{
 "metadata": {
  "name": "",
  "signature": "sha256:a18be62d14ce934a49f181450958d112885d7f28b7da332afbb367d49f880989"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "# Wikipedia edit stream demo\n",
      "\n",
      "A demo of [Snake Charmer](https://github.com/snake-charmer-devs/snake-charmer), originally presented at a [PyData London meetup](http://www.meetup.com/PyData-London-Meetup/) in August 2014, showing:\n",
      "\n",
      "* Reading data from a [WebSockets](https://www.websocket.org/) stream in real time, in a background thread\n",
      "* Pulling this data into [Pandas](http://pandas.pydata.org/) for analysis and visualization\n",
      "* Training a [Vowpal Wabbit](https://github.com/JohnLangford/vowpal_wabbit) language classifier on this stream, in real time\n",
      "* Visualizing the classifier's behaviour using [Matplotlib](http://matplotlib.org/) and [D3](http://d3js.org/)\n",
      "\n",
      "The demo is designed to be run using Snake Charmer [release a18945d](https://github.com/snake-charmer-devs/snake-charmer/tree/b86184ecd92cd785b443163f6ecef4366548f591) or later.\n",
      "\n",
      "Wikipedia streaming is powered by the `wikipedia_updates` function, which is based on the [stream.py](https://github.com/edsu/wikistream/blob/master/stream.py) script provided with [edsu/wikistream](https://github.com/edsu/wikistream)."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import re\n",
      "import json\n",
      "import time\n",
      "import threading\n",
      "\n",
      "from collections import defaultdict\n",
      "from itertools import chain\n",
      "from requests import post, get\n",
      "from math import sqrt\n",
      "\n",
      "from wabbit_wappa import VW\n",
      "\n",
      "import numpy as np\n",
      "\n",
      "import pandas as pd\n",
      "pd.options.display.mpl_style = 'default'\n",
      "\n",
      "import seaborn as sb\n",
      "import matplotlib.pylab as pylab\n",
      "import matplotlib.pyplot as plt\n",
      "pylab.rcParams['figure.figsize'] = (12.0, 8.0)\n",
      "pylab.rcParams['font.family'] = 'Bitstream Vera Sans'"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 36
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## wikipedia_updates\n",
      "\n",
      "This function talks to the `wikistream` server, which provides push notifications of Wikipedia edits via [Socket.io](http://socket.io/).\n",
      "\n",
      "Call it with a callback function of your own devising, that takes a single argument. Every time a Wikipedia edit occurs, your callback will be called, with a dict containing the details of the edit, e.g.:\n",
      "\n",
      "```\n",
      "{'flag': '',\n",
      " 'namespace': 'article',\n",
      " 'userUrl': 'http://en.wikipedia.org/wiki/User:137.44.83.167',\n",
      " 'url': 'http://en.wikipedia.org/w/index.php?diff=619400522&oldid=618259633',\n",
      " 'wikipediaLong': 'English Wikipedia',\n",
      " 'wikipediaShort': 'en',\n",
      " 'user': '137.44.83.167',\n",
      " 'comment': '',\n",
      " 'newPage': False,\n",
      " 'pageUrl': 'http://en.wikipedia.org/wiki/Battle_of_Trafalgar',\n",
      " 'unpatrolled': False,\n",
      " 'robot': False,\n",
      " 'anonymous': True,\n",
      " 'delta': 2,\n",
      " 'channel': '#en.wikipedia',\n",
      " 'wikipediaUrl': 'http://en.wikipedia.org',\n",
      " 'wikipedia': 'English Wikipedia',\n",
      " 'page': 'Battle of Trafalgar'}\n",
      "```\n",
      "\n",
      "Your callback should return `True` if it wants to keep receiving edits, or `False` to stop."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def wikipedia_updates(callback):\n",
      "    endpoint = \"http://wikistream.inkdroid.org/socket.io/1\"\n",
      "    session_id = post(endpoint).content.decode('utf-8').split(':')[0]\n",
      "    xhr_endpoint = \"/\".join((endpoint, \"xhr-polling\", session_id))\n",
      "\n",
      "    while True:\n",
      "        t = time.time() * 1000000\n",
      "        response = get(xhr_endpoint, params={'t': t}).content.decode('utf-8')\n",
      "\n",
      "        chunks = re.split(u'\\ufffd[0-9]+\\ufffd', response)\n",
      "        for chunk in chunks:\n",
      "            parts = chunk.split(':', 3)\n",
      "            if len(parts) == 4:\n",
      "                try:\n",
      "                    payload = json.loads(parts[3])['args'][0]\n",
      "                except:\n",
      "                    raise ValueError('Received non-json data: ' + chunk)\n",
      "                if not callback(payload):\n",
      "                    return"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 37
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Watching stream activity in real time\n",
      "\n",
      "This is a callback function that collects the edits into a list called `edits`, for offline analysis later. It runs until `max_edits` edits have been received.\n",
      "\n",
      "A `timestamp` field is added to each edit, as they don't have this by default.\n",
      "\n",
      "Run this cell to start gathering the data in the background.\n",
      "\n",
      "### Note\n",
      "\n",
      "If at any point you get an error message about non-json data, wait a few seconds, then restart the IPython kernel and retry from the first cell. This seems to be an intermittent problem with the server -- or possibly the requests library?"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "max_edits = 500\n",
      "edits = []\n",
      "\n",
      "def simple_callback(edit):\n",
      "    edit['timestamp'] = time.time()\n",
      "    edits.append(edit)\n",
      "    return len(edits) < max_edits\n",
      "\n",
      "thr = threading.Thread(target=wikipedia_updates, args=(simple_callback,))\n",
      "thr.start()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 38
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "We can watch the data as it arrives. Run the following cell repeatedly, until it tells you it has finished."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "reg = re.compile(r'^.*wiki/(.*)')\n",
      "\n",
      "c = len(edits)\n",
      "n = min(5, c)\n",
      "print('Last %d of %d edits...\\n' % (n, c))\n",
      "for edit in edits[-n:]:\n",
      "    wiki_name = edit['wikipediaLong']\n",
      "    page_name = reg.match(edit['pageUrl']).group(1)\n",
      "    comment = edit['comment']\n",
      "    print('\\t'.join((wiki_name, page_name, comment)), '\\n', flush=True)\n",
      "if c == max_edits:\n",
      "    print('Done.\\n')\n",
      "else:\n",
      "    print('Still streaming.\\n')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Last 5 of 500 edits...\n",
        "\n",
        "Wikidata\tQ1022537\t/* wbsetdescription-add:1|nn */ Auto-description for Norwegian (Nynorsk) \n",
        "\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Wikidata\tQ1029016\t/* wbsetdescription-add:1|nb */ Auto-description for Norwegian (Bokm\u00e5l) \n",
        "\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Wikidata\tQ17744235\t/* wbeditentity-create:0| */ \n",
        "\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "English Wikipedia\tWikipedia:Tutorial/Editing/sandbox\t \n",
        "\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Wikidata\tQ2068332\t/* wbeditentity-update:0| */ Added: [[eu:Rhynchobatus luebberti]] \n",
        "\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Done.\n",
        "\n"
       ]
      }
     ],
     "prompt_number": 43
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Import into pandas\n",
      "\n",
      "Let's add all these edits into a `DataFrame` to make them easier to work with."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df = pd.DataFrame(edits)\n",
      "df['timestamp'] = pd.to_datetime(df['timestamp'], unit='s')\n",
      "df.head()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>anonymous</th>\n",
        "      <th>channel</th>\n",
        "      <th>comment</th>\n",
        "      <th>delta</th>\n",
        "      <th>flag</th>\n",
        "      <th>namespace</th>\n",
        "      <th>newPage</th>\n",
        "      <th>page</th>\n",
        "      <th>pageUrl</th>\n",
        "      <th>robot</th>\n",
        "      <th>timestamp</th>\n",
        "      <th>unpatrolled</th>\n",
        "      <th>url</th>\n",
        "      <th>user</th>\n",
        "      <th>userUrl</th>\n",
        "      <th>wikipedia</th>\n",
        "      <th>wikipediaLong</th>\n",
        "      <th>wikipediaShort</th>\n",
        "      <th>wikipediaUrl</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td> False</td>\n",
        "      <td>       #en.wikipedia</td>\n",
        "      <td>                                         /* top */</td>\n",
        "      <td>  -5</td>\n",
        "      <td> M</td>\n",
        "      <td> article</td>\n",
        "      <td> False</td>\n",
        "      <td> Android version history</td>\n",
        "      <td> http://en.wikipedia.org/wiki/Android_version_h...</td>\n",
        "      <td> False</td>\n",
        "      <td>2014-08-31 18:32:09.978374</td>\n",
        "      <td> False</td>\n",
        "      <td> http://en.wikipedia.org/w/index.php?diff=62360...</td>\n",
        "      <td>  JordanKyser22</td>\n",
        "      <td> http://en.wikipedia.org/wiki/User:JordanKyser22</td>\n",
        "      <td> English Wikipedia</td>\n",
        "      <td> English Wikipedia</td>\n",
        "      <td> en</td>\n",
        "      <td> http://en.wikipedia.org</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td> False</td>\n",
        "      <td>       #ru.wikipedia</td>\n",
        "      <td>                                  /* \u041f\u0440\u043e\u0442\u0438\u0432\u043d\u0438\u043a\u0438 */</td>\n",
        "      <td>  12</td>\n",
        "      <td>  </td>\n",
        "      <td> article</td>\n",
        "      <td> False</td>\n",
        "      <td>  \u0427\u0451\u0440\u043d\u043e-\u0436\u0451\u043b\u0442\u043e-\u0431\u0435\u043b\u044b\u0439 \u0444\u043b\u0430\u0433</td>\n",
        "      <td> http://ru.wikipedia.org/wiki/\u0427\u0451\u0440\u043d\u043e-\u0436\u0451\u043b\u0442\u043e-\u0431\u0435\u043b\u044b\u0439...</td>\n",
        "      <td> False</td>\n",
        "      <td>2014-08-31 18:32:09.978403</td>\n",
        "      <td> False</td>\n",
        "      <td> http://ru.wikipedia.org/w/index.php?diff=65199...</td>\n",
        "      <td>     \u041a\u0430\u043c\u0430\u0440\u0430\u0434 \u0427\u0435</td>\n",
        "      <td>    http://ru.wikipedia.org/wiki/User:\u041a\u0430\u043c\u0430\u0440\u0430\u0434 \u0427\u0435</td>\n",
        "      <td> Russian Wikipedia</td>\n",
        "      <td> Russian Wikipedia</td>\n",
        "      <td> ru</td>\n",
        "      <td> http://ru.wikipedia.org</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td> False</td>\n",
        "      <td> #wikidata.wikipedia</td>\n",
        "      <td> /* wbcreateclaim-create:1| */ [[Property:P1459...</td>\n",
        "      <td> 346</td>\n",
        "      <td> B</td>\n",
        "      <td> article</td>\n",
        "      <td> False</td>\n",
        "      <td>               Q17744225</td>\n",
        "      <td>                http://wikidata.org/wiki/Q17744225</td>\n",
        "      <td>  True</td>\n",
        "      <td>2014-08-31 18:32:09.978426</td>\n",
        "      <td> False</td>\n",
        "      <td> http://www.wikidata.org/w/index.php?diff=15466...</td>\n",
        "      <td> Reinheitsgebot</td>\n",
        "      <td>    http://wikidata.org/wiki/User:Reinheitsgebot</td>\n",
        "      <td>          Wikidata</td>\n",
        "      <td>          Wikidata</td>\n",
        "      <td> wd</td>\n",
        "      <td>     http://wikidata.org</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td> False</td>\n",
        "      <td>       #nl.wikipedia</td>\n",
        "      <td>                                     Linkonderhoud</td>\n",
        "      <td>-232</td>\n",
        "      <td>  </td>\n",
        "      <td> article</td>\n",
        "      <td> False</td>\n",
        "      <td>                 Beersel</td>\n",
        "      <td>              http://nl.wikipedia.org/wiki/Beersel</td>\n",
        "      <td> False</td>\n",
        "      <td>2014-08-31 18:32:09.978446</td>\n",
        "      <td> False</td>\n",
        "      <td> http://nl.wikipedia.org/w/index.php?diff=41990...</td>\n",
        "      <td>     Smile4ever</td>\n",
        "      <td>    http://nl.wikipedia.org/wiki/User:Smile4ever</td>\n",
        "      <td>   Dutch Wikipedia</td>\n",
        "      <td>   Dutch Wikipedia</td>\n",
        "      <td> nl</td>\n",
        "      <td> http://nl.wikipedia.org</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td> False</td>\n",
        "      <td> #wikidata.wikipedia</td>\n",
        "      <td> /* wbcreateclaim-create:1| */ [[Property:P1412...</td>\n",
        "      <td> 405</td>\n",
        "      <td>  </td>\n",
        "      <td> article</td>\n",
        "      <td> False</td>\n",
        "      <td>                Q3035478</td>\n",
