nbs/congress-links.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Tools for Link Analysis: urlExpander\n",
    "View this on [GitHub](https://github.com/yinleon/links-as-data/blob/master/nbs/congress-links.ipynb) | [NBviewer](http://nbviewer.jupyter.org/github/yinleon/links-as-data/blob/master/nbs/congress-links.ipynb?flush_cache=true) | [Binder](https://mybinder.org/v2/gh/yinleon/links-as-data/master?filepath=nbs%2Fcongress-links.ipynb)<br>\n",
    "Auhor: Leon Yin <br>\n",
    "Updated on: 2018-10-01\n",
    "<hr>\n",
    "\n",
    "## Intro\n",
    "This notebook will walk through using links as data with the [URLexpander](https://github.com/SMAPPNYU/urlExpander) package.\n",
    "1. What kind of link data does Twitter provide?\n",
    "2. How to extract link data from Tweets (`urlexpander.tweet_utils.get_link`)\n",
    "3. Processing data by expanding shortened URLs (`urlexpander.expand`)\n",
    "4. Avenues of analysis with link data (`urlexpander.tweet_utils.count_matrix`, `urlexpander.html_utils.get_webpage_meta`)\n",
    "5. Using links as features to predict political affiliation\n",
    "\n",
    "Software for this tutorial is found in this `requirements.txt` file, and can be downloaded as follows:\n",
    "```\n",
    "pip install -r requirements.txt\n",
    "```\n",
    "\n",
    "Download data here:\n",
    "```\n",
    "python download_data.py\n",
    "```\n",
    "\n",
    "NOTE: at the time of this writing, `download_data.py` does not work! Please go to [OSF](https://osf.io/36b5w/) in the meantime."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What Kind of Link Data Does Twitter Provide?\n",
    "If you have yet to look at the backend of a Tweet, here you go: [https://bit.ly/tweet_anatomy_link](https://bit.ly/tweet_anatomy_link). <br>\n",
    "In addition to hashtags and the like, Tweets contain metadata fields for urls. The code below will show you how to extract and work with links from Tweets."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We're working with Tweets from members of congress collected by Greg Eady."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "import glob\n",
    "import itertools\n",
    "from multiprocessing import Pool\n",
    "\n",
    "from tqdm import tqdm\n",
    "import pandas as pd\n",
    "import urlexpander\n",
    "from smappdragon import JsonCollection\n",
    "\n",
    "from config import INTERMEDIATE_DIRECTORY, \\\n",
    "                   RAW_TWEETS_DIRECTORY, \\\n",
    "                   CONGRESS_METADATA_DIRECTORY"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1950"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# config setting\n",
    "pd.options.display.float_format = '{:.0f}'.format\n",
    "\n",
    "# these are the files we'll be producing here\n",
    "file_raw_links = os.path.join(INTERMEDIATE_DIRECTORY, 'links_raw.csv')\n",
    "file_cache = os.path.join(INTERMEDIATE_DIRECTORY, 'cache.json')\n",
    "file_expanded_links = os.path.join(INTERMEDIATE_DIRECTORY, 'links_expanded_all.csv')\n",
    "\n",
    "# this is the raw data we're working with\n",
    "files = glob.glob(os.path.join(RAW_TWEETS_DIRECTORY, '*.json.bz2'))\n",
    "len(files)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's preview one file. The file is saved as a newline-delimited json file like this\n",
    "```\n",
    "{\"tweet_id\" : \"123\", \"more_data\" : {\"here\" : \"it is\"}\n",
    "{\"tweet_id\" : \"124\", \"more_data\" : {\"here\" : \"it is again\"}\n",
    "```\n",
    "and bzip2 compressed!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'data/tweets-raw/1089859058__2018-03.json.bz2'"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f = files[2]\n",
    "f"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The file structure is new-line delimited JSON. We developed software (like [smappdragon](https://github.com/SMAPPNYU/smappdragon)) to work with Tweets like this:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "collect = JsonCollection(f, compression='bz2', throw_error=False, verbose=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "smappdragon's `JsonCollection` class reads through JSON files as a [generator](https://wiki.python.org/moin/Generators)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<smappdragon.collection.json_collection.JsonCollection at 0x10d28f588>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "collect"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The con is that generators are hard to interpret, the pro is that they don't store any data in memory. We access the data on a row-by-row basis by iterating through the `collect` object. Here we only get the first row, we can see the contents by printing row:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "for row in collect.get_iterator():\n",
    "    break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#print(json.dumps(row, indent=2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## How do we get the links?\n",
    "Each Tweet can have more than one link, thus we need to unpack those values! urlexpander has a function to do just this:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\u001b[0;31mSignature:\u001b[0m \u001b[0murlexpander\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtweet_utils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_link\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtweet\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
       "\u001b[0;31mDocstring:\u001b[0m\n",
       "Returns a generator containing tweet metadata about media.\n",
       "\n",
       "The metadata dict contains the following columns:\n",
       "\n",
       "columns = {\n",
       "  'link_domain' : 'the domain of the URL', \n",
       "  'link_url_long' : 'the URL (this can be short!)', \n",
       "  'link_url_short' : 'The t.co URL', \n",
       "  'tweet_created_at' : 'When the tweet was created', \n",
       "  'tweet_id' : 'The ID of the tweet', \n",
       "  'tweet_text' : 'The Full text of the tweet', \n",
       "  'user_id' : 'The Twitter ID of the tweeter'\n",
       "}\n",
       "\n",
       ":input tweet: a nested dictionary of a Tweet either from the streaming or search API.\n",
       ":returns: a generator of dictionaries\n",
       "\u001b[0;31mFile:\u001b[0m      ~/anaconda3/lib/python3.6/site-packages/urlexpander/core/tweet_utils.py\n",
       "\u001b[0;31mType:\u001b[0m      function\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "?urlexpander.tweet_utils.get_link"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Once again we have another generator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<generator object get_link at 0x11b8a3f10>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# returns a genrator, which is uninterpretable!\n",
    "urlexpander.tweet_utils.get_link(row)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'user_id': 1089859058, 'tweet_id': 976517212063322112, 'tweet_created_at': 'Wed Mar 21 17:53:40 +0000 2018', 'tweet_text': None, 'link_url_long': 'http://bit.ly/2FC7bMz', 'link_domain': 'bit.ly', 'link_url_short': 'https://t.co/5P1JAaxwQV'}\n"
     ]
    }
   ],
   "source": [
    "# we can access the data by iterating through it.\n",
    "for link_meta in urlexpander.tweet_utils.get_link(row):\n",
    "    print(link_meta)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'user_id': 1089859058,\n",
       "  'tweet_id': 976517212063322112,\n",
       "  'tweet_created_at': 'Wed Mar 21 17:53:40 +0000 2018',\n",
       "  'tweet_text': None,\n",
       "  'link_url_long': 'http://bit.ly/2FC7bMz',\n",
       "  'link_domain': 'bit.ly',\n",
       "  'link_url_short': 'https://t.co/5P1JAaxwQV'}]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# to unwrap this we'll do this mess of code\n",
    "list(itertools.chain.from_iterable([urlexpander.tweet_utils.get_link(row)]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First let's a generalize the workflow into a function. Below is a boilerplate function you can use this as a starting place for your own workflow."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def read_file_extract_links(f):\n",
    "    '''\n",
    "    This function takes in a Tweet file that bzip2-compressed, \n",
    "    newline-deliminted json, and returns a list of dictionaries\n",
    "    for link data.\n",
    "    '''\n",
    "    # read the json file into a generator\n",
    "    collection = JsonCollection(f, compression='bz2', throw_error=False)\n",
    "    \n",
    "    # iterate through the json file, extract links, flatten the generator of links\n",
    "    # into a list, and store into a Pandas dataframe\n",
    "    df_ = pd.DataFrame(list(\n",
    "            itertools.chain.from_iterable(\n",
    "                [urlexpander.tweet_utils.get_link(t) \n",
    "                 for t in collection.get_iterator() \n",
    "                 if t]\n",
    "            )))\n",
    "    df_['file'] = f\n",
    "    return df_.to_dict(orient='records')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can iterate through files and run the function iteratively. From there, we can instantiate a Pandas `DataFrame`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 2/2 [00:01<00:00,  1.03s/it]\n"
     ]
    }
   ],
   "source": [
    "data = []\n",
    "for f in tqdm(files[:2]):\n",
    "    # read the json file into a generator\n",
    "    data.extend(read_file_extract_links(f)\n",
    "df_links = pd.DataFrame(data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Advanced (but practical) usage\n",
    "The for loop is is slow! This task is not memory intensive (because we're using generators).\n",
    "\n",
    "We can parallelize this task if we will use the `Pool` class from the Mulitprocessing package to have each core on our computer read a JSON file of Tweets and filter for links. The \"if\" statement is to prevent repeating work once we have already read the files once. We cache this intermediate in the `file_raw_links` file path declared at the beginning of the notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>file</th>\n",
       "      <th>link_domain</th>\n",
       "      <th>link_url_long</th>\n",
       "      <th>link_url_short</th>\n",
       "      <th>tweet_created_at</th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>tweet_text</th>\n",
       "      <th>user_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>/scratch/olympus/projects/mediascore/Data/json...</td>\n",
       "      <td>frc.org</td>\n",
       "      <td>https://www.frc.org/wwlivewithtonyperkins/rep-...</td>\n",
       "      <td>https://t.co/l9dXT0L7oT</td>\n",
       "      <td>Fri Mar 23 14:38:34 +0000 2018</td>\n",
       "      <td>977192888781168640</td>\n",
       "      <td>nan</td>\n",
       "      <td>2966758114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>/scratch/olympus/projects/mediascore/Data/json...</td>\n",
       "      <td>thehill.com</td>\n",
       "      <td>http://thehill.com/379188-watch-fund-governmen...</td>\n",
       "      <td>https://t.co/YbdvepWNQ3</td>\n",
       "      <td>Thu Mar 22 15:21:32 +0000 2018</td>\n",
       "      <td>976841314024206339</td>\n",
       "      <td>nan</td>\n",
       "      <td>2966758114</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                file  link_domain  \\\n",
       "0  /scratch/olympus/projects/mediascore/Data/json...      frc.org   \n",
       "1  /scratch/olympus/projects/mediascore/Data/json...  thehill.com   \n",
       "\n",
       "                                       link_url_long           link_url_short  \\\n",
       "0  https://www.frc.org/wwlivewithtonyperkins/rep-...  https://t.co/l9dXT0L7oT   \n",
       "1  http://thehill.com/379188-watch-fund-governmen...  https://t.co/YbdvepWNQ3   \n",
       "\n",
       "                 tweet_created_at            tweet_id  tweet_text     user_id  \n",
       "0  Fri Mar 23 14:38:34 +0000 2018  977192888781168640         nan  2966758114  \n",
       "1  Thu Mar 22 15:21:32 +0000 2018  976841314024206339         nan  2966758114  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "N_CPU = 4 # this is the number of cores we canuse to parallelize the fast\n",
    "if not os.path.exists(file_raw_links):\n",
    "    data = []\n",
    "    with Pool(N_CPU) as pool:\n",
    "        iterable = pool.imap_unordered(read_file_extract_links, files)\n",
    "        for link_data in tqdm(iterable, total=len(files)):\n",
    "            data.extend(link_data)\n",
    "    df_links = pd.DataFrame(data)\n",
    "    df_links.to_csv(file_raw_links, index=False)\n",
    "\n",
    "else:\n",
    "    df_links = pd.read_csv(file_raw_links)\n",
    "\n",
    "df_links.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## How useful is this data?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The bulk of URLs we encounter in the wild are sent through a link shortener. Link shorteners record transactional information whenever that link is clicked. Unfortunately it makes it hard for us to see what was being shared."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['http://goo.gl/kDUwP',\n",
       " 'http://bit.ly/12clU3p',\n",
       " 'http://nyti.ms/Z4rdlU',\n",
       " 'http://goo.gl/LxkrY',\n",
       " 'http://www.huffingtonpost.com/rep-diana-degette/reducing-gun-violence-mea_b_3018506.html']"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "links = df_links['link_url_long'].tolist()\n",
    "links[-5:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is why `urlexpander` was made. We can run the `expand` function on single URLs, as well as a list of URLs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'https://degette.house.gov/index.php?option=com_content&view=article&id=1260:congressional-lgbt-equality-caucus-praises-re-introduction-of-employment-non-discrimination-act&catid=76:press-releases-&Itemid=227'"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "urlexpander.expand(links[-5])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By default, urlexpander will expand every URL it is shown. However you can pass a boolean function (one the returns True or False based on an inputted string) to the `filter_function` parameter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['https://degette.house.gov/index.php?option=com_content&view=article&id=1260:congressional-lgbt-equality-caucus-praises-re-introduction-of-employment-non-discrimination-act&catid=76:press-releases-&Itemid=227',\n",
       " 'http://www.rollcall.com/news/house_hydro_bill_tests_water_for_broad_energy_deals-224277-1.html',\n",
       " 'http://nyti.ms/Z4rdlU',\n",
       " 'http://www.civiccenterconservancy.org/history-2012-nhl-designation_25.html',\n",
       " 'http://www.huffingtonpost.com/rep-diana-degette/reducing-gun-violence-mea_b_3018506.html']"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "urlexpander.expand(links[-5:], filter_function=urlexpander.is_short)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What's happening behind the scenes? \n",
    "['abc.com/123', 'bbc.co.uk/123', 'abc.com/123', 'bit.ly/cbc23']<br>\n",
    "--> Remove duplicates<br>\n",
    " ['abc.com/123', 'bbc.co.uk/123', 'bit.ly/cbc23']<br>\n",
    "--> Filter for shortened URLs<br>\n",
    "['bit.ly/cbs23']<br>\n",
    "--> Check the cache file, did we already expand this? Unshorten new urls<br>\n",
    "[{'original_url': 'bit.ly/cbs23', 'resolved' : 'cspan.com/123'}]<br>\n",
    "--> join back in<br>\n",
    "['abc.com/123', 'bbc.co.uk/123', 'abc.com/123', 'cspan.com/123']<br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "urlexpander parallelizes, filters and caches the input, which is essential for social media data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\u001b[0;31mSignature:\u001b[0m \u001b[0murlexpander\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexpand\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlinks_to_unshorten\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mchunksize\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1280\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_workers\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcache_file\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrandom_seed\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m303\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfilter_function\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
       "\u001b[0;31mDocstring:\u001b[0m\n",
       "Calls expand with multiple (``n_workers``) threads to unshorten a list of urls. Unshortens all urls by default, unless one sets a ``filter_function``.\n",
       "\n",
       ":param links_to_unshorten: (list, str) either an idividual or list (str) of urls to unshorten\n",
       ":param chunksize: (int) chunks links_to_unshorten, which makes computation quicker with larger inputs\n",
       ":param n_workers: (int) how many threads\n",
       ":param cache_file: (str) a path to a json file to read and write results in\n",
       ":param random_seed: (int) initializes the random state for shuffling the input\n",
       ":param verbose: (int) whether to print updates and errors. 0 is silent. 1 is progress bar. 2 is progress bar and errors.\n",
       ":param filter_function: (func) a boolean used to filter url shorteners out\n",
       "    \n",
       "\n",
       ":returns: (list) a list of resolved urls\n",
       "\u001b[0;31mFile:\u001b[0m      ~/anaconda3/lib/python3.6/site-packages/urlexpander/core/api.py\n",
       "\u001b[0;31mType:\u001b[0m      function\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "?urlexpander.expand"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The above is a toy example with 5 links, let's see how this works for 1.7 Mil links. For reference, this took me an hour on an 8-core computer with reliable internet connection."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "resolved_urls = urlexpander.expand(links, \n",
    "                                   filter_function=urlexpander.is_short,\n",
    "                                   n_workers=64,\n",
    "                                   chunksize=1280,\n",
    "                                   cache_file=file_cache,\n",
    "                                   verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1700150"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(resolved_urls)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_links['link_resolved'] = resolved_urls\n",
    "df_links['link_resolved_domain'] = df_links['link_resolved'].apply(urlexpander.get_domain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_links.to_csv(file_expanded_links, index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analytics\n",
    "With the links resolved, how can we use links as data?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>file</th>\n",
       "      <th>link_domain</th>\n",
       "      <th>link_url_long</th>\n",
       "      <th>link_url_short</th>\n",
       "      <th>tweet_created_at</th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>tweet_text</th>\n",
       "      <th>user_id</th>\n",
       "      <th>link_resolved</th>\n",
       "      <th>link_resolved_domain</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>/scratch/olympus/projects/mediascore/Data/json...</td>\n",
       "      <td>frc.org</td>\n",
       "      <td>https://www.frc.org/wwlivewithtonyperkins/rep-...</td>\n",
       "      <td>https://t.co/l9dXT0L7oT</td>\n",
       "      <td>Fri Mar 23 14:38:34 +0000 2018</td>\n",
       "      <td>977192888781168640</td>\n",
       "      <td>nan</td>\n",
       "      <td>2,966,758,114</td>\n",
       "      <td>https://www.frc.org/wwlivewithtonyperkins/rep-...</td>\n",
       "      <td>frc.org</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>/scratch/olympus/projects/mediascore/Data/json...</td>\n",
       "      <td>thehill.com</td>\n",
       "      <td>http://thehill.com/379188-watch-fund-governmen...</td>\n",
       "      <td>https://t.co/YbdvepWNQ3</td>\n",
       "      <td>Thu Mar 22 15:21:32 +0000 2018</td>\n",
       "      <td>976841314024206339</td>\n",
       "      <td>nan</td>\n",
       "      <td>2,966,758,114</td>\n",
       "      <td>https://thehill.com/379188-watch-fund-governme...</td>\n",
       "      <td>thehill.com</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                file  link_domain  \\\n",
       "0  /scratch/olympus/projects/mediascore/Data/json...      frc.org   \n",
       "1  /scratch/olympus/projects/mediascore/Data/json...  thehill.com   \n",
       "\n",
       "                                       link_url_long           link_url_short  \\\n",
       "0  https://www.frc.org/wwlivewithtonyperkins/rep-...  https://t.co/l9dXT0L7oT   \n",
       "1  http://thehill.com/379188-watch-fund-governmen...  https://t.co/YbdvepWNQ3   \n",
       "\n",
       "                 tweet_created_at            tweet_id  tweet_text  \\\n",
       "0  Fri Mar 23 14:38:34 +0000 2018  977192888781168640         nan   \n",
       "1  Thu Mar 22 15:21:32 +0000 2018  976841314024206339         nan   \n",
       "\n",
       "        user_id                                      link_resolved  \\\n",
       "0 2,966,758,114  https://www.frc.org/wwlivewithtonyperkins/rep-...   \n",
       "1 2,966,758,114  https://thehill.com/379188-watch-fund-governme...   \n",
       "\n",
       "  link_resolved_domain  \n",
       "0              frc.org  \n",
       "1          thehill.com  "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_links = pd.read_csv(file_expanded_links)\n",
    "df_links.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can get an overview of the most frequently shared domains:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "twitter.com           255532\n",
       "house.gov             199218\n",
       "youtube.com            93986\n",
       "facebook.com           90061\n",
       "senate.gov             78645\n",
       "washingtonpost.com     29886\n",
       "instagram.com          28460\n",
       "nytimes.com            25014\n",
       "thehill.com            22925\n",
       "politico.com           13488\n",
       "foxnews.com            12045\n",
       "cnn.com                11611\n",
       "wsj.com                11289\n",
       "twimg.com               9633\n",
       "ow.ly                   9463\n",
       "Name: link_resolved_domain, dtype: int64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_links['link_resolved_domain'].value_counts().head(15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Text-based URL Metadata\n",
    "We can also look at the contents of each URL. Twitter provides URL metadata ([if you pay](https://developer.twitter.com/en/docs/tweets/enrichments/overview/expanded-and-enhanced-urls)), we provided a workaround!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "OrderedDict([('url',\n",
       "              'http://www.rollcall.com/news/house_hydro_bill_tests_water_for_broad_energy_deals-224277-1.html'),\n",
       "             ('title', 'House Hydro Bill Tests Water for Broad Energy Deals'),\n",
       "             ('description',\n",
       "              ' In February, the House did something rare: It passed an energy bill unanimously. Unlike the previous Congress’ standard fare of anti-EPA, pro-drilling measures, the first energy bill of the 113th Congress promoted small-scale hydropower projects and the electrification of existing dams.'),\n",
       "             ('paragraphs',\n",
       "              ['',\n",
       "               '',\n",
       "               'In February, the House did something rare: It passed an energy bill unanimously. Unlike the previous Congress’ standard fare of anti-EPA, pro-drilling measures, the first energy bill of the 113th Congress promoted small-scale hydropower projects and the electrification of existing dams.',\n",
       "               'In other words, the Republican-controlled House passed a clean-energy bill.',\n",
       "               'Of course, few Republicans could object on ideological grounds to legislation that aimed to expedite or remove regulatory requirements for expanding hydropower facilities. And it certainly helped that the chamber passed similar legislation in 2012 by a large margin.               ',\n",
       "               'Nevertheless, hydropower proponents say that developing the resource represents “low-hanging fruit” that members tired of the partisanship that has permeated energy policy can all agree is worth advancing to President Barack Obama’s desk. Hydropower represented nearly two-thirds — the largest share by far — of domestic renewable-energy production and 8 percent of total U.S. electricity generation in 2011, according to the Energy Information Administration. More than half of that total powered the Pacific Northwest region, and hydro is one of the few renewable resources that can provide baseload electricity — power that is available at all times — to the grid.',\n",
       "               'But just 3 percent of the 80,000 dams in the United States generate power, representing great potential for growing the resource, according to legislation championed by Reps. Cathy McMorris Rodgers, R-Wash., and  Diana DeGette, D-Colo.',\n",
       "               '“If you can work on regulatory reform for those projects, then you can have small hydro throughout this country,” DeGette said. ',\n",
       "               'The House lawmakers’ legislation would let small hydroelectric facilities generating up to 10 megawatts of power bypass Federal Energy Regulatory Commission licensing requirements that currently apply to projects producing more than 5 megawatts. The bill also would require FERC to study the feasibility of carrying out a two-year hydropower licensing pilot program at unpowered dams and would allow the commission to extend preliminary permits for two additional years.  ',\n",
       "               'Jeff Leahey, government affairs director at the National Hydropower Association, said House leaders probably moved the bill so they could promote energy legislation that “checked the boxes” on encouraging the development of a resource that is renewable and reliable. It also helped that the bill — along with another measure passed April 10 that would designate an Interior Department agency as the lead regulator of small federal conduits — moved through the House last Congress and didn’t need much additional work, he said.',\n",
       "               'A significant factor in the refocused spotlight on the “original renewable” is the new leadership on the Senate Energy and Natural Resources Committee and its representation of key hydropower-producing states. Chairman Ron Wyden, D-Ore., promised industry representatives at the hydropower association’s annual conference this week that he plans to “quickly” mark up hydropower legislation after a panel hearing Tuesday. ',\n",
       "               'Wyden attributes the rise in hydro’s profile to better environmental stewardship on the part of facility operators and a more cooperative relationship between hydropower lobbyists and environmental groups focused on protecting river ecosystems. The effect of dams on fisheries and riverine habitat, as well as operational costs, has compelled organizations to promote the removal of dams in some instances.',\n",
       "               '“Hydro’s environmental performance has improved dramatically,” Wyden said. ',\n",
       "               'Association President David Moller of Pacific Gas and Electric Co. acknowledged the role that historically low natural-gas prices have played in limiting hydropower expansion in recent years. But he said opportunities for hydropower still flourish because of state renewable portfolio mandates, coal-fired power plants being pushed into retirement and technological advances in powering existing dams and water channels.',\n",
       "               '“The price of natural gas has dropped, but it will never match hydropower’s fuel price of zero, or its attributes of being renewable and non-carbon-based,” he said. ',\n",
       "               'Hydro proponents in the private sector and in Congress said this week that they will continue to promote hydropower development, possibly in future legislation. That could include examining additional regulatory issues that contribute to long lead times for completing electrified projects or adjusting current benefits that exist for clean power in the tax code. Prospects for he latter — which would involve extending the production tax credit for a multiyear period or making it permanent as Obama proposes — are dim outside a comprehensive tax code overhaul.',\n",
       "               '“I think at the end of the day, it’s all about making sure that hydropower and the benefits that come from hydropower projects are competitive in the marketplace with other energy projects,” Leahey said. ',\n",
       "               'Whether the bipartisan camaraderie that has surrounded promoting hydropower can translate to moving broader energy legislation is anyone’s guess. But members close to the debate express optimism that the current spate of legislative action could beget compromise in the future.',\n",
       "               '“I’m not sure we’re any closer to that comprehensive policy, but I’d think that common ground on these issues like hydro can only be helpful,” DeGette said.',\n",
       "               '',\n",
       "               '×',\n",
       "               '$${CardTitle}',\n",
       "               '$${CardTitle}'])])"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url = 'http://www.rollcall.com/news/house_hydro_bill_tests_water_for_broad_energy_deals-224277-1.html'\n",
    "meta = urlexpander.html_utils.get_webpage_meta(url)\n",
    "meta"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Categorizing what is shared\n",
