44dPnw4R48eJSwsDHd3d6Pcw8ODdu3a8eOPP5IrV875NePEiRPEx8fTunVrPDw8MnXs+fPn6dWrF5cvXyYsLIzq1asb+yxNOouIiIiIyIPRcHgREZGn1KpVq7CysmLs2LFUrFiRNWvWkJSUdM/jfv/9d3788Uc6duxolgC9U7169cibN2+6bSxZsoRXXnkFb29vatasSdu2bfniiy9SnX/Xrl34+/vj6emJm5sbfn5+aSZq4+LimDp1Kr6+vtSqVYuePXty/Pjxe14LwM8//0y3bt2oUaMGNWrUwN/f32w4/8iRI3nllVcAGDRoUKamD/j333955ZVXiI6OZtGiRWYJ0JS27xwOf/r0aUwmE2FhYYSGhuLn50fNmjXp3Lkz+/fvT9X+ihUraN68Oa6urrRs2ZI1a9akOcT+iy++oHXr1tSoUYNatWrx/PPPM2/ePIuuQUREREQkJ8g5XTRERETEYjdu3GDTpk14e3tTokQJ2rdvz6RJk9ixYwf169fP8NiffvoJ4IHmsjx16hRt27alRIkS2NjYcPDgQSZPnsyFCxcYOnSoUScgIIAmTZrQt29fcufOzblz5/j+++9TtTd16lTc3d2ZMGEC165dY+rUqfTv358NGzZgY2OTbhy7du2iT58+VK9enUmTJgGwYMECevXqxfz58/H29iYoKIhq1aoxbtw43nzzTTw8PHBwcLjnNf7999+8++672NnZ8cUXX2Sq1+fnn39OhQoVeOuttwCYOXMmAQEBbNmyxZhiYPny5bz99tu0aNGCUaNGce3aNYKDg4mLi8PKyspoa/369bz//vv069cPT09PEhMTOXHiBBcuXLA4HhERERGRJ52SoCIiIk+hDRs2cPPmTdq3bw9A27ZtmTZtGqtWrbpnEvTs2bMAlChR4r7PP3LkSOO/k5KS8PDwIDExkbCwMIYMGYKVlRWHDh0iLi6OcePGmc0t2q5du1TtmUwmJk+ebGzb2NgwdOhQDh48SM2aNdONY/r06RQuXJiwsDBy584NQIMGDWjSpAkfffQRy5cvp3Tp0lSoUAGAsmXLUqNGDYuuccaMGdja2rJhw4ZMD3svUKAAISEhWFvfHrTj4uJCx44d2b59O61atSIxMZFZs2ZRq1YtZs6caRxXu3ZtmjVrZjZf6d69e3nuuecYMmSIUebr65upeEREREREnnQaDi8iIvIUWrVqFQ4ODjRr1gyAQoUK0aBBA7Zs2cLVq1cf+vkPHTrEwIED8fX1pWrVqlStWpWZM2cSFRXF5cuXAahSpQq2trYMHjyYDRs2cP78+XTbu7tXaspq9CkJ27TcvHmT3377jebNmxsJUIA8efLQokULDh48yK1bt+77Ghs1akRCQgLjxo0jLi4uU8c2bNjQSIACVKpUCfjf9fzzzz9cunSJli1bmh1XvHjxVElfNzc3Dh8+zNtvv82OHTu4du3a/VyOiIiIiMgTTUlQERGRp8zx48fZt28ffn5+xMXFER0dTXR0NM2bNyc+Pp5169ZleHzx4sWBjBOMGTl9+jRdu3blv//+Y8SIESxZsoSVK1cSGBgIQExMDHB70aCwsDDs7Ox46623qF+/Pq1atWLNmjWp2nRycjLbtrOzAyA2NjbdOKKjo0lOTqZIkSKp9rm4uJCUlER0dPR9XSPASy+9xNixY9mxYwcDBgzIVCLU0dHRbPvu64mMjASgcOHCqY69u6xdu3aMHz+eI0eOEBAQgLe3N927d+fAgQOZuBoRERERkSebkqAiIiJPmZUrVwKwdu1aPDw8jJ/hw4cDt3uJZsTHxweAbdu23df5t2zZwq1bt5g9ezatW7emVq1aVKtWzWweyxTu7u588skn/Prrr6xYsYJy5coxYsQIdu3adV/nvlOBAgWwsrLi4sWLqfZduHABa2trChQo8EDn6Nq1K2PHjuXHH3/MdCI0IylJ30uXLqXal1ZZx44dWb58Ob/++itz5swhMjKS3r17ExUVlSXxiIiIiIg87pQEFREReYokJCSwdu1aypUrx+eff57qp3379hw+fJhDhw6l24arqyv16tVj+fLl7Nu3L806O3bsSHcoeUqy09bW1iiLjY1l7dq16Z7TxsaG6tWrM2LECACOHDlyz2u9F3t7e6pXr86mTZvMeozGxsayefNmqlevnuEK95a6MxE6cODALEmEli1bliJFivDNN9+YlZ89ezbdZwKQN29eGjZsSK9evbh+/Tpnzpx54FhERERERJ4EWhhJRETkKbJ9+3YuXrxI37598fLySrW/aNGihIeHs2rVKqpWrZpuO5MmTaJ379706NGDTp064ePjQ/78+Tlz5gybN29my5Yt7NmzJ81jfXx8sLW15Y033qBv377cuHGDBQsWkCuX+a8lS5cuJSIiggYNGvDMM89w69Ytli5diq2tLd7e3g92I/7fa6+9Rp8+fejZsye9e/cmOTmZBQsWcOXKFaZNm5Yl54DbiVCA8ePHM3DgQIKDg40h7vfD2tqaQYMG8fbbbzNkyBDat29PdHQ0wcHBFClSxKxX7ZgxY8iTJw+1atWiSJEinDt3jk8++YTixYsbCz6JiIiIiOR0SoKKiIg8RcLDw7Gzs0tzhXWAMmXK4Onpyfr16xkxYoTZgkF3KlSoEMuXL2fZsmVs2LCB8PBwYmNjKVKkCO7u7ixZssRsRfc7lS9fnhkzZjBz5kwGDhxI4cKFeemllyhSpAhjxowx6lWuXJkdO3Ywc+ZMLl++TL58+ahcuTKhoaHGQkEPytvbm88++4zZs2czbNgwAKpVq0ZYWBju7u5Zco4UdyZCBw0axOzZsx+ovc6dO2NlZUVoaCgDBgygRIkSBAQEsGXLFrP5Wt3d3QkPD2fDhg1ER0fj7OyMl5cXgwcPfqBErIiIiIjIk8QqOTk5ObuDEBEREZEHl7LAVePGjXn//fcfyjlu3YpjwturH0rbIpI5fQIbUbxkQa5cuZLdoaTi7OwM8FjGJtlH74WkRe+FpKVYsWJprhnwINQTVEREROQJdP78eUJDQ/Hy8qJAgQKcPXuWsLAwbty4QY8ePbI7PBERERGRx4qSoCIiIiJPIDs7O/799182bNhAVFQUefLkoUaNGowbN46KFStmd3giIiIiIo8VDYcXEREREYslJibx74lL2R2GiADFijtibZ38WA4h1fBWSYveC0mL3gtJi4bDi4iIiEi2sra2onjJgtkdhjxGbG1tAYiPj8/mSJ5GySQkJGR3ECIiIk8EJUFFREREJFPUU0PupB48IiIi8iSwzu4ARERERERERERERB4mJUFFREREREREREQkR1MSVERERERERERERHI0zQkqIiIiIpmSMgekCPxvYaSc9F4kJCQQHR2d3WGIiIhIFlISVEREREQslpwMkVdvZXcY8ljJWe+Dk7MDufSvJBERkRxH//cuIiIiIhZLSEjk202/ZXcYIg9N0+bVcHTKm91hiIiISBbTnKAiIiIiIiIiIiKSoykJKiIiIiIiIiIiIjmakqAiIiIiIiIiIiKSoykJKiIiIiIiIiIiIjmakqAiFjKZTBb9nD59mvDwcEwmE3/++edDjSnlPCaTiQMHDqTaHxsbi7u7OyaTiZEjRz7UWO4lNjaW2bNnExERka1xpOXixYvMnj3b4ue1f/9+Zs+eTXR0dKp9JpOJCRMmZHWIaZ4nvZ9JkyZl2XkiIiIwmUxZ+tzWrVvH888/T7Vq1YzPzN38/Pws+rxFREQYMX733Xf3PLe/vz/+/v4PfA0p57zzx8vLi5dffpmNGzc+cPsZOX36NCaTibCwsPtu4+eff6ZGjRpcunTJKIuLi2PJkiW0b98ed3d3atasSePGjRk8eDC7d+/OgsifXDdu3ODjjz+mdevW1KxZk1q1atG8eXPefPNNDh8+bNTL6LvhUcno+2zUqFEMGDAgG6ISEREREcl+Wh1exELLly832546dSonTpwgODjYrNzFxeVRhgVAvnz5WLVqFdWrVzcr//bbb4mJicHOzu6Rx3S32NhYgoODGThwIF5eXtkdjplLly4RHBxMiRIlqFy58j3r79+/n+DgYF588UUKFCjwCCJM2/PPP0+PHj1SlWfHO2ipy5cvM2rUKBo1asR7772Hra1tmvEGBwcTFxdnbC9cuJANGzak+hxWqFCBQ4cOWXz+d9555/6DT8Obb76Jh4cHAFeuXGHFihUMGTKEOXPm0KRJkyw9V1ZJSkpi4sSJdO3alcKFCxvlw4YNY8uWLfTq1YshQ4aQK1cuTpw4wbZt2/jll1/w9PTMxqizT2JiIn369OHo0aP06dOHatWqkZiYyD///MPmzZs5fPgwlSpVAh6P74aMvs+CgoJo0aIFP//8M97e3tkSn4iIiIhIdlESVMRCNWrUMNsuUKAAdnZ2qcqzQ8uWLdmwYQNvvfUWefLkMcrDw8Px8/Nj586d2RjdwxUXF/dYJHmzQ+HChR+L9y8zTpw4QXx8PK1btzaSh2mpUqWK2fbXX38NpP4cZlaFChUe6Pi7lS1b1iym+vXr4+Pjw9dff/3YJkG3b9/OkSNHmDt3rlF26tQpNm7cyJAhQwgKCjLKfXx86Nq1K0lJSdkR6mNhz5497Nu3j8mTJ9O2bVujvEGDBvTs2fO+701SUhKJiYnY2tpmVaj3VKpUKerVq8enn36qJKiIiIiIPHU0HF7kIbp+/Tpvv/02Xl5eeHl5MXDgQM6fP5+q3po1a+jQoQPVq1endu3aBAUFceLECYvP06pVK2JiYvj222+Nsv/++49du3bRvn37NI85deoUQ4cOxcvLC1dXV1q1apWql13KkN8NGzYwdepUfH19qVWrFj179uT48eNmdQ8dOkRAQAB16tTB1dWV+vXrExQURFRUFKdPnzYSXsHBwcbw4dmzZxvH79u3j759++Lu7o6bmxudOnVix44dZueYPXs2JpOJQ4cOERgYSK1atejTp0+69+XKlSu88847tGzZkho1auDr60vfvn3Neg5GRETQrl074PZQ0ZTYwsPD02xz9uzZTJw4EYDGjRubTYNwp/DwcFq0aEH16tVp06YN27ZtS9XWsWPHGDx4sPEMWrduzbp169K9nvvh7+9P27Zt2b9/Py+//DLVq1encePGzJs3L1Xy5tixY/Tp04fq1avj5eXF6NGjuX79usXn+vnnn+nWrRs1atSgRo0a+Pv7mw2jHjlyJK+88goAgwYNwmQyZcnQ9BRxcXH3fE/vHg6fmJjI3Llzad68OW5ubnh4eNCuXTtWrVp1XzHkypWLPHnyEB8fb1Z+5coVxowZg4+PD66urjRr1oyQkBASExPvq97dYmNjGTRoEB4eHvccur58+XJq165NiRIljLLIyEgAihQpkuYx1tbmvy6cO3eOUaNG4evri6urK82bN081PN+Se/vbb78xdOhQGjVqhJubG40bN2bUqFFmw/Thf5/9w4cP079/f2rWrIm3tzfTpk0jKSmJ33//3Xj3mjVrxpo1a1JdgyUxp8XSe3Ov74aUqTIWLlxI06ZNcXV1Ze/evYDl3wX3ugZLvs/atGnDTz/9xKlTp+557SIiIiIiOYl6goo8RCnDfqdNm8Z///3HlClTGDZsGJ9//rlRZ9asWXz88ce8/PLLDBkyhOvXrzN37ly6dOnCV199ZdHQZkdHR/z8/AgPD6d169YArF69mkKFClGvXr1U9S9evEiXLl2wsrJi2LBhFClShG+++Ya3336by5cvm/UEg9tD/93d3ZkwYQLXrl1j6tSp9O/fnw0bNmBjY8ONGzfo3bs3VapUYfz48RQoUIALFy7w008/ERsbi4uLCwsWLKBXr1689NJLdOzYEYBixYoBsGPHDgIDA/H09OSDDz4gd+7crFixgoCAAObNm4evr69ZPIMGDeLFF1+kR48eGSaHIiMjsbGxYfDgwRQqVIhr166xZs0aXn75ZVavXk2FChWoWrUqkyZNYsSIEfTv35+GDRsCULp06TTb7NixI9euXWPhwoUEBwcbiZE7n9OWLVs4cOAAQ4YMwd7entDQUAYOHMjGjRspVaoUAIcPH6ZLly6UK1eOsWPH4ujoyMaNG3nzzTeJiYkx7lFGkpOTSUhISFVuY2ODlZWVsX3+/HlGjBhB7969GTx4MJs3b2batGm4uLgYCZOLFy/SrVs3cufOzTvvvEOhQoVYu3Yt48ePv2ccALt27TISqClzkqY88/nz5+Pt7U1QUBDVqlVj3LhxxjByBwcHi9q3xL3e07SEhoYyb948Bg8eTNWqVYmJieHo0aNERUVZdM6kpCTjGVy5coWlS5caiaoUt27dwt/fn/PnzzNkyBDKli3Ljh07mDFjBqdOnTLmkLW03t2uXLlCUFAQFy5cYOnSpRn2do2Li2PXrl306tXLrLx8+fLkz5+fWbNmYWVlhY+PD88880yabZw/f56XXnqJfPny8dprr/HMM8/w008/MXnyZCIjIxk6dKjF9/bMmTNUqFCBVq1aUbBgQc6ePUtYWBhdunTh66+/TtXLe+jQobRv3x5/f3+2bt3KvHnziIuL4/vvv6dPnz4EBgayePFiRo4ciclkMoaDWxpzWqpWrYqtrS3jx49n0KBBeHl5UahQoVT1LPlu2LhxI0WLFuWNN97A3t6e0qVLW/xdYMk1WPJ95uXlRXJyMj/++KPxRwkRERERkaeBkqAiD1HDhg0ZPXq0sR0VFcWUKVO4cOECLi4unD17lk8++YRevXoxfPhwo17t2rVp3rw5CxYsYMSIERadq0OHDgQGBnL27FmKFy/O6tWradu2bZrJn7CwMC5dusSaNWuMuewaNGjAtWvX+OSTT+jWrZvZfHYmk4nJkycb2zY2NgwdOpSDBw9Ss2ZNjh8/TmRkJCNGjDDag9s9VFO4uroCtxOfdw9pHj9+PFWqVCE0NNToVVW/fn06dOjARx99lCoJ2rFjR/r373/Pe1KuXDnefvttYzsxMZH69evTqlUrVqxYwVtvvYWDgwMmkwm4nSi413DrYsWKUbx4cQAqV65MyZIlU9WJj48nLCwMe3t74HYSpV69emzYsIF+/foBMHnyZJycnFi0aJFRz9fXl6tXrzJjxgw6dOiQqvfd3RYtWsSiRYtSlYeEhNCoUSNjOzIykvnz51O1alUA6taty549e1i3bp2RBA0LC+Pq1ausXbuW5557DvjfcN///vsvwzgApk+fTuHChQkLCyN37tzG8U2aNOGjjz5i+fLllC5d2kjQ3T2MPCvc6z1Ny969e/Hx8aF3795GWYMGDSw+56BBg8y2c+XKxciRI2nRooVRtnr1ao4ePWr2XHx9fUlKSmLhwoX07t2b8uXLW1zvTv/++y+vvvoq9vb2LFu27J5/NDl8+DAxMTGpphuwt7dn2rRpjBo1yvjOcnFxwcfHhw4dOphNXRAcHExMTAzh4eHG+erWrUt8fDzz58+nZ8+eODo6WnRv77xPAAkJCXh4eNCoUSN++OGHVFMKdO3a1ejJW6dOHbZt20ZYWBhLlizB3d0duP1dU7duXdavX28kQS2NOS2lSpXivffeY8KECbz22mtGmY+PD126dDG+8yz5boiLiyMsLMws+T969GiLvgssvYZ7fZ8VKlSIokWLsnfvXiVBRUREROSpouHwIg9R48aNzbZT/nF69uxZAH766ScSEhJo06YNCQkJxk+hQoWoUqUKe/bssfhcvr6+FCpUiNWrV7Nnzx7+/fffdIfCpwxzvzNhCdCuXTtiYmLYv3+/Wbmfn1+G11GmTBkKFizI6NGjWb16daaGWZ48eZITJ07QqlUro1ddQkICiYmJ1KtXj0OHDnHjxg2zY5o2bWpR28nJyXz55ZfGatdVqlTB1dWVEydOpBomnZW8vb2NZAbcnruzUKFCnDlzBrg9dDkiIoJmzZphZ2dn9uwbNGjApUuXLIqvVatWrFy5MtVPSjIoRdGiRY0EaAqTyWTEA7B7924qVapkJEBTpPQszsjNmzf57bffaN68uZEABciTJw8tWrTg4MGD3Lp1657tPKh7vadpcXNzY/v27UyePJk9e/ZkOs4RI0YY9z00NJROnToxadIks+R0REQE+fPnN0tMA0YCOmX4uqX1Uvz222907tyZUqVKsXjxYot6jV+4cAEAZ2fnVPsaNGjAli1b+OSTT+jduzclS5Zk7dq1dOvWjdDQUKPe9u3bqVOnDs7Ozqne3bi4OA4cOABYdm+vX7/OjBkzjCHzVatWNa4/rc9ASs9GACsrK8qVK0e+fPnM3nlHR0ecnZ3NnrulMaenQ4cOfP/998ycORN/f38KFCjA8uXL6dChA+vXr8/w2Dt5e3ubJUAz813woNdwp0KFCqU5NYuIiIiISE6mnqAiD9HdPYtShnbGxsYCGPPe3bnYxp3S6kmUHhsbG9q2bWskIatXr56q11iKqKioNId7pwzfTJkDL4WTk1OG15E/f34WLVrE3Llz+eCDD4iOjqZkyZJ07dqVXr16mQ3NvlvKPZgwYUK6w32joqLIly9fqjjvZf78+UyZMoWuXbsyZMgQnJycsLKyYsyYMUbsD0NaPcrs7OyM1c4jIyNJSEhgwYIFLFiwIM02rl69es/zODs7U61atQeOJyWmZ599NlU9S+51dHQ0ycnJadZ1cXEhKSmJ6Oho8ubNe8+2HsS93tO09OvXjzx58rB27Vo+++wzbG1t8fX15c0330z383On0qVLmz2DevXqcfr0aaZOnUrbtm0pUKAAUVFRZquwp7j782ZpvRQ//fQTV69epVOnTmafj4zExMQApLuYWN68eWnYsKGRbDx27Bg9evRgxowZdOrUiQIFCnD58mU2b96cKrGeIuXdteTevv766+zZs4cBAwZQtWpV8uXLR3JyMp06dUrzuRUsWNBs29bWNt33+87jLY05IwUKFKBFixZG79W9e/fSp08fJkyYYNbrPSN3f0Yy812QFdeQInfu3A/1O1BERERE5HGkJKhINkpJ2syZM4eiRYum2p/ZVc/bt29PaGgoZ86c4Z133km3nqOjY6qFR+D2vJAp+zPLZDIxc+ZMkpOTOXLkCEuXLmXSpEkUKFCAl156Kd3jUu5BUFBQqp58Ke5ODGWUVL3T+vXr8fLyMhsSD7cTD3cO93/UChQogLW1Ne3bt+fll19Os07ZsmUfaUz3eicyUqBAAaysrNKse+HCBaytrbP1fmckV65c9OnThz59+hAdHc2OHTuYOnUqffv2ZevWrRa/a3eqWLEiP/zwAydOnMDNzQ1HR0d+//33VPXu/rxZWi9Fv379+Ouvv3j99deZMmUKLVu2vGdsKZ83S+c8LV++PC+88AJhYWHG9Tg5OVGlSpVUUwGkSPnjzb3u7bVr1/jhhx8YOHAgffv2NY7/999/LYotMyyNOTNq1aqFj48P3377LZcvX05zntC73f0+Zea7ICuvIb0/eoiIiIiI5GRKgopkIx8fH2xsbDh9+nSque/uR/ny5enSpQvnzp3j+eefT7eet7c3n376KYcPHzYbEr927Vry5MnzQHM1WllZUalSJd5++21WrlzJ4cOHgf8ldFN6oqUoW7YspUqV4siRIwwZMuS+z5teLLa2tmZlP/zwA+fOnTPrCZtebOmxpIdhRvLmzYuHhwd//vknlStXJleu7P8q9vLyIjQ0lL///puKFSsa5ZasVm9vb0/16tXZtGkTb775pjEkPjY2ls2bN1O9evWH3gs0KxQoUIDnn3+e48ePM3v2bKKjo1P1PLTEkSNHgP8lHL29vfnmm2/Ytm2b2VD3tWvXYmVlhZeXV6bqpbCysuKDDz4gb968vPHGG8TGxhpD59OT0gPz7kTj9evXsba2NpvGIUXKcOyUXoz16tVj165dlC1b1uKFrdK6tzY2NiQnJ6f6Y8+KFSssajMz7ifmFFevXsXBwSHVd0liYiInT54kT5485M+fH8j8d0NmvgssvYZ7fZ8lJCRw7ty5VNO1iIiIiIjkdNn/L2+Rp1ipUqUICgpi6tSpnDp1irp16+Lg4MDFixfZu3cv5cqVo1u3bplq8913371nnZ49e7J69WoCAgIYPHgwLi4ufPPNN2zevJnXX3890732tm3bxtKlS2nSpAklS5YkMTGRdevWGfN6wu35IUuVKsW2bdvw8fEhf/78uLi4ULRoUd59910CAwPp168fbdu2pUiRIly9epUjR45w8eJFxo0bl6l4UjRo0ICQkBBmz56Nu7s7R44cYd68eal63ZYsWZK8efOyfv16KlasSJ48eShZsmSq4dUpUubNXLx4MW3atCFXrlyYTKZM9dx96623jEVeOnfuTPHixYmOjub48eP89ttvzJ49+55tXLp0KdX8rXB7yHBme5L26NGDVatW8eqrrzJkyBCcnZ1Zu3YtJ06csOj41157jT59+tCzZ0969+5NcnIyCxYs4MqVK0ybNi1TsTxKgYGBPPfcc7i6uuLk5MTJkydZtmwZ1atXtygB+s8//xjPICoqii1btrBjxw4aN25MqVKlgNtzei5ZsoRhw4YxZMgQypUrx44dOwgLC6NTp06UK1cuU/XuZGVlxdtvv03evHkZOXIksbGxdO7cOd14ixUrRunSpTlw4ABdu3Y1u46+ffvSunVr3N3dKVy4MFevXmXDhg388MMPtGnTxlgtfujQofz000906dIFf39/nn32WW7evMnJkyfZtm0bn332GTY2Nhbd29q1azN//nycnJwoXrw427dvZ/v27ZY9vEywNOa0REREMHHiRFq3bk3NmjUpWLAgFy5cYOXKlfz111/079/f+Ozfz3eDpd8Fll7Dvb7Pjhw5wq1bt1Il1UVEREREcjolQUWy2cCBAylfvjyLFy8mPDycxMREXFxcqFGjBm5ubg/lnIUKFWLZsmVMmzaNKVOmcOPGDcqWLcv7779Px44dM93es88+i4ODA/PmzePChQvkzp2b8uXLM3PmTLPVoMePH8/EiRMJCAggPj6egQMHMmjQIHx9fVm2bBkhISGMGzeO69ev4+TkRKVKlXjxxRfv+zqDgoK4efMmy5cvJzQ0lMqVKzNjxoxUCcbcuXMzYcIEZs2aRc+ePUlISGDixInpLizl7u5OQEAAq1evZtmyZSQlJbFly5ZMDUetVKkS4eHhzJkzh6lTpxIZGUnBggUpV66cRcOaATZs2MCGDRtSlTdu3Ji5c+daHAvc7uW3ePFiJkyYwDvvvEPevHlp0qQJY8eOJSgo6J7He3t789lnnzF79myGDRsGQLVq1QgLC0u1UNPjxNPTk82bN7N8+XJu3LiBi4sLTZo0SXfI8d2mTp1q/He+fPkoWbIkI0aMMEsw5smTh4ULFzJ9+nRCQkKIioqiePHivP7662bDwC2tl5Zhw4aRN29e3nnnHWJiYujRo0e6dVu3bs0XX3xBfHy80bvx2WefpUePHvz8889s2rSJq1evYmtrS/ny5Y0kXYqiRYuyatUq5syZw8cff8zFixdxcHCgTJky1KtXD2tra4vv7bRp03j//feZNGkScHvF988++yzdqTHul6Uxp6VGjRq0a9eOiIgIVq9eTWRkJHnz5sVkMjFp0iSz3rf3891g6XeBpddwr++z7777jsKFC1OnTp0HvKsiIiIiIk8Wq+Tk5OTsDkJEREQejfPnz9O4cWM++ugjmjZtmt3hyCOUlJREkyZNaNu27QNNPxIXl8CKpbuyMDKRx0vT5tVwdMrLlStXsjuUJ5qzszOA7qOY0XshadF7IWkpVqzYfa3RkJH0uz6IiIhIjlO0aFH8/f0JCQnJ7lDkEVu7di23bt2id+/e2R2KiIiIiMgjp+HwIiIiT5kBAwaQL18+Ll26ROHChbM7HHlEkpOTmTJlirGQk4iIiIjI00TD4UVERETEYklJyVy8EJXdYYg8NE7ODiQnJ2pY5gPS8FZJi94LSYveC0nLwxgOr56gIiIiImIxKytwdMqb3WHIYyRlkbX4+PhsjiRrJCcnkpCQkN1hiIiISBZTElREREREMkU9NeRO6sEjIiIiTwItjCQiIiIiIiIiIiI5mpKgIiIiIiIiIiIikqNpOLyIiIiIZErK8GcR+N+coE/6e5GQkEB0dHR2hyEiIiIPiZKgIiIiIpIpt24lZncI8hjJCe+Dg0Pu7A5BREREHjIlQUVERETEYomJSRw4eCq7wxDJUtXdSpE3r012hyEiIiIPkeYEFRERERERERERkRxNSVARERERERERERHJ0ZQEFRERERERERERkRxNSVARERF5ZPbu3UtgYCD169fH1dUVX19funbtyrx588zqmUwmJkyY8NDjCQ8Px2Qycfr06Yd+roykdb3Jycl88MEHmEwmpk2bBjw+8YqIiIiIPGm