notebooks/3-Exploratory-Data-Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LckNd8k7KIbi"
   },
   "source": [
    "<div align=\"center\">\n",
    "  <h1>3 - Exploratory Data Analysis</h1> <a name=\"0-bullet\"></a>\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "UE6hozVVKIbx"
   },
   "source": [
    "The analysis is focused around the general characteristics of Eminem's songs (e.g., the average number of verses in a song, the number of lines in a verse, the general sentiment of a song, and similarly). In other words, the goal of the analysis is to determine what should be an \"expected\" output of the lyrics generators in the following notebooks (4a/b-Lyrics-Generation.ipynb), at least from the perspective of song structure. \n",
    "\n",
    "Therefore, the question of the analysis is: **\"What does make Eminem Eminem (structurally)?\"**.\n",
    "\n",
    "The notebook is structured in four sections, including \"Setup\", \"Song structure analysis\", \"Vocabulary analysis\" and \"Lyrics sentiment analysis\"."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "UtRtmZS_KIbz"
   },
   "source": [
    "- [1. Setup](#1-bullet)\n",
    "    * [1.1 Set the working directory](#11-bullet)\n",
    "    * [1.2 Load the data](#12-bullet)\n",
    "    * [1.3 Missing data visualization](#13-bullet)\n",
    "    * [1.4 Define helper functions](#14-bullet)\n",
    "        + [1.4.1 Splitting lyrics into lines, verses and words](#141-bullet)\n",
    "        + [1.4.2 Contractions and dash/hyphen words](#142-bullet) <br><br>\n",
    "- [2. Song structure analysis](#2-bullet)\n",
    "    * [2.1 The average number of verses, lines, words, characters and features per song](#21-bullet)\n",
    "        + [2.1.1 The average number of verses per song](#211-bullet)\n",
    "        + [2.1.2 The average number of lines per song](#212-bullet)\n",
    "        + [2.1.3 The average number of words per song](#213-bullet)\n",
    "        + [2.1.4 The average number of characters per song](#214-bullet)\n",
    "        + [2.1.5 The average number of features (collaborations) per song](#215-bullet)\n",
    "    * [2.2 The average length of a verse, line and word](#22-bullet)\n",
    "        + [2.2.1 The average number of lines in a verse](#221-bullet)\n",
    "        + [2.2.2 The average number of words in a line](#222-bullet)\n",
    "        + [2.2.3 The average length of a word](#223-bullet)\n",
    "    * [2.3 Summary](#23-bullet) <br><br>\n",
    "- [3. Vocabulary analysis](#3-bullet)\n",
    "    * [3.1 The number of unique words](#31-bullet)\n",
    "    * [3.2 The average ratio of unique words per song](#32-bullet)\n",
    "    * [3.3 The most frequent words](#33-bullet)\n",
    "    * [3.4 Word sentiment analysis](#34-bullet)\n",
    "        + [3.4.1 The average subjectivity and polarity of words](#341-bullet)\n",
    "        + [3.4.2 The most frequent positive, negative and neutral words](#342-bullet)\n",
    "        + [3.4.3 The most frequent words including sentiment annotation](#343-bullet)\n",
    "    * [3.5 Profanity analysis](#35-bullet)\n",
    "        + [3.5.1 The most common profanity words](#351-bullet)\n",
    "        + [3.5.2 The relative frequency of profanity words](#352-bullet)\n",
    "        + [3.5.3 The average number of profanity words per song](#353-bullet)\n",
    "        + [3.5.4 S-word/F-word graph](#354-bullet)\n",
    "        + [3.5.5 Profanity words through the years](#355-bullet)\n",
    "    * [3.6 Word count by the first letter](#36-bullet) <br><br>\n",
    "- [4. Lyrics sentiment analysis](#4-bullet)\n",
    "    * [4.1 The subjectivity/polarity of lyrics](#41-bullet)\n",
    "    * [4.2 The polarity of songs through the years](#42-bullet) <br><br>\n",
    "- [5. Topic modeling](#5-bullet) <br>\n",
    "// TODO\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "-b3VyNG5Ib9U"
   },
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "8v-nD1mPIdRY"
   },
   "source": [
    "You may have to install [PyEnchant](https://pyenchant.github.io/pyenchant/index.html) and [Pickle5](https://pypi.org/project/pickle5/):\n",
    "> `!pip install pyenchant` <br>\n",
    "> `!pip install pickle5`\n",
    "\n",
    ", or in Google Colab:\n",
    "> `!apt install -qq enchant` <br>\n",
    "> `!pip install pyenchant` <br><br>\n",
    "> `!pip install pickle5`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 420,
     "status": "ok",
     "timestamp": 1615900639601,
     "user": {
      "displayName": "Mario Štrbac",
      "photoUrl": "",
      "userId": "03507842523962903593"
     },
     "user_tz": -60
    },
    "id": "qhFWtlbpKIb2",
    "outputId": "8f53d041-4b54-4d88-873c-debd0fdc611b"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to /home/mario/nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n",
      "[nltk_data] Downloading package wordnet to /home/mario/nltk_data...\n",
      "[nltk_data]   Package wordnet is already up-to-date!\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import re\n",
    "import json\n",
    "import string\n",
    "import enchant\n",
    "import numpy as np  \n",
    "import pandas as pd\n",
    "import missingno as msno\n",
    "import pickle5 as pickle\n",
    "\n",
    "from typing import Union\n",
    "from collections import defaultdict\n",
    "\n",
    "from textblob import TextBlob\n",
    "\n",
    "import nltk\n",
    "from nltk.corpus import stopwords\n",
    "nltk.download('stopwords')\n",
    "from nltk.stem.wordnet import WordNetLemmatizer\n",
    "nltk.download('wordnet')\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.style as style\n",
    "from matplotlib.patches import Patch\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "KxzUt8_sKIb8"
   },
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "673qXDH5KIb-"
   },
   "source": [
    "# 1. Setup <a name=\"1-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "lF63shSOEAGo"
   },
   "source": [
    "## 1.1 Set the working directory <a name=\"11-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 21293,
     "status": "ok",
     "timestamp": 1615900366436,
     "user": {
      "displayName": "Mario Štrbac",
      "photoUrl": "",
      "userId": "03507842523962903593"
     },
     "user_tz": -60
    },
    "id": "UPhbmZq8D9Y-",
    "outputId": "2bc464b1-06ab-483b-ef27-2f0f5e6bce7f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mounted at /content/gdrive\n",
      "/content/gdrive/My Drive\n",
      "/content/gdrive/My Drive/eminem-lyrics-generator/notebooks\n"
     ]
    }
   ],
   "source": [
    "ROOT_DIR = \"./eminem-lyrics-generator/notebooks/\" \n",
    "IN_GOOGLE_COLAB = True\n",
    "\n",
    "if IN_GOOGLE_COLAB:\n",
    "    # mount google drive\n",
    "    from google.colab import drive\n",
    "    drive.mount('/content/gdrive')\n",
    "\n",
    "    # change the current working directory\n",
    "    %cd gdrive/'My Drive'\n",
    "\n",
    "    # create a root directory if there's none\n",
    "    if not os.path.isdir(ROOT_DIR):\n",
    "        %mkdir $ROOT_DIR\n",
    "\n",
    "    # change the current working directory\n",
    "    %cd $ROOT_DIR"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "jvNBXUedKIb-"
   },
   "source": [
    "## 1.2 Load the data <a name=\"12-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "ffMv9TE4Kl2L"
   },
   "outputs": [],
   "source": [
    "# specifies paths to all files in the project\n",
    "SETTINGS_FILE_DIR = os.path.join(os.path.abspath(\"..\"), \"SETTINGS.json\")\n",
    "settings = json.load(open(SETTINGS_FILE_DIR))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>song_id</th>\n",
       "      <th>state</th>\n",
       "      <th>title</th>\n",
       "      <th>artist</th>\n",
       "      <th>featured</th>\n",
       "      <th>popularity</th>\n",
       "      <th>album_id</th>\n",
       "      <th>album_name</th>\n",
       "      <th>release_date</th>\n",
       "      <th>lyrics</th>\n",
       "      <th>url</th>\n",
       "      <th>comments</th>\n",
       "      <th>comments_total_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>235729</td>\n",
       "      <td>complete</td>\n",
       "      <td>Rap God</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>672689</td>\n",
       "      <td>The Marshall Mathers LP2 (Deluxe)</td>\n",
       "      <td>2013-10-14</td>\n",
       "      <td>[Intro]\n",
       "\"Look, I was gonna go easy on you not ...</td>\n",
       "      <td>https://genius.com/Eminem-rap-god-lyrics</td>\n",
       "      <td>[It’s clear now to everyone who the King and t...</td>\n",
       "      <td>459</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3958196</td>\n",
       "      <td>complete</td>\n",
       "      <td>Killshot</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>None</td>\n",
       "      <td>2018-09-14</td>\n",
       "      <td>[Intro]\n",
       "You sound like a bitch, bitch\n",
       "Shut the...</td>\n",
       "      <td>https://genius.com/Eminem-killshot-lyrics</td>\n",
       "      <td>[Eminem Sucks haha, GENIUS NOT EVEN READY LOL,...</td>\n",
       "      <td>1252</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5180439</td>\n",
       "      <td>complete</td>\n",
       "      <td>Godzilla</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Juice WRLD]</td>\n",
       "      <td>3</td>\n",
       "      <td>594809</td>\n",
       "      <td>Music to Be Murdered By</td>\n",
       "      <td>2020-01-17</td>\n",
       "      <td>[Intro]\n",
       "Ugh, you're a monster\n",
       "\n",
       "[Verse 1: Emine...</td>\n",
       "      <td>https://genius.com/Eminem-godzilla-lyrics</td>\n",
       "      <td>[This definitely takes the spot of most unexpe...</td>\n",
       "      <td>624</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>207</td>\n",
       "      <td>complete</td>\n",
       "      <td>Lose Yourself</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>4</td>\n",
       "      <td>452012</td>\n",
       "      <td>The Singles</td>\n",
       "      <td>2002-10-28</td>\n",
       "      <td>[Intro]\n",
       "Look, if you had one shot or one oppor...</td>\n",
       "      <td>https://genius.com/Eminem-lose-yourself-lyrics</td>\n",
       "      <td>[This was Eminem at the top of the entire rap ...</td>\n",
       "      <td>265</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>235732</td>\n",
       "      <td>complete</td>\n",
       "      <td>The Monster</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Rihanna]</td>\n",
       "      <td>5</td>\n",
       "      <td>672689</td>\n",
       "      <td>The Marshall Mathers LP2 (Deluxe)</td>\n",
       "      <td>2013-10-29</td>\n",
       "      <td>[Intro: Rihanna]\n",
       "I'm friends with the monster ...</td>\n",
       "      <td>https://genius.com/Eminem-the-monster-lyrics</td>\n",
       "      <td>[Why do people hate that Eminem is different n...</td>\n",
       "      <td>171</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   song_id     state          title  artist      featured  popularity  \\\n",
       "0   235729  complete        Rap God  Eminem          None           1   \n",
       "1  3958196  complete       Killshot  Eminem          None           2   \n",
       "2  5180439  complete       Godzilla  Eminem  [Juice WRLD]           3   \n",
       "3      207  complete  Lose Yourself  Eminem          None           4   \n",
       "4   235732  complete    The Monster  Eminem     [Rihanna]           5   \n",
       "\n",
       "   album_id                         album_name release_date  \\\n",
       "0    672689  The Marshall Mathers LP2 (Deluxe)   2013-10-14   \n",
       "1      <NA>                               None   2018-09-14   \n",
       "2    594809            Music to Be Murdered By   2020-01-17   \n",
       "3    452012                        The Singles   2002-10-28   \n",
       "4    672689  The Marshall Mathers LP2 (Deluxe)   2013-10-29   \n",
       "\n",
       "                                              lyrics  \\\n",
       "0  [Intro]\n",
       "\"Look, I was gonna go easy on you not ...   \n",
       "1  [Intro]\n",
       "You sound like a bitch, bitch\n",
       "Shut the...   \n",
       "2  [Intro]\n",
       "Ugh, you're a monster\n",
       "\n",
       "[Verse 1: Emine...   \n",
       "3  [Intro]\n",
       "Look, if you had one shot or one oppor...   \n",
       "4  [Intro: Rihanna]\n",
       "I'm friends with the monster ...   \n",
       "\n",
       "                                              url  \\\n",
       "0        https://genius.com/Eminem-rap-god-lyrics   \n",
       "1       https://genius.com/Eminem-killshot-lyrics   \n",
       "2       https://genius.com/Eminem-godzilla-lyrics   \n",
       "3  https://genius.com/Eminem-lose-yourself-lyrics   \n",
       "4    https://genius.com/Eminem-the-monster-lyrics   \n",
       "\n",
       "                                            comments  comments_total_count  \n",
       "0  [It’s clear now to everyone who the King and t...                   459  \n",
       "1  [Eminem Sucks haha, GENIUS NOT EVEN READY LOL,...                  1252  \n",
       "2  [This definitely takes the spot of most unexpe...                   624  \n",
       "3  [This was Eminem at the top of the entire rap ...                   265  \n",
       "4  [Why do people hate that Eminem is different n...                   171  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "EMINEM_DF_DIR = settings['EMINEM_DF_ALL_CLEAN_PATH']              # or 'EMINEM_DF_SONGS_CLEAN_PATH' \n",
    "\n",
    "with open(EMINEM_DF_DIR, 'rb') as f:\n",
    "    data_df = pickle.load(f)\n",
    "\n",
    "data_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "eh_WfybmKIcB"
   },
   "source": [
    "## 1.3 Missing data visualization (songs sorted by popularity) <a name=\"13-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "q_S1BoCRKIcC",
    "outputId": "2296e96f-a465-4f3a-a847-3ea74c1c44dd"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# sorted by popularity\n",
    "msno.matrix(data_df.sort_values('popularity', ascending=True), \n",
    "            figsize=(15,6), \n",
    "            fontsize=13)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "E_ywShvNKIcF"
   },
   "source": [
    "Some of the columns, like \"song_id\", \"artist\", \"state\", \"album_id\", \"comments\", etc., are not very useful for the analysis, so they can be removed from the dataset. However, they can be still quite useful for inspecting potentially corrupted lyrics and deciding whether or not to remove them from the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "oyavQT5ZKIcG",
    "outputId": "dfee9bae-1476-44db-c3f0-f24f8c4342d5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "complete      415\n",
      "unreleased      0\n",
      "Name: state, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(data_df['state'].value_counts())\n",
    "\n",
    "# remove the state column from the dataset\n",
    "states = data_df.pop('state')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "IPrK3OY8KIcG",
    "outputId": "4f1879d8-f09c-460b-ae58-035dbaa93214"
   },
   "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>song_id</th>\n",
       "      <th>title</th>\n",
       "      <th>artist</th>\n",
       "      <th>featured</th>\n",
       "      <th>popularity</th>\n",
       "      <th>album_id</th>\n",
       "      <th>album_name</th>\n",
       "      <th>release_date</th>\n",
       "      <th>lyrics</th>\n",
       "      <th>url</th>\n",
       "      <th>comments</th>\n",
       "      <th>comments_total_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>235729</td>\n",
       "      <td>Rap God</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>672689</td>\n",
       "      <td>The Marshall Mathers LP2 (Deluxe)</td>\n",
       "      <td>2013-10-14</td>\n",
       "      <td>[Intro]\n",
       "\"Look, I was gonna go easy on you not ...</td>\n",
       "      <td>https://genius.com/Eminem-rap-god-lyrics</td>\n",
       "      <td>[It’s clear now to everyone who the King and t...</td>\n",
       "      <td>459</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3958196</td>\n",
       "      <td>Killshot</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>None</td>\n",
       "      <td>2018-09-14</td>\n",
       "      <td>[Intro]\n",
       "You sound like a bitch, bitch\n",
       "Shut the...</td>\n",
       "      <td>https://genius.com/Eminem-killshot-lyrics</td>\n",
       "      <td>[Eminem Sucks haha, GENIUS NOT EVEN READY LOL,...</td>\n",
       "      <td>1252</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5180439</td>\n",
       "      <td>Godzilla</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Juice WRLD]</td>\n",
       "      <td>3</td>\n",
       "      <td>594809</td>\n",
       "      <td>Music to Be Murdered By</td>\n",
       "      <td>2020-01-17</td>\n",
       "      <td>[Intro]\n",
       "Ugh, you're a monster\n",
       "\n",
       "[Verse 1: Emine...</td>\n",
       "      <td>https://genius.com/Eminem-godzilla-lyrics</td>\n",
       "      <td>[This definitely takes the spot of most unexpe...</td>\n",
       "      <td>624</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>207</td>\n",
       "      <td>Lose Yourself</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>4</td>\n",
       "      <td>452012</td>\n",
       "      <td>The Singles</td>\n",
       "      <td>2002-10-28</td>\n",
       "      <td>[Intro]\n",
       "Look, if you had one shot or one oppor...</td>\n",
       "      <td>https://genius.com/Eminem-lose-yourself-lyrics</td>\n",
       "      <td>[This was Eminem at the top of the entire rap ...</td>\n",
       "      <td>265</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>235732</td>\n",
       "      <td>The Monster</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Rihanna]</td>\n",
       "      <td>5</td>\n",
       "      <td>672689</td>\n",
       "      <td>The Marshall Mathers LP2 (Deluxe)</td>\n",
       "      <td>2013-10-29</td>\n",
       "      <td>[Intro: Rihanna]\n",
       "I'm friends with the monster ...</td>\n",
       "      <td>https://genius.com/Eminem-the-monster-lyrics</td>\n",
       "      <td>[Why do people hate that Eminem is different n...</td>\n",
       "      <td>171</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   song_id          title  artist      featured  popularity  album_id  \\\n",
       "0   235729        Rap God  Eminem          None           1    672689   \n",
       "1  3958196       Killshot  Eminem          None           2      <NA>   \n",
       "2  5180439       Godzilla  Eminem  [Juice WRLD]           3    594809   \n",
       "3      207  Lose Yourself  Eminem          None           4    452012   \n",
       "4   235732    The Monster  Eminem     [Rihanna]           5    672689   \n",
       "\n",
       "                          album_name release_date  \\\n",
       "0  The Marshall Mathers LP2 (Deluxe)   2013-10-14   \n",
       "1                               None   2018-09-14   \n",
       "2            Music to Be Murdered By   2020-01-17   \n",
       "3                        The Singles   2002-10-28   \n",
       "4  The Marshall Mathers LP2 (Deluxe)   2013-10-29   \n",
       "\n",
       "                                              lyrics  \\\n",
       "0  [Intro]\n",
       "\"Look, I was gonna go easy on you not ...   \n",
       "1  [Intro]\n",
       "You sound like a bitch, bitch\n",
       "Shut the...   \n",
       "2  [Intro]\n",
       "Ugh, you're a monster\n",
       "\n",
       "[Verse 1: Emine...   \n",
       "3  [Intro]\n",
       "Look, if you had one shot or one oppor...   \n",
       "4  [Intro: Rihanna]\n",
       "I'm friends with the monster ...   \n",
       "\n",
       "                                              url  \\\n",
       "0        https://genius.com/Eminem-rap-god-lyrics   \n",
       "1       https://genius.com/Eminem-killshot-lyrics   \n",
       "2       https://genius.com/Eminem-godzilla-lyrics   \n",
       "3  https://genius.com/Eminem-lose-yourself-lyrics   \n",
       "4    https://genius.com/Eminem-the-monster-lyrics   \n",
       "\n",
       "                                            comments  comments_total_count  \n",
       "0  [It’s clear now to everyone who the King and t...                   459  \n",
       "1  [Eminem Sucks haha, GENIUS NOT EVEN READY LOL,...                  1252  \n",
       "2  [This definitely takes the spot of most unexpe...                   624  \n",
       "3  [This was Eminem at the top of the entire rap ...                   265  \n",
       "4  [Why do people hate that Eminem is different n...                   171  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "UkBJjSxDKIcI"
   },
   "source": [
    "## 1.4 Define helper functions <a name=\"14-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "p4WllqtGKIcI"
   },
   "source": [
    "### 1.4.1 Splitting lyrics into lines, verses and words <a name=\"141-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "jM-NuZ24KIcI"
   },
   "outputs": [],
   "source": [
    "def split_lyrics_into_lines(lyrics: str) -> list:\n",
    "    \"\"\"Splits lyrics into lines. The end of a line is specified with the newline character.\"\"\"\n",
    "    \n",
    "    # remove section headers, lower all letters and split it into lines\n",
    "    step1 = re.sub(\"\\[.*\\]|[\\'-]\", \"\", lyrics) \\\n",
    "              .lower() \\\n",
    "              .split('\\n')\n",
    "    # remove all characters except letters\n",
    "    step2 = [re.sub(\"[^a-z]\", \" \", vers) for vers in step1 if vers != '']\n",
    "    # remove multiple white spaces\n",
    "    step3 = [re.sub(\"\\s{2,}\", \" \", vers).strip() for vers in step2]\n",
    "\n",
    "    return step3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "uGb_vRUEKIcK"
   },
   "outputs": [],
   "source": [
    "def split_lyrics_into_words(lyrics: str) -> list:\n",
    "    \"\"\"Splits lyrics into words. Does not clean contractions or dash words.\"\"\"\n",
    "\n",
    "    # remove section headers\n",
    "    step1 = re.sub(\"\\[.*\\]\\n\", \" \", lyrics.lower())           \n",
    "    # remove all characters except letters, single quotation marks and dashes\n",
    "    step2 = re.sub(\"[^a-z\\'-]\", \" \", step1)             # or [^a-zA-z0-9]\n",
    "    # replace consecutive identical quotations and dashes with only one character\n",
    "    step3 = re.sub(r\"([\\'-]|\\s)\\1{2,}\", r\"\\1\", step2)\n",
    "    # filter consecutive white spaces\n",
    "    step4 = re.sub(r\"\\s{2,}\", r\" \", step3).strip()\n",
    "    # reduce multiple consecutive identical letters into only two letters \n",
    "    step5 = re.sub(r\"([a-z])\\1{2,}\", r\"\\1\\1\", step4)\n",
    "\n",
    "    return step5.split(' ')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "3wT_0No8KIcK"
   },
   "outputs": [],
   "source": [
    "SPECIAL_CASES = ['Book of Rhymes', 'Just the Two of Us']\n",
    "\n",
    "def get_only_eminems_verses(row: pd.core.series.Series) -> str:\n",
    "    \"Removes section headers from the lyrics and returns only Eminem's verses.\"\n",
    "\n",
    "    assert({'title', 'featured', 'lyrics'}.issubset(row.keys()))\n",
    "\n",
    "    # clean the section headers that end with double newline characters\n",
    "    lyrics = re.sub(\"]\\n\\n\", \"]\\n\", row.lyrics)\n",
    "\n",
    "    # songs that has no features (with few exceptions)\n",
    "    if row[['featured']].isnull().values or row['title'] in SPECIAL_CASES:\n",
    "        return re.sub(\"\\[.*\\]\\n\", \"\", lyrics)\n",
    "    \n",
    "    # split lyrics into verses\n",
    "    song_verses = lyrics.split('\\n\\n')\n",
    "\n",
    "    em_verses = \"\"\n",
    "    for verse in song_verses:\n",
    "        # check if the verse has one of the Eminem's section headers\n",
    "        if re.search(\"\\[.*Eminem.*\\]\", verse) or \\\n",
    "           re.search(\"\\[.*Slim Shady.*\\]\", verse) or \\\n",
    "           re.search(\"\\[.*B-Rabbit.*\\]\", verse) or \\\n",
    "           re.search(\"\\[.*M&M.*\\]\", verse):\n",
    "            # append the verse to the previuos\n",
    "            em_verses += re.sub(\"\\[.*\\]\\n\", \"\", verse) + '\\n\\n'\n",
    "\n",
    "    return em_verses"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "da2eHnFOKIcK"
   },
   "source": [
    "### 1.4.2 Contractions and dash/hyphen words <a name=\"142-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "id": "-lLDTwEhKIcL"
   },
   "outputs": [],
   "source": [
    "CONTRACTIONS_TABLE_FILE = settings['CONTRACTIONS_TABLE_PATH'] \n",
    "\n",
    "# load the contraction table from a text file\n",
    "contractions = {}\n",
    "with open(CONTRACTIONS_TABLE_FILE) as f:   \n",
    "    for line in f:\n",
    "        (key, val) = line.split(':')\n",
    "        contractions[key] = val.strip()\n",
    "\n",
    "\n",
    "def edit_contractions(word: str, contraction_table=contractions) -> str:\n",
    "    \"\"\"\n",
    "    Expands a contraction word if the word is in the table, otherwise \n",
    "    removes everthing after the apostrophe. If the word isn't a contraction, \n",
    "    the function just returns it.\n",
    "    \"\"\"\n",
    "\n",
    "    # expand from the table\n",
    "    if word in contraction_table:\n",
    "        return contraction_table[word]\n",
    "        \n",
    "    # otherwise remove everything after the apostrophe\n",
    "    if \"'\" in word:\n",
    "        return re.sub(\"'.*\", \"\", word)\n",
    "    \n",
    "    # return the word if it isn't a contraction\n",
    "    return word"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "id": "_G0hfZcvKIcM"
   },
   "outputs": [],
   "source": [
    "def edit_dash_words(word: str) -> str:\n",
    "    \"\"\"\n",
    "    A custom function for processing dash/hyphen words, specifically \n",
    "    constructed for the words in this project. If the word doesn't have \n",
    "    a dash, the function just returns it.\n",
    "    \"\"\"\n",
    "\n",
    "    # return empty words\n",
    "    if word == '':\n",
    "        return word\n",
    "\n",
    "    # \"delete\" the word if it has a dash character and the length smaller \n",
    "    # than four or more than one dash\n",
    "    if ('-' in word and len(word) < 4) or len(word.split('-')) > 2:\n",
    "        return ''\n",
    "\n",
    "    # remove a dash from the end\n",
    "    if word[-1] == '-':\n",
    "        return word[:-1]\n",
    "\n",
    "    # remove a dash from the second to the last position in the word\n",
    "    if len(word) > 1 and word[-2] == '-':\n",
    "        return word[:-2]\n",
    "        \n",
    "    # if a dash is in the first two characters, remove it\n",
    "    if '-' in word[:2]:\n",
    "        return word.split('-')[1]\n",
    "\n",
    "    return word"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "A-8jyHS6KIcM"
   },
   "outputs": [],
   "source": [
    "d = enchant.Dict(\"en_US\")\n",
    "\n",
    "def expand_ing_contractions(words: Union[str, list]) -> list:\n",
    "    \"\"\"\n",
    "    Expands contraction words that end with \"in\" suffix, by \n",
    "    adding a \"g\" character to the suffix.\n",
    "    \"\"\"\n",
    "\n",
    "    # if the input is a string, put it in an empty list\n",
    "    if type(words) != type([]):\n",
    "        words = [words]\n",
    "\n",
    "    processed = []\n",
    "    for word in words:\n",
    "        # check if the word ends with \"in\" and isn't a valid word already (like \"main\", \"certain\", etc.)\n",
    "        if word.endswith('in') and not d.check(word):\n",
    "            processed.append(word + 'g')\n",
    "        else:\n",
    "            processed.append(word)\n",
    "\n",
    "    return processed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "id": "gf1uNSPOKIcM"
   },
   "outputs": [],
   "source": [
    "# define stopwords\n",
    "ADDITIONAL_STOPWORDS = [' ', '', 'u', 'cannot']\n",
    "mystopwords = stopwords.words('english') + ADDITIONAL_STOPWORDS\n",
    "\n",
    "# init lemmatizer\n",
    "lemmatizer = WordNetLemmatizer() \n",
    "\n",
    "def dash_and_contractions_cleaning(words: Union[str, list], mystopwords=mystopwords) -> list:\n",
    "    \"\"\"\n",
    "    A general custom function for processing words in this project. Includes multiple \n",
    "    previously defined funcitons, like `edit_contractions`, `edit_dash_words` \n",
    "    and `expand_ing_contractions` functions. Removes english stopwords as well.\n",
    "    \"\"\"\n",
    "\n",
    "    # if the input is a string, put it in an empty list\n",
    "    if type(words) != type([]):\n",
    "        words = [words]\n",
    "\n",
    "    processed_words = []\n",
    "    for token in words:\n",
    "        if token in mystopwords or len(token) < 2:\n",
    "            continue\n",
    "\n",
    "        # lemmatizer is mainly used to process plural words,\n",
    "        # this step ends with splitting the expanded word with a whitespace\n",
    "        step1 = edit_contractions(edit_dash_words(lemmatizer.lemmatize(token))).split(' ')\n",
    "        # remove new stopwords\n",
    "        step2 = [word for word in step1 if not word in mystopwords]\n",
    "        step3 = expand_ing_contractions(step2)\n",
    "        \n",
    "        processed_words.extend(step3)\n",
    "\n",
    "    return processed_words if len(processed_words) != 0 else ''"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ixn_Pg9UKIcN"
   },
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "fSF0hfqdKIcO"
   },
   "source": [
    "# 2. Song structure analysis <a name=\"2-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "id": "MpKPluZ9KIcO"
   },
   "outputs": [],
   "source": [
    "# use a special matplotlib style\n",
    "style.use('fivethirtyeight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_v5dH3chKIcP"
   },
   "source": [
    "## 2.1 The average <ins>number</ins> of verses, lines, words, characters and features per song <a name=\"21-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ge4e4yM9KIcQ"
   },
   "source": [
    "### 2.1.1 The average number of verses per song <a name=\"211-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "2jCcn__hKIcQ"
   },
   "source": [
    "#### a) all verses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "hTHdbxPNKIcQ",
    "outputId": "4fde465d-67fa-4318-dbf3-fcbe238ef5bd"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 4.38\n",
      "std: 3.47 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# splits lyrics into verses\n",
    "count_verses = lambda lyrics: len(lyrics.split('\\n\\n'))\n",
    "\n",
    "# total number of verses in each song\n",
    "verses_count = data_df.lyrics.apply(count_verses)\n",
    "\n",
    "# show results\n",
    "print('mean: %.2f' % verses_count.mean())\n",
    "print('std: %.2f' % verses_count.std(), '\\n')\n",
    "\n",
    "verses_count.plot.hist(bins=35, rwidth=0.89, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7qYNxUy0KIcR"
   },
   "source": [
    "#### b) unique verses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "v6iKgvRGKIcR",
    "outputId": "201c27b0-30e2-4962-ab78-5ed47df10d37"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 4.31\n",
      "std: 3.39 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# splits lyrics into verses and converts the list into a set to remove the duplicates\n",
    "count_unique_verses = lambda lyrics: len(set(lyrics.split('\\n\\n')))\n",
    "\n",
    "# total number of unique verses in each song\n",
    "unique_verses_count = data_df.lyrics.apply(count_unique_verses)\n",
    "\n",
    "# show results\n",
    "print('mean: %.2f' % unique_verses_count.mean())\n",
    "print('std: %.2f' % unique_verses_count.std(), '\\n')\n",
    "\n",
    "unique_verses_count.plot.hist(bins=35, rwidth=0.89, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Jx_MD33lKIcS"
   },
   "source": [
    "#### > show outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "RWN84yjhKIcT",
    "outputId": "7b26f2c9-f01c-4e41-9d46-bb88baa87138"
   },
   "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>artist</th>\n",
       "      <th>featured</th>\n",
       "      <th>release_date</th>\n",
       "      <th>verses</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>320</th>\n",
       "      <td>First Word Freestyle</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>NaT</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>364</th>\n",
       "      <td>MC Hammer vs. Vanilla Ice</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Proof]</td>\n",
       "      <td>1990-01-01</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>340</th>\n",
       "      <td>2004 Tim Westwood Freestyle</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Stat Quo, Tim Westwood, D12]</td>\n",
       "      <td>NaT</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>241</th>\n",
       "      <td>Dr. West (Skit)</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Dominic West]</td>\n",
       "      <td>2009-05-15</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>304</th>\n",
       "      <td>Just Rhymin’ with Proof</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Proof]</td>\n",
       "      <td>2000-12-01</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           title  artist                       featured  \\\n",
       "320         First Word Freestyle  Eminem                           None   \n",
       "364    MC Hammer vs. Vanilla Ice  Eminem                        [Proof]   \n",
       "340  2004 Tim Westwood Freestyle  Eminem  [Stat Quo, Tim Westwood, D12]   \n",
       "241              Dr. West (Skit)  Eminem                 [Dominic West]   \n",
       "304      Just Rhymin’ with Proof  Eminem                        [Proof]   \n",
       "\n",
       "    release_date  verses  \n",
       "320          NaT      31  \n",
       "364   1990-01-01      26  \n",
       "340          NaT      26  \n",
       "241   2009-05-15      25  \n",
       "304   2000-12-01      21  "
      ]
     },
     "execution_count": 17,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# find outliers\n",
    "N_TOP = 5\n",
    "outliers_idxs = ( np.abs( unique_verses_count - unique_verses_count.mean() ) ).argsort()[-N_TOP:][::-1]\n",
    "\n",
    "# show results\n",
    "pd.concat(objs=[data_df.iloc[outliers_idxs][['title', 'artist', 'featured', 'release_date']], \n",
    "                unique_verses_count[outliers_idxs]], \n",
    "          axis=1) \\\n",
    "  .rename(columns={'lyrics':'verses'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1Xrc3URuKIcT"
   },
   "source": [
    "### 2.1.2 The average number of lines per song <a name=\"212-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "pOXxHGEBKIcU"
   },
   "source": [
    "#### a) all lines"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "ITT-3gVWKIcV",
    "outputId": "e583a5d7-9789-452b-8aa3-253fb33b3a52"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 78.34\n",
      "std: 54.28 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# total number of lines in each song\n",
    "lines_count = data_df.lyrics.apply(lambda lyrics: len(split_lyrics_into_lines(lyrics)))\n",
    "\n",
    "# show results\n",
    "print('mean: %.2f' % lines_count.mean())\n",
    "print('std: %.2f' % lines_count.std(), '\\n')\n",
    "\n",
    "lines_count.plot.hist(bins=80, rwidth=0.7, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "nypW4YndKIcW"
   },
   "source": [
    "#### b) unique lines"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "7PFKnAJtKIcW",
    "outputId": "5a6b7725-bf54-413d-d97a-6c4bcffe6778"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 67.99\n",
      "std: 51.07 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# total number of unique lines in each song\n",
    "unique_lines_count = data_df.lyrics.apply(lambda lyrics: len(set(split_lyrics_into_lines(lyrics))))\n",
    "\n",
    "# show results\n",
    "print('mean: %.2f' % unique_lines_count.mean())\n",
    "print('std: %.2f' % unique_lines_count.std(), '\\n')\n",
    "\n",
    "unique_lines_count.plot.hist(bins=80, rwidth=0.7, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "XrQbynatKIcX"
   },
   "source": [
    "#### > show outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "U6UM7YkCKIcX",
    "outputId": "edc340c2-0a6d-48f2-ea62-c96a3102993a"
   },
   "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>artist</th>\n",
       "      <th>featured</th>\n",
       "      <th>release_date</th>\n",
       "      <th>lines</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>340</th>\n",
       "      <td>2004 Tim Westwood Freestyle</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Stat Quo, Tim Westwood, D12]</td>\n",
       "      <td>NaT</td>\n",
       "      <td>781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>Eminem Freestyles on Tim Westwood | 2009</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[​dEnAun]</td>\n",
       "      <td>2009-05-27</td>\n",
       "      <td>246</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>190</th>\n",
       "      <td>Kick Off (Freestyle)</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2018-11-30</td>\n",
       "      <td>202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>334</th>\n",
       "      <td>Baruch College Session</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>1997-01-01</td>\n",
       "      <td>202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>Campaign Speech</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2016-10-19</td>\n",
       "      <td>187</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                        title  artist  \\\n",
       "340               2004 Tim Westwood Freestyle  Eminem   \n",
       "121  Eminem Freestyles on Tim Westwood | 2009  Eminem   \n",
       "190                      Kick Off (Freestyle)  Eminem   \n",
       "334                    Baruch College Session  Eminem   \n",
       "56                            Campaign Speech  Eminem   \n",
       "\n",
       "                          featured release_date  lines  \n",
       "340  [Stat Quo, Tim Westwood, D12]          NaT    781  \n",
       "121                      [​dEnAun]   2009-05-27    246  \n",
       "190                           None   2018-11-30    202  \n",
       "334                           None   1997-01-01    202  \n",
       "56                            None   2016-10-19    187  "
      ]
     },
     "execution_count": 22,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# find outliers\n",
    "N_TOP = 5\n",
    "outliers_idxs = ( np.abs( unique_lines_count - unique_lines_count.mean() ) ).argsort()[-N_TOP:][::-1]\n",
    "\n",
    "# show results\n",