    "If we want to know more about what kinds of information are being shared by members of congress, we can enrich the dataset by joining metadata about domains. In this example we will use the [Local News Dataset](https://github.com/yinleon/LocalNewsDataset) to inspect the local media outlets that members of congress share:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>state</th>\n",
       "      <th>website</th>\n",
       "      <th>domain</th>\n",
       "      <th>twitter</th>\n",
       "      <th>youtube</th>\n",
       "      <th>facebook</th>\n",
       "      <th>owner</th>\n",
       "      <th>medium</th>\n",
       "      <th>source</th>\n",
       "      <th>collection_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KWHE</td>\n",
       "      <td>HI</td>\n",
       "      <td>http://www.kwhe.com/</td>\n",
       "      <td>kwhe.com</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>LeSea</td>\n",
       "      <td>TV station</td>\n",
       "      <td>stationindex</td>\n",
       "      <td>2018-08-02 14:55:24.612585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>WGVK</td>\n",
       "      <td>MI</td>\n",
       "      <td>http://www.wgvu.org/</td>\n",
       "      <td>wgvu.org</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Grand Valley State University</td>\n",
       "      <td>TV station</td>\n",
       "      <td>stationindex</td>\n",
       "      <td>2018-08-02 14:55:24.612585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>KTUU</td>\n",
       "      <td>AK</td>\n",
       "      <td>http://www.ktuu.com/</td>\n",
       "      <td>ktuu.com</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Schurz Communications</td>\n",
       "      <td>TV station</td>\n",
       "      <td>stationindex</td>\n",
       "      <td>2018-08-02 14:55:24.612585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>KTBY</td>\n",
       "      <td>AK</td>\n",
       "      <td>http://www.ktbytv.com/</td>\n",
       "      <td>ktbytv.com</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Coastal Television Broadcasting</td>\n",
       "      <td>TV station</td>\n",
       "      <td>stationindex</td>\n",
       "      <td>2018-08-02 14:55:24.612585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>KYES</td>\n",
       "      <td>AK</td>\n",
       "      <td>http://www.kyes.com/</td>\n",
       "      <td>kyes.com</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Fireweed Communications</td>\n",
       "      <td>TV station</td>\n",
       "      <td>stationindex</td>\n",
       "      <td>2018-08-02 14:55:24.612585</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   name state                 website      domain twitter youtube facebook  \\\n",
       "0  KWHE    HI    http://www.kwhe.com/    kwhe.com     NaN     NaN      NaN   \n",
       "1  WGVK    MI    http://www.wgvu.org/    wgvu.org     NaN     NaN      NaN   \n",
       "3  KTUU    AK    http://www.ktuu.com/    ktuu.com     NaN     NaN      NaN   \n",
       "4  KTBY    AK  http://www.ktbytv.com/  ktbytv.com     NaN     NaN      NaN   \n",
       "5  KYES    AK    http://www.kyes.com/    kyes.com     NaN     NaN      NaN   \n",
       "\n",
       "                             owner      medium        source  \\\n",
       "0                            LeSea  TV station  stationindex   \n",
       "1    Grand Valley State University  TV station  stationindex   \n",
       "3            Schurz Communications  TV station  stationindex   \n",
       "4  Coastal Television Broadcasting  TV station  stationindex   \n",
       "5          Fireweed Communications  TV station  stationindex   \n",
       "\n",
       "              collection_date  \n",
       "0  2018-08-02 14:55:24.612585  \n",
       "1  2018-08-02 14:55:24.612585  \n",
       "3  2018-08-02 14:55:24.612585  \n",
       "4  2018-08-02 14:55:24.612585  \n",
       "5  2018-08-02 14:55:24.612585  "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "local_news_url = 'https://raw.githubusercontent.com/yinleon/LocalNewsDataset/master/data/local_news_dataset_2018.csv'\n",
    "df_local = pd.read_csv(local_news_url)\n",
    "df_local = df_local[~(df_local.domain.isnull()) & \n",
    "                    (df_local.domain != 'facebook.com')]\n",
    "df_local.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "# this is a SQL-like join in Pandas that merges the two datasets based on domain name!\n",
    "df_links_state = df_links.merge(df_local, \n",
    "                                left_on='link_resolved_domain', \n",
    "                                right_on='domain')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This dataset unlocks insights regarding the locality of news articles shared, as well as media ownership."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TX    43962\n",
       "CA    35029\n",
       "NJ    33442\n",
       "NY    33145\n",
       "MI    28912\n",
       "FL    21519\n",
       "NC    19841\n",
       "PA    19060\n",
       "OH    18857\n",
       "MD    16533\n",
       "GA    13419\n",
       "MO    12660\n",
       "TN    12324\n",
       "AZ    11052\n",
       "MA    10767\n",
       "WA    10325\n",
       "NV    10261\n",
       "LA     9519\n",
       "MN     9510\n",
       "OR     9404\n",
       "Name: state, dtype: int64"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_links_state.state.value_counts().head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Nexstar                                         7543\n",
       "Advance Local                                   4464\n",
       "Tegna Media                                     4265\n",
       "Sinclair                                        3457\n",
       "Hearst                                          2571\n",
       "Tribune                                         2086\n",
       "Gray Television                                 1817\n",
       "Fox Television Stations                         1580\n",
       "Hearst Television                               1528\n",
       "Acvance Local                                   1522\n",
       "Raycom                                          1396\n",
       "NBC Universal                                   1344\n",
       "Georgia Public Broadcasting                     1248\n",
       "New Jersey Public Broadcasting Authority        1156\n",
       "The Philadelphia Inquirer                       1103\n",
       "ABC                                              978\n",
       "Evening Post Publishing                          820\n",
       "Meredith                                         750\n",
       "Oregon Public Broadcasting                       740\n",
       "Georgia Public Telecommunications Commission     624\n",
       "Graham Media Group                               588\n",
       "E. W. Scripps Company                            540\n",
       "Cox Enterprises                                  526\n",
       "WGBH Educational Foundation                      425\n",
       "Piedmont Television                              333\n",
       "Name: owner, dtype: int64"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_links_state.owner.value_counts().head(25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "66891808              100\n",
       "818948638890217472     68\n",
       "1065995022             66\n",
       "1058345042             64\n",
       "90651198               63\n",
       "368948092              56\n",
       "27676828               48\n",
       "2929491549             48\n",
       "2987671552             44\n",
       "58579942               44\n",
       "Name: user_id, dtype: int64"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_links_state[df_links_state.owner == 'Sinclair']['user_id'].value_counts().head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is one example of a dataset enrichment, you can create your own categorizations and join them in. Alexa.com is a good starting place."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### User-Aggregated Acvitity\n",
    "The frequency of domains shared per-user make rich features. We can aggregate the data using the following utility function:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>1011fmtheanswer.com</th>\n",
       "      <th>10best.com</th>\n",
       "      <th>10tv.com</th>\n",
       "      <th>11alive.com</th>\n",
       "      <th>123formbuilder.com</th>\n",
       "      <th>12news.com</th>\n",
       "      <th>12newsnow.com</th>\n",
       "      <th>13abc.com</th>\n",
       "      <th>13wham.com</th>\n",
       "      <th>13wmaz.com</th>\n",
       "      <th>...</th>\n",
       "      <th>yorkdispatch.com</th>\n",
       "      <th>youarecurrent.com</th>\n",
       "      <th>youcaring.com</th>\n",
       "      <th>youngcons.com</th>\n",
       "      <th>yourconroenews.com</th>\n",
       "      <th>yourdailyjournal.com</th>\n",
       "      <th>youtube.com</th>\n",
       "      <th>zeldinforcongress.com</th>\n",
       "      <th>zeldinforsenate.com</th>\n",
       "      <th>zpolitics.com</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>user_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>813286</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>198</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>939091</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>106</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5496932</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>427</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5511752</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>73</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5558312</th>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>53</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 2376 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         1011fmtheanswer.com  10best.com  10tv.com  11alive.com  \\\n",
       "user_id                                                           \n",
       "813286                     0           0         0            0   \n",
       "939091                     0           0         0            0   \n",
       "5496932                    0           0         0            0   \n",
       "5511752                    0           0         0            0   \n",
       "5558312                   16           0         0            0   \n",
       "\n",
       "         123formbuilder.com  12news.com  12newsnow.com  13abc.com  13wham.com  \\\n",
       "user_id                                                                         \n",
       "813286                    0           0              0          0           0   \n",
       "939091                    0           0              0          0           0   \n",
       "5496932                   0           0              0          0           0   \n",
       "5511752                   0           0              0          0           0   \n",
       "5558312                   0           0              0          0           0   \n",
       "\n",
       "         13wmaz.com      ...        yorkdispatch.com  youarecurrent.com  \\\n",
       "user_id                  ...                                              \n",
       "813286            0      ...                       0                  0   \n",
       "939091            0      ...                       0                  0   \n",
       "5496932           0      ...                       0                  0   \n",
       "5511752           0      ...                       0                  0   \n",
       "5558312           0      ...                       0                  0   \n",
       "\n",
       "         youcaring.com  youngcons.com  yourconroenews.com  \\\n",
       "user_id                                                     \n",
       "813286               0              0                   0   \n",
       "939091               0              0                   0   \n",
       "5496932              0              0                   0   \n",
       "5511752              0              0                   0   \n",
       "5558312              0              0                   0   \n",
       "\n",
       "         yourdailyjournal.com  youtube.com  zeldinforcongress.com  \\\n",
       "user_id                                                             \n",
       "813286                      0          198                      0   \n",
       "939091                      0          106                      0   \n",
       "5496932                     0          427                      0   \n",
       "5511752                     0           73                      0   \n",
       "5558312                     0           53                      0   \n",
       "\n",
       "         zeldinforsenate.com  zpolitics.com  \n",
       "user_id                                      \n",
       "813286                     0              0  \n",
       "939091                     0              0  \n",
       "5496932                    0              0  \n",
       "5511752                    0              0  \n",
       "5558312                    0              0  \n",
       "\n",
       "[5 rows x 2376 columns]"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "count_matrix = urlexpander.tweet_utils.count_matrix(\n",
    "      df_links,\n",
    "      user_col='user_id',\n",
    "      domain_col='link_resolved_domain',\n",
    "      min_freq=20,\n",
    ")\n",
    "\n",
    "count_matrix.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Are links good features to predict political affiliation?\n",
    "Let's see how the `count_matrix` features fair in machine learning. To do so, we need to enrich our data with an output variable. For our purposes, we will try to predict political affiliation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/leonyin/anaconda3/lib/python3.6/site-packages/numba/errors.py:102: UserWarning: Insufficiently recent colorama version found. Numba requires colorama >= 0.3.9\n",
      "  warnings.warn(msg)\n"
     ]
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import umap\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These files have the affiliation of each account."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "meta = []\n",
    "for f in glob.glob(os.path.join(CONGRESS_METADATA_DIRECTORY, '*')):\n",
    "    _df = pd.read_csv(f)\n",
    "    _df = _df[~_df.twitter_id.isnull()]\n",
    "    meta.extend(_df.to_dict(orient='records'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_meta = pd.DataFrame(meta)\n",
    "df_meta.twitter_id = df_meta.twitter_id.astype(float, errors='ignore').astype(str)\n",
    "df_links.user_id = df_links.user_id.astype(float, errors='ignore').astype(str)\n",
    "look_up = df_meta[df_meta['twitter_id'].isin(df_links.user_id)].drop_duplicates(subset=['twitter_id'])[['twitter_id', 'affiliation']]\n",
    "df_links_ = df_links[df_links.user_id.isin(look_up['twitter_id'])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(971, 971)"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df_links_.user_id.unique()), len(look_up)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Unsupervised Learning\n",
    "This will be an exploratory step, where we will try to visualze the dataset of counts of links shared by user. We will reduce the high-dimensional data (where each domain is one dimension) to two-dimensions for visualization using the [UMAP](https://umap-learn.readthedocs.io/en/latest/) algorithm (much like the populat [TSNE](http://scikit-learn.org/stable/modules/generated/sklearn.manifold.TSNE.html) algorithm)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "def color(party):\n",
    "    if party == 'Democrat':\n",
    "        return 'blue' \n",
    "    elif party == 'Republican':\n",
    "        return 'red'\n",
    "    else:\n",
    "        return 'black'\n",
    "\n",
    "def viz_umap_embed(count_matrix, title=\"Members of Congress Embedded by UMAP\", threshold=5, **kwargs):\n",
    "    '''\n",
    "    Visualizes the count matrix in 2 dimensions using UMAP.\n",
    "    '''\n",
    "    if threshold:\n",
    "        count_matrix = count_matrix[count_matrix.sum(axis=1) >= threshold]\n",
    "    parties = look_up.set_index('twitter_id').loc[count_matrix.index].affiliation\n",
    "\n",
    "    embedding = umap.UMAP(n_components=2, **kwargs).fit_transform(count_matrix.values)\n",
    "\n",
    "    plt.figure(figsize=(14,8))\n",
    "    ax = plt.scatter(x = embedding[:,0], \n",
    "                     y = embedding[:,1],\n",
    "                     s = 100,\n",
    "                     c = parties.apply(color),\n",
    "                     alpha = .4)\n",
    "\n",
    "    plt.title(title)\n",
    "    plt.axis('off')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "# domains to exclude in our count matrix\n",
    "exclude = ['youtube.com', 'twitter.com', 'fb.com', \n",
    "           'facebook.com', 'instagram.com', 'ow.ly', \n",
    "           'house.gov', 'senate.gov', 'usa.gov']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "count_matrix = urlexpander.tweet_utils.count_matrix(\n",
    "      df_links_,\n",
    "      user_col='user_id',\n",
    "      domain_col='link_resolved_domain',\n",
    "      min_freq=5,\n",
    "      exclude_domain_list=exclude,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>frc.org</th>\n",
       "      <th>thehill.com</th>\n",
       "      <th>iheart.com</th>\n",
       "      <th>c-span.org</th>\n",
       "      <th>news9.com</th>\n",
       "      <th>speaker.gov</th>\n",
       "      <th>foxnews.com</th>\n",
       "      <th>frcaction.org</th>\n",
       "      <th>koco.com</th>\n",
       "      <th>okcfox.com</th>\n",
       "      <th>...</th>\n",
       "      <th>mullinforcongress.com</th>\n",
       "      <th>therepublicanstandard.com</th>\n",
       "      <th>barbaracomstockforcongress.com</th>\n",
       "      <th>thefriendshipchallenge.com</th>\n",
       "      <th>tomreedforcongress.com</th>\n",
       "      <th>sincomillas.com</th>\n",
       "      <th>detodopr.com</th>\n",
       "      <th>thedowneypatriot.com</th>\n",
       "      <th>garretgravesforcongress.com</th>\n",
       "      <th>about.com</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>user_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1004855106.0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1009269193.0</th>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 5301 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              frc.org  thehill.com  iheart.com  c-span.org  news9.com  \\\n",
       "user_id                                                                 \n",
       "1004855106.0        0            0           0           0          0   \n",
       "1009269193.0        0           12           0          16          0   \n",
       "\n",
       "              speaker.gov  foxnews.com  frcaction.org  koco.com  okcfox.com  \\\n",
       "user_id                                                                       \n",
       "1004855106.0            0            0              0         0           0   \n",
       "1009269193.0            2            1              0         0           0   \n",
       "\n",
       "                ...      mullinforcongress.com  therepublicanstandard.com  \\\n",
       "user_id         ...                                                         \n",
       "1004855106.0    ...                          0                          0   \n",
       "1009269193.0    ...                          0                          0   \n",
       "\n",
       "              barbaracomstockforcongress.com  thefriendshipchallenge.com  \\\n",
       "user_id                                                                    \n",
       "1004855106.0                               0                           0   \n",
       "1009269193.0                               0                           0   \n",
       "\n",
       "              tomreedforcongress.com  sincomillas.com  detodopr.com  \\\n",
       "user_id                                                               \n",
       "1004855106.0                       0                0             0   \n",
       "1009269193.0                       0                0             0   \n",
       "\n",
       "              thedowneypatriot.com  garretgravesforcongress.com  about.com  \n",
       "user_id                                                                     \n",
       "1004855106.0                     0                            0          0  \n",
       "1009269193.0                     0                            0          0  \n",
       "\n",
       "[2 rows x 5301 columns]"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "count_matrix.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/leonyin/anaconda3/lib/python3.6/site-packages/sklearn/utils/validation.py:475: DataConversionWarning: Data with input dtype float32 was converted to bool by check_pairwise_arrays.\n",
      "  warnings.warn(msg, DataConversionWarning)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAzYAAAHiCAYAAAA6f1lRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAIABJREFUeJzsvXeYXVd5t30/0zQa9Wp1ybIsWcKy5V4xBuFgjB0bCMVgsAmBJKR8+ZJ8IYU3IXnhDeQNCYRA6BgwYBtjmsGAjXvvvSDL6r2PRtOkmef747e2zpkz55wZTZ/Rc1/XuWbOrmvvs/da6+nm7gRBEARBEARBEAxnKga7AUEQBEEQBEEQBL0lBJsgCIIgCIIgCIY9IdgEQRAEQRAEQTDsCcEmCIIgCIIgCIJhTwg2QRAEQRAEQRAMe0KwCYIgCIIgCIJg2BOCTRAcZZjZa83s5W5ue6GZbeyj8641szf2xbHS8W41s6v76nhHcN7zzGyVmTWY2RUDff5imNnHzey6wW5HIUPxXo0UzOwuM/uDPjpW2eenN+9uuX3N7Foz+0RPjhsEQVCMEGyCYIRSakLh7ve6+5J+OF+NmX3GzDamiexaM/tsX58nw93f7O7f6q/jl+FfgP9297Hu/uPClem6W81sasHyJ83MzWzBALVzKFD2XgGY2XvM7LH0zGxJAuv5A9zOASH9/gfStWafvxnsdg13zGxBurdVBcsPC05mdk3a5j8Ltrk8Lb+2YPnY9PvcWuR8a82sKa3fls4zth8uLQiCIyQEmyAI+oq/A04HzgTGARcCT/T1SUwMZt81H3i+i23WAFdmX8xsOVDXn43qKwonh72k7L0ys78EPgv8H+AYYB7wBeB3+7ANXdLH19wVJydBL/v82wCe+2hnNfDOgt/7auC3RbZ9O9ACXGRmM4qsv8zdxwKnon7vY33d2CAIjpwQbILgKKPQvSxpH//azJ4xs31mdoOZ1ZbY98/N7AUzm1Nk9RnAj9x9s4u17v7tgm1WFDuPmU0ys1vMbIeZ7Un/Hz5Hcrv5pJndDzQCC/NdcZI29j4z+/e0/xoze3Pe/sea2T1mtt/MbjezL3ThevMhM3vFzHab2U/NbFZavhpYCPwsaWtHlTjEd4D3532/GuhwL8xsVGrv+qT1/ZKZjU7rLkyWr78xs+3JknGFmV1iZr9N7fr7gnPWpnu638yeMLOT8841y8x+mO7vGjP787x1Hzezm8zsOjOrB64xszOTFaU+te0/+vpemdkEZNH5E3e/2d0PuPtBd7/F3f8m7x591sw2p89ns+Pk3aO/yrtHH8g7/hQz+1m6hkfN7BNmdl/eejezPzGzVcCqtOwEM7stXcvLZvbOvO0vSc/+fjPbZGZ/nZZPTc/r3rTfvdYDwTv9Dj9Iv8N+M3vWzBab2d+l69tgZr9TsNtxZvZIusafmNnkvOOdbWYPpHY9bWYX5q071szuTue5DSi0Lr7PzNaZ2S4z+4eCdRVm9rdmtjqtv7HgvCX3LcHUdM/3pzbNT8f5gpl9puDcPzWz/7cbxyzFVuBZ4E3peJOBc4GfFtn2auBLwDPAVaUO6O6bgFuBE3vRriAI+ogQbIIgAHgncDFwLHAScE3hBmb2j2n569y9WNzNQ8BfmtlHzGy5mdkRnKcC+CbS8M8DmoD/Ltj3fcCHkTVoXZFjnwW8jCZp/wZ8Pa8N3wMeAaYAH0/HKoqZvQH419TWmelc1wO4+3HAepK21t1bShzmIWC8mS01s0rg3UChIPUpYDGwAlgEzAb+MW/9DKA2b/lX0QTrNOC1wP8ys2Pztr8c+AEwOV3vj82sOk2yfwY8nY61EvgLM3tTwb43AROB7wKfAz7n7uOB44Ab++FenZOu70fFjp34B+DsdI9ORtbAfM34DGBCuq4PAl8ws0lp3ReAA2mbq9OnkCvQc7PMzMYAt6F7Nx39Zl80s2Vp268Df+ju49Ak9o60/K+AjcA0ZHX6e8DLXFM5LkNC8STgSeBX6N2YjYTALxds/37g99G9PwT8F4CZzQZ+DnwCPQ9/DfzQzKal/b4HPI7elf9N3r1J1/s/6B2Zhd6ZfEXGn6H79rq0fg+6193ZtxjvTW2YCjyFnj+AbwFXZkKiybXzjantveHb5JQO7wZ+giwzh0nC1YWpLd+lo5KCgm3nApeg3ysIgsHG3eMTn/iMwA+wFnhjkeUXAhsLtrsq7/u/AV/K23YT8B/AfcCEMuerBP4EuB9NFDYDV3fnPEWOtQLYk/f9LuBfCra5C/iD9P81wCt56+rQ5HIGEpQOAXV5668Dritx7q8D/5b3fSxwEFhQ7r4W3nc0Af9XJMjdBlSlNi0ADE26j8vb7xxgTd59bwIq0/dxad+z8rZ/HLgi/f9x4KG8dRXAFiQAnQWsL2jj3wHfzNv3noL19wD/DEzt4hnr8b1CE9qtXRx/NXBJ3vc3AWsL7lFV3vrtSBCqTO1YkrfuE8B9ed8deEPe93cB9xac/8vAP6X/1wN/CIwv2OZf0OR4UTfeSQfqgb15nzfl/Q635W17GdBQ5BmYmPf8fypv+2VAa7r2jwLfKTj3r5AAk70PY/LWfY/0PiAh+vq8dWPScd+Yvr8IrMxbPzPd66qu9i1yP64t2H4s0AbMzTvXRen/PwV+UeI4C9K9qSpy/E/k9RH3AaOBbUggfgg4Lz0b1+bt9zHgqfT/7NSmUwre8Yb0+60DvgiM7ur3j0984tP/n7DYBEEActHIaEQTjIyJyFLyr+6+r9QB3L3N3b/g7uelfT4JfMPMlnZ1HjOrM7MvJxeWejSxnpisHRkbunsN7t6Y/h2LNMe785Z1daxZ5FmE3L0B2IUmOEfCd4D3oAlVoUveNCR8PZ5chfYCv0zLM3a5e1v6vyn93Za3vomOv9Pha3L3dmRFmIWsYLOy86Rz/T2yLnTaN/FBZE16KblxXVriGntzr3YhN6Ry8S0djp/+n5V/DHc/lPc9e6amoYl2/nUV+83zl80Hziq4T+9FwjEo5uISYF1ymTonLf+/wCvAr83sVTP72zLXA3Cqu0/M+/wqb13h77uzyDNQ9DdH96YaWT7mA+8ouJbzkRAyCykNDhTsmzGLjs/SAfRbZcwHfpR33BfRxP+YbuxbjPztG4Dd5H7jb5FzA7sKvVPFyJ6B6oLl1UjoOoy7NyFr1seAKe5+f5HjvZ9kOXK5mt1NZ4vfFen3m+/uH0nHDYJgkAnBJgiCrtgDXAp808zO684O7t7k7l9I+y7ranvkzrMEWSTGAxek5fnubD1179kCTDaz/OD9uWW234wmb2qAXJSmIMtVt3H3dSiJwCXAzQWrd6KJ6mvyJrgTXMHIPeXwNSX3nTnoWjYgS1D+ZHqcu1+S39yCtq9y9yuRS9angZvSfSikN/fqQWTZK5cGusPxkbVhczeOvQNNdvPdoIr95vnXvQG4u+A+jXX3PwZw90fd/XJ0T35Mcs9z9/3u/lfuvhAlPfhLM1vZjTb2BfnXNA9N4nema/lOwbWMcfdPofdhUsHvOS/v/y10fJbq0G+asQF4c8Gxa5MA0NW+Za/BlFlsMrnf+DrgclO82FJ034uxJV37goLlx1LcbfXbqM/pFGdnZucCxwN/Z2ZbzWwrsnq+pwshPAiCIUAINkEwsqk2s9q8T48GZne/C2mvbzazM4ttY2Z/YQroHm1mVaYaM+Ponu/5ODTR35sCev+pJ+0sRhIwHgM+bkpJfQ5y8ynF94EPmNkKU6D6/wEedve1PTj9B5G7U752PLOofBX4TzObDoqLKIh7OVJOM7O3pd/4L5DQ8BCKLdpvZh9Nv02lmZ1oZmeUOpCZXWVm01I796bF7UU27fG9Sta/f0RxMVckq121mb3ZzLJMYd8HPmZm01KMxT9SZDJa5NhtSJj8eDruCZSJk0jcAixOwe/V6XNGipOqMbP3mtkEdz+I3MnaAczsUjNblOK59iHrRbF71R9cZWbLkgDxL8BN6dqvAy4zszel37s2vZtz8t6Hf07XdT4d34ebgEvN7Hwzq0nHzZ8rfAn4pOWC/KeZ2eXd3LcYl+Rt/7+RS+UGAFcs36PIUvPDUlaRdM0/TO2akn67K5FSpVO6ZmSBuQj4fJF1VyPX0WXIJXYFiqkaDby5yPZBEAwhQrAJgpHNL5DAkH0+3tMDufttKFD5Z2Z2apFNGoHPIJewnSje5u3u/mo3Dv9ZNHHYiSbjv+xpO0vwXhTDsgv5099AQcBwhrvfDvwvNFHagoLn392Tk7r7and/rMTqjyIXpoeS+93tyGrVU36C4kT2oODtt7myjLUhi9sKZEHaCXwNxRiU4mLgeTNrQIkE3l1sUtnbe+XunwH+ErkF7UDWgD8lp5n/BJqEP4OyWT2RlnWHP0XXuBVNjL9Pid88tWU/8Dup/ZvTfp8Gsmxu7wPWpt/qj9AzBdLu345iLh4Evujud5Zp19PWsY5Nb2o9fQfFkWxFiRj+PF3LBpQQ4u/J3df/j9yY/x5khdiNlAiHXSXd/Xn07n4P/aZ7kFtjxudQFrFfm9l+9L6e1c19i/G91IbdKDFGYQaybwHLKe2GlvGRdIxnUKzVnwJvcfdthRu6+I27785fbsrS+E7g8+6+Ne+zJp1/wAsCB0FwZJh7T707giAIhidmdgPwkrv3mWUoGNqY2aeBGe4ek9NhhJldgCxQ8z0mLEEQdEFYbIIgGPEkl6LjTDU4Lkba7FL++sEIwFST5iQTZyK3wHKppYMhhplVA/8P8LUQaoIg6A4RCBcEwdHADBRzMQW5xvyxu0fdiZHNOOR+NgtlG/sMctcLhgGmbIqPofpLH+hi8yAIAiBc0YIgCIIgCIIgGAGEK1oQBEEQBEEQBMOeEGyCIAiCIAiCIBj2hGATBEEQBEEQBMGwJwSbIAiCIAiCIAiGPSHYBEEQBEEQBEEw7AnBJgiCIAiCIAiCYU8INkEQBEEQBEEQDHtCsAmCIAiCIAiCYNgTgk0QBEEQBEEQBMOeEGyCIAiCIAiCIBj2hGATBEEQBEEQBMGwJwSbIAiCIAiCIAiGPSHYBEEQBEEQBEEw7AnBJgiCIAiCIAiCYU8INkEQBEEQBEEQDHtCsAmCIAiCIAiCYNgTgk0QBEEQBEEQBMOeqsFuQBAMa8xmAkuAOmAv8BLuuwe3UUEQBEEQBEcf5u6D3YYgGH6YjQV+DzgeaANagRqgEngM+DnurYPXwCAIgiAIgqOLsNgEwZFiVgtcDUwD1hWuBU4HajG7Aff2gW5eEARBEATB0UjE2ATBkbMcmIVczyYgN7QMR8LOScD8gW9aEARBEPQcM2rMmGHGTDNqB7s9QXAkhMUmCPIxqwSmA9VAQ6d4GTMD3gosBFYgQQZgH/ASsC19bwDOBtb0f6ODIAiCoHeYUQecmz7Z/LDNjEeAe91pGLTGBUE3CcEmCADMKoAzgAuBsUhgqcDsVeB23NenLc8GLgC2AvlCz2g0GDwDrAbqgdkD0vYgCIIg6AVmjAE+AMxA41sWI1qFxrZlZnzNnX2D1MQg6BbhihYEEmreClwBNAMbgI3AemS9+UPMTsBsOnAZssY0FRylCQk6y4GJ6N1qG5D2B0EQBEHveDOKG11PTqgBOITGxDFojAyCIU0INkEAr0EB/2vpLLDsBrYD7wLOQx3+VjrG1WS0AweBY4HJwAv909wgCIIg6BvMGA+cDGwusUktGhtPMWP6gDUsCHpACDbB0Y1iZi4EdpKLlymkCcXcvDFt9woSbKzItg3APJT2+Yk+bm0QBEEQ9DVz0HhWOAZOBM4C3gS8DikAP2rGqWYxfwyGJvFgBkc7Y4FjgP1dbLcPZUJrQxac1cBM5I88E5iKhJ9qNBj8DPcd/dTmIAiCIOgrisVbT0PCzGTkubAHjYMGvAO4LISbYCgSyQOCo51KSltq8mlD1pixQCMqxlmLOv0K1NkbsAu4G/eH+6W1QRAEQdC3FCYEqAHORGNefryNIQFnO0qksxp4biAaGATdJQSb4GjnAIqLqU5/SzEOeBC5mS1HAs2rKJ3zaHLJAmYA+zCriOKcQRAEwTBgA6rLNhYJMzOR0i8TaqqRsNMG7EDKwF0oQ2gINsGQIsyIwdGN+0HgASSQlMKAUcAt5JID7MmOgCw4Dajz34Bc0Rb2U4uDIAiCoM9wpx2Nb9PQWDcbZQgdByxCCXZOAsajOJuJqKTBLDPGDUabg6AUYbEJRh4qslkHtON+oBt7PAKchuJkdhYeDZgPPI77esz2AJuQxaYNCTpV6bMbeBSlxTw71cCZD0xAGdM2dir4GQRBEASAGbVIqKhCcZ9b3LvlKt1r3HnJjO8Bb0dlDiYCU8hl+/wtstJMQQl3HkaKvcqBaF8QdBdzH5B3Jgj6H7M6JKC8FsW/VABbgLuBF8q6hplNBd6DEgm0oNz9tajTfhj4BerE/xnl+Z+IMsmMRlnTNpMr2FkFHI8EoIn5ZwGeB27Bvb63lxsEQRAMf8wYBbwexa1kgkIFKi3wC3dWD2Bb6oCPogQBu5CAtQ+NiRnVyHrzFPBP7mXduINgQAmLTTAyMJsA/D6yumwlZ3kZD1wFPIbZj3AvXjTTfSdm/40sLCcggWUH8PxhK4sKeToSUPamz+EWIC3XuPR3CfADJARZ3n6LgA9h9hXcu8rEFgRBEIxgzKgB3ovclzfTUYAYD3zQjO+68/xAtMedRjO2oAQBG0psdhB5JuwuJtSka5pKSjbgTmN/tTcICgnBJhj+qBbNO1Hg47qCtfXpcxqy3txf8jiy6KxJn+LrzX6LLDW78tZMB05FvskGLEUxN29H8TcHUZKCVWjgOgZZlX7R7WsMgiAIRiKnI4XX2iLr6lEA/++ZsWYgBAQzKpE7XLGxLqMuta26YN9aVMj6PPLml2Y8Dtzt3kEZGAT9QiQPCEYCs4AFwLYy22wGLsCst8L8g0iLlhXnnA6ci4SX3cgyk8XcTEJCTGNafgrKIrMXOAuz0b1sSxAEQTBMSULEBcjLoMMqcmNMMxIglg1QsyrTuZ9EisLJyKW6DnkkZPGl96P4GzXYGA1cA7wBjYUb02cLGvv+yIypA3QNwVFMCDbBSGAxHc33xWhBpvOZvTzXatShL0Ad/anIB7kVpcOcj4SY+rS8HWm/WpHmqxZZjypQEGYQBEFwdDIJjUvNSJiYAZwP/C5wefp/BvIAWDJAbTqIxq5qJNz8hlzigE0oi+hv0Di3PW+/N6Cxbh0da9+0IcViBfAOs8MCWxD0C+GKFowERlO+Bk2GU2A6P2LcHbNfoPibdyCLTZb/vxElEnByRT9b0rpR6f96JNBMJAiCIDiayeIvK5CSbC4aR7JENHXAOai8wJaBaJA7bsZ9wMUoxqYBeLnIpuOBn8DhhANnIQGmFDuR4m82suQEQb8Qgk0wEtiDBIeuqECxLr1DsTgPJ1eyKtRht6bPjBJtyQQbkHVpGsV9l4MgCIKjg3pk0ViG4lkKyw00pc8C5NY8UDyNBKpiJRBA7t/rgVfyvhu6lnK0IeEmBJug3whXtGAk8CIdfZKLMR7YjHu5OJwj5RAamLYjq01WeyDTwJViNLAJ96Y+bEsQBEEwjHCnBXgCOJlc0edCKtC4MsWMMQPUrgbgm8h6NB8JOOORcHUsHG73gpQBrbu1bNqRy3YQ9BthsQmGP+57MXsA+SOvg04FzWqQ+9fX+/jMm+jYoR9M516PNGzN5DRYmRAzPi17sI/bEgRBEAw/NqOxow65feVTQa6EwUnAf5ixFdVDe9i9UxbQPsOdXWb8N0pDfTJyqT4GjadjUAyQISHnabqnKK9CbtxB0G+EYBOMFH6NAvNPRx3tPtTRTkLCxg24v1J69x6xDrmTTUjna0Kd9njgVeRaMB65v41L+2xD/tOP9HFbgiAIguFHGyp0eQzKONaOxqwsO5kj1+VqNL5sQQlzTjHjN8Bv3Dsp8/oEdw4BvzVjFfBWFEezGyntdqW216DCohOR58K+EoerQQLcqv5oaxBkmHu/vA9BMDiYzUZZx+aiTvd54BncS3W2vT3fHODD5AI+p6D0nXuQVmsU8CwSevYiQWcD8A3i5QuCIDiqMWMJ8D5k6Z+ALDRVyHozHwk7+9LfR8glEahAngHXu/NUP7avAtWJ+wg5V2tDrtgvo0yhIIuSo7ibwno7VcA84EfuPNxfbQ0CCMEmGMakGgCjgIPFqh8PYENmAZeiQagdad6Wotibh9EANRYNWuuB7+Je6HIQBEEQHGWYMQr4KLKA5I9j44CVaXkFUordSsfSBnVp3WfdczVl+rBthlJP/z4SZvIVhJXII+JV5Io2jZx3wihy9dvGoHHxVvdwwQ76n3BFC4YdZkwGzkBm8WrAzXgWeNB9ELKtuG8GvoLZMchi044GqOOAFajz3w7cAqzCffCEsCAIgmDI4E6LGXcBbwbWkosRnUkuRnMy8AKd67U1IkvIDMqnWu4pC5GbWQWdXczakNC1EMWb7kbCzaeRq9zxab8NwPMpIUEQ9Dsh2ATDCjPmoerGFUhYOJT+X4J8jm9y54lBaZwyruVnXVuNYn+CIAiCoBT3IYvMecjqsQdZY2qRsuxVVCSzkCqkODvZjLHAWvcOxTF7yznI/Qw6J+XJlrUg4eYxwFKmt2fTJwgGnBBsgmFDSnX5PtTx789b1Y6EnBrg98zY5s6mIgcYB5yIXMYcBTG+hHuhP3AQBEEQDAjutJvxcyQMnAksQq5e9UhgKKx5ZmmbxUjwOYhcnluT9ee+PnJNOx7F9OwilwinAjocuwEVqp6ELE5BMKiEYBMMJ5ahGjCl0kW2oiD9s4EfdlhjdjZwCRoQDqS/y4FDmN2E+3P91OYgCIIgKEvKbLYufTBjGvAXFC/kfBKykjQiN7AX0vIa5NI20Yyf9VG2NE9tWEEupXMrGof3kBNyxgM39cH5gqBXRIHOYDhxBvLjLccOZJbPCe1mpwNXoFoAG9IxdqFA/l3AezBb3B8NDoIgCIIjxZ0dyIIzt2DVNCTU7Ebuai/mrWtFVpNzUCHN3rIWlVBYhqxCRi4pwBxk0ZmQtn2KXIa0IBg0QrAJhhN10GX2s3bU+VYDYFaDLDWb6Bx4CfIP3gm8BTPrs5YGQRAEQe/4MRIWFqA6MZVImKhGrl/P0jlpgCP3sLP74Pw7kPv2bhTjswONw6PSeSagDKA/BW7uj8xsQXCkhCtaMJzYhbRVLWW2qUbCT7bNcagTLrdPA8osMwdZdIIgCIJgUHGn2YzrUDzNuWiMmovqx6yhdDHM3WmfHpNSPS9CdWmmpGNuQglyxqH545i0/LZBLbkQBHmEYBMMXWRBmcThYmX+MPBeVOiyFNOBe/M0R5PIpcwshyMf4SAIgiAYErhzCHgpfTDjn5GVplz8TFZEszdMRQLNA8gqs5Bc4oAK5AHxBBqPl5Fig4JgsAnBJhh6mFWg4MgLUcfqAOuZ88IZPNq0jZkzULxMIeOQEPN43rJWZL7v8qx0TwAKMiR4zkHCJCgz3UYGoOqvGXOQ7/cJaJBdi4qhrgl3iCAIRjBrURHocvGmfZGhLHM3awOeQ1lEp6IEBVnygFZUY2dcL88VBH1GCDbB0EJCzWUo+HE7sOEeXjvpZt66aB3zLziNx6vv4sKmRsZuQ+b4veT8jZuAb7izJ++I3XEtq0Ad+MAX9xyumM0B3oYG2Hy2YvYj3PvlXib3iJXA65F74W6kQVyAstw9acbNScsZBEEw0rgf+ADlBZu+yFDWSMc47BYoUkZBtXb2FFkeBINCCDbBUGM5EmrqgWk38I6TruOq+Y7tm8zujXU0jXonN476Du9raKP6ZBQ8uRW4G3jBnY41ady3YbYamE3H4pn5zAIexz0qI3cHCTUfQmmzC90PJgIfwuyr/STcnA68MZ0338K2M31ORjFTv+iHcwdBEAw2r6AMZCchZVx+P1iJYnB6naHMnd1mrENKw3Lu39VEMc5gCBFZ0YKhg1yb3or8ec9/gaVn/oo3nXcKT049g8fmjKOhrpna5onsq1rB05uBh4Bm4Fp3Husk1OS4GWmf5tBRmK9BxTq3Ar/ur8saUeR+owMUH+z2pnWX93WWuZTCeyXyLy/lNrgBOMcsXCOCIBh5JFfbm4H7gJloDJuX/s4E7gV+2EcuuXcgwaamxPq5SKFYzDU8CAaFsNgEQ4kT0cR1K1B/C29ZUMPBBqeiBdqrZrJl0WZmvdJKTfMiXpn9OKc/hjrzE5CGqjjuezH7CnA+quqcxdy0ALcDD+Le3I/XNXwwG40EyyXoPm0EnsE9c3uYjdzP1pc5yl70u8ymb9375gBjKe+CkaX7XgQ82YfnDoIgGBKkDGS3mnEvCuofjZR3r7pzoA/Ps9qMG5HbcTvKTNqO+uHxKAX0zX11viDoC0KwCYYG0u5fhiwwzQeoq3iRZRNmsakR4BBVhwCbxo6525m+ropD1WnPvcBplBNsANz3A7dididyl3JgN+6RojLDbAnwbuRa0IAGsMXARem+3YHSbXeHdhRo2peCzWjKZwLKaCMy3AXBoGFGBapxYkB9xLz1D+40AM/08zmeNGMDsAK5v1Ujq/nNwGr3SLoTDC1CsAmGCnPQQNgCWBOjKxyoyJvHHqLqYC3NY8bSMHEjc7IgxlZUMKx7yDITZvNCzBYAV6M4pKa8NXuR5WYlEhi6GyTaH8VOW7p53EroO61lEATdw4wa4FTgAnKZsprNuB94BCknTkHW+awm2UvIuvpqTJKHJu7sRN4Ntw92W4KgK0KwCYYKU5CQsh6YM5aG+kra/RCVVkXbYenGMa+hddSzLM+yndVR3i0q6B4XIyGmqci6NnSPXw98i+4JF4ay2vWYpPWdgdKONqH4mSZy6UZLnRd6GTgbBMGRYcYoVGfseKQ8ymLwalDCj8tQ7ZPRKIvhKKT9vwilDn7MjBvc490NgqDnhGATDBVkoFHGlzm1tIw6k4d3PcoZk2ey9XD8Sw2tdRuZ8/I6FtSnRWOBRwehvSMHs+koCLRcgbU29PtMRO5lkyke6zIeaWMnAB/AbAsq8PYK7t1yR0kpnVcgK1HmNmgo69k64DWoRkMxt7Q5wFMFKb+DIOh/VqJ4jzUFy0cDxwLnoTiQdpQ2eCeyrk5CBR4XAivN+DLwS3fqCYIgOEIiK1q39jK6AAAgAElEQVQwVMhSMTegbC+Vb+EXbRV4TRO1tTW0jq6leUwTo/fdwRseS9vOQAPkq33WCrNKzEb1dUavIc54ulectBW5j/wYaVunFKxfCLwFZeh5BLmtTUMubldhVtvN9rwReBfS7q5Hlpr1qL86EdiHJkpTyFloxiMt8Grglm6eJwiCPsCMOuAsOtc5WYgsvctR4PlE9M7OQgqKZagfr0fvdxPwQeAvUhHeIDiMGVVmLDHjUjPeasbZkQEzKCQsNsHQwH0rZmvRRHgncNscNk3/Y/5nyRf4yJKNzDk0lob9BxizYxWLDWXd2gxc1yeBqWZzUf2cE8kGWLP7gSdwH+maw0N0P3alGfctmH0ZuBz9Do6EjJORtvZJOKxt3ZM+C1Ga6O+XO4EZxwJvQBaZwnSlDSi5xGzgpyixwaK0bjPwS+ClCFQOggFnDuo3MwXJJKTgWIEUImNRPzEaKSay9PvNdKwvVon6oxrgGjM+587+gbiAYGiTBN33IiVWExofTgXeYsbtwD3u3UouE4xwzD2eg2CIIJeoD6MBcEe2eANzRt3Jhcse4Nw5P+GKR7cycz2qYfNKHwk1Z6FJemM6ryOLxDTUgX4D91LFPY/0XKPQ4N6Ke6m6OwOLWQ3wt0ijWi5L3HzgS7hvOLzEbAYwHbgSBQJv7mL/z+FeMvbGjCuRELSj1DaoVsMT7vwsxeHQRzUbgiDoAWYsQ33AfpQcILOgTkZxNDWoLx2D+tn8RCDbkfAzCrm3NiPliCOXtPsH6jqCoYkZM4A/QsqtQkVjJRpbbnXn7oFuWzD0CMEmGFqYTUPuTMeT09hnsTc/Lzcp7sG5atJ53odcnYoJSZORFvJzuJcKWO/OuY4BzkUazAo0qP8WuA/3vnOl6ylmK5GlpFSczTQUDPxFCjsNCaR/TtdJHGYDd+J+V/EmYMC/oBiech1TDTDWnU91cb4gCAYAM+YBf4WUEq3o/T0X9alOLvVzBepHxqXtWtBkdR9S+GxDypXHkaKl2p1/H8hrCYYeZrwHuR+XUnhVIoXX/w0LXxCuaMHQwn0HcC1mU5ElAGA77jv77Bxms5Hb2ckoL/8YlF1tLRpk89mNXCoWAS/08HzHAdegAXsLuSKSs4EPYfZj3B/u0bH7jnuRa8gJKKNRlh2tChXkbASu7yTUiDo6u40VoxVNcEph6dOVtqWd6LuCYCixEfVnkHMzk+uqyJKP1KB+0JElp5VcweQKpI2vS39bkGIpOIoxYwIqGl2uJlobGjuWAYM9lgaDTEwOgqGJBJm+E2YyzE4G3okm7tvRO7ANaYOORRm8dhXs1YB8eY9csDEbh/yCd6EBP8PTsn3AFZhtwX3w0la7t2L2PeB0VINiampjO3L7ewD3fSX2bqZ7MTpZ4c8STaDdjG1Im1tO6zaezkHKQRAMHjNRfzYP9amZpTujGb3XB4FacsqdWvSuZ38rUb9fjyw4Iz2+MShCiqc5HSnaJiHPimb0bJRSfDWjRBTBUU4INsHRg9lM4B3IapIV9swm73uRNvFsVISsJW/PVmTV6QnL0YS+VDzNIVRM8mwGux6P+0HgQcweQYNJBbAP95byO7IdWbbGUkZwQffhxS6OdS8SPMsJNhOAH3VxnCAIBo6pyE1oB1ICjUZCTHVa76iPrUPzjixBQDUScFrJpY9/Ku+Ytw5A24MhQnJHXoky6bWgZyIrM3A+ypD5BMU9BCooHyMaHCVEuufgaOJMNIBmsTLZX8v7XknOpSJjNPS4Lsqp3dh3B7Acs6GhaHBvw30n7tu7IdSAezvwGxSHU1liq5nAKtzLJRcAWcXW0fk3yJgLvIxirrrEDDNjghmTUwHBIAj6j/VIMbQDWXAMKTu2oMQiu9K6LFkAadku5AJ7b9p+PNLAPzuAbQ8Gn9NRuv+N6HloReNnffo7B7mbFaMWWDUAbQyGOENjIhUE/Y1ZBRIytpGL4ziENECzyLk8HEDZfPID+utQMGtPqKWj9acYWQHKTJM5HHkWCTYr0T3MXAbGIs3rBuCmrg7iTqsZ1wFvB5Ygd5ZWpN2tAp4GfuJevu5Oypa2DGn+jkEaPjfjEeBB96LFRYMg6BnbySmIGlA9qTrU385Glpl29C6vTd/HICFnCxKI9qL+ci7qM6+NQPDBI1lP5iOF4ELUn68CHnVnQ7l9e3i+KjR+bKajG2Mbep6WIgvOwtSO/HF1Anp+Bj8RTzDohGATjHwk1CxFaUgzi8JupPVfgwbSWjQwV9LRkjkTDbqF1bS7y07k91tOuMlSofY869pgo6QCv8FsNSrUdyK6n9uBG4EXuptVzp0DwLdTis8TkHC0D9WoKZcGGjgs1FyKEkTsJOfiVwmcAZxqxtfc2XIEVxgEQQnc2WLGelTPahfqW7P370BaPg1NjuuRVdaBLyFF0tlIAKpH9aieCaFm8DCjEpVAOAP9fnvTqhOB0824B/h1H6fZn4P6+t3Immdo3PQZbN5q+OJWao49SHVbI2NmHKJ6Xdom8xT4elcKr+DoINI9ByMbs0rgCmTiPgV1mm3IvWwMMnc7GlirkFbxILIM7EApma/DvWeDrNkJKJ10qTTKoA79Ttzv7NE5hiJmynAmN7UBPjUrgHcjYbRYBzcxLf/PKOYZBH1DUkR8GPWfu1B84SIkrFSj/vVeNBmdBFzvztOD09qji5RZbDzyCNjRVb9nxkpkPVlHLpXydKT025OW/dCdB3reJqtAFqFlwBh4+zj41OthUWbNo5qW9jN5pOJknqky2tvXcOykVSyaMZk9e/cx4bGXWLYVuS/f4U7f1JoLhj1hsQlGOhcAp5FzfzgBdcxNSBt0KrKUPIey9sxDrhFjkBB0S4+FGvEK8heemY5byMTUjp66ug1NpDEZcK1Jcp+4EFmKSp1/LxpQFwEvDUzLgmBk485WM74C/C56v+pR9sIFqI97GbkMrQducg+3of7GjFlIQFlCLhPdATPuRi5lnQQcM+rQuLkBCTNnIEGmBfWpM9P3cWY86t51wL4ZtcBxSLhqgU/uBd6M3MBboLoNWl8Ln3kNLN4IH361mipbyosn7GbymLu5YN9ynls9jZ27mhm153Xcs+4Sbm137K4z/NHf9PY+BSOLEGyCkYvZKNRBZwUfN6DOdTQSbKYhDdQoZMXJglXvTt8nAO/C7D9x75mJ2/0QZtcB70GDfWM6dzXq5PcCX8c90pr2DZOR20tXPuAHkFtFCDZB0Ee4sxX4SrLeZHXIsni7aqDRvR/S+AedMGMh8AE0rm0gp+ipRcLnAjNuLOK+tRDNDScid8J6OmYba0Tj5onA24AbyrTBUKHWNyKX6zY4UAsrToVP7oT/ug+2tcLKE2CRw4Sd8PRY+K95s/i95jE0Vh2gbk8jdaNfZvHcE3luTQXuV3L9S8fzykHgQswejfEzyCeyogUjmfloMM20Uk2oTk0FCmhP2iJqkDtYC6rZknX0+5DLxLG9aoUsPl8FvoGCHhtRgOT3gM/hvrVXxw/yqaJ7lqJDaIAPgqCPcWerO8+kz2Z3trizPoSa/sWMWjOmmDEVuBJ5J+ygY5/YjNx0T0YeC4WMSdsvR2NVMYtMOxofL0zWmFKsBC5LbVgHbIQHgVda4NQx8O/nwJxaOH4R7NsN9dthkcHzkybwxOwmapsA6mhs2seE8WtZcMxJPLP+eF5pIjeuL+/OvQmOHsJiE4xkiqX33YfSkS5BwaqOOvqXUH78Qu1VCyoO1q30wiVRrMnq9Bk8lFJ6MrkaNU2D2p6+5wByt8gy35WiDrmrBUEwAjBjLJrknkku4chDwAvuJeuIjQjMmAachwQVQ5kg56DrL1VbbCvwOjMeL0gC0Izu3wQomz0y22cx8EyJNr0eCTRpXG0zeHkRjKmHrYdgzhR4/zI4UAlt7bB7O1TXVDJx7m5eqJ3EknoVOaqubqK2rpbmlo/xyfwU4PuR4vH+Mu0MjjJCsAmGNmYzUOxLG7AZ9+Yj2LvUpP0QuRSje9BEvzDFZEYbuSJzwxezGjTgvxa54gE4Zo8C9+G+t+S+wwh3Gsx4Fg225QSXaooMxkEQDD/MmAu8H1lhd6F+vRYljllpxrVDNbjcjBrUX51GLrbzceDV7mT5MmMe8PtIkbMFjVnHpNXnoAQ4zxfZtRFlBJ0MHSxpryIvhnJUpPPsRN4PxTgFWXvyrqGhGg7WwIRDUDEK9rbCuQvh9pQx04GtG6vwg0bTsgOMrTOc0TQ1zWHjxjdwx8uT2FsYFxQZsIIOhGATDE3MFgJvQlqnLOjxIGYPAnd1M3XwOiTc1NA5lXIT6pwtHb9UGuHRMDQHxC4xqyYnxPweCpbfggZ+UADo6cCJmH0N95HiJnIP8BqkdSymrZwPPJXiAYIgGMakjF/XoIl6fj/ekD6TgGvM+PxQs9ykWKT3o3jLLJZlIbI8bTDju+XSXic3sKuQ5SJ/u0p0P5qR0JQVQC3EKQhJcGe/GU+lNpSy2ExCXg5O6VIGx5KrD5eomQQT5sGkPIv6pDqYuhP27skWtXBg73T27zyJu188RNWhCpyNzJ48m00HCs4xjt56UwQjjoixCYYeZkuBP0AT03Uo8HE9GrQuAN6bLBBgNg6zczD7MGZ/gtlbMZuHmeF+ELgTuZwVPutNaJI/B7mHFROUKlFPW0zbNXQxm4LZJcA/AH8N/A/SXB6g4yAkK5juzbtSiuZhTxJYvomEuvkoCHYcyuYzn1Tkc9AaGARBX3IKssCWEgD2oLGkVMX6QcGM8cAH09esQGlW3HgdSr7w/lS4shQnICtPdu21SKCYjMa9unTM44vsm41vxe7b9Wi8nUZOOUY6XlZw+bfpGGtLtK3QkjILRp8L01th30E41KhPewuMngyzF+ZqvNbXNXHyCxOon1CBc4jKikra/CJuzxdca5BScniNz0G/ExabYGhhVge8E1lJCl3JDqEBYDFwOmZ7gHehznUvmqgvR+kpn8bsR8jHeAJywWpCGihHGqeGdJ5NRVpSiSbBt+Jeykd56GE2F2XCqUCuWBXIKtMGnI+yvhVquHaiNNdzyRWzHNa4s9aMz6CBfzkaBF8CnnRn86A2rj8wm4A0vdlE5pVh9dwGQc85l9IW94w9abvH+r85XZOKCF+ChI9NaLwpdDvbStdp6U9GsUSGilAvSssrUHbIMWjc24tiTvMVW8cAj7t3dtl2p96MbwBvQdakyWnVLqQY2kZOwCnVn65CMTYN6dynqR3LWuCORVDXCqMq4UATrHsVZi+Dhr2wtR5gDVc9vo/HTh3L/skvsbTyIm57bgbbMgVkXWr/jdHPBYWEYBMMNU5Ak9ByQe1bUbrK9vR/ftxNDerw/gC4GPgRGsyeQzEmi9EgsAZlJWtHxRznp3O2k9NQ/RK5NQ0PzDK3hANICFyMXLLmo3vUjHyu99F5ItCGJsYjQrABcKcZeCp9RiayXF6ChFdDv7sKzcpt89e4D3oR0qR1nozauC/9NkHQK8yoRJPcrtxom9BEf9BIqY9nI7fg1yEL0h5UgqAZCQKvQIdA/npUPLqUYDMKvfMnIqFmFzlLySg0FtYgpVUdOcFmYtrvvjJNvh1ZuY8FHiZXx6YyXccB4EagwgwvSEAA6nffgPqj2Rzun+bvhyXb4OVjYGEFPLEG9u2FmrVQsQiad8AfPtDESe3fZ9GaFTx9wju5sfXP+HwLUsBZum/fwf2FMu0PjlJMdfSCYIhgdiWaiJfLxgJwKeo4N2Z7okn8cajDbkQTqSfRQHEXcBvFHnhlCluEBIFqpIF6ftjlxjc7BXg76vTPR9fShgamxvR9LNKy/YCOrgIzgIdw//VANjnoBXpur0QZ/vLrVIA0tvPQO3JTyso34KTA6LOQxTRLC+toonS/O/sGo13ByMGM/4XGi3KFIusAc+c/B6ZVHUmC/RUo9fGxqL3Hk7PSbMz7+zgdhZNR7vx7ieNejpRVJ5GLncxnFnJpGwfcgRR6ten83+0qztCMUeTe3yzLqCNl4U6UhW1mWv4qyk62KhNyzDgHuBwVaa3ksBKyvQI2zIMHauA7u5QRrd1gRQWccT28sxb1YWuBJxzLEh1UIte59YPVpwVDn7DYBEONSuik+SmkDmmc8idFi9BAka+xakWT+VVIc9SA6th0RBrtlxj+xRpPQRq+M9E9yDIDZY7LB9OyBemzJm/fGrrWegZDi8VI67umyLp2NCk4BU2UBjzNeJoUvQ9N5LaSe76qkBZ6uRlfcy86IQuC7vIwir3MFFGNdI6ZnALcOpCNKuDNSNmUuW85EmqyYpfz0Dg1FynWstjHaUB7SjKwzb1T3MqTyB27VPa0zeheNKFx81nkvbCmOxnX3GkB7jHjAXJlAhpRYc5z0XibWfmnIzfoR834sTtt7jxoRiPwceQWNxowqHCY/yTMeRFOqIOGSsXeLJkC/Mi9MOmAA4R1JugWIdgEQ40NaMJWLv3wGDRxy9zVqpDWOoufyTiYt+1GYCVmj3Uzo1r/o2D96aizbwW29lILNRr5Q48lZ/FqRvdhKrn6LqPRQPsMGvi2oXs03AW7o43z6dqyWY80ugMm2KRMTUtRYPRrUKxXNXoHm5FFdSOatL3DjC8XmbAFQZeYUYdiKM9HrlJZ374eBbc3IWtFK5rU90cbqtAY1FrEHQszJiFBHtTPZtvsTW1rSZ9ZaPw7Hgkhx6N+exXwp8BWM37p3iFGcgNyCZtWonm1qP+/Fwkm97h32Wd0wp1DpPT5ZrwNeUYUKlR2I8XZaUjguSvt+7QZ30Lud/VICNsNtMJBg8ZKaKqE5lrUPwypzHXB8CMEm2Co8QzwOxQPpszI6s5ksQNTyeIKOlJJzj2hFXXyC9CAN7iYnQBchFzA2pAmbC9mdwBPFnWZ65qdaADNrrmCnG/1uPS9Jv3NkgtMRcLQl3GPAWW4IKF4LprYlGMvcu3s4WmoRLFXc5FQvAW5mhR1+0kVz69GE63jkaXGkGVpCdKuZ5abHalts8m54gRBtzBjDKrfMh1lvzw9rdqPntfZwMto4n9tubTJPTz/fGS1eA0p7bEZ96OA/HyLw1LUx89IbcnYgZLYtKI+uy4dZynqlxuQoPAkEnzGAx804wZ3xQ2642Y8ghR4k9GY2Ir699FIULgvnXcecLEZxyFFw/a07gX3ollBi13zROR+VqrfcTQ2X2DGA3nHfRQl9dmsbQ4afH0B/GoJHBgF5jB2Iux5DO5cAB4pnIMeE4JNMLRw35Mm9xchrVth4PM01CHvRgNBI5qsFxMEsniZw0dP+wwuZmcBb0UD27q8NXXAO9A1/qoHR34MJULIBMK5aODci+7RFDTJbEL3YiHSuj0JLMVsPu7rCg/axbVUIe3d6Wjg3Y9cn15J6baD/iWvHkTJ9T07sLEAPY8T0cTL0XPUmFxNni/YfhQSakaRq8mRvb970r7noElolskoiwELwSY4Ui4m59oFeq7moeepEk3sq4D/ci/rAXDEmHEeivNsIOdaVoOUcm9Lrls7UNzJVCSYFL6rjei5n0PuPZmM+tHN6D16gFzAf336/21mrMmLT3sKKQjq0fWPQ+/es8ga34aEpdcg9+3tadlYlIF0kxnf7qbgt7DIdRSSuYDPI2XgdGeTGY8Dp0PbOvin5fDQcTB1H8xqhLFj4dBW+PVO4INmdqO7P9mN9gRBJ0KwCYYid6COeSU5q0tl+mxAmViORdll1lLcsjMWdeL5ZnejeL2agcNsOgogLSa0NSJB50LMVuH+6hEefQ1yOTqdXErrA+i+1ZBL+XkAvfuZYPUsGlBXAt84gmuZhGIojkEDfDNyC1kK7MDsuhFU9HNo4e6YrUJa6XIxKlPogW96quT+QfTMFAq7o4Gr0mToxbzli9FzlNXgKKQVCe/zydWeaKPrKudB0IFUA2YFHVP1N6Bn/QVksWhHz1ptpwP07tzHIaFmAx378NkoO9kYZK18GLgQvRNN6D2dTcckBzvTuumoH52L3ol1SCgqFDYyIeck5F4Gut5LUP/7TJEmT0cJAO6mo6IvK+o5A7jSjK92wyW0js7jVilGFXz/KdAKd74Fnl8OJ+yAqjFQWQn79sAvHoWGRjRuv83M1rr7nlIHT1nmZiHXt+y+Pg8819fWuWB4EYJNMPSQG9Y9mD2G3FemoEnRq8CmNKnbizq181Dn7EhwqUAar2bgEXKapcy1be3AXUhRTkUDQ6nBoR1dz7noeruPeztm/w18AQ3oVei6xyENWjvS4I1O/1eggfi3SABciNlk3Lv2wVZq6avJFVHNOJCONRW4BrMvhotbv/EAEj4KY8syKtBE5NEeHPtSOlczz2hCLmaXm7Eq+d+D3CAzzXgzxa1F9Ugp8UJqczWRtCI4crL0waViErPljiwiZbN/HSEXoMl3fh++ENWU2Y3egcmpDZtSO89HgsWCIsc7gCw3e1G/eSfFE4Jk7EOunfcCuHPAjBuRkmkvHeNTR6GxZA2lx76sXk536pjtR2NKV2SeAYdJ/cQtZlcdB3MdKtqh5SCs3gardud1Ya3oywp0LzofXHFNlyOhpiW1qw4JeBebcWOhRTk4egjBJhi6aEJc3BytSfwtyEJxARoc5iOt2MuoE88vRjYHuUidjFlWsHEzct/a0MOYlp6wnPIadtL6JZhVHHEyAfe1mH0T+AgS5LLJ5W40EGfxSNvICT0XIU1XPR0TD5RjKXKZK+W6lhX9XI40l0Hfsxrd27PQBCrfGjkKTf7uwv2IahOZMRO9L+XcEjMt80JyMWsTyNWUqkcTrNF0nOBkAnUVucxQgx/zFgw3KijvDpXRTh9aBM0Yi1xv89+pGqQg2k1OoGpGgsIW9B6diGq0rUeC/Q46tn8CuYLKXcXNtVEgXLjzohlfQa5w8+ko8O0AHqT8/WpBVqCu+orV6djlYmAzr4D1cLgY6SygDl4E6o+Bx57qolbqbhh3khkbkcWtGViXF7NzCRJq1uVdVzW5hEF/ZsZn3FnVxfUEI5AQbILhi4QRuR6YfR+lvVyIOvJMqBmPLD67kQboFDThakM+x6cDz2B28wBlS+tOOut861NPsqT9FAl7JyJNVguqNTAZdfyb07IaNAhNR0LIVuBnmHVH0DuPrgWgnWm7EGz6A1kuf4p+h9eRm8AZcmv8EbJaHimZtrkr2uhY9LABTSqyd+85VP/iEB0TWnj6zAN+VqzyeRB0wT70LHVFJd1T1HSXYvGcM+lsPWoj54pVjazvb0F9YjWa6O9G78oY1CffhoSmSsq7e40Hnihc6M5a4CtmTCMXZ1MB/BmyZmQZCYuRuRGXxZ2GFD90AR2FiozKdG0/dueQGa9BwtYUoB3mV8NXz4ZHq+BLq6GlyDhTZfBHC2DlYjpme2s14y6khDsr7/wVSGg8jtwzUQv8uxmfBh4qlq0uGLmEYBOMDNz3YfYNZKK/AE2aHGnMsgrI2+ioPc5cpE5EA9EPDq8xG08KlO5jV6pNSBtezmozBtjV44rx7q2YfQX4EDlBKqtKnbntTUifNnID3iwUTDoXs5/gXq7OwRR0P8vRCMzDUl3qoO+RRe8ezB5Cz/wo9Iyv7/Hz031hulDwfgTVt8jc13YhTfEZ6DlsQRbBPWgy+EuK1ZUKgq7ZhCwc4yjuLgl6Fxrp21Tn2ViQ/+xPoHPsZg0SXMajsWc8EmqeR1bzCUiBsAH4JnC7OzvMeC3wJkpbTgxdVyfBJiMdZx9S2r0JCQH7UntfTZ/mgt1qKV9iIZ/b0Xt8GlJm7E3tmpKu+w7gETPORkVJt+euZ5RB81K4eAXMHQ9/+yQcyhsbKoB/Xg7Ll0PbKjpbxt6MXOvayCkAsxibPXTsj6ai8WwCR1jDKFmZZqRzNgHbIyX98CEEm2DkoIncM8gCU5OWtWJ2DRr8SmmG1wMrMLsbBdy/DpnzNbE3exa4F/e+8NN+CKUoLSfYTANu6uV5fossMAeR//bx5DJbjUETAicXnzEKDUCNaEBsAH5d5vjNSPPYUmabKqA1hJoBQNbGvkqRuoVczaOuMq7lB2+/iFxQJpArnrsNCTAz0eQD4LvA3XlZnYLgiEhpjn8G/AFSyhT27TXoebu+VGryHp632YynkeW/BsXMzEYT/fXkBIhRyEpyFXLHbEl/F5AKZKbtZgGb3NmRTvEE6n+PobPiKCvk+bB7hyQAHTCjJp13ERoD1pKz2CxK7b2fjvViRtHNOj/JEnMzcu0+K12TA08Dj6YMaFMpmiSn0mHhS/DMclgxH67YAjdtya1/41Q46zh49gBcXNiftaZreSM5N9kZSFFYLE6vDY1przXjWfeuMy+mhASnAq9H2SAzt7stZtzmzstdHSMYfLpjyg2C4Yd7axJqJqLOvCt3hENoMLgGdWjrkECwGSUw+BPMFvVBy15F7nPzSqyfhSaLvQt81ET3OiTEzEGT1Szd9RT07u9H1z0WDQy70eC7ATgfszFlzvA40oiVYzpdOFIHQ4+UGvd5NGkoRVZL6rBgk1zKvoUmfHPpmBWpLW37MXd+GkJN0FvceRU9bxNQfzoN9UnZ/zdl9V76GPWPiklpQ0qqLL3xEtTvjUN9eSWyJDQhQWIncptagibqm9HEuzpd0wGUmXILUq7NQe/hychaMRtYZsZbzEq+n69HbllrkTDzErkyB3tSW0/L234mErS6iu05jDvt7qxx53p3PuXOp935sfvh/mAFJZPknLoRRjfApgq47PiO6y5dCJtHw8yNsLjYmO1o3FqQvi+iY22gfAwp85rJ1TgqSRJqLkbZVg8hoWwjmgvUAteYcWZXxwkGnxBsgpHOGKR16U4ay9eizizfJN+OtD67gPdiNrZXrZF7141IO7YADVzT0CA4H3Wi38K9nCWku+fagDKkPUfOIpMFnjahCegk1Pk3IgvLobxtjitz9CfRwFEqlWpNOt7jvb2MYFD4BZowzKLjOGFo4laNJo4d3qukSf48svaNJTfJfAT4vPuRp54OglIkDfqngRtQP0oKcl0AACAASURBVPci8GPg0+59r1QxYzoKXL+XnPvuTPSsZ+5ls9BkOnOXAimMdqR9dqJY0MlIuBlNzpqJO/vc+Rrqu+9Nx61D1pyH0cT+dBQgf15B+2pRrah8S+oO1F9PSsc6gJRbs9EYtBO4oY9drZYhIaoIE1rhrQ9A+zaoWwDjsj5iDkw/Hsa+BJc/WWZ6uhr1QUYulXYh1eg3qEcKu8XdaPNxyFtjLR2tWaTjbESZIKcRDGnCFS0Y6WQBlF1xLOq8SsWVNCJt4InInaznSGi5IRUiXYosRAeAF3Ev6WLQw3PtxOxbSIDah65hJbmUz1nhuDlISMmy0GUF3Eoddxdm1wPvQQNIluXH0jnGADfg3lUcTtAVZuNQdrlzkCa4AcWuPIt7fcG2U5G29AT0W6wBHsd9C0eAO/tSlqWL0vEyKtDk8Vd57jOF+9ajCdm9ZlS6l3yngqDXuNOCFEXdcqXqJWehfrMZ9XeV6J2oR4JDG+rPt6Z1mZW8kY6uZS1ozMlcgTtlbnNnsxnnp30LXaC2oPnbZWbszUttPJviyQfWpnNlaZ3HpM+3gJfzso11wgzrgdDTReKbaU3w/vtg+x6YfD/sr1b7fmcUTFpTft/DsaKZcFOMCSi29kjafS4aI0vtkxUcPoXybtrBIBOCTTDSyVysxlDaZF2FzP1dBRjuQZO83gk2Ge47oPjksE9xb8LsXpRUYSYaeAt9kqvQgHEyGqQzq065476A2RdR5rOT8ta8ANyXLEZBbzCbjdwja9FvthVZ3y4G3oDZtYfvs9n5aXkbOW3pacC56ff/1ZGkD08Cyg/N+DW5ScQu91Ka2KLHKCvUmFERGYuCgSYFhy9AmTFHo77vOffSCVFS7ZTT0TiyEr2To9F7sR9ZScYiZdViZP3PXHu30nGy3kgu45dRZGxKcSonUzrt+iFkhX+jGS8k4aPcnK6enAA4E7jXvbgwaMYoZHU5HzjGjIPkxdCUOUfGenRvC5MU5FFRBzM2wNpfZoJTKn6aWYmPRWNV5sa6Ad2niWiszrwF8sf2WvQbrCVXt2cSdCgkXOx6q9Bv1tWYtQspN0OwGcKEYBOMbFTv5i6UHWUNxbUx05AWvKtCgYfoXE15uHAncn94A7niiVmqzKzWyFp0fUtRB991gVBZmH6A2c/I6g24lxnMgm4jS801yF0lXwBuQr+P1pt9Dv22l6PfMD9YuhH9xheg371owbtypCrefVbJO03YTgPOBGrNOICCmZ+K2JugvzFjCvBeFKDfjN6XZcAbUmKAH5ewYIxC79JpSChqRpaXrNDsDPS+ZfVoHkV96S6KjztOLr6xmKV+KaU9CDIakLvnMUh4qqd7HgolC+OaMR4VX56R2r4eKbpOAs4w4+fuXWYzfAwJgeUSkEwDflpgDXoU+BSam7agvq4CJb9ZjKwwlcDNyGq1O7W1Jp1rH3LXy2JKDVnMukp7n5+GvhztxLx5yBM/UHA08CTSAp1PrjI0qDM8Bml7nkCdfblaNmMoXxF66KJECi+goPDjkOtZExLWNqH70pa+LwRuw737k0wJMyHQ9C3LkbBYyqq3H00O3gdcigbnpWiS9Cq5hBntSBC6ELOHcB+wujEpKHoRmhSNRu/YUjQB205OWbASBVFf253sRUHQE8wYB3wQzX2KWUJOBqrM+H4R96tWpP0/lc6phQ+hSfR49JzvR8/1b9GkvDAL5hjkmjYN+HYJV69iaaSLkbm7gYSbLakd9SW2r0J9fafilSmA/kp0nfn3py0dO3N/2+letrDuBnLCzXpy92oyEsTmp/bVmTE+WYdB/cQhJLwcIFfEd2869+tRjOor7rSb8W1yNek20dHLoCqd6wG6sMS402rGXnJug6UYT/nCxcEQIASbYOSjQoY/R0GHr0U+xiCN0J1IS3QmChwsN6kaQ88KHg4VJqEJ72Ootsg8NLhkHXlmxt/NkbjbmdUhIbFlICfNRwHnUt6KmLm9nI80t9vT8mnoGX+GXA2PQ6i/X8TAxCKQsja9D7mONKBn5PVosrY+r70t6L0bD1xtxmdTdqgg6GtOQ89ZqTox65Cr0TwKJrDuHEyplPPjRzJhZiy5eI8x5IrX3oWE+QXIMpRZICan49/s3tFNKlk0F6PMaQsoX6ogozm10c24FaXBbqWzsqkyXdvPSxTGnVfs2vM4hMaH10NpwSa146fp/Oek8y5CLnCZouU55EFwoRn/P3vvHSTXdaV5/k75AlBAoeC9I0ACBL0naCSKImVbo9a0pJY3rZ7dMb290zuzsTu7E7uzETs909tW7WTYlKEoipShyCYpUvTeG4AkQAIgPAgPlPd194/vXtyXme9lZgFVhSogv4gMoNI8e9/x5zt3ogDO+ajMaw0KsIVrHTI/m4D+UL7qqad/iub1XE5uz80A8CCili+n1+ZJNHenmOPSxEiVolcwaqg4NhWcPjBrQT0wFxN7Ep4FNnn6443ARm+Iq4ckDDE0exk1hk4nnc1lITISJ2bGRuhFz/wQctB2oOxNoG0+htL4tRSPWglmi5FRvZqQ9tfMn2dwrpw67AqKo4n0EhWQwr8CGVAh0xbQge7f+chxDRmfQWSEjTrMmE40roKhsNYfTysycC4mN1DQhoyqtWgdVlDBiMGMaiSvSs0j60KBnxwD12d7BpD8DKVLgSEwZFZCb2IVev5eReVT21GWYhbKxNwHfNe5OIbAjEZUTnoecRbOOehZ2Y4cgfzStEnoeTp+Ts6x1YzbgN9D8j4MZZ7ij+sh4OmMc19L8dlk+P0tNqMlefz5cI4B4H4zngb+FXLsnkUBjVDaesSf5xdRwKXfn/ubKKM0EwVE+pEc6/X7numcgj6+bPBes+Pl1mGo5lbn5NiZMQVlddag+7ULeM25nLWwAbgSXe80spXFqOJhItsAZwQqjk0FpwfMVgOf938dRmnsKcDngP2Y/fB4aZVzhUa7c62Y3YJm2SxBxmEfUhwNSMjeOZzm63GITUixgxTdPv+yxHsN6JyLs5mZXYj6ljpRtD3UM68CLsDsJzhXofY9ObSj+5EWWZ2FHJ9WYs9UEkPIQFtFdGxKE0KMHK5A+uVA4r2l5Bo081HkOmkcHUGR14pjU8FIoxEZ0aXKu9qRcZv2+0NIjq4gllb2IYchBI3C87YFVQEEpsmjSN4+AjyfJM3wJZtfpDBbMhM968uQwf4ysQ+kCvXB3JlPwOEcb5mxDRnyq/x3dwHrS5B/TKO0Y4M/zyyq/3zU+e0+SXoPSy+SY9eSWzHRC6lEBUPE0rvjcI4OlKXOgRmrkR1QTWQ+vRy4xowXUfZqwA9f/QHSa8v991vQPTDERHlvhexk/KPi2FQw8SHmqC8hYzyZem/zrznAlzD7Rz9HJh3O7cfsb1DK/AJUUrAFKaZdODeSPP+nAjvQNZpBbs138rzmAfcUvU5mc9EQs73kGgkOGbINwOcx+2ucy68tr6B8PI9KLNLqwxehtT4FRRFXUkixGijKG4jGyvZROtbj8OU6V5Dr1FT7V/L4+pEhl3RselDpWgUVnDA8y9UKlPGoRQ7FFrLpgZOoJt247yWSAjShtRuomgeRcd6JjO524GEUJHsYrfFuYEcGMcFqZExvz3v/VVTKNR05N9uRczUTPfuPER2nHDhHF3KEhjPPpxXJklKoovwgybmUniV3FGW42ilN4lNFeb1HmLEUlcPuI9c26EBr4Qokh+4HEaWY8U8oc/Y5YhntPuSc/bEZd6XN4/KkC+eigM0ACoZudi6HzKWCMUDFsangdMA6JGR7kdHeiIToUeL8gCVIcRQ0TObAuX5CydrpBjHE/Qz4IxTp20MkUpiGIntbgW7MmnAuiwnrUqQMspRLmPFwEVLqFZwY1qNa9iYKWcnqiXMydiBlv4rCJuVAATsDWI9zmaUjI4hQ6pksjwvlctXEcppetO6SaIQKM1oFJw4zFqL5WtOQXhhERvNHULAqq9w4YDopfRR+ttM25FRsQzNjjhHlYBhYXIt6awaRnOx3rjCTkIfrKXx28dt+GgWcLkIlcm/7/baisrVlZrwMvO2dmZPBBsgd+pmCZuSglUv7PpPyiGUOkJKJQVmTpX6/1cjJKNepuomY1c5HkJ1Xm/Gsc8d14XUoUDqJmB1aju7FFuBLZtzinPoXPeHCdcCN/veBifJyoNOM250b/YBSBRHl0AJWUMH4hVkjEu51wEdROvsSZHx/GCmCehRJu+QUHeWph1k1ZlcD30CKdybwcRSZuhk1cYZG2N8D/iNmn8Isjd76EkrP3zmEeigqOFFo+OYPkWJdSKQar0cG1GRUs96NBvi9j+5r8p7VIqNoL3DvmBy3HJg03fIectICqqAgmjkdlXxUUMGwYcZsxHo2hIzWA8hh2O1fk5GDkJW5aUAOyZsZnz+BDOxd/tWDns9J6LnrQM7Bdv/9IWSMFzvmalRVkBVIGvTH/oA/rgPIaQsDliejpvc/9k7dySDQ/M/P+LwWPaPDoY3v9L8rhQ50z+b4v6uR/r4OBeLM77sO+PdmrC22MU/CsJjiTmygeF7tf9ME/Am6pkfyXr2oB2kq8DHv0IBsjo8iGbsLrbeD/lx6gG+aZV7PCkYBFcemgomOs1DfyDeQEDwXCcE+JIzmoAjUEIXR4TMDZlXAp4FPIuWxBSnoO5AQXohKHl5ECnkHyuZcBvw+ZrWJbRlSLPmTrfMxQPk12BVkwbmdwN+gzNckpKinAL9CTckhAzOISmRe9v9vQRHlA8BdwA9Se8tG5ZDpQKxT+SVlO5ARERyvRnJr6ltQ6ejply2tYKzwAbTG0qiOh5CcA5WoTcn7fAaxZyWVKtk5tgC/RrqkGjkBodF9L5KbTxPLrmooXVoVjOtSZXK1iBxnMnqWjiFjux09b33A181OvJTTs4fdgaoclqLzrCIGSOYhJrctw9jsJko7NoEE4Xvo3i1FGY8FSEcNIYfiLeAVdE2/4EvNsjAFyuqH6SEOS/0kWgeHiPekgejwHkElgUuBeWZMRpmanaTPHApELh8u4zgqGCFUStEqmLhQBuJ/RMK+i2hsL0QRp81IAUxHZTqPjOC+A4XnZALzk3PjlaJ2Dcpg5Q8onUIsq7iA3B6lISSsz0bO4utAoM4+Smm+/0mkl1ZUMFyI9OJJ4EnMzN+DKnT/FhEbbAON6i5kBMwCvo1zp2IuzJNouGgb0bjoQtmYK5HTE5iOphCdmh9k0NBWUEFReOar8yhO2T+EaIq3oTW4iDioeAvwE+cyqaABcI7nzdgN/B+oj6cPZSVeQ85NyELWI8ejqBPg57G8i/RWMZm5Cj1DB/3xhnLPINPb0HN/MfBosX2WOJ4OM27x+7sGBQf7iWXcHzDjRiR3ngPec67oINGdxGxympNnfh+/9OV+30VZmv+Arl8LKj97BZ17HXI0hpDD8L2M/WZljvNRC3SbUeXPNww5nYlY70LGzfn99yPZOhWtnxZiqW2YB5fEYWBlKRa5CkYOFcemglMLs5lIEK9F63Evqm9+rygDmdly4CtIsXQQB3mBlEw1Esx9yGA/C/jLETheQ5GkG4k1uAYMYfY88LCnlh4f0PF+gBiBSmIJvgbc/72QQiV8ELgeszcS5AlPochWMQOgBWUKKhhJhHugfqk7EJPSEhS17SD20wyhLM2pGnb5DuozCL0DIQJ+FPUOLUSGUZP/7G5GpkeggjMXUylvenwX0hE/QAZqLdCZ6LEoCefYbcb/C/whMmY78r5Sh7INd2WQBeTjOVRCdyTj+KtRpuBdpC8XIr0zgLJGO/x5HUA9I4+VObslFf6Y3/QvzLgKlS4PEDMoC1ClxNtm3OlcOpuad9x+6r+7yB9j+G4z0hXPIscF5+g2ow94Bjk0jtjrciFxDp0B68zYAzyccp33ofscHMws1KLrOgVlkUF2xWRi727Y32yizr/AX5PFRDKCfpShyg8iOpT9qjg2Y4CKY1PBqYPZZag2OAjLfiQsPgv0YvY6Eqwv4Vz+PI9rkWAFpbCbiUO8IEZQFiGB3wfMx2yJ/2wrcp5KlVTl4wYUJdpNbp9JNRqoOMtTHQ93u6OFSSgalsasNYvYhNmNygzyHZsOdA3ridmcDej6Z0XgZiPlVSkpKgXNXjqPyKSzEXi3qHOsvrJz/Kvb/2Yautd9wG+BN47Tm58C+AF9D6J190Gk/INBsB74u7wZEhVUcLII66sUqlFDf2BxPCE4x07PoPV5FEzoRvqnEemYXzgnY70MbEWG/DqU3Ug2u09CJXLdqBekG5WhOeLgy+XIOThKJDAYkQCbGWtQL+ZOckuQD/vX2f7zO7O24RxHzPhH1J+5DukIkB69F9iU54i1oGsQgpVNSOdUIX0fvlsNfAyYa8ZtSefGD+98EvgE2bNnZqFg6k6/j2PEnt38CgyHbIkpKGh6GQoohdlcIJv6AlQl8jaRSCcZeK1glFFxbCo4NTBbBfwuEmz9SEBdjJyVQGO7HK3RKzD7Lc495n87CTWAhubjfhQJaUECJGQgHHJ45iCh8jFiGv8aoBWzO3wfQznHPAf4EIqOBSFVhYTjUhThuRjowewenBsPJTXl9tEVa3LNNRac68LsVkSjGWb+9CLnZwpSzLfhXDnzEM5MqJTsw6jkYhAp0SqkFLsxux3nCpWx2dnIkKpDSnWIOHjvUeDx8TJryRsqb5nxNnoOQ2R8vJZsVjCxcQg9Rw0UZ+GaxAgFXZzjPTP+jOhcVCGdtmk4JZU+EHA/0k8fJGYGqpCx/SjSee3kMoIN+s8bUInnYxQO7M2Bp8JuQXK9NQyxzPiuoeqEg0W2uRO4wIxHw9DMjHPsAJ7wAzsbgaEiGdouYl+OISdikEKChSHUF7oCOT6PJM7xLGRPTEEBybeJg0xrkbPYDtzhr38HctTqKe6ETEL3ZZ//7lL/fhW6rrP8vs9C92YbWpeVQM4YoeLYVHCqcCNyRoITspbYKAgSBPNQs+AB4GbMjuLc68ioq/Pf60DCpNNva7J/hexNL7GB8x1yhXMT8E3MvpOSEUrDJSgKFoReA+LBb/b76fPb/wawHLMfn8JSoIAupAjT0vEHkGPSihRNmoPXBOzHuVzl59xhzP4WCe9LkPLYi5pzTyQTdqbhBlQiuIPcBtfD6Fp+A7N/yFmXZkuBr6LSyqRxcww5pTeh9ffk6B328OEdnHKpYSuo4ISQiNB/kuwIfSh9HE7ze6n9jsiIAD/48QUzXkG6sB4953uAzyADfgHpVMc9SA+eC9yfNkTSjAZURn0tkvfNwHwzulBm9W1iz0z4/RzkZJUK/g2ibNJTZZ5uDeDMqMoYeLmROEy6BZUZ5vcf1SH9Fpy9dWY8hbJnX0JZ7C7kaNYS+2feQbrwCeDFQBThS+Y2E1ncGokZF5DjEqjsd/vj2kEcBj7f/6YX2SV1yCa5ChHzVMi6xggVx+ZMg9ls9BD2Ae8XHcQ4uscwnygsG1HUI19wDSLnZjOKdtyI2XokxPqR83KMGNnq93+HWukGJJQD/3y+sd3rP//PmD2NFMd6T7ObhrOIBloVipAFWsiAVmLm6OuY/S3OnTqjzrlBf243UaicdqBIlyFZkKa8ZgG3Z2x7ANUTbxqpwz0jYDYV9Z7sJJ21pwOt3Q8BP068/xG0ttMMm0G/vRsxe3msGNAqqGCc4UUk09aQW9JlSJbVALc4V5LV8ZTBH9uO8LcZtWgOz9soy5DVM9KLem9eyP/AOzVfRSWh+/33lvrfTEfypg9dt1fMuNsfR5gJVwq9ZAzW9U35TX4f56CS7TCvpt07o6/m9cjsRIGyWajkOe1+TUV9OaFKo9Zv/1P+eJL67CgKuq1EtsBtGYMzNyE7pAo5JjOJFQtDyIlp99sLc8ReQuXzoZQtfHea/+5W/7uPoF7CCkYZFcfmTIHZWci4XUCsRe7A7HHgxTEuX2kk16CbQ7rwHECOAyjyshhYgHO7vLF+PhIwu5CQDsPYAib7/RykMEK3kDjPoBEpwVUoM/QbnHsm5XiSJVmzyW4GDHSjcxAb2W9TvjOWeAWl8kPvS0CIXl2BMmPJ5tlqdI3egsIpy2MJP8m+EdXFnw4G+7n+32JBBdWumzXj3DFfBrmIhMGTgsACdDYZ08grqOB0hs/a/BTJtOuQYRxoe98CHnOO/afq+PzMmtVItvagyP+2EqxioSSrG2VU1iF5GBgHA0NiA7A+o3ftw36fO5Dxv4Tc/sgmRE7wDCqnPoYo5sstJ64jj0DBl4NdiII4s1CZbQPwBtLHff48fgdYa8aPAwFBHuHAXHKzHY1It2+hsHf0MqS7soggNqMAZeitycdh1EMaGPP2Eq9/F1pDgego6KI+YsAp6dzVIt25Fa3BS814xLnMeUUVjBAqjs2ZALMLgc8hIzxpGDWg6MZ8zO4eQ+emj0JBlbbvGnJrpUNjJiiN/EmU0dnnf78QnZPz/waHJV9Az0UC8CjRuBzw360BPolZH869lHc8W1HZVTdypNJqk8Mw0AHkRFyF2aPHM2PqrViKFO9Cv/+3gFdxrtTQyxODc52Y3YLWwBJ/bP2JY/05UhSL/S/MH9cziOXtlEQ3/YC1K5FzWI043rYAT4WpzxMUsyk9iTswAU1FSrOJ8mYy9KEoYwUVnJHwmYZnzHgBGbDVQFvWbJqxgM9a3AB8i9zhl0dRH9qPncsMIPUieV2DdPgjSIeEnh5QduIIKSV2ftbKZchIr0FZi/wqgnZUwtWEnK1rzY6zkrVSmt6/hkTm3meZfh85cfv9tgf89lYiHRyGC29HeukmEkOEE4QDnwe+EDaNnI/1FPas1KCgTlYZYkAfcrjSHJs9frsHUTncDLR+ev17g8jhSQYJZ6D7s8//PpSxHUFsa8cvC9Kxb5U4vgpOEhXH5nSHWTNq0t9DIUtKDxIql6PI/Vg9cPuRsRaEZS/pjevV5AovIzgozh3A7P8B/isS8AdRNKYZCedWJOD2kGtEGmI9aSM3Yh6MxgEk2D/iKY6T1+wVVC8bImRpqewmNCQRYnq8HujCrA7VSp+HHIpWv62rgGsxuxfnCsoIRgSK+n8XZexWEefMbPSf1aKIXWDW2n4qy5nMWIyidQ6tl3Cv5gJ/YMYDzo2vXpJhoI8S08g9qojnXW7JaDUjxIZUQQUTGd7Bef9UH4d3ar6I5FkX0rkBk1GwbLoZf+ccG/J/7xyD3km7ChnjXSgTsBE970P+tYSUMjQk80OwKmQ/0oIkDhnpbX67y5zjbTMeBX6Pwn7AgHnAZudyrvUHiE7GJHL7Z4+ggM1FKAOFP6/LPAHBcXIRP1Pnp/7cAnNqmt5tQfd6FqVlZYc/5gJ4EoH70L3aS+FMpJCpeZ9cAqGAAWKlybu5P8VRelBpBSOAimNz+uN8/2+WseOQwLiOsXJsNIMjKSzTKDenolR5sl+mkziMEJzbjNk3gU+jdPZ0otD/Z+QU/X7edqcjQRtKyKoonFTd5/e3gmRDqHPvY/YkYq0ZJNc4NeRU7SdGgiaj6M6nMetBwn0+mjuQRIik/QvM2nFudEq/NANlN2kD7Jzrp1AQnxKY0YgY19opZME5ghzCj5qx27mCazkRsJnYGJuF0Dgcymb2IoVZQxHGI//5RLwmFVRQNsyYi4zZIWDfcGbQ5G2nAR/M8axdo4Hz0FiDDgpLpDqRvlkEfM6MzRksZS+hLH8TUSY6oiyYi3TYUTPq8vpVknZelmEdgnUL/TYbkQ4ElbXORiVl7checMQBz3tQ1h84fk2vJurqGRSWmodS7XA+g/4YFpNHwuAcnWb8Fvgo6X2ggTDo18iBLIWiwR/n2GzGT4B/ia5XO/F8QX2Pi1BJ8S5yq0EM2Rh7KLRrjEJ9VsEooOLYnP44l+x604BWYBFmjWNIUZwUlm3IcJuLHJOQcQmZD0MRlrsLyqI0q+MHmP0IOUMSHs4N+LKvI8jhSDpISSE7DRmC+VGg0PyXj9/6Y/wSuraBv94hofum/+2lqKb4KMoo1SGH6KA/lvzenAHkyH0Ys42JYZhnIlYjxVpQmldPj13D08319DbU0fcZ7He/XcDYNv6xDRkHyXWZj3nAA8fXu3O9mD2D1lB+n80kdL2mk+W4VlDBaQAzFqChiEtIzK0x4y3ggXIdHDPmoF6Vi5DsrjJjO/CkcyMX4PF0yTchnZY1M6ffH8M8lOV4I/8LvizrByjgE+RGHzK2z0ZO3ibgXwMDZjwLPO+dtTZif2haUGQ20r2hn3UG0qW7zdjkHF1+JtW7yGE5x3/vEBrC/HbecM6FyCEI+6ohvYd2CDlGSWM/y/F6CjkkNxDnyQSWsm7gViRX96Nr7ZCTNMf/fj/Szx1Irz+UsR8AnONNM95Dumglun47gDedo833fbajapcaJH9DtmgbChInz7nB77tYj2QFI4SKY3P6I6Sqy/3u2MA5h9mDqCb4KiRs5iFBtwEJ0UGikH0GRa2ytjdEvpEYp7P/ARLeB4nKsIqoIN5J2WIVaVEd7ecpzN4E/i+/vTZkqPYg4XYjcSBhEH69/v/9SKE+SXSKAtqRwp7LOCihOIW4mMJrw1f4wcJP8etzp9DR4DDaabq0h/qpDWZPoPkt45bpKAdiqrsdrct5yOAJZQ2N6P4H6lVBZYwH/WdXIKNiP3JmZiEFH3qnPoDZM0WHfFZQwQSDGYvQM9NNroFoqLx2gRnfL+XcmLEM+Bp65vYQ9eMM4OtmPOgcj4/QYU9F0f0+irOL9SAduJgUxwaODwT9SxRQuxQZ8Cv9th8jDpSsRRUYF5hxCzrHg/77h4iEA0NI/oThnxCzLINIrnzZO1QOyZhk+firiBo6n2Cghlybo5t0quO00qzUjIanjH/MjFeJoyEGke5+N2SozHgM+E/omoeh3eF8jqLA417KYPL0RDWv+Ff+Z33AvWY8gSo7tqF+0A0UDvasQZUad45nNr7TCRXH5vTHblSOVqxfot5/PrY9FXJu2lH2opvcUrjFyEh7FzUZvndCWQzn9mL2BEmDzAAAIABJREFUj8jZWIPWfGjE3uK3n1azC8W4+507itmfo1pckFE5DzktC4nK9wgS6meh63yESEuZVg89SCRIOFPRQN49+SP+asXvcM/5h5jZupcFXQDHaHZdNB5poFcD7cx+dkroy08Ezu3D7B/QTImL/btiKoR7gJeOO2pmLcBXkKGxGTl9F6KAQD/KOL6ByC0G0VpfgtlPfIlhBRWcFMyYhbIba1EAbA+SX9sy5pBkbSdk35uQzNxTbEBk4ndViFI3THpPwqFA0DyUHbmzyHYmo2z7MQoN0KN+2zebsWuECEpCqXMpOLIzG/FLMrZfAl4y46PIccrXU/1I788BPuMct/i+kT9AOn4/CqAd89/pRNmaQD8/AzkM2/33rkHO1BziOIVaxLR2gxm3OZdDWtBBriNzkDiEOymfq4n3oBHd26LzcpyjFQU5szADObnN6F4Gp6sWrY/ZwH8qZ82VA09I8ZoZr6MSuo+gYFMIzE1D1+LXzvHqSOyzgtKoODanPwLVbxhYmYY5wG/GfGK52cWI2KCfONl4MsraHAZuzZkBYzYJOQRrkaDaA7yGc8XpO507ANzuiRSakUOznNwmziTmAu/gXOYUZb/dnZh9G0XQr0ZlQmGo17tERzFkk5b5czvg99FI4UyS9EzRmYVD6P50Aaxlw5RPcN/a95l3ZIDaIYBBqquqGHKT6O5BDuR5KBq3/lQd9LDh3GHgbsweQgbKEHA4xzkzq0dOzWRilLoVXZ/XiFnWrcS1tAMp9yuAp0f3JCpIhdlkJKO6xjJz5vsbpiNZfySv1+JEt3k56mEcRHK5D7FyrUWMXneVsx8zVgIfI1IwAwya8RyiYS62jaXEgYhZ2Aec7zMuBRlfjzWoLDjfqQkYRAbxOhgRxybMPQlVAlk6NjBUbi9no74P8UrSmb0C9gPLzViIgm2NRPKbZiQjQkYjkOzMRNc4VDEcBv4tkjXJaz+IHJipwL8z408T5AF7iYOGO/x3N6LS7CPEjNEg0vu1SB/eUYL2uih8edi/Rvr1fWTXTEp8JbCnftyMezPm2JwQfEbpKV8SeSHS9aBy+jdOtAesghNDxbE5nWBmxAe52zsqu5FQuhgJpnznZg4SQmMbTTBbgprz9pJryHf612zgy5j9ve+XWQn8IUrpOmTEdSA2sReB+0qWIjmnaJPZT9CwsiVI2OVHdQ5R7iAt5w5j9jZSGMF5DAxj+4mzAtqQAzfbvxeoq5OOTYhancllaCBlcB7+2v0O9ywapGooODUAnUyeuoxt7zXQGxThIeB6zDZMuP4ksc9lZUvPQYZg0qiYjdZY6NOajkovkr0B76Nn4/kJU6J3OsDsbFQGFHpABjF7CXge59JmXo3QbmlCkfUriNHy0Gvx7InOfzLjbETOspvcLOoh/1qDaPd/UWI75yHK3kPkRuVrUNZynhk/KWJshqb2YnD+NYeUUlaPiyjdc3oEWGVGw8lG9p2jz9MmryBSt6ehAZVHldvfswDd51LXxFCmpg5le99FOm4OWiuT0D3pRRmrV4j6CWKgMUkPvQj1noR+1Sbgv5nxt8BLfgbNb1BAps+/3iNWKgTd9w6SZQbc7Vx6Cd4wcCG6LoHu+ag/73CdBlAgaAVy6kacLMk5jgCPesdztX8tNmM3cnBGTQZUEFFxbE4HiKr3AqRQp/t3j3kGrzeQkd6LBNkQEjS16P7vAH42JtS+Os4VKDpzoz+GLEV2ABkHyzFzwJ8jQdqDBNRU5OQc8e8PAPeVdRzOdWP2A9T7sg4ZjlVIiK9HvQ1ZEb38c1qDShsWIyepmZhyX4QUx05/3MeQkqjzv05Gpwxdl19lllMpeh8mN8/w23wZeCMnszXx8R6wDYYWncXWzit4/mKgsZlj03uoP9rKtH7DDV3Ky8l5Be3oejdQmAWbGFBgYjp6LjtxrhM5zPn3di65wYA29FwljaJeZDTMgFM3kPCMgtmHUXPzUaLxXoPkzCWY3YJzxSLsJ7hbpiHjtRk5tMHYrUO0u+eYcWuSRrfM7YbG90MUL9e9yIzHnTtO55u/nSmI5n4vWpchgLQAXZ82VFXwnhlPZ5S2Wcp7aSgV1Ai6otQ2QmnYCcGMehScuRbpqXmoLCnMVktiHnLE/mkYGbZyj2223+cTifd2+lfomTHUTJ92XZaQ2x90NnJmW4lyqQ3p0M8AM8z4jXNsNOMuxAZn/vt7kc6bj5ysN/2/60coo7GKwvuffz0Dm+kKRokF1gcDPo/WechYrQQ+ZMbjwKPDKd+sYPioODYTHTJ2v4h6OA4SJ/FOQmVeFwC34dy9mD2NIggtyPh7B9gzJhFuOQCfJmaUrkHOwxJknKcJtg7UJPlh/7u0TMY0QtTP7NmiBr4asNVk7VwH8CRmzyHlvQ45CsuR0GvH7GE0ODP9+phNQXXf+5BgD9OgZ/ltdRDnAhxDymQqcm5CiQLofjSjsqGXSYPK6L5GjJ4dRoLzBtQsfjvOjQu65pOFcwx93W79RS91/62HhjV91M4E199HbUM/dXOm0dZ3Jc/fPY99ac54uQbQ+IEcmrXICJ1DILgw20Ccfp1EfknLIDJi82vYHWNJCHImw+wc9Czmz/oYQPdvGspA/+UolKZ9FhlOVcjpPYxKi3qRzFmAav+LZlVSMButx2J9D8EJWIOYq9IQJrX3Inl4pf+7G63bs/371wG/NJMcdC7Hod9HefZKNYWMk0kcQNeqGemCUF6Xz8zVT+khuqnwjtxX0XUP89X2oWDecqQfgvw2pIf/zLnSDe0JhPK2ooeCKgfSdYoCHmHdzCE5SiFiOtJffeiarSbSPQcoM6l1fh06n/ec41U/UPk8FJCrQs31rxHZQpcAV5ixhwQJwAmiZI/SaMOMpeje7yc3wBZm1n0IXavHxvzgziBUHJuJj5uR8Nqe934XSskuQUrtbm/0PzumRwfBqfkycaLvJCSYjyJn4FrSWcL6kFJcRfZsjlbkGExHDY6FPQVm01Hk+zK05g2znUgRn40yWfkDTBvRnJ1pwKMZ+w6NtD1Imc7z5xSikkE5zkaOzRAyEg7581/ot7MV+CWwOdWJMqtBaf0p5BoZg6hERHNfzP4O5/KnMU88mNmt8NEBqg88woe2tzP5uiY65lfh2tfy5s7lbOttoPd8pGDD7IlJaC1MLOpnOTUfRQbBQeL9DUxPl6B1syvxq1ZkNAXHLjg6SYO6KvHdCkYfHyD2D6ShFWV1VzJCkWIz5qO18z8QG98DbfBaFBF/DwWELjTjt77ZuVxMojxDsQcZvVlY5Y9tKso2d/u/pyP9FHotWlBQCeAqn2UK634rCgal9SUGzEKDIlN7I30Gqg8FytqJvR6GHJ7X/LZnA08XY7Ayowbdy6uRMzkAvO638Ql/LNsTP2lHlROzULBuM2I+fBXYcAIR/L1EprOs2Sgh6JflmO5G+q/fn0OaY1NP7DVaSqSmTkM/ksdXEvV1IJrYheYN9Xnj/xtoPYRsUD3Qa8av0oaUlonN6F4W6ye2xHdHAzcjXZ+2RoP+/6AZLw43g1pB+ag4NhMZyhhcRrpACtiDyiAexbnhKLWTg9liVIJxAcrOHIHjk6AHiAImUEGupZDtpB4ZA6VKizqRoTcr5TjmoTKNGhRFCcqqGfhjJPwfo1AQBlazD2G2KaOEZC3RcNyDHKsqpNDOIjo9k/z+A1vKw8BPCQZp6R6Is1BELatxthsplauRgzQxISO/CTl8a2sY3H6zxg08gogZjhDvUyhrDHXZs4BfjzkBxsljNXJqtpG7Bh0ytt5Bz8/PieVAYa0FBd5EYf/cHOB1X85WwWjCbCpas0UZnZDRdxHDcWwU1FiBMncAe1bxTudmVn0csV1ehe5/LQp87EZOThWSvYNobZg/xuEM/+2jvAxoLcVLdwMz2Ep/PL3IQVmCnPOwbkMQ6A1kBH7FjL90ji7nGDDjlyjAc4DCnrTpSN4+UOQ41iGdtIvCfpfp6Dl7Az1nmaMFzJiEqiSW+W0c9fu+CjmaTaQzXoZn+nF0TX54oiVJzuHMuB85CH1QQLlcjdZM2oy2gA60Fi8jZhTC8VQjnbqeSHpzLpLBvXn7m0LU64eA1d7pvhE5tWHEQr8Zm4n9k/n6rB74gmdZOxHn/3Xk8LVAelmkP4+tjMIwajNmI3ulGMFFGER6Dik00hWMDCqOzcTGEoozrUCMSi2BE46EpMOsAUUHq4E2zz4GZtej8q5u9CCHfy9BiuA5JAhnIOHaiQzT/OjTZOJk5mLoRYZcrnBXT8+XkcLMj1wfQ/XWc8meGzPo930J6ewzyRlBPchoOA8puneRkm5Bgv8c/52fAXcNs6H7EkpPLD4AXIDZfTiXr+TGN8yq0XW7HimelWg91CMnsRUpo7OIUfF2pETe9t/dTcb8h3GO6yks7UhiCzJQlxOZirpRo/EaFPmuJjc63IyuUbKuvoLRQy3lZTb60ZouD2ZnIYKVJrwMXM95LWvZcC7wxmZWtSADsQoZ+nXo2dmM1sgxFHwJA1uHW5b4Pnr2imVJQOdfrIxqJ37ODLEvYyaSr+G6BT3WiwzyN9E6XoMvpXKOTWb8EJU0h2GIgW1sH/Aj59L7yXwf0s3+WHajrMlsYnP7ANIDrcD/WaLn4/eQk7g98V4oOWxBWZD3SBkw7NFFHB55wkQxzrHZjNtRb0voVQqBjmrUc3o1hSWqSRxF130tkjE96Brt89trIfZzVqP7Nssf937/Xj0cp3t2KFj4r9C93EW8x3XIMW0nNwMd0Ov3+ykz3hnuzBfn6DTjR4gZbSaxhC7se5o/178/Gfa1IphK9nVOoo+0IGwFI4aKYzOxUUf5HPlxEJZZI3GOwLFhR7nV1/MBYq00QBVm25FA/zCKWoSmuRCl60XC5VJkkM4nkgEMIUcmGPBB6HcihVOMaCAot3zlutLvLy2CYkj4HUSKKEvBHEZR9XtTPlM2LJZDbfHnei4S+IGkIQxXfB0Z53+C2T04V66j2UzpEqvgwNZTGL0bv5BT8xnE2ncAKbzFyMA4GznkTyNDpw8ZSMEIakbK+CXgngno0E1Bz0Cakg/oReWja9CzcQA9L1uQ0XEWepbqUZChDq3ZH3k66QpGEmZVSG7UAB0+C96Jn15P8SDTZMiZ91FsP2GI5GESPSP/nf+wqoqhrpt48OIe6tnFkuRU+T50/wND3oA/zmBEDStj79mtHkPOVX5GMWAesMW5orTDbyDmtHCchtZuMuvSQGSonOLfO4Z0xfEeEed4x4z/j9irEno7dnnK3Syc548/GMvP+WNY7PfXi54jR5EeHTMWIBmUFZWvRs/pOWQ7NqB1Ur6TmwHn2GDGe8gxWen3/xJi4Drq6b+vQPqtl9x7eA6SK9XA7f54pvnXZFRp8C6SOeuIRnsvklsN/lxfJTqsgaFxO1EvBjT6855CIdlJQCgFXEH64OxS+I3fz8fR2oRcSusfO8eLJ7DdcpCsRCmGaiaSjp6AqDg2ExudlPcgGdCF2UyUbr8ECTgDjmD2OGLWKu3gyKn5Mqq33Usuk0oLUoKvEIVgSEMHtCLDwCEBfKn/TliLTX47B4DbUMRpGYqQZc2VmY2ilNvz3j+fQuF6/Ez8vz1+f1lRyWCwpOFVcmmeQQbALnR9rvL7f4JcOu0G4AuY3YFz5WQZOpBzV4y5LmncTCRcicpzkj1UYU0dRevhMnQN30HXNxA1zAD+BueKpf7HM8qVv4dQD9pGVEoTmmTvRVnYFmL0eQuwbQKW5I1vyKG5GAV0QkasCrN3UA/ea8h4LtbjVk85tPoqyfwUeUMkn+PKafuZ27yAPUfamFp/OS8t3s2ivY6qAaJT1YcM0iDPHHpOdlDcgc7CKyjIdA1yjEI56CRi5P7nxTbgHIf9hPbL/Dnl66x6f9yH/f9DAKsPGdj52xtARvFwyomWkKsLnN9fvvO/EBn2BzK2cx7ZATaQjB5Az+QksmV2VZHPhgXfq/ECifI3M+rNuApRIF+Bzvcoul8H0Tme699/DTko7UjWhGDTQSRz2lG59mHUDxtKC+uB+8l1BFf436Xp3dCz1eq/FwYKJ2H+2D7h5x7tAzaW24/iHINm/MJfi8tR4GcIBcZecS7zvo4E9qK1USwIC5Lf5QU4KjghVBybiY1tSDgWi9LX+c96UIrWiLSLoOjJZxGt8q/KMIiuInu4Zb/f/hqkRAeRIst3DAZRWcJbSMgtQRGnBiQk7wI24VwPZi8jpWr+N+2Jc61CQrAWGbj5x16M2nPI77vB/z+rTKOZyIufC+f2YvYKcs6SPQ5BsfUgwZxfL9yD7sGnfP9OqejNy2gGRDFK55nAmzg3cZrnzWYBn0PXYw5SqoEQYS6ROa7Fv44gY2cPuu/bglNjRi0ySsLwvb0TgFKzC62/+ejZOUr6ep2CAg8P+GGeMgQr82nGBnJqPoUMxH3kEjwsQs379yIjMgwlzMcilFHenfJZPhai5zmnZ2czK5uCiOmjrqeRrpqF7J60i8V7UCAlZI7wx9GPdPxU4L4TeR4SvRyh1+ssv49jqBl+Q5mzXu5BTuEatMYD5b2h67WDSOMfekYbKZ71GA5CRrscFMv8lMqev4/O0aFzTHNephKJdEYcfobKV5Be3Y9K0m5GTtnl/phq0fV+gcJKhxnE6orpREKcl9E1nIecnBnESpCgn3uKnFcIpg4Se8KSJeLNyPkNmedGJOs+acZDaB5TyQoV/51dnJgjf8LwxAhPI+azrGDbLKT7y5EDFZwgKo7NRIZz/Z6S+F8gJZgf/ahGSvKfkfPSTWGvSehxCcZ5FjVk6FlZR3bZVg2xHCKwrBxBxmkyIzKADLqw/yPAd3DuzpRzPIDZ3Yi6uh0ZgWFWTy1SsP+Ac6+nHM9BJNyzmKE2I0EfemnSMAV4PuMzkMLuQw5fKLkLPTU7gRdJdzp7iTXxbxbZfjjOI0TjPh91KLI59ox3JwKVYH0SXbOLiCUy/cgJDFS1YW7RIHJ8wrkbUqr3mVGN1uR1xIFxBhz1E8hLXdtTA7PliK1wIVrT7ejY30NGZNJpiZF+zTga/ZlTFSRxLnJq0ggeDqFI9M2IEOQzaG0eQ+s2ZE82od66ckqHWyhuXFNHf5fDaKK9HhmZVaisKpQZTyHqgzs4CRYobyhuAbZ4NrCq4dLy+rK276Pyum60huciAz/ohUb/fjCMpyHdNRLYjByOYlTQ9f5YigWQ2ohOWRo6kPxaTXqQogatjx+WY6SfID6GnIzt/u/VyLF7l5jpdsT+xJXkZr9CidwAkk3hejhkH5yP1lqT/7wW2RrriRTThnRSFbEMPehhy/sXv61rkC5tQ+st9DnVIH0BhQRD4w1PoSDGOeSu7Rqkw7qAO0bx3ldAxbE5HfAiUp6BHz0IoWYkcH6LFF9WrwlIYO0HrsPs1SJZmxnIeMyKyIQa0zAcsJtIabzUH9t+tO66iHMX2lFtbMbRuRcxO4rmRHQSh4vuBB7CuY0Zv3wDlbIlEQTMWUiYLkLGSR25zk2V/+w10rNToWRkFjJadhDn0UxD96QUs0sfUkDFjW/n+jD7MfD1xPF2I2USptDfiXNjGqE6IZhNQucxAwn+5KC3GlTuU40iieuISjbIqmn+t098lp9tQg7vJRTSdU8GvmTGr53juVE7nxOB2WpUznkUnecH0LPTi9ZlM3KmB5EB8RbjMcInNrCZ/q+DOFeK4GLiQc/4B5DMyzJGutB1mAT8BTJqLkBGZAgW7RzGvLD88l0AVrC1w2EGUMXQQDUDrQNUT0HyPTRKh+z9AWL0/FzgM2b88mSbpofb0J2Ht1BJ07XICA6kC1VIFg8hw3UQybX9jBwt79uIhrlYdcNc4MES5/gWkkvF8J7fz0z/bxs6xxlIjt1dzswaM6qGm2XzJAkXEuXFDLQeDxEzbW2oQqIPrZE1aL0ERyKskWRmjcRnr6Hgy/n+/xvQDJrDZlyN2NAuRus/YA9ynnYTe1+Spd/B+aryx5EkbxhAuv5mM14vtyzNjCZ8cMy5kuQ7IwLn6PeEDpeidR7K7oeQTH/WuQoF/2ij4thMNIj+cyYy/lr9oMnHMHsTRb9X+m8+B7zmMx6foXSUtwsZzTPIdlxCbW0WutEDPM9vaxWKYPUjwbnAv9pQfW2gk3wwlYpaGaKVKKsy3R/jIyjFfAw4XMJY2IMU2jn+N40oSzCVyLYWsgOfRkprH1K4QygD8mCqoyfj9CakgMPnXagXZAPxPhSDUbzZOEL38e9QOcF1yKEaQDXwL02g+TWXIwNiB4oqJw24ARRRXYsc8sdRLfZaZDQuRvfrPmDjXXx2NVIgaY3NnUgZftwse7bFmEPEHZ9F63cmyijWoABAFzr/hchZ24+ej7vHZIhuuTBrQcbLeYl3HWZvAA/j3OmkuJvQM14qaNAGnI9zryJj77WT2GfIiOfM41jHs8fmsu/oMZonNXOsawqdh4/Q0olk9gwkn9uJvTmT/fGvRuupxYwfO3dqMn6JsrZtiA1wB3q2q5DRuwWd82KkL37kXNFeheHsu8v3XnyBOEcsoBplHnaSTtOcxE7/mgOpDGzVKDDx35F+uRLpuQFk2L5arM/DU0mfj7IX083oQ9nal5wr2r8VEJhSw7pZhpyXpPwIAcjJKLDUjwKPoeohsDTWkc2G14Nk8U+D02BGFZLXq1G5VQhYGZL585GTvxhd/+BANiKbIfSKpdEgh89WU6yqRMexHK2v5f48qsx4B9lEfUjetjqXSQl9UvBr9jkzXkR2iwFtzkWH2owZaO3PRNdyI7B9ApRPTwhUHJuJAjk0l6MHNkRCwnTyxz3V8kP+lY9kQ2YxOIqviRBFSaOPXIwig03IqA+9P43IKAiNmouRs/MjYGvmbB2zZlQnPAcZDd3IEP4QUto/LmnsOecw+wUiNDiXOAE7DHobQMbzYSSAViHn5j00LDMem9i7lhEjYJf77yWzYPUoZb4eCeFG5DRlCas6inPe559PBxLOz2FW7cuSJg60hpOljKEMMVmmGK7VAmTobED3/7vIkU3cE66jOFXyAFqnFyFHaTxgNTKUF6LnqJMYbZ+BFN1u/9n3UcTxRt/nsRv1npWa6zR6EAHJt9Da3U0uXe95wDLMvo9zxehyJxLK1ZGhb+Dk4dwhzDYjWZljPP8b/v71/5v/fF0/tZNbmbb/ELOfR9mhFcRewVYkh/cTZXYP6mebasYtzhUtyRo1+BKct4G3fUR9CjKqL0LBi8Oo/GzLSU6hT9v3BjNuRaVaS4iZMYcqH36bND6zjt+Mn6KSuiXISepE6z9k7R5IzGEpO+Nkxky/3WYUXNyB7udFwBU+41aKfCJ/Dc6jsBQ79Nw2+s86/PdeT3y+AzlYWWx3C1AzftJBvBhlf54lZoSG0PVtQ/rxSuT0b0R2QiALaELO7dtkE/50I+coE2ZchrL4rcRgRA2ym77l93sA5WLfAx5ybnSy4T47mhNQ86WcHyeWwHeje3Y1sN+M28dNEG4Co+LYTATIIPw8Ms7fJzK5GKLEXeONiawHdL//XjFjI0zszU7ZOteF2atIkSb7bJYi4RtSyKHpewgJ+pVIyNSiaJcGpGU7NbXIqckvn+vx250KfB2zb5c0oERA8BNi42svcvI2+XMIDt9R5KhU41xuxMhsJcroTEMG3ZX+HKcjIR0UR68//88iIR6aI7ejKGVSYDcRM1dhPyEbV4XmAmUJePVaaI7QahSZwu9j4yk1fItjGoWljG+jlP0A8V70EAepLQaewrkcAgcz6ilvKOIRdI3Gi2NzGXoWD5NbOtfpX1PRc9MJfBM9kyHiegUwiNk/41zmAMFRxqfR+szvsxtCRtBcVPJz2xgf12ihg0ibXKxEaQrFZ7kMF/cAf4gMzv14h/8qnm/7U/63TX/Hv1n1cz5zGBmYk5GxdhAZUkv9/5MBldDXMgsNQfz7Ux0d9kZxO1pLY1Iu6hzv+iGRC4hzR3YPZwq8c7SZ8V1kvF+LZJRDAa0XnCspkwrgDd4vIf2S/P0gqiKoA/6lGYdKbD9fZ+Rk/RI47LdZTTrzZytazzPQug9Zvkko2LiVRPm4z9Zch9ZhuJar/f77/fZrkA3woHM8YcYUtB7nI91ZygmspkiA1ox5qN94d+J71ciJmI3W2UJURtePspg3mfEg6o3ZNkrzbZL4lN9v/kBl0LX+phn/4NzwqNkryEXFsZkYuBw5NfnsXKE3ZirwRcz+PIMpaT3qT8kSciABs76oMS08iaIyM4jCMQyldEipbkMCrgUZ/5OQ4n+S2GOzDrOnMvp5ViLhmZXNaEPnfAkqTcuFotvLUdnZMn9cLf64itEsHgLOxWwSznX5ba1EPSEhghYcxGPIoLjGn1c7Kr+7BAnxo/69yf6aLEWRrMP+WCYDt3gHpRYZrdeSrEs2W4+ycYURHLO1qFG5lqh0LgIGMLs7g0xhPOIwKv+4FF23brSmmpHB8ALpWchiazkJx/AHE44mzibS8qahDSn6lYhpK7/8pA74XcyGChzw0YbZXLSOi2UZ9wPnYNaCc6ckKzCiUH/biyiYsSfjW4buy8mUn+Xv9yhm30Ezwc5P7utC3njlz/mT//p9vtWIDLbr0TpvQJHfKvT8hL6J5HNyDDmfS8mlWD9j4LNGJxWld44eM14jBMNU2nQyGaazkA7Oerb6kNNyDZo5k4VtKBASekbTBqzWIP20EQUp69BaqUUBt2koe/KnSF5dh3SbofXzSzQnJ3m+05FeC07XZv//uX57A8Q+tTkAztEBdJixF5W2lqJJricZCCzE5cTAZcAy9IyEYPAQekYCeZGh8sRVwCEzflpiJtMJw4z5KKuV5tTgj3EROo+HR+MYzhRUHJvxDkXxr6f4hOI2lBZPH2rl3MGEct5O4UPVhITbkyWPx7kjmH0PZZDC9OQm4hyFI8i4CZOY9/rPpiLHIRh1jUi4pNV7X0U2k1nAfuBqzB7LcY50vT6DGijbiNmBS5Bz1YDKzdLRKo/4AAAgAElEQVQEiyMaCF2JbR1IHOf8xP87/TbXoszDxUQ2pNnEXpsWf843o9k9m4HbcG63d2q+gMrb9hIFcBVyIFf7bFy8/2bnAF/0388vnagHPodZP86VIi8Ya7SSTk/+PnJeFqAIdSPqVfpJkd6hwLJTbFYEyMAbH8xoZsFhK1UW2oJkc1pJQh+67x/HbAPOjeXcovmUdibDMzSf4gxUEwnPoaBBGiuhIdn7Ks6NrEGkbPRdmD1IHLJ5COdap+Lp/4yjqMRsGTFL04HWz1JkwO4kll31EssGz0jH5mRhRh3SJ9cRh4n2m/Es8Lw32IeLSyhWLSEcBlabMSmrT8pTDj+EMqs7kK65gujYVCEn5BX/+SGURd6DdPRe5Lhs9dmLl8x4BclZkKx1wAyfNe/2ZY01FJZch5K2JJrJIyRwjl4znkEkHVnZqKlItxZzbC4kd/5QFdK/SVuiATkxLxPJChaj56EKuMGMvwGeHk4Wr0xcjOR3MRm6D7jajCdGqr/sTETFsRn/mIkMvVKNbl3IOM6a1nsfenAvRYZVJ7r/U/z/b8G5tGbIQshR+lsUXfia3+5egrNRWLYRlOokJGRCw3hWaUcL2U2LAf3IQK4jd67ATYljOAdF047577cT+3/SlHooxwtG99nI2N5HnCidnynoQM7dkD8fh67LgsR7nUigdqOyqnsSv7/a7yc/Gzfk99uMhnn+lc/uVKFSnwNk00jvAz7pZ+SMnz4c5wYwexrRHOcrvF50T3Yjp/B7maWKHK91fxKl9rOinIaenVNVtpWPBmRIzCF7Sn0DUuI7yX4+etHaX0lp5r2RRLlzQGwY3x09mIXo59nomLahBu7twxpg6twxzG5BwYQl6DkeQGurBq2vkaIlTtt/G2SWpgQa3FZy+x4HkGxqJpbIHkMyrJpokFcwDHhj/otwfHhkaJCvRY7OBb6Habg9ZlMoPVw5BA3qKR7MeRGtzRv9321I5gSndj1RZk5HJWU/yWKD8yWLHQBmrEF9rkHnVZmxg0jDnCXXAiYD3Wb8rt93G/C0f61ATkayRNyQDVQLfD+rVMwM899Jfj7FvxcczVqUQQp9rw2osmMyeia2+2P6FnCpGT8c4f6bMIevGMLsqckUbx2ooAgqjs34R7n3aIhi/PoqUfsVZs+i9PNcJEjfBN4dduRXjfs7MXsLKdEQXd6FMhjFmjBnAO8U2WepoaMQBWgUxpqP8jniFPYO4pTs2UjJb0QOzw4KCRBmoBR8H2Y3EBVY4OU/4vfXRG7UfQgJyL3I8GlGAjs/4j4H+D3M7kuUoF1DdoMmSLgt8cewBQn+Zor3lnT78w2/GU94Ca2Phei8k0qwEa3LXxZzahJYj6KN8ym8hiES9xLjhyq5D63pTcSZGvlGwHT/XlbZU0A/MYo/Vig23+NkvjuyEEXzTSjT3UMkmFiKIrOv+mHE5VMXO7cfs79Gz/kaJJ8OARtwblQGLZaCGS3ofLYhGbc65WtdSO4Eml7Qc3a6ZNPGGjegNZAfiOpHcmYOyvLfMszttqKAXrFsTwi8HQ/6mTEV6bMwVHMz6hl63Iw3URbjfZRxbEByso1YWrYeuKccGm8zbkDP1UFy9c90RF8fgjZZ1SV1qNz6Bv93ILv4I0Ti85coq7GOWD5cjXT2w8VY4Xyg6xByCEKmJZ8yfRrSCx1+uyv8+11EoplAJATwNTO+PYL0zAOUVxYdhpgWfqC5bcvQ/ZyGzuVV4L2TpGI/rVBxbMY/QoajVE9BI8XL1QRlZdJ6FtJhNhkp8vnowdyC2MzCQ7QVCaNgxIfhZMl62cCi1un/30TxYZIvo6xEMSE/G3g9zzi5CTki2yhsnN2JhP9yYv9P0iCpQxGeZ1CZ3Wp07duIRtokJPSqyI2gVqH7MwU5HUOQSul5CBn0K5FxO4fic4ECelH6fIvffrm9JdPK+N7YQmQOP0BZm4vJVT7twE9xbn15m6LHjB+gsoswB2GAOB/jKcR0ND6oknXuG5FiWo+eq0D5XUVUuocoXnKB/+5YK7LtaN0XK/+bitbzbj/nZi1yMIeQ0TUWrG6XAx/0x5uUA4eQ4XIxus4PDGuryn5uZuRmq5wslvl/A9NYmGs1lXh/6pATtpkoZ+rQ+qtgGDCjEZV1FQtE7QeWmzG3mCFuRjMKMJ5FfJZXIedogEJyEfB9sF7uVaPMybX+s14k9z4E7DLjTs+u9bB/YcYs5AQ1Id32TrkMXGYsRf1eaQHBY2i9LUM6fyqFGcYaVHLdjNZeMjBYhZ7XWcD/hOR2yGofc45jZtSZsRb1mzUg/fqaczkBoKeRLgiOTZAzwXZq9Ns8gHRjmKeXlGchK9aP9PkFlFOiXx7eRBnWYo5SYDMssH38jKIvoXLtbhS0mYWCG/vNuO1UsR2ON1Qcm/EO59r9jJqVpBvLEI2ikeslUNTzatQXEpq6q/x7bZj9FOd2IgO9jzglvgelwq9CwqEDCZF3kKBoQbNhitV3v4WE6BTSnZs6v784c0DHex2RjS0fDl2f85FgbkKKPjnk8hf+/6uRUTQTlbEEdPlzWoEEShh65oglaYHFKi1iXYcU1jp03QIjTSkkqWQHSRnel4LMqM8phwzbX2H2CIoc1qD7tmO4pXO+Dvo2r7RXonXRigbGjceBkU8jY3+Xf81GUcZADVqHMg2lnN1qSjPCjSyUZbwblZ/up7BcNDj+30dr/Gb/fidaj+cD/ZjdWWSo7slBfXE3ogBLlhzYAVzlyUtOpB9ivKCBeI5DaCDxMRTggdxMelhPgRVqVBqkT3MsQM9dqYBCyOCnOjZmXIkCd0PIAViKyIEW+7/f959tRXpiEN3rBmLJ10fRM7aDwnU+C/iGGd9JykDnOEhpuZKFq5GOy5LPoUJiI6owWIKe+wEk35Yhp+b1xDYC1XMtOu+LgY85x69IyDYzFiKDvok4/2YRcJXPSv3S03S/iSogZiNbqc9vZyHSCQ3+vTakK/qIujS/7MuhaxXIgUYCb6KAXj6ZQ0Aou7s9PxjnSyC/Qnq1xmF0z7/u2Q7HKyvqmKHi2EwMPI6M7fBgJ1GFhMhjIzw74ioUXUir9W8CvonZd3BuL2Z3IcFz2B/fAdT8vYrY09KPhMuvcS6rD0hwrgOzHyM2sql+e2FAV5gV8HOcS0Zr5iABWkzp9BFLCBxSJP2oROMlZKz9L8TZEYf9sQenDf/9g0g4TidSrXajyMke0qNaIIEWBpSBrkc1HK8Pnps4h4NEYduQOKbdlJfBO2nmn1GHys1GpEfkJJX22MG5nZj9EkUWe4nlePVIOfUgquQVZN+/Gf53pYZGjjyce9c/m7+LjjdEleuRIXMrMiw+QaHsOIyegS97QozRaF5f7PdRbC2EEpgVyBkIgZEFqIl7CXp+3wLewLlTV1ZXHG0UlrbsQP2Ul/q/Q5/NJHRe7wF3jZss5sTCcGYapZaFm3EhoiQOz8a5aB2+j9bsWnS/3kf6czqR8OcnzrHXjDlIP4f383EQGfOXk8YaOkx4KufVlNYnh1Flx1+gTNS56DrsR1mG3ejahDLhZf5zh+TIIPC/mvFUyCT52T5/5D8PTKNhLg5+H6BBoSGD/yW01rvQdTwLBUkPIDk1RKwomYxkabK6JPSnDQJzzKgeCRpo5+j0M5C+6s8jmV0JZdgvka4TzyEOtk7DQeTsnYcCy2c0Ko7NRIDqu29BqdzwwA4iZVUFPMZI0gOaTULR1qwG5nYkID4M/BDn3sbsn5AxEzj9DQmVR1CUqRXnWjGbjNmVSDA3I2PoBWRARMfMuR2YfRsJ58uJa/UN4PmUmT0NSLDOo7jRP+Rfv0DXcdvx0hizGchpC8fhkCNyDXGYIsiJ6UQRtWnIiWtHgvpQxjULTulBZJiGpuR3UKYppN5DHe65SPCFuvi3/W+O+gze2WRHXeeivqmRH/Rl1oSU7xp/nLsRI1Q5U7ErAHDuZczeR2UtFxLr5h9B93sA+AZSVO8T11MVchoGgbtKDqgdLTi3EbM/QwbDIvQ87UKlkjWoNy2U1OSjG63rj2P2t6NwDpMoPwuqBnpleX4HOQM9KOBQh3oBPuTp08eWWrs8bCV9xs5epA8WoszBIIqkP0tku6pg+AjkMaVQS0oPky8f+wjxmW5Ggb/QAzaIyrTmIEO+Bt3Dx4BfJ8qMLkKGeLF1fgBYZ8aTI8CuFYJvpZ6rQWCS399G/8KMWuB/Rs9VDdIfc5AsCBnTavTMLQH+rRl/5b/z75FRHwK6nSjjGPT/TmCtGQucY48vW/sHpIsvRHr3duQ8VqOezBl+X7UoEJmshGlGunbQH2vvSD4vfo7Sd1DJYBiqa8hR+xXwcsZ8qXWU7os7jGyVimNzqg+ggjLh3C7M/hwZE+egBzPQKRuwGLP3ca7o5OQyEdjE0gyTKUTjYc3xeRXObfHNtQuJvPV7cC5mmDQH42t+G4f8sdehh/wDmP0oZxCjc4eBBzB7CDlS/TiXJaR7/D63I6GWJgRCjfEmFDUDDTx8GimPKgqF91FU83s+Eoghul7rf3Pf8Yiu2RIUya4nMifVIqerDTEyhbK8gF5/TPsoTE9PRVOy/wLn2jxd8DlENpelKL0dolfVxEnTv069SicDs3PR8NEqv48hpCjWYfYCuhYVo6kcKNv4S1/aVYPWdlx7ChRcj5yfwDAWmn0f9c/GqYOIP94mONwBoiKvpTjDU6Cnn8fIl0T1lP4KoGcl1NXfhJya/PkSnfjBiJh1lMw0jzGco8uMx9Hxbyf32APz4wBwh3PHp8pXUASeyrkG6EkxMPeiwFRa5URAcDLT+rCWIN0XdNNSCh2UfnTvNiKdMA1oyOudWFhk/wF96DmcwkkSeThHvxnHKE2t30R6n2/y/JYhPdyZ9/4g0idN6Lp8i0iAs5tYblePdE4T3nFC53oRnnDFOyJbSBDn+KzTEqQvP4F07WRiZreaOPsp9DfOZhQGx/oBq7eaMR3dn37gQImBuTMpXZHQBSw2o+pUD9891ag4NqMJNd6vRcZJYLB4ATHpDL/+X43ym4BNmC1ECu0jRK9/wLOePXmSDk6ghkyiGWURZhIF0lTg05jdjnPd3jALvQO5UBboa/5YkzWivUhwTQa+itnfFAz2k7GcK1CVXTmb2Kj4DopYhFraBf43XejaLEeC6giKjIcGvhrEn9+CnIEhCiOgx1Cd7VR/nLOBh3Au33l40O+nH0Wy6/1+3iBGheYiBwfMZiKH6X5iJKkXXd86YtarD7OrgH/nt9tPFPSBLOE9/7vnEaX0yPaXmC1HKf782Tkd/jhDT9XwGrLPdIh2uNAJ0IDYBzB7DBkCVcCREb+vI48WyuvtGiLOzRhJ7ETrMAwoTEMgbNiK2TTUP7CT9Ih0HwrC3ITZu6csS5aNJ1AZyzp0rMfQ+U1Hxto/V5ya0jBjGbqGoaey18+l2YjKq1qQLnkJlWgHhsMkqpF8vj+jzyGfYns22Y5CGCDZiozVmgTrVSjnKgcjZeA+RSxNz8JUNAMnB84xYMZGZAfNII5FyEfotZmNArj3o+ucPIde/97ZSKceJva3ZsIb+tuAbWY8hpybLxKzYyGrucX/PxANjFqm1jmOUr7T2YvskmKytRroP9OdGqg4NqMHZSe+jgzhw0g51qOmvxsw+yGKupyPovBVKGL4WsmyHhmZ3yA2HHcjQRGGeS7x2Y8THd4X+lkCZhAVZ9LpqEKsIfWY3VqC7WgNEuxZgrETOU+XAL/N3IpZAyobOZ/IXlKLMhvHkFB81e/nLKSQGtB92Ikim0lWkgF03S9C0fAXyJ4y3oaEaC1qAs/HDtQPdR0qYUsqvnqkIJ8i1sleQGzwfcgf+1xiRmSP//wrKHLV5o8fJMQC+1EfEsp3jyLj1M1ICKc5zKEhex1mz+LcSNFjVuBcD6eil+bE0U/5RtfIZ/ec68PscSRnt5NuQC0CXvZZ0Cv8e8WMgXYU7Z1LOcyTYwhvxNxvxmtIdi71Hz0FvF4u69WZDDOuRwHCdhRkC7NivoFkdiBlqCZm+uaj9dvmv9/kP/st6boBhsdiWOy7G5FhXkzOhszQSAVC1qPg1Rxiv2cSCwmMqR5+tkwIEt6FrnHaIM+AOpTpmkkkzEkr/Qs9OcuRbVVL8dEQuT92vA98z4zn0T0GOT09/vjmIxl22zh6fl5BDHjFxgAE2+eMR8WxGQ0oU/N19HAmDfmQQWgC/nckdMLwNIciGtf4rMsDqUPk1OPwH5EgCehD81d2+tdyFIV8/ATP4D3kIIEe8Mv8cSeFR+gH2YQEwbUUp5G+gtI1ogcQW9HDqZFR1cJ/HtWm7iLXaDF0TYaQ4dKJ0shDSCCHetoNyCEITGZhG0dQfeqdyGmaSeEcmmpk4KSXAjnnMPuN39YNSAmEfqNO4G7gxcS5LSLWGA8ioynfcJqBrv82coe0DiJHoxFlA89H2ZKRd2zMZqNrm9W4CFFZrSZkpCo4E7GD0o5NYAMsNavnRPE0en4vQ4bnESL9eQuSWff777ZQejAiaH1PHpGjM5uPjBCAAzh30lkrb6yN3pDQ0xRmrEJOcD7hyyoke1uR7NuFqgCWoWxYBzGjMIACWW+UoNvdSSSuGESlRfNJH74atjMN2Js3oySwayVJbfIxG7hzpKL3ztFtxq3A7yMdGDInYUj2JuDnzjHoS6wuQQHCBhTseBGVrX6Q3Oetyv9+EjFwOJdIyBMqMPLPswOVsoKcuDdO4Jw2mPFf/LFeha5ZONaXPCHNeMHryCbLmu83Gcmy3WY0nunMaBXHZnSwlpghSMNkxF7xCrkDFENZzzXoQc5lNFE513/wv91FZPKoRYbtQmTMdwFf9NmN/WgYZrHa2HxsQ4K1GQmVwHiUxHTkTA2gcpKrMHvSR5jTMJXiESaIJSQ1kNrwuMq/tqd8FsrglqLG2RZ0H0Lt7MvICbiBGAXqQxGi7cRBmO2Irvbz/u8B/6r3+3iIYg6jnJYXMHsFOS5BKO9KGQhYDnXzar/vrIGV3UgBT0elgs9kfO9kMIXyous9BGKECs5MiCVxB5FyNQ3zEQHIcGTScI5hyPcubUCZ5hXoOduFGMPeTTyLnZSnB43yHKAiW7D5wKeQnA7BDcNsF2KLHFfZoDME1yNdl5RvU9GaCYGtRagMq8O/wiDOK5FBf2uxmTUBztFhxsvI4d6F9M6SvK81I4c/PBst5JX3+u0EJtIj5OrVOmKWadjGfonjbzXju4gg6Fxi/86bzqmk1IwlqOQ8zIvpR8/XFQSyHl3bWmIDv/PncBTp7xnI6etAJeZXUujYhIDhVPQMv3uC53QUeNiMR/yx9I9HxkDnOGTGz5Bj2UUcNdGAaLJXoLX4acCZ8SLw5DgdeTDqqDg2o4MryY2u5+Nc9NAvppDazyGH6HrMnse5pEPxMWJtadLw7/f7m42auzuR8/Qh5N0PYfYw8HRqFigfMgzuAP4A9XAk9xUchcNEYRKYvGaRXTbTgQRWEBqtFEYeaogRsDSso5BvPh/HgBU49z3g557G9b+gsi+QgxCuQS1yEmcTmUQM5w5j9vdIoZ1FnDK+sey5FzKc8qdT5+MddD+zzimwxHRT3KjqQtd2blnHNnz0Uz4bULnN2xWcvvg5kh0LUGAlPM/1aI3uZARoaItCck7DNCUDLEP2bSHO28lCPXrGTjyzIqfmD9GznJ/5bAH+0NPnV9gFxwhmTEXyNT8AuZjcNTsb6askE+dhYub/q2b8tXNlyb5QcrwcPRtbkY5pQ45CKyr7qkYO8NvIYM2Bc7xlxvfQ2g0DcMMMtd8Az44G+53PAG0nJbjor+dXUHAwqSdD8LMXneur6BpMR89Vcj5OKF2bgoz299Ezepb/XrjGoQdmCnCLn2NzMucVytvGLZzjTTP+EVXHhCHoFyN770FiIKkaOZLnmPE95zKDoqctKo7N6GAa2QwWzcT61xko8rMQPag9SMgeRA/tKgLdr5pcL0BG8KyU7dai8ovJKG05RJyFUo88+amY3VdWA6wir/+IMkQLiGtlEAnaLRQ6IOnGr9k8f8wfJDe6tAsJ7iCUxEKSfXwLKW1chNKB5DEtIHcWTEBwCOega/v/s/fmwXVd17nnb2EkQHAC55miKInUSM2SrSGJp8R24il2bD87sV88vO5UpVKvqqtfV3V3pf/o6levk5dOHD8nni3LsmXZ8iBbtmRbsuZZpkiRoiSK8yAOAElwAAEC3P3Htxf3uRfn3HtBgiAA31WFAol77hn3WWt9a/jWmtOsXjoHL+3LXoshY3hD/O1Tv8+E8ng9Aqs+OGw2KctzHD2/NqqX8Hmz40gw4uXJHuSQFaXBXZo4w8hZXSaQhNAddcctyMC6XuhDPQhPjRB7Y63n40N08z7bjdlmpCPy3l9DEfCf52RcaxPpjA+idyivWbgbgZv3Af9+Rseoy5mIz1Apl06SA+3BuPLtnPHyOLIfq0jU/IUSZ63cgRzS25B93ofWmLOczkc6/UngV2VlaNl9bQa+ZMYcVN5+EtgzAvTOZypXkObW5EkXykZNR37KAUqDpj6w8zVkSy5GIGhd3PYS9Gy8rPRe1AczlkrGzqmEwA7gLjPagU+i+1ceTB5EIHwe6ke+c1RPcgxIHdicGzmCXvC8OkcvaZqEHHDv/5gUf19KilLMyHxvKXrxD8bvlZeGzSLRrDr7SkAKdzl61tcBqzD7BbChavYmhH2Y3R33043A0hGGliX5wMihmQext32GVFN8Ct0X74mZiWrivS6/EguJs8FUikSVf76ENPenSA6h+/6lCtt4j8/7SVGSw+g6rgfehdlO9OxOIGX8RkVnKIRjcbjpZ1Httmc8BpHidkOxn6FMbVlpid85N3S0IQxg9igCYVsLtpqDlOnYHgpal1Ix68R11UiSPmj46v2YPYQcGemQYrr28yn3oJ7IJcjJdKfWe3Je4OxoXxciZ7VSj1o3In2ZXy9JGzXxcq8GShvavcwJZFcHyI/m+2yXbgTgqwIbgBDoB56O5UJT4n76UBlWWzzW1hCoqTogBPZRXPY5mnIDQ/tSy+UpVOZ9El37KUqJAtYipr9rke+yHt0b7z9tQyBwD/Cls83UjGNpQ0HQSsQye4FVZswMoWIF0YSTOrA5N/IMakjMW3SDKFW4ChnQ+SQav4Ca6GYiI/tA5ntN6MXfjRzerDJuQNH+XrTgO1AZ1E2k/gznib8A1eY+g9lPaihN2xD3e4T8vhdQxGpddGaSmDUAf0ECAccRqYE3Xx5CjsNbEdPLt6sMlXyFVIpXJDNJ/PagezUZOSkr4nXsQwBxgBQl6qa4j8XlnQjUZOddNCHHZXH87Hmk3K8GujG7kxCKIlggoOAG1mmeQc/rGaTEF6M1UxSZmopK6c7FJHeXp9B1ribVP4POdy56vt8bg3S4dckTs1Uog7oA6YUGzF4HfksIlRzw4Yl67sZ2eZXY0b6C1vZtpIz4btQg/kpNJbzFMofahoaeitvWgc0oSJwDtJahNmUPCnT1k4J25VnzVhL5j7OiDff4pyitYHilaNtxIs4AWyRLUN/rjLhde/xOP+oHWpf5/jPIj1mB/IQ+dM+bUADv+7/HoAZki6GyXvHPFlL5uUw4qQObcyPrUH9L3iCvw8h4OWOWR4N6SLTNXv5wFalR3R3JEwhsXE7Koni0wx3jQdIwyWzpQ388biuq974kZgy2FBruEI5jdj8qZdvJ0MhVZzzOQ6f/otKLxcB7EBjwYZw7UXP7VYjswMvvdgLfJYQ3qCzPIMBQHmFzaUTK8pl4HkuBD8dz2Y4A1nzUaDeI0uK9CBBUniotNrqbGTrv4pq4zwPo2S5GpXoBKfDPYPZFQijqo7kyXsvPEUCZFM/tUOYcF6H7PhOtEweYFo99DPjiOR2OGcIgZj9ARuV2dJ0BGZyHEdtbbf1HRaJ1s4gsYxTsrIOlERazP0K1+ftJpZaucz4f51K9fL5O77yIiAyexOxppJfCOSM3KJZa+tjGnZjRSixjHYPO6ONIB2eHT+5CwKYBnXc3QyskpqCZNpBAzu+79KB3J88OLEO22wOa+zLbTUW+aHlfzgbge8geTEf3+JUQzhmb4niSxhq3C8PYdsJIHdicCwnhSJxT81ekfhuPONxCoik8QZoZ48MZDyBwsgf4E8y+Eg3sZqR4W1FTrJeZGVq4bjwmxR8vweomMWhcFI+3NR7rNqRUdmD2nZKMi1kbqZ71JeR8v5s0+K4hHncf8P3TmRazZgSCViPn1531ZSSGsV7k/A+gNTgfNc/+EyGkUiYzK3FqQ9geBxa+jdT34dIW9/PruN30eP+74nWcRJHjfcgotcd79zBpTkWljM3FpMF+Lp3IGfRoiKfXZ6D7fhBFS66neDbPbaRMTE/OORxBxvfSeP6zEZBtQs/7FeD/Lrlv50oEnNZg9hK6zgbg6Bn3HmRFJYsfpHTQmgFvYnbvqFzf74NoBtY7kA7IAuFAGnb3Ecx2VgDjY1/MGklTzk8hfbGjKkhWgKfciT1b2UdtoMWoPl183IgZC1CG/qrM39YCTziL1vkSs9M9HLcgfX0rcrjXoxKeLShotZdS0OLkObtIdOUzgIfMmIXs3nK05l4D1oZQlfBmosiTqFS7HNi0IvB4EPkD65Gt9u16kC1dET9rR36TAa+GMIRgacTFTKXw42i4pZfBVxOjOhvthJM6sDlXIuf6C0jRvQU1crmR3YhARmP8/ykENFridnsRkFmIhne+GPscHgD+HEVZN8Xf85CymIfAkU+gP4aAUyeptKIRlce54XYHfDbwScy+jADCrahe1l+cUygL8q/xnOYiJ34rsL0s2/NeFJmZHq/RJzY3osjLAKpZ98xPP3rxTgKfik3Hi+I5zMFsAAGr5whhF6Jy7kZlNLNIGa7DqFbeB1StRut7czwPz/J4+v9wvH4HCluqlMFNY2iPyxKGZrCy5WSgZ3kzZr8d0l+gnp0pVJ8+vB09ywdQn9RMZIQfQSWAZytKh0UAACAASURBVA8shiNyDkeOaUWg5rNoXZaXQU0HPhsBfh3cnL3cTH6fnItH1K9C62v8idlFKLgyjdSX1wDswewH56GHZRcKSkynmAGxE2Unz6vDP1JixqXAx9F62oWXO6rsa7UZd42Gw1pwbi2oRPpSpHvXIhCyCunXXSjg9bP4/3eRehwHUDDJg4szkC6cA3yINLQT4O3AO8z4WQiximBiywY0TqGT0tK9BfF3M7p/L6P7lp1P432uM5B9m4bu8d+Z8QjyGaYiMDoF2YrXa+kdKQItEdyuRL7Gwvi3TQigbRrjIGcb8mGymcZy8c9ORIKJrnPBlDcWxepVHqMkZi2IYawHOf8LkbL3NCxIKQ4gVrMTCED8d0K4P7OfW9BwrlMkBTofKdETSEEvQS+/O93TkLLYi8rkfHHPAB4lzXB5HinxqQi07I77aELAqRv4GiHkp93NZgL/OW47L17LBejl6kCGvQFlWp4mZT9mIsAyNx6rD0WPj5BopFuB+wnhiXisBqQIr4nnfgwBgOfi7/+CIkJ9yJheGveZXfAt8WcD8O+EUDR3iDih/E8pZUm7nURR7TIDKcYsSFoE/ENOD5IBf4+iyZWUaBvQRAj/WGGb8Sm6B3+DlHCR0zcdrZl/rZelnYUISP89Q4fblks7KsX659E4rREVgZpPo8xH3uytVuDLo06rLPD+OVIpTlY6kZP35YkAbGLm4m+RDiynQG5BIGAWImt5ebTnhpjxfgRY8vS9oeqC+0PgMTMMBQM+hnT0FlIv6yz0PF9GmZ/yIZ8g+7AEuCsE1o74xYwxiQ70p5APcRDdq5tJg0iforS3dgDdw6nIRr+BdNNuZMsnkVjiAnoGXunRhEDmD0MofdcjaLkEVUT43Kg3UCn8JgSyPha36YrnZkhHTIvH/vFYBjdmrEL02k6lnZXpKDi9F12fDwl/BF2bfAo4PgbLQ89a6hmb0ZMFaDHtRz0lFyCgcBw5+4Ze3vb404yifKVR/hAex2w9iqheGP/ahIZ4eVbIh2K5wz2FNAQrq3idjcUn17+HVI5yFWq8X4+yHjvjNbwbuLvgGlfHbZahF2qQRJbQgRTSqfjvWahEoyNuexSVbHUCP8zsczDehybgTzE7QAivItaUD5DKNwZQ9OVaxE4zlVQi9lrc7pJ4fG8K9eGj36gIaiSbSI2kbogHKc3ONMbzKM/A5DO5hRBiWdflVG6wnkmloaDjW2phjHLgvZA669rZiNdaV3MkB9F7Mb5EwO1DSK/kRTEPIr3zbuDro3hmEMLOmBH/AIkm3jPiu4A7JgKoiXI10otT489RpHNXIbsHskn/G/CMGfeHwOujcWJmTEM2oohNKqDn8QdmPBMZzJ404xV0XW6jDqGMzqsIxO0gPws6gAJXf2zG+okeMQ+BfWZ8AdnityAH+ySqUnmF5JN0Iyd7JSpZv4DkA7nP4QyqM9Ha+RVy4qehcr/FCBx9yIyvo9k9+2NP10fjvrtJtmUuCno8h2zyhZTOmnOGu4PoOe8HHhuZOzPyEgKvmHEXKuFuIem8WcifexWBbl9zbQh0fhb5G4PAYBwa+9REYk6rA5vRk1ZSVP5VFOFpJLGVuXgpk8+hGcp0FcJB5Oj+NhP13xZ/LkZKoD3uy2mU9yJH3mUycgAa47m4o581rg2k+ujNSEFfgdkDJfX36se5Db00V5J6hrxJ0OfguFIbRErGMzLrUaRkKsXMawMIqPwRZnOBv4vn61GcbUhJ7UcAayEyUP1xm40oQrcQKdtT8fy6CGFTwTGTaGjnOhRVcud6R7w/HpWcQWlGjHis7WWDVrPyDIoeFtE5+zXWRCU6DsXLCatJiNvWgc2ZSz9657IlIHnSwfi8z8uRw1xp7tMB4ELMZhPC6PazCNx4Oa+XB+8Ddk+UTKQZncBfI73n+swDW70kyv8eZKOagE/H0rTRIKxYjuxTpfvdj+z1EhTQ8gn1D5ElyeF0yV0zxXYLdN2z4/6qDW0e9xICvche/Q7AjOtQ70152XYPYvPsQT5PH2IhzN7LWSR654vQc7sBra1DCITMJjKWxhlBV8Rty++1g5Zb4n6fKLoE5DvcngG3Y1JCYJ0Zr5OCuvPQ/fLBrllfZDlagy3Il9qK3s1r0b37egjjUu8PkYbqm9RlhMQzM6AX8jlSaURWnARgE3IuqinCRhJAGkRRkZ+iUq+DaHF3I+XqEdvmeIyN6IXwv5WnXU/F715G6TCzNADTrB34j0hZ+GDRnni9UxGA2UUCTs3xPKbF/T8et/c63D4UWfBytiyN5mEEWv6eNLjSaZqXkPputsXzWFx2PcdR3e5zpD6fx6ldfoLAzDJkqPeQZs7MQooiy+xmKLr3aOEe1Tf0k3iuM0lRXEMKew5wdwSzY0fMDLOlmF0Xf5ZGkF2XsSpynh8nsc4VyRSkP8abeA9fNXHWyNGXEAIh7CSE38WfXRMI1MwEPo90/gFkfw4iXTyX1M8IegZNCGjvAf48Dh0819JO5ZlmLj5rrpp4NUQt0lF9kwkpG9A9n4ICqi1ln89G7+16hgLE5SgI04ve75tIJDt+348gIH0Qzcx7K5Uzcv3IX6jk//aj51/uQ4xF6UDXvBy9Z/Pjv/+YVNUzHwW9u1AwZQXJb9yDfKNPxP6zcS/1jM3oyS4EaLyh6zn0Mi5GL7u/pL2oAb4LUSBXG6I5gFl5FHaARKu8JP59atx3J1rMT8X/LyDVvOY1hHvTZ7ZcKEsf+DYEQJyUIFuudQIpsU6ktJaQSjB2A7/JXHdbvA+nUPQFUsapCzV4niJlSLI9LE4I0IIGpf0m7r+TVOZWLlMRiFqX81m+hNAb2e4uRf018xD4XIJA4quZ65mCFPYjVB+cuQ7d0xtRSt5L5V4GHo/gZ+yI2WJU8jO77JP9kcGsWllfVobDGDUWhtCNd1mDat5nk8/AtQgFVc7lTKRzJdnyrkqSLSety3BF/Y0LkS0TOYCIUd6H9NheUgTeAzSHEKhZjNZXC6kHqg/5IpeiPs8aToF58Rwa4743h1ATqD1GZfrb6ahU8BKg2YwXUVZhW0Ev0ElqDxCPxeG0oyELkM/xAXT/TyJ/aBMCJZ3IfrvOaYnfWYHsvbPFTkclVAOZ7aaiddWG/KaZ8d+VShsb0TPrpDILYWCMl+SaMR2BuQbkn12GAF5P/JtX0CxA995HjDQjH8WziZPQ/bgaxj/RRR3YjJaEcAqzB9EibEVKuQE56W2khbgVNZ8/RAiljpzZDPSiX4UW5h60CJ9EYCAbpRhEL/pmVJp2K8omrCWRAnTGbZuQssnSJ2flJMpKQJY+UNma60nla91I8cwlUQz2I8DSR2J+24MyUVlD0RzvySaGNvlPRqVuB+I9Kmp282MtQvdzPTJS00mU25PQi9wHfLOQCKFIQuhHzuGaSG0dUCTpdmSYHYztBe5CrGWlBlFUtPNQxu4qEpW000m/AvxmyPMfC6Im6M8gsFjeFzMNze35CiFUmoicJITdmO1kKJNOViYUY9R5lRCOYfZ1NKR3KXrn+5EOakEA/YfndCbSuZPdVJ/Z4O9ZpaG5dSkSs6sQXbiX8zYAvffbu18y7rsw0LgN6fBrka71gdE+ZLEj8/eXMns+isqHKgKbWOr2ARSRdr1qwFEz7quhnG1T5rzLg4arEKAJKPi4C+nmq4HnzPhJtkcm9nJ0ozVXCSw7+2mhTozMXRegjIQDgZeBF0PgQHRg/Z7vCyG/lNQMpzlfFK9xF/BGEegzOz2OoTeEkZnFEwkXfJzCatS7ewh4EPWTTkb39VLkj6xDPsbJ+Nlb4vePImd8gFSStjde10IEYrwUvR0FWaE6GOmN59ZcZTuj2CcaK3IrpS0ETaR16BU3VzA0MGjo/s8i+R2Tgf8S+5UeGMsleNWkDmxGV3qQYr8QRSp6EACYGX8eA/6v3LIjsyvQsEnQYu1DmYLLkIPZR75zeAgBmoNIQRwhRY4CWswNqFzqSvLrhV1pd8T9eER+Pqnhz+U3iG1kMlIKTrHstMZt8Rhey+ksOTNJM3fK5Rh6eW9GTn+ldOlR5LBtRODpJwg83IQAVw+iTV47hKVsuJLom3cD38WsA133SeBgDqBpQPXBf4CudzVSwjsQCDuAjOAqYEEECCNHq3y2olIzj7rlceM7t/77MRsOg9mPEGPUTIZOSJ6JnvePz+ic6zJUQuiOvR4XIKM3Gb17CnqM39KobWj9TKN4dsNcYMO4ntFzvkSMnO9FwatsVrZ1kMYP3MzTU5/iph2BxqnINjWRSnXdUT1Jes+ztNsONiocPlK/y36UB1XagP9gxt0hsKZoHyFw1IynUOn01sxHF5CazTvRu9BHyihfF8//QTPmIlvkw6Ivjj/PkT/hfSFqzs7ts4wA6cPI0TwSfxpQedGfmNGF7N+peC4DZjwNPBr7WXw/S4GPkGjOQ7xXx+I1N8b9OFvdtQggDgINZrwa91mJyKVQIqi6DAX55iKbfyUCq0eR/d+LAnruBxjwLcTu1YDsdBPJj+lBgI54Xxah9eMVKu3xOH3x/kxHzrqXWeXJflLmokgmITtXW4DuPIgZbWhdZomHelCpvIuvmcllX5+P7v8+EhDykr8bgWlmfHe8kl3Ugc1oiRrsP4GiMF6WNQ+9gFsQGcAU5OQfLPvuMgQW9lCarXBa5CXIYDgPfA/K1BxFL/5e4Ivopb+V1ITXhBb2GpJCvjEeP7ugmxBAmo2yHB7pyjNEe1Fk5iakvLy3Zh9K6bfFfzeQaJr97wdJjDPlchIBq63xe0UNm4PxeI3A69GBeYTRmMkRwlHyy94c1HwAKaI9KDJ4HCmldmRon0UgaSdaG+9FWZ+xIgvRedXCYLaIWo1CCG9Gxqg/IzFGgdbONuCno07PO9FF7/AblPaEjW9RVvz7yPltpDTIY0i39iEGyboMR0TY8icI0JQ7hH2HmL57HrtXzmXvbW+yYAbqq1iG9LqPMfCs/5uoFDobEZ5CZb0CCgi1Qe5wz9749/eZsbEooxHlwXi81chZPoxs5ylkP19jKFvWDuBWM3ahGTgDJGapfWjcwntRb5qXHrcjB38TGmdQJH+GQNXWsr8fJTZ2x+97lrEJ2fHlZnwzBHrNWIwy6YdIoNNHLNyMsiYvxn04iHsFgTEv4VwEfN6M7w2XmjrSK38I3dMD8Ryuj9dwGSpBfAK9f9n30ntL16AZOFMoBYcHSMM8DyPbPh35OPPJgDcSmG5D2Z08+9OKnPwdDHX2Xbz0/kc1ljeeL/EhpllfbQ8Ck9kMYi8pM+MVJl4ympWW+Lft6JldhALE407qwGb0ZCXJwQeBm/IBZQOo5Kq8J+OP0IucV4LVhjJArjh2IMV1DQIKXwCejdmFg8AWzCajF7wXRYX+ECmT3Si6shotfC8da0NK/zuRatnlEPngZgPJWHQgZbU5XsNP0LDNgFlDdEYuQoDqufh7JnGwVNx/B1qru+P/XyUp0HJpidtuOg+D+CrJpQjUbEUKdS5JgR9HRv5apPRPIAVzGWadhFCJ5alUzKag2uRpSNm/UfP3zZagjNKl6D7vRM3mm+IQ0OEymNUe7RJw+TJm80jN7fvqgKYuw5JEq/weBJKzQZhXgJ+POSKO8SHXoEBSrqO3mB3HuphlbfReTioxe43UMD8DOactKLiXDQC5H1LY72jG5HgOlXS626uVUDFrM2DGPcjW3YxsyVRkV7aSXxLrAbP/GfVvZOnE+1AVwHKkw5tI1Rg/Bn5XVNZjxux4/HJQZ0gXexR9JckRHUDO52IE9n6BZqx5tmd23H4Ocva9pGoV0uneL7sw/n0d0tk+6uGvzfgpidV0Zw2zhm5BlREOCB0c+Ls2DT2/p8q+1x3//m/oWUxCa6Uf+SDNJDtiyNluj/vvR8/7EGk9daBnczvwHZK96ozfnY+e9aZ4TlfH/bsv04l8o4eQPzKWJe+Z+CzDlST/ogWtRe+zdkKncmkmgeJDyDesA5u6VJSrKS6PcFG022zyaXpgs+lIYZY3ZDu7mGdGDqIXcgMpCtuJIjtryWY3tG/f/+MIGC1BEah+BBjmIuf0CPAPqOenNAoWwn7MtpLKyLKyCwGR2fHnC8CWktr9lPnxht9+pGQXIKU9DSlxp7L2jNSG+NkF8To8He/NqeuADbEfZKywDv0BqXdoes7nA+g+LCQNKQvx/9WBiWZ4vBMZByMNMbM4K+enQ55f+q6h+uQ/QkbBU/WdwF8BGzH7HsNruD6zey4gM6bAjJktIA2CHUDrb20Iodr7XJfzISLbcJDsfYRvDitAUJdyWUUFPXQrj3UfYjqT6G03Bi3Q6L0PPUgfX4Sc1uPI4X4lftV19s9DqGgfncWuWmnMcWTLCoENnJ5C/wbwhtnpgdjVSE/8HPJmJA0gh/IwcpR/BBypYcDjKvKvaRYKgLlz2okChNk+mD3AjWZsRHZiG8q6XE8iSTgZz7cRZdDeFfcxiFOfC8x5E//VyPmfhUCKAW+acW8Iuc4wkUnrVoaOisgSeRwmladlr2EAaA2Bk/E6upED3oH8gTfj3xahstk2tI764jUdJIGaVvQcdyBdvYxUtXIdyS4+j3T4TBI7mvcvrQWeHye0x13oPrRSGvTeSKoCcSbCE+h+7YjfK3+XO9F68mqZQ+jejEupA5vRk1ZqoyL11KpLO3rhWtELPwcpMW/KnIMUUH/cNivdyGhcRRGtcQgnItPXexFt85S4Lx+MuRspvyfJn33xC9Qf0c/QMqymeI53VpkV04WUzIUkBeNRpldJ0awuElX0S/H8VsTvNiHF2Ydeyj+N+3gTs58SwhnVDQ9blA1bRKIy3Umir3bDWTRH4TgCddnyoGrN0F7m9n7kfG+ntCnWkEGYitm3Mn1BWbkGlVJsLfvuofizAt3PJ/g9YjAzkTy8B2UR+5FxbkUA8p1mdm8IYaLOFxr/MgZB8jiWirNfGjnFDA727GXOHCO0hlJb4RmGFcgmLUE6shG9Vz9CZbiVpPzYhmxFE9L5PZm/DzeoUuvk9QUMrbIol33Ijh2rcWr9jILjz6TUX/D5dlkZQM/lIlIDvVdqeNbcg36DyL6soHQm2iDyEd5EjvAJFJScRLJX04HPmfHlAnCzMJ5bNis1GP+fnc92Cj33Y3H7E/GzrfHzw8jBzmM020EaEn4Ret7H4j6b408/sp398bN74/34SPz/rngszyJ1kUZN/GsIFbOBY05i5vExROu8NfPRDJTtm4R8sjYEYr0MbTa67pPInjkR1QuZfYzr0Q11YDN6sg9FZ4oGNUJSAtmIUDNKK16HFMhUEt1xG1KKi9ELnmfE96FoSqV5LYYiHM+QqIaPZ85jDvApzP4tJ2uzM7IsfTTu40Q8F6+tvpcQqqV0fXCUUztnr/0aUur4IHAH8PF4zYfRyzgzXuMeVD+dvcfTgM9i9nVCOHcUtmaTEFvQ9SQnwJAhf5RSY+usLOWSHahK3KaWJuelqARiK0ONuteHL0cZr5dKPpXz/jZ074oM8U4UyXsIGbu8Jn+XmcCOCcJg9naUAdtK6X31kpqPmNmxEMJr5+Hc6lKX0ZTtqEG+MGAxmwMDUzjy2itc7gOaPRPRiHTe3cheXYZIZrqATVX6YVzcMWtCuv8SkqNvKIi0ATlqwx2CuR05eZUGbVo8djWg7HqiaOByubiTXy6NDNXlefvLzj3zmSuDcZ/lYNT7MdyxBdnrGShL42VvDZT6hl6m9WfAl3LOIY9dLCCbvYo0y2gOAltdJLKIqcC3zWhGPUBXUGxbQjzv3egZe0BTlONaA6dQMLUrnvdWRMpUZI9OIp/nejT/b7zJM+hdWIqucQAFOdvRGvJ5f4Pofh9Gz2AJel/2o/VfXp7biTKQ41LqwGb05HnkpFeSOagfRpEPs6moUXFx/KwDvYhNSHG1o8V8Ai3EvBKnPmAuZs0F0XoQBeMMilPx+3Dn2WwQZUb6UWRlCyFsxewfUDRoeTy/ncBGQshL2ycxm4XmHzxOonn075xEBm0lUp7/EoHUv6CX+fp4T2aSepbK0/qH498+jNk/xl6RkRWzVuCTSFnsLjuHDuDP4+99JE7+XoZGudoobfzMMtBVkhuQs10pUtmF+rdeKvv7AmRcKh3Hy0ouQTXjn0fKsbzHaSZ6TuOewcz07t2CsoV599VLNt9lZq+HsVHuWJe6nCt5FvWCFGZEVvO7vrv4D3uRjplNIgs4jN4VZ616IoThkbnEBvlnEXvW1LjPbABrEmmkwbAcshDoi5Hvt1P8vi9CjmC1mTU+v6dWqtxXUI9ruRwmAYYWZC/KGTItns8m1FOykHRP8q7BgcE0Et21k/t0kMqTWnOO1Q0sNWNBCENAgs+vK5ftKHvVgdaDEzP4caaj+70c+CDKsLyJfJ08AN2KwMoUdH+KeuUmoXUwSHEvblb2ocDduAM2ce3egUrdb0a+os93GkSAr3wtTkJ2/+vkA3lD97i8H2rcSK2Dpepy9rId1T4WTbJ1isbsxO/3Imffp/Z6k/lJUlP/bPTCH0QKv5zpwxd4pdrkW6j88htSSv8HKs25HIGKTwN/h9kCQhgghI2EcD8h/JQQXqwKaiSr47l1o8jKSQTSZiDF14n32YSwAVD5XAgvEcJXgW/G/UxG9cN/jFhBpqVDcBTdm+U1nM+ZyPWonncHQ+/zUaRcOpDCARmYtfGcssGFJrROJiHg8NOqA1oly6jev3UEAdzyYIYTQ2SlgaQb/LdY6ULYC/w7CewuiT9L0Rr68gRp+F9J4vcvkiOoxHDeqJxRXepy/mQbiqgvJd9vWHALTzz3KhcfIZWivh5/9pHeoxlUrh6oJHtQ8CSPxCAg3eujBYYrj6ASrQviObpMjX97A/hK/H8lmQM8WWMZGqTsQ7kO2YuupykecyNDwcpM4NUQ2IxAl5dxQ8pEeCmzkwXtQr6DAxEvV8oCk3YElsrFy9vyruEAuu+zUGZvZTy/p0ml2AMI9E2Jnx1EGYfNyA9YCnybNLJhSjx/n/s3G2WMXieNAchKM6k3aH88r1aq92UNomGs47L8KgT6QuAB4L8iYHgEAcA8UANprMRtDL2HLcifeJLhZz7HjNQzNqMlYgG7B/HVr0Qv+DHSBNge4GunG1zNpiEQ0YIW4gJSn84gSXF1IgWxlRQZyUayZgO/q+Igz6RyxP5iEt99OdOVD2X8EiFUmuJbJFeQIjiHUbmTZ58a0H05ACzCrKkk46JMyWdQD9E+UuRocTzf9aR63X4UdRvZ9KqAwq1UHvjXFz93ILofGelnUaSoCRmTXeh+zkRK5X1x/7uQgehBoHIOytKtRwbXI2+1SLlxPEEaWjgPPet58TyccWZ//P2o9hD2Al+NFLCz4372x79PFHFmnmrig83qUpeJK7JfP0bO8U1IX3j1QAOw/jYe+/EALbcjh6ncqWpC+ncNZ66Dr0UAxKPSJQM6UWCsBenUSvTKQyQEBs34ISI6uIU0C+RNRLnvZAfOKpbXjzEd6fqKQ0bLjhvMuBv4FKmUezqpt/YipPvL7e5UdK2/iv//OaLjnhfPOSCb44QngWTjj8f9e9n4fhKwmob8iZp7JOM1vAj8n/FYJ0llgyfjMQ4gO3+ENB+mm/QMe4CbQ+DbZnwJlbC9lTRM+zEEPLtJwz2dHMTLvvviNgHNDTphxn4EdhyMH2doELAd6KqB+W1MS7zeQeRTVmN+PI7u5TTkl/g9PIHW0pPj+X7Ugc1oihr170QK/mr00vai1P3rp0vQJD4LZCcCPs7j3oEWIiR6y6PIiHgjnUsLAkPVGjP7KK4JbkMlSIfIrwU+HLf5A+CeKsfJk8ac43aTz8BTvu17UTTtEKVgrgcpsstJE6Th3DTEuREqqgt28ZrfRvRsHZiuQ6V93rw3BdWgz4vf8ebO96FntBYZh2Zk6A8iQ7acylSonZSz0kl2oXvkRqQ1/nbwPCMeay5wK2ZrT4NvAZmJBGay0ktt+tEdvLrUZWKLgko/x+wx5FxOQ7rjdUJQxt94ANmEtyFd4g7TAOqreWQY2YzTEpm3liLnfC8K6njw6zjJkWtH+nNYwEaXxymUGdkYh01a+RwTM+5EPZ7LkN31kmIvj/taCEPKuKrJMeB7wP+ESsqakf71HpFF6BrdrrVmjrUnnvshM/4b8PdIZ7tTehLd/y2kfpqHkN1cjCiNT6LgjCGb8yz5WY5cUpg4sPQPUdbIqyJOkIZ6X4GIZx5maGDNxeefEQJ9CADnMtuZ8X0U0DyI7r/7BU5l/Abw2zjQdSqq5PBe1Yb4vZcyf5vFBCifjrIXrcdq75iz4v03VHHRhp7ZthDGvz2rA5vRFtXi76D6jI95pDpYV659JAYVr3V2CsWppCZNQy9rO2rer9bI/QLq08hu5/NnFqMXZTLFw/z2AVdi9os4pLK6mM1BxqEjHuNV8oFVI3rxVgD/C2YnkcLbjMDh65RO2nVxqtFV8bpayOduF93x2fVI1Drb5QRKtXvzaxsyJBsI4SBmq1ANefkQPFc8/ryza2c6il5Ni/9uQ4BpDykiZvHzHw09qzCI2VF0f/eRjGjWMM5CpQmngE9g9sUcgDTRZBMyiJXEa98nAlHChBIzJqF37AakG3uQw7YxOk51OVMJoYeCrEQEB0+a8TzSy5OQPtpaI0lAkZSXv/WSGL+ycopamCSrSNHE9RA4YsZXETC4DgGFbhTlfq1oXk2emLEclZqvQlUHzag0601SpgOiw4/m//QRaapzzvFl4Aeol9ed9uPIl7gI6fb++PMLBOJmk9hVNecu357NRE5vXpnxO5G9ejmem1d4GMoGbSI1sxf1uNbMZhcCXWb8GwrG3UACe6Dn8Byyg58hDeNsJ/UNtaGs4uPIB+6iOtvdeJHfAP9rlW3a0frYGUHMxBnSHKUObMauHCv790lSvwyUOp4b0EJdRUr5vwI8Tgi1NJ8/jxSsZ3iuQIruFFKKU5BTXsRudoo0n6UysFFT9gdJFJXTUObhqy1sOQAAIABJREFUgngdWeayFkS1ewEpDd2ImuQ+SqIR3UMp04uLl+rNxYdVpvOYFY8rWkSzQyiqtLZmcCbpiedRTgQAehbz47UuQODtIBpQ+quSLTVL5u0oipJV/tMQEPKM0CJkkPwcjyADOyuew0F0X1egNbEGGd/nyKPRNGuP330eZd2yWa1mtB58rs1htB4uIL8G2/fp/USN6P6cn1lCmmPivWu7a+z5AiCEsNvMtqDnV5SVWgD8MhSTctTlPIgZc9D8penoHfDyng8Dh8y4I4TxT0c+liU6+CNZ9ttHqg7IAzQuPmjznEkEFJuopAMrSOzleCfSt0dIWZVDpHl3T5KCmNsQSHwjhOIhpiFwyowfIDt0Nann6BiyGYeB+4HdIZT21JrxIBrb0MnQ6oNZSJf/JOdaOlEAwcHDNXF79wl8/k47xYMhidvV/NzizKP7zXgI6fhTwEHPsJnxPgTYdiKbekM8xsn404BYTH8M3BlCxTU1nuRVBDBXkW+3mhDYewL5WxNS6sBm7MobpOFSx5HzvgwpKXcSG9ELvRm9qP8PUoj9hcMY8ySEfZj9AJEBOCV1FynzA3Jur0W1rnmpdqeSLBY5vH+NFI7PldkZr3EhaiBsJim465AjvwMBtYH444M/r0ARpg2oz6WD/Fk6C4B/Os0Kp8zIx+P57keKrg3VKN8eqaFrK7EKoR+zJ1AqPgsi2xAAm4qe31FkXK4BbsTsx4SQjXrOQYq/HIguoRTonELOtoOUa5Dh8MnRl6Aoqafm56B0868L+qwuQPfoDVS614QMhZG4/3uQUfC5CKvJM+pmbchgX0s55bXZegSwJiNDswbYfk4Aj9ky9CwXoWcsEgCzZ4HfDOPduAe9E0tRFNUdDW+SXYMMRF3GiMQJ9Z9Gzzw7u8od407g02Z8IYTcQYt1GYMS+zgeQ5TDWytsOpnqpdfnW1Yje7EV6fNbkV7tRzp8KiKkyZIsdMftCoENnAaUPzDjEVSSN4MEbHbk9U2YMRtljNpJ80+2owz+SWRj7wsh11H2sjdnpWtkKDAKiNDnQvKBTUM89jOVri1PYhawRJ+bMTUez4dsnkB+y0wUJHQippPAAyFU7UcZNxLfk/8d+CKyUYOkKh6fc/gQ8ifXnpeTHAWpA5uxKiG8idnjSAEOoki8O+nOANNGooV8FHiwRhatvOOtwWw3cq47SMww20jAajJSyo+WfdsZV6qRB9zIUKKCgErhTiLgdkM8XgfKdGxC2YTyFLZnsS5BzuVjyMn36dDZeQJ3EYLAkqL4H0dGpAVlIJxFZEe87r/C7F+G4QA/jRSpN25avI62uN8ZyCAdiT8twIcw687M1vFm0cnoeQ4gMDSN0gFuA6Qeq04ECFuRwt6X2dZJKSahlP1j5Ec6W0iNntla9TxpQve9Y8gnmuPzV/F8spTXU1Bk8kPI4diF1vCNwGuYfX84mZSKoqzXingePZQ6tk3Au4E/x+x1ZOzWImKN3P6oEMJhM/syAmq3IJBoKBL2XeDlcKbvW13OlVyB1lxRptoHRF7BGThSdTmvsgY5/K5ns+Kz2H5HbRT550TMaEM6aCrSt1uyjrMZDcim7yWVmk2jNFjYg5xSzziC7MaiWs8jBPYDv63hfC9F1Q+nENDahPTcImRX7giBlysdKv6+GNmcPHbOg+h6r0K2PluN0orswSMhjNhzm1N2bv7vgXiOPhZjKvCf4uDRjSN07LEga4D/D5VSzycBHCcbagbujVmvCSl1YDO25bvopV+MnMl+5Bh2kuokfwLccdYDEeXwT0VovoVSRpN3IAf5WDz2NEoV2HzgFxVnxJg1Iwc7r0Z3EBmk1xBAOB6P9RTF5Qx7kTG7HDky/cihfh29uE7Ve5TSqPq7EQDK1vQaUr5r4u+lCDCVz3zJlxCOYfY14AOkptrFyBh1xP1mqRN9iv3tpNK7+ciBhpTp6EGKP+s8N5DAyzIEhGaiSJmXAmYBTAcq/duD2TcIoby/wLf1wayFVxm3mUQ++HEO/SyYcFA1gEDjhchwuhFfDnwMs2/GXp9JpLr8E8DWquDSbD7K7F2NAN+VKLuXfR+akUM0J+67B933W1GG7l5CyE7jThct0PWYmT2J3rkAHKvPrRmz8haqz604ELerA5txJJH16ZtoLtgK0nT7VqS7ngHuz2YlYmP7tUgvNCG98CQaDDpiM80iYLkdBXGaSAMRzYyXUMbDM+idlIIvZ7XM6pRBBOCyTe8jOoPNDA/y7aM067Ev/kwH3mNWcYjqm+h6l1E5KObHWBG382vtReVgI5lly7Njc5CN6ied5ymk0z9lxoMh8NAInsN5k0xJ4i7US+R2y8kfHgghpyx9Akkd2IxlCWEzZl8APoZKrzy6vhW9nL8GHh6hcp65JGXjDYYuzyIHtQm9HM4A04Ic8tconb+TJ1ORg1mpufIYaRputnmyXBoRgHCQtSOe10Xx90vxb0uBh043upstRc353ZRGjUAv/60os3EIZVxqAzaSpnjMDqTk95Mc7DyD1A1cGHuOLgH+lOg0k+7RJPRcGkmlZ9np1z7rJyDjntc3cBQBn8tQtq3cmdtCYtR7E93T8nI+7+06gDIypfdFoPUtKKvncwdOofvvfT8g4zyfZNR3InBzIWYLkRLOsvqdjAxMj+SSFZhdj6Ys96P7PTd+fxUCWU8iw+l1313x/Jah0gyf6fCRSEqh0osQhhjooOMfGXIOdRlr0kl1YpZeFDCqyziTEDgCfMPsdO/iZKSvN5aXFJlxI9KrA0h3DSI98JfAG2bcNYK9FX+Csro7KNX3HnybYcY3GBqoAumhxZQGCwdIDfEg+18pc3ImchOys0WgxZnKVlLAUBYCx814HfkHeUymIDvWi4KXryD74X/bPpIAM4qX0TtYnIQqBLyfyaUR2ef9wDvM2BpnAo17iX1gj5vxNNJ1zci32DueaZxrlTqwGesSwkbM/ity1i5BL+MO4CVCOFTxu8OTStH6blR+tjKew9z49xOIR/9JqjdQ1/oyeablCKWKPSuXIUOwARmNyUhhOZ3yW5Hj+hOcvUdlSh9CBqMc1EDi9l+NQFptQ97MGhC16e3IcB4lDVW7EjndRQMrTyGg8D5k3AZQ9qGLxKK2AznmM+L2+0gGsJXUDFjN6Q6Irvm5knLFEPowexhlsjYhcHeCUuM8AxnVOQhUl5cM+Nyh2Qio+IDPhZTWVPfHbbLfPwr8J5SFKncKmhChwvTYk5TO28ynVe8kGas56NkeRc/vJjRUcD6p7nsQrZEOUjPyUjTc7EWgIZZkPkwIPruiLuNHjlM9gOJsdnUZpxJpjgvp7c24GAU9ynXKIZLD/iHgzrM9FzMWoMDOVobauYB01AXIHmxnKGvb1vi5s6CCdJ+v0UbknI9YVsOMJmRXKo0IAAWlrqcA2ER5DPgIsgE9lAI3L612BrKBEM6ehSsSMMwnsYDuzrDE9SB7+Mfo3k1Dtn0byVa4jtiPbEIP8hsmBLBxiaCxthI/BSiXonvVh4ainw2T4XmTOrAZDxJCL3K6XjyHRzlAZXBzGEX7DwI/RAZjbw2AJvv9Y9TGavMoAgM35Hzejpznrrivl5FDvoI0rGtX/P4vM9ksn1zcS6kBycrxuI85lJZUVZLbgT+K2/s+uxDoOIGc68fIn3NjCCwOICW7A13/xfG7R5Ei3hWv7zVK10APcuB7KQY2Lei+D5IUfHlG5gnk6N+KFPsKEquNR7VOonv63Zw+rlnISPqsBadcXYb6GdrjtQWG0rZOQgb/HoY6BQPovl6PonxZg3gbWlPZ9eeEBaD7MRNRYRdFBLPlCSI50DVMQ31WvyaEYc/DqMt5lWfROi5iXwKt1/I+wTEvmbkcA8C+cxDpnkjydvQ+F92jncAqM+YVUBgPR65FjmCl4N1epLP+CYEJpyAH6bGXUd/XkbivxrjdFGS3fgscNqPpTJ57ZApcQJrjso9aCH90LtWCfDuQXZqBAo6nD4uAw3p0jUsYAeBgxmWoPH4WqcTqqBkPo17WjyHQ48RLbaR5fLviuUxBAUy//m7gEjNahkPZPSFEQd+bkB8zCwWutVbMNiDwv2aYbLHnVerApi4uDgayDYvl4mVfTw+7/E09FI+ioZpbC7ZqRYrmZdJgzbmU0hbOQ45zA4oGvRQ/30lpnfKS+F03WnPivrfFz4oa51qRoTqB2UeRoiwfnioxm0xiQ8s6+ztR5ONIvI5LEbjJSidy1FdQCnrWx+tZjpQzyAjtAh5AoMeV+S5SiUWRNMdtiiMvAiq/wGwdyhjtQ4qtNZ7LJlTW9foQIGvWiLI9PjfI74M3ax5HytLZ/MoB3goEtIrWU4j7fQsObMymxO+VR6K6EYD1jFx//H9WITuwOoWAs5cnzEAg0CdTHwHehtlWQjgjWte6nBd5EQGbogCKz4PK7akaixIzAm9Djpn3Y/Sa8SjwdB3glEpk+VpI9eDUSZT9P1tgs5xie+JyHDn9bcCDwH9E69P16Sakt1aigNBepJMa4za3oIxCvxlPAs/GsryKEumY34f0ZVbHHkb2PEv0kidFPZWnJfY+PY6c4/UkVk1nA4WUWamZAjzOomoD+pzB0Iyb4/XsYygL6ftR/1UvqubYgWx5B7r2gMoXN6PARtav8M+bqJztnYjyDgRqBtH9CWhdBLS2/w54PrLFjgua/DqwqYskhIDZj4HPkmaWZMWHgH71NKgR4cAitI4OAm9UJBBQWdilSHHvolShdqDI1N2nIwNmdwGfQiDhAFL8k5GybUCgo1w5uXhjoEtT3H8jAgytcZ/uiDcgw7OEVBZ2ASp1O4LZnYRQHgVeGb+XvY5J8Tz6SCDRufz74r8nxc++j6JL5aVxB0gZNC/NWwLcFa+pAxm7buSM307pADLidbaRmjaPxN95ZXiSEHYiUDacKczL47W8Ev+dbczsiefViwDpHoZG0hdRvT/rIDLMLpPIz7jtRlFPz8j5fIhs6cdU5PDMjtu5Y+HT0V38/G/hDOdV1GX0JQS6zfgO8An0Xu4jlR/Oib+/HUJhP8CYEjOWISfYSzVdx7UC7wGWmfG9OrgpkTaqZyIg6eizlbxMdJ4YsrSvmXEvoq4eQHr8FHKot6CyrV8joLAK6SG3x67vrzXjayHkVgLoYMrwfRYFt8pBXgcK/BmVKaSnoWGe1eRR5BTPQno+a4unoOz5HbVkQyKQfyvK5AegwYxXEVh5D0MHWINsTBcqT74//u0IynQZsj8n4v6OMHTGS0v8/HQA0IxWkp3oDqHKjL7xKGaL0Hrqjr/L+5D2omc6m8QWO+YHHNeBTV2ShLADs6+i2uOllLKGdSFQsw2zTsQAtjx+7o7Dccx+Rgj5Tfea+fJtFH28MX7Hsw9dwLcIYWNm+8OY/TsCF7fGc5qCXsIXqBxJasAjL2YXo0jOZQgwHESgZRkyJAfitficlgcz++5Cyv2vMftiGTXwbFJ0pwMBHafjbkZRQ2d4W42Ue0e83lfjOTXEa8q7Fo8iTQa6SA3sKVIn8DcDlWtNppQAwo3ZtniMB8/B3JhL0T3biwDjlMz57SexF01DJW9ZpegU19UY/crP2csJy5mE+lDE8Ep0P5tJhAiQAOHr6H55RL81nnO54eoGVmDWOh6UOeCzokTdnZdl/D2Q6Dj+K9Ix15LIL55Hke5xEXU0owWxVh1i6NrsQ7rrUrSWnxrdsxvTcoIENBqQ3nWwcxBOzy/y9/5s5VVU0ppHWz8V6cTJSJ+cAAiBZ83Ygkp4r0Dv7E4U5Nkc/74SPeOsjuuP280CPm7GF0MoJNl5OwoC5fXRHEXVDm9HFRR592EWul9VsywhcNSMr6Hs/RWZjwzZ16/V0ltjxkoUlDiBgmCeoVyEQE0PxT0j89AzuJBSIp2Xkf1xUoFFqDKhFwHGN1HQ46HIKNaGygZvQs8lkNjtHhovQZEa5TqkS5YhsJjXWnAM+TonUNZ4zM+/qQObupRKCNsx+2dSKRdISWwjhFOYVYoCTQI+ilkjIeT3A8lBvB+z35KyPUeAnblOt5rXnkep0AakvP6GyqBmElLcezC7EGV9diGF6OVVB5HiXkbKRu2In5UbgsPICX8LcF/m7ydIzYm3xL8dIhmiQ/EaL0ZO9W7UhLktXnM7cvybqeyYzAJ+VPDZi8B34j4uRMrbe0ba4rUcQsZpOCxvtYoPAz2BIo3XICXooMxZ2bpJbFUN6JpagZ+RsjpF0km2NjuEo5i9hoBjecTyDWQML4vf2xp/L0Zr4AVk/NzZNfT886h/s+UJtQMbUVAvIGUyN1fJZJ69mF2CjLEHJAYxew6VjXZltmshDdU7OIweuXElEbzcZ8YvUDS2fxxmNS4mDbQtkr3AbWY8i9b9YhRgaER6e0umqfr3RfbHn4uQTiwnodmFshTNjMz0dS9/dJ0C0ikezAooM7QB+JwZPw2BPXHWzAPx57RkqKP3UlyiewC960spHSXg+5iC5sZU6jU7QMq0d8X/+7DqTrSe1gBvNeMA8HoF2mdiRuP7ZjyAdHMjsj07a2Hiihmmj6Nnl7UHXsLcjOz16+SX/rUhuzqH0j7aA+hdWIx0n4+HcKKRFvRsXjCjHQ349QoD1xkNyKZcEufeDJnZF5n6rkP3vTke8zHE2FdLpqojc+5dIzU4NAZIJgODIQwZrn4xiQW2qAWhF63jl+N2dWBTl3Eo6rnYSnkvjFkTUrht5EfZPcryPsw2Vhy8qM9SJMisMYKQpaR5A6+WsHLovHZjthkpzrzaaGdL+TFSbH+KlGIvUtK3xu2Ok6gelyAldpxigoZ9wHWY/SpzTm+g+tTrSP0kWTkZ/zaAHOfylP/x+Ld3IQOcF9HyYXT5VJ8Cg7/BbCPwF3FfU5Di34wAxdPAr86RE7sfRY1B9/gJBBI7SQM9jTRHZkr8/1oEhCYDfx3Ps8j4TQV+UPa3R4HPIUNWbjS2kBpwfxvP4yZkbNzY9ZCiVBvIX88tcT9pDar88hrkDAS0hn9HCAcwm4Uymcvi1qcbWzH7KSGsL7i+sxMzr5E+SIpmNqJ7fi1mX0dG6y0osuylef1oPs9TkaBkwkkEM+MN0LispFLpqKQXBQkuRTNUPBjlBB49Ztw70edWZCVOX9+Lyny3M/QezkcBp/uo7PjXerz90Zl/D9IjLcjOeIZoBgIQL8R/fy46x0WMZHMQEKrGZtVHyuoAYMZkVOHwnngO++LnO8kPzuxFWe4DSF9MR/rOECC8OV5HM3DSjF8Cz1QCKnHw45kMf1wdfxfpIp+ldgH5LG3eKzkZgYsDJHDyety/02072YPPiwMFGT2gW04ZfyruaybwETP+R9m8pBtR708fiWmtAw0/3WXGHUU9URGEvh3ZFZcGMzaimTNnlGGOQPFGZPsa4z53I9u5Pp6/kcZ4VAOf/dTKFnuepQ5s6lJZVNpyFcpIzECKbj2KxJejf9Dib0QKtzYWN7MlSAFMI81UmYQcr58Rwgtl37iH1HuzD72QnhmZikBAT/x8FslAHEIv9dWkSBoow7AZRVeKFLIPXJtMcnR3wumhazsLvrcYAcRliIK6HFz0IXAzEM/3ZPxbS/zZBnyvKu2i+n/+O2b/QhrceRJl2qo5R2cj61BpYVYx9lC6NmahYbM/xRv0HfQqC/cieiY7KK2Nb0T373eUg74QtmL2fVTO53MqBtB1dyIw+C1C6InHeRSVqN2KjFc/enbPUBwRnwP8NhJfNADvjN/3Aaug9+L2CBCuQmukPJPZBnwCs7sIoVI9+/DFbCW6/1sp7TsaRE7WVAQcj8d/Z6OQLfG7q+Lw1uJARF3OhzRTW69IO5rNsp+hznA78GkzvhlC7Y3b41mio3gpercvQuvcCUpaiU4esGOkZnqEwGNmHEeBrttIc1qaUanaa+j97Eb66QPA/yjYXTYAkyfTSBPlzYwXQ2BvzBh8ilSSfCzu53JUQvQ0pbNm2uO2J0PgEeARM6Yh+v3jMCQr0Ywa9JtQUGqk5Rry2UNdDiFbvZihwGY+8jkWo3Ofh/S8D/9ejmx8MwJvB0hlffvj3z9IIuUpki5kpxcRwY8ZF6HnWd7742Xj84G/iD1RJevNjKnAZ9Az9dI74nksBT5vxldqZe6LfUGXoAD025APsC7uOyAb8AngSTN+FnRPVpEImfLWXRu69y1UH348JqQObOpSLGZzUVp2Mnr5vbRoEXrpXiDfoe9FCqY6sDFbgByvHoY6hC3AhzELJaVtIfRg9hUUgfkAikoYcuTWIYX+CaRkmsv2eRhF8aejiArI0G2iepSptMFchAsvIyVSzuHfjMBTH1Iq05BSyVPcXcjg/II4zA0pxPXAjmH1xajkafSa3UPYj9kLqJdhO0OjPu3x5+HTIKP0+6cw+xG69z4E1iNJg+hZPZRDMQ0hrIkzZ65GRtEV790o29eX2dZBzDORrz+gNXJhwZV1IgP5fPy/AEwprTdxmyZUnrmX/DXfi9bmBzB7bYT7dbzxs8gR6kFgbAdDSwi8D2shGjD4wxE8r7qcvexGTke1XsKLkC7OKyU5joI/HzTjH8ZhOd6ZyOVIf6xHQP4CtMYN6dUXkJ64xowH43eWIz3VD2w9k0bxEHjBjDeRffRASx7ldDew1IyFIeQ60cfIJyNoRXpuLtKNk9F7/7dmvBGPe4JUcu3MlN2k7MvDSLd5efQ0YGosV3oMZW3aKXXuW0mMpduBd5mxNqes6WxlEsXlUCAn3HtYs074HOQDeC/aYVJPWhPS3c3x3GcCDzEUvPTFzxqonikbRKDJszp/SGVqcV+Di3P2/S7kF5SfT0DvbSfwYTP+tRoIj6QLf4myKleSgo03IR/jWdKMn7fE838W2c/taP3kPdPJpD6l+3M+H3NSBzZ1yRfV4v9l/J+/wM74dRiBgeuQAjmbQaF/QmLtKpd+9MK/F7P1QxxVs61xm/tIJV8u3SiCvhQpxHKl4IPaQBGVSVXOcyrKApS/+O4wd5CGUxKv6WUSRSsUM+fIkQ+h4tC5URGzdhTtvBIBhTeRs14JYN2H1sNq9DyOkHqP+oFv5jDKJREYexCzx9E9nITu3+aqWQTRTw6pU6/yHWXNzO5GGZ+VlNZci6wB7owgehIq9dpBPoBwBrq5FEe9PAt3MZVZiGoXs6nkG8usdMSftgrb7AFWxzLLkXZW6nLmsg5lCYvWFMhh6qOy3jiOIt0XwO9FSdpykmPbTSKbKS+3mYXurwdF/PNTsWfpwTOYaTIL2RUvbTUSKyZIN3rZ7RxysgOR2W9z/NwzLI3IQc0GxxpQsOIoCgrNAX4eP3MWvUXxfPrgdFO896IcQxH9lxGAvhrpvlfjPqYh+zUvc3pHkP2/jJEnrDgQj1kUYOxCzv4SSomNrkT3YDKqihhEIOUYielsWfz/BoozMoPUVmp1ihgwjXTaS6itbPCK7HYxW3NlhfOBCIKpoufNmIECxCeQT9JIWidHUeD1BpRpCygI9wcf4e5//j5/8SwqY22OP1mm0BnIBxhAz35cZH3rwKYuRXIxUjLlU+I9WjKInMELkdHIShu1DOJST8JyKs8bcIfwIob2mdyCXuQiZ+w1BG5mUjmF2oWuK8voVXKmyGDdmePcO23khng8B3/H4/eWofesgXzWHJCxOv9zNdTj9HEUnTuMnvGVKIq3HrMf5mYbBBTuieVY1yCg2I+igxtqLnHSdvm9ROdCBI7vRNHcq0jzdtZRSl1+IVL4RRG5jvhZC3JiitbaCWSgRqoczbNblcSJAsozl1nJMg+NRDN1XUZAooP7KHI6yjOFIGdlErUNEx5AwPv3AdgUSfm7shzp/NcpLRF2EDHTjO+cRZarE4GFjszfDOnWan0TvwE+j2zGCaRTZ5Cc1Zmo/NQBnAcv5pB6TzciQOvDQBuRXX8WrZsONJ/sJLJjC5C+fy1+7+b4WTZjOAmBnT8z4+mRKuWL8iSyP0XAJiAQsJFUZtaO7s0RdN2/Q2t9HvIZZuBl7bItlda/l8FXkyZSmZ6PHpiGdPu0eHyfC+hrpw89B+B0/8u7SOXze0l+SLkMUj2AdX08r8MowFhup33sxGy09o4Bi+/hIzPhL+5D9+/9yA6C1t1gvAYfGfGt8cK0WQc2E1kUfXfH5sAwF+X1pChPH3p5BxBguRApuyPIKXyJ9AIPZxDXDGpTJCeRwk5i1kb1aEcvilpdhQxFnsxB/Rs+NG0yepH9vHzy8+OotMGPPxOVPCxEBvIkUgLZEoZAohbeRn4jcCN6D/OaIUdPVBL4KaRcsxz/fj2XIUV3d+E+lJU562bcURUBVZ/fUyQdVF6nIfO7nIEpK7U0aA5HjpOaxIt6MXx9VaO1dfa3uowt+RV6NrejZ92LnlMriub+AGUdqslIr71RlVgqtQLp4gHk5O1EjuUqlI0C2aedVC/hm4V09xMMDVh4ydVKlL0eDgvUflKWxgNv5fTAbSgA9O2inYTANjO+jQhhmpD+PYmcYC+rygZIJiGndgUJ2PQiu3UVArVLSCXSR9G1Z/tojsZjLY7HO8ZQYpYT6L6uRABnIyMnr6HMoxPmlMtMBNC+iIJIlyAnfja6F9kAZ9YWtQIfIZ/KOCsnULaq0pDyFnRPHCANoHvVFv/dRyIw6kMgsiueQ48ZjagP6xa0RuagZ7cC6ejnGKqrK85JMjtNjuN2u6hHy2mdHVQr4KUxEr/B7CkExK9CoPB43OcTKEA5bghm6oZsIooomW9HfQ8uAzGi/kTFBaom6SsQq8oUpAy8b2MbeqHnopffIyuNcZvTg7hq7COoBdRAfimGl9ZU28c6BIAWUzpl2amje4EfEkI3Zl9ECsfZWQwpgbuAdbGnpglx9d9Iojk+gpyL3ahBM6uYuuI2zqGfdS5aUJT8l4Rwvrnxb0OKryi7sg24ErNHY8ncmYn6WwR+qxEimM1Byj8AbxLCmTDtjIT0U3kAn4O/8mGt5VJrdL02CeEEZr9DhqjomZyIx91HZee2geLBXG48AAAgAElEQVTMZ13Ok8QZJQ+a8QxyoOah9/RVxHY1BUV+qwGXJsZb0CGKGVcjdkuPvDeQCACcit3ZF51ZaiZyjouCeatRxqNSNuYAsqM1A5sQ2G3GTlKJdZGt7UYBr+cLPicEXjHjH1AAbTWcZhvbylDH2+3t1LK/H0clY7OQ3dqPCHQOMnS9HIvbX4HuZdG9a0V+wG2MILAJgX4zvoV6H5eSSslaSPPr7sj0Pz1ldjq4Wkl39aG14FmuPOlETvyPEdvmSYYGIpuRvf5BpkTxVmTPjlP6rHvRfXoL8Ej89zq0Lt6K7MBBBDT83CfHzx4p21c2Q5QnkyitKDhCmumTlX5S9tDQ+5MCsaqYeCL+jGupA5uJJsokfIY0w8QdrWbU5HZxZEAaqnDNGtFwzqvRS3As/jjL2RLSwr8s/r8DvUQBvXxfI4Sqg7iieFSmUsQZtE7LHUKno2yisnEKqP9iH6oxtcxnLyLWKxkJ9Wvci9kvScMce1BG5uOYzUZRorb4XXfMu5CSug4BwifQ/ZsUv/+f0T29NB7fGUj6gJ8gMHT+RMx3l1E5awG6pkpOdKVjzEJg8Hp07Q2YvQo8TghbyrZdQJyqTraW2mwtAoGjDXC2kggN8pzHfhRBvZSh0VmXdmRoRroU6Ank9JSXUTYj5+A65OzchN7l1xjKPtdOKVV0XcaYRArdJ3M+OmTGBkoj9eUyjXzGtLOSmEVpAU6cK1ICM64jRdsXoWsZRGt5OXrfHiI5oQeQXfCyqvWUOokeTW+ieknoEWCJGS3D7LV5HlFN55WbtcXzexxYZMaCEIoHFEcn/mkzXkB6N8ualQ2kbEG6uQhI9ZIYwop01CDKeL2V4ooLd6BfAxab0R5CYTBs2BICPWb8O8rAXYsCqPtQufumEIZkXRysVwP2W5COW0SpX2SkapDvh8CbZnwDZcp8YHcgDXi9NwSBUTPmIiD9FAIwXtni4mX0N6DMa2/cbmvc5xHkO7htOBav90JSSbaDpko+VbYnxsv1VuZs10gCwLOADeeAAGJMSB3YTDz5MGkOTFZE/asX+x2IerdcbkZO0hb0Eroz5Hz8U5Cj9Bhy7LvRS/QIiiDtGiaD1zHMnkcKrMipno6MsoCN2WKkeC9DL+81pPKDboZGmTrQkMJXMHuISJEJdBNCPvONIhfHMZuMytMuIDXEX4SUw9sopQr2c7gS3a8XkTLaFHs1XsGsEznrzXF/b4yRifYdpGGUlcRnZgxPzJYidj1vWvTBmIuAz2F2HyE8GbddjID5CUrBrOF0nmZfGVVwE8LBCKoup3idHkOgxSPFARmsSfF3E/CNEZ8lJFa6rwOfRFHHw8jheSsq9dyM1unqeE6rUYTzGeSgtKIM7B25zHN1GQ/yCxRlnsfQoY6daA1+tcKE+mGJGYuRrbiC0mb7Z0MYOTrYOAH+AyjbPhXpn6NIF1xLYv1aSWmfZx8qj16E3r05lN6TtUi/1PoutplxOdLtrSRClaLBkw4g5iB9ACkwcgSBGrens4i2Ol7vsniMXsTO5vZhPerHGED2aFHcZx/SO/uQXSnKyjWSnN5K4kG6qQztOWlD9/MpUjBxxH3IOFB2EzWwe4bAQTPWo4BjEUCci3oH70H9ah7g9J8NwK986GYIvGHG/0sa8NqA7tuGMhB3Jak/aU38/wBaoz6yogU9z/sQ8DxJ6VpchzJfp0j9whegTFgTeqe/kwPosvegz4xXUaD5AFpj21BgK8vC2oYAqdul3xbtc7xLHdhMJDFbiIxApXKXPWjQ5G9K5puk4Zsejd+HXrJsJPgIUtTTkUIdBH5ACNmejOHKr9ELmRdJmY2UyrcjLfA1iMXqGHIwW4A/jt8HKZ/NSCF2IaPejStIZanKB2/li0ryPhbPa2v86+UoenI4HvtmpBz8/vQjBTcbDcQsBVkqNzvfJWd50k9pJqtImqF48nSuqM/rEyROfxefJn0Ysd7tRhHlj5AoOynbfg+KxL4L+P6wzuPs5T4EFJahd8MNnE+LXouoMN+BonIXoPeEuO3vgE7MGkYcQISwHbN/RA7ealROeQxlc7Lr7QpkQN0p8szN3YTwyoieU11GTaJz92VU5nIZKUjhM5V+VikrMBwx41qU1XcdHJDTfANwoxnfCKGw3Ge4spJUFpVdxx1o3R5H9uhi5PhndZM33d9Dmt8xCGwPgUNm/BnKolfKPk9BjurfUkqochW63pfN+GFBNsfp+n3SPUivZa/D4HSPxB+i97Yx8/lJMx5GQOgZ4L1I13ifS0AO6ltJFQbOGLYhHq8F2aMAfAfpz0r9dg3A95AOm0Fy/onX/yJynp09ayzMvvoZAi+LKZ3T1Rj/fhT4cQgcA34e7+kc4rqK2dASiUBiA5XJVOaQrn8LpQxmDaSeGX+mi2AIjfgh9HyvRf6KM7OtQM/4rhBqItTxKppD6PrXxnNYHP9/Cvn7TXH/3xwpnTAWpQ5sJpYspXrPiUfLF1Kabl6AlKTXcp5CJVI3k2Z6uAN8GXqR7zhLUONZm68hxZ4tFWtASuXXhLAPs/logNbueB4LkdE7GLcN8Tx9zs52lDH59hlGyZch5zQLEqeRUrleanYhpY3/A0iRdXC+QYxoilciQNaMlP4aQigvWTmIjFURK5xLO8Nn9LoURYqKaoQ9wvUW9Bw7qQ7Mr8Dsl6NKTSwGtW+iqNztyGBAmpuzjhAGYs/LdSiK2oPem0NwOvq8CLMfnQNwcwJYg9kOZBR3UBoZ3ILenYXI0ZmGCDO2AKcwm0EIlZqt6zKGJQQ5QZH2dR7So10hcHb6OSNmLKRUB7v4MNgO4JNm/NOZzIHJkSsRICkPRrWQwJvPH2knP+jSEUIuMcvzqCKhEpX2snisjeQTqlwev1tOqLKXZMcOUpnEYD+6p6sRUMyW9DUjsDoN9VS5v9ZPAjUXkjID21Cg7RIU/NkY9+8VFgb8DcWUyktRpusZBJCejds2oHub/c5c4NGxMBcpBI6Y8VWU+fCZdqB79ALw22zJVcy6bB2BQ/dR6kMfRoCivCfLwcUA+b2a3SjAO4tEKHAP8EKtJZAhsMWMn6MS7h4SxflW5CdNicd4GHhlJMsHx6LUgc3EkiZqa8jPY9nwoYVZOY7KzOYhBdoe//4y8JXTvSlnKyr9+jlmHklpQKVi2f3fgCJE3gB3LVIkXfFnNlK2s5AybwHuiX0zZyLXMDQa5bS4LkeR0nqZUoNUyViOjpipL0gOdQ9yPpYCt8byv/tO0xmLFOHhuL1TepfLDGSga+2fcrmG6nOOuhCL0V4q90uRObe5jHazu8oGn4sDSdviufSeLr9U6eJHkSNWXuvei0DEdSiD+NI5OsulFNeb96GM5mbkSP0hcoDU82W2HQ1DHb0Br3UZUYkA51wB1JuQg1vkbB1FjtnljEzf4GLy9aiXdYFswhSK5zTl9m7GJv9HUZBiF6XX1Ega6LmLfDZLiIQqZjwaQknmZ3f8nlcM5Mms+P2pKHO0JWebk8gxvRkFE9egBvOVyN4tjds51bFnBh5HgKQV+JesUx8BwCfjd48ifTsZ2f81qOfTsxUrye/L6kTr4LmCaxt1iUD6/piN8fK/7nPswL+Mnl2l8ssOko+ynmK2Pu9RPobWzjPDLR0NgcfN2I0IDS6K+zyFMnBPeand74PUgc3Eki5KU9lF0sjQiM3Rgu8OUEqHuxh4bMRATVYEcLYO+buZs914psHT7e4EH0dGYg+6hl8hMLaK4TviLjMZ6pzuQg6h/92BTkvmXDqQgjpfDF7efP9pZFQd2LXHn2kIwMzF7KuZWS3rUGTvVlKJGEhHzEWG/1uZ7bPHa0DPxOlE9wKvRyDQRvVadnfC88B14VXWuN3Ii7Itec6OZ8aKGnhFuw63YbZ2WP1otUsLlYk4QJnIK9F5vvr/s/fmUXbV153vZ1eVSlKppCo0T2hmEoh5EGAwNja24ySOYzvxkMTx2El33upe3e0k/br7Jat7vV6rX2eePBvjCeJgYzB2zGQjgxGDMCCQACGE0DyWVCXVoJp+74/vb+t37r3nDjVKgtpr1Srp1rnn/M5v2PP+7sznrcCnYkTpyTEY2wSdoRTTpS6hfB2DUxtyOg3JsIlABOeiCG9ejWiWOils0Ail0RpPoaoEiHI/co68AxkBfs9BFOG4AqpGvEoAVUIgmHEXqnuaRWGdA8ioqUcIXL9GZcePK6dXIRAc1bLK0GkkoW6CDLzFJPlzdhzbI5mx7THjr5Di600996P1etZrhsz4FxRhXhPfsRvxtia0xt/MS+EqRxHm2MpFeOL+WhnH7OuWBxZQkUKgm+ogOKNFW9F6lIOHdmCCO+KeeAk5lxwsII/mAd8bbj1cCHJcmdFElEXDaDJ7xtOEYfPGoq1IiXSs9TxqAfaUQPaGsB+z3VTu/FtHKrYbGQmSehkSKMdRIX25+o16CpHTlpCfMuVNEt37cTnKvx0OdZFqJJz2IoWweH6zzHoucMcYKa210k1I2DkE6oUoPWmQBJf6O8DcaNzsiVGbHyPP4Y0UGo+PAU/kGrNm8xGCzFxSLq/mx+xHSJFfQnllHxKy3U5kWFUjRfROP1pNdYP2GJqP6YxNxKmDytDULUjZOUapp/Eo2jO/jtmOnJTFCXrzUgMJ0bES9ZIi+zWRGctJ0WUHq/E+Ka1on2ZBVnrRPm9Ge/0opYbNAuCFGMHKpag8PmbGBmTsN8V7b4/3dsCNSpQLqBICeyO616+j6EjWEHsNgfccis+thlbXQaksmo9kXLEcmpL5/0GU4vsIhXRW/HxZHNciVLt6vRk/CIFdEbTgDjPWkZoX96A0q221pKCZUYeM1RuI0SUz9qEskBf9Hmasis+fnnmfRqAr1jCVQEqbMRkp7ieGavyMFoVAvxnfQmA3C0nAOJD6360npqaFQLcZ3wH+L7QGrqvsi99tideOuFn3Gz3VrBpNGDZvJArhRFROP4CYZTHzmYoY5J1l7nAfCZGqGK3LkEL2yMnaBrMZyEu9GhkeO4FnKvY5UWPN9yKh4YLSUJ+dnyL430LhqdqF44hp91A+5a6R5EmXkWNmwzQynkHCNSsY+1D+sUM7goRTL5rb+Sh3eygN3UaXtCarSXnpFyGjpliJbUDM+FOYfR7N2wwU4fkyep8GoLNsjZLgrz+D3r+4LmYyQpQ7hOZxO/LC5jHcech42hLHMZXyhtAcFA0aNfSlUaRJVI+WQJWGayOkV9E+9eLeYlqGp9DlG4d98edyBIYwQcMkM4y4JyLS05lMvST42koe4CbgcFRql6GzPYiUtx3FKGJmLER84giFdXhH0R5dhZTiXRTyjt2It9WjqKPzfUeSaqNGp1b0aGcjl6441xIV9j46effdB3zJjPkk4+cAciosR9GJ2chhVkk57yNFoHz+plLq3JtEYSS5B5hnhmUiMXNRJGmQUp7dCnzWjK+EIGMrptiVyPMYETgfzXUfMtZe830eIzDvR3wkCyU/AxmxL5txO5JBv48iT3mNTH/PjK+HoHpgM5YgkIUL4jUDZjyFUrdyZYIZM+JYZ6G12grsLINoNySKBuw/onTB1MpAhsq3EJLaYBxHXRxHHTJ83AGwEMmN7wL/8gbgFaecJgybNxqF8FTsR/Mr6AB1xt9T0aG+raRvSPruq5h9B3lP3BMW0CFsROHq+wEwW5Nz3VXA9Zg9AfwodrRNZNaIIgVLKC1ubkCFkk3AT3JG9whCQNuJIjyTKfXSNZGiSU3A4RFETraQPGXZSMVB5HFajcL5u+P7dKDUgqdOMWzudCS0AvI6riQ/B9iBIKahPjtuBDsk6c+BDbmpZ4neGX8Xe0Wb0V6YiQTQMVTUeRzVd2wiGaYz4rOfIoQ+zH6A9sh+So2bVqTI5O2P04F2ofeuVDjtBke5vP2RkZwbDyBP8euUGlpL0Fl7gfLe6EMod3zCsBkGmTEdzd/1ROXFjI3A4yHUiMp4mlEIDJqxHkVzK0EGt6Bajf+IvNLuuDLgoBl3hlCQKnQzCca5mA4DLyIe1o8U0yxvfRzxhP0oBYv49/XAIyMEMKgVUGUaSv25HtWBNiF58RiKSvREA2dfDvqZQ96vRKnAL5N/Jl2uzSLx8hPxHtn5aKAwDasRpSJl7/leUkpsluqQTF0J/KUZdyMn3dZsdCYa62uRLK5HMrge1SodMeOOENiN9kle7VBH/FmF5P1SUqS4mLqRvP31mD53BULNbERr0hu/exVwVTSAtmfGWof2141ornsQ/30HsNuM20MYeeQ/3uNHZjyI9sMA0JFjOPnav4DW22GXB5HRvgxlV4xV/eWbhuzUZsxM0IjJrAUp2R4q3oK8I9PQIVmMDs5WYEtNfVMKIzEN8X4pEmO2EkV29pAf2VkGrCOEQgXU7EqEALO9zJMdnvDvS9Jg1ETy38X7N6Ei7Gzu8nT0/usQY1kC3E0Iwy9wFBLbJxHD30/ynM2Kz1tHAg44WGLInQoym4fmaRdK5TiP/GJF7+Ts8N3fJxkSHn3ahNLq8upqWoDPUWqgTiUJkk5k3DyP0h0cvWgrEuKtSKH5JiHsytz7AuTtm4bm1pDwPIAAIU5PmErVNv07dF7KMdaz0dl4cAzHYUiIvh2dfU+Pa0Vpio9QGRWoDphPCH8+ZmN8g1L0iH8S7d0DiD/WIf48FbgnBJ44dSMsT2ZMQYpmI7FusUipbQX+CL1TXsrlPHROG5HiXGxYtCCnxxdjrUcr4iGVzssU5KTbh5wiTpPis36MFPCZ8fO2EIYISV+GzLgYRRdeKzO+1vjTG8dymBS9d7CV20LgUFSyfxNFMLINcuejDACPoDxb9CyXp/cguOd9aP5XIfnsvL0l/jvbNX4RMvAejO8zB/gPlEZqmpCxMgPJ0OmoxqifFHnojP+/Gngf+RkhLfFeXwU+hYySco4xT5H2PjyVaClyqP5bEmy3R9MGSKh1U4G/jrDOmPErKA1uB6UOnjlInn9hPJpUmjEN+BMKW1oUk/c7+6vR6jn1ZqWJiM2ZSorK3IKYIiRvxE3I4LiDEJ6AYQhRpZo9Rn6naxCSUhv5IfiAGOf1mD2WSVszpPBWQinzGpDSNJgQjmN2K/B7SHD2IUW5BxkeHciDN4CY1hGoCf89kSJK56I0OYe+vhsJn2vjc0FK+S/KRr5OLR1EczEVCZpyQr4RKSI9yPuV5QWO4HUheu/iHG1I3thigb8CrYcL3D7EsB9Fyt1SZHS+gCAtXyqprVIz1VeQ93BOfM5uYMcprl2qTCHsiYhpV6IzUDzW2UjZG9vCfM3RT2NT0UtQ2ktAKEbHqJ4u10z1oukJKqJYAP+78b/ZyMwg4nsNwG+YcSCEXBSsU0IxknAT8iZn+UCXGQ+hxpsh9n/5GvBxkqfdFfnpSOluQfwnL1rSjgyAXwW+RFJSK53pHqSse+3NnPj72Tgun+excHZUAlSZi+RtnhHnPbtmAR834x8Qz7uC1HXeaT/aK4vQOd1FSslrQI6QJ5Bs60ARi2xfn1ZSzeFLSFZBagT5y8yz5lKawt2AdAgHIgDxb08fvBj4ZpyLgAyq58g3WNrjd38rzk2laL/DVZdDs5uJ5mMWkjU3xLFuo3D+6hHAwRQ0PxcCT0Yj7nrEh/OMhINobq9BYENjTavQOlXivV1on1TrRThBVWjCsDlz6T2UP7hzUO3EF0a914ciAgupXPDo47mAZFi5F6taKsZRxARKSf1s/g6lgN2IhPFkFKXag7ykjvLyrdiQszaSt/13kYA+jgTDQuSh2gL8Vbyyr6Tx5niQIlYtiDGWjw6pkenPUcRjgPxajiYSZGoXhYhAWRJ0pNn6k1Ebrf+lSOBdHO+/j+S1XEahN9fzwh3Q4WC878uEkNdfwt+jHxmQL5e95vSke9DeWYvm32vCPOr3nXHrv6M6pIcKPjPrAD5E5fSamYxhul/ssH4B2j9T0Lw8zSjlvZ9CWkXlPkz9SPl6C/nwvuNOEa3qg0g53E2hMjoZ+A1SDwxH1fprVCtwJeK5e5CxPoCMnkrOqzZgaaw7qbUHigEvhcCPs/UiY00RycoBVbK9q/qRoTENzVu5lLfDyAA8D6VldVDKZ9346Ir3ugpFyh245QFgXXznTWb8DXJWXBavOw8ZVlNIwCvT0Jx9j+pOjIUkSOIsNZHkawMpSjIb6R3Pkr+HHbSnFh43GO+fJUPyZSmSKV1Id1iO3nMekjcN6J0Nzf8q5Li5Eu3FSxEfrhT52Adca8a6cUAOc9ldjQJavwkaAU0YNmciSbm8llLvj9NYeiOayzyzmHpJ6QHVaDIyxiaRCu/ySUbFJmATZl9CDO8yxDiOIyGxrUptSCGZzUSh8x4KDTZX/lYhZfAb414/YzYXCZg1JGOkE7N1qC4l7z03IM/Pu5BS4vnLXrQ4iJSRucQ8bPJznE/Ea+Zjtg8Z02uRwOhBDPhKtNZPkiDDs3M0idJ87mNIoL7xSOvxQ8y8E7SniL4EvH6K669ANQt7kGc3D/VsIXI+bMn524jJjGWohmoKqZP7RUihe9GMfxmtVKJTQJdT2WAEKfbnmtF0miAXnYcU5Twl9QTih28zY7N3Ko/r8yyFjYkx45oanzmIzsVL6Gw4FH09Kc0tK2MaiftxvA3f+LzNwGYzmpHO1IWMjv9OZSMO5KhbiyIy5aJKA0imvYoyBu6Kz9hWfBYiytvD8Qcz1gB/jCIbfXFcz6Az3Ap8xowvR2jmg5Q6ulZRapjVxc9PoMh7I4qeO4x0JzIc8pAVAymKV426SLVGvq4XIKMma2i1xmd1Ib1mLsnYchqIY3Sn2tk571VMffE+49FMu5vaWnFAZXCOCaqBJgybM5MuI4Way9E+4FrM1o1yhMHrTKpRA4VpUN2ICTaTGE49Uv6WkZibOk2bvQd4oKKBoqjF1vgzEromjrccVO9OJHCWUGvHYqUKLkHv24fSqIamyJgtRgbXIEpR8PWegvofrMSstAZGUZvvI8/xnyEl9kT8/nYkQM+J951BaV53wd3QOt2CjOlsitWLKPWsC6UzFKc9TiEVgGapWkj+zKcQ2shP4Tu1JHCBbwAfQfvT0a4mIwVmO0pjTTzDrAHtl7cgw6cfKWJPEUKlQvICMmMeqj85SqFC6PzgPOCDZnz7DI3cTKO6UuLvNZnyvSzGk26kslI3gN7pCqqnew3FaPd5OEZpBPE4Mib2IqW2jdMgwpUFI4j1SA5TX4m6kdOtFuoDDobAhloujnU7t6DUsA4KWyJAciC9C/huCBwwYxuKerhBMo1CcByHi/YaSdD6n4Ucjo5k2o14Qh4wzVHESyZTBjEujvUYWucF8d6NyKDK1oQ6uIzq/jSXdfF93QjujeO+mpR6N0hl5Emve1sIrDHjxRDyjdRo0E4BekYASOH7t1zzZEjpe2ckwMjpRBOGzZlG6m5+PtUZ6lh5I/YgplaJaRGfnYoC1SdlHVLUDTHXJYjBvY6EvPfJeRaF1adjdueYernNJiGGWK2moAtFJ7bn3MMbVF6KPGezkFeplyTsByNa3IM1Ajg0AB9DwqU4rN8Tx7E6jr20Fkpz9hRmfwT8Idove0iwor1xjJty3ymOgrQm11FaN7IFCYcZ8X7L43g9faER1dYUr98MosdxgoZBijBegrzAA8jALK1VKkchHMPsy8iruQatRzvKo99VUMckePaPkhSO/WhPXAJchdl9hPDzGkf+FrQPyykHO5DHdhFj2GQvIjvNRIrE8RGiZ2XpCKmOqhy5slV7muwYkRmTELhMtT4qbcixU41qrcsy5Oz4EFIsXdluR/xpMsmJshP4xmkIgduHzl6xMVFMkxFPPIDetTjlK0tnMbSG0svRPt6LlP4GxIcPZ8a0DynuP4lF8vcC/4bUOLQ/8w5TSKhu2RSxeqLRRUoT70RZFsV6gBt796Faqu2U8n9DvOdnKH3ss8i4aUDnw5H0ZsR/96FMg2ZSXecAqdbpMKnx5fyYXrkpjjWvceZClAY7Pf7/FuBdZmwBfhCCvhOjyzciAy6gSuFXgJ9n0ddqoQggsRk5b8o5gxYCP409hMaXVANdP6RMl9OYJgybM4WE0HUDUkTWkAp8XyY/pQQSkxg9Uk+ZdQipZnuZq+YiYZkOsA6OGzOtyFDwaMZ5iDl1I+WqPf5chjwwI43IVCLv11LtQHei9yokGZofIUUu5iKm6VCUj5NStK4F5mH2zbK9YRKtQoy9UhHhPtTF/smyDCmE7Zj9NULIWk0SGkeRYfIM5T1IM9HcLyIfKKAXGVXno/SBlUhJWRW/t4lSwTKV5PGfoKGQDOh3IGE7QIJyvwjowex2QqjtrMh42UF1pfb9KKKa9ZgPImVKsPJmbYRQEagj9r2opXN9Lzr3o27YRA/3GoQYN4ukrDyPlJWRNiT9JXrHSjQHdXg/HdLtaom8D+XanUgmlevEDuKPryBF9mK0r15H/OMcJBNAe+ws4H8PZ10iCpUrxodG2zAKgQEzfon2avm+bXqHdUiefRIZiXn81pCivX4Iw1iC5s1rHeuRLOtDUZxtJPk/D0EQ7zfjS+hcL0UydzGKnhxHUeYLKUybmh7vdRA51byfHJSmVy1ErQJ+gXj925DR4e/digyQDUiJH4jjeSfSKaaRZNRuZBCeg5x7Z5FqfepINTSzkOzdgwytD8b3mY32YdYxuBilvbYjOfQ0iQeeDXw6jucc1A/wGClbwpAs/GyELc8CM9RCPyABbxwkRWynx7FuRHtl/Ei1xVejfVyP2TG0BzYSwmg5fMadJgybM4HMVqEDcQIdsjoUHXDPljy2heQFgWNRqLweHfBLUSja0wgcaeso8M9FCFZrEZO7H3n/biI1swrI4/QkhdCPHej9krKmSMbgKEZxavVQNFA8lymqsggZeU1ovLsRw52GCi0fJuWrn4PmrRoM9XlUT1XpQYrSbMobt0To7O9EGG8P7R9EGP9vIx+AYlp8nwdRpKqcItaLGPJLaF2/jwycVTnfmYmY+LJ1DsUAACAASURBVDeGnJY3QaC1yluvNrRWn8Dsi4RQzVipjVTfdSHljesBpHjcjNkBkrGwnxCKIca9GV21c9tJTif3kVI0at6LzuMBUrqHoX17YeyDMZKUp9fifcvVL01FPPIXOX8bdwqBXjP2U71XSys1OCJisf33kAfe4Y+dDBk1/Shq8FFSGtMgmrvtiO/UId4xFynvWZjnimTGbFTo7wamAZ1mrAOeysJXD5fMmIkU5OMk5TkbgWtG69yEePhm9D5PozqsXRTKHUc/e4oaU+4ikt17kENtHwk9DDR/y1Hmw89ISjlwsnHo52Nz1PPQWuwk7dlOtCdOxLEZKVr/JIq8TorXuoPOGz7vRE6CADwQoxRXojNWh5xpT5Bp1hojJP9ixl7kJNyD5rYb6Q170Xx3x3mtJ+kO3gTXkTgvi3PxShzv+9D8/zKO8TK012cgOZ11oOxHsvx9cV72UJha6s/tR41Dd4dQO3pkCHRFVMFLkZP67Dh+gcqomef4RSZT+40etPYetXsPAg36GiEUp5GfETRh2JzuJDSsjyEh4crgXuR57I+fX4CUm2yO6FxkXIx+GlcIA5jdiaJFWbSYHoTCtKHA2le619sRE+lDQvJcxDScUXgucjZ6chRYhtkU9L43IuYdMNsMrCeE7SN8ly7MtiMBldfvxamFYoQpKfDLSJGrxYiZ+px3xvGeTTLO9gNvxezpKmvjkJtV34BaixKFxpWMM7MHEWO9gRQBqEeCuRu4lRB2oWas1XhFL/KCbUEen6vR3pgTx1iHhPZ3Rk3xfjOR2XTkDNhB/r7oQgL7naiPxGjQaqob/nXI2FqGlBE1YzTbBPxrxsDpo3LOu9MkyhvRI6HVKJ2yuCdJQHyzGfioGX853GhK9D5/G0HSL0XnoQudnZlo3b4xCpGh0aSfAx+mvGFjyCCrqR9YRE37IkpDWkbaq3VIXvwIre88SqOFgcI0vqNItj1cy7Ojov7p+N89mWdPJdYkxgaSwzJuYgf7X0XGvu+hSfHeLyFn0blIsZ4an/8wcmZtRB77dhKstt/DwQOagf9idjK9dENsdlk8jnORMXIzki+eLdBGOq/1KDLRj+a5JAUugkHsMeMg2gNzkLG5g1QfOhmtva/LERRVuCaO2+Gl+xHfX5c9P3H8tdbhvYDW/BBJrjkKWmt812NI+fb+UJ4S6DU13lh1b/zZjebb0U5bKQRrKOalDpLzGsmoqUeGzkqSgTgD+HMz/pJkkA+i1Nay9YExzewJM56K7xFQ7c741hSaLUVGTbHx1oOM01nA72H2dzVkl5x2NGHYnP50IWKeWQ93LwrlrkUKaDfykB9AjGghSjHamHtHGUvN8T5HhtUbRIX7z2L2XLxXHdBZJiVqGWL0Wet/ADGpYkz6OaTQviEPzWcRgzuImK4hZn4xZg8Qwk+HPP5CegSlCbSTrzQ2owP/YtHn11AYxZlHaZTlGCk1CxKySzOVo2l7UYpRJfKO3rVF5WRgnoO81vOQUHgWuA0ZZQvRumxGcMye6/ty/E4lmoIMo/1xb6zH7ClSv4djsaB+bMhsPkqJa0br+GKEO36j0AXxdyWPXhuwHLPZI3p37RP35jZWuLIFGcVNiPe48mRICfgsZl8mhLYQaDdjT/xOOZAO0PqNauftWFNzE0lhyqPjSJhfgHjnsCgEjsVUllWIP89F/Pl+YON4NAMcIm1G3m3voVLMj5egCH3NqYEhsBf4cmxW6tG3/SFof8Ri7FpkTrUC8JMUIxgfQ3NdvL+6STWJa1HdXy33bEby13tzXYL4d7ah6A40N2+N9z6KeP7zSEFsQFGIeSFwP4piPIrmewoyCC6Nz+lAirnXsF0d+wj9zBVfM96OakLOjc89mwR93IIcZ96r6li8z6s+93kUAi+Y8QXE413mDMaxrac0rbAPGXLfRgbGYHzuiZEo6LFP0jNxzDtJRkRAhoY3eO6MP+5E7Ef6RQdyIGTX/wAyKM9D83U03qscH3VjzVNr65Gcn4vWxvWDDsTj/gzx3Y443kMxOrixUvQlNuA8lVkLNyCeVw7sxKHKV1Gq95z2NGHYnP50CfmRhH1IIb8AHfgVSBnuQF6xx0t6nSif8q2IwXsB7XHMnkRF7eXyosuTjKJqEKd5uOz7kQemGGY4q0jNRAp4L4XpMN4TpQ24BbMD1XL8K1IIW2L04mY01z4Pbmg1AF/L6Yszt2j8eXno/ci7U0sdT5ZeQGg2leqk5gAvEEK1+c9G/pag9zscx3QtYnLfJ4Sflfn2diQgvOA0jxYA9xTsORm5Y9E4L5FqnD6ABJcjODUC70YNKu+uuaj+9Ka5VAbrcArIOzl0w0apldeh/eBd6JfF35sphba9nNS/KLu3lY6ms/srqHM5yHv9ccQv8va0N3cc7Zq6GUhZqRYpbEeK3bANG4AQ6EPKwGmvEIRAX4wy/QpCPjNSUbz3UXl4OAprRJnKQ5rqQnwzW6tRTJOJjXzNmB5CVRmzAnnjq9Uk3mDG4+WiNjGVbREyNNaSmo2ei5xCrjRnn9NFUu7/lVJlsQtBZm8NgW0h0I3OE2Yn056KU9D2kdAo24GnzTgHRWR3omjMQlJ6lNeeLEIyzJ2I9aiOrL6Kor0L+Gcz7kFzPwnVqiwmIaRlI4/fDIEtMUp2NdJTJsXozyMorWo4fPdexAdWIR7mjcfdgXcYyRR/58747KMkCOji9LDB+F13elVL98pCSS8lARRkqQ7Nv9f4/BitezPw28AFZnw33qcR6D9tADDMmkiGXiU6hnjCac/HimnCsDn9aTLlFeLDyPvUjDxAt1Kuh4vZSuD3SQ0DLyB1Tr4e+ABmdwH3Dqn+QaAAZyOGFxCDaQd6MgplnkK2DYWIiw0bH7uRYIrL5bGqWSW8DbNNI+pKH8KDmO1C6W5LSIrXL4HHCCFPQLtS5+SoMcXvm01P88hGXt+Y7HiOYPYI8jTn1cBMR+v4cMX7qOB8EUqPmYc8bT7HA0hINAIfwuwoIZTmsws++naEaLcIrYffo5lYEI3yr8ePzBpRT5SF5Cs0FwFTMPtW2YamZw71UFvKoSumQyPBk3+AVODvPStmIz5xPVpfTyuZSeoddZz8KMwB4ALMZsZo3YvAT1HqWhvJgeDoRgPAV4faLC+mCF2E0lUnxfFvIDX8LO6tVI5cCRlVGs+mksOhmB5zlxk/JUUSOpGnf9S9yiEwaMbPUc1T8bmdgiKvS9D+MmC5Gc8BD8U+Lnl0HtWR5rwmMZsVAIAZrcCvx/vMRvvJC+pfiN/xbIHLEf/zs+Cw6XWIxxbD9Q7Ge11Hpl4oPvMqyhvcA3GcN8dIxg3onE1BMldZEhpTEymC34zOwT4SytosqvfcIRpd3XF8X0bG3No4JyeQobsROGbGdWgNT8R7D8RnfwAZkF+P/XNqphDoMeMbyPl6I5q3lcjw+wnaG32ktW6MP23IsN1Avr7UgWRmtf46DaRokJGAC4ppJVrXg8hBuwDtZeeHlyOeNo2YcmbGs8Dj3g/qFJLzuGo8yQ21M44mDJvTn/ajw1WJafcixlLOqGki1eksQkz7CIUK0GzESGbHorHq3hazZagztSOdzY2/jwAvx0jQY+jAZ2ElQd6YbcjT5iHeED9vjOM8QHU0nmPx+bMp7ZcyNArhJeClWGTvZ6Of8iHjZ5BHzf++EwnGLFZ9MxJOrljNB35YUdGWMbIAea5nIu+hh67rkRDrAL5KCOWLF81WA+9GXvfLkVBchoTdFtJaeH3MTZQr1A3hIGZfQGH5taTGam3AdxGKyngbD97MbXuZv+9EgnAlY9RwchxpKzIIKlEjUt4qoTSVozUoJSbrOe6I95qLjJArEA/pIfXmmE55g9YN+vlAWywuvx/xA3cgeIf1J4AnQhgaNL0ZFwK/hZS8o2hPX4gKlp8z4y6kqAxSHZq3GaVdjpjMaEGFymuB6WYcR2k9zzqc7OlGUQl9tuqFo0PPoP20kBTVnYqcXY5U+RwyhuvQmq404ytl0qoaqb2ze4GDIBrGn0Z8dSfiKzuRAt2IDIoGkiHejs6L8/WFJJj7WeT3ITkMnFcUOTknM6Zy1I2MqlVIVu5ASv8JpDT7+WmPY5yE5uF4HK+PZci6XoxqFUQezViBjJnrSIhwL5HktCv281DN2hdj2tVQnuuobs/FtfksMg4OIP3gSjTPA8ioGIw/T1E+ClFHakxcKVI4H2W8LCBBjxfL/6nxuQ7Y1BO/50Z6C4rwXYHS4Lwx6kXAlWZ8LwSerjINY0n+7tVQc6cw1hkXY0QThs3pTxuQwlFJaZ8L/LwCBrnX6YCYYhulG7oTHchF6EBWRu4xWw78KTrQ7rEYQMy0CUFQNqEw9a0I9vgG0uEPyPPThZjATHSI5iFm/kPklVpecRyiQQpx94dPikAtQorX2XGcfZg9hhoSZj1QGxH87tQ45uNICKxGxlpAzGEbYvwLkZcvP9VFz74CGRitJK9RF2LYPXAyzWVrRcx5s6uQ58y9dI6QV4eMrxlIEPg+aANWYNZS9I6JlKp4H2Y/QwrtANA+okjZyOhGKveFAAn86znzDZsdyMlRKR1wIWpqO7RiT+27t5LPY55BxuwsONn35DV0Xmcgg6RicfBd/EbLb0ohGgT2hcBLwEvZTu7lojTRq30BOg+dwBYvvo99Jj6GPNNZRcUVkYuBvhD4vhlPIl5USVA3wpAhXPPGvARFxxtIKbOTEYDKjWbcFkLFlKk3PIVAtxm3Ih51LtobFyAl/hiFSJ8OLz4XOdLywDH2ITlZiR84pPJas5OAN6+S0Bp3x+d7xArk9DmOHCS7SH1VZqAzcZDSVMzcV47PX2jGZFLTy1rOaoCTdUmGZKL3fMo2O+4nOQi7kLPvfqrXc1alWKf2DrSHHVVsb3z25ch59BgpW2E/cjotZQTNVUOgw4yvIrCEpfH+zyI9YSaSwVPR2lVKV5yBdJpHUZT/EIVAFYZ0mDbg9njNfPJpLuI3WTnphl0Dcmb0oj3UhHSDQbRHG4EPmHEghFPUiDOEnpimfT6Ve09Ng1NqgA2bJgyb05+2Iy/iKvI9QWehQ1YJteZixNjOJnk3iqk73msjgvorrdFxUtrKn8YxOdKIpwjMRozXPU37gf8J/DckrC4kKdkBHfZBxFAeiJ+/Tgh9mL2TVDSdpSakKHiBZB2jUYgnBe9XkAHWRko9aEVIKddh9oWTaWkhtGP2HZTm1YWY5ctIWF2OGMMr8XczKii9Jzcapme/Cxk1+yhMT2hG8/AjQqhc+Kr0LFcAdiEGO40keAfR/C+ktDnfIBISldMH1JW+mkExtqSoVh6yUjF1IEP1zKaUDvhptHb7SQrNVCSEN5NTGB2hjkOFdKhmdG7z+Iv3KpqHvNSr0V5/Pv6tbIRjPWtb7uL9l97KJ5rROTVgwIz1CD2pbJ+EDJztNYhP9CLD6l2xSd6dqCaug/Le1x3AFWY8ghwrV1C+x8pS4LmRpolED/PHkVKTVbQcqr8Z+LgZfzvUNJ03GsX1vy2CDFyGosmPIv6Xl758AFhhxvwcZLlNaL9U8kJfg2TcUZK8uho5SJ5DRq8jVWXJ071aMt8LpLSmo/G+kyjPF5cjRfIPSH1aZsbn7ct5Zpa8vsSL5OtJdTTLKDSOHCp7ShxLIypkH2lPkjXovG0nRaY8unoYzc2VFDpEu0m9igpfSDxpCQmeuQsZLK/EiM1JCoEjZnw+Xn8hOkMbgOdDYJ8ZVyGEr3KGzWQ0v89HcI9bEZKdR4zdKHke+HG85jYk168n1W566r4jzWXTy7fHf89Ha3Q4fl6cPtwb3/Va8vnteNEvUNqxO2WLaR46DzVDrZ9ONGHYnO4kheafUZfm8xHD70Zr14yYy9fLetlFfhjnUD0PuRcxmumUb7J2Iwqrvoa8HM6U56KD3EViAu3xuZ8D/itiXp6GAgnDfVNOBGITUvSdZqI5cAhhp2crjHUotAYZNa+RPGPnkqJBKqg2+5OTqFMhvIzZP8XvOQxoP3BXvM8RHPygMirYcuQ1306pYD6OhNV7MHuVEEpTjQQMcS1iVktQ2sImxJi8ripLHfHddpLmso7aCtTHh2RANwJ9OXsjkIRSJaXA4cTHYnwz0Hxfh85LJ4pgPJPTx2XkpHTAzyNhezUpp/4YcDeCWe/X0JiMjJAbiPUrZmxC6V7F0QLvJVWOvBarA+gihC/EtflPyGguqRd7mLfO/Fv+/Tv6qT92iDkvZe7fEMe0LEYuSoyS6B1+P/LAZxGonJYB/xbxg0qC18/ihSHwsBlfQZ7YpXHM/chJMgnxkLsr3KtWugTt2XKeUHcEXYKglt/0FAIHzNiNeF81ZS8gw77AsImIWnk1ic4fLkSpX/dSCMbjqVPed6uL0vTnAeS0mlv0XXcsvIachieKxxVpGZJ5T1JYU3QcKdg9yLDKO4OT0V7dHr9/XeZvW5FMmk8ybibh6JRSrusZYdPHaITcHO85iM5wXXyWP9flfNZx0Iv4YvH9iptadyPHyoXA0Vhjcyj+fUH82h7gtRBy046fQY6/81DkJsvvp6N1WwfMN2NaCLxixt+gNWtB67g7myIa99M/xPf6VTSn3cg50YR4E6R+Op4Ct5TkZDXy5elBYI0Zd8f6tvGnEPZg9k0UCTO0V/oQPz8LRePO2NrUCcNmvEge5pnoEHQMCalJocNvkZpizkaH7DmqpSSJ9iFDpKBRVxF5SLu3wjVONyFmHNA79aDDUE9idO7Rco/UAqSs/BMhvBAjC+79zycdvheRAj6APG7dJE9RPZyEFH0vZvcWpEUJ5ekclLftaWUvIu/t60XXGqph8LoeR53xXhSQign/a1wPb8R2CLgHpc9NBXpjzxgK7i/s+EsQ8z+O1u+12M9mLeXRokDMuhdFgn5UdO/VKCXnBBIAy+J7LERGzsukVEQnT4WYjNavFdiZq5BrrQZr2Gf5ZDaZlLp2tGrqmtlspLhfjdY4YPZL4ImTRl0IIe6N5VRO05zFKKQX5YxxMUo3mhyffwTthbcCN2J2GyEMOwWjLCkd8EeYPUTqrH4kK4BiitfH0fofRsZBHTpHl5nxQAhkIdI70d6ZTGXDdjqeHqReVt8HPpG5BwBdTK37An9wQytHBtZx02MUKmzu7VyKzuWDOc9Zgvb5dvKVvT2In02jukexB+0BQmCvGX+NlFgBS+icPDeURntVaC3Va/0OIifEhGGTyJst1nptHj0Qf9+C5IJDzXtd4gPkI4wOIj5/HlKA8/ZcG1KCZ5Cgff1enYif7qY0tWwaMkb2UNh82u+5CUU19pExioxBAhjULQTuDoF+M55AUZEjpKaqG+P/FyH+cxSl182Kz/xqRKcbCc0l1Q4tRg6TVUgWe5Pew+hsLyAZNlMoSoEzox4p04sprY1sQzLpc/F7WeCkeqDdjDuLm+jGubkDGY/Xk/TaerReA+i8XQvURSP6vhBOAivkUgS5+D4pPdHn0SNA9XG8m0k6go95ClqfPIezO+UmcSodiXLK/jXSKT3D5BBC9ttSUTc7zWnCsBlrknJ9BVJ4puObWoX1v6gZYlnK4C6G0E8gQ08jRf0AinjkGVUzEENsRAcyPydXntqsV8LhJpsohLn0Lrb+7z5S+tPOIRya7yHl6SOIgXv0wXvnPIEUpevj+DfHcU5GTcfOj9/bTeqvIW+p2U8ySvYsUjrOKsSgiyFzAzJIrkWK0WPx3ZyZ3Y8Q1AoFo6CWP4yUcI+4nY2YyT7Mvo08TtXW9jCKKiXDRkbAh5E3rYfU8drXsR4ptF2U7zDuMJ4/yNx3SnzHG+PfwGwrShV5taa6GrOz0FxdTVJc2jB7GHiOvAalqt36fTjZONFBJy4GrsTsdkLwLujr43wcJt8gbEBrs6HqWIdCWs+Po/nOKg2u3EwDPo7Z345J5AaIjpFy0Y7fRsI46x0ejGOtB26JOd4vxHv1RwS+Wyif2uc9pZKRGMJWzLxOYjbRg/04a+d0MGPS01zx4GusKBdJ3gNcZ8YjOV7Lq0jIROXoIFKwnqhwDRT1AIsF0S+R6jdGm2qpZziBUj0mKFGtEfesQVFAURF9BdWBDKCz6H1WLkH8rAMKACqOkxRNb068m0IF3Z/7KFLyz0E8x3v0HEOp1ucjfjRA4sXnxr8/Qn7k+FmkHK+tp+/hS9g49UJeWDVI3cLjNE85ly1P/Df+38PYsboQQlusN/kjxJdBPOcokq/9yNNehwz+PxulvkmOonUFklvH0fnzPjwLkIzYR2Gt62RKe1J5U+tyNWZ1yIG5Mee7zcAnzfhqceQm1ug9GKN2i+Oz34HWdQ+FBmcL8Ckzvh0Cm6hAMS3tKwj6ejlaW9WWSg9aT6HB2on2RSPl65QdpfHUZ0dI/3yYGpvgnik0YdiMJcmo+QgSwPso7I9yFWow+RVCGBmaV3XaiSArLyOFTrOKYFP8XYcgL3cAn8NsPVJAs8qJQ0zOjf8/QTJgsuR5xD2kcHrADZtaKYRuzJ5HYeppJA/Mq/E+LjxmAB/F7H9FuOr3IsMBJDi6STC2h5FgaCMpRpPi+Ooo7FeQpUnIm9yNGGUvyTvbgOZuIWYvxTHuitd+jAQHWUxzkOHmz69EjuyUpctJjewuJSHEnYUUugNxrPvj5+pdlIpgW9H6/ZAQVGCfFPeFFDZFXYAgnx/F7OdIiK+J774bGdC7YjRlLvAZUlpOFh76t4BVmH2/INSt1K7fRefE89pBc+lC88OY/QMh7CeE1yKQwdvj37N1Vg5D/aPc1L2R0Ro0Z+XObSea18uBh0b52dVoEUrhKKc4uIf1ZjM2ZepunkaKyzxK06jqkBB/jBAKU21k3PwFEvrzgMH/zv84fz3X7g3UV1JW+0gwz8X8YAnVjYMD6LzMqHJtI+Pbh8E7o1dK+S3xZE8QO5EjqZzzBSSn2imDgmjGWaguIgtzDJI5akEgR8tDJKdAL+JvnhrdiGotZiAe6oAr/SQ+9gpSBB1NcntEOns2QmZfRIrKezQoz5k4JT7jpUa6V/06964Blh/hrJ7F7N7+Mb6z7RbuH6xn8FPAE5jdHULYY8afoxrKjyGeuDfOyTESbPrXRrEZbDdJrrizz9Fa+xDfnRKvcZCWBXFMxY6Sa6BsvY8hHeUg4mMbKZSJx5H8e58Zf1dcMxhT5hahOV0dx/d8znPakd7yITO2h1C59UJMUfuKGQtIzVrvRjzvSiQnPYp3AsnYn1EeXnsu8HRxLdEEjR5NGDajRUo1a0ab+3hU2K5HnvjilBTPWZ8FfCQqakOCRBwSSdH8HmKyc5DX5DBiGpPj55OQYrwbeWXnA/8eqMfsXhTG3xGL+l8jeUUOxvsV02TEaHvRQX8GvXdNnaRRJ/nFaI/ehJhk1tiYjoxDN3QGkcesLkbDPkFCq+mP9zkXMeRfImHwdsyejilWx0kFopPIZ77zSP0B+uL/XbmdGefhOpTqdgLNr0NsF6chOB1EAqGBykKd+K67ogFwMVJE3xO/4yAS7UgQNcX3XogEj0eUFiKGf3Yc0wvAk4SQjRZ9iFKPv79LG0opfB+aS4fYvRhFZl7E7FVU5N6APIfFwslx/ndQ6HFfQwJaWIWEZR9ae+8R0I+EiUetHohzeDOFxaAHUY5wRY/cMGkt1cETDsbrxtuwWU11pKXjaK7m4MI3hE7MvobW3j2T/aQUxofJTxsj8rqt8YfHjDlon9VCefzAmw1WooD2xDy0//OcAnMoBeIYa1qPINYrOW/moLP4hqYIeb2UxE9fK4d+F6HA70GOkwFKwWCmIIXwtgqNDi+Dk80ks+ROlROILy6msAnsy6TUtf543aNIiT0f8WU3dtYBj5Xr7xPTvk6meZqdTAHPUiNy1Dm/4rf57vIPcue+Rey+71Ke219fGIBuQ7z1MPDzOIffNeMXyFC7GPHNRhQZerpCv5/h0FFkqGWNs+Po/HmKt6d8BhSR2Q3ckTU+IiCII9+tQMbAbrReZ5HWYDdJDhfvl3Y0bwUOUjNWIWPvrDiGtUgnaEW6R7E898yPi6ge9QWUykohlP6zEQjlsjimAeAOJPPK9cuZRso0maAxognDZqRk5gbBW0kQvZ0RHvgmKsOLHkYHYgnle3GMDoVwArgDs3VIMb0Rrf8epET2IiV3B3qXJnT4DUWdzgdewOy7iLkvjz89SJFuITGhBmRUbCUVVe9BCn5lhdBsJkptWRY/GUQGTD/yKu+OY3sLSdl2crSYP0WG1Qs5T5iDvEa/QExUub4hHIs1G5eSryTVI+PFm4M1ksLuC0g1Mu3x+x7leC9JCSuXfufGzSzyDRuH1nU0rP+JmP7+eO9VJDQzBxo4iJS+ThJqltc39KE0ui9T3IxVIASrKO/xnxnH2oXWw+eqC63ZZ0g9HTqQQO5F6WBZD9Y+4CbMNmSiNjcjgU+ch04kBFYgo/HJeI+rMPuxcL5CAJ7B7DlSj4IuYP8YwlBPp3odRS/QhFndmDotSmk6taU4DFIcaVVk9iuZPdCElKqXSmrGKtMetO6VyNDa5ilgmxDAQCUY6Va0H/YiB1K24edkdLbbETDJbDPmxWceGuMGeRsR3y8XSZqO9kZxms0bhsyYitAl3dBwZ0NP7GP0ZB5CXwhsi2hUH0R82p1DXvv17RAq9hkqV9/k7Q3qEH9cTqFh04P4ocu9pSTI5J8jefcqcHSozWMRv82mf3tD7GZiS4DZHJz6VtbZPua3beWc6wZoePJqnsru/YDOwk0IrbQXIAR2A3ea8QNivcZQe8bUSIuQEbEivofz6zY0RzPjNT4330Gw7Cfnyoz5CLjjsvg+3Yhfe5pyO5oTbx9RrQZ5ZhwTZpyL0pcPIbnlusuReM8bkMFXbNx0oLqqYRsZEZ3vX7OfxWayv4f20VES2NMsJHtvG4W6pwmqQBOGzUhIRdUfRYdjP8kzOAUx51XAj6vcpQ950LePzSCLSGk5/4TZl0gFvP1IYZyK8uyPU2gwOGrSClTPcTvycp0gv19fHQAAIABJREFUNdKci4S2Y+l7OksPimB4rntWoBSS6iv+kJQj7J62WXGsVyNY6znEYsLMtx24oCV+v5n8RlxH4/3mxXfKKnfrkGGT1xNnCmKmA3EsU+L7GvJatcfn1yOli3h/r8FZSfmUmG40lzvRfLpQa0GRDS9abSQBN+yJ4zg7juVQ/L0cpQPsQfM3H839TGR4hPj375cYNaLzqYwidj7aH5PjuFyZXIqE1i60nx2Guw/th2uRF9T3lSPhyKtu5s35jlPocR2Mz5iE1v9nJMM5jVPGw3g1E/PC1koQ441A52gaNWZMJ3bGrqBgHUX7oAHtvXLGXR3l0qVC2MPI5vJFlJbZQPm9NAfYVCZd5hm0z7OoS1lyqNx7EB/aipSXJaTc9YcQT30/4ls+D2bGLuCeqBiOKmWgYj+OzuYhEjCDd2+/dRTThE4rimh8v0dShrP7rxGtRxM6xyUUAlvM+D+IhyyNH++kSFHOea7F+xbXRYIU7m3xnm2k1OsszQS+jnjUIlLD19dHaCw8hSLrbuCtRLz8ZJ3PajbPm0HHkU6ae2bQ0f84ay+/gBcPTOd4du873PASimRorBsbG+RHkcNYt6NapUFSbVI92tPrEe/fjhyRv2pGPzL0tyG9wQ36ixDvbEJGjMuJDpJz2Ps+7SKh0h2Kf2sAlpidhIH/ENJBsiA/TsfRfK+mtJFwXmr3iCkE2s34Itpv16G91Y10wRdCqJiVUUJmNCKnZgNwPAfqfIKKaMKwGRm9A23e4lQz7/59CVKUy3XmhqT4ji8p/epVzP4IMXKPGLi35CgJOvE4Ug7uQ8baQiQEfhsd2KlIEboSMYrDKM1pG/KgTYrf+XYuspY8xO9CXr65SCk+L37/ZRKK0tH4DKCkQ3lLvPYqxBxnorlvR0zvCKmuqAsJmB1kDZ8QdkeD7/8jRZcGEQNeic7LPpLHajdKb1uAGHgfpfDD3gV6ZZyTPAHkOdy3ITz+C9CcXhqf34tHAJIBNRt5EXvi87tIwqI5PtPRajzH/Cng0So1XU2UT2Vqis9tQ0qKC4V6JKyOkLzl7jF3IXM8vs9D8ZrppHXaF9/ZCzPzyA2kcxGi3VgK8mr0OIIArZTiNJdRKMiMCtsFyJu8iAhhHBtOPp5NOYle0bORQuB9L7YhL2bWCJsB7AmhatRpWBQCnWb8hFSvV7xWrWjP56bphcDhmJb0fqTMZBWBKWi/P4GUXUc6fNGMJpLBNhP4N6RoaZZmAp+JXeyHA8ZSkUJglxl/h1Ir1yInynHk2X1+qIrNGUbejyYPEVCw9/AOM54PIdcIIdYebMaBYGqgmMrm9U153v6XEF/0Gkkn7+H0CvCvEcgid1zDpB0o9dmdPisodMi1zORwzzQ6eztppoH+/kHq6jZy8YLreSyPv+TVs441eW3LdjQ3Syk0/rbF37cgWfMyCbjmeuCT6J2fRnN/AUrJWkSqvZ1FqmdyIKKL0Vn11gTd8fmrkU7RHb/nICKvktLiBuLzB+L3F8b7ZvngdATeMOoUjfBN8WdYZMYk5LB5C5K3gYTq9pMQzsweM+NBE4bNcMmsCQmtLnRovMmke5V60KFdSGmvB0Mey/lI6O2jUsf3sSKza5Ah4DCVC4nhcWQoTEHeoX6kDIGY3NWEcDtmX0GMzAv77yd1Q3bY6MXx398hhNLUMEEgfxIxoKkkL18dMgZmoTSB3SRIy0YKI0rNJKShlYiRJQQ6KXutSNgOkvKBN1OaC78d+C/AH8d3cq/3flJqRDNSxn2t+xCzfxkp7VmPyuuIKVu8X94azwY2xgjKtzCbB/xnJBQPoHO6Ov7OMuazSXVMc0jerdmksPvUOPatwPM1AFUcpbyhnW1eZyQDyGFVpyDD12upsmNtQob+AGkvtQCfwGwDWtsXkUFbLg3hGIoY3V7lHcaankfGQwv569mMzsyIYKajUfNeJNgcthm0D64BLo8IQftiLv+H0Zq8gtakE62HdwVvi9+dDdw6krHVQOvRXnl3fGYPqfdFG/CVSukYIfBkVFRvQWfL6266UKTmiWJPerbuwYxfi88vl5rUCvymGX9foXHpsCkaL4/FnzcFxeLtG8nv5eLktVuXUq5ma/i0Hngn+fVNA8ix04/4iIMFHAS+i6KHo17MHQ2uH6D9/3bkOISEJrp/Cif2TaLvYv/OJPpO7GbRHEodJ0b1PnRjQQ5F7TWgeQr7WnSmHiClwjpojTsbd6D5fhL172kkyYhBZMRMxlsm6LO5pJrWlcix+RSpMfBUJOvWxH8/j9Z6G5K7bZn7N2eeV4fWYEwMm5FSrEf6MNKn9lCYxt4CfLoWVLc3K00YNsOnm1CY0ZVd99S/hhS0TnSIFyMl2g2bFqSUuGd8Oor6fC7Wvzw0Ljn5KkB/LwldyBmtC3kvQHcPvSubnUiJdgjqnWQFiZDgViBlpA4ZJPmY6Lr2w8gQ8SJlf776c4jZnUsCH7gYzeGhOOYGUsG815T0kupgBuJnzaT0iHpkZDwU+3E0IIX5RlLR8zZSnvflpD4lc5Gx4kWEFp83BTFvR89x2oWU9XIh7wbEkIvzfL3ju6Hak+K0J28uGOLYWpHQPB7v1xTH1IkiDO4Rq0YvImU0r/Flf/zc62f8fs1ovZaROitPy3mfBSiyszH+fQsyuM5DgvExUiGpw1YHtL7tJKCLU8vMVWj/dZTXfTYyOhz9bzZ619tqhnIvTxcjo+Y1SnvB7EFr/jtm3E6C/D4Rx3N1HIvvm+tJKEN3ValVGDFFY2F9zDc/j1TjtQ0VkVflcSHwohkvxe9OQ/O6J6belKXYyX4l5evEQAa8p3KOJ7jAG5mmUbmxs1M7MrhH27B5FkU2K9U3vQ78Hckp0zsWhm2WosH0QzO2AH9CckbtBzoe5S2TP8Gta+oYsEHqg+UPZzKSaZX29JAoOk6o9v7ROLsPAfKcoLRWtAWd8acore+bgnh2N5KxB+PPLpKxAZKDrgN4f5h+JAccpns2Sb6fHB6S8W1oT+1B/G8b0r08e8Gv9fsvRSAPh81YjIyvVfHvrwBPjUWq6hDoMuTMzIt8DgnV7c1IE4bNcMjsCpQy5AfKqQ7VOLQg79GLSMl3Ja8ZKSp98XszUc7pq/G7b4+/7xvzdxCzMKRYribf89yNlHtPDYCY51n2rkoR2kKCfaxEnm/8OsmTVUztaA63IOOmHRkah5HSti+O511IoE5H898Zx91MMnRmkXJuHwA2xDopB0fIesUdDMFD2UdQNGdhfIZDZrsi3oDm6l4KU286SQLXYV6dwboxci8hZL2Ms0lpc1OQgl/OKOkjecYg9QzqQIrsAXyOQ6hesBjCYcyeRYy1WIgeQwr8TAqbCw6iPe4RpYAMvxmkvXJWvK4xvtMk5HULaA1PIC/uYbQ3PdVgEK1bPVqbp6mOSDb2pOaxnm50LTLajqGo5XNDLLYvoah03ERyPOSRK+fvQ2vuSoUbxa7gt6A9dBC4fRSbUValGEV5ZgTf9700FJpL+Ua3BbeP104YNuNPYxEl6zDjVlJ902FSfdMsxJtuDaGC/Bpbep0UhT8pI3Zx9on1rN12Pb9YtYvFh3tpnDyP/cV6xULgnpGm4Ea+shLxrHNRzdluxM9fLuc0iLVP30XpoYZ4zyDi8efGd8urIfXUYgcZ8HQwryn1lLFpJGSxXYhnObKYj7kDAQK1Zu5zlKTHnkD6l6/7L1B90yykB0yL9xxEfHodcu5eG6/3tN6LgavN+Dlw/1BqrGI9zEok73qBHUMFCoiRz7dSCrufJUd1u5DKpQ5vSpowbIZKinS8DzGoBUV/9SjDbKRwbEOhUfcQLEFKXR06YK+QGsUNIuZwA2ZPjlljv0QrkCJ2EDEmb+joYWCQ8GmM4/Vc9LNQgWV5UpreaqQce0Hnk8C2gr4lYgAeCfJixOIeO57WNB3NbT9CXTmXBLiwgBRtOowMjF7EaL2fyTQSus5zCBEsYPZu8uukPEpwKWKgzjy2xmevIDWd9LzuzeRj1w8AX42/zyMh9LwO3EUIxVDQeULfDay8Rqh70f5qj+/vaFHE62cwtPSte9BcXURq1lqHBNNRtFZZpns4PscbfXrBfz/aL56G54WfDpua9exuRGh4+5Eh6LnvjcjA6Yj3yG/seSoohOPIgbF+DO4+E+3bakp3NzKAiouxBynsZt4M9I2nUXOKqdYu9rVeN0HVqRPxoOJahmJqYYxQ4WJ909+S6ptmI95xLyrcPmXe7RDoMeMJpEgX1Hb9b/508//i/558AZuXNdA/6RKe3Ufip03I8BgRRHBUmN+LMk064hg87fx3gJfNuCOE/FTgEHjGjG1I8b8A6TVbkKxxqOdi6iaBzbhjqwvpBCtI7QI8NbAHyY3DJORRB0vwYv9A2mMH4zO8V5731yP+/REkrzsQnz6MDLhOM25CkeztRWPfByeNiw5qTCc14yrUfmEySW8xM7aiKHmtEfyW+FOt318HWocJw6aIJgyboZBZK0r5WI0OptcVFDOCDqQs70SK2t/E7/xnlJp1BDGVYiY7GH8uQgdyPMihLj2t7izk/a1DB7QehZg9Na2PSqlAZktQk8WppI7IS9D778Ds21EhBO0/V1L7kHGxgnwIWIv37AHuRFGWpSj03EBiTCE+txcxxmzKlDfS+jYh7MPMe+FkhYzXPy2JzzPEGJ+M89OPDJgtpCJoBy3IKzr1PhvfJ4SO+EwZj+VTlfaTFC5/Zj0yms6Nc3USQhXtI79vY7x+Ngn28lZCqN0rHUIvZrej1LK1KKTfjwTDPyHBfC0JKccbfc5F+6QtfieQ0No8pfEQMoyLax9cITJSZ+dsFNHBCspF9t5oNInaog4DpJTLStRLaXrgG5UOUrvBMtbNkd80FAKDsfP7b5CfQgMJ0XDMahtiRGasHA4jpUeRLJxPphapm6bBz/F/Nl3Nk+HP+fPdrXR4jejLyKDZOQrw9WuRnN9OoSLvvHYlAkW5s9wNQqAd6SYn9RMz3kOpkzdLW5B8HCRFqnYgHclTm1uQfMvy90nxug5SretMUlYA8ffTyEA5QeG5NySTDgJfzBoWZkxBhksxcp/TINLVbjZjQzWYbzOuRQ7vXZSm6i0GPmXGl2oEDsmWBFSiMUF1eyPQhGFTC6n55tuRZ3QNOnDzkMK2FEVdsqkn2dqZ7xFCO2avI4OgGgpPF6oFGWt6FUUPjsafn6IDeA2pqaPn7O9D79sIfD1jmBSS2SwEBOCRIKdupAQvBD6G2Vdi5GYPYnhOr8TntJL6wTjVExu0EcJxzL6B1mMtqc8L8Tl3o7U5nxRB8FqYOwjh8XjtcgojRFPj+7dSaFCchULw95A8kf0kQ6aN1D9laeZ5nup318nUpBCOUbkJp6eDbUHG1YF4j/ORwbcDpVl4IWknSQl+ECFxNSNBsAV4Obe+qRopKrIt/hSS2T1orW5EeclXoP3vvXRc8Lhg8bnyJnd5aY/eLHIArZmnEtaTanh+CszFbBYhnPp0tLElbxibV+uUpSloPhsp3ycJkqPhDU8hsN+M15EXudw+mY7O7bilocWC4EUkZ9iuCs0mz1R6BsnIpUgxzBrnk9H7/ziE0yCd9BRQhAP/CjL+zvWPAetnUtdjXP+Fd4YHRt3oi3vvbUjmluMnu4DLzHhoiA0+N6EU+3K0G8kyB1kCycCtCFDmBIr6HENnFsSvukip/sdJzZs7KdS3DiFD6/p4f5ePhgzDH+ZES1YguVkpta8X7dllVEitjzD87yEfARKkR52NjMpaygyOkZojVwK0GDNUtzOdJgyb2ugmVMC9Ayn/00iKr5GKvPzwTEUK3MOktK1aumlD6scy1vQCOoyuEPWQuofPQ0x3GWI481Gq0KOEUAnx5pr4u5zi7r1XlmN2FM3j2aRGod7x+SI0z8RrjiDGdjshSNFWw9H7IuDC2cAfIGMqm5u8FzFCL0qcTGHKzmQSk29AUYiplCpDryMm8jZUm5NVIOsR0/4WKtJfQYp67cqtaxHq2RUoza0RMb5HKTRCfgR8lgRWsATtKR/bHKTQzopzdDfwxWg4jS3Jcyg4VjOvNduJvH1e4+TUH/9+EEVk9pCP7DMd7cFH0JysINVHbUbCsRvti+mcDnU2Y0ghcNyMTchwrJSj3YD2ykVUdpq0AD8ZvRGe9nQPgnueSSk0/Aw0H18do4aGBRTrGq5G/KOZZPQfN+MhYMNYF7CPF4VArxnfQjWPVyDHEeh9O4E7Q2DDqRrfaJMZddk9FCMB55GU+O3IiThARJWMUY/bzJhDkn/twKvVgDFGQGcj2VYpQhnimFehLI1aaSfSjeaRXxdSFz//1zgO7zfVgVKSvV+bgwE5ONBOkiF0kJTW/BqlOlI7kvcHST3G1iFI/Dx549kBtVBe36Mseb1ypbXbC6w14+EIKV6W4hnyhrHl0tHq0FxNGDY5NGHYVCPV1LwNHdwBpGSfgza7hwE9RHoUKecHkCFwbyZ8fAgx9nI4+05NlG/kOHoUwjHMfohAEPZRqGw6xPCzwB1AGyFUhpk08waK1XL4TwD/gRRe7kMHuB15V15EoeXNiAlOBr6MoIpLhX8IPcArmHlhYzuFaTmOwb8UuLsoetFFYm7zKTQcisf8HKoZuhwJKu9vA2LYv4jjqwyaYHYl8tZ5FKMfKTsfBvZg9g1COBajNl9COdFeT3Q+ijIdRUzbEfkeAtaV1J4o0rgSeYqWx2tfQQbYayNObTCbgjxufjaeQntgH4WpdC0oCuZ1T3nkjd72x3GW86R7qtqbgdaRoNTzagOWoH35ANobDsdaTHPQvq4F0OMNQREC+0uobssbPbpBcRDBTY8awlQ5ikbNuxGP20OhkTUV8V+HyT0jKb5jnUefYo3G3dFoW4ocS50IDe9U9p8aFTKjFcmBtcA0s5PIk30os6MRyXgHAGkmOT3NjH3obG8eq15SOZR14lWiAcRHaqaImnY7Qk1bis6Xy9bZ6P3vC4Gfxjqf2Ui/OBYjWFOB/yde9xoyDLsolePEz4trUpvRGTuGMgy60BrcjIyJb+QgnFU0LoqoWsbDMvL5c5Y8+6OV6joSKP3wUuTYLHZsuTN1XbleUG92mjBsqtPq+NvRO+bFnyOkw9GFvAnTUC1GK/BgAYKJYIXXoaZ15fKPp6MDXHxwx4ZCeAqzE6TGmFnG54ZZpSLQLE0lNcQqRw3Iu9EM/CB+thMp+GtQWtoc9P6TkAf6nwmhFlSkJ+MY3hHHcDS+Tyticg8g4ZOlbYjhNCADII85eQRtK1JM5iGjqx5FETbVjIBltgIpWrspZJZeB7QA+HBM1Qsx3eobmM1G62PIUzMDCaojKMpTukaCsP5NZIx1kPK5l6G5fhyzH46wEL+VQkNjN2LIl5Aaik0lNXDdTb5wnRo/31bm706uMLwpOi+HwN6I8vQxFJlzo70Zrf9G4AfRw3cr6vh+Fqmf1lRkVB4GbqvmKXyjUZy/f0TpT35+DiGkovGKkCxFKZvbKa2Zcvjet5nxcgjDT4szYxoqJJ5P4lc1wWqP4JlLUJR+DWoceBQVuT8fAt2x1qWkHjMqt3NJEMYHz5SIlRlLEepaA1I4DyG+9NtIV3iAhCY5CRk/C9C5fQjJpRnAR4HnzfjuOBl7bmhVI48eDYkiIt0X0V64ESneHtlfH4KAfuJ+PFD03e5oGH0UyYAdaC7nkeRBHUpz/CEJuGAwjvcyJF8eJ52xHiT3WoBPmPEPRelo3tOuGKwoS67PbK/2+mU+b0L8uI7Ux6umfR4Nvq8Cv4V4iGc+uFx9gFFoAP1GpQnDpjrNI0VYViFlbhsKNXvXXEiFa+9AHvw8pIqnkAJ9ISn1CpJnoxH4KiGMRyqaKISNmG1CCu90dIB2DqNZqCvqlWoCvKdPsffkNWTEzEPRsM3ABmB7zYq3og8Px3e5LN4HVED6TG5KWAjd0di8hUK8eydDjOk5xFSOxf/fPUyD4G0kYIM82ovWYQlZqOUQDjH0btjvQB6fYiP6EFJ0r0EC7OEh3jdLeXOwBxke7qnzvfAMalj5QVKvl0BKG/h71FyvUl3EQuC+cT0fp5hCYJsZf4GE+cWIR2xFc7nbFcIQ2GnG3yDHwZXIyXIUpZ9tebMZNU5xfnZRvbZxrOhaxFfK8YtBtP+vZhj1PjFacj3iYa5A1SPlss2M20NgzzDGXe25b0EOsS505j2C/T7gejNuLa5riGNdgzzpM0komPvNeCCEkwihpyWZ0YKMmuMUyooBZFAeROl3P0UG28VIXuxFc7MGpdq6I+tCUnrzWNPOOOZKGSNu+AzLsRojdU8BT8WansEhGNYvIJ52eRxfG5rDPtKe/mYIbDSjGaUqT0UyoaHMmJvQuqwE/tiMu1CUrDOm+j6J5GC5c7cYeCTb+LcMbUV816Oxk5FzbyGFvXNARm5N8M8hcCRGnRfFd5iC5PfLpxCy/IygCcOmOvWiTdmEIgqOrd6PPC/9JM/0ZMQ8fpbrRQ+hH7M7kBfnBlKUxFCKzk+r1LCMDamQ/9UR3qMnFrwvprxieg6ar7wUENWk6O8nTtbSDH0cBxFG/f01fuNhpARejdbZ+9x4uPwVkiGm6M1wjBoh6i2nuvJyAjHF4afJmE1D6Wfl8nMDeqcbMVsf65WGQy58HEbbyb1yzsCXAq8QwlbM/gIxaYf33AVsJYQ+zA4Bn0aMfB8pEjQVCajNqDfBm4pijvgv40+l6zpRxGxEsLATNKp0PtX78ByO1w2H3oLSVfMKl1tRh/LPj2bKkxnnoXq64md2Ib41F/hYfG6WV96MnBf7KeRNC4A/N2MzUnBfALadhmlrl5Dfx21u/LwDGTKLEX/1XjqguZlFYbrobuA6Mx4pB7E8WhQCAzE18ANojfIyK5agmpQR9eCKzxvS2kVEvceRQX4NhRGZI8AXQmBjvPY4ilZjxmcoNRQMpbOdj96zHxlMg8B7zbgzBF5AhfxnIafRYRIgQQsyvJ9HUbZqtBmdQU/3ewvSF7N60EzkZPxwrMuqqbfXaeCYOSNpwrCpTq+g9LEbkHLlDRa9420/UvDqUKRmGmLUG3PvpvS0RzF7HDHEeqBjGBGS05EeRQXv7ZQK2UYkaI9RWdC3o6jF+FAIg5jdiwTOx0jNIduRcj0NFcJuRhG6isplBfIOzNWoh4TwNlxaQfW0wD60JksZbu2FDPVHUX5zOUNsKjJ6Xjz5HdVSvZxzv4OYfR55oK8mIdscQ+AIG0baoG6CJmicqZb0k8AwYFsjGtM7Ka+oHkUy6ybgX4Z6/wr0dqS0lTuLBxBfWUqMGJuxLH5ve2as9chY8LSlmxC/uAJ1hP/2adZ3aS35kXNHoQTxqhWI5zUiI6cVvavFv21Cc+dp0EsYn/q3DXEsb0Ny5jCpj80MpLOcEoARM+YjRNXXSU0469A89gJvMWNnjNh4I/RLUWZCO9pX/j6r4t86kcz1tdiD1uGjZnwtBLaa8R1kAL2VtA/3AD8GXqoFtTAEusz4Pkqlm0kCPAKteQs6i5vjO/1GTD2tNc1/goZIE4ZNdTqBIg0ekWlFSmEX8tw0Ie/+/cjo8a7qlUkK2qinCIw7mbWgfdRJCK9FQIJfQwzeGU0DErCTUB6sF8y78NP3ZUS2UxlcYfRJjTpfIfVeOYHWcRYS0seQAN4LfHGYT/Ec22o0GUbsMfOalWpkwALMjgDtw4KFVnTgPGSM7qFQ2WlFTP22CPJQndTb50eYPYSE7SByHEwH1mK2Kl75KgKUeFNAGE/QGUs7kFe40j5tpXozvjxaTXUwjf3AxWb8pMYeGhXJjLNQRLWWxrEXk1Jhr0E8PjvWyyiM8LcgpfAFpCB+2ox/HEJjw7GmZvLrTzylDsT/piMleSnSE3riNTNQ+tlZKEXao9yNYzfkzCDl/X/AjBdR9skFaP/siOPZNh4ogWXo3ciA8fktRjLcA7zPjL2o7sRbUjSi/bgYGROvIkdkce2MO/IcQvrdcW/1E6OEZtLbQhg6Km0IPB/RAP8SrbUjuAVkrG0mycZ6dHZz0QHNqAfqq/XOmaDyNGHYVKf3kvpxeLfbyYh5nSB1yvWDOIXaUC/OTBLS1hJkvFyD5mQf0IbZBpQq9EXkdT+fFNl6hCTcF6E6gAFkHA6gOb0M7ckvlHm2e7zeFa+tQ56an6Ai/uF5QGSc/SckjKcjBnsijmU5WnMHixheZC2Eo7GXUTUlx4vCR0LdVC4UrUeGyFVxPIeAgRhFfGxIkNFq5vlNUk8h5yl1KHz+XULYPrTh42h3MobMrkGpL5CMvpXALZj9mBBOx0Z8EzRBIH74u1Q+8y3AXcO490LyodOz5NkFHi0fKXkH+WrUS0TXirU1xXDkrZSmLR+Pn72A5OkiJGNq6f0xHuSOS2+Q7EZaG4nfTkL81aM22fXpR84xT31+hASDPW4UwsnUpuHsuVEnM2Yj53Gl9OsTaN7+IynlERT59/qWFUgODSJ53Y3WaQpaq+tRVkkHMjrnk8keyTNozJiJoj8XoXXdheqIduQYgW2ottp1xEAhyJRTJ9IrCgwbM5ajmrzV8f/eDHXjRHRnaDRh2DhJYT8bHQJHkVqAvANHEAO+GAkht7yb0CbdgYTMUXSoxh6u+VSQajc+gnKlF6J56kNK8lzkEbkU+BohfBuzyYjR98S0pd3AH6JDXZyudiLeayFwPmYNBWlHevYnUXHqJFKe81Ik/DZg9mVCKAYmqIU+gdb6FbSmc9E6O67/ZP5/9t47SK7rSvP8nfJVKLiC9yAMSZCgBQl6UhIpQ/mWF1sSKdc9EzMTE7vbMbsxuxOxsbsdExvT3dvdw5amRVlKFCmJogwlUqK3oAVoABAgCILw3qOA8nX3j+9e3JdZ773MLANUAfVFZJCozHz5zL3Hn+8o8rgfZSf6yx3/JPBNf+5pZRzTiDMBBoJNRBa/4mhuDXJApiLHYQ2xJOJ64CLPylZ+lLRwptBMYnnlwB18s0sRjXdxPf8Rf86fwqyuV2tlAAAgAElEQVQT51YO+LdGMYrBh4bkyhHfTmEm1ZDOeQs1IFeKHsrLAhvlOSPloK3M32ygsBynmFRmLn1nkYR+04A9iK73if5E0QcTZtQhG+BLRDl0COmM3UgXNiFdUYWIBMYSr7sRGb7d/rOTkM5p5RQOiR2mmEh563MCypo9k/jbDtQ/czHSaY1oXQVyisCOupU4tPs1tHfGkVMWb8YyNJrBoWfXg/T/ZcCbZjyQkVUpzjaVhBk3o7K6VmQDhDXzCeCaNDKOUWSjHAF1ZsPMMPsUcC9wD5qZ8jPgH1DPRQ3ywDuQEXjAv3ajqIEhQ24cMrIfx7kzj7HCrBo5NZcRSycOI8fuAFpLi1Ck7i8xa8C5DpxrTTgobyGh00TfUqk6JIxeQsJrYeK3a4E70CY/ioTUQf/ahQTOxcC3fPalkuuaiebEhGjmCZQFWgOs9/99CzkdXUh49g/ObUQ017OQUgt19aEs7zAaQjowI8S546i0YHbKu+chpeqItd4Q2PAkTD/Xz99tx7lNOPfOIDk1NWiIbHGJW0A3Umwf8Z8dxSiGFXypy72IWXGuf81ADs0c1LP3i3Jq+VOwARlrWQjzLi4C/o0Z/4cZnzZLlQtlwQ+XfBfJ6jyczDwnKH6bE+9PoG8ku4HCjHgIuIzp7/kOBvycldtRGVk3cu4OovO9GvUErUJ6LVxjK3LMmv3fOyg0orvRc3m0n8/+TEK5VN/z6Fsx0YBsjyb/3za0ZoLDMQ3p1Tb/3TAbL3egphmLkR7cgwISoaRwL7IPLkLVPEkE56eULhpDgq3UjKWoV24rcojD/Qh08M3Al31v0SjKwNltDKi06T+jBXqAuNhqUerx0xQ2C7YhZTIfGeLVaDPN9e89glKHZyIWoEzLBApprgOOo8jLJHQvzkPKPIl6YgP+QrT+QpTuBErj7vDHWEZsMD8XOVRdpKftQwRspv/cUxVcV2BqK0YPhdmOQH4wsMihcy9htg2VgV2K7sE+4H7grQEwlBXjUeRsX4oE+iF0DUvRNbxLepR4D3AOZtNPC0NfIeYjoZ7FsgcyGJrQ+uzbgKs93oz26fERTT6g4MJcdD1dwNZ+l1+O4pTBU23fb8YTSN40o0DKO86dzGr0BxuRAZ02mDUMTF6E2J22I311MXClGY84x9P9/N0ngb/2v50WsZ6F5Euy9OxZ4PPETHs3fZ2yJvpmw4fDQN5PIIN4AzJsr0G67jgyZueja1mFru9W/36X/3fI4Bh6LmPw1+ocIz7T7EsN56IZdKB7tK2C2URBz+TNlAHtmx3IqT4HBQjmEUvPqpGNUYt03XFkv40rOnYLMSiWhQ8ivZPV57IVWGbG084pQ+McHZ7Z7Xqye+bq0Hp+C07eu/ej9ZG1zvcQgyKbc855FB5nt2MjHv6Po9Kd4kaz/WjTzEQbL6QBj6NFOc6/Qk3t3+JcfxmzRgKuQvdlAtmp1qPIYXkBKdA0x6YHZULeQcI/RFcOUTgTKJl5uR4Jo7xa5Fbk3FxLZY5NU+K38wRrmLnSPxrqgiO5nYjl63eYmZ/BM7hQ6d+vkLK9DimCiegZvQqZ9K9N/rP/gtk4tLZXI8frxUGbIWM2Fq2Ra9CzbkXZutcTZXDJOTh5cBRGg0Np6YWo5joMeuvEbAXwckV9RMMBZpegBtuxxGBAr++LeqyfxA/DBj4aOQVfZnomll14A2jQqLido9sPNvwGkk2BrAUUODkHORhv+r93I11WDdxqxgFPe1vp72424xeIOriX0KOnPTgeGX2/KDJs1xKHVe5CTs9lxB6Uif7cknJpPJrVdNr2qidLuIRYLnYYzamZhe5vM5KpHcDf+c+0EHXkMZQFn4/0E0iHtHMG0LKbMQetg+DUBNm0z4xfO1eaFMPPlFmJ1kOWszEe6f9pKNDZ4V9VaF1PRrIjzFTaT8wINiFbLciUqcDvsii2zZiKnm/o46lB63Yxkr89aP22o36Y5xJffx6RMsxE6zy5Bxr9cX6R6JmZ5K+pVDliJ9Jnm0t8bhSczY6NDJ8vI2WQNzwtcKKvIRrWDkXBjxAN7oE2fA93TCU242UhpGENKdpihCGmIQqXNXSynsLp8tOQMs5rlO1AyrGpT39OPgJV9ybklGVFUJuQ0B3ceuihcGrisXuRA6nhZWbnI0rKLKdmChr0twit/VYiRemngOcx+79wri9VcyVQ+d/XiQPHtqNI1geAmzC7G+feI6dUIAXJfqwq1JdzhT9+eGZ1iNbzcsx+gHMV10KfFphdjbLHuyhcn9XIkZ+K2T0jcXCpd2guRVHLCfgAgxkbgSecO+v7D3LhHFvMuBsF6JagPdtIbE7eSN991IOi6reYsbaCyHryd183Yzt6dpeivbUXTYZ/p3iOiXN0mvFTVN4ThifXI5nTi2Tra0RdbMjoe6jScxtkLCASMAR0IH2RDHLNBSZ5CuENFJLEdNI3yDcXBYxGLLxT823kvBU3/o9HrHZ3ebKCUngMOX+zkJxL2mRhePnzSD9t9e+HIE+XP4fpaD1tImbFkgQCtei5PEx+8LOJmD1poDD4FuyFWf5vO0k4Nt5J+wHK8l1IXDeGHOCfOsfaxG9VQsbRXPJTowDOZsdGhvq8jSzc/TbntvRitXPZVn0+60/U01WHNkgz2gzjUXT5KFIUHcS+mgbgn0d0iUt5CNN/S2U2QPek73Rd59owW4sciLyBcc1QkKLvRsIrWTtbjCokPLqorHRhPRJC7yAntQU5rOEYhtZCK3BXvxwRszpUylCLSu62D7iPpu9vTENNlIvQM3oHeA3nks/hGNl9dc2oeXGWP889yGFvREqlEWUMzsfs74EH+5X1EAnEHeg5JtdAB3JwxgC3Y/bPxCbK4sbjJOqQ47sQs8lIqU1HEev3ir7X6X9jCvBlzL476M9hsGE2kewBjD3+7+eiqHIqfehwhS/D+BjKKO6m79DGvzbjp8N9Iv3pgmdseh9yLECyrwHt8xUk6vjRnm7y/9/qX3PR3ulXyalz7EcG6WNlfr4V+LEZM5ETthU9++3IMA57tRnp5+fwJTunEY2Up08ckba5FEnMdHS9/aH4Hhbwe/fTaK2lMYUeQXL7055WOVdveofg+0gHXU4kkTBkc/0ZzchLMn4mj9mO5HuT/+2dyCGagp5hmDX0Z+B7Jcgouoh68kr6DtvsRbZgNXC5GYud88FDXcsx4Od+f85GdssRYHNKP9UJKifjGEUJnLWOzU/42pS9TJ22kQUtV/LymPNZ39xGo73OJVVz2OpmsvcoirhMQQbeamScXoyEUhtKCx7h7JiG/hoq3duCFGKaMGtGSrLOfz4Nz6HGu+Jp9QFTkFH9buJvr6PI2Qyyy+CakSJ5pSLnw7kjvjzpBjRjZwGx/wckvFqB/1rkJJSGGtpvRKV0dUQj/TBmf8a5gUfsVC72RVQqGIgPelGU6SbMHgOe9PdkJ5Gtp9gpmY+MiQakIE6g6Fbo52jzx52AGmlnYvbDfmQ9LkKKJitbF3q1Lse5xzB7A+25tKjfAn+d+/1xq5FiXIrKPLLWwT5GTs3yJcQyoizsBW7EbNWwd9QKsQQZtsUOKMiQOIEmdf+dN4pH4eHLZb6N5FQov4GYlaxBhnMtCnbMJxqEPUi+dgIXmHEVMUu+JvQMDBWcY6c/58fNeBA5Z0uIQbP9wH2I5nboMtrl4Rjl2UmhTxSftfkNynKHAE4o1ZuEslP3DYNrGwhmER20LBxG5fyzSJffBfB7/LdmPOaPbcBB5zjgneEaRLV8GXJkkhUkdUQdNYY4YHsv0mvr0Lr6bhkMe4HlbiaFZYXFqEKy6/0QHZvE9QSCo7xrPmTGZrQu8hyXWqi8bPRsxVnp2Jgxbgr/7TN38IOqv+JfW6axr7GbaucwHNZwhIlU08M0DtQigyJw1q9Dht0+tKDr0PDBs0HprkFp4F3IIGygcJBmlf9bG6qnThdkzm3H7D5Eox3K0bqQMJqEDJqfFWXAViHlN510h6gKGcubkOCrFI+gZ3klEmjPION/or/GX+BclqOWDjV6fxYZpjspzDQ1Abdh9juce7Ef5wtmjahJ9ZMoYh+E4hx0/7ciQ/9DyFl4yQ8i/SOKJoZZPSAFsoTYdLmXyPCTvNdd6P6HEpL+ZD2WU5oOMzToPoZKUaYiBbk7cT7no5r9rchZDvd3vD/vK9BzzOrL6vLH2FzBuZ8OnEv+DBSQQTUH7aGR1Dt0E4V9IcVoQ+tsKQo6jIKT5XtfQvKzOPMdspI3AJ9B+iqUegU9VY3k0iy0VwKz5FLgw2Y8D/z5VLB1eSfn52aMJQZRDgwjo38j2dT5AY1o353MwDjHy2ZsQ3LoEiRb9yKHbf1wGr5oxhQkN3uBXc6VnI0E2pflPKNelDkppxwNOOngFBPbNBCHXR5HcnEKWu8tRJtsPHoek/xvN/r/vg3c71xupUj4/R4zngT+E9kBpXHIdtkCzDVj4gCIQJ5AQYrjpFekzAHW+b0yijJwVjo2wA37mFp1K3/qmsHucUcZd8JRBbjqKlwvuN6DTGqcyGFXJ1n2EDL0Ajf6YuBp4BGcy+RBP6Pg3DHM7kHR+vXoHgQGtHq00TcjA+T+3KyJc6sx241Szsv89w8Cv0bMYO1Fn9+N2f2ohGkxcjaOIUUT5s5sBL6HcyUFV8r5dAO/xexlpIjmICHzJPBm5kwXs0mo/2o8cog2JH7/AlQiEspBxiAjI1BSbgc+gdk7OHeg6LiBaW8+2qN7/LE7/Pv1wNf8Z1rQfQ9Ktxbd0xrk6O0AbvHR/C6c2+if4+f9uQTWnhZiGUsryo4VR7Z6/Wdr/DXMpPKsxzjyWc5Awr0es2qcO4HZD5HTeT161rXo3r6IIs/J86xGhn41Un5ZDmk36X1gww15g1aTKJ4BMqxhxhi0fkqV4xxC2bhRxyZiLtoHaf1H45GMnIju7yG0Pxaivb0Z7d+pxH7HILPCoMkb/L9PWX+LL98Zdk65cxw341kUlU+WywVUI1n5q2JH0Dl2oZ6jB0/FuVYKT/t9K9IzPXhCEjNeQv1tqc31HuU6nnllxJUgOT9pv381IP2/DDkwbUg3vI6CQYHW/F+B1yp0llei570M6YrQ7F9LrHgIpb+95NOu58I5NhWRcQTa6GakLzci22gUZeKsc2w8J/3yq1nRtpiNjcdo7gSrAbrNb5wqHMcZ03ucMY11HD2MjMOdaEG3oEjB2/SzNnlIoV6LS4mG6VpgfR9noT+QUfwdZGB2okjMdKQYf48U4Y6ySsHkAPyZcqdKO/cyZodQTf6VcDI1fQTNHfrtgGeniK3s9yU/J8fik+g+96L7XAt8DLPVaFZNiEY3oFKqmeFX/H/bUATvUuDxxLGno9KyqUig9vpjd2H2R5x7FQnbecjBq6EwktiFBONSFIltQ8J/PiFd7twazDah5sZz0brf4n+vkcj0Vxw9Cr1VvQSWscqzHmF6dx7DXR0a6trjz7cdeBazF5CgX+avOa0MIhBUHCUOP0wreQwsUkMH9fxcTCS/eBdYi3NHKzjKe4ggIC8r3Iju5ymdYD5AVJf+CKDnXDuUJzICsYD07EETKu1rR2t7IVr7YR+PRXIgUC2vR3JkPXGPOLSfrzPjJeeGeI+MDDyB7u1V6N4eQjKmBa3NhwebttmMGlRCeA3SsR2IgOD1wXgmZixABC7HKZSjNf4355jxkxznZi/lBVKMtH7byrEH6bVQcg56Fu0oyHweWtuBgKTe//0x5yr/fefo9VmbOhQsmOyP20GkTw/7KgQr+w1PxrEN2QOXoP25E3gAeHd01lFlOOscGxTJsi9z78JerKOThh21dM0wXL3FCFZ1Dd1VRxnXPlGOzWS0YRagRTwODe9cjtnDONenvvKUQ/0cH0OlPmHeSzXKHLRh9nOcq5yqWHNAZFg753yG6le+pKkR6Dxl1Lm6z/+IWQuRuWTPKe0rUInZl5DS2UrfaNQStFZmIMf3Rjg5tTqJOuRU3ExwbMymoJR0J32N9jrg8z6bc6M/9rSMswyOx2zkzDiKh9xp/skrhNI9sxPA3xD3QBpCpqYdRcQC80wleAkxOOUZ4VMhZcaGMmsHfV9R1vfDwNhwvY30dWzCUN2haUzWXvwIKpUbhwyTQEfaidmvgd/jXDnlGa8hJzmvFGYaInMYScrvBFrndWQTgoDu35nOOFkpaklfC+egdXIMGYFz0T4IzFFtaG/VILlwGBnnE5DhGBB6ui4x43VkbIUm6oOImGBtgrL2jIYvTfo9Kom+EhnQDs1dW+kcAwuoFcGMJmRfnIMCd0fRM7sJeJ8Zv3Ku/3vCjFqkww7SV46GQc3zUAAzlRjCOXZ6Vry8HpRJwFafuRoQvKPxKLovbRSu/07kbLQiZ+CPqDdnoJTxq1FQaQ1xZmFoTQiYgEgBBkxP7x3Wx0kGOkfRL5yNjg0A89jS0k1Ndw/Vrb1UbaqmZ1w1PS1VuKZerKOH6o4u6o4RnYMQqQmLegNSyt/A7F6cO93K91YUUSpOlx9Eyu3rvh+ivDpNsTEtQxu7Duj2pVqv4txebxifHsWmhvXTRdW7CGUp3st4f7v/zDgUIW0kvSmwExkWl2LWiHNtyMkh5/PbUN28Q9ef11DejoyYd5AhX6qm+xmUKbqAdLkQBHsrkTxjEtkkAFlYg2idx5NNQNFLdgkZ6LrzooXrUDlNFX0dT0OGyYohpHv+CNo3hsqC2ojzGRpQycFizO7CuXznyrl9mD2JSBGKmdEMZQIDVe6IgZ/BsgIZa3kOXgOM/CGGg4y9RAaugGpkCCezgUeRnJqA5JBDTk810ZEJDfvFOI4ixzeidbYf7f1zgP8VaDPjZ8Dzp4rYwWcxzkHGdA+w81T1Hfgypm0MMZOZZxv7AgpKbU681YX2ST0i1DjqXL/7AxcjmyBPdu8CrjXjmZx+oN8gprLQG5vEZKRHftfPc+wD51htxp/Q8MwOdP69SNdORLru3jJ7hPrAjMnIoatHumkj2j+zkcNUHEwI9NGjZWLDDGejY7Mf6G5lbA+A4XBYTzc1h7qpOVRH5www10VdzTiOthOVQjBMJiADIwyI6gQ+i9mm00YioF6Pq0mvAQYpqSZUJ3xPGcebh/pZqpASDVTLy4GrffbnbKVgvY50gzyJ/Uh5NJDvgNUhQ+N8Xxq2lHzF2YmewzgkcMMMprw65hDpz2OvCQb0/w38vT/3kNVw/jer0XW/hyJZ4f21qcfL/p3jmP0ElUHMQfeqHd2LKf5cf1LC6XgHrcUsHCAOJp3oryHMPahDWaqHKzrvcqHys2vQvboJPf9kRrEdGYhiszP7R5wr1XT6OHKObkHKtJdYyrUaZX8GXmp66vEK6mlLM4zCNPM3qKDx+CzB28SZYcHRraOQhr8arbV9FBIMjEWVBwFGehnNOFTe85x//xz0rEJGqBlleG8043fO8fKAryoHZixBLGNjiTLPfPnOr8tpCh8hmI3kb5a87kAO6/uAH/fzNxZRunSqE8maKWQMzXSO3WZ8D5Vlz6NwZst7wIODnc1yjqf8jKsrUSl1NXI6UucnlQMzmtHauoCoK0Ip9nPoPsxHsrwV7bvguD3gXAGD6yiGAc46x8YPClvxLDcsvpTX2hppazxB08lN3k3NQcNNq6KHFg6EBuOdSPnOpXBeyha00Ks5vcw9S5Giy+tt2Q8swWxcbo2/JsJ/jdhEHtCN7kMjYvT6794YbkD3JdAtbjnDZ/rMJn8GDygT046EX1ZULDgGm5DRcIhYQpaHI0jZBFa6rej+FxvHDWh9zgJewLnS/Rcidbgdsabd5r8bqJ93oKbMNf5v84CX+xAflAPndmL2T6j/5Fq0t44jA/6NTLKGiI3+88l66zT8HTKYL0SybgOwquysZf9wMdor85GMSCuTPIbKx0KP1ZO5R1TP2vOYrUR9E+OQgbN5xAwZTYFzHPXD7G5Da7gD3bNG9LxWIeNouDBkDTp8z+eFaB+0IBn6Kmp2Tn22znHCjEfQ/K0tSPaH5m+QLO5C5apJ5wf//x3EPrpjpLPuLfbfN5RhXea/10t0REMk/zNmdDvHqgovvyyYcSHwFWKfRRKTgb8y43+cIf1AF1M6u34QWGTGeOdKBtnSUE15QyFLEpI4x27ge2ZMQ3oJYN9gOzRFv7kdBTt+k/y7GWbGXBQwq0F6emMeA50v+/sG2nvFzmQtyg49DjyLqgCmob01JGWIoxgcnHWOjceKJ3n/5Z/g9wsXsrGmgfb6dho6ADqo7+mi9uh5rO+pwU1DhmSgwg30fr3EeSev+s+cx+lzbKZRuiwsGbXOa16+CCnGrA0buOKXY9aO6nBriULwOGaPU+k8mZGDpAGRhZDpGk96yVWYqv0ukQkoRCFLwVDPwTS0FtcgA38Keq6d/vfrkQLbCDxaxnEF9VD9P5j9vygq+Hl/zB3IkZroj70K1TL3D8purvCvSr/b6SnDv47WatLYqSGybT3uMxmF+1Jsa+X3o8h5H4+e/aES3w17cTbZmb2wL44jZsB8x+bkt1w7lWbIhjmcY78ZdyJH8AKUWd4PrD6DovCp8KUvd6A9dQA5ErWo/OtGM+5xjg0ZX1+B9vmH/L9Dz9IsJAeeR/fychT0SE5SDzPZukifwRaapUN/1yK0loODZP79XhSUeQG41YzVZcwIqQhm1KHy212Q2si+H/WwfRBRKY84mDGBKF+mUF4jeqAy7o9jsxM5qnkI5Yll9Y54A3/IjXyzkyWXE9D92uEce71j9QW0FkKQtwZoN+NB53g945BXEHVpMULw+n3A3zvHdwfzWkYxdDgrHRtFvM678xWW11bT/ckWDo5por1ZVBodHTPZsWc8rT0oDXkxMkDCvJWAw+j+XYlqwCvtNRhMdFD+ELFS2ZTlSBHOQFH5RqRQNiNjPcxOuB0ZzduLjlmPFNFYzswmuLVoTeQx4k1CDu9xdB+nIUFbiyJDY5ExPgE9t0eI9zYZYa1CBkRoAD6M7u9vkEM5CympF/z/L/afH+d//25gDc5VbmzoO49i9iLKCC4lUkivpFz2u4FCjfg1iKQiRhmd24TZv6J+lvlEZyHs26cKyrNEzHClf9V6soTngNczM5jqM7seKb9AqtCK2TPIcU/bS4EhL62/pxiitT7L4Ry9aF1VTm4yQuEN9tuRY540qjpQEKER+KoZd6ZFhX0W61kz3kAZn+DQXIYckrDnq1CwKpSr1hGH9B6iMONpyLBuQgGTpcTenORad8iRGoP23lP+OwsRw9pgIpT05hnNe4ClA8hgnBaYMR31VS4hBrYWIn37CtlkIaDnmsb2WA7WIqKh4mxeEnNROVmjGceHQ9bUjAtQ2dtYEtkkM3Yhhyb0fyYRepLMucI+RN+zdQP5urwH3aNLETveKEYAzkrHBuTcwP/5t+tsyTPVdH+mkfYpYzlaO56j7RbLYraicpZgdBaj27/OI7A4mU0gpkKPonKRoWYrWkd+zwHEIWKloqCT0fU0IQHbjQzl5ei+vIgEyyx0j4oFXgcSLh/A7K0zcM7Pq+heZLFUVaEMyrMoCnsNipBegbJ8x4izV5qRArkSKZs3EWOYIadnEnoGnf5v9f77+4Gf+M9eTMz0bPTHfRQ1xw9cGYnx7gX/OnVQn9c1yGhTH4DZc6iU7Jg/t+3A931fS4jg7erTb2K2BPgyMh72EmfY3ALcgNmP+pSnmU0FvoUMwd3EZ92IlOsizO4rcBrNAlPf+Wj/dJE+myOUbULafpQzF6heJ/vPrgJW94uB0KwWGYhL/G/vQvOZRmwZWznwtfM1wPHBziQMAhaTXv4S0IbW6dXkNGA7x1ESe9OMD6LSsYPEnrjdKBvWgrKsb6D1fxNaYyeIvZRrkMP/N0j+5N23TiTDGtG6n5jz2f5iNqUNeOdfk+hfBuOUw4w5SL4EUpggq7tR9qkGPdc0HTMeZSr6NRDSOVrNeBj1lWyj8Bm3oCxfmBf2H4H9nvr4zdPl4JhxESpZ3U3fcsSbkfxPqyDoQMHXT5qxroi+eixymtNsorFIN89G63uud6D61cczilOLs9axCVji1oX5GOcQJx9v8QMp70CNmmHQWRqOI0dmE2ZfQIZmgAHHMHtoiFnT3iVmALJSxzOA3+RSI2ua/QXEgZkBoc9iDKoFBzltWUKux3/ncgZSrjQc4dx2zB5DJSC7KSwBbET3+Umc24zZfmTkLkQK411ibXPgx9+F2Mg+h5TaGOQ0TSTWzu9BBrJDfSLfBO5HBspGImVuK/BevzI0wwlm1yGnrRUppTCX4BbgGsx+iHNxNoFzYWBb2rGmIYW4j8JSlqDwxgO3Y/ZPnukvDEj9kv/d4n6cNmQsLkF74Wn/+euRgq1DRsEYpBRD+WrymYxDUfXxFO8P9bh9xX/3KFpftYj18BbM7sG58ptVzeb66x+LZFWvP/dbMHseDRkeSTTRufCsUhegcq6Z+GyDHzr40mDQsg4SllPaCN8LXG7GQ3mOmXfgLkLZmgakA5qQDAkMgo8gh2Uu6s2RbtI6nIb218POKUBnxg4kv/PuV61/fzwK6AyF3Cm3RBcGZxDkkMNnCm5D+7s4ULGb2De5D9kfSVQjB670vLV8vIju7a2EjLjWwSVI9/+BWLLejAJDM8z486l2bsxoIJYjjkXyawyS4Xv93zpRQPaNlEMEWvni97OuYx5xRl3IaI4BvgpsMuPnZwvV+UjFWe/YAGE+RtosmjD1daL//+KSlSr/3nZE4TqWwugLyNj9Mmb1OPfKIJ+54FwPmib/LaTM9xAjPaH/YiVhZkk2LkRCbWbG+8eRUJ1O6bTsISRIzizHBsC5JzA7iAztucS0+FHkcKz0n2vF7AfAnUghTfBHMGR4GDJEjiFh/Ta6x4uRER6e4TlofYZejGtRjfSacEaosf+RM8CpWYScmm0Ulkl0+L+1AF/F7J/LvNbl/jhZrGFH0DO8gDhJei7ZddcBO4HrMQuUxUk65kOoLrsJGX2L0PMLM7C2EYRXwu0AACAASURBVOlKYw+FZiTdhvZrcXlSYDa8A7M7yxpGK6fum2hdJo930J/LTWjt/KnksUYAzKhCJTbXoWxpYBiswe8ZM76fN7DPlwddjvZgFSqPe9W5dGaoASAQQOQh9PPVk+E0+EGLX0FOxiH/uTr/2ovKVvc6R5sZlyA9VRylP+S/f6sZh53jDVRS9sms3yUSFLQSyzSHggp5G32prYsR+kFGSk9WIABJky9hPs4NSMbvIN7jUCb4kHN9HJ6K4J2Tl3wpYxhu+Rco219cZdGKAizvQ/shq+9rqHA+ciwWIfnfhZyV8cihn0wcNruOdPKFNmTXJB2b4Fg2EjPoU1B1RTuxT6cezTzbjILYn0Nl3qMYphh1bPJxBC3655Cym0KcvB4mvq9FEdoW0iewhxkWn8Bs3ZBRQju3G7PvIAV+JRKEVUjB/xKxTZViQrkWCa0JSJCkMWmF0rRSSixrPsKZAedex+xNlKEJfUg7U+5xE1JgK5ETHErVlhDpms8n9kl1oHt8DBnPoS+qHin4q5Bgr0eGcZs/5iXAXMy+PyhrTFmIZv+v46cwqn8jhY3KxTiIFNgipMSyoWtYRumm1kPIAQqOzWJKMxN1Iqf0fESjHtipQFHsx5HSvRLJjSoUANns398N3FvknJ2DnKosh+oE2nvXAL8tcX4gQySrHK7Xn8sNmL1UBuX0SMDFyKl5j8LgUjeSwS3AV8z4p+JJ3j7T8wGUdeskNtxfClxlxvMoozFYw4CPEcsMsxD6tFIdIN8wfbs/12QEuYMYoPoIcJfPEnwUGa1pzkoXkjcfNWMtcsTXIkOug8hAWI1kTzcycsf41/ohYojaSKSXzpJr04FVp2qeziAgzLbKwgkUxLoM6YxQ4rcaZR3zqftT4Nm/liCnCrT31zrHceANn/XbRl+nJqAXya3rOfWOzTnI+TK0rmtRKXzoUR1HJMmYTXqvXp+eRz/081mUwdzsj/V+tG+6i74XdMMO4DwzZp6qGUqjqByjjk0+XkER1EOIPSawxdQgo3+v//dM+pasJBGYqoaWElo183/wpVLNSDEfrqDXYhK6jhXIeGpBQjbMsRmDrmU7pdfORGJGoS/UjzCFMPztdM0AGgjkxJSK5DYipRBYi2Yg5yTMhmknRmRD03kbcoh6iAqwiWj0t/vPNhNZ6nYgoX4zAxmKpv6OS/1vTSQM91TZ0go/SHRoYNaM+pC2lvhkK1L6+Y6N7lEIQOShA63tgPoyvhOwFCnB4s8fQ/voNeSsLAIeIs7Y2ZziBF9BejAhib3AZZg9jHPZEX+VtC0lfwZMrz/vi9CA1uEDOaVzSbAf5fUE+WzN+9H9yZJ3wSleQN8M/XLU25B0UEGGniGD7jhqlB8MvIJKUPNKvabi+2HMqHKuj4NzjT/X4NRU++8sJJZVLwBeQnummb7zgpJoR5nK+ahsdhvaizNRBmESWtc7kE5s8N/bCzyQd7H9hR/k+kuUeQyztAIMOTXHGFlENWE+WB4CE+X3kU7u7a9TbcbFKFNXjdawQ0GAj5nxB+d4BQXGSgU3DgILzGgo6lUZakxHttc2JNMXEW2wEIALVRBXkt5LXE/6UO2VqFLlHORwhtEJDt2vRrS+l6Oepz1I3i8l3+YbxWnEqGOTj41IEYR5JEcoFKwNKFqwm9IR3mNIyQw9JbSap/sjeNqQ4GhFEaMZ6JwbkfH3JtrkC4lKLg2hxKpv6ZvZOGSALCv6+xuInne41MCXhlkoMwrRxI04V1w3346E7xQkHJuQw3gcGSGN6J4fRYZIyHL1IsciOBIN/rMhKppWd74LWIbZY2XNrel7PXWolvp8JNzr/fU1IaNvF2b/DVhZRvavP6ijvPkKnf6cSqGL6JTnNXw2UFhmuo/y2MoMPcs8p7wNlRieAH7t+4GyMJHS+7YXrZGwJ7MwjthUnYcTyHAYPjA7F5VBTSQxjBGztcAfMljsWvyrVCa5DRkyJx0bn824GRmSaQ5tmDr/PjNeHCSjbgNyalpIH+LbgiLs45ATjxnrgOedY4sZ9f7vgdGpARENjEfX2I4Ms9nA/w78jPJ6UBwwzjsUD6LA3gZU6rPQv0KfRyPwQzSkMG+EQEl4lrhzkcFY66/rDefY6xybUgZBhvLftcAfB/r7pxi7URAjD6G87/BAmtX9YNMvIyM8KS8OInn7GbOTwzjLCeYEg3/IYMZ4JH/bnDuZ2QznNh2da9CLvUiXjvXfOYRsi0eIuqQZyfc+vYl+ruFPEUnCPP/nRv/fLuQMHfa/eQXwZ//bE8iBn091vj/3cJytw4Fd7mzAqGOTB83LuBvNy5iLhMFx4kReA34NfPi0nePg4lUUmduBDMFt9DUUZqLI83QU5QjN3QHV6F69QHG0WE7Nt5Gy3kUUViGbtQizu4Y9W5McgI8iARqaZjURXkMUH8K54OjuRgL2fWj9dCGHJhjdvSitDjKA64jlZ7WJXw0sWxCdmmJlHmryp9E/6tz3I+NiFyrpCQrhODL2pwL/BfgpZg8MQXlaUFbJCeppaCS97LMQzvV6uupryI+uTSDM+jEL1NoziCxqaaU7Leged1GY7clDKaetFe2NPASDp1QgpZx5S6D92l/a2MGH2YXAXxJJF06+g5zsb2P2vRR2uDrKM9y7iYZLwHzi/Jy879Uiw37As4Sco8OMpG7ZTySKWISclrVojQUCgHOAi8z4LZIVVcTnfBVah0nZ2eOP65C8Kjfb2u3PcbUZ9wGf9ud1EMmDCf5cf+Ycb/Xj8gvghyp+xZ9/qz/vhciRfBk5Llv8vKOZRId35zAig6gEb6FesCxmTZB98dZAHLZEz9ke0vd4J5L1H0XrbBH5a6Tevz8kWXszPXMUTO0BqszYgNZFG9q3oYokiaPIselF1zQB6ard/vNjgB/kOIhhzMIfUIDjGJLryeBgl/+NGcRS8bRrMKRvPoyeb6f/7weB3Wb8Iq/HbxSDg1HHphScO4DZv6Ao33VIsLYhOt/X0AIPbEh5xkYz6anQ4YTXUAlSmGpfjDD08RV03R9DZUtBwVajTf8YmiNSbGh8HAmZ4vKtXiRgp6IGxh8MwrUMDdTg/QUUjSkmijAU1WnG7F6c6/HGdTCKuoiGaUAnuv4wDboV3cPi9aThp0KyAT31LPtxXY2ox2oHivw20DeSvBdFaq9FjkLacL/+w7k2zFajeuo84d8IZU85fxUpmqwa/UkoIrcRs1v9Z0MJwi0oAvg2ilqHZ92ElNw9SB6cR35JUaBaL8WC9Zo/Vl5JSAvw9kkGt2zsQ+slay8HNNGXeen0QINQP4sMkuJzdv7vs5GMKiYlaUVyKMxryUIjfctUmkp8J3kOzSU/VSb8YME7USng9cjB6UGG2RMUztdw6LwPIZreHxNlSaA7TwsIVaO1uR8FLdaRfa1BbpwsBXWON8x42393tv/uZjTRfcBEJb5PKBBcJB3Lw0SHDeB3Ptq9g9Llv8MCfgDrVP/P/cGgdY5jZjyOGMmOIHlejdZwK3K0W4A3zZjhXGbfSymEUs680t5QfrgXlaPlETBMBR4dxD6zkzDjKuRAHyYGNAzJw+uRfTAN6dFG/14P0oH1xEqG8Wg/n+f/vgW4x7ncTG6LP2boMaohXWZ2oHuwn+zgRmD0DEQySUwCvm3Gd51L3aujGCSMOjblQH0FrxIbyAqh/oP3ky1AapFCyO45GQ6QE3cviph2ImEXSl+mIEFxL87tw6wK536N2eMoktiAhPK7qUaXhh1eSL6Q3QsswGxaWaxPWTALw0XrUCnhxtx+hMqwCF1HmpPq0PVd6D/3NmYt/jxWotKSLmKPDP69UBYwHgnNzcjYCcI1ZLhq/Wc6SO8vCYQR/RGas/13xyBDKasWPwzru8k3nQ82p38YiptkqkliJro/5TXQak3/GPgaivSGQbsNSKEdRgw3n0LPbTtSmFvRejwflRBORve8Gd3/H+HcNs+OdyvaG1lrbLq/rkWY9SKSibRSwQ3I0MmibQ+Nss+Vcd09mD2NyneyAirj/e9tLHm8U4Pz0D7Ic8R2AcsxeyLZ7+UcR32p1nzyjbMa1ISdRJgTVQrGIGe3fPP2i8CLPtq7DK3FrKGB3chhvRyV081E15wVVGtE17sb3d85ZMvgGcDq4iyIL717078GGzeRHQF3aJ8vN2OFc+nP1Wcm5qOAzHyks9YjNru8HrPw/WYk/6qRg7V9ICVD3qH5GHIGgxNQZcYmlH3ajQzjT6JsCUgmTSHK+mdQkOwqMzYC9/uyrEowvszPOaS730LyLm19TEH3ptyAUkn4ElCHHJYwUyfokxakAyei53IVkgtT0VoPz0fyVPdsB7KxFqI1/4cyCS2SwZD1aB2FgGMSzp/XSlJKXs0YizI1aU4NSKfORIGZcshfRtFPjDo2g4MXkFE0E22wpFBs8H9/oF/D9U41nFvnM1TXoGxM2PRvIsE3E7O/ABoxa0XX/lpKb0kxZiFBUUphOP/Zyh0bORGfJSq3kAnp9M3Wg0G3fR350Xn8+9ehSPg4fx5vI0N5PnEw50Eined5SHkcQQ7Ma+henINS728iRfceonZOM/6mAOv6yXAVHK1p5JdMhWfYhNZ1qUb/yuDcLl/++ZfImThInO8TKFLvrajHRzOF/hE5TFci5XQINTwHY/giCh0A59/bgozBxeg5/QlYf3IIqHPHMbufWD6V3OM1xD6J65HCFMxeAR4rCAKo9PWnwDeQobWXaHQHqtff4ly5md9X8OVLyLBNlvqFSPIPhsA57S/mU7rMJZReTaIvMcJTwL9B9yktozUPeD3F2NlKzDhnlQYFOThkWXfncGYso7R82Y/0zc+AO9DeSMueNKL7sAft241IvsxDeio4Q3XIqdmNynFOCRLzd/KcjxBIuZgUggBvHH8Gle4dQ/euyn9+uRlPAI+nOSqeKexDyJkMz9eAA76pPm0ERKlrmgz8tT9OcfBlKvBXngzhL5Bs+yW695chmb4ZyY0u4n2ZA9zhqcorKQMrt1TY/O/9yp/XRRQG4BrQGrpnoMxz/nmdj0reQwl2C4UELFOQ/jyBnIFWJMdAgYUOtI4D+UkIUj3n32sFXqyApe8IsXdoF3I6L0wcK8xRm4Zk6n0Zju9S/988eboHzaZ61Ac1RjEEGHVsBgPOncDsh4g2cClRQII2xr04lzY4anjCuV3AA5j9AQmNDmRofRVd1z5kdNUjitQbMPsxzuWleyuhfq6cJlqN/GFSfLFCqQM+i1kNzr3Q57uVIRgFeThMbETsJl7PIf96DdURX0R09Hb5z05FAn0Tiszej4ypE4jt7GOkOx6BSvqRSi/II8xKKNVEGljbqijsARo8OPcOZv+A7s/laJ3tQjM53u2XIa6G8+dIy3aYXU92+dcJ5JQeBPbh3Ospx17j5xXdip57eD7jkAH+JoVZtGrkpM7G7EcFTHOibb8TXfd1ROap1cCLOFe+I+lcN2a/RGvpRn8uAW8CT+PcSJn9URLOscOMn6CG90kowtyDAgn16Jr7MAY6xwkzVqB7lJUJnA2sdE7lhGbMRMZWyApvGCRSgWQvXRaCzNgOPAj8b/48wrOsRmuvC5WLBmewG5X5XoLKSQNBRgcqH37pFA8eHEv5BBdTM977EArAFTucu9F9uAUZrgVBLT/08Q6UTd1BoUwdC3zDjHucq7jKIgw/TSulPYAM8P/Z/2bQI2HY9Xr/72rkmAXnczeSK5dQGflQcIxKlWeCslQdwH1mPIVk7yR079cAmwdaguYJIr6IZoYdJAbFriI6KG8Qmc2C4x1mCI4l9toEhrcw4ygwj4bS27KdUudo9wN8r0b3bAMKHpyDnC/zx14P/NecvqdZpAdUkgi6dQKlGTBH0U+MOjaDBdEV3+tLruage3sU0boOl4hoZVADfKfPhHyN9HkJ25HACdPbs7JSoWa6FIzSEcs0XI8MmLToXycSorditmaAmbNymrKTze8hUl5cqrQJKbq5aL2MQ9mA3xMHrB7rM4TS7ChSng1Ep6kaKdK7ca6/jYnb0X0fQzbrTbiuvSjKOHSCWY7I8wx2H08xRCs8j6hka4n9UMVDDBeSBec2eodkBtoPhpR4cEqT6EH3ey5yXh4rOtZR4CnMnkGKu7vfMkTfe9FniCYRym36w5o39NiCjJo8hLWZWm7pHBvN+DtkPF2MnuW7qHxkR06J0RMoUnwh2pfJqetT/DEeNmMi8HkiO1fon+s248/ACwNkPgrlKnlOUojodzjHCjP+Cfh3xHkn3cgZ30rMgE0C3va1/U+a8RyxVOnIYPTL9AOBFKEUwvUWwJf+XEPfrHENko+hd/NmM14rah6/FhmiaY5saB7/rBnvlpslMWMKKkHOK5M9jrK4SXKXeRReXw+StdOJ17YXuNGMl8t1MJzjsBlvor2QRp5S5c+lHWURdgDv+FK5k6WQ3on/kM9GnUBBlk3Fs6DKwIdRZULSCQ1EKAdR8KABybxjifenEllDDyM5PAXdpzB8OLDqNaC9+nUz9qN9v7WMe/Y8cuamoADBQf9a6Y99kT/uNWbsRTOA0sh7ygnKGuWxf46inxh1bAYbKgM6E4bdJXG5/29WNOIYikAsRaVpadiGoiBjyaaJbvLvRcGnpvYlyHEJdLivovK3g4nPLCc/kxKcgAsZGOX2BiRY8yLdLYQhZoqaP4Oi+ZuLPncEKYmdSOHdWbLEyrk3MFuHFOhUInvd1grmFaUdtxezP6EyqDThHKiN16CI2W76Uy44fDERlZvNIGZcd6HIX3n7WfdfBoTYvWrJj+DtAq7F7NnUHjCthcGZFyEGu+HOxrOeOHw2r1/plTzyBJ91yO6JTP9Ol2cAW4L6PuaidRAGHId1/01kPBUbrzWoZ6IWeLrc303BSyiIlNcnNw3RPgdD/VmUtWhAhl83hRH6WiR3nw1/8I5MHgvcqcABJAOzSgcDGknvT11EoZHYiGRzKFsKkfZj6HluAjCjFjk2efoilGEtofy+kmmUzowEmvpm4jNO0voHdPq/B8emDZXmNlA6K5BEYDCdg2R2cKCmIQayXuBllK2oAzrMeMA51vqs1meRY9RJHAexDNjnM1plZXy9E3olfQOPjph5OYTK1JL9ZWEWWRg0vAk9tzDg+oT/7jiUcV2BHIwxyElZBmzwbGR9HFRPm34uclxa/XGnonXZg55BcAwbkI3TgAbZPgU8mXCa3qE0jXfoITzde++MxqhjcybAbAJSbMuQcD+ANvj6QWqav5rSRtEB/7l0x8Y5h9nvUf1xPRJmSUHTgITt3SdphM0moXKBFn/8PUjQ3QjciNk9OLcBOVWBJSUPrUjBDcSxeRGVBAQaVfzvB2alI0hpJX/jBaRYLvLXEBRTDbrmdirpG1Em7S3/qhxmRuwbqffXst6XVN0H/E+o3yEo/gYUKQ9R4JnADwfkSA0XyKHrRCUtx4jD2QLb1ExUxtJFZexhpahTIdKET2aEMD0NKZxrx+wBVEqW3CcB09AzGojjkPPzdKNAw2pv8BjKijgAM25BBlRaVrgbOTsfNOONAdAQv+uPP510AoFxSM6ddNoSszi+4b+3FzmGoZeqAc2a2dzPcxoS+MnvTyHjOY/g4hjpBBdjiTJ/LAp+hUqJIEvrUJbgM2b8o3/GgdClFGX6cbSPy3Vsysk+JWfwBIRe0OJjFcvXiiP9ztFqxveRI3ctkYAkDDheS2EQoQH4ihk/QwHN80jPQE1C5XrfLZOS+hyyafzfQfbLQSIjZXHPbh0x69RG7DU1//kr/fUkn1XoB1oAfNGMu5OZGzPmod7IMWiNdaOA5WS0NtYjUqin6OuIBApnI2bcNyCZleeozwAeO00Z0rMGo47NSIfZYjQDwJDxfwBtrC8A+zD7ST+bycPxq5CwK7UROyis4S8+zlQkTJuREKpFwmsTUkSdwD04t85/vg5FLgNlY/J3diBh9lVPdFCsKDLPgizFYFaLDP2rUJTmBIpkrSvogdC5PIuaH9tRhCnMHqlGCvY5kqxisdfhHRQNnuPPuRdFaFecssGkKiv8MjLYu4kR8k9g9ixq0P33qAn7WsLU98jKNhX4Jc5V3Fg7LGE2BRmDKv2LcP7fNSgKtxH4aQVHLme6ePidyum5z1Q4t9oTKHwCRVyT92cD8LuMAZ2DfBqFGSPv6Cwnm60MtIYcCmA8m/O5vN/t9kblV1CJUitxvk2oy/9hMV1sgjr6UiSbAgnI68DL5bCDVQozbBAGDq5CQZQrkDMbSiSrkaxxZM8gOeE/Z0inOPqWMXciPbMAyfbn6bvfQvn4GP/5Hcg5dFTW73kg5djFCMZ2MhO7G+mepKNVS2FVwARUUlVxBtdnMB8z4xnk1H0dlT2nMV+2+/O5Ha2frHloB1Dp2BWojLMU8gYe70TX3+x/P/nZEPxpJZaHhmcS9uJSf65b/TFm+f+GOT07/PHn4qsmPM34N+hLM37UH3cuWi97Sc+u9Phj3WTGq85x2AcY7kP3t5XCrGsNcmo2o6DzKIYQo47NSIbZdGT876cwOnzMv6Yi4/87A6jR78UspHvzIlzF09uT5zkPCZEeFBFd489tNjIq1yN2pqQxsQg5SlnN0m1I6F0FPOz/v5xZQjHyJ2drHHIEP4gU3FF/7DGIbec4ZnfjnCLqyjz9yX/n3/lrakcKrQ0ZEm1okOBdJ40w3f9X0ADP0Oh/7CS7ViVQH9cCf45HEZ11abYaDUj9FlIUxVG4KhSdqsG5h4C/xWwWygLO8df5FvDmKXPCTg2WIeVVja7zEIXOr0Nr5ACVMcBtI5ZwZiFJKjGKAOfewuxtZPCOR89jO86dzvKN8cQ5XXkIZU/9hp9zchfa41ciOXgEBR3WZ/V8eMaq58x4Hun2noE0fHv66TB89oTPsNShYM6NwHQzelDU/yXnyqRgLzznXjMeQAb0+/BBn6nsqf0Wdx36D9xZPZ09X8U4iDLfbxOHH4dJ8pNRQClrjk8XCird6JvED6M11Yiy79OIQZ5qpJeOoz1cdnbQOXaasR3J96w93YgyC0njfSsqhwqsfHX+XEKVRCgDHhBjnTe8q5AezHtWbaikr9Tz3Atca8bTZfTbnCDb6QskF1cRe0ab/H9riXI3ScpylEjDvxjJ6QvR8+wgPsvz0D3dCXzcZ5h60PotDmYFhJlZXySFbCSBwBB6MpDhHO+a8T3UTxSIZEI1yfPAU8VBk1EMPkYdm5GNq9EGzip52Ys21yIi40p/8CJSOnlRv0nA772zcAEyBFqRwroNKeak8b3Lv8JE7YUUllYtJ8tRitiHjMeHUJbkZrKFcSMhvSxHKwgeh4yHasSatIM4+OuYv45vYPbfEwZ9NVJEDyIFFRTRQaLwnYUY4wr56lVuljUjJh8aYPhxFJWFODG5F7MVwKMlHNirySZY6EXRpOswewXn9nln7tSUSMWM2UXonu4FXse5vAj5YGAZCgzsRsp3EYVR2m5U9tNTYendWyjrUEO2MTwdWDVMG/lPL1SO+m7Jz506lJtZyyq3qezHlKHYQOjVq+y7jtIZ9kx4St6LUXa5xf/5mHcKliLZFhz9KiQLL/NDJ1OplUucby+wyozXgYk/4o75n+eXnxpDW3BWupAc/iKw3weaDjrHIf+dz5C9x1qQDJuKjOBzneMtM15GZdFZc7/GIjn795VcCyJ/+Su0BorLqcb68/lHxNY2C+nAE4gN7HJiL+gLSL7Xo+z6ywxMhwdMpLz1Oab0R+hEuq+R9OHHSWxC15YlD9vQNe9Ga+sa9Nw2Iz18HcpadREHah5GDmjopVpMLFvsQfd3ov/NhSjg0OzX6UWkEyoEBHKeWeQPVj5BpK0GwDm2And5MokQmNk5SKyJoygDo47NSIVKtS4jvzQCtNGvZGBCcRWqX85q/J+AHIElaNBWN1HofQkZcFk0xKFB9yYKHZswiDIPgYWkHqV3wwC6nRT224xDwvBuJNxuR8JqC1IaPf4czkeC8OXE94/4676cmHJfhAR/yJS1oMhSiNT1oudyOWaPDorhKsP/NuQEbqOwzKkKlZ+MwezXqQa41svV5K+XYBBdRv9poyuHMkNfQc+pFa2fBYhG/HVUflSqHr6/qEfXHObWbEROemhYPUBUkuVD823+iOZCbKdvJnEyWt9D0i8yikFHYITMIzYABQ4qdkaGC3xG5stIFu4jDiJsBP4XJGv+RFzPgSWxGgWW9iEjvWI4Ry9moIDAYQqN5VBSNhX4mq9C6ERZjBuRHE8G+cag0p9qf47TkE75t54RbpP/TJpsrkHPebO/D3kGcPE17PDZts8RA2ehpO0Q8H3neM+Mraiv7zKiM7wV6ZAwSmGuP7/fo4zYYDBphQxCKZTL8FVF6SwmztFmxtOoMmIzfct0q9AzvN8z/S1HhDtjkB2xGT2Laf7fh9GadMgxDb3FdciJCU5heL7BwWhDYysayA/U9vrrGlvi0qrICCJ4YoUzhk5/JGHUsRm5aECbqlQKuJ3yJxCnw7nDaHr77ciJ2Y8UTT2cpIA8SjrV5Rx/rtegrEqaEDwCzMVsXKJ+/iiqsQ9MLDORwArD6LYQ5w90+uGGP0aC8wqi8DZUu/19pKD+E1IcQQEu8OcfnJvpSCEla4v3IvaqJ73TMA8ZN4uJJQThtzqIgx3x1zAYEfkl6P5uTnmv1//eMpRdSPvMWPKzBxAjX7MHcJ6VwWwyYppqQ9dg6H4eQ9cVhsT+cojO4ABaU8GI6qKv89dMf7Jszr2EWS+aLl6L7n012rc7Ua/SaBnaCIDvfXkWGVtZWeHQi7julJ3Y4ONmJNeKm/kDO1UPinavLHo/sO7dbMbqARjhy/1/szIAoQphMaLcbTfjV8igDRkRiCVo7xF1pCG5vgRl7F9CAYswI6eXaKiuwrMWlllqdRLOsd2Mf0a6bzpxrs3JWTB+OONvzHjUn0M1MtZ3+e+EeUY7M/qL+otACFNqtk05BnkLon0uNxPxNJKlVyM9EyogQgbmepdyigAAIABJREFUaTzpjnO87KmqA/tnJ/AdFACb5P/d7s/zc8QMUyfSuV0U9lsFRyeMqPg4KovPClIcpzw68iZG9n4/IzHq2IxchA1ZSkCF6MbA4NxWzP4JKbWrkbN0FJVjdSCygs0p36xFkaoJKIKWHOIZaocnoUjMpb4Mqg0xUX0JCarriMbnYaQEFiGj9zcno/nOtWH2KDIa5yHB9BbqQXGYXYqEW7J0bwyFwq0bZbg6iVHD4MTVEifBL/HXdIhC57IWZXcaiWnywcBN5BvXzp/rVSSfg9kYpBjGE4c9JhHmBITZAA3ATsxWAW8NOFOifiANTUs34q9FxkQnyrgtRPfQIcf1PeASzJ7DubIjpxXgOVTKkldK0YKGg1YO517B7E1kiE0m6ZSfCaxyZxdeRlHj+UjGJA3OkBW+5xQPuUyFGc2o52AaWnMbKTF7xIwmJNvTyk9nIDl3GAU+1lHI/GTIyFsIfNiMNWjoY/lr3Kwayd5SDJxHkAO01v97NQq2vY7kRjNy0LYSdeMYf9zj/nWdP84KokNU7a9pH1Gm1yP5VVFwyjswWyjRp+L7oorZFksNgO43/GybNSggl5W9n4Tu5QSkx9JK3auQTnmggt/uMeNBf+zlaK04VAL+qncIq81YgJ5hN7AlDEn1c3TG0Dcj2urPJ+jaMDYiIGTfQ8b1MHKKisvfkwhVF3lOZbM/Zhpj3yhOI0Ydm5EK5zowW40EVJ4iGI96UAbjN9OHJprdQXY/zBEUuWpF5xocm8BrPw4JkTEoGnozZn9GmYejSEF1U2jU9/jj1QELMJuJsjI3+VcNsSb+CuBlf8zz6VtK10mkmJyDBF89Mrg7/XHfJKam8cdfhIybOgoFf+i1WYIUVv/6aZIwC7TQpZrXjyCjC8zGonsXGtgNlT1MRaUiwem6FDmBbf68JyIj4QvAJk+pXdaAuqJzXoTICEIDZRVmW4EncG6j/0wDyjKdQD1ctcRp8eaveQ6KsF1KBSUhFWAtWjNTSd9HU9F96R+1NuBJMSqdYD6KYYYErfIHkAMQSnWqkBH0A+dOb1+Qb/i/EfVwBEKTalRKfMiMe53L7JubS2xgL0Y9hUZemEoPCohchmToWJShvB7Ybcb9zpVtqIfgUakMRRtxGGkw1p9H172ZSHYSnJo6f+xkOXYYuLueSLZTjGAoj8wB29l4CGWJZqJ1G7JrQeb2AD9Hjs3tSFccIN7PseiZP+lcZWWX3tHdSoouM+NiZAMky7+cGatQ+WMt6f1Bh5GenpQ4R/OvEMx7F62B8P11qKx7HemB4fFIfwfGul0UBpMn++P90LmStOGjOMUYdWxGNlYggy+LDawFCfChrvmeSXZWaCsq9zrqzyfwzl9P7K8Zjwy/94iD7qqQA/VhYjlaYDppprDB8hqkfK5GjlNSEVWhLMYE/+9iwbjFf28SElzH/XkdQlGdFuAjwE8Ah9nViA1thn+v039nN5EJp5dAKDA41LTlRj1FZy32s28jp3EX0VAJvTjjEYvLdOQIBbapoDje859djJ7FLyo6W7OrgE+j+5FUYBOAb2H2G5x7CT3HGhS966WQSSjQLVchA+RCBstBT0KzU36E2AUvRgZTyBgdRobPTypy7jQwthpoOzmTaRRnBDyj0cN+/kooIToG7BoE6uPBwI3EcrnitTce+KYZ/8O5VCe+Nue4bUR7wSX+fwrKfiTpbbcgeT8R+Csz/tVPsy+FTmI/Sl4pWx19MyiP+O9di4JOvShyH5rAV1AoXzb5z9aTPQS3BXjnTGOxco6jvg/oZuSQhqqPKhTAecQ59gN7zfguWlNLiMHCvcDPiBmzAcP31HwG6avkc6pCwbmpwK+JmZnkXtuJ9PdhtCdr0b5sQGtyn//eYaJzsgvZE/OR/guObTWRKv0HhDJ02SsN/nerkV543LkhCbaNYoAYdWxGMpzbjtkvgM8T6SF70AacgpTN3UPYeB0QegfSIluHUWnDLCI94nn+80eIDYChp6UbTjZWbkPDsZqQshrr338PKc+Qgr4JKd5N9HUCQlPmBUgYNVPohO1GQrODWFPriI7icWSQNyBn4TZiU+dE4hyYMSizsNv/xnFCRFMDMWeiMr6J/rxXA1vLGsrpXA9m76EoUV5PxkSU5v8IulfFkdltKAq1GDkTDYnjNSJHaCdSeGE+xKWY7QaeLetcRUH+Sfo6mBBL+z7pszfJadpZk9Z70T1eUPK3+4/QdFyNZGKg5wzUq3Ulj6BnfD5yHMPU+nbMngNWlkXHPYoRA0+5PKxKUPx091tId2pA8rYBZVLTghV55Va7UXABtD8Cxf3lyCgM5brHiNn7Q2h/fxT4YckL0Lyv1xHzWp4jNBF4svCr9AB/9Gxn/xkZp21Inu2iryzagfZoA+mOTTWSh78qed4jEH6oZujxCT1GB4sHy/r5Rz/3ayv0tx4cTCfejHGo52UbfRvxe/3f5yO74W2UkUtSv+9Ea7MdBXEnoExM6NMErcPVie80o2fbToJm3L9WAc975w7gaTNWoPtUDRwdwADeUZwCjDo2IxViuVpCjCTMJZal7UazXd48RQbVGmQoZ0UvXkOGdjjPRcjAbUGC5TkKlWpwMM5DzsIesidT9yKnISulHHAQCbfAkx8U/3h0vyYh495QhCfML6kBnkEGwwLkJLUgpyUwooXIYB2xMfxNoAazJtTceB5S/h3+c6plN/t5mXNhnkODvw5nXGc1Miw2oAbLbSmf6UVNv4eRAQ5SEJqpo3symVgOBsrqfAOY5hnXSmUgQn9SVvlGYMy7EvVnNWVcTxJi6DFr7FdZXB5U5veXKAP3WsonJgDfxOxOnEun/dQQ21B+c5CYpapH6+ZKzH6Ic1nO2yiGMcwsZBaPO+eG88Tw4Hjk7dG9wFIzHk6ZGB+CRWl9FceIAarkhPbARlWFZPxLRd/bDyw0Y4pniSqFF5GzlMU+Nx7pitSGbefY5/s4biIGdppQ2dEEYu9EKKseh+RxkPn4zy1Aem2SGV1oOOZwyMgNKnyPT0kbwbnMcr3BwEVI7+btrd0oc/RzRKd9nLhGO1BP7jVID79NzPrVoWe8iegs16E9ssY5jnuSgglIh7amkSE4RxenavTBKAaMUcdmJEKlRl9DxthBNHzsPZSlqQMex7lXT+EZrUTp2izWrVDe9Ftk4C9FCm87MctUjBNIidVDbjNuFZGpLQ9iXhPzygeQ8dmDHJNjSLHNJmZhJiDnYBMSop9Fgq2DWBrXjpyYBqQ8O/01PY+exRaU4ZlDehPpVOAOzP61DIP9HdS8fIX/zWQWrgE5d48S68KzlHCvP9Y4/1rtj7XMX0+xkmtH9/YydD9eKHGel1CaUWef/9xKdG8XomeQtnYa/PkFKs/BdWxkwMwlu8k3NEsvI3vC9qWoHOc9Cu97B7pnU4DbPEXtYFC2jmKIYXJWL0DG1Az/5y7TvKiX3eCUmA42ZlB6f4SodCB/iW+oufsR1F+3lb778Q3/G4ECP5T3hgDQatKDWyFgUtqxcW4nZr/055CsQqgjsmP9qIS8fB3J+FoURDvXX3MHko+zUcT+H5BBfCNyCh0KcE335zoVUU8bsM+MX+X0J42i/1hM6Zl17UiOHkNl4bf5fwfyHof6aGr9Zy5Ccn0vysAEIolG9Hx/5ZnpQt/PyUoIP8dpASpNbkD6740KesVGcZox6tiMNEjhfhkZ5EljrIvYzP5ZzA7i3KaUIww+nNuD2R9QX8V+CoVUYOV6wtPfbkPKJo9DHuSwvIOEUF751WQUiSm3l+FJdK8+4H+jBSnpWnQ/HyCWpQUjdDaFlMB7UAQwRJja/auJ2EfUjEogbiadLQ4kdOciAVoc6SyEc72Y/Q7d35uI5VHmz+t+5ChcQHk9OW3+GPuRMh9POtFBtf/sXuAmz1qX10xbS+lnMZFY3z4V3dsr0XMMDqeh5xJ6qRoYfKcG1H9VKmIZ6L6f6uOYaD++H62JrPu+Dz3nuWSvhVEME3in5hMoAryfmP2sRXtvmZn9wDlXKphSdFxaiHTEu320fDARhjtWE3vFOomztU6eChk9LM6xyrOj3eqPd4ToCNUC/4LW8DXIIK1HwZ8tZA8yrKL8PkFw7k3M9qK9eTlRBj2FyjrzBibiHAfNeAj4D0g/JPdmmHu2A1UZrHSO+8wYg5ybL6LAVPGzmQB824zvDaSvwozp/pouRfbXHpSN3+CzAmcUzKhHjuXlSD8eRMRA7/WXEtw5NpjxdyhAejHSY+uBnyL9MR7plfcjmQsKLhpyeu4JTGsp5zsNVTyEqowuFHi7wYzVwG/OtJ6rMxGjjs3Iw1y0SbMYssIgs/dROItlaOHci5gdRnNk5hJnAhwE7kOlWSBDoYu+hAdhfkmIvjSh4Ws3I+WUZkTUowjMU0SGlyxMALZ4o/wpzF5BPRHLUMRuDdl9HpP99YQU9VH/aqbQ2A4sLPNQlGge+VOL8dd1A3mOjdk0f56BKOIgqu/f4c9j68kSMbNDlDdY7QS6/42oHj3NWQkG0F5ixGwa+Sn5vei+ZJUtTEf3W+et44amzHlIKW0j1lZv9MdbNUS9YpMpTeUa+gca6Js9nELM7pU6xhJGHZuRgMuR4V6cgetCz3kScJuZ3enKyMB5Y+nDKFvdm/j7a8Cjvsyn+DtTUNR5Glo7a4GNJWaabERlr8uIRABhttbb/nrq0J7LZNJ0jufMWIfkzbn+z+uRExC+t86MJ4Bvkc/WGNip9uR8Ju0kdgO/w+xBAl1vZfToa5AMqifBoEbs53wL7dtbgB8Tey12kB7oCGMGPg58r4LzOAnfIP9JCgf/jkOlsFvM+FnIIpwJMGMmGoY5Fsn7LhTUvBjYasY93rnfhDIkeboykDwchZMldC/6VzEO+9cGMyYie6naf3dzFuW5mS87jnPhkscDOVLVZvz8TCxLPJMw6tiMPFxC/uRrkOG7ELPxpaJbgwrn1mP2NjL2mggKNKn8NUjzeWI5WCNqDFxApGnei6Lc65CiuQMJpwPEMrDJSFj9HAmsf0s+m04L6ukI53EcWOkps6eSnw2Y4M8pUJE6ZKDOIkZgQZkHhxjHHgX+htJ1ySeAOZjVpGZCzJah6fXdyAnqRob+9eje3F3U97LLv8aTrSga/Xk9gogn6knPsrQgYyk4dEk2pCw8i6Keadddg4yuXiL98T60XseiHpdAa70GGXRN6DmvKPG7/UUH6fN9ksijfa2jvGh0F7qWUQxj+GzN+1CQJOu5HkBO+HxKBI/MmIFIR3oonKtShWT5OWbcFfpdfBnMxyhk7OpGztYRb/z2yRh4mufzkJw8RGGWuwY5KU3IaH+0VHbAOQ4Aj/tXFrYSWS2zZM1U4C3n0rPu/rxDf2gjXu6fbM6W7uhPQOMC5EytRI5okHEHE8c7ACxKZNLGkz935gAwz4xpzlXmqJmxGMnx7RT2khzxr5nA5834yZlgNHsn4RtIviYd3+C4zQT+0ozvo6Dnh8gfID0dsbVVRL3t1125Q5ADoU5WNclWlNXLCyyPYhhg1LEZeRhPNj1lEoEd7dQ5NoCPqpUasPY8KmO4EDk0RmxYb/Z/awOW4twqzL7jP3s9EiptqDxpFVrDlyGn4uMoEreVqLxqkAPyGn0HoQVH69cosrSfvtG6acjIaUfCNThAIaOwGxnloSHxv3gqYzAL9NR5RkQwmvs6ZGYLUAR2B4XKPWSMpqPejbtOOo8aRPogarDspm82ot5/72c4t9bPu/k6imq2E+uQ6/31Jee3VFG6bGs9KomcQd9Bc9ORk7gj8Z5D2apriLMx5qB66TBc8Mc4V2pN9Revocb/vOuaDKzLyBgdp7wMWZKBbmTBbAYyFMehdfcWzp2p9eZTaqDlHDhSA41HoGNn+tpoR/ck07HxRvsX/GeLCUJ60T6YibI5gX3rM8gAHkNcVyHj0Yvomr/jHY8kFqDSrUeQUzSWGFzo9r9/BRo0W1GQwF/HbP/9+f481qO98yska6opzHiHmSgdiMgm7bgtqKx6JpFwpAH4mGehqtiQTWAqUZ7llQz2onU9hfLKmXuRnKosA6VA3iGydcFOpBNnUbpM+5TBjAb07GvRPthRZgnZFUgnZsntMER7oS8t+xPS36GP9eQpIF2yG/VDDQl8QOFqSj/XNhScG3VshjFGHZuRh9AEXwpVlOcAnXpouOgvgP8PGdBdSLlUISX0IjIYP4fZHpzbgYRaFGxis/o4kYVrG1Kuy5GCfwtF2FR6Bk9mNm7//+y9d5Bd13Xm+1ud0N1oZKCRAQIEAxjATIKZFGWZCrYly5Yli6YsSxpbHs+U57157/015fdmqqYmyEnBlixTCSYlSqJEiiLFIEaRYAJJEIkkAAKNnBtodKMbHff749sb54Zzzr23c4Pnq7oF9A0n7rP3+lb4lnNvY/Y9lCawlEivH1Tn8y/AXxClNeQaqKEh5znIu5kr2vAWSjNLS92aDWxOOLbb0P1O8lge9MebX7vh3C7M7kFG1Wy0UDh//D3AfTi32X/3WS/n/P+ixcvQPdjhzyt4D2cCO3EuveGoiOIPgE/7Y+shSjdbie5J6D8U0I2U55pR7VJYSB8FNuFcfvRHXvVl6D7PQ/f4LeCtMhXmcrERpU82Ei9SEfomxRuDzrV6Ke45JKcyQlRcPThImUvStKMlHZ2v6NeL7mUd8AHM3gEewLk0YY+JBbOaH8PN98HqpdFYsBNw7GXY8m7+/e1DDoA0LEFjOi0KcABYZcbjyGn1Of/+CfKdHTPQGNuGHDwPFWznBkQ6W9GztBI9G+H57UeOnb2VNBQ0oxrVG11LRNAMOSJuRo0Tv4Xm4tCMF/+dd4FHnCt+LrzE7xeJahvzPvbnWAP8otxjLUA35dk3RlRPWY6DAiqpF+IMgSvHy9+D0p3GnNiYUYfWnxvQHAi6VsfMeMy5eFU6/9tqND5KkYQONIdv9emP3ShyE5Qywxq8CXjYuVQRoaGikahXXho60HOVYRwjIzYTDxvQIpOGGcgIHd1oTWVYjgz+19CCXoW8IblGWyeaIH8a8/s70XXYRbTQbEC56AuRt/Bp4KWyJIKd24rZ36HFeQYyBPbjnJR8RMQ+798PHY77ifodbAf+e0Fa2HqiQv84Y6IaeWaL84TNpqGixTSjCGRsXEZh7YZzOzD7CsppXoqu7x5ga1HkQSmEa5Ax0ULxwh163PywxLGE7bVj9q9oMb8URWnakPduBvGplEGG9SAiNt/DueIiXbN6RNguRGOlHV3HO4A7MPsJzpVPIJzrwOxelO441R9jMM5CesojONeSspWniCRI485tKfDmmbFUCcyWoDEUlJ0Ms+3AcyWOKfw+eNsvIUoZ2gTsTa1ZMKtFUcwFxI/BFcBnMfse41sCuTzIUfLp1XDlPdDTCseDldsIkz8Mt0yCtRsiD3Q98WIbuVhE6ShAMODmIfGVJuIN4A6/z3OATjOe8L10QkTlfCKD+CSKgjb41wB6TgzNKZXgDmR8tpA/L4SGnR8DfuQc3/KF8WFuPBhHaHJwLYoqxdWmOTTmrjPj1TKbexZiG6rlS0OoNzqE5pBqNN5Dk8eT/vjCOmpoHq00WhPuQSl0EynNjRnMqEWRtPPRGp0bNWsC7vYKcW8kbCK0Pig1L5xCjjcAnOM1M9ajMdrk97u7xDgaLgTBmlIolYGRYRwgIzYTD7vQZDuP+GL5WjQ5xpGB8YRrUOSjj2QD4SiwCrOH8ownsxmI8OSSmoCQy94BLMO5JIneYihqspO4njky/r+NFvLFaAGvQ4vRz4EfF0UWnDvq09z+CC2OuRP0VDSpP5ZgoNZTXmrEaWSwxp1PH0oZeaeM7TyO5oNrEQkLhtQKtNjvQIbsq6iQP32xkdG8m1wjzew6dP3SDMJqdE+TIi+fQCkbLQXvd6KI1B97+ezCz9OO9T3MvoG8k1eEo0XpcD/FufRGjM61eHL0KWT4BAnSKcgTuJHBeJ7NLkORr5PIaA1G8ALgzzH7Kc69nvL7JjT2zkXjtBtdo5uAnZjdXzRmI1yACFlLwuf7kJF9PoUdyEWmlC9fWcH3WOIK4OLF8M5imHEIps30aZyd0NkLvbfCtTvg8Q7NWTVEgihJCGM5CYbm66DUtZr0qN9pNO/MoFi4hJh9dRV8p5ryoxKYnanly60NykUfijj9lhkbPAEpSUK84XxDie86ZEBeQUIqWwm0oHqdmSRf0wXA487Ra0Y/SnduJEqLnoGenX0o7W4O8M4gDO3TlHfdJzHaqePxuBw5juJ6x3Wg9eHjZmyP6YUEkeEfml0noZaCVGlf+1XOejWs8P1sdiPbKUQlofj4p6GWERnGMTJiM9Eg2d/7kId5KTIST6FFK/Sx+TnOvTdmx1geQhfjNAwQLf65372YSD0tCa3AMsxmDltjROe2Y/aPiNhMI+TJp6U/ObceszaUY72MKOXhAPB9nEsK6Y/uYigS9BBmr6Pi5ZvR4nYC9c/Zhe7DzcDNmN2Lc1sr3Mtmv+20kP88YF1sipPZPBR5SIpidaNr8QHK6XSeC6kw/QyzRwk9iSpJ+VK90t+iCNUqNLe+jSKSeyo28M3moDSw/eRHgRx65tuBT2C2zx974e/rUMRlLvHkZCFwt6/Piosm3ky6kQ0aGzcRiI3ZdJR/Hhrl9Xn1wddGsEZq6FBq4614T/yfwDv/FW6th55GP057oXcyVF8K81/SnPR6GXLPh4lfY+uIGhVPQnOJI78uJgk9iNyccXo4hzNjD5FRloTpVKbKt4L8ZsZx6ELrTlKvrjg0UV7aTzsapxXDOQbM+CFKd1uA7m04j9DLZAuw1ozZqNB9PZFjI0j4gyJv0xG5eWQQx3LMjP2kiyyAxsXmlM9HHGbkPQsJ6EHPwKWoXjYPztHjVfXOIb2+aQaSuh4v+A3wn9DcFdo2nEBNrw/593vJrzvNMA6REZuJCOdOYvYtVLx6M5p4e1Ex/boJUtjbSlTgmQTJfBan98yOeS8OA8hYGL5QdhSJiGBWjRZg1UCI7PTn/GYncI83/BqB7jJqVdowa0FGTFrReQOlPcflwawZefib/X4PIGPgcnR+ryHvZQNwF2Zfr8hgVdrXI0TqQIUGdTOKvDyfsIVLKW0MtQLLMZs+iHobcC7XoKn0t6GbedFiPwhcia590jjvQdfiaiSLXojzkbHZkvD7A8gxshLVJxViAaVz/dv998BsMXK21CKDvgcZxVcDqzG7L4XEjzWmE8mMcwsc/ytY+y249jhUT4HOKnAnobYWLl0IP94I+zD7KxR5bUepX5u92mLAe2g8NxBFTuopLu4PAgQhGpN2nWqRsVg4J7yAiGzamJ9KiQa7Pq1tOpp755CsUpWLIM9fLspN+6mi/P5kRXCOQ2b8M1ojrySSnu5AEdTXnKPPjBv9+3uJmjsGiX/Q+dUBDw0hLepp9HycIv6azkPPaqxsvJkF8Ybz/FvvoUaxw1aP4+/9uYiQ7EHzYFIKXRtyfCXNdWvRdTxB/Pk2obkttp/MaMPXBV1M1AOnlSjCvRqNi+3Ad84mSe6zFRmxmahwrht5kN4c60MZJF5F+v1pC3EzqpEpXNxOU35haOl82KgYPUhOHwTeLVkYrZSbq1CEYCpRqlA7Zk8jkhl56mVoV2JsP4M8ju0kL4Z7Kd9TmgyzqUjD3yGjdID8Y52Jcu1fQEZaP0onrMyDqSat/UiJbBJRU8EqtLA/kFIbNovySMcAighWTmzGD64k3dsJSrW5knhiU07EpdV/L47YhPuSZlhWAb0+5e1uZMTnRhz6EYGqRymCXxunkZuiBpIfh8PXwhMPwYLXYGEvVJ8H+z4GOy6HRVUy6o6jMVaPhEfuwOz7oTbMpzg9hOa5A2jsXoWITqv/XS0y7k6jeedc4hUFA6ajmpbCCOBWRIjOR0Zp7udViORuBA6b0RDqcwK8p34VKhgPdTILEAFro7QaYiWSzO1obDeV2O40kp0cZcETkYfMeBLN0f1Aa+hl4lW/riRKi2tDc9wUf3wOpaZNQ2vEoLz1zvGOGb9E895p9OwGFdBZKDJbdF+9/PiHUL3QaSJCewlwtZmtBR4tp59SGsw43+/nHDQOliHDfivF/Zwgv7i/CM7RYsbDSHgiiFo45Oxo9v9+d6xIghdHmIzuQTuKUl2DarOmoVTcILYTSM5W50hPS84wLpARmwxjhW0ovNtMvCRkk/93Xcxn7yCDLKAaTUKOfM/oKUoVepotQIWSM4kM+pBG8xjwckoa0QdRce0B8qM4DUi2dRZSDRocVPvxIDKaetBiGPrYBMnR+xLV3irD1ei496JFs5DUnfD7nIPu12HgGsx+VfH+nVuH2QaU6jITkc/dZUQaO5DntBSMwfW+GE+oo7S3vA+ow8xixug8KNkhvQNFe+OwEY2DtDqIOYgUXYyemSRxhNCLpXIiPDpoRwZOXh+NBdDzZWj5chT1OgdFubaS71nv8K9pwOcx+2qoXXKOTWbci6KUy/02ThJJm79IRMD3ouu4EM1d7UQGZR2RVPqDhSfgIw8/Qobz1URqX9X+tycR6fm/kEtmB/Ab59jmSc3HUX3d4ZxzawN+D5Gd3xCfRhWKxEs1qM09VmfGc6gmLYnY1PvjHxaPvlfUinNUBVntQgLfTj5Jr2GIalhe+asFXedV6N4cQfWwWxI62l+PjO5d5EdPDqJ7fKM/zkHXfZhxJepndgxFJVYQraOXoXG9nnxyM5USmQLOsdaMg2itPs//fgBllryU0+x11OD761yHojA1RCqgF6J1dSZ6Vt5EY7sBjcUTwAVmTE2oK8owjpARmwxjA+d6vSzw3WixP0HUfHMmMoa+m5CytRtN7AvRBLscLRIhzWCr385DMdGeCEq9+qLfb2HUowYt6kac1K/Uqm5HRk+hYR+2dytm71RUyF4IRTh2I6PwcnReYTF826dODQ3yCt5AvuJTLjFoICI1zciY3UlkNCUfg7a9hPyapFZf11Gp93MzWuje7DY0AAAgAElEQVTT0IQWqqN+//Vo0VrmP29huK5bgNkkdF9OE9dkdXA4ioyutJoLnWs88Q6pYGnHk6bw8yoykJNqLMK9fw2R+FI9egZPhEcakp9/FRk7SdLsVSiS0kpyJLANRUZWkZOi48nNViIHSmhKexQZe9PQHLgIRUmq0fwzk2hu6UEOlHucix8TXsb5QTOeIVKWqkPe/jrkCAn3uxn4MzN+7c/nWoo98+3IiXSOvzZPUjzXLUYG7nKzM31Oyqkn24CI1uUU15HNQNdkzSh49PsoLy2uhvLSn1PhHHsRgf2ZGZZ2rUx1cnegMRn3zDhEKG8zs1fcIOY0b+h/gvxeadtQmnsYo+eg9TY4nqrQ9VhfavvOsQPYYXZGTrmrErnx4YQZYb2fhJ6FMDcuQ+Sxg3zhgh7/akNrXwOyNUqed4axRUZsMowdnDuB2TeRN2c1WtC6gIdRvnq8N08CCo8B/4gW7wNEk/JUFEl5ifhoTy4+iBaMOE9kHyJQd2L2VkHuPMgQ6CQ5B3kAeV2vpbKC3WIokvELBt/ToRQmEYXdQUQldOpuRsbWgH/f0H0KXuU0yeDzUCrCzNx3MdsM/NLXpFSCXcgImkN8dKDKH++9wAzMVqP0il5kvDmUCtSL2c9xbmi1SWbLUPH8BX7bfZi9jIrlh1rX9SLyoqYRm1lIkS8OG1CKTVoULERciuHcXsweBz6MSEnuszjF//YRnNuH2WRKpyqFiEgN4zOathYRktkUpwBWoUhNOZGJo8hJkFd74AuqDxNFqgOWo1qEfkQON/m/56IxHpoKd6DmmyWbazpHG/CGVx/7a//bQjLWiua9D6Gxe4j4Z3kDkUGX2y9rJvLmhyjXhUR9Tn7lXLqylXP0m/FTv9+bEfFy6FrvAu53bojzZnk4iYz20Hw2CU0kPSuDRBkEcBn583Ic+hBhWEZ6bVYSVhGlHwfsIhLIaUPz/PloLqlB4+Bp5xIjtEUY4R40JeFraD5L1Bi3ENVonCep6J1A12TGSB1jhuFDRmzOVsjYmI8e2LZY5aTxAHnuN1OJGoyiAL+NvMVTkHdyMqG+RZ7FOhTliDcE1CfmItINleDNW0kxSbqI5NSbgGP+t+MD8gCej67LFLSQv0Yk6xnkOXcS1QLkEphgxJxCi2kPSn0pNq7NLgLuQoZebpqeoVSHL2H2Lylyw8WIFAH/DC2uh4nI1kxknLyOvMBfRot2SDPagxb+o8hY+AxmfTg3OIUbs5uQwlvodRGuzw3AtZh9BzWWHSzeRkbfXOLTKZvRuSQ9N+tQdCsvvSoHNUQRl3g49xxmR5EDYAmRot9hYA2hyasW/SbSDbAQ2RufPSAk1nEPUWPZXnTdJvlvPI1IbKmeWEElLA4nyE+lnIvIQSuRg6QLFYa/hZ69amAN8HaccejTyEKj0M4CY3kFSl9LqsHr9/u7hkjAoBA9yEl0Eq0nYQws8O9tIP+eNgGfM+PHzsXXf3oj80IUSVpIlDa3DljnXCoZH1b4tLhn0VyVm/qXiynIiVWpCuRQERpVlvvdweBiiklvDyLmV6J5Bv/vuegZfowh1j6NAZYjp0XcsxBsh9PomYwjNmHcx7dWyDCukBGbsw3qFn4HWqxCiL0Ksz3AozhXqvvxRMASlO8czmUbMkBCVAGUTnUrZq8mpAdN998vtXCcJig/5aOo4DgGA0Rdm8cWkg++G3mcTqK0imaUHtOKvFizkLF8AC3k5yPDJzfPfx+RcbYOuAqzZ3EuSkVSatYfIIM8rtdGaMJ5G4rOlQ/njvso3+XI0zsHjfOtSGzhQ+i6NxOlV4QmlbOJxA8OAb+D2bup6YpxMFuOSM0e8klDn9/nVCSl/PeDTnlTetT3kZdxKbofp9G4bkT36F6Sms86dwCzh1DtRPDOB0xDRPDhkuRLMtZb0PUMqn9dwPmYfdD//11EftI83s3A8+O6r436Tn0DzS/nEdUNve2VKK9A80zaPa2DxPSpt9G4CQ6EC/13c6O+9ehehX5aS4FDztHpCcE56DmtQs/yRURNHY+b8Tywwaf7XEbpSFoHMtxzldsK0YcMwn3AA8jg/RjxkehABv+7GS/637wMbPN1QDXAJ9Hz20o0h09GRHyBGWtG2cO/Gc0LN6H7HZwtIQJchdSwhpyKViG6KS9Nzhh8mlwN8VkH3YjQTkHz5nzgZ2hslW54Pf5wEcnjOzyD/eg5mETx9awiaiKdYZwjIzZnE8wagM8jr0OQ6g2Yjpr6fWcC9LgphfPJ9xL2U1wYehoZvUE5rBChR04pVBHv8d6LjMM05a0zErJjCpHdP0XPey6x7ULHP9u/aomapm5B3jyIJvlwvRqRsdKOxtWF5MvIno8MvLQF8ABwNWZPlVSfK4TSAl/E7CW/nwE0Hv6KiMgE5SmIGn6GKN3raLzMRoZipc/DTciIT6pfOYkM0gsYSvqKogjfRGkmVxClhqwHdpYkZKrPOo5U+3IjLvuAX+Dcu2k/z9mOUpXMahBxvJ5IcTDU2iwhqgMpxBR0rZKbiY4X6Fx3Ee/ZfRkRuDSJ3TkkeLOd44QZL6GC76Po2cn1DlejiEduGlsPcIkZDagWYqr/3qWI2OwnehYnI9KwygsWNFA6Qhae7VIOmODVvgCR5TjFvjmoMLvKb2+G3/9dwD4z1iCH22UUN3885V8LkVjKj0ocz7DBR20eQUTtdqJnBVTo/mIlaVfDiBa0tqX1EgqftQxyH3tQZDtpDm5H97ANyWOPr/q48hHEKOLQhcZzqAON6x83A12r4WsdkWHEkBGbsws3I89KXFTmBFokP43Z/05oyjc6kBJZyB9uBbZWaNymTVK5COlBcThMlJ+ctvjXE2/4vgB8jtIN8eKkeEcbl6JJOyladxSRla0oZB/GSgtRp/t6lK6yBU3w4X71ImMrF8soLcscelkEdajKoSJ07UdiDvOQQZqk9NWGDKctaDEzKk0tkPPgPEr3eGlDnu2h5eXrHN+jcvIVfr8V2IrZTGToni7ZQykOkjb/XSQosJtiL68hg7UBeb1PI8ITJLq/lxfVm5jYgKKMk4mPyoTUoTdStvE4IZociUNUEaUSvU6+SmQ3UdrWEXTtL/ef7fDbuBmRqQ5EGJYBd6JU2MWkR9J6/f6mkRzdWYTI2FR0j1ej8d2CnqU+/9n16Lr0+HOchaJ5x9GzeTdyJqRFCfcjIjfLOSofp4OET+HbZMZmdC1qgY6xjE445zpNzpubSE4nXAi85CppJpyPdagONEQR4zAXeHQCkxrQs3NRymfH0LhdjOawYB81EmUq7KKYkGcYh8iIzdkCpf9cT3qxcPBSn8dYdDg2m4EKopciw0hytSq6fgp4oUzFpKNo4SyFKpIWa6X6vIg8dEmLRqhDiTMqt1HcN2ISWgQakIfnTUY/LzsOwUOchlY0iX8HGVGrkNF0EkVjthMvslBLfIH7aKccLSS/oV4aphFFkypdrEORc6l99DD4vPfhx9DFDEKDwF3En3t4Rjah52Y6MnJ/CWxKFAKZSJDYyRrk0JiOCEEvWkfnouuyJo3A+XSsnyGD/28IohNSY9pLMclvIIr8daJ5bwmRQ+UUut4XoyahICPsKuA+FEFJwxxUM7GE+L4yQTHqbf9qQPPACf/ZNFTHeC5yVgSDsLDPyUFEyII3PNToHSXfSRXG1nIYPWJzZufuTHR3vODXKDPgEvJTSqci4rgF1ZQOFvsQmb6K+Gc7iFhM1H55ARtRin4cgetH13E1uhat6LxBz/gu9Iw96dw4rRHMkIeM2Jw9mEO6fGvAabQIjS6xMZuCGkDWEy+t/BG02D1Vxta2+O+neZlmAC04l2bQv4CuxVI0wYcFNkQSJgH3xNboONeP2f0o1/wqlNK0gGjhPoyM7T/C7MEYVbXRxEySozUBp4AFOLcN2ObFFf5PtJCmFfhXUUzedlHaoAry3MMV2s+teSo3MhDkUitBqDlKSw8BaOyhduckYzm6FvOJ+nKsH0IH87HCNeTXW8XhAPLM/+24rqUZCpzbidlXUTTueqI+LmtRQ96SxnhOdOBpRCaSGtKCDKx2IsIzi2jeM2RwzUTzWDVyuBwlkuR9Fzmy4sZ58Eb/Es11d/ltdaCxPRNFCzYRGbbd/nyr0LM7C5GQxeQTggbyIzPVaGysJJKVDqIB7yLHSRgzodbh/QdFRmeie3fKOddhZj8iitotRtdtPyKk74CrMWMq0F1pHZBPw3sI2QXXo3vQjcZFNbpXPw41T14eehUiqUFC/BXgvdD0dDzCOY6Y8TrJzpmDaLweQ+M6pOlOR8/Jk0gCP8MEQEZszh6UUy8CyTmkI43rSa45CcWpH8DszZLeZXlO1xKF6AsnqQbk0bo/9vdaPFagSMZSREquQB7TI35724AnUptGKurzM7+/Bcir2oUmyWCInI8Kyb+Dc6NdfBpwmtIpd3XkeotV3/FLlNffQbxBuwh54wvJ47vIGxtXhBkwDxmCxYRPqnfT0fzUnlcgr3u3CBl81ciIa/H/hjqBTmRkN1NsNFb5z+ehFMjKvMLO9WC2jhJSyieZMvU2np2PyHwninxVo1Sm2814wLkJ1Q9hCekGOOg8FyPnxUQsMC4Pmp9+7aPMUp0bHJF7GtW+nSI+tTYoj+VGcYJDYBKKmNQTEY2FaMwfR4ZvLfAT4I8Quen0+wr1Lz2o+/tBADO+guarS4jEWF4jv7dHSI1chcZDF4oo1RL1+gj1iyFd04gUtvrJj8RU+/3VI6866Jq+v5ogal67FGUQzCYQP7NNTqIbm4HNZlYL4JzrNWMh6iF1afi+GW8Ba50r2Zz3DJyjD3jEiz2sRHNvF7A1dztmXITGUnBIdaO5+EJgpxn3jbWscwk8jMbb5Wjst/u/p/m//watdVciR4FD9WtvjEUz0QyDR0Zszh4EL3VQ70hC6C4/epDMcKk0uX7/WgU8W8ZWn0Dj9zq0sIdc9Wn+7zWxjTFlNH/MH89JRIxa0EK/BE1s38e5cheGhSh39yXijf99iDytIk1atxR03HOQAXGqwnqF19D5pp1TXNHzOv/+Tej6HkPnOAV5ancS0wUd5057wvfHyKNXuNjNRdf52bx3dY5XoNSXGYTrafYGiq4FRaUF5F/rk8BDyGgL6k4b/HHPQAbYAEqtO06UYvhQyvVIw0soSheXugOw8L/w32as57J+iqOTnYhE/pEZ7c4NsnZm9FGuQyQtijo6kJG4ABlcs9D9bkPkdztqEjv0Y9Q2Bp2a4hzvmvEwmo+6kVOlH42rWcgJtJP8prTd6D6EpsTBMRD+346cOpcCPc7RZcYP0Bx0DSIXnehZ35LbANN7+zf6F2Z8nmKH2TxkzC5E82Uveq7UVkDHH56z1cjL3eh/cxiN/VwnSz+6L+eiuTJsY1uZl3HiQ+P1o2i+Okzk/DNESC/B7Ls4t8M516ufcCkiGafReh56AK0ErjDj35yrrK+Nc5wgXwQm5xBZjBQaD5FfP9mN1oUlwKfM+H6ZDVpHHT6N7CeewF2J5oge5GDIlVN/dIwOMcMwISM2Zwuc68DsLeT9SiIQtchAGVzvjsEjKPmUKvhvRxNkaSg97Be+Y/gVaKHtA55DzT2TUr+uRYZCC/kG2HH/WoRC/uWq8lyNJvq0yfwIkp5eV7FBFRn7txNJVFdhtgN4CueS6oNy8TpaNINcbyEaiCt6Vt+YR5HH9gbkzTUUkfoR8E6iCIVzG309wu8gcpSbc78NeBDnogiAzvMTyPjKXdyrUcHy9WjhPkYxWWhCdQ/PI1J0EJGb3yDDIAhV1Ptr8SrygpbfQyf/3I5i9l2i1J0TyFBrBJre5PLWf+WLjY7qpHvTg8baHQxWFGD0sQUVqad5ZKcBewctcT0cMJsKfArd80UoglSLjOVDBBlls58MQ93RIA6PBkQOqoETzrHWjBYitbBqf5w/RJHPJei6BwShj3rySXXowwEiOKEAHl/0HeSjK0EhkT0f1fKcQuRnMZp3p/t9N6E5fCd6hhvQvNPjf9OInr9zyO/dA3p+lqJn6elcwvU+wEp0nUKKXoBD13Ey8FnMvoJzXWY0ozF+iPyI+IB/rx74tBlfHUYBhlvRPUx6tveiuXYR40EJNAU+ClV2RCvDxENGbM4uPIPSAmZRXHhZhyadn1csrzt0FBaSJsGotJhbjUd/VdZ3JVf7Uf/XQmSkHSd/MdmLPGSzS9TnBJzDSKXoRJ68G5Gxnlsn04zku9fgXLJnToShHkU8PozO9xBRj50QBboXqMPsI8h4qUYL1MtIWvg9fzxWpsADOLcFs3fRNQqkbC/OxUmnXolIYuHi3o+I+h2IIP8i5rcdftuXI4Pwd9D16UEGVBcynH4BbEzs/aLzW+qPY77//Ub/m3wS5NwuzP4eReuuQEZcC/Da9by0tJv6W+IvyhmcAJaMtvrTELAekeukhp+hLu2R0Tyo/COwBpTaNR0Zg81o7ITxOhOl+XQAX8DsWzg3KilPntDcjiLMFl5m7AAed46HiIkgmrETjZUpRM1m+8gnHfVE5Dr0t9mAIiZDSXcMfXROont7MfmEJPS+OoycN4sQUQ+kpAutO8uQM6LTH1e739YA+RGnC4B/QB709xNuJYqGx+EUus4rkfPpajQOktJ8A/m4AokPDAlmTEH3plSmRzci5+Oa2GQ4+5ERm7MJzrVi9m3kzVmKJr8BdJ97gQdwbvDpUIPHCWRMpDWBAy3e5fXWqBRm01HH+tvQQh1C9x2oNibX2HbI+1QOsYlL0TFkQIV88xMMLkVnJSI1O/3v56EUlGno3h4BPo/ZfysyvGWkX4mMqVxJ43OQ0X7EH9ubiLysAP6EiAwMoDF0CbARs5/66Exl56BeK+lRCZGv2xDhitt+ExobDhGxuCZpnf6zLuB/ofs3j0ggIL3vi1QFP4nOtxONiyZEBj+E2f1FBFIOgnX+dQbdxmWUlruGKD1u/BMbRakeRj1MDpMfLahDaR3ryK/HGG1chshMKyIQhcZiKzrO9xBJXo1SWkcUntT8qd/3AfKJ4Vzgz824x7lidUbn6DfjJ6hWK6R7dSEjcz5ySpwm6otTjaJr7wFzzbgCzRkgcrGlgjqI9cihUE3UdT7XqTEdzX0hQuP8OR4kqsvrQs/lPhRR7UfpgIdQNGouUTH8EeDh8ZrKNCKQqM5CSou7tKF05jdQGmypmo+j/ntDJjZEMual7ksXch5kyDCmyIjN2QbnjmD2T8h7tgQZHUeAbWNWvC4FsedRH4ykdIh6tOgNv1qbFL6+iK7JSfKVuCYh8vASWmxBhsfkMrf+NkojCMbCfJTf3kAUpZqEDJGkVLAk3IKMs1qUQjcbLR6dyKBYhIyLP0Qyzbn4LdSY8SD5HrTJyPh4AXjGF8NfivL8d5NvdB31r4v9PovraYYHs0hvZtqEFtUedH2Tuj/3AXO9stsWKku5/DiKvrQUvN+O7uVd3sNfygABeVjryvieEcnjjn849zJm7cBvo7klRGK7UF76S2VH84YbIvI3o7kuNFiMM8S6UQRhPXA9Zs+OQk+vm0k2Xo8h0v5pM/7WF3LnwTlazPg2ikQu8d/vQM/BAJrTetDzsxuN2WXIsaGaPOFy4GNm/NK5fDIeB+doNeMJpEC5gGICPstvO7y/F80vh/z5hvG9D82Tuc6XdjTXh/l+LiJd7x9SI1RTnrOoH6gzOyMcUSqtu5fhU5YLdV2lkC9AkyHDGCEjNmcjVMexh/EVEl6HwtkrkHcud2Keijw9945QmtztyDg+4Pefi25kHFyJvLf9yBgoN0VlPUolCL0srqWYPM32730Bs2+Xlf5i1oiMgz3I+zyTfMMiRIJ6gD/E7Hmc2+5/ew465xaKU/tOocZ+twCbMDuASNAhkhfL3cA1mD03Qk0WSy3u4bMQfUzC4ArXzeYib2hSTUwXMiRvA35Qxha3ILKbhlBgPfxqO2bziaIXp1E63XZ84fGQ4NxmzN5Ghm4jwXAdjm0PDZPQPLIHPStpaToz0Fiv9r8pJzI7KJgR+oul5fS3o+joChIiXs6x24x/QsT+EIpIvke8IdmMxt+b5JOpY8j4/AMzep0rq3lsiLJcR/GzdRzNJyESOoCu++soshDqKkOqZhIMjaWxyCYYa3Sia5SU4hnQBGz38syh51jaWtnIMEnpO8cJM/Yh51Na2nUjQ21InCHDMGAsZH8zjAXMpmC2GrNPYvZ7mF2El44cFcjwuQ8tlHNQzclitKCfRv1iRiJa04RC8gcREQgd0XPR699rJoqylNdYU/UijyIP6bVEZAO0sM9CxsUmZETdWuaRB+niaciYSSIUfeic7sh573qiupNchHMLjfSu8dueSXwn9QDnt3VhmcdeKTqI6g7iEIhgHaV76qQp7yXhUkqrW7UCF/joXykEp8LchM/Nf/bMsHbzNqvD7FPAXyFDdC5KIboL+E+YLRiW/Tg3gHN7cW4rzrWMA1IDGtNh/AyQPJaMyBCvvKavcjRT2mgFPcPL077gHM4XPv8UPQdx193QfNeJHBiF6EEk66NmpR2bPoLyInL6rEc1Mm+gvh6/Id/RUE/0rIbGo43+OF5Fc301+ahB6bEvMr4ccaMDRQtfJXmuCJhEJO7yInKWpWEmuj/DhWfQWlZ4/wJmo2hp3JjLkGFUkUVsznYoReMm5JU3ojSma4FTmP0Q5ypVyxkcNIk/4dPSFhD6lMDBEWzoF3oCBANmCzLoC1V5+tBiUAe8inP53afl1Q+pfceRB1wExrm1mC1Enskp4RfI8NhM1HzuIHA1Zk+VEZnqQt7PUCuVhHqU6rEYs1m+L8uFRAZ+aKA33x/TSaImfhcjwlXOte8lv1YnHYo4XYJIVuhC/yrwVp4aGgRFv40ooneIYnT58zmf5ALWqei+tJR9jBFmUzpF0BFJ2aaKRXiv6v3A59GYOUJUWzYTXcfnGExht+qRlqLndwEatxuQp/S3EUkrjDwd8/v8AmbfGAs1sBGHc71eKbAZPWdzifdoT0bjvxHdx5HuMh+iQguJoqxxdYalopFn4Bz7zfgZqgkL0eGQFrgCOY6eJNmjfxpdn+WU4cDx4/lZlJKWu1Z0+1foV1VPvmFr6Jnega57FZoTgvw6aF55DPjN+zANLeBlREanEz8el6Coa2h4uhGlN8aJBIHufytUJvecBud4x4zH0ByTK/0/CT1zJ4F/G89NOjO8f5ARm7MfNyBlrcL6CVB4+8987cDo9baRHOxYeXb2IqPmInQ9QhF0Iyo2/zm5OvYSHfh95PkGGbfVQDdmj+PcK/79HtSXpQ+Rnz5kaOdO9MGrPJNSucjO9fkmpH9JejShBhmyk4EGT2RDL6OVyLDoQQtmWIiu9sf2nv+sHMW6QEJLw6wZGfVT0AJ4CF2TDwK3eyW3wvv/HCJaQf0pb4to4dyJxuwp8snYDHT/7hlkjUcX5c2FgayWhE/f+BZKCbsFGRsgkvsAPq2koqOUwMEfEEnuhgZzd6AaoSaS03na0Bi5nrFULhtZvIAEQvYiglfYlDY0twzRtAdHsibIjPMR+biSSLAk7H8z+ely9VQgQesc68w4igzc84meh0Mo7bdU5NIhwlUu3kJryRwioZV+5Ky4ETkH9uV8Ng2lw3aj8wq1QNVofvq1f3+n75/z/oVzxzH7VyTespSoYWsjGsMbkJqp09fpNON7wN1EqnWBWE5B6a1rnKuonrOMw+RZM3ahOeQiNIY6gceB9c7F9vTKkGHUkRGbsxnymn+IeFIDMuon+e8UFp+fLQj587mNS7ciA2ApUSTjFPBd4Ikzxo4Ua76AFphCL3gd8AnManDuRb/90AxwMvKK7vL7GawX6zVESM4hPlVsFjLiQtPJbpxzvm7mQv8qVIYKXtb5aFHah8ZBKWGDasppmhdJ7hr5+f2hkdxk4G7MvuajS4JzhzC7h6g3TAeRiEMtSqt4AbgTkTWIlO32ovqswZLzTSgCkoYmdC2L6zEURRGZVX8lf0p0Ai+Z8TK+k7tzQxIL+D1EVlsK3u9EHt8LEPlLqhk5hGqlfj1mQiIji21ojNyIiMNl6Np0oRqPychInI0it6+P1IGYcQlqaHgEec7nIgPUUPRmGkop6kHrsKPC/mLO0QK0mDEZja8uFIX5TDmHSEoanhl1QH/wwDvHKTO+g57PJeh57vH73Y7mjzZ/bqFR5H40h+XOP8fQ9b8KePV9T2oCnDvo5ePPRaS8Hj3H631Lg4Kvc9SMr6Faq6sRSd2Hrvf2OBGK4TlM9UPyaYw1QPf7ONKWYZwiIzZnNy6kdGPMY8DynDSmswvJjUvbkJGzARk9U4GnCzy4N6D0gLjc7x5kuN+J2W5kGF+LjMdevz2l+ynVIHjXHeXK+zrXjtlXgf+NSEwvkbhBtd//W4jUHCIyaF9AnuKjJKeZ9aBxMQflT3+c4qalAQuALThXTqH7hchoSyrED80Dr0aevgjO7cbsK8gDfQk6z4MofS3sew1mM1F0zYBWnBtMXU0udqLrN5t4UmAo3eKHeSmTiuZdjTyYk4ABzDYAL+eSLL/wD817qijY5SSn2jUgw/lCdP/j0I+MzskkF9dPXIjUP4Lu5a0oKrwMjZWT/u9T6Pr8JpeEloTZbKJeTAfTUkm9vPNdfp+9SBBgLpHc/QkUZTwXkYLFwIMVyDCH3iIXIyLRi5w124nSlcoR0sib1/xxr0JRoGkokfkdYK1z7PAqaV9H13QVkQDGW36/TWjeW47GWUvCfo+ic14FrC3vjN8H0Hh8lzJbHjhHL5WrPw4LPHEaEfKUIcNQkRGbsxtzKc+AGUALWWXERhGhUCvTNgwG5kjhKZR73kyxCtVkZNyvyZN9VdrPatJTOkKjvP8DEaVjiDD0+n87UbTndpQeMh14KrFBZDw2otSllYjM1CHjLERaqv15rQGW+s7rzf4YJhMvJzwDGX9HkCEcBB1uJD9nvwkZ+7tQil45WE1pNZ7DwGrMnixKBdI92ORf8VCNyPDVie6nY6wAACAASURBVDg3gNl9KDq3GF2bcN2mExXibjjzG9VU/RkiX4f994On+grMfoZzwxkRuIT0NLggIjGLdMWkKs5mg0TE8zXMXkeEpo6IyPUB+yuSd5bgwp2IhISxOoDZq0guvTP/66xEKUU3EdVLiJxHzZP70fN1BXqGf+4cr1AGvNzvLSitE0SUqpBYRDtqULsBpQolpbbNBbY5F/XuMmMaSh8NReB7EDlaAlxsxlPAU17s4j3ie1O1A+1m/Dala5eOADeb8VLm8c+QIcNwIiM2Zzd6SFYxyUVl6kBKN7oDRSRCfUYVZvuBR0dNjKBcOHfCNy79PURwchfSE8D3cK6wiDY0uytlBDaiVLFf+b+vI+or0USkOrYQeSrbMDuOen6UTlGTF/pHwOf8to4SqYg1I6NtMzK+puccexMy5oIiU5AUNZS69ZY/riZv2D+CPMs3oPQGQ8bHT4DNFRiDU0lXWMMfS60/nvHRx0UNKP8JRWBuRBEYQ6TuF8C7Z6I1Gv93IyKRK3YwQFRP9EnMDuPccCk9zSA96rMPGayhjiqO2ExFhn25UuYTFyLMZdesxMJsKSKvIa00oBo95+dg9t1Absz4ICIcCxDZDc/BLPTsvozG/my/jQbgHp9SVi5uQk1jd1Gc4jrFH+8aROqWkC+tX+3fbyenJ5UnS5/2v89NH3Vovmn153UIOVpKoZnS9Xin/fdqGS9zQIYMGc4KZMTmbIE8ixcj4+UkMnZ3oCaNaQiGb1LTw8L91COjbhFaNHMX12nAlzCLIwpjC3n5v+tTehYgL+cJoCWhgDgU+gZUoYV4GSITvci4WE4kcbofNfq8FBkVC4kiOL3Aw367HwUWYvbTPHKjWo3ZhEJ959r9sZ/0xOwilGITmiNu9OdwOzKkglHS5o9zwO+3xW+zCxknociz7syxy2jfDmz3x1FVUapOhFC3lWas1PjjGl+RA13vZ7xqXwPQnxBduxARx6R0uxCtu57hk7DtJL3p5wEiMh13XQ2NrX8bpuM5u2FWg4z9kxQb6f3IObAYPY+/MmMFcva0IBLalPP9DmTAX4PUykJkfBEVGPVmNCF1y93E1+21o2fvBuBfUUrZaiLnlgNeQQpkuee0GM0pSeN5AM0vHzBjUxkRluBQSYswVhFJQmfIkCHDsCEjNhMdZpNR5/nz0UJyGhUe3oaKadvQQpvUB2UB8LxXKisHN5C8CLahhfrTmP3PcVmgrFqNcmpFjqNrGTz3q1EkpIuoaPYGlHq0CaUK7SHql/O76Pr3IaO0yf9uANV0XI5yqdd7InElumfTCKRKjRCfwbkD/v68AbyBWbXfTj3w/yBClXut24iiOo3+vbjUrmriJEFF9AarFvUKqtdJU8hpRpLaY9OlvhRENtOOP6jKpeEIcClmDw5TZ/tSTT/7iEQQJqNoQbi+U1Cq4QtEnd4zpGM5chIlGfsgMnkdZs+AuwmRoAEU5VhEfuSyFxnzC5HDqcZ/t5KUygspnUp4DK0Fk5zjcS/TPMt/dty5WKnpiylNsDrQvD+H0vPnm8ihlibmMRvYMKx9nDJkyJCBjNhMbJjVoSLVBcQXai4jUrwKHatr/PeXoQWvBdiDWVVJQ1P7u4H0upMutPhdQG5NwkSD+mK8iHT7lyJjMXhaG5DhM8X//zKUg++Q1/Skf38XMm6CIlvu9T0C3OqLzT+BjOVDRB5+Q/foQp/uEqX3hSiP2Uq8Mk3MGbyN0mXaiIqUcw2ixcAG32B0OLEFeZWTulQ3oOsxkbuMN1Ja9jl4tYcr1WY3MhTj6sRA46UWCU3MQIXZQTXuMGqOu3EE+0WdbTiX0vWJfUD1Ji5ehFJcQ8T0ALr+1eRHVjrRc7cDpYS9XKEk7zzK67cUaiaPe9WxUil5welSCgOkRw0D1iNik6S0WIOeoZfL2FaGDBkyVISqsT6ADEPCBcjo3pfw+T5k5DyNPLUXI7Wsm9DCtxUZ0ncBn/O1A2kIDSxLLYKdqE5jouNltDAvJj99awXyOE5DHtcaZBxMIsptr0PXYB7Fuev47c1G6UrXIIKZWxfhEPk5AfyxTwEsxLkk17PsR6lqU/0r1AzNRITpPeCh1LMfDFRv8H10/ouRcYP/e6Hf/5oRIFSjieOIoKWhBhm1wxO1FCH5IRojS4muK+ianoNEIJ7CuZ8A/wP4OvB3wNdwbkNGaipCiIqWgmtjWi35dXs9yKkzA5HNgAGiOpc2Ku8M38tI1ExqPE8q43tVxDcXzYNznEBjtRk5uXJTemeieeGXzg1bmmaGDBkynEEWsZnYuInSqQzHUZrUgygq8CwyjqpRusQl/ntKvTD7aooBVC4RdpRagNVIchah0Nm5Uqk9ow/nTmO2B3lYZxHVKTSjc9yHSEsVOt/gnZyNDIDjKDrWSrIRcwvyqCdd8w5kDFyACv4rwXZEjq6EM70uHCKzJ4AVmG0dplSpCM7t9zLVl6FC/LnoevwGeLNIVlyRwMmICLRPAAP8VaR8lfbshXS74ashkAjGN1EaY27Tz53Az4BtZ66dc6coLeKQIRn7UPppGgyomknrQURgc50+obh/FZojQFGKE8jJ8WBBnUs52I7uexrqEAEqr2ZS2ITq9NIwHdjtXHnKmc7xthnfRHU+oZmjobH6U+fYXsHxVQQzZhHJwR9zrmSz0sHsoypLo8uQYXwiIzYTG/NJjtYEnETG9U3I0DmMIjcriOo/QIblHwJHMPthgnFZi7yQfcioSzKIGymOUAgiNJegepK5yLNYhdlO1EdmfCmqySh5Ax3nLKQO1I5ISzVRY7ppRIZNNyI34dp2EwkMBNShhTc0fkxDB7pmhcSmBZGHtN+fQrU8T6C0ump0D2egSFEXZvfjXOnmm5VAhfgvAC9gZrHjyWwaMh5XE81FhzF7FimxjVfDYTtKN5pLvipaQCgcf3XY96yI2FrMQtPP/mEnphlAz0w/6RHqOcCWle7t4xhr0ZyWW1eyF0VOZ6M5cS7wz86xbpDHtBM969NJllOeDzxTSSNY5zhoxhZUwxMXRalD88UDlRysc+wFfugbiDagZo6VkrmyYcZM4GOoxigQKTNjD/CwcyXXylLbb0TCMDcDM8zoQ/VErzo3RAW+DBkyDBuyVLSJjdBHJQ0hcrIKee/PRSlSrcgg7yW3P4N6GeR778wWYfYl4N+hRfoOJC+8iuKc62CgJhUp34E6ctcR1Q3sRkbClzC7rMT5jDaCGlU3MiZyIzMhHakXkY/w3WAQnUYSyt1EUswBc1Fvm9KSz5E8ciG25OwrIETCLkKk5xpEbn6XqD/FEf/a5Y/7c5idU8ZxDA7xpGYO8GVUsxWOaw86l88CH/eiCuXBrAazWZjNxizuWg0fFIVZg8bDUnRvJyFCswQ5Cb6Hc3HNPofrGAZwrisjNSMEEchHUFQ7rq5kOnrWfu3/Xoee/5kF3wuKYv0oYvnG4A+JAeBHRCmvuSledSgdcSdKSawUD6D01FB7WYfmt4VorvqJc7G9a8o57lPOcXQUSM2fo+cxzCW70Rw3Dfh3Ziwe4va/DPwOmo93oajYZcBfmXHNkE5ghOClvDNkeF/Bxn/WR4ZEmH0ckYu0tIP5aME633/vThRxCGRmrn8ZUV3AUeAfUG+W0IjwFCJDDSgdogYZ+SeBF5FxX4cMgV/gXHFHabNzgS+iRSHOGz8JpfD8vZdnHnuYXQH8ATrmSUjqOhTsT0PGrPN/H0PXcRKRotI+ZAS9TFT03YxIz3eB/4juSxrBWYC62T8Wc3zXAL+PIgg1iMhMQ4tvUESrR/U264lS3kKxOeiensK5fyp5PYYDIiz/0e83LtpkyEj7Oc6lRz1Ue3QNikiG8z2NOpq/ktYhfsgQgToPiTTMQul261C0KU1VLcNEgZ6vjxA5N6rRc3YEuD+3KbEZc1CK4iw0X/aiMVmHHBw/TVAlq/CQmIOiQ0EgwtCYfxFYW0m0pmC71SgleTWac3pRrdCb5aagjRXMuAsde9JaGObEf6w0hcyMGuAvUfQ+rjawFq17/+IcY55xYMYUNDZuROfdhaLHbzjHyDlbMmQYJ8hS0SY21iF519CLphA1aFF9FaUZzMGnBKBFa75/7xRaIOuQB7+LqOHjIpR2FQy1LpRidBXyTi5CC+Euv82fk5yCcwORJGocuv1xXAY8k37qo4a30bmH/kAnENnrRmTE/Ct4I4MS1T6UsrTA/3YaUVPKPcgoOo7ZOnQtk1IZAgFZH/upc69h1oPIzc3+3dM5+zmAOpwv9ce7BRXvriCKOLUD7ZgtxLkhpWuUiWVo3CVJ6TqU4nULZusSU9IkdvE5dD4HiUhSHYoMXorZd0aMZDjXi67nlhHZfoaxh56vjURCIP0oKrKrcFw6xxEzvooM7FUocncUpZDuK6P/S5mHpMa5ZjyG5uh+4IhzJZX6Sm23H0nUD29a6gjDR1NWkpT+LLShOXAJ8QqiaViOnH9J81UvWhtugbElNmbMQ47IBjT2dqH58EbgJjPudY53x/AQM2QYcWTEZiLDub2YPYryio+Q30huKvIcPoIMbIeMvfmInNQib+IR8qVpjxN5JG/w779esOdTKN1hOoo+zELFy1sS++Go4d0FlG5Y2IqMgvFBbCQgsAal6E1BkY9l/tMBdM0OEeXhh2fqEJI0no4W02fRdd8G7M1Jz3oREbm4vPkQuXgN55Kjcs695Ru0zkILeGjMORtFMpr9+1PR/W8gqhMCRXQuAj6D2T8UFbwrwrKEqBD6hL8OewZZ6L+S0rK1QRq3mWQv7J2IOBYaHD3IyFmAUvDuq/gIzaaiuqbQ6HQ7isSMXASoEqhWbTEi2VXoed0xrGIFGQTNaRv9q8RX6UNqkyPeoNindo1YetcEQjN6RkvNRf0kt0ZIwxWUFuJoBc4zo8m51B5YIwYzgqOnj/x1tgc52hqAu8z4mnNl9XIrtb9GNJdfjWyJ46iP2fahkuwMGYaCjNhMdDj3AmYngD9CtTGT0QT/LjLo1qKGiXOJ6j5CLnidfz+E13tQdCJ4AWuJmrvF4YR/LQaOlGjymdv9Og39lCc9OnoQgfw6iqyEmpCFqD6oiyjyUY/ISCuqMapC1/m7OFdIDsO2WzH7VyS5vRTdm36/zRqUwvZI6vFJVexaFDnoQyTkOrTYn0b3tdq/vwAtPrkG8Gl0H89FndSfytn2NOCPUWTuNIpUrUAewPe88EClC3k95XUcdyTNUWZTkNpbWoTpAHARZrOKlNjSYHY9Sj+CqNHpRcBHMfsZzo1tfyaz+Sg9ci5R35IaoMM3BC1uupohw9mLSupIBlNz0kRp2XbnX5NIb+47klhJfGuBgC40714L/HIoOzJjEUrLDoSmF81HdwEHzFjjXGwfswwZRhwZsZnoMGtEqWDVKLLSjQydScijfQ5SQXsdGaYfRpPQJP93FVHzzpCCMIDSeuagMLyRTkhC+lU4pslokl2IjPTt/tWODPa0PPMmSkd1Rh9KG3sKEcZtwGfQte1E5x/y6o8iz24vIhJPUapg2LmDmP0DutYXIcP/MGqgWU5O9BR0//tQlOYqtNgEgjSf6P41IA/bESJJ6kBItwI3Yfaij1TVA39KcQf2EFlaAnzWp3tV4qE7gqI/AbX+1UvU+NLQmEoyEhZTul9HqD9YQmnlOb9Xuxr4PWQc5JKvVvTMfAazHpx7p6ztDTfM5gJfQs95oQHTANyN2Rqcy9LjMrxfcJTyhJCqiVcxLIU2lIaYFh0LKclDrqEaAq4lisIn4TBwtRm/8qmHFcOMGSjdrRPyanZO+/3PBe4245tZ5CbDWCAjNhMfv0+UNxyM2EbkpT+GvCrrkJG3Ccl2Xkqk3uXQZHwATUwzkdhADzJgS42RMKEf