0MJKIiIg8Etu2bSMoKAhvb29GjhxJoUKFOH/+PPv372fTpk0EBARkd4iPjcTERMaOHcuqVat48803efXVVwFo2LAhy5cvx8XFJZsjFBERERF5sigJKiIiIo/E/PnzKV26NKGhodjY/G8V5jZt2pCUlJSNkT1e4uLiePPNN/n222957733ePnll419zs7OODs7Z2N0IiIiIiJPJg2HFxERkUciMjISZ2dnswRoCmvrtH8lCQ8Pp0WLFlSvXp02bdqwbds2s/0nT55k5MiRNGnSBDc3Nxo0aMDgwYM5efJkqrb27dvHyy+/TLVq1fD19WXy5MnEx8enqpeQkMDcuXNp2rSpMWR/7NixXL161ajzwQcf4O3tTXJyslE2fPhwTCYTq1atMsoOHz6MyWRi+/bt975BwK1btwgKCmLr1q1MnTrVLAGacj/uHg6/a9cu/P398fT0xM3NDT8/P1577TVjf2JiInPnzqV58+a4ubnh4eFBu3btzOIUEREREcnp1BNUREREHonq1auzcuVKPvzwQ1q1akWlSpXIlSv9X0W2bNnCgQMHGDJkCPb29oSGhjJw4EA2btxIqVKlALhw4QLOzs6MGDECR0dHLl26xNKlS+nUqRMbNmygUKFCAPz111/07NmT0qVLM2nSJHLnzs2SJUvYsGFDqvOOGTOGNWvW0KtXL3x8fDh+/DgzZsxg//79fPnll+TJk4e6deuycOFC/vzzT6pUqQLAzz//TJ48edi1axcdOnQAbicobW1tcXd3v+f9uX79On369OH3338nODiYhg0b3vOYU6dOERAQQJMmTejbty+5c+fm3LlzfP/990ad0NBQ5s2bx+DBg6latSoxMTEcPXqUqKioe7YvIiIiIpJTKAkqIiIij8Qbb7zBv//+y4IFC1iwYAF58uShZs2aNGnShM6dO2Nra2tWPz4+nrCwMOzt7QGoWrUq9erVY8OGDfTr1w8ADw8PPDw8jGMSExNp2LAhdevWZf369fTo0QOAuXPnYm1tzcKFC43h5A0bNuT55583O+fRo0dZvXo1vXv3ZsSIEQD4+vpSvHhxBgwYwJo1a3j55Zdxd3cnV65c7Ny5kypVqnDs2DHOnz9P3759+eqrr4z2du7ciZubG/ny5bvn/QkPDwdg0qRJFiVAAQ4dOkRcXBzjxo0jf/78Rnm7du2M/967dy8+Pj707t3bKGvQoIFF7YuIiIiI5BQaDi8iIiKPhLOzM4sWLWL16tUMGzaMBg0a8McffzB+/Hg6d+5MbGysWX1vb28jAQpQuHBhChUqxJkzZ4yyuLg4QkNDadWqFTVr1qRKlSrUqFGDmzdvcvz4caPe7t27qVu3rtl8mjY2Nrzwwgtm54yIiABuz1N6pyZNmpA/f35jv4ODA9WqVWPXrl3A7R6fJUqU4OWXX+bixYv8/fffxMfH88svv1CnTh2L7o+HhwcODg7MmjXL7BozUqVKFWxtbRk8eDAbNmzg/Pnzqeq4ubmxfft2Jk+ezJ49e7h165ZFbYuIiIiI5CRKgoqIiMgjVaVKFfr27cusWbP44YcfaNWqFYcOHWLlypVm9RwdHVMda2dnR1xcnLE9ceJEpk+fTpMmTQgODubLL79k5cqVODs7myVVIyMjKVy4cKr2ihQpYradMkQ8rbqFCxcmMjLS2K5Tpw6//vorcXFx7Nq1izp16lCqVClKlCjBrl272L9/Pzdv3rQ4CVq5cmU+++wzoqKi8Pf3tygRWrp0acLCwrCzs+Ott96ifv36tGrVijVr1hh1+vXrx2uvvcZPP/1kzB3av39/jh07ZlFcIiIiIiI5gZKgIiIikm3y5MnDq6++CsDff/+d6ePXr19P27ZtGTp0KD4+Pri5uWEymVLNd5kyX+jdLl68mKoekGbdS5cumSVm69Spw61bt/j111/ZvXu3keysU6cOu3btYufOndjb21O9enWLr6d69epGIrR79+6cPXv2nse4u7vzySef8Ouvv7JixQrKlSvHiBEjjF6quXLlok+fPnz11Vfs3r2bSZMmceTIEfr27Wu2sJOIiIiISE6mJKiIiIg8EhcuXEizPGXY+t29Mi1hZWWVai7RVatWkZiYaFbm5eXFzp07uXLlilGWmJjI119/naoewNq1a83Kt27dyrVr1/D29jbKatSoQd68efn000+5du2akQStW7cuu3fvZseOHdSuXTtVfPeSkgiNjIzE39/fokQo3B7eX716dWMu0yNHjqSqU6BAAZ5//nnat2/P2bNniY6OzlRsIiIiIiJPKi2MJCIiIo9E3759eeaZZ/Dz86NMmTIkJCRw8OBBPvvsMwoVKkTHjh0z3WaDBg1YvXo15cqVw2Qy8euvv7Js2TIKFChgVq9///5s3bqVHj160L9/f/LkycPixYtTzUNavnx5XnrpJRYsWADcXhTp+PHjzJw5k0qVKpktOGRnZ0ft2rXZsWMHJpPJWIne29ubGzducPDgQYYPH57pa4L/JUJ79+6Nv78/ixYtonjx4qnqLV26lIiICBo0aMAzzzzDrVu3WLp0Kba2tkbCNjAwkOeeew5XV1ecnJw4efIky5Yto3r16hQsWPC+4hMRERERedIoCSoiIiKPRGBgIN999x3z58/n4sWLxMfHU7RoUVq2bElgYCAuLi6ZbnPMmDHY2NgQEhJCbGwsNWrUYP78+fTv39+s3nPPPceCBQuYNGkSI0aMoGDBgrRp04YWLVowduxYs7rjxo2jVKlSrFq1ikWLFlGwYEGef/558y5swwAAXmtJREFUXn/9dXLnzm1W19vbmx07dlC3bl2jrFChQlSsWJG//vrL4vlA03JnIrR79+58/vnnqepUrlyZHTt2MHPmTC5fvky+fPmoXLkyoaGhVKpUCQBPT082b97M8uXLuXHjBi4uLjRp0oRBgwbdd2wiIiIiIk8aq2RNBiUiIiIiFkpISOSnnUezOwyRLFXdrRR589qYTZkhD8bZ2RlA91TM6L2QtOi9kLQUK1YMKyurLG1Tc4KKiIiIiIiIiIhIjqYkqIiIiIiIiIiIiORoSoKKiIiIiIiIiIhIjqaFkURERETEYjY21lR3K5XdYYhkKQeH3CQmJmR3GCIiIvIQKQkqIiIiIpmSN69NdocgjxFbW1sA4uPjszmS+5eYmEBCgpKgIiIiOZmSoCIiIiKSKVq9Ve6kVX1FRETkSaA5QUVERERERERERCRHUxJUREREREREREREcjQNhxcRERGRTEkZ/iwC/5sT9FG9FwkJCURHRz+Sc4mIiEjOoSSoiIiIiGRKUpJVdocgj5HY2JQFhR7+e2Fnp3++iIiIyP3RbxEiIiIiYrGkpGTOno3M7jDkKVW8uCPWmtBLRERE7oN+hRAREREREREREZEcTUlQERERERERERERydGUBBUREREREREREZEcTUlQERHJccLDwzGZTJhMJg4cOJBqf2xsLO7u7phMJkaOHJkNEd7byJEjcXd3Nyvz8/MjKCjogdsOCQnBZDLh7++fqeNu3LjBxx9/TOvWralZsya1atWiefPmvPnmmxw+fPiB47ofKc/69OnTj/zcp0+fxmQyERERcd9tREREGO9qRj9+fn4Wt/Ug8YiIiIiI5FRaGElERHKsfPnysWrVKqpXr25W/u233xITE4OdnV02RZa9Vq9eDcCePXv4999/KV269D2PSUxMpE+fPhw9epQ+ffpQrVo1EhMT+eeff9i8eTOHDx+mUqVKDzv0VBo2bMjy5ctxcXF55OfOClWrVmX58uVmZb169cLDw8Ms4f20vqsiIiIiIllFSVAREcmxWrZsyYYNG3jrrbfIkyePUR4eHo6fnx87d+7Mxuiyx549ezhx4gSNGjVi27ZtrFq1itdee82i4/bt28fkyZNp27atUd6gQQN69uxJUlJSlsSXlJREYmIitra2FtV3dnbG2dk5S85tqYSEBKysrLKkLQcHB2rUqGFWZmNjg7Ozc6pyERERERG5fxoOLyIiOVarVq2IiYnh22+/Ncr+++8/du3aRfv27VPVj42NZeLEibRu3ZpatWrh5eVF165d2bVrl1m99IYdpwyPDg8PN8r+/fdfhg4diq+vL66urvj6+tKnTx9OnjyZxVdrmVWrVmFlZcXYsWOpWLEia9assSiBGRkZCUCRIkXS3G9t/b9fKUaOHJnm8O3Zs2djMpnMykwmExMmTGDhwoU0bdoUV1dXdu/ejbe3N6NGjUrVxn///UflypWZN28ekHo4fP/+/WncuDHJycmpjm3WrBl9+/Y1tq9cucKYMWPw8fHB1dWVZs2aERISQmJiolEn5ZnOnz+fmTNn0qBBA1xdXfnvv//SvA8P43lfuXKFd955h5YtW1KjRg18fX3p27cvhw4dsuj47du3061bN2rWrEmNGjXo0aMHBw8evO94RERERESeROoJKiIiOZbj/7V353E1pv//wF+nlZTShhIq46CkhaLQYomkTNlJtk9DGmGYLNnHmNCQyjahYXzItFkmjH2+lokZy4yZsYxEGKNFhahOnd8ffuf+ONo3kdfz8eiPc5/ruq/3dZ9L6t21aGnBxcUF8fHxGDx4MIBXS8F1dHTQq1evEuULCgqQm5sLPz8/6Ovr4+XLlzh+/DgmTJiAbdu2wd7evsox/Oc//0GTJk0wf/586OnpITMzExcvXsSzZ89q3D8AQrLxxIkTFZZ9/vw5jhw5gu7du8PQ0BBeXl4ICQnBmTNn0Lt373LrmpmZQVlZGcuXL8enn34KOzs76Ojo1EofDh8+jObNm+Ozzz6DmpoaTExM4O7ujri4OCxcuBBqampC2cTERIhEIrnZqK/z8vJCQEAALl68CFtbW+H6r7/+irt372LGjBkAgBcvXsDHxwf//vsvAgMDYWxsjDNnzmDdunVIS0vDihUr5O4bHR0NsViMRYsWAQA0NTWhoaGBGzduyJWri887OzsbioqKmD59OnR0dPD06VMkJiZi5MiRSEhIQLt27cqsGxsbi+DgYAwaNAgTJ05EcXExtm/fDh8fH+zdu7dEUpqIiIiIqKFiEpSIiBo0b29vTJkyBQ8fPoSBgQESEhLg6ekJRUXFEmU1NDSwcuVK4XVRUREcHByQlpaGXbt2VTkJmpWVhdTUVERGRqJv377C9QEDBlS/Q28orR9lSUpKQl5enjAL1tPTE6GhoYiLi6swCWpkZISlS5dixYoVwvJ5IyMjODg4YNSoUTXaD7SgoADR0dFQV1cXrnl7e2Pnzp04cuQIPv74Y+F6YmIiHBwc0Lx581Lv5eTkhGbNmiEhIUEuCZqQkABNTU3hc0hISMDff/+NTZs2wdnZGQDQs2dPFBcX49tvv8XEiRNhamoq1G/SpAk2b95c7vOuq8/bxMRESL4Cr8Zl79694e7ujr1792L+/Pml1svLy0NISAhcXV0RGhoqXHdwcICrqysiIyOxfv36GsVGRERERPS+4HJ4IiJq0Hr27AkdHR0kJCQIBwGVthRe5siRIxg1ahTs7OzQqVMnmJmZ4dy5c0hJSaly282aNUPr1q2xevVq7N69G7dv365JV0p19OhRueX+5YmLi4O6ujr69+8PANDR0YGjoyOOHz+OJ0+eVFjf29sbp06dQlhYGHx8fNC0aVPExMTA29sbBw8erHYfunfvLpcABYCOHTuiQ4cOclsLXLp0CampqXJJ0TcpKyvD3d0dhw8fRl5eHgDg5cuXOHz4MNzc3IQDhpKTk6GhoSEkQGWGDBkCALhw4YLcdRcXlwoTznX1eUulUnz//ffw8vJC165d0alTJ5ibmyM1NbXccXnlyhXk5ubCw8MDEolE+FJWVoatrS0uXrxYK/EREREREb0PmAQlIqIGTVFREZ6enkhISBBOin99ht/rDh06hOnTp6Nly5b48ssvsXfvXsTGxqJXr17Iz8+vctsikQjR0dGwtrZGeHg43Nzc0LNnT6xduxYFBQU17VqVpKSk4PLly3BxcRGW/efm5sLV1RWFhYU4cOBApe7TtGlTDBgwAMHBwYiPj8d///tfqKiolFg+XhVl7TP68ccf4+LFi8J+n2/O5iyLt7c38vLycOTIEQCvEsVPnz6VS37n5ORAV1e3zFhke6DKVOb0+br6vLdu3Yrg4GBYWloiNDQU33//PWJjY9GhQ4dyx2VGRgYAwN/fH2ZmZnJfBw4cqFTim4iIiIiooeByeCIiavC8vLwQFRWFBw8eYPHixWWWO3jwIIyMjBAaGip3+rdsRqGMqqoqAJRIbJWWVDI0NBSW2N+5cwf79+/Hxo0boaCggMDAwGr3qapiY2MBAPv378f+/ftLvB8XF4dx48ZV+b7W1tZwcHDA0aNHkZmZCR0dHaioqJSa9Csr6VbWSeseHh5Ys2YNEhMT8Z///AeHDh3C4MGDhdmcZXl9FunHH3+MxMREtGvXDhYWFkIZLS0tXLt2rUTd9PR04f3qqIvP++DBg7Czs5NbEg+8StQ2bdq0zHrNmjUDACxevBidO3euVttERERERA0Fk6BERNTgmZqaYtSoUXj06BHc3NzKLCcSiaCkpCSXlLt16xauXLmCFi1aCNcMDQ0BADdu3JA7YKmiw4mMjY0RGBiIffv24fr169XtTpVJJBLs378fJiYmWLJkSYn3ExMTER8fjz/++ANmZmal3uPJkydQV1eHsrKy3PWioiLcvXsXjRo1goaGBgCgVatWyMzMREZGhjDbsqCgAGfOnKlS3Nra2ujVqxcSExPRtm3bErM5y/Pxxx/jq6++wq+//opz585h9uzZcu93794dhw4dwsmTJ+WWxO/fvx8ikQh2dnZVirU0tfV5i0SiEs/9p59+wqNHj9C6desy61lbW0NdXR0pKSkYPXp0tdsnIiIiImoImAQlIqIPQmnJvzc5Ojri6NGjWLp0Kfr374+0tDSEh4fDwMAAxcXFQjk9PT306NEDW7ZsgaamJgwMDHD+/PkSe3Nev34dX3zxBQYOHIjWrVtDSUkJp06dwoMHD/Cf//ynVvpVmdPhT58+jfT0dEyePLnU5F7z5s0RHx+PuLi4MpOgycnJWLlyJQYPHgwrKytoamri8ePHiI2Nxc2bNzF16lRhhubAgQOxfv16zJw5E5MnT0Z+fj527tyJoqKiKvdPdtr76tWr8dFHH1V6RqNsFumsWbMgEong4eEh9/6QIUOwa9cuzJkzB4GBgTAxMcGZM2cQHR2N4cOHw8TEpMqx1tXn7ejoiE2bNiE8PBxdu3bFjRs3sGXLljIPh5KRnVIfHByMnJwc9OvXD1paWsjIyMC1a9egoqKCGTNmVDsuIiIiIqL3CZOgRERE/9/QoUORkZEh7AVqYmKC4OBgnDx5ssRBOatXr8by5cuxZs0aFBUVwdnZGaGhofD29hbK6OnpwcjICDt37sS///4LkUiENm3aYNmyZRgxYsRb61d8fDxUVFSEQ3/e1LZtW9ja2uLgwYMICgoSlvu/ztLSEkOGDEFycjISEhKQnZ2Nxo0bQywWIyQkRO7eRkZGiIyMxNq1azF9+nTo6elhwoQJyMrKQkRERJVid3Jygra2Nh49elSl5fra2tro3bs3jh8/DicnpxL7jjZq1Ajffvstvv76a2zatAk5OTkwMDDArFmzMHny5CrFKFNXn7e/vz/y8vIQExODqKgodOzYEevWrUN4eHiFdb29vWFgYICoqCgsWLAAL1++hK6uLszNzTFy5Mhqx0RERERE9L4RSaVSaX0HQURERETvh6KiYqSmZtR3GPSBMjDQgoKCFFlZWfUdClVAW1sbAPhZkRyOCyoNxwWVpkWLFmWeHVBdPB2eiIiIiIiIiIiIGjQmQYmIiIiIiIiIiKhBYxKUiIiIiIiIiIiIGjQejERERERElaagIIKBgVZ9h0EfKBUVJUgkhfUdBhEREb2HmAQlIiIioipRUOC5mvQ/ysrKAIDCwrpPTkokhZBIJHXeDhERETU8TIISERERUZXw9FZ6HU/1JSIiovcB9wQlIiIiIiIiIiKiBo1JUCIiIiIiIiIiImrQuByeiIiIiKpEtvyZCPjfnqB1NS4kEglyc3Pr5N5ERET04WASlIiIiIgqrbi4GKmpqfUdBn0gjIyMoKTEX1mIiIio5vgTBRERERFVWn5+PlavXl3fYdAHYs6cOWjbtm19h0FEREQNAPcEJSIiIiIiIiIiogaNSVAiIiIiIiIiIiJq0JgEJSIiIiIiIiIiogaNSVAiIiIiIiIiIiJq0JgEJSIiordOLBZX6uv+/fuIj4+HWCzGX3/9VacxydoRi8W4evVqiffz8/PRtWtXiMVizJ07t1bbTk9PR3h4eKl99PHxgY+PT622R0RERET0oeHp8ERERPTWxcTEyL1es2YNUlNTERERIXddX1//bYYFAGjSpAni4uLQpUsXuetHjx7Fy5cvoaKiUuttZmRkICIiAoaGhujYsaPce4sXL6719oiIiIiIPjRMghIREdFbZ2lpKfe6adOmUFFRKXG9PgwcOBBJSUmYP38+GjVqJFyPj4+Hi4sLzp07V2ttSSQSiESicsu0a9eu1tojIiIiIvpQcTk8ERERvReePXuGRYsWwc7ODnZ2dggICMC///5bolxiYiK8vb3RpUsX2NjYwN/fH6mpqZVux93dHS9fvsTRo0eFa//88w/Onz8PLy+vEuXz8/OxcuVKDB48GNbW1rCzs8OYMWNw/vx5uXL379+HWCzG1q1bERYWBkdHR5ibm2Pfvn0YMmQIAGDevHnCkvz4+HgAJZfDV7W96OhoREVFwcXFBVZWVhgxYgSuXLlS6edBRERERNQQcCYoERERvRfmzZsHZ2dnhIaG4p9//sHq1asxZ84c7NixQyizfv16bNy4ESNHjkRgYCCePXuGDRs2YNSoUdi3b1+lltdraWnBxcUF8fHxGDx4MAAgISEBOjo66NWrV4nyBQUFyM3NhZ+fH/T19fHy5UscP34cEyZMwLZt22Bvby9XPjo6GmKxGIsWLQIA2NraQkFBAUFBQZg6dSqcnJwAAK1bty41vqq2t2PHDrRr1w7z588HAISFhcHPzw/Hjx+HhoZGhc+DiIiIiKghYBKUiIiI3gtOTk5YsGCB8DonJwerV6/G48ePoa+vj4cPH2Lz5s2YMGECPv/8c6GcjY0NXF1dsX37dgQFBVWqLW9vb0yZMgUPHz6EgYEBEhIS4OnpCUVFxRJlNTQ0sHLlSuF1UVERHBwckJaWhl27dpVISjZp0gSbN2+Wu5dYLAbwKvFZ0ZYAVW2vadOm2LRpExQUXi0A0tfXx7Bhw3D69Gm4u7tX6nkQEREREb3vmAQlIiKi90KfPn3kXssShw8fPoS+vj7Onj0LiUQCDw8PSCQSoZyOjg46deqEixcvVrqtnj17QkdHBwkJCbC1tcW9e/dKXQovc+TIEURHRyMlJQXZ2dnCdRMTkxJlXVxcSk2mVkVV2nNychISoADQoUMHAK+eGxERERHRh4JJUCIiInovaGlpyb2WndKen58P4NUJ6wDg6elZav1WrVpVui1FRUV4enoiISEBaWlp6NKlC0xNTUste+jQIcyYMQODBg3C5MmToaurCwUFBYSFhSElJaVE+ZqeeF/V9ip6bkREREREHwImQYmIiKhBaNasGQAgMjISzZs3L/G+LPlXWV5eXoiKisKDBw+wePHiMssdPHgQRkZGCA0NlTvpPS8vr0rtVdbbbo+IiIiIqCFgEpSIiIgaBAcHBygqKuL+/fvo27dvje9namqKUaNG4dGjR3BzcyuznEgkgpKSklxC8tatW7hy5QpatGhRqbZkCdqXL19WWLY22iMiIiIi+tAwCUpEREQNgpGREfz9/bFmzRqkpaXB3t4e6urqSE9Px6VLl2BiYoKxY8dW6Z5LliypsIyjoyOOHj2KpUuXon///khLS0N4eDgMDAxQXFxcqXZatWqFxo0b4+DBg/joo4/QqFEjtGrVSpjdWtvtERERERF9aJgEJSIiogYjICAApqam+O677xAfH4+ioiLo6+vD0tISFhYWddLm0KFDkZGRgb179yI2NhYmJiYIDg7GyZMnceHChUrdQ1VVFStWrMD69esxfvx4SCQSrFy5stTDmGqjPSIiIiKiD41IKpVK6zsIIiIiIno/5OXlYfr06fUdBn0g5syZg7Zt2yIrK6u+Q6Eq0tbWBgB+diSH44JKw3FBpWnRooXc9k+1QaFW70ZERERERERERET0jmESlIiIiIiIiIiIiBo07glKRERERJWmqqqKOXPm1HcY9IEwMjKq7xCIiIiogWASlIiIiIgqTUFBAW3btq3vMOgdoqysDAAoLCysk/tLJJI6uS8RERF9WJgEJSIiIqIq4cEF9DoeaEFERETvA+4JSkRERERERERERA0ak6BERERERERERETUoHE5PBERERFViWz5MxHwvz1B62JcSCQS5Obm1vp9iYiI6MPDJCgRERERVZpUKkVGRkZ9h0EfAF1dXSgp8dcVIiIiqh38qYKIiIiIKq2wsBD79++v7zDoA+Dh4QFdXd36DoOIiIgaCO4JSkRERERERERERA0ak6BERERERERERETUoDEJSkRERERERERERA0ak6BERERERERERETUoDEJSkRE77xNmzZBLBbDx8enUuXDw8MhFosr/Jo7dy4AwMXFBf7+/nXZhffC3Llz4eLiUmG5goIC7Nq1C15eXujatSusrKzQp08fTJ8+HRcuXKhW2y4uLsLn8bb5+PjIja3k5GSIxWIkJydXuW5Znj9/jo0bN2Lw4MGwsrKCtbU1XF1dMXv2bFy/fr1G8RMRERERUcV4OjwREb3zEhISAAAXL17EvXv30Lp163LLDxs2DL169RJep6amIigoCFOnToWTk5NwXVtbu07ibejmzJmD48ePY8KECQgMDISSkhJSU1Nx8uRJ/PLLL7C1ta3vEGvEzMwMMTExaNeuXa3cr6ioCJMmTcLff/+NSZMmoXPnzigqKsKdO3fw448/4vr16+jQoUOttEVERERERKVjEpSIiN5pFy9eRGpqKpydnXHy5EnExcVh5syZ5dZp0aIFWrRoIbxWVVUFALRu3RqWlpZ1GW6Dl5aWhsOHDyMwMFBu9qyDgwPGjBmD4uLieoyudqirq9fqOLl48SIuX76MVatWwdPTU7ju6OiI8ePHN4hnRkRERET0ruNyeCIieqfFxcVBJBJh4cKF+Oijj5CYmFhnSaNTp05hyJAhsLCwwIABAxAbG1uizF9//QU/Pz/Y2NjAwsIC3t7eOHr0qFwZ2XL8N8XHx0MsFuP+/fvCtfPnz8PHxwe2trawsLCAi4uLXJI3Pz8fK1euxODBg2FtbQ07OzuMGTMG58+fl7v3/fv3IRaLER0djaioKLi4uMDKygojRozAlStXSsQSGxsLV1dXmJubw9XVtdS+liY7OxsAoKenV+r7Cgr/+9FC1t9z587hs88+g42NDaytrTFr1ixkZmZW2NajR48wb9489OzZU4gzOjparoxs6XpSUhLWrFmDnj17wtraGuPHj0dKSopcWalUii1btsDZ2RmdO3eGp6cnTp48WaLd0pbDV7ZuaaryzMrakqC0MSUWi7FixQrEx8djwIAB6NKlCzw8PErElZWVheDgYDg6OsLc3Bw9evTA2LFjcfXq1UrFT0RERETUEHAmKBERvbOeP3+OI0eOoHv37jA0NISXlxdCQkJw5swZ9O7du1bb+vPPP7F69Wr4+flBR0cHe/fuxYIFC9CmTRt069YNAHDr1i2MHj0aLVu2xNKlS6Gmpoa9e/ciICAAISEhGDJkSJXaTEtLg5+fH/r27YvJkydDVVUVjx49wqlTp4QyBQUFyM3NhZ+fH/T19fHy5UthKfq2bdtgb28vd88dO3agXbt2mD9/PgAgLCwMfn5+OH78ODQ0NAAA33//PYKDg9GvXz/MmzcPOTk5CA8Ph0QikUvIlcbU1BQaGhpYv349RCIRHBwc0LJly3LrzJ8/H05OTli3bh1SU1Oxdu1a3L59G7GxsVBWVi61zr///ouhQ4eiSZMmmDlzJlq2bImzZ89i1apVyM7OxowZM+TKr1mzBl27dsWKFSvw9OlTrFmzBlOnTkVSUhIUFRWFZ7Fx40YMHz4c/fr1wz///IOFCxcCAIyNjcvtQ03qmpmZQVlZGcuXL8enn34KOzs76OjolFunso4fP46rV68iMDAQampqiIqKQkBAAA4fPgwjIyMAr7YvuHfvHmbOnAlDQ0NkZ2fj6tWrQnKWiIiIiOhDwCQoERG9s5KSkpCXlwcvLy8AgKenJ0JDQxEXF1frSdDs7GzExMSgefPmAIBu3bohOTkZBw4cEJKgkZGRkEql2