        "      <td>                 http://wikidata.org/wiki/Q3035478</td>\n",
        "      <td> False</td>\n",
        "      <td>2014-08-31 18:32:09.978467</td>\n",
        "      <td> False</td>\n",
        "      <td> http://www.wikidata.org/w/index.php?diff=15466...</td>\n",
        "      <td>          Jura1</td>\n",
        "      <td>             http://wikidata.org/wiki/User:Jura1</td>\n",
        "      <td>          Wikidata</td>\n",
        "      <td>          Wikidata</td>\n",
        "      <td> wd</td>\n",
        "      <td>     http://wikidata.org</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 44,
       "text": [
        "  anonymous              channel  \\\n",
        "0     False        #en.wikipedia   \n",
        "1     False        #ru.wikipedia   \n",
        "2     False  #wikidata.wikipedia   \n",
        "3     False        #nl.wikipedia   \n",
        "4     False  #wikidata.wikipedia   \n",
        "\n",
        "                                             comment  delta flag namespace  \\\n",
        "0                                          /* top */     -5    M   article   \n",
        "1                                   /* \u041f\u0440\u043e\u0442\u0438\u0432\u043d\u0438\u043a\u0438 */     12        article   \n",
        "2  /* wbcreateclaim-create:1| */ [[Property:P1459...    346    B   article   \n",
        "3                                      Linkonderhoud   -232        article   \n",
        "4  /* wbcreateclaim-create:1| */ [[Property:P1412...    405        article   \n",
        "\n",
        "  newPage                     page  \\\n",
        "0   False  Android version history   \n",
        "1   False   \u0427\u0451\u0440\u043d\u043e-\u0436\u0451\u043b\u0442\u043e-\u0431\u0435\u043b\u044b\u0439 \u0444\u043b\u0430\u0433   \n",
        "2   False                Q17744225   \n",
        "3   False                  Beersel   \n",
        "4   False                 Q3035478   \n",
        "\n",
        "                                             pageUrl  robot  \\\n",
        "0  http://en.wikipedia.org/wiki/Android_version_h...  False   \n",
        "1  http://ru.wikipedia.org/wiki/\u0427\u0451\u0440\u043d\u043e-\u0436\u0451\u043b\u0442\u043e-\u0431\u0435\u043b\u044b\u0439...  False   \n",
        "2                 http://wikidata.org/wiki/Q17744225   True   \n",
        "3               http://nl.wikipedia.org/wiki/Beersel  False   \n",
        "4                  http://wikidata.org/wiki/Q3035478  False   \n",
        "\n",
        "                   timestamp unpatrolled  \\\n",
        "0 2014-08-31 18:32:09.978374       False   \n",
        "1 2014-08-31 18:32:09.978403       False   \n",
        "2 2014-08-31 18:32:09.978426       False   \n",
        "3 2014-08-31 18:32:09.978446       False   \n",
        "4 2014-08-31 18:32:09.978467       False   \n",
        "\n",
        "                                                 url            user  \\\n",
        "0  http://en.wikipedia.org/w/index.php?diff=62360...   JordanKyser22   \n",
        "1  http://ru.wikipedia.org/w/index.php?diff=65199...      \u041a\u0430\u043c\u0430\u0440\u0430\u0434 \u0427\u0435   \n",
        "2  http://www.wikidata.org/w/index.php?diff=15466...  Reinheitsgebot   \n",
        "3  http://nl.wikipedia.org/w/index.php?diff=41990...      Smile4ever   \n",
        "4  http://www.wikidata.org/w/index.php?diff=15466...           Jura1   \n",
        "\n",
        "                                           userUrl          wikipedia  \\\n",
        "0  http://en.wikipedia.org/wiki/User:JordanKyser22  English Wikipedia   \n",
        "1     http://ru.wikipedia.org/wiki/User:\u041a\u0430\u043c\u0430\u0440\u0430\u0434 \u0427\u0435  Russian Wikipedia   \n",
        "2     http://wikidata.org/wiki/User:Reinheitsgebot           Wikidata   \n",
        "3     http://nl.wikipedia.org/wiki/User:Smile4ever    Dutch Wikipedia   \n",
        "4              http://wikidata.org/wiki/User:Jura1           Wikidata   \n",
        "\n",
        "       wikipediaLong wikipediaShort             wikipediaUrl  \n",
        "0  English Wikipedia             en  http://en.wikipedia.org  \n",
        "1  Russian Wikipedia             ru  http://ru.wikipedia.org  \n",
        "2           Wikidata             wd      http://wikidata.org  \n",
        "3    Dutch Wikipedia             nl  http://nl.wikipedia.org  \n",
        "4           Wikidata             wd      http://wikidata.org  "
       ]
      }
     ],
     "prompt_number": 44
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Visualizing the total number of edits in each wiki\n",
      "\n",
      "This is a one-liner in pandas!"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "p = df.wikipedia.value_counts().plot(kind='barh')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAyYAAAHdCAYAAADo/W3BAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3XtclHX+///HIIwQykEUEcUAYQbUpNRSS9cS07LDttt6\n2IOtqdumZvVRQSotUxMQzbUNbdPUcstvurdOupqbpBmrVkqaooAIaqgjUKAgx2Hm94c/J4nxjA21\nz/vt5u3mNdf1fr9f13W99DaveV/vGYPdbrcjIiIiIiLiQm6uDkBERERERESFiYiIiIiIuJwKExER\nERERcTkVJiIiIiIi4nIqTERERERExOVUmIiIiIiIiMu5X2mDkSNHsnLlSsf2li1byMvLY/To0Y0a\n2OX6/vvvWbFiBZMmTbqmfg4fPsyiRYuYO3cuAOnp6bz22mu89dZbuLm5cfToUf7+97+TkpLC9OnT\nmTVrFpmZmaxdu5aEhIR6fe3cuZOCggIeeuiha4oJIDU1lR49etC7d29ee+017r//fjp06HBFfaSl\npV1zHCIiIiIilxIbG3vVba+4MDEYDFc9WGOrq6ujVatW11yUAHTs2JHi4mKqqqrw9PQkJyeHDh06\nkJeXR0REBNnZ2ZjNZgBmzZp10b569uxJz549rzkmOHu9z13zxx9//Kr7ScgwkDIkgpjglo0Sl/y8\n+fv7U1JS4uowpIlRXogzygtxRnkhzmRkZFxT+ysuTC7m/E/34YfZlczMTNasWYOPjw/ffvstYWFh\nPPnkk8DZE1i5ciXNmzfHZDJRWFhIQkICubm5rFixgtraWoxGI+PGjSM4OJgtW7bwxRdfUF1djd1u\nZ/z48SQlJTF//nwKCwtJTU2lqqoKgDFjxmAymS46/jlubm506tSJgwcPctNNN5Gfn8/gwYPJyclx\nFCYxMTH1zut8ubm5LFmyhEmTJnHgwAHHLFJqaioeHh7k5+dTUVHBn//8Z7p3747NZuPtt99m//79\nWK1WBg8ezMCBA7Hb7Sxbtoy9e/cSEBCAu/sPt2jGjBk88sgjhIeHs3TpUg4dOkRNTQ29evVi2LBh\njXkrRURERER+UldcmNTU1BAfH+/YLi8vd8wO/Hg25fztw4cP8/LLL+Pv78/06dPJzs4mLCyMJUuW\nMHPmTNq0acPChQsdbdq3b8/MmTNxc3Pjm2++YdWqVUyePNnR17x58/D29qawsNAxhp+fH9OmTcPD\nw4MTJ07wyiuvkJiY6HT8rKwsoqKi6sVrNpvJzs7GZDJhMBjo3Lkz77zzDkOGDOHgwYOON/8/Ps/s\n7GyWL19OfHw8AQEBHDhwoN7+7777jsTERCwWCy+++CKvvPIKn332Gd7e3iQmJlJbW8vzzz9Pt27d\nyM/P58SJEyxYsIDS0lImTZrEgAEDGow7YsQIWrRogc1mY9asWRw9epSOHTte7m0UEREREWlSrrgw\nMRqNjnUY8MMak0uJiIigVatWAISGhlJYWEjz5s1p27Ytbdq0AeCOO+5g06ZNAJw5c4ZXX30Vi8WC\nwWCgrq7O0Ve3bt3w9vZuMIbVauWNN97gyJEjuLm5ceLEiQuOX1RU1KAwMZlMrFu3jtzcXCIiImjb\nti0Wi4XTp09TVVVFYGBggzGPHTvG66+/zvTp0/Hz82uw32Aw0KdPHwCCgoIIDAzk2LFj7Nmzh6NH\nj7Jjxw4AKisrsVgsHDhwgL59+2IwGPD396dLly5Or+e2bdtIS0vDZrNRUlJCQUGBChMRERER+dlq\n1Ee53NzcsNvtANhsNqxW6w8DnfdIkpubW71Cw5l3332Xm266ibi4OIqKipgxY4ZjX/PmzZ22Wbdu\nHf7+/kycOBGbzcYf//jHKxo/MjKSQ4cOOWZNAAICAti2bRuRkZFOx/T396e2tpa8vDy6d+9+0XM6\n59zMx5gxY+jWrVu9fRkZGY5reCGFhYWsW7eOpKQkbrjhBhYtWkRNTc0lx3X3cMff3/+yYpRfPuWC\nOKO8EGeUF+KM8kIaW6MWJoGBgeTl5dGnTx927tx50eLDYDAQHBzMyZMnKSoqok2bNmzbts3xpr2y\nstKR8Js3b76s8SsrKwkICADgs88+w2azXVH8Xl5etGrVis2bN/Piiy8CZ2dR/v3vfzN48GCnbW64\n4QbGjRvH7Nmz8fT0pHPnzvX22+12tm/fTv/+/Tl58iSFhYW0b9+emJgYNm7cSJcuXWjWrBnHjx8n\nICCAzp0788knn9C/f39OnTpFZmYm/fr1q9dnRUUFzZs3x8vLi9LSUr7++usLzqycz1pr1UI1AbRo\nUZxTXogzygtxRnkh10OjfitXbGwsc+fOJS4ujptvvhlPT8+LtjMajYwdO5Y5c+bQvHlzOnXq5Nj3\n4IMPkpqaynvvvcctt9xy0XHP7Rs8eDDz58/ns88+u+T4F+ovKiqKnTt3Oh77MplMrFq1yvGNXD9u\nazAY8PX1ZerUqSQmJjJu3LgG47Ru3Zpnn32WiooK/vKXv+Du7k5sbCxFRUVMnToVu92Or68vcXFx\n3Hbbbezbt49JkybRunXreuOeExoaSlhYGE8//TStW7du8EiaiIiIiMjPjcF+qeeGrrNzX88LsHTp\nUoKDgxkyZIgrQ2pUixYtokePHvTq1ctlMaSlpenrgqUefdIlzigvxBnlhTijvBBnMjIyftrfMWls\naWlpfPbZZ1itVsLCwhg4cKCrQxIRERERkZ+YywuT++67j/vuu8/VYVw348ePd3UIAKQMiaBtS6Or\nwxARERERccrlhYn8NPQIl4iIiIg0ZW6uDkBERERERESFiYiIiIiIuJwKExERERERcTkVJiIiIiIi\n4nIqTERERERExOVUmIiIiIiIiMupMBEREREREZe7osJk+PDhvPXWW47tjz76iDVr1jR6UFfqtdde\no6Cg4Jr7iY+P5/DhwwDU1dUxcuRIPv/8c8f+qVOnkp+fz+rVq9m3bx8AEyZMoLy8vEFf06dPv+Z4\nADIzM0lKSgJg586dfPDBB1fVz57jZVjKqhslJhERERGRxnZFhYm7uztfffUVZWVlABgMhisazGaz\nXdHxl+vxxx+nQ4cO19xPVFQUOTk5ABw5coTg4GCys7MBqKqqorCwkNDQUIYNG0bXrl0v2tesWbOu\nOZ4f69mzJw899NBVtY1bn8vJsppGjkhEREREpHFc0S+/N2vWjNjYWP79738zYsSIevsKCwtZvHgx\n5eXl+Pj4MG7cOFq3bk1qaioeHh4cOXIEs9nMN998w8yZM/Hy8mLMmDGMGjWKX/3qV7z66qv079+f\nLl268Pbbb7N//36sViuDBw9m4MCB2Gw2li1bRmZmJgEBATRr1oy77rqL3r17M2PGDB555BHCw8NZ\nunQphw4doqamhl69ejFs2DDg7MxG//792bVrF3V1dUyaNIng4OB652Aymfj6668ZNGgQOTk53H33\n3Xz22WcA5ObmEh4ejsFgIDU1lR49etC7d29H25qaGubNm0fv3r0ZMGAAI0eOZOXKlWRmZrJ69Wq8\nvLywWCx06dKFsWPHYjAY2LNnD2vWrKG2tpa2bdsyfvx4PD092b17N2+++SZGo5GoqCjHGFu2bCEv\nL4/Ro0ezc+dO3n//faxWKy1atODJJ5/E19f3yu6+iIiIiEgTccVrTAYPHsznn39ORUVFvdeXLVvG\nXXfdRUpKCn379mX58uWOfSUlJcyePZtHHnkEs9lMVlYW3377LW3btiUrKwuAgwcPYjab+fTTT/H2\n9iYxMZE5c+aQlpZGYWEhX375JUVFRSxYsIAnnniCnJwcx4zN+TM3I0aMIDExkZSUFA4cOMDRo0cd\n+3x8fEhOTmbQoEF89NFHDc7NbDY7Zkyys7Pp3Lkz7u7uVFVVkZOTg9lsdox3/piVlZUkJyfTt29f\nBgwY0CCm3NxcRo8ezcsvv8zJkyf54osvOH36NO+99x7PP/88ycnJhIeHs27dOmpqavjHP/7B1KlT\nSU5OprS01OnMVHR0NC+99BLJycncfvvtfPjhh5d5B0VEREREmp4rmjEB8PLyon///mzYsAGj0eh4\n/eDBg8THxwPQr18//vnPfwJn36D36dPH8eY6OjqaAwcO0KZNGwYNGsSmTZv4/vvv8fb2xmg0smfP\nHo4ePcqOHTuAs2/6LRYLWVlZ9OnTBwA/Pz+6dOniNL5t27aRlpaGzWajpKSEgoICOnbsCECvXr0A\nCAsL44svvmjQtk2bNlitVkpLSzl+/DjBwcF06tSJgwcPkpOTw7333ut0zLlz5/LrX/+avn37Ot0f\nERFBYGAgAHfccQdZWVkYjUYKCgqYNm0aAFarFZPJxPHjxwkMDCQoKMhxLTdt2tSgz++++44FCxZQ\nWlqK1Wp19H8x7h7u+Pv7X/I4+d+gXBBnlBfijPJCnFFeSGO74sIEYMiQIUydOpW77rqr3ut2u93p\n8ecXMNHR0Xz88ccUFxfz+9//ni+//JIdO3YQHR3tOGbMmDF069atXh8ZGRmXjKuwsJB169aRlJTE\nDTfcwKJFi6itrXXs9/DwAMDNze2C611MJhPbt2/Hz8/PsZ2VlUVubi4mk8lpm6ioKHbv3n3BwuT8\nGQ+73Y7BYMBut9OtWzeeeuqpeseeW3x/KcuWLeOBBx6gR48e7N+//7K+hMBaa6WkpOSy+pdfNn9/\nf+WCNKC8EGeUF+KM8kKuh6v6uuAWLVrQp08fPv30U8ebbrPZzH//+18A0tPT6xUa5wsICKCsrAyL\nxUJgYCBRUVGsXbvWcXxMTAwbN26krq4OgOPHj1NdXU1UVBQ7duzAbrdTWlrK/v37G/RdUVFB8+bN\n8fLyorS0lK+//vqKz81sNrN+/XrHY1smk4mtW7fi7++Pl5eX0zbDhw/H29ubpUuXOt2fm5tLYWEh\nNpuN7du3Ex0dTWRkJNnZ2VgsFuDs4voTJ07Qvn17ioqKOHnyJHD2WjpTWVnp+KRiy5YtV3yeIiIi\nIiJNyRXNmJz/yf8DDzzAxo0bHdujR49m0aJFrF27Fh8fH8aPH++0HUBkZKRjxiIqKopVq1Y5FnnH\nxsZSVFTE1KlTsdvt+Pr6EhcXR69evdi7dy+TJk0iICCAsLAwbrjhhnr9hoaGEhYWxtNPP03r1q3r\nLRy/0Hn8mMlk4s0333TMjvj5+WGz2S44W3LOo48+yqJFi3j77bf54x//WG+MTp06sWzZMiwWC127\nduW2224DYPz48SxcuBCr1QqcXR/Trl07HnvsMZKSkjAajURHRzuKlPMNHTqUBQsW4O3tTdeuXSkq\nKrpofCIiIiIiTZnBfqHnr5qgqqoqPD09KSsr49lnn2X27NlN/puoMjMzWbt2LQkJCS6LIS0tjYQM\nAylDIogJbumyOKTp0BS8OKO8EGeUF+KM8kKcycjIIDY29qrbX9UaE1dJTk7mzJkzWK1Wfve73zX5\nogQafoOXq6QMiaBtS+OlDxQRERERcYGfVWHywgsvuDqEK9a5c2c6d+7s6jA0UyIiIiIiTdpVLX4X\nERERERFpTCpMRERERETE5VSYiIiIiIiIy6kwERERERERl1NhIiIiIiIiLqfCREREREREXE6FiYiI\niIiIuJzLCpORI0c6/p6RkcFTTz1FcXGxq8LBbrczZswYKioqACgpKWH48OFkZWU5jhk7dizl5eW8\n9tprHDt2DKh/Hud8//33vPzyy40S15YtW1i2bBkAn3zyCVu3br2qfvYcL8NSVt0oMYmIiIiINDaX\nFSbnfg197969rFixgueee47WrVtfsp3dbsdut1+XeCIjI8nOzgYgOzub0NBQcnJyADh+/DgtW7ak\nRYsWPP7447Rv377eeZyvVatWTJo0qdFjvPvuu/nVr351VW3j1udysqymkSMSEREREWkcLv3l9/37\n9/P666/zzDPPEBgYCMC6devYvHkzALGxsQwZMoTCwkJeeuklIiMjyc/P55lnnmHbtm1s374dq9XK\nrbfeyrBhwwBISUnhu+++o7a2lnvvvZeBAwcCZ2c27rvvPnbt2oXRaCQ+Ph5fX9968ZjNZrKzs7nl\nllvIycnhvvvu48svvwTOFipmsxmAGTNm8MgjjxAeHu5oe/r0aebOncvDDz9M+/btSU5OZv78+WzZ\nsoUvv/ySyspKvv/+e/r168fvfvc7ALZu3crHH3+M1WolIiKCsWPH4ubmxubNm/nggw/w9vbmxhtv\nxMPDA4DVq1fj5eXFAw88wKZNm0hLS8NqtRIUFMTEiRMxGo3X61aJiIiIiFxXLpsxqa2tZd68ecTF\nxREcHAxAXl4eW7ZsITExkZdeeom0tDQOHz4MgMVi4Z577mH+/PkcO3YMi8VCYmIiycnJ5OXlceDA\nAQDGjRtHUlISiYmJbNiwgfLycgBqamowmUykpKQQHR1NWlpag5jMZrNjhuTQoUPcdtttfPfdd0D9\nwuTHsySnTp0iOTmZ4cOHc8sttzTo99ChQ0yePJmUlBS2b99OXl4eBQUFbN++ndmzZzN37lzc3NxI\nT0+npKSENWvWMHv2bGbOnElBQYGjn/PH7d27N4mJiaSkpNC+fXs+/fTTq7oPIiIiIiJNgctmTNzd\n3TGbzXz66aeMGjUKgKysLG677TbHJ/+33XYbBw4coGfPnrRp04aIiAgA9uzZw549e4iPjweguroa\ni8VCdHQ069ev56uvvgLgu+++w2KxEBERgbu7O927dwcgPDycb775pkFMnTp1Ij8/n+rqaqxWK56e\nngQGBmKxWDh48CAPPvhggzZWq5WZM2cyduxYoqOjnZ5rt27daNGiBQC9evUiKysLNzc38vLySEhI\nAM4Wan5+fuTm5tK5c2datmwJwO23386JEyca9Hn06FH+3//7f1RUVFBVVUVMTMzlXXgRERERkSbI\nZYWJwWDg//7v/5g5cybvv/8+v/nNbxocY7fbHbMEzZs3r7fvN7/5jeMxrXMyMzPZt28fL730Ekaj\nkRdffJGamrPrKpo1a+Y4zs3NDZvN1mC85s2b065dOzZv3ux4TCsyMpKMjAxOnTrlmNk5X7NmzejU\nqRO7d+++YGHy43M6p3///vzhD3+ot/9cUeXsePhh1iQ1NZWpU6fSsWNHtmzZwv79+y85truHO/7+\n/pc8Tv43KBfEGeWFOKO8EGeUF9LYXLrGxGg0kpCQwAsvvICvry/R0dEsWrSIhx56CJvNxldffcXE\niRMbvDm/+eabeffdd+nbty+enp58//33uLu7U1lZibe3N0ajkWPHjjkey7oSJpOJf//73441KyaT\nib///e+YTCanxxsMBsaNG8f8+fP58MMP+fWvf93gmG+++Yby8nKMRiM7d+5k3LhxGI1GUlJSuP/+\n+/Hx8aG8vJyqqioiIyNZsWIF5eXleHp6smPHDkJDQ4H6C/+rqqrw8/PDarXy+eefExAQcMlzs9Za\nKSkpueJrIr88/v7+ygVpQHkhzigvxBnlhVwPLp0xAWjRogXPPvssL7zwAo8++ij9+/fnmWeeAc4u\nfg8NDaWwsLDe+opu3bpRUFDAtGnTAPDy8mLixIncfPPNfPLJJ/zf//0fwcHB9YoJZ9+e5UxUVBQb\nNmxwtA0LC+P7778nNjb2gudhMBh4+umnSU5OxsvLi1tuuaXeeBEREcyfP9+x+P3cbMzw4cOZPXs2\ndrudZs2aMXbsWCIiIhg6dCjPPfcc3t7ejqLk/LHOtX322Wfx8fEhIiKCqqqqyzo/EREREZGmyGC/\nHt+9Kw5btmwhLy+P0aNHuyyGtLQ0EjIMpAyJICa4pcvikKZDn3SJM8oLcUZ5Ic4oL8SZjIyMC36Y\nfzn0y+//I1KGRNC2pb5OWERERESaJpeuMflfcOedd3LnnXe6OgzNlIiIiIhIk6YZExERERERcTkV\nJiIiIiIi4nIqTERERERExOVUmIiIiIiIiMupMBEREREREZdTYSIiIiIiIi6nwkRERERERFzusguT\nkSNHXlHHmZmZJCUlXXFAV2P16tXs3bv3mvtJSUnhq6++cmw/9dRTvPfee47tefPm8eWXX/LJJ5+w\ndetWAGbMmEFeXl6DvhITE6moqLjmmAoLC5k8eTIAhw4dYvny5dfcp4iIiIhIU3PZP7BoMBiuZxzX\nZNiwYY3ST1RUFNnZ2dx6662UlZXh6elJTk6OY//Bgwf5y1/+