9+kxqxF5qkIG+nyi9KW1yPOflgs9hfHYkdpsEmpuejG6Ni+i63QpMoA3EtUnnUAk4RWcS8rLzodSiAabwhJqfUAiEV1EYgRdiHAFydk+RB5Po+hesz+f1/z3ZiHSsA2lYjWTnFu+D0WZzqOyOpNNwG+hsbgCRZEC8Trg912N+rIkydrW+FeDP/6kRdoRryhXDEW+7kTnFRdR6kak7CO+/89YyFH/NjrXOAOmC6XtfQKzbRWSzQwZJiR8bdN7iHwkOYKaEEEZTA3MmygdLQ2zgS3OMZapqlMoTaz60bxZS3mKnHG4hkg4KA6H0LqwAtWoZsgwqsiIzUSGUlJWogjHKuTxH0BGWQ3yMC9AKUb70WT2BprQLkPkpt3/G+SID6CGlCFtrdVvK8lImosiCycxW4283YdRmtBMZHT2oIn/T5BBFppEFmIxMuzHXFkmDyJsn0TRlFwj/ziauKci0vCNilKeivfTiIjTDeh+dCLCsT7FwActMK2IlDRTHJ04iBaZGnS/g7HfhojBJCJJ21wicDPJhkLACZSWVj6xce4YZoeBT/nthzFhKEq4FBH1NbG/N1sBfMjvt83/tgWNm8IIj5G8ABdiOboeafnnHciRUIqgDz/MZiHimhbR7EJj4FykxJUhw/sBj6J2BDMpbpzbhObG+4GlJhfQsQpSpXaitOOkvlWTkJMoqQnzaOEkIjdpqnhhLR+scl4Vcl6WqtFp89/LiE2GUUdGbCY2LkaT1OXEp9vUI2PtBjSxdyJD7wjyQl2HJuQaogLDGcib7vzvn/Tb3kPxZDgTEalnvELVhxExWo0M/dzj6UJEqRaJCNT64+nx+6lFxvEDODdYT9JIYRGKXrQkfH4SXcfrGWzusvq6BIGCVrSA1gEfAG7z6UXxfSScG8DsOZQ2FhdFmIYWmmqiFK9GFIE47F/ziHpznPKCAbMobbx3oAW/fOhc56JFbyEihiEtMiy8vcTltZvdiqIqJ9E16kdEfql/rSUybCahMVZu/41QW1YKzh/zaGMWUS5/GnqQQyMjNhneF/BNRr+NHFBLcj4K6WFHgS+gZ6cTGDBjE/BkKSlr5xgw415UmL8UOXM60Fw1x+/jPufGPIX6FdRkO82R0wy8UqnkdQ4mEaUNp6EL2RIZMow6MmIzsTENkYKkGoIm/6pCRu2Jgt+9hchJ6GGzAHn0T6BJab//twtN4NVoQqv22zwE/BjnjmJ2uX9/tt9+3PGEVKe9wHP++03IEN2Y2xNinOFKSqt4HQauwexJnCtVaJoP1bJ8Dl3T3MXxNLpWjah24mspNTdvIbJ6FbqHIZWqCpGI0yhy04GiEieRJzIscL0oqrMNqZ0NYNaP7tEUotTFE+Qb/2nRvCRc5X/zGorQLUGkog/VMB0gqhWKeveoD9KdiGz1o5S264iKYmsRqX7C/2IhIsrFx6dalen+/A94AYReysvVNwafxjEUlGuMVFXw3QwZzgo4xwEzvoGe+3noOWhAvdOWoLnPoWd3B4rAn2vGv3gJ7bRtnzTjWyhiehOan7qROuh46fPzDop+x0WtQOuIoXl3sOglqqlNm2PqYEzT8jK8j5ERm4mNdlTAHmdY1qMJvg0RjZlERvMpZOhOQ0X9t/ttnUZEYwAZmO/6/y9BvWl2II9P6Ly8O0fudx7yFF9IuipMUOo6F+e+XuH5jhWaKS332Y8m+8mUVtApxAWIQCbVsnQiI/xalHJRDOf6vSz1+SiCFwzvOjQ+tqMowzy0MB0kf2HqQ5Gpb+TUjhxHcslBStXQPd6GoiBBNOGlSk4WEcVA0NqIl9E9SiGxUWpcW865hXF/KZHBEua0jcDDOJe/iJstRopnS/x5SYDA7A1EDEupJlWRnEo50gjpH6WMimrGowBHhgwjDN8raC+w14wFwD+itS838lyN5slW5Fj5JPDNMrbdg5wpm4b5sIcFznHajB+gyP0SdH6daA2YjeaMHwylSadz9JmxEa1Zaelo0xHpy5Bh1JERm4mNLSR7Z2YShd0b0b2uJjIK+9FEF1JW1iOPdx+KtuSSpX3IyPw1ziVN6sHbPZXSyiwdKAI0URA6jpdCFeXJGBfiOspTs7kGs8cSi9ad68PsHuShDPnjDhGCTv+agshrKCA1ND4agRdx7i0gSB6f5793mMiQrkFEYgqStK5FhKAYqhm6yH+/FkUA30TpDHFkvAYRQ/yxRrLfSnVcQWSgLERE7xQSx5iOxl49IqJbKSRcZsuQmk8H+SSyGpGtxSjdcD7JvXMWAK+PWNPPNKiObQNKQU1q6Bqev/FVp5YhwyjC14J8Hq2Dhc9KP1FN4iyg3owFziU+UxMGznHYjK+jOeJ6FK0/Bfwa2OBcUa+0weAlJKZQR3ytzhTkvMyUGTOMCTJiM7GxDxmdsygu8p6JDPJGVMuwyL8XFLNCrmw1SmNKk8EMKUlLSM7b34EiP30ke5SDt7uD8lJ+xgs2IA9V2qIwDTXeLM5vVs+VkPZ0BOcKF4OplI7yhFSrIOoQD+e2Y/ZzFGnpQfVUJ/zx1SDDfTNa8ELNxnZ/bA/4450DfNR/rwMJU5wiSvk6iiJz1cC3cK7Ycydi9Jf+vDuQcbESqfQtR+T5kN93m38vSIvjj3UnZpNx7pQ/9xAdrEdE+wQRkTzqXyFl7mJEqjZhNhPVSH3Bn0chKelHXt7FiOA3oeflYM726xDh2UuU6jYWeALl+c9D1y+33mYmSr25Z4wU2zJkGC9YhtarNJWw42jO2YwcFrHExoxa5HDpBzp8VGjcwiuzvcbQUs7Str/XjAdQpCtIPodeWs1oTvqucyWzHDJkGBFkxGYiwzmH2b3Af0BGagcykI2oweN+VE9xBKWWLfDvtxEZm+U2S0yTzd2FJrhWFPaOU5yZhlKYZjBCk+4I4R2iaEdcLwND1z+/kaYIwu1EqVIG9GL2IvBCTi1OB6VT2EKfmtL1LM69gtkeJMt5Obr/zSidMEgZ5xLZJjRe3vV/X0FUlP8eOucLiYpBDd3rlqLGo2a1KBXic/53HWjcXI/GxW4ilbx2lK8+A43P40SEeI7/3Rcx+w4RIa9B0Zqg5Fbrt9eGrl8dIvtHgdsxW44iYjMQIQjpm8dQpCfX8DmA7tXX/DW4kShS142ENF4u0Yh2ZOFcG2bfRrLPq4iITRV6Bh/BubSmpRkyvB+wHM1faamlA+j5nkxMRN6MqSgqfAOadww4bMZzwKYhFOBPGJhRTVRfe9I5rX/O8boZh9C8HnqS9aJozmvOxWaRZMgwKsiIzcTHyyjVqAoZbEGp7CjyaB9HRuRGZLjlFuhPQV7pcgumkz0wKja/H3m1z0UGbUh7q0ae+2MosjOHUk0rxxPUrHINSmOaTH5q1lREap4jV9pSUtxfRNdtL5EBGpTOlmP2A09uXkHer7QGcM0oBaq8onXn9gMPAQ95ueqPoHHSSKSaU+W3Ww18J8dgv4j89MagnNZIRCS6gCWY1RVEoD7mX0eIIiPT0HUKBGs7IkohBS6oooUC3Cn+u2+isfIxnPshZq8iInQTkZT4gD+PBWisn0IG/kkkCb3T728KIj8nc/ZxI8oDD8ffRzB0nPs1Zs8TyaAfHzd9YZxrA36M2ePoeTOgFecG03wwQ4azETWk13oGhEL4PEPcjFkowtuE5r4wR0xBzZnXmfHzs5XcmFGDHGO3ojXPAVVeSe4Z5zjkHHuBn5jxIFoXugfb9DNDhuFERmwmOpw7jtkDwB8gD3ToU7IYeapno/SjuJqBmcAvkExzDcmRmwaK6xLijuUAZl9BC8LHkLF6GhGcd5DhOg8pVSXVMIxPOLcbs39C3rsr/btVKOr1K2DLGSEFSSV/Bnn5C9PXetB1XArcgqIA76D7Nj3m+xBF314Z5LE7zB71+72NqHAelHr1QkE6WRXxkaFClZtgFAiKUN3ojzXXUJiPxkEgDnNQ5GYKSonrQOQwkI52RNh7ERG/GLPpiAz/JYq+5BLngHnICGlF13Ie6i0RomW5KSTtaPwvQcSn+JxE2Er1axg7iOCU24sjQ4b3Ew6h+eMk0NBAZ3cNfdZJY39/vtlTi+aLHeENX5/zGf/Z3oLthij0Nf6zwc3J4xie1HwKRa8PEM3lhuocV5pxj3Oqd3TujDx/hgzjAhmxORvg3OuYtaP0lMVEIfZuZHivp7j3xVxktL2FoggfRQQo1wNlfjvzkaxzaW+Mc0eA/+EN6Q/54znlt7sfeA7ndqRsYfxCxv+D3lM+GehLaJy5DBnqaUTwAHADZs/jXBdm30fFrotRtK0LLazN6D7cG1vLUv6xO1RvshlFUGqBDpyLy0HfjSIqaftrRCQsN33uUqI6qoB6on4y+O9PI4quhPqtav+7V1C0J4zDsK3FhLqbqOi3nSi3uwZdt05E5pcQEWv8sS4rOIeTaKEOCm9GiH5kGHWYMRNFeyeje7s1pL5kyFAh3gH6b+G5A1fz2gcuZktdFQN9XdT3PcPtR57mjtZjzG5CjqYfFkQalqA1L2n+dmj+vs2MdWdhlOIaNJcXCpA4tCZMAT5rxlc8qcmQYVwhIzZnC5zbitk2RFgakQelE/g0mqg7kVE5yX++H/g3n2b1IooK3ErUmHMFUY+bx6jUc+3cBmADZkGp6nRsYf1EhMhAWlHqckp3du5Fz988YBfOHcTsayhf+UZEaLqAF4E3UvrXVAYRnFLKOK+hqFRhlANEgutQ6tf9OXLfEC+LHQhLIYIIwilEJJrQmExKp6pGEa6dyKj4MFF/pk6/jVCrdC66rlty9n3Q/z9XGbCPqEdPN3p2Np4143SCwIx6JHZxGbpHfejZcGa8BDzh3KDUBjO8T+Ecnd+2Lx2dxbEPdVF/eCvnz+6npqaertoP8utzrmHd8n/grw/uZ+F/cS4nhVi4kNLzd+jtNpdkhcIxhRnNqFZwEZrztgCb04r6fU3NLSSrQoKcDkuRamamfJZh3CEjNmcTZGTmT0hm/4w81VcQ1Tm8Cew8o5ykf5/AbAfwn1GtTjtKbduPiNCXMXsE59ZWeEwnSe+EfDailvIbOEZFq5IQXgusxcwKSENlkDzySqTmVo3SJjbgXHE0QjU4i1AkpR8JDLyOlMd2o6jITkjOJwAAIABJREFUdNT7YR4iIZ3A1ZidwLltfktdRPU3QQq0n/gC3hrUa+XCnL/TivJPEqWwdaBeEksRocm9Tl3+XII4RkAvUre7knw1NYj68fQDT6UcQ4Zhhlecugvdy93k38sqVE812YwHztZ6hgwjALOVX4CFz3PLC29x+crptB3so6aum0kNB5lfs4g9PU/zgecuYOtvYvwu9ZSXWjVAuqDOmMCTkw+jwv5etJZXIefBR8z4sXOJhGQWcg6Vaj/QidaXjNhkGHfIiM3ZDpGW9/wrGaoL+S1k1BYW9p9EY+V3MDuKc1sHdSzaR/Wgi7DNZhA6Gjs3nlNUDpHbgyUeIe0pPnoyNFJzHapFWYJW7S4UcTuC2ZPAU2dIrdkKVA81O+zZ/7uBKHIzGwkK9CPy8R4ix1OAL2D2S5x7AYknXIvU1a5CqWGnEcEJNVyByLWiRp/LiGTHc4UtAhoRSWkhv1bmLX88y4i8/NV+W+3APageKpfI7ULGyCr/vXBM8/2+f4xz46GD+PsJF6EIZ0vMZwPonl2JnCxZb54M5eL2Kjh2G8+3XcaG3Zu4ZN4R5kyvpn9gCbuPXMzmw3X0LkYiLu1oTpgDdP81f1f3z3y5oZv6UvuoojyBgtHGB1HUv4V81taGnFJfNOM7zsU2Gi3XJuxnHJK6DBkgIzYZIixFnu6kvOI+5MW5DTU/LB9mi5CQwWVAFWYdqKj7rbIaHZqt9PtdSOgWb/Yu8CzOFRZ3jge8gxaUNEGGWcC22AjKUGB2G/BfUYrXAaIFaA4qum/wx/SMv65/gkQdcjtzV6G+L8eA7wP/N1LV60D1L4FU9vj/f8TLS7+HiEQtKsZd7n9zABGQThT5CUW3A4hAfdB/vzCyNwmlevzAq+697bdzJOe321BaXEi/BEVd1iKP5STy64D2oEjOXJRK8SukjLZnSGTy/Qo5K85Bz/Y0ZDy9haTAy4mw3EKkhheH0PdqNRmxyVAO5ABbgJ/TZnCi+2Ze2EXx2tYN/AVyvoTU7Zq/4f+bdSkbLv8Wf/HYq1yXJM4xE9jpXOrYHXV4ieqb0LnmzmchRXcpWgP+pxm/xKt55vTmyVXMTHt+G4l3RGXIMOaYSE0SM4wsVpGeCgSKLizxC0d5UPTgy8gzux8tNp3AncC/90paab//AGrq2OR/u9e/lqD0uJVlH8toQQ0lH0fHGOc8mIIM7ieHdb+6lv8ZEdCjROlwvf69PkReP4TZLOAPUXSpMPo1gCJ3c/7/9u48OK7ruvP49wAgNoL7Kq4StdLaZVuyrMW24jjerUkydhLbcTJxkqmkklTNUjXJzKSmMpNkktQ4E8/EcVUSexJbkld5kWxHimVLoqxoJ7VQGxdxX8AVIEgQAIE7f/zeYzca3Y0mCRDd5O9ThSLR6G48AN3v3XPvuecAH0HpjS+g4KP0vsMoiLo5Ky5xF1o52Y9m2YfQeeY4SmPbAdyXPU++v+VhNCC+FgUu+eb/ecA9pJTnwP8Leh0Un7f6UUD1IqpudhyVxe5BDUeXoAF3sVb0+/8mcDcpbXNQcxoiZgG/iaogXoleL1dmn/9Gtr+uysNpQQHmeKuvPej1YFaLVqoPynOXor2kW9G16TCwfzY9r13Ilq2/y2c/cAWvTC/zuHZ07vrhRB3wBHpT9m/xzz+bQjXMXgoBySJ0bX1fhNKFU6IPnUsXVvke+d7GFyfsqM0mkAMby3VRvUFkboTx06xEzRHvRIPkvRROtsfRzHkT8Ekiyq8c6vE/jQKa4pmzhAbO3cDHxhtATZHHgfvRYH4FGqwvRDNmLahvzERvOn03hX1U5eR7XxaginWtVC+CsBv4KcYvKbwflWRuycp4fw6tyAyjYGM98DUU4H4eBRrL0ettHRrYHkWvq0vRvqCXgD/LilBISm8AP0IrBHPQa7Y1++oM9Hu+/2Qp8ZSeB/4OzfjnK5Ir0Izlt4Fv17iqYKUi2lAT1vloYNiNXifd2ecLgE8R0RbBtIiTf6ditQaTQW0DVTPQxFme6lvJbHROKJuhcCs/eWo52/e+jx/chFJVu9C5dQVarflySuO0P5ga8xl9HW9G2RJ5E+PiapODaBX0dgoBEWiiCcZOCEGhX94aN+G0euVUNMv1oKXq8eSz77W4HQ2yK6VjHUAXl0tQ+lapW7LHV9qI349O5Fej6mH1QysAjxGxDs1iL0G/hw3ARlKa2CpPSgm6ifFzvo+iYGs1Y/vSlDqB0sraqD6rnu99aUYlsA8B3yfiIZQCMQz0jloVUcGCO1DgugetBuVlwDuB69Ag+Ucl32sT2jvzbgoB9gG0L+wfSWn0ZtaUNgF/k61mTUcrSHtqbnRqlbwJzfiWHdwN0dL9Y951w3/lv/9nsgm0CPaj1Jf1KTGYEsMRbEYBf7VB0lwaqaGvTa2UjmSpyitR2mo5y9F5q2wqcytDI7ez5uHLeO2Su/j4890sWoTOh2vQ67ce99aAronF47pFaCKnXMrcMIVJwnehCShSYl8Efw98HP0O+7P7dqBz/MNMdLaB2QRyYGO5F9CehGrmApsr9G4ZLaILBSzbx7lnH9ocPDqwiWhFVbjG20NzEFV8q6/AJqITBVxvRylVA6jS2MEJD2qkhbE9ZMo5gVIphmu4LygQ6kIXv0rasvuNLpGa0gCVVwGXo9WgrYwNXI9lt7+biA2kpNeQ0hrvREH4N9AM6jS0WtNBtRla9VeqNMg5v2kP3FvQ+zWhVbanx1lRvJXCYCnQa3wR0DxEc98f8/sLn+TmlcvZtvApbvpBdr8ulP74lgi+nNLJcua/glIly70em9Gq3LNn8iPaeedhtELcQflV6QtRJkG1a9nwYrqP7WXxoxO+F3LybEBBSm4pYyci872f+Up8L7Aigtkp6feREjsj+Aya7LwCvQf3AC+lNG7FNLMp5cDGctvRsvQyytflb0VL01+r8fla0UBlvMHzIJpJL5VXz6ploF4uzWXqaP/Kr6J0h/0ola4FVQx7OxFfG5ViNTHyfTR5c9ZKKxLTsvs+gzZkV9v82oxeF+OdJxYCD5ziPpUbUQBT6TiH0YDkRmB7Nvj+cHY8eWCYX5j3o+Dqo0T8VQMNQqaWVvnej4Lv4xRKvF4L3EjEI8A/V/i7LkSTDitRgDoPvfYGNrNq5C0827Wc7Vte4uri105f9rEc/S2/igZiT6OmgDsZHRx3oDSgB1Oqz14hVqdS2kHEl4GPoVX9fI/hdBRg76Z8Jb5yGml1dzv62RagiZxyrQdmoVTfk+mdbRxv+kP+aAnxp21o8m0o6x31WvZh1jAc2JiklIj4CqqSdSG6EPShwe189Fr5Biltrvgco/VTyHOuNuBtp3w6y0D2HKVVrUrlRQXqQ8Q09DtsZfRxnUABYxvaF3RwQiu66e/3OPrbLaBywDIP7X/5MQoa8n4z5VyANsguQjN/Oys83xG0V+ZUXMn4TV/3Z/f7ZnasA1ROaxxAr7XrGJu+ZuXdTvmysHvQ6t8d6G/7L2UeO4iCkdtQoHwAYIiWaS9y9dIOjvav5tUlbQyWmxHfCVwTwT+nxMEIvo1eC+9AwcxI9v2PAF/HaWh2OlJ6hYjPoJXz69C5bivwFFrt/dlxnmEm2hvaMH3YUiJFcA/waTSBcAxlWhxH154ZaEJiM8A89rffxBOXtnP8mn/L50Hn14GsafcTpFRr2rlZ3XBgYwUp9RHxdygF7FY0KzuIBjZrSWm8gWjxc/UT8UL2XNUe10HxwEWBwTJ0EXodXZCqBS5daKBeLy5BgUWljaUD6GJzC5qxnkjPZM/bggKOXgolkJtQgHIc+L+kdIyI+4GfQwPN0uBxIQpsH0YXu19A5ZuPUyhCMB3NCn65prLdo41XThTy1Sftx7mO8Tt870dpiQ5sxhPRjiolbaf8xMMIGgDdQcQzZXpPHUV7uvLXAwD7WNA0xLQTMxlu6aNr+gq2Hb2OtTPWcX3xHq18Jfdi4GDWePOxCJ5Eg7G8qMWOlBpqttzqjRpE/4TSVGW9/t+Prh/lzl15euVdjVQxMYJZqFR/QtehFdm/zejnfJbsPb+IPZ138u1bDjF7/kW88cpcDm3JnqYV7X28goh/IKVqBWbM6o4DGxtNA5j12ceZegyltVTKc16MTrJbiGhGKTG3w8nOaK1oxn5GheNZhvYDVO9vEbEQzVqNoI3jkzkD92bGL1+7H7iKiO9M6IxYSgeI+CKqVpWXS56B3ucdKO3nT0lpZ3b/p4kYAj6ANuwPU0gB3Azcm5VNhoi7UEnwq1B6w0HUr2Tzae4Z2oVmRKtVXJuFBtfN1BYInUADlckVMQP9Do5m+4ga0Sqq91kCTWp0oFXADSdvVRXDBdnXDo5+wLSs0mYMNTE8fZDWvg9y//J1XF/aoTzfjHxSSgxRKCBhNnlSOp6d034Vnfv2UQjwu9DEzk+YmOvgWRHBajQB1YRWUNej6+vb0Tn0MYomsH6GB27op336YWb3fYyvbSh6qkE0MbccFWm57+z8BGYTw4GNTZ6UdhHxJeAX0QzYAQp5zrPQ4Pbu7N53og3MOxm9ybsXbYacjVZwBtGFqBXl/n69YoWriKVoVu5C8saekLKVpAcmKcCZyfhls/MLaDu1V5irTUpvEPGXKKC8CaUR9qJqVM+R0rGS+68jYj0a6M5Bf58dJ0smqwjEbdlz5eXhE3ku9+kXQlgDfILqgc0c4D5SOkFED5UD5Fy+gjTxtGp0BUqXWoZ+ByNEPAs8TkrViivUo84zuO8S9P7bjf5Gg9kH7QwMB6k5GGk7wbTeYZrSCrbOLvOcLTRQio+dg3Su/Bx6T+erHPl16ivAC42yWhPBMnQ+3cvoCZ4e4HtoH9x7gKfb6e//MN9Z9S5+9LaZ9Ay+jSc3LWHPAvQeLr527QLeSsSPst5sZg3BgY1NrpRezQba16DVjBlo8Hk/sIGUhrImm29hbK4/2X2/RaHB5wkKqwU7K154IlagRoH9jE4LC7QKtIKIv52E4CZvJljtQpAHCJOTv6xVlkezj1ruP0S5DaJamfg1tNq1h8LsfgsqPLCaiL87uaoz9vHz0aB4EOgu6RnzOlptu4jylfOWo1WCfCbxMbSyVC0tcTbw/SpfPz0Kat6DAuwDRcfQjCr6XU/EF05Wb2sMlfZV1XLfNrTi0oNWHxeRDaQWsI9WBgeOMKOnlcHUwolpQ0wrnXjIi1tsPL1DN5sgmsD5KhHfQ6/hIbR5viECmiK3o2O/GJ1Ti3sUbkMTSRetZMv+P+EP3r6CLSuWsvPQBezZ0s5goMBuNUrrzidphtH1cik6X5s1BAc2NvlUHrraQPs2Kpd7BZ2w9wG7SOmBcb+fqj19FKWElQYuCQVIS1Ae8TfHfb5T8ywKwqpVG5uPZgPrfWPmB1GwUFrkIA8urwH+mogXUeDzBOozcyFKYVie3b8VOEDEA8A6UkrZKsw9wEey5zmBBtBt6KL8ImqgmQdTL6D9Q3m1n1KL0d91Mi7AV6CgZguj0+GGs+85A/gEEZ9poNS0N9B7oVqKXx6AlO4X60d/q2PZc2xCqXnNzYwMX85rXU/wtgsX0j3SzvGmZ3nz7qLHBgr8H67jXiB2vtEewYZ4PUacLHGf0ORC3lrgwuz24kacgVaYF01jcONn+Z2rP8z9r6Pz1moKkxYD6Dx9M9pXWZxO7XGiNRS/YG1qqV/NCsavbHYInYjHD2w0YzV7nOfcA1xHxIOkdCQ7lrloEDsbXeReP5mSVbt8wPhudNHIV4zyHO42lIL2+Ck+79kVMQetbJVbhViCVtjynwdUQe1TKAjtRIPe2ehvG+iieRvwTSK+mAU3/cBXskael1PYc/PamNSulI5m+4c+gUoMH0UX447sYxtwd5lN7hPhHShQrRQAHMmO6XIUgNU/NTF8BlU2q/Q+WQasGZO+qEHRoexxV6IZ3qHsgyt55dAxOls3seriAdp67udDh9DrZBZKF3waVdszsxpF0IVSgm9BEwmgc88LKJV4OozpMZMHP11X8dK1XfQNAmvR+bqp5L75nrpVKCMCdO52yqg1FAc2NtUqN1UcbYTCyXw8K6m+KTp/PoBFRAyg1Yk3oxnqQTQQ/5msg/W9NVX9UurVJygMtvP9B0tRoLQZDf7uObmBv34tpXwformo9HIP+h3PQxfUHSi17mdRENCSPbaXwu+6C5Uh7QfuOfmMtTbPVHGEv0YX3uuz59uCLtRbSlLdJobS8ZYxfuDdi/Y1NUZgIz9Ar9HLUXCSz9LOQn/nF4GHxjwqpREi/hl1Js9XPw9T9J57K8+2XMLGl/89/+u73SyagQKbV1Gp3e0p1dQc1sw4We3s0+j9uofCSksHStG9jupNbPtu4onL+5jxKsAArQeO0tm5g2Xz++hqbyKl2Rw+vJjdB2fTuwL1ucmLKtT7tcpsFAc2NtUGUVrTdKrvS5mFBka1qmXglG8W/ddoNWhbmcetAj6Z7aGonGYU0Yk6qOdFDTahWbFL0AB8FkoT+ONxOrrXi2bKB52XoQAmH8Tm6Uyg1LM+lIr3OmMLA/Rlz/lLRDx5Cj2RCpSa9jpnL+d7GrW9lk5QWL2qTGVmF6Pfb8+UFh1IaTBrYrgarUotR3+fnSjoeaViYY6UXiBiFvC+7JYF6HqSr0iun0PPn30h/dqmL0zuT2F2Pvh5FGiUTrD0o4IBTWgCplKbAeZxoDUY6T1C17R7+dkbZ3Ck7UrWz4K+nmGaOczs2QeYO+8i3uhbwp4ZaMX9Cw2438jOcw5sbGqpseSjaKa/UmAT6KT+VI3PugcFGNUEuhjMQBsnK5WM3oVyl6+keqPAaxid/nYCpXEVp3KtRAPARghsyhUEaEcbxYtL/AaFSmVLKZxTOtEsfqkj2f1uZzJL+2qlZQb6O+w/g9Wco+T9dKp3IJ9Otc3wEW2ob8zbKASNTURsAh6c0Gatp0KB4ovAi1laqAKe2h67hogNaKXzKhS8d6ONymsnZQXN7DwTwWI0wVYpaGmisHK6k/LZCtOP035sHgf67+ND1x9m9txEbNrFkp5lbF8K0M7xEwnSMTqXbeDiiy9l02dJyQU+rOE4sLF68CLK9S/X3b4J7dNQYzH10JiBBptHKgyeNqJVhWrllOejWf/V1NZ35naqBza3Mn461SHUU+D5ce5XD7ahwKR4Ja2V0asX09DvN191aEbBXd7lupJ+4DIi2ie8gELEYrTR/00UVpMOZ8HzM6c82E5pgIjnUOpbtYC0DaXElTumNuCXUWCbV/bLLQJ+M1sRrN6PabLVGtCMfsweVE72exN+PGYGqnRW7bzVj87R+1CWwCEKhRBa0N7F/pe4at0n+VLnTpYuaWVw4BBz5vXRNbCbC15ezO5ZM+jraqe/7QVWv/Ab/O2mA8x/xUs11ogc2NjU0+DxH4B/hVZGhtHgL08Degw1S3sXCgza0Iz3wWzAum5UPxWl2HwL7QHYw9jeJ7PRIPwBlD423ubIPlQeuqVs3xZVYZtDlTSAzFG0J6X+aR/F/aggwC60Ub/4Z29BM/RPUAh2DqNUtSDbSF5GHgwNM9Hnn4gLUcO9IbTnJz+uTrQiuJKIb57GSsJPUA77DMoHwcuBVyhfaAG02fdCyr8+DqDf7S8S8ReTVPzAzBpXG9X3jPagc+80NEk4EwU4oHPLemDv49wyfy033DhM00X7WHByz2g3C4f3sWDHPPZvW8yeOZ/jtx87wPyZaNJlN2YNxoGN1QdVXrqLiAVoX0oHOmG/joKQchsnp6PN+m/PApmdJwetKb2U7R+4E3WRPoEG3M0oXeZLpLSXiLxWfzX518tPYCkIGGL8dKUWxm/eWT/Ug+huFBS0osDsBLpo9qPqVsUXvjdQtbQOyqehgYKD9Sj4mLjVGq2KfBy9ZkoLPRxDaW83oL1P1Vbe8ufLy6QuQ3/XR1CAMhcFwnmj2TYU1HyjbC660rtuofoAoQ8FvJcCL497bGZ2PjnA+KnVL6IiAj1oL2pT9pFPEK46wkz+B//l0O/xV4sX0p32sWBgkLaRBM1BuryT/u57+bk1D/HuA2gyqH3yfiSzyePAxupLuQpZEZ9m7MbJBag08zyU4/9W4DUi3kAz50dQStqfo6X8hSjo2AFsKxqErkclNKulGc1FpZ+rBS3PoVn9auWh51Nr08x6kdKL2T6K1SjPuw/9nM8zdiXsABrAX075nhAzs9uHgCfLrn6dvstQQFUtHbAbeAcR66qu2kQsQpt1l6AUkBEKQemzFAKajejvvqPKBtv5aFAyXprXMRzYmNlYG1CA0kLllZtB1H9mGKW85j2mOtD5awuwcBOXvPFH/OH09/Bg8208uqCdgWaA57m2+3P8Vv9rXJGnggeT1UDabJI5sLH6FrEE9aUpTuNZhgKZo2gw3YsGtotRlaataIY+oVSpB0mpUkW159CMeqWLRqAB+dfHOdKnUBnkSoPYfPar/D6MeqZ9MGvJjz3iHcB7UZCyn8IFdAEK3PaiYDIvCd2cfRykUM2s1kIQtbqK8RvsHUVpY7MZXQChQCW7fx0dd2nqWDt63d1DSrWWdW6itqpqefB0SiKYh/b/XJp9r83Acymx91Sfy8zqT0oci+Ah4P3onFQ6wdaGJlC+kH39MnT+bUKTfBuB30OTTh37WXDl3Xx871f52N4O+psGaU2DtCU0gbcEndMPUH2SzqxuObCxereK0SfyTpRSdCi7PdBgdS4qPLAHneTzTe23AF1EfL3CLP0+4PvAh9GMfvHguAMFS4+hFKbKUtpDxL1opr8PXRjyctIL0KD4blIqP6BuJCk9QsRW1KX6TehnPII2kOcD/l9HK2lN2dfy38cgSgMsbSR3pqZRPQ0wl6h+3nsP+f6tsY6jlb07iXi1xs32hylU4Ku2t6eTU+wXEcGtKMAczr5PQquPt0bwCPDDlKp+TzNrDI+hSbN3ofd7LzqvzMo+vyclNmT3fSn7AE5WVWtGkzVH0Dls8TAth/qYUXx+GECZDU3AXe41ZY3KgY3Vuw5Gr6Qsy/7NB7Ezs49+Cv1U+tE+nW60BJ83L1PpSlVWuwK4DVVig8I+h7kUApJe4F5UTWv8k3xKzxGxH1VIW130PC8AP2mQ/jW1SWkLsIWIfDVmaNTvKOKzKCXizej3ehytjr1etR/Q6duDguByZapzzehvUr4KXsRs9HerVnp5AA0wLqNo8FBRSn1EPI+KYlTaZ5Ofh9eP+3wnD5UbUFPZbYx+fxxD74M70PtgTa3PaWb1KZug+GEEa9H1bAWaKFkDvJTSuKvVxftI16IJkPnofJFfO7uyr9+VEi9O4OGbnVUObKzelW6cXMrofjcL0SrANAoDvKNolSTfzN+LVhc2ZpvMfxHtAzlIYd/OXHRy34DSqfImlNNRRbTdNc3Qp7QNuDtr2NkG9E94SeN6on1HY1dKtDr2BpX7A020dajJZFA59Wsh8Bwple4Nys3N/h0viB1Er8PxAxt5GAXSc9BKY7EWNEi5n5TGKzsOQATNaGVpN+XTJ0fQ6/qOCJ5OybnyZueClDgAPHSKDzuI9jVOy/4dAh5HVc8uQam5I2hS6Isp8eMJO2CzKeDAxurd64xukFi6Z6ETDTSPM3oze97DJE/TWZnd/n50Mi8dcB9Eg85LUWA0B60O5d9riIjHgDU1leRVlbdjtfyANgFS6ibiKTQTuZWxwcks9Dp6rNqz1PjdguppZaXHtp+IvwV+AQUxJ9Drsi379z400KjVSjS7Wi2tMa+GdDGnsBJkZueWlBiM4HE08ZOXpB9BEyP5KnK+T+fhs36AZhPMgY3VN6XyPAK8Gw1Ye9BqTB5cNKH9KxuKHjWN0X1XNIuvVKM3U7nfSMoe81vADxhdha01O4alRNwzwRW9bGJ8D/0Nb0QBQz8KiDtRcPv3pHSgyuO7s8ePV7a7lfF7Fo2m0uL/BwUlq7Ln2AO8lgXBp2J6rd+VQnqJmZ2/Hkf7IZegYKZ0cnAR8NWUPBlnjc+BjTWCH6OB4K1oFWQGGni2o9Wag4ze9D8T7WvJT97zUK+RVVSvUtWK8pePo8FtsUG0X+dNKDh68gx+HpsMCja/S8TjwDUo9WwAlVDeNG4wmtJRIp5FxSkqbeTvQqmNm0/j+CYqPa/WJp5NjF9m2szOcVlltS+gjIVrSr58GPhSSi41b+cGBzZW/zQg/AERz6BBZydqEvkSKgpwY9G9Z6LN4fkG8CY0G7UfNTC7GKWbbUSD1+KB3+Ls/kNUfm/sBW4j4unT6GBvZ0NK+4EfneajH0IB8BK0olL8N56DApu/n+IVu20osK+2spTvNdpylo7JzOpYVmDgaxE8gCqJNqNJmq2unmjnkqil2JNZXVE394+g1ZVh1OdmJUo96gaeyf7fhoKeFrTPoA0FRkfQqs9RtESfL7/fhDZSdqC9GPsrHMFy4C/PidLNNlbEDFRGuXhmM+8J8T1SqpTKeNZE8D60glkpJW4psD4lvnb2jsrMzGxqecXGGo+qk32diIdRatg8FNgsQalH89GMdd51eQ3aO9NGYR/NQRTc3Ag8QqHYQBNKRasWtOT3tXORqpN9nYgH0MpgE3CIlOqpYd1DKNXuCtSLKU/F7EArlFuB+6fm0MzMzKaGAxtrXCntQ0GJRExDVafaUHDyHpQ6lqcNDaBGm5eiMtJHUFA0D63O9GRfW0PlqlctFEpI27kspV7q9O+cVTq6C7gaVTtajgLuXuA7wLqUmIx+QWZmZnXLgY2dS06ggORatJJzKyoX/QaFFZhXUOCzkkKvmgvReyGhme5qFa8Wo2ab3pRtUyolTgBrI1iH9p0FcMz58mZmdr5yYGPnhohmtO/mLWjvTB/aZ7MIzWZvRPtsRlDn5a1ok/hS1OPk+2hvzpXAOyls0C62IHvef5nUn8XsFKREYnTTWjMzs/OSAxtrLBFNaHY6AccoVL+4AwU1eXPGdhTEHEEz2ZeiQGdT9vUD2cd2YC0p3Zc9/y60ivMOCuVym9F7ZQfw1SxFycz5dldPAAARPklEQVTMzMzqiAMbawwR7cD1wG2o5G4TsI+IR9FqzG0oSMkDneNoU/UsNJt9CG203sLolZg24PmTn6mE80NEPJndf0H2XBuBHbiMoJmZmVldcmBj9S9iOvArqOpZN2ooBgpwPpp9Po2xqWOvoZWXAbQK04IKBXRnX1+CGi2OLd+bUh9KTTMzMzOzBuCStdYIPoJK225F6WS5PlQY4ApUDa3UQeApVNZ5Dgps2lGvmpXAbuArXoUxMzMza3xesbH6FjEfVTir1hRxJ3A7ej2XdoTfiVZ0lqOGi50oGLoP2DDFHeTNzMzMbII4sLF6d1EN99mb/Tuv6P/FjqI9Mj3An5HS8Qk6NjMzMzOrEw5srH5FtKLyyzegwgEJ2IMqmx0ouucAWplZRvnAJrKv/dBBjZmZmdm5yYGN1aeITuCTqITzNBTQgFZllqLCAC8XPWIL2jNzEaqAlhcYmIv21DwDPDLZh21mZmZmU8OBjdWrn0NVy54HFqNCFyOoYMBR4HLUo2Y72jdzAPgc6ldzKyoOkFDVs3uBjVkpZzMzMzM7BzmwsfoTsRhVOtua3fIaSkk7gIKVBPRm99mDAp+7s0IArwCvZI08kyuemZll1A/scuBa1MOrG3gO9+gys3NE+FxmdSfincAdwI78FhTYXIKqnvVnty1GQc8GFOw0oZS0J4GtvlCbmWUiVqD03g40MTQMTEcBzsvAN0hpYOoO0MzszHnFxupRFzBY9HkCXgK2oX4187Pb8gCnA/WsSSj4uQ54hojvuJyzmZ33IhYC/wal7+4r+sqR7N/VwM8TcbcnhMyskTmwsXrUA7SWub0XBTigPTQXotWaQ0X32QfsB96M9uL806QdpZlZY7gdTQjNyT4/iNLQhrPPt6F+YUsprJSfFMEFqJDLFWhlfCvZynhKeO+imdUNBzZWj14B3lvl600oR3wPo4OaXEJFBd5OxGOk1Dfxh2hnJGIBSoMZAvaQ0vA4jzCz0xFxDfDbaIV7KLv1kuz/z1IokT+IyurvGP1w3gn8dPb1fGX8YtTweG0E30ppTGNkM7Mp4cDG6k9K+4l4AbgKBSilFqAyzg9WeZZhoBldwNdN+DFa7SJmUkgvnAW8B/UVGkaphH1EPAw87cp1ZhMo4lLgV1GvrwMlX50G3Az8BK10HwUWjn44b0aTTFsprO5AYWX8+uxx35+EozczO2UObKxefQdoR6kPPSgNrYlCH5vXgV3jPMcJNKC2qRCxDHgXqsI0AiwCVgEvAq8W3bMDuBO4gIjvOrgxmwCqDPkRFIRcVuYeQ6h8/rXAQ2g80F94OM3AT6H+YOVWVBNKYbs5gjUpndyvY2Y2ZRzYWH1K6TgRX0YpD7egYOYEKk36IPCBGp6lGThW67fMLuR5IHTEueNnIOIy4FNo4LQdBak3oM3Kq9Fscb5fqh9Vs7sJBawvY2ZnaiVqTrwNNSzuoChwyQyg1e+56Nz3QtHXlgEzs8cX60Dn47yIS54a/NjEHr6Z2alzYGP1S/suXs8+CiJaUc53K6Orp426V/bv5vG+TcTJQfftaN9HAD0RPAqsTani97ByIrqAX0Kbk/OB1NLs36Mo4HwHytHvQ4OuDdm/t+PAxmwizCn6/+vA24DjKBgptQC9X4vPtR1l7nsJKjJA9lz5Y389gqMpsfZMD9rM7Ew0TfUBmJ2ylAaBR4ElFAKYUsuAtaR0uNpTRdAJ/BpaATqGVhe2oYDpTuCXs8DHanc1JWktKLDpR+W6L85ua0ObkdvQas1qYCURHWfvUM3OWcUrzrtR0DIfBSy5JrRS0wZ8OTu35gYZfX69CL23e1F6cH/20YcKEHw04mTQY2Y2JbxiY41qDdpv81Y0038wu31mdvsG4Hs1PM+H0IbZrSW39wNvoHSOnwbuO/NDPm9czdhqdc0o3WUeGggBdGa35wOk+UW3mdmZ2cXowGQ9mry5FL3XhtGKzB7gf5NSaZnn7eh92Zrd9wYKaWiR3dZd8r3eH8GrTuM1s6niwMYaU0rDRHwLXaxvQbOJoIv03cCrpDRU6eEAEVwMfBBd7OehWc39jE6/2AW8NYIfpcTRif0hzln5QKhYD2qc2lNye/HAqw8FmT4vmZ2plLqJ2IhWr6ehAgLFxVT2ovPdy6S0YezDGYrgEbSaPQuttPahYCffW3Mhhb5jh9GK7HLGThSZmZ0VHkBY41L1rFeBV4loBoKUxu2nEME0dLH+ALow70OrBBehC/dTFDpyD6ML+ArUX8fGtxullRUXbuhFv+M8aAyUKlMcAM1As8QXgXP1zSbAd4HPoGqExX2/mrPblgH/UOXxj6Oqhr+Lirfk+2paUPra/ux5b0aV1UDvYzOzKeE9NnZuSGm4lqAm82HgRnRR7kUb2ntROts0tAJUus9j2gQd6fngGZRSViz/HU/PPu9AaSx5oDMju8+uMo81s9OzAvWvWY/ec3Oyj5loUugx4ENElD2/ZSllPdnjB9F7sxNNSmxF6bpH0YrN4uxhbtZpZlPGKzZ2XolgCcoV34r6qpQWHziKSqRexOjqXH1YrbahUs5XUkhJGUSBTDtwAQpoBtHvuhkNvp5BqWj+XZudqYhA1Qd3oPfUegqpaH2o1DNoH+GllKlGGEELql74HHqvHkDnzNJqacfQOXNT9v3MzKaEAxs731yHBtQJpaANoYF1cUpUL0rTeA3NRPbhnPHapZSI+Ab6nV6DfseD6HwzgFL6dqAB0iDK9T9E4e+wcQqO2uxckxfr2J59PkAhmCnWB1xF+TLrreh92otSzuZl/y91AhUk+FpKnpgws6njwMbON0sorAgMo0H2tSgNLa/kM4LSNKeji/VXUirbedsq6wKeRjO4C9Hm4yG0Afl5xhYXCLTp+BFScpEGszPXwtiVlRnonNaCVlm60XuxtcJzDKD3bQuwDqXpzkXpafl7uAWtfm8HHpi4wzczO3UObOx8c4LRe8s2o/0zV6CAJu+9MhNdrO9NiXVn9QgbWcRiVB77ckYHik8C3wZuA96JNiEfRAOv2SgQeoZ8A3JEJ/q79Jf01jCz2vShCYMmFLhcj85pI+h9l6+Q7keTDWOkxHAET6M+U7tQmf1VaIIiL8s+iFZgP5PSyeICZmZTwoGNnW9eQo03i8sOv4YuzMvQ6kIrqsr1JymN6cdilUQsAz6NZni3o8FTJ9pX806Ug/9FlPLyFhT8BAoun0DpfquIuA0NnBIwTMRTwJOkdOBs/jhmDS2lo0Q8j1akL0Pvw9L3UBN6H86s8kxPocCmCwVLr6I+YXmBlS50PnXVSDObcpFS6Uq12bkrgk7gP6A88WPl7oIG4PekVH4W08pQue1/h4KRXlQh6XK0GpPQ7/UEcC8plS8vG/EO4L1okJQ3XG1Gs8wJ+AIpbS/7WDMbK2Ih8BcoBXdPmXvke3D2oSad+8s/DSuBX0Ylng+iVZoO9P7uBv7Rk0BmVg9c7tnOKylxDDXwnI8u6sVV0TpRX5ungBfP+sE1tlVokNOLZodvRitfB1FhgDzt7NNE3DDm0RGXoKBmG4WgBpQqswvNFH+CiPbJ+xHMzjkHUXGO4+h8Nx0FJLOyz7ehvTPDqLBKWSmxFfhL4H60ItuJGnLeA/yNgxozqxdORbPzTkpsjODzwLvQ3pphFOD0APcCz2b9G6x2q9As7jxU5rm4GEPuCJrx/RQRb5BS8WDoNkZvSC51BJWlvQK858msRl2oAMAPUJrtUnTd70UrNXmFs170/qooq3b2RPZhZlaXHNjYeSkldgBfiji5cf0E0O2A5rTlG5EvRrPDlX6PeRB5LfAwkBcKuBjNHldzGM0qO7Axq02ea34CrXzuqnC/0n5eZmYNyalodl5LicMpsSMl9jioOSO7UYrLYqo32GxCg6vitJdpjC1LW06eAmNmtTmCVkLHe9/MQkUBzMwamgMbM5sIr1IoIVtJFwqA+lAwk+tHKznNaF9OF0pZKzUdbXI2s1qkNAI8igpwVNKK3rveV2hmDc+paGZ25lLqJ+I+4FYUlJR2OO9A55uXUeCyi4hZwDWo2MBy4A6UxjaIUmP2Aa+jPhugWednJ/cHMTvnrAXeBFwK7ERpabkuYAHwdVLqLfNYM7OG4nLPZjZxIj6NysIeQQOoQAHNUeBptE9mNeobdAUKgI4Db0XBzfHsa3spVHBah9LQtgD/L5uFNrNaRbSiYik3U1hVDdTX5p9IyT1ozOyc4MDGzCZOxHTgd1Cz02FURGB/9tGOBldzKFRmakYBzWYU1KwCZqCKTd3ZY+YD3wQ+T0rleg+ZWS1ULn0ZSgU9AuzyRIGZnUsc2JjZxIqYC3wcFRLoR6lls1D38h40UzyU3b4INQ8EeAMFO3Ozx76GVniOAWtI6Vtn74cwMzOzRuPAxswmXkQTanZ6DUopuwytxOxD+3DyJpxvQqs6Ce3NeSn7fB7wJCo20IwCnf+GT1hmZmZWgauimdnES2mElDaT0reBfKVlA2OrnTWhQGYEreTMym4fQftroFAxzecrMzMzq8gDBTObbHPRuSbfc1PsOIXqjMNodYfs/nn1pg6gh5SGJ/k4zczMrIE5sDGzsyFPITuIgpv83LMP9dEovXPazeKB17hs7j7mXThM05qzdJxmZmbWoNzHxswm2wEKzTsHUdnmVSjIOYKKA7SjYKfvELOWbeDStse55W0L6O4YoH3kP/E/d+4L+lLipan5EczMzKzeecXGzCZXSv3AU8AF2S2voGBnPgp2NqOAp+MAs5dt5qILdnNB75Wsp5mR7r/gP96/j4UjwCcieOcU/ARmZmbWAFwVzcwmX0QX8Buo0/leNKmyAnVDnwF0HqZrx/f44OIhWncfZs6RH3HH9h/y7t5+Ohdk90mo/8afp8S2qflBzMzMrF45sDGzsyNiJvBB4EoKe26aUCraD2bQu6yPGdehEs8AF6Ny0HkhgSYU4GwCfj8lus/i0ZuZmVmdc2BjZmeXGnguRYFKD7CNlEYi+APUoHMQuAS4mkKxgZOPRj1tfgz8TUon++GYmZnZec7FA8zs7ErpIJQNSFpQyed2tFJTGtSAVnqOo5S024FvT96BmpmZWSNx8QAzqxe7UarZErQyUxrUgIKeXrRP54aIk31vzMzM7DznwMbM6sVPgDnZx0CF+0wHNqKVHYBZZ+G4zMzMrAE4sDGzevE6KgwwB63YlJoFHKJQXCAoFCEwMzOz85z32JhZXUiJExHcjfrbfBCdn05k/zaj9LPn0GpNK9prs39qjtbMzMzqjQMbM6sbKXEsgs+g1ZlZaFWmH63S9BTddQnwYEoMnf2jNDMzs3rkcs9mVnciuBj4VaCP0RXUWlBQswX4UkoV9+KYmZnZecaBjZnVpQiWA+8FVqIKaYFS054AHnZQY2ZmZsUc2JhZXYtgAUpLGwZ2p8TxKT4kMzMzq0MObMzMzMzMrOG53LOZmZmZmTU8BzZmZmZmZtbwHNiYmZmZmVnDc2BjZmZmZmYNz4GNmZmZmZk1PAc2ZmZmZmbW8BzYmJmZmZlZw3NgY2ZmZmZmDc+BjZmZmZmZNTwHNmZmZmZm1vAc2JiZmZmZWcNzYGNmZmZmZg3PgY2ZmZmZmTU8BzZmZmZmZtbwHNiYmZmZmVnDc2BjZmZmZmYNz4GNmZmZmZk1PAc2ZmZmZmbW8BzYmJmZmZlZw3NgY2ZmZmZmDc+BjZmZmZmZNTwHNmZmZmZm1vAc2JiZmZmZWcNzYGNmZmZmZg3PgY2ZmZmZmTU8BzZmZmZmZtbwHNiYmZmZmVnDc2BjZmZmZmYNz4GNmZmZmZk1PAc2ZmZmZmbW8BzYmJmZmZlZw3NgY2ZmZmZmDc+BjZmZmZmZNTwHNmZmZmZm1vAc2JiZmZmZWcNzYGNmZmZmZg3PgY2ZmZmZmTU8BzZmZmZmZtbw/j/4s/5u3GYrgwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x111be18d0>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "viz_umap_embed(count_matrix, \n",
    "               title=\"Link Sharing of Members of Congress Embedded by UMAP\",\n",
    "               threshold=5,\n",
    "               # umap params\n",
    "               n_neighbors=50,\n",
    "               min_dist=0.1,\n",
    "               metric='dice',\n",
    "               random_state=303)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Supervised Learning\n",
    "In the unsupervised case, we already see Democrats and Republicans sectioned off. Here we will fit a [logistic regression](http://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) model to predict whether a congress member is a Democrat or a Republican based on the `count_matrix` we just created."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(820, 145)"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# filter out independents, and accounts that sent less than one link!\n",
    "dems_repubs = look_up[look_up.affiliation != 'Independent'].twitter_id\n",
    "count_matrix_ = count_matrix[(count_matrix.index.isin(dems_repubs)) &\n",
    "                             (count_matrix.sum(axis=1) >= 1)]\n",
    "parties = look_up.set_index('twitter_id').loc[count_matrix_.index].affiliation\n",
    "\n",
    "# create the training set\n",
    "X, y = count_matrix_.values, parties\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=303, test_size=.15)\n",
    "len(X_train), len(X_test) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9517241379310345"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "logreg = LogisticRegression(penalty='l2', C=.7,\n",
    "                            solver='liblinear',\n",
    "                            random_state=303)\n",
    "logreg.fit(X_train, y_train)\n",
    "logreg.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from sklearn.metrics import confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_confusion_matrix(cm, classes,\n",
    "                          normalize=False,\n",
    "                          title='Confusion matrix',\n",
    "                          cmap=plt.cm.Blues):\n",
    "    \"\"\"\n",
    "    This function prints and plots the confusion matrix.\n",
    "    Normalization can be applied by setting `normalize=True`.\n",
    "    \"\"\"\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        print(\"Normalized confusion matrix\")\n",
    "    else:\n",
    "        print('Confusion matrix, without normalization')\n",
    "\n",
    "    print(cm)\n",
    "\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=45)\n",
    "    plt.yticks(tick_marks, classes)\n",
    "\n",
    "    fmt = '.2f' if normalize else 'd'\n",
    "    thresh = cm.max() / 2.\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, format(cm[i, j], fmt),\n",
    "                 horizontalalignment=\"center\",\n",
    "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label')\n",
    "    plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix, without normalization\n",
      "[[49  2]\n",
      " [ 5 89]]\n",
      "Normalized confusion matrix\n",
      "[[0.96 0.04]\n",
      " [0.05 0.95]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x15b7235f8>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x15bc95470>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pred = logreg.predict(X_test)\n",
    "class_names = logreg.classes_\n",
    "\n",
    "# Compute confusion matrix\n",
    "cnf_matrix = confusion_matrix(y_test, y_pred)\n",
    "np.set_printoptions(precision=2)\n",
    "\n",
    "# Plot non-normalized confusion matrix\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix, classes=class_names,\n",
    "                      title='Confusion matrix, without normalization')\n",
    "\n",
    "# Plot normalized confusion matrix\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix, classes=class_names, normalize=True,\n",
    "                      title='Normalized confusion matrix')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "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>twitter_name</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>twitter_id</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9.410800851211756e+17</th>\n",
       "      <td>SenDougJones</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23820360.0</th>\n",
       "      <td>billhuizenga</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>136526394.0</th>\n",
       "      <td>WebsterCongress</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4615689368.0</th>\n",
       "      <td>GeneGreen29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19726613.0</th>\n",
       "      <td>SenatorCollins</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>242376736.0</th>\n",
       "      <td>RepCharlieDent</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16056306.0</th>\n",
       "      <td>JeffFlake</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          twitter_name\n",
       "twitter_id                            \n",
       "9.410800851211756e+17     SenDougJones\n",
       "23820360.0                billhuizenga\n",
       "136526394.0            WebsterCongress\n",
       "4615689368.0               GeneGreen29\n",
       "19726613.0              SenatorCollins\n",
       "242376736.0             RepCharlieDent\n",
       "16056306.0                   JeffFlake"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# these are what we got wrong!\n",
    "df_meta.set_index('twitter_id').loc[\n",
    "    y_test[y_test != y_pred].index\n",
    "][['twitter_name']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "indep = look_up[look_up.affiliation == 'Independent'].twitter_id\n",
    "count_matrix_indep_ = count_matrix[(count_matrix.index.isin(indep)) &\n",
    "                                   (count_matrix.sum(axis=1) >= 2)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "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>twitter_name</th>\n",
       "      <th>preds</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>twitter_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1068481578.0</th>\n",
       "      <td>SenAngusKing</td>\n",
       "      <td>Republican</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>216776631.0</th>\n",
       "      <td>BernieSanders</td>\n",
       "      <td>Democrat</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2915095729.0</th>\n",
       "      <td>AkGovBillWalker</td>\n",
       "      <td>Republican</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29442313.0</th>\n",
       "      <td>SenSanders</td>\n",
       "      <td>Democrat</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3196634042.0</th>\n",
       "      <td>GovernorMapp</td>\n",
       "      <td>Republican</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 twitter_name       preds\n",
       "twitter_id                               \n",
       "1068481578.0     SenAngusKing  Republican\n",
       "216776631.0     BernieSanders    Democrat\n",
       "2915095729.0  AkGovBillWalker  Republican\n",
       "29442313.0         SenSanders    Democrat\n",
       "3196634042.0     GovernorMapp  Republican"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_ind = df_meta.set_index('twitter_id').loc[count_matrix_indep_.index][['twitter_name']]\n",
    "df_ind['preds'] = logreg.predict(count_matrix_indep_)\n",
    "df_ind"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## K-Fold Cross Validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "logreg_cv = LogisticRegression(penalty='l2', C=.7,\n",
    "                               solver='liblinear',\n",
    "                               random_state=303)\n",
    "scores = cross_val_score(logreg_cv, X, y, cv=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.96 (+/- 0.02)\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy: %0.2f (+/- %0.2f)\" % (scores.mean(), scores.std() * 2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}