    "pd.concat(objs=[data_df.iloc[outliers_idxs][['title', 'artist', 'featured', 'release_date']], \n",
    "                unique_lines_count[outliers_idxs]], \n",
    "          axis=1) \\\n",
    "  .rename(columns={'lyrics':'lines'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "-VJLz4fEKIcY"
   },
   "source": [
    "### 2.1.3 The average number of words per song <a name=\"213-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1lDp_qH0KIcY"
   },
   "source": [
    "#### a) all words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "DgvX1tyYKIca",
    "outputId": "cde3441b-2176-40e1-dc47-748370a2aa59"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 675.10\n",
      "std: 403.67 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# total number of words in each song\n",
    "words_count = data_df.lyrics.apply(lambda lyrics: len(split_lyrics_into_words(lyrics)))\n",
    "\n",
    "# show results\n",
    "print('mean: %.2f' % words_count.mean())\n",
    "print('std: %.2f' % words_count.std(), '\\n')\n",
    "\n",
    "words_count.plot.hist(bins=80, rwidth=0.7, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "hGB3nJhNKIca"
   },
   "source": [
    "#### b) unique words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "CD5wXmSHKIcb",
    "outputId": "63b4c904-f9f7-4cf7-c0f1-818f6bf8deac"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 283.57\n",
      "std: 148.14 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# total number of unique words in each song\n",
    "unique_words_count = data_df.lyrics.apply(lambda lyrics: len(set(split_lyrics_into_words(lyrics))))\n",
    "\n",
    "# show results\n",
    "print('mean: %.2f' % unique_words_count.mean())\n",
    "print('std: %.2f' % unique_words_count.std(), '\\n')\n",
    "\n",
    "unique_words_count.plot.hist(bins=80, rwidth=0.7, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "mOPphr15KIcc"
   },
   "source": [
    "#### > show outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "MPFLZI2PKIcc",
    "outputId": "a49a3053-905d-4b89-f0b8-4d9af6640a4a"
   },
   "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>artist</th>\n",
       "      <th>featured</th>\n",
       "      <th>release_date</th>\n",
       "      <th>words</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>340</th>\n",
       "      <td>2004 Tim Westwood Freestyle</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Stat Quo, Tim Westwood, D12]</td>\n",
       "      <td>NaT</td>\n",
       "      <td>1253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>Eminem Freestyles on Tim Westwood | 2009</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[​dEnAun]</td>\n",
       "      <td>2009-05-27</td>\n",
       "      <td>870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>190</th>\n",
       "      <td>Kick Off (Freestyle)</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2018-11-30</td>\n",
       "      <td>843</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>Campaign Speech</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2016-10-19</td>\n",
       "      <td>782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>277</th>\n",
       "      <td>Em360 Rapcity Backroom Freestyle</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Big Tigger, Slaughterhouse]</td>\n",
       "      <td>2013-11-06</td>\n",
       "      <td>729</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>334</th>\n",
       "      <td>Baruch College Session</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>1997-01-01</td>\n",
       "      <td>719</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                        title  artist  \\\n",
       "340               2004 Tim Westwood Freestyle  Eminem   \n",
       "121  Eminem Freestyles on Tim Westwood | 2009  Eminem   \n",
       "190                      Kick Off (Freestyle)  Eminem   \n",
       "56                            Campaign Speech  Eminem   \n",
       "277          Em360 Rapcity Backroom Freestyle  Eminem   \n",
       "334                    Baruch College Session  Eminem   \n",
       "\n",
       "                          featured release_date  words  \n",
       "340  [Stat Quo, Tim Westwood, D12]          NaT   1253  \n",
       "121                      [​dEnAun]   2009-05-27    870  \n",
       "190                           None   2018-11-30    843  \n",
       "56                            None   2016-10-19    782  \n",
       "277   [Big Tigger, Slaughterhouse]   2013-11-06    729  \n",
       "334                           None   1997-01-01    719  "
      ]
     },
     "execution_count": 25,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# find outliers\n",
    "N_TOP = 6\n",
    "outliers_idxs = ( np.abs( unique_words_count - unique_words_count.mean() ) ).argsort()[-N_TOP:][::-1]\n",
    "\n",
    "# show results\n",
    "pd.concat(objs=[data_df.iloc[outliers_idxs][['title', 'artist', 'featured', 'release_date']], \n",
    "                unique_words_count[outliers_idxs]], \n",
    "          axis=1) \\\n",
    "  .rename(columns={'lyrics':'words'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Sy0rnt7kKIcd"
   },
   "source": [
    "### 2.1.4 The average number of characters per song <a name=\"214-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "1aPnkuMiKIce",
    "outputId": "0c4e32cb-d1ec-45ec-871a-1ebe91338a81"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 3569.95\n",
      "std: 2107.28 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# total number of chars in each song\n",
    "chars_count = data_df.lyrics.apply(lambda lyrics: len(lyrics))\n",
    "\n",
    "# show results\n",
    "print('mean: %.2f' % chars_count.mean())\n",
    "print('std: %.2f' % chars_count.std(), '\\n')\n",
    "\n",
    "chars_count.plot.hist(bins=80, rwidth=0.7, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Rb-6sBD9KIce"
   },
   "source": [
    "#### > show outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "u4kUXtWzKIcf",
    "outputId": "0cf8d240-1760-441c-c7e4-d613e30c0e35"
   },
   "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>artist</th>\n",
       "      <th>featured</th>\n",
       "      <th>release_date</th>\n",
       "      <th>chars</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>340</th>\n",
       "      <td>2004 Tim Westwood Freestyle</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Stat Quo, Tim Westwood, D12]</td>\n",
       "      <td>NaT</td>\n",
       "      <td>20444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>Eminem Freestyles on Tim Westwood | 2009</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[​dEnAun]</td>\n",
       "      <td>2009-05-27</td>\n",
       "      <td>12544</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>190</th>\n",
       "      <td>Kick Off (Freestyle)</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2018-11-30</td>\n",
       "      <td>9484</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>277</th>\n",
       "      <td>Em360 Rapcity Backroom Freestyle</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Big Tigger, Slaughterhouse]</td>\n",
       "      <td>2013-11-06</td>\n",
       "      <td>8872</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>334</th>\n",
       "      <td>Baruch College Session</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>1997-01-01</td>\n",
       "      <td>8604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>Campaign Speech</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2016-10-19</td>\n",
       "      <td>8156</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                        title  artist  \\\n",
       "340               2004 Tim Westwood Freestyle  Eminem   \n",
       "121  Eminem Freestyles on Tim Westwood | 2009  Eminem   \n",
       "190                      Kick Off (Freestyle)  Eminem   \n",
       "277          Em360 Rapcity Backroom Freestyle  Eminem   \n",
       "334                    Baruch College Session  Eminem   \n",
       "56                            Campaign Speech  Eminem   \n",
       "\n",
       "                          featured release_date  chars  \n",
       "340  [Stat Quo, Tim Westwood, D12]          NaT  20444  \n",
       "121                      [​dEnAun]   2009-05-27  12544  \n",
       "190                           None   2018-11-30   9484  \n",
       "277   [Big Tigger, Slaughterhouse]   2013-11-06   8872  \n",
       "334                           None   1997-01-01   8604  \n",
       "56                            None   2016-10-19   8156  "
      ]
     },
     "execution_count": 45,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# find outliers\n",
    "N_TOP = 6\n",
    "outliers_idxs = ( np.abs( chars_count - chars_count.mean() ) ).argsort()[-N_TOP:][::-1]\n",
    "\n",
    "# show results\n",
    "pd.concat(objs=[data_df.iloc[outliers_idxs][['title', 'artist', 'featured', 'release_date']], \n",
    "                chars_count[outliers_idxs]], \n",
    "          axis=1) \\\n",
    "  .rename(columns={'lyrics':'chars'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "5asVcx2WKIcg"
   },
   "source": [
    "### 2.1.5 The average number of features (collaborations) per song <a name=\"215-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "_n7NjRj9KIcg",
    "outputId": "75f990f2-1933-40b2-92cb-7564ea8a8bce"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 0.44\n",
      "std: 0.81 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# number of featured artists in each song\n",
    "features_count = data_df.featured.apply(lambda featured: 0 if featured is None else len(featured))\n",
    "\n",
    "# show results\n",
    "print('mean: %.2f' % features_count.mean())\n",
    "print('std: %.2f' % features_count.std(), '\\n')\n",
    "\n",
    "features_count.plot.hist(bins=5, rwidth=0.98, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Hxv5XOvhKIch"
   },
   "source": [
    "#### > show outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "Q7d8JhPhKIch",
    "outputId": "11c45fd6-f721-4812-d1c5-021e3664a57f"
   },
   "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>artist</th>\n",
       "      <th>featured</th>\n",
       "      <th>release_date</th>\n",
       "      <th>featured</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>174</th>\n",
       "      <td>My Syllables</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[JAY-Z, Dr. Dre, 50 Cent, Stat Quo, Ca$his]</td>\n",
       "      <td>NaT</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>Detroit vs. Everybody</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Danny Brown, Trick Trick, Royce da 5'9\", DeJ ...</td>\n",
       "      <td>2014-11-11</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>Yah Yah</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Royce da 5'9\", Black Thought, Q-Tip, ​dEnAun]</td>\n",
       "      <td>2020-01-17</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>290</th>\n",
       "      <td>We’re Back</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Obie Trice, Stat Quo, Bobby Creekwater, Ca$his]</td>\n",
       "      <td>2006-12-05</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82</th>\n",
       "      <td>Bitch Please II</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Dr. Dre, Snoop Dogg, Nate Dogg, Xzibit]</td>\n",
       "      <td>2000-05-23</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>218</th>\n",
       "      <td>8 Mile: 313 Cypher (Chin Tiki Girls)</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Omar Benson Miller, Mekhi Phifer, Obie Trice,...</td>\n",
       "      <td>2002-11-09</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                    title  artist  \\\n",
       "174                          My Syllables  Eminem   \n",
       "46                  Detroit vs. Everybody  Eminem   \n",
       "99                                Yah Yah  Eminem   \n",
       "290                            We’re Back  Eminem   \n",
       "82                        Bitch Please II  Eminem   \n",
       "218  8 Mile: 313 Cypher (Chin Tiki Girls)  Eminem   \n",
       "\n",
       "                                              featured release_date  featured  \n",
       "174        [JAY-Z, Dr. Dre, 50 Cent, Stat Quo, Ca$his]          NaT         5  \n",
       "46   [Danny Brown, Trick Trick, Royce da 5'9\", DeJ ...   2014-11-11         5  \n",
       "99      [Royce da 5'9\", Black Thought, Q-Tip, ​dEnAun]   2020-01-17         4  \n",
       "290   [Obie Trice, Stat Quo, Bobby Creekwater, Ca$his]   2006-12-05         4  \n",
       "82            [Dr. Dre, Snoop Dogg, Nate Dogg, Xzibit]   2000-05-23         4  \n",
       "218  [Omar Benson Miller, Mekhi Phifer, Obie Trice,...   2002-11-09         4  "
      ]
     },
     "execution_count": 29,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# find outliers\n",
    "N_TOP = 6\n",
    "outliers_idxs = ( np.abs( features_count - features_count.mean() ) ).argsort()[-N_TOP:][::-1]\n",
    "\n",
    "# show results\n",
    "pd.concat(objs=[data_df.iloc[outliers_idxs][['title', 'artist', 'featured', 'release_date']], \n",
    "                features_count[outliers_idxs]], \n",
    "          axis=1) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1wxEzgc0KIcj"
   },
   "source": [
    "## 2.2 The average <ins>length</ins> of a verse, line and word <a name=\"22-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "-ZxYw59oKIcj"
   },
   "source": [
    "### 2.2.1 The average number of lines in a verse <a name=\"221-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "qGYePKfCKIcj"
   },
   "outputs": [],
   "source": [
    "def get_avg_verse_length(lyrics):\n",
    "    \"\"\"Calculates the average length of verses in `lyrics`.\"\"\"\n",
    "\n",
    "    # remove section headers and split lyrics into verses\n",
    "    verses = re.sub(\"\\[.*\\]\\n\", \"\", lyrics).split('\\n\\n')\n",
    "    # calculate the length for each verse in the lyrics\n",
    "    verses_lenghts = [len(verse.split('\\n')) for verse in verses]\n",
    "    # return the average length of verses\n",
    "    return np.mean(verses_lenghts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "p61T4XmwKIcj",
    "outputId": "a00ad4c7-4d69-4c12-f4b2-fedcba50c82a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 22.06\n",
      "std: 16.00 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# average number of lines in verses per song\n",
    "avg_len_of_verse = data_df.lyrics.apply(get_avg_verse_length)\n",
    "\n",
    "# show results\n",
    "print('mean: %.2f' % avg_len_of_verse.mean())\n",
    "print('std: %.2f' % avg_len_of_verse.std(), '\\n')\n",
    "\n",
    "avg_len_of_verse.plot.hist(bins=100, rwidth=0.70, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "iRG0HA3QKIck"
   },
   "source": [
    "#### > show outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "DhAsolC-KIck",
    "outputId": "49f788ad-3d59-4356-ca7d-072dbe65a8ad"
   },
   "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>artist</th>\n",
       "      <th>featured</th>\n",
       "      <th>release_date</th>\n",
       "      <th>lines per verse</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>Detroit vs. Everybody</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Danny Brown, Trick Trick, Royce da 5'9\", DeJ ...</td>\n",
       "      <td>2014-11-11</td>\n",
       "      <td>164.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100</th>\n",
       "      <td>Still Shady</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2014-11-24</td>\n",
       "      <td>132.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>Campaign Speech</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2016-10-19</td>\n",
       "      <td>93.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86</th>\n",
       "      <td>Arose</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2017-12-15</td>\n",
       "      <td>89.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102</th>\n",
       "      <td>Castle</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2017-12-15</td>\n",
       "      <td>87.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>Bad Guy</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2013-11-05</td>\n",
       "      <td>82.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     title  artist  \\\n",
       "46   Detroit vs. Everybody  Eminem   \n",
       "100            Still Shady  Eminem   \n",
       "56         Campaign Speech  Eminem   \n",
       "86                   Arose  Eminem   \n",
       "102                 Castle  Eminem   \n",
       "19                 Bad Guy  Eminem   \n",
       "\n",
       "                                              featured release_date  \\\n",
       "46   [Danny Brown, Trick Trick, Royce da 5'9\", DeJ ...   2014-11-11   \n",
       "100                                               None   2014-11-24   \n",
       "56                                                None   2016-10-19   \n",
       "86                                                None   2017-12-15   \n",
       "102                                               None   2017-12-15   \n",
       "19                                                None   2013-11-05   \n",
       "\n",
       "     lines per verse  \n",
       "46             164.0  \n",
       "100            132.0  \n",
       "56              93.5  \n",
       "86              89.0  \n",
       "102             87.0  \n",
       "19              82.0  "
      ]
     },
     "execution_count": 32,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# find outliers\n",
    "N_TOP = 6\n",
    "outliers_idxs = ( np.abs( avg_len_of_verse - avg_len_of_verse.mean() ) ).argsort()[-N_TOP:][::-1]\n",
    "\n",
    "# show results\n",
    "pd.concat(objs=[data_df.iloc[outliers_idxs][['title', 'artist', 'featured', 'release_date']], \n",
    "                avg_len_of_verse[outliers_idxs]], \n",
    "          axis=1)  \\\n",
    "  .rename(columns={'lyrics':'lines per verse'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "h9Xjx-30KIck"
   },
   "source": [
    "### 2.2.2 The average number of words in a line  <a name=\"222-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "fTC4bOSKKIcl",
    "outputId": "f9d34a5e-00f2-479c-a048-6f00488a80ba"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 9.00\n",
      "std: 5.67 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# average number of words in a line per song\n",
    "avg_len_of_line = data_df.lyrics.apply(lambda lyrics: np.mean([len(line.split(' ')) for line in split_lyrics_into_lines(lyrics)]))\n",
    "\n",
    "# show results\n",
    "print('mean: %.2f' % avg_len_of_line.mean())\n",
    "print('std: %.2f' % avg_len_of_line.std(), '\\n')\n",
    "\n",
    "avg_len_of_line.plot.hist(bins=70, rwidth=0.70, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "5XhkBhquKIcm"
   },
   "source": [
    "#### > show outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "7WBLfln4KIcm",
    "outputId": "86a273bf-854f-48d8-b3c1-2c73639322c0"
   },
   "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>artist</th>\n",
       "      <th>featured</th>\n",
       "      <th>release_date</th>\n",
       "      <th>words per line</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>412</th>\n",
       "      <td>Fuck Obie (Skit)</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2006-08-12</td>\n",
       "      <td>10.538462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>348</th>\n",
       "      <td>Power 92.3 Freestyle</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>NaT</td>\n",
       "      <td>7.878788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>309</th>\n",
       "      <td>1-833-2GET-REV (REVIVAL Voicemail)</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2017-10-25</td>\n",
       "      <td>61.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>409</th>\n",
       "      <td>Whoo Kid Freestyle</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>NaT</td>\n",
       "      <td>8.419355</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>310</th>\n",
       "      <td>Curtains Up (Skit) [2004]</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2004-11-12</td>\n",
       "      <td>3.384615</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  title  artist featured release_date  \\\n",
       "412                    Fuck Obie (Skit)  Eminem     None   2006-08-12   \n",
       "348                Power 92.3 Freestyle  Eminem     None          NaT   \n",
       "309  1-833-2GET-REV (REVIVAL Voicemail)  Eminem     None   2017-10-25   \n",
       "409                  Whoo Kid Freestyle  Eminem     None          NaT   \n",
       "310           Curtains Up (Skit) [2004]  Eminem     None   2004-11-12   \n",
       "\n",
       "     words per line  \n",
       "412       10.538462  \n",
       "348        7.878788  \n",
       "309       61.000000  \n",
       "409        8.419355  \n",
       "310        3.384615  "
      ]
     },
     "execution_count": 36,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# find outliers\n",
    "N_TOP = 5\n",
    "outliers_idxs = ( np.abs( avg_len_of_line - avg_len_of_line.mean() ) ).argsort()[-N_TOP:][::-1]\n",
    "\n",
    "# show results\n",
    "pd.concat(objs=[data_df.iloc[outliers_idxs][['title', 'artist', 'featured', 'release_date']], \n",
    "                avg_len_of_line[outliers_idxs]], \n",
    "          axis=1)  \\\n",
    "  .rename(columns={'lyrics':'words per line'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "cym1f15zKIcn"
   },
   "source": [
    "### 2.2.3 The average length of a word (not cleaned) <a name=\"223-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 4.00\n",
      "std: 2.05 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# get length of each word in the corpus\n",
    "all_words = np.concatenate(data_df.lyrics.apply(split_lyrics_into_words))\n",
    "all_words_len = np.vectorize(len)(all_words)\n",
    "\n",
    "# show results\n",
    "print('mean: %.2f' % np.mean(all_words_len))\n",
    "print('std: %.2f' % np.std(all_words_len), '\\n')\n",
    "\n",
    "pd.DataFrame(all_words_len).plot.hist(bins=50, rwidth=0.70, alpha=0.5, legend=False)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "eQ5utIh5KIcx"
   },
   "source": [
    "#### > show outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "_gWmB5EUKIcy",
    "outputId": "560fdab1-087d-40dc-faf1-5253e06baae3"
   },
   "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>word</th>\n",
       "      <th>length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>antidisestablishmentarianism</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>pooaaooaaooaaooaaooaooassee</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>sensitivitytoothpaintight</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>fingernailsdramatically</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>hahahahahahahahahahaha</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>balalalalalalalalalala</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>achecoughshortness</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>womenragedecreased</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>appetiteincreased</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>responsibilities</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>extraterrestrial</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>hydrochloroquine</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>allergiesringing</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                            word length\n",
       "0   antidisestablishmentarianism     28\n",
       "1    pooaaooaaooaaooaaooaooassee     27\n",
       "2      sensitivitytoothpaintight     25\n",
       "3        fingernailsdramatically     23\n",
       "4         hahahahahahahahahahaha     22\n",
       "5         balalalalalalalalalala     22\n",
       "6             achecoughshortness     18\n",
       "7             womenragedecreased     18\n",
       "8              appetiteincreased     17\n",
       "9               responsibilities     16\n",
       "10              extraterrestrial     16\n",
       "11              hydrochloroquine     16\n",
       "12              allergiesringing     16"
      ]
     },
     "execution_count": 38,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# find outliers\n",
    "N_TOP = 130\n",
    "outliers_idxs = ( np.abs( all_words_len - np.mean(all_words_len) ) ).argsort()[-N_TOP:][::-1]\n",
    "\n",
    "# get outlier words\n",
    "outlier_words = all_words[outliers_idxs]\n",
    "# filter dash/hyphen words\n",
    "no_dash_outlier_words_idxs = np.vectorize(lambda word: not '-' in word)(outlier_words)\n",
    "\n",
    "# show results\n",
    "words_len_df = pd.DataFrame([all_words, all_words_len], index=['word', 'length']).T\n",
    "words_len_df.iloc[outliers_idxs][no_dash_outlier_words_idxs] \\\n",
    "            .drop_duplicates() \\\n",
    "            .reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "YgGmtkqpKIcy"
   },
   "source": [
    "## 2.3 Summary <a name=\"23-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "SLXyYeqxKIcz"
   },
   "source": [
    "**I.** The average **NUMBER** of:\n",
    "- **VERSES** per song: \n",
    "    - *all verses*:\n",
    "        - mean: **4.38**\n",
    "        - std: **3.47**\n",
    "    - *unique verses*: \n",
    "        - mean: **4.31**\n",
    "        - std: **3.39**\n",
    "- **LINES** per song:\n",
    "    - *all lines*:\n",
    "        - mean: **78.34**\n",
    "        - std: **54.28**\n",
    "    - *unique lines*:\n",
    "        - mean: **67.99**\n",
    "        - std: **51.07**\n",
    "- **WORDS** per song:\n",
    "    - *all words*:\n",
    "        - mean: **675.10**\n",
    "        - std: **403.67**\n",
    "    - *unique words*:\n",
    "        - mean: **283.57**\n",
    "        - std: **148.44**\n",
    "        \n",
    "        \n",
    "- **CHARACTERS** per song:\n",
    "    - mean: **3569.95**\n",
    "    - std: **2107.8**\n",
    "- **FEATURES** per song:\n",
    "    - mean: **0.44**\n",
    "    - std: **0.81** <br><br>\n",
    "    \n",
    "    \n",
    "**II.** The average **LENGTH** of:\n",
    "- **VERSE** (in lines):\n",
    "    - mean: **22.06**\n",
    "    - std: **16.00**\n",
    "- **LINE** (in words):\n",
    "    - mean: **9.00**\n",
    "    - std: **5.67**\n",
    "- **WORD**:\n",
    "    - mean: **4.00**\n",
    "    - std: **2.05**\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "FUn9enE0KIc0"
   },
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "JvD6s6aoKIc0"
   },
   "source": [
    "# 3. Vocabulary analysis <a name=\"3-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "DE6aTt5zKIc0"
   },
   "outputs": [],
   "source": [
    "# use a special matplotlib style\n",
    "style.use('fivethirtyeight')\n",
    "\n",
    "# get only Eminem's verses from each song\n",
    "em_verses = data_df.apply(get_only_eminems_verses, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "3pEDP6ktKIc1"
   },
   "source": [
    "## 3.1 The number of unique words <a name=\"31-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>\n",
    "\n",
    "This could be seen as Eminem's vocabulary. However, the number obtained in this analysis should be probably somewhat smaller, mostly becuase it includes various interjection sounds (like \"agh\", \"uh\", \"haha\"), different verb tenses, words from unpublished songs and freestyles, people's names, misspelled words, etc., all recognized as unique words. \n",
    "\n",
    "**The reference number**: <br>\n",
    "The size of Eminem's vocabulary measured in [this](https://www.billboard.com/articles/columns/the-juice/6642050/eminem-largest-vocabulary-music-industry) article (**8,818 words - 24/07/2015**) can be used for the comparison of obtained results, mostly to see how close or off is the analysis."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Kt-mhrlPKIc1"
   },
   "source": [
    "### a) Total words said"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "3XiTDNXAKIc1",
    "outputId": "ce3a9494-2ca4-41a8-98d8-8f07cccb6386"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "123994"
      ]
     },
     "execution_count": 18,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_words = np.concatenate(em_verses.apply(split_lyrics_into_words).apply(dash_and_contractions_cleaning))\n",
    "len(all_words)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_ByaHcH3KIc2"
   },
   "source": [
    "### b) Unique words said"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "AKXsr-9zKIc2",
    "outputId": "362d26a7-16d5-43e9-878b-47552e2c22b3"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "13320"
      ]
     },
     "execution_count": 57,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(set(all_words))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "cKaofI7KKIc3"
   },
   "source": [
    "As expected, there is a lot more words than mentioned in the article, but still this should be in the acceptable range, especially considering six years (2015-2021) of difference between the two.\n",
    "\n",
    "Still, just to see how many new words Eminem has added to his vocabulary in the last six years (according to this dataset, and with some gibberish words included as well), the following analysis is performed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "ypeo0nUTKIc3",
    "outputId": "d243b900-4e9f-4044-a50b-c4c9593feeda"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11174"
      ]
     },
     "execution_count": 58,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# get words from the songs before 2015, including the songs with no release date (mostly his earlier \n",
    "# unpublished songs and freestyles)\n",
    "RESEARCH_DATE = '2015-06-1'       \n",
    "all_words_before_2015 =  data_df[(data_df.release_date <= RESEARCH_DATE) | (data_df.release_date.isnull())] \\\n",
    "                            .apply(get_only_eminems_verses, axis=1) \\\n",
    "                            .apply(split_lyrics_into_words) \\\n",
    "                            .apply(dash_and_contractions_cleaning)\n",
    "\n",
    "# create one list with all words (including duplicates)\n",
    "all_words_before_2015 = np.concatenate(all_words_before_2015.values)\n",
    "\n",
    "# number of unique words before 2015\n",
    "len(set(all_words_before_2015))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "fLDKNFJIKIc4"
   },
   "source": [
    "which differs by... from the number mentioned in the article."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "Ku_IoL7AKIc5",
    "outputId": "b02d9b3d-393c-44a5-f405-93538747d171"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2356"
      ]
     },
     "execution_count": 59,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(set(all_words_before_2015)) - 8818"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "08b_Aiq3KIc5"
   },
   "source": [
    "And the number of new unique words after 2015 is..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "URhFUwPqKIc6",
    "outputId": "6c4d2620-3500-4209-8190-d736980e0aef"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2146"
      ]
     },
     "execution_count": 60,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(set(all_words)) - len(set(all_words_before_2015))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "AAfK6kgmKIc6"
   },
   "source": [
    "## 3.2 The average ratio of unique words per song <a name=\"32-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "O51F2AnBKIc6"
   },
   "source": [
    "Calculating the average ratio of unique words in lyrics, including each song in the corpus.\n",
    "\n",
    "The equation used to calculate the ratio is as follows:\n",
    "\n",
    "$$ UWR\\ (lyrics)\\ =\\text{ ... }\\ = \\frac{N_\\text{unique words}}{N_\\text{all words}},\\qquad UWR\\ =\\text{unique words ratio}$$\n",
    "\n",
    ", where *N* is the number of words in the lyrics."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "t2Zp4unhKIc7",
    "outputId": "2697c2c3-546e-4a78-e9fb-981000e38f50"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 0.68\n",
      "std: 0.13 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAboAAAEJCAYAAADiqeJeAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAg+ElEQVR4nO3de1DU5f4H8PeCOhKiKC2LhWDJxm1SgRKFOgaeVOQkUlBYZtEdsQOZoGtnykzPgkJHJNxQdFLTk0JodvVkUBqhON5Ls/UCI44sAqKsQQbs7w/H/bkhynf57oUv79eMM/I8D/t8vs/AvHl2vxdZY2OjAURERBLlYOsCiIiILIlBR0REksagIyIiSWPQERGRpDHoiIhI0hh0REQkaQw6IiKSNAYdERFJGoNOgrRara1L6DG4VsJwvbqOayWMJdeLQUdERJLGoCMiIklj0BERkaQx6IiISNIYdEREJGkMOiIikjQGHRERSRqDjoiIJK2PrQsgImlRH7xstblUQQOtNhf1XNzRERGRpDHoiIhI0hh0REQkaQw6IiKSNAYdERFJGoOOiIgkjUFHRESSZtOgKysrQ0JCAvz9/eHq6oqNGzea9BsMBqjVavj5+cHDwwPR0dE4fvy4yZg//vgDaWlpuPfee3HXXXchISEB586ds+ZhEBGRHbNp0F25cgUBAQHIyMiAk5NTh/6cnBzk5eUhMzMTJSUlkMvliI2NRVNTk3GMSqXC559/jjVr1uCrr75CU1MTnnrqKbS1tVnzUIiIyE7ZNOgmTpyIt99+GzExMXBwMC3FYDBAo9EgNTUVMTExCAgIgEajgV6vR1FREQDg0qVL2LBhAxYtWoSIiAiMHj0a+fn5+OWXX/D999/b4IiIiMje2O1ndFVVVdDpdIiMjDS2OTk5ISwsDHv37gUAHDp0CH/++afJGE9PT/j6+hrHEBFR72a3QafT6QAAcrncpF0ul6O2thYAUFtbC0dHR7i5uXU6hoiIeje7v6mzTCYz+dpgMHRo+6vbjdFqtaLUZs96wzGKhWslzO3Wq6G+r5UqAbRandXmMgd/toQxd72USuUt++026BQKBYBruzZPT09je11dnXGX5+7ujra2NtTX1+POO+80GRMWFtbpa99uUXo6rVYr+WMUC9dKmK6s1xC99Z5eoFTa79ML+LMljCXXy27fuvT29oZCoUBpaamxraWlBeXl5QgNDQUAjB49Gn379jUZc+7cOZw4ccI4hoiIejeb7uj0ej1Onz4NAGhvb0d1dTWOHDmCwYMHY9iwYUhKSkJ2djaUSiV8fHyQlZUFZ2dnxMXFAQAGDRqEZ599Fm+//TbkcjkGDx6Mt956C4GBgXjkkUdseGRERGQvbBp0Bw8exGOPPWb8Wq1WQ61WY/r06dBoNEhJSUFzczPS0tLQ2NiIkJAQFBcXw8XFxfg9//73v+Ho6IjExES0tLTgb3/7Gz788EM4Ojra4pCIiMjOyBobGw22LoLExc8Guo5rJUxX1otPGL+GP1vC9MrP6IiIiMTAoCMiIklj0BERkaQx6IiISNIYdEREJGkMOiIikjQGHRERSRqDjoiIJI1BR0REksagIyIiSWPQERGRpDHoiIhI0hh0REQkaXb7hHEi6j6xnyTQUN+30yeI2/OTBKh3446OiIgkjUFHRESSxqAjIiJJY9AREZGkMeiIiEjSGHRERCRpDDoiIpI0Bh0REUkag46IiCSNQUdERJLGoCMiIklj0BERkaQx6IiISNIYdEREJGkMOiIikjQGHRERSRqDjoiIJM2ug66trQ2LFy/GyJEjoVAoMHLkSCxevBitra3GMQaDAWq1Gn5+fvDw8EB0dDSOHz9uw6qJiMie2HXQLV++HAUFBcjMzERFRQUyMjKwevVqvP/++8YxOTk5yMvLQ2ZmJkpKSiCXyxEbG4umpiYbVk5ERPbCroOuoqICkydPRlRUFLy9vTFlyhRERUVh//79AK7t5jQaDVJTUxETE4OAgABoNBro9XoUFRXZuHoiIrIHdh10Y8eOxY8//ojffvsNAPDrr79i9+7dePTRRwEAVVVV0Ol0iIyMNH6Pk5MTwsLCsHfvXpvUTERE9qWPrQu4ldTUVOj1eoSGhsLR0RGtra2YO3cuXnrpJQCATqcDAMjlcpPvk8vlOH/+vNXrJSIi+2PXQVdcXIxPPvkEBQUF8PPzw9GjRzF//nx4eXlh5syZxnEymczk+wwGQ4e2G2m1WovVbC96wzGKRcpr1VDf1wKv2XDTdq1WZ7E5O3N9Tnsl5Z8tSzB3vZRK5S377Tro3n77bcyePRtPPPEEACAwMBBnz57Ff/7zH8ycORMKhQIAUFtbC09PT+P31dXVddjl3eh2i9LTabVayR+jWKS+VkP0l0V9vYb6BgxxG3LTPqVyoEXmvJXrc9ojqf9sic2S62XXn9H9/vvvcHR0NGlzdHREe3s7AMDb2xsKhQKlpaXG/paWFpSXlyM0NNSqtRIRkX2y6x3d5MmTsXz5cnh7e8PPzw9HjhxBXl4eEhISAFx7yzIpKQnZ2dlQKpXw8fFBVlYWnJ2dERcXZ+PqiYjIHth10C1duhRLlizBm2++ibq6OigUCjz33HNIT083jklJSUFzczPS0tLQ2NiIkJAQFBcXw8XFxYaVExGRvbDroHNxcUFGRgYyMjI6HSOTyaBSqaBSqaxYGRER9RR2/RkdERFRdzHoiIhI0hh0REQkaQw6IiKSNMFB19jYaIEyiIiILENw0Pn6+mLmzJn48ssv8eeff1qiJiIiItEIDrpXXnkF+/fvx4wZM+Dr64u5c+di3759lqiNiIio2wQH3XvvvYeff/4Z27Ztw6RJk7BlyxZMmjQJwcHBWLp0KSorKy1QJhERkXnMOhlFJpNh/Pjx0Gg0+O2337B69WoolUosW7YMwcHBiIqKwkcffcTP84iIyOa6fdZl//798cQTT+CNN95AVFQUDAYD9uzZgzfeeAP+/v5IS0vD5cvWu5s5ERHRjbp1C7BTp05h8+bNKCwsRFVVFdzd3TF79mxMnz4d/fr1w0cffYSCggKcO3cOmzZtEqtmIiKiLhMcdPX19fj000+xZcsWHDhwAP369cOUKVOwdOlSTJgwAQ4O/79JXLx4MRQKBdRqtahFExERdZXgoPPz80NrayvGjBmD999/H7GxsRg0aFCn45VKJe68885uFUlERGQuwUGXkpKC6dOnY8SIEV0aP3nyZEyePFlwYURERGIQHHT/+te/LFEHERGRRQg+63LDhg149tlnO+2fOXMmTzwhIiK7ITjo1qxZA4VC0Wm/h4cHCgoKulUUERGRWAQH3alTpxAYGNhpv7+/P06ePNmtooiIiMQiOOhkMhnq6+s77W9oaEB7e3u3iiIiIhKL4JNRRo0ahcLCQsyePRv9+/c36WtubkZhYSFGjhwpWoFERLeiPmi9Oy+pggZabS4Sj+Ad3Zw5c6DVajFp0iR89tln0Gq1OHnyJD777DNERUVBq9Vizpw5lqiViIhIMME7uoiICKxcuRLp6elITEw0thsMBri4uCA3Nxd///vfRS2SiIjIXGbd6zIhIQHR0dEoKSlBZWUlDAYD7rnnHkRGRsLFxUXsGomIiMxm9k2dXVxcEBMTI2YtREREojM76JqamlBdXY2LFy/CYDB06A8PD+9WYURERGIQHHSNjY1IT0/H1q1b0dbWBuDa53Mymczk/w0NDeJWSkREZAbBQZeamoovvvgCL7/8MsLDw+Hq6mqBsoiIiMQhOOh27tyJV199FUuWLLFEPURERKISfB1dv379uvyIHiIiIlsTHHQxMTH49ttvLVELERGR6AQH3euvv46amhq89tpr2LdvH2pqanDhwoUO/4iIiOyB4M/oQkJCIJPJcOjQIWzZsqXTcWKddVlTU4OFCxfi22+/hV6vx/Dhw5GdnY2HHnoIwLWzPDMyMrBu3To0NjYiJCQEWVlZ8Pf3F2V+IiLq2QQHXXp6uvFSAktrbGzEpEmTMHbsWGzZsgVubm6oqqqCXC43jsnJyUFeXh7y8vKgVCqxdOlSxMbGYt++fbxLCxERCQ86lUpliTpuasWKFfDw8EB+fr6xbfjw4cb/GwwGaDQapKamGu/SotFooFQqUVRUZHIvTiIi6p0Ef0Z3o7a2NjQ0NKC1tVWsekx8+eWXCAkJQWJiInx8fPDQQw9h1apVxjuxVFVVQafTITIy0vg9Tk5OCAsLw969ey1SExER9SxmBd2BAwcwbdo03HXXXfDx8UFZWRkAoL6+Hk8++SR++OEHUYqrrKzEmjVrMHz4cHz66ad47bXX8O6772L16tUAAJ1OBwAmb2Ve/7q2tlaUGoiIqGcT/NZlRUUFpk6dCoVCgYSEBKxfv97Y5+bmBr1ejw0bNmD8+PHdLq69vR1BQUF45513AFx76Ovp06dRUFCAV155xTjur58Z3nhLspvRarXdrs3e9YZjFIuU16qhvq8FXvPmJ5pptTqLzdkZrVZn9fmEjZfuz5YlmLteSqXylv2Cg+69997DiBEj8N133+HKlSsmQQcADz/8MDZv3iz0ZW9KoVDA19fXpO2+++5DdXW1sR8Aamtr4enpaRxTV1fXYZd3o9stSk+n1Wolf4xikfpaDdGL+/TthvoGDHEbctM+pXKgRea8FaVyoNXn6yqp/2yJzZLrJfitywMHDmDGjBno37//TXdNd999t/Etxe4aO3YsTp48adJ28uRJDBs2DADg7e0NhUKB0tJSY39LSwvKy8sRGhoqSg1ERNSzCQ46BwcHODh0/m06nQ5OTk7dKuq6WbNmYd++fcjKysLp06exbds2rFq1Ci+99BKAa29ZJiUlYfny5di+fTuOHTuGWbNmwdnZGXFxcaLUQEREPZvgty5Hjx6Nb775Bq+++mqHvqtXr6KwsBBjxowRpbjg4GBs3LgRixYtwrJly+Dp6YkFCxYYgw4AUlJS0NzcjLS0NOMF48XFxbyGjoiIAJgRdHPmzEFcXBxmz56N+Ph4ANfuXrJz505kZWXhzJkzyMvLE63ASZMmYdKkSZ32y2QyqFQqq17fR0REPYfgoIuIiEB+fj7S0tKwadMmAEBSUhIMBgMGDRqEgoICPPjgg6IXSkREZA7BQQcAcXFxmDJlCkpKSnD69Gm0t7fjnnvuwYQJEzBgwACxayQiIjKbWUEHAHfccQf+8Y9/iFkLERGR6AQH3dmzZ7s07volAERERLYkOOhGjhzZpacXiPWYHiIiou4QHHQffPBBh6Bra2tDVVUVPvnkE7i7u5uc/k9ERGRLgoPumWee6bQvNTUVkZGR0Ov13SqKiIhILN16TM9fDRgwAM888wxWrlwp5ssSERGZTdSgA4C+ffvi/PnzYr8sERGRWUQNuqNHj+LDDz/s8MQBIiIiWxHtrMtLly7h8uXLGDBggKi3ACMiIuoOwUEXHh7eIehkMhlcXV1x77334oknnoCrq6tY9REREXWL4KDTaDSWqIOIiMgiRD8ZhYiIyJ4I3tFlZmYKnkQmkyE9PV3w9xEREXWX4KDLyMgwfkZnMBhM+m7VzqAjIiJbEBx0v/76K+Lj4+Hn54ekpCT4+PgAALRaLTQaDU6cOIHCwkIoFArRiyUiIhJK8Gd06enp8Pb2xqpVqxAUFAQXFxe4uLggODgYq1evhpeXF3dvRERkNwQHXWlpKR555JFO+yMiIlBaWtqdmoiIiEQjOOgcHR3x888/d9p/5MgRODjwZE4iIrIPghNp6tSpWL9+PbKzs9HU1GRsb2pqQlZWFj7++GPExMSIWiQREZG5BJ+MsnjxYpw5cwaLFy+GWq2Gu7s7ZDIZdDod2traEB4ejsWLF1uiViIiIsEEB52Liwu2b9+Or7/+Gjt27EB1dTUMBgMmTpyIiRMnIioqyhJ1EhERmUVw0F0XFRXFUCMiIrtndtCdPXsWZWVluHDhAmJjY+Hp6YnW1lZcvHgRgwcPRp8+Zr80ERGRaMxKowULFmDVqlVoa2uDTCbDyJEj4enpid9//x3BwcGYP38+kpOTxa6ViIhIMMFnXa5YsQIajQbJycnYtm2bye2+Bg4ciOjoaHzxxReiFklERGQuwTu6devW4cknn8S7776LhoaGDv2BgYEoKSkRpTgiKVEfvGzV+VRBA606H5G9Eryjq66uRlhYWKf9Li4uuHTpUreKIiIiEovgoBsyZAhqamo67f/ll18wdOjQbhVFREQkFsFBN3HiRKxbtw719fUd+g4fPoyPP/4Y0dHRohRHRETUXYKDbsGCBXBwcEBYWBgWLlwImUyGjRs34oUXXsCjjz6Ku+66C2lpaZaoFdnZ2XB1dTV5fYPBALVaDT8/P3h4eCA6OhrHjx+3yPxERNTzCA46hUKB77//HpMnT8bnn38Og8GAwsJC7Ny5E0899RT+97//wdXVVfRC9+3bh3Xr1iEwMNCkPScnB3l5ecjMzERJSQnkcjliY2NN7sNJRES9l6Cg+/PPP1FWVoZLly4hJycHZ86cgVarxYkTJ1BZWYnc3Fy4ubmJXuSlS5fw8ssvIzc31yREDQYDNBoNUlNTERMTg4CAAGg0Guj1ehQVFYleBxER9TyCgs7R0RHTpk0zuXzgzjvvhLu7u0UfzXM9yMaPH2/SXlVVBZ1Oh8jISGObk5MTwsLCsHfvXovVQ0REPYeg6+gcHBzg5eUFvV5vqXo6WLduHU6fPo38/PwOfTqdDgAgl8tN2uVyOc6fP9/pa2q1WnGLtEO94RjFYq21aqjva5V5rtNqdRaZs6G+4/Wz1+e71m+947TUMd5qPmHj+XsohLnrpVQqb9kv+ILxWbNmYcWKFZgxY0aHgBGbVqvFokWL8PXXX6Nfv36djpPJZCZfGwyGDm03ut2i9HRarVbyxygWa67VEL11LxhXKgeKPmdDfQOGuA3pdD7AusdpiWO83Xxdxd9DYSy5XoKDTq/Xw9nZGcHBwYiOjsbw4cPh5ORkMkYmk+Gf//xnt4urqKhAfX09xo0bZ2xra2vDTz/9hLVr12LPnj0AgNraWnh6ehrH1NXVWTyEiYioZxAcdAsXLjT+f/PmzTcdI1bQRUdHIygoyKQtOTkZI0aMwJw5c+Dj4wOFQoHS0lIEBwcDAFpaWlBeXo5FixZ1e34iIur5uhR0b775JmbMmIGgoCAcPnwYBoMBFy9exMCBAy36OB5XV9cOlyrccccdGDx4MAICAgAASUlJyM7OhlKphI+PD7KysuDs7Iy4uDiL1UVERD1Hl1Jq7dq1GDt2LIKCguDl5YWGhgYEBQVh69atHc6EtLaUlBQ0NzcjLS0NjY2NCAkJQXFxMVxcXGxaFxER2Qezt2M3Pp7Hmr788kuTr2UyGVQqFVQqlU3qISIi+8bHgBMRCdSVRy411PcV5YxQPm6p+yx3lTcREZEd6PKOrrKyEvv37wcAXL587a8UrVaLAQMG3HR8SEiICOURERF1T5eDTq1WQ61Wm7Slp6d3GHf9Yu2bPX2ciIjI2roUdHl5eZaug4iIyCK6FHRPP/20pesgIiKyCJ6MQkREksagIyIiSeN1dEREdq4r1+2JRYrX7XFHR0REksagIyIiSWPQERGRpDHoiIhI0hh0REQkaQw6IiKSNAYdERFJGoOOiIgkjUFHRESSxqAjIiJJY9AREZGkMeiIiEjSGHRERCRpDDoiIpI0Bh0REUkag46IiCSNQUdERJLGoCMiIklj0BERkaQx6IiISNIYdEREJGkMOiIikjS7Drr3338fERERGDZsGEaMGIGnnnoKx44dMxljMBigVqvh5+cHDw8PREdH4/jx4zaqmIiI7I1dB92PP/6IF198ETt27MD27dvRp08fTJs2DRcvXjSOycnJQV5eHjIzM1FSUgK5XI7Y2Fg0NTXZsHIiIrIXfWxdwK0UFxebfJ2fnw8vLy/s2bMHUVFRMBgM0Gg0SE1NRUxMDABAo9FAqVSiqKgIiYmJtiibiKjHUx+8bLW5VEEDLfr6dr2j+yu9Xo/29na4uroCAKqqqqDT6RAZGWkc4+TkhLCwMOzdu9dGVRIRkT2x6x3dX82fPx/3338/xowZAwDQ6XQAALlcbjJOLpfj/Pnznb6OVqu1XJF2ojcco1istVYN9X2tMs91Wq3OInM21Dd0Ot+1fusdp6WO8VbzAV0/xs7WSuic9nyMYs5p7u+iUqm8ZX+PCboFCxZgz549+Oabb+Do6GjSJ5PJTL42GAwd2m50u0Xp6bRareSPUSzWXKsheuu9FQQASuVA0edsqG/AELchnc4HWPc4LXGMt5sP6Nox3mqthM5pr8co5pyW/F3sEW9dqlQqfPrpp9i+fTuGDx9ubFcoFACA2tpak/F1dXUddnlERNQ72X3QzZs3D0VFRdi+fTvuu+8+kz5vb28oFAqUlpYa21paWlBeXo7Q0FBrl0pERHbIrt+6nDt3LjZv3oyPP/4Yrq6uxs/knJ2dMWDAAMhkMiQlJSE7OxtKpRI+Pj7IysqCs7Mz4uLibFw9ERHZA7sOuoKCAgAwXjpw3bx586BSqQAAKSkpaG5uRlpaGhobGxESEoLi4mK4uLhYvV4iIrI/dh10jY2Ntx0jk8mgUqmMwUdERHQju/+MjoiIqDsYdEREJGkMOiIikjQGHRERSRqDjoiIJI1BR0REksagIyIiSWPQERGRpDHoiIhI0hh0REQkaXZ9CzAiS1IfvIyG+r5Wee6WKmigxecgopvjjo6IiCSNQUdERJLGoCMiIklj0BERkaQx6IiISNIYdEREJGkMOiIikjQGHRERSRqDjoiIJI1BR0REksagIyIiSWPQERGRpDHoiIhI0hh0REQkaQw6IiKSNAYdERFJGoOOiIgkjUFHRESSxqAjIiJJY9AREZGk9bF1AWIpKCjAihUroNPp4OfnB7VajbCwMFuXRV2kPnjZanOpggZabS4isj1J7OiKi4sxf/58vPnmm9i1axfGjBmD+Ph4nD171talERGRjUliR5eXl4enn34azz33HABg2bJl+O6777B27Vq88847os5lzZ0HwN0HEVF3yRobGw22LqI7rl69iqFDh2LNmjWYNm2asX3u3Lk4duwYvvrqK9sVR0RENtfj37qsr69HW1sb5HK5SbtcLkdtba2NqiIiInvR44PuOplMZvK1wWDo0EZERL1Pjw86Nzc3ODo6dti91dXVddjlERFR79Pjg65fv34YPXo0SktLTdpLS0sRGhpqo6qIiMheSOKsy+TkZLz66qsICQlBaGgo1q5di5qaGiQmJtq6NCIisrEev6MDgMcffxxqtRrLli3Dww8/jD179mDLli3w8vKydWmiKygowMiRI6FQKDB+/Hj89NNPnY7dvXs3pk+fDl9fXwwdOhRhYWHYsGGDFau1PSHrdaNTp07B09MTd999t4UrtB9C18pgMGDlypV48MEH4e7uDl9fXyxcuNA6xdoBoev13Xff4dFHH4WnpyfuvfdeTJ8+HSdPnrRStbZTVlaGhIQE+Pv7w9XVFRs3brzt9/zyyy+YMmUKPDw84O/vj8zMTBgM5l8gIImgA4CXXnoJR48eRW1tLX744QeEh4fbuiTRCb0wvqKiAoGBgVi3bh3Ky8vx4osvIjU1FYWFhVau3DbMvZHA1atX8cILL/SqO+uYs1ZvvfUW1qxZg4ULF6KiogJbtmzpNWsmdL0qKyvx9NNPY9y4cdi1axe2bduGlpYWxMfHW7ly67ty5QoCAgKQkZEBJyen246/fPkyYmNj4e7ujpKSEmRkZCA3NxcffPCB2TX0+OvoepMJEyYgMDAQK1asMLYFBwcjJiamyxfGP//882hra+sVOztz10ulUuHSpUsIDw9Heno6zp07Z41ybUroWmm1WowbNw5lZWXw9fW1Zql2Qeh6ffbZZ0hMTMSFCxfg6OgIANi1axemTp2KU6dOwc3NzWq129Ldd9+NpUuX4plnnul0zPU/nn777TdjMC5btgxr167FsWPHzDqbXjI7Oqm7evUqDh06hMjISJP2yMhI7N27t8uv09TUBFdXV5Grsz/mrteOHTuwY8cOZGZmWrpEu2HOWn311VcYPnw4du7ciVGjRuH+++/Ha6+9hgsXLlijZJsyZ71Gjx6Nvn37Yv369Whra0NTUxP++9//Ijg4uNeEXFdVVFRg3LhxJru/CRMm4Pz586iqqjLrNRl0PYQYF8Z/8803+OGHH/D8889boEL7Ys561dTUICUlBfn5+XBxcbFGmXbBnLWqrKzE2bNnUVxcjJUrVyI/Px9arRYJCQlob2+3Rtk2Y856eXt7Y+vWrVCr1XB3d4eXlxeOHTuGzZs3W6PkHqW2tvama3u9zxwMuh7G3Avj9+zZg5dffhmZmZkICQmxVHl2R8h6vfLKK3jhhRfw4IMPWqM0uyNkrdrb2/HHH38gPz8f4eHhCAsLQ35+Pvbv348DBw5Yo1ybE7JeOp0Or7/+OhISElBSUoIvvvgCAwYMwPPPPy/5PwzMcbO1vVl7VzHoeojuXBhfXl6O+Ph4qFQqvPjii5Ys026Ys167du1CZmYm3Nzc4Obmhtdffx1XrlyBm5sbPvroIytUbRvmrJVCoUCfPn3g4+NjbBsxYgT69OmD6upqi9Zra+as1+rVq3HHHXdg0aJFGDVqFMLDw7Fq1SqUlZUJ+uihN3B3d7/p2gIw+yYgDLoewtwL48vKyhAfH4/09HTMmjXL0mXaDXPW66effsLu3buN/xYsWAAnJyfs3r3b5IbhUmPOWo0dOxatra04c+aMsa2yshKtra0YNmyYReu1NXPWq7m52XgSynXXv+aOztSYMWNQXl6OlpYWY1tpaSmGDh0Kb29vs16TQdeDJCcnY9OmTVi/fj1OnDiBefPmmVwY/+6772Lq1KnG8bt370Z8fDwSExPx5JNPQqfTQafTGf86kjqh6xUQEGDyb+jQoXBwcEBAQIDkT+ARulaPPPIIRo0aheTkZBw+fBiHDx9GcnIyHnjgAQQFBdnqMKxG6HpNnDgRhw8fRkZGBk6dOoVDhw4hOTkZnp6eGD16tI2Owjr0ej2OHDmCI0eOoL29HdXV1Thy5IjxUoy/rlVcXBycnJwwa9YsHDt2DNu3b8fy5csxa9Yss9+6lMSdUXqLxx9/HA0NDVi2bBl0Oh38/f1NLoyvqakx+Qt706ZN+P3335Gbm4vc3Fxj+7Bhw3D06FGr129tQterNxO6Vg4ODti8eTPmzZuH6Oho9O/fHxEREViyZAkcHKT/97PQ9Ro/fjwKCgqQk5OD3Nxc9O/fHw888ACKiorg7Oxsq8OwioMHD+Kxxx4zfq1Wq6FWqzF9+nRoNJoOazVo0CBs3boVc+fORUREBFxdXZGcnIzZs2ebXQOvoyMiIkmT/p9eRETUqzHoiIhI0hh0REQkaQw6IiKSNAYdERFJGoOOiIgkjUFHRESSxqAjIiJJY9AREZGk/R9c2zEpzgWNvAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# calculate the unique words ratio (not including stopwords) for each song \n",