gq+vr+O1C12XZ555plFiOl+nTp3o\n1KnTVbXdc7yMti2NBLVs3shRiYiIiIhcuyv+5ffMzEzWrFmDj48P3377LWFhYTz55JMA7N69mzff\nfBOj0UhUVJSjTXl5OYsWLaKoqAij0chf//pXOnbsyOrVqykuLqaoqIji4mKGDBnCvffeC8DWrVv5\n+OOPsVqtREREMHbsWAAWL15MXl4eBoOBAQMGMGTIEFJTU+nRowe9e/fmX//6F7t27aKmpgaz2cxj\njz0GnJ3ZiIyMJDMzkzNnzjBu3Lh6MQKYzWb++c9/ApCdnU2PHj3YvXs3cHbmwmg04uvry+rVq/Hy\n8uKBBx5wtLXZbCxevJjWrVszfPhwJkyYQHJyMhUVFcyZM4fw8HDy8/MJCQnhiSeewGg0kpeXx1tv\nvUVVVRUtW7ZkwoQJ+Pn5kZeXx+LFiwGIiYmpd+3Xrl1LQkICubm5rFixgtraWoxGI+PGjSM4OPiC\n9y1ufS4pQyJUmIiIiIhIk3RVa0wOHz7MqFGjePnllyksLCQ7O5uamhr+8Y9/MHXqVJKTkyktLXXM\nJqxevZrw8HBSUlL4/e9/z6uvvuro68SJEzz33HPMmTOHf/3rX9hsNgoKCti+fTuzZ89m7ty5uLm5\nkZ6ezuHDhykpKWH+/PnMmzePO++8Ezg7a3FurHvuuYfExETmz59PTU0Nu3btchxjs9mYM2cOo0aN\nYs2aNQ3OKywsjG+//Rar1UpOTg4mk4ng4GAKCgrIzs7GbDY3GA+grq6OV155heDgYIYPH96g3xMn\nTnDPPfewYMECvLy82LhxI3V1dSxbtozJkyeTlJTEXXfdxapVqwBYtGgRY8aMISUl5YL3oH379syc\nOZPk5GSGDh3qaCsiIiIi8nN0xTMmABEREbRq1QqA0NBQCgsLad68OYGBgQQFBQHQr18/Nm3aBJyd\nfTi3TqJr166UlZVRWVmJwWCge/fuuLu707JlS3x8fCgtLWXfvn3k5eWRkJAAQE1NDX5+fvTo0YOT\nJ0+ybNkyunfvXm824Zx9+/axdu1aqqurKS8vJyQkhB49egDQq1cv4GwBUlRU1KCth4cHHTp0ID8/\nn4MHD/LrX/+awsJCcnJyyM/PbzDDAmC323n99de5/fbb+c1vfuP0egUEBGAymRzXZcOGDdx88818\n++23zJo1Czg74+Lv709FRQUVFRWOsX71q1/x9ddfN+jzzJkzvPrqq1gsFgwGA3V1dU7HFhERERH5\nObiqwsTd/Ydmbm5u1/Sm+EJ99e/fnz/84Q8Njp83bx67d+/mk08+Yfv27YwbN86xr6amhjfeeIPk\n5GRatWrFmjVrqK2tbTDWxWI2m83s37+fqqoqvL29iYyMZMOGDRw5coS77767wfEGgwGz2cy+ffu4\n//778fDwcHrMj7ftdjshISHMnj273r4zZ87U27bb7U7jfPfdd7npppuIi4ujqKiIGTNmOD3ufO4e\n7vj7+1/yOPnfoFwQZ5QX4ozyQpxRXkhju6rC5McMBgPt27enqKiIkydP0rZtW9LT0x37o6Ki+Pzz\nz3n44YfJzMzEx8cHLy8vp2+6DQYDXbt2JSUlhfvvvx8fHx/Ky8upqqqiefPmNGvWjF69etGuXTtS\nU1PrtT1XhLRo0YKqqip27NhBnz59ruhczGYzb731Fl26dAHgxhtv5ODBg5w+fZqOHTs6jjs/9gED\nBrB//34WLFjAlClTcHOr/4RccXGx49Gw9PR0oqKiCA4O5vTp047XrVYrFouFDh064O3tTVZWluO6\nOVNZWen4D2Hz5s2XdW7WWislJSVXdD3kl8nf31+5IA0oL8QZ5YU4o7yQ6+GqvpXL2TdReXh48Nhj\nj5GUlITRaCQ6OpqTJ08CMHToUBYvXkxcXBzNmzdnwoQJjn6c9dWhQweGDx/O7NmzsdvtNGvWjLFj\nx+Lh4cGiRYscRcGPZ1S8vb2JjY1l8uTJ+Pn5ERERcVnncz6TyURhYaHjsSw3Nzf8/Pxo06bNRdvf\nf//9VFRU8OqrrzJx4sR6+4KDg9m4cSOLFy8mJCSEQYMG4e7uzqRJk1i+fDkVFRXYbDbuu+8+OnTo\nwLhx41i8eDEGg4Fu3bo5vfYPPvggqampvPfee9xyyy1N+lvTREREREQuxWC/0LNC0igKCwtJTk5m\n/vz5LoshLS2NhAwDKUMiiAlu6bI4pOnQJ13ijPJCnFFeiDPKC3EmIyOD2NjYq26vX37/CWg2Q0RE\nRETk4lSYXGeBgYHMmzfP1WGQMiSCti2Nrg5DRERERMSpRln8Lk2fHuESERERkaZMMyYiIiIiIuJy\nKkxERERERMTlVJiIiIiIiIjLqTARERERERGXU2EiIiIiIiIup8JERERERERcToWJiIiIiIi4XJMq\nTEpLS1m4cCETJ04kISGBadOm8eWXXzZK35mZmSQlJV30mPj4eA4fPgxAXV0dI0eO5PPPP3fsnzp1\nKvn5+axevZp9+/YBMGHCBMrLyxv0NX369EaPe+fOnXzwwQdX1c+e42VYyqobJSYRERERkcbWZH5g\n0W63k5KSwl133cVTTz0FQHFxMTt37mxwbF1dHc2aNWv0GKKiosjJySE0NJQjR44QHBxMdnY2/fr1\no6qqisLCQkJDQwkLC7tkX7NmzWr0+Hr27EnPnj2vqm3c+lxShkQQ1LJ5I0clIiIiInLtmkxhsm/f\nPjw8PBg4cKDjtdatW3PPPfcAsGXLFr744guqq6ux2+0kJCTwxhtvUFBQQF1dHUOHDqVnz57YbDbe\nfvtt9u/fj9VqZfDgwfX6BMjNzWXJkiVMnjyZwMBAx+smk4mvv/6aQYMGkZOTw913381nn33maBMe\nHo7BYCA1NZUePXrQu3dvR9uamhrmzZtH7969GTBgACNHjmTlypVkZmayevVqvLy8sFgsdOnShbFj\nx2IwGNijV8I4AAAgAElEQVSzZw9r1qyhtraWtm3bMn78eDw9Pdm9ezdvvvkmRqORqKgoxxhbtmwh\nLy+P0aNHs3PnTt5//32sVistWrTgySefxNfX97rcGxERERGR663JFCbffvvtJWciDh8+zLx58/D2\n9uadd97hpptuYvz48Zw5c4Znn32Wm266ic8//xxvb28SExOpra3l+eefp1u3bo4+srOzWb58OfHx\n8QQEBNTr32w28+677zqOGzp0KP/973+pqqoiJycHs9kMgMFgwGAwONpVVlayYMEC+vfvz69+9SvH\nMefk5uayYMECWrduzZw5c/jiiy/o3Lkz7733Hs8//zxGo5EPPviAdevW8eCDD/KPf/yDF154gaCg\nIBYsWFCvr3Oio6MdsydpaWl8+OGHPPLII1dyyUVEREREmowmU5j8+M33G2+8QVZWFu7u7iQmJgJw\n00034e3tDcA333zDrl27WLt2LQBWq5Xi4mL27NnD0aNH2bFjB3C2aLBYLLi7u3Ps2DFef/11pk+f\njp+fX4MY2rRpg9VqpbS0lOPHjxMcHEynTp04ePAgOTk53HvvvU5jnzt3Lr/+9a/p27ev0/0RERGO\nmZk77riDrKwsjEYjBQUFTJs2zRG/yWTi+PHjBAYGEhQUBEC/fv3YtGlTgz6/++47FixYQGlpKVar\ntd7Mj4iIiIjIz02TKUxCQkL44osvHNtjxoyhrKyMhIQEx2uenp712kyZMoV27do16GvMmDH1ZkkA\n9u/fj7+/P7W1teTl5dG9e3encZhMJrZv3+4oXEwmE1lZWeTm5mIymZy2iYqKYvfu3RcsTM4vuux2\nOwaDAbvdTrdu3Rzrac45t/j+UpYtW8YDDzxAjx492L9/P2vWrLlkG3cPd/z9/S+rf/nlUy6IM8oL\ncUZ5Ic4oL6SxNZnCpGvXrqxatYr//Oc/DBo0CIDq6gt/i1RMTAwbNmxg9OjRAOTn5xMWFkZMTAwb\nN26kS5cuNGvWjOPHjxMQEIDdbueGG25g3LhxzJ49G09PTzp37tygX7PZzPr167nrrruAs4XJypUr\n8ff3x8vLy2ksw4cPZ82aNSxdupSxY8c22J+bm0thYSGtW7dm+/bt3H333URGRvLGG29gsVgICgqi\nqqqKkpIS2rdvT1FRESdPnqRt27akp6c7HbOystLxH8KWLVsufGHPY621UlJSclnHyi+bv7+/ckEa\nUF6IM8oLcUZ5IddDkylMAOLi4lixYgUfffQRPj4+NG/enD/96U9Oj3344YdZsWIFU6ZMwW63ExgY\nyNSpU4mNjaWoqIipU6dit9vx9fVlypQpjnUhvr6+TJ06lcTERMaNG0dERES9fk0mE2+++aZjdsTP\nzw+bzXbB2ZJzHn30URYtWsTbb7/NH//4x3qzJJ06dWLZsmVYLBa6du3KbbfdBsD48eNZuHAhVqsV\ngBEjRtCuXTsee+wxkpKSMBqNREdHc/LkyQbjDR06lAULFuDt7U3Xrl0pKiq6/AstIiIiItLEGOx2\nu93VQfySZWZmsnbt2nqPpP3U0tLSSMgwkDIkgpjgli6LQ5oOfdIlzigvxBnlhTijvBBnMjIyiI2N\nver2TeoHFn+JfvwNXq6SMiSCti2Nrg5DRERERMSpJvUo1y9R586dna5l+alppkREREREmjLNmIiI\niIiIiMupMBEREREREZdTYSIiIiIiIi6nwkRERERERFxOhYmIiIiIiLicChMREREREXE5FSYiIiIi\nIuJyTaYwKS0t5W9/+xsTJ04kISGBxMRETpw40Sh9r169mrVr1170mPXr17NixQrH9uuvv86sWbMc\n2xs2bGD58uXk5eWxfPnyi/a7evVq9u7d2yixT5gwgfLycgCmT59+1f3sOV6Gpay6UWISEREREWls\nTeIHFu12OykpKdx11108/fTTABw5coRTp07Rrl27a+7/cn55PSoqivT0dMf2kSNHsNvt2O12DAYD\nOTk53HrrrYSHhxMeHn7RfocNG3bNMTtzfqF0peLW55IyJIKgls0bMSIRERERkcbRJAqTzMxM3N3d\nGThwoOO1G2+8ETg7+7Bz504ATp06RUxMDOPHj2fr1q18/PHHWK1WIiIiGDt2LG5ubuzevZtVq1Zh\ns9nw8fFxzDIUFBTw4osvUlxczJAhQ7j33nvrxXDjjTdy4sQJamtrqa2txWg00q5dO44cOUJoaCg5\nOTmMHDmSzMxM1q5dS0JCAvBDcbJp0ya++uorJk+ezJIlS+jRowe9e/dmwoQJ9OnTh927d2M0Gnny\nyScJCgri9OnTLFmyhOLiYgBGjRqF2WymrKyMhQsXUlJSQmRkZL0YR44cycqVK6mqqiIlJYXy8nLq\n6uoYMWIEPXv2vA53RkRERETkp9EkCpOjR486ZiF+bNiwYQwbNoyKigqef/557r33XgoKCti+fTuz\nZ8/Gzc2NpUuXkp6ezs0338w//vEPZs6cSZs2bThz5gxwdkbm+PHjvPDCC1RWVvL0008zePBg3Nx+\neJKtWbNmhIaGkpubS3V1NZGRkQQFBZGTk4OPjw92u51WrVo1eLzMbrfz8ccfs3fvXuLi4nB3d8dg\nMNSbTfH29mbevHls3bqVFStWkJCQwPLly7nvvvuIioqiuLiYl156iQULFrBmzRqio6N5+OGHycjI\nYPPmzY5+zvVpNBqZMmUKXl5enD59mmnTpqkwEREREZGftSZRmFzqUSu73c4rr7zC/fffT1hYGB9/\n/DF5eXmOWYva2lr8/Pw4ePAg0dHRtGnTBjhbEJzrv3v37ri7u9OyZUt8fHwoLS2lVatW9cYxm81k\nZ2dTU1ODyWQiKCiI999/Hx8fH0wmk9O4tm7dSkBAAPHx8fUKnfPdcccdANx+++28+eabAOzdu5dj\nx445jqmqqqKqqoqsrCymTJkCQPfu3R3ncD6bzcY777xDVlYWBoOBkpISTp06ha+v70Wvo4iIiIhI\nU9UkCpOQkBB27Nhxwf1r1qyhdevW3HnnnY7X+vfvzx/+8Id6x+3ateuCfbi7/3Cqbm5u2Gy2BseY\nzWb+85//YLVaueeee2jZsiUFBQX4+PgQFRXV4HiDwUBISAhHjhyhuLiYwMDAi51mPXa7nTlz5tSL\n63Klp6dTVlZGcnIybm5uTJgwgdra2ku2c/dwx9/f/4rHk18m5YI4o7wQZ5QX4ozyQhpbkyhMunbt\nyqpVq9i0aZNjncmRI0eorKykvLycvXv38sILL9Q7PiUlhfvvvx8fHx/Ky8upqqoiMjKSpUuXUlhY\nSGBgIOXl5bRo0eKy4zCZTKSmphIQEICPjw8APj4+7Ny5k0mTJjU43m63ExYWxqBBg5g7dy7PPfec\n03+k27Zt46GHHmLbtm2YzWYAYmJiWL9+PQ8++CAAhw8fJjQ0lOjoaNLT0/ntb3/L119/7Xgc7XwV\nFRX4+Pjg5ubGvn37HOtULsVaa6WkpOSyr4f8cvn7+ysXpAHlhTijvBBnlBdyPTSJwgRgypQprFix\ngg8//BCj0UibNm0YNWoU7777LiUlJTz77LMA9OzZk2HDhjF8+HBmz56N3W6nWbNmjB07loiICP76\n178yf/58bDYbfn5+PPfcc8DlfTOXt7c3vr6+hISEOF4zmUzk5OQ4FuOfv37k3N+joqIYOXIkSUlJ\nTJs2rUG/Z86cIS4uDg8PD5566ikAHn30Ud544w3i4uKoq6ujc+fOjB07lqFDh7Jw4UImT56MyWSi\ndevWjn7OjduvXz+Sk5OZMmUK4eHhtG/f/mouuYiIiIhIk2Gw2+12VwfxSzZhwgSSk5OvaOamsaWl\npZGQYSBlSAQxwS1dFoc0HfqkS5xRXogzygtxRnkhzmRkZBAbG3vV7ZvMDyz+Ul3OTI2IiIiIyP+6\nJvMo1y/Vq6++6uoQAEgZEkHblkZXhyEiIiIi4pQKk/8ReoRLRERERJoyPcolIiIiIiIup8JERERE\nRERcToWJiIiIiIi4nAoTERERERFxORUmIiIiIiLicipMRERERETE5VSYiIiIiIiIyzV6YTJy5MjG\n7vIncfjwYeLj4x3b6enp/OlPf8JmswFw9OhR4uLiAJg+fToAmZmZJCUlNehr586dfPDBB40SV2pq\nKjt27ADgtddeo6Cg4Kr62XO8rN4fS1l1o8QnIiIiItIYGv0HFg0GQ2N3+ZPo2LEjxcXFVFVV4enp\nSU5ODh06dCAvL4+IiAiys7Mxm80AzJo166J99ezZk549ezZKXAaDwXFNH3/88avuJ259br3tlCER\nBLVsfk2xiYiIiIg0luvyy+9VVVWkpKRQXl5OXV0dI0aMoGfPnhQWFjJnzhzCw8PJz88nJCSEJ554\nAqPRyL/+9S927dpFTU0NZrOZxx57DIAZM2YQGRlJZmYmZ86cYdy4cURFRWGz2Xj77bfZv38/VquV\nwYMHM3DgQEpKSvjb3/5GZWUldXV1/OUvfyEqKoo9e/awZs0aamtradu2LePHj8fT09MRs5ubG506\ndeLgwYPcdNNN5OfnM3jwYHJychyFSUxMDHB2VmjlypX1zjk3N5clS5YwadIkDhw4QF5eHqNHjyY1\nNRUPDw/y8/OpqKjgz3/+M927d79g/Ha7nWXLlrF3714CAgJwd//hFs2YMYNHHnmE8PBwli5dyqFD\nh6ipqaFXr14MGzbsetxKEREREZGfxHVZY2I0GpkyZQrJyck8//zzvPXWW459J06c4J577mHBggV4\neXmxceNGAO655x4SExOZP38+NTU17Nq1Czg7Y2Cz2ZgzZw6jRo1izZo1AHz66ad4e3uTmJjInDlz\nSEtLo7CwkP/+97/ExMQwd+5cUlJSCA0N5fTp07z33ns8//zzJCcnEx4ezrp16xrEbTabyc7Oprq6\nGoPBQOfOncnOzgbg4MGDjhmTH88KZWdns3TpUuLj42nbtm2Dfr/77jsSExN55plnWLJkCbW1tReM\n/8svv+TEiRMsWLCAJ554gpycHEc/5487YsQIEhMTSUlJ4cCBAxw9evSq7pWIiIiISFNwXWZM7HY7\n77zzDllZWRgMBkpKSjh16hQAAQEBmEwmAPr168eGDRt44IEH2LdvH2vXrqW6upry8nJCQkLo0aMH\nAL169QIgLCyMoqIiAPbs2cPRo0cd6y8qKyuxWCx06tSJxYsXU1dXx6233kpoaCiZmZkUFBQwbdo0\nAKxWqyOG85lMJtatW0dubi4RERG0bdsWi8XC6dOnqaqqIjAwsEGbY8eO8frrrzN9+nT8/Pwa7DcY\nDPTp0weAoKAgAgMDOXbs2AXjP3DgAH379sVgMODv70+XLl2cXuNt27aRlpaGzWajpKSEgoICOnbs\neJl3CNw93PH397/s4+WXR/dfnFFeiDPKC3FGeSGN7boUJp9//jllZWUkJyfj5ubGhAkTqK2tBRrO\nNhgMBmpra3njjTdITk6mVatWjkeuHEH+/48zubm5UVdX53h9zJgxdOvWrcH4M2fOZNeuXSxatIj7\n778fb29vunXrxlNPPXXRuCMjIzl06BDZ2dmOwiUgIIBt27YRGRnptI2/vz+1tbXk5eXRvXv3y7g6\nP1wDZ/FnZGRgt9sv2r6wsJB169aRlJTEDTfcwKJFi6ipqbmssc+x1lopKSm5ojbyy+Hv76/7Lw0o\nL8QZ5YU4o7yQ6+G6PMpVUVGBj48Pbm5u7Nu3j+LiYse+4uJix+NJ6enpREVFOYqQFi1aUFVV5ZhF\nuJiYmBg2btzoKFSOHz9OdXU1xcXF+Pj4EBsby4ABA8jPzycyMpLs7GwsFgtwdg3MiRMnGvTp5eVF\nq1at2Lx5s6MwMZlM/Pvf/yYqKsppHDfccAMJCQmsWrWK/fv3N9hvt9vZvn07drsdi8VCYWEh7du3\nv2D8nTt3Ztu2bY6ZkMzMTKfXt3nz5nh5eVFaWsrXX3/9s/3SARERERERaOQZk7q6Ojw8POjXrx/J\nyclMmTKF8PBw2rdv7zgmODiYjRs3snjxYkJCQhg0aBBGo5HY2FgmT56Mn58fERERFxzj3Bvw2NhY\nioqKmDp1Kna7HV9fX6ZMmUJmZiYfffQR7u7ueHp68sQTT+Dj48P48eNZuHAhVqsVOLtGo127dg36\nj4qKYufOnbRq1Qo4W5isWrXKsb7k/BjO/d3X15epU6eSmJjIuHHjGsTbunVrnn32WSoqKvjLX/6C\nu7u70/jj4uK47bbb2LdvH5MmTaJ169b1xj0nNDSUsLAwnn76aVq3bn3BoklERERE5OfCYL/Uc0NX\n4PDhwyxZsoSXXnrJ6f7CwkKSk5OZP39+Yw3Z5C1atIgePXo41sm4QlpaGgkZ9WdUUoZEEBPc0kUR\niatpCl6cUV6IM8oLcUZ5Ic5kZGQQGxt71e0bbcbkP//5Dx9//DGjRo266HF65Mg1UobUn4Vq29Lo\nokhERERERBpq1BkTaZrS0tIue2G+/G/QJ13ijPJCnFFeiDPKC3HmWmdMrsvidxERERERkSuhwkRE\nRERERFxOhYmIiIiIiLicChMREREREXE5FSYiIiIiIuJyKkxERERERMTlVJiIiIiIiIjLXffCpLS0\nlL/97W9MnDiRhIQEEhMTOXHiBJmZmSQlJTlt89prr1FQUHC9Q6tn/fr1rFixwrH9+uuvM2vWLMf2\nhg0bWL58OXl5eSxfvhyA1atXs3bt2gZ9rV69mr179zZKXBMmTKC8vByA6dOnX3U/e46X1ftjKatu\nlPhERERERBpDo/3yuzN2u52UlBTuuusunn76aQCOHDnCqVOnLvoL8I8//vj1DMupqKgo0tPTHdtH\njhzBbrdjt9sxGAzk5ORw6623Eh4eTnh4OHDhX7EfNmzYdYnx/ELpSsWtz623nTIkgqCWza81JBER\nERGRRnFdC5PMzEzc3d0ZOHCg47Ubb7wRgP3791NdXc3LL7/Mt99+S1hYGE8++SQAM2bM4JFHHiE8\nPJyRI0dy3333sWvXLoxGI/Hx8fj6+nL69GmWLFlCcXExAKNGjcJsNrN//37HzIfBYODFF1/E09OT\njz76iO3bt2O1Wrn11lsbFA833ngjJ06coLa2ltraWoxGI+3atePIkSOEhoaSk5PDyJEjyczMZO3a\ntSQkJDjGANi0aRNfffUVkydPZsmSJfTo0YPevXszYcIE+vTpw+7duzEajTz55JMEBQVdMP6ysjIW\nLlxISUkJkZGR9WIcOXIkK1eupKqqipSUFMrLy6mrq2PEiBH07Nmzke+eiIiIiMhP57oWJkePHnXM\nLvyY3W4nPz+fl19+GX9/f6ZPn052djZms7neTERNTQ0mk4kRI0bwz3/+k7S0NH7729+yfPly7rvv\nPqKioiguLuall15iwYIFrF27lrFjx2IymaiursbDw4M9e/ZgsVhITEzEZrMxd+5cDhw4QHR0tGOc\nZs2aERoaSm5uLtXV1URGRhIUFEROTg4+Pj7Y7XZatWrFiRMnGpzHxx9/zN69e4mLi8Pd3R2DwVDv\nHLy9vZk3bx5bt25lxYoVJCQkXDD+NWvWEB0dzcMPP0xGRgabN2929HOuT6PRyJQpU/Dy8uL06dNM\nmzZNhYmIiIiI/Kxd18LkYo9rAURERNCqVSsAQkNDKSoqwmw21zvG3d2d7t27AxAeHs4333wDwN69\nezl27JjjuKqqKqqqqjCbzbz55pv07duXXr160apVK/bs2cOePXuIj48HoLq6GovFUq8wATCbzWRn\nZzuKoaCgIN5//318fHwwmUwN4rfb7WzdupWAgADi4+Nxc3O+ZOeOO+4A4Pbbb+fNN9+8aPxZWVlM\nmTIFgO7du+Pt7d2gP5vNxjvvvENWVhYGg4GSkhJOnTqFr6/vhS61iIiIiEiTdl0Lk5CQEHbs2HHh\nwd1/GN7NzY26uroGxzRr1qzeMTabDThbFMyZM6deHwAPPfQQPXr0ICMjg+nTp/Pcc88B8Jvf/Kbe\nI2XOmM1m/vOf/2C1Wrnnnnto2bIlBQUF+Pj4EBUV1eB4g8FASEgIR44cobi4mMDAwIv2f74LxX85\n0tPTKSsrIzk5GTc3NyZMmEBtbe0V9eHu4Y6/v/8Vjy2/HLr/4ozyQpxRXogzygtpbNe1MOnatSur\nVq1i06ZNjqLgyJEjVFZWXnI25VJiYmJYv349Dz74IACHDx8mNDQUi8VCSEgIISEhHDp0iOPHj3Pz\nzTfz7rvv0rdvXzw9Pfn+++9xd3fHx8enXp8mk4nU1FQCAgIc+3x8fNi5cyeTJk1qEIPdbicsLIxB\ngwYxd+5cnnvuOaf/SLdt28ZDDz3Etm3bHDNCF4o/Ojqa9PR0fvvb3/L1119z5syZBv1VVFTg4+OD\nm5sb+/btc6xTuRLWWislJSVX3E5+Gfz9/XX/pQHlhTijvBBnlBdyPVzXwgRgypQprFixgg8//BCj\n0UibNm0YNWoU33///WUVJxc65tFHH+WNN94gLi6Ouro6OnfuzNixY1m/fj2ZmZm4ubkREhLCzTff\njLu7OwUFBUybNg0ALy8vJk6c2KAw8fb2xtfXl5CQEMdrJpOJnJwcx6L989ePnPt7VFQUI0eOJCkp\nyTHG+c6cOUNcXBweHh489dRTF41/6NChLFy4kMmTJ2MymWjdunWDa9GvXz+Sk5OZMmUK4eHhtG/f\n/pLXUURERESkKTPY7Xa7q4P4JZswYQLJycm0aNHCZTGkpaWRkFG/wEsZEkFMcEsXRSSupk+6xBnl\nhTijvBBnlBfiTEZGBrGxsVfd/rrPmPyvu9ZH1hpLypCIetttWxpdFImIiIiISEMqTK6zV1991dUh\nAGh2RERERESaNOffbysiIiIiIvITUmEiIiIiIiIup8JERERERERcToWJiIiIiIi4nAoTERERERFx\nORUmIiIiIiLicipMRERERETE5X4Whcnw4cOJj493/CkuLr6m/lavXs3evXvrvXb48GHi4+Md2+np\n6fzpT3/CZrMBcPToUeLi4gCYPn06AJmZmSQlJTXof+fOnXzwwQfXFOM5qamp7NixA4DXXnuNgoKC\nRulXRERERKQp+Vn8wGLz5s2ZO3fuFbWx2+2A819eHzZsWIPXOnbsSHFxMVVVVXh6epKTk0OHDh3I\ny8sjIiKC7OxszGYzALNmzbro2D179qRnz55XFO+FGAwGxzk8/vjjV93PnuNl9bbbtjQS1LL5NcUm\nIiIiItJYfhaFyY9VVVWRkpJCeXk5dXV1jBgxgp49e1JYWMhLL71EZGQk+fn5PPPMM7z77rvk5eVh\nMBgYMGAAQ4YMITU1lR49etC7d29Hn25ubnTq1ImDBw9y0003kZ+fz+DBg8nJyXEUJjExMQCMHDmS\nlStX1ospNzeXJUuWMGnSJA4cOEBeXh6jR48mNTUVDw8P8vPzqaio4M9//jPdu3fHZrPx9ttvs3//\nfqxWK4MHD2bgwIHY7XaWLVvG3r17CQgIwN39h1s0Y8YMHnnkEcLDw1m6dCmHDh2ipqaGXr16OS22\nzhe3PrfedsqQCBUmIiIiItJk/CwKk5qaGsdjVoGBgUyaNIkpU6bg5eXF6dOnmTZtmmOGwmKxMHHi\nRCIiIsjLy6OkpIT58+cDUFFRAdSfhTif2WwmOzsbk8mEwWCgc+fOvPPOOwwZMoSDBw863vz/uG12\ndjbLly8nPj6egIAADhw4UG//d999R2JiIhaLhRdffJFXXnmFzz77DG9vbxITE6mtreX555+nW7du\n5Ofnc+LECRYsWEBpaSmTJk1iwIABDcYdMWIELVq0wGazMWvWLI4ePUrHjh0b43KLiIiIiPzkfhaF\nidForPcol9Vq5Z133iErKwuDwUBJSQmnTp0CoE2bNkRERADQtm1bTp48ybJly+jevbtjxgN+eNTr\nfCaTiXXr1pGbm0tERARt27bFYrFw+vRpqqqqCAwMbNDm2LFjvP7660yfPh0/P78G+w0GA3369AEg\nKCiIwMBAjh07xp49ezh69Khj/UhlZSUWi4UDBw7Qt29fDAYD/v7+dOnSxek12bZtG2lpadhsNkpK\nSigoKFBhIiIiIiI/Wz+LwuTH0tPTKSsrIzk5GTc3NyZMmEBtbS1wdj3KOd7e3sybN4/du3fzySef\nsH37dsaNG3fBfiMjIzl06JBj1gQgICCAbdu2ERkZ6bSNv7///8fenYdHWd3//39OlklC9rBFlgSy\nzIRFtoAsCmjRohFUWoF+tVhFEAERPkACVLACUkgCIpSApcgitbWAiGJpFSmgbFUIoATIEIJCCCFA\nEpOQDMkk8/uDH1NiJoBAmNi+Htfl1czc57zP+77vo807514oLy8nMzOTTp063VD+V1Y+nn/+edq1\na1dlW2pqqtOi6Wq5ubl8/PHHzJkzh3r16rF48WLKyspuaOwrPDw9CA4O/lF95L+Lzr84o3khzmhe\niDOaF3K7/SQLk5KSEgICAnBzc+PQoUM1PqWrqKgId3d3unbtyl133UVKSso14/r4+BASEsLWrVuZ\nPn06cHkV5e9//zt9+/Z12qdevXqMHDmS119/HW9vb1q3bl1lu91uZ/fu3fTu3ZuzZ8+Sm5tL06ZN\nad++PZ988glt2rTB3d2d7Oxs6tevT+vWrdm8eTO9e/fm+++/Jy0tjZ49e1bbfy8vL3x8fCgoKGD/\n/v01rqzUxFZuIz8//0f1kf8ewcHBOv9SjeaFOKN5Ic5oXkht+EkUJj+8p6Nnz54kJiYyceJEIiIi\naNq0qdO2eXl5LF682LEC8dRTT9UY84qYmBj27t1LSEgIcLkw+etf/+p4ItcP+xoMBgIDA5k0aRKz\nZ8+utiJjMBho0KABv/3tbykpKWH48OF4eHjQp08fzp07x6RJk7Db7QQGBhIfH88999zDoUOHGD9+\nPA0aNKgy7hUtWrSgZcuWjBs3jgYNGhATE3PdYygiIiIiUpcZ7Ne7bkhuyeLFi4mNjaVr164uy2HL\nli1MTq1aiCXHRdG+ib+LMhJX01+6xBnNC3FG80Kc0bwQZ1JTU+nTp89N9/9JvGBRRERERET+u/0k\nLihU/pQAACAASURBVOX6KRs1apSrUwAur5BcrbG/0UWZiIiIiIhUp8Lkf4Qu2xIRERGRukyXcomI\niIiIiMupMBEREREREZdTYSIiIiIiIi6nwkRERERERFxOhYmIiIiIiLicChMREREREXE5FSYiIiIi\nIuJyd+Q9JoMHDyY8PNzx+d577+Xxxx//UTHS0tLYuHEjkydPvt3pOaxcuZJGjRoRFxcHwKxZs6hf\nvz4vvvgiAO+88w4hISGEhoaSlZXFE088QUpKCrGxsXTr1q1KrLfeeot+/frRrFmzW85ryJAhrF69\nmry8PFauXMn48eN/dIyD2UU1bmvsbyTU3+tWUhQRERERuSV3pDDx8vIiKSnpTgzlVEVFBe7u7tdt\nFxMTw+7du4mLi6OyspKioiKsVqtju8Vi4dlnnyUqKorOnTsDYDAYnMa6UszcDlfGCAkJuamiBCB+\nU0aN25LjolSYiIiIiIhLufTN76NHjyYxMRE/Pz+OHz/On//8Z373u99x+PBhVq5cCVz+pXz69OlV\n+mVkZPCnP/2J8ePHc+rUKT744ANsNht+fn68/PLLBAYGsmbNGs6ePUtubi4NGzbk//2//8eiRYsc\nhcbzzz+PyWSqEtdkMrFq1SoAsrKyaN68OQUFBVy8eBGj0cjp06dp2bIl27ZtIzMzk6FDhzpyBHjv\nvffIy8vjxRdfZMaMGTzzzDNEREQwZMgQHnzwQb7++muCgoIYO3YsAQEB5OTksHz5cgoLC/Hy8mLE\niBE0adKE3NxcFixYwKVLlxwFEEBubi6JiYnMmzeP3NxcUlJSrrk/IiIiIiI/FXekMCkrKyMhIcHx\necCAAXTv3r3G9hs3bmTYsGGYTCYuXbqEp6enY1t6ejorVqwgISGB+vXr4+fn5/jlfcuWLXz44Yc8\n88wzAGRnZzNjxgw8PT0pKytj6tSpeHp6cubMGRYuXMjs2bOrjBsSEoK7uzvnz5/HYrFgMpnIy8vD\nYrHg4+NDWFiY05UXu93O6tWruXTpEqNGjQKqrqSUlZURGRnJb37zG9atW8e6desYOnQoS5cu5YUX\nXiA0NJRjx46xbNkyXn31VVasWEHfvn3p1asXn3zyidNjFBQUdN39ERERERH5qbgjhYnRaPxRl3KZ\nzWZWrVrFfffdR9euXQkJCQHg9OnTLF26lGnTphEUFATAhQsXmD9/PgUFBdhsNho3bgxcLgxiY2Md\nRY3NZuPtt9/mu+++w83NjTNnzjgd22QyYbFYSE9Pp1+/fuTl5ZGenk69evUwm83V2tvtdt5//32i\no6N54YUXnMY0GAz06NEDgF69ejF37lysVivp6em88cYbjnY2mw24fMlYfHw8AD179uTdd9+tFvNG\n90dERERE5KfApZdyubu7U1lZCUB5ebnj+yeeeILY2FhSU1OZNm0ar7zyCgaDgeDgYMrLy8nMzKRT\np04ALF++nP79+xMbG8vhw4dZu3atI46X13/um/j4448JDg5mzJgxVFZW8vTTTzvNKSYmhqNHj3Ly\n5EnCwsKoX78+GzdupF69ejzwwAPV2hsMBiIjI8nMzKS4uBg/P79r7rPdbsdgMGC32/Hz87vpe29u\ndH9uhIenB8HBwTfdX36adM7FGc0LcUbzQpzRvJDbzaWFScOGDcnMzKRDhw7s2bPH8X1OTg7Nmzen\nefPmHD9+nOzsbOrVq0e9evUYOXIkr7/+Ot7e3rRu3ZrS0lLHvxjbtm1zxLDb7VXGKi0tpX79+gBs\n377dURD9kMlk4qOPPiI0NBSDwYCfnx8XL14kKyurxhvaO3ToQIcOHZgzZw5Tp07F29u7yna73c6e\nPXvo0aMHO3bsICYmBh8fHxo1asSePXvo1q0bdrudkydPEh4ejtlsZufOnfTs2ZMdO3Y4HfNG9+dG\n2Mpt5Ofn33R/+ekJDg7WOZdqNC/EGc0LcUbzQmqDS+4x6dChA0899RQDBw5kyZIl1KtXj9atWzu2\nb9q0ibS0NNzc3GjevDkdOnTAYrFgMBgIDAxk0qRJzJ49m5EjRzJw4EDmz5+Pr68vbdu25dy5c8Dl\nlYyr7/Po27cv8+bNY/v27XTo0KFa8XBFWFgYRUVF9OzZ0/FdeHg4ZWVlNa6GGAwGunbtSmlpKYmJ\niUyZMqXKdi8vLzIyMnj//fcJCgpi3LhxAIwZM4Zly5bx/vvvU1FRwb333kt4eDjPPvssCxcu5MMP\nP6Rz585V9uPKzze6PyIiIiIiPwUG+w+XFuS2e+aZZ3jnnXdcNv6WLVuYnOr8scZw+XHB7Zv438GM\nxNX0ly5xRvNCnNG8EGc0L8SZ1NRU+vTpc9P9XXop1/+Kmt51ciclx0XVuK2xv/EOZiIiIiIiUp0K\nkzvgyrtRXEkrIiIiIiJSl7m5OgEREREREREVJiIiIiIi4nIqTERERERExOVUmIiIiIiIiMupMBER\nEREREZdTYSIiIiIiIi6nwkRERERERFzuuoXJ4MGDSUhIYMKECbzxxhuUlZXdcPBvv/2W/fv331KC\nd8q3335LQkKC4/OOHTv49a9/TWVlJQAnT54kPj4egGnTpgGQlpbGnDlzqsXau3cvGzZsuC15paSk\nsGfPHgDeeustsrKybirOweyiGv/JKbp0W3IVEREREblZ133BopeXF0lJSQAsXLiQTz/9lH79+l03\ncEVFBd9++y2ZmZl07Njx1jOtZWFhYZw/fx6r1Yq3tzcWi4VmzZqRmZlJVFQU6enpmM1mAGbOnHnN\nWJ07d6Zz5863JS+DweB4c/yLL75403HiN2XUuC05LopQf6+bji0iIiIicqt+1JvfY2JiOHnyJMXF\nxSxevJhz585hNBoZMWIEYWFhrFmzhrNnz5Kbm0uDBg1IT0+nrKyMo0eP8sQTT5CVlYWPjw/9+/cH\nYMKECUyZMoUGDRqwbt06duzYQUBAAPXr1yciIoL+/fvz2muv8cwzzxAREUFhYSFTpkwhJSWFyspK\n3n33XQ4fPozNZqNv3748+OCD5Ofn8+abb1JaWkpFRQXDhw8nJiaGgwcPsnbtWsrLy2ncuDGjRo3C\n29vbsW9ubm5ERkZy7Ngx7r77bk6cOEHfvn2xWCyOwqR9+/YADBkyhNWrV1c5NhkZGfzpT39i/Pjx\nHDlyhMzMTIYOHUpKSgqenp6cOHGCkpISfvOb39CpU6ca87fb7SxfvpxvvvmG+vXr4+Hxn1N09bFY\ntmwZx48fp6ysjK5duzJo0KCbngQiIiIiIq52w4VJRUUFBw4coGPHjqxZs4aIiAgSEhI4dOgQixYt\ncqyqZGdnM2PGDDw9Pdm2bZvjF3SAtWvXOo2dkZHBl19+ydy5c7HZbEyaNInIyEgAx2rBD/3rX//C\n19eX2bNnU15ezquvvkq7du348ssvad++Pb/4xS+orKykrKyMwsJC1q9fz6uvvorRaGTDhg18/PHH\nPPnkk1Vims1m0tPTMZlMGAwGWrduzV/+8hfi4uI4duyY45f/H+aUnp7OihUrSEhIoH79+hw5cqTK\n9gsXLjB79mxycnKYPn06CxcuZPv27U7zP3HiBGfOnGH+/PkUFBQwfvx4fvazn1Ub91e/+hV+fn5U\nVlYyc+ZMTp48SVhY2A2dSxERERGRuua6hUlZWZnj3otWrVrxwAMP8MorrzBhwgQA2rZtS1FREaWl\npRgMBmJjY/H09LzhBOx2O+np6XTp0gUPDw88PDyIjY29br+DBw9y8uRJx/0XpaWl5OTkEBkZyZIl\nS6ioqKBLly60aNGCtLQ0srKymDp1KgA2mw2TyVQtpslk4uOPPyYjI4OoqCgaN25MTk4OhYWFWK1W\nGjVqVK3P6dOnWbp0KdOmTSMoKKjadoPBQPfu3QEIDQ2lUaNGnD59usb8jxw5wn333YfBYCA4OJg2\nbdo43f9du3axZcsWKisryc/PJysrS4WJiIiIiPxkXbcwMRqNjtWQG+HlVfO9Cu7u7tjtdsfn8vJy\n4PIv71d/f/XPbm5ujs9X2l/x/PPP065du2rjzJgxg3379rF48WL69euHr68v7dq1Y+zYsdfMPTo6\nmuPHjztWTQDq16/Prl27iI6OdtonODiY8vJyMjMz6dSp0zXjX3Fl5cNZ/qmpqVX235nc3Fw+/vhj\n5syZQ7169Vi8ePGPeijBD3l4ehAcHHzT/eWnSedcnNG8EGc0L8QZzQu53X7UPSZXxMTE8MUXX/DL\nX/6StLQ0AgIC8PHxqfYLtY+PD6WlpY7PDRs2ZN++fQBkZmaSm5uLwWDAbDazdOlSBgwYQEVFBamp\nqTz00EOOPsePHycyMtKxugDQvn17PvnkE9q0aYO7uzvZ2dnUr1+foqIiQkJC6NOnD+Xl5Zw4cYIB\nAwbw9ttvk5OTQ2hoKFarlfz8fO66665q+YaEhLB161amT58OXF5F+fvf/07fvn2dHot69eoxcuRI\nXn/9dby9vWndunWV7Xa7nd27d9O7d2/H/TdNmzatMf/WrVuzefNmevfuzffff09aWho9e/asErOk\npAQvLy98fHwoKChg//79Na6s3AhbuY38/Pyb7i8/PcHBwTrnUo3mhTijeSHOaF5IbbhuYeLsHo+B\nAweyZMkS4uPj8fLyYvTo0Y62V7dv06YNGzZsICEhgQEDBtC1a1c+//xzJkyYQFRUFE2aNAEgMjKS\nzp07M3HiRIKCgggLC6NevXoAPPbYY8yfP58tW7bQsWNHR/w+ffpw7tw5Jk2ahN1uJzAwkIkTJ5KW\nlsZHH32Eh4cH3t7evPTSSwQEBDBq1CgWLFiAzWYDLt+j8cPCBC4XXXv37iUkJAS4XJj89a9/dTyR\n64fHxGAwEBgYyKRJk5g9ezYjR46sdvwaNGjAb3/7W0pKShg+fDgeHh5O84+Pj+eee+7h0KFDjB8/\nngYNGlQZ94oWLVrQsmVLxo0bR4MGDYiJibneaRQRERERqdMM9utdN3SHXHlM76VLl3jttdcYMWIE\nLVq0cHVat2zx4sXExsbStWtXl+WwZcsWJqc6f4gAXH5ccPsm/ncwI3E1/aVLnNG8EGc0L8QZzQtx\nJjU1lT59+tx0/5u6lKs2LF26lKysLMrLy+ndu/d/RVEiIiIiIiI3ps4UJi+//LKrU6gVo0aNcnUK\nwOVVkZo09jfewUxERERERKqrM4WJ1C5dqiUiIiIidZmbqxMQERERERFRYSIiIiIiIi6nwkRERERE\nRFxOhYmIiIiIiLicChMREREREXE5FSYiIiIiIuJyKkxERERERMTl7lhhUlBQwJtvvsmYMWOYPHky\ns2fP5syZMzW2Lykp4dNPP72h2EOGDLnl/C5evMjzzz/v+GyxWBg8eDB5eXmOfIYOHYrdbmf27NmU\nlJSQm5vLhAkTqsXKzMxkxYoVt5wTwJo1a9i4caPj52+++eam4hzMLrruPzlFl25LziIiIiIiP9Yd\necGi3W4nOTmZBx54gHHjxgHw3Xff8f3333PXXXc57VNcXMwnn3zCz3/+8+vGNxgMt5yjr68vQUFB\nZGVl0axZM9LT02nZsiXp6el0794di8VCdHQ0BoOBKVOmOHJ0JiIigoiIiFvOCaru26BBg246Tvym\njOu2SY6LItTf66bHEBERERG5WXekMElLS8PDw4MHH3zQ8V14eDgAVquV5ORkiouLqaio4Fe/+hWd\nO3fmL3/5C2fPniUhIYF27drx5JNPOm13tZpi5ebmMnv2bGJiYrBYLISEhBAfH4/RaKzS32w2Y7FY\naNasGRaLhbi4uCqFidlsBmD06NEkJiZW6Xv27FneeOMNRowYQWlpKRs3bmTy5MmsWbOGs2fPcvbs\nWYqKinjsscfo06cPAB999BG7d+/GZrPRpUsXR+Gxfv16tm/fTmBgIPXr1ycyMhKAlJQUYmNj6dat\nG+vWrWPfvn2UlZVhNpt54YUXbuMZExERERG5s+5IYXLy5MkaVxCMRiMTJ07Ex8eHwsJCpk6dSufO\nnXn66ac5deoUSUlJAFRWVjptdyOxAHJychg3bhwjRoxg/vz5/Pvf/6Znz55V+pvNZg4fPszPfvYz\ncnNz6d69O5999hkA6enpDBgwwOk+ZGdns2DBAkaPHk1YWBhpaWlVtp86dYpZs2ZhtVpJSEigU6dO\nnDx5kpycHGbPnk1lZSVJSUkcOXIELy8vdu3aRXJyMhUVFUyaNMlRmBgMBscKysMPP8yTTz4JwKJF\ni9i3bx+xsbE3fE5EREREROqSO1KYXOtSq8rKSv7yl79w9OhRDAYD+fn5fP/999jt9htqFxgYeN02\nAI0aNXKs0kRERHDu3LlquZjNZjZs2EBubi4NGzbE09MTu92O1WrlxIkTREdHV+tTWFhIcnIyEydO\npGnTpk73vXPnznh6euLp6UmbNm3IyMjgyJEjHDx4kISEBAAuXbrEmTNnsFqt3HPPPY7VnB8WX1cc\nOnSIjRs3cunSJYqLi2nWrNktFyYenh4EBwffUgz56dC5Fmc0L8QZzQtxRvNCbrc7Upg0b96cPXv2\nON22Y8cOioqKSExMxM3NjdGjR1NeXn5T7a7VxsPjP7vq5uZGWVlZtTFCQ0O5ePEi+/btw2QyAZeL\nmK1bt9KwYUO8vKrff1GvXj0aNGjAkSNHnBYmzlwp1AYMGFDl8jaATZs2Vfn8wwINoKysjLfffpvE\nxERCQkJYu3at02P2Y9nKbeTn599yHKn7goODda6lGs0LcUbzQpzRvJDacEeeytW2bVtsNpvjsii4\nfPP70aNHKSkpISAgADc3Nw4dOsT58+cB8PHxwWq1OtrX1O5qN9LmeqKjo9m0aZOjMDGZTGzatImY\nmBin7T08PJg4cSKff/45O3bsqLbdbrezd+9eysvLKSoq4vDhw0RFRdGhQwe2bt3q2Me8vDwKCwtp\n1aoVX331FWVlZZSWlpKamlot5pUixM/PD6vVyp49e27LAwBERERERFzljqyYAEycOJGVK1fy4Ycf\nYjQaadiwIc8++yw9e/YkMTGRiRMnEhER4Vh18Pf3x2w2M2HCBDp27Mjjjz/utB38ZwWiplhXt6np\n8xVms5kDBw447uuIjo4mNzfXceP7D/saDAa8vLyYPHkyM2fOxMfHBx8fH0cbg8FAWFgY06dPp6io\niCeffJKgoCDHE8CmTp0KXC7ExowZQ8uWLenevTvx8fEEBgY68riar68vffr0YcKECQQFBREVFXXj\nJ0JEREREpA4y2J1dKyS3zdq1a/H29qZ///4uy2HLli1MTr3+ikpyXBTtm/jfgYzE1bQEL85oXogz\nmhfijOaFOJOamup4+uzNuGMrJv/L6sJlVslx119VaexvvG4bEREREZHaoMKklg0cONDVKQBoJURE\nRERE6rQ7cvO7iIiIiIjItagwERERERERl1NhIiIiIiIiLqfCREREREREXE6FiYiIiIiIuJwKExER\nERERcTkVJiIiIiIi4nJ1pjD58ssvGTx4MNnZ2TfVf8iQIU6/X7NmDd98880NxUhOTuarr75yfB47\ndizr1693fJ47dy5ffvklmzdv5vPPPwfgtddeIzMzs1qs2bNnU1JS8mN2wanc3FwmTJgAwPHjx1mx\nYsVNxTmYXXTL/+QUXbrl/RERERERcabOvGBx586ddOrUiR07djBo0KBq2ysqKnB3d6+xf01vV3cW\nqyYxMTGkp6fTpUsXioqK8Pb2xmKxOLYfO3aM4cOHExgYeN1xp0yZcsPj3qjIyEgiIyNvqm/8poxb\nHj85LopQf69bjiMiIiIi8kN1ojCxWq1kZGQwffp0Zs2a5Sgm0tLS+Nvf/oafnx/Z2dm8+eabJCUl\nkZeXR3l5OY888ggPPvigI86qVav4+uuvCQoKYuzYsQQEBJCSkkJsbCzdunUjIyODVatWYbVa8fT0\n5NVXX8Xb29vR32w28+c//xmA9PR0YmNjOXDgAHB55cJoNBIYGMiaNWvw8fGhf//+jr6VlZUsWbKE\nBg0aMHjwYEaPHk1iYiIlJSX8/ve/JyIighMnTtC8eXNeeukljEYjmZmZvPPOO1itVvz9/Rk9ejRB\nQUFkZmayZMkSANq3b+8YIy0tjY0bNzJ58mQyMjJYuXIl5eXlGI1GRo4cSZMmTWrvJImIiIiI1KI6\nUZh89dVXtG/fngYNGhAQEEBmZiYREREAnDhxgjfeeIOGDRsCMGrUKPz8/CgrK2PKlCl069YNPz8/\nLl26RGRkJL/5zW9Yt24d69atY+jQoRgMBgwGAzabjQULFvB///d/REREYLVaMRqNVfJo2bIlp06d\nwmazYbFYaN26Nbm5uWRlZXHixAnMZjOAI+YVFRUVLFy4kPDwcAYMGFBt/86cOcOoUaMwmUwsWbKE\nTz75hLi4OJYvX86kSZPw9/dn165d/PWvf2XkyJEsXryYYcOGERMT4yiUfqhp06bMmDEDNzc3vv76\na/761786LvkSEREREfmpqROFyc6dO3n00UcB6NatGzt37nQUJlFRUY6iBGDTpk2O+0AuXLhATk4O\nUVFRGAwGevToAUCvXr2YO3euo4/dbic7O5ugoCBH3KtXSq7w9PSkWbNmnDhxgmPHjvH444+Tm5uL\nxWLhxIkTxMTEVOtjt9tZunQpPXr0cFqUANSvXx+TyQRAz549+cc//kGHDh04deoUM2fOBC6vuAQH\nB1NSUkJJSYljrF69erF///5qMS9evMiiRYvIycnBYDBQUVFxrUMsIiIiIlKnubwwKS4uJi0tjVOn\nTgGXf0E3GAyOm9m9vP5zT0NaWhqHDh1i1qxZGI1Gpk+fTllZWbWYdru9xns/rsdsNnP48GGsViu+\nvr5ER0fzj3/8g++++46HHnqoWnuDwYDZbObQoUP069cPT09Pp21++Nlut9O8eXNef/31KtsuXrxY\nbV+c+dvf/sbdd99NfHw8586d47XXXvuRe/rjeXh6EBwcXOvjyJ2hcynOaF6IM5oX4ozmhdxuLi9M\n9uzZQ69evRg+fLjju9dee40jR45Ua1taWoqvry9Go5HTp09XuTHdbrezZ88eevTowY4dO6qsbhgM\nBpo0aUJBQQHHjx8nMjKS0tJSvLy8cHOr+mAys9nMO++8Q5s2bQAIDw/n2LFjFBYWEhYWVmW8K372\ns59x+PBh5s+fz8SJE6vFPH/+PBaLBZPJ5MitSZMmFBYWOr632Wzk5OTQrFkzfH19OXr0KDExMXzx\nxRdOj1tpaanjPwhbt2697nG+HWzlNvLz8+/IWFK7goODdS6lGs0LcUbzQpzRvJDa4PLCZOfOnTzx\nxBNVvuvatSs7d+6kR48eVVYbOnTowObNm/m///s/mjRp4rg8Ci6vrGRkZPD+++8TFBTEuHHjqsT0\n8PBg3LhxLF++nLKyMry8vJg6dWq1S7pMJhO5ubmOy7Lc3NwICgqqcjkZVF8F6devHyUlJSxatIgx\nY8ZU2dakSRM++eQTlixZQvPmzfn5z3+Oh4cH48ePZ8WKFZSUlFBZWcmjjz5Ks2bNGDlyJEuWLMFg\nMNCuXbsqY135+bHHHiMlJYX169fTsWPHm14hEhERERGpCwz2mq4VktsiNzeXxMRE5s2b57IctmzZ\nwuTUWy9ckuOiaN/E/zZkJK6mv3SJM5oX4ozmhTijeSHOpKam0qdPn5vuX2desPjfTKsZIiIiIiLX\n5vJLuf7bNWrUqMoTwlwlOS7qlmM09jdev5GIiIiIyE1QYfI/QpdgiYiIiEhdpku5RERERETE5VSY\niIiIiIiIy6kwERERERERl1NhIiIiIiIiLqfCREREREREXE6FiYiIiIiIuJwKExERERERcbk69R6T\nwYMHEx4eTkVFBU2bNuWll17CaHT+Ur9t27aRmZnJ0KFD2bx5M15eXvTq1ctp2zVr1uDj40P//v1r\nHHvTpk3k5uby7LPPArB06VLOnj3LtGnTAPjHP/5BTk4OvXv3Zvv27Tz33HM1xl2zZg2tWrXi7rvv\nvomjUNXo0aNJTEzEz8+PadOmMXPmzJuKczC76JZzaexvJNTf65bjiIiIiIj8UJ0qTLy8vEhKSgJg\n4cKFfPrpp/Tr1++6/R566KFrbjcYDNeNERMTw44dOxyfv/vuO+x2O3a7HYPBgMVioUuXLkRERBAR\nEXHNuIMGDbrueDfjZosSgPhNGbc8fnJclAoTEREREakVdaowuVpMTAwnT56kuLiYxYsXc+7cOYxG\nIyNGjCAsLKxK26tXLjZt2sRnn32Gu7s7zZo1Y+zYsQBkZWUxffp0zp8/T1xcHI888kiVGOHh4Zw5\nc4by8nLKy8sxGo3cddddfPfdd7Ro0QKLxcKQIUNIS0tj48aNTJ48GfhPcfLZZ5/x1VdfMWHCBP70\npz8RGxtLt27dGD16NN27d+fAgQMYjUZefvllQkNDKSws5E9/+hPnz58H4Nlnn8VsNlNUVMSCBQvI\nz88nOjq6So5Dhgxh9erVWK1WkpOTKS4upqKigl/96ld07ty5Vs6DiIiIiMidUCcLk4qKCg4cOEDH\njh1Zs2YNERERJCQkcOjQIRYtWuRYVbnCYDA4CoQPP/yQlJQUPDw8KCkpAcBut5Odnc3vfvc7SktL\nGTduHH379sXN7T+32Li7u9OiRQsyMjK4dOkS0dHRhIaGYrFYCAgIwG63ExISwpkzZ6qMbbfb+ec/\n/8k333xDfHw8Hh4eVfIB8PX1Ze7cuXz++eesXLmSyZMns2LFCh599FFiYmI4f/48s2bNYv78+axd\nu5ZWrVrxy1/+ktTUVLZu3VplPwGMRiMTJ07Ex8eHwsJCpk6dqsJERERERH7S6lRhUlZWRkJCAgCt\nWrXigQce4JVXXmHChAkAtG3blqKiIkpLS2uMER4ezsKFC+nSpQtdunQBLv9C36lTJzw8PPD39ycg\nIICCggJCQkKq9DWbzaSnp1NWVobJZCI0NJQPPviAgIAATCZTtbHsdjuff/459evXJyEhoUqhc7V7\n770XgB49erBq1SoAvvnmG06fPu1oY7VasVqtHD16lIkTJwLQqVMnfH19q8WrrKzkL3/5C0ePahDp\nLwAAIABJREFUHsVgMJCfn8/3339PYGBgjcfldvDw9CA4OLhWx5A7R+dSnNG8EGc0L8QZzQu53epU\nYWI0Gquthjjj7N4Ou90OwOTJkzly5Aj79u1j/fr1zJs3DwAPj//sqpubG5WVldVimM1mPv30U2w2\nGw8//DD+/v5kZWUREBBATEyM0zyaN2/Od999x/nz52nUqNEN76vdbuf3v/99lbxu1I4dOygqKiIx\nMRE3NzdGjx5NeXn5j47zY9nKbeTn59f6OFL7goODdS6lGs0LcUbzQpzRvJDaUOcfFxwTE8MXX3wB\nQFpaGgEBAXh7e1dpc6UosdvtnD9/njZt2vDUU09RUlKC1Wp1bL8ek8nEsWPHKCwsJCAgAIPBQEBA\nAHv37sVsNldrb7fbadmyJcOHDycpKanGf0F37drl+N8rcdq3b8+mTZscbb799lvg8krRlZvw9+/f\nz8WLF6vFKykpISAgADc3Nw4dOuS4T0VERERE5KeqTq2YOFsJGThwIEuWLCE+Ph4vLy9Gjx7ttJ/B\nYKCyspJFixZRUlKC3W4nLi6OevXqVbvnoya+vr4EBgbSvHlzx3cmkwmLxUJ4eHiVsa7+OSYmhiFD\nhjBnzhymTp1aLe7FixeJj4/H09PTcTP+c889x9tvv018fDwVFRW0bt2aYcOGMXDgQBYsWMCECRMw\nmUw0aNCg2vHp2bMniYmJTJw4kYiICJo2bXrdfRMRERERqcsM9htdTpCbcvV7SFxly5YtTE69fmF2\nPclxUbRv4n8bMhJX0xK8OKN5Ic5oXogzmhfiTGpqKn369