LFjB3R1dQEAzs7O8Pb2RmhoKDw8PCpMIr7ujz/+QEFBAZYtWyYkKAHIJVM1NDSwcuVK4XVRUREcHByQlpaGXbt2lUiCNm3aFJs2bRLi0NfXx7Bhw3D69Gm4u7ujuLgYYWFhsLCwQHh4OEQiEQDA2toaAwYMEPpfFjU1NYSGhmLevHlYsGCB0IaDgwO8vb2FZ/U6a2trLFmyBADQq1cv6OjoYObMmThy5Ajc3d1LbSciIgIvX75EfHw89PX1AQD29vYoLCzE1q1bMX78eGhpaQnlxWIxVq1aJbxWVFTEjBkz8Ntvv8HKygo5OTnYtm0bBgwYgOXLlwvlTE1NMWbMmHITmTWpCwBGRkZYunQpVqxYIczyNTIygoODA0aNGlWj/UALCwsRHR0NNTU1AK8Srr169UJSUhI++eQTAMClS5cwY8YMuXHVr1+/ardJRERERPQ+4nJ4IiJ6Z8XFxUFdXR39+/cHAOjo6MDR0RHHjx/HkydParWtTp06ySUAVVVV0bZtWzx48EC4lpycDHt7eyEBCgAikQgeHh54/Pgx7ty5U+U2lZWVMX36dCQlJeHff/8ttdyRI0cwatQo2NnZoVOnTjAzM8O5c+dKLPcGACcnJ7lErCzB9vDhQwDAnTt3kJ6eDnd3dyEBCrxKyllZWVUqbtlnsHnzZkycOBGtWrXC/v37MXbsWERFRZUo7+bmJve6f//+UFJSKvck+dOnT6NHjx7Q1taGRCIRvhwdHVFQUFBiKfebS8hlS8dl/b5y5Qry8/MxePBguXJdu3aFgYFBuf2tSV0Zb29vnDp1CmFhYfDx8UHTpk0RExMDb29vHDx4sFL3KE337t2FBCgA6OrqQkdHR27cWlhY4JtvvsG2bdvw119/oaioqNrtERERERG9r5gEJSKid1JKSgouX74MFxcXYUl4bm4uXF1dUVhYiAMHDtRqe6/PKpRRUVFBQUGB8DonJ0cuASoj2+uxqsuLW7dujejoaKioqGD+/Pno3bs33N3dkZiYKJQ5dOgQpk+fjpYtW+LLL7/E3r17ERsbi169eiE/P7/CfqioqACAUFaWPC5tf8qy9qwsTePGjeHk5ISgoCDs3r0bBw4cgJ6eHtatW4fc3Nxy76ukpAQtLa1yn1dmZiZ+/PFHmJmZyX1NnDhRrh8yzZo1k3v9Zr/L25ezon7XpO7rmjZtigEDBiA4OBjx8fH473//CxUVFaxYsaLS93hTZcbt2rVrMWDAAOzYsQNDhgyBvb09li5dWuJzIiIiIiJqyLgcnoiI3kmyg3r279+P/fv3l3g/Li4O48aNe6sxaWlpISMjo8T19PR04X3gf6fRFxQUCMk4oGTiDng1m7Br164oKirCtWvXsHXrVgQFBaF58+bo0aMHDh48CCMjI4SGhsrN3MzLy6tWH2TJQlnMpfWjOkxNTTFo0CBER0cjNTUVFhYWZd5XIpEgOzu71ATe63F26tQJn376aanvt2rVqkrxydoqq9/l3a8mdctjbW0NBwcHHD16FJmZmdDR0SmRwJSpycxnbW1tBAcHIzg4GP/88w+OHDmC0NBQPH/+XG4LASIiIiKihowzQYmI6J0jkUiwf/9+mJiYYMeOHSW+vLy8cP36dfzxxx9vNa7u3bvj3LlzcolQqVSKAwcOoHnz5jAxMQEAGBoaAgCuX78uV7+808QVFRXRpUsXBAUFAQBu3LgB4NVyeyUlJbkE6K1bt0o98b0yjI2NoaenV2IJdlpaGi5fvlxh/WfPnpWZgJUtz39zdmRSUpLc6x9//BESiQS2trZlttOrVy/cvHkTxsbG6Ny5c4mvN2d+VsTS0hKqqqolZhD/8ssvwpL5uqgLvEpgFhYWlrheVFSEu3fvolGjRsKesK1atUJmZqbcGCsoKMCZM2cqbKcyWrZsifHjx8PS0rLE+CQiIiIiasg4E5SIiN45p0+fRnp6OiZPngw7O7sS7zdv3hzx8fGIi4uDmZnZW4tr2rRpOHnyJHx9feHv7w81NTXExMTg2rVrWLNmjZCodHR0hJaWFhYsWIDAwEAoKioiISEB//zzj9z9du/ejeTkZDg6OqJly5Z48eIFdu/eDWVlZXTv3l2419GjR7F06VL0798faWlpCA8Ph4GBAYqLi6vcBwUFBQQGBiI4OBgBAQEYOnQocnNzsX79+kot7b5z5w4mT56MwYMHo2vXrtDV1cWTJ0+QlJSEn376CR4eHiVOi7906RKWLl2KPn364M6dO1i7di06dOgAV1fXMtuZMWMGzp49i1GjRsHHxwdt2rRBXl4e7t69i5MnT2Lbtm3Cqe+VoampiYkTJ2Ljxo1YuHAh+vfvj4cPHyIiIqLCftekLvBqL9mVK1di8ODBsLKygqamJh4/fozY2FjcvHkTU6dOFWYMDxw4EOvXr8fMmTMxefJk5OfnY+fOndXex/Pp06fw9fWFu7s7TExM0LhxY/z666+4dOkSJkyYUK17EhERERG9j5gEJSKid058fDxUVFTkTrN+Xdu2bWFra4uDBw8iKChIWH5e10xNTbF79258/fXXWLhwIQoLCyEWixEZGYm+ffsK5dTV1fHNN9/gyy+/xJw5c6ChoYFhw4ahV69eCA4OFsp17NgRZ86cQVhYGDIzM9GkSRN07NgRUVFRwoFGQ4cORUZGhrAXqImJCYKDg3Hy5MlyDxYqz7BhwwAAUVFRCAgIgKGhIaZOnYqLFy9WeM82bdrA19cXP//8M44cOYInT55AWVkZpqammD9/PsaMGVOizsqVKxEbG4vAwEAUFxfDyckJCxYsgLKycpntNG/eHHFxcYiMjMTGjRuRnp4OdXV1tG3bFr169ZI7/KmyAgMDoaamhv/+979ISEiAiYkJli1bhm3bttVpXUtLSwwZMgTJyclISEhAdnY2GjduDLFYjJCQELlxbmRkhMjISKxduxbTp0+Hnp4eJkyYgKysLERERFS5z6qqqujcuTMSEhLw4MEDFBcXw9DQEIGBgZg0aVKV70dERERE9L4SSaVSaX0HQURERA1PfHw85s2bh8TERHTs2LG+w6Fakp+fX6nkL1FNeXh4QFdXF1lZWfUdClWDtrY2APDzIzkcF1QajgsqTYsWLeS2BKsN3BOUiIiIiIiIiIiIGjQmQYmIiIiIiIiIiKhB43J4IiIiIqq04uLiEod8EdUFXV1dAFwe+b7i8lYqDccFlYbjgkpTF8vheTASEREREVWaSCQSklNEAIRDzgoLC2v93hKJpNbvSURERB8mJkGJiIiIqEo4U4Nexxk8RERE9D7gnqBERERERERERETUoDEJSkRERERERERERA0ak6BERERERERERETUoHFPUCIiIiKqEtkekETA/w5Gquy4kEgkyM3NrcuQiIiIiEpgEpSIiIiIKk0qlSIvL6++w6D3lLq6OpSU+CsIERERvX38CYSIiIiIKq2oqAiXL1+u7zDoPWVlZQU1NbX6DoOIiIg+QNwTlIiIiIiIiIiIiBo0JkGJiIiIiIiIiIioQWMSlIiIiIiIiIiIiBo0JkGJiIjeEfHx8RCLxcKXhYUFBg4ciDVr1uDp06eVuoeLiwvmzp1b6bbu379f07Dfusr28fVn2aFDB/Tu3RtTpkzBH3/8UavxzJ07Fy4uLnV6z/v370MsFiM+Pr5W2yEiIiIi+lDwYCQiIqJ3TEhICNq2bYsXL17gp59+QlRUFJKTkxETEwMFhfL/fhkREQF1dfW3FOm7z83NDb6+vigqKkJKSgrCw8MxYsQIxMTEwMzMrFba8Pf3x7hx42rlXmXR19dHTEwMWrduXaftEBERERE1VEyCEhERvWPEYjE6duwIAOjRoweysrKQmJiIy5cvw8bGptQ6BQUFUFFRQadOnd5mqO88XV1dWFpaAgBsbGxgZGQEX19f7N69G1988UWpdWTPsrLeRmJSRUVF6AcREREREVUdl8MTERG94ywsLAAADx8+BAD4+PjA09MT586dg7e3Nzp37ozNmzcDKH2p+OXLlzFy5Eh07twZPXv2xKpVq1BYWFhqW4mJifD29kaXLl1gY2MDf39/pKamVhjj3bt3MXfuXPTt2xcWFhZwdHTE9OnTcffuXblysmX4Fy9exKJFi2BnZwc7OzsEBATg33//lStbUFCAkJAQODg4wMLCAiNHjsSVK1cq88jK1KVLFwD/e5ayeM6ePYvZs2fD1tYWAwYMqNLzeHPpenJyMsRiMZKTk+XKlbWkPTY2Fq6urjA3N4erqytiY2NLxF1W3cuXL+OTTz6Bra0tLCws4OrqirVr15YoM3nyZHTt2hUWFhYYPnw4zpw5U8knRkRERETUMHAmKBER0TtOtm+ntra2cO3Ro0dYsGABpkyZgjZt2qBJkyal1r158ybGjx+P1q1bIyQkBKqqqti1axeSkpJKlF2/fj02btyIkSNHIjAwEM+ePcOGDRswatQo7Nu3D/r6+mXG+PjxY2hrayMoKAhaWlrIyMjA7t27MXz4cCQlJUFHR0eu/Lx58+Ds7IzQ0FD8888/WL16NebMmYMdO3YIZRYsWICDBw9i0qRJ6NGjB27evImAgAC8ePGiSs/vdWlpaQCAZs2alYinf//+WLt2LV6+fFnj51FZ33//PYKDg9GvXz/MmzcPOTk5CA8Ph0QiqXDrg9OnT8Pf3x9isRgLFiyAvr4+0tLScPnyZaHMmTNnMGXKFNja2uLLL7+Eqqoq9u7dCz8/P2zZsgU9e/ascR+IiIiIiN4HTIISERG9Y4qKiiCRSPDixQucOXMGe/bsQfPmzdG1a1ehTHZ2NjZs2FDm8niZDRs2QEFBAd9++62QRHVycoKbm5tcuYcPH2Lz5s2YMGECPv/8c+G6jY0NXF1dsX37dgQFBZXZTrdu3dCtWze5Pjg5OcHe3h4HDx6Er6+vXHknJycsWLBAeJ2Tk4PVq1fj8ePH0NfXx+3bt7F//35MmjQJs2fPBgA4ODigWbNm5cbxJqlUColEguLiYqSkpGDJkiUAgEGDBsmV6927N4KDg2vteVRGcXExwsLCYGFhgfDwcIhEIgCAtbU1BgwYgObNm5fbr2XLlsHIyAh79uwRlu/36NEDw4cPF8otX74cnTp1QlRUlJBU7d27N7y9vbF27VomQYmIiIjog8Hl8ERERO8Yb29vmJmZoWvXrpgxYwY++ugjREVFQVVVVSijra1dYQIUAC5cuAB7e3u5WaSKioolkoBnz56FRCKBh4cHJBKJ8KWjo4NOnTrh4sWL5bZTUFCAqKgouLu7w8rKCp06dYKlpSXy8vKQkpJSonyfPn3kXovFYgD/W6YuW0ru4eEhV87d3R2KiooV9ltm586dMDMzQ+fOneHp6Ym7d+9i0aJFJU5z79evn9zrmj6Pyrhz5w7S09Ph7u4uJEABwMjICFZWVhXWvX//PoYOHVrm/qV3795Famoq3N3dUVxcLPShqKgIvXr1wh9//IHnz5/XuB9ERERERO8DzgQlIiJ6x6xZswZt27aFkpISmjdvLpfAlNHT06vUvbKzs6Grq1th/YyMDACAp6dnqfdp1apVue2sXLkSMTEx8PPzQ7du3aChoQGRSAQ/Pz/k5+eXKK+lpSX3WpbIk5XNzs4GgBKxKykplVjKXh53d3eMHz8eIpEIGhoaMDIyKnWZeW0/j8p48uRJqW3LrskSwuXVbdGiRZllZH1YsWIFVqxYUWqZnJycMrdSICIiIiJqSJgEJSIiese0a9dOOB2+LK/PHCyPbH/ON6Wnp8u9liUWIyMjS12GXdFp6QcPHoSnpydmzJghXCsoKEBOTk6l4nyTLEmakZEhlwiVSCRCArAytLW10blz5wrLvfk8a/I8ZDN2CwoK5K6/GbesjTc/i7KuvU6WGH/06FGZZWT39/f3LzHzVaa0BDkRERERUUPEJCgREVEDZmdnh1OnTiErK0tInBUVFeGHH36QK+fg4ABFRUXcv38fffv2rXI7IpEIysrKctfi4uJQVFRU7bgBYP/+/ejQoYNw/eDBg9W+Z1XU5HkYGhoCAG7cuIFevXoJ10+cOCFXztjYGHp6eiX2TJUdblTenqBt27aFkZER4uLi4OvrW+LZy+5vZGSEGzduIDAwsEp9ICIiIiJqaJgEJSIiasCmTp2KEydOwNfXF1OnTkWjRo3w3XfflViibmRkBH9/f6xZswZpaWmwt7eHuro60tPTcenSJZiYmGDs2LFltuPo6IiEhASYmJhALBbj119/xZ49e9C0adNqxW1qagoPDw9s374dCgoKwunwW7duhbq6erXuWRU1eR56enro0aMHtmzZAk1NTRgYGOD8+fM4evSoXDkFBQUEBgYiODgYAQEBGDp0KHJzc7F+/foKtzsQiURYuHAh/P39MXLkSIwbNw4tWrTA/fv38euvv+LLL7+ESCTCkiVLMGXKFHzyySfw9PSEnp4enjx5ghs3biA9PR3Lli2rtWdGRERERPQuYxKUiIioAWvfvj22b9+OkJAQBAUFQVNTEx4eHhgwYAAWLlwoVzYgIACmpqb47rvvEB8fj6KiIujr68PS0hIWFhblthMcHAxFRUVs2rQJ+fn5sLS0xNatWzF16tRqx75ixQro6uoiPj4eO3bsQMeOHREREYFZs2ZV+55VUZPnsXr1aixfvhxr1qxBUVERnJ2dERoaCm9vb7lyw4YNAwBERUUhICAAhoaGmDp1Ki5evIgLFy6U24ajoyN27NiBDRs2YNmyZSgsLETLli3h5uYmlOnZsyf27NmDTZs2YdmyZXj27BmaNWuGDh064OOPP67mkyEiIiIiev+IpFKptL6DICIiInpfBQQE4OHDh4iPj6/vUN6KwsJC/N///V99h0HvKSsrK6ipqSErK6u+Q6E6JtuChZ81vY7jgkrDcUGladGiRaXPQaisksejEhEREVGFHj16hEOHDuHnn3+GpaVlfYdDRERERETl4HJ4IiIiomr4/vvv8e2338LW1hYBAQH1HQ4REREREZWDy+GJiIiIqNKKi4uRk5NT32HQe0pdXR3FxcVc8vgB4PJWKg3HBZWG44JKUxfL4TkTlIiIiIgqTSQSQU1Nrb7DoHeIsrIygFf7xVakuLgYEomkrkMiIiIiKoFJUCIiIiKqEs7UoNdxBg8RERG9D3gwEhERERERERERETVoTIISERERERERERFRg8YkKBERERERERERETVo3BOUiIiIiKpEtgckvZ8kEglyc3PrOwwiIiKit4pJUCIiIiKqkqKiovoOgapJVVW1vkMgIiIiqhdMghIRERFRpRUXF+P+/fv1HQZVU6tWraCoqFjfYRARERG9ddwTlIiIiIiIiIiIiBo0JkGJiIiIiIiIiIioQWMSlIiIiIiIiIiIiBo0JkGJPiB//vknPv/8czg7O8Pc3BzW1tYYPnw4Nm/ejKdPn9Z3eGWKj4+HWCyu1h50P/zwA6Kjo2s9poKCAuzatQteXl7o2rUrrKys0KdPH0yfPh0XLlwQyt2+fRvh4eF1tn+eWCxGeHh4ndy7vrm4uGDu3LkVlhOLxcJXhw4d0Lt3b0yZMgV//PHHW4jy3RUdHY0+ffrAzMwMYrG4zHKyf19lfd28ebPcdu7fvw+xWIz4+Pja7kKtGTBgQLkxzp07Fy4uLsLr+uzT3Llz5Z5/165d4eXlhb179/IwIiIiIiKiGuDBSEQfiN27d2P58uVo3749PvnkExgbGyM/Px9XrlzBzp07kZmZifnz59d3mLXuhx9+wPXr1zF+/Phave+cOXNw/PhxTJgwAYGBgVBSUkJqaipOnjyJX375Bba2tgCAO3fuICIiAra2tmjVqlWtxgAAMTExaNGiRa3f933j5uYGX19fFBUVISUlBeHh4RgxYgRiYmJgZmZW3+G9dX/++SdWrlyJUaNGwd3dHUpKFf93HxISgrZt25a43rp16zqI8O355ZdfcOfOHQBAXFwcvLy86jmiiqmpqWH79u0AgCdPniAmJgYLFy7EnTt3EBQUVM/RERERERG9n5gEJfoAXLp0CcuWLYOTkxPWr18PZWVl4b3evXtj0qRJuHz5cj1G+H5JS0vD4cOHERgYCH9/f+G6g4MDxowZg+Li4jqPoaCgACoqKrC0tKzztt4Hurq6wrOwsbGBkZERfH19sXv3bnzxxRf1G1w9+PvvvwEAw4YNq3QSWCwWo2PHjnUZVq0qLi5GUVGR3Pez0sTFxUEkEqF37944ffo0UlNTS032vksUFRXl/m336tULAwcOxO7duzFr1qwK+0xERERERCVxOTzRB2DLli1QUFDAkiVLSv3luUmTJujZs6fwOi8vD1999RWcnJxgbm4OJycnhISE4OXLl3L1xGIxVqxYgV27dqFfv36wsLDAsGHD8Oeff0IikSAsLAy9e/eGjY0Npk2bhqysLLn6Li4u8Pf3xw8//ICBAweic+fOGDBgAPbv31+pfiUmJsLb2xtdunSBjY0N/P39kZqaKrzv4+OD48eP48GDB8LS0teXvObk5GDFihXC9gDOzs5Yt24dCgsLy203OzsbAKCnp1fq+woKr761xsfHY9q0aQCAcePGCTEkJycDACQSCTZs2IB+/frB3NwcPXv2xMKFC/HkyZNSn9PBgwfh7u4Oc3NzHDx4EEDpy+Fv376N6dOnw87ODubm5hg8eDAOHDggV+bFixf46quv4OLigs6dO8POzg7Dhw/HyZMny+373bt3MXfuXPTt2xcWFhZwdHTE9OnTcffuXblysiXWFy9exKJFi2BnZwc7OzsEBATg33//lStbUFCAkJAQODg4wMLCAiNHjsSVK1fKjaMiXbp0AQA8fPhQuLZ7927h+dnZ2WHWrFmlblNw+vRpjB07FlZWVrC0tISvry9+++03uTJz585F165dcePGDfj4+KBLly7o1auXsPXC2bNnhbHp7u6OM2fOyNXPyspCcHAwHB0dYW5ujh49emDs2LG4evVqhX07fPgwvLy8YGFhARsbG3zyySe4fv268L6Pjw/mzJkDAPDy8oJYLK7UtgKV8e+//2L69OmwsrKCjY0Npk+fjvT09FLL7t27F66urjA3N8fAgQORmJhYYtk5AISHh8Pb2xvdunWDjY0Nhg4diqSkpBL3k32/+fbbb4V/M5cuXSo33ry8PBw+fFgYewCqvcT9999/x4wZM+Ds7AwLCwv06dMH8+bNQ0ZGRon+iMVi3L59GzNmzIC1tTXs7e0xb968am87oqSkhE6dOuHFixfC99G//voLfn5+sLGxgYWFBby9vXH06NESdR89eoR58+ahZ8+eMDc3h6ura51sEUJERERE9K7jTFCiBq6oqAg///wzzM3N0bx58wrLFxcXY8qUKbh06RKmTZsGCwsLXL16FRs2bMCNGzewdetWiEQiofyxY8fQqlUrBAUFQSKRICQkBFOnToWDgwMKCwvxxRdf4NGjR1i5ciUWLlyIyMhIufauXbuGGzdu4NNPP4WmpiZiYmIwZ84cKCkpwc3Nrcw4169fj40bN2LkyJEIDAzEs2fPsGHDBowaNQr79u2Dvr4+Fi9ejCVLliA1NRUREREAABUVFQDA8+fPMWbMGGRnZ2Pq1KkwMTHBb7/9hsjISDx48ACrV68us21TU1NoaGhg/fr1EIlEcHBwQMuWLUuUc3JywuzZs7FmzRosWrRImJHXrl07AEBwcDASExMxYcIEODg4ICUlBevWrcOVK1fw/fffo1GjRsK9fvvtN9y+fRv+/v7Q19eHvr5+qbFdv34do0aNgomJCRYuXAgtLS0cPnwYs2fPxsuXLzFs2DAAwMqVK3H06FEEBgbCxMQET58+xfXr14UEb1keP34MbW1tBAUFQUtLCxkZGdi9ezeGDx+OpKQk6OjoyJWfN28enJ2dERoain/++QerV6/GnDlzsGPHDqHMggULcPDgQUyaNAk9evTAzZs3ERAQgBcvXpQbS3nS0tIAAM2aNQPwKjEVEREBLy8vfP755/j3338RFhaGkSNHYt++fULcsbGxCA4OxqBBgzBx4kQUFxdj+/bt8PHxwd69e+X21iwoKMDMmTMxZswYTJkyBbGxsVi5ciWePHmCo0ePYsqUKdDS0kJkZCSmTZuGY8eOCYnzOXPm4N69e5g5cyYMDQ2RnZ2Nq1evVvj84+PjMW/ePLi4uCAgIADPnz9HZGQkRo0ahdjYWJiammLx4sU4ePAgNm7cKCxx19bWrvCZFRUVQSKRyF0TiURQVFQE8Cpx7uvri8zMTMyePRutW7fGqVOnMGvWrBL3iomJwaJFizBgwAAh+RcREYGCggK57x/Aq0T1mDFj0LJlS0gkEiQnJ2P27Nl4/vy5MF5lDh8+jObNm+Ozzz6DmppahUv1k5KSkJeXJySN27dvj4SEBAQGBgr9qqwHDx6gXbt2cHd3h6amJh4+fIjo6GiMGjUKP/zwg/C9RSYgIABubm4YPnw4bty4ga+//hrAq3971XH//n0oKSlBU1MTt27dwujRo9GyZUssXboUampq2Lt3LwICAhASEoIhQ4YAeJW0Hjp0KJo0aYKZM2eiZcuWOHv2LFatWoXs7GzMmDGjWrEQEREREb2PmAQlauCePHmCFy9ewNDQsFLl/+///g/JyclYuHAhxo4dC+DVMu8mTZrgyy+/xJkzZ9CrVy+hfFFREb755hshYffixQvMnTsXDx48wLfffiuU+/vvv7Fjxw68ePECjRs3Fq4/fvwYP/zwA0xNTQEAjo6OGDx4MMLCwspMgj58+BCbN2/GhAkT8PnnnwvXbWxs4Orqiu3btyMoKAjt2rVD06ZNS102vnPnTqSkpCA+Ph4dOnQAAPTo0QONGjXCl19+CT8/P3z00Ueltq+mpobQ0FDMmzcPCxYsAADo6+vDwcFBmNEGANra2jA2NgbwKvH5egx///03EhISMHHiRGGPv549e8LAwADTpk1DYmIiRo4cKZTPzs7Gnj17KtxXdNWqVWjWrBl27twJNTU14b5PnjzBunXr4O3tDQUFBVy6dAmDBw+Wa6NPnz7l3hsAunXrJvQPePX5Ozk5wd7eHgcPHoSvr69ceScnJ+EZAa9m365evRqPHz+Gvr4+bt++jf3792PSpEmYPXs2gFfjrVmzZlXa+1AqlUIikaC4uBgpKSlYsmQJAGDQoEHIycnBN998A1dXV7kElLm5OT7++GNER0fjs88+Q15eHkJCQuDq6orQ0FChnIODA1xdXREZGYn169cL1/Pz8zF79mxhZqO1tTVOnTqFrVu34tChQzAyMgLwamx4enri6NGjGD16NIBXW1TMmDFDSFYBQL9+/crtY3FxMb7++muYm5tjw4YNQjLRzs4O/fr1Q2RkJL7++mu0a9dOSA5WZYm7t7d3iWtaWlrCzOWEhATcuXMHW7ZsgaOjI4BXy7RfvHiBuLg4oU5RURHWr18Pa2trhIWFCddtbGzQv3//Egn81z+T4uJi9OjRA0+ePMGuXbtKJEELCgoQHR0NdXX1SvUpLi4O6urq6N+/P4BXM2O/+uornDlzRuhDZQ0YMEDutUQiQbdu3eDs7IyffvoJffv2lXt/xIgRwl7E9vb2uHfvHuLi4vDll1+WSASXRpaQzs7Oxn//+19cu3YNrq6uaNSoESIjIyGVSrFjxw7o6uoCAJydneHt7Y3Q0FB4eHhAQUEBERERePnyJeLj44Xnbm9vj8LCQmzduhXjx4+HlpZWlZ4DEREREdH7isvhiUiOLOHh4eEhd12WrJG9L9O9e3e5GYuyZKaTk5NcOVNTU0ilUrnlyQDQoUMHoQ7wain5wIEDkZqaWuYy27Nnz0IikcDDwwMSiUT40tHRQadOnXDx4sUK+3n69Gl06NAB7dq1k7tH7969AUDuhPfSODo64vjx49i8eTMmTpyIVq1aYf/+/Rg7diyioqIqbL+s59y3b19oaGiUeM4dOnSoMAGan5+P5ORk9O/fHyoqKnL9cnR0REZGBlJSUgAAFhYWiI+PR0REBK5evYqCgoIKYwZeJaGioqLg7u4OKysrdOrUCZaWlsjLyxPu/bo3E6uymZSycVDWc3B3d6/STL2dO3fCzMwMnTt3hqenJ+7evYtFixbBxcUFV65cQX5+fok2OnbsiPbt2wsxXLlyBbm5uSXGlbKyMmxtbUuMK0VFRbk/CDRu3BgtW7aEsbGxkAAFABMTE7k+A6+e/zfffINt27bhr7/+qtSp3ykpKUhPT8fgwYPlkmj6+vqwt7cvMWaqas2aNYiNjZX7ev0PGRcuXEDTpk1LJA/ffK537txBRkYGBg4cKHfdwMAAVlZWJdr9+eefMWnSJNjb26NTp04wMzNDbGxsqeOpe/fulU6A3rlzB5cuXYKbm5vwhxdPT08oKyvLJW0r69mzZ1i3bh1cXV1hYWEBMzMzODs7A0Cpsb657F8sFiM/P7/E8vnSPH36FGZmZjAzM4ODgwM2bdoENzc3YX/b5ORk2NvbCwlQ4NWsXQ8PDzx+/Fg4COr06dPo0aMHtLW1S3w/KCgoqNT2C0REREREDQVnghI1cM2aNUPjxo1LJB/LkpOTAxUVFTRt2lTuuqamJlRUVEos19XU1JR7LdtztKzrbybbXv8lXka2NDk7O7vUfTdlSQRPT89S+1CZU9gzMzNx9+7dMg+NeXNfztI0btwYTk5OQsL39u3b8PX1xbp16zB8+PASz/B1OTk5AErvv66ubonnXNb+o6/Lzs6GRCLB9u3bhZOl3yTrV3BwMPT19XHgwAGEh4dDTU0Nffr0wZw5c8rdNmHlypWIiYmBn58funXrBg0NDYhEIvj5+SE/P79E+TdnmcmWDMvKyvr55nNQUlISlrJXhru7O8aPHw+RSAQNDQ0YGRkJe7OW1Qbw6rnKls7LxtXrh1297s3Ze2pqaiX22FVWVi7xub/ZZwBYu3YtNmzYgB07diAkJARaWlpwc3PDzJkzyxw35e1Fq6enV+FS+oq0a9eu3Fmj2dnZZT7D0uIsa2w/ePBAeH3lyhVMnDgR3bt3x6JFi9C8eXMoKSlh9+7dpSYqK/PvQEZWv2/fvsjNzQXwalz16NEDJ06cQFZWVqW2CZCZNWsWLl68iGnTpsHMzAxNmjSBVCrF8OHDqzX2y6OmpiZsGaGqqgpDQ0M0adJEeD8nJ6fcz0L2GWRmZuLHH3+s0fc5IiIiIqKGgklQogZOUVER3bt3x5kzZ4QlyOXR0tJCQUEBcnNz5ZIxOTk5KCgoqPWlk6XNisrMzBRiKY0sORYZGVlqwu7NvfnKuoeamhqWL19e6vsVPafSmJqaYtCgQYiOjkZqaiosLCzKLCvrW0ZGRonETkZGRolkVGWWzzZt2hQKCgrw8vKSW+b+OtnyfDU1NcyYMQMzZsxAZmYmTpw4gTVr1uCff/7Brl27ymzj4MGD8PT0lNtLsKCgQEjqVtXrz+H1pI5EIqlSgkZbWxudO3eusI03paenC+/LxtXixYvLvFdt0dbWRnBwMIKDg/HPP//gyJEjCA0NxfPnz7Fq1apS68jiK22G9Ov9qCtaWlr4448/Sm37dbI4S3veb15LSkqCkpISNm3aJPfv9vU9Y19XmX8HwKsl+fv27QMA+Pn5lVpm//79wnL1iuTm5uKnn35CQEAAJk+eLFy/d+9epepXlaKiYrljULYf75tkn8XrY7pTp0749NNPS71PZf5gRERERETUUDAJSvQB8PPzw+nTp7F06VKEhYVBSUn+n35eXh4uX74MBwcHdO/eHVFRUcLSbhnZie3du3ev1dhu3LiB27dvC0vii4uLcejQIbRt27bMWV8ODg5QVFTE/fv3S+zD9yYVFZUSp9oDr/Yy3Lp1K3R1dUs91Kg8z549g4KCgrDn5utky2JlsctmCr4Zg52dHYBXz/X1hOeJEyfw9OnTaj3nxo0bo1u3bvjrr7/QsWPHEp9zWXR0dDBs2DBcunQJP/74Y7llRSJRidmPcXFxlVrOXZrXn4Nsb1bgVbK1uvd8k5WVFVRVVbF//3658XL9+nXcvHlTSJJZW1tDXV0dKSkpwt6db0PLli0xfvx4HD9+XO6U9zcZGxsLs3d9fX2FhGBGRgbOnz9fqT1da8LOzg6HDh3C6dOn5ZbEy743vB6nnp4eDh06hHHjxgnXHz58iMuXL8v9gUF28NLryc3MzEwcP368RrH+9NNPePz4MXx9fUt9Lp999hni4uIqnQRVUFCAVCot8QeWvXv31ijO6urevTtOnDgh98cDqVSKAwcOoHnz5sIWDL169cL58+dhbGxc6W0EiIiIiIgaKiZBiT4A1tbWWLhwIb744gsMHz4cI0aMgLGxsbAn3J49ezBw4EA4ODigZ8+esLe3R0hICJ4+fQoLCwvh1PSePXvCwcGhVmPT09PDJ598IpwOv2fPHvz9999Yu3ZtmXWMjIzg7++PNWvWIC0tDfb29lBXV0d6ejouXboEExMTIYH70Ucf4dChQ4iJiUHHjh2hqqoKsViM8ePH48iRIxgzZgx8fX3Rvn17FBQU4MGDB/jpp5+wcOHCMg+TunPnDiZPnozBgweja9eu0NXVxZMnT5CUlISffvoJHh4eQmK1Xbt2EIlEiI2Nhbq6OlRUVGBsbAxTU1MMHTpUWLbes2dPpKSkICwsDB06dJA7MKcq5s+fjzFjxsDHxwcjRoyAgYEBcnNzkZKSgt9//x3h4eEAgOHDh8PJyQnt27dH06ZNcf36dRw5cqTCw2IcHR2RkJAAExMTiMVi/Prrr9izZ0+5S//LY2pqCg8PD2zfvh0KCgrC6fBbt26ttaRN06ZNMWXKFISFhWHevHkYOHAgHj9+jLCwMOjr6wuJsCZNmmD+/PkIDg5GTk4O+vXrJ8y4u3btGlRUVGrlNO2nT5/C19cX7u7uMDExQePGjfHrr7/i0qVLmDBhQpn1FBQU8NlnnyEoKAj+/v4YPnw48vLyEBkZCQUFBUybNq1Gcd24caPUpdrGxsbQ1NTEkCFDEB0djTlz5iAwMBBt2rTBqVOncP78+RJxfvrpp1i0aBECAwPh5eWF3NxcREREQE9PTy7h6eTkhOjoaMyePRsjRoxAeno6NmzYAF1dXTx//rzafYmLi4OysjI++eQTYXuN13l7e2PTpk34/fffKzXrV11dHTY2Nti6dSuaNWsGAwMDnD59GqdPn652jDUxbdo0nDx5Er6+vvD394eamhpiYmJw7do1rFmzRnjGM2bMwNmzZzFq1Cj4+PigTZs2yMvLw927d3Hy5Els27atSnvvEhERERG9z5gEJfpAjB49Gl26dEF0dDQ2bNiAzMxMqKiooF27dhg3bhxGjRoF4NXMrA0bNmD9+vWIiYkREhfjxo3D9OnTK70ctbLMzc0xaNAgREREIC0tDYaGhli1alWZJ8PLBAQEwNTUFN999x3i4+NRVFQEfX19WFpayi1D9/HxwfXr17F69Wo8ffoUhoaGOHHiBNTV1bF7925s2rQJu3btwsOHD6GmpoZWrVqhZ8+e5S4tbtOmDXx9ffHzzz/jyJEjePLkCZSVlWFqaiokIWUMDQ0xd+5c7NixAz4+PigqKsKOHTtgZ2eHZcuWwcjICHFxcdi5cyc0NTXh5uaGWbNmQVVVtVrPs0OHDoiPj0dkZCTWrFmD7OxsaGpqwsTERO6gGltbWxw7dgzbt29Hfn4+WrRogdGjR5e5H6ZMcHAwFBUVsWnTJuTn58PS0hJbt27F1KlTqxUvAKxYsQK6urqIj4/Hjh070LFjR0RERGDWrFnVvueb/P39oa2tje+++w4HDhxAkyZN4ODggNmzZ8vtC+nt7Q0DAwNERUVhwYIFePnyJXR1dWFubl7mFgNVpaqqis6dOyMhIQEPHjxAcXExDA0NERgYiEmTJpVbd8iQIVBTU8PmzZsxffp0KCsro1u3bli7dq0w+6+6goKCSr0eGRmJvn37onHjxtixYwdWrFghJNp69uyJr7/+usSzGTFiBEQiEaKiojBt2jQYGhrCz88Px48fl9ufuEePHli+fDmioqLwySefwMDAAL6+vsjMzERERES1+pGVlYVTp06hT58+pSZAAWDYsGHYvHkz4uLiKr31QWhoKL744guEhIQIsW/btq3EAUhvg6mpKXbv3o2vv/4aCxcuRGFhIcRisfBZyTRv3hxxcXGIjIzExo0bkZ6eDnV1dbRt2xa9evUS9s0lIiIiIvoQiKRSqbS+gyCiD5OLiws6dOiADRs21HcoRFTHcnNz4erqij59+ginnNP7qaioCLdv367vMKiaWrVqBUVFRWRlZdXaPWV/TKrNe9L7j+OCSsNxQaXhuKDStGjRotYnYXEmKBEREdWqf//9F1FRUbCzs0PTpk3x8OFDREdH4/nz5/D19a3v8IiIiIiI6APEJCgRERHVKhUVFdy7dw9JSUnIyclBo0aNYGlpiWXLluGjjz6q7/CIiIiIiOgDxCQoEdWbEydO1HcIRFQHmjVrhs2bN9d3GERERERERAImQYmIiIio0hQUFNCqVav6DoOqSVVVFRKJpL7DICIiInrrmAQlIiIioipRVFSs7xComiQSCZOgRERE9EHi6fBEREREVGlSqRT88ZFeJzu5leOCXsdxQaXhuKDScFxQaUQiUa2fDs8kKBERERERERERETVoCvUdABEREREREREREVFdYhKUiIiIiIiIiIiIGjQmQYmIiIiIiIiIiKhBYxKUiIiIiIiIiIiIGjQmQYmIiIiIiIiIiKhBYxKUiIiIiIiIiIiIGjQmQYmIiIiIiIiIiKhBYxKUiIiIiIiIiIiIGjQmQYmIiIiIiIiIiKhBYxKUiIiIiIiIiIiIGjQmQYmIiIiIiIiIiKhBYxKUiIiIiIiIiIiIGjSl+g6AiIiIiOrXn3/+iYiICPz222949uwZDAwMMGTIEIwfPx4qKipl1gsPD0dERESJ67q6ujh79mxdhkxvQXXHBQBIpVLs3bsXMTExuH37NpSVlWFiYoK5c+fC2tr6LfWA6kJ1x4WLiwsePHhQ6nvGxsY4fPhwXYVMb0FNvl8cOXIE27dvx+3btwEAJiYm8PX1hZub29sInepQTcZFXFwcoqOjcefOHTRt2hR9+vTBrFmz0KxZs7cUPdWVX3/9FWFhYfjtt9+goKAAGxsbzJ49G2KxuFL17927h6+++grJyckoLi5G165dERQUhHbt2lVYl0lQIiIiog/Y7du3MXLkSBgbG2P+/Plo1qwZfv75Z6xduxZ///03Vq1aVeE9tm/fDjU1NeG1srJyXYZMb0FNx8WCBQvw448/YvLkybCyssKLFy9w7do1vHjx4i31gOpCTcZFREQECgoK5K7dvHkTCxcuRN++fes6dKpDNRkXCQkJmDt3LlxdXTF16lQAr5JfM2fORF5eHoYOHfq2ukG1rCbjYtu2bQgJCcGwYcPw+eef49GjR1i3bh2uXbuGvXv38ueM99iVK1fg6+uLLl26YM2aNSguLsaWLVswduxYxMbGok2bNuXWz8zMxOjRo6Gjo4OQkBAoKipi48aNGDt2LBITE9GiRYvyA5ASERER0Qdr/fr10vbt20vv3r0rd3327NnSTp06SQsKCiqsm5OTU9dh0ltWk3Fx+PBhaYcOHaSXLl2q6zDpLavJuCjN8uXLpe3bt5empKTUZpj0ltVkXIwdO1bq7OwsLSoqEq4VFRVJnZ2dpWPHjq2zmKnuVXdcvHz5UmplZSWdOnWq3PXk5GRp+/btpTExMXUWM9W9CRMmSB0cHKQvXrwQruXk5Ei7desmnTVrVoX1Q0JCpJ07d5Y+evRIuJaVlSW1srKSLlq0qML63BOUiIiI6AOmpPRqYZC6urrcdQ0NDSgpKUFRUbE+wqJ6VpNx8d1336Fr166wsrKq0xjp7avN7xcFBQU4cOAAbGxsYGxsXKtx0ttVk3GhpKQENTU1KCj8LzWhoKAANTW1CpdL07utuuPi1q1beP78OZydneWu29raQl1dHUeOHKmbgOmtuHz5Mrp3745GjRoJ15o2bQobGxscP34cRUVF5dY/duwY7O3t0bx5c+Fas2bN4OzsjKNHj1bYPpOgRERERB8wT09PaGlpYcmSJUhLS8OzZ89w7NgxJCQkYMKECXK/mJbFzc0NHTt2RM+ePREcHIzMzMy3EDnVpeqOi8LCQly5cgVisRhff/017O3t0alTJwwaNAgJCQlvuRdU22rj+4XMsWPHkJ2dDW9v7zqMmN6GmoyLMWPG4Pbt29i4cSOysrKQlZWFjRs34s6dO/D19X2LvaDaVpP/R4DSt9ZRUVHBzZs36zRuqluFhYWl/oFDRUUFL168QFpaWpl1X758iXv37qF9+/Yl3hOLxcjMzKzwZ1DuCUpERET0ATMwMEBMTAymTZsmty/flClTMGPGjHLrGhkZYdasWejYsSOUlZVx6dIlREVF4fz584iPj4empmYdR091pbrjIjs7GwUFBUhISECLFi2wcOFCNG3aFLGxsZg7dy4KCwsxfPjwt9ADqgs1+X7xpri4OKipqWHgwIG1HCW9bTUZF3379sXGjRsxZ84crFu3