    "unique_words_ratio = data_df.apply(get_only_eminems_verses, axis=1) \\\n",
    "                            .apply(split_lyrics_into_words) \\\n",
    "                            .apply(dash_and_contractions_cleaning) \\\n",
    "                            .apply(lambda words: len(set(words)) / len(words)) \\\n",
    "    \n",
    "# show results\n",
    "print('mean: %.2f' % unique_words_ratio.mean())\n",
    "print('std: %.2f' % unique_words_ratio.std(), '\\n')\n",
    "\n",
    "unique_words_ratio.plot.hist(bins=12, rwidth=0.95, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "qd3KdGrrKIc8",
    "outputId": "619ad2ad-4abf-42d5-d95c-888abb993b70"
   },
   "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>artist</th>\n",
       "      <th>featured</th>\n",
       "      <th>release_date</th>\n",
       "      <th>unique words ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>326</th>\n",
       "      <td>My Salsa</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2004-03-14</td>\n",
       "      <td>0.091837</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>292</th>\n",
       "      <td>Alfred (Intro)</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2020-12-18</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>246</th>\n",
       "      <td>Alfred (Outro)</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2020-01-17</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>255</th>\n",
       "      <td>Public Service Announcement</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Jeff Bass]</td>\n",
       "      <td>1999-02-23</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>285</th>\n",
       "      <td>Steve Berman (Skit) [2002]</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Steve Berman]</td>\n",
       "      <td>2002-05-26</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>Superman</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>[Dina Rae]</td>\n",
       "      <td>2002-05-26</td>\n",
       "      <td>0.380855</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>321</th>\n",
       "      <td>1998 Wake Up Show Freestyle</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>NaT</td>\n",
       "      <td>0.971154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>FACK</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2005-12-06</td>\n",
       "      <td>0.389078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>298</th>\n",
       "      <td>It’s Only Fair to Warn [Freestyle]</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>2010-01-01</td>\n",
       "      <td>0.964286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>352</th>\n",
       "      <td>’97 Blazer</td>\n",
       "      <td>Eminem</td>\n",
       "      <td>None</td>\n",
       "      <td>NaT</td>\n",
       "      <td>0.952703</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  title  artist        featured release_date  \\\n",
       "326                            My Salsa  Eminem            None   2004-03-14   \n",
       "292                      Alfred (Intro)  Eminem            None   2020-12-18   \n",
       "246                      Alfred (Outro)  Eminem            None   2020-01-17   \n",
       "255         Public Service Announcement  Eminem     [Jeff Bass]   1999-02-23   \n",
       "285          Steve Berman (Skit) [2002]  Eminem  [Steve Berman]   2002-05-26   \n",
       "32                             Superman  Eminem      [Dina Rae]   2002-05-26   \n",
       "321         1998 Wake Up Show Freestyle  Eminem            None          NaT   \n",
       "76                                 FACK  Eminem            None   2005-12-06   \n",
       "298  It’s Only Fair to Warn [Freestyle]  Eminem            None   2010-01-01   \n",
       "352                          ’97 Blazer  Eminem            None          NaT   \n",
       "\n",
       "     unique words ratio  \n",
       "326            0.091837  \n",
       "292            1.000000  \n",
       "246            1.000000  \n",
       "255            1.000000  \n",
       "285            1.000000  \n",
       "32             0.380855  \n",
       "321            0.971154  \n",
       "76             0.389078  \n",
       "298            0.964286  \n",
       "352            0.952703  "
      ]
     },
     "execution_count": 62,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# find outliers\n",
    "N_TOP = 10\n",
    "outliers_idxs = ( np.abs( unique_words_ratio - unique_words_ratio.mean() ) ).argsort()[-N_TOP:][::-1]\n",
    "\n",
    "# show results\n",
    "pd.concat(objs=[data_df.iloc[outliers_idxs][['title', 'artist', 'featured', 'release_date']], \n",
    "                unique_words_ratio[outliers_idxs]], \n",
    "          axis=1)  \\\n",
    "  .rename(columns={0 :'unique words ratio'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_3Mh9ttPKIc8"
   },
   "source": [
    "## 3.3 The most frequent words <a name=\"33-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "5GQR7O0MKIc8"
   },
   "outputs": [],
   "source": [
    "# count word frequencies\n",
    "word_counts_df = pd.DataFrame(all_words)[0].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1224x504 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "style.use('fivethirtyeight')\n",
    "\n",
    "# words per subplot\n",
    "NUM_OF_WORDS = 50\n",
    "\n",
    "fig, axs = plt.subplots(1, 2, figsize=(17,7), sharex=True)\n",
    "\n",
    "# left subplot [1-50]\n",
    "word_counts_df[:NUM_OF_WORDS].plot(kind='barh',\n",
    "                                   title=\"Most frequent [1-50]\",\n",
    "                                   ylabel='word',\n",
    "                                   fontsize=9.3,\n",
    "                                   ax=axs[0]).invert_yaxis()\n",
    "\n",
    "# right subplot [51-100]\n",
    "word_counts_df[NUM_OF_WORDS:2*NUM_OF_WORDS].plot(kind='barh',\n",
    "                                                 title=\"Most frequent [51-100]\",\n",
    "                                                 ylabel='',\n",
    "                                                 fontsize=9.3,\n",
    "                                                 ax=axs[1]).invert_yaxis()\n",
    "# title\n",
    "fig.text(x=0.5, y=0.95, s='Top 100 Most Frequent Words [Eminem]', ha='center', fontsize='x-large', fontweight=555)\n",
    "# x-label\n",
    "plt.xlabel(\"count\", x=-0.1, y=0.042, labelpad=12)\n",
    "# signature bar\n",
    "fig.text(x=0.03, y=-0.045, s=str('_'*160).expandtabs(), color='grey')\n",
    "fig.text(x=0.03, y=-0.07, \n",
    "         s=str(\"github.com/mariostrbac \" + \"  \" * 183 + \"Lyrics Source: Genius.com\").expandtabs(),\n",
    "         fontsize=7.9,\n",
    "         color='dimgrey')\n",
    "\n",
    "plt.subplots_adjust(wspace=0.13, hspace=0)\n",
    "plt.savefig(\"../reports/figures/vocabulary-analysis/100_most_frequent_words.png\", dpi=200, bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "fMSDsrDuKIc9"
   },
   "source": [
    "## 3.4 Word sentiment analysis <a name=\"34-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "yFmOe0ZiKIc-"
   },
   "source": [
    "### 3.4.1 The average subjectivity and polarity of words <a name=\"341-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "id": "RMNu2l6WKIc-"
   },
   "outputs": [],
   "source": [
    "# get subjectivity and polarity for each word\n",
    "words_df = pd.DataFrame(all_words, columns=['word'])\n",
    "words_df['subjectivity'] = words_df['word'].apply(lambda word: TextBlob(word).sentiment.subjectivity)\n",
    "words_df['polarity'] = words_df['word'].apply(lambda word: TextBlob(word).sentiment.polarity)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "bsQGZTT3KIc-"
   },
   "source": [
    "#### a) Subjectivity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "ghY6aJITKIc-",
    "outputId": "05642a3d-d350-4383-a42b-b9f97391a61d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 0.08\n",
      "std: 0.22 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# show results\n",
    "print('mean: %.2f' % words_df['subjectivity'].mean())\n",
    "print('std: %.2f' % words_df['subjectivity'].std(), '\\n')\n",
    "\n",
    "words_df['subjectivity'].plot.hist(bins=8, rwidth=0.95, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7W0TDNzxKIdA"
   },
   "source": [
    "#### b) Polarity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "TyOgabRYKIdD",
    "outputId": "0505c0a3-b1a3-4099-e061-e934332cd27f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: -0.01\n",
      "std: 0.15 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# show results\n",
    "print('mean: %.2f' % words_df['polarity'].mean())\n",
    "print('std: %.2f' % words_df['polarity'].std(), '\\n')\n",
    "\n",
    "words_df['polarity'].plot.hist(bins=6, rwidth=0.95, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Roo5pwLCKIdD"
   },
   "source": [
    "### 3.4.2 The most frequent positive, negative and neutral words <a name=\"342-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "id": "lu_4goHnKIdE"
   },
   "outputs": [],
   "source": [
    "# count word frequencies\n",
    "word_counts_df = words_df.value_counts().reset_index().rename(columns={0:'counts'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABLYAAAHYCAYAAABQoGEYAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOzdeVhU5fv48Tc77ogLJAhuuFDihqnghoCo4JblAu6pifuGlVmZS2QuWKZmuSuofUozQcQlFzRxS4VccipFwdREwVQShPn94Y/5Muwih2Hgfl2X1+XMnHPmPvfMOQ/nmefcj0FiYqIaIYQQQgghhBBCCCH0jKGuAxBCCCGEEEIIIYQQojCkY0sIIYQQQgghhBBC6CXp2BJCCCGEEEIIIYQQekk6toQQQgghhBBCCCGEXpKOLSGEEEIIIYQQQgihl6RjSwghhBBCCCGEEELoJenYEkIIIYQooODgYCwsLDT/YmNjFXuvtLQ0goKCcHFxwdraWvOe7du3V+w9S4Pi/Iz0mb+/vyZHTZs21XU4JVLTpk21vksZ/yIjI3UdWo4yx+vv76/rcIpUbGxsjp+FhYWFrkMTQpQA0rElhFBcbn8Y5vWvJP9Bdu3aNWxsbLTiDQwMzHX5n3/+GT8/Pxo1akSNGjWoX78+/fr1Y+fOnS/83pkvRDL+rVy5Msdlf/rpp2zLent7v/B7FoXAwMCX/iO0oN+dknrBoY9e5nP7/ffftdZdvHhxtmUGDBigtUxMTIzW60+fPsXKykrz+qRJk15md/TO559/zieffMKlS5f477//dB2OlpzORfp2Li+Ma9euERAQQJs2bahVqxY1atTAwcGB119/HV9fXz799FNUKpWuwyxw+1SaRUZGFkvbkPU8mds/XbW/QghRFhjrOgAhhNAnaWlpjB07lsePHxdo+VmzZmXreEpISODgwYMcPHiQ3bt3880332BsXPjT8erVq3nnnXcwMjLSej63Di8hikOjRo2oWbMmd+/eBeDYsWPMmDFD83paWhonTpzQWufYsWNaI0dOnTrF06dPNY9dXV0Vjrpk2bZtm+b/tra2DB06lHLlylG9enUdRlV2HTlyhIEDB5KcnKz1/D///MM///zD1atX2bNnD9bW1jg4OOgoyoLr168fjo6OAFSuXFnH0ZR8bm5udOnSBYC6devqOJqczZgxg4cPHwLQpEkTHUdTtKpWrcq8efMAOHfuHDt27NBxREKIkkQ6toQQisv8h1aGDz/8UPP/OnXq8Pbbb2u9XlL/IFu2bBknT54s0LJbtmzR6lxq3749nTt35tdff2XPnj0A7NixAwcHB95///1CxxQbG0toaCi9e/fWPHfmzBmioqIKvc2SLqfvTIaCXnA8fPhQLuYU5urqqhmZeOrUKVJTUzExMQEgOjo623nh2LFjWiN8jh07lm17SlGr1Tx69IhKlSop9h4v6saNG5r/Dxw4kJkzZyr+noXNw/Tp03Mc1VdSz+UvKj09nfHjx2s6tSwtLenTpw82Njb8+++/xMfHExMTw5UrV3QcacF5eHjg4eGh6zD0xuuvv87EiRPzXW7kyJE5tkM2NjZKhKVl2LBhir+HrlSuXFmT/+DgYOnYEkJokVsRhRCKGzZsGBMnTtT6l5mNjU221z08PLh8+TJTpkzB2dmZWrVqYWVlhZOTE2PGjMmxcymn2w527tyJp6cntWrVws7OjgEDBnDhwoVC7Ud0dDQLFy4EoFevXnkum56erlkWoEWLFvz000/MmDGDkJAQfHx8NK+tWLEi2wV+QWWM0so6Ouurr77KtkxuUlJS2LBhA7169aJ+/fpUr14de3t7unbtypdffsm///6bbZ24uLgcb8dp374948ePZ9euXcD/fSaZcwHat8kU5lalnL4zGf9sbW2B7LeHJCcns2DBAlq2bEmNGjUYO3as1jZ/+uknBg4cSOPGjalRowZ2dnZ4enqycuXKXG8DO3PmDP369aN27drY2trSs2dPIiMj86zxk9/tffnVSHnw4AGLFi3C3d0dOzs7atSoQaNGjRgyZEi2jiDIXpckODiYI0eO0KtXL2rXrk2tWrXw8fHh9OnTmnWK6nPLXAvqyZMn/Prrr5rHmWO1trYG4JdffkGtVmueP378uOb/derU0Xy2ULjvrbe3t9ZtQfHx8YwbN47GjRtTrVo1Nm/erFn2119/1fpse/XqlWN+M0tLS+Obb76he/fu1KtXj2rVqmFnZ0eLFi0YOHAgixYtKtBoz4w4M+di8eLFOd5apnQeCmro0KE5Ho9ZO06+/PJLfH19adGiBXXq1KF69erY2trSpk0bpk6dytWrV3N9j1OnTuHv70/Lli2pVasWr7zyCs2aNWP48OHZRv9lplar2bx5Mx07dsTa2pp69eoxatQo/v777wLv3+XLl4mLi9M83rRpE0uXLmX69OnMmTOHb7/9lqioKM6dO5drB+yRI0cYMWIEr732GlZWVtja2tKhQwc+//xzkpKSsi2ftQbWw4cP+fjjj3FycqJmzZq89tprzJkzh5SUFM06GZ9tZgsXLszxfJRXja2s35Hr168zcuRIzXHYp08fzp8/D6B5rW7dulhbW+Pp6cnBgwdzzMGzZ8/YsmULffv2xcHBgRo1alCnTh18fHwIDg4mPT092zpZz4l//fUXo0ePpkGDBtSsWZN27doREhKitY6FhQU9e/bUeq5nz5657m9R6tu3b47HwhtvvJEtxszH9NGjR+nRowe1atWifv36jB07lnv37gFw9OhRvL29sbGxwc7ODl9fX/74449s751X+1GYPGb2om1k1veLjo6mf//+1K5dG3t7e/z8/Lh27Rrw/O+rAQMGYGdnh42NDb1799ZqL4QQIj8yYksIUSJt2LCBgIAAUlNTtZ6/ceMGN27c4H//+x8zZszggw8+yHUbS5Ys4fDhw1rPRUREcPToUf73v/+9UAHmp0+fMnbsWFJSUnB0dOTDDz/kp59+ynX58+fPc/PmTc3jvn37Ymj4f78lvPnmm4SGhgLw6NEjDh06pDXiqqC8vb356aefOHnyJGfPnqVVq1bcuHGD3bt3A9C8eXMSEhK0YsksISGBN954I1tnX1JSEqdOneLUqVOsW7eOnTt3an6BvnfvHl26dNHcYpYh43ac3377jevXrxdqf5TUr18/fvnll2zPP336lOHDhxMeHq71fGpqKqdPn+b06dNs376dnTt3YmlpqXn9wIED+Pr6al1URkZGcvz4cTw9PRXZh/PnzzNw4EBu376t9fydO3fYvXs3u3fvJiAgIM/jYsuWLURFRWl1mhw7doxevXpx5MgRGjZsWGTxZj3Gjh07Rps2bTT/h+e/wo8YMYLAwEAePHjAxYsXee2113j69ClnzpzRrJu5s6Aw39usbt++jbu7e7ZcAhw6dIgBAwZofbZHjx7l2LFjeY5wmTx5Mlu2bNF67uHDhzx8+JBr166xd+9e+vfvT4UKFXLdxotQOg9KWLZsGffv39d67tGjR/z+++/8/vvvhISE8P3339OhQwetZWbPnq3VYZ8hNjaW2NhYGjVqRLt27XJ8zxkzZnDgwAHN4//++4/vv/+e6OhoIiMjMTMzyzfurB2SGR1YBgYGWs/nlGe1Ws306dNZt26d1vNPnz4lJiaGmJgYgoOD+fHHH3P9nB4/fkzXrl21RoTFxcWxbNky7t69q+it53FxcXTp0kXrczt8+DA9evRgzZo1jB8/ngcPHmheO336NG+99RY//fST1jkgMTGRt956S6sTPeP5Y8eOcezYMX744QdCQkIwNzfPMZbo6Gg6d+6s9WPQ5cuXGTduHAC+vr5Fss/FLSIigkWLFmk69p48ecK2bds4f/48kyZNYsKECVqdfnv27NGMzM7cLhXUi+SxsG1kZhcuXMDLy0vrNt6wsDBOnz7NkiVLGDVqlNZt50eOHMHHx4ejR4/SoEGDF94/IUTZIx1bQogS5+TJk0ybNk3zR1yVKlUYNGgQZmZm/PDDD8TFxaFWq1m0aBGOjo707ds3x+0cPnyYNm3a0KlTJ65cuaLpiEpOTuadd97h3LlzmJqaFiimefPmcenSJUxNTVm9enW+62Utgl2vXj2tx3Xq1Mm2fGE6gsaPH6/ZrxUrVrBu3TpWrVpFWlqa5vW5c+fmuv4777yjdVHcsWNH2rVrx6VLlzSdY9evX8fPz4/IyEiMjIzYtWuXplPL3NwcPz8/bGxsuH//PnFxcVqjWurWrcu8efP4+eefOXTokOb5jDoZULhbleLj41m+fHm256tXr86gQYNyXOeXX36hRYsWdOnShdTUVE1ds9mzZ2v+YDcwMMDb25tmzZpx584dtm3bxqNHj7hw4QL+/v5s374deP4dGjdunFbHR79+/ahXrx779u0jIiLihfcpP48ePdLq1KpWrRr9+vWjRo0anDx5UnPhvmjRIl577bVcv08nTpzAwcGBXr16ER0dzf79+zX79PXXX7N06dIi+9xyqrM1ffp00tPTNbfKtmvXjo4dO2pGIR07dozXXnuN06dPa40CyHyBXJjvbVZ//vknAN27d6d58+bcuXOHmjVr8t9//2k6sTNk/mz37duX474+evSIrVu3asXUoUMHUlJSuH37Nr/++isXL17MN2cAb7/9Nt26ddO6ZTtzfZ/XX39d8Ty8qE2bNuU4AnHQoEFaNcFeeeUVXFxcsLOzw8LCAmNjY+7evUtoaChxcXE8ffqUmTNnao3AWr16tVanlrm5OX379qVu3br8/fffuY4OynDgwAE6duxImzZtCAsL49KlSwBcvXqVsLCwbCNpctKoUSPMzMw0F98ffvghy5Ytw9nZGScnJ1q2bImrq2uOtzd/9dVXWp1anTt3pl27djx8+JBt27aRkJBAbGwsgwcPJjIyUutHkAz3798nMTGRAQMGUKtWLTZu3KjpaNq2bRsfffQR1tbW+X534HmtohcRGxtLlSpVmDBhAvfv39eM6nny5Am+vr6Ym5szZswY0tPTWbduHenp6aSnp/PFF19kO24zOrVMTEzo27cvDRs25Nq1a3z33Xekpqby888/8+GHH7Jo0aIcY7l48SJVqlTB39+fp0+fsmHDBs3fCsuWLdN0yMybN49r165p5T3zLYJK3oa+c+dOzWi2zLy9vbP9LZDh/PnzNGzYkF69ehEVFaVpR69cucK4ceOwtbWlf//+XLx4UdO+3L17l82bNzN58uQXjrGgeYTCtZFZXbp0CRsbG/r376857jL2YciQIVStWpUxY8Zw584dvvvuO+D592vVqlUsWbLkhfdPCFH2SMeWEKLEWb58ueYPLAMDA0JDQzW3DYwbN46WLVtqfj1ftmxZrh1brq6u/PTTT5qLuQ8++IAVK1YAzztGwsPDC9SZdPz4cc2v4bNmzaJp06b5Th+fdURC1no1VapUyXP5gmrTpg2tW7fm9OnT/PTTT1y8eFEzYsTGxoa+ffvm2rH122+/aY1i6NWrF5s2bdI8zpyvS5cusW/fPrp3756tmHfWPzrVarUmP7a2tkycOFEzKi1DQeqU5OX69etaF24ZXnvttVw7tnr06MHmzZu1Lu4TExNZv3695vG7777Le++9p3ncvn17hg8fDjz/Rf3SpUs4OjoSHh6uNWJt2rRpfPTRRwAEBATQoUMHfv/995fax6xCQkI0nVrGxsbs379f6yKpf//+mk6XoKCgXL/bNjY2HDx4kMqVK6NWq+nYsaOmI/bs2bNA0X5uOdXZunjxoubWq/bt29OqVSvKly/PkydPOH78OGPHjtW6DTFjO1D4721O5s2bl22fduzYwZ07dzSPC/rZPnv2TNOhDPDtt99iZWWltczff/9doE6FjI6WzN/xrPV9lM7Di8rt4rNTp05aHVvHjx/n0aNHnDlzhtjYWB49eoSNjQ2dOnUiODgYeD5yJD4+HhsbG9RqNUuXLtWsX758eQ4dOkSjRo00zz179oxbt27lGluPHj0IDg7GwMAAf39/HBwcNJ/V2bNnC9SxVaVKFd5//33mzJmjeS4hIYGIiAhNR4OZmRkDBgxg7ty5mk6+9PR0vvzyS806vr6+WqOr+vXrp+l0unjxIvv27aNbt245xjBv3jzGjx8PQKtWrRg8eLDmPc6dO0f37t0L9N0pjJCQEM0x+Pvvv2vOFfD89tL+/fsDzzsqMn5syTzi8tKlS1od/l988YVW50mjRo00x9mGDRv44IMPcuwoNTAwYNeuXTRv3hwAU1NTvv76a+B5R+W///5LpUqVmDhxIpGRkVodW3379s02ElAJWUfmZahbt26uHVsWFhbs27cPCwsL/v33X+rUqaP5jhobGxMWFoa9vT1paWk0adJE0/5kzvGLKGgeC9tGZmVsbExoaCh169ZFrVbTqFEjrTZ069attG3bVvP+GR2Dhd0/IUTZIzW2hBAlTub6Wc2bN9eqhWFlZaV1K1B0dDRPnjzJcTt+fn5anRgZFwEZMv9hnpt///0Xf39/0tPTadeuHZMmTSrQPmS+zaswr7+IjAudZ8+e0b9/f01NnTFjxuQ522LWOmV+fn5aj7PmK2N5FxcXzYiCgwcP8vrrrzN8+HDmzp3L999/zz///JNtRFpJEBAQkG3EypkzZ3j27Jnm8WeffaZV9yTjD/YMGaOMstb+GDBggOb/pqamBbpQflGZJwN49uwZLVu21Io180iiCxcu5HpcDBgwQDNawcDAQGv2tsTExCKPO/OIjcePH3Pu3DmtUX3t27fH1NSU1q1bA/9XZyvzMnZ2dtjZ2QGF/95mVaVKlWx11qDwn62FhQWvvfaa5nGbNm148803CQgI4JtvvuHChQu88sorud5i9aKUzoMS0tPTmT9/Pg0aNKBPnz5MnjyZDz74gA8//FDTqZUho6NKpVJpdTQOHDhQq1MLnl80Z3w/cjJq1CjNLYOWlpZUq1ZN89qLfOenTJlCSEgIbdu2zXYLIjy/ZWvTpk30799fc47/448/+OeffzTLhISEaB23mUdSAblO+mFoaMjIkSM1j7PeMqzEsZvB1tZW61Zge3t7zf+NjY21flzK7XySdb/GjRunlYeMTi14fotbbh0arVu31nTGQPHmQUndunXTdORVqlRJ6zvapk0bTc6NjIy0blct7P4WNI+FbSOzev311zVxGxgYaB2vtWvX1nRqQe7fISGEyIuM2BJClDiZa3XkdEtM5ufUajVJSUmUL18+23JZR0tkfVyQP5iWLFnCjRs3qFSpEqtWrcrxFpGcZP6jFMhWHD7r48LUyMjQs2dP7O3tiY2NJT4+HoCKFSvmOztS5jxD/vnKWL558+YsXbqUTz75hAcPHnD16lWtgs9GRkZMmjSJjz/+uND7lB9XV1fNrQwFlVPtqBcdKZdRyDdroecaNWpoPX6RW7nUarXWRXLmEXGZvUisarWa+/fv53hcZO0AyHxbbU6Fm19WTnW2Tp06BTy/HcjJyUmz3JEjR0hISODChQu51tcq7Pc2q3r16mlmaMzsZT7bdevWaW5zTkxM1BpRBdCsWTN++OEHrRFMhaV0Hl7UhQsXtDo8cvLtt9+yePHiAm0v4zbUrHHn9x45KcrvfI8ePejRowf379/nzJkz/Prrrxw+fFjrgv7UqVOcPHmStm3bFvock1XNmjUpV65cjvsAyhy7GbLO5pe5JlmNGjW0vj+Zf0zJ/ANOUeUhr88SlM1DQe3evfuFR4blleNatWppvZY5x4Xd34Lmsag+t+LePyFE2SMdW0KIEqdq1aqaX7izFijP+pyBgUG22/oyZP6VP6fHOd3mkNs2/v33X61fN7NauHAhCxcuZOzYsXz22WfZZlz666+/8nz8MjM0GRkZ8c477zBr1izNc76+vvnuX9ZbovLLV+blhw8fjp+fH7/++iuXLl3i2rVrnDp1iqioKNLS0ggKCsLd3f2FCvQrLaeC3Vlz4OfnR+PGjXPdRkZto6zfuX/++UerczKn722GrJ2jycnJmg6opKQkrdEducVavnx53n///VzfA3KvIZO1EyOnkSdFKWudraNHj3Lu3DkA2rZtqxlFl/m7EhQUpFVkOHPH1st8bzPLqdMPXu6zbdiwIYcOHeKvv/7i3Llz/Pnnn/z+++/s2bOH5ORkLly4wEcffVQkhb6VzoMSfvjhB83/ra2t2bBhA82bN8fc3Jx9+/ZpbmfLLGvc+d0GnhMlvvOWlpZ07dqVrl278t577/HZZ5/x2WefaV6/efMmbdu2zRa/t7e31uiUrHKrXVfcx21meY38zeu1zLLmYcKECdk6XzNr1qxZjs/rMg9KKoocv4iC5rGwbWRWxb1/QoiyR84kQogSp23btprCx+fPn+e3337T3OJz584drVEQTk5OuV6Ybd26lYEDB2o6ErLOVtaqVSslwgeej2qytbXVTA+/c+dOJk6cqIkl8wVehQoV6Ny580u935AhQ/jss894+PAhhoaG2ab5zknWi6vg4GC8vLw0j7PmK2M2u4w6T9bW1rRp00bzvFqtxt7eXjMa7dy5c5rOiqx/RCcnJ2uNPtCV1q1bY2xsrLnV4r///suxFk1SUhL79+/X7GvLli21Xt++fbvmVpqUlBR27NiR63tm7XA8deqU5vNfvHhxrreptm3bVlOr6smTJzg6OuLu7p5tuUuXLpGUlFQkxZGL6nPLXGfryJEjmn3M3GGVuc5W1hlHM3d6FfZ7W1Av89meP3+eZs2aUa9ePa1aOjNnzuSbb77RLFMUlM6DEhISEjT/b9asmdY+ZBSMzsrBwUGrY3Tbtm288847WrcjpqWl8ffff2Nra6tQ5M/roy1dupS33347xwv7rHUUM0btOjg4UL16dc1Ilvv37zN27Ngcj62dO3fmOevmi8h6XtO1rN9XExOTHM+1d+/e5eTJk4WaVCSrrDkuCXnQN4VtI4UQorhJx5YQosSZMGECYWFhpKeno1ar8fHx0cyK+P3332tNuz5lypRctxMZGYm3tzcdO3bkypUr7Nq1S/NarVq1ci3Qm1nmQvWZJScna2aTg+ejUho1aqQZeWVoaMjMmTM1NbnOnz9Pr169cHNz4+zZs+zZs0ez7rhx43IddVZQlSpVYseOHdy+fZtKlSrlOmV8Zq+++ioeHh6ajsKffvqJ3r1707ZtW61Z1QAcHR3p2rUr8LxWz/Dhw2ndujVNmjTBysoKExMTfvnlF61bLDP/0pv1VoPRo0fTunVrDA0N6d69u86m87awsGDYsGGsXbsWeN7h+Oeff+Lm5kalSpW4d+8e0dHRnDx5Emtra958803g+QxymS+2ly5dSmxsLPXr1yciIiLPwvHOzs5aj4cMGUKXLl24efNmtvpOmQ0aNIglS5Zo3nPQoEH4+PhoLvBv3LjBqVOnUKlUvPvuu7Rr167wifn/iupza9++vaZjK3PHXeYOq4w6W5k7vuB5fZ/MNdsK+70tqJf5bL29vbGwsMDV1RVra2ssLCy4deuWZhY5KNhI0YJQOg8vKrdZETPPVOrg4KCZhXHfvn1MmDCB2rVrc/jwYa1ZEDMzMDBg2rRpmoLVT548oXPnzvTt25c6depw9+5dDh8+TL9+/fIdxfgyUlJS+Pbbb/n2229p1KgRzs7O1K5dG2NjY65evcqPP/6oWbZy5cqaC3xDQ0MmTpyouTX7xIkTuLi40K1bNywtLXnw4AGXLl3il19+0cwyWBRq1arFjRs3gOd1vUxNTalcuTKWlpbZ6rEVh1dffRVPT09NuxkUFMSZM2do06YN5cqV4/bt25w/f56zZ8/Srl07evbs+dLvmfX8NW/ePKKjozE1NcXJyYlOnTq99HvkJLdZEc3MzBgzZowi76mUwraRQghR3KRjSwhR4rRp04alS5cSEBBAamoqiYmJrFq1SmsZAwMDZsyYkeuMiPD8AjU8PDzbBZO5uTlff/21Vo2H3IwePZrRo0dnez42NlbrVok+ffpku6gaOnQoFy9eZPXq1cDz2kKZC2LD85nMZs6cmW8cBZG1w6QgVq9ezRtvvMGFCxeA56Npjhw5orVMnTp1CA4O1iq8rlarOXXqlKZWUlb169fXmpXP09OTihUr8ujRIwBCQ0MJDQ0Fntf60FXHFsCCBQuIi4vTzNh1/vz5fEfVlCtXjpUrV+Lr60tKSgrwf6PwDA0N6dq1q1Yx98xatWpFp06dNHn+999/NZ2urVq14saNGznejli5cmW2bdvGoEGDuHPnTr6jh4pCUX1uOd2SWqlSpWy3G2XU2cos86iuDIX93hZEXp+tgYEB7du3z3YcZxYfH5/r6CMjIyMmT578QvHkRck8vKjcZkXMPFPp9OnT+fnnn0lJSUGtVmuNKvPz88tWQD7D2LFjuXnzpmaWx+TkZK3OwuL2+++/59rBaWxszJdffql16/OkSZP4888/NbNWqlQqVCqVojH26dNHMxvjP//8w6JFi4DntzrqomMLnn9f33zzTU0HfmRkJJGRkYq9n52dHS1bttS8X3R0NNHR0cDztl2pjq3cZkWsXLmy3nVsQeHaSCGEKG4yK6IQokQaPnw4R44cYfjw4dSvX59y5cphampK7dq1eeuttwgPD+eDDz7Icxvjxo1jy5YttGnThvLly1OpUiW6du3K3r176dixY7Hsx8KFC/n+++/p3r27psiupaUlbm5urF27lo0bNxZJ4ebCqlatGvv372fp0qV06NABS0tLjI2NqVy5Mq1bt+aTTz4hMjJSawTY66+/zpw5c+jRowcNGjSgSpUqGBkZUblyZZo3b05AQAAHDx7UujWnRo0a/O9//6NDhw7ZbtnRNXNzc7Zv386mTZvo0aMHtWrV0oxuaNiwIV5eXixevJjw8HCt9Tw8PNizZw9dunShYsWKVKxYkfbt2/Pjjz9qderlZMuWLYwePRpra2tMTEyoW7cuM2fOJDQ0NM9Z81q2bElUVBSzZ8+mdevWVKlSBWNjY6ysrGjWrBnDhg1j69ateY5kfBFF9bk1atQoWz2dNm3aZKutklMHWE4dW4X53r6IrJ9thQoVcHFx4fvvv9d00uRk8eLFDB06FCcnJ81IRnNzc+rUqUP//v3Zt2+f1u2CL0vpPBS11q1b89NPP9G+fXvKly9PhQoVaNOmDdu3b2fgwIF5rrtgwQIiIiIYNGgQdevWpVy5cpibm2Nra0vPnj1f+nbu/NSuXZt9+/bx0Ucf4enpSePGjalWrRpGRkZUqFCBxo0bM3LkSCIjI+nTp4/WugYGBnz55Zfs2rWLfv36YWdnh5mZGRUrVqR+/fp06dKFefPm5ToTYGF88MEHTJ48GTs7uxJTw8jS0pJ9+/bx1Vdf4e7uTs2aNTExMcHCwoImTZrQu3dvVqxYwcaNG4vsPTdv3swbb7xB9erVCzz5i9BW2DZSCCGKk0FiYmLRzTkvhBA6FBkZqXX7QmFmJhKiKAQHBzN+/HjN44LMGCeEEKJkadq0KTdv3gTAzc2NLl26ANC3b19Fa7qJ7B4+fKjp9Dx37pzWiOWCzHIthCjdSsZPOEIIIYQQQghRQh06dIhDhw4B/zdBjCg+Dx484MMPP9R1GEKIEkrG5AohhBBCCCGEEEIIvSQjtoQQQgghhBAii5iYGF2HIP4/e3t7ueVQCJErqbElhBBCCCGEEEIIIfSS3IoohBBCCCGEEEIIIfSSdGwJIYQQQgghhBBCCL0kHVtCCCGEEEIIIYQQQi9Jx5YQQgghhBBCCCGE0EvSsSVEKeLt7Y2FhcULrRMcHIyFhQXBwcHKBCWEEELvSVshhBCiqFlYWODt7a3rMEQpIB1bQhQBCwsLrX+WlpbY29vTrVs31q9fT1pams5iy7gYCQwM1FkMQghR2mSc7x0dHXny5EmOy3h4eGBhYUFsbGwxR/fipK0QQghllfR2I+MHcn1os4TIyljXAQhRmrz77rsApKWlce3aNUJDQ4mKiuLw4cNs3LhR8ff/+uuvSU5OfqF1fHx8aN26NVZWVgpFJYQQpdetW7dYvny55vxfWklbIYQQRaOstBtCFCfp2BKiCL3//vtajy9evIiHhwe7du3il19+wcXFRdH3r1279guvU6VKFapUqaJANEIIUbpVrlwZU1NTvvzyS4YNG4a1tbWuQ1KMtBVCCPHyylK7IURxklsRhVDQq6++iqurKwBnz57VPH/nzh0CAgJo1qwZNWvWpG7duvTv359jx45l24ZarWbLli107dqV+vXrY2VlhaOjIz179sw2CixrjS1/f3/Gjx8PwMKFC7Vul4yMjASy103577//sLe3p169eqSkpOS4X3PmzMHCwoJNmzZpPf/XX38xceJEXnvtNWrWrEn9+vXx8/Pj/PnzL5Y4IYTQA+bm5rz//vs8fvyY+fPnv9C6L3q+/Pvvv/H396d+/fpYW1vTvn17QkJCiIyMxMLCAn9/f63l//jjD+bMmUPnzp2pX78+NWvW5LXXXmPSpEncvHlTa1lpK4QQongUV7vh7++f622FsbGx2doNCwsLjh8/DkCzZs00bUDTpk01y2RcZ1y/fp3ly5fTtm1brKys8PX1BSApKYkvvvgCHx8fmjRpQo0aNahfvz4DBw7k5MmTL7SvQrwoGbElRDGLjY2le/fu3Lp1C1dXV9544w1u377Njz/+yIEDB1i2bBlDhw7VLD9nzhy++OIL7Ozs6NOnD1WqVOHOnTv89ttvbNu2jWHDhuX6Xt7e3iQlJbFnzx5cXV1p37695jU7O7sc1zE3N+eNN95g/fr1hIeH07t3b63X09PT+e677yhfvjx9+/bVPH/kyBH8/Pz477//8PLyon79+vz999/s3r2bAwcOEBISgru7e2HTJoQQJdKwYcP45ptvCAkJ4Z133tG6CMjNi54v7969i6enJ3FxcbRr1462bdty9+5dZsyYgZubW47vsXv3btatW0eHDh14/fXXMTU15fLly2zevJnw8HAOHz6MjY0NIG2FEEIUp+JoN17Uu+++S0hICDdv3mTs2LGaEbo5jdSdOXMmJ0+exMvLi65du1KxYkUArl69yvz583FxccHLywsLCwtu3rzJnj172L9/P1u3bqVr166FjlGIvEjHlhAKunz5subXj5YtWwIwdepUbt26xXvvvcd7772nWXbChAl4eHgQEBBAly5dsLW1BWDjxo288sornDhxggoVKmhtPyEhIc/39/Hx0VystG/fPtutkrnx8/Nj/fr1hISEZLtYOXz4MLdu3aJ///5UqlQJeP4LzYgRIzAxMeHAgQM0btxYs/zvv/+Ou7s748eP58KFC5iZmRUoBiGE0AfGxsbMnTuXAQMG8OGHH/Ljjz/muXxhzpdz5swhLi6O8ePHs2DBAs3y48ePz7Vja8CAAYwbNy7bOXf//v0MGDCAxYsXExQUBEhbIYQQxak42o0X9f7773Ps2DFu3ryJv78/9vb2uS4bExPD0aNHsy3TsGFDrly5QrVq1bSev3HjBh4eHnzwwQfSsSUUI7ciClGEAgMDCQwMZP78+YwePRo3NzeSk5Px8fHB1dWV+Ph4fv75Z2rVqsW0adO01n311VcZOXIkT58+Zfv27ZrnDQ0NMTExwdg4ez901oajqDg7O9OoUSMOHjzIP//8o/VaSEgI8PyCJsO2bdu4f/8+7777rlZjC9CoUSOGDh3K7du3OXz4sCLxCiGELnl5edG5c2cOHz5MREREnsu+6PkyJSWFnTt3UqlSJWbOnKm1vKOjIwMHDszxfWrVqpXjBY6npyeNGzfm559/foE9zJm0FUIIUThKthtKmzhxYo4dX1WqVMnx2sTOzo7evXujUqmy3QovRFGREVtCFKGFCxcCYGBgQKVKlWjWrBlvvfUWw4cPByA6OhqAtm3bYmpqmm39zp07s2LFCi5cuKB5rn///nz99de8/vrr9OnTh3bt2tGmTRuqVq2q6L4MGjSIOXPmsH37diZMmADAw4cPCQsLw9bWlg4dOmiWzbhv/uLFizlOFf/HH38Az4coe3l5KRq3EELowvz58+nYsSMfffQR7u7uOf4YAS9+vrx69SrJycm8/vrrOd4S0qZNmxxn3VWr1Xz33XeEhITw22+/kZiYSFpamub1nNqgwpC2QgghCkepdkNpzs7Oub4WFRXF119/zenTp/nnn3+y1WD8+++/CzXZlRD5kY4tIYpQYmJinq8/fPgQgJo1a+b4esY06hnLASxYsIB69eqxZcsWvvzyS7744gsMDQ3p1KkTc+fOLdB9+YUxcOBA5s2bR0hIiOZi5ccffyQ5OZmBAwdiaPh/Az7v378PwObNm/Pc5uPHjxWJVQghdO21117D19eXLVu2sGHDBkaNGpXjci96vvz3338BqFGjRo7L5daezJo1i1WrVmFtbY27uzuvvPIK5ubmAJo6KkVB2gohhCgcpdoNpeXW7uzevZthw4Zhbm6Om5sbderUoXz58hgaGnLs2DGOHz/O06dPiyVGUfZIx5YQxahy5crA80LAOblz547WcgBGRkaMGTOGMWPGcP/+fU6cOMHu3bvZvn07ffv25dSpU1haWhZ5rNbW1nTp0oX9+/dz/vx5mjdvrrm1JGP2k6z7dfjwYZo3b17ksQghhD6YPXs2O3fu5LPPPqN///45LvOi58uM+lRZb/XLkFN78s8//7B69WocHR2JiIjQbCPDDz/8kO/7FpS0FUIIUXhKtBuA5keFzCN1MyQlJRUu2P/PwMAgx+c//fRTTE1NOXToEI0aNdJ6bcqUKZq6w0IoQWpsCVGMnJycgOdDinOaHv3IkSMAuTZalpaWeHt78/XXX9OvXz/u3btHVFRUnu9pZGQE5Nyw5SfjomTr1q1cu3aNqKgo2rVrR7169bSWa926NQAnTpx44fcQQojSwtramkmTJnHv3j2WLl2a4zIver5s2LAh5cqV4/LlyzlejOQ0hfr169dJT0/Hzc0tW6dWfHw8169fz7aOtBVCCFH8lGg3ACwsLACIi4vL9tq5c+dyXCejHUhPTy/w+2T2119/0ahRo2ydWunp6flerwjxsqRjS4hiZGNjg7u7O/Hx8XzxxRdar12+fJl169ZhZmam+cXm6dOnHD58OFsDo1arNb/eZ9xakpuMIo45NWz56dGjB1WrVuX777/X1HAZNGhQtuUGDx6MhYUFixYt4tSpU9leV6vVnDhxIsfOPCGEKE0mTpxIrVq1WLVqFbdv3872+oueL01NTenbty///vsvixYt0lr20qVLbNu2Lds27OzsgOe1TjJ3VD169IjJkyfz7NmzbOtIWyGEELpR1O0G/F9n2IYNG1Cr1Zrnb9y4oakJnFVGO1DYW9Xt7Oz466+/uHXrllZsn332GVeuXCnUNoUoKLkVUYhitnTpUrp168aCBQs4evQorVu35vbt25qaJF988QW2trYAJCcn06dPH2xtbWndujW1a9cmNTWVY8eOERMTg7OzMx07dszz/V5//XUqVqzIjh07MDU1xdbWFgMDAwYMGKC5+MmNmZkZ/fr1Y82aNaxYsYLy5cvTt2/fbMtVrVqVTZs2MXjwYLp27UrHjh1p3LgxJiYmxMfHc+bMGeLi4rh+/XqRFSwWQoiSqHz58syePZtx48bl2ElUmPPlnDlzOHr0KF999RVnz56lXbt23L17l507d+Lh4UFYWJhWLSsrKyv69evHDz/8QIcOHXBzc+Phw4ccOnQIc3NzmjZtSkxMjFZc0lYIIYRuKNFudO/enUaNGrFjxw7i4+N5/fXXuX37NuHh4Xh5eeV4S7qbmxs7d+5k8uTJ9O7dmwoVKlClShXGjBlToP0YN24cU6dOpVOnTvTq1QtjY2NOnjzJ77//Trdu3di7d+/LJUqIPEjHlhDFzN7ensOHD7N48WL27t1LVFQUFSpUwNXVlUmTJmnNIFWhQgXmzp1LZGQkp0+fJjw8nHLlymFvb8/8+fMZMWJErjOoZKhSpQrBwcEEBgayY8cOHj16BDyfmTG/ixV4PlX7mjVrSE1NpW/fvtlua8nQsWNHjh8/zldffcXBgwc5deoUxsbGWFlZ0bp1az7++GOt2mFCCFFaDRo0iNWrV2vNcJvZi54va9asyb59+5g7dy779+/n3LlzNGjQgEWLFlGhQgXCwsKynV+XL19OnTp12LFjB2vWrKF69ep0796dWbNmMWTIkGwxSVshhBC6U9TthpmZGbt27eKjjz7S1ECsX78+n376KZ06dcqxY2vw4MHEx8fz3XffsWLFClJTU6ldu3aBO7ZGjBiBqakpq1atYuvWrZibm9OuXTtWrFjBTz/9JB1bQlEGiYmJ6vwXE0IIIYQQJc28efNYsmQJy5YtY/jw4boORwghhBCi2EmNLSGEEEKIEu7vv//O9tzFixdZvXo1JiYmdO/eXQdRCSGEEELontyKKIQQQghRwnl6elK7dm0cHR0pX748f/75J/v27ePZs2fMnz8fKysrXYcohBBCCKETciuiEEIIIUQJt3jxYsLDw7l27RoPHz6kYsWKtGjRgnfeeYdu3brpOjwhhBBCCJ2Rji0hhBBCCCGEEEIIoZekxpYQQgghhBBCCCGE0EvSsSWEEEIIIYQQQggh9FKp6diKjo4mKCgoz2X8/f2JjIzM9nxwcDCJiYkKRSaEEKKkiIyMxN/fP9vzBw4cIDg4GJA2QQghSrvY2Fi8vb0JDg7mwIEDADRt2lTHUQkhhCisUtOx1bRpU6ZOnVqodUNCQkhKSiriiER+VCqVrkMotSS3ypHclk4eHh74+fkB0ibokhxfypHcKkdyq7/8/Pzw8PDQdRilmhwf+ZMc5U3ykz/JERjrOoCXERsby9ChQ2nSpAl79+6le/furFq1ilOnTjFz5kysrKywsbEhOTmZVatWAfDTTz/x1VdfERcXx7p167h16xYxMTG8/fbb1K9fn9WrV+f4XuvPrC/OXdMbI5xH6DoEIYTI1dWrV5kwYQKmpqakpaUxduxY7t69y+jRo7l06RK+vr6MHz+e4OBgbty4Qdu2bQvUJrBsWbHuR1nhoOsASrFccztlSjFGIUTJEhgYiJ2dneaHDYCYmBhmzZrFl19+ye3bt5k/fz4GBgbUqFGDlStXUq5cuRy35bShSnGFrWec4biuYyjpJEd5k/zkTz9yFD1cuR+O9X7E1o0bN/jss8/YvHmz5rmZM2eyfv16tm/fTo0aNbSWt7W1Zfv27YwbN45Nmzbh5uZG06ZNWbt2be4XMEIIIfTS/v376dmzJ6GhoYSHh1O1alX+/vtvVq1axb59+zQ/emSQNkEIIcqu/fv388knn7Bx40bq1KnDjBkz2LJlC6GhoTg7O7NhwwZdhyiEECIHej1iC6Bx48ZYWFhoPXf//n3q1q0LQOvWrblx44bmtRYtWgBgZ2fH8eN60K1Zwr3ssEcZNqkcya1yJLfKcHAo+vE6gwcPJigoiDFjxlCrVi3at2+Pk5MTxsbGGBsbY2BgUOTvKYQ+kfNZ0ZA8Fj0l2oT8zJgxg+DgYCwtLbl37x43b95k8ODBADx9+pQOHToUe0xCCFFavGxbmVe7oPcdW4aG2QedWVpacu3aNerWrcvp06e1Xst8EaNWqwE0t6iIF/cyf3SoVCqd/NFSFkhulSO51S8mJibMmTMHgAkTJpCcnJxvZ5a0CaIskfPZy5N2ofTYtWsXo0ePZunSpbz22mvUqVOHkJAQqlR5fpthSkqKjiMUQgj9pWRbqfcdWzlZuHAhw4cPp0aNGlhbW2Nqaprn8n369GHy5Mk4OjqycOHCHJeRWlJCCKF/vv/+e7Zu3YqhoSFmZmaaHzTyUpA2QeoSKUM6CJQjuRUif3Xq1GHTpk2MGDGCuXPnsmjRIoYMGUJ6ejoGBgZMmTIFd3f3HNdVsnaMPpNzT/4kR3mT/ORPcgQGiYmJ+f+Vr2dSU1MxMTEBYP78+VhYWDBhwgQdRyWykgNQOZJb5UhuhVCOHF/KkdwqR3IrRO7k+Mif5Chvkp/8SY5K6Yitw4cPs3TpUtLS0qhWrRrffPONrkMSQgghhBBCCCGEEEVM72dFzBAbG4u3tzcAnp6ehIeHs2/fPrZu3UqlSpV0HJ0QQoji9O6779KtWze2bdtW4HWaNm2qYERCCCH0UXR0NJGRkboOQwghRB5K5YgtJaw/s17XIeic1BkTQuiL8PBwoqOjlXuDZcuU23YZVrYH0ReA1HYTotjFxMRw48aNfGdEdNpQpZgi0jfOIBPR50NylLeizY/UwyudSs2IrQx//fUXPXv2pEePHgwYMICEhAQOHDjA1KlTNctMmTKFn3/+mYcPH/L222/Ts2dPunXrxtmzZ3UYuRBCiKIwZcoU7ty5g7e3N7a2tprn/f39Nb+6r1mzBnd3d3x8fFiWpZMqJiaGnj17cu3ateIMWwghRDFJT09nwoQJeHl54e/vT6dOnYiNjWXBggV0794dT09PNm3aBMDy5csJCQnB29ubK1eu6DhyIYQQOSl1I7Y+/PBDpk2bhpubGxs3bmTp0qXMmzePjz76iOTkZABOnTrF0qVLmT9/Ph4eHgwaNIjbt28zePBgDhw4oOM9EEII8TKWLVvGwYMHCQsLy/H2wsuXL7N9+3b27NmDmZkZaWlpmtf279/P6tWr2bhxI5aWlsUZthBCiGKyZ88e0tLSiIiIIDExkWbNmnHo0CFu3bpFeHg4aWlpdOvWjR49ejBx4kRu3LjB+++/r+uwhRBC5KLUdWypVCratWsHQNu2bQkNDcXQ0JAePXoQGhqKWq2mR48eGBoa8ttvv3H06FG2bNkCQFKSDEvMi0ql0ottiuckt8qR3CpD6dlc1OrnkwBfuXIFFxcXzMzMADAyMtIsM2PGDIKDg6VTS5RIL3vukXOXciS3RU/JNuGPP/6gdevWAFhYWODg4MDDhw85deqUpmbvo0ePiIuLUywGIYRulNbzdWndr8zyahdKXceWg4MDJ06cwM3NjaioKBo2bAiAn58f06ZNQ61WExQUBICjoyPNmjWjb9++AKSkpOgsbn1Q1H9gyLSkypHcKkdyq1/UajXJycmYmJgQHR2Nn58fTZo0YdWqVaSkpGBqakpaWpqmc2vXrl2MHj2apUuXSjF5UeK8zLlHzl3KkdzqHwcHB0JDQxk5ciSJiYmoVCoqVqxIx44dWbJkCQCpqakYGxujUqm0RvYKIfRbaTxfSztUCju2PvnkEyZPnsyiRYuoUKECq1atAqBu3bo8e/ZM83+A6dOnM336dNatW4darcbJyYlPP/00x+1K4XQhhNA//v7+eHh44OjoiLW1NQCNGzfmrbfeolu3bpQvXx4PDw+m/P+i3HXq1GHTpk2MGDGCuXPn8vrrr+e8YSnirQj5w0wIURy6d+9OeHg4Xl5e2NvbY2trS/fu3UlISKB79+4YGRlhbm5OcHAwbdu25dtvv2Xo0KF8/PHH1K9fP8dtSkHqnMl5PX+So7xJfkRBGCQmJqp1HYQom+QkpRzJrXIkt0IoR44v5UhulSO51U+pqamYmJiQkJCAm5sb58+fx9Cw1M2rpXNyfORPcpQ3yU/+JEelcMSWEEIIIYQQQuRl1KhR/PPPP/z777/MmzdPOrWEEEKP6cUZPDY2VlPIMT+RkZFER0drHgcHB5OYmKhQZEIIIZT07rvv0q1bN7Zt21bgdXKqjXXgwAGCg4OLMjQhhBB6bOPGjezZs4fIyEh69+6d63JZry2EEEKUPKVuxNaxY8ews7PDyckJgJCQENq3b4+FhUWB1ler1RgYGGR7fv2Z9UUZZrGS+mBCCH0VHh5eJBcUHh4eRRBNJsuWFe32BAAldhC91FQToszKem2RG6cNVYopIn3jDMd1HUNJp7scSW04UVroxYgteD7l7tixY3F3d2fOnDk8e/aMqVOn4uPjg6enJxEREdy7d4+QkBCWLVuGt7c3O3bsICYmhrfffpt33nkHgHXr1tGtWze6devG559/Djz/JaZv374MGzaMadOm6XI3hRBC/H9Tpkzhzp07eHt7Y2trq3ne39+fyMhIANasWYO7uzs+Pj4sy9LZFBMTQ8+ePbl27RrBwcEEBgYC4O3tzZw5c3jzzTfp0qULd+/eBZ63Dx06dMDPz4/hw4fLCC8hhCil0tPTmTBhAl5eXvj7+9OpUyeuX7/OjBkz8PLywsPDgwMHDmS7tnjw4IGuQxdCCJEDvRmxFRsby65du6hcuTK9evVixYoVWFtbExQUxOPHj/H09OTYsWP4+vpiZ2eHn58fAGvXrmXlypXY29tz9epVQkND2bNnD4aGhvj6+hITEwNAfHw827dvx9TUVJe7qQiVSqXrEHJVkmPTd5Jb5UhulZG16OWyZcs4ePAgYWFhOd5eePnyZbZv386ePXswMzPTmo59//79rF69mo0bN2Jpackvv/yitW6zZs2YM2cOgYGB7Ny5k379+rFmzRoOHTqEiYlJnreliLKntBzzpWU/SiLJbdFTshDynj17SEtLIyIigsTERJo1a0Z4eDhJSUlERESQkJCAh4cHv/76a7ZrCyFKE305d+lLnLpUFnKUV7ugNx1bDg4OmtsJnZ2dCQ4OxtTUlKNHjwLPZza5d+9entu4dOkSKpWKnj17ApCUlMSNGzeoXLkyzZs3L5WdWqDsHwYvQ2ZvUI7kVjmSW91Tq59P5nvlyhVcXFwwMzMDwMjISLPMjBkzCA4OxtLSMsdttGjRAgA7Oztu3rzJ9evXady4sWZbrVq1UnIXhJ4pDce8nLuUI7nVP3/88QetW7cGwMLCAgcHB54+fUq7du0AqFatGpaWlvleWwih7/Th3CXn2PxJjvSoY0ulUpGUlETlypU5e/YsgwYNwtDQkMmTJwOQkpKCqakppqamWr/aZ37cuHFjHB0d2bZtGwYGBqSnp5Oens6JEye0LohyInWqhBBCd9RqNcnJyZiYmBAdHY2fnx9NmjRh1apVmvN/Wlqa5ly+a9cuRo8ezdKlS3Mc7ZW5lqJaraZOnTpcuXKFp0+fYmJiwtmzZ/P+A0FqLilC/jATQhQHBwcHQkNDGTlyJImJiahUKszMzIiKimLkyJEkJCRw//59qlWrlu3aIjdSqyhncl7Pn+RIiJenNzW27O3tCQgIwN3dnVatWjFp0iTi4uLw8fGhZ8+ejBs3DoAuXbqwZcsWhg4dSmJiIn369GHy5Mm8++67NG7cGG9vb3r06IGPjw/9+vXT1FYRQghRcvn7++Ph4YG/vz/W1tbA8x8r3nrrLbp164aPjw/Lly/XLF+nTh02bdpEQEAAp06dynf71atX5+2338bDwwNfX18qVapUakfxCiFEWde9e3eMjIzw8vJi5syZ2Nra0rt3bypWrIiXlxf9+/dn4cKFGBoaZru2EEIIUfIYJCYmqnUdhCib5NcJ5UhulSO5Lb1SU1MxMTEhLS2NXr16sWDBApo3b67rsMoUOb6UI7lVjuRWP2Wc8xMSEnBzc+P8+fMYGurNb/56Q46P/EmO8ib5yZ/kSI9uRRRCCCGU9OWXX/Lzzz/z6NEj3N3dpVNLCCFKsVGjRvHPP//w77//Mm/ePOnUEkIIPVamO7ZiY2MZN24cYWFhug5FCCFEEck4t/v6+mJlZYWHhwcrV67U3LKemJhIWFiYZoar4OBgbty4wfvvv8/06dN1GboQQohi0rVrV00bIYQQQr+V6Y6tF7H+zHpdhwBIEXshhCiozFOzr1q1StOxlZSUREhIyMtN3b5s2UtGV0ZIkX0hRAn1Um1ADpw2VCnS7ZUeznBc1zGUdAXLkUxQIETu9K5jKzY2lsGDB9OkSRMuXryIr68vcXFxnD17FisrK8aMGUNISAirVq0CoGnTpsTExJCamsqMGTO4cuUKJiYmTJ06lQYNGvD48WMmTZpETEwMLi4uLFiwQMd7KIQQoigEBgZiZ2eHkZERd+/exdvbm1atWvHkyRNiYmLw9vZm+PDhWuucOHGC+fPnY2BgQI0aNVi5ciXlypXTzQ4IIYR4Ienp6UyaNAmVSkW9evW4dOkSGzdu5KuvviImJoa0tDTee+89PDw8NG2En58fTZs2ZciQIRw7doz09HS+//57zM3NmT9/Pj///DN2dnY8fPiQqVOn0qFDB13vphBCiCz0rmML4Pbt24SHh2NsbEyDBg3YvXs3gYGB9OrVi0uXLuW4zsaNGylXrhwREREApKWlERcXR3x8PLt376ZChQq0bNmSd999l8qVKxfn7rwQlUql6xCKVGnbn5JEcqscya0ylCp6OXDgQBYsWKC57Tw2NpbLly9rHgcHBwOgVquZMWMGoaGhVK1alRUrVrBhwwb8/f0ViassKMyxIseXciS3ypHcFr3CtAl79uwhLS2NiIgIEhMTadasGeHh4SQlJREREUFCQgIeHh78+uuv2dZ1d3dn5syZ+Pv7c+jQIWxsbDh58iQHDhwgNTUVV1fXotgtIQqtLJ9nyvK+F1RZyFFe7YJedmw5ODhQsWJFAKpWraop8GtjY8Orr77K+fPnNcuq1c8nfbx8+TLdunXTPG9kZKTZVqVKlTTrJyYmluiOrdI024HM3qAcya1yJLelV0JCAjdv3mTw4MEAPH36VH6Zf0kveqzI8aUcya1yJLclxx9//EHr1q0BsLCwwMHBgadPn9KuXTsAqlWrhqWlJffu3cu2bsuWLQGws7PjwYMHJCcn06pVKwwNDTEzM8PJyan4dkSIHJTV84ycY/MnOdLTji0DA4NcH1+8eJH4+HgA4uPjuXPnDgCOjo4cOnQIT09P4PmIrZy2ldERlpXUthJCCP1lamrKs2fPMDY2xtTUVNMGZFatWjXq1KlDSEgIVao8r5WSkpKS8waldpQQQpQ4Dg4OhIaGMnLkSBITE1GpVJiZmREVFcXIkSNJSEjg/v37VKtWLdu6ma8J1Go1Dg4OrF+/nvT0dJ49e0Z0dHSe7y31j3ImF9z5kxwJ8fL0smMrL46OjkRFRdG9e3datGhBjRo1ABg6dCjTp0+na9eumJmZMWXKFBo0aKDjaIUQQhSHfv36MWDAANq1a8e0adOoXLkyQ4YMYcCAAZplDAwMWLRoEUOGDCE9PR0DAwOmTJmCu7u7DiMXQghRUN27dyc8PBwvLy/s7e2xtbWld+/eqFQqvLy8ePbsGQsXLsTQ0DDfbTVt2pQ2bdrg7u6Ora0tr7zyCqampsWwF0IIIV6UQWJiYs5DlIRQmPw6oRzJrXIkt0IoR44v5UhulSO5LVlSU1MxMTEhISEBNzc3zp8/X6COrLy29d9//9GxY0fCwsI0P5qLgpHjI3+So7xJfvInOYLCneVLocDAQE0BYSGEEPojMjKy0AXeX2ZdIYQQJc+oUaPo0qULTk5OzJs3r9CdWgCzZs3C29sbT09PRo0aJZ1aQghRQpW6WxGFEEIIIYQQZdPGjRuJjY1l3Lhx9O7d+6W2tWjRoiKKSgghhJL0tmMrPT2dSZMmoVKpqFevHpcuXWLjxo189dVXxMTEkJaWxnvvvYeHhwd//fUXkydPJi0tjUqVKrFy5UqqVavGunXrWLduHbVr18bMzAw7O7tc32/9mfXFuHc5kwL2QggBV69eZcKECZoi8GPHjuXu3buMHj2aS5cu4evry/jx4zl+/DiBgYGaelnr1q3DysqKU6dOMXPmTGrWrImtrS0ASUlJ9OzZk6NHjwKwfft2VCoVs2fPzjmIZcuKaW/1iBTUF0KUQN999x1hYWEYGBhQo0YNbty4QVxcHOvWraNRo0bs27ePhQsXYmJiQtOmTfn8889ZsmQJtra2DBw4EBcXF4YNG8Y777xDjx492LhxY44jt5w2VNHB3ukDZziu6xhKurxzJBMTCJE/vb0Vcc+ePaSlpREREUFgYCDXr18nPDycpKQkIiIi2L59OwEBAajVaj788EOmTZvGnj176NGjB0uXLuXevXt8++23HDx4kC1btmhmTxRCCFGy7d+/n549exIaGkp4eDhVq1bl77//ZtWqVezbt49Vq1YB0Lx5c0JDQ9mzZw99+vRhzZo1AMycOZO1a9fy3XffYWlpCUCVKlVo1qyZpmNr8+bNDBkyRDc7KIQQokh8/vnnnD59mnXr1lGuXDlsbW3Zvn0748aNY9OmTaSnpzNz5ky+++479u7dS2JiInv37qVz584cPnyYO3fuYGdnx7Fjx3j8+DHJyclyO6IQQpRAejti648//qB169YAWFhY4ODgwNOnT2nXrh3wfNp2S0tL7t27h0ql0jzftm1bQkNDuX79Ok2aNMHMzAyANm3a6GZHXoBKpdJ1CEWuNO5TSSG5VY7kVhkFLXo5ePBggoKCGDNmDLVq1aJ9+/Y4OTlhbGyMsbGxZsr233//nQULFpCcnMzDhw9xcnIC4P79+9SvXx943ib88MMPALz99tssX74ca2trypUrh729vQJ7WXoV1XEhx5dyJLfKkdwWvZcthKxSqbhz5w6HDx/GyMgIgBYtWgBgZ2fH8ePHuXfvHtWrV6datWrA8zbh6tWrdO3alejoaA4fPoyPjw+7du3iyJEjuLi4vNxOCVEIcn6RHBREWchRXu2C3nZsOTg4EBoaysiRI0lMTESlUmFmZkZUVBQjR44kISGB+/fvU61aNRwcHDhx4gRubm5ERUXRsGFD6tSpw5UrV3j69CkmJiacPn26xM8kUNLje1Eye4NyJLfKkdzqnomJCXPmzAFgwoQJJCcnazqzMlu0aBHTp0/HxcWFb775hnPnzgFgaWnJtWvXqFu3LlFRUZrlmzdvTnx8PIsXL2bo0KHFsi+lSVEcF3J8KUdyqxzJbcnk4ODA5MmTGTRoEJs3bwbQaivUajXVq1fn3r17JCQkUK1aNaKionjjjTcwMjKibt26rFmzhg0bNhAXF0dQUBAzZ87U1e6IMqysn1/kHJs/yZEed2x1796d8PBwvLy8sLe3x9bWlt69e6NSqfDy8uLZs2csXLgQQ0NDPvnkEyZPnsyiRYuoUKECq1atonr16owaNQoPDw9q1apF9erV83w/qW8lhBAlw/fff8/WrVsxNDTEzMwMtVqd43JvvfUWU6ZMoX79+tSqVUvz/Oeff86IESOoXr06tWvX1lrHz8+PBQsWsHLlyryDkHpSQghR4nXt2pUKFSowaNAgKlasmO11Q0NDAgMDeeuttzQ1trp16waAm5sb3377LTY2Nri5ubFkyZI8R2xJHaScyQV3/iRHQrw8g8TExJyvCPRAamoqJiYmJCQk4Obmxvnz519qSl9RvOQkrhzJrXIkt6Xb1q1b+euvv/jggw90HUqZJMeXciS3ypHcCpE7OT7yJznKm+Qnf5IjPR6xBTBq1Cj++ecf/v33X+bNmyedWkIIIQpt7ty5HD9+nK1bt+o6FCGEEEIIIUQB6XXH1saNG19o+aZNmxITE0NwcDA3btzg/fffVygyIYQQ+uajjz7SdQhCCCGKUGxsLOPGjSMsLKxQ68s1gxBC6Ae97thSq9U5FgxWwvoz64vlfXIjNb6EEKIEWbZM1xFok5pfQgihU04bqug6hBLKGY7rOoaSSeqyCVF09K5jKzY2lqFDh9KkSRPi4uKoWLEijx8/BuCLL76gXr16zJ07l9OnT/Pw4UPc3d1z/RX+zz//JCAggB07dgDPZ9CysrKS2bCEEEIPxMbGMnjwYJo0acLFixfx9fUlLi6Os2fPYmVlxebNm1mwYAHHjh3j2bNnDBkyhKFDh3L8+HECAwNJT0/HwMCAdevWYWVlhb+/PxUrVuTGjRvExcWxbt06GjVqpOvdFEII8RIePXrE2LFjUalUdOjQAT8/P6ZNm0ZaWhqpqaksX76cxo0bExcXx9SpU3ny5AmGhoasWbNGs4309HRmzZpF1apVeffdd3W4N0IIIXKidx1bADdu3GDXrl3MmDGDESNG4OrqyoULF/jwww8JDg5m2rRpVKxYkfT0dLp27aqZ1j2r+vXro1aruX79OnZ2duzcuZP9+/frYI+EEEIUxu3btwkPD8fY2JgGDRqwe/duAgMD6dWrF6tWreLWrVuEh4eTlpZGt27d6NGjB82bNyc0NBSAb7/9ljVr1miKxdva2rJo0SKCg4PZtGkTCxYs0OXuCSGEeEmxsbHs2rWLypUr06tXL3r06MGPP/6IkZERERERBAUFsXr1ambPns2wYcPw8fEBnndmATx58oQRI0bQvXt3Bg4cqMtdEUIIkQu97Nhq3LgxFhYW/Pbbb3z66aea5zNuS9ywYQNhYWEYGBhw48YN4uPjc+zYAhg+fDgbN27E1dWVtm3bUqFChWLZhxelUql0HYIiSut+lQSSW+VIbpVRmNlcHBwcNFO4V61alebNmwNgY2NDUlISp06dwtvbG3j+q31cXBzp6eksWLCA5ORkHj58iJOTk2Z7LVq0AMDOzo7jx/Xn3onS9J0sTftS0khulSO5LXpFNcOXg4MDFhYWADg7O/PLL7+wbNkyEhMTSUlJwdzcHIDLly/j7u6uWS9jUqqdO3fi7OwsnVqiyGU+b8g5JG+Sn/yVhRzl1S7oZcdWRkPj6OjI+PHjadWqFQApKSkkJiayYcMGTp48iYGBAZ6enqjV6ly35e3tzZIlS7h06RKzZs0qlvgLozRO3ynTkipHcqscyW3JkrXOYubHzs7O/PPPPyxZsgSA1NRUjI2N8fX1Zfr06bi4uPDNN99w7ty5HNfPq+0oaUrLd1KOL+VIbpUjuS3ZVCoVSUlJVK5cmbNnz1KxYkV69+7NwIED2bt3L8uXLweeX1ccOnSIHj16AP83YsvPzw+1Ws306dNZvHhxsdX3FaVfxnlDziF5k/zkT3Kkpx1bGRYsWEBAQABJSc8L73l4eDBp0iScnJzw9PSkXr16VKpUKc9tGBsb4+XlxcGDB2nWrFmuy0nxdiGE0C9mZmZYW1vTvXt3jIyMMDc3Jzg4mLfeeospU6ZQv359atWqVbiNS7F2IYTQC/b29gQEBPDHH3/QsWNHPDw8CAgIYOfOnVoXgvPmzWPKlCl89dVXGBkZadXYmjVrFitWrGDs2LGsWLECY+Psl1BSCDxncsEthCgOBomJifrzk7RCFi5cSK1atRgyZIiuQylTpKFTjuRWOZJbIZQjx5dyJLfKkdwKkTs5PvInOcqb5Cd/kiM9H7FVFMaPH8+tW7fYvn27rkMRQgghhBBCCCGEEC/AUNcBFKV3332Xbt26YWlpWeB1Bg4cyCeffIKpqamCkQkhhNCF2NhYTfH44lhPCCFE6bNy5UpdhyCEECIPpWrEVnh4ONHR0Tm+plarcyz2eOzYMezs7LRmxcrJ+jPriyTGvEgdLyGE0BPLluk6AqnzJYQQxWTVqlWMGzcuz2WcNlQppmj0jTPozyTDRUrqrglRfEpNx9aUKVO4c+cO3t7eXLhwgbi4OCIjI1m6dCmVK1fG0tKSzp0788UXX1CuXDmsra1ZuHAhISEhmJubExISwpYtW6hataqud0UIIUQRevToEWPHjkWlUtGhQwf8/PyYNm0aaWlppKamsnz5cho3boxKpWLChAlUqFCBJk2a6DpsIYQQL2Hx4sXY2toycOBAXFxcGDZsGO+88w49evRg9OjRrF27ltTUVCwsLFi/fj3ly5fH29ub1q1b89tvv3H//n22bdvGzz//zN27d/H29qZVq1bMnTtX17smhBAii1LTsbVs2TIOHjxIWFgYTZs21TwfHx/P9u3bMTU1ZdCgQSxZsoQWLVqQnp6OoaEhvr6+2NnZ4efnp8Pon1OpVLoOodiVxX0uLpJb5UhulaFU0cvY2Fh27dpF5cqV6dWrFz169ODHH3/EyMiIiIgIgoKCWL16NXPmzGHWrFl06tSJTZs2cf78eUXiKSql9XtYWverJJDcKkdyW/Retk3o3Lkza9aswc3NDTs7O44dO8bgwYNJTk7G09OTvn37AvDhhx/yww8/aCaRatasGXPmzCEwMJCdO3fyzjvvsGDBAsLCwl56n0TZ8iLnBTmH5E3yk7+ykKO82oVS07GVm+bNm2vqZ82dO5cVK1awatUqXFxcGD58uG6Dy6KszWQgszcoR3KrHMmt/nFwcMDCwgIAZ2dnfvnlF5YtW0ZiYiIpKSmYm5sDzz/bNm3aANC2bdsSP6lIafweyvGlHMmtciS3JVOLFi2Ijo7m8OHD+Pj4sGvXLo4cOYKLiwtRUVEsX76c1NRU/vnnHypUqKC1HoCdnR03b97UVfiiFCjoeUHOIXmT/ORPclQGOraMjIw0/7exsWHZ/6+L0qJFC3r37o2pqSlpaWn5bkfqXwkhhH5SqVQkJSVRuXJlzp49S8WKFenduzcDBw5k7969LF++HHj+B+jp06fp0KEDUVFReW9U6lsJIUSJZmRkRN26dVmzZg0bNmwgLi6OoKAgZs6cyccff8z27duxtbVl9uzZqNVqzXqZa/JmPG9qasqzZ88wNs790knqKeVMLriFEMWhVM2KmJ+PPvqI7t274+XlRZcuXahatSpdunRhy5YtDB06lMTERF2HKIQQoojZ29sTEBCAu7s7rVq1Yvz48XzxxRcMGDCAY8eOaZabM2cO8+bN44033uDy5cs6jFgIIURRcHNz4+HDh9jY2ODm5sb58+dxcXHB19eXN998k8GDB5OQkJDvdvr168eAAQNYvHhxMUQthBDiRRkkJiaq819MiKInv+AoR3KrHMmtEMqR40s5klvlSG6FyJ0cH/mTHOVN8pM/yVEZG7ElhBBCCCGEEEIIIUoP6dgSQgghhBBCCCGEEHqp1BePLyrrz6xX/D2kQL0QQhQdtVqtVQS4SP3/iUh0RorXCyFEieK0oYquQyihnOG4rmMofjKZgBDFq9SM2IqNjaVz585MmjQJNzc3PvjgA549e8bUqVPx8fHB09OTiIgIAFxdXXn69Cnnz5+nWrVqJCYmcv36dfr376/jvRBCCPGi5s+fj6enJz4+PoSEhODo6EhAQADe3t7cv3+f3r174+3tTefOnfnll18ANK97e3tjbW1NTEwMrq6uPHnyBIATJ04wZswYXe6WEEKIYpKens6ECRPw8vLC39+fTp06cf36dWbMmIGXlxceHh4cOHBA12EKIYTIRakasRUfH8/u3bupUKECLVu2pHbt2lhbWxMUFMTjx4/x9PTE09OTNm3acPLkSX799Ve6d+/O0aNHuX//Pp06ddJp/CqVSqfvrwtlcZ+Li+RWOZJbZRSm6OW+ffv4/fffiYiIwNDQkLS0NCZOnMj48eOpU6cOqampfPfdd5iZmXH58mWmTZtGeHg4ixYtAmDlypU0adKEpk2b0qdPH3bs2MHgwYPZsGEDw4YNK+pdLDKl+TtYmvdN1yS3ypHcFr3iLIS8Z88e0tLSiIiIIDExkWbNmhEeHk5SUhIREREkJCTg4eHBr7/+qtxIYFGqvOg5Qc4heZP85K8s5CivdqFUdWw5ODhQqVIlAGxsbIiOjiY6OpqjR48CkJqayr179+jUqRNHjx4lJiaGjz/+mDVr1vDgwQNmzJihy/DL3EwGMnuDciS3ypHcliyXL1+mc+fOGBo+H4BsZGSEtbU1derUAeDhw4fMnDmTW7duYWhoSHx8vGbd7du3c/bsWb799lsAhg4dyvDhw/Hx8eHy5cu4uroW+/4UVGn9DsrxpRzJrXIkt/rvjz/+oHXr1gBYWFjg4ODA06dPadeuHQDVqlXD0tKSe/fuUaNGDV2GKvTEi5wT5BySN8lP/iRHpaxjK+svKM7OzjRq1IjJkycDkJKSgqmpKZ06dWLp0qVYW1vj6OjIX3/9xd27d3n11Vdz3bbUvxJCiJLH0dGRzZs3M3LkSAwMDEhLS9N0cgFs27aNRo0asXbtWi5dusSAAQOA5yO9fvjhB7Zs2aJZ3srKCisrK2bPnq1ZLldS40oIIUoNBwcHQkNDGTlyJImJiahUKszMzIiKimLkyJEkJCRw//59qlWrlus2pKZSzuSCWwhRHEpVx1ZWbm5urFy5Eh8fHwwMDLCysmLNmjVYWFhgZGSEi4sLAPb29vLrixBC6CFPT0+ioqLw8PCgfPnyDBo0SOt1d3d3Ro0aRVRUFM7OzprnZ86cSdWqVenbty8Ay5cvp169eowYMYL+/ftz+fLlYt0PIYQQutO9e3fCw8Px8vLC3t4eW1tbevfujUqlwsvLi2fPnrFw4UKtH06EEEKUHAaJiYlqXQchyib5BUc5klvlSG5Lt+PHj7Np0yZWr16t61DKJDm+lCO5VY7ktnRITU3FxMSEhIQE3NzcOH/+vHRkFQE5PvInOcqb5Cd/kqNSMCtiYGAgwcHBBV6+adOmCkYjhBBCX3399dfMnj2bmTNn6joUIYQQxWzUqFH06NGDPn36MG/ePG7evIm3t7euwxJCCFEApfpWRCGEEKKgxo4dy9ixY3UdhhBCCB3YuHGj1uPY2FgdRSKEEOJF6UXH1uLFi7G1tWXgwIG4uLgwbNgw3nnnHXr06EGTJk2Ij49nz549/PnnnyxZsgRXV1fOnj3LrFmzMDQ0xMbGhq+++gpzc3PNNh8+fMjUqVO5e/cuqampLFiwgFatWuUaw/oz6xXfTylQL4QQBZeens6kSZNQqVTUq1ePS5cu8eWXX/LRRx/x7NkzHj9+zKeffoqLiwvBwcGEhoZiaGjI1atX+fjjj/nf//7HH3/8wcCBA5k4ceKLtQvLlhXrvkqxeiGEeDHz58+nUaNGvPXWW6jVajp27MjKlSv56KOPSE1NpXz58qxcuZLq1aszfvx4YmNjSUpKYujQoYwePRqAx48fM2nSJGJiYnBxcWHBggW5vp/ThirFtWt6xhmO6zoG5cikAUKUDHpxK2Lnzp05fPgwd+7cwc7OjmPHjvH48WOSk5OpXr06ZmZmBAcHs2jRIr7++msApk+fzldffUV4eDj29vbZfoVZtmwZHh4e7N69mw0bNvDuu+/qYteEEEIU0p49e0hLSyMiIoLAwECuX79OxYoV+e677wgLC2PVqlXMmzdPs3xaWhrBwcF88cUXTJs2jZUrV3Lw4EFNPS1pF4QQovQYNmwYmzZtAuDo0aM4OTkxc+ZMgoKCCA0NZciQISxZsgSAhQsXEhoayv79+1mxYgX//fcfAPHx8SxYsICDBw8SFhbGw4cPdbY/QgghcqcXI7ZatGhBdHQ0hw8fxsfHh127dnHkyBHNrIYtWrQAwM7OjgcPHgBw//59TQG1tm3bsnfvXq1t/vbbbxw9epQtW7YAkJSk+952lUql6xCKXVnc5+IiuVWO5FYZL1r08o8//qB169YAWFhY4ODgwMOHDxk3bhy3bt3C0NCQ+Ph4zfLNmzcHwMbGhgYNGlChQgUADAwMgJLZLmQoS9+5srSvxU1yqxzJbdF72ULItWvXpkKFCly9epWNGzcybtw43nzzTcaPHw88/7HDzs6O9PR0lixZQlRUFIaGhty/f5+7d+9qYqhUqRLwvO1ITEykcuXKL7djolQpqmNfziF5k/zkryzkKK92QS86toyMjKhbty5r1qxhw4YNxMXFERQUxMyZMzlz5ozmogRArX4+yWPVqlU1swNERUXRsGFDrW06OjrSrFkzzVTvKSkpxbdDuShrMxnI7A3KkdwqR3Jbcjg4OBAaGsrIkSNJTExEpVLxyy+/0KhRI9auXculS5cYMGCAZvnMbUXm/2coie1ChrLynZPjSzmSW+VIbkuuESNGsHjxYq5fv46zszOOjo6sXr0aW1tb4Pl5PiYmhl9//ZXw8HCSk5Np3bq15noia1uR8bwQGYri2JdzSN4kP/mTHOlJxxaAm5sb3377LTY2Nri5ubFkyRJcXFw4c+ZMjssvXryY8ePHY2RkhLW1NQEBAVqvT58+nenTp7Nu3TrUajVOTk58+umnub6/1L8SQoiSpXv37oSHh+Pl5YW9vT22trY4Ozszffp0oqKicHZ2fqHtvVC7IDWvhBCixPP09CQgIIBJkyYBEBQUxJQpU3j69CkAgwYNok+fPhgYGNCtWzcaNGiApaVlod5Lai3lTC64hRDFwSAxMVF+ehA6IQ2dciS3ypHcliypqamYmJiQkJCAm5sb58+fx9BQL8pHihzI8aUcya1yJLclV3p6Op06dSI0NJQqVaS4uy7I8ZE/yVHeJD/5kxzp0YgtIYQQIqtRo0bxzz//8O+//zJv3jzp1BJCCAHAxYsXCQgIYNCgQdKpJYQQpZx0bGXStGlTYmJidB2GEEKIAso64y1AbGws48aNIywsLN/1IyMjqVKlCk5OTkqEJ4QQQkdeffVV9uzZo+swhBBCFINS3bGlVqtzLBBcGOvPrC+S7WQmdbuEEEK3jh07hp2d3Yt3bC1bpkg8uZKaXkIIUaI5bZBRYTlzhuO6jUDqnwlR+pW6jq3Y2FiGDh1KkyZNiIuLo2LFijx+/BiAL774gnr16jF37lxOnz7Nw4cPcXd356OPPtJx1EIIIYrS48ePmTRpEjExMbi4uDB8+HCmTZtGWloaqampLF++nOrVqxMSEoK5uTkhISFs2bKFK1euMH/+fAwMDKhRowYrV66kXLlyut4dIYQQRUStVhMQEEBMTAxpaWm89957nD59mvj4eB48eMCff/7JkiVLcHV15cqVK7z//vukpqZSvnx5Vq5cSfXq1XW9C0IIIbIodR1bADdu3GDXrl3MmDGDESNG4OrqyoULF/jwww8JDg5m2rRpVKxYkfT0dLp27cq1a9eoW7eursMWQghRROLj49m9ezcVKlSgZcuWzJw5kx9//BEjIyMiIiIICgpi9erV+Pr6Ymdnh5+fH2q1mhkzZhAaGkrVqlVZsWIFGzZswN/fX9e7I4QQoojs2bOHpKQkIiIiSEhIwMPDg7feegszMzOCg4OJjIzk66+/xtXVlalTp7Jq1Srq1KnD7t27WbJkCYGBgbreBSGEEFmUyo6txo0bY2FhwW+//aY1VXvGbYkbNmwgLCwMAwMDbty4QXx8vE46tlQqVbG/Z0kjOVCO5FY5kltlFOVsLg4ODlSqVAkAGxsbYmJiWLlyJYmJiaSkpGBubp5tnYSEBG7evMngwYMBePr0KR06dCiymF5GWfrOlaV9LW6SW+VIboueUjN8qVQq2rVrB0C1atWwtLQkISGBFi1aAGBnZ8eDBw8AuHz5MuPHjwcgLS0NOzs7RWISytKH41MfYtQlyU/+ykKO8moXSmXHVsasWI6OjowfP55WrVoBkJKSQmJiIhs2bODkyZMYGBjg6emJWq3WSZxlfUpOmZZUOZJb5Uhu9UPW+oorVqygd+/eDBw4kL1797J8+XIATE1NSUtLA55f4NSpU4eQkBDNDFopKSnFG3guysp3To4v5UhulSO51S8ODg7s2rWLkSNHkpCQwP3797G0tNRqNzKuDRwdHVm9ejW2trZAyWkTxIsp6cennEPyJvnJn+SolHZsZViwYAEBAQEkJT0vGOjh4cGkSZNwcnLC09OTevXqaX7Rz48UehdCCP01fvx4AgIC2Llzp1bD36VLF959910OHDjAl19+yaJFixgyZAjp6ekYGBgwZcoU3N3ds29QirkLIYRe6tGjBwcPHsTLy4tnz56xcOFCzp49m+OyQUFBTJkyhadPnwIwaNAgfH19c1xWCpTnTC64hRDFwSAxMVE3w5VEmScNnXIkt8qR3AqhHDm+lCO5VY7kVojcyfGRP8lR3iQ/+ZMcgaGuAxBCCCGEEEIIIYQQojBKXcdWYGAgwcHBug5DCCFECRMdHU1QUFCur0v7IYQQQgghhP4p1TW2itL6M+uLfJtSt0sIIYqPk5MTTk5ORbOxZcuKZju5kRpeQgihV5w2VNF1CCWUMxwv/neVmmdClC16PWJLrVYzY8YMvLy88PDw4MCBA5rX/vvvP0aNGsXmzZtJTEykd+/eeHt707lzZ3755RfgefHIuLg4AG7evEmPHj10sh9CCCGK3tWrV+natSs+Pj50796dXbt24e/vD8ClS5fo2bMn3t7e9O7dm/T0dM16mdsPIYQQpU/W9uHMmTP07NmTHj164OnpyZUrV0hPT8fV1ZUnT54AcOLECcaMGaPjyIUQQuREr0ds7dmzh6SkJCIiIkhISMDDw4O33nqLhIQE/Pz8mDBhAm5ubqSmpvLdd99hZmbG5cuXmTZtGuHh4QwbNoxNmzYxa9YsNm3axJAhQ4o1fpVKVazvVxJJDpQjuVWO5FYZRV30cv/+/fTs2ZOJEycCcPToUc1rkyZNYvHixTRv3pz09HQMDZ//zpO1/dCVsvwdK8v7rjTJrXIkt0VPyULIWduH5ORkfvzxR4yMjIiIiCAoKIjVq1fTp08fduzYweDBg9mwYQPDhg1TLCZRtPTtmNS3eIub5Cd/ZSFHebULet2xpVKpaNeuHQDVqlXD0tKShIQEQkJC6Nu3r+ai5OHDh8ycOZNbt25haGhIfHw8AH369MHDw4Pp06cTFhbGwYMHizX+sj5zgczeoBzJrXIkt/pj8ODBBAUFMWbMGGrVqkX79u01r929e5fmzZsDaDq1AL755hut9kNXyup3TI4v5UhulSO51T9Z24cBAwYwb948EhMTSUlJwdzcHIChQ4cyfPhwfHx8uHz5Mq6urjqOXBSUPh2Tcg7Jm+Qnf5IjPe/YcnBwYNeuXYwcOZKEhATu37+PpaUlM2fO5NSpUyxdupRp06axbds2GjVqxNq1a7l06RIDBgwAwMzMjM6dOzN9+nQ6dOhAuXLlcn0vqYclhBD6xcTEhDlz5gAwYcIEkpOTNa9ZWVkRHR2Nk5OT1oitrO1HrqQGlhBC6K2s7cOoUaOYOHEiAwcOZO/evSxfvhx43lZYWVkxe/ZszfVDbqSmU87kglsIURz0umOrR48eHDx4EC8vL549e8bChQs5e/YsRkZGLF++nFmzZvHJJ58wYMAARo0aRVRUFM7OzlrbGDlyJC1bttTU3RJCCFE6fP/992zduhVDQ0PMzMxQq9Wa17744gsCAgJQq9WYmJiwc+dOgGztx8cff6yr8IUQQigka/uwcOFCAgIC2LlzZ7ZOmBEjRtC/f38uX76so2iFEELkxyAxMVGd/2KlV2xsLOPGjSMsLEzXoZQ58guOciS3ypHcCqEcOb6UI7lVjuS2dDt+/DibNm1i9erVug5FL8nxkT/JUd4kP/mTHOn5iK2X9cMPP/DVV1/x6aef6joUIYQQQgghRAny9ddfs337dtasWaPrUIQQQuTBMP9FSpfAwECCg4MB6NevH4cOHaJdu3ZazwshhNBvsbGxeHt76zoMIYQQOpJTOxAdHU1QUBAAoaGhxMbG5rmN3bt3s2HDBurXr69YnEIIIV5emR6x9SLWn1lfJNuRIvRCCFEKLFum3LalML0QQijCyckJJycnAMLCwqhSpQr29vZFs+0NVYpkO6WPMxwv3neUQv5ClD2lumNLrVYTEBBATEwMaWlpvPfee1qvr1u3jnXr1lG7dm3MzMyws7PTUaRCCCGU8NdffzF58mTS0tKoVKkSK1eupFq1ajRt2pQhQ4Zw7Ngx0tPT+f777zE3N2f+/Pn8/PPP2NnZ8fDhQ6ZOnUqHDh10vRtCCCEK4dGjR4wdOxaVSkWHDh1wd3cnJCSECRMmcPDgQWJiYqhRowY7d+5k165dLF++HHNzc5o2bUpgYCAA69ev57fffuP+/fts27aNmjVr6nivhBBCZFWqO7b27NlDUlISERERJCQk4OHhwVtvvQXAvXv3+Pbbbzl8+DDGxsb4+PgUS0wqlapY3kdfSD6UI7lVjuRWGUoUvfzwww+ZNm0abm5ubNy4kaVLl7JgwQIA3N3dmTlzJv7+/hw6dAgbGxtOnjzJgQMHSE1NxdXVtcjjKYiy/v0q6/uvJMmtciS3Ra8o2oTY2Fh27dpF5cqV6dWrl6ZT6tVXX8Xd3R1fX186dOjAgwcPmD9/PgcOHKBKlSqkpaVpttGsWTPmzJlDYGAgO3fu5J133nnpuISy9PF41MeYi5PkJ39lIUd5tQulumNLpVLRrl07AKpVq4alpSUJCQnY29tz/fp1mjRpgpmZGQBt2rQplpjK+mwFmcnsDcqR3CpHcqtfMrcDbdu2JTQ0VPNay5YtAbCzs+PBgwckJyfTqlUrzfTvGberFLey/P2S40s5klvlSG5LLgcHBywsLABwdnbG2tqamJiYbMv99ddfvPbaa1Sp8vx2QiMjI81rLVq0AJ63FTdv3lQ+aPHS9O14lHNI3iQ/+ZMclfKOLQcHB3bt2sXIkSNJSEjg/v37WFpaAlCnTh2uXLnC06dPMTEx4fTp03l+GaQ2lhBC6B8HBwdOnDiBm5sbUVFRNGzYUPOagYGB5v9qtRoHBwfWr19Peno6z549Izo6OvcNSx0sIYQo8VQqFUlJSVSuXJmzZ89q3UZoamqqGZlVr149Ll26xL///kulSpVIS0vTdG5lbStyI3WdciYX3EKI4lCqO7Z69OjBwYMH8fLy4tmzZyxcuJCzZ88CUL16dUaNGoWHhwe1atWievXqOo5WCCFEUfvkk0+YPHkyixYtokKFCqxatSrXZZs2bUqbNm1wd3fH1taWV155BVNT02KMVgghRFGyt7cnICCAP/74g44dO9K0aVPNiK3u3buzcOFCtmzZwpo1a5g1axZ9+vTB3NwcJycnTY0tIYQQJZ9BYmJi7j89CKEg+QVHOZJb5UhuS7fU1FRMTEz477//6NixI2FhYdSoUUPXYZUZcnwpR3KrHMmtELmT4yN/kqO8SX7yJzkCQ10HoKR3332Xbt26sW3bthxfb9q0KQDBwcHyq4wQQuip2NhYvL29tZ6LjIzE39//hbfl5OSEu7s7np6ejBo1Sjq1hBBCCCGEKOFK9a2I4eHheddIEUIIITKpV68eK1euxN7eXtehCCGEEEIIIQqg1HZsTZkyhTt37uDt7c2FCxeIi4sDwN/fXzO174tYf2Z9kcQlReiFEKLoPXr0iLFjx6JSqejQoQPu7u6a11atWsWePXt48uQJDRs2ZOXKlRgYGLBr1y6WL1+Oubk5TZs21Rq5e+PGDSZMmMAnn3yimRFLy7JlRRe8FKIXQohic/XqVSZMmKApHr9lyxbmz5+PSqXi6dOnzJgxAy8vL27dusW0adN4/PgxAF988QX16tXLdbtOG6oU1y7oGWc4XjRbkgL9QojclNqOrWXLlnHw4EHCwsI0txwKIYQonWJjY9m1axeVK1emV69eWjNfDRkyRHNb4vDhwzl69ChOTk7Mnz+fAwcOUKVKFc3MWADnzp1j7dq1fPXVVzJySwghSpn9+/fTs2dPJk6cCMD69euxtrYmKCiIx48f4+npiaenJx999BETJ07E1dWVCxcu8OGHHxIcHKzj6IUQQuSk1HZs5SavaXqLg0ql0un7lzSSD+VIbpUjuVXGyxS9dHBwwMLCAgBnZ2esra01M1+FhYWxadMm1Go1N2/epGvXrlSsWJHXXnuNKlWe/8KeMa07wOzZs1myZEmxdWrJ90mb5EM5klvlSG6LnlKFkAcPHkxQUBBjxoyhVq1a3Lt3j/Pnz3P06FHg+SQi9+7d47fffuPTTz/VrGdgYKBIPKLgSvNxVpr3rShIfvJXFnKUV7tQJjq21Go1ycnJmJiYEB0djZ+fn85iKeuzFWQmszcoR3KrHMltyaRSqUhKSqJy5cqcPXtWa8TWJ598wpkzZyhfvjxDhw5FrVZTr149Ll26xL///kulSpVIS0vTdG5t3ryZDz74ADMzMzp37qx47PJ9+j9yfClHcqscya1+MTExYc6cOQBMmDCBevXq4eDgwOTJkwFISUnB1NQUR0dHxo8fT6tWrTTPC90qrceZnEPyJvnJn+SojHRs+fv74+HhgaOjI9bW1oXahtTGEkKIksve3p6AgAD++OMPOnbsSNOmTTUjtt544w26du2qVRulatWqzJo1iz59+mBubo6Tk5OmxpalpSXbtm1j2LBhPH78ONuMi4DUxRJCCD31/fffs3XrVgwNDTEzM2Pu3LkEBgbi4+ODgYEBVlZWrFmzhgULFhAQEEBS0vO6Th4eHprOr5xI/aecyQW3EKI4GCQmJur23jxRZklDpxzJrXIkt0IoR44v5UhulSO5FSJ3cnzkT3KUN8lP/iRHYKjrAIQQQgghhBBCCCGEKAzp2Pr/AgMDZaYTIYQQQgghhBBCCD1SJmpsFYX1Z9a/9DakTpcQQpQcarVaa5arrI/ztGxZ0QQhtbqEEKJUcNpQRdchlFDOcPzltyI1zIQQedGLjq2rV68yYcIETE1NSUtL49tvv+Wjjz7i77//Rq1WExgYSIsWLfD29mblypXY29sTHBzMjRs3eP/99/H29qZ169b89ttv3L9/n23btlGzZk3WrVvHunXrqF27NmZmZtjZ2el6V4UQQhRCQdsJf39/zM3NuXXrFr6+vsyePZsBAwZw5swZ2rZti4mJCdOnTwegb9++LF68mPr16+t474QQQhTU/PnzOXLkCGZmZvj6+nLr1i327t2LWq3m7bffxtfXl+DgYEJDQzE0NOTq1at8/PHH/O9//+OPP/5g4MCBTJw4kYcPHzJ16lTu3r1LamoqCxYs0MyQKIQQomTRi46t/fv307NnTyZOnAjAqlWrqFu3LuvWrUOlUvHOO+/w888/57mNZs2aMWfOHAIDA9m5cyf9+vXj22+/5fDhwxgbG+Pj46P4fqhUKsXfQ99ITpQjuVWO5FYZL1P08kXaiVdeeYWgoCAAZs+eTZcuXZg9ezZPnjzBy8uLadOmcf36ddLT0xXv1JLvUnaSE+VIbpUjuS16hWkT9u3bx++//05ERASGhobExMSwZcsWIiIiSE1NpXPnznh5eQGQlpbG1q1b+eWXXxg+fDjnzp3DyMgIZ2dnJk6cyLJly/Dw8GDQoEHcvn2bwYMHc+DAgaLeTVFApf0YK+3797IkP/krCznKq13Qi46twYMHExQUxJgxY6hVqxZ3797ljTfeAJ7v3L179wCy3VKSWYsWLQCws7Pj5s2bXL9+nSZNmmBmZgZAmzZtFN+Psj5TQVYye4NyJLfKkdyWTAVtJyD7+T7jcfny5Wnbti0///wzkZGRDBs2TPG45bukTY4v5UhulSO5LTkuX75M586dMTR8XkZYpVLx+uuvY2RkhJGREU2aNOH69esANG/eHAAbGxsaNGhAhQoVgP+7nvjtt984evQoW7ZsASApSW6F06XSfIzJOSRvkp/8SY70pHi8iYkJc+bM4ZtvvuHevXvY2dlx4sQJ4PmHWK1aNQCqVq1KfHw8AOfOndPaRtZOrzp16nDlyhWePn1Keno6p0+fLqa9EUIIUdQK2k4AmgueDEZGRpr/jxw5km+++YaIiIhiGckrhBCi6Dg6OnLkyBHND9z16tXj1KlTpKen899//3H58mXq1KkDaF8b5FRf0dHRkfHjxxMWFkZYWBjHjxdBoSghhBCK0IsRW99//z1bt27F0NAQMzMzPvjgA95//326d+9OWloaixcvBmDs2LFMnz6dunXrYmlpmec2q1evzqhRo/Dw8KBWrVpUr149z+Wl8LsQQpRcBW0n8tOkSRMePnyIp6cnpqamuS8oRd+FEKLE8fT0JCoqCg8PD8qXL8+gQYPo1KkTnp6eqNVqJkyYoPVDR16mT5/O9OnTWbduHWq1GicnJz799NNcl5fi5jmTkSRCiOJgkJiYqM5/MSGKnjR0ypHcKkdyW/r17t2boKAg6tWrp+tQyhw5vpQjuVWO5FaI3MnxkT/JUd4kP/mTHOnJrYhCCCGE0u7cucMbb7zBq6++Kp1aQgghhBBC6Am97diKjY3F29u7QMsGBgYSHByscERCCCH0mZWVFTt27NDcatK0aVMdRySEEEIIIYTIj17U2CoJ1p9Z/1LrS40uIYQoRZYtK5rtSK0uIYTQG2q1OsdC8wBOG6oUczT6whleou6+1C4TQhSEXndsPX78mEmTJhETE4OLiwvDhw9n2rRppKWlkZqayvLly2ncuDEAJ06cYM+ePfz5558sWbIEV1dXVq1axZ49e3jy5AkNGzZk5cqVuTZWQggh9ENsbCyDBw+mSZMmXLx4EV9fX+Li4jh79ixWVlYsX76cYcOG8ezZMx4/fsynn36Ki4sLd+/eZfTo0QC8/vrrOt4LIYQQLyI2NpZhw4bh5OSkuTb45JNPCAgIQKVS8fTpU2bMmIGHhwft27fn0KFDmJubc+rUKb755hvWrFnDunXr+O677wDo0qULM2fOJDIykqVLl1K5cmUsLS0JCgrS8Z4KIYTISq87tuLj49m9ezcVKlSgZcuWzJw5kx9//BEjIyMiIiIICgpi9erVAJiZmREcHExkZCRff/01rq6uDBkyBH9/fwCGDx/O0aNH6dSpkyKxqlQqRbar7yQvypHcKkdyq4yiLHp5+/ZtwsPDMTY2pkGDBuzevZvAwEB69epFfHw83333HWZmZly+fJlp06YRHh5OUFAQb775JkOGDOHw4cNs27atyOLJjXyXspOcKEdyqxzJbdErTJuQ9dqgdu3aWFtbExQUxOPHj/H09MTT05Pu3bsTGhrKm2++SXBwMH5+fly9epXQ0FD27NmDoaEhvr6+xMTEaLa7ffv2vGfLFYooK8dWWdnPwpL85K8s5CivdkGvO7YcHByoVKkSADY2NsTExLBy5UoSExNJSUnB3Nxcs2yLFi0AsLOz48GDBwCEhYWxadMm1Go1N2/epGvXrorGKrTJ7A3KkdwqR3KrHxwcHKhYsSIAVatWpXnz5sDztuLBgweMGzeOW7duYWhoSHx8PPD8sx0x4vlt423bti22OMX/keNLOZJb5UhuS46s1wbR0dFER0dz9OhRAFJTU7l37x5+fn4EBATg7e3N6dOnCQoK4qeffkKlUtGzZ08AkpKSuHHjBpUrV6Z58+bSqaUjZeHYknNI3iQ/+ZMc6XnHVtbbBlesWEHv3r0ZOHAge/fuZfny5Tkuq1arAfjkk084c+YM5cuXZ+jQoZrncyI1soQQQn9kbR8yP75w4QKNGjVi7dq1XLp0iQEDBgDP/3g+ffo0DRs25OTJk3m/gdTGEkKIEifrud/Z2ZlGjRoxefJkAFJSUjA1NaVmzZo8ffqUlStX4u3tjaGhIY0bN8bR0ZFt27ZhYGBAeno66enpnDhxAiMjo3zfW2pB5UwuuIUQxUGvO7ayGj9+PAEBAezcubNAJ9A33niDrl27yrTuQghRhpiZmbF161aioqJwdnbWPD916lRGjx7N9u3bad26tQ4jFEIIURTc3NxYuXIlPj4+GBgYYGVlxZo1awAYNGgQU6dO5fTp0wA0btwYb29vevTogZGRESYmJqxYsUKX4QshhCggg8TExNyHKQmhIPkFRzmSW+VIboVQjhxfypHcKkdyK0Tu5PjIn+Qob5Kf/EmOwFDXAQghhBBCCCGEEEIIURjSsSWEEEIvvPvuu3Tr1u2FZits2rRpod7L39+fyMjIQq0rhBCi5EhMTCQ4OBiAAwcOaP4vhBCi9ChVNbaUtP7M+pdaX4rPCyHEywkPDyc6OlrXYTy3bNnLb0MK0AshhOKSkpIICQnBz88PDw8Pxd7HaUMVxbat35zh+IuvJcX4hRAvQq9GbMXGxtKhQwfGjBmDq6srK1as4P3336dr164MGTKExMREevfujbe3N507d+aXX34BIDg4mBEjRjBs2DDatGnDzp07SU9Px9XVlSdPngBw4sQJxowZo8vdE0IIkYspU6Zw584dvL29sbW11TyfeWTVmjVrcHd3x8fHh2VZOp5iYmLo2bMn165d48SJE3h7e+Pj48OIESNITk4GYP78+bi5uTFs2DDu3LlTbPsmhBBCOcuXLycmJgZvb2+WLFlCYGAgAN7e3gQEBNC3b1+8vb3ZuXMnb7zxBi4uLly5cgUg1/ZCCCFEyaJ3I7Zu375NeHg4xsbGNGjQgN27dxMYGEivXr2Ij4/nu+++w8zMjMuXLzNt2jTCw8MBePz4Mdu3b+fGjRsMHTqUvn370qdPH3bs2MHgwYPZsGEDw4YNUyxulUql2Lb1meRFOZJb5UhulZFX0ctly5Zx8OBBwsLCcry98PLly2zfvp09e/ZgZmZGWlqa5rX9+/ezevVqNm7cSNWqVRk8eDChoaFUrVqVFStWsGHDBtq3b09UVBQHDx4kJSWFdu3aKbKPmcn3KGeSF+VIbpUjuS16RVUIeeLEiVy+fJmwsDCCg4O5ceOG5jUXFxcWLVrEhAkT+OWXX9ixYwfbtm1j8+bNzJ8/nxkzZmRrL/z9/YskLpG3snZMlbX9fVGSn/yVhRzl1S7oXceWg4MDFStWBKBq1ao0b94cABsbGx48eMC4ceO4desWhoaGxMfHa9Zr3rw5BgYG2Nvb8+DBAwCGDh3K8OHD8fHx4fLly7i6uioat9AmszcoR3KrHMltyaJWP5/Y98qVK7i4uGBmZgaAkZGRZpkZM2YQHByMpaUl9+7d4+bNmwwePBiAp0+f0qFDB1QqFc7OzhgaGmJubk6LFi0Uj12+R9nJ8aUcya1yJLf6K+Ncb2NjQ+3atTX/P3r0KAkJCTm2F6J4lKVjSs4heZP85E9ypIcdWwYGBrk+vnDhAo0aNWLt2rVcunSJAQMG5LhcxoWQlZUVVlZWzJ49W2vZnEiNLCGEKBnUajXJycmYmJgQHR2Nn58fTZo0YdWqVaSkpGBqakpaWpqmc2vXrl2MHj2apUuX8tprr1GnTh1CQkKoUuV5PZSUlBR+//131q9fT3p6OqmpqZw/fz7vIKQ+lhBC6IWMNiEnma8Psl4rVKtWLcf2IjdSEypncsEthCgOetexlRczMzO2bt1KVFQUzs7OBVpnxIgR9O/fn8uXLyscnRBCiKLg7++Ph4cHjo6OWFtbA9C4cWPeeustunXrRvny5fHw8GDK/+98qlOnDps2bWLEiBHMnTuXRYsWMWTIENLT0zEwMGDKlCm4u7vTpk0b3N3dsbW1xd7eXod7KIQQoqhYWVlRuXJlhgwZ8kK3mRsYGOTaXgghhChZDBITE9W6DkKXjh8/zqZNm1i9erWuQylz5Bcc5UhulSO5FUI5cnwpR3KrHMmtELmT4yN/kqO8SX7yJznSs1kRcxMbG4u3t7fWc9HR0QQFBeW5XpcuXZg8eTIzZ85UMjwhhBDFzNvbm9jY2EKvHxkZKQWChRCiDAkODtbMmCiEEEK/lKpbETNzcnLCyckpz2UaNWqEr68v9evXL6aohBBCCCGEEEIIIURRKTUdW48fP2bSpEnExMTg4uJCt27dCAkJYdWqVZw6dYqZM2diZWWFjY0NycnJrFq1CoCffvqJr776iri4ONatW0ejRo1y3P76M+tfKj4pPi+EEMpJT09n8uTJXL16lbp16/Lo0SMAFixYwLFjx3j27BlDhgxh6NChHD9+nMDAQE3NlHXr1mFlZaVpK2rWrImtrW3eb7hs2csFLMXnhRBCEbGxsQwePJgmTZpw8eJFfH19iYuL4+zZs1hZWbF8+XKGDRvGs2fPePz4MZ9++ikuLi6a9dPT05k1axZVq1Zl+vTpBAQEoFKpePr0KTNmzMDLyyvH93XaUKW4dlHPOMPxvJeQwvtCiJdVajq24uPj2b17NxUqVKBly5a4urpqXps5cybr16+nbt26BAYGcuPGDc1rtra2LFq0iODgYDZt2sSCBQt0Eb4QQoiXEB4eTkpKChERESQkJNCiRQsOHTrErVu3CA8PJy0tjW7dutGjRw+aN29OaGgoAN9++y1r1qzhgw8+YObMmaxdu5b69eszf/584uPjdbxXQgghCuP27duEh4djbGxMgwYN2L17N4GBgfTq1Yv4+Hi+++47zMzMuHz5MtOmTSM8PByAJ0+eMGLECLp3787AgQPZsGED1tbWBAUF8fjxYzw9PfH09MTQsFRUcxFCiFKj1HRsOTg4UKlSJQBsbGxITEzUvHb//n3q1q0LQOvWrbU6tlq0aAGAnZ0dx4/n83PCS1CpVIptW59JXpQjuVWO5FYZL1P0UqVS0aZNGwCqVatGw4YNefjwIadOndLUYHz06BFxcXGkp6ezYMECkpOTefjwoea29fv372tuTW/bti0//PDDS+5R3vGKnElulCO5VY7ktui9TJvg4OBAxYoVAahatSrNmzcHnl8jPHjwgHHjxnHr1i0MDQ21fsTYuXMnzs7ODBw4EIDffvuNqKgojh49CkBqair37t2jZs2ahY5NZCfHj+QgP5Kf/JWFHOXVLpSaji0DAwOtx2r1/032aGlpybVr16hbty6nT5/Odb3M6xS1sj5LQU5k9gblSG6VI7ktmRwcHAgNDWXkyJHcv3+fq1evUrFiRTp27MiSJUuA5xckxsbG+Pr6Mn36dFxcXPjmm284d+4coN1WREVFKR6vyE6OL+VIbpUjuS15sl4XZH584cIFGjVqxNq1a7l06RIDBgzQvObn54darWb69OksXrwYR0dHateuzeTJkwFISUnB1NS0eHaiDCnrx4+cQ/Im+cmf5KgUdWzlZeHChQwfPpwaNWpgbW1dqAZJamQJIUTJ1b17d/bu3UvXrl2pU6cOdevWxd3dnYSEBLp3746RkRHm5uYEBwfz1ltvMWXKFOrXr0+tWrU02/j8888ZMWIE1atXp3bt2nm/odTIEkIIvWRmZsbWrVuJiorC2dk52+uzZs1ixYoVjB07li+//JLZs2fj4+ODgYEBVlZWrFmzJsftSp2onMkFtxCiOBgkJiYqN0yphEhNTcXExASA+fPnY2FhwYQJE3QclZCGTjmSW+VIboVQjhxfypHcKkdyK0Tu5PjIn+Qob5Kf/EmOysiIrcOHD7N06VLS0tKoVq0a33zzja5DEkIIIYQQQgghhBAvqdRO6ZGYmEhwcDAAnp6ehIeHs2/fPrZu3aopMi+EEKJ0i42N1RSPz09gYKCm3fj+++/p3LkzH3zwgZLhCSGEKAHyayuCg4O1JqYSQghRspTaEVtJSUmEhITg5+dXJNtbf2Z9odeV+lxCCKFftm7dypo1a2jQoEHOCyxbVrgNS20uIYTQOyEhIbRv3x4LC4tcl3HaUKX4AtIrzpBl4nmpRyaEKGqltmNr+fLlxMTE4O3tjZ+fHz/++COPHj3C2NiYL7/8kjp16uDt7U3z5s25dOkSKSkprF27Fmtra12HLoQQogg9fvyYSZMmERMTg4uLC8OHD2fatGmkpaWRmprK8uXLady4sWb5tWvXcvr0acaOHcvQoUMZOnSoDqMXQghRXG7dusW0adN4/PgxAF988QWxsbHExMTw9ttvU79+fVavXq3jKIUQQmRVaovHx8bGMm7cOMLCwnj//fdp3Lgxw4YN49ChQ6xbt47NmzdrOr18fX3ZvHkzFy9e5LPPPstxey8zYqt9lfaFXlcIIcoKJYpexsbG4uHhwa+//kqFChVo2bIlR44coWLFihgZGREREcGOHTtYvXo1gYGB2NnZ4efnh7e3NytXrsTe3j7nDRdyxJaqgLdFCiFEWVechZAzrhteeeUVRowYgaurKxcuXODzzz8nODg4/zYBGbH1In5wPaPrEIQQeiivdqHUjtjKTKVSMWLE89sB27Zty3vvvad5rXXr1gC0adOGXbt2KfL+ZX2GgtzI7A3KkdwqR3KrfxwcHDS1FW1sbIiJiWHlypUkJiaSkpKCubl5scYicifHl3Ikt8qR3JYev/32G59++qnmsYGBgQ6jKb3keNEm55C8SX7yJzkqxR1bpqampKWlAc9PnidOnKBhw4ZERUXRsGFDzXJnzpzBwcGB06dP515LRQghhN7KemGyYsUKevfuzcCBA9m7dy/Lly/XUWRCCCFKEkdHR8aPH0+rVq0ASElJAbSvK4QQQpQ8pbZjy8rKisqVKzNkyBDeeOMNgoOD2b59O4aGhnz55Zea5aKjo9m+fTupqamsXbs21+1JAXghhCgdxo8fT0BAADt37iz8r1tSBF4IIUqdBQsWEBAQQFLS8+LmHh4eTJ48mT59+jB58mQcHR1ZuHBhjutKQfScyUgSIURxKLU1tgqiIPfLC+VIQ6ccya1yJLdCKEeOL+VIbpUjuRUid3J85E9ylDfJT/4kRy8wYsvJyemF7zM3MDDg/PnzLxqTEEIIIYQQQgghhBD5KnDHlqura7aOrfPnz3P58mUaN25MgwYNUKvV/Pnnn1y5coUmTZrQvHnzoo63QDJmNvH19cXKygoPD48clwsLCyvmyIQQQpQkwcHB3Lhxg/fff79AzwshhBBCCCFKlgJ3bK1atUrr8d69ewkLC2Pnzp107txZ67Wff/6ZESNG8PHHHxdJkIXl5+dXZNtaf2Z9odaT2lxCCFEKLVtWuPWkNpcQQpRKThuq6DqEEsoZjv/fI6lFJoRQQqGLxy9YsIDRo0dn69QC6NKlC6NGjWLevHl4eXm9THwvJTAwEDs7O6ysrAgLCyMoKAiAKVOm0KtXL5ydnZk6dSp3794lNTWVBQsWaGZBEUIIod/mz5/PkSNHMDMzw9fXl1u3brF3717UajVvv/02vr6+WsuHhYURGBiIjY0N1apVo3bt2jqKXAghRFGKjY1l8ODBNGnShIsXL+Lr60tcXBxnz57FysqK5cuXM2zYMJ49e8bjx4/59NNPcXFxITg4mAMHDpCens6VK1d477336Nu3r653RwghRBaF7thSqVQMHjw419dr1KjBH3/8UdjNF6kuXbrw0Uf/r707j4/p3v84/prJxrXFUlw0RMUSRRW1L2kSJBNUb1t7rFX72li63WqptkrSamlRUsTW9qdKhFpqK1qUiq2mLYml1JJEo0hk5veHh7lNZZHImEy8n4/HfVzOOXPO97ydpfOd7/mc17h27RoAP/zwAzNmzGDy5MkEBATQrVs3zp07R8+ePdm4cWOebttsNufp+goa5WM/ytZ+lK195GXRy2+++Yaff/6Z9evXYzQaiY2NZfHixaxfv57U1FTatGmT7ocXi8XCK6+8wubNmylZsiQDBw7Ms7b8nY6d7Ckj+1G29qNs815eF0I+d+4cMTExuLq6Uq1aNVavXs3UqVPp2LEjZ86cYcWKFXh4eHD06FHGjBlDTEwMAFevXmX58uXEx8cTGhqqjq17pHMlY8ola8onew9CRlndF3LdsVWpUiW++OIL+vbti7u7e7p5KSkpfP7551SqVCm3q89TRqOR4OBg1qxZg9VqJTg4GKPRyKFDh9i2bRuLFy8GsL3aNy896G8nyIre3mA/ytZ+lK1zOHr0KG3atMFoNAK3/t2eeOIJXFxccHFxoVatWpw8edK2/MWLFyldujQlS5YEoEmTJpw/fz7P26VjJ2s6v+xH2dqPsnUOPj4+FC1aFICSJUvaagFXrFiRhIQEhgwZwtmzZzEajZw5c8b2ucceewyDwUDlypVJSEhwRNMLFJ0rd9I1JGvKJ3vK6B46tkaNGsXw4cNp06YN/fr1o1q1ahgMBo4fP86CBQs4duwYH3zwQV629Z706NGDMWPGYLVabY8k+vr6Uq9ePdsvLykpKZl+XrWyRESch6+vL4sWLaJfv34YDAaqVq3KvHnzsFgspKSkcPToUapUqcKxY8cAKFOmDJcuXSIxMRFPT092796Nt7d35htQrSwREafyz5dg/f3vP/30EzVq1ODTTz/lyJEjdOnSJcPlrFZrputX7aiM6Qu3iNwPue7Y6tmzJ0ajkUmTJhEWFma76FutVh566CE++OCDLB9VvN+8vb25efOm7c8AY8eOZezYscyfPx+r1UrdunV56623HNlMERHJA4GBgezevZuAgAD+9a9/0a1bN1q3bk1gYCBWq5Vhw4ZRunRp2/JGo5HJkyfToUMHypcvn26eiIgUbB4eHixdupTdu3fTsGFDRzdHRERyyJCYmJj5Tw93IS0tjR9//JFTp05htVrx8vKifv36uLrmus9MHhD6Bcd+lK39KFsR+9H5ZT/K1n6UrUjmdH5kTxllTflkTxnlcsTWtWvXaNq0KYMGDWLQoEE0atSIRo0a5XXbREREREREREREMmXMzYcKFy7MlStX7iga7yhxcXGYTKZcf37NmjXExcXlYYtERMQR7vV+cFtUVBRTp07NgxaJiMj9lpiYSFRUFJD59TwqKirbt6HXqVPHLu0TEZG8levnBdu2bcs333xDv3798rI9DhEdHU2JEiWoXLlypsss2Lsgx+tVwXkREedgtVrvKCycpYiInG1AxeZFRO6bpKQklixZQo8ePTJdJqt5uVE3skSerq/gaAjfqbi+iNhXrju2Ro8eTd++fenTpw99+/bF29ubwoUL37HcQw89dE8NvFvJyckMGjQIs9lMy5YteeWVVwgLC8NsNnPjxg1efPFF2rVrxyeffMLy5cspXLgw9erVo0ePHmzatInY2FgeeughVq5ceV/aKyIi9nH16lVGjBhBbGwszZo1o0+fPowZM4a0tDRSU1OZOXMmNWvWZOrUqcTHx5OQkIC/vz8VKlRg6tSpVKxYkdKlS/Pwww87eldERCQXZs6cSWxsLCaTiSeffJLjx4/Tu3dvjh07xoQJE+jcuTNTp07Fy8uLHj16UKdOHXr16sWOHTuwWCx88cUXFCpUyLa+rVu38sEHH/DJJ59QpkwZB+6ZiIhkJNcdW02aNAHg6NGjfP3115kud/ny5dxuIkfi4uJYtWoVxYsXp2PHjnz00UeUL1+e8PBwrl69SmBgIIGBgSxevJiVK1dSpkwZLBYLRqMRf39/unfvTsuWLfO0TWazOU/XVxApI/tRtvajbO0jr4penjlzhtWrV1OkSBEef/xxxo0bx1dffYWLiwvr168nPDycTz75BAAXFxeWLVuGxWKhQYMGbN68mZIlSzJw4MA8acttOmbunrKyH2VrP8o2793LPWH48OEcPXqU6OhooqKi+P7771m+fDnx8fGEhobSuXPnOz7j7+/PuHHjGDx4MN9++y1BQUEALFq0iG+//ZbFixdn+CO+3B2dI1lTPllTPtl7EDLK6r6Q646tcePG5eyxDTvz8fHB09MTgIYNGxIVFYW7uzvbtm0DIDU1lYsXLxIeHs7rr79OSkoKISEhdOzY0a5tkszp7Q32o2ztR9nmfz4+PhQrVgyAihUrEhsby6xZs0hMTCQlJSXdr/BPPPEEABcvXqR06dKULFkSuPXjzfnz5/O0TZI9nV/2o2ztR9nmf4899hgGg4HKlSuTkJCQ4TKPP/44AF5eXrZlkpKSePvtt9m0aZM6te6RzpHM6RqSNeWTPWV0Dx1bEydOzMt23DOz2UxSUhLFixdn3759dOvWDaPRyMiRIwFISUnB3d2dokWL8uGHH3L9+nXq1atHx44dcXd3Jy0tLcv1q16WiIhz+OePLh999BGdOnWia9eurFu3jpkzZ9rmubi4AFCmTBkuXbpEYmIinp6e7N69G29v78w3oppZIiL51j//2/7v9wWr1ZrhZzJapkSJEixcuJDQ0FA+/fTTLB9RVw2pjOkLt4jcD7l6K2JG/vzzT/7888+8Wl2OVa5cmbCwMPz9/WnQoAEjRozg9OnThISE0KFDB4YMGQLACy+8QHBwMCEhIQwYMACAoKAg3nnnHdvfRUSk4Bg6dCjvv/8+Xbp0YceOHRkuYzQamTx5Mh06dODZZ5/FaMyz26OIiNxn5cqVo3jx4vTq1YukpHvrcKpfvz7vv/8+/fr149dff82jFoqISF4yJCYmZvyzxV2Ij49nypQpbNiwgcTERAA8PT1p27YtL730El5eXnnVTimA9AuO/Shb+1G2Ivaj88t+lK39KFuRzOn8yJ4yypryyZ4yuocRW2azmTZt2vDFF19Qv359Bg0axAsvvMDjjz/O559/jp+fH7/88ktetjVTcXFxmEwmoqKi2LhxY5bL1qlTJ8Pps2bNskfTRETEifTr1w+A7du3M3jwYAe3RkREHGXq1KlERUU5uhkiInIXcl1ja9KkSVitVr799lvq1q2bbl5sbCydOnVi0qRJLFq06J4bebd69OiR68/Onj3b9riiiIg8mObPn+/oJoiIiIiISA7kumNrx44dDBo06I5OLbg1Kur555+3vU79fpk6dSpeXl706NGD+fPns2DBAry8vHBzcyMwMNDW8fXuu++yY8cOLBYLX3zxBXPmzOGPP/7AZDIRFBTEsGHD7lj3gr0LctweFZwXEckfpkyZwo4dO7h58ya9evViyZIlzJ8/nwoVKnD27Fn69+9PTEwMderUITY2NvsVRkTkrAEqNi8ikm9ZrVbCwsKIjY0lLS2NCRMmALBr1y7Wrl3Lr7/+yvTp02nevHmm66gbWeJ+NdepfJl5ZCIieSbXjyKmpKRQvHjxTOeXKFGClJSU3K7+nly8eJF58+axceNGFi1axKVLl9LN9/f35+uvv+bhhx/m22+/ZcSIEZQtW5bo6OgMO7VERMR5bd68mbNnzxITE8O6detYtGgR7du3Z9myZQAsXbqUrl27OriVIiLiKGvXriUpKYn169ezfPlywsLCsFqteHh4EBUVxbRp0/j4448d3UwREclErkds+fr6snz5cvr27UvhwoXTzbtx4wbLly/H19f3nhuYGydPnqRmzZp4eHgA0KBBg3TzH3/8cQC8vLxISEiwWzvMZrPd1l1QKCP7Ubb2o2ztw15FLw8dOsQPP/yAyWQCIDk5mYYNGzJx4kRGjx7NypUrWbt2rV22fZuOmbunrOxH2dqPss1797MQstlspmnTpgCULl2aUqVKcenSJerXrw/Y/ztDQafzI3vKKGvKJ3sPQkZZ3Rdy3bE1ZswYevbsiZ+fH/3797dt5Pjx48yfPx+z2czixYtzu/p7UqVKFY4dO8aNGzdwc3Nj37596UIwGAy2P1ut1jum5ZUH/c0E2dHbG+xH2dqPsnU+vr6+tGrViunTpwOQmpqKq6srtWrVIiIigtq1a2c5Ajkv6Ji5Ozq/7EfZ2o+ydX4+Pj6sWrWKfv36cenSJS5fvkypUqUy/M4gOafzI2u6hmRN+WRPGd1Dx1ZwcDBz5szh5ZdfZty4cbYLv9VqpVy5csyZM4egoKA8a2hOlClThv79+xMQEEDFihUpVqwY7u7uWX6mZcuWdOvWjXbt2tGnT5875qteloiIcwoICGD//v0EBQXh4uJCoUKFiIqKomfPnnTu3Jmvvvoq5ytVzSwRkQIjODiYTZs20a5dO27evMk777zDvn37crSOg32S7NQ65/YADCIRkXzAkJiYeE8/P9y8eZMDBw4QHx8P3Bqq+9hjj+Hqmus+szyRmpqKm5sbaWlpdOzYkSlTpvDYY485tE2SnnqW7UfZ2o+yFbEfnV/2o2ztR9mKZE7nR/aUUdaUT/aU0T2M2KpZsybNmjWjWbNmNGnShKeffjov23XPPvjgAzZv3kxycjL+/v7q1BIRERERERERKWBy3bEVFBTErl27bI9wFC9enCZNmtg6uxw9amvs2LGMHTvW9vfx48fz008/8cMPP3D58mWHtUtERERERO6PxMREoqOj6dGjR6bLDB48mO7du9OyZcv72DIREckrue55Cg8PB+Dy5cvs2LGDnTt3smvXLt544w0sFguFCxemUaNGuatdYgcxMTEcPHiQOnXq5OrzC/YuyNHyqsklIuJYVqvVLi8GASAi4u6XVT0uERGHSUpKYsmSJVl2bOWFupEl7Lp+Z6J6YyJyv93zkKpSpUrRsWNHOnbsyKVLl4iJieGDDz7AbDazbdu2vGjjPRs1ahTnz5/HZDLZXtU7cOBA+vfvT+PGjfnrr78ICAhgx44dGI1GB7dWRESy8t5771GpUiW6du1Ks2bN6N27Ny+88ALBwcHUqlWLv/76i4SEBPz9/SlcuDDz58/HaDQSHBzMmDFj2L59OxEREZQqVYojR47QvXt3hg4dyh9//MHzzz8PQOPGjVm6dCmxsbEO3lsREbkXM2fOJDY2FpPJRKtWrdi+fTsWiwWDwcD8+fMpV66cbdmEhAReeOEFnn/+eWrXrs2YMWO4evUqAO+//z5Vq1Z11G6IiEgW7qlj69SpU+zatcs2WstsNuPh4UH9+vXp2LEjzZo1y6t23pOIiAg2bdpEdHS0bcRW3759WbBgAY0bN+bLL7+kU6dOedqpZdYrQO6KcrIfZWs/ytY+7rboZZs2bZg3bx5+fn54eXmxY8cOevbsybVr1yhTpgxnzpxh2bJlXLx4EZPJxNatW3Fzc6NDhw60bdsWgN9//53ly5dz48YNGjduzNChQ5kxYwbPPfccPXr0YPv27SxdujRP9kvHS84oL/tRtvajbPNeXhVCHj58OEePHiU6OpqrV68yfvx4AObOncu8efN4+eWXAThx4gTvvPMOU6ZMoV69egwYMIDhw4fTvHlzfvrpJ1599VWioqLypE0F3T/PB50f2VNGWVM+2XsQMsrqvpDrjq06depw5swZihYtSuPGjenSpQtNmzalQYMGuLu753a1903Tpk155ZVXSExMJCoqivnz5+fp+h/0txLcDb29wX6Urf0oW8erX78+Bw8eZMuWLYSEhLBq1Sq2bt1q+zHliSeeAODkyZPUrl2bQoUKAdCoUSPMZjNlypShbt26uLq64urqantc8ZdffqFfv362ZfOKjpe7p/PLfpSt/Shb5/Hzzz8zZcoUrl27xpUrV6hbt65t3rvvvsvIkSOpV68eAIcOHeKtt96yzbfbo+0F0N/PB50f2VNGWVM+2VNGkOshSqdPn8ZgMNCkSRNatWpF69ateeKJJ5yiU+u2rl278uqrr1KqVCkqVKjg6OaIiMhdcHFxwdvb2zZqq0GDBoSHh9OmTRvbfIAqVapw+PBhrl+/jsViYc+ePbabfkZfUHx8fNizZw+A7f9FRMS5ubu7k5aWBsC0adMYO3Ysa9euJTQ0FKvValsuPDycjRs3snz5cgB8fX154403iI6OJjo6mv/7v/9zSPtFRCR7uR6xtX//fr777jt27drFggULeO211yhSpAiNGjWyvRmxYcOGeHh45GV781SXLl2YNGnSXY3WUjF4EZH8w8/Pj7lz51KxYkX8/PyYPn06zZo1Y+/evbZlypQpw9ChQwkKCsJoNBIUFMSjjz7K9u3bM1zn6NGjGTBgACtWrOCxxx7L+ocaFYQXEXEK5cqVo3jx4vTq1Ytnn32WUaNG8cgjj9zxo3ahQoVYtGgRgwYNIjk5mSlTphAWFkZS0q1C6AEBAYwcOTLT7ahguoiI4xgSExOt2S+WvfPnz7Nz50527tzJ5s2bOXHiBB4eHvz+++95sXq7SE5Opl27dmzfvl1F4x1AQybtR9naj7ItuNLS0jAYDBiNRrZs2cLHH3/MsmXLHN2sB4rOL/tRtvajbEUyp/Mje8ooa8one8ooD96KCJCYmMiPP/5o+19cXBxWqxWLxZIXq7eLrVu3MnnyZMLCwtSpJSIiXLp0idDQUAwGA2lpabz//vuObpKIiIiIiGQj1x1bq1atYseOHezcuZNjx45hsVgoUqQIDRs2ZNy4cTRt2jRPi+/mpcGDB9O9e3c2bNjg6KaIiIgdxcXFMWTIEKKjozOcHxUVhclkwtPTk7Jly7Ju3br73EIREbmfxo8fz08//USfPn3o2rXrXX0mKiqK+Ph4Jk6caOfWiYhIbuS6Y6tPnz54enrSpEkTunbtSrNmzXjsscdsRXvtzWq13te3kyzYu+CullMtLhER57FkyRJatGiBp6dnzj4YEZH9MqrDJSKS78TExHDw4ME8X2/dyBJ5vs78TDXFRCQ/yXXH1o4dO6hdu3ZetiVbcXFxhIaGUqtWLU6fPk3RokW5evUqAO+//z5Vq1bljTfeYM+ePVy5cgV/f39ee+01ACZPnsymTZvw8vLizz//BGDhwoVcuHCBsWPHAtC5c2fee+89Hnnkkfu6XyIiYl8rV67k008/JTU1FU9PTxYsWMD3339PbGws/fv355FHHuGTTz5h/vz5rFixAoAnn3yScePGObjlIiKSV0aNGsX58+cxmUy0b9+etWvXYjAYqFmzJtOnT8dgMDBlyhR27NjBzZs36dWrF6GhoY5utoiIZCPXHVv3u1Prtvj4eFatWsWLL75I3759ad68OT/99BOvvvoqUVFRjBkzhqJFi2KxWGjbti0nTpwgOTmZ3bt3s2nTJlJSUmjatCkAzzzzDO3atWPMmDGcPHkSi8Vyz51aZrM5L3bzgaG87EfZ2o+ytQ97Fr0MDAykc+fOALz66qt8+eWX9OrVizp16jBr1iwqV67M8ePHWbNmDWvXrsVoNNK9e3diY2OpU6dOrrap4yR3lJv9KFv7UbZ5zx73hIiICDZt2kRUVBSdO3dm3bp1eHh4MH78eNauXUvhwoU5e/YsMTExpKWl0b59e4KDg/O8HQVBTo55nR/ZU0ZZUz7ZexAyyuq+kCfF4++nmjVr4unpyaFDh3jrrbds028/lhgZGUl0dDQGg4H4+HjOnDnDH3/8QcOGDTEajRQqVIj69esD8K9//YsmTZqwefNmtm/fTu/eve+5fQ/62whyQm9vsB9laz/K1jnt3r2bmTNnkpqayoULFyhSpMgdyxw5cgSz2UyHDh0ASEpKIj4+PtcdWzpOck7nl/0oW/tRts7nt99+48yZMzz99NMAXL16lWrVqnHt2jV++OEHTCYTcOsN6qdPn3ZkU/Otuz3mdX5kTxllTflkTxk5YcfW7TcY+vr6MnToUBo0aABASkoKiYmJREZG8v3332MwGAgMDMRqteLj48OCBQuwWCykpqZy4MAB2/r69evH66+/Tnx8PC+99FKm21XtLBER5/Xf//6X5cuXU6lSJV555RWsVisA7u7upKWlAbd+OPH19WXZsmUYDAYsFkvmb/dV/SwREafl7e1NlSpVWLlyJe7u7gCkpqaydetWWrVqxfTp023TXF1dOXz4cLbrVM0pERHHcbqOrdumTJlCWFgYSUm3biIBAQGMGDGCunXrEhgYSNWqVSlWrBgAderUoXHjxvj7+1OpUiUqV65sW0+tWrW4cuUKgYGBthubiIgULN27d+eZZ56hWrVqFCtWzDZi66mnnmLkyJH4+vryzjvvYDKZCA4OxsXFBTc3Nz766CMqVKjg4NaLiEheKlmyJKNGjaJz584YDAaMRiOTJ08mICCA/fv3ExQUhIuLC4UKFSIqKsrRzRURkWwYEhMTrY5uhKN16tSJ8PBwqlat6uimPFA0ZNJ+lK39KFsR+9H5ZT/K1n6UrUjmdH5kTxllTflkTxmB0dENcKTz58/z9NNPU7t2bXVqiYiIiIiIiIg4mXzXsRUXF8eaNWty/Ll+/fplOX/79u0MHjw43bRy5crx+uuvExQUlOPtiYhI/pKYmGh7ZGTjxo339PhIbgvGi4iIc/nnd4+oqCg2btwI/O9ekNH3CBERyT/yXY2t+Ph4oqOjCQkJuWOe1Wq1vf3wn+bPn5+r7cXGxhIfH0/Lli2zXG7B3gV3tT4VmRcRcYykpCSWLFlCjx49CAgIyHCZrO4jORIRkf0yKjAvIpLv/fO7R48ePXK1nrqRJfKyWfmSCuSLSH7l8I6t48ePM2zYMNubqYoVK8aBAwcwmUyMGzeOnTt3Eh8fT0JCAv7+/ty8eZO1a9fy119/Ub16dWbNmoXBYKBOnTrExsby119/MWDAAK5cuUKtWrWIjo7myJEjAPzxxx88//zzHDlyhO7duzN06FBmzpzJ1atX2bFjB9OnT6dmzZoOTkRERHJj5syZxMbGYjKZePLJJ0lJSWHixImYTCbq1auH2WzmxRdfZOPGjezYsYObN2/Sq1cvQkNDbfcHgCeeeMLBeyIiIvZyN989vLy8ct3BJSIi95/DO7Y2bNhAhw4dGD58OADbtm1j6dKlzJ49G4CdO3fi4uLCsmXLAEhOTrYNBe7Tpw/btm2jdevWtvVFRkZSp04dJk6cyIkTJ9KN5Pr9999Zvnw5N27coHHjxgwdOpThw4cTHx/PxIkT82R/zGZznqznQaG87EfZ2o+ytY97LXo5fPhwjh49SnR0NFFRUcTHx9vm1a1bl7feeovNmzdz9uxZYmJiSEtLo3379gQHBxMeHs4zzzxDr1692LJli+2ecy90nOSOcrMfZWs/yjbv2asQ8t1895CM5fY41/mRPWWUNeWTvQcho6zuCw7v2OrZsyfh4eEMHDiQChUq0KJFizuW+fuv59HR0SxcuBCr1cqpU6do27ZtumV/+eUXgoODAfD29qZ06dK2eXXr1sXV1RVXV9e8eRQlAw/62whyQm9vsB9laz/K1jk1btwYgEOHDvHDDz9gMpmAWz+WnD59GrPZTN++tx4lb9KkSZ5sU8dJzun8sh9laz/K1rnczXcPyVhujnOdH9lTRllTPtlTRvmgY8vNzY3XX38dgGHDhnHgwAHS0tLSLePi4mL786RJk9i7dy//+te/CA0NxWq1plvWx8eHPXv2EBAQwIkTJ7h06ZJtXkadWbeHIWdHtbNERPK3rK7nRuOtd6X4+vrSqlUrpk+fDkBqaiqurq62e0f16tX5/vvvs9+Y6meJiDilu/nukRuqPyUi4jgOfyviF198QVBQECaTibNnz9KlSxdOnTpFaGgou3fvvmP5p59+mrZt2xIaGprh+nr37m2rsfLhhx9Svnz5LLffpEkTtm3bRmhoKL/++mue7JOIiNx/5cqVo3jx4vTq1YukpIy/YAQEBFC+fHmCgoIICQmhW7dupKSkMHr0aFasWEHHjh3Ztm3bfW65iIjcLzn97iEiIvmfITEx0Zr9Ys4lNTUVNzc3fvnlF/r27cv27dsd3STJgIZM2o+ytR9lK2I/Or/sR9naj7IVyZzOj+wpo6wpn+wpo3wwYsseOnbsSHBwMM8//zxvv/0248ePp3379nlSDFhERPKXuLg4W82s28LDwzl48CAAderUASAqKoqpU6fe9XoHDx6sH0ZERAqIjO4V2dm+fbvtpVUiIpJ/ObzGlj3ExMSk+/vgwYNtX3CyYrVa7VZUXkRE7p/Ro0c7ugkiIiIiInIfFMiOrb8bNWoU58+fx2Qy2TqtkpKSCA0N5fnnnycuLo7Q0FBq1apFcnIyixcvznA9C/YuuKvtqci8iIjjrFixgujoaAwGA/3796dly5YZLrdr1y4mT56MwWDgoYceYtasWRQuXJjJkyezadMmvLy8+PPPPzPfUERE9o1RgXkRkXznt99+Y+TIkaSlpVGsWDFmzZpF6dKlWbx4MfPnz8doNBIcHMyYMWPSfe6DDz7g5MmTTJs2Ld2LrW6rG1nifu2CQ6g4vojkZwW+YysiIoJNmzYRHR1NcnIyRYsW5fr16zRp0oRevXoBEB8fz6pVq/D09HRsY0VEJNfeffddLly4wPz58xk2bFimy1mtVl588UXWrFlDyZIl+eijj4iMjKRFixbs3r2bTZs2kZKSQtOmTe9j60VE5H549dVXGTNmDH5+fnz22WfMmDGD0aNHM3PmTLZu3YqbmxsdOnSgbdu2AKSlpTFmzBgqVarEjBkzHNx6ERHJSIHv2LrNYrEwffp0du/ejdFo5PLly/zxxx8A1KxZM886tcxmc56s50GhvOxH2dqPsrWPeyl6aTabOX/+PFu2bMnwl/S/u3TpEqdOnaJnz54A3Lhxg5YtW2I2m2nYsCFGo5FChQpRv379XLfndpsk55Sb/Shb+1G2ec9ehZDNZrPth4smTZqwZs0aTp48Se3atSlUqBAAjRo1wmw2U6ZMGb777jsqVarEu+++a5f2OIt7OcZ1fmRPGWVN+WTvQcgoq/vCA9OxFRsby48//khMTAzXrl2jUaNGWK23XghpNOZdDf0H/W0EOaG3N9iPsrUfZZs/+fj4MHLkSLp168aiRYuyXLZ06dJUqVKFJUuWUKLErUdHUlJS+Pnnn1mwYAEWi4XU1FQOHDhwz22SnNH5ZT/K1n6UrXPx8fFh165d+Pn5sXv3bqpXr06VKlU4fPgw169fx93dnT179vDss8+SkJBAq1ataNq0KX369OHTTz/Fw8PD0bvgELk9xnV+ZE8ZZU35ZE8ZPUAdWz4+PhgMBtq3b0+1atUoVapUjj6v2lkiIvlb27ZtKVKkCN26daNo0aKZLmcwGJg2bRq9evXCYrFgMBgYNWoU/v7+NG7cGH9/fypVqkTlypUz35jqZ4mIOKVJkyYxcuRIpk2bRpEiRZg9ezZlypRh6NChBAUFYTQaCQoK4tFHH7W9GTc0NJRixYrRs2dPIiMjKVKkyB3rVQ0qERHHMSQmJlod3Qh5MKln2X6Urf0oWxH70fllP8rWfpStSOZ0fmRPGWVN+WRPGUHePYMnIiIiIiIiIiJyHxXYjq24uDhMJtNdLbtmzRri4uLs3CIREbGXgwcPEh4enqvPmkwm3QNERERERJzUA1NjKyvR0dGUKFEiy3oqC/YuyHY9qsMlIuIYdevWpW7duvdvgxERWc9XDS4RkQdK3cgSjm6C3ah+mIjkdwW6Y+vq1auMGDGC2NhYmjVrxiuvvMLQoUP5/fffsVqtTJ06FXd3dzZt2kRsbCwPPfQQK1eudHSzRUQkG8ePH2fYsGG4u7uTlpbGoEGDWLduHbNnz2bw4MEULVqU+Ph4Tp8+zfz586lRowbR0dFMnTqVihUrUrp0aR5++GEmTpyYbr1Tpkxhx44d3Lx5k169ehEaGuqgPRQRkXv1z2v6kiVLmD9/PhUqVODs2bP079+fmJgYVq9ezaxZszAYDNSsWZPp06cTHx9P7969qVu3ru27xJQpUxy9SyIikoEC3bF15swZVq9eTZEiRXj88cepVKkS3t7ezJ8/H7PZzAsvvMDmzZvx9/ene/futGzZ8p62Zzab86jlDw5lZj/K1n6UrX3kpOjlhg0b6NChA8OHDwdg27Zt6eZXqlSJadOmERUVxcKFC3nzzTd55ZVX+Pbbb/H09GTw4MF3rHPz5s2cPXuWmJgY0tLSaN++PcHBwZQpUybH+6JjJPeUnf0oW/tRtnnvXgshZ3RNN5lMLFu2jDFjxrB06VK6du1KYmIiM2bMYN26dXh4eDB+/HjWrl3Lo48+esd3ifHjx1O8ePE82kPnca/Ht86P7CmjrCmf7D0IGWV1XyjQHVs+Pj4UK1YMgIoVKxIbG8vTTz9tm3fx4sU8357cPb29wX6Urf0o2/yhZ8+ehIeHM3DgQCpUqECLFi3Sza9fvz4AXl5efPfdd1y8eJHSpUvj6ekJQKNGjTh//ny6zxw6dIgffvjBVp8xOTmZ06dP56pjS8dI7uj8sh9laz/KNn/K6JresGFDJk6cyOjRo1m5ciVr167ll19+4cyZM7bvCFevXqVatWo8+uijd3yXSExMfCA7tu7l+Nb5kT1llDXlkz1lVMA7tgwGQ7q/16lTh127dhEQEIDZbKZ06dIAtkdZRETEObi5ufH6668DMGzYMK5du5Zu/t+v/1arlTJlynDp0iUSExPx9PRkz549eHl5pfuMr68vrVq1Yvr06QCkpqbi6lqgb5MiIgVWZtf0WrVqERERQe3atSlevDje3t5UqVKFlStX4u7ublv27Nmzd3yXsFqt930/REQkew/Uf7H7+fnx9ttvExQURFpaGu+99x4AQUFBvPPOOyxevJh58+Zl+FkVhhcRyT+++OILli5ditFoxMPDI9svG0ajkTfffJOQkBAqVKhAiRIlbF9gbgsICGD//v0EBQXh4uJCoUKFiIqKwsPD484Vqji8iEi+ltk1vWfPnnTu3JmvvvoKgJIlSzJq1Cg6d+6MwWDAaDQyefJkSpTIWTF4FVgXEXEcQ2Jion56EIfQkEn7Ubb2o2ydV2pqKm5ubgAMHDiQkJAQOnbs6OBWyd/p/LIfZWs/ylYkczo/sqeMsqZ8sqeMHrARWyIi8uBatmwZS5cu5fr169SsWdNWd0VERERERJyX0dENEBERyQtxcXFZdlb16tWLtWvXsnnzZmbNmoWLi0uGyyUmJhIVFWWvZoqIiIiISB56YEdsWa3WOwpCZmXB3gXZLqM6XCIizi8pKYklS5bQo0ePzBeKiMh6JarBJSLyQKkbmbOaXM5AdcNExFkUiI6t48ePM2zYMNvbDYODg0lOTmbixInArbchxsbGEhUVxTfffENqaiq1a9fmySefZPLkyRgMBh566CFmzZpF4cKFHbw3IiJyL86ePcuYMWO4evUqAO+//z5Vq1bljTfeYM+ePVy5cgV/f39ee+01Lly4QN++fbFarVgsFqZPn878+fOJjY3FZDLRp08fnn32WQfvkYiI2NvkyZOpUaMGzz77LFarlVatWjF16lTefPNNjEYjFStW5MMPP6RQoUKObqqIiPxDgSge/9FHH2GxWBg+fDgAixcv5tSpUxl2bH3++eesXLkSgBYtWrBmzRpKlizJRx99hNFoZPDgwRlu425GbLUo0SKP9khE5MFzr0Uv4+LiGDJkCP/+97/p27cvzZs356effuLdd98lKiqK5ORkihYtisVioW3btsydO5dDhw6xbds2pk2bBoDFYuHUqVMMGTKE6OjozDeWzYgts+p3iYjck/tdCPn2tX/16tVs3bqVFStWcPjwYebOnYuPjw9vvvkmZcuW5YUXXsjw8wVxxNaXzfc6ugkiIjZZ3RcKxIitnj17Eh4ezsCBA6lQoQLVq1e3zfvnK+AbNmyIwWDg4sWLnDp1ip49ewJw48YNWrZseU/teNDfRJBTenuD/Shb+1G2+d+hQ4d46623bH+//dh5ZGQk0dHRGAwG4uPjOXPmDO3atePYsWMMHDgQT09P2w8i90rHSO7o/LIfZWs/yrZgePjhhylSpAjHjx/ns88+Y8iQIfTr18/2b9ukSRPWrVvn4FbeX3lxXOv8yJ4yypryyZ4yKiAdW25ubrz++usADBs2jJo1a3LmzBkAfvzxx3TL3i4WXLp0aapUqcKSJUsoUeLWLywpKSmZbkP1s0REnIOvry9Dhw6lQYMGwK1re2JiIpGRkXz//fcYDAYCAwOxWq3cvHmTsLAwAN59912WLFnC008/TVpaWtYbUQ0tEZECp2/fvrz33nucPHmShg0bUrJkSdsXxt27d6f78fyfVI9KRMRxCkTH1hdffMHSpUsxGo14eHjQunVrlixZQqdOnWjevHmGnzEYDEybNo1evXphsVgwGAyMGjUKf3//+9x6ERHJS1OmTCEsLIykpFtfMgICAhgxYgR169YlMDCQqlWrUqxYMQB27NjBjBkzcHFxwWKxMHv2bMqVK0fx4sXp1asXXbp0ISQkxJG7IyIi90lgYCBhYWGMGDECgPfee4+hQ4fi4uJC+fLlbT+EiIhI/lIgamyJc9KQSftRtvajbEXsR+eX/Shb+1G2BYfFYqF169asWbPG9kSH3BudH9lTRllTPtlTRmB0dANEREREREQc6fDhw4SEhNCtWzd1aomIOJkC07GVmJhIVFSUo5shIiIOcv36dTp27EhISAiHDx++q89ERUWxcePGO6ZPnTpV9xQRkQdI7dq1Wbt2LUOGDHF0U0REJIcKRI0tgKSkJJYsWUKPHj3STbdarbY3Yt2LBXsXZDpPheVFRBwvNjaWsmXLMm/evHTTs7oP/POecdciIjKfp8LyIiIPnLqRBW+Ulwrii4izKDAjtmbOnElsbCwmk4np06fTu3dvunfvzltvvcXgwYPZvn07ANu3b2fw4MEAHDt2jM6dOxMSEsJzzz3HxYsXHbkLIiJylxITE+nUqRMmk4k2bdqwc+dORo8ezbZt2zCZTJw4cYLWrVszaNAgevXqhdlspkOHDgQHBxMYGMixY8eA9COz5s+fT4sWLejWrRs///yzI3dPRETy2HvvvceyZcsAaNasGZ988gkAwcHBrFy5kpCQENq1a0eXLl3466+/+PXXX3n66adtn582bRoLFy50SNtFRCRrBWbE1vDhwzl69CjR0dFERUWxfft2Vq5cicFgsHVk/dPo0aOZPXs2VapUYfXq1UyfPp2pU6fmeNtms/lem//AUnb2o2ztR9naR06KXhYpUoQVK1bg4eHB0aNHGTNmDFOnTmXJkiXMnj2buLg44uPjWbVqFZ6enly7do2vvvoKFxcX1q9fT3h4uO1LDcDFixeZO3cuW7ZswdXV9Z7ehKjj494oP/tRtvajbPNeXhdCbtOmDfPmzcPPzw8vLy927NhBz549uXbtGoGBgXTu3BmAV199lS+//JJevXphtVo5efIkXl5erFy5kg0bNuRpm/K7vDqudX5kTxllTflk70HIKKv7QoHp2Pqnhg0b2h49+fsjKFbr/14CefToUYYOHQpAWloaXl5eudrWg/4GgtzS2xvsR9naj7LNH65cucK4ceM4e/YsRqORM2fO3LFMzZo18fT0BODcuXO8/PLLJCYmkpKSQqFChdIte/LkSWrVqoWHhwcAjRs3znXbdHzkns4v+1G29qNsnUP9+vU5ePAgW7ZsISQkhFWrVrF161aaNWvG7t27mTlzJqmpqVy4cIEiRYoA0KdPHz777DOaN29OkyZNbNMfFHlxXOv8yJ4yypryyZ4yKkAdW+7u7qSlpdn+7uLiYvtzyZIlbV969u/fb5vu6+vLJ598QqVKlQBISUnJdP2qoyUikn8sW7aMGjVq8Omnn3LkyBG6dOlyxzJG4/+etv/444/p2LEjXbt2Zd26dcycOTPdslWqVOHYsWPcuHEDNzc39uzZk/V/IKiOloiIU3FxccHb25t58+YRGRnJ6dOnCQ8PZ9y4cfz3v/9l+fLlVKpUiVdeecX2Q/jtEidHjhzhpZdeynL9qkclIuI4BabGVrly5ShevDi9evUiKSn9jSU0NJQPP/yQLl26cPr0adv08PBwRo0aRYcOHejQoQNffPHF/W62iIjkgr+/P19//TVPP/00X331VbbLd+jQgffff58uXbqwY8eOO+aXKVOGAQMGEBAQQLdu3ShTpowdWi0iIo7k5+fHlStXqFixIn5+fhw4cIBmzZrRvXt3nnnmGXr27MmlS5dsy7u6utKuXTsuXLhAvXr1HNhyERHJiiExMdGa/WIieU9DJu1H2dqPshWxH51f9qNs7UfZFmzvvPMOFSpUoFevXo5uilPS+ZE9ZZQ15ZM9ZVSARmxl5ODBg4SHhzu6GSIico/i4uIwmUx3tezf33T4d+Hh4Rw8ePCut2kymYiLi7vr5UVExHmcP3+eCRMmZLnM0KFD2b17d4aPu4uISP5RYGpsZaRu3brUrVvX0c0QEZF8YPTo0Y5ugoiI5BPlypXj7bffznKZjz766D61RkRE7kWB6diKi4ujd+/e1K1bl9jYWJo1a0b79u1tr34/cuQI48ePx2Kx4OrqysqVKzl37hxjxozh6tWrALz//vtUrVo1w/Uv2Lsg022rsLyIiP0lJyczaNAgzGYzLVu25JVXXiEsLAyz2cyNGzd48cUXadeuHQC7du1i7dq1/Prrr0yfPp3mzZszePBgunfvTsuWLfH19eWpp57ixx9/pHz58kRGRmKxWBg5ciTHjx/H29ub5OTkzBsTEZH5PBWWFxHJVywWCyNGjMBsNlO1alWOHDnCwoULGTJkCF9//TUtWrTg22+/pVChQvzwww/MmTOHefPmMX/+fFasWAHAk08+ybhx4zLdRt3IEvdrd+xOhfBFxNkUmI4tgDNnzrB69WqKFCnC448/TvPmzW3zRowYwXvvvcdjjz2GxWLBaDTy2muvMXz4cJo3b85PP/3Eq6++muHjKyIi4nhxcXGsWrWK4sWL07FjRz766CPKly9PeHg4V69eJTAwkMDAQAA8PDyIiopi+/btfPzxx+nuBwAXLlxg6NChVKxYEZPJxOHDhzl58iQpKSmsX7+eS5cuUb9+fUfspoiI5LG1a9eSlpbG+vXrSUxMTFcI3sXFhaCgINasWcMzzzxDVFQUPXr04Pjx46xZs4a1a9diNBrp3r07sbGx1KlTx4F7IiIiGSlQHVs+Pj4UK1YMgIoVK5KYmGib98cff/DYY48B/3sF/KFDh3jrrbdsyxgMhlxt12w2567BouzsSNnaj7K1j+yKXvr4+ODp6QlAw4YNiYqKwt3dnW3btgGQmprKxYsXAWydUl5eXiQkJNyxrnLlylGxYsV0y5jNZho3bgxA6dKlqV69eq72Q8fHvVF+9qNs7UfZ5r28LIT8yy+/0KhRIwA8PT3vWHePHj0ICwvDZDKxZ88ewsPD+frrrzGbzXTo0AGApKQk4uPjH4iOrbw+nnV+ZE8ZZU35ZO9ByCir+0KB6tj6Z8eU1fq/Fz6WK1eOgwcPUrduXduILV9fX4YOHUqDBg0ASElJydV2H/Q3EOSW3t5gP8rWfpSt45jNZpKSkihevDj79u2jW7duGI1GRo4cCdy6hru7uwPp7wd/vxfcltH9wsfHhzVr1tCvXz8uX77M8ePHc9VOHR+5p/PLfpSt/Sjb/O/v1/fExMQ7vgA+8sgj3Lhxg1mzZmEymTAajdSsWRNfX1+WLVuGwWDAYrFgsVgctAf3V14ezzo/sqeMsqZ8sqeMCljHVlbef/99wsLCsFqtuLm5sXLlSqZMmUJYWBhJSbeeIw8ICLB9Qfon1dESEXGsypUrExYWxi+//EKrVq0YMWIEEyZMICQkBIPBQLly5Zg3b16u1x8UFMS6deto27YtVapUwdvbO/OFVUdLRMRpBAUFERMTQ7t27ahcuTKVKlXi3Llz6Zbp1q0bo0ePZs+ePQDUrFkTk8lEcHAwLi4uuLm58dFHH1GhQoUMt6G6VCIijmNITEy886dskftAPcv2o2ztR9mK2I/OL/tRtvajbJ1Damoqbm5uXLp0CT8/Pw4cOGArTyL2o/Mje8ooa8one8roARqxJSIiIiIiD6YBAwZw4cIF/vzzT9588011aomIFCAF5ooeFxeHyWS6q2XPnz/PhAkT7NwiERFxNJPJRFxc3F0tO3XqVL0ZV0SkgPrss89Yu3Yt27dvp1OnTjn67MGDB9m+fbudWiYiIvfqgRuxZbVaKVeuHG+//XaOPrdg74IMp6v2loiI88uowHyWIiIyn6f6WyIiBUpsbCzx8fG0bNky02XqRpa4jy2yH9UKExFnVKA6tiwWC82bN2fDhg3861//YteuXSxYsIA5c+bg6+uLyWTi8OHDfPTRRwwbNozo6GhOnz7N6NGj+euvvzAajcybN49y5co5eldERCQXLBYLI0eO5Pjx43h7e5OcnExiYiIjRozg5s2bXL16lbfeeotmzZoRFRXFN998Q2pqKrVr17Y9lnL9+nWGDRtG69at6dWrl4P3SERE7CEuLo7Q0FBq1aqF2WymZcuW+Pj4EB8fz8SJEwGoU6cOsbGxzJw5k6tXr7Jjxw6mT59OzZo1Hdx6ERH5uwLVsWU0Gnnqqaf4v//7P3r27ElkZCS9e/cGbj1+OHToUKpUqZLusZRXXnmF3r17ExISApDj1/j+83XBkjPKz36Urf0oW/vIi6KXMTExpKSksH79ei5dukT9+vUpWrQoK1aswMPDg6NHjzJmzBhiYmIASEpKYuXKlRgMBqZOncqlS5fo0aMHw4YNw8/PL1dt0PFxb5Sf/Shb+1G2ee9+FEKOi4tj1apVFC9enI4dO1K2bNkMlxs+fHi6Dq+CzB7Hss6P7CmjrCmf7D0IGWV1XyhQHVsAoaGh9OnTh5CQEI4ePUrz5s0BKF++PFWqVLlj+aNHj+Lv72/7e04LST7obx+4F3p7g/0oW/tRtvmb2WymcePGAJQuXZrq1atz5coVhgwZwtmzZzEajZw5c8a2fMOGDTEYDLa/z5kzh86dO+e6Uwt0X7gXOr/sR9naj7J1Xj4+Pnh6egK37geenp4kJd16FC/Hj6gXEHl9LOv8yJ4yypryyZ4yKkDF428rV64c5cqV45VXXqFLly626Zl1WPn6+vLtt9/a/p7TEVsiIpJ/+Pj4sGfPHgAuX77M8ePH2blzJzVq1CAmJoZp06al+7Li4uKS7vPjxo0jISGBGTNm3Nd2i4jI/Wc2m0lKSsJqtbJv3z5Onz5t+/Hjxx9/tC3n7u5OWlqao5opIiLZKHAjtgD69u3Lc889x9GjR7Nd9s0332TUqFF8+OGHuLi4ZFpjS0XiRUTyv6CgINatW0fbtm2pUqUK3t7e+Pv7M2DAAHbv3k3Dhg2z/LyLiwszZ87kpZdeYtKkSfz3v//NeEEViBcRcXqVK1cmLCyMX375hVatWjFixAiee+45OnXqZHvqA6BJkybMnTuX0NBQ/vvf//LII4/csS4VXRcRcRxDYmJigRtn+91337Fw4UI++eQTRzdFsqAhk/ajbO1H2YrYj84v+1G29qNsnVNcXBxDhgwhOjra0U0p0HR+ZE8ZZU35ZE8ZFcARWx9//DHLly9n3rx5jm6KiIiIiIiIiIjYkVPV2Bo/fjzt27dn2bJlmS4zaNAgvv322wyHCIuISMESFxfHmjVrcv35CRMmcP78+Tumm0ymdG/QFRGRgqVy5coZjtaKiopi48aNd0yfOnUqUVFR96NpIiKSQ041YismJoaDBw86ZNsL9i7IcLpqb4mIOE58fDzR0dGEhITk6vNvv/127jYcEZH5PNXfEhFxWj169MjV5+pGlsjjljiGaoWJiDNymhFbo0aN4vz585hMJipVqmSbPnjwYLZv3w7cesPhSy+9RPv27enTpw9w69f8Nm3aMGLECPz8/Hj55ZcBGDhwIN9//z0Af/31F82aNdMbEUVE8rnjx4/Ttm1bQkJCCAoKYubMmWzatAmTycTWrVvT3RO2b9/O4MGDgVv3itGjR/Pss8/i7+/Pzz//DPxvZJbFYmH48OG0a9eOQYMGkZyc7LB9FBGRvPfP+8fevXvp0KEDwcHBBAYGcuzYMSD9yKz58+fTokULunXrZrtviIhI/uM0I7YiIiLYtGkT0dHR1KlTJ8NlLly4wNChQ6lYsSImk4nDhw9TtGhRzpw5w+rVqylSpAiPP/4448ePp2/fvixYsIDGjRvz5Zdf0qlTJ4zGnPfzmc3me921B5rysx9laz/K1j7upujlhg0b6NChA8OHDwdg27ZtLF26lNmzZwNk+ah6uXLlCA8PZ+vWrUyaNIklS5bY5sXExJCSksL69eu5dOkS9evXz/V+6Pi4N8rPfpSt/SjbvJfXhZD/ef+4du0aX331FS4uLqxfv57w8PB0L566ePEic+fOZcuWLbi6uuZ6ZLCzscexrPMje8ooa8onew9CRlndF5ymYyszVuv/XupYrlw5KlasCICXlxcJCQkULVoUHx8fihUrBkDFihVJTEykadOmvPLKKyQmJhIVFcX8+fNztf0H/e0D90Jvb7AfZWs/ytaxevbsSXh4OAMHDqRChQq0aNEi3XyDwWD789/vDwCNGjUCoHHjxoSFhaWbZzabady4MQClS5emevXquW6jjo/c0/llP8rWfpStc/jn/aNLly68+eabJCYmkpKSQqFChdItf/LkSWrVqoWHhweA7R5R0OX1sazzI3vKKGvKJ3vKyEk7tqxWK9euXcPNzY2DBw/anoX/+xea28tlNb1r1668+uqrlCpVigoVKmS5TdXSEhFxPDc3N15//XUAhg0bxoEDB0hLS7PNL1myJGfOnAFg//796T67d+9e/P392bNnD9WqVUs3z8fHhzVr1tCvXz8uX77M8ePHs26I6miJiDiVf94/BgwYwPDhw+natSvr1q1j5syZ6ZavUqUKx44d48aNG7i5ubFnz54svziqNpWIiOM4ZcfW4MGDCQgIwNfXl/Lly+d6PV26dGHSpEm5Hq0lIiL31xdffMHSpUsxGo14eHgwfvx4Bg4cSGhoKEOGDCE0NJT+/fuzcuVKvLy80n32999/55lnniEhIYFZs2almxcUFMS6deto27YtVapUwdvb+37uloiI2Nk/7x/vvPMOYWFhrFy5MsMOqzJlyjBgwAACAgKoUKECZcqUcUCrRUTkbhgSExOt2S9WMCUnJ9OuXTu2b9+eq/pacm80ZNJ+lK39KFvnNHjwYLp3707Lli0d3RTJgs4v+1G29qNsRTKn8yN7yihryid7ysiJ3oqY17Zu3Urnzp0JCwtTp5aIiIiIiIiIiBMq0D0648ePp3379hm+Jat169Zs2LCBp5566v43TERE7qvZs2drtJaIiGTp/PnzTJgwIcN5gwcPZvv27fe5RSIicjecssbW3YqJieHgwYN5sq4FexdkOF1F5UVEHkAREZnPU2F5ERGnVK5cOd5+++07pv/zLbsZqRtZwh5Nui9U+F5EnF2BHbE1atQozp8/j8lkolKlSrbpf/+1xdfXl5deeon27dvTp08fB7VURESyM2XKFIKCgggMDGThwoVMnDiROXPmADBo0CA+//xzACZPnkxgYCAhISEsWbKEuLg4TCaTbT0mk4m4uDiOHz9O27ZtCQkJISgoiEuXLnHz5k1Gjx5NSEgIgYGBrF+/3iH7KiIi9mexWBg2bBjt2rVj8ODBtG7dOt09Y+rUqQwePJiuXbsyb948B7dWRESyUmBHbEVERLBp0yaio6OpU6dOhstcuHCBoUOHUrFiRUwmE4cPH6Z27do52o7ZbM6L5j6wlJ/9KFv7Ubb2kVnRy82bN3P27FliYmJIS0ujffv2LF26lN69e/Pbb79RqFAhnn32Wb755ht+/vln1q9fj9FoJC0tjdOnT2e4zg0bNtChQweGDx8O3Po1/rPPPqN8+fKEh4dz9epVAgMDCQwMzHEdRh0f90b52Y+ytR9lm/fsXQh57dq1pKWlsX79ehITE6lXr94dy7i4uNhKmgwePNiu7XEkex+/Oj+yp4yypnyy9yBklNV9ocB2bGXm70OJy5UrR8WKFQHw8vIiISEhx+t70N8+cC/09gb7Ubb2o2zvv0OHDvHDDz/YfkVPTk7m9OnTDBw4kP79+3P48GEAjh49Sps2bWwdUS4uLhgMhnTrun0P6NmzJ+Hh4QwcOJAKFSowYcIEDh06xO7du9m2bRsAqampXLx4kbJly+aovTo+ck/nl/0oW/tRts7pl19+oVGjRgB4enpm+G/4xBNP3O9mOYQ9j1+dH9lTRllTPtlTRg9Ix5bVauXatWu4ublx8OBBevToAZDpF56MqJaWiIhj+Pr60qpVK6ZPnw7c6nBKSkpi1KhRzJgxg7Fjx7J48WJ8fX1ZtGgR/fr1w2AwkJaWhqenJ+fOncNqtXL16lXbr1lubm68/vrrAAwbNowNGzbg6+vLww8/zMiRIwFISUnB3d0940apjpaIiFPz8fFhzZo19OvXj8TExAxHO7i4uNz1+lSnSkTEcR6Ijq3BgwcTEBCAr68v5cuXd3RzREQkBwICAti/fz9BQUG4uLhQqFAhACZNmkTr1q359ddfmTlzJsOHD2f37t0EBATwr3/9i27dutG9e3fat2+Pv78/jz76qG2U7hdffMHSpUsxGo14eHjQokULihcvzoQJEwgJCcFgMFCuXDnVVRERKaCCgoKIiYmhXbt2VK5cmUqVKnHu3DlHN0tERHLBkJiYmP1rPkTsQEMm7UfZ2o+yFbEfnV/2o2ztR9k6r9TUVNzc3Lh06RJ+fn4cOHAgx3UVJWs6P7KnjLKmfLKnjArYWxHHjx9P+/btbUUeRURE7sb58+eZMGECU6dOJSoqCiDTF4+IiEjBMGDAALy8vAgODubNN99Up5aIiJMqUFfvmJgY1q1bR9euXW3TsqqbJSIiArdeJvL22287uhkiInIfffbZZ9SpU4cVK1bQqVMnRzdHRERyqcDU2Bo1ahTnz5/HZDLx008/MWjQIPbu3csHH3xAeHg4ZrOZGzdu8OKLL9KuXTvOnj3LmDFjuHr1KgDvv/8+VatWzXT9C/YuyHC6isqLiDi/uLg4hgwZQosWLe7uAxERmc9TYXkRkXzLYrEwcuRIjh8/jre3N8nJySQmJjJixAhu3rzJ1atXeeutt2jWrBnBwcHMmTOHSpUqcerUKV544QXWrl2b4XrrRpa4z3uSN1T0XkQKggIzYisiIoKyZcsSHR1NyZIlefLJJ/nqq6/YvHkz5cuXZ82aNXz11VdMmjQJi8XCa6+9xvDhw1m9ejWTJ0/m1VdfdfQuiIiIiIiIHcXExJCSksL69euZMmUKJ06coGjRoqxYsYLo6Ghmz57Nm2++CUDv3r1ZuHAhAAsXLqRXr16ObLqIiGSiwIzY+qfGjRsDcOjQIXbv3s22bduAW0UiL168yKFDh3jrrbdsyxsMhlxtJ6NXA8vdU372o2ztR9naR0Epeqnj494oP/tRtvajbPOeve4JZrPZ9j2hdOnSVK9enStXrjBkyBDOnj2L0WjkzJkzADz11FMEBAQwduxYoqOj2bRpk13a5Ej349jV+ZE9ZZQ15ZO9ByGjrO4LBbZjy8XFBQBfX18efvhhRo4cCUBKSgru7u74+voydOhQGjRoYJueGwXli5gj6O0N9qNs7UfZSnZ0fOSezi/7Ubb2o2ydi4+PD2vWrKFfv35cvnyZ48ePs3PnTmrUqMGnn37KkSNH6NKlCwAeHh60adOGsWPH0rJlSwoXLuzg1uc9ex+7Oj+yp4yypnyyp4wKcMfWbb1792bChAmEhIRgMBgoV64c8+bNY8qUKYSFhZGUdOu58oCAAFvnV0ZUS0tERGxUR0tExCkFBQWxbt062rZtS5UqVfD29sbf358BAwawe/duGjZsmG75fv368fjjj7Nz584s16taVSIijlOgOrZiY2PT/T/cGrk1bdq0O5b997//zeLFi+9b20REJP+qXLky0dHR6ab9/V4iIiIFg9FoZObMmXdM37Fjh+3PL730UrrlmzVrRq1ate5L+0REJOcKTPF4ERERERGRvPLll1/Sp08fXnnlFUc3RUREsuBUHVtxcXGYTKY8W1+dOnXybF0iIiIiIlJw/Oc//+Hbb7+ladOmjm6KiIhkoUA9imhPC/YuyHC6am+JiDgfq9Wa67fhAhARkfF01d4SEXF6ublH1I0sYafW5D3VAxORgsapRmzdtnLlSkJCQmjXrh1dunThr7/+AtKPwJo6dSpRUVEAzJ8/n5YtW9KjRw/69Oljmw7w7rvv0rFjR0JCQrh+/fr93REREcmV9957j2XLlgHQrFkzPvnkEwCCg4PZt28fHTp0IDg4mC5dunDp0iXg1j1i8uTJPPXUU5w6dYqQkBBMJhNBQUEcOXIEgClTphAUFERgYCALFy50zM6JiEiu/PMa3r59e86ePQvA2bNnCQoKAmD16tUEBQURHBzMmDFjsFqtxMXF0bp1awYNGkSvXr0YOHAg33//PQB//fUXzZo1w2KxOGzfREQkc045YiswMJDOnTsD8Oqrr/Lll1/Sq1evDJe9ePEi8+bN49tvv8XNzY1OnTqlm+/v78+4ceMYPHgw3377re2Gd7fMZnPudkIA5WdPytZ+lK195OQ1xW3atGHevHn4+fnh5eXFjh076NmzJ9euXWPGjBmMGTMGPz8/PvvsM2bMmMGUKVMAePLJJ3nllVdYvXo1tWrVsr1cxGKxsHnzZs6ePUtMTAxpaWm0b9+e4OBgypQpc9ft0rFx75Sh/Shb+1G2eS+nr67P6BpuMplYtmwZY8aMYenSpXTt2pXExERmzJjBunXr8PDwYPz48axdu5ZHH32U+Ph4Vq1ahaenJ7t27WLBggU0btyYL7/8kk6dOmE0OuWYgDvc7+NV50f2lFHWlE/2HoSMsrovOGXH1u7du5k5cyapqalcuHCBIkWKAKQbMmy1WgE4efIkNWvWxMPDA4AGDRqkW9fjjz8OgJeXFwkJCTluS05vuvI/ZrNZ+dmJsrUfZZs/1K9fn4MHD7JlyxZCQkJYtWoVW7dupVmzZmzYsMFWD6VJkyasWbPG9rnGjRsD0K5dO44dO8bAgQPx9PRk4sSJHDp0iB9++MFWyzE5OZnTp0/nqGNLx8a90fllP8rWfpRt/pDRNbxhw4ZMnDiR0aNHs3LlStauXcsvv/zCmTNnePrppwG4evUq1apV49FHH6VmzZp4enoC0LRpU1555RUSExOJiopi/vz5jtq1PHc/j1edH9lTRllTPtlTRk7asfXf//6X5cuXU6lSJV555RVbJ1bJkiU5ffo0lSpV4sCBA1SuXJkqVapw7Ngxbty4gZubG/v27Uv3j55RZ5iIiORvLi4ueHt7M2/ePCIjIzl9+jTh4eGMGzeOkydPsmvXLvz8/Ni9ezfVq1dP9zmAmzdvEhYWBtx6JH3JkiX4+vrSqlUrpk+fDkBqaiqurk55mxQReeBkdg2vVasWERER1K5dm+LFi+Pt7U2VKlVYuXIl7u7utmXPnj17x4isrl278uqrr1KqVCkqVKhw3/dJRETujlP+F3v37t155plnqFatGsWKFbON2Bo1ahTPPvssPj4+tmllypShf//+BAQEULFiRYoVK2a7ieWEisSLiOQvfn5+zJ07l4oVK+Ln58f06dNp1qwZ3t7ejBw5kmnTplGkSBFmz559x2d37NjBjBkzcHFxwWKxMHv2bKpUqcL+/fsJCgrCxcWFQoUKERUVZRvxm46KxIuI5CsBAQEZXsN79uxJ586d+eqrr4BbP4SPGjWKzp07YzAYMBqNTJ48mRIl7iz+3qVLFyZNmnRXo7VUkF1ExHEMiYmJBX6YUmpqKm5ubqSlpdGxY0emTJnCY4895uhmPfA0ZNJ+lK39KFsR+9H5ZT/K1n6UbcGVnJxMu3bt2L59e4Gpr3W/6fzInjLKmvLJnjJy0hFbOfXBBx+wefNmkpOT8ff3V6eWiIiIiIhkauvWrUyePJmwsDB1aomI5HMF7io9fvx42rdvb3sNPMDYsWOJjo5m69atvPbaa+mWj4uLS1dYWEREnEtcXJytWLCIiEheaN26NRs2bOCpp54CICoqiqlTpzq2USIikqECN2IrJiaGgwcP3vXy8fHxREdHExISkuVyC/YuuGOa6m6JiDygIiIynq7aWyIiD6S6kXfW6MqvVA9MRAqaAjVia9SoUZw/fx6TyURISAghISG0bNmSuXPnAnD8+HHatm1LSEgIQUFBXLp0iZkzZ7Jp0yZMJhNbt2518B6IiEhu/fbbb3To0IHg4GC6dOnCpUuXAFi8eDFPPvkkAQEBzJgxA4CdO3fi7+9vu1cAXLlyhf79+9OhQwfat2/Pvn37HLYvIiJyb+Li4mjTpg0jRozAz8+Pl19+mZs3bzJ69GhCQkIIDAxk/fr1AJw9e5auXbvSoUMHOnTowG+//QZAdHQ0LVq0oEuXLnz33XeO3B0REclCgSseX6dOHWJjY0lOTqZo0aJcv36dJk2asHv3bj799FMsFgvDhw8HwGq1smPHDpYsWZLhW7P+LqMRWy1KtLDLPoiIPIhyW/QyLi6OIUOG4OnpyYABA/Dz8+Ozzz7j+PHjjB492vbDhZubGx06dODdd99l0aJFNGvWjE6dOmGxWDAajbzxxhv4+PjQrVs3zp07R8+ePdm4cWPGG81kxJZZj0SKiOSJey2EHBcXR0BAAD/++CNFihTh8ccfZ9CgQSQlJTF+/HiuXr1KYGAgO3bsYODAgfTt25fmzZvz008/2e4TDRo0YPPmzZQsWZKBAwfi7e3NxIkTM9yeM43Y+rL5Xkc3QUQkx7K6LxS4RxEBLBYL06dPZ/fu3RiNRi5fvswff/xBz549CQ8PZ+DAgVSoUIEJEybc03Ye9DcP3Cu9vcF+lK39KNv8y2w207RpUwCaNGnCmjVrOHnyJLVr16ZQoUIANGrUCLPZzKhRowgPDyc6OpoaNWowZswYDh06xLZt21i8eDEASUk5f1RDx8a90fllP8rWfpRt/uXj40OxYsUAqFixIgcPHuTgwYNs27YNuPXm9IsXL3Lo0CHeeust2+cMBgMXL16kdOnSlCxZErh1Xzl//vz93wk7uJ/Hq86P7CmjrCmf7CmjAtqxFRsby48//khMTAzXrl2jUaNGWK1W3NzceP311wEYNmwYGzZsoGzZsqSlpWW7TtXTEhHJ33x8fNi1axd+fn7s3r2b6tWrU6VKFQ4fPsz169dxd3dnz549PPvss5QoUYJ3330XgI4dOxIQEICvry/16tWjc+fOAKSkpGS+MdXSEhHJ9wwGQ7q/N2zYkBo1ajBy5Ejg1nXe3d0dX19fhg4dSoMGDWzTXV1duXTpEomJiXh6erJ79268vb0z3ZbqVomIOE6B7Njy8fHBYDDQvn17qlWrRqlSpQD44osvWLp0KUajEQ8PD1q0aIGbmxunTp0iNDSUIUOG0KRJEwe3XkREcmPSpEmMHDmSadOmUaRIEWbPnk2ZMmUYOnQoQUFBGI1GgoKCePTRR5k2bRqbN28GoFy5ctSsWZOxY8cyduxY5s+fj9VqpW7duul+wRcREefm5+fHrFmzCAkJwWAwUK5cOebNm8eUKVMICwuzjdQNCAhg5MiRTJ48mQ4dOlC+fHlKly7t4NaLiEhmClyNLXEeGjJpP8rWfpStiP3o/LIfZWs/ylYkczo/sqeMsqZ8sqeMCthbEe+FyWQiLi7O0c0QEZFcyItr+NSpU4mKisqjFomIiDOqU6dOjqaLiIjjqWNLREQKHKtVg5FFRERERB4EBarG1uTJk9m6dSseHh50796ds2fPsm7dOqxWK/3796d79+5cvHiRIUOGkJycjKurKx988AFVqlTJdt0L9i64Y5oKyouIOIbFYmHkyJEcP34cb29vkpOTAWjdujW1atUiOTmZoKAg4uPjba9mr1OnDrGxsURFRbFx40YsFgvHjh1jwoQJtoLxANevX2fYsGG0bt2aXr16ZdyAiIg7p6mgvIiIQ7z33ntUqlSJrl270qxZM3r37s0LL7xAcHAwb775Jq+//jppaWkUK1aMWbNmUbp0ads9AWDw4MF0796dli1b2tb5xx9/8PzzzwPwxBNPZNuGupEl7LNzeUTF7UWkICswI7a++eYbfv75Z9avX8+aNWuoU6cOmzdvZv369URHR/PBBx9w6dIlpk+fjslkYu3atYwePZpXX33V0U0XEZEciomJISUlhfXr1zNlyhROnDgBQHx8PG+//TaLFy/O8vNXr14lMjKSFStWEPG3TqpLly7Ro0cPevTokXmnloiI5Ctt2rRhy5YtnD9/Hi8vL3bs2MHVq1e5du0aM2bMYMyYMaxdu5bg4GBmzJhxV+sMDw/nmWeeYdWqVTRv3tzOeyAiIveiwIzYOnr0KG3atMFovNVXZzabeeKJJ3BxccHFxYVatWpx8uRJzGYzffveGmnVpEkTJkyYkOttms3mPGn7g0wZ2o+ytR9lax85KXppNptp3LgxAKVLl6Z69eoA1KxZE09PTyD9a97/+WjiY489hsFgoHLlyiQkJNimz5kzh86dO+Pn55fj9uu4yBvK0X6Urf0o27yXk3tC/fr1OXjwIFu2bCEkJIRVq1axdetWmjVrxoYNG2jatClw67/916xZc8fnM3p8/Z/fGZydI49RnR/ZU0ZZUz7ZexAyyuq+UGA6tnx9fVm0aBH9+vXDYDBQtWpV5s2bh8ViISUlhaNHj1KlShV8fHzYtWsX1atXZ/fu3bYvQ7nxoL954F7p7Q32o2ztR9nmDz4+PqxZs4Z+/fpx+fJljh8/DmD7cQOgZMmS7Ny5E4Aff/wx3ecz6/QaN24cP/zwg+0X/py2Se6Nzi/7Ubb2o2wdz8XFBW9vb+bNm0dkZCSnT58mPDyccePGcfLkSXbt2oWfn1+6//a3Wq1cu3YNNzc3Dh48SI8ePdKt08fHhz179lC9enW+//57R+xWnnLUMarzI3vKKGvKJ3vKqAB1bAUGBrJ7924CAgL417/+Rbdu3WjdujWBgYFYrVaGDRtG6dKlGTNmDIMHD2b58uUYjUY++OCDu1q/6mmJiOQfQUFBrFu3jrZt21KlShW8vb3vWMbPz4/Zs2fTqVOnu36MxMXFhZkzZ/LSSy8xadIk/vvf/2a8oOppiYjkK35+fsydO5eKFSvi5+fH9OnTadasGd7e3owcOZJp06ZRpEgRZs+eDdyqqxUQEICvry/ly5e/Y32jR4/m+eefZ/ny5TRq1Cjb7auGlYiI4xgSExP16ihxCPUs24+ytR9lK2I/Or/sR9naj7IVyZzOj+wpo6wpn+wpowJUPF5ERERERERERB4sBaJjKy4uDpPJlG5aeHg4Bw8evOt1mEwm4uLi8rppIiLihNasWaN7goiIiIiIEygwNbb+afTo0Y5ugoiIOKno6GhKlChB5cqVHd0UERHJJ6xWa7qXj4iISP5Q4Dq2VqxYQXR0NAaDgf79+9OyZUt8fX156qmn+PHHHylfvjyRkZFYLBZGjhzJ8ePH8fb2Jjk5Ocv1Lti74I5pKigvIpI/xcXFERoaSq1atTCbzbRs2ZIePXowZswY0tLSSE1NZebMmdSsWZNPPvmE5cuXU7hwYerVq0ePHj3YtGkTsbGxPPTQQ6xcufLODURE3DlNBeVFRPKlKVOmsGPHDm7evEmvXr1YsmQJ8+fPp0KFCpw9e5b+/fsTExPD6tWrmTVrFgaDgZo1azJ9+nTi4+Nt95Pk5GQWL16c4TbqRpa4z3t191TYXkQKugLVsfXuu+9y4cIF5s+fz7Bhw2zTL1y4wNChQ6lYsSImk4nDhw9z8uRJUlJSWL9+PZcuXaJ+/foObLmIiOS1uLg4Vq1aRfHixenYsSPBwcF89dVXuLi4sH79esLDw/nkk09YvHgxK1eupEyZMlgsFoxGI/7+/nTv3p2WLVs6ejdEROQebN68mbNnzxITE0NaWhrt27fHZDKxbNkyxowZw9KlS+natSuJiYnMmDGDdevW4eHhwfjx41m7di2PPvoo8fHxrFq1Ck9PT0fvjoiIZKDAdGyZzWbOnz/Pli1bcHFxSTevXLlyVKxYEQAvLy8SEhIwm800btwYgNKlS1O9evVcbVPujTK0H2VrP8rWPvL6bS4+Pj62LyENGzZk586dREREkJiYSEpKCoUKFQJu1WR8/fXXSUlJISQkhI4dO+Zqezou8oZytB9laz/KNu/l1T3h0KFD/PDDD7Z6vMnJyTRs2JCJEycyevRoVq5cydq1a/nll184c+YMTz/9NABXr16lWrVqPProo9SsWdOpO7UcfXw6evvOQBllTflk70HIKKv7QoHp2PLx8WHkyJF069aNRYsWpZv3z2fhrVYrPj4+rFmzhn79+nH58mWOHz+eq21K7um1pPajbO1H2ToPs9lMUlISxYsXZ9++fRQtWpROnTrRtWtX1q1bx8yZMwHw9fXlww8/5Pr169SrV4+OHTvi7u5OWlpajran4+Le6fyyH2VrP8o2f/P19aVVq1ZMnz4dgNTUVFxdXalVqxYRERHUrl2b4sWL4+3tTZUqVVi5ciXu7u62Zc+ePYvR6Nzv23Lk8anzI3vKKGvKJ3vKqAB1bAG0bduWIkWK0K1bN4oWLZrlskFBQaxbt462bdtSpUoVvL29s1xe9bRERJxL5cqVCQsL45dffqFVq1YEBAQQFhbGypUr0938X3jhBS5dukRKSgoDBgwAbt0j3nnnHRYvXsy8efPuXLnqaYmIOIWAgAD2799PUFAQLi4uFCpUiKioKHr27Ennzp356quvAChZsiSjRo2ic+fOGAwGjEYjkydPpkSJu6udpTpWIiKOY0hMTLQ6uhHyYFLPsv0oW/tRts4hLi6OIUOGEB0d7eimSA7o/LIfZWs/ylYkczo/sqeMsqZ8sqeMwLnH1YqIiIiIiIiIyAPLKTq24uLibAUfRUREslO5cmXbaK3t27czePBgB7dIRETup5x8f5g6dSpRUVF2bpGIiNhLgaqxZU8L9i64Y5rqbomIPKAiIu6cprpbIiIPrLqRd1eLyxFU/0tECjqnGLEFt165O2LECPz8/Hj55Zcxm8106NCB4OBgAgMDOXbsGHDrF5fBgwfTpUsXWrZsyc6dOwEYPHgwo0eP5tlnn8Xf35+ff/6ZX3/91fZKX4Bp06axcOFCh+yfiIjkjePHj9O2bVtCQkIICgri8uXLtnn79u2jXbt2BAUFMWDAAK5fv86KFSt49913AejatSsvv/wycKuo/I8//uiQfRARkXuXnJzMoEGD8Pf35/XXX8/0+wPAjh070n1/SEpKolWrVrb5y5cvZ/LkyY7YDRERyYbTjNg6c+YMq1evpkiRIjz++OOMGzeOr776ChcXF9avX094eDiffPIJAAaDgeXLl/Prr7/Sr18/tm7dCkC5cuUIDw9n69atTJo0iSVLlmC1Wjl58iReXl6sXLmSDRs23HWbzGazXfb1QaIM7UfZ2o+ytY+8Knq5YcMGOnTowPDhwwHYtm2bbd7YsWOZO3cuPj4+vPnmm3z22Wd07tyZvn37MmbMGFJSUjh69CgAsbGxPPbYY3e9XR0XeUM52o+ytR9lm/fy4p4QFxfHqlWrKF68OB07diQ4ODhH3x/q1avHtm3baNWqFYsWLeKjjz665zY5gqOPT0dv3xkoo6wpn+w9CBlldV9wmo4tHx8fihUrBkDFihWJjY1l1qxZJCYmkpKSQqFChWzLNmrUCIBHHnmEhISEO6Y3btyYsLAwAPr06cNnn31G8+bNadKkCUWKFMlRmyT39PYG+1G29qNs87+ePXsSHh7OwIEDqVChAi1atLDNu3z5su3fr0mTJqxbt46yZcty9epVtm3bRsOGDYmLi2Pbtm088sgjGI13P7BZx8W90/llP8rWfpRt/uXj44OnpycADRs2ZOfOnURERNz194f+/fszc+ZMypcvT+HChalcufJ934e84MjjU+dH9pRR1pRP9pSREz2KaDAY0v39o48+omPHjqxdu5YXX3wRq9Vqm7d3714ATpw4QcmSJe+YvmfPHqpVqwaAyWRi06ZNzJ07l969e9t7N0RExM7c3Nx4/fXXmTNnDhcvXuTatWu2eSVLlrT9orV7926qV68OQNOmTZk6dSpPPvkkrVu35s0336RNmzaOaL6IiOQRs9lMUlISVquVffv28f333+fo+8Njjz3GmTNneO+99wgNDXXIPoiISPacZsTWPw0dOpSwsDBWrlx5R+/kzZs3ee655/j999955513bNN///13nnnmGRISEpg1axYArq6utGvXjk2bNlGvXr1Mt6dC8SIizuGLL75g6dKlGI1GPDw80n1xee+99xg6dCguLi6UL1/eNnrXz8+PxYsX07BhQ7y8vBgyZAgff/xx5htRoXgRkXyvcuXKhIWF8csvv9CqVSsCAgJy/P2hR48eTJkyxfbdITMq0C4i4jhO0bH199e2A7Y/79q1K8PlW7VqRY8ePe6Y/p///IeWLVveMd3V1ZW+fdVxJSJSEISGht7xy3rHjh2BW4+afPPNN3d8pm3btpw6dQqAChUqkJiYaPd2ioiI/VSuXNlWZ/fvMvr+MHHixEzX4+rqSq9evXB1dYqvTSIiD6QH/go9dOhQzp49y/Llyx3dFBERERERySfeeOMNvvvuO5YuXeropoiISBacpsbW3UpMTGTRokWUKlUq3fTZs2dnOFrro48+YuXKlbi7u9+vJoqISC6cP3+eCRMmOLoZIiJSwMTFxWEymdJNO3jwIMWKFWP9+vXs3LmTuLg4B7VORESyU+BGbMXExHDw4ME8X++CvQvumKa6WyIi90+5cuV4++23Hd2MWyIi7pymulsiIgVG3bp1qVu3LnCrDEqJEiWyfCti3cgS96tpOaLaXyLyIChQHVujRo3i/PnzmEwmfvrpJ06fPs327duJiIigVKlSHDlyhO7duzN06FCuXLnC6NGj+eOPP0hNTWXKlCk0aNDA0bsgIiKZiIuLY8iQIbRo0YIzZ86QkJDAr7/+yvTp02nevDmTJk1i+/bteHh40LZtW0aOHMkbb7zBnj17uHLlCv7+/rz22msA+Pr68tRTT/Hjjz9Svnx5IiMjAZgyZQo7duzg5s2b9OrVS2/BEhF5QFy9epURI0YQGxtLs2bNaN++PUuWLGHYsGFs2rSJ2NhYHnroIVauXOnopoqIyD8UqI6tiIgINm3aRHR0NHXq1LFN//3331m+fDk3btygcePGDB06lIiICAICAujWrRvnzp2jZ8+ebNy4MUfbu/3KeMk9ZWg/ytZ+lK19/PMNVVnx8PAgKiqK7du38/HHH9O8eXP+7//+jx9++AEPDw8sFgsAY8aMoWjRolgsFtq2bcuJEyfw9vbmwoULDB06lIoVK2IymTh8+DDnz5/n7NmzxMTEkJaWRvv27QkODqZMmTJ31SYdF3lDOdqPsrUfZZv3cnJPyAtnzpxh9erVFClShMcff5zmzZsDULt2bfz9/enevXuGZU3yu/xwbOaHNuR3yihryid7D0JGWd0XClTHVmbq1q2Lq6srrq6uGAwGAA4dOsS2bdtYvHgxAElJOR+me79vuAWN2WxWhnaibO1H2eYP9evXB8DLy4uEhAQApk+fzsiRI7FarfTq1YsWLVoQGRlJdHQ0BoOB+Ph4zpw5g7e3N+XKlaNixYrp1nHo0CF++OEHW52V5ORkTp8+fdcdWzou7p3OL/tRtvajbAsGHx8fihUrBkDFihULzNtxHX1s6vzInjLKmvLJnjJ6QDq2bndm/Z2vry/16tWjc+fOAKSkpGS5DtXTEhHJP/5+XbdarVitVpo3b05AQACnT5+mS5cuREdHExkZyffff4/BYCAwMBCr1XrH52+vw9fXl1atWjF9+nQAUlNTM3+9u+ppiYgUKBndF25zd3cnLS0ty8+rlpWIiOM8EB1bGRk7dixjx45l/vz5WK1W6taty1tvveXoZomISC7cvHmTZ555BqvVyo0bN3jhhRcoUaIEdevWJTAwkKpVq9p+ic9MQEAA+/fvJygoCBcXFwoVKkRUVBQeHh73aS9ERCQ/CgoK4p133mHx4sXMmzfP0c0REZF/MCQmJlqzX0wk72nIpP0oW/tRtiL2o/PLfpSt/Shbkczp/MieMsqa8smeMgKjoxuQH8yaNcvRTRARkTw0fvx42rdvz7Jly+6YFxcXZ6ujJSIiBZeu9yIiD4YC9yii1WrNsKZWVmbPns2QIUPs1CIREbnfYmJiOHjwoKObISIiIiIiduZ0HVsWi4URI0ZgNpupWrUqR44cYeHChYSGhlKrVi2Sk5N59913GTNmDFevXgXg/fffp2rVqrzxxhvs2bOHK1eu4O/vz2uvvcYHH3zAH3/8gclkIigoiGHDhmW43QV7F9wxTQXlRUTyn1GjRnH+/HlMJpPth46kpCRCQ0N5/vnn0y27YsUKoqOjmT17NsuWLWPFihUAPPnkk4wbNy7zjUREZLThPNoDERHJa7ev9y1btuSLL74A/netv3nzJmFhYZjNZm7cuMGLL75Iu3btmDp1KvHx8Vy+fJmzZ8/yzjvv0KxZswzXXzeyxP3cnWypmL2IPEicrmNr7dq1pKWlsX79ehITE6lXrx4A8fHxrFq1Ck9PTwYMGMDw4cNp3rw5P/30E6+++ipRUVGMGTOGokWLYrFYaNu2LSdOnGDEiBHMnTuX6OhoB++ZiIjkhYiICDZt2kR0dDTJyckULVqU69ev06RJE3r16mVb7t133+XChQvMnz+fX3/9lTVr1rB27VqMRiPdu3cnNjaWOnXqOHBPREQkL9y+3k+YMIGXX375jmv9vn37KF++POHh4Vy9epXAwEACAwOBW29LXL58Ob/++iv9+vVj69atDt4bERH5J6fr2Prll19o1KgRAJ6enrYiaTVr1sTT0xOAQ4cOpXvD4e1f7CMjI4mOjsZgMBAfH8+ZM2fw9vbOdVvMZnOuPyu3KEP7Ubb2o2ztI6+LXlosFqZPn87u3bsxGo1cvnyZP/74A7j1b3j+/Hm2bNmCi4sLR44cwWw206FDB+DWCK/4+PgcdWzpuMgbytF+lK39KNu8l1f3hL9f7zdu3Jjhtf7QoUPs3r2bbdu2AZCamsrFixcBbN87HnnkERISEvKkTfdDfjom81Nb8itllDXlk70HIaOs7gtO17Hl4+PDmjVr6NevH4mJibZ/QKPxf3XwfX19GTp0KA0aNAAgJSWFxMREIiMj+f777zEYDAQGBmK13nohZE5rcv29LZJ7enuD/Shb+1G2ziM2NpYff/yRmJgYrl27RqNGjWzXfR8fH0aOHEm3bt1YtGgRNWvWxNfXl2XLlmEwGLBYLFgslhxtT8fFvdP5ZT/K1n6Ubf729+v9yy+/nOG1/vz58zz88MOMHDkSuPXdwd3dHYC9e/fSt29fTpw4QcmSJR25KzmSX45JnR/ZU0ZZUz7ZU0ZO2LEVFBRETEwM7dq1o3LlylSqVIlz586lW2bKlCmEhYWRlHTr2fKAgABGjBhB3bp1CQwMpGrVqhQrVsy2fMuWLenWrRvt2rWjT58+GW5X9bRERJyLj48PBoOB9u3bU61aNUqVKpVuftu2bSlSpAjdunVj4cKFmEwmgoODcXFxwc3NjY8++ogKFSpkvHLV0xIRcRq3r/eTJk2iffv2d1zre/fuzYQJEwgJCcFgMFCuXDnmzZsHwM2bN3nuuef4/fffeeeddzLdhmpaiYg4jiExMdHq6EbkVGpqKm5ubly6dAk/Pz8OHDiQbsSWiIiIiIjIvZg6dSpeXl706NHD0U0REZEsON2ILYABAwZw4cIF/vzzT9588011aomIiIiIiIiIPICccsSWiIiIiIiIiIiIhjqJiIiIiIiIiIhTUseWiIiIiIiIiIg4JXVsZWHp0qUEBgYSEBDAt99+6+jmOKVy5cphMpkwmUy89957AERERNC2bVvat2/PwYMHAbBarUyYMIGgoCA6duxIfHy8I5udbwUFBfHII48wderUdNNzkumNGzcYOHAgQUFBdO3alcTExPu9G/lSRtnGxcXh5eVlO4YXLlwIKNucOnDgAO3btycoKIh27dqxd+9ewPmOW90T8obuC3lH9wT70T3BfgrKPQF0X7hN1/U76fqcPV1ns1aQrpX3gzq2MpGYmMjMmTNZvXo1K1as4KWXXiItLc3RzXI6ZcuWJTo6mujoaF588UXMZjPffPMN69ev58MPPyQsLAyAzZs3c/nyZWJiYhg9ejSTJk1ycMvzp7lz5/LGG2+km5bTTJcsWYKXlxcxMTEEBQXx/vvv3/f9yI8yyhagTp06tmM4NDQUULY5Vb58eZYvX05MTAzh4eFMnDjR6Y5b3RPyju4LeUf3BPvRPcF+CsI9AXRf+Dtd1++k63P2dJ3NWkG5Vt4v6tjKxL59+2jWrBmFChWiVKlSVKpUiRMnTji6WU7n0qVLmEwmnnvuOQ4fPsyOHTto3749BoOBatWqkZSUxI0bN9ixYwfBwcEAtGnThv379zu45flTpUqV7piW00z/Pj04OJjvvvvu/u1APpZRtgBHjx4lKCiI3r172379ULY5U758eUqUKAGAu7s7rq6uTnfc6p6Qd3RfyDu6J9iP7gn2UxDuCaD7wt/pun4nXZ+zp+ts1grKtfJ+cXV0A/Kry5cv4+npafu7p6cnly9fdlyDnNTBgwcpU6YMBw4coF+/fjz33HM89NBDtvklSpQgISEhXd4Gg+GB/cUrNy5fvpyjTP8+Xcd11sqXL8+BAwcoXrw433zzDUOHDmX16tXKNpdSUlIYO3YsY8aM4eDBg0513OqekHd0X7Av3RPsR/eEvOXM94R/tsGR7cgPdF2/O7o+Z0/X2Ts5+7XyftGIrUyUKlUq3TOoSUlJlCpVynENclJlypQB4LHHHuNf//oXLi4u6XK9cuUKJUuWTJe31WrF1VV9rnfrn8dqdpn+fbqO66x5eHhQvHhxANq2bcvJkycBlG0u3Lx5k/79+/P0008TGBjodMet7gl5R/cF+3K2c8uZ6J6Qd5z9nvDPNjiyHfmBrut3xxmP8/tN19n0CsK18n5Rx1YmGjRowK5du7hx4wYJCQmcPn0ab29vRzfLqSQnJ9t6i0+fPs3ly5cxmUysX78eq9XKr7/+SokSJfDw8KBFixbExMQAsG3bNurXr+/IpjuVFi1a5CjTv09fu3YtLVq0cFjb87ukpCSsVisAP/30k+0/3JRtzlgsFgYNGsQTTzxB7969Aec7bnVPyBu6L9ifs51bzkT3hLxREO4JoPvCbbqu3z1nPM7vN11n/6egXCvvF0NiYqLV0Y3Ir5YuXcr8+fMBeOmll/Dz83Nwi5zLvn37GDVqFEWLFuXmzZu88sortG7dmoiICGJiYjAYDLz77rvUrVvX9iaHgwcP4ubmxocffoiXl5ejdyHfGThwIAcPHuTatWtUqVKFVatWAeQo0xs3bjB8+HBOnTpFiRIl+Pjjj9MNpX9QZZTtmjVrePfddylatCgAU6dOpV69eso2h7788kuGDRvG448/DkDx4sVZunSp0x23uifcO90X8pbuCfaje4L9FJR7Aui+ALquZ0bX5+zpOpu1gnStvB/UsSUiIiIiIiIiIk5JjyKKiIiIiIiIiIhTUseWiIiIiIiIiIg4JXVsiYiIiIiIiIiIU1LHloiIiIiIiIiIOCV1bImIiIgUUAcPHmTq1KmcOnXK0U0RERERsQt1bImIiIgUULGxsbzzzjucPn3a0U0RERERsQt1bImIiIiIiIiIiFNSx5aIiIjIPfjjjz948cUXefTRRylbtiy+vr4MGDCAs2fPApCQkMDYsWOpWbMmZcuWpVGjRsycOROLxZJuPZ6enkydOvWO9ZtMJkwmk+3v27dvx9PTk2XLljFz5kweffRRypUrR0BAAAcOHLAtN3XqVIYOHQpAUFAQnp6eeHp6sn37djukICIiIuIYro5ugIiIiIiz+uOPP/D39+fcuXOEhoby6KOPcvHiRb755ht+++03SpcuTYcOHTh69Ch9+/alRo0abNy4kVdffZX4+HimTZuW621//PHHpKSk8MILL5CWlsYHH3xAz5492b9/P25ubnTo0IHz588TGRlJWFgY1apVA6BGjRp5tfsiIiIiDqeOLREREZFcmjRpEqdOneLrr7+mVatWtulhYWFYrVbmzp3LoUOHCA8Pp2/fvgA8//zz9OnTh7lz59KvXz9q1aqVq20nJCSwe/duChcuDEC1atXo2bMnmzdvpl27djz66KM0atSIyMhInnzySZo2bXrvOywiIiKSz+hRRBEREZFcsFgsrF69Gj8/v3SdWrcZDAbWr1+Pp6cnPXv2TDdvxIgRAHzzzTe53n737t1tnVoALVq0AODkyZO5XqeIiIiIs1HHloiIiEguXLx4kStXruDr65vpMvHx8VStWhU3N7d0028/DhgXF5fr7T/88MPp/u7p6QncGsklIiIi8qBQx5aIiIhILlitVuDWyKzcupvP/rPI/G0uLi5ZtktERETkQaCOLREREZFceOihhyhevDiHDx/OdBkvLy9OnDjBzZs3000/fvy4bf5tnp6eJCUl3bGO+Pj4XLfxXjrdRERERJyBOrZEREREcsFoNNKhQwe+/fZbtm3bdsd8q9VK+/btSUhIICoqKt28mTNnAtCuXTvbtKpVq7Jjx450y0VHR3PmzJlct7FIkSIAGXaYiYiIiBQEeiuiiIiISC699tprbNmyhf/85z+Ehoby6KOPkpCQwIYNG3j55ZcJDQ3ls88+Y8yYMRw6dIgaNWqwYcMG1q9fz/PPP0/NmjVt6+rTpw8jRoyge/fuBAYGcvz4cb788ku8vb1z3b569ephMBgIDw/n8uXLeHh40KpVKx566KG82H0RERERh1PHloiIiEgulStXjk2bNjF16lSio6NZuHAhZcuWpXnz5jzyyCN4eHiwevVq3nzzTb7++msuX76Ml5cXb7zxBsOGDUu3rp49exIfH8/ChQvZvHkzjz/+OF988QUTJ07MdfuqVKnCO++8w4cffsjw4cNJS0tj9erV6tgSERGRAsOQmJioCqMiIiIiIiIiIuJ0VGNLRERERERERESckjq2RERERERERETEKaljS0REREREREREnJI6tkRERERERERExCmpY0tERERERERERJySOrZERERERERERMQpqWNLRERERERERESckjq2RERERERERETEKaljS0REREREREREnJI6tkRERERERERExCmpY0tERERERERERJySOrZERERERERERMQpqWNLRERERERERESckjq2RERERERERETEKaljS0REREREREREnJI6tkRERERERERExCmpY0tERERERERERJySOrZERERERERERMQpqWNLRERERERERESckjq2RERERERERETEKaljS0REREREREREnJJrVjOvXLnCokWL7piempqKm5ub7f//OQ+4Y7rmaZ7maZ7maZ7maZ7maZ7maZ7maZ7maZ7mad7dzAPo1asXxYsXz3DebYbExERrZjMtFgtXrlzJcgUiIiIiIiIiIiJ5rXjx4hiNWT9smGXHloiIiIiIiIiISH6V4xpbZ8+eZd++fQC2/weYM2fOXX3+bpe7F19++SUpKSn3tI6/72d2rl27xuHDhzOc9/nnn5OQkHBPbRERERERERGR3Nu4cSO//fbbXS+/evXqHG/jxx9/ZNasWXzyySd88803Of78vUpJSeH//u//mDNnDh9//DGbNm3K0edzs8/5QZY1tjJSoUIFKlSoANzq2GrQoEGeN+pepaSk4O7unuvPWyyWdPuZnevXr3PkyBFq1659x3pERERERERExHlYLBY6dOiQ48/t3r2bQYMGYTQauXbtWq63nd2jd5nZuHEj1apV4+mnnwbg5MmTOfp8bvY5P8iyY+vq1assWbIEV1dXXFxcaNGiBQC//fYbJUqU4Pfff2fOnDl07twZi8XCqlWriIuLo127dtSoUYM5c+YwcOBAEhIS2LhxI88++ywWi4Uvv/yS33//ncDAQGrUqGHb3p9//snnn3/OzZs3eeSRR2jTpg0rVqzgzz//pFixYjz33HMcOHCA48ePc+3aNUqUKEGJEiU4duwYTzzxBE888QTnz5+nbNmyJCQk8OWXX1KoUCGuXLlCkyZN2LdvH56enjz77LMcO3aMbdu2cePGDfz9/fH19eXzzz/Hw8ODy5cv06pVK3777TcCAgL4+uuv+f333/Hw8KBr166cP3+eNWvW4ObmRv369UlISOD48ePMmTOH0NBQ5syZQ9myZalYsSIA27Zt49y5czzyyCMEBASwZ88e9u/fT0pKCh07dsTLy4vDhw+zdetW3NzcaN++PQ8//LAd/9lFREREREREHkzx8fEcPXqUdu3akZiYyDfffEPDhg3ZsWMHFouFFi1asHnzZgYOHMjJkydZt24dLi4uNG/eHKvVyrfffou7uzutWrWiZs2atvWmpKRw+vRpHn74YQoXLgzA/v372bVrF0ajkU6dOlGoUCFb/8hvv/1m63f44IMPbP0IpUqVStc/4OHhwddff83Nmzdp2LAhDRs2ZMuWLTRo0IBixYrZtn/69GmCg4Ntf69SpQpwq4Nr/fr1WCwWnnzySWrUqMHHH3/Mv//97wz7cG7/P2D78/r16zlx4gQGg4Fu3bqlK+h+e56rqyt9+vQhNjY23T7/+9//ZubMmZQtW5Zz587h5+fHnj17MBgM9O3bF4PBcE//nll2bO3du5cmTZpQp04dIiMj081r1KgR+/fvt+3sX3/9RUBAADdv3uTrr79O12H1d8nJyXTr1o1ChQoRGRmZbrktW7bQunVrHnnkESwWC4cPH6Z8+fJ069aNzZs3c+TIEQBKlChBt27dmD9/PrVr16ZNmzZ8+umnPPHEE/zyyy/4+PgAcOPGDfr378/WrVs5f/48zz//PPPnz+fatWtUrVqVmjVrcuPGDRYtWoSvry8A3t7edOzY0TZE8dSpU9y8eZMXXniB/fv388MPP3D9+nXatWtHgBVPxgAABolJREFUtWrVsFgsJCUlkZSUxLPPPgvcepvkoEGDcHd35/PPP8fb25tOnTqxYMEC/vzzT+rVq0ejRo24fPky69evp1KlSmzdupXnn38eNzc3jfQSERERERERsRMvLy82bNgAQGxsLHXq1AFujZbq06cPAJs3bwbgm2++oVevXhQpUgSLxcLnn39O165dKVOmzB3f3Tt37szmzZu5ePEi7dq1o3bt2uzatYsXXniBK1eusHr16kxHRd3uR3B1deXjjz9O1z+wbNkynnvuOYoVK8aCBQuoX78+bdq0uWMdtzuIUlNTWbBgAcnJyYwZM4bNmzfTt29fXFxc+Oyzz6hRo8Zd9+HcduLECQYOHIjRaMRq/V+p9rNnz5KUlMSgQYOwWq1YrdY79jk0NJTk5GQGDx7MoUOHOHToEP3792flypWcO3eOf//739n/o2Uhy46ty5cv2zp8ypcvn+WKihYtSpEiRYBbj+b93d93ukiRIpQoUQLgjuF1Fy9exMvLyzbv0qVLtlFPFStW5Ny5cxQtWpSyZcsCUKxYMcqWLYurq6vtH/D06dM0bdqUpKQkHnroIQwGA8WKFaNo0aK2z1y/fp2EhAQ2bdqE1WolOTnZ1obb2/t7Bn9vw86dO3nyySfZvHkz+/bto3nz5rb9vu2hhx5K9yjk7X+kf//73yQkJJCUlMTOnTsxGAwYDAauXr1KyZIlba+3zO2wQxERERERERHJXqVKlThz5gxms5mmTZsSHx9/R38A3Oosuv2d32g08uSTT7J161bb6KfSpUvblq1cuTJ9+vTh2rVrzJs3jypVqlCyZElcXFwoWbIkN27cSLfuv/eV3O5H+PPPP+/oH7h48SLLli0Dbg0q+uuvv9KN1PonNzc328grgHPnztkGK/31119A1n04GY2g8vPz4/PPP+df//oX7dq1s/V5XLx4kcqVK9s+l5ycnOE+ly5dGldXV4oXL85DDz0E3Oqfye0jm3+XZQ9KqVKlOHfuHIDt//8us+Fit/9xUlNT7/js1atXuXLlCikpKXf0bpYpU4ZTp04Bt3pKS5UqxZkzZwA4c+YMpUqVumO7f/9zWloaBoPB1jGU2XJw6/HA5557jtDQ0CyXy6gNhQsX5qmnniIoKIhNmzZhNBrT7cs/1/H3DEuWLMn27dvp378///nPf7BarRQpUoSEhARu3rxp23cRERERERERsY/HHnuMLVu2UKJECVxdb435yayP43bni8ViwdPTk//85z888cQTfPfdd+mWu3TpEgDu7u64uLjYvuunpaWRkJCAh4eHrVwSpO8rub3tjPoHypQpQ/fu3Rk4cCDDhw/PtFOrYsWKxMbG2v5+u2/h3//+N3379rV9/p/+3sEG2NqYnJzMn3/+CUDVqlXp0qULRYsW5fjx47Zly5QpQ3x8vG09Ge3zP93ro4f/lOWIrYYNGxIVFWV79vGfHTjFihUjKiqKoKCgDD9fvXp1Pv74YypVqmSbVrRoUb755htbjS3ANhyvTZs2fP7552zcuNFWY2v58uXMmTOHokWL0rp1aw4cOJBpe0+dOpVuW1nx9fXls88+S/fsa0YefvhhfvzxRz755BPc3d3p2rUr33//PYcPHyYlJYU2bdpQrFgxrl69SlRUFM8888wd6/j111/57rvveOSRRyhWrBg+Pj588sknPPLII8CtHthWrVrZtqEaWyIiIiIiIiJ5Kzo62vb9v3fv3ly8eDFdTaqMBAYGEhkZiaurK82bNycuLo74+HhSU1MxmUzpll29ejXXr1/HarXStGlTjEYjjRs35pNPPsFgMNCpUycKFy5MiRIlmDdvHqVKlUpXqwoy7h8ICAhg+fLlWCwWChcuTM+ePTOssRUQEMCaNWts9a28vb0BePLJJ1mwYAEGg4GyZcvSqVOnLPe5UaNGREZG4uXlZRvVtWjRIlJTUzEYDDRq1Iiff/4ZgBo1alC8eHE+/vhjXF1d6d279x37bG+GxMREa2Yzb3diGY1GIiMj6dy5s+0xwvwoPj7eVlBeRERERERERCQz8+fPp0+fPioH5OSyHLF148YNFi5cyM2bN6lWrVq+7zC6XZ9LRERERERERCQzn376KTVq1FCnVgGQ5YgtERERERERERGR/EpdkyIiIiIiIiIi4pTUsSUiIiIiIiIiIk5JHVsiIiIiIiIiIuKU1LElIiIiIiIiIiJOSR1bIiIiIiIiIiLilNSxJSIiIiIiIiIiTun/AUk5PrYiEWdjAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1296x432 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "style.use('fivethirtyeight')\n",
    "\n",
    "# define sentiment colors\n",
    "POSITIVE_COLOR = 'darkseagreen'\n",
    "NEGATIVE_COLOR = 'lightcoral'\n",
    "NEUTRAL_COLOR = 'dodgerblue'\n",
    "\n",
    "# words per subplot \n",
    "NUM_OF_WORDS = 40\n",
    "\n",
    "fig = plt.figure(figsize=(18,6))\n",
    "\n",
    "# positive subplot\n",
    "plt.subplot(131)\n",
    "positive_word_counts = word_counts_df[word_counts_df['polarity'] > 0][:NUM_OF_WORDS]\n",
    "positive_word_counts.plot(x='word',\n",
    "                          y='counts',\n",
    "                          kind='barh',\n",
    "                          title='Positive',\n",
    "                          legend=False,\n",
    "                          fontsize=9.3,\n",
    "                          color=POSITIVE_COLOR,\n",
    "                          xticks=[0, 500, 1000, 1500, 2000],\n",
    "                          ax=plt.gca()).invert_yaxis()\n",
    "\n",
    "\n",
    "# negative subplot\n",
    "plt.subplot(132)\n",
    "negative_word_counts = word_counts_df[word_counts_df['polarity'] < 0][:NUM_OF_WORDS]\n",
    "negative_word_counts.plot(x='word',\n",
    "                          y='counts',\n",
    "                          kind='barh',\n",
    "                          title='Negative',\n",
    "                          xlabel='',\n",
    "                          legend=False,\n",
    "                          fontsize=9.3,\n",
    "                          color=NEGATIVE_COLOR,\n",
    "                          xticks=[0, 500, 1000, 1500, 2000],\n",
    "                          ax=plt.gca()).invert_yaxis()\n",
    "\n",
    "\n",
    "# neutral subplot\n",
    "plt.subplot(133)\n",
    "neutral_word_counts = word_counts_df[word_counts_df['polarity'] == 0][:NUM_OF_WORDS]\n",
    "neutral_word_counts.plot(x='word',\n",
    "                         y='counts',\n",
    "                         kind='barh',\n",
    "                         title='Neutral',\n",
    "                         xlabel='',\n",
    "                         legend=False,\n",
    "                         fontsize=9.3, \n",
    "                         color=NEUTRAL_COLOR,\n",
    "                         ax=plt.gca()).invert_yaxis()\n",
    "\n",
    "# title\n",
    "fig.text(0.5, 0.95, 'Top 40 Most Frequent Words for Each Sentiment [Eminem]', ha='center', fontsize='x-large', fontweight=555)\n",
    "\n",
    "# x-label\n",
    "fig.text(0.5, 0.004, 'count', ha='center', fontsize=17)\n",
    "\n",
    "# signature bar\n",
    "fig.text(x=0.03, y=-0.045, s=str('_'*170).expandtabs(), color='grey')\n",
    "fig.text(x=0.03, y=-0.07, \n",
    "         s=str(\"github.com/mariostrbac\" + \"  \" * 196 + \"  Lyrics Source: Genius.com\").expandtabs(),\n",
    "         fontsize=7.9,\n",
    "         color='dimgrey')\n",
    "\n",
    "plt.subplots_adjust(wspace=0.117, hspace=0)\n",
    "plt.savefig('../reports/figures/vocabulary-analysis/40_most_frequent_words_sentiment_eminem.png', dpi=200, bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "CYeIWxAuKIdE"
   },
   "source": [
    "### 3.4.3 The most frequent words including sentiment annotation <a name=\"343-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "id": "1Sw4Ulp2KIdF"
   },
   "outputs": [],
   "source": [
    "# for colors encoding\n",
    "COLORS_DICT = {-1: NEGATIVE_COLOR, 0: NEUTRAL_COLOR, 1: POSITIVE_COLOR}\n",
    "\n",
    "# returns a sentiment color for a given word based on the word's polarity value\n",
    "get_sentiment_color = lambda polarity: COLORS_DICT[polarity / np.abs(polarity) if polarity !=0 else 0]\n",
    "\n",
    "# assign a color to each word according to their polarity value\n",
    "words_sentiment_colors = word_counts_df.polarity.apply(get_sentiment_color)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1224x504 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "style.use('fivethirtyeight')\n",
    "\n",
    "# words per subplot\n",
    "NUM_OF_WORDS = 50\n",
    "\n",
    "fig, axs = plt.subplots(1, 2, figsize=(17,7), sharex=True)\n",
    "\n",
    "# left subplot [0-50]\n",
    "word_counts_df[:NUM_OF_WORDS].plot(x='word',\n",
    "                                   y='counts',\n",
    "                                   kind='barh',\n",
    "                                   title='Most frequent [1-50]',\n",
    "                                   xlabel='',\n",
    "                                   fontsize=9.3, \n",
    "                                   legend=False,\n",
    "                                   color=words_sentiment_colors[:NUM_OF_WORDS],\n",
    "                                   ax=axs[0]).invert_yaxis()\n",
    "\n",
    "# right subplot [50-100]\n",
    "word_counts_df[NUM_OF_WORDS:2*NUM_OF_WORDS].plot(x='word',\n",
    "                                                 y='counts',\n",
    "                                                 kind='barh',\n",
    "                                                 title='Most frequent [51-100]',\n",
    "                                                 xlabel='',\n",
    "                                                 fontsize=9.3, \n",
    "                                                 color=words_sentiment_colors[NUM_OF_WORDS: 2*NUM_OF_WORDS],\n",
    "                                                 ax=axs[1]).invert_yaxis()\n",
    "# legend\n",
    "plt.legend(handles=[Patch(color=NEUTRAL_COLOR, label='neutral'), \n",
    "                    Patch(color=NEGATIVE_COLOR, label='negative'), \n",
    "                    Patch(color=POSITIVE_COLOR, label='positive')],\n",
    "           loc=4, \n",
    "           fancybox=True, \n",
    "           shadow=False, \n",
    "           fontsize='small')\n",
    "# title\n",
    "fig.text(x=0.5, y=0.95, \n",
    "         s='Top 100 Most Frequent Words with Sentiment [Eminem]', ha='center', fontsize='x-large', fontweight=555)\n",
    "\n",
    "# xlabel\n",
    "plt.xlabel(\"count\", x=-0.1, y=0.045, labelpad=12)\n",
    "\n",
    "# signature bar\n",
    "fig.text(x=0.03, y=-0.045, s=str('_'*160).expandtabs(), color='grey')\n",
    "fig.text(x=0.03, y=-0.07, \n",
    "         s=str(\"github.com/mariostrbac \" + \"  \"*182 + \" Lyrics Source: Genius.com\").expandtabs(),\n",
    "         fontsize=7.9,\n",
    "         color='dimgrey')\n",
    "\n",
    "plt.subplots_adjust(wspace=0.13, hspace=0)\n",
    "plt.savefig('../reports/figures/vocabulary-analysis/100_most_frequent_words_sentiment_eminem.png', dpi=200, bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "-D2CMK4XKIdG"
   },
   "source": [
    "## 3.5 Profanity analysis <a name=\"35-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "HzGs6darKIdG"
   },
   "source": [
    "### 3.5.1 The most common profanity words <a name=\"351-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ljxK48joKIdH"
   },
   "source": [
    "After inspecting the previous graph, the most frequent profanity words are written down and stored for the further analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 255
    },
    "executionInfo": {
     "elapsed": 546,
     "status": "error",
     "timestamp": 1615901872856,
     "user": {
      "displayName": "Mario Štrbac",
      "photoUrl": "",
      "userId": "03507842523962903593"
     },
     "user_tz": -60
    },
    "id": "eq1-YzfMKIdH",
    "outputId": "c59395b3-0517-4946-aa25-1398f6ca24aa"
   },
   "outputs": [],
   "source": [
    "PROFANITY_WORDS = ['fuck', 'fucking', 'fucked', 'shit', 'bitch', 'dick']\n",
    "\n",
    "# df object for profanity analysis\n",
    "profanity_df = data_df.loc[:, ['title', 'featured', 'release_date', 'album_name', 'popularity', 'lyrics']]\n",
    "# split lyrics into words and clean dash and contraction words\n",
    "profanity_df['lyrics'] = profanity_df.apply(get_only_eminems_verses, axis=1) \\\n",
    "                                     .apply(split_lyrics_into_words) \\\n",
    "                                     .apply(dash_and_contractions_cleaning)\n",
    "\n",
    "# filter out profanity words in each song to a separate column \n",
    "profanity_df['profanity_words'] = profanity_df['lyrics'].apply(lambda lyrics: [word for word in lyrics if word in PROFANITY_WORDS])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "aq2SBKLgKIdH"
   },
   "source": [
    "### 3.5.2 The relative frequency of profanity words <a name=\"352-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0Ac3P_OkKIdH"
   },
   "source": [
    "The relative frequency refers to the occurrence rate of profanity words in the corpus and is calculated as follows:\n",
    "\n",
    "$$ f_\\text{profanity}\\ (corpus)\\ =\\text{ ... }\\ = \\frac{N_\\text{profanity words}}{N_\\text{all words}}$$\n",
    "\n",
    ", where *N* is the number of words in the corpus."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 238
    },
    "executionInfo": {
     "elapsed": 874,
     "status": "error",
     "timestamp": 1615901873203,
     "user": {
      "displayName": "Mario Štrbac",
      "photoUrl": "",
      "userId": "03507842523962903593"
     },
     "user_tz": -60
    },
    "id": "uJIbAXA1KIdH",
    "outputId": "bb60983a-7118-4983-dd47-c1fb04c57c21"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-the number of profanity words in the corpus: 3678\n",
      "-the relative frequency: 0.03\n",
      "-the ratio of non-profanity words to profanity words is 32.71, i.e. there is a profanity word every 32 words\n"
     ]
    }
   ],
   "source": [
    "# total count of profanity words in the corpus\n",
    "profanity_count = word_counts_df[(word_counts_df.word==\"fuck\") | \n",
    "                                 (word_counts_df.word==\"shit\") | \n",
    "                                 (word_counts_df.word==\"bitch\") | \n",
    "                                 (word_counts_df.word==\"fucking\") | \n",
    "                                 (word_counts_df.word==\"dick\") | \n",
    "                                 (word_counts_df.word==\"fucked\")] [\"counts\"].sum()\n",
    "\n",
    "print(\"-the number of profanity words in the corpus:\", profanity_count)\n",
    "print(\"-the relative frequency:\", np.round(profanity_count / len(all_words), 2))\n",
    "\n",
    "p_ratio = np.round((len(all_words)-profanity_count)/ profanity_count, 2)\n",
    "print(f\"-the ratio of non-profanity words to profanity words is {p_ratio}, i.e. there is a profanity word every {int(p_ratio)} words\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Cph6cEIPKIdJ"
   },
   "source": [
    "### 3.5.3 The average number of profanity words per song  <a name=\"353-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "KgyTABqmKIdJ"
   },
   "source": [
    "#### a) All songs included"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 8.86\n",
      "std: 8.40 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# number of profanity words for each song\n",
    "profanity_cnt_songs = profanity_df['profanity_words'].apply(lambda words: len(words))\n",
    "\n",
    "# show results\n",
    "print('mean: %.2f' % profanity_cnt_songs.mean())\n",
    "print('std: %.2f' % profanity_cnt_songs.std(), '\\n')\n",
    "\n",
    "profanity_cnt_songs.plot.hist(bins=8, rwidth=0.95, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "AmkWNODUKIdK"
   },
   "source": [
    "#### b) Only the songs that have \"release_date\" column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "id": "0sSeKlb5KIdK"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean: 9.55\n",
      "std: 8.61 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# number of profanity words for each song with release date\n",
    "profanity_cnt_songs = profanity_df[profanity_df['release_date'].isnull() == False]['profanity_words'].apply(lambda words: len(words))\n",
    "\n",
    "# show results\n",
    "print('mean: %.2f' % profanity_cnt_songs.mean())\n",
    "print('std: %.2f' % profanity_cnt_songs.std(), '\\n')\n",
    "\n",
    "profanity_cnt_songs.plot.hist(bins=8, rwidth=0.95, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "5y1GAjLPKIdK"
   },
   "source": [
    "### 3.5.4 S-word/F-word graph  <a name=\"354-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vUW3ZU2VKIdL"
   },
   "source": [
    "The S-word/F-word graph shows s-word and f-word ratios for each song. Since those are the most frequent profanity words in the corpus, the outliers on the graph can be seen as songs with the most profanity. The relative frequencies are calculated as follows:\n",
    "\n",
    "$$ f_\\text{s-word}\\ =\\text{ ... }\\ = \\frac{N_\\text{s-words}}{N_\\text{all words}}$$ <br>\n",
    "$$ f_\\text{f-word}\\ =\\text{ ... }\\ = \\frac{N_\\text{f-words}}{N_\\text{all words}}$$\n",
    "\n",
    ", where *N* is the number of words in lyrics. So, for example, if the f-word ratio is 0.10 that means 10% of words in the song are f-words."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "id": "xbmzeC_uKIdL"
   },
   "outputs": [],
   "source": [
    "# calculate the ratios for each song\n",
    "profanity_df['f_ratio'] = profanity_df.apply(lambda row: len([word for word in row.profanity_words if word[:4]=='fuck']) / len(row.lyrics), axis=1)\n",
    "profanity_df['s_ratio'] = profanity_df.apply(lambda row: len([word for word in row.profanity_words if word=='shit']) / len(row.lyrics), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {
    "id": "9EdKNCMtKIdL"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAxgAAAIPCAYAAADw2XjzAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOzdd1gU19fA8e/Se1MQFRAUVOwVhagoxq7YxRaNxo41ajRqEo1GoyYaE0uMJcbYYom9FxR7b1hBBLEgCFKkw+77By8biZgfKrgC5/M8eeLOzM6emYFlztx77lXExMSoEEIIIYQQQog8oKXpAIQQQgghhBCFhyQYQgghhBBCiDwjCYYQQgghhBAiz0iCIYQQQgghhMgzkmAIIYQQQggh8owkGEIIIYQQQog8IwmGEEIIIYQQIs9IgiEKvYyMDE2HIIQQQghRZEiCIQo9bW1tAA4fPsyjR480HI0QQgghROGmo+kAhMgPGRkZ6sTi/v37jBkzBoVCQfny5Zk2bRoGBgYajlAIIYQQonCSFgxRqCxZsoSLFy+ira1NQkICADo6OowfP56tW7dy/vx5Dh48CIBSqdRkqEIIIYQQhZIkGKJQyKqzKFWqFAcPHmTVqlV06dIFPz8/zMzMMDQ0pF27dri6urJ69WoAtLTkx18IIYQQIq/JHZYo0AIDA4F/6iyqVKnC48ePCQgIYNy4cVy6dImHDx9y+vRpfvrpJyZPnoyuri4XL17UZNhCCCGEEIWWJBiiwIqOjmbBggW8ePGC0NBQzp07h52dHcWLF6du3bo0bdoUpVJJQkICCoWCIUOG8M033/D1119Tu3ZtTYcvhBBCCFEoSZG3KHCUSiVaWlpoaWlRq1YtZsyYwe3bt9HW1mby5MlYWlqSkpICZLZohISEMGTIEBo0aEC1atU0HL0QQgghROEmCYYoMFQqFQqFQl07kVVbERcXx+rVq7l8+TInT56kY8eO+Pj4EB8fz5EjR5g8eTJaWlqSXAghhBBCvAfSRUp88G7cuEFiYiIKhUK9bMGCBaxatYqPPvqIpKQkzMzM8PT05PTp05ibm9O2bVsePnzIvHnzqFWrlgajF0IIIYQoWiTBEB+skJAQhg8fzoQJE7h27Rrh4eFs2bIFgE6dOnHgwAFsbW3R1dVl48aNpKen4+TkhKmpKV988QWzZs2iTJkyGj4KIYQQQoiiRRIM8cFavHgxbm5ubNu2jfr161OsWDFWrVrFzZs3sbe3p379+hw+fBhfX1+++eYbBg0aRIUKFYB/RpUS4kNTtWpVLCwsNB1GgTBr1iwsLCxYu3ZtrrYfOnQoFhYW6v/mzp2bzxG+mdDQUCwsLBg6dKimQ3ln8+fPz3au27Rpo+mQhBAfEEkwxAcpIiKCa9eu0adPH3R0dEhLS0NXV5c+ffqwdOlSAMqUKcONGzews7Nj4cKF/PLLL/Tp00fDkYv/RalUsnr1atq2bYuTkxPFixenXLlyuLu7M3ToUHUrVVF1584dLCws+O677wBYu3Ztthu5nP6LiYnRbNAfmCFDhjBhwgQ8PDyyLc9KWP7rvwYNGmgo6oKlfv36TJgwgSFDhmg6FCHEB0iKvMUHSUdHh3r16hEfH4+JiQk6Opk/qt7e3ly4cIHt27dz+vRpypQpg4mJCU2bNtVwxCI3lEolPXr0YP/+/ZiZmdGyZUtKlSpFbGws9+/fZ8eOHVy4cIHOnTtrOlSN2bFjBwDt2rXLtrxKlSqvfUpsYGCQ73EVJEOHDv3P7pEfffTRaxOJEiVK5FdYlCpVinPnzmFmZpZvn/G+uLu74+7uTmhoKL/++qumwxFCfGAkwRAal5GR8UqXpoSEBG7duoW2tra6uDspKYmIiAhat27N4sWLGT9+PHXq1NFEyOItbd68mf3791OlShV2796Nubl5tvXJycmcPn1aQ9F9GHbu3EmZMmVeGfWsatWqfPnllxqKqnBp0KCBRs6lrq4u5cuXf++fK4QQ75t0kRIal5VcPHr0SL3M3t4eR0dHZs+erV62ePFigoKC8PT05K+//pLkogA6e/YsAD179nwluYDMJ/FNmjTJ9f4qVqyY4w1bvXr1sLCwYMKECdmWBwQEYGFh8Uq3jri4OKZPn07dunUpUaIEDg4OtG3blp07d76y76x+9G3atOHx48cMGTKE8uXLY2Vlxa5du4DMIZV/++036tevT4kSJXB1dWXcuHHExsb+5/E8ePCAa9euvdJ6kRemTZuGhYUFBw8ezLZ8wYIFWFhY4OjoiFKpzLauUaNGlChRgqSkpGzLt2/fTps2bXBwcKBEiRK4ubkxY8YM4uPjX/ncNm3aYGFhQUhICL/88ov6nPTs2VO9zZUrV+jcuTN2dnbY29vTvn179c+KpmXVzKSlpTF79mxq1KhBiRIlqFOnDn/88Yd6u2XLluHu7o6trS2VKlVi5syZr5zP19VgZNWOHD9+nO3bt+Pl5UXJkiVxdHSkX79+2b4bXxYbG8t3332Hu7s7JUuWxM7OjpYtW7Jt27ZXtj1+/Lj6s+/fv0+fPn1wcnLCzs6Ojh07cvPmTQAiIyMZPnw4FSpUoESJEjRu3Bh/f/93PItCiKJGEgyhESqVSv3vPXv2UKtWLebOncuqVavUyydPnszNmzeZNWsWbdq04cGDBzIDdwFnZWUFwL179/Jkf40aNSIiIoIbN26ol4WHh3Pnzh0Ajh07lm37rNeenp7qZTExMTRv3pwff/wRQ0NDhgwZQqdOnQgICOCTTz5h5syZOX728+fPad68OdeuXaNDhw58+umnWFpaAjBx4kS++OILnj9/Tp8+fejcuTNHjhyhQ4cOpKWlvfZ4Xtc9Ki80btwYgKNHj2ZbnnVOYmJiuHLlinp5dHQ0AQEBuLm5YWhoqF7+7bff0rdvX27evEmnTp0YMmQIBgYG/PDDD7Ro0YK4uLgcP/+LL77ghx9+oFq1agwePJgaNWoAmUlnq1at8PPzo2nTpgwcOBADAwPatWvHxYsX8+z431X//v1Zs2YNXl5e9OzZk4iICEaNGsWaNWv44osvmDNnDrVr1+aTTz4hIyODOXPm8PPPP7/RZ6xYsYJBgwZRpkwZBgwYgKurK1u3bqV9+/bqyUOzPH78GC8vL+bOnYuFhQWffvopnTt3JjQ0lE8//ZRZs2bl+BkPHjygadOmPH/+nF69euHu7o6fnx9t27YlMDCQpk2bcvPmTTp37kzr1q25evUqXbt2JSws7K3PnRCi6JEuUuK9iY6O5tKlS3z88ccoFApUKhWxsbE8e/aMPXv28OzZM3788Uc6deqEmZkZ5ubmLFu2jMDAQFq2bEnNmjU1fQjiHbVr146ffvqJlStXEhcXR6tWrahRowZOTk7Z5jnJLU9PTzZu3MixY8eoXLky8M8Nc5MmTfDz8yM8PBxbW1sA9ZPYlxOMqVOncvv2bXr16sXChQvVcYwfP159A9eiRYtXktubN2/i4+PDokWL1DVCkHnDvHTpUhwcHPDz86NYsWIAfPXVV7Rv357w8PDXHs+uXbuwtbXFzc3tlXXXr1/P8abx448/pm7duv/zXNWrVw8DA4NsSVdqaipnzpzBy8uLI0eOcOzYMfW8McePH0epVGY7V+fOnWPevHmUKlWKw4cPU7JkSfU5HDp0KBs2bODbb7/lhx9+yDF+f3//bLURKpWK4cOHk5SUxB9//EH79u3V65YtW8b48eP/53G9jRMnTrz2BrxBgwY0bNjwleVPnz7l1KlTmJqaAtCjRw+aN2/OpEmTsLGx4dSpU1hbWwMwfPhw6tSpwy+//MLw4cOz/Xz8l6xrULFiRfWyAQMGsHnzZnbv3k2nTp3Uy4cOHUpwcDDLly+nS5cu6uVxcXG0bduWOXPm0KZNm1e62p08eZLp06czYsQI9bLRo0ezatUqmjZtSq9evZg5c6b696BKlSpMnz6dxYsXv/acCSHEv0kLhnhvLl26xP379wkLC2P69OmsWLECCwsLunXrxrp16xg3bhyGhoacP39e/R4LCwvq1q0ryUUhUa1aNX777TdsbGzYtGkT/fv3p1atWjg6OuLj48O2bduytW79L1k3vy/fNB87dgwTExPGjh2bbV16ejqnTp3CxcWFUqVKAZCWlsbGjRsxMjJi2rRp2ZKc0qVL8/nnn6NSqVi9evUrn62np8eMGTNeuXnMGlJ17Nix6uQCQF9fn6+++uq1xxIREcG5c+do06ZNjslWQEAAs2fPfuW/l39f/ouBgQFubm7cuHGDZ8+eAZkJQ2JiIj179qRcuXLZWjdyau1Zs2YNAJ9//rk6uQBQKBR8++23GBoasn79+hxbaUaMGPFK4fXZs2cJDAykXr162ZILgM8++4yyZcvm6tje1MmTJ3M8l7Nnz+bEiRM5vufrr79WJxcAbm5uODo6EhcXx9ixY9XJBWSOcOfu7k5UVBSPHz/OdVyDBw/OllwA9O3bF8j8/sxy48YNjh07Rps2bbIlFwBmZmZMnDgRlUrFpk2bXvkMR0dHfH19sy3r1q0bkDkIw1dffZXt58/HxwfITBCFECK3pAVD5KuEhASMjY2BzJvL7du3c/fuXRwdHdXdH3x8fIiOjmbfvn189dVXxMbG5lj4LQqHjh070rZtW44fP87p06e5ceMGZ86cYf/+/ezfv5/mzZuzZs0a9PT0iImJYcmSJa/sI6vPup2dHc7Ozpw8eZL09HR0dHTw9/fHw8OD+vXrY2pqytGjR/Hx8eHixYvEx8fTvXt39X7u3r1LYmIiderUoXjx4q98Tla3oqtXr76yzsHBIdtNZZasbT/66KNX1tWvXx8dHR3S09NfWbdr1y6USuVru0f16NEjx3PxsrVr1/LgwYNsy15+Ip/Vn/7YsWN07tyZY8eOoVAo8PT05NSpU6xbt47k5GR1S4eZmZm6RePlY2vUqNErn21jY0OlSpW4ePEigYGBVKpUKdv6nGqm/utcaWlpUb9+fYKDg//zmN/GhAkT3rjI+98tAQC2traEhIRQtWrVHNdBZlcmBweHXH1GVrexl5UuXRog21DEWfUp8fHxObYqREVFAZk/3/9WpUoVtLSyP1vMirVs2bIYGRm99jiEECK3JMEQ+eLFixesW7eONm3aYGxsTGhoKGXKlKFq1aocPnyYuXPnEhkZyaRJk6hduzYPHz6kQ4cOVKtWjSZNmkhyUcjp6uri5eWFl5cXkPnkdMeOHfj6+nLgwAFWrlzJkCFDiI2NzVbon6Vnz57qyeo8PT1ZsWIFFy5cwNramocPHzJ06FB0dHTw8PBQd4vKeiL/8s1xVr2AjY1NjnFmDVmaU13B696TtW1OyYe2tjZWVlZERES8sm7nzp1YWlq+0zwM69at4+TJk68sz0owXm7x6dy5M/7+/lSqVAlra2s8PT1ZuXIlZ8+epVy5cty7d49WrVpl+13M6/P1X+fqvz5HE3IaWjbr3LzcsvHvdf9Vc/Mmn5GRkaFeFh0dDWRex3/XGb0sISHhlWX/FWtO67Ja6N7kOIQQQhIMkeeUSiUmJiY8evSIffv2cefOHQIDA6lSpQrTpk1j//79hIWFYW9vT40aNahQoQLDhw8nPT2d+vXrazp8oQFaWlp06NCBgIAAfvjhB44ePcqQIUMoU6bM/5xErlGjRqxYsYKjR4+qb1SzWh48PT3Zv38/gYGBHD16FC0trWz967Nu6HK64YfMfvcvb/ey19WMZG0bGRn5ykhZGRkZ6pvDl8XExHDixAm6dOmS6/76Odm9e/d/rq9RowZmZmYcPXqU+Ph4Ll68yODBg4HM86ilpcXRo0fVBb0vd496+dgiIiJynI38Tc/Xy+cqJ6+7LkVd1nmbMWMGw4cP13A0QgjxKqnBEHkuq/l90KBB/PHHH9SqVYutW7cSHR2Nn58fnTp14scff2ThwoXcvn0bIyMj6tSpI8mFUD9BfZM6jKwb46ynuTY2NuqC76xEY9++fVy4cIHq1atnuzEuX748RkZG3Lx5U92t5GVZT4dz6rryOtWrVwfIsSXhzJkzOXaP2rt3L2lpafkyetTLtLW1adCgAQ8ePGDNmjWkp6erz5GlpSXVqlXj2LFjORbDwz/Hdvz48Vf2/ezZM27duoWxsTEuLi65iue/zpVSqeTMmTO5PraiJGsQgKI+Z4wQ4sMlCYZ4Z0+ePMn2Oi0tjVGjRmFoaEidOnUICgoCMmstHj58iI2NDYmJiRgYGDB37lxNhCw0ZPPmzfj5+b0yPwBkPv3OKqbOqU/+62TdGF+4cIFjx45luymuVKkSJUqU4OeffyY1NVV9M51FV1cXHx8fEhMTmTZtWrbE5smTJ8yfPx+FQkHv3r1zHU/W/A4//vhjttaKlJQUpk+fnuN7du7ciYmJibrLWH7KOgfz5s1DT08PDw8P9TpPT0+uXLnCoUOH1PN3vCzrPMybN0/dWgGZCeHXX39NYmIiPXr0QFdXN1ex1KtXDxcXF86ePcv27duzrVuxYkW+1F8UBjVq1OCjjz5iz549/PHHHzkm5EFBQTK0rBBCY6SLlHhrR44cYenSpVSsWJFJkyYRGhrKs2fP8PDwoFq1aly5coXBgwer/wAGBQVRtmxZ6tWrR+3atdVzIoii48KFC/z666+UKFGC+vXrq0cVCg0N5cCBAyQlJeHm5sbAgQPfaL9ZN8axsbGvPHVv1KiRejSdf68D+Oabbzh9+jSrV6/m2rVrNG7cmNjYWLZt28bz58/54osv3mhSx/r16zNo0CB+++033N3d8fb2Rl9fnz179mBubo6trW22oWoTEhLw8/OjRYsWGBgYvNFxv42scxAZGYmHh4d6EAbITD4WLFhAdHS0emShl7m5ufH5558zb9483N3d6dChA2ZmZvj5+XH16lUqVar0nyNl/ZtCoeCXX36hY8eO9OvXj3bt2lGuXDlu3LiBn58fH3/8MYcOHXr3g/6X/xqm1sDAgDFjxuT5Z+a15cuX0759e0aNGsXSpUupW7culpaWPH78mNu3b3Pt2jXWrFmDvb29pkMVQhRBkmCIt3LgwAGWLVvG559/jpOTE/r6+jx58oRdu3bh4eGBra0tUVFRODs7k5ycjK+vLxkZGdSvXx9DQ8NsE3eJomPEiBG4uLjg5+fHzZs38fPzIzExEUtLS9zc3OjQoQO9e/fO9RPwLJ6enixYsADglVaKrARDX1+fevXqvfJeCwsL9u/fz4IFC9ixYweLFy9GX19fPSGct7f3Gx/n7NmzcXZ2Zvny5fzxxx9YWVnRtm1bvvrqq1eKuA8dOkRSUlK+d4/KUqFCBUqWLMmTJ09eOVf169dHX1+flJSUHEeKgszhWrOGG960aRMpKSmUKVOGcePGMWrUqBwLhf9L/fr12bt3L9OnT+fw4cMcPnyY2rVrs2vXLg4fPpwvCcbJkydz7JYFmfUNBSHBKFmyJH5+fixbtozt27ezZcsW0tLSsLGxwdnZme+///6dBgwQQoh3oYiJicl9Z2dR5CmVSrS0tFi1ahVaWlr06dNHve7+/fvs3r2bLl268Pz5c6ZPn86ff/7JwoUL1aNDCSGyGzBgADt37iQoKOiNb85FdkOHDmX9+vVcvXr1lTk3RP4IDQ2levXqfPTRR/9zkAEhRNEhNRgiV+7duwdkdmlQKpVcuXJF3T87q2jV3Nyc5ORkwsPDSU5OJjg4mKCgIEaOHCnJhRA5SE1N5cCBAzRu3FiSizyUVcwvNV75Z/78+VhYWKgL9YUQ4mXSRUq8lkqlQqFQcPPmTebPn8+cOXOwtLREoVDg7OzMli1bqFu3rnr4SSsrKypWrMiqVaswMDDgr7/+kqeIQvwHPT29VybGE2+vTZs22Sa1e7mAXeSt+vXrM2HCBPXr3E4mKIQoGqSLlMhRdHQ02tramJubExwczM6dO2nQoAG1a9cGMls0Vq5ciY+PD9WqVSMgIIBjx47h6+tLTExMjmPkCyGEEEKIwk+6SIkc/f7773z++edAZjFhcnJytiE3ixcvToUKFfjmm2+YP38+48aNU6+T5EIIIYQQouiSBENkkzWe+pgxY0hOTubGjRsYGhpStmxZdu3apd7O3NycPn36MHLkSIyNjfn999/x9fXVVNhCCCGEEOIDIV2khDqpyKqlyKq9WLVqFWlpaeo5Cdq1a8c333xDnTp11KNJCSGEEEII8TK5Q9SgwMBATYeAUqlEoVCgUCjUo0FlJRqlSpVCpVIRGRkJQKtWrdDRyRwXQJKLt/MhXHPx/sl1L3rkmhc9cs2LJrnuOZO7xCIqq9UiK1H45ZdfGDNmDPfv31dvY29vz5MnTzh37hyrV6/myJEjlCxZUiPxCiGEEEKIgkESjCImKCgI+KeVAmDGjBk8evQINzc3vv/+e/VyV1dXmjdvzvnz57l48SLz58+nRIkS7z1mIYQQQghRcMg8GEVIbGwsCxYs4NtvvyU5OZnAwEAaNWrEiBEjyMjI4Oeff+b8+fMcPXqUxo0bA+Du7o6bmxva2tqaDV4IIYQQQhQIkmAUIQqFgtq1azN37lxu3ryJlpYWKpWKWrVqMWPGDFq2bEnlypXZs2ePOsEAJLkQQgghhBC5JglGIRYQEEBISAht27YFwMzMDBMTE54/f87y5cu5d+8ehw4donbt2gQEBODu7s6ZM2fw8PBQjyQlhBBCCPEm0tPTSUhI0HQY74WBgQGxsbH5sm9jY2P14DoFTcGMWvxPSqWSsLAw4uPj8fPzY9u2bTRr1ox69eqxe/duLC0tqVevHgsXLiQlJYX27dtz7NgxRowYgbOzs6bDF0IIIUQBlJ6eTnx8PBYWFkXiQaW+vj4GBgZ5vl+VSkVMTAympqYFMskoeBGLXNHS0sLGxoYjR46QkpKCp6cnS5YsYe3atZiZmbFmzRr69u2Lvb09FhYWDBkyRNMhCyGEEKKAS0hIKDLJRX5SKBRYWFgQFxeHubm5psN5YzKKVCGgVCqJiooCMp8cqFQqtmzZQu3atUlNTaVy5cp07tyZDh068OeffzJ69GhmzZpFv379KFeunNRYCCGEECLPSHKRNwryeZQEo4DbvHkzffv2JSAgQL1MoVCwf/9+duzYQY8ePTh58iQAXbp0QalUYm9vz4IFC1iwYAGfffaZpkIXQgghhBCFkCQYBVRqaipjxoxh586dTJw4EU9PT9atW8ekSZMAGD9+PKdOnaJ+/frExsYyZ84cpkyZgqWlJTo6OrRo0QIzMzMNH4UQQgghhChsJMEooPT09DA2NqZmzZpUrlyZFy9e0K1bNx4/fkxQUBAlS5bEwsKCe/fu0b17d0JDQ+nYsSN9+vTRdOhCCCGEEKIQkyLvAmzcuHH07t2bmJgYTp8+za+//krz5s0JCAjA0dGRixcv0rBhQ5o2bUqPHj00Ha4QQgghhCgCJMEowCwsLKhVqxbnzp3jr7/+wsLCgvDwcDZs2MCxY8fw8PCgSpUqBXL0ASGEEEIIUTBJF6kCrkGDBnTq1AkjIyMA6tevT8+ePenduzdjxoyR5EIIIYQQ4n84efIkH3/8MaVLl8bBwYGmTZty8+ZNAHbs2IGHhwc2NjZUrlyZH374AZVKpX5v1apVmTt3LqNHj8be3p5KlSrx888/Z9t/UFAQrVu3pkSJEtSpU4cDBw5QunRp1q5d+16P832RFowCJi45Dv/7/qSkp6Cvo08l10ocP36chIQE9PT0UCgU1KtXT9NhCiGEEEIUCOnp6fTs2ZNPPvmEZcuWkZaWxtWrV9HW1ubKlSt8+umnjBs3jm7dunHp0iXGjBmDqakpgwcPVu9j8eLFfPnll4wcOZKDBw8yYcIE6tevj5ubG0qlkt69e2NjY8PBgwdJTk7myy+/JCUlRYNHnb8kwShg/O/7E5kQmfkiBW5yk4yMDCIiIrC0tNRscEIIIYQQ7+jfD1M9nTwxNTDNt8+Lj48nNjaWli1b4uTkBED58uUBGDhwIB999JF6lE5nZ2fu3bvHggULsiUYXl5eDBo0CIDBgwezdOlSjh07hpubG35+fgQGBvL3339TqlQpAGbOnEmLFi3y7Zg0TbpIFTAp6SmvvJ4xYwYVKlTQUERCCCGEEHkn62FqXEockQmRHLt/LF8/z9LSkp49e9K5c2e6devGwoULefjwIQB37tx5pWeIu7s7jx8/Ji4uTr2scuXK2baxtbUlMjLzgfDdu3cpWbKkOrkAqFWrFlpahfc2vPAeWSGlr6P/yuvC/AMqhBBCiKIlp4ep+W3x4sUcOnQIDw8P9u7dS506dTh8+DAqleq1M2q/vFxXV/eVdVl1Gi/XaxQVcmdawHg6eWJtbI2ZvhnWxtZ4OnlqOiQhhBBCiDyT08PU96Fq1aqMHj2a3bt306BBA9avX0/FihU5c+ZMtu1Onz5N6dKlMTXNXbetChUq8OTJE548eaJedvnyZZRKZZ7G/yGRGowCxtTAlLaubTUdhhBCCCFEvvB08uTY/WPZajDyU0hICKtWraJVq1aULFmSkJAQbty4Qf/+/WnWrBleXl7MmjWLrl27cunSJRYtWsRXX32V6/03adIEFxcXhg4dyvTp00lOTmby5Mno6Oi8tnWkoJMEQwghhBBCfDDe98NUIyMjgoKC+PTTT4mKisLGxoauXbsyevRodHV1WbVqFd9//z3z5s3DxsaG0aNHqwu6c0NLS4s1a9YwYsQImjZtioODAzNmzOCTTz7BwMAgH49McyTBEEIIIYQQRZaNjQ1r1qx57Xpvb2+8vb1fu/769euvLNu9e3e2187Ozuzduzfbe9LS0ihbtuxbRPzhkwRDCCGEEEKIfLRz506MjY0pW7YsDx48YPLkyVSpUoXq1atrOrR8IQmGEEIIIYQQ+ejFixdMnTqVR48eYWFhQYMGDZg5c6bUYAghhBBCCCHeXI8ePejRo4emw3hvZJhaIYQQQgghRJ7ReIKxfPlyqlWrRokSJfD09OTUqVP/uf2NGzdo3bo1tra2uLq6Mnv27FcmMNm0aRMNGjSgZMmSlC9fnkGDBvH06dP8PAwhhBBCCCEEGk4w/v77byZOnMjYsWPx9/fHzc2Nrl27EhYWluP2cXFxdOzYERsbG44cOcL333/PL7/8wsKFC9XbnDlzhsGDB9OjRw9Onz7N2rVruX37NgMHDnxfhyWEEEIIUWQVxZmr80NBPo8aTTAWLVpEz5496du3LxUqVGDu3LmUKFGClStX5rj9pk2bSEpKYsmSJVSqVIn27dszatQoFi9erL4I58+fp1SpUvj6+uLo6EjdunUZNGgQFy9efJ+HJoQQQghR5BgbGxMTE1Ogb44/BCqVipiYGIyNjTUdylvRWJF3amoqV65cYcSIEdmWe3l5cfbs2Rzfc+7cOdzd3TE0NFQva9q0Kd999x2hoaE4OjpSr149vv32W/bu3UvLli2Jjo7m77//plmzZvl6PEIIIYQQRZ2Ojg6mpqbExcVpOpT3Ii4uDjMzs3zZt6mpKTo6BXM8Jo1FHRUVRUZGBtbW1tmWW1tbExERkeN7IiIiKFWq1CvbZ61zdHTEzc2N5cuXM2jQIJKSkkhPT6dJkyYsWbLkP+MJDAx8h6N5e5r6XKE5cs2LJrnuRY9c86JHrnnRlJyc/N4/08XF5b1/5pvQeFr07/F/VSrVf44JnNP2Ly+/ffs2EydOZPz48Xh5efH06VO++uorRo8ezdKlS1+7X01cqMDAwA/+B0TkLbnmRZNc96JHrnnRI9e8aJLrnjONJRjFihVDW1v7ldaKZ8+evdKqkcXGxibH7eGflox58+ZRq1YtRo4cCUCVKlUwMjKiVatWfPXVV9jZ2eX1oQghhBBCCCH+n8aKvPX09KhRowZ+fn7Zlvv5+VGvXr0c3+Pm5sbp06ezNUX5+flRsmRJypQpA0BSUhLa2trZ3pf1WgqOhBBCCCGEyF8aHUXK19eXdevWsXr1au7cucOECRMIDw+nX79+AEybNg1vb2/19l26dMHQ0JBhw4Zx8+ZNduzYwU8//cSwYcPUXaRatmzJnj17WLFiBSEhIZw5c4YJEyZQvXp17O3tNXKcQgghhBBCFBUarcHo1KkT0dHRzJ07l6dPn+Lq6srGjRtxcHAAIDw8nPv376u3Nzc3Z+vWrYwbN44mTZpgYWGBr68vw4cPV2/Tq1cvXrx4wbJly5gyZQpmZmY0bNiQadOmvffjE0IIIYQQoqhRxMTESL8hDZHCoKJHrvmH7ezZs6SkpNCoUaM83a9c96JHrnnRI9e8aJLrnjONdpESQogPwYsXL4DMVtIjR45w6NAhkpKSNByVEEIIUTBJgiGEKLJiY2MBOHDgAKmpqVSsWJEmTZpw//59zp07p+HohBBCiIJJEgwhRJH166+/olKpiIiIoEmTJowePRo7Ozv09fWJiopCqVRqOkQhhBCiwJEEQwhRpKSlpan/PX78eMLDw9Xz43Ts2JFy5cphZWXFyZMn0dKSr0ghhBDiTclfTyFEkZCamsq6desICwsDICwsjHv37rF27VqcnJyYPn06Fy5cAKBt27ZERkby6NEjTYYshBBCFEiSYAghCrXnz5/z559/oqenx7Vr13j8+DGxsbEsWbKE5ORkTE1NSUpKwtbWlmLFinH9+nUuXbpE/fr1KVmypKbDF0IIIQocSTCEEIXW6tWr6dKli7rVolWrVqxevZoLFy5Qo0YNqlatiq6uLnfv3sXBwYEXL14wePBg7t27x/3799VF4EIIIYTIPUkwhBCFTnh4OC9evOD69ets2bKFSZMmAeDp6UmpUqWYPn06JUqUADKTjj///JP09HR8fX3Zvn07Xbt2ZcaMGVhaWmryMIQQQogCSRIMIUSh8vjxY2bOnElGRgbXrl0jLCyMO3fusG/fPgCaN29OQkICS5YsYcWKFcTExDB06FCeP38OgLW1NUqlEn19fU0ehhBCCFFg6Wg6ACGEyEsqlYpKlSphbm5OmzZtmDVrFnfu3MHFxYXbt2/ToEEDJk2aRMmSJUlJSeHu3bu0b98+2z5k9CghhBDi7UmCIYQoVBITEzl06BBDhgxh5MiRtGnThnLlynHz5k2WL19OsWLF2LNnDzVr1sTR0VHT4QohhBCFjjymE0IUSEqlEpVKlW2ZSqXCxcWFatWqcebMGQDKlSsHQGRkJCkpKTg6OjJu3DhJLoQQQoh8Ii0YQogCJTU1FT09PXU3pujoaBQKBZaWligUCuLj47G1tSU+Ph6lUsnNmzf54YcfiIqKYuLEiSgUCoyMjDR8FEIIIUThJQmGEKLACAgI4MmTJ3h4eGBoaMj06dMJCAjA2dmZWbNmAWBqakrJkiUJDg6mQoUK2Nra0qxZM3r16qXh6IUQQoiiQbpICSEKDAcHB5o1a4axsTFPnz7FyMiITZs2qesrsjRt2hRbW1t+//13IiMj1cmFUqnUVOhCCCFEkSEtGEKID1Z4eDgZGRmULl0aADMzM2bPnk39+vUBuHr1KqNHj8bW1hYvLy/1+4yMjGjfvj1lypTBzMxMvVxGhxJCCCHyn/y1FUJ8kJKSkliyZAmJiYk8f/6c1atXExkZiZeXFxs2bOCjjz7i9u3beHp6snTpUq5du8a9e/ey7aNGjRrq5EQIIYQQ74ckGEKID8qKFSs4e/Ysenp61K5dm9mzZzNlyhTu3r1L//79cXFxwcbGhvPnzzN58mT27dvHxx9/zKNHj9QjRgkhhBBCc6SLlBDig3D16lUmT56Mvb09MTExLF68mD/++INff/2V7t2706dPH1atWsX69etp2bIl169fZ9CgQVhZWVGpUiWsra01fQhCCCGEQFowhBAfiOvXrzNixAiWLFnC2LFj0dfXZ8uWLYwYMYIrV64A4OrqSunSpUlNTcXS0pL09HQ8PT2xtraWAm4hhBDiAyEJhhDig3DkyBFSUlLUr319fTl69Chly5ZFX1+fEydOsGPHDjIyMvD09KRr167o6PzTCCsF3EIIIcSHQf4iCyE0ImsWbpVKhVKppHr16tja2qrXZ2RkYG1tTZkyZUhPT2f9+vWUKVMGb2/vV/YhhBBCiA+H1GAIId6bu3fvcvbsWT755BMUCgUvXrzAxMQEhUJBYmIiFy9exM3NDQArKyvCwsIwMDCgdevWVK9eHSsrKyAzsVAoFCgUCk0ejhBCCCFyIC0YQoj3xsrKio0bNxITE8Pq1avZuHEjaWlpAPTo0YO9e/fi7+/PlStXmDlzJg0bNgSgSZMmWFlZqVssJLEQQgghPlzSgiGEyFdKpVJdH1G8eHG6devGvn37ePz4MSNHjkRXVxelUomjoyPDhg3j6tWr7N+/n27dutGnT59s+5LEQgghhPjwSYIhhMhXWcnFxo0biYuLo2zZsly/fp2oqCiMjIyAzHoLLS0tWrZsScuWLRk8eDB6enrAP92hhBBCCFEwSBcpIUS+Gzt2LHv37qVVq1bY2dnh5+fH06dP+fLLL7l8+TIJCQnZttfT05PuUEIIIUQBJS0YQog8kZCQgLGxsfp1VFQUYWFh1KhRA21tbX7//Xf1OhcXFzp37oxSqeTEiROYmppiYWGRbX+SWAghhBAFk7RgCCHeSUZGBvPmzeP7778H/hk69s6dO3z77bcAPH36lODgYPV7WrduzcqVK/Hw8GDEiBE4Ozu//8CFEEIIkS8kwRBCvLVr167h7e1NQkICI0eORKVS8dlnnxETE4OHhwfVq1dn//79dOrUie3bt7N9+3Z69uxJw4YN+emnn9T7kfkshBBCiMJDEgwhxFszNDQkNjaWr776CmtraxQKBTVr1mTp0qUA1KpVi/DwcGrWrMnTp0/ZtWsX48aNQ0dHByMjI5RKJSDdoYQQQojCRGowhBBvzcXFhSZNmvDNN9+QlpaGvr4+Q4YMYd68eQDExcVx69Yt+vbty7Rp09DX1wf+GRkqa4QpIYQQQhQe8tddCPFOunfvzpYtWzAzM+Obb77BwMCAsmXLMmrUKIKCgqhUqRKADDsrhBBCFBHSgiGEeCelS5emb9++NG/eHAATExM6d+5MQkICffv2pVixYsA/3aAkuRBCCCEKN0kwhBBvLC45Dv/7/qSkp6Cvo4+ugS5hYWFUr14dLS0tihcvzueffw5Ii4UQQghR1EgXKSHEG/O/709kQiRxKXFEJkRiWM6QU6dOAa+2UEhyIYQQQhQt0oIhhHhjKekp2V4Xty/OgOkDNBSNEEIIIT4k0oIhhHhj+jr6r7zW1taW+SyEEEIIIQmGEOLNeTp5Ym1sjZm+GdbG1ng6eQLSHUoIIYQQ0kVKCPEWTA1MaevaVtNhCCGEEOIDJC0YQgghhBBCiDwjCYYQQgghhBAiz0iCIYQQQgghhMgzkmAIIYQQQggh8owkGEIIIYQQQog8IwmGEPkkIyND0yEIIYQQQrx3kmAIkccCAgKIj49HW1sbgF27drF27VoNRyWEEEII8X5IgiFEHoqKiuLIkSNcvnwZgM8//5yDBw9y+/ZtNm/erOHohBBCCCHynyQYQuShYsWKYWhoiEqlQqVS4erqyowZM4iOjmbPnj2kpqZqOkQhhBBCiHwlCYYQ70CpVGZ7HRISgomJCcePHycpKYlNmzbRrVs3mjRpwq+//ip1GUIIIYQo9CTBEOItREZGEhMTg5bWP79CDx48YPr06XTu3Jk7d+7w9OlT2rRpQ+3atalatSojRozg3LlzGoxaCCGEECL/SYIhxFt4+PAhhw4d4smTJ3z77bfcvn0bBwcHXFxcSEtLo3fv3qxcuZLhw4ejr6/PjBkz6NmzJ56enpoOXQghhBAiX+loOgAhCoqMjAz1yFAqlYqFCxdy4MABHBwc2Lt3L3fu3CEqKgptbW2aNGnC0aNHCQkJYeLEiejq6gIQGBioyUMQQgghhMh3kmAIkUtZycWlS5cICAjAyMiIsWPHUqFCBQ4fPkxAQACXL1/m0aNH2NjY4OPjg5OTk7ob1b/rNYQQQgghCiPpIiXEa2SNBJXl3r17eHl5sX79enx8fGjbti3BwcEANGzYEDc3Nx49ekSxYsUwNTWlRo0a2Wo0Xv63EEIIIURhJXc8QuRApVKhUChQKBTEx8cDcPr0afr06cPcuXPR0dGhadOmrF27ltDQUB49eoS7uzsdO3bk4sWLGo5eCCGEEEJzJMEQ4iVZLRYKhYK0tDS++uorRowYwa5du6hWrRpnz56lX79+dO/eHXNzc9zc3OjVqxf37t0jPT0dbW1tqlWrpuGjEEIIIYTQHKnBEILsLRZZ/P39sbS0ZOzYsbRs2ZJNmzaxZMkSAKZPn87Vq1cZOnQoAwcOxNDQkEuXLlGsWDHMzc3V+xNCCCGEKGqkBUMUaS+3WACcP3+eX3/9lZSUFC5duoSWlhYzZsygfPny6OjocOfOHXx8fAgICKBmzZro6upiaGgIQPXq1Rk9ejR6enqSXAghhBCiyJIWDFEkKZVKtLS0siUCS5cuxc/Pj2rVqrFp0yYqVKjAyJEj2bt3L05OTly7do3k5GRGjRqFh4fHK/vMGmVKCCGEEKIokxYMUeQsX76ckydPql/Pnz+fHTt2ULx4cebPn4+lpSVLly6lYcOGdOjQgQ0bNtCiRQuCg4Np1KiROrnIyMjQ1CEIIYQQQnywJMEQRcaOHTto3bo1gYGBVK5cmdjYWCZPnkxMTAzu7u7s3r0bHx8fVCoV27Ztw9zcnK+//pry5cuzadMmunfvnm1/0mIhhBBCCPEq6SIliox79+5RvHhxZs+erV4WGxtL1apVsba2pkWLFsTFxdG9e3dmzJhBhQoVGDRoEL169QL+6VYlhBBCCCFeT+6WRKGXVcjdsWNHKlWqxPnz5/nhhx+4fv06/fr1w8/PDwAfHx88PDwYO3YsNWvWZPDgweoaDZVKJcmFEEIIIUQuSAuGKJSePXtGQkICZcqUUScJurq6pKSkMHToUHr37k2lSpXQ1tamYsWK7NixA29vb3x9fbONApXT8LVCCCGEEOL1JMEQhcrly5extrYmKiqKmzdvEhISgoGBAfXq1aN06dLY2toyaNAgBg0aBGQWateqVQsDAwMA9PX1gZxHmRJCCCGEEP+b9PkQhYpSqcTCwoJixYoxa9Ys5s2bh729PUqlEoAaNWpkG0FKW1ubpk2b0rx582z7ke5QQgghhBBvR+6iRIGWlpaW7XXt2rU5ceIE58+fZ+DAgdSsWZNSpUqpWyLq1atHYmIia9asUb/H2NgY+KdWQwghhBBCvD1JMESBpVQq6devH0+fPgXg6tWrQGaiEB4ezogRI4iIiODx48fZujotWbKErl27vrI/6Q4lhBBCCPHupAZDFDiHDx/m3r17DBo0iNatW/Pdd9+RmJjInTt3mDZtGiYmJujp6REeHo6Xlxfz58/H3t6eESNGoFAoKFasGAqFQl3ALYQQQggh8o60YIgCQaVSkZCQwJQpU/jpp59wcHAgIyODRo0akZycTJs2bVi0aBF37tzB0dFR3WrRtGlT0tPTqVq1qjqZ+Pf/hRBCCCFE3pEEQxQICoWCtLQ0FAoFW7ZsoWXLlmhra2NiYoK9vT3m5uZUrlyZy5cvo1AoMDIyYuvWrVhaWjJ//nyaNGmi6UMQQgghhCgSpIuU+KC9PHt2YGAgycnJHD9+nMePH6Orq0vnzp0pUaIEL168QFtbmwYNGjBnzhx+/PFHUlJSctyPEEIIIYTIP5JgiA/WihUrSEhIYOTIkQBUrlwZR0dHdu3aRXBwMAYGBpw9e5bx48czfPhwvL296dOnD56enujq6qKrq6uus5DkQgghhBDi/ZAEQ3xw0tPT0dHRoVu3bgwaNIg+ffpgYWGBkZERvr6+PH/+HEtLSwDatWtHqVKlqF27Nvfv38fR0ZEyZcrIDNxCCCGEEBoij3XFB0dHRwelUsmZM2cwMzNj+/bt2dZbWlqSlpbGggULqF69OgCTJ0/GyclJCriFEEIIITRMEgyhcSqVKtskdykpKUydOhU/Pz9q1apFZGQk0dHRAKSmpnL+/Hk+/vhjHj16pO4+lbUf8d+USiXBwcGaDkMIIYQQhZh0kRIal9XacOvWLaytrbGysiIjI4NRo0aRkZHBunXrePHiBVZWVujp6eHo6MjixYupXLkygHSHegNaWlqsW7eO9u3bo62tjYuLC7q6upoOSwghhBCFiMZbMJYvX061atUoUaIEnp6enDp16j+3v3HjBq1bt8bW1hZXV1dmz579ypPr1NRUvvvuO6pVq4aNjQ1VqlTh119/zc/DEG8gNTU122ulUsncuXMZOHAgf/75J3PmzKFixYoolUpKlSrFvn37uH79unp7a2trKleujEqlQqlUSmLxP2SdpywmJia0atWKzZs3y7kTQgghRJ7TaAvG33//zcSJE/nxxx+pX78+y5cvp2vXrpw5cwZ7e/tXto+Li6Njx454eHhw5MgRAgMD8fX1xcjIiBEjRqi3++yzz3j06BELFiygbNmyREZGkpSU9D4PTbxGTEwM06dPZ/LkyVhZWXHt2jUcHR1JTU3lxIkTrFq1ioCAAKKiokhKSiI9PZ20tDRevHjxyr6k1eJ/e7l15969e9jZ2VG2bFm8vb3p2rUrOjo6MqO5EEIIIfKURhOMRYsW0bNnT/r27QvA3LlzOXz4MCtXruSbb755ZftNmzaRlJTEkiVLMDQ0pFKlSty9e5fFixczfPhwFAoFR44c4dixY1y+fJlixYoBUKZMmfd6XOJVWfNQ6OvrU65cOYKDg7l9+zb+/v4MGzaMEydO0LJlS2rXrs3333/Pixcv2Lt3L35+fqxZsybHhFO83suJRVJSEitWrOD333+ne/fu+Pj4YGJiwu3bt3F1dZXkQgghhBB5SmNdpFJTU7ly5QpeXl7Zlnt5eXH27Nkc33Pu3Dnc3d0xNDRUL2vatClPnjwhNDQUgN27d1OzZk0WLVpEpUqVqFWrFl988UWOT8DF+5GRkaGehyIlJQVjY2OioqI4cuQI3t7e6OjoUKNGDQYOHMj48eOZMmUK4eHh9OjRg99++w17e3uUSqUUcedC1jl6OWlYs2YNZ86cwd/fH21tbV68eEFSUhLx8fGsXLmSa9euaSpcIYQQQhRCGkswoqKiyMjIwNraOttya2trIiIicnxPREREjttnrQMICQnhzJkzBAQEsHr1anWryLBhw/LhKMTrKJVKdu7cCYC2tjbJycl88cUXbN68GRMTE3bs2EF4eDiVKlXCyMiIunXrcuvWLTp27IizszMVK1bMti8tLS150p4LWefo0KFDTJo0iZCQEGxsbHB1dcXY2Bhra2v27dtHmzZtSElJ4cyZM5QqVUrDUQshhBCiMNH4KFL/vmn8X/3Bc9r+5eVZRb/Lli3D3NwcyOx61alTJyIiIrCxsclxv4GBgW99DO9CU5+bH7KunZ+fH25ubiQnJ3Pnzh20tLRYvXo1SUlJ1KtXDwMDA7Zs2UJwcDBTp04lNTWVZs2a0bJlS9q3b4+BgUGhOi//ltfHlpGRgba2tvr1vn37OHnyJB4eHqxevRpdXV2ioqIIDAykXr169OvXj6pVq9K4cWMaN27M8+fPef78eZ7GJF5VmH+mRc7kmhc9cs2LJk1cdxcXl/f+mW9CYwlGsWLF0NbWfqW14tmzZ6+0UmSxsbHJcXv4pyWjRIkSlCxZUp1cAJQvXx6Ahw8fvjbB0MSFCgwM/OB/QN5EQkICxsbG/PbbbyQnJzNy5Eh++OEHxo0bx9OnT5kyZQoODg4ADB8+nOXLlzN+/Hg2bNhA9erVsbKyAl69YS5M8vKaZ7XsQGbXs6CgICpXrszt27cpVaoUERERnD9/nkWLFjFjxgyuXbtG586dWb16NeXKlctxPyJ/FLbfdfG/yTUveuSaF01y3XOmsbsKPT09atSogZ+fX7blfn5+1KtXL8f3uLm5cfr0aZKTk7NtX7JkSXUhd/369QkPD89Wc3Hv3j0AKRTOR3/88QedO3cGoG3btlhbW5ORkUFMTAznz5+nQYMG6qfkT58+pXz58tSqVQvIHPUrK7kACm1ykVeyhpzNSgpCQkJo1aoV06ZNY+fOnYSGhvLrr7+iUChYt24dpUuXZtSoUerJCl9OLl7ejxBCCCFEXtDonYWvry/r1q1j9erV3LlzhwkTJhAeHk6/fv0AmDZtGt7e3urtu3TpgqGhIcOGDePmzZvs2LGDn376iWHDhqm7SHXp0gUrKyt8fX25desWZ86cYeLEibRv3/61LSPi7Z08eRIAKysrIiMjCQoKIjExkdu3b6OtrY2rqyvPnz8nNjaWc+fOMX78eCZOnEh4eDjDhw/H2NgYkFm430RWQnD9+nV69uzJjh07mDdvHhMnTiQpKYnixYvj4uKCl5cXX3zxBdOmTcPV1ZWBAwdqOHIhhBBCFAUarcHo1KkT0dHRzJ07l6dPn+Lq6srGjRvV3WjCw8O5f/++entzc3O2bt3KuHHjaNKkCRYWFvj6+jJ8+HD1NiYmJmzbto0vvvgCLy8vLCwsaNOmTY7D3op3ExoayqZNm9DV1aVMmTIYGBiwZ88eevfuzZ9//klsbCwmJiZER0fTr18/Tp48SUhICPPmzcPS0hKQWbhz69/dxhYuXEhQUBA+Pj4cP36cwMBAmjRpwvr165k6dSpmZmb8/PPPNGvWjE6dOqnfJ3NeCJE7T548ISUlBUdHR02HIoQQBY4iJiZGHh1rSEHvt6dSqTh27Bjbt29n2rRprFmzhrCwMMqXL09UVBRDhw4lLi6Ozz77jA0bNmBmZqZ+b1Ht9/8u11ylUnH69Gnc3d3ZsWMHixcvZv/+/ezatYv09HQ6dOjA5MmTsbW1zTbxJBTd8/2hKOi/60VJWloaACtWrMDZ2ZmPP/74rfYj17zokWteNMl1z5nccYi3plAoaNy4MdbW1nzzzTfY2Ngwfvx4wsLCWL9+PUFBQZQsWZL58+djZmam7gYlN7v/m0qlytZt7M6dOzRq1Ih9+/YxePBgGjZsSN26dQkMDMTOzo4zZ84AMGXKFD755BP1+/5dryGEyO7l37PAwEBWrFiBlpYWxsbGREZGvrKNEEKI/03uOsQ7Gzp0KI8fP+bXX38lMTGRjh07UrFiRfUf5woVKgD/DCUsN7v/LWuoZYVCwYMHD4DMgQpmzJjBt99+S1hYGEePHqV69eoEBwdTo0YNTE1NefHiBfr6+lhYWKhviORcC5GdSqXiwYMH+Pr6ApnfSwcOHCAkJISYmBgUCgXJyck0bNiQzZs3k5iYKN0KhRDiDWl8HgxR8MQlx+F/35+U9BT0dfTxdPKke/fujBo1ikePHlGvXj2WLVuGkZGRpkMtkLS0tHj+/DkbNmxQ1xMlJSUxYcIEKlWqxJdffkmjRo34/fff1YnE5MmTs+1DboiEeFV8fDympqakp6dTsmRJkpOTmThxIoGBgbi7u+Po6Mjdu3eJjIzE0dGRYsWKsXjxYsaNG6fp0IUQokCRBEO8Mf/7/kQmZLZOkALH7h+jY8eObN++nZIlSwJgZGT0yiSIImf/7jKWmJjItGnTcHZ2pnv37hw/fpwWLVrQvHlzfH19uX79Oj///DNdunRRn2+QAm4h/svx48c5cuQITZs2VbcQpqeno6Wlxe7duzl9+rR6CO0ffvgBNzc3XFxccHNz03ToQghR4EiCId5YSnpKjq9XrVqVbbnc7OZOVnJx6tQpypUrh4WFBdHR0QwbNoyoqCg2bNhAUlISDRs2ZPjw4Tg7OzNu3DhKlSoFyEhcQuRGjRo1MDExYcqUKVSpUoU6deoQHh6unoPHzs6O+fPn89dff/Hbb78B8Pnnn8u8PEII8RYkwRBvTF9HH1L+9fr/SQF37rx8niIiIvjmm28ICgrC3d0dd3d3mjZtyoMHDyhTpgynT5/Gw8MDLy8vqlWrRtmyZQFJLIT4X9LT09HRyfwzZ2JiQs2aNZkyZQqbN2/m+vXrtG/fnsuXLxMREcH9+/dxcnICYODAgfI9JoQQ70C+QcUb83TyxNrYGjN9M6yNrfF08lSvkz/K/y0jIwPIPE9Z/46MjKRfv34cPHgQQ0NDgoKCePbsGUeOHGHGjBk8f/6c7du3A6iTi6xCcCFEdkqlkhkzZvDw4UN0dHR4+vQpqamp6t8Xd3d3PvnkE/Ukn3379mXRokVMmzaN9u3bo1Ao5HtMCCHekXyLijdmamBKW9e2dK7ambaubTE1MNV0SAVGVneLDRs20KdPH9atW0flypVRqVR07tyZgIAAtLS0aNq0qXpEqGXLlqGtra1OSEASOSFyklVTYW5uzurVq1GpVEycOJHo6Ohs21lYWJCRkUFgYCDe3t58+eWXHD58GA8PDw1FLoQQhYvcpQiRj1QqlXouCoCEhAQGDx5MeHg4Y8aM4fLlywQGBvL8+XP69evHxIkTOXLkCDY2NvTq1YtmzZoxevRoHBwcpC+4EK+RNaBEVncoHx8flEolq1atwsvLC1tb22zbOzo6kpycTFJSEgAGBgbvN2AhhCjkpAZDiHwQFxeHmZmZukbi/v37JCQkYGxsTEpKCkqlkjp16hASEsKlS5fQ0dFh2bJl2NraMmrUKEqVKkVERATbtm1j4MCBtGjRQtOHJMQHK6v7k7+/P/v376dx48Y4Ojpy4MABvv7662zbZtVlTJkyBT09PU2EK4QQhZ4kGELkkayi6wMHDjB48GACAwPR0dFh6dKlLF68mPbt26Ovr8+kSZNYuXIlaWlpGBkZERERgY+PD+XLl6dq1arq/dnY2DBx4kQNHpEQBce5c+eYO3cuY8aMwcvLi9WrV3P9+nU2btyIUqmkQYMGNG7cWN3KIcmFEELkH+kiJUQeyXqK+vTpU/T09Ni+fTsZGRk4Ojpy6dIlPDw8CA4OJjY2FmdnZx49eoSrqytGRkYYGBiok4uXay2EEP9QKpVERUUBmS0RAH/99RcA169fZ+TIkXh5eQFQpUoV6taty5dffknt2rVJTU3VTNBCCFEESYIhxDtKSkqiW7du6tl+a9SoQZs2bfDz8yMuLo769eszduxYVq5ciUKhIDExkUePHhEfH4+TkxPdunXLNiKU1FoI8arNmzfTt29fAgICsi0/fPgwO3bswMnJidOnT6uX16pVCwMDAy5fvkyrVq1o3rz5+w5ZCCGKLEkwhHhHKpWKMmXKEBQUxPHjx7l+/ToVK1bEw8ODO3fu8OLFC+zt7VmzZg0RERFoaWkxaNCgbN2hsopUhRDZpaamMmbMGHbu3MnEiRPx9PRk3bp1TJo0CYDx48dz6tQpbG1tUalUrFy5Eh8fHzZv3sz8+fOpVauWho9ACCGKHkkwhHgLL48MZWRkRPXq1bGyskJLS4vg4GACAwOJi4sjJiaGsLAwgoKCaNGiBR4eHjRs2JCSJUtm25/MaSFEzvT09DA2NqZmzZpUrlyZFy9e0K1bNx4/fkxQUBC2traUKFGC1NRU2rVrx6lTp+jZsyddunSR1kAhhNAQKfIW4g2pVCr1PBSJiYkYGRnRokUL1q9fj76+Po8ePaJ06dJUqlSJv//+m59++gk7Ozv09fWJiYlR70OSCiFyZ9y4cfTu3ZuYmBhOnz7Nr7/+SvPmzQkICMDR0ZFTp05Rq1YtPD09Wb58uabDFUKIIk9aMIR4QwqFgqCgIMaOHcvNmzdJT0/H2tqa5s2b4+/vz4QJE7C2tqZRo0bo6ekRHx9P6dKlsba2Vrd8SHIhRO5ZWFhQq1Ytzp07x19//YWTkxMuLi74+fkxfvx4PvroI6pXr67pMIUQQvw/acEQ4n9QKpXZZs5++PAh8+fPp1SpUtSpU0e9vGfPnsycORMDAwMGDx4MwJw5c7LtS2bgFiJ34pLj8L/vT0p6Cvo6+tR2q42DgwNGRkYA1K9fHy0tLXR0dKhdu7aGoxVCCPEySTCE+B+ykoLly5djYWGBm5sbNjY2NGrUCPgnAUlPT8fFxYWoqChsbW3Vy/+doAgh/jf/+/5EJkRmvkgBPUs9Qs+GkpCQgJ6eHgqFgnr16mk2SCGEEDmSBEOIf8ka0SmrG1NKSgpjxozBxsaGFy9eoK+vj0qlIiwsLNv2JUuWpF+/fhgaGgL/JCaSXAjx5lLSU7K9NjAzICMjg4iICCwtLTUUlRBCiNyQOx8hXpJVfK1QKHj27BmnT58mPT2dmjVr0rt3b86cOcPNmzfp0aMH69evJzY2NttINVnJhRDi3ejr6L/yesaMGVSoUEFDEQkhhMgtacEQguyJRUpKCrNnz1aPUGNtbc3+/fs5cOAAAwYMoHHjxujr69O5c2fu3bsn4+wLkQ88nTw5dv+YugbD08lTWgOFEKKAkARDFGkvJxZZ/P39iYuLY968eQwdOpTixYtTrlw5KleujJ6eHoMGDeLTTz/l008/1VzgQhRypgamtHVtq+kwhBBCvAV5HCSKpKSkJOCfOoszZ87w/fffAxASEkLjxo2xs7OjcePG/Pzzz0yePJno6GhWr17NxIkTadKkicZiF0KIguzEiROkp6drOgwhRD6SBEMUOYsWLaJHjx4AxMfHs2zZMpYvX87Vq1dZv349tWrVYuvWrQB4enpy+fJlnj9/zujRo/n9999xdXVVF3YLIYTInayk4tixYyxZskTD0Qgh8pMkGKLIiYuLIyUlBX9/f/z8/Lh69SqTJ09m1apVbNu2jerVq6Ojo0O/fv34+eefcXZ25smTJ+r3K5VKmShPCCFyKSUlhcjISHbt2gXAyJEjCQ4O5tSpUxqOTAiRXyTBEEVGVqtDpUqVcHNzIyQkhISEBGJjY0lMTERfXx93d3dWrVrFjz/+SN++fVm5ciWmpqYUK1ZMvR8pNBVCiNy5f/8+a9euxdramvXr19OnTx+uXr1Kq1at2LFjh6bDE0Lkkzcu8r5//z4HDhzgwYMHADg4ONC8eXOcnJzyPDgh8pJCoVAnGSNHjuTQoUOcP38epVLJunXr+O677/j44485d+4cBgYGlClThl69emFtbY2FhYW6IFwIIcR/y/q+dHJyonTp0sTExKCrq8vz58+pV68eurq6LF26FH9/fxo1aiTfr0IUMm+UYEyePJlff/0VpVKZbfmkSZMYMmQI3333XZ4GJ0ReUygUJCYm0q5dO8qUKYO3tzfbt29n//79DB06lEqVKlGlShUADAwM8PHxoVOnThqOOvdiYmKwsLDQdBhCiCIqKCiIqKgo9SzrMTEx+Pn5YWlpycCBA9XrbW1t6d27NyEhITRq1EiSCyEKmVz39Vi0aBGLFy+mdevWHDhwgNDQUEJDQzlw4ABt2rRhyZIlLF68OD9jFSJPVKlSha5du7J+/Xp69OhB69at6dSpE3Z2duruTxkZGZQsWbJAJRc3btxg1qxZnDhxgoyMDE2HI4QoQqKionj27BlxcXHMmDEDgGXLlnH8+HHCw8O5d+8eHh4eREVFYW5uTlpaGsHBwdjb22s4ciFEfsh1C8bq1atp3rw5f/75Z7bldevWZfXq1XTr1o1Vq1YxbNiwPA9SiHcRlxyH/31/9YRd6SHpREVFAZnN+H369HnlPS/Pzv2hCwsLQ6lUUrlyZXx8fLhw4QLR0dF4e3trOjQhRBHw448/sm/fPsaPH0/z5s2pWrUqe/bsITk5mRYtWpCcnEz//v3x8fHB2NiYESNGYGNjg729PS9evNB0+EKIfJDrBCMkJIRBgwa9dn3z5s2ZPHlyngQlRF7yv+9PZEJk5osUUBgrsLa2JiEhAWNjY/V2Ba0P8J49e6hcuTLh4eFcvHgRKysrbG1tqVy5MsePH6dRo0bSXUoIka/++usvHj58yNatWzExMQGgc+fObNmyheTkZJRKJWZmZjRt2pSFCxcyceJEjhw5QrVq1YiPj6dUqVIaPgIhRH7IdRcpS0tLAgMDX7s+KCgIS0vLPAlKiLyUkp6S7bVxMWNGjRqVLbkAClRyAVC2bFnKlCmDtbU1P//8M4cPH6Zx48YUK1YMfX19UlNTNR2iEKKQWrNmDTt37sTAwIAKFSpgYmJCaGgoqampWFtbc/PmTSIjI5k3bx4PHjzA29sbS0tLUlJS8PLyonjx4jg5OaGvr6/pQxFC5INcJxitW7dmxYoVrF27NtskYyqVinXr1rFy5UratGmTL0EK8S70dbL/ATPUMwQocJPlRUdHc+XKFfXr8uXLs27dOi5dusTEiRPVfZkdHR05c+YMcXFxGopUCFFYZX1vxsbG4ubmBsCVK1fo0KEDXbt2Zfz48ejr61OnTh2+/PJLXF1dWbFiBbGxsQwePFgSCiGKiFx3kfr66685d+4cI0aMYOrUqZQrVw6A4OBgIiMjqVKlCl999VW+BSrE2/J08uTY/WPqGgxPJ0+g4LRYJCUlYWhoyJEjR0hMTKRy5cps3rwZU1NTrK2tefz4MZ9++iljxozh4cOH2NnZ0aZNG4KCgnB2dtZ0+EKIQiTrezMyMpLTp0/ToUMHqlWrxpUrV+jYsSOtWrUiICCAUqVKsWXLFr766is6duyofn9B64oqhHg7uU4wLCwsOHLkCKtWrco2D0a1atVo0aIFffr0kScT4oNkamBKW9e2mg7jraxcuZJ169axcOFCypQpw969e9HV1eXRo0cMGTKEO3fuEBoaypMnT3B3d2fixIlUrFiRtLQ0GjZsqOnwhRAFXFxcHLGxsa+M9lSnTh0yMjJQKpU4OTmp58Jq1qwZCoWCfv368ezZM/X2WYmFJBdCFA1vNA+Gnp4egwYN+s9ibyFE3vj99985e/YsCxYsoGLFiuplkyZNwt7eHhMTE2xsbAgNDUVXV5d27dpx7949BgwYQGBgIGZmZvK0UAjxTp4+fcq2bdvw8fEhKiqKatWqoa2tTWxsLE+fPkVLS4u0tDROnDjB2rVriY6OxsfHB4UiczANSSyEKJpyXYMhhMhfKpWKtLQ01q5dy5MnT7h06RKTJk2icuXKPH/+HIAhQ4awbt06rl27RlBQECVLlkRHR4e9e/diaGjIl19+ibW1NR4eHhQrVkz+qAsh3tjL8+i4uLiwb98+WrZsmW2bdu3acebMGUJCQtDV1eXy5cvUr1+fv//+m9KlS6u3k+8gIYqm17Zg+Pr6olAoWLBgAdra2vj6+v7PnSkUChYuXJinAQpRVCgUCnR1dbl16xbNmjXDzs6O4cOHU65cOe7cuUO3bt3o0qULo0aN4uHDh+zbt4/bt2/z3XffYW5urunwhRCFRNY8QNevX6dEiRIMHDiQXbt2UbNmTQDS09MxMzOjS5cubNiwgWbNmvH555+r35+RkVGg5hISQuS91yYY/v7+aGlpoVQq0dbWxt/f/38+iZAnFUK8udTUVPT09FAqlSQmJmJgYICVlRXjxo2jQoUKPHz4kJYtW3LkyBH8/f2xs7PDwcGBzp07k5iYiJGRESDFk0KIt/Pyd5CWlhYhISEMHToUS0tLypYty4gRI4iLi+P8+fPUrVsXHZ3MW4du3bpx+vRpnj59SnJyMnp6emhpaUlyIYR4fYJx/fr1/3wthHg3SUlJ7Nu3j+LFi9OwYUO0tLQwMTEhJCSEPXv24O3tTbt27dR/rP38/KhcuTKpqanqOWeMjIzUNwWSXAgh3tSuXbs4c+YMM2bMQEsrs9f02bNnad++PUOGDOGvv/5i48aN2Nvbc/bsWc6cOUOvXr2wsrICwN3dXZPhCyE+ULmuwQgLCyMpKem165OSkggLC8uToIQozNLT01mzZg0JCQk8fvxYPSFeWloaAD169CAoKAilUklGRga3bt1i7NixPHnyBB0dHcqXL4+1tbV6f1k3BUII8aYcHR3R1dXl8ePH6mU6OjoolUogs7uUnZ0dNWrUIDw8XN3CKoQQ/yXXdybVq1dn165dr12/d+9eqlevnidBCVGY6ejocPXqVfz9/fHy8mLLli0A6OrqAplDQhsaGnL9+nUSExOZP38+Dg4OrF69Gjs7O02GLoQo4P49waiJiQlmZmY8evRIvUxHRwcrKysyMjJ48OABL168wNHRkenTpzNw4MD3HbIQogDKdYLxv2Y9Tk9Ply4aQryGUqnM9js0evRoNmzYgIuLC3p6euzevVu9rnz58tjb23P48GFUKhW//fYbo0aNAgre7ONCiA+DSqXKsU7L0dGRlJQU9dxWAHXr1iU5ORlvb28uXbpEkyZNgMw6y6z9CCHEf3mjvhWvSyBiY2M5dOhQtm4bQgiIjo4GUNdI3LhxgzNnzlC6dGlatGjBnj17GDZsGAcOHFC/x9TUlJYtW2JjY6Nu3cj6gy5JvBDibWTNRREXF8eCBQu4c+eOep27uzubN28G4NmzZ5w4cYIOHTqwcOFC1qxZk63lVOa0EELkxn9OtPf9998zZ84cIPNL5X9Nsjd48OC8jU6IAioqKooffviBx48fM3/+fCwsLPjxxx/ZuHEjnTt35sqVK9SsWZNr165Rp04dnJ2duXTpErVq1QIyuyj07t1bvT/5gy6EeFNZA0BkWbFiBWfPnkWlUmFqakqFChUA8PT05Pfff6dDhw7o6OjQtm1bLCwssLCwAGTYWSHEm/vPBKNmzZp8+umnqFQqVq1aRaNGjShXrly2bRQKBUZGRtSsWZMOHTrkZ6xCFAiPHz9m7NixeHh40KdPH7S1tYmPj8fNzY0OHTpw+PBhIiIiSE5ORqVSkZKSgpGREcbGxtn2I8POCiHeRlY3ppeTi5iYGM6dO8eMGTO4ceMGQUFBREdHqwu2Bw0axKpVq5gxYwY2NjbZ9ifJhRDiTf1ngtGiRQtatGgBQEpKCv3796dOnTrvJTAhCqqwsDCKFSvGiBEjsi2vUKECo0aNwtzcnIyMDBwdHdmwYQMtWrTgk08+QU9PL9v2klwIId7Eo0ePePz4MXXr1kWhUHDnzh2mT59Ov379qFSpEg4ODujo6ODg4MDatWvp2LEjkJmQeHh44OHhAWS2fEhXKCHEu/jPBONlixcvzs84hCg0wsPD1SOqpaSkoK+vD8CLFy+oWbMmw4cPp0uXLgQHBzNnzhxMTU01Ga4QopBQqVTq0ej+/PNPDhw4gLu7O4GBgbx48YKMjAy0tLSwt7cnJCSEU6dO4e3tna219N/dqoQQ4m3kOsHI8uTJE65evUpsbKx6nOyX9ejRI08CE6IgyKlvskKhYN++fQwcOFDdKpGens7t27e5c+cO3t7ejBo1Sj0yixBCvK3k5GQMDAwAsLOzIzU1latXr2JgYICRkRG+vr4sW7YMBwcHnj9/zrRp07h8+TIff/wxO3fupF27dtkSCkkuhBB5IdcJRmpqKsOHD2fLli3q5tOcRraRBEMUJTn1Tfb29mbPnj3s3r2bNm3aoFQq2bBhA23btsXa2prq1aurbwikzkII8bYWLlyIubk5n3zyCRcuXCA2NpbQ0FDs7e1xcHDA1dWV1NRUypUrx8qVK/nll1+IiYlBW1uba9eucffuXfn+EULki1w/qpg5cyZbtmzhyy+/ZNeuXahUKpYsWcLWrVvx8vKiatWqnDx5Mj9jFeKDc+zYMVq1asWyZcs4f/68enn37t2ZNWsWO3bsoEePHhw7dgwAKysrDAwM1K1/8sddCPGm0tLSAGjatCn79u0DYN++fZQtW5ZixYoRHByMi4sLMTExxMfH4+XlhYmJCc+ePQPgp59+Ys6cOVSpUkVjxyCEKNxynWBs2bIFHx8fxo0bh6urKwAlS5akcePGbNq0CSMjI1auXJlvgQqhaf/uEhgVFcXevXtZvHgxtra2rFu3Tr2ucePGTJ8+ndDQUD799FOWLVumHvIRpBuCEOLt6erqkpiYSFpaGlZWVkyYMAEAJycnmjZtytGjR4HMBxi7du0CYNasWRQvXhwLCwu8vLzYvn07devW1dQhCCEKuVzf5URERFCvXj0gc4x+yOz7CZlfYu3bt2fHjh35EKIQHwYtLS3S09NZv349jx8/RktLi4cPH7Jz506WLVumHnEtIyMDgCZNmjBixAhatWoFyCzcQoi38++HG1evXqVz587cuXOHQYMGce7cOQwNDYmIiMDExITWrVuzYMECvvjiC3x8fNTvS09PB+Cjjz56r/ELIYqeXNdgFCtWjJiYGCBzpmFDQ0NCQkLU69PS0khISMjr+ITQmKzRVLLqJPbs2cORI0d48uQJQUFBfPzxxxgbG/P06VN27NjBhQsXshVcZsl6v3SHEkK8jawWz8DAQHXXJ0tLS7y9vVEqlVStWpXo6Gi+/vprtLS0+Omnn4iOjsbQ0BD45zso6+GgEELkt1x/21StWlXdx1yhUPDRRx+xePFiqlWrhlKp5LfffqNq1ar5FqgQ79PLk1RlZGSgo6PD2rVrqVSpEj/88AN//fUXz549o2LFipiamtK/f38SExNZsGDBKwmGJBZCiDcREhLCnj17qFu3LnXr1uXRo0fMmDGDq1evMmHCBIoVK0aVKlVISkpCT0+PEiVK0KFDBypXrsz169fR09PD1tZWHm4IITQm112k+vbti0qlUneLmj59OgkJCbRp04a2bduSmJjId999l2+BCvE+KRQKoqKi6N+/P99++y1Hjx5l+PDhvHjxAqVSqU5ABg0ahI2NDe3bt2fDhg2UKFFC06ELIQqwZcuW4evrS2RkJN9//z2nT58mMDCQli1bsnbtWvbt24eTkxPPnz/n7t27GBkZ4ezsTFRUFEC2B32SWAghNCXXLRitW7emdevW6tcVK1bk0qVLHD9+HG1tberXr5+tiFWIguTp06fZkoOMjAxWrFhB9erVadmyJYsWLaJBgwa4uroSExODi4sLS5YsoU2bNnh7e6vfJ8POCiHexdmzZ/H29mbw4MHs2LGDv/76i59++okTJ04wceJEbt26xdGjR+ncuTM//fQT69aty1ZnIYQQH4JctWAkJycze/Zsjhw5km25mZkZbdq0oWXLlpJcfKCynraL19u5c6d6pvodO3Zw5coVFAoF4eHheHp6UqFCBapXr46pqSlXr14lOjqa2rVrM2/evFf2JcmFEOJdjBgxgk2bNgHg7OyMg4MDABcvXmTx4sWsWrWKpUuXUrduXWbPng3IABJCiA9PrlowDAwMmD9/PnPmzMnveEQeermO4ObNm7i4uKCrq6vhqD4Mly5d4s8//6R69er07duXtWvXMn78eIKCgqhWrRo3b96kdu3a6vN369YtWrVqhUqlwsDAAJVKhZmZmboQXAgh8kL16tWpW7cuS5cu5cGDB9jZ2QFw/Phxbty4gZmZGdOnT0dbWxt7e3tpNRVCfJDeqMg7ODg4P2MReeTlwr6oqCh+/PFHDh48SJcuXWjUqBHu7u6aDlGj1q1bx59//kn//v3ZvHkzurq6jBw5kqlTp3LgwAHOnz9PSEgIERERHDt2jIULFxIfH49SqczWTRBkPgshRN6rW7cu3377Lb169WLo0KFA5mS39+7dUw97nUWSCyHEhyjXCcbXX39N3759cXd3V4/3Lz4sGRkZaGtrZ/uD88svv3Dz5k3Onz/Prl27ePjwYZF/4hUVFUWdOnXo2rUrNWrUYMyYMfz66684OzsTFhaGnp4ez549o1evXkRERKCtrc3AgQOl9UcIkW/ikuPwv+9PSnoKseaxtGzdksGDBwOZD43Kly9P+fLlAaTlVAjxwct1gvHzzz9jYWFBjx49KFWqFI6OjuoxtrMoFAo2btyY50GK/3bp0iXs7OywsbEBYNu2bQQGBvLJJ59Qs2ZNjIyMgMzrExgYWOSSi8TERIyMjEhLS0NXV5d+/frRvHlzBg8ejIuLC+7u7igUCmrXrs3Tp0+pU6cOc+bMoWnTptn+qBf1xEwIkX/87/sTmRCZ+cIMzgecJzQ0lKpVq77yvSPJhRDiQ5frBOP27dsoFAp1f9AHDx68so3cfL1/iYmJHD16FBsbG3r37s0vv/zCmTNnqFGjBrt37wYy/xilpKRgbm6urh8oKtfq+fPn+Pn50alTJ3ULhImJCT169GDChAmUKFECExMTSpcuTWJiIrdu3aJOnTosWLCA4sWLZ9tXUTlnQoj3LyU9Rf1vhUJB34l9ZW4pIUSBlesE4/r16/kZh8gllUpFamoqmzdvplevXmhra+Ps7Ex8fLx6JJEvvviCqlWrsmjRIlq2bMmPP/7I8OHDiYyM5PPPPy8SN8pZXQgsLCy4ffs2AwYMoGHDhvTt2xeAtm3bsmHDBqpXr84XX3wBQNOmTXn69CnAK8mFEELkJ30dffgnx8DK0kpzwQghxDuSdtYCRqFQoK+vzy+//IK/vz/6+vqUKlWKCxcuEBsbS0REBJaWlmhpaREaGkpkZCT9+/cnOTmZbdu20ahRI00fwnuR1YUgNjaW7du3Y2hoSKdOndTrTUxMaN++PR4eHkBm/UqlSpVo0qSJRuIVhU9SUhIgQ4iK3PF08sTa2BozfTOsja3xdPLUdEhCCPHWct2CIT4MWYXcX375JQcOHEBLS4vFixdz6tQp+vTpg4uLCytWrKBhw4aEhYVhY2ODs7MzVlZWnDlzhvr162v6EPLF48eP0dXVxdraWr3syy+/ZMCAAXz99ddERUVhamqqrsOwtrYmKSmJxMRElEol2traGoxeFCZKpZLbt29jYGBA2bJli0SLoXh3pgamtHVtq+kwhBAiT0gLRgGgVCrV/866Ea5WrRpXrlxh+vTpTJ06lS1btvDzzz/Tp08fatasydatW+nfvz/Ozs4olUo6depEyZIlNXUI+SYuLo6MjAwWLVpEZGQkiYmJnD17FoCaNWuyePFiWrVqxf79+0lNTc02EtSYMWNo3ry5FEyKPJGSksLy5cuJj48nNDSUVatWMXnyZK5cuaLp0IQQQoj3Su6sPmAZGRlAZnef9PR0dZcLAEtLSxo0aMCUKVMoX748tWvXRqFQcPDgQTp06MCiRYvUwwlraWnh6elJmTJlNHIc+SEjI4O5c+cyfPhwMjIycHR05PDhw0RFRTFjxgwAunXrRlpaGqGhobRv35527doxYMAAUlIyOzqbmZlp8hBEIaJUKtHX16d9+/aYm5tjbGzMsmXL8PLyokaNGpoOTwghhHivJMH4AJ08eZInT56oWyvWrFlDjx49mDlzpnobCwsLTE1NuX37NgDp6el88cUXeHr+02/35ZaPwiQpKYkvv/yS58+f891336Gnp8cnn3zCsWPHMDY2xt3dnd9++w2AChUqcO7cObp168aoUaOYO3cu+vr6Gj4CUVgEBgYC/9T8WFpa4uPjg4ODA4MHD+bFixcApKWlaSxGIYQQ4n2TBOMDEhAQQFBQEKVLl8bMzIyMjAw+//xzTp8+zY8//khISAhbt25Vb//xxx+zcuVK0tLS0NHRoWLFiujp6amLSgtr15/ExEQiIyOZOXMm9vb2xMbGYmBgQJcuXfjrr7/o06cPd+7cISAggKtXr5KSkoJKpaJ169ZYWloW2sRLvF83btxgzJgxAOzevZt9+/aho6NDhw4duHz5Mr6+vuzatQuQIY6FEEIULYXzDrSACgwMZNGiRTg6OrJq1SoiIiIwMjLC3NwcBwcHRowYgb+/P5A5Mk2FChVYsWIFurq62UaqKew3M8WKFSMjI4Px48fTv39/xo8fz7Fjx6hWrRoqlYq0tDTs7OyYOXMm/fv3p0+fPtnOSWFNvMT7kdV1sXLlytSuXZuuXbuyadMmLl68yNSpU3FwcOD58+fo6enh6upK3bp1sz0YEEIIIQq7144i1a5duzfemUKhYMeOHe8UUFGzZcsWvL290dHRoWrVqoSGhvL8+XPMzMzYsWMHY8eOZciQIUDmjXXlypWBf2aVrlSpElD4k4p/mzZtGgcOHMDBwQEdHR32799Px44diY2NRaFQqJ8sZylKkwuK/LFq1SrKlClDkyZNiIuLw8zMjB49ejBkyBB+//13DAwMWL16NQkJCTx//hxtbW0GDBhAxYoVcXFx0XT4QgghxHvz2gRDqVS+ckP26NEjQkJCMDc3p0yZMqhUKh48eEBsbCxOTk6ULl063wMubDZu3Eh8fDyffvop1tbWPHr0iODgYNzd3dm5cydKpRJnZ2cGDRrE06dPad++PVA0n8LHJcfhf9+flPQU9HX06darG5YmlqSkpPD5559To0YNYmJiKFGihPo9WRPuSXIh3lbWz5CTkxNr167l6dOnLF68mGHDhlG+fHnatm3L3bt3qVWrFrdu3aJLly4cPHiQx48fU758eVq3bq2u1RBCCCGKgtcmGLt37872+ty5c/j4+LBgwQJ69uyJjk7mW9PT01mzZg1Tp05l6dKl+RttITRlyhS+/vprKlasyB9//MHx48epUKECXbt2RVdXl+joaOrVq8f169cZPnw41apV03TIGuN/35/IhMjMFyngF+iHQ4YDS5YsoUSJEiiVSpo1a5btPUUxERN5Q6VSoVKp1D9Dnp6e7Nq1i8OHD/Pzzz9z/fp17t69i76+PgEBAaSkpJCUlIRCoVAPPiCEEEIURbm++/rqq6/o3r07ffr0UScXADo6Onz66af06NGDKVOm5EuQhVnVqlWxtrZmyJAh9O7dm6NHj3Lw4EEMDQ159uwZ4eHh2Nraoq+vT7Vq1VAqlUV2ZuCU9JRsr6Njovn222/x8vJi5syZMjqUyFMKhQItLS0CAwPZvn07KpWKKlWqULp0aWrUqIGJiQlGRkZUqlSJdevWcfToUXr37o2pqakkF0IIIYq0XM/kfe3aNbp27fra9eXKleOPP/7Ik6CKmtatW1O1alU++ugjAEqXLs3SpUsZM2YM5ubmACxZsoQnT54UysnycktfRx9eyjFsbW3ZunWrejhfqbMQ7yI5OZk//vgDV1dXGjVqhEqlYu7cufj5+WFnZ4eJiQkWFhbY2NgA4OjoyPHjxxk5ciQATZs21WT4QgghxAcj1y0YxYsXZ+/evTmuU6lU7N69GysrqzwLrDCLS45j161dXIi7wK5bu3BycSIlJYXw8HAgs4C5V69e6uQCMhOMopxcAHg6eWJtbI2ZvhnWxtZ4Onmira2tbtGR5EK8raNHj+Lp6cmtW7eYO3cuoaGhKBQKqlevzt69e6lWrRobNmygUaNGHD9+nO+//565c+dSoUIFQJILIYQQ4mW5TjAGDBjAkSNH6Ny5MwcOHCA4OJj79++zf/9+OnfuzLFjxxg4cOAbB7B8+XKqVatGiRIl8PT05NSpU/+5/Y0bN2jdujW2tra4uroye/bs13YZOn36NMWKFcPd3f2N48pPWbUEyapkIhMieaB4wOXLl4mJiQHA1NQUKyurbMdlYGCgoWg/HKYGprR1bUvnqp1p69oWUwNTQBIL8W4yMjIICgqiR48e/PTTT1SsWJGDBw8C0LBhQ3x9fQkKCiIoKIiQkBA6duxIdHQ0M2fOpEWLFhqOXgghhPjw5LqL1KhRo0hKSmL+/Pn4+fmpl6tUKvT09Bg3bhyjRo16ow//+++/mThxIj/++CP169dn+fLldO3alTNnzmBvb//K9nFxcXTs2BEPDw+OHDlCYGAgvr6+GBkZMWLEiGzbxsTEMGTIEDw9PXny5MkbxZXf/l1LkJqRyqJFi7CwsMi2XG6chchfSqUSbW1t7O3tuXLlCuHh4Tx48IAWLVqQkpJCbGwsFSpUYOTIkUybNo3Nmzfz3Xff0a1bN02HLoQQQnywFDExMW9UMRwdHY2fnx9hYWGoVCocHBxo0qTJW3WPatq0KZUrV+bnn39WL6tVqxbt27fnm2++eWX7FStWMHXqVO7evYuhoSEAc+fOZeXKldy8eTPbDXnv3r2pUqUKKpWKHTt2cPr06TeOL7/surXrn9GQAGtja9q6ttVgROJ9CQwMlDkRNEipVLJixQrs7Oxo1apVtrodHx8foqOjcXFxoXTp0tjY2NC1a1cGDRpEdHQ0w4cPp0OHDm/1uXLdix655kWPXPOiSa57znLVgpGUlES3bt3w8fGhd+/edO7c+Z0/ODU1lStXrrzS8uDl5cXZs2dzfM+5c+dwd3dXJxeQmaR89913hIaG4ujoCGR2u4qIiGD8+PHMmTPnnWPNa55Onhy7f4z4xHhMjUzxdPLUdEhCFHqHDh3ixx9/xMHBgb///htXV1f1dwbA0KFDWbFiBYsXLyY+Pp7WrVvj6urK0qVL0dbWxszMTHPBv2fh4eFcvnyZWrVqUaJECRlAQQghxBvJVQ2GoaEhV69eJSMjI88+OCoqioyMDKytrbMtt7a2JiIiIsf3RERE5Lh91jrIrNGYPXs2v/32m3p0oQ9NVi1BHbM62WoJhBD5Iz09nXXr1tGnTx+WLl1Ko0aNOHDgAJD54GLAgAHUqVMHW1tb/Pz8MDU1Zfbs2dja2mJpaVmkkovLly/j4+PDsWPH1N1eJbkQQgjxJnJdg9GgQQNOnTpF37598zSAf//h+l9PynLaPmt5SkoKn332GdOnT8/2ZDI3NDXTrszwW/TINX+/smbidnV15cCBA9SpU4dTp07RrFkzzp49y6FDh3B2dubJkydYWFhw6dIl7OzssLa2RqVS5dn1+tCve1paGrq6ugQGBtKsWTN8fHyYPHkyCxcupEWLFurzKHLvQ7/mIu/JNS+aNHHdP/RuWblOMGbPnk2nTp346quv+Oyzz3BwcHinPzbFihVDW1v7ldaKZ8+evdJKkcXGxibH7SGzJSM8PJzbt2/j6+uLr68vgHpiumLFirFp0ya8vLxy3LcmLpT02yt65Jq/HxkZGa+0YI4fPx5PT0/69OlDu3bt0NfXZ+HChfz555/qbXr16vXGDydy40O+7gEBAfzwww9UqlSJnj17oq2tjY6ODi4uLowbN45vvvmGYcOGSXLxhj7kay7yh1zzokmue85ynWDUrVsXlUrFokWLWLRoEVpaWujq6mbbRqFQ8Pjx41ztT09Pjxo1auDn55etcNLPzw9vb+8c3+Pm5sbUqVNJTk5WD9vq5+dHyZIlKVOmDOnp6a8Mc7tixQr8/PxYs2YNDg4OuT1cIUQBlpVc3Llzh5IlS6q7OA0bNowLFy4wadIk0tLSePToESkpKejo6KCtrZ0vycWHbtOmTTRu3BhjY2OmT5/ODz/8wO+//05ERARubm7UqVMHf39/GjdurOlQhRBCFBC5TjA6duyY5/1wfX19GTx4MLVr16ZevXqsXLmS8PBw+vXrB2ROOHfx4kV27NgBQJcuXZg9ezbDhg1j3LhxBAUF8dNPP/HFF1+gUCjQ1dWlUqVK2T6jePHi6Ovrv7JcCFG4vNy9MiwsjGHDhlG5cmXGjRun3sbFxYW4uDj279/Ppk2bcHZ2Rl9fX1Mha8yFCxcoV64cJiYmREVFMWnSJPT19Vm/fj2nTp2ie/fubN26lcGDB6Ovr4+dnZ2mQxZCCFGA5DrBWLJkSZ5/eKdOnYiOjmbu3Lk8ffoUV1dXNm7cqG5pCA8P5/79++rtzc3N2bp1K+PGjaNJkyZYWFjg6+vL8OHD8zw2IUTBolAoSEtL4+7du1y+fJn+/fvTsWPHbNtYW1tz9uxZNmzYwODBg4vcfBbR0dF0794de3t7tLW1+f7777G0tGTLli307NmTMWPGsGXLFgYOHMiCBQtYt24d169fx8fHR9OhCyGEKEDeeB4MkXek317RI9c8f23cuBETExPu3bvHs2fPePDgAdbW1kyaNEk9keXFixepWrUqenp67y2uD+W637p1iwMHDjBq1Ch+++03AgMD6dWrFzt37mTMmDE8efKE3bt3M3r0aI4cOcKOHTvo3LkzDRs21HToBc6Hcs3F+yPXvGiS656zXLdgQOZM2r/88gsHDhzgwYMHADg4ONCiRQuGDx9epIZyFEJ8GIKCgrhz5w5t2rShWLFinD59milTphAbG4u5uTkjR47k+vXr6pvk2rVrazji9ycxMZGZM2fi6OhIs2bNSEhI4MKFCwD07duX0aNHc+vWLWxsbLh79y6mpqY8ffoUyJyT6HWDYgghhBD/JdfDgoSHh9OoUSN++OEHkpKS+Oijj/Dw8CApKYm5c+fi6elJeHh4fsYqhCjicpqLJzU1ldOnT7Nt2zb09fWpVq0akNkdaNCgQURERGBvb/++Q9W44OBg+vfvT0ZGBnp6eixduhQ7OzvKlSvH9evX0dfXp0GDBpQqVYqKFSvyyy+/MGDAAKpXrw78MwS4EEII8aZy3YIxdepUnj59ytq1a2ndunW2dXv37qV///58++23LF68OM+DFEIUbVlzMGhra5OamsqzZ88oVaoUAJUqVaJTp05cv36dvXv3UqJECby9vQkPD6dBgwb06dNHw9FrhpOTE59//jlubm5cuXKF0NBQTE1NKVasGGFhYVStWlU954enpyfFihXD0NCQcuXKATK5nhBCiLeX6xaMw4cPM2jQoFeSC4BWrVoxcOBA9cy4QgiRF7KeomfNwXD8+HHatGnDrFmz2LNnj3q76tWr07hxY4KDg4mPj0elUuHu7l6kkovU1FTS09OBzJYehUJBnTp1ANDX1+fWrVsYGxtTunRpgoKCOHToEGlpaZibmwNQpUoVdXIhhBBCvItct2DEx8f/51CFdnZ2vHjxIk+CEkKIf0+Wt379ev744w/WrVvH06dP2bx5Mx9//DF6enpoaWlRpkwZxo8fj52dXZF7+p6RkcGVK1ewtLTExcVFfd5eTsyyko1mzZoREBDA/Pnz+eijj6hRo4amwhZCCFFI5boFo1y5cuzYsQOlUvnKOqVSyc6dO+XplxDinSQlJXH48GFevHiBtrY2Dx48YNGiRdy4cYMOHToQGxtLeno6JUqUwNbWlpCQEOCf7jzNmzcvUnPeZH0fKxQKYmNj+eGHHxg9ejSPHj3Ktl1iYiLdu3dn9+7dPHnyBHd3d9auXcuoUaM0EbYQQohCLtcJxuDBgzlx4gQdO3Zk//79BAcHExwczL59++jUqRMnT55kyJAh+RmrEKKQCwgIYM+ePZiYmHD//n0GDBhAcnIyP/30E/v27WP8+PGcPn0aCwsLQkNDSUpK0nTIGqFSqVCpVOoWCi0tLQIDAzl+/DiDBw+mdOnS2bbfsmULn332GX/99Zd6eF5dXd33HrcQQoiiIdddpPr06UNUVBSzZ8/m+PHj6uUqlQp9fX2+/vprPvnkk3wJUghROCmVSh4+fMikSZMYO3YsdevWZdasWdy/f5+QkBAaNWrE2LFjuXnzJr///jtNmjTh2bNn6Orq0qFDhyI79nhWi010dDTz5s2jfPnyDBw4kKCgIAwNDYHMmgw9PT3u3btHQkICU6dOpWnTppoMWwghRBHxRvNgjBkzhr59+3L06NFs82A0btwYKyurfAlQCFE4BQcHU7ZsWRwcHAgODmbTpk1YWlrSqVMnYmNjsba2Vhd5Z2RkUKlSJSwsLAgPDyc9PZ169epp+Ajer3/XpGzdupUbN25gb2/PmTNnsLW1pW3btixatIi5c+eqWyrKlSvHpUuXNBW2EEKIIuiNEgwAKysrOnXqlB+xCCGKkOnTp+Pm5