Lnp/nVqxeS/0Y2s1NwJyXFRtxyjsb/z\nl12KiIiIiNwqFSa1bNGiRa5OAUArHSIiIiJSp9X5m99FREREROS/nwoTERERERFxORUmIiIiIiLi\ncipMRERERETE5VSYiIiIiIiIy6kwERERERERl1NhIiIiIiIiLndHCpP169czYcIE4uPjSUhIICMj\n47aPMW3atGtuHzJkyDW32+12nn/+eUpKSgDIz89n8ODBHD161NFm2LBhFBcX89Zbb3H69Oka4+bl\n5fHGG2/82F1watu2bSxfvhyAzZs38/nnn99UnIPZRbX+T07RpduyzyIiIiLyv6fWX7BosVhITU0l\nMTERDw8PiouLKS8vv+3jzJw585rbr/cGdoPBQHR0NOnp6XTs2JH09HRatGiBxWIhJiaG7Oxs/P39\n8fPz48UXX7xm3JCQEMaPH39zO3INDz300E33jd90+4vBH0qOiyLU36vWxxERERGR/z61XpgUFBTg\n7++Ph8flofz8/BzbRo8eTffu3Tlw4ABGo5GXX36Z0NBQ9u7dywcffIDNZsPPz4+XX36ZwMBA1qxZ\nw/nz5zl37hznz58nLi6ORx55BLi8crF69Wry8/N58803KS0tpaKiguHDhxMTEwPAe++9x759+zAa\njSQkJBAYGFglV7PZ7ChMLBYLjz76KF9++SUA6enpmM1mAF577TWeeeYZIiIiHH0LCwtJSkril7/8\nJU2bNiUxMZF58+axbds2vvzyS0pLS8nLy6Nnz548+eSTAHz++ef885//xGazERUVxbBhw3Bzc2Pr\n1q1s2LABX19fwsPD8fT0BGDNmjX4+PjQv39/PvvsM7Zs2YLNZiM0NJQxY8ZgNBpr4xSKiIiIiNS6\nWr+Uq127dly4cIGxY8eybNkyDh8+XGW7r68vc+fO5eGHH2blypUAtGrVilmzZpGYmEiPHj348MMP\nHe3PnDnDK6+8wu9//3vWrVtHZWUl8J+Vix07dtC+fXuSkpJITk6mRYsWAFy6dAmTyURycjKtWrVi\ny5Yt1XI1m81YLBYAjh8/zj333MOFCxeAqoXJD1dJvv/+exITExk8eDAdO3asFvf48eNMmDCB5ORk\ndu/eTWZmJllZWezevZvXX3+dpKQk3Nzc2LFjB/n5+axdu5bXX3+dGTNmkJWV5Yhz9bjdunVj9uzZ\nJCcn07RpU/71r39d/2SIiIiIiNRRtb5i4u3tzZw5czh69CiHDh3izTff5KmnnuL+++8H4N577wWg\nR48erFq1CoALFy4wf/58CgoKsNlsNG7cGLj8i3mnTp3w8PDA39+fgIAACgoKCAkJcYwXFRXFkiVL\nqKiooEuXLo7CxMPDg06dOgEQERHB119/XS3XyMhITpw4waVLl7DZbHh7e9OoUSNycnI4duwYjz32\nWLU+NpuNGTNmMGzYMFq1auX0GLRr186xUtS1a1eOHj2Km5sbmZmZTJ48GYDy8nKCgoLIyMigdevW\n+Pv7O47LmTNnqsU8efIk7733HiUlJVitVtq3b3/tEyEiIiIiUofVemEC4ObmRuvWrWndujVhYWFs\n377dUZg4s3z5cvr3709sbCyHDx9m7dq1jm1XLgm7EvfKiskVrVq1YsaMGezbt4/FixfTr18/evXq\nhbu7+zX7AXh5eXHXXXexdetWx2Va0dHRpKam8v3339OkSZNqfdzd3YmMjOTAgQM1FiZXs9vtjp97\n9+7NU089VWX7V199VWN7+M+qSUpKCpMmTSIsLIxt27ZVW4lyBQ9PD4KDg12dhtwgnStxRvNCnNG8\nEGc0L+R2q/XCJDs7G4PBwF133QXAt99+S8OGDR3bd+3axRNPPMGuXbscl0qVlpY6Jvu2bdscbX/4\nS7oz58+fJyQkhD59+lBeXs6JEyfo1avXDedrMpn4+9//zqBBgxyf//CHP2AymZy2NxgMjBw5knnz\n5vHhhx/y+OOPV2vz9ddfU1xcjNFoZO/evYwcORKj0UhycjL9+vUjICCA4uJirFYr0dHRrFy5kuLi\nYry9vdmzZ49j1cdutzuOgdVqJSgoCJvNxhdffEH9+vVveB9ri63cRn5+vqvTkBsQHByscyXVaF6I\nM5oX4ozmhdSGWi9MrFYrK1as4OLFi7i7uxMaGsqIESMc2y9evEh8fDyenp6MHTsWgIEDBzJ//nx8\nfX1p27Yt586dAy4XATU9XevK94cOHWLjxo14eHjg7e3NSy+9VGX79cTExPCPf/zDUYi0bNmSvLw8\n+vTpU+O4BoOBcePGkZiYiI+PDx07dqwyXlRUFPPmzXPc/H5lNWbw4MG8/vrr2O123N3dGTZsGFFR\nUQwcOJBXXnkFX19fR1Hyw/0fPHgwv/3tbwkICCAqKgqr1XpD+yciIiIiUhcZ7DeyDFFLRo8eTWJi\nYpUndf232bZtG5mZmQwdOtRlOWzZsoXJqTdWmN2K5Lgo2jfxr/Vx5NbpL13ijOaFOKN5Ic5oXogz\nqampNf4x/0bckXtManKjqxhy65Ljomp9jMb+elyxiIiIiNwclxYmixYtcuXwd8T9999/zRv97xSt\nZIiIiIhIXVbr7zERERERERG5HhUmIiIiIiLicipMRERERETE5VSYiIiIiIiIy6kwERERERERl1Nh\nIiIiIiIiLqfCREREREREXO6OFSaDBw8mISGBiRMnMnfuXKxW622L/dZbb5GVlXXLcRISEvj2228B\nqKioYMiQIXzxxReO7ZMmTeLEiROsWbOGQ4cOAZffXl9cXFwt1rRp0245H4C0tDTmzJkDwN69e9mw\nYcNtiSsiIiIiUpfcsRcsenl5kZSUBEBKSgqbN2+mf//+tyX2iy++eFvixMTEYLFYaNGiBd999x1N\nmjQhPT2dnj17YrVayc3NpUWLFrRs2fK6sWbOnHlbcrpa586d6dy58031PZhddJuz+fEa+xsJ9fdy\ndRoiIiIiUge55M3v0dHRnDx5EoDXXnuNZ555hoiICAoLC5kyZQopKSmcOnWKJUuWYLPZsNvtTJgw\ngaCgIObPn09eXh6VlZU8+eSTdO/evUqMZcuWcfz4ccrKyujatSuDBg0CLq9s9O7dm3379lFRUcH4\n8eNp0qRJlbxMJhP79+/n5z//ORaLhYceeojt27cDkJGRQUREBAaDgZSUFGJjY+nWrZujb1lZGXPn\nzqVbt2787Gc/Y8iQIaxevZq0tDTWrFmDj48POTk5tGnThmHDhmEwGDh48CBr166lvLycxo0bM2rU\nKLy9vTlw4ACrVq3CaDQSExPjGGPbtm1kZmYydOhQ9u7dywcffIDNZsPPz4+XX36ZwMDAGo95/KaM\n23b+blZyXJQKExERERFx6o4XJpWVlXz99dfcfffdABgMBqftNm/eTFxcHPfddx8VFRVUVFSQmppK\nSEgIU6ZMAaCkpKRajF/96lf4+flRWVnJzJkzOXnyJGFhYQAEBASQmJjIp59+ykcffVRtpcVsNvO3\nv/0NgPT0dAYOHMjOnTuxWq1YLBbMZrNjvKvHLC0tZf78+fTu3ZtevXpVyykjI4P58+fToEEDfv/7\n3/Pvf/+b1q1bs379el599VWMRiMbNmzg448/5rHHHuOPf/wjv/vd7wgNDWX+/PlOj1GrVq0cqydb\ntmzhww8/5JlnnrnR0yAiIiIiUqfcscKkrKyMhIQE8vLyaNiwIQ899NA125tMJj744AMuXLhA165d\nCQ0NJTw8nNWrV/Puu+8SGxtbZTXhil27drFlyxYqKyvJz88nKyvLUZh07doVgJYtW/Lvf/+7Wt+G\nDRtis9koKCggOzubJk2aEBkZybFjx7BYLDzyyCNOc01KSuLxxx/nvvvuc7o9KiqKRo0aAXDvvfdy\n9OhRjEYjWVlZTJ06FQCbzYbJZCI7O5tGjRoRGhoKQM+ePfnss8+qxbxw4QLz58+noKAAm83miC8i\nIiIi8lN0xwoTo9FIUlISZWVlzJo1i71793LPPffg5uaG3W4HoLy83NH+vvvuw2QysW/fPmbPns3w\n4cNp27YtSUlJpKam8t5779G2bVuefPJJR5/c3Fw+/vhj5syZQ7169Vi8eHGVmJ6engC4ublRWVnp\nNE+TycTu3bsJCgpyfD569CgZGRmYTCanfWJiYjhw4ECNhcnVKx52ux2DwYDdbqddu3aMHTu2Stsr\nN99fz/Lly+nfvz+xsbEcPnyYtWvX3lA/V/Lw9CA4ONjVacj/T+dCnNG8EGc0L8QZzQu53e74pVxG\no5HnnnuOBQsW0KVLFxo2bMjx48eJjIxkz549jnZnz56lcePGPPLII5w/f56TJ0/StGlTfH196dmz\nJ/Xq1WPr1q1VYpeUlODl5YWPjw8FBQXs37+fNm3a/Kj8zGYzmzZt4oEHHgAuFyarV68mODgYHx8f\np30GDx7M2rVrWbZsGcOGDau2PSMjg9zcXBo0aMDu3bt56KGHiI6O5u233yYnJ4fQ0FCsViv5+fk0\nbdqUc+fOOfZ/x44dTscsLS11/Adh27ZtP2ofXcVWbiM/P9/VaQiX/89E50J+SPNCnNG8EGc0L6Q2\n3LHC5OpVgxYtWhAaGsru3bt57LHHmD9/Plu2bKFjx46Odrt37+aLL77A3d2d4OBgfvGLX5CRkcGf\n//xnDAYD7u7uDB8+vMoYV56YNW7cOBo0aOD0Uq8f5vJDJpOJVatWOVZHgoKCqKysrHG15IrnnnuO\nxYsX8+677/L0009XGSMyMpLly5eTk5ND27ZtueeeewAYNWoUCxYswGazAZfvj7nrrrt44YUXmDNn\nDkajkVatWnH27Nlq4w0cOJD58+fj6+tL27ZtOXfu3DXzExERERGpywz2K9dRSa1IS0tj48aNTJ48\n2WU5bNmyhcmpNRdjd0pyXBTtm/i7Og1Bf+kS5zQvxBnNC3FG80KcSU1NpU+fPjfdX29+r2U/fIKX\niIiIiIhU55L3mPwvad26Na1bt3Z1GiTHRbk6BRr7G12dgoiIiIjUUSpM/kfoEioRERERqct0KZeI\niIiIiLicChMREREREXE5FSYiIiIiIuJyKkxERERERMTlVJiIiIiIiIjLqTARERERERGXU2EiIiIi\nIiIuV6vvMVm/fj07d+7Ezc0Ng8HACy+8QFTU7XvR35AhQ1i9ejV5eXmsXLmS8ePHO22Xm5tLYmIi\n8+bNqzHWxYsXefnll3n77bcBsFgsTJs2jSVLlhASEkJJSQkvvfQSb7/9NnPmzGHs2LEUFxc7jZuZ\nmcn27dt57rnnbnkf16xZg4+PD/3792fNmjW0atWKu++++0fHOZhddMu51LbG/kZC/b1cnYaIiIiI\nuECtFSYWi4XU1FQSExPx8PCguLiY8vLy2zqGwWAAICQkpMai5Eb5+voSFBREVlYWzZo1Iz09nZYt\nW5Kenk737t2xWCxER0djMBiYMmUKAMXFxU5jRUREEBERcUv5XHFlHwEGDRp003HiN2XcjnRqVXJc\nlAoTERERkf9RtVaYFBQU4O/vj4fH5SH8/PwAyMjIYMOGDUycOJGvvvqKBQsWsGrVKioqKpgwYQJ/\n+MMfyMnJYfny5RQWFuLl5cWIESNo0qQJubm5LFiwgEuXLtG5c2fHWFeviJw6dYolS5Zgs9mw2+1M\nmDABNzc3Kisr+eMf/4jFYiEkJIT4+HiMRmOVnM1mMxaLhWbNmmGxWIiLi6tSmJjNZgBGjx5NYmJi\nlb5nz57ljTfeYMSIEZSWlrJx40YmT57MmjVrOHv2LGfPnqWoqIjHHnuMPn36APDRRx+xe/dubDYb\nXbp0cRQe69evZ/v27QQGBlK/fn0iIyMBSElJITY2lm7durFu3Tr27dtHWVkZZrOZF154oRbOooiI\niIjInVFr95i0a9eOCxcuMHbsWJYtW8bhw4cBaNmyJd999x0AR44cISwsjIyMDDIyMoiOjgZg6dKl\nDB06lDlz5vDrX/+aZcuWAbBixQr69u3L3LlzCQ4Odjru5s2biYuLIykpiTlz5hASEgJATk4ODz/8\nMPPmzaNevXr8+9//rtbXbDaTnp4OXC52unfvTmZmJgDp6emYTCanY2ZnZ/PGG28wevRopyslp06d\n4ne/+x2vv/4669atIz8/n4MHD5KTk8Ps2bNJTEwkMzOTI0eOkJmZya5du0hOTmbKlCkcP37cEcdg\nMDhWUB5++GFmz57NvHnzKCsrY9++fdc5IyIiIiIidVetrZh4e3szZ84cjh49yqFDh3jzzTd56qmn\nuP/++2ncuDGnT5/m+PHj9OvXjyNHjlBZWUlMTAxWq5X09HTeeOMNRyybzQZcvjwsPj4egJ49e/Lu\nu+9WG9dkMvHBBx9w4cIFunbtSmhoKACNGjUiPDwcuHyp1blz56r1NZvNbNiwgdzcXBo2bIinpyd2\nux2r1cqJEycchdPVCgsLSU5OZuLEiTRt2rTadoPBQOfOnfH09MTT05M2bdqQkZHBkSNHOHjwIAkJ\nCQBcunSJM2fOYLVaueeeexyrOVevDF3t0KFDbNy4kUuXLlFcXEyzZs2IjY2t+YSIiIiIiNRhtXrz\nu5ubG61bt6Z169aEhYWxfft27r//flq1asX+/fvx8PCgbdu2pKSkYLfbGTJkCJWVlfj5+ZGUlHRT\nY953332YTCb27dvH7NmzGT58OI0aNXJcUnYlr7Kysmp9Q0NDuXjxIvv27XOsjkRERLB161YaNmyI\nl1f1+x/q1atHgwYNOHLkiNPCxJkrqx4DBgzgwQcfrLJt06ZNVT7b7fZq/cvKynj77bdJTEwkJCSE\ntWvX3vb7d1zBw9OjxpUwuf10rMUZzQtxRvNCnNG8kNut1gqT7OxsDAYDd911FwDffvstDRs2BKBV\nq1b84Q9/4P777ycgIIDi4mIKCwtp3rw5cHl1Y8+ePXTr1g273c7JkycJDw/HbDazc+dOevbsyY4d\nO5yOe/bsWRo3bswjjzzC+fPnOXnyJI0bN77hvKOjo9m0aROjR48GLq/AvPfee3Tq1Mlpew8PDyZO\nnMisWbPw9vbmvvvuq7Ldbrezd+9eBgwYgNVq5fDhw/z617/GaDTyt7/9jfvuuw9vb2/y8vLw8PCg\nVatWLF68mCeeeIKKigpSU1N56KGHqsS8UoT4+flhtVrZs2cP3bt3v+F9rKts5Tby8/Ndncb/hODg\nYB1rqUbzQpzRvBBnNC+kNtRaYWK1WlmxYgUXL17E3d2d0NBQRowYAUBUVBSFhYW0atUKgPDwcAoK\nChx9x4wZw7Jly3j//fepqKjg3nvvJTw8nGeffZaFCxfy4Ycf0rlz5ypPrLry8+7du/niiy9wd3cn\nODiYX/ziF1y8eLFK26vb/5DZbObAgQOOG86jo6PJzc113Pj+w74GgwEvLy8mT57MzJkz8fHxwcfH\nx9HGYDAQFhbG9OnTKSoq4sknnyQoKMjxBLCpU6cC4OPjw5gxY2jZsiXdu3cnPj6ewMBARx5X8/X1\npU+fPkyYMIGgoKDb+ghmERERERFXMNidXSskt83atWvx9vamf//+Lsthy5YtTE51XojVJclxUbRv\n4u/qNP4n6C9d4ozmhTijeSHOaF6IM6mpqY6nz96MWr3HRC6raXXmTkqOq/urKo39jddvJCIiIiL/\nlVSY1LKBAwe6OgUArUSIiIiISJ1Wa+8xERERERERuVEqTERERERExOVUmIiIiIiIiMupMBERERER\nEZdTYSIiIiIiIi6nwkRERERERFxOhYmIiIiIiLhcnXqPyeDBgwkPD6eiogI3Nzd69+7No48+et0X\nFK5fv55f/OIX12yTkpJCbGws3bp1q7HNypUradSoEXFxcQDMmjWL+vXr8+KLLwLwzjvvEBISQmho\nKFlZWTzxxBM1xn3rrbfo168fzZo1u5Fdv6YhQ4awevVq8vLyWLlyJePHj//RMQ5mF91yHq7S2N9I\nqL+Xq9MQERERkVpUpwoTLy8vkpKSACgsLGTBggWUlJQwaNCga/bbsGHDdQuTG3n7ekxMDLt37yYu\nLo7KykqKioqwWq2O7RaLhWeffZaoqCg6d+58zbhXipnb4coYISEhN1WUAMRvyrht+dxpyXFRKkxE\nRERE/svVqcLkagEBAYwYMYIpU6YwaNAgtm3bRmZmJkOHDgVgzpw5PPbYY+zfv5+ysjISEhJo3rw5\nY8aMYfv27Xz88ccAhIeH89JLLwFw5MgR/v73v1NQUMDTTz9dbZXDZDKxatUqALKysmjevDkFBQVc\nvHgRo9HI6dOnadmyZbVcrhQO7733Hnl5ebz44ovMmDGDZ555hoiICIYMGcKDDz7I119/TVBQEGPH\njiUgIICcnByWL19OYWEhXl5ejBgxgiZNmpCbm8uCBQu4dOmSowACyM3NJTExkXnz5pGbm0tKSoqj\ncHr++ecxmUy1eEZERERERGpPnS1MABo1akRlZSXff/99jW2efvppPvnkE8dKy6lTp1i/fj2zZs3C\nz8+PixcvAmC32ykoKGDmzJlkZWWRlJRUrTAJCQnB3d2d8+fPY7FYMJlM5OXlYbFY8PHxISwsDHd3\n92o52O12Vq9ezaVLlxg1ahRQdSWlrKyMyMhIfvOb37Bu3TrWrVvH0KFDWbp0KS+88AKhoaEcO3aM\nZcuW8eqrr7JixQr69u1Lr169+OSTT5zud1BQEFOnTsXT05MzZ86wcOFCZs+e/eMOsIiIiIhIHVGn\nC5MrbuQyrP+vvXsPj6q69z/+nskk5EaukKRcAoTcwAs5kggIeGlQMEWrbbG2Fop4pAJCFRIuLR4v\nFEOIkOoR9BRRKtXWUMUSC1YauRSBIkQphJARgoQAIQkJkISEZJL9+4MfIyETBCTsAJ/X8/g8mdlr\nr/3de77uh++stfacsXPnTgYMGICvry8APj4+zj4SEhIA6NKlS4vFTnR0NHa7nfz8fIYPH055eTn5\n+fl4e3sTExPTrL1hGLz//vtERUUxduzYFuO/7bbbALj99tt56aWXqK2tJT8/n/nz5zvbORwO4PSU\nsZSUFAAGDx7MO++806xPh8PB4sWL2b9/P1arlcOHD1/Q9RERERERaYvadGFy5MgRrFYrfn5+WK1W\nGhsbndvq6+td7nO+IsZm++Z0DcNw2SY2Npbdu3dTWFhIeHg4wcHBZGVl4e3tzV133eXyeD179qSg\noICqqipnQdQSwzCwWCwYhoGvr69zpOdiffTRRwQGBjJx4kQaGxt55JFHLqmfq4HN3UZgYKDZYVxz\ndE3FFeWFuKK8EFeUF3K5tdnC5MSJEyxatIhhw4YBp6d1rV69GsMwOHr0KHv2fLOY283NjYaGBtzc\n3LjxxhtJT09n+PDh+Pr6XlCxcLbo6GhWrFhBWFgYFovFOR2sqKioxQXtcXFxxMXFMWfOHGbOnImn\np2eT7YZhsHnzZm677TY2bNhAbGwsXl5ehISEsHnzZvr3749hGBQWFtKtWzdiYmL47LPPGDx4MBs2\nbHB5zJqaGoKDgwFYt25dk6LtWuOod1BRUWF2GNeUwMBAXVNpRnkhrigvxBXlhbSGNlWYnFnEfu7j\nguH0SEbHjh2ZPHkynTt3JiIiwrnfkCFDSE5OJiIigokTJ/KjH/2IZ599FqvVSo8ePVyu+2hpZCU8\nPJzKykoGDx7sfK9bt27U1dW1WOBYLBb69etHTU0NaWlpzJgxo8n2du3asWfPHt5//30CAgJ46qmn\nAJg4cSJvvPEG77//Pg0NDQwcOJBu3boxevRoXnnlFf72t78RHx/vMu6hQ4cyb9481q1bR1xcXLNi\nSERERETkamIxWprTJJfNqFGjePvtt007fnZ2NtNzLnydTluTnhRJn07tzQ7jmqJvusQV5YW4orwQ\nV5QX4kpOTg6JiYmXvL9++f0KuJjF+yIiIiIi16M2NZXrWnXmt1HMlJ4UaXYIlyy0vYfZIYiIiIhI\nK1Nhcp3QVCgRERERacs0lUtEREREREynwkREREREREynwkREREREREynwkREREREREynwkRERERE\nREynwkREREREREynwkREREREREzXJn7HZOTIkSxdupTS0lLy8/MZNGjQeduXlJSQlpbGvHnz2Lt3\nL+vXr+fRRx/9TjF8/vnnrF27lpSUFACWL1/OmjVreOWVVwDYunUrn376KY8//jhvvfUWkydPZu3a\ntRQUFDBmzJgmfa1evZp27dpx++23f6eYAJ577jlGjRpFREQEqamp/PrXv8bb2/ui+9lErFRyAAAg\nAElEQVR+qPI7x9JWhbb3IKx9O7PDEBEREZHvoE0UJhaLBThdcGzYsOFbC5Oz9ezZk549e37nGGJi\nYli0aJHztd1ux9vbmxMnTuDn54fdbicmJobAwEAmT5583r7uvvvu7xzPGWeuDcCMGTMuuZ+UlXsu\nRzhtUnpSpAoTERERkatcmyhMznj33Xc5ePAgU6dO5c477yQhIYFXX32V2tpaAB577DGio6Ob7JOb\nm0tWVhbTp09nz549LFmyhPr6ejw8PBg3bhydOnVi7dq1bN26lbq6Oo4cOUJCQgK/+MUvmvTj5+eH\nl5cXR44cITQ0lIqKCvr160d+fj4JCQnY7XYefvjhJqM1Z8vJyeGDDz5g2rRprFq1Ci8vL+677z6e\ne+45unfvzq5du2hoaGDcuHFERkZSW1vLm2++SVFREQ0NDYwYMYL4+Hjq6upYuHAh+/fvp3PnztTV\n1TmPMWHCBNLS0vD19SU9PZ2jR49SX1/Pvffey5AhQ1rpUxERERERaX1tqjB55JFHWLFiBdOnTweg\nrq6OmTNn4u7uzuHDh3nllVdITU1tcf/OnTvzwgsvYLVa+c9//sOf//xnpkyZAsD+/fuZO3cuNpuN\np556iqSkJIKCgprsHxMTQ35+Pg0NDYSFhREVFcWXX35J37592b9/P5GRkZSXlzc77pYtW/j73//O\nb37zG7y9vbFYLM6RDovFQl1dHXPnziUvL4/XXnuNefPm8cEHH3DTTTcxfvx4qqur+c1vfsNNN93E\n6tWr8fT0JCMjg8LCQqZNm+byXMeNG4evry91dXXMmDGD/v374+vre0nXXURERETEbG2qMDEMo8lr\nh8PB4sWL2b9/P1arlcOHD593/+rqal599VWKi4uxWCw0NDQ4t9144414eXkB0KVLF0pKSlosTBob\nG4mJiSEyMpK//vWv7Nu3j06dOmGzNb9cO3fuZO/evTzzzDN4enq6jGvgwIEA9OrVi5qaGk6ePMl/\n/vMftm3bRlZWlvNcy8rKyMvLIykpCYDw8HDCw8Nd9rly5Uo+//xzAMrLyzl8+DBRUVHnvT7XKpu7\njcDAQLPDuOromokrygtxRXkhrigv5HJrU4XJuT766CMCAwOZOHEijY2NPPLII+dt/95773HTTTeR\nkpJCaWkpzz33nHObu7u782+LxUJjY2Oz/WNiYvj4449pbGxkyJAheHp6Ul9fT25uLjExMS6PGRoa\nSklJCYcOHSIiIuKizi85OZnvfe97zd4/t0A7V25uLjt37mT27Nl4eHjw/PPPU19ff1HHvpY46h1U\nVFSYHcZVJTAwUNdMmlFeiCvKC3FFeSGtoU09LtjLy8u5ngSgpqaGgIAAANatW+eymDhbTU2Ns3pf\ns2bNRR+/c+fOlJeXs3v3bnr06AFAt27dWL16NbGxsS736dixI1OmTOHVV1+lqKjI+f6Z4sIwDDZu\n3AjA7t278fHxwdvbmz59+rBq1Spn+3379gGnR1U2bNgAQGFhIYWFhS7P08fHBw8PDw4ePIjdbr/o\ncxURERERaUvaRGFyZj1Gt27dsFqtpKSksHLlSoYOHcq6detISUnh0KFDTaZKnf20qjN/33///bz7\n7rtMmzaNxsbGJm0uNI6oqCj8/PywWk9fmujoaEpKSposuj+3306dOjFp0iTmz5/PkSNHmrSxWCy4\nu7szbdo03njjDZ544gkAfvzjH+NwOEhOTmbKlClkZmYCcM8991BbW8vTTz9NZmamy1GYuLg4Ghsb\nefrpp3n33XebPRBARERERORqYzG+bd6QfCfPP/88I0eOvOhpXpdTdnY203Murki7mqQnRdKnU3uz\nw7iqaAheXFFeiCvKC3FFeSGu5OTkkJiYeMn7t+k1JnL5pCdFmh1Cqwlt72F2CCIiIiLyHakwaWXP\nPvus2SEAaERBRERERNq0NrHGRERERERErm8qTERERERExHQqTERERERExHQqTERERERExHQqTERE\nRERExHQqTERERERExHQqTERERERExHSm/o7JT3/6U7p16+Z8PXXqVDp06HDZj5Obm0tWVhbTp09v\nsc3XX3/NwoULmTt3LgAbNmzg9ddf5+2338ZqtVJYWMj//u//kp6ezjPPPMOsWbNa7Hfr1q0UFRXx\nwAMPfOfYFyxYQN++fenfvz+vv/46w4cPp0uXLhfdz/ZDld85lmtBaHsPwtq3MzsMERERETmHqYVJ\nu3btnIXAuQzDAMBisVyRWMLDwykrK6O2thZPT0/sdjtdunShoKCAyMhI8vPziYmJAWDWrFnn7Ss+\nPp74+PjLEpfFYnFegyeeeOKS+0lZueeyxHO1S0+KVGEiIiIi0ga1qV9+LykpYfbs2URFRbFv3z5m\nzJjBxo0b2bRpEw6Hg4SEBB566CFKSkpITU0lNjYWu91OUFAQKSkpeHh4UFxczKJFizhx4gRWq5XJ\nkydjsVg4deoU8+fP58CBA/To0YNJkyY1ObbVaqVnz5589dVX3HTTTezbt4+hQ4dit9udhUmfPn0A\nGDlyJEuXLm2y/549e1i0aBGTJ08mLy+PgoICxowZw4IFC3B3d2ffvn2cPHmSX/7yl9xyyy00Njby\nzjvvsGvXLhwOB0OHDmXIkCEYhsGbb77Jjh07CA4Oxmb75iN67rnnGDVqFBEREbzxxhvs3buXuro6\n+vXrx0MPPdT6H5CIiIiISCsxdY1JXV0dU6dOZerUqbz00ktYLBaKi4sZNmwY8+bN4+DBgxQXF5Oa\nmkpaWhoFBQXk5eUBNGnn7e3Nv//9bwBeeeUVhg0bRnp6OrNnzyYwMBDDMNi3bx+jR49m/vz5lJSU\nsHv37mbxxMTEkJ+fz6lTp7BYLPTu3Zv8/HwAvvrqK+eIybmjOPn5+bzxxhtMnTqV0NDQZv0ePXqU\n1NRUZsyYwaJFi6ivr+fTTz/Fx8eH1NRUXnzxRbKzsykpKWHLli0cPnyYjIwMnnzySex2u7Ofs4/7\n8MMPk5qaSnp6Onl5eRQWFn7HT0NERERExDymjph4eHg0mcpVUlJCx44diYyMBGD79u1s376dqVOn\nAnDq1CmKi4sJDg4mJCTEuT4lIiKC0tJSamtrqaioICEhAaDJaENkZCRBQUEAdO/endLSUmJjY5vE\nEx0dzUcffcSePXuIjIwkNDSU4uJiTpw4QW1tLSEhIc3O4eDBg/zhD3/gmWeeISAgoNl2i8XCgAED\nAAgLCyMkJISDBw+yfft2CgsL2bx5MwA1NTUUFxeTl5fHoEGDsFgsBAYGcsMNN7i8dhs3biQ7O5vG\nxkYqKiooKioiPDz8Aq66iIiIiEjb06amcsHpdSdne/DBBxkyZEiT90pKSpoUHVarlfr6+vP2e277\nhoaGZm2ioqLYu3cv+fn5REdHAxAcHMzGjRuJiopy2W9gYCD19fUUFBRwyy23nP/k/r8zIx+PPfYY\nN998c5NtOTk5zvU1LSkpKeGjjz5izpw5eHt7s3DhQurq6i7o2Nc7m7uNwMBAs8NoE3QdxBXlhbii\nvBBXlBdyubW5wuRscXFxvPfeewwaNAhPT0/Ky8ubFBhnMwwDT09PgoKC+Pzzz0lISKC+vv5b/5F/\nNi8vL4KCglizZg3PP/88cHoU5e9//ztDhw51uY+3tzfjxo3jd7/7HZ6envTu3btZXJs2beKOO+7g\nyJEjlJSU0LlzZ/r06cM//vEPbrjhBtzc3Dh06BDBwcH07t2b1atXc8cdd3D8+HFyc3MZPHhwkz5P\nnjxJu3bt8PLy4tixY3zxxRctjqxIU456BxUVFWaHYbrAwEBdB2lGeSGuKC/EFeWFtAZTCxNXT9w6\n+72bb76ZoqIiZs6cCZwuHCZOnOhy3zOvJ06cyB/+8AcyMzOx2Ww8/fTTTZ5sdb5jA8TGxrJ161bn\ntK/o6Gj+/Oc/O9eXnLuvxWLB39+fadOmkZqayrhx45odp0OHDvzmN7/h5MmTPP7449hsNhITEykt\nLWXatGkYhoG/vz8pKSnceuut7Ny5k8mTJ9OhQ4cmxz2je/fu9OjRg6eeeooOHTo0m5ImIiIiInK1\nsRgXM6QgF23hwoX07duXfv36mRZDdnY203OuzGOX27r0pEj6dGpvdhim0zdd4oryQlxRXogrygtx\nJScnh8TExEvev01P5ZLLJz0p0uwQ2oTQ9h5mhyAiIiIiLqgwaWXjx483OwQAjRKIiIiISJtm6u+Y\niIiIiIiIgAoTERERERFpA1SYiIiIiIiI6VSYiIiIiIiI6VSYiIiIiIiI6VSYiIiIiIiI6VSYiIiI\niIiI6Vq9MDl27Bgvv/wyEydOZPr06cycOZMtW7a09mEv2sqVK1myZInz9R/+8AdmzZrlfL1q1Sre\neustCgoKeOuttwDIzMwkKyurWV+ZmZns2LHjssQ1YcIEqqqqAHjmmWcuS58iIiIiIm1Nq/7AomEY\npKenc9ddd/HrX/8agLKyMrZu3XrBfTQ0NODm5tZaITrFxsayYcMG5+v9+/djGAaGYWCxWLDb7SQk\nJBAREUFERAQAFovFZV8PPfRQq8R4dqF0sbYfqryMkUhoew/C2rczOwwRERGRa0arFiY7d+7E3d2d\nIUOGON/r0KEDw4YNA6CxsZF33nmHXbt24XA4GDp0KEOGDCE3N5f33nsPX19fDh48yK9+9Svee+89\nfHx8OHDgAP3796dr166sWrWK+vp6UlJSCA0NZevWrSxfvhyHw4Gvry+TJk3C39+fzMxMysrKKC0t\npaysjKSkJO69994msXbr1o3Dhw9TX19PfX09Hh4efO9732P//v10794du93OyJEjyc3NJSsri+nT\npwPfFCf//Oc/+fzzz5kyZQqLFi2ib9++9O/fnwkTJjBgwAC+/PJLPDw8mDRpEmFhYZw4cYJFixZR\nVlYGwOjRo4mJiaGyspKXX36ZiooKoqKimsQ4cuRIli5dSm1tLenp6VRVVdHQ0MDDDz9MfHz8eT+L\nlJV7vtuHKU2kJ0WqMBERERG5jFq1MDlw4AA9evRocfunn36Kj48Pqamp1NfX8z//8z/cfPPNAOzb\nt4/58+fTsWNHcnNzKSwsJCMjA19fXyZMmEBiYiKpqamsXLmSVatWMXr0aHr16uX8B3p2djZ/+9vf\nGDVqFACHDx/m2WefpaamhqeeeoqhQ4ditX4zk83NzY3u3buzZ88eTp06RVRUFGFhYdjtdvz8/DAM\ng6CgIA4fPtzkHAzD4OOPP2bHjh2kpKRgs9mwWCxNRlN8fHx46aWXWL9+PUuWLGH69Om89dZb/OAH\nPyA2NpaysjJmz55NRkYGy5Yto1evXvz4xz8mJyeHNWvWOPs506eHhwfJycl4eXlx4sQJZs6c+a2F\niYiIiIhIW9aqhcm5U50WL17M7t27sdlspKamsn37dgoLC9m8eTMANTU1FBcX4+bmRmRkJB07dnTu\n27NnTwICAgAICwsjLi4OgPDwcHJzcwE4evQoGRkZHDt2DIfDQWhoqDOOW265BZvNRvv27fHz8+PY\nsWMEBQU1iS8mJob8/Hzq6uqIjo4mLCyM5cuX4+fnR3R0dLPzMwyD9evXExwczNSpU5sUOmcbOHAg\nALfddht//OMfAdixYwcHDx50tqmtraW2tpbdu3eTnJwMwC233IKPj0+z/hobG3n33XfZvXs3FouF\niooKjh8/jr+/v8vji4iIiIi0da1amHTt2pV///vfztePPfYYlZWVzmlQZ947M0pyRm5uLu3aNZ0m\nY7N9E6rVanW+tlgsNDQ0APDmm29y33330bdvX3bt2sWyZcta3L+xsbFZvDExMXzyySc4HA6GDRtG\n+/btKSoqws/Pj9jY2GbtLRYLXbt2Zf/+/ZSVlRESEnJB1wVOFzUvvvhik7gu1IYNG6isrCQtLQ2r\n1cqECROor6+/6H7k0tncbQQGBpodxndytccvrUN5Ia4oL8QV5YVcbq1amNx44438+c9/5pNPPuGe\ne+4B4NSpU87tffr04R//+Ac33HADbm5uHDp0iODg4Es+Xk1NjfN/krVr1zrfNwzjgvaPjo5mwYIF\nBAcH4+fnB4Cfnx9bt25l8uTJzdobhkGPHj245557mDt3Lr/97W9d/k+6ceNGHnjgATZu3EhMTAxw\n+txXrlzJ/fffD8DXX39N9+7d6dWrFxs2bOBHP/oRX3zxBdXV1c36O3nyJH5+flitVnbu3OlcpyJX\njqPeQUVFhdlhXLLAwMCrOn5pHcoLcUV5Ia4oL6Q1tGphApCSksKSJUtYsWIFfn5+tGvXjl/84hcA\nJCYmUlpayrRp0zAMA39/f5KTk5ut0Tj39dnO3jZixAgyMjLw8fHhxhtvpLS09Fv3P5uPjw/+/v50\n7drV+V50dDR2u51u3bo16+vM37GxsYwcOZI5c+Ywc+bMZv1WV1eTkpKCu7u78+lkjz76KIsXLyYl\nJYWGhgZ69+7Nf//3fzNixAhefvllpkyZQnR0NB06dGhyrgCDBw8mLS2N5ORkIiIi6Ny587eem4iI\niIhIW2YxLnQ4QS7JhAkTSEtLw9fX17QYsrOzmZ7z7YWZXLj0pEj6dGpvdhiXTN90iSvKC3FFeSGu\nKC/ElZycHBITEy95f/3yeyu7kJEaEREREZHrXatP5brevfrqq2aHAJz+hl8un9D2HmaHICIiInJN\nUWFynbiapx2JiIiIyLVPU7lERERERMR0KkxERERERMR0KkxERERERMR0KkxERERERMR0KkxERERE\nRMR0KkxERERERMR0KkxERERERMR0F/w7JkuWLCEkJISkpCQAZs+eTXBwME888QQAb7/9NkFBQYSF\nhVFUVMQDDzzAggUL6Nu3L/3792/S1+uvv87w4cPp0qXLZTwVmhzvUo6xYsUK1qxZg7u7O25ubtx7\n773cfvvtlzVGs2w/VGl2CHIFhLb3IKx9O7PDEBEREbloF1yYxMbGsmnTJpKSkmhsbKSyspLa2lrn\ndrvdzujRo4mMjCQ+Ph4Ai8Xisq8zxczlZrFYnMe82GN88skn7Ny5k9TUVDw9PampqWHLli2tEaYp\nUlbuMTsEuQLSkyJVmIiIiMhV6YILk+joaP74xz8CUFRURNeuXTl27BjV1dV4eHhw8OBBevTowdq1\naykoKGDMmDHAN8XJX/7yF8rLy3niiSd44YUXGDVqFBEREYwcOZKhQ4fyxRdfEBAQwE9/+lPeeecd\nysvL+eUvf0l8fDyNjY2888477Nq1C4fDwdChQxkyZAiGYfDmm2+yY8cOgoODsdm+OZ3nnnvOeYw3\n3niDvXv3UldXR79+/XjooYeand+HH37Ic889h6enJwBeXl7ccccdAOzYsYM//elPNDQ00LNnTx5/\n/HFsNhsTJkxg4MCBfPnll1itVsaOHcu7777LkSNHuP/++7n77rvJzc0lMzMTHx8fDhw4QP/+/ena\ntSurVq2ivr6elJQUQkNDKSkp4bXXXqOqqgo/Pz/GjRtHhw4dWLBgAd7e3hQUFHDs2DEeeeQR+vfv\nT0VFBb///e+pqamhoaGBxx9/nNjY2EtMAxERERERc13wGpOgoCDc3NwoKyvDbrcTHR1NZGQkdrud\nvXv3Eh4ejpubW7P9DMNg6dKlVFVVMX78eKxWa5ORlLq6Om688UbmzZuHl5cXmZmZPPvssyQnJ5OZ\nmQnAp59+io+PD6mpqbz44otkZ2dTUlLCli1bOHz4MBkZGTz55JPY7XZnv2cf4+GHHyY1NZX09HTy\n8vIoLCxsEuPJkyepqakhJCSkWfx1dXUsXLiQp59+mpdeeomGhgY++eQT5/aOHTsyd+5cevXqxcKF\nC0lOTmb27NnO2AEKCwsZO3YsGRkZrF+/nuLiYlJTU/n+97/PqlWrAHjzzTe56667SE9PZ9CgQbz1\n1lvO/Y8dO8asWbOYNm0a7777LgAbNmygT58+zJ07l5deeonu3btf0OcoIiIiItIWXfCICZweNbHb\n7eTn5zN8+HDKy8vJz8/H29ubmJiYZu0Nw+D9998nKiqKsWPHug7AZiMuLg6A8PBw3N3dsVqtdO3a\nldLSUgC2b99OYWEhmzdvBqCmpobi4mLy8vIYNGgQFouFwMBAbrjhBpfH2LhxI9nZ2TQ2NlJRUUFR\nURHh4eEXdM6HDh0iJCSEsLAwAO68804+/vhj51qbM9PWwsPDOXXqFJ6ennh6euLu7s7JkycB6Nmz\nJwEBAQCEhYU1Od/c3FwAvvrqK6ZOnQrA4MGD+dOf/gScLrASEhIA6NKlC8ePHwcgMjKS1157jYaG\nBhISElSYiIiIiMhV7aIKk9jYWHbv3k1hYSHh4eEEBweTlZWFt7c3d911V7P2FouFnj17UlBQQFVV\nFb6+vs3anD3KYrFYnNOxrFYrDQ0Nzm2PPfYYN998c5N9c3JyMAzjvDGXlJTw0UcfMWfOHLy9vVm4\ncCF1dXVN2nh7e+Pp6UlJSUmzUZNz18kYhtHkPXd392axn3nd2NgI0OR9q9XqfG2xWJqcY0vncvb+\nZ9r06tWLF154gW3btrFw4UKGDx9+zSzUl0tnc7cRGBh4QW0vtJ1cX5QX4oryQlxRXsjldtEjJitW\nrCAsLAyLxYKvry/V1dUUFRW1uNg8Li6OuLg45syZw8yZM51rOC5Gnz59+Mc//sENN9yAm5sbhw4d\nIjg4mN69e7N69WruuOMOjh8/Tm5uLoMHD26y78mTJ2nXrh1eXl4cO3aML774wuXIyoMPPsjixYt5\n6qmn8PLyora2li1btjBgwABKS0spLi4mLCyM9evX07t374s+h28TExPDZ599xu23386GDRvo1avX\neduXlZURFBREYmIi9fX17Nu3T4WJ4Kh3UFFR8a3tAgMDL6idXF+UF+KK8kJcUV5Ia7iowiQ8PJzK\nysom//jv1q0bdXV1LkdD4PSoQL9+/aipqSEtLY0ZM2Y0297S6zN/JyYmUlpayrRp0zAMA39/f1JS\nUrj11lvZuXMnkydPpkOHDi6nk3Xv3p0ePXrw1FNP0aFDhxYXiN9zzz3U1tYyY8YM3NzcsNls3Hff\nfbi7uzN+/HgyMjJoaGggMjKSu+++22Xsrs7j7CeFuWpzZtuYMWNYuHAhWVlZ+Pn5MX78+PNek507\nd5KVlYXNZsPT05Mnn3yyxVhERERERNo6i/Ftc6Hkqpednc30nJaLKLl2pCdF0qdT+29tp2+6xBXl\nhbiivBBXlBfiSk5ODomJiZe8/0WNmMjVKz0p0uwQ5AoIbe9hdggiIiIil0SFyXXiQr5FFxEREREx\nywX/jomIiIiIiEhrUWEiIiIiIiKmU2EiIiIiIiKmU2EiIiIiIiKmU2EiIiIiIiKmU2EiIiIiIiKm\nU2EiIiIiIiKmu+K/Y/LTn/6Ubt26OV8PHDiQH/7wh5fU18iRI1m6dCnl5eUsWbKEyZMnu2xXUlJC\nWloa8+bNa7Gv6upqJk2axOLFiwGw2+0888wzvPbaawQFBXHy5EmefPJJFi9ezJw5c/j1r39NVVWV\ny34LCgpYt24djz766CWd19kyMzPx8vLivvvuIzMzk169enHTTTdddD/bD1V+51jk2mErrcFR7zA7\njCsutL0HYe3bmR2GiIiIuHDFC5N27doxd+7cy9KXxWIBICgoqMWi5EL5+PgQEBBAUVERXbp0IT8/\nnx49epCfn8+AAQOw2+1ERUVhsViYMWMGAFVVVS77ioiIICIi4jvFc8aZcwR46KGHLrmflJV7Lkc4\nIle19KRIFSYiIiJtVJv55fcJEyZwxx13sG3bNhoaGpg8eTKdOnXixIkTvPzyyxw7doyoqCh27NhB\nWloavr6+zn3PHhE5cOAAr732Gg6HA8MwmDJlClarlcbGRv7v//4Pu91OUFAQKSkpeHh4NIkhJiYG\nu91Oly5dsNvtJCUlNSlMYmJinLGmpaU12ffIkSPMnz+fX/3qV9TU1JCVlcX06dPJzMzkyJEjHDly\nhMrKSu6//34SExMBWLFiBZs2bcLhcJCQkOAsPD744APWrVuHv78/wcHB9OzZE4AFCxbQt29f+vfv\nz1//+le2bdtGXV0dMTExjB07ttU+GxERERGR1nbF15jU1dUxdepU53+bNm1ybvPz8yMtLY177rmH\nFStWALBs2TJuuukm5s2bR//+/SkrKztv/6tXryYpKYm5c+cyZ84cgoKCACguLmbYsGHMmzcPb29v\n/v3vfzfbNyYmhvz8fOB0sTNgwAAKCgoAyM/PJzo62uUxDx06xPz585kwYYLLkZIDBw7w7LPP8rvf\n/Y6//vWvVFRUsH37doqLi0lNTSUtLY2CggLy8vIoKChg48aNpKenM2PGDPbu3evsx2KxOEdQhg0b\nRmpqKvPmzaOuro5t27ad97qIiIiIiLRlV3zExMPDo8WpXP369QOgR48ezsIhPz+flJQUAOLi4vDx\n8Tlv/9HR0SxfvpyjR4/Sr18/wsLCAAgJCXGubYmIiKC0tLTZvjExMXz44YeUlJTQsWNH3N3dMQyD\n2tpa9u3bR1RUVLN9Tpw4QXp6OsnJyXTu3LnZdovFQnx8PO7u7ri7u3PDDTewZ88e8vLy2L59O1On\nTgXg1KlTHD58mNraWm699VbnaE58fLzL89y5cydZWVmcOnWKqqoqunTpQt++fc97bURERERE2qo2\nM5ULwN3dHcA59eoMwzAuuI9BgwYRHR3Ntm3bSE1N5fHHHyckJASb7ZtTtVqt1NXVNds3LCyM6upq\ntm3b5hwdiYiIYM2aNXTs2JF27ZrPTff29qZDhw7k5eW5LExcOTPq8eCDDzJkyJAm21auXNnktatz\nr6urY/HixaSlpREUFMSyZcuor6+/oGOLXM9s7jYCAwPNDqNN0/URV5QX4oryQi63NlWYuBITE8Om\nTZv44Q9/yPbt26murj5v+yNHjhAaGsq9995LWVkZhYWFhIaGXvDxoqKiWLlyJRMmTABOj8D85S9/\n4ZZbbnHZ3mazkZyczOzZs/H09GTQoEFNthuGwdatW3nwwQepra1l165d/OIXvzpVyx8AAAsmSURB\nVMDDw4P33nuPQYMG4enpSXl5OTabjV69erFw4UIeeOABGhoayMnJ4e67727S55kixNfXl9raWjZv\n3syAAQMu+BxFrleOegcVFRVmh9FmBQYG6vpIM8oLcUV5Ia3hihcmZ9aYnBEXF8fPf/7zJm3OfhLV\niBEjePnll1m/fj3R0dEEBATg6enZrN2Zvzdt2sS//vUv3NzcCAwM5Ec/+hHV1dVN2p6779liYmL4\n8ssvnQvOo6KiKCkpcS58d3Xcdu3aMX36dGbNmoWXlxdeXl7ONhaLhfDwcJ5//nkqKyv5yU9+QkBA\ngPMJYDNnzgTAy8uLiRMn0qNHDwYMGEBKSgr+/v7OOM7m4+NDYmIiU6ZMISAggMjIyJYut4iIiIjI\nVcFiXMw8KRM4HA6sVitWqxW73e6cwnS1WLZsGZ6entx3332mxZCdnc30HNeFmMj1JD0pkj6d2psd\nRpulb0DFFeWFuKK8EFdycnKcT5+9FG1+KldZWRkZGRk0NjZis9n41a9+ZXZIF62l0RkRERERETmt\nzRcmYWFhV9UIyblGjBhhdgjA6W+KRc6wuduu219+FxERkbapzRcmcnlo+oqcTUPwIiIi0tZc8R9Y\nFBEREREROZcKExERERERMZ0KExERERERMZ0KExERERERMZ0KExERERERMZ0KExERERERMZ0KExER\nERERMZ1+x+RbLFmyhJCQEJKSkgCYPXs2wcHBPPHEEwC8/fbbeHt74+7uzg9/+EMWLFhA37596d+/\nf5N+Xn/9dYYPH06XLl2avL927VoKCgoYM2ZMizHs2rULm81GdHT0JZ/H9kOVl7yvXHtspTXX5Q8s\nyvkpL8QV5YW4ciXzIrS9B2Ht212RY4m5VJh8i9jYWDZt2kRSUhKNjY1UVlZSW1vr3G632xk9ejSR\nkad/Wd1isbjs50whcyl27tyJl5fXdypMUlbuueR9RURERMySnhSpwuQ6ocLkW0RHR/PHP/4RgKKi\nIrp27cqxY8eorq7Gw8ODgwcP8vXXX7N+/XrnqMeZ4uQvf/kL5eXlPPHEE7zwwguMGjWKiIgI1qxZ\nw4cffoiPjw/dunXD3d0dgK1bt7J8+XIcDge+vr5MmjSJU6dO8c9//hOr1cq//vUvxowZQ3V1NR98\n8EGTdv7+/uZcIBERERGRy0CFybcICgrCzc2NsrIy7HY70dHRlJeXY7fb8fLyIjw8HJut6WU0DIOl\nS5dy6tQpxo8fD3xTrFRUVLBs2TLS0tLw8vLi+eefp0ePHgD06tWL+Ph4ALKzs/nb3/7GqFGjuPvu\nu/Hy8mL48OEAVFdXM3v27GbtRERERESuVipMLkB0dDR2u538/HyGDx9OeXk5+fn5eHt7ExMT06St\nYRi8//77REVFMXbs2GZ9ffXVV/Tu3Zv27dsDcNttt3H48GEAjh49SkZGBseOHcPhcBAaGtqk3zPO\nbRcSEtIapy0iIiJiOpu7jcDAQLPDkCtAhckFiI2NZffu3RQWFhIeHk5wcDBZWVl4e3tz5513UlVV\n5WxrsVjo2bMnBQUFVFVV4evr26Svc9egnF1wvPnmm9x333307duXXbt2sWzZMpfxXGg7ERERkaud\no95BRUWF2WHIFaDHBV+A6OhocnJyaN++PRaLBV9fX6qrq7Hb7cTGxjZrHxcXxwMPPMCcOXOaLJQH\niIqKIi8vj6qqKhwOB5s3b3Zuq6mpcX4jsHbtWuf7Xl5e1NTUfGs7EREREZGrlUZMLkB4eDiVlZUM\nHjzY+V63bt2oq6trNiICp0dF+vXrR01NDWlpacyYMcO5LSAggBEjRvDb3/4WHx8funfv7tw2YsQI\nMjIy8PHx4cYbb6S0tBSAvn37Mn/+fLZu3cqYMWNabCciIiIicrWyGGfPJZJrUnZ2NtNzXD/GWERE\nRKQtS0+KpE+n9maHIRcgJyeHxMTES95fIybXifSkSLNDkDbE5m7TD6ZJM8oLcUV5Ia5cybwIbe9x\nRY4j5lNhcp3QNw1ytsDAQC0klGaUF+KK8kJcUV5Ia9DidxERERERMZ0KExERERERMZ0KExERERER\nMZ0KExERERERMZ0KExERERERMZ0KExERERERMZ0KExERERERMZ0KExERERERMZ0KExERERERMZ0K\nExERERERMZ0KExERERERMZ0KExERERERMZ3N7ADkysjJyTE7BBERERGRFlkMwzDMDkJERERERK5v\nmsolIiIiIiKmU2EiIiIiIiKmU2EiIiIiIiKmU2EiIiIiIiKm01O5rmFffvklS5YsobGxke9///s8\n8MADZockJpkwYQJeXl5YrVbc3NxITU2lqqqKjIwMysrK6NixI08//TQ+Pj5mhyqtaOHChXzxxRf4\n+fkxb948gPPmwfLly1mzZg1Wq5VHH32UPn36mBm+tBJXeZGZmcmnn36Kn58fAD/72c/4r//6L0B5\ncT0oKytjwYIFHD9+HIvFQmJiIklJSbpfXOdayovLer8w5JrU0NBgPPnkk8aRI0eM+vp6Izk52Thw\n4IDZYYlJxo8fb1RWVjZ5b+nSpcaHH35oGIZhLF++3PjTn/5kRmhyBe3atcsoKCgwJk+e7HyvpTw4\ncOCAkZycbNTX1xtHjhwxnnzySaOhocGUuKV1ucqLzMxMIysrq1lb5cX1oaKiwti3b59hGIZRU1Nj\nTJo0yThw4IDuF9e5lvLict4vNJXrGrVnzx7CwsIICQnBZrMxcOBAtm7danZYYiLjnCeDb926lTvu\nuAOAO++8k88//9yMsOQK6tWrV7NRsZby4PPPP2fgwIHYbDZCQkIICwtjz549VzxmaX2u8gKa3zNA\neXG9CAgIoHv37gB4enrSuXNnysvLdb+4zrWUF3D57hcqTK5R5eXlBAcHO18HBQU5k0euPxaLhVmz\nZjF9+nT++c9/AnD8+HECAgIA8Pf35/jx42aGKCZpKQ8qKiqa3EOCg4N1D7nOfPzxx6SkpPDaa69R\nXV0NKC+uRyUlJXz99ddERUXpfiFOZ/IiOjoauHz3C60xEbkOzJo1i8DAQE6cOMGsWbPo3Llzk+0W\ni8WkyKQt+bY8UJ5cP+655x5+8pOfAPDee+/x9ttvM27cOJdtlRfXrtraWubNm8fo0aPx8vJqsk33\ni+tXbW0t8+fPZ/To0Xh6el7W+4VGTK5RQUFBHD161Pn66NGjBAUFmRiRmCkwMBAAPz8/br31Vvbs\n2YO/vz/Hjh0DTn+r4e/vb2aIYpKW8kD3kOubv78/FosFi8XC97//fef0C+XF9cPhcDBv3jxuv/12\nbr31VkD3C/kmLwYPHtwkLy7X/UKFyTWqZ8+eFBcXU1JSgsPhYOPGjcTHx5sdlpjg1KlT1NTUAKe/\n5fjPf/5DeHg48fHxrF27FoB169aRkJBgYpRilpbyID4+ns8++wyHw0FJSQnFxcVERkaaGKlcSRUV\nFc6/t2zZQnh4OKC8uF4YhsHrr79O586d+cEPfuB8X/eL61tLeXE57xcWw9VqFbkmfPHFF00eF/zg\ngw+aHZKYoKSkhPT0dAAaGxsZNGgQDz74oB4XfB36/e9/T15eHidOnCAgIICHHnqIhISEFvPggw8+\nYM2aNbi5uTF69Gji4uJMPgNpDefmxYgRI9i1axdff/01FouFjh07MnbsWOfaAuXFtW/37t08++yz\nhIeHO6fe/PznPycyMlL3i+uYq7z42c9+xmeffXbZ7hcqTERERERExHSayiUiIiIiIqZTYSIiIiIi\nIqZTYSIiIiIiIqZTYSIiIiIiIqZTYSIiIiIiIqZTYSIiIiIiIqZTYSIiIiIiIqZTYSIiIiIiIqb7\nf8I0mts635bUAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x7f46e0375390>"
       ]
      }
     ],
     "prompt_number": 45
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Checking and filtering bot activity\n",
      "\n",
      "Wikidata is very busy -- is it all bots?"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df[df.wikipediaShort=='wd'].robot.value_counts()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 46,
       "text": [
        "True     188\n",
        "False     22\n",
        "dtype: int64"
       ]
      }
     ],
     "prompt_number": 46
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Time series analysis\n",
      "\n",
      "To make time series analysis easier, let's turn the raw edits into counts per second.\n",