DgCgpqaGsLAw9O7duw6jprpW3XFhbGwMBQUFXL58GUOGDBGu37lzB1lZWdwP9D3Xrl07XL16FVKpFCKRCMCrxOjvv/8OAHjy5Anatm1bat2cnBxIpdJSf77U0tIC8OrnEB0dnTLbZxKUiIiI6AP24MEDTJkyBXp6eoiMjISGhgYuXryIzZs3QyQSlfuLyuu/nABAjx49YGlpiYkTJ2LXrl3w9/ev2+CpzlR3XBQXFwMA8vPzsWXLFhgaGgIA7O3tkZaWhsjISCZB32M1+X7xukePHuHcuXPw8vKSO1SN3k81GRdnz57FZ599hkGDBsHV1RVFRUU4cOAAZs2ahfXr18PJyemt9YNqV3XHhZaWFgYPHoy4uDiYmZmhf//+ePToEYKDg6GoqFilGef07hk7diwWLFiAL774An5+figuLsb69evx6NEjAKjU5ytLnlYHk6BEREREH7DQ0FA8f/4ciYmJwv5MdnZ2AIDIyEgMHToUrVq1qvT9HBwcoKenhytXrtRFuPSWVHdcaGpqQiQSwcTEREiAAq9+YenVqxc2bNiAzMzMcmdp0Lurtr5fxMfHo7i4mEvhG4jqjgupVIqgoCB0794dy5YtE6737t0bjx49wvLly5kEfY/V5PvFkiVLIJVKsXjxYixcuBAKCgrw9PSEnp4ebt269db6QLVv6NChwrYX3333HQDAysoKEydOxDfffAN9ff0y68p+xsjOzi7xnuyabEZoWZhCJyIiIvqA/fnnn2jXrp3cBvUAYG5ujuLiYqSkpFT5nlKplDM13nPVHReNGjVCmzZtSn1PKpUCqNkMDqpftfH9QiqVIiEhASYmJrC2tq6rUOktqu64yMjIQHp6OszNzUu8Z25ujvv37yM/P79OYqa6V5PvF2pqali9ejV+/vln7Nu3D2fPnsVXX32FlJQUdO3ata5Dpzrm5+eH5ORkHDhwACdOnMCePXuQk5MDQ0NDtGzZssx6jRo1gpGRUan7wt68eRPa2toV/pGVP50SERERfcD09fVx69YtvHjxQu765cuXAUDu9M3KOHPmDDIyMtClS5dai5HevpqMi379+iElJQX3798XrkmlUvz0008wMjKCtrZ23QRNda42vl9cuHAB9+7d4yzQBqS640JTUxOqqqr47bffSrx39epVaGlpQVVVtfYDpreiNr5faGpqokOHDtDW1sbRo0dx9+5djB49uk7ipbdLRUUF7du3h6GhIe7fv4+kpCSMGTOmwnp9+/bFuXPnkJ6eLlzLzs7GyZMn0a9fvwrri6SyP8kSERER0Qfn2LFjmDZtGmxsbODr6wsNDQ0kJydj69at6NatG7Zv3w4A8PHxwYULF3Djxg2h7pAhQzBkyBAYGxtDSUkJly9fxtatW6Gnp4fY2Fg0bdq0vrpFNVSTcfHkyRMMGTIEjRs3RkBAADQ0NBAXF4cjR45g7dq1cHNzq69uUQ3VZFzIfP755/jhhx9w+vRp6Orqvu0uUB2oybj48ssv8e2332Lo0KFwdXVFcXExEhMTcejQIcyYMQNTp06tr25RDdVkXCQlJSEzMxOmpqZ4+fIlkpOTsXPnTkydOhWffvppfXWJasH169dx7NgxmJubQ0VFBX/99Re2bNmCTp064ZtvvoGS0v927SxtbGRkZMDT0xP6+vqYNm0alJSUsHHjRqSmpiIhIQEGBgblts8kKBEREdEH7ty5c9iyZQtu3ryJvLw8GBoaws3NDRMmTBAOLSntB9FZs2bh2rVrePz4MSQSCVq0aAEXFxf4+/tXuCcTvfuqOy4A4P79+1i1ahXOnz+Ply9fon379pg6darcCcH0fqrJuHj27Bl69uyJHj16YOPGjfURPtWR6o6LoqIifP/999i7dy/u3bsHBQUFtG3bFmPGjIGHhwe3z3jPVXdcHDlyBBs2bMC9e/cAAGKxGOPGjeMf0RqA27dvY9GiRbh16xby8vJgZGSEIUOGYMKECVBRUZErW9b/JampqQgJCUFycjKkUilsbGwQFBSEjz76qML2mQQlIiIiIiIiIiKiBo17ghIREREREREREVGDxiQoERERERERERERNWhMghIREREREREREVGDxiQoERERERERERERNWhMghIREREREREREVGDxiQoERERERERERERNWhMghIRERERERG9JWKxGHPnzq3vMIiIPjhK9R0AERERERER0duWnJyMcePGyV1TU1ND27Zt4enpibFjx0JJqeq/Mufm5uLbb7+Fra0t7OzsaitcIiKqISZBiYiIiIiI6IPl7u6O3r17QyqVIiMjA/v27cPKlStx+/ZtLF++vMr3y83NRUREBAICAkpNgv72229QUOCiTCKit41JUCIiIiIiIvpgderUCZ6ensLr0aNHY+DAgfj+++8xc+ZMaGtr12p7qqqqtXo/IiKqHP75iYiIiIiIiOj/U1NTQ5cuXSCVSnHv3j0AQHFxMTZu3IgxY8bAwcEB5ubmcHJywuLFi/HkyROhbnJyMvr06QMAiIiIgFgshlgshouLi1CmtD1BZdcuX76MsWPHwtLSEnZ2dliwYAGeP39eIsYLFy5gxIgRsLCwgIODA7744gvcunULYrEY4eHhdfFYiIjee5wJSkRERERERPSatLQ0AICmpiYAoLCwEFu3bkX//v3Rp08fNG7cGL///jvi4uJw6dIlxMXFQUVFBaamppg3bx5WrlyJfv36oV+/fgCAJk2aVNjmX3/9hSlTpsDLywvu7u64cOECYmNjoaCgILcs/5dffsHEiROhqakJPz8/aGho4NChQ7h06VIdPAkiooaDSVAiIiIiIiL6YL148QJZWVkAgPT0dOzZswd//vknOnfuDGNjYwCAiooKzpw5g0aNGgn1Ro0aBSsrKwQHB+PYsWNwc3ODrq4u+vbti5UrV0IsFssts6/IjRs3sGfPHlhaWgIARo4ciWfPniE+Ph5z584VEqlfffUVRCIR9uzZAyMjIwCvlvD7+PjUxuMgImqwuByeiIiIiIiIPljh4eHo0aMHevToAQ8PD/z3v/9F//79sXHjRqGMSCQSEqBFRUXIzc1FVlYWunfvDuDVYUc1ZWlpKSRAZbp37w6JRIIHDx4AADIyMvD777+jT58+QgIUAJSVlUucdE9ERPI4E5SIiIiIiIg+WCNGjMCAAQNQWFiImzdvIioqCo8ePSpxgFFSUhK2b9+Ov/76C4WFhXLv5eTk1DiO15OaMlpaWgCA7OxsAMD9+/cBQJih+joTE5Max0BE1JAxCUpEREREREQfrDZt2sDe3h4A4OjoCBsbG4wePRqLFy/G2rVrAQA//vgjZs6cCQsLC8yfPx8tW7aEqqoqioqKMHnyZEil0hrHoaioWOZ7svvXRjtERB8qJkGJiIiIiIiI/j9ra2t4enoiMTERPj4+sLa2xr59+6CqqoodO3agcePGQtnbt2+XqC8SieosNtls0Tt37pR4LyUlpc7aJSJqCLgnKBEREREREdFr/P39oaioiPXr1wN4NUtTJBKhuLhYKCOVSuX2DZVRU1MDUDtL5N+kq6sLc3NzHD9+XDjBHnh1ev2OHTtqvT0iooaEM0GJiIiIiIiIXtOmTRu4ubnhwIED+OWXX+Dq6oojR47A19cXQ4YMgUQiwbFjx/DixYsSdZs1a4Y2bdrghx9+gJGREXR1ddG4cWO4uLjUSmxBQUGYOHEiRo4ciVGjRkFDQwOHDh0S9imty5moRETvM84EJSIiIiIiInrD1KlToaCggLCwMAwaNAjLly9HXl4eQkJCEBUVBWNjY2zdurXUumvWrEGbNm2wdu1azJo1C1988UWtxWVra4tvvvkGhoaG2Lx5MzZv3gxzc3MsWrQIAEoc6ERERK+IpNxZmYiIiIiIiOi9duTIEUyfPh1ff/01Bg0aVN/hEBG9czgTlIiIiIiIiOg9IZVKkZ+fL3etsLAQ27dvh5KSEmxtbespMiKidxv3BCUiIiIiIiJ6TxQUFMDZ2RmDBw+GsbExsrOzkZSUhBs3buA///kP9PT06jtEIqJ3EpOgRERERERERO8JJSUlODo64vjx40hPT4dUKoWxsTEWLVqEMWPG1Hd4RETvLO4JSkRERERERERERA0a9wQlIiIiIiIiIiKiBo1JUCIiIiIiIiIiImrQmAQlIiIiIiIiIiKiBo1JUCIiIiIiIiIiImrQmAQlIiIiIiIiIiKiBo1JUCIiIiIiIiIiImrQ/h+1SoM53tgb7AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 864x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "_, ax = plt.subplots(figsize=(12, 10))\n",
    "\n",
    "# Most rated books\n",
    "high_rated_works_top = by_work_weighted\\\n",
    "    .sort_values(by='weighted_rating', ascending=False).head(20)\n",
    "high_rated_works_top = pd.merge(high_rated_works_top,\n",
    "                                works_data[['title']], how='left',\n",
    "                                right_index=True, left_index=True)\n",
    "sns.barplot(x=high_rated_works_top['weighted_rating'],\n",
    "            y=high_rated_works_top['title'],\n",
    "            palette=sns_palette, ax=ax, ci=None)\n",
    "ax.set_xlim(8.5, 9)\n",
    "ax.set_ylabel('Work titles')\n",
    "ax.set_xlabel('Rating')\n",
    "ax.set_title('Top 20 most rated books')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Save"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "books.to_csv(os.path.join('data_prep', 'books.csv'), index=True)\n",
    "categories.to_csv(os.path.join('data_prep', 'categories.csv'), index=True)\n",
    "roles.to_csv(os.path.join('data_prep', 'contributor_roles.csv'), index=True)\n",
    "publishers.to_csv(os.path.join('data_prep', 'publishers.csv'), index=True)\n",
    "series.to_csv(os.path.join('data_prep', 'series.csv'), index=True)\n",
    "authors.to_csv(os.path.join('data_prep', 'contributors.csv'), index=True)\n",
    "works.to_csv(os.path.join('data_prep', 'works.csv'), index=True)\n",
    "book_cats.to_csv(os.path.join('data_prep', 'book_categories.csv'))\n",
    "book_authors.to_csv(os.path.join('data_prep', 'book_contributors.csv'))\n",
    "work_ratings.to_csv(os.path.join('data_prep', 'work_ratings.csv'))"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "data_collect.ipynb",
   "provenance": [],
   "toc_visible": true
  },
  "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}