sbQoUNp164d9vb2HDx4kJs3b1KlShUsLCywsLBg27ZtnDx5EhsbG/r164eHh4emQ9eIrOQiJCQER0dHbt++zb1795gyZQo3btxg48aNTJs2jbVr11KjRg3mzZsnE+UJIYTQiFzXYDRv3pypU6eyf/9+YmJi8jEkIURhllWYPHnyZA4dOkRCQgIvXrxAqVRSvXp1wsPD2b9/P3Z2djRt2pQ9e/ZQrFgxxo4dq+HI37+XJ7t79OgR3t7ezJw5k++++46xY8cSHR3NkydPsLOzw9TUlMePHzN16lRWrVolyYUQQgiNyXULhqmpKStXrmTBggVoaWnh6uqKh4eH+j8bG5v8jFMIUUhkFSY7OztTr1491qxZQ6tWrbh06ZK6q2VsbCyQOYneokWLilxBskqlQqFQoFAo1F2j7t69y5AhQ2jdujVubm40b96cbt26cfz4cdq3b4+pqSkqlQp7e/siOXO5EEKID0euWzC2bNlCSEgIR44cYdq0aZQpU4YtW7bQr18/KlasSO3atRk5cmR+xiqEKICynsIrlUoiIyPZtm2bet3IkSO5evUqBw8exM7ODmdnZ8qXL8+NGzfU2xSl5OLlYWcBfvvtN8aPH094eDi3bt3i119/pXv37nz66adUqVIFU1NTkpKS0NXVzXH0KCGEEEITFDExMar/vVnOMjIy2LhxI/PnzycwMBCFQkF0dHRexleoBQYGFohRcOKS4/C/709Kegr6Ovp4OnliamCq6bAKpIJyzfOCUqlUt1ZkPYW/ffs2u3fvZuzYsaSnp6Ojo8PZs2f57rvv0NbWZuvWrRqOOn+86XW/cOECYWFhhIWFoaOjw7Vr15gwYQIDBw5k+fLlJCcnc+vWLcqVK4ejoyNmZmb5GL14G0Xpd11kkmteNMl1z9kbFXmnpaVx8eJFTp06xenTpzl79iwvXrygdOnSdO3atcgWXxZ2/vf9iUyIzHyRAsfuH6Ota1vNBiU+eC/PIn348GF69+7N06dPX1lfr1492rZti46OjvoJfta6ouDlRCwqKop58+YRHBzM8+fPmT17NtWrV2f48OEEBAQwYMAAvvzyS5KSkhg/fjzVqlXTcPRCCCHEq3KdYLRp04ZLly6RkpKCs7MzHh4ezJ07Fw8PD+nvW8ilpKf852sh4NXJ327fvs2MGTPw8vKiTJkyLFq0iJkzZ3Lw4EFSUlLQ19dX31x/9tln2YZgLQqy6iy0tLTULTwHDhwgIiKC9evXM336dC5evEj16tVp2bIlYWFhDB06FHt7ez766CNNhy+EEEK8Vq4TjFOnTqGlpUXHjh3x9vbGw8MDa2vr/IxNfCD0dfQh5V+vhfiXrKLk2NhYzM3NefjwIaGhofTv3x+AwYMHM336dCpWrIi+vn62ZKQoJRcvF3ADrF27lr///psvvvgCAwMD9QObTz75hIkTJ9K1a1dsbGwwNc3slijJhRBCiA9drvshbNu2jXHjxhEZGcmwYcOoUKECdevWZdSoUfz111+EhYXlZ5xCgzydPLE2tsZM3wxrY2s8nTw1HZL4AKhUKhYtWsS+ffsASElJYfr06XTo0IEDBw5gZWVFixYtuHv3LgATJkzg+vXr7Nu3j5SUFPUNdlGiVCqzHfexY8e4cOECX331Fbt27cLKyoq7d+8SGBiIo6Mj1atXJyAgADc3Nzw95fdOCCFEwfBWRd7p6elcunSJ06dPc+rUKc6dO0dsbCx2dnZcu3YtP+IslKQwqOgpTNf8xo0bdO/eHW9vb2bMmMHy5cvJyMigVq1anDhxAnd3d06ePEnHjh0pV64cANevX8fJyQkTExMNR/9+vXzdw8LCWLZsGXXr1kVfX58HDx7w+PFjTp48yebNm9m1axeHDx9m+fLlxMTEYGFhodngxVspTL/rInfkmhdNct1z9laVlDo6OhgbG2NkZIShoSG6urqoVCoePnyY1/EJIT4gd+/eZerUqWRkZFC5cmUaNmyInp4egYGBDBgwgKZNm7JixQqWL1+OtrY2RkZG2UaGqlq1apFKLrKK1rPExcXx+eefY2trS7t27Xj27Bm///47jRs3Zvfu3QQFBeHj40O3bt0AJLkQQghRIOW6BuPy5cucOHGCU6dOcebMGWJjY1GpVDg5OdG8eXPc3d2lb/AHQIaUFflp165dbN68mVq1atG4cWOSk5OxsrIiOjoahULB5s2bGTRoEF5eXixbtoyFCxdy/vx5TYetES/XmNy7dw8bGxvCwsKoVq0aw4YNA6Bly5asW7eOyMhIOnbsSKNGjahSpQrNmzfXZOhCCCHEO8l1guHl5YVCocDV1ZUuXbqoZ/AuUaJEfsYn3pAMKSvyQ9ZoTx06dGD79u3s2rULd3d3nJ2defbsGUePHqVmzZo8fvyYXbt2ER8fj5ubG/r6+jRo0EDT4b9XWedKoVAQEhLCwoULuX79OqdOneLzzz8nLCxMPQeIlZUVixcv5uDBg0yZMoV69eppOnwhhBDineU6wVi3bh3u7u7SZP+BkyFlRX7Q0tJCpVJRtmxZ2rdvz9mzZ9m6dSuXL19m2bJl9O3blz59+jB48GCOHDnC0KFDsbGx0XTYGqGlpUVycjLJyckMHDiQsWPHMnjwYObPn8+FCxfw8PDg559/JjIykhcvXvDLL7/w2WefaTpsIYQQIs/kugajVatWklwUAP8eQlaGlBXvKi4uDsicjRugdevWODs7U6FCBezt7VGpVLRu3ZpNmzZRpUoVRo4cWWSTC8gcBOPTTz9FS0uLihUrcvv2bQDatm1LcHAwzZs359GjR5QsWZKff/5Zw9EKIYQQee+tp8uNiIjAysqKY8eO5WU84h3JkLLiXWQlEVmCg4OZNWsWkDm4A6Ceq8HExIRp06ZhaWnJ4MGDGTVq1HuPV5OUSiUq1T+D8J0/f54LFy6go6NDkyZN8PPzY/z48dy5c4fU1FSMjIxITU2lRIkSzJkzh5EjRxbJoXqFEEIUfm+dYADZ/riKD4OpgSltXdvSuWpn2rq2lQJvkSt37twB/pnwbtWqVdy9exdHR0d1gpGlZMmSPHr0iLi4OPXkb0XxuyCrziIqKgqA+Ph4lixZAkCtWrWIiIhAR0eHihUrEhQURI0aNfD29kZbW7tITSwohBCi6Ml1DYYQovB5/PgxOjo6jBgxgh07dpCWlsaECRPUNReHDx9m6NCh6u1VKhX6+vp06NBBPbcFUGSexGfNwg2QlpbG+PHjuX//PlWrVuXTTz+lRo0a3L17l+LFi3P9+nXat2+PtbU1ZmZmWFhYSDdTIYQQRcI7tWAIIQqm1NRUnj9/zsGDB7GxsaFly5bs37+f1NRUSpUqxYwZMzh16hTp6ekAREdHk5CQoL65rl27dpG6Wc6az+LlROrmzZsYGBiwfft2XFxcWLRoEaVKleLevXs8ePCAS5cukZCQQM+ePbGzs9NU6EIIIcR799YtGGZmZixatIiKFSvmZTxC/E8y18e7mTVrFhEREcyfP5+ePXsSFRWFu7s7AQEBPHr0iEuXLtGzZ098fX1p2bIlAGvXrqVLly4YGxtrOHrNyJrPws/Pj4sXL9K+fXtevHhBYmIiAA0bNuT58+eYm5uzY8cOrKys2LRpEyVLltRk2EIIIYRGvHWCYWBgQM+ePfMyFiFyReb6eDfPnz8nICCAtLQ0tm3bhlKppEKFCqhUKpRKJba2towcOZKwsDC6dOnCggUL8PX1Vd9kFxUZGRloa2ujUqmIj4/nl19+ITExkVKlSjF//nz69OlD7dq1SUxMJDAwEC0tLT7++GPq169fpGYrF0IIIf4t13cMJ0+e5Ndff822bNOmTdSpUwdnZ2cmTJig7kYgRH6SuT7eTEREBA0bNlSP+FaiRAmioqLYunUrdevW5ezZs9SoUYOgoCCsra3p3Lkza9euZeXKlcyaNQsHB4cilVxkfY9pa2uTlpZGUlIShoaGrF+/nrZt2+Lr60udOnUICgrCwMCAAQMG8NNPP+Hs7AwgyYUQQogiL9ctGLNnz6ZYsWIMGTIEgLt37zJs2DCcnJyoWbMmy5Yto0yZMgwbNizfghUC/n9uj5R/vRavZWNjg4GBAWvXrkVXV5emTZtiZ2fH5s2b1aMaHTlyhCZNmjBt2jR+++03GjRogL5+0Tqv8fHxmJqaqpOpI0eO8M0339CoUSMGDRrE5MmTOX78OO7u7hgaGmJhYUGrVq2wsbGhVq1amJuba/gIhBBCiA9Drh9L3r59m9q1a6tfb9y4EUNDQw4dOsSmTf/H3n3HNXW9Dxz/JGHvIUMUhSoOXLgV/YJ7K45at3W02jo6raPL2tqqtVNt66jaumrds+6Fe1sHDhQVF4ogRDYk+f3BLykBVNAgoM+7r77aJDc35+Ym4Tz3POc8y+nRoweLFi0qkEYKkVVh1vpQp6jZcH4DK8+sZMP5DTxMefjcXvtp6HQ6UlNT6d27N/7+/mzYsIGff/6ZUqVK0b59e/bu3cuIESNYsGABbdu2pVu3bgAvTXCh0+lISUlh0KBB/PbbbyQkJACZy/ROmTKFRYsW0aRJE3bu3Im/vz82NjYkJCRga2vL/fv3AWjatKkEFy+JLVu2EBkZWdjNEEKIIi/PAYZarTZaNWbHjh00bdoUBwcHABo2bCg/vOK5KMxaH/r5H+pUNdGJ0ey5WrQKTc6fP5/hw4ezYMECIHPVI0tLSxwcHHBxcaFTp05ERERw8eJFXFxciIyMxM7Ojj59+pCRkUHr1q0L+QieL4VCgZWVFSVKlODatWuGytpBQUEolUrc3d1xcnLCzMwMnU7H/fv3iY+Pp1OnTvTr16+QWy+el8TERACio6OZMWNGIbdGCCGKvjwHGB4eHoZiXHfu3OH06dM0a9bM8LharZbiUeKFV5Tnfxw4cIBdu3YxcOBATpw4QWhoqOGx//3vf2zcuJGqVavSv39/GjRogL29PRYWFtjb29OyZUtDpe6XTXJyMg4ODnz77bdERERw6NAhXnnlFdq1a8eFCxfw8/Nj69at+Pj4EBISQqlSpQq7yeI5iYuLIyEhgSVLlgDQt29f7OzsWLVq1UtZXFIIIfIqzwFGx44dmTNnDqNHj6Zfv35YWlrStm1bw+Nnz57Fx8enINooRJGRfb5HUZr/ce7cOSpVqkSdOnWoU6cOVatWBTLTgNzd3alfvz7fffcdgwcPpkqVKjRp0oR+/fphYWFRyC1/fnLrFFpbW2Nvb49araZs2bJ8+eWXPHz4ECsrK6Kjo3FycqJXr15YWFhQs2bNQmi1KAxRUVEsXboUOzs7jh07RpMmTVi8eDHdunVj8+bNpKYWnYsLQghR1OQ5wBg3bhydOnVi2bJl3L17lxkzZuDu7g5kjl6sX7+epk2bFlhDxYurOM1rKMz5H0/y6quvsnfvXkaPHs20adPYuHEj9+7dMxSHGzJkCBEREdy7dw/IvbP9ItNqtTkqjmu1WnQ6HU5OTrz77rvExMRQvnx51q1bR3R0NDdu3ACgXbt22NjYFEazRSHx9PRkyJAh3L9/H5VKhbu7O507d6ZKlSpYWloaRjWEEELklOecCFtbW2bPnp3rY3Z2doSFhckfYPFUilNdC/38j6LI2dmZcePGsXr1ambPns2ff/5J6dKlcXd3R6vVYmdnx6xZs7C2tgbI0dl+0SmVSqKjo/n7779p1qwZ/v7+hhWjEhMTqVatGp9//rkhJaZ3794yKvsSCQ8P58iRI/Tp0wfInG+xbds2XF1defvttzlx4gS3bt2iQoUK9OnTh7NnzxZyi4UQougyyeL2SqUSR0dHzM3NTbE78ZIpyvMaihu1Ws3//vc/AgICKFOmjGGlI31HWh9cvAw0Go3R7atXr/LWW2+hUqlYvnw5K1euNDxWq1YtGjRoAGReMBkyZIgEFy+Jq1evEhYWhqurKwsXLkStVrNs2TI2b96Mra0thw8fplq1aty5cwc7OzuSkpI4efIkVlZWhd10IYQosl6e6lmiyCrK8xqKg6wpZrdVt9m+czsffvghe/fuxc/Pr7Cb99xdv34dwLDoxIYNGwgLCyM+Pp5mzZrRoUMHzp49a1QY9MCBAyQnJz/V68XGxhIVFfXsDRfPVWpqKhMnTmTQoEFcvXoVFxcXOnfuzMKFC7l9+zbt27cnJCSEixcvcuHCBapXr87YsWMZNmwYdnZ22NravnRphkIIkVcv57IxokgJ9g1mz9U9pGakYmlmWWTmNahT1IReDTVq1/NcEjevsqaYmbuZE9A+AKtoK7755puXpp6FXmxsLHv27OHVV19Fp9Px9ddfExYWxueff45SqWTevHkcPnyYKVOmULp0acPz3nnnnXyvoqXRaNDpdOzYsYMrV67QqlUratWqZepDEgUgPT2dAwcOEBcXx65duwz3t2rVisWLF3Pz5k3D4geDBg1i+vTp/PLLL/j4+FCxYkXUajUODg4vXZqhEELklQQYotBln9egvyJf2B374jI3JHtKmWtpV7q17VZIrSkcOp0OhUKBi4sL/fr1Iz4+HjMzMzQaDStWrMDMzIwzZ87QsWNHOnTowLZt29i5cydz5szBwcEh38FFWloax44do1q1anTv3p2DBw/yww8/SLHRYmDz5s3cvn0bb29vQ22nf//9Fw8PD1xcXLhx4wYxMTGMGjWKnj178sorr9CiRQuSk5Px9/cHwMnJSYILIYR4DAkwiqHicmX9aRWVjn1e5oYUhXNhaWYJqdluv2T0nb2LFy+yb98+HBwc8PHxwd7e3hA8VKxYkaCgIGbPnk3JkiWZP39+vhemuHPnDqdOnaJWrVo8ePCASZMmYW1tzWeffcaSJUtYvnw53bt3N/nxCdOJiIigWbNmqNVqYmJi6NChA1FRUfj4+PDdd99RrVo16tWrh7W1NQcOHMDNzY0uXboY7UOCCyGEeDwJMIqhotIBLyhFZdJ3XjruReFcFNUUs4Kk0+nQ6XSGyesAJ06c4ODBgzRq1IhFixbRpUsXvvjiC2JjYzl79iwXLlxgwIABNGrU6KlTx1xcXAgODsbGxoYDBw7wzz//MG/ePCBzmeBjx46h1WqN2iWKFmtra7Zv386IESOoXLkyp0+fplGjRrz77rts3bqV6tWr8+eff/Lrr79SvXr1wm6uEEIUS/JXsBgqKh3wglJUJn3npeZFUTgX+hSzbtW60aFyhxdqNCs3+nQopVLJzZs32bRpE3Fxcfj5+XHw4EECAgJISkri+PHjvP/++3z++eeMHz+eChUqYGFhka/gQqfTsWzZMhISEoDM1bguXLjAxIkTeeedd2jUqJFh3kVcXBw2NjYolUqZ/FsE6Guc6On/v2zZsri6uqLRaLC3t6dRo0YAVKhQAVdXV4KDg/nuu+9yPE8IIUTeSYBRDD2pA16cCtflpqgUs8tLx72oBEMvA/2qTwqFAq1Wy8KFC3n//ffZuXMnq1evJjw8nBYtWgCZRQVnzpxJ8+bNmThxIrt27aJJkyb5er3bt2+jUCi4evUqBw4cICoqijFjxuDv709YWBgA/v7+/PbbbwA0bNiQQ4cOGQIgUbiUSiUKhYJjx44RGxtrdE5OnDiBSqVCo9GwceNGXn31Vfbv329YqtjGxsbo8yaEECJ/JMAohp7UAden7ahT1UQnRrPn6p5H7qsoBiPF6Yp8UQmGXnTZ06G2bt3Krl27mDx5MuPHj8fW1pa1a9eyadMmUlJSqFq1Kj179iQjI8MwkTevQkNDadeuHXPnziU1NRU/Pz80Gg3Hjx+nQYMGWFlZ0aVLF7Zv386gQYPYsGEDzZs3Z9++fQQFBUmHtBDoRxmy1j5JTEzko48+4rPPPuPq1auGbZo1a0ZsbCy7d+9GpVKRkZFBr169WLJkCaVKlTI8X9LchBDi6ckcjGLoSdWk85O2UxTmEBRnRbmyd3F36tQpXF1d8fb2RqFQcPjwYf744w9atWpFxYoVAfDy8sLa2pqmTZty/vx5IiIiSElJwcnJidatW+fr9VJTU0lISGDmzJm8//77tGzZEgAHBweOHDmCTqdj8ODBALzyyiscPXoUa2trxo4di5mZGdWrV+fSpUumfRNEnqSnp2NhYWGofXLq1Ck8PT2Ji4vjn3/+QaFQkJGRYZjw//bbb7N27VpSUlJo06aNIW1ORp+EEMI0JMAoph63elF+VhUqCnMIhMgqNjaW7777jr179/LTTz/h7e1NZGQk06ZNY8SIESxevJiyZctSokQJli9fTv/+/XFzc+Ptt9+mevXq+R6xALh27Rr79+/HxcWF+vXr07JlS7RaLSkpKTRr1ozly5eza9cuKlWqhKenJyqVisTEREPldL2aNWua8J0QTxIfH8/x48dRKBQ0bdqUTZs2MXPmTBo0aECXLl2oVKkS/fr1MwSkn332GQB16tTBzs7OUOleH1hIcCGEEKYhAUYx9biRh/ysKiRLnIqiZtu2bdjY2LB3717DfXfu3KFVq1bcvn2bQ4cO0bt3b0aOHMmoUaPo2rUrdnZ2uLu751hONK8iIiK4ffs27dq1Y8KECcTGxrJq1SoCAwPp0qUL77//PtHR0VSoUIGZM2cyaNAghg8fjpWVlakOW+TTb7/9hlarZfjw4aSkpJCRkcGJEyd444036NixIwCVKlUiJSWFf//9l+PHjwP/BROVKlUqzOYLIcQLTQKMYupxIw/5Sdt5GZc4FUWXVqvlxo0bhpGAkydP4ufnh52dHZMnT+btt9/mwIED3LlzB5VKxbBhwwxpMc/Cy8uLiIgInJ2dmTNnDtevX6djx44kJyczY8YMFixYQMmSJXF3d2fq1KmG+hmSUvN8paWlceDAAQIDA4mIiGD48OEkJibyxx9/UL58eby9vTl8+DArV64kICCAOnXqsHv3bnbs2MHYsWMBmbQthBDPgwQYxZSpRh5kDoEobPq6EfqJ3CVKlGD9+vX88ssvJCUloVKpWL58OW3btsXS0pJ3330XKysrxo4dS3Bw/gPi3OpUREREcObMGQCqVatGtWrVADh06BANGzbE0tKSjz/+GA8PD6PnSWf1+Tp58iQnT540TOb28fEhLS2N9PR0nJ2dad26NbGxseh0OmbPnk2ZMmXw8vJi69atmJubF3LrhRDi5SHLZBRTsnqReFEolUpSUlIMnfUBAwbwzTff8MUXX7B7926aNm3KokWL+Oyzz1AoFAQFBfHjjz/m6Ozn5/X09EuRtmvXjtu3bxvSaG7fvs23337L119/jbu7O4DRCkOicPj6+nL37l0OHDhAmTJlALCwsKBChQosWbIEgIMHD/LWW29x7do1PD09GTRoEObm5oZzLYQQouBJgFFMFaelXIXI6uzZsxw7dsxwe+7cubz++uvExMQY7nNxcaFevXoAmJubU7NmTZydnRkyZAi9evXK1+udOnWKlStXGm6vXr2aHj16cPbsWaNlTQcOHMjy5cuBzCvlsbGxzJkzh969ez/VcYpnlzUo0Ol0uLu7U758eebMmcONGzcMn6N27doRHx/Pvn37sLS0ZMiQIcyaNQsLCwvD83NbdjYmJoarV68W/IEIIcRLRgKMYq4o1rEQIjfXrl3j0qVLODk54e/vD8DixYs5cuQIP/zwA66uroZt4+Li2LBhA+3bt+fmzZtUqVIl36+nr3vg5ORkWJJ01qxZbNmyhY4dO7Jx40YSExMN27dr1w6dTseaNWto3749kydPxtPTU658FwL9uVMqlTne/z59+uDv74+Xlxd//fUXY8eOZe3atYwbN46MjAxatGhhWGL4cVW4b9y4QdeuXfn777+fuK0QQoj8kQCjmMtPUT0hCtPixYu5fPkypUuXZtu2baSmpvLgwQOGDx9OqVKl0Ol0hk6eSqUiKSmJUaNGMX36dFxcXPL9evqUK1dXV6pUqcLly5cZMGAA3333Hffv32fmzJmGeRfp6elAZn2E06dPk5ycbNiPFFx7fvTnX3/uZs+eTZ8+fdi6dStqtRoAa2trevbsiUql4vvvv6d3796EhYXh5+eXo1p7bnNkbt26BYCnpyc//fQTMTExMllfCCFMTP5yFnMFUcdCRkWEKeh0OqOrz//73//YuHEjAMuWLUOtVpORkcGpU6eAzMrLDx9mftbs7e157bXXaNq0ab5eU/96+tSnRYsW8cYbb3Dx4kUiIiKwtLTk559/xt3dnQkTJjB9+nQgMw1Lp9Ph4+PDZ599hrW19TMdu8ifu3fvAsYBwY4dO7h27RqTJk3i8OHD7Nnz38WTatWqcefOHW7evEn16tUZN27cEwOEe/fu0apVK1q0aAH8l3pnaWnJrFmzABnFEEIIU5EAo5jLvnqUKepYyKiIeFYRERH88MMPhqv/cXFxBAUFYWFhwZ49e+jTpw8zZsygf//+LFiwgN9++402bdoQFhb2VK+nH/3Qv55+6dp///2XiRMn0rVrV86dO0dycjJXrlzBzc2NmJgYqlSpQkpKitEVbLmS/fxER0dz+PBhBg0aBGQGAdu2bQMyR44UCgULFiwgLCzMKE2uZMmSDBkyhNKlSz/xNbLOsxk7dixt2rThzz//NNw3YsQITp48SVxcnJx7IYQwEQkwirmCWE1KqnuLZ+Xj48ORI0dYs2YNPXr0YPTo0SxdupT33nuPLVu20K5dOx48eEBCQgJff/01ycnJhgrMT0NfhfnatWsMGTKERYsWERUVhb29PX5+frRq1YrNmzcTExNDp06dWLNmDfXq1WP8+PFYWVlJx/I502g0TJ06lR49euDh4UHTpk0ZOHCgoVL7kiVLiI+P599//6VatWr89ddfPHjwwPB8Ly8vfHx8nvg6v/32G5988gk7d+7E3d2dZs2aMXjwYA4dOkRSUhKQOVoWEBDAv//+W1CHK4QQLx2pg1HMFUQdi/zW2FCnqAm9GmpUrE9WtXq5KZVKRowYwfvvv8/06dOpVq0aH330EV5eXpQvX56zZ8/SsGFDTp48SUhICPXr18/3a2SvZ7F161YWL15MYGAghw4dok2bNty6dYvDhw9Tv3597O3tmT9/Pp999hmdO3c2PE/y75+v9evXM3nyZJRKJevXr8fJyYkBAwYwcOBAvv32W6ysrDhw4AD29vZUrVqVpKQkXnvtNSpVqkTNmjXzXFhx//79hIaGMnLkSD799FPDSmQuLi6UL1+eY8eOERQUhEqlwsHBwWjCvxBCiGcjIxgih/yOikhKlcgqIiKCSZMmUa9ePUqVKkVGRgZ2dnbUq1cPMzMz7t69S2pqKsHBwYSEhDz16+iDi6NHjwIQGRmJr68vQ4cOxdfXly1btjBixAi+//57mjRpQuvWrfnoo48Mz88+oVgUvOTkZK5du8bUqVPx9vYmISEBADMzM+rXr090dDTOzs6cOnWKatWqMWLECO7du8eYMWP48ssvnxhc3Lp1i8jISCAzLa9jx44EBgYSHBzMtGnTALCxscHe3h5ra2uio6OxsrJCq9Wybt26gj14IYR4icgIhsghv6MiklL1clq8eDE7duwgKCiIAQMGcOrUKTw9PbGwsMDe3p7bt2/Tt29fbt68CcCFCxfo1KkTDg4O+Pn5YWmZv/lC2edZnDt3jkmTJpGenk7fvn2xtLQkIyMDwFAvo0WLFvz888+kpaVRtmxZw370KVXi+bK2tmbkyJFA5vyYr776ilmzZmFvb4+npyepqak4OTnh4uLCH3/8wejRo3n//feBJweEv/zyC3/99RfVq1enY8eO2NnZsW7dOnr37s2oUaN49dVXuXr1Kr6+vtjb2/Puu+9So0YNPv/8c2JjY586PU8IIUROEmAUcUUh/ehJbchLStXzOo6i8H696NLS0hg7dixxcXG8+eabzJ49m5SUFEqUKMGVK1do3749arWatLQ0PDw8WLRoEadOnUKr1aJSqahatWq+Xi8lJcUwT0KhUHDz5k0ePHhAREQEvXr1okqVKowaNYrPP/+crVu3sm/fPho3bkz//v1Rq9X4+fkB/6VUSWBRNAwdOpQhQ4Zw6tQpAgIC8PX1ZenSpbRs2ZJx48YZbZuXNLaEhASWLVtGamoq77zzDsuXL2fRokUcPHiQhg0b0qpVK27evImnpyd///03jRs35pNPPsHR0ZGRI0fKcsRCCGFC8otaxBWF9KMntSEvKVXP6ziKwvv1orOwsMDKygo/Pz8aNmzI22+/zZkzZ9BqtWi1WqysrHBxcWHOnDnUqVMHGxsbevXqxdSpU3Fycsrz62g0Gr755hvefvttDhw4AMCmTZv44IMPOHfuHJGRkaxatYoPP/yQ9u3bU716dQYMGMDixYsBeO211wzBBUg9i6JGq9USFBTE6dOnAWjevDk1a9YkNTXVMBL1uFGL3bt3M2nSJA4dOoRWq+XkyZNoNBp8fX2pWLEiv/32Gz169GDHjh0A3L59m1KlSmFtbc2cOXP49ttvcXR0zDGXRwghxLOTX9UiriikHz2pDfqUqm7VutGhcoccIwbqFDUxSTGP3cfzaqswjY8++ogdO3aQlJSEubk59evXp0KFCpw8eRKtVkvZsmU5ffo0Go2Gn376iYCAgHzt//Tp04SEhKBSqWjVqhVjxowB4NSpU9SoUYOePXtSokQJoqOjWblyJRUrVuTHH38kKCiI77//vgCOWDyr7PV1kjOSsbGx4cGDB8TGxgIwbNgwLC0tMTPLHFzPLbBITU1l1qxZ/Pzzz7i5uTFz5kzCw8Np1qwZK1asADKXno2Pj6dixYrExsbSp08fbt++jZWVFTqdDjc3txwpd0IIIUxHUqSKuPyu6FQU2xB6NRStTmt0X0EdR1F4v14Gzs7OtG7dmnfeeYekpCTat29PQEAAGzdu5M0336RmzZr8/fff2Ns/XXqatbU1cXFxhsDi2LFjANSrV4+///4byByh2Lt3L0OGDCEmJoavvvoKc3NzzM3N5ap0EaQfXQQgFfZc3WNIjcprpfZvv/2WKlWqUKlSJfr06YOdnR3JyckkJSUREBDA9u3buXfvHomJiahUKry9vRk7diwnTpygTZs2RvuSVDkhhCg4EmAUccG+wey5usdoTkFxa0P2UQSVQlVgx1EU3q8XWdY5LmmuaVyPvM6HH3xo6Ly99957JCQk4OHh8Uyv4+fnR5MmTRg/fjwPHz7k+PHjLFq0iL59+7JixQo2bdpE27Zt+eabb7h//z7ly5c3er4EF0VPbqOL5SuXf8TWOUVERBAWFsZrr71mVANj9+7dNG3alBIlSuDs7MzWrVuxsLAAMtPs3N3dDZ9PCTyFEOL5kACjiCuIOhfPuw3ZRxVcbFwKbOJ1UXi/XmRZr0I7lHHAxceFBg0aGNJNbG1tsbW1Nclr9erVix49etC3b1/WrFnD999/z/bt2+nZsyfbtm2jbdu2ODk5GeZ1SOexaHvW0cXSpUvj6upqmJeRkZFBXFyc0cIBHTp04KeffuLGjRtMnDgxx7K28vkQQojnQwKMIuJFXv1IRhVeHFmvQts52aFT6Lh58yZOTk4mTzkpVaoU/fr1o3///jg7O2NjY4OXlxflypWjVq1aObaXzmPRlp/fgaSkJJRKJVZWVkDmSISFhQW1a9dm2rRp/Pjjj5iZmREfH0+TJk1Qq9UcPnyYli1bMnHiRKytrZ/XYQkhhMiFBBhFRG75yc/7SnxBBTnPa1ThRQ7SiorsV6H7vtOXqlXyt+zs42Q/hxbWFqxYsYKzZ8+SnJyMi4sLlpaWWFpaSgXuYiY/vwPLly8nPT2d8uXL06RJE8P9vXv3JjQ0lGPHjlGnTh02b97M3Llz2bBhA23btqVly5aG+ioyoiWEEIVHfn2LiKKw+lFxX+K1uLe/OMi+JHHTck1Nuv/s59C+kj2bNm2iRYsWLFiwAE9PT8O2Ely8WDQaDWfOnOHhw4e8/vrrNGzYkOXLl3P37l1UKpVh6dqQkBBOnjwJwPHjx2nSpAnz5883FPDTBxUSXAghROGREYwioiisflQUgpxnUdzbXxwU9GhU9nPmXNKZ9evXG3Lp5ar0iychIYFNmzbRrl07rly5QmhoKOnp6bz22msEBASwcuVKhg0bZli61tnZmbS0NGJjY5k1axbm5uaAfDaEEKIokV/jIiIvxeoKWvagprgt8Vrc2y9yP4cqlQqtNnOZY+lAvlh0Oh12dnbUqVMHW1tbHjx4wMSJE6lRowZeXl5Ur14drVZLamqqYXK3p6cnmzdv5o033jAqyCefDSGEKDpkBKOIKAqrHxX3ydjFvf3i0edQOo8vliNHjlCiRAleeeUVANzc3HjzzTf55ptv2LVrF76+vkDm6EZkZKRhXgWAj48PI0eOpHz58oblaCVdTgghihYJMIRBUQhynkVxb7+Qc/iiu3PnDiVLluT48eP4+Pjg6urK119/zWeffYabmxtRUVEMGTKEX375halTp9K8eXPWrVtHVFSU0fwbf3//QjwKIYQQTyKXBYUQQhSoq1evMnToUD788EMA7O3tUalUXL9+nVKlSmFvb09QUBCnTp0iICAArVZL27ZtmTt3Lp6enkbBhRBCiKJPRjBEkSJLzQrxYklJSeHLL78kKCiIvn37AlC1alVmzpxJtWrVqF27NgAODg5cu3YNOzs7+vfvz5kzZ+jbty9btmwpzOYLIYR4CjKCIYoUWWpWiBfD9evXWblyJefPn8fBwYGBAwdibm5OfHw8AQEBlChRgqlTpxIWFsbZs2cpU6YMx44dIzY2lho1ahiCkdatWxfykQghhMivQg8wfv/9d6pXr46HhwfBwcEcOHDgsdufO3eOdu3a4enpSeXKlZkyZYphdRGAdevW0aVLF8qVK0fp0qVp3rw5//zzT0EfhjARWWpWiBdDWloahw4dombNmuzevZvPP/+coKAgRo8ezcqVK/noo49o0KABrq6u/Prrr1y8eJEpU6bg4uJS2E0XQgjxjAo1RWrVqlWMHTuW77//ngYNGvD777/TvXt3Dh06hLe3d47t1Wo1Xbp0ITAwkJ07dxIeHs7w4cOxsbExFFnav38/QUFBfPrppzg7O7Ns2TL69u3Lhg0bCAwMfN6H+MIzdUpTUagHIoR4dmZmZpQpUwaAv/76i/DwcGrVqoWLiwsLFy4kICCAevXqUblyZTp27GhYEUoqtAshRPFXqAHGL7/8Qu/evXn99dcBmDp1Kjt27GDevHmMHz8+x/bLly8nOTmZ3377DWtra/z9/bl06RK//vorI0aMQKFQMGXKFKPnjB07lq1bt7Jx40YJMAqAPqUJgFTYc3XPM60CVByXmpV5I+Jll1uRu5SUFHbu3MnIkSPx9/encuXKKBQKLl68SNmyZfH19SUkJIRy5coZPU+CCyGEKP4KLUUqLS2NU6dO0axZM6P7mzVrxuHDh3N9zpEjR2jYsCHW1taG+5o3b86dO3e4fv36I18rISEBJycnk7RbGDN1SpN+mdJu1brRoXKHYtFRl3kj4mWl0+lyLXKn0+moXLkyVatWZe3atQDExMSwcuVKxo0bh4WFBQqFIkdwIYQQ4sVQaAFGTEwMGo0GNzc3o/vd3Ny4d+9ers+5d+9ertvrH8vNnDlzuH37Nj169DBBq4sPdYqaDec3sPLMSjac38DDlIcF8jpSPVvmjYiXl0KhQKFQcPbsWd577z327t0LYKi83qJFC8LCwkhISODcuXNs376dcePGMXr0aBmpEEKIF1ihL1Ob/Y/Mk/Jvc9s+t/sB1q5dy+eff87cuXMNucCPEh4entcmm1ReXjdZk8yl5Euka9MxV5pTwboC1irrxz7n34R/eaj5/6AiFbac30INuxqmaLJRm9K0aShRYoYZlipLyijKFNp7WVh0Gboctx/3Hrxs74/I9KKcd41Gg0qlAiAjI4Nt27Zx6NAhfHx8mDJlilHNChsbG5RKJVOnTqVv37588MEHAFy6dAl48dOhXpRzLvJOzvnLqTDOu5+f33N/zfwotADD1dUVlUqVY+Th/v37OUYp9Nzd3XPdHsjxnLVr1/LWW28xc+ZM2rVr98T2FMaJCg8PN3rd7Ln8tb1qc/z2cWKTYtHoNACkaFKI1EXSwe/x8xxOnzkNmv9uK8wUJj3GDec3/BfAAI62ji9tBWbPFM8c80YeldqV/ZyLl8OLeN6vXLmCn58fs2fPpm3btvTr148xY8Zw8OBB+vfvb7hYVKtWLd5//31sbW3x8vJ6aSZxv4jnXDyenPOXk5z33BVaipSFhQUBAQHs2rXL6P5du3ZRv379XJ9Tr149Dh48SEpKitH2JUuWpGzZsob7Vq9ezdChQ/n1118JCQkpmAMoANlz+bdf2U50YrQhuNB7UgqOOkVNUnqS0X2mTl2StKD/FMd5I0LklT7dSW/Tpk20atWKUaNGER4eTuPGjbl48SIAb7/9Njt27ODhw4coFAq0Wi0qlYrvvvsOLy8v4MUftRBCCFHIdTCGDx/OkiVLWLBgARcvXmTMmDFERUUxcOBAACZMmECnTp0M27/66qtYW1szbNgwwsLCWLduHT/99BPDhg0z/NFauXIlb775JuPHjycwMJC7d+9y9+5dHjx4UCjHmB/ZO+karSbX7Z4ULIReDSVDm2G4baY0M/lqTDL3QoiXg34Ctz4dNTIykj/++IMBAwbw7bff0q1bNy5evMiFCxfw8fGhdevWpKamGj1XvwStEEKIl0OhzsHo2rUrsbGxTJ06lbt371K5cmWWLVtmmC8RFRXF1atXDds7OjqyevVqRo0aRdOmTXFycmL48OGMGDHCsM28efPIyMhg3LhxjBs3znB/o0aN2Lhx4/M7uKeQvQaESqkyChSUCiWuNq5PDBayByo25jYmv6peHJeTFULkTdZlZ9PS0vj888+xtrZm/Pjx9O/fnz///JPNmzeTmprK/v376dGjB9u3b6dSpUr07t27kFsvhBCisBX6JO833niDN954I9fHfvvttxz3ValShU2bNj1yf0U9iHic7J12/RyMx+X251aDIT/F6p62hoM+LSjrfjac3yC1IJC6GKL40wcXmzZtwsnJCTc3Ny5evEhoaCgBAQHcuHGDNWvWMHfuXHbt2sVbb71ltBT4yzLPQgghRO4KPcAQ/8neaQfo4Pj4idO5Fbp70uhC1g5wUnrSf6Mkz1Aoz9QF94ozeS9EcZR11OLEiRMsWbKEEydO4Ofnx8SJEwkPD2fJkiUEBgayY8cOIiIiqFy5MkOHDqVEiRLAf4GFBBdCCPFykwCjmMttsnVugUpWRh3gJ+zvWdrxsnra90JGPkRhUiqVREZGcvv2bTZt2kRqaio7d+5kwoQJaDQaAgMDmTt3LkeOHGHOnDmo1WoaNWoESGAhhBDCWKFO8hb5l72AnkqhMno8L5OtH9fhze35eSnaJ5O+//O074VUBBfPU/bVofbs2cNbb73F/fv3adq0Ke7u7gDY29uzYsUKAAYPHkx6ejrVqlXLEVwIIYQQehJgFDPZO6EKhQI3WzccLB1ws3XL02Tr7B1eM6XZY5+fl45vsG9wvtvxonra90JGgcTzkjUd6ujRowDcunWLqlWr0qFDB5ycnHB0dCQ9PZ3evXuzadMm1Go1gYGBBAcbf54luBBCCJGdpEgVAn0qzMOkh1w8fzFfqTDZO50Z2gxCqoQY9puXgm+5zdF43OvnpeP7pLSs4pz+k9+2P+m9eJT8TM4XIr+OHTtGWFgY/fv3R6lUcvnyZSZOnEhSUhK9evVCpVIZKnRXqVKFGTNmcPLkSerVq8eff/6Jg4NDIR+BEEKI4kJGMArBtvBtRCdGk6JLIToxmq3hW/P83Mel3+Q1xSa/heFMkf5UnNN/nlfbZRRIFAS1Ws0777zDxIkT2bhxI3/99RcAFy9epGPHjkybNo2FCxdSsWJF7t+/z9atWzE3N6dLly6GUY4SJUoY6mAIIYQQTyIjGIVAnap+7O3HedwKUQWVYmOKmhfFOf3nebX9aUc+Ctq6deuwt7enadOmhd0U8RS2bt1KcnIya9as4dy5c0yePJlevXpx8+ZNjhw5wpIlS2jVqhUBAQEMGjSI2bNn06pVK9q3b2+0H0mFEkIIkVcSYBQzj+qEqlPUJKUnGd2Xl5GGvKT/mKLjW5zTf4pz259FUlISVlZWbN26lU8++aSwmyPyST/5umrVqmzevBmAc+fOUbJkSe7evUu1atVYvnw527dv5/jx43z33XeMHDmS6dOnF3LLhRBCFHcSYLwgQq+GGlX9NlOa5Wmk4XnVbCjOlb+Lc9ufhk6nY8OGDSQnJ+Pk5IStrS0lS5ZEo9EYcvRF0aRPY8q6ZGylSpWwtLSkTp06+Pn50axZM1599VX27t1LjRo1ePvtt7l16xZffvkllpaWWFpaGk0CF0IIIfJLAoxCoECBDp3R7WeVPW3HxtwmT5Ooi0r6T1GeBF5UU5dMLSkpiWnTphEUFIRSqeTSpUuUKVOGgIAAAAkuirisQYE+GNT/9+OPP6ZXr14sWLAAc3NzTp06xYULF/jyyy+5c+cO5cuXN9qXBBdCCCGehfwVKQRKhfKxt5/G007Ezsvz8lIHIz/b5aY4TwJ/ERw4cID27dvz4MEDVCoVbdq04fz58/zwww+4uroatpOJvkWPvp6FPiiYMWMGI0aM4MyZM4bz5eXlxWuvvcY///zDggULuH//PiVLlsTW1tYQXGSviyGEEEI8LQkwCoGDpcNjbz+Np12BKC/Py2vn/1mChOI8CVzvWQKswpA1WNi/fz9ffPEFU6ZMoW7duqhUKgYNGkRcXBzr1q3jk08+ISwsDLU6c0EC6YwWHVlHG+bMmUNMTAz9+vVjzZo1nD9/Hsg8X46Ojnz99dccP36cKVOm4Ojo+Mj9CCGEEM9CUqQKQfPyzdlzdQ8Pkx5ib2Nvkpz+p03jycvz8tr5z0+QkD0lykxp/FEsjhOpn9d8FlO4du0aTk5OODk5ARAVFYW1tbVR0NG0aVMGDRpEYGAgaWlpzJs3j5SUFGbMmCGd0SJm+vTpVK1aFTc3N8zMzFi1ahUxMTGGatwqlYr69esTEBBAtWrVAGSehRBCiAIjAUYh0Hfqw8PD8fPzK+zmPFFeV1HKvl1SehIPUx7mOpcie2fc2coZN1u3Yj2RujiMwmi1WlJTUxk9ejT29vZ88803ODs7G1JmFAoF6enpWFhYcP/+ffz8/Lh58yavv/467dq1K+zmv/SyTrTPyMjg0qVL+Pv7Y2trS6lSpTh58iQ7duxgzJgxNGzYkMuXL+Ph4QFAhQoVDPvR6XQSXAghhCgw8hemGMpLKo4p03Xymn4V7BtsNBKRoc14ZJpU9s63RqfJtfhfcUo7MkVBwoK0YsUK+vfvz759+yhVqhSlSpVi+/btpKenU79+fb777jsALCws0Ol03L17FzMzM0qXLi1zL4oIfXCRmJjIwYMH+eGHH4iPj+fOnTvcvHmTWrVq4efnR1xcHL169WL37t257kdqWgghhChIMoJRzKhT1KwNW/vfkrSPSMUxZbpOXtOv7K3ssTG3MSoc+Kir+HkdFSlOaUdFdTnbtLQ0xowZQ2xsLKNHj6Z69ercunULe3t7zM3NmTdvHiNGjKBFixbs3bsXe3t7Jk6cSHBwMEOHDsXCwqKwD+GllZCQgEajMcyX2L9/P1999RUtWrSgfv36tGvXjh9++IGOHTuiVqtp0qQJCoWCvXv38sUXX1C1atVCPgIhhBAvIwkwipns9S4g9058YaXr5DVwyGtnvDikHekV1eVsLSwssLW1pWzZslSvXp2oqCgsLCyoXr06s2bN4uzZs1SpUoXvv/+e9evXc/DgQYYOHUpISEhhN/2lpVarcXBw4MiRI1SqVAmdTkd6ejr79+/nt99+49q1a8yaNYtFixaxceNGJk2aRP/+/QEIDg4mODjz+6QvtieEEEI8TxJgFDFPqgeRWwc7t058YVSfvhN/h9ikWMNtB0uHRwYOee2Mv6xVtE1t1KhR9O3bl9jYWC5fvgzAt99+ywcffMB7771Hp06d2Lp1K5999lkht/TlptFo+OGHHzh79iyzZ88mOjqar776Cn9/f15//XUOHTqEUqlk//799OrVC4Bhw4Yxffp0zMxy/pxLcCGEEKIwSIBRxOy6sovY5P/vpKfCzis7Cany35Xk7B3uR1XsLox0ne1XtqPRaQy3k9KTjIKjpymmV1TTjoobJycnatWqxdGjR1m+fDlXrlwhMTGRwMBAAD7//HPMzMzkinchSk5OZvz48ZiZmTFx4kQsLS3x8vJCrVbTr18/6tWrR8mSJbly5QqrV6/m6NGjREZGUrt2bWbPni2pbEIIIYoMCTCKmLiUuMfezq3DnVsnPa8jBKasoK3Rah57O7/zKYpyde/iqHHjxnh7e2NnZ8e+fft48OCBIcDo3Llz4TZOkJSURHR0NPPnzzfc/t///sf48ePZsGEDDRo0oFmzZly+fJmBAweSlJTE1KlTgcw0OFl2VgghRFEhAUYRo0Dx2Nu5BQ5ZO+Jmysyr0Bqd5omd8rxOGM8rlVJlND9EpVQZPZ7f+RTFaYJ3UZQ9QPOv7M+ePXtISEggJCTEkColigZXV1c0Gg0fffQR9+/fx8zMjL59+xIUFMTFixc5duwY/v7+lC9fHn9/fzp27Gj0/OISXEggJIQQLz75lS8E+qVXj6mP5Vh69WmqfGetoB2bHMuDlAd5qqad1wnjedWiXAvMlGYoUGCmNKNFuRZGj+d3GdfnNcG7OC2Fmx/ZK6uHJYSh0+m4efMmpUuXpkmTJoXdRJHNhAkTKF++PLVq1aJ27dr8888/3L9/nzJlyjBkyBAuXbpEjRo1DMFFcayorg8url27VrgNEUIIUWBkBKMQZJ1nkZKYYjTPQl/l+0lzDrJenU5IS3jkaz2uU57XCeN5oU5Rc/z2cWzMbR45cpLf+RTPa4J3biMlQb5BxT49K7cAbeLEiXL1uAjz9fVl8ODBmJmZkZqayvvvv4+npyc9evSgVatWODs7G21fnM6lfn7PqVOnmDZtGq+88gpjx47NdXK6EEKI4k1+2QvB4+ZZ5HXuhFGn+DEe1ynPbcJ4ba/abDi/Id8d6+yd9DVha+js39nouflZxlWdoiZDk4FKoUKHDicrpwKb4J1bR/x5pGcV9ByT3AK04tQhfVlk/xzU86xH5JVIfvvtN0N1dZ1Oh7Ozc7FLL8paeVy/eMDy5ctp27Yt3bt3L8ymGWRkZEiQI4QQJia/qoXgSfMs8uJJ6UIKFJSwLZFrp1zfoUlOT8ZMaYalyhIbCxvDCMPTdKyzt0dfxTs/AUXWTlaGJoMHKQ8Mj6uUqgIbQcitI/480rMKOoiRFbiKh+yfgy1ntrD8u+X06NGDPn36GG1bnIIL+K/y+PLly3F3d6d69erExsYWieDi5MmTjBs3jp49ezJgwIDCbo4QQrxQJMAoBA6WDkad57zMs8gue6c4uxK2JQjyDcp1xansox/O1s6Gju3Tdqxza09+OuXZO1kqRe4TxAviqn9uHfE9V/cUeHpWQQcxRbXwnzCW/bzbONuwevVqQ+e8OC0dvH79elJTU6lWrRoVK1YkJiaGQYMG4ePjQ0xMDJMmTcLBwYEdO3bQvHlzDh06RIMGDZ5rG2/fvo2Xlxc+Pj706tWL1NRU4uPjDdXShRBCPLvidTnsBdG8fHPcbN2wUljhZutG8/LN872PYN9g3GzdcLB0wMXaBQcLB1QKFUqFEhdrF4J9g3NM8t1zdQ/qFDUxSTFG+8rawcnvROys7TFTGserWZ/7pInU2TtZOnS57iu3Y3pW+o54S7+WAGwN34pGq8HZyhkHSwfcbN0K5Or/077X4sWS2+dApVKh02V+B4pDcHHlyhV69OjBypUruXv3LiNHjuTs2bPExcXRsmVLfv75Z9q3b8+SJUt44403mDlzJgMGDGDy5MnExcU9lzampaXx8ccf07NnT7766itu3bpFx44dSUtLIzk5+bm0QQghXhYyglEI9B3a8PBw/Pz88vXc7FfwW/m1euQV/EfNLdDqjFee0Xdw1ClqNFoNSoUSBYrHVuLOrT0Olg45lsjVe1I6UPYRECcrJ1RKVY70nuzHFJMUw8ozK00ympF9ZMfN1o3OlTs/9f6eRFKYBDz6c1AcAgu9w4cPY2NjY6jhYWZmxr59+/D09OTUqVMA9OrVi759+/Luu+8yZMgQtFotrVu3fm5tPHLkCKmpqYSGhrJw4ULi4uKoUqUKN27c4Pbt23h6eha7OS5CCFFUSYBRzOy8vPO/9KpU2HF5B52rds5127zMLVApVIYOTejV0P+qiANmKrPHdthz1NEAXKxdsFRlvs6eq3sMnf4npQPltYBg9mPS6rSoU9UmmcPwvJbF1ZMUJgEvxuegVKlSNGzYkKioKDw9PQkJCWHIkCEsWLCAjRs3sn79eu7fv4+3tzdWVla0bNnS8NysE8FN7fz583h4eODi4oK9vT3h4eEAxMXFYW9vj0KhIDAwkJkzZzJ79mwJLoQQwkTk17SYUaeqH3s7q+xpVBmajBxL2rrYuBg68o8aHVh7bi1rzq7Jkd6UWx2NuJS4XFOYnpQOpO9kdavWjQ6VOzwysMl6TI+ap/G0JGVJiCfLrfZG5cqVOXv2LCkpKQB4enryyiuvcPToUUaPHk1oaChHjx5l6NChOZ5r6uBCq9WSmJjIgAED6Ny5M5GRkQDUqFGDwMBAAgMDWbFiBdOnT+fWrVuEhISgVCrZtWsXgCE1TQghxNOTAKOYyT43IfttPXWK2mhEIDUjlQcpD4zSo7LPLcjeYdePDjyqeF9uHfrsK2Lpt8kaGDzLnIasgYiLjYvRY88aEJiqjUKYSmpqKvfu3SvsZhg8KoXI3d2dUqVKMW/ePMN9KpUKb29v/Pz8+Pzzz/n111/x8fEp0A58RkYGSqUSW1tbBg4cyJtvvsmePf/N03rjjTfo2bMne/bswcfHh5UrVwLQpk0bLly4ABSv1DQhhCiqJEWqmHGycjJKY3Kycsp1u+xzHnLToXIHw+Tr1IxUEtMS89QGfdCQWx0Newt7oxWy9J3+gkgDMfUchhchVUW8OG7fvk3fvn2pV68ekydPLhLzA5RKJSkpKXz66aeUKVOGFi1a4O/vD8DIkSPp06cPp0+fZs+ePTx48AA3NzcA7O0zRyQL8hiWL1/OH3/8QZcuXejevTvBwcHUqFGDLl260KVLF8qUKcPWrVu5c+cOAE5OTlhbWxMXF8fff/9N586dC6RdQgjxMpIRjGKmWblmRlfZm5Vrlut2eU0Xyroqk0anydNz9EFD9iv+nf07U9+7PmZKMxQoDIX7Ckpe06qEKE7OnTtHREQEXl5eTJs2jejoaGJjYwsluMieDhUVFcXHH39MiRIlKF++PJs3bzakRdnY2FC3bl3at29PfHw8P//8M66urkbPL6hjuHXrFlu3bmXSpElERkaye/duHj58iJOTE6+++ip//PEHAJ06deL27duGUYz27dvj5OTEL7/8Qo8ePQqkbUII8TKSEYxCoF956WHSQy6ev5jn1Y+yryBV26v2IydGP6lOhj4d6nGBiJnSDBtzG8yUZrmuDpXbFf89V/cY5mVkaDM4fvs4HRyNtynoCtZCFDc6nQ6dTsdXX33F8ePHsbS0ZOzYsdSuXZvatWszZ84cxowZ89zbpA8Ijh8/TokSJShbtixffPEFd+/e5ZNPPsHJyYn4+HisrKwA6N27Nz169KBcuXJAwY5YREVFcfHiRYKDg7GxscHGxobq1asTHR3NoUOHCA4ORqfT0aZNGzZs2MDJkyepVKkSs2fP5ubNm4Y2Ari4uDzmlYQQQuSXjGAUAv2oQYouJV+1HLLXgNh+ZbvR7TVhawwTsIN9g3G2cjbUxrA1t0WlUBnmSJgrzdlwfkOO2hV6KoUKewt7Wvm1IqRKCJ2rds7TSEFeVmJ61loWT6qpIURxo1AoUCqVaLVali1bxuDBg/nll18A6Nq1K4mJiYbUnoKmH7VQKBSGQOLDDz/kxx9/ZMGCBahUKr7++muGDx9Oeno6UVFRhuf6+PhQrlw5tFqtUYBiajNmzOC1117jl19+Yfr06UbzJpo3b87ly5e5cOECCoUChULB1q1b+fjjjwkLC8PS0tIQXMiEbiGEKBgSYBSCp1kOVZ2iJjYp1ug+jdY4pSlDm2HorNtb2WOmMkOj06DVaUlMT8TFxoUStiUASNFkBjc6nQ6lIufHQKPT8CDlQb47/3lZiSm3489P0FAQxfZAAhfx/C1evJivv/6a7du3k5aWxs2bN7GysqJNmzYkJCTwzz//4Obmhqenp1FHviDoA4usQcFff/3F9u3b2b17N2PGjOHff/9l//791KtXj5s3bxIXF0dMTEyOfSmVygKZLL1582YOHz6Mm5sb27ZtY+HChSxYsAAnJyeio6MNk7ZbtWrFpk2bAJg/fz5VqlRh0aJF1K5tnLIpE7qFEKJgSIBRCJ5mOdTQq6E55kiolDmXd8zaec+tI5/9PnWq+pETxfXPyU/HOy8rMeV2/PkJGgqqXkVBBS5C5ObMmTMsW7aMxo0bM2nSJBISEnB1dTUUq3vvvffYvXs3SqWSjIwM7t+/D+S+TOyzuHjxIklJSYbAYsGCBQwdOpRTp07Rq1cvKlSowJ07dzAzM8PX15eSJUuSkZHBiRMnmD9/Ps2a5T4PzNRCQ0MZMWIEZcqUoU2bNlhaWpKamkq9evUAeOeddwxzLZRKJfXr1wfgww8/5Ntvv8XV1dXk750QQojcyRyMQqBf/ehh0kPsbewNnXD93ISktCRSNalYmVlhbW5Nba/axCTlvEqYvQYF/De3Qp2iJik9yegxQ8c+S39co9Og0+lws3UjNSOVpPQko/1m7fzrn/u4gna5zcvIbe7I8dvHjeZgbA3favScxwUNuRUQNIXsr5mUnmRYYUvmighTiIuLAzJXMIqIiKBbt24EBwfTrVs3vvzyS0aMGMEvv/xCeno69vb2eHt7o1AoDEvAtmzZ0mRpR2fPnuXjjz/G3NwcT09PfvnlF2bMmEFERASdOnVi69atNG7cmCZNmnDy5ElcXV25cuUKw4cPp1q1aob9PK/VrerUqcOQIUNIS0ujZMmSANy4ccMw/6NBgwZUq1aNt99+mzNnzjBt2jQg870GCjRlSwghhDEJMAqBvhMeHh6On5+f4X6jjjyQkJZAQloC269sN6pf8TgKhSLXCttmSjNDILPi7Aqj56Rp0gzVwB+mPMwxcTw/nf/cZA9Qjt8+niMIeVLQkDVIUSlUmYUDtRkmWZ72UW0wWrrXBJXCxcvtr7/+YurUqbRt25YaNWrg7+/P5MmT6d+/P8OGDWPAgAFERkbi6+vLggULiI+PJyMj8zt89epVGjZsCGR2lJ81tSc2NpaZM2fSpk0bhg0bRqtWrTh69Chvv/02iYmJ/Pzzz4SGhhIYGIhOp2PdunVkZGTw6quvGr12QQYX6enpmJubG14jLS0Nd3d3IiMjKVu2LAA7duwgKCgIyAzevvrqKy5fvkyFChVyvEeSDiWEEM+PBBhFyKM67tnnWjxOhjYj1wrbNuY2hqvvZkozo8dTNf+9bm4jENk73upUNX8c+wOVUkWLci0o6VjysW3KS0rTk2paZA++3GzdCKkS8tjXza/sbUhKMx7NkREN8Sxu377N0qVL8fDwoHv37sycOZNKlSoRGhpKUFAQtWvXxt3dnV69evHll19ibm7O6NGjgcz0HwsLC8A0HWUXFxdKlCiBg4MDAIMHD2bWrFn8/vvvfPXVV9SoUQMrKysuXLhAw4YNOXHiBL///nuO/RREcKHValmzZg23b99mxIgRhuPVj0TcvHnTsO2dO3eIj49n4cKFvP7663Ts2JGKFSsa9iMjFkIIUTjk17cQ6Oc0HFMfM5rT8KhUn0dV686Nvmp3bvfrWZlZGT2m1WofO78i67yKrG3K0Gaw/cr2PLXpcbfhyTUtCmrexePaYGNhk+M1ZY6GyKtz587x7bffcvHiRQAOHz5MYmIijo6OdOzYkT///JPq1asbKkhHRESQkZGBm5sbkydP5ocffsDT0xOdTmcILp5W1tWS9PMQOnfuzNy5cwGwsLCgdevWJCQk4OHhgVKp5NatW9y6dYsqVaoQGxtrmDRdUDQaDSdOnECpVNK6dWuOHj3KiRMnUCgUhpGc4OBg1q9fT1JSZvrntm3bOHbsGKNHj6Zjx45G+5PgQgghCo/8AheCXVd2GS1Tu/PKTgBqe9U2WjbWxtwGN1s3rM2sjZ6vQGFYftbB0gEXaxejSdXZO/BZ06MArM2N96dFa7TUbfZgI2vHW7/MrV5eRlfyMvH7SZ5mYvyzyt5uS5XxaxZEkCOKP51Ox5w5c/j4449JTk5m/vz5nDt3jh49enDo0CEAWrRogYeHByVLliQ6Opo+ffoQFRVlKExna2tr2Nezjlhk3UdMTIyh4x0QEICvry9Vq1Zl48aN1KxZEwsLC/z8/Ni2bRt9+/Zl/PjxAIwfPx4fH59nasejaDQaDhw4gEqlYvfu3axfv574+HjeeustfvvtNwDMzMzQarX4+vpSp04d1qxZA8DMmTNZu3atYaK3EEKIokFSpApBbHJsrreP3z5ulJJja2FLh8od2HB+A8kZyYb7S9iWeOxcgNzSjbKOCOgfj0mKyTG3I0ObgTpV/cg5ByqlyqiNua1klV1uaVf59aQUqvzKS7G/7O3ecH4DiemJhtvPI8gRxcucOXPw9fWlYsWKdOrUCQ8PD77//nsAnJ2diYqKIiYmhoyMDC5dusSwYcOoXLkya9eupVu3boYJy3rPElzoAwuFQkFCQgITJkxApVLx/vvvU6JECVQqFcOHD+f+/ftMnz4da+vMCw8hISGEhPyXfqjRaKhcufJTt+NJoqOjDQFVfHw848aN44033uC9997jl19+YfXq1XTp0sWQ8tSvXz/69+/Pw4cPGTJkCCDpUEIIUdTIL3IR8qg0IP2VdFtzW8yUZiSnJz92udgnpRvpH7ezsHtse5LSknIsT9uiXAvMlGYoUGCmNKNFuRbPcMR596Rjyq+nWZLWFCMx4sUVERHBP//8Q+XKlQkKCsLDwwOACxcukJaWhq+vL6mpqRw9epTo6GhKlChBSkoKdnZ29OnTBysrK5MWftMHJ3fu3GHVqlU4Ozvz9ddf4+HhgUqVeWHA1dWVgIAATp48meP5+lQq/bamdPfuXcP/e3p64uzszI0bN7C0tKRfv34MHjwYgCFDhhAWFgZkjmIAODo60rt3b9q2bWs4RgkuhBCiaJERjCLkUSsp6TvXG85vIDEx0bC6VNYRhrxckX/S62WXqkklMTHnKkr9avV7puPM6mnabQpPM6fDFCMxplJY75v4T/b0pVdeeQU3Nzfu3LlDqVKl0Gq1qNVq4uPjqVmzJr6+vjRt2pTVq1ezb98+pk6datIRi9zs3LmTu3fvcvbsWerXr8+nn35KtWrVaNGiBe7u7nh4eFC7dm2qVq2a47kFNYH73r17DBw4kNWrV5OcnMzp06c5d+4cSUlJfPzxx3z//fekp6cDmStDeXl5GZ6rVCqxsLCgXz/T/QYJIYQwPbnsU4Q86Qp59roWyen/pU2Z4op8G782mSMlFpkjJdnnVzzLnINHFesrrOJ2hTGn42nl9t5JUcDClZGRYRQM6Cch9+3bl19//RXI7KDHxcVRt25dIHNJ1Zo1a/Lpp5+yc+fOHFWln5Z+pEE/+vHgwQNOnToFZE40d3JyMsyh6N69OydPnkStVgNgbW1NSEiIYTWpgpKWlsaXX37J0qVL8fT0pFmzZuzatYvTp08TGxvLm2++yZEjR4iKikKlUrF48WJ27dqFtbU1sbGZKaQySiGEEMWH/GIXAmW2t11/O78rKaVkpACZHdDshfjyc0Ve/3olHUsS5BtEakYqGdqMHKtXPUsn/FEd4uexOlRuilO6U27vXWG9byLTwoUL+fXXXw0deX36TlBQEGXLlmXv3r0AHD9+nKVLlzJy5EiOHDmCVqs1pByZoqp0dHS0YRUqfcBz8OBB/vnnH3bv3k2zZs24efMmNjY2+Pn5sW/fPsLDw59rRevz58/Tt29fEhISaN68OZD5Pl2+fJnZs2dTu3ZtzM3N6dKlC0uWLGHo0KHcvn2bbdu2ER8fT9OmTZ9bW4UQQpiGpEgVAkcrRx6kPDC6DU9Oe7FUWRpX2f7/VY1Cr4bmmKxtpjRjzdk1qFPV6NDhZOVEs3LNnphGk1sNDQUKStiWyHcnPOvxJKQlGD2m7xAXVFXuJylK6U5PklswUVjv28vu7Nmz+Pv7M3DgQPbs2cPWrVtxcHDglVdeMXTaa9Wqxb179wBYv349fn5+DB48mJYtWxrt61muyB8+fBhvb29iYmIICwvj33//pVSpUgQFBdG6dWs8PT1ZuXIl6enp/O9//0OhULB9+3YiIiL4888/cXR0fPo3IY/u3LnD6dOnKV26NEFBQYwYMcLwmKOjIzdu3CA5OdnwvgUFBTFv3jwSExP54osvsLSUz7QQQjwvZ8+e5ZVXXsHGxubJG+eBjGAUgublm+Nm64aVwgo3Wzeal8+8qvektJfc0npyG71QKVTodDoepDxAo9Og1WmJTY7NUxpNblfCnaycANgavvWxk8uzy3o82QMg/bEUp5GEwpLbedcvaayfbF/byzTpNiKn9PR0/vzzTwCOHj3KX3/9xZIlS3B0dKR06dKG0QqlUmmYI6AfTZg1axZLly7Fx8fHpBO4VSoVXl5eWFtb88UXXxAaGkqlSpUMj9WqVYsuXbqwZcsWTp48iVarpXPnzvz00084Ojo+lxGMGzduoNPpuHz5MhEREaxYsYKPPvqIP//8k0qVKmFnZ0ft2rWZO3cun3zyCTt37uTNN9+kRIkShuDClO+ZEEKI3KWmprJ3715mzJhhsn1KgFEI9FfP6zjUMUqFelLaS/Y/thmaDNaGrc3ReXexcUGjy1mfIi9pNLnV0FAoFE+V75/99VQKVY5AwtSrQ72IcgvC9Esa6wseHr99vLCb+ULS6XSYm5tTo0YNAMzNzXn33XdRKpUEBATg4uKCRqNBp9Oh0WR+5wICAli5ciXdu3c3LJLwrPUs9IXl9OrUqcOsWbOIi4ujd+/e1K1bF3d3d8NcEP02nTt3pnXr1obgBwp+SVf979SxY8e4desWwcHB1KxZk507d+Lk5MQff/zBvHnzCAoK4u7du3z11Ve88cYb9OvXD09PT6N9mXrSuxBCiP/o/25ZWlry9ttvc/r0aUPq77Ne4JEAoxBcib7C/GPz2Re/j/nH5nMl+grw5InH2YMGdZo6RzqTSqHKtdgegDpVzR/H/mDhiYXcib+T6+Th7J3Zzv6dc7xGXvP9s7fBxcZFAomnkFsQJnMwCtbly5e5du2aoYNbtWpVxowZQ61atXjttdcoXbo0kJnqs3PnThQKBSqVCq1Wi4eHB6NGjWLWrFm4uLgAz9ZRjouLY+DAgYbb+h9/Nzc39u/fz9ixY9myZQvp6emGuSBpaWkATJgwgfr16xvtryCCixMnThiK3+n/YFWrVg0rKyucnJzo168f3333HZ988glTp04lIiKCJk2a4O/vT0ZGBr6+voCMWAghxPOknxMYGRkJQM+ePZk2bRrw7Bd4JMAoBKHXQ3O9/aR0obzk2bvYuGBvZU+wbzDOVs6ZFb+znGb9Fe/tV7azLXyb0cjE1vCtuXZmHxX4PGplKD1Jfyo4xWkVrOLk/v37aLVa1q1bx507d4iPj2fKlClERkbi5eWFRqPhnXfeYePGjQA0bNiQihUrcuPGDeC/znvNmjUNwcXT0o+KODk5UbNmTUaOHEnfvn2ZPHkyU6dOpVy5cqhUKhISEmjfvj2DBw/mu+++A8DCwsJoPwXN3t6eBQsWkJycbAhyHjx4wKVLl4DMURMbGxuOHTvGDz/8YBgReuuttwzbg4xYCCFEQdLpdEZ/E3bt2kXLli2ZOHEimzdvpkOHDmg0GrZt2/bMryUBRhHypHShrB12lSJn8SszpRnBvsGoU9TsuboHjU6Di40LTtZOObbVaDWZFbuzyH47t9fNGig8ac6IpD8VHAneTCsmJoaPP/6Y4cOHo1AosLa25v79+yQmJmJtbc0rr7xCyZIluX79Ora2tri4uNCnTx9++eUXUlNT8fb2Nllbrl+/zu7duw2jIgBvvvkmFy9epG/fvixdupQSJUpw6dIlrKysiImJoWvXrpQsWZJmzZrl2N/z6LT7+fnRokULduzYYbivQ4cOnD59mrNnz6JUKpk7dy4ffPAB3bp1o3v37gXeJiGEEJn06bMKhQKFQkF4eDiXL19m48aNjB8/nnfeeYfLly8TExNDp06djNJtn5asIlWMZF35aMP5DUQnRhs9rtVq2XRxE6ma1P/SmlJBqcgZR2Zfgjavr5uVpOkUnuK0ClZRl5KSwqeffoqfnx8fffQRCoWCtm3bMnbsWE6dOkXDhg0BcHZ25syZM3To0IFWrVphbm7O8OHD2bRpE/Ds8yz0Hj58yLRp06hSpQrff/89DRs2pG7durRo0YKEhMzV2B48eEDjxo1ZvHgxYWFhlC9fnilTpjzzaz+NEydOoFQqeeWVV4iPjyc1NRULCwuUSiUDBw5kzpw5/Pzzz3To0IFBgwYZ3iNTvV9CCCEeLTo6muPHj9OmTRsyMjJYunQpixcvZvz48Tg7O1O9enUgc95cr169uHz5Mv7+/s/8ujKCUQhszGwee/tJqUfw3xXsrMGDFi2J6Ym5LjObnZXKKsd9uY2KPI6p0nTycrxCmJJOp+PmzZssW7aMqKgoEhIS+OCDD3B2diY+Ph4fHx8aNmzI3LlzOX/+PPfv3ycwMJD9+/cTGRlJjRo1eO+99wBo27Yt8GwjBVlXdapatSq1atVi8ODBODo6EhkZyaZNm/Dw8DBcVXrw4AEZGRm89957dOjQIdf9mFr2VKvo6MwLHFu3buXBgwd4enpy584dw6gLQKdOnfD09GTx4sVotVoUCoWhjRJciMK2Y8cOQ9AuxItG/5vt4uLCqVOn+OKLL5g3bx5Xr16lZ8+eNGjQgKtXrxIfH4+DgwNOTk4A+Pr64urq+syvLwFGIUjRpDzytjpFzdqwtbmmHmXtiO+5uodg32DsLOye+Hr6NBr9f1+t+mqu6Uoty7fM5dmPZqo0HVNVpZZAReSVQqHAzMyMY8eO4ePjw82bNxk9ejQhISGMHj2af/75h759+9K1a1cSEhL4+uuv+eeff5g9ezZlypQxaVuyruqkL5pXq1YtnJ2dGTduHP/73/+wtbWlUqVKrFixgnHjxqFQKPD29sbJyQmlUmn4Q1JQq0NpNBqjgOD06dMsWbIEyKxfsXz5cgICArh06RKXL19GoVCQnp4OwIgRI2jTpg0lS5Ys0DYKkVcPH2b+bbhx4wZ//fUXW7ZsKeQWCWE6+nkW+t/s6OhoVqxYwa1bt+jcuTN2dnY4ODgA0KxZM2bMmGFIr3VxcaFEiRLUrFnzmdshKVKFIPuIQtbb2y5te+SqTfqOeOadsOfqnhwF1/TMlGbYmNsYaiZkXcY0ITUhRzqTlcqK47ePE2wZnOe5Es+SppOXInz5ldv7I2lEQi8xMZGIiAiqVq2KQqEgNjYWd3d3AGbOnMnRo0epWrUqVlZW7N69G1dXVypUqEBwcDCVKlUy6RX3a9euER8fT40aNVAqlYSHh/PNN98QERHBJ598go2NDVWqVCEjIwM/Pz9mzZpFz549qVevHh07dqRy5cpG+yuI0YDk5GRSU1NxcnJCpVIRHx/Pzz//jKenJx07duTatWs8ePCAwMBA/vjjD/bs2UObNm2YO3cuU6dOxdzcHMicAC5EUZCUlISZmRlHjhyhefPmdOrUiStXrrBlyxYqVqyIj49PYTdRiKf28OFD7O3tDX8PLl26xL59+3j99dcZMmQILi4uuLu7U7NmTRYuXEiXLl3o3bs3LVq0IDo6mipVqgDkOpfvacilpEKQPT0p6211Ws6J1vqCerFJsUb3p2akGkYR7CzsjFKctFqtoQDb9ivbjUYItl/ZnmNCd4om5YmjB+oUNWvPreXP43+y4PgC1pxdYxglyO/oQV6K8OXX854TIiMmxUdkZCSDBg3i+vXrhivr7u7u7Nmzh5iYGCpWrEjPnj3p378/9evXx8LCgjp16lC3bl0qVKhg2M+zrsiUlpbGRx99xLBhw/jkk0/4999/Adi7dy+vv/4606ZNY+PGjVSqVIkbN26wdOlS5s2bR6lSpQAYPXq0IbgoyHSohQsX0qxZMyZOnMiff/5JRkYGPXv2xNnZmXPnznH//n3KlSvH/PnzAahcuTJbtmyhU6dOfP311wXWLiGeVmRkJDNnzsTCwoLNmzfTu3dvFixYgIuLC56enpw+fbqwmyjEU8nIyCAyMpIBAwYY/katWLGCb775hoMHDzJhwgT69OnDunXrAGjSpAkqlYpVq1YBmX8L9cGFKf+uSIBRCB6XIpWbYN9gQq+G5qiDYWlmaRhF6F69Oy42/y2LqSWzevfm8M05RkSy384qNin2kR3m0KuhxCbHotVp0eg0PEh5YAhI8pvmlJcifPn1vJduNVVqlyg4J0+eZOjQoZw7dw5HR0fq1KljSNEpUaIErVq1Yv/+/QAkJCSwY8cORo8ejaWlJTqdjtq1a6NSqQxXhJ51pGD37t3cuHGD9evX06FDB2bOnAnAwIEDSU5O5ssvv2THjh2cPn2akJAQ5s6di52dHaNGjTLso6DTobZu3cqkSZNYtWoVQ4cOJTQ0lNjYWFauXEnnzp05d+4ckydP5q233mLPnj28//77WFhY8M477wCZS+RKPQtR1JQpU4Z3332XW7du4ePjw9mzZ3nvvfcoV64c8F/9FvnsiuJmzJgxODg44O3tzcKFC4mMjGTBggV8+umnzJkzh8uXL5OQkEDt2rXp3r0733zzDRMmTDCav6dnyr8rkiJVCLJfsc96W6VQGQUSKoWKhNSEHCtGASSnJ7Ph/AaCfYNzLb72NDS6/1++NpcUo9z2r78vL6MHWdOiktKNKxO72Lg8czpTsG8we67uITUjFUszywJfujW/x2xpZkkZhWnz90Xu1Go1Dg4OHD16lP79+9OoUSN++OEHunbtyooVKyhZsiRarZYyZcpw//594uLiOHv2LH/88QeDBg2idevWJm2P/qqQm5sb1apVQ6VS4erqSs2aNcnIyMDMzIyzZ8/y119/sXPnTn766Sf++ecfqlevbkjj0s/VKOjJ0a1ataJq1arcvn2b2rVrU6pUKW7fvo2dnR2TJ0/m66+/5ueff+b48eNMnjyZ5ORkatWqBfy3MpRM4BaFLS4ujq1bt9KwYUO8vb25du0adnZ2jBs3jlmzZnH16lV2795NkyZNKF26NKtWraJLly7y2RXFhkajQaVS4evri0qlYtiwYXz11Vf069cPLy8vjh8/Tvny5enSpQtr165l+PDh2NjYEBwcbJgTV5Cr+ckIRiFQKVW53lanqLG1sP3vfoWKluVbsv3K9lz3k5CWYHTl3NRX7LN3mHPbvz4tKy+jB1mv+GdoMzBTmpm0jsPzrruR32OOTozmUvKlAm2TyMxDXbx4MZC52tKkSZN46623aNeuHV27dsXLy8tQY6Jq1apYWFiwevVqGjduzMKFCw3BxbNeybx16xZHjhwB/hv5qFmzJjdu3KB79+5MmjSJsLAwlixZglarZd++fbz99tvs27eP8ePHA5lD11qtFp1OV6ATuPX0gVDnzp3Zt28fGRkZxMXFUaZMGU6ePEn58uWJj4+ndOnSpKSkULlyZUNwkfU4hSgs165dY8+ePahUKi5duoS5uTlHjhxhzpw5pKWlUbduXe7evcurr77KmTNnuHfvHuXLl6dRo0ZAwaYeCvGsHj58yL1794DMKtwZGRkcOnSImJgYKlSoQOXKlTlx4gQjR45k69atAHh7e2NlZYVOp+PNN9+kQoUKhr9vBfmbLSMYhSDQO9ComnegdyCQ2RnNOjfCxcaFko4l0Wg1OfaRVUxSDA9THhLsG8yOyzt4kPLAJO3M3mEO9g1m9bnVRiMs+g9n1tEDlUKFRqth5ZmVhpGE3EZYbMxt6FatW57akn0kQL/P5yW318/LiEn2Y07Xpj+vJr909Ff47e3teeutt0hOTsbS0pLY2Fi+/fZbfH19WbhwIdeuXTNM5ixXrhwlS5ZkwoQJ/Pvvv9SoUeOZr8JrNBomT57M7t270el0TJ06FTs7O8NIxeDBgxk9ejQnTpwgJiaG4cOH4+bmxowZMzh79qxh2Vu9gl51Sb+sbFpamqECuK+vL5s2baJp06Z069YNFxcX/ve//7Ft2zaSkpL47LPPZPK2KFJ0Oh1ff/01+/fvp1OnTgQHB6PT6Th06BAqlYoWLVrg5eWFhYUFt2/fxsvLi8uXLxtWi7t58yYgq5yJouvhw4fMnTuXhg0b4u7ubvib0qhRIw4cOICPjw+lS5cmPDycTp064e3tzYkTJ6hVqxYNGjQw7Od51SCSAKMQnL9/Psftcm7lHplyo1KqHjtvQqvTsuPyDsxUZmh0mhxpVkqFEmszaxLTE/PUPgUKStiWyNFhtreyx9bC1igI0rfrkUUAs6RaZV/xKj8jLoW9QtSjXv9Jbch+zOZK8wJs5csr+xX+rVu3EhMTw5AhQ/Dy8uLixYv4+vpy79490tLSDM8BsLGxYeLEiYZVj571h/enn37iypUrbNu2jSVLljBz5kxGjRqFmZkZf/31Fx06dKBx48asW7eOTp06MWLECE6ePEnbtm0NFcGzLl1ramq1muXLlzN48GAADh06xMSJE/nwww9p0qQJCoWCV155BVdXV0JCQnj11VcB8PT05Oeff8ba2hqQQnmi6IiPjyc6OpqkpCQ2btxo+O707duXDz74AA8PDz7//HMAPDw82LJlCxMmTGDMmDHY2tri6OhIcnJyYR6CEI+l1Wqxt7fH2dmZ8+fPU79+fcPn3MvLi/j4eDQaDaVLl2bLli107NiRzp074+/vb7hw9LxTWCVULwSPCiQelXLTolwLzJRmKFBgpjQjqGxQjqJ4+hQcdao6x2RwrU6b5+ACoIRtiUemGOUlLehRx/e4uhlPWpGpsKuGP+3rZz/mCtYVnvwkkW8KhYLo6GgmT57Mzp078ff35+DBg9jY2FCyZEni4+OxtrbGwcHBUHk76w+tPrh4FvqARb8aB4CjoyMBAQEkJSWxatUq5s6dS3x8PHXr1uX06dOkpqbSuHFjRo4cabQvUwcXOp2OjIwMfvvtNxwcHNi9ezenTp0iLi6OH3/8kS+//JKmTZsa3g8PDw8cHBy4cuUK8F8alQQXoqhZv349a9euJSYmhmPHjqFUKjl27BiXLl3C19eX2rVrs2fPHubOncuyZcto2LAhCoWC1NRUPD09cXR0JCMjA2tra0mPEkWWUqnk5MmTnD17lmXLlnHr1i3D34myZcty//59zp8/T5MmTbCzy6yPFhAQYAgu4PmnsMoIRiEwU5rlevtxaUad/TuTkJrA9ivb2Xt9LzqM88Oz31YqlDkmkz+Jg6XDEydH5yUt6FEjFY+rm/GkEYpnGf0whad9/ezHHB4ebuqmvZT0k9v0Nm3axM6dO3Fzc+P7779nyZIlODg4sGbNGurXr8+6devQ6XSMHDnS5D+y2a8K6WtBtGvXjoSEBFq0aMH48eNZs2YNXbt2BTLnhlSqVAlLy/8+RwU5aqEvLHju3Dk2bdrEsGHDmDt3Lt9//z0pKSn8+++/7N69m6pVq9KqVSsABg0axK5duwCM3mv9/oQoTPrvXalSpYiKiqJ+/fo4OjrSu3dvzpw5Q1BQEDVr1qR37944OjrSuHFjNm7cSM2aNfniiy+M9mVmlvk3WNKjRFGhT3/Si42NZcqUKQwZMgRXV1fmzZvHZ599BmQGEpGRkezevZu7d+8a7i9sEmAUguyTR/W3n5Rm9CD5QY5UKX1QoNFqiE3+r06Gq40r9xPv5wg8slOgQKVU0aJcC0o6lnxi2/NSXO9pVnN60gjB0+zTlPM2nvcKVSJ3t27dolSpUoYO7507dyhZsiTr1q2jcuXKvPPOO5QoUYKFCxfy/vvv8+677xISEkLdunVJS0vD3NwchUJh0ivwWfejn8cwbNgwvv76a3bv3o1SqaRPnz5cvHiRihUrAlCtWrUc+ymIzo1WqzUKfoYNG8aKFSsYO3YsGzduJDIykm+//ZaoqCgcHR356aefaN68OSqVCh8fHwYOHGjyNglhCvrP9IULF0hKylyV8Pfff+fmzZtUqVKFtWvXcu3aNWJiYoDM+Vaffvqp4fkFGdAL8bRu3LiBt7e3IbjYtm0b/v7+pKenY2dnR7NmzShfvjxz587l7t27eHh4ANCxY0euX7/OmTNngJwX4QqDBBiFIE2T9tjbkHuHO/tkbwUKwyTphykPc3SA14SteezcDTdbt0cGC3npnD9qm6ep8P2kEYKn2acp5208S9VyYRp//fUXYWFhfPXVV5w7d45Ro0bh6upKnz59GDx4MFOnTuWdd96hS5cuTJ48mfj4eLp06UJ8fDyvvfaa0b6eJbjI2mnX6XRERkby999/M3r0aMNwdLly5QgJCeHGjRucPn0arVZL+fLln+n48yMpKQkbGxtDB+rcuXM4OTnh7OyMl5cX58+fN6Rpde3aFZ1Ox7fffkvt2rWl0yWKpOzBsj5ACAwM5IsvviAuLg4nJyccHR2BzIsR+lWjqlatmiO9Tz7noqhZs2YNcXFxDBgwgBMnTvDBBx8QEBDAokWL+OKLL6hRowbR0dFoNBrOnz/PnTt3DAGGQqHAx8fHsIBJYQcXIHMwCkWqJvWxt9Up6hx1IhLTEnNNi9LPV9B3gFv6tQRga/hW7Czscrx2XgraqVPUrA1b+8QicqYsNPe4+RlPq7DnbYhnl5GRwcqVK0lLS+PMmTOGCcfHjx/n008/ZfLkyfz666/UqlWLsmXLsmPHDiwtLQ3Lz7766qs4OTmZpC36q6T6WhTh4eGkpaVRtmxZQwCjH410cHAgPj6eQYMGsXr1al577bXn8oMfFxfHxx9/zPTp04mLiyMjI4MxY8YwdOhQZs2axa1bt1Cr1eh0OhwdHXn48CHh4eHMmTOHJk2aFEgKmRDPSr+Ig0Kh4P79+0Dm91Cn0+Hs7Ezt2rX5999/Adi5cyd9+vThwIEDhkJi1tbWz2VZTiGehn6OW0BAAOfOnQMy0/YWLFjAlClTUCqV/PPPP7i4uDB16lSmTJmCtbU1Xl5ehdnsJ5IRjEKg0Br/wGVoMwxzLWp71Wb7le05Rh6yT9zW03fs9VfXja7akzm/I+u+citol30kIkOTkeP1H1dk73Hb5JUpRgiyH0f2uS7Pe96GeHZ3797l/PnzWFlZkZ6eTo0aNdBoNHTv3p0dO3YwZcoUrKysWLZsGa1bt+bChQs0b96cfv36mawNGo2GGzdu8PHHH7NkyRLDMrSbN2+mZ8+elCxZ0jAKoO+8ODk5UatWLWrWrEn79u2fy9yb48ePM3HiRAIDA+nUqRM6nY64uDgCAgKYMmUKw4cP59atW1SuXJlNmzYxfPhwli5dSo8ePfjuu+8MbZcJ3KKoyDq/KT4+nk8++YRbt27Ru3dvunfvjkKhMHS0Ll++TM2aNTEzM6Nbt26G+U568pkWRZVKpSIuLo7169dz+vRp/vrrL3r16sVff/3F0qVLqVixIjqdjkaNGhEVFUWtWrXo2bNnYTf7iWQEoxCkk7MWgn4UILfg4kmS0pIMKzDFJsUaPWahsjBagaq2V+0cz88+EpF1GVq93DrneVlR6nnKfhw6nQ43WzdszW0xU5oZKp9nX6FKFF329vY4Ojpy9OhRw4RolUrFw4cP2b9/P0uXLqVr167cvn0bT09P+vfvb3jusxbKg8yJdb/99hs+Pj6UKlWK0NBQ4uLi0Ol07N27l6SkJC5evAhkdmAyMv777rZr14727dsDBVu8S3+ct2/fpk2bNnz00UdUrFgRZ2dnSpQoQWBgIP379+fu3bvMnj2btm3bcuTIEVJSUpg6dSpWVlaGdC/9cQhRFOg/i3fu3GHPnj3UqlWLH3/8kblz5/LwYebvuIWFBfXr18fOzo4ff/yRoKAgQ3CRtYikEEVFbn8Pfv75ZzIyMvjwww9Zu3Yt6enp3L9/n6+++oqaNWty/fp1XFxc+PDDDw3BRVH/fEuAUcTkN7gASExPfOQStUnpSWRoM9ChI0ObwfHbx3M8P/vIQ/ZULDOlWa4pS/q0puwd+Dvxdx675GxByX4cGp2GDpU7YGNhQ4Y2I0flc1H0ZP/hdXBwoEyZMqxYsYK4uDjWrFkDZI4QRERE8Oabb7J3717at29PlSpVsLe3N0lHecuWLcyfPx8XFxfeeOMNADp06MD58+eJj4/nzJkzdOrUyVBrA+DXX3/lzp07ue6vIPO99ce5b98+PD09AUhN/e+7EB4ezqBBg1ixYgUajYZ9+/Yxf/583N3dTfZ+CVFQNm3axJEjR2jbti21atXiyy+/JCUlhd9//92wTenSpenRowevvPIKSUlJht+RopCHLkRWWWs27d+/n5MnTwIQERHB66+/TqtWrQgICGDHjh2Gi2Znz57l/fffx8HBwbAPKPqfb0mRKuJUChWOVo48SH7wxBWh9JSKzA9vbsvUxibFGuZs6GWfYO1k5YRKqXri6kv6tKYN5zeQmJhIQloCCWkJxqMwz7Eo3qMmistcjKJPv+KFUqnMsfpF69atmTdvHk2bNuXYsWMcOnQIHx8fPv/8c8LCwujevbvRvkzRUV67di1hYWEMHDiQPXv2kJKSgpubG1ZWVkRHR+Po6Mj7779PpUqV+Omnn3jnnXfo1q2bYcJdQcv6Huknu9aoUYP9+/cTEhJiVNcjPDycHTt2MH36dEaNGkXjxo2N9iWBhSgKsn6mY2JiuH79OrVq1eLs2bP4+flhbm7Ohg0b+Pzzz4mOjmbevHk8ePAAZ2dnQyqVKVMjhTClrOl+N27cYNasWZw6dQpnZ2c6depE165dOX78OC1btuTChQuULl2aPn36EBAQgJ+fX459FAcSYBQRtua2jyyGl1tw8bg6F642rqRmpOaa6qTRaXJ0+HNbgjU/y7nmGDnIttrV8+rQP2op2cKuoSGeTN+xWLlyJbNmzWL06NE0atQIa2trrKysCAkJ4dixY0yaNInIyEiWLFlC5cqVDUXtnmXegE6nQ6PRMGnSJPr27Yuvry9ubm4kJCRw+PBhnJ2d2b59O++//z4bNmygQ4cOVK1alc2bNzN27Fj69++Pi4uLyd6LvMgagOmPu3LlykRFRXHu3DmqVKnCw4cP2bRpE2+//TYNGzakQoUK2NjYGI65uPyREi8HlUpFcnIycXFxXLp0icOHDxMVFUWzZs0MBR8PHTrEvXv3SE1NpWfPnjg7OwMSJIuiK2tQoL8YtGjRIg4ePMiOHTu4e/cuH374IZ07d+bAgQNs376dO3fu4O3tjUKhMAQXxXFZZQkwCoGdhR0JaQlGt63NrXMNMLKnPGUthveoZWj1HW0e0a/P3uF/1gnW2TvwKqXKqF3Pq0P/qOOQGhZFU9ZCQhqNhnHjxpGens67777L5cuX8fLywt/fH4CuXbvywQcfcOnSJSpUqMDYsWON9vUsHQyFQkF6ejpz5swhNjaWDz/8kI4dO6LRaJg2bRqLFy/mt99+4+bNm5QvX57ff/+dMWPGcO3aNT766CNsbW2f/k3II33Kh/4PzNGjRxk/fjzvvfceQUFBWFlZGVKefv75Z3r27MmsWbPw9vamQ4cOBAQEAMXvCph4cWXvMK1fv55ff/0VDw8Phg8fzmuvvcZ3331HQkKCYfW4CRMmcOTIEfr37/9cvndCPC21Wo29vb3ht3batGlcunSJNm3aMHLkSPbv309UVBQeHh7UrVuXevXqYWNjw9GjR1m9erXhYpBecQsuQAKMQmFtbm0UYFibW+fo9CsVmUvwZR25yFr3AqBFuRZsDt9s9Dw7CzvsreyNOtX6eRh6pu7wZ+/A1/aqzfHbx4tMh15qWBQdUVFRqNVqKlSogJmZGdeuXSM6Opo6derg4+PDgwcPaN++Pd9//73RPAJLS0u+/PJLSpcubfI2abVazMzMGDNmDFFRUcyaNQtLS0v69OnDxo0bOXr0KEOHDuWXX34xLBGYnp5O2bJlUSgUBVrQKDY2FhcXF8Mfl8jISK5evUpERARdunTh4MGDVK5cGW9vb0qVKkXfvn2xtrZm06ZN9O/f3zDJXE8CC1HYbt68ibOzs1GAcPjwYWbOnMnPP/9MWFgYly9fpm7duvTu3ZvPPvuMa9euodVqqV27NrVrZy5UUhyv6IoXX3p6Omq1mgsXLtCoUSPUajWXL1/m5s2bfP311wwePBhbW1u6du3K1q1b8fHx4fLly5QpU4YyZcrQrl07oGgUyntWEmAUAn2H/GHSQ+xt7HMdcXC1cc1RuVulNP6wlXQsiZutm9GytNbmmcWEsnaqcyvCZ0q5deA7OEqHXuT02Wef0a1bNypUqMDq1av58ccfCQkJYfv27VSsWBF7+8zUvDJlyjB37lxmzJgBZK5jb6rg4urVq+zatYuaNWtSs2ZNlEolSqUSFxcX3N3dUSgU/PDDD7i4uODv709YWBghISH4+vqSmppqVA0YCmainX70ZOPGjcybNw9vb2/mzp3L/v37SUlJwd7entmzZzNx4kROnjyJt7c3GRkZWFtb07dvX/r27WvYl6RDiaLg5s2bTJw4EXt7e6ZOncqNGzeYOXMmPXr0QKlUUrduXWxtbalSpQpTpkyhZ8+eBAYGEhQUhJmZGUql0hBUZJ0oK0RRcP36dWbMmEHt2rXp2rUr8+fPZ9asWbRq1YoHDx7g5OTEvHnzsLS0xM3NDbVazZo1a1AqlYa5Q/rPt1arLfbBBcgqUoVC3yGv41CHDpU7GEYcsheaa1GuhdESsy3Ktcixr7wUqNO/Xrdq3QyvJ8TzcvToUUP9iMTERJo3bw5kdvQXLVqEv78/N27cQKVSER4ejk6no3v37vj7+6PRaEyy3Kze6dOnGTx4MNHR0fzwww+GFTwAqlevzs6dOwkJCSE4OJhatWqRnp7Ow4cPUalUvPvuu9jZZRavLMhlZ0+fPk3Hjh1JSUlh8eLFlClThoyMDKZPn86wYcP46aefDHNPQkJCmDlzJjqdzpBupierQ4miQKvV8u2339K8eXPKlCnD1KlTuXr1Ku+++y6Wlpbs2bOHixcvkpGRQXJyMn5+ftja2houLrz66qukpKQA/6WJyGdaFCXnzp1j6NCh+Pr6Go0aR0dH07dvX2rXrs3KlSvx8vJi8eLFODg44OHhga+vL3PnzqVJkybAf5/vFyV4lhGMIiK3UQB7K3v61Xr8qhiS/iOKOisrK+bPn0/58uVp1KgR5ubmpKWlkZCQQL9+/ahSpQpjxoyhbNmybN68mT179tCkSROGDRtm8racOnWKHj16MHToUJYuXYqjo6PhsSpVqhiK9k2cOBGlUkl6ejoajQZr68yRQf1oQEH+AbC2tkatVjNu3DgALl68iK+vL2+88QYXLlygSpUq3Lp1iytXrlCtWjW6devG/fv3KVGihFHHSzphoijYuHEjFhYW9OzZ05AWdffuXRo2bMhHH33E2rVrUSqVVKpUiWnTpjFt2jQmTpwIQFJSErNmzaJFi5wX14QoKq5cuULv3r2N6jD9+OOPvPPOO1y/fh1vb2/69OnD7du36dq1K/Xr12fMmDF8//33HD58mPr16xdi6wtOoYdJv//+O9WrV8fDw4Pg4GAOHDjw2O3PnTtHu3bt8PT0pHLlykyZMiXHFc59+/YRHByMh4cHNWrUYN68eQV5CEKIx6hWrRotWrSgd+/ehhQohUJhmOcwefJkFi1axIoVK/juu+8MV3MKQvXq1Vm1ahU7duxg+fLl3Lt3j7S0NMPj7777LosWLSIxMRGFQoG5uTnW1tbPdTTAz8+Ppk2bMn78eD777DO++eYb7t27R+XKlQGwsbGhZMmSrFy5EoDBgwfj5uYmAYUoktq1a8d7773H559/zurVq4mMjCQxMdEQbPj6+rJlyxb69u2Ls7MzSUlJ2NjYYG9vj6WlJV988QWdOnUq5KMQ4tFcXFz4448/WLZsGQMGDGD58uWYmZnRuXNnjh07BkDv3r1JTEzk888/Z8yYMQCMHj36uS1tXhgKNcBYtWoVY8eO5cMPPyQ0NJR69erRvXt3bty4kev2arWaLl264O7uzs6dO5k8eTLTp083DKUCXLt2jddee4169eoRGhrKBx98wOjRo1m7du3zOiwhRDYdO3YEYPXq1fz9999ER0fTsWNHtFot7dq1w9bWlldffdXQ6TBlWlRWAQEB9O7dmx9//JH69evz119/cfz4ccNr+vj48NNPPxkKGuk77c+7896zZ09WrlyJlZUVf/75J6VLl8bNzY0TJ06g0+l4//33GT16tGH7gnq/hHhWKpXKkFPeo0cPVq1aRc2aNdm6dSsnTpxg3rx5hpS/CRMmYGNjY8g/V6lUz30JaCHyq06dOgwYMIDQ0FAqVqzInj172LZtGz4+Pixbtox33nkHKysrPv30UwICAtBqtWi1WurWrYuPj09hN7/AFGqK1C+//ELv3r15/fXXAZg6dSo7duxg3rx5jB8/Psf2y5cvJzk5md9++w1ra2v8/f25dOkSv/76KyNGjEChUDB//nw8PT2ZOnUqABUrVuTYsWPMmDGDkJCQ53p8QohMXl5ejB07FnNzc1QqFUOHDiUwMJChQ4fSs2dPnJycgOezjGq5cuUYMGAAr776Kj/88EOO6tsVK1YssNfOq1KlStGvXz969eoFwObNmwkKCqJevXqkpqZiYWFhmOwqy86Kok6fUtikSROWLl2KtbU1Y8aMYfv27bi4uPD2228btpXVoURxY2VlRf/+/Q0pUvqVBmvUqMHEiRMNtSyAl2qBgkILMNLS0jh16hQjR440ur9Zs2YcPnw41+ccOXKEhg0bGvKhAZo3b87XX3/N9evX8fHx4ciRIzRr1szoec2bN+evv/4iPT3dqMJtYZl/bL7h//cd2weAm60b6hQ1qZrnW2XaSmWFhcqCxPREo5obChSolCpalGuBraUtoVdDcxTiU6eo2XVlF3HJcWj5b9KrEiVO1k40K9cMHbrHPzclDgUKbM1tMVOZkaHNQKlQok5Ro0NnaENJx5JG7c76fP1KvjYWNlibWxsVClSnqB/5+rndnx9Z92GmNMss2KbTYGlmSeUSlTlw4wAareaRx/Ciy/r+WKgseJj0kPK+5enYsSPVqlXD0tLScHVS36koqI5y1rZcuXaF+IvxWFlZcfjwYcPV08LupGf/TFpYW7BmzRrOnDlDVFQUVapUMQQceoXdZiEeJbffWDs7O8zMzDh9+jQNGzakQYMGhs+w/jfgZel8ieJP/xl/mPSQC2EXaFiqIaeOnuLo0aOG5ZT1wYV+2dmX6Te70L7JMTExaDQa3NzcjO53c3Pj3r17uT7n3r17uW6vf+xx22RkZBATE2Oq5ptcdGL0cw8uAFI0KajT1DkK+unQkaHNYPuV7YReDSU6MRp1qproxOjMJXWB0KuhxCbHGgUXAFq0xCbHsufqnic/V6dFo9OgTlMTmxyLOlVNXEpmwJK1Ddllfb7+n4S0BKPX0G/3qNfP7f78yLqP2ORYHqQ8MOwv9HooGdqMxx7Diy7r+3M/6T425W3Yty8zoK5YsaLR0HBBdyqytqVExRIo3BX8888/jBw5ktatWxfoa+dV9s+kfSV7Nm/eTKtWrVi/fj3e3t6F3UQh8iy331hvb29eeeUVw3df39l6ma7qiheH/jOeokvhftJ9Ro4dyZw5c/jggw9yLEzwIiw7m1+FvopU9mjuSWu257Z99vvzsk124eHheWvwSyZDm8HDpIdG9z1Mekh4eHiO+7PL7fG8Pjd7G7Kfn8c9X/8auW33qNfP+py8etpjeFk+a9nfHwcPBwYOHFgox5+1LQqFgiqBVajTpg6AYWncgr6y9KTjzv5+2bra8sMPPxiW75XUkeLnZfmu5+ZRv7H16tVDrVajVqsLqWUF62U+5y+b7J/xlr1a8r+S/wOez9+VrKlXRVGhBRiurq6oVKocoxX379/PMQKh5+7unuv28N9IxqO2MTMze+xksed5ovRpUcWBmdIMe2t7UhJTDPfZ29jj5+fHxfMXje7Pzt4mM+XoaZ6bvQ3Zz8/jnq9/jdy2e9TrZ31OXj3NMYSHhxf5HwVTye09rlSpUqEUfnvU+X5enfa8nPdHvV8SWBRPL9N3PTeP+419UYs/vuzn/GWT/TPu7uaOn5/fC1GF2xQK7a+WhYUFAQEB7Nq1y+j+Xbt2PXJN4Hr16nHw4EFD0R399iVLlqRs2bKGbXbv3p1jnzVr1iwS8y8exc3WDUuV5XN/XSuVFQ4WDqgUxl+GrMX9HlXML9g3GBdrF5TZPkZKlLhYuxDsG/zk5yqUqBQqHCwccLF2wcHSAScrJ5Qon1hgUP98/T92FnY5ig0+7vWfVKDwSbLuw8XaBWcrZ8P+gsoGPbFI4ovuUe9xYXQsHtWWotRxLw5tFCKvHvcb+yIGF+Llo/+MWymsjD7jElxkUsTFxRXa+oarVq1i6NChfP/999SvX5958+axaNEiDh48SJkyZZgwYQLHjx9n3bp1AMTHx1O3bl0aN27MqFGjuHz5MsOHD2f06NGGyeLXrl0jMDCQ/v37M3DgQA4fPsyHH37I77//XuRWkZKrHS8fOecvJznvLx855y8fOecvJznvuSvUORhdu3YlNjaWqVOncvfuXSpXrsyyZcsoU6YMAFFRUVy9etWwvaOjI6tXr2bUqFE0bdoUJycnhg8fzogRIwzb6Ncd/vjjj5k3bx6enp5MmTKlyAUXQgghhBBCvIgKdQTjZSdR78tHzvnLSc77y0fO+ctHzvnLSc577iS5VwghhBBCCGEyEmAIIYQQQgghTEYCDCGEEEIIIYTJSIAhhBBCCCGEMBkJMIQQQgghhBAmIwGGEEIIIYQQwmQkwBBCCCGEEEKYjAQYQgghhBBCCJORAEMIIYQQQghhMhJgCCGEEEIIIUxGAgwhhBBCCCGEySji4uJ0hd0IIYQQQgghxItBRjCEEEIIIYQQJiMBhhBCCCGEEMJkJMAQQgghhBBCmIwEGEIIIYQQQgiTkQBDCCGEEEIIYTISYJjI77//TvXq1fHw8CA4OJgDBw48dvtz587Rrl07PD09qVy5MlOmTEGnM17Qa9++fQQHB+Ph4UGNGjWYN29eQR6CeAqmPu979+7Fyckpx7+XLl0q6EMReZSfc56SksLbb79NYGAgJUqUoH379rluJ9/1os3U51y+58VDfs773r176dWrFxUrVqRkyZIEBgaycOHCHNvJd71oM/U5f5m/6xJgmMCqVasYO3YsH374IaGhodSrV4/u3btz48aNXLdXq9V06dIFd3d3du7cyeTJk5k+fTozZswwbHPt2jVee+016tWrR2hoKB988AGjR49m7dq1z+uwxBMUxHnXO3ToEBcvXjT8W65cuYI+HJEH+T3nGo0GKysrhgwZQqtWrXLdRr7rRVtBnHM9+Z4XXfk970eOHKFKlSr8+eefHDx4kMGDB/Pee++xfPlywzbyXS/aCuKc672M33Wpg2ECzZs3p0qVKkybNs1wX61atQgJCWH8+PE5tp87dy5ffPEFly5dwtraGoCpU6cyb948wsLCUCgUjB8/nvXr13PixAnD80aOHMmFCxfYtm1bwR+UeKKCOO979+6lY8eOXLlyBVdX1+d2LCJv8nvOs/roo48ICwtj48aNRvfLd71oK4hzLt/zou9ZzrvegAED0Gg0hqva8l0v2grinL/M33UZwXhGaWlpnDp1imbNmhnd36xZMw4fPpzrc44cOULDhg0NnUzI/GDfuXOH69evG7bJvs/mzZtz8uRJ0tPTTXwUIr8K6rzrNWnShIoVK9KpUydCQ0NNfwAi357mnOeFfNeLroI653ryPS+aTHXeHz58iJOTk+G2fNeLroI653ov43ddAoxnFBMTg0ajwc3Nzeh+Nzc37t27l+tz7t27l+v2+scet01GRgYxMTGmar54SgV13j09Pfnhhx9YuHAhCxcuxM/Pj5CQEPbv318ARyHy42nOeV7Id73oKqhzLt/zos0U533z5s3s2bOHAQMGGO6T73rRVVDn/GX+rpsVdgNeFAqFwui2TqfLcd+Tts9+f162EYXL1Ofdz88PPz8/w+P16tUjMjKS6dOn06hRI1M1WzyD/J7zp91nbveLwmHqcy7f8+Lhac/7oUOHePPNN5kyZQq1a9d+4j5zu18UDlOf85f5uy4jGM/I1dUVlUqVI8K9f/9+jkhYz93dPdft4b8r2o/axszMDBcXF1M1XzylgjrvualduzYRERHP2GLxrJ7mnOeFfNeLroI657mR73nR8Szn/eDBg3Tv3p1x48YxePBgo8fku150FdQ5z83L8l2XAOMZWVhYEBAQwK5du4zu37VrF/Xr18/1OfXq1ePgwYOkpKQYbV+yZEnKli1r2Gb37t059lmzZk3Mzc1NexAi3wrqvOfmzJkzeHh4mKbh4qk9zTnPC/muF10Fdc5zI9/zouNpz/v+/fvp3r07o0ePZtiwYTkel+960VVQ5zw3L8t3XQIMExg+fDhLlixhwYIFXLx4kTFjxhAVFcXAgQMBmDBhAp06dTJs/+qrr2Jtbc2wYcMICwtj3bp1/PTTTwwbNswwFDdw4EBu377N2LFjuXjxIgsWLGDJkiWMGDGiUI5R5FQQ5/3XX39lw4YNXLlyhfPnzzNhwgQ2btzIm2++WSjHKIzl95wDXLhwgdOnTxMTE0NiYiKnT5/m9OnThsflu160FcQ5l+950Zff87537166d+/OwIEDee2117h79y537941jFKDfNeLuoI45y/zd13mYJhA165diY2NZerUqdy9e5fKlSuzbNkyypQpA0BUVBRXr141bO/o6Mjq1asZNWoUTZs2xcnJieHDhxv9yPj4+LBs2TI+/vhj5s2bh6enJ1OmTCEkJOS5H5/IXUGc9/T0dD777DPu3LmDlZWVYZ9PWk9fPB/5PedAjnXUg4KCAIiLiwPku17UFcQ5l+950Zff875kyRKSkpKYPn0606dPN9zv7e3NmTNnAPmuF3UFcc5f5u+61MEQQgghhBBCmIykSAkhhBBCCCFMRgIMIYQQQgghhMlIgCGEEEIIIYQwGQkwhBBCCCGEECYjAYYQQgghhBDCZCTAEEIIIYQQQpiMBBhCCFHEnTp1irZt21K6dGmcnJzYu3dvYTcpzxYvXoyTkxPXr18v1HZMmjQJJyenQm2DEEK8LKTQnhBCFGEajYaBAwei1Wr58ssvsbW1pWLFioXdrCLp1q1bLFiwgPbt21O9evXCbo4QQry0ZARDCCGKsJs3b3L16lWGDh3KoEGD6NGjB+7u7oXdrCLp9u3bTJkyxVBFN6uPPvqIqKioQmiVEEK8fCTAEEKIIuz+/fsAODo6FnJLcpeWloZGoymQfSclJZlsX2ZmZlhZWZlsf0IIIR5NAgwhhCii3n77bZo3bw7A8OHDcXJyolq1arluq9PpeOWVV/jwww+N7q9SpQpOTk5cu3bNcN/KlStxcnLi0qVLhvvCwsLo2bMnZcqUoWTJkrRs2ZJt27YZ7Wvv3r04OTmxbNkyvv32W6pWrYqnpyc3b94E4OjRo7Rq1QoPDw+qVq3Kjz/+iE6ny9Ox6udIXLhwgbfeegtfX18aNGgAQGRkJB9++CF169alZMmSlClThh49enD+/HmjtrVs2dLovXJycmLSpElG+89uwYIFBAYG4uHhQfny5Rk6dCh37tzJU5uFEELkTuZgCCFEETVw4EDKli3L5MmTGTBgAA0bNsTW1jbXbRUKBQ0aNGD//v2G+65du8atW7dQKpXs378fHx8fAA4cOECJEiWoUKECAJcvX6ZNmzZYWFgwbNgwbG1tWbJkCT169ODPP/+kY8eORq/1448/olAoGDp0KAB2dnZcuHCBzp07Y29vz6hRo7CwsOCPP/54ZHsfd8xlypThk08+IS0tDYCTJ0+yf/9+OnbsSJkyZbhz5w7z58+nXbt2HDp0CA8PDypWrMjYsWON3ivIDLAe5ccff2TChAkEBgby5ZdfcvPmTebMmcPBgwcJDQ2VSeFCCPGUJMAQQogiql69eigUCiZPnkzdunXp0aPHY7cPDAxk06ZNxMTE4OrqysGDB3FwcKBu3bocPHiQPn36AJkBhr4DDvDll1+SlJTE9u3bDUHH66+/TmBgIOPGjaN9+/Yolf8NeD98+JDDhw8bBQ/vvfceaWlpXmNroQAAEDBJREFUbNq0CV9fXwD69OlDrVq18nXM5cuXZ+HChUb3tWzZkpCQEKP7evToQcOGDVm4cCGjRo3C3d2d5s2b5/m9iomJYfLkyTRu3Jg1a9ZgZpb557BBgwb06dOHGTNm8Omnn+ar7UIIITJJipQQQrwgAgMD0el0HDhwAMgMJOrXr0/jxo0N98XGxnLhwgVDgKHRaNixYwdt2rQxBBcADg4ODBo0iJs3b3Lu3Dmj1+nZs6dRcJF1H/rgAqBEiRK89tpr+TqGwYMH57jPxsbG8P9JSUnExsbi6OhIuXLlOHXqVL72r7d7925SU1MZNmyYIbgAaN++PX5+fmzZsuWp9iuEEEICDCGEKHbu3r1r9G9ycjIANWrUwM7OzhBMHDx4kMDAQAIDA4mIiODu3bscOHAAnU5HYGAgkDmJPDEx0Si40NMvhxsZGWl0vz7VSu/+/fskJSXh5+eXYx/ly5fP17Fl3zdASkoKn3/+OZUqVcLLy4tXXnmFcuXKERYWRnx8fL72r6c/ptyOu0KFCjmOWQghRN5JipQQQhQz2etg/PLLL/Tp0weVSkXdunU5cOAA9+7d4/LlywQGBlKzZk2sra05cOAAR48excHB4ZGTxbN61ARta2vrXLdTKBR53sejZN83wNixY1mwYAFDhgyhQYMGODg4oFQqGTduHFqtNl/7z4v8tlkIIYQxCTCEEKKYWbNmjdHtSpUqGf4/MDCQyZMns2XLFqytralZsyYWFhbUrl3bEGDUq1cPlUoFZKYx2draGq0opRceHg5AmTJlHtseNzc3bGxsct3HlStX8nt4OaxatYqePXsyefJko/vj4uJwcXEx3M4twHkU/TFdunQpxyhLeHj4E49ZCCHEo0mKlBBCFDNNmjQx+tfT09PwWGBgIBqNhmnTplGnTh0sLCwAaNiwIdu3b+fMmTOG9CgAlUpF8+bN2bJlC5cvXzbc//DhQ+bPn0/p0qUfuxKTfh/NmjVj8+bNXL161XD//fv3Wb58+TMfr0qlyjGqsGLFihzLyernasTFxT1xn02aNMHS0pKZM2ca1fHYtGkT4eHhtG7d+pnbLYQQLysZwRBCiBdI7dq1sbS0JDw8nK5duxrub9SoEVOnTgUwWkEK4LPPPmP37t20bduWN954w7BM7c2bN/njjz+MVpB6lI8//pidO3ca9mFubs4ff/yBt7f3U8+T0Gvbti1Lly7F3t4ef39/zpw5w6pVq3LM1yhXrhwODg7MmzcPOzs77OzsqFy5Mv7+/jn26erqytixY5kwYQIhISF07NiRW7duMXv2bMqUKcOIESOeqc1CCPEykxEMIYR4gVhZWRmWhs06UlG3bl1DNevsS8f6+fmxefNmatWqxS+//MLEiROxtbVl6dKldOrUKU+v6+/vz+rVq/H29mbq1KnMmTOH/v3789Zbbz3zMU2ePJl+/fqxevVqPv74Y86fP8/KlSspVaqU0XaWlpbMmjULS0tLRo0axeDBg1m7du0j9/v+++8zbdo0YmNj+eyzz1i0aBEhISFs3rxZamAIIcQzUMTFxclsNiGEEEIIIYRJyAiGEEIIIYQQwmQkwBBCCCGEEEKYjAQYQgghhBBCCJORAEMIIYQQQghhMhJgCCGEEEIIIUxGAgwhhBBCCCGEyUiAIYQQQgghhDAZCTCEEEIIIYQQJiMBhhBCCCGEEMJkJMAQQgghhBBCmIwEGEIIIYQQQgiTkQBDCCGEEEIIYTISYAghhBBCCCFMRgIMIYQQQgghhMlIgCGEEEIIIYQwGQkwhBBCCCGEECYjAYYQQgghhBDCZCTAEEIIIYQQQpiMBBhCCCGEEEIIk5EAQwghhBBCCGEyEmAIIYQQQgghTEYCDCGEEEIIIYTJmD3Lk9VqNQsXLsxxf3p6Oubm5ob/Zn8MyHG/PCaPyWPymDwmj8lj8pg8Jo/JY0X/MYB+/frh4OCQ62OKuLg4Xa6P5IFWq0WtVj/t04UQQgghhBDFkIODA0pl7slQzxRgCCGEEEIIIURWMgdDCCGEEEIIYTISYAghhBBCCCFMxuQBxu3btzl+/DiA4b8As2fPztPz87rds1i5ciVpaWnPtI+sx/kkycnJnDt3LtfHli9fzoMHD56pLUIIIYQQxdX27duJiIjI8/br16/P92ucOHGCX3/9lVmzZrF169Z8P/9ZpaWlsWrVKmbPns3MmTPZsWNHvp7/NMdcmJ5pFanceHl54eXlBWQGGLVr1zb1SzyztLQ0LCwsnvr5Wq3W6DifJCUlhbCwMKpUqZJjP0IIIYQQIm+0Wi0dO3bM9/MOHTrEW2+9hVKpJDk5+alf+1GTmp9k+/btlC9fnq5duwJw7dq1fD3/aY65MD1TgJGYmMiSJUswMzNDpVLRuHFjACIiInB0dOTOnTvMnj2bLl26oNVqWbt2LdevX6d169ZUrFiR2bNnM2TIEB48eMD27dvp3r07Wq2WlStXcufOHVq2bEnFihUNr/fw4UOWL19ORkYG5cqVo0mTJixbtoyHDx9ib2/Pa6+9xqlTp7h06RLJyck4Ojri6OjIhQsXqFevHvXq1ePu3bu4u7vz4MEDVq5ciZWVFWq1mgYNGnD8+HGcnJzo3r07Fy5cIDQ0lNTUVJo3b46/vz/Lly/H0tKS2NhYgoKCiIiIoEWLFqxbt447d+5gaWlJz549uXv3Lhs2bMDc3JyaNWvy4MEDLl26xOzZs+nfvz+zZ8/G3d2dUqVKARAaGkpUVBTlypWjRYsWHD16lJMnT5KWlkanTp0oU6YM586dY8+ePZibm9OmTRu8vb2f5dQJIYQQQhQ5kZGRnD9/ntatWxMXF8fWrVupU6cO+/btQ6vV0rhxY3bu3MmQIUO4du0amzdvRqVS0ahRI3Q6Hbt27cLCwoKgoCAqVapk2G9aWho3b97E29sba2trgP9r715fmv77OI4/v3OectM2pzg0j5WRSEVqiZAHDKkgi8oyygPioRvd6U+pOxoeZiIo3gi0bmg5lKjoaKAGnUkzFkzHdJqHtf1uyL7XptZ1kV7QL96PO3Pue/rse+f13vvz2RgZGeHx48doNBpKS0sJCwtT8+jHjx/VnHf9+nU1txmNxoA8FhoaSm9vL263m6ysLLKyshgaGuLgwYPo9Xr1/F++fOH48ePq8+TkZGC10Ojv78fj8VBUVER6ejqNjY2YzeYNM7PvEVD/7u/v59OnTyiKQnl5ecBXx/pe02q1VFVVMTo6GjBms9nMjRs3iI2NxWazUVhYyLNnz1AUherqahRF+a37uKkC4/nz5xw+fJjMzEwsFkvAa9nZ2YyMjKhvwsLCAsXFxbjdbnp7ewMKB38ul4vy8nLCwsKwWCwB2w0NDZGfn09aWhoej4fx8XHi4uIoLy/HarXy+vVrAKKioigvL6e1tZWMjAwKCgpoaWkhJyeH9+/fs2vXLgCWlpaoqalheHiYb9++UVtbS2trK9+/fyc1NZU9e/awtLRER0cHe/fuBSAlJYWTJ0+qrbzJyUncbjf19fWMjIzw9OlTFhcXKSkpYefOnXg8HpxOJ06nk3PnzgGrvx/S0NBASEgIPT09pKSkUFpaSltbG3Nzc+zbt4/s7GxmZmbo7+8nISGB4eFhamtrCQ4Ols6HEEIIIf5KiYmJ3Lt3D4DR0VEyMzOB1e5BVVUVAFarFYCBgQEuX75MREQEHo+Hnp4eLly4gMlkWpeVTp8+jdVqxW63U1JSQkZGBo8fP6a+vp7Z2Vn6+vp+2iXw5TatVktjY2NAHuvq6qKsrAy9Xk9bWxsHDhygoKBg3TF8QX1lZYW2tjZcLhfXrl3DarVSXV1NUFAQ7e3tpKen/8+Z2efTp0/U1dWh0Wjwev/z5bBfv37F6XTS0NCA1+vF6/WuG3NFRQUul4srV64wNjbG2NgYNTU13L59G5vNhtls/u83bQObKjBmZmbU4B0XF/fLbXU6HREREcDqlCF//m9GREQEUVFRAOvaUHa7ncTERPW16elptQsQHx+PzWZDp9MRGxsLgF6vJzY2Fq1Wq97YL1++kJubi9PpJCYmBkVR0Ov16HQ6dZ/FxUUcDgeDg4N4vV5cLpd6Db7z+b8H/tfw6NEjioqKsFqtvHjxgry8PHXcPjExMQFTtHw3z2w243A4cDqdPHr0CEVRUBSF+fl5DAaD+kMnv9ueE0IIIYT40yUkJDA1NcW7d+/Izc1lYmJiXf6C1dDuy1gajYaioiKGh4fVbkB0dLS6bVJSElVVVXz//p3m5maSk5MxGAwEBQVhMBhYWloKOLZ/NvXltrm5uXV5zG6309XVBax+mL6wsBDQuVgrODhY7UQA2Gw29UP6hYUF4NeZeaOOQmFhIT09PWzbto2SkhI1Y9rtdpKSktT9XC7XhmOOjo5Gq9USGRlJTEwMsJqHf3cqGWxykbfRaMRmswGoj/5+1lbx3TTfLwT67zs/P8/s7CzLy8vrqk+TycTk5CSwWskajUampqYAmJqawmg0rjuv/98/fvxAURQ1oP9sO1idtlRWVkZFRcUvt9voGsLDwzl16hTHjh1jcHAQjUYTMJa1x/B/Dw0GAw8ePKCmpoYzZ87g9XqJiIjA4XDgdrvVsQshhBBC/I3279/P0NAQUVFRaLWrn4X/LFP6QrDH42H79u2cOXOGnJwcHj58GLDd9PQ0ACEhIQQFBanZ6sePHzgcDkJDQ9Vp8xCYTX3n3iiPmUwmLl68SF1dHVevXv1pcREfH8/o6Kj63JflzGYz1dXV6v5r+Rc6gHqNLpeLubk5AFJTUzl//jw6nY63b9+q25pMJiYmJtTjbDTmtX53StRam+pgZGVl0dnZqc7VWhuk9Xo9nZ2dHDt2bMP9d+/eTWNjIwkJCer/dDodAwMD6hoMQG1bFRQU0NPTw/3799U1GN3d3dy8eROdTkd+fj6vXr366fVOTk4GnOtX9u7dS3t7e8BcvY3s2LGDly9f0tTUREhICBcuXODJkyeMj4+zvLxMQUEBer2e+fl5Ojs7OXv27LpjfPjwgYcPH5KWloZer2fXrl00NTWRlpYGrFbIR44cUc8hazCEEEII8Te5e/eumrcqKyux2+0BaxY2cvToUSwWC1qtlry8PD5//szExAQrKyucOHEiYNu+vj4WFxfxer3k5uai0Wg4dOgQTU1NKIpCaWkp4eHhREVF0dzcjNFoDFjLABvnseLiYrq7u/F4PISHh3Pp0qUN12AUFxdz584ddf1DSkoKAEVFRbS1taEoCrGxsZSWlv5yzNnZ2VgsFhITE9UuR0dHBysrKyiKQnZ2Nm/evAEgPT2dyMhIGhsb0Wq1VFZWrhvz/8umfsnbV0xoNBosFgunT59Wpzf9iSYmJtSF30IIIYQQ4s/U2tpKVVWVTAv/l9pUB2NpaYlbt27hdrvZuXPnHx/cfes3hBBCCCHEn6mlpYX09HQpLv7FNtXBEEIIIYQQQgh/UhoKIYQQQgghtowUGEIIIYQQQogtIwWGEEIIIYQQYstIgSGEEEIIIYTYMlJgCCGEEEIIIbaMFBhCCCGEEEKILfMPw1XuNzHP2yYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 792x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "style.use('fivethirtyeight')\n",
    "\n",
    "# text rotation angle\n",
    "ROT_ANGL = 30\n",
    "\n",
    "# plot the graph\n",
    "profanity_plt = profanity_df.plot(x='f_ratio',\n",
    "                                  y='s_ratio',\n",
    "                                  kind='scatter',\n",
    "                                  figsize=(11,7), \n",
    "                                  title='S-word/F-word [Eminem]',\n",
    "                                  color='darkseagreen') \n",
    "# top f-word and s-word ratios\n",
    "top_f_ratio_df = profanity_df.sort_values(['f_ratio'], ascending=False)[:14][['title', 'f_ratio', 's_ratio']]\n",
    "top_s_ratio_df = profanity_df.sort_values(['s_ratio'], ascending=False)[:8][['title', 'f_ratio', 's_ratio']]\n",
    "top_fs_ratio_df = pd.concat([top_f_ratio_df, top_s_ratio_df]).drop_duplicates()\n",
    "\n",
    "# plot annotations\n",
    "top_fs_ratio_df.apply(lambda row: profanity_plt.annotate(row.title, \n",
    "                                                         (row['f_ratio'], row['s_ratio']),\n",
    "                                                         fontsize=8, \n",
    "                                                         fontstyle='oblique', \n",
    "                                                         fontweight='roman',\n",
    "                                                         rotation=ROT_ANGL),\n",
    "                                                         axis=1)\n",
    "# legend\n",
    "profanity_plt.legend(['song'], loc=1)\n",
    "\n",
    "# x-label, y-label\n",
    "profanity_plt.set_xlabel('f-word ratio')\n",
    "profanity_plt.set_ylabel('s-word ratio')\n",
    "\n",
    "# signature bar\n",
    "profanity_plt.text(x=-0.032, y=-0.017, s=str('_' * 110).expandtabs(), color='grey')\n",
    "profanity_plt.text(x=-0.032, y=-0.02,\n",
    "                   s=str(\"github.com/mariostrbac \" + \"  \" * 112 + \" Lyrics Source: Genius.com\").expandtabs(),\n",
    "                   fontsize=7.9,\n",
    "                   color='dimgrey')\n",
    "\n",
    "plt.savefig('../reports/figures/vocabulary-analysis/profanity_ratios_graf.png', dpi=240, bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "p3_4IFGVKIdM"
   },
   "source": [
    "### 3.5.5 Profanity words through the years  <a name=\"355-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "l-yXvIlOKIdN"
   },
   "source": [
    "The following graph shows the profanity ratio for each song, along with their release date. The profanity words considered for the ratio are the top five most frequent profanity words in the corpus, from [Section 3.5.1.](#351-bullet). The relative frequency of profanity words in lyrics is calculated as follows:\n",
    "\n",
    "$$ f_\\text{profanity}\\ =\\text{ ... }\\ = \\frac{N_\\text{profanity words}}{N_\\text{all words}}$$\n",
    "\n",
    ", where *N* is the number of words in lyrics. So, for example, if the profanity ratio is 0.20 that means 20% of words in the song are profanity words."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "id": "2PZtj2oDKIdN"
   },
   "outputs": [],
   "source": [
    "# calculate the profanity ratio for each song\n",
    "profanity_df['profanity_ratio'] = profanity_df.apply(lambda row: len(row.profanity_words) / len(row.lyrics), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {
    "id": "78DHxHNhKIdN"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 792x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "style.use('fivethirtyeight')\n",
    "\n",
    "N_TOP = 8\n",
    "N_TOP_POP = 6\n",
    "\n",
    "# only songs with release date information\n",
    "profanity_rdate_df = profanity_df[profanity_df.release_date.isnull()==False]\n",
    "\n",
    "# plot the graph\n",
    "profanity_plt = profanity_rdate_df.plot(x='release_date',\n",
    "                                        y='profanity_ratio',\n",
    "                                        kind='scatter',\n",
    "                                        figsize=(11, 7), \n",
    "                                        title='Profanity through the years [Eminem]',\n",
    "                                        color='darkseagreen')\n",
    "\n",
    "# highlight songs with highest profanity ratios\n",
    "profanity_rdate_df.sort_values('profanity_ratio', ascending=False)[:N_TOP] \\\n",
    "                  .plot(x='release_date',\n",
    "                        y='profanity_ratio',\n",
    "                        kind='scatter',\n",
    "                        figsize=(11, 7), \n",
    "                        color='forestgreen',\n",
    "                        ax=profanity_plt)\n",
    "\n",
    "# highlight the most popular songs in the corpus                        \n",
    "profanity_rdate_df.sort_values('popularity')[:N_TOP_POP] \\\n",
    "                  .plot(x='release_date',\n",
    "                        y='profanity_ratio',\n",
    "                        kind='scatter',\n",
    "                        figsize=(11, 7), \n",
    "                        color='forestgreen',\n",
    "                        ax=profanity_plt)\n",
    "\n",
    "# annotate songs with highest profanity ratios\n",
    "profanity_rdate_df.sort_values('profanity_ratio', ascending=False)[:N_TOP] \\\n",
    "                  .apply(lambda row: profanity_plt.annotate(row.title, \n",
    "                                                            (row.release_date, row.profanity_ratio),\n",
    "                                                            ha='right',\n",
    "                                                            fontsize=8, \n",
    "                                                            fontstyle='oblique', \n",
    "                                                            fontweight='roman',\n",
    "                                                            rotation=0), \n",
    "                                                            axis=1)\n",
    "# annotate the most popular songs in the corpus\n",
    "profanity_rdate_df.sort_values('popularity')[:N_TOP_POP] \\\n",
    "                  .apply(lambda row: profanity_plt.annotate(row.title, \n",
    "                                                            (row.release_date, row.profanity_ratio),\n",
    "                                                            ha='right',\n",
    "                                                            fontsize=8, \n",
    "                                                            fontstyle='oblique', \n",
    "                                                            fontweight='roman',\n",
    "                                                            rotation=0), \n",
    "                                                            axis=1)\n",
    "# plot the average line\n",
    "profanity_plt.axhline(y=np.round(profanity_rdate_df['profanity_ratio'].mean(), 2), \n",
    "                      ls='--', \n",
    "                      color='darkslategrey', \n",
    "                      linewidth=1.3, \n",
    "                      alpha=0.3)\n",
    "\n",
    "# the average line annotation\n",
    "profanity_plt.text(x='1987', \n",
    "                   y=np.round(profanity_rdate_df['profanity_ratio'].mean(),2) + 0.001, \n",
    "                   s='average', \n",
    "                   fontsize=8, \n",
    "                   color='darkslategrey')\n",
    "\n",
    "# change number of ticks on y-axis\n",
    "profanity_plt.set_yticks([0.0, np.round(profanity_rdate_df['profanity_ratio'].mean(),2), 0.05, 0.1, 0.15, 0.2, 0.25])               \n",
    "\n",
    "# labels\n",
    "profanity_plt.set_xlabel('release date')\n",
    "profanity_plt.set_ylabel('profanity ratio')\n",
    "\n",
    "# signature bar\n",
    "profanity_plt.text(x='1983', y=-0.045, s=str('_' * 108).expandtabs(), color='grey')\n",
    "profanity_plt.text(x='1983', y=-0.054,\n",
    "                   s=str(\"github.com/mariostrbac\" + \" \" * 220 + \"Lyrics Source: Genius.com\").expandtabs(),\n",
    "                   fontsize=7.9,\n",
    "                   color='dimgrey')\n",
    "\n",
    "plt.savefig('../reports/figures/vocabulary-analysis/songs_profanity_through_years_eminem.png', dpi=240, bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "6oVBAaxzKIdO"
   },
   "source": [
    "## 3.6 Word count by the first letter (no stopwords)  <a name=\"36-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "id": "Df5iE9iSKIdO",
    "outputId": "0266ef5a-2365-4dfd-f81d-55aa6452643a"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "style.use('fivethirtyeight')\n",
    "\n",
    "# init an empty dict object\n",
    "word_distribution = defaultdict(list)\n",
    "\n",
    "# sort words by their first letter\n",
    "for word in words_df.word.values:\n",
    "    word_distribution[word[0]].append(word)\n",
    "\n",
    "\n",
    "fig, ax = plt.subplots(1,2, figsize=(13,6))\n",
    "\n",
    "# plot the word distribution (with duplicates)\n",
    "pd.DataFrame(word_distribution.items(), columns=['letter', 'words']) \\\n",
    "               .apply(lambda row: pd.Series([row.letter, len(row.words)], index=['letter', 'word counts']), axis=1) \\\n",
    "               .sort_values('word counts', ascending=False) \\\n",
    "               .plot(x='letter', \n",
    "                     kind='bar',\n",
    "                     title='Word distribution',\n",
    "                     legend=False,\n",
    "                     xlabel='',\n",
    "                     rot=0,\n",
    "                     ax=ax[0]) \n",
    "\n",
    "# plot the unique word distribution (no duplicates)\n",
    "pd.DataFrame(word_distribution.items(), columns=['letter', 'words']) \\\n",
    "              .apply(lambda row: pd.Series([row.letter, len(set(row.words))], index=['letter', 'word counts']), axis=1) \\\n",
    "              .sort_values('word counts', ascending=False) \\\n",
    "              .plot(x='letter', \n",
    "                    kind='bar',\n",
    "                    title='Unique word distribution',\n",
    "                    legend=False,\n",
    "                    xlabel='',\n",
    "                    rot=0,\n",
    "                    ax=ax[1]) \n",
    "\n",
    "# title\n",
    "fig.text(x=0.5, y=0.95, s='Word distribution by the first letter [Eminem]', ha='center', fontsize='x-large', fontweight=555)\n",
    "\n",
    "# labels\n",
    "fig.text(0.5, -0.02, 'letter', ha='center')\n",
    "fig.text(0.0, 0.5, 'count', va='center', rotation='vertical')\n",
    "\n",
    "# signature bar\n",
    "fig.text(x=-0.015, y=-0.055, s=str('_' * 129).expandtabs(), color='grey')\n",
    "fig.text(x=-0.015, y=-0.08,\n",
    "         s=str(\"github.com/mariostrbac\" + \"  \" * 139 + \" Lyrics Source: Genius.com\").expandtabs(),\n",
    "         fontsize=7.9,\n",
    "         color='dimgrey')\n",
    "\n",
    "plt.savefig('../reports/figures/vocabulary-analysis/word_distribution_first_letter.png', dpi=240, bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "CG1aOKH_KIdP"
   },
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "k3cPvNQbKIdQ"
   },
   "source": [
    "# 4. Lyrics sentiment analysis  <a name=\"4-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {
    "id": "TQvHuGwjKIdQ"
   },
   "outputs": [],
   "source": [
    "sentiment_df = data_df.loc[:, ['title', 'featured', 'release_date', 'album_id', 'popularity', 'lyrics']]\n",
    "\n",
    "# split lyrics into words and clean contractions and dash words\n",
    "sentiment_df['lyrics'] = sentiment_df.apply(get_only_eminems_verses, axis=1) \\\n",
    "                                     .apply(split_lyrics_into_words) \\\n",
    "                                     .apply(dash_and_contractions_cleaning)\n",
    "# calculate the average subjectivity for each songs                                    \n",
    "sentiment_df['subjectivity']= sentiment_df.lyrics \\\n",
    "                                          .apply(lambda lyrics: np.mean([TextBlob(word).sentiment.subjectivity for word in lyrics]))\n",
    "# calculate the average polarity for each song                                          \n",
    "sentiment_df['polarity'] = sentiment_df.lyrics \\\n",
    "                                       .apply(lambda lyrics: np.mean([TextBlob(word).sentiment.polarity for word in lyrics]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "erRCSTsXKIdR"
   },
   "source": [
    "## 4.1 The subjectivity/polarity of lyrics  <a name=\"41-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "1dQ39hYiKIdS",
    "outputId": "5034e570-b5a5-427f-dd0e-4640be7e72c9"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "style.use('fivethirtyeight')\n",
    "\n",
    "# top subjectivity examples\n",
    "N_TOP_SUB = 4\n",
    "# top polarity examples\n",
    "N_TOP_POL = 3\n",
    "\n",
    "# subjectivity-polarity graph\n",
    "pol_sub_plt = sentiment_df.plot(x='polarity',\n",
    "                                y='subjectivity',\n",
    "                                kind='scatter',\n",
    "                                figsize=(9,7),\n",
    "                                title='Subjectivity-Polarity [Eminem]',\n",
    "                                color='darkseagreen')\n",
    "\n",
    "# highlight songs with highest subjectivity\n",
    "sentiment_df.sort_values('subjectivity', ascending=False) \\\n",
    "            .iloc[:N_TOP_SUB] \\\n",
    "            .plot(x='polarity', y='subjectivity', kind='scatter', color='green', ax=pol_sub_plt)\n",
    "\n",
    "# annotate songs with highest subjectivity\n",
    "sentiment_df.sort_values('subjectivity', ascending=False)[:N_TOP_SUB] \\\n",
    "            .apply(lambda row: pol_sub_plt.annotate(row.title, (row.polarity, row.subjectivity),\n",
    "                                                    fontsize=8, \n",
    "                                                    fontstyle='oblique', \n",
    "                                                    fontweight='roman'), axis=1)\n",
    "\n",
    "# highlight songs with highest polarity\n",
    "sentiment_df.sort_values('polarity', ascending=False) \\\n",
    "            .iloc[:N_TOP_POL] \\\n",
    "            .plot(x='polarity', y='subjectivity', kind='scatter', color='green', ax=pol_sub_plt)\n",
    "\n",
    "# annotate songs with highest polarity\n",
    "sentiment_df.sort_values('polarity', ascending=False)[:N_TOP_POL] \\\n",
    "            .apply(lambda row: pol_sub_plt.annotate(row.title, (row.polarity, row.subjectivity),\n",
    "                                                    fontsize=8, \n",
    "                                                    fontstyle='oblique', \n",
    "                                                    fontweight='roman',\n",
    "                                                    ha='center',\n",
    "                                                    va='bottom'), axis=1)\n",
    "# legend\n",
    "pol_sub_plt.legend(['song'], loc=4)\n",
    "\n",
    "# signature bar\n",
    "pol_sub_plt.text(x=-0.158, y=-0.048, s=str('_'* 91).expandtabs(), color='grey')\n",
    "pol_sub_plt.text(x=-0.158, y=-0.057, \n",
    "                 s=str(\"github.com/mariostrbac \" + \"  \" * 86 + \"Lyrics Source: Genius.com\").expandtabs(),\n",
    "                 fontsize=7.9, \n",
    "                 color='dimgrey')\n",
    "\n",
    "plt.savefig('../reports/figures/sentiment-analysis/songs_subjectivity_polarity_eminem.png', dpi=240, bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "9pAJXbB5KIdS"
   },
   "source": [
    "## 4.2 The polarity of songs through the years  <a name=\"42-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "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>release_date</th>\n",
       "      <th>polarity</th>\n",
       "      <th>fit_points</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1988-01-01</td>\n",
       "      <td>-0.003256</td>\n",
       "      <td>0.000472</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1990-01-01</td>\n",
       "      <td>0.007623</td>\n",
       "      <td>-0.001941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1993-01-01</td>\n",
       "      <td>0.006466</td>\n",
       "      <td>-0.004984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1996-02-17</td>\n",
       "      <td>-0.000833</td>\n",
       "      <td>-0.007337</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1996-11-12</td>\n",
       "      <td>0.003346</td>\n",
       "      <td>-0.007337</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  release_date  polarity  fit_points\n",
       "0   1988-01-01 -0.003256    0.000472\n",
       "1   1990-01-01  0.007623   -0.001941\n",
       "2   1993-01-01  0.006466   -0.004984\n",
       "3   1996-02-17 -0.000833   -0.007337\n",
       "4   1996-11-12  0.003346   -0.007337"
      ]
     },
     "execution_count": 169,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# filter out the rows with no \"release date\" info\n",
    "sentiment_rdate_df = sentiment_df[sentiment_df.release_date.isnull() == False]\n",
    "\n",
    "# compute coefficients for the approximation of the best-fit line\n",
    "degree = 2\n",
    "polarity_df = sentiment_rdate_df[['release_date', 'polarity']].groupby('release_date').mean().reset_index()\n",
    "mean_polarity_per_year = polarity_df.polarity.values\n",
    "years = polarity_df.release_date.apply(lambda x: x.year).values\n",
    "a, b, c = np.polyfit(years, mean_polarity_per_year, degree)\n",
    "\n",
    "# quadratic function that approximates a best-fit line (trendline) for given coefficients\n",
    "quadratic_poly = lambda x_df: a * np.square(x_df.release_date.year) + b * x_df.release_date.year + c\n",
    "\n",
    "# best-fit line points\n",
    "polarity_df['fit_points'] = polarity_df.apply(quadratic_poly, axis=1)\n",
    "\n",
    "# show restuls\n",
    "polarity_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x648 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "style.use('fivethirtyeight')\n",
    "\n",
    "# 1. STEP\n",
    "# polarity-release_date plot \n",
    "polarity_plt = sentiment_rdate_df.plot(x='release_date',\n",
    "                                       y='polarity',\n",
    "                                       kind='scatter',\n",
    "                                       figsize=(12,9),\n",
    "                                       title=\"Polarity of lyrics through the years [Eminem]\",\n",
    "                                       legend=False,\n",
    "                                       fontsize=9.5,\n",
    "                                       color='darkseagreen')\n",
    "\n",
    "# trendline (best-fit line) \n",
    "polarity_df.plot(x='release_date',\n",
    "                 y='fit_points',\n",
    "                 kind='line',\n",
    "                 legend=False,\n",
    "                 linewidth=1.2, \n",
    "                 color='salmon',\n",
    "                 ax=polarity_plt)\n",
    "\n",
    "# 2. STEP\n",
    "# highlight/annotate most negative, positive and popular songs on the graph\n",
    "N_NEGATIVE = N_POSITIVE = 7\n",
    "N_MOST_POPULAR = 13\n",
    "most_negative_songs = sentiment_rdate_df[['title', 'popularity', 'release_date', 'polarity']] \\\n",
    "                                .sort_values('polarity', ascending=False)[:N_NEGATIVE]\n",
    "most_positive_songs = sentiment_rdate_df[['title', 'popularity', 'release_date', 'polarity']] \\\n",
    "                                .sort_values('polarity', ascending=True)[:N_POSITIVE]\n",
    "most_popular_songs = sentiment_rdate_df[['title', 'popularity', 'release_date', 'polarity']] \\\n",
    "                                .sort_values('popularity')[:N_MOST_POPULAR]\n",
    "# merge the dateframes into one dateframe\n",
    "top_songs_df = pd.concat([most_negative_songs, most_positive_songs, most_popular_songs])\n",
    "\n",
    "# preprocess the data before plotting\n",
    "top_songs_df = top_songs_df[top_songs_df['title'].isin(['Lucky You', 'Not Alike']) == False]\n",
    "top_songs_df['fontsize'] = top_songs_df['popularity'].apply(lambda popularity: 8 if popularity <= N_MOST_POPULAR \n",
    "                                                                                 else 7)\n",
    "# highlight the top songs on the graph with green dots\n",
    "top_songs_df.plot(x='release_date',\n",
    "                  y='polarity',\n",
    "                  kind='scatter',\n",
    "                  legend=False,\n",
    "                  color='green',\n",
    "                  ax=polarity_plt)\n",
    "\n",
    "# annotate the highlighted points\n",
    "top_songs_df.apply(lambda row: polarity_plt.annotate(row.title, (row.release_date, row.polarity), \n",
    "                                                     fontsize=row.fontsize, \n",
    "                                                     fontstyle='oblique', \n",
    "                                                     fontweight='roman'), \n",
    "                                                     axis=1)\n",
    "# 3. STEP\n",
    "# plot horizontal lines at the bottom and in the middle of the graph\n",
    "polarity_plt.axhline(y = -0.135, color = 'black', linewidth = 1.3, alpha = .7)\n",
    "polarity_plt.axhline(y = 0, color = 'grey', linewidth = 1, alpha = .4, linestyle='--')\n",
    "\n",
    "# plot horizontal line for the average polarity value\n",
    "avg_polarity = np.round(sentiment_rdate_df['polarity'].mean(), 2)\n",
    "polarity_plt.axhline(y=avg_polarity, ls='--', color='darkslategrey', linewidth=1.3, alpha=0.1)\n",
    "polarity_plt.text(x='1986.7', y=avg_polarity + 0.0006, s='average', fontsize=8, color='darkslategrey', alpha=0.8)\n",
    "\n",
    "# set a tick for the average value\n",
    "polarity_plt.set_yticks([-0.10, -0.05, np.round(sentiment_rdate_df['polarity'].mean(),2), 0.0, 0.05])\n",
    "\n",
    "# x-label\n",
    "polarity_plt.set_xlabel('release date')\n",
    "\n",
    "# add a custom legend to the graph\n",
    "polarity_plt.legend(handles = [plt.scatter([], [], lw=0.6, color='darkseagreen'),\n",
    "                               plt.plot([], lw=1.2, color='salmon', ls=\"-\")[0]], \n",
    "                    labels=['song', 'trendline (best-fit line)'], \n",
    "                    fontsize=10, \n",
    "                    loc=2)\n",
    "# signature bar\n",
    "polarity_plt.text(x='1984', y=-0.175, s=str('_'*117).expandtabs(), color='grey')\n",
    "polarity_plt.text(x='1984', y=-0.182,\n",
    "                  s=str(\"github.com/mariostrbac \" + \" \" * 243 + \"  Lyrics Source: Genius.com\").expandtabs(),\n",
    "                  fontsize=7.9,\n",
    "                  color='dimgrey')\n",
    "\n",
    "plt.savefig('../reports/figures/sentiment-analysis/songs_polarity_through_years_eminem.png', dpi=240, bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Nze06PdRKIdU"
   },
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "6y-bs5kKKIdV"
   },
   "source": [
    "# 5. Topic modeling  <a name=\"5-bullet\"></a> <a href=\"#0-bullet\"> <sup><sup><sup>^</sup></sup></sup></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "dmNmiiRBKIdV"
   },
   "source": [
    "`// TODO`"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [
    "_v5dH3chKIcP",
    "ge4e4yM9KIcQ",
    "2jCcn__hKIcQ",
    "7qYNxUy0KIcR",
    "Jx_MD33lKIcS",
    "1Xrc3URuKIcT",
    "pOXxHGEBKIcU",
    "nypW4YndKIcW",
    "XrQbynatKIcX",
    "-VJLz4fEKIcY",
    "1lDp_qH0KIcY",
    "hGB3nJhNKIca",
    "mOPphr15KIcc",
    "Sy0rnt7kKIcd",
    "Rb-6sBD9KIce",
    "5asVcx2WKIcg",
    "Hxv5XOvhKIch",
    "1wxEzgc0KIcj",
    "-ZxYw59oKIcj",
    "iRG0HA3QKIck",
    "h9Xjx-30KIck",
    "5XhkBhquKIcm",
    "cym1f15zKIcn",
    "eQ5utIh5KIcx",
    "YgGmtkqpKIcy",
    "3pEDP6ktKIc1",
    "Kt-mhrlPKIc1",
    "_ByaHcH3KIc2",
    "AAfK6kgmKIc6",
    "_3Mh9ttPKIc8",
    "fMSDsrDuKIc9",
    "yFmOe0ZiKIc-",
    "bsQGZTT3KIc-",
    "7W0TDNzxKIdA",
    "Roo5pwLCKIdD",
    "CYeIWxAuKIdE",
    "Cph6cEIPKIdJ",
    "KgyTABqmKIdJ",
    "AmkWNODUKIdK",
    "5y1GAjLPKIdK",
    "p3_4IFGVKIdM",
    "6oVBAaxzKIdO",
    "erRCSTsXKIdR",
    "9pAJXbB5KIdS"
   ],
   "name": "3-Exploratory-Data-Analysis.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}