      "\n",
      "We can use a bit of a trick here. Pivot the dataframe, keeping the timestamp as the index, making the wiki short names the new columns, and the value of the `delta` field for the cells. **What is delta?** That doesn't actually matter! We'll see why, shortly.\n",
      "\n",
      "### Usage note\n",
      "\n",
      "The following cell originally used\n",
      "\n",
      "```\n",
      "pivot(index='timestamp',\n",
      "      columns='wikipediaShort',\n",
      "      values='delta')\n",
      "```\n",
      "\n",
      "but this doesn't allow duplicates in the field to be used as the index. So if you're unlucky enough to get two edits in the same microsecond, it'll fail.\n",
      "\n",
      "`pivot_table`, however, handles these properly."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "ts = pd.pivot_table(df, values='delta', index='timestamp', columns='wikipediaShort')\n",
      "ts.head()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th>wikipediaShort</th>\n",
        "      <th>ar</th>\n",
        "      <th>ca</th>\n",
        "      <th>co</th>\n",
        "      <th>cs</th>\n",
        "      <th>de</th>\n",
        "      <th>el</th>\n",
        "      <th>en</th>\n",
        "      <th>es</th>\n",
        "      <th>eu</th>\n",
        "      <th>fa</th>\n",
        "      <th>...</th>\n",
        "      <th>ja</th>\n",
        "      <th>ko</th>\n",
        "      <th>nl</th>\n",
        "      <th>no</th>\n",
        "      <th>pl</th>\n",
        "      <th>pt</th>\n",
        "      <th>ru</th>\n",
        "      <th>sv</th>\n",
        "      <th>wd</th>\n",
        "      <th>zh</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>timestamp</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>2014-08-31 18:32:09.978374</th>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> -5</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>...</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2014-08-31 18:32:09.978403</th>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>...</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> 12</td>\n",
        "      <td>NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2014-08-31 18:32:09.978426</th>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>...</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> 346</td>\n",
        "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2014-08-31 18:32:09.978446</th>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>...</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>-232</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2014-08-31 18:32:09.978467</th>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>...</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> 405</td>\n",
        "      <td>NaN</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>5 rows \u00d7 24 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 47,
       "text": [
        "wikipediaShort              ar  ca  co  cs  de  el  en  es  eu  fa ...  ja  \\\n",
        "timestamp                                                          ...       \n",
        "2014-08-31 18:32:09.978374 NaN NaN NaN NaN NaN NaN  -5 NaN NaN NaN ... NaN   \n",
        "2014-08-31 18:32:09.978403 NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN ... NaN   \n",
        "2014-08-31 18:32:09.978426 NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN ... NaN   \n",
        "2014-08-31 18:32:09.978446 NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN ... NaN   \n",
        "2014-08-31 18:32:09.978467 NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN ... NaN   \n",
        "\n",
        "wikipediaShort              ko   nl  no  pl  pt  ru  sv   wd  zh  \n",
        "timestamp                                                         \n",
        "2014-08-31 18:32:09.978374 NaN  NaN NaN NaN NaN NaN NaN  NaN NaN  \n",
        "2014-08-31 18:32:09.978403 NaN  NaN NaN NaN NaN  12 NaN  NaN NaN  \n",
        "2014-08-31 18:32:09.978426 NaN  NaN NaN NaN NaN NaN NaN  346 NaN  \n",
        "2014-08-31 18:32:09.978446 NaN -232 NaN NaN NaN NaN NaN  NaN NaN  \n",
        "2014-08-31 18:32:09.978467 NaN  NaN NaN NaN NaN NaN NaN  405 NaN  \n",
        "\n",
        "[5 rows x 24 columns]"
       ]
      }
     ],
     "prompt_number": 47
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Then, by resampling the time series to second resolution, and applying the `count` aggregate, we get edits-per-second.\n",
      "\n",
      "We used the `delta` field as the value to be placed into each cell, not because we care about delta, but because the `resample` method only works on numerics. "
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "sec_counts = ts.resample('s', how='count')\n",
      "sec_counts.head()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th>wikipediaShort</th>\n",
        "      <th>ar</th>\n",
        "      <th>ca</th>\n",
        "      <th>co</th>\n",
        "      <th>cs</th>\n",
        "      <th>de</th>\n",
        "      <th>el</th>\n",
        "      <th>en</th>\n",
        "      <th>es</th>\n",
        "      <th>eu</th>\n",
        "      <th>fa</th>\n",
        "      <th>...</th>\n",
        "      <th>ja</th>\n",
        "      <th>ko</th>\n",
        "      <th>nl</th>\n",
        "      <th>no</th>\n",
        "      <th>pl</th>\n",
        "      <th>pt</th>\n",
        "      <th>ru</th>\n",
        "      <th>sv</th>\n",
        "      <th>wd</th>\n",
        "      <th>zh</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>timestamp</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>2014-08-31 18:32:09</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> 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> 1</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0</td>\n",
        "      <td> 1</td>\n",
        "      <td> 1</td>\n",
        "      <td>  6</td>\n",
        "      <td> 0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2014-08-31 18:32:10</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>2014-08-31 18:32:11</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>2014-08-31 18:32:12</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> 3</td>\n",
        "      <td> 1</td>\n",
        "      <td> 1</td>\n",
        "      <td> 1</td>\n",
        "      <td>...</td>\n",
        "      <td> 1</td>\n",
        "      <td> 0</td>\n",
        "      <td> 1</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0</td>\n",
        "      <td>  9</td>\n",
        "      <td> 0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2014-08-31 18:32:13</th>\n",
        "      <td> 2</td>\n",
        "      <td> 0</td>\n",
        "      <td> 2</td>\n",
        "      <td> 0</td>\n",
        "      <td> 1</td>\n",
        "      <td> 0</td>\n",
        "      <td> 4</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> 2</td>\n",
        "      <td> 2</td>\n",
        "      <td> 11</td>\n",
        "      <td> 0</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>5 rows \u00d7 24 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 48,
       "text": [
        "wikipediaShort       ar  ca  co  cs  de  el  en  es  eu  fa ...  ja  ko  nl  \\\n",
        "timestamp                                                   ...               \n",
        "2014-08-31 18:32:09   0   0   0   0   0   0   1   0   0   0 ...   0   0   1   \n",
        "2014-08-31 18:32:10   0   0   0   0   0   0   0   0   0   0 ...   0   0   0   \n",
        "2014-08-31 18:32:11   0   0   0   0   0   0   0   0   0   0 ...   0   0   0   \n",
        "2014-08-31 18:32:12   0   0   0   0   0   0   3   1   1   1 ...   1   0   1   \n",
        "2014-08-31 18:32:13   2   0   2   0   1   0   4   0   0   0 ...   0   0   0   \n",
        "\n",
        "wikipediaShort       no  pl  pt  ru  sv  wd  zh  \n",
        "timestamp                                        \n",
        "2014-08-31 18:32:09   0   0   0   1   1   6   0  \n",
        "2014-08-31 18:32:10   0   0   0   0   0   0   0  \n",
        "2014-08-31 18:32:11   0   0   0   0   0   0   0  \n",
        "2014-08-31 18:32:12   0   0   0   0   0   9   0  \n",
        "2014-08-31 18:32:13   0   0   0   2   2  11   0  \n",
        "\n",
        "[5 rows x 24 columns]"
       ]
      }
     ],
     "prompt_number": 48
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Visualizing activity over time\n",
      "\n",
      "This is easy, now we have the data in the right format."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "p = sec_counts.plot()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAr0AAAHzCAYAAAAtnqVZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\npZRSSqmAp4NepZRSSikV8P4Ph93UH58XSJgAAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x7f46e020d710>"
       ]
      }
     ],
     "prompt_number": 49
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Not so easy to read! (Even without taking into account the terrible default legend placement.)\n",
      "\n",
      "What if we make it a cumulative sum instead, so we can see how activity 'mounts up' over time?\n",
      "\n",
      "That's much nicer. Although you'd really want to select just the most active or interesting wikis and show those, because with 20+ lines, the colours become hard to tell apart. This is left as an exercise :-)"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "p = sec_counts.cumsum().plot()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAsMAAAHzCAYAAADFHhByAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTEREREQV6xensRpArPrJbQAAAABJRU5ErkJggg==\n",
       "text": [
        "<matplotlib.figure.Figure at 0x7f46e02a57b8>"
       ]
      }
     ],
     "prompt_number": 50
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Training a model in real time with Vowpal Wabbit\n",
      "\n",
      "Here's the fun bit. We're going to restart the stream, and as the data comes in, we're going to feed it to [Vowpal Wabbit](https://github.com/JohnLangford/vowpal_wabbit) to incrementally train a language recognizer. Or more strictly, a wiki recognizer, as we want to teach it to distinguish e.g. French Wikipedia articles from Polish Wikipedia articles -- using only the distribution of characters, and character n-grams, in their titles.\n",
      "\n",
      "**Jargon buster:** Character n-grams, sometimes called shingles, are subsequences of length `n`. So if `n == 2`, it's all the consecutive pairs of characters in the input string.\n",
      "\n",
      "We'll use an adaptive online learning approach: only update the model when we find an example that it can't correctly identify already. VW makes this very efficient and effective.\n",
      "\n",
      "First let's define a helper function to generate character n-grams, for `n` from `n1` to `n2`. That is, if `n1 == 2` and `n2 == 4`, then you'll get bigrams, trigrams and tetragrams (is that a word?). If `n1 == 1`, as in the example below, you'll get all the individual characters too. Also, if you set `include_words=True`, it will include all the complete words yielded by splitting the string on whitespace.\n",
      "\n",
      "The function returns an iterator over all the resulting n-grams. Any which consist purely of space characters are dropped."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Helper function to get all length-n ngrams\n",
      "def _ngrams(text, n):\n",
      "    return (text[i:i+n]\n",
      "            for i in range(len(text) - (n-1)))\n",
      "\n",
      "# Helper function to get all words, padded by a space on either side\n",
      "def _words(text):\n",
      "    return (s.center(len(s) + 2)\n",
      "            for s in text.split())\n",
      "\n",
      "# Helper function to filter out tokens that are all whitespace or numerals\n",
      "def _filter(word):\n",
      "    return (not word.strip().isspace()) and (not word.strip().isnumeric())\n",
      "\n",
      "# Get all ngrams for n1...n2, and optionally whole words\n",
      "def ngrams(text, n1, n2, include_words=False):\n",
      "    if n1 < 1:\n",
      "        raise ValueError('n1 must be 1 or more')\n",
      "    if n2 < n1:\n",
      "        raise ValueError('n2 must not be less than n1')\n",
      "    return filter(_filter,\n",
      "                  chain(_words(text) if include_words else (),\n",
      "                        chain.from_iterable(_ngrams(text, n)\n",
      "                                            for n in range(n1, n2+1))))\n",
      "\n",
      "print(list(ngrams(' the quick brown fox 1997', 1, 2, include_words=True)))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "[' the ', ' quick ', ' brown ', ' fox ', ' ', 't', 'h', 'e', ' ', 'q', 'u', 'i', 'c', 'k', ' ', 'b', 'r', 'o', 'w', 'n', ' ', 'f', 'o', 'x', ' ', ' t', 'th', 'he', 'e ', ' q', 'qu', 'ui', 'ic', 'ck', 'k ', ' b', 'br', 'ro', 'ow', 'wn', 'n ', ' f', 'fo', 'ox', 'x ']\n"
       ]
      }
     ],
     "prompt_number": 51
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The next function contains filtering logic, defining what kinds of edit we are interested in. It also performs some simple cleaning of the input data.\n",
      "\n",
      "We'll filter out Wikidata edits (wd), as these just have IDs, as well as Commons files (co), non-article pages like user homepages, and also English Wikipedia (en) -- to keep things interesting.\n",
      "\n",
      "The function returns just the page name if it's one we want to keep, or `False` if it's not. Parentheses and commas are removed, and hyphens and underscores replaced by spaces. A space is added at the start and end of the name, to help with ngram generation later."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "skip_reg = re.compile(r'[(),]+')\n",
      "split_reg = re.compile(r'[ _-]+')\n",
      "\n",
      "def page_name(edit):\n",
      "    if edit['wikipediaShort'] == 'wd' or \\\n",
      "        edit['wikipediaShort'] == 'en' or \\\n",
      "        edit['wikipediaShort'] == 'co' or \\\n",
      "        edit['namespace'] != 'article':\n",
      "            return False\n",
      "    name = split_reg.sub(' ', skip_reg.sub('', edit['page'].lower().strip()))\n",
      "    return name.center(len(name) + 2)\n",
      "\n",
      "test_edit = {'wikipediaShort': 'xx', 'namespace': 'article', 'page': 'Testing, Testing_ (one-two) '}\n",
      "print(page_name(test_edit))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " testing testing one two \n"
       ]
      }
     ],
     "prompt_number": 52
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "We'll use a [ringbuffer](http://en.wikipedia.org/wiki/Circular_buffer) to keep track of predictions and actual values over the last `k` tests. In this kind of online learning scenario, we don't really worry how much about how the model *was* performing on historical data -- we just want to know that it *is* performing well on current data.\n",
      "\n",
      "Here are a few convenience functions for dealing with the ringbuffer, which is implemented as a Numpy array with `k` rows, and a column each for the actual and predicted values. It's initialized with -1 in every cell, so we can tell when a row has yet to be filled in.\n",
      "\n",
      "`add_entry` uses a Numpy trick (`np.roll`) to redefine the start and end of an array, while copying as little data as possible."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Create a new ringbuffer for the given number of predictions\n",
      "def new_ringbuffer(capacity):\n",
      "    return np.zeros((capacity, 2), dtype=np.int8) - 1\n",
      "\n",
      "# Calculate accuracy (% correct predictions) over those rows which have been filled in\n",
      "def accuracy(ringbuffer):\n",
      "    # Filter identifying rows which have been filled in\n",
      "    valid = np.sum(ringbuffer, axis=1) >= 0\n",
      "    # Filter identifying rows where prediction == correct value\n",
      "    matched = ringbuffer[:,0] == ringbuffer[:,1]\n",
      "    # Accuracy is number of rows which are valid AND matched, over number of valid rows\n",
      "    return np.sum(np.logical_and(valid, matched)) / np.sum(valid)\n",
      "\n",
      "# Return the number of valid (filled-in) rows\n",
      "def valid_size(ringbuffer):\n",
      "    return np.sum(np.sum(ringbuffer, axis=1) >= 0)\n",
      "\n",
      "# Add a new test result to the ringbuffer, deleting the oldest entry if necessary,\n",
      "# and returning a reference to the modified buffer\n",
      "def add_entry(ringbuffer, actual, predicted):\n",
      "    capacity = len(ringbuffer)\n",
      "    # Overwrite the bottom row of the array\n",
      "    ringbuffer[capacity-1, :] = (actual, predicted)\n",
      "    # Roll the newest entry around to the top of the array, pushing the rest down\n",
      "    return np.roll(ringbuffer, 1, axis=0)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 53
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "All the action happens in the next cell.\n",
      "\n",
      "It starts the VW server with the appropriate options. If you get a memory allocation error, reduce the value of `hashbits`, although this may impact accuracy.\n",
      "\n",
      "It also defines and starts a new callback, which breaks the title of each page into n-grams, and tests them against the current model. If the model classified this edit's language correctly, we don't change it. But if it classifies incorrectly, we pass it the same edit as a training instance.\n",
      "\n",
      "The reported accuracy might fluctuate to begin with, but then should settle, and gradually rise (not necessarily in a monotonic way) as data comes in.\n",
      "\n",
      "*This may be an over-estimate of accuracy if multiple edits to the same page appear within the last `k` edits. For a more conservative estimate, we should filter out duplicates. But this is just a demo...*\n",
      "\n",
      "You can set a maximum number of training instances to process, or just **interrupt the kernel if you get bored of waiting**."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# How many classes to allow in the model (must be provided up front)\n",
      "max_expected_classes = 30\n",
      "\n",
      "# When to stop, if not interrupted\n",
      "max_edits = 100000\n",
      "\n",
      "# How many tests to calculate score over\n",
      "k = 500\n",
      "\n",
      "# Ringbuffer to hold results of each test\n",
      "rb = new_ringbuffer(k)\n",
      "\n",
      "# How many bits in range of VW hash function:\n",
      "# larger = potentially better accuracy but more memory\n",
      "hashbits = 29\n",
      "\n",
      "# How many edits we've received, and trained on, respectively\n",
      "received = 0\n",
      "trained = 0\n",
      "\n",
      "# List of the wiki names we've seen, to be populated as we go along\n",
      "labels = []\n",
      "\n",
      "# Maps each wiki name -> its index in labels list\n",
      "label_to_idx = {}\n",
      "\n",
      "def vw_callback(edit):\n",
      "    \n",
      "    name = page_name(edit)\n",
      "    if not name:\n",
      "        return True\n",
      "\n",
      "    global rb, received, trained\n",
      "    received += 1\n",
      "    \n",
      "    # Set up mappings for this wiki if this is the first time we've seen it\n",
      "    label = edit['wikipedia'].replace(' Wikipedia', '')\n",
      "    if label in label_to_idx:\n",
      "        idx = label_to_idx[label]\n",
      "    else:\n",
      "        # If we have exhausted our preset class limit, just skip it! Bit kludgey...\n",
      "        if len(labels) == max_expected_classes:\n",
      "            print(\"Ignoring class %d\" % len(labels))\n",
      "            return True\n",
      "        labels.append(label)\n",
      "        idx = len(labels) - 1\n",
      "        label_to_idx[label] = idx\n",
      "\n",
      "    # Generate binary features -- unigrams, bigrams, trigrams, and whole words\n",
      "    features = set(ngrams(name, 1, 3, include_words=True))\n",
      "\n",
      "    # Test example first -- does model get it right?\n",
      "    raw_line = vw.make_line(features=features)\n",
      "#    print(raw_line)\n",
      "    result = vw.send_line(raw_line).prediction\n",
      "    prediction = int(result) - 1 # NB VW FEATURE LABELS START AT 1\n",
      "\n",
      "    # Add prediction to ringbuffer\n",
      "    rb = add_entry(rb, idx, prediction)\n",
      "    \n",
      "    # Train on this example only if prediction was wrong\n",
      "    if idx == prediction:\n",
      "        #print('%s [%s] OK' % (name, labels[idx]), flush=True)\n",
      "        pass\n",
      "    else:\n",
      "        actual = idx if idx >= len(labels) else labels[idx]\n",
      "        predicted = prediction if prediction >= len(labels) else labels[prediction]\n",
      "        tested = valid_size(rb)\n",
      "        print('%s [%s] Failed!\\nPrediction: %s, Received: %d, Trained: %d, Last-%d Accuracy: %0.3f'\n",
      "              % (name, actual, predicted, received, trained, tested, accuracy(rb)),\n",
      "              flush=True)\n",
      "        raw_line = vw.make_line(idx+1, features=features)\n",
      "        vw.send_line(raw_line)\n",
      "        trained += 1\n",
      "\n",
      "    # Stop if we've hit max_edits\n",
      "    return received < max_edits\n",
      "\n",
      "# Start up local VW process\n",
      "vw = VW(loss_function='hinge', b=hashbits, ect=max_expected_classes, hash='all', q='nn')\n",
      "print(vw.command)\n",
      "\n",
      "try:\n",
      "    wikipedia_updates(vw_callback)\n",
      "finally:\n",
      "    vw.close()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Visualizing language similarity\n",
      "\n",
      "When the model mispredicts, which wikis is it most likely to get confused?\n",
      "\n",
      "This could give us some measure of the similarity of those languages -- in the limited domain of \"Wikipedia pages\" at least.\n",
      "\n",
      "If we visualize these relationships, do they roughly match what we'd expect from a non-expert idea of language similarity? Or do they reflect noise in the model -- artefacts of VW's feature hashing process?\n",
      "\n",
      "First do some preprocessing to get a dense confusion matrix, sorted so more popular wikis come first."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "confusion = np.zeros((len(labels), len(labels)))\n",
      "for (x, y) in rb:\n",
      "    confusion[x,y] += 1\n",
      "samples_per_wiki = np.asarray(confusion.sum(axis=1))\n",
      "sorted_idxs = np.argsort(samples_per_wiki)[::-1]\n",
      "sorted_confusion = confusion[sorted_idxs,:][:,sorted_idxs]\n",
      "sorted_confusion.shape"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 33,
       "text": [
        "(30, 30)"
       ]
      }
     ],
     "prompt_number": 33
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Visualize the confusion matrix in Matplotlib.\n",
      "\n",
      "The left edge represents the pages' real languages, the bottom edge shows what the model identified them as, at least in its last `k` tests. So, shading along the diagonal represents correct predictions, and shading off the diagonal represents errors.\n",
      "\n",
      "The values are absolute counts, so this shows the relative popularity of each wiki as well as the error distribution."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fig, ax = plt.subplots(1)\n",
      "p = ax.pcolormesh(sorted_confusion, cmap='Reds')\n",
      "fig.colorbar(p)\n",
      "\n",
      "labels_array = np.array(labels)\n",
      "\n",
      "ax.set_yticks(np.arange(sorted_confusion.shape[0])+0.5, minor=False)\n",
      "ax.set_xticks(np.arange(sorted_confusion.shape[1])+0.5, minor=False)\n",
      "ax.set_xticklabels(labels_array[sorted_idxs], minor=False) \n",
      "ax.set_yticklabels(labels_array[sorted_idxs], minor=False)\n",
      "plt.xticks(rotation=90)\n",
      "plt.show()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAArEAAAITCAYAAAAQDkTIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQqFQKJQGBw1iKRQKhUKhUCgNjv8Dkrv6PDKWc6gAAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x7f46e044cbe0>"
       ]
      }
     ],
     "prompt_number": 34
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Alternatively, we can turn it into a labelled list of edge weights, write it out to a CSV, and render it in D3.\n",
      "\n",
      "For this visualization, let's normalize by the number of actual occurrences of each wiki in the test set, so the weights indicate the proportion of the samples for each wiki `source` that were mistakenly predicted for `target` instead. Also we'll need to set the diagonal to 0 to remove loops from the visualization.\n",
      "\n",
      "**Mouseover the nodes in the graph to see the wiki name. Drag the nodes to rearrange the graph.**"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "normalized = sorted_confusion / sorted_confusion.sum(axis=1)\n",
      "np.fill_diagonal(normalized, 0)\n",
      "edge_weights = pd.DataFrame(data=normalized,\n",
      "                            index=labels_array[sorted_idxs],\n",
      "                            columns=labels_array[sorted_idxs]).stack()\n",
      "edge_weights.index.names = ('source', 'target')\n",
      "edge_weights.name = 'value'\n",
      "edge_weights.iloc[edge_weights.nonzero()].to_csv('cm.csv', header=True, index=True)\n",
      "from IPython.lib.display import IFrame\n",
      "IFrame('graph.html', 650, 650)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "\n",
        "        <iframe\n",
        "            width=\"650\"\n",
        "            height=650\"\n",
        "            src=\"graph.html\"\n",
        "            frameborder=\"0\"\n",
        "            allowfullscreen\n",
        "        ></iframe>\n",
        "        "
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 35,
       "text": [
        "<IPython.lib.display.IFrame at 0x7f46e04ad5c0>"
       ]
      }
     ],
     "prompt_number": 35
    }
   ],
   "metadata": {}
  }
 ]
}